From 15d2abf93eed3cacc16ccab58da092d011c59afa Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 1 Apr 2024 11:41:00 +0800
Subject: [PATCH 001/183] Utilizing subprocess for nnUNet training. (#7576)

Workaround for #7575

### Description
- Due to the impact of #7575, the operation to set the device within
nnUNetV2Runner will become ineffective. This PR is intended to resolve
this issue.
- Add a version check for #7575, will revisit after the update from
pytorch team.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/apps/nnunet/nnunetv2_runner.py | 94 +++++++++++++---------------
 tests/test_set_visible_devices.py    |  3 +-
 2 files changed, 45 insertions(+), 52 deletions(-)

diff --git a/monai/apps/nnunet/nnunetv2_runner.py b/monai/apps/nnunet/nnunetv2_runner.py
index 44b3c24256..8a10849904 100644
--- a/monai/apps/nnunet/nnunetv2_runner.py
+++ b/monai/apps/nnunet/nnunetv2_runner.py
@@ -22,6 +22,7 @@
 from monai.apps.nnunet.utils import analyze_data, create_new_data_copy, create_new_dataset_json
 from monai.bundle import ConfigParser
 from monai.utils import ensure_tuple, optional_import
+from monai.utils.misc import run_cmd
 
 load_pickle, _ = optional_import("batchgenerators.utilities.file_and_folder_operations", name="load_pickle")
 join, _ = optional_import("batchgenerators.utilities.file_and_folder_operations", name="join")
@@ -495,65 +496,64 @@ def train_single_model(self, config: Any, fold: int, gpu_id: tuple | list | int
             fold: fold of the 5-fold cross-validation. Should be an int between 0 and 4.
             gpu_id: an integer to select the device to use, or a tuple/list of GPU device indices used for multi-GPU
                 training (e.g., (0,1)). Default: 0.
-        from nnunetv2.run.run_training import run_training
             kwargs: this optional parameter allows you to specify additional arguments in
-                ``nnunetv2.run.run_training.run_training``. Currently supported args are
-                    - plans_identifier: custom plans identifier. Default: "nnUNetPlans".
-                    - pretrained_weights: path to nnU-Net checkpoint file to be used as pretrained model. Will only be
-                        used when actually training. Beta. Use with caution. Default: False.
-                    - use_compressed_data: True to use compressed data for training. Reading compressed data is much
-                        more CPU and (potentially) RAM intensive and should only be used if you know what you are
-                        doing. Default: False.
-                    - continue_training: continue training from latest checkpoint. Default: False.
-                    - only_run_validation: True to run the validation only. Requires training to have finished.
-                        Default: False.
-                    - disable_checkpointing: True to disable checkpointing. Ideal for testing things out and you
-                        don't want to flood your hard drive with checkpoints. Default: False.
+                ``nnunetv2.run.run_training.run_training_entry``.
+
+                Currently supported args are:
+
+                - p: custom plans identifier. Default: "nnUNetPlans".
+                - pretrained_weights: path to nnU-Net checkpoint file to be used as pretrained model. Will only be
+                    used when actually training. Beta. Use with caution. Default: False.
+                - use_compressed: True to use compressed data for training. Reading compressed data is much
+                    more CPU and (potentially) RAM intensive and should only be used if you know what you are
+                    doing. Default: False.
+                - c: continue training from latest checkpoint. Default: False.
+                - val: True to run the validation only. Requires training to have finished.
+                    Default: False.
+                - disable_checkpointing: True to disable checkpointing. Ideal for testing things out and you
+                    don't want to flood your hard drive with checkpoints. Default: False.
         """
         if "num_gpus" in kwargs:
             kwargs.pop("num_gpus")
             logger.warning("please use gpu_id to set the GPUs to use")
 
-        if "trainer_class_name" in kwargs:
-            kwargs.pop("trainer_class_name")
+        if "tr" in kwargs:
+            kwargs.pop("tr")
             logger.warning("please specify the `trainer_class_name` in the __init__ of `nnUNetV2Runner`.")
 
-        if "export_validation_probabilities" in kwargs:
-            kwargs.pop("export_validation_probabilities")
+        if "npz" in kwargs:
+            kwargs.pop("npz")
             logger.warning("please specify the `export_validation_probabilities` in the __init__ of `nnUNetV2Runner`.")
 
+        cmd = self.train_single_model_command(config, fold, gpu_id, kwargs)
+        run_cmd(cmd, shell=True)
+
+    def train_single_model_command(self, config, fold, gpu_id, kwargs):
         if isinstance(gpu_id, (tuple, list)):
             if len(gpu_id) > 1:
                 gpu_ids_str = ""
                 for _i in range(len(gpu_id)):
                     gpu_ids_str += f"{gpu_id[_i]},"
-                os.environ["CUDA_VISIBLE_DEVICES"] = gpu_ids_str[:-1]
+                device_setting = f"CUDA_VISIBLE_DEVICES={gpu_ids_str[:-1]}"
             else:
-                os.environ["CUDA_VISIBLE_DEVICES"] = f"{gpu_id[0]}"
-        else:
-            os.environ["CUDA_VISIBLE_DEVICES"] = f"{gpu_id}"
-
-        from nnunetv2.run.run_training import run_training
-
-        if isinstance(gpu_id, int) or len(gpu_id) == 1:
-            run_training(
-                dataset_name_or_id=self.dataset_name_or_id,
-                configuration=config,
-                fold=fold,
-                trainer_class_name=self.trainer_class_name,
-                export_validation_probabilities=self.export_validation_probabilities,
-                **kwargs,
-            )
+                device_setting = f"CUDA_VISIBLE_DEVICES={gpu_id[0]}"
         else:
-            run_training(
-                dataset_name_or_id=self.dataset_name_or_id,
-                configuration=config,
-                fold=fold,
-                num_gpus=len(gpu_id),
-                trainer_class_name=self.trainer_class_name,
-                export_validation_probabilities=self.export_validation_probabilities,
-                **kwargs,
-            )
+            device_setting = f"CUDA_VISIBLE_DEVICES={gpu_id}"
+        num_gpus = 1 if isinstance(gpu_id, int) or len(gpu_id) == 1 else len(gpu_id)
+
+        cmd = (
+            f"{device_setting} nnUNetv2_train "
+            + f"{self.dataset_name_or_id} {config} {fold} "
+            + f"-tr {self.trainer_class_name} -num_gpus {num_gpus}"
+        )
+        if self.export_validation_probabilities:
+            cmd += " --npz"
+        for _key, _value in kwargs.items():
+            if _key == "p" or _key == "pretrained_weights":
+                cmd += f" -{_key} {_value}"
+            else:
+                cmd += f" --{_key} {_value}"
+        return cmd
 
     def train(
         self,
@@ -637,15 +637,7 @@ def train_parallel_cmd(
                 if _config in ensure_tuple(configs):
                     for _i in range(self.num_folds):
                         the_device = gpu_id_for_all[_index % n_devices]  # type: ignore
-                        cmd = (
-                            "python -m monai.apps.nnunet nnUNetV2Runner train_single_model "
-                            + f"--input_config '{self.input_config_or_dict}' --work_dir '{self.work_dir}' "
-                            + f"--config '{_config}' --fold {_i} --gpu_id {the_device} "
-                            + f"--trainer_class_name {self.trainer_class_name} "
-                            + f"--export_validation_probabilities {self.export_validation_probabilities}"
-                        )
-                        for _key, _value in kwargs.items():
-                            cmd += f" --{_key} {_value}"
+                        cmd = self.train_single_model_command(_config, _i, the_device, kwargs)
                         all_cmds[-1][the_device].append(cmd)
                         _index += 1
         return all_cmds
diff --git a/tests/test_set_visible_devices.py b/tests/test_set_visible_devices.py
index 7860656b3d..b4f44957a2 100644
--- a/tests/test_set_visible_devices.py
+++ b/tests/test_set_visible_devices.py
@@ -14,7 +14,7 @@
 import os
 import unittest
 
-from tests.utils import skip_if_no_cuda
+from tests.utils import SkipIfAtLeastPyTorchVersion, skip_if_no_cuda
 
 
 class TestVisibleDevices(unittest.TestCase):
@@ -25,6 +25,7 @@ def run_process_and_get_exit_code(code_to_execute):
         return int(bin(value).replace("0b", "").rjust(16, "0")[:8], 2)
 
     @skip_if_no_cuda
+    @SkipIfAtLeastPyTorchVersion((2, 2, 1))
     def test_visible_devices(self):
         num_gpus_before = self.run_process_and_get_exit_code(
             'python -c "import os; import torch; '

From ec4d946571937ad162b415b1ea5977efc5adcfcd Mon Sep 17 00:00:00 2001
From: Vladimir Chernyi <57420464+scalyvladimir@users.noreply.github.com>
Date: Mon, 1 Apr 2024 07:28:56 +0300
Subject: [PATCH 002/183] typo fix (#7595)

Fixes # 1.

### Description

Fixed typo.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] In-line docstrings updated.

Signed-off-by: Vladimir Chernyi <57420464+scalyvladimir@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/data/dataset.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/data/dataset.py b/monai/data/dataset.py
index 531893d768..79e066303e 100644
--- a/monai/data/dataset.py
+++ b/monai/data/dataset.py
@@ -427,7 +427,7 @@ def _transform(self, index: int):
 
 class CacheNTransDataset(PersistentDataset):
     """
-    Extension of `PersistentDataset`, tt can also cache the result of first N transforms, no matter it's random or not.
+    Extension of `PersistentDataset`, it can also cache the result of first N transforms, no matter it's random or not.
 
     """
 

From a7c2589133ab01afd75a762885dce69a73abfe1a Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Mon, 1 Apr 2024 08:13:31 +0100
Subject: [PATCH 003/183] auto updates (#7599)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 monai/transforms/regularization/array.py | 1 +
 tests/test_regularization.py             | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/monai/transforms/regularization/array.py b/monai/transforms/regularization/array.py
index 6c9022d647..0b495c8623 100644
--- a/monai/transforms/regularization/array.py
+++ b/monai/transforms/regularization/array.py
@@ -22,6 +22,7 @@
 
 
 class Mixer(RandomizableTransform):
+
     def __init__(self, batch_size: int, alpha: float = 1.0) -> None:
         """
         Mixer is a base class providing the basic logic for the mixup-class of
diff --git a/tests/test_regularization.py b/tests/test_regularization.py
index d381ea72ca..22faf1027d 100644
--- a/tests/test_regularization.py
+++ b/tests/test_regularization.py
@@ -19,6 +19,7 @@
 
 
 class TestMixup(unittest.TestCase):
+
     def test_mixup(self):
         for dims in [2, 3]:
             shape = (6, 3) + (32,) * dims
@@ -52,6 +53,7 @@ def test_mixupd(self):
 
 
 class TestCutMix(unittest.TestCase):
+
     def test_cutmix(self):
         for dims in [2, 3]:
             shape = (6, 3) + (32,) * dims
@@ -76,6 +78,7 @@ def test_cutmixd(self):
 
 
 class TestCutOut(unittest.TestCase):
+
     def test_cutout(self):
         for dims in [2, 3]:
             shape = (6, 3) + (32,) * dims

From c885100e951e5b7b1fedf5e9bc965201592e8541 Mon Sep 17 00:00:00 2001
From: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Date: Mon, 1 Apr 2024 15:56:38 +0800
Subject: [PATCH 004/183] 7540 change bundle workflow args (#7549)

Fixes #7540  .

### Description

This PR:
1. add logging file and meta file into BundleWorkflow
2. add the sequence form of meta files check for ConfigWorkflow

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/bundle/workflows.py     | 70 ++++++++++++++++++++++++++---------
 tests/nonconfig_workflow.py   |  4 +-
 tests/test_bundle_workflow.py |  8 ++++
 3 files changed, 63 insertions(+), 19 deletions(-)

diff --git a/monai/bundle/workflows.py b/monai/bundle/workflows.py
index da3aa30141..804b3b06f0 100644
--- a/monai/bundle/workflows.py
+++ b/monai/bundle/workflows.py
@@ -46,6 +46,9 @@ class BundleWorkflow(ABC):
             or "infer", "inference", "eval", "evaluation" for a inference workflow,
             other unsupported string will raise a ValueError.
             default to `None` for common workflow.
+        meta_file: filepath of the metadata file, if this is a list of file paths, their contents will be merged in order.
+        logging_file: config file for `logging` module in the program. for more details:
+            https://docs.python.org/3/library/logging.config.html#logging.config.fileConfig.
 
     """
 
@@ -59,11 +62,40 @@ class BundleWorkflow(ABC):
         new_name="workflow_type",
         msg_suffix="please use `workflow_type` instead.",
     )
-    def __init__(self, workflow_type: str | None = None, workflow: str | None = None):
+    def __init__(
+        self,
+        workflow_type: str | None = None,
+        workflow: str | None = None,
+        meta_file: str | Sequence[str] | None = None,
+        logging_file: str | None = None,
+    ):
+        if logging_file is not None:
+            if not os.path.isfile(logging_file):
+                raise FileNotFoundError(f"Cannot find the logging config file: {logging_file}.")
+            logger.info(f"Setting logging properties based on config: {logging_file}.")
+            fileConfig(logging_file, disable_existing_loggers=False)
+
+        if meta_file is not None:
+            if isinstance(meta_file, str) and not os.path.isfile(meta_file):
+                logger.error(
+                    f"Cannot find the metadata config file: {meta_file}. "
+                    "Please see: https://docs.monai.io/en/stable/mb_specification.html"
+                )
+                meta_file = None
+            if isinstance(meta_file, list):
+                for f in meta_file:
+                    if not os.path.isfile(f):
+                        logger.error(
+                            f"Cannot find the metadata config file: {f}. "
+                            "Please see: https://docs.monai.io/en/stable/mb_specification.html"
+                        )
+                        meta_file = None
+
         workflow_type = workflow if workflow is not None else workflow_type
         if workflow_type is None:
             self.properties = copy(MetaProperties)
             self.workflow_type = None
+            self.meta_file = meta_file
             return
         if workflow_type.lower() in self.supported_train_type:
             self.properties = {**TrainProperties, **MetaProperties}
@@ -74,6 +106,8 @@ def __init__(self, workflow_type: str | None = None, workflow: str | None = None
         else:
             raise ValueError(f"Unsupported workflow type: '{workflow_type}'.")
 
+        self.meta_file = meta_file
+
     @abstractmethod
     def initialize(self, *args: Any, **kwargs: Any) -> Any:
         """
@@ -142,6 +176,13 @@ def get_workflow_type(self):
         """
         return self.workflow_type
 
+    def get_meta_file(self):
+        """
+        Get the meta file.
+
+        """
+        return self.meta_file
+
     def add_property(self, name: str, required: str, desc: str | None = None) -> None:
         """
         Besides the default predefined properties, some 3rd party applications may need the bundle
@@ -233,25 +274,26 @@ def __init__(
         **override: Any,
     ) -> None:
         workflow_type = workflow if workflow is not None else workflow_type
-        super().__init__(workflow_type=workflow_type)
         if config_file is not None:
             _config_files = ensure_tuple(config_file)
-            self.config_root_path = Path(_config_files[0]).parent
+            config_root_path = Path(_config_files[0]).parent
             for _config_file in _config_files:
                 _config_file = Path(_config_file)
-                if _config_file.parent != self.config_root_path:
+                if _config_file.parent != config_root_path:
                     logger.warn(
-                        f"Not all config files are in {self.config_root_path}. If logging_file and meta_file are"
-                        f"not specified, {self.config_root_path} will be used as the default config root directory."
+                        f"Not all config files are in {config_root_path}. If logging_file and meta_file are"
+                        f"not specified, {config_root_path} will be used as the default config root directory."
                     )
                 if not _config_file.is_file():
                     raise FileNotFoundError(f"Cannot find the config file: {_config_file}.")
         else:
-            self.config_root_path = Path("configs")
-
+            config_root_path = Path("configs")
+        meta_file = str(config_root_path / "metadata.json") if meta_file is None else meta_file
+        super().__init__(workflow_type=workflow_type, meta_file=meta_file)
+        self.config_root_path = config_root_path
         logging_file = str(self.config_root_path / "logging.conf") if logging_file is None else logging_file
         if logging_file is not None:
-            if not os.path.exists(logging_file):
+            if not os.path.isfile(logging_file):
                 if logging_file == str(self.config_root_path / "logging.conf"):
                     logger.warn(f"Default logging file in {logging_file} does not exist, skipping logging.")
                 else:
@@ -262,14 +304,8 @@ def __init__(
 
         self.parser = ConfigParser()
         self.parser.read_config(f=config_file)
-        meta_file = str(self.config_root_path / "metadata.json") if meta_file is None else meta_file
-        if isinstance(meta_file, str) and not os.path.exists(meta_file):
-            logger.error(
-                f"Cannot find the metadata config file: {meta_file}. "
-                "Please see: https://docs.monai.io/en/stable/mb_specification.html"
-            )
-        else:
-            self.parser.read_meta(f=meta_file)
+        if self.meta_file is not None:
+            self.parser.read_meta(f=self.meta_file)
 
         # the rest key-values in the _args are to override config content
         self.parser.update(pairs=override)
diff --git a/tests/nonconfig_workflow.py b/tests/nonconfig_workflow.py
index 7b5328bf72..b2c44c12c6 100644
--- a/tests/nonconfig_workflow.py
+++ b/tests/nonconfig_workflow.py
@@ -36,8 +36,8 @@ class NonConfigWorkflow(BundleWorkflow):
 
     """
 
-    def __init__(self, filename, output_dir):
-        super().__init__(workflow_type="inference")
+    def __init__(self, filename, output_dir, meta_file=None, logging_file=None):
+        super().__init__(workflow_type="inference", meta_file=meta_file, logging_file=logging_file)
         self.filename = filename
         self.output_dir = output_dir
         self._bundle_root = "will override"
diff --git a/tests/test_bundle_workflow.py b/tests/test_bundle_workflow.py
index f7da37acef..0b0d51cbfb 100644
--- a/tests/test_bundle_workflow.py
+++ b/tests/test_bundle_workflow.py
@@ -35,6 +35,8 @@
 
 TEST_CASE_3 = [os.path.join(os.path.dirname(__file__), "testing_data", "config_fl_train.json")]
 
+TEST_CASE_NON_CONFIG_WRONG_LOG = [None, "logging.conf", "Cannot find the logging config file: logging.conf."]
+
 
 class TestBundleWorkflow(unittest.TestCase):
 
@@ -144,8 +146,14 @@ def test_train_config(self, config_file):
     def test_non_config(self):
         # test user defined python style workflow
         inferer = NonConfigWorkflow(self.filename, self.data_dir)
+        self.assertEqual(inferer.meta_file, None)
         self._test_inferer(inferer)
 
+    @parameterized.expand([TEST_CASE_NON_CONFIG_WRONG_LOG])
+    def test_non_config_wrong_log_cases(self, meta_file, logging_file, expected_error):
+        with self.assertRaisesRegex(FileNotFoundError, expected_error):
+            NonConfigWorkflow(self.filename, self.data_dir, meta_file, logging_file)
+
 
 if __name__ == "__main__":
     unittest.main()

From 264b9e4700667e895c272f8da6a5a43d0c50d865 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 1 Apr 2024 18:09:35 +0800
Subject: [PATCH 005/183] Add "properties_path" in BundleWorkflow (#7542)

Fixes #7541


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/bundle/workflows.py                   | 23 +++++--
 tests/test_bundle_workflow.py               | 10 +++
 tests/test_regularization.py                | 16 +++++
 tests/testing_data/fl_infer_properties.json | 67 +++++++++++++++++++++
 4 files changed, 112 insertions(+), 4 deletions(-)
 create mode 100644 tests/testing_data/fl_infer_properties.json

diff --git a/monai/bundle/workflows.py b/monai/bundle/workflows.py
index 804b3b06f0..d876f6d7ae 100644
--- a/monai/bundle/workflows.py
+++ b/monai/bundle/workflows.py
@@ -11,6 +11,7 @@
 
 from __future__ import annotations
 
+import json
 import os
 import sys
 import time
@@ -24,6 +25,7 @@
 from monai.bundle.config_parser import ConfigParser
 from monai.bundle.properties import InferProperties, MetaProperties, TrainProperties
 from monai.bundle.utils import DEFAULT_EXP_MGMT_SETTINGS, EXPR_KEY, ID_REF_KEY, ID_SEP_KEY
+from monai.config import PathLike
 from monai.utils import BundleProperty, BundlePropertyConfig, deprecated_arg, deprecated_arg_default, ensure_tuple
 
 __all__ = ["BundleWorkflow", "ConfigWorkflow"]
@@ -46,6 +48,7 @@ class BundleWorkflow(ABC):
             or "infer", "inference", "eval", "evaluation" for a inference workflow,
             other unsupported string will raise a ValueError.
             default to `None` for common workflow.
+        properties_path: the path to the JSON file of properties.
         meta_file: filepath of the metadata file, if this is a list of file paths, their contents will be merged in order.
         logging_file: config file for `logging` module in the program. for more details:
             https://docs.python.org/3/library/logging.config.html#logging.config.fileConfig.
@@ -66,6 +69,7 @@ def __init__(
         self,
         workflow_type: str | None = None,
         workflow: str | None = None,
+        properties_path: PathLike | None = None,
         meta_file: str | Sequence[str] | None = None,
         logging_file: str | None = None,
     ):
@@ -92,15 +96,24 @@ def __init__(
                         meta_file = None
 
         workflow_type = workflow if workflow is not None else workflow_type
-        if workflow_type is None:
+        if workflow_type is None and properties_path is None:
             self.properties = copy(MetaProperties)
             self.workflow_type = None
             self.meta_file = meta_file
             return
-        if workflow_type.lower() in self.supported_train_type:
+        if properties_path is not None:
+            properties_path = Path(properties_path)
+            if not properties_path.is_file():
+                raise ValueError(f"Property file {properties_path} does not exist.")
+            with open(properties_path) as json_file:
+                self.properties = json.load(json_file)
+            self.workflow_type = None
+            self.meta_file = meta_file
+            return
+        if workflow_type.lower() in self.supported_train_type:  # type: ignore[union-attr]
             self.properties = {**TrainProperties, **MetaProperties}
             self.workflow_type = "train"
-        elif workflow_type.lower() in self.supported_infer_type:
+        elif workflow_type.lower() in self.supported_infer_type:  # type: ignore[union-attr]
             self.properties = {**InferProperties, **MetaProperties}
             self.workflow_type = "infer"
         else:
@@ -247,6 +260,7 @@ class ConfigWorkflow(BundleWorkflow):
             or "infer", "inference", "eval", "evaluation" for a inference workflow,
             other unsupported string will raise a ValueError.
             default to `None` for common workflow.
+        properties_path: the path to the JSON file of properties.
         override: id-value pairs to override or add the corresponding config content.
             e.g. ``--net#input_chns 42``, ``--net %/data/other.json#net_arg``
 
@@ -271,6 +285,7 @@ def __init__(
         tracking: str | dict | None = None,
         workflow_type: str | None = None,
         workflow: str | None = None,
+        properties_path: PathLike | None = None,
         **override: Any,
     ) -> None:
         workflow_type = workflow if workflow is not None else workflow_type
@@ -289,7 +304,7 @@ def __init__(
         else:
             config_root_path = Path("configs")
         meta_file = str(config_root_path / "metadata.json") if meta_file is None else meta_file
-        super().__init__(workflow_type=workflow_type, meta_file=meta_file)
+        super().__init__(workflow_type=workflow_type, meta_file=meta_file, properties_path=properties_path)
         self.config_root_path = config_root_path
         logging_file = str(self.config_root_path / "logging.conf") if logging_file is None else logging_file
         if logging_file is not None:
diff --git a/tests/test_bundle_workflow.py b/tests/test_bundle_workflow.py
index 0b0d51cbfb..9a276b577f 100644
--- a/tests/test_bundle_workflow.py
+++ b/tests/test_bundle_workflow.py
@@ -105,6 +105,16 @@ def test_inference_config(self, config_file):
         )
         self._test_inferer(inferer)
 
+        # test property path
+        inferer = ConfigWorkflow(
+            config_file=config_file,
+            properties_path=os.path.join(os.path.dirname(__file__), "testing_data", "fl_infer_properties.json"),
+            logging_file=os.path.join(os.path.dirname(__file__), "testing_data", "logging.conf"),
+            **override,
+        )
+        self._test_inferer(inferer)
+        self.assertEqual(inferer.workflow_type, None)
+
     @parameterized.expand([TEST_CASE_3])
     def test_train_config(self, config_file):
         # test standard MONAI model-zoo config workflow
diff --git a/tests/test_regularization.py b/tests/test_regularization.py
index 22faf1027d..c6f727cb54 100644
--- a/tests/test_regularization.py
+++ b/tests/test_regularization.py
@@ -16,9 +16,15 @@
 import torch
 
 from monai.transforms import CutMix, CutMixd, CutOut, MixUp, MixUpd
+from monai.utils import set_determinism
 
 
 class TestMixup(unittest.TestCase):
+    def setUp(self) -> None:
+        set_determinism(seed=0)
+
+    def tearDown(self) -> None:
+        set_determinism(None)
 
     def test_mixup(self):
         for dims in [2, 3]:
@@ -53,6 +59,11 @@ def test_mixupd(self):
 
 
 class TestCutMix(unittest.TestCase):
+    def setUp(self) -> None:
+        set_determinism(seed=0)
+
+    def tearDown(self) -> None:
+        set_determinism(None)
 
     def test_cutmix(self):
         for dims in [2, 3]:
@@ -78,6 +89,11 @@ def test_cutmixd(self):
 
 
 class TestCutOut(unittest.TestCase):
+    def setUp(self) -> None:
+        set_determinism(seed=0)
+
+    def tearDown(self) -> None:
+        set_determinism(None)
 
     def test_cutout(self):
         for dims in [2, 3]:
diff --git a/tests/testing_data/fl_infer_properties.json b/tests/testing_data/fl_infer_properties.json
new file mode 100644
index 0000000000..72e97cd2c6
--- /dev/null
+++ b/tests/testing_data/fl_infer_properties.json
@@ -0,0 +1,67 @@
+{
+    "bundle_root": {
+        "description": "root path of the bundle.",
+        "required": true,
+        "id": "bundle_root"
+    },
+    "device": {
+        "description": "target device to execute the bundle workflow.",
+        "required": true,
+        "id": "device"
+    },
+    "dataset_dir": {
+        "description": "directory path of the dataset.",
+        "required": true,
+        "id": "dataset_dir"
+    },
+    "dataset": {
+        "description": "PyTorch dataset object for the inference / evaluation logic.",
+        "required": true,
+        "id": "dataset"
+    },
+    "evaluator": {
+        "description": "inference / evaluation workflow engine.",
+        "required": true,
+        "id": "evaluator"
+    },
+    "network_def": {
+        "description": "network module for the inference.",
+        "required": true,
+        "id": "network_def"
+    },
+    "inferer": {
+        "description": "MONAI Inferer object to execute the model computation in inference.",
+        "required": true,
+        "id": "inferer"
+    },
+    "dataset_data": {
+        "description": "data source for the inference / evaluation dataset.",
+        "required": false,
+        "id": "dataset::data",
+        "refer_id": null
+    },
+    "handlers": {
+        "description": "event-handlers for the inference / evaluation logic.",
+        "required": false,
+        "id": "handlers",
+        "refer_id": "evaluator::val_handlers"
+    },
+    "preprocessing": {
+        "description": "preprocessing for the input data.",
+        "required": false,
+        "id": "preprocessing",
+        "refer_id": "dataset::transform"
+    },
+    "postprocessing": {
+        "description": "postprocessing for the model output data.",
+        "required": false,
+        "id": "postprocessing",
+        "refer_id": "evaluator::postprocessing"
+    },
+    "key_metric": {
+        "description": "the key metric during evaluation.",
+        "required": false,
+        "id": "key_metric",
+        "refer_id": "evaluator::key_val_metric"
+    }
+}

From bbaaf4c7d75976b0289152541afa9fa03fa44f7a Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Mon, 1 Apr 2024 15:39:10 +0100
Subject: [PATCH 006/183] Auto3DSeg algo_template hash update (#7603)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 monai/utils/misc.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index c30eb0904d..dd0ccada3d 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -527,7 +527,7 @@ def doc_images() -> str | None:
 
     @staticmethod
     def algo_hash() -> str | None:
-        return os.environ.get("MONAI_ALGO_HASH", "c51bc6a")
+        return os.environ.get("MONAI_ALGO_HASH", "b910ab8")
 
     @staticmethod
     def trace_transform() -> str | None:

From 5ec7305abfa406fec7f013953ec4da7cfcae15ff Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?L=C5=91rincz-Moln=C3=A1r=20Szabolcs-Botond?=
 <lorincz_szabolcs@yahoo.com>
Date: Tue, 2 Apr 2024 04:13:28 +0200
Subject: [PATCH 007/183] ENH: generate_label_classes_crop_centers: warn only
 if ratio of missing class is not set to 0 (#7602)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes #7594

### Description

Only warn users that the ratio of a missing class is set to 0, when it
wasn't already set to 0 by the user, in
`generate_label_classes_crop_centers`, function being used by
`RandCropByLabelClasses` transform.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`. ⚠️ See notes
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`. ⚠️ See notes
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

### Notes regarding tests
Some tests were failing, see details:
<details>
======================================================================
ERROR:
test_cuda_0_2_batches_1_dimensions_1_channels_2_classes_2_mixtures
(tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 288, in
test_cuda
gmm = GaussianMixtureModel(features_tensor.size(1), mixture_count,
class_count, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/networks/layers/gmm.py", line
44, in __init__
    self.compiled_extension = load_module(
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'gmm_1_2_1_Linux_3_10_12_22_12_1'

======================================================================
ERROR:
test_cuda_1_1_batches_1_dimensions_5_channels_2_classes_1_mixtures
(tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 288, in
test_cuda
gmm = GaussianMixtureModel(features_tensor.size(1), mixture_count,
class_count, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/networks/layers/gmm.py", line
44, in __init__
    self.compiled_extension = load_module(
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'gmm_5_2_1_Linux_3_10_12_22_12_1'

======================================================================
ERROR:
test_cuda_2_1_batches_2_dimensions_2_channels_4_classes_4_mixtures
(tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 288, in
test_cuda
gmm = GaussianMixtureModel(features_tensor.size(1), mixture_count,
class_count, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/networks/layers/gmm.py", line
44, in __init__
    self.compiled_extension = load_module(
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'gmm_2_4_1_Linux_3_10_12_22_12_1'

======================================================================
ERROR:
test_cuda_3_1_batches_3_dimensions_1_channels_2_classes_1_mixtures
(tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 288, in
test_cuda
gmm = GaussianMixtureModel(features_tensor.size(1), mixture_count,
class_count, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/networks/layers/gmm.py", line
44, in __init__
    self.compiled_extension = load_module(
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension
'gmm_1_2_1_Linux_3_10_12_22_12_1_v1'

======================================================================
ERROR: test_load (tests.test_gmm.GMMTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2096, in _run_ninja_build
    subprocess.run(
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned
non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/home/szabolcslorincz/MONAI/tests/test_gmm.py", line 311, in
test_load
load_module("gmm", {"CHANNEL_COUNT": 2, "MIXTURE_COUNT": 2,
"MIXTURE_SIZE": 3}, verbose_build=True)
File "/home/szabolcslorincz/MONAI/monai/_extensions/loader.py", line 89,
in load_module
    module = load(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1306, in load
    return _jit_compile(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1710, in _jit_compile
    _write_ninja_file_and_build_library(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 1823, in _write_ninja_file_and_build_library
    _run_ninja_build(
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/utils/cpp_extension.py",
line 2112, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'gmm_2_2_3_Linux_3_10_12_22_12_1'

======================================================================
ERROR: test_spacing_35 (tests.test_spacing.TestSpacingCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_spacing.py", line 289, in
test_spacing
    res: MetaTensor = tr(**call_param)  # type: ignore
File "/home/szabolcslorincz/MONAI/monai/transforms/spatial/array.py",
line 525, in __call__
    data_array = self.sp_resample(
File "/home/szabolcslorincz/MONAI/monai/transforms/spatial/array.py",
line 223, in __call__
    return spatial_resample(
File
"/home/szabolcslorincz/MONAI/monai/transforms/spatial/functional.py",
line 178, in spatial_resample
img = affine_xform(img.unsqueeze(0), theta=xform.to(img),
spatial_size=spatial_size).squeeze(0) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/nn/modules/module.py",
line 1511, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/nn/modules/module.py",
line 1520, in _call_impl
    return forward_call(*args, **kwargs)
File
"/home/szabolcslorincz/MONAI/monai/networks/layers/spatial_transforms.py",
line 584, in forward
grid = nn.functional.affine_grid(theta=theta[:, :sr],
size=list(dst_size), align_corners=self.align_corners)
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/nn/functional.py",
line 4418, in affine_grid
    return torch.affine_grid_generator(theta, size, align_corners)
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate
840.00 MiB. GPU 0 has a total capacity of 8.00 GiB of which 0 bytes is
free. Including non-PyTorch memory, this process has 17179869184.00 GiB
memory in use. Of the allocated memory 6.47 GiB is allocated by PyTorch,
and 156.63 MiB is reserved by PyTorch but unallocated. If reserved but
unallocated memory is large try setting
PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation.
See documentation for Memory Management
(https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)

======================================================================
FAIL: test_seg_res_net_1_cuda
(tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
108, in test_seg_res_net
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 233543 / 1451520 (16.1%)
Greatest absolute difference: 0.0014852285385131836 at index (0, 19, 12,
13, 22) (up to 0.0001 allowed)
Greatest relative difference: 589.0405883789062 at index (0, 32, 21, 16,
15) (up to 0.001 allowed)

======================================================================
FAIL: test_unet_4_cuda (tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
57, in test_unet
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 6705 / 49152 (13.6%)
Greatest absolute difference: 0.0015408992767333984 at index (11, 0, 21,
27) (up to 0.0001 allowed)
Greatest relative difference: 51.80112838745117 at index (14, 0, 25, 19)
(up to 0.001 allowed)

======================================================================
FAIL: test_unet_5_cuda (tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
57, in test_unet
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 6218 / 49152 (12.7%)
Greatest absolute difference: 0.0015670061111450195 at index (2, 0, 23,
14) (up to 0.0001 allowed)
Greatest relative difference: 7.987473964691162 at index (8, 0, 27, 8)
(up to 0.001 allowed)

======================================================================
FAIL: test_unet_6_cuda (tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
57, in test_unet
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 6743 / 49152 (13.7%)
Greatest absolute difference: 0.0015552043914794922 at index (1, 1, 9,
11) (up to 0.0001 allowed)
Greatest relative difference: 2.0317020416259766 at index (11, 0, 19,
21) (up to 0.001 allowed)

======================================================================
FAIL: test_unet_7_cuda (tests.test_convert_to_onnx.TestConvertToOnnx)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_convert_to_onnx.py", line
57, in test_unet
    onnx_model = convert_to_onnx(
File "/home/szabolcslorincz/MONAI/monai/networks/utils.py", line 709, in
convert_to_onnx
assert_fn(r1.cpu(), convert_to_tensor(r2, dtype=r1.dtype), rtol=rtol,
atol=atol) # type: ignore
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/torch/testing/_comparison.py",
line 1520, in assert_close
    raise error_metas[0].to_error(msg)
AssertionError: Tensor-likes are not close!

Mismatched elements: 6823 / 49152 (13.9%)
Greatest absolute difference: 0.0018431544303894043 at index (10, 0, 9,
19) (up to 0.0001 allowed)
Greatest relative difference: 4.297887325286865 at index (13, 0, 12, 13)
(up to 0.001 allowed)

======================================================================
FAIL: test_shape_2 (tests.test_multi_scale.TestMultiScale)
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/parameterized/parameterized.py",
line 620, in standalone_func
    return func(*(a + p.args), **p.kwargs, **kw)
File "/home/szabolcslorincz/MONAI/tests/test_multi_scale.py", line 59,
in test_shape
np.testing.assert_allclose(result.detach().cpu().numpy(), expected_val,
rtol=1e-5)
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/numpy/testing/_private/utils.py",
line 1504, in assert_allclose
    assert_array_compare(compare, actual, desired, err_msg=str(err_msg),
  File "/usr/lib/python3.10/contextlib.py", line 79, in inner
    return func(*args, **kwds)
File
"/home/szabolcslorincz/MONAI/venv/lib/python3.10/site-packages/numpy/testing/_private/utils.py",
line 797, in assert_array_compare
    raise AssertionError(msg)
AssertionError:
Not equal to tolerance rtol=1e-05, atol=0

Mismatched elements: 1 / 1 (100%)
Max absolute difference: 0.
Max relative difference: 0.
 x: array(0.715212, dtype=float32)
 y: array(0.715228)

----------------------------------------------------------------------
Ran 14392 tests in 2252.916s

FAILED (failures=6, errors=6, skipped=450)

---------

Signed-off-by: Szabolcs Botond Lorincz Molnar <szabolcs.lorincz@mvision.ai>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/transforms/utils.py | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index e282ecff24..455b0cfb7d 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -625,9 +625,12 @@ def generate_label_classes_crop_centers(
 
     for i, array in enumerate(indices):
         if len(array) == 0:
-            ratios_[i] = 0
-            if warn:
-                warnings.warn(f"no available indices of class {i} to crop, set the crop ratio of this class to zero.")
+            if ratios_[i] != 0:
+                ratios_[i] = 0
+                if warn:
+                    warnings.warn(
+                        f"no available indices of class {i} to crop, setting the crop ratio of this class to zero."
+                    )
 
     centers = []
     classes = rand_state.choice(len(ratios_), size=num_samples, p=np.asarray(ratios_) / np.sum(ratios_))

From 763347d0c945d5ea87916b86240c4368d818a286 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 3 Apr 2024 21:12:58 +0800
Subject: [PATCH 008/183] Update base image to 2403 (#7600)

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Nic Ma <nma@nvidia.com>
---
 .github/workflows/cron.yml          | 16 ++++++++--------
 .github/workflows/pythonapp-gpu.yml |  8 ++++----
 Dockerfile                          |  2 +-
 3 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/.github/workflows/cron.yml b/.github/workflows/cron.yml
index 792fda5279..0f9e6cd480 100644
--- a/.github/workflows/cron.yml
+++ b/.github/workflows/cron.yml
@@ -19,18 +19,18 @@ jobs:
           - "PTLATEST+CUDA121"
         include:
           # https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes
-          - environment: PT191+CUDA113
-            pytorch: "torch==1.9.1 torchvision==0.10.1 --extra-index-url https://download.pytorch.org/whl/cu113"
-            base: "nvcr.io/nvidia/pytorch:21.06-py3"  # CUDA 11.3
           - environment: PT110+CUDA113
             pytorch: "torch==1.10.2 torchvision==0.11.3 --extra-index-url https://download.pytorch.org/whl/cu113"
             base: "nvcr.io/nvidia/pytorch:21.06-py3"  # CUDA 11.3
           - environment: PT113+CUDA113
             pytorch: "torch==1.13.1 torchvision==0.14.1 --extra-index-url https://download.pytorch.org/whl/cu113"
             base: "nvcr.io/nvidia/pytorch:21.06-py3"  # CUDA 11.3
-          - environment: PTLATEST+CUDA121
-            pytorch: "-U torch torchvision --extra-index-url https://download.pytorch.org/whl/cu118"
+          - environment: PT113+CUDA122
+            pytorch: "torch==1.13.1 torchvision==0.14.1 --extra-index-url https://download.pytorch.org/whl/cu121"
             base: "nvcr.io/nvidia/pytorch:23.08-py3"  # CUDA 12.2
+          - environment: PTLATEST+CUDA124
+            pytorch: "-U torch torchvision --extra-index-url https://download.pytorch.org/whl/cu121"
+            base: "nvcr.io/nvidia/pytorch:24.03-py3"  # CUDA 12.4
     container:
       image: ${{ matrix.base }}
       options: "--gpus all"
@@ -76,7 +76,7 @@ jobs:
     if: github.repository == 'Project-MONAI/MONAI'
     strategy:
       matrix:
-        container: ["pytorch:22.10", "pytorch:23.08"]
+        container: ["pytorch:23.08", "pytorch:24.03"]
     container:
       image: nvcr.io/nvidia/${{ matrix.container }}-py3  # testing with the latest pytorch base image
       options: "--gpus all"
@@ -121,7 +121,7 @@ jobs:
     if: github.repository == 'Project-MONAI/MONAI'
     strategy:
       matrix:
-        container: ["pytorch:23.08"]
+        container: ["pytorch:24.03"]
     container:
       image: nvcr.io/nvidia/${{ matrix.container }}-py3  # testing with the latest pytorch base image
       options: "--gpus all"
@@ -221,7 +221,7 @@ jobs:
     if: github.repository == 'Project-MONAI/MONAI'
     needs: cron-gpu  # so that monai itself is verified first
     container:
-      image: nvcr.io/nvidia/pytorch:23.08-py3  # testing with the latest pytorch base image
+      image: nvcr.io/nvidia/pytorch:24.03-py3  # testing with the latest pytorch base image
       options: "--gpus all --ipc=host"
     runs-on: [self-hosted, linux, x64, integration]
     steps:
diff --git a/.github/workflows/pythonapp-gpu.yml b/.github/workflows/pythonapp-gpu.yml
index a6d7981814..f83d52f8e3 100644
--- a/.github/workflows/pythonapp-gpu.yml
+++ b/.github/workflows/pythonapp-gpu.yml
@@ -29,10 +29,6 @@ jobs:
           - "PT210+CUDA121DOCKER"
         include:
           # https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes
-          - environment: PT19+CUDA114DOCKER
-            # 21.10: 1.10.0a0+0aef44c
-            pytorch: "-h"  # we explicitly set pytorch to -h to avoid pip install error
-            base: "nvcr.io/nvidia/pytorch:21.10-py3"
           - environment: PT110+CUDA111
             pytorch: "torch==1.10.2 torchvision==0.11.3 --extra-index-url https://download.pytorch.org/whl/cu111"
             base: "nvcr.io/nvidia/cuda:11.1.1-devel-ubuntu18.04"
@@ -47,6 +43,10 @@ jobs:
             # 23.08: 2.1.0a0+29c30b1
             pytorch: "-h"  # we explicitly set pytorch to -h to avoid pip install error
             base: "nvcr.io/nvidia/pytorch:23.08-py3"
+          - environment: PT210+CUDA121DOCKER
+            # 24.03: 2.3.0a0+40ec155e58.nv24.3
+            pytorch: "-h"  # we explicitly set pytorch to -h to avoid pip install error
+            base: "nvcr.io/nvidia/pytorch:24.03-py3"
     container:
       image: ${{ matrix.base }}
       options: --gpus all --env NVIDIA_DISABLE_REQUIRE=true  # workaround for unsatisfied condition: cuda>=11.6
diff --git a/Dockerfile b/Dockerfile
index 7383837585..d5777104c8 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -11,7 +11,7 @@
 
 # To build with a different base image
 # please run `docker build` using the `--build-arg PYTORCH_IMAGE=...` flag.
-ARG PYTORCH_IMAGE=nvcr.io/nvidia/pytorch:23.08-py3
+ARG PYTORCH_IMAGE=nvcr.io/nvidia/pytorch:24.03-py3
 FROM ${PYTORCH_IMAGE}
 
 LABEL maintainer="monai.contact@gmail.com"

From 195d7ddc3669426e39567d90d083684cd948e388 Mon Sep 17 00:00:00 2001
From: Lucas Robinet <luca.robinet@gmail.com>
Date: Thu, 4 Apr 2024 09:16:17 +0200
Subject: [PATCH 009/183] simplification of the sincos positional encoding in
 patchembedding.py (#7605)

Fixes #7564 .

### Description

As discussed, a small simplification for the creation of sincos
positional encoding where we don't need to use the `torch.no_grad()`
context or copy the tensor with `copy_` from torch which doesn't
preserve the `requires_grad` attribute here.

The changes are simple and are linked to the corresponding comment
#7564, the output is already in float32 so it doesn't seem particularly
necessary to apply the conversion previously done.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Lucas Robinet <robinet.lucas@iuct-oncopole.fr>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/blocks/patchembedding.py | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/monai/networks/blocks/patchembedding.py b/monai/networks/blocks/patchembedding.py
index 44774ce5da..91bd73ebbb 100644
--- a/monai/networks/blocks/patchembedding.py
+++ b/monai/networks/blocks/patchembedding.py
@@ -120,10 +120,7 @@ def __init__(
             for in_size, pa_size in zip(img_size, patch_size):
                 grid_size.append(in_size // pa_size)
 
-            with torch.no_grad():
-                pos_embeddings = build_sincos_position_embedding(grid_size, hidden_size, spatial_dims)
-                self.position_embeddings.data.copy_(pos_embeddings.float())
-                self.position_embeddings.requires_grad = False
+            self.position_embeddings = build_sincos_position_embedding(grid_size, hidden_size, spatial_dims)
         else:
             raise ValueError(f"pos_embed_type {self.pos_embed_type} not supported.")
 

From 625967c483e6ad98c2e41d7d046040c91329cd8c Mon Sep 17 00:00:00 2001
From: Lucas Robinet <robinet.lucas@iuct-oncopole.fr>
Date: Fri, 5 Apr 2024 11:21:24 +0200
Subject: [PATCH 010/183] harmonization and clarification of dice losses
 variants docs and associated tests (#7587)

### Description

This PR aims to clarify and harmonise the code for the DiceLoss variants
in the `monai/losses/dice.py` file. With the `to_onehot_y` `softmax` and
`sigmoid` arguments, I didn't necessarily understand the ValueError that
occurred when I passed a target of size NH[WD]. I had a bit of trouble
reading the documentation and understanding it. I thought that they had
to be the same shape as they are displayed, unlike the number of
dimensions in the input, so I added that.
Besides, in the documentation is written:
```python
"""
raises:
      ValueError: When number of channels for target is neither 1 nor the same as input.

"""
```
Trying to reproduce this, we give an input with a number of channels $N$
and target a number of channels of $M$, with $M \neq N$ and $M > 1$.
```python
loss = DiceCELoss()
input = torch.rand(1, 4, 3, 3)
target = torch.randn(1, 2, 3, 3)
loss(input, target)
>: AssertionError: ground truth has different shape (torch.Size([1, 2, 3, 3])) from input (torch.Size([1, 4, 3, 3]))
```
This error in the Dice is an `AssertionError` and not a `ValueError` as
expected and the explanation can be confusing and doesn't give a clear
idea of the error here. The classes concerned and harmonised are
`DiceFocalLoss`, `DiceCELoss` and `GeneralizedDiceFocalLoss` with the
addition of tests that behave correctly and handle this harmonisation.

Also, feel free to modify or make suggestions regarding the changes made
in the docstring to make them more understandable (in my opinion, but
other readers and users will probably have a different view).

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Lucas Robinet <robinet.lucas@iuct-oncopole.fr>
Signed-off-by: Lucas Robinet <luca.robinet@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/losses/dice.py                      | 42 +++++++++++++++++++----
 tests/test_dice_ce_loss.py                | 18 +++++++---
 tests/test_dice_focal_loss.py             | 14 ++++++--
 tests/test_generalized_dice_focal_loss.py | 14 ++++++--
 4 files changed, 73 insertions(+), 15 deletions(-)

diff --git a/monai/losses/dice.py b/monai/losses/dice.py
index b3c0f57c6e..f1c357d31f 100644
--- a/monai/losses/dice.py
+++ b/monai/losses/dice.py
@@ -778,12 +778,22 @@ def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor:
 
         Raises:
             ValueError: When number of dimensions for input and target are different.
-            ValueError: When number of channels for target is neither 1 nor the same as input.
+            ValueError: When number of channels for target is neither 1 (without one-hot encoding) nor the same as input.
+
+        Returns:
+            torch.Tensor: value of the loss.
 
         """
-        if len(input.shape) != len(target.shape):
+        if input.dim() != target.dim():
             raise ValueError(
                 "the number of dimensions for input and target should be the same, "
+                f"got shape {input.shape} (nb dims: {len(input.shape)}) and {target.shape} (nb dims: {len(target.shape)}). "
+                "if target is not one-hot encoded, please provide a tensor with shape B1H[WD]."
+            )
+
+        if target.shape[1] != 1 and target.shape[1] != input.shape[1]:
+            raise ValueError(
+                "number of channels for target is neither 1 (without one-hot encoding) nor the same as input, "
                 f"got shape {input.shape} and {target.shape}."
             )
 
@@ -899,14 +909,24 @@ def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor:
 
         Raises:
             ValueError: When number of dimensions for input and target are different.
-            ValueError: When number of channels for target is neither 1 nor the same as input.
+            ValueError: When number of channels for target is neither 1 (without one-hot encoding) nor the same as input.
 
+        Returns:
+            torch.Tensor: value of the loss.
         """
-        if len(input.shape) != len(target.shape):
+        if input.dim() != target.dim():
             raise ValueError(
                 "the number of dimensions for input and target should be the same, "
+                f"got shape {input.shape} (nb dims: {len(input.shape)}) and {target.shape} (nb dims: {len(target.shape)}). "
+                "if target is not one-hot encoded, please provide a tensor with shape B1H[WD]."
+            )
+
+        if target.shape[1] != 1 and target.shape[1] != input.shape[1]:
+            raise ValueError(
+                "number of channels for target is neither 1 (without one-hot encoding) nor the same as input, "
                 f"got shape {input.shape} and {target.shape}."
             )
+
         if self.to_onehot_y:
             n_pred_ch = input.shape[1]
             if n_pred_ch == 1:
@@ -1015,15 +1035,23 @@ def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor:
             target (torch.Tensor): the shape should be BNH[WD] or B1H[WD].
 
         Raises:
-            ValueError: When the input and target tensors have different numbers of dimensions, or the target
-                channel isn't either one-hot encoded or categorical with the same shape of the input.
+            ValueError: When number of dimensions for input and target are different.
+            ValueError: When number of channels for target is neither 1 (without one-hot encoding) nor the same as input.
 
         Returns:
             torch.Tensor: value of the loss.
         """
         if input.dim() != target.dim():
             raise ValueError(
-                f"Input - {input.shape} - and target - {target.shape} - must have the same number of dimensions."
+                "the number of dimensions for input and target should be the same, "
+                f"got shape {input.shape} (nb dims: {len(input.shape)}) and {target.shape} (nb dims: {len(target.shape)}). "
+                "if target is not one-hot encoded, please provide a tensor with shape B1H[WD]."
+            )
+
+        if target.shape[1] != 1 and target.shape[1] != input.shape[1]:
+            raise ValueError(
+                "number of channels for target is neither 1 (without one-hot encoding) nor the same as input, "
+                f"got shape {input.shape} and {target.shape}."
             )
 
         gdl_loss = self.generalized_dice(input, target)
diff --git a/tests/test_dice_ce_loss.py b/tests/test_dice_ce_loss.py
index 225618ed2c..97c7ae5050 100644
--- a/tests/test_dice_ce_loss.py
+++ b/tests/test_dice_ce_loss.py
@@ -93,10 +93,20 @@ def test_result(self, input_param, input_data, expected_val):
         result = diceceloss(**input_data)
         np.testing.assert_allclose(result.detach().cpu().numpy(), expected_val, atol=1e-4, rtol=1e-4)
 
-    # def test_ill_shape(self):
-    #     loss = DiceCELoss()
-    #     with self.assertRaisesRegex(ValueError, ""):
-    #         loss(torch.ones((1, 2, 3)), torch.ones((1, 1, 2, 3)))
+    def test_ill_shape(self):
+        loss = DiceCELoss()
+        with self.assertRaises(AssertionError):
+            loss.forward(torch.ones((1, 2, 3)), torch.ones((1, 2, 5)))
+
+    def test_ill_shape2(self):
+        loss = DiceCELoss()
+        with self.assertRaises(ValueError):
+            loss.forward(torch.ones((1, 2, 3)), torch.ones((1, 1, 2, 3)))
+
+    def test_ill_shape3(self):
+        loss = DiceCELoss()
+        with self.assertRaises(ValueError):
+            loss.forward(torch.ones((1, 3, 4, 4)), torch.ones((1, 2, 4, 4)))
 
     # def test_ill_reduction(self):
     #     with self.assertRaisesRegex(ValueError, ""):
diff --git a/tests/test_dice_focal_loss.py b/tests/test_dice_focal_loss.py
index 13899da003..814a174762 100644
--- a/tests/test_dice_focal_loss.py
+++ b/tests/test_dice_focal_loss.py
@@ -69,8 +69,18 @@ def test_result_no_onehot_no_bg(self, size, onehot):
 
     def test_ill_shape(self):
         loss = DiceFocalLoss()
-        with self.assertRaisesRegex(ValueError, ""):
-            loss(torch.ones((1, 2, 3)), torch.ones((1, 1, 2, 3)))
+        with self.assertRaises(AssertionError):
+            loss.forward(torch.ones((1, 2, 3)), torch.ones((1, 2, 5)))
+
+    def test_ill_shape2(self):
+        loss = DiceFocalLoss()
+        with self.assertRaises(ValueError):
+            loss.forward(torch.ones((1, 2, 3)), torch.ones((1, 1, 2, 3)))
+
+    def test_ill_shape3(self):
+        loss = DiceFocalLoss()
+        with self.assertRaises(ValueError):
+            loss.forward(torch.ones((1, 3, 4, 4)), torch.ones((1, 2, 4, 4)))
 
     def test_ill_lambda(self):
         with self.assertRaisesRegex(ValueError, ""):
diff --git a/tests/test_generalized_dice_focal_loss.py b/tests/test_generalized_dice_focal_loss.py
index 8a0a80865e..65252611ca 100644
--- a/tests/test_generalized_dice_focal_loss.py
+++ b/tests/test_generalized_dice_focal_loss.py
@@ -59,8 +59,18 @@ def test_result_no_onehot_no_bg(self):
 
     def test_ill_shape(self):
         loss = GeneralizedDiceFocalLoss()
-        with self.assertRaisesRegex(ValueError, ""):
-            loss(torch.ones((1, 2, 3)), torch.ones((1, 1, 2, 3)))
+        with self.assertRaises(AssertionError):
+            loss.forward(torch.ones((1, 2, 3)), torch.ones((1, 2, 5)))
+
+    def test_ill_shape2(self):
+        loss = GeneralizedDiceFocalLoss()
+        with self.assertRaises(ValueError):
+            loss.forward(torch.ones((1, 2, 3)), torch.ones((1, 1, 2, 3)))
+
+    def test_ill_shape3(self):
+        loss = GeneralizedDiceFocalLoss()
+        with self.assertRaises(ValueError):
+            loss.forward(torch.ones((1, 3, 4, 4)), torch.ones((1, 2, 4, 4)))
 
     def test_ill_lambda(self):
         with self.assertRaisesRegex(ValueError, ""):

From c0b9cc0b00459563196b9ce1d430f5d86406c5e6 Mon Sep 17 00:00:00 2001
From: Lucas Robinet <robinet.lucas@iuct-oncopole.fr>
Date: Fri, 5 Apr 2024 14:56:25 +0200
Subject: [PATCH 011/183] Implementation of intensity clipping transform: bot
 hard and soft approaches (#7535)

Fixes Issue #7512.

### Description

Addition of a transformation allowing values above or below a certain
percentile to be clipped.
Clipping can be hard or soft.
With soft clipping, the function remains derivable and the order of the
values is respected, with smoother corners.

The soft clipping function is based on this medium article
https://medium.com/life-at-hopper/clip-it-clip-it-good-1f1bf711b291

It's important to note that I've chosen to switch from Nones values to
percentiles to take account of the fact that soft clipping can be
one-sided or two-sided.
In fact, providing percentiles of 100 or 0 doesn't change anything in
the case of hard clipping, but it does in the case of soft clipping
because the function is smoothed. Hence the interest in introducing the
possibility of putting None to avoid smoothing the function on one side
or the other.

To implement this we had to define a `softplus` function in
`monai.transforms.utils_pytorch_numpy_unification.py`. One of the
problems is that `np.logaddexp` do not exactly yields same outputs as
`torch.logaddexp`. I've left it as is and lowered the tolerance of the
tests slightly, but it's possible to force the conversion to numpy and
then switch back to torch to ensure better unification between the
frameworks.

I've also added the `soft_clip` function in `monai.transforms.utils.py`
with the associated unit tests to ensure that the transformation works
properly.


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Lucas Robinet <robinet.lucas@iuct-oncopole.fr>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 docs/source/transforms.rst                    |  12 ++
 monai/transforms/__init__.py                  |   4 +
 monai/transforms/intensity/array.py           | 148 ++++++++++++-
 monai/transforms/intensity/dictionary.py      |  35 +++
 monai/transforms/utils.py                     |  37 ++++
 .../utils_pytorch_numpy_unification.py        |  15 ++
 tests/test_clip_intensity_percentiles.py      | 185 ++++++++++++++++
 tests/test_clip_intensity_percentilesd.py     | 203 ++++++++++++++++++
 tests/test_soft_clip.py                       | 125 +++++++++++
 9 files changed, 763 insertions(+), 1 deletion(-)
 create mode 100644 tests/test_clip_intensity_percentiles.py
 create mode 100644 tests/test_clip_intensity_percentilesd.py
 create mode 100644 tests/test_soft_clip.py

diff --git a/docs/source/transforms.rst b/docs/source/transforms.rst
index bd3feb3497..8bd5bfd99f 100644
--- a/docs/source/transforms.rst
+++ b/docs/source/transforms.rst
@@ -309,6 +309,12 @@ Intensity
     :members:
     :special-members: __call__
 
+`ClipIntensityPercentiles`
+""""""""""""""""""""""""""
+.. autoclass:: ClipIntensityPercentiles
+    :members:
+    :special-members: __call__
+
 `RandScaleIntensity`
 """"""""""""""""""""
 .. image:: https://raw.githubusercontent.com/Project-MONAI/DocImages/main/transforms/RandScaleIntensity.png
@@ -1405,6 +1411,12 @@ Intensity (Dict)
     :members:
     :special-members: __call__
 
+`ClipIntensityPercentilesd`
+"""""""""""""""""""""""""""
+.. autoclass:: ClipIntensityPercentilesd
+    :members:
+    :special-members: __call__
+
 `RandScaleIntensityd`
 """""""""""""""""""""
 .. image:: https://raw.githubusercontent.com/Project-MONAI/DocImages/main/transforms/RandScaleIntensityd.png
diff --git a/monai/transforms/__init__.py b/monai/transforms/__init__.py
index 349533fb3e..ab9adb6a99 100644
--- a/monai/transforms/__init__.py
+++ b/monai/transforms/__init__.py
@@ -92,6 +92,7 @@
 from .croppad.functional import crop_func, crop_or_pad_nd, pad_func, pad_nd
 from .intensity.array import (
     AdjustContrast,
+    ClipIntensityPercentiles,
     ComputeHoVerMaps,
     DetectEnvelope,
     ForegroundMask,
@@ -135,6 +136,9 @@
     AdjustContrastd,
     AdjustContrastD,
     AdjustContrastDict,
+    ClipIntensityPercentilesd,
+    ClipIntensityPercentilesD,
+    ClipIntensityPercentilesDict,
     ComputeHoVerMapsd,
     ComputeHoVerMapsD,
     ComputeHoVerMapsDict,
diff --git a/monai/transforms/intensity/array.py b/monai/transforms/intensity/array.py
index 0085050ee3..f656475a36 100644
--- a/monai/transforms/intensity/array.py
+++ b/monai/transforms/intensity/array.py
@@ -30,7 +30,7 @@
 from monai.data.utils import get_random_patch, get_valid_patch_size
 from monai.networks.layers import GaussianFilter, HilbertTransform, MedianFilter, SavitzkyGolayFilter
 from monai.transforms.transform import RandomizableTransform, Transform
-from monai.transforms.utils import Fourier, equalize_hist, is_positive, rescale_array
+from monai.transforms.utils import Fourier, equalize_hist, is_positive, rescale_array, soft_clip
 from monai.transforms.utils_pytorch_numpy_unification import clip, percentile, where
 from monai.utils.enums import TransformBackends
 from monai.utils.misc import ensure_tuple, ensure_tuple_rep, ensure_tuple_size, fall_back_tuple
@@ -54,6 +54,7 @@
     "NormalizeIntensity",
     "ThresholdIntensity",
     "ScaleIntensityRange",
+    "ClipIntensityPercentiles",
     "AdjustContrast",
     "RandAdjustContrast",
     "ScaleIntensityRangePercentiles",
@@ -1007,6 +1008,151 @@ def __call__(self, img: NdarrayOrTensor) -> NdarrayOrTensor:
         return ret
 
 
+class ClipIntensityPercentiles(Transform):
+    """
+    Apply clip based on the intensity distribution of input image.
+    If `sharpness_factor` is provided, the intensity values will be soft clipped according to
+    f(x) = x + (1/sharpness_factor)*softplus(- c(x - minv)) - (1/sharpness_factor)*softplus(c(x - maxv))
+    From https://medium.com/life-at-hopper/clip-it-clip-it-good-1f1bf711b291
+
+    Soft clipping preserves the order of the values and maintains the gradient everywhere.
+    For example:
+
+    .. code-block:: python
+        :emphasize-lines: 11, 22
+
+        image = torch.Tensor(
+            [[[1, 2, 3, 4, 5],
+              [1, 2, 3, 4, 5],
+              [1, 2, 3, 4, 5],
+              [1, 2, 3, 4, 5],
+              [1, 2, 3, 4, 5],
+              [1, 2, 3, 4, 5]]])
+
+        # Hard clipping from lower and upper image intensity percentiles
+        hard_clipper = ClipIntensityPercentiles(30, 70)
+        print(hard_clipper(image))
+        metatensor([[[2., 2., 3., 4., 4.],
+                [2., 2., 3., 4., 4.],
+                [2., 2., 3., 4., 4.],
+                [2., 2., 3., 4., 4.],
+                [2., 2., 3., 4., 4.],
+                [2., 2., 3., 4., 4.]]])
+
+
+        # Soft clipping from lower and upper image intensity percentiles
+        soft_clipper = ClipIntensityPercentiles(30, 70, 10.)
+        print(soft_clipper(image))
+        metatensor([[[2.0000, 2.0693, 3.0000, 3.9307, 4.0000],
+         [2.0000, 2.0693, 3.0000, 3.9307, 4.0000],
+         [2.0000, 2.0693, 3.0000, 3.9307, 4.0000],
+         [2.0000, 2.0693, 3.0000, 3.9307, 4.0000],
+         [2.0000, 2.0693, 3.0000, 3.9307, 4.0000],
+         [2.0000, 2.0693, 3.0000, 3.9307, 4.0000]]])
+
+    See Also:
+
+        - :py:class:`monai.transforms.ScaleIntensityRangePercentiles`
+    """
+
+    backend = [TransformBackends.TORCH, TransformBackends.NUMPY]
+
+    def __init__(
+        self,
+        lower: float | None,
+        upper: float | None,
+        sharpness_factor: float | None = None,
+        channel_wise: bool = False,
+        return_clipping_values: bool = False,
+        dtype: DtypeLike = np.float32,
+    ) -> None:
+        """
+        Args:
+            lower: lower intensity percentile. In the case of hard clipping, None will have the same effect as 0 by
+                not clipping the lowest input values. However, in the case of soft clipping, None and zero will have
+                two different effects: None will not apply clipping to low values, whereas zero will still transform
+                the lower values according to the soft clipping transformation. Please check for more details:
+                https://medium.com/life-at-hopper/clip-it-clip-it-good-1f1bf711b291.
+            upper: upper intensity percentile.  The same as for lower, but this time with the highest values. If we
+                are looking to perform soft clipping, if None then there will be no effect on this side whereas if set
+                to 100, the values will be passed via the corresponding clipping equation.
+            sharpness_factor: if not None, the intensity values will be soft clipped according to
+                f(x) = x + (1/sharpness_factor)*softplus(- c(x - minv)) - (1/sharpness_factor)*softplus(c(x - maxv)).
+                defaults to None.
+            channel_wise: if True, compute intensity percentile and normalize every channel separately.
+                default to False.
+            return_clipping_values: whether to return the calculated percentiles in tensor meta information.
+                If soft clipping and requested percentile is None, return None as the corresponding clipping
+                values in meta information. Clipping values are stored in a list with each element corresponding
+                to a channel if channel_wise is set to True. defaults to False.
+            dtype: output data type, if None, same as input image. defaults to float32.
+        """
+        if lower is None and upper is None:
+            raise ValueError("lower or upper percentiles must be provided")
+        if lower is not None and (lower < 0.0 or lower > 100.0):
+            raise ValueError("Percentiles must be in the range [0, 100]")
+        if upper is not None and (upper < 0.0 or upper > 100.0):
+            raise ValueError("Percentiles must be in the range [0, 100]")
+        if upper is not None and lower is not None and upper < lower:
+            raise ValueError("upper must be greater than or equal to lower")
+        if sharpness_factor is not None and sharpness_factor <= 0:
+            raise ValueError("sharpness_factor must be greater than 0")
+
+        self.lower = lower
+        self.upper = upper
+        self.sharpness_factor = sharpness_factor
+        self.channel_wise = channel_wise
+        if return_clipping_values:
+            self.clipping_values: list[tuple[float | None, float | None]] = []
+        self.return_clipping_values = return_clipping_values
+        self.dtype = dtype
+
+    def _clip(self, img: NdarrayOrTensor) -> NdarrayOrTensor:
+        if self.sharpness_factor is not None:
+            lower_percentile = percentile(img, self.lower) if self.lower is not None else None
+            upper_percentile = percentile(img, self.upper) if self.upper is not None else None
+            img = soft_clip(img, self.sharpness_factor, lower_percentile, upper_percentile, self.dtype)
+        else:
+            lower_percentile = percentile(img, self.lower) if self.lower is not None else percentile(img, 0)
+            upper_percentile = percentile(img, self.upper) if self.upper is not None else percentile(img, 100)
+            img = clip(img, lower_percentile, upper_percentile)
+
+        if self.return_clipping_values:
+            self.clipping_values.append(
+                (
+                    (
+                        lower_percentile
+                        if lower_percentile is None
+                        else lower_percentile.item() if hasattr(lower_percentile, "item") else lower_percentile
+                    ),
+                    (
+                        upper_percentile
+                        if upper_percentile is None
+                        else upper_percentile.item() if hasattr(upper_percentile, "item") else upper_percentile
+                    ),
+                )
+            )
+        img = convert_to_tensor(img, track_meta=False)
+        return img
+
+    def __call__(self, img: NdarrayOrTensor) -> NdarrayOrTensor:
+        """
+        Apply the transform to `img`.
+        """
+        img = convert_to_tensor(img, track_meta=get_track_meta())
+        img_t = convert_to_tensor(img, track_meta=False)
+        if self.channel_wise:
+            img_t = torch.stack([self._clip(img=d) for d in img_t])  # type: ignore
+        else:
+            img_t = self._clip(img=img_t)
+
+        img = convert_to_dst_type(img_t, dst=img)[0]
+        if self.return_clipping_values:
+            img.meta["clipping_values"] = self.clipping_values  # type: ignore
+
+        return img
+
+
 class AdjustContrast(Transform):
     """
     Changes image intensity with gamma transform. Each pixel/voxel intensity is updated as::
diff --git a/monai/transforms/intensity/dictionary.py b/monai/transforms/intensity/dictionary.py
index 5b911904b0..5dbac485fe 100644
--- a/monai/transforms/intensity/dictionary.py
+++ b/monai/transforms/intensity/dictionary.py
@@ -26,6 +26,7 @@
 from monai.data.meta_obj import get_track_meta
 from monai.transforms.intensity.array import (
     AdjustContrast,
+    ClipIntensityPercentiles,
     ComputeHoVerMaps,
     ForegroundMask,
     GaussianSharpen,
@@ -77,6 +78,7 @@
     "NormalizeIntensityd",
     "ThresholdIntensityd",
     "ScaleIntensityRanged",
+    "ClipIntensityPercentilesd",
     "AdjustContrastd",
     "RandAdjustContrastd",
     "ScaleIntensityRangePercentilesd",
@@ -122,6 +124,8 @@
     "ThresholdIntensityDict",
     "ScaleIntensityRangeD",
     "ScaleIntensityRangeDict",
+    "ClipIntensityPercentilesD",
+    "ClipIntensityPercentilesDict",
     "AdjustContrastD",
     "AdjustContrastDict",
     "RandAdjustContrastD",
@@ -886,6 +890,36 @@ def __call__(self, data: Mapping[Hashable, NdarrayOrTensor]) -> dict[Hashable, N
         return d
 
 
+class ClipIntensityPercentilesd(MapTransform):
+    """
+    Dictionary-based wrapper of :py:class:`monai.transforms.ClipIntensityPercentiles`.
+    Clip the intensity values of input image to a specific range based on the intensity distribution of the input.
+    If `sharpness_factor` is provided, the intensity values will be soft clipped according to
+    f(x) = x + (1/sharpness_factor) * softplus(- c(x - minv)) - (1/sharpness_factor)*softplus(c(x - maxv))
+    """
+
+    def __init__(
+        self,
+        keys: KeysCollection,
+        lower: float | None,
+        upper: float | None,
+        sharpness_factor: float | None = None,
+        channel_wise: bool = False,
+        dtype: DtypeLike = np.float32,
+        allow_missing_keys: bool = False,
+    ) -> None:
+        super().__init__(keys, allow_missing_keys)
+        self.scaler = ClipIntensityPercentiles(
+            lower=lower, upper=upper, sharpness_factor=sharpness_factor, channel_wise=channel_wise, dtype=dtype
+        )
+
+    def __call__(self, data: dict) -> dict:
+        d = dict(data)
+        for key in self.key_iterator(d):
+            d[key] = self.scaler(d[key])
+        return d
+
+
 class AdjustContrastd(MapTransform):
     """
     Dictionary-based wrapper of :py:class:`monai.transforms.AdjustContrast`.
@@ -1929,6 +1963,7 @@ def __call__(self, data: Mapping[Hashable, NdarrayOrTensor]) -> dict[Hashable, N
 NormalizeIntensityD = NormalizeIntensityDict = NormalizeIntensityd
 ThresholdIntensityD = ThresholdIntensityDict = ThresholdIntensityd
 ScaleIntensityRangeD = ScaleIntensityRangeDict = ScaleIntensityRanged
+ClipIntensityPercentilesD = ClipIntensityPercentilesDict = ClipIntensityPercentilesd
 AdjustContrastD = AdjustContrastDict = AdjustContrastd
 RandAdjustContrastD = RandAdjustContrastDict = RandAdjustContrastd
 ScaleIntensityRangePercentilesD = ScaleIntensityRangePercentilesDict = ScaleIntensityRangePercentilesd
diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index 455b0cfb7d..14f35e1219 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -38,6 +38,7 @@
     nonzero,
     ravel,
     searchsorted,
+    softplus,
     unique,
     unravel_index,
     where,
@@ -131,9 +132,45 @@
     "resolves_modes",
     "has_status_keys",
     "distance_transform_edt",
+    "soft_clip",
 ]
 
 
+def soft_clip(
+    arr: NdarrayOrTensor,
+    sharpness_factor: float = 1.0,
+    minv: NdarrayOrTensor | float | int | None = None,
+    maxv: NdarrayOrTensor | float | int | None = None,
+    dtype: DtypeLike | torch.dtype = np.float32,
+) -> NdarrayOrTensor:
+    """
+    Apply soft clip to the input array or tensor.
+    The intensity values will be soft clipped according to
+    f(x) = x + (1/sharpness_factor)*softplus(- c(x - minv)) - (1/sharpness_factor)*softplus(c(x - maxv))
+    From https://medium.com/life-at-hopper/clip-it-clip-it-good-1f1bf711b291
+
+    To perform one-sided clipping, set either minv or maxv to None.
+    Args:
+        arr: input array to clip.
+        sharpness_factor: the sharpness of the soft clip function, default to 1.
+        minv: minimum value of target clipped array.
+        maxv: maximum value of target clipped array.
+        dtype: if not None, convert input array to dtype before computation.
+
+    """
+
+    if dtype is not None:
+        arr, *_ = convert_data_type(arr, dtype=dtype)
+
+    v = arr
+    if minv is not None:
+        v = v + softplus(-sharpness_factor * (arr - minv)) / sharpness_factor
+    if maxv is not None:
+        v = v - softplus(sharpness_factor * (arr - maxv)) / sharpness_factor
+
+    return v
+
+
 def rand_choice(prob: float = 0.5) -> bool:
     """
     Returns True if a randomly chosen number is less than or equal to `prob`, by default this is a 50/50 chance.
diff --git a/monai/transforms/utils_pytorch_numpy_unification.py b/monai/transforms/utils_pytorch_numpy_unification.py
index 0774d50314..020d99af16 100644
--- a/monai/transforms/utils_pytorch_numpy_unification.py
+++ b/monai/transforms/utils_pytorch_numpy_unification.py
@@ -52,9 +52,24 @@
     "median",
     "mean",
     "std",
+    "softplus",
 ]
 
 
+def softplus(x: NdarrayOrTensor) -> NdarrayOrTensor:
+    """stable softplus through `np.logaddexp` with equivalent implementation for torch.
+
+    Args:
+        x: array/tensor.
+
+    Returns:
+        Softplus of the input.
+    """
+    if isinstance(x, np.ndarray):
+        return np.logaddexp(np.zeros_like(x), x)
+    return torch.logaddexp(torch.zeros_like(x), x)
+
+
 def allclose(a: NdarrayTensor, b: NdarrayOrTensor, rtol=1e-5, atol=1e-8, equal_nan=False) -> bool:
     """`np.allclose` with equivalent implementation for torch."""
     b, *_ = convert_to_dst_type(b, a, wrap_sequence=True)
diff --git a/tests/test_clip_intensity_percentiles.py b/tests/test_clip_intensity_percentiles.py
new file mode 100644
index 0000000000..f5fe07a323
--- /dev/null
+++ b/tests/test_clip_intensity_percentiles.py
@@ -0,0 +1,185 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from __future__ import annotations
+
+import unittest
+
+import numpy as np
+from parameterized import parameterized
+
+from monai.transforms import ClipIntensityPercentiles
+from monai.transforms.utils import soft_clip
+from monai.transforms.utils_pytorch_numpy_unification import clip
+from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
+
+
+class TestClipIntensityPercentiles2D(NumpyImageTestCase2D):
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_two_sided(self, p):
+        hard_clipper = ClipIntensityPercentiles(upper=95, lower=5)
+        im = p(self.imt)
+        result = hard_clipper(im)
+        lower, upper = np.percentile(self.imt, (5, 95))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_one_sided_high(self, p):
+        hard_clipper = ClipIntensityPercentiles(upper=95, lower=None)
+        im = p(self.imt)
+        result = hard_clipper(im)
+        lower, upper = np.percentile(self.imt, (0, 95))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_one_sided_low(self, p):
+        hard_clipper = ClipIntensityPercentiles(upper=None, lower=5)
+        im = p(self.imt)
+        result = hard_clipper(im)
+        lower, upper = np.percentile(self.imt, (5, 100))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_two_sided(self, p):
+        soft_clipper = ClipIntensityPercentiles(upper=95, lower=5, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper(im)
+        lower, upper = np.percentile(self.imt, (5, 95))
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=upper)
+        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_one_sided_high(self, p):
+        soft_clipper = ClipIntensityPercentiles(upper=95, lower=None, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper(im)
+        upper = np.percentile(self.imt, 95)
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=None, maxv=upper)
+        # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result, p(expected), type_test="tensor", rtol=5e-5, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_one_sided_low(self, p):
+        soft_clipper = ClipIntensityPercentiles(upper=None, lower=5, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper(im)
+        lower = np.percentile(self.imt, 5)
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=None)
+        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_channel_wise(self, p):
+        clipper = ClipIntensityPercentiles(upper=95, lower=5, channel_wise=True)
+        im = p(self.imt)
+        result = clipper(im)
+        for i, c in enumerate(self.imt):
+            lower, upper = np.percentile(c, (5, 95))
+            expected = clip(c, lower, upper)
+            assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    def test_ill_sharpness_factor(self):
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentiles(upper=95, lower=5, sharpness_factor=0.0)
+
+    def test_ill_lower_percentile(self):
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentiles(upper=None, lower=-1)
+
+    def test_ill_upper_percentile(self):
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentiles(upper=101, lower=None)
+
+    def test_ill_percentiles(self):
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentiles(upper=95, lower=96)
+
+    def test_ill_both_none(self):
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentiles(upper=None, lower=None)
+
+
+class TestClipIntensityPercentiles3D(NumpyImageTestCase3D):
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_two_sided(self, p):
+        hard_clipper = ClipIntensityPercentiles(upper=95, lower=5)
+        im = p(self.imt)
+        result = hard_clipper(im)
+        lower, upper = np.percentile(self.imt, (5, 95))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_one_sided_high(self, p):
+        hard_clipper = ClipIntensityPercentiles(upper=95, lower=None)
+        im = p(self.imt)
+        result = hard_clipper(im)
+        lower, upper = np.percentile(self.imt, (0, 95))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_one_sided_low(self, p):
+        hard_clipper = ClipIntensityPercentiles(upper=None, lower=5)
+        im = p(self.imt)
+        result = hard_clipper(im)
+        lower, upper = np.percentile(self.imt, (5, 100))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_two_sided(self, p):
+        soft_clipper = ClipIntensityPercentiles(upper=95, lower=5, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper(im)
+        lower, upper = np.percentile(self.imt, (5, 95))
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=upper)
+        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_one_sided_high(self, p):
+        soft_clipper = ClipIntensityPercentiles(upper=95, lower=None, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper(im)
+        upper = np.percentile(self.imt, 95)
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=None, maxv=upper)
+        # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result, p(expected), type_test="tensor", rtol=5e-5, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_one_sided_low(self, p):
+        soft_clipper = ClipIntensityPercentiles(upper=None, lower=5, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper(im)
+        lower = np.percentile(self.imt, 5)
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=None)
+        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_channel_wise(self, p):
+        clipper = ClipIntensityPercentiles(upper=95, lower=5, channel_wise=True)
+        im = p(self.imt)
+        result = clipper(im)
+        for i, c in enumerate(self.imt):
+            lower, upper = np.percentile(c, (5, 95))
+            expected = clip(c, lower, upper)
+            assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_clip_intensity_percentilesd.py b/tests/test_clip_intensity_percentilesd.py
new file mode 100644
index 0000000000..193fa8b487
--- /dev/null
+++ b/tests/test_clip_intensity_percentilesd.py
@@ -0,0 +1,203 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import numpy as np
+from parameterized import parameterized
+
+from monai.transforms import ClipIntensityPercentilesd
+from monai.transforms.utils import soft_clip
+from monai.transforms.utils_pytorch_numpy_unification import clip
+from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
+
+
+class TestClipIntensityPercentilesd2D(NumpyImageTestCase2D):
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_two_sided(self, p):
+        key = "img"
+        hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5)
+        im = p(self.imt)
+        result = hard_clipper({key: im})
+        lower, upper = np.percentile(self.imt, (5, 95))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_one_sided_high(self, p):
+        key = "img"
+        hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None)
+        im = p(self.imt)
+        result = hard_clipper({key: im})
+        lower, upper = np.percentile(self.imt, (0, 95))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_one_sided_low(self, p):
+        key = "img"
+        hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5)
+        im = p(self.imt)
+        result = hard_clipper({key: im})
+        lower, upper = np.percentile(self.imt, (5, 100))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_two_sided(self, p):
+        key = "img"
+        soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper({key: im})
+        lower, upper = np.percentile(self.imt, (5, 95))
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=upper)
+        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_one_sided_high(self, p):
+        key = "img"
+        soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper({key: im})
+        upper = np.percentile(self.imt, 95)
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=None, maxv=upper)
+        # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=5e-5, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_one_sided_low(self, p):
+        key = "img"
+        soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper({key: im})
+        lower = np.percentile(self.imt, 5)
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=None)
+        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_channel_wise(self, p):
+        key = "img"
+        clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, channel_wise=True)
+        im = p(self.imt)
+        result = clipper({key: im})
+        for i, c in enumerate(self.imt):
+            lower, upper = np.percentile(c, (5, 95))
+            expected = clip(c, lower, upper)
+            assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    def test_ill_sharpness_factor(self):
+        key = "img"
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, sharpness_factor=0.0)
+
+    def test_ill_lower_percentile(self):
+        key = "img"
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentilesd(keys=[key], upper=None, lower=-1)
+
+    def test_ill_upper_percentile(self):
+        key = "img"
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentilesd(keys=[key], upper=101, lower=None)
+
+    def test_ill_percentiles(self):
+        key = "img"
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentilesd(keys=[key], upper=95, lower=96)
+
+    def test_ill_both_none(self):
+        key = "img"
+        with self.assertRaises(ValueError):
+            ClipIntensityPercentilesd(keys=[key], upper=None, lower=None)
+
+
+class TestClipIntensityPercentilesd3D(NumpyImageTestCase3D):
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_two_sided(self, p):
+        key = "img"
+        hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5)
+        im = p(self.imt)
+        result = hard_clipper({key: im})
+        lower, upper = np.percentile(self.imt, (5, 95))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_one_sided_high(self, p):
+        key = "img"
+        hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None)
+        im = p(self.imt)
+        result = hard_clipper({key: im})
+        lower, upper = np.percentile(self.imt, (0, 95))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_hard_clipping_one_sided_low(self, p):
+        key = "img"
+        hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5)
+        im = p(self.imt)
+        result = hard_clipper({key: im})
+        lower, upper = np.percentile(self.imt, (5, 100))
+        expected = clip(self.imt, lower, upper)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_two_sided(self, p):
+        key = "img"
+        soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper({key: im})
+        lower, upper = np.percentile(self.imt, (5, 95))
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=upper)
+        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_one_sided_high(self, p):
+        key = "img"
+        soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper({key: im})
+        upper = np.percentile(self.imt, 95)
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=None, maxv=upper)
+        # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=5e-5, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_soft_clipping_one_sided_low(self, p):
+        key = "img"
+        soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5, sharpness_factor=1.0)
+        im = p(self.imt)
+        result = soft_clipper({key: im})
+        lower = np.percentile(self.imt, 5)
+        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=None)
+        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
+
+    @parameterized.expand([[p] for p in TEST_NDARRAYS])
+    def test_channel_wise(self, p):
+        key = "img"
+        clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, channel_wise=True)
+        im = p(self.imt)
+        result = clipper({key: im})
+        for i, c in enumerate(self.imt):
+            lower, upper = np.percentile(c, (5, 95))
+            expected = clip(c, lower, upper)
+            assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_soft_clip.py b/tests/test_soft_clip.py
new file mode 100644
index 0000000000..de5122e982
--- /dev/null
+++ b/tests/test_soft_clip.py
@@ -0,0 +1,125 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import numpy as np
+import torch
+from parameterized import parameterized
+
+from monai.transforms.utils import soft_clip
+
+TEST_CASES = [
+    [
+        {"minv": 2, "maxv": 8, "sharpness_factor": 10},
+        {
+            "input": torch.arange(10).float(),
+            "clipped": torch.tensor([2.0000, 2.0000, 2.0693, 3.0000, 4.0000, 5.0000, 6.0000, 7.0000, 7.9307, 8.0000]),
+        },
+    ],
+    [
+        {"minv": 2, "maxv": None, "sharpness_factor": 10},
+        {
+            "input": torch.arange(10).float(),
+            "clipped": torch.tensor([2.0000, 2.0000, 2.0693, 3.0000, 4.0000, 5.0000, 6.0000, 7.0000, 8.0000, 9.0000]),
+        },
+    ],
+    [
+        {"minv": None, "maxv": 7, "sharpness_factor": 10},
+        {
+            "input": torch.arange(10).float(),
+            "clipped": torch.tensor([0.0000, 1.0000, 2.0000, 3.0000, 4.0000, 5.0000, 6.0000, 6.9307, 7.0000, 7.0000]),
+        },
+    ],
+    [
+        {"minv": 2, "maxv": 8, "sharpness_factor": 1.0},
+        {
+            "input": torch.arange(10).float(),
+            "clipped": torch.tensor([2.1266, 2.3124, 2.6907, 3.3065, 4.1088, 5.0000, 5.8912, 6.6935, 7.3093, 7.6877]),
+        },
+    ],
+    [
+        {"minv": 2, "maxv": 8, "sharpness_factor": 3.0},
+        {
+            "input": torch.arange(10).float(),
+            "clipped": torch.tensor([2.0008, 2.0162, 2.2310, 3.0162, 4.0008, 5.0000, 5.9992, 6.9838, 7.7690, 7.9838]),
+        },
+    ],
+    [
+        {"minv": 2, "maxv": 8, "sharpness_factor": 5.0},
+        {
+            "input": torch.arange(10).float(),
+            "clipped": torch.tensor([2.0000, 2.0013, 2.1386, 3.0013, 4.0000, 5.0000, 6.0000, 6.9987, 7.8614, 7.9987]),
+        },
+    ],
+    [
+        {"minv": 2, "maxv": 8, "sharpness_factor": 10},
+        {
+            "input": np.arange(10).astype(np.float32),
+            "clipped": np.array([2.0000, 2.0000, 2.0693, 3.0000, 4.0000, 5.0000, 6.0000, 7.0000, 7.9307, 8.0000]),
+        },
+    ],
+    [
+        {"minv": 2, "maxv": None, "sharpness_factor": 10},
+        {
+            "input": np.arange(10).astype(float),
+            "clipped": np.array([2.0000, 2.0000, 2.0693, 3.0000, 4.0000, 5.0000, 6.0000, 7.0000, 8.0000, 9.0000]),
+        },
+    ],
+    [
+        {"minv": None, "maxv": 7, "sharpness_factor": 10},
+        {
+            "input": np.arange(10).astype(float),
+            "clipped": np.array([0.0000, 1.0000, 2.0000, 3.0000, 4.0000, 5.0000, 6.0000, 6.9307, 7.0000, 7.0000]),
+        },
+    ],
+    [
+        {"minv": 2, "maxv": 8, "sharpness_factor": 1.0},
+        {
+            "input": np.arange(10).astype(float),
+            "clipped": np.array([2.1266, 2.3124, 2.6907, 3.3065, 4.1088, 5.0000, 5.8912, 6.6935, 7.3093, 7.6877]),
+        },
+    ],
+    [
+        {"minv": 2, "maxv": 8, "sharpness_factor": 3.0},
+        {
+            "input": np.arange(10).astype(float),
+            "clipped": np.array([2.0008, 2.0162, 2.2310, 3.0162, 4.0008, 5.0000, 5.9992, 6.9838, 7.7690, 7.9838]),
+        },
+    ],
+    [
+        {"minv": 2, "maxv": 8, "sharpness_factor": 5.0},
+        {
+            "input": np.arange(10).astype(float),
+            "clipped": np.array([2.0000, 2.0013, 2.1386, 3.0013, 4.0000, 5.0000, 6.0000, 6.9987, 7.8614, 7.9987]),
+        },
+    ],
+]
+
+
+class TestSoftClip(unittest.TestCase):
+
+    @parameterized.expand(TEST_CASES)
+    def test_result(self, input_param, input_data):
+        outputs = soft_clip(input_data["input"], **input_param)
+        expected_val = input_data["clipped"]
+        if isinstance(outputs, torch.Tensor):
+            np.testing.assert_allclose(
+                outputs.detach().cpu().numpy(), expected_val.detach().cpu().numpy(), atol=1e-4, rtol=1e-4
+            )
+        else:
+            np.testing.assert_allclose(outputs, expected_val, atol=1e-4, rtol=1e-4)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 87152d106557bdc37de0ee923be4e9a25c4c003c Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 8 Apr 2024 12:51:23 +0800
Subject: [PATCH 012/183] Fix typo in `SSIMMetric` (#7612)

Fixes #7610


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/metrics/regression.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/metrics/regression.py b/monai/metrics/regression.py
index 9d29654ee3..4c8b8aa71b 100644
--- a/monai/metrics/regression.py
+++ b/monai/metrics/regression.py
@@ -303,7 +303,7 @@ def _compute_metric(self, y_pred: torch.Tensor, y: torch.Tensor) -> torch.Tensor
 
         if self.spatial_dims == 3 and dims != 5:
             raise ValueError(
-                f"y_pred should have 4 dimensions (batch, channel, height, width, depth) when using {self.spatial_dims}"
+                f"y_pred should have 5 dimensions (batch, channel, height, width, depth) when using {self.spatial_dims}"
                 f" spatial dimensions, got {dims}."
             )
 

From e9a5bfe4693ce490c75616a97eb0812e1828539b Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Wed, 10 Apr 2024 04:03:54 +0100
Subject: [PATCH 013/183] auto updates (#7614)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 tests/test_clip_intensity_percentilesd.py | 2 ++
 tests/test_regularization.py              | 3 +++
 2 files changed, 5 insertions(+)

diff --git a/tests/test_clip_intensity_percentilesd.py b/tests/test_clip_intensity_percentilesd.py
index 193fa8b487..7e00ef09de 100644
--- a/tests/test_clip_intensity_percentilesd.py
+++ b/tests/test_clip_intensity_percentilesd.py
@@ -23,6 +23,7 @@
 
 
 class TestClipIntensityPercentilesd2D(NumpyImageTestCase2D):
+
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_two_sided(self, p):
         key = "img"
@@ -124,6 +125,7 @@ def test_ill_both_none(self):
 
 
 class TestClipIntensityPercentilesd3D(NumpyImageTestCase3D):
+
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_two_sided(self, p):
         key = "img"
diff --git a/tests/test_regularization.py b/tests/test_regularization.py
index c6f727cb54..4df60b9808 100644
--- a/tests/test_regularization.py
+++ b/tests/test_regularization.py
@@ -20,6 +20,7 @@
 
 
 class TestMixup(unittest.TestCase):
+
     def setUp(self) -> None:
         set_determinism(seed=0)
 
@@ -59,6 +60,7 @@ def test_mixupd(self):
 
 
 class TestCutMix(unittest.TestCase):
+
     def setUp(self) -> None:
         set_determinism(seed=0)
 
@@ -89,6 +91,7 @@ def test_cutmixd(self):
 
 
 class TestCutOut(unittest.TestCase):
+
     def setUp(self) -> None:
         set_determinism(seed=0)
 

From 54a6991c7e319c30751d017da670391eb8551179 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 11 Apr 2024 18:26:50 +0800
Subject: [PATCH 014/183] Fix test error in `test_soft_clipping_one_sided_high`
 (#7624)

Fixes #7616


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 tests/test_clip_intensity_percentiles.py  | 60 +++++++++++------------
 tests/test_clip_intensity_percentilesd.py | 60 +++++++++++------------
 2 files changed, 60 insertions(+), 60 deletions(-)

diff --git a/tests/test_clip_intensity_percentiles.py b/tests/test_clip_intensity_percentiles.py
index f5fe07a323..82471e25ce 100644
--- a/tests/test_clip_intensity_percentiles.py
+++ b/tests/test_clip_intensity_percentiles.py
@@ -12,12 +12,12 @@
 
 import unittest
 
-import numpy as np
+import torch
 from parameterized import parameterized
 
 from monai.transforms import ClipIntensityPercentiles
 from monai.transforms.utils import soft_clip
-from monai.transforms.utils_pytorch_numpy_unification import clip
+from monai.transforms.utils_pytorch_numpy_unification import clip, percentile
 from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
 
 
@@ -28,8 +28,8 @@ def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=5)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = np.percentile(self.imt, (5, 95))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (5, 95))
+        expected = clip(im, lower, upper)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -37,8 +37,8 @@ def test_hard_clipping_one_sided_high(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=None)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = np.percentile(self.imt, (0, 95))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (0, 95))
+        expected = clip(im, lower, upper)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -46,8 +46,8 @@ def test_hard_clipping_one_sided_low(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=None, lower=5)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = np.percentile(self.imt, (5, 100))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (5, 100))
+        expected = clip(im, lower, upper)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -55,8 +55,8 @@ def test_soft_clipping_two_sided(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=95, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        lower, upper = np.percentile(self.imt, (5, 95))
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=upper)
+        lower, upper = percentile(im, (5, 95))
+        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
 
@@ -65,8 +65,8 @@ def test_soft_clipping_one_sided_high(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=95, lower=None, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        upper = np.percentile(self.imt, 95)
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=None, maxv=upper)
+        upper = percentile(im, 95)
+        expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
         # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=5e-5, atol=0)
 
@@ -75,8 +75,8 @@ def test_soft_clipping_one_sided_low(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=None, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        lower = np.percentile(self.imt, 5)
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=None)
+        lower = percentile(im, 5)
+        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
 
@@ -85,8 +85,8 @@ def test_channel_wise(self, p):
         clipper = ClipIntensityPercentiles(upper=95, lower=5, channel_wise=True)
         im = p(self.imt)
         result = clipper(im)
-        for i, c in enumerate(self.imt):
-            lower, upper = np.percentile(c, (5, 95))
+        for i, c in enumerate(im):
+            lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
             assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
@@ -118,8 +118,8 @@ def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=5)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = np.percentile(self.imt, (5, 95))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (5, 95))
+        expected = clip(im, lower, upper)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -127,8 +127,8 @@ def test_hard_clipping_one_sided_high(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=None)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = np.percentile(self.imt, (0, 95))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (0, 95))
+        expected = clip(im, lower, upper)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -136,8 +136,8 @@ def test_hard_clipping_one_sided_low(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=None, lower=5)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = np.percentile(self.imt, (5, 100))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (5, 100))
+        expected = clip(im, lower, upper)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -145,8 +145,8 @@ def test_soft_clipping_two_sided(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=95, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        lower, upper = np.percentile(self.imt, (5, 95))
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=upper)
+        lower, upper = percentile(im, (5, 95))
+        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
 
@@ -155,8 +155,8 @@ def test_soft_clipping_one_sided_high(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=95, lower=None, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        upper = np.percentile(self.imt, 95)
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=None, maxv=upper)
+        upper = percentile(im, 95)
+        expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
         # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=5e-5, atol=0)
 
@@ -165,8 +165,8 @@ def test_soft_clipping_one_sided_low(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=None, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        lower = np.percentile(self.imt, 5)
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=None)
+        lower = percentile(im, 5)
+        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
 
@@ -175,8 +175,8 @@ def test_channel_wise(self, p):
         clipper = ClipIntensityPercentiles(upper=95, lower=5, channel_wise=True)
         im = p(self.imt)
         result = clipper(im)
-        for i, c in enumerate(self.imt):
-            lower, upper = np.percentile(c, (5, 95))
+        for i, c in enumerate(im):
+            lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
             assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
diff --git a/tests/test_clip_intensity_percentilesd.py b/tests/test_clip_intensity_percentilesd.py
index 7e00ef09de..2b49383182 100644
--- a/tests/test_clip_intensity_percentilesd.py
+++ b/tests/test_clip_intensity_percentilesd.py
@@ -13,12 +13,12 @@
 
 import unittest
 
-import numpy as np
+import torch
 from parameterized import parameterized
 
 from monai.transforms import ClipIntensityPercentilesd
 from monai.transforms.utils import soft_clip
-from monai.transforms.utils_pytorch_numpy_unification import clip
+from monai.transforms.utils_pytorch_numpy_unification import clip, percentile
 from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
 
 
@@ -30,8 +30,8 @@ def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = np.percentile(self.imt, (5, 95))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (5, 95))
+        expected = clip(im, lower, upper)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -40,8 +40,8 @@ def test_hard_clipping_one_sided_high(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = np.percentile(self.imt, (0, 95))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (0, 95))
+        expected = clip(im, lower, upper)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -50,8 +50,8 @@ def test_hard_clipping_one_sided_low(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = np.percentile(self.imt, (5, 100))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (5, 100))
+        expected = clip(im, lower, upper)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -60,8 +60,8 @@ def test_soft_clipping_two_sided(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        lower, upper = np.percentile(self.imt, (5, 95))
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=upper)
+        lower, upper = percentile(im, (5, 95))
+        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
 
@@ -71,8 +71,8 @@ def test_soft_clipping_one_sided_high(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        upper = np.percentile(self.imt, 95)
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=None, maxv=upper)
+        upper = percentile(im, 95)
+        expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
         # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=5e-5, atol=0)
 
@@ -82,8 +82,8 @@ def test_soft_clipping_one_sided_low(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        lower = np.percentile(self.imt, 5)
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=None)
+        lower = percentile(im, 5)
+        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
 
@@ -93,8 +93,8 @@ def test_channel_wise(self, p):
         clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, channel_wise=True)
         im = p(self.imt)
         result = clipper({key: im})
-        for i, c in enumerate(self.imt):
-            lower, upper = np.percentile(c, (5, 95))
+        for i, c in enumerate(im):
+            lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
             assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
@@ -132,8 +132,8 @@ def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = np.percentile(self.imt, (5, 95))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (5, 95))
+        expected = clip(im, lower, upper)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -142,8 +142,8 @@ def test_hard_clipping_one_sided_high(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = np.percentile(self.imt, (0, 95))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (0, 95))
+        expected = clip(im, lower, upper)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -152,8 +152,8 @@ def test_hard_clipping_one_sided_low(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = np.percentile(self.imt, (5, 100))
-        expected = clip(self.imt, lower, upper)
+        lower, upper = percentile(im, (5, 100))
+        expected = clip(im, lower, upper)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -162,8 +162,8 @@ def test_soft_clipping_two_sided(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        lower, upper = np.percentile(self.imt, (5, 95))
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=upper)
+        lower, upper = percentile(im, (5, 95))
+        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
 
@@ -173,8 +173,8 @@ def test_soft_clipping_one_sided_high(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        upper = np.percentile(self.imt, 95)
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=None, maxv=upper)
+        upper = percentile(im, 95)
+        expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
         # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=5e-5, atol=0)
 
@@ -184,8 +184,8 @@ def test_soft_clipping_one_sided_low(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        lower = np.percentile(self.imt, 5)
-        expected = soft_clip(self.imt, sharpness_factor=1.0, minv=lower, maxv=None)
+        lower = percentile(im, 5)
+        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
 
@@ -195,8 +195,8 @@ def test_channel_wise(self, p):
         clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, channel_wise=True)
         im = p(self.imt)
         result = clipper({key: im})
-        for i, c in enumerate(self.imt):
-            lower, upper = np.percentile(c, (5, 95))
+        for i, c in enumerate(im):
+            lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
             assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 

From 3856c456bd8140c70decaeb89c4f58d1dfabb74b Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 11 Apr 2024 21:19:18 +0800
Subject: [PATCH 015/183] Fix deprecated warning in ruff (#7625)

Fixes #7623


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .pre-commit-config.yaml | 2 +-
 pyproject.toml          | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 14b41bbeb8..08b75b1e97 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -27,7 +27,7 @@ repos:
       - id: end-of-file-fixer
       - id: mixed-line-ending
   - repo: https://github.com/charliermarsh/ruff-pre-commit
-    rev: v0.0.261
+    rev: v0.3.5
     hooks:
     -   id: ruff
         args:
diff --git a/pyproject.toml b/pyproject.toml
index cd8a510b04..50d0b09672 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -38,8 +38,8 @@ exclude = "monai/bundle/__main__.py"
 
 [tool.ruff]
 line-length = 133
-ignore-init-module-imports = true
-ignore = ["F401", "E741"]
+lint.ignore-init-module-imports = true
+lint.ignore = ["F401", "E741"]
 
 [tool.pytype]
 # Space-separated list of files or directories to exclude.

From da3ecdd52e0ba68c765eb5010ec202811c3cb1ec Mon Sep 17 00:00:00 2001
From: binliunls <107988372+binliunls@users.noreply.github.com>
Date: Fri, 12 Apr 2024 15:11:52 +0800
Subject: [PATCH 016/183] 7601 fix mlflow artifacts (#7604)

Fixes #7601 .

### Description

Support not save artifacts.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: binliu <binliu@nvidia.com>
Co-authored-by: Nic Ma <nma@nvidia.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/bundle/scripts.py     | 17 +++++++++++++----
 monai/bundle/utils.py       |  6 +++---
 monai/bundle/workflows.py   | 26 ++++++++++++++++----------
 tests/test_fl_monai_algo.py |  6 +++---
 4 files changed, 35 insertions(+), 20 deletions(-)

diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index 2565a3cf64..598d938cbd 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -778,10 +778,19 @@ def run(
             https://docs.python.org/3/library/logging.config.html#logging.config.fileConfig.
             Default to None.
         tracking: if not None, enable the experiment tracking at runtime with optionally configurable and extensible.
-            if "mlflow", will add `MLFlowHandler` to the parsed bundle with default tracking settings,
-            if other string, treat it as file path to load the tracking settings.
-            if `dict`, treat it as tracking settings.
-            will patch the target config content with `tracking handlers` and the top-level items of `configs`.
+            If "mlflow", will add `MLFlowHandler` to the parsed bundle with default tracking settings where a set of
+            common parameters shown below will be added and can be passed through the `override` parameter of this method.
+
+            - ``"output_dir"``: the path to save mlflow tracking outputs locally, default to "<bundle root>/eval".
+            - ``"tracking_uri"``: uri to save mlflow tracking outputs, default to "/output_dir/mlruns".
+            - ``"experiment_name"``: experiment name for this run, default to "monai_experiment".
+            - ``"run_name"``: the name of current run.
+            - ``"save_execute_config"``: whether to save the executed config files. It can be `False`, `/path/to/artifacts`
+              or `True`. If set to `True`, will save to the default path "<bundle_root>/eval". Default to `True`.
+
+            If other string, treat it as file path to load the tracking settings.
+            If `dict`, treat it as tracking settings.
+            Will patch the target config content with `tracking handlers` and the top-level items of `configs`.
             for detailed usage examples, please check the tutorial:
             https://github.com/Project-MONAI/tutorials/blob/main/experiment_management/bundle_integrate_mlflow.ipynb.
         args_file: a JSON or YAML file to provide default values for `run_id`, `meta_file`,
diff --git a/monai/bundle/utils.py b/monai/bundle/utils.py
index b187159c89..a0f39d236f 100644
--- a/monai/bundle/utils.py
+++ b/monai/bundle/utils.py
@@ -113,7 +113,7 @@
         "experiment_name": "monai_experiment",
         "run_name": None,
         # may fill it at runtime
-        "execute_config": None,
+        "save_execute_config": True,
         "is_not_rank0": (
             "$torch.distributed.is_available() \
                 and torch.distributed.is_initialized() and torch.distributed.get_rank() > 0"
@@ -125,7 +125,7 @@
             "tracking_uri": "@tracking_uri",
             "experiment_name": "@experiment_name",
             "run_name": "@run_name",
-            "artifacts": "@execute_config",
+            "artifacts": "@save_execute_config",
             "iteration_log": True,
             "epoch_log": True,
             "tag_name": "train_loss",
@@ -148,7 +148,7 @@
             "tracking_uri": "@tracking_uri",
             "experiment_name": "@experiment_name",
             "run_name": "@run_name",
-            "artifacts": "@execute_config",
+            "artifacts": "@save_execute_config",
             "iteration_log": False,
             "close_on_complete": True,
         },
diff --git a/monai/bundle/workflows.py b/monai/bundle/workflows.py
index d876f6d7ae..471088994b 100644
--- a/monai/bundle/workflows.py
+++ b/monai/bundle/workflows.py
@@ -506,13 +506,19 @@ def patch_bundle_tracking(parser: ConfigParser, settings: dict) -> None:
                 parser[k] = v
         # save the executed config into file
         default_name = f"config_{time.strftime('%Y%m%d_%H%M%S')}.json"
-        filepath = parser.get("execute_config", None)
-        if filepath is None:
-            if "output_dir" not in parser:
-                # if no "output_dir" in the bundle config, default to "<bundle root>/eval"
-                parser["output_dir"] = f"{EXPR_KEY}{ID_REF_KEY}bundle_root + '/eval'"
-            # experiment management tools can refer to this config item to track the config info
-            parser["execute_config"] = parser["output_dir"] + f" + '/{default_name}'"
-            filepath = os.path.join(parser.get_parsed_content("output_dir"), default_name)
-        Path(filepath).parent.mkdir(parents=True, exist_ok=True)
-        parser.export_config_file(parser.get(), filepath)
+        # Users can set the `save_execute_config` to `False`, `/path/to/artifacts` or `True`.
+        # If set to False, nothing will be recorded. If set to True, the default path will be logged.
+        # If set to a file path, the given path will be logged.
+        filepath = parser.get("save_execute_config", True)
+        if filepath:
+            if isinstance(filepath, bool):
+                if "output_dir" not in parser:
+                    # if no "output_dir" in the bundle config, default to "<bundle root>/eval"
+                    parser["output_dir"] = f"{EXPR_KEY}{ID_REF_KEY}bundle_root + '/eval'"
+                # experiment management tools can refer to this config item to track the config info
+                parser["save_execute_config"] = parser["output_dir"] + f" + '/{default_name}'"
+                filepath = os.path.join(parser.get_parsed_content("output_dir"), default_name)
+            Path(filepath).parent.mkdir(parents=True, exist_ok=True)
+            parser.export_config_file(parser.get(), filepath)
+        else:
+            parser["save_execute_config"] = None
diff --git a/tests/test_fl_monai_algo.py b/tests/test_fl_monai_algo.py
index 54bec24b98..c8cb3451fc 100644
--- a/tests/test_fl_monai_algo.py
+++ b/tests/test_fl_monai_algo.py
@@ -75,7 +75,7 @@
             tracking={
                 "handlers_id": DEFAULT_HANDLERS_ID,
                 "configs": {
-                    "execute_config": f"{_data_dir}/config_executed.json",
+                    "save_execute_config": f"{_data_dir}/config_executed.json",
                     "trainer": {
                         "_target_": "MLFlowHandler",
                         "tracking_uri": path_to_uri(_data_dir) + "/mlflow_override",
@@ -201,7 +201,7 @@ def test_train(self, input_params):
         algo.finalize()
 
         # test experiment management
-        if "execute_config" in algo.train_workflow.parser:
+        if "save_execute_config" in algo.train_workflow.parser:
             self.assertTrue(os.path.exists(f"{_data_dir}/mlflow_override"))
             shutil.rmtree(f"{_data_dir}/mlflow_override")
             self.assertTrue(os.path.exists(f"{_data_dir}/config_executed.json"))
@@ -224,7 +224,7 @@ def test_evaluate(self, input_params):
         algo.evaluate(data=data, extra={})
 
         # test experiment management
-        if "execute_config" in algo.eval_workflow.parser:
+        if "save_execute_config" in algo.eval_workflow.parser:
             self.assertGreater(len(list(glob.glob(f"{_data_dir}/mlflow_*"))), 0)
             for f in list(glob.glob(f"{_data_dir}/mlflow_*")):
                 shutil.rmtree(f)

From 12684884a93173bed23190ff7a3d0e674a3e6a9f Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 12 Apr 2024 16:19:16 +0800
Subject: [PATCH 017/183] Uninstall opencv included in base image (#7626)

Fixes [tutorial
#1689](https://github.com/Project-MONAI/tutorials/issues/1689)

### Description

Uninstall opencv included in base image.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 Dockerfile                  | 3 +++
 monai/data/video_dataset.py | 2 +-
 requirements-dev.txt        | 1 +
 3 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/Dockerfile b/Dockerfile
index d5777104c8..45673fd775 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -22,6 +22,9 @@ RUN git clone --recursive https://github.com/zarr-developers/numcodecs.git && pi
 
 WORKDIR /opt/monai
 
+# remove opencv-python before opencv-python-headless installation
+RUN pip uninstall -y opencv && rm /usr/local/lib/python3.10/dist-packages/cv2 -r
+
 # install full deps
 COPY requirements.txt requirements-min.txt requirements-dev.txt /tmp/
 RUN cp /tmp/requirements.txt /tmp/req.bak \
diff --git a/monai/data/video_dataset.py b/monai/data/video_dataset.py
index be3bcf5bd5..ed5c37d777 100644
--- a/monai/data/video_dataset.py
+++ b/monai/data/video_dataset.py
@@ -177,7 +177,7 @@ def get_available_codecs() -> dict[str, str]:
             with tempfile.TemporaryDirectory() as tmp_dir:
                 for codec, ext in all_codecs.items():
                     fname = os.path.join(tmp_dir, f"test{ext}")
-                    fourcc = cv2.VideoWriter_fourcc(*codec)
+                    fourcc = cv2.VideoWriter_fourcc(*codec)  # type: ignore[attr-defined]
                     noviderr = writer.open(fname, fourcc, 1, (10, 10))
                     if noviderr:
                         codecs[codec] = ext
diff --git a/requirements-dev.txt b/requirements-dev.txt
index af1b8b89d5..2855b56ad5 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -57,3 +57,4 @@ zarr
 lpips==0.1.4
 nvidia-ml-py
 huggingface_hub
+opencv-python-headless

From 9e2904afdae311b91fa7b5a1f2171e1cc3336e19 Mon Sep 17 00:00:00 2001
From: Mingxin Zheng <18563433+mingxin-zheng@users.noreply.github.com>
Date: Fri, 12 Apr 2024 22:41:37 +0800
Subject: [PATCH 018/183] Add checks for num_fold and fail early if wrong
 (#7634)

Fixes #7628 .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Mingxin <mingxinz@nvidia.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/apps/auto3dseg/auto_runner.py | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/monai/apps/auto3dseg/auto_runner.py b/monai/apps/auto3dseg/auto_runner.py
index 52a0824227..05c961f999 100644
--- a/monai/apps/auto3dseg/auto_runner.py
+++ b/monai/apps/auto3dseg/auto_runner.py
@@ -298,9 +298,13 @@ def __init__(
                 pass
 
         # inspect and update folds
-        num_fold = self.inspect_datalist_folds(datalist_filename=datalist_filename)
+        self.max_fold = self.inspect_datalist_folds(datalist_filename=datalist_filename)
         if "num_fold" in self.data_src_cfg:
             num_fold = int(self.data_src_cfg["num_fold"])  # override from config
+            logger.info(f"Setting num_fold {num_fold} based on the input config.")
+        else:
+            num_fold = self.max_fold
+            logger.info(f"Setting num_fold {num_fold} based on the input datalist {datalist_filename}.")
 
         self.data_src_cfg["datalist"] = datalist_filename  # update path to a version in work_dir and save user input
         ConfigParser.export_config_file(
@@ -398,7 +402,10 @@ def inspect_datalist_folds(self, datalist_filename: str) -> int:
 
         if len(fold_list) > 0:
             num_fold = max(fold_list) + 1
-            logger.info(f"Setting num_fold {num_fold} based on the input datalist {datalist_filename}.")
+            logger.info(f"Found num_fold {num_fold} based on the input datalist {datalist_filename}.")
+            # check if every fold is present
+            if len(set(fold_list)) != num_fold:
+                raise ValueError(f"Fold numbers are not continuous from 0 to {num_fold - 1}")
         elif "validation" in datalist and len(datalist["validation"]) > 0:
             logger.info("No fold numbers provided, attempting to use a single fold based on the validation key")
             # update the datalist file
@@ -492,6 +499,11 @@ def set_num_fold(self, num_fold: int = 5) -> AutoRunner:
 
         if num_fold <= 0:
             raise ValueError(f"num_fold is expected to be an integer greater than zero. Now it gets {num_fold}")
+        if num_fold > self.max_fold + 1:
+            # Auto3DSeg allows no validation set, so the maximum fold number is max_fold + 1
+            raise ValueError(
+                f"num_fold is greater than the maximum fold number {self.max_fold} in {self.datalist_filename}."
+            )
         self.num_fold = num_fold
 
         return self

From 049744842adf24803324a269c7573661cbe06882 Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Mon, 15 Apr 2024 07:45:31 +0100
Subject: [PATCH 019/183] Auto3DSeg algo_template hash update (#7642)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 monai/utils/misc.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index dd0ccada3d..4611ff958e 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -527,7 +527,7 @@ def doc_images() -> str | None:
 
     @staticmethod
     def algo_hash() -> str | None:
-        return os.environ.get("MONAI_ALGO_HASH", "b910ab8")
+        return os.environ.get("MONAI_ALGO_HASH", "bd09a2a")
 
     @staticmethod
     def trace_transform() -> str | None:

From 605ffe1c25a4416b3050f324a30210cebe744590 Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Mon, 15 Apr 2024 10:26:42 +0100
Subject: [PATCH 020/183] Auto3DSeg algo_template hash update (#7643)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 monai/utils/misc.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index 4611ff958e..4388c1387e 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -527,7 +527,7 @@ def doc_images() -> str | None:
 
     @staticmethod
     def algo_hash() -> str | None:
-        return os.environ.get("MONAI_ALGO_HASH", "bd09a2a")
+        return os.environ.get("MONAI_ALGO_HASH", "f65128f")
 
     @staticmethod
     def trace_transform() -> str | None:

From bff4b1559a147f570fc881717eadc7b3e8a98b96 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 15 Apr 2024 21:06:59 +0800
Subject: [PATCH 021/183] Remove source code of numcodecs in the Dockerfile
 (#7644)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- Remove source code of numcodecs.
- Add check for only build from source when arm architecture.
- Removed the hardcoding of “python3.10”.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 Dockerfile                  | 9 +++++++--
 monai/data/video_dataset.py | 4 ++--
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/Dockerfile b/Dockerfile
index 45673fd775..fc97227351 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -17,8 +17,13 @@ FROM ${PYTORCH_IMAGE}
 LABEL maintainer="monai.contact@gmail.com"
 
 # TODO: remark for issue [revise the dockerfile](https://github.com/zarr-developers/numcodecs/issues/431)
-WORKDIR /opt
-RUN git clone --recursive https://github.com/zarr-developers/numcodecs.git && pip wheel numcodecs
+RUN if [[ $(uname -m) =~ "aarch64" ]]; then \
+        cd /opt && \
+        git clone --branch v0.12.1 --recursive https://github.com/zarr-developers/numcodecs && \
+        pip wheel numcodecs && \
+        rm -r /opt/*.whl && \
+        rm -rf /opt/numcodecs; \
+    fi
 
 WORKDIR /opt/monai
 
diff --git a/monai/data/video_dataset.py b/monai/data/video_dataset.py
index ed5c37d777..9ff23ebeff 100644
--- a/monai/data/video_dataset.py
+++ b/monai/data/video_dataset.py
@@ -173,15 +173,15 @@ def get_available_codecs() -> dict[str, str]:
         all_codecs = {"mp4v": ".mp4", "X264": ".avi", "H264": ".mp4", "MP42": ".mp4", "MJPG": ".mjpeg", "DIVX": ".avi"}
         codecs = {}
         with SuppressStderr():
-            writer = cv2.VideoWriter()
             with tempfile.TemporaryDirectory() as tmp_dir:
                 for codec, ext in all_codecs.items():
+                    writer = cv2.VideoWriter()
                     fname = os.path.join(tmp_dir, f"test{ext}")
                     fourcc = cv2.VideoWriter_fourcc(*codec)  # type: ignore[attr-defined]
                     noviderr = writer.open(fname, fourcc, 1, (10, 10))
                     if noviderr:
                         codecs[codec] = ext
-            writer.release()
+                    writer.release()
         return codecs
 
     def get_num_frames(self) -> int:

From 16d4e2fcdbc6a7be5bf9df02fc12f34eee5f26b9 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 16 Apr 2024 22:05:35 +0800
Subject: [PATCH 022/183] Remove memory_pool_limit in trt config (#7647)

Fixes https://github.com/Project-MONAI/model-zoo/issues/568.

### Description

Remove memory_pool_limit in trt config

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/utils.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/monai/networks/utils.py b/monai/networks/utils.py
index 4e6699f16b..7e3d2e2170 100644
--- a/monai/networks/utils.py
+++ b/monai/networks/utils.py
@@ -840,7 +840,6 @@ def _onnx_trt_compile(
 
     # set up the conversion configuration
     config = builder.create_builder_config()
-    config.set_memory_pool_limit(trt.MemoryPoolType.WORKSPACE, 1 << 31)
     config.add_optimization_profile(profile)
     if precision == "fp16":
         config.set_flag(trt.BuilderFlag.FP16)

From d6e6b249b1acfd043d2f599be3bef526127131ff Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 19 Apr 2024 11:16:03 +0800
Subject: [PATCH 023/183] Add version requirement for mlflow (#7659)

Fixes #7658.

### Description
set mlflow>=1.28.0,<=2.11.3
https://github.com/mlflow/mlflow/pull/11740

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/requirements.txt                     | 2 +-
 requirements-dev.txt                      | 2 +-
 setup.cfg                                 | 4 ++--
 tests/test_clip_intensity_percentiles.py  | 3 ++-
 tests/test_clip_intensity_percentilesd.py | 3 ++-
 5 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/docs/requirements.txt b/docs/requirements.txt
index e5bedf8552..5acc437391 100644
--- a/docs/requirements.txt
+++ b/docs/requirements.txt
@@ -22,7 +22,7 @@ sphinx-autodoc-typehints==1.11.1
 pandas
 einops
 transformers<4.22; python_version <= '3.10'  # https://github.com/Project-MONAI/MONAI/issues/5157
-mlflow>=1.28.0
+mlflow>=1.28.0, <=2.11.3
 clearml>=1.10.0rc0
 tensorboardX
 imagecodecs; platform_system == "Linux" or platform_system == "Darwin"
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 2855b56ad5..bfa8961ecf 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -34,7 +34,7 @@ pandas
 requests
 einops
 transformers>=4.36.0
-mlflow>=1.28.0
+mlflow>=1.28.0, <=2.11.3
 clearml>=1.10.0rc0
 matplotlib!=3.5.0
 tensorboardX
diff --git a/setup.cfg b/setup.cfg
index d7cb703d25..c8ae1630f7 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -66,7 +66,7 @@ all =
     pandas
     einops
     transformers<4.22; python_version <= '3.10'
-    mlflow>=1.28.0
+    mlflow>=1.28.0, <=2.11.3
     clearml>=1.10.0rc0
     matplotlib
     tensorboardX
@@ -125,7 +125,7 @@ einops =
 transformers =
     transformers<4.22; python_version <= '3.10'
 mlflow =
-    mlflow
+    mlflow>=1.28.0, <=2.11.3
 matplotlib =
     matplotlib
 clearml =
diff --git a/tests/test_clip_intensity_percentiles.py b/tests/test_clip_intensity_percentiles.py
index 82471e25ce..f584a0bb41 100644
--- a/tests/test_clip_intensity_percentiles.py
+++ b/tests/test_clip_intensity_percentiles.py
@@ -18,6 +18,7 @@
 from monai.transforms import ClipIntensityPercentiles
 from monai.transforms.utils import soft_clip
 from monai.transforms.utils_pytorch_numpy_unification import clip, percentile
+from monai.utils.type_conversion import convert_to_tensor
 from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
 
 
@@ -29,7 +30,7 @@ def test_hard_clipping_two_sided(self, p):
         im = p(self.imt)
         result = hard_clipper(im)
         lower, upper = percentile(im, (5, 95))
-        expected = clip(im, lower, upper)
+        expected = clip(convert_to_tensor(im), lower, upper)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
diff --git a/tests/test_clip_intensity_percentilesd.py b/tests/test_clip_intensity_percentilesd.py
index 2b49383182..97c08f9f4e 100644
--- a/tests/test_clip_intensity_percentilesd.py
+++ b/tests/test_clip_intensity_percentilesd.py
@@ -19,6 +19,7 @@
 from monai.transforms import ClipIntensityPercentilesd
 from monai.transforms.utils import soft_clip
 from monai.transforms.utils_pytorch_numpy_unification import clip, percentile
+from monai.utils.type_conversion import convert_to_tensor
 from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
 
 
@@ -31,7 +32,7 @@ def test_hard_clipping_two_sided(self, p):
         im = p(self.imt)
         result = hard_clipper({key: im})
         lower, upper = percentile(im, (5, 95))
-        expected = clip(im, lower, upper)
+        expected = clip(convert_to_tensor(im), lower, upper)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])

From ffd4454b576abd4eaae30b364f41c213e30dca4c Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Fri, 19 Apr 2024 05:28:48 +0100
Subject: [PATCH 024/183] Auto3DSeg algo_template hash update (#7674)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 monai/utils/misc.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index 4388c1387e..de4bfdc2bb 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -527,7 +527,7 @@ def doc_images() -> str | None:
 
     @staticmethod
     def algo_hash() -> str | None:
-        return os.environ.get("MONAI_ALGO_HASH", "f65128f")
+        return os.environ.get("MONAI_ALGO_HASH", "def5f26")
 
     @staticmethod
     def trace_transform() -> str | None:

From 224c47a5741f0e0b454b6ffda5a65bf598af6a7c Mon Sep 17 00:00:00 2001
From: Han Wang <freddie.wanah@gmail.com>
Date: Fri, 19 Apr 2024 07:11:26 +0100
Subject: [PATCH 025/183] Fixed four test issues within test code. (#7662)

### Description

This PR fixed the three issues within the test code that could cause
problems during CICD.

Three issues:

tests/test_pad_collation.py line 120: supposed to be assertEqual, but
not assertTure? or it will always be true.
tests/test_to_numpy.py line 74: Same as above.
tests/test_auto3dseg.py line 370: typo? two same asserts next to each
other.
tests/test_compose.py line 719, 725, 727: Should be assertEqual instead
of assertTrue

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Han Wang <freddie.wanah@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 tests/test_auto3dseg.py     | 1 -
 tests/test_compose.py       | 6 +++---
 tests/test_pad_collation.py | 2 +-
 tests/test_to_numpy.py      | 2 +-
 4 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/tests/test_auto3dseg.py b/tests/test_auto3dseg.py
index e2097679e2..6be33bf6ca 100644
--- a/tests/test_auto3dseg.py
+++ b/tests/test_auto3dseg.py
@@ -367,7 +367,6 @@ def test_filename_case_analyzer(self):
         for batch_data in self.dataset:
             d = transform(batch_data[0])
             assert DataStatsKeys.BY_CASE_IMAGE_PATH in d
-            assert DataStatsKeys.BY_CASE_IMAGE_PATH in d
 
     def test_filename_case_analyzer_image_only(self):
         analyzer_image = FilenameStats("image", DataStatsKeys.BY_CASE_IMAGE_PATH)
diff --git a/tests/test_compose.py b/tests/test_compose.py
index 309767833b..3c53ac4a22 100644
--- a/tests/test_compose.py
+++ b/tests/test_compose.py
@@ -716,15 +716,15 @@ def test_compose_execute_equivalence_with_flags(self, flags, data, pipeline):
                 for k in actual.keys():
                     self.assertEqual(expected[k], actual[k])
             else:
-                self.assertTrue(expected, actual)
+                self.assertEqual(expected, actual)
 
             p = deepcopy(pipeline)
             actual = execute_compose(execute_compose(data, p, start=0, end=cutoff, **flags), p, start=cutoff, **flags)
             if isinstance(actual, dict):
                 for k in actual.keys():
-                    self.assertTrue(expected[k], actual[k])
+                    self.assertEqual(expected[k], actual[k])
             else:
-                self.assertTrue(expected, actual)
+                self.assertEqual(expected, actual)
 
 
 class TestComposeCallableInput(unittest.TestCase):
diff --git a/tests/test_pad_collation.py b/tests/test_pad_collation.py
index ee6e001438..17f49611df 100644
--- a/tests/test_pad_collation.py
+++ b/tests/test_pad_collation.py
@@ -117,7 +117,7 @@ def test_pad_collation(self, t_type, collate_method, transform):
             batch_inverse = BatchInverseTransform(dataset.transform, loader)
             for data in loader:
                 output = batch_inverse(data)
-                self.assertTrue(output[0]["image"].shape, (1, 10, 9))
+                self.assertEqual(output[0]["image"].shape, (1, 10, 9))
 
 
 if __name__ == "__main__":
diff --git a/tests/test_to_numpy.py b/tests/test_to_numpy.py
index 8f7cf34865..f92b7c0075 100644
--- a/tests/test_to_numpy.py
+++ b/tests/test_to_numpy.py
@@ -71,7 +71,7 @@ def test_list_tuple(self):
         assert_allclose(result, np.asarray(test_data), type_test=False)
         test_data = ((1, 2), (3, 4))
         result = ToNumpy(wrap_sequence=False)(test_data)
-        self.assertTrue(type(result), tuple)
+        self.assertIsInstance(result, tuple)
         assert_allclose(result, ((np.asarray(1), np.asarray(2)), (np.asarray(3), np.asarray(4))))
 
     def test_single_value(self):

From 7a6b69fc22b25fbdc747e0e6d28f85c90947cfd4 Mon Sep 17 00:00:00 2001
From: Han Wang <freddie.wanah@gmail.com>
Date: Fri, 19 Apr 2024 10:15:38 +0100
Subject: [PATCH 026/183] Adapt to use assert raises (#7670)

### Description

Some test cases use try/except command, instead of catch exceptions with
assertRaise. This will cause extra execution time according to the
experimental tests.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Han Wang <freddie.wanah@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 tests/test_decathlondataset.py | 5 +----
 tests/test_mednistdataset.py   | 5 +----
 tests/test_monai_utils_misc.py | 9 ++++-----
 tests/test_tciadataset.py      | 4 +---
 4 files changed, 7 insertions(+), 16 deletions(-)

diff --git a/tests/test_decathlondataset.py b/tests/test_decathlondataset.py
index d220cd9097..70a2a6c06c 100644
--- a/tests/test_decathlondataset.py
+++ b/tests/test_decathlondataset.py
@@ -80,7 +80,7 @@ def _test_dataset(dataset):
         self.assertDictEqual(properties["labels"], {"0": "background", "1": "Anterior", "2": "Posterior"})
 
         shutil.rmtree(os.path.join(testing_dir, "Task04_Hippocampus"))
-        try:
+        with self.assertRaisesRegex(RuntimeError, "^Cannot find dataset directory"):
             DecathlonDataset(
                 root_dir=testing_dir,
                 task="Task04_Hippocampus",
@@ -88,9 +88,6 @@ def _test_dataset(dataset):
                 section="validation",
                 download=False,
             )
-        except RuntimeError as e:
-            print(str(e))
-            self.assertTrue(str(e).startswith("Cannot find dataset directory"))
 
 
 if __name__ == "__main__":
diff --git a/tests/test_mednistdataset.py b/tests/test_mednistdataset.py
index baf3bf4f2d..1db632c144 100644
--- a/tests/test_mednistdataset.py
+++ b/tests/test_mednistdataset.py
@@ -65,11 +65,8 @@ def _test_dataset(dataset):
         self.assertEqual(data[0]["class_name"], "AbdomenCT")
         self.assertEqual(data[0]["label"], 0)
         shutil.rmtree(os.path.join(testing_dir, "MedNIST"))
-        try:
+        with self.assertRaisesRegex(RuntimeError, "^Cannot find dataset directory"):
             MedNISTDataset(root_dir=testing_dir, transform=transform, section="test", download=False)
-        except RuntimeError as e:
-            print(str(e))
-            self.assertTrue(str(e).startswith("Cannot find dataset directory"))
 
 
 if __name__ == "__main__":
diff --git a/tests/test_monai_utils_misc.py b/tests/test_monai_utils_misc.py
index a2a4ed62f7..f4eb5d3956 100644
--- a/tests/test_monai_utils_misc.py
+++ b/tests/test_monai_utils_misc.py
@@ -92,12 +92,11 @@ def test_run_cmd(self):
         cmd2 = "-c"
         cmd3 = 'import sys; print("\\tThis is on stderr\\n", file=sys.stderr); sys.exit(1)'
         os.environ["MONAI_DEBUG"] = str(True)
-        try:
+        with self.assertRaises(RuntimeError) as cm:
             run_cmd([cmd1, cmd2, cmd3], check=True)
-        except RuntimeError as err:
-            self.assertIn("This is on stderr", str(err))
-            self.assertNotIn("\\n", str(err))
-            self.assertNotIn("\\t", str(err))
+        self.assertIn("This is on stderr", str(cm.exception))
+        self.assertNotIn("\\n", str(cm.exception))
+        self.assertNotIn("\\t", str(cm.exception))
 
 
 if __name__ == "__main__":
diff --git a/tests/test_tciadataset.py b/tests/test_tciadataset.py
index d996922e20..5a16bb4816 100644
--- a/tests/test_tciadataset.py
+++ b/tests/test_tciadataset.py
@@ -108,7 +108,7 @@ def _test_dataset(dataset):
             )[0]
 
         shutil.rmtree(os.path.join(testing_dir, collection))
-        try:
+        with self.assertRaisesRegex(RuntimeError, "^Cannot find dataset directory"):
             TciaDataset(
                 root_dir=testing_dir,
                 collection=collection,
@@ -117,8 +117,6 @@ def _test_dataset(dataset):
                 download=False,
                 val_frac=val_frac,
             )
-        except RuntimeError as e:
-            self.assertTrue(str(e).startswith("Cannot find dataset directory"))
 
 
 if __name__ == "__main__":

From 03a5fa695ad02fcb916d5495e84f8f01883806e2 Mon Sep 17 00:00:00 2001
From: Fabian Klopfer <fabian.klopfer@ieee.org>
Date: Fri, 19 Apr 2024 16:21:14 +0200
Subject: [PATCH 027/183] MedicalNetPerceptualSimilarity: Add multi-channel
 (#7568)

Fixes #7567 .

### Description
MedicalNetPerceptualSimilarity: Add multi-channel support for 3Dvolumes.
The current version of the code in the dev branch already largely
supports that besides the following:
medicalnet_* require inputs to have a single channel.
This PR passes the multi-channel volume channel-wise to the networks and
concatenates the resulting feature vectors.
The existing code takes care of averaging over channels and spatially.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Fabian Klopfer <fabian.klopfer@ieee.org>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/losses/perceptual.py    | 56 ++++++++++++++++++++++++++++++-----
 tests/test_perceptual_loss.py | 34 +++++++++++++++++++--
 2 files changed, 80 insertions(+), 10 deletions(-)

diff --git a/monai/losses/perceptual.py b/monai/losses/perceptual.py
index fd61603b03..02493f412b 100644
--- a/monai/losses/perceptual.py
+++ b/monai/losses/perceptual.py
@@ -45,6 +45,7 @@ class PerceptualLoss(nn.Module):
 
     The fake 3D implementation is based on a 2.5D approach where we calculate the 2D perceptual loss on slices from all
     three axes and average. The full 3D approach uses a 3D network to calculate the perceptual loss.
+    MedicalNet networks are only compatible with 3D inputs and support channel-wise loss.
 
     Args:
         spatial_dims: number of spatial dimensions.
@@ -62,6 +63,8 @@ class PerceptualLoss(nn.Module):
         pretrained_state_dict_key: if `pretrained_path` is not `None`, this argument is used to
             extract the expected state dict. This argument only works when ``"network_type"`` is "resnet50".
             Defaults to `None`.
+        channel_wise: if True, the loss is returned per channel. Otherwise the loss is averaged over the channels.
+                Defaults to ``False``.
     """
 
     def __init__(
@@ -74,6 +77,7 @@ def __init__(
         pretrained: bool = True,
         pretrained_path: str | None = None,
         pretrained_state_dict_key: str | None = None,
+        channel_wise: bool = False,
     ):
         super().__init__()
 
@@ -86,6 +90,9 @@ def __init__(
                 "Argument is_fake_3d must be set to False."
             )
 
+        if channel_wise and "medicalnet_" not in network_type:
+            raise ValueError("Channel-wise loss is only compatible with MedicalNet networks.")
+
         if network_type.lower() not in list(PercetualNetworkType):
             raise ValueError(
                 "Unrecognised criterion entered for Adversarial Loss. Must be one in: %s"
@@ -102,7 +109,9 @@ def __init__(
         self.spatial_dims = spatial_dims
         self.perceptual_function: nn.Module
         if spatial_dims == 3 and is_fake_3d is False:
-            self.perceptual_function = MedicalNetPerceptualSimilarity(net=network_type, verbose=False)
+            self.perceptual_function = MedicalNetPerceptualSimilarity(
+                net=network_type, verbose=False, channel_wise=channel_wise
+            )
         elif "radimagenet_" in network_type:
             self.perceptual_function = RadImageNetPerceptualSimilarity(net=network_type, verbose=False)
         elif network_type == "resnet50":
@@ -172,7 +181,12 @@ def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor:
             # 2D and real 3D cases
             loss = self.perceptual_function(input, target)
 
-        return torch.mean(loss)
+        if self.channel_wise:
+            loss = torch.mean(loss.squeeze(), dim=0)
+        else:
+            loss = torch.mean(loss)
+
+        return loss
 
 
 class MedicalNetPerceptualSimilarity(nn.Module):
@@ -185,14 +199,20 @@ class MedicalNetPerceptualSimilarity(nn.Module):
         net: {``"medicalnet_resnet10_23datasets"``, ``"medicalnet_resnet50_23datasets"``}
             Specifies the network architecture to use. Defaults to ``"medicalnet_resnet10_23datasets"``.
         verbose: if false, mute messages from torch Hub load function.
+        channel_wise: if True, the loss is returned per channel. Otherwise the loss is averaged over the channels.
+                Defaults to ``False``.
     """
 
-    def __init__(self, net: str = "medicalnet_resnet10_23datasets", verbose: bool = False) -> None:
+    def __init__(
+        self, net: str = "medicalnet_resnet10_23datasets", verbose: bool = False, channel_wise: bool = False
+    ) -> None:
         super().__init__()
         torch.hub._validate_not_a_forked_repo = lambda a, b, c: True
         self.model = torch.hub.load("warvito/MedicalNet-models", model=net, verbose=verbose)
         self.eval()
 
+        self.channel_wise = channel_wise
+
         for param in self.parameters():
             param.requires_grad = False
 
@@ -206,20 +226,42 @@ def forward(self, input: torch.Tensor, target: torch.Tensor) -> torch.Tensor:
         Args:
             input: 3D input tensor with shape BCDHW.
             target: 3D target tensor with shape BCDHW.
+
         """
         input = medicalnet_intensity_normalisation(input)
         target = medicalnet_intensity_normalisation(target)
 
         # Get model outputs
-        outs_input = self.model.forward(input)
-        outs_target = self.model.forward(target)
+        feats_per_ch = 0
+        for ch_idx in range(input.shape[1]):
+            input_channel = input[:, ch_idx, ...].unsqueeze(1)
+            target_channel = target[:, ch_idx, ...].unsqueeze(1)
+
+            if ch_idx == 0:
+                outs_input = self.model.forward(input_channel)
+                outs_target = self.model.forward(target_channel)
+                feats_per_ch = outs_input.shape[1]
+            else:
+                outs_input = torch.cat([outs_input, self.model.forward(input_channel)], dim=1)
+                outs_target = torch.cat([outs_target, self.model.forward(target_channel)], dim=1)
 
         # Normalise through the channels
         feats_input = normalize_tensor(outs_input)
         feats_target = normalize_tensor(outs_target)
 
-        results: torch.Tensor = (feats_input - feats_target) ** 2
-        results = spatial_average_3d(results.sum(dim=1, keepdim=True), keepdim=True)
+        feats_diff: torch.Tensor = (feats_input - feats_target) ** 2
+        if self.channel_wise:
+            results = torch.zeros(
+                feats_diff.shape[0], input.shape[1], feats_diff.shape[2], feats_diff.shape[3], feats_diff.shape[4]
+            )
+            for i in range(input.shape[1]):
+                l_idx = i * feats_per_ch
+                r_idx = (i + 1) * feats_per_ch
+                results[:, i, ...] = feats_diff[:, l_idx : i + r_idx, ...].sum(dim=1)
+        else:
+            results = feats_diff.sum(dim=1, keepdim=True)
+
+        results = spatial_average_3d(results, keepdim=True)
 
         return results
 
diff --git a/tests/test_perceptual_loss.py b/tests/test_perceptual_loss.py
index 02232e6f8d..548613e2ae 100644
--- a/tests/test_perceptual_loss.py
+++ b/tests/test_perceptual_loss.py
@@ -18,7 +18,7 @@
 
 from monai.losses import PerceptualLoss
 from monai.utils import optional_import
-from tests.utils import SkipIfBeforePyTorchVersion, skip_if_downloading_fails, skip_if_quick
+from tests.utils import SkipIfBeforePyTorchVersion, assert_allclose, skip_if_downloading_fails, skip_if_quick
 
 _, has_torchvision = optional_import("torchvision")
 TEST_CASES = [
@@ -40,11 +40,31 @@
         (2, 1, 64, 64, 64),
         (2, 1, 64, 64, 64),
     ],
+    [
+        {"spatial_dims": 3, "network_type": "medicalnet_resnet10_23datasets", "is_fake_3d": False},
+        (2, 6, 64, 64, 64),
+        (2, 6, 64, 64, 64),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "network_type": "medicalnet_resnet10_23datasets",
+            "is_fake_3d": False,
+            "channel_wise": True,
+        },
+        (2, 6, 64, 64, 64),
+        (2, 6, 64, 64, 64),
+    ],
     [
         {"spatial_dims": 3, "network_type": "medicalnet_resnet50_23datasets", "is_fake_3d": False},
         (2, 1, 64, 64, 64),
         (2, 1, 64, 64, 64),
     ],
+    [
+        {"spatial_dims": 3, "network_type": "medicalnet_resnet50_23datasets", "is_fake_3d": False},
+        (2, 6, 64, 64, 64),
+        (2, 6, 64, 64, 64),
+    ],
     [
         {"spatial_dims": 3, "network_type": "resnet50", "is_fake_3d": True, "pretrained": True, "fake_3d_ratio": 0.2},
         (2, 1, 64, 64, 64),
@@ -63,7 +83,11 @@ def test_shape(self, input_param, input_shape, target_shape):
         with skip_if_downloading_fails():
             loss = PerceptualLoss(**input_param)
         result = loss(torch.randn(input_shape), torch.randn(target_shape))
-        self.assertEqual(result.shape, torch.Size([]))
+
+        if "channel_wise" in input_param.keys() and input_param["channel_wise"]:
+            self.assertEqual(result.shape, torch.Size([input_shape[1]]))
+        else:
+            self.assertEqual(result.shape, torch.Size([]))
 
     @parameterized.expand(TEST_CASES)
     def test_identical_input(self, input_param, input_shape, target_shape):
@@ -71,7 +95,11 @@ def test_identical_input(self, input_param, input_shape, target_shape):
             loss = PerceptualLoss(**input_param)
         tensor = torch.randn(input_shape)
         result = loss(tensor, tensor)
-        self.assertEqual(result, torch.Tensor([0.0]))
+
+        if "channel_wise" in input_param.keys() and input_param["channel_wise"]:
+            assert_allclose(result, torch.Tensor([0.0] * input_shape[1]))
+        else:
+            self.assertEqual(result, torch.Tensor([0.0]))
 
     def test_different_shape(self):
         with skip_if_downloading_fails():

From c6bf8e9c52ccecbdfe289151772940df4356f358 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 22 Apr 2024 16:07:18 +0800
Subject: [PATCH 028/183] Workaround for B909 in flake8-bugbear (#7691)

Workaround for #7690

### Description
Updated with flake8-bugbear, causing the B909 error. A workaround fix
was made for the flake8-bugbear version because the 24.4.21 update
contained some false positives.
Fore more information, see
https://github.com/PyCQA/flake8-bugbear/issues/467

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .pre-commit-config.yaml | 2 +-
 requirements-dev.txt    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 08b75b1e97..b71a2bac43 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -58,7 +58,7 @@ repos:
         name: Unused noqa
         additional_dependencies:
           - flake8>=3.8.1
-          - flake8-bugbear
+          - flake8-bugbear<=24.2.6
           - flake8-comprehensions
           - pep8-naming
         exclude: |
diff --git a/requirements-dev.txt b/requirements-dev.txt
index bfa8961ecf..b207b56b19 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -11,7 +11,7 @@ scikit-image>=0.19.0
 tqdm>=4.47.0
 lmdb
 flake8>=3.8.1
-flake8-bugbear
+flake8-bugbear<=24.2.6  # https://github.com/Project-MONAI/MONAI/issues/7690
 flake8-comprehensions
 mccabe
 pep8-naming

From 178ebc8cc9352f3d7f67286a3856de67d355576f Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 22 Apr 2024 22:49:41 +0800
Subject: [PATCH 029/183] Fix AttributeError in 'PerceptualLoss' (#7693)

Fixes #7692

### Description

Fix AttributeError in 'PerceptualLoss'

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/losses/perceptual.py                |  1 +
 tests/test_clip_intensity_percentiles.py  | 16 ++++++++--------
 tests/test_clip_intensity_percentilesd.py | 19 +++++++++----------
 3 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/monai/losses/perceptual.py b/monai/losses/perceptual.py
index 02493f412b..a8ae90993a 100644
--- a/monai/losses/perceptual.py
+++ b/monai/losses/perceptual.py
@@ -125,6 +125,7 @@ def __init__(
             self.perceptual_function = LPIPS(pretrained=pretrained, net=network_type, verbose=False)
         self.is_fake_3d = is_fake_3d
         self.fake_3d_ratio = fake_3d_ratio
+        self.channel_wise = channel_wise
 
     def _calculate_axis_loss(self, input: torch.Tensor, target: torch.Tensor, spatial_axis: int) -> torch.Tensor:
         """
diff --git a/tests/test_clip_intensity_percentiles.py b/tests/test_clip_intensity_percentiles.py
index f584a0bb41..01820e7115 100644
--- a/tests/test_clip_intensity_percentiles.py
+++ b/tests/test_clip_intensity_percentiles.py
@@ -31,7 +31,7 @@ def test_hard_clipping_two_sided(self, p):
         result = hard_clipper(im)
         lower, upper = percentile(im, (5, 95))
         expected = clip(convert_to_tensor(im), lower, upper)
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_one_sided_high(self, p):
@@ -40,7 +40,7 @@ def test_hard_clipping_one_sided_high(self, p):
         result = hard_clipper(im)
         lower, upper = percentile(im, (0, 95))
         expected = clip(im, lower, upper)
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_one_sided_low(self, p):
@@ -49,7 +49,7 @@ def test_hard_clipping_one_sided_low(self, p):
         result = hard_clipper(im)
         lower, upper = percentile(im, (5, 100))
         expected = clip(im, lower, upper)
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_two_sided(self, p):
@@ -89,7 +89,7 @@ def test_channel_wise(self, p):
         for i, c in enumerate(im):
             lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
-            assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+            assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     def test_ill_sharpness_factor(self):
         with self.assertRaises(ValueError):
@@ -121,7 +121,7 @@ def test_hard_clipping_two_sided(self, p):
         result = hard_clipper(im)
         lower, upper = percentile(im, (5, 95))
         expected = clip(im, lower, upper)
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_one_sided_high(self, p):
@@ -130,7 +130,7 @@ def test_hard_clipping_one_sided_high(self, p):
         result = hard_clipper(im)
         lower, upper = percentile(im, (0, 95))
         expected = clip(im, lower, upper)
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_one_sided_low(self, p):
@@ -139,7 +139,7 @@ def test_hard_clipping_one_sided_low(self, p):
         result = hard_clipper(im)
         lower, upper = percentile(im, (5, 100))
         expected = clip(im, lower, upper)
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_two_sided(self, p):
@@ -179,7 +179,7 @@ def test_channel_wise(self, p):
         for i, c in enumerate(im):
             lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
-            assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+            assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_clip_intensity_percentilesd.py b/tests/test_clip_intensity_percentilesd.py
index 97c08f9f4e..fa727b6adb 100644
--- a/tests/test_clip_intensity_percentilesd.py
+++ b/tests/test_clip_intensity_percentilesd.py
@@ -19,7 +19,6 @@
 from monai.transforms import ClipIntensityPercentilesd
 from monai.transforms.utils import soft_clip
 from monai.transforms.utils_pytorch_numpy_unification import clip, percentile
-from monai.utils.type_conversion import convert_to_tensor
 from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
 
 
@@ -32,8 +31,8 @@ def test_hard_clipping_two_sided(self, p):
         im = p(self.imt)
         result = hard_clipper({key: im})
         lower, upper = percentile(im, (5, 95))
-        expected = clip(convert_to_tensor(im), lower, upper)
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        expected = clip(im, lower, upper)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_one_sided_high(self, p):
@@ -43,7 +42,7 @@ def test_hard_clipping_one_sided_high(self, p):
         result = hard_clipper({key: im})
         lower, upper = percentile(im, (0, 95))
         expected = clip(im, lower, upper)
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_one_sided_low(self, p):
@@ -53,7 +52,7 @@ def test_hard_clipping_one_sided_low(self, p):
         result = hard_clipper({key: im})
         lower, upper = percentile(im, (5, 100))
         expected = clip(im, lower, upper)
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_two_sided(self, p):
@@ -97,7 +96,7 @@ def test_channel_wise(self, p):
         for i, c in enumerate(im):
             lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
-            assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+            assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     def test_ill_sharpness_factor(self):
         key = "img"
@@ -135,7 +134,7 @@ def test_hard_clipping_two_sided(self, p):
         result = hard_clipper({key: im})
         lower, upper = percentile(im, (5, 95))
         expected = clip(im, lower, upper)
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_one_sided_high(self, p):
@@ -145,7 +144,7 @@ def test_hard_clipping_one_sided_high(self, p):
         result = hard_clipper({key: im})
         lower, upper = percentile(im, (0, 95))
         expected = clip(im, lower, upper)
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_one_sided_low(self, p):
@@ -155,7 +154,7 @@ def test_hard_clipping_one_sided_low(self, p):
         result = hard_clipper({key: im})
         lower, upper = percentile(im, (5, 100))
         expected = clip(im, lower, upper)
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_two_sided(self, p):
@@ -199,7 +198,7 @@ def test_channel_wise(self, p):
         for i, c in enumerate(im):
             lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
-            assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-7, atol=0)
+            assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
 
 if __name__ == "__main__":

From ac9b1862f540bebdc3ff45a1655a0667602a0bce Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 23 Apr 2024 09:28:13 +0800
Subject: [PATCH 030/183] Always convert input to C-order in
 distance_transform_edt (#7675)

Fixes #7660

`indices_` may not always be a shallow copy for F-order input as it
constantly converts to C-order

https://github.com/Project-MONAI/MONAI/blob/ffd4454b576abd4eaae30b364f41c213e30dca4c/monai/transforms/utils.py#L2209

https://github.com/Project-MONAI/MONAI/blob/ffd4454b576abd4eaae30b364f41c213e30dca4c/monai/utils/type_conversion.py#L262

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/transforms/utils.py                | 2 +-
 tests/test_clip_intensity_percentiles.py | 3 +--
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index 14f35e1219..560dbac346 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -2190,7 +2190,7 @@ def distance_transform_edt(
         if return_distances:
             dtype = torch.float64 if float64_distances else torch.float32
             if distances is None:
-                distances = torch.zeros_like(img, dtype=dtype)  # type: ignore
+                distances = torch.zeros_like(img, memory_format=torch.contiguous_format, dtype=dtype)  # type: ignore
             else:
                 if not isinstance(distances, torch.Tensor) and distances.device != img.device:
                     raise TypeError("distances must be a torch.Tensor on the same device as img")
diff --git a/tests/test_clip_intensity_percentiles.py b/tests/test_clip_intensity_percentiles.py
index 01820e7115..af157446f6 100644
--- a/tests/test_clip_intensity_percentiles.py
+++ b/tests/test_clip_intensity_percentiles.py
@@ -18,7 +18,6 @@
 from monai.transforms import ClipIntensityPercentiles
 from monai.transforms.utils import soft_clip
 from monai.transforms.utils_pytorch_numpy_unification import clip, percentile
-from monai.utils.type_conversion import convert_to_tensor
 from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
 
 
@@ -30,7 +29,7 @@ def test_hard_clipping_two_sided(self, p):
         im = p(self.imt)
         result = hard_clipper(im)
         lower, upper = percentile(im, (5, 95))
-        expected = clip(convert_to_tensor(im), lower, upper)
+        expected = clip(im, lower, upper)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])

From a59676ff1be98190fbd8ab1d804aa822cc40ca14 Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Tue, 23 Apr 2024 04:47:12 +0100
Subject: [PATCH 031/183] Auto3DSeg algo_template hash update (#7695)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/utils/misc.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index de4bfdc2bb..2b4b30fe48 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -527,7 +527,7 @@ def doc_images() -> str | None:
 
     @staticmethod
     def algo_hash() -> str | None:
-        return os.environ.get("MONAI_ALGO_HASH", "def5f26")
+        return os.environ.get("MONAI_ALGO_HASH", "4403f94")
 
     @staticmethod
     def trace_transform() -> str | None:

From ec6aa33a4a528d96733c380b50136c81f662bf38 Mon Sep 17 00:00:00 2001
From: Han Wang <freddie.wanah@gmail.com>
Date: Tue, 23 Apr 2024 06:15:20 +0100
Subject: [PATCH 032/183] Merge similar test components with parameterized
 (#7663)

### Description

I noticed some test cases contain same duplicated asserts. Having
multiple asserts in one test cases can cause potential issues like when
the first assert fails, the test case stops and won't check the second
assert. By using @parameterized.expand, this issue can be resolved and
the caching also saves execution time.

Added sign-offs from  #7648

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Han Wang <freddie.wanah@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 tests/test_affine_transform.py                |  27 +--
 tests/test_compute_f_beta.py                  |  36 ++--
 tests/test_global_mutual_information_loss.py  |  40 ++--
 tests/test_hausdorff_loss.py                  |  22 +--
 tests/test_median_filter.py                   |  21 +-
 tests/test_multi_scale.py                     |  29 +--
 tests/test_optional_import.py                 |  27 +--
 tests/test_perceptual_loss.py                 |   8 +-
 tests/test_prepare_batch_default.py           | 100 +++-------
 tests/test_rand_affine.py                     |   7 +-
 tests/test_rand_affined.py                    |   9 +-
 tests/test_tversky_loss.py                    |  11 +-
 ...est_ultrasound_confidence_map_transform.py | 183 ++++++------------
 tests/test_vit.py                             |  99 +++-------
 tests/test_vitautoenc.py                      |  97 +++-------
 15 files changed, 243 insertions(+), 473 deletions(-)

diff --git a/tests/test_affine_transform.py b/tests/test_affine_transform.py
index 6ea036bce8..11464070e0 100644
--- a/tests/test_affine_transform.py
+++ b/tests/test_affine_transform.py
@@ -133,28 +133,17 @@ def test_to_norm_affine_ill(self, affine, src_size, dst_size, align_corners):
 
 class TestAffineTransform(unittest.TestCase):
 
-    def test_affine_shift(self):
-        affine = torch.as_tensor([[1.0, 0.0, 0.0], [0.0, 1.0, -1.0]])
-        image = torch.as_tensor([[[[4.0, 1.0, 3.0, 2.0], [7.0, 6.0, 8.0, 5.0], [3.0, 5.0, 3.0, 6.0]]]])
-        out = AffineTransform(align_corners=False)(image, affine)
-        out = out.detach().cpu().numpy()
-        expected = [[[[0, 4, 1, 3], [0, 7, 6, 8], [0, 3, 5, 3]]]]
-        np.testing.assert_allclose(out, expected, atol=1e-5, rtol=_rtol)
-
-    def test_affine_shift_1(self):
-        affine = torch.as_tensor([[1.0, 0.0, -1.0], [0.0, 1.0, -1.0]])
-        image = torch.as_tensor([[[[4.0, 1.0, 3.0, 2.0], [7.0, 6.0, 8.0, 5.0], [3.0, 5.0, 3.0, 6.0]]]])
-        out = AffineTransform(align_corners=False)(image, affine)
-        out = out.detach().cpu().numpy()
-        expected = [[[[0, 0, 0, 0], [0, 4, 1, 3], [0, 7, 6, 8]]]]
-        np.testing.assert_allclose(out, expected, atol=1e-5, rtol=_rtol)
-
-    def test_affine_shift_2(self):
-        affine = torch.as_tensor([[1.0, 0.0, -1.0], [0.0, 1.0, 0.0]])
+    @parameterized.expand(
+        [
+            (torch.as_tensor([[1.0, 0.0, 0.0], [0.0, 1.0, -1.0]]), [[[[0, 4, 1, 3], [0, 7, 6, 8], [0, 3, 5, 3]]]]),
+            (torch.as_tensor([[1.0, 0.0, -1.0], [0.0, 1.0, -1.0]]), [[[[0, 0, 0, 0], [0, 4, 1, 3], [0, 7, 6, 8]]]]),
+            (torch.as_tensor([[1.0, 0.0, -1.0], [0.0, 1.0, 0.0]]), [[[[0, 0, 0, 0], [4, 1, 3, 2], [7, 6, 8, 5]]]]),
+        ]
+    )
+    def test_affine_transforms(self, affine, expected):
         image = torch.as_tensor([[[[4.0, 1.0, 3.0, 2.0], [7.0, 6.0, 8.0, 5.0], [3.0, 5.0, 3.0, 6.0]]]])
         out = AffineTransform(align_corners=False)(image, affine)
         out = out.detach().cpu().numpy()
-        expected = [[[[0, 0, 0, 0], [4, 1, 3, 2], [7, 6, 8, 5]]]]
         np.testing.assert_allclose(out, expected, atol=1e-5, rtol=_rtol)
 
     def test_zoom(self):
diff --git a/tests/test_compute_f_beta.py b/tests/test_compute_f_beta.py
index 85997577cf..43ebb6a6d5 100644
--- a/tests/test_compute_f_beta.py
+++ b/tests/test_compute_f_beta.py
@@ -15,6 +15,7 @@
 
 import numpy as np
 import torch
+from parameterized import parameterized
 
 from monai.metrics import FBetaScore
 from tests.utils import assert_allclose
@@ -33,26 +34,21 @@ def test_expecting_success_and_device(self):
         assert_allclose(result, torch.Tensor([0.714286]), atol=1e-6, rtol=1e-6)
         np.testing.assert_equal(result.device, y_pred.device)
 
-    def test_expecting_success2(self):
-        metric = FBetaScore(beta=0.5)
-        metric(
-            y_pred=torch.Tensor([[1, 1, 1], [1, 1, 1], [1, 1, 1]]), y=torch.Tensor([[1, 0, 1], [0, 1, 0], [1, 0, 1]])
-        )
-        assert_allclose(metric.aggregate()[0], torch.Tensor([0.609756]), atol=1e-6, rtol=1e-6)
-
-    def test_expecting_success3(self):
-        metric = FBetaScore(beta=2)
-        metric(
-            y_pred=torch.Tensor([[1, 1, 1], [1, 1, 1], [1, 1, 1]]), y=torch.Tensor([[1, 0, 1], [0, 1, 0], [1, 0, 1]])
-        )
-        assert_allclose(metric.aggregate()[0], torch.Tensor([0.862069]), atol=1e-6, rtol=1e-6)
-
-    def test_denominator_is_zero(self):
-        metric = FBetaScore(beta=2)
-        metric(
-            y_pred=torch.Tensor([[1, 1, 1], [1, 1, 1], [1, 1, 1]]), y=torch.Tensor([[0, 0, 0], [0, 0, 0], [0, 0, 0]])
-        )
-        assert_allclose(metric.aggregate()[0], torch.Tensor([0.0]), atol=1e-6, rtol=1e-6)
+    @parameterized.expand(
+        [
+            (0.5, torch.Tensor([[1, 0, 1], [0, 1, 0], [1, 0, 1]]), torch.Tensor([0.609756])),  # success_beta_0_5
+            (2, torch.Tensor([[1, 0, 1], [0, 1, 0], [1, 0, 1]]), torch.Tensor([0.862069])),  # success_beta_2
+            (
+                2,  # success_beta_2, denominator_zero
+                torch.Tensor([[0, 0, 0], [0, 0, 0], [0, 0, 0]]),
+                torch.Tensor([0.0]),
+            ),
+        ]
+    )
+    def test_success_and_zero(self, beta, y, expected_score):
+        metric = FBetaScore(beta=beta)
+        metric(y_pred=torch.Tensor([[1, 1, 1], [1, 1, 1], [1, 1, 1]]), y=y)
+        assert_allclose(metric.aggregate()[0], expected_score, atol=1e-6, rtol=1e-6)
 
     def test_number_of_dimensions_less_than_2_should_raise_error(self):
         metric = FBetaScore()
diff --git a/tests/test_global_mutual_information_loss.py b/tests/test_global_mutual_information_loss.py
index 36a1978c93..22f5e88431 100644
--- a/tests/test_global_mutual_information_loss.py
+++ b/tests/test_global_mutual_information_loss.py
@@ -15,6 +15,7 @@
 
 import numpy as np
 import torch
+from parameterized import parameterized
 
 from monai import transforms
 from monai.losses.image_dissimilarity import GlobalMutualInformationLoss
@@ -116,24 +117,33 @@ def transformation(translate_params=(0.0, 0.0, 0.0), rotate_params=(0.0, 0.0, 0.
 
 class TestGlobalMutualInformationLossIll(unittest.TestCase):
 
-    def test_ill_shape(self):
+    @parameterized.expand(
+        [
+            (torch.ones((1, 2), dtype=torch.float), torch.ones((1, 3), dtype=torch.float)),  # mismatched_simple_dims
+            (
+                torch.ones((1, 3, 3), dtype=torch.float),
+                torch.ones((1, 3), dtype=torch.float),
+            ),  # mismatched_advanced_dims
+        ]
+    )
+    def test_ill_shape(self, input1, input2):
         loss = GlobalMutualInformationLoss()
-        with self.assertRaisesRegex(ValueError, ""):
-            loss.forward(torch.ones((1, 2), dtype=torch.float), torch.ones((1, 3), dtype=torch.float, device=device))
-        with self.assertRaisesRegex(ValueError, ""):
-            loss.forward(torch.ones((1, 3, 3), dtype=torch.float), torch.ones((1, 3), dtype=torch.float, device=device))
-
-    def test_ill_opts(self):
+        with self.assertRaises(ValueError):
+            loss.forward(input1, input2)
+
+    @parameterized.expand(
+        [
+            (0, "mean", ValueError, ""),  # num_bins_zero
+            (-1, "mean", ValueError, ""),  # num_bins_negative
+            (64, "unknown", ValueError, ""),  # reduction_unknown
+            (64, None, ValueError, ""),  # reduction_none
+        ]
+    )
+    def test_ill_opts(self, num_bins, reduction, expected_exception, expected_message):
         pred = torch.ones((1, 3, 3, 3, 3), dtype=torch.float, device=device)
         target = torch.ones((1, 3, 3, 3, 3), dtype=torch.float, device=device)
-        with self.assertRaisesRegex(ValueError, ""):
-            GlobalMutualInformationLoss(num_bins=0)(pred, target)
-        with self.assertRaisesRegex(ValueError, ""):
-            GlobalMutualInformationLoss(num_bins=-1)(pred, target)
-        with self.assertRaisesRegex(ValueError, ""):
-            GlobalMutualInformationLoss(reduction="unknown")(pred, target)
-        with self.assertRaisesRegex(ValueError, ""):
-            GlobalMutualInformationLoss(reduction=None)(pred, target)
+        with self.assertRaisesRegex(expected_exception, expected_message):
+            GlobalMutualInformationLoss(num_bins=num_bins, reduction=reduction)(pred, target)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_hausdorff_loss.py b/tests/test_hausdorff_loss.py
index f279d45b14..f2211008c2 100644
--- a/tests/test_hausdorff_loss.py
+++ b/tests/test_hausdorff_loss.py
@@ -219,17 +219,12 @@ def test_ill_opts(self):
         with self.assertRaisesRegex(ValueError, ""):
             HausdorffDTLoss(reduction=None)(chn_input, chn_target)
 
-    def test_input_warnings(self):
+    @parameterized.expand([(False, False, False), (False, True, False), (False, False, True)])
+    def test_input_warnings(self, include_background, softmax, to_onehot_y):
         chn_input = torch.ones((1, 1, 1, 3))
         chn_target = torch.ones((1, 1, 1, 3))
         with self.assertWarns(Warning):
-            loss = HausdorffDTLoss(include_background=False)
-            loss.forward(chn_input, chn_target)
-        with self.assertWarns(Warning):
-            loss = HausdorffDTLoss(softmax=True)
-            loss.forward(chn_input, chn_target)
-        with self.assertWarns(Warning):
-            loss = HausdorffDTLoss(to_onehot_y=True)
+            loss = HausdorffDTLoss(include_background=include_background, softmax=softmax, to_onehot_y=to_onehot_y)
             loss.forward(chn_input, chn_target)
 
 
@@ -256,17 +251,12 @@ def test_ill_opts(self):
         with self.assertRaisesRegex(ValueError, ""):
             LogHausdorffDTLoss(reduction=None)(chn_input, chn_target)
 
-    def test_input_warnings(self):
+    @parameterized.expand([(False, False, False), (False, True, False), (False, False, True)])
+    def test_input_warnings(self, include_background, softmax, to_onehot_y):
         chn_input = torch.ones((1, 1, 1, 3))
         chn_target = torch.ones((1, 1, 1, 3))
         with self.assertWarns(Warning):
-            loss = LogHausdorffDTLoss(include_background=False)
-            loss.forward(chn_input, chn_target)
-        with self.assertWarns(Warning):
-            loss = LogHausdorffDTLoss(softmax=True)
-            loss.forward(chn_input, chn_target)
-        with self.assertWarns(Warning):
-            loss = LogHausdorffDTLoss(to_onehot_y=True)
+            loss = LogHausdorffDTLoss(include_background=include_background, softmax=softmax, to_onehot_y=to_onehot_y)
             loss.forward(chn_input, chn_target)
 
 
diff --git a/tests/test_median_filter.py b/tests/test_median_filter.py
index 1f5e623260..516388afce 100644
--- a/tests/test_median_filter.py
+++ b/tests/test_median_filter.py
@@ -15,27 +15,20 @@
 
 import numpy as np
 import torch
+from parameterized import parameterized
 
 from monai.networks.layers import MedianFilter
 
 
 class MedianFilterTestCase(unittest.TestCase):
+    @parameterized.expand([(torch.ones(1, 1, 2, 3, 5), [1, 2, 4]), (torch.ones(1, 1, 4, 3, 4), 1)])  # 3d_big  # 3d
+    def test_3d(self, input_tensor, radius):
+        filter = MedianFilter(radius).to(torch.device("cpu:0"))
 
-    def test_3d_big(self):
-        a = torch.ones(1, 1, 2, 3, 5)
-        g = MedianFilter([1, 2, 4]).to(torch.device("cpu:0"))
+        expected = input_tensor.numpy()
+        output = filter(input_tensor).cpu().numpy()
 
-        expected = a.numpy()
-        out = g(a).cpu().numpy()
-        np.testing.assert_allclose(out, expected, rtol=1e-5)
-
-    def test_3d(self):
-        a = torch.ones(1, 1, 4, 3, 4)
-        g = MedianFilter(1).to(torch.device("cpu:0"))
-
-        expected = a.numpy()
-        out = g(a).cpu().numpy()
-        np.testing.assert_allclose(out, expected, rtol=1e-5)
+        np.testing.assert_allclose(output, expected, rtol=1e-5)
 
     def test_3d_radii(self):
         a = torch.ones(1, 1, 4, 3, 2)
diff --git a/tests/test_multi_scale.py b/tests/test_multi_scale.py
index 6681f266a8..0b49087216 100644
--- a/tests/test_multi_scale.py
+++ b/tests/test_multi_scale.py
@@ -58,17 +58,24 @@ def test_shape(self, input_param, input_data, expected_val):
         result = MultiScaleLoss(**input_param).forward(**input_data)
         np.testing.assert_allclose(result.detach().cpu().numpy(), expected_val, rtol=1e-5)
 
-    def test_ill_opts(self):
-        with self.assertRaisesRegex(ValueError, ""):
-            MultiScaleLoss(loss=dice_loss, kernel="none")
-        with self.assertRaisesRegex(ValueError, ""):
-            MultiScaleLoss(loss=dice_loss, scales=[-1])(
-                torch.ones((1, 1, 3), device=device), torch.ones((1, 1, 3), device=device)
-            )
-        with self.assertRaisesRegex(ValueError, ""):
-            MultiScaleLoss(loss=dice_loss, scales=[-1], reduction="none")(
-                torch.ones((1, 1, 3), device=device), torch.ones((1, 1, 3), device=device)
-            )
+    @parameterized.expand(
+        [
+            ({"loss": dice_loss, "kernel": "none"}, None, None),  # kernel_none
+            ({"loss": dice_loss, "scales": [-1]}, torch.ones((1, 1, 3)), torch.ones((1, 1, 3))),  # scales_negative
+            (
+                {"loss": dice_loss, "scales": [-1], "reduction": "none"},
+                torch.ones((1, 1, 3)),
+                torch.ones((1, 1, 3)),
+            ),  # scales_negative_reduction_none
+        ]
+    )
+    def test_ill_opts(self, kwargs, input, target):
+        if input is None and target is None:
+            with self.assertRaisesRegex(ValueError, ""):
+                MultiScaleLoss(**kwargs)
+        else:
+            with self.assertRaisesRegex(ValueError, ""):
+                MultiScaleLoss(**kwargs)(input, target)
 
     def test_script(self):
         input_param, input_data, expected_val = TEST_CASES[0]
diff --git a/tests/test_optional_import.py b/tests/test_optional_import.py
index e7e1c03fd0..2f640f88d0 100644
--- a/tests/test_optional_import.py
+++ b/tests/test_optional_import.py
@@ -13,22 +13,20 @@
 
 import unittest
 
+from parameterized import parameterized
+
 from monai.utils import OptionalImportError, exact_version, optional_import
 
 
 class TestOptionalImport(unittest.TestCase):
 
-    def test_default(self):
-        my_module, flag = optional_import("not_a_module")
+    @parameterized.expand(["not_a_module", "torch.randint"])
+    def test_default(self, import_module):
+        my_module, flag = optional_import(import_module)
         self.assertFalse(flag)
         with self.assertRaises(OptionalImportError):
             my_module.test
 
-        my_module, flag = optional_import("torch.randint")
-        with self.assertRaises(OptionalImportError):
-            self.assertFalse(flag)
-            print(my_module.test)
-
     def test_import_valid(self):
         my_module, flag = optional_import("torch")
         self.assertTrue(flag)
@@ -47,18 +45,9 @@ def test_import_wrong_number(self):
             self.assertTrue(flag)
             print(my_module.randint(1, 2, (1, 2)))
 
-    def test_import_good_number(self):
-        my_module, flag = optional_import("torch", "0")
-        my_module.nn
-        self.assertTrue(flag)
-        print(my_module.randint(1, 2, (1, 2)))
-
-        my_module, flag = optional_import("torch", "0.0.0.1")
-        my_module.nn
-        self.assertTrue(flag)
-        print(my_module.randint(1, 2, (1, 2)))
-
-        my_module, flag = optional_import("torch", "1.1.0")
+    @parameterized.expand(["0", "0.0.0.1", "1.1.0"])
+    def test_import_good_number(self, version_number):
+        my_module, flag = optional_import("torch", version_number)
         my_module.nn
         self.assertTrue(flag)
         print(my_module.randint(1, 2, (1, 2)))
diff --git a/tests/test_perceptual_loss.py b/tests/test_perceptual_loss.py
index 548613e2ae..b8aa2e5982 100644
--- a/tests/test_perceptual_loss.py
+++ b/tests/test_perceptual_loss.py
@@ -113,12 +113,10 @@ def test_1d(self):
         with self.assertRaises(NotImplementedError):
             PerceptualLoss(spatial_dims=1)
 
-    def test_medicalnet_on_2d_data(self):
+    @parameterized.expand(["medicalnet_resnet10_23datasets", "medicalnet_resnet50_23datasets"])
+    def test_medicalnet_on_2d_data(self, network_type):
         with self.assertRaises(ValueError):
-            PerceptualLoss(spatial_dims=2, network_type="medicalnet_resnet10_23datasets")
-
-        with self.assertRaises(ValueError):
-            PerceptualLoss(spatial_dims=2, network_type="medicalnet_resnet50_23datasets")
+            PerceptualLoss(spatial_dims=2, network_type=network_type)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_prepare_batch_default.py b/tests/test_prepare_batch_default.py
index d5a5fbf57e..9aa498866f 100644
--- a/tests/test_prepare_batch_default.py
+++ b/tests/test_prepare_batch_default.py
@@ -14,6 +14,7 @@
 import unittest
 
 import torch
+from parameterized import parameterized
 
 from monai.engines import PrepareBatchDefault, SupervisedEvaluator
 from tests.utils import assert_allclose
@@ -27,85 +28,48 @@ def forward(self, x: torch.Tensor):
 
 class TestPrepareBatchDefault(unittest.TestCase):
 
-    def test_dict_content(self):
-        device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-        dataloader = [
-            {
-                "image": torch.tensor([1, 2]),
-                "label": torch.tensor([3, 4]),
-                "extra1": torch.tensor([5, 6]),
-                "extra2": 16,
-                "extra3": "test",
-            }
+    @parameterized.expand(
+        [
+            (
+                [
+                    {
+                        "image": torch.tensor([1, 2]),
+                        "label": torch.tensor([3, 4]),
+                        "extra1": torch.tensor([5, 6]),
+                        "extra2": 16,
+                        "extra3": "test",
+                    }
+                ],
+                TestNet(),
+                True,
+            ),  # dict_content
+            ([torch.tensor([1, 2])], torch.nn.Identity(), True),  # tensor_content
+            ([(torch.tensor([1, 2]), torch.tensor([3, 4]))], torch.nn.Identity(), True),  # pair_content
+            ([], TestNet(), False),  # empty_data
         ]
-        # set up engine
-        evaluator = SupervisedEvaluator(
-            device=device,
-            val_data_loader=dataloader,
-            epoch_length=1,
-            network=TestNet(),
-            non_blocking=False,
-            prepare_batch=PrepareBatchDefault(),
-            decollate=False,
-            mode="eval",
-        )
-        evaluator.run()
-        output = evaluator.state.output
-        assert_allclose(output["image"], torch.tensor([1, 2], device=device))
-        assert_allclose(output["label"], torch.tensor([3, 4], device=device))
-
-    def test_tensor_content(self):
+    )
+    def test_prepare_batch(self, dataloader, network, should_run):
         device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-        dataloader = [torch.tensor([1, 2])]
-
-        # set up engine
         evaluator = SupervisedEvaluator(
             device=device,
             val_data_loader=dataloader,
-            epoch_length=1,
-            network=torch.nn.Identity(),
+            epoch_length=len(dataloader) if should_run else 0,
+            network=network,
             non_blocking=False,
             prepare_batch=PrepareBatchDefault(),
             decollate=False,
-            mode="eval",
+            mode="eval" if should_run else "train",
         )
         evaluator.run()
-        output = evaluator.state.output
-        assert_allclose(output["image"], torch.tensor([1, 2], device=device))
-        self.assertTrue(output["label"] is None)
 
-    def test_pair_content(self):
-        device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-        dataloader = [(torch.tensor([1, 2]), torch.tensor([3, 4]))]
-
-        # set up engine
-        evaluator = SupervisedEvaluator(
-            device=device,
-            val_data_loader=dataloader,
-            epoch_length=1,
-            network=torch.nn.Identity(),
-            non_blocking=False,
-            prepare_batch=PrepareBatchDefault(),
-            decollate=False,
-            mode="eval",
-        )
-        evaluator.run()
-        output = evaluator.state.output
-        assert_allclose(output["image"], torch.tensor([1, 2], device=device))
-        assert_allclose(output["label"], torch.tensor([3, 4], device=device))
-
-    def test_empty_data(self):
-        dataloader = []
-        evaluator = SupervisedEvaluator(
-            val_data_loader=dataloader,
-            device=torch.device("cpu"),
-            epoch_length=0,
-            network=TestNet(),
-            non_blocking=False,
-            prepare_batch=PrepareBatchDefault(),
-            decollate=False,
-        )
-        evaluator.run()
+        if should_run:
+            output = evaluator.state.output
+            if isinstance(dataloader[0], dict) or isinstance(dataloader[0], tuple):
+                assert_allclose(output["image"], torch.tensor([1, 2], device=device))
+                assert_allclose(output["label"], torch.tensor([3, 4], device=device))
+            else:
+                assert_allclose(output["image"], torch.tensor([1, 2], device=device))
+                self.assertTrue(output["label"] is None)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_rand_affine.py b/tests/test_rand_affine.py
index 23e3fd148c..2c827b7426 100644
--- a/tests/test_rand_affine.py
+++ b/tests/test_rand_affine.py
@@ -152,11 +152,10 @@ def test_rand_affine(self, input_param, input_data, expected_val):
             self.assertTrue(g._cached_grid is not None)
         assert_allclose(result, expected_val, rtol=_rtol, atol=1e-4, type_test="tensor")
 
-    def test_ill_cache(self):
+    @parameterized.expand([(None,), ((1, 1, -1),)])
+    def test_ill_cache(self, spatial_size):
         with self.assertWarns(UserWarning):
-            RandAffine(cache_grid=True)
-        with self.assertWarns(UserWarning):
-            RandAffine(cache_grid=True, spatial_size=(1, 1, -1))
+            RandAffine(cache_grid=True, spatial_size=spatial_size)
 
     @parameterized.expand(TEST_CASES_SKIPPED_CONSISTENCY)
     def test_skipped_transform_consistency(self, im, in_dtype):
diff --git a/tests/test_rand_affined.py b/tests/test_rand_affined.py
index 32fde8dc0f..950058a9e9 100644
--- a/tests/test_rand_affined.py
+++ b/tests/test_rand_affined.py
@@ -272,13 +272,10 @@ def test_rand_affined(self, input_param, input_data, expected_val, track_meta):
             self.assertEqual(len(v.applied_operations), 0)
             self.assertTupleEqual(v.shape, input_data[k].shape)
 
-    def test_ill_cache(self):
+    @parameterized.expand([(None,), ((2, -1),)])  # spatial size is None  # spatial size is dynamic
+    def test_ill_cache(self, spatial_size):
         with self.assertWarns(UserWarning):
-            # spatial size is None
-            RandAffined(device=device, spatial_size=None, prob=1.0, cache_grid=True, keys=("img", "seg"))
-        with self.assertWarns(UserWarning):
-            # spatial size is dynamic
-            RandAffined(device=device, spatial_size=(2, -1), prob=1.0, cache_grid=True, keys=("img", "seg"))
+            RandAffined(device=device, spatial_size=spatial_size, prob=1.0, cache_grid=True, keys=("img", "seg"))
 
 
 if __name__ == "__main__":
diff --git a/tests/test_tversky_loss.py b/tests/test_tversky_loss.py
index efe1f2cdf3..0365503ea2 100644
--- a/tests/test_tversky_loss.py
+++ b/tests/test_tversky_loss.py
@@ -165,17 +165,12 @@ def test_ill_shape(self):
         with self.assertRaisesRegex(ValueError, ""):
             TverskyLoss(reduction=None)(chn_input, chn_target)
 
-    def test_input_warnings(self):
+    @parameterized.expand([(False, False, False), (False, True, False), (False, False, True)])
+    def test_input_warnings(self, include_background, softmax, to_onehot_y):
         chn_input = torch.ones((1, 1, 3))
         chn_target = torch.ones((1, 1, 3))
         with self.assertWarns(Warning):
-            loss = TverskyLoss(include_background=False)
-            loss.forward(chn_input, chn_target)
-        with self.assertWarns(Warning):
-            loss = TverskyLoss(softmax=True)
-            loss.forward(chn_input, chn_target)
-        with self.assertWarns(Warning):
-            loss = TverskyLoss(to_onehot_y=True)
+            loss = TverskyLoss(include_background=include_background, softmax=softmax, to_onehot_y=to_onehot_y)
             loss.forward(chn_input, chn_target)
 
     def test_script(self):
diff --git a/tests/test_ultrasound_confidence_map_transform.py b/tests/test_ultrasound_confidence_map_transform.py
index f672961700..63ce7d58e4 100644
--- a/tests/test_ultrasound_confidence_map_transform.py
+++ b/tests/test_ultrasound_confidence_map_transform.py
@@ -15,6 +15,7 @@
 
 import numpy as np
 import torch
+from parameterized import parameterized
 
 from monai.transforms import UltrasoundConfidenceMapTransform
 from tests.utils import assert_allclose
@@ -535,162 +536,92 @@ def test_parameters(self):
         with self.assertRaises(ValueError):
             UltrasoundConfidenceMapTransform(sink_mode="unknown")
 
-    def test_rgb(self):
+    @parameterized.expand(
+        [("all", SINK_ALL_OUTPUT), ("mid", SINK_MID_OUTPUT), ("min", SINK_MIN_OUTPUT), ("mask", SINK_MASK_OUTPUT, True)]
+    )
+    def test_ultrasound_confidence_map_transform(self, sink_mode, expected_output, use_mask=False):
         # RGB image
         input_img_rgb = np.expand_dims(np.repeat(self.input_img_np, 3, axis=0), axis=0)
         input_img_rgb_torch = torch.from_numpy(input_img_rgb)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="all")
-        result_torch = transform(input_img_rgb_torch)
-        self.assertIsInstance(result_torch, torch.Tensor)
-        assert_allclose(result_torch, torch.tensor(SINK_ALL_OUTPUT), rtol=1e-4, atol=1e-4)
-        result_np = transform(input_img_rgb)
-        self.assertIsInstance(result_np, np.ndarray)
-        assert_allclose(result_np, SINK_ALL_OUTPUT, rtol=1e-4, atol=1e-4)
+        transform = UltrasoundConfidenceMapTransform(sink_mode=sink_mode)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="mid")
-        result_torch = transform(input_img_rgb_torch)
-        self.assertIsInstance(result_torch, torch.Tensor)
-        assert_allclose(result_torch, torch.tensor(SINK_MID_OUTPUT), rtol=1e-4, atol=1e-4)
-        result_np = transform(input_img_rgb)
-        self.assertIsInstance(result_np, np.ndarray)
-        assert_allclose(result_np, SINK_MID_OUTPUT, rtol=1e-4, atol=1e-4)
+        if use_mask:
+            result_torch = transform(input_img_rgb_torch, self.input_mask_torch)
+            result_np = transform(input_img_rgb, self.input_mask_np)
+        else:
+            result_torch = transform(input_img_rgb_torch)
+            result_np = transform(input_img_rgb)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="min")
-        result_torch = transform(input_img_rgb_torch)
         self.assertIsInstance(result_torch, torch.Tensor)
-        assert_allclose(result_torch, torch.tensor(SINK_MIN_OUTPUT), rtol=1e-4, atol=1e-4)
-        result_np = transform(input_img_rgb)
+        assert_allclose(result_torch, torch.tensor(expected_output), rtol=1e-4, atol=1e-4)
         self.assertIsInstance(result_np, np.ndarray)
-        assert_allclose(result_np, SINK_MIN_OUTPUT, rtol=1e-4, atol=1e-4)
+        assert_allclose(result_np, expected_output, rtol=1e-4, atol=1e-4)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="mask")
-        result_torch = transform(input_img_rgb_torch, self.input_mask_torch)
-        self.assertIsInstance(result_torch, torch.Tensor)
-        assert_allclose(result_torch, torch.tensor(SINK_MASK_OUTPUT), rtol=1e-4, atol=1e-4)
-        result_np = transform(input_img_rgb, self.input_mask_np)
-        self.assertIsInstance(result_np, np.ndarray)
-        assert_allclose(result_np, SINK_MASK_OUTPUT, rtol=1e-4, atol=1e-4)
-
-    def test_multi_channel_2d(self):
-        # 2D multi-channel image
+    @parameterized.expand(
+        [
+            ("all", SINK_ALL_OUTPUT),
+            ("mid", SINK_MID_OUTPUT),
+            ("min", SINK_MIN_OUTPUT),
+            ("mask", SINK_MASK_OUTPUT, True),  # Adding a flag for mask cases
+        ]
+    )
+    def test_multi_channel_2d(self, sink_mode, expected_output, use_mask=False):
         input_img_rgb = np.expand_dims(np.repeat(self.input_img_np, 17, axis=0), axis=0)
         input_img_rgb_torch = torch.from_numpy(input_img_rgb)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="all")
-        result_torch = transform(input_img_rgb_torch)
-        self.assertIsInstance(result_torch, torch.Tensor)
-        assert_allclose(result_torch, torch.tensor(SINK_ALL_OUTPUT), rtol=1e-4, atol=1e-4)
-        result_np = transform(input_img_rgb)
-        self.assertIsInstance(result_np, np.ndarray)
-        assert_allclose(result_np, SINK_ALL_OUTPUT, rtol=1e-4, atol=1e-4)
-
-        transform = UltrasoundConfidenceMapTransform(sink_mode="mid")
-        result_torch = transform(input_img_rgb_torch)
-        self.assertIsInstance(result_torch, torch.Tensor)
-        assert_allclose(result_torch, torch.tensor(SINK_MID_OUTPUT), rtol=1e-4, atol=1e-4)
-        result_np = transform(input_img_rgb)
-        self.assertIsInstance(result_np, np.ndarray)
-        assert_allclose(result_np, SINK_MID_OUTPUT, rtol=1e-4, atol=1e-4)
+        transform = UltrasoundConfidenceMapTransform(sink_mode=sink_mode)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="min")
-        result_torch = transform(input_img_rgb_torch)
-        self.assertIsInstance(result_torch, torch.Tensor)
-        assert_allclose(result_torch, torch.tensor(SINK_MIN_OUTPUT), rtol=1e-4, atol=1e-4)
-        result_np = transform(input_img_rgb)
-        self.assertIsInstance(result_np, np.ndarray)
-        assert_allclose(result_np, SINK_MIN_OUTPUT, rtol=1e-4, atol=1e-4)
+        if use_mask:
+            result_torch = transform(input_img_rgb_torch, self.input_mask_torch)
+            result_np = transform(input_img_rgb, self.input_mask_np)
+        else:
+            result_torch = transform(input_img_rgb_torch)
+            result_np = transform(input_img_rgb)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="mask")
-        result_torch = transform(input_img_rgb_torch, self.input_mask_torch)
         self.assertIsInstance(result_torch, torch.Tensor)
-        assert_allclose(result_torch, torch.tensor(SINK_MASK_OUTPUT), rtol=1e-4, atol=1e-4)
-        result_np = transform(input_img_rgb, self.input_mask_np)
+        assert_allclose(result_torch, torch.tensor(expected_output), rtol=1e-4, atol=1e-4)
         self.assertIsInstance(result_np, np.ndarray)
-        assert_allclose(result_np, SINK_MASK_OUTPUT, rtol=1e-4, atol=1e-4)
+        assert_allclose(result_np, expected_output, rtol=1e-4, atol=1e-4)
 
-    def test_non_one_first_dim(self):
-        # Image without first dimension as 1
+    @parameterized.expand([("all",), ("mid",), ("min",), ("mask",)])
+    def test_non_one_first_dim(self, sink_mode):
+        transform = UltrasoundConfidenceMapTransform(sink_mode=sink_mode)
         input_img_rgb = np.repeat(self.input_img_np, 3, axis=0)
         input_img_rgb_torch = torch.from_numpy(input_img_rgb)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="all")
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb_torch)
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb)
-
-        transform = UltrasoundConfidenceMapTransform(sink_mode="mid")
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb_torch)
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb)
-
-        transform = UltrasoundConfidenceMapTransform(sink_mode="min")
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb_torch)
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb)
-
-        transform = UltrasoundConfidenceMapTransform(sink_mode="mask")
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb_torch, self.input_mask_torch)
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb, self.input_mask_np)
-
-    def test_no_first_dim(self):
-        # Image without first dimension
+        if sink_mode == "mask":
+            with self.assertRaises(ValueError):
+                transform(input_img_rgb_torch, self.input_mask_torch)
+            with self.assertRaises(ValueError):
+                transform(input_img_rgb, self.input_mask_np)
+        else:
+            with self.assertRaises(ValueError):
+                transform(input_img_rgb_torch)
+            with self.assertRaises(ValueError):
+                transform(input_img_rgb)
+
+    @parameterized.expand([("all",), ("mid",), ("min",), ("mask",)])
+    def test_no_first_dim(self, sink_mode):
         input_img_rgb = self.input_img_np[0]
         input_img_rgb_torch = torch.from_numpy(input_img_rgb)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="all")
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb_torch)
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb)
+        transform = UltrasoundConfidenceMapTransform(sink_mode=sink_mode)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="mid")
         with self.assertRaises(ValueError):
             transform(input_img_rgb_torch)
         with self.assertRaises(ValueError):
             transform(input_img_rgb)
 
-        transform = UltrasoundConfidenceMapTransform(sink_mode="min")
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb_torch)
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb)
-
-        transform = UltrasoundConfidenceMapTransform(sink_mode="mask")
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb_torch, self.input_mask_torch)
-        with self.assertRaises(ValueError):
-            transform(input_img_rgb, self.input_mask_np)
-
-    def test_sink_all(self):
-        transform = UltrasoundConfidenceMapTransform(sink_mode="all")
-
-        # This should not raise an exception for torch tensor
-        result_torch = transform(self.input_img_torch)
-        self.assertIsInstance(result_torch, torch.Tensor)
-
-        # This should not raise an exception for numpy array
-        result_np = transform(self.input_img_np)
-        self.assertIsInstance(result_np, np.ndarray)
-
-    def test_sink_mid(self):
-        transform = UltrasoundConfidenceMapTransform(sink_mode="mid")
-
-        # This should not raise an exception for torch tensor
-        result_torch = transform(self.input_img_torch)
-        self.assertIsInstance(result_torch, torch.Tensor)
-
-        # This should not raise an exception for numpy array
-        result_np = transform(self.input_img_np)
-        self.assertIsInstance(result_np, np.ndarray)
+        if sink_mode == "mask":
+            with self.assertRaises(ValueError):
+                transform(input_img_rgb_torch, self.input_mask_torch)
+            with self.assertRaises(ValueError):
+                transform(input_img_rgb, self.input_mask_np)
 
-    def test_sink_min(self):
-        transform = UltrasoundConfidenceMapTransform(sink_mode="min")
+    @parameterized.expand([("all",), ("mid",), ("min",)])
+    def test_sink_mode(self, mode):
+        transform = UltrasoundConfidenceMapTransform(sink_mode=mode)
 
         # This should not raise an exception for torch tensor
         result_torch = transform(self.input_img_torch)
diff --git a/tests/test_vit.py b/tests/test_vit.py
index a84883cba0..d27c10f95e 100644
--- a/tests/test_vit.py
+++ b/tests/test_vit.py
@@ -69,75 +69,40 @@ def test_shape(self, input_param, input_shape, expected_shape):
             result, _ = net(torch.randn(input_shape))
             self.assertEqual(result.shape, expected_shape)
 
-    def test_ill_arg(self):
+    @parameterized.expand(
+        [
+            (1, (128, 128, 128), (16, 16, 16), 128, 3072, 12, 12, "conv", False, 5.0),
+            (1, (32, 32, 32), (64, 64, 64), 512, 3072, 12, 8, "perceptron", False, 0.3),
+            (1, (96, 96, 96), (8, 8, 8), 512, 3072, 12, 14, "conv", False, 0.3),
+            (1, (97, 97, 97), (4, 4, 4), 768, 3072, 12, 8, "perceptron", True, 0.3),
+            (4, (96, 96, 96), (16, 16, 16), 768, 3072, 12, 12, "perc", False, 0.3),
+        ]
+    )
+    def test_ill_arg(
+        self,
+        in_channels,
+        img_size,
+        patch_size,
+        hidden_size,
+        mlp_dim,
+        num_layers,
+        num_heads,
+        pos_embed,
+        classification,
+        dropout_rate,
+    ):
         with self.assertRaises(ValueError):
             ViT(
-                in_channels=1,
-                img_size=(128, 128, 128),
-                patch_size=(16, 16, 16),
-                hidden_size=128,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=12,
-                pos_embed="conv",
-                classification=False,
-                dropout_rate=5.0,
-            )
-
-        with self.assertRaises(ValueError):
-            ViT(
-                in_channels=1,
-                img_size=(32, 32, 32),
-                patch_size=(64, 64, 64),
-                hidden_size=512,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=8,
-                pos_embed="perceptron",
-                classification=False,
-                dropout_rate=0.3,
-            )
-
-        with self.assertRaises(ValueError):
-            ViT(
-                in_channels=1,
-                img_size=(96, 96, 96),
-                patch_size=(8, 8, 8),
-                hidden_size=512,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=14,
-                pos_embed="conv",
-                classification=False,
-                dropout_rate=0.3,
-            )
-
-        with self.assertRaises(ValueError):
-            ViT(
-                in_channels=1,
-                img_size=(97, 97, 97),
-                patch_size=(4, 4, 4),
-                hidden_size=768,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=8,
-                pos_embed="perceptron",
-                classification=True,
-                dropout_rate=0.3,
-            )
-
-        with self.assertRaises(ValueError):
-            ViT(
-                in_channels=4,
-                img_size=(96, 96, 96),
-                patch_size=(16, 16, 16),
-                hidden_size=768,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=12,
-                pos_embed="perc",
-                classification=False,
-                dropout_rate=0.3,
+                in_channels=in_channels,
+                img_size=img_size,
+                patch_size=patch_size,
+                hidden_size=hidden_size,
+                mlp_dim=mlp_dim,
+                num_layers=num_layers,
+                num_heads=num_heads,
+                pos_embed=pos_embed,
+                classification=classification,
+                dropout_rate=dropout_rate,
             )
 
     @parameterized.expand(TEST_CASE_Vit)
diff --git a/tests/test_vitautoenc.py b/tests/test_vitautoenc.py
index cc3d493bb3..c68c583a0e 100644
--- a/tests/test_vitautoenc.py
+++ b/tests/test_vitautoenc.py
@@ -82,83 +82,30 @@ def test_shape(self, input_param, input_shape, expected_shape):
             result, _ = net(torch.randn(input_shape))
             self.assertEqual(result.shape, expected_shape)
 
-    def test_ill_arg(self):
+    @parameterized.expand(
+        [
+            (1, (32, 32, 32), (64, 64, 64), 512, 3072, 12, 8, "perceptron", 0.3),  # img_size_too_large_for_patch_size
+            (1, (96, 96, 96), (8, 8, 8), 512, 3072, 12, 14, "conv", 0.3),  # num_heads_out_of_bound
+            (1, (97, 97, 97), (4, 4, 4), 768, 3072, 12, 8, "perceptron", 0.3),  # img_size_not_divisible_by_patch_size
+            (4, (96, 96, 96), (16, 16, 16), 768, 3072, 12, 12, "perc", 0.3),  # invalid_pos_embed
+            (4, (96, 96, 96), (9, 9, 9), 768, 3072, 12, 12, "perc", 0.3),  # patch_size_not_divisible
+            # Add more test cases as needed
+        ]
+    )
+    def test_ill_arg(
+        self, in_channels, img_size, patch_size, hidden_size, mlp_dim, num_layers, num_heads, pos_embed, dropout_rate
+    ):
         with self.assertRaises(ValueError):
             ViTAutoEnc(
-                in_channels=1,
-                img_size=(128, 128, 128),
-                patch_size=(16, 16, 16),
-                hidden_size=128,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=12,
-                pos_embed="conv",
-                dropout_rate=5.0,
-            )
-
-        with self.assertRaises(ValueError):
-            ViTAutoEnc(
-                in_channels=1,
-                img_size=(32, 32, 32),
-                patch_size=(64, 64, 64),
-                hidden_size=512,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=8,
-                pos_embed="perceptron",
-                dropout_rate=0.3,
-            )
-
-        with self.assertRaises(ValueError):
-            ViTAutoEnc(
-                in_channels=1,
-                img_size=(96, 96, 96),
-                patch_size=(8, 8, 8),
-                hidden_size=512,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=14,
-                pos_embed="conv",
-                dropout_rate=0.3,
-            )
-
-        with self.assertRaises(ValueError):
-            ViTAutoEnc(
-                in_channels=1,
-                img_size=(97, 97, 97),
-                patch_size=(4, 4, 4),
-                hidden_size=768,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=8,
-                pos_embed="perceptron",
-                dropout_rate=0.3,
-            )
-
-        with self.assertRaises(ValueError):
-            ViTAutoEnc(
-                in_channels=4,
-                img_size=(96, 96, 96),
-                patch_size=(16, 16, 16),
-                hidden_size=768,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=12,
-                pos_embed="perc",
-                dropout_rate=0.3,
-            )
-
-        with self.assertRaises(ValueError):
-            ViTAutoEnc(
-                in_channels=4,
-                img_size=(96, 96, 96),
-                patch_size=(9, 9, 9),
-                hidden_size=768,
-                mlp_dim=3072,
-                num_layers=12,
-                num_heads=12,
-                pos_embed="perc",
-                dropout_rate=0.3,
+                in_channels=in_channels,
+                img_size=img_size,
+                patch_size=patch_size,
+                hidden_size=hidden_size,
+                mlp_dim=mlp_dim,
+                num_layers=num_layers,
+                num_heads=num_heads,
+                pos_embed=pos_embed,
+                dropout_rate=dropout_rate,
             )
 
 

From dc58e5c2b7997e8cefdef4698e342937fccea956 Mon Sep 17 00:00:00 2001
From: Konstantin Sukharev <50718389+k-sukharev@users.noreply.github.com>
Date: Tue, 23 Apr 2024 11:30:21 +0500
Subject: [PATCH 033/183] Add ResNet backbones for FlexibleUNet (#7571)

Fixes #7570.

### Description

Add ResNet backbones (with option to use pretrained Med3D weights) for
FlexibleUNet.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Konstantin Sukharev <k.sukharev@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/networks.rst             |   5 +
 monai/networks/nets/__init__.py      |   2 +
 monai/networks/nets/flexible_unet.py |  13 ++-
 monai/networks/nets/resnet.py        | 145 ++++++++++++++++++++++++++-
 tests/test_flexible_unet.py          | 135 +++++--------------------
 tests/test_resnet.py                 |  39 ++++++-
 6 files changed, 222 insertions(+), 117 deletions(-)

diff --git a/docs/source/networks.rst b/docs/source/networks.rst
index b59c8af5fc..249375dfc1 100644
--- a/docs/source/networks.rst
+++ b/docs/source/networks.rst
@@ -491,6 +491,11 @@ Nets
 .. autoclass:: ResNet
   :members:
 
+`ResNetFeatures`
+~~~~~~~~~~~~~~~~
+.. autoclass:: ResNetFeatures
+  :members:
+
 `SENet`
 ~~~~~~~
 .. autoclass:: SENet
diff --git a/monai/networks/nets/__init__.py b/monai/networks/nets/__init__.py
index 9247aaee85..de5d1adc7e 100644
--- a/monai/networks/nets/__init__.py
+++ b/monai/networks/nets/__init__.py
@@ -59,6 +59,8 @@
     ResNet,
     ResNetBlock,
     ResNetBottleneck,
+    ResNetEncoder,
+    ResNetFeatures,
     get_medicalnet_pretrained_resnet_args,
     get_pretrained_resnet_medicalnet,
     resnet10,
diff --git a/monai/networks/nets/flexible_unet.py b/monai/networks/nets/flexible_unet.py
index ac2124b5f9..c27b0fc17b 100644
--- a/monai/networks/nets/flexible_unet.py
+++ b/monai/networks/nets/flexible_unet.py
@@ -24,6 +24,7 @@
 from monai.networks.layers.utils import get_act_layer
 from monai.networks.nets import EfficientNetEncoder
 from monai.networks.nets.basic_unet import UpCat
+from monai.networks.nets.resnet import ResNetEncoder
 from monai.utils import InterpolateMode, optional_import
 
 __all__ = ["FlexibleUNet", "FlexUNet", "FLEXUNET_BACKBONE", "FlexUNetEncoderRegister"]
@@ -78,6 +79,7 @@ def register_class(self, name: type[Any] | str):
 
 FLEXUNET_BACKBONE = FlexUNetEncoderRegister()
 FLEXUNET_BACKBONE.register_class(EfficientNetEncoder)
+FLEXUNET_BACKBONE.register_class(ResNetEncoder)
 
 
 class UNetDecoder(nn.Module):
@@ -238,7 +240,7 @@ def __init__(
     ) -> None:
         """
         A flexible implement of UNet, in which the backbone/encoder can be replaced with
-        any efficient network. Currently the input must have a 2 or 3 spatial dimension
+        any efficient or residual network. Currently the input must have a 2 or 3 spatial dimension
         and the spatial size of each dimension must be a multiple of 32 if is_pad parameter
         is False.
         Please notice each output of backbone must be 2x downsample in spatial dimension
@@ -248,10 +250,11 @@ def __init__(
         Args:
             in_channels: number of input channels.
             out_channels: number of output channels.
-            backbone: name of backbones to initialize, only support efficientnet right now,
-                can be from [efficientnet-b0,..., efficientnet-b8, efficientnet-l2].
-            pretrained: whether to initialize pretrained ImageNet weights, only available
-                for spatial_dims=2 and batch norm is used, default to False.
+            backbone: name of backbones to initialize, only support efficientnet and resnet right now,
+                can be from [efficientnet-b0, ..., efficientnet-b8, efficientnet-l2, resnet10, ..., resnet200].
+            pretrained: whether to initialize pretrained weights. ImageNet weights are available for efficient networks
+                if spatial_dims=2 and batch norm is used. MedicalNet weights are available for residual networks
+                if spatial_dims=3 and in_channels=1. Default to False.
             decoder_channels: number of output channels for all feature maps in decoder.
                 `len(decoder_channels)` should equal to `len(encoder_channels) - 1`,default
                 to (256, 128, 64, 32, 16).
diff --git a/monai/networks/nets/resnet.py b/monai/networks/nets/resnet.py
index 34a4b7057e..99975271da 100644
--- a/monai/networks/nets/resnet.py
+++ b/monai/networks/nets/resnet.py
@@ -21,6 +21,7 @@
 import torch
 import torch.nn as nn
 
+from monai.networks.blocks.encoder import BaseEncoder
 from monai.networks.layers.factories import Conv, Norm, Pool
 from monai.networks.layers.utils import get_pool_layer
 from monai.utils import ensure_tuple_rep
@@ -45,6 +46,19 @@
     "resnet200",
 ]
 
+
+resnet_params = {
+    # model_name: (block, layers, shortcut_type, bias_downsample, datasets23)
+    "resnet10": ("basic", [1, 1, 1, 1], "B", False, True),
+    "resnet18": ("basic", [2, 2, 2, 2], "A", True, True),
+    "resnet34": ("basic", [3, 4, 6, 3], "A", True, True),
+    "resnet50": ("bottleneck", [3, 4, 6, 3], "B", False, True),
+    "resnet101": ("bottleneck", [3, 4, 23, 3], "B", False, False),
+    "resnet152": ("bottleneck", [3, 8, 36, 3], "B", False, False),
+    "resnet200": ("bottleneck", [3, 24, 36, 3], "B", False, False),
+}
+
+
 logger = logging.getLogger(__name__)
 
 
@@ -335,6 +349,120 @@ def forward(self, x: torch.Tensor) -> torch.Tensor:
         return x
 
 
+class ResNetFeatures(ResNet):
+
+    def __init__(self, model_name: str, pretrained: bool = True, spatial_dims: int = 3, in_channels: int = 1) -> None:
+        """Initialize resnet18 to resnet200 models as a backbone, the backbone can be used as an encoder for
+        segmentation and objection models.
+
+        Compared with the class `ResNet`, the only different place is the forward function.
+
+        Args:
+            model_name: name of model to initialize, can be from [resnet10, ..., resnet200].
+            pretrained: whether to initialize pretrained MedicalNet weights,
+                only available for spatial_dims=3 and in_channels=1.
+            spatial_dims: number of spatial dimensions of the input image.
+            in_channels: number of input channels for first convolutional layer.
+        """
+        if model_name not in resnet_params:
+            model_name_string = ", ".join(resnet_params.keys())
+            raise ValueError(f"invalid model_name {model_name} found, must be one of {model_name_string} ")
+
+        block, layers, shortcut_type, bias_downsample, datasets23 = resnet_params[model_name]
+
+        super().__init__(
+            block=block,
+            layers=layers,
+            block_inplanes=get_inplanes(),
+            spatial_dims=spatial_dims,
+            n_input_channels=in_channels,
+            conv1_t_stride=2,
+            shortcut_type=shortcut_type,
+            feed_forward=False,
+            bias_downsample=bias_downsample,
+        )
+        if pretrained:
+            if spatial_dims == 3 and in_channels == 1:
+                _load_state_dict(self, model_name, datasets23=datasets23)
+            else:
+                raise ValueError("Pretrained resnet models are only available for in_channels=1 and spatial_dims=3.")
+
+    def forward(self, inputs: torch.Tensor):
+        """
+        Args:
+            inputs: input should have spatially N dimensions
+            ``(Batch, in_channels, dim_0[, dim_1, ..., dim_N])``, N is defined by `dimensions`.
+
+        Returns:
+            a list of torch Tensors.
+        """
+        x = self.conv1(inputs)
+        x = self.bn1(x)
+        x = self.relu(x)
+
+        features = []
+        features.append(x)
+
+        if not self.no_max_pool:
+            x = self.maxpool(x)
+
+        x = self.layer1(x)
+        features.append(x)
+
+        x = self.layer2(x)
+        features.append(x)
+
+        x = self.layer3(x)
+        features.append(x)
+
+        x = self.layer4(x)
+        features.append(x)
+
+        return features
+
+
+class ResNetEncoder(ResNetFeatures, BaseEncoder):
+    """Wrap the original resnet to an encoder for flexible-unet."""
+
+    backbone_names = ["resnet10", "resnet18", "resnet34", "resnet50", "resnet101", "resnet152", "resnet200"]
+
+    @classmethod
+    def get_encoder_parameters(cls) -> list[dict]:
+        """Get the initialization parameter for resnet backbones."""
+        parameter_list = []
+        for backbone_name in cls.backbone_names:
+            parameter_list.append(
+                {"model_name": backbone_name, "pretrained": True, "spatial_dims": 3, "in_channels": 1}
+            )
+        return parameter_list
+
+    @classmethod
+    def num_channels_per_output(cls) -> list[tuple[int, ...]]:
+        """Get number of resnet backbone output feature maps channel."""
+        return [
+            (64, 64, 128, 256, 512),
+            (64, 64, 128, 256, 512),
+            (64, 64, 128, 256, 512),
+            (64, 256, 512, 1024, 2048),
+            (64, 256, 512, 1024, 2048),
+            (64, 256, 512, 1024, 2048),
+            (64, 256, 512, 1024, 2048),
+        ]
+
+    @classmethod
+    def num_outputs(cls) -> list[int]:
+        """Get number of resnet backbone output feature maps.
+
+        Since every backbone contains the same 5 output feature maps, the number list should be `[5] * 7`.
+        """
+        return [5] * 7
+
+    @classmethod
+    def get_encoder_names(cls) -> list[str]:
+        """Get names of resnet backbones."""
+        return cls.backbone_names
+
+
 def _resnet(
     arch: str,
     block: type[ResNetBlock | ResNetBottleneck],
@@ -477,7 +605,7 @@ def resnet200(pretrained: bool = False, progress: bool = True, **kwargs: Any) ->
 
 def get_pretrained_resnet_medicalnet(resnet_depth: int, device: str = "cpu", datasets23: bool = True):
     """
-    Donwlad resnet pretrained weights from https://huggingface.co/TencentMedicalNet
+    Download resnet pretrained weights from https://huggingface.co/TencentMedicalNet
 
     Args:
         resnet_depth: depth of the pretrained model. Supported values are 10, 18, 34, 50, 101, 152 and 200
@@ -533,7 +661,7 @@ def get_pretrained_resnet_medicalnet(resnet_depth: int, device: str = "cpu", dat
 def get_medicalnet_pretrained_resnet_args(resnet_depth: int):
     """
     Return correct shortcut_type and bias_downsample
-    for pretrained MedicalNet weights according to resnet depth
+    for pretrained MedicalNet weights according to resnet depth.
     """
     # After testing
     # False: 10, 50, 101, 152, 200
@@ -541,3 +669,16 @@ def get_medicalnet_pretrained_resnet_args(resnet_depth: int):
     bias_downsample = -1 if resnet_depth in [18, 34] else 0  # 18, 10, 34
     shortcut_type = "A" if resnet_depth in [18, 34] else "B"
     return bias_downsample, shortcut_type
+
+
+def _load_state_dict(model: nn.Module, model_name: str, datasets23: bool = True) -> None:
+    search_res = re.search(r"resnet(\d+)", model_name)
+    if search_res:
+        resnet_depth = int(search_res.group(1))
+        datasets23 = model_name.endswith("_23datasets")
+    else:
+        raise ValueError("model_name argument should contain resnet depth. Example: resnet18 or resnet18_23datasets.")
+
+    model_state_dict = get_pretrained_resnet_medicalnet(resnet_depth, device="cpu", datasets23=datasets23)
+    model_state_dict = {key.replace("module.", ""): value for key, value in model_state_dict.items()}
+    model.load_state_dict(model_state_dict)
diff --git a/tests/test_flexible_unet.py b/tests/test_flexible_unet.py
index 404855c9a8..42baa28b71 100644
--- a/tests/test_flexible_unet.py
+++ b/tests/test_flexible_unet.py
@@ -23,12 +23,11 @@
     EfficientNetBNFeatures,
     FlexibleUNet,
     FlexUNetEncoderRegister,
-    ResNet,
-    ResNetBlock,
-    ResNetBottleneck,
+    ResNetEncoder,
+    ResNetFeatures,
 )
 from monai.utils import optional_import
-from tests.utils import skip_if_downloading_fails, skip_if_quick
+from tests.utils import SkipIfNoModule, skip_if_downloading_fails, skip_if_quick
 
 torchvision, has_torchvision = optional_import("torchvision")
 PIL, has_pil = optional_import("PIL")
@@ -59,101 +58,6 @@ def get_encoder_names(cls):
         return ["encoder_wrong_channels", "encoder_no_param1", "encoder_no_param2", "encoder_no_param3"]
 
 
-class ResNetEncoder(ResNet, BaseEncoder):
-    backbone_names = ["resnet10", "resnet18", "resnet34", "resnet50", "resnet101", "resnet152", "resnet200"]
-    output_feature_channels = [(64, 128, 256, 512)] * 3 + [(256, 512, 1024, 2048)] * 4
-    parameter_layers = [
-        [1, 1, 1, 1],
-        [2, 2, 2, 2],
-        [3, 4, 6, 3],
-        [3, 4, 6, 3],
-        [3, 4, 23, 3],
-        [3, 8, 36, 3],
-        [3, 24, 36, 3],
-    ]
-
-    def __init__(self, in_channels, pretrained, **kargs):
-        super().__init__(**kargs, n_input_channels=in_channels)
-        if pretrained:
-            # Author of paper zipped the state_dict on googledrive,
-            # so would need to download, unzip and read (2.8gb file for a ~150mb state dict).
-            # Would like to load dict from url but need somewhere to save the state dicts.
-            raise NotImplementedError(
-                "Currently not implemented. You need to manually download weights provided by the paper's author"
-                " and load then to the model with `state_dict`. See https://github.com/Tencent/MedicalNet"
-            )
-
-    @staticmethod
-    def get_inplanes():
-        return [64, 128, 256, 512]
-
-    @classmethod
-    def get_encoder_parameters(cls) -> list[dict]:
-        """
-        Get parameter list to initialize encoder networks.
-        Each parameter dict must have `spatial_dims`, `in_channels`
-        and `pretrained` parameters.
-        """
-        parameter_list = []
-        res_type: type[ResNetBlock] | type[ResNetBottleneck]
-        for backbone in range(len(cls.backbone_names)):
-            if backbone < 3:
-                res_type = ResNetBlock
-            else:
-                res_type = ResNetBottleneck
-            parameter_list.append(
-                {
-                    "block": res_type,
-                    "layers": cls.parameter_layers[backbone],
-                    "block_inplanes": ResNetEncoder.get_inplanes(),
-                    "spatial_dims": 2,
-                    "in_channels": 3,
-                    "pretrained": False,
-                }
-            )
-        return parameter_list
-
-    @classmethod
-    def num_channels_per_output(cls):
-        """
-        Get number of output features' channel.
-        """
-        return cls.output_feature_channels
-
-    @classmethod
-    def num_outputs(cls):
-        """
-        Get number of output feature.
-        """
-        return [4] * 7
-
-    @classmethod
-    def get_encoder_names(cls):
-        """
-        Get the name string of backbones which will be used to initialize flexible unet.
-        """
-        return cls.backbone_names
-
-    def forward(self, x: torch.Tensor):
-        feature_list = []
-        x = self.conv1(x)
-        x = self.bn1(x)
-        x = self.relu(x)
-        if not self.no_max_pool:
-            x = self.maxpool(x)
-        x = self.layer1(x)
-        feature_list.append(x)
-        x = self.layer2(x)
-        feature_list.append(x)
-        x = self.layer3(x)
-        feature_list.append(x)
-        x = self.layer4(x)
-        feature_list.append(x)
-
-        return feature_list
-
-
-FLEXUNET_BACKBONE.register_class(ResNetEncoder)
 FLEXUNET_BACKBONE.register_class(DummyEncoder)
 
 
@@ -204,9 +108,7 @@ def make_shape_cases(
 
 
 def make_error_case():
-    error_dummy_backbones = DummyEncoder.get_encoder_names()
-    error_resnet_backbones = ResNetEncoder.get_encoder_names()
-    error_backbones = error_dummy_backbones + error_resnet_backbones
+    error_backbones = DummyEncoder.get_encoder_names()
     error_param_list = []
     for backbone in error_backbones:
         error_param_list.append(
@@ -232,7 +134,7 @@ def make_error_case():
     norm="instance",
 )
 CASES_3D = make_shape_cases(
-    models=[SEL_MODELS[0]],
+    models=[SEL_MODELS[0], SEL_MODELS[2]],
     spatial_dims=[3],
     batches=[1],
     pretrained=[False],
@@ -345,6 +247,7 @@ def make_error_case():
             "spatial_dims": 2,
             "norm": ("batch", {"eps": 1e-3, "momentum": 0.01}),
         },
+        EfficientNetBNFeatures,
         {
             "in_channels": 3,
             "num_classes": 10,
@@ -354,7 +257,20 @@ def make_error_case():
             "norm": ("batch", {"eps": 1e-3, "momentum": 0.01}),
         },
         ["_conv_stem.weight"],
-    )
+    ),
+    (
+        {
+            "in_channels": 1,
+            "out_channels": 10,
+            "backbone": SEL_MODELS[2],
+            "pretrained": True,
+            "spatial_dims": 3,
+            "norm": ("batch", {"eps": 1e-3, "momentum": 0.01}),
+        },
+        ResNetFeatures,
+        {"model_name": SEL_MODELS[2], "pretrained": True, "spatial_dims": 3, "in_channels": 1},
+        ["conv1.weight"],
+    ),
 ]
 
 CASE_ERRORS = make_error_case()
@@ -363,6 +279,7 @@ def make_error_case():
 CASE_REGISTER_ENCODER = ["EfficientNetEncoder", "monai.networks.nets.EfficientNetEncoder"]
 
 
+@SkipIfNoModule("hf_hub_download")
 @skip_if_quick
 class TestFLEXIBLEUNET(unittest.TestCase):
 
@@ -381,19 +298,19 @@ def test_shape(self, input_param, input_shape, expected_shape):
         self.assertEqual(result.shape, expected_shape)
 
     @parameterized.expand(CASES_PRETRAIN)
-    def test_pretrain(self, input_param, efficient_input_param, weight_list):
+    def test_pretrain(self, flexunet_input_param, feature_extractor_class, feature_extractor_input_param, weight_list):
         device = "cuda" if torch.cuda.is_available() else "cpu"
 
         with skip_if_downloading_fails():
-            net = FlexibleUNet(**input_param).to(device)
+            net = FlexibleUNet(**flexunet_input_param).to(device)
 
         with skip_if_downloading_fails():
-            eff_net = EfficientNetBNFeatures(**efficient_input_param).to(device)
+            feature_extractor_net = feature_extractor_class(**feature_extractor_input_param).to(device)
 
         for weight_name in weight_list:
-            if weight_name in net.encoder.state_dict() and weight_name in eff_net.state_dict():
+            if weight_name in net.encoder.state_dict() and weight_name in feature_extractor_net.state_dict():
                 net_weight = net.encoder.state_dict()[weight_name]
-                download_weight = eff_net.state_dict()[weight_name]
+                download_weight = feature_extractor_net.state_dict()[weight_name]
                 weight_diff = torch.abs(net_weight - download_weight)
                 diff_sum = torch.sum(weight_diff)
                 # check if a weight in weight_list equals to the downloaded weight.
diff --git a/tests/test_resnet.py b/tests/test_resnet.py
index ad1aad8fc6..449edba4bf 100644
--- a/tests/test_resnet.py
+++ b/tests/test_resnet.py
@@ -24,6 +24,7 @@
 from monai.networks import eval_mode
 from monai.networks.nets import (
     ResNet,
+    ResNetFeatures,
     get_medicalnet_pretrained_resnet_args,
     get_pretrained_resnet_medicalnet,
     resnet10,
@@ -36,7 +37,14 @@
 )
 from monai.networks.nets.resnet import ResNetBlock
 from monai.utils import optional_import
-from tests.utils import equal_state_dict, skip_if_downloading_fails, skip_if_no_cuda, skip_if_quick, test_script_save
+from tests.utils import (
+    SkipIfNoModule,
+    equal_state_dict,
+    skip_if_downloading_fails,
+    skip_if_no_cuda,
+    skip_if_quick,
+    test_script_save,
+)
 
 if TYPE_CHECKING:
     import torchvision
@@ -191,6 +199,15 @@
 ]
 
 
+CASE_EXTRACT_FEATURES = [
+    (
+        {"model_name": "resnet10", "pretrained": True, "spatial_dims": 3, "in_channels": 1},
+        [1, 1, 64, 64, 64],
+        ([1, 64, 32, 32, 32], [1, 64, 16, 16, 16], [1, 128, 8, 8, 8], [1, 256, 4, 4, 4], [1, 512, 2, 2, 2]),
+    )
+]
+
+
 class TestResNet(unittest.TestCase):
 
     def setUp(self):
@@ -270,5 +287,25 @@ def test_script(self, model, input_param, input_shape, expected_shape):
         test_script_save(net, test_data)
 
 
+@SkipIfNoModule("hf_hub_download")
+class TestExtractFeatures(unittest.TestCase):
+
+    @parameterized.expand(CASE_EXTRACT_FEATURES)
+    def test_shape(self, input_param, input_shape, expected_shapes):
+        device = "cuda" if torch.cuda.is_available() else "cpu"
+
+        with skip_if_downloading_fails():
+            net = ResNetFeatures(**input_param).to(device)
+
+        # run inference with random tensor
+        with eval_mode(net):
+            features = net(torch.randn(input_shape).to(device))
+
+        # check output shape
+        self.assertEqual(len(features), len(expected_shapes))
+        for feature, expected_shape in zip(features, expected_shapes):
+            self.assertEqual(feature.shape, torch.Size(expected_shape))
+
+
 if __name__ == "__main__":
     unittest.main()

From 1c07a176409bdb99bd06505dd59296c7ac274a92 Mon Sep 17 00:00:00 2001
From: Han Wang <freddie.wanah@gmail.com>
Date: Tue, 23 Apr 2024 15:38:28 +0200
Subject: [PATCH 034/183] Refactored test assertions that have suboptimal tests
 with numbers (#7671)

### Description

As discussed in PR #7609, I tried to break the suboptimal test refactors
to smaller pieces.

Suboptimal Assert: Instead of using statements such as assertIsNone,
assertIsInstance, always simply use assertTrue or assertFalse. This will
decrease the code overall readability and increase the execution time as
extra logic needed.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Han Wang <freddie.wanah@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 tests/test_bundle_get_data.py            |  2 +-
 tests/test_compute_regression_metrics.py | 10 ++++++----
 tests/test_handler_stats.py              | 16 ++++++++--------
 tests/test_invertd.py                    |  2 +-
 tests/test_load_spacing_orientation.py   |  4 ++--
 tests/test_meta_affine.py                |  4 ++--
 tests/test_persistentdataset.py          |  2 +-
 tests/test_rand_weighted_cropd.py        |  2 +-
 tests/test_recon_net_utils.py            |  2 +-
 tests/test_reg_loss_integration.py       |  2 +-
 tests/test_sobel_gradient.py             |  4 ++--
 tests/test_sobel_gradientd.py            |  4 ++--
 tests/test_threadcontainer.py            |  2 +-
 tests/test_warp.py                       |  2 +-
 14 files changed, 30 insertions(+), 28 deletions(-)

diff --git a/tests/test_bundle_get_data.py b/tests/test_bundle_get_data.py
index 605b3945bb..3731752b65 100644
--- a/tests/test_bundle_get_data.py
+++ b/tests/test_bundle_get_data.py
@@ -53,7 +53,7 @@ def test_get_all_bundles_list(self, params):
             output = get_all_bundles_list(**params)
             self.assertTrue(isinstance(output, list))
             self.assertTrue(isinstance(output[0], tuple))
-            self.assertTrue(len(output[0]) == 2)
+            self.assertEqual(len(output[0]), 2)
 
     @parameterized.expand([TEST_CASE_1, TEST_CASE_5])
     @skip_if_quick
diff --git a/tests/test_compute_regression_metrics.py b/tests/test_compute_regression_metrics.py
index a8b7f03e47..c407ab6ba6 100644
--- a/tests/test_compute_regression_metrics.py
+++ b/tests/test_compute_regression_metrics.py
@@ -70,22 +70,24 @@ def test_shape_reduction(self):
                         mt = mt_fn(reduction="mean")
                         mt(in_tensor, in_tensor)
                         out_tensor = mt.aggregate()
-                        self.assertTrue(len(out_tensor.shape) == 1)
+                        self.assertEqual(len(out_tensor.shape), 1)
 
                         mt = mt_fn(reduction="sum")
                         mt(in_tensor, in_tensor)
                         out_tensor = mt.aggregate()
-                        self.assertTrue(len(out_tensor.shape) == 0)
+                        self.assertEqual(len(out_tensor.shape), 0)
 
                         mt = mt_fn(reduction="sum")  # test reduction arg overriding
                         mt(in_tensor, in_tensor)
                         out_tensor = mt.aggregate(reduction="mean_channel")
-                        self.assertTrue(len(out_tensor.shape) == 1 and out_tensor.shape[0] == batch)
+                        self.assertEqual(len(out_tensor.shape), 1)
+                        self.assertEqual(out_tensor.shape[0], batch)
 
                         mt = mt_fn(reduction="sum_channel")
                         mt(in_tensor, in_tensor)
                         out_tensor = mt.aggregate()
-                        self.assertTrue(len(out_tensor.shape) == 1 and out_tensor.shape[0] == batch)
+                        self.assertEqual(len(out_tensor.shape), 1)
+                        self.assertEqual(out_tensor.shape[0], batch)
 
     def test_compare_numpy(self):
         set_determinism(seed=123)
diff --git a/tests/test_handler_stats.py b/tests/test_handler_stats.py
index f876cff2a3..52da5c179b 100644
--- a/tests/test_handler_stats.py
+++ b/tests/test_handler_stats.py
@@ -76,9 +76,9 @@ def _update_metric(engine):
             if has_key_word.match(line):
                 content_count += 1
         if epoch_log is True:
-            self.assertTrue(content_count == max_epochs)
+            self.assertEqual(content_count, max_epochs)
         else:
-            self.assertTrue(content_count == 2)  # 2 = len([1, 2]) from event_filter
+            self.assertEqual(content_count, 2)  # 2 = len([1, 2]) from event_filter
 
     @parameterized.expand([[True], [get_event_filter([1, 3])]])
     def test_loss_print(self, iteration_log):
@@ -116,9 +116,9 @@ def _train_func(engine, batch):
             if has_key_word.match(line):
                 content_count += 1
         if iteration_log is True:
-            self.assertTrue(content_count == num_iters * max_epochs)
+            self.assertEqual(content_count, num_iters * max_epochs)
         else:
-            self.assertTrue(content_count == 2)  # 2 = len([1, 3]) from event_filter
+            self.assertEqual(content_count, 2)  # 2 = len([1, 3]) from event_filter
 
     def test_loss_dict(self):
         log_stream = StringIO()
@@ -150,7 +150,7 @@ def _train_func(engine, batch):
         for line in output_str.split("\n"):
             if has_key_word.match(line):
                 content_count += 1
-        self.assertTrue(content_count > 0)
+        self.assertGreater(content_count, 0)
 
     def test_loss_file(self):
         key_to_handler = "test_logging"
@@ -184,7 +184,7 @@ def _train_func(engine, batch):
                 for line in output_str.split("\n"):
                     if has_key_word.match(line):
                         content_count += 1
-                self.assertTrue(content_count > 0)
+                self.assertGreater(content_count, 0)
 
     def test_exception(self):
         # set up engine
@@ -239,7 +239,7 @@ def _update_metric(engine):
         for line in output_str.split("\n"):
             if has_key_word.match(line):
                 content_count += 1
-        self.assertTrue(content_count > 0)
+        self.assertGreater(content_count, 0)
 
     def test_default_logger(self):
         log_stream = StringIO()
@@ -274,7 +274,7 @@ def _train_func(engine, batch):
         for line in output_str.split("\n"):
             if has_key_word.match(line):
                 content_count += 1
-        self.assertTrue(content_count > 0)
+        self.assertGreater(content_count, 0)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_invertd.py b/tests/test_invertd.py
index c32a3af643..f6e8fc40e7 100644
--- a/tests/test_invertd.py
+++ b/tests/test_invertd.py
@@ -134,7 +134,7 @@ def test_invert(self):
         # 25300: 2 workers (cpu, non-macos)
         # 1812: 0 workers (gpu or macos)
         # 1821: windows torch 1.10.0
-        self.assertTrue((reverted.size - n_good) < 40000, f"diff.  {reverted.size - n_good}")
+        self.assertLess((reverted.size - n_good), 40000, f"diff.  {reverted.size - n_good}")
 
         set_determinism(seed=None)
 
diff --git a/tests/test_load_spacing_orientation.py b/tests/test_load_spacing_orientation.py
index 63422761ca..cbc730e1bb 100644
--- a/tests/test_load_spacing_orientation.py
+++ b/tests/test_load_spacing_orientation.py
@@ -48,7 +48,7 @@ def test_load_spacingd(self, filename):
         ref = resample_to_output(anat, (1, 0.2, 1), order=1)
         t2 = time.time()
         print(f"time scipy: {t2 - t1}")
-        self.assertTrue(t2 >= t1)
+        self.assertGreaterEqual(t2, t1)
         np.testing.assert_allclose(res_dict["image"].affine, ref.affine)
         np.testing.assert_allclose(res_dict["image"].shape[1:], ref.shape)
         np.testing.assert_allclose(ref.get_fdata(), res_dict["image"][0], atol=0.05)
@@ -68,7 +68,7 @@ def test_load_spacingd_rotate(self, filename):
         ref = resample_to_output(anat, (1, 2, 3), order=1)
         t2 = time.time()
         print(f"time scipy: {t2 - t1}")
-        self.assertTrue(t2 >= t1)
+        self.assertGreaterEqual(t2, t1)
         np.testing.assert_allclose(res_dict["image"].affine, ref.affine)
         if "anatomical" not in filename:
             np.testing.assert_allclose(res_dict["image"].shape[1:], ref.shape)
diff --git a/tests/test_meta_affine.py b/tests/test_meta_affine.py
index 95764a0c89..890734391f 100644
--- a/tests/test_meta_affine.py
+++ b/tests/test_meta_affine.py
@@ -160,7 +160,7 @@ def test_linear_consistent(self, xform_cls, input_dict, atol):
         diff = np.abs(itk.GetArrayFromImage(ref_2) - itk.GetArrayFromImage(expected))
         avg_diff = np.mean(diff)
 
-        self.assertTrue(avg_diff < atol, f"{xform_cls} avg_diff: {avg_diff}, tol: {atol}")
+        self.assertLess(avg_diff, atol, f"{xform_cls} avg_diff: {avg_diff}, tol: {atol}")
 
     @parameterized.expand(TEST_CASES_DICT)
     def test_linear_consistent_dict(self, xform_cls, input_dict, atol):
@@ -175,7 +175,7 @@ def test_linear_consistent_dict(self, xform_cls, input_dict, atol):
         diff = {k: np.abs(itk.GetArrayFromImage(ref_2[k]) - itk.GetArrayFromImage(expected[k])) for k in keys}
         avg_diff = {k: np.mean(diff[k]) for k in keys}
         for k in keys:
-            self.assertTrue(avg_diff[k] < atol, f"{xform_cls} avg_diff: {avg_diff}, tol: {atol}")
+            self.assertLess(avg_diff[k], atol, f"{xform_cls} avg_diff: {avg_diff}, tol: {atol}")
 
 
 if __name__ == "__main__":
diff --git a/tests/test_persistentdataset.py b/tests/test_persistentdataset.py
index b7bf2fbb11..7c4969e283 100644
--- a/tests/test_persistentdataset.py
+++ b/tests/test_persistentdataset.py
@@ -165,7 +165,7 @@ def test_different_transforms(self):
             im1 = PersistentDataset([im], Identity(), cache_dir=path, hash_transform=json_hashing)[0]
             im2 = PersistentDataset([im], Flip(1), cache_dir=path, hash_transform=json_hashing)[0]
             l2 = ((im1 - im2) ** 2).sum() ** 0.5
-            self.assertTrue(l2 > 1)
+            self.assertGreater(l2, 1)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_rand_weighted_cropd.py b/tests/test_rand_weighted_cropd.py
index 1524442f61..a1414df0ac 100644
--- a/tests/test_rand_weighted_cropd.py
+++ b/tests/test_rand_weighted_cropd.py
@@ -154,7 +154,7 @@ def test_rand_weighted_cropd(self, _, init_params, input_data, expected_shape, e
         crop = RandWeightedCropd(**init_params)
         crop.set_random_state(10)
         result = crop(input_data)
-        self.assertTrue(len(result) == init_params["num_samples"])
+        self.assertEqual(len(result), init_params["num_samples"])
         _len = len(tuple(input_data.keys()))
         self.assertTupleEqual(tuple(result[0].keys())[:_len], tuple(input_data.keys()))
 
diff --git a/tests/test_recon_net_utils.py b/tests/test_recon_net_utils.py
index 1815000777..48d3b59a17 100644
--- a/tests/test_recon_net_utils.py
+++ b/tests/test_recon_net_utils.py
@@ -64,7 +64,7 @@ def test_reshape_channel_complex(self, test_data):
     def test_complex_normalize(self, test_data):
         result, mean, std = complex_normalize(test_data)
         result = result * std + mean
-        self.assertTrue((((result - test_data) ** 2).mean() ** 0.5).item() < 1e-5)
+        self.assertLess((((result - test_data) ** 2).mean() ** 0.5).item(), 1e-5)
 
     @parameterized.expand(TEST_PAD)
     def test_pad(self, test_data):
diff --git a/tests/test_reg_loss_integration.py b/tests/test_reg_loss_integration.py
index e8f82eb0c2..1fb81689e6 100644
--- a/tests/test_reg_loss_integration.py
+++ b/tests/test_reg_loss_integration.py
@@ -99,7 +99,7 @@ def forward(self, x):
             # backward pass
             loss_val.backward()
             optimizer.step()
-        self.assertTrue(init_loss > loss_val, "loss did not decrease")
+        self.assertGreater(init_loss, loss_val, "loss did not decrease")
 
 
 if __name__ == "__main__":
diff --git a/tests/test_sobel_gradient.py b/tests/test_sobel_gradient.py
index 3d995a60c9..a0d7cf5a8b 100644
--- a/tests/test_sobel_gradient.py
+++ b/tests/test_sobel_gradient.py
@@ -164,8 +164,8 @@ def test_sobel_gradients(self, image, arguments, expected_grad):
     )
     def test_sobel_kernels(self, arguments, expected_kernels):
         sobel = SobelGradients(**arguments)
-        self.assertTrue(sobel.kernel_diff.dtype == expected_kernels[0].dtype)
-        self.assertTrue(sobel.kernel_smooth.dtype == expected_kernels[0].dtype)
+        self.assertEqual(sobel.kernel_diff.dtype, expected_kernels[0].dtype)
+        self.assertEqual(sobel.kernel_smooth.dtype, expected_kernels[0].dtype)
         assert_allclose(sobel.kernel_diff, expected_kernels[0])
         assert_allclose(sobel.kernel_smooth, expected_kernels[1])
 
diff --git a/tests/test_sobel_gradientd.py b/tests/test_sobel_gradientd.py
index 7499a0410b..03524823a5 100644
--- a/tests/test_sobel_gradientd.py
+++ b/tests/test_sobel_gradientd.py
@@ -187,8 +187,8 @@ def test_sobel_gradients(self, image_dict, arguments, expected_grad):
     )
     def test_sobel_kernels(self, arguments, expected_kernels):
         sobel = SobelGradientsd(**arguments)
-        self.assertTrue(sobel.kernel_diff.dtype == expected_kernels[0].dtype)
-        self.assertTrue(sobel.kernel_smooth.dtype == expected_kernels[0].dtype)
+        self.assertEqual(sobel.kernel_diff.dtype, expected_kernels[0].dtype)
+        self.assertEqual(sobel.kernel_smooth.dtype, expected_kernels[0].dtype)
         assert_allclose(sobel.kernel_diff, expected_kernels[0])
         assert_allclose(sobel.kernel_smooth, expected_kernels[1])
 
diff --git a/tests/test_threadcontainer.py b/tests/test_threadcontainer.py
index 9551dec703..568461748b 100644
--- a/tests/test_threadcontainer.py
+++ b/tests/test_threadcontainer.py
@@ -62,7 +62,7 @@ def test_container(self):
 
         self.assertTrue(con.is_alive)
         self.assertIsNotNone(con.status())
-        self.assertTrue(len(con.status_dict) > 0)
+        self.assertGreater(len(con.status_dict), 0)
 
         con.join()
 
diff --git a/tests/test_warp.py b/tests/test_warp.py
index bac595224f..55f40764c3 100644
--- a/tests/test_warp.py
+++ b/tests/test_warp.py
@@ -124,7 +124,7 @@ def test_itk_benchmark(self):
         relative_diff = np.mean(
             np.divide(monai_result - itk_result, itk_result, out=np.zeros_like(itk_result), where=(itk_result != 0))
         )
-        self.assertTrue(relative_diff < 0.01)
+        self.assertLess(relative_diff, 0.01)
 
     @parameterized.expand(TEST_CASES, skip_on_empty=True)
     def test_resample(self, input_param, input_data, expected_val):

From 07a78d20b30ab4abe8fd776063ab6903439df264 Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Wed, 24 Apr 2024 03:31:28 +0100
Subject: [PATCH 035/183] Auto3DSeg algo_template hash update (#7700)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 monai/utils/misc.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index 2b4b30fe48..84e979abf4 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -527,7 +527,7 @@ def doc_images() -> str | None:
 
     @staticmethod
     def algo_hash() -> str | None:
-        return os.environ.get("MONAI_ALGO_HASH", "4403f94")
+        return os.environ.get("MONAI_ALGO_HASH", "07acb39")
 
     @staticmethod
     def trace_transform() -> str | None:

From c3e4457f98e80785a946892bdd6741b9fead9113 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 24 Apr 2024 17:15:58 +0800
Subject: [PATCH 036/183] Update pycln version (#7704)

Fixes #7703


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/pythonapp-min.yml | 2 ++
 .github/workflows/pythonapp.yml     | 2 ++
 .pre-commit-config.yaml             | 2 +-
 3 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/.github/workflows/pythonapp-min.yml b/.github/workflows/pythonapp-min.yml
index bbe7579774..dffae10558 100644
--- a/.github/workflows/pythonapp-min.yml
+++ b/.github/workflows/pythonapp-min.yml
@@ -9,6 +9,8 @@ on:
       - main
       - releasing/*
   pull_request:
+    head_ref-ignore:
+      - dev
 
 concurrency:
   # automatically cancel the previously triggered workflows when there's a newer version
diff --git a/.github/workflows/pythonapp.yml b/.github/workflows/pythonapp.yml
index b7f2cfb9db..d4043585cc 100644
--- a/.github/workflows/pythonapp.yml
+++ b/.github/workflows/pythonapp.yml
@@ -9,6 +9,8 @@ on:
       - main
       - releasing/*
   pull_request:
+    head_ref-ignore:
+      - dev
 
 concurrency:
   # automatically cancel the previously triggered workflows when there's a newer version
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index b71a2bac43..b9debaf08f 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -69,7 +69,7 @@ repos:
           )$
 
   - repo: https://github.com/hadialqattan/pycln
-    rev: v2.1.3
+    rev: v2.4.0
     hooks:
       - id: pycln
         args: [--config=pyproject.toml]

From bfe09b8c3c9538cee577d3608330879952a54fde Mon Sep 17 00:00:00 2001
From: Han Wang <freddie.wanah@gmail.com>
Date: Wed, 24 Apr 2024 17:09:21 +0200
Subject: [PATCH 037/183] Refactored others type of subotimal asserts (#7672)

### Description

As discussed in PR #7609, I tried to break the suboptimal test refactors
to smaller pieces. In this PR all asserts are checking textual content
or instance of a parameter.

Suboptimal Assert: Instead of using statements such as assertIsNone,
assertIsInstance, always simply use assertTrue or assertFalse. This will
decrease the code overall readability and increase the execution time as
extra logic needed.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Han Wang <freddie.wanah@gmail.com>
Signed-off-by: Ben Murray <ben.murray@gmail.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Ben Murray <ben.murray@gmail.com>
---
 tests/test_bundle_ckpt_export.py     |  6 +++---
 tests/test_bundle_get_data.py        | 17 ++++++++-------
 tests/test_bundle_trt_export.py      | 12 +++++------
 tests/test_bundle_workflow.py        |  6 +++---
 tests/test_component_store.py        |  8 +++----
 tests/test_compute_ho_ver_maps.py    |  4 ++--
 tests/test_compute_ho_ver_maps_d.py  |  4 ++--
 tests/test_concat_itemsd.py          |  8 +++----
 tests/test_config_parser.py          |  2 +-
 tests/test_cucim_dict_transform.py   | 16 +++++++-------
 tests/test_cucim_transform.py        | 16 +++++++-------
 tests/test_detect_envelope.py        |  2 +-
 tests/test_ensure_typed.py           | 32 ++++++++++++++--------------
 tests/test_flipd.py                  |  2 +-
 tests/test_freeze_layers.py          |  8 +++----
 tests/test_generalized_dice_loss.py  |  4 ++--
 tests/test_get_package_version.py    |  6 +++---
 tests/test_grid_patch.py             |  6 +++---
 tests/test_integration_bundle_run.py |  6 ++----
 tests/test_inverse_collation.py      |  2 +-
 tests/test_load_imaged.py            |  2 +-
 tests/test_look_up_option.py         |  2 +-
 tests/test_matshow3d.py              |  2 +-
 tests/test_mednistdataset.py         |  2 +-
 tests/test_meta_tensor.py            |  4 ++--
 tests/test_mmar_download.py          |  2 +-
 tests/test_rand_affined.py           |  2 +-
 tests/test_rand_bias_field.py        |  2 +-
 tests/test_resnet.py                 |  2 +-
 tests/test_to_cupy.py                | 16 +++++++-------
 tests/test_to_numpy.py               | 12 +++++------
 tests/test_torchvision_fc_model.py   |  4 ++--
 tests/test_traceable_transform.py    |  4 ++--
 33 files changed, 111 insertions(+), 112 deletions(-)

diff --git a/tests/test_bundle_ckpt_export.py b/tests/test_bundle_ckpt_export.py
index 8f376a06d5..cfcadcfc4c 100644
--- a/tests/test_bundle_ckpt_export.py
+++ b/tests/test_bundle_ckpt_export.py
@@ -72,9 +72,9 @@ def test_export(self, key_in_ckpt, use_trace):
             _, metadata, extra_files = load_net_with_metadata(
                 ts_file, more_extra_files=["inference.json", "def_args.json"]
             )
-            self.assertTrue("schema" in metadata)
-            self.assertTrue("meta_file" in json.loads(extra_files["def_args.json"]))
-            self.assertTrue("network_def" in json.loads(extra_files["inference.json"]))
+            self.assertIn("schema", metadata)
+            self.assertIn("meta_file", json.loads(extra_files["def_args.json"]))
+            self.assertIn("network_def", json.loads(extra_files["inference.json"]))
 
     @parameterized.expand([TEST_CASE_1, TEST_CASE_2, TEST_CASE_3])
     def test_default_value(self, key_in_ckpt, use_trace):
diff --git a/tests/test_bundle_get_data.py b/tests/test_bundle_get_data.py
index 3731752b65..f84713fbe3 100644
--- a/tests/test_bundle_get_data.py
+++ b/tests/test_bundle_get_data.py
@@ -51,25 +51,26 @@ class TestGetBundleData(unittest.TestCase):
     def test_get_all_bundles_list(self, params):
         with skip_if_downloading_fails():
             output = get_all_bundles_list(**params)
-            self.assertTrue(isinstance(output, list))
-            self.assertTrue(isinstance(output[0], tuple))
-            self.assertEqual(len(output[0]), 2)
+            self.assertIsInstance(output, list)
+            self.assertIsInstance(output[0], tuple)
+            self.assertTrue(len(output[0]) == 2)
 
     @parameterized.expand([TEST_CASE_1, TEST_CASE_5])
     @skip_if_quick
     def test_get_bundle_versions(self, params):
         with skip_if_downloading_fails():
             output = get_bundle_versions(**params)
-            self.assertTrue(isinstance(output, dict))
-            self.assertTrue("latest_version" in output and "all_versions" in output)
-            self.assertTrue("0.1.0" in output["all_versions"])
+            self.assertIsInstance(output, dict)
+            self.assertIn("latest_version", output)
+            self.assertIn("all_versions", output)
+            self.assertIn("0.1.0", output["all_versions"])
 
     @parameterized.expand([TEST_CASE_1, TEST_CASE_2])
     @skip_if_quick
     def test_get_bundle_info(self, params):
         with skip_if_downloading_fails():
             output = get_bundle_info(**params)
-            self.assertTrue(isinstance(output, dict))
+            self.assertIsInstance(output, dict)
             for key in ["id", "name", "size", "download_count", "browser_download_url"]:
                 self.assertTrue(key in output)
 
@@ -78,7 +79,7 @@ def test_get_bundle_info(self, params):
     def test_get_bundle_info_monaihosting(self, params):
         with skip_if_downloading_fails():
             output = get_bundle_info(**params)
-            self.assertTrue(isinstance(output, dict))
+            self.assertIsInstance(output, dict)
             for key in ["name", "browser_download_url"]:
                 self.assertTrue(key in output)
 
diff --git a/tests/test_bundle_trt_export.py b/tests/test_bundle_trt_export.py
index 47034852ef..833a0ca1dc 100644
--- a/tests/test_bundle_trt_export.py
+++ b/tests/test_bundle_trt_export.py
@@ -91,9 +91,9 @@ def test_trt_export(self, convert_precision, input_shape, dynamic_batch):
             _, metadata, extra_files = load_net_with_metadata(
                 ts_file, more_extra_files=["inference.json", "def_args.json"]
             )
-            self.assertTrue("schema" in metadata)
-            self.assertTrue("meta_file" in json.loads(extra_files["def_args.json"]))
-            self.assertTrue("network_def" in json.loads(extra_files["inference.json"]))
+            self.assertIn("schema", metadata)
+            self.assertIn("meta_file", json.loads(extra_files["def_args.json"]))
+            self.assertIn("network_def", json.loads(extra_files["inference.json"]))
 
     @parameterized.expand([TEST_CASE_3, TEST_CASE_4])
     @unittest.skipUnless(
@@ -129,9 +129,9 @@ def test_onnx_trt_export(self, convert_precision, input_shape, dynamic_batch):
             _, metadata, extra_files = load_net_with_metadata(
                 ts_file, more_extra_files=["inference.json", "def_args.json"]
             )
-            self.assertTrue("schema" in metadata)
-            self.assertTrue("meta_file" in json.loads(extra_files["def_args.json"]))
-            self.assertTrue("network_def" in json.loads(extra_files["inference.json"]))
+            self.assertIn("schema", metadata)
+            self.assertIn("meta_file", json.loads(extra_files["def_args.json"]))
+            self.assertIn("network_def", json.loads(extra_files["inference.json"]))
 
 
 if __name__ == "__main__":
diff --git a/tests/test_bundle_workflow.py b/tests/test_bundle_workflow.py
index 9a276b577f..1727fcdf53 100644
--- a/tests/test_bundle_workflow.py
+++ b/tests/test_bundle_workflow.py
@@ -138,11 +138,11 @@ def test_train_config(self, config_file):
         self.assertListEqual(trainer.check_properties(), [])
         # test read / write the properties
         dataset = trainer.train_dataset
-        self.assertTrue(isinstance(dataset, Dataset))
+        self.assertIsInstance(dataset, Dataset)
         inferer = trainer.train_inferer
-        self.assertTrue(isinstance(inferer, SimpleInferer))
+        self.assertIsInstance(inferer, SimpleInferer)
         # test optional properties get
-        self.assertTrue(trainer.train_key_metric is None)
+        self.assertIsNone(trainer.train_key_metric)
         trainer.train_dataset = deepcopy(dataset)
         trainer.train_inferer = deepcopy(inferer)
         # test optional properties set
diff --git a/tests/test_component_store.py b/tests/test_component_store.py
index 424eceb3d1..7e7c6dd19d 100644
--- a/tests/test_component_store.py
+++ b/tests/test_component_store.py
@@ -48,17 +48,17 @@ def test_add2(self):
         self.cs.add("test_obj2", "Test object", test_obj2)
 
         self.assertEqual(len(self.cs), 2)
-        self.assertTrue("test_obj1" in self.cs)
-        self.assertTrue("test_obj2" in self.cs)
+        self.assertIn("test_obj1", self.cs)
+        self.assertIn("test_obj2", self.cs)
 
     def test_add_def(self):
-        self.assertFalse("test_func" in self.cs)
+        self.assertNotIn("test_func", self.cs)
 
         @self.cs.add_def("test_func", "Test function")
         def test_func():
             return 123
 
-        self.assertTrue("test_func" in self.cs)
+        self.assertIn("test_func", self.cs)
 
         self.assertEqual(len(self.cs), 1)
         self.assertEqual(list(self.cs), [("test_func", test_func)])
diff --git a/tests/test_compute_ho_ver_maps.py b/tests/test_compute_ho_ver_maps.py
index bbd5230f04..6e46cf2b1e 100644
--- a/tests/test_compute_ho_ver_maps.py
+++ b/tests/test_compute_ho_ver_maps.py
@@ -67,8 +67,8 @@ class ComputeHoVerMapsTests(unittest.TestCase):
     def test_horizontal_certical_maps(self, in_type, arguments, mask, hv_mask):
         input_image = in_type(mask)
         result = ComputeHoVerMaps(**arguments)(input_image)
-        self.assertTrue(isinstance(result, torch.Tensor))
-        self.assertTrue(str(result.dtype).split(".")[1] == arguments.get("dtype", "float32"))
+        self.assertIsInstance(result, torch.Tensor)
+        self.assertEqual(str(result.dtype).split(".")[1], arguments.get("dtype", "float32"))
         assert_allclose(result, hv_mask, type_test="tensor")
 
 
diff --git a/tests/test_compute_ho_ver_maps_d.py b/tests/test_compute_ho_ver_maps_d.py
index 7b5ac0d9d7..0734e2e731 100644
--- a/tests/test_compute_ho_ver_maps_d.py
+++ b/tests/test_compute_ho_ver_maps_d.py
@@ -71,8 +71,8 @@ def test_horizontal_certical_maps(self, in_type, arguments, mask, hv_mask):
         for k in mask.keys():
             input_image[k] = in_type(mask[k])
         result = ComputeHoVerMapsd(keys="mask", **arguments)(input_image)[hv_key]
-        self.assertTrue(isinstance(result, torch.Tensor))
-        self.assertTrue(str(result.dtype).split(".")[1] == arguments.get("dtype", "float32"))
+        self.assertIsInstance(result, torch.Tensor)
+        self.assertEqual(str(result.dtype).split(".")[1], arguments.get("dtype", "float32"))
         assert_allclose(result, hv_mask[hv_key], type_test="tensor")
 
 
diff --git a/tests/test_concat_itemsd.py b/tests/test_concat_itemsd.py
index 64c5d6e255..564ddf5c1f 100644
--- a/tests/test_concat_itemsd.py
+++ b/tests/test_concat_itemsd.py
@@ -30,7 +30,7 @@ def test_tensor_values(self):
             "img2": torch.tensor([[0, 1], [1, 2]], device=device),
         }
         result = ConcatItemsd(keys=["img1", "img2"], name="cat_img")(input_data)
-        self.assertTrue("cat_img" in result)
+        self.assertIn("cat_img", result)
         result["cat_img"] += 1
         assert_allclose(result["img1"], torch.tensor([[0, 1], [1, 2]], device=device))
         assert_allclose(result["cat_img"], torch.tensor([[1, 2], [2, 3], [1, 2], [2, 3]], device=device))
@@ -42,8 +42,8 @@ def test_metatensor_values(self):
             "img2": MetaTensor([[0, 1], [1, 2]], device=device),
         }
         result = ConcatItemsd(keys=["img1", "img2"], name="cat_img")(input_data)
-        self.assertTrue("cat_img" in result)
-        self.assertTrue(isinstance(result["cat_img"], MetaTensor))
+        self.assertIn("cat_img", result)
+        self.assertIsInstance(result["cat_img"], MetaTensor)
         self.assertEqual(result["img1"].meta, result["cat_img"].meta)
         result["cat_img"] += 1
         assert_allclose(result["img1"], torch.tensor([[0, 1], [1, 2]], device=device))
@@ -52,7 +52,7 @@ def test_metatensor_values(self):
     def test_numpy_values(self):
         input_data = {"img1": np.array([[0, 1], [1, 2]]), "img2": np.array([[0, 1], [1, 2]])}
         result = ConcatItemsd(keys=["img1", "img2"], name="cat_img")(input_data)
-        self.assertTrue("cat_img" in result)
+        self.assertIn("cat_img", result)
         result["cat_img"] += 1
         np.testing.assert_allclose(result["img1"], np.array([[0, 1], [1, 2]]))
         np.testing.assert_allclose(result["cat_img"], np.array([[1, 2], [2, 3], [1, 2], [2, 3]]))
diff --git a/tests/test_config_parser.py b/tests/test_config_parser.py
index cc890a0522..cf1edc8f08 100644
--- a/tests/test_config_parser.py
+++ b/tests/test_config_parser.py
@@ -185,7 +185,7 @@ def test_function(self, config):
             if id in ("compute", "cls_compute"):
                 parser[f"{id}#_mode_"] = "callable"
             func = parser.get_parsed_content(id=id)
-            self.assertTrue(id in parser.ref_resolver.resolved_content)
+            self.assertIn(id, parser.ref_resolver.resolved_content)
             if id == "error_func":
                 with self.assertRaises(TypeError):
                     func(1, 2)
diff --git a/tests/test_cucim_dict_transform.py b/tests/test_cucim_dict_transform.py
index d2dcc6aa5f..3c5703a34c 100644
--- a/tests/test_cucim_dict_transform.py
+++ b/tests/test_cucim_dict_transform.py
@@ -80,8 +80,8 @@ class TestCuCIMDict(unittest.TestCase):
     def test_tramsforms_numpy_single(self, params, input, expected):
         input = {"image": input}
         output = CuCIMd(keys="image", **params)(input)["image"]
-        self.assertTrue(output.dtype == expected.dtype)
-        self.assertTrue(isinstance(output, np.ndarray))
+        self.assertEqual(output.dtype, expected.dtype)
+        self.assertIsInstance(output, np.ndarray)
         cp.testing.assert_allclose(output, expected)
 
     @parameterized.expand(
@@ -98,8 +98,8 @@ def test_tramsforms_numpy_batch(self, params, input, expected):
         input = {"image": input[cp.newaxis, ...]}
         expected = expected[cp.newaxis, ...]
         output = CuCIMd(keys="image", **params)(input)["image"]
-        self.assertTrue(output.dtype == expected.dtype)
-        self.assertTrue(isinstance(output, np.ndarray))
+        self.assertEqual(output.dtype, expected.dtype)
+        self.assertIsInstance(output, np.ndarray)
         cp.testing.assert_allclose(output, expected)
 
     @parameterized.expand(
@@ -116,8 +116,8 @@ def test_tramsforms_cupy_single(self, params, input, expected):
         input = {"image": cp.asarray(input)}
         expected = cp.asarray(expected)
         output = CuCIMd(keys="image", **params)(input)["image"]
-        self.assertTrue(output.dtype == expected.dtype)
-        self.assertTrue(isinstance(output, cp.ndarray))
+        self.assertEqual(output.dtype, expected.dtype)
+        self.assertIsInstance(output, cp.ndarray)
         cp.testing.assert_allclose(output, expected)
 
     @parameterized.expand(
@@ -134,8 +134,8 @@ def test_tramsforms_cupy_batch(self, params, input, expected):
         input = {"image": cp.asarray(input)[cp.newaxis, ...]}
         expected = cp.asarray(expected)[cp.newaxis, ...]
         output = CuCIMd(keys="image", **params)(input)["image"]
-        self.assertTrue(output.dtype == expected.dtype)
-        self.assertTrue(isinstance(output, cp.ndarray))
+        self.assertEqual(output.dtype, expected.dtype)
+        self.assertIsInstance(output, cp.ndarray)
         cp.testing.assert_allclose(output, expected)
 
 
diff --git a/tests/test_cucim_transform.py b/tests/test_cucim_transform.py
index 5f16c11589..162e16b52a 100644
--- a/tests/test_cucim_transform.py
+++ b/tests/test_cucim_transform.py
@@ -79,8 +79,8 @@ class TestCuCIM(unittest.TestCase):
     )
     def test_tramsforms_numpy_single(self, params, input, expected):
         output = CuCIM(**params)(input)
-        self.assertTrue(output.dtype == expected.dtype)
-        self.assertTrue(isinstance(output, np.ndarray))
+        self.assertEqual(output.dtype, expected.dtype)
+        self.assertIsInstance(output, np.ndarray)
         cp.testing.assert_allclose(output, expected)
 
     @parameterized.expand(
@@ -97,8 +97,8 @@ def test_tramsforms_numpy_batch(self, params, input, expected):
         input = input[cp.newaxis, ...]
         expected = expected[cp.newaxis, ...]
         output = CuCIM(**params)(input)
-        self.assertTrue(output.dtype == expected.dtype)
-        self.assertTrue(isinstance(output, np.ndarray))
+        self.assertEqual(output.dtype, expected.dtype)
+        self.assertIsInstance(output, np.ndarray)
         cp.testing.assert_allclose(output, expected)
 
     @parameterized.expand(
@@ -115,8 +115,8 @@ def test_tramsforms_cupy_single(self, params, input, expected):
         input = cp.asarray(input)
         expected = cp.asarray(expected)
         output = CuCIM(**params)(input)
-        self.assertTrue(output.dtype == expected.dtype)
-        self.assertTrue(isinstance(output, cp.ndarray))
+        self.assertEqual(output.dtype, expected.dtype)
+        self.assertIsInstance(output, cp.ndarray)
         cp.testing.assert_allclose(output, expected)
 
     @parameterized.expand(
@@ -133,8 +133,8 @@ def test_tramsforms_cupy_batch(self, params, input, expected):
         input = cp.asarray(input)[cp.newaxis, ...]
         expected = cp.asarray(expected)[cp.newaxis, ...]
         output = CuCIM(**params)(input)
-        self.assertTrue(output.dtype == expected.dtype)
-        self.assertTrue(isinstance(output, cp.ndarray))
+        self.assertEqual(output.dtype, expected.dtype)
+        self.assertIsInstance(output, cp.ndarray)
         cp.testing.assert_allclose(output, expected)
 
 
diff --git a/tests/test_detect_envelope.py b/tests/test_detect_envelope.py
index e2efefeb77..f9c2b5ac53 100644
--- a/tests/test_detect_envelope.py
+++ b/tests/test_detect_envelope.py
@@ -147,7 +147,7 @@ def test_value_error(self, arguments, image, method):
         elif method == "__call__":
             self.assertRaises(ValueError, DetectEnvelope(**arguments), image)
         else:
-            raise ValueError("Expected raising method invalid. Should be __init__ or __call__.")
+            self.fail("Expected raising method invalid. Should be __init__ or __call__.")
 
 
 @SkipIfModule("torch.fft")
diff --git a/tests/test_ensure_typed.py b/tests/test_ensure_typed.py
index 09aa1f04b5..fe543347de 100644
--- a/tests/test_ensure_typed.py
+++ b/tests/test_ensure_typed.py
@@ -33,8 +33,8 @@ def test_array_input(self):
                     keys="data", data_type=dtype, dtype=np.float32 if dtype == "NUMPY" else None, device="cpu"
                 )({"data": test_data})["data"]
                 if dtype == "NUMPY":
-                    self.assertTrue(result.dtype == np.float32)
-                self.assertTrue(isinstance(result, torch.Tensor if dtype == "tensor" else np.ndarray))
+                    self.assertEqual(result.dtype, np.float32)
+                self.assertIsInstance(result, torch.Tensor if dtype == "tensor" else np.ndarray)
                 assert_allclose(result, test_data, type_test=False)
                 self.assertTupleEqual(result.shape, (2, 2))
 
@@ -45,7 +45,7 @@ def test_single_input(self):
         for test_data in test_datas:
             for dtype in ("tensor", "numpy"):
                 result = EnsureTyped(keys="data", data_type=dtype)({"data": test_data})["data"]
-                self.assertTrue(isinstance(result, torch.Tensor if dtype == "tensor" else np.ndarray))
+                self.assertIsInstance(result, torch.Tensor if dtype == "tensor" else np.ndarray)
                 if isinstance(test_data, bool):
                     self.assertFalse(result)
                 else:
@@ -56,11 +56,11 @@ def test_string(self):
         for dtype in ("tensor", "numpy"):
             # string input
             result = EnsureTyped(keys="data", data_type=dtype)({"data": "test_string"})["data"]
-            self.assertTrue(isinstance(result, str))
+            self.assertIsInstance(result, str)
             self.assertEqual(result, "test_string")
             # numpy array of string
             result = EnsureTyped(keys="data", data_type=dtype)({"data": np.array(["test_string"])})["data"]
-            self.assertTrue(isinstance(result, np.ndarray))
+            self.assertIsInstance(result, np.ndarray)
             self.assertEqual(result[0], "test_string")
 
     def test_list_tuple(self):
@@ -68,15 +68,15 @@ def test_list_tuple(self):
             result = EnsureTyped(keys="data", data_type=dtype, wrap_sequence=False, track_meta=True)(
                 {"data": [[1, 2], [3, 4]]}
             )["data"]
-            self.assertTrue(isinstance(result, list))
-            self.assertTrue(isinstance(result[0][1], MetaTensor if dtype == "tensor" else np.ndarray))
+            self.assertIsInstance(result, list)
+            self.assertIsInstance(result[0][1], MetaTensor if dtype == "tensor" else np.ndarray)
             assert_allclose(result[1][0], torch.as_tensor(3), type_test=False)
             # tuple of numpy arrays
             result = EnsureTyped(keys="data", data_type=dtype, wrap_sequence=False)(
                 {"data": (np.array([1, 2]), np.array([3, 4]))}
             )["data"]
-            self.assertTrue(isinstance(result, tuple))
-            self.assertTrue(isinstance(result[0], torch.Tensor if dtype == "tensor" else np.ndarray))
+            self.assertIsInstance(result, tuple)
+            self.assertIsInstance(result[0], torch.Tensor if dtype == "tensor" else np.ndarray)
             assert_allclose(result[1], torch.as_tensor([3, 4]), type_test=False)
 
     def test_dict(self):
@@ -92,19 +92,19 @@ def test_dict(self):
             )
             for key in ("data", "label"):
                 result = trans[key]
-                self.assertTrue(isinstance(result, dict))
-                self.assertTrue(isinstance(result["img"], torch.Tensor if dtype == "tensor" else np.ndarray))
-                self.assertTrue(isinstance(result["meta"]["size"], torch.Tensor if dtype == "tensor" else np.ndarray))
+                self.assertIsInstance(result, dict)
+                self.assertIsInstance(result["img"], torch.Tensor if dtype == "tensor" else np.ndarray)
+                self.assertIsInstance(result["meta"]["size"], torch.Tensor if dtype == "tensor" else np.ndarray)
                 self.assertEqual(result["meta"]["path"], "temp/test")
                 self.assertEqual(result["extra"], None)
                 assert_allclose(result["img"], torch.as_tensor([1.0, 2.0]), type_test=False)
                 assert_allclose(result["meta"]["size"], torch.as_tensor([1, 2, 3]), type_test=False)
             if dtype == "numpy":
-                self.assertTrue(trans["data"]["img"].dtype == np.float32)
-                self.assertTrue(trans["label"]["img"].dtype == np.int8)
+                self.assertEqual(trans["data"]["img"].dtype, np.float32)
+                self.assertEqual(trans["label"]["img"].dtype, np.int8)
             else:
-                self.assertTrue(trans["data"]["img"].dtype == torch.float32)
-                self.assertTrue(trans["label"]["img"].dtype == torch.int8)
+                self.assertEqual(trans["data"]["img"].dtype, torch.float32)
+                self.assertEqual(trans["label"]["img"].dtype, torch.int8)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_flipd.py b/tests/test_flipd.py
index 277f387051..1df6d34056 100644
--- a/tests/test_flipd.py
+++ b/tests/test_flipd.py
@@ -78,7 +78,7 @@ def test_torch(self, spatial_axis, img: torch.Tensor, track_meta: bool, device):
     def test_meta_dict(self):
         xform = Flipd("image", [0, 1])
         res = xform({"image": torch.zeros(1, 3, 4)})
-        self.assertTrue(res["image"].applied_operations == res["image_transforms"])
+        self.assertEqual(res["image"].applied_operations, res["image_transforms"])
 
 
 if __name__ == "__main__":
diff --git a/tests/test_freeze_layers.py b/tests/test_freeze_layers.py
index 1bea4ed1b5..7be8e576bf 100644
--- a/tests/test_freeze_layers.py
+++ b/tests/test_freeze_layers.py
@@ -40,9 +40,9 @@ def test_freeze_vars(self, device):
 
         for name, param in model.named_parameters():
             if "class_layer" in name:
-                self.assertEqual(param.requires_grad, False)
+                self.assertFalse(param.requires_grad)
             else:
-                self.assertEqual(param.requires_grad, True)
+                self.assertTrue(param.requires_grad)
 
     @parameterized.expand(TEST_CASES)
     def test_exclude_vars(self, device):
@@ -53,9 +53,9 @@ def test_exclude_vars(self, device):
 
         for name, param in model.named_parameters():
             if "class_layer" in name:
-                self.assertEqual(param.requires_grad, True)
+                self.assertTrue(param.requires_grad)
             else:
-                self.assertEqual(param.requires_grad, False)
+                self.assertFalse(param.requires_grad)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_generalized_dice_loss.py b/tests/test_generalized_dice_loss.py
index 7499507129..5738f4a089 100644
--- a/tests/test_generalized_dice_loss.py
+++ b/tests/test_generalized_dice_loss.py
@@ -184,7 +184,7 @@ def test_differentiability(self):
 
         generalized_dice_loss = GeneralizedDiceLoss()
         loss = generalized_dice_loss(prediction, target)
-        self.assertNotEqual(loss.grad_fn, None)
+        self.assertIsNotNone(loss.grad_fn)
 
     def test_batch(self):
         prediction = torch.zeros(2, 3, 3, 3)
@@ -194,7 +194,7 @@ def test_batch(self):
 
         generalized_dice_loss = GeneralizedDiceLoss(batch=True)
         loss = generalized_dice_loss(prediction, target)
-        self.assertNotEqual(loss.grad_fn, None)
+        self.assertIsNotNone(loss.grad_fn)
 
     def test_script(self):
         loss = GeneralizedDiceLoss()
diff --git a/tests/test_get_package_version.py b/tests/test_get_package_version.py
index ab9e69cd31..e9e1d8eca6 100644
--- a/tests/test_get_package_version.py
+++ b/tests/test_get_package_version.py
@@ -20,14 +20,14 @@ class TestGetVersion(unittest.TestCase):
 
     def test_default(self):
         output = get_package_version("42foobarnoexist")
-        self.assertTrue("UNKNOWN" in output)
+        self.assertIn("UNKNOWN", output)
 
         output = get_package_version("numpy")
-        self.assertFalse("UNKNOWN" in output)
+        self.assertNotIn("UNKNOWN", output)
 
     def test_msg(self):
         output = get_package_version("42foobarnoexist", "test")
-        self.assertTrue("test" in output)
+        self.assertIn("test", output)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_grid_patch.py b/tests/test_grid_patch.py
index 4b324eda1a..56af123548 100644
--- a/tests/test_grid_patch.py
+++ b/tests/test_grid_patch.py
@@ -124,11 +124,11 @@ def test_grid_patch_meta(self, input_parameters, image, expected, expected_meta)
             self.assertTrue(output.meta["path"] == expected_meta[0]["path"])
         for output_patch, expected_patch, expected_patch_meta in zip(output, expected, expected_meta):
             assert_allclose(output_patch, expected_patch, type_test=False)
-            self.assertTrue(isinstance(output_patch, MetaTensor))
-            self.assertTrue(output_patch.meta["location"] == expected_patch_meta["location"])
+            self.assertIsInstance(output_patch, MetaTensor)
+            self.assertEqual(output_patch.meta["location"], expected_patch_meta["location"])
             self.assertTrue(output_patch.meta["spatial_shape"], list(output_patch.shape[1:]))
             if "path" in expected_meta[0]:
-                self.assertTrue(output_patch.meta["path"] == expected_patch_meta["path"])
+                self.assertEqual(output_patch.meta["path"], expected_patch_meta["path"])
 
 
 if __name__ == "__main__":
diff --git a/tests/test_integration_bundle_run.py b/tests/test_integration_bundle_run.py
index c2e0fb55b7..60aaef05bf 100644
--- a/tests/test_integration_bundle_run.py
+++ b/tests/test_integration_bundle_run.py
@@ -135,9 +135,8 @@ def test_scripts_fold(self):
         command_run = cmd + ["run", "training", "--config_file", config_file, "--meta_file", meta_file]
         completed_process = subprocess.run(command_run, check=True, capture_output=True, text=True)
         output = repr(completed_process.stdout).replace("\\n", "\n").replace("\\t", "\t")  # Get the captured output
-        print(output)
 
-        self.assertTrue(expected_condition in output)
+        self.assertIn(expected_condition, output)
         command_run_workflow = cmd + [
             "run_workflow",
             "--run_id",
@@ -149,8 +148,7 @@ def test_scripts_fold(self):
         ]
         completed_process = subprocess.run(command_run_workflow, check=True, capture_output=True, text=True)
         output = repr(completed_process.stdout).replace("\\n", "\n").replace("\\t", "\t")  # Get the captured output
-        print(output)
-        self.assertTrue(expected_condition in output)
+        self.assertIn(expected_condition, output)
 
         # test missing meta file
         self.assertIn("ERROR", command_line_tests(cmd + ["run", "training", "--config_file", config_file]))
diff --git a/tests/test_inverse_collation.py b/tests/test_inverse_collation.py
index f33b5c67eb..bf3972e6bd 100644
--- a/tests/test_inverse_collation.py
+++ b/tests/test_inverse_collation.py
@@ -133,7 +133,7 @@ def test_collation(self, _, transform, collate_fn, ndim):
             d = decollate_batch(item)
             self.assertTrue(len(d) <= self.batch_size)
             for b in d:
-                self.assertTrue(isinstance(b["image"], MetaTensor))
+                self.assertIsInstance(b["image"], MetaTensor)
                 np.testing.assert_array_equal(
                     b["image"].applied_operations[-1]["orig_size"], b["label"].applied_operations[-1]["orig_size"]
                 )
diff --git a/tests/test_load_imaged.py b/tests/test_load_imaged.py
index 699ed70059..914240c705 100644
--- a/tests/test_load_imaged.py
+++ b/tests/test_load_imaged.py
@@ -190,7 +190,7 @@ def test_correct(self, input_p, expected_shape, track_meta):
                 self.assertTrue(hasattr(r, "affine"))
                 self.assertIsInstance(r.affine, torch.Tensor)
                 self.assertEqual(r.meta["space"], "RAS")
-                self.assertTrue("qform_code" not in r.meta)
+                self.assertNotIn("qform_code", r.meta)
             else:
                 self.assertIsInstance(r, torch.Tensor)
                 self.assertNotIsInstance(r, MetaTensor)
diff --git a/tests/test_look_up_option.py b/tests/test_look_up_option.py
index d40b7eaa8c..75560b4ac4 100644
--- a/tests/test_look_up_option.py
+++ b/tests/test_look_up_option.py
@@ -56,7 +56,7 @@ def test_default(self):
 
     def test_str_enum(self):
         output = look_up_option("C", {"A", "B"}, default=None)
-        self.assertEqual(output, None)
+        self.assertIsNone(output)
         self.assertEqual(list(_CaseStrEnum), ["A", "B"])
         self.assertEqual(_CaseStrEnum.MODE_A, "A")
         self.assertEqual(str(_CaseStrEnum.MODE_A), "A")
diff --git a/tests/test_matshow3d.py b/tests/test_matshow3d.py
index e513025e69..e54bb523e4 100644
--- a/tests/test_matshow3d.py
+++ b/tests/test_matshow3d.py
@@ -78,7 +78,7 @@ def test_samples(self):
         fig, mat = matshow3d(
             [im[keys] for im in ims], title=f"testing {keys}", figsize=(2, 2), frames_per_row=5, every_n=2, show=False
         )
-        self.assertTrue(mat.dtype == np.float32)
+        self.assertEqual(mat.dtype, np.float32)
 
         with tempfile.TemporaryDirectory() as tempdir:
             tempimg = f"{tempdir}/matshow3d_patch_test.png"
diff --git a/tests/test_mednistdataset.py b/tests/test_mednistdataset.py
index 1db632c144..c1b21e9373 100644
--- a/tests/test_mednistdataset.py
+++ b/tests/test_mednistdataset.py
@@ -41,7 +41,7 @@ def _test_dataset(dataset):
             self.assertEqual(len(dataset), int(MEDNIST_FULL_DATASET_LENGTH * dataset.test_frac))
             self.assertTrue("image" in dataset[0])
             self.assertTrue("label" in dataset[0])
-            self.assertTrue(isinstance(dataset[0]["image"], MetaTensor))
+            self.assertIsInstance(dataset[0]["image"], MetaTensor)
             self.assertTupleEqual(dataset[0]["image"].shape, (1, 64, 64))
 
         with skip_if_downloading_fails():
diff --git a/tests/test_meta_tensor.py b/tests/test_meta_tensor.py
index 1e0f188b63..f31a07eba4 100644
--- a/tests/test_meta_tensor.py
+++ b/tests/test_meta_tensor.py
@@ -222,9 +222,9 @@ def test_stack(self, device, dtype):
 
     def test_get_set_meta_fns(self):
         set_track_meta(False)
-        self.assertEqual(get_track_meta(), False)
+        self.assertFalse(get_track_meta())
         set_track_meta(True)
-        self.assertEqual(get_track_meta(), True)
+        self.assertTrue(get_track_meta())
 
     @parameterized.expand(TEST_DEVICES)
     def test_torchscript(self, device):
diff --git a/tests/test_mmar_download.py b/tests/test_mmar_download.py
index 6af3d09fb2..2ac73a8149 100644
--- a/tests/test_mmar_download.py
+++ b/tests/test_mmar_download.py
@@ -142,7 +142,7 @@ def test_load_ckpt(self, input_args, expected_name, expected_val):
     def test_unique(self):
         # model ids are unique
         keys = sorted(m["id"] for m in MODEL_DESC)
-        self.assertTrue(keys == sorted(set(keys)))
+        self.assertEqual(keys, sorted(set(keys)))
 
     def test_search(self):
         self.assertEqual(_get_val({"a": 1, "b": 2}, key="b"), 2)
diff --git a/tests/test_rand_affined.py b/tests/test_rand_affined.py
index 950058a9e9..eb8ebd06c5 100644
--- a/tests/test_rand_affined.py
+++ b/tests/test_rand_affined.py
@@ -240,7 +240,7 @@ def test_rand_affined(self, input_param, input_data, expected_val, track_meta):
             resampler.lazy = False
 
         if input_param.get("cache_grid", False):
-            self.assertTrue(g.rand_affine._cached_grid is not None)
+            self.assertIsNotNone(g.rand_affine._cached_grid)
         for key in res:
             if isinstance(key, str) and key.endswith("_transforms"):
                 continue
diff --git a/tests/test_rand_bias_field.py b/tests/test_rand_bias_field.py
index 333a9ecba5..328f46b7ee 100644
--- a/tests/test_rand_bias_field.py
+++ b/tests/test_rand_bias_field.py
@@ -39,7 +39,7 @@ def test_output_shape(self, class_args, img_shape):
                 img = p(np.random.rand(*img_shape))
                 output = bias_field(img)
                 np.testing.assert_equal(output.shape, img_shape)
-                self.assertTrue(output.dtype in (np.float32, torch.float32))
+                self.assertIn(output.dtype, (np.float32, torch.float32))
 
                 img_zero = np.zeros([*img_shape])
                 output_zero = bias_field(img_zero)
diff --git a/tests/test_resnet.py b/tests/test_resnet.py
index 449edba4bf..22890daeea 100644
--- a/tests/test_resnet.py
+++ b/tests/test_resnet.py
@@ -228,7 +228,7 @@ def test_resnet_shape(self, model, input_param, input_shape, expected_shape):
             if input_param.get("feed_forward", True):
                 self.assertEqual(result.shape, expected_shape)
             else:
-                self.assertTrue(result.shape in expected_shape)
+                self.assertIn(result.shape, expected_shape)
 
     @parameterized.expand(PRETRAINED_TEST_CASES)
     @skip_if_quick
diff --git a/tests/test_to_cupy.py b/tests/test_to_cupy.py
index 5a1754e7c5..38400f0d3f 100644
--- a/tests/test_to_cupy.py
+++ b/tests/test_to_cupy.py
@@ -62,8 +62,8 @@ def test_numpy_input_dtype(self):
         test_data = np.rot90(test_data)
         self.assertFalse(test_data.flags["C_CONTIGUOUS"])
         result = ToCupy(np.uint8)(test_data)
-        self.assertTrue(result.dtype == cp.uint8)
-        self.assertTrue(isinstance(result, cp.ndarray))
+        self.assertEqual(result.dtype, cp.uint8)
+        self.assertIsInstance(result, cp.ndarray)
         self.assertTrue(result.flags["C_CONTIGUOUS"])
         cp.testing.assert_allclose(result, test_data)
 
@@ -72,8 +72,8 @@ def test_tensor_input(self):
         test_data = test_data.rot90()
         self.assertFalse(test_data.is_contiguous())
         result = ToCupy()(test_data)
-        self.assertTrue(result.dtype == cp.float32)
-        self.assertTrue(isinstance(result, cp.ndarray))
+        self.assertEqual(result.dtype, cp.float32)
+        self.assertIsInstance(result, cp.ndarray)
         self.assertTrue(result.flags["C_CONTIGUOUS"])
         cp.testing.assert_allclose(result, test_data)
 
@@ -83,8 +83,8 @@ def test_tensor_cuda_input(self):
         test_data = test_data.rot90()
         self.assertFalse(test_data.is_contiguous())
         result = ToCupy()(test_data)
-        self.assertTrue(result.dtype == cp.float32)
-        self.assertTrue(isinstance(result, cp.ndarray))
+        self.assertEqual(result.dtype, cp.float32)
+        self.assertIsInstance(result, cp.ndarray)
         self.assertTrue(result.flags["C_CONTIGUOUS"])
         cp.testing.assert_allclose(result, test_data)
 
@@ -95,8 +95,8 @@ def test_tensor_cuda_input_dtype(self):
         self.assertFalse(test_data.is_contiguous())
 
         result = ToCupy(dtype="float32")(test_data)
-        self.assertTrue(result.dtype == cp.float32)
-        self.assertTrue(isinstance(result, cp.ndarray))
+        self.assertEqual(result.dtype, cp.float32)
+        self.assertIsInstance(result, cp.ndarray)
         self.assertTrue(result.flags["C_CONTIGUOUS"])
         cp.testing.assert_allclose(result, test_data)
 
diff --git a/tests/test_to_numpy.py b/tests/test_to_numpy.py
index f92b7c0075..f4e5f80a29 100644
--- a/tests/test_to_numpy.py
+++ b/tests/test_to_numpy.py
@@ -32,7 +32,7 @@ def test_cupy_input(self):
         test_data = cp.rot90(test_data)
         self.assertFalse(test_data.flags["C_CONTIGUOUS"])
         result = ToNumpy()(test_data)
-        self.assertTrue(isinstance(result, np.ndarray))
+        self.assertIsInstance(result, np.ndarray)
         self.assertTrue(result.flags["C_CONTIGUOUS"])
         assert_allclose(result, test_data.get(), type_test=False)
 
@@ -41,8 +41,8 @@ def test_numpy_input(self):
         test_data = np.rot90(test_data)
         self.assertFalse(test_data.flags["C_CONTIGUOUS"])
         result = ToNumpy(dtype="float32")(test_data)
-        self.assertTrue(isinstance(result, np.ndarray))
-        self.assertTrue(result.dtype == np.float32)
+        self.assertIsInstance(result, np.ndarray)
+        self.assertEqual(result.dtype, np.float32)
         self.assertTrue(result.flags["C_CONTIGUOUS"])
         assert_allclose(result, test_data, type_test=False)
 
@@ -51,7 +51,7 @@ def test_tensor_input(self):
         test_data = test_data.rot90()
         self.assertFalse(test_data.is_contiguous())
         result = ToNumpy(dtype=torch.uint8)(test_data)
-        self.assertTrue(isinstance(result, np.ndarray))
+        self.assertIsInstance(result, np.ndarray)
         self.assertTrue(result.flags["C_CONTIGUOUS"])
         assert_allclose(result, test_data, type_test=False)
 
@@ -61,7 +61,7 @@ def test_tensor_cuda_input(self):
         test_data = test_data.rot90()
         self.assertFalse(test_data.is_contiguous())
         result = ToNumpy()(test_data)
-        self.assertTrue(isinstance(result, np.ndarray))
+        self.assertIsInstance(result, np.ndarray)
         self.assertTrue(result.flags["C_CONTIGUOUS"])
         assert_allclose(result, test_data, type_test=False)
 
@@ -77,7 +77,7 @@ def test_list_tuple(self):
     def test_single_value(self):
         for test_data in [5, np.array(5), torch.tensor(5)]:
             result = ToNumpy(dtype=np.uint8)(test_data)
-            self.assertTrue(isinstance(result, np.ndarray))
+            self.assertIsInstance(result, np.ndarray)
             assert_allclose(result, np.asarray(test_data), type_test=False)
             self.assertEqual(result.ndim, 0)
 
diff --git a/tests/test_torchvision_fc_model.py b/tests/test_torchvision_fc_model.py
index 322cce1161..9cc19db62c 100644
--- a/tests/test_torchvision_fc_model.py
+++ b/tests/test_torchvision_fc_model.py
@@ -195,8 +195,8 @@ def test_get_module(self):
         mod = look_up_named_module("model.1.submodule.1.submodule.1.submodule.0.conv", net)
         self.assertTrue(str(mod).startswith("Conv2d"))
         self.assertIsInstance(set_named_module(net, "model", torch.nn.Identity()).model, torch.nn.Identity)
-        self.assertEqual(look_up_named_module("model.1.submodule.1.submodule.1.submodule.conv", net), None)
-        self.assertEqual(look_up_named_module("test attribute", net), None)
+        self.assertIsNone(look_up_named_module("model.1.submodule.1.submodule.1.submodule.conv", net))
+        self.assertIsNone(look_up_named_module("test attribute", net))
 
 
 if __name__ == "__main__":
diff --git a/tests/test_traceable_transform.py b/tests/test_traceable_transform.py
index dd139053e3..6a499b2dd9 100644
--- a/tests/test_traceable_transform.py
+++ b/tests/test_traceable_transform.py
@@ -33,12 +33,12 @@ def test_default(self):
         expected_key = "_transforms"
         a = _TraceTest()
         for x in a.transform_info_keys():
-            self.assertTrue(x in a.get_transform_info())
+            self.assertIn(x, a.get_transform_info())
         self.assertEqual(a.trace_key(), expected_key)
 
         data = {"image": "test"}
         data = a(data)  # adds to the stack
-        self.assertTrue(isinstance(data[expected_key], list))
+        self.assertIsInstance(data[expected_key], list)
         self.assertEqual(data[expected_key][0]["class"], "_TraceTest")
 
         data = a(data)  # adds to the stack

From 8c709de12010929307060628a450741e1263e8e5 Mon Sep 17 00:00:00 2001
From: Matthew Vine <32849887+MattTheCuber@users.noreply.github.com>
Date: Wed, 24 Apr 2024 23:10:30 -0400
Subject: [PATCH 038/183] Fix download failing on FIPS machines (#7698)

### Description
This PR fixes downloads failing on FIPS enabled machines due to insecure
MD5 hashing. The two solutions are to disable MD5 hashing (SHA1 is
allowed and faster), or use the `usedforsecurity=False` flag. This PR
uses the second method. However, the `usedforsecurity` flag only works
for Python 3.9 and later (which was accounted for). Let me know if you
have a better implementation to solve this issue.

The error thrown on FIPS enabled machine is:
```ValueError: [digital envelope routines: EVP_DigestInit_ex] disabled for FIPS```

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not applicable items -->
- [x] Non-breaking change (fix or new feature that would not break existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u --net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick --unittests  --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/` folder.

---------

Signed-off-by: Matthew Vine <32849887+MattTheCuber@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/apps/utils.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/monai/apps/utils.py b/monai/apps/utils.py
index db541923b5..0c998146a3 100644
--- a/monai/apps/utils.py
+++ b/monai/apps/utils.py
@@ -135,7 +135,12 @@ def check_hash(filepath: PathLike, val: str | None = None, hash_type: str = "md5
         logger.info(f"Expected {hash_type} is None, skip {hash_type} check for file {filepath}.")
         return True
     actual_hash_func = look_up_option(hash_type.lower(), SUPPORTED_HASH_TYPES)
-    actual_hash = actual_hash_func()
+
+    if sys.version_info >= (3, 9):
+        actual_hash = actual_hash_func(usedforsecurity=False)  # allows checks on FIPS enabled machines
+    else:
+        actual_hash = actual_hash_func()
+
     try:
         with open(filepath, "rb") as f:
             for chunk in iter(lambda: f.read(1024 * 1024), b""):

From 6a130cc7f659efe66cf651369e35d4a93b2ec28a Mon Sep 17 00:00:00 2001
From: binliunls <107988372+binliunls@users.noreply.github.com>
Date: Fri, 26 Apr 2024 13:18:06 +0800
Subject: [PATCH 039/183] 7713 Update TRT parameter (#7714)

Fixes #7713  .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: binliu <binliu@nvidia.com>
---
 monai/networks/utils.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/monai/networks/utils.py b/monai/networks/utils.py
index 7e3d2e2170..152911f443 100644
--- a/monai/networks/utils.py
+++ b/monai/networks/utils.py
@@ -986,6 +986,7 @@ def scale_batch_size(input_shape: Sequence[int], scale_num: int):
                     inputs=input_placeholder,
                     enabled_precisions=convert_precision,
                     device=target_device,
+                    ir="torchscript",
                     **kwargs,
                 )
 

From 4c193ead5aa16c23bee55f62cb225349df340c72 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 29 Apr 2024 14:18:00 +0800
Subject: [PATCH 040/183] Fix itk install error when python=3.8 (#7719)

Fixes #7716

set python 3.9 for pythonapp.yml

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/pythonapp.yml           | 4 ++--
 tests/test_clip_intensity_percentilesd.py | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/pythonapp.yml b/.github/workflows/pythonapp.yml
index d4043585cc..b8b73907d4 100644
--- a/.github/workflows/pythonapp.yml
+++ b/.github/workflows/pythonapp.yml
@@ -70,10 +70,10 @@ jobs:
         maximum-size: 16GB
         disk-root: "D:"
     - uses: actions/checkout@v4
-    - name: Set up Python 3.8
+    - name: Set up Python 3.9
       uses: actions/setup-python@v5
       with:
-        python-version: '3.8'
+        python-version: '3.9'
     - name: Prepare pip wheel
       run: |
         which python
diff --git a/tests/test_clip_intensity_percentilesd.py b/tests/test_clip_intensity_percentilesd.py
index fa727b6adb..ed4fc588cb 100644
--- a/tests/test_clip_intensity_percentilesd.py
+++ b/tests/test_clip_intensity_percentilesd.py
@@ -96,7 +96,7 @@ def test_channel_wise(self, p):
         for i, c in enumerate(im):
             lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
-            assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-4, atol=0)
+            assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-3, atol=0)
 
     def test_ill_sharpness_factor(self):
         key = "img"

From 56508992c32e328667740e18f99b82a6a19f2c2a Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Mon, 29 Apr 2024 09:07:04 +0100
Subject: [PATCH 041/183] auto updates (#7723)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/nets/resnet.py | 2 --
 tests/test_median_filter.py   | 1 +
 tests/test_resnet.py          | 1 -
 3 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/monai/networks/nets/resnet.py b/monai/networks/nets/resnet.py
index 99975271da..de16cd4cca 100644
--- a/monai/networks/nets/resnet.py
+++ b/monai/networks/nets/resnet.py
@@ -46,7 +46,6 @@
     "resnet200",
 ]
 
-
 resnet_params = {
     # model_name: (block, layers, shortcut_type, bias_downsample, datasets23)
     "resnet10": ("basic", [1, 1, 1, 1], "B", False, True),
@@ -58,7 +57,6 @@
     "resnet200": ("bottleneck", [3, 24, 36, 3], "B", False, False),
 }
 
-
 logger = logging.getLogger(__name__)
 
 
diff --git a/tests/test_median_filter.py b/tests/test_median_filter.py
index 516388afce..02fa812380 100644
--- a/tests/test_median_filter.py
+++ b/tests/test_median_filter.py
@@ -21,6 +21,7 @@
 
 
 class MedianFilterTestCase(unittest.TestCase):
+
     @parameterized.expand([(torch.ones(1, 1, 2, 3, 5), [1, 2, 4]), (torch.ones(1, 1, 4, 3, 4), 1)])  # 3d_big  # 3d
     def test_3d(self, input_tensor, radius):
         filter = MedianFilter(radius).to(torch.device("cpu:0"))
diff --git a/tests/test_resnet.py b/tests/test_resnet.py
index 22890daeea..ecab133154 100644
--- a/tests/test_resnet.py
+++ b/tests/test_resnet.py
@@ -198,7 +198,6 @@
     [model, *TEST_CASE_1] for model in [resnet10, resnet18, resnet34, resnet50, resnet101, resnet152, resnet200]
 ]
 
-
 CASE_EXTRACT_FEATURES = [
     (
         {"model_name": "resnet10", "pretrained": True, "spatial_dims": 3, "in_channels": 1},

From e1a69b03c86ce065db2816b696ea4a6b57d46435 Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Tue, 30 Apr 2024 06:27:19 +0100
Subject: [PATCH 042/183] Auto3DSeg algo_template hash update (#7728)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 monai/utils/misc.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index 84e979abf4..ab9fe259aa 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -527,7 +527,7 @@ def doc_images() -> str | None:
 
     @staticmethod
     def algo_hash() -> str | None:
-        return os.environ.get("MONAI_ALGO_HASH", "07acb39")
+        return os.environ.get("MONAI_ALGO_HASH", "e4cf5a1")
 
     @staticmethod
     def trace_transform() -> str | None:

From fe733b0ff1951ee752ab87ebfe5c4b7c82d30579 Mon Sep 17 00:00:00 2001
From: Pkaps25 <43655728+Pkaps25@users.noreply.github.com>
Date: Mon, 6 May 2024 23:55:31 -0400
Subject: [PATCH 043/183] Propagate kernel size through attention
 Attention-UNet (#7734)

Fixes #7726.

### Description
Passes the `kernel_size` parameter to `ConvBlocks` within Attention
UNet, creating a net with the expected number of parameters.

Using the example in #7726 on this branch:
```
from monai.networks.nets import AttentionUnet

model = AttentionUnet(
        spatial_dims = 2,
        in_channels = 1,
        out_channels = 1,
        channels = (2, 4, 8, 16),
        strides = (2,2,2),
        kernel_size = 5,
        up_kernel_size = 5
)
```
outputs the expected values:
```
Total params: 18,846
Trainable params: 18,846
Non-trainable params: 0
Total mult-adds (M): 0.37
```

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Peter Kaplinsky <peterkaplinsky@gmail.com>
Co-authored-by: Peter Kaplinsky <peterkaplinsky@gmail.com>
---
 monai/networks/nets/attentionunet.py | 12 ++++++++++--
 tests/test_attentionunet.py          | 20 ++++++++++++++++++++
 2 files changed, 30 insertions(+), 2 deletions(-)

diff --git a/monai/networks/nets/attentionunet.py b/monai/networks/nets/attentionunet.py
index 5689cf1071..fdf31d9701 100644
--- a/monai/networks/nets/attentionunet.py
+++ b/monai/networks/nets/attentionunet.py
@@ -29,7 +29,7 @@ def __init__(
         spatial_dims: int,
         in_channels: int,
         out_channels: int,
-        kernel_size: int = 3,
+        kernel_size: Sequence[int] | int = 3,
         strides: int = 1,
         dropout=0.0,
     ):
@@ -219,7 +219,13 @@ def __init__(
         self.kernel_size = kernel_size
         self.dropout = dropout
 
-        head = ConvBlock(spatial_dims=spatial_dims, in_channels=in_channels, out_channels=channels[0], dropout=dropout)
+        head = ConvBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=channels[0],
+            dropout=dropout,
+            kernel_size=self.kernel_size,
+        )
         reduce_channels = Convolution(
             spatial_dims=spatial_dims,
             in_channels=channels[0],
@@ -245,6 +251,7 @@ def _create_block(channels: Sequence[int], strides: Sequence[int]) -> nn.Module:
                             out_channels=channels[1],
                             strides=strides[0],
                             dropout=self.dropout,
+                            kernel_size=self.kernel_size,
                         ),
                         subblock,
                     ),
@@ -271,6 +278,7 @@ def _get_bottom_layer(self, in_channels: int, out_channels: int, strides: int) -
                 out_channels=out_channels,
                 strides=strides,
                 dropout=self.dropout,
+                kernel_size=self.kernel_size,
             ),
             up_kernel_size=self.up_kernel_size,
             strides=strides,
diff --git a/tests/test_attentionunet.py b/tests/test_attentionunet.py
index 83f6cabc5e..6a577f763f 100644
--- a/tests/test_attentionunet.py
+++ b/tests/test_attentionunet.py
@@ -14,11 +14,17 @@
 import unittest
 
 import torch
+import torch.nn as nn
 
 import monai.networks.nets.attentionunet as att
 from tests.utils import skip_if_no_cuda, skip_if_quick
 
 
+def get_net_parameters(net: nn.Module) -> int:
+    """Returns the total number of parameters in a Module."""
+    return sum(param.numel() for param in net.parameters())
+
+
 class TestAttentionUnet(unittest.TestCase):
 
     def test_attention_block(self):
@@ -50,6 +56,20 @@ def test_attentionunet(self):
             self.assertEqual(output.shape[0], input.shape[0])
             self.assertEqual(output.shape[1], 2)
 
+    def test_attentionunet_kernel_size(self):
+        args_dict = {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 2,
+            "channels": (3, 4, 5),
+            "up_kernel_size": 5,
+            "strides": (1, 2),
+        }
+        model_a = att.AttentionUnet(**args_dict, kernel_size=5)
+        model_b = att.AttentionUnet(**args_dict, kernel_size=7)
+        self.assertEqual(get_net_parameters(model_a), 3534)
+        self.assertEqual(get_net_parameters(model_b), 5574)
+
     @skip_if_no_cuda
     def test_attentionunet_gpu(self):
         for dims in [2, 3]:

From ecaf5a1c9df5a462e3e1989db381865a3bed56a2 Mon Sep 17 00:00:00 2001
From: Simon Jensen <61684806+simojens@users.noreply.github.com>
Date: Tue, 7 May 2024 16:48:58 +0200
Subject: [PATCH 044/183] Fixed misguiding weight mode documentation (#7746)

### Description

DeepSupervisionLoss has a small typo in the documentation for 'exp'
weight mode. The correct weights for this mode should be starting at 1,
not 0 and then decrease by power of two. Might be worth fixing as it
made me check the source code.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Simon Jensen <61684806+simojens@users.noreply.github.com>
---
 monai/losses/ds_loss.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/losses/ds_loss.py b/monai/losses/ds_loss.py
index 57fcff6b87..aacc16874d 100644
--- a/monai/losses/ds_loss.py
+++ b/monai/losses/ds_loss.py
@@ -33,7 +33,7 @@ def __init__(self, loss: _Loss, weight_mode: str = "exp", weights: list[float] |
             weight_mode: {``"same"``, ``"exp"``, ``"two"``}
                 Specifies the weights calculation for each image level. Defaults to ``"exp"``.
                 - ``"same"``: all weights are equal to 1.
-                - ``"exp"``: exponentially decreasing weights by a power of 2: 0, 0.5, 0.25, 0.125, etc .
+                - ``"exp"``: exponentially decreasing weights by a power of 2: 1, 0.5, 0.25, 0.125, etc .
                 - ``"two"``: equal smaller weights for lower levels: 1, 0.5, 0.5, 0.5, 0.5, etc
             weights: a list of weights to apply to each deeply supervised sub-loss, if provided, this will be used
                 regardless of the weight_mode

From 32b7754e98486b5edf01b8de3c08798a5e176566 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 8 May 2024 09:17:56 +0800
Subject: [PATCH 045/183] Enhance logging logic in `ConfigWorkflow` (#7745)

When using nvflare with monai bundle, the logging logic may be overrided
by the logging logic in the bundle.
Add an option to disable logging logic in the bundle.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/bundle/workflows.py     |  9 ++++++---
 monai/fl/client/monai_algo.py | 14 +++++++++-----
 monai/fl/utils/constants.py   |  1 +
 3 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/monai/bundle/workflows.py b/monai/bundle/workflows.py
index 471088994b..11c9bf0562 100644
--- a/monai/bundle/workflows.py
+++ b/monai/bundle/workflows.py
@@ -239,6 +239,7 @@ class ConfigWorkflow(BundleWorkflow):
         logging_file: config file for `logging` module in the program. for more details:
             https://docs.python.org/3/library/logging.config.html#logging.config.fileConfig.
             If None, default to "configs/logging.conf", which is commonly used for bundles in MONAI model zoo.
+            If False, the logging logic for the bundle will not be modified.
         init_id: ID name of the expected config expression to initialize before running, default to "initialize".
             allow a config to have no `initialize` logic and the ID.
         run_id: ID name of the expected config expression to run, default to "run".
@@ -278,7 +279,7 @@ def __init__(
         self,
         config_file: str | Sequence[str],
         meta_file: str | Sequence[str] | None = None,
-        logging_file: str | None = None,
+        logging_file: str | bool | None = None,
         init_id: str = "initialize",
         run_id: str = "run",
         final_id: str = "finalize",
@@ -307,7 +308,9 @@ def __init__(
         super().__init__(workflow_type=workflow_type, meta_file=meta_file, properties_path=properties_path)
         self.config_root_path = config_root_path
         logging_file = str(self.config_root_path / "logging.conf") if logging_file is None else logging_file
-        if logging_file is not None:
+        if logging_file is False:
+            logger.warn(f"Logging file is set to {logging_file}, skipping logging.")
+        else:
             if not os.path.isfile(logging_file):
                 if logging_file == str(self.config_root_path / "logging.conf"):
                     logger.warn(f"Default logging file in {logging_file} does not exist, skipping logging.")
@@ -315,7 +318,7 @@ def __init__(
                     raise FileNotFoundError(f"Cannot find the logging config file: {logging_file}.")
             else:
                 logger.info(f"Setting logging properties based on config: {logging_file}.")
-                fileConfig(logging_file, disable_existing_loggers=False)
+                fileConfig(str(logging_file), disable_existing_loggers=False)
 
         self.parser = ConfigParser()
         self.parser.read_config(f=config_file)
diff --git a/monai/fl/client/monai_algo.py b/monai/fl/client/monai_algo.py
index 9acf131bd9..a3ac58c221 100644
--- a/monai/fl/client/monai_algo.py
+++ b/monai/fl/client/monai_algo.py
@@ -134,12 +134,14 @@ def initialize(self, extra=None):
 
         Args:
             extra: Dict with additional information that should be provided by FL system,
-                i.e., `ExtraItems.CLIENT_NAME` and `ExtraItems.APP_ROOT`.
+                i.e., `ExtraItems.CLIENT_NAME`, `ExtraItems.APP_ROOT` and `ExtraItems.LOGGING_FILE`.
+                You can diable the logging logic in the monai bundle by setting {ExtraItems.LOGGING_FILE} to False.
 
         """
         if extra is None:
             extra = {}
         self.client_name = extra.get(ExtraItems.CLIENT_NAME, "noname")
+        logging_file = extra.get(ExtraItems.LOGGING_FILE, None)
         self.logger.info(f"Initializing {self.client_name} ...")
 
         # FL platform needs to provide filepath to configuration files
@@ -149,7 +151,7 @@ def initialize(self, extra=None):
         if self.workflow is None:
             config_train_files = self._add_config_files(self.config_train_filename)
             self.workflow = ConfigWorkflow(
-                config_file=config_train_files, meta_file=None, logging_file=None, workflow_type="train"
+                config_file=config_train_files, meta_file=None, logging_file=logging_file, workflow_type="train"
             )
         self.workflow.initialize()
         self.workflow.bundle_root = self.bundle_root
@@ -412,13 +414,15 @@ def initialize(self, extra=None):
 
         Args:
             extra: Dict with additional information that should be provided by FL system,
-                i.e., `ExtraItems.CLIENT_NAME` and `ExtraItems.APP_ROOT`.
+                i.e., `ExtraItems.CLIENT_NAME`, `ExtraItems.APP_ROOT` and `ExtraItems.LOGGING_FILE`.
+                You can diable the logging logic in the monai bundle by setting {ExtraItems.LOGGING_FILE} to False.
 
         """
         self._set_cuda_device()
         if extra is None:
             extra = {}
         self.client_name = extra.get(ExtraItems.CLIENT_NAME, "noname")
+        logging_file = extra.get(ExtraItems.LOGGING_FILE, None)
         timestamp = time.strftime("%Y%m%d_%H%M%S")
         self.logger.info(f"Initializing {self.client_name} ...")
         # FL platform needs to provide filepath to configuration files
@@ -434,7 +438,7 @@ def initialize(self, extra=None):
             self.train_workflow = ConfigWorkflow(
                 config_file=config_train_files,
                 meta_file=None,
-                logging_file=None,
+                logging_file=logging_file,
                 workflow_type="train",
                 **self.train_kwargs,
             )
@@ -459,7 +463,7 @@ def initialize(self, extra=None):
             self.eval_workflow = ConfigWorkflow(
                 config_file=config_eval_files,
                 meta_file=None,
-                logging_file=None,
+                logging_file=logging_file,
                 workflow_type=self.eval_workflow_name,
                 **self.eval_kwargs,
             )
diff --git a/monai/fl/utils/constants.py b/monai/fl/utils/constants.py
index eda1a6b4f9..18beceeaee 100644
--- a/monai/fl/utils/constants.py
+++ b/monai/fl/utils/constants.py
@@ -30,6 +30,7 @@ class ExtraItems(StrEnum):
     CLIENT_NAME = "fl_client_name"
     APP_ROOT = "fl_app_root"
     STATS_SENDER = "fl_stats_sender"
+    LOGGING_FILE = "logging_file"
 
 
 class FlPhase(StrEnum):

From f278e51b8c13fa46d3f5e763452d9e70e24df955 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 8 May 2024 12:32:50 +0800
Subject: [PATCH 046/183] Add version requirement for filelock and nni (#7744)

Add version requirement for filelock and nni

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 Dockerfile           | 12 ++++--------
 requirements-dev.txt |  5 ++---
 2 files changed, 6 insertions(+), 11 deletions(-)

diff --git a/Dockerfile b/Dockerfile
index fc97227351..8e255597d1 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -18,18 +18,14 @@ LABEL maintainer="monai.contact@gmail.com"
 
 # TODO: remark for issue [revise the dockerfile](https://github.com/zarr-developers/numcodecs/issues/431)
 RUN if [[ $(uname -m) =~ "aarch64" ]]; then \
-        cd /opt && \
-        git clone --branch v0.12.1 --recursive https://github.com/zarr-developers/numcodecs && \
-        pip wheel numcodecs && \
-        rm -r /opt/*.whl && \
-        rm -rf /opt/numcodecs; \
+      export CFLAGS="-O3" && \
+      export DISABLE_NUMCODECS_SSE2=true && \
+      export DISABLE_NUMCODECS_AVX2=true && \
+      pip install numcodecs; \
     fi
 
 WORKDIR /opt/monai
 
-# remove opencv-python before opencv-python-headless installation
-RUN pip uninstall -y opencv && rm /usr/local/lib/python3.10/dist-packages/cv2 -r
-
 # install full deps
 COPY requirements.txt requirements-min.txt requirements-dev.txt /tmp/
 RUN cp /tmp/requirements.txt /tmp/req.bak \
diff --git a/requirements-dev.txt b/requirements-dev.txt
index b207b56b19..e8792c86f3 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -46,15 +46,14 @@ pynrrd
 pre-commit
 pydicom
 h5py
-nni; platform_system == "Linux" and "arm" not in platform_machine and "aarch" not in platform_machine
+nni==2.10.1; platform_system == "Linux" and "arm" not in platform_machine and "aarch" not in platform_machine
 optuna
 git+https://github.com/Project-MONAI/MetricsReloaded@monai-support#egg=MetricsReloaded
 onnx>=1.13.0
 onnxruntime; python_version <= '3.10'
 typeguard<3  # https://github.com/microsoft/nni/issues/5457
-filelock!=3.12.0  # https://github.com/microsoft/nni/issues/5523
+filelock<3.12.0  # https://github.com/microsoft/nni/issues/5523
 zarr
 lpips==0.1.4
 nvidia-ml-py
 huggingface_hub
-opencv-python-headless

From d83fa5660e87d99f9ddcfcfb4695cbdf53a3884c Mon Sep 17 00:00:00 2001
From: NabJa <32510324+NabJa@users.noreply.github.com>
Date: Wed, 8 May 2024 17:51:47 +0200
Subject: [PATCH 047/183] Add dimensionality of heads argument to SABlock
 (#7664)

Fixes #7661.

### Description

The changes made add a parameter (_dim_head_) to set the output
paramters of all the heads in the Self-attention Block (SABlock).
Currently the output dimension is set to be _hidden_size_ and when
increasing the number of heads this is equally distributed among all
heads.

### Example
The original implementation automatically determines
**_equally_distributed_head_dim_**:
(qkv * num_heds * equally_distributed_head_dim = 3*hidden_size
in this example  ->  3 * 8 * 16 = 384)
```
block = SABlock(hidden_size=128, num_heads=8)
x = torch.zeros(1, 256, 128)
x = block.qkv(x)
print(x.shape)
x = block.input_rearrange(x)
print(x.shape)

> torch.Size([1, 256, 384])
> torch.Size([3, 1, 8, 256, 16]) # <- This corresponds to (qkv batch num_heads sequence_length equally_distributed_head_dim)
```

The propesed implementation fixes this by setting the new argument
**_dim_head_:**
```
block_new = SABlock(hidden_size=128, num_heads=8, dim_head=32)
x = torch.zeros(1, 256, 128)
x = block_new.qkv(x)
print(x.shape)
x = block_new.input_rearrange(x)
print(x.shape)

> torch.Size([1, 256, 384])
> torch.Size([3, 1, 8, 256, 32]) # <- This corresponds to (qkv batch num_heads sequence_length dim_head)
```


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: NabJa <nabil.jabareen@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/blocks/selfattention.py | 12 ++++++---
 tests/test_selfattention.py            | 34 ++++++++++++++++++++++++++
 2 files changed, 42 insertions(+), 4 deletions(-)

diff --git a/monai/networks/blocks/selfattention.py b/monai/networks/blocks/selfattention.py
index 7c81c1704f..7b410b1a7c 100644
--- a/monai/networks/blocks/selfattention.py
+++ b/monai/networks/blocks/selfattention.py
@@ -32,6 +32,7 @@ def __init__(
         dropout_rate: float = 0.0,
         qkv_bias: bool = False,
         save_attn: bool = False,
+        dim_head: int | None = None,
     ) -> None:
         """
         Args:
@@ -40,6 +41,7 @@ def __init__(
             dropout_rate (float, optional): fraction of the input units to drop. Defaults to 0.0.
             qkv_bias (bool, optional): bias term for the qkv linear layer. Defaults to False.
             save_attn (bool, optional): to make accessible the attention matrix. Defaults to False.
+            dim_head (int, optional): dimension of each head. Defaults to hidden_size // num_heads.
 
         """
 
@@ -52,14 +54,16 @@ def __init__(
             raise ValueError("hidden size should be divisible by num_heads.")
 
         self.num_heads = num_heads
-        self.out_proj = nn.Linear(hidden_size, hidden_size)
-        self.qkv = nn.Linear(hidden_size, hidden_size * 3, bias=qkv_bias)
+        self.dim_head = hidden_size // num_heads if dim_head is None else dim_head
+        self.inner_dim = self.dim_head * num_heads
+
+        self.out_proj = nn.Linear(self.inner_dim, hidden_size)
+        self.qkv = nn.Linear(hidden_size, self.inner_dim * 3, bias=qkv_bias)
         self.input_rearrange = Rearrange("b h (qkv l d) -> qkv b l h d", qkv=3, l=num_heads)
         self.out_rearrange = Rearrange("b h l d -> b l (h d)")
         self.drop_output = nn.Dropout(dropout_rate)
         self.drop_weights = nn.Dropout(dropout_rate)
-        self.head_dim = hidden_size // num_heads
-        self.scale = self.head_dim**-0.5
+        self.scale = self.dim_head**-0.5
         self.save_attn = save_attn
         self.att_mat = torch.Tensor()
 
diff --git a/tests/test_selfattention.py b/tests/test_selfattention.py
index b8be4fd1b6..0ebed84159 100644
--- a/tests/test_selfattention.py
+++ b/tests/test_selfattention.py
@@ -74,6 +74,40 @@ def test_access_attn_matrix(self):
         matrix_acess_blk(torch.randn(input_shape))
         assert matrix_acess_blk.att_mat.shape == (input_shape[0], input_shape[0], input_shape[1], input_shape[1])
 
+    def test_number_of_parameters(self):
+
+        def count_sablock_params(*args, **kwargs):
+            """Count the number of parameters in a SABlock."""
+            sablock = SABlock(*args, **kwargs)
+            return sum([x.numel() for x in sablock.parameters() if x.requires_grad])
+
+        hidden_size = 128
+        num_heads = 8
+        default_dim_head = hidden_size // num_heads
+
+        # Default dim_head is hidden_size // num_heads
+        nparams_default = count_sablock_params(hidden_size=hidden_size, num_heads=num_heads)
+        nparams_like_default = count_sablock_params(
+            hidden_size=hidden_size, num_heads=num_heads, dim_head=default_dim_head
+        )
+        self.assertEqual(nparams_default, nparams_like_default)
+
+        # Increasing dim_head should increase the number of parameters
+        nparams_custom_large = count_sablock_params(
+            hidden_size=hidden_size, num_heads=num_heads, dim_head=default_dim_head * 2
+        )
+        self.assertGreater(nparams_custom_large, nparams_default)
+
+        # Decreasing dim_head should decrease the number of parameters
+        nparams_custom_small = count_sablock_params(
+            hidden_size=hidden_size, num_heads=num_heads, dim_head=default_dim_head // 2
+        )
+        self.assertGreater(nparams_default, nparams_custom_small)
+
+        # Increasing the number of heads with the default behaviour should not change the number of params.
+        nparams_default_more_heads = count_sablock_params(hidden_size=hidden_size, num_heads=num_heads * 2)
+        self.assertEqual(nparams_default, nparams_default_more_heads)
+
 
 if __name__ == "__main__":
     unittest.main()

From 258f56de80b38378ca4dfbf86436924bbb93ff7e Mon Sep 17 00:00:00 2001
From: Pkaps25 <43655728+Pkaps25@users.noreply.github.com>
Date: Thu, 9 May 2024 09:46:02 -0400
Subject: [PATCH 048/183] Add activation parameter to ResNet (#7749)

Fixes #7653 .

### Description
Includes an `act` parameter to `ResNet` and its submodules to allow for
passing the `inplace` param.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Peter Kaplinsky <peterkaplinsky@gmail.com>
Co-authored-by: Peter Kaplinsky <peterkaplinsky@gmail.com>
---
 monai/networks/nets/resnet.py | 28 +++++++++++++++++-----------
 tests/test_resnet.py          | 19 ++++++++++++++++++-
 2 files changed, 35 insertions(+), 12 deletions(-)

diff --git a/monai/networks/nets/resnet.py b/monai/networks/nets/resnet.py
index de16cd4cca..93a77d7b2a 100644
--- a/monai/networks/nets/resnet.py
+++ b/monai/networks/nets/resnet.py
@@ -23,7 +23,7 @@
 
 from monai.networks.blocks.encoder import BaseEncoder
 from monai.networks.layers.factories import Conv, Norm, Pool
-from monai.networks.layers.utils import get_pool_layer
+from monai.networks.layers.utils import get_act_layer, get_pool_layer
 from monai.utils import ensure_tuple_rep
 from monai.utils.module import look_up_option, optional_import
 
@@ -78,6 +78,7 @@ def __init__(
         spatial_dims: int = 3,
         stride: int = 1,
         downsample: nn.Module | partial | None = None,
+        act: str | tuple = ("relu", {"inplace": True}),
     ) -> None:
         """
         Args:
@@ -86,6 +87,7 @@ def __init__(
             spatial_dims: number of spatial dimensions of the input image.
             stride: stride to use for first conv layer.
             downsample: which downsample layer to use.
+            act: activation type and arguments. Defaults to relu.
         """
         super().__init__()
 
@@ -94,7 +96,7 @@ def __init__(
 
         self.conv1 = conv_type(in_planes, planes, kernel_size=3, padding=1, stride=stride, bias=False)
         self.bn1 = norm_type(planes)
-        self.relu = nn.ReLU(inplace=True)
+        self.act = get_act_layer(name=act)
         self.conv2 = conv_type(planes, planes, kernel_size=3, padding=1, bias=False)
         self.bn2 = norm_type(planes)
         self.downsample = downsample
@@ -105,7 +107,7 @@ def forward(self, x: torch.Tensor) -> torch.Tensor:
 
         out: torch.Tensor = self.conv1(x)
         out = self.bn1(out)
-        out = self.relu(out)
+        out = self.act(out)
 
         out = self.conv2(out)
         out = self.bn2(out)
@@ -114,7 +116,7 @@ def forward(self, x: torch.Tensor) -> torch.Tensor:
             residual = self.downsample(x)
 
         out += residual
-        out = self.relu(out)
+        out = self.act(out)
 
         return out
 
@@ -129,6 +131,7 @@ def __init__(
         spatial_dims: int = 3,
         stride: int = 1,
         downsample: nn.Module | partial | None = None,
+        act: str | tuple = ("relu", {"inplace": True}),
     ) -> None:
         """
         Args:
@@ -137,6 +140,7 @@ def __init__(
             spatial_dims: number of spatial dimensions of the input image.
             stride: stride to use for second conv layer.
             downsample: which downsample layer to use.
+            act: activation type and arguments. Defaults to relu.
         """
 
         super().__init__()
@@ -150,7 +154,7 @@ def __init__(
         self.bn2 = norm_type(planes)
         self.conv3 = conv_type(planes, planes * self.expansion, kernel_size=1, bias=False)
         self.bn3 = norm_type(planes * self.expansion)
-        self.relu = nn.ReLU(inplace=True)
+        self.act = get_act_layer(name=act)
         self.downsample = downsample
         self.stride = stride
 
@@ -159,11 +163,11 @@ def forward(self, x: torch.Tensor) -> torch.Tensor:
 
         out: torch.Tensor = self.conv1(x)
         out = self.bn1(out)
-        out = self.relu(out)
+        out = self.act(out)
 
         out = self.conv2(out)
         out = self.bn2(out)
-        out = self.relu(out)
+        out = self.act(out)
 
         out = self.conv3(out)
         out = self.bn3(out)
@@ -172,7 +176,7 @@ def forward(self, x: torch.Tensor) -> torch.Tensor:
             residual = self.downsample(x)
 
         out += residual
-        out = self.relu(out)
+        out = self.act(out)
 
         return out
 
@@ -202,6 +206,7 @@ class ResNet(nn.Module):
         num_classes: number of output (classifications).
         feed_forward: whether to add the FC layer for the output, default to `True`.
         bias_downsample: whether to use bias term in the downsampling block when `shortcut_type` is 'B', default to `True`.
+        act: activation type and arguments. Defaults to relu.
 
     """
 
@@ -220,6 +225,7 @@ def __init__(
         num_classes: int = 400,
         feed_forward: bool = True,
         bias_downsample: bool = True,  # for backwards compatibility (also see PR #5477)
+        act: str | tuple = ("relu", {"inplace": True}),
     ) -> None:
         super().__init__()
 
@@ -257,7 +263,7 @@ def __init__(
             bias=False,
         )
         self.bn1 = norm_type(self.in_planes)
-        self.relu = nn.ReLU(inplace=True)
+        self.act = get_act_layer(name=act)
         self.maxpool = pool_type(kernel_size=3, stride=2, padding=1)
         self.layer1 = self._make_layer(block, block_inplanes[0], layers[0], spatial_dims, shortcut_type)
         self.layer2 = self._make_layer(block, block_inplanes[1], layers[1], spatial_dims, shortcut_type, stride=2)
@@ -329,7 +335,7 @@ def _make_layer(
     def forward(self, x: torch.Tensor) -> torch.Tensor:
         x = self.conv1(x)
         x = self.bn1(x)
-        x = self.relu(x)
+        x = self.act(x)
         if not self.no_max_pool:
             x = self.maxpool(x)
 
@@ -396,7 +402,7 @@ def forward(self, inputs: torch.Tensor):
         """
         x = self.conv1(inputs)
         x = self.bn1(x)
-        x = self.relu(x)
+        x = self.act(x)
 
         features = []
         features.append(x)
diff --git a/tests/test_resnet.py b/tests/test_resnet.py
index ecab133154..3a58d1c955 100644
--- a/tests/test_resnet.py
+++ b/tests/test_resnet.py
@@ -107,6 +107,7 @@
         "num_classes": 3,
         "conv1_t_size": [3],
         "conv1_t_stride": 1,
+        "act": ("relu", {"inplace": False}),
     },
     (1, 2, 32),
     (1, 3),
@@ -185,13 +186,29 @@
     (1, 3),
 ]
 
+TEST_CASE_8 = [
+    {
+        "block": "bottleneck",
+        "layers": [3, 4, 6, 3],
+        "block_inplanes": [64, 128, 256, 512],
+        "spatial_dims": 1,
+        "n_input_channels": 2,
+        "num_classes": 3,
+        "conv1_t_size": [3],
+        "conv1_t_stride": 1,
+        "act": ("relu", {"inplace": False}),
+    },
+    (1, 2, 32),
+    (1, 3),
+]
+
 TEST_CASES = []
 PRETRAINED_TEST_CASES = []
 for case in [TEST_CASE_1, TEST_CASE_2, TEST_CASE_3, TEST_CASE_2_A, TEST_CASE_3_A]:
     for model in [resnet10, resnet18, resnet34, resnet50, resnet101, resnet152, resnet200]:
         TEST_CASES.append([model, *case])
         PRETRAINED_TEST_CASES.append([model, *case])
-for case in [TEST_CASE_5, TEST_CASE_5_A, TEST_CASE_6, TEST_CASE_7]:
+for case in [TEST_CASE_5, TEST_CASE_5_A, TEST_CASE_6, TEST_CASE_7, TEST_CASE_8]:
     TEST_CASES.append([ResNet, *case])
 
 TEST_SCRIPT_CASES = [

From 4af23069ccbbd26d7816bafc3762365ce4c5da77 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 10 May 2024 13:33:37 +0800
Subject: [PATCH 049/183] Revert version requirement for mlflow (#7742)

Revert https://github.com/Project-MONAI/MONAI/pull/7659

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/pythonapp.yml | 1 +
 docs/requirements.txt           | 2 +-
 requirements-dev.txt            | 2 +-
 setup.cfg                       | 4 ++--
 4 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/pythonapp.yml b/.github/workflows/pythonapp.yml
index b8b73907d4..d1e77bb567 100644
--- a/.github/workflows/pythonapp.yml
+++ b/.github/workflows/pythonapp.yml
@@ -100,6 +100,7 @@ jobs:
         python -m pip install --pre -U itk
     - name: Install the dependencies
       run: |
+        python -m pip install --user --upgrade pip wheel
         python -m pip install torch==1.13.1 torchvision==0.14.1
         cat "requirements-dev.txt"
         python -m pip install -r requirements-dev.txt
diff --git a/docs/requirements.txt b/docs/requirements.txt
index 5acc437391..468545dc19 100644
--- a/docs/requirements.txt
+++ b/docs/requirements.txt
@@ -22,7 +22,7 @@ sphinx-autodoc-typehints==1.11.1
 pandas
 einops
 transformers<4.22; python_version <= '3.10'  # https://github.com/Project-MONAI/MONAI/issues/5157
-mlflow>=1.28.0, <=2.11.3
+mlflow>=2.12.2
 clearml>=1.10.0rc0
 tensorboardX
 imagecodecs; platform_system == "Linux" or platform_system == "Darwin"
diff --git a/requirements-dev.txt b/requirements-dev.txt
index e8792c86f3..83bdae2a5c 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -34,7 +34,7 @@ pandas
 requests
 einops
 transformers>=4.36.0
-mlflow>=1.28.0, <=2.11.3
+mlflow>=2.12.2
 clearml>=1.10.0rc0
 matplotlib!=3.5.0
 tensorboardX
diff --git a/setup.cfg b/setup.cfg
index c8ae1630f7..711d5c5558 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -66,7 +66,7 @@ all =
     pandas
     einops
     transformers<4.22; python_version <= '3.10'
-    mlflow>=1.28.0, <=2.11.3
+    mlflow>=2.12.2
     clearml>=1.10.0rc0
     matplotlib
     tensorboardX
@@ -125,7 +125,7 @@ einops =
 transformers =
     transformers<4.22; python_version <= '3.10'
 mlflow =
-    mlflow>=1.28.0, <=2.11.3
+    mlflow>=2.12.2
 matplotlib =
     matplotlib
 clearml =

From ab4bd43ae4faa983d48b67e17bc0a2ddb597d038 Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Mon, 13 May 2024 09:35:15 +0100
Subject: [PATCH 050/183] auto updates (#7760)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 monai/data/video_dataset.py  | 2 +-
 monai/networks/nets/daf3d.py | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/monai/data/video_dataset.py b/monai/data/video_dataset.py
index 9ff23ebeff..031e85db26 100644
--- a/monai/data/video_dataset.py
+++ b/monai/data/video_dataset.py
@@ -177,7 +177,7 @@ def get_available_codecs() -> dict[str, str]:
                 for codec, ext in all_codecs.items():
                     writer = cv2.VideoWriter()
                     fname = os.path.join(tmp_dir, f"test{ext}")
-                    fourcc = cv2.VideoWriter_fourcc(*codec)  # type: ignore[attr-defined]
+                    fourcc = cv2.VideoWriter_fourcc(*codec)
                     noviderr = writer.open(fname, fourcc, 1, (10, 10))
                     if noviderr:
                         codecs[codec] = ext
diff --git a/monai/networks/nets/daf3d.py b/monai/networks/nets/daf3d.py
index c9a18c746a..31c2bf4c31 100644
--- a/monai/networks/nets/daf3d.py
+++ b/monai/networks/nets/daf3d.py
@@ -196,7 +196,7 @@ def __init__(self, in_planes, planes, spatial_dims=3, stride=1, downsample=None)
         self.conv2 = conv_type(planes, planes, kernel_size=3, padding=1, stride=stride, groups=32, bias=False)
 
         # adapt activation function
-        self.relu = nn.PReLU()  # type: ignore
+        self.relu = nn.PReLU()
 
 
 class Daf3dResNetDilatedBottleneck(Daf3dResNetBottleneck):
@@ -287,7 +287,7 @@ def __init__(
             n_input_channels, self.in_planes, kernel_size=7, stride=(1, 2, 2), padding=(3, 3, 3), bias=False
         )
         self.bn1 = norm_type(32, 64)
-        self.relu = nn.PReLU()  # type: ignore
+        self.relu = nn.PReLU()
 
         # adapt layers to our needs
         self.layer1 = self._make_layer(Daf3dResNetBottleneck, block_inplanes[0], layers[0], spatial_dims, shortcut_type)

From daf2e714bb8a1cfa7c17d9aa7fb066dd37f5429e Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 13 May 2024 23:26:14 +0800
Subject: [PATCH 051/183] Fix HTTPError when Too Many Requests for huggingface
 hub (#7765)

Fixes #7764

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 tests/utils.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/utils.py b/tests/utils.py
index ea73a3ed81..d1939e590b 100644
--- a/tests/utils.py
+++ b/tests/utils.py
@@ -156,6 +156,7 @@ def skip_if_downloading_fails():
                 "limit",  # HTTP Error 503: Egress is over the account limit
                 "authenticate",
                 "timed out",  # urlopen error [Errno 110] Connection timed out
+                "HTTPError",  # HTTPError: 429 Client Error: Too Many Requests for huggingface hub
             )
         ):
             raise unittest.SkipTest(f"error while downloading: {rt_e}") from rt_e  # incomplete download

From 1bcf97f808718eb39b9a42ec770afbf581a5d83c Mon Sep 17 00:00:00 2001
From: johnzielke <j.l.zielke@gmail.com>
Date: Wed, 15 May 2024 22:29:51 -0400
Subject: [PATCH 052/183] Add direct links to github source code to docs
 (#7738)

### Description

This adds direct links to the respective source code of classes,
methods, etc. to the docs.
In my opinion these nicer and more useful than viewing the source code
in copy in the docs.

![image](https://github.com/Project-MONAI/MONAI/assets/20091488/4a6a2650-9fd2-4cd9-a2a1-084cc3a5fb36)

I currently added it in addition to the links to the source files copied
into the docs. If desired this could also be done instead of.
I also used a github logo instead of another `[source]` link.
I am not sure whether there is any easier way to change that into the
logo than the way I have done, but this one seems to work.


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: John Zielke <j.l.zielke@gmail.com>
---
 docs/source/conf.py | 55 ++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 54 insertions(+), 1 deletion(-)

diff --git a/docs/source/conf.py b/docs/source/conf.py
index fdb10fbe03..782b585c9f 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -13,6 +13,8 @@
 import os
 import subprocess
 import sys
+import importlib
+import inspect
 
 sys.path.insert(0, os.path.abspath(".."))
 sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "..", "..")))
@@ -95,7 +97,7 @@ def generate_apidocs(*args):
     "sphinx.ext.mathjax",
     "sphinx.ext.napoleon",
     "sphinx.ext.autodoc",
-    "sphinx.ext.viewcode",
+    "sphinx.ext.linkcode",
     "sphinx.ext.autosectionlabel",
     "sphinx.ext.autosummary",
     "sphinx_autodoc_typehints",
@@ -162,3 +164,54 @@ def setup(app):
     # Hook to allow for automatic generation of API docs
     # before doc deployment begins.
     app.connect("builder-inited", generate_apidocs)
+
+
+# -- Linkcode configuration --------------------------------------------------
+DEFAULT_REF = "dev"
+if os.environ.get("GITHUB_REF_TYPE", "branch") == "tag":
+    # When building a tag, link to the tag itself
+    git_ref = os.environ.get("GITHUB_REF", DEFAULT_REF)
+else:
+    git_ref = os.environ.get("GITHUB_SHA", DEFAULT_REF)
+
+DEFAULT_REPOSITORY = "Project-MONAI/MONAI"
+repository = os.environ.get("GITHUB_REPOSITORY", DEFAULT_REPOSITORY)
+
+base_code_url = f"https://github.com/{repository}/blob/{git_ref}"
+MODULE_ROOT_FOLDER = "monai"
+
+
+# Adjusted from https://github.com/python-websockets/websockets/blob/main/docs/conf.py
+def linkcode_resolve(domain, info):
+    if domain != "py":
+        raise ValueError(
+            f"expected domain to be 'py', got {domain}."
+            "Please adjust linkcode_resolve to either handle this domain or ignore it."
+        )
+
+    mod = importlib.import_module(info["module"])
+    if "." in info["fullname"]:
+        objname, attrname = info["fullname"].split(".")
+        obj = getattr(mod, objname)
+        try:
+            # object is a method of a class
+            obj = getattr(obj, attrname)
+        except AttributeError:
+            # object is an attribute of a class
+            return None
+    else:
+        obj = getattr(mod, info["fullname"])
+
+    try:
+        file = inspect.getsourcefile(obj)
+        source, lineno = inspect.getsourcelines(obj)
+    except TypeError:
+        # e.g. object is a typing.Union
+        return None
+    file = os.path.relpath(file, os.path.abspath(".."))
+    if not file.startswith(MODULE_ROOT_FOLDER):
+        # e.g. object is a typing.NewType
+        return None
+    start, end = lineno, lineno + len(source) - 1
+    url = f"{base_code_url}/{file}#L{start}-L{end}"
+    return url

From 5e9ac1bf4e3bfa73dcdb1c604554c5f7f41075b3 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 17 May 2024 16:08:14 +0800
Subject: [PATCH 053/183] Fix matplotlib 3.9.0 has no attribute 'get_cmap`
 (#7780)

Fixes #7776

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/conf.py      | 13 +++----------
 monai/visualize/utils.py |  4 +---
 requirements-dev.txt     |  2 +-
 setup.cfg                |  4 ++--
 tests/test_matshow3d.py  |  1 +
 5 files changed, 8 insertions(+), 16 deletions(-)

diff --git a/docs/source/conf.py b/docs/source/conf.py
index 782b585c9f..543372ebc7 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -139,7 +139,7 @@ def generate_apidocs(*args):
     "github_user": "Project-MONAI",
     "github_repo": "MONAI",
     "github_version": "dev",
-    "doc_path": "docs/",
+    "doc_path": "docs/source",
     "conf_py_path": "/docs/",
     "VERSION": version,
 }
@@ -167,17 +167,10 @@ def setup(app):
 
 
 # -- Linkcode configuration --------------------------------------------------
-DEFAULT_REF = "dev"
-if os.environ.get("GITHUB_REF_TYPE", "branch") == "tag":
-    # When building a tag, link to the tag itself
-    git_ref = os.environ.get("GITHUB_REF", DEFAULT_REF)
-else:
-    git_ref = os.environ.get("GITHUB_SHA", DEFAULT_REF)
-
 DEFAULT_REPOSITORY = "Project-MONAI/MONAI"
 repository = os.environ.get("GITHUB_REPOSITORY", DEFAULT_REPOSITORY)
 
-base_code_url = f"https://github.com/{repository}/blob/{git_ref}"
+base_code_url = f"https://github.com/{repository}/blob/{version}"
 MODULE_ROOT_FOLDER = "monai"
 
 
@@ -208,7 +201,7 @@ def linkcode_resolve(domain, info):
     except TypeError:
         # e.g. object is a typing.Union
         return None
-    file = os.path.relpath(file, os.path.abspath(".."))
+    file = os.path.relpath(file, os.path.abspath(os.path.join(os.path.dirname(__file__), "..", "..")))
     if not file.startswith(MODULE_ROOT_FOLDER):
         # e.g. object is a typing.NewType
         return None
diff --git a/monai/visualize/utils.py b/monai/visualize/utils.py
index f6718fe7a5..88c9a0d66a 100644
--- a/monai/visualize/utils.py
+++ b/monai/visualize/utils.py
@@ -24,11 +24,9 @@
 from monai.utils.type_conversion import convert_data_type, convert_to_dst_type
 
 if TYPE_CHECKING:
-    from matplotlib import cm
     from matplotlib import pyplot as plt
 else:
     plt, _ = optional_import("matplotlib", name="pyplot")
-    cm, _ = optional_import("matplotlib", name="cm")
 
 __all__ = ["matshow3d", "blend_images"]
 
@@ -210,7 +208,7 @@ def blend_images(
         image = repeat(image, 3, axis=0)
 
     def get_label_rgb(cmap: str, label: NdarrayOrTensor) -> NdarrayOrTensor:
-        _cmap = cm.get_cmap(cmap)
+        _cmap = plt.colormaps.get_cmap(cmap)
         label_np, *_ = convert_data_type(label, np.ndarray)
         label_rgb_np = _cmap(label_np[0])
         label_rgb_np = np.moveaxis(label_rgb_np, -1, 0)[:3]
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 83bdae2a5c..09aeb22cac 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -36,7 +36,7 @@ einops
 transformers>=4.36.0
 mlflow>=2.12.2
 clearml>=1.10.0rc0
-matplotlib!=3.5.0
+matplotlib>=3.6.3
 tensorboardX
 types-PyYAML
 pyyaml
diff --git a/setup.cfg b/setup.cfg
index 711d5c5558..086722914e 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -68,7 +68,7 @@ all =
     transformers<4.22; python_version <= '3.10'
     mlflow>=2.12.2
     clearml>=1.10.0rc0
-    matplotlib
+    matplotlib>=3.6.3
     tensorboardX
     pyyaml
     fire
@@ -127,7 +127,7 @@ transformers =
 mlflow =
     mlflow>=2.12.2
 matplotlib =
-    matplotlib
+    matplotlib>=3.6.3
 clearml =
     clearml
 tensorboardX =
diff --git a/tests/test_matshow3d.py b/tests/test_matshow3d.py
index e54bb523e4..2eba310f4e 100644
--- a/tests/test_matshow3d.py
+++ b/tests/test_matshow3d.py
@@ -114,6 +114,7 @@ def test_3d_rgb(self):
             every_n=2,
             frame_dim=-1,
             channel_dim=0,
+            fill_value=0,
             show=False,
         )
 

From 7429e2a0633defa415f644e23e83a85c07dc3ba1 Mon Sep 17 00:00:00 2001
From: johnzielke <j.l.zielke@gmail.com>
Date: Fri, 17 May 2024 07:44:00 -0400
Subject: [PATCH 054/183] Fix doc source links for read the docs (#7779)

Fixes the current docs build.
Currently no [source] links are shown anymore.
It seems like read the docs uses a different working directory,
therefore breaking the source links. It also will not have the correct
git tag referenced. This should fix both of these issues.

If there is a way to test if this fix works without merging to dev, let
me know.

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: John Zielke <j.l.zielke@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/conf.py | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/docs/source/conf.py b/docs/source/conf.py
index 543372ebc7..a91f38081f 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -140,7 +140,7 @@ def generate_apidocs(*args):
     "github_repo": "MONAI",
     "github_version": "dev",
     "doc_path": "docs/source",
-    "conf_py_path": "/docs/",
+    "conf_py_path": "/docs/source",
     "VERSION": version,
 }
 html_scaled_image_link = False
@@ -167,11 +167,24 @@ def setup(app):
 
 
 # -- Linkcode configuration --------------------------------------------------
+DEFAULT_REF = "dev"
+read_the_docs_ref = os.environ.get("READTHEDOCS_GIT_IDENTIFIER", None)
+if read_the_docs_ref:
+    # When building on ReadTheDocs, link to the specific commit
+    # https://docs.readthedocs.io/en/stable/reference/environment-variables.html#envvar-READTHEDOCS_GIT_IDENTIFIER
+    git_ref = read_the_docs_ref
+elif os.environ.get("GITHUB_REF_TYPE", "branch") == "tag":
+    # When building a tag, link to the tag itself
+    git_ref = os.environ.get("GITHUB_REF", DEFAULT_REF)
+else:
+    git_ref = os.environ.get("GITHUB_SHA", DEFAULT_REF)
+
 DEFAULT_REPOSITORY = "Project-MONAI/MONAI"
 repository = os.environ.get("GITHUB_REPOSITORY", DEFAULT_REPOSITORY)
 
-base_code_url = f"https://github.com/{repository}/blob/{version}"
+base_code_url = f"https://github.com/{repository}/blob/{git_ref}"
 MODULE_ROOT_FOLDER = "monai"
+repo_root_path = os.path.abspath(os.path.join(os.path.dirname(__file__), "..", ".."))
 
 
 # Adjusted from https://github.com/python-websockets/websockets/blob/main/docs/conf.py
@@ -201,7 +214,7 @@ def linkcode_resolve(domain, info):
     except TypeError:
         # e.g. object is a typing.Union
         return None
-    file = os.path.relpath(file, os.path.abspath(os.path.join(os.path.dirname(__file__), "..", "..")))
+    file = os.path.relpath(file, repo_root_path)
     if not file.startswith(MODULE_ROOT_FOLDER):
         # e.g. object is a typing.NewType
         return None

From d0d129246f3759f88bc0c23488e819cbac263889 Mon Sep 17 00:00:00 2001
From: Mingxin Zheng <18563433+mingxin-zheng@users.noreply.github.com>
Date: Fri, 17 May 2024 20:48:31 +0800
Subject: [PATCH 055/183] Restrict Auto3DSeg fold input based on datalist
 (#7778)

Fixes #7777.

### Description

Lower the maximum `num_fold` allowed for user inputs in Auto3DSeg
AutoRunner

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] In-line docstrings updated.

Signed-off-by: Mingxin Zheng <mingxinz@nvidia.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/apps/auto3dseg/auto_runner.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/monai/apps/auto3dseg/auto_runner.py b/monai/apps/auto3dseg/auto_runner.py
index 05c961f999..5b6b501555 100644
--- a/monai/apps/auto3dseg/auto_runner.py
+++ b/monai/apps/auto3dseg/auto_runner.py
@@ -499,8 +499,8 @@ def set_num_fold(self, num_fold: int = 5) -> AutoRunner:
 
         if num_fold <= 0:
             raise ValueError(f"num_fold is expected to be an integer greater than zero. Now it gets {num_fold}")
-        if num_fold > self.max_fold + 1:
-            # Auto3DSeg allows no validation set, so the maximum fold number is max_fold + 1
+        if num_fold > self.max_fold:
+            # Auto3DSeg must contain validation set, so the maximum fold number is max_fold.
             raise ValueError(
                 f"num_fold is greater than the maximum fold number {self.max_fold} in {self.datalist_filename}."
             )

From 25e78a2b3d2333f7effad7ae3ca2d45e5081e82b Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 17 May 2024 22:13:12 +0800
Subject: [PATCH 056/183] 7753 update changelog for v1.3.1 (#7773)

Part of #7753


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 CHANGELOG.md | 95 +++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 94 insertions(+), 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 61be8f07c1..38336505ed 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,6 +5,98 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ## [Unreleased]
 
+## [1.3.1] - 2024-05-17
+### Added
+* Support for `by_measure` argument in `RemoveSmallObjects` (#7137)
+* Support for `pretrained` flag in `ResNet` (#7095)
+* Support for uploading and downloading bundles to and from the Hugging Face Hub (#6454)
+* Added weight parameter in DiceLoss to apply weight to voxels of each class (#7158)
+* Support for returning dice for each class in `DiceMetric` (#7163)
+* Introduced `ComponentStore` for storage purposes (#7159)
+* Added utilities used in MONAI Generative (#7134)
+* Enabled Python 3.11 support for `convert_to_torchscript` and `convert_to_onnx` (#7182)
+* Support for MLflow in `AutoRunner` (#7176)
+* `fname_regex` option in PydicomReader (#7181)
+* Allowed setting AutoRunner parameters from config (#7175)
+* `VoxelMorphUNet` and `VoxelMorph` (#7178)
+* Enabled `cache` option in `GridPatchDataset` (#7180)
+* Introduced `class_labels` option in `write_metrics_reports` for improved readability (#7249)
+* `DiffusionLoss` for image registration task (#7272)
+* Supported specifying `filename` in `Saveimage` (#7318)
+* Compile support in `SupervisedTrainer` and `SupervisedEvaluator` (#7375)
+* `mlflow_experiment_name` support in `Auto3DSeg` (#7442)
+* Arm support (#7500)
+* `BarlowTwinsLoss` for representation learning (#7530)
+* `SURELoss` and `ConjugateGradient` for diffusion models (#7308)
+* Support for `CutMix`, `CutOut`, and `MixUp` augmentation techniques (#7198)
+* `meta_file` and `logging_file` options to `BundleWorkflow` (#7549)
+* `properties_path` option to `BundleWorkflow` for customized properties (#7542)
+* Support for both soft and hard clipping in `ClipIntensityPercentiles` (#7535)
+* Support for not saving artifacts in `MLFlowHandler` (#7604)
+* Support for multi-channel images in `PerceptualLoss` (#7568)
+* Added ResNet backbone for `FlexibleUNet` (#7571)
+* Introduced `dim_head` option in `SABlock` to set dimensions for each head (#7664)
+* Direct links to github source code to docs (#7738, #7779)
+#### misc.
+* Refactored `list_data_collate` and `collate_meta_tensor` to utilize the latest PyTorch API (#7165)
+* Added __str__ method in `Metric` base class (#7487)
+* Made enhancements for testing files (#7662, #7670, #7663, #7671, #7672)
+* Improved documentation for bundles (#7116)
+### Fixed
+#### transforms
+* Addressed issue where lazy mode was ignored in `SpatialPadd` (#7316)
+* Tracked applied operations in `ImageFilter` (#7395)
+* Warnings are now given only if missing class is not set to 0 in `generate_label_classes_crop_centers` (#7602)
+* Input is now always converted to C-order in `distance_transform_edt` to ensure consistent behavior (#7675)
+#### data
+* Modified .npz file behavior to use keys in `NumpyReader` (#7148)
+* Handled corrupted cached files in `PersistentDataset` (#7244)
+* Corrected affine update in `NrrdReader` (#7415)
+#### metrics and losses
+* Addressed precision issue in `get_confusion_matrix` (#7187)
+* Harmonized and clarified documentation and tests for dice losses variants (#7587)
+#### networks
+* Removed hard-coded `spatial_dims` in `SwinTransformer` (#7302)
+* Fixed learnable `position_embeddings` in `PatchEmbeddingBlock` (#7564, #7605)
+* Removed `memory_pool_limit` in TRT config (#7647)
+* Propagated `kernel_size` to `ConvBlocks` within `AttentionUnet` (#7734)
+* Addressed hard-coded activation layer in `ResNet` (#7749)
+#### bundle
+* Resolved bundle download issue (#7280)
+* Updated `bundle_root` directory for `NNIGen` (#7586)
+* Checked for `num_fold` and failed early if incorrect (#7634)
+* Enhanced logging logic in `ConfigWorkflow` (#7745)
+#### misc.
+* Enabled chaining in `Auto3DSeg` CLI (#7168)
+* Addressed useless error message in `nnUNetV2Runner` (#7217)
+* Resolved typing and deprecation issues in Mypy (#7231)
+* Quoted `$PY_EXE` variable to handle Python path that contains spaces in Bash (#7268)
+* Improved documentation, code examples, and warning messages in various modules (#7234, #7213, #7271, #7326, #7569, #7584)
+* Fixed typos in various modules (#7321, #7322, #7458, #7595, #7612)
+* Enhanced docstrings in various modules (#7245, #7381, #7746)
+* Handled error when data is on CPU in `DataAnalyzer` (#7310)
+* Updated version requirements for third-party packages (#7343, #7344, #7384, #7448, #7659, #7704, #7744, #7742, #7780)
+* Addressed incorrect slice compute in `ImageStats` (#7374)
+* Avoided editing a loop's mutable iterable to address B308 (#7397)
+* Fixed issue with `CUDA_VISIBLE_DEVICES` setting being ignored (#7408, #7581)
+* Avoided changing Python version in CICD (#7424)
+* Renamed partial to callable in instantiate mode (#7413)
+* Imported AttributeError for Python 3.12 compatibility (#7482)
+* Updated `nnUNetV2Runner` to support nnunetv2 2.2 (#7483)
+* Used uint8 instead of int8 in `LabelStats` (#7489)
+* Utilized subprocess for nnUNet training (#7576)
+* Addressed deprecated warning in ruff (#7625)
+* Fixed downloading failure on FIPS machine (#7698)
+* Updated `torch_tensorrt` compile parameters to avoid warning (#7714)
+* Restrict `Auto3DSeg` fold input based on datalist (#7778)
+### Changed
+* Base Docker image upgraded to `nvcr.io/nvidia/pytorch:24.03-py3` from `nvcr.io/nvidia/pytorch:23.08-py3`
+### Removed
+* Removed unrecommended star-arg unpacking after a keyword argument, addressed B026 (#7262)
+* Skipped old PyTorch version test for `SwinUNETR` (#7266)
+* Dropped docker build workflow and migrated to Nvidia Blossom system (#7450)
+* Dropped Python 3.8 test on quick-py3 workflow (#7719)
+
 ## [1.3.0] - 2023-10-12
 ### Added
 * Intensity transforms `ScaleIntensityFixedMean` and `RandScaleIntensityFixedMean` (#6542)
@@ -943,7 +1035,8 @@ the postprocessing steps should be used before calling the metrics methods
 
 [highlights]: https://github.com/Project-MONAI/MONAI/blob/master/docs/source/highlights.md
 
-[Unreleased]: https://github.com/Project-MONAI/MONAI/compare/1.3.0...HEAD
+[Unreleased]: https://github.com/Project-MONAI/MONAI/compare/1.3.1...HEAD
+[1.3.1]: https://github.com/Project-MONAI/MONAI/compare/1.3.0...1.3.1
 [1.3.0]: https://github.com/Project-MONAI/MONAI/compare/1.2.0...1.3.0
 [1.2.0]: https://github.com/Project-MONAI/MONAI/compare/1.1.0...1.2.0
 [1.1.0]: https://github.com/Project-MONAI/MONAI/compare/1.0.1...1.1.0

From b16f54ab82e665f8643043ffba5c917fa417121b Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Sat, 18 May 2024 23:49:49 +0800
Subject: [PATCH 057/183] Pin transformer's version (#7782)

workaround for #7781

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/requirements.txt | 2 +-
 requirements-dev.txt  | 2 +-
 setup.cfg             | 4 ++--
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/docs/requirements.txt b/docs/requirements.txt
index 468545dc19..007281ac35 100644
--- a/docs/requirements.txt
+++ b/docs/requirements.txt
@@ -21,7 +21,7 @@ sphinxcontrib-serializinghtml
 sphinx-autodoc-typehints==1.11.1
 pandas
 einops
-transformers<4.22; python_version <= '3.10'  # https://github.com/Project-MONAI/MONAI/issues/5157
+transformers>=4.36.0, <4.41.0; python_version <= '3.10'
 mlflow>=2.12.2
 clearml>=1.10.0rc0
 tensorboardX
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 09aeb22cac..a8ba25966b 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -33,7 +33,7 @@ tifffile; platform_system == "Linux" or platform_system == "Darwin"
 pandas
 requests
 einops
-transformers>=4.36.0
+transformers>=4.36.0, <4.41.0; python_version <= '3.10'
 mlflow>=2.12.2
 clearml>=1.10.0rc0
 matplotlib>=3.6.3
diff --git a/setup.cfg b/setup.cfg
index 086722914e..7b82784a8a 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -65,7 +65,7 @@ all =
     imagecodecs
     pandas
     einops
-    transformers<4.22; python_version <= '3.10'
+    transformers>=4.36.0, <4.41.0; python_version <= '3.10'
     mlflow>=2.12.2
     clearml>=1.10.0rc0
     matplotlib>=3.6.3
@@ -123,7 +123,7 @@ pandas =
 einops =
     einops
 transformers =
-    transformers<4.22; python_version <= '3.10'
+    transformers>=4.36.0, <4.41.0; python_version <= '3.10'
 mlflow =
     mlflow>=2.12.2
 matplotlib =

From 96bfda00c6bd290297f5e3514ea227c6be4d08b4 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 21 May 2024 02:03:26 +0800
Subject: [PATCH 058/183] Skip failed tests (#7783)

### Description
- skip release_tag_docker for resource issue
- increase tolerance for tests/test_clip_intensity_percentiles.py
- skip tests/test_regularization.py for non-deterministic

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 .github/workflows/release.yml             |  3 ++-
 README.md                                 |  1 -
 tests/test_clip_intensity_percentiles.py  | 15 +++++++--------
 tests/test_clip_intensity_percentilesd.py | 14 +++++++-------
 tests/test_regularization.py              |  3 +++
 5 files changed, 19 insertions(+), 17 deletions(-)

diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
index c134724665..60b610565e 100644
--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -119,7 +119,8 @@ jobs:
           rm -rf {*,.[^.]*}
 
   release_tag_docker:
-    if: github.repository == 'Project-MONAI/MONAI'
+    # if: github.repository == 'Project-MONAI/MONAI'
+    if: ${{ false }}
     needs: versioning
     runs-on: ubuntu-latest
     steps:
diff --git a/README.md b/README.md
index 7565fea1b7..5345cdb926 100644
--- a/README.md
+++ b/README.md
@@ -12,7 +12,6 @@
 
 [![premerge](https://github.com/Project-MONAI/MONAI/actions/workflows/pythonapp.yml/badge.svg?branch=dev)](https://github.com/Project-MONAI/MONAI/actions/workflows/pythonapp.yml)
 [![postmerge](https://img.shields.io/github/checks-status/project-monai/monai/dev?label=postmerge)](https://github.com/Project-MONAI/MONAI/actions?query=branch%3Adev)
-[![docker](https://github.com/Project-MONAI/MONAI/actions/workflows/docker.yml/badge.svg?branch=dev)](https://github.com/Project-MONAI/MONAI/actions/workflows/docker.yml)
 [![Documentation Status](https://readthedocs.org/projects/monai/badge/?version=latest)](https://docs.monai.io/en/latest/)
 [![codecov](https://codecov.io/gh/Project-MONAI/MONAI/branch/dev/graph/badge.svg?token=6FTC7U1JJ4)](https://codecov.io/gh/Project-MONAI/MONAI)
 
diff --git a/tests/test_clip_intensity_percentiles.py b/tests/test_clip_intensity_percentiles.py
index af157446f6..a821558fb7 100644
--- a/tests/test_clip_intensity_percentiles.py
+++ b/tests/test_clip_intensity_percentiles.py
@@ -22,7 +22,6 @@
 
 
 class TestClipIntensityPercentiles2D(NumpyImageTestCase2D):
-
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=5)
@@ -58,7 +57,7 @@ def test_soft_clipping_two_sided(self, p):
         lower, upper = percentile(im, (5, 95))
         expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_one_sided_high(self, p):
@@ -68,7 +67,7 @@ def test_soft_clipping_one_sided_high(self, p):
         upper = percentile(im, 95)
         expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
         # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result, p(expected), type_test="tensor", rtol=5e-5, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_one_sided_low(self, p):
@@ -78,7 +77,7 @@ def test_soft_clipping_one_sided_low(self, p):
         lower = percentile(im, 5)
         expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_channel_wise(self, p):
@@ -147,8 +146,8 @@ def test_soft_clipping_two_sided(self, p):
         result = soft_clipper(im)
         lower, upper = percentile(im, (5, 95))
         expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
-        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_one_sided_high(self, p):
@@ -158,7 +157,7 @@ def test_soft_clipping_one_sided_high(self, p):
         upper = percentile(im, 95)
         expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
         # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result, p(expected), type_test="tensor", rtol=5e-5, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_one_sided_low(self, p):
@@ -168,7 +167,7 @@ def test_soft_clipping_one_sided_low(self, p):
         lower = percentile(im, 5)
         expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-6, atol=0)
+        assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_channel_wise(self, p):
diff --git a/tests/test_clip_intensity_percentilesd.py b/tests/test_clip_intensity_percentilesd.py
index ed4fc588cb..98840419a0 100644
--- a/tests/test_clip_intensity_percentilesd.py
+++ b/tests/test_clip_intensity_percentilesd.py
@@ -63,7 +63,7 @@ def test_soft_clipping_two_sided(self, p):
         lower, upper = percentile(im, (5, 95))
         expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_one_sided_high(self, p):
@@ -74,7 +74,7 @@ def test_soft_clipping_one_sided_high(self, p):
         upper = percentile(im, 95)
         expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
         # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=5e-5, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_one_sided_low(self, p):
@@ -85,7 +85,7 @@ def test_soft_clipping_one_sided_low(self, p):
         lower = percentile(im, 5)
         expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_channel_wise(self, p):
@@ -164,8 +164,8 @@ def test_soft_clipping_two_sided(self, p):
         result = soft_clipper({key: im})
         lower, upper = percentile(im, (5, 95))
         expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
-        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_one_sided_high(self, p):
@@ -176,7 +176,7 @@ def test_soft_clipping_one_sided_high(self, p):
         upper = percentile(im, 95)
         expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
         # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=5e-5, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_soft_clipping_one_sided_low(self, p):
@@ -187,7 +187,7 @@ def test_soft_clipping_one_sided_low(self, p):
         lower = percentile(im, 5)
         expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
-        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-6, atol=0)
+        assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_channel_wise(self, p):
diff --git a/tests/test_regularization.py b/tests/test_regularization.py
index 4df60b9808..32df2f7b41 100644
--- a/tests/test_regularization.py
+++ b/tests/test_regularization.py
@@ -19,6 +19,7 @@
 from monai.utils import set_determinism
 
 
+@unittest.skip("Mixup is non-deterministic. Skip it temporarily")
 class TestMixup(unittest.TestCase):
 
     def setUp(self) -> None:
@@ -59,6 +60,7 @@ def test_mixupd(self):
             MixUpd(["k1", "k2"], 6, -0.5)
 
 
+@unittest.skip("CutMix is non-deterministic. Skip it temporarily")
 class TestCutMix(unittest.TestCase):
 
     def setUp(self) -> None:
@@ -90,6 +92,7 @@ def test_cutmixd(self):
             self.assertTrue(torch.allclose(output["lbl1"], output["lbl2"]))
 
 
+@unittest.skip("CutOut is non-deterministic. Skip it temporarily")
 class TestCutOut(unittest.TestCase):
 
     def setUp(self) -> None:

From 244148d76e3ce2fe83c36226027d9f02acd43c02 Mon Sep 17 00:00:00 2001
From: ytl0623 <david89062388@gmail.com>
Date: Tue, 21 May 2024 15:16:38 +0800
Subject: [PATCH 059/183] Add function in monai.transforms.utils.py (#7712)

Fixes #6704

### Description
Combined `map_classes_to_indices` and
`generate_label_classes_crop_centers` to a single function
`map_and_generate_sampling_centers` in monai.transforms.utils.py.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: ytl0623 <david89062388@gmail.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Yu <146002968+Yu0610@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/transforms/__init__.py                  |  1 +
 monai/transforms/utils.py                     | 65 ++++++++++++++
 .../test_map_and_generate_sampling_centers.py | 87 +++++++++++++++++++
 3 files changed, 153 insertions(+)
 create mode 100644 tests/test_map_and_generate_sampling_centers.py

diff --git a/monai/transforms/__init__.py b/monai/transforms/__init__.py
index ab9adb6a99..ef1da2d855 100644
--- a/monai/transforms/__init__.py
+++ b/monai/transforms/__init__.py
@@ -671,6 +671,7 @@
     in_bounds,
     is_empty,
     is_positive,
+    map_and_generate_sampling_centers,
     map_binary_to_indices,
     map_classes_to_indices,
     map_spatial_axes,
diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index 560dbac346..d8461d927b 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -108,6 +108,7 @@
     "in_bounds",
     "is_empty",
     "is_positive",
+    "map_and_generate_sampling_centers",
     "map_binary_to_indices",
     "map_classes_to_indices",
     "map_spatial_axes",
@@ -368,6 +369,70 @@ def check_non_lazy_pending_ops(
         warnings.warn(msg)
 
 
+def map_and_generate_sampling_centers(
+    label: NdarrayOrTensor,
+    spatial_size: Sequence[int] | int,
+    num_samples: int,
+    label_spatial_shape: Sequence[int] | None = None,
+    num_classes: int | None = None,
+    image: NdarrayOrTensor | None = None,
+    image_threshold: float = 0.0,
+    max_samples_per_class: int | None = None,
+    ratios: list[float | int] | None = None,
+    rand_state: np.random.RandomState | None = None,
+    allow_smaller: bool = False,
+    warn: bool = True,
+) -> tuple[tuple]:
+    """
+    Combine "map_classes_to_indices" and "generate_label_classes_crop_centers" functions, return crop center coordinates.
+    This calls `map_classes_to_indices` to get indices from `label`, gets the shape from `label_spatial_shape`
+    is given otherwise from the labels, calls `generate_label_classes_crop_centers`, and returns its results.
+
+    Args:
+        label: use the label data to get the indices of every class.
+        spatial_size: spatial size of the ROIs to be sampled.
+        num_samples: total sample centers to be generated.
+        label_spatial_shape: spatial shape of the original label data to unravel selected centers.
+        indices: sequence of pre-computed foreground indices of every class in 1 dimension.
+        num_classes: number of classes for argmax label, not necessary for One-Hot label.
+        image: if image is not None, only return the indices of every class that are within the valid
+            region of the image (``image > image_threshold``).
+        image_threshold: if enabled `image`, use ``image > image_threshold`` to
+            determine the valid image content area and select class indices only in this area.
+        max_samples_per_class: maximum length of indices in each class to reduce memory consumption.
+            Default is None, no subsampling.
+        ratios: ratios of every class in the label to generate crop centers, including background class.
+            if None, every class will have the same ratio to generate crop centers.
+        rand_state: numpy randomState object to align with other modules.
+        allow_smaller: if `False`, an exception will be raised if the image is smaller than
+            the requested ROI in any dimension. If `True`, any smaller dimensions will be set to
+            match the cropped size (i.e., no cropping in that dimension).
+        warn: if `True` prints a warning if a class is not present in the label.
+    Returns:
+        Tuple of crop centres
+    """
+    if label is None:
+        raise ValueError("label must not be None.")
+    indices = map_classes_to_indices(label, num_classes, image, image_threshold, max_samples_per_class)
+
+    if label_spatial_shape is not None:
+        _shape = label_spatial_shape
+    elif isinstance(label, monai.data.MetaTensor):
+        _shape = label.peek_pending_shape()
+    else:
+        _shape = label.shape[1:]
+
+    if _shape is None:
+        raise ValueError(
+            "label_spatial_shape or label with a known shape must be provided to infer the output spatial shape."
+        )
+    centers = generate_label_classes_crop_centers(
+        spatial_size, num_samples, _shape, indices, ratios, rand_state, allow_smaller, warn
+    )
+
+    return ensure_tuple(centers)
+
+
 def map_binary_to_indices(
     label: NdarrayOrTensor, image: NdarrayOrTensor | None = None, image_threshold: float = 0.0
 ) -> tuple[NdarrayOrTensor, NdarrayOrTensor]:
diff --git a/tests/test_map_and_generate_sampling_centers.py b/tests/test_map_and_generate_sampling_centers.py
new file mode 100644
index 0000000000..ff74f974b9
--- /dev/null
+++ b/tests/test_map_and_generate_sampling_centers.py
@@ -0,0 +1,87 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from copy import deepcopy
+
+import numpy as np
+from parameterized import parameterized
+
+from monai.transforms import map_and_generate_sampling_centers
+from monai.utils.misc import set_determinism
+from tests.utils import TEST_NDARRAYS, assert_allclose
+
+TEST_CASE_1 = [
+    # test Argmax data
+    {
+        "label": (np.array([[[0, 1, 2], [2, 0, 1], [1, 2, 0]]])),
+        "spatial_size": [2, 2, 2],
+        "num_samples": 2,
+        "label_spatial_shape": [3, 3, 3],
+        "num_classes": 3,
+        "image": None,
+        "ratios": [0, 1, 2],
+        "image_threshold": 0.0,
+    },
+    tuple,
+    2,
+    3,
+]
+
+TEST_CASE_2 = [
+    {
+        "label": (
+            np.array(
+                [
+                    [[1, 0, 0], [0, 1, 0], [0, 0, 1]],
+                    [[0, 1, 0], [0, 0, 1], [1, 0, 0]],
+                    [[0, 0, 1], [1, 0, 0], [0, 1, 0]],
+                ]
+            )
+        ),
+        "spatial_size": [2, 2, 2],
+        "num_samples": 1,
+        "ratios": None,
+        "label_spatial_shape": [3, 3, 3],
+        "image": None,
+        "image_threshold": 0.0,
+    },
+    tuple,
+    1,
+    3,
+]
+
+
+class TestMapAndGenerateSamplingCenters(unittest.TestCase):
+
+    @parameterized.expand([TEST_CASE_1, TEST_CASE_2])
+    def test_map_and_generate_sampling_centers(self, input_data, expected_type, expected_count, expected_shape):
+        results = []
+        for p in TEST_NDARRAYS + (None,):
+            input_data = deepcopy(input_data)
+            if p is not None:
+                input_data["label"] = p(input_data["label"])
+            set_determinism(0)
+            result = map_and_generate_sampling_centers(**input_data)
+            self.assertIsInstance(result, expected_type)
+            self.assertEqual(len(result), expected_count)
+            self.assertEqual(len(result[0]), expected_shape)
+            # check for consistency between numpy, torch and torch.cuda
+            results.append(result)
+            if len(results) > 1:
+                for x, y in zip(result[0], result[-1]):
+                    assert_allclose(x, y, type_test=False)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 66a2fae406f8d83a4421edd86a04fa787dbbd091 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 21 May 2024 19:05:34 +0800
Subject: [PATCH 060/183] 7753 update releasing 1.3.1 (#7788)

Fixes #7753

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 CITATION.cff | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/CITATION.cff b/CITATION.cff
index cac47faae4..4754c5b2e3 100644
--- a/CITATION.cff
+++ b/CITATION.cff
@@ -6,8 +6,8 @@ title: "MONAI: Medical Open Network for AI"
 abstract: "AI Toolkit for Healthcare Imaging"
 authors:
   - name: "MONAI Consortium"
-date-released: 2023-10-12
-version: "1.3.0"
+date-released: 2024-05-21
+version: "1.3.1"
 identifiers:
   - description: "This DOI represents all versions of MONAI, and will always resolve to the latest one."
     type: doi

From 373c003d94009d021b66fcf08a1b650228d7079b Mon Sep 17 00:00:00 2001
From: Pkaps25 <43655728+Pkaps25@users.noreply.github.com>
Date: Thu, 23 May 2024 03:00:22 -0400
Subject: [PATCH 061/183] Add norm param to ResNet (#7752)

Fixes #7294  .

### Description
Adds a `norm` param to ResNet

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Peter Kaplinsky <peterkaplinsky@gmail.com>
Co-authored-by: Peter Kaplinsky <peterkaplinsky@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/nets/daf3d.py  | 24 ++++++++++++-------
 monai/networks/nets/resnet.py | 44 ++++++++++++++++++++++-------------
 tests/test_resnet.py          | 19 ++++++++++++++-
 3 files changed, 62 insertions(+), 25 deletions(-)

diff --git a/monai/networks/nets/daf3d.py b/monai/networks/nets/daf3d.py
index 31c2bf4c31..02e5bb022a 100644
--- a/monai/networks/nets/daf3d.py
+++ b/monai/networks/nets/daf3d.py
@@ -13,6 +13,7 @@
 
 from collections import OrderedDict
 from collections.abc import Callable, Sequence
+from functools import partial
 
 import torch
 import torch.nn as nn
@@ -25,6 +26,7 @@
 from monai.networks.blocks.convolutions import Convolution
 from monai.networks.blocks.feature_pyramid_network import ExtraFPNBlock, FeaturePyramidNetwork
 from monai.networks.layers.factories import Conv, Norm
+from monai.networks.layers.utils import get_norm_layer
 from monai.networks.nets.resnet import ResNet, ResNetBottleneck
 
 __all__ = [
@@ -170,27 +172,31 @@ class Daf3dResNetBottleneck(ResNetBottleneck):
         spatial_dims: number of spatial dimensions of the input image.
         stride: stride to use for second conv layer.
         downsample: which downsample layer to use.
+        norm: which normalization layer to use. Defaults to group.
     """
 
     expansion = 2
 
-    def __init__(self, in_planes, planes, spatial_dims=3, stride=1, downsample=None):
-        norm_type: Callable = Norm[Norm.GROUP, spatial_dims]
+    def __init__(
+        self, in_planes, planes, spatial_dims=3, stride=1, downsample=None, norm=("group", {"num_groups": 32})
+    ):
         conv_type: Callable = Conv[Conv.CONV, spatial_dims]
 
+        norm_layer = partial(get_norm_layer, name=norm, spatial_dims=spatial_dims)
+
         # in case downsample uses batch norm, change to group norm
         if isinstance(downsample, nn.Sequential):
             downsample = nn.Sequential(
                 conv_type(in_planes, planes * self.expansion, kernel_size=1, stride=stride, bias=False),
-                norm_type(num_groups=32, num_channels=planes * self.expansion),
+                norm_layer(channels=planes * self.expansion),
             )
 
         super().__init__(in_planes, planes, spatial_dims, stride, downsample)
 
         # change norm from batch to group norm
-        self.bn1 = norm_type(num_groups=32, num_channels=planes)
-        self.bn2 = norm_type(num_groups=32, num_channels=planes)
-        self.bn3 = norm_type(num_groups=32, num_channels=planes * self.expansion)
+        self.bn1 = norm_layer(channels=planes)
+        self.bn2 = norm_layer(channels=planes)
+        self.bn3 = norm_layer(channels=planes * self.expansion)
 
         # adapt second convolution to work with groups
         self.conv2 = conv_type(planes, planes, kernel_size=3, padding=1, stride=stride, groups=32, bias=False)
@@ -212,8 +218,10 @@ class Daf3dResNetDilatedBottleneck(Daf3dResNetBottleneck):
         downsample: which downsample layer to use.
     """
 
-    def __init__(self, in_planes, planes, spatial_dims=3, stride=1, downsample=None):
-        super().__init__(in_planes, planes, spatial_dims, stride, downsample)
+    def __init__(
+        self, in_planes, planes, spatial_dims=3, stride=1, downsample=None, norm=("group", {"num_groups": 32})
+    ):
+        super().__init__(in_planes, planes, spatial_dims, stride, downsample, norm)
 
         # add dilation in second convolution
         conv_type: Callable = Conv[Conv.CONV, spatial_dims]
diff --git a/monai/networks/nets/resnet.py b/monai/networks/nets/resnet.py
index 93a77d7b2a..2cd7c8102a 100644
--- a/monai/networks/nets/resnet.py
+++ b/monai/networks/nets/resnet.py
@@ -22,8 +22,8 @@
 import torch.nn as nn
 
 from monai.networks.blocks.encoder import BaseEncoder
-from monai.networks.layers.factories import Conv, Norm, Pool
-from monai.networks.layers.utils import get_act_layer, get_pool_layer
+from monai.networks.layers.factories import Conv, Pool
+from monai.networks.layers.utils import get_act_layer, get_norm_layer, get_pool_layer
 from monai.utils import ensure_tuple_rep
 from monai.utils.module import look_up_option, optional_import
 
@@ -79,6 +79,7 @@ def __init__(
         stride: int = 1,
         downsample: nn.Module | partial | None = None,
         act: str | tuple = ("relu", {"inplace": True}),
+        norm: str | tuple = "batch",
     ) -> None:
         """
         Args:
@@ -88,17 +89,18 @@ def __init__(
             stride: stride to use for first conv layer.
             downsample: which downsample layer to use.
             act: activation type and arguments. Defaults to relu.
+            norm: feature normalization type and arguments. Defaults to batch norm.
         """
         super().__init__()
 
         conv_type: Callable = Conv[Conv.CONV, spatial_dims]
-        norm_type: Callable = Norm[Norm.BATCH, spatial_dims]
+        norm_layer = get_norm_layer(name=norm, spatial_dims=spatial_dims, channels=planes)
 
         self.conv1 = conv_type(in_planes, planes, kernel_size=3, padding=1, stride=stride, bias=False)
-        self.bn1 = norm_type(planes)
+        self.bn1 = norm_layer
         self.act = get_act_layer(name=act)
         self.conv2 = conv_type(planes, planes, kernel_size=3, padding=1, bias=False)
-        self.bn2 = norm_type(planes)
+        self.bn2 = norm_layer
         self.downsample = downsample
         self.stride = stride
 
@@ -132,6 +134,7 @@ def __init__(
         stride: int = 1,
         downsample: nn.Module | partial | None = None,
         act: str | tuple = ("relu", {"inplace": True}),
+        norm: str | tuple = "batch",
     ) -> None:
         """
         Args:
@@ -141,19 +144,20 @@ def __init__(
             stride: stride to use for second conv layer.
             downsample: which downsample layer to use.
             act: activation type and arguments. Defaults to relu.
+            norm: feature normalization type and arguments. Defaults to batch norm.
         """
 
         super().__init__()
 
         conv_type: Callable = Conv[Conv.CONV, spatial_dims]
-        norm_type: Callable = Norm[Norm.BATCH, spatial_dims]
+        norm_layer = partial(get_norm_layer, name=norm, spatial_dims=spatial_dims)
 
         self.conv1 = conv_type(in_planes, planes, kernel_size=1, bias=False)
-        self.bn1 = norm_type(planes)
+        self.bn1 = norm_layer(channels=planes)
         self.conv2 = conv_type(planes, planes, kernel_size=3, stride=stride, padding=1, bias=False)
-        self.bn2 = norm_type(planes)
+        self.bn2 = norm_layer(channels=planes)
         self.conv3 = conv_type(planes, planes * self.expansion, kernel_size=1, bias=False)
-        self.bn3 = norm_type(planes * self.expansion)
+        self.bn3 = norm_layer(channels=planes * self.expansion)
         self.act = get_act_layer(name=act)
         self.downsample = downsample
         self.stride = stride
@@ -207,6 +211,7 @@ class ResNet(nn.Module):
         feed_forward: whether to add the FC layer for the output, default to `True`.
         bias_downsample: whether to use bias term in the downsampling block when `shortcut_type` is 'B', default to `True`.
         act: activation type and arguments. Defaults to relu.
+        norm: feature normalization type and arguments. Defaults to batch norm.
 
     """
 
@@ -226,6 +231,7 @@ def __init__(
         feed_forward: bool = True,
         bias_downsample: bool = True,  # for backwards compatibility (also see PR #5477)
         act: str | tuple = ("relu", {"inplace": True}),
+        norm: str | tuple = "batch",
     ) -> None:
         super().__init__()
 
@@ -238,7 +244,6 @@ def __init__(
                 raise ValueError("Unknown block '%s', use basic or bottleneck" % block)
 
         conv_type: type[nn.Conv1d | nn.Conv2d | nn.Conv3d] = Conv[Conv.CONV, spatial_dims]
-        norm_type: type[nn.BatchNorm1d | nn.BatchNorm2d | nn.BatchNorm3d] = Norm[Norm.BATCH, spatial_dims]
         pool_type: type[nn.MaxPool1d | nn.MaxPool2d | nn.MaxPool3d] = Pool[Pool.MAX, spatial_dims]
         avgp_type: type[nn.AdaptiveAvgPool1d | nn.AdaptiveAvgPool2d | nn.AdaptiveAvgPool3d] = Pool[
             Pool.ADAPTIVEAVG, spatial_dims
@@ -262,7 +267,9 @@ def __init__(
             padding=tuple(k // 2 for k in conv1_kernel_size),
             bias=False,
         )
-        self.bn1 = norm_type(self.in_planes)
+
+        norm_layer = get_norm_layer(name=norm, spatial_dims=spatial_dims, channels=self.in_planes)
+        self.bn1 = norm_layer
         self.act = get_act_layer(name=act)
         self.maxpool = pool_type(kernel_size=3, stride=2, padding=1)
         self.layer1 = self._make_layer(block, block_inplanes[0], layers[0], spatial_dims, shortcut_type)
@@ -275,7 +282,7 @@ def __init__(
         for m in self.modules():
             if isinstance(m, conv_type):
                 nn.init.kaiming_normal_(torch.as_tensor(m.weight), mode="fan_out", nonlinearity="relu")
-            elif isinstance(m, norm_type):
+            elif isinstance(m, type(norm_layer)):
                 nn.init.constant_(torch.as_tensor(m.weight), 1)
                 nn.init.constant_(torch.as_tensor(m.bias), 0)
             elif isinstance(m, nn.Linear):
@@ -295,9 +302,9 @@ def _make_layer(
         spatial_dims: int,
         shortcut_type: str,
         stride: int = 1,
+        norm: str | tuple = "batch",
     ) -> nn.Sequential:
         conv_type: Callable = Conv[Conv.CONV, spatial_dims]
-        norm_type: Callable = Norm[Norm.BATCH, spatial_dims]
 
         downsample: nn.Module | partial | None = None
         if stride != 1 or self.in_planes != planes * block.expansion:
@@ -317,18 +324,23 @@ def _make_layer(
                         stride=stride,
                         bias=self.bias_downsample,
                     ),
-                    norm_type(planes * block.expansion),
+                    get_norm_layer(name=norm, spatial_dims=spatial_dims, channels=planes * block.expansion),
                 )
 
         layers = [
             block(
-                in_planes=self.in_planes, planes=planes, spatial_dims=spatial_dims, stride=stride, downsample=downsample
+                in_planes=self.in_planes,
+                planes=planes,
+                spatial_dims=spatial_dims,
+                stride=stride,
+                downsample=downsample,
+                norm=norm,
             )
         ]
 
         self.in_planes = planes * block.expansion
         for _i in range(1, blocks):
-            layers.append(block(self.in_planes, planes, spatial_dims=spatial_dims))
+            layers.append(block(self.in_planes, planes, spatial_dims=spatial_dims, norm=norm))
 
         return nn.Sequential(*layers)
 
diff --git a/tests/test_resnet.py b/tests/test_resnet.py
index 3a58d1c955..e873f1238a 100644
--- a/tests/test_resnet.py
+++ b/tests/test_resnet.py
@@ -202,13 +202,30 @@
     (1, 3),
 ]
 
+TEST_CASE_9 = [  # Layer norm
+    {
+        "block": ResNetBlock,
+        "layers": [3, 4, 6, 3],
+        "block_inplanes": [64, 128, 256, 512],
+        "spatial_dims": 1,
+        "n_input_channels": 2,
+        "num_classes": 3,
+        "conv1_t_size": [3],
+        "conv1_t_stride": 1,
+        "act": ("relu", {"inplace": False}),
+        "norm": ("layer", {"normalized_shape": (64, 32)}),
+    },
+    (1, 2, 32),
+    (1, 3),
+]
+
 TEST_CASES = []
 PRETRAINED_TEST_CASES = []
 for case in [TEST_CASE_1, TEST_CASE_2, TEST_CASE_3, TEST_CASE_2_A, TEST_CASE_3_A]:
     for model in [resnet10, resnet18, resnet34, resnet50, resnet101, resnet152, resnet200]:
         TEST_CASES.append([model, *case])
         PRETRAINED_TEST_CASES.append([model, *case])
-for case in [TEST_CASE_5, TEST_CASE_5_A, TEST_CASE_6, TEST_CASE_7, TEST_CASE_8]:
+for case in [TEST_CASE_5, TEST_CASE_5_A, TEST_CASE_6, TEST_CASE_7, TEST_CASE_8, TEST_CASE_9]:
     TEST_CASES.append([ResNet, *case])
 
 TEST_SCRIPT_CASES = [

From e5afa4341b18149ca8d44f680f669b33ab423722 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 23 May 2024 22:35:37 +0800
Subject: [PATCH 062/183] Fix tests/test_warp.py (#7794)

Fixes #7793

### Description

The Update() method in ITK is used to trigger the execution of the
pipeline. Add it to checks if the output data is up-to-date. If it is
not (for example, if the input data or parameters have changed), it will
recompute the output.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 tests/test_warp.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/test_warp.py b/tests/test_warp.py
index 55f40764c3..0e5f2466db 100644
--- a/tests/test_warp.py
+++ b/tests/test_warp.py
@@ -217,6 +217,7 @@ def itk_warp(img, ddf):
     # warp
     warp_filter.SetDisplacementField(displacement_field)
     warp_filter.SetInput(itk_img)
+    warp_filter.Update()
     warped_img = warp_filter.GetOutput()
     warped_img = np.asarray(warped_img)
 

From ad6a4339cd5c8d9272c3381a22fce7441e3e0c01 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 27 May 2024 14:54:14 +0800
Subject: [PATCH 063/183] Fix Resnet (#7805)

Fixes #7802.


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/nets/resnet.py | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/monai/networks/nets/resnet.py b/monai/networks/nets/resnet.py
index 2cd7c8102a..6e61db07ca 100644
--- a/monai/networks/nets/resnet.py
+++ b/monai/networks/nets/resnet.py
@@ -94,13 +94,12 @@ def __init__(
         super().__init__()
 
         conv_type: Callable = Conv[Conv.CONV, spatial_dims]
-        norm_layer = get_norm_layer(name=norm, spatial_dims=spatial_dims, channels=planes)
 
         self.conv1 = conv_type(in_planes, planes, kernel_size=3, padding=1, stride=stride, bias=False)
-        self.bn1 = norm_layer
+        self.bn1 = get_norm_layer(name=norm, spatial_dims=spatial_dims, channels=planes)
         self.act = get_act_layer(name=act)
         self.conv2 = conv_type(planes, planes, kernel_size=3, padding=1, bias=False)
-        self.bn2 = norm_layer
+        self.bn2 = get_norm_layer(name=norm, spatial_dims=spatial_dims, channels=planes)
         self.downsample = downsample
         self.stride = stride
 

From 94ab632dca0fb39be4a119683183bb7027106f1a Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Mon, 27 May 2024 11:00:36 +0100
Subject: [PATCH 064/183] auto updates (#7807)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 tests/test_clip_intensity_percentiles.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/test_clip_intensity_percentiles.py b/tests/test_clip_intensity_percentiles.py
index a821558fb7..cab2a89a47 100644
--- a/tests/test_clip_intensity_percentiles.py
+++ b/tests/test_clip_intensity_percentiles.py
@@ -22,6 +22,7 @@
 
 
 class TestClipIntensityPercentiles2D(NumpyImageTestCase2D):
+
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
     def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=5)

From 762b5253efa364b1b740876eb2f9bf838ac7bed0 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 28 May 2024 22:36:50 +0800
Subject: [PATCH 065/183] Fix precision issue in TestClipIntensityPercentiles3D
  (#7808)

Fixes #7797

### Description
Ensure the same dtype when test to avoid precision issue.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 tests/test_clip_intensity_percentiles.py  | 75 +++++++++++++----------
 tests/test_clip_intensity_percentilesd.py | 55 +++++++----------
 2 files changed, 67 insertions(+), 63 deletions(-)

diff --git a/tests/test_clip_intensity_percentiles.py b/tests/test_clip_intensity_percentiles.py
index cab2a89a47..77f811db87 100644
--- a/tests/test_clip_intensity_percentiles.py
+++ b/tests/test_clip_intensity_percentiles.py
@@ -18,9 +18,32 @@
 from monai.transforms import ClipIntensityPercentiles
 from monai.transforms.utils import soft_clip
 from monai.transforms.utils_pytorch_numpy_unification import clip, percentile
+from monai.utils.type_conversion import convert_to_tensor
 from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
 
 
+def test_hard_clip_func(im, lower, upper):
+    im_t = convert_to_tensor(im)
+    if lower is None:
+        upper = percentile(im_t, upper)
+    elif upper is None:
+        lower = percentile(im_t, lower)
+    else:
+        lower, upper = percentile(im_t, (lower, upper))
+    return clip(im_t, lower, upper)
+
+
+def test_soft_clip_func(im, lower, upper):
+    im_t = convert_to_tensor(im)
+    if lower is None:
+        upper = percentile(im_t, upper)
+    elif upper is None:
+        lower = percentile(im_t, lower)
+    else:
+        lower, upper = percentile(im_t, (lower, upper))
+    return soft_clip(im_t, minv=lower, maxv=upper, sharpness_factor=1.0, dtype=torch.float32)
+
+
 class TestClipIntensityPercentiles2D(NumpyImageTestCase2D):
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -28,8 +51,7 @@ def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=5)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = percentile(im, (5, 95))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 5, 95)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -37,8 +59,7 @@ def test_hard_clipping_one_sided_high(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=None)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = percentile(im, (0, 95))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 0, 95)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -46,8 +67,7 @@ def test_hard_clipping_one_sided_low(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=None, lower=5)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = percentile(im, (5, 100))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 5, 100)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -55,9 +75,8 @@ def test_soft_clipping_two_sided(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=95, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        lower, upper = percentile(im, (5, 95))
-        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
-        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        expected = test_soft_clip_func(im, 5, 95)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -65,9 +84,8 @@ def test_soft_clipping_one_sided_high(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=95, lower=None, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        upper = percentile(im, 95)
-        expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
-        # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
+        expected = test_soft_clip_func(im, None, 95)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -75,9 +93,8 @@ def test_soft_clipping_one_sided_low(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=None, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        lower = percentile(im, 5)
-        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
-        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        expected = test_soft_clip_func(im, 5, None)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -85,7 +102,8 @@ def test_channel_wise(self, p):
         clipper = ClipIntensityPercentiles(upper=95, lower=5, channel_wise=True)
         im = p(self.imt)
         result = clipper(im)
-        for i, c in enumerate(im):
+        im_t = convert_to_tensor(self.imt)
+        for i, c in enumerate(im_t):
             lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
             assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-4, atol=0)
@@ -118,8 +136,7 @@ def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=5)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = percentile(im, (5, 95))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 5, 95)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -127,8 +144,7 @@ def test_hard_clipping_one_sided_high(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=95, lower=None)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = percentile(im, (0, 95))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 0, 95)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -136,8 +152,7 @@ def test_hard_clipping_one_sided_low(self, p):
         hard_clipper = ClipIntensityPercentiles(upper=None, lower=5)
         im = p(self.imt)
         result = hard_clipper(im)
-        lower, upper = percentile(im, (5, 100))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 5, 100)
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -145,8 +160,7 @@ def test_soft_clipping_two_sided(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=95, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        lower, upper = percentile(im, (5, 95))
-        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
+        expected = test_soft_clip_func(im, 5, 95)
         # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
@@ -155,9 +169,8 @@ def test_soft_clipping_one_sided_high(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=95, lower=None, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        upper = percentile(im, 95)
-        expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
-        # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
+        expected = test_soft_clip_func(im, None, 95)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -165,9 +178,8 @@ def test_soft_clipping_one_sided_low(self, p):
         soft_clipper = ClipIntensityPercentiles(upper=None, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper(im)
-        lower = percentile(im, 5)
-        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
-        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        expected = test_soft_clip_func(im, 5, None)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -175,7 +187,8 @@ def test_channel_wise(self, p):
         clipper = ClipIntensityPercentiles(upper=95, lower=5, channel_wise=True)
         im = p(self.imt)
         result = clipper(im)
-        for i, c in enumerate(im):
+        im_t = convert_to_tensor(self.imt)
+        for i, c in enumerate(im_t):
             lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
             assert_allclose(result[i], p(expected), type_test="tensor", rtol=1e-4, atol=0)
diff --git a/tests/test_clip_intensity_percentilesd.py b/tests/test_clip_intensity_percentilesd.py
index 98840419a0..3e06b18418 100644
--- a/tests/test_clip_intensity_percentilesd.py
+++ b/tests/test_clip_intensity_percentilesd.py
@@ -13,14 +13,15 @@
 
 import unittest
 
-import torch
 from parameterized import parameterized
 
 from monai.transforms import ClipIntensityPercentilesd
-from monai.transforms.utils import soft_clip
 from monai.transforms.utils_pytorch_numpy_unification import clip, percentile
+from monai.utils.type_conversion import convert_to_tensor
 from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, NumpyImageTestCase3D, assert_allclose
 
+from .test_clip_intensity_percentiles import test_hard_clip_func, test_soft_clip_func
+
 
 class TestClipIntensityPercentilesd2D(NumpyImageTestCase2D):
 
@@ -30,8 +31,7 @@ def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = percentile(im, (5, 95))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 5, 95)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -40,8 +40,7 @@ def test_hard_clipping_one_sided_high(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = percentile(im, (0, 95))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 0, 95)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -50,8 +49,7 @@ def test_hard_clipping_one_sided_low(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = percentile(im, (5, 100))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 5, 100)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -60,9 +58,8 @@ def test_soft_clipping_two_sided(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        lower, upper = percentile(im, (5, 95))
-        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
-        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        expected = test_soft_clip_func(im, 5, 95)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -71,9 +68,8 @@ def test_soft_clipping_one_sided_high(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        upper = percentile(im, 95)
-        expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
-        # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
+        expected = test_soft_clip_func(im, None, 95)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -82,9 +78,8 @@ def test_soft_clipping_one_sided_low(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        lower = percentile(im, 5)
-        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
-        # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
+        expected = test_soft_clip_func(im, 5, None)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -93,7 +88,8 @@ def test_channel_wise(self, p):
         clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, channel_wise=True)
         im = p(self.imt)
         result = clipper({key: im})
-        for i, c in enumerate(im):
+        im_t = convert_to_tensor(self.imt)
+        for i, c in enumerate(im_t):
             lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
             assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-3, atol=0)
@@ -132,8 +128,7 @@ def test_hard_clipping_two_sided(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = percentile(im, (5, 95))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 5, 95)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -142,8 +137,7 @@ def test_hard_clipping_one_sided_high(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = percentile(im, (0, 95))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 0, 95)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -152,8 +146,7 @@ def test_hard_clipping_one_sided_low(self, p):
         hard_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5)
         im = p(self.imt)
         result = hard_clipper({key: im})
-        lower, upper = percentile(im, (5, 100))
-        expected = clip(im, lower, upper)
+        expected = test_hard_clip_func(im, 5, 100)
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -162,8 +155,7 @@ def test_soft_clipping_two_sided(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        lower, upper = percentile(im, (5, 95))
-        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=upper, dtype=torch.float32)
+        expected = test_soft_clip_func(im, 5, 95)
         # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
@@ -173,9 +165,8 @@ def test_soft_clipping_one_sided_high(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=None, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        upper = percentile(im, 95)
-        expected = soft_clip(im, sharpness_factor=1.0, minv=None, maxv=upper, dtype=torch.float32)
-        # the rtol is set to 5e-5 because the logaddexp function used in softplus is not stable accross torch and numpy
+        expected = test_soft_clip_func(im, None, 95)
+        # the rtol is set to 1e-4 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
     @parameterized.expand([[p] for p in TEST_NDARRAYS])
@@ -184,8 +175,7 @@ def test_soft_clipping_one_sided_low(self, p):
         soft_clipper = ClipIntensityPercentilesd(keys=[key], upper=None, lower=5, sharpness_factor=1.0)
         im = p(self.imt)
         result = soft_clipper({key: im})
-        lower = percentile(im, 5)
-        expected = soft_clip(im, sharpness_factor=1.0, minv=lower, maxv=None, dtype=torch.float32)
+        expected = test_soft_clip_func(im, 5, None)
         # the rtol is set to 1e-6 because the logaddexp function used in softplus is not stable accross torch and numpy
         assert_allclose(result[key], p(expected), type_test="tensor", rtol=1e-4, atol=0)
 
@@ -195,7 +185,8 @@ def test_channel_wise(self, p):
         clipper = ClipIntensityPercentilesd(keys=[key], upper=95, lower=5, channel_wise=True)
         im = p(self.imt)
         result = clipper({key: im})
-        for i, c in enumerate(im):
+        im_t = convert_to_tensor(im)
+        for i, c in enumerate(im_t):
             lower, upper = percentile(c, (5, 95))
             expected = clip(c, lower, upper)
             assert_allclose(result[key][i], p(expected), type_test="tensor", rtol=1e-4, atol=0)

From a0935d95a1746687576bca34666407ebbe38ad71 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 28 May 2024 23:34:47 +0800
Subject: [PATCH 066/183] Fix negative strides issue in `NrrdReader` (#7809)

Fixes #7798

### Description

Add copy to avoid negative strides when save spatial shape.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/data/image_reader.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/data/image_reader.py b/monai/data/image_reader.py
index 2361bb63a7..f5e199e2a3 100644
--- a/monai/data/image_reader.py
+++ b/monai/data/image_reader.py
@@ -1331,7 +1331,7 @@ def get_data(self, img: NrrdImage | list[NrrdImage]) -> tuple[np.ndarray, dict]:
                 header[MetaKeys.SPACE] = SpaceKeys.LPS  # assuming LPS if not specified
 
             header[MetaKeys.AFFINE] = header[MetaKeys.ORIGINAL_AFFINE].copy()
-            header[MetaKeys.SPATIAL_SHAPE] = header["sizes"]
+            header[MetaKeys.SPATIAL_SHAPE] = header["sizes"].copy()
             [header.pop(k) for k in ("sizes", "space origin", "space directions")]  # rm duplicated data in header
 
             if self.channel_dim is None:  # default to "no_channel" or -1

From 0d7f77225669d1165a0d923d4ff54fd4bfa28607 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 30 May 2024 21:58:51 +0800
Subject: [PATCH 067/183] Ensure deterministic in MixUp, CutMix, CutOut (#7813)

Fixes #7697

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/transforms/regularization/array.py      | 52 ++++++++----
 monai/transforms/regularization/dictionary.py | 80 ++++++++++++------
 tests/test_regularization.py                  | 83 ++++++++++++-------
 3 files changed, 146 insertions(+), 69 deletions(-)

diff --git a/monai/transforms/regularization/array.py b/monai/transforms/regularization/array.py
index 0b495c8623..a7436bda84 100644
--- a/monai/transforms/regularization/array.py
+++ b/monai/transforms/regularization/array.py
@@ -16,6 +16,9 @@
 
 import torch
 
+from monai.data.meta_obj import get_track_meta
+from monai.utils.type_conversion import convert_to_dst_type, convert_to_tensor
+
 from ..transform import RandomizableTransform
 
 __all__ = ["MixUp", "CutMix", "CutOut", "Mixer"]
@@ -53,9 +56,11 @@ def randomize(self, data=None) -> None:
         as needed. You need to call this method everytime you apply the transform to a new
         batch.
         """
+        super().randomize(None)
         self._params = (
             torch.from_numpy(self.R.beta(self.alpha, self.alpha, self.batch_size)).type(torch.float32),
             self.R.permutation(self.batch_size),
+            [torch.from_numpy(self.R.randint(0, d, size=(1,))) for d in data.shape[2:]] if data is not None else [],
         )
 
 
@@ -69,7 +74,7 @@ class MixUp(Mixer):
     """
 
     def apply(self, data: torch.Tensor):
-        weight, perm = self._params
+        weight, perm, _ = self._params
         nsamples, *dims = data.shape
         if len(weight) != nsamples:
             raise ValueError(f"Expected batch of size: {len(weight)}, but got {nsamples}")
@@ -80,11 +85,18 @@ def apply(self, data: torch.Tensor):
         mixweight = weight[(Ellipsis,) + (None,) * len(dims)]
         return mixweight * data + (1 - mixweight) * data[perm, ...]
 
-    def __call__(self, data: torch.Tensor, labels: torch.Tensor | None = None):
-        self.randomize()
+    def __call__(self, data: torch.Tensor, labels: torch.Tensor | None = None, randomize=True):
+        data_t = convert_to_tensor(data, track_meta=get_track_meta())
+        if labels is not None:
+            labels_t = convert_to_tensor(labels, track_meta=get_track_meta())
+        if randomize:
+            self.randomize()
         if labels is None:
-            return self.apply(data)
-        return self.apply(data), self.apply(labels)
+            return convert_to_dst_type(self.apply(data_t), dst=data)[0]
+        return (
+            convert_to_dst_type(self.apply(data_t), dst=data)[0],
+            convert_to_dst_type(self.apply(labels_t), dst=labels)[0],
+        )
 
 
 class CutMix(Mixer):
@@ -113,14 +125,13 @@ class CutMix(Mixer):
     """
 
     def apply(self, data: torch.Tensor):
-        weights, perm = self._params
+        weights, perm, coords = self._params
         nsamples, _, *dims = data.shape
         if len(weights) != nsamples:
             raise ValueError(f"Expected batch of size: {len(weights)}, but got {nsamples}")
 
         mask = torch.ones_like(data)
         for s, weight in enumerate(weights):
-            coords = [torch.randint(0, d, size=(1,)) for d in dims]
             lengths = [d * sqrt(1 - weight) for d in dims]
             idx = [slice(None)] + [slice(c, min(ceil(c + ln), d)) for c, ln, d in zip(coords, lengths, dims)]
             mask[s][idx] = 0
@@ -128,7 +139,7 @@ def apply(self, data: torch.Tensor):
         return mask * data + (1 - mask) * data[perm, ...]
 
     def apply_on_labels(self, labels: torch.Tensor):
-        weights, perm = self._params
+        weights, perm, _ = self._params
         nsamples, *dims = labels.shape
         if len(weights) != nsamples:
             raise ValueError(f"Expected batch of size: {len(weights)}, but got {nsamples}")
@@ -136,10 +147,16 @@ def apply_on_labels(self, labels: torch.Tensor):
         mixweight = weights[(Ellipsis,) + (None,) * len(dims)]
         return mixweight * labels + (1 - mixweight) * labels[perm, ...]
 
-    def __call__(self, data: torch.Tensor, labels: torch.Tensor | None = None):
-        self.randomize()
-        augmented = self.apply(data)
-        return (augmented, self.apply_on_labels(labels)) if labels is not None else augmented
+    def __call__(self, data: torch.Tensor, labels: torch.Tensor | None = None, randomize=True):
+        data_t = convert_to_tensor(data, track_meta=get_track_meta())
+        if labels is not None:
+            labels_t = convert_to_tensor(labels, track_meta=get_track_meta())
+        if randomize:
+            self.randomize(data)
+        augmented = convert_to_dst_type(self.apply(data_t), dst=data)[0]
+        if labels is not None:
+            augmented_label = convert_to_dst_type(self.apply(labels_t), dst=labels)[0]
+        return (augmented, augmented_label) if labels is not None else augmented
 
 
 class CutOut(Mixer):
@@ -155,20 +172,21 @@ class CutOut(Mixer):
     """
 
     def apply(self, data: torch.Tensor):
-        weights, _ = self._params
+        weights, _, coords = self._params
         nsamples, _, *dims = data.shape
         if len(weights) != nsamples:
             raise ValueError(f"Expected batch of size: {len(weights)}, but got {nsamples}")
 
         mask = torch.ones_like(data)
         for s, weight in enumerate(weights):
-            coords = [torch.randint(0, d, size=(1,)) for d in dims]
             lengths = [d * sqrt(1 - weight) for d in dims]
             idx = [slice(None)] + [slice(c, min(ceil(c + ln), d)) for c, ln, d in zip(coords, lengths, dims)]
             mask[s][idx] = 0
 
         return mask * data
 
-    def __call__(self, data: torch.Tensor):
-        self.randomize()
-        return self.apply(data)
+    def __call__(self, data: torch.Tensor, randomize=True):
+        data_t = convert_to_tensor(data, track_meta=get_track_meta())
+        if randomize:
+            self.randomize(data)
+        return convert_to_dst_type(self.apply(data_t), dst=data)[0]
diff --git a/monai/transforms/regularization/dictionary.py b/monai/transforms/regularization/dictionary.py
index 373913da99..d8815e47b9 100644
--- a/monai/transforms/regularization/dictionary.py
+++ b/monai/transforms/regularization/dictionary.py
@@ -11,16 +11,23 @@
 
 from __future__ import annotations
 
+from collections.abc import Hashable
+
+import numpy as np
+
 from monai.config import KeysCollection
+from monai.config.type_definitions import NdarrayOrTensor
+from monai.data.meta_obj import get_track_meta
+from monai.utils import convert_to_tensor
 from monai.utils.misc import ensure_tuple
 
-from ..transform import MapTransform
+from ..transform import MapTransform, RandomizableTransform
 from .array import CutMix, CutOut, MixUp
 
 __all__ = ["MixUpd", "MixUpD", "MixUpDict", "CutMixd", "CutMixD", "CutMixDict", "CutOutd", "CutOutD", "CutOutDict"]
 
 
-class MixUpd(MapTransform):
+class MixUpd(MapTransform, RandomizableTransform):
     """
     Dictionary-based version :py:class:`monai.transforms.MixUp`.
 
@@ -31,18 +38,24 @@ class MixUpd(MapTransform):
     def __init__(
         self, keys: KeysCollection, batch_size: int, alpha: float = 1.0, allow_missing_keys: bool = False
     ) -> None:
-        super().__init__(keys, allow_missing_keys)
+        MapTransform.__init__(self, keys, allow_missing_keys)
         self.mixup = MixUp(batch_size, alpha)
 
+    def set_random_state(self, seed: int | None = None, state: np.random.RandomState | None = None) -> MixUpd:
+        super().set_random_state(seed, state)
+        self.mixup.set_random_state(seed, state)
+        return self
+
     def __call__(self, data):
-        self.mixup.randomize()
-        result = dict(data)
-        for k in self.keys:
-            result[k] = self.mixup.apply(data[k])
-        return result
+        d = dict(data)
+        # all the keys share the same random state
+        self.mixup.randomize(None)
+        for k in self.key_iterator(d):
+            d[k] = self.mixup(data[k], randomize=False)
+        return d
 
 
-class CutMixd(MapTransform):
+class CutMixd(MapTransform, RandomizableTransform):
     """
     Dictionary-based version :py:class:`monai.transforms.CutMix`.
 
@@ -63,17 +76,27 @@ def __init__(
         self.mixer = CutMix(batch_size, alpha)
         self.label_keys = ensure_tuple(label_keys) if label_keys is not None else []
 
-    def __call__(self, data):
-        self.mixer.randomize()
-        result = dict(data)
-        for k in self.keys:
-            result[k] = self.mixer.apply(data[k])
-        for k in self.label_keys:
-            result[k] = self.mixer.apply_on_labels(data[k])
-        return result
-
+    def set_random_state(self, seed: int | None = None, state: np.random.RandomState | None = None) -> CutMixd:
+        super().set_random_state(seed, state)
+        self.mixer.set_random_state(seed, state)
+        return self
 
-class CutOutd(MapTransform):
+    def __call__(self, data):
+        d = dict(data)
+        first_key: Hashable = self.first_key(d)
+        if first_key == ():
+            out: dict[Hashable, NdarrayOrTensor] = convert_to_tensor(d, track_meta=get_track_meta())
+            return out
+        self.mixer.randomize(d[first_key])
+        for key, label_key in self.key_iterator(d, self.label_keys):
+            ret = self.mixer(data[key], data.get(label_key, None), randomize=False)
+            d[key] = ret[0]
+            if label_key in d:
+                d[label_key] = ret[1]
+        return d
+
+
+class CutOutd(MapTransform, RandomizableTransform):
     """
     Dictionary-based version :py:class:`monai.transforms.CutOut`.
 
@@ -84,12 +107,21 @@ def __init__(self, keys: KeysCollection, batch_size: int, allow_missing_keys: bo
         super().__init__(keys, allow_missing_keys)
         self.cutout = CutOut(batch_size)
 
+    def set_random_state(self, seed: int | None = None, state: np.random.RandomState | None = None) -> CutOutd:
+        super().set_random_state(seed, state)
+        self.cutout.set_random_state(seed, state)
+        return self
+
     def __call__(self, data):
-        result = dict(data)
-        self.cutout.randomize()
-        for k in self.keys:
-            result[k] = self.cutout(data[k])
-        return result
+        d = dict(data)
+        first_key: Hashable = self.first_key(d)
+        if first_key == ():
+            out: dict[Hashable, NdarrayOrTensor] = convert_to_tensor(d, track_meta=get_track_meta())
+            return out
+        self.cutout.randomize(d[first_key])
+        for k in self.key_iterator(d):
+            d[k] = self.cutout(data[k], randomize=False)
+        return d
 
 
 MixUpD = MixUpDict = MixUpd
diff --git a/tests/test_regularization.py b/tests/test_regularization.py
index 32df2f7b41..12d64637d5 100644
--- a/tests/test_regularization.py
+++ b/tests/test_regularization.py
@@ -13,29 +13,31 @@
 
 import unittest
 
+import numpy as np
 import torch
 
-from monai.transforms import CutMix, CutMixd, CutOut, MixUp, MixUpd
-from monai.utils import set_determinism
+from monai.transforms import CutMix, CutMixd, CutOut, CutOutd, MixUp, MixUpd
+from tests.utils import assert_allclose
 
 
-@unittest.skip("Mixup is non-deterministic. Skip it temporarily")
 class TestMixup(unittest.TestCase):
 
-    def setUp(self) -> None:
-        set_determinism(seed=0)
-
-    def tearDown(self) -> None:
-        set_determinism(None)
-
     def test_mixup(self):
         for dims in [2, 3]:
             shape = (6, 3) + (32,) * dims
             sample = torch.rand(*shape, dtype=torch.float32)
             mixup = MixUp(6, 1.0)
+            mixup.set_random_state(seed=0)
             output = mixup(sample)
+            np.random.seed(0)
+            # simulate the randomize() of transform
+            np.random.random()
+            weight = torch.from_numpy(np.random.beta(1.0, 1.0, 6)).type(torch.float32)
+            perm = np.random.permutation(6)
             self.assertEqual(output.shape, sample.shape)
-            self.assertTrue(any(not torch.allclose(sample, mixup(sample)) for _ in range(10)))
+            mixweight = weight[(Ellipsis,) + (None,) * (dims + 1)]
+            expected = mixweight * sample + (1 - mixweight) * sample[perm, ...]
+            assert_allclose(output, expected, type_test=False, atol=1e-7)
 
         with self.assertRaises(ValueError):
             MixUp(6, -0.5)
@@ -53,27 +55,32 @@ def test_mixupd(self):
             t = torch.rand(*shape, dtype=torch.float32)
             sample = {"a": t, "b": t}
             mixup = MixUpd(["a", "b"], 6)
+            mixup.set_random_state(seed=0)
             output = mixup(sample)
-            self.assertTrue(torch.allclose(output["a"], output["b"]))
+            np.random.seed(0)
+            # simulate the randomize() of transform
+            np.random.random()
+            weight = torch.from_numpy(np.random.beta(1.0, 1.0, 6)).type(torch.float32)
+            perm = np.random.permutation(6)
+            self.assertEqual(output["a"].shape, sample["a"].shape)
+            mixweight = weight[(Ellipsis,) + (None,) * (dims + 1)]
+            expected = mixweight * sample["a"] + (1 - mixweight) * sample["a"][perm, ...]
+            assert_allclose(output["a"], expected, type_test=False, atol=1e-7)
+            assert_allclose(output["a"], output["b"], type_test=False, atol=1e-7)
+            # self.assertTrue(torch.allclose(output["a"], output["b"]))
 
         with self.assertRaises(ValueError):
             MixUpd(["k1", "k2"], 6, -0.5)
 
 
-@unittest.skip("CutMix is non-deterministic. Skip it temporarily")
 class TestCutMix(unittest.TestCase):
 
-    def setUp(self) -> None:
-        set_determinism(seed=0)
-
-    def tearDown(self) -> None:
-        set_determinism(None)
-
     def test_cutmix(self):
         for dims in [2, 3]:
             shape = (6, 3) + (32,) * dims
             sample = torch.rand(*shape, dtype=torch.float32)
             cutmix = CutMix(6, 1.0)
+            cutmix.set_random_state(seed=0)
             output = cutmix(sample)
             self.assertEqual(output.shape, sample.shape)
             self.assertTrue(any(not torch.allclose(sample, cutmix(sample)) for _ in range(10)))
@@ -85,30 +92,50 @@ def test_cutmixd(self):
             label = torch.randint(0, 1, shape)
             sample = {"a": t, "b": t, "lbl1": label, "lbl2": label}
             cutmix = CutMixd(["a", "b"], 6, label_keys=("lbl1", "lbl2"))
+            cutmix.set_random_state(seed=123)
             output = cutmix(sample)
-            # croppings are different on each application
-            self.assertTrue(not torch.allclose(output["a"], output["b"]))
             # but mixing of labels is not affected by it
             self.assertTrue(torch.allclose(output["lbl1"], output["lbl2"]))
 
 
-@unittest.skip("CutOut is non-deterministic. Skip it temporarily")
 class TestCutOut(unittest.TestCase):
 
-    def setUp(self) -> None:
-        set_determinism(seed=0)
-
-    def tearDown(self) -> None:
-        set_determinism(None)
-
     def test_cutout(self):
         for dims in [2, 3]:
             shape = (6, 3) + (32,) * dims
             sample = torch.rand(*shape, dtype=torch.float32)
             cutout = CutOut(6, 1.0)
+            cutout.set_random_state(seed=123)
             output = cutout(sample)
+            np.random.seed(123)
+            # simulate the randomize() of transform
+            np.random.random()
+            weight = torch.from_numpy(np.random.beta(1.0, 1.0, 6)).type(torch.float32)
+            perm = np.random.permutation(6)
+            coords = [torch.from_numpy(np.random.randint(0, d, size=(1,))) for d in sample.shape[2:]]
+            assert_allclose(weight, cutout._params[0])
+            assert_allclose(perm, cutout._params[1])
+            self.assertSequenceEqual(coords, cutout._params[2])
             self.assertEqual(output.shape, sample.shape)
-            self.assertTrue(any(not torch.allclose(sample, cutout(sample)) for _ in range(10)))
+
+    def test_cutoutd(self):
+        for dims in [2, 3]:
+            shape = (6, 3) + (32,) * dims
+            t = torch.rand(*shape, dtype=torch.float32)
+            sample = {"a": t, "b": t}
+            cutout = CutOutd(["a", "b"], 6, 1.0)
+            cutout.set_random_state(seed=123)
+            output = cutout(sample)
+            np.random.seed(123)
+            # simulate the randomize() of transform
+            np.random.random()
+            weight = torch.from_numpy(np.random.beta(1.0, 1.0, 6)).type(torch.float32)
+            perm = np.random.permutation(6)
+            coords = [torch.from_numpy(np.random.randint(0, d, size=(1,))) for d in t.shape[2:]]
+            assert_allclose(weight, cutout.cutout._params[0])
+            assert_allclose(perm, cutout.cutout._params[1])
+            self.assertSequenceEqual(coords, cutout.cutout._params[2])
+            self.assertEqual(output["a"].shape, sample["a"].shape)
 
 
 if __name__ == "__main__":

From 4029c422f1da84cdf5e4ce546dbff33245cbe025 Mon Sep 17 00:00:00 2001
From: Suraj Pai <b.pai@maastrichtuniversity.nl>
Date: Thu, 30 May 2024 23:26:46 -0400
Subject: [PATCH 068/183] Refactor Dataset to use Compose for transforms
 (#7784)

Fixes #7646

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Suraj Pai <b.pai@maastrichtuniversity.nl>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Ben Murray <ben.murray@gmail.com>
---
 monai/data/dataset.py      | 52 +++++++++--------------------
 tests/test_arraydataset.py |  2 +-
 tests/test_dataset.py      | 68 +++++++++++++++++++++++++++++++++++++-
 tests/test_profiling.py    |  4 ++-
 4 files changed, 86 insertions(+), 40 deletions(-)

diff --git a/monai/data/dataset.py b/monai/data/dataset.py
index 79e066303e..871b523289 100644
--- a/monai/data/dataset.py
+++ b/monai/data/dataset.py
@@ -36,15 +36,7 @@
 
 from monai.data.meta_tensor import MetaTensor
 from monai.data.utils import SUPPORTED_PICKLE_MOD, convert_tables_to_dicts, pickle_hashing
-from monai.transforms import (
-    Compose,
-    Randomizable,
-    RandomizableTrait,
-    Transform,
-    apply_transform,
-    convert_to_contiguous,
-    reset_ops_id,
-)
+from monai.transforms import Compose, Randomizable, RandomizableTrait, Transform, convert_to_contiguous, reset_ops_id
 from monai.utils import MAX_SEED, convert_to_tensor, get_seed, look_up_option, min_version, optional_import
 from monai.utils.misc import first
 
@@ -77,15 +69,19 @@ class Dataset(_TorchDataset):
          },                           },                           }]
     """
 
-    def __init__(self, data: Sequence, transform: Callable | None = None) -> None:
+    def __init__(self, data: Sequence, transform: Sequence[Callable] | Callable | None = None) -> None:
         """
         Args:
             data: input data to load and transform to generate dataset for model.
-            transform: a callable data transform on input data.
-
+            transform: a callable, sequence of callables or None. If transform is not
+            a `Compose` instance, it will be wrapped in a `Compose` instance. Sequences
+            of callables are applied in order and if `None` is passed, the data is returned as is.
         """
         self.data = data
-        self.transform: Any = transform
+        try:
+            self.transform = Compose(transform) if not isinstance(transform, Compose) else transform
+        except Exception as e:
+            raise ValueError("`transform` must be a callable or a list of callables that is Composable") from e
 
     def __len__(self) -> int:
         return len(self.data)
@@ -95,7 +91,7 @@ def _transform(self, index: int):
         Fetch single data item from `self.data`.
         """
         data_i = self.data[index]
-        return apply_transform(self.transform, data_i) if self.transform is not None else data_i
+        return self.transform(data_i)
 
     def __getitem__(self, index: int | slice | Sequence[int]):
         """
@@ -264,8 +260,6 @@ def __init__(
                 using the cached content and with re-created transform instances.
 
         """
-        if not isinstance(transform, Compose):
-            transform = Compose(transform)
         super().__init__(data=data, transform=transform)
         self.cache_dir = Path(cache_dir) if cache_dir is not None else None
         self.hash_func = hash_func
@@ -323,9 +317,6 @@ def _pre_transform(self, item_transformed):
             random transform object
 
         """
-        if not isinstance(self.transform, Compose):
-            raise ValueError("transform must be an instance of monai.transforms.Compose.")
-
         first_random = self.transform.get_index_of_first(
             lambda t: isinstance(t, RandomizableTrait) or not isinstance(t, Transform)
         )
@@ -346,9 +337,6 @@ def _post_transform(self, item_transformed):
             the transformed element through the random transforms
 
         """
-        if not isinstance(self.transform, Compose):
-            raise ValueError("transform must be an instance of monai.transforms.Compose.")
-
         first_random = self.transform.get_index_of_first(
             lambda t: isinstance(t, RandomizableTrait) or not isinstance(t, Transform)
         )
@@ -501,9 +489,6 @@ def _pre_transform(self, item_transformed):
         Returns:
             the transformed element up to the N transform object
         """
-        if not isinstance(self.transform, Compose):
-            raise ValueError("transform must be an instance of monai.transforms.Compose.")
-
         item_transformed = self.transform(item_transformed, end=self.cache_n_trans, threading=True)
 
         reset_ops_id(item_transformed)
@@ -519,9 +504,6 @@ def _post_transform(self, item_transformed):
         Returns:
             the final transformed result
         """
-        if not isinstance(self.transform, Compose):
-            raise ValueError("transform must be an instance of monai.transforms.Compose.")
-
         return self.transform(item_transformed, start=self.cache_n_trans)
 
 
@@ -809,8 +791,6 @@ def __init__(
                 Not following these recommendations may lead to runtime errors or duplicated cache across processes.
 
         """
-        if not isinstance(transform, Compose):
-            transform = Compose(transform)
         super().__init__(data=data, transform=transform)
         self.set_num = cache_num  # tracking the user-provided `cache_num` option
         self.set_rate = cache_rate  # tracking the user-provided `cache_rate` option
@@ -1282,8 +1262,10 @@ def to_list(x):
         data = []
         for dataset in self.data:
             data.extend(to_list(dataset[index]))
+
         if self.transform is not None:
-            data = apply_transform(self.transform, data, map_items=False)  # transform the list data
+            self.transform.map_items = False  # Compose object map_items to false so transform is applied to list
+            data = self.transform(data)
         # use tuple instead of list as the default collate_fn callback of MONAI DataLoader flattens nested lists
         return tuple(data)
 
@@ -1432,15 +1414,11 @@ def __len__(self):
 
     def _transform(self, index: int):
         data = {k: v[index] for k, v in self.arrays.items()}
-
-        if not self.transform:
-            return data
-
-        result = apply_transform(self.transform, data)
+        result = self.transform(data) if self.transform is not None else data
 
         if isinstance(result, dict) or (isinstance(result, list) and isinstance(result[0], dict)):
             return result
-        raise AssertionError("With a dict supplied to apply_transform, should return a dict or a list of dicts.")
+        raise AssertionError("With a dict supplied to Compose, should return a dict or a list of dicts.")
 
 
 class CSVDataset(Dataset):
diff --git a/tests/test_arraydataset.py b/tests/test_arraydataset.py
index efc014a267..b61b3c139c 100644
--- a/tests/test_arraydataset.py
+++ b/tests/test_arraydataset.py
@@ -41,7 +41,7 @@
 
 class TestCompose(Compose):
 
-    def __call__(self, input_, lazy):
+    def __call__(self, input_, lazy=False):
         img = self.transforms[0](input_)
         metadata = img.meta
         img = self.transforms[1](img)
diff --git a/tests/test_dataset.py b/tests/test_dataset.py
index 1398009c63..0d37ae2efd 100644
--- a/tests/test_dataset.py
+++ b/tests/test_dataset.py
@@ -23,7 +23,7 @@
 from parameterized import parameterized
 
 from monai.data import Dataset
-from monai.transforms import Compose, LoadImaged, SimulateDelayd
+from monai.transforms import Compose, Lambda, LoadImage, LoadImaged, SimulateDelay, SimulateDelayd
 from tests.test_compose import TEST_COMPOSE_LAZY_ON_CALL_LOGGING_TEST_CASES, data_from_keys
 
 TEST_CASE_1 = [(128, 128, 128)]
@@ -99,6 +99,72 @@ def test_dataset_lazy_on_call(self):
         data[0, 0:2, 0:2] = 1
 
 
+class TestTupleDataset(unittest.TestCase):
+
+    @parameterized.expand([TEST_CASE_1])
+    def test_shape(self, expected_shape):
+        test_image = nib.Nifti1Image(np.random.randint(0, 2, size=[128, 128, 128]).astype(float), np.eye(4))
+        with tempfile.TemporaryDirectory() as tempdir:
+            nib.save(test_image, os.path.join(tempdir, "test_image1.nii.gz"))
+            nib.save(test_image, os.path.join(tempdir, "test_label1.nii.gz"))
+            nib.save(test_image, os.path.join(tempdir, "test_image2.nii.gz"))
+            nib.save(test_image, os.path.join(tempdir, "test_label2.nii.gz"))
+            test_data = [
+                (os.path.join(tempdir, "test_image1.nii.gz"), os.path.join(tempdir, "test_label1.nii.gz")),
+                (os.path.join(tempdir, "test_image2.nii.gz"), os.path.join(tempdir, "test_label2.nii.gz")),
+            ]
+
+            test_transform = Compose([LoadImage(), SimulateDelay(delay_time=1e-5)])
+
+            # Here test_transform is applied element by element for the tuple.
+            dataset = Dataset(data=test_data, transform=test_transform)
+            data1 = dataset[0]
+            data2 = dataset[1]
+
+            # Output is a list/tuple
+            self.assertTrue(isinstance(data1, (list, tuple)))
+            self.assertTrue(isinstance(data2, (list, tuple)))
+
+            # Number of elements are 2
+            self.assertEqual(len(data1), 2)
+            self.assertEqual(len(data2), 2)
+
+            # Output shapes are as expected
+            self.assertTupleEqual(data1[0].shape, expected_shape)
+            self.assertTupleEqual(data1[1].shape, expected_shape)
+            self.assertTupleEqual(data2[0].shape, expected_shape)
+            self.assertTupleEqual(data2[1].shape, expected_shape)
+
+            # Here test_transform is applied to the tuple as a whole.
+            test_transform = Compose(
+                [
+                    # LoadImage creates a channel-stacked image when applied to a tuple
+                    LoadImage(),
+                    # Get the channel-stacked image and the label
+                    Lambda(func=lambda x: (x[0].permute(2, 1, 0), x[1])),
+                ],
+                map_items=False,
+            )
+
+            dataset = Dataset(data=test_data, transform=test_transform)
+            data1 = dataset[0]
+            data2 = dataset[1]
+
+            # Output is a list/tuple
+            self.assertTrue(isinstance(data1, (list, tuple)))
+            self.assertTrue(isinstance(data2, (list, tuple)))
+
+            # Number of elements are 2
+            self.assertEqual(len(data1), 2)
+            self.assertEqual(len(data2), 2)
+
+            # Output shapes are as expected
+            self.assertTupleEqual(data1[0].shape, expected_shape)
+            self.assertTupleEqual(data1[1].shape, expected_shape)
+            self.assertTupleEqual(data2[0].shape, expected_shape)
+            self.assertTupleEqual(data2[1].shape, expected_shape)
+
+
 class TestDatsesetWithLazy(unittest.TestCase):
     LOGGER_NAME = "a_logger_name"
 
diff --git a/tests/test_profiling.py b/tests/test_profiling.py
index 6bee7ba262..649d980ebf 100644
--- a/tests/test_profiling.py
+++ b/tests/test_profiling.py
@@ -35,6 +35,7 @@ def setUp(self):
 
         self.scale = mt.ScaleIntensity()
         self.scale_call_name = "ScaleIntensity.__call__"
+        self.compose_call_name = "Compose.__call__"
         self.test_comp = mt.Compose([mt.ScaleIntensity(), mt.RandAxisFlip(0.5)])
         self.test_image = torch.rand(1, 16, 16, 16)
         self.pid = os.getpid()
@@ -82,7 +83,7 @@ def test_profile_multithread(self):
         self.assertSequenceEqual(batch.shape, (4, 1, 16, 16, 16))
 
         results = wp.get_results()
-        self.assertSequenceEqual(list(results), [self.scale_call_name])
+        self.assertSequenceEqual(list(results), [self.scale_call_name, self.compose_call_name])
 
         prs = results[self.scale_call_name]
 
@@ -98,6 +99,7 @@ def test_profile_context(self):
                 self.scale(self.test_image)
 
         results = wp.get_results()
+
         self.assertSequenceEqual(set(results), {"ScaleIntensity.__call__", "context"})
 
         prs = results["context"]

From 7449c4d23137e6fd264b35cc71c5396e2fce03c7 Mon Sep 17 00:00:00 2001
From: YanxuanLiu <104543031+YanxuanLiu@users.noreply.github.com>
Date: Thu, 13 Jun 2024 17:08:06 +0800
Subject: [PATCH 069/183] change blossom-ci to ACL security format (#7843)

Fixes # .

### Description

Requested by security to prevent DDOS. The new format is provided by
blossom team.
### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YanxuanLiu <yanxuanl@nvidia.com>
---
 .github/workflows/blossom-ci.yml | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/.github/workflows/blossom-ci.yml b/.github/workflows/blossom-ci.yml
index 1d6ee8a46c..bf507bab3b 100644
--- a/.github/workflows/blossom-ci.yml
+++ b/.github/workflows/blossom-ci.yml
@@ -29,13 +29,15 @@ jobs:
       args: ${{ env.args }}
 
     # This job only runs for pull request comments
-    if: contains('\
-      Nic-Ma,\
-      wyli,\
-      pxLi,\
-      YanxuanLiu,\
-      KumoLiu,\
-      ', format('{0},', github.actor)) && github.event.comment.body == '/build'
+    if: |
+      github.event.comment.body == '/build' &&
+      (
+        github.actor == 'Nic-Ma' ||
+        github.actor == 'wyli' ||
+        github.actor == 'pxLi' ||
+        github.actor == 'YanxuanLiu' ||
+        github.actor == 'KumoLiu'
+      )
     steps:
       - name: Check if comment is issued by authorized person
         run: blossom-ci

From 08d572839ede5d087e49d4dfadba033e512528df Mon Sep 17 00:00:00 2001
From: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Date: Wed, 19 Jun 2024 02:40:09 +0100
Subject: [PATCH 070/183] Fixing Numpy requirements to exclude 2.0 (#7859)

### Description

This fixes the version of Numpy to be less than 2.0 so that we can get a
working setup guaranteed. Numpy 2.0 isn't supported yet by MONAI and
will need a number of changes to be compatible.

This is only sometimes a problem because a number of dependencies we
have require Numpy<2.0. When these are absent however, such as with
minimal installs, Numpy 2.0 can be installed and causes problems. This
is a temporary fix until we have worked out an update that's compatible
with versions before and after 2.0.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Eric Kerfoot <eric.kerfoot@kcl.ac.uk>
---
 requirements.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/requirements.txt b/requirements.txt
index 1569646794..1d6ae13eec 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,2 +1,2 @@
 torch>=1.9
-numpy>=1.20
+numpy>=1.20,<2.0

From e801540166655fb602df8f749089db2a4cabc0de Mon Sep 17 00:00:00 2001
From: Adam Klimont <adamkli@gmail.com>
Date: Tue, 25 Jun 2024 09:39:39 +0100
Subject: [PATCH 071/183] Change deprecated scipy.ndimage namespaces in
 optional imports (#7847)

Fixes #7677 .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.

---------

Signed-off-by: alkamid <adam.klimont@cydar.co.uk>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/apps/deepedit/transforms.py             | 2 +-
 monai/apps/deepgrow/transforms.py             | 2 +-
 monai/apps/nuclick/transforms.py              | 2 +-
 monai/apps/pathology/transforms/post/array.py | 2 +-
 monai/apps/pathology/utils.py                 | 4 ++--
 monai/metrics/utils.py                        | 6 +++---
 monai/transforms/signal/array.py              | 2 +-
 7 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/monai/apps/deepedit/transforms.py b/monai/apps/deepedit/transforms.py
index 6d0825f54a..5af082e2b0 100644
--- a/monai/apps/deepedit/transforms.py
+++ b/monai/apps/deepedit/transforms.py
@@ -30,7 +30,7 @@
 
 logger = logging.getLogger(__name__)
 
-distance_transform_cdt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_cdt")
+distance_transform_cdt, _ = optional_import("scipy.ndimage", name="distance_transform_cdt")
 
 
 class DiscardAddGuidanced(MapTransform):
diff --git a/monai/apps/deepgrow/transforms.py b/monai/apps/deepgrow/transforms.py
index 9aca77a36c..c2f97091fd 100644
--- a/monai/apps/deepgrow/transforms.py
+++ b/monai/apps/deepgrow/transforms.py
@@ -27,7 +27,7 @@
 from monai.utils.enums import PostFix
 
 measure, _ = optional_import("skimage.measure", "0.14.2", min_version)
-distance_transform_cdt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_cdt")
+distance_transform_cdt, _ = optional_import("scipy.ndimage", name="distance_transform_cdt")
 
 DEFAULT_POST_FIX = PostFix.meta()
 
diff --git a/monai/apps/nuclick/transforms.py b/monai/apps/nuclick/transforms.py
index f22ea764be..4828bd2e5a 100644
--- a/monai/apps/nuclick/transforms.py
+++ b/monai/apps/nuclick/transforms.py
@@ -24,7 +24,7 @@
 
 measure, _ = optional_import("skimage.measure")
 morphology, _ = optional_import("skimage.morphology")
-distance_transform_cdt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_cdt")
+distance_transform_cdt, _ = optional_import("scipy.ndimage", name="distance_transform_cdt")
 
 
 class NuclickKeys(StrEnum):
diff --git a/monai/apps/pathology/transforms/post/array.py b/monai/apps/pathology/transforms/post/array.py
index 99e94f89c0..42ca385fa0 100644
--- a/monai/apps/pathology/transforms/post/array.py
+++ b/monai/apps/pathology/transforms/post/array.py
@@ -33,7 +33,7 @@
 from monai.utils.misc import ensure_tuple_rep
 from monai.utils.type_conversion import convert_to_dst_type, convert_to_tensor
 
-label, _ = optional_import("scipy.ndimage.measurements", name="label")
+label, _ = optional_import("scipy.ndimage", name="label")
 disk, _ = optional_import("skimage.morphology", name="disk")
 opening, _ = optional_import("skimage.morphology", name="opening")
 watershed, _ = optional_import("skimage.segmentation", name="watershed")
diff --git a/monai/apps/pathology/utils.py b/monai/apps/pathology/utils.py
index d3ebe0a7a6..3aa0bfab86 100644
--- a/monai/apps/pathology/utils.py
+++ b/monai/apps/pathology/utils.py
@@ -33,10 +33,10 @@ def compute_multi_instance_mask(mask: np.ndarray, threshold: float) -> Any:
     """
 
     neg = 255 - mask * 255
-    distance = ndimage.morphology.distance_transform_edt(neg)
+    distance = ndimage.distance_transform_edt(neg)
     binary = distance < threshold
 
-    filled_image = ndimage.morphology.binary_fill_holes(binary)
+    filled_image = ndimage.binary_fill_holes(binary)
     multi_instance_mask = measure.label(filled_image, connectivity=2)
 
     return multi_instance_mask
diff --git a/monai/metrics/utils.py b/monai/metrics/utils.py
index e7057256fb..340e54a1d7 100644
--- a/monai/metrics/utils.py
+++ b/monai/metrics/utils.py
@@ -35,9 +35,9 @@
     optional_import,
 )
 
-binary_erosion, _ = optional_import("scipy.ndimage.morphology", name="binary_erosion")
-distance_transform_edt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_edt")
-distance_transform_cdt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_cdt")
+binary_erosion, _ = optional_import("scipy.ndimage", name="binary_erosion")
+distance_transform_edt, _ = optional_import("scipy.ndimage", name="distance_transform_edt")
+distance_transform_cdt, _ = optional_import("scipy.ndimage", name="distance_transform_cdt")
 
 __all__ = [
     "ignore_background",
diff --git a/monai/transforms/signal/array.py b/monai/transforms/signal/array.py
index 938f42192c..97df04f233 100644
--- a/monai/transforms/signal/array.py
+++ b/monai/transforms/signal/array.py
@@ -28,7 +28,7 @@
 from monai.utils.enums import TransformBackends
 from monai.utils.type_conversion import convert_data_type, convert_to_tensor
 
-shift, has_shift = optional_import("scipy.ndimage.interpolation", name="shift")
+shift, has_shift = optional_import("scipy.ndimage", name="shift")
 iirnotch, has_iirnotch = optional_import("scipy.signal", name="iirnotch")
 with warnings.catch_warnings():
     warnings.simplefilter("ignore", UserWarning)  # project-monai/monai#5204

From e92949df0799e8f96981947facb9448a18a0c17a Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 25 Jun 2024 23:02:33 +0800
Subject: [PATCH 072/183] Cherry pick #7859 for 1.3.2 (#7873)

### Description


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Eric Kerfoot <eric.kerfoot@kcl.ac.uk>
Signed-off-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <eric.kerfoot@kcl.ac.uk>
---
 requirements.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/requirements.txt b/requirements.txt
index 1569646794..1d6ae13eec 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,2 +1,2 @@
 torch>=1.9
-numpy>=1.20
+numpy>=1.20,<2.0

From 59a7211070538586369afd4a01eca0a7fe2e742e Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 26 Jun 2024 00:51:31 +0800
Subject: [PATCH 073/183] Update changelog for hotfix 1.3.2 (#7877)

Update changelog for hotfix 1.3.2

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 CHANGELOG.md | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 38336505ed..804508c262 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,6 +5,11 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ## [Unreleased]
 
+## [1.3.2] - 2024-06-25
+### Fixed
+#### misc.
+* Updated Numpy version constraint to < 2.0 (#7859)
+
 ## [1.3.1] - 2024-05-17
 ### Added
 * Support for `by_measure` argument in `RemoveSmallObjects` (#7137)
@@ -1035,7 +1040,8 @@ the postprocessing steps should be used before calling the metrics methods
 
 [highlights]: https://github.com/Project-MONAI/MONAI/blob/master/docs/source/highlights.md
 
-[Unreleased]: https://github.com/Project-MONAI/MONAI/compare/1.3.1...HEAD
+[Unreleased]: https://github.com/Project-MONAI/MONAI/compare/1.3.2...HEAD
+[1.3.2]: https://github.com/Project-MONAI/MONAI/compare/1.3.1...1.3.2
 [1.3.1]: https://github.com/Project-MONAI/MONAI/compare/1.3.0...1.3.1
 [1.3.0]: https://github.com/Project-MONAI/MONAI/compare/1.2.0...1.3.0
 [1.2.0]: https://github.com/Project-MONAI/MONAI/compare/1.1.0...1.2.0

From bef5eb8f311835830d0ddcbe7cb006cb8246a4e5 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 26 Jun 2024 17:51:16 +0800
Subject: [PATCH 074/183] 7870 update releasing 1.3.2 (#7879)

Fixes #7870


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 CHANGELOG.md | 8 +++++++-
 CITATION.cff | 4 ++--
 2 files changed, 9 insertions(+), 3 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 38336505ed..804508c262 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,6 +5,11 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ## [Unreleased]
 
+## [1.3.2] - 2024-06-25
+### Fixed
+#### misc.
+* Updated Numpy version constraint to < 2.0 (#7859)
+
 ## [1.3.1] - 2024-05-17
 ### Added
 * Support for `by_measure` argument in `RemoveSmallObjects` (#7137)
@@ -1035,7 +1040,8 @@ the postprocessing steps should be used before calling the metrics methods
 
 [highlights]: https://github.com/Project-MONAI/MONAI/blob/master/docs/source/highlights.md
 
-[Unreleased]: https://github.com/Project-MONAI/MONAI/compare/1.3.1...HEAD
+[Unreleased]: https://github.com/Project-MONAI/MONAI/compare/1.3.2...HEAD
+[1.3.2]: https://github.com/Project-MONAI/MONAI/compare/1.3.1...1.3.2
 [1.3.1]: https://github.com/Project-MONAI/MONAI/compare/1.3.0...1.3.1
 [1.3.0]: https://github.com/Project-MONAI/MONAI/compare/1.2.0...1.3.0
 [1.2.0]: https://github.com/Project-MONAI/MONAI/compare/1.1.0...1.2.0
diff --git a/CITATION.cff b/CITATION.cff
index 4754c5b2e3..b535a77a9f 100644
--- a/CITATION.cff
+++ b/CITATION.cff
@@ -6,8 +6,8 @@ title: "MONAI: Medical Open Network for AI"
 abstract: "AI Toolkit for Healthcare Imaging"
 authors:
   - name: "MONAI Consortium"
-date-released: 2024-05-21
-version: "1.3.1"
+date-released: 2024-06-26
+version: "1.3.2"
 identifiers:
   - description: "This DOI represents all versions of MONAI, and will always resolve to the latest one."
     type: doi

From 586c65961f8154619ebec3a54dd69060e2e4395d Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 28 Jun 2024 00:21:51 +0800
Subject: [PATCH 075/183] Fix 'load_module()' deprecated in Python 3.12 (#7881)

Fixes #7880

### Description

use `exec_module` instead.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/utils/module.py | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/monai/utils/module.py b/monai/utils/module.py
index 6f301d8067..4d28f8d986 100644
--- a/monai/utils/module.py
+++ b/monai/utils/module.py
@@ -13,6 +13,7 @@
 
 import enum
 import functools
+import importlib.util
 import os
 import pdb
 import re
@@ -208,9 +209,11 @@ def load_submodules(
     ):
         if (is_pkg or load_all) and name not in sys.modules and match(exclude_pattern, name) is None:
             try:
-                mod = import_module(name)
-                importer.find_spec(name).loader.load_module(name)  # type: ignore
-                submodules.append(mod)
+                mod_spec = importer.find_spec(name)  # type: ignore
+                if mod_spec and mod_spec.loader:
+                    mod = importlib.util.module_from_spec(mod_spec)
+                    mod_spec.loader.exec_module(mod)
+                    submodules.append(mod)
             except OptionalImportError:
                 pass  # could not import the optional deps., they are ignored
             except ImportError as e:

From ac84a4eaf13456ae7adccfc69eb161ac8c417fa7 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 28 Jun 2024 13:37:38 +0800
Subject: [PATCH 076/183] Fix Ruff type check issue (#7885)

Fixes #7884 #7832

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/transforms.rst  | 6 ++++++
 pyproject.toml              | 1 -
 tests/test_pad_collation.py | 6 +++---
 3 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/docs/source/transforms.rst b/docs/source/transforms.rst
index 8bd5bfd99f..a359821679 100644
--- a/docs/source/transforms.rst
+++ b/docs/source/transforms.rst
@@ -667,6 +667,12 @@ Post-processing
     :members:
     :special-members: __call__
 
+`Invert`
+"""""""""
+.. autoclass:: Invert
+    :members:
+    :special-members: __call__
+
 Regularization
 ^^^^^^^^^^^^^^
 
diff --git a/pyproject.toml b/pyproject.toml
index 50d0b09672..53ca608d20 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -38,7 +38,6 @@ exclude = "monai/bundle/__main__.py"
 
 [tool.ruff]
 line-length = 133
-lint.ignore-init-module-imports = true
 lint.ignore = ["F401", "E741"]
 
 [tool.pytype]
diff --git a/tests/test_pad_collation.py b/tests/test_pad_collation.py
index 17f49611df..9d5012c9a3 100644
--- a/tests/test_pad_collation.py
+++ b/tests/test_pad_collation.py
@@ -89,7 +89,7 @@ def tearDown(self) -> None:
 
     @parameterized.expand(TESTS)
     def test_pad_collation(self, t_type, collate_method, transform):
-        if t_type == dict:
+        if t_type is dict:
             dataset = CacheDataset(self.dict_data, transform, progress=False)
         else:
             dataset = _Dataset(self.list_data, self.list_labels, transform)
@@ -104,7 +104,7 @@ def test_pad_collation(self, t_type, collate_method, transform):
         loader = DataLoader(dataset, batch_size=10, collate_fn=collate_method)
         # check collation in forward direction
         for data in loader:
-            if t_type == dict:
+            if t_type is dict:
                 shapes = []
                 decollated_data = decollate_batch(data)
                 for d in decollated_data:
@@ -113,7 +113,7 @@ def test_pad_collation(self, t_type, collate_method, transform):
                     self.assertTrue(len(output["image"].applied_operations), len(dataset.transform.transforms))
                 self.assertTrue(len(set(shapes)) > 1)  # inverted shapes must be different because of random xforms
 
-        if t_type == dict:
+        if t_type is dict:
             batch_inverse = BatchInverseTransform(dataset.transform, loader)
             for data in loader:
                 output = batch_inverse(data)

From 06cbd70d6cce86b36a5137774901e79ba983706a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Vahit=20Bu=C4=9Fra=20YE=C5=9E=C4=B0LKAYNAK?=
 <bugrayesilkaynak@gmail.com>
Date: Fri, 28 Jun 2024 08:20:28 +0200
Subject: [PATCH 077/183] fix implementation mistakes and add conjugate
 gradients solver (#7876)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes #6767
### Description

Fixed two minor implementation differences between the official MATLAB
code and the MONAI implementation, to confirm also added a new test case
where two images and their confidence maps calculated by the official
code is added to tests and the MONAI implementation results are checked
against there results created by the official code.

Also fixing the issue: added the conjugate gradients solver option. Now
the users can utilize it to run the algorithm faster with a trade-off of
accuracy of the end result, a range of speed-ups can be achieved with
little to no quality loss by tweaking the parameters, the optimal
parameters between quality and speed in my experience is set as the
default parameters, namely 'cg_tol' and 'cg_maxiter'.

For the CG solver installing PyAMG (https://github.com/pyamg/pyamg) is a
requirement, this is because we use it to generate a preconditioner,
without it CG does not provide any speed-ups, even slows down the
algorithm. This part can be changed if the requirement is not ideal, yet
this was the best solution as far as my knowledge goes.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: MrGranddy <bugrayesilkaynak@gmail.com>
Signed-off-by: Vahit Buğra YEŞİLKAYNAK <bugrayesilkaynak@gmail.com>
Co-authored-by: ge85evz <vanessa_share@nevarro.ifl.campar.in.tum.de>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/requirements.txt                         |   1 +
 docs/source/installation.md                   |   2 +-
 monai/data/ultrasound_confidence_map.py       |  45 +-
 monai/transforms/intensity/array.py           |  27 +-
 requirements-dev.txt                          |   1 +
 setup.cfg                                     |   3 +
 ...est_ultrasound_confidence_map_transform.py | 907 +++++++++---------
 .../ultrasound_confidence_map/femur_input.png | Bin 0 -> 125674 bytes
 .../femur_result.npy                          | Bin 0 -> 974976 bytes
 .../ultrasound_confidence_map/neck_input.png  | Bin 0 -> 100996 bytes
 .../ultrasound_confidence_map/neck_result.npy | Bin 0 -> 1356512 bytes
 11 files changed, 531 insertions(+), 455 deletions(-)
 create mode 100644 tests/testing_data/ultrasound_confidence_map/femur_input.png
 create mode 100644 tests/testing_data/ultrasound_confidence_map/femur_result.npy
 create mode 100644 tests/testing_data/ultrasound_confidence_map/neck_input.png
 create mode 100644 tests/testing_data/ultrasound_confidence_map/neck_result.npy

diff --git a/docs/requirements.txt b/docs/requirements.txt
index 007281ac35..6caddce666 100644
--- a/docs/requirements.txt
+++ b/docs/requirements.txt
@@ -40,3 +40,4 @@ onnx>=1.13.0
 onnxruntime; python_version <= '3.10'
 zarr
 huggingface_hub
+pyamg>=5.0.0
diff --git a/docs/source/installation.md b/docs/source/installation.md
index d77253f0f9..644dd623c1 100644
--- a/docs/source/installation.md
+++ b/docs/source/installation.md
@@ -258,6 +258,6 @@ Since MONAI v0.2.0, the extras syntax such as `pip install 'monai[nibabel]'` is
 ```
 
 which correspond to `nibabel`, `scikit-image`,`scipy`, `pillow`, `tensorboard`,
-`gdown`, `pytorch-ignite`, `torchvision`, `itk`, `tqdm`, `lmdb`, `psutil`, `cucim`, `openslide-python`, `pandas`, `einops`, `transformers`, `mlflow`, `clearml`, `matplotlib`, `tensorboardX`, `tifffile`, `imagecodecs`, `pyyaml`, `fire`, `jsonschema`, `ninja`, `pynrrd`, `pydicom`, `h5py`, `nni`, `optuna`, `onnx`, `onnxruntime`, `zarr`, `lpips`, `nvidia-ml-py`, and `huggingface_hub` respectively.
+`gdown`, `pytorch-ignite`, `torchvision`, `itk`, `tqdm`, `lmdb`, `psutil`, `cucim`, `openslide-python`, `pandas`, `einops`, `transformers`, `mlflow`, `clearml`, `matplotlib`, `tensorboardX`, `tifffile`, `imagecodecs`, `pyyaml`, `fire`, `jsonschema`, `ninja`, `pynrrd`, `pydicom`, `h5py`, `nni`, `optuna`, `onnx`, `onnxruntime`, `zarr`, `lpips`, `nvidia-ml-py`, `huggingface_hub` and `pyamg` respectively.
 
 - `pip install 'monai[all]'` installs all the optional dependencies.
diff --git a/monai/data/ultrasound_confidence_map.py b/monai/data/ultrasound_confidence_map.py
index 03813e7559..5c716b62be 100644
--- a/monai/data/ultrasound_confidence_map.py
+++ b/monai/data/ultrasound_confidence_map.py
@@ -21,7 +21,9 @@
 cv2, _ = optional_import("cv2")
 csc_matrix, _ = optional_import("scipy.sparse", "1.7.1", min_version, "csc_matrix")
 spsolve, _ = optional_import("scipy.sparse.linalg", "1.7.1", min_version, "spsolve")
+cg, _ = optional_import("scipy.sparse.linalg", "1.7.1", min_version, "cg")
 hilbert, _ = optional_import("scipy.signal", "1.7.1", min_version, "hilbert")
+ruge_stuben_solver, _ = optional_import("pyamg", "5.0.0", min_version, "ruge_stuben_solver")
 
 
 class UltrasoundConfidenceMap:
@@ -30,6 +32,9 @@ class UltrasoundConfidenceMap:
     It generates a confidence map by setting source and sink points in the image and computing the probability
     for random walks to reach the source for each pixel.
 
+    The official code is available at:
+    https://campar.in.tum.de/Main/AthanasiosKaramalisCode
+
     Args:
         alpha (float, optional): Alpha parameter. Defaults to 2.0.
         beta (float, optional): Beta parameter. Defaults to 90.0.
@@ -37,15 +42,33 @@ class UltrasoundConfidenceMap:
         mode (str, optional): 'RF' or 'B' mode data. Defaults to 'B'.
         sink_mode (str, optional): Sink mode. Defaults to 'all'. If 'mask' is selected, a mask must be when calling
             the transform. Can be 'all', 'mid', 'min', or 'mask'.
+        use_cg (bool, optional): Use Conjugate Gradient method for solving the linear system. Defaults to False.
+        cg_tol (float, optional): Tolerance for the Conjugate Gradient method. Defaults to 1e-6.
+            Will be used only if `use_cg` is True.
+        cg_maxiter (int, optional): Maximum number of iterations for the Conjugate Gradient method. Defaults to 200.
+            Will be used only if `use_cg` is True.
     """
 
-    def __init__(self, alpha: float = 2.0, beta: float = 90.0, gamma: float = 0.05, mode="B", sink_mode="all"):
+    def __init__(
+        self,
+        alpha: float = 2.0,
+        beta: float = 90.0,
+        gamma: float = 0.05,
+        mode="B",
+        sink_mode="all",
+        use_cg=False,
+        cg_tol=1e-6,
+        cg_maxiter=200,
+    ):
         # The hyperparameters for confidence map estimation
         self.alpha = alpha
         self.beta = beta
         self.gamma = gamma
         self.mode = mode
         self.sink_mode = sink_mode
+        self.use_cg = use_cg
+        self.cg_tol = cg_tol
+        self.cg_maxiter = cg_maxiter
 
         # The precision to use for all computations
         self.eps = np.finfo("float64").eps
@@ -228,17 +251,18 @@ def confidence_laplacian(self, padded_index: NDArray, padded_image: NDArray, bet
         s = self.normalize(s)
 
         # Horizontal penalty
-        s[:vertical_end] += gamma
-        # s[vertical_end:diagonal_end] += gamma * np.sqrt(2) # --> In the paper it is sqrt(2)
-        # since the diagonal edges are longer yet does not exist in the original code
+        s[vertical_end:] += gamma
+        # Here there is a difference between the official MATLAB code and the paper
+        # on the edge penalty. We directly implement what the official code does.
 
         # Normalize differences
         s = self.normalize(s)
 
         # Gaussian weighting function
         s = -(
-            (np.exp(-beta * s, dtype="float64")) + 1.0e-6
-        )  # --> This epsilon changes results drastically default: 1.e-6
+            (np.exp(-beta * s, dtype="float64")) + 1e-5
+        )  # --> This epsilon changes results drastically default: 10e-6
+        # Please notice that it is not 1e-6, it is 10e-6 which is actually different.
 
         # Create Laplacian, diagonal missing
         lap = csc_matrix((s, (i, j)))
@@ -256,7 +280,14 @@ def confidence_laplacian(self, padded_index: NDArray, padded_image: NDArray, bet
         return lap
 
     def _solve_linear_system(self, lap, rhs):
-        x = spsolve(lap, rhs)
+
+        if self.use_cg:
+            lap_sparse = lap.tocsr()
+            ml = ruge_stuben_solver(lap_sparse, coarse_solver="pinv")
+            m = ml.aspreconditioner(cycle="V")
+            x, _ = cg(lap, rhs, tol=self.cg_tol, maxiter=self.cg_maxiter, M=m)
+        else:
+            x = spsolve(lap, rhs)
 
         return x
 
diff --git a/monai/transforms/intensity/array.py b/monai/transforms/intensity/array.py
index f656475a36..3b813809e4 100644
--- a/monai/transforms/intensity/array.py
+++ b/monai/transforms/intensity/array.py
@@ -2789,6 +2789,9 @@ class UltrasoundConfidenceMapTransform(Transform):
     It generates a confidence map by setting source and sink points in the image and computing the probability
     for random walks to reach the source for each pixel.
 
+    The official code is available at:
+    https://campar.in.tum.de/Main/AthanasiosKaramalisCode
+
     Args:
         alpha (float, optional): Alpha parameter. Defaults to 2.0.
         beta (float, optional): Beta parameter. Defaults to 90.0.
@@ -2796,14 +2799,32 @@ class UltrasoundConfidenceMapTransform(Transform):
         mode (str, optional): 'RF' or 'B' mode data. Defaults to 'B'.
         sink_mode (str, optional): Sink mode. Defaults to 'all'. If 'mask' is selected, a mask must be when
             calling the transform. Can be one of 'all', 'mid', 'min', 'mask'.
+        use_cg (bool, optional): Use Conjugate Gradient method for solving the linear system. Defaults to False.
+        cg_tol (float, optional): Tolerance for the Conjugate Gradient method. Defaults to 1e-6.
+            Will be used only if `use_cg` is True.
+        cg_maxiter (int, optional): Maximum number of iterations for the Conjugate Gradient method. Defaults to 200.
+            Will be used only if `use_cg` is True.
     """
 
-    def __init__(self, alpha: float = 2.0, beta: float = 90.0, gamma: float = 0.05, mode="B", sink_mode="all") -> None:
+    def __init__(
+        self,
+        alpha: float = 2.0,
+        beta: float = 90.0,
+        gamma: float = 0.05,
+        mode="B",
+        sink_mode="all",
+        use_cg=False,
+        cg_tol: float = 1.0e-6,
+        cg_maxiter: int = 200,
+    ):
         self.alpha = alpha
         self.beta = beta
         self.gamma = gamma
         self.mode = mode
         self.sink_mode = sink_mode
+        self.use_cg = use_cg
+        self.cg_tol = cg_tol
+        self.cg_maxiter = cg_maxiter
 
         if self.mode not in ["B", "RF"]:
             raise ValueError(f"Unknown mode: {self.mode}. Supported modes are 'B' and 'RF'.")
@@ -2813,7 +2834,9 @@ def __init__(self, alpha: float = 2.0, beta: float = 90.0, gamma: float = 0.05,
                 f"Unknown sink mode: {self.sink_mode}. Supported modes are 'all', 'mid', 'min' and 'mask'."
             )
 
-        self._compute_conf_map = UltrasoundConfidenceMap(self.alpha, self.beta, self.gamma, self.mode, self.sink_mode)
+        self._compute_conf_map = UltrasoundConfidenceMap(
+            self.alpha, self.beta, self.gamma, self.mode, self.sink_mode, self.use_cg, self.cg_tol, self.cg_maxiter
+        )
 
     def __call__(self, img: NdarrayOrTensor, mask: NdarrayOrTensor | None = None) -> NdarrayOrTensor:
         """Compute confidence map from an ultrasound image.
diff --git a/requirements-dev.txt b/requirements-dev.txt
index a8ba25966b..517c842d1e 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -57,3 +57,4 @@ zarr
 lpips==0.1.4
 nvidia-ml-py
 huggingface_hub
+pyamg>=5.0.0
diff --git a/setup.cfg b/setup.cfg
index 7b82784a8a..05bf181c70 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -84,6 +84,7 @@ all =
     lpips==0.1.4
     nvidia-ml-py
     huggingface_hub
+    pyamg>=5.0.0
 nibabel =
     nibabel
 ninja =
@@ -162,6 +163,8 @@ pynvml =
 #     MetricsReloaded @ git+https://github.com/Project-MONAI/MetricsReloaded@monai-support#egg=MetricsReloaded
 huggingface_hub =
     huggingface_hub
+pyamg =
+    pyamg>=5.0.0
 
 [flake8]
 select = B,C,E,F,N,P,T4,W,B9
diff --git a/tests/test_ultrasound_confidence_map_transform.py b/tests/test_ultrasound_confidence_map_transform.py
index 63ce7d58e4..87c08b3ac3 100644
--- a/tests/test_ultrasound_confidence_map_transform.py
+++ b/tests/test_ultrasound_confidence_map_transform.py
@@ -11,11 +11,13 @@
 
 from __future__ import annotations
 
+import os
 import unittest
 
 import numpy as np
 import torch
 from parameterized import parameterized
+from PIL import Image
 
 from monai.transforms import UltrasoundConfidenceMapTransform
 from tests.utils import assert_allclose
@@ -32,7 +34,8 @@
         [1, 2, 3, 32, 33, 34, 35, 1, 2, 3],
         [1, 2, 3, 36, 37, 38, 39, 1, 2, 3],
         [1, 2, 3, 40, 41, 42, 43, 1, 2, 3],
-    ]
+    ],
+    dtype=np.float32,
 )
 
 TEST_MASK = np.array(
@@ -47,474 +50,435 @@
         [1, 1, 1, 0, 0, 0, 1, 1, 1, 0],
         [1, 1, 1, 0, 0, 0, 1, 1, 1, 0],
         [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
-    ]
+    ],
+    dtype=np.float32,
 )
 
 SINK_ALL_OUTPUT = np.array(
     [
         [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
         [
-            0.97514489,
-            0.96762971,
-            0.96164186,
-            0.95463443,
-            0.9941512,
-            0.99023054,
-            0.98559401,
-            0.98230057,
-            0.96601224,
-            0.95119599,
-        ],
-        [
-            0.92960533,
-            0.92638451,
-            0.9056675,
-            0.9487176,
-            0.9546961,
-            0.96165853,
-            0.96172303,
-            0.92686401,
-            0.92122613,
-            0.89957239,
-        ],
-        [
-            0.86490963,
-            0.85723665,
-            0.83798141,
-            0.90816201,
-            0.90816097,
-            0.90815301,
-            0.9081427,
-            0.85933627,
-            0.85146935,
-            0.82948586,
-        ],
-        [
-            0.77430346,
-            0.76731372,
-            0.74372311,
-            0.89128774,
-            0.89126885,
-            0.89125066,
-            0.89123521,
-            0.76858589,
-            0.76106647,
-            0.73807776,
-        ],
-        [
-            0.66098109,
-            0.65327697,
-            0.63090644,
-            0.33086588,
-            0.3308383,
-            0.33081937,
-            0.33080718,
-            0.6557468,
-            0.64825099,
-            0.62593375,
-        ],
-        [
-            0.52526945,
-            0.51832586,
-            0.49709412,
-            0.25985059,
-            0.25981009,
-            0.25977729,
-            0.25975222,
-            0.52118958,
-            0.51426328,
-            0.49323164,
-        ],
-        [
-            0.3697845,
-            0.36318971,
-            0.34424661,
-            0.17386804,
-            0.17382046,
-            0.17377993,
-            0.17374668,
-            0.36689317,
-            0.36036096,
-            0.3415582,
-        ],
-        [
-            0.19546374,
-            0.1909659,
-            0.17319999,
-            0.08423318,
-            0.08417993,
-            0.08413242,
-            0.08409104,
-            0.19393909,
-            0.18947485,
-            0.17185031,
+            0.8884930952884654,
+            0.8626656901726876,
+            0.8301161870669913,
+            0.9757179300830185,
+            0.9989819637626414,
+            0.9994717624885747,
+            0.9954377526794013,
+            0.8898638133944221,
+            0.862604343021387,
+            0.8277862494812598,
+        ],
+        [
+            0.7765718877433174,
+            0.7363731552518268,
+            0.6871875923653379,
+            0.9753673327387775,
+            0.9893175316399789,
+            0.9944181334242039,
+            0.9936979128319371,
+            0.7778001700035326,
+            0.7362622619974832,
+            0.6848377775329241,
+        ],
+        [
+            0.6648416226360719,
+            0.6178079903692397,
+            0.5630152545966568,
+            0.8278402502498404,
+            0.82790391019578,
+            0.8289702087149963,
+            0.8286730258710652,
+            0.6658773633169731,
+            0.6176836507071695,
+            0.5609165245633834,
+        ],
+        [
+            0.5534420483956817,
+            0.5055401989946189,
+            0.451865872383879,
+            0.7541423053657541,
+            0.7544115886347456,
+            0.7536884376055174,
+            0.7524927915364896,
+            0.5542943466824017,
+            0.505422678400297,
+            0.4502051549732117,
+        ],
+        [
+            0.4423657561928356,
+            0.398221575954319,
+            0.35030055029978124,
+            0.4793202144786371,
+            0.48057175662074125,
+            0.4812057229564038,
+            0.48111949176149327,
+            0.44304092606050766,
+            0.39812149713417405,
+            0.34902458531143377,
+        ],
+        [
+            0.3315561576450342,
+            0.29476346732036784,
+            0.2558303772864961,
+            0.35090405668257535,
+            0.3515225984307705,
+            0.35176548159366317,
+            0.3516979775419521,
+            0.33205839061494885,
+            0.2946859567272435,
+            0.2549042599220772,
+        ],
+        [
+            0.22094175240967673,
+            0.19431840633358133,
+            0.16672448058324435,
+            0.22716195845848167,
+            0.22761996456848282,
+            0.22782525614780919,
+            0.22781876632199002,
+            0.22127471252104777,
+            0.19426593309729956,
+            0.16612306610996525,
+        ],
+        [
+            0.11044782531624744,
+            0.09623229814933323,
+            0.08174664901235043,
+            0.11081911718888311,
+            0.11102310514207447,
+            0.1111041051969924,
+            0.11108329076967229,
+            0.11061376973431204,
+            0.09620592927336903,
+            0.08145227209865454,
         ],
         [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
-    ]
+    ],
+    dtype=np.float32,
 )
 
 SINK_MID_OUTPUT = np.array(
     [
+        [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
         [
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-        ],
-        [
-            9.99996103e-01,
-            9.99994823e-01,
-            9.99993550e-01,
-            9.99930863e-01,
-            9.99990782e-01,
-            9.99984683e-01,
-            9.99979000e-01,
-            9.99997804e-01,
-            9.99995985e-01,
-            9.99994325e-01,
-        ],
-        [
-            9.99989344e-01,
-            9.99988600e-01,
-            9.99984099e-01,
-            9.99930123e-01,
-            9.99926598e-01,
-            9.99824297e-01,
-            9.99815032e-01,
-            9.99991228e-01,
-            9.99990881e-01,
-            9.99988462e-01,
-        ],
-        [
-            9.99980787e-01,
-            9.99979264e-01,
-            9.99975828e-01,
-            9.59669286e-01,
-            9.59664779e-01,
-            9.59656566e-01,
-            9.59648332e-01,
-            9.99983882e-01,
-            9.99983038e-01,
-            9.99980732e-01,
-        ],
-        [
-            9.99970181e-01,
-            9.99969032e-01,
-            9.99965730e-01,
-            9.45197806e-01,
-            9.45179593e-01,
-            9.45163629e-01,
-            9.45151458e-01,
-            9.99973352e-01,
-            9.99973254e-01,
-            9.99971098e-01,
-        ],
-        [
-            9.99958608e-01,
-            9.99957307e-01,
-            9.99953444e-01,
-            4.24743523e-01,
-            4.24713305e-01,
-            4.24694646e-01,
-            4.24685271e-01,
-            9.99960948e-01,
-            9.99961829e-01,
-            9.99960347e-01,
-        ],
-        [
-            9.99946675e-01,
-            9.99945139e-01,
-            9.99940312e-01,
-            3.51353224e-01,
-            3.51304003e-01,
-            3.51268260e-01,
-            3.51245366e-01,
-            9.99947688e-01,
-            9.99950165e-01,
-            9.99949512e-01,
-        ],
-        [
-            9.99935877e-01,
-            9.99934088e-01,
-            9.99928982e-01,
-            2.51197134e-01,
-            2.51130273e-01,
-            2.51080014e-01,
-            2.51045852e-01,
-            9.99936187e-01,
-            9.99939716e-01,
-            9.99940022e-01,
-        ],
-        [
-            9.99927846e-01,
-            9.99925911e-01,
-            9.99920188e-01,
-            1.31550973e-01,
-            1.31462736e-01,
-            1.31394558e-01,
-            1.31346069e-01,
-            9.99927275e-01,
-            9.99932142e-01,
-            9.99933313e-01,
-        ],
-        [
-            9.99924204e-01,
-            9.99922004e-01,
-            9.99915767e-01,
-            3.04861147e-04,
-            1.95998056e-04,
-            0.00000000e00,
-            2.05182682e-05,
-            9.99923115e-01,
-            9.99928835e-01,
-            9.99930535e-01,
-        ],
-    ]
+            0.9999957448889315,
+            0.9999781044114231,
+            0.9999142422442185,
+            0.999853253199584,
+            0.9999918403054282,
+            0.9999874855193227,
+            0.9999513619364747,
+            0.9999589247003497,
+            0.9999861765528631,
+            0.9999939213967494,
+        ],
+        [
+            0.9999918011366045,
+            0.9999588498417253,
+            0.9998388659316617,
+            0.9998496524281603,
+            0.9999154673258592,
+            0.9997827845182361,
+            0.9998160234579786,
+            0.9999163964511287,
+            0.9999743435786168,
+            0.9999894752861168,
+        ],
+        [
+            0.9999883847481621,
+            0.9999427334014465,
+            0.9997703972600652,
+            0.9853967608835997,
+            0.9852517829915376,
+            0.9853308520519438,
+            0.9854102394414211,
+            0.9998728503298413,
+            0.9999642585978225,
+            0.999986204909933,
+        ],
+        [
+            0.999985544721449,
+            0.9999296195017368,
+            0.9997066149628903,
+            0.9753803016111353,
+            0.9750688049429371,
+            0.9749211929217173,
+            0.9750052047129354,
+            0.9998284130289159,
+            0.9999558481338295,
+            0.9999837966320273,
+        ],
+        [
+            0.9999832723447848,
+            0.9999192263814408,
+            0.9996472692076177,
+            0.90541293509353,
+            0.9049945536526819,
+            0.9051142437853055,
+            0.9057005861296792,
+            0.9997839348839027,
+            0.9999490318922627,
+            0.9999820419085812,
+        ],
+        [
+            0.9999815409510937,
+            0.9999113168889934,
+            0.9995930143319085,
+            0.8370025145062345,
+            0.8358345435164332,
+            0.8358231468627223,
+            0.8369430449157075,
+            0.9997408260265034,
+            0.9999437526409107,
+            0.9999808010740554,
+        ],
+        [
+            0.9999803198262347,
+            0.9999057164296593,
+            0.9995461103528891,
+            0.7047260555380003,
+            0.7023346743490383,
+            0.7022946969603594,
+            0.7045662738042475,
+            0.9997017258131392,
+            0.9999399744001316,
+            0.9999799785302944,
+        ],
+        [
+            0.9999795785255197,
+            0.9999022923125928,
+            0.999510772973329,
+            0.46283993237260707,
+            0.4577365087549323,
+            0.4571888733219068,
+            0.4614967878524538,
+            0.9996710272733927,
+            0.9999376682163403,
+            0.9999795067125865,
+        ],
+        [
+            0.9999792877553907,
+            0.9999009179811408,
+            0.9994950057121632,
+            0.05049460567213739,
+            0.030946131978013824,
+            0.0,
+            0.019224121648385283,
+            0.9996568912408903,
+            0.9999367861122628,
+            0.9999793358521326,
+        ],
+    ],
+    dtype=np.float32,
 )
 
 SINK_MIN_OUTPUT = np.array(
     [
         [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
         [
-            0.99997545,
-            0.99996582,
-            0.99995245,
-            0.99856594,
-            0.99898314,
-            0.99777223,
-            0.99394423,
-            0.98588113,
-            0.97283215,
-            0.96096504,
-        ],
-        [
-            0.99993872,
-            0.99993034,
-            0.9998832,
-            0.9986147,
-            0.99848741,
-            0.9972981,
-            0.99723719,
-            0.94157173,
-            0.9369832,
-            0.91964243,
-        ],
-        [
-            0.99990802,
-            0.99989475,
-            0.99986873,
-            0.98610197,
-            0.98610047,
-            0.98609749,
-            0.98609423,
-            0.88741275,
-            0.88112911,
-            0.86349156,
-        ],
-        [
-            0.99988924,
-            0.99988509,
-            0.99988698,
-            0.98234089,
-            0.98233591,
-            0.98233065,
-            0.98232562,
-            0.81475172,
-            0.80865978,
-            0.79033138,
-        ],
-        [
-            0.99988418,
-            0.99988484,
-            0.99988323,
-            0.86796555,
-            0.86795874,
-            0.86795283,
-            0.86794756,
-            0.72418193,
-            0.71847704,
-            0.70022037,
-        ],
-        [
-            0.99988241,
-            0.99988184,
-            0.99988103,
-            0.85528225,
-            0.85527303,
-            0.85526389,
-            0.85525499,
-            0.61716519,
-            0.61026209,
-            0.59503671,
-        ],
-        [
-            0.99988015,
-            0.99987985,
-            0.99987875,
-            0.84258114,
-            0.84257121,
-            0.84256042,
-            0.84254897,
-            0.48997924,
-            0.49083978,
-            0.46891561,
-        ],
-        [
-            0.99987865,
-            0.99987827,
-            0.9998772,
-            0.83279589,
-            0.83278624,
-            0.83277384,
-            0.83275897,
-            0.36345545,
-            0.33690244,
-            0.35696828,
-        ],
-        [
-            0.99987796,
-            0.99987756,
-            0.99987643,
-            0.82873223,
-            0.82872648,
-            0.82871803,
-            0.82870711,
-            0.0,
-            0.26106012,
-            0.29978657,
-        ],
-    ]
+            0.9999961997987318,
+            0.9999801752476248,
+            0.9999185667341594,
+            0.9993115972922259,
+            0.9999536433504382,
+            0.9997590064584757,
+            0.9963282396026231,
+            0.9020645423682648,
+            0.965641014946897,
+            0.9847003633599846,
+        ],
+        [
+            0.9999926824858815,
+            0.9999628275604145,
+            0.9998472915971415,
+            0.9992953054409239,
+            0.9995550237000549,
+            0.9972853256638443,
+            0.9958871482234863,
+            0.8006505271617617,
+            0.9360757301263053,
+            0.9734843475613124,
+        ],
+        [
+            0.9999896427490426,
+            0.9999484707116104,
+            0.9997841142091455,
+            0.9321779021295554,
+            0.9308591506422442,
+            0.9299937642438358,
+            0.9286536283468563,
+            0.6964658886602826,
+            0.9106656689679997,
+            0.9652109119709528,
+        ],
+        [
+            0.9999871227708508,
+            0.9999369646510842,
+            0.9997276125796202,
+            0.9006206490361908,
+            0.8987968702587018,
+            0.8965696900664386,
+            0.8941507574801211,
+            0.5892568658180841,
+            0.8892240419729905,
+            0.9590996257620853,
+        ],
+        [
+            0.9999851119906539,
+            0.9999280075234918,
+            0.9996788394671484,
+            0.778755271203017,
+            0.7763917808258874,
+            0.7737517385551721,
+            0.7707980517990098,
+            0.4788014936236403,
+            0.8715671104783401,
+            0.954632732759503,
+        ],
+        [
+            0.9999835837292402,
+            0.999921323618806,
+            0.9996389455307461,
+            0.7222961578407286,
+            0.7186158832946955,
+            0.7146983167265393,
+            0.7105768254632475,
+            0.3648911004360315,
+            0.8575943501305144,
+            0.9514642802768379,
+        ],
+        [
+            0.9999825081019064,
+            0.999916683268467,
+            0.9996093996776352,
+            0.6713490686473397,
+            0.6664914636518112,
+            0.6613110504728309,
+            0.6558325489984669,
+            0.247299682539502,
+            0.8473037957967624,
+            0.9493580587294981,
+        ],
+        [
+            0.999981856118739,
+            0.9999138938063622,
+            0.9995907248497593,
+            0.6331535096751639,
+            0.6271637176135582,
+            0.6206687804556549,
+            0.6136262027168252,
+            0.12576864809108962,
+            0.8407892431959736,
+            0.9481472656653798,
+        ],
+        [
+            0.9999816006081851,
+            0.9999127861527936,
+            0.9995832399159849,
+            0.6133274396648696,
+            0.6086364734302403,
+            0.6034602717119345,
+            0.5978473214165134,
+            0.0,
+            0.8382338778894218,
+            0.9477082231321966,
+        ],
+    ],
+    dtype=np.float32,
 )
 
 SINK_MASK_OUTPUT = np.array(
     [
+        [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0],
+        [0.0, 0.0, 0.0, 0.9047934405899283, 0.9936046284605553, 0.9448690902377527, 0.0, 0.0, 0.0, 0.8363773255131761],
+        [0.0, 0.0, 0.0, 0.90375200446097, 0.9434594475474036, 0.4716831449516178, 0.0, 0.0, 0.0, 0.7364197333910302],
+        [
+            0.0,
+            0.0,
+            0.0,
+            0.09080438801405301,
+            0.06774182873204163,
+            0.038207095016625024,
+            0.0,
+            0.0,
+            0.0,
+            0.6745641479264269,
+        ],
+        [
+            0.0,
+            0.0,
+            0.0,
+            0.01731082802870267,
+            0.013540929458217351,
+            0.007321202161532623,
+            0.0,
+            0.0,
+            0.0,
+            0.6341231654271253,
+        ],
+        [
+            0.0,
+            0.0,
+            0.0,
+            0.0006444251665178544,
+            0.0005397129128756325,
+            0.0003048384803626333,
+            0.0,
+            0.0,
+            0.0,
+            0.6070178708536365,
+        ],
+        [
+            0.0,
+            0.0,
+            0.0,
+            5.406078586212675e-05,
+            4.416783924970537e-05,
+            2.4597362039020103e-05,
+            0.0,
+            0.0,
+            0.0,
+            0.5889413683184284,
+        ],
         [
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-            1.00000000e00,
-        ],
-        [
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            2.86416400e-01,
-            7.93271181e-01,
-            5.81341234e-01,
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            1.98395623e-01,
-        ],
-        [
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            2.66733297e-01,
-            2.80741490e-01,
-            4.14078784e-02,
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            7.91676486e-04,
-        ],
-        [
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            1.86244537e-04,
-            1.53413401e-04,
-            7.85806495e-05,
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            5.09797387e-06,
-        ],
-        [
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            9.62904581e-07,
-            7.23946225e-07,
-            3.68824440e-07,
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            4.79525316e-08,
-        ],
-        [
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            1.50939343e-10,
-            1.17724874e-10,
-            6.21760843e-11,
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            6.08922784e-10,
-        ],
-        [
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            2.57593754e-13,
-            1.94066716e-13,
-            9.83784370e-14,
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            9.80828665e-12,
-        ],
-        [
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            4.22323494e-16,
-            3.17556633e-16,
-            1.60789400e-16,
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            1.90789819e-13,
-        ],
-        [
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            7.72677888e-19,
-            5.83029424e-19,
-            2.95946659e-19,
-            0.00000000e00,
-            0.00000000e00,
-            0.00000000e00,
-            4.97038275e-15,
-        ],
-        [
-            2.71345908e-24,
-            5.92006757e-24,
-            2.25580089e-23,
-            3.82601970e-18,
-            3.82835349e-18,
-            3.83302158e-18,
-            3.84002606e-18,
-            8.40760586e-16,
-            1.83433696e-15,
-            1.11629633e-15,
-        ],
-    ]
+            0.0,
+            0.0,
+            0.0,
+            4.39259327223233e-06,
+            3.6050656774754658e-06,
+            2.0127120155893425e-06,
+            0.0,
+            0.0,
+            0.0,
+            0.5774279920364456,
+        ],
+        [
+            0.0,
+            0.0,
+            0.0,
+            4.0740501726718113e-07,
+            3.374875487404489e-07,
+            1.9113630985667455e-07,
+            0.0,
+            0.0,
+            0.0,
+            0.5709897726747111,
+        ],
+        [
+            3.2266922388030425e-17,
+            1.801110982679718e-14,
+            9.325899448306927e-12,
+            3.913608442133728e-07,
+            3.9581822403393465e-07,
+            4.02383505118481e-07,
+            4.14820241328287e-07,
+            4.281640797396309e-06,
+            0.0023900192231620593,
+            0.5686882523793125,
+        ],
+    ],
+    dtype=np.float32,
 )
 
 
@@ -527,6 +491,21 @@ def setUp(self):
         self.input_img_torch = torch.from_numpy(TEST_INPUT).unsqueeze(0)  # mock image (torch tensor)
         self.input_mask_torch = torch.from_numpy(TEST_MASK).unsqueeze(0)  # mock mask (torch tensor)
 
+        self.real_input_img_paths = [
+            os.path.join(os.path.dirname(__file__), "testing_data", "ultrasound_confidence_map", "neck_input.png"),
+            os.path.join(os.path.dirname(__file__), "testing_data", "ultrasound_confidence_map", "femur_input.png"),
+        ]
+
+        self.real_result_npy_paths = [
+            os.path.join(os.path.dirname(__file__), "testing_data", "ultrasound_confidence_map", "neck_result.npy"),
+            os.path.join(os.path.dirname(__file__), "testing_data", "ultrasound_confidence_map", "femur_result.npy"),
+        ]
+
+        self.real_input_paramaters = [
+            {"alpha": 2.0, "beta": 90, "gamma": 0.03},
+            {"alpha": 2.0, "beta": 90, "gamma": 0.06},
+        ]
+
     def test_parameters(self):
         # Unknown mode
         with self.assertRaises(ValueError):
@@ -683,6 +662,44 @@ def test_func(self):
         output = transform(self.input_img_torch, self.input_mask_torch)
         assert_allclose(output, torch.tensor(SINK_MASK_OUTPUT), rtol=1e-4, atol=1e-4)
 
+    def test_against_official_code(self):
+        # This test is to compare the output of the transform with the official code
+        # The official code is available at:
+        # https://campar.in.tum.de/Main/AthanasiosKaramalisCode
+
+        for input_img_path, result_npy_path, params in zip(
+            self.real_input_img_paths, self.real_result_npy_paths, self.real_input_paramaters
+        ):
+            input_img = np.array(Image.open(input_img_path))
+            input_img = np.expand_dims(input_img, axis=0)
+
+            result_img = np.load(result_npy_path)
+
+            transform = UltrasoundConfidenceMapTransform(sink_mode="all", **params)
+            output = transform(input_img)
+
+            assert_allclose(output, result_img, rtol=1e-4, atol=1e-4)
+
+    def test_against_official_code_using_cg(self):
+        # This test is to compare the output of the transform with the official code
+        # The official code is available at:
+        # https://campar.in.tum.de/Main/AthanasiosKaramalisCode
+
+        for input_img_path, result_npy_path, params in zip(
+            self.real_input_img_paths, self.real_result_npy_paths, self.real_input_paramaters
+        ):
+            input_img = np.array(Image.open(input_img_path))
+            input_img = np.expand_dims(input_img, axis=0)
+
+            result_img = np.load(result_npy_path)
+
+            transform = UltrasoundConfidenceMapTransform(
+                sink_mode="all", use_cg=True, cg_tol=1.0e-6, cg_maxiter=300, **params
+            )
+            output = transform(input_img)
+
+            assert_allclose(output, result_img, rtol=1e-2, atol=1e-2)
+
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/tests/testing_data/ultrasound_confidence_map/femur_input.png b/tests/testing_data/ultrasound_confidence_map/femur_input.png
new file mode 100644
index 0000000000000000000000000000000000000000..0343e58720d7bf2dd741d273111170f75de8ed34
GIT binary patch
literal 125674
zcmV)GK)%0;P)<h;3K|Lk000e1NJLTq00G<p00ICA000004cvm@004jhNkl<Zc-k!e
zNsp#Wwxx&d#@e^}8h4L4aWeC^RU{<{;y@Y*0yNQ7&`^M&fhPRP)PMlVs=9SE^E46e
z{xxrB?cN$3{sXgt**2{8tY!FrcmWB)FcL9qGaezQrCvt>-5*V3d~C*8%WMdGR~2bE
zaBaP{VYUWQ>XGvEhq(D<3;+`%^w#zBJ-630Y0_Y=bxf_Iq|)s?O$^7DkN0()W={aU
zcbNuDB_Gclj>G63W2-5bw3#XR@873s@PNix##;8B{o5?*{#?umo$0=x-J<7t3|_iO
zs)5`6ukIf%6KboaTWYQL)?1fxKmfMZf+$FDRTV+qO!e4BYXhF5cLsW`r6OBrWZ!o?
z1?Oz3RX}_1(&K1o-`6^Mt6d=-LsUr(+{{fh9oFor;$X!Df>O8S{P{FFOv`@E7v|a_
z|H#HzrtiPM%&Co5ZN%O?sPwMZ83toPzyClAyjO_ld33hsZC$G!W(Q;dQYJ<M<Cv@$
zd9Gam8Rhw7&!-Fi-+;(FrY4Bg8?b7r0JUS!U2HJ?xm_nfkg7l&gY?>gtzlGVHh|jJ
zLi?@aID40hV04@wwW~7K9&JZwz?fEZE>(MWuP;}}y)Vl%#~~W9fvF^gV70fj)h-%>
z;5@LDru(CV^G6+=qIKT7R*mh-q*BuMxVa(a=nZ3h87=hIQ(E`l8%1jn6r4FK%c{X3
zlQH{YNWmGFN>-(c_l}zxA*cPhL~p`%_ST{<_xI1AUKLP#T1&frwWhU__EgLSb$&r;
zWvib0cI#-Z;ZnrRqp!=lu^ERdL`c*NS%lu2jzeI5eCP3U1;`ESeQ)foISwL%)<v;3
z>*rbP_lJ#_XaQ<T$y%oJ{DptI%+{v;{TYK{Lj>BJ_S-fLL-dA2MA-IHI#`2c+jrxP
zL_iQwXzP(vb}?8$WZ%~3rWVndqiuC*(_rhaMhFE6T{O_^wtuYS(8Y$qU4X68RK+<<
zD#*cm$Uk;7jnP;GE$3QODu<RmhN{xm=dM#29QncGBdx_cMr{~*->P--JR4Ne{oFA4
zo^z?aC@|Z>n3mSPlyq#3Y?NG7TbYct1T{Y&%MNCmCI@5)rS@KQ4qi%=TEwxTx~1-?
z;8B(%`#BnGj0@3uK@}6cwOHCB*GM_5pRTa)arOp4>b~#mL+pqagT`$;N~-8CXVX4Q
zh=JPYFMSBp96=FKB`JAlHQzH)N|}DLh!4;%k{)~aAbDS1h}@;8&CK(89&x(^8lYh0
zncSHHnP0y)gU9o@Jx$cIHX_*WMcEiGN3F&>VMo@cENl=R4IK^~$LNZjXDjJA)cE0a
zqMr7nV>`fmKMv5+e(=jBLi+yO*8SiO4zClJ{rT9du|x=BhH-LK#0R4-Z_Dndo~ndk
z4I!|B-n)RImbdNbWW95QIW;p|Aybdk=Gj2oA39HgkPI77Qqz&Z`Vefd$B}DOWM)Qa
zWiJG^WD#k#`{_IYurV=0Ex9%k0W-{_$MSeIbVh4UxwHarFN_|JqBa-+DUA$E<-x(W
zvX%TyLonmnY0Au6x4qcw>xt`;(U38kj{R{rXK$JsZ`*T8ia^6;2&6UbwYGR3f;X*}
zTspLh)_aDmvrX^c-Or|qqN<dX_5*zMs!EPh#op}{y&<VfiGfLv=llKT+c-?4GnyX1
z{L&dI`no>MUtV2JrPXdNqe>U?<LK-1Bmexx6SSIhx*NyPMY0haBQiX16@sz$mL|{W
zhWT>BvfS^j-k-rj74|X8eI*<~!Chw7+NB&|z4vyMoSbu_+k5EG&++wq)+*8~0J32=
z2HWDy-o|fpr2gZx617C3r19l6In3))xhFENXJL}GrVdDad138>)@tAP0`tsvQue{f
zcKh`)U)}iSq_wQ;(^~6iO+TK<oDa*W`%y7^+COQYO|EaR-XOLn2rYv_W)x*>v}_A`
z%R!bK#Yc2xh|sv<nI8FOsb(c+$d3c;DfqS?x#2XLYws<k+h^z36QBi8mdF?aB&%Hn
zP*QK!SUi>v=k`O?#o(m^SxcQ=T-Vciu(h<DK^0QTI)8DbEj|9)rvTv`xEzunKmO)^
zdOPb;V3;l=4;Ssno=a9B)>e<Zf!2rDZ!9%yZnfh7eLz&n)i4v0+}_n)Uq+L2GY(tR
zI8fh@Qp&cvi-8bDxNS0wq2~(cEBCbA4)o&~K>G2^eLgYt=bgjn$It(Cj@~jz?*gC*
zNQ9c#Os6lG*dG7+$9yu{1l=$WjN0?D=3djf#i3*Y!!(@mSQhrdlZ}>9o*$Xc7ek23
z-opOKFwbRMvh}V!pF7xjl=OhN8T0e8sDHi4{$sVqy!?EPX}7}|u$OI7@14PI+4dxM
zn6UM}?aqTXw3ZB)O#X6ascF02X`TiYDcfEJeKbl)x;}2^U%wgYt*%QEW398+I;Vo9
z=nQJDy`+5uqb(nq&m)Mm1{8xKp<uLBk;nS_b#j`Pj|ZJj14_<Ss3jW88qQ;IxNZeK
z*4t;Mmvh}7`Z8la+zGq%{QSP3uagmxJ%w>Pod<0-9l7dmrl2Z~Tk8yJ-0X@@W3*F0
zl)I{w+8TtH(D!w3aSpg18b@MDsmAjZ5NvrqPGbl*t=nPV<o3ktbnX?0i-X=>umI5V
zx-(<XFmSYexn15Mn@ez(bbE-M25qJF*4)MS_ofv%Rp%`YvxW9}fSso35cT?|hOn0^
zmb8=qGK|=c_MhH=`g!s{9b4|I8s8=-TVXe#Kb^(9$7k`7nh#?G09As#<IChhT5FoO
z?FKl{NJt&VVe*!AeO6*a+MANG(`ZZS72<`Ppygv9Mnjv};0RSp(Yo*2i+9d}sP=vk
z5xJh9pX+cUC>i6)<qpwFmqbil1ervxCsVVcDD<8yAV_|H^p?Zx;B8wEhOPK$IiCJO
z0ZDpAuWKKt3*=PTjYa&-MFElFsCNV+RNz+e|LmEWm;qHquoiMttP-rWwk>wz{@i%{
zayHPOH?^#Ln<k5WeH426VpSBkzoyr>i^q~uAMF19zpSU%7b^;edTajc*`jREB;x@2
zwi5xu<I_(J4dNKBl>A)KsMz2wk#%&e!d;OKbfNJ&`JPov=_Yzs>1|nCuNb5E4%C^n
zmi^H#*U1tyN~^g@+qbRZID6`4gO{Nk$I<35(V(i-*7N>YO+<(h0n+DzD$~g_VaFJ(
zF^H}8-fP~M=H@w=w*U5dx_avvS|z%D%ohh4q9dkyl;)iFo_81qqn)VbZ7+0+uzVzV
zc_V1xg4bg^;+duATk=+p!<`(q=HuW2q#WxO=k5Nib7&AWiPqu=KU`0iq=SvYF@o0P
z@mQL17691>Pr3BGrB-rlRg5uiu*`UTnty*ghjH*0dn+j)N2Yvhh0ddunoG^SyGh%R
zPa^HDtxCpQHUxo`2dep~jPv<42kK^udB2yf)-B9FSfi~-%jG^&dED&;)zN8$-dc-q
zz1|P>>H^ToeqC-)^>5c``ug#Kfb(pasMYj*+<#asEz<2NCY)a8(2kEA`d~Ge!<4-$
zG2lSF9xV@7&uCPi2adBb%srPQvHLPWUf1OZm}pb(HJAGQ$Wq}S|6vU5Cu)MMtviES
zG&5<dg8j$GuH)$K5V7<)48dYk!~MBbW60Q{A3GQebzM1}W|rgeF43ZjinOJ*p#EQ?
zGv<%;@6cf<kgC$VpRreTMsppe!PHL1lhk`I9o*Uau|JNQ>^Qr4ok_Zo4VG1@e!f3b
zSM55S&Xcu<wWh7#epSabTJOkM6a)gJ8_H_NKuCZgrbFuYAAW`d01WqaDSv+PnsVzx
z#!zYF+4<3fg0!?g4<C#<8|=^Cx}xq!Zr6*OhTxs)MU~UzKGnxsjbr;Vfgr-M7Im!n
z)7QaUEorOGUC~`nh7@#7$o)ACr0rPNOeW>e<un$&@cMK<D7=32R%$wOA}Q<fdbN;`
zqp51sV0vdWUdF(Z)6u|^BTOb<U+VMrJV412wARKUH4anoRPn7o)@+f<vNQz;pnYGS
zYlASq3{Za^F8(zBY1x-nd2qV?-fb;`EBN4@WjLn-<G`XzYRsnOcs_;FKxw=Ls0B1v
zFqTY9l3PWXFVpDbU>RgzyFyM!Mw>g(TiCM=(HWsxfDE<vdACqGkj8KvSpWWMr>mFT
z;yF6fnoB;?k(2WX?1HHk8IZz=|9fbCYp-WClS2~-JimXc@eF84jy3fdoiR){8p!)v
zrrDTafi2aK-|yw8FW$1Gl%=mrn1W%1-ppW0+kO-nP3gG&s3gsdFT?s+Xv!beyuJee
z{3TFN>-S$_yqqmV&jk?x9DV<O{(26Qw}OPw*WCu^BiCFMOF5qzdO6lr;yg$K9wUmB
zq=tDMT0`SPG_tQ^828i_*o~7TZ8>dwR%-yMx6-VwP4fr*dG5!N+YqasP8NG@0`>TQ
zpRd6gEbqVCx5?IHPlDD*BWZ2DOEk&+{N-dNAGI6fSU&&DJzib{mg>gK$!M)z6tHCG
zD#kI&b}y2O3&g>(4~|9Bk@Jz8+VNxnjiugQ@YrhU=xopHvK7_>yf5Qb^HF^q&lj)h
zP#Okrr0n<4L~$^>KG)3H>}mAY@@nRob_?eT%_vJV=hyS_b6F2GYz?&MBJ3&kIE-;H
zvfOxcvNs@f1J)cbF?uNXqqo}oe$-w)8SevP?X~X34bi%e(@$u6>kv84Rk4hFKc6l0
z0Zutw4G!KpdETDQd1npK%-6TcQ_HQccd?f%vA5c=-ZRBv`O$Fn+8U^+59`+C?c$*3
zec6EGV2v?GY88KqyzfwMM^z1`=GKL&0V1eEFr?60k0d%wZQU3i8=f3DtsO1%nVrFW
zEW3lpN&?rvSj$Yk^|sybJ9uv$G%4%KLkOgKe?P82{WMT3t!*1Y%Na}zWv?B{I?vib
zJI7#iQlY(<3Mj1!u%E7{(KGZ~m8~%bL4b`Z`B<N8RfO(LxkXd*Pk$P`Ck^v{<lcrD
zhv9TIHyUW^{@Z_DU(cs0I%}<wI4S8-*)hjP!9hc^T~u;L;=x7FU95K&0mAiEZnpyl
zPse9b8)Ed7w@3DtQW>tj)qTZpUq<eYf+a%eCHHu~1k&D0Zi*-+7Xd((?Fn{>_B^W+
z_v7Oer`HKV6s}*LRYAi1`2JnTm)GblmgTY0%XJF9bR`Dq3dcfzxPluTg3;PO;c>kJ
zQa!fkt`lSQMmISO8O7y8?De%bAZ$lU98har4x*1|w2Z;viMRg+z_5y__I9-3sUH5#
zY3nAQPZqjpX-Y^Y3}=h_@f{ezdPmyY@ULZbqymiX<Foginm<QdlA?ELyoF<F?47Rn
z9H&4nWjhdRS1>W+`cPFS=v9!En8;d!-t%_Qzz}0egcopywG&zoa$Ci}T*uQXEx-QH
zJO)#t7GMGZl#c(87i+NJ7v&Iw^WK2evR7+B0F8ByOI1;{-V!lE`Y*rGco|^*=uS0`
z^MIherRC#3y=o&Ud^tyN4O?d*9qHJ`PN#s|nyUgLfGAqqtF*K}OB@H&Q#QWWLx&+6
zBuv|p8;+xewA>rJc)oaRpd3pk7p)rxROwBm=KHPM)5#DM0I8%y*sua?+1D&#3_e(`
zTaUq!-rntLjyZS0dK`^d6=iuYIt|Wy?}ID*wmh;4!JE93@fL}UkHO0Nc&2f3S{w8V
zmY_6dhFY4k4Z%4JB7jo6NH?a|R`apt?nB$z$8iYEOnPi;v_AI`7zH3_cbcP?_k9SY
z0Dp)*P#e@$3@95S)Z=IzWaHp%hcWcdjASg=V_A-7E`ND*kiY+y0ku|ilP4r&cr&k8
z1cs^g*3A&yABsaQ$i^|Cbp}v+If{l;w3hRRS3g+GCLGyK!{EI$lGnVfJ>8~p9D_ye
zwZ;%5Dw`19o^KE6&s%|cAad3?=S;7?bQ}8pc^YilmksO~kLMyWSV@nE8&(JTnc(^|
zd(#?PBVFrmCPUE!U{6(gEdu5Jqqof0lamT$Y1^w9$2<=u{g#`t)^HH1P;U=d45{Q=
zg{)&FVw}PYM+>zT)z)ecZvokQ`u^Ma?dt`42b5e(eLmCnA!hJwM{9_BzTJyD7qp;@
z1CvUt+kO-^-s0zaIeTx8s;!f;TFvCJE0`JKc{CbNaI8mD?7hQynxx%-O+&P;HabrR
zkE|wAFUOHO8isl_!7@9C2J+FAoWou-^s>QG-u*C~#(hI9oqg1F<PJ8QVH71)X-$xN
zVv^p}D3h}W6*^ltNiS*bS$&0V?cU1P1t1sWEHPtCskkpMEIN$d8tkpF3&iuqv9!8B
z?n~{<(?nx3u{8iCg;CqFubD?^B@CDP_hpxzmIK4&h|TyiMUb{Esn8fav1L$1Pz-0=
z>$7!exm(L;D`nr(wrEO&rjF24AE#lATwC7}vx-XkzPUGF_gw+qc`zmMJQ_r2RoFNK
zJ>56)A+n!_)5vVFmd`kyt_;1Ewd@66zT*CT>hLlS(P1rRyQkE^1n2yDw5Y07mLE&_
z_A+6MTiKmEjqUj%w$#CcN?Z5J0eiPYdOnCms)-JM{QKM28J5ilL;3j_F491HS)fKY
z$SixY23n~Ntueu}I6*bms3@3t0%`p&s$n!{8lfH~lSNa7yYT}6fb=FJ(v*$%)|#?+
z!7=u3g10I)Z67~+3^teA8KoTH7}JL}LjdWD?kA(4w_u<i3V_Dn{%&W(mfH3>glWG|
zvp3E<AFVz=?a8s)X*4AJip{wfxgDOw+G+NX%KY^*l9o^HcXn8XkG>zxgm?+Qwc1))
zs|~E3W^U!!);VhqHq??1HE28^=>UiGC`hJ$e}0GkGE7&a&+R~eeT^qBwPUXuIY0I+
zJV|HgjN#)Da*n2d*}tbx1Jc^Qbe`Tk=VN=^_OHl5d0(G0nbO<dxjt7>wODdCLv#$l
zr(w-L7?iucXA`d{-k|NvqcK4k$mdk5^nMDVxBKxN`)}>%x%GNf>B}96dtblD*XwBt
zD#ReYw~~>W*bUf4RC50D{^d`z*Szfg$1mpfdfFUo+EzS8W^d8_d!K@{D7CiQnzFM{
z4m#@A&|8wOO>1wtl;i%;>rdl^AYFQEtu+L8@_eq9hB114|J$@ofshdP?>j=OJqE<n
z*|L89M-IU=5t3qjY0U;l`6xn`a=tM}t7GVChta{d8Lf#lxA5nG8AfV#{e09~ah?ZB
z=r6P9`!)_<Q$9>DWqr2Zg*<Lg|Kd4){Ah_LXM$&{8Op~0fauNx)nZ&-(RtgYm!r3%
zjlLYY(|8`ZKXXN9t%5dqQ}gmS07WfH%rH1-m~cDlaSmMfAJ5YZDYRX<-+HZ$bnk<O
z^0`=oT*3h8j3`dY4w2I#Qq9Y;R+DiWu28n68B>JZQ19xmK<k0S^=w#(Y;a%qXQz_q
zIa-6#N(X}0a%CGFNo#GVcTwz_!MZkli*FwGJ5$>Besgi0Vl)5a@27DbywTozZ+#%K
zJ`Bg66HrBw*3xn0RBBFJsZ>s<D7EBrY&1-BpmllPS9*h9A=S_H^=&kuZGE>ty^bvl
zG83`&h7drHeXs1o_*EJ(w#Th0mZfnRPbUP!n%BZ0r%Cojnt;Df1NANmEL+yJJnZp2
z{b}|#n7tV$?C-tY?m%6Un|I%T`SxYh+HduA8uSn#*`Ckr=Gl{Vnzm#73DHtL6b*Yr
z$K$v3)7t>Y_y6hZKlkm&G7OA}`aH~dowXiCh|wL-8!%f22qX7Grq%T`glM!)`+npq
z3XBe;Bh;3(-`9Dj@G^0?^B`IX0(6VE#@lk99ob+}rGEdrM)L;fgK5is8^Y|Q)Xp|q
zK7D)q`1t;Kj^laI<NhqumCKS(r^%DH)`$6PYC<UNyp3-$<_(9@Y{nhA^}1^}mvpEo
z*<p;qK-;p4kKPQnVXClgO1-4CAL%&4#XT3EM^!q0+{^feCYO%~*q^40DMyhLc?U%-
z^S8OG9k6%R-fBk+s#4Oi7c)mfLDO>I-*49T^|KFQ{`SR6tA)E^1aH?nf>mQJ2%ys=
z6|Ifq%%)f5XPPE&QhL^7S<!CgbRJ@)-pf9D-O{H2^fr0P>*Gd_#?oX>(;15&CuD0F
zC2#A|#_M@3;h9fgUE@^XLE$fGDdz;1nLI)mhhsgG8^<ajvTr5?k1Pq&j^}}y6|}T{
zUCZm2!5Kz2(6;p`?&7hRL=1|Q){H-W_0XgjJ2qp=)?^;7Y;o?x5FL@!$I++DB^q%{
zIw%=5wmGK$EBn`r?c3vKX460Se{^C%rFXV7ZO_saINFr^<Jes6ApL-ElS7~#|6eX0
z&)esU!{suM9-)>j;{a%+Ds>sAA(9;XwkIG%y(0=zd29{L`Gpl!-0Q{f-*<X>mDC07
zd9+3!ADMiNj<l9~?4=VS8t09)PS&8dUOt~ysTH+8T&|AtkxLO&3X}ISkd%EBW`g$n
zZ%5;v+}Cq3*wb$x(`Do%^<LA#J_H*tmjP?q9}nZ(@m~J&g#n>`em7He*q7&KPO_hV
zo-7ebhG=DdiV0qF(mARU$>+Phe7it7wk->3O>D;qE$=&eXL@G~;EYPG4Rqb-OLUf`
zw^~#lcRx?jqL3lkm*){DMnLGfbnnad?0^2tWO{Gu04~e>I$tjXwqrlY)#`@9mQ>I=
z#{`nr=b;=(r|TSgF6mL{$<}Q>{HPre0HO76$03sBB4`;wMO1QXbt^o78CgZ{CY)4`
z{r=c;eF%Ay5KBWrsM~GRZ~uB7gJn&n^|o)_PNTN8e8Ah!+^edY-_F!hzCRu~QLZ&a
zwt;);g*?R}Sj5;9){cH~K6>1KU;Q*W)LxHV3cX$fX<FX@XXa9EoVZlnkFS?Vy@PcX
z0FTxF15!Wka6K8etd~WN=eafkV^yqUASh6a&a-r#{o}{4WSn*UlXb=hZ%|s<*S52Z
z4v$+gFE6${J|CYPz6L`E+P-BVO)Dr~|71}6a^IVeOdwKn&PQVB1F@DhDPfbX21`WN
z`cvTT^Rf4>Qw$!JCXR$19JiW|WCuG80DngpE!0lZ{_yZiuLUN9+q;R;Q+V@)NJwzH
z+}@vbonSZzOIkjD`R(<~bsYRB8E!uwHdx5Dl68LY&=p&pf>8k|dEZTsfomhS*5rfi
zbov=kaszfCWSLM}>v7x<dM{HnprERv%sxiQsi_&J(dxcQxn<8~`S{#M^H}ZlX6k-#
zMacRopl&ti%A<d4J&ev-F8$;vSjLXpj;(O*)+tjhM;!vSr+qm&DfeG%?-Z@FAAH`;
zU`U;}Z<DbWK~)7nTUxhx9jzfKkH=O^Y2!tUiinmFOq|ETTOwpOs9ojugTg#vwkxje
zdugR!zK(&J^8L8kf5!UwAn4)CB^o8N8#W?h?&(p&Ano}%j@(L0xq6>B?||dL(hoT*
zxqQ3p*Ri{E^ds!+)`n;X#+LK>8eqBa?%SC?Fc}_2v6b3jyhQ#|v@SdJqp^<!WB{36
zIA00+j?#|z@1J0~>Cy96_Fb+XOSNADrDHEl2(?MGSW`;-R!rz=>mq=MNrt;I>^!T8
zC^8^nFSi`$pE&>W_aEbF6z~kl0<s_2O4f6*E}$qJ`}eiGGi`l5jpG?`Un}@vqche~
zFZtN+@ACS3X6c1NrS;bPLz#Q+N?)No?tyhZPA|O=p??1xl4or{en}84*vS){nu7#y
zt#|Xl)}kUH07#*l?RXg-wRP{$=A(`SxZaL1Fog5@G&%-Vt@~2I$H~vBdvE8Lv~IU!
z?|_%DKYM4XAK<>ZC}l6yK929pXjD4-?UwL*4p4goCKEMyXI~$`iu>}jMd@9T343Yc
z2Q2%%w@<i+fj~rp2^MNaa&B-WW{NHlBau2o{c+nW6A3rRo&>Ez%iUP%!zk;o+v}In
zaNdrrdMq%Hrap56!%Vi}4#zUS&QWshWE^a_zx$s?Zu$3LKkM6<?x5yvd9J3lo)uI?
zDNZbG33_YCy}^KN-#qzXTWO66v>k__^)K_}K|ytB<v1@#R~>B(-e9k}1@GfP(zG*L
zX$Z9c`|{_L&1GMo$p@=l(K<rwPm|gG_Yd-qPniI!;_1KLmm_+2y#UDZNJnlR{1|3K
zL~Pp9jrZO<BGOv+WIP-NK}&{+(ODk|pkERKaIms%y`gc;(J7(W^VEshnfa2AR9ma9
z{OY_TsSO6_ozvdZTT&#cKn9lIe%ZsD2UL3Ag@-`I!3}4K=d*=gYt2n+Rtkf&0Oj%9
z14i4@aXj}JGiyDzGF$?ykmdG!Cu5Cqo{oR}h3B`))DNU%$MAB}-rBKm6|XN;njfQX
z^;jPJd@|F3jJ*`&4f>Nx>h}DVO)Er}d(G|Gs}YXQ6n$wmfg@0mw3=ZA=&j@}s|idl
z7;K$r;%It4{9ndMTK2U$BNgn4T?hv2TAR8tc<9G*R77Zb|L8c5lQ&RHZ3(@v(0eW2
zM6p0dOWwP`j!+JF^3Hq0o*${5uEPwa_9o6dgL$pw$KbKtZ@6p_rz!a8O3uW*=flpI
z)1-jfORivbKYGc;odz-l?1Pauw%W1h$Ce5k!{pC%K!GA%Rhx*SMKz)JvKEM*5TuFt
zaU4(@o&RLC<s@i$GI{&<#cOS~<f?|H)rZ(|h<s!c=^Y4FtZ^ZDgZ(n!9{JZUL>}@p
z(J5-XSDU1C?_mE(j=*?tb^p;{zMVbPN@ysRl?H<%1(@3%R80ENv{w;e&!Ezp7(_@_
zDEO=O&h%ry=f>P}(HJkoVCwojjK|glt)by`{xE;J{vrHt+rDiyxX}Qpa`YCpfeGGF
z7qvDR6=6#VJ#CLe0f!PDAbCm!Y#6A3F@|N`?im5XV6+rwb<EM3*6b;~g~97yc$ghZ
z)4J{}eY;MM8TTbOWXk}cfY@3SdpXtA-RY#IrCfwuhYt0zeV-?zIn|D=r7`8bD?ld)
z&5xFHPBMC_W!u$}(e0yr`(hagOWOCG*5ZS;)_Ydb!8s-*gjV*ym!>>kJ9tvEt#-Zd
z-~4s5(3L_!sU<yrIgmu^mVA&_)p^T|pq->?@5b@|Y0l@X$Nlk{qHPmfi;%KE1^C)9
zZyobJ$=5H}(2i`+mQ?EU>yNh=KTh+jYT6%L<H<HbAGJP@;2l$7Z8^2xrB(8%wWS}w
zh_lA{OJr^PBc<)Ij;$M12|AcFoB4`JL?FG@3XGuIyP{z&JI26yUDmz%n7uIuTg_d8
zzF5mll-4E9lgJL&%lMC_6wz7~J%h2%LhsbX4>kln_WJhKk%4@;@j6-P%aM+ye|ftO
zB!X;W$wdZ7(p&4jHw-7s?N~Nh>$YdL^B4@YvM;yg^-G}rXkZwboh8GnYP9#d{+6@W
zNB{Ya*?LAqs41t9>X2Q4^89|^X&6TzM#oI3(t0O0OibmrWU}M(BLvU2D0BO5`}4)r
z3fdYl!JeSX`E_;-kAj4PJkQRMmgkbhxn4`PmnZ~v98a%*E;VPhabVq2h?cO|eb4nc
zFn+mmE;aulGMFyq_w99J8VJ0>fLL?2<K)|REG`ZXkeC#ZQF;m$s^;hIp3Ql-#u_cl
z{hnbskCE75T|YA^axs5hL~M+~+Q~rAxgry*v<!Lew%4|P|EP58$w!N=0Rpl+PceAb
z_5S;<?^Vy!c^+8P(WNT@s3L<(@9D9D?z<&5=Pz#_>z?a4odR}n0eVeWM*={gXh73`
zxX7&(MIKOl(R}~B(QH)NkJG?%EX%rVB~0w+x##03f})p~Xqljwf>OyF48>&jA#g9!
zTghvBS3jReY%rS|#=(IL(K3{@r`rCw<V&sb#XG3$(^y<?@$F<Se&1iBw03NzGaYr9
zr^|V?)U#baLoi|R2BGEqcX44r0fn+{>+xQu-klvrB(|1u+)Js3C5u7J_Sk><4ywKN
zE<L5@zWv9e)`zmS)?8l^+qNC{GC3)!0D_eL;L!mhNQ0d`V=1kz#a{dK$==}QgvL6C
zmQxi8kr~@!{--l^px`-7uBYTq-mvys!1E}Q57Cz|#EiZ5RC95j2)4&~y<Qz_?;faO
z3{IXYyEu?a2kVWNTyq0M(t(M?G=vdV>#-E9TQ#6O#M89)k0V;t-jPwE{b;X)w1%wx
zSU)!<ErrW4UFS&K=h3wn8oBtsK8sjVrIu1Tj&X1nRR!bd9BIzA1Gi0@SZ7dM2GW>n
zt>JXhqukm7F?fTuk#)8;73mtT-g^TgiqNoO%XIO7{q5@+%CQ~n5R|g<$B(-q*&p8@
zuU|u56<vT@<9L|{tXZJh;kvH*y+QQE7Nk|=TADb+*lQ8V$L*Gy8zzS~dMD|ZrTOU(
zf*k>wsWfmloPz-{jFL-wepYGBj<G8-F$FZIqq4TvF<BcYPf}}3%V!Fwm)Wx(fdAKX
zaMDE(kuCA*YKgGdN>jl7_T0e#^u<%%o~aY&ZEw?QaOkJe^<2a`mYR!d*&h4(>u6hA
zn=^pgYs!is=w?sWl6Emvk9%SiJ2-<#lD1u(C6HQ;H`H>bad28o`uMRPRn2&g1jyDo
zXwO~z*@FU+)b#k<Eld+v#V+iEBPQ!D^~T)u(oU0wwm(y|HXj8qvn6Z=OdP1)9;Fm;
zmJJg-Z<tVlYDJ4^tudB0ub(&J)-3GJI?z^%^#rxYX&x8>RJF<wVg3Cs@ifIDT9%qS
z8gGzbzyEl|DX=P*<-X4^!8xPSMAk<)+FGq0jkDz9G+MoX9y&&E4G2i7+P8Jl;WSh4
zc)76T+*%XqYC?<_rFBsOBCMa^e_!W0KvM((1T-XSjZsl-tC_FP2S=b~E8wT|B^bhX
zJZi~ldHOHjjYjkG@x%Q2;_J55*SEmh>w5qA<QTl<rABYGr1JX@X9>E<_Mo!^kgD^@
zRC{RxP%AO?l1n<&GI>{0g43B<{#Y#5&-{8}m6meGX<+Zfo+fWt+x`9hxcuy`se6$a
zu%Z0pPe{<fL_%vFwFYi<a+KU^dUWwd%k$Pv%3J^P?RA7!js{p7N+m->oGp?ul$Iau
z<;>7)CPC<;pwPBQvMQ2)-(xh&W;9mSa%)OhOYcB&{1Oco_+;#BJq};@Wz8_3&(Wf+
zk6n#3&O4JM^Ei&tVDDX<lm?xwAEI~EbM6^}e>ralJ7W!MZ^&A=k~%=6-n1l=t|nf&
z<<zC+20gD;iNatojE=~kU908CcL-x-H+t9~x5bREnQs#r0wh3QpDC3ZJ)x%Cvz1KK
z>$ga~mAovOp;e(bL#1Osmit3vyWMWBgFDS3gkh+qB*G4C&r&OaW?^G+D|@QME$33y
zQAe*^mid%+ip0c(1b`s9?db=XhlFYVGI-~)G<IQ}xV_IWay@h1SLtOroYinPHEoH<
zOK?MEs8z^ObM$O+KBe-XJ8?ON6J+0CE~8T-qV3tFs&kQvyP&Ey+v6^A04d!#*K6xZ
zdRLJcddc3EIL~8nmZ^0j)ZTl|Nm@f=t>?loBWf4uxk2EZw}xsYy+2<6?1+tXXyVI|
z7Z6n`TV*@7gsez3*t|ar+D-=fd;t)J-ny7z@$|KH0kqaKVQ-~FJAyNmj=j{@MND8w
zONj1IuiLXW8+^K;ojiAK(4U`qIA2{dUoS+hwkAqZJCz@*iq=5TlaJoE?^`a=jxCke
zb~|6fkDDJ%TMI{mZcbzv>Yf*J!}NmE^79a8tw_C_XqsxPV!h?8&(F-~Da5%iV$P?&
ze;!ZmPHlh=2wmEeV0)bY<#Zl`qu!fHSw0w0&7Zwx0W%CgQ7fg51g-U^#*bqd0;=|0
zrPX{K*`9)Y|NAfHGLG|LH0-+3U=hF$teV>}U@y-b4qDjxaYQ~=AnP5Y)>29XV@SpN
z69PrY><mGNnrl;I*hP@EmbDx=*{T~!IvQiJt&iegr@=_7rHG2M9Baw7^`>?7tdH+E
zJ%5?bOm)9MKVjE6pB*8VPQ*mea;{}daUO?JLBUOP)Uq!h&!Y-^MkH_p)YQhWV|&~?
zMb|xUkTUU~{sUvL3IN?-%@{1*O=rWAL=jX$TS=v<l0OYh(96F3?N_$7lv)KvT^w90
zz^oOlb7WozXQ0<E$I^;8k1er3w$H{sl!FW**%$^NxG!qn&<@c%Bh5J0e(=5SS(=|E
zbv1YtH<;_cUcSr@5z6E6(MY;~O8Dtpv^`5z8^!?n{Zo;V%uLSPntp3x^q7v^dfIcn
z@1w<6yQVhk2B5ZlE`|UU0Ik6WiBQapPOS;&vy<&va0~_%NE!{^F)BG@30sHKl?@Z-
zz1G?gO*}b3ga-Aw+At}QbB>^Fzx<%nd^Ot5Ft7XXr;Fpf9rofNWn)hhy<DCd-Qapk
z&(CKMuV3CyzOK&&**IjB-bHo&ou_XvmO!QEW7KvWaGJF)pW>W9U(QiJ|H~Rx8XJ73
zquI%|M<zE}5&na|!QPb6;Q#QIC<G+ZT6^nC%mhFLEp4fkbblh(k~<pXJxb}rpuLsc
z*^gs%e+1iDYi??+L6P2jIkvrHT0Rzbc6`0g4$`u<^CdVYRF$TH$e2>y?+K?X<|f`L
zpXPB0mI?B9+x&QnM$)qT%Xy-<AGxU?yn*`pnZd)_tbtyS^|$-Wi#MbqBE7egs<6Rc
z&B*P@(@Z_@iBFU3$B}#A`P-W_RI?qXaR`tfOARNFy_Ms!`(dzzBHDY;&!y5Boi*CK
zDg`eMkn-ccpT3y<{pJR5IG#N9a(^frLn?xCwEgxl!{q9=PJf<FuSZrpM%KDL9=jqc
zpdZc=?myO+$~+k@=qw^Jw<_$y;8@#w-*c(p-d;z~<tU<x$Y$t2-X9s}*MaVLGo2?)
z_wTpUpQkX4o>l6xJQg$XVveUYqAt>N&bh|R92uZ4TM<<yG^gjC5Zw7Z2g6j#A97AA
ztrcTHdKYbk<S6H2ECYf9deb|@x*)UWRjEAg)qo@Hs;#E=$L;#((E-il)QvSr-Usi@
zANCwdgZqE^{@?$3V(8Hyesry-WzFaT4ppVE+RMRt2h)n^gzbkPqBp485kYdwX@_A9
zR7*+D6B*CeMjF?(%iw-`_1Kf~=Xun!B|Xp6XtC8}^D7x_Eo}{qX~+Bf=KkUtYu>g4
z-Dxz8D984E@>TPGprEmqwD-ClFRxU~kyA!vna%nc$4H_4`*<20*Wd2;bRA&*_~mzV
z`S!}_DczTU`%nI7OCWjqd;an=N8^0(Sblu&*VhR1{CGZEtF0^fnL(NIb~_;c{Q6i+
zJO@7i!#G*2`TVkqcX6I9N#`mUP=U<fZ_c$pfMP`1kK-OEUmm~h!{r*uT5TV<y@;%4
zVr$(QKtV;Nm7E2PWytF$!(iIJ>~W43YWn<ce>#~O>|2a~fYma*raT&ITV`t8V=28J
z85o0SE?WgYPs3;t3^M`&v-1X#2)pbbov4*mQB{$xbuo|%u^t-O#j&pM+xhG*0!T;f
zGHE%MW}UHv@d^cD445Cc@%-g4&N3od*_+j^+>brkfA}^yLaAG-Hs(F&{dPAmFGHM8
zUw+@^nW5D3@fqU3)nN)DxcGI#&z4qYKRJ+{A*~<#DCCE8FUOIOVz|rR#mTY7Eppy>
zSyG=l%Z%UuEvH@$){^1BdHw21dJQ&N@0>SSa$W!W&X>Qu4&nM0kL9rusJ7l}W7~^h
zr8xL`Io|DXx{MA5(Ov+6*c#NHiwu!K(c%|-zF0er!JxLZJ)WKBQMb?j_VZ5xkr49o
z>yNe3^g4R)y*CD+tAc{os#^BPQr$TK=<-BkOz^P%=%@3<Kx{1{YX@d3dnzs8>uH32
zb@Sj_12Y6?plr)-V;J8q-jMcIQp#m%U%y00EiKEQORXJk=#LhrDcJpa^wVq&ThHWN
z-uKMLFiCA)+L4`OXtj3q5nKO5Nte9u<Lk@G8#q39!;%&s9yf~TFK-jHTJyFwAW26`
zM{aJuT%tF+-0xcl0BFMS*oJ|X;>EOlq-MepJ@;eJy)^5G=$tnM((1Z(IWih2M>laa
zkot92VAgyb`?6%boG`D=|MU+t_fi@%=4Zw*cvEva_9Lmge3^&nrS1FP{1`E>YpPX^
z^Kr0!e;!3?J`LV;2jjgX&{j&`*0tN<E$a=xI3{D~(HTNzs(UGqjfnhtp1fh~UCMJ4
zGX#TxWW6ycrJ*r~w3q$+kDI()VtSa%DYoPH_56Bv{o_}D9iubS1sOXUXODH;j%wq1
z@>+p<CpV4G8Z6tg9c-Hnrzuj;Ilcd=9ERwKh){vJrsc8X+XQ}&rl!Z^k$Akkyo{E~
znp%-WfT*R?L<B-`BsEoK%cQD`(wcM~FM}s&>)(H0?HM=4=X!oSMFN6Wj@%Fd38brv
zfRE0C<f1}Egb3R9Y_RN0vy2RD{`xh-zHQIvn&Tz9t|5*NTWhMSJ$Glc7O;j9(9I)j
zZ!M**jPc_>pG<xB;d+6ztW|A{PFpv@5+eXX{`{q#0}+Z6mf9#dXSwCp0g)XmB2&$^
zbzwGc2w{xW^Y+{Oo+}!HFyLxl&OMj%y{j3AIL>EJ`TLKAA(;N|#%Q;tf`5@#VZc%w
zvNhTZSm+d3j`hb9BOkeg?96NkT#OM=b@Bwgq|&7ocbY@fY0llo(-^4UKGjX9c`yi_
zdb8JgZ2Rr@433rjxPLyuUP4LTPeTZV#9WScK8>E+wp9O*CO2E&KcTfa)CB%Ap#pjC
zz}7p**lJT0K|(@cZN2ONHXp%^bC+D#Zm$1fuq|yc4GvWW+Q<HSb)bOAY)~Gz1;4#{
z>Y^o;yk_qx(zX;FM>bI5_;Igf&lPJ?8ypfr`s{9q+C304VlTCo!tg^)7%b9!A`<j@
z8~rO-U+R%M^hb&3Yv6jT-3A{9Bdl9?)^cx%(tBEI9OF38R@;8;ou=#Lw5`jt7E6%U
zL*3=1&tHE>Yq)>j;@7blE#yxQRXI3fOx3$Te@T@sK^J8j0~rVD^8qG?h(w55xwdq8
zMzZk|9UVDd#<DEO;AcM&0{W{M>%8qPZO@#IpJs2N@7c#8a6b<3=NQ}{Y5sBO4aA;$
z-**780WN;s8<F9(ei+YKnh!5+{qcC5t+odK^a2fAD!t~Kv}|RXVjR3BM6w1tpa2s}
z2l_*<4Jy5rN3D-V*!!RM&!t^|ewj4Yu9}slY`5C(#W|`+qd3n7YtNem44~4}_W7Aw
zm`<nZ^tNq?6XN@_HhYSWRHTFVY*;%E{u*sr{`xUq=D}$#$98XhD{h#k)5%9#o{!xI
zAE1`*c{(x6?c;gtz8~AVSC}quVLSBftYISTE!VCs-*0sJa-CW7vRB&n>(zk=*P~}`
zd;9J8>z^D>9+>Ncn5k#;EXJyPy*SI7j{Vmk{(2s??-gw1p1_6>xa&kXM8{;Pww7|v
z=F2p(mR;JT&$Eq@a=1nh<ynh!)3EO}%!B84tb13AKJdQ5e4c~BVuYG1NdIGegQWZ7
zr}*_)*LBYoO>nlo7h$Wb*!@}Q<t0LXF77p;VKJbV5sWp46tzA+AGi55IkFyvttu#x
zGfXNXtyQ(2q~>io^7}{eUoTY3{N?3j`tzAe@0)h*&s*oyDPr25dmCT((V5{-r?vd<
zEJ+7E`^V;n(`5>fp3mgNmNs}hnO26Ec{JGb@_5hahtp_4^RIvV^C%+hP#{wnQSy&R
zNjGE4@z_jA^U$~E<LCqSE|7X5heQ=TmiE-^C9<|k44tr&M-xX&j?0nFE7tYzhZ%@s
z=w(MX5j)=g$J5*E7pG;d&Gn@$<{yXN+x~GFb_BfwZ~J~>>3J9D9W(Stw~ow<4~Dc$
zc`mItzIe}RTeJA<2`Aqr)7RH8p0Te7m^k{L%@Cuv+DmQwvtrGAM$C>Bg{;eKF4i!V
zTsj~Wc|695y@S(UjuMHXR}}Z+%;uLq#$ziffjZxF6<r((8W-lXhxK`wSMB@O`t&wB
zR4vO6#vANC=h7K#6=12Qv5i0P`(OVPUH|FlGe1ieoGp>|TI<W7(MQ*hM-DCqCpYVz
z8SPO>^Rahg^ZKXLG&)kzQIJ}$=V3Ov9xQC3msHA;jg-xL-^;hle6p?dUb0-R4_uEE
zhWZd5M_<;mX3JO?J5Ijtx2^X+e0h!1kU(XxX>ZPZ=zvUCsnc+ooaC~n8p6PBSxwE>
zqUQH4i~}`=o>ywC=bAIXzN+H?`_yvYe%z|?c$pl@ejpobNks&SP;XB=p5`$IgHn(E
z_5|ZitF`r<<MmHJ1(M#jg~l&m9253hinz%cYmG4^wbfcnYrRP?xd|EzwUpY3K%}N+
zao6+YaDRUPEI^i3ZM1)&<IfLziQLM53=sg32%Wk8ZlZ$c29`L^R^Okem$84Y3KUPT
zuMYCDBtnH=OUj*JCRsm9upEXEEG&Qf|9yP@5C7qU_1N0mm(i29k~{gp*lJDdk|<it
zxwYdEo=2-`TR-1d`PcvW)7iD{sI`~e^4?B!#FA21;@Zd#fl!_w#hkB%){siiIoEEi
z3IH2#u$QjG9BtaS-~N~W=ULej<o!_xTJ-eO&rTG2tH+vZPSY0;BJ2lK*F~dchGik|
zopB!Wwt+K@C=`Qj+j9Y9=*U8jdhfYnHP5bE3I&fmUPf;%v~A0^^zKi#*V@IN@(x;e
zBh>}f)7Wa+lCTTsmuZ?u1HIL>b(*{-#@5T@Nz>pB0f6-0TN6Mu29@}DF5Q-uhB0b7
zj=f>8VVtL6PeWOEJ#*U2X>bk!k;ceYOYLMn+Sj+>tsalxc8HT=492jw^pOg<ApiB=
z#;+6P=l;XMUWX|dFwQwpM8i^3QL@fR)23hkn}7P#KYZ~jUCB8}pP!rh7((^jQqC<z
zLImleJwK5rZ^m(+uOI92pYIdpM^d&3peA@@jh}kn_pJ#dI*$daI-^3W9UV=*HL1B&
zC1cnyGePUAl;gQMGrbHBb24`1*{1KmJnxgyc2u%<vUN42P*UxLn%9bo<Eb~#kQ+K|
z`=$(Lo}<^>(*5}AS?acK$ExhcGxmr1%@_mp<)4Fd#t<PA=A)$j@gR3`2({IkWelKw
ze~d27meTX1p2vBb=F$EUbtdEIR=HHby4*hQn+9jCF${mykhWa<a1Pc$&ihh?XX{*P
zt&_FJOtWp99=rL=zp(80*<0&OtNW3~c|q-2&Dh(GR8(&tZ~r{fejJ%yw4}${FR%Xi
zew)Vn*q%=?j2s$rn8tA&458%G6k4i1KlVL0WB(gJ&XWUT(qp;3e_vp@$kvPYR<aJ$
z9I54H5fl@V1W=L2V1L{`{wo4|Mi&{o8Vg8~og>M)=8xz8IJqV#6S*4#5$%X<1N2&|
zs`O(KU~kL19Dv>bceA&wJ#Wu41Ojy7`CsQ(tLflq{YaP6kph<ckIhZfpgBp8!CM2E
zKWG>yg$CYWE8qW9dHLyUH=rie^>f4d#k8`Sc>u5>#^^1osw&VhqW0R9eekAk-*2b$
zb%b7XRk95AfN>h<d2m>B2URqT2ms|M<yb#gAqzEQ9LRawyQn1S{qbR}bw*0w#;>P9
z<;W05ElPbm`o2||PqBTL*J*@WOWlv{{z&#TkKyDUV=ZaDZ-+Y99>;A_L^E8@#P!kD
z;M+Iq&-;%|FK5f%v`6be1f7)d4~_@&<NLdq+Wj=cR+3{TtdI8}i+>%R)Y9Ag_r~)X
zq4wHZH;4d(;F3sHq!sB^iNZV@@3cG?;tInM$5XKASm0Pay~f!Rh*+bltsI$S3}NW4
zmLqRRQL?riYnkT|0vwN0N;!UiUM|7e_#fJ;Cb&y)g2%23h-};?yjJpaVAR$H7-|)D
z{v4Q~9NWSS^7W|SmuUbX4&N?EMswWv-Te3GoSP^D0D=l2qP2zqq3pdXi}YG*JG?cX
zv?HXl$+y?h8$ys)jw3h5o--IjJ+E8kQ^elDc&i<3z;-NsnEUU)9BRWf&}p#Tj+*v8
zK^%Q>SX3LH2Fo==JP%Zk-+vS|s%CUC3{$kMy^3<gXiWdNlo}3$h4zq0+@95%5Tens
zi@#p2F<1(&#}8|)VP~!P2HNrX{{FOx%<LmElR=f<@b%j`P}xeb_RBwrI}IfHx4)Wi
zXN#a9$L-eqX&O-Cz}V0mfm47^xe=u0H+#J-!CBI3PXeNL7>w2mnk#6NzJ8xyJ(Jc7
z4ku$->t>@EgWE^ZoOp<57@5&tU$3q|J|F8eTax_D$ehx6K7HlB^&wdAQJa7)AK&v7
zd~i&KdwCubuU}eMhxZ!{Uthmq%juZcoB8Q0f^<MaM3FA4$b`^a$+>|shNkwseTM64
z@Ro@HX@<d&WDyk=HMKsEecc+J&YpM{;$BL3ru9@@bVk=te&M!1#w)3;Cd_}-L8?4g
z|MHi~YfaTg?+wuj5!<ocfB)|MqngWQV8in09om|hX=K(V)75*nfsW@d=wWnL@^^i`
zdJ}8E@7|B&<fI(yk((R5Wixo_Y1_x2u-uQ!m)DbJW|ErmJcmF%la9_gLhrfn+r2LV
zj{9^rT8f*)`DOB`E&cg(sfIyy2OHdr^lLik6saH2W*m4^_Is)@u-0-s4(AB^_WsCI
z3{KI{^B62k*|%eRe+WP@-k1>l1TGB0xDf3a<!GG|&FG1hgtYA?Wox}QvgEbaRu6H(
zSPz}Y_`0u0)6Og{XYJef)jDoR6*HS(VH|?@mK0T0L`6YCrPZ8k2Q7t9$QfHF#@1Wg
zQ}3S`EP~dOvncev3Yy>|r|uoLBPEJ*5D7uU*4BfU`b=NmUNEogvcz$WKAcD%ANRf*
zkA2%+XeCCr(0fi2|K*>``nUc3GGI@&Rz*gd*cfO)1E4#F+}Ro%w(O=kmin_bBvqv-
zS4Hg(>(TgR5hNj3OCmIn6V;H1;EXc}nz!Y!9-8(lgpdzWwy3q_)BleAazAq9T-<y*
z&CweM_=lWxDS26N93x2UfWTb$<+<GF3$>npJpHxH(EGkrdzy^wW(4QQU|WN5o-8rQ
zRO`wp{xW%zwA~*aijJl)iytX`n|n{k+KnM?sT{k<$H5b6Fo#IZ`?hBqoiXWAdK{e_
zt}U%cX)U+&pRlZ-pL)_QchC6$JcST^@XlIGgh;BrOK&1Q?MK(6jH8w1vDz0Wy?2#f
za_MMX2=&PMdAKl6(-=tevE<VU>QQ=}p(q$;((=4Lcx>tVHQHXc<@wl#Z<lch2Ga9$
z?PQ@heEGtvfKr7K*jn5_{(ZfkZO^%Mbr$UqL9|too*#>~oyI8!hte9eaC9n)sLBSk
zrDvvg&^V2RNX&+bNV_6PTDN{a1#7LTsY4tXdY777ExC6Uv@tqwds@~bR|uYvtRKc0
ztTUu_e?IXvdV<~+0DAAoa-?;SQS!2E-Jjo16ZZ9R<8?A^TUI*<r@g5Oqh}%z6)Dfh
zD)DrloaWE>@9+0^Jvpfy*jS@AJ#TUe{r;gIRom9b7<@FEw$J5s#$JjHr~Ksk^~Dh&
zYb{LxIC#yUzyI21*P8Uh3|?~vK(^jH!-U%RZ7*W19?ODZo++oc-+sxXCGUxyvsTM?
zq*DDD`wPZto(7iN_r1<P&k;n?7?#>a%krqwFF$#wVm)IwB1Ftk_qx@l;B6NKL>{P>
z(h!Yx-k|hG=YM+U&5;@wMTE920?slBIn1BU%>mg)Z!|w1Nvt93ottKd(DMd6!mLm~
zyBIq}Mg-}lwchr}ugmNc%`cZ=$0>S?(xurobtdaAl5@sc4P#F0$3C32)wD0pFf$=&
zZ+$Xt6C^g&%28TZKmBBwdh2D+Qp@KZz!*|sYZ1+G4b+;F`=eM~dMOIlX>m$6TAg3N
zeZD`@jod_QqF`ChV?Vjv*7yeL`#wyg0GB{$zbBQ7WC_E`*0rbOMjS%F37<yd0E))j
znzMu#*Bui`U-AF@ibxEK0BD$1rSI$Byk~^o)p&cX_eAB9$ao*LWf|vaAumVsqj$z%
zPLU-okA=t6DH;GahCn23kN3ZQn)A0W7YDr^xl219n@H`V;B4#Gd(1RO%do9oIymnb
z5JZd(4)XoEH)O-qTHSwt`16;uN7bJ8qu4k`YfNu`EZb3#px4~ypH4$Cb-O1wMye?v
z>u--Q-%egyuT4Qjd#!kU<PgI&#b{lO21WH)g_Wk+kti^us!H7wqUax%P?x?P0BD13
z-3+5O*7Vi|*)jn-tD-IEwA_wX<TQwm(-16S7`$O(V~s%+X*KPw_Pn$qdV-o$>0N47
zcHS9jfS{6YI}xtW(;KEA4-=<pfRszFUAmG9-t>&rbC*IdXKS3n-hiBuavV8VG|=fV
z-<!1|4j!a+ZMD^|VoyCUV#mR$pn5Y2F@h+F^!&N@E&FKCX2~TEPFLBnwLZL9hPIM5
ze;670@%es}X+BSsIwLaH+)C#_z4cuEWEhE#JNPDX7{93G{Sbh<t)0JK21j~LS=d;j
znvR=>uisuKYmId2Y2TX<7W-b3p1wpQtsKj;rvK{VJofds@&(CYaNcE@qN7%F6*54U
zOQX|e+tGy}_tb3g-1e=ak8HMm@~Q|#%*GhD$F|`#MS~)}`U6pWdw%*?#~`B7Fp)@a
zDXB`UX+4?}I*0PC?HwtM7Hu5I!5cy}h|HuN$@=M9kMz8K98H0UiRux5y4Zl*dKA>w
z*AA`MT~o@rk#$Uc{qdM)ZRlf&Sf95f=U0Q$5Sz&Ud|%$aI>_5nI2fsQHGluk#<(dO
zBoOJn2^h=uwxJ(}&`LO+XG_2qd#j}={H3(ky4c7)ZTFk?yJx21G<nU>kKd1<-YikC
ztsRH?bBrE)O2llGKqz{pUIbL*6dhwbE(L(XCEAh>+Try|_up^DpS@HFmp_aV=q>O6
z@v!rGu!gm_l8&Rh;IyW^AG+f-8EvJc&)=V)Hzv@*%3%z(<uU(!9{;FwQ3Vkw%Q{U%
z+``*LEgz|t1n~Tx%`lB2-QwU$+55n32t<47y&y{~rDib}$|DV4D|(h(5VX|>+Ut5K
zY26FBVF-}F-^Y^$fxTrkK;uX%0!S`;=PZJ#_MDFIzx5bGuz*$!=KIQYECqu1T5>tk
zwpz`6IXTl>7hpHM1OtHjD988T!dDyTXo;kk1j1(j4cnGE#^77oZuoz4Ig2Py^WY5v
z8zgU0ni2xYu{E5h5WKUfy?lO}X$&Z;K-QVE{Qje&H@zu4LrmN4m*?A;X^Mu3$vGn(
zNmY7lommv2<&;us%CBF>!BMZJb+p0Pl)5wEP9yK{|Ni&+YcOEu!5d~`5|wuBJDo?v
zSdJqrfU1bHF_M4$IOyw(w~Re)$NJ01csUIgglW6|SpVUlCsR|VU{HFisgMbd>Yl6h
z2(9k-A9)&W-ilr>AmA<&hcO1rjHUEeI%sb-9kr`!%~{=GfSuu*_U^C)wvy802*Y4i
z1lYN@t>w#?XqmL-Bi+~e6dXZUP$3h7rP7d$mfL$X!AfU0I#H=5cOQ+hkhVimYJ(7s
zL9If}hDB@$j{E+&FYE1L#vzX55S?Y|tz*0WxUa>1xfpFJJ)h>kJ(t|uR;PKKMrW82
z5JXU#AgJ{9@pu<|Jx5ED-+!g?^`rtse+b8Xt->lG<(V@XfuH}S(r_MPFxY$EghKQT
z($k)j(l~HGQrh><^Ed^Iok$dkS~~1B286csrY9-4d5$qyA|ggWQElB=$0U7QQ@5s+
zl#)G9-gzFD=Fxc1(9UJuJ_Xr%R&DFA?q7d?^N^(~8SDz}w*CBdv|1kj`F}nC>ww0)
z_Sg67t^*dFrf55iV+h`0skH%F8yH&YB0|*j?fd<hhA@SmTW;NYODa7*9&7Ily<VKA
zBO?KhUqdd0sM_}fzsz$ChEZBN)G%}iE;!bF{hD)1Td}?vay~kJDn_ie+P}<^WqBM8
z>vK2WdA8A_^jr*7bD8S5B)zunvB=xkm&le%+wO^9$4Ic>I@{6Q(NEa<=xb{|7{h<y
z-sLztv58Tw(P#cWj`M(R=W_3dfm&-y73Q-=)e1HSOVW>{=yhf@I5L4GJwNyBYm6Rh
z-cx!!!8N{29$Tr3Y^^0|`+A5HXld!8{b&Z1ETU01mbGgqB4dcKrH2e7NN;?;63UUr
zjy5o<oKGqp5Q}O}`xgGOyLj^4bDRHlemxDugoq#|9U1G7-Y$cY(t%wxmSlk6GgSX2
zj)t96MJjb`wd5^Nr$2V4)(2-f-M$~=phr^+puh&Dq+?r(xGzpWQXHK+yW}dyMmUbr
zn~*bx<^8+!hOn0gF0d*%?+gL3AKJ%0$9aI3AxvXie|&VS=BF=TPPV7z{{FFLy<8?o
zy8W<I@Hwr@@$?5fvab8*zaPW-?6u|3-~aX=FQf09Hirzx{HK7>rFJyIIq!q>jtL~E
z{c+0_2WIEK<ooY+{&{BfM(Q$hS@)uLe0hru+C|ZX;2E`+qmb@L9MNDi9;VT?T(;-h
zdO8L^etVyPzQmJ76-AQX({EddPD(y*PX#3(yy2E_kJHN(5hU$qnn&;8`0Yos=ZNgb
zdGsLe1PIz%ZH<^2Yu=Wuy$Qbl2Wy8w(raT!2&l-`p;TjxxBooX{O-qKzaXV<Ev2Q!
zFuk3l#a`;ODNCuXR)Cy7thWq`?1y3WrraO;CDvv6{bre|t0V2{d2=t<*~ql}FgiRQ
zpX>NG$DYl)VPG;`w&yWU4qHlKtRW^yS;%`I;yCb&@z;|lZQaKj&^8(_uh+?H0(&y^
zmTjB@Yx}4i<_PI=uwyCK8H(pU<i`kG#+D8-P97@=a)^3p!T)1;e0H8^?<_ML<Bai+
zLEG{E%SJ>l{QPyG^y9Dc<?N*+9)}<kPG2r#j7+U*E2RNj8>7*DthZm<+w1J<`MJww
z)Vvok-cqZrWBq8aZ<jfe^wx4p`~8Q$ovbt<LRAGataW>Qf6sIZwQ0*d1`Dbxe?%O^
zFilfLa+ai)rqsm`mZjIUR~2#5JA;VE2Wy>!)>`XLK)R@cYTG{V*-l>bhBl{~x4lT1
zFQchRp{AnViUKG|I`-l$8*gA=&|AycD0<WP_3+ae)LbSk$K%oLJOw(QKkCcbNGfu=
zT1fkPbhd3TJ&w#SSgh&Tvm&75+9(X3@~&<eqc_%vpm|%C!+A0>dTUT>6HuvDfI(Vq
zVuQ1qABVA+H}uw8)62T&+Q>RDt>x!qJ!)&=j8&>hWZ&X^o*j{K2*iX$0NTBc6-E5`
zb{!r5%YU~OrJ8%b-`ns{Kc8P-CP$=F8$hqMHNfBvk3Eg{A&+9=erI1m)s5pcSSTr1
zDe>T5jt(tvpSR^`6l2*3KgJlW*5lY7_s$>_F3(?D7{+tdRF=p6z(0MP|9Csgx&T72
zC6|Wt)zR33V=zNR0x-C*U4dO3LLg%zKyS6yR$CWg+&=Dk7|iyGr^}R|zn9ZDX!-bE
zO<p_3K-LCBL{JlJ)nN`s^Lqaf)2g*>har0NV*C&T+hC+{FIh90dAiQ`vQ6XbG&pK$
z)kfAD=Zza@XGBGMt4)!ShyaisNj*MqM?POsNvova{wthLug)N}`(5HV7*;{uj@HVN
z{Aq?>*PX}NwN2Bpo9Q%M{}I!+9oYuxQpah1{5YzJftt-@8>aJV3V+lUk#&YZdhgv2
zZvfI$DVudX4tY(rmF4-dogyjrjMZ62gf6|7bkvR<qrqO;?8gJ46mE%3IM30@QP<_D
zfMSOL$1alAAHRR>;pJpoGM;2#d&|eRe*eXOo1Ga7JoC?|^XRGOG+iIbE3oqxwDn_G
zJ+{39+nf!>Az09+NCb#bWSp#_dINUBdS?^>$RgC`-uiOHw?Osyl!lDfh09yyx;+cF
zydLv;2!WxM)`_{_wt4ngmZy9PE+45`XI&h}F)|>+_Uqb^r2v(@-J+wRwm(3V)|z<8
zHD~*J9fAXuURz~voMmGOrPU^gFbDhBk9dmI^Rh2bH@sX1N7%J~U;Nuo6F_g}SZi;!
z*X$j%+w%1^axGAHwN5}P+d-w3+Ul~Vl9taNO>dGD{$B<W@5zp1@XnA6Od!%l6b+-4
zj~|7ulQv;%prw7!mDyT949+-9*juZ))csleQ7$jRNKKsxL|ZK_4la(KiL@Te?Fq&?
zXR+5K8{3w@ehk-ZBp<M~R=bF*iZDw_={Pb>(-<8pAOdM?9oUc*q+YU^FpmN9vQ`{M
zgHlqit+&qB4cBwDgaFX1K=1&7Y>npI``Ul{(>e6C-ET>4i01jkaJ>YJNFcd@F_t84
z%a&59CF8d%h&;Z_FgU44pF&l~UDP@QW&hax%g-lgjrNkB*$qSV2J5k|jm{Ibx~Jjm
zc^V@DsI*)=fC4i9Z`FYXP2Yd}_~|Fk%U(+Lm*|7H)M{DppAY{sctb63%?D>0l^EKR
zq7#9-mrBN4X&nvZ)}^cd;lP?<=2FVO;Q!|;Mcdv_mpK|DYYY(rsCGb*ULUvh5CD+)
za+)kjZ>2O*00DV?uKnd=ptqKjAcD0FT{uh*Yi>2yar%S77c$yQTK7E_^3-zmQ^eb&
zf+IAx9L6dd??~C?=lzkIi-V~}{HWDW%ogjB#0Mv-h&4t0IC-f1kxlTHR20g36kuxs
z*m^@YSm?DWv$T{NT1M%@-@cfhtF(-!cQjCLn;qxNb@Gh8<=WD^Rq1tme;0qfcubXg
zt`Y~5)F)d~N&C^SUkC1u-1+4tB3o&#q~->vDK({DmNHC^vFF^!>*ZvLP)aHSs=e2B
zCu8M!EdF)0dEbv*3ZLh391OOcN|&w#y?10|Soc1kPlJV)k7X+z+g{WT(KD1IRWy!z
zS3u(&)vRPK_FT&Tj{n0ev$nk;VLXq)S!0+HrMK2gD$;sA-uKZ9kEb~p1nnv!JumzE
z>tAo>>$kz7bW!6iBN@#5CeB-DjbTJ-t@l<_6CVRDAHOfgxnYX#SUZB&4y`n)eVh@U
z=`{<cO_<1;UiM01a5i{pT}tWThSB4;k(Ji+Gn0?gXi>U?vNgySrMDwnM_Pg4k1vL$
zmXu2Cz}AlgOWq#+G=_1ay65QfwkDX~Ue1<U*_LH_Hb0Jd{O#W*M4L+!2vghQ#XwF4
zoUR&Um`CVDXrsm6fD9oyXS5y5p8YBE_N-#Z*_GRS{CXJ?SaWUy@f3`tA0N4cYRw$I
zrSkm);xvuXTZX=`_xy5ZLRB|};EVxj!rohyyncS(^Ay_p(D{6_*lP!4NJ`3`;^0UW
z{t&{ow@*_NA*xDx-1w(2ucuj6x@xQC*h*IrWu8r64%?x7qqT@Ly_Mzt_H6D=?a^4J
zcOmZG52GvhUsg9?E~kJ5$NJn#2kl3iuLJL^IOCji7PWP-+`Gm%miwpdy;<C39v!RL
z5TGm>;xvR9N%q`@0&A_++Q_OMYAM~HuA{3d<xW;M9iz?LvI)Oj9F*1-!(a(j6bzLk
z3kPQ`7_3KeQ*fer6*FJzv8mPyQ0uWh>dz<3RM%=6dTt#g*IZiJqhadZ$0?p7rLCdo
zoXWnM6YJ#-MJ0)qy{P`qr1iG?%a@_a7<75IPEdulsu@N{G#;CsPJ^j?=@9<$pPaXv
z_s=FG&VP+Bv*z`_j1ha4%+~d`Z2&;xCe|4AIO9B6#NHGDw5w=Wl&sPbpf|KGqMFq)
zsz56+*58-U^X!c?prsz)|5gUa3W_XymQXi7_9_Km&Q$m1k(A73Fl@24miF8g%eMaU
z8`f#E*q`NC_O1d{cW~{#;-1++)9v4~4D&o0ERP>H%;?4=XV|K6?L6}BeK>iZNAhDZ
zhI_H$D9U6UIn5Lh3`r}(#u|1$=G3$-`Htl{gvMDK+W|zx5JK;09b&pkBF}20!+kwO
zpU+lv8&vFzS77TwdK)f(!pZex&3U>Ic?(jG?QySY>=3o))HK**$-n*n`X{hje!usZ
zFaWOry>56hh}z4M*~oaFnJ!Z_T52FeB2w1xzg6}j;(PsbJfBCBu1L`vt!dk~brW~K
zI?}xDsrmZ%Qc6{h(O}C-3>#wO*@o%QabS_&nn-QZZ4f{t{Ug7vEjLkL%-`3`)fs0H
zv23Y5KY+Y5So8Y(Uy}>g^Ik*rHkiEZw?~-9m^w)=)<ACAqH$0<``nK4Y`}6E-E^7<
zX#18LM&r02YfVp8&wF*z8>;}u#b6zIhP+hgPg6OLTxxmltJ9P{s%Xn=2?)wY>Tyso
zL$ug(t;Uiwq}ppqWSA8Dx))J*p7MAJ-dSSOT2f9#y@5Xk>9Ka!Bu_U+a*)&V%sQO5
z%<lB^dUA%Oi{#R^m+k(U_~p$L6SQMBXefFcg0`}JtoyUHc6e9{N`c8wr0{rlWJvN6
zoB>74#>H`T>`#$OK5|#*Nn6_Kb=gvVbTe{#U!$i|!pWFkb16bLDAmuFmd8_1wjLY0
z)}70DsQD=Iq>1c!0tBSonXUB*(z^84KotQ|rMI$!*7tYs%>BMe^v)90+FLE9DqD;B
z`LSD%DEEEqm#dU2gn0hTI62++R8t4&t(GH6uHwf2d?ptR!2h47KihI7NwzFOns^Wq
z0IF(c?s4;GX4R_IZ~gyYQxDyhcgP#z?q;eAKty<;p&vB*53m-%0wLma_6GKOx4wX?
zrI+s%bKgHs9cM6#LAkxGB~e5Gq64k`c>Kh&6viaA_Z&Vjz0lLOwN5Jyr)`7Ttup3C
z0)4cgBsph>@IIJ7J^&T#?07A_Zi3)*Z|89iV8q0Vp9MUkrbTcZf64b-Ve)DH7^$pt
zfB$^DR6zhg_T%i+#wlA;62O#22w}qZ<63?8{rYGwObqXz$oP86bALR77gw9wKcQ?(
z;_tU2LI~0Su5+5(^PqaUEV5i!CophLVgCGM&f~0?%bM6BVc=@>9HZ^`2Vx$vun?4N
z^EjXM_Oilw49c-Tb@T~&z0@pBh~N>3fZ?O>`*AkliRWE@Cwab4Gd1@JynV}53I|}f
z+m`0(X~WA>{rQ}i-+%j7*)&xmawZHnpZh<~Zzblu{Lhyw+J5>xW1%sJ(|UVZ`d|O{
zj;37DpY-zm`?X5SNzXCo=ql`3&_p<~oQ8EHFeAL*mtwJ{<Q+4ID&l^g$7_a&*^j~X
za&elioy|j?(@T!_h!NE9oWvfFuGaf<6RJu3-T3<BnuIat^B<qvO>$jtQ_VRg9OK{r
zBQZ)!4Bnm}b6cu^JZln?Li71P$i_>_%nXj->bhi_`~UM_-+t7RV*leG2j&9dKse76
zmhF1I)P(Lf_V>plT{FLp)|+#wUjt3fXy*7F`{C*F_g!DUe|xEk7#YBanY_rl){K20
zK6Un?X}#IsDT#2~Rr}E`AeB^>B?|$92ZZ<YIG#<_s0wM9QDZ)zQwLo#jB%80T~;(S
z>v^gCRb>RS<MZ9;&W|uaSvCYFLhr|OzFsQMK3?9gg^}U@d<<}B8Ar&=+U`$W$q|T4
zdAs4`CnlI?M5Fas*7fqnLIg0*J|-msT9)!UBzXt{AmJS5KHGrVBf$IkyvsHlxy?B>
z=<QohNkXUkXdfRTOB(dIYxZG*5RggF9RUbD>^yVk0gSXtCLdkfy#=vPU?j<<{O)e&
z@Rw@>kM=dRru6t<IW5VcUSBVTeaHX){o8l(e!@343Ueqm3ykg}#6naPxQ!p@$yEq_
zJbxYIBVS$@4!{`d#JqTQr>Xe4Kdw1_Y1&ZFyKV)C&Ninja7o*SsL$v1$IEK%J{{|I
z6$}W>B_)Q480XaEu(-Vmg{uV){6BLpI$aHGNf>>sFP8<tEhv{#N@9XBbhdt~*Tu#d
z(?ytsB`my857$2C_U&4Q&d)zTV;=HySzvrVrcGSdmE63pYtr*^2GMN!<+g%a<ho>X
zH82Z(?&CZ?Z?~5Vc!cqi3CX6p*_b+&S23NvH!a2c1FmaE0IExs2tWes&yPE@(%biM
z*Bbr)`8b=sTyZ@2xqV;7=g}mUoEQ-$*%*D^|MR|n|J#+pb+m!`vXG5*x!$%CI!0IP
zXKNpy*S9L1*KJLNq?Qtf&vSSLNqmKIz&wbpP3>B@l9QyE7LLN;qwU9;Z;P0k&8ZGH
z9DKZg5Y6*|>z0M*3`()jl3d5xx3`R+?d4WUo$I!3YmNClrW%(l!tOdwWd?)@9sAzt
z?N)O_(|P`soEVr8)IdE8bQEjoBg%Ezil4tmDFmj0$qj{psekUuxg>T)Vi;Ww*58(7
zy~p$gnylN)n!&!eHi3yb7liGCcHB{P4*pi_b`esW`k^+O1#xB<Vei(@MCdx2UoTZ)
z4nWsA&w-@Ou5^82vHM}N+$zO+gdN~mOR~Nn>0dW~oEhB4+~0paj&2{zC0V&-qD(I7
zIMggjD8T4QSZ;d0ujR6ou(sciy}oQOwZO-?l)MtUkNrHKtu8UopZ{OKB;&eNvh$oy
zQW6ug)I<OQu7OAiwZH$`E=w(-bDrb&Qp@J!`T6{5h-m7r=Y2@6X!|3Tlo!5Tvc&m#
ze@Lwr+vgEfwwIN!$N89<^JU9;=tw9T1psBM3G{sb_0v+GPpFyqdrN6^SMYKC`JeKv
z8v`T`Q!WzqH2~-Ar^-I(Iv<bz`j#kc9*_HDePc?r;L-<Wv_R`eGi{T8e=RKHp6ja(
z=PY*Y$Asl|&4NTsvMr$P_T%~M6T+VNKjr0u-p^w=5(dJjb1wPE@$u{qJ{^L_vk}ZG
z2|PkagmcX)%kkNi(!XrCDxv#c$otbSw+;J@a$Rur4h+dBUy44Ttu0$7JAUz+gt3qv
zqh_+vKku;q_V(@kZ2B?BR%DU9Kl-14pXFBIalZYyT`J9~kbY;zagOtUpKlkTJ7NT5
z_oW)coK%;!W+oJ$>OjOIjM~E)K_9>VO}@WKjNTvj0ZR#YdwzZ%P%i6o{YHpLoO41K
zI6m+Dw-*E*cD=6N;yeehYi0Q3o-Rvvcg(diEti_W3VFImrK4+ImW9%mqU-3-`+mya
zMkh!NfW!<b+wAABr{ow<-`f*1W6ar(rnTyW(ppkZ(T$kF`)>A9V~)$^wiGr6L~xt^
zd_4Q;b_Ej2FpqPtFI$mqy<R^){`#qsivmm8Mpt#le)QI1D{wx%`gt6$uWKUT9hpTE
z_rURtjTLmn`>)3T^0sbUVee<jIbHJSAAkLJ;gsWX(3G70)AROwK|dcKXRgvdKb}Dr
zzx|Nd_m<yosXsnGJn6io*N;E{%GX=XVZEKl^JwYib-921N4;EbS+SH{v#UD&%l9z@
z`}2NYzFlt1QZqu#Hnbm2oC<?$KictmzFxdrd41V%68^9M`th5%%9iqY9yu(=Uwymn
z$9esRESWJdriN+M8_mAgbX_FZTIc?AUta2xVa!2r?oQY1QZti~1x4_38MbU!%(>)@
zKF8N}bODsNG!Ntnie)^_b4Azo*@`T6e#Ynh<@e||+6)hbEd!*aHmy%bA&m6;a=k9B
zyw)P9>S3zm{PhRjzAf6#ZJ#N$Js!Axe_K-oCAw^N%fc96?EYb#!ma;$KMtWvGM)<k
z93(MAE^8K%(2^OJTh{hy{5O&{gDUGdKA#!WniBS|W~NrJuiuvV`0Ks=?b}U2u2ogL
zmA4DPdG^1A&_CznvQoVMMzQ~$2*-ZR<t9j|4s>}F?Po9l^8VTPKk|Cz!W82e{NsBW
zzkYr84y7jV*#m^kY2I7cIil9YG25?SN{#2R^|$9LV1D^dpVMSZDUZL$iYQT*4O5FP
z|NOYTZiOX>>1++Z|0ZQA0x=%<@SGMa>!lLG?QdqE58m=t1YHjWOx%vgKBt$<?N%r}
zEG1EQ*Kv3Je=n^`1|J@Jo+?W%glbNeoaH=Te`M%`QjX)}@$xqgp*hsS_1q_1Zz9|E
zS|LJJr#oO`xAFOY`=7U@{dsgp7ACOcbh)iI`UEcIy$w$pi9|$*kYe2b`Z=#FfzJud
z;gJOa5SfVx=JA-aY;o>wy~vkpHq87y3`;VD!1mM#di(MHg6-q8mDe>VAw)Zl2DxT}
z_BeXpF1R0a>h;HxMc@DYXi(Sazz`lNWmyZ&zW1d(9)MB5eY<Uo#ON-sSJwUUbcv|N
zY|b&=-DAvJ=jiHbD`<1f%SJvdWpP!PENJr>W(N3p@9_P{n#ud~Q`7Z28`GRrFKly+
z8Tt0|daViO{`u@s*CjKto%=a0w6zXz*POHOkF(y^96{i{MJ80wOOiwe5T^b73;$=H
zGl^q{vzz+uN6iegw>J5*EkbTKHF7C>zwbut6518ZTF~12$L)Jgb$eN9w&U^hSBRAB
zN(45STSCNe2O?&oXpg_{>!s3w>}@}ebK>hIXxpmLXy*Xm_m0bDEfU_lqYx3E_vaai
zyks|1#RO)KU-J^2UCSlM&%b6M0Ve|0>2)jA<Fc&`5tN9+y9q-oF<Uc8rDQ@scFOGp
zIQOG_y{?6jaI`)#CP)&9Il)g&Wx0O4F6754!cv5gmc=z^J@1E&!&5CHIoX`;K7`3l
z%4K6vbMW5IE_GWaMmrvlV=@nYeJjGzkLLO1ZB4rW`K!^A!h-PeeEV@-3eMiSAaGfV
zkMsWPW4Ca)EuVX?waEGNy}kT?y?%eYB%kNw^RK@?$AA0pOI->=xS7rm+`<D8WzEh@
zDVHm$&e59!2`3_R5+*>_?qObvwmyYfQlimv1q)ucg~#*r*GD(W>ykvkJW^SfOxC(#
zSnua?c3EC8H6`sM0s~7TP7HHQ?bCgnj|Szo<eZM(__kt<`>(+OKtifOg5l2M=Co`}
z5u2FE{`$#fm9IC@gA1StZ@=BXXo-L<6qLz^n)OlEL=Gv79j(28C?x@pe7)TgnznPy
zaeP>*Im5?&y-FyS^-{R)N6(q^b(O?IXIr-z4}X2VElCoF+f;4OXY0d>IOEYAG3DsT
zaX&5>veBo<oOxT5&*4P?X}N4`Nj}f#v$JqOlIIhkcMY(({q2$n2zopEOsnmCf4$wG
zWy#Rq|2!ermv3(`Yi4W5^Z7i+cF9RlJsbezz>9s60fEYPTdR=i@FY`pzc&lu^7^u*
zhyZ3u8NuCjPIUq-OQzAzZoC%iyTv@;_x0tv3NrvPIN1L2t6@r@qXP+9dc7_jNU0<-
z1Gj$8kQA!Z#$#{6sa`MML6#+fs$ym&3ZMPZ(C&yx(FdinUEr8zHe<V8iyn`ohkW~v
zp33FA@#v~*rUv@hm8y405bIhK*c`(d<GBAiUM~YKFRz!1_PjrO&d?u^R^KW<o~bSg
zQ)TP>&!3*Zy_Wj25TXvlWno0hw;#8h5`@k%ZI02;=Zw+Of9=aw3yu4)KR?gQrpGxU
z30S#&zs)nkf;lBYu4^qEs_v#nINNdm$0w9cJS{aR23O~_EcyN?Ue+Z7z{m3(8WD-k
zPUR*uT`Jr8`9Sh<Gz$-^BAikN4_BM_BPB+Qr7F1fcx47Z8=@3Jopx$21R}!WV0xTa
zX3A;l`}s2xQqItH&SM2ETOsg%?5E1xP%4CyL~!04PQio`(<gdA&c_2+4m0(XU%r#;
zP?(q`<s^tm+Bp#tAusdM2PAL{1MlPchkeB}iWv}dOkLh?*#Zfzq4CN<FxuG&Kl?;y
zD5YfK3{B4#%UYH##-psamuq3yIYl_7ZCQ_W-OB$-$NllFne_9yA5IK5d1MyZN-c?z
zP>=K9)|cxds5%uBmGyER<NZwQw_Gx6d#CdKB`%k+u3^*ML!Yu0u`$}GLqJ(?x38#x
zx@!bg%Kt*Cu(#u{y<QlQWwu}a{o`8!N~p@IA`wE&^Vb71Ny#;%85R#p#YgLkE#d!B
zMP{E0(bbU<152_f?CX!0CB?@lFPBx&Mmw7Ypa@}TZ~HM`zBAU!G4{LHr6#h`j;<su
zDGLAzk|c@N#=gtf`}9yOC4sxPAvEV46EHl$A=k^6<J_NnuS?E3`T2RM84ytfl?&Vl
z5~kEU)10Bx2_k~a?YbnJ<N4fvMBD9hDWpd0l|d4td(7v@r&p0>5%Lh!em<Tv92f_F
zds~W79J^sofP#P=MrD!b=d&rGw&6LuC2_i}nPq#qtU{o@^=|pLCU7?kVF)wr`+xl%
zFMoTBIUK@B7;`2tH68Bl@#`T4i6zeWGu8`Z(7Gfhw?IHn3^u2lWSXC!A)uqHL#XO<
zsTufk-7?w?Ld4idq7gI?pmVRe4`FlfhXICsXU3z?ZUI1?K*OheyRFHAqECP&)shp(
zYC4anMVNx?P`<!|t&y-F9V-j>9w;x~_ikn^5K`W*;1;3#{RyA7EIA>kO9r3kW7_b#
zep@mF0AX)kp&zri){NGgU_3tVUY9(MeMDg13l<5_b@kcW*&%06iNFJ%KZgVJ`gU_k
zu`E1~#{C&{=yqERI52W9DYMT&prE|k`A*ALBw{Xly8U`zUP>t`F^gF5?H`Bp<<{<>
zv3=v5bV7LW?B-IJWm^O+_*yfe&gl`*T>Ck^?WZ3<|B+$r`Coqjwq)T%VP>jx_L#N9
z9GaI}uFzio&l-=%@yU`1MMMZ9=4gH3^78gwRoii&kCR2D;(2JJf3X3rO%G(E;RQ}F
z{Z0#$oy~J467NpX<eeNazB~sk3F90uKYqMaLF;|&NB3L72uJVz{P|0kbrG~VoKyN|
zLA5=6r9S=ntXnPla&<=vBWlNAf7F6t$J0|?Byk)e9BTdi>wT?r7_BUc040s{JfHg+
z$*hQU;q&o0oF%1@giL157IrG7AgtFSdu%lcC-(W=yJyCfh_1ggj&_`rmxR3ZE*?nC
za&aO;-FN#68+p1V65@g}j(85cf99TEmZb<HMv}O<?WJIjIl$mNzSiyR7BR=W=jC!;
zYs#!$!>#uq!TJ2v0O)q3@qVPs;@(<D7NO|@p!;zSt@Jqah2~!q=<UiFk|>7GQiz#m
zXQ3jsKmLCIZ4>LqX?f#7mUO$~cpT@y!8~YPtI+6u8s>sVEZ60D1|ikVo>iyIPcX=d
zizG=29XViGxG6I@Ao$tSrfgvI+#kO_uzq_<xvn-QiI9{!W?x(g+Wt8D5Gh$Cl{L{k
zkcewVa48{|OJ?hzXArYJ-Www>wJdC{g?Y{tBq?Ls;&kjij>qnlp>-yD5mPhw5lc>N
zGqwfu3*?e9m?joq(ozNd_>b@ZTHtv9;~s0SQWpw$u<+?&&B5pA-NN&=3a9)M;V2aP
z*zd+BZ7iGhrKUtcY5nj2T4bE)W51h`q~vr>k_%GX+quuzA8!{yJ$Glu(I&mEw(nDx
z>&rF6cg<<;dyK;g2t5$yao-!)oFA=TX@B~5yQcHwxIYG@B<d3o&bF+H2{6w0zhAc5
z6>r~v^VWLgx~y_Op3mo$<z-u!qP-oD!-oRrvfZvP=Xo4U1_W)-hcK;K4Z#&9Ck&66
z!$1TPnQ_H){GEwh19Z@f&A<O=$ZcVtl)rv+(;?+V>-FWF=HosLxNKR@BhvTREaVC_
zOnvtAXvdk}^1R>E>icC)5}F9n$3ZyyF#!h`*G2yNWo}eJL29z`JWvv&b?ZoFp|L+y
zz#Fl!MWzv^fOdY;a>Ox~40(|RBfR&~&o)so0Bs&UXSRpmYK<8_``sw|c`{!z@r`^Y
zM`G=Jlj_HFZ_mRPQqDOO`p~kbn7?{}%=`Q`6(jPBffD+cO?Ap?zW-9i{Cc4bA^tda
zu50#jJRiqc1IJ_n%-6|29D_ldCkA4hEWIO3Su0QI`(Mf8?Qg%oWj*)({y6WoGOlS+
z%C!h5N!#oB__bTTW^A;*ZD(sklL)RgiLlUu;JuS#CGBG(I@J{BQWEz4<Ka}Y<g9~B
z@kjI75K~DbJZJKA%1SAHl|0-29EcG1@yM6kEz@}RHXx;>Pw$@(_plE?Cd5CTa867F
zF`6odI&;b?2?lZipoNBpTQ^MN2E-9*A==wvH8VoEg$KmwyA%fgv9$L?mT!xYhiji>
z4(*RKq}GBqM!786oa(xyARX54rz8M$Uw&Wnx|E!f#Q6B*h*B7ouj_Y^7!DYreMIZx
zkC~P=_o>I-udlC54u#O;{od1cE3A+G7+L-;-LC6e$=szRnCJdHV%wI$OA-qYv+>N?
z)zXGb0quL^%DH#JH6;mv2_-M~u$uT<*J~T+p73ZcMRk_`ki@(u`Pl1~$LBxRjksO_
zB^2Pg;7X2533UsuF&wvRO#$R*Q(&-j?@Hxz*(!$y<m?_{!IIDb!bU&)<aJ%vB?;)+
zIc0GhkGu1d-MoK{QrB(!(!2NJ2>VB5SIL$8S133z%g~e(c)Ey7QbkZF0-dePl5$Qk
z9;X8W<eczmX??pcS-$qMZA2g?#*|>Z@5^Odw`5~Xn+^m%`!O*u(H`x;{p*5#Xx?ff
z7~OOlk&m;-^~dY&x)t_u-hcl2>+;){0dq>I<J_NPwythLMDuhh%bFffbK~{pwNf}?
zXgiLm+nS*t=S=p;zP^5Yy{-k_X17eHz?oe8=<32qLS*CpgQ;s-E6nqlK{&0hOU>Sg
zArZ{;*L?f_RtZV_7`p%aPZkyucCRZ?q7FGRs808xQr3btETZEZx&j0cA)=|QFR#}H
zb-zEGt05y&-qw_}5ThGWc$hg8g%SB|=QtjttV=?l>Uq1a8EnqhT2l)Efby~ly4j2(
zeteANN=cO&!>ua<q(nF-L{}~a72s=dWfpL^X_6B|cu-8BIZ=cLuCKq{vX42B=V!lN
z6X@s;?h%X*k~0|!VkzNnHu_YXpU?I6$4h2Pb*V`L9FPb?`{QHR+uN4GXM5i7QD5I^
zKF8&{ineoqyni;&+&U1AU8yW<Q8hrDvX)dbAq)C=e){F@TZZ1v`fy4N%e8Xb4_j{E
zt`hzI=RRvqOhmW8T}a3C4153Um%l6s7yt+;3?4v|(MKQ0{fO)Dw<N?wsK@(#w4-So
zoSBk9M81eQGHd(%e8eRgkWNC`@@VQ25iCS3sZcio4NRdeZzZoKCt@LDvT5_|I>&iN
zLeG^r7v?-8B|G<{+4?dDTUXuB%Zqnag(|JV`J7Hnh+#qN+v{c{qEEAFuv`;qe?0qi
zfH~BViOh){ND9n;I*`;NQu0;s{%g+618baEI3PlgLu#QwpODtdk9!*`Tg^44#AE@S
z%F?>#^7?#tec71Pa$7P}n0WxBpZCB2*T4PWm-1|7$tYz@&*#VcQPyp#nH^)+8l&AU
zZI}ScC9weLp{`9ubxux+bfoz}Sxa4mi&)z~KOXm5!;}9z{$GD9e*XPGJH6#p;<*2H
ze_3n3T<kpJXn^asZA%gapS^cTjB@}=xwa>zvSprw()#wV+V;oe^Zj}Io=8~tzs^}J
zh4%9}z49f8DMAGPj}#W6!O&+=&Z!jU%mJXicd0o?I0yj(AW`U0K%0H2su|F>SbkmI
zVYJ5+X=tg$!t78tA<%XnM{`18$-Fm6TLP5r!XDQ4=c!z>tSjmLF`!Z#?oOPOsRN;N
zqR=rFiMXV+y<E~f?vEM7oJ$fyo1;DFy5*8VgL6h}N1NeL)-_Yucpk&d4Y+J;b^{EM
zRIV?DbBfuzQ_hLzumAmj|NG0^m{c!oq8R7>5d=2*_2p8<&3%716JZ}AudlPYJsagE
zqkEFk6~s@x6a>uMw{Mw!w%+%rQ^3pXqJQ3HSu4-RwUkWOo<IMDm+K|Nc>Kpdc6$Bo
z_aEP0u4@h*V+;j$RW*+Yn~vZZqk+^C5dibNKdGc7@cj5Smu+D?pT__$wU!*y{J4vS
zWN>nH0@FE5!&;an<x(U|*zDc?9IYG5x-pim<RYMRjCPFC_E3(j4btM<`h1R-@@)~a
z_y1M463>0m_seBnn844Y87JxYkM~>!BZ(m4IA>H0FQvMxmjbcWb_Qb@P{|2Khv*8X
z!Lg(Z5Q&fbJLWWZ1(LM1C)c$m4p(z|S=u?VY?*W%kI&~YPC{O&pF-no6J#kFqZEM%
zl9ZCq-qYoBOY`T`W#itT#|#n?8t?bATrM%Tm6Q0fx29tT1pqNaP?D6uJz{PKQ`O2u
zC+l5lSuPy!pUzbOwtaLd%lDEv958O*e*DX`*}Zqyv22{|UC&>2-D)OWRz`3$n>zXk
znQh;L6QgxMPfwRMzm<=_@>cMSA1N`I^`pKqy3dK$gJqKwl0YD+M&4rD$viyL<pMb|
zpilsfK8J;uTalb|q8L5gv6L5Z?dN`fT>MGP92j^-m~6qI{Tze+^!0Lid0QAU(fRl|
zjPi2X@8?n1x~#)cylYdLREv#CYssq2Ukg1z4I;@j2f{g>@64BLDmjJL)`6x@n7+u-
zDRItBDJ2%@NAqM>7_6ymDa!3lCgO1lz8@7&W}K4RSl65qg|_3VvHc%RqfN6hmTmq1
zydS4IqUaOy?I6Bf*hbqQ=ZyJ~Bz%j}@9A>cl87MKXszwLBA)Y4i)CB4OJSUpD^4(~
zk#TBP^+00E%kT5~cyPL16KFr~kB`Tk>&y4=uh&u%x%uce_x+q;-7j@X3FCa~@!6=Z
z<ooeLOf^XsjOXdsA3xaI@!Z>aswI|_1|lVbEKH7bs<CiciwNlv01*25=;mD3b<H`u
zTW@DOn?hdJf}l}ok!`d&0whty`H-4q`5(l>;Qjphcy5Uh!N%ELz2*xX@1+yy(0t2w
z|MYab<rtdFn#px`EUia?h@dMnr8y`eLLX;;yOASFO&)H5`=OC5M9l7nvR0ou&2ibX
z<dhS6?~s7fnCUxORE8zJtfgd~km`~MMg#x#Q%X0^|C5gWqr6ql-`?i)aALx86}z98
ztLXE28ZGts6Uw(&w4k!QT#7K^nA-QV!xaIcO|E%a3k(Eb)>2$2+MGR7WiC03BN90h
zioAYL1YuL>+z-ylfB=vn%=-NUm#dJg+4H$B1<7V?ZJ(zbzf|kd&Q4pwW8eKs3vb)H
zpVR00c!-dx;DExQ4uO)(vaU-Z2l8_4(;yVV0|^0BDToy9IAGtE5@ArzDd)=Cj&4{g
z<Jum_eR;iZCFK;_9-rghEBIr)y}rVjHr?!!na{Hu&p(kzP+Bi3iX%%(F+WdU!Y8<C
zOyYm0gQjEbpYx8+QsCK<h;J19mCNv%3wwY_$RJZw#VBhM!MXo>d`57&WU~q%daI$*
z+y0oaY?md+=+ol)!6J}zqJBI8mg~#wclx>!jprxv=R4pRbvJub)2_w|Y3!O4f{rnH
ztt4O`U6`9S@u5`Kb#@!QjcMnBd08))wGeoF8XO;=?Z@xi_Hw&c2|b=45BI+B0X_p`
zo>poVa69+^_5b<)|9Z(2>arHF+1i{#9fgkeA&e3IzaEz*%pAl?h@=oe7=yR40HBnr
z#PA^ufq-o4A=~N^5H04BYKCw`DWMS%9vb8LNSQ)5vr-F9^(gDIUJ}PNgUhmi9w$WP
zx?RM_`P_}Z)r7O{liEY`w<=UJ3;F1y#q5tA2|Q@E!-zT8HL~P+fDk_0#}jB3Ri}~#
zAeGtMK%y$O<YbCyI)_e+vXsn-AT;;x8y$nZAG_)<2t<7_{l(v23%UBXo#)i~{+SZ2
z<`&v=St>{+oei|qIgA7}1p09BU+=_87+}3EIUJ*V%@}F`QJKM_p*=pLtSh7IF^Rb!
zp0=_7tTmB8{)%PZoBeUTz3d-LW??<we>vC8F?vTPNQoc<AawtEKHupg>x+;iiqVdJ
zKRL;7|9T#gh1vi5c?W{vA_!0bU;c?Xk3fijo=^0kRKe4NHWi@_1PzkJF?GpPISggC
zw19;>i1$bcmlr|8n9utoe|x*Fg<Mtp(RDNp0FukLl+5;g&{7jd*wNh;a}qEI7996y
zS990tkP;GG4{^h^RR~0IU_75)6(k9fWbeEzsU6LWz-U-%hKPs=cS>Je2e7VT!!Z}c
zMCV92oCrnGdq4fMkmMvxbC^y8B4&mFH;a-eRSHV3h^FB-+Vt6uMhv_r$0B@r6=>aq
zD1ALZ)l$wh_xCQ=xh|P7bVSYx?fqxkmJC4NXOJYx36a0zg3L_XKkpsZY!S1ax~&*4
zivaTZs5i;mr7*ggd6@R0PJ(*iS|sD+F#<t>N+G}#FYKO@Z~!L(FDr)5K86~SFt}RH
zeFRWii%{$blByYS@+n0varZ?G8G<u2qxNI+<=eGVj5((E{dtrd8_-ZmRJJ8&$-*&v
z9{`jx`<#7fK#cQ#s?<zw?g$jqzP>1|Y0e-ClL|U5D3z@c<N0ZFS!0GQ)Q$sK2aEPY
zYfkGWV>>-&KjB)5GkO3b*K2kUVvP2<KSpG7u1o$p?wGc<BoX$Z<8eQx<ebZPsRC{R
z-jB!8hEf77nK);um13$#&PxuPqn*d0+ZE@>L$NGNDLI8c=X$#o^!6YBE7tEXZZXf{
z=F9746&TapJXF&n*w3j59?_mowE%>ymrJ3rcK&&P`&MhE(Yn!kK_Bhxb83BA3$w5Y
z0EDw3;)g8Z;e9eukR&M!M+Z}jbxC>*QwOM*;Ih;t7@>2{e$EJ^WT%58#xBJkjtERC
zm-Hry{KI;0BbZ^v?e;Bsz*k<Rq*Q8>h+JAX5)lU1*5>Hv{ZKPAP1+Fyvt(`45yK~o
zBnl#MHB5j+QfiU-{qz0vSI%UZ)5|TT(ZCO|biJ+#0xqwi$G(^yC`K_S5~gFxxezc~
z&Z%oA_W&kxuz}`XQdw3aLCPObQ}KDmP5?+8eSaJZxfZhVP}-J^;%#`YWz8x2{`vmv
z!_(ynqsi~LT*^3x_Wka^-%^e~%a0E8M3TIm_colg6`AKl3z;lSWhxtjo2u!zfN4dJ
z<MIBwP)_YJ{O4yXC3C1_Dw!cnH7I34B0>yz!azodEXqt~i5zo6!2TRVRSilhglh>v
zq>KYn7Usaj?I5IJh?KO=xp%;!U@daV2pH~eQ_p>OAS!tkqouH`dxX@K1XcS|1w{~^
z`<#lIAoM(sZXuOs9}c35iA(Xr(9|P<Q%0u(Q-T@v@%iWF`|Fw)q<?%?&LySfswyvN
zId8BQ23IWt08+?QO@%Sg&ra*ht#SYW03?>gu3Zu$lw|reN_p8bGhz95H4kKaa4MO=
zbc|uVtXV)cmaHFEu4^d}!y7Z|{_(!!Wt(^5#5QpJd_Fb8jebE}2;jPW-p8B@!u_+s
z#&f@J$P!%Zx(Ol-`|5w1_CNphTrW%d`Rjj|*VhYQUTI7<pCl59&A>gaKQ8|aVNF#X
zM$CYK&~eKQoH9FP@0j#d7Y4v^$^^_MfzRGOGqQxSI0z#lIAT9~+Xu4{LIl>8VoWs)
zM3^%zUlbXZBGAs(hHp#o%41^PazX&@kB`SN1WL(NBy*5}0|A(OK)IFxoIS$EIS`SE
zWmzF`{(SQF+qDu>{P71Zml_b(&n5*yH~{sWh@eBwcwIP5-7LUN2=gUb_!#EQr7k@8
z$LY%&)7{*7;Xou3p}Z8fW-vM>k2Q_%?X%-@CF`?p*Tp{ndiv#ck?<^t0bpt#`ti7K
zZEv`Ydt|d(nS?0S?Yblzb9{b2CToX$xvtAv$=U$sT$l(wSV$7+JpTRvfo-iYJ`dl1
zTx&`tldD<4JkJi2Wwc)Y$B#uAVVs{&17yjC0*H3}tq>xHo1;!caFLo3*g9tZ+G6AA
zqaTJuA_#Te7Qz7CpKYqAa#<4S`8<2MuElf)5tssH9?e0Pmo*X0{(L<508T5L8VjKr
zV9YjxkccEDcMA+BNi-)Qg!6W(s1d!Ho*j@<CSWG6(ll?^LT+Om9Z8@K18@xldA()w
z_I!3JKH5O3G?bF;pD2#k)h*f{8MWK>?b|iSczirty@Y-(z+H8Ypd@8^x#c9GV~+m$
zAaX}UfLyf?c>T5&o{!JFr)`t8ZH!2a){pm}k8xSx_-G_4m0CH~r2G5HsocKZR<Ps#
z*GDsuT9!q6BWOz6^X;;P4s|3cIfd=}*$^DkaxF;c&Sb$t2*>+5HMZN9r)5Sr)$r#t
z<We~mAtGuLlL)ML0{}o`v0)Q9bCxwhSf9k9(+d+q-4YW~xK9t4E2pxpC4sm7et)__
zCNK{(M3yXTa9Xx4L)bjoshqY}M&JY%-fTEw%DL8z`!l9$a3ZwU5nP=pP+K8)QzfK|
zge=-1)hrp#`}5cm%+F5{LbNsm3CVrCE^H=Sy$#5j+X-MDITcx7k(@4d_IYr+rFa~-
z*DZtd#xIVfQ=J*LJ5P%cveWil@=~wl^ZNOIH=qbEX(K}>*bhnD<zE3Qj>BzTYG&`o
zK7)qph+3~N6lFo+B;S7Pec$i4SK)xXr9hwE4}ae;TV@A|-TV1GhLBm53k9WeV<P4(
z)(yiR`%||CF~$*s(C<{Pf-y(``~QZzEVUK}P~ZRxX&68hOh`y|jcGQXjwx4GC*lpm
zJV)MYVj>2Nc?`m!bgj9RoH+u#EZ3LE=X2N0qt7`pF-ro>k%xvw9G^emL0D#ROown{
z_fnV{p{(<%cG`ADAU6Qc6n(gx_k$3prYec@b)m68J*}5Dl*ya*1=O;KK{(U`(|xwl
zM`=y-wIH-fb0E>Wm|~KY1SpGnxJ%*aGl@;7Bnu%1K)6|$0?lULS}vL9qnB^jC8s*N
zwP}H>K{=H?2GMY+g=3u0Pot7pvS7@UqY>ls_OexJJH{9qWx2d%@2$^f!Mt3Gn9b+s
z(dftRl4EMj)<*DlU8J289nmy`r9|IdRZ6YIc~~dy(@Q3w`}`CswFZ^-<MAqmL<|^B
z5e~d?I7?y2#FTgr<T+Jma7s*wCG{TevaLmsVDvefmStTpHK)V`;BKns)_(n@q_g({
zNhmC07&9^Z8H9j<=w_G&W>*K2ye^313fw!gN9HjDN!Sdj4+P*uHiyIXT<fwHvi96*
zxh=D61TqU#N~8B_4sOWWhjOmT$4-}uKA(F8%2shUDoIM=v?NnglbRUZwNJv}EX*RL
z?Hnvp7aEWI<8-;M*gvO~B6uFXPmSfR%yt^)gnV5Xi8(X^z^y$#da4yB#Nz?=+i%}5
zLS*V-cAkC0<zIfgCex{gBm_Rj`PZLOY7yU$q4ssLDU!$?QYk6M{qxx+5kz3BwI-U+
zk3C#@tp&_<YC8^G6C{2<WXU2-lEDGyQ`A{P4G@Wt&=jSVJmGARO411s4j2Vt+N_uG
z#}hKAx?Pu?gedGQjG?BQ2-{<rGs~i<1wuFD4UP^kzyG+#IQNf6Wrd-xCKOLM_Xx;!
zEjqMMsKSiLS?ZQp^0KncG5g2+{^;!y6H`n(O!tTKvMskYVa(Z{gDFPmm6w%;)XMe0
zWdS`O!VcCu2^h8{5S*?n*gW^hA{gei3Nx5N0Qmmwp0D4kjpOhCRj<J7CF74dk=j0a
zyJd#7QU8oXThw(XCo&|;i3z<W&Fhx??r_ZmaD1KwC~2tzHrn&!bN}=E>&to-#_(~R
zdVD_p^~b{geE;koMJUx;G6CnM)<ixZ_s8L!1R>ga91*D9XxWy+G5Tmj)niV8>)UU?
zV|c$0A>b@DPuvU`!Ql%_oMIX<K|q8lpqnHQB8=b+UW(7DSZ-TgE;R{#jlFGH-}l`y
zB7iaGOpBjROa|+k8O+M{<w`#G&wB$Q2zO8q2RC3&gcyBb-qtVt-?}X8Dh!ke5c+XE
z?*}LJ)@G_;Ssp6DQvd*f07*naRK}1*<+2ru+3)vBwF(C1EZ{a3a$&~yIQp32pdHri
z<`yF6DKk3K?9M{Omn|`YyU%evPg@p}g#Gz2Da*QUi(q)Twc~zJeOcW6e4InwkywN!
zF?ystx`*~deq6+~ZV2J3hV=Z|`1k*IU1}2V`{&1FeXD6*7|GRro@Z}%y^ubh4VD;6
zuiuwU04#(7rmADOhuLhO?@nWIUNYxpUC7&U0Ixt$zcgWU(6ig}^6kbk+w&ZKP(ket
z-w=Y35Zw(@_DK|OLCi6RaVkk?NQSN`1kvVp$s#G0l4$nxJkM#!3C4L4o8@Jtb~Z!6
zltdt1SIL>J9a4pS_T#e)(^Bv6hMbb`gVz<tR6x#a(zD_E?OPFHx2f(yLdXmdraJbA
zQe8_zps(<@Q_4c_(-8uY6M-2}rZML67`E?{8Ho}>fN3fO?iRS*D#_Qk3PJ%$<l}zt
za9!;9Y_9P3-MRyEP6!C#<NY2fv#O~9L70VUr1e(dd>Un82qb2V?e%scz|eN~_VHK4
zdU^Zt{cTBL<N2I+Hh|!=)|$zW`#=7o<h)(BD)agNxc#W&^E}$bx{(o?fgVTit!pVZ
zyOoT{%gd6$+!5v&`^V(#>t$OC=y?9RH-j)&2?1bK6LN?#%{U<-#ZctLsSpB$BIm-|
zreU%JanJ(jF`5OZX%UjjdSgBhfJsm<0LUV$BIFQrjOU|!sa4!%wzLUxl2j8|npUcV
z05O4|PMf6I8<?sEfX!i-Z`U=$Fz-)Yw3}JF)S;9zI>^_F`1RRBYULmzGy|BkJBPbA
zTo()_N}1xg*XrlT+z6J-T9|C?=R_7J1WXIgbKg(J^0pP4^Ju4Hy)3Lf;%sP&K8GQT
zh*_$04D(Bs`uTkMe%qFW{l2Sg70kPWKA%JWuSCc5_|xaQ<h))%+rHoTX<=;uLOaXv
z{W$v1+kdAvyIhMNc0S%e4!V5Dj8cd&#+d460Bt{y-Y;|<&&xG&Sf3stKcr-__jd&3
z*%ao<O9k-}&SZwfOg4kTYVm0hLAk^*GZ30ef&l>#=kt@3q`K8i1<|x0hXN9jbw!9y
zOU;5dZO&l<NQB@By>~@oS=;Amp7^&cVte_qnfZAHP8B<SEgbIq>_a^ivJg<p9IZba
z=~-SY=xm>yvl!Leg>}FG{ipR`N4>sW$~+Vhd~`$Yd8v4NY09kKWLcK$qRqit8Rb$a
z6cYz6Nl_Ayy|ttD-kO34+wthF0}^)y#kO;{X;HI?xdVnO<}KUvocF%0fOy`PEeE#K
z&3bq1V;&#6UcPNI&vyR(KkDmmFH2$%sZof@j>j`Hge0lg{rwYufqGtSALBfpy4K5z
z>SqWOU_jz|>g>~niGq;ei~A_0q-D*Zqdorfk8*vfZ&z)*0)u$+7=eU@VIjnUnN53?
zBq=!!fB<(O8xaDg0m3o*{+Y_UB>@C54P&N$1Yw{snd9g|9(@9MAoF&;ML^2iR_yuL
zUo(|OV(brjEn6m2ZToW{WA^Z}dY>AA-X_YjUapJov!C7b%iG%}+qoa--(RZr=le?Y
z=O6ceO;~^b?UMX%lnE}Rv){YCRX^po?B`!YvdH@Kt)k7LpskI^nG;ihq{I^Kd2eI%
z05xWW*>wy9N;#pm!+^c>5`or$3wVT^?Qy%j_~@F~*Z%=jGmQ3pJf0&8LWCask#fes
zFt_KYF2Dcwddc3WVlLd<=iw=%KK9$KlyulB^H2Z?1mN4cEo8^z^MN9SoD#REMnI`G
z9y6~iV=Cd4R&liMZtbb%`u%k!Fdh5zI7A#+kkNv`1I>sJ51^Pv0AGZ+p#l1}<G9q|
zHBZCY&U>5NCd@^2XrN?6az~D4Q!OG0F)=g59EZQ$LJ`-ug~nz1>;q;qNB@ss-@d)o
z+CkOjZT$7wyW0W6OdyVWoag!JqbsMp-BxxrYae$VH#B=VK&-ct;CB@|AII}3m&`0>
zv=MsD;Fs$q`u9Q1MATh~lMr6p{`t!!7YGpyJ?;YudLJ492tY(saxt9>pa!8_a=n%F
z{_sj|+U2G1J|4Cu6iVrG5w!mC_}sg}@kin$ZZjkv{g+ULd%EoB9}2MeRI7}7?#(kz
z*Zq-JHg~fz&vSSXA@O>xD-6Q$l-B}`*1M;agoV0IBs0IhF?iqSm`Xx%w*33H)q=rb
zm2l##10?~Q!4w|M;1MOoJZB(Ea!+I-=z&Zgm#d_;Ad;P}1?9w`v$ykTT!m~{2u=qx
zpojn>EVniF-OKA6I;6L2Mm<0N{$B%!r9_etFW)i-NMe!s@#|N!-jD!@f|9jvPPf~%
z%Vk><gu2d;cg3Zqa;cfqy44gO^Z9r_9?t`$7UXKbdd(oq^;Sg9wPWh<9hP;e?0pjG
z=U+3uzA(7W^Ybv`@UyuS0%91~QklN&2#yS}T!aMsInp;dKAvxX^U=E+es!ak6(jo3
zJ0Nl@CA{R(XFs|F1+FV`<u+-3-LlQ8>SUj1gb@cIZ*@L?wki>r1j21hkgX)4=iPZ(
zuQei$`@~#1Vm_X2I*HDFxvemtr_Di2;<{YcFXOZJ9^hR%i)as&L}4sUrU4k{ln?;}
z#Z*#Y@`>WJ35!YqBGM%#h;Y;Av(0eLsh`6V887U}j6*q5PPyLxTPigZ33z{wwm%Q{
zepw#RnwEm_sk5xrGa|p_?bl!CBE2&?NZwwcwOT7=Zf8W862mt7dxPcm^>z_NM2vYp
zAIH91IUXq?udL}J#zJ=J^Oz1I`=6R>O+wL?8G4s>-3qyOwjw^B2VN1;jSHNL%-Wsu
z3gIX@qs{Z@5wdOT`gZ%^5SA1+d0P?!j!>O&Te7RwEhnM#*j*^4oD+Be0RzoH0s&e(
z8pY`~@$J>ydJ~_sX?`h^*Nq|c=yu+J9a3jo){+J1P-XNvwuMW&^!;P1PVBq7dCH3u
zOJ27kp|kfHfW(}I2&S16W0;~7N+tpzDLOP9h!|&cNHB3M-%6pFhpKZ)g(=kMxew=3
z1p4X1K!lvp%_Xf=G6C33wIueiK0cp1+u5AbR!ErP)YKo7YUP-*76c!hV5z-%m_yLI
zf|>RiKGo{CZ8Zaw%slO5|6^Tq$^zC}8#)bf`Z;nr!bmxSmKEmxY3ah}-U-?mCgpP5
zsyCb;?Vbg|+hdY;PXG|+0F<jKXzS~BSvLU#jB#RMb;o`m>=gl1;R4V3HBgz&N8hen
zvghZyU$5%N{skTNh%mE00YT4hNIC&AtxE!pE=7Er%VMS6mQpey_{Hp8us=VEeO%^*
zM*{$n^EltKBxYN*(+b1S31wSzV7Pw#8MLnC(^(ji0dHtd<W7XH!IBV4Apr2%90D8&
z%p4d|>gAFcLN$PhQnFzYuIB1oq7Sdzn!`PSpq;W1G9dy=u1pZ4eLUuuGy8exf}uee
zNb?9vCBe}VBLI+v2)qsR2+H+~rVkLN<#l5l(*l^>9RSe{2@#3VZH_S~U{0tVvXG5j
z1*WB%&%c_ii|d!&D|3wJ*&P571Hz^U5`hB{N==-^rt|eyGZ7OZ+B7%S^J6Y?N=0We
zx~-s15dfpN4&Pog>GR_`@*=2?5dBmL_Ye#$nZZ@t8G%IQWr=Z~vt9~Y5HnG|B+6OP
zBYckI=Rcmq9k}M0XDnH+m-aqkDHAt%T1s*rr6QN53WT=({#3jyV4AKekmA4#1tPK!
zV+ay7kCgt2+~y4G!K6_VM5HQ27{J6tVF(}Z<x-0<SZ}S3bp7qe%SMD?=l$8FgaJ~y
zR3-$Uqn-DC(pnSa@p1Iv#Oeu#$9Z5X1ps`SP9-uIGH3P(cc7ZX2~9N+002;gnGlFg
z%f>Mj!hyZFj-t~+2&AN#=Q(dLYt3ShU;O>;#)$U$)bm}dpFzTbnbo5|Cjh3jEL;8J
z7f^sCM%2bHmnBnZYmWmUbUgR^iWBjE<m*d1hpCt9!*D4!`<SZkm}<>r=P9cm$W5<X
zV5R_l-h=Ebf#SV2S(h}mUSC#lYb`Aph@pK9^;CQuJ@auW=DMv*f9{7uUe}t1(H(OU
zP8>P|$9dn~kJGW-z9%K_PXWkjYH+roi~)kkB5-H`f+1l@O2JG3(U1@=n9&WGP&dJ8
z!eYnsFrxFBK{J!}=O@2xi}r|Mkt(EP231~<J^>?)Q{-Bb2ldb9b#)ylv4A?UP7jn+
zG5o7ZhA;ua+i(B}Vx+`wVVYJ#W{II;r@CuU%|hyuQ_2F(o$~s6VTiKAdRb$P^S%?k
z0*x@hi14)eiHUjLa(;PPGV8t{=ct!NZU!YU1-kZjw&PUHzMoEGT?H?#_5VuOJVtl7
zsrF$*J(av@*EvrB3MU7tTbo;%s;X0m*Oicng<?cVTCxp`<G^u@84<x4;aDzjmrC}$
zPtD(ig4lNBm&@a1KYBuSU;+>Z)7EI^7(IMO&M7{1<Z|3`gQfN!SV^&>QA!k!F~$f6
z<>bamXig=GFXfB}C34E8GKP<#w7}DnYhFs-YIr|SdHMEMS!e6B4_em*riM8CeV_S?
z?Hrfak~68<`Isz1_Vd@Nh=5-ibY2#m`}swwQH@%_)Bqfig%G}6%e1Z)-9ZF>j&__O
zH4~TvIHXd^0Z_Jrfb%#zvFq89Q!U~;f%<8|c`HJaOD)VfVR*a$^~V!`{B6le07VeO
zy*Jh4{&5mxT9~$LCi3(1FZf>O(AJI<2?>Ze)xt5_{mhwinY&`jHAf&=--nmW?V1I_
z5p_^18RI<L{y5bwW$mS$4brk+swCkU$LG1ezO3x?^Mfy2(erV7L}0k~vllsA001zj
zR7$Bi9-ogXNpuK`J_C@<9Zo`u5YVF&S}3{><~bRMQ5Z5J5)-DCj){7tM0^zliWfJt
z?e)4PGVOf;`%j0Od-HkfPO#pV<i>%4V4<|#u3?5Y47Py&@Mz|dm&!gyU_?LNJ(v;(
zh=(u<004)FOTqxB>iRs3V1NjLStPNrhLn@_0U$;Uc8p(S1}qfk<g{J^fTB+Ww~)1_
zoJ%b^Gq~%0>ipWL7{zVNNzj2(N=fSTes)<QU$zAu<JSS8=ixJv>l_o^g{p)(T{84c
zl_a4m5QaWMT@o9XEDSy-*}btfT5IzD+|4bRU^b}<0wgb&t7z{Yswr>RDoB6|foa*6
zAvxy~s`}gsx-x-|-e=S$v-XZxNd(p<ff{~=pv;h&IweHmFbhEVav}tk<Q;QPfJr<+
z)lq#(5bimroT3Nidi}yV_4C(1?jtX^jj*gYx93ABW$*p|d87n_7|)-MYZXMBvweIj
zAgTft!c0@=FmqnT!T}I+3cjZ2IpzdmVs`6>7*Im<Vr`%1?o`$aW}E~jy=_~eaVSQc
zI-Qr-1c|oWTEY=r`)JSSaCXTFho0`p*86!5iXlIu#W-QdEDn-rH1+cYUZeX|NHEV|
zKP}80j$A;43PMafg^2B0()22As&y=aOKnGlIgzk<*8_>{aaw>yDhOs8EJQdh0CVwa
z<1rJM1*eK!@oA6~CH;KnOEq5Bnlq@*<?TyKGUw?1{<QsW9<aVv64X21${b&B7VIMs
z90^Fs%^W150|ez-5)neEb_KB3JSR*UZL0C>%XPbCPHCLeh|osw&pWNP)<TQ~dhCr$
zW^JGM6V^+ePqoLsZj!d_<2;U`z$N(@p0j|?84;MpY<d79u;jX=&~~1|gy8@|1(cVZ
zvh-tb9_A>el!z?7dHwC%MswdSMxWDxB@xt1$xDGC0=M(|>-`*Qxm}V#SZk@qber*|
zq;s{L%_s@Q`Fx&1%R>G<!^ZMfAi%UuNnjf6u7;q&j8e!x->DF|O?5L_QrfPWy+0m1
zZ#YoCly51TVt7Cwk_iKeGh-k1aw%;<`)C9%b*ob)T8HJeGWxWVJqW)%p@zIv2HQV6
zGHMGyI#8-h0%RBxGZ-;U1X0u?5^Dn>LI&>$4k(0k5);}u`v7DhVj_ZZwjqn#vp<>F
z3)y}@rf^`s{+<&PbA;``9wRMmPd$5jedA2J@2BDIvJmK)<2*YuU|LXHQ~@JI3t|e7
zj6ozRY`p&hM+&Dv#&8dE;@CSQm^S_Yg9}sWh^d@??%<w*!Vw}o05}&tChK$$$<ZB=
zQ(d;2$@Rc_h%f;}Ud52X5t_MC%;Skaics{&(G%FCOIo%h)2<iN^E_4O45TD=5o?bl
z+?c_T!VN>``MzCb9*2ULJbDmGDgq@qr~Z5(&^0$-YZ5x1;{p7V?OZ;evI*DRTg<bA
z&dcp~WgkPwF;tCH$zb<S517pbL@r|#GRi!L8zdB05+WoFbrjN&!Z;lf-2#A+MCJsY
zj$l9l04^&uE#4xf=(dNTt73Y6D=+IXz_b+hLDHOH{QA%Sw%rnF2U!<{=!4!~eBB5c
z+H)E(yCOG3q_l{eBpaG|Y7r(9#;_Uo6&#1M2n#-+Q%4YDcOXO-B#EJ_PB9e_`W`5W
z3BNp#;KT&pdK+VmF=M$E^xS4m=!k$^<FB8$-_{gnNb%2R+4CsH(C9`<;_=zXa0Am;
z*3`$jkNdj#KH`gzunNSqd3O1>t~H}RPvru$wPBy1j4m5!zMKu@vWeYWO4Lt_Z5>X4
zY1<eS94I`(<=eXRmL+1HJ5nGqo&-xG;k*G7Nv0V6Je$?+B}th3Y)6bqC`lODaRcHA
zLbu=~<i>2K7KlK(gc7S7ElEK_$7Bh!K%dUbrR0V0d#W|D;Nv5|TvmZLF(#1ClzMkN
zp6B}C|8^yv^Xy1A+l;(1l+>Heb{q<3fdqY|oXI)G3`B&O#uR2{ozLKum>G-;fjjsZ
z;p!{`KBnPy4IR_eEF`bR4FM>LAOV;v<je&9)Ueqmf~lhj2}`<OYR&@SK7Rgs`P=J7
zm<Vi+aXuf%@RA{rlQ8=G2T1_hu9z~#9I8eODHqhK8Yw3UBy@nBOJNwpk^BA7y0|+P
z*3mHwrc!0x|2mk%VY!I6(|}W{bv|cs%7PK5ff172B4%V7n1tp`1yFKb*V{FL>rn7+
zOb{`K#W;U`oKZ5DMG(4>2u(EzBveR9UmGTjui{sP>M&l)tu8o5f9^p+s~9iaT2hwr
z{Lt<7wurW~_XiO`jFYz%pgMX#C&T+$YAKT0=l=P?+bWgOeLn6FtxKk_j+Z0CIL<~y
z_$9{>2@hxrEW*SU`*?gUr6A0J+yNk0^Cf$vvZQDSt*=)f<9xJ+lnaMifLV&;@pM6V
zq*SmQLb#VN{i_n84P28Mq)g*+4EOW-ad&DQZd}%lT;A%M$=c^JA{2P(K`@Y+IjcdS
z&tp2)CB_p|-d0}7dV6*cXLPd|SoFTUWYBqkOa$2y3M1zHd6M;=Z7$c)eU1s?C|_9D
zK8Z(wJ~@Z?U1tz)-@d<9DYde&S?HL`=z;JrKf0wD!7+s-XlknN0N_4500baWc)&l$
z5kUkW7L3h2!ui|V_e;rz!X3QZueL@2uixhReB3`LuXW1_eKc9jBW#?%9?Ng9H<CH;
z?;WN~VmHi-S*P^6q?DN$T>JC=(aeCI$}-@Rozi@s(`|HGwxu5Ty``FxPIW?12i&eX
zNjzPoWO>25B(k~ty$#H%!u+cGq0{RRaD?7mZVLkXynl|*zT}*eF!V!3pefk-yvJoJ
z`TO?z^9i|<8jda{d^n?z8BW<1fml#Mpc9iYjX7t;7<(wtvWl#khQ(5`ABfWsIU7KK
zr*H?Toa!cU>?mQ9EhwfV3rW&E6%$jgi9Et}3@7ryoLol>z{j8Ux|G*%Yh4O66K5|~
zZHdIL^W2-_%3Y`h4n+h26c74><e(!bHVs0bh-fX2PdhbIsoUGz>-xn3p6A)-mU$m_
z%gn%;=b!!i+iEA1VNE45*l3T(z@`}UJf<;&Dg)$_^78mRmh0PPNn~o;m$kLW{XB!V
zRXAo)MA4yhM&<Qew)6eZQ<dC-B#93oeb!7iT@Rz>?YB$8n2wMyo|YvOm;r!|W0%X4
zgv)AtyJW%kxevp<5+dj{1wo3wKR-U>?Z;a|rsNWiwHT0%vcP!!8mi{N1;?oXFpvby
z>Xs;=tV2<ohEbqC=TQ=khq1^~R718(I$?GFr?DoHJrLCFoXVL(9Wb6pE)+}&h+s}8
zbs_CTgM`T;K-Q9A9_PvB=l{;7)@{j|nLJ#_m@~Y!sUeAm00Y<r08TiAh=3x%El|xJ
zL0C=5G3M?;WxJFjdML7hX@CC+O04_o5>YQ%(p!Ij99S-y%d)N|6S(V-aUNp^q|9>W
zAFpf4h!nkw?lwFn7~}EOa1U7W_1Hh39&go*!H~Bc-UrcaC>!TLe|ori=m>BKcR=El
z$VW`XJ%MXKM_O~vM4Z4qrVhW~f8D+nDJx4(L?o1!$w-Oge4aBciF}qy`6UwXA45_M
zzC?5&No6H;Sy^?obEr)eMzg>~W)6h$RTJ|$)Xb(%S9_SDBpiD#j3T!M`(bEch8UA5
zz}<6cXK)dl;ecWNlr^;hh(zwWgb(i%0V0O6T(;rYMd$Ts<(d#?GvHJb19MnEKKB3%
zF12@e+{BGJB9Of~BL)bm0ip*N@KlI6`gAjpy1u<6wApaU6m~v89*{2!obRUtO1<4G
zJR8rW{_SO5)};zTfV+oBK!mqnf4z4~<?^~FpE~=w_qhFS#r<)APACavT?^{Rhw|G^
zSi-)BnK5R&YZWIrcgZ?K+fWo&a|=v0BS{PgA~&<hZ!c>tC8vaeVY5F!PD?dO{c@c3
zcI-gse#Yy?&wHcWx7(UH34))WpU)m;Eg4-s34_mRdNz<ky-!txWi5iC0pp=e$Z}mF
zM99wGhnhzqIEZjcHh~C?)*KEtkvt&Ijw#1s$Y>@^U|$n`5<=!V9SDOsC!4*i8Bk~d
z^6Pb4tvF9hb=y|-(AMXi=0q5CbjvBe+SBG=jRQgiR|AAVBm`uEX@LYtvMgk6;MYL$
z@>ZZdK6G8{Qs(*oJ|nY^5$SRzb)d3c*7Fbi{o8G=IT3n5I6%1RIFB~Y$NNX)?|*yQ
zOd}BE@ekVyo7KXp<djlQgx2@N-6?~*n-hcU@&1^tm0RJ=(>Tt(ySQ`GX^sS`+A$0<
ziJM_b<#J2FjIwPBbq;k4Ktgcq&;2=RS$s6#RsxW+F5Aj6Mr(b1oxZreBv7+(J<c{}
z>&Phyx+-w4B~i>VMn?`u5e8ymP&0F49qvH6WTEf~fG~9gL|#|4Ia}928z82@Ob7u4
z5ttJNeD%pV`#@Hm=5B%D5gzM8fl_D!&z@Jw*VX4Zjxo<2H~=XpL@?))R0xrDP{Oaz
z79iAYSy&IHPzY5UNo2WJrg+}%GkloPey*buq^jrBfH>zQmG7TR+OCOXv_C#C*PIy5
zJlx#o*x%nz{&B6}R#>-fDHL<M?hX1~qrP0ezpObWrmy|U`S|tcKb{zK4D(VF$2^Z?
z!1S$7J7<+?#6(_}&~T^9bDs*!w&gKh$N3zjcjtQf=9$6WhGR<LIvyYIy581I?$=9V
z5fO0(bRXxK<__fUZtg_Zj?udL>><L)3!EW!OB{Z*F#w1K9b$GQLbeXMEW(`$j&8Wb
zZWNd#^yrDqWpuQoAreHs6h5Ek!`BiH?&?QG<m`}OrYkC<n*&jTARH$pXb6xf;n>f^
zk;}`Kh=E}CDUwnGH5MU|loh`Rgc=YL6NWPr4kP}0ut2y+V9BL!m#rr2s?Wx(V@!_-
zfEe9q*{XOuJuIXynRE6we7y>Ojf`{5K*IC%{BmK*wGg^vOoQ=jSD<6hm$z@<HVC9t
zDwF%{&-cIoY8n8lQWCaFgp$C5Y)&O~Dw$)N1*E!N7V&m}9P3TcASEKMi!~5ohBi#c
z^m^GA0qf8ET}v%VP%c{*VRStopATS*QmFkpF$u<)L^jSz2&seu&S~LuK&mBi@21n4
z8By4#eN~ry&d8O)9Y8Ilc%Q}`9&W%GrYr(&NG8OzX4V$iXA%dNG|xa5cLxNU&WQqi
z8h{5TL8x_f13<!1jUWk-QtFyO)!d;}W{PQFOS}``#Er?-C^3#9h2e|uMTFo?s(^t^
zvA(@rmm)9?Z9nFmV}1qK5t$in|J-|Ct_yJ~3B!RX#;FPR?xW49D<dO#aH3EhYC&nk
zX9E)VWzo1UV+P5(6#<Bd`FxzTE=hY3&p)1*l@W9tjWc@dz|71vTleku?PbXv)0`Oq
zIh9(N2}w9~Jbyh#P-t3ey;K{m_deCtQq74-glx?A*FPrRZfnK!C;hm|JU%}9Xk9Uv
zrB*@L*2mD`ub8)5KmZZ_Gd}<X=D<1Z>%Qfm_?R#=#BfI;2TqweOK%?N=71O>46c!c
z*9$8D(^9P-eV?E+i5UPO5=pALosuFT)ZK%{VcnLsCil?MyNB~yz|<l_Pw9@Kz=9}b
zC}sI-hX+J>U|I+Yw-K)6ujiKJa{ba^j2=_ZMrmc#qmLj{yxd;bWl5;RkRWunLzj9>
zMW53n0yCY%Ok2~aMYXshW(*G^=*J@oCxPc)DiKlWbR?1`O1OJHfBq>j0ZdaWWPyRv
zk*J_83k+14L)Oc!Bnm_dHD^L1ChwlgtF)g#ac!5FM{`VeeC%aQ1a80t#F|H^%caoe
zEniE_ez|?z%bcpcr&_Bi75ECZv~zm6ss|u}n=PB5tIt#6OEW@}89e~qfFnu`AVi|F
z<dntwejFX*Yc2_hV5B%KB_Rqoz&;r>L8>AEB#Eu<q33`UGfBAXzzh>5*QE%uozu+Q
zIp&0t$W*d5AqX&I$>b)Lg%NBz28byndtP6eE|M5#GdMnvb1P*1JbOP5$i&krrYE9s
z<ZS0S*k|uE5MuiELYS9|o4Z>eS(e=bC`&0qVAF{V&=f+pp{bT!z=#Oc?EE}DsIFk9
zW}~Z_Ps&W7lUOu>IiRcrHs-M{Ic)#+GQTZ_5nN+Vpf9&`_0Ug_VB2@Z`8X}Dx?8G+
z!9yny*_>v`lGjD7J${<a{dxAt7r}WBJG&WxLVz(@KuiOK0Dv%qaF0H74#V*HB5!)i
zK8FLLIcp{ZhH?Cq#6o!Pr!o20A*?kgbbY=D=OlQZfSA_2NM<y1H+OYli#dhEdoTuH
z(U^uKdm9L|O*P21AJ1`~!AXsaWN=`Dm=K&xo+nA4;0nl9WSs5Ea#<=nyoR;2cVlpC
z`wW+|#Iz|Gm~+me-P;%z$f#X`g%V(#cgQt^OF|4JLcmL9oRh`eN4Nt9P)ag%wvlJT
zICT#5`RoXA+=cy%BS1O^uNdY4HUPjpEtgA*_dUdQTV4tFf4@Jjm(1a&-JF@2NLKp%
zuM^j6j4^$Tt}||byIlmW@6V$v;yAn3jT>2`^Zk$0dz%`N)^9bzW%SnC;8b~jx+yS3
z0Ac{cbPL#@5so<{hAB`9!l{-R0uV8wpn`+XXLCSAW^>?-4un|p?R6EMWB)vQCjfxp
zb-NbI8Rk@<eLp;tn<6&@N;&6TQhQneKs0wf4d)5zc0sD9y#q02#0w*YhbxFM2BNu;
zt5PizW|3<tOZNU8oD0t9AyqhRnz{9nGl97QrsZ}49sU03YQ#){$V3D-r_p*XX(?u?
z&s{xIiGe()d6<GAcrasF0F{yit)HhlCjte5C<_z8H1G+`L=>>_F%$Nt<+3d_e|<U<
zalKr(wnKh<{Q`RT=dr!kMBymo{y3gzx~vuTdHVAG``enp+W!96Uk{K>-iKo?0f;z{
z;bS72A*UpSjOQUa*Hy-_aZdN(B+zGs1EISG5E6KVLm)B6RAf#Z7C`*<r_A925r}4p
zoQgL?qPo>8g7X+ajB%dC$fZc=R5NomH+66eB!Y0-k_9K$uE4ou#u!b}Er^BCU0v1k
z0z=H10f`ADx<f92C7(*d=zXVRgJ`Ip&yL$A+TN~E!9#o~%t?tbAYl-Umc{Lw&+cDA
zr}ID&?24#Ih^&h%&X1EAwG-I{@PPjxQGc@~$+0A9f<(+LqH5;u5x~vLte(xxy#Gth
zYP!2JZ(ac5Ze}VX{4)!2)L<9q0Dcf|rlKrNL?20`R0?B(!i<^}$u%8B8H$IZRw!eH
zAha9CHSy!e?YDP+eO<SohpuHe_On&SAOBiI&i)>b81wa?|JVEf@%y{P7!E1^_5Z#7
zho>dG{`NoLZ48gO^rOyI5*~9^RfeOn5NJd%kL$6s&(iO`ndx2Q`S}&ndV|2x)G8uX
zm21wDCh4M0=1MsaD|dIN(i&HqvWjI@w0308aOx<p!uS(Eew^Z&l@qyq&F5EYZ^_$R
zf$7TuR$yE+sb(s5+|KiCQt9*c9OLP#l|(||>uHU#dIQK>A?@f$<Jkz*Tp}1(`WnO0
z`^h<Qwt7`cn_p?rHAu5=;;+BtIE3fzH)=dy3plW<b<3M&#JZkpk{(uBz|pHx8a-$S
zvmLL8DZ(>XNNcyF&R$s+Y=^A@i8=)&OC+FrzWsPR>-H0^Ykt0}|Ih#U`PSp>uRs6#
z_1CZdW-5iuc7OeU{`J$izWzJjEaHDQ`~20d5~)w|@p?UGzxR(dufP7~cDo%G)33ig
z3e7TrV$+q2=lzCiJx7VozrIHAb$x9+cZ~#?rCLRMlNCWkv|e^vq_o;Yb&kusvNE!=
zc=St1SmS8w*RSWQ$A5`xC36MgU<Gl!Mcqy_#x<^pYpA?Z+mV-al_~f8+o83t8dV^L
z*957OW4V=95?eYD)QogUvjTV=1gcC9iPYOEAvcv;9zbQ80dtC)kV=f_Q%%HfH}TiR
z`L1=n)`Ee(yhZq|VwV+#@<&0WvRGBt6?&>xwj$|Ej75(~h}KmenUI#AXpMRQdApy7
zDf9Mm<m+Gme65tf=;L<W`^WzrfBX^uSvbUQ{rx|l&n%Grx9R-Htz$heZJMv=zb>ow
z`RAXXdYq@Qh8Klbpq!0>nMFi^?ErZDc)J@j=gNo$<NN#kdOTBv5WusvO+xHEiL9zI
zAq{gZh^h)IbInp=W>zKCnu^(t{Z=^o;#0aBicAO8EC<?gm=>(^B~oh^*5D0=Qi(F_
zNY=w2htwETK7Nxbt;Z{>vV?x|ZW(|eQdF(Y7d3gbC=s!S>a9}z;k{`j8FP5k^eC9t
z_gA~vRDG^QjWDIQ``i1QB<BD6?*zoQWq(yWwT2g25z#Hm;8EH(Sya@7s(vLiSrJ}r
zJqB$qvnp7>-MjVPHIt&fA9nrmfB&y)hKcu%w}1O_Kf6Ev@4w#vv8{`!oHJvN`FPm-
zp{Q{^U!GLEzn^w}y+*8;i(p+}uSB37_e0A|NKxd-+u2mr`YDxxLe+Xbo_aQZeqJ-7
z{fyyNC1OpRG3J$3)fBW`dc-K8ASRW$>rA80LqJefq-hn_=Zm)=>Zs4ph4Vb?v8*4B
zV>UjXDMpAa@9i2QS?Kq-V_v{>swV8X|M;zoiLbRhe7st}{@@QU6njGzbUP49qB1}-
z{Jfp|8f7=kAW%UN6;5MXGkIRb3|Q-J&8kSLY(GxvN2T%Wb3{=qm&4SK^Cok8=5%Og
zQOZ+UX64Jd$#na3W`%=d&18*4(?ByN%uwmC)!*Llht-<UIRmX9n5%yN$Hyt_k5B#h
z$G`o!cd3jumoN9vKjL;j1hszs>%Sfias0e@flw;4a&JdxR6m-s^7e7-T64y{f+~p2
z<^3=<Yw*X{LOYDl2^P{rS_DNZ-Dwq+*3=9$Q0X&@BBnJ%$ttoc#aLjKNVn%~t<YL)
zjrDpy<KEp9LZXt&c};nHd*hnpdU-WXYU|BdzC5r-9Y6o^+ecTy7;BC>#|6g4-!ouJ
zssN}Itz2G3YCqs}6_s;|2`dY1tzrd8QBhT?@a5^jes-+sW};FNP(wkACgF2=l-SWq
zj~0>cbE#$3_H7|W!i!B-qOu5(WT2ahpvKT{3S_z`l{AZ#<JN1<)bqGr$H$L1i|5zt
z{%>zxxW>TS+i@I<a8DxU<IjKD?;p4Fg3mG2hgT-pkE3@<Us*dgr1bFR(kcqW8NWvB
zCb(XnDv>Hmt`!yNO?6j0dsIJnDZwi4RlAqfOvX%53aUkl5}BSV(*cN<Yd21}-ngdg
zDU7Ic-yc7P;X%W~d4FqleLW}p$9qT2Yn7Qhu{`WN+S|{wORhE7xSp>opvt>tsRE!B
zprVb>O9i6+G*n$LuO{xuq9Us*yma$lNL68B+tAcxtbV8+@oOowa1FNilj)I^3J*^(
zp>WZvA~NORNHxvS))-^iT{H8#63tBEGllHE(o-lt*xz<rbd|q8e?8}lvj6!%x~e}@
zkDKc4WVz?OcqnBnUXTAA_xpid(;dg}?<x{wj4_5=+YA)>{ojAS9VS&7udl!U_`m)a
zEP0--_ud2*sMj?-Xt_LA+}=!3`TAPYbWUv*0W+DeQa}|IULYX?5mrHOM?-*XH5+3k
z)wmx6m?~pD7g#8y<IqUhaXYbI>(Jpz(MI;>`0e)_pRd>FC$u7b)o~tDq}kC`D%V_V
zEl2TG)#b*5s-!3PU5QMHYQ=JG_r6v}kYogf<)yh2+N7dXsg}#p&dMlNAt<?4P^uiw
zs9NqTV4MzBgisg>HHAA&K)MqZv%OnDkk?YJ2oH9$H(x33XgNaO-fzb#TMJ%m&MC*+
z=}D%jD&;((_g+y6(SGRG%IwJJ^Ht0>>;2x$P+}FY=ZM~D&8M{UcE9&-lCMAi_1|C5
z%k8+|dZP#-A~knMSNb&^zx{T|8rRn^d4D?<^ZN7G3K9_kvrLldC53*5i>x53oyyDv
zYt5pmA?Kuu6v~botD2z_-D+Jf{Wy+$yRL|IFJ$%GA<ua`o8)+nC1w(-bh#PBm*4JY
z4P&lY<8tSZM!IK_Wk*df$gR)pUEHh2Xa}HNPRbq)qYhnJ8KS}p4^T*Cp&ebvh$`vl
z%{WF0{p%vxj+%2xySLX8)m(ODD6<SBsG2UUnAv2eP+DucR(_Amxld_<qcuZB_Kx^^
zPRH9nKF*xEKA*{|pxtg=*HSan*5Lha)+9Xbt;Jve8o8YCFtaA>rRUKMpW`^2>3sb8
zE8CBs_gj-3k82+1pZ{w2k5eijsLHBX&k+#TShpW<H&Y^gvTbMludnoLI4!du^U6vZ
zq5!IlE1}q)Qeg=TqEhZ5f=U=CZ<bNkRol_)>+2bMbhY!qct*qE9YMLSh${BOUoe3R
z^w!Gmr<$xKf?}=f@w}F2@aLf#;id2-sRXLzc)K%KRN-j9{+hcs%S^bYsG2c6iLDA&
zWx;yW9leW4>BrG+J)a|s$V9*A06kTfqf$&rvLXcuA*##}?nSL-Z}PXe-H%4oHKRNv
zhpTEcA0<FlwYPu#`|o%9di{C=`Sk~W`~8Q8hZ)sWMTI_N{`&I%qvsfFeti{I-S5YF
z)>uKkAK|dhbv>`aex64YBt#hV>+2P7Ki+#5tJ1F#^YbxxUDf^Gv~^>+uO<EH%sG8L
zuhcHA^yOaC>Dqi)8$&eNa-_8;EcT{3ua{G;5+Z_jGt6s+TI+8uGuJba*$$JERx`AB
zAFoM7sF{&)pPqWYecW%|npFzYbB*g7o~*h1=R*Z-6@ZfE1%KwPF$$0<(RRwysTP4o
z)z(m~C^8BxKu9lLxx6-KFb9F0zRKERn%7v?E2FAViOis6DQ6N@C0aWbM4<61nE3*!
zy-91M4*hjGs2TKD<v8!}_xt^Jp1rY#qLY)!A9wSCcHZyz(^M_4uRs6K|NGaa=hY*M
zyx)*tv#aE?+nI^uL{7gTM~5`cYKj6_>+$Dn&Gq_xKM<=>ndu(w&7l2utWiBK!F9b>
zY;3#u>p6q18n0JL72*lVq}1doj4F2NEYeK1eAUsJ$!{mdiY#Gyh1c!4$%yByiglh%
z;p1A=SdgaHGDjShY~j{#2YtAxtq~dSz9X68g9k-7S(^}*n1zB=@H{mxN~NbAU1lKD
zOUz8k^fi}PRqg*e-3y)+KwDFA{}a>Aw6*y;b6)|}Xyz|yK@^pl5Te<**w5Y-v960P
zLnYB~?}ucp<^CGN0szs><Nk3!ZXY+T1W?!KfBt%VV<nYXOZxr34L&u;<Li%K!clLx
zmh-qv*1A^Lh_Wuyy9)g4^MU*Qh}92mXHQ>We_pX^kSgiJD-uv~9QU`wP^_=7_xINm
zl$<NFGSp6c4Tzbt*4&H$P%>x~5S8J2N)!sMRg_OT#krZMK(2koFa;<o<pU;+Y>gG;
z;gU{MC9PRLUBy%*DW}ddC3(G`7b>N}^^lyCH&U8KkQHuc&XS@^L}@3ETe`<|K#_?(
zS)jb6Dyu5%=!$TUq!21qu-+|KsR(j<=US1e7OS#~?P#2%*cwPcPb1mw&;OosJ4CI<
zSlY%s?r&$4%)Fk7xx8eTuSeMV_I@_6qp1>A<*z@UZ$ID9#&V|#;g9E-KF3&W_nWO>
z*YWmoHmUh~JYTL-5j09~NFc^zviC~1A8&^8>+^Hz$8Wbzh=2X{=~_+=PxfZ*$k&X(
z*?e(Vmr8H;e3c#=Ns)9ATQ)*wQi`oH%V(gfpr(HdgA|ZN6_pVIv39hC@p?VeOpqv|
zwrVCK8XigO5~*tGuUGCa76L#CqqMw~1sd&LBbilE1;ne+8>QLep=BMJ;S)8pkW^E_
zHGqT)#Rycqgp~=a>gQT&(DQakxX%H=_WfB9?3oGC991I3687HD(@<dEetuq$XTcZ{
zN$KbP&^5+20+8eFc`4nisp@&VS;Slk=}lm*<R8CFYZ{pe75MA%cszzjL2h@xGQzl?
z+2CWohMbCo&#NE`iZp#YR?Vu0uP^tEh_cgC&Tcu@F-SHd+ZMW3R0R#ev*%;%g-h9O
zWGW)rC=p=S^5s?NM@H)bJ$?Cdst^#_R!BxzLuCboLS-z79T+K<DdFIrhefCdlLD!7
zfvQ&3x|}9~W=3hP^SXRx6tXDFHx(gS;bA9M>2V6Gw4EBu*D73VOK(9<7ugbJ=us;n
zbWbq}T6rYp<4x35n$EB9u1o=$DGjAuY4yr3MGZGA-p{72oQw5(`1Zf52`TquTx&Ux
z`}^6AudC+2o}%r%-)={*sBfi>AoOhIF|UQzeOAoSd|XK}HC`j_qm5uJYOHqNT2=b;
zpjP$<T03$=UZ2*VSJxG#?Kp+!f%HUrq&4_lWKwFTR05zN<_xr>JgRksmvDO>A=V_q
zGiek-#Hx1mrhbVbhab9}WO`KZNDn}c8&;I)*?cZEIJia>2r44=cK1O<QlvbnrcyCi
z6)5LTs&}mCl~w5_LKg3$yH1jk1-9NB5m7O-e3gC+b6SxABC0hb!l~tvg(j=cMj$fy
z_VL?03gv~k{`$i+lSmq>_43`gv2X|=V3rD~txKsk8>4`jbCs#8l+S9W+HQA6<-A^F
z4wshx_Tzr4iip2nQ(CPNZ#S;zNVNNHy{3C=U9tV5cAjSH&*$>UihYl#GBQO-B6%Dd
zL;8GCGgujQ{`TWdGUM^M^3+O*DX+_`sv=5=eIU7m&9ppHtkW0BbWq4Dm=YAp<tqzS
za&|Q<&m>j8E4j0>RI>oKo-@!JYR<IN(jDc|59>FXuM4(b4pD-YyM>NY2{x8mO?PCZ
zm?_0nFzv9+jFdO4we(Py3n@N6Cjka<bbvV_<#v~v-t4(Z2URv-X5xnF@$KYi2Ih*g
zyT^1DRDwQGpvp>NM1rO(Z*BWc?YJ3_VvKdW9@R~RqpZEZ9a!_v*C4Mc$J_rft<OK^
zuUCJ&YhIs!UUl3I&+*wbqpYswB2sI7&j(d{e!S7gnBlywSE6#YQ)Vh_fs&cBUXt#s
z_|{o7gsk;kL{w@E__Z=qEFH+b$)W5E+*tvttVN9hU8^A4krWTHrb()Lh<4SQYj`i@
znDO@JZj1<=z9Lkum=UugFh)d^SiD?s%98W`e){yTqgo%AYQ-u$dRzg=dD;f4ujN-!
zn}}FD4n~r{BeQ6lkyU09y@f}Dq7n{KAu2P1tpa+su>>pIwun>(n^j4AGD7Tp`*Bvx
z@mg7ops1uT&rFDD+wI#mp{&f-TWczNe~y~hD-4U8%j-BqAj+I$Ob~Lsef<1k<MHdS
zFZQ!DR{#jV21HEATpp0#_j*@E5k=81zOooes}*>AYs`oYQ6C`COsyFoznrK<JMSIy
z>osWtRZyjTMMiG?Y*IFxYp)oSDp*S`W>r;*_2vUYhzk6h10V&YM+r&7y|?lRsv&##
z1&|a$31W;%?S=?wv#PiEZt&NVrfVFxa*y;?QuCTnH4zbQTI(7#NK<@=I}&b9Gd)#B
zstRbQipqqjl4>jZxkjWaEA+^i$sS2lKrNltD1!A2P^2M#{q^Ue!gV<c#cnww5)D-#
zu!F!nh_%4!?RGYgd_GeY&jITy{Wb%&l2sYfRg@5D(|jevOC;wkXg@k;vQk7q1u0Hd
zwtni!)5@2$wP;jfX82$i?V3K)kj_FC1+=3n-jB+!FPGM9)$40PQ*=y<5|9F=O06pP
z_Y}wytRRLcjiqOmAhSH#g&x>yQ_(CgkW4)!JQLQCRa-_XS~!@{#$4J>>;TCGq;YAJ
z_?*ux+Q+R!kQt%?YYy+G*31K#(^v^W=!>-yF=+z+zc?pZ?9J#|y|1b2rSUaEhKptH
zE??1oz{^^MC;L&JP|jcl^f<fK`npsp4a+05$|N(}{m@uo_=v8tMzyogVdgH@RMO|H
zGR{ynwfh|_^LdqY@lvujdB~WhXYb7Gx@JL@v4Y2uj;)QA_nY86YkpqMn01`5$NfIt
z>dH1BXuYj*IfWt_+NB~(PK^{*WcW&x>7c32A(l~9>C&V~N@kHLJGv1lUu$IoVpgTC
znUv2FXouFEicu%8AW#rJZR&^3$T>yEP{UqM79zM4$_3HmJR0LQia>$Auj@5ym!1Gp
zW2%aDxZawX&5;%JwSZDBcTaTPB7^zG{~=Lk6ltx=x>iI)3b&m)s|qA5wQEN0)J2pq
z6M(4jdnwiW?bf)KAU(sg(hHFRRa$3e6)NHzyufxC$H-V4XFxr=Zbpb6{$eMFQ&9nA
zw!Yaq0aJPZDG~GW@+~-L_0vQ4BS8^qT{$PpQ_R#VBPj}pv7{lFM`Teny(svW#B!nm
zrtVTJc-%PLL2HV!L1$U2ks>JAc^oF!<25Lxq(g;O-8?0OO{IdI%h4rw2vNSTKT%jq
zG+czL43FHq4wYP`NU>&$E39d*B;cO3rm7XS!G0N0)P9=Y`tz|OAS|frcol`21k5Ct
z^WTD2MKvSCXTapOn!5eeF@67~P!O#(<GigagH%;Pc!}<xwS6PVz^xCryj)OO!WCl1
z478T3Dzn(mn^ct9yfS7=d69Cyzu&fIoPYh-;M;9YWtOyF5m8BkVurLcb}4WP5~|$x
z`J$AgvB*92fDp+z`dZc~h>&jL3)*E2FIIsl+FhSRY6hqx`+pHCvHA^3y|t`b*V+Q7
znUoKn?>|m?em-8oF1frY+*Fl<nlWR?C9A3u{XonGHA5`GEUMdb$tfp;bT1S|Dial6
z1ZZ6{D}xG}WQv_qkV01abfBnasxhxgI}B<$$LL4rYmr&3V)2Kd0Ks;HQG|$u=!T$F
zdSF{Ni+e)MqyR-JeB;%^qkt@m2#{a}SW#4EK@P2=NMVggGfI=J%p`%02^FGnp0NAd
zS<k=z<gMEpk+A#yruF&P47B?}QgvNpzD8C2O=VXTyj}^ZWNWP=AsAsQ<q_5@MMbqD
zD)-+|A}W*ZcCWA|@%0SAH0BD~U7LW1^;optv|`R6C3D|cY4?x!LwzBhueAzF%%N7;
z{@WDI{Tr;vBE(o@iBft4P&K{Ta$s+jSy>f@<6Y-k5oT3Yh*^@Z))Y}mP((FcO@w+#
zK#^1+{hC>ciWCzH5{9*0TLs??h%!d(?DmSVMo|`|Dq<f=*k;_d#4=SfvWQY`y{Pe8
zR26;uQ0BT`&$Tzp0A-Vfw3UnOlA-|nX&Mo1#+W$H-fg}9_{;UY-{SF#^Op5-zaFnT
z?;nR;Un^FauF|7jUn8?1f=G~}t!d3!nJFge*3;9I&6LR42Z*q=*P63L#Y20?fE_&(
z2#;9F-g9~EZzx;O0RSSgx5_FRDzPF8CS|=fQRQw#=vHe%h;4pzX&Kxb#S&pA_HP4d
z4G}@fB0U8~K~e~*kj7XGN@*6J5!x}cC<&v7$_6y4YBLi6Yf5XX7;||hGAWtDDx#oB
zfIpa7rAi=ktG^#VEHB4;EMUKJ*)^s6bH%>+GlNP39{uQi&8V{DJe87<rzbMSR8a}R
zN*p&SV0qIySGD839fWnIuhQeZ-#X@h{nuaXZu9HYS=HZu|A^P;EA_bDr4)~^&sXiK
zhg#KIqmPkl63dL*Enwx>RicP8Q{nlcIfiMhH5aruf9>a8+MOd%Rq0NOvMNICY_FjP
zwx;EvD&iU?t%-;SnBgUi<sJoVCYjs^BvBcuQWPRQOE_a+&|qzh-ZO<LX7pRbH4lA_
zqL{8(?9F|@eMthMqRg_J)wqU3yJjgOvEyfekl|UB?Y*+P()LY|Qt4%?RV=>kMo%Ck
zliCmR?N)xxSR-i?0VdQ`WnI)LvLKu8B%DK3)ykpB2<})F>l;N^->k4*mhJ&<^tI4>
z@7qO=uF`L}gSnpndNf<(icDCW3z9xUj-x5!x@MVW281Mq0fU258Bs+uV^$TiK-(Rk
z{U)J`Tyv}}f|a6xAVJ-wfs&O8)mpKg`EqL|9h(nZ?pY?t$US_~47s8*3kB1^v9hA-
zt4i&lmsJ9+s-o~4AQS(8ouD;kf$-%~WMqj+B~Z0ri>&m3=stQpRU^oxw3cg|yjZnY
z1I4B#4QmyUyJm)h1W5sJknf*Zl+2)-RFqVtum~~}6>>^YncUPUq7V-56Nc8Cp@51p
z6F`Xo`F(>Cu~=KSNI<NHC)IlIkyROK{jfr_Ii`Hv)U5pZ_4Np`8rg)AktMAm3zbP}
z7AxfJySgmL<q2zyB!yKpAtH$ck7F*BqiJSE?x^&dVJD|HKsi0Z@)XfxGf0z&0y2pa
zx3)rJes3`3XoQqUnjxpNyrhDbL81!A2#6BGBH#(p@YYrcS3pb<Wrq~7jfhaoAR<X)
z0YuheuTk3IUhUlFCRx>vo%$_QRuxUfy>;VUEce|^08|kbNwZqKS8L%EA+%L^v=;l3
z7ePh!Gb7wfRn&YzNU(B2b45YBd8n#s<^Ea&QJD#5itR()P}uIt@}eq2q^ci^HJ_Ku
zaW+$+;kDY^&yQn#J)UDPN{LutWIr3m0)-{k)IDGV2;d=VCStY5wJHlN*lvQY%FGfW
zWwYqk0*PF!1W^SwnJ(5etAN(hue7|XZ8J!giKhugQTrKfO>569`$Q6nRfuYqMJ!mY
zs@mI6wp@XJ>@;ZstbMgxGXW5=-qkllLr%?|@9xc5#2Nu=*X2-ClD@6Os_+y_x^8Jb
zkSP}4PnD|lk&3dGlRq|ItW-e&l%{iJfl2^H0`>Q2jc=wzMhOa{x~E>Lpyo;tBGwvw
z%`ALh9cBv2tq}vLf^6sgZY+xE@BjG6y}85x`t_>wY_HE5V3*QPUyA}9mc9yAP-S`5
zcKS#Ga^CMfLj7xH`1i^M08vq(DY03RX$MnTRIIT|D!D5fRTaXB4aSWG1<7xIsFHSU
zF4vq6QzA3W+*PwgfXdkBRi&p^K&mp031da7ZRC!!G9jv{45UL5RI;Lq8<hKf9Bz+c
z#wM>tK+I&vfeU-s#eSRDE{%z##<U}oOi!317Qvt0r)VUQRS<49zz+3|<lR3|72&0=
z5x&`hMDhM!{u(L~(awBr6|wauHLl%fog37F63ZmAK+RbyV12JhtRMH^{?Vl}#$OMg
zV=fO-Rke2BFg%uPV~&~B^kTPg$IU#YS*(5nUB=7|n!tUfYGo8ziSuq`MM=D7Mru<5
z%$h%WG!x12KrAm3IhXd{s<`{0&pX+BV|ZLwQt&n3j_}B!>}X`x{=#nxw3%*=u=<Id
zE)1wiIv|n&L^Kx#P-D(&@$|}~m?o5&mAT;qRZtaLi%CWRD2K6nXI)oG(-IX_MeRVC
z7b+u`gQ_IR2oyhz$RZ-LAi4#0-AEOtduxo9%KYv=$gTSathRbeVP#rF6blp+&x*(-
z1)v$y6MzR5qLP);`f=D=nZBl+x85|pVqPOq5hdEJlG+<7G1e-eP?1R$spQ@!tF|WP
z;qFlx0H#1$zj8JZYYL}}?6;<v7J4_4^^5|ljI~phrj$$qV9sDAji>|!zK`;LSpPV4
zUgMfx1f{)o_gafKFSUru0xAn?m7dL*nVG5EPaZ`T<mMv)1=UTP7a@eT)10FFPO|@C
zEDxdp5mUr+ZgmlCr)iVxqNGnW0^7SpA)+8erN`zt6vfClQ1>t($ShFBt6@|(2`>Tm
zbq#>E^cBL4gqf%ypO;z_Aop>RSpYJMRvx8N5D^qLR}`t%Fhkn;{zJ>x>zcmGx`Af3
z9z!tGn(g~n1d6JR44}|8e6lOTsa+!is?uzH)fTWRMAepEcJvC`1cCF`alJ;Mbzzl9
z`qI--IVT0EM7v2@H@QZUOrKBzd7f6+wdCAx@2o7c!V_!eo`#87#b%WeWbGR0^pw-W
zgCY`(C<vo=9o3FliB!^5D)*kjOH9+PtqI5;ooibV1yy2jh-z<DlCPJ;wq6HTqV`H5
zfGX}#VWpAx+QPG$lrOi#<eHRZV@8@?vmc7m=3g#mh$xd)y?3Us-6p_Hs_eWtRvH2u
zQk0caDNTh{sfw7@C29Bj8ivnDCd8Q1@3%D;?sBcF%xsOhRzfQvWG2}fzsDnynyZkN
zgqlxC6X(e}1KOK=)8JVz-kLS@IBZ;V?Z`*fH|2mpeHWL3s%owGrj*`@%y@oX4l%h#
zO6#qenbhO!%HF<jvJqLZo6K!bjTDT~eNBx(MXFi3v<lI3uVxhqDv{e-LD>t+0*uSS
z&ASU=KZdA+9_^aKm1<IZ@e7|TO)8SI$7^Ya2{1>JqN=${E2?;c+)!eQ#-JjFq^L@|
zXBN9<u>fR>qT+k{R78B|N)R(B@X*#U7KA{hh=;1eX{>fKr&osD8zL(_%x0(nEKn*u
ztr>FPLab@IQ$!>~81AV8a5ptn&hLFLvhaOL++CGbFcnmxLXO^=&ZT%<!tbv)kuYWE
zT3MnZR1;LiEWN#-$L**!m%o0!mbLIi>qpH1XgbGCv(|d764o36V~MKx+<Rmop-p{}
znX1glJ$W)|SUcTS5oi+9nVt#LOcLcKz56Ty5@14AR7Iv!h)tzqR%_U(pjcF5zfC0;
z#PnNH&SzP<kZ&Nf1Q{ibkqJ_ax!(!({lRBOyLGu<$ts~#(wM3$A|!In0rhD|MP*3m
z^4Nx!NFlU|C|*yo-k4tW>19@IW-*{XBf0786cKllvLfw%JFpVImZwLOv{p%WjZ$XC
z^5Z^-Fg(z1_ZnB+-$<!ZR$f3x*6w*_&H*h7t@S3vW;A6XQ(Hctv)$kC4S9WyKxLh`
z&XK0v!rY3gEHy)9Y3p)b1uxOPdM`lf9<<(KB+VGhYrn9dN6Sd4U%m%<fIy+dj<)vX
z0+^K-$Q9p*tDqoKJgvrSZMDPB#!69<HHn09Ee`-{{9Y~yie&MpRC!>(Fts57fK?f_
zojB>Ch*G7<_DHfSw}^rQpe!(R8)CDl4HW@WO!l_5sNBkfDAsZ+DO!m7-noX0ep|m(
zd`XKi%W6{6SQ(qM1r;A9N-?Eplkf=@DF`zlDnQQJPct<^#$0PT@$LJnEP9Hy8gBhK
z4+c30;x$tadRE~)ZAXEmo3=)Vd!(2k*0qkF5XG7>Q{iwSbCqUj+pPnvDgdg;ie#s2
zm-G!?QZ*}|F4|;Ifzld6)Lj5JMnZI-wy2<`Lo-2P{JmFm|7JLffF30FQm*$Hfp4!P
z1WD%N8z87CF+mnlSt^;RC{?Rc@lZ<$YK8U8+@CL4sZzPEx%zkK<_ggUFKrTQ=VZ2}
ztd?sNGWXPG-7>0{r#2QU)0!{?WH_}-bn}J;wU%fkwHabY6)DD*B6~Y7!bl>EYw01o
zfkP3Ij8$loi7jVoGZ98)l<0A6$r(hh$YSeE$H!gZ^YZEkBEecwVw>t>+oD};O~swx
z$XFBrlk}oVMtDf;GG>V>+9A0n;7;j?U5X?oc}*hQk<(LHQ5!EtR$A|tP((yAq98dP
zsv>(%%q)=frHw$WG&7=#5+$mr2x@v2Pa({f%io`uyIDn5L9DTQd(GCEUPv$OB_dH`
zl_jR`;1--wFx|^7Rq5w?F0TUrUZ-y>?f0e!qU7JQ`J!bV2ZgmtSZl5{LpgMpAVMk~
zLfDzZQ_N%qtQ#UB#@7q!rKoo6t$nx4&604nzb^uj1>Vl<B~ma#tYR)eWDDb>;%nl5
z7tn{h*vC7v)@xF=JVESZs4A@kfCs>S@65Fph#0X^5VD66q!&aeW|2bVasd5^I1c1K
zO0lQKN{2F|YKJDJ<e+<?1evIeDz=^uL{bcql@JB008vSAM~+Bf%hXz4$&a2vQDkK%
zw42X_GNGz_0Ue%ERbt587?AIq6C#>bf>|YU^xPG39*J*ZJBL$D3X%zWfJDizCh($I
z7l+0xpt9dh`@WVDsu@|V5><&X-5Mf&6#<j!Nl`J)$c%t!KbquRaooDf{CasM#VkA^
zjBTSASz+7@aEL@6J!7NDP~|?C!z7E=Rc`mLwH}XGib_Tm#Ka4^K+`U-i`ILS@>xY9
zhl>i!S}u@quSAuYYG%S%`MoF-dMX@RK@~B{NMvEV>WRP6f~CFNm<b`>H#1``h%zXd
z+F2xkT)7b=*vsXt5{<=o&D6FdUEI0MF)6gJtZ*+f-HR%_P1lx-D#AlreMj`NM6_vm
zg4|##pu~h()=iPIzC|rk)l|%UiiDgN%R!$JfNmsKp`RPrRNq4sTW`nRR?7G_v0E#O
z%w+rVft4j<$H%=J*XxhJM)srCbEL1ml&yFzNHgml6;a4_d1X+KE_(3wa_vUo==1ej
zDnS(r#XjB$|NQ4=%G?x?OsGngsA=sz<^FCF9u%Rd#!|hea0R!!xJnTrYI_SMj4V00
z%A_LHntCYw8z}&klxI>Z)JPyArPcJ(e%PA(wA%M7RS~Aoz_)0^eYqkF$@^|}*=#pt
zBn1>vR)UD2#_lj>?K0BZx+4nTwv>P>gt^lkq%?tQ&WOm35a{KZb3#aEZuwkI#e=H;
zLf<p32WTpU&|Nhlh^&f0JKyg&yxj72xd`JLuqLqE+f5**dftED4T=2x>+!gv^)BIl
zy{<*ca8Gs<Qf16y*YfFJJk0anURR2!Y3rSpYlgCXhTPvjZpiU`j0Aw~?>^7NH0Sdn
z?RM;5TKK}g7pZ`%87Oz?VNCbPQj1_ERfsBNDhp)gZrx#~X?Ye~+rvR=(`|XFwRA5?
zw8h$%J+k(`x{5+=hICI#RWn~+1Y7q}X)2XfB30jsqx5drD$@p3P)xU9*Y~0xwW=F{
za#iHsWs)?R3(VcP<*`Dv3aE4u1@`lhNadc`_HLi2kI3z%O=-O;V`Aq;Ald{Gg|BB5
zPi3W7Ew_%GVuZhrL*@2%za3o!<oR<v|M+#iUfN`H2{K_l(-*Ri^SnJqHj#u?2`#TD
zY)glpvs)x30qF&)B4g#L11fNz)vOuUZi$C|d-}E>q9$iY%m|dWR?w!EEfQdPpaLdU
z%xVT?Ko1F;fOyPv6Tz*=+7*TsMI*@;krYa`o8-3PNJf>35ABtJyYl;hsKRbXU`c11
zYQ^486g9J=_gtCSV-J|6m=LLqLV+|xXyy9*-%wTtMKW7QW|^tA`L#;yXrv0mCq=OH
z^AM$3461wZP?x4&Nt{P`0ZhPMPIu9cX>X^*TJw4a8&^;?Q_i<H`YfsI!H?hjy-R%k
z#eN*k*eh<%_0LPOiUMESMAgSCNV2!^T#Pbnmf7W+BH>d|D?(7{;wz;Iu4`TfC+ory
zbC*o|qpBirKar9u$%X({9&cuc)fHowknJ#8U&}r|`ktVtgB5FeQCrl_LeAclQsb54
zOUxoF*^rU#`@VvGSDZ<0x`;XCG)0Qf#R^|4nSzuqB`SVp*G%PR0}uYxh(dkOwB_Df
zhO+kajJ+#PG0jy%WHrkKDz=%7sBd#7_X|e56|}PCI0{YISmW~4_6;%PjVqDyeUnCM
z$D%bwPAB49v#z~(DQ%5wLAzYDBG63waU8c(nHit|{PUr=pC4MD?6<QYy~5YmGh)7;
zs|q65Iy!PIazz<q69L3pZ<zpVIU%h}t}#~b)D&rZB|wVBD2V`7O~M1(x}efi)ovfB
zmS3;6u4pn-5u)LcE}nM0|9IDQU-NlorjHUzX40CV>}E_K*IFATxw1?nMD_(ON<=eN
zDBUox5b*4;0f|DESV8E?9lG44SXIP^iMCu;M2Tcp5iZ)>4sL<Af5%jnSc#Her4k;Y
zrf`0@UsvstUI@~~j>DL8HnnC#AV<5%xlm>miH3;fNhO4WDzFN|Sg{orNg?1%wA@8v
z0u<BxZNA0~Xglw>W<s;_)y{j(=lZKk`x*Wu;`uaNGkM4zN`&8Qc~!OB+fg%y581O}
z=JZ&RDk*0dI4MU}R8S2NDq=3u*%a3H`9z3GB*bR|3@Vy6J8tj4{wyzR$J=2dmE#(#
zpTEC#YdQTfM)<W@r6vWKh!9-Q*n;R*i6QCV5&#?Bx71qFMCci!+7HRKG7BcGQY+sf
z?*wygs6lb-h6`oqF{-NV?PbDx#oWJqYb*y+N^E6`mXAGD0n!X>+R0itn~;^6UJ<Nz
z+z+epT-YueFqRYN-LJWfuMkK?JCvR(<%>;a&5dTSoYrzdDw|bBWbrr{myeOTR@iOM
zZs8>6=Sb>}B8k?-1GpZ~tDQ|%lAX05z|R#>Yjyd`I`*<gb`CP#p^E9XW<psMO<GMc
zMwJ}h5Sd9aVIh}ScvLz{Rqbe!ug^lEj{9#W(#~5K)k>dVpRW?TTX;tDK)tP*TvxF5
z-lPKbsx%EJ{(ekat@A#IolVL;%NG=Zn2zJ#s9Mz}5*f*ef-(V77KJ@RNNhjlw$=Wo
z8Q7Pl?`<Z-31;Y7YwPkv)K&{qj8taIz5ua`r#j9WiSUG4KTZWI!nYky8N*@Sv~j#P
z8z3_hYJ01)wxGzp+ODMR0dG5`BCvNK!m0{KJ8$ndpJT;%hAL5NK$Z|9Ya~GF_nuj6
z1S5~OW=U%rR}`q<E3>8;m$ocXshP65Y3_Ro;DGF9w^)v{h;}>rsdLORqLz!9HWjmY
z&RkRadH?a&>~`L|DkH1r<3IoTl6Q-wT5Cj&r`O}p>fOS>8GR}$Ym)9DEBkF-)Cf@%
zfU&%K%M?VGb*)NeMD4u3ichmDC9qB9nGvbnp(U{RmZBye0o84)Q`5@WT|lChrWIS5
zqG%+f%k(OUMye*7Wg!=oWF-!Dm-F<;DoC+g#VW0=5@Sr4b|6%rk5aR$Z|RqI9E!2<
zt=5}j__x%YDG1(=$V!)^pYM09$LG_Po>hO}93kgDXFta)<<QM;5}-Brl&<A705)|O
zAyD5pU8?4Di8ZA#=CvT-DI_&kpmWI?gYE3CHF|PAo+Ec%MdV)fsV3ik{P_8P>#{}k
zit+i!e|=SE;qBw&%`k>T!qxos^#Usc6fw<I&P5jWIOCej44L&!UNCico-Jd0Vi=_;
z(AwGNwNgYGu>b9gL_jOu#aR5YhiDq2q(Y3c(<py?6DpEoTwdC?RpIYvtSE%1nZztb
zP%Ddum2#X|R8gVi8nckm+H#QyZN?O>$Sfw*TDhlM`u2Ey+cx?4A4F?KLPRO3>gRc=
z!5?2Qbj;MIYl21(n4u_<B7mePP)GOmoyTb{XQ1hDJ3B^gwugeuRYX99y97W~GqY5M
zkt*Ay%J9;&u^s+!codsS&Zy#f%K^K;-QVBZ_Yju39Dc3z>-k)HbdxAm?Y-8#uB)m_
z`Aycw&ckR^Y!STzVpt{L3M&+`v-FWf6qN-P<XXE2ij~VDs%lNb9n3)Swq-zciGu9a
z*xbJkKy&*9Oo+@vmKrjlQc)#jm2w}PR*~>j#ij<FH&&Es`8B-6Ot$+<1dLU3JJ*OT
zF=mA9MGsV1P(xHk3TwxBH)@B4Drv`|Sj&Bdi6ST>l^ZZp;nKpXJI9t0VEax;6vP^0
z=W^79$N&);nMqaVj8ZM>hTQIz1iMxQw8Bo~oXVA^DiE31E=OT4$J<ZEadd+;@ks3^
z5$<EA^ctUkUGp_)y?2I~59{eGP=wU>Jy&uYDIEgrK#vI{3XtBduHl)YDwU{0Q&!a8
z9t0}T;Zc;Pfm5z*P3EP-H;7P>3x^f8N_fp?83FE+lGY6?v8~Of%#8S^x)E~d^wyLq
zlFL0|hTP_%DpWc;Cakq}2}_#M=Zey<+n$nZ70kk|0R>diMD}O10-`K!J0ALYPBa!J
z*Kjjd+7TmRs+FYzcPdIb7e$2WR6zSpA45d^f;GMT?CGA&fEtIJMK#l%)-2xcH(SZ@
zIhIpdUMl1Uw;&@*dzTbkBioNRU9Yk1?8kXZc?nXFqbqY=V~&{;)AuV*dXrqWYjIk5
z&d2NWYWMR{R>HmX^{b4v9EB>eaA?yFXx?AfRLN>q+O^yhrd3%^%Oy<^5fR^|Zi2;k
zRuxTF?$0ADL9xc2mZFl3NT^_k#6&ap<CK61Ky4bag-j%xA+yBi-=>3Z$g1#!n5cX&
zwhMq@Z!%HHQpsW^w2Ic6QgW;)>8DETx9+cB&jksmqC^D|D1=i)%&eUjb6(f}HB<tb
zrL7ytoWrxKv>CU!?0>IWqq=2n&nba!plM}Q`I;_@^b+f}zK?rKr5Z}~`IT~<oe{o*
z+Ko9w%x)j|qr%7Mdb>%CF=iBx-Zed|Jl}u+xZe-1*XJL9{CZ`-9oCxC*DS8*%t}D_
z&S=|M!xPrPLSk<eP^Eie|81@VA^~o7012dvNm-JTIknZClr2gcDox?qEBbLwsA${5
zyF0}+sX$2U;RV)2<Q~ZZs_oG>sys{qg}c*qd#h0d;fpAco?=>&X)PSsF2O()9J+f3
zxYA-gB(oZPvBjbtZAO>1R;nr4j{B*(D2@`5W)V^Fpa94!IkS*uZH>Kq5O!-#N#oZg
z3{No;@UYh?WTJH+=t_sODYVT2Y??^=D^i%E7%#N*xUXx?P~&oi_zR9$*JCDb{Z{_?
zHP+?T)>tqfE7$x|=GSYX)9dvpHDMOi<dmzlihyiWf91Xh6f!6zgp7=uR1{1#vj{?v
zRZ!tBQ>{f6@E=xL;Xx95biY7A%9AVtBFt3D3^2X4S;Y28GVA^}UaISFC#EvI2s^C+
z_c`|$ZRgxn1T-tcua(jY36;V&ft%@^f?4OrH;D{aR>91aFd0i)<DB0Tu$AWySZkzO
z>8;I`nNiwAJOv@WA;QDj(VI2NHRqbs!@Z<6Yi25xdORIw0+{702rE{Is!w9AqRBR>
zVrR@vay&{aAxE!RF2_y%@m1Ozas_YijRfM?SK9ghcFKH=>ouTFBA}Mp4^&ar`XO<>
zrdKVewyhp|3a6_ueZTrD@$GjiN;kw@TL>0Y3d((xvaKR=?=iOGUItKt3Rq_fDCq!^
zC^A!7N%bH=&_o1EXERZW;I(Sagh}T9fsl;pMl#(KhyWofS`%`%ct%#Gho(?iA+}kj
zHIl+qWQdxrwWOOOSA{64T3FA`z!Iv6M?`L*S4BbPdd?uLa?Rd_9?6t<4M;zlsbsov
z4j0vbJENLmJReDM2SP-0u69&-0(&o11(CwH4QM5#q#Ls;aBR$NES7OyrAS6lPO&jQ
zUq_3TnTh4&arT%zrOWZw#`BVMuX&dzXBi=_^YJ`xsESJ8H`PS!U{R%Lid1|XP@2Hk
zH`kQRs8nvVl57zdC=WojVkO_VyB*M@hF4XlN@4eDf4@jYD5k1fIY#lg{kREY&BtSh
ztBf9woeA3))Ba&lyGE6-fSBeAw8r!m(2aj!Rhd+=BEk%4jk!X4XH{uyQsLv7R1+{?
zSLn$Vd)c}DN}?M{c($BZ9Zh0+(e8(PZsWPgdF!qGb&Wa4jMDQz{$W+IUXSO<8WW^^
z>(d0_Wki<s3els(XUgvXiBJt_YglVR$R=6eT1-_{@;G}wUUA;nGhs#qw72z=<9;(`
zt>-K4I81ZB#*Y25ADh;cKD_>>GmwC(C^M&5ZDbz+7579zgc)i+t+^Mp9}*EX5;;~v
zC<KCcB-CWHfcK?Bwuw}%Jh6wWDq1t~Arffk{np^u>p4XCS&&uRx|E5ENbM@71#(Ug
zA>GBQAgooU^|u=}B4NrZQBO0L=f);M>`-^Vt}-bUflS)ok7ubSxkroM!p%rKPAxM1
zaUDI`51U`Fk!a_~yAk2Os{Z&xZ)X=|IOtogqX;uZ!qqag^?J=BDk#b*Y=@%|<z6Uh
zO6yWtChn-r!UjI(<f#Fwg^P>{8azOWy5B1jM3g{8HVmYEl?oHq0BlfsEJ~M^-B1Pk
z%8D#fRraI0L%d3#+lgnp?u8^MyBnR{A0LgPRcl)$lQc`O%oLOvlA=ZUiZZ`q75Yi=
z@hJzEQ;u!dQdEh^d6!(%3&watvu-z$;|%xZ3AKoD6|#a^MJjmI@?<l)9%6zj5yTRV
zAset?Mgg{sMl;x(tO+S7ik1sTIlHR$dzW%Z`N!+)E8Fd$sA0H|dTGD6CbF(l&A(i%
znPg-M0%p6$>ZlN%LPBl_DrO$XHK1Cmx4fLzdU{k0z4vZfm1|DLN~*QKR5+Z{q~=%d
zd!sguaj6}=-9Qs$uKD=%QrmyM+BlJ0A4_-pX;xt|Yb7(tSo@NXtV%WNR^KE>8kt2F
zddB}g4w+?JlZl|1o<5a(T|zMg6qyk~29eb?yuMQ*25R+|YanvqxV7<GMM+e+p0Rv}
zJ4^wgWn_p5g^AKuS5^{%LZ)h_h#;YK&jDzKAfSvDyM;4SwD*Rn3SWU8Sx#g|wx)s%
z_DmN8DHtnbft1SYSGD-0S{>q4yk6Pqfp9wn)Z=U_uLp1C$Xr0h!mIW5@<at&BOv#q
zckdP{rY{d+bs5tu^O3dC?Rxk|(TZMFfmp9y%2}wQHqErtQhxi#$6IfPjQRZh^RLIC
zG_xwbiyVE<-pN>%QDPJfH_cL7R*_Qs*s6*Y0<z7b?Pdtia95_>%qmt;y8;yx%9_p(
z0}8vGg34H8`K`<E+66XKYkikT=5np)%h${ezt0ktEXkl|BuzXhs!|@Aa%-!UUIg6>
z5R(Xe1J+B+x5c+;N_bQWK(%%pon!b~(pW%*ubqJR-Be0Z>vg5Z+!RFHH$xG#N?#c%
z2hzF}ML3s7WzmlAOO%xbpVpbkO4RP1W7YBsN8cr(y_b&>CCKHVhuPZGiwe=yQ|M8V
zkV=t~n~gea)g~+HVf)B#KmPIi&s%R&zFuE{{_**>)(jJ|o>L@Nf><yu8Y{eNUvVv`
zXFtMA8@%?xln|8f6#33!3X2VqI~FF$_=Y2<&0#I`=!ggbOJYE_6dNT{9=G>)e7v2<
zVMM+j*K2C6k*9Jl>!yd*_60v*8@VzT1nFxPmIMzgZR?dZ$#hf!B?OdNQtqOyqJo50
zeUlsd*{J%QF-L}Ix&YFYDG?&3+qBZx%-zGph2h^eQpj<{STTI+spJm4siuR(9EXe$
zrM1d^`3D4h21!NPq0Rc+M;BzyIme0w*_u=?b~Qze3o@W)?S{xeRnA2`KzSk=v5L3T
zjLay?`V{!>?4%u#<9`17N)(g>9M^U%s;RKNGE%jGoKT^lJfji<(I{&fq9(D{n#*cK
zOwdi%`ew;i6gI)Rf_D`bPg`D@wUh80F|?^xq==rs{kR>*oMX<fudh*5f$18nj<cBw
z!{_u;a~`?^;ZQYYMnqQauXZMxrNSLwQZ<HqZ8N1OdT%O#N-WhesK`|+s8UI^-Y{1X
zo80|<K6yaNW);5aeD+;yTCrA^Y;X=L(-Twx(ALj4E59Bu(MXeIvs{B(Xa>-_@i^#X
zR8EgIl0u5LzgLjCGs+^?PtL44Hqka=QZ$5z+SWs@vcI2A?RGnQGehN?&;R;M@t8`G
zK0HZ$@7S9U3L(tQNZn`ztBOs@D^PaWKw_3u5tZLw&DuPR%n}wLf-vhS@Ox+thl=d2
z2%+vpp?f{A&n{Ux2WpK`=h298o?If<Oe%9-Da>YSM~zv<^RBGQ`8A5gwgWP{V>>8j
zW+}U6BDnwmR7@+pB4)XBTk;^AS;d_2aO?D)U8&3nXw#_Opp8s`%tAt%>dh*&HO@+K
z8xyOD(%X@*SZlf|HbNlLYz>-qR@J!5Y`&%gA_|Mm(c8qWA}Nh-8X03UE5o6xg)9lL
z)ZSp~YsEqna^7x*m0xJhm>Ki+=NGYps!TKncmB3AQS?$-32Cj?Y&~663nf|9Dqt6w
z?xdS4RtZr;<nBJPElU+h@}0%rysR6mP~po9+PHjw{*`5tdnKj1dvhA5gQ9jEy)_As
z<!h}Oe7{iyYpuvE=+SBo7m@TrZ<*hgs|wZdqFO@)`gwGZir4yf;Cr@)xE429H#QsD
z+TTVl)nb#FZinO;32-+xlzZ~*&cn2I=A4h07sOgKm{xk4Rpwen6ySfSMXgm(L1lPE
zsR`YihC|y<WyOjhlDoygH=71UQEM#E-UKnn@^&bdzGf0ps>jEV+tDPG$aTG5pI?c+
zwA0=BCG;c(j1+Z;*!TW1A>UtQ8p96Vxxn<bio#|^N|WzWRPO4ra;Q`~l5d5b#nuLg
zY>Fb0iK^{d5bH)!MkJM}l3I~j$6ciL#xd6zUbW-Sy^9z2?u615F$?KMwSqFr=wPO`
zF1}W#9!+oWozGXT&xf}+jm(J6uN7pbnL9v~_L>0-&?LPmD%?p6VxurBJhdH%WWbE5
zd9B5*%>&jd&ht1-RapM@*XK*SpzOp7CU55&5zABib_uXnwBsf@ui9ph$_z^F(?OMg
zCl8nB*&|tAm8psnf!**5m}&p`acd-F%=y<JE{BbgWIL7Vuh;ypGeB)TZgq`Fh}pUz
z`CZY<4JjMjh6$)<6x4*uS}Q==11T$Z#?#6w9RU5JRTNu0I9E|rE4Tc!>M&$x_@efX
z5;3BkO?-Y%j|k7oI?kBx&q|n5+R>YM(kya$rLtmWNk~ETP(6FBv6kJ=cAj+NdJdkw
zZEjsGw(~sVwOC7!jI65Ao-3j-0RZ(~`qfQpJ_ae2CU18^OkcB#LV0ho65^L5*7<QX
zLF=T1$4ZL!)^6{;vLf%db<OMRvDzIoP?(N(bkCTh^f-)bMMx$jt7)>ya%QZTpj0_m
zS3oj?X6?M6Tm~!mnIeqB>+{NIeEeqlC87Q7O<$i4;@uo4J&`UVag~UPn872JQ4OmU
zaX}RlDMB@b=!#rmlPcrN1*nKf`i7qJH_ap?$)c5KQH2FCDHRk_<h#O2kXdpZ#;ko<
zg}a9kqB)_h?O@{mE(h4nqpB*hK(JXx{*B#Kf+bBUnrTPQ>l)5}!&<eYOCVKqd6MLQ
zotD~$Ef;$?rU&&rO?<vaKnZDTXV38%-)fP{?z%9c+M3!7&r3}i5mDA7?{9Cn!&EgE
zFkaUzAF#Nx#%f(fV~j<s+VcQKMCH2xUSy|pVJ+*%KgN5DF@t7_+pSButFD*JKYl-&
z)O!5&9M1baeO#|Rnv&zS&Fv^`rfX7=C8pSG$n7@Y8(CjPQ{=8n&rHxXvPhw7%q{5e
zuw90mkh0lB0?(@4ptP+qQ8e9DYkFi-wUdfMdU7B<g2#;n1t%0mszsqt9-dfMzN_m>
z6;)_QW)&g}5mnkvN7{O_#fmJ_zp5Y=Vok$W>_v|PT00)N03K%6-tN{8C4ii-NA%VV
zuZJA|^#!oh`e_36e$CfKQ)?!zM+T)u+-@BnnLgH4q-t$i+1Kaikn`<q^3&(z_4Z~W
zeq9%=0)(VjeY@=DM7iABTmL@e=xsfq1*J;FfE$Z`tp%<ZMU--Ty=Imd-)_g#uUQJ}
zCT<}-ir+hjDuA{PN0nd|E5k)2Cskx;=KfCfR8b5C%fGjwC=|tBqKs(t+$$g<cT|-Q
zv6`N1>ISn^ux-rfm>yIm1k6k`J<eR@9Mo7NG+kSG=JrEFm{V;|YHy6;6o@5y%Qeaq
zB7sAF2WJK0rN>c4G0~NksHT;WMvv5X+}?it{&73HO8C6K9xEYA|NP^MaV5}x+^W)d
zv1B{I^P|lf+A7L^??&l2%QffqXS}`lqiM~Tuje!LZVT-9hRT1GNCEh`K0_)Y4Ufl|
z!CpjodLpjJyjgF#R+xxgGql4Cq|e)d@NuosPUfmuQ9dDHYXa-9%P2d<>5+2SD%oG*
zDA3M>eqAELZ=1Xb*Z``E8gqcH(v#iFrLhWUTg%;HsCJ6;CM0NBL7^H`!~<aMij#x_
zq#^;I+EIHMQLx^TRlY``-}XZ^2dZL}F|C(Jw)TBNT-P<e?Z3H-MA3w~LS(P($|Orf
zkl|ScGqTh5j4P!I8`)yr*c$cY$H)8ab|@p<J*_Exz5e{?zdln{B8j?wd4GF1wc}{o
zI_Frq{_CGV{_W%5u$=u=N<FV`!Bf|`{`%$e{^Q;d^SV;inlYy<{q^M_wH^9+&gty;
zd()$#e7q(i#9~fAk1R=#^vPy+?B<{`(bV2QPFZ7K&zW}g^RU_)onOxjva^R}oe`;8
z=?M|tkS7*QGVnLVZu@{Yo}AkvEZBK_`3{Mq#K>|p7KI*SKtZ6wBc>Or&=$)qQYmDt
zG^X#xT2~4tC>kXd!@o*3mS}<O1+lvNJ(D!+tutG@#p|LNymT)%jutCRz-H;-e%nXo
z9{venp`@tg&0<^1Jdazy-EV?F{%G$%ZwzNUj-w%7QTFjB{b0rQ^_XAlIeTfp-HxNF
zD09tM=s*7bG(lCi<1k^a>+5p`SNP*!Z~yooza3KZ^YNGder&h{O%xhQY5i!wN5m|%
zc7Hp%(#PxbkH3D+C=@GZh){Cx7Lrd_ChfR2J-dn=^ylmH8uNa?9nFOP*K=I>uKQ*T
zes8M~s8}Uh3CT(YtGt4?<JCZlsq8KmU)-CpZ+ADXR%WiEq12F-uEtz}vStlZ(41bi
zGxi%nMw(I%c*gSdeNe3U#?7Lr70oD$l^N+gj*NN9zJJ~BLN7JtqKRhYfEWtxIHaoF
zD?LatRBqNoLdUBZRbt)Dm_SDV`ENfM^Y!)5KaQjP_5tQ7f4%1M+wG8?<Hgka$Nv+x
z*6nTAA`(QU{J($R8c?W8k4Rs0Jg*s<VXgmu9`$-gj>ogTy}vb9&Wste#L$o4Hg9(p
zg{p{RWX#v+uUEA5eq)rEHR;+jr$2=7q8^qZ8KSq&`h7fKujeW;tywuQ-Yg=~t#$K9
zu-RN@JHii2#HLuVn2EwtTV9k(2!oBX-?(T;dcjnfu`A!Ck{?N_6egQQ#*P?fd}oA;
z8xK^a+c3H_`qQa;yKBYzdj4+}xOM$j9CHmw=IumgWP~1q%|Xpf@VGJ3U&FMh)~%`f
zTr0~<6{5K^3(Zt&jZ#xU+wDAzqN2C^*`(IvzyJGJ9#ZG~d7h`B#BOhm>Ercv!OpWc
z882w;O;AQe@Z*R0*!1b~yc~EwqEz&FyY;T?dR<P#tkzpQKHkgc_4s<cM)Lmg^B!|~
z7WFW_-3-3w_5AvLyaLDV)Oih86O>x6h~Q!2A!dZU-Hi!WWv->|Z0_zefKs`hNo}SX
z3o6{z@)?okt3-v|ip*vaB1)xXyQDTO6#%rC*EjW5r8fMCZ<;|7H7RICVN6P9Niz}d
z>t&*yZ?{8=H72a%>*4L=cCf0h$LFO4SasaGyk4Gr++e0OGZD*aH~Q-ZYg$RsrvBO?
z{0*jLGrv5FDnhFI6iM_$NT7K1ug})?_;|lHQLbwx;_*U8)_6{5|LsRR-+%tN-;9+R
zo}xP;<L+~MH52Tgj*RE$Utg=F;`P*)kwxprdB4BiYJ7gpY|T`CE^oJ=KkqtU&p-d`
zj|ckiKh>Y>FflVh`kM0^E5D&W8@d646pmrdt2c&wsb#+18P922v1rHH)mMa96G3HG
zUaK<n-b4-GHAb~YiZzo6FChf!s*$QDw7$M#GvkCC7TtQSf(N8qRf^T(-Kv^qvXV_G
z4BwdfELsyr0M&lHpZ60>_4Qm4YrI~*Q`z53YC&d2X1aferZY^J@1#6cg=@MeWK-?t
zkbZU`VtG+jwmDojaX<O~1F)tdq9o$^;_>l*zuyh>`ux0>(#wus&&PAD$1C;yw}1b)
z-`)-*QDUZ8b7iHEF|OhL?W0AC2$AdS^Yb}8X%*M#E%oOe&%fHwxAQni5o+59!2k7M
z@Bj86zn?O{{`Jp)K32cqJM`4b^7+Rf6P8~NJ8wPL24*G{Srsc_hRjS+L_nBn?dbiq
z`BgeE)r`v8kU&*LB`UQy@kt#M+w-etG6uNYIY@x2;gQm<RxwkpOS(glE4MYy_>D6s
z46zZfUb<O)nUJLFH=3$g`s2s@t*Mm6T)wW)r?cH*tzupq(oTemKWNsnD0Uo7(ei5`
zm)$H9?HI2sY}@FxsjZ1hwi{5n3Sywt4*&SW&qIo;C~e_1?YFlsxzh9fM?s~+BUk@P
z)~|p4=j-?X$M0`8hkGGI+U;P)=EPF4uE(!>@aTs4`n<-trYF#k<E$~uU$&Olb-Uf}
z`_*N=-#+mE`@jC{&-?qy^z+}}M+$hm9S!A=Sv<^A21StTDruS?dYaEMT}&mbGR0(b
zXW)@#B_BU?U6*H&6%jKE1u>B-<ecdaY+$3dCK)3N1%w>A2^P!}`m3KGSf1J%7P}e!
znracvBqG8Z62;$3WQDLO-;`WZg~g5RE9#Hm{_RI^2KV*#%VUg;h4Vb`_wno3BT*Ga
zY0MSJskt2e?NAe+!msCY96xo=rIrrqO>?9IYwjS&VH@NHYi&1|P}ggEu^mUF54)KP
zq3<_p%h$|yqI#2BbJ_iVGtTEfA0NN}_}>@n_4%as<6vrMR?O$W{`=2$+}=(bkFQtL
zh^UA<-`;Pve*Jp6p0}gDeKZw3UnBLrpZz>Fe*O996VlG}ekw8-?(erI<*&!{`M6p)
zs~DN{RrPumQa^6rSp|`VHEWIKvGQ!|^>Wqz<9A&fDy!wi@Hxk7Z8}uRENc;71#F66
zB{EAS^{6S9k+god<<d{AIhGSEMIa^=e%E;w#H7lNj7dZRl1@ptF1xCr(vx!B?k84G
z#8~s1R5@=ab9z2UOgRB}B!r@;m;xunnpAl;U(!oduX2x~s%ovV!JV6K57zP$(Sk^h
zGGgk%L+7mH{oe2#k1w(N{p_;l!hEd+m*3uh`{?KG_Vac$_4)75D~GecogJC$>(77w
z>p%7PUX)C2h>GV_|9Z)I@%{H7Y6pFcsDi!y=Fi8aZ(WYwB*yg`r7_?h+LXap-4DC{
zJd}CO%@Sa{tJ&FaXE)3_U(f6ESsCMM-AJnxAxxzfrP8n_)llV+P-HZd8j*Tv5~6T1
ziXzFHq7^`($Xo;<(!?uhqG}a2s`n$tRaLQ;s4*fKLZnLWLWy`KO>Ii1>Gy7#R1+gs
zYGbV4sK;5qUJbiX%9`3KsEU{~s!Cf0*z|zwp?E#dn`y%;SuaPkCYq9KPFls@-H8ZN
z8+TQGWw<t?TljGJ1!}TZ-f!*pafITpFXy=ak2^iCUsKNyt*rC@cE5K!Zrv1~dd$Zk
z_WQ^Ct(Py}%dcOTo{V*k-aUE#`>ZuK2i9S|p@fgWV6czd&$pFYIpgRkUrF_^<l{5#
zc0|yQQT{q_P2=^yK9Bp+m>KI@XH^w*J%9bW*2?8_s7kvV+|42-$8fashmY5#W``m3
z1qkN8&F|rs1}N5AYPAv|&;;olgsn}*r%08=YMr&#aPNe#l`ImCq^6euP$0mK2aaWK
z+EER;u@)-fD#|YIy&VeFaFfUay4_BxT(9RfJzxzG&H*)yN=~gYo<eCtVmx#E^J5iD
z$fgm(%79i}&wf}ZDwiJR)FR4SRkkyS-CDk%k3TMXU1PcS`;U*eni^#5t!v5A`S=p+
z$9Wvhox*gxe;o8!bB^f$@sFfP%rWQd^Y!uD|B&VR_@Xt%U&_q5o}P13n-<V+y-7|O
z*LnrQwI8Qh%^7EVJR5M_n_~R0zwGVfKvD6pKacmfLn_x8v;T~_<hS2$M{lCCJYqaQ
z9}vrMUxPO>gRhkp1<75b6-gD6q<a)HNh7|=a5+mzq$fcN1j^&~9m=lOWL?D!kGTRw
z=3YUjQW+G{=|=eCcR$R=N@~@ct!wJ7VWoTJm82YRhxT5+#>)K!D#EimL0^?cYqWVm
zyRkA2mBUr}H_c9t4NR((rXf|izhW)CxGroJTrGIq=P=^>HP=d5gI#}2&GqtJ{?*R6
z`~Q!rzgdzaNtQK1W)V>}cMkwrb<XY6)YL@t{x3B|v*=!2>RUgVK!m%QiU`j_9cN};
zC`uqa+)P!3`TK5nL$&JpB2XIse9G!YC&YkH-dwF(=hx4F{PXFN!_05Z!TR;pfN_0c
z+<urc>w4Ds@%M+lp62UXr^H_4m_t4VK~4Sq*Qz+~|F7FIEi(-tdH(sF@FR|?<5-v9
z|Nh6v?QoHKetllopZ~7on_l(0E=E}9nIH$$K_HlH(QF3<O;FLs=%!ZLEqXPHUV?j@
zZBdrfAoN<tW4inK>_y*$tZkEVL3T&c!uf~oK+Ai-D|mp1y%v3LnD2dD>1OvIL*|`i
zUau>`x@gyRyQ#L#PG;FYj_K?wALnybGsf6$-fM-OCoIr*Ot0(B)Z9@ZY~MwMaok+j
zqRv+~Gkqu}`Tg!O?)Sj$`}gl3w-Khb&gZXx|NGyc$I<rf$M<iyIh`t1aSZl4U(e^S
zzyADMwX(r=#xSfb&fC!^;&1;kxyoMu_?6r{LOibs5ZAL{@%T9KIx{=w>$B|s$N%=9
z-)@7t&U2lgfBxhc$FzK%F*1+)w~vq82sb=`{y+cg=X1%%eOl+bvVpFKS;SDU0!5;?
z?xo@3eQr#n+bS_4w5m5qsx2H{<7nM)xztz3yxpeq^D`wstg6k{m2%UTd8!-%*&u{T
zRl&O?=SzM=y=<T?lwi~PbI$W>8zBI$YOAzIzZ{^MZKBOMLb*YjtGFEtyYHEKQ49lv
zSn0<uLPwK+zq!umwbU!_R#(!#p2_+67R^EB>z{xAtpEBSe|-OT%y~q3L=d?G`3-Sj
zna6P-eRZl<wz@0xy4JeR;{AKRzFyaPwt0-mbz$&@xPN>ca~zZTf1c~>ufKj4ZsU^Q
zW<N2)TKUZ2Bal9zh7W&595+W@=lM1Kb-n&1t2Ksnz1H;{fm+w|`PZ+7+ayc3VcpMX
zRf)aC^_tO(l-g%(-?^zN1I^YlFHNuW1Q|xcR3M5+z_eBsbxC1kv~K=<t;!KjRkzW?
zxhnIk5sU>RNQN!Zf=y6>^C~ksY1!2c2X5c;MI$$stzjlwpi2>27tI3wY0BCONApd4
z*R#kSxf-=D7&|p@ckf))Y8zu3Y^+2#`s{2Xdsr{KTMV<nzy9$T?)S&z<2Hw*d*`yQ
zuRs4?{NwM}YsG*5n9B3(T3JHp^XH$ZJUZ=o+@n4(U0<i-2>d){Q_u8PFOS>dZujH3
zKR*8T>q3k<W383O`_1av-yT11ZZy8WK0kZB&g(v%IL4emzCM5cdf_&sRxUYox?P{2
zUu_@%wPR8d+jEZRl{Qrk=scxvfO(m-%plVutsC^-0AeOa1lb_GRqbp6rIkP<+fY@0
z<mFWittO^fi#9^nOD5yTF5fp$3^(J-&Bu0AHNs4QQtoCRqt?o7_LAG)G6A(UH$^+f
zwMkpf@XDpf_*yq^3=!0O@(igpkKq7>wGgVTW4M~G>q;7I9$n2z+`d8IzSZm1kB{%)
zJ`Ue=br_N7=Rf|}Q}>4f<J*mWU29J@RoD3{c4gUo9LQ^RzMik!H`cGG0H1wj(}ojs
zw?nSybN%??=KVUeXz{q!wMyf0AH7=Z_4yk2?;mr9JLzUXJ%9cCsj$~)y&>1^3qEci
zN3Lo}(irPp>+`jL2<*}}Khm_?He?#+%AyBBU8^~~VUU^m<_siE*>8!F0h`$hbVCba
zE!mqpXWW?fi~rGhn_eNh0W?sOh4uNgfao4E=P@CV&h`5B>`KHg?_O(*(85s(9up0L
zJZ7HP3g=FQU?X55aYxP6fZ1?#F_c%0*;RhGypm3sG47AM)_Bx84hgrveZSp?6Uw!&
z^XqE~o!5B*tH$HUAs8(jbzLPIjqAMHn6J;1F|Fy~h^&78#5cQE)=sJ%k6W;FJ=Y)K
zA)*?`eVF0-LcV$)162(+Z};1D<CYVVhV}LH=hs@V^Zqa?uQq`G@nB2NYyoq8{q?*S
z9Qj`8kRB$Pc^3ezH22Pi1z~&mX0gS>9^!C=x>lO4D~6juWGUipc&*-wx$NMTl;$00
zW;}i44@fFCKuA?`J3%?5W-I1x%wba5Kz+oS2UQXVx8+Bji`@+$s;dyU>jZB1E5o~y
zC#TaHp1CDdcAaij8pkXNfks7~z_`~%WMvJX-~RIs>HPfs)#HEs{g011g6Qo0`t{fU
z`oDhq@Nqlh>#H=|8#V&U^Z7a#rokt1Gn{|^<Eioe+w3u~D~7LA$X{mjW5#-ZUAq$3
z=HSmi0nXd7?X@p`eXaH}A0IR3p6<W_$*@lKx~}u}dOc5#xJ4YH%GXz$&h0Z~m@=ht
zYPfGvqO(`6Q$~zK3v@pw`YLlMyJ5FUsZTQY8eg9@w_2%j>&kP(NVp}BK=Bx7OKh5<
z-PxW0rqg4SRe!@MM5t5lYQciFV5IeP#ipkcm8GVUgV%)-8`IFeO#^yD?GtAS=JY@V
zS*FE7c6DPnhjGM!JZ$)So+Yz!8*Ef|kkH{~U+ek(yLTDeZ`1PCxzpG3`K{J?R##qw
z_Xx+=FM0Gb7mnMM*Hw%Nf4*uAivxvK@qpI3{12##cHH~^?JD_NI|H&x>eo5!cyu|c
zWv=Uq`|H~`ugrD6dbrcb%&N@my0QzX>vansrlq{}Dkw3mWS!aOm_lZ@LvgFDeC2Sl
zy2Z2RtP~lemaLZ6TTeuEzj=h6ve~Y%N&-MO4WdnUmO+<MqO}+nZ9}5?zpDT?h*fY`
znJwLCt}2)VX6}b2rq1(RC4?2mU{xIJ1-f<99K{G;3u@UIIvWyw4oK?lh}r#qZ7qjW
zh(U=%*`SSa4DZ+1Q*Jim%-dM!HU8tr$F!>I_Qx^Z(ZVr}eg5@7|F5qZzN24QUnj<q
zr`-I06JRmtF|Gdf?^DC7A&p~3=W0f<U!Q9SNDZ`lwHwaYM#0FB$HN3xL3y1^)$?{6
zz0RxscALfs>bwN4*J?o!A0Hn#Gb5fqKcB^x>FjuXV_7)X>aHpqj}I2=d7?`j5}2SY
zoIZ3l(2WSwhQ{pBsG!=0<t-p@sSDQIF5W3Z7Fnbe2J^7Gi1?k(FfrRlHRS5E+N{X?
z9f&b4Wj*ixdM+KaN@&=+Kp=VeRV^9UelEay3D)3kqRt*py7Rooab{S{Jk-W(U2AyN
zC&%q>q?w90>UGV>AKwVGJfEL`F5Tw5v#+Z2jGKW>sn=S^fBuMJ*XOS<pYDH)Bec#d
z*ZOxg{PFkMRj;pKS6j?Y(6iw48Fqc4R<=~-Mjyj^6(m2Z$3P|2*Yl@r?^8Ic`>fNU
zD`n!~+thM4c4)~VU8*PJHY4t|=jB#qfPXw@c&^N<^?c@fVe5u^Esc3|SDq(haODC=
z1T&p$ZS0g&Yi$95bsI^+ot3wT<yz`9?{OwQ(6(Dbo#NFcDcS?)HSBEAD&rPjjm{=r
zs6~nN&dO}*2)RdRHCPEF<{(CFWzuc{z@?(pSm;bJm)zNHJ|NtXaad)I5R6_`#ltY7
zitd&rcIkG%d0wwS|CxX0w>fV^l?4l`>+8&hjlccxp}c;LJdJU?O-F*hZhgJ3w)M|{
zVcf1)Z;FSTalOE6`ykd?7{iNZhIAW4D>){1WFOD3%rnj1;nU5h&+w{yixz4-pEQ5m
zV5r)9W?btpaNci&e0*HnbC0(5(t`R*>vf%rnM9*i7!-$f6&(EpRA5(G)vH?=yY_`Q
zn%)`&x<<5&<Nn>3l<Vu_+YlpZBj?a3wrTkLK$qOB+6U@*9NwimuO)Ud#_*WI^?Y4d
zTexlW6!H|@)x&bzX0q<zP}RMgrRx&HVTN^G1;HIKrhV{f_j;|e2(w<isUCkDem}Y-
zYE@n5*Ev7_#}APCOn&?O$9=QjE7y6QuYdgdAOCd^Gh5HsI$vL3U#Cd4k8o*>^{VPT
z_fqREhZSnAg57Us6X)v6U4NfZm|p7=$Rp0n-H=ztT&MZqDwvPk{m!PY4o5EGb$$EW
z$8Co7s(SuhMg9Ct#yD<wg9q#D*>((5cR!zGCW)8^_sX@@#i@m0XKh`O8C7PRtx!!L
zXk%&f3DNzV+tp$WM;yoOM8D2GUMAXhG<JJnGGmo=ySb@WCS7MwbJtw2vh}sDo1x2C
zXZua9N>D8khh#xrmpStCdFxP2aG>nAetlJG$G5INsH%u{(Wsvm{K^zjj>aV7I0T+&
z={SyKI_mYWf0q6Hq92dHkGS1$$8f1~U8ah-{=8pz@8{2d{680oZX8m{b!Hf4-+o-H
zwn{g`x%r=v8_kDzpS~rwc{OpK!@kJ%f;*d8W}tYB_!ujzqpy`1=W<uCLIkzW)$25l
zaE<OUuXNlVYn|8i{G{lH=d1nO#}Ttu*p<ZMF&~5K2<Iio;&zEqZ8Nqxdw22;FxM5g
zx;zZo#c%>G4^t(T#VXDi4tLPlYsL8dbnEWn5G^ve??79ZK%RfX$DH1Rn~iL8W5<kB
zD>33)jqT^|esve-217EK=^qbg1NAyPvrDxJRgPBE&8=59&2B`uxy3*=u5_-8#}Au+
zzlB>~=j-$H*YnSK+~)m$zkl2*!R9#bmGj5H|KI=hA9nqG)~fEt<8e&ubzbYDUFVNG
zVGN<W#M^xvRg~40=d(&O-`HhWz0!y{7+q}w-L>Q++zmuG5!U^aHs>y=`HV^hR;)F)
z*~83-Wma|87xq~<c=q*cM9g8cn;7Wla}1;LI-73{5~;2pvBe_0f7x$nWu4$|=E-CZ
z5ft2S(>3p5V|1}bD~{;b>vlg5VKwLMR~tKy$h9sAK4Q9^UHv*qddvu)(RnTUkg_|Q
z>b%T@%5DnXx5#xIPS3pEu2)~vTYRNBysk0Z?zRxlykIc`dA+K_gyGf0vWk}LI9}JG
zpI`Ia-@o5x#Po=8e?I@8)cxBrzkM9{$6?NKbjEo7Pu54wzkPiD^}1TG#t8LbopLUH
zUS}$HWa#FP<H%<spVilE_3*Wdv1e1YMoLXFy0>(Nv_@Z5Tff?L?V~6F0Du5VL_t)$
zJz`OxdJDhD-GzCx{^$Rq8qC+3xB2U<;J2oeTkI64t|g2cIu%|Gu~XUd5QA%7@UVg*
zmw9ZVo7qZZuT*s&M}9Wxfhv>CE7L~M+!Lev>uG#0Oa<(2b$?8ri*B7+y7+gIQfU`(
z<$>xdxdC*$Jsj)GErK#rUgd5^*4CbL)0R<<((QJhGN3n_p+%M~9N9Kp-5}in)m3eA
zn=5zSmtA{~)`;Wr{rksb#wyh_-E6(YT&4K<ufKmB!<6UeKmR}fcYXUm|HDzwuh+S%
z8MnvCb^iK{TQINZYau8f>M@VO&T6O{C>)_emrXZe#?t85Yl+}yB^jzgo-df2;B0zC
zt=02zA7i-1FrwG<*Grv@e6DsEkLe!AfnGI7Rp+a`&eej&@J5(AGmB%;25M~@Kyy!f
zyDd-vIG81Ov@tK)2#_k!_S~&uh6JGO@M?35=@b|dBD%YweDb4Xva5IaobJ?{>f7Bm
zj<8ypDK|<M<^$(t?D?Tnyq1sQ*4MRtpEY<cN5>sY8h&0*3j0uFRt3%xsFif*Dhr~|
z(;^p9J?Es3AKxC2LvqS)_d8du&tJcO{mhS#$H$Mu@_POHzyD`G{_6)l)=MU^o|h@J
zvd5^E*;Nv--FIfYSLJp2=(335?o+P}orymaZXCx+<V;kOyQFgFzRz6cN;1dXH#0-b
zJ$I-arHELrNE;^1+(HO*o3YK6pCe;rSS<DPKRmDJ`97cT`}6rQztP)<`Z9Ga%%tVw
z3Li3kY`#7$kp@_ztSgZGKT0XJKN!IS(W>sFaRONE0)Jz|-?j#%x`}OYfgxr5@yE&X
zMnV7nwE737E%d)U`SuYWoY3PWB-;S1>qGXM4v<CT)>yB+k4eo79vJ1;t=)#X;jOBi
zGJ-=#njnxuPDS)FejPXsojE?<+ETB1j^$C+-D_=eomaLU>OHn9-7SC9XfAFW$mX%x
zTrJS`LL^%@xl{P8sk)u}><%P}gE8>1)00CWY8>@WBq1U7OXZ*AO6}F8Os0mj-(Fij
z$=g}O*Fp7x<e%dyyU>S;>K!o`+n~KBZ~Zqr>ZUTX_82YDPle8-Uu**#qMOqEiiqmH
zGssVRZa**NmLZ0vLiNH;Z6zs@-MZ5=|20u-<a!qCHx7OOzP-{#M<|c@RE)HS3l2*l
zb02jN&xdr?wh>ovq5Tmzxg3q1-)X%KX<`Qc=u1?C5iBe(0M}`Kn~1S-H^P48VapN=
zRPL~?LBUr&E7K!Smz;M_vHe970%DvQ{c^hwDtkf^s>ZM%!1C|@-jjQ&t}8+#?))mK
zVj)1LX9D3HXQvrdyHWBe%jx&&<F*`>UP2u|g|Evhz(w+X5K#jA<(ZyGhuyTx7D%uY
z+Vas3o#iFadz5-3AP-XB=#<(|X|nC?e!D`7oD2=SWPg!<a=68+_L3_utUF#-z!CvJ
zx8~G15Oc@8HFTA$5ri9<14m~tMAmVna@9$&JhuE={wOx}!Hrndk6B9Um8#(Pmu7Os
zpX~8}YxA_8+X7mn!I$_3F`_vh`FN^i*Z#UaqM>q(D#e>o2Y=6O?IN%xG8r%lb#beQ
zWNUcoX2*4&FPEy4e}(h1vEiFf<KvVpez)-BS4$Q@f~>7n92$0VsqNqKhB;L)1Xiiz
z8knUol}0b_97ZS+vprm>vSv3P^?Xa?vNLvo&jqwydV8<6jv1Kx`LwQ4IU?FuFZiv|
zQMk=&pIsRCTRqO#;b9%wJ~BEpet4i|VO{><y}Z(W94Y~(v!LVoCSSvs9}B?0(5Pw1
zOK#8IR4dY5OMdT?Sk!qI6^uyxjIjh4S_v85t|bh5pg3s3ZB5VGSZsb$MC92#w{=T3
z{1-y4b@@H4_HrSe;DP$yUvn~6NzI@C^fXtvNoP>=WL20rO)?F%yA|<hZ=%>~zlJ#n
z@Fj~P<st@|L=-xHHkBc^p1DCi+$Nstp7zkc%+fT%Pl39>GY6Y4PhQ(L&*L)XqaT9y
zBKS*X)<7)CZ_EwT?HCYD%QUB8IIrj$+3-r{wrwyoSLw~|+%oD<v9L)v16gz0P{ILQ
z8~qG?2S^GU<6#s__^XLR5-eZcB+c&x_d2hdF&R^b?bL?kKWnX{7oE)od@96jpIN0v
z@YEAEb!`)Y4tJNgTm_3SDtWnHOEc|-t<8<NV$9NN5>B5*Ypv_Il)1uk1`1x|TN1E*
z<U&4?{GX|tlXBx>wRJd(udgagB``3wLDM3)?Wgr@FzQpg>GW=TRQt1E9j^GnQ`YWi
zb@GPZ4&B46hWuOam75!?5#9Dl098B<cl!raG^dqqrKOJ%4WcDlJ~KgNAFy!%9$H=_
z<nYWy+XcObU2oF<3TTFFv*bOo<S3TMU<-DV_YrnM(>nzi@mCw<xWC`jFBS`6h?r>3
z#yaxhA+Y=J{#|Xz{ZhnM*6*_vEJx<gYw6!$JWq&1b`m$%T-lyHf-(iYekj_N-r2X)
zQHl#%>^_ppH3_$YaIOX=kU**JU-<mv8{O_<i6MS2s7805O%qO0%;)SF*LmEjZkoQl
zzGawGr|Da2L=7H=k_i#)b6_h{9}xrB!(Zoq&DqT%7AQHR@REtNLHE2hdq>%2XYu`G
z=ft{^JIj33&{XQ5L!hrCZw#8z(Tl;Rr>_)J&MtWC@b=O+%SVk#9RC*`A4U2<dP*k@
z!7e9)-GUUhYqh!i;*k%5nRKuBTZd4%IW<CECYy)}&LLg|#dVE!#}C_&KK^@3ar+lv
z2R0-uiF}9F+F;cCY{@Ug-LhC4x46bapQ?NAL$9Wko0-~vc2aa2sUg&QP($K6H|vnz
zd}aN+U}YBb?5cJjJ`>xZ2QhSIZnI6Qg<gwKf1t{~*`z5y$Y*+!F>@ufJ%MzSR7xK2
z{(v8Dcs`|S=y5fPpVv{_vlc3s@F)1+YN)nII{6symT-EuOr(iV{fj`=p-TKpG)&PN
zXDZG=ebWM)ZE4)FGoaYc#bw?$U>VfC5P};R7Qi}slKEvzl{jz;B(8bu5t7vCa{^~S
zPq1yjtAkyCK{kaT`BU)|))SJ6)0HOds6#NTx*MUdW{#;PdTyQ}mFY!-b>m!4yfAuz
z5>hP$V4VppQjtk=eQ98w6_Doy*vhM1w%Xgn$maOEtfTAu4=$pw@U(th%iekU?gAoF
zVZhyOBW5wjYX<sLp#*<#$t|A$#bDM-ZrJ3CzJzsPIrHH>bK~kRW`i8P4(bsxy#`*F
z9FHjZ@S^m3?bx_!rnYq4pHPygZ8yU|X(lqYWL#!(_xYe~&EP0-3+;+}p^0jOQ<nKD
zrY^}n=dXuD-EIN>%km4QvWeeGqmUpZd49kpP^SBhvRn7CcsK<miEi@yyWZ8%_L}g~
zOEll5=%dDMav@4eAywBNuc@ck?Fu8|uU6QRrDX@;paUw{(a5bZxzSA3@1Vcc?G|FB
z>dlwQlzRdKc4oZ#jH<Rb#nP@O=Y8T5v<`(JZ@)-To>?ef<m)t3II^1KQmFehv0T)!
z679n%AzI<O#shlkWt(D$&oc>W@g4?65){g?P?x=}oD1#JC~)pFe;=qaizwSWxT;&7
z_v5UG1VtA|4QCIoT-iBpQ;gQ?9X&YFeshUSg1mEXz4gCoi5>1*L1n;}di@$&`99$A
zVj+WW@qNG$7Er&n>dI#Gc{48vBGs|hbbi@cqFO-Vf{%zZ>zin~7#cz@^K1Z8nA{+5
zC<SHqoW$r*0)1RNa?~NM)l>uo_~GXJ?&o~5vkiWB61;h!lOPTae`~a{e_kPUC$7Bk
zWByL)>j<-Z#!TIgyS#pHSQ6=M9#v)r@*K{BE|=3+H6u$kUzIzKtj4RK#9So-i`={}
zjsA+u@=9yncsLGi=l<Z=w-1n&hrC8Tq0&Y5C<f)`Wb$#0PEo=?t^MOe=-IOm$#IJR
z`v)ut9`H>?J-r-cIQg|6LD}ME^=LyF<!P;4tbdZk(#GD0T<~T@x5Xs1My?ywT3O-S
zR3VEk9_NXeu4DSbaseTKkKf}JaYHFS*jn3@lwEf}^@iTR7#uB_05bd;W$i)ka79&0
z<<!18&C<ZE1e<@>_nl1gD5S@DZh@632`6;56`EY<G%V#3wkH4mQgQOqZzyDX;=^9r
zt!f}4V1$o#a^bM87kA*%EOAI5fXAz&uI3AerNtf`QZKKs@XUS{0*rwh$Eb(Nu~-c{
z_2c+p-(xPdzT?8;^u1ND+P3&kyQ>d;3so$b!y=2{O%zx6vZ}y1+v8=K1TYnff>LwE
zWKCbta>sD-pCcHPaAnsA#m2xM;i36EoiE~}uC`w?;=|T^^MDz*xyCc~n;SBBXx1dn
zq;WqBf_gUjQ`+8XdRS^;r)~m_FQ+wO!G<y&$6fRGTDi368c{CaJTU_jNsxbUkDNuP
zT-x1&hMhlbf6)9Uh-m4#*bw5AWb$h_NjOmDepX(1lHiw<S!LBDd{_Rn^XeQPiipaz
z05lV28^D(?VUk*LG(MLz*I=wQ9x2f(k&29_|5E59BhEGlWWB%f-CQ+!Ut1xO13Do1
z_{PTB0DO^m+pG)^vSB!SannHTq%!2fLIZZ@nf!x3)LIuCMPXy$X%Rw#bA;WE-A^{^
z8d!5ns<0h;K`H5!YGMR<9);RU+=IW*w32LGTm2g~^Ck?QS5I8|7IT&)!Gg><eGe8Y
zOMB!U40fCrdg%8#Qqdv4;OXikEh0%WX`7Jwf@Fco6(>qIVi>;g_8_u<j{dV|YB`Co
z5Q8RptF(-|cj2DaT}}t&ssAyfsqIE`>O%wSR|banNGu<8T(K9I`nSozb&)FE<_jW*
zjw)#w6#aKR=w;nGR^nFLLuvzC2YTq1o8h)lehM|<5ZGUBOegZZmlIgdD3T+5sDoHz
zkSDOrr|VFU82mDZ<<)jj9*&E>)LbT>2rB#uiY-c=2q7cjon>0c_6CidRrwlExTO3=
z{{WJRX@SPrnb)IXFd;rip%72jcJnh+bzC37F8P{3+M+jXHqr7Iy#g;j(67bGQlwY&
z?*@PPaJ>xyyWa<}XJ)^RWjoXVIVy;n$3aX7YLDi6Yui;i5J?c4R0rTKaXo!+wq7D)
z1d9vq>5ZS&avkxUnWs?lbqaz6tCtp3Vdz3+c13>7@lw}9*J<d>73kkTWR%-Wng%3t
z<M2e=(XC}cWH8ZoqqQMWr-a!ds?wdOih7`MVP)GHxfGh0){0;T<Xqp@iBb8Zmf`d&
zLMj>7^Vo@5ywDicvca)iua-#RmmVa2t3VLwJ+J`=|BF3wY0o1vspMYKvyG=|rbp-4
z_;`t1ApCvHw(7Z)KQ-&K`5Y@~V4vg5F$l@`D&2YCpg>EBjsXU9U;UPJ4#<NBwvYX~
z{W0ZsPpMqIs4dz{f;EP}jd0}eds&0_PrT`fvr?g=XdN!^x#J$lG#58QOXxBvnq*gf
zE~2&utxMyc{`FeV{w3`?g@Q?+KlS2|btz<vVQ~0&LTg;d=yfev1I|XJVQORV!#52=
zhw+WyS(B7Xhe4A--<5-E=_jRJHRo<bHyw4&-tZj%!h(u#B(|f6cIM)5UHX_Pr!-MK
zP@Bt)@^LyTJICrZq-~8;xO=j?N?eV=234YBIze!^l8y^p3%BxO3mL9fMQW$*BkIzC
zxYE{~xhDmD?(Uq`NaodyHoKzd^k3(ppNvGc+~o>?3F_0_n|EwwtQR<|oFxSM7!ClG
z=JcCZ50#%Ujs^v1iXrruAfgszIo?jqY(q)gb1phvg8@(Q<kWz_J3Ol*4jK?@l#rFP
z;EBZ3v)#GMKWaAxOFnDIDskg*2`zz`z~5o5=QH+3d&R^(6GQ+(k-=#z+G>VZ4wkA7
ztC=G+t`ys+`>;V2{&Zm5t|++)BsvCH95mY&YRn#~31g^2UAdj#6iAy0rBR2CsRC$H
zXdFr0lk>2N`r|m~5u-wbh^Q(55cbRtCm4}4bATYd3z6zqK=bD%gZHIAi?>}E4|=`R
zgdsD!cpCXgdoH8juh0Cjtk>cGL@3X#jv!h-YOOdJJ9b~R?L5q1ci7W0QRWTu_)fTT
zP3eJCuc6CTMuMi7U9<aigSXYcg6=Wm=z3>~y~@TZakU^DtLcWazdGwTrm;YpJ>Gby
zE~K1uolF9%KB*<$10XKF?cZ|-?2i9unSOzPw}-jcZg7`Nc=5#%Zcm?G*6^C=%|`!N
zQoD+!>+x7YL;EIW(;_M}S_meq^J8O!Wg6YSyuB1Of89v63An&F_3QWC;m6@|H~Utm
zb7asgs8x`&VG|j%dcP$FAe%!7DL4c-x|FISei})8if&#{vl|pU!=kB^`1Vx%Xb_;Z
ze#%;C^F&cv#VbRO-uaJjGyA2TO32Vos@B6#c|9D-1j^tqa7B}*;C~8VMVn%D5?o4{
zhqNI}B4KGtjayVD1<|g*aJj<KZbtbfM>_SV!T~m=+`jO-BysI&`DMQE5uYa3AbcXg
z3~fPge-;6eL?Zbq<kB-xZF$nI2AvN4xw2D^L{`A()4iIb*Qin1^{BXiEnJ7xTV12%
zHbrckm-fY6xV+>|V#{o_E^#1>|1!szYb)PlpIbETXOEKw6#F7<F_&@0T3suy0!j`Y
z&jW+7=dzT)^Upf6b7P>d{}3a-w7yn=&g%LlQk*9KyLgq~L%k#<mjw9=<4#L&+#bMf
z4u6Y1&$|w>frk8o1gu1T@Z^ApqaU3N*5Z*^FRMPNtF*c5Ab*)%7pM&psN+_!6pU(2
zjYj%C|EM9C#qY)P;Zd0QmPPjCck_Al%m8RlUu4enCb^)yE^90irv6W4al!b?*3Qru
zkYQZcArVkd!(Wor%O0CKbuiD&*}b4+UC6x2TQ@xjeKGj4Uz<W(F*!IUFLih2-)a+_
z)V3hE$(8NId%@XySs!mTe>S)ELUxvMd7Q7mhv56ZqxNod^5yLD*hKYLNXF9AkP2j^
zjsTM&Ii6bmeHa|<XV<r2y)z)!#x9RpT2{k22fxv=2xdAFYK|_zNpl^}zMk~@R%NZY
zjtN%V9X?b1=}7UowD50ud&S=GQkrOjW5hBrh$Uz=ca+#P?q=SDMyz#ReMf(22|eG2
z;LsIX&}Dt286<}ERBz1s%=aybU$Ud5-#3sb_uhL1JpS*f(*F`bD=f~M8m~8jD?L0u
zOvbehYwP~~9l>E{X=8(;o;?BH{8}1!%JRG=h3dPE#5&CDUG9&d?f-2ir{m8x!sQ3N
zF+dMNS-0MSYJLgZ>BoOon_buI0KERYOkV1<QbkU5e!HhvJ2+USb^l25{s$bBSaKX=
zZ5S2?2%+H~AGz(;psJ(2jWWf;74!&YmN9@Mb(J`nk#ZIgQyFxxwg&p^Y(-C5IzdQv
zb3ZVu8>lmhb_>0WY<N6Rx;VQ`pt$&BT@%X?>T!(+nCYq3wK<4bKzP&gZ1j2vc_hcx
zbNhh_(^LEL&wqy90pm`7Ro@|Pd%xpLtZ<c~rxvy*4_*Z8bs3*s#3e(k5kj;fS2s65
zMyQK<X#GN9Eg{4Nsx<KZtw!Ay@2(O;a9(JBti2d<XSAkHwI;Q7sD{w{Hb{!<`ey}#
zPB$slY9qq^w*^zyl6P2C7zt)Ojy2rd3+8^o9ZGE;O*?}gb3jTDK`0IGdt;)UmhADy
zvvV2!AI1wyA3X~#kMDfHiOdN&P;l-3lcXIMBe|`nPo?E@mPQ75n{XiVd`l~2NnxF`
zH_eLbr1slb{qZl#BDvGoXXJ(^BdG(}j4V5X04_q!)Qw~P)t6xldA^176T51L8+n9B
zieWh+y@@hF4~LxKl$GNXT8Vwd*i(v23t))JnD)<A6WOki1OIv>^ZSnvPwj(zgs;oA
z9@;?)+f=<Q+&_Kee*dP(uznW@=4^kOy2f!^<B4zSb@5kt0xphCfz!ol3`}P^`ujX(
z_4~_PDV|0o(Ik`~fGPG=YXcZAkR?J5iA>=nAO0cxMYn86h|c1HwQ!TBK-BRNM4?7~
z4ym2#wM{N*USm<_c;i@^gZJ^O-?qE=GVbTjkNjBK9E*lXWYe@Ed-_QhR|BA@k;Uh)
zZVl!3{)Y|lJ51vc*|Ffu_;~yGY%Xx`85mH+b9#8F`EG%N38_i=gB+y&in+R48xQ<3
z_a+rB@Sm=Do8#?;!(sl#5$(&l+R0*YeD3vzFC2FqdBF|FZ2cDmy`C0Dwqb_`>hqWr
z*OON)c&s_U%b3f1cO?w^b&OZ9Z`NjhY>UB*!h{BeSiY5x;qhUTCl=8R>pV4QUUktW
z(ZS+)-LvYWPDS!BCT3O-Qpy{s9bBij;n*tvK@d~x{J)QqJXJb~5nG`XGp+2dkBcS7
zhXwcmFT&W!?tgkmF+Zi=m!;T`%il!R2q)RTWXyUzFU4sp7xbI)?z}!)^<*yM6_YZD
z=}7kAX*>yHT}P#!fo{>gU<>qMLD*bO)!Bw~#$|NEjLjoA&bp}%w>Oce1;rRugpIGL
zZvct6l@0N&IAG6KT9*4hz3R!(qAX>p`mzV+j~31T%VBd^>mlEfNi`jVXv|GH$4u3)
zyCS<u>x%>8T%`}k-?icZLVYs5^Bbtl-2^>P&OvA`ej{vWfe@>`(<kO)A9>_U3#?V|
zwd))#i-`=o{bZALIbjNIwyx!T<FmlgQjw#|q-5HvEMGi+qkhTw?PEeGhf)&A<|ccO
z&fY1Eg)nW;Its%dY^J@I)r<f03)4p$!UE4=F!D3)O6l%@1v2V&d0TnDJ;NqZ2PVdm
zp`JZbXbxkokjWqg=Xy<=@e&tEedTg1?_Fk;Vk96jfi$zwx=Ww)h<VIb0p2xUm=~Hc
z6HQ}z9dzls;W@J{Tq$Gcg04Yz^A>>{iLPgA_Bz}HES&sa-IG5&JRUjAGLlbhKbMDt
zogUt~uv|rc)|2+|mX;QF<Ijl-_fmHLu#Cw_t%0nFgJW84IvYrqDx9C&+Mw{G?S0?-
z>;bD0=_THbbDTxrKYeK>)h^Cg;4HVVefh>(VMNoFUQdeYY<OVU!xi20arLN0m=>Ha
zn1yp$$Z|!M=#4k=s83tCQ1T>|g1An1M6~ZMv40xF?m*9tY6R0S=5<hyft+83qU9n6
zcwT{gig?GIRg#xRPlq!Wot(lQR2YYD@9W%Mh*PUIW5nv*poo-R{1EWq_feHiR&+Nv
zQzpvKGqufYoopdmMf$W)#1KP^hTFQv7B|3x(>TQ2x_8aBo#u&8P1iI?=ffI-xE0Ik
zWCQP!dgWsMtF5`ET8I}l?9#2fZsv(U;*;LCz95}bbWcDYFMj6`0et<CH*N7yoW0Ld
zc+&nQL*~vjt`inVs$eU-8kQZBGf_2&YY4ue>GKVb&x89g2~c%TmRzhH9QZQZp?9X%
z^-(tiyfS~H-f2%ra723fqU^aU<twIJa`YBz*Ds{5$IeJ)y!y}AzX&?C(PA}eUB6cc
z7?Y~NN)n2m-Zx}Z&@>Mf8VnUO8N}NIn%}ISkpcDS8O!cb*GyvKlHb{wlwqAB(s#b8
zZ5EJcOoP{tvGb7_d=^E>9g8V(;?jL6<F`wPQty-Y`s|@sg(ZS&{a^hztTAE!cPweT
z#wYgbZbcCZX}C*<HJ+QSRCh4CN)z&Uc?bSkzGp+Z1@&REv%e4KBFpVZ&}S?(8Xg>#
z&*?k!0SDL$8PLYvV!BD0>TY6>PhF=98R@J`(E!oNa-U&xgS8W9pn86doJPyiAFIjY
zjW^sthO}I-;Ph*w6GGcMg3-I^h)r)!#`VLzG?vzlsf8vhDz-UTbQ)|Cc)0Nd7*q`o
zx^UfnC8E6+HGqGE=}y?T%%0Um9eMrUQ7AxyYV+9-G%PEL+vo;beE@Q@CFSw07JK39
znQZp{(JnqoCECgYBDah=X!z}<5g?_M--6T2l?%v!1bEP1*__8iSB_pk2)uEBKKYqF
zoqxqU%C|n^zqb(|c4v9gmC{x%Xh7tsMa#m3v<juKYz(jTQlVCGArD9Q4WFBR9eG=}
z<r8Okk$AYPF*$X&?-pnt+xR!3UDP8Ry1(vKP74CW-cfaSe9HHncJJ{J34r673bEyY
z%L@E7k|4R)DAGcv(rtdn>^Qor#NEA2jSymz2qXXbV`dfn8X^5Jwo-c+)Abb=NMJEz
zU%5n2HfX-H&#;oc1?Cy;kx4eWQ3K+TLQ7bDl#VR<ou&8edY}Q0a{Ki}zG0t*W6%nM
zFx5TVRr1Cpf*8f|b>fH6BvRe<-6}R~LWRvl4;%n~m9(6Opr8j<RZZM=rso75x+J!r
z%bo}Em#K0*aAwwQV<>7Qw~T9~i+QXSjbD61sA0Yw`CVW7wEE3bX;!!@g5kY!fi0kc
z=9x3#%TGlFMPicMCEGK_>Id0{E^{d_2z>WhyxCmw`qj+NYJ3#__YpsO0#9wtC|~0Z
zpnsSRH}oC(UeG+t+t-G**q?k^Q~vA*q;J2?X0FimiJ5O=yZfz6fB$4w%Zw!woQ)Z>
z;Cv}f=|}4|D!vl0jGLl=1HFTzsbefXrX9b6(MKJu;TtD5M(Rq&xdV04+ac`%fLe)`
zmy|8+$#1W@kaZLay|AAKfY^qHeEPGQJ>0TdL)*qL%nTa(gtiMC$h*0fIF8$wUAIDW
z57(OgP=93({(HwvKMXiLLD$&6jf;(7!pID@W?cg%_Pv!<TFU;ur~qms+`U+5(;S_8
zg9rRL+|3-8cU|wVI!PGXsXLzU?Mopi=Z0k-zC}u?osF@AN;6+`O{-~gWAz%VgvDz>
zen76bEs>(1`4|b1U@`w{_5RvSRPd)VFe6{oy%y%L4|WveRQf6j)b^EDZa1zlrigRK
zgz6W7f&IRib~Bn~p_Kf9Ne@pkH@={bEamuNX9lV&ERpQRE6AA|)bgageg4;~$rl{7
zF8Vx=;rJOp^@zI<6yxCZH3iQrQw6A!8oq#Ja#U<X+ED;0JLi&1BjoZrcQa-yX(CJK
zZ(%>#wSIZm4z#->FXNlqVDI@!XJKysn)od05$xaVYd^*sZQqGj2LjYktPe)TD)rTl
zHVhC)L66qu4bR@%Ao0leb>H3|*;T|Qn8dQ^L@;V(W`hJblQqav-C0x_=9hJ<;pzR&
zQ-gD|Y*)8mF&qw`rc!Es2Y0VItNj59CKfgrd&9+CGNtA_v>Z$9wc|;C5-!&hZA|Ta
zK?TazvV}vO(7OHS_Rl{GdUaH`Y#qk&)<<uGghC>tZeGN|<dS0@5@aWB<5Jn@e)v<X
zgP8BVmwcu$rc_litV#Q&XaTUK)k9x2?aju{^0=y`kH&7-yF|w8yfG~(dL0GdcNbl1
z8eojFM$b|?%C=^8hru+tfb(a+L8vRA^*bqzQjlMMmY@x`iQldp=HPpYEv@wi>b`o$
zSaQ14hU4HzMAFz<Np9ZHNNtTzZhmM3MO^Ik+8fk9zX=U59VuUvAN^?_LG!&-l_p%F
zYz<iNLD>uC1Z?3OS67P~$K6hM26F~I7-ntUN)mL!YV&k2I>D%DcBhgTVUyf7fzNps
zR&*f_O&v4%b4SjZgP|^cC;^t9ny@#QgQHw58UF0)bpDcmYs)6+Dm)v5=Mt~(932DY
zM0&nl^ud0pK(4E4HG)2!ZtSSnAZ@!pgw;e`Fc@JUKhl6IoW#h69iPhA;xC(cbPqNL
zJV@jWEFf^^p|2e}_1P?swdoTM=~$f<0_JUuAY_4B-p=8@tLCW&?#@;Vi_i;ZyV_T3
zhh$8oEEI`WEC!hK&nk+uSMfGC6q?A#)$$X4V{4mtjzQe^p(IFOU)_e^84Q2aGVW20
zJ(H!#RKV8Q(!<O;*tfL#i#yj5M{ZvSm5sxn!TnyuY_v(q#{X0DHnEN7WGY4U2T#8a
z@Z~u)(`Y{!gA4b~=3cjJJ}!o_gMi=!d+DmrVXNBKjyM4?A+-^JV_3dOB(=wR9_B((
zZq)Z<M=AY5%k)O;6iu<53)$#y3-DIY`%zb!0;zupxcqqqonM^c@%^d&ry1r~S{+iK
zK}OfId|~L%&ob0h8e@efS|h61t-Uy(90g}82Y@CYKlRy)s0km*eRNBVbDXrYnWeJM
z3f*XU9Q_S?KgdgjmZCU&G93C({fBzh2J~v;Az~@brG{Xyy?s8&?_@_6KV%);|4x|Z
zim5>1C3dt1a760#Q`D`Yy)g+&zJi_!UID~5JH$GIo8wN0eLbvwbx(^0H$Uhg=r!mM
zBe^{ay;ewqzgplXH;!g@=w%x;;0dkAw~!n_jf<?+x!kirS;ULPD{AWloEa-eCK{Ju
zd1V^3)rN-aj5KWzm)HZ%W|n<fi#gXT+~IzqiaZOkcAZCsj&zBOUMXfztpKwZU$GO!
zrO!3VSzD?pMb_?I8;d~d^#12WaasP*WnaVv?@U<=KF>P{n~M*gixRR$<pGZ4T#;8i
zS~ruOnhW1ht?Nje=g&lnoFbGt-x#ove#qS;lw9}sS2dmrO6*yWyqbnAWsy;}Bi4Kq
zek1{p)!7uuBS~0zPOn689(;wWh_kf$D64|UI^qOo&cwXY0?At5HWlBw;VP{m3qh0Z
z&Y~RF#3*>;=BC`&n)lVYq7R-|^R9AzZrPVLs1P)|oOq>ru0Ha?GyUfuS>M;GdYW1Q
z`@hiM-dc8MQ<CT?y3~6#goBgIlYGZq19m$nE@E-?cltzC8{9Ju>sBQpS@0w8OnJc1
zReG^0!}mM@8Cti~FEN@iW<wH5p$^1!+SS&`mR(5A$?-bBe+aERONjbsashHa{vFWQ
zcl4h-07JBrykhzd=xf8y7xqAWU#fkj%?^vuxlzw!=5bXx{H<^_<vRO>5R|ZOUaf;*
zCM`AIoa~cHyMY3HtDMX1&;S#iM+bwG(&;C_%rfW^%uc-HUR1Nz+g@G7lt65BR^+51
zQf5Gbd{%NK8j+8bi3wP0Z?)&xl$(ox7Kt{oWfL$hiTU&s=>v%cWo7v<%tnv~sGuT<
zi_@*;brCE%`8wQD8JStcTYBW|Q|+~Q8?Dq*Ii{-mc064Cx*_!HZo)tQu=jO%Xsr>7
zLdJ#bL1;y1E(0n!#=*zypOv(9Id6OvhyI-eq0WXQ?>>NXO4$`GxFiJxwac{T8*MW=
z&}}kwnw6ATuLshVPph_5C>|RGDGnv2U=1ozU{HvXPbOS8e%WECmz~!q5%C5@?Q|q>
z=|r#Rb0rt~5b_A`EZaq05K5N)3z>7hA`uuOFryEl|1(my_v0dL3l@nA@>1R9y)|HD
z*G^+`h$%(Vl#+<A<_UviAT(G#6VZ5qw@#7F@{C5{-A}BCFM{4HtvR={i%_P=a0Gkn
z%$2dO5Ud8v7uVLh<meJswT<YcxtIEpGeA#p10c`3LG1nH!)J*ut(kDvp>H8}Q;_l<
z4{9#0pZ{rhjEkUY8muWS{PA=5dJ^uE$6%H^1yeTM?203k*1$f!XGyW0u6{`?+k1jS
zD+@*sZAJ`;UhQ^fCa;OgmsW<s1-HDIhJ+u<@OxlCPI4iSWanqxRD*(JNBup(f{62k
zS^^gHR4oeEm8{~es%;+3Xzt+_U7tcKZTFix2e+4IsWpZ#+`b!mfnhrX9)LH=Mp>!+
z3ANCQa+Pk(TYxk|tLwdr9|CpU6b#pR?8wQ`pxxfrs1ePbJr^c+!@NR&If|VF51U}m
z1vaH{LQmlMBc%VHc*huQyxi_kBuF3q@-{uEtpbTK5W%<({?erJc^SpU>7|9`?MK(5
z#zCJ~sPUO+=C$Z9T{pBiuD(8R4F5F4?z-2b)CrE7;rv3t2&YaC(2-vawl({1?({?T
z<U<;DYJGpVq0R~aH`_eKp3UT&?a6EWUS8BH<r%Dvki~CP%Qxyh`;c;(KSKBH27XMG
zo7y?cVg5Y#1xbHAU$lMM)jVpXlrdm2n<(V5S#xHD7}*G&J0xF$iOf)2<-w7F(T(Nk
z^=J?_DPQ!L*&x8z+UGLvVA)B_t?)Fy;d+!mr&v8ZoUe~G&M}Y`MUu!|=`?Vzh2IFP
z#9Hb^W+iSO^gRSL18?mnXhXy^>^8{9z)lw<>XAB0oFqF=J%)ZpHYL<qLFx|%LNh`w
zh^`0c%4B|r04NDqcR+AO`rA`oiEgcPaAVQfi@9t{{u=~4PgAv0et+u)7qNJ9k$2Qf
z!oPN<0XWCf)rW+-+@pA+|2xqKd6}}4Z7mn{&I#~`H8vxR-)NJrW0E=tV>_R&`kvrC
zUiG#&z#>%#lOR3mi#xxzI*-TU2MWXqYs8WMEM}BJ()SlpKS>#h3|e#m*kX<@C(9>~
zvYug+xvzIiGP&+l$EyX1xi3mBXuYWzBc3hCB&*w9OHjz8i9`=Jp~!G-$5<e`G(HFF
z;fsLGFRFn}@7IkQDH~p6jR@0Ih@@+!Ir+`^JTVGRGC=dFvQGU1j-_=HZG^fEu3QYN
zTCcOKKu1pZMNkBvokHbyR@=C>@Q~q%+ZyFjR1hwfn^TjoyvPLF-u~r@?#AL{@BEN<
z9-iv|LPK=M!E1=yUSpc}u?5#JP4Bl2WVzs|^jcitb0a9uk-iX&`Hr^VTAzKGaoPNf
zUfkb}?}1Ca-`rQbMRoX1y9Ea35{q<PGhTr=bh5U#BKwv+`>r3!W=9X45)GNRPX4rQ
znz(ar93K`cd}10V0Z!2YNdT??qC>Z{1!2G!=x_WB>!lzMA*!Z_37~lzyul3nqZ;yR
zfzE%SJ~ED(&NHg!Rti-0>oOxLjS1WH8=^sKmFZ#h;u+@h)Ow_LF+BMEr^Y7nRQll^
z$Cl6SUwqc-2lqS-!L-hyUQQR}AG-J)9M$;#E8knS#n{19jt%04J4)L-F6)a$l4*Qg
zv8BzCD2b<s2l+H!+#lj!wo<jtN1SRMBd)gB77872YNQdi)rKH~fe{)7^64?b*33=v
z*>tXRvt*PHWCbAaS$pLtoq0>9CEvI<@*89M_&i>!2R26>VPuB{2WOc^_7%!LwK~GT
z@dX$BsueWsxD)c8>P(4c=!b9|)>r-X3|9?^!WXp@Zpz>ZN<;i(E)3A1XJ<Bg2l!g5
zqePS5e{;uq@HfSE7LAm8Iks4n?jn!(OQzF_<QSS_X~&0+0$<(oE<$8>Me{2dCn(R+
zKF^&JeES8r%(p7_2Wd1dUIABQCBQ@Zy&P2zJA>WPC^ZZD20laM*phnw^9-)7J)QH3
z3&C4;Xjzr1j)fo5&EG;3gjQ_Nf+selnE}KMjdba8dlihyLT{@a%Vf}I3pWbt+H`ut
zdS#M%41OymRiPo1ZeMJ+yr5kYrgNQVW-2XTzJq%DV5u_+R596Plew3x^d*ILE_EJu
z`k8jq2(o#Ua@sav{ZMhKvisW}Y3UB*oXq{OU0za({reptM?Y~?^K|xJcR;5d_BnZ&
zI{$W{!*UHu(R}DFz!+w4&qkjthv`LYvzFtd^K#^@?`k%Dsla8Y?QF^R>=t=SOwFuY
zV_I*uHP;P9ys2KfV(X);O}Ffd-1aob$V~LE;Ll+<PQ~M5R5*{d@n*oE>g#)<?x!cx
zuce7|$|`YsovBsOox|wcxR7bxR*D$OL~U1iZ!B(-<Y@0XNIklyZa}kgkn<X073_>L
z?D{fY5k&tn$z{r+jKQ8RUf>!Qxlv#A`kKp4hiiR=id>O_x+;Ni;@yr@eWEgj);Be<
z>8xWIA&)XHZzS!_W`D~Oy<I(y+5i=Dp_EmZXXkIh-RmodL&>UPzCHj<D}gg~L)v3|
zIJe)f$Z-4a(zDpMGOdISWHq>vrqyS$Fs$%4B(fQ?Y_&wvPMsF0<bB$%w7&fOJMO-V
zeTqk7AL9YK=|f;eVbYJyXNwh93a_MLmzri}710JHxa6GJttQ7EcI}scTabvbozX18
zSpeq+uG{ms+8@T)3cA;uIK1rJ2kJ(w7>d0Lx{rLwM~Gt&K6(sb46A=P01@0lz}_`m
zM_~4f-x_`80)Mko-3DLVU+b$Ai9QL*14@hy@+bTWsg?H9etKA7M&2;i^j2#E6Quhr
zJ!!3c3$5vAan27h_Ajf>n?9iXwhuMtoHq4q(OW)A1znbAHbakO)6Y*gxm@;98MI{|
z7&bEck9n$>`I`_#*Y8?597S@vZSYPPk{swW5Y-ujar;bhaeDPytK>)EP@4G5NtIP?
zT44VSa`ar_r%b06m$6(R-#6{LYX;9J*ZNxQX^ZwnPIa~W0oUhW7^vIx;6j#a0~(&y
z!R`@er|-WKS?Fsx{JR9y-H{TQk&dj2-f)*JY!AMi`1#?snqnSDS34y|`iv}5FjXg$
zodXk4YIg=E`m0?A!j&FS$W5#x=hSwth9@5aOPPKI%^=dUEPt)$EZq^m{@)v|V|=dQ
z(|)WlPL0}SW%8S4U*Iz(GC-%ONWi<9OZ{BH-$hoH2aCI%HQiQ9j+|uDT_C~T9Ns?P
zKh`{#_z|J{u!#nH$ub4wHeX;YcdC{9Ep!A)?!@0)uvBY#Vjl!+iiko)aWc4^lj#t2
zmq-cv!*%hCaL3#QQ-{)@3jh(pG0YL{eY3etyF9%};lTIVWkDgkLTLAfS(F>Ai|&Ed
zpB7@rv+)4m;WKt*9F=t`uxq2>jIDx2ttDR{<?0*>h~;ZTdBlQDls@AQ+~tz=CW6`v
z7{hD;>zgFDWrbKkyVi@xBNryvYruLZAaz=&BA$L{QPI1TgC<@)<;p=ybMc!+$6PpS
z$0FeMzdJt@6Q;y+uwFHucRB<0G|&GIw>vjA>l=R6<%PV>-j}ZTh3jKBnN0Ir&VebX
zH0n^I{7JQ6P<Mb}WpioU&`;^-r^P>9XODFuHG!Xw(wUqbTSa}uV3~Ah)S?o?4p$6M
z1MFVji7$Q4ARcpA^;h`cuO0Q<9#wn=CF9UeMBC3gY38pvUbW(`*G3;p@3>5xx3U+n
z9$(osEkzppwxTUZZrcv&)pHPMa?3_rF~yw(9a0SUDh6JISd+8n-FS@jc-s~z9c)DP
zKz?qv9#!buk7fh4`b-&JPvXZjGuiehAy@j!%q~h#CU7UlFI~ixU#ivlDj&|4%@^5Q
z@X9@DjCW8Tyk7^Uyw`+z9%kDYv;o9cYG!n~Z=e-k_=nuvj#Tu2G`>iF0TxY>jF^{;
zX*<Ma$g>;KK3oPzRW*#xTl)5kVbA?k8`c>&Up<>H^h`>rh+&(qI%FPztK<k6g;5{<
z?wS=^nksWim(yFeKGKqqg0tDJag9Q$_@b)keP)ZyZa7vaxzB%8eOk?XY2Ixvv4Xa1
zhd&1#h2``2P0)`G%6Nao#cdO<4A9<9t_R}vr~(@Yt(YLOi6dIe%&XUNtK(~{S)3DN
z%bG8~D+emER@r>=uQ|VBa(lG30d|nv1DVW|kGMT?Me|MH!)sZx^I|>(c(i_lzQ+>Z
z(LMYJz0}OszUQn@%rE%?P?0S)PB(ieW7?HRBKaDhGplV~H?Sl8I`b@Rt@C$!S328N
zkK^;aicxFGAdz8nC(4gfz5rHn4vOrgStxl6^B@sm@-;eXN=jL@x#-)n)1}5o%(1DH
z0P6R=VEB{?2q{t2?B*$Z(&BUy)bd#PM02#>s+B(ILMG!DYyds9cCel(^V3Ga)eDLv
znd>n)xnJ%vuF1<r-=c>|-b}4eyZ7S8wFj^->-gZtY8Ml9R{j&YSFyH|=E_Aq^B^xY
z?cUvj#RU{Z-<i;1f$|MxwH~d#c;#81y7jdc3vPTw!oh=52oFm35MXOxy5bdVRXXV2
z-n%$#Jp{e|cV`eoKijSy4+~>DyIfJoh-}eyYay##n6Ek+K|@Hk7+n=ESZzyG1AL#G
z8_(^*=Kt=C`NwHteg0)y+mz5(Ifn{+UPYxQ|M>wv@uPejS@q?VoiF*JyBVDII<=0-
zq&nqh(q{>BCb8~2k>yJ{Vq?49E{wX6sg-*4fyrA7xlo(ncTvq^(C}Q}(?N!r`znw5
zt97o|-exqVFx%4F^BP0Ty4IpX8g*WQczn}91<;Y+Ib=UAc5wax7FT}M_AneOX#wDz
zr3ckg@&_I4GqGvhen}vPh?o0x9+n`Tf?7QOOoLz$>dGfc&apDZ{L+nenFnau!gn@d
zEkJZ2c~!ld1i-^kP3ttrhnhAJD_*?fh~knObyzil+f?mN{jjQuuBS-e4ElV9(K1lP
z5hTQIPdjaq!$}c~GA!H5oHo*!l77|ko#QtczW>f&@JEjgf5w!;hfmjMbUERU5L|OY
zrY~sT1Gy;Z-ZRgY$CW~C`$}VQ-mQJie&sofJ~tV;=~1KOvlJq$GWo1UG~zG9dfeeu
z-2nwS(0I!8EI?Acx9;+8wN>P0-z@zhI9lM+j%uy_XBCs%&;Wk_gWUx6q?5xOH{(0n
zElF_-``*<IC!FMGu%S&BN$%>{yg1DO9-gWW-IBctNxh4~@fRXLe_<eoMhet(y{C(|
zMTSg$ATD$9VAl)q4N)nzXELSd9LobBe<bAT^0TsG5hmG!hP-QFqiYc!(Rh{3>7_si
zrO0yjBFAo`Q;K7EV1DWh@50PLGS8lCCRfT&IsX5wZ0xOK8>^WUu6d`*31jfBbmnqS
z7ZH;gkdbv^`}u5HM7qA|uFALO%p_~4@Sr#+rK8$%W5HR)QfbfH1qEgQ%Z$5fYWn18
zTnjAL0e<z6yuaEtwt~8>8}1v#exUji9C!^-7;ah^sp_uvOcvUWyARoY$ylU^otxkB
z-$@067rl0b)D~9-2Y35emeXZD;PKA)M0dww6*$r*Oc~ix)nt83;r_ec*BVtf%!h3<
z+cLM-=SCW6uZepi`kZQqf3?)xmT-aFpL>NEa1B;pVFT#4<5;H4i|{P|yIoo@>03Xn
zAHEX)pa}l!Y3AFL22i@`ydfAq6RV!bR&ufQd|7PH5CRq}*2K6J7;K$9S{?XkiS9Dm
zW0D1an|5n{Rb(v2HNb0(?z}+;X{BXtR>pT>?IK3}HUw8J-y^@}Un05jWHO|eAPv+G
zpp))N`(7!%$EBO!%YzD3Ia38swnFaAf-)H%OqS*im@MBvD35%?tQJ1_kvVMWw%yU+
z$Gf-X*=UH2tXml8Rmf?_@)#t`>E4Z_@5boIg}x`JS85}cP?P=h947p2*VSD8P8b16
zb;$;KUYF9EhxncjQT`t1?tUJXTL?R(NBGbsjl+X(GD7oia^lx|I)M%RkB=#^8&pg8
z%@f!ouk|XI3+*o;HKQ_2#hWCWsY#~$x;6p<)fw4c%GQ9v$%jh2IVp;<U8Gr%B*pZu
zo8O$Obb-@7k5v57^EcBZ3+V0dgQ85adLQ_`vR9UW47Niif=q}4E-k^P<Gf|OUw*c?
zuKwvF5@XRj+R|XZAB^5pDde_+$?Lzme9GI~lW>^VhgFaL*sQ>tTa;ob(zw8{E9oV;
z45!GA>1qhDss;uB#c|=9d%;q*4d|dWr<cv$8kQ%nI+u-tHm$1CuP3z`1#m*7x;k%o
z&Be3AgG?H&2=$iM#@|{^!d-(Xcc+Ax;}w@JNAJAYef?kd<%=n;1lVB)C}m=Ouq;mQ
zwO2W|s|$<hJ72-2=SD%xou9hXc7s3v=y|T!{m&uU3af{OaLJ-C(?LJdpJc7TWwQ;L
z!zab&tC1mEb?C?aldZBzl4?yqs*y2j6yA2f^W_?yfzg+g4TorZ&>ZyzT+n2mj&@i{
z4@bD~`_2vC6`YpIL+a{g;8}07&9<pQvwH8A)b?D=5Uq#N8^xOf4~}qk2=?jwoAajq
z(o!WzNyz}Mlen;B$B#9k@oVk-znxk8l0HhB5!tlh4uA1qG4axCvar#<$M={)wSX~@
z-x(t0{!N|I=<yxh;IXnp`^j<!l^#QnIcM7Jatt;9IEO=4lAh%8KX@s;7n;nIbhBif
z$Awag1F)~24HplDfuW?`P4e()D}t0ynlkZZQm#w&2;OAeoZ3<C(-dso8oiRq|0*0_
zZ>hQ{R6I96U6-2mMKlSr{!5ov{%C^LGSVIeBcVt)o)o4B-Vt4}bdeuw3B9O1ISQ8M
zkv*yok~kJg>k_|Ch$LpM8J}m!vpA=eRl2Ki$w79Y>nqei7Dzmh`9UKM%W%FZt5KwM
zMLBNIeAQy_W|VQNK~wJ+4~DEE#$G)BNnNcEEgFl_Ib&bqkA<ockVdcj$Wp{nS!C){
zcaKZ$!%a6HGGN!gcF}Rn$bM1Jjz;&~-@iPdw;h5y%_zv5N9)b!t9j+^x?iV?H!O1L
zdSNuzW;3r3??974?Ve#f?7y^}SGu$ek{=rZqW)%cPm|OxMVL~+Mzi@*k6jc$H9XAi
z3bV}**~X|W`N*2d+rP_P7)aE&t!fR5;g*T>**+XC5?i$GmCPYX=ASO~|E*biYneND
zLnHeO91dj5_!}GVthfY=P_k=16!gu`fEGTE{o%3V;iIP;CuDH*jL>rB2=}@-0eiRf
zc3#!Qbr072W6^*o(C<%^A$#yOm0K97iMfinOaVZ3$5w5?9rfO*ZjUvhUoEx@p^ja`
zgeM&+0Pv02*4TIZt0Fui)>W}MpnaKAkzD{EAAKL@;=WpR+Gg87xAPnGFOBj6K(KXZ
zg0fBH0)9E6dssi&am)hZjnw>cS?+8_0|Uz8mY7p_k-=f#RDpPhhY+(9P&h`&H(mPi
zmT*cI<>rr`j@*ONrwJT!(&Ae+awE^?T6g(IZ_aU<`7m@MM~_q=BFp5O76#9JF-926
zN+I#ytt+qb(6|Iw@h0oLN-uORehwZM-pbiT|JPpf+FYnBSfAuoM$FK@tba`}+iJf4
z)bYl>f<^_BOikF~uKu_x&!pre?KbOrc0AlNcDi5ex4NgFPN=L+^cqDp^eRR%k)ZIL
zmccwLJo%K*B+3<6+6Y?mtUyK;ZN<|oe_1^soLMx6;GvBEvz<@GuLXBK-N>emh6mqO
zyXiBMLr9Dqn>xzBsZE^Fvb$Z#C3i8x?lAr)r%ceRM}n9m=5;})BRjoX(r0{Juw6n9
zcSJq|{WQD&nF48bweP5_eJrpR`oXa%p&Bi7bWj=ezHaTw)4mvqUH}D<(MMaYL}J?;
znK(R*23$Qf^LxJ_oN!=<w90`rH-r0S?;+u?f&DxMVJ<8xy{wH2Y6rFbQ<@yH#eQ>f
zIxW%9g68r)rw-x=f*IKtKhx^4kjZ306@75<MD!sa-hf5*J?vL!@(p69>{YDP&%0vn
zvG1xSWA3$1^)f0mRkE0Q06u!QHGF)xxoh@FqS}msENtq5j_$)wcD8w`+jEyc(mbB^
zs;&(AG^|5sw?EWf80j4xUHg6L6T!Cn`*+I{#nhgbrU|90c73>ylwJR#CGX-xpeGLt
z3zNYQnCn0Mv5fIIjXE?gqKTaa&#I;PmtB5Cc;>ky@r6Ca24cv}y^D43##^;6^I=|d
zg{=)M<h0uVFwb|4mTU8%-RrHT%eo@ZIT=xGd}z(a{Oy4$heKcU6w^J|387%hH7oUb
z0j=L#UJ?;MSR~WjG|FvmdF%er{PMXWbso4r0TQkIMz2A?`LNCM&8Y@3SEhjMt(%+H
zMl4bv6r(RJTfTkdGJl}<YG23nd5==xVpw6PG(<f>A;+WSkz?kHZnm&0)XuJM+ad>8
zf)7#N;KFTk`A_l&2C=`UHO5&T?m?4ED`+vM^0lf&&E*U3UlKa<b`Q?^V;e_h-(2iT
zUJzNSCf8sVK^$wuvm)p3A6Heq#Pf~o@F>6CnEL_ZkE|<snz=)!+AnvFTv`|vo^96o
zw~!HLJ0=LarWE1-#QU(_iDTz|0b<#tF^?`ox4M7HrRb0D-Fn{j{2)Zy1@x3lwTb$(
z+4xAtx!QVI{Jp4BTwh(k!qz;5BIxg5#4+E#`+$LvF1I5V#a1*i*O`#*dp{KGp*41`
zSAi)Sj>mOQnVwg_@LgN5F!`9+gT8m;psW$Qp#293S^hd~-g$2Yw+%`gRL~q1sCiql
zwhr*PP;+6HwHlxqH}4H+9vXDGxtQIKkM5e0I0Mnc_}K=mrySK%eYB7^!RsH0n~g_q
z_Ku;|??6i251<_yk3HL;x4$i1HOZ(CkoB<>cur~FgU4^z#L<6Db#h#JxQjFhORhtc
zr^P}UkjpGXy`Dc>Cw3;2gR<VlUdl6C+gl$S9Fue0v(X0^jxNx>Gdm(Kr@k7@Amap`
zQWZaY^K1e3R~+zWq66E_8ZstcBH(JL0EVrW`Q}<3y?5W!avCswcSAaJX+)+%4&}NJ
z`TMT!-T(f2txD$op?^4e3=*Y({R`oJ0`MHE)!^i*!5^o+gNVs`adz~5+*7You+X>f
ztN(zsz~`}^D-P`fTXc^FUIs@_jr5`g#;^(z>jpS_;*`8VZEg`KiDwo5TMZBJPTkmA
zx+~l8GOaUT{G-eyR_OZ{CcyA(*S#-|3WKxwF8n?Ou*(;|rYq71)_eM0?zekVl;DF+
zyK5{sp$fC^JQqPfpOzPMw<f+QRb`K5e^-oD@)Qb-`nE{aCl7bzPt=~;xClPZjY1V~
zPKxdPuE*nM(4@WwHrxIKm#*i}1qM(4i?o}rpDZO(=REIm#a{U4e=a5{+L`0&lC2T6
zKZNq(mXg<f7e6j%p-t%ZR>!davJqKxJ<3Gc{{dk@p1)?(DvSd~-^v=R6N6c`S#5f^
zayz>awm%-KcjrE}-SS%qWji{o4Z~=gdxf;?7Vkb28P;#=nekl$Qa6Xm$2vWBO^j~n
zOww7sO&mMP4HK&1`@tL5%`UOq$A}%zr^?3G-c)ZPrqy=j!gjskn6>RYl=(4tlG@*Z
z!1e1_tE;=leIQ@0VdtqG3^g;~Pe*l@psVIk*IpOJ+)0Q%$Za(bS{M>Hze?t2#+CT}
zI~ajpwSAn}0TPCd`{)w5t}JR5&5nD%;;?=_Pt5>O9n(=AyE|z(Pamk1k6^vCOx^Dh
zF1R(~O?Omp$fj0r^`~_M@@eaJHORfzgHKkqI{=j2Sm{&!rr%Jl<8E~=su3p8ESI4$
zca7MLT1kNHl$%jk#-QM?T_xDZBP=n32&=VS3U<tMJ?8T(8OWUkhol|NSHA}**xMM@
zZpwk??p%Guc~SB)29=roxLe$;#|`;9tLy*E)VnQMj@-zWW&oUw$SO(QXXgJua=KL+
zqyw;dfbY_I>WidOX2jm4J7Bif{`uMK56AtL+gaJExcBdY`Zcfm-;B;|U(33y)vX5T
z*D_W&SH9i+ChUkwG>sz}3vtdmXWcrd6;{!lJ+)kka^G1ZZC%<jI%ql+L3dTjm~!K>
z!&glL_5%@_|56UA)Y$Mbh99TBb9{Q!1A$spo5jqjW^+LjEho5GRr4S|#!7RdJJ?4T
zs+k6ojvBEfDY>#Sm5CX8bqf^2x+X{<`ee`>?x-za{%&X288uf~P#85!r-qaom_ML8
z;?vroalazyCW&l(|M_*_S4^fJ-PQGG?kCZ0>+9-Xf$sNxp6+UBYcczDRncIW)IipK
z@0{@!q^`J}%3hX=qc}iT#KJbz9)U9OHHts=eY3LW)8X!J*TQfB)z9nKTHgD4cLHw?
z9Vu94)gJw0Knw0tTeQfSlhjBRWw-g>^6ScHKu@mfZ5oGFi`Q+cY&6`x-<+3o3uTUW
z>!%6Y^~K+P<L$R}T_#oaM<?UXU8N#Qr7KDcvWwklEZ^B?Y72Gukmk@uDhYK?WdVAs
z@hZYl7DIEGr2oX&sa~@NVFcf*S$bLYyS~G-BP8}N`2EM9%Td@ZbX7r>DJQt|{r!G_
z?}zuFSO5P1-ZFEmD!qPr`sLVrr!21<dz=T)>kd<OHNgL4#+`z@Vbqj0sTp_ApT<$a
zL26<mMUqr@4N}X8@@XEPd0lVbzH?{pJ<nVX;z&3|KeR~e7Trc5iUA~Y7BUG7BUy)s
zOw3mq<iX$VmrgUYPPp2*_nh4uPz{X(J!?HmZt9$hrdgMU6i&kw0VvpL0(v(GXicNq
zeeeN8f)P=5#f-tInb;|;Gsm1d*{lh(nABKa1rKLeRn4jra)+*ObZ2IBU0?To*_6<v
z7OLFeEuE*@$yhMfg?GPW2pIFjbZsT6-?rIa<uHusodOh_tw|GUxD8eAnda=$nz$lW
zfUSC9m7qg1q}lUBH!G2)P_925)wCF@g;XW`vfe%oZMpznoZb2aF-Cq_+XCtMN%@6G
zc2H_C4>SriLZ)gJ84_wYe4%XhZl^(J*3FxekfrdEb(?kr0fV_J8yZfUz=-|M`=|ju
zkXshbPoC9V;JiNpnVY*oDLW4Uq%QB=`Pv$`N;P{3MsnuUFe3N-J{oJOcB2bq)m+?&
z+6n00v+>2*sP((MVSdHC{k|;a`PbI_g2t59=BigTnp3Q%N2o%#r!Hi}gB>MvFpz^K
z^D;*&0#Ek+%axm^9?`YOQokvKS}2yTFOD=cE@Sx4KcHrByKR=%zU-|eJ83IW$raMu
zK3RNYHCipOYNHVjq4&H;h!-&Dx9Zh|uR*!JuTYIHXj#2x(q*{|<^rJ9wPp>bOq(-x
z)xZy~+v+LwL~z1~qYw)Zeciuv;A5bhDoZp+!5C|Gp-W3OH>u20n&^6SyGf0bLE;s4
z1kr?g`=us!Z@I7cuL1K-PlfMazPiJn7lgUCEY?&XpK^~{dtWX~p`PE5UFNm3&|1L~
ztK1v+H&yEE4^_S|z6LLE;2&<*qN*pCT-`mCc=(Dq#p}7rE<^1J4rzTkBAnf-odpXw
zh?ni>Cq;IOW^S1W8sUp`bv_gXbj^z&m3D%b%MAi1<VR<}%?LNg-bBOKLeI#_*eV7z
zzHzh4vY3*^9H;Uyr1c;FovDwq&71>W>>;mCi@jk)jX=woIV%^;SVgBJpCZRSOj%$g
zyC*>y5au#zZdn9rgDlL}IYYT~(t6$99XGJ-Se@0amIux1sbS`jdK0v%Q;=bqW^pa{
zeyYuC+A`Ja@^SKx&t1)bwuWLt^H96JKeK1C`I~oD)685&cY@;OwJW~Z=0>l7|E}D-
zEX*R@DwC_90!SY<tx>)PE>FFMSVQ`BN>DXo)pnmg*?i->`F-{DRkGGXX9KV}t>~CZ
zu&G;L*TF#2@@;<IsChI97BLk0o*1$*x6G_9piE#4p4m;h@APox>5mvsbekO0JP68Y
zYS<Lj<(Gc<C>&Av<!`wv)34$FOC6W?6gd;2&1KxVmbke0hMD8Ua(sjatyqR#psL|e
zw|Z-#%%HXA!C^EaGh%Oy_?C39USa#EoF;YiPaoQ;{IWj8a-VKV0Z=Qvfy%?rbC0j9
z-+%L0zp~kI&8mjsd$%n=N!6aOfEnPy9@hSHwcOF!Zs^2f!_nC*rlx0*S<5}}XkiPp
z@y_-Y+Jbtm`9P}LFO0vS8Lot79s`73C{gn9^o+IbzG_RNbvB0_*KIS*tb)<hvTzzJ
zBcN<foW0s(GGj3796sT6G@$}pvf!>WA2CWZ5gPaj$caEs8y3TTh03D4aYuHzg&TJf
zI~l%~G2dO;x9xU=dVL{(KZq~>?&Xxs_LN@e9=(5R5J{k;D)|qei{#jR`$PNDNK4cn
zlo~BAM^__)=IZsm|Ke2%3o}}*R;-`jTM^2)&AH#kNtl>t9>*L<t4WnPG@vJB(VUWG
z010G=Wf@7rW_C}|0%~R6SCxkvSrfjCpi<G>E@aoF6y&aIm{mcXoP`CHJ(XP!I8mT4
z<hZ9sWI5~!0eP5X=#u&TE(N;w5HekfWtAnl4IiNzFkz@a<4*X$38A&yA^`N8vSpmq
zJZbK68^U{c{(fK<>(~AK&lcFc7gbVbmCXrPj5Sc|;SM-^?^ZBzS<^b9kY<Nm)da(R
z@3CV|@ehVd5^-y5b)%o(VBCLvEmrS$U*5cz8Q)iIME&nJRKdK<3~*Y#)o`$NS(d`I
zZ5ayp(({*AfLix(6S@TvZrbq6y5(W;c<S=pFG+*F&)m-F(Qm`O-^*A(F!m7bFYD~#
z832Yd4>wfS9%hLA<{>P(iKRrW2tFI`R+;lfEiP*X1^3uL*Fu$D{TmSr*zI<8HwZVg
z(h6h0Pgi*Efmn=P%G_GU>X!SUaaMGS_cs!qB{1&#clWx(DsAoUwL72R4O2phex>)3
zeO;ZqNo9Y1PhvH9t2;|yYd;07P-b0={~YH$IZ^}QT2*c>)2<TM)iUdm<&lQxg-5Ii
z?W%XXWBub-ke&H<=l=cBtg~@lo$wW?toEguDGlscZK$b}$k5&4S*P?oE{0*uYmRzD
zuaLSWgk=t-hvt;o<Y1-tFUR{vRGlTP&d6y9jm{~mzzps^2XGEPdn7`FO|f*3=fQ(w
zSHsNk?zpIFHye4a9vw1_nAS6C^TlW#;F+<L0XYX$fpg^)&1Rgh0w#nKrlzsV1E>?}
z1A)VJV3kETf!WP+5*^CsaCyv)IkkJTiwJCUsNClmSY-}Yxd~Sl3{`ST7Qd8nU#{Bw
z-LH9VMgwrqzdXV{>n9<*NW|)nD3JZqg1ev5X`7Ycp@!94Sl6XDMQ>yVj@lA0t#R2?
z2c}xqHnTg?(4(`S7JC@!?jxYv?KlPBqudYyePkCA+B;#(_DeM84+4tjhTM%XBmmqX
zO)|ah)0UZ^)tD~F){tE%w@daEiPZ=C*0JWR0$W`o9B7)Wj~+*r432Ok;;5Isv$`Y!
zO-E<N8se4qgQYS@k*OH$%e#ksXXcJMy~b6?mrWIakuK1jCGj&1ozxLDK|a#y*UcJW
zYX1J~9m{$%zHch!BvJ31Bi}ibq^SymRFgA|hBS^qH;fo|auSnU#qg15B>5S(5A(fe
zB(XSlHe!}~U?+G-_QXyYJ}5{4XrkLmCBi6l@1!peyhFC}b5S%i6j{A|Jg(uaT-Wn7
zmWx4TUh932(#8UPsu&jeqp@-72rtHAX(wQ0%50zW3W>p3sAU6%nCRY{Yth3{xV}v9
zR-bq=QDfuz&(FXjrqPgZX~lf&XFX-#>wSySNyp|d*I?8RxuwY&Cm+7joCH9wFBeS`
zGKXey_4WvJy;*xtXrZ-GjVgvD9y$a_m!xbXh0G==*fREWvfD#i_VHjOmAd(qfwL7V
z{9i<oVPrc+o4KvcoC)n-3ANugl2}M~;@oJo#LM4atb(s`6Al!=udm)cY!I5V=Y1(N
zbj8)>rk!$X1ijtjWhLtbKLjDNtJcjLM_J87qYsj|8a$Aw=4EyAX)Sm}2=?ordVIOw
zo%^je<hJ>-u0NvRT>xSL*R4vA{e}&RAC;m{1`?qX3^5Cwx%6nJm8L*pu|78Yp<EGI
z_X_oHsP;?$bxZf^+F4cVIji*k5e>aPpe|!giHz)`Usi24k#-u|F3U9YN=o`I(cQzu
zS+F>yj{8k_lM_-zK4&3>0SaEk-nSR-PDXDy8fvAy+kA!$Xt3i7{`&{MVe9JlIIshd
zCgekO(%NprhMT*jdJ!n8lSYJva&KEIFqrcG;f;*gTLw1GE#iG=tbX+ilgzc(VrS(P
zpd?_=l$vb2u4^sJETrG>{}8ArrP_kAQ>NUn%?JgJUG7lty4U6JGatX2d3nnG>Q~lD
z@?W9DEFE0xB-nVYd81&Wh5P-zEFwI!@_Bxql*gj~x+YfOoq`$aefG{VzpnToSKW==
z>8>hX*Q&SMtlDhNK7ptX*8DR*jMS17K=*Xiy5)|MWT~_`&vk{Xa;Uf#y+<p%t1(?r
z?K&6NISESyY#FLbW@yfAsAQ+IOEzp8c|%mBaiX{&2CX;vff<}m#trHTbh{4rZQxMu
zmI*>FLL{)3WY}%ygu4Kvsdk}tf7gbge=78B-M`px`VcSNXXg#j?|#a{I~!rMe>L~&
z%TtCNO#}aNor&P=Nlx{Uq22DTks9c7tY6>X_v*~5_vf#la+^eB{p7mu`R7eh0Q;}E
zXD$e(*u`a>D!l~12Ht*NrZ`FcYP#83Ik-zrqG^W8@NTH(RLNr*de^vY(ZyKYrHOZz
zK-$UB#&V26ibDXTUG}`rnwNLBn;3@1ED6X4k%AjU%wIvmESJV>4cPf#1|H9$XW_$y
z=)<!6oWVIG3l)9jrTDuk7O1%`*tx6PTIz$An9{a8g>F%xdr)sV_J>u3gFX~w@xhC7
z{ws3x4{@raQ<ftzQY}msRr5zSAnxy9*P77Et|s@pz75LXPs*`N^VE-s{(03J2%_2W
zUWem0CaBsXt+b+YV;Ke*IvVCITHM$Bh{mj%B_m1gc2vUyeR*XyXY-!LSPm80S-4kb
zHO<i1S8dYpY|&yVv!`U-ngPk75(xysoEFsq@xy&@WBV81A;yEjEcn7WEDLs9_aG@I
zG2n}t<<T{5B1ekmRWKWo&i(cK-!O47!8s^NHRH<_H_NO=^Zt3;Y(Lp%G4g1RKX0pd
zwJ+qf<ee#p!25oFcKb;{NZ>yQ#6N9XDMA8T#3$=?63=AB6;4jD75V=4+m^xn3U<Ld
ztNwa0CJdjuUuVM(1ga_)p>Bo}UUh`P=Eh2q%%0Gq&GDM*F@w@ROxiyB@rI52JLZW1
zPW`yuArYNjaHE<&=bS?&sRGFYH9Q;E0L)00`*DrPkwZ?%o}Q3_LN~eaBqOM*9?bT9
zW6ysx&P|r+ju^NNiRm>B)`{Wy6e8x5&?3$4-QqH>+lppJW#&#>s!W*2ZFy3I^F2O{
z71VIw$u5|Crl$I#0@?hpkO0S915n1VU##58@Rl+sWGOBO`1L(X;(Gr5RPF}p>%PDJ
zpfLaa?=2|X*F;p&O&Z&V`8>Z{qmo9NWi~HY6%TVv;6(Vc=6N_!_JRCb3}|IoO(XgB
zu>CTeKbvkq-2IMW3j$C(t^rMd5;-*68|&NNx7MAn$Swt^KZibw=pMO(8({Uct*gsD
z%coc~tHkAX05cdsi;-2Dir?Y6(QKwLQ*;jFG*JDpvecT`>j-WyXLWd0cNJotj0c_U
zoUAkrt*S=2ii6+RY`{=5g3G5D^%wtTYS`kt&(qsi7_~dF>-3kq;rG3ubuD||@cjKh
zwH9*^&!!vZkhwIQ`$G!Y8BWk#`Bp5p5bWx%>G938g9pek8?i2$&0uC;vh)^F>24~9
zTP2wtuKD0kRQoIsKtuJ&7eqot>RO$<B~zPe4i@0vZL=hSG|ZgZ-8RJSp{R$?hO_L0
zKjr7!$y2pc&NY2D36j*M0XMhI%&pQ^_kjJ~kO{l|d0V8<XqY3@n{4gz%b{(By$@}K
zWgvjBFVvg!L+dFsQCS4yo`!okY9|Qydie)i=59^z-4?drw)-+=clG33BHaA?4hO?~
z|9<w~TnE#w+CMq&T&`tv!>YH@Ra!n{xB2{yPH>tx8f1{2pl@gCa^wV37J*hJ!rHyd
zY$l5H1d_QLF!z2-v$T6UlrqRYYmyeIio55zRS8Fy%&S*GC4}c5akPy<V2xim6Yi#}
zqXL+q%3=7PFW{(^1;*@iZK?}~f)TX`eBE7*SWuTl4jC_D*WFboY->89vV)arg}+-y
z^YGq<<+(xV*jU`EX6I4=vHSoMu1mWcMDJlx8u+g+nUx~EpW<4>8U<Y`>o#j%SGe7O
zTxMoO{(t}TvjyaPfB*ayk^6&{=z(eklGjNlvoeboSL|mVc`?<9p^@H!eC=NLJT)HM
zaAelC-tC5Nx{=&z&Z<%I5~Y7!|MRScF3|(3+Hw}%0qt@FhMhR^EtR-#?3~i%5uL?M
zV{o!aysrJaWyx$UC0MmasyJmKLKq0~bQtG>{E<GkoF$p-IM_HPM9{sf8!*506k;v=
zdGD{`d1GK}*>QcDg#F%My?+vubDBK8uZwTX`1`rOou;aUuc&OAU(xxr%}}u90h?~M
z@y|~Z!Ge`#F=iX#9MaT{xbFM9!l5Q}=kxEMtVwj2nPz#Zo(0-6YEA_PWsk+$mmj_5
z*p9lor4`iMb4i4;3h@=)l?ecAsS1A36f~%%ze9IF?>XvqA9>CU+RZ7QAEtMkX91AU
z;4b%CJ7;lT7RG9ubPTMgU_4ASaT1+5Y_i#c0zAJC8+M4MbtttWNWqy>w+DI;ia_bB
zlb^H;%~?geE-Hs`Ow(||rUPLE90j`9msOX@ShHBjv>>S3HK*KaGk5Dv@1y~{J%}2E
z6#sCIz{`T_%%YoX<W(_pj+X!WUhM9ywVqtug?zd4^~)#FL3FuSo%7d8cL3nP8ya@t
zK%>DpGq3Jq^xL*#VsHYw&31WUw6vdOhMBryHSHBryKF_jQ#o<$QzJgjgW=WIFkF@^
z8@yJ&=9)pEn&v*+&J@v*8D~FS4UeF@vUmbS*xSa5OzR%LfN1zTFXbp$CoigU4F`O>
zUTSk-vIvKvb8KN2^%kz6GB+IUi|BKHkh`6X$A!&149L{nu7~S?S~LdD#cecgXS|pE
zr$b;Nm~CP8&OJu0Zn}vitJK9njdz6<UB%UD205!nkYR^vJe;yBqXRN?=kqYCVyb=J
z7Rz&ITU>gxZN)qfk+T^y$1u(y3+mk^4;$+#aCqKggto8lZpY-*y;8KZRDiL1t!|qN
zbnaMthY4XK>uFt)WfX0k6vjCdnj5uq90!CdhvLkWu$6Qzsdd=WM)x<z=x5=Taj6S|
zJ>S;3P&+i=`N`KfvQeoaWWmh^zrL30d4=}1f9665TSn4-<{Ft|;9w;H^pW}jkhx~I
z{Pjye{KJRkDGmLF2MlaOpgsU_<h%Z{YG(s_b6#i70ltlOfZm7L*L|2;r<Q~HU&p{e
zV6<)Eum<V~_`s72JT#wT7`odyiDxr{f<>36vVrh{_kJ=Lg9;sEJ$Vq3*UFp^M0NX`
z0oa;(WD95&udulVRh_Bf{3{Ws2HIT}q0aenx~arFa;ng?@;NVIfUeU>`j}6~YUSrL
z`1gLA%vTP$_1w!rvshhS^Hk_e4*(Vp3uo?KZaw#{IY^li4Wly0Cn<FU@WAFDutC4W
zV!i*CK`UPUS}E2~Pe3HTmVf<u;pg8%=exV%i=Am!XU7j3nROm0{8Um!yII+E_o{|-
zzM$ENHK0VSJ0IA8<W|jbOZuD<OLTWue};>6Q%P+RUb_kmtKz&uW7VDry>-P>s{oqU
zqi}=5-=rIwR=HM|W8?Ci3~cpsbk59}=6><n<Xn{c=*EUYziT-KgUX|Gej5Y&#6aLd
zhvH`E3wFyE_5^jVdkQh3wCHjxSwD)A?M(t}N(v&)MC%vVirS5Fdf@Lejs>$-=~v~~
z{qr;fDGCJtnAgZOsZ^7iZ(e+dkOj5e<Nl7O*;qunGBaD15iLW_!)dL(!*;?#32%k(
zJStN&RrRvBM}%fZp&C;@Fb9cU^cm&OcfOjJL)i+Sq8vV_6)_A#AI+~M3!7&M^ueyr
z{!?-r&{q(9KHh5h05*ohquvc;*HDM|#E|CBIzwlyHHA<&o?e#{cN9!jitF?P=8|3&
zK8LNe(21o>C!H}!;+~7{;QV1AhF3WUhEs?d1fX*liFN0`ujl7beqj1oMs)JE*Ba66
zd=#7CMucSnqOGf*>G!wJPw?zXeJKm{@Hs?OzBNl1(|@Ki5f~#!r|3Qw7J<ULm1=bJ
zvg&E7&S$;P%M~!Qxr<H%a^5JP%R<j<%Sg3*&!wb#s+*0uSzRg`X5&~Gb^K(OX+h;X
zV0%ZwwZSS6KZW3_!^23<?1o)qS8rM;W9)nmoR)W&d>&kqufuliYD&_zDnXrqAlzGq
z0S8Xf{U|c1_Rn>p-*ey+iSgLCR=nE`G$>ZpTKQ(-^Q9B$8fUX9JHj}aQ&slnKWxmC
zQ#}E?N?akk_4I%r4ts#UIwu`8_*NU<>rNPrwHs#Q<uxPY+B%41jR^#u<spdZGLu%!
z1p}4Y5in;qG!1+WOUof&T}I;VA@<ANwG(7_TyvK=F&Fi)#%6&-s>Yu%fad0Xbv6xq
zC?1+1z~Li*yg+mnqo$*QEF&WJZx3SeZ9o?J48@A7S?Z&OShww6De+!aHmia81e=+w
z%e*$stp<|On=ysIu9S^Daxzwn)iN6tf)g!UAKb~s-K58|De)EJTKM^G+*|M|#CP#u
zSLNJuU3*YYG@W~pcrai`1#i|ZO~=Yc)v^=5SVA|^9ae1wL1o73I{la2u$k`f0?|#U
zCh;+%x$-!h+wKGDK+7!Zg+XSH&e6;8(^}xC**s?ooP!3>hYGWvJyhYNEohAnNWN6=
zJzr0p!wEE*q&fW^kP%gJVQ109=0<-KG?}t+L7Cy>Hl6F+fS2)~lNHX+lV&!M*U2y+
z*K<z=kpqWOH#s&Cn<9EOjGg3pGOB%rwV+*K12#0veKrcBxs(55t6R1#r>_V&cD6-~
zLya@FaaN`IJnE*pUX>BMme)*80Wt8Dvu7l5o?V0mgbBM~w*a&56>l8@riH58IVvNr
zJNEz%Fw(<M84&dF+Gc{W8Xon)Xf5}m*G9zZ>SNt<e3BT$Qbh@x0D#(O98h7$jRKgr
zro(GpbgEKzakoM1&UdcZ)q?JeRb4H=t)-lJlbUXfyWX8XtAFwwDu7N<y)IR01yQAj
zuS}9^Ud}Ed+kEDqa9_T6P2$^F8@9sq^QO+fWh{NPa7+!IwqgV$ZQ`FY%Pm`0A7#^I
z;&+Qy%{izV$nbmw!F{$@Isw!9bC{~#)f@xmkQK7pF0Vt@<LGU=n^tvJ22L~y?6Q;9
z1Wg^j>u@ov<H}udpCAc6Ae+{}WX#;VT5cnKog#_J3}ZjHsrJN#rG(SLl!V5;&hmG*
zH!}r5S2f84vW?n`YgG<1l_TTTIoSA2CZ@VJCS9Gkgu78~*1dcAnbGxd^2|_~9+*o+
z&$IX3G-%elxvtu~ZDJvl(0A+-*$449#d(rd^lSB0a)_=jep&e0&5SpEa^KXhVeBBn
z7_lNufBpMSn$dh=XS?rRxV$x;-E?m`_dFbVo?YGS)ZD`HwmCXML)Hv5M}CDK?eJ$7
zGO<0PTONce+sx0=N6TgaN6D=khqUg_>Y4OgP@0(r$+@~tFXwb{N18Rac%sdf1G2Fh
z#8dM;8i+%7AC%}AOuDCR&))5v!{-!%N0?}wGK0-=O1;aC^E8-5WWn7!dFrZj&9eb3
z{G_p=EG|>lDG`|o_o|%aRTk-CH41ETKCZ-k)gV>zS9D+Zvdo4<XHGPdH@HXjuWzTt
zGIVF|TA?<wc5Cllv8K+PV-X&^DOFwJ)qelN|K!cS?)TpUp(_8v+ceI-AsZg5hIzp>
zEl4%i8W3kO?)CmmFk4;mg>2mM*N>Txr5rk#KF)Z*OU{P(h&CO)`(xCbUEmB&j>r==
z$HiDx7@c4hJz|gegJI*QPXKC!13h7Ax6(3Y&XiZv&OP!hxBf|f@+g<T<)qX#tH>5C
z3<MN>(!=X*i_2Kq4WmUq^X;c5{)Vb>HI4Cw+<WlvI}a0VUC$%leB~%m&3}T{aEx~I
z3|ZzGKqal`wqWeE%PARXfwyT^>29p*&SsT)pW6$%#AD>qtQ4W%)pC=q-~<__%G?c)
zE>e?}AFM@~Y61sN)^xEBt>)mCh_*29P3wV54K`PpTAz;0al$tqZnXih;AZSDU-Rql
z8ndxWm;LN>oNAc?yJo%S#va}5{KW;b+Sx;UpAp9_dJjV1SVN;GE5&8mIFf}btN^@o
za(eyunFRtk4O0WVOXK>ik@bY_?PfEQ0`{JB8f>B9&N|vo)rJ}PkC;u7IS&Ei$PNH@
z0qSZq$W(s2&3DY^lS$s|=K-y=YMM^Vs?S<;=TKu^qppe-eRPqoc0ZSt(YBrop<j;o
zLAdou4}7-LkUqo9Nun$4`1-mg^*2O*NWzLkH?*l88yc`!tlZN)6QiG5C&O1rIy)gq
z1Gcd&I(ue2n#26yqI1eY-C)V*gfcqx8LCcFPK$-*U1>{Q;#mxv`Dnamj5z;;m*;ph
zRVCSVdA~DrjX>{&1&XmIaMb%f5A`F_;6e>Ef=Y#}3JmhktLOh$hvjtiIU_!@71q`B
zj%r+YzG66}-qVkKH#s#2&l*s>QG`LNrY%S9fqjJ=s!LGWOq=ZH2@qDT=$g~ohE1AP
zg99582^b&O(U2@#8h%?uPD00Mmd~M1*_?BPiK1bP>*{wFR}H`K_ct9Pdc4nFQsHZ5
zYUIZnHz!Y18WyZ_RmsD>ny7Ynwa{q9*)C2_7i(c?bX{gMGc;tkEEX!uQCxc9lbn4!
zDD$pL3>dH9kJ3l5@ZRt$8BJWwR%_j}#YhQ00R#<hRs@3l2TBpH>;ao_=*>=oE%$Jr
zh5LNN2*GVaEeCM8tV(V(l<Ja<hBm7`VY96uP8;}WlwCPPR!7Z4yIKfWy?JaeZjyhJ
z1|Oy0{BGwW1h?+4@-_90<0P0@Fb?(2s1ZJs?3!o%C^ofjoh)QcQQ8i;nj~0h0adR7
zP?{-6(|7(DUFY8pnb%9FT48F=9)M_3rHIiK&DmNIs;^76>ulfh$XWLwnC5P6hs82w
zbdp0l>1wk|v+)*Y40L(ca07v#xzlGZ^DlNA8@G36$-?=jvCp5f$f*nyX4W;}k#`=#
z*Lq+170L}$U0=D+SX(k?noARWa%H7?SP6hf?b-UuZ@dk_Eo%j(EYLins$1}Is?<j_
zHpyIkCKK|QgNY!5rhT&HXxMQq%7Z``FY8SCSL|9*!#E|wcF&#<>jt{E%un0jSPEvV
zQ@mI0Y7z`1**Tbv;2sJL(Y$A@)olS}8JdQ6pYRWqBhIQ4f+RZ`$K?7M$KE#@X4<=8
zKG+wA-Z$+=!M?AbpYBA2<)&v@oCPH?fC}MU<g56Pm?QF6R7$AvDb%s6NbcSAWnJ=6
zow{-#;G%JMcio5!Wz4pzkkhJK*BYVj!AZLb;h2j>x3Lgb6V?RJ+}|!VwCk*zrf5**
z^SwF0UL9o#=el#|RgUwI8s*|#T>#kmHo&54Fc|$=o1fl$L~Zl)a@R~F<(w1WjF>yD
zdR7sO=-TD$3RBHz&ux<S4Mt>zch5-`gJ34QH_W*6zE5Vo8!k<y$GQ@JRk~q-R=$+~
zmo5dQb3y_d7_3#zJG9T;eO3r=X6T*#<0N0^QW~n(itY(sLi}J3l?fYZRAXTjW}J|X
z^@}j=S2hStcmKp*cm`Kzr8>Z?Srmgnr7dqM%gzU`{W9%+CThpEYCgZ^87G=eM&&aC
znsb-nb6bcxiqDVFQ9IqKY`8mi;e;@trdNaKSI|X26n?6*1PkU2WF}E@G9GGM1GO4c
zhM~Fu$4`2ga%SZ2=;9i&+N8AU-n$z7zH{gC5oj(?X?N&-YL5Gy5+Kw<*HE+wktO%(
z`Y_z)uE<uOor>m_M94i~{^M^f4{WLY3hW);v*DgPM2;g}ikNSa?8c67Wi@blXEFLG
z<)O*O1Kr_NCWfxwUHH0o&e6y~x7-bPb<-tyc!T}sOn9jYM44K#h95N$Ziga<FHN;r
z4HuSWwVew~=a4Edw))^&!+X=aPAH*!N|tk4x%keOcg>=0tUD7vr-f_os+neD1eGOU
zf%j+lyr+dTU$W7tTh?)A&r+|dVb_hF?NN28&ND=#UvXYXs7yO$)Ih>ks7r<t&9Z0`
zwH3w@<2EiASg(bW1d74jO=I&<$I)@iyf!>+_-HKILRF0-R~DbZdkMDK=C!9-D;&()
zwPr~nb)N=&Shv)4jIS7V0z5oL0ZkwYXXk0b+;v@5=v0KG)FVT-`PXZ1htt57=u-8}
zl5{lS#HiR&m8Er|lff25W4D>0ZnG--39EFE48fo(Mlg3=RpRp2Oq|DeLd4on&b6;Z
zL$=JqG_h<8?JW2H`zdqe<FILG(QNNx*i<M*%ceQHDlXT~(}1&>!Gnb%CqGLFS#HTd
z(K<_J5+Ybt2T_P)N^;|$HmW4Y8LgTFHL$O*@ArS-79!9!KWGf;wo7whJRxwK`?y&q
z=5;qq1B=DI3l`D65ByxSJH0NbvV3Vb$z6oF?n_T<Mn7-4k$B$`LO5s2%f=Pi>c-@I
z>`jls{7$ZzK3E7CS?v+}95%ST#iJycg7EAM*WG*SKOi8FO3dV8+#)^Odr0!A7p(L!
zcLLaN!+n4$^figQm{-C40VUY7ERCvy*d5MK8Z66fS~gQNCS#d^&hBn^()&F(&v$mS
z1c6j6DpU09LKd9Wg5uUIK>VlCCI(Nndu`FpNNudJ+B+HNF;C|i2%mabwHXfq#-MDU
zZ;@S2<9=)ry9oH|XKF!EXT?=7opfZ6!&5_#X&E7zXPaYau19z@J$2*zGSp?4Ws!DW
z`It>^NWMB}MsR$W905vC72jHOj3qIJ?|JKViwuX1nPl<os=Cb`JLQ3ih#C5=Sv69f
zGo%oWZkdJ4y9C8DR=dxwtZmL8^J|;1dNXWbqgwIqSk=>l+py1At}<n$`?9Ttr)Jz&
zHoD<Cr~UayrnFNQXc*9qF27t*H~+PuETVW@VqASj0<P+|v&#kqbeVN}kUd!4!Fv-B
zV(A?j@SLLk<NO@UT4DH)?%3_*tk%Xjmm^A@2~<}$Bf5gw=BBP*7h7WQsb3!k83i*p
zlXl(lj*7BtU6qFUncH;3=Je$d$K6|rh4z)tDf97EO(w}XQ<(Wr3*<alx@h9EI>}|w
zr4~DzZf-eiohepL?{eBn?%e4y+?2EOD&3K#5efs&T{N22=EZ3R5XZ+hxJG7tGZ7Jv
zhE!ZeA%b`_v#Tq6Zt&f5cM`=vZ4A_~#Y|jx=Lp*_?5x(hkUu*N^dbIrPw+?Nc-wco
z8wE}2V{7qk%-4Aa0NC5^V71VKU)mNz36tkJQ%2Ry$UzTPm3W;R|JkFDj6Bq4rr&ih
z()N#V0*0*#H_!GaAiD;n48a0jz3zM(Ffi0{W1ku#3dbdWM;4V$(0#@G7T2<T=Po;&
zEi-oRlzp#PcXa7}&8zNJ3CY*~EkB|{bX+Jv$|-~Oq1p)WFIQ#hy547Mq~5*0dRLoS
z%f6QFN2($Wc74?k)}OzSf8}j?|Klsx0!@2YzVdvY*X_-MQ&olA`RngSlcesCgcoBA
zo$_A8jhz@$B56cr+0ZoRU}FK5)|&LJ=5G)_SGcI%c6?17Gc?>apEQQ)%!`|7U&g)3
zLurIXygwELPBHc5*=BBI>a3Ak!zN$qXF4+m_n|))%uQRo*7RunJl6owgE=(E+EWEK
zW_`9s-89#M&JJyC66ibWqu<6SDizj)9<w8wn<5Ls)u~xW3OtG(xd9E$EPpO~6L22c
zpr*<yL#CPE-wQCmzT)}G{XVbO_4VsJ3~=n{$q`m@ahDn1zuvyUf5n;k%;da!PQ|T{
z*;%wJpvr{X2}BsXrAn<;WlY%eeoJmZ@LYFhne7;V%hvFSXIu`fYj2~}w!(V1j1AZ3
zjkkI(mpzA|#=54RYRZy%w>gzXs<oE7`=E^hb=C@W$uYwDgi#wo9XZStJEn9H3v%?p
z<37x8(+JSo+>r3W?z>l;-XhxE=B-&NdU$ji-!n1C^{d{N+i_-@cf!BoEm8#WZgT^=
zWRTlwwGL2#nx1FFwNKs3*L8pYTJGUg|Gd3_|LpznT8rlQ7oWd=Qs!>H?uBllHm!d6
zPvWyzp4?H(t>i~|hVjx8f>qh<OVxH+?>)YYYQ%l{&tG<RPSnt;+<>rxs_yY@8nVsK
zw8#F|Qi34H(^hg+%h&ORSXk|}=$bnfy2OYo^2km#7^rF)xo&nf*`l)PQ*U*$vm~3C
zAt}jJ2&J(6t$Cu%qYGx;FjLW_nJ?3q8Z(`ln2p6W$Hj?mm0xNMhA!Wx+ulI9Q-s>1
zs)sUSoQ)b$2zEDl$MX)9%vD3)xz}-g5AI-n|KkgBA@}pw@0<xnENFch>8{_u-}rw2
z9oM_p6^?oeYd!zV?_ovR={iZ%u0P@){_P1CppKYps%#;fv~Kr?p<$g2vx8cn)!az)
zAc%l__L=oXjA(pxRLE*xtF&yFskqPt`uyzFlU;4jp@iX?G9Pa)`jEqchPm3DtRbPy
zxa`x^6((H)V-9-(=h4(O%w(gcRjCO-TLiOg(04o<m?;yV|7;)2a+49%E(@<c$$7~?
zvvkt53uu{Ic3-H$NM<#C#r(wP)llw2_<WG<USnqYGtzFox&QiUE;sX7Ve1O-eBY<J
z*8W*vhlAWDSm@=q83r)AXOBxGm%Z&QxOrvc4nX_C8mK}Qef4faqe|+06S35dU)N6t
zUio}YYk2YNtJ9;(xif*dS=m=)VqI463+winyAiJI{qfZkmP=Vwud8#nkHQM5zAl>W
z-&+vviqvIUgJOyrVh9~Au2q@MxR>Qm%DoDZ5I46)<-AJRg$S>l53@Q)s=5443$L^*
z9}GE*vDW^JGXU-eEIxEwboQij%NPEB3zfx+<5_+mMnu<)olZx0&<cSD8{c2=lX=qC
zv1*?n02$YQXbl=j;{JMnWUilAabb((@|~d3?(Zpr@iZirFm$!AKHQ?emoJ6S0a>Fu
zEn9lTt~Jwfs>;@2iDxgwPDu=(1?FsId|Jq8xn+IbJLh2tvwiYcEvh}bm%nvc?c{JB
zjM{DPs_a=vet-_s$X9Cuv-r=%>v&I~)S8J=thsp_YSJCUU&twUnH!CLlJs=$I;<$?
z&bZjSWM>E@%*b)}WvrqLK5&@9^f>n9P3FpT*+|oo<vtLFwLG7iWC<5r?Ot`1N^S(1
z^Z2xdj_djDp`5b-%)43!JBtD9GMf#eU6-;q|6p7@V-Oj{G`<U$Wy=7C_pUY~Ze^))
zQkt|09v;S=x)s-D&oedfAq)(bIszN4q#1`)lCW+w5S{H#j<lN+z}C7lS5%v6h#oyh
zn>x^vQK8%P_>Glz1`_&A_hayE12#L!wY8MfDxG%nhn2_B+dkMqdRU7Sbup6)j`q8r
zwrf>RKFg-Q7T9fL^pmcYSqP*+H=ANnHEcQdGoLa^aSeBCrg$-Jw6IExQ2V9#Nn6)2
zKH1f@<HSN6YlF+SB0^dCx;ts%$hU39{0zF9hJGG1-i$R6hB_+6ntcp1{mU8rdChu4
zq+Q?`N?$G;5d*@#6J80Uj4eac9GNGrI7_kvmZr?D=VnOSxfx(GjI$vbAj))x%=#Vu
zjcC?5=wfxY{6bfs5sMjFlYfQm;*+y=x4mJDTj5q&4a&aw47&P#+(-QwYbkS!GbHpm
z+An`o{>9!KR2xff^Rtp$CS-B-D~hpH)4Dfltt_{b=I`AnnR5Q9=h<(;w?LmqM6hHl
z+@81L9+cm%=O_Gq=8+DIGa}*CN6P)(F{>~-=l+K(sj92P`46w68TKc}MoP|sdX0T`
z)=#}@=ec*<DXhZjvyRAXz~QuUsl938RHGASIJC{VH&3Xe&bjSlpkdRG2#}ZW!Uzd8
zog_*{tL;IIiJ&bEdt01plsRfbRx><|kv|T#oP)U~WPt}?%(1*yA6Ef6k?YVDH3jPl
zN}l8vh{OtH>hA+t@j?p5)&#PMoT(jfFc&q~XiiOCA2hElkkw{k)uLlJ!`Uig@HdQw
zcFQ4mj|0J&P0gF7hYGKp?K?m-$*Jnnx_a|p_xsEb928v*yP`|ocIQuX*3oaD)wg@e
zk#Yl>wXUv7*TH>q8uK@0wl8$^;w~d=Pn|Zk^lbP#eGScQiOx(y=FDmj;rsr1#-c@`
zaV^!}4G!pQW?zSSku{Soi-l6}be7g^1<Yw*_G~j6a(e(n<|F#u^e`Z-s<PuU%*2W+
zUyk=RU)VMM75~2XI}CyD_E@nuJ{C79+q`&roo1B3dw|bw1gpanw|+X*ZZo${t`U<R
z_odpbD%OD?b;;KnF<Mpi!LX0D2T-?6$9gfWu(<_HRg-4ouK*m`w`G=2&!Yv&Z{#El
z&{Oj2kJO=LgMCxa?f6vW=J--Winy+RC&X>0O2K_jWl+N~rs&xNj)Jx%&@a?X&fhY&
z7^_~3?&_pdaT)fTeK=L5$3(8%V(1p#ZVemS_kipOo2%=`x&%Q6Rr!)y127uFMfKEo
z%y!;unhIUjxUOm_vwT^1`v(Jca1_Vg4m#~@wCBr9<Eb{a5J8bur&jv$>iSVLPr-9m
zHv~w&??^ez?wPZ+EWUl$85Ekq!X&zN?q2bhq3M+bzi?fLH$`geXmg;U&}X*$pq{5r
z4QC$(!=_2_FMoRi{Zl~|2=_j>WC8A|L;W8g*f<8lst!_)E%40eWL~+hEFeYTx_f*f
zYA&jp?v_0rE?9Km@w}qXjh}QMOghhoZnkD<A+CHrO$M~S|9JkIjZ)ik?pK)h8+_$6
z#r#6}xJ7oidDuBD&YpbGf~tZ`(~W96TGDYUT<lhrS#^xrs!~`%(xZO2ceX_Hpdiq_
z>aela9n8J6s@mPwHAk76bXm(Ust@uA&|00C(G%z2m;5t$@Fa*BwvH5bWW^_)ea^%1
z=mKOH-=505%@0d8by&4Vo0%XWTqDo~oc{rVVWM~A5Sp2i5k{lv`D0)8(!qr%ZMf)_
zAG61B8vHAMzZRk0EqvX*Q@5UC_;9KR+GXIisvS=S22fT)_h1Gs7Mj*DJvxJ5_p!Aq
z-%><Z+p@Eo+gl#g`tF|z-d3L;v)N!Tk7;9ie*U++R#`?hVsU4mMTY>#r0IN@d-KxM
z96ua7TfFY;`OQ#^aoxcoHwaPxn}I#h!6{t=L&KRho*QA&FaJeB%`wx?$8Y2U=xRo&
zOX2G3!+0BV@~pG$0MKVQuVZyIvrb0%3?gQ)I%6XoZ!rm!X6!W7@H&?R7Q7S2vp_q0
z<k>}4x5c!qK%4Oo&|0fSP3V4W!cW-@oKQx7{8Th&YcQxbz@5_w5@wo7f~-WSU^4rV
zBYiZCnELA^Z6%ISl$mBgRUu;7AkDv<<u<+8BRVy6d*30#a)zzLwCd?K%@COC()~y8
z$xPFn^l&{tG-E(&HB@Od%7T>%KWjE;7s-lEa{*Q`&YiM2{0%S+DH&CBc=Pv}rN?pR
zxv??zYrjyRg1pl@TC?z!n!(yLs_M2QF$E0OWSdzJ3WiX=7J9(oe0)shOGOXux@=mu
zRBe4-%4%MB=kB+Ho42LrVyt}z2dWln*R{nS1-GFo4^z44cHc>^A<Z|?Hi{sEAw{qa
zLdD9*7g=&VbnD4JpBq|jFlRh$!G?D=+F~SnbB!_Y?v~%3Zf3xOHq>O{iruEvLQ5~)
z8*b%=eO<i^7k_>Tu2715mwiVerX4aA^q{wOt=MIwY(Z1}3Ulghx3)_;l8xD^xeu-8
zu)*k5lUZOm=F>F!=PC90|AizT7!|0)x2kSmi%%}=hMEw1MyPfK<fm>oAmPsW#Gifn
zJP{<B_x3Au@Q@A7(qWgO<Hmzf>JS$A`P$433!e)9%;QExZ^oFSr}@OP8|aONZOfQ@
z2x<V-*X4SPv+NiZlW7l5W|F&W+03v2^GU&u_C>JUo_EexV~X$N6m?eWSbgWrH_#7^
zB--~Yy91fLwWsBmn)u2zoLI7T^-s?cT6a{|*?s)O=B+`s1?xaU7}izhD#=*$E9hBC
zokbitpv?KcjtHjfQ{WJ0h=CQRZfVNCG`2wNZ8p=`i5*Z)G{Q`6SHCruCt0UP$7xVD
zETphZV^#!=Ia)Gu4&F!3i!(B<j+o<lH!OOeozGlnCem?THt?gfMjh$*5oB}<c5aGm
z?P-}-D$WPlX0b(mTr+H_%fqPp<hQqAE>g{<9JU>Y*0EMSH4Dah(#){g`c3yB2I|{%
z>-e50YqjesqdA|SNv+RYg}q7Pa?SI-lmD;1Z`+^Ozh5^<u|{f7vk8^?u+QwWmQ{HW
zSyi~c`zK+cp(L1h-QW9}M!Du7i#D5NcA58XTv3{3*GvH^q36RdDm$z+3qQM8?WY{=
zSHAR_caPW1_a1!Pbcg<)T4>Jvk{h}d2C&-f?B02jnzz3>v}O-~UOTOpIgyBz1)Hoi
zKPyRT$j6NVfwIe3)rb{?CBmQq3x0=kpSiqWd_^Eu=hI)7a-!3mbZ0+Fi%Qzpwcnha
zxa#;=q__j5S-#aJuG>+&TU3i{h2eeASh*{H4e?WI_-G|M`A@c{9Oi6bU54DJfOp&#
z9G@u744D!wD8{Ei-i+Z5t?1Ngs!=i@Ep^K5bYXqIL<J8`?a{6&>(b^<QLwdUSNrTk
z;V>7V=4FCa4RS)`6o_!|%AAzlZ1#ATrE`l7-)}SAJLAsR!uf829+e)@uA$f?0-beO
zHe*=&+($Vh{9$<M1WD(@{GZzBK@tja)m!a0IS%85LLoYI{x&FFe4E`CCN)189$J@i
zYsEC))J;k~Ke{S&XUh!3u~ZK&M(dt&k9xpk4WjucVTQMS=8TnQjqv_l_N9o6&s(Jk
zB~e;o>YA%7jz*}8-bI%t8o$CTPY8e{vf`rl;A{r~cQ~97O^C=^why7aiR!jB;U<qv
z8t87lVPLlrw#H=BHQukobNVoJ4b_~S@kf6w^V)8IcTxx`9Md@=Ei#AkHE$f};j?~D
z9rEK95Hwe}cZ+13tNqXry4tS;h^ibZ?5GMGj-$Zt>N!iQ#Nfb)Xu&D7eNupuF}E*2
zyr$h)mx0FKsa@CiY+%)`6W%d1RU2jA$^Tjaw9}Cc4j0>kDf=L#+yOwePSM<K*?zaI
zYcK@V?vVq}@YI3{-gl+&rasNOxY*A=WQ>E+CT9*lPu4>CXxpb5XYAtTR@p6FHrnw6
zzF&59T-}ZEQm{w&2rgGu<GS}>qlp|bBh`Yl+Sb)i(U~d6{L$w+6p>9dUFmBimozu_
zdlbAdk9>-;BBVZ>c{AHN*9bGs#jO=~4wi!|<{48JSqRIf*@}GW(gdVkdu}6|_d%7f
za3DN0Z6Rl7aaoBfO=S#pr}jH+AUkNfg);CDcYE`A4hG!L^Asj+MjkkLis8V*S;IU#
z5Z2-9cTTi-{He!1G9P8eGEQKH;2G9|<NSbN)ZQGf_KB^YPC|+9IvL<ynAa?kT6l{d
zXFoijX05>|>QHA#(V(V(K<92*<9-$#ldlcO>2eOJd!D_Gqo|cFx60=7%;}%BL@qyF
zP!R~vHp{|XpCp@0SI@#%poa~nZMq#p!iI&a8h$V2uC`$q)WG?9fS3I6lLXZ+UC|<7
z2ZBcs@X*e_ujv+<8HD1mHtSwa(X2)e!^OA4$l<ZV2Pvj*MZ6oQf(5`w?ZVRutfspM
zo<b*Jae3~Bp#^7l3_aOS5dtdS{WA19u1*E2%e-}kl$qjnohc$0UU&qd-deWzOlwhn
zdF7O-nI#zt?52fx)0VMQfL8pxSJaq+fwihH|LMhSwA?h7;d)PfdpJ_=W@ec}0cvxU
zFszXw|G>O7A}cUvQhtWXUA|PpHM~w(c=y+Oc34&l9(!BdE32y4H#;jVvxV0Ez4ZHs
z$od<zE36pSzif31fxR2C*6hAvU;L|Ay|(J%FB`V{Afg46n%5A<hln_Z$K@cwZgdX~
zus$lGp6-NvWu6WpyKB-x*)a6foJ@3zR8Gg+Q49*mTI-Y^N7V8$MzR~Wymt%m)30p0
zW8-U`Rf*uNQd~awhAQ6x0BVg%L_t(F0}aywGtmIU#tUrI5*m*oW?y;2pw2Gyb339*
zc_^N=(+@{wk#05#WXbHjjbfq9_6wSks-in{k6v{)w+3}r&tLzX0o-i>uEcejI(N2*
zcJ+#qAFhUo%0f(Q-Aou(EDra=&Q)IxhVLzz)7eF1@()Jz?w(zZyBVvCm%Vk^N8~Xm
zkP$VH`m?W79r98$Dq+h_?_081st9U#uzJ>dus9|8ZaV4{9JbDe5nP{q{9%N#HT5xM
zM_|BHGtTrdG{Vj8`JHeI?qpXsB1)6)=((mGW7fRk!!>ylgBo(GCXOhKV!}0MGl3`_
zweTlyVy2H`K$BUXkGV$`z%wbZ7UbSKM`tm7XLFiGGVcCbyQcl&fBfpK?y)M&u48s?
zQM(y|2AB2uDEiN%VMvVdrDPW9s&4v<O!;Cdh~E6ejF}$6s<x#D$6F?yZb-L}4h<v9
zF!Rt);kxTxx4P;5<Y0L}Ro_`-R*hB22qMdTPIbe>7$^GZX%6*^Qcvq-MZ=jh;Hxft
zU6i`UK&@)~Wq&=K3^``$QCVeC1-}+L3z$0=_a4-$9qI1O7nZAIIp3WywXb}?zH(0y
z%iJf%tUg4l5#}9TE?Die{h<>oD1yimX*QoQNSd6<snf%kt5RxQS0H!ULL<Jvetz%i
zDR#S6FIP@ji&=gIUS&>sbagjl1w<$JR&8C=xoTNNEWYmv`S72RV2fEk2~xqGYWa9C
zt4h~s=wrQ)>?DXa-6MVmOfXkVTlaP8*@q`JMwy|jH`H1_YP3mgSzwoF#@uF&Ku)UF
z!<ijO#fO|SmI;zju;!+rP_64g_0DM&1^d`O2CYw(rV7outu|fRSc^T~WeN-cqP9Ft
zDX!Sty<{pLc~;fODtew|m?}Z|6cM|#eZI5iYLGP!A)czj6e+rBf8U>lh&Ro;VBXp7
zlROwU(WG5$3Z$^`ce@E|yjj2`w8u#lr{8G1G<dfFdq9N0;Ocn=B`Baq1yC};MOu{z
z)$Bn$uC<dCm#fRJz0G=`HXQ(&R1b&EE2uggoifQBGG4C96uALXbvD8)TazZ29isb|
zL|VT#Q1fk0iy02efRvd79xHZMH9RD=BKEVlnFJ3&O0LD+bve3tF7Jzdu8YS>8)nOY
z{u1ezS6ii;(HoPwp_x~&S#=0;Q9I{VvM6If-C9;vV1$a4s@*QAyDg6?xVhFsJv%uj
z`LXE&cJ(VWq^|q<(K)K>>wfZ%HzS;5Z}u!_W?5u4%>P)=n+P<)z&H*gR8`&lmyPW~
zNB9ae$Ge@4V|AHjnVYF5oo<9SV<?Qq+$D1x4(gi%ZF;M%F1c$e$}Mx}E}LXZ_<WT=
zJM}}h0*!LZw|IFLR3Crtq%@xSPw>dB8w~Fx2zJV@=;S>0VT6aea@2*x@8+^fbu%LJ
z#ZU<WGb7klGqvx@I7O8t_}Lrw6+8PApj`g*z`_BIHJ0o)5)HY1z*7Lk?vlB)VD3aG
z)P=#U&%&kfWl!Rxg8uwQj%^`jYoRj7@Ht}&PCssFpf<v{b-sF`O{+H^JN&M(c}&qk
zM)+`Z&di4dW}=Fb?|l&dZ~Dx$dRsXy1cLQ8+@$b)HE{$0%1_))E$y`M(G{3O>paGr
z6tlo;O@ademkpd=Q(>7L*i0-rKha@R?J5grg1%7wIB63c?-HY{lrvgIRr_j?fh^?-
z?bu*1$lmly+M<LiSSbhA4|Q3%S(qs+7L$Xo9OpHdZPqeFZ8_$Yt3nZah6Zkg9psz4
zfc{PMZgEVo4wM*FcS#Y;RS(ZmB1b*uMs+_jGq6j;AX5n}LKPQplVIVeaM1;Z0q{V|
zbBe*>?1pEbGMF}AHw$=n_>^&5lC1?uoiz`$ul0LJgz0@wO9}$RD}gT9ZEwllKz2Kx
zgkRP>J>1@O6Hczh{xg;vr6^9$*v0VYrMY`=Sqmag)#>nRXeuri{r!>oH05VNazNbr
z-17!Z&m)+toCaj8Y4Zf4xq7}qv0P8t9MdNv$1Yo{!m^zs<`_k!Wz)i#k34MRjC~3o
zP(#HP?DWW->&Y;dkXv+z(C|>U=4O?Cb6_HKr4*{}`+a;y9e&m8&VO0e;r_Sm`0cwn
z85S7kRX(WVX2jn#pl}ZCa$3{ETv5(`elYD>WS~yU;{=-o5}Dz9(|M5m)6}A-auwG{
zS+8-c<49A;IR|LSg1}7229!Auv&l`2fXS*N0BAxRR5!06tloxM^|iWA3+^x!26;N}
zbdMf-Gd_)L9;~i9wfs|TQ+K$k;mWx5if)O()9yH&bQT;3V@gPzN63-Bk4$L(inCPn
zwX}DegF~&RB!qJmptRoR;q!zy!+r;Z=Tuh{GR8`{jo-dy%cHZqg?d4?TP23LX|8k|
z^oi(Xy)8`0dD@zW@coY1G*H0&=5~wiA?-AGCfF5=5@Wgve<SO4Ez2Ld0}12E6nf5H
zIKiIJ;<#;vD>EhV^67JSBd*-zHE~l{T(c{%W|t6`TDut^aV}$E8Rt4?CQ&I76F?1T
z7L84+SCFXP>=t8!C6hMSR9?CfOfF*Q3U#!~X%0KS*^cFE4j##C5a%RF^~UM2IZHI}
z->O3(B5k2d>P&Zl5lXSU959vzHg6#8-DT_++|m5kxIL^y%FgtB6c(o|G6C7Su&KxS
zg$Wcf94A0A+PPi0Jx58>FAl@&^r5FhS6pmqt?U(5UEMgi>sfAN$~xyT5(Zkg^U;@Q
z&%Byru?}P@R_-3x-n4{9lg_S#=6KBjaj8nz&3g8%GW(Q5xJTCOR^>hft5Xrzsq{^n
z{~+ptQ<NSJ+NGm85gofEaD_duso<1GuOLls+$}R!Zu>C9goyVo-j*`$@eR1SGSlK(
zQuoQ1xp(erTb6b^)w!jyjKItzU#SO-<^y)VIS3%wg=*E<rKWP#BRn<9h7F0cPN+UP
z`m?Vbd7MX|j0^QS_JK5=dAS?lddp4S_uVNqZ@pNko~F=DOkw7Ou0M$fok3uW#+d4z
z_MWZtA%OARZhAVS2GPU!OSYWqvQYc#8X7anMyoM<JawBpt6(wmAc>})Z{^Vj&bhcj
zx5|PqvJv%STB1Cg9~C&%*MaA<icGA4I0lJXoD6rBN7*$aci9!Kz?LiJZX9hlb(*SU
zhPSyh3!=KuVq&%%!$d_F!7rJA-@o5!Xf>Tg@0_chGHElOMODdyD1EI%qnhp2wa~j-
zSWAMohPIIwXpQ4Yh?qh3(Fyakt3{YaptA3k*@FRwz^^N-c;hLVjZ`<A?n67GEmjFW
z@UE}xXWM`iVoj2_?V4L`&y}${ZdRtSe2n8D%?~A@o1wne`+VX;iwb}1_AGGnJY1G-
z1TAd8?$ygP3+7hatu;$ZLuZ>UKe*8`VrNH41z!IAhDC&-b0-WmUwSr<4~uT^*82AM
zhyXg=lrr^9XXHyib>uX2&ht?Uy3fU@8(@I{@B<4tj<$1QBiIG2zOFn`!ZDLVGtfCS
zy3QVMkK*e6_nK|8$yHO_=7dLDj$LL%A4LE(-Z-Iqt+An|T{Gw#B%lEBKtkxKv01cW
z9<JRme{x;ci8Y;!StAmFvSaPMqJ~G~zTa2}g?||P7|vLXmufH@vk{uhlTFA?m80L=
zht<dMcS|K(*El^uH|<y9*3}4i?dHL)Oq*$D&h<)+u+HR25N3wjP3P%bku>X_vuU=a
zx6QEA7vSvKd5;Ed4kf5LK=!Qb5tlaqJN71lFsYzc!+aUc^`1^UiBZwbNL{(h!rI(E
z*;Sv)Au<Q-c({@7GXwbS7+HOS4YaT>K@1m=n_gPBbEbM`eA3)T0?=jV5S;9M(v}V|
z3^)nQQx^PtYBvczY24|sPZ-&SYwa`xhvw9<v!U_mw>p-o_P|ynw=v~z+7p4!_R~H*
z>kiJ===`5;JkI)=qw#PE>2`i2pBP<Owy05+7`x6>U+i*AT``xalJMyppNr7xsD6eV
z^M$k#cWUaGv+gp8|26I9+GQqk#GP%I72^u!-Y|FOg9|euk&Sq!VfN;A9gznCjsLXI
zu8~)q+sr(gY9R14Gf@@3R~Nbwm9a8Q-5i4M5JX3(q^jBB)HFP&M~cC}kDN%gb;#rQ
z)iw3l7R*(%ge2Nsvv+J&N`MC?hF-a8wz<Y>E!0l<3guhop}oGD=$&ng4a;qzZhS4)
zGwnRnPLgZQ7t&Bw@Nkvbb>Nxs>C=RncELkcy<$F3V<BryKGaRg2v)WXv-H#b_5AdJ
z5f}utGnSz~)h}bixpVjM`F~x2<fxoozlW=Kv2%)Be~0H!bd&RK)j21*eX1n1OhLBd
z6al8I07z{MH6wwZDJ0RIz5Jt-YzQdP!q|0s4CZK>3N|&&0i0j?2xxI#Q9q@1*PfgV
zC&U$Ac~*AP6h{vtV~Clj-;=#=>KI=p9Wn-+`TY~^Bb;b4IGF^x&AX$fYCvOd#PKB`
z70(3oZDyzb0chx>4Xld8g*3CyiEW;Klgd~JKij-q=SEKrvuv%fH;=4C1-Tp9Ipv=b
zV+ad*aOwBJd4|oQ=K8vJI-lKQ70tax>oYQxMQ}H@MYpOye*ONd={EGoP*;1b*?_9L
zK@Z3FyWh3;22xbT)yWu}48XWp&r1u!1HDspuae*RNdoR4D&1f^U}MHvm&VM&Iz<K=
zpXa{zo~3-kR?NEyJcFg~6#UDok>yS_LvO+s1T;9smV?8JWp4teY=99kVz}P6x3hYV
zHwkt}0G-9aM_4xDc%|UW&6F*VdYWSwgvwXF671LAxl<DT_>oY4F`sufixV>&RaDnx
zc(V{+>#v{BlFEoKSIMt?c|KkIL-euc`q4e1wY+*A1~$yk;0M#lpBrA4Md!d^LRnZx
zpEen**+FTy2cRzccySKbvg))MY^+NdsA+dw7~f~DN?BM-or2pK(1N{l-IOD<P6}Kn
z^ds3q8@rcP&1cp$2Fkf2wah2n3+TG*DP7mWNz65JcF(7j%8zDoR;4yD904{uu9j?h
zb&uuX_AJ_re2`^kp9<9Z6gAMXR3-PG{_UJ&+)m?ac$3qaTg;xUmmfk#mD#()t?LMW
z&^<(;gsX_`iPtR_h~j@Qm2ms+OS_?=-SAwy=7%m&HDTy*X|j>o(<dLDMZ<hR9T)@Y
zxHftu^Mg@YQ9~rv+h>$5nv`V`lx8hqZk3vtYD{0hX@)!$tdmk%p^2ZM`XCHAcZpAb
z6K+$mCJo4v4gJxe&ZdWY>c>5BU7dH1@jxqhid=5^ux{DAU~|On^TW{uttQ_HSJFpX
zu)M2Ez8u|EXC`JePT#R)l%H5^a)niXv^F0aBR@USc{6m)w`FXi=*AV@1`XW1s<F5m
zRer1ZUw7@q*M%<8ta36u;6Y+_k*TR6*j$F{e4op_%iU18to9ymq-3V92=$PL4WM_;
z3r!ZzPM8nfzo&k{bC+;P*&dk|M$?;!g>I}hVb1dP{s7_~d&?r6z4N+;Y~h)pvq}Vg
zdA+o1V9ep|&Y@Txp<AtZUb(Y<A*-iIJfsx=U_Fb==%IXaMV6r$j@)GPm;c_w$Y~()
zz23J!tnginPZ@CC%Isjzd379Z;Z<jS?vWWjeFvG<Zdxoobx8GQStXl5cWGJS!c*o3
zy7xd#Y><5*vIaT`C;y;ib}U4LemiN?#`UE=53{3JXjM-_?d&mu(&a5<Nq}?E8xm;Q
z9`o29mY?hN7@A(`*PsE4aaNM34*#3F6G82S2R}y?*2D@w1D&}$xDOL^?jPOe)nU6h
z{$P)3X~6<A8(-h_M+o8cf=&?yFqcYH(Z@o7le*VwZrQT4q3Cn;!ogqEkqNha)gJJP
zNOQOSPFqyw=*E;|$P>P?$%vTud=3drnv5SjV+_2XJ~gBH*V^oT<AOLm6SaBkw!dCL
z<|-%ubRK%R)f<0c6Isq(GE<h^g%w0uDDw!7VS%a=+!`@)Lbvy!m6n$#14F@WF6hJa
zqYf$DXR|cWw&@&b#0;1N>Djm1n(9qVb8Qao=A(M{<=DGy4M<!(>BZ_55~<9iwJAq`
z=yAJ7=Fx(CZCisK)bO1Lvs{AP#|k%8o!0O<VlnO5Jq>TtEFvF-8fZA%mTA8*BC(eD
zE|KcOvL0mltjOJ+2Y0;F-A2?qN<ymo^oa-mGUT#0kAT=4*A<iJuyDQY05pkCf*2D`
z@&A0W=OCd{T%GHt8I`+i|JHPcI<a`Vdgy`Qu**YQk+WZN%U4789X<w_6HnDVs^Q+F
zRHU1uhbv~rE?-p1nbi!XjI$iHli4>5d$*?2rurn`Ev(+=-v{}>(T;Nh05ngib8t)0
z1GS3=i%!4T)%L+6fn=t|!c%l_UDvy5?9CfR+wWojP7u<=>UlNG{Fy0HS!KhbGPi=w
zl*#SDhWa?d&BGej8w;Z-b?a9#GZ15m7`G+Ox?3wunKJ7NZ_O7yeN|1x8$Dd87MBC4
zq@npQo2%92_CJl*2(Ox9M4jmb&U1oO+9D*6?yiZ{9Ugi3h~Bv(w|$&OQ?nclYnNym
zUdC*8<P(HB{!5{XMhnYSw%l12%dyXThFLUM)@1K6tY)}F94X-ZY0Xcq-^WqtP!Hkb
z9D(Sl)KEU$YCHg(r}NAmtYHHcIxO9Zq@D2E7*Es|Dv}6naYUg+<I6JJrw`<&OmLRt
zK4N-PiPH3xN%D}(jR%o5*;yj{@4$DT4l|CBM5PHUosIMC>rAld&HuK{ZjbROt7|M9
zTz`1Jd3`zlde?>g!J^cAjB@xiem*t|RckG@I$_MYF0IJD<!oAb?ra)8tl^_BY=N$=
zc@q&`c4M#a`q5;!PG4^pwae;}M;M-0Y>+`z$5lPV4&xyUU(uOuHtM3H-ClB(+w&m4
zF6%dY_}T7_(A$2m_cy|*+_L+-eqg!H2NYZaM*E-vC8KM8kb|C4CB*5M<{Mu~)mveS
z<>(Zy3;pEy5^0|r^r;Xph`o8?m8s0sVV^{F*@r>^#>&kTNab2+GPjz9-`1@uu}o;g
zS3JM@|9tId%iZa!mS9%xCK?h(Wu3CS1(iK($-=?AW7_<kS#~YoFJ```Zmb)6b9(Vm
z3nlk4p7*rw*SN$Wx|+(V^UUk+EQRgE%kP3A=_|T1a74`9TAH9k&gngP3OucDu2u7E
z;0H9r$%o7x%A8e4u`tE$@wDlqiI}rxf~Rywx8WdYm$_rerRtVBw=V0__n-BgJvDtm
z+hc+!QFVRgp1s60?+}rF*_#KCGx`8R7&h9()}}`9V}`3a(^{vvqK25%4R8eJ^M_XQ
z|Ex*bN%je_UXks1Ht(<eN!@FH!41j2@7i<q93CAt;gObycIj*N-h;IG+af&yq*zlF
z*R20fy==yTb|e+12Qgl88%SiIJ$Vhe^xW|gHh>UcaVVTr*9li}M(yKyatK)pHmF(`
z@Rm`wG-!fy?mEeugXEqb{uIqy4O1b^AyvwQom~7Rr8NZTbO+~XBM*Jp{n`bL!Z`kJ
zHkt`;Rn3`gm3xE8rqA*hRdhhJWBprLG%G)-?F&!REjQQKwck3^rU1j<oVqbh{GO)i
zd<$lLvpkDNG*5{Feb^XrQR+id@3N4V?Ih$EDr>9{`;e##G;4ULg^&0{F&Wn`3-p_`
zKv__kv7D8OP}R9Y2&^p&?>gq2%!Wr*^I~@JZ?eV7e|hW|Zsyo>_o)%L0P~nCAsZMh
zms;UfG^zb|)SgpvmE!mbF>mGhS-GDD=h(>HGbgL$7*i1g8#ne&zll-NLoKY!#&scj
z)Lz${bVoMgX1}iun|ge!9@C(pdGFfuVNHPag)<bwG+L&}0n9DfAPgSjL<pOgX526(
zdKpz&K8eA`_=7RJO>_1D9l_{q9y2>bn-_af+;rCnf5sB6Ja~k2v#x?KqcQjK03HvK
zAJd0&5*0?eVZ3WL<xQZo+=qZQWw#F}dkQ{a*Yv2S$bcRsfY0vuL^ce4l9+T?8$?r8
z$Eq5$^W+W2_|!dDu=yP??Dj6dhLlHOvT4&*S%&U9$0BlU#{#d|<IU!<WRGd4EpQ?z
zH8B9l{DY!1W7oUghkpyqV%0dLJeZkbmrqjPqqjznOLTP``3KgW`}1iq6jW^({q(FX
zNV~b)R79plur%Qas&{)#u;j4HEEY>vuSEZj-_McyMTAvl2@aT7i5?^>r}$~un~<5B
zE-ZV?*yi<$_o{Z*M5{&ESNzV)cFWuY>dJY@()Dp{s9X0}r&$$z<?ph!@}+xN?<QbY
zWfL2l9wH66qH8z6DL&|AE*+mtBFC4tC?Kpg85LYHt;g}jDBI(x+~#+NN-VxJ0|m0(
zIGkCFud8;+=e^PeOm`NvwQ>)9Uqn|+{(bFUc*hRl(O9;pz&n3Zn+a(b$$y@=)c81}
z^11?b+*S2%i@|}?lnEc<h_UFI3`1e@si207Z3pZ9H!<RXO6PVr+=@vxb3%=CU1F=6
zr>RA<lE-&3xn&V4pHS}3xX?RkX7H$tnBf%WR#@%U@~NK2DCK7#&^2Z9V9pu*{1E+%
zS?Zc8=XBL5byrQA;D}PGw^kg(Ff|&@X#%2JqFe2D!oB+RrAK=s^w~f=HNU_yxp}>F
zNgku7Re21B(NQhozOsqz0}W{TA)B-Xc*Xk;CnqQpN;ymC(-DkHv(NM>;Ssg7{kqui
z<}#buWpH}+4Rz60fD{UD(6TMKgJIRCD)>UXt9QkuFb$Q76?D^(B%R7^&Yx1+0=nro
z`M)`Iv$rPd0Hr$TI1)@r)+JdrxZm(G1O|s=e6GsK?WqO~>6hMWn%gcJP^xM(35*~e
z<)-!-@^r$0!Rm|;7^Cy@Ix-I*P6E&=YjajDfA`mFHQ4#VU`;S#8kCdJp5Ed~m-=^E
zsS|@caYub{%p*uds+(+}_J|Zl9|u@(zXs;vop!Ujhkq23WB~ZDFhbd;msf!7$^@^)
zZu)gSKQwiwCh;!OzwFr%uCM&fpo%x%Bp|F+W7w6&>fH)Uxe=gs<;V_c>mv3J-wDoc
zu<3@Kg9Y*#CBGGhswrc*L$e?@kLndws+19fq?*|WzJArusNxHzV`7HJdIEqkgzKqs
z(jAcbkSycs-1>w`X%?2b__ix~pgOnjJhB$I7IlC%?%j?z?MzjOq3kk7X8ZF@_0Lt{
zyU&{*hO<>SqKk`!D$R*(fac+Y4}!5%KA)Af2vokKxiuRv`xFV?Y0QVWt2h6F9#8qC
zJD@5LXJ<9zTDIRhm)jFtp$#I!txw>^mo=<ka{8&$wk4GP70-sjOnt5WCKeNmHC&0K
z=d)qQKp10;1J%ps(_W+X8}6!+oh{iNZ%zTAYS?qe`9`Tf#2F4UZUwuVoxId;`H@_b
zgmLZx=N=uF8E1VnFo>C%b@^TU)Ki);w@mlBIB@81XZ`D;Jdj1S+(?<_9(SBg8({8-
zFk|m+D=V&WR36$`&8Ftmokiy^i$nELzI_E~N|1av?XrU=L}LgjMkU?E|LQ9D<Iw#C
zLYzH4rjre~+pDFHRn>!#y;lC!-gW+z7Djbf;p=8stEMwB$aT7KeZ50tSMJ1TGs}~~
ztBM7o)0Z}_Ydl&Y#wmk|gN4mchwK4C^v-tAL7~LbZuI1*c;K*RX6zhSpxI8q*`I$V
z=xl~{R^z&$VY#$sZBCl&X*V;gGVU}qBfPQ)2sEQ19dE~)lymloODV;wr*L0=0zDEV
zx{f5VHhn5#*&~i<_byFWM=kiF;|>##=q(?j>D)`RJw9L(bQ1}8tGA8de~o;1p7~XO
z$Avy$U+G#x=q;M%Q*#S;1B|M%4hrzF;c3judv;>b6nQlZ`xWZ4TMvU18;)RR9yMxn
zLiqI}+<+~^Mi}M+A)Mq!tf~GV!h#MKD4Ql@XA8cDOKgsclL)jCR=03gXlLyL59`_c
z)bVqvlS@f=6oLusq-zG7lhd0oZ8rreVw`+Zcg@P{@?Dp`iMZ;qH4~XES&)($*i+>2
zu#Z#|G_~C0fW;yCrL2xX?~!9RzOJgCi9HRh6HnC)xCaU{Gh6Y$)P%=SBs&>uRbtsx
zB6IxQnor%JO}Zpk!Mi5}cr4VgA!dorIo6DDK}`$axmHyJDhng9<!cZPlYSh**qF{S
zqak7tx=pCc&S)HFf;p!jIT*?*=52<-P>6+or)Nc^44W6zJk#6;pENTX(=LPJW>xmt
z3xjO<n$Ee~*&z>_LogLUbK83;gTp5BW$3JPFt5g>cU!Rx-Knt5o=G7Mpm}e*oVCZ9
zd>+G+#%lu7-Ls|}QfI!x+xMXX$UMu9n&-K)PcmNP@e==R;rk1<cfs6H-L#2+&!?(q
z?L83D^%cMO@|`prMC8o=fX{6o6DSn8)`xBei*WS2uTZ7XT30@U7NgV7@#em|HQOt6
zDW>QVYd^7SHUT8HGoq$G=XbvrKaa8C;q@TqH}=+?Mjkqvs3u@b)w1|B4pb>D&rMC0
zWX}>H3+}3Q4KDkwbI6;R1B^2ob=z5R&iLNGuKj8zKM|_Ny>^>+!!INBB;#i796Op&
z3?u_*r(5!{dSi8FiQ%XkhLgG5k+5{RXUT57G8}PoiBzn73lYL{^Oe8b^5Gq`D|cjL
zwjh6e)!zG+t86*D%s66gj{)mxl45REZlgYI8m2<dDJMAL$e!x<tZN`T8jgku?t6>w
zR@-84cOY0ISvhx<qrn_3s<|ocvaUAkLN^yBtTR2R32MWD;9I^N$)R^N=$*FMKuy}t
z!nAX0{PS=jn$|LQ)>Qa>p5I3~IlW>zaugTRd{2(8XVPhy7#3|Rrbl2f>HR(tMy`eK
zR?9+}iUo1-DmC?Px{(5>5Uf5BpgDz&;_y_RPY%?h6FhD{DV9o9@|UZq!q;Mt{9}FH
zs+}<IA5ooSj#Yagv%as~(_U)YOxcd<alVNb);l$!*sA#^75#X@hu#GZU9ae<HW=)(
zzn+OIdnMhODRbmFxLV7)n?adX9oc*BMuah3o!PdOILxK6-23{fEtReqm<lkZP=Y?<
z;?we)ZlT+F$m{S7W&%4Ou4;ERW9jLGaj?l*xuFL&=V9ikcUz1#-iOsw&5;};d~kES
z5lnffP4k`_Y9C-W=~KFQmX_yk(PM_#rCDtSa(xGOLD@4l59iJ}HyHy={O>q*1sW7<
zzt>mwT8PU^s{K4~&8@mTref$ev(0lUFB<Z*b}%%Ssd8RTEt^5O9ZGIBKO#q9$m4B6
zx>u@;%QEW-keNE|FRElHIP0`R;StygjOV1ysPp!#`O9UVHiux#+F}4Uy|ae07}u5W
z<W`4bBr{i51cz^EXnsZRuD0leAHS>$Hly})dpTHOfzIl=sH3uN1x|<GsGicERMl8P
z^yKx8>QtDz{X&~PuV_jkuHoJhh)Y}kb^rbu?+J+s$Q!b(i*Yh`Q(feo2L7K#WnZ3k
z=&Cx~$uPPNct7x@(LAc4XZs3d%fr3W>`5o+W|`CDC@waz(?SeWH>`4K5l*RV^KaZn
z&Djm8G8+2y*R$mj*av>l9AHkhbo?4-+Oh~Vv^Q)y&~!Ad8}mYf8dH_e20bjCK)=Pg
z@a4{9LHPWRSqNj}%rkv?SI^MMjLM>gw#BIG54iTgcu_i>Sjp|+yBlCn_=4xfUEN%s
z&2u?t_t?WhbOLTh9Pv|rnit-$8~}qmDVQU(eQ~bpW@hGK*4RcxQ$<-9^w+tL=yd%1
zr+KY@{TlAo%zX>q9oF3_S^wZzXNyF}MMJk1D~q7hRm<}=VnQQ3C+8J(ixG`1-(oi`
zJFco$Ga{6ur6Yg~o%4-1w=!l5ux=D=5&dXZd?;R$t;QAalMTO-^Eg9i4+*t@SFA<v
zZ5Jw4a;s@HfZgg;%;msBxr=z@Y`<#zZKxD%ni;Te#g{#MYSF&mQ~X%pdU{x%Gq>Sb
zirBDz6_khJ-D_3l`U<{Jm8d);qMa5_a!%jr3wjH-)aXnGNA9hnqr-vAU|u^ryW!Sw
zx7vW3|2-prOu>(k8SUaOkAu7?dUA@V3O(}#)gWUf`~Qo2gC$FnTgg%tfVoE$tBNxt
z-J6;J|H9X-40i*dvw+<sMYv#5$jpqq*WC<IJ!c4-ouBgkNTKwRe}L{z)NWld>RU;g
z8ndl_#<A#@#c)*0C|nDjjc(D`y`JZ^2~~pxQnz2s2@~qrs|{1nSNFM#0Yui!09ymN
z@Y~)(HC-ngGH0MR&S`kmCn=#x*`-}DivcL8=oC#2i+=K~KKj4!8aI`2HtbN#W4Aj&
zNtm;6_f%jR(YyLIIUDQh)W@UjrZS?oE}y8HWu8z+M8AWzYnU%(bq!#1CL#jC{|m9Y
z%^a0&-es2MZc14Aey%&~vM1GjdI7FXjf8Wc2<jQT^{nD`xC1))!-6{vIYB+RfvD{<
zN_Rpq7#-+FeBo#3=XXdJOvM?{4BFbyQ}5lskBe&aK$d_?I!!pC4QOOjS0sa~8s6=)
zs#&J$Tx)@@2sA*j%dg(mBTZ^s{cu&rmE*G$N198$J)dA!!zYLLS&9~@F26V-<94TH
z(Qi+`unS?#J=?ftf^`;RnPo4&Ij$4SU9fd|XWBA6ug+ecJ%q5%qAfeKFLaB7&~ow*
zn>`Xaam*K;wJ!ALHJCPY_FFa-rfiQ{=Y;xSfA#*^g=N`o?p|36?sg8|{G@(4B<Jv*
zXA9PSU73RQ?%|Cwe9}X6Q?X2US91k_{UiDzuLeIm|90$|Vv2{g)!iDbw~c;`=CNP5
zI)`~MBz{(JQxU1`p)I(p<(Kc8(TWpORuXJQJ_ASKr?Gln${!604G~f+rm)>w7g60h
z?Nu7Md=@8zT{@;L`Px6?z4jh7eatm;#mw1ZHYLNd+Y-!I2=8J1b9S}^D<?a}udn^?
zx42AQRs45U&bdC6IBbr=ms@R*E+;MUwxPOx0aCDQb1mO{Q1}LAO--5uI2zVLojtLk
z!NF|c^Oc9il#7NaeS~w9A|tCW&yH|(wGHY-E!hDm3P#d8HF3ldoZD~|7%VvDt5XB+
zQjb_KY?`>5_B?6&nL8qV`a!}uQUb<ZW4h}ZJ5DBxT@9NydN-{z&U1lia7attB~#g@
zPo8Y@5@DU6+3Cba`|$n<T5kGLxR3A@gu2qcmMV9}Wd~_#JzY4zZ`;4x*%~PCq-u=H
zfcrI3|Jf!pXm~c3tJ<`o1ml9Rx@<|8@2vGNJ&Jx4=3OPT))r&Ainn$49<HqgyEKEW
zt|?S~1{KT(FXl`SibA-fGN*=7InsFfZLlf@vPz72bNo(0&~qkkQg^ruyvCDjjqa`r
zS0Om(aH*2+uG$vb^DH>xS(UrtNq594r0yJ{ICH}n0BQSje>S|UuW$RMrIO`q-f9aQ
zZZO~UxMpTImyQXZw4m7**f3$sFVk)y{IY7m$jjb#RO7n(%`(6i`YF%{Pu&~e+*xy{
z-_=!3(EZtht=p;;U!MF9bt*#Qy8B66>qrq%QYJ~ba=fx}k9b53H=xT*GUgDofd~#|
zsUgMMk8If)K8Nq!avzJsL>)9tvf<C0zzIV8(TjUO6S%@mu+0a<Jz67D=7UBvBTzN#
z93y)l5HvxvX-aWE4euUU%Bd??1A~2ldCMj^rP9GY9+xXX&<$Cn=(8~aorxL9YX&2y
zmrm`900eZ>9bKSBSm!<_#i=k^%ueN8JC3kZ#*ap7{7K#Q4&_*F=9A=XMs!!(5Q9G9
zJ~Np^fHnV&&ejTv>*^oa5Uz(=_gS>*+0#+9aTz;iI9Q;$TkQwyT1uckBSc~sxSIxB
zY;$0A5hbIt;~G3(!YX0likuLH=%ETfbNGS$Pm~)|*|_d5Oe|JKlRgW!9(JLdjR`rT
zQj-ow(}6)~4`!~w6;-W<PtKw;0)o2Iql*?sWgm)k*7$KcE{9pZ(+(kR#R0Qu4=e>;
z<rDVSK%?OVT`6{fD>Rx}^Z9*a<fj<LSDV3ZtDQW~o=IF9dd@cN-K2qJ70t}7JFCN6
zP&RFRi2p0h!fWrcq4%eZ*@xq|uH3OEMR?Xy<48$$H?8te8AoyjB`$6)>}L0T7Uqp#
z)d#;27x&1s18+IIWIhvXUjM>V$IJrSy7BJe9R)RmY-NY5V1fFy9>m|@27riD6cVfl
zN?$~Q63heI^6+{}W^O6B>W-_nnT^qQJiqJOX>3GzXR8Yq%svByf%2X+QrD<oHQ8^_
zz2w|-_uLGsdURigh7I#b3~gb;b*TvJ`G3a=GMFDLo3Dk+2DE#*;3_z}bX`ODv2ja@
z%P|N%kmUb&n|=8|{~%VMPITw<BCcUh)2hP9h^*uIGoM<r81Zf+ZKlUH&+2JAkp!2&
z$<xF}wliyl57sFJ8$@w-Q07*Xq3TY}k!v8SYPkdAEE?;8AE!R+3&@gY`%2?^HlKat
zSk%s^e!ggW+S%k-m>Y4eev59EXY@X+YFZCTB2YQXWG5<{WNBT}AoBH5p5Mc>ruJS@
za@%sFK<mslG6MB6x(^|*KB!n&uwg=$JJd~(tO=@fbcYS22#qxAWJu;?(9KVBa;Um*
zt&9KfmmB)Y2sg`ZtFp3Mep}M^Pb-sjlLczF*&iOI8rPL69<;1!mYyK(DBTZ83Pwgq
z-Bf1(UVDsr)fQ^@EAKvF_4(V#Jo#FhBgcTwP)qlu{@ezG%bL8Vlk*e`L(lfNjOuVZ
z%m<mFcbhuCtaCTqEpFy6^TX1ZUM`fqR!_&jlT-1W3)Y;ECIESW6kqGt<5rTnTRx)_
zGV^(|%%#Cs^fR_BXBK@an-Q4d_{?3anzz5@m-jA#7R&lTBFugFq?kxUXRXMhO~N7E
zes*{R=J`PU_l<6a@9cq|8O;N4A1PKQPumx2`W5f{DR|T@afhSmP^^6CM>>t52@~kU
zUuAgqREO>K{}_0RK45yaOk2)Fa#l+Xu0>Vb-0aMu%veK7bkg|z_%iGyIN0mz6QZXH
z-DaoDIpftK0tI`l?A2z$!a`Nj4O?N;&~CHT3d>*RPAIHvH+;>pe}INVijlc6T3G9D
zT-SahBva}{_ZL&ku{WH^UY;cjGm#cv)j%gWsT0mK$Sq%*=vHmOy7!NKh4;@E2!#*0
zafr>$Dc1`yH|O*4drh!%H`R=ZNf=w~zP?fSg{msCxA<E)YNi|l?9}|+CU1dvw+wwS
zR}er~*;4ixY;42bs^}OKpib+8C|;Izz6@svY`1~zI_)Ar*@&3VP>XK4Vdkf0(UoI>
z7eDCoGZqo(>KR<k&8cSFCY2_Ak0IW+G`Ox6Z+f(-@-?lD(ZFaxD?|C5kLk1?8WzgZ
zy27&NcT643%HiMuwDB6Y+JpUWbHI64Y3{)%#|F7jzirfvz_{PKjxKg8RW%rF8fwol
z)C}#>FTd%7MWT1jFERc0Y4+o~(7P{g%+3)JYWbqp7yH@EQE7MmBhT{aoXr=yEZEt;
z7K)ZX=G|_DEDY}(mnjuYcs;XVt44hDLF~t(wY;j`dpEDEvgA`28Bmy;b=rj*WzZFs
zFE8bk*8HQH_)L7y_IcDRlMWskY@qgGC=-BRe3FnY6<w%SeB0Yn*)(ud7G1Dwbwg9|
zjfqGdaDZFZ*Y!LiC(=K>B<G3x4?iZtm5t8d^A~52=iGF1xfUzO`=X+C+FF)-%dOY9
zURlfC9YStfnTOK{S1}g&Z+LXys|8zg6zLw4ckY~EMT*2^ZY6in=v^??P76cAJ7vS5
zB+jv#2dcY}EzPaYYSw=A+7LXffG%IqjA@n$*zZs#0oU8~I3uVznSv$buCM{_KoW!5
zwDS-+@DTclCqI&Ic}x{-cX*&AvdV1!fSjQw*qEa8Ecwe&@aZMW?xmJl{xrk8+(u{Z
zharaUCf$1vCx?ghHr4`Py*1}MRh7jM$Rno1NH`GDDQFJWfLm^3$>L%P#>_dLbaz~p
z5m`DIO|`-gObZ6Of~JR7QX^>Sj+oeh!NmpBR!sa9Hu-xI{Z<udwt5eO!Gsl6(~O#N
zJX#*;REzFAEqoBEkU_|v$7V6Y&&5l*3G!>Sb6|?$*VQ{a%s{m-Y;95)xTS^>+9;Yf
z++f}#HtFP9`ON0H&9i3WWOxOOQC-q*pmn}LL^oIOA-`8Og`x&a0?qvN&Yjt42RsPD
zGR?vlI(L_^s{5Lhkvra3l<ea`0t7eK9=>tE9Bhjw*-wc5;tEskIuyS8HO-9NW8=tk
zWoX*YNov%XZo}#fVw!(z0{f2L^T2CpLits?zu#x3_RT=j4?IJZXb~%NmqbU+WyzRB
z4E4qCY#C{x-Kt)fquNowV@5+AbgMmAEHe%Ax^_-el<r4hIlH}0cY9v>;?4nzP67c-
zZyg~4c;uRraK%n%&l%H)q&CRYK_iWq^~?t%O!-!mW~_GZ<hq$PInQo51P90GG%OIP
zr&k2On0+u3pGz-2@+EWXX<ZhYu0y4GKn(8O<wO^p)^aP{j@rA#FsQ_Ci{(eq19^10
zSDVf0%V&^3GMV<PUV(pcfV{@@T?B|}6zXtv=h`(}&JSb_q`Mh3hXkHwsXKeH03N5z
z)pg>+IyTSY$4&t;vl-5+Y_m_NS{9=i*QuC`!?`AboeaaIqd8NH2z0|x^eP`ewo3Gv
zD=`lUkNdIZ<9eEh8En>iLuE0uNwuPL)LG)`2>4K2D|>jku#f86Ty5hDMsVUEYOIwd
z$ilh`HdCf*Oj4siNlx`7c0x~b0?VTXaO)`+6Nj&c^61@i%Sn|D>$|x^ea1IeWES1p
zthRywrn}>vW}HfsC}D{zTfIN(84;>jRaux_^}$&BcL=QAjLWn$*6M$H<&W%KzC|;m
zvv$hEdXFF!gI@DMnN?20W%B!o>-O9!r=%)(X?e*fkLuPnyrCMT*|7E8QFYEzhpmq)
z+PsG0rVR;<i%((>&F*vWoGoEBmp5EH3y1)}*Yoqj-Bg4u@YcvW7*W~0unXb61^Smy
zH55<{nUolcju0_~Pj>OhF{VDz`k;KxjkRYwh<OE>shv-n!%W%JA(}4C%m;y)({^*+
z#{vLKYgzFR6T(L)y|CroyTio`FP9-RJj>ns9T3zMW)7b8z~9YLt^3a1%_HrRIaE{1
z=!DS>yUtvY<0LrcWw_@IuOS|Flg}uF)lC?yuhm&IkvUO?a6>ngWrh}sX<XJ2$!IpS
zvCqi4XV~0`Ftgh8ML%%l^<{c87Khw3(?6<Ouyt$i%^5mYd-QHZREE#?&s2rw`XCgk
zpBF}7*d@9&Ah+><v@ARm<1J8OW~V&6&}za|vWYwD{L7|ohVcX0%{|cR5!6@1FjHw)
z!l&ZIe@!(}Su~sP5ba|&_K8%fYg#{c7V1Gwsjst##Yd+MDoK;2XON)btuW1Bw(*mL
zY|Rd(C%)tSjHI-zg*<eNGkcidM#KnIjpslpTiA4TAv}xWs+M~Vy|PDH7t?#d{6a?e
zYG=`UH>RmEvp>;hUirJE4>^Cfg{^C@OEcm)!pC8X%|g9sB2`WFL}F*zF;0Tfd75eb
zx^x<59U)pcx{KBlt#>SgA<Nx6t1+tzI8P#YG;*T>#foPWiRGXx!L0EYt1~y9e+~Rt
z%PMny_bWn$upXVJSvPS`r65p}F9T%0V!tgj<g2xo`Y5NnTd{JlMZKGG1!~X68a(@4
zsQ?SJdg0x11)fca3vE_?$ImM@)<`Y|8sc03OeDPRYrTn17}N@v6yMfe?{K5(lP((g
z`GgMBB6F>QFASbQbH<>1{u&w4Aq5I_ohCTdt#zr+B5){pRT43;_T6%;4#$A6y1%Y0
z=SRWqtre`FjiXU1*%dQ|r`b-j5$?5Ty+Ft=xA#XNzSh&1XLiecxQOh;TG1O*ROH2~
zgIJuNsDak}w-K3b*FqP)GrY=qI+z-cHQd-qpP%~-k<Bzra<^((bWYQ+W<jEC9g4aI
zdOjui>=Rq9Rz#qNED%<iFiz||jPE|hAbvO=ZtlcbW6yb)i_OzhH;f{njk8lC8r1BQ
z1iEU{2%26F*!a4B{(;4~b1jG-26RtxA*zX%*S5<<%xGPMIU>W({>Z&I(72{&c|2f}
zt;wnrw34$*pZX>tS*JJ`V^hj)^sL=nR<d>VTZbJ#(j-=!8?qhAwPHWCvTu9dK9n-Y
ze4GF?H2)Q9vnq491KW#Oc;9oe&-QCsR|@I~HBmV-VYFgVyUks>i+0ER0*$HDqjg2L
zI!nHsRVO0jo-h4$ozxAVRgVqpfabyf{G1vzrzU7$v3GemD`8L-$B{hGubyu2bzzWp
zGY4v2EB9Fn507RXD?Q%cYvo%qW+B#4b`4D*B#2}dc++eyZUet#!%zEdW+lTCX$|eQ
z=>YNOfGo^Il_j<ii*s(76qVf2JyP^xU=9tzM7J5(plWwSw>)ePphiqH0RXXg+t)&!
ziidmX=OtMQQ2g5m^QV^DV~VD`Y`!qpmrima8NKVc@s1H^t-T|<X|{jar8+=WbJCqX
zv_z<yYgM}MvMK8D=m86UMLlx|94=DX75sU`;nz<jJKS<wT&Imkr{vTbcEE=4603G$
zE!Uf{sBP8-hN}uZxq7ov)|qA9cjnY7eC|1G8KtfQGihrgE6A3P#9`UYo@=~a8ys*r
z4yv?4T8S3Q_85fgY5NeP<`QHq+J}bG@O}&GSZI<%ZLoTNi8M?G&0<75VNwH~(>!_m
zTCALHXLcY*dPVa87QDv(Q0<o^TgYM9K;{vZlaA@?ZZ-XKj4+t26qieNs6&&LJ25jl
zeqIlgEZCW5(3oj>Kr%#c-mIFvi)MSN%7P&k4zteD-0V|<-gEVB8e47ANtb~0qSf%R
zt7J~9+swmIdml(CtImYZnIqb@#ipsoB(uD`fe9%$oGF`AHqG!VOhWb4MnBV=K}Pc9
zZ5Mo!L=&PvnbYBW%&ebvv>pbvoS&`d5q>ly0-2gy6MAd3G&VAA9V(kG{7R4Dbor}!
zb96+2_*a4xI^^8Us;)piiKCb`B8Qq(iW%UurpcOpU(aji?$Gd#%!Dz5Tlc-5Z9d_6
z%;Y(kv!$&TEiV1!@S?Y9Y}vZ{r|AyTFShhPW8|SZ8m@}l@I1xE>S<9S^Eg-tFl@iO
z&0{tqf|+yggvHwLTGdC|g62r9es_slYKFlBmF3ncyFFX#S?2gC`7G3@-QrGp?8ieb
zCETr51%@wEXC|YHe))u)QgGX86kKm%bV^knx3{h5XI+ySFsij~Dzgf9nYzI%_M7tX
zetlJ5>v@fcZ`Mw;C$D(RUD<Fn|FSBY&0rm8f2Wpe=0#x24SlfhM9uDFzMSJzh-J!}
z>t#<sSTs{<`I-9>MnfV(wZ(x0FG~bk7SWjvn>`7>|CPTASD<#E_{=DL)0WC&amt_0
z+Idjj=A1)<0X2yXj>u(9AQQGV%ZfoMS6g7(R0UX<xzwYFu{C6Z`8WNzddvV;%BhHI
z$M87Xya%4s5Qc$fEG)0jo>I#B#tUS%Q*c%_J^a3Y{pdtjB+U3btY+jn4-oodmtr~g
zDAQtLXPH-#ZLas&Dfrfk(!}g}rRZx}KjRv9q*j<D?F|c3Lp5?@DsaSc7a87XIpqBg
zZS8nOFq>Rg$J(=4>l)|FZo``7orS6`nAs`6S8w^TK|f9QyE)xs2wY<zndw=LEJvg)
zz}DrO#8%A%(k?qh!<Jc=><)I7Bp>GxK|^ueMc7p>TS~|&E>*IdT&~*VWS`KbvANuP
z=Sv%tUcqydonb>|+oZt_`C@FpimQ^^(0P|>D%5n7Qf^smY1-v(Rpyhr2qIHHz=57_
zM~xmyBka@OYgMVY+Bo9T2o!B#bKj#NS2lEpx^Kn;NM$121Sv~?4Uk16TIw199Ts&R
zdUf|KTH%Jw^QIBayITiwc(8dR(f}>;;X*+$(FnJE%~7P;NIaS|i}Gcl<=gW=G#+}h
z-&f~ulW3L4PSm7zjyGojbuV1{x^mB{xT?*iW-PB)2Aukff8!rH(4$_nBUT%J+LTgz
zCwzHjH#`jU33Bb><e2RYS5>tb7YYo^f-O{*2YM***w3C#<tN)mfF(on4oDMr-%YfC
zO%-Lau#hb{$3XtE@Xj4>RBNbo@?YZ}+==BU-Y^H7#-RQM`@Kq~y?Meif;-^o!fztO
z`phH+@0lins4BWU<X&~UrSphGt7U8AQ6-`oOgP#slcWt-=BylS2Ju<uH5F1+;}eK9
zNitXO!@@X0xvI%Z12ATKI8$>MA)yj4uY<gYp*1D5FlK&+f>CcNVs?vyO)T+5TcFWr
z4?idlOr+=wKZ6d)Jz(sK-nH5BcRA`!^;nb!E7*PAYIHKN+rRGT2bhw~ymVMi530I)
zvJs~cJmID)qbX_iy76xtnfF#W>uojSf-UaiT!*2_ycewAob?dRz4HE6Kkv928nfu~
z>bB@dMhcCKyCC`niOWV_d-Bi(!TB-K&oBpldh^2cS2?XoHWuD-L!U#Dh0-nu`k&Xs
zD6@R&5LIx;$5mi9#w3TTPVo0Ia%RY?H`}9&B&vf)dEp-JdUyJb$};C<h<jUfFW+fS
z_j*q3)fyf?%l4`8ITUxl)lhZVM3e!-ma9knZ0<QnYH+sVL}@wRe+urI<|I@lzn8x+
zr{M{WlzS43fu!eltD2HZ)0F%(hB7&`?Gz}$V>Fvq+4{PEc3}6xd{xKf{6{h?$u4fu
zfX=30!##AbP3!6G>*ONCXT@}2*WqYsG?+cdG?Kq{L^O&TeV>$VAGrsHAaCrFH~#qQ
zU76F;0!REeDDvsa8d1Oq8Q(1nOvHP*t(@SA_W;gjo8HyEZc1FX_kfBp&qU{mQ_;*@
zO?Eb~r4(T#t4a}_#fu3uzT5mWSm9JhasGbjamC)*^LQfZ?3(G<03=LsJCiy9*gcP=
zd8kml`BTAad1MJ;0=wf<^4~6<Y~v6?#^u|h`TeE$-4R_?+3qghD>pu)EY(0MAu#I2
zjB{fmTB`H>IrF}10P$tN^0;dzL)bF4MQzx!E|EU9b<F1)mGtHB3y$aM^JCHY`e?T1
z9$O=7wiAh&T-cD`r>T%OK}s8j#|)H)xi<M6iCmV=xpUc&B)#8v{zi3Rq>!MUHp)>3
z+U)W-HJ)X+Rz#K|e@miuo)Tv2h|ZKnP+gew#H?;T0ATOJVm7X;la$<Pf!^=rEP$?@
z20wB573!JQw&2sSuuA@E;klcaH|U;wvUIyN1US>qi3-8JhhB6tQV_zs9lJ;VVPE(A
zQ{eTbr{Lju>IAMUX_V;LBLST-%vqv(9G{nR?otzHA7sB8;XBR4(1!+O_8G#A#}=n(
z3Atr<%P%XqDs@GdV7=j^C>qEEs;X(NSm@CZ>59FLZOi6fhx|y~jb>d^80*oDC;enu
zb>KIWnIDuxwTc&Ct>w9|`~5jPe}nqEYSyx5#`lLl(BD9stE!<7Sg0c^IXxOQ=>gnP
zrOwujK!UFp8-B&le`0l|)?oS|-Zqo<L1_nSQZ<?e_Z%afjq~hzUh~h+7Lx!S(1j+i
z+f{S;>h{%gdvmr9V{Nv{%&`jAP2yn2i$2`Pftfg!eY}zY*zE=-%#R}%#~E1-AUAC~
zsCB-ODBjpyoeeR(WYY+8ALu`hEu~IhbKtRpGxM0AyXI`Tb=gRtYn(%)pD_#r^K4`%
zYH?2SM!m_yCl$slsBXnVYT8{U7~6b5GEi`L)Nlr49ZI8yA#-RiM!AvhlM_Ps=a~z?
z(6M?;piQQ*(#XHqy06@=8rexKN6GuAmm{-ruWk=lHt+lSsjhZ2LseWeWdPm0x0DLk
zq}NqI+uzs!`I%8_H+(HVX^>6CuQ@9sT5WRxy>YYQgyCmLNF1<mV|ng`xg$A(i=v0N
ztP!5&Gc~M|KB_!`fFpCd&#cXGo2Wx`8NR;XUk&wEP(sAs_g$HD1}3V_0)u9*neaZF
zakJ>eAj}JwcXi_vkJh}XC}_8{=Kt3sraY$<yGs#*(I({Anqs_wxYXf^i|^Zqcn349
zmp2Nmf{q(!ZgUNi>+T1EYBlfB$FRkJubDVOS)<epo-O9NXViI|n=}*Vf*&|b&#QDg
zEJlsjB$zo&89TylMp)JrJHdITPECy6MaO&@;e3l`vFe2t!iV8$n9vDt&)qR)W)Hox
zdPVYH@7yYN1ivJ)u|{uIJ##hE%=YUstxb~Zh*QPS5wxIX6KJm*s6yG7eE?B>)7OQc
zgj=V_1jFcFIbl;X>n)m1_kG}y3j$l=j@-%XQ@^TD+5ZR#x@$Hns1|0sFv!mw)tANL
zG(JhNYu%Ua<LTO5%8%k8RMjx+=p0b_XnH2RX*9&ey~_jjiF2AE%Z*qwCN%G^_C>Ov
zgh$Mlo^?761YU%xsceT;bx*|~ow525ZJpMtG)pjK!QG(7Ofs;lq!CF!B}fvP#nht9
z*6Q7ct`cNufF;neMYXwwM0k})|D;%>lx?uP#{CnsNM2}Es}U?gHLaDBRBmo)vvuKp
z+JNfo)_%&V{fzaqtIaXjDU~zqq_-Pwj(Vjo%PD_#Pf1`-oH?^BLYXiyMqN*H&rO!#
zB=Qw#Fbj6IqqbwLzjLi5oD`J>UR1vPGa2&;Bj6XYH#m0S(cT{)*a>AF?^NabvwsQ_
z{a&gpfcN!2A7--6!BHwjME*1)pt0Tnl&T@xwNP6is(hVw&#cX>j-Dx2zO)t~|MGk{
z=@_ElvL?g#KE1-Vv?tYIZe8%sqmr5w%53htW}M(2dpoyQaA=kLjmSKfD34gd+z&2h
zp$D^FDT~zwc7<gdKkw`*g)tm^i?C~{&UdhToNSy=NT%Cx?dkD8>v|p;JxrN&vv5N7
z6yoYP+x)UBSkzPJjyyHIuD&kKIIDlShL|yjekxo1L-cC+vN`&JZZLKAG#|{X2&raa
zHp;?oU3a|!LMOf0QwWDI=P+rXAM>NG>@x2pN<QI}v(5sCI^KnelIAO)${&fYf`xjb
zyUS9X4oD+B+YfS=%Q$pa;@z0p&UC!0BhH5ecA(%#MN)J#=P<5@d6HT1+pG9#C4Lm2
z_<WvWR0<>c-IJit9A4D?(42!o#QCbK2N5tq<-{1odZ02SAv?J)w?W|A>?~_whdk}9
z$v;g7c%GhiAdBz|HIv;BjAuB_bE`Wyck(Zj%yIw*5W~&QuuHIsthiMTtJCZxx<Tu=
z{rLTSv4>(;<9QP%Mb(W%ijQNM;RJ)}F_`ULfc!$c?YFsHUAFRV_zGulvGz1FMrB9$
z>bM*QpFrBq`<7<<q*lcA&7o?<@<BLA7O3u_qlQ(Z%BR9Jgz0&!s%w(PK$dW{huwam
zgs^PKD-M3Lj-3P^d)vrEQE>Mz)C`Rs{kfgzIm!%&-)cf??zz9dDtk0cHkeUK#X!S!
z^uJzD@w&RrcFuJ=RKr^~72FI5*M|wHvH91WDQX~!<(7M%_RAr5R!5A@WTd6Mf)H&#
z_T9s>p}KtW<h(WHw|SdR*8Cg<_P7W-X^tFmmClNL#m`G3zVIvQrt<g{#b(%WOq#{z
z=CkGNYDtV`S)(<8+E?y_nj|2z6^9yZrp%+!E;y3sa@GwCqLL@pubOd}gt^2+iZyYC
z^Nf((pwjdk3v}DNK(mO>VRD97m47++i&K*tRK8_O%?q0$#nqWT;fU<^81;0HDs6sO
z1Yl;}YpJQd6YlIXr!PFkk3&;)iJFuV|MvN1TX5H|!y^YzVBhV$BP9*g-_E805ZtSb
zBS1!Xw-83!6z?(kSkqXFy4x-vA0(=;s18$i)(}*k)ug(^_x1x_B(3t`!01#fR;DnX
z<<A~sQWBXKD8ue+sX?nP4td+v^sezfXJHPN!`V0x>v{Uby7cpwFSi`~>CA`+KrG0D
zZGNz!SS+#TlrRrYq^1Ngp&?ptzyP=v%&oV8&wc_9tX1=IG^?Dw+dRHidy?~o#X7w*
zcFk;X8WT39!daoK+J`tGfN1{xh0bgS)0Q6bx~gw`pA;pF@2hrpT&w;`#OkbeR$GJ;
zsCFL^fko_XEPdUEw*=(iox0)k_zvslSj%QmQO%jjbQk{kB2cfwmCuL2AvzK?2e#Z9
zuH=Z>$aS-N`g7d6ta1pySXONj?5b)zMU)BPQ8HzbSa&^ZJ)3r6k5pKsyWZWPhl*D1
zP8Dxc<u3Z|S!EGXIYJ1a;oSrEa066(8iTjhPPnQRYw_7|BiB7n<A@k;NJ}#@!)(Eh
z=2HNAZ${YO@UK91?)&TKS2g2n4e8L9JsY-Iur4Ueom7|1EI0oJku1eHE6j|N^L-~p
zH3y1g|3wv`fh!6go#JSxTQqZ<Z~Z|Rs@$iwGj&N50$BttPLj_@%V1rNh~q*q&-A5A
zg{2@(Hd3}MMKkx@P-#xYE9`kJoT}uMB6-I@4}7_IonUXBYPA%R<!lS=U7t9b4kbx(
zoZLLdO$)E4H1X)dy{vYfX5Wxw++7nKw6S5iF>@Ho09(G)MxaN?LBnV*)rhe7r*Qck
zI~)9Mz|^jT9CD~X$kjlXU@?#6`KQzAJC6!*x`lMF&A%qUK5N;L4h&xaqnMf|I(oOQ
zr6Q^u;Z`+#q0Uhi7d^OlSQo|~VliXcn0+mj1qz|rE>4NMvgxs+-YPd7U7?YgsfkKD
zMP!e(`V{D@r*|Rqu;(>vn2G^;5SdVFQV2MW;}JhkI$+_|C($|afoX`~WRs_XGLUT+
z9c|INuU}6rJHTdBpIXC5(jLy<MKh1iD#2EmveunHg?p)<W~X0nIi8w<Ubl>ybHP|>
zH6es%rqDEZ^iHjt-7+`k?{R5j(0%OY^byFaLO8p^`2VaKLR0);qNf;ewj?^adqWTJ
z7swiuwGS-2VS_a_&K>!HQ0JFyInj!R;ktdqt2NLTIW=^i>SG6@t0#p|bG>$T&5or9
zfm{WnT~<>%7uJ>YmElzWX5u?2imHzU7a(DL_jP3nj9XbfQYHW75XU^>v&iBMM&PkD
zN2YS@f_WuOpNy0w#ih!Q>(W!K@-@fqK*{DOVOI1Rxt~7uSTkdG#RgDf7@x;navr8#
zZrSBsq~G`ZV?N2rqSv-K4tF!i{e9NTJ$5`(FXx|5R?}#3&!2+%@|^-EYA}&s8?;##
zc!f%%+M;#})6^&0S8ZT$4{0%fD2X$aJMQjjG6DGDdOz@*DP1`bQ7?Ulnou=PVNVS|
zy-u98s|F$Ixy@$SXaB{~)%3qLWEHFPQ-F=QfIW6m9#qJ<-#`<5;N9J#X}1|G<J!4a
zrX4}4BD!FP%BHXAe2czL0hH&)3^UGfQnEw0ESjTx4ufOZv@jJk5ZNg#__C7mL}LX3
z9$ig~gYO#aY2&il6%%mg#v}&gQu2$0f4HrxEP9ZgT8`R`Yq6`1Sp|b;MnH_1KAg`E
z8wuT}2hrKbl;dj>4K0wBwxqCgbm_jXZV0_ab12r%ksZbTl|Py+zJ&1@TrJLDa+kPb
z?(YNhs+KTroH9etMB?B~2eAe)B7ZTe)^#!r7P)<~tHWSc?}&yB-S(S7Xu*^#!SJj*
ze(K9R+ZS}|*PYqx9xY+P+3@gpn@c4QUd&=fS;HKubvbgD*<8lG*XpVQ%p3@>U2P+a
zf|;Qvh^!XizPw6gSA6w5qmF~@J5C*%v9@UBzkS9OO;b*8!tGN_>{!$;`?w0o>bVh2
zh5byr2D*gmwQ?UdT@?+CP*06hUGw%RyG_GmMQ6=)41=g0^wyB7Nu1f?h_Q$o|DAgm
zsoo5eIs1LjpKu~qXUOCitK@S>nT?rwr8_jC73aYuz(P;ijL`zS9VMG^HRtX_@0uNs
zMpebr+MDOX!P=DFd{Z=XdGCfz95a@&sy|jHnohTIFs1L=bkD<biVZvz`mvWWY=E3T
zd3$HigP?~2272Ufp3CRixQl<|%&ed)M8wWP7;vupZw{@);+ko;aknfC=AD^5{30DR
zI*o~i{jt8A%@$$mNa6i!+X}q`tw*sme}C@k=0Xlt5Hpd$MqKNCn?o-)H#g!<*ho5g
zt<1t;mPxCQXkGYGU^l_#SyQXN*gB3uj-jPqm=RpP3ucpxn>OF{`Fb3l7;)zA<<cA<
z&Dbe5muXG`#%go+F2y1oSyIhdYH?S1w=7gZ9sHs|Uq^Umu7V8>Xwdmo(JIec+d>&j
z+3hBkohC6j=Fq`H0T5&21G)L<>8F0%`<CC<`yC6o8}}=<zOKDpDN_xu*nB+CojEl4
zi}Myk%K~&%clBC*be0@cIJ*4b>;L>ybmwrjr)`(wb?uDl->ik+d+>{dWcRInYep3g
zRo~b1^6M}B&ksa%pcRyPzrKcu$lg0;Yq6`R9UKu5`R*EB$;6Vf;FovKrocLz7vG`*
z%34li?-W0aelD*$GkglIc79`{=9qJWhY);oQked}O?tTXmVj%$n^rAz;<}aH>@2u*
z!26xD>q~C|rt_R8W*svibb*M0HE{WElPtoOrDd;uAq5rx<foRXh5(yli);n+DYVYG
zbL@8$IID)4+xsJiPvsWp<N(1{G9BRKJBDn*L)rf4dfvHo@?#@C&g@o<Ps>m_TI7+8
z3wFic?q;F4dqj-Byt9WW#d%fE%<NQh$GB)zR`?vX8eF#b2d~_c`MB=;Xhv8~oBuH<
zL<*R4;Ix9_ZJRUH+z)F26Yg1NBIb5XlVU`SbAgt?W!WHmcp0NVvfF-xJ-V8~nS;z;
z4XB=VO3tl|P7<a!WrnuMO#*H$Q^)E}w&sLhtuX%k<k<&3g>H`(?5%4p<-2Df?*>{#
z?TH#xPWd+_aFB60zjK<p$ti&!u$p3Zf(>7e#(5omypnLM&58A+YVL6JOz$cP(h8O2
z>%yymcFmzP!n+2oE_%NR9PvT01mozfTFX(@-H27XVh=A@AkEot%8qsn45k)4+b8ma
zFafAfykIDU{1C4eHW$^MZH{BkF#_5(S_8}1m$RmlJP#!Cimqh1<(}_Vu&af2p&OTt
zt?XcFW#R7LZ^1)VH6zn;wOLsAmPP#hc#MLfyT!}-znQgSX2HGgdYkv!ujm8)236Dc
zKcgy0w)nO?N7)7wy5!fg=fQLXXlJcRn)f4D&tc5b=4i;Q3Z$xK2CS<}v#M(0d?A8a
zZH~@fUh^HD1CinF6*K?CRA8nK=~EM^9zoo=smo|8H~{A1?PC(@9@R9*POLTY)pNl=
z!Naw<Wn}J>g#+R|e5bKlTmxk*T~RWgKo&@r%&*?#UDSb3(yWUN*N80`0$B>1v5CwY
zYgSC%@N>0eGY&{{rhr0g-S58FRycwEo)p4CRLj7Bi{7*PX4mhj^Q`2%vAk=38Z|NT
zfxcokR>o0NTNiyo!(*(ndSM;|Ae)!7XlAar%&qr`zIw=AhXOmphQkD7*zW0y9#r)R
zmNY?Z+Po<lksZEg!)9S@Wz8cqhR4RU$f>oufL*f1!cj8=?mnusIoA>0a^H{5;haYz
zv(ubPEYwVKstDSm>JZPfF(bVy`85m0$#04_I(If%TPG!poOys5ZG_sHw6G7LBy12|
zJ=OZbiV1U4FL`tsQtoD5!}=cT>O4ky(M(v;xUHTyZjD1t$xBm8*IC&B#;WZ50$qC-
zX@ko0+x7my$;+z>K6;Ny49>n+RTj%pyKT5AVQSG=sH(Gk6xAYqxO2<j9E~imD|aEn
z(V4WCd*JQR>(>O;QQ5ABOIKutRkkg*P#vynv(D-4j{~+DO0%$P>)qy_uX(roSuvfn
zUdfkUxf?t4;EWHtzs!iT)swkNANh>AvNY<GitqYqxC_g2P+F>ycMpWia!VdJo|+sU
zmhCcj^ZSqdXWXCi+nc_(8Rs$Hi6FwO&{e=Fm%R#hIXcaI^AC0*hSLSnz3S_#O=Zi%
zt=cWe_{*q0o*G%q65LEezr{7z3UrS<h{GBkvipeTY<k7v*som%psFz>hz*2KY=aVb
z0{ED%Mh)L`4nHE?ft_WSVVd))2n_S02hfi|4FjgT>E5I5IoDcj#S!pLXh<=Gd!w4e
zv2>RlRpVqk;sX0`eVTs6lp}S4fKeU=Wwu|UIVz#D7`4-2w?%i`H5JZSJ}&Qx>C$X~
zIxWcl`0M$fn%+H|YenwXTA^QwPyNyG+W=#f>ZqnI0?EIvPUlT@Hq=^+Xp4S$FRKE(
z8=9XT?e?0hL98y}UaI|W`!el>h`}30kTy%CnbI{eG<nd_^BZ%^!t`guGtFVcIpf*2
zO<cWY_S6-b-7tfNzptiziTO|>+*HlMchioihJG9y9>{9*>4b)#G?d%k9x>os_o}cO
zdVmuDB}bbR7_W)Ve$*ixfSaS~`;A+D=%s|aR8#-&v6**^etS2%gs-*VbDi>)>C<-v
z{L1%u4&anu{WN;G`6)9$3%bg~wmUNvey$z|l_vA$`S4FOTb{ebsYd`^9?Jo)_5Rb^
zJ(*OiwCE0h6Mi9m?H92km^+&`L=bpH{YtK@Gi*;Q%40^<VO=dTD51po!=H|NY@y#=
zm-Vj=@QNY#e?nI{&H=t1&b?1-v5O9&d7$1#wal;B753IiuEjEBwPZ1Ue@HYXe6U^9
zooEYmPjYF7Z<D<70jtNwuATw6L1L`T#InjF&D3Al^T2G7%<hLgh#4|cCtZ756&8kp
zm&{dPhTlt9^58Yxh|JmS%NDA7U9;gE+)kURh0AAeV(M7ohTR}*6a0!5)A2EM+?BC1
zWy{;vvgg-%N|YpnDHARL1`s<fELJG@7@DCRn_AYmXfX9B!Gi1-98<+H+OQ(cK#?=*
z>baZtz_rY+&lMc*8r&#U?LJpZoUn*X-GiYy*A0;v+Ij2>ua@LjXM|MqW~YT`ngJ!5
zg{yNOj6Mo&^{8#C;Ok<)d)N%5GS?a=Y+yGuZxfY{1qw5|F&iSMF9AaCK1e_g7oyXa
z3bWXp%FfZV3N!P<Q~^=EEi=zr4`Luw?rPd)nMaE^UxTI(VS|6f@}CD5!P+vTTQ>xY
z>$WLdR_<L5uDf2Lc`wvEDo-=;_FWQY)gSnIu)er4L(sLfQ`Mr!#h?0$XA6VAQ`bSp
zxdF?cM+_&C+r2u5&SxC9@1R(a<*qhpYR*$O0b$W^+5MG2yH|7-7z-_C&C|@$bMI+M
z{V{vdT86fZohdHg6%I{qW)8m??;PLFSfP5bST#CuT;Y?md|vm-1zv{g@T$aOmKx@c
zDnoO{5TdK)9`*V+vb);j+n=Z6%6VMdg$JRITO(OD)E4x4-NcP2A-YM1hK1FZjwajK
z2@mpb*O}~0JB}@LQ_GjH%h>rmnUc*LH)2`<Hjx1-%MbHQu;BQLjhNit;58=b=~Qz?
z?Zmn~H!#R>ba_bE>e7UncxG4jAvl5(lnQH7`tL*Az{%duijAiRI@dPUG=j#8wvZDe
zE%~(5U{%JP{q?j*AcD&IxH<P3ul0$n&Ps%C0vqb$Wqr0WI(R9b^ekuZO{(2mr*LA2
zn2|LQl`sOSu$;APHBd4K+`??X6H~dcd(0{=We+vH#d&CnF~88{V{IyOPX4`i`kF@T
z$Q_#&R1ZMLI!PjN=`D;)aJNlUZc=-nq@#f9ea=9ban#;zS9teK(yG-MT+B?2)<{jM
ztRcI0rWi3m6k1?29I8wsPGGrbcE9r+CWOUqGgD_-%$~!}^>RF+qyH5mCc|gIR4L9H
zs%oAY=Iov7X1LX4{66`s5o)+WDO*dbY%J`y<-Lcfc(UD-gQOZ^H10H>;4EAP(ZZ^G
zvcPLqJKscHtj3UMl3$j7y}t%%<l25QJ#gAz=R-V#+aV%UL{?aqp?sk;`L{{MYh{mK
zy{@n4T`r(rv1LJ9*iG{K+P`K|8hZCs?c@>MWfZJSxK@@lRf%!$GERF(-Hu&|FUy^A
zOX?EA2xU&W!Or%<ta`qxAgD~x-1_B0a^?=Ee2TD1Fpe->>b2gz7Wyqt5gO($p>j-v
zQ+E=fuA(`iv>SuvV1%;dy%C126~|UPDia8{IPvEskFcf7FY6KcPBE~FtYN;1-c`ES
z9`ef+cdFdY7Hc2ap?L9W=jVq&Q+ChN4<vTd=&`rQz!zJzL2uaZ;YiuFe%|kUzgt|<
z3GiQ{<NS=~mSvx3j7_N5_~!65f-vmF-0-?1ROC!cFnMZoXr`6Sl!Z_BwB0**-`7A$
zS!v;Jc++e+ygCWk>#Kg6W0pHnv`!C>2IWIbnHd2IdU)qg`ef}6TTwcuZ+01{WIao|
zylzw~Itv58Qy1J6p*?BN6LXzSg4;|$th<Vo1ru`^(7Csp)Unvlb|<>pS6J?OyLI*N
z8>l#IQ{zsL;8;~rg$ZKKd!aR;ZXGRQ1J0V>Q|&p@4dla&(`)n@rCss9v=OyK2=J@@
zGNca6(*whaLtCunwx3e<1j9RSLuD2HTGlB5BkQcj_aFObH=Pma+LLe>S{$Tc_w@f&
z$uLUhKKO!s=*Lb%nsxTnx(k@2vQ1rnjBi8R8g-=8y=&}?!-1%V`}}uq=p1BR%~Y{U
zTr<=i@yh-+E6`@RL&UpB0Y!iYR*yov=;-X22CzA_uFOLl<LcdT8>=Hb&EwO@`6NW<
z=tGZw3+4w_51$~CIaxC0AVJKIwN@gl&BV*qYI)=~>xQj<n=9(fcE-j#kvG3w^@>jZ
zIY_BWhL359I>UMkt35PD4-Ufh*m@=cn>pkEa8rTG!ckG6h3o2F!(5ZC8-jK^4-Q{F
z^9krQO`M6ZOCv$0x?qltm{?J3u%9}Ke=8Y^0py<3@LXTpj=t}6up9xWS-<e!=9j-|
zDlTK!ToehTa{C%AI{*(i$XR=*=JU=Nk8Z)yvQw*Y*UlmJ+KlUG63PhE?l5?0XIfm=
zH4;AVjL;0{#^-&yLWbILsRj4YBd?dHUI7)1tehYmYL@FWp)z5<m{U><66Agu>~PC%
zJ|^H1O0nAf!i$BhhI{nrT+ve=AIf+*&9oEYRbvD<GAyeBi@`jG0d1?g(rv#YCp6tM
z!**~?_vaR22EkEX?sNTbB%tQ$>KR5pg>E1Pt2wlr4YOrv19NcH;f@VnvK7(*`|PTQ
zlB_!5yrHsBncm|hRl1C5yTdB6E))3d@Ht-1BT?G#9B<4&01}$#k+M1pm;RB#P}_-4
z#zL7BLvI9Z=4{8PimK<jXPsjCioROrj$P`qOJhezKz*(+)o#VWF7?ETP^E@d82>P{
z)}*?-3vf#<&#@?)b$NLIdQT{1*m}ERQ&CdH@f&^r`Rjk=gftY>tZJ=@erg5w>ydSN
zwk$B6IPa>GMfbWo^Hkx@ROu0xnebq4yCC#1^7bxzIMk(T9J-Y-%iY{H1`>41Rn73;
zB^u*WT9|%b51>cYEWyl|ca`6IO9KHdDa)dBEdj=Uzy8AeJVG;+aPqo;{p#s$UN>u}
zc-uP~6!X)BW7$;?E{tuBC8EJKYewxXSzOECO*bN6i~Z0*=eTzWx&v;Pp0_OAh}`F1
z);*%8+mkomU2X*EdJ-1IW&Csc100+)Kx233p8v3B8jI$#-g#(Nz*tqx+$Pk}vcn+-
z4V4z!(`7epELXlXAT0B6U0X8Z)u(4XW8sqdOc0EnshExWY-qN^?R@Kp?B4So=z>vP
z)-^*OI<u*gnO7f(<3Lf|O1CAfUWAG#N~!SovE?SAvN;3jS>yV^td7OAUd8bzTNiCP
zp}ozdYPT-MLTR#^bTq6rfx!_HhOrb;DM5&9l$&QH)U~s088Yn}ez`C@hs>Oy^4XKm
z?k^KG8%hk^hl)7&Q}M4r_oxXXCO*Djut5uqKd^_>%ubvS%wiQ`)3Tr&QtIM71SX7Q
zHp-|y>6w}MaQ+H%)6AYdfl7m_Vs#U(qW64`d7QKB&bN6jezM!p-skqOf_2YJe~%b`
z0Pl>&oK_j_AQh_3vaXvtzl=F$!!<Q@?OK=X*@j}=U;8IK-)WW14d*Pb6X%3K+^w<I
zI}bs0<gAdwTQ>LXNlk*w1_qDRWlx78qJCQNWxK*jbafkAZVl^NSH41vakEBqvb-k~
z;3vDB`dHX>U7ZQ%j5LdX$8b^Lw|C|+T+DrV-R}#ZjsQDpRrfuRsQD0YolhLA=#mir
zmPJs$x^|heIXKm$j|Sb<*|B=WxprG-fo^#!`8B8I-_A236C;Dnw)I~Tgy%0_wQ1qm
zHsqA~o`9ybdzd`LSye<lkNGUdC+6+cZl}mji>y<m)LQ=jm}iFx(Ph^JB~NY_Fy9-T
zTs=uYp@pLNmb-;#4qSb*<HOGx8&kd~7eK95^+8O!c8Rup60ep&c3I)0J8waTohxd}
zD4L)(ctAB#8)t{|DTMzUjkS&qb!3!0!U|J;I@eHV%H6sq?FV;bSDtGm)gFU!$0rSN
z2&!Y!QPZ8-0%;=@iWr?|(K}Nf{#mg~h44O^T|mUnlRBk$-`Be@+gNP#eB7#%x9`-r
z6@D9G=Dz-RM9chmy60Y~>FUfwyST>5R9K7k%9i&=7cSSV?o{Soz;B`oAcmc1<z&W<
zG^;I(WvX`%h;3@FE$2~2RUNlDj2sGR!_LrY%0t7v_iHLN;Zt*xapwvRtS8Ga2$P0`
z-*;E3dD~A8S)%w|wyt0wtxNa1AobJ1QEt|(V*r?c-T(O?jNZTlH7}hpr412MyYJPR
zwuTZvZpq!&y{_7cFjdad;6wK9b!)e><8nP^oCJ$w(!;U#3lkl0aWv_3No@H2<#>Ox
z%)ro7@JU*jn=F*oW=Az~I%Y8kw@}I5&sG(_LK_T!hM6RR_x1B*R%K{R#j(vu@3-eT
z63;8YtQwV|x$`aO=|K+=dwd4$G9I|_wDsHyK8fFm`BJI2eJ|fXlbP*GU&?Vw_?%!V
z$-~eH>)f!$ifU)q*$dEMi#f<o6~E6I&DaxM46mXiuAYD~-ul2&G`OPv&?6<u9mqkm
zqqD+|uI_>xg_s!JP`NVAOglTi(Bbd)DXfvU(C;e}^2M@1J?YMm^nQ>N!@lS8_c=2O
z_*%WIZ8_f&QrQq`35)5^w1RIftJL+?dD5t4T*$qJX?YAc9*&vrz3@2f=+2~1ssd+~
zC&5CslkW2!TUV!Koe0!C02yQXV;5a;w_(+sa6#a6ikys8V8gW;Itmr|(tB#yb4XGP
z5t>e~KDn)gg>%foqE9v6%vhWM>)!7^fhP!nTb3E!hg>+n@Q~TMnm%M&(41X5(=r3%
z+P{Q|u?i*^!5OL3x?nR}%_d4*H<h~VS6((-G3zN@$Lbt1oe?Ub`}&?<Tv#>Tm(M^d
z8(p>tL*nwc&U}OeY@bz+8PU1t`)AYi$L>(Ymv+I1NzesTiq{%ireIm&)!$d{DFT?;
zE)u@jpy7Sx%XZ5Gh)`!wys+t%IZ353<1?RoL&-xu<43B6b>|Oks`Mao%WaMpZkVcl
z&o|3RpejN_XM#SY>zTk$182(r2Wo9B{%_~#3CM0++O4%JtBo=@G%kPNt+l*2SS9;f
zxtlTLl&ZEn1gjLoT8`c5nG}u3MAU{=#Tp3=x-1|!bnyz+yto+btck>x{p)-0A-Wdr
zBytXwrtM?eny-^F_x$lEtK!C%)$VzIS0kospAS7O%WPh`CS0-W_C3!v3HE)d8<tPu
zTD#ne3%kbDLqiSg;279+W+f#Ls^Lo6ic0rcCYIO@ck8Z{-~ApLNDYna>Q}s0;K>;{
z)@Xd`^MPWu*<sX<k*HO;@Y;>)Ca<XOZo0}N@auK&YPs6TuVL4pWU`xfj!^#0HwmF8
zEQgtD%vgOC2ZGyMM$q>UFK5+UJQZt{zc5!KbeuU;V^?oqVO_9=T{ZulFl8N~-STb$
zwph8%KfKb3Ir=Gzo*(k$x$R_S{T#)bQRzNm(Ua5_BeH27SHB7K<8{MF5eN^!R9Rp+
zhMH?2GG!}T?40B8TBgM9{k**R0eAr2W@;jbQV00;*;pml?A|A-Yl9jh6MwAt1x!6;
z%~IHq0;WA~l7@EEMIKMtbd6hvlu2>1_7*T&6p+m?i9?E-@nP)8{G+y4tb7~SvU-Z^
zQcb4D!@A7#5vS}tt8*Hw+JYsnE4)ip@Ot3tb#(`K2j1B(o>#F_p!<}|BWA*7d^5a1
zaI>oBy+)6Q*6mNO*t+N1vonHS*&fImHU$@J*fTZ?ZMJ|iYZ+zBCfsfB2C0tQ4#Gx5
z4xv*i_i+w=Eqk7s6ATP=?gNtF-nsn_K2LWx+);4jJ~K$bXluROV+IwDo)Fmbgs~wi
z`|@3cL26DvWbUdyfC(;anTm6X(4#9be>1znTIjt0(EF6ThMFr2^3R0DxiDdvO_N?R
z9?S2W?$%o#B+1$%?-&Wz1bu)iTj7I(GE!>A++79}A#oy>a|aV>dGZaQB=>CN$)0w*
z0fuh$hF{S&M^)>Fg;PHjmh$EKltd5XBvL_^0NBNyz3%-)be^ogJ5>b>L#ODmpP?Fw
z%cpzd^Jq#BRKsJ4c;-GMzr6u4=AJt5IU3awm;dKiL*~<wpRtjlTh-%E5HD91AYAco
zLA15@Bi`$M%a(nHqLVN~sAi8qjFj2j_YSLUxHDgK6=~+vDmrp*9y%GtKj2P)qE91y
zfW?g4vIUrRjaPPDw2je%oMw#~OAQ0vgM0Jieyy~i!iPV9irX!BSf~a`KQZNm^$~s#
ztV1*XuR7$Bp@QRtmPtT<SU)x56q=1icGkNkW1X$p;e$5pT-Y_+bXA#+VQ^wa>>9E0
z9Ed0l+PWKElQnewY&QVw>TFF@8MNAdFMswNE$1>ckw%90i8N*o`&1+^M^z8krw=2>
zE}t#g=_w_~)19OdX7$<~eBK!R%j6>~N0CU6ff#@K!jiiAmo2XY?`%dmA?)(C`Wf;V
z4VC^^Jii``tnj(474Uh2WZU|}?txFks5x#yyRSg*>EH-qW(&E~7G_|AY1$+F-SFFY
z30z(|N=%RFC)*tiQ*Ke+vbdrOU5&3l^=4Kd3t3~W-c4iia#gqHN!*C*e*Pnyz<aP-
zbT{W5y(jz>Vb43OW~;j5c_AQ1E%$8L7$o2^YP{mIY5|`RCRNg8GS7|6c+`+bmD-}Z
zu^b@Ln{LYfimbCStP}@WH~<M|;ebNC$|vzE@0w#;*!!xvu6i5pEmJczoRhze)+HUd
zzoA-mBMk*)#yXi>_ewA;u2I!wkK5ZvH0AY{%)4VQ7qk|+JG=({&_l(?x!0|$+mJPL
z*f71hVrcO#waia_*Gy&XcI+AQRJ-?l_jtQi0*0LaGX}G0>u7B>1z{t**Hv%N+>x`^
zYoc-co3J=a5rSS%2iq+7sXREl*XmsuY#daw>#CHI037e*fqBjh5N(_ue!|L-2>aj&
zj@8Og>6f<5%#HONuU(V0FjV&39!n?ZsD}Hbn$PHO#rTc8N{BV$IYWz`0ROf;)!`H&
zi|*_>>e#sI!z$5bD{$m!G3~rFQ=HQ>rD}IG_A?iU9s#*DAfx#!aOB)0V`X142@s7{
zgQ#}*dJ0#J6b@CsRCY4l=B}39p7+F?sRnpOXTj0FIN1JS37V;G%2bnZGVfADtz%Fu
z-IoLp$m9LlO<(wFr$$eA8g*)M#hVtvI$YyXvW<VR@`^WMMtJ1*Fl8Bszr&HFOele~
zg~(psRTk)M_vX#L&pyey<;(lsY0mjGT5#8_FUFiEo2`DgIfzPm#M%9$=u{{FMz^~;
z^`zpeXI=AFHBH}@)oU<Mb9{O??F)~>j0D~^O6B*^0o}7eQ(4R;_SBq*C8#6caZQ2R
zh=!eKJpG6zV6Ju!y%%;oxHMBSIxU34m;cmac>weFR2O3Yz|-Lg<)GHH3B(t3Z*s+(
z!L(c1@Ywq>g4{A=UH+eE#;fIvM^BMh#{CLHu(?d#>-zbbJe>9BI%`jU%*Drex9}pS
zF@{5l!S(h0+$#%oGKbr9RA>-rnPI7g4V$@`8?oW;onkc^5qReaRdsn*!3}h<+fWS)
zy<jZ<DNMq}9>qAyR?OL1k~%!gAco$Vjux(v&cmVV^6be{W|hYrx)UB@wl^fZ@L;JY
z{+%*N@KyPd##J@l1|kWfXzo?B-<T)y?*gYole{*hA{G*s+xTCKE>&ie{WLXW9_`f{
zVyq-ApX)T;_x3n|6h^1;$Ng@YqxUuks^gsV)om1R|Cyb*QOcSRVhymdb((8Fp={o^
zt1M#WE7@rSbSswB-MsO@R0mFp`h8Uf#@w!6S5<e|`4PBfxuc%4xKVvvM-$#1osb%f
d+3Je*{{skqWVwa56r2D6002ovPDHLkV1ir?DMkPQ

literal 0
HcmV?d00001

diff --git a/tests/testing_data/ultrasound_confidence_map/femur_result.npy b/tests/testing_data/ultrasound_confidence_map/femur_result.npy
new file mode 100644
index 0000000000000000000000000000000000000000..a3f322b11337ee1aa3d98976f93ddba3a2a1fc5f
GIT binary patch
literal 974976
zcmeF&cl^&){Qv(nP|@B-OIt%pX;4awq-C}g8l*%?iXu&w%nBtbtE`4iB4qE3(6CBr
zX$bYb-oKys`*QjI{kvR#|9xC8x63)tGal!09_Ml9>#cjOo3E{1zsPQn?AGq2#!Xt?
z_0UNd9COk|O)H*M{+N@RKJZZMhZ?oG{eg!XH%a;xjheS=BE8i;jUH^``{Z*goL&By
zisw~2rTj5Z9P|I_-#)wT_GH!n<$rDPzdrE4KJdRj@V`FrzdrE)uMe#FuS3;USc!Gm
zgrD#We#5qcaI0`LzC+5T&OA)VG|b0WNZvw>$1uEu*9*drgkvxo?_nrXZUz=3^}||x
zi}m;#YY_d?#smz+Af)aW1<VyLK+1$ASc)%^y7`W-A7D5>#UiBL<fUBtl(u7&srV2n
z7kSGINZHw#gp^7BlwXb1Pn*$c4wj)XFa3z^QZDi)VhFmU8`7ufH3E~7dK<6>TMI}$
zHYsc;I>k1zRqUR&V}t0Jc>4DlMj~U8_TEKoknh+(X=(Qp#J1^Ehz?`%Ip!3QaSj=;
zq{jvyA@WBR5ZSTYI7FA|_5*%I^pAhV4oNT6J7XUEr#}hfKOy>u*el-&GtM&+UDH<D
zkG|0{?WJ7o6ML^k#wWT)pOEp4A0#c~l<|m8(Jy)?UTDkMFuqeLBV*GOoeOwF7#ohp
z7{q4LEqWBnO`jwCd!*m#Yw|LV$@>Y(%iIbZka9`iR=}^qLK(>~B;}%O?3lFFFSK>)
zr%uu~BK=BOSU2@%qj3B(*0Iere1_OG<C}06Vyk(`c*G`!^E7hOm&i{)(nf5bu}q!#
zRe$uxTNsbnDD_jmuwSVcSsA;8=~LvSTqtbgn}YA?kT&A$8K01N=1uGp+eUusrO!!Q
zi^z(M#G~Kmn1SdNA4>lsBV&{@VNC(iDe^+(MTgXl%?j(sM`NeCNM7uhFuJAf$W0rC
zZKwVC?EkA*Y>;w=#13gM?MC-}CrrQ6$Bc1wi#};5VajAow&G{}g5MCI`V)WPcf_}n
zmoTJ!Y!{i4A34z{#1@5Z7uq4;nX|Ef*1L>J+RqrKd|{ts)8vQ3zDB>ak-o*osat5j
zj78$97h==cG4>jZ_Yi*?gz<<yQ$BMxa?(y=Jo==~LfxZR@-s#$lkfB=x)kb>{K7sb
zuh53EP4wP~q{r@sW081dg`}lS=3{J{`e77C6fi-!0EN1x{lfVgUDNN_HF|!A)JvaV
z$6yrB&w=8tQ4<#zaEq`#I^rQT!UK2#{qR0ge=_2~(IahTO_+iYk+z=3t+)#vka^q_
zeefpUDBx{j$as8!=#;kZMs!Qrk1!7Dclwj@44)wF+=do-9+8{&`y=f=g=g^|B0shm
zS@7Lm7`=vIIFkM`(pJiM#bdY~XW($0h%-?U4bTe{5j(_x;?oN;1+jH(7FqGV_~2AT
zk9QGUrrm^Tt0mI!#;Auoa09MEb!42<W^9`NHbYZ1!HY<r-$D8mGT!N9Cp>`kqb_Qp
zHtxlvi2b8i^j?DWWd_o(*f?VvoeO2huIYE=#!sF=dvrm@H)HS`df;)~h5OI}FQXfh
zHx%O$+YCeMW$e?J=>8U>+W@4@T*Sulx9AvOC>+!18b67>@5edV2Ssrf>YzEIPjrZ1
zrS7MQyrl)CjY$}R<Yye`A!U-6c-o$gtR1Vd5$o|4V!!!F-W;TS%0^CXJPXNRg!syu
z0#*v6L&hL=XCnSG4zbsR1;ozJ;Tb%M^rsElp))cjv0>_DE@kW{U@{`NQ2)eZv-oK2
zeGjg{r3J+I(#9A>uk<terhdjO;}ZEJ3&{8-o-s)uQ!lbJhCPrry5LpBPYcI0{f=!@
zCw-od_}p-$Oycpk-iUw1FA~NlJ}4k<ht$dZorc&hVHl0Bh@9yBF5)NA>v=qe7I+v*
z8&W{(JcZc2Bl_Ya#Fhz%7Vv(-cgA-H3db&Cd?v)+(J8(aed1rSW6H#5*5X@yhfUam
zt@yFvd%ZAu8?X^yV+9r>vg2Rrd+ZfGk`|kWj7jv(Si}agU&b%?N?+n1g>|FXJ4k-^
zoYYMlkrQ7`os3_`@g;OXJH%JI6oj3H$?t>*(HybybBJHY_U|KY#!sSiZ1E`yZ5BT(
zY%}8={ga=$w+QKDd^<9tf4);EcKHU&F&BmNG;%XGa}Xa(olw|j>c$Ua(=~Vs&*Eha
zK=e!9Wk`P7DV*=oGjh_8l#Na9!gXkfClNVKaXO0P=YP#BBrNCq6copozQ4g9;-zss
zBBveV`|-K>dUv!$Z5)hQ>U@Sn#8dtrG{fUa-Hb!Zw?W1ud99Fgm*H@ni{wSuUg(4L
zCv#>HdY~T4;0RPe6;#JHsEVVI{*}iixEc4NF>b`|Xp4u@0#76T>WEvg8y?k<^HCG8
zU<|%M<lc$s6~CE+`0q%(g6nY{qE{WHjo3am=!EE>w&H_@V-Y(qBYPxP;x9ypKZHeu
zrEmr=$5p6?ixIoUR%z!_)I<Z^f!L@r_QZN}KF6ErkA+D6^!q(zE%*=_*Ef;Ah2BV+
zM#y-^#_@%da5SP<#w+%X4k7WhTN<%JUBvdWb4#S(Y4aU)MH|#d>ct-E&waQX@qyTQ
z7N%k%1|vE>g2$1usDYF5lYYhjV#nq95Al&>a3<2{*r+Yu#V43l5cU;5hUl6(HxRMQ
zD2y#2_UVV{u>{MIIXex7a>k1%e;nRM&w}s+;c$G6PYX!<?;~>4UUW(r8(ocm>AxNE
z=lJ!txCU3@5?qGNy_;|sZbinb9U`MYqWfo<j`(SGconhn`G~L0Kzt)>!G+S^K+4a@
zY^2@*WUL-Q=J7oR-`$0gnQ?jx&*M=vM#eNWL~Hax`X7BFGj(F8*dhMV6Dbp!O%Pvb
zg{%`9=d}9_Uc?LkPuNL3b1!2P`wc^Hv_Qr)z7!i(#%VYf2V)<^kL%${M0U#8LF{@x
z9zrifZo<y!hz>}d*fKh2tkZ`Ph%Zbihz}LUcIjK;n2Z*W4Hn@`ti)IN7MrjEUt={^
z6p(TYFaweQA>KoDitMzJxe*^qdcI@F$MFcF@6(8HXAO$|(wDC2g7iIUZz4V!{WCt1
z7rooyMqGm9u{VmNG%BG!GCsE;{&@ys>x#G<_u(b<McRlSZP5-Hzvz&5qC<S;PQ=He
zfBZ3eq^}wK^eg>~-^Gt&o7ia<#vyu-M|>^gmiE%m^f&#?n8jDq&$0L%^Kl2_EAjE?
zF%+L-4wfNpEW{KHMr5WBnR7L9F-}2cB&|8hW3_peIW`Kd@gycAYgG?aK_xUm8+5{#
zI6>Xpka3HDXB_L`JZzR9Sq+i(ysWgUI3LyjpLlKYy0{J*ml{a@;>dpe9g2w`fuoVJ
zO!+2=9d5+|n66xOUW;wWy7UjCXJ3rKP8_Ig<eY_5Q3}ziUIEePUObAXD2Ldi_b&d0
zH&N2}wYU$lTYRoFGM2*-|BgKxq6o$-6I({Fw3%_Z05_l&G7gtwiTudjp^drt5%qjO
zjot-(BFui7z8s`Z^gkTm7o^8tY3pUA-_dUsVxKqA0`2e!t|%b!*!Uh~uBIR9Lm#}1
z_;ma}bK*f<j?ACqum@u2O^6@G?#D_ijYClag*xq!7vIY`*F=4^LF7M$jAi^WGQPr0
zyovbHK%6ZvI!(Y3WDZ2<(-1osMf&<F2H{zxuWj)jmLj?&oQwEIbdS%b--D1j`!1ft
zUC0<cj9y6oJIK5XuOl|e-Z=|%F%@GF8^&hOAimHQjS)XjKbs>y))C2%4YHQB#A9fO
z_*B-O^f70J9(WPcu$#1lQ5?G?HY+B~n8b#cBDTH=>EC^bt^49*jKwItg(q<{4n_Q{
zA&$fS_!SxR*d*mc`W~I{#4SjF(yzqZ<88c;l!@)q?^-wq@tgSm^+>;BqrnBl=Ta~7
zW3R`N{K(80A638s!u^r8J33@NZiyBJ+$W5EQ~zm1Ml+m@)ZL~}KOu8w9Tp>VW;_-l
zYgKL3MRYEMBXAgE*YqiSMdYUs@soy#4bs<+i2g652l^m335j<@`jok!FlFOgQ!o>=
zFb9jV2CK0QbCLXn^D!Bz*Ar>8DRKt884n`k`V=xxVuz$PC?IRx{fK{Gj*Lka)Wa<W
zJS0rLe#rPId=}~BS=a+vhcfpwr&eGqN=T3IMQ-d@64_fGLe}Abk@jM{*dXB*Nc&ge
zYScvfaxuyvGEYbJ%bX~*Va7H6ik_LHBe1A|*mF3BAo^!4(&yVy3(+rO#xG;p6$9}R
zKE*J4X5EQz6!BdRnM1MpbbNu7%Xj8^V?_Vxc_5<mp4boROZt@c=`h@2Z7++acoF?D
z5To%rR-lw}SEejJz;5cFkF1-`k$#+q*s~8#$L`n{XQD2$W@OF#5gDhfchT)w9FJ3R
z2r_=z&mO=V$T~0<(f0uPm5}*xIkIlviktBQMk9N5WF($(PF{D+M%Iwx>J`Hd@q>iX
z`$C+J*etSRqmA;T`?>l*W+z|ACE{H$2<OWC3k{Xc`j|0$8S$goW`Vp8cpejxzHh)X
zWX`9rSv%H<mqTqb_rr<kgp$&B2<Kn}4)J|CB6Big=l#)3Udml9emzb`;*X;X?#G2V
z7^k2D_P}a&((ZcHrbp)JD@Z@$Lr>sQG(qOW!HE4b&tk*n!s*D^e2nz#7tBQVxL0r~
z{?*RI!aori>37zz%%{wu%aA=~2%4cGYN0e%tGl;w2FfcJABi7iFHRey@D<XB=+O(e
zpcD?p`Dl!87>v{zgy_2x=~w34W5^nF59%Omz!}K=&v?Fy?#SBK9*^Jw#8zYQ88TOU
zBlgNVQ55m*bCLeW*UI7)RL1p4dszoEU+W|5T@74>vk*JCqG#q(CB%=v#vjPL(_-X|
zu?QKP>>;s7{PHMu<A+xv{df==-)_jdm3ez3j>8h=<0obDHJiupW2fj4yR^fd$a<c+
zk+I9TWe#TUR4RzaE@|Ul6qZe1)}4%T=3VqkTK2|Xh;8F<v2E(#j@uBsH^O%H2BH(P
zF2p}NBJ=rR`LShHWXx(IcFY*Wrm=0tc#g91nMcqTPoXoOMPw!IapWu(9d}23<W<Cv
z<5!yzTRx1)jvYd5a2ldV+K#<4cfwQ1+VCvWzxY{Gq<`su2gC*=@CiP|Fbu+w0)`6T
z$0&@$WW0km$oxsWvF%FiB_2COzq^nzy8&lo55zAr|KfuuA@k{YoQU{b`jW8?m*Pr9
z-kHd_WgO#YYYX^HI0~O&8s;H>8C~M1S)b!qC!jUrXOR<`(IL8&MY#e_7hZt)RPuK!
zpZT8tWgOF|?6Y^EHfkcajBoeA49r37H~>kDoic8bSq<ml<bp8ccQ4wbE8arveV6tc
zA-<72V{8<gW~|0yDl+f;AU=B|B6n|0Rz7}~{U~$uaB1mteC>r@=DmGn)ehq#oP;9M
zj>m=AAK8aj<2dC{#PisQ$lf7bj`;HJD35(GPx%J;o4mV(n{c)C?12-|1KA7X^W*UY
z_Aii^@hgrA>Xa2eh?LE}?Q_I$Vz-RZ`@VlyZWWFfkDYpA3|_>AD1r0P2ib=g;0wej
zrs5rB?q;uuzm~wi;{CCb+^PC-3d*4)#$qz=lb1PqnDQC(cF3NUHKG(!|2i~9>@yAN
zTL+~7Yw<TmAoe*LYsg)XUvQQ0I>;Vh7f)dXvY+<FqxjZ$V|<BnzGH*$#P<?jj)u4t
zd*CtYXQ3EQ#{>8Zu|XyJX8nB{&!8=`9;7ds8~fs2<+dt+4Zfk@P+{aZ!ELw^wUPaC
zzPy>Z8ncwGCR~XXn22i1e2){zjjfi6r$41p6dC`Ne-_<Q*>`N6v(;&+go_Hk<5Ppt
z8_{90^08k@tmD@iubh$oM(Sid+Tl68gh|MlUx@0+{#XV%Q)W%dn)WC%&oi$w&$Bii
zh$a}0@fe8MFYDV*G8bbtzC-*g>(4gCf8#S*18bmA<~0T18H?!ikFxRAjAiEF7{rfX
zL+0@mWKT>#`XFs&E_FjcWbI2^$KgcOK?~f2%aFA$cF#Ic27}0Ng<8lP%PrE1DwF;<
zMonbR%i4P|$|813U(>Hc#j{sFg_{r?l}8Z_ke@l+71{fWNsC`Sh9?mFX3b1nvH7v6
zkJ#r!<*z_#tfyD*0V`0^cWY!#ya!nyzCr3=imP!8()SmnUx_1eILhKwoL|7X!sN$~
zw%}XDA7k(M+6~BFd<I#W>lx$h4W}XYJpl3L*fr;j_(<&C6tQ{w7Td&r$0KE9&-?K_
z-o-EsLe~7g1z{gyFZ3=*OL|Xq#$8B1k3sw_y01X=i7$n;aXw05p}d(WyoY9NfA{@W
z0lS3piHvDEWRHj~vyWt~3)lOV%Ek9SN5(Vf+3cwqi(9cT9ydNK&AWcsT|8@8-U(OZ
z2VAC1J7n!!k3SK;HsKfig^a^QbVt_B%%k|yz6JUmQ4o)gSE4n>Vlv)F7j!_zAag4H
zimmoTF=XAj7#XL#k$pLJV&nKmxCv=H{vLmP2ASu>(H~DC<6Ryp^P{q_BYWLdXoby)
zZJQx$=fP~!!MZdL^~7f&>;8O<L)Pcqdw<3bG*E9EentFpG-C7YdntD>>fujyKSm|-
z&vBeGz3?L9+lQetI%7UE_Su78#Rtk?fb0i1`>u~R7>3NLjQ`KVBe6*v({P`%*=r&<
z^Jla8u}J?OM#lSI>BBJ;v14p=BkCcxiN9QcJ8(Was`H?_6WC@P&JwSTtmUh5p)y$)
zGryW6d(?UOOxi)1qJA0S6Nv6Tkv5uR65>OhkbR{wc_-pVJcZc34C1dPPyw;+8q`v!
zHin3AQTA@(bj-wC$hvft{3EbT{8f}g#<ee6;Z_uvcBe9X(D{-AGS~J&al|L{y<M5i
zt;x#&px$L@hqS#Jk-ZWpD3kg0KJxyavs-*>CT<dMftEN%xkbo2m3jY(czpYBc{w|T
zH*hL)-~SDD<j0RQ=CNsHR7MRH!%xbL#n~vYy`BH4x7~h^J;h(Zg~sqOc@q&`Vxu?k
zFLLJ1dU`Z=`R;*hknzcykU5q0GHYtq%Z%j{$QpK-@65Z#xEvd#Wlv0Ad?B`bA1kp9
z>2ub_%=fGnRdE5bm!<8jojI3Xi}-Eqa6guyytKRU44%e4cpQ=c8XiIX>rrIA+KA|L
z64})YNc<0JBT*mA$vp@+v&(mEw43<J(laO1->Xm>2O;Z9&Pv^o_G7<k$Xb}b+=2Av
zSZtU60Wwc^$DYVI#wPKtl+B#U+_@96-A;Lt_b$?xYB&;q(rGN(BYST4&FJ0?*CXes
z{`d-ek`sNeKzuAVDu=(+i=163>U)12hy$<}qT4HY9%rElQZDnTPeQVC1~?V5>t-y(
zGiZqyF$vi_vaV-Ojt+GZ--(}1#u)sGi=^euku`oG-bD7-F33Lo9A3gJ=#IPvBrkhI
zEu4bP-N=g_+9P&*2R|Wax^nnSd>G;<@#|TLPh{*eh7aKi^v2)F-6uYpbs(gE#{6r<
zudCxFHoO2gAnRQ0nsuicvNo(Y-|jZ|=KpOhgac7k+H7PmE>1>Ue2mD9eu*!|8cfIC
z*ht<Yyn$WnwZL;|<2&<UF-p=g`{oFIh}fttS|MknLy>uvz4{+ve1A9b%=L2-AJ6=K
z1hIGevlEq+>x}g{UV7$FGdzm>k+D4z@$)T+?&;IfD2uEg@&Am;c=GDd`x5hO9UI&(
z{10uUy^fw}hPQBxvVD+yOV+lR@Fj9LdIFI@Sl$)L`)l^O?6tWwZ$K~Kf8%E5uEM|K
z(~&hYYv_yUhelW+Eq9`j{iLe2^N>BJ3uYtzkMA^)|2v%zQg%=A0m$Ay3(=*u^cC{2
z6t?x9dHFiJB6iFkdj_I^Y5b;4^!fv<Z~!)`e+?SqZG4D9=uYM#>ckGGvw1_@ip<4P
z(zd9ZHT`+4#5vL`;8*cugz@{4I1`nmok8Aga);t;4ELS8?k5<Ag(#|A1=K<A*gq<F
zgfR8rME332tqVRvWW0cN$el0ej?-}h_C;OHQKt<ZCLn9UAQYZQR*CPWKO2M@o2!+*
zIQ6t!6}yo0K_7gLyty~TKB$1D^2gv+@w3&<oB8#K4Kly}Le`yEQ4$|ZAB1su6g$<+
z86m#4r~K{0#n^$Yg*nqi*4xM)A3xin{72f{k5AuZj`zS~)bzaxgXLYN%s$Fy{e6Lq
zteyX25$?d3(m%wHxV0etD0awxnE9DCJZtl@xD1yf{&OzkQ<-1kZM=yV=#44JoXpx)
zjEv0p?DNMUYeLq#=a4y&ccF~yuXJ6FbHtxR)}Qpd6i!BbsS2`YM)&Ht1UH};(&kuX
z{jY@Vh1rWXON)Nd{{u2Jc6*^a`{aByQCj9@){)dLj;!t5eUC=w?S-g~j+l$|<yvID
z&UttiI^Y0g+%h+#d-k-feOIFavL+^89*gBgUOi-P9Egll>UTo}WKR8t%jMO?Dad}2
zdaq$P1|z<;31?%f^6BSih^(ar;VNPFiPbnyxwdGAqR2R;>>6Yr*&iq2Q>4wTXE!4H
z6%oev&*5XtN5(ntAUT8Q{IS+|)}Q#oO61&@^*eiMH?&6!+=8qpjgWOcY1s=NL-yM2
zEm`M(Bx4@h<5XOP2B?ZRknze`d@Y_i)(Gv8J>WxR?6U^HfEpN&CIxvjgsGSDpN8n1
zxsv{7KK{-}-o!j)UO$7|u-5kuK6JCWmc6zrdg2&!a4QB$e_xpu%HBj)=110?tcTHY
z7`{NpZ*OE>5BuV6^$tMR$Luegq-D&P;VS)n2ZPWNv1`tA@u~P`+Q~dfn6@%T8K2DM
z?4_9(DgOlCAvgX}T>hPyiruAW{@;d-Nn_-^lzE)7`~>$Q_nTID1JS=Tu0l)ShtPSk
z{$Gp^xKTf{cQwWNNPa~;i@lY94_Vvdi_s(dR#WVQF4D{45G+wX`^R+h1_~dMmbz=P
z4w*Mc<3uz-UF5AZdvi-<-e<pWENy?XkH+C>h_o5M$)1|uRbSLb-oqD&KP>H4yoMu@
zeKvdK(ZXANS5_}`bfI|m=0%u?$eM!Gi(TGD=5&Ac%BcS{KY5CMt`;wfKg6dAJ1EnZ
zy!ds_l9@j%e8=D8f8|jhV{nwTkTqu)vIeIwNjt#z67o7D`}#@bzs5&%CY&PvxU{l#
zc@4Smo{Pj|-`A1-F8-Lk>1W^TkUMB~{3U**@DpKAjKr&GfIX#;QZHlp9^OTKz9-(o
zOl-rKxCUjFjUMIHol38q)h42vw6Pe3uY5m)gT?y@+o8UA_Q5jhABxLR1sU_>P!#{j
zFD`6|o~Va8(njJ%@vN`kU_3J4??Zib$76U2@s|<Gr*EIJT~XA+TR4`!AK)%|@rR{&
zMfo4F9__JR+DDj<qkK<5{ACih<KKetb748eH_yXqh#%dGj6(^GLv88pk^S#M^uhc1
z2($1lvZhVPJLrNV(L~+P>5#M0p~xBMU7Rg_623z0m-W9l{zLZvN{CLoe3uaBtPmf2
z4u9h${+Y93?ko#@XRRuWr<5N+e*9xMHrk9@(xUrnM8~Y(Sv#_q#XpX~p-7vtQT#b)
z;1?15tVP!K8!(ZqcF3Ii1(`FkXXZ@q_LY%yO!knhThaRrWL^1H`HXkgh@a^AGBQTT
z`@Rmb!{hiGCnz63jX!6tn2*`Wp50e|&ZAw?1H&*9S<fy(?)7;S`49`y&375?<lS$P
z@9Xg|4wM#OPTjP*5D#Gl_R~&XJc+F7>HjJmq+dC&%n|Q|4#=L8^z=XH^d_i{6Op{*
z5#Ns7?{J9k+pz>Yu>sj@C*U4rZ7+%V<H>jmKjIHe!Cz!NO7B&OEy_!aZ68JE<X(6f
zIb$9s?`Z6S>^sX*L_A}CB0BlbS-v4M9tZkvh}dzw?^$e?`S#}DeDyE;B07q<G!NfL
ze0qqo705akUx|nK?#*}%mm=#xCuIj?8S;DH-(*Mr>*Bfp#qT~te5gMLpgFF`H8>dk
z<;N%DxAFPNk3YwLu}}8QGPnd8iv!5bnzIENhm2|LxfPeo&v~a7GMD1VGZ1;Pedg%z
z^qZ#tnM*Ci<1^Vas_OgOY;h{?#$#+!Qrbs2T7Cl@t$YW>_iiMs1ER;ZWZ#MSLHxMB
z{KJ%=MaGp_PhLH8#;Kcip&~Nh(vPZGtbFFr$>K+1lX&jd@%z`2d(%16x8qyfkBrf3
zY{mn~c%P|k*7e*08%WE%%)Xy{-r*=A?^u)<zXw@wN6<6-PS*15Ul-ywWWBC~ocHp5
zoBYhRUHa1xD{z!PEkn*GcPrCK|Ks~***<S%ncv5tioEnO_P+;h&>XShTgd#l1X=I5
zD0ee?$7t&&ZS47nG5ei-;UsoktG(*-ODmVNO!nF%@j}w@JaXonfmWD|lJdV1K8@V*
z@5is?cSrW3c9<aTF73y+`=T4}#Z8EAcOm=4GG)FeZ<6vkcdVl0v-D~uj4fJ9tCUb$
zjX(Jzz2DfTUGaP5ouuv}WW00NK7zb`5x*#o+>K`=W#T6p+niPU`Of(-Z{YdOcn@he
zAmbCi87(dIJ?FF7rxC6~6T0lD?T2wa9W&RT#0k<%(I@NXH07RAFM3>r$nNgD8Im5~
z$QpDlUPjiZ%-IIW9GoL<2eM}5d#rfwquHy|-;Bi+WSk=VAZ4<TG)LotF#A~zyeq%G
zav7hT4eHSKMz(w&JH+3X{tsEdDwDns#xLUAq{Y9lMc&AB=2(q^h`*ghc5AE??}V91
zds#c$;9|^>HbdR{+L)lNJyA^j6*8ZbevbUU_zvmAZHP~N>3c5zL)PYcxF6}$2gq5i
zH>#kcGAE$|2H`4cpW@d7@^+cCz}|=-#`g}!Hf8(cGkWB$qzbOXS-$_k@$zEZZQ`;0
zdnl$}`js=l)98S#%l9jjx1-OI_5OIoZ%)H9Y3tejJAEI(=Ku1W&ccDd8%WQ3Se#A1
z#tvn_Ri`(Z(~vbOZD+4qj?&t@0=J+gvc_kRT!q*rbLtKZ#wO(cSq0bPVjPC-p@$=O
zOaHUy74=;U*;}&iuf)fQy{ci8GTHxsLOYa3_Sd4glg^{*z6WxKUx6t&1z+f6A7N};
zN!cNYKMY0A)_Fr~k~n!25kI&EuSlzaJCL>eRB79VIm6$iozCPw&4y)B9S!k>v`4g6
zgUm;iYlXw`sqc)<`QmlR$vJC3Tp&JE`g|NLFS5Ie*GF`UKABT*BR;j8@0_<YK0CzM
zNsEuYgsTx>$Q;fXJb~03jgK%1vF#KLRDUe4BO^9FTH3Xkf!vp~FIPm~{&K$>iH!9g
zI7s;m=y#8PZqw%%_|gm3!^5$`n)so0z1(lsvtQ_jmEujLUnIY}vKO&Y*c-2q(T}WF
zWX2y3#jD~M(QTc)3Ca&t_h~YRU?1&$!uB7^UxGW8&06}e__c`5GMBQ>CcGNiJMX~D
zXs7NHa<jL!lXsrFBe57&kTIB{Os#_aO480k*6^&+v!zdx{}(c@^>CrIYklu4JlA*H
zJw*8{P#qO<4=%w`@`h+XdrOt{1qZN0Z>&e=;h$u6BCD9Pmm_;X#;WihnY(88yN8f_
zW$v5t(I4@NHgdkMBYrkHw~<+!4X<Xy5$spW7{pIoYiFYTyVaeFvN+TCCgqQnw@tjE
zw9Ko^kiGQ{v_|gdIlolKd&-r^)#`kyZYS~H+Rhm$XUnXoYs6b&z4Xkz>G%pcXZ9eY
zzB(P$eUYA@qL%oz<Q*aG;d>sPSJMAFyy?5B^5fL&u71uRcOq+LH)Q;8MQ3EZ-ou{S
zu8o9seW$;zgt6-=94oyovKK~X#`QF99Li3!wD}bOZS$*rS@<Abj-pq6M9*H>pRO;U
zkvbocoj0l?WMw~Zi)(NbE<g?B3^)hTDf`%?xED_%cY)K`tSm-|XHSWa@ym?=2xPrm
zh}H6PKe!%QkFP`K{yB*K8!5AtPEF~yR9}8#yDNn+(fM@gACVQmN*fpB6Y){fG9H;D
zS;J!6i{xE^%!QI<pNA>pjnEId1J=h4I2PG2dZ|~9oSxcS%I{9p_A9<KK1a)se#@~F
z=c<!Fye0lVR$&Pal$JH^A>>T)v%00C6a5DohazNOLC>roM=Bq`ABX7pKC<upE6iCS
z>(=M;x}zF$7I|B}k~jgo`OcW^FE49H{9p{O^qq51DfM&KXo8#{@00eK{_LUsSLxFi
zr{E!Uk)E^eWO+k;-%rn(`hStIn(x^-QC{Zy=gMwTcAmC6llcx>V|UcUbfiBS=h)&l
zw8uD1$7?tPMUlMm*onM1=eNMz@neTk(uyE^*kP!Pw3U4*`)kfJDf1RyM$)tAy+pU}
zI0_Z%dIz2B;61#7n)p?J)?zyDz<wBpBb5ILPawa=e4*?QI755|nv(f3hT|IK9##{N
zVLUPqFG1Gosp{1r^8_~hjxA^4H1WF9vUZ#;|28tSj%Dw>7$>3)?j~cDzHY`*cvIRi
zWZ%zQT7DP)R{fmuv#u;CC_hNqnd0~1O2nrRmX>*$`+DAEGS_mK$li4TGB+;7naG)?
zChkK=b#KA#1z}BL#v%K2&V_k<&Tl8b`p#Y*pU;>tQsymX-=f>Tbj)3;slMOO|L0qq
zYN43<)zkWPv%Piz-u}(FOaC72l{;OXf$B^UFDITo_Xz#Wxo`?{&S~bmxb#buJ5Rme
z$*6%V#P5-Iy}WA5T}+=dkTtwCPC@P`Sx3^h`^Xtc*Sw+D70=%Aq`F&$mE<3#ZeNT?
z)}3#a*(U#a`IY6DlJ<qLm-Nif>yi6dFZ34A9dI8U=({O$_q$Q~&+!-X7Lwm<E66LZ
zuLsh9vwplIyiPxEl@=Q=#5=xQu*rYItmXHjHeSSDbb1?ylQUkJyVV)$oQ%x#{2jr`
z@^%Q@V+?ZE93%frGPfF&W!k-6INkS8>U2{+ZxXRj&QAA8`(F7og|A7gNblV3)`}Mw
zc0>`Zk$)EUz@zvcZEz|cLF)dh-&@&k7joaZ3E4Mt#>qYxKhD`?fU>t}`)lpquHWaO
zhWIboDZYxXSD-6<9e{??a&~@Qy<dc%YVT@gvyTiC-&<Pt>eBMVKH`<-ZCCeh^iw}~
zyPT7<9}h&{Eh=KNzP*kokoi4P*=Fke%a?oe<MV~P(`7t)AEFB`*5)@zo4Mz<RX2CI
zjD7Cc7bPw&=lVl%9vb6A9Ly$X;Q+)BufsZ=WS$*~tcOP;I>v|2LDrs?=q;_NywZqm
zhhig+Qg1grD1MzXIXl*++c>%|)}Q#yt@Qm)T61!8M=PQ13CLc!pR`f(vZfv{o;ffd
z!;#;}UQuox-ogam&2h5$6zrwl--T_FvqjeCXSF$#-+jhzu}?ASKcJ~HpCGbw)=!_~
z4`1x~AN!?m!;!nsQe2Pxt`NN&>%-H=r4HFs^d;{QTh)C|nb<5kbS3W}Y(y2bk#-7>
zz`n@1#IJuM?-}7@GIBOsjwcaa+Td_yKS1{DuY|GFT<J}5fqL0zFGW2pqSp@OO)l$M
z&N+R3m!s!W`Yti|DhTgli`*NE$iGq9ca+UL+0kS-M9u@*f2#;{{`doBv|kZ<OSl{F
zU^Je`Hu-PhV)3EKy*Iz(S3uVCygLt*wnLfHD317AH$?U;^0H57?N9tkER&uyV&?l@
zD2Cj}Pp0$H$er?0oPnkK{yCn-Z~ApNHse8g=c$u7wn^fTVFZf!&U##j%-n^Z!>P#I
z`+>sjxw#vChyCcVM*7EOuVlxAPzg2hH9Jlh=KbVwa<5QsPjxOud^b8Leh?Xr^=AV1
zLiYNG*n-?2^7mFf)qfZ(5gk(QHRLz6yj3i~UE(>j_eX5LJ3F0-?5~-7-y-+L%)t^k
z2H7+7PIHR-nZIA9P2^k<A4*={AM!hH#%z=CZWusT)}&#|{!XW??=R@*%j|l+KG)Ir
zxA;;COg6s`wl=N(nScCZ&-1;9^yHVvu_%XoF-hGPWc@?#q2!N1{QL&`{6PM(^0P+`
zBcmdj{n5a8OL==JS2q2|G2)pwZSbVJITuw$`uGR=C8ZxOzNfIE?<K+_@~0v1d=q^S
zCbNb9ysOQe0S=VD16l84ugsy2%4LloiigEF;!9<J^xZ~S*LT*0FUiTiHi52(m}l?C
z4qMslN8^kueUF#kN|_m`pv^agdtssY8RRuYb>vMx`{PsOeuacL`rca`+m-!XxY+k|
z((`wFhl;-<tt3X1SDX$xt6YdbkvrQy%Is9OKYbs<rQ%83Mcx7G)sprj_SWb3g;%3K
zK36XOy+D~Zn2+2;{}yJwc>vRt$sBr--HtIfl`vgA=gHikvj6PD?$R2QIYa%c)O(+<
zOR$3ugME(^zU6ywoJ8kZ>~*I2Kk^1^;~l)BZtRdTJ4%?hro)l>o_(c>aGW}^^{vPp
zy;PeytG$4ikh9t0$lR-otS?>2u1jY8b`YPdV!Xb@8ZwR};}zT~{~Pj0X=gPWtMeP#
zXOfk*`ypfxe*}*rYu8`N7mh*RwtvF{WDlyqPNR@_&pna-W0&;kIbZpl|8oZFfor7Y
z{+#)LIo`lB<c$4?Fn7-j<&V`~b$aEV*@wP4%T=W7tI|#<Cw6!qU2r7wzL~SvA;{jF
z^)CBQ&I`kR-zUs`tBCY5^DJ*m&--43uaUiUe`&eHovFQ>_}p#m^{nr=$jBT|{ReOm
zhHLX$I^_&=xv-sbId9)AZzVp#KFAu}Tizxz%NmEAeX^hBZ2T;lb(G2b_j=^5E_byq
z_|<nk9ER-UTa=A{L)9BV-dV!*Ei%TT8Dfi^*Yi&CI=)9M+^<{<<+CR5COn9~htVN>
z!F=2qeaM_C??Pd1-#h6!mLKJPX+PoRzSD>NPPSB;ca)t@-tXG0sf{DiKztn<i;u+*
z$elCilDunoLi(2VdWO8#NIx6-o`*dP*6(A4|N8EapXI0jf00{F_yU&T3VGQhW{T(T
z-Wn6cf1p!G`sJ;o118gVFXW6l6M5gwS>tr%UVAjs|D0(X;XTCX&d0m@eZKzZjWTQh
zP+TVccjaTJ?8ASfCR_BuFDRk?tU1}IOOgG$@?+U<2TF@?CZnPFC^E7><~NJcn1QTo
zeX$Oc$f>M9eUQ0!uC(}KFSJ!J>qh3p>Aoi-d(md)Iw+rWWA@tkM|@&1GDfp;Bw9+(
z+>PJm9b^ac{+_uQpUHg8c`s{X-roNd{)+sjpSe{_nbkN;nMaYm_H6touQA!1lr4d-
z^xHts+)t0x_p18+jQ($6&t~S=)7GX<Tlpcb{?)oGeGV4MJ52ZxQuZo!tB{woX3oC3
zv*%2aH`Rx=v4Y&S$~~su0x~O*mHj&Rz6tcXS=pzAc^~UXPWqI6@d5M`&mE`<iYnik
zth{$`629j<|DPlG+&aj6NPgccL7(;V4^w9a8L?$2WZZTjzs*;YpLg=xaR9xtr`A+=
zI~lph=G|nd?`p~&Ew8#VWtCm_lX<>{-+XV*2#Y$8ESEPzne_84G!(y>4!Q5<-S=B*
zTky8D+;?+kE|0^-H%iOznXQz&!FSfkoCAmZ{zCdJ^&TVdA8lO1ZVj*r)ur`N=4WO9
zRVO;<F1$tB6!o^rKS{XC_h{*N%0C=caDneL=w0RqGK}q2#^w&~9IX8QxLn<w@ej!u
zuv^xU*zk1BP<MvBL&<qp{@=*DKUKXZWQ<mKfG~Y(uiUG`r+jD3wrD43h3J;I_?!=(
zK;FwcD_0Gx@eOibc~_a;<>fv4J#}WQ+Z3zBzgO>M`aFuK@HlRx%S7c*(q4WGIRlHe
z*&1iD+5P%l7uWgzgN&1v`voJk^B+bc_x-%N<sA41?m+INA4uDToNb@OEbWh0_7I$a
z|9gf~?iJ-<H}^6}^Y+(4{6}fo<MO`ple$@ljug-N@LQ}E?<qfL!rYNMBKNP<I||uT
zDrzrhqxJ>%`Byx5l-$Ek#UA26OFIWC_q}rG((8A5e~=%!JB6K5S3GYQhtn<l(Rk&`
zD8HwAABkt(%=vJwv<Ap~Rs(+DYy%s9%fG(kD;u5B&4D7)KVj1&**d>>+^76naxNhw
zcbs#zk-cOHmT0S}w5Q3QPiAv<R;bff`J4@Zm6vn#cATaBy_kz5*m1tFEQZVLijR=9
zU-sE;<b9{Thkais{ce0Np6|^V=DUu%(@+mNk6ozVi`Y}W;@a;??@`G6)G@yQCUd*I
zox*j}TGRJ?e)B1vuN97`cjj}>?78>loqZs=^R;!JHugr&q_uo+LfXwecNPA`Vzfoh
zRPAv+u0zf`dB1!e3$YFPTjqa+J+WN=2J(JF+G&A_xKv)wclE{RYxf)aT&&C>@!oVS
zNypRhn)pxpcZ{$P{u0mMSNw_G3v$Ly{~y)2?4>UufA8?K{#VrJtaUHx|3TtS@v8K!
zVR>V(L4Nl2a@xKhy^+6(sf4Sfe@E^E%2rn=drjU`^VV5hm^<GWWMp2&7jlNrK7AW<
z2dkzZITN<jzfXj@FW!aRzq5~gK<<@@zIoqkr_NW(oveKBYj64f8CQ$vJ@ZX@eXzUu
zYBF-a%UjRW%72e#$Q}4SY{2nsdw_Bckb7$0ADff6QN6sWPC?$Ib9TBN+k7t~?@Iak
zTb8YKyd7icwmWj(D2FTc{{i-#W__Au9zF1rGcGQ**PO=IOR!wt7~uiRorf3I$v$^I
zxjCnvppCq}=WKGNHYSogOu5|8?j~n2Cip%_UVcy9n_d-!IU6-Zen-pMCUSBYUEn+G
z%|LBT(^g&dA@gSSFOxqN)v(BSHD$h$_nb04QAV3NFZ9yhG<okUGhF_IbSSBRKVqr=
zJfLju4HM+=M^^T(Im+Fp{+a6jq|8=jhqA?N`&4CX&6mO#$*!eLPi2~6hjM?CowVo4
ze1?qN%~ql-`8oS%zuO=2_0iJaz_H@F5C4OlBl`GGdMEX&kkf^{C&<kiA#Vg3hxPJ%
zD7&1jqI7;)yp-~P(DMuVId^2=zDR!V?P+@wS+~mHO82`r%hR6~<n^F??#x5g&%6Ie
z;uF-ZO6~{7A!qez!kfs+KA1hei|_dEo#OdTwv4(t_f{6>y(E5hmhfWV{iS!+#*1Y4
z62DMh#%nVk7Qa`SLzT_j(j;X6y;gcBER~n}{hYcLk#+qk^(N3QzaQmyqVwr@o$z((
z+4Eo1r&o~mVG~BO+Yn<lPk+}*&%1VY`Wz_l2l1iWYo@LAZMOcT&&QKFQ#e37TcoYQ
zN*wR|4Pn-<?9T@yb#gw;@2!6$`$5hE&GDb~?Aepm>w+F=hPvwH{4+yd-h*;($k{XT
zyd}&+ew(^Znd#aXL&xvwnsY)M@vM^>-~85ap)uW{Z>99-D&JWLKa-btw-3oUkF2S}
zy#M9AKNgvLIcw+q{3`prhRtmA9=qh+{V*CKcaZz|OE3Fu&u`QhcCq)K&42nibB<%z
z=hQ2vTy=V7o}H|3N0FcR?H}ooGhZ)Z#_R!o89~1?>g4R0y?VYjbB{h>-bc#TSNAP-
zn_0KEVmNYNcoo@C^G<q)GX0f(UY*=?O8TCHaj1l4+In1APuT<2e;t?MTxsK#?}VJ0
zN^19MI`q~5yXf;GUGiR)Gv{;i^F~wO_p8$LPEr~Dl+U~9ee!#d(?MIWk~>6Or)c9B
z;V9oH2p6KJv@U3iThSCZp(<+OOk@vzS(xAc^B&w4`FoxGALhS=uVSJ6_{C)GBc8f1
zp*n6sCB$du(dT4k9uq&5j;~@3nj`msZ*i1(e54AJmfx4Y6c)pE`j$8JyKy!0W>v=c
zzNX(<^GD+pG{-i1^VR93e$H;+A@8~;(BV$x4WWqdykq1%-<O<S>aD`w$o`SH`F85P
zrd%oIkHX%_xjyf5hw4Yp{rBo$&cXGS9h5Oq_XA`P-;?Zfk>8$jcmE_Zl-(wL8DApj
zki1{z+#4I^ycXYlko*D4UP68!bVOaW#&qOuxh<I=;x6Rglyx`1A@3o77<N-8f5%V(
zlYNgM?`-*#m7AmdiD*Z+FX_03z6atAx}U=a2lA;a%%fT6(>Fg@^TbE<$J}{85U(mf
zb9<C<E^@!io{=>)^7H1Kzj@m!tr5AGDtokYrL7Cu4?1Zh_k#BH*?>Qke_HwF(*DAa
z;#r$2W4L&J`@ET~j8h|I%*Ol9Z{#@#Z4f@AUTJBQwO3r3V%DE?>0TF$wV!*;3(Dj@
zb2c5m(2se-G4#5Iysz2e3Hgi2`azgI{W4{)_jY_e8O=B8x3%W6wQRq}99w6NT<?A%
zZzMVAYa?s?TsEpA{v?@?lkqm*Lw)ks(P1{(*&A;V&)dpZ=q9hYw86-lbD6x$rPY@A
zwZ6<G_gnoLK$l*|q?k6pv*+(hy7D<!wiK?_X6~+^(R%|OrjfBsdhGfEnUBlgD(@I|
zXV86;HM1(Yzp2wvJ9#I2oBoHXd!_hW>a<sPDso@CMcQ6?QG59<cq#T3&pqhR*pQ5p
z@=N*7zPV7^Qu2P%<}TqW<hD>|sxbG1fn?sKjZ@Y64O!cA&$u6dqlCP?M^8~Fb=T_e
zZe&K6y2d?ketCbIj?C{X)Ol2y<=T5)d+(?-3+=?$>*wL@cZB@S<n4w=;$z6`q~7t`
zIvsiI&u>y?&6gp<<H+qSZ35oG%NVNt{H@1ax-~^n-=8XXt~&Ysus3qo`bGV$yE&hq
zE<fjo+$9eb<}7-mwCwHKm$Fyq@2MV>wpknXgsrexys@%X)O&%h|KcX`i^-^q&e}iH
z_%+qP{C#6Z{ThG{@>0JvnfsER_uZTw)(G>vQ})C7{1o=dIrv5lV2?-DZLW>R*j>CS
zUpt=fzwB=N$k)#4!oBUc1Edv+FS7MAbx%|FY`R>d-P{|qx8`@Quj!umjE35(B7cQ)
z$7^$-I>j(Sn`5QreX+c<@2Pt&-Exnux62s^uOV;yr{D-2iu|tESlRroLVaQ04F)6)
zSx56eIaB&r-#P!>Oh!o@>pT8i3cKjlm`?lCrxsm`(&b<APxU?iTSxxM!f$+MPOsxX
z{mq>wY;rrBk5~6Xavsp$qvSrKop+J5TJAPQgjsj9pEk$k$o*`YyzT0wpS%0c*yrxv
zN`Br&n&Cx^Kz<M0jLw)Tf3$M>4dpwW=DRb_!=Jv-Kuh}kMdyih&fk(O(1(ZEVm_`z
z?k(A8ONtK>mO?k=j(I9tsJjocf7Qhi7(&)ow%Lf2jMd#ZT)ZZqtKqHVD1NbFjkZ@=
zJC^hL1#CN)zb`cJe2=kiZc~1$dhO5<_aJw${=zfpncpkQYikE`7u$l?`jWfsUu?Br
z-)iV*C*fpz7NOJM>YuLuGF+zqM&!4F8`Z6e|GVC+UsC<}etWdR8|aL&^0Ky#MDDJ6
zCu^^b)+k><zQ4i0WbdzBbgP25lz-m%eXUN{f-<FzQQl+T_1#zcG4gXJ%vt0fdgW}A
zHMT9?^FA;<y6Z#UeJ8TzvF1}pzSY#6I#${+KGvDf{VVLses`i5a+lnnyy!lR{B7uq
z-0gpp_q+Q2jLV+Z_e<2BqmA5QvghX=EbIEQzAsYtFvL$f%F8{XHb#l(T#$R@Ms-Um
zli#yGl{XC+$#0;}ROx>zb2gpMqUR6h#)<ULexEnCtlvZBpF^Me7)zId${w%YNcpwM
z$@|cg!c)lmmkr93Q-zEQ)|VmHng`bK_pkWHT0X#!4%Mgf+8H8#KOPgWrOtJ+71=Lq
zFK+^yedl+u+2kHUkJq$!g?K-lDt?u;eYE?ZFz<o~O3S<PQITOBc3I<U(&Hz*ri}(@
zA}?px>;*>(^Y<KIYWqfgYpAUvWM|+1Se+uP?0xh6t>PE-5btb`G$*s0yhi$2o2(kj
zzDv*C)3Y|dPLI4T=AAab@x)j1H%li=zeK#X_}|Fyv)K!B2N*2uEbLFthvY6Ie;(WY
zY=3`&t;$M&iZ5M8zx<}!oUWHk%i7gXpYOo0;+utetIC-rb2EQmTnwLz=gfZ1KlTc|
zhjBO>xpVwZ<^VEtHq4zkcac-X>(iwVJN6ULo{}@zQfYr`dmrC<htC~rAo8x7`{zPr
zPUd|yZ&#0@74r9Q@ug4bmfsvUi4W0!*4NK)4f58s2rnQy{;5vJJa>hhCGvME`EBP-
z`By8SJ6pHJm7OpBdg1-Z8(V#K_flsEJ#S{;L&?b5KEGjPeoZyzyX)IxeR-y!&sjV3
zF4q*Ru@(7yysQByBX1+GkylflN7a7-@tqymiScaMgRS;9evQrD1NnX>XUt<z%RRT8
z@my_OmuvShGHy0D?~<4QbCtVOetXDVsv&+NT`Eh@?>KXm`Bm9N$eyQ-f7I<y=U(!D
zRW9e9)99Bs@FLc&x6lRIPg|4ql5&ldzgYb`WX;z9deWC9U%fq%vCI3=9^`#5FZ=I-
z((^lf-tqI^vt9aX`NP%E9bmgY=lyCQa&M<!E9vpS?b^%x@$T9@NSV9~J&^gQ?6=D0
zT)LH<yw&VWc3*8QL~$%Z&J9P=|5hB0629}+{k^gm2)CoEw3etN|5IFs+?5yL66^PJ
z<hQ|2(odDv9y!+?=es#hL4G4Bh4ekY1>~(Z=ijI3c0Jv?;xOc0<OlhA(|HS%(NtQ}
zcSF{l_*dS4&SZ-XZ1b_cuNBXEEWah*f-S!LvPVU0-nuWn^DH(_3#84_j~V95Om*i9
z7g@iTIIAvmw^(7n=079ZE@zyYWc*BK<X1p9@e_pc^X_!W7_Ow#F>JL~-)iXR6T;j_
zW-60A#Sy~yF;{<ccKS}8BKo;Zndz8-d)XxC*e1vvxDnM+N%|?u=C|twct={!qq*zO
z#(vs+2rF>pO8Kkp6>F>sd_RAa^se$FmFpmXA#xASyMFw&v-JPuXFRST`yzVGr{Dhg
zk-irqe^2s}Ij{=Hvq?{LZ7AP5!uK!i-UP$UudK-@vrT2?Klj~FnYqd?6R%(lI$)Ig
z`5mn!y|#&OG%hb-nlk%Je~4Z3yTN(Ld*lM;GM0IFyp~Q6(lK_*+wM)`ljT3I-45uO
z`s(Gk*xWfv$j^CUPrRUh-p1b0PH%dx#zD&GEWaFin;lHQvij75tQ++sYj_iR$C8tK
zdF*o#9d1(hc>b~fDn7fEjtlt~K3?oDgU^=m=VjJAdBde&_62>WIfr~^P7A-_cl*p^
zcV(WlMy+9oAL*F&wk`7iukt%aH~b-e5j`rS6mGy$y56KO2kFadV_e)^FHWb|taFbb
z@1)s#+S2)V_Q>B9)zr5o#$pgg;39Q)({@`rb~Fd>o@x)K>n^@C+<2@f?+10pvd4vZ
zmX29R*3f4d9rNDX9c8c&J^zxIx7lKBm-ToU9Us^31H$Kg=l*%6?`P!ay?Q_Li`v)w
zXlo*!FQU^2bXtqD;s@$mTk?96-BEjcpbr_j1Kpynyj^`Sp8N3@GV6<HuR4e?=MMOg
zzQ3Z*o$6ew?Agj~)b6>$yOqt~4ZNdXDSd3Cz4x?Lo!l+>Kw5r-&mPr`jGc5iI_<OR
z4dnL2aaiyBX5r<?`+3e||0tKUzy@tD)%Iu7Mw_cIn`fnwzuCH$ez&3#R>_agKdsJq
zVLyFnfXUMDlb^pO%6soc;_>Tp!tA{ti{~7e-xe0rxf%|q@3Y!HPP@Cr8|uf|Z2Ob7
z;QQ6e+Z%Jo8Lr+&3@7J9<n5`YupRP#nBO9|utyp7>tY^$!E`J(E~gl~%Eqn@E_Ob<
zZml_lOZd|f{Be#pGynHv!!zl1E_qw&xn92pkpG0RmG73)d(iVNysF$D*r46B)ooK?
z|J>7aZpmJlGr?_am_6hrYgc>ZuHIh$Y_gge+nvUEC>ei}mA_j^|6;!d!qQ~a*8j};
zFX@rr0CwPd`S%L{kai%}$<P0t$a_~?eO`}4#B+DOgRc3FC1=oi;(tr)#DB)=^Xcqz
ziuUuCl6Utt<m3+Xr*^w*BflSy(oRERb$!TP^iX{mD*Qou?%lbY7n8pUhf6OnEThc(
z$Ud6>e=IF)c4_4AB)X!Aw9DmnM^WV(;urD!twC*hdDE@KmU*L`OXs}#Rg{+XbPwd-
z(Ng|p^u7rniEoj1wsP+apY>f?dftv-R==-)ZZ%GiBk%3gq!&fb$S+7cim&bcrS)l%
zHWnD0Ir?MW8;##)^NacB(qewU%$mHynDC$6i}S9L`8?cMo+iFYcn@ZxH~D+fVH9pd
zG2hp#^B`_i=NI3%pdQxa3O<p!AM*Q3#(to16*h>U$8Pca8dxu$zfrghc~{T-`P;}_
z`z!Jnvtj<8BEPxhZz1ns--Q^jkB_ZpZwy+ce^~jIHQ2m3j9;9nt&f$Tu54T6uADRO
z1=`t-?5gzGi+(xZKcv4o?@!g&3;6IK{H_!kS=XPjE^gA#Y04dk2GVNFtD)Ru=p}wR
z{~Iox%>MhUTMd`RHnAhQQ^Z@->jGm_Lb+dj=l>h!|F--mzlC_-nh(}~eGFH>jc^KE
z=Wl49W{>=R>(R<A(Pk0lr|Mt+o}rWe{3HJua-w(ULRD=HK@IWT1GBc~?(`srikG2h
zKRVn@Uj7a#d;F2|`jC@#FmES0<ImOC!TjUdmDZ9a%FK7xm~GzU;(7dfv38f~%ThL8
zXuX?dZMDx<{mlL?9Kq*)o-19Mm#t9)wVAzSfwppw%Wp*!F+f`Gg?TqU2H)vR&K{$L
zCD|eO^f%Q%bEElfJZjSGH}QL8Bl@1t1}lVbs#8z?1p59(@1A&K4ZHA-3+dYbbGDvh
zf6$Mr(w37Mc_r9mci~89i{GrHlk`95#lCdRJIFK0nYn?q+=E{*20M}cwG|zIM&?i6
zJ(kkp4CHK7PJh24_Zu=_H1_+@W!MTbF%Nb5>!tSL+^;Jb&)jSCe%wJjQ^~uH{1s$R
zMt*Z|tK8MvZAktX^vQcz){wmQWbG&?{Ra8H`Pyw?vd=v8A73tUt}#y@q(>z(N<=m;
zG{zsNFXDO6&f7!YGaj(N6hr<lXeZlGQYJbr)7GhMU0;~L>&lzTHoE2a&X1(!Z+F|7
zX9u8>c>cbiHu4*6e#3c0`S0}O1oFyhbGdpsE97q}E><S<J@>mt$bG#U-WAXLRerZ^
zsNO&H*@I1A;@|Pxw4dJ=Y8ks$`dDq1Go5kI-<3V1Uf!G@BqzVm<Q+WoDR2LswBM4v
z{LRz(#{Yc%o{Uv^4Npnyp`A8t))b@2Z)*G=z((<mTX%jqLAfHv`dIRM(IvnCK1J6P
zw6jc@`(sgQ&&oR*dGlG#2Cry0ziZ^|R#kd)WpkIw-=pL%TDVWO*u^(+4*f5cpT9LI
ztA022bFL~$-pA6r$*-*5jl!%4xo6iPGk*(}{6)x~a=h;^=(Js%`7O31-Jd6?yMFZ3
z@AdR4E-k+;=Wid&Vv@4oNxPiQYv}V=?6R9W4VB4$JeJH>Z2KAAOVHsO<V>?&+FjDl
zL4JdOTKF3Bf1u{lWsG{G=(tRHr1B;7vp@F54SXoS59T+=o$BN_(Y&wR?mPGAyW~&9
zRush_@@^or8hx^^KhH+b(>bzphs!%?-pEU9FYD?iW%9d2{&r=s^b^$0?<yzA?<0L5
zea>%BHS}{n+cdx$GV^Zsuzp`e_KDI?GS`|dw5H9oF3cxyHXW@qcUb!dh+ixH5{#Tn
zXZhWh+DEKAmm2%(?0pNFxhubitp6>LeRKvL?xkD)rl_y>nkgF}okgF6d@n>r<?^>G
zM=QHpcsA}={%w6|i^cd{pAKM~oXyY0Qt^Sv9-Mdeyrbn@n78&*QBi!PaGbK$*tNW|
zn1NDb&GGTpVC(vVMObMs`pVv`|4*>PLG=Dk-HPg+sNV62j3czOPI}g;*7VCBdWb#@
z<x9QT;1Rx*bL*uz&3bbyULkLtGJE4n^Xw<(4lx$L>;JG-e0i|FtA{nAvvF^!uMZgG
zj_UPw=N!(4qv<-%czkM|U2F{1ufSe4*ePd~8^kw~w}>uh(7QFc7YTc@-5mD#1=*j!
zmtKJmP2}ZWDL#IU{^W0Uzp<X*$_9(s;C4FSpg((RXD=LyVb+M}*RT_PuM+=?tkboZ
zzuQ<thjnx*jZ?{PD(@OHT3TmbvUW8lx5yIf)I9THmi2F{^UY-I`V8}Df%R>Xe)FY)
z^2X}ps0sXXJU^afotbGZT56p5SXa7d{<c8&sl11D)@JT0cOdWihhnMp{N2*?SVzbF
zeO4Jdzd)Z-#^(pR-G;n_<-9tG9cr_|L+V~;+-5C<zV1u+qlIPkd5?wqJcXYchZg#A
ziawpdS1K!Ck*z8Ut7zj;cATe;`=!mnF616E7C+N}hc<69RzKqjwz`>)^U;SLbI-}|
zc<I9<;=i)v5VEpnpGofdtI1kGcXn$18QJuiZ*CUXZW-g5d6GLt)`82&JDL0*WIu>+
z$@`SNAL&qC{%88#U!UX4Ig2fpegz(9m#3%OFF&##4YzK*Zw!XgbCmm)@#tgRPMF6x
z*d=Gg>-6JOZRMT!c<if<`|P#5;XBM>`@B!r(MEpPICwk16)x3&3FSA^tEsWQamD}K
z6>=}xANhNpyj|^uztw4>&dti_-hC%MuN1Z+^LBC~FKfZGI9%Dm%8f?;uCzAt4p`iG
z&b{sEGul|4M)%?LyItE)nv+Y6bshFSV3jkh@Db@1lpV@;c{|HG{hPKnv(Mkl`N0x8
zEas=eiOcj|dhV+Ey=bIxg1Q5x6*qRv$zMx;IX1f%V~y7a<2hY^ZDn_`&;D%KPFo#}
zO?kSmM}8xCL?2I;{y97Srfdgw-k{rceD7fT7g6^B`8O(?zkRQ-zk9@XY&aer3hcQ>
z`jPV6k{`d%?*l!MJ6wLB%Kno1m38TGW!{$dvbKxqb2arkBfkyy*UpL3?~$M18CI%u
z3B7jm)Aqs!>c`jf?oyg9vQD%_ao;Dh$!=_an6lY}(q{g@S>Ayf8KV#BxI5i1#7XSE
zR9fy){|a+|x?H?2x{K%ib~;^-RR1;MuhPHJ*9N=zqcDHt)}NgG&D-ztnxYvl_uWx?
zN%=402kh|ukGxuBo<&~npbh9&Q(s=zztiYij=!|<U5#zBC-)a_R_<8gvD$xJAL=1*
zSovG1iN<rT@@K2xSN-(=hxoukK0n7AFxweUc>X+n6SfwP6P{siFEsz2_}p1<rgc=g
z_vUt}S`|B$$(!J%tIc<Q_3ajI(d$tDus6T@>nm%xHm>7e*9wRD-pJ3YDU&<iU&?<#
zpZwmLJ-8J8$Dt&Bb2rJ~x#VvvS0nf0++)(8*Mz;5$s6Xg?38z_!%&^Qa#s2s`F*BO
z<SCcG-yAIO4z|v_!P~~9+!*~FXFr(W+%w7eTf55Id(T_yoXRKi_ab>`Ia8gC&4GAJ
ze3P;6CH+$}@2B51zSL4b_SB!o{OJn5)KdM0=EWfM;c9kiD(!b^Z~8u#ZC0?y&tLKV
zMbgIG3*R(vUSiiLjN5&Dxsko;7W4I1dNncU+Op#t=JZIuN7u9H`u<dN*0_wGZEliN
zhuj+G&NfWf-VkHb^-JTy?v?0wzOm2mOb?=!?-9y3G-ub*r5#!+ciu{4MX!wIVfwO&
zoGMt2p~|#h^FQZ<p28#Ke@RY$n_k8az1TVD!lT*kMr~Bp|DUZhrODlg+*=p3^IU7y
z=gtE9Fh{?CWAEA0KA6uYbJ%YNAF$s(Y)lT7zj>m$ul)E0#)^#mtw|&8=6B7l+8JPc
zH)5xDGjGcXtNG5G+d0T@kbf%sIbFJ_e*k^*o7n}v&so8@=rnl=+bq`~`F+`=wlX)c
zXI1vz@C<u(;8Sg^m%@X)8LyGTdFH436WC-6Su5#uw{RAl4`$bX>~%YTn1LNQlbwd3
z7hNtjJ|);}5t`Gl5`90x590az-!5!FW|eaQy?)T|ZQ5F89Ir4gUoGP^#=G1M^U3&^
zp3Il0=nor}RK6HJMw4-ozJEu48SyRnMB9;DQ<-(@)nebxsrQ9*0)L-D_UFd*GWvhS
z7G;(B<U<TMhQq9JL)`&J+2@Q$7j@s1e~@|~v1NSVK5Z1Ie`D?D?-hm_^Hy}7$Bx~|
z{GL5>7JA*>dEeZ5&HVh#{K;G7QhKh^evg^#Z_IBP!)M33OXypPh5Uvu?X6D!UN3it
zzt}J5+v>vXO>asoMQ-NlL)d5g|D4ftc6mhIuj#iEPim*LdU?M%iQZG#>NI+0-^_3G
z*=two)7RR(QCRF7KA=o~!_M!Q`FpMW->S97Is5Hc^(rhj7Q$QfvjqLy(0lSoc75Oa
z$!<T0mt?E$E6fe{JwZQj#mW4pIr+nJ0iNL3Ih)s^V-bDI`n>>otGI=Y@_Wt{{aK+O
z^_06xot$~68PnPN^ai`1rtCO2$~*plbUkjTalnJf|G&%|PG$OEOIAPaOk$h-jo9_#
zcab}e+?LUY9VXJXm-cFFdkndG`_H@TaA8B=|Ix7yIeY!$Y(YkTfBC=qlaV*l?!I@T
zxHi7Pb;$oqSQS4|W+eOHuUzf~`P+ha<lbxyb2p!d{N|p&ySz$z{%+}eZFbRq?mFp5
z{x0)Y-}xKc-Q~9trcU<Rdb_L-!jtg2elKT>ALKPe?#hSx{s)8PU9Ze0d@G)FT_rN}
z*7PBJ?nk%$&H77p-h<AM(>Z@vlD}KZJJ`v1N_?1d*(>ty*I3$b_>qjPL8mGIy}E;p
zQ!h4sWvM-MzW&b9p1JoNpV<?8&-$M;Y#nK5&*0bA#VY3X0`a}%eI)%Pbt=p=pBK|(
zm9vU<>K`^ejO<^@j$if10em`db0ehHH<mq&>E3j2iu@nO+vqgG*q%?n{FX47uKE9x
zPatnL6X>7c<?i!65q}_en-k?N*Uw?>op+zEc$eMtzSu$jKl1Mve&9Rj;_Q(*r(VRq
zU5&}wPkfIv-V^xPB>rrE?`Ex^GnFl~e(B%9MgNDU`;Px={Qn0oM5ToCK~qI4l}Oqy
zq%;&tp&@DSMDzW6y&AN4Nn1;MX{WvS-a8s-Qi+WC-Oug){r&O0opY{po$Gq;$Md?*
zdA;9q0NHwy?I*Ff-+WEXmxM|hJNh>DKk`n{C)$61f<B^pW!wSrNA}v<hwl+Dj$asK
z1NvH=yiXtRy1!yCo8uwRIC;6B8X}2qn$c@Lepj1L>%h*2_V_!=BctHj9QhT$L-3lw
z=g5t8x*hN4j`kStrN^;o6y07v%HGDmK#l|K)k~k&Y#;4&MfGT0|LXAN1b#ZA5AZbA
zGcTgi$JBv%U2B|cXx#<zh2Q9X{ZmKq=ll{sjPm${;x*dCh`yIRZ|-QHxf{EtG*<^>
zJf-i>L&+GHf2Y25+jE(CoFdmw!q-x|(3T0G9UYsp%Z~YEMOhc9_4WOm-&H?|Z~W#o
z??>i{-hIzEcNX;Rb$Vy#Xz%@TRN3_n_p73H`hJF&ck$jpUYkvxLl?vpIqu#s{@E(e
zVR7buKI5;`=PF))9nIf+lDi8&+soyhVNPFsvq33xEZ4vPDgN!*-Fx_=I7*Ah33OS<
z*eSJ*rT0hJDANYIG1mZN*WgEA@cB5t>)<axextf!n)BXxjlLU;*3%>K^@=fi<K>c=
z`;%|}y2xI_TzG_<%!HBC_}C1(n9iRkh*3ULeT2G;KmAF+Kh74L^TY-E{9FD@*H5qH
zTaJs3RSacZELYHZc6y$Uen%y~Q_C)tAI$UGEP66WSz}CJZXL=0=C6UHd)bN}+p~2O
z@;v<uy`hw@NAjgEbdhrp|L057=zpg^mB}CNk)1R~P2)FV<D!m3@N)+*CtSC~=OA;<
zH+LzzE<%PosD(bC8o%5%^IT?kx{7?R%uR76z9-yf(?9vEWAtt<y}r>mOsDjXYJvuy
ztLmAj_|HFdJoyNG*&znm=7lY=fREi`tCe*99$Ova3%%LuEw*WAtigQhL-R!6Eq=h}
zt@rWW-Qs_bT!<bsMl-tm5dWiR;e~w)k)aE|%ekNa1WZFk#X}}O@}Br>hjyDME4@sR
zM{40SnlGJsMx%V_akAVU`3D}Ny8arSFZN6^I<D`TsJ?25w>$bp->NSm|AoKgqJP*O
zHAZ9p<_p@-81JIK(L0662eW(j50uBak&Q~>Clwwi^Xvcg*-N+4dNx{ro^`w*6){hQ
zC(--IlAdjW5^Ns5Ye?tW8>p@rJE`rmaihKU(rA(EN_4-I?(^&Wu<P{fSH&DzjsF_{
zUS_{er~`_=JsD)aXWXA@&U$2vYN>3fsJ_wnpV9BqGq@k^okricWkk{UJ@1hJE^-ei
z_de%)JyQ<VMA7%I#Ze}-&p5fz@5cJYTy2aOeM3AM{Yk!PE%+{;qkK_?OfQjRvwLmK
zcMp2P`E>p6B}a4TuWO6G0ck`(ziFGzfAW#F5_|QfmyZteYd+I@2mbk4UOrQp-~F`F
zQQKeIqVopp<VnXK?p@HQ?>05wb{M*ouJ)*v_KClv;s#%H=yMT!?KXbZll;K>MZVF>
zF<Qe6+YbkvC$aCNbRNkNeV-S7Yjgt7(V8@>F{0mKF3>gv|Etm4uD?ZRQMB(Iy*FJ&
zAJK0}qVFf7?}nm%nCP3q2%Dqdg+}j*_n>I)wT~_b=rfM~!yWbR67zp&{uTPI;+w0*
zn>@0CtWR%3c#ZZsmmQP$#KXUI7wt_BGgeddk@F|<Kajk~^r^4!X8K6ID|x;7ptdO2
zL^|#Ctz0vcou;tO9<?6bbUs85_gXFB!=uGgA2#a5-<w+3e<Z)v=D&6LT0^nh(LTdS
zwx5HK<?3(zj==8?ypF_cy1nuNn=ZO+Z1!#9{Ds5TFl_r7ISQ~>Ve@`Uj!k%u_Bh+%
zp{!@~$~QCEwY9$Q@`rRs_$9vU{XmDtdX1eHkTEw~e&pWk_)hKFggMS=Pt6}6qKj^9
zb=<K5n{1$~_b>2mV?^)7#(U-!`pG~)C-_lOx_@GVJ?!E3<VLCK$Mcc-+UdXbVl{>H
zWoR8LwiI9Z%XI*+{Ai4EN0Rf4Yw{2}ZoabD`5W46{wK5*;Zt9dG5VVV(Z2p7d=8<H
zBKwWIUCpzP&0XK%-~CZrl!}guxz`8%8|iDPbw+nKZIAaRu(mnf#I&)&esX+-pAYEd
zKe9#Z@A33EN?YIK_%!|>#yM`>KiA8V>@>hLkq-aBPqdd7?TJpuYk7K$_>b0y-{ZR)
zUhgqRPh;IQW`5(G-oQ`T{o`->-z>3*7EEW~DRR>^`C%&GrNdYG$ai!)Z!{h!;bR7U
z;5YkR{(*YY?M&n3rPGTG=^3S21b<LYHhDz*B$R6zUAPzF`s$f{)m(XI!w!8Gpc(ux
z32#UEQ&c}hes|+q^4fhK{wMF17s>l1AN>_iy9Ub5L-@;AbcMHt<JcO%qb8ac|0B(l
zLmbSQpNv=U&r=_0D~R_@=G}z`x_5A`ey6O{FUieUx`aSy(f4K@Zpx>Q(R-DyzwvL^
zH}O+azv!E#OOB`L`#;Z|!t><)>K(Q@#^2`B$MY-5vQnGtd2}7c)JAPjlYI@I?;Bw6
z(Qzq#_T|65^-1V|o-S&!{p08tKDvhPlJLD&zyF~c_<u}Wp4~7`OeKkh=N(HP6$i)R
zgBXh9Hu`q?OR-j#9;4iL-!1tRJ%yt0IHTXK4f+K??hoMqSI_Wu<Los~=_4@A7)$kg
z;}15ytq%E943hCSJFMKN20Mg*bHA)#l&=aI<F22{YP{&(f3y}IB+efauV?YKlb;^p
zr$^~7n0KhLe@DGh9`r08qxZB=dUhC{M(<CGks;z~h<T!K&!TrJU5sA|^+VCVXG!f*
zzxH@oZtN?@j^07tCTH|pk*xS`joP5UTxT@b3uJGsPY$w0-<#a$UK+=qznk0rop_0`
z?*U_cfsSjB-Ys1Jn>;9b9~gbB7wvz(h+aTx(cLI|C!7P__Dr;fj(%qn{r%mUs5suz
zk*^Th+R|4o`q@Q>mG1q6r}mDM>0la)bTrTPJ$TuSUU&bNzL`Dy6Imal_eS(`;(&2>
ziIZ*eH-8z;&vx^r6e!~k>z%E}bY5XCUin#8eL8HRXZ_c0=g0bF^?cvG>N0lEY3vU4
z`JtF-Z~SX$mh-vtWKPE_`cGv4l<e684^d4Q?N?4?k7!@yMSK^)_cj!*U(-59?>?h<
z&DmT>-(5xT3ZFwi=^NFq9og_M$LRMfO|>m>FZw2Woc_^wJByr0HE=Z)t%tjrZ#(@}
z5GzqFpKb;I#Lh!<!?WUTFIlIN`4l-{+>H*(C3ubM_N~T`#wme{I{%r^MESA#F}ZLr
z-`&B_>E(kx>_#8+j`ANm%keFLn<<{A;1_)|Pp(-cCoWcxO(W}9a&0g5Lq~N%6FUD$
zT~kYrt!@5>Nc-UtWSMQh?mPGJJZlwR<?$Yy)C}f)VKUpz;kWLuWvfqivIE<6b1&NK
z>a?D3Yk%MMcL!iQUf=Tk0W=klwaA*AtPKzG3;oy7K~4N;W24LBFnXso%Do7yhG@Sc
z=HF-QXuo_f{YQIKf3v}C=h6G#tQU;wxi>x6#dFViZkIWq!^7VT+5c<rCWgb{QT*Nc
znd$VrfRCcL=gUK=CK^Of<=N+cwkeF)?BvL7jHO$RL&o>%xkogYc^|?<c|1h#Q+wj!
zQ55YR-pxigjPv9GwrwD1C9&Nyb(k@VqW_`aQ6ai%gpO=un<3(-JN>mMdt-KMCU1By
z|2Oatj}dksAzO=m^uY%0JpY=u9ImIV;8U~dc7n0V_qhJI$^In2UxJ^acrAv{A5gTu
zSkEt`_n3XON8hpjL%)HIqJ8!)c<D{%M(p||KRP&*45-~SwIp7ickD6M7~|z7J~HV`
zHI6ykO_W328;Flz=NNA;JYB5rGtLuye<K@=Mt`C`baTtSAGNnz>N)f$p1XN&zcK2u
z&s?^;GmtMC`&oV8p{K=%lh<v_=wS<fYqC#-)3@<+o9v56vll(y=efL&d%jZNXe%&I
zy@Cef`7wQ);q3wQHJvX{ij7}gXG0h9{>Lo7=vdsnJN#g<I4es>vo0B%-Ui)peN+Cw
z&epEm(BUBdFh!rFQ)2&^aqxMKpNu~)zEJc1@)w^kV7xcUx`i){Sx)z>*pIFsVb3mX
z7~$pwa*ydp-eGDYx*aqK=JevDN&MwgzL;QV`sq2HUFmBbe)~CZi5fWno*Xmr{wy6g
z+9HSHV;m}mkAK8aQG6~HCppiv8<~C}!x{4Qy~TG>^u79Rvi0I0(Kqw;$kxkqdHBdu
zauqrxK6la6S~ZG!UZ>Y2yzU;#S4O~QIzCRvS@D)<A-#~f%?`5j#VSA2?PYn*xY2vz
zb9mWi++X(?_W;b?CtjUz)4wYJZpjy;?=U;!=jv7Q?Agcp@XVjt&9f!+UxLy&kLt=3
z#!anH9dzm<f57{Lr}5){^c~;<GNmQ!F>>zugAe@47f=hd_I7d|wZnWh^-1|V|JJXN
z^XOgCCn);+^y`h0g%A8hPti9%yY(w&tWL&g_m}=Czw?~NJE1+=|A@X@iM|!wr|)^s
zr8j0yV{|g+b>nq5ZuFb?S!6p;j>`B>O+N>n4`-)!^2uH3C^_<zYbRZ_^UU9#?`WK8
zFTMoXuj{{6zj5r9fxS2Fkwdr06PwkvXfQt;wE+f-o2S0xe+%XJ#c-7$rr0QFx&NX5
zchUQGRC&8G*kTiVH{L6L_9wq1YJZeXjX(Vuevim+2dyjIudaW;eQFo+`4L`w;yvPN
z8{VRQi7fbx*5=WB^oYv2{+z9<qf}(+N|uEvTJwM7-es~jG{&s6Y8^WHjE?HNzsLQ&
zDB6dg?79GbmPEUa8GTPWPv3fr#QsuwhMdjCR&z35Ud{j3iZMDkM9vnw#R=Z4;Pn`N
zN4|629Pb+^+8>*s?L+qp@%wFb@E-Ykqw#d|65ZajpG}N)dmbO2Azr?rm&t6~#~LaI
z2io$1c4|Gby9+IIeW4L-{ZP&D9vf7bW9rHG`rP=EzNf>sIbvWYd==wS4s3wVeFeKl
z*~fA3=rX!xs}^{ew39vXbAoL@TcZshC(J)a-~5B*y57!V%859gc2=w4AtYeQkmUGH
z!;S^W_sc1IW6ygI(ZxQt*(*-ae724D#ir0p-K%_=PMSL(%cdjP<)UYEop8@{X+1X#
zPq*l!5x=TFj!%<m;z&G@Wf<9(ERaj)!-8+g?wD{a!cGO{h<{d-nO|m>52G<6Uu{FK
zKiH=ndq(dvw(^HH&JUXN#4782`p%4AGWI0;ztEpg)D@$h%tuevjIqGjtwwl1Qys50
z<iW<`n=V2>aoGj7ehQmE##<dceQ2(C;MMDLOnI0<kH_)%<O(?wzjfR{Qb4?x5<BjH
zS<70fFWa(9gROiPU*FJ8MmoBUD&RT#4y+12-*<qlbo|{(dUYM;yJ!tLU>lyud=0-l
zjMI9o_+{@jBh_c)tjVUzOK947IwnUhI(-CRWhW+MWsli-^IV3x^p4k)OOxxoscg_0
ztwQ@yIrsiXwU?MfpIL0N#Ir4?(33f<8Mov>IAD(b`X3-q;{$Yx*U88DDSs%-K85($
zOmpuV!^idgd$9ZTzOJ|RM^9t-l6&c-A^$r&3<l`af*jk-m5C1bdVV2ZH{p4_wg|uG
zIX>v#(v|o#PqgnJ?PEvzI(nC|^H;jQ#@DZie?E|t4|Lb3(J4NQx4wt1OOEg-`rYsP
z@OCwdvD%a4E_22C##FS~`F8U6TrM`uSC<@fdKf1rzYoDbe#_(g201EsmJ`J52T9^s
z4CVWrP0=FvejliIAXi2<n(ny~s4kmbCCe`37u=G(&Ruj;PQ%k}d_MDoH95Wy{p_4?
zynYq$WQyo(=Ra#3?B0vwYP|l@d!3@<YC0SA+@)q%OMi6Mh~D4E=c1wd;o&B~IKw}p
zeU1#|jn?JAk-s}%oNK=5Zv?cZ%SV1xpPrWQ_RDEI#M*B8(zxgKt<Bca_d6Ha_x_D+
zzK@>x@j}nGr0?iAD*f=2Yma{JH9lia?0)plHog0woE3}Y%Y@F8Ia+JYBTE<d+j>Jh
z-s0zfnCG?{5Dh`8&<i)&g>7GU9H39H-;#0hMLH>O?7#3_16?s!dAz(~yq)@P{yiCQ
zqu+_;_|x;ou8iuU=-be5(4#1!Uxc^O8aw)ZMu+R>Hb!G}9x+Gs{qJMO9!%%&&{gz3
z`9SB(&hcF`%_T>(NKXCU_Du9{x3+O=86*16BzWc-Hi~MM)?4U!8?1FKwpsmvKBUu*
zV*4_iq+4v>`Evb2{JS2pK`wCrQT>;Beuigvpp5Jht#h00(Vwky9*{Taypr)=TFoby
ziEBQV1AW6@(R+{RyNrcwQUuRWMzYXZ^v&QGDEhr>wyXRK|G6BOvC(SsjVD7ZvP65z
zyT}#&-G}-57V!Lw#_8o4{T{Ku=e~45dVg2nG1_N}*10G2zaH60-=*g6+f;pAj~{#_
z4p2!{r2&6xC02S`>rGcD(Z#D|iuT`k<MCI|UuK7B4ZfXykD7BPUyE|(IXZceZnkcO
z%bs047A`sVrN7Se^q)bu-@u~@^xjKuh>7=4T(^VKVs;Ih(^lU`u%R|ts`BH?>YrM$
zuDRUe{-rO}lYPWkKfDYg*HAGuTu%B*Orcy8`8z#drRT^#>2~6YeX?w(?^Q5mj@&Sb
zY@-}KzaJk@cb7}!^0oZY*&HpM?=wg5Z|VN9d_?x2PVrZHVf{hzzt?lS$+JhUIl?F1
z@5rw%qcn6>6g`IiV6$`d{F(fA&NI<(%AWg<9CWdAih7%VF5&+p{GY&k)e(H{>*Sm<
z%G@VVb~Ho(kLHUXx-2`GJfFGOj9pvsuP$oI@nW4`?{<F#*|ySe9(wo`ZF0Wb@%O3x
z!TorDu~3tLbP`YIUePxZ2G&S~?e8Z-gZhb3p-m#RN+iPkVTq7+N5XF@6QS0BiBRQH
zIqzA%Q&N5LDqJl^Uq$5Y=lDWC>%gb*m7Tq_@TWX#TlWi>Cp+7<rORUGzP-q^EBLXP
zXn}Hxr)Ur1XEF5?TlJ#P4RqD_r2Ki5F8S13^xPt{<9fB7I7+bJLu2I_`uc&snmNxw
zz9!`UXQbMqH;n7XzY@meD;0*&oj#N3yQ*h@<cpn_(&=Kk#d#~V9PL1_;O`7t<lbw1
z=;Z}un4PTKlw;`GSg#m!e^+=pi(WU16a1~)%eMS%Iv$6T>v8i;r1#fH!<FIk-#~e)
zuNdZQ^?Jes$78Oa9Yp`bj6;_7jw8sEVlsW0bF4A`b6jGco-^&4I2M}A=lR2ax{UUT
z(wO6W&s3t{-QA5Hvpyns3**0~z55LQwUVB;%7b`pbeKGRcoZG4&~`sLzvEk3A{nR2
z|0vq0f0M5z$o@1QYSC@a9{6n~Z`S3rmHBWv`gw&+rPLUW$ln>J(fL!Ki#ybvj4?7V
zCD#}Dc#Piipx^1@bK?)B+mFaGi{0Njfbg^Fq@3|Rzdk3Pe<TZ9dY-&L@Cogs&&bbW
za6bO~@r~Faa-oq2-7`k_U2^|gK2DZ8Z1of0erPb9z{Bt2<H}UNL7vg%I>QH7ZHJl0
zihj@k$JykXDa}6mr<d#O_z9m|ZOjGeh`#S??||RvTf5~Ol5_Z{KZ&u+as@t1M)~20
zdwik@dsM51?|0?J_vvVW{B=ToLymQ)@s9tEj+x2W{i^F5__;+lf2hH4TR)=cdz{}~
z@4u$q*fos3mi;b~c`F)%_YR(qzAKrGzQRlNn~;fD;L}YwZoD^*8~xTe`c2i^-{lMA
zMc-MsMHO66(q0T5*C+bDT=dOnEfnnyA2Lq#JFmwaN1A6lI~>KwZoDtpPcFLZ=lC|e
z<u`9#{W5tj+W-6*&%@~SHXiHpyRq~atx?OPUi5f=Ghae$=<|Q{^$MMq<fn(kUfKD4
zV<Fvr=brX$blpwAXFV5ZlOAXfTC_vk9<h#(@9}Z3<01X_uYx<vtV5QF1(eztm-UNs
zO7#7G^!vpOY<>gf!*}%Cip=zv&Gkk+XJ*T~`iyd2fc!hy_cwjs&?oNR1;;J!ozthD
z>+2WgaxzDI5XDgRHzkVEW%Tz4c4+_U68#!Cjk%lGhdUp@6m$lSuLf@`kguftRmd7C
zw_KKkzdm9QDV4l&w_089?fO^3-;+;-6ljR{()tW9E+14P!~61L1M&QsF?;hHzW&Cy
z^gYiQGsrg?evO9VJ>}$>nyoc^;p3S`^w!u~$}yGeC;4lRxA}oFpKzY0x%1BE`NA5e
zk9^Qi?Kg=34S{)ITHAhw50p`RCH>QtA+Kkze<TNX-~-}6dRJX|4u2Xa9}RQVwzikt
z+#N4*d>TJ@65qnt)V8ppB6-S+9XcugF55Q~51->>g7xf{<oD1;j>xIvD60MTqFd-6
zR0&<7=k@gbzhnGouYQ{m9_G`<y;J1>QF2#bHYQI_d><GrS8Cfj%l-kK#Zg6l2a<0s
z9#i7&yHCVKOF5vanyoqi$LrbQd}9P3r`J7W<!SQ0M(@}0IU6;d$j{7|{TngbU;J-`
zv2P|q{Ix{b`bHwGtda=n-b;l3A1A{8)`@Tk-i{mMI6DzGZc2pLPbb3C2gs9~O<oWi
z#o=6G`K=IJ7L;T1!rG_cST?m_W;T3AoR(tK^6bwR54MpH+QT2b{zHES>HT}WlwL1x
zHqg-~F|v){?UZkK!-0c*_As9J%bBi!pzHao)sFMky=3W0XXgjVlk9zQjNC{+o5(f2
z8~?7W_TfWA%y}mvr|b6womQofO_Pi<P5jeuo4M*Qx?Q!vT3LLrM=var15mTYWI{if
zZ@s=ZCcCEFtAk+{{%^H2ZWHsHr;1|@ei37nG#$0^p=DjzQQzYDy2Q_Bk#{8<H|{An
z^WWo+i^=e>_9^a<CC7Nj2yZ))Z_x}sHrqPf+>g&B>vXzw{_h-^_$|GVVTFDj$(2qF
zd~WRR{Nqjfoyw2z7(aUFdTOc~Y`Iw8EOvLnuRZE4$LKfSZ_#IAGJj47*Nk1sn4Qrw
z=YN}{2)(2?U(7tEdf=%!->E}475Qpec|91vn3|NGLs2q3i!b&+^BCXd0~ekVr$xxm
zUv}i>6V=(UF<s)ZJ2_XGH!Z#@9Z25q9(j_#o=MJS(ceLP@kjPNuNF8fpNOw-+-s-L
zbH~`_i2bI6baIGIchD<-EAppf<JE*?&0+jk<b^qO6RoN56$97l;$HHUAy=*ye2)%h
zdOrHTDYY@9_edG(rx0I!pI;7e|B>x@H2(N)YM?FhhkFgz@l!rqZ=d@=$vqe0mN7PY
z{v+c>_2d2QF}x;Ts?t*x7*L5Uj&0t>KOOvu&uM&p2>JKoD=lBS#UC=FO`g5QmnY$S
z?hLkA3?J}&|ElB|n7Ed1%<+h4GvM#_{d95MUM3yw`GK!HZgPy?MMrzXrEl`#TWtS(
z@)|e#{qS|?6|}8!FFX4F20TK$_5T1p^9!D@XuAx<j1$#Bjg9&3DcF0|-j_L_-%DOP
z=yF(oW25MIw9$8bv#uuZ??u0(*n+Rj`bB?hpc#IrZiGjUH|a52+eY<P^sQbyGDY7I
zJ+z44*)qoh`di8l+M{>5pX!&6JxXnZ4d@r=cd^Tv-Rdzsti#WK^DNVEIz9DR249x&
zUAoK0&u{8Cg754(jc;;H!rKZmT_Rib+qTpEEPCHILHl~g%5->C-{>9tDEDuoG8gcy
z-y81LrGr#_Fsm`6TpuTM^e*fsnuN;znvA{A(?_mLYBl2?r@P1Bq`&;~VJ6RIWY1^d
z%d7mbgf&_rYv|naW)Ajvn5-$qZ3@`+ueA=m{T9}Kek&0Qq%zl&d|T{ItpEpV@RR!T
z+Gl*Xx0;7<y)%!F=E`~eC*x!_kvx&Vr<hFQ2QBFJ1M$>=z8X2AlOM}tHQ^5an!S$P
zZ`w?r=p=7^ZvV-+>H3MqL1J$R{d~!1zhXbf1MaskWj=MmzgH7}W0D9d@5LV(Q<o>#
zM`~Wrtr*I$2Ff?R*ryxa#mSH7SruXZb8;5FbbSG@#nlR>>CCekYSHB<d@_M&yc~VU
zT3EkonY)CL-r6oTEV_@M-HqQn3I9VP5pvR5A`M+;Q2%E!XI_3wo>&h)KY*{0Wh8p0
zFJ0g#(@4C}l*{J%PGmYh2dl}+bRuTHj@D)7FWu4{P5EbYwO4z3CeI^sOtsPMGggd@
zqaS)NJ%Jt_r=U&Wz$UtXp}%~&E)j}WN`&uUO@tBf_G@^%%NqA#csm*1eh+VJ!`n5p
z;OvG(ID9M-9(q8o$j#3R$ooa*uEO-lM}BlYkxUzNsfV-4-SnKKU$qkQGo3A@vs<mL
zbMRS!49|W6x5?Q6Pc6R{7vgH$dbwyD|JW(F?PmW2eB%)N?^Q!+uf36dS6Oe(=T{T$
z-}DqWboLJ2bQ~_%4&!^|n#PU|>(X;0b#fce$Hf`@&UU{Xo&4q5ucqR620hbp<yqom
zHk_E_xB!2P^hN7X^qz0N_L6LH1+P`;x9b4UcO@r3snjI7HhjVH@n(Fh1$&UOHJ|>b
ziyVZv|IE7!6(esxd{pQz#!*}RM>WF{?O(Yc<GW29zcxp&ulXE4Z%$HUvDsVG=oZz%
zZ(ID7|5iWO+4RXg3eK|qRy>{S;5mG4HUC`W-({?l`bBuVdK0{)+g)@U{cig7R(?;P
zH;i@Nm|e-(mERnv%U{IB4SenCEr;{-m)ntv{T|_K`RlOXt8%}%$j0U`<tJxJHh;<a
zvwR~fzfUi3->262O)W0=3ObI_*8E*{V`I6K546PF<tg&z5^~~e<zYEU4l9BR9FxD;
zzKZs<ThzRGe-dp+!(F%L579T7&GxHV4ru$%z1ha++kesB&M|z(_-E;*E;)WB(>U!F
z>EbT3j^IPldz4aiFu^>lj*BbvW;@8o_V8=-j9nv^*Cp#I#SXSIXHj!s-Jn)nsWzW0
zPmh+{`@x&e_~^vHzQprhHrA&O+t#g4$MVC}N-(v898*OOp@XFjtou5uW%$TkcDw&`
z^Y?;heP9NiUvpl@`9}WGbcVLY?9GNJ%sY2A-(8CbbNq+TmD}{$Cx+;()hWJvTI`$=
zb3dr-F6e(nO#Nc)8?f;fU-&}~L!E!;kH5hd?E}%J>*DEGcGveh8huR+dH%=q$+$iA
zoI1_8El-%2t<IVA7rH9vxSI~=9g)-6<MwWH@vm~|8v20l(&F<!{C~iI=T1>$u){<A
zdMZ0*;NzS5;E6?IfG-?bAeLOOneTb$CD^GbnLpmhSLi=2d3&>4c{F*uykWj*Uw^*2
zJJS6KbJgb8Z+Y$+;}xg739gUn7vb$e`Z@8vdf}}4fK0o7lsi!~Z7YvkuhU`lJx+Sp
z9rQ0wuhDNhE|c#k6dmt!{WTptjZ)~dnx3Ays^+J+)#z>%{f&?!C=>nH{u!}HYh%ZL
zi~Ac_@zDNqoL%Dnu3J1zzY`DRA4r71A5MhQk0nCqY>6<tU?Q}z7J64+F4e$#r(GgE
zJ30~aok)bVV(X13>7<xkUO|qpVZM6euN7T<VO{q%|DKy1*9+<T2|BJZlAU|dWoI(B
z5D#y%{|oYR4!-gbKS&|A;riX#`BDir-uuRH3Nt#`$LKDv_LP_V7<-_cG{l^Iedch!
z=)4`@zpt=y;9X~UyGhJHcrV*!V(T)n;yu1xU+tG9$99DSXo6=lwxsjQa#(Trnm0L@
z40~R!@e*6(;l(%ja83T-S{$_#C(YQRY%&JTdjdYDmJeZkfU#Nr#=Dy2e#h*0+1E$|
z^D@W>ndPxOZ2bZMq?`Bple@PVM<>YBh&&H}MK8{0#reCKEJ9DkKbIF`=I+SPK2bB_
zwL?>Up^@$7ib2Wwwb*F-9BYno>^?pjCl0ypKQuWX=Mwuxu3N{hOoac+C&IpxiIAgg
zBAlw62v0?LTQ3p5X{DBew|54S1Kw7Gw=sBIg$y^ImJ9y}Te6Ciyz0H@*s_3J?L3wX
z{%KE77xkZlS^VJy{q!^Eoj{%nu%Id&<{KZ;VTw^=b)Fh)iGFL<PTSN0JNfJeId}zI
ztg@G|-kMGO<J$hGZ$`e<l8%b;fg`=-i9zZud1)W{mh}^7-Q=RW^w_{W>@({#KBBF!
zK0ne!+leq}DxaER&jy{CDZkIQUYR2%#Ch)paDSn=L~-ZUwV&m)o#^rwyZp^wUAkE3
zw7}z=+6sH-IqS6-jr+11h+Uho%`5D*a6mGK#?Ac{p5uJBBwh}qU)Z!wUw*8;p8M5}
zUDh$eiRxs04xg9A$&zp6Uh^lR&!+QT#}Dz{%rS<i_x0aQ2W|N9F|kqpQ#gRP*U6s;
z{n|thZmrI8Z>M=n(otJ_|9Z0XSzSW*d9Z+tj~jcd{MSzmt)|QE_*&f?&L!}|*O!oK
zs2KdByc{AHYUYP6ImJ;HIF?B)JW4Kc`%4;m`(C~#4wl?Vgkoav`c3B#svD~rzj5+D
z$1Z&QK(^>xizX}K%|3bIAphDgHssBeyTrhDJZ-U0vx!fkjcADLs4iQx6JPp1cR)-n
zGX`09(AoQJ+;b$G(Pba{ZG~3K39GanW%m`c<*NC7WGT$yTW@c*Z?S{FnEwSd+`JEK
zFTKVyt9?_!4`0XMD!zD!j~>uI_d8f^zVpUR*+<>g(LO<YxReBU*`W+O5A96n?;8JY
z^U_&X)U~p6I>=O?uJoTxx9{s;wY%KnnJb@@osClRua(X#vq436s)*m-%fuF5I^t<M
zK5}kIu0M*=MJjxsJ&K=`;^370;JEmhJut^iXZZR#zId5G{05(X=l7^78hf)#h`w#F
zc17-V|3~!8FMI~Yuff6#;?(oecoR`YbjP^w9CmHaHG9OY<A6i(=7>6xJ<jjO4;{XS
zengM&z%PCNN`~LaQV``}hx;dK<7?AeTR+y81FN!S1-?<5uIMa^p?CYkw<ThS%+)sH
zo$W8M*Si1niD=z>H~Ej4vpt`A2S4Tc$kCNzf_;}R7WWJ33bl8Aa<e$bZ(+R7Isre=
z>hq&mMw!2tJC2J%GEQ?ok=)gsC+Kyy{>#xJbOA+sD^Jov=7Vwu*?)Gu{WSmmS>C0$
zQ&-g!DEe(kw8!7?yqI>szB!+Uw>PfDL(QM#;p~}sr~rG{!Q0{R_9nb73~x(hON3s}
zB*GWPVF<j<3~y_}+i%Bszjq`NrirgL+1a!xdP6L~i;oYjmm86@i+l8XbR4};pbz?v
z;$pA1{Cw-jPU@Z(u%Zk<Hcsl>34a4f-oKFWw|Em_XjV05F~0mhA8bOlcI@f7rajq%
zuMg;FjWvY7e#tiIRq^%nXR!4-ISSU@xniwmytnRh&seu#QETz}eD&4xjm38xd90&-
zzz%As=5V*NvCNl?Ja2lg*OEk7FxWb|MI!XG_Dfqs?rG&2bJX<wQw8ucJP|q$ppUtU
zFnEjHvO5tjA0oqvM0omC!tZ&laStZ^?FVb^`_vW@Z{(@^xw?Q1=f!FBifmj;-Zyt4
z{AI>ZwD#{xf8h)E>ni_=;Uw+t@jQsmo@feV@NjnnHAhW)dJmS?l~d>^(#eWZ)(m6i
zmC^1GrhB^kz5<=ef%i;GgtEC3VPr<UrAvgXDdFxv@o+V&zwSwdnb{NJMrk~=b={u!
zF}}6Vl|%o|W#0eE`&rmCuNW%G_vm1f^9Jtyt-V}E_TK3Eoe95VOoXNx)Dth^^L6W`
zcjSP2a)~(GFhWk9#!u$kgB4TJn(KG*IdFp*S|MMp6!&Z8J9O`2dYG%`9VcIk!&D#B
zJ-N&Bu|dwu)Pi}&PE#I+(BISb#XwVbq^oD>aaLTM@tF?h>@it=V&0rH<nx*MnJtga
zQTu#rZM=XE7xFd74)b8KKKs8>Kb!Z_uhe+v?w7znUpZQdeG21)J?lC)C@mk=k+<1q
zKRZ<&V4R_NMh_0O?&%@^_}bvMd=N!@HO;?J7rQr6pBw0`IetLf@j03ON{n@Hf|_co
zIj4Ib-xr-%!Bdy7@vPs4L3rgKYs~qLaqHr37x{leQ^iafx_+*`eG$I&llUlO+(C43
z-z>O5rwhpbpK%}KC)xN-dcM+^oI~(iF(&62^MG;ky+<GPdv6IfR(_b26aQIY&lB{L
zLBI5LnTAg8kq7w7Sa_R@uPuPL3-34oV|3-Yctf&v7w=@KW$vGr!?>;dW+%Pv;os=6
z^B1<$!xlP_LsOx(sGsW(x3Q=DxA(b54=yz)JI$h}>STz%<v35L<9o3&c}J6{B%OXY
z4Q|fH!+i2DkiUnqh4zdi;KE4bdG?)2#&B;ve~kWyT3&t_R@(nX9nG`CJTbc4O81Wp
zg=?Q1qXW6=cU5O~5FHO;)A60mQO)|G3cl#<(>KIPWj52cp1;=CKNmkrrT-}B&(pDF
z87g`%+EbWg+$Q{I6`tnPQ%}Cx8b6KkHAOsR*dPzlM-F<6e)~M>D4U<OhC3yH92bZ5
zvh;}OPQ%1==DjR0-gN!Dyov7r4aWWoQ!c@G_q(1DkB7-}fSu3t?W^*}1@U<XZWwo`
zF-N*LlubIDa}XQsa^Aw+jSi8?voGw2mAm=5_U-z-vJo%%tht6y(RuX#Bjs2aPS-Ed
zUB7<jZG(^6<fy8it{`Sg>-Uyi+!7zX`R^3?w}@Yp`RC32-nh{_+SIG$F}8dImE^-O
z(a(bm#lsSDjvhw2<&4|roiv^8d&n(vc6IY*a6IZd+gZN$qdJbfr^xvb88_2I^Y8U@
z{lhV~qsJv={Aw>7IUj`{MY~+rppV@L<zVz8{r};ajP$vj?*6#Kp3b)#zoB`4r^C;T
zpXz!%EIAtwf5F<E2jU^$`gr(fOFZ;E6c0Pj#r-Y3czE>gM92ekQ^MOx)<OH=&Rrd?
zcLyfI`DxZahZA8sY%83eUu9;qr|GAFV@>?>oe}&$<v4PWS6__5=V&(9R-*-stWM6d
z)@jeM>pl2^x7*>|mUD^l04xvt5}`*nHhI^2kS&@O<rl{OITwF=ikyY%S3IY<sxE+M
z<$l)x$3%GBamMe7u=sK!)OiK&Js%ILo{5LGIpU#r-gtOR{qV~~STZLO<}HGw3#{Q#
z9=O=cIMt0;<ai>~TWG%=-tL9B*{UVNi8AoJf_>Ji<gN{?8YjY>_S(DBMW00IpznO^
z{!|mJt*2Yt8gsC<<lnI8CcX;aEnnp3yYGqn?r@9UFN?94D#G_tZ1SS`D#ov$W&2G0
zCzV(^>HX@NL`aju+D`k?m-*WVVxW=O{0KH_TQOffP{n%UW!TA9{qMI2%(i=`#KXCc
z@i4e~JbYh29tJgtha9nZSf;L;kTnsi!P{GD6X6KF-S>Cg-y4jF=iu$A?B3g!mMiPa
zn_u{Tik+|QN`(4%;Pj(xmzA&Pl`9L%HS7`XC%?lc?`07W4@diMiO`+QbM|^iyv;Z>
z65$NpywfWY27G3WJ8|E2#KU28UCNUP&sR)@@^#d1@kFR8c9yJA?=Q8+lVeJWwUhkj
z&$eQhzr0z-8mET3t-hF{gGL|o?OJ^QZNAT!*3)H)x^id(xzD{Gt*kpednsm*WT<`B
zNn(8}oy)&FzY&vDtofXeoN25%Vtt+(3S|(-JH+aj`dl2#M~BFfU&w`>t>2nj8=EHw
z`Sa)D=X6lNr250WUp0XjpUMRT_^A1!?{@xm{x*Bh{LJ3lC*;9TVE1Tmx_EE+(;pA|
zt?MiP^?m+Jxp#y(6)Pvj&0YBY8;`}&Ea%VRsrlDr&}YpcabcW|9kdy@Su?p0AD!QU
z#g)lki41aCN%}dU)``B8>)B7<qldT1n2p?NjGJ*dy$`nTF<!Lq(uHr9ptCl{98IRo
z?}^p&a!qmf^TUsv>KgWb><Rfd13XG^eU)17;-|042N`c9LT|BL*|E%hZ1<Qv@43I&
z>eY_&nfzOBj{eKFZzbDK_3j?)q}|E+`d4;uvW8w!@fB)K{XW+A-4;He?~r}=jnHVh
z_11KAPJp4K;1S)O8)#3rm-SRny6DC)*lqO?HXFmXWW9g3d@{_w&sWKK`#%`4hORT=
zsRy2}EjGqHIotIncKeIHx4x^!;Cs1)HSO!Lf<At3&Tg&M&~#M3J)O|k<Suy9exeHd
zqi5-?UOBbbo9x9VFR{b4s>wMZLmjc(5E-L{=L@q<wz%im^9B5uWs|z>w12MJWEs1!
zW*0iRwB9@$;m>CM=<Kbda_&j?KV@&`7<u=q2X?Dr50KUM^Jmma=lJi>{QIhBZbWrA
zAHC#Ud!G~Zao98Zw=`ZJ&!;xl2WaDQ;~M{%wu|i4-<)r;LG^9yVUEEE>CLl?*y6-4
za%`8Q^(&6Q{2O38`$fOC=rB${_Gvl7Tx?Odf0qzG6Av}{Dc!U$#g_`gc>Wr1L6_vu
z&ZhNd(F1$*HO|gW$>WY(={Q^bp=al+1C4**LiMhG+h)kEo+~h2jpqC<`umst&ztMD
zHso?V{y96G=G#BY7w6fSexi5y?a481KmO^VHM!>PwWmWLbIBR~jgHgKr)nFAa_N&-
zpL5!Na9z-|o9MR<{T94z5ALFQFYsq$H)op{>G7ML@sQzYJY+tO_QXTdy0~vs;-Lq;
zy#;Uc!P~tlV2^d~;g=F&Vr4nBp0!T2zLkfYk4%KNd*o|-r;Q(Ek4);Vr|2NR{9A=T
z*0&eZSX}VG@A_(^>xW0no#pYlJK=B6!$Wx4=5c)9W3BTC%o3xS&n7}*zi)Yr{a#i%
z<Xw6w4X2A*KNk=ixzth*lSy13xNiM<Mb3tk`QU6OxYqj|Y}lFzugr~ykBi5{8--v`
z?sympZ>wZiUzCW4QqAJw=$G*@Y;D}%xQK@jAB2ZwIt(YzS)Y|SAXmWKw1e!mHYIO4
zxci##i7NPpw<@ZY2(LDT>G0#vF5Z9kfx$3k!l*=eAKtb#PDePH0^T-*JxlPlFonF4
zTP;^Z4J;4T?!kXM%JXHdU!IgBFDAmruyrumSFcwC*w5W&{26a0LSAzggnw@jfPv%G
z74xj4*CoO%{VvXciB%He5qSFqysdQ~{{D)G|KRP(_H5eJK1RKG*!*!kyasP?cz>|^
ziA3m@0sg_;mhkq;KjWeH_w4Q7Qx7M?mqimHz`*8Y%QD_x#wK&zu@25C*4g>KobqFK
zv6-G6Y@gBbzW<4XhhP{#xqedZMz{6n$HOyUtLvu6L;4NzaEhN@<a1B)uifzdaXkD*
zw^_+J=6!n+!_-{gB>V<C5ss<l{x$DQo3+92QR=6svZzt`&+U(3P(ykAW4<AW?|EBI
zT}Do<D4sr)x3rIOo%3Zpy&`t%@Wp{@898+M6!_`5;!AmKketLfTXgoGpsCzi-@Z#-
z_Icmf#{BCIYlPS3;kT@P*)81%a+ULveDUdLVeB(<7@w?P#F|L|lX6ARPp#7(=d=`S
zZSdmyrS|;S@v~3llUDN1XYyylo&z69(FcFM#k=Fvczg<9<;BW&vGVG8u`*fi!1JA{
zYFfwIc%K0~1`U>fqi4Fx8y&2@jD4Y@cQ<%kg_pCHVR$9|E6A64UP5OnI;rLQ;a47-
zKNG&r!2dM4Y#8~B*OfkA>2H03uLy73wx!!f_Rp)sM!NX4I30`c%{kNyPq2YFd?k&V
zKBc^UxB35NOL2Mi7e2~<CH5r3Dt0@k&zpSz(Q5F7UP_SjRu}oJ5DdF%E&Q9h_@6`=
zelLBclVgotBp-~Y=LGwese)(xztmbS)XlyJKBI4nW_}F|zfoUJAj2qjqq~#Fj^2@f
zO@<u$&i#TuhR{2mMSCjQXVD8;YZ@ars_ML!v0gUDKeO>CZ#SApkFF1Nhp)}(fc$Uq
zr<MHatYZywysv#<D|krnTiCM*AG*p8W3?yf{<FfK!~Zz`Q|nD}|F*HTcX&tseUH6r
zi<<`gx4C@wsk!BX?sT5JE6m3K`@`sotv{WEmhj)z^2AzvueTSrQNCEmXLr!oQDdFt
z>!;L0$CB5)f9+E19Wbx!$IjSaItS0;!3S6A=DPiuU&(NZ?ApJ8w^I+p9R1(k1BZ88
z8|_WT>u(Op&FDAnleemYHuJYF$?tz&*0170dUHSe-G0O6h)wH|B?p;`(8I&6<-GRr
znI2as!5{YuY0pCLyX(RXah8H^8kL~yJjUaz-+T7{UhsRo^~o&!EY?o;@nr75LC#sJ
z2K`1|!Ong3n@v|s&_U-9^i#))n`qBw0N*}If5*)IUsL{L-0#i5+4Z8~Y{$Ok@Vt($
zX3^nWbnpedO+^pW9S4v(3mKz#(bvh>XqoxGh3QNAApQKM-(G#9-zQ}C>~BB8>MQJb
z*|-<!?R=LIJa>JK`4-UU<aj99HXbgwu$TM^>=_#mxi`i`et7$fd{_|P?tr&_{wL2B
zR;Rw12qiz3=izN<*whY9Y662l@@+(O-`*sw7h%Ny2k4-(JW-F%8{vswGvuQ)_}b^V
zb>(^Z1#d5l#TQ0cAASH|G9|*Ig6iT%Fn2Xvfg|U)S?@edhSdCGhxO~Ncqn-y9&)|w
zxqHOe?S%h37z;Cd$3oL@VqxV&@$mZd`c#bjJ{F#qj)ymkkqO?eg10T;?XGO`@GlJh
zqai#R7!MaV#eF9q5ATuj+~;u7TJGAKL};`s;cqP_!p~pAqNcvdE=$HdF#Y*NSX;t+
zuzVuiuAK-k!o&73sbnAL@OC`Btpaar!Q1NacHxSIzpG*G{G)y;)OF9Q-D-%%Zge6a
z^bwzJ%kZh}^6y3K?$a<0-Y%ho?s%#Ei97^L|9Ho}hV<4y5q=nB-967b8{Q5_^(NY5
z$m+d%7WfHA7o;-(@9{9Rm3IY;W8vbYSa^APEG&7~8a5X94I}KmLpSLY;dmOg%DrkP
z{imObhu<#7!<C2F=q2A&)k%c>WE=Oj{SCPH!5ujMkeE&{*QQfnrxk1Ho(K7;wwK(S
z4Hs|XVeXN57`H|(18+}_i-#5PHp7N^*avTq!P`alRSxj4FYi}(6}QeM-xRWT8fq`$
zn?x8r(_X>+L>Q)Cd2S=Q_{s+QS^v1a@{(Ly5GLh-N15gP$Lx##4}QqedFXO>dGTFN
z&I@v(IxXcZauvLda(uIi);LqdtyrrzRIChu-@QH8$r`yyGTvVPP!5w1FTJI2S$XR<
znBx3gb-7Z0__B(cvIyDpdPn$-z2N-nN!QihluPtk)mScUA!ndMVtiaHxul&u+{Sv{
zbz1G!#r@vSFqI$NmMi+<A#)!wiJ$Z^<BbvaW#x@L<JAR|)zWw#FUR!3f2(ow+9;Si
z7>0Q++AGc~-^Yx*roL|&K2Q(f_xP*oh*Gfg1v${M$h+iiDDQPu?;GcioYa*bo;KI-
z<~>9|?>dg?Cr9_tKc)tNCFMM`oLoP@Ax{^_Uw*Z24tjeOzNDq^`>i)q%HMaB=U@9@
z@OH*8Vi?|zR%;eIkO({YbBj6NOU+J%dzZptemi7~_ebpd>&Zk&x~6XZ)3+A)SSLIn
zf3ov7`n{)Um(Yd%DpvM>@gsSwwReB`FFr`_{tAANwdNi|{=RI2vh=58GzI_PeXg#e
zo3i@Xqo>2#50LTWVRTD(nT-+sR%?-Crs3B4@VGVI_M-O!?yr-V?;`(7^8ZKvie;<;
z;cdmIwei8xe5wt5Zf4iP^q-SG&v($?%$n{^dkBTu75`md?-GLRsOE{jQO!b@eO1&7
z=u}<L8Y4e{i17API{y#P8TrJ+<T@~(e@thiC43g<bX|+@b#P;Yy38^6R%_*LFl`fC
zxOevoHIMeQyU20C+UqE}PN+3b+Y9>v{$BO`b$$d3-qcp)l>W!@eAv7&Cfy!&!ETto
zm%cnd@(}+)^R%}$SM>KMx@{)c7VEd|^k?oToqu5N=<h33#QWK{;<G7T(A)D3_;P*U
zCVUL<Kg4$pxJUj(dHUf`?dc{}V7p>sCO12jgKLe%M|(UNYp)osK9O(F<CA3DYn~yF
zRVRzb0dTShc~DoBp_jU+H<`6J=t{Tjk&aFuX~D+$iDGXEJumErPxfy4HJ@5chh(_#
zh$9{3po18SzQ>Ddk0kdqx1$?2eH>r?jMK++xt4f-y_%N}I@4F>gKS3Esp+DWXG)s8
zB42pk{4dHaT|bu>;O*Q9XYDsGkee3`j)zH8;~~?6c<8z!Zv8H2?uJ#z;^DOXxZghH
zTmQzx5ZKf#pS7So`7GR-YF)CRF+7M{Bf^=E>eT`;rtw<&QO?`|58j;Y8=d8TcXG_{
zOsqdsi@7lyV&Q%`nR#3+EE*9DBiqG7-xje@HYpZnjf#c)PRIPM_L#pZ9SbM!#6tcw
z@INhF_%9a5{T2&FkHo^Iv9XZ0Ys~K~+=naQua1S+PR7Ed`{Q9?PV3nM@h}$F-h#Dt
zv&KV8c$+^xtTk4F$K-zP_v=%!LOi@UCLYqACL_E_Iah4}Z*RlfT<~`5>+(o$7*xn!
zcQN(GD~a%7Memu(o*U+V*jrscSpQLKf$`R$Q(*SDiBJRHcA|^ZWRAWg>HMr5D9-MD
z#P^zr@5bWjRr{w|eXk++boXqI;qV2eHP-Gn-UXX`S?YxUos0kVaB_-o+TM$Yuk*!2
zH?maxE9P%@#=@XWu`u^oEPSvl7G7Ez3%RGnLhZq^@IyQn(tJtRCt~4~PVw;28MV~y
zc&Kbm`y89rhreZ($3y>D;^7{8_@`SutX~-q_0(L4GP9}m>Kp2!ZdViGnTN!5dcOXE
zcuOIEuWN_LnP6ugSeohz8{PCC_;5VThqqPW?KOVY1kY>nzEPjO@U|0wn(LW~{3-e_
zucUe2dDi+C-WG(nCE;ybzS9xjJ^^oIbTb;>UVF^C=6U*hN?uMQUT(<c7p(se+P6Q%
z$LR7kvHXm<-T1OtC{4fOxqNMFr7z_1L2@Y?ERLgZX_kr2D`GUS_{box?h;qu>wB-f
zu;eW@b{YAxn7sO&byQxtDUV$K3><$Ne&$yDqObG8{DR)+Re-Uc8^os5<d{?qVICXz
zZe>5r*f(0k4>5n(99`QQH_4b{zi&@ijfc~YZ_69s<MU(JNBSnO!Pbost75$JB)BmV
zc8;@7g16@f!-QUNSY9e3-`A2?#+&yoczctNwR}yim$WaAx5g!`5v$qD!_xq{X#7Yx
zQO<e~?^$xfq)PH+fAtG}4DM$wgNKWB@<bc?zJWDgb^B6pz+QNJ<yo>kDgQmHc1T0^
zKh(>Y)rOZ6;R0I9PCL)48MmrGr`i|mY41mFctx$4P(zLv`$gbwUwE6uu4hm99-eQd
z{Znq>Q#Bq?&pj?z;q9%W?DUEp38$iW{}=IdH~vTAasM!OqKDD!IH6B6<~~Y>Lg*fP
zY%IsEq?;=V>qdF<U;3^>uCDqIpu30p!J>ioK^*`7lKn^PgFd61pS8{E&MxL#KySOt
z%AtJWGJl$o!#Mc*v!-0!%-#pPKBZq$3o%?reO2Cj%!=%hm;b$BkK=VU3jY6QlWCRs
z6#Anw9$e?(YfrG(71YG@KXqc)xHS<zr;>3W84sYMcrJ;j!IRY*Gx@@7zOxKwZM3f1
z0;f0e&9%w<^S`d)?>p5i2iW)!Y&)(#J8hh!$h|@9VC_C*9OL(=;FJ5m!Hei`W)I)Z
z*LLxTz1EAKALRLTsK*|@wAFa#`EWBVG3UYUYQ1gnaRZ+=caM3_*(Q4TGl0I2H-t+K
ztRw2P<HvZQo6h2+581oXQ3?K3=ymtSX&3sO{*;<8pZy_vzF3v)^zkU!D@{@ZkoS${
z$-f=?7~b{}51oIrwyQ7y$j1}2S(g;BuOfz`Z>?J3<8)_w?qH9hxma#Owr1i2Z&TfC
z&)$<Jvd4Vye#y{^{DsK=1=$<Z!`<jL*Bux05zoBr+2v%ZLzbnV&>wrh-cn7+_78n(
z56Q8Kc}|kMODprWmxIh5;r$)-EnoU)qCD6pZVeg_CF;e)-vi>IExdgI-gbbu8{lnk
zc)J|lo|PYW%FPGm$Mo{!hxzO;78hr5XZg$O);E05410RFmy^}B8@AgghPN~2%wLE2
zer9nzY|S9wZjXhYOJd>2EwOO*yI8n6ITlvI%=+*k(KZ&^H;aXq9b(}(IMES46oR)G
z;cYs2yZT=1Za6v!-o6cQZyt?>1D>tZ5yp0lg~1bJVGFz+3~$Hk*ASLofJsZA_MG$I
zvc&Cy#Y3NT)^Puaw>!1p&L0oeUng5Nm{2z!vNwj+pTU9t_Fm_DhrQ3b@P7N6Pg@JZ
z+dD<=v66dMSv7g7MEFGe_{ZUKe)89bt?jJYVdc*-d!HKPMR;2k-d_9Ho<O1G?^+)(
z#V=lwzbaZ625p(uxN=WfcsmEik00bcaBI)Q+OhT)&ZM%=>m3hi8pK1RDr$~Wu$El?
zAB_9oA2I(sEarbCs8itWyYRLbynP$qemp1^K7_Zg4UL7?^JD)0O)UKSFngc0#)Y{*
z**|UtZ=Z#?mrGlN(nH}c-t(@c19+P*3p>7rf4IMlZJ)Rc26juv=0Z8+q1sdNP^?5e
zyz+58#J_^m>*L|>2z%jeYgoH|1bzp_{oQQ)${XpQKMmb)pW7ZvB|Ho{<Czoi`R8~j
z`=<TluJD%Zea4uV{wA;R?s$Xw_jundkDvS>-!7=$%r2Lvwzh${`{3<GYpA>Tsr~3Q
zn^?{QOUjDl?4`x5<L#=}WbMSr=hh1H*f#m=OH^EJj{U-#PafFJ7UkvD)UVouEbhJZ
z^J-(%0<9_lzw_w(jQ2CBLoRqG?+$Y8lusVR)7fIyIBePWZ82R>%`D$c73Wc0t`L*y
z#pHZ(`!wqRiTok|N9)_-a>Pwk4ZYk0mWj<l+8a53JrFjIkiX@ZXiZuktr-c2wNL6J
zU+Vw1acYR|9r9Cc`C!Pq>LEOCNN4TwfOyYheg3#y?tIQYo>ljKXs_a7{`eIteHR&0
z$$ORbkrN-5`zzDqpY)a^(&uZi8ZD*E<1gazS-$&Z@;>r%xSHaWcPOV5q4oFh7Ug!G
z_NepCa^V!eCF`NK?qo0Kb9+4M(ofZrZP>T?*S_6@$KM~bcLi^!-L{98A{oO^KPtD;
zYu%!BhllcS%3Wl;L)MIBPEG!fL)qpld2Wcj$A<H|sDI+>8}yj-P4ex`n6*m6cWLtA
z2G6F^FR%U&4fZXBF`{=2Ukrj%^6#`!^5kf9((fnQ{_Lp!FlU`=$vi~++qW~z$2s6R
ze@ICmYs$k{_iwdQlhOY%_*tehTZ{A5dH4(8Iar2o(e0cn{Iw$NcfLlSdahrurB?EM
z8rQA)=+XDtybZqbxv`h=@bUF9?^*Hu5?_tpYu^6cI&J`;SqiIHk$V}RM15WF`_7ze
z%(oi;uYil%URVM%(7BcD>Rz68`Z+$lg+Fdpf9!zgJB_u^{?0D3<-FW(?P#v+<@%go
z-z9_%?6Ao?Z!;|0=-M3L%%{)sYD;$bkv~4!9&ew($dA<kHR!$sEXxfa>27*WIZoRX
zrSyG{4f2Zx?IZJw1G@S*r*-Ld@9oamM-<CT)no5%^S2)O#diMiJiU+5e_~-VT}%v#
z{oC-iVF`PD+2Czf^;^e8cnu#}*u1@Y4(Xe_A)S1r&c@R{c#FOxA2QrH6V(Cue~*qz
z(dm2I^NtY<WW2VI?~{KTedd9;(Yu484LmD0&X#Ahx6~SBiLrOu`pIjjEBfr_XVY3)
z2Y*VYcKoJ;daWZ}ch^2K9zJay4<FTzhm0xh?X8K0dHrKyZzXHT9I<fke@WrvACf}H
z%}Jry;-s*1Vp4dZuVe3|@b}Q9u;j5=C_T(xELls!gz9DBMCC;Ir-JWr%G<lGF4w1s
zg<-!Yg<HQRh4(VW!h~wE@c8&xcx+=VY=*ZJr&^Q2%#{OT;ZU1cC=74Y!P}Sn#X|bN
z+OoooY)PTe{Ym)g9E$weIqdqUbI6|~DI9N<6l!-#3g3@U3g3-O3V)4G@_VbK5X`e<
zb}VeN*L*8WJp2h0j>FZ4&aXXgZ_^sN3A|nRKRBR$I=p=d&b$C`+ris&>W?_Q&EHSo
z+4e;Dsxj{KyQo}=u;MxS0p6y4#r|#CMEDNg=6M<(JgzpkKC20D--5SY)EHk|r`=v`
zU1oij@~CJ3O@z<H;U9e9NxtyDJYoYTqz-aerf9!L-m<>?ucf{Ay4H8E+Pf}hA3@#r
z*ImZy1^XNL{t4bLEfo(f;qA-t_V6Ece2p%Dh=mdG^ndjFz_+pRXqQ+BrDNgETCs3N
z{c>4bw%z*O<6GnhU@eS({zxp8(*K=Vv2gd{SeWvYTIhJpetOI|IkE60f4JHt7Vau*
z-CR5tau$w-8>o00_Z!4QMwtGcc~8-8;d<VePxIdYiuwkg-Z1V*YN9R!<Dm;2KDz}D
z@3en!jhheVK8tegpx^EA9p2V`Pn>nPHv(@b$j3jz+d1l$@~hOZtNm8#G@buX{x4{~
zlU@9!fi?D0+RE|w?e^{)jrdE=o8BJOlk`~xuF~^jG*d2{!v9|qQ>**h-;&3kkaK_Q
zXMH2L{q&wTwkT6pj(AmWk|+LRt7Ry|^YWhDFh$;&nG5ciyCQpyd_XSyTfK>g-M8%F
zq?9l8|EmhTtY^(5&L=dHV_V4eXh1VqDrR?~X5zj=lDybWy!JF7dKNv^!(3vst@alk
zs}8h=83DV$me0r8PaA1%HH_@qKb2SZ;C)97?ig>SxWA3BL<YGdoi)-uaz-j^HP@#e
z(sn=oQpkmO=ozKukN2XB&Qt4GuYf*$a2vh9U~gvsQW!|@|Ed??hKc8Y@P6b+*mzmp
zb<Mtl^Ezy@>!h`<V>Nbo5T#!4{r@y}{7QVl+QOgup0Bxg90>$(cd>77)OWBx6MchW
z>{jr0_HB6tfB&VnR((YMkp=#+^<erRRNQ`~d>LV8aXLyx{^%RCa$mut!SWhi_2>de
z;&g^q>es5XwG`bvE%*M*4)1IG`*V4Ih;=IaFEqw0=x4MB-i{u{zN6`Yj;p!<xm=l=
zPUpX^AH8OO7(V0e!)(@H{NZ1IRj0f)9KX5qvK(AcJm=$kdGJJ+hn}W)$NAdd;TIWT
zM4fVKGI*P_GR&yLR-SEJnq1n;^3iec^M%%GB6e*qcfa=uANtgKzmvTaaxLc<50T-+
zzG@1Xv8F%1d&7QomR!;Akt2-w617CFmcSeIExOBfg>i7GFquEIrY7HaTk!yIU)rgT
z*=LQsi=R86ukEX~)|G3lDHo~jmcsP)Fn*&r+2rW_pXKm)o?JE&zR=$sJpU!ehPF}L
z(p%pee6~1Us_*vmor2b2dk=YM_?%h@WuwFYkGJdTtXB?e|6jfDhqn*H+hv>WtMR`X
zv+Naq?HkUK{BO85iS@`c@?<$abnavI9K1c7g-@gRb9)acKM$_OxB99D$J%$jAV0I;
zN&3HEPDmj)#2&O?z~}yQuQop({u92`^M9E82p;$Gz243EA^qn-SJ~ofJ$oMI#aKyx
z$A(Y7t4^xLM*Q-BP57Mt%k&%Dik;bP4P7jv<53;e-R#!ATRfz;f4IC59Fns)!MIO~
zp(kUZeA-x;4kNdnObXAz+XC=*1H8S{Cn+Ss+s^%yLO7TdS`UwfG1fMlv&n_<A=+bZ
zAg|{3PIW5W9C<4#)PT3YUQP;?zD^1sbWIA)$|i-=g_A<cmy^O%ZzqL3HDckOB<ru{
z@_ADjYhU>VSbMf-%=bUBP!Zn#dtXxM^>61;=XU3C`LE7l`aMaZNSdT@{o$mr=BcF6
zs!&q6Rw^m%eIqISS0yRTsHHAQcxG=>C{fjV_fjm(wyxXzSUfa?wTDu{V0ilxy!`~;
zerzwb5zHv*e#(N{)FB1n?cN#sS*Ja2@4eQtc(}RQ_de`er?fn;z8F%Hyx9{W!+qAi
zzr@4Ca!a~)-r33}dnegLo1*@JyBVg#1i7peA8N{n+Aiili};h{q-8L9fqbq1xvsFb
zhMf1Zbu*kjZ?19~`GUQL!#&|x{do8W-li-`j?BJ$x-TBe--`LB8pfWncf8GB!7_XC
zur(vRT~#U;UVSeXI>F5qYLEwa#zKI%tH|Hs0v%iP_8p;pLM&80=y}+=2S(O~x1WxS
zg@@RoPs3Q)T7W%s$3nxr#(f@szv^Cn&)c7P!ZVrZF0ESS<~Q-s_;NgCfwwd95sa0B
z9{a-EQ;}c5+uZQ>*M0UC&;#)H#x^yMwlxQRS7oic>U`Ya{k9gqS8P2ew!TP&iW98&
zXL{cZ3tpP7CXvsN)9XaIyjNatiGHBNy72Z&`Yxfam?|f>2FQs|h>z$u2gmvUF*$jr
zT-9z6>_e~hwx1@S9x5x>$pQPysApfbzL6)A#D9~lFi!0KfNK6v-b|%7y31bSAMA2T
zY>U;me-ZD$sj2=H1F-ASwDP$;yIy>zZ!8y!!;a0>){ZsA<`Z)5<J!B*pHFlb|8RB#
zijL9REuFT7`lj*B-vg|<hN}lg!jv&$*zr)z-fC$z^<&o65ihrVlP*uq#6waF`-~6B
zn{Z+7-Rc4u`^&>>@rU4$^NjAlKu0UpCTHkw_?Q0H*lF+P^<7A}BkA^k^4r3B-f6(w
zL45U}9efcco{%#qkoP`vKYSThJKud$pTpK5dt7h#J^DiT*r@`0Emw0s32#?7Q-fB7
ztMbDD^iMf!lAh`#F_Lnk_cKQm;br4JhySW_^ZoezSU!C2QF%QtzUi{?^YEht%nQ~g
zu<1GWjLGK%Yr;LbZAmxwX4|xMb3}fN_V5}<`fFe<)=(e%uAHPDR+S&@y*#~cK+*g1
ztpm*gFY1r7&pFyY<p}RP*`kKrxrYs_@sD!y^&fQj{CzN(jW7RBw(H(u$k9Fj;fr`Y
zf?mjHy__5VJp*%d7*E^ug4RoXs{((#_Np9Q5mr`~qtP@nEqTFrJ}|J`i!hhpWpsbm
z$8gg85#CNF-vD`gBmVn$mAm=J(muw*^NRlXbRPYN{}9=J?xaq0yht|{%Ui#dV*i&s
zXO17}@Gi&J=BkXB@Bv>pN0WBeUwvW4IQTr5Uw#K$*IL`JSJy0qzYFPdiD$J9+n|Pl
zxtrJVujPDZp7rcR^}`Ul>dYVcP^r%L_Uu(n&&OZbz7+qsXFuBRUH&%vg6m-K8nxU?
z?-iD-yX3zMJMFXZ<K<#$5zI>qZ|83IEe!u#&L11X+A+iAL(~Ip<y%7sTI0am3NR;+
zJXtfdxpLb}g||g2@bOyew}f|kFm-xLHB)9a6#1`bSF`->n<N-p*S+*VdY9!q?a#hH
zI>k@r>1_C1b%I~AbGOv=lEwO%9apnYKQ_x!NnKbC)l|o^Rr)6M*G!!8von0I6`w2B
zC^`P4znd8CLH8e-6mnV<W*U<ex|fgnZqhz(QSVB#$HJE>V&V42q_BN#QaIW(DSYxt
zQfOZxDQtc)Da5Tan<gcNg$I&+#}o6dpS4qJdxMW9!sf^A1;VA4uyPW-tpKx2oK6Z&
zmL-LOwUa`fmy^QoeCB*M$(}$`X!JTB+9rikE0RL<jNZpKiiM})?G<?4#TdV3wHKN#
zDeQ;0>F;z7KmFc0Y_`UI6yD~7w^QJ4np{cYk%F+bcv4sj%V$@%madr;TH-O1A@|+h
z5x*4ko5WZ+>V5GY81QEb`{VGo7`#pQOgxN%Tb2H`c6%Tm{+Bx*&NuNJnqJ;z!;h--
z>~~JG2TiUq>%5bM8`bQCHoOrJk37O39%U!kkO|&C_^JB7ojtSO^6q#!5GHg&@50i2
ze5U$Pw&y3W_hH{J65%oU`g+`+bCUYLsc(!+yPgpDy^{Un?7sgh8V^mawNDL=hs8bp
zR;y7wTrcN4V{6-;_UC`FFR|X<dU1GrAH3ZPZ->u}`E7Pm7^Bws_1;+6{w95@rMko0
zKX$-Ic>4gnoqR497SD@?&%TO<(XjRKUV9GT$HKM!u~2PJEVPK(AAq$bt!KYw$I-c9
zZytDD$a)vv?yM6F=Z*W5`5sss3t9QV6Lsij3SWS$nHJgWUlR*CdeAMr%}8(QwWoxw
zD`9O4lpof<iK=PK4sRbh5)U1{zvu;bKY+WbvciDwax`6boI!VD@6VamT=4cZy$&1$
zlfSaZ4>$UalKbiX9`V>-9^V94I+v99#K$EVwDWs)kk~)H+rH%vwe1#l>{{QMi~0VG
z*g~%OYN2oNTu0xPHx%pH94oK#PGpT*NWLgxjC$wf1o%@$Jf=VI-GUskA6EQ!n@#WN
z1CvXolj}0UP`Rz8I=H#mn=kGv$#JRwlrxMo<C2{4uR1=b-1~;TsOoZoJo%zno-Z#S
z);6$>96VI}NV#H+_54@H65pNVhN5`L8PQ39vqa7qhli5?ktdDXU5;pRxBapE<(7xV
zI;?ojc{BV~mHQf!XARxGNMFqdkeNQu&5`@y?RWHh*zqU&_+x=wCO>xHWpDq8Jaz_V
zntMOq?~r}aao=(rg0=8=m>d=1?I_P*{hHn2X?nJ+BqxSv)Fi*g{Vj6u><`C7XZ7pg
zL_GZ5o;{7*z<6cx{|^4|!RHeX;`<SM>S~HBPbR-FKOxuUg13vx!GfxEKv!wyyM`^*
zQEHc>@OCClihiel`crr$|Fx0(R?B@;>cMH(&%>f^{a_EAnmPc6p}YgF3;WVdLhaJY
z`b58ZjoAqnrK{zg0=&&F-$pttC~vR$$9oHTc&8d?|4-_JYjjAj`RKe8UFUsV|7`Xo
zJhQ_ZZiGC28s6rDai#J9Lm9RyYc6sO<D0W#zzTE=O)4s{zu_GMe%sfFWzD@`<O5mA
zQyy<y@R+t69($^(@IB@K@U{cFqTlSV(5G2z_!;TOIJHWt)n2j}V(g>OThF5C8^6=f
zsx$Dk=QV3gvi9v`oeXaa&V|QItW{_IKbFoqE~@tF!z$Q{g@KBof`yG;gN23NC}0<$
zuzk)hU}7P5pxA-kVv8Uu0tPmU@z@xsC?@KA?aS|v^VvRe$6WKBnfol9-bU<)_j+#T
zypG``#L=b+j6Zn0GoJ6CLc77+Tbqc>a4tR^+jIavMc|8g5#w|twt<ZdD9vjPV~cs!
zojGUatqbGN%G;5=xA{1l<0NY_vl(OWVex;|!Z1enz}s5P!JCXHPsV6IBfo0{KWm^@
zN-}1O;^!2?mn%pg{K9UO;y5nY-BxIs9_(vmsECEJ%>HWSCkONk-NIP=Pa!@%m2tWQ
ze=iAMGZzcO-HFWWdKrvE-n+w=!2CY`6HR5uyrVB?(U(11bM38|)A%xV`f}|&j-X%1
zxDX$rUHj3VS!3|a8FMX|^F?XfRJJz3_+jw28oX`IW?5;Ll84Mv;H+5=y)}#PE3-JD
zX=_HB<p4aqI@Bx)XwtNj=ux=miY~31YL@-I&_+uwVtiwfvN7adsabl)HVsxrz*{@)
z^nV#<*#~dS!o$%yD*qh@H_@pJY|Y{ge{aKbM~*wR9JR*ScjI7-)HzBl8(xOM=-&=z
zSq5)!5=XxJRh4nss#NCpH4B*K1N-<@*33Pw%+kTtEQ&qamhbf7^I`n<+j+CBfVX$y
z?HhQz1KtjTw?qG+`OtOlv2X*;*9BJH$+bv2yglfG-G{?|?&vLYlZzwZ-bQrS?hvu;
zhkxgY<HFk^j3q67?P5<ZDnGXKG`j65aoPs-FLvglN?%1%Q@j}#Z^6E=#uhJQZ1JAY
zeAX<@q82NRxyqPGW(>R-kNsmjxvpn@Qu7e03vZ+0?IS(wm&mP`oJsxK0JL9gbU3^<
zl;(T?@RML<Wqg#8aCOujv-F^?UE|E+cg-xr;O#+Au5AiA-9milH=h+Jo=}1Ms_L|L
zLyHvbL+(6`JVG3NBpyAR_UH^l6=DvT=#M%bVDCON{~2qRulzm}?rQ77<l4kU%2>p$
zphf!R67y%w1ajP@BW6it?0joQuHY=kwE8!v$Qx%^q!~>1z?b-qPWPD?BF?nek|ekb
zZ=G*ZYlSZUdmn!d-hTZrL>3jto-@CK@KyfoMEApmC)o1owB=+tTa>NPO1QX!I1oN>
zxq0}cJa^rN_KHOt=fYv^<N_F&2`k#chPBsVE4)3y+@Aq&T~48okHX4*@SOc0+XM?a
z#%><lu?>qRzD57_JOR`2yE~=Bpd0w=JdeJQMq~~j<#_;on~(k*`HkE+TzP&CABk(r
za;D$?_)LSAqtAUO5bNrSrfiLW>OnlWFTVU3YKcs+jBDt3j5@?@#z$FfV=3&PEqeJc
zV+!t1V(b*(Pu}JNx{>SX#`T0gAl8n(nEy3Y;<?_0-;AYS_<aAkroz}+uD=|csSrAd
z&wAa&F2$i?X{X#^evf^8&-IR6i|vaF<z5=J8NN(0+Is}Y${)|MuX1dzvoLLV<TYa$
zK5oZ8T}{U4Im30(b~kCOKlDX7eb6bC^I@xIrk}X>618X}c>9)lGZo%i;B66jyKxG6
zC&u?ZeB0EOjA{JZl8lLJwA=UR)B^rs3}8o}w<Di2k?W=pf<Mwm#bB``>!h%Mi2=kR
zE!a7>nb`a^Ltee_VR+ed9QqL3_Yj}wk`Eek8q87ibYXv3`)D?P5jHZVBkhCjY!9ar
z{D^I0BdxV=Md58~Z(^h{D}gc85*`eIwPkXsQGvISw1?Xp#uV*c4c=<t?LvHn=2gfK
zl;^m0VLo<xU~PP$8t@DayMp80>;fO*u7%fWFmMGtyv7y@Z@a?c4MWjHoVy*@<_>S`
z!rQuB+X;C4dpb;ox0hkuh6v)<u+pAuww`MozZ=Q*dU4JQ$FOSKoehY!H%9w5BBol0
zI8H6b1iWp)>)jkbp7ZF)Z;OqFz5Ld?W+!MPdU-4U0KARmd9A&~^iPM0B+y2Q#9HC)
zYIxiD0D3r@ea}Vf>1k`)&U-MiBib^SzF$S37b{F0z@2&AknviKF~)n-h}YGGl~?OA
zmxt4yjG4NL_*>_QSHZ!jTSFyoJ24QpPuR~Ei-;*08QaVmFYH0r63pKs<Sp389$P+R
zoRr|bH`VbITk;!t8#)p#6aarGU=!Ga!|?eS*B2GW?<B?&@8$4#obL>_&I9J(dA`$)
zIof(PbLuuW{ReqLcsm&0E}|a;>7P^d<&Z3X$2{s`^)F*W#fkl@%$v(=m|M)F_3N=6
ze75BhIj(E?v+4Mi%gn4jG?P<-t85Do!lpCCZZE@FZ0n5!W+}SGEV}V#vF~M;8BNfq
z@b)Y$`-#q6^-`6zyQ+*yMsK1qcSc~l(Mb!}VY|^jy3*M5_h$b4gZ;Db`R!rmJ5?5a
z=C|yl2fV!9*(`6?p?~3GMR<F#hDDq{o8`+Ucn@#K_`p<nI}qMJ_)RSNqbkoHtMVdM
z<vw((*qKzh?XAj6FI9>*gTr-HnL0p~B@0wJcu<uj`0F>yEVDnD#m$NQB>LTB3|cId
z+M(qZ`Fq+Thv7?DCV5JDTMXVls{rS!!2@_318-Yn3+hLQ$Ub-*2yau)hDb;3#!z@$
zy&!e~|1As~dxFm@qKBWsio!zQk7k@LU{2BxF^rj8^kJiAXoX1D*}&W0Gx4eD`!n?Q
z93O1(AZ$~6{5N!Lz9!VJIbv(zZ94X7*sKr<oklGZwO;A(h##j}Bnk~Wy({|lHGb0^
z+9}8^Dbt7@Y&T2#IpY3T&GH;y!ym4m+lGHcAN;v%=KoUhootA)S44ZdS)>74b)Y|S
zjl;A}2wWu25Yg2vo%);Q%rN?3Dn8jRvmAq?`A?G{xM-Hkf5_>RC%;>YHiYSo;cx=H
zZAd?!<$E*s<HPOXyNshW_C0SUdE&)9;uq+rS>%YqF}vWi975k>TV3I8Rd{<A-e$ww
z9<<?km}LuZFFIlS+j5^P+G{&wJBjujL|gj7+JbCN;l@zrNY-N3?1mF-#^#URlcz&>
z{ADh~n)2An!_1FUJMpojSmQkd?gSAd^(N-unEa3<{gfLbE#8NSQ$AwX%$F_jqb6I0
zXc$BP%{>l7QepLVblw@}`d<7fj=P{6u`oO4XC`(17ecsq6W6kl`ZLDOkYLufPlhWB
zpRI)-X!$Q$dH4D_Y|s4Wxb0lg>)4&yO~~(aojp3kx895?#zR>jd<Izj-(<L_Vp|wb
zM`7FG3wiUHKbH{S2l)Cg;s0mq@nFG@-^6!+5ntolKNiG3U@zR+Y~bw%?8kxn#9TOT
zIySH`HYaQ$d2+^Or{!=BdohXj{l1lJqOE&#jrQl^G1uMY9zNq^_M3tJz<w-!#Qv|~
zdz{8!;hg+wtGl#)$XqyWfJc7(emKX#_C-6xDeUaRn<40QYDKifX{LtAYT9O>hFB6Y
zklFZ@&7#TiGbUzVL<hs$f!NUz%&!_A*s(zT1NvYfwr&Qz9RP3Z!ISaW_u8=bE?YAb
z*NEL8gq|n}Z>{%kodQD^45GFN-a5nED0AL#3bcbEXq6v=ErqueJ<+o8N`)Ci;B6Sc
zi-bdduxky+Im{R`y~8JhwMk#eUw?wLv_~VhOwOqmp9jI)x)rIfL*s7e9P2p29T(1%
z?WF^8X};6DFy~5t6yx;?xHk#j&W5$4c+7A|=fXd|7Yt*I<k-*p;wNz)sW7!OytUpZ
zr5e2Tn1_}RAw~|5t#R`$-uMJVU}=9gew)^XoCD__11Fx=ffaD}HRoZCcPDY&4-H`_
zynV#+toLZMfwwie=1IZiOakx~rt%wn6Ss|U7`@y7B%i}u7q$jGZ+!&+Wj7ixikvgN
z&CmI*rtR9(AKpEP(U@T!ecrhczH52L8FQpYF0nP{N!Dw8t9;l$H#pQC7EWPouZOFQ
zqeZYQWgYX6@#wG=jzyqv!|<#9@#|VK4&m*e-`EHCQwg6u3Y$^!4Y9xC#Cfn46`C?$
zJJ2U1U?Te+#s0Un!T;pF0v)jjo{ayI@RhOu0p6|%VC-Xqt+iIWI8JNq%tt+WLEfLG
zf%~EOxZjD_&|W|3pQsXCBXhumu~;0YT6ychITzT#IvwUu_-e*!G`>GJ<9RIO{u*P1
z`I7xV-VTDb_h*@9UO%%Og{Oy#qe;t{#l3=AHdltXPUKs8+*uwQo}V}+tnI<J>y;|k
z;O&Desw_XGil?H=c8x0GY!{5=T6d#i&#SD>R#^+C@?Y1)p`WO75oX%J+gjvd+oP5H
zmLiti3?2+NOGiJmboWOmp}AcY^zc}-yn?s<IgS(~j%<0R$~^WLvO|S2s#G7ZihqAq
zwtA}6B&hPQnkwsfZ3Ayx?N{X$=dAjok>P$Rc>5lmdutSNRJ7ZWh16p0BnEp4?!3(7
z?F@3eYvFCX{KRVC5nFvl{So}TybAvbJF#vrng!m@e@uSyJ4}SPo3RI5b3*vPv=C{z
zm3Z?#YH+L52lVL&xVnsSG=3xB!>0LRpVQeQW-$Ijh+ELdarAW;pHOkazD{Y6PMg8;
zep&cWc#C+Jg;9slmh?})c4mnyXO`vo5wkX_qI_3ncu%uD8fq3BZ?klv?b2vF|BYtp
zd&VryCKEFlV3vb=;t)%SUBKHx_oyj@x3l5xQh56ZojO;@33yN=$FXL>=?-1YlFMf0
zZ5+I<GX_7)VwOdR$%&tW=dosNv03`fXJ4FG&TX^w{EF`cZ+l(AZ|iQBJ{-3~xLH>6
zT?cqO0^a(<+Zph71iU?@K@-E=QTS+r_8s?-HDT{rBisn>GmyIY?Zl(utvl>({h2y*
z=F8XKtd)X0S7s9*2x4wAA41tKv3=)t04&-XPA-Uf+Gaa`&>rUMN!s>2diN5aClj-W
zx9(9ec{<GHH-qV;!i}*9RpBYNVEjq?(U-MKqp73g9A4oET&9oHV2E|y=-5=|A@gVU
z37Eq?tImG<b|H>i0>7L2KM~%(!!NUgx6k2iQFxm<k^LB{_m6@z_!fI$x^Ds2DVKvw
zjvU_=`%;f}=S>(NZSey-!;_w{y)W^y!En!)bLG0fE9Ao%Pi<FVGZwOD*@H30I2y-$
zw$Y5U_xKI4U`{rE8{Dmsi&n^oKE}59C<#yM(uTz77yb>AaV7CRINpNs#KqayBF1KO
z+ARWmadSNyAKp4}eU5w4Z?s=BZMP<kaeN0o&$i}18u%fa{UVRF@yK2HF`Sq00yg@<
ze;V;+&imsqbe|`-sG5~Y#5UlqhIup|-j0B``TJ4N7zl4?hlp+gIfvE6mY8Ej67ek`
zU`J`wN59CQFgG@13kz_Zu~%s;u63wAc|&;8dq7?-?6cw2!@$_N8nhk^DmIDQU^sFE
z1{H_5JFVP=LBG+Lh2ZO-w!9DHLVnXupBW>svCXtwPAQn@%0AID%iysadggH<a&fe6
z&NKM%6(5=H(kJZdBXkn&@P_s%#TZ**9dE~o=^sU_@mY;()Z)V4l2wQ`R)E2<wjuj^
zN}n8o%Ln0~V8;_UkNarQL^#-rV<)4hg3FU5Er)%_U-aSm*GBwy0-S}hw&AcD#<hco
z)>_yv<M1!I{!d(I8F+gM-j?TDk2QyTbzwHw(5e=?n&a8=xR2u`bKIxwb16KwZ$R6?
z{5x$pub$MXje?_7@oCt%b`~{$VYC;=3gvZCxSTD-xH-?poLiOR_z{1YKXGV=M<L>v
zhTn7rU+FGm{Uf&UPu_T2SqS!g;l4`D>)m{BJj}IcES7eIg(H}=s~B&)@vC6h&2{*0
zj7KYP!y{<td9a&t70+C=z}x!F)hCRN67Y5+y!B;V=Pyj#GtQD45tDAgd}z-cU~U^)
zqc_?7JJ5g3nKkUAEAuNKuiMt5@0j~fYct3B&F&u9U(RJ%AU?K#s5t3h_G0YHdSXy;
z){S{|y%2GM4CdWAYG7j-H#^`tj2*xnO5;(__A45{9Nr%0^G?_c>pIx6%gnMGeY<ro
zJez5jv3<?50N#3jQYDm_ZFl0bec<W>?0Q4^SOq@bf_c8ggdNb9K>@0)pQ1|bAXRc!
zs8SN%u7$Nmwx>E(O3Y+Gk!au}stkp<W_a5ZZQ34<yMX=ow(cuem5T5-yfxYj8{G>w
z&YFn*MyFlleDL6866chB2afwwpWFy-{a%r&iHfv4tH_WHMUon;(%M6n-1@2%hqnQJ
zRp|zAi^AJLS9F#_J=IRLl=}vI99c&+7#(Z0$obi<Q-!yaE?YP^YK73ZtKjXB(!_+}
zUFEmLT>m3ZP8|0TwjlT%>wL*chHr+!n^=29EZUbbG_N#v7v5&>gVmeh12%QWvk+-A
z945@Czv-(>Y*yA5U@Y`~$a)yYqxb0$>LSswgQ;C=jxB0}&)<--=|&yYP4eH9sSz(p
zO#T>cL;vh-Zk8B$`*EEr3!C6?Tveq(RkUmiv+SXrYT_>~g^dN!qmPD@FYjR%JKD4o
zyv<pRRy_bi;q8uWd>VKg4sS2P+aJV>@59_F@OBWqt?YqcWyklERq;NmN+B<^OefZG
zWHz;PjG248i4~(=)8OeE_CKqhng4Ax%YbUIwhA@uZQwuehjPyC;B9ylV$uW1@p{vy
z<EY<)x9i|-QFxm<m)!q0i#XAinecY%U+SlAnMdAeTX?rAl^6oNE!q|h*p1jcEUMj!
zImld#XO4A(T_w>#-tfcT0UcPG*t3B=%qHe7tOz^_PvC{aCA4f3^NzVOXf*LASoaBk
z$(}i1KAqZhIGMbSxJ3~5pfA2Nyfwhe@iFk=4A0?`=K<y(bF9y0j)Ofd2a^h$(8Yqi
zC_-GCJ|DiEI4J$Nw+3~HY|ZG;Nh8Ryqv1}jp}!d$TF$vXyzR?5^kA+(c1FWjq28`2
z_P!XoC$1qGTTp`Q{n(2bnirb29Wm5o_%n~Xw;yH+xMP;6>-bX_sA<HH`n{eQ5WdT6
zSTq9LZ~lND_(^Q+JN5^Lb%-Q4vKt?Ub2MV(Dxfut<r$mo`_f$O7q;RE+wVx?0$kG|
zj&0MF7<zsDL)xqX*ZeJoSi&vt9d?^oLL%`$-urg}ZlAz5Vk7Mq@!53lQ=($ieCQ8%
z`0SY1F9m3yNUl4P>vv|{?_G}%GJx~yL%eY~YZBpaCVpnl66)!8hRA_S+!Ky|`SgbR
zb^7;91+>Rx>@N2B6ZY%KN9>p*_8va`?8SN&81W6BR)#n4;Lh?1aA;!Q{YiY`ari*u
zwJ^xK7WNmsO@l4H{t`d<MO=b*IuCEVzoM<5;LmY=AMh=F+5X(X_Oh9;6Jz1E4ZQWs
zWc)qndiief!@O8wQ9jR4n{;Blxxm|EjE|A&v^|U+!M36@Jmj$!yv>Kbz5{R1!P_l0
z(bBN)JFGRcS^sCr6^=!~=U7<wA2vQ5o$LW`-@#;SJ+!rs>zhA0YnV6=f5Lev+P5Eh
zX|D4D*O^Q^WOhYUqu)!z(>m<$CcJG7XPx1#_1<u;VQ>h%P37|xj+xE&nEgA@o&#v3
ze(dKd$BEYvONF-$;cW`M?FVnCj77t<{{Z$^gmc_dlschYG$Xui4M)9iFkk7vf47MN
zzND6xF=w4)SMar3T;qKnU9f?ne3lD0ooZ8`&N#2UkT^&TJ|^?(=30Ce#$&?>bni_3
zQO1*@8Zn(Rv@3Ju-dFhgmNllZ`)V5bDdLS63o^z^&>z@i+gijC+vEQ*m;Q1dN1AhP
z%&89^jC*)%<@Uu^)Ie0E@0o+!IEM>-XH*Y<18<j1U_3DAS7IyH<I^UrXUubb&r{i6
zur4tJe~R&bXBK0E?f6{$N#@SE6?y*fdmhWc+cigsmmk7^-OE@&7kwE=9TYm~W`J2D
zdZTAbn8o=8`VkF!5$^u@49j3{et6ptJ{ovTft5dE@_0K9-uh2crQSqU&cWZmu(v9_
z-3o7K&rl^ILKS<M8yKca{8~7O4X>D~lAmHf>}wwTtPIC}iow^q=%GRAva#qhbXvNT
zSq8(|u^cOXKK%4kWm`kI`%aO@@HXI#B1f+&GUbILKe7}F`k{zY2_K=iDr>alm^ZVz
zQ_}@+1JPsM&|d}f!_Riy8w2L9gtx_4pzY9JJ6Bm`D>2%?S*%6+gYEwbPl+)vBRBe_
zAG+%>@o7Iryn_`9U8+d@9!0VrE0W)b{h)uZN26UfllyIkKD$fo6dr$_$$dHIP*XIE
zny*pBUfYoCtc+%^Ky7b%*hbve1mnVqyW5mP_t{Y|#&iGq@E<>*-+0#9l%US(DDmPM
za17p_g}2srypNA@Dpi#~@GSw}M#I|z@U{uObqvHmp>L<7Gh4#S1V3`zp=SQajab1g
zwCiWHIMEJ{9DnRTbm|IX`Q6C7(%x>w8@@Voy#Mhw^PnpCqE*><S(VXf*^=-!1>Syu
zx8d+siw5@kA8$Lu+qw>nsRr=A4~*xWI~}7I0Nyr+x32KE?pXLbl{z|j+jS!G>(Ru+
zd7RHRzj?rQ!`n1?djj6x-axzvzr-0n^leRzL^tBeow4)qb~e2232!H3!*$=#qs58W
zW0R9$$ZO`?gu>*GvY7Ai^f*3ncjjXzbNnUy$`41xo}3T7wRy#{um$<xt%c)Nn})v)
z16GeDXRo6^eFqx*1T|iqM-=mR9e$N3b9UZR{Mjwo2G~B3c{~c9gkx6*z}p(cOY(88
z?Rr~eB4cNsj`daeXj@K@2gheSe1_+YkKHryckrn$FosGFqL!aL`qDPkiPt9n!Z?^x
zkh-@I)FV7K%aaFYnS(DhJcU|@`@DCX7(cwthPP8-?PRu`|IG3;n|Ktl{WNTi?`LcN
z8Q)<hZ9uH{0ltIhef$UXh5@@2OB+mVfK5Ase*^EHHs+o;HssGa$Av-oKC4+jRuJF7
ziTjf@Am)P~QML<tH$VKv$uJQ<W<;RFkE5x#;lJ~I1-$J?AJwW)E*u-y+>!V*ZSy6A
z{>08Mzsx;W;B6=7iVuFxSz<|U@YaXpKVg6I@U|1Y{YZZ}qmO#i$Bodp+4SvCY~(Eq
z`Fm_z2yA=?r}BBA5wZQv;O!$glW64){MqeC9g;8H<+&#uy2!Q|-U=+K#qUbNyJ>&W
zBR^R;%r=U)3Vn&agrNzy@srbtp|It`+QDpkczTz|E>DS7z9Clf9Dj!Ij!DNa!Ur*3
z#+DzXhAIl*WCOq1#dq$a*<fo3j8)3PC|mNnHk@xk-YZFt8Qv~;g2!-c9=z)UZ$r?v
zx*ur8$LQrd_%~P4WEWroK88zY;<aeko9Ombc&m65JLh^mxvq)ub|IgQ;qfM~KU=R8
z-hPCsl{ts3Qq)*gM`ydBqj_9V6CSgb;P`#;1r*N3iE})}<Ar+oDT9eC!}HQG_tXUH
z2sq|6cw38hyy}M!(v>(8{p*FF_^k{Y4+ah_z&&|>5|2owzBC>`iTM)1oQQ#mX6Bgn
zemy_=Y#`%c$pgk9eXFByt6Rqjy!E1AD{Uvnw-x^n-fmb*U(X@FG8R5EPK#F~KUyhN
zoH*9uuiS$P8?gE@bM!j*hGl*Z!H0XtSl>AtUk-f~%~*PJfcq#g<~(6(wGNCK=I#{6
z(OfobY<eu?D$xeLTM8f38DFt8V;9~Igtv{x<7dI!t?>5Zdh7+f{gVRs@v&0q>l-03
z*^E63%gYU>GXBTF+4^h`*)DFPJ@zxd4v<gW3!jO_ezbCuEriW&vMR^Xq>`b^d2H<P
z)##{|s!V6|hF!Wns_g#{rov)(;+fxw8Q&kE%3JnRh2KAiyL+ec7_3TZc)Qs|%#+76
zy!JMzayW`(5qp0A26n>RCh#^D?cC%hk7zZ|m#U2X2fzEWo)#aVDfYj|ZE~^1jt|XK
zr7XNH-3b1}+d1)yX!a}QqZJ8$rO1BX8}v<)cjeHyJzzb&y@d{(e+(bt986E=xFgX6
z<Q^Bbg{ea<GHfL*;C#!Xo2%k0c;YKq%&G)0BVK;OEcG8-M7_@1R?hXmv5G9sHOZ@g
zCTZ%T$m?N>v|p^qhF^*l>7|N34xNXM8k7&6bju=PD_BznR}-;^rghYPp+TOflL!Ax
zO&4ta1!o76ul`vE&hUCXtbKu&9nSW|mYix9bxn_n>+}0E=*mjNi9^&ei-!2X)K#hs
zszqNUD01u(zK5GCVVv`&7_J51sC8HL4(zqYg}?POOABA@H0|j`++YfAu%jnjElVte
z7=ZI&SUL=D`w}0(H#&-c^|Tsmo#E~+o-4;y@uJ@}pA`v#^M6wCjneQvo~z=K!*v%n
z%XPGES++v_t|5JR{1tig`eu1gKY9?4&Ns>|ansDQdkgXCG_$;bx7FZnVf=|_zT|T!
zSXj15%wQKf`zD(HH~t7VeH%VZ<HgL|FmxkxsKzq5x{Y{1Eb)g#@_fwOC-}(4;O$@L
zZ%0_$fi1(1d72*`ivJt>AM=vW7Tm+fPM}T$=9~${mcL*P_;ZWw`$!DqGsnyzj_}$d
z<=PQ%?M;lq7cB_`uEATc2spH#*cg7a)vtCGYIyzebMas2>ahjPTMIVe47SB<Ir*wY
z`gAyX3oq6(_hn6cphZqBqPA=Ye#Rm4=J-<a_!sAR|I0XzNgEV_x6}E4_x2WiZHweG
zCYF(V?+$Mlz}qx<`*bh$Y715FrLM@|(u$0JYLc4wOfn(SB(eCO9iPG7x8xJxtqtc?
z16_EVHd}}da_1V7xR%A(y^$Zdw@3+cwg)ZJitqXsx5%t2)S}m=Pdup6XMQwECT7C8
z=|`NR?i%Vj;Co|yv;P>=Pa3gCkNH!L_(V9@F;P!W0Dp3rAMv3n_ziH{L>rXMrarD4
zEQPH_u!-*U*X@0@t-^nQjz+U~p)KvHD{Vm>fi{0MjQ*UC7UBCf<M4gp?HSIs5B9j1
z8$MP?;&h?Jk~X3j$`D_Lo0Wz9B)oNpx8?ev-_W;-Fm@5VeKr!l@w^)hItGiz!yyOw
z?CFZ;D}g4?<v%`t5N99;aN+}eeTm+ItKIhzyI%?C0`aK^5EpMp>>B%gHkEk89qeW}
zd5V$vbnW3R?=^tSEidzU3_Y_Np8(tZZ6|Tc)7WzC@*N(FC!v$>!IiJX4`A*{c-t7Z
zCBVBW@OB2gtx_NV0ruDV2+QGZ369nBD)EC0#B12DU4jW5YYX<?7LA?ykA0zi-{3pc
z<@r>8dx>jV_zh;{!06A!c(|_9R?hRi&1|k{+gNzp3D(-N72^1c1E0~Jt=LZwV&6}i
z5eJ|hN5Wg{`tcZe>o6L&`>>8}2sLVc_^WVf2mR6k{n)fD=gjyY32$$}+XR@pjsCk!
ze`o&3b9}9hDa3u^(YLT~6!WMUb8^*l?tz2PJJW^pgtz5r*M(@>d$gh8w=9FVHRt8!
zS;x}{w>e(n@^GvqaRm6<<vbb(ziebO^Phc(twZC6u(p2|;|Si`MdD||;GdTmNBG+L
z8Ba|(r)g{%*FwbaD)Eyq_|P_JQ})@0Yl-DJ-@M7;z}xch)^9ZR4m0QvxZf#)<HOOU
zHTZk<t3e+saYLw^m`sd{ak-e!`}28Gw#QsY*1AyH$XMvXW6E-T+i}E!C%_h1>BCkW
z-c~@9&U{E6Fubh}Z-=Z<<s@6;R#lu1!>?;_D-7-HuSm*0qbxpVl-ehaGA!LB`?1~r
z@GAzsmSStFp@x{(&3V2U{{0%NO0To99o|0v%)Vgl6KwXQG*$ePRH=)tPDkq=Z%=-3
zH1X~7*zq`3I$-aI!s{y|(O&ha&q`P1c$^}a_bD>%AiRxH#CNA6m2WEYvMj&viuQ){
zo!~>@G4724Z(G3gv6W%bG==+CC{k;+iJC-{3><Ef=95gaHpnD<!b}nxWs)-IOmg<K
zN$N!?%sWN;MJm#~nIe;Z@;#VasfHr0hbnT4&m&=cPIs>NJT<v+q-!VcFWCZy+(c_)
zAEUzHFwDE{OkG+b;=&(b*aLFf<hA$gQKjv6mH*484dB=yTbPO-|2Ytiik3{+MNalC
zYo>T^2XCLlTlXjA%~#TnuCx{B5LHf*%!Y~_>`QylCl9tOvH&0C*KSotG$r;6TOaz6
zPldHB(Yv23V>3CPFaC?OH|H^k{lVMc@YZn-zQ#%75${zQ_)e9Uu-zDo4@Fx}#-2Z-
zAM&LVqoBWL5(7E%g&aKBR2aYLCEGJcn85q1$+3U7#wyUbw(vI14`#qyfw$qQ#8J@J
z4o%5T_vc=nXggnc>j7`Oz}r3WcGNEz)sX+0@W3yH6AJdZg@!tdaOUg^{!b$cpP2b|
zoO$~htu%t~jAH)2w8z&hjV>sHP5Q+>Jzl{`=6xjdZiE-}(1rCUW%#`<wXm?ZFM9gX
zMb;Bvh7mk(1aGIeMeAYv7Y)H*9*@473a96DUlf0A!)zYcU_%5uFc#ZN?CHiiwD&5s
z_CkKQAy2EPYq$=0d#$5IGPusmy}8z5X#8Ka|9rD-tcV{%Jwo(bd`ae0$wR7SjbdEa
zRHbk}#_v5v&KecTrC+aAR%GiBlT<uqlH#WsmvJU(a?vF9ADH;x5tG=#+p7sCVm<hR
zaJW9c=M-Wf@2WDloai^^bUS=VuchRDm`mR2^eev9!qs_ojAn8y9pJ6*AMt3;Ys@X?
zE$5V<elEiIPBUhHET_iq2IKoLH4x+=IyQ!*L&?9u+gZf7yTIE+^ka`!7P&f!dU@hS
zrxR!ke3>}<@Bub28rwFR{tDce*Pp}sQCHTHb}NdX)B?WtVSU<A{tI*@_bwZUKA9XM
z4+HTF@V6S!uk)LsA>eJVu4qy0*@Q#r=fmW+W63$Qoj6H;7bd-inOayH;mjJ_ro>I*
z@RGi;8RkwS7JIt`bv^Lbx<<MAFKlWS`+%9fu%V7e;Rk%P*6dX07}ol%nFEMlw<Sgh
zZ>|4%wI3}TWJU)KCAQfH+sb>dZtxp;8_DnOVMP!&_B_1Jcbap=E??yF2cKK+aU57`
z0}JnD6Mx}(vG=_9hP>fxG~*ZI`QOPo@qYFr_I(FFb3T2zu6YUQZg@Ky-Y$BG{(glm
ze@Bk+17nbDXb5X3{g1by@YbGV?1#73oPOwEY8GJa7PghJcJx=;jNku&*-iQUOnx*h
zZFZ9FG23tU^PJD_(x$s<<6mouSs2lKur?dr+-efcfoGk9$)7Pt&X2{Pnn%qebHfpz
zZu1fRPUhXeGsKln;<NHt^CErEoIMEp@-YsqF{KFnQ~NCHoM7Hf{8%^GTj~enf$h{M
z){HS`LX+r!ezPi(ISX62uvKSYz44t>IfjPUDX=&4EbIBvvoY<duW3w71V7KUKluvI
z?M5hegL68`*vr3~bq|ZUuMu<k@pf{A|KshZQshD$7(?~xgSz-zEm-dhYh!(B<3Ws*
zp7>0i$V;J*Yq5Fp{J-w_LVao5$uKu0RGb#%@i-T@W-Ow8!|@x~{O01H^V$g==;4Jv
z?WanC;i_oi@7n3AL}Obc^d`yrV3b}PjZ$ueQ989U$_)pjlr3VE#BT;{lR>&&G03wh
z1HO+zTx<<;#@ir+e;K4hGuT^%{RN|kreZ^QJjIspw~6~88)VNagCzAZNWJqqxiViX
zL)vJ>Hb%q$3TP!_r&fN&nB+qQ@n-lw;{-MWJ3m{`xeb9&_3;VdZLK4UjGb(fuLF$I
zp_oy!HAXq%kL@Xeje)nX$D*M>D%2gIOCwFvrJPBE-x_7@C!@swF^Xz$lFO}4@_rb0
z4BlGcZCiM2Sz{6twyWNAlMMN55;uEA)CP*wX#`iXdlO*pqH4sH`zbOiMByGzZ2U${
z260N#i_b~DMDN$jvW<Fqu}CikOnMpHMK7uK^^#CZF9(X~<>?=t^!l!oGFdv=_*N&g
z-|J*he!YxqpqKSJz5I*Pa}NT&+$n01sof1S@sJ{(#K_ChCMz}(dzeLR*O|KP{)+4?
zqexRxxYv^+vD1mGZ&YMCyzKzXPuD_Edct{ldwGIc{t+{_^44J=zQ}xH6ZE^Urz(-1
z@P+!Q(uV$Th<~yg)~<jB!(jVNj(5^uW$wXtuJ_yoMQQ{pV#WvAy-AT_=V9(;e3?Xi
z8tn8A?D?46s@T@14glW%M_jwTH?_^gxhpTGCXRe=U_R=rn-Ej&M@=+#THtLdc)NGE
zMH1lc_8-JzYY}_I9v6hA8<;a~I$)0*GOwGE8*jz^NreA<!an~T!1tK@(>{_DWbV|=
zC4N*GI|FasU$dr)d0B2bbG<J$ST58sm4TnO)N8=ov*{f3Jh~1o*5V=SpvqG>32)2v
zU_P?%bJ&s|@U|khWX2Tcx0#qTT>U*7`*4JK|9<iStGH(x=ib-IeYVi62_uOGds7RB
z{=1D<JlTLXXGK^Wy$b)ah*|WseKh@dVz4SV+*Mg-&%F4e@PFNkOgOE`7y2(WToErV
zISX$^-Z&`o7XPp5A(Pm^+mi71`azRC^)QJ?JCjr%YLbX(6ZhXV!6b!ywki_1L6H(R
z_&hbqr;MPV**5gT&uyd1o%X79=GgJ%5d4X^#Gm7wIufI}NxjB(@<MDOSMjs1sIn!8
zF;vnlT`N&XS%W+TF^iziW?AA(zb@eYIBF}2zc`|i17NL=K3xoNRd^c>Zy&?kl@+j;
zdg6X->7NtW;=R~YmHZvNJx!k$E`-g6w@u;g7TSCkyqz|VJTJU`5kNcxKg)X^->nCi
z;BC%GbmSPc!$h=1APlw;^P305_7dNOYvZ32(}Sg%=(b&r(Pc2UFS>U)yfyROS*Qzw
zxB1{oO?dkiUhTUAQ%1lxS90`>ovc6b;vaR>*x;(wh+{Y7IGxdsLpcVvRWpj?c*6L`
za4&^=@l<#h&wk+2vlVEiXf(}Ea)R*61Kvh+UXL$hli_U_j;DvKOX2NOxcU@keujO0
zId5G?sQkN+zX;POz*yU}u=^BoWFAMt+s2oOQD0}xARG>7U%%k)_BYg`v(MgNiD6$Q
zS9t_J(-s5v@EsmM@m|~?Y981}>qPvCB;o{5$;)%h_nGYTJ?9T^chP3$iF?PBi~SC7
z1NrPI=N|)aKX6^Kdgc@{=F#vrmYnapgCX*Sxe-NQMKM3eGDj}2ArBA3?jPoVN$^9T
z9j7mk;a|bJLwvtHuP4CD#y3Lb+kO1shr}+P)4$ByKiT-fhxiO8_S(<9<vRv=TLIoy
zgSSf)LnQ73`(rE4^H$93BJj2`uj}v~N5<ChNNVibF-IFS_jx?WJpML@x>p0^hphzT
zuAv1!6GJ&R1OIs|emckAXG0&+*Q4RB<um8<jd%<F|D_Q9*8pE@Fk=edzGeIsg|*e#
z(s}OYNqvMDv1rCbqYz>d;V_zQ%tCU+u(uM=7tUrpuyvnH?FqK_3)<`@w%iw;b3w>^
zwq=g5G>Hn!o}?LNTCh<TiBZP58Krb3qs%X46itpn^1m{ObCN;&!`qpD2GRc3%kCh(
zeEp}BV#zx3*rp@CtrM5NI=Szzlf+M2>3>BlL5sCAw1ZYY{?y3NM2(b+)X3(18hN%u
zBiFuYWXwJ-_qEeW|5Uxq$kNM6cY{=#VvzkDYj#zm9CJ6y%t1!(L2Z-;!;Mm>ol#N?
z8YK<hPUoEel2dhlqlnu%h5IKfQf9D8YQxw$FO0H3(<rNX{v_8ZN=cJcC~p#@qe;%!
zMK829$qas9XcZa)yH*R{THviKye$B4HE4~9ER*!8V3I#wjN-r6Kpl`l(w^$&^?kjp
zzM+@>m-SKw-k#s5=RU}K=@X$BlZCCTUNYeAUU<6!-j@8KlaPlx>2gsgz0c{SJ=+pq
z7pnwwr|PBQ3cbvHq?bf^d#1BN-v2eofgq!dEnt${b4;?*4PC8M<Pe;_N?g&O$5!z6
z61<%rM~)eP;{Z1OIC`@oy!}iJ`Sf`11p;q}R5Q!#qpYQx4O8j&wu}K2yp4vpJMp>t
z^(PO|NR>AQ`HWa~YuH_tZ9}Xgy{0NM+gFjz=;)V=IS#x%v_g>>7*OSaNy=U~@&69Q
z9+pu@2kX1S+XL`+7}|Uyyq$`TF1Fh&VQ0uiKF4+wFCBsY+luy!4YkPcbrx|sKn@qZ
zx$+Kr@P$R5+fobOoVXA3XL3XC3s#@?o6M&i{MU&vwD>@3s+d1h;q74NN|C4d#Hqxl
ziP0tzpLHiT>PGzYPc3TDUD4CzkAK5fe|Vb<Z|lNaHI~OS@bMG6x-#>O<K69wZScWH
zOe7~1#IZQe)GEYev-uwTkBNt=@V0s+Iu3pJXCn6jY0kRkW)}VrkM&L*yJmf~^B-dA
zg{fnLx6_LfyEs5?IdOzEe8-(F@R`aGzxkwy0p13~+n4ax2i``&TbmK&3edM5;cbU`
zCaJT-C~uOCvg5T;CKfiyRA-YUw`80eY5x<%XPPPE0B`57Q)JdNax3_rK6UYf+7ioX
zsmdpwJJjIV#Chs8R3(;vD!+vd)~2#)Qdnn)KQL?$u@_=Ojo@um3HlG-ZiBbYh*=aL
zVV1l2Z`FxE6+cV;A-t^#Z->L%7PNEa(fCr!$U9%f7kN*OeFb6*I{pI_TYT;aoW&lS
z&>K$ZkS=!Aj@g9B#zxeadXlT}M{aL8_7;DpE4;l1Z!f^xKJfM!yd95?dkJq{;BChL
zc)JPSh7s$l$#%rb-5uz>i)hbR<X)*c`rVDVu@^DqUgR!e#ooH`)`dEsmuQsr=)2M6
z+MLj<*wYg=&^CpLTf*C8f3g21iQ~fCFnGJV6Y);AW}HJQ;;A3IqH}7ZTi`~yGcbjC
zX*D$4`W!eyJ|l_W&qpH_!`5zvxx?f6J)E@ucW7Z)TNX{){sSC_w>g|^D|q`goqF+P
zYMbF)I{G~V)?G)tS31L%z`1Z;cj0IHP1pyMv)B%!cfVYLk9_`~-}UBL1@^%ScpJ;s
zm(9rQhUfBl`x5R)u(`m8o7K^6@T%-?Vnc9l%sZaf#h++E9-DLj4p*A><N9gy*6=p5
z2HY<LJD6|1Ea=;*<R?baPF?8VCe(>Kv(aZQ=;QHcnJ-7FkBGrP-G*Ph3--a`PcX8{
z72dmxHhsYV<UE0;*U-3pN5_1A=SZzpG4@05#i=zu9AoNYPwq*JFWZjwo7KM)GZB1e
zJ?FX<&Yvb84Z{z^@-O(pFIv<7*oVPwsZ;EUpWYoE>p?86DK!AD<e{8+uM)@f#joB%
zzg@={q~EMHAcgYbr(i>}Ur<+206&Pa5Lb-)Iof4Y7U%RGJx{;hpbZ{0A*a9?ux{NL
z6ZiY$y9BZRf&QMrI4rn;nua;V^3B9Kd|@zc>CsOoeuWM4AlD!PKVXobNzSY=Njvml
zL)fzk-qwY;!{BXGc<TUfBjK$tynX!AK))H#gD}_EAU(gM$9(lN;;v5O7Qok$I+@o5
z4OauM<kv~b7h0K~q@}i6OEgX^ec^3wc>8w|kN<*YF}yvO4ujpb(zc*Z_AW#ZzSoNi
zZ$HA@fRhGk_rxG~Irj33M%tWMb}yqeZflfZHb!}7^$XCcn({_zt~biMXrm0?ZIrKv
zjM5;%C>3t;-Xo*zL*vH$<hR94@(YdI9Nxy)LL)SQqs`F|Xx`m*Op<-cDCGmu^5cwB
zxhub~Zj>n6si+qAS24)#UwV1V`R#t9mz(f*%Q?NcAJ$9V9eVD6u9p%<&V7ttq6EEL
zOE1n2dYNFSmo@M<18r-0pyS>mI`Q7FlZ+)gaZA_92zcuaZ|~35%d4e&v0JT|`eKk5
zwE32fMhSwsk@yOaR+?nlS(D^A5qB&FbMY(wp|@WwQe+9debt4yIWa`P2dcPW<6pW^
z!$fTUb{4t6O!DRAc^8BckM<%4&>60FBsL6tFA?vLK?l{w7N^47_Ez5Fw`9!5f6*vn
znoPVB-VTMgAK`7i^~4zJDRN}HiU066Ndb6Un)rLy<D7$mJo8I*^lw$BP*c?!-cEtH
zk9WcSQ{)8S5F@7^s1<zbLu_&dx!i%%xwj-S<!+aVN#3Sr4JKM^rxf`2upRTJ0a^&&
zZpS_^U_N<bqx<+VheOH5FrP*--v^^ri?W|Ca5WL;g}{Ml<RvG=+v_gm9Xpd#-AbPF
z8Fh2SkmtlPm*H*FVZMKY`*U#Y&Tu%i6E=4wwNvo6(lq9DALbhSznn|1_c1lp#G!9+
zuA_;qRxeB5@*O(w7`ksGHD%-$#>4rQZtU+hHEu1aD|<qWo;Zysd6l;waL<XjO;O^y
z4-|Py9zOY)BBl4ywzO&KIf`_Gw>iV%Ci?a?zD)k@MseN1xKB1pAiV7WZ(ZT-Jb2q0
z-hPWUN#{0-yf>3mzz=eGrO1tvs?340GtspRVEcKVzpqYy0N&cr$GhO|8@BHYV5|>$
z0+@V)?;Rw+V7Gy~i+A|nKgcgIZobo(Ij_ijkV824m2>BP55`n5@sjdv-fYpWS$76+
zo70|;kJI+=Sm*VFc1OcH7Y&iH5@?X(#OdkNo%l9Y>E9yx;V!Sc@>xPHv<kd!4sRVN
zk(-0L_So-FA>{C3?sqp>HHiG)XyUCC&?LdE_kbHl>{@xYmTW7xpm*co!%OPLsQ+o_
zN!%FTX7^wn3%uP0Z<ElzXI_wJy-oZOHb(9VkzI^2>-}9_-C5HIgDSvH@8Z<HIHM7<
ztp{4eT<mIXw%^##<gW0l8acT$#6Fhg#U>Y_u@<HghaooF7~A{{PG6q|PhsG@^X!M?
zrNY~+&s+z*ee(|Pp(BIgXSrwSQnal*?EG{BJAH`v;q5kfdjNf0ncv$bkXt+p6X0iM
ze1$8-z)!Jl+r<6S_|B5!_#As_gI%2Gj!+5P&G+D~%Il2f<QKzev#Ics>xsmt@PQ+1
z(TPuC;fB@JGvN;%szDqb-X_A^A0D(Jyv;zvjX8~96+}(mBD36yBA$_tdk5iz_8Ucx
zr7QgjZ%5nH&&<sb=1Q&c#BZ0g?;YfCx3OQ2@fhB&i(?L5qJ3l0#c;Og6Jl7jf7O-L
zD&yCB^{2m=;IqQlZFR`al_A&jm$jkU)S>V=x)}blJ?}A}*WlkSC1>-Z596#l<0b;T
zU<XG_F`lpwhk6kE@`RH@yqWW=UK382XI?YjX1NfzVoU{GNB6?p#Q$i6XY7MEIgAhW
zADi1l><jM~$fSNFD@2^=(;&7t=;XsjV&l_@Yj-B@(Ux&on7%7Oj-Su1zSiVMXx^rb
zReZ9&RmnNnQ|mQUD~|oN^43c$mfl)f)L$!SbXqyOKuhh7R(7RmrNMKpc>UH&wQ}&G
ztxnF5*YW?(I=OjECoR6}q%sU~uZ(6Zgtmhj<9F&LX1Y$w4%bOrc>A%sPCn$*5iilo
zpi5fGT&9&5@HPS7ZoQz9v+&m66fA)?G*SfKuCJ>VPd<OrN6T7kt*pt`i%XzE2A(j;
zlg9?R0B<){FiKblqYUV26t^};Ih)@omEIbp09z8DEr7S@;cZ`ddmi5U#`B#U_zVw?
z@&b+99NzYWw@=~io~kC93vZ{w+pjz>LNo8dcer5IOKG_GgV=5q3|)<OeeqB)4#f>}
z-oYUA;B94i`-}J5r|D(XS#&A99lsr3;wv13xAyRMCcNDQZ<Ar^ejB~4{Xm<gpl=uJ
zWR*@Q^C#-$HNMG+eL8XNsFxa}@LM=;F1#%PZ^POdq?9e1#=|Hd#RUJ+-^8!`z}vpr
zCdn<VNI1NGJdijqwz?6#t=AQss8ahxY_~ML&49P>ti1g|UE5aHQ-x5^q@l(fwkpIl
z<Lj%ksTHsLpox1D=WKxfg|~a|DiTL5yZ8*^#uH&LHDW(Rk*pHbsiF0q;A;-rcK0@u
zWRL^cd!KlFJNQiea`R#8xt^2j|3%$h18UyULxbS$nl;2FlUUnC4NwJYflRji=LLCO
z72XCBlN~dSn%yzfa`U<?nl}o~+dPf>vcKray3|?J$FFS|BLAA>XSOGH)SEekJwC;J
z9|R94+(W|}(38|)Cnl4l+`>FX-`jVhX1h1_*`26`j^chEFxUSmY)04m5F57P^(q+A
z2j1RbA9LVwBHO8E<bVdj0giKx{deX(w2!C<g13POSm#9i_^2cI??T6B#d2>_csnDS
z_3SIna*Aytaocsxxesa@z7D=kT#6!pmMP-lu1L8n_<D;>^1{a?qg?TE(~aV1G|Kzt
zMrlDD=OA&EN8^o>97=3v4dZ<&WBQ^&&ZQWr4JG#EWRel^Hc@Zlf6~a8^rW9)^~%+X
zxWMiO<b~QLDRKbqd<Fhi{H;i#3iNj+7+Q?jF<Yrp<PplL@~|Q`kmM-Z5Zib^fE<IK
z9PShtP0YiV{_HYV6?GE#b_*Z|K1-F|)ORGIjUVE>HMxd<&QHvl*vTGvYY%Vl(v~*x
z*6}<34a^-=1ct)Y7d-l+X)`}kBS-ss!re!Feii>}m5%<XLGByhr9SmpNo>#Vv(FdQ
zF4$0~k5AMX8{Y%F*e#b>G<GbbJ#|5S(426@3EpnsNbL?xY6x$eP?OuzmfBw!=lvAM
zp$8YkrbmOYi|Ceh1<`AP<nS0rX^bg<czXkF^kpmTy-N%d-oC4j=7djit*DKIwbj{L
z!rNV*u#?Z$!jXJS;MVrMczI`doBIXZ{F?pn{2})DF1B_IHo0jk>#AYwDR}$q1F=h<
zYoBm#oYzb^o0ZI(BI3O<@HQIWdc#`{yqyR~KTXTiRL?iSNycMa#^t#x_%5)eegQNc
zWApS>*dK*n<M?|vpoih@bXXg^li$MIHPi7K{Hf1v$vMCT^%e65{vF?f=A4Zl9fN<<
zhy6QpP4wHt?aWg@YCrIoy@^v*qz0pQBz2s;AAW@Oe<`dX;asM{+ZD{e7v-ocryb4g
z<McS{Czi6`^{_M&tsl;@@JmMQrRL&1|9yQF7GA}dg|}DH*G*R8GY7!pqQnU{P)F)d
zf5Y2vrK!h*vjyR;4Uc2st$lgcIJlAra<_=(f4udEw@wk<3+OlT6+7lDylvQnW5e6^
z9bkR~SPE}v!P{o=_SQFxe7wPZ`pIGK!_W4{zq^-=fBl?mfwd3Vs=(W{r|k0$?U{v-
z@Qr)=vb}=0``~TaU~JD2a>D(|htPhDljwWC|C~NQSc~!DhQ@`rXBej~?1<xk36_@K
zv~swYR;I$+9c*F0H0TBx_%2vtQi3JrNU+>l!*fe8_gV~=1>U^&36_U2a^~J(F~1EK
zyT%&%7Op{)Yh+0at-PA1m6H)#sgIV8vuMR(j8?9*??AL`dIha~v(rjPc<TvouSaMk
zaayn}Tf(`pj{#LV4-c)_^d*n05QB!dv*E24eOm>6dkWq*gsJCS86~F~dFA&8)?XQ9
z-$R46Of<-l1$wzRL@)aGIw^cXD<`70Qkiqu$MNmhcggP>Dfm$%;jOgtd5M--2Iuru
zE9D31<T7#WsJeQIAE%cOs$Twv>!te|z05hNmrBWcYS{JSM4sBN9Q=p3sqfIB8G2~}
zZ`CtA#^|LLvEYUgdN~en3&Go6@OBxzEmKi1)8MTR-hL+DJ{{g#;H`0jPF@mcI1F!J
zc#z-q)5{-tTO&a)|4JG};B6P;^{=_U%uK!be$k5%%g)f5B=eI=9=$Qi)<-7UMa^_u
z>}>ogMFzs#y6`qCl3e#fmGyIIS@di@>}X@O>^B{H7TzYn*!Wu1#Z@6s4r5P~TR7L0
zx@1SzTzw=jeUG}U!-`CvrLfMP8t|d$uMRvf#yY5vCMgGR*TCDgTZnaTrH%~VrZ<B9
zl@#<bbvw6A+^+)NP2R66yj{PLbzbP(&)2BehPTz>Z3F8ZDn)&EW$x{a{@l`xxKTIO
zMEhCf3cTF{Z#yJYpTj(Q&79g>3qKX^-hj6j?4*S`bsxWYCePb2-vcu&avXoRE&O|a
zku}%wb~-W89HEwr^Vl7R*6+^RYxvq9-ab7|40Jtfa3*lyGGeW-FOk0_&)oPHTFsZ5
zAY$8I*nkqu+ld^%FSacU-j0N~PvLDBc-x>F>%RW69_tj^7QRlHZ{|Nsu}3`L96?Oo
zjk@yd@C)9yO;zx#SW8B(!US(`!P_prCUL;mtA;<gR58l>21aRCpIQTWdkfx{hPRtc
zMtOY1AglET8BRQ@j~8RGltI${4YH`SNiJS8$<SVkq`}*IYZUnkZ@0o*M|ise-adx6
zdwwZ0gMMw0gN8<1*MPUB;ca7h`vBe!gtw0HHjZ(!R!dDje4PnzUwadS9!>2dHI=UL
z)~K+yekFB&FgIvBwf$G9ZU4hqqTVWozO!FUn_$BRy~Ji>*POB4JG1dkzFOo#LGos_
zVfY*Bub;5y=`lHjx%gE+_&;rk=NHHJ!Po3KVj-JZdp4VxwL&}_Ti<LFbqIs_-;`$D
z?>$4649=x6{^H#K(DlR&PQjB-k<=(<lOv|j23n}S(^H!_g>zv4HwJL5-rV1%EBA2i
zfNxcg^?;7t6R#cL8N`?)*3l5&j@w3^&uzw2(LAR1gtxol?JQXP$%A9S_}=K<D0JYI
zL#%5?J8fBq{yxFjOryRU-ql4jwL3-)54QIdyj|Ogd@6o`53$_0DR3Ws?UO-W-vw%)
zFLHkpcsm!~?%=t41ic%BU$BQ9F5G>KKE6Ga-(brpVb?cd-xt8!vJT7x{FmqWE`9Hi
zL&4A49K*FTXU67dt;=faoQQ4CzYNcqPXme1b{@k2Vvl0XPb79UpBj1m%V5sknf->1
zrfvbhvH)CqMW5|~s|Ed-qwu!f0Q~kLW~rjX2ii#uD7^K?Z&sh6duhjOX4)2i${VgV
zZV@6o#^u$#)zMQMwE&LIfXfQ&<>uoX!1uOj+6o)-VNI~FA6l~`8^7N}4q|O8;~)Tk
zmi^WzcC;dw|EBpwe(D?Zn0Y+_AN?x2cLZ^(hU4&wH{hr3;hqZ2OZ_i=UmKVWZ>wM%
zuHg&pfVWk!4avl_JQKN3?LF=}ki-3bu2E~TjeGFSv&a+1M{Xl(F>1nJ{DI5x*1F~)
zls12d@8tA>*bL+9Xbudgtq0uV9s<`b@~;=y;f`+qKy4&rdKB$>5ZmC5y)swEHeoZ)
z(jG-<*U0SwqB$ABI17+DDFG7kB0x$F36##psgluYnvA+RO<Hvbl7R<<q-D)u)+z)`
z6R%)#9TqGd^}*r|Z%f15wv9B>aW1i8@~;upJUtt$m71YiahVNwO=#%RS{aOn-p~+j
z`$!`vPibV-HjUKC7cBd}1WBnr!7_1SuyjccmJTH~5;t5UX8eFmtyX-h>10k*gM1*4
zdrD)FZvh4=GS+|%HOSe?)Ly~iV+jVCyul#l9f>zL)${)hI(bu8C%VU4$-fG{j%M$K
zPch3u%e_}MvhbZot~Q0Y;adLpn$O;#y$9>0*&Lk=IjobvdYz~mozyn#_-`+r%uLeB
zqJnxUyIC)d3K&?+V-TO8@HAb|y=aMhU(t*4lwRs>)yv9dda>!RmriB$a;mIe8rRWF
zEoU^cJ>0#`emF+GV4eJC{}Xu(CSH)-i(|vvVV(3c*av3MVV_C(40hC?xf<k0A-!Dc
z!13o3OW=5K`g7eSjFOLB-YRNmCcxYAuy5Q~ll<JyJwV{CExfHUpSojgl07W>JchM9
zuyrl=-W`n<TbsHjcxzJ{HlHTfjV_LPr^r3NlK^k4VWSHesl6hu`8-gO9ju4y25)20
zuN!Qr*S<iG`>;vc@1jn2Gw}gxc6R17N!20bwDET|)P4mO)5}HtpWehyW2x8jSOxR;
zupTLu^}CO$i-otz%pdnsFcsdGbRoven4i@GJ&Wz#fgSwO4~?{)HCYF!flb0@VxwE&
zziz9>+^$bPrZqK*Es0I>d}Uc;lZ^L<*w=BzxPRC~V%BH4kEk~s6z-D+_beBw9hyg6
z6n5lLFIJrEJ@ZA8=w#M&lS{SgH|x46m!Em-qQ(*%=v)Jz-ky9V+s<;VCu2({UeMNo
zdR^kETDEyrxE}~P>)+wT+OSdnM%I}VKe*_{dhjLGC@n_=wkM8E`xhog&=lTk;jR60
zle8Q_yS+fm5-%Q&wk>aGlmo?#vV^$C2I>pnW*ek#hC#L*GH`z`gP45C0rfI)k0R<Q
z;O+N#gQT1%4nkgKSFBOW{4+{ueUpr&2C#4vwS2@QQbrKBCI_FD$?@Usg(B2Q{*Skn
z;q5JWI}P4i*V{$I+bj5ERjHr5Z?DQmYVfbUWL(`*#EJ7yA?^{zer|sy=UkawFh1Q)
z@&=7d<3nxanz>djkK2lK-;}A?GJKi~pJ4%8(tp^rr`S_CJCMF?hmG`rw?BsFad0#I
zbGS|05+nG$jM(K&?%M)yjquhE-mVzPnsQ<l=SmPeFmry*@L}L>vmdNa_zxf8HU0we
z@6#XgA$SaY%lPJT>o@LW_m_EK$D00TaAY3q(yp-1{0DgsbkY(yoTX(=7<SYHTWtp)
z%3ef6mf;@1m01shRtaiMJlCDq?Wi|`w;SQ@V0hc4HZhP@XxWq0*TUP&*k#{a)EI20
zHYbccr<wax#)e4T73#FnP#J6=*#`05i_fs<=qFcKbol`GA4vYy1mC03M#O|GvUOU*
z+OCqsdE<$%Oh>oDTU&UWS&99^+gTrpCBT5;_>Xhp$Fjx56lhbgNaFgi_w;)13(q{d
zWY1c!JN(B>3TyP3g9WbUt>wPrX^}|!$@Mz<953dTGyYIv{0+wU`B-Ww>DT{gv!!iV
zGtr*=xK8B0uJldC1FZFiw{HtmC*cNTZd;@`zwOJpErz#`CX=flOFq<x`a)lr3vY|k
z9<?^nzPtHvlMURb7TykkMV+gl%Q>bk+*%56_ru%n@b+W@i`2$e45zQU-=tRW8FTL+
zHS>J-!-IJ9Sn4a`?L%@qN0TiQhF|w7m;X^`KYsAFB-ivfo_h!Koh1>(r54~1(pFdD
ztu4IWMSBIp+i~!=Df=9gO^wA?vz$1FkBlx@o59*tj@gd9(8(-p$anPqM|{7p#Oi+Y
zejWU7`oKF2zQWsl*q?nIPhos7{KHtsRyB9yzTvcM=Eea23n@T`!`rIxHa$H+l40&Y
zww2HGHoI2=(*0Y2JS`k30ZxIkyFs9IY!@i^x&}(=_JOj*HITZd0Qs{oKyDiYWT_=U
zd|~*jrvXySF;Loe50oLJ17+ykK)F0+s<bT}B%MnHkyi_noxg*m&Bq`)=@~4c^Mj?z
z#bB9BTzTGjjm&6{zU@f7H9#kiX6ab}2tQMGasZw9oBXH_&3W*jUV>u{)P@_R^J-X(
z-h4`Z@86y}=?-sCKhVmx6<TpcCvSuI26$WPv4;DzXe6bDMzV8)WqoF_O!<U<Z;rNJ
zppn2M8gaX)krU|Pz%p8~_0&mqjynT>+HTZKC~SRYr<3%8TJhVgkv-8GiDYZOMI*zg
zv35A15yK&kv^uJhJMkJBTT&|r(bB<aes01cYjd>nFi6M!vvjhQICASg<hgt3q@s^b
zHjUQFPBiaQboSZ#Itg2(lY7hfJcI8yBE~!&j-}|OUp)i&lQhVv`9`^4)Ffe3Ofs4^
zA#c9$xJ{Aj@HWk$N_HrDSYm<kFf|*SwC$F{|6H<W7q-me`H>9xivE1bS}FrN_QM5o
z!PFraT8kD!A7vS+y@9tb@U|ALty>eVnrkALLj5+pO`&GCC%V=e<MtvJ`k5NGwgU~~
z1#jEeC*BTkKa+2$xtBgek0h*Qp5PlFWWHZv4)uh$XYpN+6vMZy$ebXSc)1~S1>W9<
zw?p8qzAt&4an!qpQ^UMEPY>BJhG$fx#)EOZ5Z;c1w+-O!40!t<-_pGi+J*hi&cQyP
zhJEn%U<bHfm$gd?<TTOx`8pBnw&gyg_!zldd(LI*gA2iomgJn#p)S3ttsl(#^pUFg
zX;mp>!~agf*^O+zWyy(O=ia5)(0MoUXYR2+i|gw`yncWuI=(GjCmvf=So79`x^nh8
zqB+NOBesCA<2*=_MC#nzp#R6<FU=36?%<_SDsD7NL3sNQU#bzj9Sd*Qz}rN48$^7g
zz(L0Nc!QYh!d_x3+6_7}7u3nCGCGN1oT_KFvgw3YoYrZH<!Slf8?7{<j?Z;6wf5B2
zsVeKn;AA&=+nj5C4|8{xB;L>XXjv8?4d#}_4`WX<3}0j`KAN?DdO5uP0dI@K+htc2
zIdfH!)w$HC+mY*a#fJ;VpTMX12V*ZTru`Pdy;al@?1gdXd7ZAx6WT28A=(Jmu4N0l
zPs|xxm5knMN$jv6y#3pS7(Tr9xXwDj9>gUIvxfT>_3`|^7<T^|`(B41xy&DbiN3Ms
zreD#2TWQlW@U||zjfc1YGGQ&h8}<SX%vP3uv3z4)6ujLJZ~wNWc5ef<*$=oE&M#QP
zdAPybGLu>BG?=m02cM4DZ_vD@u;&w|QqyN)EHVaLZ$uNnu*kh)u*06XBN}F@D>>U%
z)D8?KM(a<k1*T+QBDNS!t!pUt^6+*#ygdbP=fPPc+k_;(!($(q`{Xw3lHZc|L<?8?
zomVeg%7J?TR-k6KI6C(sx^E%lJs4JVjDsCvJ7fQo12KAH1vg<!`Mb0syfwiO(_(Vx
z@b>K@)_}v?$o2R+@OC}Cy#a3x%s=bqg+85qiF?K!W^5a{Pb_hYj!UUa$A9vVBEGSf
z{QV)?|2}K<8Q+1lYx^eDTaL}+ZSFp5?&0l(Li|4s{<6-AwyBEF#D_k~V<<d)%&`^b
zWE$+dG?@BymHasIsNtW;>%51jFZr)>ShSIA$tuJ(9%o%1ZNCs(U@1v_1pAQt2-b40
zO>a_55Bsje+i$SZ%G)yZ&$;EWel0#mH1&Bq@#}J_RisT0!qvVP@ZXNWW%{iuynR9+
zX5bh2!NYQ}cn7|qmbQ8@l6FHc`(u+rEabjA;?Gs3UCQIDH^uMMlb2de&geKfskg*f
zz7VT}x6!VQk^1-<#AQk{Zhek&|G(X|X$=1P4&r36d&VZ>YU{aIaWtHb36RBG0;JDo
z7`Z7x+HMPw7P|xFI1JX_43M$U0;CAMeFATr!P}bfHnwe`wC@r~{3lQfI|s^y*8$Rd
zM}S<I79g|C0o>0FKEqohylo9{7r|Sb8G#agAyA(AO_ky?Q)TMtX)?$qNd9dNk}I8q
z<@ms0xiU_}|7&W<_rl0#T6qU=XTsa88RUz1>m--F=i3V8M&WI(1L(_Fdg(lyI%d}0
zTql+r${L#2@OE=O>X0^SW#>$-93o~MV9?6%?pm4ONGlED?V^bqiHEQ0?}Nn|O<jq&
z`A>BETZ=}nz{+ay_6l)zKlJa3!a7NLsQZ7tI${Kx^QA^QoPn`hH8NL+Z!iS5vYx3_
zHI3xhXr%bpV5#;!Sc?1&mXSsItb#_uR%)oR)=CJxon2KYQ@BR2{8}mPs+DqGby8!O
zMzrlTauhyqaM4PvKRNSlS}F2TD+_(OZt~&<;cdm4<jJq<WnOKAtSe`f!h4Leya+ke
z2_}i5t~wUpUeYKsbBrRNe2D)J!7dJ_ehE9fWr-rCj=>e;iv!@TFT8C7Z#To+m8*z5
z5}OV<LH#o|+c{aRmx8w&y1-eqm}h%*3~}OJ)!6S3G!wk73U4>U+dpef(x@Kmyr`X>
zA7PLa=yU(AIw@aU&ws)Zn~%{+pH;*R$oEI|<XG6b_007H@U}i<doA;+6TI~<Po9Q3
z+P?vs2j0fP+u^Wxb0f3Vi^i^^nWAl3Q*;lTQ4;$FZ@sZOwy^eBbL<h%d*UM(tiYV<
zN_~Yh^8^2UYI9<d_~ibr@WX4Pk>G9p5Nvis>pEfT&C-ZP!rKY>908vdS<{$$ZFpOw
zw<-mP5KD!(|AL6i!=F3O)MH;Je(p|8VL$7g=dcdyD)+mDx7K^a6vnQ9;@C~$?Ra?G
z!(EjRp^A*Kr=1I-!H5g^6;))hD{cRadM9|>tStEe>dv1g8fEW2gVfqZ9sVrV*|D}R
z2mNaMf-y}E!J<ie{x?-m-dZo=l^OFna5hpWe{89jxS^Hs-CAiw&BZEsJAW-}>=^SQ
z&xzw))5!f4jkGwSk^K0Ct(R%23!rX-{EA~6;w#GyvcpM{Dy(7eaZiz-Xx@p|HsJc-
z5kt0WUkkjo>f4966$xYfyoYbIi&M+og?07#b61E7RGAMG*K=<fm{?&Exqo<DXAkQi
z&XJqA!S<fotfH*P9>iMg1H>j_?-JO$qCZ*#zTNoCcv#2!wLn<w!{@Z)7<l^uZpOpg
zf$;WE1CIHKe8vMs#&NE>jNvzo<;+Ayo~%$<i=lAO6zb;bQ?)2{;9N%&&QU*#jq?dT
z&KeB#?x1eeGhhR|aSq?$?fTmApgVTc8+$zvzH&a5vH98X_7L_wDv)@KiUyiRJa!EX
zxk7z9I%--Ov?aXV<j6Q=>`kjq9MdI_zq`=9_Tw3&iP(KukbW7K^W8=0zFj-WPh2Ly
z32!s6v3?ca-u)kM3;rhWOKy7EYnXh2d%WzX_Gbm_U*Ptz(cCX>AZuMZ;a4)=SHP&f
z`MF;rZ1IJ+RpD&~c<Tah`&k(RUt*S`w`P+Mt%80lMcfzOK1M4ZIfF(snq^fPe9<HP
zXWN>*`z(B0f`0;U6D@gpDBFWD7v4rOzDuMr4?p6U!J^Xv+zU9CH3s)t-;V};?7)15
zw|_WJzCY+@=JHl}_}6Ta)%&?G%2j;nLFC=hvo2G}^=855cX>I5pTy&H(CWFwfnDKk
zcQ5i0%;WDbi7RDr&y~x>H{ux|@b(zI&4jll;O#%YmpF#;qvcxXz)$+dF_On#_!7SO
zQfB=6BCPv!CJ)>m8)L%n;ClsWpVABYF6`Y#+Z4Bf{pi(+Y3v_H79C3d%bl@W2|HDs
z`|`otJsn`|5b8cpai3av8wYPE{bp?=ydCdGzN<9Xj~@PU1T7A4tHj`wu&?s)R$;8I
zg|{2J`OBlQ0EuH;wE}Is5lsqjkM9hSZ|L4q@c~ljPJp!e5+DVN1WE;X`>1{(>%0SH
zSI0nk;ua{|Yyzbmy4aQ2^s0aW>EQvlC$cYi+oE!y<Zm1(rG~M=+e-<7az#7!|5!Tf
zxTx0eZQGt>9J`O*-5s|rC?KE$O2aV4Fhz^q-QC^Y-Cf7-j$_BM+xPPQ{{GmXA%~g0
zH_x@!z3#OhT{~k}$^H(lF5%R?c}_ja@1joTRE1X#8Am(h-^ii=PTJLQj$H@(+x4NN
zT_q~P>AZI4=EiBwnK%`A6es&6yOO;UH5cAag15)fS~Dk7W385;CDd&n^k+?B-O0@1
z=NYet@U}=cJXjswI#11cF^nw*Yj2ZRJ8n{o?&8+WC~|AKo4-N4g5mY0`|)z1J-7bJ
zdUX%Y{JdMW;jQ;#)~%lO48z-k@U}9%y$Ek#zjeuvygBbTm&WmVI*tYpo%(O4L(SGZ
zRP?f2<6(AgcwUVBBNV=m{Opnu-nw?X^{7O=#@XUk2i|6dxBJO~`(8;<ml}!cFosy;
zM52zNF>i;rz3_v*NKdawkrejb;t}668*OWRc&ZM-+l{bxatm@2n7Pphy)BGh6W0Gm
zbMUa9M=QQTT@T(KX06HuZ)?#9H3Z&PW$oXK7y1~v)?Pf>_cp=N3HYSDB&#c$_2>W4
zt;wf{z}va-c3YklZ9)S-5>4H@d!h==Oi-oVc*(XUXns(FO5w9ht4dwq6aKppVg`DO
z3mcO(co;hA+9c(;o}`}-lho>El7fHnST0!|Jd%~ZDmpbWc_S-p{{sB*O_G$cPm;2)
z;N#Uv%1OT00*!St#{`aC9DVs(sohC>i-&#v9lZ7U;<BzxQP7}Nb!M$z{h9T+G3yJw
z9RqJqa~>k$ZF|0daRRwInn9q8+QkF31oHAK_|C`Yrk+9n+vStVnoBO~gtxChCu<{K
zyoau29eIp)%Nl<X4t%|wqA}-_^$Z{R(=T|;`Z5;<-u8jF1>x<6Ow_jd_|bEE&#1XB
zXXeTsuHm}f)NaowY1TsC*8r|Bye$H6Pu@w?nJJ0d25+y!+sE+M0B={r+v+_Nbf9;<
zT5gEft@-iVOHP}KI?kLf+&9PF+VT%R40ziR-lpH;*7I3z6@a&yqTTvc-o^Z8mkN5g
zG-MLGQdPGOn%$ZUlW(If#q<36P9^HbTf7Whmt(x2)`e2^7%ipdeeN|_x0yV6F!4xl
zuFoI+dTKmO`!tBWnH;p`1bPftps&E&erKr*6N^N^&jG?C*OM6QI(u{$#kVHrWZ-$J
z&Kg>adc#uk+Bd{VABmUxlaIEzgdPvK7oZ&vhw+Cw$HUQKZtaJ+__|A@OZ11g;jUCo
zNn`F7anmZU#iDHJ54F%j;jIzgnt5$);cYwe->&(|KbkYE-3oUX5SQH|w_|;O4rklL
zTOS@P!QV~o$ctdcpoZjX<%kLX<Pl!=-_2{!#B5H!w>ud=WJf24rLWLZ*OtNa3UfQd
z+w#PMEBT$@GgMReq2~~1^q5PIHUVvR2(h3W%`$=Yn;LZnzF&cLJM(VxfArT&Fm}~Y
z{7~eAZ$g-RO%C|2EcIX*Tbf$rV*Y0RtmMK!&@5k2PlUGx;O)$X#0YS9NME>TCFcqt
zPN;#mi2tKqe=$px*Rqb7Ck)?vmJWCZ3Z~HaLT`5s{22K54xObhvNG{_YO2z`;GQkV
z|49#O57ZZmb3N_wHaooS>Y(>}beh89?PIRpGp^@hI39b4T<!z047`ox{B-5DblQxz
zeUZnG)I^i<OQAFGc*W~^O0D@idTw>TUWoTcew%WT>(KdsuebQWa24KOI>NQ${QL`V
zOa9>AC_xNEExI8w(GqxDF5!Q(@P|(&Z-t{98Zy7O2ih;!cg#HUVR&1N*D@xSb7R9x
zFb4L*$*DQ<N@k^o%j+KkZwJ8JT`)2Sye$V~-{d1U;eLt!NZj2zO?|t=WS+A!_rRmi
zcp!-pe{f&yAzmxBli5V8m<z*kZh4w=!`q*W@p!@83Uko#r{hKF9;N0jqvTg7N-=q(
z6wxtCdmlxqYH*CQ9g5N4Ct{R*EJiiY#3=S*j9TS3>PZEoYBn*dklU#1f1A))O|rf;
z(GP7=#)McM`w*+k^=-^xjMevXv6?fH?$A}S%GNDblfGKCJDo+{OIg&XxrLbk7VR11
zRN!|f`msfUWz1^6*`#B=OuApzq_MY+%Dm2~6}^qx)WxXH*^J7T#Yo<2($$qFH9KRH
z?LCh>EZVv~R)c=U>Gp7^wi#g=9;uV|c-=eh)WTU#Z5`p%&>l|qgK?^p2W))k(s}C8
z4ePj+o}VeZ$%&@wRLYzG)syGI+c|w%$NX8xvchh7`{piM^EtO#j^=MPBL*NII9}DE
zRg2)iIgY-8IJ6y`h7q5*vc(Zo#_2gaK%RmQY61@3Tj$V(6Alg8>(KiIhsyBznp#dx
zN7Me^)UAzU-D>l}&FsZ^)ne_8`5Le0=*%h9eXCLr?G0~NP}eP52>%rPxj~Ooq%~Q=
z`<O+7&)V&iqFkM*2lDZN=gIoyN6zGf2kT>k+K^{{h21e|?N3{f|6fW_V0-dZ@~t`K
zTMzL+ZP`Je5jEJw_|MbO7Wz<MF0`5aIx<lomM6+wGFfHe?Mrx@xhHGyg(SJpCCODf
zNpZR0DBkK-@b;dS=Uzo#i!OdE0Dn4pR3yB8@|2#iCHU9you7mWZL5;QXGHhB=~f?j
z;JeU`=HpgMUo`odZZ#n`T98Oj_CS|v<Z<a%vQsO#hSvg|D%XvVyEs+G->EsiPWndp
ze1=oIt~r(Wy;B|V26z^AsR`HjBp$!_33&bHxHNbL`omV2X6$rnR3EoGcSur8)|Qq7
z@fgj<_cbP2ZQ*TkJMVcPzI|e})*0f}BX_*+@_J*?9yS}P*Tcs(@YdBKSts!d4-7|7
z@13k{6O-9<E?Fgh(z6b49}wT{&zH&^;#6(RKwb!M)8K8yLp0u@DVkW1c~R6UN}~xE
z-Wad<Z{zhDjd(vEzDoEH=JFc$qSZu<<Zt6c=oo|VFeF}O%<=4*9j}lej`j2y@H3mZ
zXMRyTDarTKU3KeI7Ps!UajNxhhs;|YIzYTP<%~n4-Z}KPl~a%JIyHh?PTt8bdW77X
z8iB?FZ=0e^&n|&4A}vYp7bodUXL{|HCsS8SR&z6+>eLkV9zic1Jz&H9nbAQ^(QXO#
z-GA`Pqf0-EMt?__Y|Z+-9^TG@x9{QY?JUfjY(kBjHS59wau)KfCoS;rz}vE{c}*vy
z?RqfhssXjJQp{!fg@67ae)%{wd2-u~XxLTJWU|8BOYrtHy#09u?^g;sIWcB-besru
zhehx<5Z-pINnaDZ?E!Ct;O)oF^c3IkshB@mGyf)cu7al;c1(^+)7*Zn(c`G;u4K=v
zd#tDAYx!9>I}L=5{mH5Nz>{8hhd3UNM@J-=InO#?-i2P7jeHee{O;XeL7saHF4lmH
zUGI>0^1e<T<oQ{9Jr<GU%!IN3$J>3r<UpN!YB%rg8XEocJ*m3NYd;&#o}~-WOSYiJ
z>_=<Bb1)OFcho67<kT>F{zFVdt~sqVubJ0ga7>yUTbbK-jk&LP@kj9bOTwvhm(%2Z
zmRjUK{`LyieExo;6!LlQ_vOU07xA|b&qsWii5L>zI)m}uabK3I$lR2qRONk!*Dep|
z3=QM=tiU}>(3Sby`QT8drp&_gBUaePxf+2tk#kjH4=lV*eH`9;5x<P9M;#X4W`no&
z&~}ft<Q(CR+K86i;uh!b1pRp&H>m~v;QTE<MowG{&rw;lt~_X6@V3qkczYO6#9nF<
z@V4;(@%B(tJd4@T!r)p_;xvzm)E>Bg1xMo7;Qpz>Yg-#kt<_1+3F{&op%;_mj)1qF
z;cW_>?EZ}N&vh*q${aqfMYlHOto4aSD*omuI(c3g3va{W?Ok}=8{THJ;z1#|AI<ZI
zCKFf8#yd8T`vOkagSGQH&ucl)*ZKSt*Zk)KyaKa$-J`f)0;81H0=~jq%bQ4L+sF|Y
zse>Mox|}ys*1q97lsihD>&B?o?HFdd8ubsn{nFT|{~ShHt{K(+g;BY_8nxPLQlWoL
z`s1NVhfkVRbEZlCVDZ*{COvv;Vm~Rfu6vv1anh{+id*nrS?EQ!=-VF_6`2lCZ<+Nh
zpILWTn-rO5VowVbdpR4ma*a{<dl{J{Wt8`yM%6XO=yJtq4ZRaBpTj0)erHnM@@6&o
z+pKeS%?gP#%W}i4?#(PJ^^b*|+e&|btUAH_mUCj&E8eEW0}i<B(z4DjX5zTC_n1?!
zHagXOjgvi&o%*uGsYlbD`Z$368%~!R=~Q(%?ilLSs@hInAh*tx&91ul;cUe?-lI*^
zTyWNEQ#&;J25)UL!P_Kwo2w!9ZFGRW{LJFL)VnL%)ovJhISg(b<sfHrsMH;&j*N8?
zQ@b_o6+UlxTa>l4C2Q`T2sA^q*!899;cbfF8KxzY%Rj+4)1_3hjPUjw`Dk7kn}^<z
z9sKXyeUq$vcwNuK+ZEpE?(bmVRCMe5c(V#qGyWW}PS@hK_5i-N)$zKzj>j$W`gDP{
znjAD+C4L^>W{*!$BiP$GkQ@)**1bf|cagrR?etmKLf@{MsMK<as+x}8C4TNiw|J$(
z+v&gg+6BG_!gjo1llZ(Myj}SyUVG-@qeE{vOdK=1O1u(aS^gVn2Ak0a$iFrWp)bsi
z@2;C$r5d<(?}$@Rxeh&gJM}insny+`ih#8Zx;m8^-rn_dGV9B!BJlR+e5bzObm}S|
zyu9$XU2bagRb1><=+aksdwRM{N0$-XY;kEG$I&q^dEtxydd;N{3FJxi)$}EQz7vLy
z+lLy0A3i_o?;p_qU*w9HXIXTI7UYf#;x%~_c{|tSucvqh?k8z}Sh8-w+gM_tB;uH;
z-pqeOFQ|Ek*)cHI`71@|)1_)LIqr92n<@iSbc%eU54_Dwy`%uwp*MQ)F?gGTM*QQ|
z|9E?OSG=CUQq!<_y|Kc2diZ8>o`1sKsql6MpI0D%y+cjn?=rj}Gxu>5r^;`1s4THy
zfn5$&fw!HvIJ9@QLs{W%(p!f%Hg>AoPPC=gF4c#(y}0k@5bJq0i&qDD>xXaP+JE@-
zmlAvUCd)gG9&lo~!tgc@-oAjhd(bwbsVy|M<D*-gsu}S1ZUOQEA9C5g)WXr1W2xaj
zIgJMSnsqxfepc4A+(G16<R49a$Q3<Vry}u;TUe{e_59%NCw|6K67SVF^5hM8>EZW4
z>ck(up%=j0uwCfi@yr8>MYFJ^>YFK5QDMyC>V*$sU8)jbmS<M-TR3KF(Njmgs7rT1
zQ*MPej%RIYQ|6^Lp`Omii}<>yH(&3HSB%HX>Z#3qe*nyzGl6}&=b%HMWEMyx*4HFh
zo5su*c$@nGxei>o;zU=bhB^uFz*;oX_i&?W4m9TPXwUF=_G@zJXn0s1?(%-7-ezAk
z*tu~BaS1WW1YT3t$MkLCtN7i&H45%N`<r<&=mZZp@p|BGJiP76?+k;tHQ=oc=G7#I
zS<dHm_<9@~z=KbmANZSr^I0(jPZ>vwD@_af<7<L91K{BdIN$|uGZB{z#!pwV1o1Y!
z{Y7k;1aF&lXa3I#beGc9!V}TF;O%DW2}3zAEBfF~fVa~RQJen{F9upgDHxOj(~iK~
z%P?oeM%a6mzfZ2%0p6ysi-rtu|J?vX58=@`MQn0~8o~qk3vXwl%P)_^>(zrEZF1Uw
z+o0RS+a{dXR^*;t;NeK}<}`S_dPtg1SHs)(yr-V-hJjqC)5HQ7Ial|p(Z}e)IYVp7
z!}-cs0S3d{ZDFvNdn#%Mc{sf71aH%KLzkcyJYy2CoBaGy9Gvb-+|Y_RuQ7FuTHK@M
z@j#Y^nR(zb@mYV)ai4wc#YQd4yKb6>!rN7FcRuHL#vtMwc=`*rw%{BWhqr++);R|a
zfH-LUFqq4c#Bt|bfFd3RFbgk0gNg^r>=`KAut0sz7sRaMAkB>o(r|Z>)~5t9!!1Zb
zt%CIGd!Rb(4b<1(fqL5@P}MU9s?41Ld5#QFQ%8V;7X&D$J5Uw8g5=S=ySCm9rUn%v
zSHDo*yA!I>p<ybRK3ugvhil&42>mtRpqOz+`j(^B*CSfx?a`{+GDcfMj0zrZ)aL<4
zZH+amZ#$zNk!y!NiDFMH1D+s*CLK1Y&pf_<+N6-NChZz&()iCN-PmVVURmg8v1noo
zEA@1%PCl@z%Y<0PC)o6BpF@wxD`$}}-Gi?Y<hEDIZEvl1Dw5Byz}q_0oLbzBytl1`
zz5~0mcD8F{PP_KMiPP7A<77P_r|e(whaJUN21~o!`M9P{Z*Ewbb86Mx_OZIqH&#2B
z#cKH<Hm&xzX)3%8tY%lep?2LU;LzF#ho%p4Fh|#+rL2FB{!Z;L;?ghJ+nO40jy>`6
zfVWO^(Gw+^P1Ph(9fK0lz3A($N6%1sKA)7Rkh_U$-zZ6a3Mb2tKI40*lQm>Dc^ZAm
zSK#6!c)bYE*~rQX+KP{CI%|5%%kirCJAZ>68z;mo5e|N)4mc7&*Uk#m%yG@+hqsUC
zC+G|POmq0VC-6_dZ%JP=M-b0DtqK}4{<XtD6Z8kX^{vbEk+*IwlpxQN30e$uH^SRO
z4d~4xXUl&czxs51=j3VG{Nr`GG9SaT1N2?*#a}iR-hPF*f5F?{@U}L*T?udR!P{49
zPPIsIvZs|(*J15*So@LOc0If`c5>>dldr9C>JNB(8s6Uf<y3Ka+qkkzb$rR+qFjnf
zb;(Vx?1HuZIC776vG=P>zO!AbmfNjB{O9-PCa5yKS}AzbFZssnLgsk6Kf5(42XR7~
zc-<u5HIqM@rjlQB9^a*KPOITxg17tO?N?%;1@N{3ysfq!&-TR>HToBgj=Hx6-X`HU
zD@9&g65e+Ek*qI+5>zcQUI$@lgZOy8eHO2R84@%D-nJpXeYX>T-5mH1Z-2nsLh!XK
zKYJYBE^8aFECbv+b_`DfxqQ=k4z;DPpi(7=?*C_3t_OB~+vHH42@cI0?@-QF4h5ca
z;16`D8ux#RUQX?K<kX>|#3JzaHs1X{MdLLS#@D5f-gcE9xuhgL2uN0=Dam?;H*ZEv
ziUz{k+($XkIUe$!Qi`XNtHQi;^nl~5+w+S&zBlXQEP9Am;rV90ntqo#AAgaDl%fXa
z#o7dK%X`x!U5T|hKlQQPc!hJ4f2E_w{tVwS3^DM#W5$x(C7=lhQzxemaDqDW8~DD)
z4ddZ#do%SB9v=jysx2JxpO`AYuP~U`un*ogfwvi9>XdfmwXiJ##@>asw|K4&a5aot
zeRVJD)bREvkFDzCmHRJMv(xZb!NY#=cGos))U3C=yYX7dX@-p>2i{MO4^L0&?)Y2{
zXt`E&P1w6KmHnWG)06Uz8t^05>(lsWBJtU<h6ljJQP-$%!;5S4$h`-W^YePz942nq
z$;?=ITX7RPIv%j0u%j#Qc?w(^-UzPny6QAwjy$jXCpo+4aQtj=_#MYU&Q;sb)X&S1
z(|V-o4_KRlTr3MroW*<WfLHRj=3a0MzVyPxDa7W%?~}C<|3_31bAjkVi-fl$gScPu
z%t!uBJ<63TZ|d~#3sKi?*;6ICo^uxS+Bpy1?-Gw>CU1kcY5bkeuw*d2?E!Cf0lx*D
z{8^a&E9&5T<UDqv=HF~Zsv4s8*ItU=k0$Y9F02d1i_CfZ&N<u(Z|gNdPtS`coEtxT
zcC>n)XFR+ucbIccoHOBXa%v;J{Z*+i=OE62x9i|-`O56W4sS1W&f3FUXAxqeD%5)`
zlY8=bye&OTiD<^X*n@5awd-|g8u0e)XzCgfTo>ZL8y(WLf%|Vq7QFBLerb5yE*Coa
zX|Dff`Ze~_PYG`i)=87Wf*;IEUp>6N+MnEZI$H63SPO46tw^KBM$Le8vSKJY1CQG|
z8u9VvZ2>Y}36LMW?U*4@*M0>kW0pWgTnJRpz#v7z+d0-C&5I3E_0S;oZWDxmE>OpJ
z2kJyxpbEg-C+P!~8Qy*j3{Z|--Bc}0SB=imMXSH~tF}j11-Sw=$q=Y~UjsFMbC8yg
z>#j0y;cof}rNG;tGYtAQ%qSbYE!rkpJJLlf=v|av{1>H~4Wro;9?kn)q=xQ})XHa(
znsYZwshJIG-QJ+uV+=a-&Y;<EVq|V%lsm?#jJ1u52rz0v4U_t8Fl%W!i|QII?8$9W
z3VQaaLvh-7#Gyjusj={OHhf*a(W#a2_R>nHCNzb!(`@uo#j5V@SpB>js}YxCb>>^F
zHV(CEA|A39H*Fe_9IHb-U;7^xwM}Q0Er(V4OIg*iv{j1^Sv4{sRxf(RvX^J9_QTtV
zP@DSV;jT>X`>mH<CyTNEwQ=ZGM~A#0IMk=QQx)Ouq+;}2?RVi#jMw>p@Oz;r7a2w$
z5&6c5I|<s3PkJ4^Roz7Gq_^r$b&lfrqw#1hN_Ok%X}6leue*G2`gONv>~pK`4m@bz
z+!|XQp5oIUVTjkGGpyI}HVEE6A(t*d4mA<pUVe?{TRK5+(442bVe@o&IgNfL7*wSf
z^G<f4Yr>j>=&Fap$p2XTtG4HB_0hh0{+=-E79O)MCD5?p?Oycjq&m#2%!t?R9R6&0
zTP~4VD)6>3`E7{|@!H4hyaR6sz}s8!_DY;v7rMGt9^RHY?9`fAr%pPY3XOMa6TICB
zZ*Rie2k>?rymfduC9Y?&%uZ%#-~r%#e!T9G9~|EO9~$#V@?aQTiKF@lm|V!E!{{4n
z@b@RIiJaxqf1GpQdoG>JL5|miKJr++M)bIq+u+u~kJQ}ZZMV|#GStGy7Yc)CquCO_
zROH<FqVIpTCaKUfdeq?Uw2o*EwiK0_%Ir=2+pG8S`A70y*wphGdkmn%CQ`S)xt#v9
z?(s@PQ~yD|;>8&}?$lSUT*odpH!%XQA;gVOmRNw}247o<m+wL$n9FPE2yeH-+a>rB
zMpB1eQ<Zv1JL(kNTm4?!_4ywD2Y6c>-p+xy=NCD2h|kLoM^oYWGKG&<I+&Y>M!-2(
zfH!|Ld3=9(dx`qPBzSx3Ewfo4CF&izxtB2+A84}vxR@++BRpdfDH=DDT=z(duH?XT
zU6y^RJz)iT=eMtDlm5(Ui$`N`pTs`D2})ms7-naZexWlTs*Qfvk~Oe2dw{X7jm^Z|
zv0uy~{lVT1Pno-Q5iUBZ50}K33_qL0!28tSJ+7y+4?6y8{zjaUUgH@2$p&&UZ}#kZ
zpQ2~OVJ<)a9NsR6w<Y22>h{crhPQ9wMo=@j32(Q;t(UCZpI+gS!^cqfQYyI=o(5|A
zh51_-sRg8idmqU0DzC?P#X9~F-mV~L>eh-idm?$`W7bo&@SfBV^TLg<4rX}3+fMM-
z_bVK}fsS~J`XM>sty28{1M2q2d0qRc<8Ng)3BG|JYl#owY$162eh0Ht;ce7R=Cci?
z_li98`7-9o;43h3FPZ7P8@!De2fgHG)l~V^Pu1RSJXZ>{SzYvZ6El>ohu&MB_luuw
z3bD_PD=GShuVYv;b8_&hpFuMo25+O@(Mz}{Svh@}q1T#t3f^99NKgMBJR!@diCo2>
zL7do|ztPU0eg`W$1)BU9VyYf(@z0~<+?<}MH{|9)u%!KD^y*dA1ODMXKZCbP%u0z)
zQR+|n!>QS9@<~zy{zpm;Ptx<5%o4avO>8_mu7$m*;B7*Ce97>(F?nsUojNh+D-D*e
z;%HbLeIykR;!E^(*k|Cn`M}%4Ioa=n?|-a>cES7J)2S!>08*oW1&c@H`znC%F$=l>
zH)<Vks5{)oUvZXulv>2{r>V-!HLtvkzBlsQIUUjz0&iDwAJ)kTk27%1zM`+3BsSW_
zETAoT2WpVV!sml#?pb)-CZ66TSQ-dl_x<MW(*N}mEuIRK`T9e0@{WV?OmSGrgHIOl
z)X$urs`1@}US?0(!#b+L>yDb)#7oDWUW(u0rP5*EcsqP_D7dq{cXuX!_f=YyukOzE
zRhRUB%H7pZMJM`c-5EcXdFiM7nfzg#zeXJMmyfZlI(H4wgtq~@Fg`?2x`fJ=7%Ic<
zP!&xI)8?w-Dl;Zr4a!7oPs?cLvqo$1wJ5cKx9_V&sS8?miR_Uw&5G2#rI9-JZ={;u
ziPE-@(b`RZdv>NlAKw`CVoi(!(73A^j2ckGsG;5A?pl-b+%aotX^V2e+ZhKedfnfq
zE~Ra%Q^}@^RcyM)k-nNuwi-6sIl5KF>s8#Q;3=`{GsmpH8%=85*`)V%Ov=FHeTPZE
zo|*JI)~rucEE@8`qC?*-+V#t#TiLB@o8PK`3zA2}TW4Ub{(-mq=f<+<QLH>eY}9aU
zTK>kS4*l%<uaH9_^&QISNj^dySC1U>eD2WTLQaLyd+bR*lK>CP&^y#=D6ECIi{Nec
z?7!zSC#V*_=GoLDZ#8u5GQ1uAz@_fw#ea`-$%anqn&Q$2a=1{uTLJCe%7B07P$^j3
zF<#&4kDdW<2Y+Qw39P#bPkPUSv+(u{c~uK|8-T~@3O!3Ny%Mx;CHd4c`jr>s$HAA=
z8)lYwk{82Ue|X!i9W_{Z`?gxVR>FX7WANn6b;xdomE@D<PQp1jSO6_^;7*5@&vIz{
zDg1E9;$+$!r$~J4v7_TO8~yWsey958aw$^@moCHFrqk^@*T$gO7SS5BH(HygTNLTC
zYDjOZ9?Y@Aq*xuHHhGhLt$ik^f}fz%-*+${$f1SQ$qP<(sMtjKgQnW(u|v%(IoY?&
zsgc1h3EmEYx0PnNG!Wi4hPPYj|DMp;t<n}Z@hQH#mDDp=yVd$K_4TIY{@&!Fedsag
zJk9r`Z)-+^a?rQ6m|BCK_{}elx~S2qC1K=_!~@$nMy_<qbDc{E@gyY}@#5WwrElE&
zlk<L`+}X%=D1w%|;uM~F&fR(|v&4hkD(TNP^QC`{V>L%6{Cv3`ZapLhh$bF$cO~vg
zc4=;y3$L(C`y#1JP%AlI#3AP^yPn^+%X@-d?r9FS9Oclt`DiNi1#~RS{TAm?i4hKF
zFQGA8h)dw@XuNy-F1e|dGlQiV{_^3>jJuBSKN&CDEc|V+=zWK`UoRx<0QJA+0j&Mx
zyLpK%5^pB6J|-$Vy#dc3$7?tD?%Wyp9)^>f_hM$c8?PdDk}jrrHQR!2^AlZzpDTMX
zUKiI=FCNPMNE{f+&-^fOZ{rE<6-rI~B|XaIu7jvK&W1^4x8wDLvlpx|hn)J`Zu|yw
znQP7w^Dadlvcn))H5d+r<RTA)s|Hx>Lrxe6Z!5#wZSZz2U%$+{-4fpZb(vg_94r{#
zZiiDXL)iDko}wxJ(9F>y=E3cs<UL2apf?7Re}AVReK9#UuPX>1jk$}TD;%E_ylvbT
zz48t{ROIUqh$qUB1I9eWH^#bu^CA93c)JYVZY0-ix}Nt0Z!`10M!;Hcj#g7rHSc%N
z0o*f;PuApv$=bxn8@JLgHkDbPco?%}Wo8~TUjNR7w~uS0i^O^-lC+!|^Nu^r1$xA_
zf@i&%|1+50>|%TIAa22vxS3f&cnJ&PxzHbY`U)n?)+|~5Ir4NP*X@J<kH4{ly1fBE
zV^BD=%)+?#2G|HIBchU2za6u61Cum9g8gP{C+R!0f!c5`!+5`g;nu%p@HX~i-g_5n
zP-_yIT}jU-`j3O(tHX2TfVa(OFe_n6l9K6jYeh`fbTxX<bZRSUDSERsSx1l3ckv`y
zzo?BJAr83C*DrKQ)kk>y5ij+<vhd%R-o~}WSJZ40bJI6K+_9JI+;TU%JJ+KXyiEsh
z&7Ajp@b<=Qe2QG>pZwgwdiX|(4eRnamb%0~Y8Jz)!^mvt&+v9ndg6sIeC;{D`Zeex
zGs$b=?S6Rs65bwww=Ll9-$v^H?f>Up9^8z&h&Q?d->)+;P2uqNHoVOTXY+IX<ZIFS
zJY~!QZ(n*SYkyC5S?(#@B~RUgx7Vy*%6`{Nv-f(d<7OWfXw+GQw{%wP(!TmP+E-cU
z`Kk}R{o2J(YbN;7i|B{m?x#~Qcft!l)j#H^Uq}4Z=A6GK_6^YXF(JAOZy&>3%dJqg
zXd1@;$Kh%^DqQn7MM<xMb#Y3l5*CN5?x#>a4Gh<Vbdmb=k4P1Vx4y$8wJAA5_EQl$
zFf>x(2P4%feUvUoM(ZTJ?NrgIZ5@o-Ufrk%O^v$O%qV+DKE7yFP!||E&7_&DOe!+U
zq=bHG&Iu;PnW;7N7}3|Hd~2u~XEbYH8?$Nzm{q~UtmV{_gF2d2tg=ay3Y+vSyGaYv
zoAmRSQ8Q|rm|bPU(`;7IcZ*(Tv&uK8Rr`3%2ybTu#i~i4Sbdrmt1n;Bm4j{CHq)k!
z__lkDx2rn5O{?qBhg=Sw%;8YFvv{bUIy5K`wMkgh35|2eA?A3Ma<lIk^~_?dPpnl9
zMo`DAo}kz^)O4xCzRT&-aB{V#y<7@H8@<m_AKpHuCRc2;OPl|5>0=>0Uh`bKQV#ZI
zaOrblm)4U{=5I!g6mQb}|J-^&&t^4vYk?y<;BB9}_@uHhck&o>Lgq0Sq-VT_!q!T4
z=&$0nyyACO-Xyoevt8z)TXo*TjmK{FtVf;<GbYX9{k)CSNLQRvh)W7?Gb?>@vsz3w
zY29#>CP$m_vEq%R?mv{eK#kP~ZLMWch5QB;yBN*P+-N1`j@B2CC}nvRsUn_Hsy{1A
zqjyHB|Gg-c`5L8%Yoj$`YP6m-h}KiDXkG6?9K`t!KN71OdGM_L6Q`M3(GSpAzZ2I~
z_j0Jg2IjaFb?R$J>Ii5X$#2M)|D=Ao$)%BZVHUid3vWB&)txlRty9cpd5;&p_-VHc
zy!Tn2@d_umU)m2Yz&vArmxfJ%t<;vwQ$s#TjBy>_4uQ9MVQmKT?_GRcYKl`aYn{6B
zmrF@-zSR(XO8B-aJ#lG#1Nz|l;Z2_dcSpL}L)NXW@b)5%P3a6@Ip*^5GWdIqJhv-#
ze$To*4st11SvdL2q1%5t^@F-?U+T4Mb2~KgF)<yy9lh9&ZbU7Fn#j8<4n>DMG$XS^
zy*=!@>rGt)FTq=0i!+d(wCTKOc)Kfif{gHX{!nI+ZcfyOWr?z;QKRpfsG+He%r8h(
zCF%zcXD4XvEb?HqkEMeX<kgj$LBRwad4!GvZ|A|=6Y#b*yv+}9&za+Og}AI!qj;Sr
zcA1UVvWt6aL^uqFvwof8l>z-=Z*J<-#Du@=)*;Mqt8ytxO->{$xd}dVI9R1TbG$9g
z!z@frjdsux)~=fc|MB{?hPNy6sMVz&>hTv|BUsQ2*4}DO{>R66$Fg=ePS&D2$;y=t
z9rbjQsxd2e*Zd?M$Mb$1COtWrB=a+3i|)xPNq)ZMHMRL-%vpoCVFCDIzOpCBa(G^c
zwfY(wM&(qUhr9h7!0VGKnh$Tjlhf~QjIVzK^Xc%p|2#tMG7_)%MCQG&NYvlY*#oS3
zk`@M|B@;t9VQp4oh?NZ!m6e$K)+1^%BNMbOJzoF&#3HwdVTh3r?8Uoxmm0r~`E*C9
z(_BD9x}TuBFYzdTqHl6{f(D|S45of`nOOZyY=X-6VBZ+(7Hifg=$~B)iffRlF7S3C
zyv^ApS(jjV^D>+ZH?zy)dH+0Chquk(?W_=H_jxC2EAv&Jkf)Y`o4#ECH8q&0$LsWf
zw@dMp>|h?A^(x-XVe}{9qfhap)&*~kuh<{0AUy=syVmgBv*2w3c$;H!lG?A~`8fBB
z@#R;;1Am^Jxj8yYspHA)*^2iB-e#^%PEOoo@WOAHmE3wkipIL}>=6fzo0*~!TkuzB
zWM0i!`pRK_RruJ5`nC_eb)rpFos9pC`(e;;O{5z=k)`kee5Y0oZ;Qj*1wDv$h|fk(
zMkhyKDF|;r!CA)*X7tP@Hd~1=Z3w!4AoFU<(c^}`ewgEY0-m<V%*G+63Ti>k8s6IB
zZ6w~S$JEFLZ_mNo?{GFFynV;x-n^d53vVkt@lf6f5B-_NLk%L@D`;L@oo?Ay1HZIU
z^oh2*JGmXXTzlPo;Xz&8Q)kO|)I85ls`bN5*^2on?UIlFecV|)Zun|?RX+`C?k95_
zKl=Cl*xQzmzx(T4=|Da29;kZJff^YXs8#Mj%?=3E?H}Eg`AdNM6zr}XgL`OI(_nSz
z6s+^MVEJYa)BYCW$~+)ktHy<EaQ|?8OcB~UI8v?QZMiH_T6a8}eQ09zDk?@<avGK1
z5~Bg}(W>tqtvg$y6};V`HdhQv_u8Nhzc_NnXmz0&&8irq@b)p<7#_piniv&u#^_%E
z7|qxiqr~?yYFpB%77L9^y>C>#LU^DHz=&+HBArQxK9EDd!Y6&zsKj$d^(a9eJk_Fx
z=*pA+v})^ct4igI)iy({>f2-0?MkdpqfO5kVpBhOYa3@*-#;DXkq-R{YvWk|Y&jk3
zo8F-&Ssc2?#|2q4r{;0!>)-s`W`}(3)GN2Jev|VZIExOBr#dhG(q7qJ>JM*sr@8dc
zFg!xTT`Eh?*2Cdak#%U~)be6h#<9-?b?kXI)fgV9AB*YB+D$JPp6J|TS>Ipcy&l9G
z$p3S$7v#Bp<8=ylE;>p-(+D0@-1IHD^(~x!sVZ*u^P|@H%B_0n)~jlueX{=l*~_h*
z^W8cElje~R2JrP|tnDq~?TsrYbtc|<|Inzi<4hU}Z?9%1Ug;R4tJLcI95CqEDuWgl
zH?V(EH2Zl+sg8G)!eMO5myz1=AX4Tsc<4=0`qc~GuB6`Zo4Mbklu{&GmEmn>>JizS
zM5|n*X!eVaR?d`Yy?nrPEjK8KBZeG4M(6Lw$h|N|*N4QYP<g!f<818pVAo6ZhGdhA
zdKKQOKj@i;w_flzBfNFYWA5lMJknnDY!@K+J%`76yGuKkxHKl5^X$)F8{_Z-z}hNR
zoElr)sSgF5N}toI-00heZ(LtF<pq-);Kw@zZ{y+ZQW!T5{b9~gVxNmHO+ttGFRxqA
zs=0NYIOu3cbPFF|1FT)j(GT8^g0~akZ6?n5G<aLHC)d58Q~CBdG%^RC1OEPorVe%B
z-kE#BuB5tg+JHaaLw5EB!b_jWuAjN=cnsq-xV>Ey;cXFvU5DN~*c%leATdMSb=V7U
ztCN4e|H#_@FTS?p>@kuQucI(C6S_vZ(|EOD?HLX`ygg1IaC7S2xy}=tH6$MC?be8I
z=q-iaVsTU>xaxqXh2U)r@mQn(<Lx0&=Bc6=tnUnK@uqD@d#ON;p(OWXzrPYx1CQOk
zRM>wkK@;Fmee}y$cokm%V2*iNa<%SBddGTwwkb2Y;$iNC6dgF1qCYp$*Br$>>9Aya
zJn=Q$VkXY@BvmEn3q<d~U!Iwn704TzD^lqixgO6kV|J2a;O#+pn-1R2{Fk~lwao#E
z$@s6x0}C-{y*qn%HlvsN3pHbEkS}YKtM<aD?TNmgj$CvueQHBstR;z_CVJJ*;k8)D
zoElzt!|lw=F)?#(Jlu<dfp{7^v>-Ok8L#fV=XR}`bBhL5z>7YLa}NC)>`)#P{tjRC
z2jcS0B^<iIn(^c_aZ)XZR$Opsc6z6LD%0PPh+ls^^^^_x8BRDAb%S*Q4`x5shsay}
z%xS0o-cN4^U-zS)(x8@0-&(uLu@dx)^W#%BS?%EMcPBk)p-CzWXZ^@!Uy<96fVT_b
z?R<DU3*PPz`rozAO?<U38@|b|_&OdZs9O$Z@0G<*3Co+dPgE0lyO!RDKEzDXYnb&z
zP3*RV=YhBLrYEV^5}tci68{6tZ~2EDn%B`6-uB{J*1(_lBPLmg$?pr%*WdLYG?lmH
z_UPz>xAoy|4tSe3F-6A@aGmLWoHL64YjlSknem?$rp8+W?fDb+n+41chqsCFHaBcN
z+LT`MPs}QZx7WFT4~dZm@wJs)*WxYkxwoQ@-iCRrt?3=)v8z&*zZ3oWeEnmS|J6&1
zM>E%ed*EgW^WvuAc{`6+p)s|6>c9oyZAUl$IO_bVeW+y*Am8oJ!N)0CJmvk>gZ#lm
zJ6e0Fe)bM{%-X6}a$8NP-&T#Ew9$eyZB=S+J8B6X<Tb!UHGMqQ^@At1xQ@(V?4(Wc
zUV2!<TOS5_t8aTB_T=iUXAON-;i|8GRPocIA%5EM#!uU7_^V=ozp|wIYxy33EiBwc
zKd*FA<B#3c@Kb=|b9Pt6pdQ)*Z{Nb(Rfb?yOA6M*ufb}RGnATWsOszpqjngfO+z9z
z<#?oi4IvLYX3*u*F?t4Xvu8DGVGQdiy!{*AUV^t7;O)|j2Bm*tPz!iF2j0$wx9Q>S
zvUV}*3UAZF+Y0bD{eT!9hqswN#Arcjqq_7lYGQt)yiUbv9ADoxEk?!1^ZWEdc`n1_
zwcRNHvqp7=w-3;lhoLJ^hHGOjR%MuK)y@}IEh`_Z%)?`K{bsC+^hUqFhqr2WoaX(2
zf#e+Z-jjd)va2WmM~mUD7rea;Z#R?ox;Zjjvg<{lUF)bvE+ucOhflZ7GCZj8_5uG(
z-SNqcH{u6^w+&ds>$3*uGPo4z<-!9;Jvggf8E#l)3$dv6S&P!jSrr@{tBX71^o_jg
z*i^SHFWg!~{<`%PeLBz4ve8Qq!P~9ywgar~mfNko<lBAcI~CxDq2=tl6l2q&{TA7G
z$Ljv^SmkJKQ*mmZmur&?Z*UXmvG;_X-X~`qdmQ3jM_0H4Zw)pyh3`h?*b<|1Hw{YI
zV9>5&2HpJ_rO7?uY=tQKq=U63*x|Nhlt#RW)X4jhst#{2z}wD!qSSD8lnOnKVo&)f
z-6<HYYW1UagL*^~-0gKYQrBNocX$=4Rh=SK!4aXGm$)Cg;;)M)_L^p3F9U;aEFh0w
zXwXe!HFrvkrhSi5zb{6e7-3NdwCX3*t;`#+YKmv9Hpj)X9}>FtzF75q6|1c2nZ;I&
z88P&|O>Jk>qrZq<8qu3xf!-U=`QLe+8kohY`x%^y{_3D_#-YTX4t4i-Xf!$YKJ@f$
z=ZT}{I%Q=h`M%RGm3imVG|uyEczYb)F7c!f+#Bw4<b=2J@U}V(emc#rPRv`)1Sc03
z;hrjE(<-#`Rn%zTv~(~d8g|0lVMpxDw6-a7f(`$YP1E7+TzLBx-UgvNtgV8dioSNA
zRdy};YF7^KzwaOHdhpJ!$Zk9rzWr(KT{=ZwINS%lg?lW03_3`4X02z1jhWpldd#J4
zQ(cN_iQlY|OK14`ye@c+PVf?LA1cU<wV}*%qux9+0{?pyx&WTK_b~Y|`R!(SYlXKp
z;caPn+X>$K-FCAdv|D!KxiaYPv!1(ED2Q2e<YzOd;%9@mVZWH8QXYMh9_4~N=vVeg
zmW5j6ozcv|L>FEBg`RQprBbZnGwADX3~!sk+jsP@#kNUQ+zooe-zVzaPk2#<+!@Yf
zn*)pC?KXHj>Oc}RPm`2zgt@KcU|--t`FbgOj-TlFntz{l+;)`Qmi4-P0={!4t2?|M
z32&Fe6C---U~;+#hnQ0WZ}*Jn(a!vIdiwgn+rp95g{apQg13hTIh7S%>kyiMTbDzb
z0;sp3#cXfo(6T1nCw1T|G4niX@`L_xC<p7syr<|D#8cM}au2+*>lxR(LVG;tsoYy*
z`2U;<o2kpsJw^Z5PadP4y0gtmpF2MP_fGbIcIhS>m!kk$RU9=VVvm`;{&vI(f5F>B
zaP}p7@nIgl;jMQIX0+nV%-$UzJu%P$VxZ8o2}-=eHTefm<!km{A=k};r?Xb=MBQmi
z&w?-W0^sc#xZa33(b<E0`8(c+zMLO;TOZ!`g||cCt&#J)1wEk;{e~6Ex8pmpR|mY^
zJ&|kqf4p4>Z<l8zhy7i{cF{{Wgued??2)pUp1Qou$s3ufd(;q0XU9uckU3)1d~<M}
za>Co4@OBrBtPz;1xh>J1KcxKsS$S8veh=VnD!g3=Zx1#{Uxu%DU~3w@P0!Jjug`3R
zPXOP=(?uz2$KUrwSMi;cs@_?9>KA!&_>-Or@JQ2Kcv}SC{ta&p@HVVB-spaKu6P_t
z{n`6oJ9VniLE}1hQ24+Oy8lN9HHEh=hqTohV_OC7ZKDD3Hs6AFXnh@+spmmW$y3Q+
zJ(cHSN4e*B(s-Ab_7wG&_h4@gYU88NMLKi+e6@X#ul(}+>2MuCwXKid?BS<`UVfUn
z$4@nW`srN*f883^MVa93_~-x)jt`L6njmGU5zJnD!D>AxSo1FjYhCLQz04G<hetv+
zEHX^yjN$Cj5UwIQBXx31lzuLa)<4wun#?lDmNiDj;O)~5(HdJRLQ}?uYg$COws91-
zhii*(xC)?QcgPd2Uw?$F-={G8)WXzmSD2pl57VxHLv^7)dbV$<3bzQ6aeS~`6+={c
zZ-`=AhicV~P+dRB^A(CzUUI4LN24_^gF)s^2E4`Ox*g04o=tDD*{Us5teWr=?pBZ0
z+99#(aw}G$O>D{-PT!C*PMJT&>6isi%L}{av(A0{*RJ!M`J8%Nh=X-80LGq-Qx034
z8ZqN5m_DYau)AG>SXFvzRi-tuib}HS;Z~d89=7QW$CkY|g)pxq-~L!l&laoh+pJ2=
zYgL2u7Bz;q8|ja_Sed#fUhT^lV$}#;`(7k_Ml8i|<MTTPWlhgTt#TN($TIW`Ggl=1
zbgUXav}$Qf^5X??dx%B7`dd_em_=9MZE~VjH_^^LTG@2JWt{xd#VLq+Cl%<q&REB)
zfc$2~PB!Tc-f^#cM)tchGPlRb93HgeQqg+$DT>%BN~;$~sXYDQx5`Ip+>b~F;gOrY
zFH-CA$#sFD8GF*BCj4@JBUI*mIJs81eA<L7zbD#90H1SxPYev#)H&fw_c$D%YlJq}
ziBJJogl4yo(#`_WI#eZ^{ehx&=3caF_ciEYKZEkb7?dmAK)uSK19J@OHP@hd_yP)Z
zoLy^B)#nCz6PGoePc9x|)a4~cMeR4LIDK#rtC(~!*`yA<zdY+r8otJ&_h~kz!&CQ^
z`KYUQbF8%K)L3c=U2S@qmpN?VvC8GPYN~^KBgHC@7n~>N#T;ga>fYBJc-}`a%WZPG
zIL(QQQ#^Xa&WUl-XL3@`Z^UPd(s7R&URcmw@CS^GQ}M@fD!JaSNhj>YxDI7O7b&sR
zuBAC`>?O@>VLn`Op090>IEBrR)8UtKdR@UTe^0w2$Z>z!?JD?>U5mI6_rTlshwVz+
zVAr^1)Yz$?2XdcG9cx$W9++Igp|cGfy5n$Y*Es6o_||;L-6s<NO=;&+26!8Po7($F
zr%rTtDk#y#9^KTN_tH}hV;7)*l;Btox2vHahrruv@b-%@Gf&~~5qiV&!rX|5Xe;D<
z;nY1gu5jxFM{AyA{1;-wQ+TJzq288)jjY@A;ceqp^g!3d%SFwy@<8@5X-fZGIKHw`
z_y$ePfotN@e7I93#;H$j(P#L2b$Z!VIY}qbVH@B}Z_)?$a4ckhvu9|zTkuP?hC6@5
z-sF+V`U~b>XU(62pLd>Lk`7#9AJR(6%7-3n{FS61o#FaHa&WZwmEGZM2lo167G7U?
zn<qR`)r+Am!=Q!X%&enFE~YUah!LE-QuyhK$>t@}j}LEc@OCu3oe6K(z}q$Ob|SnT
z3U5DheQrKMlR0mfWt&}p%(iRgOuN>uvnvdp>?rZp`9Ny)$#_Cn(`QXw7f~A@0sR&~
z&{k&L<UFLOH=7#C!{X=_cJ@vs-yZ{Sv%=d;@O4mk_I>#iCZb#4n~aCA2&~M)+`TPm
zUDW3fU5Qs1=gWqkT@T*&g|}bGb(2aYY8i9zF4Dg+qg|p#&@)#Bo$DNWn3Z*AJ++bn
zT-O4`Ro!FAiPI&jqJN?WqC@8+_NmY7@NI~93*Nq<mvLgRWR<>}%$}mucuS*QG^Mu=
zzrZ(Y3a1BCw}rQZm-6|W6wMt@9(n=q*(tPnw1=99Qq^k>oSnc-8+bbt-qsJmcL#5?
zQs>V{?V<BwW(iD6(K%)b{1uy`ss8u?@H%F1mZC)*woZ8d@F?{0V(w04ijIxJBLHu6
z!`tfcb`ZS%0&n-<#wU&k(3ih8kiLUyeW`IIr)VRZ%OH+hY2?%R5Z;YxrUf~g)2H8D
zSL?Ubm3}Rm)7?s=`nT0Zc-wnN8#OuIR)>$aQ<+@kw!J;{xwEINuRYZxdnYYe(uvu!
zUh*m6t(;@Mb*7$=hAi>XsBb>_YCG${HJ$a>AYbJ<=Bvco@HWm*-S+w^=MO)nhqqmV
zy6A9RH!bNApfq@UY)pXquMN<T^8w7j4pjT_K%HOMU8`4zXdApud=#py(?fM6AXJ??
zhw7|5RKxlD?kQpFVhPvC9ca$i!j+!&_TMw%ItXtUgoZ1mQMg9rg_A?WRL?(<xhg>l
z>lDPS%OGt_*IhF|25IiwAT|9EgihaGJ^t#UANPAG)Eq2x?hrlOAEG_*wkChu7v2W#
zjnG555Hu}HW*>uIeKx4v{TP|x?bC%8bu(CHB}W?YpH+glQ&V8%)mSZRVAJa`4rXJV
zL(@Ca*RF%as_*z88h6Jo*AmvV+ICI<!>(!%<1}m%^QWzG+6Zq~Inh2lvA0CQSbBx5
z8konbM#&cTG_|Nf1#;`^JTf!nP#cR<Cs-6Z*P;(|EbM~}7hM+J4z}nFJyNI1zwPkW
z{MxGTc8lJ0;rl%-x*lQC0-HsjE5XoBW^I~jR${nWZEa?`N0`;+hgqJzEZQ;HqAu_@
zWGMa9@OF-s*aF^;ykyndt1$}9Xw>7TCKc{ulJ^Ibn$lB!{IgO2zBH;Byxn=wsLqRx
zn*10&|4B5xK+#%<PEm(GuW1dV^p3eJ&FKYCJR7NBdn0Av8L6TZB6Y4{qzb=`V7_96
z*4_-)HD|a+w+vS?k8s^>9<IK`Qnp0mFnIg$-*7zR)E#O^s8xJ~u6BshDR^6?QZ)bj
zc;K5EG%D4goxKd2ZZK$fkU^8-t`Xk$fVTm#_Bg)zgx|b<z|ZE7(fGMB`T}o%>l#7u
zwi>)0g9m;Az2Xhw?MmwZQ!bd~`ic)Ams!IKn^mi(S*4t2EvRDEwX;^`JdA(6A;(Lz
z@?13QpQF^XcbKW`@-s{E0pQWgGTki0M11uld5+0uz0GdneOOejsfB&B&=bNeS~St3
zr|@<Ku~%PsyLWD!TC9oFh4Of#oObmXWmlt>=*Ky1x&!B<iN!LPkJBS~yJLJD{@Xa6
zENj=m`qXZ{$a9<6nW1D?0UnD*^Rdw`Bm5o(Z%>j>mw~lWBkWo;mAq<)T^;J-M}xOT
z#yV7--nv`V$Gb2O&Y$1OPQBP`o}E4C?dnz7t}EnQ#X|7O!@TyX)E}tN#-M?8g14tC
zGpFM-bIS1oJbpucoJVi`d(W>jgYp8t0^-2Ti_oND!(Z?=?;d)@-m{kp>wM*AiCUPI
zS|GewRg`_UCgClBuf<cDjYYk9F}lr~6sNlOaLVQj)9|#f2yyCjxKmZp&9|L%Y8bVc
zN;~PT#;b4y{#|XKAcr5l&w=!VF|(x!dE6*!ph@KOMq&fk$avlUN<EsHWA#5W>kZzv
zpa$HUd^q<?>W}c&eVCaq3lnu2-ZmsAu#8M(FEIM)Lf}x*MEUPX(1m=|zE{y(hOYiN
z3%>sK%$?(Zen>iWWDm|OeG8vkpno*RO9^j(R%Bj4VF!5!H79&<Q|{Z<_$(R+jIBM<
zE-UNE=FWEA-X6#7Ci)V~662M_16tLl_4RCe%(^qMqfL$QaCUBK)2A-X2KWej=^0q+
zNYHY8S8d5Nzg56TK91h|<K&?gxi%-=%x-hbR)V?{dPe=dXla$v`l%~q<GfVbiKp^R
zf(|CY-R#tu2cb)Ft$PyBd`F|n*x1e9p2Rwp+zLTUdjPvTK4R8c&3LVPMqPnga!ulp
zqj>TjQ6s1aZ(G1yA9(u}5A+`;nHip%q{M6J%}+Sj6^RSJFiQr1-f0+_=_9p)6)9Rm
zUs}!5Xv0OwU&&cVke3?#={s`LhdqwZd(mU{f*QkJ=4stwpJ@1eipMkft&-8Dz3JWa
zgttH7?Z2MvaSGQ5!Q0cYw|Q6U#qH3nbEK%NSF#SSr<Oe?NhR@+-Ss67>xfUSAv5DI
zC+akDR!#aQvvY5Dqy90MJih?GvDFF0TxcK#M>JFKea#fHx|vSrY_4K(x20zb%_!Vb
z2@6``r)q=et*!cfZL2VK(4JHe_4M=9>8u?U)2@?-uk0j8V=t9k<fYN~yi~KWxAJ%N
z*7gD3TKmOY=QsFh-p0<%n)Ow6>bmXW?c9BS%Ae6+Ya06NX6G*28rMY~PIb|>j9vAn
zSAbmb_QZt%{g)CbztTavw5q#)E(+24{h_)>|8C;gP#td(D*xi4nv@>D^QjQo8;5AY
z`(QPh5{!Q(SntOL>)`2N{bLT*z8>K^){wmCY?z$XiD!?;S3SpHm9P4%$lw0jl|Mk2
z?%<#97NmU}gS5L@cfBgqL)#zp&{b2g_GIJv?}sR<a+vy+38%j&T+7I@_CAbM!zO5&
zWeoZPZ)^M$qeTUb3iUMUY*F%@rxrzgvdRbE7Wau&&?M$xk>84$7|(VwlbYOhJ^v>?
z;q0Z;c2#CAJK5N-CR^k5FMYiIf5z(3f3fUE7prw1HeGVqw9eb6?Bp~Z$Tu^=*t8-R
zMU^1mr01z7+^q(0m%!Actjl#;u`YX9bPC?q^tb3PAGd5|VJ?P6S&Lb8pn*kU{LDnY
ze*oTYB>%nI!J=9ZS@SoWnS*22*if^4EN1N*&iem{h1o?G_BXPqBAj1UiQEo;=AKDS
z7R~+R6Kdj@V(|XP=n$VTDQi>+KJg7EdcE!%$*GJg$-J;X;ca(#I}P5N;cW(Zd$SL{
z*W;s@`yHhj@HQ*FeS9WT3)e)dF5bB6j!3=FAE|usb_~3|0&mm7+XY9$v|)OfCNBzO
zK1{f#Cxk0(X1F$ArMH`Yvp{&;aBPGw-ig$6H0JvmqExK_y0tw@t`vjzSq(D5UIiNT
z4(?9tV^E<4gWB?SbC^L%FgFd}X3Z0WS0YAM^4y0LW7KUWHH$OU9MaQwh9@s&G&5T8
z!>_wUeDTAi8t~QvZ~un3i+#+hI@qjA6U<7UZzh&EE8V|lW^<V}_lQ}I;B6UryJCr1
zoo1QUVX9fn#`C*_%qrB~tafe9^23uC-QUc<CT6Yg&bc3F(MRIB$X`}G=2k7iD{!bX
zHG&qgTGldF&6>xmYW7&wTpvrl7v1}aP1Qc&*K5x{Bk*=Cx<xMZWm`qN{wieGsXBJ`
zhqnjn+O><1ccM2R&(8WEXje1t(Y`(HYL#l&BkJ6(hrnfcJ{l%3MT7BZYiIuxbZA(=
zojQX7*8atD96pcA#61mf2M6N^i{VIOKHC)XRkU#*YSoASa5JA4o}#Jv<Z<iYK=gFh
z?7}OVwK|g;^>}8SPNZHv-K|i(0nN6Mmo0JY)mN7mzo37bnUuS#(MwHD_)YV8{S9yD
z{(*Kq1V0%*aj!_UT6o(U-WDehe+_fjk#k4kBR>dlT^Z57-{ON{mh6jx%qipR&4SoR
zg}Qs6+tf~9JM@ny^-=Wg?(o(Z-u8l%X4rn<Jo}sMrLGf<=YB1kz##nZ_>l(hWF`%~
z-3M>~qF?>XL}rEIZSurNcJq{5S>SrpEO;Fkus0|%(LB!iic4|2%KxGl{S+G`t;|@q
zs#dgBL;bD90Mw-7EZWk@tit<E+O))^_8X|l5G&7~W7jg`nvz_j27Y$Ugts~1ZAAT8
z70PE(+%YpVU(GT*%o=innLyjEYU68ThBz_*I)_fw!AH4_+Qm8i39!Fc8T#AkH5lB6
zebw^fSA)0JRuc;qM_0(iY|uSu73G*4KbAe-#^ALK$6JTyk)Hm>ho$KyXu)j0&MsYV
z!z_7tJFAgfCU{#B-X5#Z?@<?7g@!So7^mkGc$=HOS<2IU*eyZll9_3@fIf_0^b!U#
zD||J3oMvH8E4+=kN}nBiv#7W4-A>PXW_(oSkJb<Ps^&4D?L2eg@Im!0l&XG3sqvD-
z-hsDOqo~QEGpEAaAJ0=%9LE0MyR`ki6m6#td?T2-tRC?7KfKU~s2lGguf`kQ^8sGE
zQYkvY_oj8BC#o&I^Pid3_kj84)E744-#*wlNs+_|D~S~bY-8psz5AQ{6JPP#{hH$O
ztA}Td83j+!G)#Ee8gMVP8s1DtcQ%u0Q8Sp)Oz-=X=XPqL;qZ3%yjFVjZ)?5I(Ms#K
zwp5N;Ew$iyD;>+#Mup3FP?hwZ(BM01Rzok%Ti~Tgyv%<T^42R4Z|&*hjmOwq;mdq9
zp=W1}c-C2eqbHYs=cgSl=o7N|tJrvd{aLz;`bBk-0p5O3-&LEAbyN8P0a^=hPjwDt
zwnaA$n%qq#vj)=>8LDpELba=DsPg<1qA_zr^lM?T7BB6gBU5|m<L4f_;~lJqEzzNi
z2D6^^(8HeHb$4Nad<u8fhBm%Rm*Ok?XkYgJ@x}XqkGibCR%Q=S;GzKCEEI^}G*DA=
z1nR<z04-h|paLrcwEJm*t}9Tz2M4PAp+JqU9Hjk6dg$1#2>ngHacmX*I3*%9{{`zl
z-klo#qcsK1u{Z1D#hX?wU>*EK4R9!PE$6`79`JU_9_oKJ;+WgY_w$mMZZWIkSTpS@
zW)=11|D%ps{d$<R2oLXpT;xcL%qo7wstWh5>Uqzq)zlTYm#427-YzN*PvLD7^3oRY
zb|t($32$rj|GN;zu7;-<d|@#k?{7%{ODtgKh=jLQ_!(<ka^8g|Jsv|&I>f{b50hGM
zgthQ?v9FnZS<M<e$D}=FO!N?#<o_r}4KK&&O#zcW!O;iI+8hO2OWli6`<*db!uQ+0
zj8Vm+MvZF@YXgnwaz-5_A1nlKyB#vB6@Ijf_?T8aiq_YR(P|hSr8%9VRG?mz=E2)u
zurVIq9$FnKFKYE8O?-Sdf_X*}+7KL}kXzyCnc->yZ^upxQ?rSj`wwCI(+lQK4p+-7
z;rjF|T>kL(7`%0&M?c{?<1$96D@^zK9Hm30qvbm`8r{X94*>?%fw#F_2KJ^mXl`eN
zK9Tb}Oa{F^7p>dpqxE>FK}DM5yQ1D-0iRbl*xL{G8tDDbOpKBK_nc9SNt?-UFaKp$
zIe0swhFQm$sdh5atP8`4U1q@4L-6*lS<}y(+3VD-AKS@cSDO_NZ=1ne!P`FYb~ubK
z5yadUcssbOStSRU<)WW_HrIXi7z_Srt5$r$BS1`72euY3L4BbqeQfZy61?3&J>lh|
zSPf%+^*_!%_ui&?@U{e+Np5<u^1X;tOL$v`T=(u}W~344Meo5=z&w{+@b)OY?Fnzo
z{9eo9ZOs(Bs=(V(gUPE#+SQi&ajb`3S>SEe#5m@7(qBjJw3)uT()8FRP+wTSC038A
zAD@`vQ2T8Tjo*(S5N-3%R=@j%(VFoFoq~nW4fp~^uwO|xv`2i}s}`UE!`lJywm9qh
zPjc0I@OJcO`lR6PHF!G~-fqTkKPZQbdzOCer%ol@$20VgQ}2(#S*ugE;BB%GbIID!
zd+kqOJ-mHSPqsbYp?r7?o|7xOHpOYea^|iUi_<2LI8`L~n~sLxYllNuPCB&Xs6(&m
zD|@q+{x-Dww)q_DpPimFeD8HyQWNJKdRCzRgYRw-AGhseSD~exD{4dO_M_G3MXN7D
zJVWf!V;S>s=;g0f!Y1QD=D^$R%3qUMdq43F|BL(2qHVAa#IOc@%1+FC$3zbue62~n
zg1(4~8<{PT#{bkGMt!SmRN8!__EB3{LmXcAj7`m-+jJ7%mhKp<Z>%dN_Ly}C-ahVT
z)*|A*_zSEFp;p~19m_d(s%V4%@pb{ctz8taWIB2f?x7=8b?X=2_jqQ+gmU!X<zkO^
z;v*+B?^EJ5Y6m*oWbUccHWlI?$OLb9b5B&;%RN(>bHS|Jk>O}0iyi8fpPumw%<?YH
zUKu>^Z{F^Nw+48-GZ*vg;B5#!hijA2*WvByuhfUTC29%0-ItEMwki2-Vv;6OSMUBk
zN#(XtYlgSG6H?R|-sXgBbMeYoBCgp-&-jH6^iR%9Q~=&9ud?X;SD0%{?z(Fd`RQ!-
z<DNxcJDA+KYKqd~e_rt{S?$l0C+|y^a~DiKoUF~ak~O^m`(?Hv-}NOghPSUiv0oRw
zEx$HN7l;9t!Q087@xooiXTCL2GiN1g3;At&`~g{+ah{*o)ezp!EzIZn`MKN7-0Rs)
zPswf1^lc{h-Db?jZ?67(Tc~oemh}6yQiVsY^%&lE-qBLWXSP)8*p_lXXhEMsLp?P&
z)Q|G5<S5x%Bdx8qu67&wjclXpGumk3+BT{i(N-<aw$%qyd&Ot;ROm5sVs|H{@8_fA
zT{|<6zq1ye?kw|5{8i<ARWi(1`5#jEt>LGv=R2#Ezpq~4an4cKUuPb6(T*S8WO&+5
zWBPPeg?{w;9`RL?o^bPkkBW}=QSwRn|JX}SzI$od|6}Q_<EqTQw{3UFsH0;iVItT!
zilhkAa5^|Nc6WEj=-BPponv=*j;&+I*zNdUzTe*;&j&mL2hMr+zSg?ez1D8rR>O;i
zsn_)|dg;Qo{e3t-o_0Fgt(|hg+mDyqsdTvr`Or5!2kzH98>Lm{qm^fTjOtg5)qno6
zS_t>+tcca^7qJRz9H(i6;#39R29}7|$&7fl`JSL=bq(rT)2ODWEXsIb(U6i>HO}YI
zv0_P_Ly@M4yy?uF{+XYWf8<J6`ieBo68j_??l9KZ>b$9{^(;j%*QTiK)D*4lz@F=g
zhTDaH-x%~sc>7bE+}eUYw?~S`^d-l6!y0OwszL0#TS}*L#!9L_R!LQ*8mSrpcNc!*
zYrdo?|JM{1A;z5*Ox_yK*X~cz@ip`~txQp2=5J-f=K17I%ibsJ7`%<dGwexU)M@u*
zHN=noJs-0%!&4LjZ=LYANx@_trst{~bF|Nr$G$3p{|VlXV%BEKR^)kcE*;u|_jjF3
z&B*_1!qnj{cpYNktA%-;t&=o{{Qu8iUHD7s^T~m}d)1-e<~ejE(?K22Ay0T4T^#m)
zw3CC`b!e4cNqEs+@OIi6`oPB76oUt9!$GU6CR+9VkVW4nTXdeQ?mrd{Z(vn%YW2r=
zTh-x}RW@IniqExi9vvR4v37J-H0jz7eLe0__M%SCSa8boD!<1)=9+zRs;yU|s!$)O
z9-63*@HPwvf7|9%_9^(DiKF_Oo!VEzslz;Wfc~WhdlGf;Od@Bm;<<yj3*hYnc)Jwd
z?k~ljSv5twJjhkaPdklZp6MdqJ6`)_@|c$Jb|k#L0B^spN?}%Biax;O2jlqn&=ke=
zOi?|1ir(j>MiH5!TJUx>yv?YLeleE0E%5f;r!?}HG=;xP)uZdFYBoPrZK)%Kd8Wzt
z5xMW8bXh*6%aAui<IzXHP#69>lQX?wOqFMustO;9!rOW9c7MT4b<2w8!ucZ?@=)h3
zm#G=SnF?8#sYS53&XG)&f(t%;uYYUK2z!*l3_>&)=BouiM2C5hu1v1vSIkXE$E*T-
z2UElE!km~#XqzeMn%|mdsG&JSe?>9(lsI4{Om24~9q(wGQg@`PXj$sn%{yz9gC49X
zJoEA7v?Hk94ui31nZK>V!%ZDFbw@`f|6t~7W#($bTPJf&9zSNL$z7PbiTtz`wc~!&
zo?(G;HGRlc(UZeD`xD;YhPORBcTnBj<nm8>9jM#Xg|~nGlBu(JCI0S|sipANYkCKb
z+5lH~<3HQNS!QE8DE(S{?Y%?34{yEUZB=+XlGm^>yxk3Ncl_2~9p7g%GbU4yd0mSY
zYOg2ob`san1N8joruQG-jxU#?5$IiitY!VOh9h1OU*BZbE%k-z!&5XZG)4QqCaVH#
z@h*B$^?}TeXP)g{=IQ2s?$T|%eHZC>yfl^h^(UB>i>~5zi}lOBH*8;uW*taT+J|IS
zVb<PK*88x6DXKD%wcI^bho~jh`#|no7$0Y9y3UQ|9$7}bcJhDy6P;l8Z?7{{?_&mi
zj?AfjgQtLcR_(sjC3w#)#8t(Kp?37fk5Ga4@D{TL)^dLO-b{^tN9<CT*9hMFk7%zB
z+c|4BKQUA_JR%j*GSH%(ADN+BlNo;KC8qnFA$Kla_2}X2?vJN5ikU*_C)v=;<36-k
zN`nrvTt{F3lQ}n=@rGXRh^E5(Ux)M7iObqhi+H<%8Z_SL0n8^D3vW+t#JhE|tL}V)
zyOr^S<CnGtaXu&vH>2q<OzEcb)YG%W%8S(Qr{E8pQja-&^p)>e(L?t);2T?rZcR_J
z;~r;A?_oa26neq?;_0Hkozekcm!rF^_2Dixh5Q#db975LO@z1G;O$28*wW;*Hyn8P
zThmY3h<Qa-x~h7&uKJPWrM=<a>eAbb9Hg1@MmN{=sm+;(-;92ZrmDHCi4Ko#qUg9L
zYM-yM`YiELg|0qo9`2*FRekjLCU5oH?k%@l-g0<1Qofvxl_R3D3S~6b#@0=Aw01Lf
zztT$eh6TxezrQT&0<^Ahpz5XuGUqu^;j4n=g2nBwg)+yvttMOu!(SM#j9uZ%648!%
z#1RVfiBx@j%%<Uy%DK0#CS7i;WfAzYs1>{3geex@MvM-V$MH5QLcX2eAXKvowN-VG
zFjcx8MxGe1GjqZfbSGTVUD~P2*mkM`ZwHi#P*wVdU6UiV>spkClLME9w>9AHV!v4J
zWiHkua@;E~W0|KGr$O-ck3;d=fZkl!k)SHu5|s00f{YUkvYa;P_q;~k{LiS)p%%^W
zhnEU1Fh8}saqw?=7Q9mbrOCkFH~`*mgSWq;-Il9B{qr0;5&P}au_?OVD@7gLS$o}*
z>1$@K)q;f;$UQ42(^F0#a44U5VrKMm>V${Me~PAR2)w-xZ?~Z@_lLLhdA?%*$6G&m
z+Xvn@hwb%up+zoaum3$o&*A#wO60DEnP2@NS>AJ!^=1Ok#XOJWtnm!A-$%sKd4|AI
zpCn}=U!4nQH=dy8x|x}p7hQ6dVjo!I(wuOYrs4IjPmQzhDwj6F+w$;s;#-%-z}rFa
zwk*7z3~%fIaN)&t={5P|su$F1&pLE_4?SJW9J<ljp&p(Nm4>(7;O&tkc+wWy6*<VR
zV0i0E-*vSRn`W=IDmbf^vwJP_USv_?7>j%+Sv2pOMF;9x={L73U=KC?w^n_~urcG^
zrl6eEfNR@1GlZXo=lMIl{hH6Ip?%@!6{n)`Pd)hTR1mx^2yeT!Nz?^&?uZjk_1Nmv
z^U37l@U}6$ZAJg@q;n26s7enKdGF=UF1cTD>3(s%?A}STvp4o9?>NZ&_Msm$z<IAc
zeAw?`=vjFCZVw;P_UaP<oS{zsj`zJn#S|T>nW7vMQ*;U5PJ*oi;q4}P>qalt9eTm{
z^4M8)hfsK17ERv+-`bB(srs{Gs{X(me_|tg1-$)1Us{Fxd``__HGBF`P5Ee!47tHT
z|E3w*0P9|F%+!*Dc$8fk%I%%5582Xmy&Chqf3T;*+^Eb9<?WP#4>dz`29m=JA$MZG
zJ<enO5;L^VpFJ3#_6<1Ond>c__27E(Pr61K(p3$=paD<cSN72<^uiTem!^3O@EkCo
z<k`$LCF0%hGmUv{lhR}vkfsG8c=6t(>g_n<#Q~`*RX<g2axo9C4e<u|>i2aitQ}^n
zQ{!%o?($>^Y(guYjo$mS2HW}Hjyl&C-);F0+H$bH3XW>8$K>Rj1F0j!+fnzZg~HFj
z$!jOt@Db6cRtw&qBIbCqAX9D0YkmHYw<qB3*gO2)%c*^i!r$0ELx<6%UMHas#AT>8
z%=xf_kBc*PsvP+#yq%AhKKEyS3va{W?VOi*)%n>y;Q2s)&Rl$xU(!?6sSBFdgj6-f
zSMl#x-Y2hAT|#SW18?))NYOo5S(!cHP9ye$a>TjO$y%|MGh+JVF@(1(>Axs_#ihaz
zT{;181MK+r4&WbvxBG8!zRNTAhNsL2;A1?zZTK-+!{P0!+$rkL+Fp4pNxO;J=NXyL
z)05g!f93{oA2r4c*_XOiW4v`vcst`&hL(Nc@BYNRfPeA$!P~OD|MmIX2gYXTZ_YWt
z33uIpV@^E0?T;69)zEbET;|V|U{AuE@X-w0d5&_IseSZLQ%|(K@vrcVpveSq?Lw!E
z&I9kOr)&M-bQ#fRX6?jR!aectkTmV!{y4`y(`^lPlQU_GJPeb`mz}eSgY50q6J2Fr
zG_{Pv=os|US$lU?_}s3l0B;lE?Lm0EiuvZv@oxoFM_m)mx#9433O!jhGP`kx1$|5~
zxMWjiWId*5neR<@c2&ez&PnaYxv3SI^^TuxC|6`v`q^r8PAYYaO5XIW;VVCZzx20E
zynl0;=klzJ&cgcr8Qrvt=W*cE-T-e0q!VjAN@s>Pu~j(Q&#zvZRMJZY3f5EpYxVUY
zyqUU1HP^mho9SzorfRvqi8haHqB*X{3a{Kq*XQ|YWoI9a4fSDmfsaNm_14&F-nu-}
zTOC(>GmqRyvl{qlT0<Yr3-+OBw5dXC(Mv=>(b>KJ8ZngK-hzR0z*`f%ozWynZ>=F3
zbSOmMK85JW-Ztu4AzZVTglpQ@aK#=D)9HR;+EFu1dfHZrJKA!-3w7a*!Mc(WqR1P;
z%u@}<ZxO7P{X>{v&_+GHLUpxfTOGnHef~+97A1tMFuZm9CtO|OZ2@>2bg7+;xgzvW
zz6foq6`}T5BD6dtO3fEXvG>L3+l&~_tBBQ)zhX7uPOOU5i&HwheReorkKpaJjtS!6
zIrV>$pr@k^`g{t16`J=MqfuKX<F6WHQ!-pkCU5+C9)sr`7)wm~aCsW}7x&S$RAuW5
zYw^BJVZST_Z%Y!__N<knj;q+8nZa?UEjcDR<VxmjY+~<z?}JX+CRyY7xo`eNXF5TS
z^pf1zlC18W7jdl$@$4Yxc2M8xw<lTYx0B^YjyjYatGs`T2BT-rU4SkPZ&$L1)h>f(
z+%Z}1rex)Yw<gwO7c}f@O_F6j%{feH;;)%qy&c}(dFoQ5J=A9DTix5+h2PqxN=ID^
zoA1(oc-s};22<nwu++u8Xct~R7yTLZGDp*IRi0S<59*@SM3+*pT}VFJ70tNzXS+_$
zvZI~CRc4vYEMV6yW}6(&YuCWSc6BXp*DiOv23hRtmt@!d6E?kRZqxBeR(1Ps(G7T;
zeF6Pp{VWQbXu-=4Z}VF9*vqP+c&5BASvARJ({OluDF-uC;O%jETNd7yJ3-$Oy#3gb
z+VK*nMjdo&`YVSv6?f`lV;<{@&V1Ra7Vx&=PCQoRx4#ZXS2j6SoIE__7l#6=342F6
zG?bVl?xRb!;B7wlBz;F;uLN%&&|lo-58@48za_g}8bwWec$Xx-@0}!1=7rarN*~#D
z`o*b#SKpVUjp*zJxSAiM&kX(j2z#UPI{LxPWM<-0*X@&{CTHL@+D5UV<g+lh8oUj7
zoT4ke(IN(sznnsggtsq_()$$-*YL+|W4@Wscev;VGvRIjHsmdEc3~=YKD6KH+T;|k
zQ}ulo`Ej9C9cGXF{V)2+*#m1<!D~(Z_XmvK1B1tM*=+nAa@9rD2&?lu_>oWNfR%yF
zR~u3|O_S=TG3S~(A-pyAPS=*p=_(Fym-2H5!P`gh)(UTnv!7O-0Dt?WDbSdvmUY;J
zucm5q80VMNPE}2Kd$o9qrjE3$=XRTF54Y*Q%_i3^tIpy{4=Z3*EsH~6&L?X&x!WnO
zyVPTPCcze4imn+`=u1ywmLIj2rFeg*(NC9w=RTCPKKsyn9G|LO)nN@amOKV>w&Iyu
zwh@mJe*dyJV9Lr2Z66I|y73X--sd%I32!%r5WDQ+%(JcOIzB61A%n^9(cW)W#KQz{
zpAmOVYM-L<lT-9;8!>K{RO$+;njFjX@P1{5w-t$xf{1l%P&-H%!<jC402cg$7lIh|
zAbsN5slz<wo;vY{p1Il7W|q4Y$oji^)}`yzaz;|O=|o@S!aIo?hF(-?Aa#vVE;VP}
zO<Ir7;J8Z-r{L{lHo*OdN%FUm17|W<W?Pb0opoyOJSS&LJN0_JQ=Pc4TT)jk@|L<#
zu~fxhqdtKicW+d>p6pE5$@A$Nj{fax!7SRg{H^e|MLd0Gq39IlU@kt;=eBgMMxSsa
z2J(XMw$H>uA>4!f&F^NS#aYv|3BG#5+Xn1|-(IKb!<96>+>yro$~5iazL-0b_urnz
zJl8aBhPP?lA14<xs}J5zo0X>jV6k&Gy&A-dotR&LrbT-#hqr~`?dgY{0|swzPw1*j
z<dMD=@wkobqNe!V>rkI|Tg-gc-po#;4$-JHvvTQ|^rr{@IlXMH_H|N;7oBt;-Ftf=
z{%aUHZW|skzV{FwzC$^grB@n%UwP*1Rp}!ALO&djy9(kf@a&`qZ#a9)t0OZs&`%yQ
zGv_ltZai>j@Z8(e=$D7LIiBH}B^Isly)K^Ydb*jTo=O(3Cy!xuwEc2zMg3V@T_b91
za$iq*|K*|abv<;XzlTO%^3b(fp4yq|Np4z4rI&cq3*fDmUA^^UsJDh5_a?scR)g$5
zy5i>}FL>+Y=C5DiZ52O%`EBx3i-N7S>}6}^o*1YbUO{?h2+@_zAzI?yMn18jvb93{
zT^UY4ak%y!2veEwZRsy)tLdxTDs*LAE$AMovC{)}&=jnH{tVXp`@t#?Z(F5?s9tCr
zRj3lGXnb1*D~F-`g~`V|9ItFRv(Lj->1H@_N;}ow(oRFNMWFpMhq6h81}}@0<KIX<
zT^OYlcsmK+*1<<LVMnYspNZ9rWpSF{B3|!%#Vc(?ygt=Q(2p$%dhs?va|ReRcb`F>
zUK=>Km^mHH(7N2!thVJXy441anU5QLz=vHHv;g{EzQeJ}c$VO;BQ90$#IWni;UDVd
zQsxntMzh||aX!QB`AKRDZ{6VSEO?t*hx%%ZWOPpYPT+;DyF>rdL-Uq3*QhD899FwD
zg!p$$-6ZWZBuS@OV|mzr;O*+xDcVNdTpz}kIm_IMKj8H!@}^YIQ8Xqih&_E4IdxLP
z|MKWgr>Mu`X&OeZX7Xjt{_D~rm~mm4iyDNBGc{aV=EwPrWtbb0#idzy6RG<oYWdDY
zJ=~f|?*=`s^eGpIw>2xcRAP1_Gm{dv<u9iqFF5o|IfwqB=V#$`yGD($>tR{DO60Ta
z3cTG1ZzsW9GrXM(Z`U-j>)0E7QoN=$Z<%DeLLHy^+jDYT^s%NzgPrtqjj|~Jc?&c5
zEh?77sw;3eq^gzo+bZ|HHucR)FRq(i0p!W~D>*b0Z&nRhd83g-f48Qemvgd?k@F6K
zxBtM~PhFfEj9<Fs28aCCqH`~G=y;@4^JCyKyv<+6p=;NeeMTN$h@VxfoKqQ5iTd<~
z9&h4{?nzGVWah@Oc8QvB9DTYl`oJ}p#xiTv1KxVTTQ_*y0^YXd<8$VRe`hbu4sWOK
z;{7Amsm#~hW4}Dg`}Fluva-HQX8sIwLb9aDkv9dO6Iw|16itjHUxBx)2B0+)gJrOn
zS38)h*t66u)~0F_{lvX`r|RfR9^aeFnNWPb3!d@sd&GMq;ZBqbzZ@)>TLAw#InhaC
zz8k~~C4-VVn}@T$wvjjSeBXGk(<$WY>EvLU<mL~#$10PPHNqR%nsYX1;~}U`uH;Bl
z2lVV?*VFVmGfkP~b=`8L>c%x{(?4^2^v?3+a&~h3spvcT9}ok=*Qv0ta2s-554;K4
z_?htb<q*3rt+%PqQ2M&y?XMrL%u%HG;48KJ;#SqDZq<rY_}1N+|N4d*IMn-#B-*ue
zwoOC%o+aCur_zM}DRRInxzPRhI+gX4T{n)|w1~KPnU_tMdEbA=r3Di3wguAD7m%#p
z+`FdME*<6yG`jS~%K2z67v*)AuCZ3HWkt8)HEQv*&N7xW(0IQL%yS=swz40O=A6-1
zCDY5W_dfnTADfnswCVOo=HqpB(5s0?0dMbgAy*GdQb)9(5O_PCSu9r#^q#Sv10Uks
z%j=SHrb{A*J%z4hz@t|P-lpYe-az(5J#Ct(nynL6J|t21(PfO>gDdE5yV8aGZV*2E
zRe0=~^>gBRq6WW9)R3Vr6@<56<~mhinp211?K*Uu8UviV&1{&p^fn$W2gC6SJ|0cJ
z05g;52MjKgp)&Bc5WMw(x9^BMJmGB|dh-73={l51e!3&||NHHqphGl3V|)4o_VQkC
z9fIz*AXNtzb8nm@hPp^T|07jZ&cR@KTL9i}hPNX|r>ROq^gpx?7x}OY-VTPhM<=H#
zE9cGl?Bf65r0bGrhT6c}3h=fn{(_0{Hsnett)&m`%vG4T6`jM_QQ_{KyV{W6<Nxp&
zJYuHH)%MCB#^>biK4?7e_ogYqk9&;R=u{eK))A|o`j5I^!Bj0Mi~d-RSh9Yqj(ex7
zVAE8cZ$`|-$G1FwHFv5yAH$cy{W~W#RaNH@Q{rtM{j+8gk)g@NZ!^)0+7jpfth3B}
zLJ$AFbPfJiSHbXhC%nA?dw(xjn{$ml)#bSdGrB#rv$u!lMS7@TEf3Bb^Wf|U4_$$`
zneg`7xH>x7*;^MUdFx$AZ(>Yu)xO}Z;t#!5_lvh&|9R_8l#lKoYN`U)eHGEuPY?U~
zX-WY<-FeeWV_&yY%9mF1%I2r8-&@K4z7@XjR%&^;m9q9}rP{?>sg0u*b=E*dB)3(m
zDQ)$Ad0Y9oU~RiV9qk+_zy5)${a2v26$sLy>A`y3rj4TWhAJ&0R7ZP<YRJq`_3~<~
zY%}@(++hlC7N+`W&-bf@tH{`JEq>fiGrc3Udqt!^G>TH;ol&|2Z);DD(Z7XaRcagB
z^4VA&d=jS<iq~6Lyqfio*OlS%^!RdS($NI1g|}Yk44U@BpmF8#Mz_GP)zhfdIYzbV
zX;P2Nc&VFP)NQd%?{+eKp@Cfu$?M9qUXoZFJO6enZ*G?oI=HxBUFtwgdl=rP!rNDc
zIo}Z8elMTQ8B?6U&^lSSIIrPQS%(bp);iEZFFgAUvmj3W;QYjI%$lZewflBFM&ymF
z7TDG6s9iQ@NIbJSnJJm52Pv$-eJ(xrW1c0v{rE0Pr+B{M{+wrAFG*p{)!Ij$^=^?Q
zyjn>b_@9gUoGxN|>aNTKsoLG8s5H)JWFP7UZ%_W9Zx!Z0oCa%5cFkI5)7^G9^;nIr
zOTSJv;*5^dEV|y!qOJGLs=Ugqb<>#J+1IScVf0viGwIY}6X(~Mlx4n&J`R&?qfMI8
z-9*m;|5x0s_Wk+3E@tJrX~K(WhO1^Z$!XELN*0xAYtf=!7G}^`l;;UC4L#LoKl2!S
z@oaj)Vq&c7hnIQ(aI4NQv?}`wtD@a(GIZp1LA&l;*P%?<>(Rj>Z>LqKH(T}1o4q>K
zregTeFHkeSzR0Fqr)`S(&!z$;@uknOD;KYGEH&m&)SshPa<1h?=1Vf~>Ox+6e3`qk
z1>XL4h<VSu6ZQ88dU<o;-)5iwItQ(t-2dM*E`54NEJ8ffnb-fb37>a7b#C<cNu0Ma
zo3pV>oJE(0x1Zr{9(X$t-kyNBG4OT^{Cnk|qIK~p%!Pq}#8Y2jS&Q+hN`|LHPoydz
zyp4yqS38hX#gG@Xf97kEs*rlA`U20Q+*1`!%vT-W&ViR(YvASjH(7(&YcG)>jUPwc
z1q=MqE>D&sP9UBu_>|a=-@P$8)}lMi<b=1S#-%9R^Az10hX0P*aUSy3Dwoo<<|?z+
z;B6xN)}uQq%qAu-V4tcEZ)fktca1*ofmYn;8~P5c|D!H3L36ZDey=9vh0TZ)U00}`
zY)jNidbzIRKmVT3s;ze|_yeff=d^0)C5!Uiv#9kQtICz8zK?hDa7CLA1lrU_HZ_O4
z+57QcyW8Yf)TWQ6U~E~u>#c0c+t;SVMmFWfTNlQAmevhl9C=^&Eu2x7HBpfj=*uph
zsL|~lTCl~cS^F(&I1j(TEQ>zRwdnm9iyr63pNCg3%SL{WwH8I5u;{Rndw-0TvwP|B
zzfAlWVO8E{)Hps^w1w}>7;Di|KdTB4q_42CP2+yEsh@>9RER}W+W*gU@3zmP2hmoA
zPqk}mp+rq5-d)Yyoo8!#-AB+9et{SP-Y$l>xq=fl9?!wx$VBEiGyezP79E_ZBk;EU
zW#;|B+dlAi{}6nAcn`++a_WLFF=8&xgIPn5g+IK-!;tkqr<Rw+k6$ZMS6d`1As|WB
zl6d_c$$HX?INv>4<;o>%b=hR?ASPKwFP!;YvTAYPZ2N%!WJ-#Pli#l9eclXj`=M2=
zfw#Nnr^?DTpIX6nSXMnJ_u@O=`y$+r#C!vZ2Qwe?SeZ0tu>8-Ly$F3~DfiLYbk24|
zW40%xYX;i&FYtCN{cocK@%$Zvw{OuDHh0n=@b)(K-Thmb2?TG$Jn8>x$BdAi_-A1C
z7qrD_xP718`EqKSLV2F^#5~K0uiS0aDY!3JeMHkIE}T>{Rm&?ggO(mkH+VaOcx@yx
z<#}|9ng3AtIY>N#_hJLQUBkzgth@dXQ`pN=^n;(d2n}lY9%^_ssFmO^-|vy8A8m<o
zhkD|1^3>D0p0d96qz>b$7<ilc!b5HwJXF1}hqkx#P?H)SYQyIv?|bNETTf*y^wc4E
z8~nO1T6jIR%+f$*gS~WQazpv9Z>W_28p?RyTW8?y_H-ZWt3G;Mzp;iqYpkQjraJbl
zsj8w`-?-FVneAJs+U^$GbGe1yoDuVuq+*4A@w)h`&uCvQKj^E`hg#~rrIiL<Zl%*#
z160RE{kKz~%D~&2I|Ai?FHi@H1}Qu?NUO+oyG6Cpv;v`;krt}BDs6RebX)DsAEw`%
zhbi-9nEtLBu7hL3_3>^yy`eX_*1QOnZ^)eIa#8wGE=FFHV$==Zw%8P_-SBqZgE-}B
z60gJdcn#_sui@}E3EuvBB!St~25tS@puzC=+iynw+R~_f@U~6>^PqhaRPJNE&VP<)
zURwg^{F+p|J~8-cyK=(Df5tHP_Y(SL)kGEf-J!O99okAQ^DeUrOVKBK0^R3qd#9S5
zc5v>FLq>0MF8DgSs6%^4+qB}9MU(bebfS|*=6G^Ee+zXR3$<{og4)?N9^PI&ZdZ>d
zcJ<+BzuE0TkAew*qUqL8RPzVqr-SIvaV5zKZ_mQpU&<$OjzyBn;DJ2!1<e>;YXV%@
zx|6(i5OH=lw9#0X+MOb%CU0%i9i9wNWUVFY&jfP5Nj6RM<J_mIR*fak9J|w^n^P^S
z($S)9^l^1cFwskH;<YiUVk47o`<is!#ypY6CJm}(Qc!IZd8bL7=AdWHI;j(C(i@9O
z`$w8IjovSxb7tA$?W!CWU9M=+lh)L6`Ps)#TQumgMVnymFkXXb^u|nhTjZui7s{e7
zm$NF@BG|i}*Jh+u4j1#e+mQd(x2iO`@NV|M@;qia$UcZ}odItP!Q0*N_A<N;K5EmX
zn>O7+o9-BE*PAVN?cZrv;!Jw5;q9N-nAHJq$HLpgz7A#Bs6#JuXwMDa&tlA%q~80j
ziBnE$`<Kz)@4~D%yso*4Ynl<)3^KzuczXlh-k?@Dre6|Xtt9ywla$uMr7rW5RPG6$
z^{>eqnu{E_9Qh%2f|GW1OZ?9f<ha{e_h*t)wS5^H=4NWT3sco-WU4A>rpf?se`<U_
z@azuz+|VQNxhwuH7qwyhs{<2>=g?ZxBa`(DT1OPM`was)YdYDb_fG2b)Ty6Ob1CAs
zOQ-qWy|@>S29oa;fxqx}!+Si{X{l-n<2(II-E9#2O(lAzRx-z}0ln!(5}DQH)Ijp8
zls@!<F)OA^+eG<ZVAj`T=6Q1NNa1|MBRpsB%p{$|qkWHD`W<WJ%w6J|?N%)sYUVsr
z=FhA(@!HS>mW_EQ)DhnD-tRqY(PFskB=4O`zy0%UHZA0R+}p&aGDYBUF`Et*CeF%Z
z)7ei}wJ*r}X<(CkUFr>Ac+VGGHIKO|fp`O~k1Tq_x*D{>qLuS4>OaAvmU!}B(|fi6
z)?S@yQEq(kPw~ZP;2p^1|6DwFdW}W>@bHCjJzs25Tl)Qr@ikAbTGWC2@aJru67cpj
zdd!tb3-g@Oz<xug`Dj+pcV>mOw5W6%4Cia+ZL;X@5er%laVj;GKw|ZJ2he2R;VBrJ
zsGfKSV&Uyvc$*#G4uZG0A`(>+uYFt2PP34wK0e9ZwMF!sFLcVr#}O->+AxXQNl#|K
zw1ls@nf<aF-$ZMtw!z!tZ=IU`iF3uj@-^^wFl*a`KK1I%e3?mJ{jNlklB!b+;bU@8
zqE;_(vag`~q%yOuXo}2eEF)K>C?~xAa4$uBh_Ni!Q}pjidM_5IXlz|-BDGR<xd3|q
z+Z0VHN*%)sy&2vPJD<vVjr3gvGuI8)e)30qhPNK@w$=DFjmkqkj{Ep^IQMY?S~a<;
zC-?Jud@!|JQ4{Amb5JW=L9J|2M?9uz2Y0!DE>Kgb*^=4qJ}L67m!dwkx%eHbRlqBn
zJw<uTl8;;Pv@GNCODVcYoVT|SI!h(uR&<tv@OBIQ{DJmjfVXe%({n()cooiu!rLsv
z(KzNacWE-3*8+#uu)p?vfi9MxeVZ6=T{v|p>Q>j{l9g+Sr<%aqTky8u8&55{>8bAt
zp0XYA(1{@)s_5^b9SuD2;dp3V4-Y&f9y$eYKQ8vvT6o(U-tIEgQ=7x}^>6(KGXBv(
z-p?B_Pt;2ZJsRr3%!c}ax5^FP)(vT_d@Y+W*S4vmpET9P;>}cUTyyoP+(L|G)63=3
zf~(TIw^CMkTN~cCgSUMT`0B&fmU?SyB@?_I0dJ4O+Y#*p^}TnXe(4vet+N8Px=0ZE
zSdhxi3E{lF5bY`1Mt571?^4q(d!>yQzip#gC76X36smRUq00G!GZ_wqt6v;vFZ7Af
z@a9n}@Fhw~W25Q2h}PqRvHG(GGoBZ6j^M*MB{q&%R%^T}wU1Y|-tqc-R=irm+lu73
z+eaBx^0`5I3mG--n?W<q#%uMl7){O=tx;8?H7g)m2RBA*%5`$*>~S)+iPM#f@oHEk
zK_~x7(6MlXo?bUF=gX+i%+&g;34HN4X*H~Q!5qu74h!etSY$yvO@<5AXWMmsj$Ik8
z(bd=&YQx)+aTYazJtqcORHKMhef#4NS_9M3Y(vmdi-y}(4vqF|472}#$CC?h8_!Ra
z108ArbBk}sIF*O~n~KbIdUFdOZ(fJSz>OL=5;bL_OO74nw$zEbpe3)gy6~7WiyGeM
zWliqyN3PDtyLLOZ7~*yEiXt<t8hOB?hwV+w7c(gnj{1IMURM^AR(P4n0qJSx>d+L{
z!r$-pOp2^&QYpT!G3<TMd@nzEyD`|LwP7ZW8^qV|GHJP=S+nPv@i&>(xfJ_TGmExo
zqABzGWHVV*J&B9`OtURYEo#*u_P8_fb|Spp3~#%1gR6Z0oY&|OOfKwcRXTmu?`ku9
zjC_1vE33ZvGdBfIFn==VX^;aAg10}85JTLusYW5YzVccgT1H)Yn_aax+O_qBT?;Ot
zRll~&(UAIZONUb7ZAW-J4&M5}+qUFq&D%Kj41KRpahE)KE%(geb>!OiyG!f-Acmn<
zR~7bNz;~4aZ$~cT>}Ou<Ms}B$z*|3fTjX1^*5pi4*FyAGwqlQoPSFl{o6(cGBI8r^
zmc7A1P4x=?sR8h|*9dB&XvW>hX%mUr=lq?*nK$frL)aJDleWU!IDW=Z7xPKrZ6kO)
zAT(Lc7c!>}_N{}r`C7O%j(Xv~tmv_5=D|-Bb^0&f`<8gp-y|x>4X4hdO_nsUf3Bjx
zERwUc3UIauIdC6%+Zj&JBp%z0_pBwneMo&gJGGVf4u>|PFVuX5KaP3mhnQ*Nd~KF#
zwplZJP?PD&{1OkV8rm%~+%W0ES(7&ZWugbtMBllY*|;WU9ghyOJWl2EutrzMs{Zy^
ztsWYWmSa%8ip0FtjG9%$s3~|X`d>=W7J3F3?l-C#>+8ZMlls;&YYKh+*Sc6Vh8(pY
zGh<gDwcxq45L2_>U|(rGdHE;8PZ;|Su3EV6Plt)Hv>3b%;yMM3d#xtNB`=-EegDf6
z)+4;VFwml@{n5V3TeJ1H=s!LeAFD)J^ryE)RpD*Pw`O^9PD{bz2G#6uVD`I3Mc{2&
zevka0EO@HW%^O<vc^&cM-*zp1NL+pf&Y`JT(7Mj^*e&*hqbYWktit-HHgmOzT~Djy
z4-BzuT`s$f?`<kY-6plJT>~Cc6Tj||<BEfG{~U5w!Jlx4^FCfVH5}f0!`qhR_sxsK
z)DoO|%5y#bhvy>hKem_p${~jqdk_b~@mc6*_iwwjph}X)4aKXu9FD@RUvF_=-sW@I
zm+p(U+zkI8yuDR3MLxufPU>2zN0YT3e^-3#pS3Q0_XX4RfLLup3*t4dGVs<f5kJ7$
zRE;7xZX1ig4Bk$Iw<{aKKr~(#pI3ypf5F?(>HPc**h;;lC-=3(N#7m34dVINRHCk*
zg<i|o<mzZR?Ree_my_jRhWf~+B;|BCwW1*!Ag{}gZJa^(fOY)YCGVX0JX@ggOh*5s
z&NF}*xZqZ5JG+t<vXk>~dL}C`ynO&~r@`As@OBWq9S3i__z)YRb2ZAryx$`ZtxWP%
z^dL_ShPMS?dCGjj6HU-lhT$H}40czaBzHAE>ZSp~ZhHO3O)pn^YT12HEiF=8k)7&j
zU+H?BWm8}6_te*z1`TAJ)_~rc2I>{yrBl<q)OxR%wwGxr{t%rT<E`L?#;UBQD*s1Q
z8HYF3@1D(cUd`0;b~7DZBx>l=O?dmNkgqDj+i#<MnIGb-ZuBFaDb`Y3YWQi|AFVau
zfxr6q3($p<<hGfCy8cI?hA$7)k!(R)S0Pvn;zP8)OdHu-wNdxwZS>#eHo5?B55n7?
z@OB8iopB&k9VUdTz=KeIZ`W429-}*lv{Uq+cItk<o!-uk)_QpRU}hZkvpCg(w=c}`
z8kkA{Ep^;m<Ky+AUcBzX;!YOy=8PD8X)&@-iB?+CD7^`!$LLO!CRU(-sb4g8qZkFy
zW0XE8mRa<1dS59{PI`Xxc*f~9XE^-YCyx1eoGZMFGab&xX<&|cB}d0=Nfo2|y)|kA
z^E?`UGBR_IE0;;RXVZV&-mF<UEb?HlC<<?X_2bMXc-w)!A{R0Cw__G{I%m<HztL8w
z5X&=ry5v?HGisO}A!4#CcItR`t(|X|`yZ@Vbkat<?0Eb+|1b~wY87%~W+C2Ac4*%h
zhkjIZsX_*MjRk#~`c}JEXja`4^`mB@8fIfQ<2`DhS8Y1j3}4fDv#x9~={3x(25&=a
zo0JG|>+rX9Ib+mi_Pi^**!Na3Bm10D*K?pHS7L^yr%6%p_9?t|=X=M)+an<+HS7VW
z2blD-x=DRznsn*7S*=R2ht;vD5_{HL_NX^Ti>%aijmya|*sHdjw`l#}Xm{*~MP1|*
z?2$inq7XF6dikw7$;WH#=V#f|d|<A(cnszagvV88m=!*ly?DP_6>Bh~iW=dGJ2w0h
zc1;YkYa_hvyVkCkX!qAn<DWin*U*2d8+ti(l-D*nnpw|1S-1PhbLl<0Qp>5HAx;f~
zx0_+s^aqIwAZDq?za8UUS_5wv&!XlHZ|l~g|D%RWQ>foHMYj*-^*%rze6KPds+sg`
zT}xKo67&c8qhCeQdqfSYb`SQmvE&$EVQC+Fy4REEQa77BobxE*tsn7vZnVDW1$_T+
z$*NQ$S^Id;9EtRajZRY0j?@mwrGMYRd%MJ?()0K@jJ~f_eh;`BLyWKi-d_8RUP-=l
z@o=J|9e4+xI~Bd%sTm`knn}L%XoFL^(Ja^YX0Lfk-7<;$&Sum5qD;)`MNW9rqUG0k
zZ-|3_eMP=k#)?1Fsws0UI$7MJ8SBk>dCbb!&8*tLnY81$k(t%Rp<j*iOfxBB0r%8W
zlk%5^%aaXSxR&#@z2mecSFG~Xj#antSZx>;tBLgRCzXxU8WZcYdz{wpi&NLy@tpr^
z(6+$_{UFZ!Z9VrFb92(SnKhmJ$w|$kc3rccRWmD)=hzi)Qur8?&VDf~4CemKY5#$>
zO}Hv?FXe-`xwyCPz~b4{i2JzjF7AcP@b(A!?VRCg&aB71u(mhXWB&e8@U}F(&Bc0k
zZ)j0*?%k{S5y~&7M)HApU~_`PY8rIIWso1QL1q=J^0c&Sai~?T$k|5@w5xYcr&2Af
zbME!x%$6Tqj@Juq=Jacurhm2Rc@|iUC%;iCy9_05%x2@9m(gZ5%F1&VuqdK7xpRN3
z+MMBMRfQ4p4y|C8KnT~rd~S~=kLLb8%QY&&p-WbDwG@Y5cXR0UD0Db{BhT>YA9MfT
zIU(oZtsfqgv(%!e&rZ=~czcEZrup!;7rc#vcPa384ZID=L;c`ovVMiPy$|87hJDA=
z=$UJuiid-9!{Mz1)&`Wo(|(Zn@DzQ-@U~cCYNwB>mH$l*gY_|tweYY#zQpcHnuq^k
z^;W!v_zYKYKR@8-r5fn53nBg_x83?aS(Wdif%Clv>J!^95(}b(TbH5f{E@5|6Ot87
zZ~eLKPTeC0am{wf_bzA8(dRcF-d--7q?^N%<PC3AS0w4j+$3d=Nm7l}Bywl=JT$J|
z@HX2VdK+Qy!+F#nVDH0KFm{JSvxreDz}uWro|+Kg$xI~=l?d}7uJzE`@*etPa@VE&
z?ppBJP07pM6yfDY54@X3ul7_<H0PUz@l;vs=(ig6wEFjYa*eG|-*E%&o=%P%-$1{5
zcxh8VFTLI4r3>)3)AEMOTiIJxH+t(&OCv>%X{@J}o6$esOyA&bpIgmz2Hrl0w_fme
zei2{IZR$(?hfyJ=eARfDFa4e^RW7lm>Q8O0I^+EHq(FdPeGE`dd!U-Z+vu5r8W|I)
zp;-dee`WwP#r>IQ>#sR{95*^ZJ$wQ+-5RK3O9M6THeRg>L3%nOR7>u0)^DY@>KzoO
zGo~=jqPE*~M?2=&MXF8|{PT%b&-_u!6BMaIO(JECiBzuvkvi5ZQl8B1$l&p1bt6@?
zV5FL6jg;j}gklRtX>3rG9+!`j;VkDgoQu}#Z856ih*kcKSnXq`#_?a`RDtV-N1TSh
z+i3&h<To}>d*SVv6LBj2G)@kBeb44kQ08WX);uz*D!eTOZ?C+7S5J*fcw$uI1EYdF
z81)~XtyRs*={mB0S>wqIVGep!et3Hl-cE$K=cX~EV>KR<EjD>9$Gg#o9-H<yMUAj&
z+jX0El(1_^MZ30mGEXr(y|M5%9^Q7PKHRYcy{uVmvTq`1tjR26d`!C@*yWez(8am@
z|6%yDiy2Ymcc<F&T<CVIznYkJWYX{TP1;n}r0wjPIpOWry+&=APaHAbsMagUdCwcQ
zBBx2g@HWPSIxcx{Of!?(wlt|1yglz{(xbXY?I2(2U4_Tt?X06_`kUcwc8kK<p97bf
z^@{mo`^bNnu(y~dn>82SmSZ-`>(?e_`_oJefF}4CXHM<6Xe_Tw;!Enh<*ge3g8i_9
zRj-;_wVhn}0JYy+gV@UknK{#*eXxyLes;4KQiER(Z#R)MTpVFnDR^sMZ`byHb_HD~
zZ+K<b>-y|bc(5v_5+B3cpn(nrvlhpdARY;Fs{cW}(SPBq-N65&<t^m3E5ZI%h}jww
z3b_;xTle3l514vec6jTYn5Z+<es}c46HdN)o4oi}E;H98JZa0@vA@CF9PswUEj%s*
zlNB~JS$kF`Q^%p6+mRYAT5_FVs7Y<3F6+%XR}-lfpciirPg0#h^djF(rcWYC72vH;
zFd8sE^Zsi&>yY^1UA$8jIs2o4AGuF7F+(QWWGAP_G7~KCW_qQ#<|a60LIe0^hD~c;
z*tBz#O{b>XbZiGP(-gFgB4)J=GwE}JL8aRoID>_2w?Y1#0d}^WLF?xv=t=Vg^(&vC
z>4pRig^4?|7-c0F9nVbFcax3!bB|H+)u=rrp@A@O?rt~qlF|m<>JzKlb)wboYLuSs
zh>|rhRt-kR>HxFq#tw+p@MW<|gtvbX6PI}&t*UpJeS^kyB`IEqp2h3S7y9Y*CD6~4
zpyb90a-2%gY-YLs{+ZvE-`O<8q{9zQ>c~A{k2Y())2usaC}X*ohEFEOgSBJ1u8%@T
z>1a@TeuE~x!b8dYog2)l%XXDm*=W+WSd;RW;q@qNqKDC_x%~Y`V#-KZ?oEB<+(NUC
z;=ddKZ;dC3EwY;M_n1`dfk|`8<vY~oEYkq$G0d`^+S9D+#FqU%E!tJUtk(62QR2<|
zZ!7a=*O*l#-K;SQW@hu7wWhOKmByO2oHfyUB6H#4tv|f|%b!~FHfqJ(vln6<%+z(L
z6_**_E{P&vi=gKX-u8sI=i%*9V(v1X9P%e0j#<jQn6oe%-p<WR4B41o`$F^@!P_@~
z;DtLupFg~<3~#HW@74>TH^LpycD@wNB)>d%KUpF_^n|x<e&PS;<BKCt-H&FR32!Sj
zO=F)!XWmYnc$^w9yuFFW`xxFH`9I#~{EhpyXp;WOmZV0ow-dZIH-n$Vqj`5H>-7<0
z$`j<!%nLE%Gsp#RxAVQ7;Oz={I~U%@!P~1dU^unu3G?aeE5JQ{f%wDjP_spxOI8NG
zfcoE}4|qDA#2u4(J!}qo&+V#5PGTZo`I5yU5r-tO##_Sap!vip@OBQo{RiHjM8|o`
zm4m(U;a+z=|Lm@gO+8f5!$U_apd;t=;LK=u-QM7)Q{8JR$-S1w-mj^mmusnfF*n67
zbd&M5n->0ykLvGwx>&HjtZDUiu0#X5{2FNA`3BkoZ~yA$rQz_lW5b3@KHX3c%6rSN
zkvHdfdTZKiZ{0rK2+vVd<@0T(*6?=z0xY$=rP3SVt~;ai)1>7mrHqokngMSc!rODs
zmin)?zt(^D*HRbx?XCbhd;`^ZVxU^j2vjn>y^t+XvnB+{_nE(}zx%8CT7Tut6`(i2
zz}k#J?N}A4jqrBy$RPQ44$_a9ApNc&)!Q1Riw}ZTsA-6vj0sVVx%3w;3sG=%8)ikd
z)7B0V%nyjr)ZZet-WW-bVWcJ{Mk*ZMc7wOS7l>5Aj|k<u9icsZ-Tm5;+76#bo{UsU
zYqYG1G5S0qhPlM#w-;kIC0CrDmP7}u$Q+GY=+;f*bk3KLt>f^=#>sX#PCK0On##T~
z^H{tR|6#WCZDMoQ?|*RC$<>N04&L5wPtFK$yX_*6q_$<@e5Le8M)iluSx=kw`HMw`
zGOg;wZ0RrGtvb9Boer%o9UisnYLYRXIR36lxw@P6%W|`lGR%ta1!Ja~m2)lY^**(w
zA^aN;NY)Kjy+1;J7;Tq7yzR|e981jm&mM=aa$jBFPwf-`N6~!vNxzv?mDu1<_K)$q
z(e^qqZ_{Q};b0gWXw(doQMs4%_!a&?w@IVm?J#(|i+s2~`A05lyN?4+YDqo!AiOPC
z!Kj@#jGF0T)*e{+dJ8Osw@F8g8euT%HN5=<)0gr+KXpbUbGL7tCRc{H+u`v%_Q+fC
zHXFQ6zKE~tDLNo=NxLw-H;*hD;f?nd-VR4+_Ir-EoSZp@ee7ZoGrEmt_G>s^ihY%R
zu4E(P7$3Y~eBPQI+j@vRjr?}}cji9la%gZ7ht~I?*S8nGYxMu}%h3<;XO%t=Z+9|B
z3*Kgjx8+%zHOEmyV?94%-DW|5ZhnOR<hO~cm&c`!LlX65Bsv0m?1rg{)KC-URt^tz
zfBHg3x)gGb+;cU(NR83A;AT9$^)Ye2bGu~StdFnQgM5Ja{oH(JsBEH#yLYn2vOiV4
zgFla&{7B+~nw1=y?&Hu9Vua5F=yj^?)X~b!{hHy>Bj$s=Ccauge!rZ)oCd@kX+v$Q
z(~X$T$EJ$ZNLNSL)O)@SZx->}6tn>}gy_z^##>ElJ`w%5uSt{Im~^y(NnP6*^yU+1
zZ$3|u^8&qfdlK|@ae`I`CTMG(1eJ%k-QjJmtVUIUw|?+8VyaPhc;DwyiwKS~DLDlW
zv5syGjn%d;u_`=0Qgi;Hmmc0u4~SL25%@0PZ5X_bhqvduk=w%Cde5SD@^-Wq!rKe1
ztE2Gt4ZQ81H$f9CCa6Ip>N_VBRE$}!>+oU>;OexVSYTp;e9~cZ%LHX}OJI&Ky$Rv*
zs(7Bfc?xTINSyYk#&M>4oX&jbxtAvB!$p4H(*$k)n4sNUx85b_{gVVOd6=N4%&0AL
zEkS$hGqbjRoU&fR4^oUcgYOH5w|(~-H3w~{9K5wmMLR!c(#cz#AO78>5>aM-$E$zQ
zV9`D9yGFUpDwo})J>2&vZW*=aqERuWObT@4Yw$~~@5K6rxf7<FG?v(57VOQNA5F}O
zmi-dF`ycw+VD2wR?5c(?Js904`zE`lw`4uT+#T@uBx`&Zyxr81`xM@egtvbZH{N@U
zuAGJ5IoADFFNf~9nB9iP(S{nyYrG)y+j0hZJN%eY^zq{f{Q4qEx!~<?czX&hb~n20
zY(DOSgBxJt%R=P9>+uQuqRAAeHsMAMm3rz+YQVk=lJP|)F>fM?qzpgzujoCaT`Gom
z`{!A1z0rBgbKhrMn2fg!%^2R6gty;LG0R{NTp!HUmwUfEXO`mu3*hq2%e)X`%pmgE
zEmMeT=@HBWZ@q>y;|{Ga>XJ<x596;JZ<lt{lX#GRMf@<fOXM0C>90T^u)yM!k`7G{
zC6{IW#F^+pkwfpQ(6{i;u6fkxn$mNThZuYM9C!6R>n>vf4=pU|p^X>a)uFt*mVR^7
z*YR%Z*0q-E)~uz$XKE@mM=foOsHIi!YiVL7H{G|pDQ+a%@~yg>b+*2q|JFc}@OJKw
z2C~B2>+tsDZZAc?C)fS8p+?MVsEgUXwaw(MT5tZx+k5b~FuYwmxtYrDY^DzznyYn*
z7Rt51g-$ZocI$CzUIAYXgtvVPw^XZ+trWMpm99Sar%o537Jmh319|I&L4o=*Ay9P+
zk>C0Rpr86{)n@YAQT}pG@z<^m{<^*=K<iHgX!FGYrGF1laQ7hn(<w-KB7@Wt-X7V)
z`M-~XwbvG+|I$Nb=tCWNX^7faX`{{P&>!=(Rh2bix>z_|c{<V~JUN`X3E|q57$GsI
zvTN>0ed8Pk-K3`%-df@9fgF(<|Atx3*`oBeaFinRMQIGVuhkr-Q}<#N3vYA5TU(7d
zUH3-&g12?x?Wk69>dXwu-p%9G_ez}F!rR}e30J%luMWix8o9xsQ-2ziv6Q)ziw&}E
zHK+}|Ey(BX@Bj2=v8VaKT=K?pb?_y%HK|BAaWZQ-2X(9dLF9ZqRw9Dj(~O4;-d>z<
z((#S>lUUnVS=)<Wk&}Lbd;giVte9CJ$nU(ept)Ver-T;UtgA&m8ku#F*t=I5n4Z_H
zfZzGqFn&5Q!TMh0uJN$+I<+%+djQ@Zfw!3!=7KaO$EBuLGuo(*%Z#eK0lkpl_2C<%
z%EH}2zmZdr<Bq^Hbr5#es$kMY_-rp})XJkq4Sa-e3f>06+eK((UOQoMtWo*Nm&0@M
zceXbvWP(XwR-rrZ!TZeLR0Zxn<8RuXgR?Hk2cl!ln&cqo4YlZBFY0Ri?I-x&9M{n*
zD=~Mwp;?J=b_?fXOy}cvt_{b@tEl;1-b}7`fa@x+&wKVkbiFV5tTw>gA@KG(G4EaW
zr<I+UvjJ}}!`njewo6fFLzi$U?N^6Zq2pz1z~%Qp?e2GC>OZ<VG=7>xxqdig?a%pv
zGZN*4&opd5{vvvmYT}{$Pfq1J%AR+CJt3!y9t#&UZj<z~7%@FDcK4YMt-Zwg3utY>
zolMd*Jf(i2$(jjo`&4sqjwc%TGh!rS`8TD|e(~c~s(^OJ?>r#^oiLHQeG#WRqOFx+
zKV5>3v4}e0EO`6(4jXe~ZK{GM@;1<>r}(@pG{fr~Zqwk&c!KGT{qYK~0UoA~@#v{J
zdCkTX1HxNhc)Jtc&I~cg0Dpdgw<%W=G=F=7N-s&!3V2&Q7hZ<=1pJT&1?4j8AiT{E
zZ~fuzU3mNSi%~tfAI2t|R2bg&gtt-f_VMqL+IBob?LI~7x9U+!9u%dHlcIRtqx7+J
zw9cY0mA?_K$z1=Oj#kf8(W-eNT8(qW=xEg#)odE0a-lKI<BU=8s2HX1iq)cMw5E&%
zRe-ng@U}u)f_C{PXfa&=>rTANU5aB4E^D+e>$Wp-vJIVzb$j<qtmfyB)5O(rs{AHS
zVQ=G9oon>HI6b_^<Ll#8l3BIE)RSruXP?PnhV1Zo)!GrS{F4&2XF~$BWDLxKFsO2b
zL22_0@|a=Lxk8*<Rv(Rp`>{l#Sye-cjlHQSq4m^)x4q%*mv=^OKWEgv+{CKoOd5_J
zP}q%_lN!xy*3&DyNvB7^VB+VEQ!RRdUu<&$`l4d#vvS(?YB(HOY*#IK`;^@6@=ZIv
z5ir(5eI$vV`Ap)m!Q9u&Ve27>t~_!m;G08_;q5f;?PSi)tP67nQe(J|r{lzO?qBAx
zWiW&7j}`PVx#P`j#XLIL`5&=g!#7Fl!i;s#SmHkueGxW#50lV|mn3ONDQ2VA#IH_$
zq*p`o(v{4w?2L|UrG`qpSc}}Y2fQ6ng1&z|drKcOcc(J*R{8uKb&oaFJ!*1o#D`bs
z2^tag8}ad)C;q=Hcp_@R=elU%0ZE!koz{4lelNTRN1iyDuR)z?s$JVI+Vq0_VdF}h
zBBN}oiuW$hQ>*G6qAnK!chR*1;q4D%>(tV$tpdzt{|1MtkMBJRYj?1AW}+8%wyV<*
z=Hmsr<F9mA{eRq*^4?whkGX41Id{eHa#M77H~bW})TTx)ecDn}lYZ3HDR_I~T`j$>
z=*B!*ca03MqdJzlN_kLM{{+=jT(5e%`@WtgPOmTDrd~StmzS<S^wRC34du6_p>CJ;
zVTMp6HMTd_mr_lbJKIFLS~OF^NzK&jd^2^})m&cicG`g!dTEs&9fY}gd}W8X-QaC)
zcw4$BJw!uV>FpLjUAx^H-L*CStNz;G$e(j!{k6>HuV!QYm9)}dpN9ME_yB(m8s@J?
zpZwJj-u}1{pi6H8<m?ut&hXYV0-sg$ApN;INF^TzYc#ygnGwQl{t(Vp4AEyi(D#D<
zwQ{DvvN%Guvwf%<Z4K4DFQJ-Dk8hy?;q09eidKY1GOw!3CT2}8BoE%j*@NV}-Qew$
z50U!!bEF3R7paQywyPnE^9!R?hW%jGx+wL06Q!UA(K<XPnltmFb@WBF>Sv45!+bG{
zULT{=<>H8o;x+qlyh7f_YcPA(Ab1-EZ%4t{(@PBcwX;EIZ3Ydym7t;>n0;kR(7lod
zIe7eSe*?Wk2DL7U->9WY&*5%y_N`CtOq}gw(#IH+Dh-CqYsg1oQ+3#M;SkpeYNuyR
z+Hi%v5=||ov{~=2;48tySs30HgSWeG;_E{1E}Gq-XZTSXk05UeN>JAr^fG2x*J8Hy
zkOssAwGFD%#Gu6i2DM8y=&#b~;6X;tp)~4hC8JtaHtH+aW&VA*I`gs{7{yc)d4%A1
zF&UKuzxT3YM!mXXU~Z8?2M-wZH~GFtZln6tH)=KPei3TabH3gdX4IY-qpH_qpV<w6
z`J2k|H<f=4cZnUwP_x@co^X~JVpMap#x-I8fuqY|`N$9C$Lv)f{m}Y`kz4!$uOIPq
zivFze;gw!%(mHaEHvHTX@HPP6zN&&x6@AV<*`n#a@zYT2t2B=~r5VkcwYG`;p)edS
zpO1PtIc{xu`wtrZ@AVui4*$AFp~;)ktI+0)O+}0P4?p@UdNZDK-s2a}?xjBJmgbau
zlv5|+?W-dW{kqSAmje9{T`xQPpM8`={n-yY`NLbdIsl%hR!66XrG<IE^X!K|89%rV
zbvGm5&+{!~A8q#=J>2XOW8rNdcsmQ;dK7l5G`yV(*I$vpj-zh4^R`tBmRj}KFste>
zvZ(29iz?vvebm&ZrLAphT-2s5|DvxRrjB{itgj`}J(uAl9%s^Cc)PnW_ZIhJ{pJSF
zyXFj%#|g~yrS`ftLEp%47tLmF8+qeKI2D{bUdJ~jC~aqg?vwAm#6OYnkU4MYLZ;5K
zD)=czf5XT-og#IJSv6}9F|+1a1fDfw-4hY&jHYz`YNTR+M5=Kr=WHL2()YWpx9QP(
zegZ!vyj^)dTDjnDvuZKg+c-w~;O!G@jLy2n=!j>G8vDj*WPjG+!Whk76{CH-V^rZ}
z40&XX-gb;r<|b;!i>Ud)+aB<?FT9N)UY<ETR^^#dcb?zT+%#6}JI2!Q!i;u$`K#ZE
z)%};TYV|%=*>c3`MzJ{grju*K+XGwiPYy~@cKF^6-kzyX%)vUz>}Jr4HwN`R{ZrTC
zUJj$54oy2Zabe~nYA@WM7cC}DzG&3!J4WU#8C4~ZNmlORsqnTmyj=!wcfi{q*81ko
zCZ*mnsm2#{@b0`<Xg(XMou6A^QMF~%)Q{oQ8fMo`c)K0VbRXL3K-h7NdP()Ha25?F
z@=xlx@V4x2^53tVb@?Y}XBuqEv6Nizsa?AlqfM8lzqdS|eEd{nWAIHA7kXF4Q$>GS
z4Vd|y`o&~&-Yn6~>cNvgojkKEytTtyx2tfpGd=G?7Inu1I=nc~Uy%MrdIPe3!egEd
zzj}L@zK^2!zcBrYbLiQ6Ol<>QX1TY6S<?<xV$IcmW>*~gOkwW(daQ%vmGFq6C4G&?
zYjKyDwhrf&<>d_XSo$^ckf*;U-XvzdGQqAUx%k`9+0?x|{ecFvN<HPZoX6|T`~UYE
zbc=YiiV<rTJWF3-Vf6S?tZR6ig|##e-LF3$%6E7>Cc)de>;c8Ub8dUV>dKy@y3F6I
zDf^phs_}O<RoPcf4OUmvlTFpsA+EX{M{BC(;+pb(Rg<}(Zk(S1pC95^e&{KmakZ8H
zt+uX|sH;j%>uEc@Jq>R|;q9#^Ui!Mli*uX3RP>3Lp26Gt1$;FB2eU8=G}gd}jrI7q
zrb>9-R2|^$_;bxPaC377*KUC?uZ23p+n4*KE;)U5Xs0iItiF2r)mNv@EjbedZ8^<P
zGbgoH)?=L2TiRdA4g6It)*pQatr_0tg0~e1`!f?5{du6jX2aa#8=1YaBY--8pz<{c
zQpd(YGPDhnKfH}|529~1NQrNQ)t>zJ3cUSMG(_v*Z8vxu18+atLv^@AsH!XpRmC%*
zDm#I58E%9z6NO%0a$P?+&KevR!8v#l@`tx0k3{GZ`@qAC5$bi1>p=v6Z=|}z+km0W
zgq{_p40u~<ca&~zjnd97Q5whfY8$nvo1EcLI$AvrM{~w;j8<QW(ba=7>ia52T_a=V
zSct|9!_4q@2D~lCUe*)d_PL36)jmO1v@hp<dW28K%W*PZBaY+qLZ6;%iC5h*@f!Iz
z-(SO^4W)^J;qBD{oO!sAIukkJii6~F@OCA<ZF<(E_^a?4-ZqD~RauXT1Mm`cqTWQ#
z_kA6C>tusQR5j=iW?Yu<P44oJxtE0#n9G;I?6Cw5Wbb(G2`3vH)E(Zw>tImsmj-Qy
zv%R=3aW#XrpSUi<+kNo%8w@Q*9z3FzQI+9tVq;ianEV%?sT=wGc6fX1jzJyCgJycd
z=oUsjfZrG4tr6aChqu4N+sANm3cURQE02(qHs%U`fTxvQw+`__)~04Wy=EEU?Im*U
zrOOS>1vaR0S@e4Hsy*k8`cRegr$Wh((@eTK$HbX6<PZGZC71E6;aB@n2_F#q!kt8m
z{NQaNYI!Sa*>vv&p69#tq)^k$+Z8`^UNpG><L!NTTZFx!Zxi-{aP~0x=LK&!jU-RF
z<k0>V<Os)ME6gngZ%rLwZBrPSgZ*m`tfn^P4R3eD=imFY&dEE*WuWJW(C<}+o?`YG
zA9y>2=lM537cuzpLJsXYMSnK4q{sBYuRfAkfW2}6dqkGk<l^tyW7soJzH?~jg`fTp
zys*@{AB@Il#@?Js&YF>znje32)hx_#xn<V%`WDqAPbv;??I$es*^%GgfCp90s>Qu@
z0p6a1x1}x_WjSL&_crJeb;_u+2^s}of1Q?~bqf;|1aGVG@7XKxFW_ye_byI_+~aiM
zw;0a$iIm^Sb{g&Cypcb{RcAUg(YuDrnHets*ml}nG(w-K_cWRpAwRUG8*3sIw?9JX
zS%2*!BV}{ZcfdJh&0((Bp=iaz+;*FzH6d#|`ZdC(l@VI}B7!rKIJ2yNq}-Vyo2O}{
z#`;F;cpyHGxJZRqBQ=H^(%j{d%zVM00jn1=gKjr@@YPP#gROB|P0m`a59gazM(2jN
zNuOi%OPg4EhQ%_E5ieIPkDHklJ2h4@Z(~&g=DvZq-$%u3^5%FAV@($KL(^V|?o5n4
zW2`~DiL1ZVhNHwQebAwP9cYr<SUe2u{pHb!J54jG1-vbH*}(bAMiuE|)crn2?OkNl
z|6}Pa!=t>uuU)K2an~3z2r&pm2&}k|XOfvXE`=h+9SViw?oymmC=@9acPQ>HUfiWP
z#ooLB-}~XYGV#geIp^Mc?X~yu+nB9(MY!K9lK;Bos902vn!wxo#6tb4^Hv#vPmtR4
zZFsvE-Zoms{FtApmz>4Z)fj(A17@;u4^$@K{hryX)k_gKQd3w*Jz;cF>Mqn}a^mx4
zNX7q=Mt{`^e1$*ox!DEsM(;Q?mOPLe!+ZQ88<`JgXhnS1s<q6pbq2hh)*kMLp#^VZ
zp2*}jT6wdD(upO!I`en22W8-&%Z<gGK0QxmX67-gFrT@y=qd0v8s4^|rt+#{8x^L{
zV*jtL^ptz{DY4}>V#i<5$xrzFFP^;QLR<@PtJh|BY*9G-OMy;IEKu9q#0wS3X<Nfs
z<`?|*oO$hh-$-|I`wRG5JPXtr-gcpuH2eiR)*So;|JpUEm0cB*GF0}jOkJq%)QWU!
z1EMc<WR?c6%F<Kzr8(5_Cx6URRrai|wX;?KKsIx%z4fSww-Vv)q<_3r`+}Et?(&jl
zm6r<O?MXiOO{kCNckopXye(ANPe)eyY0hGQ?Yi&JOzHrQEE=eruLC*9D_G%^g7pI4
zCU$73E541W%Qn)5IU(vujyvjy#;SI&iPn@4)f)Fu&S?qNdq=2Ng@mcjgr+KXI8wjB
z+r$RV<qL0%!`u72q}xTKbh}%W{OQAegGagR*l2ZY5rZEhMtPHC)bLV_u2qlKmc&@C
zo*65@KVx-1I!+50;H$F4Yj(eQErhp|;q41}d*@xe{vDa1&c7t6?cM|}eUPYSk?6^8
zNz{20^&>Um0e>c_{-s1b*@<}b6BQC4uae}-GY-Y;ms1Ix^OCBP@OI34gL<4Z=&qMZ
z_YRu03*Pz+=NzPY7QOq;qTgU{yS)~5IBn5Tcx#2X(|KF?x<zjuS~Tecc`#=aD%z?E
zfmV%dU{zF2t5(*q%IslP>L@F-C~f+wpPjQenR&I-&N(;Co2;54UmHG5V*cm0OkLc=
zIwdwfPTcj0yx<bNtpaba?sCYpf<wOOSzo=y6J4JeI-A^q^*x8R?M;2D8tjUmn@#SJ
zt-)8&?BZYxOu7hf7r@&Hc-sTsK7&!FYV-->`^|*6i_l_sk{cGgN-xhB)LflT*`G2C
z1jb*!OuSy!sdvFV2Y8#;AWJ{jr}oTVn*eVe@YV}&Nf18Fn#IV^(H8ITcWM~9PK^~#
z`Lh3J{OHspuIS}X&h217UYMmLtF!QfX6eAREG6%Q+uYZCie>A0{cQQb=9%zzH@s~K
zZ=2Y%^^q$O?%r!j9j<7O=KV(9Z*vZP_~fYGc&Mmz9pmG@pJdB@S+>4hnyv4MU;bM!
z<!5DU<8JcpBiUM0hSvh#_UnW{wttTNC*^SF9Q)iM@^|v|<=(KJJ@HvwwhohDj5(Ae
zgdc5&QPoD_PhG=au`f@1=->M8e!fnyFMYaEpjELgROU$weS(qCvX~i$7T1W!f0Lh{
zsnJrOiJ333fB5kD2n=1|*HX=?EoD<z?Er5t-DoLKc>AP3U-vDZ3^czjiEtIhhK@t;
z3MDV-K&<_e`V+i8TBfC5ldtSsh#wB#HW<@_GrX9I@~#EF+%2?(ud`CeTk)_!V;1Dg
zgI=j>WvJJ`#p6xgyDz*g2W!h;%+>Ly==t!rIs4-zyhP`w<Z3H>_Hy+3weYqiye%|8
zNAbDDX*uLu?4i3}p>yCT+WI&51$p5p=9F(Deyq8NJd^rKBJ2v8oT(PuGL_5OU?(f2
zON%%&bFNiCwZyLoZ{IIB$)8y@dvZ<qv`nf4htHR>XbN?m$8#+DcacTASbJUI?Gbou
zg||8I_8WM+8%|b=w5jGrD}4|qSw0xGW4KB7DJJEQGbyjXNh^DqWb0{?`CF6jf5Tk5
zjwU(tO*&*_zAS8>kqqafO!|7UNw3D6wVztj*~T`_3$f|fV7MM_(+YB+LGacBZ%4t~
z&0NpmZGk;Q^V2x%o4q=oI&-0d4Ap_Rby$zjPCL~1vqL}BrhZEe`vbXuk&D!_2Qn+R
zD>a!-+>1N1^)Q6m>`45Blc{4*$IlOMD>$hCoXt{uJZ2x!c;0glhO?#zZpBjxZ`V}7
z_XTg?HP2CZcpC(7UoOS_ydqcS7U$|;u7MH6i%X~n!&?X39mKVb>jWQbj#jg17IRa_
z(ceCj{v`NZnR~kL?|JI66CQqv&+vBs|JC(B)}((9E#wN<-b&2zfWJ@R?a%ObU+V(S
zj3`i?Hy(iL`EtI&KQfydGFq88{m0E&TL*cLuUKc1{C%ev=4jN19Ic=R)Ui#Tp7`eL
z6a54U_wi)^k*iR8#@ssN%gE2w6?FFf&2!nyb9HnWn$=Ry^h96!au)e;JLc8kk#Cw;
zpo#ef`ezgxI=pQ>j(fZxXQlVZ*O3Ce8zu79jGl#GLwVdOUqxLC)G~>=C7<$?`dgmH
zvH!RS+sW7MD%#Sn&79pH{4KSJ0d{#Tx9iqpyM`3cP%-ZewHLi<{LeSH$WYy#cnY}u
z;q8RN-U{97C71D@nlivs-zIzF5%E-`tDbuK(o5sv?KXd3oeT5T!LNPw(50S=d{JL3
zP4#uQU41qDvA#~Pt*_Mnft=M6Ec?e`eV5#j7^;yLUTCCg>q9i*c8FSB4biuMglNHm
z5dCo|gjp9MO4->&Q{e6A@lAE&5WcCk&6MNQT;I=Wt}EL(hqt#hx_FeXz}q8dq7>s5
zt+TD7)!Q{j{p~SY)h<Q@{)*8#c$=IQtFK4IYVzq=RaKly;bV?Ujn}5W@k;9yk5@8I
z-uL1(<4wHQz}vp?b|W?2;g=J1jGkUyNYLGz3G`JZXmzhx4a|y#OR?G$9>*D<@frfJ
z8zd(vp+|x)H%(NRn)DRG+m~kyI#SW39}k%HJ-ogCoka`gT6AfKMJ35|Z~bUd9(zEG
z<rdD>u`rLBxcibt7hhRqfVbV??I<oo2`iePRpTmJ_3;?z7Op}U<9i$zENYpGCwjb9
zg-h6EscF-~R;=lI8F+*<^%~yJ-0e`>V5h!=x3`Fme}cDLqnsKAZ*P>xi$NV}FxMJ*
zd(!DZPbS}eMIDKAj%ICjFl&eWIPQPB$CWe~*aF?JKORlE7Iq~|+e_ddrnWh`DrXOW
z%+a6Zjk{iC>&13B)dXJ_b-3S6cuL@H2Y6d>$)Q`tob-z0t={U?YCNc}<kGdm@NM$D
zI($tY-Z@Jazbuu2x8FZ;>h1w2aV|a@cv}hHHh{G=xNLYxM*PR&-TbW6JZ_gIJIuAi
z+n8%va)Y<C;q8kE>blf;tHWDgcpCz1UvX_hXTFe0EifxvvvRXl2lkrU;xFq=?#jNj
zWH9xv(R}SBJfn-U^%SlTg}2`Db_~4T4{yET?KF6s&puGCAGNg6IeNJ>ORcx@_rTkU
z@V4y#=j}dz=J)Wn4!lk6nWv@fPxsiHw(Mn3xr9IKB>mAR;MU<h*$%R|bz&dma_diB
zd~CidM#EP$u_G`x8s7G-)KVkxWB9|{@$j|?yxll~euE+O9-xf{lG{G(!`Gp|-LO$l
zi$UjutyjolUHz!VJ!Sp~IYD1?*q!jU#$$TF$XV(KQDcI)A?!<Se!<fLZ;PWJjz+IL
zvy_}AHeZju@)ZYfKf~MK;BCoFW-ArO-~Cta|6SUjs}AgQ_t^Jq4kG?QThFSQt5N94
zgXW@%Scp+BIdy{j@Xggs`JQ41`B`*ZI5ZdDzG%<PRQ9NgyX~qDZ>PfBxLxTQ*&<z8
zKIvNc*d~8?`xCsK3~x>Fwix;CCT7)4;@q!eR+ExPnspf7PHbt>kK-&F*$xlI51cc0
z6P}(lao!#mY<x4@qzims{eC88HiWS+j4F6-)QCq$nK=8TiznWUrY4Q-W>OBh^TY0Z
z9lZUwgGm+P?P7Qv18<wa+m_6TtsP;~)hv@nF(=m7!_0g#vnCg{=r1>$&X5E7!rRv9
z&`y(G+pKn_IPD7K{Z7<}YV5UZPRR`QaLdr>j+tsOl6A@)owYe=QLN8pl^k+e@6aSP
z<|k-V=UEe<h)Lr<qX{(UzWa*)xqRv;Gx0rKAWkM`+{lc(Dz~!K4&FvaXLF`Jv14Ag
zj<-f%UPzvMCtCrqY{mR+J(-uSV_Yxi&?f+wSEB0};rV~Qjbml$XM(YPVeLY$UwIr1
zZ|A|=7<fB~`?(*yU9}3I^CmQCc-w$ncj01~3u6aIGH;8xYd(EXNk?+k{;OO?;rl4}
z8hwbG%VX-G$H`fPvdGs1c@ET^t{%)%bJkMj_2^#22isdy%Rz7ZAumU_h>gEqO>ahu
zQ~k3u^~aS=m4vrl;B8@e`><`Ug7VPaxDTtt+ZyOqJJ73kxWR9{WVV;+?bo;$sf$%V
zNPS~xj(TpT=YJ2e$6EfLZsh5l?Kv)ox$N-vaLIi6lSdzKi67%MGsVL4)q?t&3Eo=a
z?dc$Vuhbf@Q+wzNZ$EXnYZ3X-SM%%|NQ`*&%M7K%+l;6TO@g=g(KOoAmr@DtzMtsD
zd5fNU+rv`@*`BIm_tfn`PgO|s)Q$_DdiIx>;vRXcVtrrD4)K*svaj+w`Re6bUsb;D
zs}*>cD?94xO5b|=`&>O$3i4CRYCq~!%*rU+NXH9@=*IjIjk^U~Z9&>lJV*zOK{~oP
zNE3Dh>2+vhg_%QDv{aZTxi!`AcbjTkbflu%a`lf?m0uz?;e9i?#74=7{Pud?Xw_{K
ztqNzN>GzLOhBZbre~eM&5oSPo$I7=;tOmo|>rrvKF)WTAkvNT?AE#fI#p%1Xam;Uy
zQ&+B6i{f;XKHiiA@!CQzyA|FZ;5vROL3?*4P}hi60KC=6Sef$UbiqdL7v3g(Mg8~N
z1Rbo#EXj_Ey5gFmVebt3#)mb%huO@hO#1GTNp)S!n%mN><#)}h;E&c7#u<aj7JY@*
z+!)?IpvHUZYm2;mEjsbata4Y)TKN}me>cnP3-VzXi-MY?jltWV@OGHd%KQkcX4kXn
z$FFT#Fu<lc`)vC62HJDcbovC)otN_V8gnRJGxQ7k)fRdX3%t<gia50KStflhnY!lU
z(2@)Xz7L1$!rRI1@o7xL1AU5Kh^J1~@*?+0CU=ClVg0hyVhoQrW@+ILdK_TdEB4Nx
zsHH7O>np+R4_7>%U!sG3MBDDlzMv-#=CZMGoOP&s5hrI_Ira5H_6c&DrLN5M2+dNH
z5Ewhft~~yZCFBJo9X5IIvMKzyO>LfX9&&H|qtDZslZtO8+b;7&yV{_;RXO6+m+-c5
zjVx`OKo7|b;sNToRsUc<zy@X+%*^6!rEE<n$940~)>V&eHHxRN!kVqB@OB%#{Sn?)
zgSV;hc4nt+<@U)|?;)&tc>4z4KKc>Q7BNOMIPL{+>+Hhsd5pL4b^#u<rCeuvF@prQ
z{JoNy*;~<&7I8k*k6B9Rs+LF2UOtN)AANz`WW`daF8<`iKj+k}I$0{q-gDK1K1jTy
ze*MS?U6>En5`QziZP$lBu6%0ajq=oo>n^#m4gb-`is)eQb{%?eVSE^G(22XEmj&R@
zNT&X~wmlwacv~IbR!>JqgVztJCk;m{{wltOPChD7KD>4HZJ{mxEtDEYZK`Ptg`*WF
zqMMb6xuaLoOZ+eKe=xp~&H0)}?vV;_*LmkFtTNo)NZwODPqS*^3oC?X%-&htgx>Kj
zn!(RJhO36sIsfO44J6+18G<H{rg1jHi5}))b`824Jl)%nI_KI9&1P@1{%+Sgm=p4l
zdg*U=onYT;ydhoRPfyp{f^@xtw~LQkH8%&Xh%-fIA2(~-C6mgM-wuMC2Vva>&IjvI
z#-xIeCUyLdHTT%8itQ~5hPQv*AO=2S(#&-xtyyGJqdClF=lgzzr_awDRpp>j&mS0d
z?UqqJelqF{ycib_7}RqEbJ`af)NYtTEqZZAN>_uneq+#Lt^vypO6q9TzL`eloG|KP
zbCY(gHIc8G>DS_{&nDE3f46Gt_w@d+O4ScMpIR|?{RVG?Gwgca+b+{f^7hhrs?ee@
zx{%}6$WYBX8F+Cs^dvAtZGtkC&3-+(C})#>ppHarGUAI2z2I}j(&;5!!(L9$|A!jX
zhI%>lX@!Hc2%I{-pL)&XEY3{BBk?(l8Nb=;LT#o4_x~;KYY%wasacL1%*a+sf8x;r
z*;)*5i@@8f@b=zs^fK(j*Z3zf=rufyv*|g$m8;c7@gmeh&kfGga2x&v?$uaw(m#J@
z&JMh7OCA5w2J&F;-vb5o6^Ehclw#h;ZSM1~x%yBFP3$wQ<35R=pQBIT=jfYmIU4pf
zOMCBSY07ExcX(SA-VXmJOE>V$ofp1P<`2w$n5ipUsVxwPT~DN5?#xsYy8T!5P5wh{
zU#f>w0lnE@203LQeqK2iR!7jY0B;Mk&R!ME)ogU&KJeBJ|3-@g_z2)_D!iS5&h>aV
zu`}GAO8z_#J!Zk3JY9piCw$?nEno3G4>feCPDuxSkXGGtw{bp0y2``bBYAf1ZHvc_
z`_>oU?wAUDsrCPt=LR1@-=KDHM#tEUo>4tBL$C6w;pcklc}q{3TX?ESo+sxmcxoWL
z9TVZH*4;gcg}qd^h>ylr^HnE!yFJNQf5Y3M8+=voA|5Ju`xU&s3vbVzsHcyCetO!-
z58t!D{yr5bvu_aReFZ5ZEJ&7!AdP^xYg2<%YetYJ?+(%>c>9Z2DCc&Ds(g#4_@|pH
zAudwJ{75bTntZogq@MSTREfTk8aM<$)wD>x+7hYgz0oRDEk?I9nZpckTO5v2fLE*z
zwTUGkiPf;UIAslr(-1h@1>WAD8;33%ht?FQ2zYxFt-1WB7&RFjqh24Q74bfrGeTqZ
z4LWt=qZl<R7OO>$Salm5s|hXQIQJ=D`-aBr>b3+quaX<rNYt<N7EL%s|Iw#JSwoU_
zb~o`Z>-!7*Q<t6@mDtayyc(?KF$Pr~U{Ify2EDtMrhe3hU*N0y72cMBx2NH4ad`U!
zye$rIJ>YFHdGfh9^e|V826$RDA<LpK3apy-!Kw@7wy)srpYV1My!9`Vu24t1zU^h#
z+Uwj074c}GUzLKlx4rR24$P1{adawc`VjHfSM%vfoR=v#nCUws6Kyh6JLbUSpEDJ+
zAyX+^=mp+MFVYd1N<Q-OG4rsv??>IkugT*k@b(V5a*q-YeN)GwugQ@Q%ynqOHS*?r
z<hJm3JKnDU>QRO3;6J6F{C!b$-`ZK~?e5U+E;jvA-=>)!Hhl?i-3n})+5<o74`_Yx
zw)2g26(4P<&Ve_kFFE|P44L8_IztX$g&v9zRkIY4fv((}e0)62qmFx&n#tealZUs*
zN7V|wd2^OV!`mkC)@aFAOLE-7=*n~8?U@!ZxeeMJZzu4v%GudkKps%QdhOGjzLTk0
zsxhBf;~f{dLC9o$UF=1x{$U>q$7`EI%#8-n71q1+b*-1;73X(f7>4iGgS`%o{^AXC
zdw6S}j^^+^KHP872t%EG-l?zvC+7v=K_!na_L4ZEEUa&Vf5pJ8h0nPXOe+d+uff~8
z@OA~f-B1}n72anv-i)L8QI@=??-}0KfVWBQ@ICdSehY60!rK&hyB*#(hPQW<m_fFm
zIUR?oCHqi|3*cPgKx(z{_78I3V%5=)3z)l&_S_ZTy29HWc)JVU)*`Nd>Ox;s;e7lk
z)c9klsTU^B_>?Orye-@S|0}t_iM*<1TY6spc4!{?Sx4%xtFkiXf(Q4EU#8keXX+iS
zt;XJSezjd2m)W&tfnAsJB9)56kF+9P``6iZ`dm6WRl2rY($%9EvsGc2<*h}zX4b=g
zvtE*;Ug(RyTOS?wrcs`M!8_KOxj1X?v5{U0qb?^JRjig#!yXuP%*&{e)r|`DHmZJK
zBRxGv9hqj-cbkmVaCytwW66^_(=!>)@-@zzoS6-82f^DC^9=gj*`N<!8{`jXt8=yE
z{n!G7Zht}j2+!n~M^p9g%T#9Er|9qZDT-~Iq7Ana>C;QnaCrOpO{xwKgx9R+V)d<B
z&)+j`pH;({Iaj7ox}LvFS5dUbO{3E_t|NWHKcvgCAYB{Rq^r_);^)KZ`gAf~sn^q4
zQ|TJ=f*Ey*cDdX}w~V95IGFxHX77Z~hX;74pZCCXfG45`yd6*7@DSSl>q6P;<(94X
z^xwI|+g_e%$y>9vgZn%H-X2J09$rN}md((cqO#PtF#qFo*(wEZ`<K9*P@cL{3_eLb
zhA+@)i_lxx-@=UN>FCSw))n4v-%fl8Z;!*<ba;EF3I40%d1`erS8puz0T8b~B9?vl
zH8X4At<8$}D>zGcI9qMmb*K6sXI+rL|1iR-lEmzVi&MwHm8nt_GSzTcrndLURAG4A
zDuI2dIr<2FU2O+)e+|vlS$e6y;*9dwvzT>q5RIfS9?3)5s<8#ls-08r_`kc5TUvZF
z)P;D)gIK0R!wlUB&d@3U44osEu@KjsN9SL}-ne(CRdJK7%;~nOIi7*LwVA2#rB#s&
zEpnyjtUr5}`OkFDZ?JRbl3h*P*ky;e?p&kVkwYY--ITNI<|q8{Kd0;J626vudc^W{
zb>5LKuj6%f_i|kgcvDvoih8Ppho^Q0cxp%^Pc6>$l;xI}?mh4(2KCV|uD*CceAObz
zSC5MODx;{c?w0V?OhY{-53ENVR!@Vz^Of}&{X<LWGy1K8-c0e=;dB0)QZGPTI|L|V
zXP_p++Y!G6DJr_5{vI5nE^k6OL!5IB3Ysbs-X_D_V0b$O-g?5@TktluPox$OjKnV=
zDgVrHZ8b;GC*M>r3&p4pydAVJhFQolGW<iXTY|iFTda0Wi`Bs@)O6iq6mcP1<0nLG
zU0$@RG>%q_n$&?mL`ln|RSMQt<QiT&M#hcIXucJrO7PY_GgkGv#&(WXqkgffvN2ZW
ztHtU3j5z)3o1m$VM4day*$MDAe_oPe)+cGvu_U=a#e?-3uXCwnb#+VDOGC2!vyy2U
zOV+c!$?DrNMUK!E)e1<Ffqvh>SIIh7J%!qBiUxm^qGJP7wep^s+3OaqCeMA#dhN#l
z+>m2cjoO^2KrO21KpS&PZ8~`Zf7NH30^n^wcpLM#o!MU*!eFm8hgi$&(pCK?>$<UB
zWol<A*h0;BZiam5Tj+qkR~O#K!P_i&`-y$>-}RYF+f4pS{cPu+OoeWuwltTXrP+Lr
zkBud+HgH{{*YLaknL0F$eg}BF3*L@`w_&#(dU(R2Df`%WiMt!ta;i1Ft@x5z2J}LN
z;uU=m2W#PNDem22)Vrp&uxU8F4H#)t;9cIQKXL<q%jYT?>OG#Af#=#D-tJh-e6N;x
zRx6=Dr)RMz<3|`rK8XhP6S-y3VDvQd^7BsmrEGYzjqD+OtZjI<LSgPW;{0P(==WWa
zrTqgrL$Nh^Hf#SndFOieoWA9<WQFBhVYm;ubFWI-`l1y&G<(G!_NZdd=|zg<EQM_9
zV5wPZ1#f4;{YGwCN~_5}SDkqMwv+qNsWb5QyT6=r#S7F2-may#T|DT2GZlLJQJ+S8
zAIx(tNgnb6-u~C8Q48KK`AGi^dE`EFiL&&s{x=`%-51RAL!)z|)ot2G56*%XdP)z%
zNwnsdf75$(huN7A3bd6x_H-d?wV5rHzl$?5hU4dq;ao$$r}_bU2(J{tZsv(`?XSY;
zYjZ|UWta<x-K*h0fw%kN?a4j)y0nhp=}%2;5x(>RIT}&}ZKEi$2v^^2*=i1NH?emo
zuFTfX*eqpEbSRwr<-h&WAD^z(oXH%WOeMGwyM4}3lb}phgtt@RZ98~7=1ICdTH{&D
zOxJHX`=_%X-GMU~7qTzGTW4?B1fQNzZ>clJqD1Ds9ocEp!tbcr)-`E6ybXk#3+@<|
zU5xk9gk$0DVR$>drct%v=DOi&3UNu(W{)(vGS4;8CsmgN(zJLz{>FVqecTRzhZz;b
z_f+&TYW^<<{hmP{%dENkc7vK*4Jto5P18T7%2X&#m#e3#IzMyc_%w}uovISkl6AaF
zqG~Or*8xw*CC>GJc_mR^pA)sCY6|`DDcW2tRi&<&RIeWW!7)}fNyCSA*`}Z;Hu@Fd
z>P5SL98R18Z!5yvjqrBy+H}o>w=e!k*E4wA9Ns3tTX%Rn1>TN;x7p;mgWD1Vqti#2
zox}vhc;5IM;#eE_bjrfpfiLJO;_+ziuUIsn6f^%fVoOs2Iy1a2!Fo7{cGZr1e<-}|
z32!UF+v4!H7`*KRZ>QtYijK%p82t%e^xp-;+c<dp5Z=Clw;u5J!8Y#E)y(^#9&$2|
zIn$we`iQ6Qi}kr`IE^_c)Ke$0hAN?(xh&4onQybT9*wFUyp3h8>_fNmndOw<0DReh
zpp6x!4t70LPbb03A<QV~o~dTk%#5*|i5fv(|7!+%Lxu))B<D9|sMda76Y|M5JDKgb
z5)Lk<t~!c14h9x$VOPTje2myV3*Hunw~6q!Fuc9A8QqLn=M8oJ4v(x_yvM4w3#si*
zwBq3=E~4&#1>U}+r}O1p>I4@p`gn_bk$$qb)&Ixa((txCzvp&0JN@%^E%8E=D`{86
z8@%tc({*7gy=dr;!??dY!rSVF-Bi-sP0rPBI()!QZ}xb||72a=fVl;;>oU*OLzUOM
z>*G~-{j}dh8}`-J%&MN+IL=dtfAiF(LSATsUYwiirL#YKsV}wNIP%={GyHVpn+CGP
z`{UK|S9kvab?F$O{W}8nA|*(R;H@jXJp*rh$2V5t)lD=A-hO=1RJGu3RcEA%=0)mp
zN~G5LMluhnsoK2?*W7?GIX-gUy**5SW`rqVS(t{bj#Aj77}?2jXFQ8hD_FU3Wej!w
z7_H2Rk+TXc{*$>>W1}@aCtB`}qM6CZJmy>U7*QWSmWE!P-&`TIXT6>!{aZOoMPEhf
zDc@h+7mXU;u2>Fp3&knTo!+DLIMw+nj<YS}wC7-)k{2ha>+wWpmL(~O9^SK0lH`Q9
z3tlCu@0KKeV2;Mkn@Oq!Z~uh1uJHC?lN2q6uYFG?>x-MoT2L-UKh;aomwYaQIU6kt
zao$4ZRL%EI)waXTphU+Cd1%m>aHGE8O-w%0qO|2UokfGq`(V?1D}DUjTSJf8b*&Wo
z#d^Cc6JJ$ZOq>O8+l1KlM@{Afkmr63&rn!j^q46ybUOJey!`}sr%q*mB&H1do<0R?
zYeV{E>L&01!RMaOBHx9#TR69&39;u!csn+m{5YAIvqGlMSIpGsTA9jcJ*=6?<I(gd
zF2$!c!l6HAIdrTrHD+@1^7%IX1#id0+Z42!)EYMB{%h4PxcgT(o0i|P=_nqVIlrZ=
z2={O@xp;g|re-8MwD|yaW_Wx1AE#a@ODlT9%CT9@-A03gq16WCS*A~F51j1>Z&#b}
zJ;8$N@bC}nzZ>vQFF53s^DN)b+<|@6!%FhAR(_y%T!?u_C754bIZL-4d=D%f8303l
zsn;}TZ-~xPUG}51JnkGx-plXVz~=_?JR0)z$FP5-!rQg*b|iaWSNb*=k7R}d*OTGo
zxA69dKqq}U_`v=_n<@{J`#II^4Gbn8xV|q(*YF%Ie~$+Q-d4x+`zQNX!a1}&YDYEU
ztsk`{^K-P`fAV#Ry!X>#X6~1!Ujnbm)?v)cU(UG=)Y7KBCC{Mt)Zl0IFwR>zvjgAg
zZe~-=<Gh7Q1zI?cdM<C<F6B(w)tqz3%*Cab=`AD=ehF_|@bi6!=juQ*y-qjCh0q3$
zSLHdv+A~G)qJM!VK;PHn^JtMj6OZuzW>>sz$LW!Qw;$2_n!(#S@U|7aZO;GxEwNvu
z@R)p{|8+j>@MhlHN1G<u(=`Cz#<CyvU&;QIl&+U0@eg;k;qSml_`OxTH{l(gWu{+?
zI?X7&$*oMfeia>fqft+9lh3|2YVviXjAzi3{xIs?6NB>M?U12qDjb)pivv@b@t2|_
z%M$c%Nvz^3C2_WAvTjaKR{zP#8aE_azQ>c47fWBhPXgz0;?bx^jj0^I$%pZpa4eoP
zW$EFcM}Hynal*={X#Ky*`tlKTWbu3q8=aswClgcx-ul%|)QN{lx@)C}|8|P9pQfk{
z_2wLlS-H%myS?73^k>YY`^TmtZJB+;dd(V^uAamm;hlIb+OcL^krSn-%P%!ul_Tg4
z4CeLvCS6Zp@%b#fI<tQLE@WuTXJ%&A;ND8(o<mQL`tROGTXDeH4j-6D_Z<DD4>2Y2
z<6G)0G4M7R-o|sk-Q<5Z%P&jMh&7{2<AwN>dy{z{54zxSOr)0b4KeAEEG_myV}67m
zVJ$uS+@DQ%<!Vp^>bRcFWviH{iC@9iBRMkSnf4o*qnqf*Q-7o1?ie%Q@jv^)+y3zO
z!$f{JKFMr)8ACogv^?3VJNP`j(ZO80u&xab4NOBHKg0}@NAyenVbhLv^hjE5I<yKc
z;0(Fv(+n*?h)<Lm*)Ka}(7Qkl)!nY|%cT?F+LZs!rrAH)<@gcS4zuegIQT3a9fkbl
zeMMrNkK9j}ton1CRX?n?s`gx~T;S}Imsah&Y~{=}>UHq;FMKiveCS2Y`16=;bn+b@
z*K~`D9Hj5;lSQ41Sru}|s`yhjHI1iltczX6lklp?q{}Tj9q&-O9z;?T<9}C)eYsIe
zy4u!D*EHVm@r9e}_`uomZhF<iU57S!=-7$6vNiD3_UU!iyRwItuXfj`EAE;PZy)Wb
zs~}HL`mjCqWGveA2Tv_%=%rh^UUG-Gk6wGpx289>5O1}a<*oV4ePry2*Q%f%v#0Cp
zu*F}en+7pwF$nE0SUq<&)V?i^wQgb)U7XoO)mJsqfo)CH`y^-j-EN}FZlUt67pm9(
zp$Zris_(uC)8!AL>hLs_K8i3!!`nyO!#J-nTz-Dx8XOg_Y<$jx2S=(^`$%P(r~waa
zs*)X=>TXI?JxJhj)23>CK0;m{BD8T{gr=8ms_5+IVl<f!z}uhdL@Dcjl>WRCEzgtD
zD!VdT!6(Rpr^aae(^%ba6{kO!#_24)%{dUKo=xM`bxOQmFOJu#rSXbi7O%e&<F&V2
zoQ$jDluLi`^Ba8bUc9=$k5@!jW^KURDN~Z<b1q4zPA6+|u@r^+q^NTH6lNUK|9dlq
zSwyK?0B?uE+v@PP7#dk#J)`pbn{+nKrkktqLZ;d@w1HWJUzpI9tmqQt24~W>0uAH$
zz;w+eU#R_=H4JY{X4<KtW~izuLviSR%_ielhPN*8_61Ds18<|?Yg>4G3f@+Ow?Ac|
z-&ivh&fDS+dLGDgtG3M42pBk!pHUm$zAK-pWAJuy)l9|LWzQup9d|NQ%c?kJu1NeI
z3roXzo(<8F8`#u@8gv^^n=Zjx8~6I;BgB-&ng8((UjuJr;ceSR>AHP6Q<Z8`qiO0;
zqjC7V@L+E&V7BZtr>4TZr|sD*zRObWzRU#3qm~OFJ>lSba?cj<b_2Y97nG%dQBIl0
zI`#G!r_LU9>N7RqcjTT6h^;f9(5v(w%^TiMEkT?QZ;SJ@catkj@WH<UQxCz|e0W<M
z-tOUXBR=oocX`#Lmx{bL3$6TDe$Geu9kIZP57w!}>|ISe;1g?)rkq8Lkc8hWz^PLe
znJWZu-AX%krHxabaBG$;`WjzbDU3Q)IQ(Z1?SK~0)s2}aAE*tiqW5b=hFl#Px)V*l
zk9N`WWd>}b7O;#Sl%?q1b(x0`ZyUke4{xZUF^_U5=Q8|vo?ySJoGaKDZ|D&Gn0PWz
zk1f#l3Cyg3x6K#v`P>368Ii9t)U9gp{RL;4S4gev#9LT;D_h&)ZAI$#U0XZ#S32{3
zyV2W?ZZQSko*}<2OHS0tjoC@?wmH0QmdCwdVy*+c?NgJP5GU~w;3L`5+^$*VbobzG
zp~~rUy<t`Ud-`Y5XHL&dSN_0sbxp)0b<-vzTE~6%#mmfb`}ww6+tHCn6KnoME?T|2
zQK#W*g%w8iUqhbz(x}GpHs_>KUHNz<yq)F7j2h2W&bCd}1$es_-hN!3p!ud)X1&BJ
zgtsT3F^e6a$FOATO)cY8i~5t>>3EHDP0*KB5@aiqpfq^9>3F<)GcTtfyv>;rFZZeO
zYHp9$tMc)(y2mrKJV8P9^`9-1D2E%pgBIq>4oQ?pc9M$JN>;Nlv@PbjH|{}g2;Lgu
z?fUEZ9pUNQzT`eV(v=Bs)1uPVg!P;1%RCO3bnPsauFGYJqid#XcvZfxJ?r^Ad2m<q
zs2E<;i5VJ6%-qu9P&3xW{u$iIyNDy7a8JV9aCrNan#)5ppyAcXZ@sdWA4I=g2>LF(
z^=ObK6TH1siaE{j_8FS=EqHqk-bQxBi-=$H;FsAN&;b8JY4qsHS-SF?`}!zrXqi(B
zC-ZnTj3=fvqv6=OEMu4#HkF=&Sx#MN4Z4$CyYf3rwxb>lZ*Ng^jacr`yRi<H8}3kz
zZyk6$nG>JP`7UT!TX-G%eKP3$&Qy5~PSJd?6nX5Y{>FTvH^dnIXHySl9Tb{uW6l-*
zk{5{CcrE{cx4rH&KlheR7wFd-6lc}n^m6r>ZPjmI*>t2Nb=ME*!}!vdZ?!7=7pv@O
zC=spEQL?P6kM2<Rq*=bx@yU#1ULYUuXrfo-CG!+oF-O5deQzf<K>AAz@b-0vm9q$~
zy2$rECyqKu&NCd|y1?7A@b-6jo5KF<@5X-2|1KNd&V~P}{|8qU)OXXssczcR(w#FH
zJmmG<L(`hp<!s2hdh*#_)!w@4!ge=pwR`A7v%2zg@l;u(r~C$cYR<X3@~+{buDv}}
z=AeiAl&Y&g+Sg^SPhGVw=B+wrZyh=5t!m|bbj{(T^M1a{3<}WP&p`?u6|6pY8fwn?
z#)=smsMr&MYE55JmB1jibp*>wZ&A^q4drqJU-R82<h-H!=|-qL--oLBWBPcjhRM_*
zO!?cwGz#8!ug}#aTt`B~^;NTQ-Hi)Z|7GF&-LI*(W;RvNz^0nx-c&WpHPzJB5qkAB
zT=sS0nzSTbmq&!F_~dYH%8KB;FPM8LN*&;B(?iipn@;{5%{=FZQJfbZrC|-C6jeV;
zD{Dt-6}hluXskwlh*eUFI8~`0r*?JYw6bZO7Qx$&*|7FnoVI(#t1!G>`6^EBFD7bv
zT#}B#+w7xB+P6K4{=X#oeN57z{A78KOV%}83U#Set%kRE|47rrvIaG)Z%}8WK_{s>
z^=1;pm{_xRMsv18kXdj4F{%Ge&N?b;)&n04XKUN!>4`QqC0$3Zq^m+zG#7kzx%3kC
zXhbhQd-m(H87lqUuFSven!d)a_p|JpF#?@ruwBc0P>Z<0n#T|QU=@1=dRQ9%=@C~g
zI*teA?IW{}mbPdHdbjbYMIA@e!$hu~GmU-+^4lFJn3)A{>s7$-0dKFt+y3x&A-r{k
zx5eS@vue!zcy3kQy~LgO@n6@r;dQs^265`KTxQ&nOP7PUE9Fqx$(#>C{dSI(TF)Y<
z(zg=N|Aq#&mRirRP7ONl)W2xK<5|-_Fn0n>+(nP)&(ux(^<{q1SoV=M>;s3Ks!X57
zTX;JQ-U?1;!`p7~b~e222yZ=_qC>;mmoD@#)n?Bkca5&VEZmaJ?1HUR%kwq<XxZ+t
zm%XYkAM^bpOAE+jr@-1Yt^ydl6vkd@fKR+GwOsPQIo0TgF2{3y=1|~ThyI#JEWg;H
zK|Jnu1^;eo{9t@-g#t7I>Ykn+)X5)ZXyH0~HQ{Z^6k;X*cdvajRQ?rO1-xxFpVzt@
z_427X>R2rg-)^4ruj7Lb$;ZEwPv0r?7khE0QBRmUkhkQv&&DwG9^Mv#w@arp=lDy`
zt;4f5IT{_C`tiy$xhjp9b2PjSxXt_zcsp`7&&>^Qk^$|O^R#B-XX?+s=lds~5ca_c
zG``e3%zGj}Xx11X9=u&RDMQ2jtvW|;_f03W;_y7>(Zl+9gjFxt@0!Eg-e?==?vPKk
zva0biwBnKYjvE;@@4kUERSeo2N1WNpphxrpyeV!}Va~6<(aorZ2aWQ0W6=7s25sqL
z(2vw<i@itxd1la9he7XMQe}5X%Wayhw&gh6WI~)W3dd@DLaZLrH<&sZJ?yt=U0WZm
zyykJtUx-u7t+5)u9pA=|SS9R?)lsg#yJK~HPpo$BkEQQ2mikM)eA<!gy2h(UnRxwk
zE>1Cn<E7#WD)%fwM_m$?@g_m1zE4z%;qdl+l01~GO&yaJ$GUEj0Xq_mvSpj(&Ayz2
zH*gb-IcQ|QjSu;3C~LNUx^`7hS4=VLe`WDyz}qO|lQpn;dnNjL@6bQh6>kH)y+&MC
ziF?dKy!Jipd&ipi3RZTzhYs@(K1XUUOW<u|cpC+8Lt)<-u9?;86Z(h;f_wVXbhH=N
z$+yH(N8zmtynW9ah#(GZTsVW7y40tx5#MpIMLg%{AEg$;8uWs<|NP)o2)rFPg0%#1
zb65j`FxPJ;o+o%ax(wcac>S6>@u0A5jSXau2VTko`u|J8+oRmO)ngs<h$m*CXQFQt
za@H$0z4S`ct~M!ZT{lJ3A115hkQDv)!Jv%J%rvOL+>(hlg$<&|E8Rvv3VqS+A=$0b
zdplYb)q#G@u@=<|K}+b#EZq~-UAfOv$z|V@*WRB)emc;q;orhb?z`de_AT1+$<kI;
z?Qd46-uT|Kt;~|Os-%}itpY7NkjOsQz@mL2%=G2j{wumN+VW)jI2Xg)AC6gN<9RH&
zh*vJ$rmu}Qm1M@szN6I44pSrBYm@Uj^L{(Ks!tzR`Pf|XO}XmV6R!Gj(}QzB>S|nB
z4~<{xuE~G8$#sOA>W^}x_svZQ*Sn!1x~o%wyLv>qtL7(nUA211G^(z;&Ggi^EuNZm
z(o;dzy~y{e{hspH`SLz$=ky_`^3l>nU-h#1%00_hFYsPX=~rJf+6F4<c%aU{3RDEV
z?U5I(!&$)^ay?j|&o$CMuMmw)rmrZnu}1Z1tTnS5<H2t%Z=Vpk@wr05A#!QP<1`+R
z57FTzA<T{sQC5Cq&T(y`ZvvZW<<lmb=@rVXqEM~e94fDqVM<#Ou2+l0^)tNfHz`~-
z>qqE7V1)XGN64>6g!0pxs%SFvt?)++-VTDd8SwVEdQln(Z#R~S(&q9}8s0lf%YKYv
z{&f^HHKG)_FN)rxD7<_z>PQXwXO~$0)iqXad!s#F<{UxhN1B-5{M)J|HCvdZR_zj0
zP%lxznbdn(m)G%LG&q^6SB=xOWM-Nkv-h~yGw4PFUXBq4MV~ilBs!YcRFm4_k90j`
zLN}%_p@f+pTbr5_`#rDA%sTYC9UY0C=ca2$8T1wExn8{g2z}%92J(oh@Q?b%%WRu`
z4K}rejlCjZbs&$4;eraYA9BBK`^QW_tXZ>Hn)Tr_a|a%nH6G^nd}r2>BNm1HK<xwH
zQ|@fL>X*T<AC{qg=snB%oyO(VEzko$GNa%l{T0R7YkFBs`kaW@G}NS9#Fd5bQLC%P
z^Iia#-`W&Ko-+?0&5{y!jVz1*KG3cS!_m?V?E9?sH(SVYH^MP^>jH0k!`n{qwoc<L
zg|LTYuon!W9+b^K@``I0{Yi;Oh^-Ty%*bc9?L?=xu_wG<g08!X-1HP*M_*DDysZ;L
z9?M=ciQF}m>%Un)IWX3rpPK@gkHF=cW$>85+BbY`bzvCK^C_~BJ*KNu_sD6d=Q;H>
z<$t{W9fsC`w=LkUdkOfv*P-5X=!-^Y-wSUWz}v~>?jLqKl=Lexes_mH|4V%x)*juT
zp<D3Q7v64+XT~6Pr=55|%D%?SieK{txns$))JP*UwIL!?y}ICoV~=yC#{9Z2xvR54
zBU>{^F0(+thH*|$Am_(iX7;s%a|!>ZH`pytk%gJbxtji_@6eB1lGi3N`*R&$ntbAn
zO1utL@O_VBFUE(|xwA=&(k(o;>eNsxey?;@G1yh_d%I%cZPY*1$<{C%qMcbo`=_dn
z2~QHSVB2nHZOgT&>thQ!la)F;=Y4cEYoLLzdrA)uUgCRy(B~R&(s2C#Ki@X!@4^Pn
zg|{nP7<Bwj8nfzBnFo@p`=dGMYi+9D4o=h4W(Mg^ny!ya)9Uf5I!!(1$mkTkt(_`=
zc>5wESq1R+8+iNE3w)5TqBZ++w7xwYt$OgbKfFzbx32Iu58e)dx3Tc{CD%lF+Y#RG
zfw#B!#p=nKSZ%%<tH3w0s^=f4?0`5quf%B}d2hfS`uySWo4k0Hm_VQYy#x*Xn5djY
z>PvV#PX3*wrL5}}WASV>N!2E2ipn2I(S_4C)gy0R1#f2;LPy?*uc5U~t?V`xJfiLb
zZ$rqBnv_V_2Y=QmTqu3lA`f~N1`UP-!6q#rzh8z|yU1X(&f@3jd<#u;gI&!M@NeAB
z(9DI*)xZmM&q!YiKG~rEYQH?b%jeS~Gt{&mb<m%Q0paba&-5Ol>GmAJ9IO)PD82Ao
z`|vxLnpI|=Svx!|T7LwVq}cTX^`7eZt@nS6SBRYZn_Rn^pm$eI#iz@9{f&B75}Z4@
zjNZo07G|uX5pU)1-E33#Mw_~?Bj@9K%oX+6CLdSMH+_NT-jX^u`$RyyRF#Lf%~mJt
z{PJY&+sL_Z)Zb2^SB#s_nW}9JdQ6S&Oa-HYjTTjhx888^1ozpaUKZJ$)c)D8elVN0
zj+|uT8IyL;HF4fN_b|O)+RJN34`;1gyoRo3b#97I*xsy#J<ZHxVs=Vxi$>SAs3QAc
z|K@Ps4c_v2FuV<IYGF2#S(7-|<_133Zv*hO;_Z6{d+%AST0?&QYbECRa)wPgI9{5o
ztOpu&nyZd<apjy<S51Jo6IZ$F-D6kH4{+0r>27LT(_KZ*xe=?mY3)ci4VvP{xh8I$
zf#J>^D0i7++|>=edCD#i?dVfiKZSTIE!UG?6i=Oe?5X@3%&xlYt@^G$8fo?6j7T4C
zx$2|1*FLhC|L1j{`lGJ`H_~quQBVGl>T9RdPd&_jocZb}OXCKrGN*w?o@=1n4Vh=<
z@Ynqb{`%nzXZF74tRl`S3QNO>6&R=<D+5)nVz9o-Zpa#Iq#3ax%K4YMRvQ{KAG(QZ
z9d4rSB|~*J0gZYT`t<xTz3U5a=Y(r*%W&No9<JB*A~eA>LWfR;%e7m$iZl#YLHTe+
z28DB;ceuP4MCdDgS|h@nYR7$a>67BT?I^8=x5XAl$@2esn+|Uu!`l|{Hm7u~XlPP@
zBl^`ZtZQap6}k|wbtMwWInlTh5)=w=uO=m`S(ik;ua~Nr)u~!`CRICw(zItpnu?<-
zkB7HA8ylEkWzf*025q}*&?@dre|Wonm`OjgXVqn1&KfVXzGLmptZmh!H&%Hwx2m=;
zb0X*9QCUfy2;TYyq{~1LPBgrI*$Ga<+q)^uc8(+mCGI=02d~x=bd;$U{kh7l*7R3+
zPcrMrpUir5&a67u%<4j~(>geOYzVO)Uw7t-S=Iw)`q+89)ugRyCe8IQY3DV(q*sh;
zl}^7Ty^JpK)(zekO~Pv$!dxz%`^~;)T_}szOx)U=eP&BjbhRJp!TiK*t@7yLtlv!P
zF{6gS-|h69Gf$uoG5y^kPJN!@R9kqv{T{Wfi?H)|blFKxHS0q^FsyCYjd=T&LyPt~
z)CAs6g||P$W9wX}YA;4phT9gt-v!=Qg}0AjZAVx-<1?P(znofg-^m%B=&6s<i9b5k
z1>PQpu??vEc6;tr{u!qd;cdWjW_El-9t&@0WH@!u%5w{&&iR44xUl}&eur9Zq<*y-
zAMb4Xo+r}tFp2pFPpD5H%;Zcv&WB=_;X1tIeet1n$Ajre$<zXJ$DZvoG#uVGgtsTW
zU@W}-9p3)BjQ){b8M=OyxCY+d<0{Sf4PAq_??G*hIK$*ZPSOnikLNk@clRX!_;(o3
zHQT8YY1HlfoLcw5AtQV4;veY^!|&RP=k~8B-t(ULIO*G}m0=`SFlsOVzf(4g5~`!g
z;kn)CZPmXu(c_+4RKFWJ#U8x*x2ai{=IrTKsp`6!vp$-b<^HEx(d>mbc)M)4MTgFt
z)z)d2-Sj`+_Tt_x#C*0)c-tS|E`hh}3mNbyr^yOmz3}_K8<Hybyi`5o{h>os<y|Bd
zKBj2ZhGf0loutn#lXPTzoPJmtr`Pl2^kQkO*5$<Nj$e!x<A<zLBu15AMr+ZfXvM?Z
z-!{_6zb;yn*G8-4>S#Tf8m-1%qLrH%tu2OV?VlShuiErJ<i)5h{r&yvGYo{c&dag-
z=VPoUdBo`e{R}U%V-ysGSMyRVa}}BUYKT{nw0O;LLI2>WL?semSFOQ3ol41C_hpLa
z)K5`9`c)J<Y5k?>So=~ml=yi*dv`p0@?4`;$@zG>mZfRuo-_?VpQcBRjp~EWSuuzl
zjT-iPZ<9LLHt7W3?6UWa3dl5R3Tyr244Z^b{T%<sCiJT>SSJC`?3(t(t^=>_T0b4_
zmDo0N8XnA%^!0~YwBVDO*vX=r^a|8(gtlMRrqA82s!)-$&e#iXdr`}TxBf%SYD(Oc
zOa9kqf?2PLzwa%FXRFOBy2-2_+o-X#{tFeuTLPDMg~7mld`92WTS1T3_vDJ9^e9JV
zq5FqYZ|5o!XH;$e)||Q22B?i*A(u{}cHqyx=4{VQA$a@g$7H3<OV*r4T(gtarg5?^
z)k)U=YRQV{HT{@~FLijT0@kJ}^HQ2}h|Qix8gzY`Q5Pf4Dw}3jOfZ_<IFs&>pPYub
zZ@QV7%WF~__NlY*Hu+zZO8S^}ky&Ma@V0ar3*J;_1QK7Y@TZTnnT7McEV}xUy|BDR
z7s6Qs@U}bq)<Bz?8l71_FugE+Y2E1uZFbkBvFA;SI!@0vapb0b^og=ICXA{hza4cn
zxU`GTS94Xfm9FXqZ#Vn7F(=GT?W(z}{#iF-3O9WRbGJ-%)1`54y3ONcxLd}}T|Jk&
zYdE|;+Z_%0Y+W7f=&69wo;rKqQ(dZis{r1Pgtr5e{>R&n@U|ShJ<9xyiF>_O-P21m
z;B=6cn(z*9wdzt&me=(Z345>mHjsU61NA!IK%0I2WlQtdpyB>nb;4g8;O)vX0jlNa
zuM#8u6@1WNr=I((NACc+mI<Ve5U9521E~=PYGrVcl12vUgQKB3HEbel>n2)tu8DT{
z3RlsI;q+&Q>s)rIOr=BheZ_G3C5CH9n{WmAhU?F=;mqs}lQStyg;K)Q@<f;lZigvv
zMYv8}h)`;T<VBlSjGN^9yty9!-b`nXH&dxq&GqP8>Fq$yTKE{F70=P9lVkKNwO!YJ
z)Pm5UcEQ^!@V0q>n46ZUn)Dcz`z}#4Iq%3yjd$)J#Oe=I^s_Tn!*8Uj=fyN;U(j3B
z!=UZI(-&A9jgNl7)%Y|QyD%H)GjSpJ*}o3-DtsC%=uay4z@!7@7=5@GeuBm6tn1Pd
z+)J#n_r!%674Se8Voty_a!{*T{VSLi47bnnvmCr#Szy++0rX=`<i1*AMsJ{}5Z-3O
z+avdRe3qZdnmf;09F@Y?zcDFfHTSQ<q)O}wuU8wjjs0R|Iir?7HmEUm=$+_(EgG=K
z@Vi*;-1mdb`ZU|B@7mgkU2IAmX4B?5#P`Q+`sa;JgZhx8zfG5|Ke-gVUHZbVexK=c
zBSv3GtbPSeY&N_-K^<!QZh8#R)|U37mIZ4+qLJNtNxu@jEgnOS%k0z_<D6<Z?|-~a
zg|}(&widkgBcJO{&RcOLp5i&=$ME(Scsm!~8sTjxcx!;SP2la<@OB)$U9k@zazm%O
zdpYGz9qm;+@>=-T<tkona@GqA9eQbZXhBbh3azFeiWvrNesIWuKcrk~2RaC6A0L9R
z@HQ9Tu7$UA;cZ#`-!t%~^$g5Zee}bgAsOm^lU~)moG15tx>gd0yj@D~=3;ssHp4~c
z?3cPs&*p$kt%SGF;cen&2eUG%y9SfDP!Dc+F+)Yz)8_7@{<k|rhOOi~T=REw#?Bso
z&K4f$W@r}m!Khjp`mlkz6@6<Xzh@6Y&yBxDzf2AGNNS}+&QeE>Flz8tgA%71)aqfX
z8azu;kxMBWPAw$9M5>;bVSf&1uj9V@0cP9(Flfi-H2SGhkRORFr=&42EluCM7!)*w
zGclPRQ<!^X8oaGZ-%TXEHNn?^N2F>8XLV()OVKFaub-BxpYu}Gs41Sqi^<yiYmzMR
zw!)4$)rPk_;q9c^vG}E9^uZ-Y)&Auy&s&^IaxPlYXQDM4-nN0aKEFn547_asZ!1oZ
zR(*IoJS18x!=rUD8y-)I*4`5E+BZh;I>sp4Ggcn>C&yDea+IcjAKu=Dx6|=urq^KJ
zT?Cwu=WG<_?Br789Wp3ho2SL=#C~c?>k~BlDRDJ5-l4%s3g4ehZ(a&#tdO_ko&L8c
zF~w9~GiRE%bWKxzcspoknp(r#kl%^7dELy!80qk9KCi#u3}$ukn$BrJ-ZtB$YpmCQ
zN}5%Tc=b1UYx{~h)3=!QOzk}45j~>t(!UdNv%{?EekK{dHtFyw>QeCb0UTNVj<;N8
zomS>Qpryjwt>~>MS@+-kV^JwMd!Z|u7P(-+Rk-2@&w8h6R(0mvUQU(g=~PWRma3zB
zQ&oChs#-rFHp)xW6#l<2pQY);!!(_JkS71fY5MAlL9Wyd{t@?(%_#LS>dDte9g8u_
zx3@v_cr7l%+l3w}s+|F6UFqYfn5@t$$?8xqS-auw0C;=8OfqN8p&tw(-eMkpu|`Rn
zav)KrqlxPDJX!Voq^e0(p5xUtEjVFR<<Iz07n0*7B`C2GbB><VVY??@FVDp*xmB{>
z!`rAnX0=~n){=u}?Q*xs7v5Hgw5T)t<U{tzR#(hAv(Bv3%h3Gb?UZ!0RuNMcDb1`g
z7~7FN+WoAF*+wSL0wjNiw?Dz#0q}P7$U3^Ry$)xOy2!1o3-^bMhH=&X%|*q}x~R`T
zE^5`>RqOk^$)k;%f?K)CoJ)?|(@lvh-1Oq3n~t=1*HL^`n|6EXbWB}UZ0D(=JD5=g
zZ*#7DY9adarfHt4cGU|HwzmdE`Dj#6AB9_dv?$zLUnY909=sir>ZwoNn6oh(PIvKE
z4ZDw~%<xfiDPR4I27R1fqnJm&8nd9DE+y5M=k)q2(a}$tr5kA5M1Q%%+dtrKQqKT2
zco9JE9!MM=DC6@$jcOdEQA>l=p<J-K_YT(S?}PPjS+Mq(3eilz5Umdh(aDb?a_nfL
z-@1mX&XrJoQ6^0OF!spfFipK4rrL>N^y-CaSj#Y7A0Ebes^R)!12y335&E<yLL-kw
z$Utql(2q^g<2ZK#-k$45X>yO$?;81Tu^8>5x7P)~MsBNE`5z|VeIBdsE%9RXWi9$A
zr~$gk3FbH7$-*b_V~PeHO3~H_DXNsjn&f^C{5O^Pa0V4(Z@M^#+Lnt^Khx)XjhMIi
zQj_}LH*waAQAO#)Kfl1FU*PRkJ~kWP4urQo;q8eHX62)S%=(-B^ac#Q!~Mg3)S4co
zYOJl}=+fKx+M}#rGrV04Zx_MaEqt!wk7gydH8b-B{bdY3mfL0>I)jG&3rzmjtgG>6
zd>Y(WtmRz`janXM<g9M$Z`59z7iLbt31$hf=eEPsQV?LI_lLg6Xp>S|zia6^E;NLm
zr-saw9&S^Wc{bHSANw2Lj_;MO*RSav=|`=3n_a!AJAWKyM~}8^K`FZ`;lD4po*5au
z?cEq3#vpnO;O%G&dFyLtaKPJgu&@1ASl7;>JhMYF84i7s=g`u@4kf>LsLxyST$nqB
zoVOIUsA;qC+@6D}Tutx7WO)0At1^0aWj>cb25%GW_8^D77f6o}ye;s6(frN>{GLMX
z@q*+!bTHGQ>+p6CyzRfz!EAVXi3c%5D3uwUUs1Qk@BJ1&)*)v#5<d(@$NTb?U6U`^
z)qX!SiIy>oX|Y|C7TD#}*{;I7sO@#9SA^Pp=NZg9+>x%s_<qasdbB16Db9?9E9|q@
z6Lu}!h=21N`foO+(;~+_lty?cFVeHU#HP3NnS(@ZV|#_>H^r*K)Zl*^L>;v}buT=@
zM`zI=P9MY}cpC$6+ZKYCHPUn~GEE2TrYYrOs&=!thWAV5ysK0#ZA3nMAw@Od;(hv3
zqd%nR{G$}jyp^J)<0+cdB}MV&QZ%GUic0-M4;;Sy<Cl^(_6(Zu-ek@DHCd-mrKods
znntldH~y5Sn`P3NS(c(Qol`XCYO=D|rYHyATAHP5(Dx~tw->)~<wRL(GnchTyheAY
zrt^c8C&pLWqLdFu%`kPuqiFTL9<49o?e3$|+6il2Hbl$uE$=srW=2>vGx5o5i_q8a
z8LjT{wqX`u18;Z2+r9N;G@>c%mp+Kf3uEY|jn%giv78AMqfLcidhr-#SC7$A*BEv0
z6|K`#V^tg8ez}bqvP}}SI4eQ!)RJC2NzfH|I}qM_!rR)NQj}LERh?tV*L(5$!P~Iq
zsq~emY5ofGH~iCUrX|bkt5j9Lo+96!De7`MMR$0euApyKfwyDvP%fyACmWAP+tX<}
zG(AllmZmDXaf)ub^LkSUy<gJ6JOMQOugywj20-kUH0_Q_R!+@i^(e#t2EKU>OqJ{B
zRN1ImeF#ibC;Vw$nz6p)%v#pNta<3IUvwq6Wq<f(fRQ>*nmptAA9N;1Xq_OhH2ywf
z3)6At@ts0vXdbVl_{Ih_OVVj()ED`Y^Vbe1$rhcg&3MR8?c-dwd_GR@znGc2NxKu(
zv22n!RYA47CMk7BlFH(586TOfziLwhf19M^r;_xkQ-b1|u`p^K-nR*{oYfkqjr<<x
zEavZSNL1bZu=YrzHp1Nb>k`#wZK8rQleBh2g8I#8hA)2d(7FloY!$El#6|0z#6|G7
zB)q)_Z#%);w&WEh*caOpV|1nNv3aRkDP=9X_Ry@f3ud{(+fB=ePuK^??>4IHXoK3|
zBPid`pi`?2%KU|XG-A;D-x<{ZTZ4L)GU!~eL7v~$VdhvJ`NP{mZC$jzi;LWQx@gfD
z7x`~>(YA*!deF>O7kjwrNozN~Z0V-#Y-+jjZfX$gCO4S-YPOqx--V~@pu4Vgz%O;W
zt^(m_@q?a9yyVIG0G?_%-czAHJT(ukd1GTw?Q7_%SIpJe@9oKqZ%<_gdny{<CcxX4
z@b+_vmu3ff=?ik-p~b!ARnbe;T6ig8h?g$^>8q6M^_XE(Uu|mm=}gH6YErv_js-SQ
zvrY|EZe0USE9Eb@n*O>o!e7<ypgR`{P=;TC2AvB);|)^U*&t2oAFL52L-a+R5Z$jA
zLVY$wt=~3M^jl_3!rRYZg)!qkObc#>>EpF9?T8N3`s!gS^*U7J-iE5ir%;WF3s*CE
zTXso=<{ZUq1#h>SBb6|<x$Zw;9_6d%Ivpx~b(u2-sT*17O>!xVzv`D*W*5Y&9UkYA
z74SCtBxoqSJ+uu^^OAVI3rm#Zlk{w9k~**C9HX^KYPTVYc^gUE`;2?>+eBSw9_5%x
z3G$wnAlE4g3ZIam?$Jrqu8n%dy6ZTMxS05JW_#HAH##8vjDxoiel;r+-u8jFli+Pl
z?yKM7?Fx8XmE5<&A(M>PO<Ky=8{q9+cpCt3r*S>-Bc>yk`)@|PWdw8dZoy%A`}NPT
z_Y!jmPLf~mH0k|NG_GvwD*wf+FATE&Wl#fV_Wb!X^8&UT)SVnRxjr)o=<7J@V^aBn
z|Bt1!j&Jh*-ZwDZy-+uxv{ce0ZIdRs(>8UZw73lh!-u=O4%dy14cPGE?#^)6;lsyp
zx8Jqj=ljR=YAJ16Li$|qbD#U%2aRkq`U_{lxqR;;{G4(8t}pnRoV?6jmFrCp7~U@G
zgRhGoaj#v>xxt`O^pC8}yS?Y{PO?jU7=G;cx5)o^o~Lm^a5#xPpH=l#1m4bhSXbu`
zGp9EJKG8F~I4Dms4zjq&SsRFMTJBCgeM2pJ2%d(c3pYfowg-}rR2;7*`foDzYNO?3
zRMV#|18<wb+tu*a1#ct6(3bh!p45`zRoIcVot?!k8mI=mT>@|G!rMOZb|kEwHI2U;
z$SiRhecZG4HDyqJyglS+6{)8)mzkX(T31)-iCt`8SGOk9A6rOojDPDacpC_BM{HxZ
zuM@p(n7eZlIS5<xRN`8m9#zXza7eBy1~6-d_w0|e%nj1-yW)#3_-Bqb(H~fj_8-$4
z4Uy;fFP`s=*<>xz%bJWH`CCA?N_1p)mA#We`7%^#f4ZKbsbr1KP|z_vNYu6+sAXD*
zz}f;Cij7OxbVItFLG&Y=q-$<!x?-!Q%VRDzL_zczYTeoJcF8Sj(8+0vted75rP*@{
zZ_hnV)hc+~@ju?ullx<Ps*1wfiOW*8Vn?bBPQDHg!{pEDT6>LM)RXv{OR1U$n|)`e
zYWBqx8AnqqE~RhE{nEbEr8#&MZ>GlS@o>o%D_QC|(Ou${1#f-fZ6SDj@+|y>x7*-t
zHd?ZwYn*n$+j;PI0=!+8kC{Pc6Y?!~YMIHYTq`}HM))!MJ9XoG_NCy_DB2`W`KXs?
zIO6pOb?x)d%=JHxlXC#QVRM{9o;kH=rBfZ|Bye9#)X8Fr`h3Nub@vlguxP^npD{e#
zB0<IWrRoRve~!AD!nu|FoSDhGF+5!b;B8lUYlpWL;cbmcsrWBabtGS!j&Dh0pJ19+
za1DpICD9rd@f`p2cZ#CX)I-^E_F^)#|6NnGvO|i>w_*kY);!HZYj>q+)7NP_xs=Zz
z4#WIXl>~366-!lZKKI(26!r6GMhd<kgSUQ$ELG3UBA?{HY>;fdv(l^D!d&8OblKyX
z8dEo2^G2qsJNKVU#a&v)tmJ~fTsn2srGHPlRJ1@M^I(Z8-vo`leWG@CP9#SGPd7Dx
zJ<dkY=y(6tJW;1>C#u!3M6D~BEbo%A89s-$Pto0(_*%H<Wh`LFDrc3|*-45$lc-Vy
zTv`Edi^1FPCM4){odoii<25}eUhg;Khx;>ud;|22^du#$Ojdrbh6|IGWl2^#yzR-)
z7ro1++JCy#sd^%Pv_vhWzjwJM=Z}M_8nYrppT*o+ncDi>fo|TWwwnA_TcgnPCZOv*
z;NDrjP?la$qyO+<{je52W^AT@{4P@$$ut^UJ5yHres8F$f4Y^Sdz>4;gSSJ<W$N{x
zp4u?OQ(jZOw8QMJYj$t>q<QP@5O1B_=&dO*_Z42M{B3>Jk-E13T^~Jo<D-2(zIu*;
z>ZtFljT3y8vWMPoVL!bK@l!xoKV9zPuZ2Sb)PG8#f;I=r2j1?6rN1@_)C_pL3Eqx}
zw;$kbx-X2a1c#poD5hPYa^Y=XczYn6ncjeE>Hu%k%T?2}R@L;hdkwYdT2npl*3`{m
z!K$4eqO8jy3Wc|~;cZQLyLeSC9eGwu&g!A6Gd@&h?uW7~E=>JuhN*s9m`b$^)5MWs
zs(L<5J!ge$db0=}To9olvm({&Yol(pGNL6IwY5AoZwMN*DO!i%>m^To&=q6k4sQ>?
zWY@x{XwBiz(;P9H_%$<0`{8Y=SgnD#@$k0CBZr2<+a58}<4yRi;O%{WFI}zU@t?)3
z{15T!ev}=(uM(B#pCofof{K#yoUoNS--(Ij>vCqi%~|YGk}lwVT=tT=z*kA?|2Row
zha_oLT#`DDO;SN#?<c47=P^n8y>pVDO-NEavw#y)GE})2^ZJYFYaB+mg`cAWsBh7n
zg8o7u98WzwGFyeIT{cb3Cg&|%ixzPY=01G#GCfuL5J&Q%H^JL6`1TgS*mf1j$%D6h
znzECH{5%J|bsVX!f6-LtUd-0u6ZlKFXDe+8y_x^x?fB>Dg6GKOnU$$oye7Z>K2sz4
zea$VOrMyX58iVhv^Z-0xd|xZPEqa}u!u(Fpil)|vw|z5mb)5S4_S8HLJc-|MQXRF$
zf3O6N=P+l4ce}|3Il#~OU7jY;qi#b^(R(hp)6C_a$kWZkd1?of9^v0;J}yt;L-Ff>
zo2TROHXpow-x3dScjoyP*3-<jWX&8VGjD%A<{Z#lN03d^hipvz+#%%e)vHubZ|B!z
zFE_p=yg_@inWsYg4S}0u9@f{F3e>F?@lro$R(T}ZH{I*Aw}<@wG7VHY9z8p+fiB=F
z4Lw?)9a#8yna9fN%g$qXJ00F8y~A^YX5Jsx_S%rEuK3rU=iqT`l&3jO@R>Iwi>O7O
zS~9EoJ-jva$7j=&x6HDhL4U7>SFRkqZBQXsRq*{f@x0xIxBtx$KE<~(6W$(alcVNt
zIV!v>TY3DS&rIcfgQgcyEmPyDD_@<;Pz8AVcwM^E&@X@TPsbyI9ymWk-p|w3V=Ob_
z+3Cz0rt4w^Ixn*~sql6-ye$E5TfxS=eBC$jc4Gz_@H6J!EM&o)ONNEiCsnvdKc=3A
zx830F0C+nB-lncXCtRAUH>24>JuFqLt*K};NxH~u+Qy8{=v2PuWUA&4OwvE?xtBKN
z@8IoD`ZLX&xKw^myb2T&zFdcTajmFp*W?}!o$4%kM91m*5m*XueGkN`{q{H+;p4vl
z`n7eaePh|JT`^8gK05KvIAxj(ujvy#JIwBs#SV7LJM{HLhuX|@=+Z8S_Fi{rQGPV5
zw0JH07^kG&ae9mXT#Lu)``oFL)18{%(W!1do$AsD?<D??CUcznst9$j318OL1jQCf
z)PSf&b?T6)kMO4a%|s=mb?*C=t_yA%>`_nEmMN*~gwN_Myj=}%Q)k1<^=bNWBu#(b
zrH;av{Tkl>4Q~VB?XW*m^vi@4<?EcH4((F39*w;Wyd4d1f2VG=SE0US2g}p*$?9H!
z+Ot9u`+AdgjKBZVEQNh?DLO}nUQ1iD!V;3@)|dS)l~eR>Rla{O^fq`K-G^TDj~TM|
zhx?UN+2N6Zx6P@9+OazF(yleZE^UFgee<EIm#2^I!7g$y_OdZM^?M-94Ps|ZokaEz
zBx)4A{TAN78OSvs22V)VJ${}C1yeMRdSl<8{2cK13f`7ME#j0O&U0NjPG?+BO*cC=
z9`EYzn`j#M<1{-vUjM+`ZpRap*c~q$T<*os;c=7u>3f$hS72A`d6#m}xbzR_vuTIO
zQQ+s9nUbiG?THGioUGH&Q&gFMU#05VD#<x?&H{Wf-_p1K9*^jq+RBEtN8IR*)#U#t
zFaz&HhF0dVGp-q)y|$TJ1aIegW$G%wBOTsxo`K&5f5l%-ZNB4kiV7xp@^ALkpTBsj
z$|X-NDD0)A4PH8(<E^eMy;X0ExAuhisL|IxdNs{Q-(B#LUiz?y)K~4r>~1|@tsLj8
z_1md&;cZ+kKlLB!r+rWS*&h|i>>qXQ#z48#&t2FhP$l7Qo4Vv#`30!;SIiOx1*qEB
z0s69%w`&44$my>`ZULJAJvmpq0#)N^pc>|@reh_m>DwpO^kIJuJ+D$zyLtxe?z&(N
zg17BYhNv>U?F4VDx2~m*Nww6bZ7uRZYw7ORTDnv*R25C3%4Z8zr{6;Lp>vql4h@&P
zBIK44AzyfFUl@UxAd<W;qq?;)>hl?+dX+R$TbPvDJxag8+dUpJT3jJUD_=#+{dP2a
zhN3kH-cE<Njp1!mc$@XFmF!%rzEq6W;`R>x)<~K?PMY^o>hl#j8tzU7#XFTf$tj<m
zPTf9$*Xj!IFN<T=lME|*LGMn*kyAo$RvniD*SfUtIyLY|_AI#5_pQYq-=7jSxJnW`
zWRu`WlAP0%^l);Ljtojt=T3a=`y~Cu@63zh$-3J!Lz(n{A_MT!`{1uCL2rb!R^8I*
zPwt$-0yDL#7O#)a*>Z62-2iXb!ok0@vot6tOC?*Q8?~q{Z#)I{AK@b{Tbn)b)SS(0
zYh=ILT8Hj@E-p*$nTu4zEcN?_o^0<dEk@%$UN=i=)VIIB&eZC&nVQr$Q$;%9u^yAD
z7YE6>#8>!W9C>`NvQ!q{Uf9I<rndHjxBXt>1EG%`g+_hwr#eaw$<y@u%r-5oBfEDU
zxsiu?3*Npsgx0f)v*vwz+3^1UNW3Cv^K=2;=7YC;;O!!Kdkx+$gtyPB3&Y{l(Z=w$
zC6D)p9D!#r>oz(xJe>k>J^y6Zbvxe7;bfj>Fn2*OGp!bRG?n3R#d>6S*3(FQoi?;z
zTS`58iS_g)KX2bM^A0ad@P1eUGH`ht{+W-VeQ&zQ9>s}u^{Dv&^4?A{Cw00?J@y?l
z&(piU4mKqlHmbhHGQ&KkL0vt|&ebSbdn%qBB6#ZoZ%e`3qVV>7ojjeYNB+?ET&>01
zdFPK@y=MNe;}&weVso`WkgsLt&~=D>|NZ23U&fn>4){%L&LzRMwU^$)aC{>*rlKc|
zptoNoTajs*n!PbY8|UE7o1CF#htf5A7FmU@($#$f*=LcQ`M<}Xga<eu_mZVK>1ymn
zUXw?<9PsuCpOdmWO;OB_Z?oVD;yzRSIC^QjRE1ng(e!9|dja22ShC7}Oj0j&)!Vm{
z$qM3rO+D+eIaN6;Qx&j;dlY=>JC^$9TXIv|k~_g`diQnekK*LL(LX)M{qzG~;tNX>
z@OH+b)ku3Xq=4em<R=cb;+*^Fp-qlFyGq|<uQu-ozp$w{ecIWl<KzQxTfy5x@OI*N
zapVh=f#VmaF~gjS3w6qOiv+gy@s2~)c02Ur5(oPi9hx`Bp$fk^*kk5kj?SU@KcqI#
zcszKUI44fc4P@wiaBA8z7(30WyG@<SXyephF!waP)hy=vnN|2@H92)J*slU_b4~Pi
z+wgW2^ZD?$Z%~r%GyhwOGsSUq(WymKlvEZyndhYAL$u%C^vB_CM|>S8(biuwquQPf
zyEnfkYR1S!H5^CI`+w*07<Ab-<nO@S<a#OklmWjSDQXFCdp$wBg}2S%trgzRfp2@$
zQ?!u!cMH$?$4O+`c8}M-Zs_J*WBbHw)IxTSpXX;zb*b2T{M>t7YPE{Y`ayOba<}XI
z%CYLrEXNFZ`yIU9ji%h=|9CqO-p=z%)P3JXwS{j<@b((K{oE>1H}H-hjYre%#m74)
zYQ#jBCT}E1w{L=4rzEKKRx*3LQn$W}(_OTSL!6tUi^S<;f>Xbg;CcSSeKns`x6uFV
zhBFJ&glxce@d`f`uU5ld>IiS&Fthc?1()(JGNTD^BX+ywzrm#fWEDMz!EMd#j$6(=
zb!qfx=0hF~NLAD_c2|7H*GY~0x*NKljk!ziGn2Y7>kV&P!P_ro$n?CQuD%=b5>()P
z$?n#RHJS4)pCUIh#zv!&54f4E%a4=gc$T95A36V6J(ZH<sb28*_ytd$E##%&*L&%X
zm$xq0_Ll!9Z@sDEqxa2y<e1{a&L<zWg|~g-?Hh-$`epiRSTSFXtLCdLC%xR$zGzc^
zdh~~%w!HUOq9H)j@GZM02I&6A0QJ}spjE2^Gz#AS^^d>$FZ9>mrv3^^^w$!zzdYdW
zPw@6<c>4t22Ebc8y!9_pP2J$_Zg{&3-ntJ6R-==_np!tRzjp{xyRIRM=o!KcbBHbu
z2vLb;A?grPOS3lB(k^(L(<oHc&W379l`xf99Y&r=nBIR0Q&?2E+V%@q#YqvmQPY4X
zVpLE|qy9c`)RYP)Eetbhd?S;p95(6p4wI%_h*JE9Xq96Qc*;M~`twAzR>9y-K`}Z(
zZ@1vu7};-GRkdcU`XOpHTt+{4m=y6?+R2#SgGY85uk!l5V`m?WUGqQLRjrIesgVv%
z{@J0}gm^uHw;MLa>HQV%C5_p=K&E6$8X7d3*t3^$I<Y)KbKH`2xia-EZ^KfPR3E+N
zS9ojV_hmM<?x)@SU9n{Jxpa9SNLSE_boF_fu5@@igI-RtlFThu&D2wP>w>orGtr6K
zWRjnd$(~gD!SJ>Ry!C^(DdG6z3*o)WpRF=M*_ysCOTL5oyOzvJmdVoW;xL(6ktO&i
zdf<Ud!5dW#-fqjwk_+ArfZrP~kU`7i|JW{*yz)$Sgtuo(W$F9j<V>F<_col|_d4tk
z9+a(0=+aH$?f1OiC*xb~#-C4}#<w~RF7UZU`2IFDuD4_Gg?%7Herc8t+#{QMKOU^7
z)W)spQD&eiH>buuLY~nfYQb;v*xQQ698BIqE_wJ3@jd@nCvtrU!#cvUW=r$b_H11Z
zhMR$SfIq;~uBpsY#KF%n>R))f@;|P^&SH1)d@=|5iubELpr?xWb2B~H{BP=#pF{7I
z&znxh{MWqS9o{bD^P=HxHF(>kOkIWIbt(sM&p#)_9Pd)TN%)wkqkryLPro%G%RYl$
zBXRh(zu-qLo2#w9%q)lHk{6RJw|4l)_&1z`x5MCV9K5Y|gV|?#-L1lNRq_J9GkE)X
zGjr4Mwimoz25(!y+lTdQt7?68erC;9wTG$j)c8lbn&hQxbwIj$^-fo})y&+i#P>2f
zT~8XO>)(~ktwdz#1H2u^*I1Zmnd+IY$;H#<RxDk$ebRLU-X_D_9`JT$G+Dw2s8P`@
zJ4EAgshO$+XqTVSIIqW%H~t|>_L7N;<({X3F0!oA{=&&=draNO?9H||3F?(EL6>L5
ztL<O$>c_b<@do$flIVtbl1}0u&Y{MsG|$O?A*c487SEUTQ(>v-F`Fv>W>c5nZ92Bo
zreb&DEWEw)+@=*P<CyV})7S<0pvH6W?jOf<5T`lZ$BV(+-tg8@&#9^hrTS=6g|E;P
z+UigMK8&st9QtaogP9oW=3@?(eM;RNiWf3RB92IXX&6@<B$Zqz^|~oddM7nPi=Nh!
z+LoSjOL*I2h*S5Q!*e*iSxy<L%__W=J{3T>E>8X1&q)@0oK{fh{YZvJPjVt>EleVZ
zF-dbjC#uqdB(3$Q_j@8qU*PeuhxoeAb6+_}E?zKe3(wosbT#*r`eebQnQaP8(IHs7
zKP5#&Z7B*bm7=2GxO8HsRItBvn{2)F2OYXYK0!=Tn~uyT_ot6TR_;-cy=?mW7WqIU
zoXm^HYs@Hga%zQ{)bu}n!UBCm^6O+*-A%D76C0~Qc$))nd&1j`)m<9Xk<5V%7kQr4
zx+Po+y3M{+d}i;cg)O}jl+50>@5l-|`GK9O@YcO&q6%ds>L<=orO=1tQxlZO*(&-5
z{cxB+o_k-=3Jbd(m>rp5(cR%Tt?kG2&wR*9ex4+FyQ@OHDzn2TmaN|Mwc_*vZNF1X
zdh}cP|K$wvw;@61mT-Jwg8uA=p3L{J0&k1J+w$ln+u&_JyjkxXkV_VssRkp^(}!j2
zGyUoiOKthW+kfe|Z2c=k<LamD6ke{LXbN{Xrs#*!DO#JFqMbhI_V9KTyuJM(Sw(In
zD+7P}u6xNUeJDk*KX~ZPKOXwa-BUO6JhgMHr{=Bm(z#0Bnw;sa3v0dE4dbl_m%UXo
z$wybF`e^?FALYLAVgA5Z)^uMLFYK#2;lA>V^VM#6yD864!w31Xm)&1|s|2vSBtUr+
z0(5S1fSzs$&}w+w9o~9A^jCd&dmi5Ymf+97(O<9a%>J$>R|Do=gweCfbS^VKP>t>e
zst3H?%skQT&o$)UyryPttf~2LYU-LLMAf^5$k{DKsqpp&y!{H^wt=@3;O&X`wPZ~T
z)ugqdsu>U_KX|+U|9CqZ-VW*$uD;{hTl73aRXZ5ex{pB}el*C}EmCdw8dbKYNr8Rv
zRl(a#YZUop(Mq@+t>7Efygx^){%_G*u{&CRp=iykV>J1uRc3f=hPN^AV|AC;_Jy{>
z&*;#TgLYlMXjgl9JMf8JM?c$@1#e^E?I;Jj6*Y0u_fA!Uw+rYqHNF_f-g){(cjNTw
zQJmhr;CG30{p@Fn8p!Lko1MHlEB*dZdjB!Z{pBX<@AYWUmtZiwEoPu+;l-R0o{|d8
z;zxxiGgq6ezwr|!6-v<(Z_Z+5PF{-6R9ATTehzi-xD0*&0j=;d9uDry*Y40y!bh<N
zpUmE5{LvX1I+90k;@1rJb7ztHKyK}UO#QJjQ+?4>9+u^8D89ro<aBJ{4EPHF3$=8D
z3mxj)Oc_RI%D5;~-SNAo`jb=OO9n2!uk!SwN>t9$Zr%@GLw(&8otxMCxA6Azscg9~
zqrNUjJ+_v*8Q#7aouxQDSMk)g<rZh^&wZJi2*;cAnEmiRPH%wsu~8keMeFF$={joo
z6b-set^%n4Zob6B36Gn?+Ya!y5xo8LXXX|+vcm;tuJB`zI5q11g7w%#Nd9k0^7AUt
zUxlG=u(Q!k=7D>XV_Uth4pSc=xke@szTvA}jrYOA-OOKYf%SNszB_{+d=s9(phkta
z_2BJ^!tBB!b8gvv<{vlJ)3tHv)$sNSye-&_y)p3in3EptCvqW}VeD9#oepGhRZK-E
zCZFz3w_Fv)>$G81t_sY`)xlo5+DJXzwmh?qSL$e16MSb=&<iZt+FJ=-EFW`N8?rQi
zV3yvwGSt6g8m~=wNUv^b0Gcj(VUg?<6&{nSLU>@V)=tydCh(J9Z2RSDT5~W>x3)3o
z<4B_yMn4q~R7+SJR5V>R(G?5V$M<uQJcp8Ls%uQsX1q=R_Dof0d@&|?yAR&3?Uo`Z
z8s(3h*?9?Xy{VIGG<B(RBbWX%x|D9?K5L}^#_Y}2=ka<uFkW_e+r2e2>rd#pa8G=>
zB0>4%;&kn*QxAta6@c9K{GCH{YuokqE4x~(wdv79n=UP|slZ&DuHUh#E4&>JZ!_}S
z<?^(vc8*=+FLTyC?NFa`Qn_?#Q>H|mvTo4#En#2>=y`{{F4>i{k#q48yY3nty4D~5
zq9+%Cw|W0Mbc?I%V~75D>(DXc9@63M(Y2BR-sZyFfWl4%qigl9;M9}=CwiAtGvMt_
zn4S{jRFBWn$`&yS|JABl%dI*U7^|<!lM&z9Cbv#x<B^ZIhU*uu>j&)0|G=)_xL2I|
z+M(H-9U8KT`1dw8eGaqH--y+mlCkPYmY&yEtB%dK%DKm`?tgRN;pZ3%)64TS`F*gF
z4IHI==}~&YjAZGhX8lmws=Cc%6)+@Le~pb*w%w|X^=2)7Y}U+O>?>bw)rje_D!h^>
zXML>7?~7H@!?CJ#JXQyCW3_O)MR|wtnO2Tg;_PT`Xu^(}iFSqI8Ev@9t{b=Q3g>5;
zm|$1WI(Aj=YuBK0Xb+S4*gtmFs^d_-Vh%NXZr3+|+SzMw*Kxl7OMknrH?phJESoMS
z#wg}tw6>ME>S1k*o^La1;(D8^?&qAt^LUl#_Y*aZ|9$B+*E`NJ$@t|uao!1TM^3^9
zr{<n?DyvkS4tU4QS(TjLG?#)$yHsKXo_4h7S6s`vl6tzdw}VTEJG<Dag3f&}5q49*
zFta(ZdlIjc6n)P*v^F~ZY;>C@)Ues)`M%$P9?e|I%*DyFu(#qSSK9St+2HLAKIh{<
z$(sAdLlF->^tr01KET_OD?OE76aN$*D(@=Z+C9UY>~eg|hrQWj=%em!eDu|RAANY`
zqjS+@FQ@tH!e<|SyTV5!EBJ6mgVT+@w04`PE*$gFH+wv^xiX$B9_tSn?zIX})*t@*
z7cX<*C4cRjLtWd%Up9DqGs0ibef{<6qrbj!1*mM_0PVOQpv=HP#cZSZd#0MAqiV3X
zxCRGWyvxk*Ht7*0^KU_V{xV40rq@)4E0}W_>}?jJ9q7sj=hxEuceT{T6sk(AL$#(*
z7`wK@RQ5!e3{#mGx)v@+Py~AkBgmYKP;fJY%=njw&NXPmQG*(V8dbTiQMPGDwb^2l
zA%^UXt5NDz2u-<Nv>N>ot;q|cb!tbnmPN+s<!Gzio>_Gr-gbRsRil;`9XMjsL|?nA
zrrUL&*V5hdb~*l!w@=~i$kGmdpw?Y`&7m!vmER7gKk-`}xd?IUK+h?JOvy!X_vt;(
z;^<fZ?uysw#R)p^aH;$6E_N^AX@|F);O({?vgkG>(Pv0f#Y!+1FXUC;o-CiNTX45~
zEwl|QS<ji=_bw-^zE_Gm;t{NbW>q*SRin8NUw@ybIlP`$P|MuenXX;%cIO-F9V0qp
zQijT=Gp9h$;@8I9m*?ZBZp#c@LaH{?i<_35rffW_ovayh;<LHL9K$>Mf%~YX7cmpp
zCo5B7%rY&fjy4z0(ji#8&<9VIR~CIe>fK6Na-cK!td^xaTk$TtXQQ7m@9+;k%+u^p
z<70uGGkcHCQZzO6`098w|IJhd{0Wn}X_kbyNB+VW_>wxBT(}zO)!}_=t5I#31&1b=
z&eb{kn^oZLL$2b~gdya><nGAT%sF|?)}YbCu&Rsk^upC$czgp3z`if!De<-}HSCb9
zb#-WOT{#NYRZ$q)oqV9Su)93jA{D7S@6ty;F+NXA$oY7(lUf?ydc)heCu9j-Vvgrg
zp2qNaz53uiW~SKtTXwBChvDeRwZhqrx`gupJ?lMoa$o7ePC8ddT{$morfwa`$L^EW
zwe-7Oy@9beNAl;f%ue;mRsGuZZExf#jM?dN8<;a3%D!|PeS(V2D12t-aafjO@OBg*
z#{70zrfTkH-eNLdQFvPu-j>3r;?^idbNiw_Ud6K#N+0$|=3AK&ZM+1J@B+^BHPUq9
z3HiV~c-@d!xUhHf|M&L!a)ukq97dH;cH_K7H~xWMEj_?P=*M}LIq%Wi@VS?y8atBI
zc<ul6m;wvnt-DKWintW8in<03)6^zj7G`X&G{p1UJ6>(F;}z14IanB5t56)XR`?Dd
zNxz>EegpVM9U6EOT`0`1dxLCf(l(kmHvQmd)A7G!bqL<Jgtw2M*<^#af5O|{t+=1Z
zIFx<Tp^0S~ML~B8!Y+yBL$|%)<u<#<z=t%>#(!+E!xB5bFo(AFbLb7c{q?9rk+Aj*
z*Mxt_frYo18~l&A@oR{hg11)K=3mIEQM_GS)Tsh6H=_}KMtEy&V$tMi?iqhatAiy*
zwczaoc$-&^{P=P<efZ9%PaSP?hqJS}x^q>7w{hrVPI$C^uT3>(#>&_wmKqlpl2cpg
zLM)j#v3fvmULg0N)`P5S+0v>FPOD}=v1rs9i{jQqtKN_3bAzIEBRNWA?wB=xiCH0B
z!&aKL_MTaxd~EYYvySrTIgiYGQZP!l+@kc!7^UfLqtqcdTF$?s)p$sZHv3rAy^lp_
zr&#o9Ym~ZIidM1t_(_j2$8j!3#~#M$MHvfwWi09)Xi=4L3%}pw`TZHA!{wqC`-GWH
zSCstoqU8Ne6uV@jlrb=hCoD=8V7GQfEAALMf(2qU3*Ppku4zFZFMM;9TEN?nqs?+}
z6{9L4k`3OzhPRbr?K-Y2yg&AmG~*wsixDjWzgzlx>X@=|S`BXteTe5dPf*)YE|o-g
z_U%VTFRUHIb+$L$ZS2yGNSEAl6I7AcWzTbQszd+b2HL?*JaTh)$7?ck-7X$$Y1w$C
z6z6k}qq{I?c5NrM5xTbl&rW`Pb9?D8++IjOfwMvLmmWHP)`LE?hh|*zAOp}-r~JGW
zv&@T~Al@20)?4l8dMjbFH{K-rxqPf^32#+<>!q-rUTQtbOP7XtsdpDI4RrU?$WmUq
zao1BLmV2uFQ4b|;^^hAq;`h@%B_A&p#gA2FwzvB4@mA3mJ{r}=SEs7@>GU=~t?-~8
z=6dqkU%w;<=n=eK3vVse0@+s`sL&bJwC`#)g}$n$0iUX=acGcY;cfE;L9#pu(%U&T
z)gmrfEw==#RekE*9`ttKhN!@uTG|0`U#$$)qc5SFIyp>z+`?70Ww<_E4Oa%do%4MJ
zb2}0Gqp3j)dKt)eG^o@@1M|*Coo#4Ty;(-R++@=H)n+y4^-&q-#$`k+v_-VG6h~Kf
zL~F6hs;Ruzw@u{?26MwpS+tNd_M4|R4T87z&33hIZ<GBc8cc^+eQ#s`U}&sLKelSc
z1&h*Nl5sUB7WSik!P{9|<CMq$Uj_ajBhJL>HoSHJm)<VSJyIlISIfoAb8Wmjl7-`(
z<I+|i&S8@i$xJ5i75%9)9<24oBzjSJ`<|dT!`qecc7Q!uoimbU{w7(G{pj7`Nob#z
zqWpMSir|Aj{fhnt{ki4v)?-(?hWw2W1}%ESW3<Op=#ZV#Rflui(XPxHc1Y5nEqEL^
z{21`|Gjsf9&!p*HII|7wm>0pvy1NY?>5I&e_~S#sOYo@^I`(gwTKp9={P6Y{cv}wM
z{snKl!P_+S?9^ge8fwjw7d7w1GW46jWXMLH>~}XqYmd>7KA52$yd8BUgBj}#l@2D4
zXdQj%z3g_mh>!MGCORU0WAx^U+p?(Pv-M(swr=6)I)&$YMCsaU?ME&;ysZIm>!Qnk
zoK0_ZKKbYq*b7*ZJzS@AG>^HM!uX74{gR_aKQXI(0UlF()+VR?TbTLl^E_GUr;bS>
zqdc8IlSMh@2|FZi<!bKV<Xb<?RWw=F3vT4G$0AR+mQnZaXCDmx;$E<J<r*@;$n`ii
z4SoAP8FgOlG=5P>HMlai*U=exd-N3Ngn4xow1eC^{v1Ls!I@6Ex(ja~j3B3d3|T#W
zbJd1if<U}a{qU5YtU+%78)jv1XDSwNaa*33FL+%dR%Pn?V6GT+czFBo#8md};OT(3
z^{gqfqJ>qa&-#Knim4yaLHe^#qkNiT-=(VlS+q9vwl$aN{f&jC{J)>)b(9lDkE%|x
zD&g(>sX!9qYN8$$;I)XRI3G>1FkVsX17=o_C1@<?#IE@2moh)%>z|;GcbHw>$b9Uy
zc&(z3@{0RerHD8U3FE%jncdcz@haGm*Lyr(8G57dYdcl;y%cwvF{o)0Pd&Mu4%y(X
zZ#$c!j#yRchE-A2V<*r<H^8M#ey8$Hi`AGGRu!Wr8)34k!XBG`%(KZ4-FXf@-tz_R
zYB!yF@OQhKZlzY<ZC9b~c5Pf|SI}O&dR^f6_d0t`$~)u?ap*eytz>m*+!}`xsc{qF
z?Jt9+1zpi-kJ>b4Hg&31a=u36zA7y|?a->e7QG8KYiqeEy_p=PLzklU#tch##;AKL
z830Et`gqzRXHhFV0j%WYQm?{VBiA$!8?&}nW%sgZ`L`CO!)#l!MT=X;=n(a1z4Ng$
z!P~BI``lnF_av*@6|!peM~h0p+v97=^?|pe;B5}PJ(&=t1!v9r^aD)&oxHvu&58{+
zYf~jNz9K%}jQbqCo&Hsn2A7J`1l~W+=dJn8tgh?e+#$0{zc6b^CXY2eN(KIiQZ~H(
z`R5q?o-sNOZ++lx=NB>BTf!ns1&fAMCNF4TjA9E$Yrz+|3UBA6M5%pnly<`1dkv%1
zylWKaUD%#KT9NQJo4RISK6-lljJh_%s4cf5)onwh;x`)Q_Q)g;_*{Y+k-m6lDn7C*
z;DA*>&#-FVG%Gu5t(tHR=F|VP?;~$Ne~g}uv#KGCyiOK==V$0A^<A1f%%!S5@K@ze
z(Cszx>WcPk<n`DY|MTyy$@A<PuW9gh<{)Op8o<)OoqD>N|I?Lrm6{Q&6|G}c15N!j
z=cV{^_;xr0EuL)E-jY`3a8Lh%d;8%{oI93ADyN}`TL0pqLMuG<?wp6p`+I8A9#0jw
zduh`PFLmibZ91LWcAU5Fq%n&NZ)X<qR`1PTs`a~<CinBwix%{IscC2b$J^3gO1bB$
ziSTy&ArGzJ;-Szsp8ENwr&h=Dm}9+kcorV4L*6oW_F?ACS7l0*+uVc9jC^D_kMU!c
z%SUxH(XN}2c{RaLYmEN#{nlSe1N?P;qQ5#(6EC0V&t7|f4Zer|e9~Xj(Vz=W^4F++
z{_L^~)V0FFTK99X#wLVl`0ZLsF@)-Qhfw9g+l<j+O7acY-sa&db~apVoy-TpTPM7=
zy@*iX#s*y=yE%BBL1kAMG;fPRYj+#;=Bz<0ZX2|vdZd0xi&Xb%k?MUuQnx=u=_9;7
zQ8${pD_XfOa-EqUDhF@-RJF*<-=ZEF79HYz*FG;sJtuNTo67Em)zNCg+|kv1F{*mY
zqD200#t*ep^V7?nX;&Szti3n+Kc)T+k8tR9rb9<3;gRI`yHh)kFPxR#@WEdXa%u4|
zF3r0`Pa->!OkVDF|KM+?x4WqpIgzF$8JUe6j`zf{HAyS~O49YR$tpn2d!jY>%Y)=>
zm{XWDqYpukZ74aDdHd)~;D=ng18p#inY{byngnlauOj#A$3#7}B<f(NM0LkIx`@Z9
zaXVSlm|G}aB8~nb{u*<-{C`H1=IgTQhwP)lww~;lzvvy8>zJvTGcq-t`m+wab;8?b
z<aTU_x9;#ZAH01Ji!Z>;F7$d0rQkHYZTdJvxy%htg0n9$$fWLNkHeqrTRY24D7<aD
zI#YY#N}~&0H!}6;0d?Ge{atvQAKs4Miy!wb^>7P>!#Zfu^r(xsM90SOUu{*ce*2gF
z^7&}b9(DBl!5sayFGpwLZR_81lml;*nMJ*hPwVG_<X845dwvj^dgR76T*tMRtez!g
z=~T*P{|mXPH*?kKcCP&J3{TleUQn4lJw~^#gP-d)dUHc&mg--m9`3^oQ#Z0J8=yI(
zcl)*@M}0sYO<7t;(eQTc`a0T#ANw*NON6<tuj4mH6Zhfmaq<X0lgTsu6?t=Kb9D08
z9QBJ~R%>;ZzB6ZOCA@9SbLf7QEDD~NjqtV>&rx}1w@aZlgu&a|<C(32w`Ir<NQAeU
z@U{}X{T<%+hPTt;ZT~W93Vcn5&(%~d$4@dC-lo9YHqDveNKaN5pJet9ByrYE(q<R>
z%0qgsE@pHFQ6s?HFHzjn;B6OtPWR}YMfOfmXL_;uoOo64axa6o6}kWIgSSPvpC!WE
z=3a5S>VuDX8`^6#bU~+6ll+`yvQWDmXWobVUE47ti<ou6h53EhXq8`ks|Iv{wO6eg
zGMrj5CRVllVUC-XIa#zOc)Jm;wbmb2#lfgJvf?_xG6#Ln75VI{0&j2r3UdeARc4r7
zma){r3wgW4u2yI4n)=bM<jVXm!duU9hn86!>bVj>73b+1Nj8ljm#*eqi=K>NF5lOp
zIctnM-PI)b_9oq0ZIa`QNv+}Sv0`Q=#+aF7He;rwzT89o*pU6-@V3_}zV4|-D+gPY
zIX6ZVV$I6$XVyk|+ZW!x?_p6+D?SI_M!?%+=+VomkN0%4sui{E-X&K3J<zHu@b+Rs
ztLDPnp^q)(aPc$ov#s=x)?Rpf$r+_lTg@`JFzdumWZ*V3Yr#2_c0V!kT$uGuMYBe{
zqJOsC#6B4l=OmMA9yDnQOzyYVtPh9D(+!ML?IiZfWYc5g4Dqg^nJiheTJbagnrK!d
zn0=zBSvUKb$rd#0A78VU^L*rgY1G)mM#Vfbl2>TPgBYc5HKTZ5qjanzJ8FiYA1*Sh
z?0hc#sw;aK>C+ok`$?qU9*<P{1(EEo{O{{c%n?SZ=3r`en17tF`>}yp6_YvBSj`&i
zW>$0w&JLbtyw+xAXPR|rj8%`RU0PL-Q`-|xmDobhe4<k;@1b%0A>B99Pp3z*GM^NZ
z>(Kj`Hl;SSDSwPjyC1}=!o*kwyJ9uV9IMCSWDM4bRj7NcqKcudZ?dXW534pul7U2@
zx#t~=ww|!);ZFXamRodcex&YXqWS*pA<JS96+7smwC5f=yxo)jx|e=_;-wSa(UYfn
zYf?LJ)sOR5J9yh0-sZsDo{PQoX@RGP_<3kh33vVD=dLq3?kc&$T`N2CxIAXFO&*%q
z)l*mAc#^fu;|-$@UhgH(YhD^$%v+u*-fFemTa}}HnD@k&MW*D`93SQgd~^xk7M$d#
z-WK}1ZT$5D-nzroR`9kxyfrN4{l)A`*oFSQ)?cdz`zwWy^~4)}sX~CZ#RkY;k2-lg
zS<rU_)Mr2-d$g*-r<$5kGgz-1*V6DA;R=MePbNfY{cf&T5!&W7Fh9*T+n^?Y7}SqC
zw-dbm7T&JD%b)!s^#$H~Pmk2|bCDYQK1zOymJ8l47{Qs9zHY}Y(Ymw8EL&%@_WSUf
z;2al-&$-%j&Wp+H`kfLj2fUrao}=6KVpMrdjH+|at$)m>!St8r&O$qaw>@uht}f-!
z4tTq<jzee1I}`?QXLduLSnX7^$MN!F$H}F6F15bx(tyYG!Y;cwM>6kLK8bxiNy>t^
zm6%04P$x-mm>H^a97e+1i+CXesYR0T44$Djz7~+kUUGbrw-fc(W8M}_(!Y4Gb{0-n
z*F#BqknPfwc6cAIxl|V3e)~-#^*!9(jF10s=E%_|Z@x^`<iE)1fVcB!q-z4a4QiC3
zy7TbQ!`oNyGuSK1o+Ikqfa>I4{0)Dp!6tKsJ)-ZMFH6he#64!)y03woBluY1OnUGc
z>ivMPg}1GbX6Wv&43%7(p>n@vXdC_JcF!_&#z+Q8HdjKXd{e1)cV_A>k5!GVymrhZ
zxn0TBp#tdD_@@iPTOW8koH>}Q&B*6!MNZaZ`n!|KJvWg#ejrElcju_)${Y=wpQBRq
zVCt+K9h*e%-GCh3>BnxazB%duZ{G#tl_rPA%bWbVMDhuH)z<2-ax}Jh4q(X9*Ko41
zC%QAf=ty`w1m2!K&fF5bEdy_d!rLF=tv|dqwI%nQ+IkYTby+xCvP2!7&Zwhr;O%#F
z*vom1oIQAZi?^rY?E`o_lh0dsp8XMcqI~)OiR2<~dP#qqv&OPrnL4{5Q<+og&vSNJ
z-JdzY?wP6xZ-d~iH@vl1W8S22svOZN3M!ML?iZOQTY+~3UAP-NBp;E7@Bq!@1iYOH
zZ<F9{lWFv<;O$e9F@P^%#fL=pg(YfrQlg^q!1Q6};_4C?*}pCwy~BRYOI!!(&%)al
z@b*m!t|jrhRe%~69VcXRoX(OHky95xQwSQ4C)(XXr(D0{IqK|GcX+$P$ElrY$=BfR
zqfK~-M@U3Ik<|n9ez59xTdM}5q0U9?Ts0s@e(Pd%luVkC{ba<stm+JNe}cD_;nRcf
zse|EdCm8%H*Q$bz(O)0pZH%_7GrTQ4#IDfM^cLsa^=1|CPqgd(*LKx(*;RsCx_vM{
zjOuoFJJ{q(v8i3CRo`-d^@C}%-7Wf+`YQ?EF4%0;#sKCV;O*C+O!~csSv%qFglMyR
zz}sktS=Zj1)B-L{K5Np={qSxVoLgto+chS7Dkk+mYEqxeCQUDFlZkWus=8$DG>g$C
zYTE_y_HE-Bl`V}=(`-@Kh1ACVsPoKL`Jlnx?;5Rpt)q3idbExgMW3Y}ZCIJSfpC)w
zM46beHLDo)qdpq(*OK?s!=!<q$)bn1|GqRS7v8Rfw|^`$=`y+Wk+CM#uS7n5wn-<Z
znN;p~lV0yOY1m1VKJxFY|IDOb7flMl>tdO2(#RPmRU2<2f77JzyPDLnnMn&=MpgX<
zPb%MgI=r7)Axhr{Md?DBD7~kr{P3Dt$?&%FYNL)W;F@V9H!V^l?io~UyFpREMry-~
zNR1COYG7@nF19i0*xx2~sm|l2n^gkdX2blU@OE@jv!=YGJ}zt4;{bZn@Yb{^j%;+!
zbPt^5;5zj@1+OzbfZrF=Q<%?lSJtXjPpf)V;S9%F?L6$PA49%Eb3R7hdRVdQ4{v9}
zTPHQ`*vD2y?BX@WYjA99tFDDvWvgUWcWS6<M=kmTZK3aP7WRxs!ha9_Hr+%2{_eqj
zz(cM(9(r`Ds>Zq86`AUx*J#1ZVd`*)x4JmJ)$J>9-FxY!TQgy;(M$KnR@L4XRaIeR
zRgL#?mw&Fi*8l9TPIuk)xt@m#uk+9?9>3BJ53M;0$NPFIa<vybAiU(xW7YEW)~?mw
z+DyIM<g>Sm!P}ETK4d%js7e*GB|Vrk`r23Xx{+_S%vV#$ZC+p1PkU1Rn2Yq&lM{Zb
z`p}Pkl78CZ<*y>}_QqL%O?l(5yy~zv9naO|0R0Yc8^GI$Le=G$UQ>;R1}phfm|AZS
zBgZpB*5MKQdmr`f>j>pbHZVJE(3x2Vwfn`O+bikk?lj2uA8!{Kw3q99YTh4aM5@G@
zNG%UGYJG1bnubxAzwo+sH))U2gg?MUwt-m|-zXKMU$U=8v|2_*>l^kZcvIiztfUsE
zhjSX<ZW<k<5O`Z0-lk`9j&5RCxp12vytiu9E{le5Bj0KoJ(2HXG>RU`qyn+b&#~v|
zA1nL5@m0jz<h_j@!8L9C-{G%#?NI(_>`MqGU#D@LwtvlQtW})UE>2NP@#}4g|NkCn
zx0O=O?>W~cvx8wcyN?#YaAuTlesRj1TBBrSy!Q8tmp?w1k<6HHg||%?uoLA5J&&S^
zI-8rQTSu88zE37}46_X6Z>)I(;~OL^a7VKCoKMx^2kaT7)@@olL)D*W;ORu;tdgnD
z1;}85qo>h{Us1b;-^`FNz228-#RZs6**_PK&d$&ca&!~<oT~T8+@Y4evx6COW{UTB
z%g~pmd>mg^L2?0S_e*E~DP8Sart7xSRjO9Het3)r4AvHA9_a%*^UwLwT6^Ki!h3x;
ziyh_g*1;T8f99A*GvjoB0U5Pq*q(;1`aVa~dFxKy`1WVA(0Tu>u{k<Vp5@9U&IN&4
z?3l|^8$Yrl$kc0M%+`%CW|ZOWMm)joDv~Meg2fMcyO>=fJWd96@<I501&#V1kJYR*
zIj!x<fN#m(rLS}3##N>d`<BSIje(<;;O#7U>jQ6(PiF^qH};7&fyc%=YEhvMSwqyd
zGs#Maw=J26Zo|y74l*|lZwJHMMYHH>qaSB9%v9BU%uKFJ)7RgTrHJpT5B{bHV^fuE
zB69&ho!df}whd)gu`ant^!+=%Ojf}qaF?ure+MON3ZD1zJCihHITyJBfy|=C_DRxf
zw3EOWoYCNICjPzIZ<%|9w{go{D&Cd+mJ0Y%{!EZ>u>?IIgI9&Vto!<SU096o)Hhzc
z>5;TK#{8=*PFG^cGv+lrAKuP`w;$lGxr<ZRsBtTyC2yoZI^(cZeLgd_V4^5j7kKMV
z=FG-6R<iY|^<Zrqu63(obgu}#L-_YTz@lRCcG_r@VwX^39WrU#K9lZ!qIU^zulQKC
z#Mi3tsE5CUpOZUM>yE%vF`0AkY`an?;NO9__u}oESp&_fEH!T-JAF!<X7sUX#})dB
zx6wrpm^H|3)`7v1WMLcC4Bk#D1h)>Dlm>?uP~!zGj#P`^BXw_LB)(+!kyeUSey>RV
zJ2Fx}vm$kEVWfsnF_GgBb2qaaji2=zU)vY1U0!BV5S(+fm=(#-Zfg>w9`M#SoxW-x
zW+lR-HLO6i${&hS9v?psZ~frA>#9++@t~}RU4|Ma#lYJ)W@_#TnE%PBN>A~HCd0}V
zCKZFXL-?AB|CnUrdrq2eV&6K<hqs;JZDr;*j>7D+4$cQHOiG{z{p%oK3)ceR?Tbrr
ze5XmD;q4@N+iILip~Fln-OZ#i@OBow?VV=S{5wW9xQ9mE+AJfw_E#6p`p7l-0-E(g
z>Qs2!4c^Y{Yt*)v23=fi&?EYM>z5lin;LYDI(M|es8(@2$90UlGsej8m{BE;qcuM^
z>K*5ZqUk23yym=9+N_qj>{_pDlC8skX9T0-6Zkr=f6x~S9EsAd`_%9Ok@|pU{=FZ4
zWqQ_!(cN#PMX3`td_QxP-rh7TqlsCIubOm&Gte=vY5aR1aweL>|JC31`5uLh+Qs=`
z$QrI6BV`#Dsmw`{3Yl(D{)d&>gIGmBBvsY1l~v`sUR5nOR8>WIyUOUHaQeAtEuOMe
z^kjw*hQ6vIzspthzF1ZLSgoqcs;b<ZRn^N8Rkf?KyGp0IYd}|bC5>~}hUe~T&12jc
z<DojYJT&x#rydOSl0UrN3~%oi@K!<Qj-J%!I_#}-pO`^v=%W-LA1y9Mh9u1OtLDS3
zp^pxJ<D*u6d=xs)N4M|z$TQv-eZW`gIlk02elk7vqdn}egx~%3<*>glKl3MZE`ZD%
z=8z@@XvT*C#U<6yld(1RYpGf~d@qz8W?_oo7RFgSOiQ9dRVp=96IzC9L#t3Fj0{!S
zgD~C3r+gXS=HF&e{5pfC!{0XWc5Z)a-g$idi9xMC8r1ocLE8r!WC(`EUg*#z4RZSw
zp|DpG$}VouolShKYNTpUHECnhXtkL}J{7!;hqq}BxIb_{jYV_rFbglm0*j_Dr1v}4
zqO2z|n!P7R2lmJ4%a1Wy18*a8Vw7x;QQMBZ{feDNb1X8fwkWkOIai%xwfu)zJ=zM3
zPsOsInO-p*F1sOCUGNwFins6{_mkW`HZAkz9G~FO`ioK-czY4vj)b>!xX;DsCvO(s
z7KXPChmhG$mcaGS2^!YQCCkq)^&_99*>&n@uJ`onZjhs6en3qPZ$rrB_^~nda{+pO
zWEpgPm8Pb%$(mP&uENmn3p12K{rMBx@_KlC1eWfEw=wXx`$ZTEZ)dE_(2yU<U4Xa0
zkk7LC1sA+cILDr@of*0eZ)?Nb`ta6&7@jP0JqEr=S3mm8HwQ5r3U6=JOlRLgn&OY9
zsRSOb!~q$aIzChLm|NOl%hJrWEQP|`pa%4gJCg?tZ<~%{&jUL}M!-Ekc)Mo;nW@b6
z#+)O^=cg>Sm_|PtU$mGxz8Am_271gRsE-TuV?Jt0mKwp^(+$XxPa+c&Z*betY^@(e
zKKhOvWy9OO)X8h$?K<B7c3h4McF9o@c)JMRHsgCvXb#`u?ZuIN|Ajdkcqm78Uy!NT
zA1#@@`VC#{X#VXSw5l8hJ}1+4FP}e~eAlzg`+O$5<UN_y@b<}KcAnrTZ@D5}gMVXY
zmU+#p^agGW!Ed}aRaMB9cu2lcb$ENOB(n-@@heTB_8d<BQQrjBMHBx18M`y^THaZ}
z3=g~=3UBMf+m7(I=}$>2JC=+>csnE>-fp0Wf!6WVhE8(VrTr_(6Zk4Y!|AmSzbwsM
zMO)Y9kRNlf2e#RCm>y@F26$G;CB6%9Hy^|YnBi38ifDTWq)QFyhvN0UITGel-_~#K
zATyCV^-7GEw6rMufJNKBSoDnABe@Rq0Mn^czk#Q5R<-RPqXvEP^UNcU6+=}n=hM<2
z7Tuvf+0q}zE;8vfe7Ly{27GPOIO@qRgQ8W`-KtP{o0DLb|7e@ap(z)Kw<CDF70>YV
z<@}zdTa;TJewU~Jb2Ub{_&y)`{f%WFu*I8bO?VKci7;&DE`w62;o{-#+kz$~!`pFZ
zsrRTG-z|#Nj0ur4HHp-#Vv#y?!yr{LDEW1`(kh3`7eD7g?oD<>nB3abQma?B=u?KO
z@uLVGfwx=GR!_r;38x~}Z(XDYc^LJhd$g|fj+QyxsKOz9-;zeXdLFJCiQ)P@FI@H?
z4Dy^}R4&XtGr_2Tix}0Jt8kD}?Wo1q2XS9YqK-9ChvOZ6(IG<9nn&p5-VoJW$DHLn
zW<b(IlyoFm?<NLo)a9D^nM2fjYq+}X3)l5m;kv&T4-EIa(?t#HQI4!!Z-dGt7&QAa
zJ-BX>`t5q8Ry~Sjp3}%~B7Xi#1~t1Dp=IzkuLt^i7qbrcM#mm(R)?R=%4%oSfU=P)
zHO(N`OnQ3MKS}U*$#{cmjK*WL-k`f@47yb=QepJ?dU0Oa)W@jhf5Op!IA0VnX*Fk-
z#~lrvRhj?n%wC)CxTkjrQ@$2qTFP@&A<C#>10(g*?nvEuABm10sja{MuipB;uR$~G
z8g!|-LDn7y#q~1Ck1KaixRw=w9sXh3z93B5eInFoY6Q8y5xO%cLYX@wH10}-t~?`0
ziRa;xk=m?{LF3!<_D*FrEm?&b{;DdyqAET2sybZTO^@zZ(y%L)H6h7O_lCJ?%P}{0
zWK_}kc2(qlqKXz5sj4&Zb}GEx)P(w$>+8I#dbhT!ZtttA4)xr1;EB7c!Q1_fJhbGz
zhZ>&sRLc*Z%3t0~E77YjHTTjoPjB^txAo!eg^%8925;pKYxAQk!`modv}Ko%4ma>o
zP5QcL+xgIo@KO7A<Xg@0Rl}XWYW9RV;F*48A+duO-gbt!v;U>9d&*yxf&;X&Wq?k=
z+ppoR=YSeYJ6A({mj<cy-!(P#Ot3asgJteeQ;V+!u`4!6zugMbjFQ1ReJ4ae_YPHz
zIb5S*YvDBp?c;4YymfWvto0>AGcQDF<&p^1+8-g;MP`cJcuja3R14lZ-bToBDMABz
zdv-qB^c{mbJ~C)5XS40Gk=k1)Qj0hz_NhRBp;eR)dPnOi{gwGO(VCe#8vRYQYT>Eq
za1zdLk6|tvEs4RvFO_2GSH>ukdiOc^pFJsPW)m!&+sSf%&7PyDR=xdT)w|4CWzoyu
zF_vD|x>#A^?W`@a?1;5#`U#un!`r4Y4h3^BYcZXf!a`!tDLzZ)`A%GsN=<R9iYZ>X
z@b>F%c<`8uT;I?oO>ybRoh}VUR}TKmr9I5G7m7qz{wYxjH<`I#nxroHXRf>;PcEKp
zzoztZFOgsGjt<;2L;sG*P$_ECte;_R+YHTbouMf$$rNv%p^J^^#pc50`Wdo}%24B3
z89MrjjI5U#>^$S^=-qyxzFh{lH+P2HO{jCp91KLae)LbeUc%c=@OHZGfAge;E0KFn
z7SHz4%&>5+Di41OWhkXuhMooUJql(hj~V4*U9z+k|8BQgS^5*Mw)#FxKeWZqt1Mmg
z&(f!o^m)tRn=Vb}GrsDjTn+s2ds*l;QzsYiPd-zBb{>6X=TUt;S@5>Cm0V1C`;OV=
z+CBLD*J<n!Bp+xmnGl7Tv0~<`3QU~PhP<s)<gDVUe(Rp4uABu}*sdo(QU~uQLklkV
z8cK$3z1k|n{C&Uj+2l#FJK$25Zf~bfU!JA4=mDNF8Ct}@=?;%MrUCu#3Yl8+7y0(l
z%yQLDAwQE@{2FBBElt*wJIslFPxdgm6MxcMeK;pUwcxEAyv^##p4V{+TJ$q{#N+5O
zj7`)uYS$@q5|#BATrHENC4VO>_+%n8<IJFSPt<|-iR#dp|9|S+q<hR)tafP<yj}H9
z`W4=eT`f(8x22fNX#j5rZMTsZYg52{8#^pv*<-uRY2?E^fV1${+QlxvB6eB%ywA+{
zOsgNO#w{!=gGQXv)uIBIEh_oJqPeB%l?7WhCf2GJ$@FN2-q$(?-$9gm+M=}a3fdxl
z%7N5t9jLW>z}vUznxEU4^xZcm^bu+^YLZpdJ)e7<$RUTr)N+~Bd1dfNx~OX->)BN-
zmEO)ju`*=BZg@NB6Fv-n7vIL?)xfXu8Q#_)6K4UutpINuZZMD+Y}BPqMt%2&d-_jC
z9madP=2N8phPPX&V?TY2km0*f?Sflt@pk^!Hbf^1hNwl65WQ!{VtzDpjYVo{&7ZZ{
zCs0e*3WO@QT&RAy7^X2}!!_zG_ZWEFd@Vkb{CGO!!qMi!_4~ST4ZRz#zaECGIJ~V(
z&+In$qf-sI-*kx7DY(?&7_~IK?G5|NeG{p&jUx3dIZ_pP+;0`GHY-APFsYUX!&`57
zYk{{u@HXUdFm-vbg16Sx_FXk)ycn#H|J^J9^Qg8CQRdg7dVeQWg^P#jbb1(jKf+XQ
zMwlE&!?ZdwLLD<B^s+wlAZ;VGgL`Kzxw=h%icq0`XpAk;7aK;Xd0K=*qaw(?#7EP@
zz}%NXPI%h_-d31wU>A--+eYzm>X;Gk5t`pDT&c0)Dm1VbxskP`4xCRWGYh(!*_3Kw
zii`+TF=v<-!rL0~_5i$1X&OeZV3>Az8&wzHRvygbrZSU~VNlb027MTSrU`Fv)#jXX
zGeogJhbr^eP(9m?{|66Hp?qPw_##vtZ-uLSj|h3-|5-I7LKom|{v8oy%tYw?(+Kq|
zYfwXY+X3FTfwz8zE6HA=k{Z0JsH=}E>WP0P<(pcGS+7dEf2lH>fty+naMRL5Zo2Z+
zjU5P8^kj^iURHEd+YmR-Pj2co+D%Q%RgneWu9)Jc*SjjKy`{1$f2yc4S(Wrhx`$f8
z+cjrAHUFunZk6@Io9m^6&AjyEc=YAnUhK;CR^DrGdY3*L$lOu$f4sHi0Qr(b$dsJw
ztrDxe^;=1H^qPIN3SYBjw2%C!!{0qVdh6w@%Cr15XP}?Tx%n!5fv;+|Mr+Qbw@Zy%
z6y8=Z5U8TFtErv6x;j>^q49@ms8jJEwJ*pngwjFGYX+%qiy#dh5~N1c*^O`}NKZcm
zsYMiXz7Ill&69Z|p3Cv@_EQ&wHXo1BQC>4%Eh9AQ+i;z_6sl(VnK#<Pd6eIaE5q1b
z#QAjMw_)rz3e)m3;X3%0LFt@hH(1#j7!s*E*^z1tZ=L1M8W3dGl1Q^|*EOpYK7-NU
z;1#SFrCFz=WXHEx2d(P6XL$36MC)O>7zM-IFOy<4b4!fU3t3d}1Af4g7B$Vbs9P6{
zZZEYcsDxEhIP<?N3TwZPmD?hE{8R8lcVTW~uuWH|k@Jjx?Qxm+AJ`Ps%dW0nr9Tfz
zf5Y2?@b(8oye1Bh*ZIlJXndQX!|4fHFd#uE@R@FBZgLlOZiS~V4Gv4xTYOaQxgXzU
zR^Mk(lH3-Pef0@H&NpPj!rL8H$OP<~p`6|sdhkt#jM40dxt*>*w!#CL*J?5QO7UaW
zJI<b&`edcSTL(IEBd+(Z|Ls4WyM|m|t_Ew#_I=IQ{>PR*>CBAd%i2v1+dNI)_u08Z
zUHd}-c?|IO*Tc-Tz}j0Q({!j$n$nqL3T3YJ2syGn>%nEdca8R0dNe#s_2KRGx$J-#
zpQX%}^m*;%%)r}L%nH|k$3AnivaBC7H8_73dtRt(;q88O<w!W$IS<|TW;(r6GM8dF
zUtA|=kDsaV(NwkfNLA?Xa27pxHoR^4B28_#vESIi3^Tpe&@=3NdY;Kno-7@Nw@rS;
zv%Q&}>M;8g`8NOdtF8B`>@EqQZrwsnPmb;!a^mV!pB4xvf0*4UCyJ-&*OSRwvO1Z)
z%E{<_$r?+awY)J|qXx4l9Tv8zK~58WfT>T&BcR_ubqtwyDK340x7Fe8^kM9q<6Px;
zDPA!H;`JyBKU1}M)$}Bfzf`>5&~Kef=1Rqrar$a>oaT}<U_Im1qn%DY%V9tC1gHMt
zI!<5g$0}q9ucyyBz^*$mYxTD_CEzXE%{-+Y4faqsn>^={Me#fNr?ECw`hcG}J61u(
zV&#32bK*6dUc+W1Jg(f1d6(}vPr9O2`)rK*p0p?i-mWWU)t~Ip><w=lCE_)7@aIi2
zO8tK<on?5GXV<my;v^Y~+hj76i6;}c6(PYXPH}fF#hv0VrC8CT1wxCv6fN%VS}4|{
z#lN%f^ZmGwgd~K?Ot{Z$@4eRAu=Y+|0@-cxN?C)Rmf2zsydBw#c__Sn2yf58+fj|;
zR0`gv)rmu|On<O#9Qpfk`am6de-nMen`oD#6Ls-*f{GtX(70!K7EW<LcE+mbDrQO4
zQg1DMAL_!|S#jFsPw&<<UNhIigauB0{i{=dta2))E;+G*h;eRFle&hf;kq!5-s~VN
z%&vH|T^onk$uJC4)xu%Qg10M-aJPP_>_;3bv(uqwI~@9BYPgciMCbxcNgf@p-^Ye)
z&Ws4{N{ZBxeUY-k$r@=<%wi)|?Ny{2l!#Ji82o!S*NI4lH794@H(Y}Xg=^>IFlMV^
z3J4|lwgt0M#j51gDBT+qrRTdMH05Nddd7t44|ce@;=8<@<j~>C4xO0cP}xHc-Q4C-
zo8=DW|K(6LybXr8gN}vj7kIn^gH90}waT$Dt$R*go*$_njz+2_ylnt)@4(x@4$<uV
zh{h`z&F?_8T2do+pq`Fv5UqOfwr^ClMwp}32k%H$Mzp3h<UacY``v=z*Vq{Lm&9lQ
z4BK#=uRTpx?+tPm2J*AD4Og$$_&?ilK7qgcE=Fjhccfk`QqSRRehco$T#4}ZtMEu|
zEFP|6Hys-Fk3)6yLpAx2aAvuYs@<O&9o~+Iw;e8qsld<({jo4Y-;RT|oyjWv$e-bD
zZ+Kgx7e5F24YT3xdVZ$EJ2^+e+cbFFf^*KtV9s7q)MrmPLllO$<xFb$#;EV%Z7#e$
zKhPwPWRw1a!S|0C_3uWbcHc9~-NmHcaVE9uVwT~US&!}THq%{qxL)istLA8v#tt&+
z$95*Ug_yMCqftlfCQWT`R?&Oz+S|`v-JI^|Al;SK1%}UZC*#&f>wY6Y5<gK-bW<&F
zkk?$)pKJ#|x$W}P3V7QS-bSwR(=bneb&K;?mCpY9W;i}zczb7+zbcLOSAS-^#jg0N
zGqc}8v&i0P6{uok1Jx6~)r94NXu*RN^u1M87X)kXeVe*AV((DCT>)@6<AGh%!$LH?
zUWk5yw{Lz3(MWhZ_kD=g`#V&imP3E72$gAbxTemHR=HKtDn|a}N_bny8m;#mBUQ6n
zgbL0Mmw9!#S`~;;@`eZ<=Ug=e-hSbX_v;Ul>dbX_W~6r0!!8SdV<V%rkbX)vN35FE
zbgDhGiqU8*ZjWU)f}XD?GmK`}<F%kZ`x9m+XxV491YHx=z$Hn~I8VRtnWR<Y=&x;N
ze)ATMGrVmCZ|l;B9{z$EuWyQ6!|)|FMMHv)ZtODr>{C)z58if!w<q9jMR@z=a+>yY
zA6)Y#O&{?H?^!6lFPEum+%umsSE~?@*Q|549v;osfRG$?XY?CalDUz>e2#q!3%Pe!
z-@rA6>>B*xQ{n9?@^9KKr{DOIJU=pwdY`MJjwPtQsBtHgb>-?+MOVvJQ8}L~dbps<
z|K(u0!`nZ&s>9n+_o+*tk<mMdybO5z8r~j&%iUpZBd*%;_948jo=QE8NBH7Kyh?j2
z>ts`E*az&)d{<e;$?t7PuKUSDl{KzsWd+u(tbO+@kp)p%Md59&Un=Vdu6JYDJJX)r
z*{YS4T#0OcYVarcfAbSzEUfM6PXD=Zb$P(sO7OPBdvsi1$j5}YACt(?g106AKW|s<
zs-jQLt7u_D6>Yy!S#Q{3^P~l=pT};nY2@XSWpk+*d3U37)HWfP`XNsX59DbeR}AmZ
z-a?krOguWwy$j;MdC;qp$``00TS5G*__-eLLi<K9_ewB1vuK(s;N>b?p`u>n<!V=`
zqKtT8zjUsk5OPhX@%;|YK%?FS?J&LG@%XeSxssm;Z|Cru`P?h7?<bVkt2#OI{2Qin
z`EZ5)g+J#+HuZY8cEj7-Ry0lCcovKC`P*6YyvW|1J>&$#+xhUeFuc79Z-@A1Y9+dp
zm_8Zg6Q^t2s&u_V^P8WXuCLtFlxrfhC^n6(c5<KK?S&d?dPd)60)3ZujZ@{?AXP2k
zt)C%Pi91p>J1R*fI^qdQPE?hH33`;3pt#&bvO)28!rNqcTN2*-p#hDYN4Dz61eKeb
zpzF*gE5SIwY5c$H7%OXY`i$POve276Qk0Lu+f6Ow@cG4Q8m!$>kDg<#IDJ9`)Qb0i
zYlE(5a-2HA+Ze8qzcK@y%KQ{I&p1wR@koqXjf~Qz^YjxBMe5!iYBvY`h>li~XVEI1
z5TlzzV>EiYQ_i38?$2_ne?_Ohpz~Q$*sjxiY`Pw8S92TlEkC<{DQcH(wN0m2kuR_`
zSR)n%>*c?}y5(n+vrCA^!nVq=c|)K>BfK5@cCSN|^2v>dFGVIev?kJ_lP|b$T?<jf
z8ES<7Avzw&9<@()rOdZ$KDh-A!h*H-rB#DhS(WnK%Kb4|7v=?Xt_)LAw4cwIb@ka9
zrd6jLa`y<)pdYD8I@%SS$oJ$iM0z;To;v6qIW%vDL$|g%^v{J54P9y1)Ns2NP{VxN
z#;%)7?Yi;KuFIuDR4$VqVz&@7m_yhL;LxufLg|KvsbZHfcIkwvUw^UzmxihI8O{gr
zwp#Tly|zWkay&{EOGo1kjMjE|8v}29!`oThdyja<XfVCSb&X<FgZ{NIAIk`iR`p8}
zD)JB6`7Og$6yCPv%I+JE);wH=gTgc+JXD{lqt=cM(f$!3T5u^uL);xIHPxY3Hyql*
zbDVVFq4l#u^>$7ud3oWAiDd_DWVk|(glXTmWGS6v4|_Z6D10vEK148^i_paj;aYq&
zTn%{J4d#ctMd*1m{HF?6EncJL!@|_HK^S#Rs1|T;I&;;i<?oEz^4h3_w~gu$Xwp}W
zO$sk*($ecj-CJhVg%(ClE@{-~9R@v{VbIjk21U*_=-e8E-rY5*)JHxRYNW<DX<|Q<
zE_XJu=gFl0@cWk4qzk=Fn$yvwLCs9swb`U;kId*Z%-VI+q)LxXiu}i<>90)c$aU|X
zNww3=dNk8aKAXGVPbJr?KRXos@gg1YRU?zHrvKuj)}!b#|KLw;;;%nD_^N*kAJy3C
zqw{uuHT^L_PZ|a4-%){DH9t_F3t6;nG}`9VL8=`TEbBKmG|e`>46ti(b-VVSvXccH
z!oJE7*@uOw_pA^#yBDITc@F)w+M)BW9m=DJQWV~n{3%-b%cJ!d`wh0&i`F#XFdg|4
zrUezksgJ_Rr({PFyuG$5LPwL?3)mdy!rOVUwmH`(c>4>y%_|(GqcGRCew5Bkj6$pL
z)Jb|z$Bc2z?VYOm$f@+j@#@YzX6?jSO@14zmc^agd&#MB%;x8;M1!y_QT@lWix=LG
zzCe!|uUYyn^i&VYm89PF!izL5I~h$!vRa~B4B}ok4Bif4w%g!u=D6tGy{R7_mQ7b}
zG||o9kTs6q;rU)PC}lHMmmHk4v$HhdCVN6hkfjQ5+bzge$r?FowiK@#yd5@|%#Bg(
z2I*X0Er+s$D==3zQ_#>1WX`vQx8KlvOwUt3*VD{A9jcheo*3S8rSQ2%O<@Nf-ns4b
zv>N^NJ+C}X?+GKuSJXjxYuH20TZ#HJud)`9UDT#zWqq4Kc5GT@Eg^@Y+wRJ;!MF%!
zlus9u@0-ErVSH#(CAq_`m^I`IZmgs?FxC^^maS4r59x>2%V5ULzj>!08JN@abZl>)
zzJs^bYE{+TwpGcd#jgZUbLkm3GE=)gW5;Vh{6^HG=iu(qPiUxbz}?x^$oZ(M?7zqa
z*;hp_oE@IP+mm+Y$n50&dvcySEhMY;%K!Aa7P4~}qba{M0<S&n4TOn(U~L+gH}8M5
zmfaL&*0&^|^W67&${LBLd=&H2Z}U`Ue6HHS={91od1{XO@pef>4t0Ev8l!(Z!E56;
z3k^~iGPrAIt6xU8Y6h?u-8)D3DrT!|UbfcdWUFB|=cX*ahPS2pcwi;G`j7d$&$G3S
z%LCqiI!;~NH(R&DVeg$R9UDl_;MeSpMLRqeuTBL!S;OhlHR2_PFk-M&Lw6U?JObW^
z!`sgA_R7+9{i1Yz(>Yb+FC}Zxd3-vh@x7Kv)2-@hT5%v%Z}U^+Jb(`l4QJUgctqpz
zmo7?@V+y`f`ofjr?bD|T`miTKtEgGS!xJ>wEkU7ZSB8q7PkI7*nDKhEB_4eRb!-wI
z5?ETC|2Oxh#%g1mSoV6yDvDXozstz4d4uMndmOob^p@f6XZV^=FZo<G7+fALEq@*b
zZ_D<Mlh1bgzJ9TKa*=-4b{IQ5Mx!1_DenSx%7G}EIZsbZcId%8hn5_4=tW2<^9*{6
zNikZ~BSyy2F?tmoE5$kG!Mybsc)PX;eEpd{Xz5nXcCqSsVXOWwVb$rjR&DDQEH8Ll
z%MgrTI!I%;1Zf2sxNX^O`eJvmJb$w3qL)L(iaOBUgec*vO<7OvYTw<iV$Va=?L-J?
zF1sfGVAJs?Htqh}ri^TxBEoGd+MAs(_#{p6_R2G>whj;0yg3eb&4%eh#aR3~)G2&T
zvj#SO*UE;T!KUwB?Rre#uNap%-~V+HYCt!K4lH%(H+Z|@T!`|P*;TW%O+8B3l>R1I
zK{q)oaOK_#R?ct1oCAY3*F9Ka1%s)TY%1~2rc!0>>h57zHfN5$0l{k1-74QNLGrhe
zCCKwqeWq1+4_oQo2CHuwn||T(*RE~TPwruQ>lveY%t!OPMXSnF@)L$fsh~4T8`nmV
zEfubh)bJ--g==I8tbO4i=ii}<@b%K55N+>6P1Th%!1p0q25&dxQ~Aj5n_SK^KW(8m
z4sY|}?fluH`gdBW3USZ>uXq0UfG{P&+Y>V#T6WYS&kqjO4-VDoF4Xhq`8OwrYPl^$
zzw`4fH`(>Fgk9Slc9oi9XNOe?{RfA>qxYTxOZx9Ks^eRuraUvUgW9MAaB^c!6JA)8
z5-u6_2fUqD&8V+0806i{pdULJ6g}9${tSc8!rK$@Htmx^BV3J2tZmc}BaHg?XQLty
z8P)rtQPvVBl?yd#N^_HH!Q1`3E#BIsxm*F;Ov?DnBrCif25YNvJ@^;a!`qp2&1zl5
zT`gVQ@$9(c&2wkg?5@!^4;8HIp<!sWZ0Nfd)nYz;##<d)kwdxOM=x#u>Nh<=t>NuT
zc<VVUP=ohU=Q=EMX&=PS9IJ9VS@B6(HE0tVk*UF|v^`kQ{pmAbv1!2|yKZl?%f81>
z-zP*>mV_v-GH1HNp=w+uRHr(H>Yve}>^9>ySr@9*JE7VIZ;$41rs~gmb5S^Yzi{?q
zM5tXiddxi{G>jg^rPUEyMGvCRRAw*m))(G}m5tKEz$jfm7_E+jWAydn7?oZTqppWy
zbnPxVSC3-V7fnUTpRrnaHCELx$FhqxR&Dn4|Mmc9b>44>4kKn=l5{pne_c${BY4{%
z-fnuIq+1@zx}J&8e19_MfD|ps#PdHfRZF*$*#>V<G)hzBJ><*6+p{&&b;gxBZb`<o
z$E6cxGL@K3w!kcOp%b&T=o&g_YMqmFsB_w9scK*R8`QpAi)8CEd1$@slFJgpF0P`v
zT2n1cuf}I9zbzRCTQikBCR58sqq+ZqU1ke1<-qs$X+xGalUI41zT^YE<b|o3>_4+7
z?Q@0}yGlJK<&aB*25U{Oa(^Q`b{D%w_T;MmA#y9j$jKtRbT+)*N8Rd(ucUD?><Jx4
zZXllF^6<7o3cF?^Drv!=>`nOq_gwKR)u^P`1Mn?f&y^)FPy68Q1$e6;^T?8bZSXb#
z-VSKZ9Jvv5;!~9f_Tzl;mYV%$F1<2l)XQ>pxqq%Uajv+%lKhzk^kS(ISEC~vnO<I_
z8kg7CWEAcsFX_unbhw;TMs-4qOLoHn*BteMx8Cs9ZFaWyuFKYhx7n(~74$G$-`s_F
zSF*Jc-c@>)tsS?rHST7%4&I=r&)bVHvehLmNB^cXV-08bCmg+G&(RO?Hfk%nw&nQu
z`S%9#?;V4;6_~wW3d~jrHK`w5Z3J&;!`nJ=HUr)^PRrKmuxx!n7kqsOxrXSIUT?$4
zzn^Tw)1n8&K07B_qD#;Z56aR&cw1-}yLVb<i2t@4Nz7X{c<UP~ZHkd1;qBBa$fG0p
z{n<+Wnw+c(ocYRagQ3h@JHXpacv}=bL7CrD*|U?PiS&S*?qD9WC|N;$lC{UqY_?Fc
z%AgTUZ;Kx3OQN>6OVmH`b`!Y>m+)zJhqoQkkKSGxuW`S|YwjL2zwq|T@;G%m24~T|
z3?~yI9o{yDw<kSfIorqRcX-=tUyOpn(NT<#QYp^CwF9U_HixNaY?vMu4AVC^Lp7mj
zn40;}liWwYb{L=M-rI6as1}V4)gR;N@&3Yn_APxy`jICR9r|XjLo4BJncEJfW^f-a
z&J#j^?COya&Dk5G8oz{ScgqkpFKAa*`(TY;WR*3;s!{N^^J^ci4`TmXsK0t~@A~-M
zS3{nl0nHCmS9p7qS=b=F79Sp2*iC5Bk()s}d)lg^_F#p=+tf+HnmROClP?8n=GGwn
z^fpKz>sZyHuT`mo;P?>s#f)Yr+Y~FinXKBe(yH5!*^BBLtl5);m0dqr9iLgW|A4=q
zpYqq+byjjgY`P3@O{MKB*}$sGZ>)-R*i<sirtQ=1+B(;+lea_2DGgDN6=V%{w5j8}
zVCMAX+#h1M%%))V|CPtLELfL@1#3FI-SOF~_qVJnyq#S9)mF9n&Z^8>R`rIpu3Uva
z1t|{ZzBO3&+YI)@!P~XXgH^E}-^UiL+usK3{7ai!U$v^)S-#IMt4=Mq>e@J~vN@{+
z*R!f5HGVv`Wm|X~0B@JT+m-OP0=&HjZ@+=JT{?znHm}7d&Mbq|9NG<U3y*PVczuV;
z8ysp{+@UV?5Vjq5s5Y<pW_Wua-aaEQ_?^k2aU(+{dZ2v=hbVe>h^z%1+WRR)*Wv3<
zyIl+6?R?G@8INqLJlsx(iCw#T+Lbtf-UZJ?z&nHf+-K0|`v$FOXGDK%R8()H20b>&
z@|!`vvkeL#Z&0m<1|4c<(2(}L9cWN}c-s!%uDxT>xEBTmnT&c-)u_$GjT#DX3&7jL
ze;PG`p7I=DlO|9%KgcxcKvR<*w=$`I3zK?9nbfwlNj==@KR05A{F_PFFVfpCXI6*$
zW`&J0YvP|~jexh!Pq?c;yj>UKp;dJ}G<1T8$~W`WfOTl3Pm>w>3GG!rJA7;SYw%#^
zxA1lryq&~cxB0R_ZJ8ISiqiu1bzY!CY=LT5F;LBJ1nNkwAa!eF)rk|qdRD=v`uy&6
zo?}y=ziq0@8SDt>$@cr$kKi4u{7RwhMhMj-cpJJXRIB0bBQEbhL&-S^)AM~{y8nGR
zJH5iS7T*4l7okaLo%{dS^IsFe4)+M<`k`w>3;iyWoU30W6#Fhh?P2b_=20rM8$IEP
zC=F*OaJU4zg09i*Ig6EhOE`;%z;|sdX9Kk5dt=%C7faS7-0k92#IIyFpGi^;c-t4=
zwu84$vR#~fEciU0h#q+1$zAChLk0@GEwDM2;G0w}u92!%yVEou-ZrmJ?~mHHNo{HW
zM)J=~le3O5Z6>=@#!rBUm)Vt289nKcZ1rx&u1|Q|^v5juZiKrXa#V9QGe2v2%?T#&
zoq8v=V5U@+KI5oN9fq}SxW@N|yGVf=&t)%gU-p_c&CtvG)YA>wxw1Y(6>p&3dY3^i
zJ@Vh#(m*t=!(jjUztO4k_e<>0B&VDGrrh@n8purW!}D1luEteT?IRV{x^j6vewL%j
z&6zKE&QYl$IXcGU>ikEJYQ4(Qwe9RQ8G(lOS}wT)aQt$vK0VCUQ#{5Mc9OvbfBRHJ
zgKK5);m2H^c)-`f*mJOc-)7jqI9H8-%vD8ryA0kQfVY+BmshWe<@EyIS`xYJ<rVgZ
zJu0v?eMYvvf`5xcvX!<qOO<=DFAO=y$Td=vBAGh&Cwrn^X6bvFc;zDXC0v~VZ-2Q?
zP8{#Igtx=t>|)*@!j<rZdNiKB$?(<`%5z>8{qLD<_IOf{^6%Azx7&wt{^`K`wfG)R
z9>YU&i3(+Cax584JjXq2NQ1jbn|q7%zSNet1MqS$8!9!MM8^IyGVyleYuyJ!H&Cy(
zl-zhJD@98?$hL11POtb7ncwt74rZpSZEU)#ky{_siQHRwJ7RYV^TI@JOi$Fg0?BCd
z(cD*{$8rST7UTDDB)@ZGQZ=n-s;ZD1u;&1IgO`)ocSshhHHmDyBpvomQY)V%jr3&C
zCA{4RZ++lx(V1i%ky#MDI*uLDPBrV})NZu6_0j$OO8=)k{~!MN72Oa2Po|?Ia>Ii$
zxoxa2d(lINw>Qu%?ySZ9Zxu7jCQ<6ctaDqbNX<X^Ki<X`2vfo}>YSAh6*=$F<U0;^
ze(xY3g<jyOP`Sa|4e&O5La1g0g=%hR^6cPk(_+*~Xf1o(a;Vy1tN!|y9pJ_wZ7Url
z?*i;PyGvf}S&Q05TNK|lP<dqg{Ls;&mgTHE^vXvg;q5U8niP23^##7jL@$N(_tctu
z{`iA~RR5JllOrtp$G|<Uc%Urz17!RwK-2G|PiY*aye>gxzy|3nc$?*C(NS-UhWe0~
z8$fP>#iFGCcq6Y_w4-*A#=+Y<7p&U5$I8BVtL|;FaJUN4%l-ap(l$V~|J%FD{d3n)
zn7<}Kg<7)1d`FNXe?kK@Gf3G7g7mr+dG>HIsvLbnc)O#WP1WJ;dU%@(Z#~xsYX$x5
ztt%}$P%Tj3{T!g0O#@VxU2-+b`Kkpz(yg_9bzzyW+Ac+xWAazux`E0b5Trjh25D6o
z^?aN~<#z_^#lb+;eaHTrgBI<)6GV0>+&*K~A3LpT{EJoXCzJEr+NyWWg0yvlMOQ~!
zG`4|7j;8Fu>26We!Z3TkO>^8sRJ9X%gz)xWZioyaFw-YQwc^p~2!`_ddGI>E`pTgc
zc<Ua>xoJSCI>OuQ@b-6jyU~T$8Qvzt+qDBi*s~a-LKQ=FU@P_dYUTjJc1?n}3;(t0
z4;bQ;Ytx-+?6oT$tnUg0v!^atJzpBMe}_Stw+-4yeR_MWk=}+;aqu=Ao_fLC0C-y(
z-d2ORHR0`>?gp~C4T|M+ySEth=_G%@tWhoCZ54Q%25+k$G_s?XnJ_hO$@@l?V?Nvs
z-u_<C#QDjj(Gm2WsjKbqwj;bP32zI++szkDI#trFoN8tzz}xB<%=*^{4OUlob_}@d
zr%(@#8{<J1mxqd>iN3PYQ-{gvD1C^$=AZqvt-8O86(`p@FhEy(1ggZ;K!q;}BtIfh
zh57`_yF{SMmJHNal>^lp-VUu9q#n)Kdk=5xR<!8>zdP|fS7qR>i?3b(j$k&inOTNM
zh)Q>$XRwC3#19TVp&y_8L#PhH+e3Fl6-BRO0=%{1H43Ds;lteiFgm%%y~1^KM!2pV
z2-mFB<T>9ASB6W3x;Bi^S9>G)U5Zc@c>5OK7J;`TnO&5+j3*!_T7OhV_wE{{<@P8I
zY80hJ^z#*WprfNVmhwX^IdCu--j2+Nv9Pt}=_H-Ko}^)aCTYPV^izB+<9L#u_eqiW
zuPLhiG)1eZGbZ;))`ia09{944PQ$lN&oAawx(4+lPoq6q0;|bve=jXf%2b02nVQv!
zeOLLJy7!XZCP`UJhqrxOv&VZ1`xB^Bx4fcO{XSdKm9x|t-Zt_kLp(8q-M|?-5RtAA
zzo+W;*c5i9kyUqytnqcsm*<ncMbCTLI=t>n63EU>kQ;T)!>VxjY6fRV>BCxTr$gNL
z;q63f<V1K|lzE}Mf3_-9)4e%DAGJV5t%SG!@HQ0QK7qHb;B8G<lhr;)Nj-D4hMKb8
zw`g$la#VXzjuxQ3oiQ&*e=p3@$DQbK;c*aG`mr4OI>`<xnWs7*a@DRYS*YpkpBW%c
zA0)-kf=`8{fK6!2TBfOT8acP`=#>^s(;n)!%Lep4S!r5bDNQkWuA+0(w1po2`ulj`
z`I-t()6^dMeHfnS!-dIEhqq<n?bjvg1(boQ%uLI}+Y#_K48F#~+bZyOIPaglnyt0V
z*p)LkTRGIms~yy~E10>S<hg*iN8s&V_UY7JMF!~lEbV2#MK+u-25%#eGbfmvt^r$_
zrS3^r-hT24Vdm~5=~{3#UB71WTEW{v6QzSc<N2r7Ed+00^Eih!Vb_NK$J>?g_G4&<
z?*5prd2Q3Rg5R5!;e5~1=`zC7zw0Hc5V<e^!rJlZ>qD8l8qpj4u#2oDc-x<^@f?w=
zZXPN6JUCw6D#yz&4?R<Pe3zBTS)yJY8J(a=dc?We3HtI3zCMcApSR=ndwRT1JMm>E
zqGwnirzYdkA0;{Uj|)D|d$C$I98U;-_)l|VRhhHtTC_}K@VAzszI_L8t6h)L%6T!W
zoEoFQ-$v<KgD9n;X`MSKlAUXjYBi3Y;ipi2IR|H{S3mOh{RM|+-F48LrLXwGrjTvy
zURe?(-}dx9dssD=`l?WbU6IsEo48*aFVSnH5Bw@ENWa6|-SD>Z7mG@A&Nnm*)ZKQ0
zI?+B*MR<P%S@ONt`RMiwA0;04(b-jG;4b%33|R%MV!f1k-BV*SJ$3c8zudpGXdwOG
z>+C<Bd5!%r@OIA10G${Zpl9UsJ;AH-jL!||=dYu2{<7x;C>GxO!rLM6)(LNyUnL9Z
z1NnXbS`<w^JMo4^n=V<@=p?$AgX9eL@z?As{u)j0P+fRiqE3K5k=xtti!WLOA7!tC
z)jv{m!`!lHPy3mynqShULoqhB&-x#4vmQ``!rIxDgEakCfc6da)tFMgN=fw9>ug_L
zD($Pomwfc*fsgKgP44efU)5bso_;Vnf*t(z)pz{cJ^Xd4qrXhve){^huU!7bKU%;~
zb94N(7=4fH9Y5{oOftF%{?H2eMAvhl-@(27-$41o+f9QldQiv0zD<i#I#|?XfJGVP
zFvL}}QwQ2L=(1hKGngscLp0wr1b<fuc|0L1ur@@4?}g|QujN}$=Hq5w`!6Ay@eloU
zR|k4W&R+1=M6dYpLz{9|+0<)-O)u!#cODn4!-Io0f%&8>yd4K`&&;-wZ5ga><AZgx
zR501ZR-I{Q)x}2!CBn|0T?}&RYM>8h&;@3-H+CBI0j~B+HSl~I)UCcjbsHJ9tA#=7
z-3)5N-|>UD^Wg1Ecw4fxQArhz`qIj%H}sUR(5GGuXT9L<seAO9nGZ+9+gG(sT3iOE
z7N^#IX;jN|M((FZoljzhJJ6(N=S>;{Zxiy&y4ByT@H1xF;q4Q6I|bfuhPNHy?Weyz
z^th3ya_|aYp5P_lBVJnU>8;kay;ZI=8I@DLHO|FH|NiZ(du{!7rE`D`VS#diw;f*x
zs35PC`G)|t9FLE8N`N+f2v94uRW173$P%?_@pGF-!rSRI%^t6}t6J$06^6Hyc86&1
z42LSg+ZptkeuB4gX!*{*4b>g`QiV>2DeZQcO1ucuwohS7w1w-R>EYT8Zy&?kqqoAf
z@*e$;m+a?tk5E7)J?G}+Iv+)2<wRH2CQ=E5BURxF-n$p<3#8|KBQ#1~(D)BQJ5a50
zj2wsYlc5Fa25-OMFIdA|_g<lRy`M{M0B_^3BxybK+=wph>pjdYjLgz2=$Id@!jr!S
z{}8^SwXKu2j=E$2^d!9CsXB&UC3H4<Dfy{dSTS9p57~Y4H{AQg9+Iyz^=<_@>-3g?
zj>(b@-Uh?l2j8*3h@GD8hh^&3P-@`;?3}h{>fBN8+jrB*fkoFjlnlG5IMsORl+RtK
z8Wu&Df?s|?A+pK)kPk2k*23E&@b=S}Bn_{gp{z?8a=_a&i=}K@b6f6ZsKKpF9VrY)
z^4VKjnw(vD+hi(dgFVa_4`r*9iJVh-c^2NjJ_uVMXOY#EqqW`Hmx|_gd?&mv@U}zW
z9L?LCt)^p`Z;qlK<eX3lp1nAfBLj?H@&~!So3rr#XHlz26M9MO;jN<@yCk-x>Rq=K
zh2clp(2J~g{2|$I$Z~_FE8y*Vc>5gQZico0%`A@vr)r&Rs*Y?+Q9rb|UhsAzIV5vh
zr0I5DcGk2)n@g=fWO<hUipo}-naox%@D|>_zev8}*=!Zy5|8i8CH&9y7fO&@GJgxa
zruLjShuv>HPZ?YbW=h{p7CuY-)2lO;=bf#8N@l6z?`hg`45q@{YVh`l-Rb&79&mej
zyB*$kfVVa7rR!wr41Hy%e$8M%+r$hVT$iCnS2A>unw8ZW^5Sv3rDtd>`X@JP*@YF@
z+X82gJR^6X9PvKrnC9ms>LIzNo#Aa?SbHKVQ3;h3wX%Amjx<Zulk*Asd1Sou`p4^G
zqj=qcvFCHiTgZX6WGyAb-Zji(Yac)>*ehQ3L&+El<LlsTAKvZ`N5B6%PI)Kz9UAM@
z1-z4mT%B5eKUPL~YlgSy;B6`VHcpruH<TQSdiYzJ-)_2wf3q{$=-Z<7DmqGkq20Y|
zjbxuyq#nCOGN*}9@j{_$#C)z?o`cL;ha%wZaCn=LM~wq-`;H0Hq#M+FPlB|+xK#tF
zt$q)&%Y(V*>Ff9$sEK3X?aa6!?eq!Kh%02`bz%n#99j6p2M?o<#!mNCQnp#a1<Yvm
z$h|${qur0ZlvvzLRbG2);u%j(i1U<>*;7yKp86hzQPHiQn!1AQp!Pm`mqs4$EPBr)
z0yM8zfL4DOpz`qcA-pZ$pE?<4zPI@6cq)5SS5SMv&S7r@WGZLT^*0uc8wj`Yc>KZV
zm*n{?y_&y<G@_2~?XPv^`S@<K5N`Qu!x>+7+vBCC@HYQ7_2{=g%Gv3omB)Sb`X3(+
zY38f)pULclw*xY5s%Ww4)_MBh^DOMCU{4&Hnnid~s=(V@Wc-HR@ln!6&J%z5s3^Q`
zUxmD*AISdgN^VjsUsY=3tKY~T?6=TY1y=iN;89<_xaq6U5BRzdzH%?)r=J)3DdV=E
z^5JctGXC2BH9zl0?(5XS16~HoRM4WYU(-K;w_D)truG)KfVUmsZ5q5SvCytjr_gkL
zq)+Y=q9`-9D!g4>KSb-oLX_LXuIpC2j&c9rFvO;118r(HfI0TJ%%<tD4~4H)m}wuW
zV%0Kw11Eo_4yTV@xk)gY)WQ1FHdv*L1nb~2_UoAGAF%hm`3oyEQma<?vZ4#IY9@8<
z$leAO>txWQ&eW<C4Vtptpw;WCd*SU5DX<jYZmwrgA8OpM;O$emTyTIv1Nqu#YYkd>
zj=%c`23I!fExi33-hSF`)K0Wm_ih-~ir#YmD*DUY=`;JARId~lKBO^xyw_2qN<1~{
za-2zV@OJPylh&0mt5ij^Zuc>(2)wQ6<*p4q+*Rs|yLyIsXwdf_D!kr9o2hR{@A1@F
zc<Tvom-&!mRnJ=m;q9JD-t2?$X1}zzI!;6bZRdPd)KAsc`77#Nfc)r{44xLC%J9|+
zZ&$<H>hN|ZynV>?-R-JPH<@3=puO9;4=o|QO`(VIRSM@E=H9QnhbkA|PK37=;ccH+
zczEIMKPT9|3va*vJ4^@R?ViMNIo5{j-<{OFXTr4s-sZyF0*}Jcp!2ye;qs*SoY5#k
zuZG|!S`wjy@sSE{h3@g6NLkQk1#XYlyj#)QkM7Z<HD0uXF?xoE>ig>WtiMF5@jvL%
z%f#r&S|_tI?ose|2fXdpIZ1t1kk>tj+3r$$biX9&HMQ_*AM`1?ae7=KPTk<`?!sia
zS4m-KXNtbTJ9N{VnyhQ8hLD@_?;`v|=V9=X4E;{_KudU=*d$90sLgf-(&MELn#w(S
zXMOUZm7#Yp*c%oOXVI-(>PCJIT9o(j_T&pZ_P5D@DS#H`5j%#-mN}H~#6#&sdzPR@
z7s<wnPgUdm3@y08443+GO;Kt{w>0&g#m<xme5@8LLwquIi^p+fa_ay2NT#nKhihB9
z4(&+SWO|*hk25t6Cf&SBzBn4%)J%L!@U|U2?~4tXZJN;sJt8--W0qdD%Tn0}WcGSw
zDwZ0$Q!X`Uar`V6>h?TxxZv&ko>G?uaCHE=vn>-<DIRTX4RS`{Z4r3;?iG7l$N|i~
z#r}y)c$-(HDwd4ZZ<12=^ZXRO381!n%xeK}t3FQC0ld-Ul2cVN1#dK(@po&}ROURH
z3e(t;{!@m=AHjNinR*u9ehX`to#ukSf$$cQlO{gRP^-Ckr-zYi)H6-pnx$z#y)*?>
zAeXFmnl65mrk88d6g2`(@Ix{i;O&&9X<9NKeej@EdR1h=Ze(_UmHd;pJa)88Z|DuL
zfw%u!GPD-nw)&ABumAD(afXJKBsX+=n##6MlQln0httyKLtT3)m#h?c>;E)SwcxEC
z-m<M!4d884c>53!O<eB;&d!{vc9XNme0A`TWI0WamxJ22HoV;oZ|73mju2kB)Ofij
z^NWNh?h<S*25akZ73ckV5&z@vWipFywk5;py;E<-IMpcLsou58DxijaNQO@>G9>EI
zLte8uRySd;-w-}tH<tZmu^M#=4`drMXAa@*jET~CJRQqA<NYwwdtzq0xLBz6<GqQT
z=Fr$s>a~}_dcj_l4qP*y1Z&LdAhn@SJQdz<|1(H;leyO#?7BBOL>;(Szp+`!>$E6w
z3fjtse!5`rQ~$qwb$YTdnpt=f>7#MIJX9mctZENU%Ku_g(J5ry9`w;~@HPhCro!9N
z@U}O+HNxA{HcwSd_RwELJk)iVhcZWd=u8(6bvf^)Pr1HYc!ye|oS)`-`pK3-&OMpC
z)28^TZ#jR>PxDv18~&>N(qGNs?U=F_rS|pLiAaB)gtxAIJ~hW*Q!0{Y(9mD+BK@>$
z7<s!N*;RANS2uR}YC9Q%Q6)XKy|a(HbobG{VLlps!bdZo@whkpX#Z4t!b^NL_*Y*Q
zS?jA)i^<d<>#OWOzN*d7eEYc%zeheQ8|SWkcOQM*n7_NnhsW)sAClOOGv2J?cu!58
zJhZ8+hvt3jAwRBso~QEN`MI+_G{)gB4^Oko`I<HTkXc@B$y1u=rKiul)F+zzdnIqO
zg925Qe8K&X$Uk!B{S1rhR<fuDyluDDqS?c2YTC}O4-4$-bi__B4Vr4^(DlaK^=g1!
zo?Y$gaKOf%B%AIpvPs<k^Wp6lSZnBM(`Z=xrK?S^#@TdiiA_oH_C2h<U6)z%upkW>
z$KLnv1C)a=ruaOb>tz8duslFM{CPmv0Ie7jpiwF4;^=#?-eu4q@N+pjsABC5y41#?
z1$_;=Q_Y~p@o+WTpxc=S4Xw)zw}C<1o5Ntf{%U`NZcTu{s}1_@tU*=bZB;Z=EgeQZ
z8faA4okop-w@c7CSKfp6c?sO@%Z!&^wOSf=XuMI*twy|&Mpb)aWKS^K<Ye^CXH2?D
zeftL9ZiKfN&YQK+(_N=KpjF=Ou2$%!?ho});ySW7?s@2;$x~?-PmRp-)TgUnWL|qq
z4ZL-zjkjL6_m;<x-sDdEXbrr5bJACJYWm4zyT39xAHJCypqo5b11H1Z_W>F|Fi`uh
z2C7p<ixPNlcfi}1VRnV>x62RS{?IE#=Sw-%wm&*@X7Rm;p{;<oL(YZj!b>#d@b=s;
zYTV!P&K(HTfVW{PTNYh<AiDC5aDAf3yeA)B725H=|7q!-lHdG1T(9qk>lh3k&v35}
z{B0kJFN|~dqmhwHMOXFbHnL0bbLHYSc=;24uKh7O1aD`(k5cVY5eg|9sR^->`e?;_
z-P|ePui|w-AxS^COHv24o-<~`**S@7T?0*XX)@u2=NI07%!^aAugN-VhaX~5f>w+|
z<NQ^sn!?+Dk*RtQ)6V^rs&U+-r{rd6EOlPqZ_*I@suvB?13xn4`SZN)=~~r@S={mj
zRi*}cv5l;=#c`^G4zvQiJrhZe-Jeb^-R9J;sZJ#hB44JhQ*);{b$7Q@oliNnXb}3)
za_kbmgwB+Ea}0c*yCOj&3n%I*HPZ9@%z4|gdj%eP)u#te?YIfw=0JG6j=Jr8c-s)(
zPI!{3Vly%nawb`eCnn<$Lz~6iGvW(1Vmi4UgR|8ASf+M-ov9cneDz{y)+=_honc;F
zmY;1iEVGk8Zy}3eN;2LA&LeAi?)sC#S&_OelYJ!p;&q0e=1!M5^(vpDGmr4Ez}pmf
zJ9cHN4vd3it>Ec`6g_~qpSLBc;Rp7GvKLNgQk3|R45Wgo3WK*1KciivMy@|UmHZ;|
zz)rFgrZ*Yc@NV)dbiT_n^ykG4jebGTw-EXycsmT<R-2Hf9^bNG9NwnYhrRT!L(<qw
z<4DtJZ~U#rVR6M2?WBiMo4knAljF!W$7fn1UWPy7wUn&y?QIfeB9rLHpP02S#Mg@6
zy$Cy5(j(}3{F|)fWb_`rLuSMgW~WaQ6ocR94!MDq4eXGcoTv>=@#fG&o?0bQZL$(I
z4aVBYYbnC(IAS-?Q=J4ghPNx>?eCletMn$TKND>eJ>hCx5%6|<R=nE5+n?d>RX(;Q
zhUXT>UURUs22Xl3m^&8UP7jRNz0K@bsTQmCHDcutM`PO4FRn?h5WEd;9ZUWk`UZG=
z7#;9G^T;v!jv4OCDD_;AUvEMTdsWCVxQ35&1iDtYDD51||2cSTy%nl5Xq@(;A??Ti
z?dnBsGXEZ=_sp8o(}PqcD@awV25EeEW|!_{vH#_#7XEk?gZ&hLlzdq)i)>bl?uS~m
z?{9K<Ui)bR{LAg(qkglzRp5=69%Xy#_Z$y>rnWu$fEgORZQa>jZYMmH{mxxp3*2>S
zsJs5bCpr8ZckQd<u2ofGO}vMa(mZsId)^JM>m%svmGRJ_Q(pM7eHC2HkBnbGMV0r{
zi_U&JH-=1lcv~U^-6*`h|H5B4VNk;<fp~ymFT6bfZ>{k5Ll*wZT)vjK%li8&ah9*j
z(ZhXsj+z$U&WE?#$Pvtkx23zW<87diyf`CxEhHy!lMgv`KI%TiM?Qn7d%0SE@1v*V
zd^C!$S@aWi@M`LJG^~Hcx~p%P4;=GR&@vw#yyl~t@b=9(Ga7P}UUVei5Z-p|=Ar33
z-PNfZx}DBu)vaUJ`s!xoo6Y!L%u2jO9R_c&%<z)Q&ztNhvW7VujF{#pzarF3zri<`
zK#fjk-d(|>N6qM^|7IayDF`odkSgW`>BYccomgYXOKO)9-rj+?)4JH%V`q~Wv*+Hk
z(XhP_V#i>RRx=lO*=f<<;q<0kT6FPDpsu=GWq`NaB7#(^caSdAGxcvsuVEB3>K_Bx
zX%c|oFhJX2t{vWPqu)P2H9#Y}1n6BOt8D-EjbUvfto^H%LB|pd@_@CuRScRGXTZm2
zKo`NxwvK^(KbXo4w*b7|25+mv+atdj^zA8w%<wi7-o6cCCR>I6F`DL9wa`1aL*Km1
zNR43B#V&Y?8sX>V${%Kw&jzDD9yBWcPowTq^Im|rZ;qLCuBcfTYvS+iXx7b*W^H?F
z)+&QL9ujw5zu>Mve9$M4^FRmUsmg=Np<L{#kJmi4zq6NKhkNTrjJK92c*|AZx>Utm
z7fyNW=reCsOCtNZnV<Hp^CP<`K+h)z&{GUh@C0@gEu_wU0Dq71dhuHM!Q0>8*yO&M
znfD)P^KRR<8QwnJ93nsdPfmF0P}cBJ-CGl?hVXXmKX|?eQPWOCv%Vlq<JO01%ZD%(
zD;2Km0pXetZyU9PrTBA??G4wnKg0C|-qwe=7vXK^LSfRAP;~yGIzUf)^tK3fp~w6y
z9`AzRM=Ri2q?XfPzE?C#txMw}B0J^~zN3bo5gO|msi$ZqpZ<ogJ0nUDdN}p`3;E#a
zI~PpG(~H;W(HiIfX*c&ib!sKN{fPHrHvatEHOaSvx1~GBY2ev7b)23=u2>5Fv;TQ^
zYh|%p8h(6REmbd5;hcqi)^4z&G#b<4_$8UmMh!}4|8as&!|dMND|>8@)1^gmY6x$$
z;q8_fa#!y;**(t=VR%~z-i~X)`z4)<gt=|ut$~`?;}xE$w|x9>r%wIjlo5WL;O+3f
z>?^6x?yuJI>iwDhFA>z;JdQ!HQxvcpzj420_5O}bp&WLDQTq%kjDPtMxvE?#BgxtG
z<jgQCRcR&Jbq#N?e?Tu}BWvMes_sz7A0sz+6T8Cd!<T(gWX?L0)GUC!`*KNY{5+BQ
z0yP}<TU>=idYgEl%f-u`f1}hsGJ_U6RlZ7`+Ic0a<fBC0c$}zGZxZ$HK$5ziOVY2f
z>B0u~ed3LtvKQ8#WPi(@6g7djbD634a!FN_im6&NlU=6p_Saphs(BnwDt=Y}f#mRy
z<D4-sL!;oW|3c0f=lJ>0rl@576x|_f|KO$+?Y@$#F7UPJ+f;4fxrm^DQ7|x>8C9|t
z<R+`=kIZm+-lsKn>Kd=ZvqSL;q$Zu*8C_G)L@nhkwh-Qa8I;Iw3tq!|3A*->-7j}J
zKV2n<mtOzoISJ&>Bq-IKplRWWTE=rxp%$F3P5yT!^i0&Yd!pD$8JMW(vWfb%n!NCg
z1g$NRph{%*75O?|zg1>;2Yj_8P|F69Ne@HE!rvC~wh_EN0B^6rTMMksw!vS%W^YR}
zX#-;M1=G{PBU%OCw*Q)*F}(GHw;%sxzT$@VcsuhN`b93&$TMh0h9G_AZl9y&EN@pc
zG9?yVvFlxV&c_Ln^7%eWM@-}|<wa_sBU};bVdN}_l4<Es%Qg-@D~tzacaRRj+a{6d
zkYa;msT8EWerT6q&@Eqfiu(D<-{2>2`kj+~eD&l$b!iVz&XZnRyU<J9e)Up`nqI1Z
z$5RD5c&b4S5B>VL8J(J03vQa&>ts@)CT5*)&b*BK+&lO(w<|2ZVOCy!cl`!$>y>v`
z`a-j+&M@m7oPJJiJm+h?l-wJCJPvo@Z6LfY1aB+D+a+E6v}m}Wj-K*SU>_g3ruwk+
z*GETrJi+<Y*E@Vvgnsa{iazLkd{pN%x%jc{c&q5EFVv*JZNzJN%2zYb`fAfUU**pB
z(3<I93Yz1k4&m^0fRFaU(&r0&)OWUzEZk?y{_Lf8KY7WAudBt+?!uq1&+}H@sXp2>
zkKL{?(Ib>|#1wZ8$n%o%r8n~@AAO`Y+GRBB{aBM6c04gpjrwg7taxkC>^%lF2xx?s
z7_=B&?uH`xU@p6=TMJj!HMy!@Emt(3p6bl^+stDt`-Gp*$h?=H<&)0=WE#<H;2z$l
zyhSIg(^Ia`PC5F^Ra=qq+s>lx^q8kKLHFGse=<GfpiVZu+hx(=B^EWGN=>}Lq5?4X
z?jUyH)CknJz(D;t312C^?UEKqjt4Al#tgkN+5TK-8U*MY`uWXF0XlWnUk%qV2T$?W
z#1{Sv8tbnS+jt9aOP}TUDA}OOFtinY;(gSwTSDm#-*J=cBR5_7;-(4_Xri(V+F1F2
zHSY7K2IbRxuH4$7yxs<FU0_g!Bj}r7p=E}*tL!k9zVaA&I|JUXhPNT`cH=yBQ{NcL
zc_&i?U6wn%{bfD1FTAaO8~;#oG*k4HTOToLgUPIG{7x0^WLDgIv%Y$4mJ!~1z}r{w
zwky0nFqXc<I}a^K$9G_fr`#@jYW7=C)hghvh7NCi$ne&^&fe<DYu^Fh-iNoD**-eL
z@7>LR@#wme2U(k$#{lNG^puY*2+)ChJZE<U)#zL1+QrarFShCLP0VfJZN^=jwlZg5
zGKkET*CE<6kW7+%hoaDpf5MBEQ8QF2dqUMNCQP^Ct!*9oD)2TH-p+%!u`%JQ8XvCj
zSA}a6yuAW%dr{98{)=Ao9s0{(LNyN7=A8*uy&?FG#)oP<o+I~VWR?CDs?tBB$=V&N
zB+lYKg_!@^sgF0pY<khdOcAn{ryq^(<I8vW|Cn#pjHHff$u7!C%uMGywf$$ZQej++
zCuDj1$LVYap2N&Ibxy`Zv4VPc7dhi+;`Ht|-ri2}`a)gOq)?)+c&5m=L#l4og0pFG
z0M5N<CV39t1{6!x?BeX?oS3MwO%l{?VS<KpkF2{RPV?byd3akE-VV6!lmo_Y^(51S
zK3_FvZPR+k>T^M-y2O(88s}7VdV_t&F^hw<n>iEsra3jq!>Nf$PL0UGQweWvT$`fU
zi(QuYAHs3+_VS0sDY{pj3X?_f#Y$c(yuG{%e!~|dyj=mi>rumxs>aR}&KO4epG!}n
zDeHiC>ujn{!`mhBb{xF5z}so?_6fZ04sSb@W!Kr2M13IB2vX|mm;`m`$XqdkTD$;R
z?3Z9V&+WQnPW632UTZYI&ZIca?ZNZegiOC$<TJP?sWoSqEidq9jZBdjI`I>?QZ%3d
zK4^H`2HyUc?-2e+icSaP-P+9=X$yS@7~O1bl74v386cZ|;RRCEj(pE4L5Vu%M_pM4
zO?qUKmUAY!aVt@IhZE(sne4F`_LiZ^jX6mUeg^&eJg4>z<8h9lf8T`1=}l&mDM8~e
zB<NcUJJfs=$WBX8Q{J!LK0zbSvCHZU+3t_x6<UlwKRa8BPoU2LZ%5b@<PLA^{Tr{z
zkEvDf#B0y7c$LiOHD654Ixt>I71&eG>(#$>0=~=t@pfNV`oZwFC%m0Nt$NKbUdsZg
zWxdI%Z;Z}9H&*@pW3~8Uj2>Nxk(FAsR9uY4w2Dzq2l8s!(ba|w-tezt)dip2l?O3e
z%2kK=kKBt<p)nC!u5faW@JN!)G^q~095S1{(XPDDv9oppk6&BnZsaqho(a=;cgY{H
zhU;@$gx1`mA64F_uN%_u%f&}i$*QwlTXqJiQN1Ajf#&Hg+StugEUFe9s3>@wJ>6Gt
z*LdrkmzPT3@>Hn-ocq^$=;}^SB{Jt)TE#;>%X`Q>-9z8XL*Mv$=;C5`MZGX<c2BdO
z(!(1%%A}a*My)iM^sF=b(M8m}$4%;Y%Y=X3tg|D`%m=v^m<gXIivZrX<FZxb&o!w3
zYBT?Z*VcAs?St2ayO>q_8JT`p&AN>4^xj@KRsYFNvwFE{N<BAxY;L-b*JJS=yq3Gn
z8eRcry7(%_;;UEmrZ4*W>hV1C4d@Ryq_&+3Z-Xa$Xb&7sg=xDD%$a+TAxQ7H%yfR%
zd=LHP?V$-g9=FQ$dZ|NW8<>?e+pLDEd>wW8oRJ<Xy~AA@7Iz&#V^m-_H!YpvrZzWB
zs&c`kXuK^NV^U$WNnY3Bz#OC2P`6Emw_n!rxrx+sz6O;kXwaBTuA14&RkK`N6;{Jl
zM}KtHFKJ%*6TS4Ts5d?=Z^buaCms5o-~GL1k78$CvPFyF?dW<I<`>kc@b*<(i%xX2
zXv0OA{C$ub<F_h=eylllY&&>c9TuDTx{<}vId2No14|$}uK>lN=Q`>ZpfB+O`mq+A
zXv91c*3RZis1Tro%)d9D^HVIf+^aY2uT7$-4{v9VX2;z!au9#<*Xpfsm>PCuZF<Sg
z4d@>5)RyGuz2>G<cilAYU!I>ZgVw{_1bBO-CR}EQ>!cnIg}eLdJ-dyDzlRMf{?wq4
z1<*$&l5vr1R55tl7T!K;h~Bvc`m46+qUstolFtpUY1E27MqW3g@?f#|Wpb>FnRGVB
zqz)TQDt+6e2$=hg-K^x_%$j!0tf~K)b>HN!p>^EVnAbU+>pv~Qk1P3o-{pzM-Ba7%
z@Mnvcs+afD&cR+vVy=5)K7E(s-U{jDt+Y%gPw=+zJ7&1>wiX(^QT@@r(^H;_X8o`F
zfhquRJHXqNqG-#?q2VTzs_s^DSx#EDYBatxvR3|qw|012bRqhP4b-~yqL$_07ib!)
zwBw<gIwg$$E&unaaqE2ylM&udilVOl5~f<@sI}++{PIVks)DZh=;KgT|2tGo(OV6M
zx2552Wq9k0|G+veR9Ep{mw>lJ@E_gT6{`9-&_$=<1sf5j%zwgEaATMb%nMV^m0_B-
zFkHR+MyW-4y!E}D8n%r7?Qj@(BUUQ|omv2I-I<$~gtvFo<J1G*PK8nHV6V^SIE}dD
z)VY!Db?(mnxjp*Ny6o@Bc4~?n->0ZkTgp1=S+jQ|EkOzG63}+Rlhz6JuoE<^af15I
zOHlk*@w#}A-19tkX&T~m^(*qvt2<RR#Hpjj*`Y+QZ#z1c9`N?h0!}jCocbZusg8W^
zPp(1boLUBF7o@Vsg!x<RtudNHet0YH!QIIas8~J*PdQoWWRiB69j9jlI3u)<!&lAE
z5FDpQA7Ki-z3D>jyf9YR@5U;E$I$=3&zEPXSzMeJ`S5R(bA8A{CP99z7Sv<s3O(@E
zMcL~K%TuXe?Y#Xmi`g?C=+9_xwr(L~_DU?7P);@O=TrsGF`o1eicjachPT-l$q=Gv
zz9cM8Md5A0batq`fV<RxkuS(d@`Z)*3EBqBIu}fkzh{D4PNHVsny7wT63I)WZ{C1^
z{}3}wYQphT5|pwaQLmRLY6zd-mIG^TiE`4f-O2a5`<`8)TNCu`XYxL0C8$_$&N|<-
z(`6d7P5P_NcJbWd8>@OCmObxe2zH=v4{r_p9&F+ab9|^%ZU@+n!g*xHYo{vlyzTcT
z%l|bw@jQ;9_2cE;klE^YWD{0LJI;Bk2)v!<mZ0+Rb_~3|0B`pljaSoE?5csciSV{X
zg?MeXaMm!z>-y(7g;k_4+=~7LyfvbUiVTJ2@U}1Y>V|N7H{`F)d&d8pRiWxMm*4lA
zVR}Li@1#*-s<1Fjowr6XtLJwBeQ^$1ir(;bSXcIfkm=Xybc}An+)s1J^a%@9foJ4E
z;L+Jt+o63=Y?_LW^mlmM4BkdIw4+~VZkKOU0p@w-sA&xKZ0giGSfg6t%c)`2#vH3c
zxvDU0>x&2GTX;JOeR22A7G0~&&d_i_?fB7GC5qA~p60FngSk4eS8ceb2G8)&+ZcC!
z{@Ptj*1PK$ddDx4Jd_pcp;N2e^)ofY<A!F1P@fh%jW=VYNh_)83Uc2oR?S2<f=O)-
zne@+flXAH4{S|E1tHay}Ew0+)Sx(3QE{m6^tiG%+iyxybeNq>lFuUl$rE=Q8y_`;*
zE2lk$T-2_ii<(bz(Yn_z8W8KM@K`rhYT>4)@A$X)o~Pr@sshI=!rP(fg(Ki?oiJae
z!P_hFb~rQGg<E}958m#Cm)3D!YE1uj!wGLaZ0(`y)ZQP;nzi(WNwa=8>BKk_nZza?
zZc3IyJ(H?ua&Cy^+z>?0pf{e+k|tIC%cxX-_UNu|%zzE_`VIPe8MDdjWI7ZvDm)Z#
zDU2Oj)u@1<4664d9@Sy4`ggF4F7<bje@_=}@8Y66Rb1ru!9_)@yK1((n?i57X;Fli
zy1w?37rZspWB+RnI9t+NmX}^S-z!i*!`th1*_{q=&(ITI4R1T(#hS+3z;Ey~$57XD
zonp@0d$C0$+|lsdWUhV2U*>)O+P@Z0G&QO_yzQ0ms{;#sH3Xg4mR@MUZurWGb}f1p
z+A}n99ZsS-CwDmEE!o6q&n`9hS9m}E-ZXaX%=g!-4gPB5FeswBK@Y1Ll#~f`OBiH4
z!@20Do5J924;RiErO`+gq}TkFL5JaU{bYltCEyXNVNmuc^vt`^Egv<g2E6rww?}e~
z8qIT82Hv_-)8226zX;xj!P{!-oMS2&)el{jZ7kPLde4`Ps$0|~8@&A*-oEFImkV#j
z@7<9#X7ZHG_&k{BTHUp57`>Jv9=hDlgR_h${PZL{+fyx0q33(ysj2~R7XMm>uf23&
zB>j!mUg&wf)X9%|0rQ1x%oP5u;j2=eeKo3^uZnhOMiGmSiaGZ^czb<iAU!IJZe#{&
z#)cquPsE!$(W;^Fw&N(f9^hYg?QYjv=I_C8@H1CJGhQu3e|<xKNgFblJClz=hFLAV
z)q(i463GzTMrOvYr6Fkk$ZgweSLQl9dIGza)nRu+XS?#rTsyweuKG9a>^#NW)gSFM
zKEF(OdlufVBVX(`*Rpn8t;i=OTXjO$FijmmKIz~vO_@!e_$uyGXtioz4^t~R{A=@Q
zWiE9p^n0hw@V397Q_0<&ioWgCj1NwImq-s8-YyD=(}}lEow(*y9ctjj{!U$ow|(0?
zl>nd5S8}Qr4BTFb-svB)=-y(rml}2E&uBiEl6i$jx-8#+PzCtbk&N16@U|UuP3FVf
z;B9{KIQ@8^OuG=L<`;J=4Bobbx3hbZ{nL%Cy{56cN{{fXs(6gc$ErvcevVXfedryw
zrQUh>8~0j#93$8X(hA;IgSS)Q?OJ$ig13v|ZM!z)^1=5w&p16_$m6w8(_bf>n#cHt
zzGLNj_%6wQ+jokL)~C@L`z2bovN0+OZ>OL^TU|4Ty;|%Wc}HK8Y)wNNd9tly^(~LF
z2cD9f@OJcne$ErIioYMLw{5A{;cXMnGHb^>nejU11#ka;;N<@kXPIt1p5dGUI*_M3
zB2Fo=cgbaX+;C=zHC`Rw#VPj<JHcF;NmhwhJv-VUKK@;E>i7S6OAXm%JpJjJaF&_s
zuOAcCA|O#i;q7IA=A-Wtbn;e$!eDF&4Brk1tBj7<0(#stcs_gaeCF-qHQIw8m28v>
z@U~bx`uTXbw-;l-#=|HTzZb>XFG?5C+m*sxdtp+vR**$o^si{`<#{npjL{#ZI4^}T
zi?8s%Tw&Lx%qT9zX~;d!Ifi&;@tAAGb4Gx(2PdF~>L0Hi?Rgx&@oK`)wjJL7Mz5s`
z*`y8O?Yk6u#CMrlkT=`nB|B$s@H<<Iv*W!`HDAeGCLmNUYsn#K?NIo44)q-7KtIPG
zn%^C|GTwo3jXb>1AsTp<EVy#`A0OGZ`lm=0{V9U^Abxl>6+>QykOPL70N###V3RG#
zt`le;$Hm&!rWhGEe0<$_o8Gs#DY=$SZZ3E=8dznoZdEJ1oihThDpJ&{GV7W5CFAWG
z9Vo{z^hnWuDhY3QHK*^olz!<XZykfTXW?yccpD3EBh$=UmSmDoK3<DPW^JPWyxZ8U
z<J5q&>Y8<(`fpWLv!=u_w;X2D%j4W@+q)^TksJL-H)Xz{hCA%4DJxv%Gr&bItzBeF
zbJ316E@*wr>ew&vc2!w;URH5C%4+(BvbuM_tR}WEr$O*`GQ16|<D%OWUG&pu7fs7`
z)$bMD)N89j{+~_kMKkLIyltOvRxWk!g@^b*es))(V;=hW%tLqXl8t|py7Y>Na<+Tu
z{BRFdq>lZS*>l8cYMGWEstIpn%E4^D-@23J1x_M6p##0~#xOV6B!6atjtG<HSxm~r
z=V^eqA0HX@fS)<Nwwr>7!dZBm1#gp&8ORAS=xBE`ljs*O_AaM6f#sAxshs++EvJg>
zxeB`INDCJwb#hV9N-o;e+(p)AFu07H{w(jNaXsABGu>VD_IapC15bX>J=Nokmumm*
z_5WBp%kU`Au4|K!WGpjDCaxsJWo9C{E!?eWad$6L3dKutcc-|!JCx!ScXzkq{+)c!
z`{O<)gb*1f_j#?o*Iru#+Qz6$ffyA*f99B!LvQI>SE9B#m|lBZ@)?T8M$5_E==aEI
zl}5YHSwqiz3-d_>>CgWerQO4$G-*zhramRhzjY36W8ZrM=aUg*sE2o_AN`1W>ETGZ
zA7ZCFelaIK@QbG-<@bR-vt{U$!`mQuTMXXLo<*O0B}~pqeY_l=<O=>eS<qj<WMls4
zgq{6ecG*tZ_0>k5*vnsIzSvdP6V}4p%zX59)BRNsPS@(e^EVC6`xn`c!~M}p<UHo(
z+>+t1)C6*;vivomxIf+<fAt;VFAwhkoheP7xhek#@OIcXyi+FwRN`}h_M`{uI=sCF
zZ?j4TsXz7PSJWqp9u3my*Fid|V4dg%bK&jpc|*_`{QvX1sP>ZD20psicSBVg-p0dQ
z(L2c;LTw-3df;6S_#DRmzi^%F9j2Mo8cH3&AGa(*OZolT<zff#*&KR>Z+WLXv$}W-
zvS-h!3TwzzW6tbO^8~H6CF)rZ=9=-cRga4&Z!1opH^%EjCK)g33G5zD(0U%9XC<f$
zbI!S&Brxlqpq=;0vP#3}NLHJD9XV@T@tJe(KK(pF1)h=1vOYmy-^A$){<vAY$R5p}
zq$RDBG-V1q6IPI&dM8Qe_mdTl-aXKgnlrpD1aDWiBU|IQRJ9q9s^7^Yjb$F&_-m@}
zz+UHyRGD|DY0Q9h&2g~v>r$oyed*)E(Pn4SGEb=6=D{;8{K@dJ1^Q+&yzK^WAB`bX
zl8>941IC)N^cl{X>atUc8OJDi+Z)yz=#BN8MNZy=OcjE+o_zi-Co@&E06Rb9(Sp`Y
z)%r?~k2`;tH&ds{XI;y`X>&MJFXxc)-ZxV|uxwvPc+h~qD^9j;I<p&Te2pA@4FA3i
z|9%A-1tZ|?KzLh??;{J|ek#E};sEl2@MMoU!Jd(Q{EQpf5wa*<*U1XMGcH~I2a{nQ
zk*;P|c8+9E*Ccbgf=80!1#cfdC5si_-t^AUwU`Vv37@lShH~}hk<44;pBb9_j{S<w
zGgXT3u@Ss&3vWBGC(j+;TH);rcx!;Si~D9N-|#Hunwq7A!&$oWgSQ5DA9J~R=hB6i
zJhubrVZ$jSoI2=%u7g88-{w*lygl-a+U`U8=m*#-3cI$)q1B6_tvAxQJWr4HBYn?F
zc~xmk9yLP8u7y`WAA{Yne00}bs?sHwygKHRogQc0ZfXTQ$7y(pUX;w#YWAwed|^-M
zl63vX&%U=lIqD_T)dAklhP7u-yHt!^l~|bji;d?kgB@06f}c5;uD4If0Y*Ri@-wtg
z<GciKJvr0(q4#r~!v_a%P3ZVLoWm-2W)274IgkIVyJwge-<P3%qch0!q=(AYxHIP@
z=E$$-=6g9tem=gy_U1IT{V!Fa@b&~cwjWu#FLM9L>PyjoYm)V$RI(<M34GU)tXy16
z;>iH2pNyxNx8E}l`Ghm;^F)<_x4k%{7Gq|wK}w1`lP^3ZE=gm%vvcH0f?C~Wh92JT
z#PgUJ=DuS-Bd&N(9e72bj~SD;eeloVG5z+L{t|f)b;sc8O^nigk0^GT(+edhXIFoA
zYm?!=oj=z}jgn_W>aNyE9dATGlOANRNIbRUgVdun_bznmj|`s-7!#T;Q1i$fXh!Ys
z0(rJ=;O#hgTMXV_hdqD8+gtGFOsHK?sg-`a;K%+jKUF*Hr~j7w={CCVKr<LK*+-k>
zY-$*3V{XZ&jmNAyu+yr(2dwfrV%3~iRy{dw)4>8h_*#9`bcK(y`uM6qHb3U8{Im_;
zekhMtIvs$YAyCsw(wl>~4dJcjP7w3sL8>t~ST$#aX#chljrlW#dTWR}Zwyf=b0AIe
zbe-SAJmzM!`$qEpYT@x(fnHwgt2E98#gly1FxgiV(tI^6!<Sv+z8c3F;aELi?cV6C
ziJLh$boOOGg|EJ#OH1VOQ;%dnd2x*n_fzL_zM7ontF4=CI+<zHAb7hT-gbt!g14pN
z?ICy@Qr}ng-ubY9$X8eR+}@#niiNiU6YLs>j;>fDP(c^?`D%q~L3{E-;O%qw2o)Sk
z&lyj20esb8pGL^>HbSpH;P1*GsRd}QBs|$O@Q?>wj#Tq^>~m+<*VH3Q)#2?%c<V8V
zv(`rZXfN=8Uy9Q5K9N#zB-xM5N0pAy<U)AI3P-3e4u;2hB6O42hY!512yb7*+XX#y
z=;b&vhv98qu4d8x_=Bhyqh&{D_t(AGb~5ViT6xH>KL1izz6f`}+qL17T|M8~6?>BF
zJ@13lWBGT@8~N*W8|HW0($8(`FI%j?_6CuW5dxdznHhq&HQ{adR{m-TZwFWcG^khr
zUO}`lye+pG&K|=r^&mi@p@GVr5~$tORL*cFEKgseGk$?`$2pUE2Fn}XhSv#J$zj22
ze?3^kvV|xoz1)b_^b7ii=*G<uX8%I9whCVQF6`u86RP@X-%j_b_e6!MZp|=exx>`<
z1^V|&xB}p9d%Ob6c5!Bzz??YQDvQ1GmCq#q<#A3jDajgT9^etP%P*~ox<hZ>f+mfA
z%C4czI0cX^R-KH{srCeUkq0?3HbHlT5@dW3ugG)cnjMc*Re0+j8LO^s<1~-HP-;c$
zRq^DpMa3yOXPkCrk)_IvR{p~Y`gw1nZjvEcv3?TSQc3J7OH%#Y<ZHv*8SvH@-gad6
z-mNKG0Iho^9?qh3m$j#6&2_dLIXUpQ%qsLQyq(oQU8$ckbRJE*0H!@=9%5lchCDu@
z@BU+d1?<^)g?${)ng73uriHgN;qAn(=um5x=5NfD-{1`WRwG0GQ<<Y^pRPrv*$Fw8
znTr{j%7cCkK}VgwfF7k^SO(sD$HLl|^wd6NsHq$C!Fe)O2i|&>%T!8H@_x{cFIur@
z5^dX*$5l0%lPsJ`7$BMH3GCq}d+#~C-GbIUissA*Z->C!NyA}dDfDhQn)D<4BJQMX
z@!oXJfDu*Zz?13eGEF7Z9Nu<?w?DyKH*a<!y0c^CwM)4Vxpa~Yza8COD$&KI{Ctix
z@b(P6y#sF(;cdPF8QRXz+Y{dQgtva(*k3k0Q;FH}=iFik<M|9`l`>SA8u;jnS?n37
zPB0`(Z>CXaUqoFQjk_O>IRf66^vuO;J(ofQbLmwfbY2$TaQ-fl{_-HMxHq|^=X@-@
z4TZNCBj_<NV8_e_7<Yrd_pMx7dXwCI-p&VKYr*Y4M{>#ik6c;}XMfbJ%ky}=9d+^}
zaE~)VhZ;Qh705t=x81zS0e!^J2K#pHVYdc(vkPHrG`wAQ$Hks{7yUDr_Tam#0&h?8
zch|FczH)JXfwyk(cDRWwZ@&LL?l9iSzU+FuemL*A7tYWxIoNZxgMC@6$c+efsYbX<
z-cFY)EOluPdtAfvXXtl$TeTJYq03QUn3|?rc=qGBrz#Q7+Ni<!^LPf{hH!2)(I=^O
z7ha-ar@-6JY0Tx25qyp;zJe<fH7GPu%a`WV7i*MS{1d6&ts~TR3e0PVcV!6upzKjP
zvnY{_h9vg;B<aYt1ogPgIoXn+$0g(NT=P42I;Xzkjr<jF&6NMRXP#u<^B}$0PdRkp
zOb%5Xn?p|$s2jUQX)yjF-@Q@v0dr{F(nz%`6|RB(LNw}9uu_cpnjX+2J;Ls@O@Ufo
zn%yqB0(GZippJ15OR;btgSYwNZ8z?1V`l|u;Y04Rm(Vgt{Mqe-KXn5=xx@bWozZFV
z_6xk7eb-L|Cg24+=%bO1ZK_rij+U^gR+>%!gxc`j*%Wcqs-hjOa_73W(W-qG8(i^`
zGrx~YezfWBEgy~l<fED<UuKtmm0a1EdM1qW^wW3v_9yk}4Xf=k9k%QGb-R`{Bg5#c
zzg7nZ>cb&sMqW~rq^4MKe6ZYJ1uLdRu+9YsEA?8CvgY}##6zokytD8-V5aYHR;f_4
zIvdRDZ!@d#4YTgGw`k&1i)uIWR;s5}meN+$m}pf-7hko1WMk)&O|}2FDRR3_Yv<Y2
zt-g)kv$sOOdy^%@&wAOa{tK=8E77KyI`FrvO(j;@WY6?bjwe2fh0RCMxz7UqbTz_H
zhjRJJ`2g>38T9tWKz*zks@APTWe8v|tQ`+q2>$B0P)#Wisura~)$ONHwV4vC?~9ou
z-59C{ox)__8K&j?!&I2MbUn0KhtJ{a2TRvbb05<)LWg@X(}Y(nsa}L$UJp~=$6;F4
zF<eIcXcORV2)uQ3;R{NNP?-kk+<s`@Y0MFyi`3w+%qeqDtBAkuG;_-(BA5q4kDh&N
zS8;gj{)+$qqjp8?wQKrLyL@)rwP?LvNvrI-&0`06TkR$_kdJm1dT7_)+jhNr$;ZLu
z01sYE-qf4}{q;WBpB_B^D>U&AAAc=%;XCW?PX?a9UW&X(YRu>0?ZX-5QLPWqy0ZbQ
zNBv>!Pl3wYGEgqgfiGLr-yapE=)KfFyyzQHH|U60zFrTQJ2+ULBZIYL0^YYJ!JHNG
z^W6;AJ!6O(7YtQP>I&WyLzN73tKkuB25%R2<~c1LuCi~#^%C#Y`oR%uR+4khG3Lzh
z_zh0scXlzeTbsxMTM?%@2jVo)7_Wt$<5h@RyLAl`RnV9yZ+Ls_ORR2_J64Ceyw^*}
zZd(+qJ!@m-^(t0bapZy~#A!dg?F_fit|MnAgwK~0r!e^XI3iB3L#bK8+i+)`suYWp
zsbQQln2BCaovSr_HO75`w`7Z@3{F-wSvL<G;el?!%n`i(O1|7r@b&_{?Le<<W#v=_
zH%iqv*!$;Ssm$u)%c|p)V@kUE{?7fioJ)@T?4x4O)OoU-&unmN!%$|8`Z$#`$f<G@
zoho+0r6<%DfBlcFyDu*Gtfs4WSh`l_NSAXW8fX~Wq+Pl~kFifWGDEt}?&LC=>I##8
zI+LlBXutgMwp&hSG6s?3V_}!{gLL{R>|93woO;Dx$xQl*@V5PbXd!fLnSz-rz&yuC
z-j@b%AHds_@b-owQ^(w?<z8gocx8shz}s%z|M$ZB`h4wru=0;faCJ?(uEN-0&J_de
zrR#fox{T}!`I+zeukdvJV@a1a8#{?#xs>mqOTO^7AH2<Af8v-&>Dmo%|EfR^0=!*R
zD?>9H@_n|*P~~?S%DkGPn#VJ=c`bQf<MGC_dvnf)481y+p*`^SW(9oK@b<)?S=s<^
zFTvYiw$gW|eo*Em9x7_~v(0ddYcSkOF|q^hEA@qsS!(n!i+gI8Y<uyMql-uJ=N0g_
zKfL`>ug$!iO9!s-Z?B_!;jIU}eN>#9b!?XQglCaaN6-6Zrml`6=V5fF4)mhO{A;H2
zwq>7pkxbR+xor+_=iebOAKuoX2H@W@U53)^Yj#o#*y+;1qwEa5#Xgq%E}bLyz57>}
z_OUOmBFsJU!lm)NZ!WI~{{lR(f$8jpa_Q0>hei!yzv-XsH*M%tt8Y$um|Pkf;9`#;
z8EFo7vMzC<<=7#93cY$RRaO7xOjwG&E%0_7yd8q4!P}mqO81j=`4su&JL$E!lbKkI
z*LhhqZ3XgZg%2?`Sq-QS?;T6-Exne{)f08jJwZ=i;#GN$9%&h&OIC6deEA*T3pe5I
z2Rtje%n9sAO;Fx*3DQ7v?%SY0?ndb_nxiB>mNL)LF19FL!t+#*`M}!ncJMp;YIEta
zmCvC^Ptd(lkvh<oEF?T4pLzyr!Qw!*UrRpE-+}BZAcOW<p!UJrONY?2Lztt@M>c#a
zKF?QV(a#QG-ygG(Z337L^jCI&e|)+D1eB0p@HaI-9&_&YS6Tc(GcVY+BFL`I9)8+h
z&zD?RUzO<Kqah<~vNg2nQVpA`6oogjHcd9#^!hWJ)(zgCvoP~)(fFDcogZXTGkCk#
z;jL$I>P;sbesCLeLDZke+VJn%H1C*Aqwm_3{KclRTYXgZi!U>|zWT^xr7wQ!|C>Ml
zLVxPi{#yPN&nx=3?l@*J2l73ZgN^%*s=Lpq%(F)Acy3gAFOzmU%)CZT^1WnI?UyFz
zInA07Y|;9G-s(KmTX74#RS4dmYi}j%oVxfIAJv1mpI6(IF_p(5HdPp4)9HFPy=d=E
zjow=!1+D6L$x4o}w|wqfl-<i)H7wpLVD(mEo44j~@m2+RyQ3)DpQ}(2UuyWiI@sM;
zwzq!zEha$wriPHyikCK9sCL8KX7ILEY^ciR4OJI-+oLMJ<|iR~P?=odjv?y1Dnxgw
zk>2L->e|D&Plst(M3|<;gy~@dzUwq{lxl=&R@X54+F`nSIgGh>m^&j(vmfw2zi|B&
z5Uv8j_;X{!b$TRq(-P?2{t>D@Jwm(oMX1Wz2n9T0UYU91TyXI@diB~JyMFuIuHIbt
zkJ$D59_q(C;OsVP<m>FpLk*!P*SfWKZ9R|f{m-t1yLNrzjCA@Ler9+Z2ydrYxSrGZ
zh1+G`*tH#9Y%`;KQ~h<V27U6yc$;&Rt6Yw=Uw!(yW0}=m7oZj^*>f<3KJO2$`<5KY
z4(us%3sS$~L5kc&|Koj-CR=$8;Frn<Z}Y(0?C|yxysfb~SYvhttLXk<nV$vA`Hp$7
zufe)8GgN-?wj8`26dFc<JWNeC;zNH=1{Kdqv%V3UJ%DozHHSd_>*w&f8)71rgIVnx
z#o0$R<Hw9$4xRclhbr$N+Z2D|Uqz#}{aJ*{AH{oyzd1+aXs!7rnyl7nb%>79>YHRn
zPRIK;E2pv_%Bf=TwjNi~E9Ue5%c&R4^*x8ZUCqosTjTJB@>o1ho14XH{tPlf(ZdmM
z$v6FoXCoq>d~BF|CRsyYa4(8Vk$*mRTGdG*j|Hz4ysZUqC%{`9ylsQV-LWK9hC6BU
zhqqqiU20t1rF!?+wS3p9EAaLzy!{v6c7nIresk)=K=v`i+fj4a3AxhAt_G)W!`n^F
zD6Z%Sqx!p86z<Y`{=5p_?mn8XZ+7<h!qZt7)0t^zHuy*14-U6NbFQKP*ZLW~<A>>*
zct2fJUcf{?c1d~$^|TCqpOm3!c)KuyyuIAiX!B>P2pagC1MLfMO_BI&;ceH~_+hs3
zvCHAxOgM*UrdTERyTDstc<avBF1(z4Y0eby`th~l?PYlTnEi-rU~Od>QQySgXPB`7
z-rnDjj_d1Ekwz|^ZR}Fp-&~r+8REA|^e_20E#c5)*mAHSJ1qHpL1od?U8rUE;*qbd
zshlt1?G6~cn6pFoAm*6bW@+Oe%-FPHpVOc$=5(12g15Wx!cdqy0?z(mZ5UT3yuAZ&
zE5X}*@U`#(yxF^%KZdslFX3a`f#-^ADa>B79EMMzo=>e|X=8MBE<SG}-Yj@)pziSN
z@l4rvvae!PCOI7F@VuEin4U@9G*kBzsp<3F)`Pc&n3cH_Pu+ZEI{VMk*{hVUG&mZ8
zCcUs1ZQGiRrDB}BR;KBE8|Gq4q-hnt-4}uEN3FrGRQ5RcXzNf)7l&flyD`46LychV
zRxW1@?>kI(+O9PHdN55_Po$}|nU5tK+zqA{hPCC+u*d8u=RdrItEnLuX-9_trz9O{
zl1x2|y{fyC^d4`6tt9WuNK(tsiCR?<z7|L(=K}vf`STxPa>t=bno$OQ3vbV@k5(P#
zRbpC1DDiT*#?VLonYq6WWZMkP7NLI*MJUfd5lUD?X5fejwL=FlVFo|B9koBMEnLI9
zFbA`ToWatOS{)RrM{~mUWmkw=lnYVwTfs7X3sQAQkQPj(KG_M@!rK<`_WN$K-^ss=
ze;TNs2iWP-Hc-Cx0+l;2nl>en%=G|Wy@!{j5dFzK^xER-&!S5bzuEEo`>T-`UKt}j
zw@|Wi`MVnM*5q!N1?`zgy)-%6SN-AbPw=*PaU16Yt6p@o>SmZ#iBI6ocyB#v?5%}I
zEvf`>y-HgYvB08v?=8wq_SQXkdk5Y|p$VV<W+P+3rVqdHb~P9{$6Lo>TGdtFnip<W
z`5abV_`~}D_3BE4O(E#ylpHqwcgS0tvy3X**r>?vM*6cxc54{*dasfE6{Aw08Fk##
zq;<cWbmXK-Ic}P?;+{#Xt!Dh_7OjJ~$A)<;W`(zwR<-IOn$&%ck4De-Q7w4uSZvd#
z!8SSHTGecgRaeV<>qt#+a;3cG;NSEb%fJ8EBI|35`or6j@b;LuH~DSedbQD;`6XXw
zi+!~p-sa`8N&#Q3e&wTgSA6tOijVTv_EFnGK6<mwS2JVChxnB&;bEMQ#sq39wTSA+
z$nM7<ey4Yk)?B7n$^2F48$qhoCs?cKW$)P<tf&LQ`g$l>P2S)YqGx^nKjxxNgtC(~
zjLf4jO`=!1ab%dHnT6>CZ%e}4QDNvfL%6;U=YqLsM}+I+#Bd#-9Ii4`!vFsq_Nvw4
zY6(AGXxr#3b}gsQkhIyZP-+d`VeLa$J8mnrdU%_6xm}%R+Eu=Tot#Cxo=vkWZWEez
zpIs$3+SPTXUAf`yXI@tm;B75<JDJ*aWnOP%P5uf@^4ER*RdKWY)jo@UExc`2k9zVL
zdbz9dEH7s^Xd-+6_Hs6)9@7ur#(g0B;?E%FAGzL<%VMLpQHfl~c=qU8m`(FT&jtl+
zMIN4)RcPOH!D{|ASpM+#4!r$kRw(<PLuG=uRlLIVz!|2|JWqGvZ9Kfa$c*+-csmi^
zmSC=IXcioeA{!%yb4XHz#!(|E_a#C<GpE+!dxTcg2k4ST&Ec1Dm7{*Y^Up|q!mrqK
zcBH0K<MQ+)+w&+hdH>~*54=74on5}nW_=BbQU3X4HIK<DFOOJ-9ATC(TO65Wad<jn
zHN9o5{9DDU8T0h6Ua?yBnrznhaVi!XuQYg@e1VKPydWh5QWTRHZTmC(pO_UITQ^lj
z5}9wP1Y_atBzn4A7qF{nDLKu<9Exn=R2QpDm4h5=V|1u+b_c$MG^M_9D5EyJQ93xa
zAKvDLw}WP}n|B0U=I<s3Ih8cVfp?aj3l|-lHo?hEflIHsXFoecjT)|=Hlhb%-d%WG
z9p27AjK<{ck?6*<@HXWp{3?S->1SBj9-q>tbakV~7&I_Lwg1Ubxo}tuZ?EJdZ;!_>
z@OCh~HN)G|p46ZR;~i#aQfIVnmozeOZR8bX&(Q5p>56(!9sC0M(qwNvos7Qi3**bB
z%aWR|39$25cv}wMdcfOaUpY^_b1HjP&K^zBVD-_r@ODFImxjaJpV00h{F|?7>AH|R
zU0dPp=94aYt|o(M3z^k$I*a;fI=cBDKhLYic(-0sEA5Uh-@x~ImtAJ%@QP7`@4pMr
z6gA~j*V(=LAWP-{Li57S6@hrGnr3NDo-B3Apyr&Or4Mbg6u6BUB6RNG@OIl4_D&3C
zUaK2#|BMH^R+h$8hqKhJC&1f5ap>Y0{LgT{xtYxDE9?QKCs2oTPTRZbYRB__l3GIG
z2c84YBW=H=>k^OWSEW-w=RARq9m(0_CA{smn4K8>Y-ixDskl>XmZd4SF8j+1lD`6P
zTaf`#iJ7yKyHd$qOx1vY@iUD{QHQ_SJ(rcN&*aILAYV4AD1Fx1%%RUt(P{j}`PQar
zM4ePUTuV=b_YdOwfWKhN1iS_wX_~eKO^d(qB-yo#A10Fdg+HkXyKBe~c);GSb<~IU
zzRW3_uG;W47LNga7vDHd{~W8j^I|lHe2Md9DLLWoDCRf^91Pcymf;G<o9Yg4uYM2L
zW_VkBZ-g3b;rAHcMv!?ksTUb@8Tcld-<gSqIrSz~XRm}R@=T~)Cz&Vg8LB?tf;E~m
z{H%WLM(M~*O=ytZCX)w%hqK2UvIk1g13h8a<FkQ!@HkLc=@llz+gY{n5X0LqiOkeK
z575yNyrQ+?Y8L+{4o?exx{lljr(CitAG|Hh^@G2||FP?6BR_R)=%;<re(XK4Y5P;F
zW^T7CJ<h6CO|9~RNlSZLbp_r&hPUV8?SPivntaeg=7L4}k}O)girOx`t^UfQ(P+Yp
z7cAO)%A)jb7EPXL(ZEED!bh1k(9Nuzu+is?Nyp&iwtXi3x7DObmq|lIP3R?~9=<W^
zw>CyS4K|XqY1I1GM$Lw|UnUz>W{**aP8(JGrcu9on6wt&j^*=&^Er>-HE};S>u9oB
zYg?MxlZ7sYowl-89dBV(t6NrGqwk!PtNtRZ_O`NWYaO&N%r6UXQ$4&@*X*rt<-E1?
zr9~^?t@k<$erSsVoA9&sx9Cr!w{oC+Z@PTtSJ+o&0^n?<k6L`TX~hYf0z-T>2;SbP
z<)f{SeRP25G+S<bNa)DN==XQjR>v%2*5(Yo&s+g2Jc9qL<J4K{YrnigfAbzbBDnJ?
zlD#?w16A+|=c^XXPSM|9P7kv`yctQ)dvNtI4XMg(btN=z$1uGi>nfRA^a|?CmtTg-
zh?lz~x-P0!m~OTX)A*)bjl%Rx?l4W?OkU>{yK+9VtL-Vfs>9oS8{q3sY8<e3I@cEX
z8wPLB&b4b+Z}{2Xu0!zl->L9;nO)5m^0BaY0iWmIWIk>l^=5cm1Ku{`b@u9wU4Iz-
z)g%#pTg9LKc>cQa#9zn#19b6cyv!r1U#tqyuS)}D=!S3H!OSx~kVrh`r#lA9V^p9T
zb1r#ug#AIh)|XModG>RV_P=7M0le)<Z=q-uJ^dWPdK-?;g~_+66U?Qqu#X<@)fLRf
zk%f`yN%qt))ZOv7j~>t4`NQ<;a;R>M4aGATs`tZ!=sAYzXk>&Iz}t60WQaF~%hke_
zkRu%LPMC)BJL}|13des{Crn*_=Y3PdG=UlK)V*O!s~fIa)bg9}k5HXgQ3|bzF9Y5l
zZyv3p<Ti&-$*EZGDK)mmDh&VQ9UtcV$o6zHProV;e{L45cz8P&-nK^LTH$RWyjgAh
z$R@4C?xKlgl`@m`^YsKRVRz$oZ~9*T?24+Gs%-E!dz(~ULHB;i<j<|SSK^=gkDXE{
z@~0{dpR=1KMa|q&l(~d_tog~>9*RDWO4YUhxR*w9zbwo?jACi(P}!mDWJ~_K#-R`A
z9hx(Vd-6r<(KdQ@F!$+g_zGk97ERaDOVnoB1sZ*X`#3tXmp?vN>XNNX(gQ5T$HChK
zc)RHq`-1T(4epnr^r6&;|H#m&9vOPlGDE$n<C>}KcFp85E<>p>^w(A{wTyA8?{xf2
zxiYldM6MRRod!pm!Q1KZ_RE5FefYuKKAZsxrK^{n8uKofX5}SQ2*!5Y>(unCPOX3A
zRBC0|+K`$kynW17ANJnsM6Yo;*$0Qv@K?z{fVq|OQiF!KpU{8}VN!v=$@hBeQVP6%
z1#e$Bz{h+dL+3hVs_d3b{e2B>Zpczv4{FP+sVl?V#4}l%ehS~&EPT?3@PS3(35S)x
z<jkVP$!wG>i+vSYDm$NCkCj;}4R=kv?+Tx1G=8ka+hi!+;_*(VKE0!t9i7F_wk&q>
zW|57QsW)i+V?6(x_ftPO$#c6qUF&x8|FW6dKbpVo9L^cDICJpad>xst|L}{=;LNb{
zmP^^-t$n^r$2QX+Xp6or;nYfa`@BvX^}ICYj7?J_ye$E5H}6c<?kxQAuanh@oWki1
z$vccEli!xCOn7@Of3n;wC2IgR;0N?uo%5;xekZ$!-s!}{WFT_IuoE>~8?t+PB*^_y
zymGUj=FrqQT`5m*g}IwcPhz!x5E+SN`J5qBA`ad@Dj0*GCPokX#AptgZzp?kUWK=H
znn!9<aXg^mk?H|&zv2%r-aMQgt>M%p=^GUam5o2oO$%38dbpyBk?Ts<T!X-H<(m_x
zoM_h%jhKT?57y7+168a*peBsMN5nlj$6`CXIMJGo?Yi2Cnq4MbtVZ3fP=FlK_?~i*
z1H`>Cb^!Ogv3Bj?e&~xv2_0|OvT=A`&>|bo(Ib1|t1b6^RUciqd7}>*sy;GK@zIfb
zJ}N^G(nychdBdg?f7w*O8X7jxrm1bM`ufdVn|4{%0^X*<+iZvFX|AUxyU1HZ`g?0Z
zNpIyF=dJG_Et(0F`j@xJEy1G0>n*Cg$D;QyVC`Lt8pGOrn=INiAB{M~qN@QGb*f|5
z+Y)9CgRh;In>2WaNu>tC+)5@bN;7E^yfuwQzjiU|W^<#)!`lgYjY{I}e@7bi3%tz@
zZ!g2!?6-`nkH$Ul(x~8hCWXS=Gq>O_*WVY>yBADw)2!V!EV^CWqAsnis!-9JtO1Lb
z-m&No*9qQ!`57%7iC!-NWB<0O^eKZ}PJ<eq_hJXAmr6YI($f$F`K<=!``e(IyG%Mz
z##?uXnl<DU{bD%0>yu4Q;q5zk>r2i25zlSDL?4+3;A=kWqs9Jsv2yvUYIR>N811Vr
zOYk{c{mA$6Q}1qmn$h2nUE}CPdH~I~1ZV(WqHyZ0+30cJsuG}(mdrbkCUcU0cKpd8
z^@O*3(6(b>?ahi|y552M^Qtfn-WI0qm%`X7&KwN$${F>;^a5?yJ_X+{^Ql#{q0zoD
zO9OAK@>&^r)~<@2-5R4|O)#_DTDvZ8wd?0Cc)!rD*Wjvk7O%CgcKJ7<hYN2T!`q5W
zU^BcO1#c_B+Y9{Ly|ZB~uiXoLj%lZu;icAGlQUf@dbuAf(!YSW{~X0*<rkpC{7lpO
z1gJKz`Aq7{*V+e=lf+Iyc)QX;u4dstS<$)kR?`QkR}yk2PysIjl@H$j_7}M;^YFDZ
zpZj2QkV^iI@BU?w>cHED@OJgI5V>~`WxsML_e6H(!rN$gt6pI$*%r@oB(p+4YR!L=
zPt`9}i{Wjz_vFI7X7>S1KCq5@|6FpzCx>cTJw7HbOtp9ppQMM8GskS*n-ICOhib$<
zJX*U$6tX5nIgXQY`XYqfiwMoUME`t4l*;03+wGP^1DJg-uqB7SZ;RGnUzs^7l~dc8
zp>G?@%z=BX9>sA_nVnO+(qnbqgPtzDEevnp#Kvn=Lc9)LjZ@kuW|L0F>n!)P&z|T5
zX8QlAoTy>!QF!x_3+8sOltPUxRqfba^f&v%Zopd$S<eq!Cv#uJgEAR?GAmhzdF&3G
zL@odt#lD)X^=!%NStwPV-5i=%fjT0*^=5~2Yk2$XWb*J%Ib}G_?8O16D*XGueF}LW
zqk~7I6Q`k-_PONsfIcajvJN`)J1icjba|Ic*MmQ)Eu+C&!Q1YAn4$c^+W~0S&KU}8
z!oJ_4ye~RKCt*a_$u9LCL=QQIUUDQo#qI1Ugn#WXyVTJwU9zQXaD8|@Bwf4a<E4cu
zc6hr2UHaY!?-U=uydvHmc>7=%n*EAXO<$8oUCG7lkxLEgxl|V3E-K|xyYem#tL0LB
z9hVB%b;*V9oeSr;^>it6346uiZRt(0cA<+IeV6_i4cF^a&pw)=;CA#*x6*Hhx34^?
z{RWWP0TXk<(5Bn7wDMq<F3+Hs3vVycbB&73Qgl=n`y8_HR-j?CvUG^|8(?b8=1l!O
z6yJ7ZW}=E`DmEu`!`U(wx{5iau^B2pf-}mH3^n?h9m%iKwSijwL3n$3J^rw+_}HqT
z<%8M5&AFiv{}1;#b3ElT^5-{i_&MKkcHx==Z{Izoo<0}N&d;@@xl11mPA%z^CciCs
z?Cj(~yiL)*$0=lTu=|tTwD`#>YT(cHgSXUL_T}WlS1^-0@BJa{679ikP7AW$ODAb5
zGjdh&4?dk7uLwL3`|9&ry-Pmc1$L|BWoY;h`&XG|zc7t{3B3K0K^%#WeGYGPbY%X9
zb7wYw4~oLtPf0NvkeNe~6Ua}h5TzvYliUhKX)L_$6~G+i*l>OOnX@owVt;r$6yBEq
z9-?DzoWYNCF8@&rwg>4iKBhl4zvkQz18$R<2yZ(UVm|a7_t_)V=(u;ao8_mLaMKKL
z9X0HlSjet1PpQ%U<gX>A{P9*ZPlE@=V?Au;V@&<Irwy^oYnWXjqp16F4>c~p8#)2c
z3cP&;Z=b^3h49vYG57yzK8kPTqaJyE^x{1}p({4I!&|@nHnL-E)a>y+efDN4hPMWm
zu#h8V(ecL?GUL6KHPl<DfA{7%=&iBYy>%bno`JVp%UIOJpL(*NMUT<4`|nxUS7%Yl
zlNL3Hx9#Rzl&hab_iLKTl{M)XYQAaNP4w1`8nN1_$qS5XG~6iTON0D&!m~L>`M}$=
zFg793sCu^z>T|=OYsa}B8`!I8RORAERqqIcdm7cazfrvkm^6HYNhe|MffFX~3nubP
zO!__6tTqMA)E&%PQPixRB`vzZ_ZfM|qHdQhY7=bHC>K8`X9goR@vtdod^1M*ry1}h
zdTGIMFZN-2skpB}p?(I{j5nxrLj$vkUSvy|Rp1CZ`h&Mt7qIGi6Et)Yo1$0Rl;@RA
zY4G+$TOV19`Ece(?^a`WxuK7)b@U;}*GCWC{q)QK@wPR*ZThdDHs178AqT9gXV>fw
zKW2J=%+atj2TyrA{-KjAf|L{9ZnznwU*9sj><VKqM40xYUHkP2(`k6y`4sHf9H!%=
z!<4-N86jx0L6<_6AH6qmX{dr>N!f*=+B(Rt&b-!E?Pe}|sa<PE+L?{AE1lQSgnlq}
zg<VVGZLTqP?XSSa+0V0^T|Fk+^$o7pgtsGL?R?IB=SSl`p9HIU|800X4&Fw<+qcx2
zPtmg|Q^a2_r!jW~Z%2H`M^%+f40xLd-u^w1-SpK1)Gj$dy&}k=vXN6Ynwj0Pfr{Bk
z9t*SIMbWpniU;X7Gvf`$1gR6u%?ocY!dqW>`|<_zL-cq*O~<!RE#3fcBmWAO$MsM>
z@(9x&Y607MesaRw_GtWXoK^M?z`G1@r=Or-!@S(gMM3o8@aym6O!JAakw(5o2A{JW
zI|0`Q>!*+q)q}TnW65q#rRSbML^bn-s6ws~9e0H&lE=f3Lsg002-`+S={mDphqEzj
z#2$|HO*xb|Z?x1YT5onltK#u!)hAQi-HkoHsbrawdAc0l-sFCCjT&<}KC5cz;I((y
zBaH{^k88{g+2WNjB2JIkyPR1dUJcm8eC8(mnBi@{VtAD+q^Lzvdb~mG>im+dc4rf{
z{u|jj*O`UD=as!myjK3q9Me^D%*M0p6W+FG$KXYH8%mDGm#@?k;q8Oo4&7btkRQCg
z9pu#11?(*%FLDZ+@$cj8F*@tiw_xT;VB^WjWK3?PM|u>m47#ZUwb=Iu@I1$&ySTr9
zLSH%I?PCl1L7~ivpmlGoV`c|dp6yLN`U^GhV=kT8?b7xQ^o(J6<19Yscc+$bgXJe&
zY7K8&!P_u+dkEeZ7=%Z87_52dQnNDnzsT0A;qFus52v1bIn^!1se9v{>I!eeUN|*=
znp5A~IQ5n9yOYzU+QrbX3Gj(}dtW?1DfwM`RTT{kbEDvG)J&H;!=y|6T)v~=7GM8Q
zc-sr!Hl)@)V=l~{!Ccc-xa-Dzu@8OLX<6#D0PiupJx_nuRhk*A1<XDD&A#GYnVPvS
zQ!}V7KXhj4VkX%Wq4Z?o?NxZY4BpOzw^7CLLc`mo+2QnZdfpxJ7bo+ztQlHI4Zjw=
zeE@H-PsCsSmrJ|zxio~cMdEgH3_CiNk>u3SvQBNU=+v-}PBm=fQkV1WnBx0y3~#qQ
z;fygGEzi%j0p1?wIhzJ=pKnf83cT$BZ+F7moA9>eIrcp7NKyTP?0FuOtghXXb;y^Q
zoE!95@z1}#m`GM?qI}q&<;!fksU@@I>=}OqZ_V&_CcNDSZ_C2lmGHI^yxq8lS@tkq
zhr`&(YGLm61^GUss3W(H)o+QgH#SBU4@b-3%MP#OQA!=b{9}1$b9#_7SQ|g`F6IiR
zk~7gPTup0+%Ydf%Ons?uJ92L>Fq4mO(tT2pa!&}<WqiaJ>8)+v&-uF;a}X<hbq`Nc
zh1!0~7viU<Wqp}N^TjuAQ$Rke%Jj8rz$7%%1hmmIW>8Fa{hDIexzcvIQv;r}2CV^a
z-QaBzcspn)f5+P%b?rJcj9%ebe<hsr)AEAMzfQ2SkJ($ljzH7(w&+?13ppYdd@mNo
zy|ZXFTJ()Kob<9NWRF=L)|%DhnpyP<Srh}G8dSB&jz0V}6*g_K=oNafx;wSkd=^d0
zV^M5ji!Q;ij&2s6I||p}?KXJZ_=H8*offs1S=qwPGQ2aX?JJW!${Ll2dT|kWI|Sb5
zfw!IN8PyM6X@a*m))=&Tg+WJl8RYY?LDk{y@9=gGyls5RpeAWX4QOH1=uSrc^ovnV
z|BttQ;B7s4`zx%C;@Sgmd%@dQ@YWCBI^peR9;@FmspUK)=Vv2x%_jN*^oQvO-~898
zN=JF$3ZuRnjLd0z$?=DmYQ%f0ny;tIzxB}b5}xe6@zn0}UV4_(z|6T>#bB<*Wz~6j
z`&VI`#x1nzXYxSGG_~QuvEebn=VitZ?QX+kMi1MD79M2N?a{oS|KpMAK5F^eN0XNN
zX)wG^<o~&Ce!EV=+i{=miisx&Y9blOC-AnN=N$E%n)Ej=`UQ!wYT1q;Rk4JrNH?;v
z;cf3dc&p&82fWP*Z_7Oh)vejVn!@~QR#zUI2I~P{?;^ZjhQrFCJMF5AUVXyrXbw8{
zEw828eC_F_VC|3Zh0CWXKB$K9w7FfCeucTeQ$yzC46yje=NJQTyNzIGce-7BHo#(d
zyPI=j0eHJ7*dGn$uaP&Ysn@{|M@@bE20Zc9(`V5i`6&lG3&@>1OYP=VIKBffcGo8b
z=*s*6<)_wM?i)FfUSvKpXH_dWNQKb(Cng3da4_?1{C?F!^Jl@^m)`6UDio|Udx90~
z#EX_j?o@cF7NO~vl_D!;0+}hy?<OY)v&b}9WywlA#$22mXPGwilApg~&MA&uXKG2C
zXOb`W7yb5KfqDsVOTh4kugQuD4N)e%J!J{eS~J;_ANU;)3Q_NzA<71COQ5f>j^t~+
z57E@%P<3#H>PhZU9n2T1j(FaNgodkojd1l^6|T{B*mcGH!5d?gio@Hlhw(F$*;+3q
zR!hmmju;uMX;))a@I$QHvE!-Ms5pHXP9Dx7ywm66^vi}gRcgylu-eJ$(l=QZ%cp1?
zdoFAZ*mY4GKj$50jxNTj1$=J#I8N2v;x*)aypB*4+#ABK#}kP<)Hg|I*!B1yb!W?4
z{4&KI>eSJpbM$X#opq>~1s~N<4z-(xH+32Jba-3mh*Rn3oN8il$vMKsen6K-^+zYd
zrrgwGYs1@)uyz{v_dW2o6TBS(ZyUkeMez2++H?h=eLIBF1BL?!VYNpswCrVk)fdps
zaB>9x*O1=uGnZ56uF!wx>z5<DZx9~YPQx7<66n+$E(@2_>QrIa5dv?I!`m+Kc3Fy3
z<7}J}o;b)_Bkz)4^wZN0`EPJg7k2P_!H$^b4*J&YWTFQB0^a6@x5Lo#3l2L~nri{y
z$CIJx+$+qyz}r4`Gn7CNdhQ%*%6OMsltZ7wTN8EQCVMhf7`ATz7kzpPzxC-%?SP{#
z+u{j(P8}Nm)qKtvH|VFf;M{QYgNyk|&L;PmPogeA?;JhyGn_Br;hvf<UE_Odkiq;?
z92)$vLk$w0%E!-G>4QTRx6z9qhi7T5L%Sz9RC$s^@1}A#y5(Ts8d=+=o!U{^sTG5r
z+B}!%60JY$yi+m%IW>5eOLf|Ew)yIieF}c3v^05dN|g)V&VNUK8N6)_Z`Y!4bHLko
z1Mm~W+rIEN3f_)F*R~5!Qq~mmo2V79AP+G$hMhXhn^*f3uaN!mnqG{(I+x<q?sA;g
zGOO-GUTwsFGV&V4=)z30-N;Z~F(^j+ipQvW5xfn#V-(_KS2g)Q&ksfGwh#V8cw3!)
zH;wW~X-PG*2Cnft18;YrZI5(g=QX|Ae=3vVM@?!rXXespLzKHOIqS9QyN#z-Oi$|K
zzX594n;P08>W4$fDGKyelNLT2fmgR!JDYNFPhQu_sztf+_V%H#c-yM!`Fxbi%U7-5
zz~?x02s&f<SG#&G$0OPU#`d%;tuwxrwsz%j$vwIbp3rnWFb3{{XZ$o0-hP0${VI6t
z4ZOVsZ_mKn0Yxp;buHSLn>uf*MeczX_7YJWUT@aIK4zKWZQNe7{wZr#8k}lX+^pi2
z%(7QE>;0c*WuJ!*JY?1}yG5=Ti)LrHsMI~P{)D$<cj6abVU}&WS>p?`)G5O(A9#D}
zhDoKHd8sA3@$4-xHTveI^Cp9$!wk#^88m*OK~{KsbBaNe7aR1Ew~dDk`lFsf*=rco
zx21v4ZII`AgR<Q>sJ#u|!o_hZ=vnI0`B$5Cm#^{UACo-c?Zt5>W&LJSYBQ4(t5Rn+
znl!qfQOTW+dfwQmyCsb(lx)<+_6AjJ>7|JUy>v0&OD{sbwAbjR3(=lR@$uB<I8S!w
zd8+?Q4_!&{)QO9px({zB#~5_-w!2(q-Be|lyQ)t&$bY^;kJcOXi8I0XohCI5vniOm
zfqOA;m7&jF_XfJ(?yWoi-t-^wWJP-`?37i(HOTC$Yt!^D%wE}j6_ekW8mq4skMLFB
z3BDS>z*i$TF*p3#Piv~tb67?#_qbi}TG5}U7aC^{)2*(|`NG>q@b>7YFx?nM@3vu>
zw%%t~?Brlo`88O99fI{QyqyAX=XA2;TcURCp#LzC%w#xP2i~?BY*&?LcI_&Tk181+
zCz5|sn2)P&*V@K-m-)D+@b)0Qy$5H1gSTyXbimt-@U{)St-$xT^|4+3vg4NuBqQSm
zp5^O!oZ;=^Uzr(NkADi@2EIk_h6QLIHRjHNWGnOba(LSg_Quu1M}@zB^8(JNOX#)W
zt@;N~)z(vZtM1UtC1><Cygi2Iw?9Xpn&{D>hyUif11Ed`!>sx1KuyMbb!!PeHGB;#
zFEW2*573mIoCAFQ6%)z#k@&xP;KEn^<o?c&*D1e4eaY(>@2~jN^q#*4$nzwhm-<3(
zc-yNp^J&Z{#T^XRpYZlkV2C=w+g<rWG@&QjmwgHy;O$EvGF9RJ9e8^ZPkQG>@+IMI
z7};Z`C*Tu*5UFDDb{o9S2XFfvj#7W-8cLRm)>JYkd!CHe%A2vu4R7z@uWG<d(UMAW
z8V+x_lxD^#f3&ir2@2*S8;kp4o#O1|U6P>twc<7YUYxqzh|>mm`-b~o=EFGU+Q_~}
zTY`!fOw^FYi7G^1*3;{W+Oj553#TSh4<I|>??lxgPv8;v)a&WYKaNUbXBasG^z2TT
zWH)FX_AW1WD(;+9ha+4XfW|7=n*LxvmxADJ1$et}KRv>;E)_qG28OqV;O%xZe&3)B
zwK{{RnLcB`BI!y;+g^mX_m8;bhnG4I-adf~P2t=Qbm`6>>;cXqV;)xi;^9y@p4s(-
z*<T26&rKvZ7slL~4}Z2h=&Qow#}2JDIn_NFW+ywD!F6iO1Ba@^+vXP>x_63M7I^yr
z)&_F@0dHsY;`1GK=wVl<ZftRCD`$!IhnSV3c3zV2<!*hvMipEt<;l+#>e9_<ekKq6
zNQ?1p@pBAyX2^z*svdpl6I}Ox&(I*e%sqBy=;9!<qU&brdLia@(5*M|HgEJL%Y%P^
zf?Dq#&Ll@~x^xQ-Is)Flhqs^L?Rt2d8{U?Jw*gHYy4jyx!@>@A;Ve{g5RaoBs(`1>
z4sSQZ+kEi$9{fGDHBAS$k;k3DS*Qc&j$!-^Go6|QZ#VL@xIO>jk8|lutW%fnJ9Kz5
z`yk=%*9~~@;O)T=<bcE5n(%fHyq&i_MN509C;{Gn;XK!Xb6UH>c!qgxWWd{;%&+y!
zpP)9$>_COLb#um(Zyzha;p~6qtQa#kR?Emmd<biomW+`bwVvj98&V#_(Cty=hDGUj
zuE|59bfY`@v$djRY!s=Z4e<`|VTOmS3AgW&+E<&5!EBLwo)V!R<Cy3BOwA5$+wfeN
zS{4n}7cv_zPYY6Id>&IzkqHZLzs8b5!Mwviy_oqgf<JSkFS{?vK!mqp^!*lhx2Y7`
zXEeNRmTuLjURG6lWYxIlHlh}MRJ5g!PA>IP&nCX=73ZfB+!wdBW)5R7bF5|2$A^5G
zx%Aa3@tDRurI%gK(#*`>uxijCt9Hhj)Zm*@ZXfA$!r5x@)|S^IUp!0;V*kh6nE%ZD
zznFAku35>)&D<Bv8d24(EHq(Wc$*L2zWvEeO~=f<npq{bP#?aIp0!!D&>fu|1XKOY
z>bBLaH}lOh!rR~A?LMbjL)V#9zk-*BHu6#fczX)oCc@kAJQnpg=&xXd%9MmvwG8Ze
zM&ouy%eFEozO6wy;jOW<K_99a^j|@P#+VI?H5#<eU{F7<{4l#9ybT!-Q>U3U65bj<
z8#TS6QFBii*oR`!uEGYrEC9C)(TC+RU#da3;|v-NGZ)<T;%j+nQ64Y$L{mRDd+A9@
zPhE}i#2e|UiyofZ_0&W2xwe+{(6jw+YCF$OCx3TSf!c0r7VoAV*?6C?n_k7bGe_&L
zj`Q6W@YY@1nq}Agt_Gc<@4Ly%sQfpK+Wo?$f6?9E%odq)cxz-Kyk~X2)o-=8F5UFj
zVvAMlL#$epNDsXNzG^Z>UQe*8ZDAi(sPCuE1>j_NyMm|(zs39eY@EOTyyh>znq)RT
z3)D4gDGLV#sR(_^liP!p6OVK1NW8o7Hl#hCUU-`g-cIB7vz_zV=3un!AwTV)R$LyQ
z<}ZhqMcX=@c5Mr>>qs~no3FVX&S(1>X4arqTpJ$4>gx^Y?V@{^P;0)^#;zvxL`uNh
zczE0K5f{34y1&2bJf|PL%C32p?ChJuYt@2Y?@B(doWEK;Vm{3mzQS88ynW-zz60vY
z6&(TcZxEpU=-e9U+%ucV*xdI&-j0N~J>czKcsmx}R{Mup;)VPy*ZlPLG~S0Gf3>0S
z(w?91Lm6u1oEKVhcG$$(p%!{QkTYq6ru1LlQX5YQ&_{kRKcDy0r?-AO4sXrywj8{j
zJdzy_Q~WiFdeM?HKWYnT@ccoVfsT$157L|nb}A%OhmI%fit9N!9HCwz8r3F5HOOi?
z&?iKR>)0Xa0dw(mIDZS(^;e-f2YV+M3DbX7>9ymV&YI6o1$f(cbEIO)m~2kQN!M27
zak$gBdLOF?@YaL*AzyfVj@sO8vdBJ>{rruc(l7ckCt;3ZcAtCUt_0<Rw>{u(NqAcf
z-j0H|-n`vx0681fmZvr*4{Ieg<wIl%P!l}NdE*`4_1%ZqbM=7xVAq`5e<i1S+1M%F
zJ66$u(qmiAo^9rZ-v=b<gfU6uJF-h<4%w1<@chEtR(<i}j3l=R-VTAcE8uMhG~?t5
z_OR?jpDjaQa&KS9yh(@bX=?d4Roh?U>3Wo^dOm4NDCS^353P!>ECp`^t~=FZ8M?KP
zQ{mLL-RUit>y)DDb5rzNutWdC+o}l;?TNyRJq2#U(O&Sj8v3>bylovtUD?g4f$%mK
zUHgX5u@m0LZ(?qxq(f}O)91Qm@AgDHA3(c*b7(4TZ&HgnW#*}_uXpMQ>|Hm5Uhp@k
z=6JX?o5$4`of^Pd<2m1BSVy$^d-w%6J3U0_^Y&Wm-X4WA)OuTnLg8&fEq1e7$j&fk
z>NPVu+u&_wKE52<_8%VS+~Dnv^i-=;Q!mFO=Z@6{*%`<4njhXe8atS&<$JrDqCx+p
zsNgvESulTH3U6GiM=3hRJlH_4Aw^U5`j2GU+)~uWi~Sa+6n(R@!;1dvdS+fj$Z9$m
z2)|pplqZCFYwE}`k!fmFoaaA_3~pPhraUFL`(cV4m#HV8O3{&pWOi>z*1%E8n%6H`
zWvyfs!`njae0{%xy2`RdvOy9xn>lsEfmqqc$7txd7*)E+PL$M~3hqp;rcA8N5%gvQ
zqtuz#a@*cf@`JZ?;q8?HQ3|fhEPQWf0L+ov{}c0g^&&N=J~Q#PBGsTedHRdU^o6&>
znGv{<I|8p<xE3<|I0;?Ryi1rCpQR^D?d}-7-ALc*-HAYrAn)ni2eM+1;0yT;jaw9t
z2wLT`51B|cd{jNzN3R+&|MJ<YE<eFW7hYNJwcY1i^$NYS2Hu7?Am3=Tk3M(v(W+QK
z&7EjhVeTVi=KASf89$wcx2xv(X!c^8T7^*Ct!UQ2`|uj|FzU)~gBJbcrMlORn(@da
zM*)lC;q9gjyu<jMhQr&Fc$sYQcG^sQz!QyRks8SbK(E#|spoTQzGIAPv(u<~Z;i^8
zWzzFbCJpIf)?IY)Xp2STs3ot5w-@QpUf4=+_P$Yb`WiKMrh%HVmsa)g!r$(td4XP9
ze#ujxcX{f{UQcza>ZQy!)NI>&sUlqb<FS`EzVT9}=CBdohQZrn6%Be)%AkMY?R0o+
zg11rdwh^o?#1#i`y{DTr2t7M-q)98wnAAD9k(oz>9#f;<HqxLfu=TLjp!Xf%eVCVG
zK6}b_%2WA^daCt&5A_)0p*bx*<XzK4Z;E)x9N?k-l|ArRd+2R;PyGgK3vmrD?xC9S
z_9VQu4|LPInr><u>Za09v#BP$jd8eZ?qGMFpY2Yrt-I3xcGtvX?%HX}u2JdP6?5KG
zP5Hm*(U@Mf@Lp5Xp7O}3Nj*%O&HsVnC_XH+RTJUuSY89WOIj84n^ln`t(rZ;svT3A
z$zql{fHPWBw2yZDc<mIVM?8}pt(*9<K9e(9$xjRa^3!mbW*y)!%T@L_;CDaV5I+$-
zZit_~*J65<u;>|lv7Gi(I=roV$4}XP@!L7jv&HNZ=e9TgXw@LNo6oKr6{#h|*?I8x
zV=XiwyscfIEGn2g8s?5{#oL?%yRXNqe2iN7L-g(|yv%U7`aQc|tbo<!$Y(AOr>RSC
z=Fb!1ZFX4sqdxvL0Dl;~t>+P-STt^PBRlEQy|v)sq0|7Cq`#XzgMC9~0~8+^sP~*H
z0`Qk*2G})YvyTQ|CTk_hM;)6o)BVa<vy1wvY$HFJn)|UY&rkEN`zb7jp9kJ<E6BWY
zTWUpb@LeB8*Pis#&9m%%+UKVhQ~YF3fV;de&lL1M{)+B}ITOI!#_;xjUiM?<rgj~M
zPwTV4(yz0#;XcfoM*hb4V0?(oEqw~sm*OEx9}vP`fe@|0UlmR6r1@WJ&FqfekEiS5
zybyI=Kz`a5c8b(wmv0v5_TX?W`iVJ0?m2<*_AxW~Bk^Wh>ynjD{kh_SNM?(o^l4%a
z?H(GfM)39yylsmI`Y61uw=;p)aJ(+lJ2S!9BzW7GtgKuM<K)$kc?;&0j0Sf7lB1P$
zFp*taWX!IjU%{N>RkC64Gpk+G9;s>Rk?OuMQeD{bRgK)yjRT@pD^HA8RfyHr<#9R{
z!W>gTtU9ek!)&8Aor(?|?qV;MLk+*Phwyot=A%CgQsW(j{&er{&?R`A0dF7Re|`BT
zRpp<v!w}wvKTg$++zx$#w}aqq4r<DOq0634VYlz^P6>TFBql{Zcw$#)G9Q(Ztl-jQ
z>XHj`mCWjIGaMQUZ&$<htQ*w6VdvIBvOM5zMR@yVkAt7n!LCh*zP56xX>o^g7jeiF
z-Zp`^|Mfr{Uv#Lq)2T(}(SFP$?}xX=;jJ(1buD%%s0*Kyk6*R`O?|>CKYG9Yio!@d
zN)HdQd-xP{y=d09@b*hVbUeJZ*F)E)P-o80d~9iU-k@uLMb8ey_q_2hmp)E|)m>aV
z&>X&2!z&GMQ{in{c$+ql=c1KULAS_ZI-aVGkW{sTx7SC|o2NgYG8w(e-i*K7C6g<F
zXSi#!iqccP`ZZoR-4c|TN_J?z1kI=3k<BYvsd$JTcIwonnLQ4o9?>&RRV(rQ7fI!Q
zo~p9&b`rdehqou;?f!Mtl?SA#7`#0*0#>7Mk9za?N0OH1OVW+riF(m4Ugy}o?NcYG
zzE{hsMX)wcl_(AQh{x$gwB~<g4_Rc4OhLRh=SS=Q9de!Etp~iF1aDWt+p!Vk|9_$<
zdn;V~F5!bZ9WDp8<5|NZsP{(b59SzGj}KQB>d5`z?I5xV_H_wnzdXNtc#N;NW$#HT
ze3tnFw5v{l9`9%8POCtL^E=t&pk2Yu{1lUodt6^%MH_wPHIytSi;p^Vu&F1!eN)w{
zyvbI5zTmCO@OI#@<S~q-&&NIX$rPKquCnR<CYzqY)eA<O+8ki6Wfyg4^zhw97G<kv
zQD~$^N%(Rz?ie+nnVL774Z8lfm)^rX_ed{=|Kz1fUkw^n(5OA!^WB>>cQe93R*^vi
z1~Fp;Z^QpGvV)RY*DCZi%Tcq1+r@SmnPWEU&zeSkPBN=jUbB|<HM8fF`JeA*-S389
z_n}GVZPa$V8TlPFD9z2FrDyq?lf9G#Z~K1plykqQdc)g^@OIX}p6d79Q;olS;v@4?
zb$HtkKkyecXCJs*9G*Tc%J+{ZeP;Qg5$Wf`S1)*5=Lc)K?EE>^k6LalT9tlp6~1=v
z6g<uFw$Eu8JK3Pqu(m#*cV;^;Z41V83~#%i^i<Gt>cbU0)Mj0FeM`%(Qcv8~zD;&T
z9M7&UO+0jbwuc^_bl3WP?rItAt`FbbRCKSK{+;5c69e2d4&KJY+f|RUF+1iaYpS~j
zB)aR7mzx%(xoL>QP5Z06sVuyWwr7{!Kf8LC%C35Ucxd(!PdzQn*(0x~HW%aP=lfi`
zh<W3hCN<$)^vejdK4y8V@oaCMDhO|fFw+cgFY#LVQ4@)Uw~u3e^m4pS6LQ+L>OGw1
z8uQUAb3U7PFw^x%N2`u4$J;j8Tk*^sPoSQAaJs)-SJ|5fGq1mZFT3FjoOy7|PsO~c
z8N=9LENED2By~OQY7Wn8=HYF;1Iaa+@qxFm{*Skn;B7LzeF=LT|ENJvLhHiYkLcS$
z5AZ3!glqTd0bjz;Oue~9S-XCMw`1V#hq?58Bhcsj$!UhS7vb%T&;X6g9-!N3+(z(r
z4Yl9dkpY^YGeG^5*jb+*pe>C9^ptbNZq5h2m-wkJK8QyfeB=giXT|yGkLxzIK1IID
zJR5sa@Qc^O2T{SMKY32hzO<_6II9{yvC#{&X`9nWT@K?_JK?AHoI8wez8dv2eRk^m
zH~4pE&cU~vQg<3b-)AeDxDWkYywFzmDF6GOS@&Y-cWMjy4l(2PyT5vN^~cM>EHZU~
zgYE2^^qCsOk`Rq`3)OJA+prnEVEW9zl93k8IVhAzC!g=02ltTX!MqNG<%2dp-<F(k
z`T`Xmg=;Wp{au{#N7snds>+d^nIcq*oY&5!qV#$c*&W-N|6=CrANDS$1msX7{!L}r
zIsOlJRs8}(n}ll!{*w22ux%&9<oSxdnDowq_atin1oGVb;=|;=_IN?Ka==@`+h=iT
zjhqqM{+JAv$<aDIJ4UaLuydIV+Qt9mR2R=!eesV~tqk@$E?~x~VS=Lb;d4e?_5UwT
z;SbZuen`{n@$`VVI%K_{rlmh>m&}{2fw$-3trgxbg|{}YpRc7VsdS3+W~Qhr*&cn{
zv8SOi*+lcwwEc`joqhSdd6Lx}-rh=2R!1jUAQpT)3lnr>B|cbq+YsJ9zwS^+czep9
zn(})G`-JIh4o=fAbJ@L?mP*Yrg<deevWa+bw~$ATZv5<%iVrnasU6t6bTw69smnI`
zfDRw&P&Yp3H5j|8ESz;Y^lDI=it}gF0eZ)Lja+u8e&cEzKyNkLseaR(GVjL^mXN7C
z<=F9n-z&EfKGR#hJsckz_2G|b>#)~pGG9(r-?QX!?&WLKL%l~`_w$ET<tdz|As6YT
zcXsOWO)~vICo4DC+0FRc9{$*gjs~5Izr1&{cE=^^t(7y=?F4#j37Sg==~H-H1>UxX
zxBVA0Pv*(|SX{C?9AoY}8LyNdyoa|xX35?cO4YNdRF#CcTW+PO=}7WbUMFeCh9tFG
zOa2p@c`&>^32!&R+iB?AQSi23|3tkENL1?8oaC72&@A?(JjsaCt;$gf8y2N$tH>Q5
z7bUL-XjwAup0tZpSL(zTc>8W?l&q_HKfL|M{M~ah^v=WEy70EuiEu?PBZGq1@8)UY
zYR~I(F1+11hQ2K|m_Mqs_mqB4z?2}FyOZmLcI^aj+vW{Wh0Fli=|Q#PUJ{PirOHv}
z9WRn4+|*Ce1IS7+_{s;~UNcft=6AdGYcdy@6a6EY+`;4CN`$wA;O!T98_Yen2sP({
zrz{#Wfg12Yi+<j0)~;Z)BDgoVA81l7dXnSOaYNu+z<w|7*yTmerI%*I+Q5xo_`JNd
z4{i7Ai6>cfp30iwsZwz7_I1zy$I@AbM|pN#dy<jZM23X8D^MsDFO+NHP~4@sQ;It+
z?(Xg`#ihk5?oy;s+@)CYgb?}8zR&mLJ|+nvgk<JEuf6u#YwL$-5Bb5{D0tg(f+w#Z
z>f*Ydx(shC;BOk)41ZBMPc2IFRQZzBwvWBwnV0g1d+T5HW%vHxx{~azTQ9wo^ACJc
z@OemHbYFNI#P{<Pyj@;_@8ykM%a7O<&%gWQK|8s5cKKa^HPo8L;q8Vej{oE0iMtLZ
z!`q+WX(>3{*xy6n^Ed$B?u50^`FJHR53X0d9{_JJFt3w0%0ttV>F>Vs(5!zwRPeS#
zzxG5E{>iSuMs`{Lt7kpoJ9X>vBR1XpotiburcE&RZa%A?K6GcM%3Z&Ach_$%sZ;B@
zE31;bVv4%U4sW++xM}@1H*&|^H20xJ=D3B~1B>1~vZ(g4Y?G@?w%OAt+l(%oZGNbh
zZI)HZHhpVno4}#jX3}yV54-FAU+$_u%&IDZb~IbNF2dUfvTI~Vy9&&=Yy4unVz$}!
zb6dO0PqE9ZHGVQYT@x%`+SSubQ!aVw;{z`Z3qrRq?4x5leeghg>u5u7Z5r#X&da>D
zq@)jd+&=P}kB;#SFWg`}=X0If^VF&5Z=A~e!6_^3Tk?{A<vudAx4Tq={$<)@m##l|
zDc}{)Q+ma7GRgLI<&d>d4jnH-JxF~TUW3{cUHE-9W_EcT!BvWn?T5KHdNUVvfc`SP
z%?D>guhRGB?{>`1pTp^TaQhEx=da<dAHCjX`<Mq$$B!L{r+!|bcGA}!$Lr-XHSV|Y
z_Ah326QYCIwHKsG@OC$B?#P*;Lrs6>`i&X4t^SJQvccPtOYz;0Cr^d5#hkW2>TndF
z+!k*wUgE8}lX#9s(F5`K*2`(08a3LX%^MwBJKaN(p-vS|3{b&50h&_7PZOHqJAdJ&
z^mH#h=JS6o<c)UW!%h=ld2#kRzk<3q2LDzrc2}0B4_uJl$msf+1DJE~fv#VVzHr?f
zS_;n()OD#Np0Jn^In{bTx<VYYP1NQ`L&(@BYrAz4J2rCX)I{c=1~(?RW>ijXUzAhT
z=@-9zKsM`Ja%|Fbs*06dbovAZ*_)I}22SWUGODYQAG#(`N3+88gv`o6TZQWZ8CssN
z@oR3SPJp*+7op$b?a6u(x(;tEoo9Cje7^pU+TlTjZoEf#*v#*Yd)nxw><exhsl-N+
znzNf7A^5<P%G0~<7NI(0$oiffNxoUEytCri{ShxGxml&*ZMEvm6_bzEc4eaSkq7;j
z?AlR(u)8oYH}lZB^(lrt>5oym-ZNTP@WJ$G7^_2*VpZXGtQN8(`iHPM_E^N}*U@-q
zHpJ=V-Z<@MzjMih@mfNk+WAMKO83aEc=kQa-Gd)&d9rpbN=8FsH}SG$HJOsgZUgGM
zW+}2v%cad^@5bEAt*Xy+D=ImUc6TA?1Ky5HBoBld@#GA47IVFXx0h~{9RhE2Qs-S{
zZu#=GuT*VoUX}bn<`Q12rN1+e3}3HshWO)UUd@EJgZkuEQm?$Swa=@Gb@J*$QGDIs
z<kj~4c{P%&`hV9~c~$HYS<LWudv-oO=j$v%8}_Wjx#Dl;YEJV#CZXwD*?WB@S+nNz
z|N5CZ)pG2$$djVV%-R(Gm`m#?vJ=NUml_1*(*AL|@Ehlne|9eAewIr^Q<78^-+EuX
zaCu85>E;>g&1Xp(25+}sNRo?d_KajLiOsDI@V0vxSrzqiYpXwbTUlR`tpaPW<Yx9M
zMLjp6b!_A7SI_l-Z#P??qW7C~DH7h+?3AL5%#J^K9It89;<c7swc4|zm9H)N{#-jc
zvlE4k*thq{W;++9!T9qdQ}H-`;(Rz3PaxSaZ7N0S37XB{XvXVHvd`>Ka^1E^YP2;{
zU%!k{i+jw}{!K0|yj|-RA?qQs_{NgST7ey4-=G6cqenX>M5o(@XgcT9m3SS4d*;-B
z@?k5%+fqgGHp1Kg^6Q4?!Siu8P>;6fz;6_wtDyn<_z_+BlD`hN@K?!Xf6b=GUHimW
zVV=JFn;u>cbj2_8$s(*whTu~?HY0sBin-VF{h1f-?4wr=d^C6^ep`6k`8o5Z3-JYU
zUmgH&+g_qSx*cx)?oi{s4qZOv(9RvOw-3DgkGIe53Wc|art{d!u0eI|+-L3jxYVJC
zn;h!Ttm#KGCNgR;s{?Nb@b-f0o~l+B*23F4CA{>B`@K&#eMop4P{3Q+XwSQo;6L9>
z&%9pR(G(vNHMHw8J=R}fC11Bqc{FAI?Pq)JI=$bncYna!mUb1MYS)?VcD(ft+44Fx
z<G4dp;H?$j2E*5y=*MMX>L|O1%5rt&?M-m@EPqyr_y2PD(1c$c>XF}}X9*77jCSao
zV24ig^*nCdwRMnPmGamX7D_F-AFX(<O%Z+J?I$Ze3wQl}#-02Lcg4;@{~hG6lo9Ua
z?ecajca3b|t{L#w0&hzNxvMR_jT_^piPW@3;O&HC78TuR(Vlx2Jnz|N9=v_kFxyms
zw;5Hl%~*Kb4BkE+l5LJH$~F^rWt&1T`J91nO8)4kGt=GK`RA_ti_o8!x+`w8yUyBd
z8vn0NNp0*p-rBCC{i($_+f~)+&|I|lQJf(T@iWMui@F{i!WHJNZwq>>J-odznRCx)
zI80CBzyKf3SjvpkD<74Ew^!#lwdjdcv*4{8y#1?u4n@2`N8aty+08D^xQ3<*Z<GFU
zX)x@)^oV^9=*?5`5cLlw#}RIhrpEmRzHa1t0dMp2*omtd*V7-kZ==yV;c7d+ZZE#p
z-c*<B^L2N^>>xDUDKNVMylqFHxW~L4+Ps&1Dr&&!{_NKsi<fFW`|A(0-w)=V$A{hb
zW}vc}Au4==Jwvee41YGiA^qU8{#piOYrxufTi63R&0p~qeN-&mi@omj$kAfrJ;)xz
zw~)*kWFfq5if1p&&0GJB_EhoV4)V<$n(&fyU_MXz9Ye$E%^BvimwH@bu8imL?ldnQ
z_!iGwUou^$`KZqfr(Q2&CXOp+Bbo!X{I$-3`s*mO)AV_^-{Y(t>(b){muA(%gT6aJ
z&-VuCTqNg>E@;l*1!?>R@+_YRDV_e(>L4=D;r8Ujocg0cPHp`@r@B-p<D^zjJPSFs
z`Df1W2g#Gh?^P1s4!NIGd7I!hTTU&#B~Za7+1=F@ogCh_=)mmKu@JIB@KDzY)%VOK
zHf@hbn{1BW`^f5e%k0DxX0^}Z*;*g2W#ob$@C?_qxL|q<<YC{YA2E$x#pnr!b53XU
zg*yI0TGNV|f4nD+VSc+R>=03;yfUKb6GtnxVT=Y|ic!JUvHFxhPNT}jsa*9qb?Y6c
zpb_ky9LkKbi)>7?YOAr2BR{+yb{@U)hgexg#<EA6ta*5QlMJq9mE#mUGERHgiFg3s
z9;TL_mLpz)57-m^1AgrGxzRAF--=PcT_hiYoa@e6WOgw#6%B8vd|;0GTe!Xt?vzN;
zgbpd1J0e9_MyKd;YO;piPF7@ciU$6kf`&z1|0YE%;Y>W7`3TRZh3A$T2!E>N)i0Ow
zlHW^>OFi0a%l~*g`)ywBhPQ2plXuuLuZ}j&tJRhBYHnHT;39eT9lRX}XDf4+<Qfid
zk8Fai3+b)mJ?b_im+Y75uRlvs&ATaj);~pUqf*qG+O*yhcCnJl{eD2Q#(E{I!mT7Z
zN+)SYkt98Nny6y<+-`7|N&Shru$Ih)4NO%0Y-U-HCn^Ts+TrbKdZ^QVlC+XpZ6A8{
z4o8x%@_3E>l<&Qg6djsdwc%}O?%eE2&aH3B{_IouE9Odb>+M#47Tx(dwMfwqb=fz$
zAz5)blgJ23ROV6go!GJ6m*2DRu^63Tt}UM5{j!DBudSJRyFxZQ+5f#Rv2W#0l$`7*
zE4~lU)BGqEfU`5k;CJG2N=14T4I<?TiPUd_oasCxHTeVcv*;(57ZIulZ~wUxp*8)&
z$gT@Pk0-mSKe_+x1IUZFqFK8jm99z#>?d?Bw4^KN0u{|!d>3A*67V(x-X5pd)RX+6
zL!8Z9!rRU8_Guk5iNXR@7v6rjOrC#Be>vdoe|{-DeY-a4P941M)RJTPSK;mXvQF9l
z^<m$ckMb4p(YPAkT87`Z&PnE7THp`b0Sg<_Cxwl(hI!D_$M<u=q3!UtB)qN0<DO9t
zRVwRX?+EoHy!{sDtwl#2IN7eE_3Vo9wv%az=hi}BajZi>|LIVls`N}b`}cnAp}(W3
zYrmnsD&VOjiJpo~^U&*K=+E0cRQq=ijW2`O_aZfLqPOl{r!H`L>bjGR<B!x1Td6tW
z?WcmUvY~?=!FGIr)VCw~w~N|Ute0I|_&!d++t3WVicE262i*S}-fn#dJFWC;sgc8N
z)UjM2xPoErM6MI?_BHRXfVbn|Z8La#8{S@qw>EhDPa1r^Wmnmmc1;SQU+iXA;&z+1
zHL)qEpjC}ytjhV!U6<kQ2zc8D?sXqX|8+PY!}GMUjl25Qch{OS?)?AV)ydyo>)*R+
za#J_mJ!z5eVT%eMwa5-{*BgtDf6g{mc$?ZF+Y~95Z6Xe3nXYiRA-pX;INRJF&Bu0T
zo6U!^O+t$-v$ai@DfN4nIdU}1yeUZSyTo0uwz#V%HE`$RRvlYl)tlB<-RflL%xPC;
zcsqT+LrWv^Si^`pVaz=ZL1WnErHgpdzJj+QQ@phY-cAc+2D%zKBb?1D!P~+0o!S9!
z&)j3yIM7FV=+$>KIdtKvOZF`;#ZPlFAA&ZEXJ|OrV00KSc)R)o9RAm(g6PfeGuZXJ
z(WMGUTxv-#^gLXSwB*ozUuJH<g_-nrPr}zHH(Z*2)1?bs0cf=DJJEAqyOay&PA|;7
zT4^$>@K>#y0>kk(*V)S4xmTdBQ@1{Xx1;}HRu_-bnX%-({0_VD^S4{%(xCZ#pU2?>
zY^z?5y^r1e^g{@<<M{7eaOS8}n{#Ib-h!i^TG@m?$rv<`rw(<$?y0V@v$hkzAG~#r
z@>HpQcC~`H1JFc%@4##|XGu3SoTv9Z6?WN6x8QA;4_?|^%u9LTt+lt0ym_3w(5Yi<
z$t)$aEPfigJ6d#^ss0+<(O>Ti`78GsU(LRVK805GEq+B?e)eQN_ScGe0Xj%exot6*
zexxp)S)2V9on2Zo!KJzjTx!1DrDV=L6Zko|r<b0O-mx2cxq0i7@I<btLXUk9+IVNO
z-dg$UNjpCkDN3Kd5I>i^E|nrPq~!D<RVEu`P;9V9M6*wL{(l)d^oHpddyfm1uOB;t
z$>Kgz9qyvb6@E+xmn&FfS2I^g&--<Ar)o1>pU$~{-3X`B)&{6un;>06=dZ==^J-?D
z%H9prsQcksl0g3S_(+{N5h*)6%I4IHk~f;+%F|J5e32b4^y&t-jMl_A^xV9n^aefr
z$4B9MgJ$^4JL(tqWxei(u6;H_4XB?V_lVHq?h#tRY^87M2)!*Dq4UexFS<KI0ka}i
zjXZ<i?Bcx29?2Sw<FvB{+BR9$t3SkR)#rG9Jc+Iivu-fgd6X=L<Mf#0#wF{>Ni@|5
z$+~unS~xL9W#^`-{L>U&c$}ia+n7(D$owe#XR7$az=p}{S~MAMC2M27WEGg2EFZEn
z^E}BT*Veq+2yaUc&a0NxwWs0jg}A)>Y2SZ)TA4>J21^I#()+%-HS0f3_Gxad+6ccJ
zeWfB+wCn`--?Shn;%15tk0(29RFd8gBom@&lI}3CwW8qv&JpXoCF&C&iz}I=w(#}<
zygkU=SJE76{EpP|O~^tSl&GTU2@8%SDj41_fwwtQ@zO0#&{_Pz+sQ!8&PvcZc>6g|
zf)YNmV=`}ocKx29_JN7|Iyc@qR}wxH_E$Ayw`4x%W7+roJ9%eKI>&2EyEuJD$GFHG
zZ0Xb(U4Is%KzM6~x8d-1AUn!d{}inexuR8udiE3NsaEiI$?gB~b}GC*0&k<>>?C-*
zkH>ZuqSPMVy5Q|rcx&M8IC#4m|I^H?q1uuhs!wIv15gP4ryPBnb~)9t6+3rE2Wd2(
zr|;SaabCy!beCLWc-tS|_Mwig{&f&FFkVOM+7Z0p^FN)8`;b>8vpFN#Q(iql@$k0u
zQO@K+|Mg|*%Vj$C+b5?2-#X<<-BFy{Her!dWokGzn%R-S>puE^f{%{p^$`ml)t$O>
zRWC37nve7NIQ&b@tA3%sx%#|AKT^+b-$H%2&7nM<9BM+n_&J69G#;f7Ywg;BKDl5V
zS}vCdR~u^E#LqSzgtv>~ZJY1yIt3SJJ3Z8?Hhi1up>FWD0=(@2Z#Tf(Y5(cM%k65I
zYFCFyzIK2^-f%qU2!~=9dTRY#7z?Xg4Dpc9Tx!%HhkElleSIA&^nh9o-ku+BSF^V^
zwftgJ*Wz|vHS~0w*d;!9`vPWL@9;gmrcQt1&>D+}4t?S8XVU-W?bFn>W8m#Rc)J}I
zr%}(B;GD3wn?p%W(Fv<_kG*fx?6GLXEv!1z$jYuhtLj8i&swe8mF})>ad0fas+8OA
z8hgxLlQzOtbmYf<-F0P{yN<!z<M8$sdU6n~Egj{qeE+&>R&h72IAYNnhlSk~7JcIF
zBi${^1#jKjW}7Bq*{1o2EYlu_dcxB?C9_RAc<Tjkn~lgewX0>CkjbCTqep4x>Eblw
zJ~YkPkEI$-OErTwrkTLkZW>A*y|kIT&X>0;zLiyf{%F@v@b)^qZGY0CsBrSM;O(mr
zdI(=UweGg3Jf<`MG|x*d7E#N=+cbFlnH-XD?t82M9B=I!@2%}K@MJBZM?aABNjE&&
za5Cuu9->w3B%B0W$=j$3Yuj+`J%^rrk9+xhmny;G;!|BJJl&-x@YV%yTfy64YS=>X
z_AOs$P%*sCb?M)xxm4@~%-+D)T@O1p@p*70_e+-+`cY38<llt1u`Q`<C!=L@&4H^y
zLzva=#mvw-vPJh(>%!YreEb@`^`7n0fj0g+-<@3TZ^=ZP?yFzg`D$AkUv2H-E9V5V
zw<o}eEBNj4%WwX~Yl5H8ho70*zU5G48;6QL<GKDTUbSg<RUC$9-^H%H@U|bkt@{K0
zh>jlWkjji2pTDNRLl?$7bY>YI`l4RC(b`L&kCRmfuRAtm=6ou=&GKQ+jC1i7r&i-_
zOK9jzZk(^aKJ2TsKYg|4m9HB5_^F#O*==|%604&{m-E$*+P>&hzU-~%dmoB7ayC0H
zy_g4FPcQibXYP~uCMWr^gU?Tw2l&A;KlT(bD>jn-7*+AgHDz9~Ir9q}eROw_zlvMY
zCcbj%#ASA6<bwal(ABbnlpV<&6xpsZ*LZsm8KE1v7vb@4{wzr6-tx#?*wVw~ofKpb
z<=g-rX&InF%#DxvigSc<>S!(U&4b7}olaKE_w>i`k4LeG_|oTKMbm$N5EY_dn2D-q
zCs%7;nEqhL;dA^g<>~pl7mHM(%8{A}i#Ofo-U@HW(1R`wZy#=n(92+EDd{JlYDpd+
zbCgxOMyNZi?eSxT#_{ozi4jUK9<D3Y?Ze-)13NdK@5!NxEgz;KWSGn!9;qhGZKgDd
z(^`7TUhviqE3SQsR|&TS&HaeJy);o{n45|UM^|l#&blH=GZIpidq#?GoKMl7<0+aB
zZ<lpTQ6D&5@I$h`fyr;WC(9S!CXuDtu2vGW97&q@idr!zGp5yYDgI_IEeM6N`}3&i
z&uG@kxwNT3E?q0g9CL22_+0WWL(MudSrhxh`<BT%_AOrzCRTaNY}ca%<thXFN|SAZ
z_sqW=8QRRvl;^x~?E;$Z`UL*337jnx#Gyt7&L?PF8uQob3A)Q!;tRF@Uqjdh>6fSq
zd@Pwi+xV2+h&lMzmLzHje|IaHW!2FI7S4><qrq_+!=A|>OUIGX9;=;uV^z0RtV)pM
z^r;nk0-mO+XJfT`daVBH6RR>kVpWw)mZk7Ejyc#R^kQ$BNd3W?XePg>rOhK$w@HMy
zlVAGe8GQ!!YlWYR(&_W;+GYpY%HNsW*&3xa@b*d%_U+&|9#DnfzoOI<Pw|mqc#GSS
zD_a<UaiuVo#aFy&U``#coKp*m<kV?+yA{1>KeGru)&?nOBC~q*lcHQ?>>Z%bguXQe
zoog8N>uetPbFWBx5vb|^vU`OY!tvw-J$%G@zM+fk*Z`gB;IF;geaXM^RkdoqWUlyX
zp2L^9f2WrH>(syRoLYF2+{CG5_>w(1JdhgqZ*nBY`Y1Q|P}c%)UE+*Br;nGu<R?>U
ztf#()w`bw)Yq(?JZGrg?wL^E@-j!M?4))}LgO}~>OR{S+yv+}1r%_*4YHdd=CCeer
zrYX;DI)=_#G`~%&K3bW-vg*YitGfKgJ%5u`t^(+<^KJU<Y1gLyaC@0uQ6KDj?*nUL
z>EST$X}KMG%2nd4|IO8GNrJWSsaap!*%f0~4R43e-G!^a+cj^nT{&LZ)CAskfwx!S
zZQ&odA9lBEHf%X@!mf?f;j_8_?apv$D!gq2Z(C$DC;J)S(-mfRS~(OR=}`U&cFo~?
zT8B36$Bb;hOsl3Hwd&b8EAuN>o&4UacJTHRyglczDlC=yb)mb4PjOd&blg8zpcC_U
zkCE;=S>9bk@3<*zf}0L@cO#S9O-sIV(*UQNB3@baS0@-7Z9&JjDDg?QX<0wpOskM>
z+`Y2RqNdqqTxhoW>wT8VI+A5RR?9L8T{F#_KQhd{dtZ!smTsDVmu_l|`)p=CPBXv4
z+WV=grv2Ph69sRB*QJ_8Us6q_{n;irb@dW>+q0onx$0VV<FdP^7eMDe<)#KB+*B)<
zo8Cp?k?!TDsk8W<&<F4B=FlmA4_{)?(c>NXzZ{A#jz{jDLq&s`sVe57r7);3ygk$z
zE#4bl_d0ukm*C$U=Ta4F+!Nzn`kgwm{&AP8!_<D%SD#_;DtJ7UKmQGW4n7TcVe)tM
zd7bd~A{<@(4c@29uoK=6+~d-tb@Ud0MO%ftbGVkm+ud+^7;`uc@j9oXMUTZlwH)5|
zgR4gd=U}IJ4qeB4)oGhc*Xr~APz#>2<1cK_>*hOum8tKiczD}9n>>|#%(UUFKZGCM
zkJrLGW~?0v<ei`ue>}+iHfOPJYtV~1yWQeBdHZklcz7EPZ}Y+1D(!gA;5YBM)utP3
zY?_2G?iJ5LJwi+9!a0@x?nQJK8@%1yl&}2`9S04kJ0AQN@b+0HvOm$F4#V47@OBBj
zU9ptB@q@m~vDH^K7Lm<8!dH3vvH!7&FPVbAiYW~zx8TcYPOd5EpM-wQspEUN;Y8D-
zZXUVJsixPN(@yeL&!Xg~a6QY1e{!8u72&N%7pEHDX6|kuIpu47G<T@ClIZ1p_}yD!
zU-_zY2kt3@eVKJ-H%M+jHG7QLj_jFF>wNSF-}72#{~GW1W+uvCJ@K`dK`(29PG0*r
zvRJ*Gs*=m8U3r|E$33Uk18QA*TY2L+D-`mT{$+nx7kqJ{_}bb9DuG`07q=jtuO6hv
z<XXqx4AnntsF`8q+OP0<6^>BPY7v?}ID$O}5pvau(jw+KZUoU|&KZWL7N)m5LOADz
zs0Ww*Y={<o#0T?lh;IM+Urtbn$}S91fsJ8WOr}>YN0?@mj?%dOQRph%*O^-$=ZKZ(
zym*~VV|M9ayeuE%^*lRXzxgDn&sPa@qPLDoO46x1NgBK)NoP*uBYKmpHWkp7_tHCs
zw}J51CnZHiKPGGMon*})n5_EV^n~M*<b`(a6P2VqPBJy~Cg}n-Xi8;vG}1S%5TC?e
zgd|lWQ)CS_W@CKdrOva5VjbFd=|o*%pLzYdd~cqyYWOHd`yR(=|BD#Keu`0-9I^66
zqkY;OJ-tmVKIB;PBx5z@DvWEDpnoSZhXrqQ!&~RN1QpwmKz;<ZyMt{17;=XPlke}F
zsFvhNlt_tJv5<I`-x??1VsScOl^yA1mpRz=dUbxRUOk1O@U{#dVZZ(SJ2hjOL5<b?
zDzUoWCR)vkM5}oK*`qU}w5vpvTI7q;)o}ERQ+V<=QG;@Jn%p}~JIH(Ki6=jSYv@UQ
zjSZN~d4i|?Oq3k(_6K+yL5*8|SCsCfRrC#y&=Ag5ja;0ecrIiFM5zM2ZC5u;{g{*J
zzndM`WX6>kOaEjrGmPZ8_Jy@m^E2m;m-AJNAeA{1sI9-VTW2e?3*_T$!rSrHPICXz
zzY^eW8F;%3-tJh&-my-2K<P7;g10XsIJ?v1ITVTqbfcf%tn{S^?JGOHeFkqw!P|@P
zo%-{&Q_=7?7Tyk=;#A*qPCavTYSJ-w-ND;NA$Vgp;YmX8I>-Eo54^Q3^UxXU(5Bp9
zYQmsHi{TM<W<0#TSk<9p+;j7|(1@?1tFE@|(|Eg5VeQ$Wb}egV*B=$pSyOGwbJ`}a
z0XCh<XVVOLdmP@bg0}%%t?U@Jsye*=in*Cv@HSTu^jUaY58n3lfw$CMow!$hTxXMO
zrA_&l+tiM09iKmbr%e|_>}ns)zZ-9tTZ&yv3**l!Vpr{dY~&8w6h9i453(tHl1<jX
zx$kYX>oB~X51W6e%b5Y*=Arh;<gq$_s=jE=lS`nF&$g-TM4JMfHgdwPYTm=D@jt?*
zwpQ&K?@qtWUEPY|Ee><ntr>2506(j=cGLal%+1blS1%ZQVv@U7(W4FF^Ol6TYjC2w
z+^@SSVVs*ReE!0^Ze;hlvBTL-IpOV@&M-IFq9}I@xiH!0ZasRk@b-XbwmC^1>+6$k
z(tgV_jRs_yqMtL(fu5P>;-L(4=I$4h25(<iOgAgXe>PJdrkS$+)6D9$RMUTUs)?MM
zYMP8mHAh!wn4d;wng!dk%z<Xv#&ct~>H253c^_!ez9@?vlPy|@_qkiJn@+&nhRJS<
zt>C5^1Kd<^mzzdawJD7=$OFz<*ZJK)g}eLtU9IO#Qf(04@V<63fnjr7o3_Kj!|?T|
zkvuQCUXO;YJT`~7nJ{m_U(~PYz;|J6;9!?tP~-m2+gIT2H0t7uXwA>@Q&oi94}8#s
zVe=NW>1``uINGxhTIrdge4RNiorbr4;O)n+nQdj}=#!U=x%B|$;&R|)D8CQS;PwDL
z%nl$A0N+*!-vf2qTnAeA-2l}_JKn^8j9iVJcrjoOI(nTrFU^3rwys{P)yhjIx0gE7
zUx{l&&UQWzHQD7*;zBY^;O)?BcCvVRzU;QijpzEWmuyOZWy5b}(>{26h_mGWA9>Dp
zw~_H+)5x<nEv#%;20z29|KewO>!Dr_Pqp7qPlO)Ei6$Ouu#}qjleeliCL6S+k47H!
z(V0D*owvYNcpET^-5#8)^T6Asyw=~QI@$e%cMUJrO}tp!x_WB`yzMj4TU$nWtH@Gb
z+kblN6?#^11pBApU!H<aeP5a#k+q%rhMKc%SF%W3utS4cgxkM(tCGW;`C3m6U*)Os
zXyBbs&`-GFt;Sc$iQ_Tj7XADn{FNSL(nOFW$6S1~LCpE>@Yc2bWUzKYJNtv0-Hy*J
z*{O2ywj7$;^!soh-nRCocISI{)u3K)z@CqQ0QoEn(EbL@H)o>xQ~T~XPR}fFPK|Jd
zXxXPwvgg_H{7aZ_le3auEJ8O=v-2Srdh_FOO@g<p;O%a&Fl~dknG-_vnOX1yhMiay
zLzra?(WtH=a!n1<qXi)vFo_J(s89v83?nxk_6}kP$GmVAT!21_jyai`&NeXPm*08(
z5Uq!WVl*Tbof+O{qPM=FPF(?Sor{wcawb`6|B?rioT9tMQnasn3Oh$rG_-b#+&xot
z;##ssKfz<PoB9~ub}y2oYw&jGDs~gkWTt9pqV8}d?M0J?w~vb^XziLfeYY5o*SuKu
z92~3T)Q6p$#;T*(h5kN9*Y3n92;R1Zw`HEkDD5M=9pP;RynPIB*SC(<AnMI&jbb&i
zL9E=b#VW)jL9JfJD}ov5(k|X+4!R)u+C_UN=vA3$_N3rByTM$|msowp=jVsh{&!+k
z|0UU>u(ax+Sj{HG>E%pvoZ#)~O|dF~_Aq2dgjzL?VDD78_PfGWCpAp<=_#~c5T@7U
zpe=|+JBT8S&P67gZy2wqFlAK^Q*n5EaS6MpVQt<5A$oZN4fjI0daY%4oNFx}y&GG@
zRh@a*VdTB+pBAbgH<PorF;smQGJnq8TSk{q4ft<&CYcl&3xni0CP@3qewt4HOb+jy
zx(sg%gyz&ja^z-lFZmh&$nYD?C$0;WeKlE$_!U;b+sRzxc{}wjGkthPDt<wSYtDVl
z-K9S=@J3Y%;I+-{2D3XQ|Mu0<KYZ1xv#-i9SJ^QJEiBEcZ}E&CyyH}F&bIYAFHhjy
zd~Kprb?DuCWcX;nVINHmAP@K%^RG8ORiArGvsJwQ%g{rm=KB`jjznMWciN%!Mfil^
zZ8P#4j(tT<coZFN4PK<-cHRDodUgzqg}1ZeZC!Y~1>RPJw>M7Ow7M5wq7+yOZ%_VZ
zB{$5f;akvM|8ZBJk#5vTZaQ+pqE=pbgyy4@{*Y}xcFs0EXQ7qebkp9=?rMCEoTWGJ
zx<D=YDz{ae@>|(G$ZX9jtM;Z@)iT+RmO!r--rjv`)5hOyx;mVD;Lou37n>{-Y+Anx
z&oKA9-S=(k6=c`_T6Sh||CeJt;ju&A(3&gasT%l~otl*I8Q%K9TYq@l72aNhw{zj`
z{F3hK0!R15Td%2ZYKaznklyP7cw29ryUJ0|ZXD#U8uVxPC7~bJaQnad)vXzR;~Exu
z`B-2}rWt!P(|EqiG}&HR<^c@s%>RF>#iEpR+2(NWY!i-tJlvjb?j>fM+1}aa%;_vM
z3EsANo@w&;$uvigW|%>@znBbodm|*>w3+zXoIjgp4)#kk=dw~w%na(<$*HDu-Bh!#
ze5QHXJj*Qqm}S1{m~HCA+b(Cb%?Ef}xQIpP=ok0&c2g^U7w;q8^yWQUwhwbi{%&d#
z;6_%QoAUAg1~C)7m<*0-%nFxv)4iFT5oTNPnOO8-xkY=|Sag;8^9gnBMOb=zIBetE
z^b57^e;2y(ck~TIxDt7Lc{i7aO?K%9n(_#E+WHK>WnK^a9?@5Pj^2!~s{b1}iFVwM
zKQ9AsKf?0lUSzRMbg9uR`nT}5J(_1*FZ#S_&W~Yj#9(I1>-(!w5r0h#VR!O=^wd}~
zG4J9V<n=Pu%G+n1YC-RLFq%)H#!kKek=(FD4*U`9lZeBM*MQk<uAwl{2cP}50h}e#
zInLSb>RK1QL!3c)Zlx@>sTO{@QSf#&{`A3DY&x~UrrJYodf&;WwjFHh4{tvYvuW66
z&Y$#JD&i>&%;cOH%ejxUtm6-I$M6(f!mHM}mxnUDdure!>fR0XgnjUh5BAcJ4ZPGR
zgE=}p<&&Oxs`nkP6P^lgjNhJfVk%tBZ0D`kKeE3H-nN2&+0?4X@b}*%2c<bYu6Un*
zGw13vWR(4T01wzvFBL!SrLW0N3&hXmpMxA4*y*6RR2$y5fVac>x^?00iqrTO50MW?
zE!%DwUxRZ&^E;dkzVJ3VHD}@bSkCP?(FlCtZRQ~#?X$6CGKt)?+)gz)!Oo6%KAOP!
zAfJz~LSgR9>eL6dm_3BIZ88INiM+7N_)>xf2P<(xuqK@+TO~(`?!nt~)xtCq-kOJD
z`nU<k{vEC|Rl}64Dqify%w6So=~*;qwW4_SS_f+0X=)X&rqnIby90HBD<AdCGJ4Wm
z+}M-FER{!_Fx@*Erd<ufwf#mo*)CD~W@VIyz}stuqV*U1Ap~z<d_$j+I<oKVIJG`Z
zzE(hz#=+YI3zJj;-X>Dt{+1KHw@8Y<hqt4uk=ZaiK_Pv}Me3ZOos|>RYc6xeWF+P5
zO?~N+s73g(=1->|3~P6DeW;b7x1|!)j;!Agzr^eF(m3^=8>_d}vY(iZ8M-Nk{uKVA
z{bY2~_x$-mjB3K$MsT<LlNfD;xd;7XbvZFsyWsA>a5lMNta|f!wL83h7^_w<;^l(3
z2ic>1dwRT9vd?it4|dg+#p4Zc8-|b@@it0-zGN@tlPHzHPA){nNHxerU%+=h0q;?_
zh)9)jMC#cQa)+6(sWKV;-aA5P@X&>j|5`UKOf}$bF?f3+UzoPi>rd<(s;2crb?>`S
z^$QJUFKq}rfSD)%0Ue?+^)0;JNY7&N3HE<14Ob(w$VZe6(`RRxj`F+j{7a~+cO>^E
z0T1yf_Ih1s_Xs|M%t|@wkK?t6wJDRym!U5^oc*K|7O*FcIl6*_@kf%|lg4aAp$wP)
zg|{Df!Cbrz_Yb?&penr&W)MfWW|ki<Y(Cmp%LU}z(W}kDJ!bCN0CKh2Jzg$Ali_U~
zyv@P=V;<+<%Jg8qfwx28tsUNm!P_SAcEjKJA1^bj0dEV9cA~R7)#!_l+QZuw@HQRZ
z_JX%{277APNzT{swrK!<q0QtYaqfP=z2GIhwZq#s?H&3B-hK~n>%v<rygdeQv-+R`
z53#EzybY>pmoHlGfloF)TVhj*M*rh&D7+mEZx8LXYW;dN&0FruJ;_bAL%63Ow5TS$
z=|3;qEUuev4#L~Jur~^wweA=TemV>PjvIR)@bNM?Gixte>>qBrxy)Tfm_Kd4%3c3f
zvugKV?twRLdPH5ZwYLr5txe1O+Ein_P2W>*+&fO56TIzs(59vIlqU>eE_aqgD;_e-
z3vXSIVE1e{MUHY)*>2RWFWof6+g+ht-*}<DwsKRC7Zz#>i_X5bC`Y<Q9lvqYQh0l_
zDxRmjZn}KdqPOsJ`mQX~a#ofJc4e8tJ2OrB>6ym#$~41rWSTn@GR+`(yZCLU8R(s5
z{(-l9Y<Q%yvW*K4&i2SQE#~s^%q&v?-hKmbLyl*eC0{bkx7RaFyUH1+On8QQbnuJ0
z`7+(iL`%Nb`m;H6Jk7j1lWKC0O*NnUq?(VDQqAq!sV1;Orung1mPwzIWtKh4GN-Kk
zd;ZzRo@CL778cdQn|1V^MW<bE`YETI3O=yt{%MQ0?6ok9VNn-&`vD$@S=}^<_uE{s
zsO3(JeulU4vn&c)U{N1<>$l9JY&hF#81-vEJjuL$c@VtqOAR}~rQ7SNWAQBu);{F@
zOO0Im9^Q6EQ!dYIpbfn3bc!1FnoG0cZR8V}3X9vX;p;7@uI2NJ!tyQM(T4kQjdf`=
z_3;m9UGz3x3OhhOO&y!{o4@83Wrvclzed4ZH}cfNLh<R}bZXjmr+!)PWY*8A{_ULF
z7T{FI6;DmM=8z}-`ocx%TO8y0?Z9LI-lmn)IJ@<+>E%y0v>TfqR<UV%Ihz{5+gW_v
zjpy~CDK;(O^Q_dmR(LyjvrPlwZF_iI4&IiBw{7^`DqWbn?nmtlb0;>o>l?Uh-RjWu
z0}j<ZfnLATp-Y_kme5cC=d43v7wB~yb7%yA9)O15)=6DVALrj7@?ApkDWW$77UFCS
zZ?D3|ba-11-u_aCS);GK6p{<g>R0wqp7B&ncXmwT&sa0cO98vRv=iPwJmMu!cpC$6
z9q{%sIc_h>RPCJQr6>5oBH(SY&7NA7gIT>mFO?4cU+&Asp<eVdz0?TaK83f<;ce5O
z;re&p<b_dJQzIv#N3VKKJsb^Z%lW8bM;}#2xBd#=M#I~APG<Yz?YauSI!3nap;ZBz
zj_=w(h21E&oJyw$xBXnOLf~yjg=pqfvf|m(5>HRAWodSFvy)|>FFO|71}L&V_rxCq
z6w;ggfxhg_8XBPcJMoZ`*VB_+f@^3Y4YSB$S%nW3-d?_s2W(rAp6z13@^X+`QU9Fz
zF-TtcrUy@C4#XoTUbdWMas{i_yI=*jq(*)fqB6bc`TiTJ=MNHft{y$y1xYFdZ=+r(
z>muH$e+s9l_|aq)`M{2*<LIq;qR$?PW6nN7GyD_C&`r=8-hSVkzA61+uknd0K;QI8
z(Rj_Z#B2U1@<QP4Q+866*cPLH@1oW5T9ig4N7J8=(Z@A0?COfq>x(f8gST~H?Y+w}
zJl|vRUd8Cl8FD+x5$S*@>2X}FZWN5AcO9$4o!Mj2HCBDjlGO=sZI9WZygOP|HbiTF
z#c28b9j|=Uy7vx6X~L^0t%bMIc+)KCxz9^SGAqMe7dc^t*GG^c5TPdjWl5kBya)`_
z)4ib@y_$aeicsC3&JL{`A-a{5K0bZ;v+qO6TZflDLp85ns6Lbo)z|Pg^lFHzpgS~+
z4^b_6d+{Y1c8A$pJps<L+j4v9aNQ)&w5ofkCN2ur<()azGn;%Evgigg|FDX4b%&kY
zS9rZ|q0iK6D}D~n`R6;KTa|WcEWG`;13ScC=Fsqe$ekOGe}j9;wYx6WLn|x5^C4p)
z^{Yn?9gfbye#;!H(1)zLVPv&4-*5^K;)$8$G@kO;4tx-gKae+IqqjnTHV=NN<P0Zs
zFL)l%kdN~5vFONw@b(#=<V)AkbdEday$jCvbt-=mr+$5dH?owEmP{lUKaPAvUUN0^
zNVWE37agzj;td^YHQS-r`y6tEw|D4Q)xbCOp(*EobhOrI>~c=B<FjBcrh{D@dfW9~
zNBX2)?TV|7u3QYCFuWZBZ@-4OPm^qXuT~X9Pd>ZOsz%{fO<dzH>tr{5Vg_bB_w{2|
ziw+jgHkY%qOpf+hruft><466qA=0ApG4wW5ENI;pWm1p*T9Z1kpGCXLh3KB_PM#Ju
zFuYCQXVYIdZ2AS>mW8)I9nq_Kyf6g)xu;DhuUnP-4|hH7>ZbW!n7?U3?b#4bxC%8~
z4z%3JOmmGhzzyo|STx_=<=k}K-HohViweAjqc1J03~RsUS_5whgwqr1!99Gx0W~sA
zjq(|0$OLrW<V;f~SEgxp4IdLs-MJytY?+p60^#il-%Rs&71)@XY4Uc%LyX32@0M*I
z@&B*=9?x-$Y*XWLmZ>l^%PcCIWy({_`V`DGmbV#Z61+WrD#NtQ_+q|$_QhPT`o-k$
zkZx)^(oKiEY39@uID0hJJU~O9(L2@Lg13?I)(hUwnVM;;b3NiZwjk3S%E&T1VeYHQ
zY_q%s-e)IuH`l0qoN@j}XNI@Q2YB4U|9ve#>v<M^o?_8Ot_4RddPz;YX@*6kVQvV#
z4OwW>Gp-KwNPqd+rA58*!w$qZ3vV04+t&TaS%R(JKT?ynr)OK0d*Jf`)tM6@3-`$}
zO#(EtVSwCw1ZW;P;1gB{=+w0U^0r(G>P+1N!+Y{}C9bwDsd0Ot3&Haq@OIgsXwxrT
zDhrc;gSXzl`{U*J*JTI39C*8mY>DCJ{Is}~Q+sPMI}2}`FK52&+W%(VCPy#}wgQjB
zT%Ol3wNf;Ut!87Ff=w@akz2xbAKu>L&sx^A>2@jpJf8C;+RDmzR&{!4mGv>b{0lZ6
z<~q5BI+*8N=>|6Cg}2AM@pX8d$Jac^`R_iw9W@zkh`z7e9CY}#<Y*tZE0gbG0=zB7
zSvRIPufsa{CCWII1Fkyo#+QP(9{3up@HQ0QzT@L*1szI&w|nXDbfXV56@SB+2Gq;U
z$$6DyCZLo<xp&yruB?a7Sv-}OT^xxUymWFOxoJEm%%yhba;@-E<WuVE;$G^Tisr@7
zX2D9B3vZXg+bZz(o;RM#a^#5*^3s6WXk(MS^nqHrB=zvi6mMNG?yc9=z4fA}w;rM`
zZ+r`Xqd41_#m^3J&u#G0SnfS<{pf3b<J9`ioOP%JS{-q64{~Ze_p=Wh{B><jfC}Vr
zX>u4IyL>Kn;XZmgfmvp<;Y&o*^X2VdzX{N=DgipcY=#%S9o8X$?Ct=)>q9L9Z?7-n
zeu?kA5xs<N+=FZJygP*-y%YE9YSdC+!`sDcxX-r@(z5133cxeHhyLCh>ZILxzm}0F
zG!EWouowAD{SXBgh#&_kLZ8M&YB0R|Gb2uKm=~_}eX<<zb~e1d1#dqePF5(qeFJa*
z*b}D-`{LB_W}HS&Wmoa6c-3X5$<+kUQK@)(Nb#ED!#>4Y(fXlKtSWtolPA0#3~!4+
ziPj-JLGzelZVb07Zw*yjYCiu8VM=HeroHTSJo<qBTmkHQTpUiO5;bWSnZQ%1b^T&h
z5#A1jx8Kx^)k%7}>Kdz6-C|XW$H5h2_5IW+-OL@W#*3oypp(5lBtoyZMd@CfD3$V!
z)L&1jXJ<xeSp_m$(HTly57waKWPLL~{Tn}z)763%S0h+EkC8KeGFWMs@x2`lRt!05
z&0u8EmtY;>`))fhRA=DrG<a+93;)O3;&Vb&VGeuNzo7rV$HN3~tCM41Y*CQ%3}+r^
zWRP5=m>(Myq>ep;RJ$;})XW?z32(E?k@3#&upGQS&-s1GN@n|K;=3Qp*_+?-TH(?E
zmbw4^^s~^)hTcXW<GO#(rK|8Z7~WQ%4Qsh4&8y6r{}{Q(RsHqI&tJR9`?<&ClkWa{
z`J=x!+WfU(imx_)c51;R>ek0RKF4E$4`JOUr%wHae#4bTO`ClSKOk2dczc@XbjD8l
zx1F3imk5iG(-R`oz=!$4lJK_fS<b#)d42urp}U-?yVi4PbaOnp-Ff}@;x*lqIv2mk
z1wJ0cYrW$sd?N668$G*2ZSBmt*maWg`k6A!;^ek#a2Q^j4>lFrVbc(pw<y7;>+h|q
zhdw(O-Zl@msx&oe`787bsTr%GN%r)$sCWKsQx(lK0DhGZ%`#ta$uupOWtz=A>RqNO
z*8n~AV3tW<lx=F_-%TEfK0C~!6MHNQ$m6E99o^VjPknmdoqS-c+B~&tEce0uhplRe
zhxkiXcTJ>j?fs*hro-Fe@OFJ8bl_@esqnUkSEk8#FT-4Jkzw9=Wf-5EU(7FSznChO
z;hN7E6ZSUU{Btke9J!osMqfxb_3o#er5+h3`TGnrwrYkc0dITe_+kzfN;e+wKbu|`
zKAZ1$d^SNVKAXd#pUuQwX=e2AX=YE=&*t2lOylT9{R?kLx1^r^m~GOZXPX!Jqw-$O
zG9xEtnWy=(%#TYm&Bv5XljmWEX}ANHm&q_K;O*kqU(EekU+Cqgn;geKn=;$e%((lh
zrqlLR6FV~1yoyLQg${o(1*@Pdm&`N`;q1dHndURsM0oq~3w1ZVodj>|GB;Fb3A4Qq
z;XS<Fcgdn_7pYe{2knNp3*l|rd<!!w77gI%{h6!j1kN7tcK39Pa>C-i{P`Jp+oU$L
zI!)jpk2T<~-ynQW)S`cMr-$1U-LE4vL2d9B2Q%06I6&(Qu`eN=U4j;B6z-dYy9em&
z(E&OMZ)@!jP$qr7(QVO*;cXjOJB!DF258eTxCdXmYZo+V>RkWvF74pVy4Qm<><NGR
zVE!8H?yrjv{OA$j6-vNs6y>eIe(}(nrufe*k;NHf*Gf2=4Og4AM2o5WziX$>rqiJ|
z?QO#ImUCJ!uETuZ_||A6@U~b<{yc*$O`g+PJjePyz%vJP55ii1u0QDCmZ!e;=j**H
zVpCjMdb^e2@Ao#Hg|{i3Nk`Ftxe0HBXX7_SE1G!FuD|wkW=)2raMK1?<6-DiUT?#4
zqA^j|p8UtIax3f_wizEE{H-7EP*1KQTrcneK4KQ2AAR;)%<0X7w^=+6g16rl#RIsI
zz7N-jdw3)|kZbydd8G{WEUTyXdywxo#8X~)!4AIm&=xXiR`l~!xv8FNFu_v|d0aG$
z9ammnYKJD)g&C;@KY6JueeN`P+YH{WU~cbjQTn{pxCwRX<8Jp>6ug}eZ_k%zR;-hc
zZm#!H>{TC~L|0Cthuj9<uIAq43vV0YFIoJjQ%<zFZ_hDXvB{~ZLFf=YoN7ibv5a2)
zNqD<8DnJ$B?Q(eA1K#GT6`(|}CG7)rt1Iu<fU)F|&SmZ>_(XtWs8x<cF>BV&rLTDY
zJmCJkU@1FV@Xcf}S9z*ZkXDhWozXf-8-HNlavnJ-{|4zQ^U8h5C#?WOww+>EB|hez
z__X@3jo|-G#yxf62==<={+9aj9-dr!yN}i}1M_vfLRT{@vpG&%ca!}KZ*O0U)3}H@
z&5eqaa}0b9iP4|E(Qi93+f+YVR(N~4RJ6wCBl~w3v!oLub*@vSZaE_qg15TGyHGi|
zu(#q{`k;It^Wg2|rD1xHcH6B3v(FpC<<U7pOS(sB2fJK<ofV~XC!^JunbvA^qIGmJ
z+VD_xVsaw7k7K?D){1KryuJSq*`PV#_3$vItqWGu9+5i!IFgw=GMV7;ukbb*rVe3N
zdjQ$uKa*Y7qH3_#bG@w;tik2T`(}T~@E2sElKE{f!j3I?o9j@p>b_v_$CeQ7xdLPV
z4k2edM1Ju0?QH7b**SHEb4`&4?8>|sq;(gAG@vtly&9-^GRC*iORY+6x-UFXUvn9F
z%hDC?E1N@O;B5<ddtt3hG5kKaPUC%E|7BV+GYD_{$GfzFt7C37HO_0DO1e~}F|&yS
znA!iYe>IEy4`=qO6aAG%rt#qt{)$Bl%S)Eb$Lq|*k+oQDg`ZB2^V7=acqxmKwMbvK
z4SMlGo}VWV!`yvNo?p!UaP_-NUkc`4KtDSPhs*avM=pR@g1mxwW*h!{J-lWPArJGm
z=Q`pCxj^<}T@U^2VwcV&vMbQkGMbTt0B>8v+wVH^8pdPvj`Q+)?gNj}+kU_+Q<D2s
zd+OIa&By~TLcY>>@HQ8jiugv7Gu-8d4%_K9zSqCqbcngst{1G@0&mlO$RXV3t|OP+
z<Z9z4KX`k-b++k^hT8TG^DfN1v<=BJAI4^yue)TLqCeB`JVR|*HOn+VnPp!6oNW>V
z=oQwr=$~%Xx+g9Asv=r1b!*$7={qh$6aAI^MDB%!xffPDVAUgd>p$9E2drdHz_){F
zwli<x_dTCw`ixCCt6`PTgH$uJ=oixl-rBc*F)QF%hlO9v${*mZ^^3{%Z@M`TZ~b?r
zn*v+XO|#AECi?4iQ?S=(({^u~d7m@Qcm#bi-!%SU?sffOa{T(i%yFffdM#5;iAAYq
z^G<xjJJQV0H`7dLzRzYwwa+Ge>}NCf-Dh*M;}<h{Vy3yhF3XhV=df~QmgyXsWnQmE
zgHFga1@C8=m0L5+;}z)1N%)e#q?^!(>E?cmbh9xb-L(HX&4lrPm;`S<N2Hp+CVepL
z%Dyv`Hh(g6*QXh8W`#U=e>Tled^UAXe>TUDeKuvMd^QWwqFY4q`OXZ}fb&iXc-s-3
z`5Nb+rPRAN7d_f2=74L%-!2wi>SNI@cx&M8kFa(n*Mqqh-5!euy(P<7io2AT=l7T@
z%<zQp{Qt$Jb*BRKXd^Z2J+wY*)=swq<ohx}c~8UEf&tp~$zL@&XI41tuZE=pxZelJ
z${wR#+%v7q(FUi&-@*Y(;$zXf12i=npD}0IioCvlfVT_atp~5SR`7OIb(d<!(yIyg
zSLb!UD#_W@%g0;8v&iV2?4h^#sN1$=FH`}vT=b9yaP$Yb+PkSu-@?bTb*MkBXvOgM
zXpl|2JZ*|ao4J)u55F;Nu3@A1Y*pLiR!zTa)n%Sz6?u-Wx<{ra`pLL6Rt^1w`7r!|
zyE3inT-c^acx&M8x{5Yc<oy@9Y|5a94%)`Q)z7Z6_#djxu&V*QomQWjwt8f7!sOz-
zT>%ZLcMIm~da+Xq{rMI?wQzVl^euj9nA-u?{>*jqCiCyVp(`z5#%m@%{i*0q<H$7`
zZCCD5%;!;0w}7`b`1eM3;&VAG??sdPeLZIm=GSlZcPM1te|=z&|I2)Hk>yqe&8sD|
z3XR6F6MCkn=6X`+`g<t{ygk^-OXJ|}wOw8+jIKOCl3ic$)*s#;t%|<9*<1Nw?w@;@
zciQW%T3i-B_KEzEcxDj8R^rdX!(0^2Vb*U>twe*H0dKvyUwwO!`_~SqKBC9ffwy7Q
z37wiT&q$xWbvwMDi>W2J$4zd*ZfJbl5f7P}il%4qji25w^3%e7eu_ZXa7^aBbumEV
zY&VWG+a1noN!*`@zjUDkkPBLyy`u+mXm%dzCp5^b%v#=ox6RM7*P=#F^@g|ijs&ap
znP7I5Ge>kdTuqog$&oi)1wMvpnJrwMPKGOSW|W%Di_(p#XhrOaQq8MTs+$epcEo7}
zyzK{X&wY&5HFg%SFBq*%xHpIz(u@w*AoeMi{E_`zRoSWNPCqn9geotN)Q7}Kwf#Oq
zdDh?y+Qx3kp`j`PZ*Rieq42f`yuAQ#JHXo?;cc6?5qi(Hm}}<O_<Zw)D~|c6EycpM
zG?5IjHDoHCC7*;mPsbZ(r~eMs%V#{o+uP)Cmqj<9!2dVBCHtVmsAtdS)cMgl**Ai>
zw`fkh>3G+uNq<cYR%>Qy9OOa-J;b*gh@W?P4t+(RbVbJ?d_6(>_?#YU{$Q1bx9ixc
zU4{BM8t#s|9is312C6(eLmuX(ZtWGMAK+~Ov)Hrv*{`e_q!;%C)$9NBHfq`bcGX2Q
zUyz^Q)i2B{j5F*vIZv&+o!{wP^2qtUj)$e)>M|=>jq^Rcb#i}b#8nC2?t-^Fi@H>r
z-jz4Jt>(@-ew076H2%6W++TO=`m0X7zr5k?(YJn@eZ@~J@E_-gx96EhZ~iTQkBv?p
z*oM~+-ln0$9D%o!cQf<D^Y97J-R3-Rf2Xb;a>L2&JkQ&n<Vxk|c^l8polfNPmFJw#
zJ*edgGIe>~_+RzZLwMT*-fo-Ae2tAvC0@7Fs9nEIv?*r`oAS}KYT}09<GEEcxc5wt
zVUHa5pCb*)dZ><W_ROlgeXR<v$i1ft*$SnYH>Ky<<rO~H18yqW)=e*SyJ`7*`hUmp
z>5R5$Odawaid$6bPPW;*BFp5b-+B0Wrg^+F)BOA{!>qoUVannQ+PpBsc=pUNX^k^X
z=H?9JOAUIKT6F%9Of$C*`tv{dt0v(`It~}y+%$Lrng4Lz%Dr$Q_rb>zR*iwTs}H$r
z1-$)$W;+4izJ+TQufH_sy1zBGR=zd8&b&2eX1p^cX1_CoX1y~h)8CnHeczcMTfQ@8
zE50+H-@P;2i@r0P>%BAK&)%6?CElCY4d0th^WU3+tKXY)_uiX9#Xp$hKYTDZ;H@9L
z%^RC)<~;vsPT&4yu69o~9sE+w=^?46=iOA(rEQw2+9S;rUXx}9!14R{znGNx4ATz}
zRsX?RX6g}qQHAgo2V|NihcnEWx@gwawR3ZE21rjgX?r;*z}o{b^Z2$jQ`(=}d26aU
zG(6Q5wR|v3_<zmY^vV3OCC$W4OEoRZrJDMcQ;mDoR5NOQs(D-QvzZF-*GHzBgYdQp
zyv+x1Yr@+L@OCP^{m8kdey~N=(Dc7fppHe)zgLsD;p?<%7PTK|(LA(<SagRroE2h=
zq20mUJTP|<nqUCD9S?6$_y(x(O@FNmK?7_`Uu^(2$P6;ks8jO2_g5gi{kX?pK@a?O
z^@_iCbM~A$gPj`mVrTq=|A{)df2RO_wJU%fr`$KGldC7t7q+4`9}CbAXppN%1!xC3
zYa15PyY(O|VwJC&{ot#j@0ne3^He8zJ8*!9R@EREY!9qkK(7_vehqJ1@f<$<)=EaN
zRYl=saxnS~x=6jfuy(suzpu7x-F&o?g;qJ>ZQqktZM$t%@%#Ag;cW|^XTz>pwE^Dd
zhPUhCtzSB?g+exRFW7MmYtzf)-7CiV6yE;4)v6`<7*6qdMR_d6``^Iu$DggbIM=2E
zRqV1?MF&PVY6Bl<_qS^yyluABuICf&dOU^sX_$Is3G+z5+7)n_`Cm93$Llr;-n#g+
zb?~+ftbIBREef`V@Oo~-^=mu3mcZM3u>8n;{1?yhT=ZqeYd3QNe(bU=O5QoVUCSKQ
zE?67MRfrn*M?75p@Or$h#f%j9jq~Wq6X5M6csnS9T)HCUpO++WtBSW8!PaDWJAJ3O
zGGT5_c>5GT)xclLEDJ_cUWi|F2YzT@r#9i6o{#R9mBf7tZ`CU9Ro$6GcJJfV^6q5w
zbYP~bI@x94Gm8Xotz`CWg16I|JN~?beH-U}wKIyn(C~H#yluMDPq(+>Z{c3rf{aev
zJ%25v_HD`@%DpwIcj!ITizN#*j^3V)zAoO?B}su=m`wi`-p)8f4mdsDG;%QqX697W
zcEM`ZGguuG$@eIYPx?lfF1?3e@OE!txEjK<4q|6&=WudjqV)Mgq*_oLrqOF%znwjZ
zCu4Q`ZmjG#VpOVpjNUwp(zV0%KnsMcSrXoAuW;2sdmU9KL}Q!rZ#)T=%NnLsdZKag
zHnCWUT2nK2t`w^2=*g|%ZT>}JYP&p)tgA3t`?Hs#2A<z8Ve}osRXRRgbKq?Vyj=iq
zr||tvUlpbjgToZ_I#dt;WzLFS>Z31FlTu^moE@ZpX2QHKLAvIZQ!hNI=gBzf0&Dk^
z2YR|N8UFCri}~vZWN}u1&nzCjqK?#xLBF8ymSjgJ=a=FCWvPY*>eu){Ra!!>TO)Ga
zz6;dNa)BCNB~W*3u!oYq>P1+4f$KYXn+$K=;q3@7zK8bsn;Hk|=WoeheU?K{&><cU
z%)y)ho=A8*^m}%E1~cz>7_DLq_W-;dF3#N};B9Akn-kuWwyHM1csS_04sfv#AlOd_
zcy3(3;iske`ftqgRVjZz)z9gt%3;j)#FJ}Y&`&WX{p44LeFF5J(%0cdhPT`CeKdu)
z&o=P9n@4}v!E^8qo<FY2@OB#?s~A8gYo@n+A9?G{U2m1T#_Qlb*{$Tg?ilZ_nf1|-
z%X;Y&{=<;9WKQ(*)Gc0L*L$NS)4MH(f2il*|9ed@d~Vagft<^`+W2{~%Ld;5_K#J!
zv+zB|+EkPKO`-GjVvkzYYPnTcdsvmHs#U&?-O2nQ=ixl_ocr8VeKfmVTQV~PZzsXq
z`P=Dz!rLM6b~C&!{&%(+0#h4x&M;1Ro3SF(Jb#^G0^sc;IAfawU%O<OI`FpImJHLI
zT5le_y$El;sCR$&r>D!kzZG@vjL#O`{l}uew^?LqXVG>aW?MIBneTrhbBWp2^=|lq
zqq5BIE%Z+tqt{-1X~N-B&GB!|@V#%1ZRk5Qdcr$1f95-rHSL{g0&nxc+c(sT4`6N2
zBJWI8-FIf$vUjFn^m{Y4_<Qs3jQ3{Z;`e57vG?Y%<()~J_|7~W`N8}=;-gu*^P{<U
z`=fc+<C9r`=ac#3Of@T#Q_b?OsV0Ki*j)$k5;vsgyq<0bUi)II#$}i<8JWf%9X%(!
zEnYO!gn49|eY?=Cn`D^2FTR*})VEtw(~Zymbh8HDUVxeXIxq`#E7eromTDHm+ooUM
zn?_yUniT~<m}waw&Bc}<%&CwM=KIhOreoL#bAI&))ARaA({9iw^T?8FQoc?#FK?!r
zDiLWWJU<@i67+)`pno??Gn+c6nUaZVrbZ}NAT_*qnmK_M@F%_Eon2DR;X<j#?0;|S
z_UC6%kQ$RKjdSIOj{%xS53LuPaK|J5I(CD6VtBi!Qh<`X1ZXq&zb)|g0KSBbDgK%Q
zZ{IB=H;J88kI3@R4{yi(%E#~cE9NUcALhFAcf0X*FE=AYDbb~0;qByu0rI96z6x)D
z@%C4(UF0P(Ka;-1SL+-5s%s;sHcliT8=hFRm<fMnQ*mZ^PQlw&@b(QFjkSzTmErAO
z*!eBo6upo0hpZX`Z*Rcc)p+6i&tdn(0;{fUu&Nup9n16oA2{p9-wC+`i}5G)+Rtn`
zJkGxi4yTgmmDfgY2RWMX_9MJKnVVX7id7zy&|LUhzroumc-wxnRrPoB?;W=44qyA%
zb5?e&T2-6xW$}OK+%9BdQ};GZpuS~Ju5~H9s-q{RmBd5gY1iAIIUlbjZv)+F=K#A(
zb%xOm?TQVwYhRpQ$?)|~8+0jnyBgk3=lx=^JOJK4N2l(H4(+($P($`~<YNz36L?z>
z-VTAaS4OhiYqBSMy*-tV7WQW)FSJm4xA6AA9I`p^c3@6#70B(aecyU(Hg#^r!ruB)
z+FQa)b-Noi?_jc~C$lRI|MM#5leTk@D%OE{UmvHwgSTZ9xSzq?CCN_RfVqJc>4SB4
zsw2GJ*50YH2F*FdNiF0=*QO?BUV28Bj~>{#2lBDedC0ID0H1UD;!Pokt2XRqKg5ML
zetJ#5$6#_dcEI0NoX4u(LOXo!uTu0dn!jNt4c=afCDXGevmxwF98WgZ)S=w>(U`kC
zf@N(Rte)_8EcIU!GoBqE2diDR5Uq9#*By9!`Rj0zaH_rVwi#K`?QIeIP&rx`nnr5@
zylr|aR_0Eu?%;Xq)Eb}jX}r}9!gRWJn0m95=2}mBr0avVv2=)jTM+#JSUStFHoInx
zLa^Y}-QC^WnYz2XySp28cXxMp_qXoeQZEoBf#4)T&gyl(ANz`sgaq<DYtPI*_h>z0
z!y!epYOB&&)e?r!HJYMHJF;r+#^`%Bhu->=U9(1I)8jL;>A)u0bOOCkjq7LAXrBKZ
zZP^_j|1ozq{g5-8jzPQBG?dvD>$B>V##yz&K>U^Pa4?24qq$NBjjutChFsKl^c3E$
zGiZgw_^*<${WB$_22IF<mM4p@W6pY0_8zzK=$>4k8h0$(#i<$e=HdX|p6;&|-f^t5
zr#|6<#-ac*W)b>SYf|&U1N;cQtqX7az}xHacId!N8rld?b9lSCF!M_?Wn}g$toFfA
zg4MPZacp0-QC~7@_<DNl>7Vy*!`>)MMxE(IFFpB*PK%fqOK(GgMEb~zlOv<X(9?x;
z4Ey(c<iG0jecAsnCvw|IOMUUtBsaX-1TUL$zIq?tR)@Fs;BC(T@pfA!W{JbwF%A8+
zXcIr3%=NPc{M31<uQn`?uSH)Uec!`JpS!c~gSDH{{S~9Xt_!*Z5489h{&?xfztm3P
z?Ll~Z7T#vrfk#e3G!kt*wGR0pWuICB{h`9!`8ABX0p8|)OYI2W?tjWLJD<ETye+VS
zUWW<f)X0Z5p=LFf81};`{E0YUj3$597~Te4a?!jqT-b}csKDF!E(Tp1V9-18_C35U
zi~gu9wbH~M4yjn$A>;fV@-*EpoeDW*Ok;-}-s_MN15#vC#S|$`eR9pK6nSHTzh6`2
z%iR>YyeUQU!Qyri4jDbpA>}hryF6f*xbk)>kjY6;wkDrB&`F1*;c0-jb|QM_)n}Y^
zRx9FYc>CdVnpA_g-<zNjN+Tw|m@2i8rplDAX|ni=S<+jZ<x(xP)GlR_5%r?w>9T0q
zwli8xhoh;tM$5OM#J&<Mf$;Wg<ycu$ELKt@W8`pBj3nA(WZ>%<v3}q&Ut^@#-xxXb
zBUVx(;>9Inf+So@kXtVkqzAlR`zBG&S`)>-BT+tQN|Ngxk|fT>CZ{58vipQx5<WQO
zN&Xah>`0ZY%TlHI9lLbf<B(ZysnYf`b>j9Z^7N`hju&-EEqLpE+b*Fk?b4z|tgOBo
zBP9c3<;5ARJX~#+0IwLSG%QA*R<+6$wMtS$s~nzbm5c{iN30Tf+$!s*S|vZc9XusL
zT!zNUBlZtNT;s$!Dpo27$I9GeF(PqR+4|BdAJ1E*(lM(X%^o8kKl8aq#mW4|@v`S-
zycl-I%al)XvbATN{3?!)0S?Z{z;O|W77E@DfVU-?p)vI;^Zx$BZ^++YFVAGb*Y@Og
z%)98>fw$AH5{vAoXO`RR$oF{Q173yOUf=!kuny3l<enW3;V$($W32!k_S9cjR>lu$
zqn~cr?~NxZo=wx3Glkc&8@>=JyU-+e_tYnK*-PTf(Ffj+wiDOF+gk(3>%v$Ad7C}u
znTME-nTarS1N@Bt=%QO+yXancy9nOqzQ7D7c<TdiuWn=x(^Y&siAATu+efSiaPh)(
zJWrmwXfqh<a?(X-5C>P?<)UR_YldaiP!{u?1>~Pcus@x|b2)w%Zl$*k-sXU}!{F^7
zj-wUuHVNKteZl?}<`;ptb=e0D8$nD;e%aKL*&Hdv$ToENN$8(lV6Gc|=kWI4MA%MD
zyA<B$g15`ztvw_COZnZjcR6y_74d<qO0AgwfeysZk^FoXj4tSazsf9+oA{yI(L$H?
z(5A1M<BT6icxL81wn0a=66S8_`fZ~=spqNJ=%0E?K6&M3PaOhp8)arzC-c32z}nBS
zHXkby-oETe?P|1_u7S5H=tcgr@Av0iG!7kfxufW)FVee992|I#+n4a!c<iN3%j03z
zmGjsfZ=FTn*+egI=?HK9chOU2#Gk{+Tq;=moIZ;;@b)dd-HC@)v*|uM65fuj=&QMz
zeYF)2tyJQbU(Aab!CFz6T;Y5_y~AGa!ZPxS_;x<6f{(|rjCw2&o*Q{HX@|?y2y@_-
zLO=HhV&6mXc;PK-BKWC%L95ffMMnL(Jd^I6npuN<vZ!l?Y%28j_M4bZH<zQ%O01bq
z-_rV4IW%|49NK(-c8$)3H-53~+RscM>Eayd8Hg{(W!InZ+q9e>mnm8G5<M;t4`$Wd
zaJ4hMEm|zAz9F`rvkQ;uGw87c;a+v}zRB6O@vv;v?y_n}IN0@UW-V|uv#vSFTvqf(
znO2jd?VDMv7SF6%ebD^cGHD6?lgvBt#-7Oh3v%f>{qW$9WbXBPU;He6iCLK!*UeAU
zdy+%#i}%=YKOMy1vh-g+Jp^ylilg7gpQ%J<d=YCi$CCZ<?r?f*YSX6|icW%F_bYel
zcOSrY=6KFPGx7MkAD)oR9((JnSD(=Tb)9+JD}B|({<l#{=3~YAYiC$F;a-5=u%ORK
zgY)Qx-SJXyzJdAIpP8c#Z#&=*cc~ToA9#D6no(`~wVel&UtzDe^rNRP`0A-^63MfS
zBzH2?OXqI#BF79D@ll+dff?Wl-a6%#x8_2#wCS3+My&GIeE3e}qHp|nGPCFWnfXdS
z<t>jd)t#9nQ_%|?qlW?3wjS)O0&flQwgbGK9gk0X6dFJ?c^otehuLLisz!Yz)~MP4
zkORV#&p()B#fA^E8`?s6+YjEBV>PF*Z)JIW3@^}EORs@9I*kw`KEXEnPaiW|W(;eo
zt4>0L8Gge>M{wTgw48GROx-jJEhgs>YZ(_!h%{*7&j$5E|1<@i@cZ^?m8y|TwmR$d
zH_n=qTIvP*c>45k)=S+SvYK3ITMzWL$#(fv&>^4UZLNb23G7aNvP_B`FPkENPNc{_
zGx{TV`}t0a41u@Hho?wHxI-R{C7&ALki>oXDwRo=XGanwMB^me%UG$Mm0a-I7)gF_
zmDk;^QtEWH{Bn(wsAd*vP|PAJ(=2lGkXeeKHA|`UX7Qh37N0(5+1bi0ktNMi58iHf
zjg~tLq9tfYw8X;Oo)@E~cjXxITOA|Ia>h!ba<Q_ZP^?^sx0ixr<jvC<=?iaHz}s%{
z)&YwyosN;|dt&6kgBaQ49WN;Z5~Nz&M9H}{QA)zwqww}bu_SRrI~?w0llf?=GFP|B
z@ESJh$69<WS)Nx;mK5UGJjdWGpZfxTo2za4J6v%{8+cnB-e!Zh&EV}zc$>T;MuHz(
zCFX=x1}}xJ&M|VWzg14a+iLLk7rdR-1ok$zN@XqwG_}gP$yV7rIZ<j2OOW0};^a!6
zIC-5OD^)DS?(nwe(HOCZSw$aOC3U}5GG(*MAO0R2hFN8-XPm4n9VgAp#L4#Zak8sT
zoMf&RBW2m!4tFNc!`?ICbAZkzA6&bzKc0PlYQF5J9quy2&ct4lTFHkz{@Sd&zaFQ~
zHV)ptg|q)=)a6)AE_jl^2KHqB2D}Y`!+Y6Bn~RW7hPS@e$U*-k?~E_tGI*Uv{-@1j
zZ|xgy)KGZ42_4YYXg59o5Wh0M)5BW3t8XLxu;`QhzL_2#^2a`TTy<<b9EFqN#FeFp
zOEa^7Tue>Y2;(-x+nwCk2p-mkwflCvXy_*LIq<eCyln?>tzX%9vf9AgQ@l<lc>4>^
z)?Y>bhsOnRJv+xuevTCzyiJC;SK)0ncsmi^R$A|(uZV+P;q5GVoBN`R4!MN}>p%9d
zyzT}1)2@<FI%js#wn@}+9Pk@0S65G0{psVXEdyNjQ%<~OI=SkSQFyq)+ZBW9iK7o`
z$jJZK<}ZY|{ow5&csmK+ZiBZ6;B5fB&69`+#}79(z}v0#75Six9GFH;7~W=qw=1k3
zdK%tdhqo^9cHb8K%i--(csm;1$DQS#n*D;Oc89kw0=#q}8s%Z|_7}P3UHGwws+T&G
zPx}RLQ|JdixXMfG?()*4?Or+oAC88HytE_v<yP>v8@&DR0zOsnHVV&-obdKfptqKV
zw=3Z7e_^oK3rz~z<^vw+LSb!FR&KMmnzQ@p;WFg>Hv8y#ysMhQ+hXu`NDV*z+JiU+
z*7gpdvKQs62Xpx89Jo85bH0@rr*V1mhqe8*R3<;I|AuqyO<!$~7l#hRPx_Fr-m)>*
z^d?@?@OE@P=03vPCGa+oT<AW$74p;XX`F=DLJsCm5&PwXd7G(I`r!2txGlSOL2Gm(
zM|Qo&><6DuS=CI>>2_k{!{m-XbGs&ePK)7fqfgA4=$=)Zti)5=DXVT`rgTkW+xfe*
z=*e|i^uM`yX%5e#Ig_&K%HvsdaP`dEMKWne^0*t=7c_LvsOLZ9zeenrk=Qb=6!p!b
zX!qIsmd-+-%op~nXt|fP_E!^m`VQ#CqY8NH&TMdSwU0iU&)*K-yzfUIVHmyX@HUd;
zsx8cQRODFfOnvjHzjh~%+yrl(1F6|OCGSeEHd87dEEj#X?jm1p678cE$@$ON0k=YZ
z^(DL=0&gFXa~irF?@)Mqk{&S6GI&P7`thUjFN6E(<fHD>*Ls84b7)TH)^+yN#4f~}
zMfpFGuP8^JqcWQIb?%;8p*Ov5?Db}x@zmCL$hDyP9Sd*s!P`mjc4;^COngr!vF0Y?
z^A2yr@ON^?Lv262-B{I|9&5Zy6TI{=G3lQ>Xbt9Yy|0%}N7v+PMYj*jb3G@AgT^WH
zE}B5r7qo$sFL6mPT-0t){nHVxzAIdf#8(~O{zB7V39s~t@OGjL#|QTZ{$^%+OXkon
zW^Mqwk>Bk7Ys1^u^bx$_+_VMW?s(v)bxWb2oQ)2XeuqlbuD;J?uRWf+6z7X)BiYw4
zanUBlTy$eNUQ3@0`u#k5<E93!P2D#1kFz!+wl&<r7oi#X<na!1t?!Ujb;uHU+uzM0
zi{M4w0<Z+$j@|E&@J9|QZF0!)(kZffXo~D2zkGgWiWHucA|Dr|$fhAFauM#l9^;Uv
z#JA6hZ}Y(0&F|tR>`a{8`4THWxnkv>#>kOvR+$QK6SG9iqv;me*3=?hGr`=)W*M{9
zEX~iDWgE<WGTtoPdzqzML$l0^kCaDkB4y^%aGA6!T(bU+k~z7e#j9boj2~f@2|r?_
zE4&Ra1KU<n7tI+ZqkdYX|4WO^K5vm~$1PI(qDAKKx5&f`7J0ujO0KwBr6alNuczbX
z?4ty^<eDh+`X<V~ZLqg^k~D(1Ps!D0&twz7N;YW(Ydf(<!rPw#$ug!Tdy~s(k@?)C
zbEU}4Hwj!ONu8)fnP5(oFHaMte%Azf0&kldqUG-WC|P?UO0Ky?%d_IqGOmbK=EK{o
zVwF4awlka^18?84=EB<*@OJn-t0dIqb+RSM^i{Evzj>^Dsm5itSecM3Mw;{7m>gDV
z?g6tGMoZpo(c-Z!S~?%L%AVXYl95$_#~pT$k)Sd$;=*wq;FN)$hyb1PE<o4A+Y0cu
zVwj&c+{<}wnV*`N1vvptOLo}XpX2!_y1>gY_Yk~Y32(2!(4uqwbvpUKSMb)ste!q0
z{yNM#K-**@XPiuK`o6zDhspmn_0!+PG0pG|f0^WqcLjZ@8}P!0w@v?Z(~Y;?bp9DP
zy+VBGe;q!-+kEH_3g&Xvp71u;6BoUF4i>VfJj(UO)LtvV(<(5y%t^lY$6WOH0W^2;
zHXpog4NFZ^;U&zRGSNj-siCZ8|9PVyJ>~sfG-8O0Zkq*D*U%?Kth;gr$Im<$jhg_2
z`C0QpFq!9dWbaxE-X`qi^?3cs@b(Pvqb1DE$?vXl2k%h0UxPis>JS$#1aI%Akh_Mr
z72vI}FJ4vt=)BN&y(0d7Y<1C;`y7vGs1xCBW_arZgKMC@$k&n24U-%2xx=XqO{7<N
zE?Tiwv1q^0HkKnt{dox<9J}Z<IPZop93EIE>c424+jgc0nBL&zPagVbvr$`LGU{Z`
zDPc9}salFwiaw&wL7qAt-u8vJ7Ycf*F)LLx=6ZF8xgUscAHdr;@OC`Bb%D1(nXQy>
z5p$<FH`d!qY`e`%a~>hig}1Tr*AEtt;c?rDg9ox=iHUy%duup(<UOoKZk*3y?bdMg
z&XMHza^PDDZ?|qChj<GQ0^+Kc)N=1M@YDFNe9x-+={kd--m~D*g=ayI*S@;`wy(ZA
z@2lqv`RQDE+ZEpK<^D<Vwg9|6vIl-&_0<(?(9jSIFC9Ri+F0~^^!L`B7ofE_252ob
zIj>J<(iaoy*`)^AJUx?I(Gmr>Wu9Bt%o^GQUupVU9-Yam8{qArJz3SG6&l}cS#<?-
znVxmds?H;`YIS&99p3sH$=jmS%}TFQz7fo$IhRG}!rM{nVeir`dV(IM@pZE32Ifyr
zL1Wu*9Qoga)K^ZC;~}=J%O2JR-B@6XuU3iiMaM^c`O#NT+-2tZeqWtQuJqC@U%duz
zJ>l)UD5HL*uDI+xvn1(3TYrrHlgDV?6VSes=P0!VZ4bxion^#|=+@$3arPGIhR7+T
zo$}JVtYR>&@JsyJ&f$3myHD=*(s$_F!|=o(oRb;^mwmSWf4p}SzwqS0p8AgOO7*;G
ze((BdHFCcr(!I4G{wVL!ulL-?%unWh9^8W8+h%W_zma)5CA`%z7*7?xYtx;`rL-V7
zi<UlNFLQLBFn1MZHGo~`vJq!4#P4N|mwMyJT5pq4>$7*<y2xEO&!bm;3B7ZC#)`s>
zxyI}z+u-+(x6IN5MzoRirng5s497=p^wh#f>5;qUslVav4Op82uCIo-!{F^i_M+!6
zGFy)RS<gyFO|N6r?){h##$I&dPjbb0S`0KBbsM~0@zSVKbBsD_99}m3zEt(l3h?$m
z@$FSK7yDb0i>mLYH4EcOT!v$|C$-s5oO|GHM`GLQgV}qRchO1L4eCR`%8i>au_LwB
zckr~CL$>vE$R+q%VWLCwk3iQu&>>U%IAnWWVpDh<25&dJ(mQIk%VhG&U*T=})efn9
zoS%J0qg#@AmglFgv&*m*c3IHSE@cbaWlO3}TEU=)*OTPL0K6P(;SW4BL4u~mOQTJ3
zQt)%EOmf7??i?}l_Fw!4BaFMF@nW*j&ts8U*(_2P-u7EymL)mOQnZ^{E)F$Ih91Pb
z@b>D<NXgtSQp&>H$vML1)$cHI3k{Ru9m1vbEcp96T(X$CJ~~2f)-lWGW@btFVUd>b
zb}YPIbkrhYdn{6SvqjExd;dL))S43|%~wZB*g>o8ERrDk{S#&L)C6%ni<1n;5@p{Y
z>a5k0<tjDYbyJe$)V?G+{2@t>g(p$s!<TVLqMS<Q^Ra(;GS({9FHkq;vz~k&CskO6
z7ID(ERIKD8&+C#WS`NWm4|wZ17$tcMM2k1P{awK-^^02Ncpa<sYhaaTu(oPl^1|##
zM#0;lxmLM5%_@^7@b?-=9(tlxHuBhmaQaYb?kjGUll(2C+^sThQMA0^_oRnKNkw>m
zsY5irB+)W-zE%8>!tyLJ(g*!<O?!ZDjG(6S+g~r9Cs+K!Pn+YF-+3<k;YohlZ5+q_
zA-H>+yc4;nkBQ7~zv!pWUioQObPXohJXZbnemgW%^ug&Ne|;M4uNmR(kBk92<fFg#
z-RG}s4*Ti&F>svoY6}M*!>icS;u&#z8h(>zYL@7Izn-M`W;ooXhB<c=dIa{69^{cn
zWN>9>vx`0@cN=sVz0+FaOLDng;O!oGXF5S{7uF^d-)?}lv&jt?fswa}RlmU73Gj9r
z-24V_8xC;M`@LM$40Dgb)H&0+B(6QP5gin~JsIetwZ?EithEp1XRtSL3Ot6%iSV{P
zG4A`LJPzKLJr7eaQ2*t9tbL6~H@pq{?n0e}I5)yYSEakCGraYKx4ZmYH7C3c_zjES
zy6Dgcyia&t_6&M2*!vW%^-X#Q`g2)NT{UkzSA7R>bFFvP6Zh#S{)0yBjjOIDAN_<n
zZ$<RWSK#dr&S68M{>@Kzr&ky+jWh1r{ELSMp_zI|J@;ZcqrM<_e;3|b*y~T*fal|B
zPffwQsy?yq89#hCG84NJ=RWW1rBBBayW)8@9Z!sPv+=&-oVb>?f3cTt=iFFs3zvtz
zG}jp~jXB7yC3rgu-Y$c+eObrh?FQ<;&B#4>=3KnS?ycL1Ykhw+ACf$CUo=<gdC`k9
z3vC~4xnT0u7V!2i-;0g#_HR{UCV1=ii&+k6s}f%L>LPfXm6&P?OzJw&SH}@+85;QN
zV>DP-E8%aj-A9+g+hy=}9lV_nZ<oQ_ZIjT*qrEzRnE7kD0yG}&P#xFI+QU1uHioz1
z@b&||ebbqKUU+*K-u~+|3R#c^w!&k#O!_b>qdt$sfAn!iypb|$a%N^~jL)D>12gC~
z^jxQ!WYAr+@rGfBaQ8F#f}dw52OiUJ4rEfN8JVcb;?n?c*VALywJy5Ear7qd@Yh@L
z_7Cx9V|ZH(-hTa!$L|MXNBp-R?IhM*!G7Y3j}AQVse@N}Y91?``D|2Ub9y|fBd&l&
zo7r34fwz~E-L(ue(zY&wTh+aE7JI1kwec5&r)BtGIZm)=I!O(_zn32A1AE(f>HM{L
zC9T4ru!vEg!r9e1@L8cI@q?Le9nmX}&rSUI6px3w#MgOzb@ekJ&EV>--*}Bu*}Zgy
zQ13s^|9*qFZh*JT;B5<dTNB<o1`)Ty+uE<u8nriSS|OuedxQpoJ@t*#%w9i^uPn#=
zxpB;;nMnNDgL$#^v^ozYzcrGXcHDn~{7nILfjW!%A@kjJ$wK;F>BlL8KkWf}Z|d~J
z<99IrF=P0?PhvjLRCNEmR)Z7h5IN2>PBrMK=?2|_S8rePPca;i&smR59INZ-Lx`ki
zvy8uES$gbznGIGA@0>T}*{A`{SNeI-T{z!!)e9WwZQaR3asMRl&rH6_mKXMx!uz-j
zduw>RiQgB<KE43FZC!{y|B>`%=1mi?TPbn{-R!Pc4#`!4c$GZyZgRerCJ?g@f~)=M
zNA2a1h`n|xjW@%!Zgw&7n8o+(;+MlAhv4m^RSs!I-qb`acRwe6uH=eycd^U%7Iqn}
zcBv6!ld5sa5*dLPFuX0%BS`{xB}#`K@$#cGHN>@XG63FIi;a<{nPQ|UvD{f0(Q9wC
zRGVm#F6Yd$vZYyqoy{`wVWi|e7AcK>MaZ_35whcQgp3&wDbr6yici-_8M`7}v|zXl
z{1ztd2Zl-MzG3ove3)2vg^Ae~CQ}UI651qEZaSGIFtb^fS24@^(-tWRZ^O3{@7A@*
zjPe%wb<-lI*->(<cZ_8F79*=uS;Jyw@Xdd^=QzptFisvWOpxUd5+y&H;pOO#j>6kr
zH<M%%ygj}pQKEwrWQcEq<bt;q9`adW?q-<w;y(G}o8*nJSn<NJ%FCS5Qt4fk6oa?V
z&PPea?kEX&iIR>{7AX{Ik;=a<;>z_08KdOak7&7A!zx!Q5wmBqO6e?CY0|<fKgL_d
zZxpMTRVrq*%6onuRKhCxxZRw;Wi@!)0N&1rxBcO5nPSmW3*KIWw}S)8TQ9dt$t70F
zvok>F(Eobkp}$_b<*$F4Av^sE+Q50lv2b?EMDllI{q)FoKix-udFgLIeH%f&=z^a<
zfVUmttxE{!#{hqgZsV^J@V3N3`p)6)LkII<>rvY+O3z$8amQ}B3~$@9Z(ojYe>*%G
z9Lxi-(<kE_Oz!kL9uoA=Sm<%h24D6Q>t&?pC#nwqQ1EsRyj}l@@AE<Q$FQ#r_0xr{
zelY6;ynV6HMGJ36GdvnS6DtO0T43sva@13EqI1fR9=QhGZS10l+ri`B==Nc6>@>8;
z#KHI0a=gIX9q{%)j-4QW_Kmd!-j0E{L-?IZ@V5FM7mXr^bmKUBLwNg7pY?_{2<`^L
z-Nv8cHoOgly+1xu7iM*T<)Z2A1<tYd^1M2{-@T{#{XB0gm!tU{nc;8LbKYk(HDW$<
z8F;&OJo}9)uA21;`RE_`UDGcdL`~a^I&j7!Zkn$bJ{;77$3LTYh~D3B72GwhBXcaj
z!rikT8g_#@P*2#S$9d@YJVqV5&Zt3;IkylC6#NfQOn6%y-UfK0pUR5|WM?m}3U8Oe
z+hOo_J-ju~r$=Z3aVs%yYj~R#-d>@eI|<f~W0l%W+{ELqEaW^-?s*5iT?TJQkar&Q
z4L@i+BeQMy*5bdswKKeJ0dE;Fq>XnIw=MD29hZr5%fTIZJFydWTX=iZ124Xx#I5jn
z+$H>|&(h~{oPMGs%y*gQt7G8pnhd@=1#L|}>UY0jLcuxAn#8y7JI}4^Nv$`)PqY2U
zjADEZN|yCSfA6Q2v(tlwPN7sX^<Vl;a?|IQ92TIRZUku1wE(qIQ%UKDPktZfcF;Fo
zjlPs8_3>w^fRA_#oaC{UkKpl*p1(P~tqpIB!Q1NSAa1z@Xq=f@H0)bj#rtc|%=m&A
zruTShfNt)N-*``aAEMCaj}Or5@b+InOKyBiKfv1nc)JGP?s(&?{x{fXz}rE@qocli
z>Hv5<H;x$WDEq6SXx?X{p_|}_7c#RXPvV(^&Y^Y;-tuO5J$Vg}73wjO<Uegyn86CW
zuFPP58$L(}-{Xb$#zS8-EA{VFyieCK`)!ejdca#(c-tbo5&g4=PJ_2)>GK`|Z%f15
zg5;`4Q`2lqz4I$NqCxkK+RvYvWN+~hK5W$UweV1Biia7i3fjEQ3GDI4dT4`8c$uzc
zwk7*LmjL?1J@7h&v-8j(l(>Q}pPbf_rta#5U(;-OdjQ@(?@vrSf<AQoBbE$t*PU(M
z^=mG7?M{#T?KC(2^c8R4W8`!={-&_6-hP&TK8~T?*N7|0dHuZZqBDJ{>nt(o<;e!-
z13BrJ4)_bh+m^6)1uF}@t;q5DlVfrj`|`ih2F+tLXhbRUUmo-YCgZywW6&*l5-*`g
zX6znkt$W^C-G4djF;@RDy!Ywp8F-m~RqC>?6`k?6OqCTwQf22%dQ{Ldh2y_)jb6|J
zu+)vbXZ5vcpvpKTF4Qh#-`Hi+bGz)iW|u~n?4qaaQho=0t3&M))737;`rE~w+T&y5
z-Prc@xK74*A6-!Woes$qYnS>7HYr!zChZLLi_+(Er%keaqnD*jSfb2an<yVHCdkhA
z337Z>yhK%vm!J)CvUG8*6xkOenKQ&l>bYoHM&2`iT9hmvWs&ZO&C;-`S=`|5u5*$2
zIYr8rpAmBYM1)Me86o*TL`dC9k+Qr~q_o=*F3l^1<LMtJgIk12r(R)FkjqAe!{p;9
zlh|s9OLEU}X*@Dq8u&+uqk>rm!Q1^uERvDfwso9YZWOS{O?X=x-fr#|Bl+QNftVP%
z-8ELS!P_+U2U{Pp&=^&BN|cFcsyYlJ{vDhktKKHyJCPuME{U=$SE4j39VO>IqvTbk
zC^_)dBEv3Qq!nyW=l4DONG=<OZh0IfS)N2m+Px^r4sScIiIN{*EmFa35pQ^FBMwjR
zZI-;H%(A0|Sxz=E%g?Kk(j=Q%?sYfI(@$op6lIo!D=e~}Jo%3@QBuBnlvHU)etJ-p
z)SeYZzg3j9B}aaY-<35uN;(kVPMJk5cMq(Gx6|Qm@?xvV4tizpQd4c@uiMHpduIga
z!ol8J`LMTMMsx7jnLTX>AHBVba|68HNp5)_v2AYpH`XqOog2`RKF32X6LB%Mrijt>
z2pzx+8E${N<)>%<cx&FL)Lu?{YwK~|nla5wNB>5T|J+Nh_r3Hn=kC$V*@KfuT7C`R
zGc-Z5S&4P`(zk{d`A`>PyEd-ci~6UL<Ky4ljN)TBrp6O@5_=kl!9>_Nh}bpX1Q-3x
zKJyu@?Z@iC<F<5ze`U!3=HOV%LY+64i}ozZezcB@Hf@cjs=te#hqq-{a=b1je&uyW
z^&-A)PV8BaxD(#Gk3fgT>(*Gu^Ep0i!P{Ws+iP&MAjkT9)^k={c)JMRe)`1oIG$(0
z+n=zu%Kvzq#(Tf8)kT-XZ0CJwDmZ6!IPIeSPZ6)f+j{VJ-7|9J7PMEj$w9;0rSP^X
zb>6YdVedQi8Bg&#eSrtZSFZnc)k_99btTqaMXmR7dpDgtp4neXc$52+ziZ*H#W&OM
zo8qny@_FdjLYzzaJMSCjq4kJ2-t?khw>I;F@xnM}_S9EK{uji#H_Cfyrf%ehCwl2d
zIQjzKwuiU(;B8xYTV(^>C9m8O*493Nt`pwgCI9@A$CZJ%f$+9FF>jH}JST|%GkU7M
zJE_GzL-RsTc@wiH@{(^(;JY!C%M;8m%3gvVBYZSk{iUDi0P!6)#A9EHTZ#21z}tJo
zwoW;HH52v5{NK@9-9j6^li5tOiM<XnLk-V@uk<1P8snqGsBOBDJIP1PyY`Nc)`hp<
z%szT6jh<k*J#hgwP78BTsgGPkV{@A2+!l|HCFH)DmGqkHMc5nU{*135-`%s_xv#3P
zR^%9H$;wulSd<#|^q<7)JZ|$>Z=H9?TMJ+J*0@{f#G3GaI^wsm-iMqz^-!LBo;s)t
zae44qUwtx>&oR|k-@(?Kc3<5MZ%eo^lNIjGZQ!9BD#F4z*yH1&qljI1z|vXNSv(H8
z>FObFdaxSDNhSKZxn1-y^?ugPgKj$E06ps4+%%eeM*)0Nrj>Kodhj-6g1c_p_y6<M
z2K{l@dB5=%{lE-2e-Cs<u$AvZANFE?@YaKVZX0{EWa8Xic+C9h=Aq47(jQaLQyaqD
z%kb6(-qwP*K6gBH$psHxjgR<|?DVDOV16sUXtkQt#|Ljcr{d?o)I)vpF?*WWIlsHR
z{)CmAm~Hx_jX@JQ1~Po0M#Sft+l_q!yv<I`I*AzfKwtW9xG$U<$;(gZ7h=#katz;a
z;dm~G=IBYP6#GC=2wq6P#^Vct#`gr;qt{3AAjChS1ibAv$)M?<o%G;4CvE-QNh3q(
ze`$@c|32=AlfChGINykwa5)TG@Hf5?&G1mF@1(c$<B2=oN&QPXYt~B6s;!;16ukBP
zi7zp{?F4V#;O%s3viX@Sx1~H<rw6H$r#5{$-l^h2ug<(rDUy9_iuCVE9+<fHm&q>O
zzuKh;oV^QgTXQ+@s$Gi1+u0}VV%`cz+uG$)Q@f1sXP2uN?Q#s>8rnHz9lGTXyB%^C
zW;d))f6Et}T&ZP~+wgYT>0~KcF<Fi*N|c1EiE=*yEp3kk8SjTqYH*y4Di<dK8RO)?
z=dtoMDn{z~#Yh5q<WJNyBjN4rkrtV}i~KLV^(Q8+3UB@EM@rqG2r08JLSpJf$lV{|
z@&%55%MdQFeZyti6&O1?OuEz!lR&#kPKKN0=R=dsUu%+nwM^2ZNw~Ctw+rWnOQSR4
z@?U(obhSpx_tj?EWHyU?MvGL>XOWMEEzAqBNKfw=$y+p5O81DB+$UqjfG(*}y*ODt
zHcl?pjF+pM6QmNneTS}U%$In{LEQVokSLRLCQ21}yPjCJUnqI;D;5d4WRa@Wh`+(x
zTky8Oca-eN5G7r*L`e-;xjBE7^l43P8FqHwZkBJ<sCBSeO7|y^UdSxvbC_j9|42Di
zD^k93IfZ>jVB<(x<{T;e>=E+H93i_yBjlPXLaf}rbS6?F!y+Z5S(MyeA0>V7M9IKY
zQF6LNlyt98F2F5ZR-}i?2lsHPb2&^(Yzfd6@U~cEf3>1ZnOcvT^sUjqEQ68g7E;g)
z9%xOk#|P%xbobRQ#12ErGgqZw!$=HVWhwd=G|b-g&$e!bw;H_7L=E@Y8v1jm`D-uY
z;;1Lob540{p}Cy5Qiz{Eqsf7_U+DR)4R8PckGHeQje2!Kw_Ao<ZfpA2DiYJd*)M6-
zT9c`>-a`wujrkcPi4Vu3iGsIdhmuQ%w>jYCZx~z|E*9kH?O6rjZQE{WoJtd8W`(x_
z<bL68H+UOZ8%<#|_Qf5DO^IDIEr+{{$ua+rw+))|e0bZnJ2CSRv{wAS!4q6`0Y6_p
zoBZ?wVrL%j#`99v(~q<ttrmGFcj~ymV5a+B_ReR?UBlFYFuD_*oki^3k@vq7-bTRe
zrhEUNwYlMK<XTwE<2D7+PsC^K9Z3F~nowu*Sig(8>SlOb7~Yy6QR8J*CC<(GlU{px
z>kMxzasGV>Zy%0xQ}1}ZTV3$GEJUu3+LJFiw^NJVwd`>;Xr(-~7rdRCmHwe?)Tyd6
zr>Kjk?j{eJVHdG3^{zQi{BOFTj~dMkA-o=YOrXvSdp{Dtwu847;cW+a+X~*6fVai?
zAFi6|rM)|lU+(UuSE=3IfX}_)t<y*1VR-eH{@w2Ib~(HaLlYASYyZNk2-c-D-kNk7
z|2*oHGr#(1$KS9LPL+g5tGB|aD(I$4!#g~+e@5YJ1wUusq|azC`qbJ!TBnhZe(A$^
z6y7?}={%b1qrX@BkP|0=ZR5QPnpCtmYpVI^yQTCoQ9CY)Ua$E}zT>U%;%MWe<G2)Z
z-tDP%PMpu2aWpz-*(WS2#ViMO+Ifw<Z(`&#<<Ze|Z1nDp9vj}e^Zs9=r7e8NTkZ5G
z_wVGbPH0^_N27g37nHv(dNNk!*8EMHpz$fo--w#|PI3#;x!J3H_T_)>p<VEGUsTUo
z!=K@czT8QBjlz%SJ>M&Q)J#iUwIX}0rdE2)OER|{K3(^LK_T>@eE9#L@6JQt_ZDg~
ztjF-S*H3!Z;qB1j?(_-4Eb_8DSw+al9y>)0`G9$Du>EK|4{g^GHumw*tkgN*hI{CJ
zeEEaVFk2kn-Ym!b&Ir6pgP6BU?s_ya>c~I%uKae>hGyoywc~rW(_MR>z$@*!yEgjF
z9N2Jo=8VH#UV9Y0ZO`wCA?MQ?e}^)D27ECM`Z1qD|Kv6>YuZU$h?9;Pg_q!HgH9Y_
z&{jSM4RAH+v^0Dm@WelS(OLhSjGni$vu4N3$1gfnoG+!y<EN>Tc`kEMOVjUC7GL;A
zc-zvq>YW8o+q}#ozw4wfU+^`2>!e=rHktm@r+C{&9Cy;}N9jAQkSa@I-OD{TX*s|q
z;SFu_h9$dgGTVSBA+^-mS9wfgni%lYoxzONdSNN@^E5eK{}k!uogy#rU&yj6MT$*M
zm3CE9Ww&RlY>jis7eD-$QteV9#4bVjE-mM>(ieF8&@N}r*_i=HE_kzD3N5wEo5}R9
zjJ8WyYvNjXI~4A2J8PGBPkFxCE;~y(B!PPGcH-R1@HPS7-m62sc674zzM3RH^mi^_
zm?&vg5@im&eNZJqawo>g->z{o5shp_hByg+7%NLcV&o2eFjvE(rEpY~WD2&(b>iEF
zJIyk>iCL!GBgLyhq?}15F1;5aNh>3yOU(#54R0sZAx{i%TYU(VNVu_RVweo96DG6Z
zZ4t9ctdC6cY=uc)v^L4-q9!TYh{wX)mZQVPh1KPGgiQNa3yw5P-E_08$!U=Z@HS6r
zi@a-Pk-v>%WJ>o~nSV4^qD#h!AH3Z?Ax;*Th?jt}R#~&jBEiJOjk-tCZyY6YnWJPV
zs~SA*;}j(^A;ilkEpmAs8tT2|hT-kK+ZHjVS!4^mtp#sm;O#7UdnaF%tm$fzX7IM&
z82X&r5Zn5iWn)^TWQ4a};Ox)Jk#asfLfp?q$lqoW@_A3V4B8qd89hU#_OlRab|ysD
zFANdywjuJTUWhzu9U>Kuhe++-p_0ebB<t$)H^>z(-P6J(cjquETf-#dUj|DB`1jgO
zZHG8y{4VCep*M|h;jI(qcr%NInoNYZCO7lZLHMHW?c%G=ZD<!q`>EFeKYamrV~6=^
z_igl<lT%J=?yn!<?byArcBQ{AK+hZuhfkx;I6l=&3l!rV&G~#kvo@NyHELZKqn>%d
zET78khd)v`*+)N2BR8!<O!tkN?ZIkzDtfzWp8GD^0*?L5+t!3vwh`o$VQhu=FsUp#
zT8lvs<R(Uik=5D}qr%7*aM;KiqQt@(U~Oh%V0e2sC$S^EEe>zfnlKv~-fn@nV-~uo
zH+<bw8NQa|wQAABTNhr}BsQ+#qWgJH8*aDhO5T^cZo)j`X?Pn&OuQ1_Hh{OS&!V#;
zMturz{ow6mZm)s2-QZ{wc-s@+`mZHl45MWgeeZmh<NVy2%Za?sclP^7-l3&VB4<5|
z9$|PJ0B@TX#9O=t+Kg@ZOWtK}(L;2RoPW2HUmk3B)yk=^`j+!=U_m#{ver#+$1>9h
z-fqrIe=aqtUW43qdm#D5%gpH{Zd!%z@`f9lk#j~}2XDW0_SE4E$O*2;i)u4-gm+Tc
zJ&Xng-aZ<IPMLn($1vA8g<eIjzvccVtTwCw)*faq%o_xU+vAY|Z=dm)asPO`6Yqcz
z=t*-iKX4DcgK?Fk=*u9UJ+Ka*O{XUkb}c>zr>@W&32#@>e`)_kp1CIZ)%(79kNWBX
z;=>9#m;oH-qigv2=&NXH_xP|UWWGx!*jnF5gL=VPv^=Hzd+TW8@H`jM;qm$E(t})=
zy+oZAeD|m`emDzzi9-wUUY4*A^x|?9yiDi2ev$V$bv~Nb4CsTZG2^ffdxg4uM)rdn
zI6ht`Q?F#EVVN#?ZS}%yi+#kR3f_1|dF!|-)G&|GpUi#E@HS_PQ3E2$;a~C44$N`u
zKhr};j`GmU{XDcpF=om1antFei9ungBiKok@Y|aSZzmH^orkxx@FHl{ff-(Z&<DcX
zZv4NiMWL&Mw~1)$l%BY{g<bU}zG_KLnQet1Y2m}v`soWjj?UizZ}Xg@21&eIvk88^
zRheBw{8(^>yN-j6_1JfnBNkl%6Pv=@ntUg^6!g$F!#EbMJ8R$_IG3I8Fnc_VBXrvh
zzC+|zoIL1<C$9b%KevOoJ*iz}UW!I<H+9x$%#ne&-@D`Y)}8NeKVr;U20czZSj>TU
zz7dZHUxRuZ4B7%;rAzTnnrjs^SgH|I;_0yEJ@eH5OOp(E@tnGlCO#+Aq~)PBiP@GW
zL9No{W&Jd%R05Bsd}&hgLMk;qJRjg;mHg>a{Wd=KPt(O@O_wWp>{W-i`>r`@hWEsS
zPw)X^A2X*Wzo()@>d@mFo!KGr!FI`d5T?P~AIy&_3TxNau*v#?HsVQ}c;}>tb*^2i
z>>=0t&)cXJ{=fJZ{Whk^K;IOJ^Fk9$Kj<?2Ch84Lm7j)GiEWc2w&xBB@OQ{=czfvI
z?})O?IWCKSw#%&l?6MEuu7<ZQHsXP_1W$&EcFEt%F5{@jTHtATE4#GC3-JNGb&s@5
z2;84K?jLK>;g+$>8vGx2(I3-sOfv5aPo%^|S+_V*w$4tFHCc%P@oaE@87FR|<K#F#
z6f+WHq+yg*nv&Bzjwa|hyiK!4NpI?wpW$uq^=4`04+Br2iK-DP_fsOIBD}2xZ_8Gv
zj{7xS&KwDo-P6M4{fIDG-6Kp2_a?@z6(+~iO;Y9nG3^YK+!??nyd7THB%wz`CHzT<
z9BUCOnWu-!;+kQi&&_f=$SmH+%+f5-ECm;vrEjuXKDD&S1$diIJ~?~0SZPx*RxWgn
zk!Gu4{i87Xw#F=Bbu3b%rbQ0aBNyF*Ddm}>q$|8F4R2SbTBOrki`3j_kqz)R$5M-&
zBgQ=oZ#~g#{rzNN&tQ>@|G1ksN|xpz-`mt8tN7e?TA1Ym{l$Lp_MZmZvwxV}Y8NIe
z7n`I(X_Ex!Vh`bKl0PnduIwR_eZ*g>W&SB06Msma%Rl68haYmj_;-o%`X;4!2g!sJ
zK{D%kkevP#B*#j9lM)@iNzEGHWJj5A(rfp3IpskA0&hQlMKc3$H(&Eow*qJlXE7V*
z9COI&6~EEYM+=a%^QS&GiCWoUVvzOq{Pa;@Kdt2Jr?p$)c}!h*K|_CLyU`~||AEUY
zG@|g<;^wbG)Pr5V`fB8HBl8N#L!tkh)(Sr)=I7;a!5mrgoxv8o4IATaSc_hm9IpDA
zSXN+ByXmkDzSe<@ds}c{OLDnA$i2ebq;@c>0$RZL2EFjZpnF*BzB7aInL(dCWw!Ba
zaz4Kd`q>*!!`nzdV$hth4|c}D)2HyZbY5me!sIHj)LMtQv=Y9`F#Kk9a?{-Qtm&e?
z_#Nj;@Z4f9>P4))ftYz%Z(eT*tenGZEa7?7lnWh3Grfhn5=^}bgG-Vt{>RgQJ{jv)
zqtRMPuMz7ikB?r4wu|dixxd~T^cCD^>&0FG?S*?6_&%7pcp}`rfVSfk`SHfCx(y$z
zyzq7)ydA;$_vSZxiz4v?fw!46(-#PDbHm%&Xs6yA+;u9v?ZL`XhS_0gsZ!COEM7~_
z?t;6vyGxuCO+Qg1YI6OJdUc>tcMZaGne_*+<_aIl1(IhROUya|eh&K|Z_jZ39gJPd
zYR&Rz?S{9#;O$}b%`KJwp#gB2+;Tm5Ti_M@Ip!xUL8n#U?x|Zdd8utPwc<S9x(?nR
zg15&GlK&v?%tijUE-`Dn75E?QCjWYjdhHf8L-4i%yd7i0qxmzQd|T1ojzZ(yk~}f7
zYAbRw!>E1Suj#F;*()5R7wXqI`poeEr~+???=$KU_7g4f`q+7$nS}I2l|9F7L;6$J
zplf~4dmD(x_0CjJtw`RqCcLc*Z<oW{sa!fSkIFC5s2wI6b@ouBF7IR1at+ZKDK*U6
z_<7aDZ-mu_>&E&<o$BnV58&A{`mc{JL}LyUC6!o}@5ej7kBbgF>5RMSl6OhEtY4Wf
z&*AN!U&OOPPU=BS7Sxsbnyj?;PC6Uj-Y!orBm;fo)I**WCI`azcU)$?M&WJgW8_m{
zV;g*S3)Oa}H<lSOyvDNq<WW~RX}J<kS~@e_#&bTbFaAZdoOOR!gZA%<e=j*vTQh@N
z8gi`Ar*ix_b72DUrVn@0i+Bc3o|r0+@CFP&mm-x1rO2yO_$Ux#t#xw9LSA#`eP@00
zgZwRdR#$pl?DUj&AeZY`i(2QMRH-~NO=>XT?4b!ShmOwVbeUN?C0(-cxf?A@mkGqC
zR$peA8u0{vpDr`tZ3}pt@isp7=ke}60cQ`UNv}C+auMEEt(7K@;%V{;-u6G2Dpz}@
z$n2vIY2V!;^~ySAmybiPy|T-x40c)i+9s}dY*OchO}@6VNrnbCX`{r2Lv504v`uQl
z)O_7+lCah$*I?^sc>A5lWUFG6C@!aMv5D!5O(x;pYlOFlPr=+9b}3AsPG#cFNVLXh
zra8nrbBb)=f{$SDRPjnnk-CLaqz=4Y*@&OR<_EXz(vqC<pJ#Rn;4*cPUA6|=rQaHA
zwHxf>Mtoawl3kW`x6AeBc6nNln708nUgGZ0=g}V%4<CGCmkAg6-Nnc)!{aNRZQ>T0
zEDc{INz0K*G6+AWx%cRQnUNrU;O!5;cqzOmPS)bhFr$5}JiiblN5iesaD`P?)wfE|
z%Xm~oNAcRkwhdqrF<zmOkusimE^=msOtgl}ki+4!gq(2gyWx`KRk*mq+avI{(NH)`
zyw;R@?cjhgS(`@Ow$~&BW}9SMUz2ofX_8U_FzQ99<Rk`M@jF=7<qMOO)zDV)yjMHJ
zWN!hJ{AUW4_O?*znb9O8s+uJ06q9&uGD*w3CaLq?glANk)Lj<_n|aRBFqyFr-gEn4
zT$oHd7cTar5mF*BQij0nU>F|M!y-js?UL2>5YtCosEb7eF8_qjUtsl6bXvI^TcjBG
z=X(s_x&H&2>%oa(Qlq^|R%YV8oCuLg)j}jBG+27{2$loI{z}NpKXQJ~A9-8(k7SJe
zEzxy;Gvn{4Y~J`&{LlT62_1jPrsm(I@aG`OzdcCy9u1N_FN0(ey!DF@l0n9AGN{lu
zNr1iXF!xMOFFldhi`pLfE_8=Q=<leI5v@Px!W-oI{t|ao_3_bJL&?|8z;6`ZHsYLT
z%8W+=v4^V{{dV}PW#ckujIWN`?yDb$dE*)Ct$S+GGsvp*##5b}d+PX-p8A}a);hsM
zPmH84gvP%EHN}oE(H*k?-bp_6jK7<v;e~eOCiz<Otuu)677(YtYX`I7ZAN&zfXkY&
zu5W#GQm`=3D}z?W-+RvqgQgudXuH##OU@XyIyIX8AJ9s85y!&YFY#y_5)B&YFlY}a
zVq|w}zj<IOaqk1JS71d^A4(~NS5-Ci$92&8!`%qZcVpqLwSbGZYQX#G%46Z}C*DIA
z^2#q3qh&roejBc4f#-+d`a6E-UCxgKc&+^~x-wj~%;$0uu`!P~FC%}wf_ydiC$O4u
zdv`zh3~#S>CMF+9pD(;Ucpg5(+oyIstjH-hp+~9?TCz#E@Q$N~d&0!rLHsmk<isO`
zUWH!k$q|H8Ba6dtBP)K7@b)Zu<;~Q=Mvft_T8%%>f6R^kf{v3s-=^N=7It}P$-U(M
zj<NqCkNDuGhnD`p9N}aS-R?=>L?bWVMP501h?hR>gI2jC@oY;k_{O}0N@%M5>D%=p
z{w<AG8Qy+@w>jbM+jU-g5j}1240NiW=-q&~E<Wg|7J6#aTj-^G68naD={bY9PV}V?
zn$24?xAfNX@V3Ef^1=9fE`_(@<da*$+h#$;w@;b-vd>42qv<<^w;PN2XzOA=`nsLB
z?jR=L{Rmyoa-;5rw+&!pN?)VC!dH69D*7r}omLoit2eXNyqNL8s>^D{V{%XS)D;t%
z4P2K!0V{7aYL?xZd)OOI89gGQ>=7Hn+n(%8uA;Zi0ax$9+g>pCXAL8H5Pnw6sPV(x
zbyH0|I|V&G^If`hfPwhUS18Dw7IQkW0=~5G(&YubJvJ^~K7Mr4-SGC+4;XyaN%IbM
zLZd~_6xJ?e`F!Sk%J+XTydAsDfPUJbUFqZOu@a9Waw6ZJGhcKBGhg8Ct}4Wg%n2I_
zZ&MGDV}-YQ;O)H?>GH;yF6}<0NwJ4%Vs4!#)9R&?=S!8M=zOjZPm{Dicni&Q)@JZF
zlzQlYcoN;p2Gii}Q#=(?TBV2ueQZy7I}P6c#hdoiJz})Q%wIW5t~3R%6mdx2>JACH
znj(?RTxkz)S4>Kma(`hlynXsCP3E^rm&57!%v<mejAX9bFIHlz1o@;(!ME`BL7H^F
zl_nR6ZP)KklQ#3y<n`1vDOm}RLHzsf;w$9nnku<E;bWL7MYimNdEFec1}#!U;>kVm
zb_=|XdTEo?+j!q@waHm{+pNA#O2XUg)oc<(4EqG8u7F!zh!g+CisS0q<S6&$t8A02
zT=v{-6T?-TnDI6W@n_ZzysZImC6UhxZ-2nsXmrNK=R4%~JYr^Y$yZZSWL(h{ss7X<
z!^9yT=!}QM$%5O7ZDHx)+-Q)g&1NQMcA9UOK-Tg(cKHBrJ>hLMUinoY+a$J%U5xwe
zVkZvn25)1Y<GnxAF5f2c9wkx6@5J+{8+zndNiqfAo~BoI&W1#Z9GM`RB|%bs<E0k7
z9pxM+M|;BCn=$es*h&uDDusz}7oUxmk%>|A0^TO!a}jcY`sJudX%QA7zo$pYRZF;h
zJrFJx`iG+(3YWZ@!X^KqFxd!itHIl69mAwuo-n!Y7bc?<VeBcBESP7K$?(?F6!yYV
zkI+!5c`8J@!P~0v_5i${+bK*ucZ5k%c)Q>`JdO#KgZ?IIRf*NkBq!Fx>6<3`@Wmw0
z3z4rThMswtcoW`sg12Si?IiNl!>)$QICwjHZloNgf9X4Vrn2p*1Fx}26?mI}3N`6Y
z)QI72XSg~N-Y$o?S>SD1cpLYKwa_egOGQZbjiKUOJyb^dhe`u@yQ^Y|j57tx?w-LC
zFyN0&ANND97y2RhFMSuw<nPkC%Xf)y@LdX3{4V|p-=u%FZ}R#=kl5D+N%H0(X$No9
zUIfW#cpDcNBs<~l>jK|o1H4Ty{Y^%F@zg_!cp$*t1AXy4pnu>H9xlsZ=E)h}`g9WB
zF5W&`0^Sxzlj1nVd=2t>1L@`XG}l*aCvZMQYwY^YTgR>^CZ6Z5U(s8Qgty*>sAcuT
zJE1VWCuv5_^v<Xc;qBqw)DPP*(`qH2!1R%ZEMm57JaOeW;>McfN#SjW>*SBg!{(k!
zoCkaI!`nyAi33?V8&S(GLrv6&{OwnRK6`1<Z&wW3>4-sB?Zj_;BQ+q_;2j2a<Gx1d
z2%VYhm=<TynF$811aG^;+ZFJ&B)mNhZ*RL1KXChtf&N)$cl6-(2NuPD$_USik5}@z
zv8+YK_`Sr&DZSt==fQbZ;ACYq&oKF4tenJoaWTAI?EzbPy(yk(s^M*O;@~`JF4m*x
zSihg#;XY!LaA!?ILz#>ovM>7qdq=d;@HUB9erA8PRHLZhK1O5ph8lDzx$<kozNg4b
z6FVQ+L@k>f;zs%v$|SflgA5N!c)K4j<^*!{$!I@^XK>eAxrt}tZF_jT1>U;SQ&D9-
zv%@~R)0^U<<y(4av*qYrHhO5_b`Q<G!$T{rL|1j&L+6Lm@9%EZM@gQ#hq(3^yxj<I
zm%`f*E#NM>R*zC%+9(@y@15x{DdD9Hs?sMyugSwXuo&KU!Be3AG*7L@o@(7Rw6weE
z;XscQJ=~~8Ccry*8ybwZ8P9^gXtpMi&;8mD_6_pX`Iqr|j-;nFmf3?&UYfu@uYLpa
zEYwPiFX6ix?WH*i;Hg1A*Ol0H)+=h#^c;VHjY(sTng!lo=xN0F9i1&bAFuKn^%VPu
zL>QaKrH2O|)8y56!rzF3cpUJUu7C0GgSThcA1>`m{L8xf%%}$PunbStmGo(?twsGj
zg*}jk|I=6M@vqq@u_x;O3e7t)V`+bbwx{O!kXfO{nKRusJ9E(t_?Sng%WQa?{cXC0
z+)0-q@Yb_dx<qGWz9w9(S&I48<(!z`=Y)qZ@gsG~jxEz9&Ld6EoKBUu%n2(`9@VJ}
zv$gTBe=<2uw&7*J4&L4rXU%&LzTQC-#GJ5Ghw$ND=A>KD{#>euSNhvjdBJ>@OVv{)
zy-2D|%#kXE@~2Ag+RQs^kR~Va?wbnpdO4-bP-ec_;qTc<YLlTU^80a$Y@AEpw30*I
zeI2qiik#|l`abDJIYO*<@d)}_cv}VDo~lX=M*VW4-68jEsbXyCtONQvYbw0G{5nk@
zPfC;SJ<{ZyGEXg@-)~Ko)yzOW21|<<PLW4E?<UV*5#f+cKOM4_nQtvVIpig;`-}Xl
zA93OvIF@%N9*0L9@@f_uUU(Z+g4mn)u=ORrx%A3(>1~sXVv|EnnJ?1}j@9EfT>Vqk
zCO=`Wa{q}YHfh3r+gYXHZKWbMxlzF;$I6jU=K26;)HFp0>=9s>5|zpAezHq%c-xh~
zLD7Z|SvJNYBWI%-4s^(^!+1=r$8Qk+Ixohzf_~Z>#KOl<+NCjCC@*+BDTiJ9^x^%%
zThl!9$MALzyd4IE?Op9Mg8hMm*>MZ=(hu|wKc$Pwa(_@Vo>Iv&@Kch!9grwZ;ca<(
zR7XvUm+x-zvMwo39(;(Cn<@Ah$Hj{ML#%x8ikESF<D^?^toY&0u;O-%tcSP3@OD>2
zs|<P^Edx)W+l99s{>R(!UXkK`HbP7qAtQ)q*Y6LPU;V=62$~<CjN$U&V3>I12$O3A
zO%g#2J9LXl1}9UujWNkuc>6fegja=0zBMsP3*yzs)H2sr4Uro!f@S@vU@1H%SQfJS
zaJ_r35Lpf1azup6E_i#jqDkW0nxqo(?In177~b|3>Yi=k`YJS8UBaY6ib+nR$NI))
z)Nzwcxn`2LKTUEz!X&HBCRzH{B+cKOB=)yS4!vd%aD(fkxxb@H5{9t{SYnbZtEs<I
zkG<$=mVNWgGL(L%w^Kr8Lltu0KA}==S%^ed3XxcNJ7!R@T<`x!rgi#3ukQ~jec`)o
z8UJ0jw)`%!@V0q|?{c-pH}QNBB+qsRk^2sk!}Ee<*P<Z#3va8z+h6c@w<kGec)JYV
z_JFrl;jIU}U6ljx?%sImjwRQJ#>E@0$b)J0Z<O{XSI7Abt;=6}=xzJyb7bB|JbIQ@
z=sLUM$5$Lp)$My|M&Ru!>b}P~|8;}6&iTEyEpbov;`9lu=lt5AbJ{#OyMcIa3R)p}
zRbV9NtZ{HFkXk8hdBlGHZxX$*=ylf+^W}xN;jrgNTYMQ|?Q!DE=7Qb{9%h8MF>vxc
zEDeLVrQqxKZ3Zm@Yg@1k+<tq&pwYy=Isg3}ZU2vJ=>FmFj{63^`Np6H&`~Xgw>x>A
zW2__a)}P*hvRn`ILW2c21K{oH+}vhmE6nfx$6c7}pJvbn@cR;LaT3ggn@+HLEi5k$
zYs;~m;H{C@-^6Fg1aDu$+j8)>HN3t4+gV5EHfX(yyce`0UW*Nyo&7@&UTZaRagD*`
zu=$%ThPRzqo6eAj-T~{;N!vFPd#91l#t-D}Mf!Au(Nwi`)AGc(@8}`?V`A2jFFpD>
z(R${mx1t8U7!A;X(i_qX-qxVrcANP^xxTq;h8oO08GxSvn$n-t8Nc9NegA}qI+?hS
zKFkG2jQSmoYDs!R>bIqj2!Gj(aOe{8U^l$okJ8th`JSgnz{Fk#`h&>L8UtV!de*-m
zjhZ~gojIOx7Tzw|gijBBqFu^+=%Owj_+P<#{0wH5B?rsv_w4AY`TBb5RCsHm274yl
zQ=i7*vzO+niR4Ol(BJd&8vD3K)Lf(RdOnP<t0=l3KKrc}_!q#~GUJW9X0%bmhZuEj
zcl@K9v(KX^<rcgh=SrM=2i?zUcYS)%T?Y^L&}`j#4<o60FLl*FJ6(0OD}O^jHyzB%
z7f+965OvJM&N^oYGembgYcKrfo5I^?k!awau|J|V-)=T~S2VKk$>&b}$t*OyY8M=1
zb_mQ|l8qd(vy)DOx6R>gnfvLo;uLJFo-XCHq|2%xc(;fc^lG|%|HjO(2<E{gr_0&g
z%n^sTZnsmV|Ljz`w}H89%wJnPB~@~6<-UE)JUyH$@o0xm!K;SEst?fZ)?_VtoGK%^
zJ>odG$-TDS!sCWB*Rv|z&6g_4S<&`nOqHmBRJr2@^PK-_ebU569`_4#zE4+U<{RF;
zRj#E-hlMFJVO9#>?hYvoZ_In`vL4<R6T38lx66*xHwx?A;cYv(I6T%NMY7{bSRD`f
z^=YzhYnt3$$7?SB7w@J?mjqt7Vv0Q2K#g;oLws*j?|jQl7Z-F#$>d?-ZQe*?#9t06
z2yZLE+sC|aYk2z~+$-*Dm&=}ZDfP!DBhJ`FUM0)uTgmeAezJW2nk=S-WEmI0Oqde*
z5y8|T)<5=|O4(#ocANAnZ<9jYx14na{(5@ZWOW{!6e(zvq{7T#D{Yfsg>147-WG?q
zb%<{_*CgLdy|oIl=3nyGiS_Vt8bz<p5Qj{L_gN1)BqzM>H^(8@2a(s#>5x*7?Q)yE
z@b{{AvBeQ5XCsz{rGFmTWXO5)-J5J;pJ$UV6N$q|+2qnmuHU4FJk}<GUHRFmWEnm<
zSsIs1mM{3<59yRBkLZD!I5&aW1kAUAx6AIw$&1Nx;sjd@!do-EoqQFKhtBlY^oWtb
zbgPs<WtFa@tx^DQhUr1k(&|c-e2^%4d)6YB;bvLuW0vrqkut|SQZ9srOP`nF5|2Kp
zXSZ;fRXJR0CXqkh6(+Z{g~`N0CJBJ4O}3k4`$v<w{xZofc$+$x`$w7NdMA?<KlxY2
zEc`3AxBivSfB#CEk-?G=){bHI8y+lybwXsPSE!VK8!AJ5crAE420c+gEpoY8Lu9gB
zh|Gu#mTw`!;$A9P=BEdX*J_g-fVWFFo8<2ylQ@o;<mC>N9NcM=U#Cry^orMc15;m^
zq~H~kJcZpq;p~(SCj1pma$_OyaTm|MV3P2cCVA8<Oh%LqlS`9BWp~L?aW)cbUkQ-`
zkN!x}Gk+vy#~&Hd<&QYX7iTg2kUk5(O8l?S5}f&q%&hT62EY0u+eUtsBTv4v7YveX
z3xXsdFh~X$3KHk6K~g3lNZS1UDg*9+l@F)BO5fvOCGGrI@z_9~Z;z+mqyMsHFE4%H
zmGf5@a))>oR3N`su_)RpczXtINf~l@_o$t@p$)z5?yI-yZ^#XAOTgQ1_vx#Iw-K}P
zn}N4G;O*ReFncg{vrE*O?vZOHHXFN#7+@tmu4Bo=_Qv}dro@fl_H24y+t4FJF7@*;
zV!Hjrckni85E|dU)Nb2SyM?!gy7bA`anbUj<ktQ(=rCC8f7PJDmkf9wa2|xWoyfZx
zmm9RrRy2-i|7)Y!kB7PE?ikbzcPpaf9}jOczGabP&YVg-3||kidcfNq!FaB@z<pSG
z3uc<QJOq~;!`q$(VKA5T;H?+D{TDAc=QXmj4w0+gO6)u*9nVyr=gsSNF~UT6JC$7V
zEb3?9x;v|P7y1mC3DWA9GhV$0-APZv=*|Xxyvd+rzK{dwbxxAI{!6Ue7`7KC_T2)*
z56AMpLkv3Lr$OIM!<#CntG-$9N*@yWceJYGV9tZ-Xr}02zCj=J7W@QqWFn5uN3Oo2
zyDmgKb!s}Cg|}JZ?KOB??wz|H&+4JKd*Sf~Z{N@I(3W#OwCW>tY*r7QLoeo?bPp{;
zEO?L}%G8R?*!A+%W>q~^SVc>CYJ-ZN`kmF|HnHqaqyCOE>PRlTz}nKRy70DRFT4oo
z2OWuCIs*OHr#bFgdM>lg@H^QHZ&!yf7dRC?bz$^GMaXk=S)4t?liqj>q|mFG>Z$E>
zdg_P0#GgN?m*Sa~iOZ1-J#;Vpj;0T4b9O%WaaS#Q2VFaTn}O`_hIBFNyz2CeTJcsW
z$^SQ+9R5vwxNo@X>80c_s<VG-#p64>>Qwe(?Yy0tO~#yWJ9^eAC#{RV^#nQE56m*V
zwT=0qyU5?}an|wG4Z8d!{-2l8zHcMn0&knZ+n=FMx_JVz><B0P@zKWCKo10OU&Gs?
z@b>Q)=4Hd%M(}n^m2^4!e=MDKbQAgCg_F`2cXt+dce#TX*Tvm+vEuITvRH9<cXwD^
zlPWbTlmgWxmGpf$zxR*tnKU-Zq-UPF_hVC>>f(|njZM?X32C|skKK7*fea?&_^&j%
zwRLL$*;H!$$Pa2u4s7RCVxXy-25-B<+p_TX0=z8>lk>pWB$g|+DUIQ6_6Ka?tqI;1
zhPQEjQx#o}S)tJ>%6|)gcy)@N&QH<pMa=46m7;rV@s1~^=+4j-ou?<#6k?O7;oaZ(
zrR&gfbLS*$|1-F=FG;UgC#hMZB)Lw<A0<ZVo0g=+(#iBUNLEjHyNP_HpldL<eTrJ*
z{Vwc_&q^Hl^b>mPqnoSMO3`fUTZ+Tmw7bd5^c=>v=QU%PH6Cu4?*qHq9kMHUsa+!$
zFfVPIT>;bWD!<XLok#52ddsf4H_?CB>{@w)*{1OJ9=!eRZP$lbyY|D|sqnTSO!aHc
zTs3&x4Bl?e&rH?y4%JDttLP8AI>6>n8R0Ixz4ezv72s?P%st5V5!`J}-OXGthx(L5
zhvEx&{F<cU>F|&9(bqU5-}4tfZ_i|1pFl6g;mP`EHXbuukAhrxJWkRVv~4;3;r@}l
zzmJ1CaSmmIx79Af^Q{g=EF?Y*Z_keZ)2Gm(-`xLxtV8wr-dD=nwecA9VT&=hb$y~{
zccxDv^{C&|C+KYNczNWGSNa!mnn2x8@~0R*86Kkt>F7&*jUK<$*EsveXa{|Q*Pi>W
zvha2s{TZsGZC8&XM;G3HasR1YrRm{Fu24X~C@ph~QvY!BkncpQV(&<qGe*jt+Mk@4
zBlHR%cvE5eTA&lxl?m5bIJX5onc@?!+!x{AMs%gmH#LN(+YW>%`_d34?+B6G^$>N9
z4N<Y1AsW*tME4qpXj#KhO)Vd)h4v8nJ`B;z2O-+kG+4ongH*a%kj739R=+91DljTo
zn|s2)*@5aB5UMZi!vl`Kg!OaQqIIW+sU)A{6}+AHDqL&PwMXD>s|(@shRZfM`!Vx3
zjq?xH@XOz{58lp%w+$YJ>mIz$w<cWsCWPrw%V16Z8K|6LU-jQjf3;lYuZD~L)oZpt
z@e+S!Nb=X_B|duZYn8`TZx!9?t$uI3^)I~b_|``+4PV8hZG-3fYEpV%Rfe-u&-<wI
zEFaZjy;<X<K|6d@3+8?%C-vU$j79+GvxXkzaN-eW>&UF->GbRVNzEO3nHg7PCJxCg
zQ*u;}!`p1PslkA^MQ>8KQHObKoPQ4{QBOdvUMO{Qy(+mIFO%pM^@=`&Jvr~6b~W0^
z(GL-ic*sJSg$FyLJRB(uL&#Oy)SlRAZ*qg-?OAx+Vl;e0^EMdE-^0lDh0BZKZ6SDj
zpXY}+pawQKaV>az^A<T##BN7l#V6iG-qbRfx6;M<33K<ruOa(k7tCD`Zzsd%Y&@^@
z5%Q^E@ikTiY<)#f@vAU-B`lr@ho63g<8aplR#t+Uhv4l$dC6xgKs*`d<%D@Yu(~jO
zpF;h@(C#iqVgquuSo!OrPfNKNs}o@AT(j|ftl8*0l=?teyMq}Zf5Th<9%kdlEVGfq
z`FY<4vyuM{^C;u+-R7Z}zp!r@{;N3G9IgJRt82j9_3(BW``!>j-0rRmc~UNhe7Sb;
z<2!$5j%G5spXlDd<M9Q^Ios3K&FHZnR#CTc4Np0JMR%h)%=JNI-hsD;575I0-kN&A
zI`m!aaM((nQ2$x!jrQ=i!wTkW6W48VHNDaNCiNWX(D(7_jedRL=(h|UhYUuOVi}EX
zCFosSoqo0O_VO)y;+;)z^zKNl$t=8A81{7sxjoa|jbIOSG`#g7H`fG5S1%?%ZU)}>
za5tkfzU-}HuEsw%>0w!gTH@+%Mz{LZnzZ15hkhP~(&+PmS6pf<xk_`1K@UxDjN<yB
zUJ$J2x!365b><Z_yaVZpGT+s>nU0x?_gsw%TZwnC=k`oj<8Ds;cf5)H6}Wb*MEo6p
zWN$s{uj}GzbcCTr_+1K_jZvA*MvU2PB*WTee{||+`UJt->hQM9IqFJgkkdqL(@O4e
zj>#s&WtoW>yvew;(PRu<31^p@jLq;i_jr@hwS~zDMc-bbK4np2nqI=u)$sNzyxnyy
zO*7%`^TN#WY~@t9Qcg7rNL4E*`L=Dy4I1UtvgJ-a!FQbpZ)eg+VH&)x-4QKI?6m=L
z%>3(9)f%0+6iqlP0eu)v&mt>5j5gDc7p>W<NV1aMlGP5K^@SL#c{DjUjg!=de4KyV
zBxyFc?>Ar#z$2ZMq%2dZQDGIDNG;6BB+VM0q;`*p(>@>$d4>2Q`8#D<d+|lL=A#!p
zS}m^!`Mtw<4c@2^4E9?Ke>W#7HdnGzdO4I*&Y`zxN-Ny@Jc^uOv|IRchbEy3Yr>{8
zJMm!eP(SmGT2n7HVxA<O_-fZVc$*8ZUVuAYX4;kLX_p1&7TjQ0>_NMHj@#7-{TENa
zhdFE;p0TUnRbFSctMYHV2IOI%<sE7aGncX2u*$&EYVh`lyF>lp?C<Y(+2ZYb4{y)G
z+r$#^xEP)?&ub2EyTa#H<WHT4x3xbxWciD{;$?W*AxU}$b91@V>o|(<+#MZ`)-45R
zhYa9*ccA7L&;8>)a*21q>6+v}MUa>E$f11XYE9~aKU~hPw`kX%E9}a%8t!u)k!1tC
zJ?D@aulw8*J{P=wLO-QSpAz-&f<)!(Nbh`jdog2zHXHGZ4k3?sS*)r=!`a<2Is;!P
zxyC5pQuvzTx5lskrH@yCsyuxe-jx3-Px5#RcaGLsw;y^F9HkC>qg21x5BUT{>EiMz
zok{<lxi^t=E{N2C2GsIUJKEx6gqpBU5VOsahnywyc5fz!DH~q!&97l<4sT0L3Rj<u
z-{hGTs`=wW<$5?oW0!=e$QpF+wGjQcIYcLxho~34wbu=ivtp=b=R)gd4OP?oA*xm(
zNUMw>HK-q?$XY@8@_`Cp|5foZUlpAxP(Nk_YS{WfJ+lR>$?8yioe;GR4%Sds`|x1p
zjtW-1DMT|%v7f*Yy*?1Cafd>6Wml*Md$A9goBZ*cE}@UR?#F*7CoB16xJs`J)1r}K
zno}=W8N&nh#U)TdH~dt4n!oxj^w<0u{wkKvU!Puk>r}i|Wqquwe#5HTkFBcs*{W<2
zR!wi@t&^?1<#E(o$M^W?mz%GC2Ki{br;iF{_ED8W-l|@~st&GJom*hh{&~dKYSHiS
zVS3{SH39j($RBCJ%$V_LiD>$tF3w~OC0DtCnLY%4ne(!T8EUua%}QUN@8|KAcV#lZ
zpP>gS=i*#5(i`cvxf`C<$kRcSbpOxI$cN_GREB(>-)Pkv<lhp<JivKwc3HGRA<lL1
z^RNfpg||7_PHsuQP*>_&;q9g|=t^{9`T;Nx7GG!!f6$o;u*eK=8=_Zd6Q3;$Z<muJ
zmAVmL!rRI4_A1MFi;Hn?pNnz&fQ#|?7~b&}v?1Cy43F6p*5-r9x6s4$c{~S94u!Xo
zyk|Rjdu}edRq!-D{4MHAeki=X3~x;lE=D-a`v`B#!_-t_uy+od4f|HJF(1B`8fG?{
z5UZ-v*=%HJXg2QEfxE1x4a|lwd7EdNnvG1<&sT%D8<v@kP4G5$3-uEIW}|sdnD327
zhqsC7+Tm#21ioh{JZ(QT^R8dyfU)n4@U{ZHZHo8o!O9J9z0lWDtGSlpy5`+ES7y|@
z88gtk2jK00e-XQ_fS!c6=?!-yV^ew^;O|a7>Ta}$x1R7e3f?Y)w{IuYH#Spx7?Iv+
zT|B)}9N*WEdXf9&lCS1B_8=ZwZX~^k(3w*b@n(zABdi*|2D9?N@}54v)Pq!@ztL!T
zyA0mmB6d)P<CgQXi*fufe0q-Al=d!0hlegkBYcIk_!zmrk$+yCxfJlW61-grLrcQj
z0xg-9)Sr0`$;>Gv56VeQVl}m+g?r!!%iYL(iQXPgSK~0e^@6wV0r)5jT#Z2qc<=mQ
zWk@y~0WZzQ)VuVC$BQV5H}Q_1LI1Hf5>sEp|Di9}e%s*f9(d~uZ?nSNCGa*H-cE<N
zJ3g71Yh^MHKV~NBHR?(hqIEq?#<;F<btbt@@U|Pg9l^3w$CI~nnoPygRCuvd+53?5
zL(WjHN#y%Xa4M3#q`w9^)qJ2+b6Y!=wLSM0Bv#szx|h;U-G{f=m!zsWUT=}06iw)l
z)<hRBo0Fn-*HcswZJP_;mM@yBVO}Xpgtev3Df$cEj)k{*!jpBHT%L{PIQN&LKQFuu
zb0wY%D|>FBhXVWzN0Wv%hiR>;L4mhB;O!lF+Xdbp^d$H7AG}XkdjN*cV(lJ84lQiG
zm^(@5h+~>|J2aZVbC8o%1n%B%28(%q9(cQ-T-!J6l9UcU$4<l>#kZ{w^De;KHt_Zh
zye&ML7$teZ-?}-}p4F}sdbSlY)FuuMB(A!%D)l)2a0uS!hPPATZC!X91aJGm+iC0V
zDtFkf1@JcGX}eaPv+EM9ox<(fu(y{5Zo%7<dD#!F{|Yk~!`jv?FBqB?-uANDwVQq9
z3ubO9%<UU)SAMq9Ir)rmcQ!h=G~RL!UYk&jcrKrD{0?HkAE?VM&oQf;q{b_gl%9BQ
z$!}=;D)d1nUfXLZF=g0m=Ne~+H?ic~NqV?FNsVf7yu%!-dY^n}G`CMzyT+HY%Y$CP
zD>m9Rakfp39X4HA@6fNq)B>CBnjdD9=MkHl^`>v|t3*AYo2ccT66K#7r{~0H-Eze%
zKfL`0Z>>L=53nmnTe~rf;QMdong3fZ8R<Q^=9jv|+fto>s$n_mZPVzd1aDWu+xJ0H
zGQ-;fMd|MVZ*Rcc5ND*S{1>U<agkc@7oqCs$R(Z@p<?GFv=iRG$QhwBWy4gYLYTfp
zhH0iHOl!TvREfWj7Y@~(457+r3DHvYZH9#*y1yF6!rL40b^-qKry?ObTq{I}zJ{nJ
zyd9b;RQ`8Ew7YzeK2!=+#IgXb`Rpgx(!TO-g&*7BS6lb`sh|Rt0B^6r+qKI>)eYVj
zgtbXQ!Lo$~>(%#Q)r$$%yWivyv(6`?i=DyLe1)pPtWZ^gx5E#8Q+i^`_q%;lqvN3}
zStdlM3kNIPoM4^28l+|cfvS}*kZUVnRUj7Y_R&vmZGN&O`)TG&Z#{*#-Mp>J3vYel
zZ8vzkCCsXo*}WB1*<0}qyj7%-RcF6h<o?N`4reUd`Oc!EA1s=D*P@F5T9h*1qAlc9
zzAT%;sC}CmJ!ppf)ZUjS<~4gvMx!^p?K3o^F`_kdA&rbiNBqOqE9l|sLCpiTB6IK;
z-FSTgQwC!T{b+nRZ=3GXGylAs@nxTz@tb_(SNVv6!P`pv$+4V=Mnk9Gt4*JnGI(&X
z^9sK0P|k&&(YhIH(r2w1J}l4aF`RlEyyoff*0T>aGiZ{YoG<^>_VmF^t_5R`;r~5@
zqt{@q-zgWP2KsiyQu2{!<MW~wP5eE253{e(xL!QZpL4z)HpjtQmmRQm8+vjVHG=R~
zu-ZDEJeSGzo|%Ea3vUnbyb}*xjMnFg2hZa9HHp!=<58P<jM|5YT(FN?2Y+~5^k1`a
zeWlr`L2l>AZf2tdynPFAAHrHkZR+;nZ4|tn1#g#7Ygwx6|8hu*^)VZ1qs)fS9r_(m
zn;7TEW7Joa34;^Fcjr-GarB$n$dYQN=M*d+<idI1#mI|J&x;l|!`tegs6U3kE?loz
zJCOT~kKnw-e1rCG#ua#b9^P((w>zp6vxHOM8@d~N$)7sbi+W7@9~?vPPGW^`pdZFm
zd}n&=O~vcYy`3IaXka($M>F1{?@+T0#+iQf_k*{A`!g8l(6~kc{A76hCJQq|;cZoF
zFprR*8@AKcNTG(>V==!W`D_Q4Fdv2Mi52L0H#~rWTr;eJw-E=)NeiSl>M!a@uhU;T
zh`LfUeQk@AN7syeI(VCs;ARXW?ptL(Ju1=2>+@6B<cID(=xTT_N83^pn&e1l1mG2C
zeu@{;ga5C_<TKgKMy_XOqw!^~ZG7?aVQ+><W@9PWIdP-S#xnedzXtGm+L(<T`OvjF
z=uzZCO^Y2p``u*RfVYF;ZQo}mqb9taM@-gd8}ZrFXk0gwu_7o<1Mz=LP&-<NzfbK=
z)2ZF$Kov;SJa~JAyrz<4oZ10v_y5D(Z+P1X*1G=f)JqsUjrx{+@OE-jr|xDUPxou8
zJQt*Dn~A(8cssLSie|xEjm3-Il%kyQHb1;I!P^d-Q*<>4{;VTe(dfeMPm=KY$f1Rc
zvxsZneCAN63l4p}=8zqqa^WF|4x!P?!rNT%*3pU@l=^TKhAs+n=;8>6iez)>-(Pku
z4X~>_anOmWb`2uWsOcQ|v%sNs+Z?L7%c0=`4rR%dq$&mIi4^0I(~lWxXt#<-9r|w{
ze&{BL;^A#ibXxXW%=(6XyIMHZs6DyAy&dv|w_jmx3)Y&B4vi#Nscc!iR{Y&@Aux5j
zUFSEWY0-VT7NT)y*!2+Jex7Pqu}OBl!&i=k!@ajM-+P~3e;=|d4K{y(y*FTSNmxAY
znO%?H*mdT;U86oR{~NwmiG;KH914lIYb1X!3`8r#S`Sw42>K#sqUS|nhl;`5aMr4d
z^l$pZTdoBx4dHK3bnz62%FU&=$lsx*ChC2Tk}Fo3xp_viUJpuE0M{Gs7Sq#W4j#f%
z>U%aNX)xC(n?oIX){C5Ht~)N2<M^8F8Vqmy!`lzDZ2EMT9Aom0k5;p5h>KnSz}tvJ
zHW}na^+wB{Z4j@&?vQUxzo<9NdK;LT*>}vjdvq1gct(smu8L8Wb}{Ob^jq$8ek<c{
zX7^P7rIPS=%J83h!K{EKH-9L4>JJ@o{h_xo_Vk`873&hE4DY^c#ena6krJsn^n<F_
zGg75}Bk&9&G;mOaM&t_Df$L$q(K<|XqrR!!@Na4{B~)uZhN;cnF!h)os&`jI)aqP_
zj=|V!7V>(t1#8LUAeB!J(&5U%8qhyj%ZQ1#I}ofzC4#l@P7q@;L$nRvF0SCKXRUpR
zPx~l&u8$&#fyR0GYU~hSKBJ!&AM#V4i+<{M$4_0~_$l?4pZeeTQx%7w$~X7d;L+p?
zlRMnUH9)7k1*k5yJL?(;YdW9h*wj!()<pN#4c6g3LCV`BNb(5M?<(l%2Z5@5EKrSh
z1S;JaJZm#t-|njxgNeuD5$CMrt1Xc}Y6fr9<s)8u*Q$dTtjf6rtvcH(r>9k;SzD@F
zm9B<W&s?qgJJ_P5&n#*LZ&&QJ=)+=*+RU?P-Ykoj!`lWe$yxn}esRp=*oW3}yk!oR
z;OP)@GMCZAyeB?kE8=s?Xjt=QG;XEKX!wnyo`QVN$+q;=>rux!lAfmh=;?}gH^`Hk
zdvwi=Cd3uXxl(82YP>6j?%;g5WiGStU};{s+8^EyM>DpAw@Z?UtD<!;)Ijrco~(p#
zTpr#og|`vC=nK&)osq91b*#iqYv8LM!2@=IF^_Mcm(ZofU}qh)>f^<nzZapSU}H&m
zI|_#Ogtw>RZ8Nx90ltn$^Pbu6VoYaM<nJahxfP6Vv;bzq+l7;e2hX6_+(dj~PyA=z
zbJhqLj5ZquYnQkY>vfuqHT1o)e>NNKKj0TXG#lxs;s=j38*}=T69sRrXxLovwJf~N
z3U7ziCzl51UXj^&+1yMX4Ci&2>;`Yk;Wyu1#5w;uS{L3v4WLJz54ky?%*IIeVS&9N
zX1<f~#|OI@x3BPdSOadsu_tJ8R^)ebuJTjQ)CO<%HaTX*bU(tYG30;m-|J@d^QN}~
zy!|snXbimV!!{N6_DAckZ02sHYX^%vP%G4v`F`*=`yl%N!M1ZSH#fZiS3aUw?{hdu
zKfsLi?pwq4)-GbG7cyit?y@cuQ`kZrv<ho?3wm_$`#+>7!`X$z=wIX6FLW`sC(&<@
z>y#mN_zj+sABJYX%jZp9PHoo_YOToosta$Iz}pS<gP(#g?g4Mz_qZ7*uHP!(L+3uA
zUo^GFg+Eg7bQ%utCJxL0m^c4l#qdy?e<j|3k^djA=`OOb3s23);)@*jwfGM6%*L3R
zW@D=-*Glkq5xmU_Z!feq8+qXE*X(BFHN0&CZ!5vuLGZQ|{fO>eq-W6)?ptdz+Rmbu
zl{}_K4(dl=rD^Z_G$l|wdi`OVF5XU49%8W}1E^u?i>~eIB>&Z^8P%QIUIq5nbgGh^
zT1dXIy)t#A6`g96->FN~x{e-_syW3|b?88fc6Cb8llsI^yD<x<HyW2&Ynhj(P-~f@
zOdV3>Q;k{U)Oq%Tw{Ow5n-1e0Q=c*q{rZ$Vq6m0ve?Z>NY4qX&@_n{Cv>XmjgSP{#
zz}hBB)Lqg?0iEUnZ|lO2qdmyk$x1#^m|b>QG|Gzhdu^wV(k?H!HIh}Hbq|(?_&D@E
z1wQ6VLRY{Zcsm>3Hifs3*iS8TfI`Mm*8;;c!rM~twpL3#S$I1I-Zq9?e_w!+=+Yu*
z?F!@eESOnuH}P6n+YOFh-(Xks6?P3=VAmFS`*;d*TvqpqcI|<;4tQID=Z^E@bHd#A
zThPF~rpr;gnw+rf8lH2OHFiZTw`&*QAp_SCHPOADSml4&b%N`OUVH3%(F|>S*rw%|
zZJGyf+rZn`_{vq`?JczE?yC6D=;Uy|$EYq2RpEP`T<Op?j&Tt)Iob#D4dRm2`CO9v
z;n}B>M^+KumYGd1^-Su3InIGulQieEL-XKmCwM!tj9r!B?U?86w=c7ASo5~qH1oEN
zI%k^}FHX?5b_r^FHD0$SQQHG=gL1}e<e4~?r(e{n>-0QKiP3Im%Zl3Ca#Lc^+cBC0
zZ>y&KRuOnRV$Uz-p+~YW{etHW`Kg3K(b@}dj|}-CM}{AI3U5DnMJc9Zl#(8O*Cu#-
z7JVCbGE!Zc`E%k?geuvHhc+e0Hz`be;O!N7>;B=J<_-O((*r`);Zlg^q9O0CN3)g;
zQ5!Vo+2~-k%^9qZ4}+B9QIL}DLE2F*SmpcT`@-8wdxCYWXs{j)4^l@oYs;O)Ks)#_
zm&Zqcwe``EecsFiL@T!U(S*NzWtm9L&1OIOpZC-2+kPtZ%uoB^ZF_k8G>+TF{P9i5
z8TRzo=$Zi<+WxD~FZ!w(SHH@yUZ8G#3RDF&ci%2Sx;s8l*ZTx%daEF<-yA5PXnzfV
z=dXuOKV3`k)eC=LUD`q{7G14|zB0kvz3Yj+p1?C+=q=ytR!ul%m3^^QR(N~8zg2ZR
zS+%07RWB=A6`#(k@9@^;iA8(%SQN3*qABn;E4-aJ!=mQ!HY2>f*q7es^UyNXhU~n?
zx%8I1kvN<>!Y=$xUBrl%%!MaTx05=4)4~jfJF&c;)bhL;Pw!q<*%9<_g}1$^4?Ruq
zn__*KoyYn1WD!^69^T&RaPo@3kSir}dhzdC;pI(4+b%=bmib8zag2*`G7e7{zcfQM
zI+X3#1T<SIX6SUs7lgOp%X8i)M=3)*8ZQXH70-7vyxqfjJan9kv3ERL6-HK_N>4aA
z89#@7s`V~L63m?fQ{CZf$!+M@Ej(wli*bIdi_w<bwRx-?yuAURr%vN@kSAOL?tUCg
z-qSF2ZC_%<@SzAg)A(XGoam*{z08N+j{dxAHu|8M>P#>jroPnq523D~?Q726c{z8N
z%4IgT6(m+muf9h0%tjY@y9M6nYQ}=M)!}WM9%dumK(jFo-j29rHjeBw8+G@PyUfb+
zj@QB4v+y=cUNkuRcHB@GquK?&+ZA|wjr9;unUy;`G2JKhR47cZ>AvLIhOli)uOWCl
z58hscQ@f}OdWY9Pu{ORkyv6x2dcfOd@OD~TcViWC*L?8y8UGIh`ZIF{-kyfHS30{J
zJ7HY)6nY8~m-n+1S7r51M_*<)=Bv0e3l0{pfVJ7^3Fb#1&Yb-Ib1Tvl%ZctphgZH~
zHtKVnzA<BaRwcZ8`YasbcUX<@G2kSgc^dUt{O?pJ&-w<u?NZ#$SlWPI+MV2tW`mgD
zHjI4X&dg|FPD5Sla@tV`)gM0hg14D@tuODhz7fZ06Wk;&esVH>ZZ~qgZ{gpcqmRQH
zvvGee@mY9#1l|V1+f1<bBfr5Rc-xk1qw%@%l;N!>wXp5cwd<b4R%&dvz}s}x*UZH)
zE;EN%EWC{%e|YfAG|gC)rp%S;;owhS2Xb)JFCZ6}*sE&~rw-s3w}iLd;O$0u8&ZRK
zEPscVaLVdPRRpy#wc+i%{q&XSldAdfwllmv3~#U2W1eS=6fN(SqCEJ@#mHkSJ&rl2
zO;hv^-hM2ZqQWjIItp*Q9-!ABG1I5`xx-**p3CUStK<d2+hy?f>lTMTY(s-SCQqm|
zxlU-@(r{oc`gV8rBt`v0ZtO5v2ybe`+y3x&g2k>JaOxGSI-G68%Es-T%V0RXeFSf7
z@!AHz*@rjzz3{gCDTl)LpjF}RK~H){c#x+AyFBtZbmSX(!=ZNFrB6e*XtXF??Age!
zUq$KLP>_5mR+~b0X3*gkm$z%?61zIX)b4X(BtCLmSX+QKb)a2FVXg_P!t2I%9cs>d
z&4JsS?8*yoJM+F@(YLS0+qJBoT`qO)GF9aEE4x1Oy$ZtGRqQ{OysD-5Y&x{nrX0I$
zdU(kucfNlac)N+u+_ktvUEpnVczYY(W<h_iF3B|p_wQ}sP~QP)-s=vXLBmdlw;kcF
z{Wv|5$KbE~QqLPj-;8eLl$FC*&XT0!aJmNH=d>%IneVvP$=tkWc*%X3chk$JP<k?^
z`;wq4^yAM!pC~)Lt&uTa?)b;c{)<zI9Pz4inwnZ>mk(GJt42XFs{1-dMbWq8CdFuC
z!x%j)5u?q?%$4iOY#RDUd9D1ZF|2X<f68rOw2sq@zW_ag%cn&t{yH(+4&PNT4YuB7
zjf~XAqzL(@MChD7LWgdUAC(x6PfmVRRG7xlzam#FH16wf>OA<H{_Pd2X5=Bijty4e
zrC`O*3|6fn!CKfZn7&-W>bWjR_m2hX-q9cpTa2C@5v2Y7g0#gFsPX^!>)RDS=IW5s
z3(tzf)@^P++V|dD`}TUP+EH(2`1)vTCTeZq?ZZueYH*xBO{e{o$oB7bep)fsPy1(3
z_Zs7)hFyI0WU7xg#QAGQ+pqey;H#S7{3;8)J@Y7#`mG>Erwh_EQ=m3R1gLGp06onU
zpz|UAdbYz~Rj85umg=iJe!g12g1!~yeYL-UuQojMQH2n1xt#LW)J5LPxY4R2C#||X
z(yD|8R^=;VRojAAaxN_TlHH0Qgr@bk==lwc(!*PiO%{!tY0(f^+n-f#wncwaC(@7e
z>xG%jnOx#-EWuMOxY^wZ?1skSTzdW?v)-8LG58?8;kJ_}x{;pvOUY%KPh1Vwri`P{
z0y!^vd(-n6-ZtW#*%jUn;(WU$KfT}39FvYyhq{({E4+POjr>q@S2Dzr7vyj;ru}d+
zw!z!^tRmbFL#O${;FWMyXxZEuh~2{3X2by3vE9w%#{*rA8E^0-(Y1Hcr90s5T6o(5
z-e#MER)x3y(Wtp5yBL4N&`N8m#f7g0x8V)LSTpDOa{S$jo&f`4ZALU~B7E+`b7#Wa
zJyZD{W6+ZDwjsQ2R}qgouZ!U{krM?kiX1T;r?$ag&f%RFl5gDDZ0xLVHgXp-8y)hZ
zopRz$W`n6Y=!pk&y(*iH$!OdM^*N_EF&nG#lsm)S93Joo?$&|PpSaz<9iC?^VvUB`
z@aDdK@M8FWGc$M1Ml<*x$0{_~#h8b`ee{Zpv5)VZ6V2^in`?q>d`EQc{k*Qm4*Fr9
zBbIrr2|WX;ZO&q)pC7DT1aDWt+xs=>b%EY>|Kw(Tc<yFQ%!2MEk13$MyHThLw|V>k
zOcuQTU5gxMc$XJ%I;JiCt!q;&Lyzm_@OE5L@`qVv;O)b^)FMUUXMaOymLk?%(8ahO
z$h8J~J(%l=v2gGs->-56{_IYZF&hn8tEJfp!aH!|n&>RQNBip3Wi@g&rkLP*X*VO8
zIBr&Y2=?ybYP9-^7ZOSBRxtgPe3=h%4L*C312uy@HkjUSFV`CI)@vS~F}k(uMEV|W
zpwB$Kb)LnC;I~=}Z%4x0N$~dSD2{V~`W1IF8^!od-jpyK=ZnJP!e*oQN3<=xT?21F
z!rQ@r^ewy{32(E&+r&a9!*xlTGL)cxCtsTGW=@m!4RtiP=smaxy}FH>Sv=!tc$*Hs
zp04UtJ^CW;fVUOvQQuP(FF7d{-#1m+cER06srnDzmM)d5-S_C@0B2j3W@akeYVh_f
zzVbzwd-Qg)h7}_Z_B$Gt_~D)I^hv0cq`%NleUlv;{N16Uw;VctfprCqb`nm&zKshx
z|D$ow!par19NIS%o=kV>F{|EWd{KCN1>V+yHx1xoUs%-W3v7f@W1iA8@EP~9y7PCh
zy2Nis;;r(2H@`Tv9iQ}<6-L3^PVjc<4sxUxu}^qg8Q!jjT|II+beH!X{vKU-mb_r}
z=}dpL9Uf|pWSe$nv+GB8_Qx8@?OjFe+6-fhEJlmM+uv|DFYD7JyV7+fM~dgwOlPN`
z6nySZ&eJ&FZ<$>!*4xz)-df=8%kl6O-kRa<b9lS2Bs#h{-y`0p0QNnfJfy&zHho*e
zoVS0Osd~Yt(Y#NiJ;a4C<5w4PD7utGui)%9c$-=VPafW;pw*3>Ft<0y@P<RZiGk-2
zN6&MdpVV<^P*b~xaUD?{-TW8V0>15t#rCl)^AQ-&_UsRvs=u<S?kVP}ue8y>k-0vV
zZK`%PQAO?~sN?hm)vuJGt$UckZi-j0&^Q%c9;Y7tnU_;1PF*s`>D7W*ohTTqQOwmz
z+!CW|6JxZoL5!}%($9ifaYgAJReveH^XvZ91bP+MJotlppdWhTAEh3f>8V8j{Wj6x
z)w1JvbxLL4&*@0*8$ll_cv}wMM$&WXDSe$bkmnm58>ZxlFzxdHrbVyl7ghh8(#3^p
zZ;wz_@e0w;gkZfMO@40@{_Nc#1?)#Zt_;#g*5+M7s%i~X!-;`%L0c}|5GcXGX+!)K
z)c8-YC10Ig>!b7N+WM|Ox-iaL@9TN1U>k2OsznY`NpF>Q@z&}0R+YL)?DPn6(_dD-
z53#E2QLC!nu<Ajiw<>w~=m@-haneU)e);el`l`|mUuAsgs|8PemHf_E1ATpUFpr<?
zMPW6M|6EOsb{;*TCi|*?HD8rG<Rh2i)X^^W)i*dBoX1Ba9`oKiz17ujRp;YY?H+HH
z&p%de9%5BML#r|sv+7|stFpq|cJ5a747X@F`quk8@mnv89<8=07S5KMZqXZ77@o2x
z=hX1t?nYDMb{>oI3<vW!>|H?4OHUK`p^F;9+ZpgSGrUbt&g+&rZpJ^8=)DMQr?Y;d
zJL2-Y(Yv3T3V8dix2rL+p{wyC7oO&ibjEmidmY~9fVT|#GD_zl|JV;N6rDO4e&&R=
z<yjfvZFRh2$5+n3@Hjb*7$v+t2xp7I+mY^Q;z$=G&>z<9BVP!P4q3o^z}w}pCl9=>
z24kHtHworO`~!nnMbNIjmcvb+-vHMBiR)g4ySd<P19+PO-VTMgh2d=|&u_}>-Qn%8
zp~Q51P_J7RU6~(-qtTY&OD<euHV&XiKXd+WJBK`{hU7$*G8^$|*?jOe6yBbPx978?
zaar}@ZnOO8;lFsS68uE#USK;4KA+EQHXgar{{r5&g-4&@t$4ra1^CSCVF$eZ8{Q81
z!&>?$JcsjByy#nW&}{5IiGPj19axU<m_%>A82Fyu)o3`*)#%ucc?K=<cF}(elIevp
zh92{i@ge9%HQtxpBp+rRvMorSaODMP$E&VJ#5($Curh9TrCu7H%9_8))hJIN4Wj||
zz4UQzk`LXwfEf>m$=y0YY#d+s5;;-wPiO4?Ko1ajTNK^~qPN46IUc_F;^@Y59D~3b
z__yCo#`ERm^WxKPfp0OxU5s?__Fs6Ly%?S{v4&x_$hG1#%m{NeLYPB1n(Ma}F0RIy
z`~08WL@N`E_uyLV+b6EKZWC+ILY{UJvvHYY{s!-$IJ_+gZ?~<Yui$Dt?3H-PXx_*C
zRwLkT3wV1A-bVG~x(43ngSVI9?G8TA@xo@~2doW$Nk1rf`}D5KXaR2vz}rat<jH8;
zy5G_?5nVg4RhlNl+ew+|N%+C3SMc^Gyd5zZ{R(ftz}v3ybwA#5N%EmO|3&|#BIHb&
zoVs^2RsL5}RS@2K|37a>luXrcc>A__3OWlvISY*KnWDbc>8D8U@#^dJfq=J5(6XB+
zkQ3ZFNiG$r$w5EaKRJ|k-J$vC9oh<SZ%#vR5<@LJ(?P!-Sh@xV!`lV$wlDXUhqZrA
zg|+Z@BtB`k%nr@`VW$_AU7o}{@4rF^!rH<A&)YumcFz;L78B1bKm32Z)oV0fFg+x!
z^lE~)o#E}V9n`nNci%5?^g8{IcwZAf^+9+$1m1?+!pnt`c3+!p__%YHGs|=-`9jNV
zY6tsXX0fX=yscXk-P#0?nde$3!^~ec?Vtv=r_ClGcpL1n={3)3+MB%OvG5n(mR)OC
zCbaI*zFZe%K-)&x)cPHMEpgHPAIVJ$u_-N_-iPpZFue68c6xWLO)D4M^bfq91aDs-
zu&dP-t}&j&=Tzd?uyqNnT>)#ClDqsLyuF%(SpmeJJ&%*STq{XWxX$r|x1Hhb$_B)R
z>)X`~?rw#-AK~pDt`QFN8U10uA92|;FKl{nlKu1D&(*PMyfabh;qCB;3Cul6P`}&>
z<QvE9YT0-lIUA>Z^W(Ih*#lkR?L=n%+*}%~R`7NSynVefM$x0G1<D$u!ti#F-!J8Y
zw_Xc>>Q<efn!YAl{ot(uZ|A>?(qC(%G!x#Qibvaa`mV>xk;-)-QtSFfY6!Y^C(K<E
zPY)?-XL45#SLa*gN6~xn+RJb1b>^E+Rrscl)2OB09-_Sih>PMMJK^omlpr;|9i#+&
z;_`!oG;2|i>g^2DBlN6gLZCu&2C7D(0LA7DkoyB)G^#H>SA4Z*m5&<3%TLd})gRtE
zs(I^GO>Z@Wx0T^-F?f3v-X>kJs?c7mR&BQGUskLac}r|l_K_znZ<T04-1mf!JmBr{
zUB2|<^wn?n;|XtXzV_8b_?rE(k9IvlOP8cZb|talxxTtTk^HEk<YpD~Rh~xF@pR($
za~~Zm;G>f8He@HfO}45YybT>qJ<vaJzoS(%s>0m7R{axa(Rix`f5f5($H|d81Y37n
z^zUwqvMsS_5S)#iYSH>>7M+8)0mSsm!P_p(5owNxyqy@{*6HZnHuN-3b~8e6llyYq
z&3L(=9_7?^UcpCxKbL$Kc)NVG8~!Bc;=;_a%+37hy_}oz^}O~luZNn%6=;uVQ|P_K
zYB`#^TViU3IY*v&>tein&-ohOn&9oz-^|Ejy@=x69pYkq@<H$7l^)tgJ`{|6wT8OL
z4QyFY$Pc~=t9QcNd^|Q4mQEOpuL^UYqfK+b+hP-4jII1VaXjyVerf_Y*TT=PQ}K=$
zF`tCjomqmOMGrZf!&Z3P8s4tOM~<tFzUDPWM!Oilhu}q{Z)?KaQSjCi-d;m9wO?#D
zS}imiqv35!&f(4m<QtcSrSP^byiJ0)t2y`I%?e9dRbfpgc)JYVjxT36PSnANhPU_N
z?N)fZ(3L!$D0-~Z+waN;lX07Ky$8H40dMo*T^E73-QjIDSnt@%`|oC7F!#YR;=}2<
zKFEiTe{D9p!P}kiwg$W%?%`_qbKJKMCa=18IwQCqeWn}JkD2_lv|(t(e)t99_yk<v
zWDKA-nPWbf>zLPJ<gO*)r`x$6Vfp=Xq3<#EPx<KwQihpF2E8u_a!j@pNA@CioI>pt
zk54X0jvG1EO}Lgy=DKKOPI!9_{(0lq!rL!gcVw(eeA<UxBYMeqV8zx)>+-!TEHfL~
zGm(o;K3sNq8_<(JU%i--M85J|czc~XoF4l)E?b$I#+pu!@D`X|hufpiP=f=zn-FV#
z;m&cb$Mp&vJvPB?te<Z-T60}AitPhD;}^5c#=#l<UQ_TF@RK9@!r!)d$aoJE;q|!`
zlhGX3=6r7=w+b(r9I0Z|_MB)-k3?!|*A%25aXypb3{KOXxoOJTfL;|j=wI;^#-4R*
z<zZs7bI`IwVJd#|Y5F!@t?g7hcpFZ=PrtnMRyq$;52PyR?o{0lMYmd0H0EK7Eay{{
zZhDG_mPt_>wW%|+(v$Z@vKFsS)}{K%(wAhdzeD~jHL$jgN!kW)x4_%9viO;9^f7|B
z<FAwd3vXw@+Zb579F~^o=g@w(4OYX|G34j+_c8cds}1_C3cj<Mz7R>|72&bA48zMM
z&e`?@anYA{m4>y=ZrgPL{uDh2N3PoCeuTK{4tO|~9t_W^_3?J-@=b?|qV?+FUpIt9
zOW@>-i#9#o2FGCO=h<kwzisN`fv&3$*XqHTK{jn$N6ZrD?BY2mm*B6qu<3F`n`RHU
zk#mLDivI17|N7=X*z^fj2HNy4+NM5_ZF&Mf7cgJlX1B=+TZ@dhYxHWnQqjSGKe6fX
z9Gg5Q*|cH2P31G#6rM(IFSB9Cu1r+g$wVEGPSo3~<RcBUX^*E(-Pvb!59YL|vfrC_
z^3CwOAEB+&sMUeDdEo8xLd108ZHaXB+2A-e<d_s1gvZ|2u7haV+_1Je*8p|vz-w0d
zs(9$-(B18cdygTHi^m?r{G=T=Wf*N!lXiT@3N~F$Ow{XMiE?AM&pCP^H;1=@<Kor#
zRh(w+jZ+kJ2fmb!)1@D=dj2+650=Er0&g8}WAuewsgxcu>N)(kCe;3=c2$3=IWt%%
zZ2qaWZGOsB?Wc|piI#usA1eARN)r}GX)nAj%S^c<%#yF;@?Gl=GGh<kn$WeOA@sGV
z6`@kpxNfgRpQvkL)QyFyzy@aD1c$2B=umlPrsjv*+Jt`8(vlxsd}OdnJA-uRP9Qx$
z12v~tkPgfX()P~y$Dh9{a%+HG&id=ZZ$H_#`>84ZaBw<bWm!jzcBzlDmG@B+ynQjo
zTP?~`*II?!?%o<3X4T<OR=t3?A?K`Wy2r}QKdX*Tf_p6UdaL@uV86#!jo5D0_<rOo
zwM2)Ovubo<yx*Kw&B|<LZoE}*3vqu@tGwAR;_uq+tcw0_Rj2CSTGG*5TaS9Ha|0h8
z?C!(&^-)3wA1%7ztyoyT(&VjA@V@patJd|kYCo@S4{t;NpSO+S?IL(P`G`fn2P_)8
z)1ud`)WxtC&hCP@?L948Fw>$c<gIpsu{H3I@1boM;Uh--QWyEm%}Dz~9@KT}N#O0w
zeQrkWZEi+scspet^WgE4>%iNY@b+CnH=~LRGo;Z&1>o%}c)N24Gj^t@GwzO~FJpgd
zXUT<n_?2_`M{021kSp_+7^*KZJoM_~pX9AZa}I~Sm+*@J%o(VC7angV2M9L4fVb6P
zZBg`SE_hHGHq=EIeOgLh(_C~cY-|EwPr=)ONiIeZ+rUw99M0~9nIBl=;cY#5J0IR|
z=l$xz*h}!%vm$ZN668Cl=k+kP0=%7)3C?F>ABBne;){M_)#kC2&(TU~qi-WPf1{)J
zH$q=lFdKu5{jZ+n0KDx5Z<FC|7kE1tmKKM%9k~65?NheVtN_?^pXXW&n2nj4@Rif?
z{2%m{hqo)>ZJsB@2@li1o*W%lSmS%$WL)6fUlYAG6Ap}~XWvP5*Y~5a8~r;M?e>Vg
z+a_btY{TIR9Es0{zh0lu?SU42!SQWKo*Ff$<;Vjc2qXPJa~xRzz*`r1``R1--U?e;
z8NR_%u78fR&cNF*hv*r?^^ts;wGEq&Qv<atjo9-?_*NG``?A@neS+L<H23-)JoYbn
zNq>BDa*LL9O4As0VEC9cU8@Ag22<C{Otp{b#O8%fMnU-XHV^qnE9rR%8#^8+eu{1{
z@r*n%V*R7}{)f<(J>l(+KXt7$(V7eJc;Riq?Iz>FY5GgR+h&0#;~Km@Pz!eR{j*IV
z<~@(!;_v@4HLxWf@;H2gX?V?($zg-HQN6g<fwyn!;yD!OS_$6H!AovUKg1{J>G{9I
zWTXwJR|WO9du1Z`&t!ahOmC>wX^NknrUre}R1j^O?TM2fvUtbv_OvIxiMu&<hk74_
zyki%5TbsP7OU3b#bJLgcdaB<33vZj?Bg0$AYx+*W+ZXWGZCZ-DmP{d^hj{Eg>S*a#
zxNaqVcxxvs*oFJf&~NwOB>59FTnul0TPNv5Y3ftpZR1zeZeFH_^qfO;PB`RF%(hw&
zhu;0|Q0Q1Z-$4$g8$uiv-X4Ot$KY*0Vv|;U)0<A10B`HU+nfG&EqMbk|FbLU3A0t<
z?S6Qha@MYSm+ZQ=2hVnqU4`Loq5jl|)*_b}z1R_twCz*2->4U5eL*kAZ?(ySc6>F{
zrXt;JY6EXK!P;S2;LTK6yV|B$bZw75HWh_ETj1@$`mnFMO<4`R<_<QkhKDWp!svrA
zcpr>Ci8j4OpMG?!iP<<?gKc=CHnm4TI{94nyOD={jrVJXZZ1#$Q87M?8@@5T{RD6C
zGe@V{@kGT%C2D3Bo2Cr5>HB1x?hLVMLNl9gm7$MQR-2l$PFa}+Fv6xNKBtrYKj-_+
zU|;8Rqt!W1r#L3R@vQg3RA2OLD7yA(4Z8w3Chb_ast`-&v1P^3%J|=hGEsxXHAA@_
zHZ>S!Q)#~Q`m#1<`<194e<$kXuLLcEx9h{>sbh^-$gengY>3mkZgEO)kJYw|v0A+$
zR=4WLD)4@cMo@G6(KAN<;cfm^zg5lnrQ>lwl?C2<ZuzM-@HU|8PrZb<tC=}>^l21*
zIncIMqBOw%T^_H$>vrGo%6^18AabM@Q)gS_9=#9i(>o%6xSrx8pSm2T+6}`raoabQ
zjSSVBk)fL6N-zIi)cRx#QOU)@iVh9Z_bfpQIvA+=@YeGxc}X5$<yqsan&teeMv4Av
zecoUH&h^uUoPHV}=Bv@4d{l3l4?Qh?G@zW1CZ5Je?%=JeFfif?{_RWhk=|LA`<YdN
zSFK79Z-;x4r!>>5cVn&OsdIasRoix0b$_Z=>qlBuyt|eDp;oPix7pyWGaHP|Y}H`+
zI;VhDjk2SOvshIP4nJsXRn!)^4R2p}_13u~?Az+CdT88paQA0=A1&SLt$^I#s__+V
zebK6RBX~Z%Ezlf(SFy?lZ_~wD)cLbTdEo6`FAJZ?qOYttR_3J^ErGRbSk@^PO~703
zSH|7=NX|?gxiW?N(EqQXyKx-ec75t*l>h8zT)yhYIgr2MZCQE+&ST!!F?twfgtsRm
z$v-|zeiV9Y^IdvBb)_d%5&Xn^)aINehIfqfIP2?i7bD~j=WKX8=L5Yb-q3UXIocHF
z`bWT1SUZ!o6ZUR}w|%(Zj<!vOwaKut_i7g-E9)7zpL5@&8L)8}?>CqnsXnlF0DTbQ
z?RGf(4&I(&duSAKPIx;P*6v`fU^#eBz&|d=UcBa-aQbaC@^;|uT6miYUv|6;IaEoU
zf8$_AAbCt@$W=nm*2R<TJBQklsb=QT5fg-K{|$u;@OC!59S(1IppX9KMy-UmVeqyN
zTx||-L$Xj~3u_;;R<Rnf{%GHBuy#RyIGcgnCbLl*-j;iVX9-){!`bLVCOljC0&lv(
z+p5G7v%}k_e`-_UZCiMo6RqgyflrOU-J%nHkp{AFc$>Km8XOJniFe!AiyD?CCS%rY
z`qu9^8LwDt_o5GX^VnwUN~v!-$v$VFg7LS>35uY{Lp9>8jp?(`_j}rqnWO0W<W2v}
zvyQ@ZJvog0p}~A6czc-FmE!YGU2HOjE;Jc4VByrU%$FNzVg@|5svXmG<PMq8?oN%X
z#;i4Z2YbTV+wis-TNijc1Kt+mcG(j2a;ic;QV;YboZB>x_uRw#@cG<$zYRm_F#&I5
z=9`Q@OL=TP^W50~czAmfeR~q#wy8lMktSxt;~%r}YC6ZGyV>~7@o3SU-=;d(Ipw)#
z;@B6+O{|vRr&oIPZKBDz$m{<-fsTf^|J>s9!P{#51}7)-dr)h;qlt+=k6bHUPt))f
z%(9tD&trI7i`eZOcsri{`t1)o)e+v#gtrgjZ7F!Ws1|eO;O)Qgc4AhiR((p<N4(@P
zeB>H~VDEc$Eww!lDyHajsT2jIqvun2vV3kPYs25va+XTcD|BneR?GufLf?dK)WOn!
zw==xm3vYYnrM?HRxW`QT?QSM_7T*3khIXBVhV>w4r#pE%qv??ZZ?6n?;B8Wu(-}<*
zZ@tP>qxzRaW6*2w;cbyWy!E4B#8bNh@l(G)=Dz>u?OoW?_tgJ*TYU+!$SLT#E_R))
zO`H;a`6nh>10D@RS7zxANBYApH0)bgIv_XRZ*Zb+Cnc(NI-6=XNKlLK@#?zDrs+JV
z1J5}Oi@loL^b-aiG{NG(h^Mx+=_5S6zaEzFX2vVL9d^Q|UN>xN0dJFG*A1BaXFKo^
zjNE9`g}R9<5t5)G+Y*%5NKnSf<od2)UU^i4Rxls-0dv{^I-IBgzeE*9TRs|q)*jDm
z;cZ!XTf8*Ph0pO^6YQ;Q(=v2#zm;&-3*GzNrg88#eNJ+)3c%Dn)J#>PUMU6rjo)3A
z9H=!F@r`Suvsvfitq;6iRFq?vM$9<Irl8koTH?i?ZEU(!$EKZy(d<7G71=yd4dN2?
zlHSO@;q8_&@hTY^r+V;qZPz$Wv&CvYv)IpUj@9@2v8wqXMl-3oeM4^YhukshL!Q)i
z=KjPq{H3R_erhj%a+wxCm7ZDbc?U=9HM}*RMCtXoC=IF{rI(4{<@%9+$?*2T)<~7@
z6sgbD*;b(!Vm0OmET`9Etpj1IF(yn;>xbzPyv_C_RIi7SC*=~Vwea?LmJkh{7p&#P
zZF^=8QrrE3x}PghZ?6Pu2D~j<^Q-FQ`l^5ge|0$JuZXjLIs$K}1=3r17G5seHe+oc
zEqU*))swunqMf(=!trR~Z6kR5?5$PhpINm7-u}DFsz*y%@b&<_ZT}B()D>3M-DuS*
zcsm&0cIgHKTUa%!oK?x_*cb5jUPjoOX3-?GRefEpDv_RCC+;tF&!UFssfo2&)Ep)^
z+vTn1U%WNb<fFjMK6>~9{`c`#Oipj*^5H$;{hh&R>ppyzW>&p|w`cQN)hEHCW1lQ)
zan7Rc@OC1cJ-^+eC)<eOE+>|YuB`)a`@-9T11x%5k{B(#{YZ|?=e+L5^fYuVeJZ>U
zQTs^j_TUwmdl(Pd3&x^r|DkWu>lVzaE#_vFbGjORPmmMUl>T@dsD124Pknfs1>W{Q
zL%q#WykmHK6W&g|M4!dm=q7aQom=>nPtjj+wZJ<x>k}8F#eMoDou?1MK{V@OUQe9$
zH*DPvXLGV#SwB|6Vz$lT?N9i5uRmPwjb>$??a6!hLW}mr8-=ymS%>-&rycY^=62xu
ziSRZF-aa5cn-{G+wk7#h<(N&7iTEwP>@av+6W-2+w`o3{i_e*j-Ed~zO8A2(Sq$FJ
zfw%qO?Sf%=kisK{v2EaMT1I+A<)L1d)sVk!aCI5?{fW=^=Jpa+e^!-0_h&-$^4KZ(
zklmHMt2EB*u+woDj>6c3oad*pN^!deybZhx5AMS*Seucx{2uJN!*eafLgVRclz{fa
zvrdEkIpO;^blOrp(TNve8oVvBgjyD~=F<J}f)%xwUJbk93q6tkLT_dDOx02P7yVq5
zss(#fnPKKsa4uLl67P2%T9VZPwz<JtE6We=*~!Umanz|46IzhZd1et_I4qpJ0EQ1U
z8JD}8jCPah%>-}Dj!IST0jXN*Ox3cdX)+4Yr-`)+*6zy7Y;|~Bp#b?)#qn<8?Rt1S
zV>sU%-jXzDc$GF8KP#g>*>6$yu^!&$UTQKL^O@%GJ>ub{*TFv?yvfMN%Fne+%hJTm
z`E916TZ1ce%nj<0;O#|@(Y{i6&iqE+{5B8Z?ZPmV(U0Rff%h?U+@GSOqna`kuP)!E
zg2@P@H`KnLX^L%^ragi5Uo4cSNao79!`m9kPG!77?awj#0`G*c3!OSf&F#~6=vrn0
zOoq1`$d#Ir&#5Bmo!aT0D(j6@CBWOe<5D%qgSi3zDVj!J@@;aGTRQ0_3~wh~N>;}n
z^eaNYc7mY~qUp_apI!}USf7jZ-aY70z)8I0vkqNag>IcnZZJ7YjoQJ^^?19~o&>kX
z=WRxP3yfVbklZNltH<M6TaY_dkD8X^<oy&R_8L!)(hs}VgxOUG-uB;U*MD`0pH?TH
z+mO1~&iKv!$Pb=HeN8E9Vmra&K4?1bckh@$4}%2V^NCm9rtyjji<8+aPU#e<toh)?
zgIM|)#cIsDSVi86)r?kg3SJebg%9G?Hz7{mb>j6QPlDEvkGr&Zq6UQ~$_w7ED$eSR
z2R#tI3vZpf(6#Wk6uf<Y-KIuY@Rg5Hy9xu3!`t5QHrpTG!qkht32-QZx|0Nj@cH`h
zrXQ0zQKw+>e&({jIE1#fB&rF0B;G$w)V-I98vHs@xvX3dpnLnd+EfDGt}cVW-N>e<
zUCF6JpSw)NZ{BNDuDdq<$jkA{&2fm}GkoKkA&NX>SbT@$)2g&x$MTcUl?$KQo&2jL
zJme6Yx<2DtfzMwIJspU)T@7y^B(q=dL~Sw>m9#xUzuPD1<kxsj92>7I5zHZ2AE(J3
z<McCAoaW*sH{XT6ts1Mx=-c^!cuTHSw>&ZG1#cH#{UxtPzhtF9a@^f$o&11@%>0~o
z)qkinHMdFUqqGIy_Nf%5Lb2a<?Dlt^v`1>(he-ACjMR->%+1>zE_01=jej4eV|&At
zeN>o+)D6?oT;EjbaH!r_2$lEi5N64Ts8EIw74IFatH*<6O%2qxZh?x)7pPnBU~Pl1
z>hnH8RksD`=%4^4_43!`3w~Od8|IR`bQ9ig+32G?wR|+|HJlsot((MZ`vhBc`mR+6
z_gNLQ#i|^ut!g~as=t<4H4NVV2h*nivgp5}R*i*uD;wbNwzg_u5A<wTd}nyOr!0&v
z$QG_1g10BqELxXhQQKsTs@pC4h{kojXHo96#8p|wLyHcq#dqH6tw=9#mE7db94c?+
z=*{gc#BaUP!l%*C_|+Nq;1%;3M(nca4qPp=)gto-i>~dq=r*ja3uh<&Yf*mg9}I6x
zOeVGqbDvIvwf*sxV_<7hW=U6dH!^3VmwYn)pbjvzoct)?bNI!q`SA8WygfwR)^z}R
zu1(wwFL-N9bu|v{qlSZie6uz)f1TPLCv~*B;BCrjVtz-^mhiSEy!C*$jp3~qysZmw
z|K|2?G;bWtosQpG1m1Rnx6gNxo3zixC;@Nx!`lV$H51E))fC=Z77|~a=wcj%pP%4q
z9$0(6JMRH+Pr}=G{Jpa+TDTj%4`FWscpEmDee(SB=-_;B%|?J98tn&qhZrLXa7ML^
zXkJ*mH3dDJK>vzZ&dryIm!gk~bVMt)re3E8%&dfFLLZ$jgJ)ZknoiiW;*Un<-0cNF
z55dkktR5^+r^)clfX~c1y&R04#R`Chonh@_IJ*oc_J)<$VomgAFd3cUZL6E~S@@S3
zF*{7gO}JQ>bN{ll<QBo(1^1cx0&CAdWoFD>ld<zQGuQB**GJN4f%i$_y&tm%^ZtXA
zVb5DQ2yZ>%ZQdos9IN3&XW;qo)3l~_nx>HF)5w;pt>o;SoSdrXvtjM#R2~1BtWWf~
z9o^BX^{t%xMQ_2wot?@t5}n4%Kbo8$a&8Z{bSeY9ZL-d(+t=wm_|Qo|AgA`uCr2nV
z@yvu2o!Q51&|N8V*-1TVzf`4sOI3IxY+?pY|N8K{QJUhqrRn@zYHH>)o307>S4~qg
zanbnp%yGMtrp4a>)5Cu7cJvUwXCIUCXfVeCmS={y)!^-Z^!175=zlnTX&%RKFV`9{
z_Xu9YOMavKF!l*rJty1McoD<N8K28BTNI8L9Spkz@bTg8W8Npn2)yeiupj>C%8t*?
zF{@!?mQOUf$nbU>ylo%Ne47-fOda^Fu}=EFJ9U)rUJl+~hPP*l-+FS~%E8+b@OD=@
z;<!1SiZDC1&4*b6^fMl@ou0<b4;Tn<d%h)xTZq^#ti2USpFsa)^*u*_D0mx54)W}q
z<PDAE{QfsF#d+krQgai2fY>R#Eq)4rcm-^QwOyy6Z$pW*ex>#_#IEJF$dNLL(RP8!
z{i$t%x1Hc^<L1Oz$un*bZx6%UM0ooc-d=>a$-eYQ+<^B9Z_mKnKOX6O@_rVu?+P>R
zD&epxI~;s@fw|_?iQegupa^(t@rzf#^6?shCcD1`Ml_03W_UXR-sXX~6V@=l2HyT^
z8K=LO#xa|L{2zGx8Qw<bLhtrXP{E@K8c{4!nLfuW?s2^QQsecdd4jg|NKm!W3933T
zL9372v>m_LYY*CUn@!zz(R=Ahf~;o~wD)9!vg}OIY_x6RQLrn2f+q1EQA*ITn+Y<P
zN>s#rdLu4Plw}q<O0f6inMC!zmZ&$(k3I4wQDFgzx*X1P(6}D(_8=N}ST&pOp|vkf
zfd6~R>$+!CB$|E)I`=iaeFblGez)m>&8EC&a-ve{Ujb*c<RH(QZCV`n<9laC|JOTi
z)8t8b0quD|JnnuU6Seg%+eY+9=CjX-x3Tc{#;kZ{3yRYgcw47KoQAl^De-lzy6lP7
zBX~Op-ado3mT@t9^&HmT|E0b2eyMolU#fKbr;dHWLtY)N#$L>Zt@cBK>3-<v$tZ<S
zj8eynQOuqAF0XsvwXfTEHTcd9pyiQzPCw*B^kz&)+_ucyFm2iurl11kCNKJ?3E7$T
zb1+o(%Y>@-tq|hk^i(7_`EmPTy`b)A>CZsT>KLd3^k}*QZ-3VLsyFWfG;2qI90LNB
zy{Et2shy3l%cA~fab;h6YxxkX#aBK<txZdBeU7(k>StoM7pyvSfOzXxbSS(nfu{Z9
zW>Fc(7hP;cp6^GCnq`2G@OE?qykmG<th-fS@U|kn?FVnm7oc7T-nN3byW#Dc6pOmR
z+mX@K#QM_r3Etj218Z3g;BD6&R^4dCz80XDqlo+RS^oQJ)v;5&=NZ-^t7g`)s-~Ak
z9<1uH_843Z-opE>CodV!X63$;n=HDp#G+MGEwZ6$M@)dh-O;(dEn3;1`=i~A5LOH7
zJ9=eyHy*rmGfv|j&xW}t&bS%R$zL7XoY?GY>P#BD8Y|t%KYmO<#ZBqxYfp}hC$nG4
z#c5NTnRzGS>>hmN{p2ihdm7uLXw|&%HueN@Ry40CKJx`MtQYqmU`?CH?d9YFuSB=v
z7nj_C_sn~=-$=a<d~E<{53!9|hTpq{8k4zH%fQ=Y*!sK&UNhU7ZSbipqovETk9saf
zaS!5_@OB%Iof*jf;O!gMPI&wACAt;uI@uSmH;TR$CDGa$xE_GFZQ<>Tcw&-Y(VrK{
zcbfLU8l9%(49|kMtKsbvc)J_ket@@;oXelX+rjX*8N59RXJ@lIvf4XMMl#xI1H8Qh
z2b0*g;5ps6uN}O-4HFwgQFj6t!V>Xct+3=WKIS$!N?h{4eP~xSTOW8Ec!fAFysZpx
zx4_#Mu)%g6P4^zI!rS&S_`$rl$!s?Mi8C3){fPI%?sz=WM!QVL{e>{PlF2wn>~Z+e
zG?k8`uMtdG3<nBahPC5UwRR1&*FGjI{9Up#Q0H0EppF#Yet;VZ<k>p$NT&}cMhS2K
zgSUss1$qZ_cdd15ARL`UK5bT3%i{E8pnmkouoNxWk)oflHfDQ@=H*URly9o`|01V|
z`E%Kdrl}9hwM?37qLqi{PScADY03_7-^1H3N615hw>9yfpI1rK*zsvv_>J2+$SWI!
z_8*7_Mfc7x%(2OaKFvtK;%O$sy8(H{wa}(qv&@9Mt@wTV!`b8b;yKtR!P^b|J`?cG
z3x=Zk;qAyrCgTHIy6$*h2XC{(+sidgM(L4hx*P&my`38RjQMmooGSVbj>6kfaa@nU
z+PR0he=oH^Y~Sy2YRx?S<binS9Jij0oN7_QsXB$6s-BVS5__tSaSR%gGv$fC{jXE1
zR+ED~fWCkkt|sXyH9m8?QNJmAExOZ>*bD7B*`b%@DK#YDCVzMGa%<q_mV}MJZTfna
zbN_mq%$_z?97LS5GcnLm>No?4y#`X-@)Mm(PVHW5a?+WogZ)kahzRnDg6z8Eg9rS<
zuGM$Q7rtWGgUj?ZVy$KsgSP<>$aR{<d+i~13MUFiqTw3zdNf+e+KKvdH9-d#B`9B;
z1eGtGpe|Ns!`+ToR&?2;eDS)x8m`WWQw99b#Z#FHM^0_+GO<dRJC^*FSb5}+)fal|
zQ{|@Vlj4*z1ueWVPI(qFBYjz%PS1{4Fs!@+Z`;D#@9=gZUhCr{<Qc)+<_{Bei+zvy
z8^8E^g3iO+9q`t%BZ0Uv^K<Z$o99T-<&=2s!N0D~=RI1Ixv^6cbz)JXn$JMvPEORj
ziHVvp58lCE^N~auXw5TW=xkTw#ifX`!rOsWVS63&xyBO9#aHeQZ>ytctEZx?!#Fk^
z^JyF(#oOf1{eL{?#qhN~YYDoxa4Z@a4L;-v@oF@BkAb|OZ2Faws4N!~6*@CfEea><
z=KchY;<F~h+su4_f397oQ47?(Z5*@r;uQZTR@dR}Ie1%q7CkOH#V8NC$@O3S)})8O
zR0`hy18*BY_^G!aq80iuTA2<;s}S|JHE%?z>X;}M?GU9OWus(?`mV`Kzw7^)I?L#$
z67Opl+A`<>gUjIVGC1tO;0}YkI}Gmb?(Xh7xH}B)NuZ?$w1rBfY0@@n)A!*&zYp(+
zyOJiiX`8N<=k9&Z*(a4bpi;5lRcIqU$#(YZvBU0gkw|8)*#{4A%jJ&HEqJ@GT(~aJ
z{-$d=$R$SCI`u`{Hh<CiW?y969Hy9<P`O)%D*0N7cKL;<`zjdQAXw#I1j%oHkPcT1
zQbBlIyz*yFn9tno@IY-@Pj7PB0CoL`jNw1YDE|JDo?v|6bFdD58@%kJR(3G!c_XtH
zr7>$pf=MU2?^DyP#B(N9zG9+(+NAo`;3{77ySDHb-Zq1`*K>Z<JalZ!Up}h4lbULT
zS+l~;@^Cx~K-=ClYtM1BuAV|K|BK(7o%x~~Xk?CbwO~1(^`Nydcsn|K7VP%o3!Bll
zlhLc?(6hPCvg9!<yec)?-t<PHXNSPv1uOVn;q3%?YlXLy#+a26-tL08gE+6^H&`0&
zZ@58b-WfFPklp`j*~%-~p~?)&LNx8nMx3*pjM{q4X~5fucj!@F$!rh29Y+uBdHPb*
zF|$)0#-`athT;}jNd5KLp0tL9{b>zf(707#ZaZ}AU)$0eit}7rSnCImBWI8eH58w>
z1GQcl`E$nS2E0v$yEkF5oufXywZYqQ^I#~RaOQ!`&BD?T?eVA^!Rq>H4TIrrP)YPO
zysg=cd7+NX(V}^G!`Rud_9Vwn?$3O}jLlK9g}9v^h2IEgZ^7GQ+%|x><KS%zc$*2{
z)|f>8Q!Dy_;H?vVbhiQ;6@PL)_16Xd)OCHS$-=lj9Qo0*^Z7a$j^1*kITNXu!qzEp
z_BNXH49{2OJYTe|8$Q;wpvm~m_kp)P(YQZMc(GTgO>UxY32!gJ+Y9@B44XN!z>O2T
zspF!**2B=<aM_P@i^1D}c#W1eAH#EAe;k^*NEjUFdlqNJ15QlU`t^98FgE)QKJ&N3
zRlMU~f2ZhM8S;$cVd7Ehg=o-U{_&{pOY-gCks&`KS-%fXR@W}c>ivaW`B0Azd}coy
zJ+ce((JM>O?9+7~{k6@bn0@SZA>XKWHvCXDT414M?aq~~{aZbncfzB*uklCOo%$?8
zir$!0)EnNuhPOF$!m@(s!xE`V18*C`+c#z4Z+WsKVDK<zTPGJ_H*CICm4m~B|Hs?P
zXwL5?srObvzr))?`O*4l(90aBilFCV?R;374ITV~Yo55=WLB3+W7tv-&w}fszP#5V
zk9hxQ;IWUz!ymzQ%rI)@@OC=KrZT*zXxD6ey-Kset7a>`a;?WR-ho#PZ+iyewa0kX
z8{V#fx0T@Rshu3V=|Sf9=`8Y)J9<^2nO8IFdL?+958nO;Z)@Nu<onF~4{sO1+xuiQ
z&aIlFLxni<kbUuz9XlP!NQAege<bPZHZlOmp<@rzGu$24`ZAyL3=Td>(4?Ep(!kqy
zE2%S1z-#3*d?CC&3KO26qTU!tU+yb(D?PQDe@&7%m02EmdjQ_*1343S61DRh_1Y_V
z$?*0roE^b2jU#wXqSnCM21C(hJLxZm6CcrW6JDX~49v4+NYKEAF3lS3R3@1Bu7XoW
zbG+`s+nMOqm(3k=7KNL?ICO3ed*UX>X>XS}Wrw#Zzs0E?yltO9P8&wR+#GnOg&Ycn
zw{76<FU-C?ZWO1jyW_QqTz;x;x(-v9^mEFeSsBM6>av^Aexu`5Yf~IGu{f3J#_M#9
z)AwTR#i<ad;bY@84gdPKo9shpyh`FLPlVgQcAz%e&8fjtojSS1snzG5+H2-_w>!1I
zs!Kcepf%CBQ{in6v}bd?<)(2iz4J{_&p#6M6<>MZT6({D@8_UrmwD*JLfd*_symWv
z9dztHc-x3Rt2$hl6naZ-7Uq_)lhYW)_0Yov*|sJq2mh;y4H9$}FL~ELXl{60Igd;6
zd!6c-&nf%ic(uw;hT~<2*7b9!8#A}pF2*ru9H;Vs#A(S>`lQIW2*_=hd%I0vBdr=w
zZ*rdyi<U1z+cvf6v^7>8*|Xm1O^h04jZwGD?2~7w{ES-B`cybtj(1Ut`!h-x*-cRE
z!*|W>`dz-aBlWO%qzV+qQ$7))+ub6_j{J|eE6Rtf`J!*iPOntxoiAFD?u%AsAlIT=
zm@1qMRgP+*dhZBPlS?7Wk%WJIFjx=EF#q#BNS_vvD^Wd23*qhXN}n}yUZ7IpZLLiK
z%2p~sRq<`Nm;R(n;U87y9<|pqA2k===3b7@g}1w5Y<_q<)o!9M&!jIfw&xL(X3aIJ
z`2>>&3^D1(AiUvWW__Dz*4@KqHH5c?(|%M$l36|C;3S-Nd^IZ|%&g2GsM+3tyNAf8
zSZUV7k!atRX5}i2|C?yi>(3^&iZW?PKC@1>=Jj~p@RR%w*Uj|(n^mS9Ge6nP$_;OS
zgSS<BnALX)=ki*o=W{JE-mF7?ng8MOOWdyei~i+?W{s?GR%C#`!FrSIA?AH5Gg~`m
zhrgjNJIjYG!#jqzPp05EH=~X_jy-4O2lajLYxw&x^Rw{Q$Q<S<c$+naT>nRSxreE-
z;s<}EuecjW(suMMjC})dHy&Z`hWj<(Z3VRJEP7to!qO)2w%kZGF}%G5BWs~y7vd#*
z7f@T}x%c!w4Md~XgRR+z!(doD6>qo)+IAL?J#EN5OFej74!^hry;$(}SYteAo?F7P
z6&5#agAa{|9C3_{WBPdmkDz1Ur7`#=;5Vj8Yp4is-Q520pn<5}M!?&vyXpCzhzAL8
zt&N!Htk3)ipV=*oFx!)r{#kfi1HR6Oq50t4Wf*%YiCQN-t?fj2a_g6t3}u*m5#An#
zheJ8nik5Z3+syE`8LYISv3iD5r+h;`FAQD2k=p1AUSkzL=2~{A!p?N?wj8`&25+ap
z+b1XK6>7lF^4sX)=V;_uv@WmT6Yd7X>kI3cQ-M>@FL*U~BboVVP=7KPHXTAwwNBAc
zcpC<9XTaO7N0QYK4SHvnM`O=>RDqt@uSY!U14B1*YYX$p6yj0oPjHi6C=Uj}Wq7-3
ztw-bF?QwWp;ebcx1h;Pbv5O|RNB0`DlWZJ(-b=15Gb}DVQvY-*dftV7aqPJHr%8%#
zaN7|7I^znjI-NyB!(ac>soGx-?F(-Q!rQ;#?Hm4%7@Joy)J$K&;_MSr(Mi;|%h9*|
z65Y#ePJC3VGIDKE2HqyY+Ip3I3~f24-t{pQ%t2l<y#2o(a{T2#!r|=&>eh$obM6CU
zhtr$Z6~7&y_)DNy-9LIYbP@V=2{S#*$*<t|%LjK$!&~z?uhzp{3&*eg&J*|J5s+=s
znddtGkGF4TdgXz)Kl72zwY}<J!K*Rw_F#IiO2FIp!My*cQdFsFif+|T(Jpv<1>SCl
zw=G||)!<Kd`B1A>RFaBqCQqR{9&ZP-`YUjok-e@@nc3W%pxg`T&7BQnr*a?O4(pwu
z<aYG>jG&(v-rmiGCWfc8@r&PNVJ}UZB<1))rvDjcRaT>MVZd<MFw{)m!dqrqpZ%x5
z#@>hHcj<e&ouIO?wLd;<!7VNYPsVqpr>6<|NB8h%2Shp*(wn*~9J`+vo&}MAbR%AV
zf5+=tFNb2_?PGX*6W(4MPc}p6IPEM>zP{bAXt!M-(6@sN#A#c>IMs%g<-B%<keO7M
zqj!Tiz2^IBaO_(Zr;N*}S(3}&V_uxRgW@y*?tJ5j<bG<eI2DAqA-u*|cpF_9{ReML
z!`luXhfc@Gt7>kT3UjMec4`{T?Feu8FLr7Rns^<&EfnQcH+XA+w{_udS9seZh1zbE
zOY`h5m4vq^$|UGjhXn0hLk<=1@4$Z&^qFgt%CPnzM;Ey27edd}=LESvP^-O8J^2JP
zPyDZvZ3)^7Z|ig261$1|FTCATpBgT_Equ|XOYk-mydAtA9nX8&_+-4U=8acDczXce
zW}<eR=R%yi@i(u4w-3p2oLt*ZU$#wSm)Vs6yH(+2C9nTs(T2tJBg5OO&RDXmVior~
zMoWK-VfPX~GCOWh)QQ&E-`Q0_Z}RkNQ3}CF?nRbVgJ$1#<$k2r^kV;ee)iEGjZnod
z5n2=xuDLV9)db#7U%(#Qq_0YS`$eTA!<1)n82f_5)c8cG!qSJzWe?G?lOc+V2-YQd
zyP$lq^ejl_(YJr1Z!5#w`58Zx{l&~{u0S=~8lYCi*`ot*Uz8&E{~Os&cbIL(zs(G9
zzoKs+mqANrG^<a%NwcjceR^tA@*z081|F_5Y0G$%(!$#&15Nb#nY6zYjDxqG{6Ffp
z*Q^bRXiKY^d^59b@OD4Eod$2G!`r6^%rdRu`H|FI;jObY{Z8=q?HQBqhM5$KhMn1(
zyo%w>^_=ASD`uU6o2^QqY2odNTxQLyz^pA?tq*T=uQcl`ynP367x%_j?#%DbbEzH8
z%FFk)gtu?a?2x(cZ@7-GJv5u0EX&D7n2*OijlQV?_{w9+u#F<~XNsR;UmI>eP^Ue_
zPL<{CV3|XXMGJQ9`7oCm5A)Es58&<Qjj)yD%vNS*(YPgG?G%n#+;4^UEwh##E%5jd
zybYZOli}%ObZSF*`xpEyjXrJA4~;qkCR2}%gT<#{=vtneX`tq+w1)ZAY}3Kh-aHml
zlh;PumM+HpEwejw>yU%dmVPQ&Td*Lz=AThheL%162|jn>Z8vz^`8j<}XxW=++6?}@
zHoTq8{od4WufD=_T*IvEX!zIyKEvCF@YeD_-e!WgS>bJUc)J9KPJpvb;piD|*TUH{
z@HRiUmw0Y4-tjQ}%lUl0HXa?zbCK}&8;qPopH2XJbk<Aik+8Jp1|Ne1#vXyU#o%q;
zHRxL&8^+g5&|ZW6Qq>$?H5~q|t%Ww60gG0osy!Nc2d}*vhUeK#?Q#K}f=$8bv{`k$
zvQ>hw8PQT4k6JTVQr@En?bzS;%A+&r*H^8Q<<peM-EPhM?$$AOn`{jxi~X=i;qZ1J
zyv>5P{X(wMjWD;q#j#Ju=~mgd%ztw1`{q_Bcst#P+HX#ef=aL(v;#ZI_IR`(-Y&Fx
z^dfDFzQ7olAw}!{Owr%)_UBI5_#yQ2P9|?D(yLtP-3jn^J(_nv_gll*)p@Do;@@83
zek#vZ?@S$aA9FZ&*&z#Wy&qFG=trtf#lzgVRGs0r8d|qWF+BMwbaVvVhRFl&r&1r|
zwn?g{7fMyjZ0vg{-?(&h-XnUOS}*kK<pLNwmz@IeHWR#cEu&5gbBDs)MKE<Rnz!;c
z{PT5Q@_ngY^F5<xlWnn`>jCPv4`+C_8s1Lm=G6mudmrBZR^F>u#qgNn?L7SC9=|2A
z<AA(D>(Bn)Bt;KQQtvtRKusm*m_B2l2V^L~+axk27MFpI1;{9DLta8T=1bx2-pBZw
ze1>OONInz19Sm>1<Eg2_+uHDU-B5Ci(4<{*QD20&4bvs*3I1{4H~;O@DY%<Ho|WVb
z+u<s_9RqK(yruu>8S^^uU?{vDaEE-sTj*Isf}Y)VX*XIn@-LTamT+nHI;W1KGmm$5
zY9_3`yeD1*JaKyUo&MPnbSeAJD_x1xp}ldkT(_(56SDMS#N5{C+n>C(*tHej&LaQU
ze9*45`<Wj-VAq1vc72AmTSLiAsu%ZvoKAt6tKqF5ycN7{3U9;EvxoB8G~rKj`6t_$
zo3iQ1B)je&vg^rLIF4@4JOjVAm_q}z#w!E7tpRUy7ovt+(W#+jo$AT&-3x6Ve#)sL
zZ>jVCpmq#zGvX=F+C)Fq1DA@0;XA|IbvBn4rscgYMNUSo1YKE8kMSP*rO%@O;q7kt
z+BKNDq0j_5x%CF3f#Ge@Yxn}ksQ2=}`o12&fn&)!&V$!uyYSxDV0M^(?TwdQnhkGr
z<mdh_r%va0>WST<UMC%7k2zF(0Jna40Nc6t;C<@?YtP)U>tZdt-jbDkvkrMu`{+q7
zZ&gh<v$m6&yX|bz?>XsB{vM-R#uydO7Nf=+$x3b?tt@4t6_F!a<zGa}w_=nYO#h+O
zTi=zdHM6&uBQ=Bl?5lD`l0zM#_8lVR6Bw?u-ft?o=^H)n-<S{js<d;y=z4IN=8@AF
z@gr0Xn}sT4LWtT2gs9Ap5LLPwth@7qwW&g|vcuZ~@U~!;AjLoUtj0-!%pM17S&2Yh
zfwz8BKdE$|PpZ*@-F9EeAHGGu)G;`_=OcS6@OI~zbs{}JEUcZ1=A7}!q*n(_a?dwu
zVGol!);7tN1Ff6Mq)gdOiZhrLF_(U&@6=~gU@F?SAH2PD4c!S_)9_sSA@n7WLYq#(
z6aE`r+smwLf0{L{JbMO`Oo}~jQk5kpwSn;#ezsp`vo8M1OwVj|asyafoZlxKnl%R*
z$py?Z7J=D37FpTM-U6812Jg5lzA~Jh(;n^Hmin#8daO<66!qFh@U}0!Ergc!_G0E|
z1{oJ4*}u}t-|#(?zoF`GvWiCf8TJ$)lL+1>?qx3>yuAu<1M2%49+QLM^NzWngY+K3
z+jg+_1sXN)dis-g!C`c7ZjO=gb|9QxPM!8FtSvQ*e&oqyUclS;=+b1kI;1Xhw=i`x
zy-?L#a=!t27FDUu!r1Y!bt=!*<$eU-^N)7q2J=|o8oUO)9bFXvIWL-*<JSW0DnR$<
zgtv9zZBKaH0UvqacII^8?c|H}H{ltNrw4iA8TJiu3_cHQInMG}Pk1{H-p*`J78JZ)
z1aH68qW=kvln>2%3WmDTuX|x?b2wWJ-u~3C9*!I^HIdsBFz}v(-eNq={oKyy>oFXI
z;q7C1+X`J<>XnaSFFJJ>daLGo^y*4{VEFnoJMheEyyQW6&eSU(_lD7__?h^z0nfY&
z3P!Weq}O*PjGE32PY>#jZT_2C831oz!KNdnIcIS)@ej$`HPxdK7<;pZM^oXgsj5d?
zDzkUAqDQ65vSY^Iqt-C9eIz?jKayVrZ=Z9Fc|%s<Ubh<Vchkedyk`hJrrsFy-mSlQ
zt|`22@|BqszIWXJcpD0D=eGA~D7@_rZx2L}K^U2=N@%!+p~?DpExYDmaGGYx<oqV<
zUBMJ(hDY1$d-Zo4n6=ZZ5dPMmUUDgT+x|P5!(2OjFUkx~t5l8IpQ<}|sQ130p9<c#
zg}3t@{JnNEpWy9?bj&Nk+sp7aGrVmMZ`a=9d@?H9@|umgW~u%HMs}u;7~i)SyiL=e
zdh~Gmp74V!!duH?{NZJA7T)%Uts`NoFTBmp@Ai_fH{mmfz}WAc|7{FC%iPa4*sJ^l
zsO`erCe*&Kk}c&$-|jBTY!H86L3sNoysZsym%-a|@b(M5?Er79!P~^?^fJvzQo?w=
z$pLtgS72X$dW7j$I-Zl<i7#j+nArFdbyYsMN6bu6#1t67`G567OEt&W%t~)hF6M0V
z(Mww|L4E&bwuK(3Zt!-*S>{;aZ6|nJvk%|%#-(2yxfJT}Qp(wQJv<dJpU3gq7tJp7
z!cJxB<J74m@Sqnp#ij8wuctP9C|+%*;mcA}wR3I;dX2sGMAf|;M-C);4Dhxvyqynk
zBj9aaczX$tH=}=?uAo1&MB6pmW!LzVcBO^4hvDs?)F8XqZF0rfw2+LXTFjT;eq>Ya
zQ#Ku5Z_}3HHXS0*A@qVxFABogY@FA~rkdkyy13b<<a&0^*kRWMc$=#Py+mW;6pyw}
zAdktvl|v24i^zM<q4v3$Y4u@7sIgNE2Ril7B>d!aPPLD4s(EFXKF@J!#0Ho0!drW&
zOKZ@%u~wI|z}tg)(XSN~bROPr*h9@0-VVVB2n%4|3FiI{YkzRq;O#7Ud+I8h7~V#4
z?sZr@fnyrHO}mPEJG?zqiQ0P-85hf3THe8>TIkzXHm7#cFV*9wLsbqq^m>~^z2U8I
zbex`Zosk#bhTgX8$}u}VL3Smw3uxT$HnliqWv14u+OJ^kAokrd162EAtbTRGXz}70
zwN4YGmUE)D8s27vw;z6sR>!00+xQ=H7y2Qm>AOb3+Z<OS)udyj=I4ymRd_pD5sLa4
zu4^^I)pp}IJ%G34;qB5{U$o<Wn5xeU)4ltlTJ&eAs>L&FYYx$iZRA^A3D!b*yR0(4
zGWFZIaY6E}8>BPvw$j-^H8~WZgQ=hN_cG>PM|@Ih?@#Jmlb+-%ANBGzn)57r^ECeN
zNwcn_54XU)vB4&J-<$OLfk{E|b|}0})5fH(mC&_GMt%1+$rs*^uVvDjYn=Dcq&jgX
z&CGAshel={TtLs%GHzjJt|sj4sYRc1O|!C(fxqxJ8Qu<nw?z~9zN02GSWH@S$)xnJ
zOq$NmUB}PA=8s0rYL?e+(xr4<2mEGMHh8<Quvx3%ZA4{w4sY)@BR{z{wcHM9YHpi0
zg}Lyy1iYPf7vK1bzu{Lr<bb~ZWIMv#NqEY5$)oU*Z~j6Jw+|VC8OUb5P44kI@`HBz
z8se7zr*HSNx2H-DGHl<XS*gkPL(^7-w;SN?KP&KQ7n6mIo}ICdz9eo}txjuryELt#
z5WM{b-d4shE(}{QREN`L;A<7;a^YzK*lR)4F0W55xGeL$+>WjUJ2{$`Mkn+2KYUNG
z%IIWjw-JTuPcDduoQv<zNv$^nKK5tw{P2zI?`0+j&URXeUrZjuZ}7GQyxj$F|Aws{
zVeR;n^diIK`-jPl+Jm<3#P^m&6E{Pj!rOU2^)0;J18;M}+okY!J-j`Jj~qm;_E-2C
z!x7Js9k#xOu^nJ*Z|-;Jer}G(91gzrp~va-R~`@aF|0$|re3Aic#zrJb@<H7@R*n5
zHP4|33g(W3x8A<=NWt3+UiPK-q6cRw`3WDrTCxXymK~3IBy%i1$&JXxfyNsSZ>=wx
z1>KpfVr`Svfj*}5<Pt@&2YVj8^}yGQ>=0W9ci$KDD69zHFuV<Mxm6*;Ed#s_+vk=x
zkrTL$TH<aFc-sly-hYRVg|#Q(tmCy?$uReGfLqOMZcR*dt9uTQ=9KcN7`z>`gP(&x
zZ3y+~5;;osa^NF}!T)v1`p$f5=?c_{@6s=HFImOolF9E)kvVIMave)i<CfG%;Z{!A
zwk4ciBl?=NWlmLEbbaaOsp{AvRV{j?($AQxq42g0tet*2RjuIfCqMeBl6bEI`JRhj
z-QI*RJeBtY9@c~BJ3FwiXO~y;e|a?m-tIG?N!xi<lCLN9hL`>5ks3-5_B5|1!QCPJ
z4)3<mKgRoAf!BG79-qT8uqUjA*^S`zvDeJ^{6^+6^Rp3n29xG-4b$7BnLRw(OrLQK
z{ZLnq;1%QPhJH$vg=5rP>ZjWi^;h;pRpm2!%IO4kUXY*)@H4J+f|3@q6ZJCvvG?(e
zA10{MGyKR0=%kT&nw{v=ZADG6F?trRwr@%f1D<D2KlBy6wZey3H=gMWdW!IWuU>H~
z2DS`a>(u$lc);kpsa2f{q855@dA#b*iq|oG%)cMSt1`Sbz}vNpoGSaPOFhuCgTC?E
zPu5Y{)($-!jJNwGPRs5v<3rBTzj(v`S>iN48w|)~Q`}ptf@@h7RmQ3f#jT3|)2@Cg
zun^vM-iwwUV^@no%#7}_X~cY+nk=!YWw=eB;cY&+yT)wOl0h~-?_|@7RyJK3YEzr8
zHeJqP(@XZKPlvbHWz&t(Hl5yKQ}w2H<=<{sEAkE9ljHREWt?)*&-;~KI)U^Wg^Xne
zi5{$4FCCinK3;>8;&q~^Q^nwIZ#?DHaHp2T+vo5$bgc^yfqt;hc)aj7gVm)WX<%$_
z=8DTE=x`&L*^9gmxO?>u9y7en3vZinRDrjb_<H3%{A74L8{Srhw*@=Uzg(aGHUk-1
z-S|KK{NI*FBPY6aZYgzR_S#;Cw{PI>PIy}p-uByvAHS8od(9k5!XLleJx&wYb9-zE
zzI!LTn%HcbJlCej1#HaKSXHQ`Rg)iB)M_yM_tIO`^lq%a+ha6rCiPo-lk-f9)|=YV
zYMMD(Q=Ug@*`_GPI({ey-qt1O;%Lq9<X=Q86@8oL*GSEVw<{Y&$o3{&m&kBzlI@%R
zob?r7<co^@^+n&`gsFVvF!?+R)%X`Mc3Ox|T@O~7qQTmIDp>F422;ZdR?(M1dIoQg
z!`qSY_SvaGg&Yi!3*L@g^hs7XHQF8YH&y?r`FZJYx{Ef1wP$w2+X?hYjWcU!EPTU9
zUJduo!`q+w_ViqnMzl7mS|!+;YSgFCMt(<pW;E~jerVI-_|7odvcaTUeN9T~k3Jr3
zl7q*~P2;g~CUqZX($Z=s<*aGa;T9&HX=l=#lO|QCHr%kcNgMcC?w2Ob=jS@9>5fNZ
zr}Lpkj0Ua)Zynjqih}2>ekVJM+lZXxIp$$j3BJx~Mb;EdzXa<8;Coeg`y9UiLnd;G
z75;_>bMcTn`x`cOVXl^(-H%=U4TBAM%MSLE=4UVIYj&N%+ij=WIk1)e;T?Pp{c4gM
zPp+{umR_g3aCI|TPOvs3M}4$yGkCjq5gA4>*Tu07&W0|Bo!m~Ci-$apbB9y2g||o0
zu$5uuyYjpae(=avc*pQ|1YF$>Z_}e!pXK4PBJ8^@P9HMbbw0eU0dEKJJ>!3;Ud%bw
zc#Si#c1dCy!`oN%INc>f5r(eV$n52OG8m^aKiUtjwxb4H4^Ox>%%vu~m44$MCFqOd
zxWn<fIK5Ono*mwnfVVr~?XcQpKHw>PU~QRH=0;NRjN$EUcv~DzJA>L_TaHd}wK+$5
zc<bi29)9m9>a>5-3+sfnX_-T*3U8~y+k)^m-!(Kaylsk}JviFOVE*IhTrBn1x9pbb
z#OHb!`f&z&wQDqa2|c_zi-%eFJv-f~8CLrj4rD_gen7+CfM3^>wTRpF<nU(P$!@nH
z^f6IqJv$P<we{!^c)J817X6hhh4l11rSYh_*R7)*74V6_plR#F)vsvUV;k8WvzENT
zb!5bDfxkTd627*CwG&^tm5JjE%)N*I{5QJzHm@-kja!M=x3%=>L|>1Rme4cxf<5SF
z_T9ize|oCIQmGXmrS^LS{d*u;56&g4%7f(pWAQtD=ZLE*D(Qv2)Ip!F@ha{czOECU
ziJzV^d#WsW&DCnAs%v-VU^}JC+cZ_*cA|U#NmU*by;c9??LK&0`3xF#Jz9A>``ia{
zoq~USwk{rWBd?Y={;$XSXD_x-3z*8!s|#;C!Q12^)N<kNKs;kN%&myl^;_*#S$I2g
zA~{iid)1Er#d#Rrg=?T$yzb$HDSGgY-SEuM7F&p~4sVB$4b_rsnvqxC`T%baz}qj^
z*&i1|w&5~-;3)}c+ytfYIczNgR}0clQ-J!Qm@|dFZO-Ev<LwsXvpo^kK7_qzxAPdE
z;mhIeGkANUCK@*!?=!%q#hqPRmE@Fe!dckbv7l31uR7%mZ~e|vccpH59L_Wx!hLxA
z2F-dL-WDAauQ@Xvnuw-qdEFtOC(P{7bKLF=^Ey5075?f_%5H~>!rM-j94bua!mMvL
zO$s8{=(<gxj@z^ct#vlYss_w~POoCs)m&Cti_$OK#isM<)nZlb)F|243U8ai+d}(n
z>I84Q!&V0@J?>-EF1J-3lC8Q?!KQ09ZR}XHF|T5k_mfpI5xia!_p{iPsR2wL!Lie(
z{>|)Kzt*m@Cj9GD>|TFlmk%E7>f#PHVxLY+e&^eh9cr;UUag<St4>0^8o}F!ec>CP
za``W4^+K@rPnW7wr{27SncO?f3h|!qi{brDrKc$;byu!udcoVw@OH^cW^E7PDW6Hu
z&DZoJ2l5<8^G|RW-WE7Wy%r|NOy~dFp4_cU^s;dsbT&8tw>s2;VS2HlE}g~4zKQ1E
z&hL~iCSHBJ#_Ql&W`FiNv~!F@b(mwG%(YG1^>O+&#IF6kZ}!IICcm+1+cE05TP<pc
zrfq-H!Yryqjn2gC)TUTDXU8gM$yojP9<8HYqP3ac<friVM5btYo<?bCjVMiN_(MB9
z-!<p>cin}zUEyszcv~+^q~0@odyAZl=damy3vZj%`l`bo`jOv#(ZQDN)_clc+qz-;
z92lw|uR>H6-nP9K%v@|RyMu$ZCVj9bSc3F}eyJSnCn#YElK=KVy&fE>-3J2HI9q^%
z7Jkxec$>QBqt+*zbzmFXb|+r&9yB4k^JH&ysMDkz5hhJ~V$!L*CM`c^(qHvqW(n9?
zz@&RcOxo!+Y7D&nc-^RThm1P4&dAP0qbhGUYS9&=#yvDD@4rSxJU6O4TKKcWsFlB<
zU-OyNv^c!2Zc?cR_{g(Ny1vDv;rmSr?8?vNXH4Sua37QU{A1FB7?b`++cu@f+rXbb
ztTbl5;aH0PzL3$Z23hc=Gn=&+?z-|ZpIa0kyO>#%;cV^(W>u)c{(=eqhC1xi8{dQ5
z*8Ya&4ai0A?{ByQcXM$+(Bfwp2XDLI^)>WjZ`vlZ8Vhe>H*I@gL)q%|C6k+bJ(^jX
z`|uR6_~(whs%YB@YtkBu!q%?vb_A?-{*SkpdF%tdQ2pmH%QKqYvcs6gg|~rlGcX64
zPp~y@ML1mp?-`z+EeTWM>)8C%fYGhR;AvUzkFU+lPc>c}y_;N)pM|zvgpXXFb85p|
z8@*5!-rxh@re1r1SzCCkdF-oU#%Bq<y#{Z$Fe7rX6nXV<bxlEbx6#X5iXPTizrtgV
z(FNFR180jF*!{z~N2-%CRgoMicpL47pYV1)y!{Q{4uGvwVC@08nhBqHHnr3)XxZOk
zZCe;yn8zPet8D^fD}7|%<O991COlwx`viU4{}8#2tC+uqw`<{Te*Drh&tXGr=1OYu
zS&uL2TN=i~gudgvilWx|uq^v#?$GxOqjJ7u<_6BTfwy(IU2!{E|7>JNr)IKppQ5J;
zKlx@Wk6vem0r6zq!`rjJc=QN9CR4ZV4l8%q-3o@cXYaGi6E;L{AU_b^4urS1HEylv
z>(QIhXouapNN-O?`e_Hj+avIH=>NytA|*W<4sT1q+ad6_*E5fn&rMcn6L!tP+jj8Q
zCKyDGcLcpx88>=#>!OFb1&{j2B&##m52GVfG>N}&XJ6)i4|~Z_hjad^O30C_jqrBa
zAE_z|YtwKngtyzKrRw*!sj{!8=1Xn%J)V2<Yt&%To>kGaJz(qRN?xUdwHrC|!rOct
zyPCjSc$)|>Z}8skg`FY%%uVoi61=SdZ=LXV?@}+<4qWdfk;Rz6d+wqJ&G|Vwp0uUj
z%lD5ukfQ#JQdF5fwB%O2H_YS|e#risLy0;@e{q`zWDn8zwA(>0!$p|7l-eo0{oDn%
z@>!k>ZMx`J{NVhsxE?*Rd_I48?$Yg-E>++&e_YG|vXBmKqu(_f?ES=ZaADF{mzKhY
z6wd2~H`xeXdjj4jZsh3BXL&@tN}rBb)U<dNt{Ja^mE!eV1@fJ2<lnz_sMs@yR>IiE
zvm7edlbRv3u_^yLRNq2oZ>mGeiw1_bf%F)!rg!)~wb(>FVl%q7YOqavuiBJ(w@q*9
zl}(xiL#v~?VlAqU-V(gMhE7eF-OenCT_>nJ?s#oepSw2Ygq8JYp-<c6iK2_YrCPNT
z-i}VR>S(N0{n;ybdyG|O%vSsWtJ-|E>Y>G|ot*m|9w*{6XF6!p#CJBW``xZFf7w-s
zTJUXn`xf5Lp@!=ZZ#UI}wNdmyuRzy6jaMalmKQPeI|g65z$&Lo!P`5<=|hIMW#R20
zc)Nl3^+^EwlOE+<DK1@rx6{$2ruz5?-RU)hx6}6VzdJ=Q6ufN&XFI&%|9LV&pVr|m
zb1fCjIg8-!iV6wJn4Z}oA9|$TxpZSZ`gx*D`G>f43f`XkA8+%(+urat<P5+27KeJY
zbErDJJ@$yK<c;)W7l_lr2%A=IvPtvMwA5^KZMNu2KZ_=1wP+wc$tMoSvb!l(pQpyk
zF(y_AyT|H$pJ>%98m*p(qv+>}Qt{SN>g*GxhUI>ckM=`#&U}}<>UZ6H5~+Le_N+g*
z@OEj%2>P_b^|Dkrb24951>Vm6@I|Mae^D~~*}t)aVEyh8eap@9ELb}x2di?yU^zmA
zv^YQY+UU<Re)z1GH9o6NxzBn#Hc(xS0gBELp#6_N>94_L7XQM&yA2;TJBb<Ct<*-h
z!>2W7U1??3jjymYz@)ubOzL#rq+jpQGX-xO!`ou8a6h~~n$;xFPlg5>b>W&(K?jX8
ztuboYBBL&iFsk}UqkfF$c9c;C|2FFF7NfS<jXIW&8f{LK+7&XXSviv))HZ2-W0T%3
zHp#xnr0iYLwy<~=JRjTLq&@I9;s^6bAE+U_@U-!rXL``v@OA~f-I9^rd$6`4hZXMj
z%R`TH5&UX+8(j*YxdOe*e~^1I6fYUxR)DuA=6)^~_cuIx=4Yr=fjQhd{)QFoXvvV*
z&u|PMIdVUHO5v?-6MJ0R_!=Ur`5I=z+oSOI$~}6G(X_eX?MH4q!`uCM#`cBeHo@AT
z^RrvfwcFrruc_o&jAZ9*fBKf`9bSlz?e-hJPg$6&rQc~j{%}T^dmWY@hquMxZ6z3c
znsd^_+XbAn4Ypq8wlI9I!m)wp4d~x1Fx!-#d?|Qqd7Z{E3f}(Qv6Fuj`4w}?p@6rY
z;qBU%<kwdsn-LwG7Tzv^x6|NlVR(BC)}G`DB%i1^8aSRQ!K(0fBfPBvZ-4F=7(q|+
zV0gO)-u?=2&r_3qOONrfcRq%fA=F^euq9yb=_qm%U*PkerJo7j{t0h~n3?I}T>m?q
zkG8#Xj=Y79%n#uoyGQYQ)yPe{lB#L<@j~TQ#_H4&VQtk?)C6J2VziZI4w*$f_V2wE
zwewHW+0*PL-;u1r$6(qe=6P-=%d{z3R~oQC^{Ge3DIQ&gx7V}$tfjgY!tEh=n+4wf
zk>u8<6t@b%+d?1QN_QRa7~UR)uNC2Kxkc>G=Jo?ex>as<T}p;8*@F4kQ{O%3*6BNB
z_+Eg`r`@W2%B{OaJzCs|{^p+itZ^PS<Y%95>d};P9^K90(P)0|u52E4hPT7utslJ2
ze2KcU7jHW{S=HgKCxYBV{$4{Ldc%&;FLjq1sgdIozAw7<W9?KGUPH}yADk<es^qC~
zc1@~o!rKQ~Q?&u!`d?+9X9VZ-p0$Rzv*GQ`KiEb40PmNak7IjM^s+K}P<6bDXv$17
z=l#`{{w8?)85Y-v!`Z5M>DS=@;Ll^!U)``aBgb=i8^!s97BI(yXMUHS>~Fk3-78Yd
zPUlu*caj#oCx`J?qGmA9a}nM)gty1wZS^?%XD*Xn$Y-%{3ubK^Ca5~RO`iq^$GS9r
zwM!rO(5J-b{hgaG`N7=teCF3h&o&*+XE>kdky**?pWxEJx5#a{M(^-lxch+fTe;)|
zZx=3wwG*8>3~zINkJptG@j5$!p4ZA`NrXG}2*yS}CU4@QLsc(1$R6PHeFj`AjqjS5
z+1YFkt%!}&o`^X0dci?2aNUY=8V7GbwTRQ@!p!(&jMF-Djcn&^3R;22U1XCR-nN6c
z8{qBU`xXt$Z&lik^v<rcD5^X?wO=hd6NV<UTXa6oqHBlQVUy0Llr%Q>=vtL04E+jU
z&p)?nTNA6^c7wYkt!lZ(s>AG1KNw?GHb0xjl!58+HXFPx^wy@HS?wx^uj~hJFTmTY
z%&VTEN4f%@a(r$6)+KNk-qv|dzcRd?1aGs~XGWKPueE>0tAi_EgG)FSyag{Hqf1A3
z!`!DX=HXoOrH{B)Cj9;aXj6D=?U*2OjWi5Tzx!mo|0j6IX0DOoZGZfT2mF7G%VF*;
z`p@wB!%NaXm5MJNMz8k{7ydTaPW4@SmfxkSq12DLF8b8NiSD9)-VMK-|4R|BefBnV
zXdC@vW_rxt?P8aF1G^fB+0+)j|8fpJ$tA4xiCffVr$yu8?X{>_6<iXlX7F}9yiJ6+
zo8j#%c-yF8v_`|*5Ab$fizsR?Q9673hwkV5p<?j%Ts7*qPb0PMuSgl-ZQ|YtH7y^(
z?u&5kY!a?>aJD?V<2Mc?+u}^9`g9MK?~f3!l|!@^-oE}fSXajfD|1wkj$IGZ%uGSr
z9PwGD;q4fB+X~)pfVTzT1!!El09A&!mB~vi2yYkM_^3HSFm|a~U0|+xm02;(=?Au;
zIpNzvc$*pC&AVk%`M>ap>%mI2X(f2MH?v8DEk-SPV$@l*Y}*4yErzkF1JSudjLO6P
zOYpWTyuHuydYe(V;O%^P`zZ%H72f8Cw@H=pjBA+mq7OY-RhTiVgMSTg;|ijwJDK#$
z5|jF!=DByw9>trK59SVuHff;IMBk}N1x+Sx3OA{X13u$5AIyr!T!7l}?`HfyW|_*E
z^#bM=$iQCjbnN%Z$sX|{>>K5H@zl@I;fkN(?Eycsr~C}N**AKN8K6Jl?bBWC`GL31
z>5~e|?Q0n3&pc2fd++u$i$i_(3&$dOyBJ;ThiCk36xqj}sl%dI3$!KosR#SwC$gtt
zL|Q}JzGPj%+dlBN!cX49$t}OYW*6Qsyi^j|5*~J>`tn?MGATGG4ZQ6JXBWfUf;{F;
zA&bJveznZhg;UAz_mWA^v5r3Ddh|(kzC|r|FFShR?Z~-s6W)gOX73$+$G73F2On95
z_!=$S?|=GsAgmov9#n!qeN|Q;Lq#-bESww^%)AcTbv}K?v1ry6u(icAAH&Dnc)-_v
z3}5fl=fq<r;B9;K>^F|8oImU-^;dX1=Oh^o^dbM8>nVGjyyWvdc8ZLoZRp%g<Xk*U
z)$6^f>b4?PvyOVzkk9+Ug~=$2PthJDo^eo$#-OE=$FR?BCRz9;sU@OsNBgEI{V8T>
zms1<9l`I1~@wAJ+V>tW1ZnEB9^(eyvkMeZ$sB31AjvZicYBRT5SH>5<&dkZBB;CJ5
z7T|O8>y2(jU84_*dTj-Gn{6fiP_xPI9SdUzFb_1ut@5MX>OC4}kD!J-3(pxI_c#P|
zPrBvdxZ#UuU6H(f{M1$>;Se=U8$2orZy&(h0q}M;w|mQYWNG11&Dr?d@HT6rN4r}j
zGcQBl2E5Ju8~$%`YNX}Z(_VvmE6kijZ&PVJ+<b7f_zX1XUgliyz}DCJ>OnB`EA?3X
z=S2^_3WK*V#*?dv22K5wSsr*>v>2M1_wwF(@||{(IT4+rQZV;3yuHY4SHdH%3vYYC
z+f!WY+%Do(zmF-@T$9N&q_>M5J~_C(HJg2TC8;z2<n5JYo$d1<Z~K(u+Qq@H_cuu@
z4R0^sWJk<ovJtzGQ&E@sRu}WD-_Stl+aXiP_HUk`m+*E#X=Y2D)bRLx_l}_-c)Lq|
zub{Cmxm4;AbD;Oh`@8N^HrV=nCwcoDTq-lcrF49q1wPL;#;f>UykT_OxkRU~FvoeA
z`JA-O=Zr2FuL{LsZRU8@B-5!hwb-A$9di)<x)C02a42@ELuKHt3r5Aj+amC`j!&FQ
zTB*rK;nULRG|6L^w_Kbq!rbb0$YzAM$1}jbP^&f_M~e=&YFv9O*}-VkU#z-*$)Yv%
z%6>IiG?qOz-g4A!;qAVc7Og#GQQrgPGHkafejjrzc*)7ftm@jss(0C~D(zvP+@JWz
zXxw;s`|Av=DlM=o=TfT%!{(6U^h<TKX$qcnEqI$JgIyVF+Lc28)wu_D#iV0)h<vN9
z<sI7dl{uQMahkp?UPIr+Ydk&6cj0aPzwxS=6ZXQ}Q!v;CZ_^o_8Zi}*fa{%t@b>Qz
zm#m4*?PN@#7DE41ZT14f+xX6S6!^r8D>9cw@A5{jm!7VK(eyvnTtbi89hdfS4RkKp
zrK}FJqhei34CVg?Z!7Zu4Xx+Wt-LNx4<V;xt5aF&O-sY?^{_|0j&cn&Zkj{8>N+&x
zW}F61r}u1&T@~SNwtF^>EJp9xWPEqLgKytp?NN)Ww8T$-5vym5V-+|rR!Ku+mGxD$
z4$q8Mwrb=z=8e|8ZBbe>CrTfiMCtY~QL1?8ho<)Up$quQ`{3;&@}*imj!;m$2z9;^
zuDdP6H8JcPc^2Pv;rbW$&xgrADoj}pg{n;JQ1U-R)M9Oj%4Q1Dz`Ma}H#k^95kWdP
zBS=RIf0m;MyJSNGWX3-}(Ir4j$t5loPfl^jC-uDXNw#)m6eoOC;&C#fE>rWIWY&<)
zc*!f7Wi4q|(t3CY^X|Sk>cA0`y20DW9ZjlQpI)h4CLPaWA~(yV0IN|o;Ozi-YuRtq
zP8j<Y-j0K})?r3X<Y+Y1s0!R>UutB}kWndk&JPwCRc578S2q0jSbDzy!6T#oNHnTl
zO_Np>r8b<O*N5>_XTZ|4CUtmg(jMx#W>`D^vq>dinDqN|6Z^YNeD2^!+f8bL*E}2E
zuFp*tRYCM`E_Ths-CpqaFZ!X1`}-R{WcD}Y@bfopVUKOafBg*E;q7C1I|1I#3P8{9
z^EEt}iGMtjoh|Hl`&Q4_u!bFN3*l{Q0=3g0uoa*9>8iAbl1s>#fVCg-j57~OYnVfQ
zwG~=*WJ`SC?(7$*Zrg&oZT)2YUbwo_oyHKw{Z||%;$bbvLYVuK8tgWB8vt(u(Y}@8
zYXW|84R~7%-X6>bo8j$9&d<&{72)k1SbLrCS;u2-o{%ARlh@f#?R6`!4R1G1B734c
z``X}bUwC^D-lqP}&xW_v;cX;2#Es#tGc#O9$99Zm_7&a+z}sW+_GeFVeK>lQ`fKC2
zK8ATum^*>DiRjc(@N+BPa9(&@n{xs<9Ne!AhpTNT2Wk^F-9uziz+3MzG%?Jru!xLe
z*kE3drhApFrT3FHg&N@Bg<-+%6iq-^m35GlaF#u8_fu51BH9i%)P)6qWk$C>PEoV8
z<ldi1R;Fd_mw~sl;q5eddwLOlI`FVY3%5oH-3eoRwRbC=!&-}4t1U?_ZZV5;kUV>M
z8~cnL-Vf|g^{`_b|JaC++^RP|axb^?_Myhx&#iOt_9>k0J<_fBL(y_m(8BO`25kO}
z*F5nk_nW$Pc$`}~9B!?lS9uTKan(ZfQ6*CsMnjjdxYaow-Z8vg(9)yJvtaXnvJ68!
z>X4KCM|j)gJ+npUc`wc|tF#B5PA$6Bn`F5^P@DdotQbeK2I2P(JjwfU8}2^gy5tRg
z!SCoR{@_&~yn#;Z>E-6yqzt@GgtvX+ZH}DuK*yx0=0EV<LVY`xJ>vduEnMT#eSU}X
zN6GIX`(x=9k0x`ylIEUA!2_s4$MC+P;Um!SS<cXljpsg`-mqQh_Vjf<`j=~qBlXE)
z9OG6oXOg_HlN9=veE@e7)#W66)Hf!Q+`zu|c<OubcFhqo?gu5v@n?cop?^MRA*UjN
z&*V=oP2}_V!(rHa-lee@$w7KZ9>P6pv&WfD+33>z{nQ(Gvy*(aQw`=gmEw(83VlEg
z@5Sr%?0Dsw7cUF-$!75O)F^VFU`n-2@w$*EUM*Y>z5BvkDt!F|&vGcd4X^Id;i?X$
zEycWSW``EhQ@o;;T|1dQS^eG4XQ*BCBkh_?U-4jijvt*S%jm3C`Fhhk+|;ID-&=L=
zuvJ|KS@jg&mLq%NWr#($=USB1$)Y*%HhTqbGgvgVkwqsj!(n*a8QxymW6^!G^~bEY
zC<5Nj=xo)T->mA1_VuOy`&$dEGLN>Z-E@5ExmN97VpUbx{3b7+YBQVe(=*iWtW6Cw
z+LasL`rfyzAlC`4o5#t=*P$6%_?te)sX`F5Mpy8)TQX0x!meGF;&lIoLv_=~lWB;9
zb)Q;w1*bY!LC?;^W7y@A<q298Ubcm|LGadAoPO|H%-3>l)FCf3%h{>X2D$X<EdBx<
zZM=%_Ta0Ffx39MHzk;{bj4n+<*G|Fn&j-H?uXQPU7`}CV@}%Hxug~lQ+U!(8czXkV
zTay1@se<V3(PUiw;m}2Ty<K?sN0^h^-k;hxK76mzHjSTV)iQYd7~Wcs(3jlDqKbE7
zRSaEQy=JWHb%~M98m%Yr_A0#13vc(r+t#zAlmp%-r(;*a*&o^jZx0>&u3sB|*MWdY
zjrf<kExb)Q87_|@T<L%A;M?|1u4`XZ2;Lfphw1Olp*sF0L|bkJlf53S&O?K>y&ryZ
zHEOm`g4C&SkPcY_>01hvAuNDh5&`n>fu@DG7aTAa-sZUdNfG%zX$`r=zlNAKbUE`h
z<LE<PPsZ?AzD|Z~>#5lmHfc1xZE%ntrPB0D{Z8#R54}=wwkN#3&g~ZXnCqDl&xpG1
z4x^q<HL7D<qk;#~Lj_|;aTJBO^*DBofvNb)GsnU3Sw_8IVbltE>)B;gdwBZ_-ezza
zrK;3yf1{R5oj02MqlVLibq+84jY*~9?LY9gB)lyQYmYoL$p_wUhPMfJJZL|&O2OMH
z@HQ>HJ@cDcN4RZN9`>?FFY|jp!=NXAhN2Js3~T0**Z9cS(B~?77B_tj*Pi(rW<2*b
zOvFd_fwxyi_!`Q=+gSFy_0P^@&a{T-Z(!#R>aOth;S^?j$Z?v4roGXc+1N(R@K9f^
zA56ZFmF%YGuoum19}1u0Vh?yb5#El2wPWCGdOP{W^bGIa#@=u0tv^=LBYc|s%*!4$
z(E}9&2h*UX(ZbE(?LxF|3|e+|TKbvM!zH5r%VRu1?RN1!*b8sp!rKw>_TC)!p~Bnq
zov7K;163E^UMh-*oR=CetnHVR*M_$bVC_S4p)#jUV@LyU_n=*SGuJw24Ee)Dn1vlc
z-|z^s0Oz7*nceX+Gn)&Z=76_jcKH}4>>!(w^B--YCcD+g@MSkW%RE1Y{^Md3(ZJ*x
z1=DZr?N0tt6Y`<*@V(v?jo+HA7WC^T!`s+GDf)0LMep1xIvJOu>!-<t!aL4oPSLN^
z@Ii~EX#VbGby-D!@6lvE=XlybS)Ecms*k=r3=`8fVP@wKx2nRwNA;Mk<+f&Pw+gp)
z>p~f~c793H_?u+u?n_b?W>cOvOVW9Gn-$)EhO?(9!Q4jp$$U@Vo^E~b>t=@DtySIK
z8pg4o$Bs{i!SJ@vCi<*)xitr!d$S14&ByI+`kLQzpFZd()QA`3TL;41)v@fg;rKT%
zj6>gMqj$O)tn-g#2H1&C><MFE!eV$k;yC?Ou=Y4d0y)Dks8dG=ChG~jO_|U8u+FOi
zw_q;3eTK$u3vX}2+wlkJUz*KxQ@l#(;MFI1TjV$9ek>`Pyo~wQEy>!>^~RSQ?8@1g
zq(@}_ubM@^KXW+~$xsaAcI=ZR&40-rfDAl-gBc#Omo{BuPb{-FLGPKb2~UJ+N!k;_
z`Mj4CjY;GUC8>=yNogLC!}vS<cxorA0dur|Hga>nkV9xB_h~0t2zbdE`_oTT2lker
z77K5q<N3Tcxzu44n(B~Chv03N3og}xx256jY<T--3*IKW*0coQ6M}C;mu<c7;M&`v
zsLc*#K+jg8PP?o>-YG085XfAKn+#rfdkEe(g16J)Z8LZqFvX$%9pEhW%`@o0?g3;M
zk@+<6v0cl4_7qpQ>$~iVfMJ*5)pn;{ec<hVc<bC|)x@z@Rqkt*uhpW-<t?h2-y&y0
zi=J?QdtHmd+ghkWT2yaqEIG)rilJ}zH9O))?64?!FS?dvGrT<oZ?~_sXd&m`BOmEJ
z>^0r8Xk1$KGJf<XxbmR`97f~%!`rdcE<3)q$_?umoU|(N26bP2=;$9-8LQH>)x@R?
z73?YuZ}WD?bM9f&6K?xlvXMh-*B>jH{UxKRYR)*k=Q!n|A1otYvt<+4CC}L<gig&1
zKgU0IX$<dsaXf}m@bX}cOZ94!=ZKcI<RLdXGy4bl-&{LM-ZFl~Jb3F^#P!eueEnI}
zeBteOc-uM<o%_Y5PFY=AmeZ-&wDg9D(;L;zsahY&%-Y26qTck7d*a1Fn;Io^9RP33
z<1?&=x9hkzs86p{f$`M96K$%L+orH7T-WfvT!y!AxOSO0)}m+ZD40GWRu8Mis$_pM
zlW#|>=9Flu2Kg3wqV;e`lmcf(DZEvbj-ziMpZlTMfz)u}ZFYEDIw+F-RdO8ngsb%M
zaP@<?4}-obYvwQdV`~^$aA9iqHdJ#~g{s8AA^+EhJZgNfO2gY3eS;NSI9Ok&1nGV1
zXQdAbR2B=ne>Mf^YwG~r%oCtaPd};e*iZ7!!)^jU@*6XL(mAp%@?2&Pw!c}0hg0`l
zZPvyyW^IMHQ43&Ps!@CI!A7`PD}lOeK{RJ>yxX)U#RQ>WpBwe?rcqh0p=mF|N_e{#
z-mYq6WVbO|71oa6$Odb-qIajl+nezA)mU^e{&EOj^V?}gMQ<~r>5VE0Z-+&|>uAo6
zrT@xfRKAMzGLJ+{qi@T;GU@hL>bI|%Pr8rK{EQm&U6ba(;aGTUhqsq~fA&|KlyQ+s
zxu(+_ot;@-cv}qK=6CrS5?}clD%|!nB(vW;WSXC$Ao<7_ulO21-|#h5gtzye`Wlw)
z@HK3h;cGYrZ;N)rLx#8A;BByrSy^~H9Y1*(by+99Pm?+N!rKe*_A0vc_8wS@w(L8e
zynmRN$ZXDK7&-?QE_%s4>_>7A!^wJry|unGBg?$b2+mm#Z%4t~1*bS4-tL6AJ!9!@
zg15`yZGLz=0p4zex6RPD)3`tC2iX#^J;QtUp2FMGm#D+S+oo&jnc`=gCXzGUIgP>9
z5UmSuds4ei%|re%yv>w@{i*QwI=meVZ|kO^4><-O8Q#u=pE;?m+B?u6)dx=*-cE+M
zMd9ryc>57g`1L<97S^`nzJv4c!Q0=s-vi$6ran7y<$t{03va)|TMxXo!Q1BWwivw4
z2XAA}vY#!S`5k6bKQ&8H+Vm;<g*>LrM^dyG4|FxRkMogNd=;IyEm;Rwz#e*hM;%U9
zW;oRnPj)4q@xI~ITpQ#6R)x1Pw{A&#V@r_9`=?u<+K|%=Z~ea}DGJ^u?1dF0lQf52
zGS+(J5p_$_hI{Ns4P(zXy7*lyykhve7zVF_w{yAQg`*s=J?1=Y#*?l9UoxSO-N8vZ
zQov0Xnp=n9ZTDwx@(^IGlQ~{^`w-sNg|%zGx;5bkKJ~w381mRw=6A<*XO8BkM-KX#
z&v1Ns;*s+dx|iPQCez$%J}OBi_#NkuqzCMKvbGFk-vYe7jQ4(S55DhOuS|<!FCKE-
z4r;&fHk|ubc$*X6-poNCR4g;V<oi#9w>jbMp~xha+LWY*FnAF2ui;b3V#50!`JP%_
z2lk~sC9nS)dvjpzmGfjG9!%7p-PB-LCF=PCvJmGcs`xx^x$j`EbwB5;0lNV1Cuuv6
z515^(H2BE*qR2V;A8$**TlX|(df@Gp8VTffx%A)`9_C0Fxs7;_+!mj~EXpRAUL57K
z8qYZ636}z3aM~5@ZkvnGxsrVWHhP{iI(0BAUimK58?=cUC3<$;)OdyVWq&}vc;$TL
z&_bBI=#@h!K9Qq<mpmiPp&anmG2Nm2@OEE8W_|E2hg^))yxwuD$!Gd5JlfgS{^RX+
zcw2#4oA!9hW$)S55?(Fc%)HOvR!!|~)dGt}b75@3d}Im2+auh+2ye5sfdv7vnh+7A
zURPtZ@D~ev0O{+2w*j!$!m$b7#?Q1U4ZJ<R*rKvAoIk*#qVS{6NQ<7Xw#dWo^i37%
z4I0Iq>3plw&$jA$Ppj^?v}*rRH1SQV7Q@>v^e+$C0FN(n9blj%5B+-;&-pIL9FCLp
zT7}@rPUiRDjqjXlvTIOzb|{=+W*7h5eu(VI&+K}4kkwp_>ze5<)xJXiFdjn%n7jKJ
z8kOfJw?)6UL;sE=gX$mhr{HZ-u7kYrb}YOd32%E>g1dOrnT|MBb0@XgrcRyV`e-bD
z$aDCA4L-*{`F;4u)#ycwaj5H8YUiIF+Cu-@j5iLI8p14)DNa-ViPPwo%o9_SzCd61
z8+yWXWwdE9e)8bLR;_}!9pUXw-meU!V^wiMj9Rb{D8v!1+83i?OSCGJ;pomuZgPw7
zT1sD3qtHnG@-9-jCPwH-u?Wr08KH0FNbP(du9CrDHMVWImXMJg)|qVNT4Cy$7^=L<
zp&E3Knr+ii6*Y&*`7IbfI#|zP?I2hiUOiYDj|Zt$%^+=sx8WZH^({6)AGZXkSxa;+
zyd8V-ljimNq*=K?>9U7iwCC8pSBU=PQ!w;B>>5m-(+cKWN0aN1C;MhBkD*<C>%-Jy
zXw}9>weD(ErR7GM(=ZD}kJM0jI|<${TLL#H8#Q>oQH5b?t=8}rPdNzA2Eo}KgN^#e
z@e#g84lrsT$60t>fUh6G+rDdzYQyVBKI7-X+XGS5gky{fb@O}$YQ`hb+Gp{UU!Y^*
z?dR9jdg1N9$Mj9@q=pM$Tk`xb$C=qZ%=ueP@`JYvrqbuNls>DLd~eyGHD5o&Wcs0+
zMEDx)%=(1C@-=+8?`s%*2_1{Z{c^$Au<I!Ewe%#<oeqBo`5OG`O}<-)OeuK#vvxb^
z0U8!vyPRHQry;H38!Y@Bz}yWiO`<p0xP(2aW6_|aVcu-K-yP%|!rO!!c(|Oi&y0Wk
zjan<b&G|TuA?^@8!RS!`S$N4S>1WzW&eL9)$>ST2!^A6T3|*<q+F!v+zIQU3cotfD
z1-Cij?Vs>An7P|~ckz|abKX8~XHwhc=Pw+<_jW<^!rN8w_BVJtCl`BrI6S}c`tUXl
ztlh|r?OP8S$JA=eU8Ls;em;e<Tie6a>df_UKP$W)yPR4syluPJ$MA~2r_!s*PUh=C
z&QC(m&V{qV9P2rn!`sM$oQL-E!|VLPY{_aejkA?!KU-?DZc`I{f-hWR6dZk@teRy~
zR3s&tES(f}qt-YW-hMfdBEMYJ9ARvgp?Jr`_&#>Ijl*-@Svgr>Ec6^TBj0JKTUpVx
z%PPB7ygGeM#qmu0C8^T{=03|MYAIj;gBJ4{f)49LUk|+PUY~t$@OA~fHNx9M@YdD9
zt=m1|Zzp!kG)$7Sags*2gUzFpH12qky29J}@b(#esS0neqw_NQFrS;3OpMxYeY}S@
zwz@SHZ@Hk0zUc30<1lpgCvvIiv+4v(ADo7}|DeII;ZNh!eqTYq(O$GQyuIF%^S-k~
zE+9!S@+YcAmP8H9hrYzK?K_w2hXjv8>8C0QZ++>XdIN7C3}KHE@54&6uCD$@##J=G
z2O4xP`g6iTv?y$SRTa%zpSjbf@oN0trEZIoG<h2DMf*hh(Ab3oZ!^H#G;NuwM9&<f
zMi+q&Ju)&uTYShdo<lu$J~_kiHWsFeuV>9l)K+c_K7rRt(inIfOdpO9wZK6K@rq$)
z>qzRb=-5D?1l29a9OfU?bHiQIPyGsKCv>LQcreU`w`G>Nv}h*`<}?4uW|xkw=6QJg
z7Tykpw}-9qx}1j3@j3C#4>Gs2j(Xn+{Nu6lYTGMbP5tS0ec;f8yAHMO9H%q%6aUR#
z8OKGt;*P>nW>!5L?V37-U8*mc?U`WH+7jeW474c*-?9z7jX>XCg14qX@|!-wTr}m{
z>GZ??h}EIjv1Dk(su7v^k@sSC>sO0D(H}deTdXcOi<N)9SiL_Mi+>)YunRHzXLgJl
zc8_5eJQlCmqWv(!J<+0X@b=jXiypw+tnfB<T9mR;XWTZ_qAd+A3W~9)L^i8h(Rbyb
zPw4VYxC?LFHnpl$PWq`nSk!5iMXiTW8*XjUhdUPfm@Q<ck(m+7j7+>mbvgg&J9>=j
z+cg{B-rZ+c6uezPJ$fI#U|AE$Uxv5m;jIa#E`qoHo09u<nf~M|^kdg$cDRs3hVAkC
zmmcRDFXP!&!@dDnyZ(qvzchri^<3mPGneH<ZpH^EGgahQ9Ck7f&s<g$r{30McOibn
zU+C!7To-kRw@J(|jlDsic2AoQG`DFv^G^?)R%JM2Rb-$=fnO~89!2jN{(6=)Rwd@J
zDmU4Ur5agPp@3C`=sjzEkZYE|E!s$L)Ythjn#n$(<#FUTj{Kpqqrb})5~-tyBbD|j
z`BCh&of;jkIrJS@ITfxV`@%J2SGa~}W}p0{aCMs)u55M5kV^U@-!WhGSDi4G`WC8y
z{p{LX9;z9JP?h=+qUT}3@;ex;o*9Cfg9=jEaO$?JKQkBqSqn{p+V?O}LGbqB#sG!2
z2+%NiTXMxG?Hd14v0tfeE;s34G^)^%MUK#m^bRheaZ@Urm97vO6+X<^z}qbQjQTwX
zTD722`<ohd7v5Tq85Q;~wO4qnm1t6bqZ$u4DgfPTLBEcIv*((@*UWG>7ro0Bjrz3}
z-wR`7yW<~sG^#~UqjLABZp-alzPBCUcZ=`e%<(V!cpuDt{sY!J;5quX*=Tx};b^~q
zneQEC()$4>?K*BETO6Nz2keKrf52OR9y1*>spbm&=-FuW$tErCNRO4^Z&`e3vKsTi
z+jfEMAV9-Ty-tqBJ!Wmmk9y6Hy&N#O<Pl#(F?yterurII4&eEA?3QQtwp}h?JYe)K
zvp}=p?Oi<LLh$w>tgX$=>$P#zLz~0f>h$|oBtN1XI#%q+sX}jXg*1jeWtnp=kAM6J
zJHmO+=|uZJLZ`yjvFOk<Fmm;Byxom>%J4Q9kGDMrPcP#CUQ1)}!dnx(4T;4^hPTb&
z?KbprKX|+IJ@s37yB*&C*aK5}tugSnEWE7<Z~ua~c7FE9{Pa-4*~;1BFs%J0D?ckE
z-gi3ANnlndh^z{DJDl3<AQ(Hd8F`D<*gaDm*5+mACmXXrCCRX8%Gb>D^n<C#;O#9~
zJ8>pEYbV0m;Xa01AIMnDl&Tl$Qgx6y(*8a1JnJ!|ax_Jwil%4^ygfWKnLT33x?Ycd
zqN~(P{plr2o2<o#6zXE^vMT~_yRnzN9N7`Dbtkj6uEEqcso7?&ovgb&u2}MmFL=oA
zM#EO3XE+UYRXow%hm-W90IZzIp0FZ`YD->GSG>-Zec;4Gv|DJRa;C>qu9c*Lc*_&X
zLAsd0-k7%Z{&a`E@OIJ#a_x^L>N54n*b|9Le=bpvAF)>(MlXlSS_{w7xwX7W+7EB1
z)WOe&w-wQ{Wzn!J;?c>}ZZihM;t%Lmcx$HS+khhressraZaGK06+F|eYr8q0`|J9<
z$xBYs#*FNO^G#IWU&u!ClJk?q4C*NQR@TuU`hpqO7wB<b^Z(Iwmf>w3O4qgHFf%sX
zFg8ry3NtfPLk&6%4Ks5Z=7yP>8cxHK9W$fAAT#P)xqZJM=aDVRI>))aUT4qj*<;V7
z$4X!M58m59h&~&$8l~Q_lN{gZw|&VvvYDKq?&uc0DTi9)1%*+YH>2-XCy=k2$P57)
z4}Hk`>z1f>KN95nI6=9XrD=COfjkO)q{Wi7zXyJoHA(shhSvQXE&B8CEjl!n?@wPr
ztL9AMUOh!Auy`Of!cMTTAg>KDCdnR-7D^<iK8yFl@OAKZP7oUQIrT_*dwu|~jYb=B
zj@Bc{Ph6R#M0$Sv_#0kt#(%j^MOu-){|h^L;H`y>!#aj|O&^5EiLBiK`mmYR;<c(-
zJQ@CIT6O?Td>Sv$0qiROKi)dv?HhRe_&}Uiu0`X)+g{AD&V;w;nYl5Ii`6eMz_^p%
zC_U4LXxJr&7?rLNLw_n-!}~{T9lUjaYf~@gJ>5=HD?Vsb5xhFx=pPmuY||ciyP}m%
zyV}58_OQm?w8{<Mz8zxKZC5Mvh*t8%_`a1@7g~^~U)8Ext;y0qYSyxGX720E>;*6@
zxUx-4{Gv5BDO&fd<C#L+E}^b$#7EWHHAX@3XBE5+gV#CWZJ*B38amOY7I?K@vX}09
z06H1oX5hnmQYl&$!faY5`pML)tGH8h-p@SiW%PC^Ge=}=^riOvk(#sF%dY--v%f5)
zmP`M18Flu;huLw&EbKkFTVOb~UjF;(dz6`7v~9J5)F0q$*qlTy<2e19Os==++M9_w
zwJuTD+S6<2f48&nc3FW0tyo8&mJG^a%s!3UX2W-7)0pZu#T@6}<RraUb`7i^8Lc1R
zqm@K1@*#NJ0Nw`C)2@vlZ7sidFMMS`{G-*L-~IkW_5g98(=`*`7Po558;hQ;jMCG|
zkt&Lwy?8Q0V_QZjm0h+!Mu+J`&M<kB)7U*cRNm`C$u=ezc}=i#?g`SEi9zgf3?dsb
zNZHu4m*ZieZX65Let5fHff}(sKxwi58kOB&30ch6UhyM$)lWsX`f7};FZ&jJWbFZO
zbCV@C*ITX2daLFhFAeVQDFu=>T=|D`tnkonC+2)Mz{blz^{9uQbt6N9xteVQJml%^
zt}z$Uvb<i`1zr2iq@ggk7e1$a=iRmbrn?gB)0e$%k~6$*(Z*fBy1J`sTX!YMT^TK@
zJ?BH?!dulvA9tg6+|XS|;ci@CcP)Xl&qlkeIFHNVZ6SF3leKF&;^(=`&Uvmpgw3(;
z%2Cxrr(u5(nziLq*!stRyxls|Lxs_=VLLqZX`hD*ZS+vgS~4^Kgs)3Hbd6eakJ)&<
z`cQLj#JxjJG%&pNDV5!j<V}xuKm5dpd~36dp~g}d!&`Xke%Zy)k>~t(vJVK}?ixmI
zxtoh&Yf~44Yi<|AyYJ*t+`<C|Z!b*XV2-w+4}M|TxF6pB0dI5Ggo_2}tHRg=t>9z>
zvYq(4KD>Pj^V&Cp$L;VFcR;UlwC_uOcPg1sFmeFA&D_l_?0$F(HwT}C$+ytW@OCNw
z<G0k9+fa+%6YOkojbRU70Ci}dpZ?a_Q0f^wfNsFfBh2DXK_ibs&+^%`s3%V|pkuj)
zS@8BWto?;!5WI~m=xq3r8=dSz-8s?8Fcsd8p)cDP-fo4PIoj}87mW*V6AzM~zt^Ep
zg~((?x84~*@0OZ#pH0lp(odXrldOdc%#UutPdqkD6`3^|LTxjk7c)J~)h-vF*yYTb
z;x8VZl%aJ03@v_;u8ODVouc^y;jI__*rP3}Ef>gSJ|RQ1htkWem7yZ=c5wmr%DQG~
zI=t;zo}3H%rrXeg7w(gF|1MQk>!hg#ybXZ2PtK;QV@<eto}8jN^bXO9)mqYj9F10l
zx8nj*G&2`7Eb!Kde(ecwGm}#EI5<V+PNt|gGo-HPQnU`<{=-r61Upg5zV85UO)XP(
z7$*PZ?Gkiu#mrQ7s+^{~^pGc?{n;l@(<U_T(OA6A;mi^7+>5Qu^>!hPxh(!>mo&Mg
zq@w59LpK2}3UBisN>lVx@`cAz^Y)<s8I!Dv#p!iM;!O!7D*&&`jam3Y;pY;b51^km
z-jf*(-nZF`SCq%!;BBi<%&;C!)|dUv&umTBUFI~dp-ZN|Owh>$Gz&R|bKtGpHf9~i
zGZ%x->6}LHU^e!+cEC&Yo!)3g@|IeXy+mJh!Q}*I@z9eu6SVR{f;N-WpB2Q;H{S$p
zLG!%Ym>{n|;&qgq!)kbdtql`Y0p7M7n4rg_5>)JHf)>N?g?AEE=N;MpqZ7&9C%<WC
zqBicq;{<QLN8*izw-LjsS<<WB1Z(^DOjIlWW|wQ5sPXm$J&MN1%Dm=-<uJiDUWa;-
z?Jx?zF}ytlZzsXqA^qd^_X=iV9<wJ0-Zn4K>@0mit07JW@eVEDN<I^e-Ow;rjdIZQ
z{2ZgkZ(%A~P780KCpX2Y9K0PrB1TcL?MHd$boje(`^mqG(c1Eo`4f0M1>Tl9fM#6-
zGv?c<f7+-K*vQhhDVL8`$6s5u=aN-5ms+(EACFT>t4_m&^b%GjlBI9oYEj>H7UfuM
zR?0xLYQWnf<<0tb)~t^1X8CuqsKgjPBWDb|RN1FK32g^&YjRDV=-XG+RL@hR4){Hq
zEQo0JgtxunZOHX#J*T!htfiHHnpH~{*)%cKrusGU7g4j$Li2h&VSj9-O~Zp@bR6E6
zugxqkex#(zv5Mf|VN4qJXll-D;O+i{cJ+Hejx+W5!|mvOcZwr3ge+scWlLMe=|U6g
z>|}a;X3p4?W9U$FD)E__T=B8BOVoqr^lizfC|ZMhH<~_jPPEEJ!`Lb|rStDK5Z<~R
zv+A!K%<CMoYW7t6-nYq&g171PwX38=%NOPjhO=|v?d9C~xY6=QsNEMKPpTI8IRmn+
za?iAA>|17S3q<MM*htku-{yh0xm=m`DH^UZbwc&xOo(2r3XxY<u%^ssXS_8?M;`}i
z$EqOy{eqOyBuLY<2kDYAP>UV}s{4^Z1#Ae^PI&9RHh?}QJzGCCZLptCU1k>nynVOU
zSHqe8F(2_^uE<AI4Bm2X;jNFQy;bahm$vlq)DOQOnnN%1`2_SJwc^Vw+%*i|+S@Y=
zJCE6(LCoQ}z%OrieZA$bvdz(%KTI0?)udlCO_~KGpM5b=w>9bF6_Xy{Ffng#QYCm>
z8Q!)u@H)Ib4{uj~GU?hwlj7c+)W~ikL(E-&qkG2=aM#m;=-N^6br{|$csqAG8XDG4
z<}kwB%iF0@-=*dp;jWpLJ@hL&H3qhRT8hqvw@=2AO*Nc*1fE|BZ%@P9yzn*}-g?4X
zA6UDXBh(E}cVKQ9jeNBRES`<;iP_q)SuTeB@bvvD7lSss7z(>EUrTScN<$Yz_FgWA
z0^3~-M|tmvAufgm?OY7=$hv3%Z?`?oX6Qigw%|NwV(Hf|#Xr2BTJMAt)RSRbd6?Q1
z&6tuOU5Uo^q=x&6$H}lR4Gy1zw;g|DuN}N?0&D#^?sg`}7(RYk3{R;=-+{Ls;B66j
z>k4nz-lVq7Ir_Y#J`Hc@g~D}sdnnr3Pyya<hqr>aryr49ao*WrJnd|Voy?|9ID4lL
zwdSU1?rP44MrEiO!&qOgZxHq1Y}9=Bp-sP`QBBl|_iZGXcp1Cjmrx^K$m1Y-vaRtM
z*CeZ=7#WPrd@h~lU|$n@72Y;>a;ObCNDV)cr||Ya-g;cbA4`pMeM54Jm~AQjEmJ4&
zX6n=lc8<@^#Iun}zk@j*cpDCHgYm|8o1La78_AVO!Q14Ou5V2;^%+m@<n!bO(xY|c
z$dGA2o?cI8SYhoHbluOo>max_2i~r^mnx69sX9P^vaJK3?A27w#P5@}IfX1>ys*RY
z#Nv-V7m`d?4|%#9QdIMw6g-Qm@@hmKl)CZTVyS8jZ#`h^ugA$lg13&dcx++ra2_ww
z!}PntOewt0OBV1sey$0;Ed*~H!Q1ilibJNr+%x~>Fz^3@R=t^~QWwagf~9k}kPB(R
zPn9E08zNIx*fUjQ;O&`tc(9hI@%I~l^UYMXb0fdM0_SN?7Lx-`22+zgOt!<{)ab@0
zvdb%xzaxq2+9^>@=aJ7hEKzR;CaM&Cx&e=Jp<}P4klmA-pt&&00{i-sL--Y~(&I03
zdoJ4*vdyk`<Ow7_v1^pou6c=ed8feKUGe%OFFqr9+p1ZD_BBsXpQh}aYs?I1*LY>2
zb!<gp963O1D)RTY3Ld03?4p}OZUhX^W1^qhfF3I6JTsUc>-+?jIFg{r@OC4-?LhwU
zSr~20;PV^_S~EXUP5LM4K8FDfK4@4HwOw-Tvl7*d-{%UNyK+>5K7M4j$DiLPB0=w&
z$#J7LT9@AH4Dxkzwzq2%Y%M+BuFBo*3S?GRec7RdUw0IF2_c`!cB&JnA$W((8<;DZ
zOs*4pHgA5uhqnV?qMy#vlchI$rwqA_)G90f%rII;e{fN>io2j2ZRkgLn+iO(DdUPw
zUr&%@^p{Nmt6;_~=04%=>Pj~KQ_Q9j-c}8Qw@ojS<vY=;>%Fb22WNZcv&w6sMOWM5
zp{;Ats!e7c8fn(pK4v`}Yi6gCS;@?cc7Pks{Vm$$X3>~Hd_TFYs*(pCH;CEXRx!E;
zcQ#WqcH0oG!0ypH);L=0`&%`<ua$Xat7g`>YI+kZ^TO!pcNRT5W>FjTbg8XYg`ctN
z#1pQiq)qo8Sk>qqS;OC~8XQ4AHrlF*4%pR=eRpkR^%+jrrf$B1<Log!z%7_-{mZIv
z!>ww~XYJWek9rUH6g)Sa!*{<`u9vM^R?4Qjce(fYEn2nF)xWpl-w&O>gn!HHRc+c`
zz@|&^))U?~gtvvcFQ@@;&%@h37x;{`R=q!LmFIk`L@oa7ahs05w`m*KGlp7osZ!DU
zj%JVl%Dqj0o38(BQOm8&`jn5<K68Y|_(o{+*$D057opH85i&Jpf4oBV<a~&3!rOw`
zL)3v?wlQ%*8u}<mu4{vIa!in7;qCaG@YfWm=C=d&@9aPwY#ykO7XxJfGeC_K{590q
zPi=bnsqQg9#pd-@#u8r{&#+_fpbxw2y_9g+Q>jBd)u*MWa_{ocNLZTpH+r0PsJH&k
z*SFD*v)z@9jy=?teFJ0gAoVB1A|JI{IBP10E-gkb)J2m<H-w#?O#0f#r1kqv61?38
zZ?D4J7oH{+nq$&uCzB3kH|b4rlS0RvboYZv`yQCI7T$i1F{wp0cYUnlt^rlazNigH
zyHGo(zB~usPJy@iVeKE(nqPK=zbkqDio0f?WPbN5`unB3ZgA~6VQ9iYG;15QZD$W=
zj=68(KEaE}Su>dpn&zQZlX>41W`_73|Kh%4U<Z1)b=mO;Z;QgsD<fSDpE+L7b1_Uk
zf^J>sVra*Xy<{7EdaAk@hOq-^>IxUbSD1Wmu#3Ty-t9ZGF6P49Nzd4kcMRS#Q#+$A
zITg(5L|lhy+sJC_>TI}JgG@zOJ2!<nRyg|@-fl0Ceyohf<+TsB(V7j>yfF7?K5<!e
z=RYvFIea_~Z~ua~H{k8FztF?*w(&W%E4=LhZwtZOZ18pujP2ovw>bnKls}sGJsOu<
z@>s5;KHSf-1@F{&XG0jgZ3S;FoMTNzaxHRE^De-7IC38&r(!p=J=>iOukb*9hPOxH
z?P+-10p6~Jw*&q^Z)>o>y(l{fW>Fg+MJ~l~cFdM!t~Q%Px89Qzh+p>AjV!&`&8{}G
z6bj%mE{Uf&<p6V<dDsohtm$ohK#s-C!cJqB<Xf6{P+P6DEKMqxrU`K8{`xe%!(-(4
zikVq>>q0Mg47^P~ouQr7jt7tv(e^k#WU>Nhc)&E~drG5gXVO!hgx@w5Rvp79n;e&_
zU9fCyi&PanP9Dw%a_-xaRo{plSmyQXl}^!CdX9CUrfApK6t#oP!PJV^Ux2aHjnBZ_
zx5rY{1>R<yVV0C5i(~pBcDupemoKRA!durmsS2P697^9e54_!tKkLvqJjy52^vn;h
z)Mxs$@YV(19yyrC-W__&ZODhrP8}KEW_h5&@kv*phkVscKL0i9)SEabweMr&*}c+<
zoj33{{Bx3SA7aLlzF0eW+YjCrYeCPg8Tt<m@jKr?fVairQ$Kinm|o_WEM`V8u@i0x
z`_AXEM|7fHet1w`!lHIO_jIaVxyh(Kd(y6h@b))&dkNkS!QcAUIbK`S*%^oK{RchV
z&JV4_v63S#&@Oj)`=wO8KB0MnnO$vCH(u{1#p~n7c-<&QwnV)I<%YM%dXqQ74D4k5
zN_+4tE#>jw80x-whf97-(AKyFZRk%88_wS8m!wx+*+V`AjXO3;C*f_q#rRiuqleeq
zl?%UWGMxQFU$y74IJv^xn(+1#ynV@RY_&Gz8UJS2pelBa<#)^nZ#Tf(=H%|yBA;SO
zV5~~R+b(NkWpRsDfex{n9)$Pk0liRWa(ZV(YYMt*$Hiz_;OzuvWDC<DeeV{n8cnHh
zR%PDP)227~ZOVkVo8j#@c>5CGww`IzFnGJ+7n{7&=+%COx3H)Ub>xw~tUAvT3Nwe?
zw5Z~Ii_XK_ChUjvyJ^;Qc$>YKS#>Is#Zc6&vgOR$G{dY~@U}-+iz4l2eH>@e$%z(i
zjxwuTEj&T~7JV9TRdkd^CyrY*W1B^@dt0@r9Xv<ZHmYycf<{)2D{obUmt^)IwP<2f
zt0LA|*^_G3UU+K>v8v-avtCidKHSEv)|2t>PUl?sdCyVTElnEQdH7cLo>g@ln3Z0X
zod4VrI-DG?{l_9y+&w~1gCn$|WTcLkjZ{fHxe-4i6z>tCIIjpjc@v?8u@O49KT@{V
z^qFUy^<lnQHw&BDKV#PX5@waFYF7WcX2qX3tI7qlcK>SDqL@f^4UN?LphyL<Q^5TQ
znTzye^X+Aq51!)s^tO+bjMhhZJ9)EBC+b<1HpHTcIb=sI4^s*~*&QLF+J8J$U0`hS
zZz0<BK15x8LS+9tL_dm!s46|$BDK)B?6b}OFi3;f2WbTQ_K&7Ps*n|^9^Qd^PrgN*
zTcA2L!<P(i-{lEVwOYQiru(Q|Pe112d~|lFw>*}4$$gi%+;e;DP|qKlo6gM4LF%aJ
zzU_<f8pGQr4arf=gm3glSHju*Bbc2%jmG5p3h2c_J<)=BsWrn}-(Q%ig}3egHYu&X
zNt<h#H2xQpUKTc~WvWr7Yod35GilFglZux$=~%LnzOPBMOPSPUj7eionY8S#NzLKy
z)F{|%GpR@M|9E?>mb+?GU#?w)nssS-pPxE&E_eBGye{dkMnllM=-esm-8J$8b?Vo6
zv^?=cXSgf277T6eq3iA7=_E3nVeK}K@$fc%s)u&)TuqKnX!dSJJTyNKdCXOq6{_x`
z#%0K6EaqYuUf#t}cMf?`JzNaeo1=9z*bNZKo`K)+PBCj6zs$vOYP^e~gu%sNY2jk{
zo##jLvokE&3}w#Z6`sy6z81{?^kt6@OuOMm@3aeg6djo(2f4-6X0P01Z|fCyqcg+W
zz7kBWP8}KEZiK-B@U}|}xCaaOw1=0IU?RM|unKLwlO8L)-E$P}dzM-=9PRiJKk-|9
zRqp6k>es`3*&%?&eGPY;!P}Zo+2eB&uQ9#f2Ak;{!`p`Nc5*NH?~0Z#PcInW#^y!e
z=3(b8bFuw*;faE`S#WmkI`r;LcK-a%jO<wEd`9zpe<#CJ^lc+pyYL&k;o$A0LF|C*
z@6d%AWatMoZ!!m8EH&eZ3Rx;YnHkxV@ON;Arm~}_ww=AP7BuPIbXm@`LuY5YUcuXU
z@HRWVEw&_0S=3k0<%3VT(W-ci{wl}*o(k!zSCKiNn(Xy~v!iyWYsnh&n6{*=V_Ez{
zKd8gP+dFqs_3S)8pzW!;Se|-vu@qIu6V>HXs#>mQ_Y9oe)rMWyXvMt^;cgFdd%u%0
zw=hL5PvgP8o1)h6wio*KEt>Wsk4NC`Q(iBN7pVGy6pcCvYxx{^9&i4vF~jM-QS3vn
z%`DLbczYLpoQB^D-e!llB@d&$Fa5l>G+pI>Vkdp&T7AgdK-(^=ou*@G@&I^y5--zE
zczd^5s<x+~mp_o}-<G;{Cie#Lc3I<O?Mi31@j3iim8959XxQO+MGGftT0>aWm@HuW
zWBofOY5_WMa94J}plzSxZTT188YZ)|hCIS6&F$LW$gVBq{|3OID?C>Y-g?ZmtNkeE
zLwNmI6FZrwFm((XafY4zOf=+d_P5UG{qXK5YjbkE;wTi#^YHk4ZgedgH<h`XirvZd
zZw-4}p?90ct15GpRfnN}nRWeeiTtKV@hS&%8^GI2@YagPO+|Zu<UV12H$1R?lk}n^
z+=aLC)RueXPEr#ecFWe}cNxk)b#k8iGh;gI6T9r_!7jr0RA+mf#s<Za9|mJv+7*w-
z=X7QE!zRQj8YcHyLeI8(ocyVe7RTc{@JuZGWZ`c1Soz@3IrfgZp2IQ9P7S#&nMIMe
zn9;cqtz}2~d)_x%XBtJzP%m0Xe~H#y{>F!+ac80R(&6pUqx`)m(`dtLn~K5PjE-nr
zv|AaoRh{7N>djV#z`!mYt*YMDsx3LJdidC)>}Iph-bKg4)ZfP7ks82`xK3tOgSSr$
zQBy5#R)ajSXm6CtqIF-tfxk=0+23N;8hCp)!mPCQX3ZL6R?q%s87i1npBn18*%pn0
zx3A#sv6?W*&!QS3XlPjbvc5&DgJCK;OQWZmwHhsbpZS>C2D9G5+ASP);q7C1dllZs
z-Gu88IPc3S#W#u4w?<KlfVYElMQBHSxMsuK8RSG<WtQdyTGHE-nVJU?8g?N<t~Vpp
z>sf@XuOk!$Z~Kl8S0(>QwTHL;ud&PhM7YY04Oe{5a82_MQ`t{ps!yHQaW`B=VE;^b
zYYK@}UwFHao$xmfMJi!hq%Kd5)B*O@+Nc$mN{G~tDp6WNHltI^DBa`Vt^=R{v{0DN
zCWXrFMySklL-k9AP|f`(MC(t4sKU??<u?Ya&52-jnH;RYD+FsAygmDX+1fQha{ePo
zx8d!MOf;=mptfFNzb(9N)|kh7{`y?mS3Ao3%FFDd;f;J0SJg+}Exk2<iI;X?^3>e{
zp3D&b(7sIOSC8Q@8bw|N`t?RRxa)9Nb(qyRlB}n{U|Bn6c;KxQJZjXJ_wah(G%}*#
ztxlTM3Y~i5s7VPhwEYgFn$9$;)>@<PY%|hNHERDcqkKOa<@3X+3s$4zVeZM2CQX6C
zTll%&i%t4A)uiixm=uf#PHJva*?}gVt4@EmK666wRwc-rqSref)-L5(0dH6Ic2|lU
zbG;nCi|O-jz_0A<t|YE~No{m$OX|Z04?P~|p>l3$UXC&FHU-}1Lf5vWXPcAX_jFMY
zwaew9ixprmyj_FmsZBl?L%t%^ibuN`ilb#4S706pA5`FHG8enM7y|l|<%lo21iU@w
zO6CQ;O{LDf0p3oecRTV38OGz`b93^es^W7x$*gH4`AQjhd*RtP7iU9lI~l|9_Wfn*
zwDIt(7<J`esT)&o-brnF63k8K>m_L5s%@F)838w^!sC_rnBi>!czXuk<~jq*Z=jvw
zZIM@GEq;UNUN9Nv7U5{kY*9%#e2MpX@G}KhGY>Qaf7WRD3vWkq50IVnUFZBuv*CY+
zwbMD)!rNZ(HVED(!P^!~@g(Da8U=3y$1v+NidmpO_^06QMm$n$uCsr3l0ze5?SkSC
z-8Hj|?FzeQm!fS)lTFk$OWn({->n37Wq6y?o?K(*OP`}xM;~F%bQANY3)8iKV7kip
zOxJ06yJs;qR(M+m)^_-ws%78dRgrX6g|~_1FE)a==Dz95#?NHTN>?R#n@}cQbKz~@
z-_rC2-kyNBURz)mye$ZCD}|Bwx{AFq%<5Znl3|pWJz-_YG^&@ZMjgmE!mm>V|E)Wj
zIduo($z>k2crx|O+~oK&)8Ditd|kjkH*{vb-|*YENB0g((eE($XH7ZJW9CxfZP^;B
z%7y<qC$;RB@HTgKW`0JbX)HRn82#E_@OJb{@~Xy@6VWS8-(1tQ79Ia810NQ={ll25
zn)H}|gSQtQ)UlVt+)jKp-l(aM$$_7N9%pu@)AA&>LHpf5K;0NEyD@*FzBIrKTAy9w
z@Yb_KqE@)#d1{%c24BbzKu6{}LM;^?TNt(;gQv^j?NT1oWmg8Z*BxlWGL5M%qX+Lc
z<u!QQ8{W=CC*JMMKG%VERe`q$;O!}Rd&Ub^`=Ev4?P{`%R!76PeDrML?Kr$NcIH_}
z!`o{u;`IaGrjsX83*PpIw~x=ntMJ`;`31yl1^Tz)Y-T^7CFqm|@77@I;_&uhpCpay
zz<wGy>*h(G{-#7#@Fnw@_Z`NIe3fflU?;!uBYWt!#_8<tI8}kS!{O}-c)M;zoQlnj
zQ`7n6_#b9>&J=o-v9ZcV_U^RzvFZeC?~!%vK~_X3`nKEb%$B?+yJ>%nM&Nnv{sR9o
zyd4a0-yX)>OP#i&AzB^t&?EhhPl{ZH9r%ui!rRK!Ry*L88nefy!|-<83>zM1>YMPk
z7`$D>zVmjQtf~fQw^PTR(!r{I85WI0Yc?wh1FxYM;cZQLo7vZlCV;1Kc5r5tj<7GT
z-Mc7FgtvpUA~oudNX^Ip<J=-r!8PGesVJ>4OXhx27zuBGgSRg5_GAf*^5wMX!&r;9
zyt3%R4~uRFlOKG<qIq@5S%Ta7_nTD!rmly#=gAYEDYGWR+I6$THS%`2+J6bx^ukff
zDj20T-J>+6Zj^Gt%AqycU&r%%bCT;1Z~NQWSqC4ltPEGPCE=>Jn4ew4Bm3H&*W*o^
z7S0};aFrj&^P_k#SrXgxhU;}enEqlH!0kW7)OS~?68;U=k=G%rtPu6w8>~;G$WMF_
zs_n4yY>!ag+z_I@BhjRp!P?dleL4@H6TFRQ-@v1iQMv<fk8+*IsKc*vGwUS09Sd)F
zUJ2F1S+KTzs5U(d(R4oNX8RBs**h@yc(5jn3D&p{!P-(fSVc`i3VRl$JM7tO=?GMr
zus}^^|IgqHfm$;wP}_eC)VjL<dXT}a&nX`{+WM$xO&`sv?4y61cq_1vm%3l})c4+=
zT8sDTNd{Sn)L;FDZx{U<S{P4JBAS&tt>;E+w*&D)p)()D)@K#Tj>y4#@F)#Bfj{`G
zNlx(g<v4n@+fAB~$D~)=jdGi2)S$UW-Cl20wmn7-gSRfXjVklfs2*mc#=+b2c}+TA
z&qQw)J-f)HuG39A*w>`C)l9nC#H7;jb_KjG25%=*TXxNf2RWUZb*f3FGEHiY2F5s`
z-t>2?&T!YkKj1ISow}SI2wYpuYIvzz;DtiB25^7S0}emp@QZd=`#5*{-qgy`@Qb;p
zh`9m7-|@(EZz?mJ>&V8wb^5V|*+YQu_)BN_iq_ptX4Jud$t;Gq$KmbM;rOS9@)}zA
zW@Ww)A=lzPImI{0Hr|)b@Olv$7I>xJS8+Dvg}2k;?Pqvf5tg-chNCe~hVOUTK?iTw
z!`l>UxU1ptyfW+$s6l20ecUzhb}qaPZR2cMINaG#4c^Y2&D_pf`oS=F3%oV_O^x{?
zjJ?7R_<!K*7iz`SlrO^C;@<eLO!Q{q?deM#@V3<oyvgvk8NAH}Z`;D#V(>O!aeB#I
zQ%$t@@sCc1rtr2RnzqSm=6jg)sRM6&<9*uKlNs3o<RmjI>(zryD0rKckNMe)4txa;
zP3+{*O)~VGMP`vn!HyYn76-%I+wit^X>tYOZD64c^~#l@)QohwTqVyC-fr5Eu2tj6
z^dFqA+3@xhyd8x<_Yk~|qsCeWZkjGK4>}MXI3M3GJ<Zh)d_>LD<qvaP%}CddQR)BV
z*yAef>EUNy-bQ0G@3Lunik4ksC+8h<amX?nQ63HJoU9c&l65i<`%DVc!>yaF3+<T$
zrLLSpUe4*t$?6=>?1^8JToaS@E4*C-Z^xl6+nh;O1w2T#$^NxBNYUxRDLM;-o50(W
z@OEWnio7ePYFZz>R{c^{E<9C!=-YPiw*JO6MIEPxe1n`Tcv}k1I|SZNfVYQW^2+p7
z{elJ`b1+pUmZs`-Q+!%Zsp`xuYtuTM%PB?MZm|n?I?RQ)(eU;Q42s>GsJY$Hw+=E4
z;jK@-MD?zdsJflWE@Yl@+CTBqOc=+k$G@ZPnhS4p!`tp~_9ohI0*~k5s`~wx!CwO2
zPWj1OdYsPH(6`iV+r!)A@b)*Dd>MT_V<BvXw>SUh`<v`}g|}<qZL*yjFCMGs<>Pg!
z95rGX8-TXG2ydVB^)q;D8_FJ5cw6Z_djamy+YRLTLd;>hp>1D~O#yEoR!>y6oZKtK
zl6CYhQLla^s^v@cI=@$gP-cACJv(<Be6Gql;AkW>q+7{KDz_T1&bByR!E^kG+VZF6
zark8a<E?EGGd-i@^cLPa!P{f+=+$0`Rn;AMrQq#Z{IR}EW7LGK;%E4AzT)T2cQ;y1
zZ$_&uymf=O)w@OOHQwIpMWa=W8nO|#zTah2;7*%H9kD6y06j&XbH$rn7_U+fc)P0#
zEQYs1@V4MKtL~s}t<CT$x3=nNIz7fG7NxegDAx(I4#L~_aL)#B6WW=zqZGVNi&F4$
zcA@)5YVVgwcArPGYc5h9>P3=k8>w=YBNb64Qdtcmb!}Fp%ASl=zkefT&L5?&qtK#w
zb6cfG>C`_K-F<CQ%t?zX?SsG7*-3ZLOcoXA%>_rP`Pv&t$a8qO-VY5|y<fuBI!Bnc
zybo26m!Vp4GgN_VLpAzjn1(*!wP)cfnh>sc*&|d3-r6H!D_V8KGIrO&+R}@{HHpXA
zjl6GZxDx&d*V2jM`U-0uBg5s^JzT@#?IHHKuTKxr`{WRfZyO?OXs{ae4c6!S!Su9)
z^<*FXs}sVWq!87v7othjep|E&(N6Z^eZ3c~dAozveO<5~Z9=DR3ns@qTs2!qXb-%N
zZDJ<>Jc`<9sG3ttUI3f-6%AFzN6gl44AH!5A=>>dSe=gst0KIe4sY+n+fv~{?1T?e
zXFO9UlLB?+QXu{5KvnxIK<9=8Xz@*dO^NhVKI+Jw`uM1KytnrK>cjpJAGK-dt)sok
zjJofsWH@`%9nM}M;~33%w+DUB`W_lfZS_WiyS`tbeoMc!Yil%aMZRzBu4d)XjU3wp
zOv>wL(lB`I_0Xs%zZ<mx{rW!LsNhXTS;reSWwucX@YcH9s5eKAx^&&B^`DLE9E!e8
zG_t4Gq#JdqFAqh}E-<M!ye--b{uVYVmHP7<cpJXlB=3FHrEi!N4QmT?^n|z7>?T#u
z;{5P-^l%tD4KMT*YTDG6SHas_hnP2N>aO{4I^Eer=izN<SbLkpImTVB;B6fqd#vE+
zmvTR`ho3))j_2zuYuvRVE1MxcJFLy?VwhIL#n7W8JAlxwYr@zo@P<7+g~*n0pbKGb
z{y{E=!}yZNRw7?2m>qoZb{)L+g15gd{+X5RY<NQ5)$1s^!4ud+x18+5ESTj4Gs$V%
zM;=7zMJGe<3r>cwukiN<I2ji6S{54E1Kz%V%x)bIr~l;{ucan^e3-N0)*sZ0;q7pE
zyBXeYfVT(#kGFrkAout)T9%q}9(dauox2p?Rz}+<ZsKRCMc-WFZ1~D`{V|Gc$A0`C
z9nid-t8+<e%J8-l_X2->cQVX7#n<pQ2N_aFMmiZv4Tinw*(%-1Ze#}NcpG;5G@+O4
z<>2p(Lz~(4zIK8`)(-5OEs9Ty8PGHMkdw$Wss(S0b|e3|EZGa6Gu8ZDhGs3zAkQpA
zY1d%nu5{g-hxZtTRU{wT{_G6s3u~**VotRHc}$j6-GV=7;B5%Jy)%{hncHL&qjCM4
zrSso<x{{}+YiC1xzZ38qwWlVVjhUK3sX9*<(#(3~2OgtVe4hOy$H<V|$$Y765+1%J
zb=jY!@^JNFwPclT%{(c*z2}~!g7=w6d6T3aUy^h$j^1G|>dn8vxnATU!P}ASgNyr&
z$JdWso$T~{(VpFQvB&iub42Ld>hN|Fy!{j2deF<ANDp=n$HvqY<)hZTiQKBY6JhPo
z8Z*3o*OE-g^3=^!Q*{oHR|S68EAz;o2uabc9?Y+Do`KBoxT3%PxHm}7kJk1|)aQBZ
zQM;3<{CIhnpk+VACur5}1hx1*K~vyYZS?GioA{J!#;dIf-3o7`MzWg~-abR8mhXV3
zr5@{*9}Npv^QFe=ay;3)^gc~F(T!#83SwvZaCkc%-uiT+cMETqPO+=rT)RH5<+I?e
zAH4lpQ=S5EyU_D(mu%O0XKKV{<5dKWyOMrx7moTccosG0kMOn*ye$N88^PPE!SQm=
zk5;Y7?w*%;vJNKb^N|Gl`pk=jB&Yzqz2ly!&6nZtuguu8<L6EvI~n_&kNo=C@HQ02
z{<De9Mf7VcH0`YQXkGYgTFTdC5M8HM+=+|_SMu!-#>dE>jt{C3GcTKC^<+D*cVw>s
z&z*s{*;|v*1aBw7+o|+wr_GL5&f(E`TbS#-i5F+9O@4FOVLs2M;N9$egSVdW_9m=d
zGS901NfvFjSu`o!qTOEfWxrY!{njGaTNY(rv}kdPMYZ6qXI}h7)RHIMH>=cIvrg1!
zUrh|!@oSW7*`m~<47=js?eZ*YqphNpPM@;dC}vvv(97f~2XEcrZBoTZU4^%$>qaU&
z%x%yL_D*5P4f^)u$S7$=l(s*k_WL|aew{4(j5llgd+O5%%?dhaR-gBgI!JwXI5psn
zgCkTI-X0vxj`il@)NsS}h1`cyr$Tj|dTgy#=);kr+T1Qw^<vSUfx+^4M!)k(uo~SC
zCf^7?HVM<FdEqL{;XEr`m!^j+8^^6Y;mm7=DDQ$`#k~vCzmo$s;d_vVd<)XvALvM1
zkWw5$3eOd+!S#am0N$=^jep6NJv{w*Zb-0>jtka|%fZ@kBv_*_25T>0FNL=$Yq^&7
zJij$q-YbH&aX_dVxr8a-olx~X79w+(5H%kgtf=<zl-&Z&+tF*?>aV50hUmhVV0Jp8
zm3Idz;C_JSuk@F7Cb`F#eRYL7+HQ+{<=@FyG5C#xzcAC|<fR%`Py7@3pB|C})z3#)
z1DUC{du#1;Z?)^^t;0RN^zy<F)lQ_AYNC#O$wRNHt@df;fydKBWmDh{TF-44v$60t
zmj3A#hl$)nd`YkIH=#`*JT<B%wPJgFwCiZ2dcj-UYNKAvGpY!jbslGwKg?a^X4FTX
zPdH@MR(M<Vtx*j^j9N%dc?_PZq&g;z7;MrRbZ$^p`nNo{uaHUG8<^Cwl}Z1=TYq@l
z?V3sRL*Q+iNsp506I)FROEM`Zyv+q~_xz3yp2%F$0KV_fUi*IT8Ub(b`I;2?*`#~t
zP3m>PgqPo>#&b=oI@6@4F#hXbCgr|p(zHO63T&ok-pyU9;U@Xui@FYPbK!mZ{3M&9
zZ=Gxgm&?>x$&9-7oL#c;b~L=bAIp4A`)r0DAF>%N>?1fvUhz|WOS2l|Z({FFzQxSW
z(BJ$qf>{|jRlEcHT=5!3zkp%x?1tq%)vhtCde+JC{50Beo|B<jbNs%r>;id9?_gaP
z_2$oC*q?*$J&)IFD7>wT#+|(m{ksEi)Cqd6_>C{1ZHL|>hmjs_5qP@<-gbwzkw)^X
z`2HZgEw~rY^UqrJV&;C}t#PEY!3J-;z}r^n+kx<QQ#8AIJe>?tT;sqKWJIB9|NJ>)
z3vc7_Muk<QF8&MiJuduiMtsOm9C~!tp}*PtHf4!JdB!?aqAk1Ma!^-3$zJjeS*i|i
ze;-M08GY+lj6H5&GPRu<+WuMTYG+K>c=T;~da%vsrt9EuJeMb3hqAFB2mP8cGffUy
zG=~}1(YsUCy+f*sdC^N9jo%l|*xjD0_Grep26o4eA>Tig9p>=(K{P$v_0*pSl0jG{
zRU6`xRVEv=rf1l5;+CW{ZId*v9vKU*IrcI?8i~J`d73t8*io5eDZtx+)6_6`lV5it
zNzY@G)D_;&hqqJykGHd);%ENCj%+x1KaF`;=66dDNzn(GYeC;;FG2Q2UTVtOQ?(Fg
zjtWotA9F{grKmzauAO>w<0fP-jbgrMF+MK*QQ3=;DTRl+!e?}IS9-yPQpk2^H)sI)
zhF6&FnUKuv2)VUcN$OXWz3b@O>~j<4=0<*S=R}Qdo2Xl8!Q1G*=Ofr<R)D<Vka(Sj
zw<VaziMYrN2R+y<c<GC0=n(befz*i~)UvbN9&d4Goc5)}sR6t_o)M=ax#-8j+nVt9
zT2-?6>)N#!-rkx+A9n?r2e7u$HoIONfx)-X!oGGr3}tT)yxo>)*JCH<T&XM1DZ=?n
zvUdmG7KgWUYB5tZI9}b!fBNfOyeh!kA@H_e`veV|Pxir5Jkwjbet28-PJ)hkCMd`A
zL=E$b*KTHu0?^6V`0Spq?fT$qSB<zhJ>5somf6{&@b><wICY{wyKp5t1=huB61+_>
zftH207Uo<YM6p+gn(?a)JW=r0cN4l6-hRUe6wBV(CFEAj!1L-utvCf=P|eZwZpor}
z)QpV9D$!a6Z||%nBX9t-G%anKTEeCTU#mX<Y1QZjR!xJq*^(`KXt8JryxjnAv%y<W
zc-sKpt~+nhejbaUY0t)4w0)IDJz-wu&CL7Mf`8U1<)d!=C%he#jm*J;QM%BXx-I?4
z4woa<8QzYjFWTA|p_LZ;q-7$t9o{a2x7X0Q_o}no2Hw`7*4%?y@1wR+YPbaEK8R8P
zyxrf-qJJ_XRi4^#_Si^fsH4<ud8FDFj?iEDp1ih)YuS)+O@OyeVPVcfK{^ymM)2oA
zjkz4C8213>e-NO(;{sHDJN-*<KiO;esdGU;vJ?C?cn_TV<S)0LfePLmr0+YKVVx1A
zY3yT9z7U{#!~Au6tDkZ<2vB!;duTkLIf42tye$TAyEz4`oHO;^66n@nU|_RgCBWQG
zxyex^OR-@gUPFKGIuxt{@OCu3-N@H%n3p{cZ}V1RKOTPK#ck=mt_#wUY(ZKzERcPM
zf%+phKnGL(HSUd{oV<M5W$UMLRYDZ=DOhC=2P=G2ke)o^_k+WxLH=@o?yK$acHTl?
z1@vJ*JiPUSw}%{_`eyOe>yMsl4R3qb_0eB`-dbJVTTjO^>oeX{a}Ay{o%^A*ATkVh
z(%YO*?h?H9fVb7)tt}WG`xcG3n_MU6di-*eX^gI1{~f<EY#e^TDD!lBvv9V1d!s6J
zHfsD3qr#RLbqd~=fw2#HZ5YhGJBRP#?HPD`?5a^m-|#+Pdbt*(mb#eK(AlIN6-{cE
zAI~zi=$C0md8Hav0-d}K_C7@C8aA3#{-{aQVeZz4CSCerQVk1pLkTARnaRviaeT_v
z+|{oLb3{(;*o}mt9(c3v!_<G_`E!$2o-rw6KR&L_Ce?zssWVJEJI_Sc4V?d#-Z9tE
zcOms{bn`2C`!1b5dhm8nE*HbnJM0r6>v+s5v?2Y}$${{TT*k85vl+I;;A^r_>%GqY
zSMrhn%$Cg%4wvs2Vjs^}`m}g>AHvxg714(6oeci%;OC#@EWtAKH!^~K$X>b%bIB1l
zbR#FCnUleRUbJ#tgl&JX!Hb0cUG*6sb0*nL)RkvaXCAPCybHLQZ#%q)w*l~W1-vbL
zn^{_zyYC*JEHv-m%-GI@w*m0>GQ1rCZ?~?dXS<O6idoKv+weBAKl4D;vSUAS9dOoQ
zb}|g-@dDSl`Z&Aa;q6$wQMda#84{}UI~8R2Kr*%B7>Bw>I8^<GgLzViy20Bsi_y1Z
z96IXikXsfrwPYLJhp|_vAIERZ)ETlV9z|tn=(r3;mtbcPzT``IrhaDPZ(E-(S9EHN
zVD`xRk*nxM_5+zpB{z@_(KJmppQY;QW@cI2!q6irs*D%-Hd#*vY-B*8gA3<KSLRFd
z^AED0XGogfz}r5|&D_?{-f4>5zLR?tP6iRYO`3)O2HyT4_ht#a?YIX92b1jzZ-4IO
z90qT_jY+z@8cz;Bq)*53Bf;B8@OFDSG6-kV-~F5UReGqG;cW+a+dr9E*P`&2y7H(!
z_?qDD%i`o*psg#pFozpUy&2zfZ$I=c+#PShchvy>3vYwqtrxs?I>oGSk5sj)pQ?zD
zc%u8DY0I#$&Q9KvDOuv2BiWx*%FaHVnMo=aMXeRyj$NFn;gb{Pz!!C{5}v4-1ieGw
z{>)QqTb{k>5o80y+xBS6%gjoig|`FX?GCiz#GYi>!`mD1wwDW!u(kr3z?<%{=gb<X
zYw2;io6Ro&GW2KJClk*5R^X+YMBjG&68z2ZcGm{GmhZN!HoQFmZ~uX}gWzrUB>c`W
zG3qM5W@dU@p{@G`(=*QC{OH{pT+f>w<Z8g%{P1=Fyd4a0+x;JJ%Wp}L%TfBx@V3Mc
zp67AZH|oY4**88HcH@^W=xNtq)QM-{BaYb-r|}Eo^yH5??z>|3!8w-P-55M7F`AN>
zxzVDrnh$SZut&fT57WO*qqKxR-{2HjM4z(_HAmA{`ks67C2g^3&>Nd>!`l|{_7u8y
z^i*od&GARU+kH1|I>K!2{=PPCfw#qT*mV3Fj9o`x7Tz|yYLO3fHvRsxD9=iZ2Cv2^
zyw)P`EfyVvwf-m2yXP!A_|T$(hb<bJMcw!rzGHaX0^TNuM#;K`-Eaq@H2(J}nc(e7
zm|5~St{G47vd<B6A%D=$u`(h;F$M7u!P{Bzwp*o0^1dT<%Nn6f>dh1SMQC-;2>m)8
z57UMSO*<Z;ua6^CfIjCWn42#yQY!-@b+>n<&bGx%91)@j`l90phO5@cFxf_js%7mU
zO@Ozr-vp}n>OihBP<!_UC~HK3zAyAw(Jy}L1aD{N_EQ&l`+NtP!TtP|>z%*eT@TQ(
zD*>8UAwX{*`K!r5f7xSvG`=61!LyiCg}H0stq1#c-b4o|D3tk{kN}PI2w>-PpzLLW
zwV-jZKGq3VZut8zKBo5tf^`e#mS`KOzqpo=n(Uq>&v@RK0P3~@($fIh*p=s6jMokR
z8o$p^RR;NKAhSAw^3$c!zRU<v1HS96{P1=ZHR<K?oab1OhQZr*WJo>S<F84#{q&jH
zon)hjw!9;!@CJE>^j;6Tu$zb8<k6M%V|`)igCF?pJo!Dmw5+zb_Kfk;ufsicsjR0;
zZ2f_M(?g5kt@}I=y@$6uTX|@slZTGNTld%al9tiG#V6Gv7uizi#Nl^M8aRL%T6nu@
z26f{RMinribvqhm8D!L+Ka3hV-l!uy?>@??Jn%N(T%)qDqKA9Ls5#e+3jK%g;q9w1
zG_Mcb{%%ym$42_6MkU=h>g-FS9tH8a=|<JaMjyAONd;HX*F6F=;q4@NyBFS0H#1)p
zk0yn;vx_n>RDpa*csmZ>euKB!;BC}BlRmsR>HaNdf>+^%?m*^7U6YDb<KCex`We0_
zqSY_P7`4L8wQvvck^b?*L9jZDzT$6WBDu07t{3}PVaJsR4uzIy{*{^6^bO92(Ra~^
z@V2y>qYpJ+Plq;@K&zH?G8EqH(3oQmrFl3sBM&=q%9G{)v&YHj%`~uEot|e0cv^4~
zp5#O1_HT7E?CC`RwycxkDy+R!-^p-*_com4WT*;nTmDPEIg|W<G-xCI%l&G?NWPA5
z0z=#KTz9<2=-!R{oDE*+>#wl32^`)GZ)YAS)8a5$Q-_!v;`M3pwluu`0&j1@+sXs*
zNyFMR=<MOJe^5BHKOgW$J%+DGoeYuOAGCqDC*f@xyq#OX$*?=ap+)Wvt-Ql7^CJ%3
zIpk3ARtIxf4sw7T+S<yY*mSac*~7XK-mY+Ds@MEX-GH}w&FrEbP4-i<46+c?wY7J;
z+`6Re&$j9Eu0(cJE_Tv+li|OVEMoed`Esz&{VcnAX5lGblBxypvoE~u(lkX4j<Pdm
z85u<_*%fmrQF+N%+FCh9;Z4}})-pwL<Uchg=jnDa`m|9n^*cHhhMp{$L@hE=*Ao(T
zvSyMZ|435rtz;PSSqC_x*O0}#p5525w#!O>W_yw{VDj;U)S%CzeeWeH#|=L33EBDA
zleGF%lH$IRLx>J_uLFah;~D1j&hR-+s8#o9fNq|bqTLVKg#&L3{h+QKkfJL2nX#=(
zjog)^H<=t=nE&DHP-@RN{zWJEOwnw3yM=!3VR(D)bh5Ug!$)x~8PnNG2XE)m&+KuP
zUM;+B)Ph;ms(6Xu?SyO8VuvN@Pk4Kddh)5O_><sU&TDWG#x|jbJQ!_zvn93VJa#>W
zwF8~)iYAx8|0Obp;BC2pI8~#z+#cQ*bFkM9_Rh&~*H(JGo2OHQ-cJ@Hy0*<}yJ}s>
zkNn<F-mP8J@J%h}`?DA9<bj}j;cdBR?4yCVjRTl_g~LnV*j3dlRylgcsyaEoqsd2@
z_ZykG?U=nCm!Jbn6J%XM7V`5rMKz7n0DG)<-Hp|jN6f<FValN1tHtC8&xf-S^hse{
zE*JWs^d&b9hi%NXoc$J~A*nIi1aC*f+nPnF6`zP93p-jDJ4UI{f(W@gBK6DNC<UP-
zCl)Yk<pQ%t?lNoqMjL;xY<lsD-mJe($(6}JzRFDMSsOVK?0CZuHN7w%srOcud2LmN
zSY}qwTb0GUX&7^%cbRp~V(#YmLl!k$!Hf>PZL!XxM0h(N-Y$T*v*7KC`xfQqJs%2L
z<nf#w3w*}j^lb}#LD#N~(tUL2%y*F*j|aC_cY3y6BjsIzS(p!afxkp(@TUm&<V0we
zbEKAMMCff$gq+d5j&Bh%^Lik?*a7giBD|eCGD36LM5rgc9ey!FZ5u_%XF|AK=7ni%
z=`dw;r#`$SR124fFq0CZr2gUhem6`XJJYAV7ph{}sN259D+F%^<1WG5pk8R)Y5_`l
z<*!er{4^_@pZcDp-s<K@?blzwUH8}DSN*l)p1*cJ@z->C>o>q(0q}M~M=#xP>7|Jr
z8(`mq9Db4m_QBf|@HRW#e0DfM1K@S)!9Wd%w-Mh1nH3CF$$QMeGK<r+QJ^-|2vm8`
z0JZrTpf0btmS^n4dlH~-djsgH`^jsBw{msy){F!%71`mX&NIDK;*ysPCFrkS_g0An
z>b{?R)Y{!gWyxI}6zQ#;ue{jf>!nhc*o!{>2i`>w&3x>kOC3B^yp0EJ_0S!gyUuK4
zzBJN9E3UEUChCU@^z~Ftv!^0Pc&ckJe91k3Xl(W$+W3hq!;$RLo68)`c>1<{#)Wj|
zT7#Iifwu)#vWurJd*SgZ{R_LgbvLQQGozj_H0s+#qZ;)yQj4N4+zg)fH|oa}qe5Wq
z#-T?2$>XxIMiqp&Yw;=9J;+?pd7}pN^Dp4+>Ht26$M*2{-6Q(D@OC%6T?217cp6pn
zu8~<_qm2CAB7P>v6Qi#3df8>vm|^Kp4Lb+kCS|}@czgR7yj0Q56wSmRJ=CQ0$LI@l
z&707;JFlQ?(Xm&!{*N`OFQaDz%b0XDH~nFFn+j*Q&{Lkv<9>KM2;R0KpVtm=yTQDo
zu<=kvmLB1~Eofr*%^LW&o|zhEb8^B;w-IPTt3x|}IJ6$#&WE>m>1DdY+ijm5vV}Xy
z0d|me3b)X?_uy?kJ~L?`{m+H?kq(e^xXB666|+HQ*ah2|o$c_pUmNCbX5&G+#QCWy
zCkL`GCpSLiiq3|X|HoT9ylnw*Bk?<pp+3BvUT)^q|9D#m&Tix0pb?L;@b=hHbT9V{
zvrFKUayS_(B$KNc%(*_pe|Vd8&dJc3x^~wvvMXN0UTVq3x8d=kCSDlceulT!pZXTw
z)=zh^Z^ofYm)T8=u06bwjEf~`+XdWTz}pc`IF4p1F_j&(1F0J`3p?yN-rN(+)8c^|
z5|^%H{^`2$CS48S?I3u&0NxfV33KbRE44IwmD=jqr|b#%i~c1!zBOl%RWYA^wD8lV
z0{hB6@aWQ0eZDM7>vO`#{p>fzL-VC|qC(ClYDRqG|9Z1N2}#=Glcb(6c!Ym3L&@yT
zOjI}efNkRN7u878<SFbATT3o4pB226y{UZN7v6@!+7!G<C*W-wj5`8vs~v^IhsY_s
zL|)-zW^k^eh2d?!m+S&P!46k=JCbwNg}2+`buZ4>+?AhkBYPj-7J;=+99KQzFB*6|
zb>oH{K_+(8Q7b<PZ|{{!r8bkIzu|3ncx!Z^rQx|hytQ<sw)_wMWOzI6N}_@nut$ZS
z?F`pMEp3)a&RU{^qRC3UK@LS9YRRy+(7Sm3zKz|jb=haOz^;<8b_$wyGF~XRf_7!2
zW_$tOzGa?f<tuj1U5HcbTX8A{ZwtZOitx66R-8iMtsnaLS~5Q1rOdQaPoBCN79XSM
zOYe5gTl%}-(Xjv8mC0;x2Uwep!^YRGpW2njh-aDpZt+OFE`GGL=bT+}-SGs&%>`s5
zPWhQJR40b~sTi^}V${V!wh(pQ9}jHw8*F+@#&9_MUh)@@)|fUisyH)7S7(s(1aAja
zAV(pBc~p4o-iyqLau%iUv&!F%jE3d^<v1-HO0I*GSzYn?7W*wiUFJvV!htXytQe-`
z_)y(09HypEsG<IDlQ+HDLY#N@JDZ;S*>n+aR26vp9Nzv(?YJxFYIDV+jhw@mJcb?c
z-5Gss^N7;<j3_Ow#7<RwPBH7ua)P(j;B9kwYlF9g;B7v5n;+i(xI-;@yG3pBEG3Z}
zbs2wfAfBduZ=>iRvKJ2CPJ*}7@FA~6)2``&uB{!RZ&f2Sd3c2Gk=d{X-d=Y^$PM1U
zgtr-J-c)#d9Ny-Kw_cgd*|d$&p%D>ETpOXJqiA1pgKzS0P;^GP#_kMLi{fD#Zwgh+
z{!pbce{+0Th?*1(p`MG!_+zNLJ>xa%yOrMsvWp^6Z5jkB-*dEN?*Kh@4q#ruUzf*{
zr<CcdO9y?mWs9$N2Ky*+uD5Eqd8^|TZ<TCAwo*}Y44-)^up7K<@1>KlHngFaiq@nT
z+R94>2l^;~imyg>_LtKJf8{tDpwHI=boEYvw!J`4`vvHMDL`R^s0nxQS2WtSRvmwJ
z{F_`S_PC$>?yH=5h0_Ck^!%QWhAsBdnNQxzbH!VKALcsX?WGxBnsdQRh2ia{mfrGx
z0B=9~s1m#_2yYw0+ne{j$b0lsy**x<R|@8S_0Y>BaJC{l@A9&XE{eUeX0nM_kyCii
zLwi=R4{t8niZ|Hd?!@d%aSv5^<-zU_=3g$7xpdb<9Y=U*2HMuWH@Ocj*xwFsV?)r2
z)RM>J9rA*=&){w8?I!+zjCc5)QH2K^HM*`*X@!mWH;o$J0OrEpwUc2eob5wxc^ft6
zyXoJhUyb^_$w;OF@4IN!LD<>}&d%{O>V&sZ#l9IepXYYM+Y<2fcX+$^2i|3PJ7J|!
zOZe<yC%9XI&!H~udX)SO>e}@}Oj=IuHv--!plhF%VrB^54uiM#hnRE<-cG$^QkRR=
ztWV&*T1x*G-afBFJ-Ps_&1q6aC+3*otp(0zgrL*82e`o3%i=jVIxD^<Gbc0C^#R}P
z!;R$iUO~6Pz(+1*JiwR<x9G1<VQzL5yId{QP!rhKTG*kzbsajp$f1AsIOOq$3`2OE
zE0ymHp<64!xiE)v#5hzVKN^=>bCo?#hHLAnHP>V3E4)oBLhqE<_SMDrOCR^kXxIyH
zyTkhP(N2bPF;0edHhzY$$7MJf+7`l#Oy9L<Y5KaAnCY2L{TO{a?HJ#~+xYxsCFerJ
z`p_d@$*vx1+|@Y-&*vPy@Md)+$G8Q|$BUd7-VP1ny5Mbbc-v$f_W@|^@+0Zj_M^rP
zZ#}rSU!om)1#g$*M_#?zp^B><8i=-iGS8vK!|_WsaOnO)cD^N3KkiP(BK!!iL0-iA
zWIg>mP1mRo&n!cp?@M;uETDecFjd<=vIFj4e7ql0HR?vH9^FaR1pLUk;BEgk@U~~F
z4mD2IU**w%<x_O!ZL(&JOID@H)P2V%>aLiBsYRAAyv@IheI^Ik+j)$vxM*_Hz1f%g
zlsrIqn>2#`ZE1-zd}qIAM4~?CB7ePU64{G*qZTAlgZ?j1p~@)OIyOne=O*d(;v{Kd
zl5Q>{HxVAE??w;9;L=-?)bj5nwe?^wG?tw9QpxOdB**yYd{HP`_&%S}B}sYVZSUWc
z*nLXRbq00oiAgd)gUMlJ?W4QDaNqEqz1^of!)vr~Wa(6OtH!RKLF~)nx;_46CaHX~
z))q?EA}c=+&z%b->Yuj>di^{>;~!FYEsQtHB|)aI@fu%@`Y~RmSZcfl@E`B#P40XT
z*xJpmN?oY=8vg6qUSJmWJhL*J@kBL*p^LoZ<jr34z{7D0x)G<vrZ{==d<wd^2b_Hn
z9H-bo_Pz34Rd{<~xm}B@at>y0?(n>goIan<)Q-`$UFZ|Hg|)TM;%&}N?Rf=yxdr`T
z=bv?IGU$V2<?SE)Ki-b_j@Br0ew&gpR0-bxfVbu0?Wat#|KZ(i7?-(;jK4o@YBG%a
z?iu>2JU{m(vo@1#|I0%RZEDe9AIzHf2<9c?wRK{*=|OZOGc~<nY-n<n3~1P#=c81F
z8J6V>qBM~Cmfw$sX&bz~6&b1l&qMVC?`_x_i<+W4E9A5)e!5k4@Dd-!tMu}FlzQBc
z)C@6e(mzVh)Oa`hM5>W<l<G2v^Pc&W0BY20Z$#-aebO?2n$`S>S(7i35mlH>iE0+j
z9l@S+_S1}py$n1l;*v#|Hq+1L^_d2<vN40xlyg0zmV97pl$uwK(xsqCIgN|d%|{XH
z1aEECs128iP)m3_ab1M|_J)a1!?ke*Jy-a-r%Sl{EeKZy^z0F4Tzd@T9)kCWpy%iQ
z8KI`|w#MlQB@aXkKg1V&E=<M#4An&Zx-;<Wwp<({H|9}~IfZB_`*7-I1nDWchF9o;
zW`nma;qA_bfm$0EpfWuIbgv;yq!(K3cR#t)zx7_~E7#4wsxkon&G6CPsowZ*y_Nac
zODk@A>F{+g#dP+PZ#&r6!b_7IcqzJumlA7Hi>>cPZ`E5SH)_3GeAHsOmvY|b^^rdM
zfMyLU?XULCyUuIvujjBfoTCD~Z4Phu!`q1;eU*IISI+}|H0F+v&b#`kN`SZgA9?FL
zye$Q9hm`Zuu}+?<a{Y(2_J^#?|J&z&;PemmM5ms3=BX1mytH~Wyp1Do@L%@cT&9k^
zi@oj@*guyaZ5zp+oX702o9eC$72P%JS9+SAn0q<vj(^NuZwfJgnva_CG7nXUxA(5H
zyDm3557eUjcJa{Ek0uQ$M!gv3EHuGD`mNik1<zZ<4DHXFZXTmNQoqZ~_d9v(-}UkC
zce!|f*V!6IZSDyhhZ}W;<Iw=4&UY|s7QD@b$Bv@(dinnNRHJULGpf}K`nPD<S@8Dc
zS6F(_sC}I48IPS`Qdjoi^Dd#E-HfVJ$H?x|@0#)p9_VTKowhIsbjYOBXxDjh%p#?r
zb(3Ky++FU?^~2j;=+k@jV+T7p7aF!_N&3JeO=OT!pRRxp8s0vJxB201S`u||e&4ns
z_`SHEXJJM;!Q1ks+3O8&_rcrhhv^HFf1hv~#=@l+*?#VdBiEq;S%jgP>foBC&~SXX
zc6PDg$?aU*!Om|7`rIL}TMq4`ro1D>p<Vv4C_Db6P={8*t9)?Frvt3!^G2sRG@bqG
z@8DG5KIHfNI&`Bpxx@5(b5nQr{maSV{Kd&o0LB)Dx7V%w3}27@Ki*D&w+G?vwh903
z`&m2!PqZsL_*yV`%WRJayqyehpTXN(u(lG%KzO?e-k#?gsx-%M3~vYSW53TD?k9NP
zPwjnZmXqPJ8=4yp?*?yO;q3^nd(~%${C=T^+&oMB8)oSr|4cP{k*?@#%#c=1)0hLP
zDgq;mjZD?|wq!+6<4wBHZ0q7=6+1%4V(<TU)49Ilbu{az`l-5Hf;rh7sd}85qC3uH
z4L3{GFzTv5Gb&ogr)Yt1in8JNjfCAV(6^<S+qu~>Nm>8m%^jDhNpRLyJyBKfC8){j
z1ZD*iG(C=46dQXpe@W1h;pF}C{Nmf}hbcw&(lyu|O#TwuxJVB6X&0s@4R4cXvNsGS
zcjTBeCP^QtHFuf>Qzs;8$~bDrBhj?<f46L+_PjYs{ZFBt;q6X%>rx_FZq@M+=fkrc
z!Y&nf`)?;S_>)8>(qna7&+f53iMn(XZJkP=^)Bam4XZhC2R<_iO?<(Pdi91Rwco`K
z(Vgs`>z<^-#n{0eL+yAg8HPRhJ&Pww>Ffi4M<yb??Fw(*3MD8sE1p~2cunpUuU+5h
zm+qwYi~l&j7a9}Zj)u3}+oCh!?RaX!yDOqivr|`QZgrkloF*INl>0@T795OI;q&xu
z(YR%J?FzhY0B=vhTMJw*?jNT=>63;!@bi9%Rlk?9+WSwe4smqk>&>s&dHy6;C3##>
z1kL+CPL-L**+hP>7g<iL@{lV}9)37Gb^72#x;!CTiw8w(13l30QC58+58};St16~h
zH0&k$5Y6bd&P8Lw+ra~ETEtAvfbUir=a5Z^Cn*E1*~^C>X^dH8;O%61yCuV{^2~v3
zzef#rSELM^@dkCme@lK}C!cWL{u-`bkHR(TXqfs`4O1_8yABQ4>S!pr521417^?Nu
zSBtI+Q?uIP%3yBg4%*j#CtTN$p$RvKYeqDm@g`FFcSMqB8==O<Bh~Ooq`Hv3&}Bs=
zGsSq2o<?a;YqO?Tg10TnvlwY%?!%&zXx`WVkEOE=Z|h*XZJ3$4p@x}pCk<1>%+xg0
zFmuv|nVFdyW@fTQcFZiv5U>m~+wxug-uvS`l4YUOYMni^XAko@3t;<E`nrY5P+GkC
zo64+<(lorp_4-HY!f#Q!LVvYBuW^mJ(kt-w0yW-7rNb4$j85(pyB1{QeRzfI!QF6;
zg15)<|0cC(wlycY5v{oI-f-<=rl-~oySg;AD{DTxPU6>nX|rh-pRaH)s|=N_dX(R)
z+IV-D^|$IzYROZ%{m;ZOee4^iZ}@~$$yqwMAXI0WJuR9Gf79$>jhq^cyDV6vi;^8Y
zhh49m0#xC<zg{m7pobfv!tk#7T0iZBw?hW8Up?AaU7z@>70mUemVCawue!tAO7(nY
zsP3x<@U}a=4ejHr^22>KogQe~N?#qk@2erNeYLlspQ6VF=*fft)!Z4Nt``IJ*VO=B
zdK{pmuLATd+V=UE0QI{NpbzNlPR!`Mn(Z%pHh;CggU+txr%SQk`tqF}cVXTd`LDO`
zt@GBKY2M7bdh13dZ*?f(trEGtbux>$=BAR9<V2TxdXe|!rEDX~g{q3aO(0{*^i}O{
zvrC70n1G683d7t(9oR9u8^12IFXLeHRrvZ_5Ob<FauCZf1C-56DTA3`od~0Pd*aXZ
zR7v<#2<BD2PK}kiaddlXz)R?{=Je3oFpKPPHsYj3U(Q%G$J-)fHTtnVJoI@4S{2?t
z>kpUflWmb`QF5R~JK=4;E*|_hNS_qm7J81K3ZBpK#mD>+-tjsO;Ol&N`|~D@J?X)Y
zV(tfTE5h4rFn)Pen7ALmRR$jCePl+WUAJAOc6=1gy4zC&)_bZOwdVOVJ;_1$R7^!Y
z$Y|Fi9v-^M>`~A1)WSQEuTg<HqErw4@{RZR6HU!$O^xzU{ZROS8U2lZf5=g`M4I+8
z)7hRnXN@Ci>Nh@JgSTg>VFPx+&CJlc((LLk<We{P485)G(wu$FypZjX4c=DcSX(Po
zR|il-hPMWEVAa4(b>Ee#iDU{lIG!mpwdSYuGqs~0^;vj31m50w;?l_a^hA69cYCIu
z56jfN1DVXJQ}cy`<LIS+zRbMPH8kxf@|@hzyc{JA@Yl?aI}h@o(7=m*=n=Ogr;#HF
z-WH*DJpOBz4!7=8_a20A>kRX;r!(bO9cJEiY5Wy(D4L<0du8eH9@b{zs10uy!rT1t
zwi~?70dE(;TWeMJ#OG&^?Uzi&<)?Q`_VLwG?8c)`{EQ5TwR6aFc+c*-BV;F%-IO^h
zMaAH)(M5I>y#0GIbFRJ#Dg$qFjAMs&KX!wUPS8`q&EhF~32*Ph+jWWfp?)T-aZs|H
zWEj^v!7MBF)?wV|N|6-3fww!zgYrc;wsA|)rSS>!u0?It9Z&8Rr$W~{bqU^1bFj<w
zN4!!?JM{+M`u|NP?;WR-itw7$VmAlV^QG?mM-KY0h0(dS*i}<EQR9B&{fvaEu+<4`
zf8jVbk)3bwb``g+p#I!sWuhj-+t0_CC4#pD;BC7S?AfV`p9+0jHI!XB^m$*y)h8Dd
zG!xzy`2}75oEq<KC%rj#d`@(#=~$<Bj&ds7Xs4DWutx>n=l+L%XMeNT6<s|I-sUQY
zR))7-;dSp3^lFRajY`1>^^DK+k5jdI`~Y}cCxfiT1I)M<X8%nVc88K%_`M8Xr;21B
z!`C(KsOyT|16-d<uQa6+zG7;{6CJS{3~yt|ikL}O{mxyn+QQuGAOF>q$(rBLKSocR
z$0)8^j8em?_0DxLZ$;kiF^8_RWBtZ{_UDk@dt@+u)!yX$lQaCHm_vn4<SDV+waKGc
zWg|PHEBEhumiv+^zj!qnP2}aSz=Kr3OtcDBM8}dHF$b33fw!^veBV_4p>wwH`e*ld
zb(;2FyF9*WMNxJEaGz>{KiF6MLq+S5%LH$$`{6r+x1X6K^@=0s!HL%G9;MA>KRnBY
z4xJato^SlT6(iIa-kw0)F3KIQ7o+V;tZFAelN^L+HnsYXw`*+bJjteOV{CeG$R>|o
zc0Cwt*Y&C4nsqx|U5<yVo!O?hr)_v*teQ00s;hTwc%LJ6@2^N5rS{ty-WGtje$S#*
z<a3l_YJOvm`kP7)`=%>b$Ss7s58&;U$?%rh*mZc5zn_TGfsIj`3~%@MjbisQ8l3*}
z%3F~-fDWyEGhB@agsXeJU6&*6x{_?ynXKWQGluJrBH_wYF<jLu!`|ZLGvx@^I_@{c
zX4gs{^P-hqZA)?Pz})Hh3wUw=v8u!Zs}B5T)eCsrB{Pg3SeSOg*_VT@>OMJ4w~B@F
z|5cb4!rQIzb_5wqcgPH06BexWFF|@YHb{FL1hH#2P>t>cst3Gn5aF+ByJ6gXfBlii
zUxndqFpT|-MlFG6Jc+(sF~e8YI{9i5yzK;UpTgNOi9Slb?W4m}eDqen$^mbO!rS?a
zef4BLyJ1iGYI%zQRjlHt_{x6j$g%iWe84sRRHGI(>Uw@U%)D#dazA?D{_GbF&}^PJ
zPY)k$Z0e(yIeqjw$y+^sc<VL1eH!9T&b7Bj!P_(N_7S}O7v8pqx5eOXLZ+9N!P~F#
za59N(r_8VVXPB4D!rOCB@`FFHSN10R@K$kDU_PeOS9NXsm3a)Zjp6Owd|vEa@lxH(
z>~Ieu@Bat+#5}GD^EAg`*3GumXLEY0E4AJEe|a#M1PhB$!*#Qe7i&@5REw5|TGaHO
zMc0p8n5DL8hnGc-D#O<r)P~b7`qPdUHCq(SZ6)As8()hGpn-F}v8c#XIQ_|@nqMuP
zV^}n|5!`!=em+GH6~<NMb!Wi5d&kMcpmyAQ67}lZ9=elcQ5|^O4&FY8(~p_YjT_;q
zQt-Cs5!kyIEemfCtzb9840J0@4MM9{d`Ca_13qZD+ZEmpY7d{`^<h|h$L1j)Sl{-O
zhpJG!p2%^W{_-~7#{-VWF{x@9!T!-m_;EN*of4=mFHBdHgmj&pn4zfZ88X}0LvtxZ
z$Eu+3Hn{Z3$E7U}mx6L<>Q{KXuRndwOK|ry`G)A#+w19XQfu`diY}zb*_!JUnlrcR
z=F*<MJZ?J~{_H4yJjbQpfi5+Lw^>$as^Bbku<s`G5#D};x7n_vGvRFwSSyaK@U|zs
zUBSF;5?s9uZ^y&h1#OtEO<{ik8u$(zK3ouwu@n6q>QddmU79@7rNR?jnz`FW{s8st
z9y|xEz4~{i^qkqA0$DnIKwqEwi`Rq2Rojw-fxc~$2WIm*e__sNX#{gU@b<%rRAuo>
zQKj!mvL8vJ&YUEF=20usdpy^c9wfZ&0dLds*1q&hVD=Dw*)Bnm-RV8T+iCDNtXK+u
z3B0`Uwi>+k`H`#?JG<sSCChp|S<gM#*Ta48q$O($&sVBdlB(fX{r1HvQ-V`Hv$99i
z4ZU~4sfJUWT2~Tpu9cnQ)K|NIj8|APyJ@C5m3foA-2w@kK8zhSXx!HD|9v(zCA_`$
z3w_x-i5l9GeR2b-FQajLa=GVs{8b!B`{Pj>o+uA$*frs8#3ptHY)I4=53-#i63GSU
ze1eROsqpqm5V!4Qhg_CKU4^#>bm6Hi399tYsjEAk>N1}L-Zp@@1K?$Pd8fvZwJ;K|
z(xt7`XwjzSpR@P-3w>4}=3jG?2QihIRakxg8@c|8?B4nRysg0FJHlJ9^mrx1+vn7i
zi#pM=c%k+dp=ON698?)kQ(Nl0F!xylaw}?)<w*bbH8Vf|{*0C2ZH~{p_6_D(nOS{%
zj69~>9PsvUczYS%`oY`d@U|JeUC_Xxf|=}8^@%2jAe!BqKh^a;KB8AYWv9p5m0EAZ
zFF*CYJNgyRQce6nKj=x`%Klwt-m?#!y!>VyZi9blKE9)D@U{~Eq+Mj=Kk7gYxAqU6
zC*yDxynVusaCe_@oplS>_k$67bt^)%_LB9G2#5aDf6Sz2nL(D&e|b!BYxf&=>U{gA
zTk!S^yv>bAcMRHdZv*P6St69|3|C8ddvP~C%Yk;i$ZJ=(Am%j5E}DAM#tgqrg(lim
zx<4AQlTER$+0E7nw&t|S7;IJU_g0Oi$9W#kzM;?gFS>9;W2>^~uqlXn)CS{h+IiEa
z=V;N{55w_EGs6?fXGiDeCNJ>}ymh0Oe5c)a<-&VBre`F3VaOD2$N2!y8+p~LAuVmv
z1e+@Fu_<AkUGKKo<(WO4JnC>2{9)H}i(O5U*oO{pccNvxz}EqAwsfU%9pmyTKPSe~
zoBn23XL{5vi&(Wft6k6c*>r5HP5=IFRdINmLQe3HChWXf8m1}Cua?<@_ZQY)tP!Tw
zA))#-FjOU|Aty`<(T{@v^GKQSJRMy~e{}$Js~&+0*um`2y8zv65J2z7U*os?D-+(@
zbNDNGnV+V>+w1Vw=Z3FdZuZsRlj+$~OTN<9SAS*r=s~8B_Bq+H?(QS&A3lof<)aN%
zd}W{RtKZ@6CHkIWeaKQ+#2lzUxq*Yd^?e!p>o$0+2FEpy78|{FdaJkm-+8MSKBW%}
zeN<xz^E$!&u9S~nXY*10|9I;KS8d)J`_^0e(YICCdh6g6cF+FiEhEpp=M}qbE}~_Z
zd8q((;ik{n6Pxi>`_jlY&OtVCeR{QU^Zgm}8|Y=$i^A`dgFb92>d5f6@N6<2y2ISx
zsWFd1%N|7s?`8i$0M8Y{F1z7hWj*aF?>6K-!P_wUsc!JLWJeD@EsQpVw^=q>v<TKd
z`eIT43l@!DZqb`Q7CpRd(X(GX6p6;|29F2B+o$dpZRu*!IPQ~#7JhxjqI-KSDtin*
z-?V7>bAIM|R)kwLxfh<~86GNi4jo#>LkAKpI^eXZ3BOyyeUAI^THLmf*9h&6FDsfp
zFua{XkGA(k{K>UE)p)swa?hgox(J`uR_2c0cxW2>d?qaa%x9@_8Q=7Kw6n=W9l|{H
zXOxGI!rNNA;Q9{w&T!W6u7{>yfaNFn-BCXOce3N{sp<f8SH2~?7a#4b+3ET%I$iY#
zvIp*$bOjGg*Hd`wU)H7O3)tt4KAj%zQf#hF1yOH4*bAP*+j;QT%|BCh;Mj{kncCEw
zT;b7~vclaWP0_#b*0KjyuW-q2giA%>ZGMYO?V4mNh3ts_RWem!0-70bl7+f--<<F?
z9~`CE8()E)JTNdvb>?|F_HexrJ>oef`8_)HD{O4OKT}m!qG{=^KF)&H_IGK_GM5Gn
z<#iXjlm*_lVNQ1cBbUCw_$|Co%MHBNM!Zny?zAF!r5k4HaIJop4v#!C)p7`*t%6Jc
zCT6HKb>u;n(>3%l`44|5vDZ0?^AlKep3EkgSPR}JHc4i7D@m{KC24DaW=Pk<w0eme
zLH)M@ygdzXd!C@y`Z9ss<77oLb6W0llB!%v(qm(icERx6JJ?UlJWo?JZk3np!(%2j
z0eyRt+VIcM>~V{wh8*nF$$d_hpX}81)=q`^#4F?pd4iY7&A$^*US2%=ahNSRht_rB
zCu+tlCmOei7q1U*Kjufv7E06&xLcqi`&i*@%@Nd#`{BicwJSL`!rP-m6ZL6qqMlM`
zUN9TpuA=waf%_~>R3r3m-#y&MJnynY)SKaL>wpC1KTEDLyq&%{R;~8@uSeSPU91*n
z!QcA?{rG?^q#5MnFOAW<+cEk`{ddGB=31YUBLQzG!P_yp;#CRWn&IuyNVG7_y~}6*
z#e!}h>{Ls5J0LZl9hlUKH<PvaCp%y}vrj!YnG`TLKa359zcDcPBG*U4+vWJ529jyf
ziMb!U4|QaCdja0wT^_3@>{`Fg9+<{&9r}bXX%^abSXqZ!US;l-%%lqw*{5C|k1k%M
zEK|t=9*-s^8@?ipoLcU?>frfl6G{(r6|7mxEGfJ#4{ryahkevr58=0c@-tG8lhA<;
z>8a-UrekFL9-i?-7vXJAc<WDwf0IzMkMI-)283%LysfhjEr-U;i~ncwl1Mxk<T;R)
zTXq&~{hLgE_SHRpP3|vQLNhzl|J)e)f4qI)n!LkHWC#92f74}`KEkZ6b}jC0*GE5_
z0^izH5{=ps-bSyqDc5+LM#0WxI9sHuP45bm0T^eM8@wI<#!6409WwAXcR#DPb+&3c
zy#1QpM)km^EaX76_+nFJ6W(*Ha80G}x@}67dbKA1f66!Q#m97R8GFv>L@56e@(l;u
zwXu&~IYVq(NS5!?9n^Ap&8lP;KG<y2yq(m2$(T5UrrnBueH~_3QF0Govmd?a6T8gM
z?W*$0u3TgdFM!G8i-xNJynXJ$c?A9Gk=yJlPrZ11Rt^K%5l_N2`%swLC`?(s$l`Al
zs=e%)o0KJ#6*Hl_mN!g$>Dl%^PZmW1W?0{ZXw!rcMZsI&{lO~vJV;yN?X`A6sumTf
zEAaLTyq#PzK(#hdLx#8c7WmWi_tR(mQh%27Q-iF2s<hmfew~ljn0z$xsgFjy_tD%1
zKKge8tcA1QQ{n7*AN|taN7c`Jt9Bwi&mmr_wArG}8F1|;nyi(l0^j4qi6wu*$?QoS
zJ<u3*R~U1hK3uM5#`B%096Zi`9Y3(Um);ij*5V7^oJ)FZ@K+f9!CQHrdn@jUw~nmw
zR^uYxTIc7*IhdDD!`mtJTf--kU2KN6E7^aSMh`NT9d|j&S<J&;x(M<Qzj<mHue+S<
zT~j=j!_3SJ$4l;0q!qqpcw1${SN(gCy#JkF_1uSkFM2p}0JY{L%=>irB%^@$45M6g
z;1yn_tT7g{$1GYq(W3WxEm~Z|qJ8MvjAa%D6t(EbFDAWzV^nlyllJvADbHk+8Z9<4
zr)Xh@&Z3{U(8KU{A-tV+#G*;?b`ZRchou?O7WL|nH+Y(dew_&i>Gyu-dHc^t2Xk5E
z9IwZHEpZlVP57&R$4}*;Ph8kj>3E;UuOwrcx^vPBcuTE1X))&r@V3%xSn9^}!d>HI
zJk99U9o+Vk+x8mJ$b7crlRdPP$Bp^hLqR;p6&O1not+=v9^rZ+`i#rbW6$VChTKi%
z9F)10%W3-ceY*D1dyKf3iUvworOO$5#cb(F7`gI}i!2$JCcxW`C8!y9V3&F)*wrJG
zIgU)tTEV>P5@us6p<f%jG<XP1Eu5+B?k*LFw>!y%uyiNeAKlr>$S&BLnL1gH-xsAu
zorB$WH(~J&mlnguf=AJoTt-sU4b7R!d2c4?Z?HIx860Zc)7Q9k>YPg+)U6-y&BQN;
zW`(ym;qA;H8B#-+s=-?`=LuiXsl)fXIOlh11?K?c=4Gnb&vZ4dmaY}-k1csVO{K{s
zez7b~3;s$|Ab*SXL(^1jI9}pGY4X3DCd(ssoUdc|D!ko#2w%_HB(i>z)X59}!rMR7
z$wazAU$Y;5O?bPLnr*q<35xHMpzj+JbZRT!rR@nC#eB<ekqO!XdxP#JXx(D;pfyRa
zR<H-a#9mo<e7!I5N}Z)Ydz8KBH=Oc#fiAQ=)%+_gg}3u3IaRcoQv;1;@;;1L^cnJc
zZ^PU~cEt6jmV3&n!Wrnv#$*GoNYJN!2};RIj}_hq!`p-KwtIQ{wP@X^P0+A?nHLiL
z?MCheyd3~<-Nw+fg||!L?JIO{?l$~xE*coU+XUYJ0dGegWM|yr1lg#|E+u=R>Th_8
z$X$5wFMGRT+98fP=VP_$19`siHZQ!b^%whTx{xi{f}FvA^hOuQ==?k~^!eSF;22G9
zN}ZgVc^-H>8s3fzacaQ|>d6(*y!pBPJ9@q(yW(N>l&R!Wz}w^5$zf!MXLOS|eQX)0
z{cV}Et<OA8?Kmxkw}avByBKzpdc)c4vD&wnTJqdj<)$yYrEQGDt~oTjBU!>_$YeUl
z-trtjHOrm6duCW}qa*E4$wc}uYo5H`4N>rvEQO(Hwn5e5OQ|RYP!FB6BuXLpi5A?B
z)Z0mDQuO8`cspYjdB2?^HJUu6v$dFW`NBLYe%sgcBGh0^1iM%8*Jh2-$FJea;u)^q
z@OC`?$gNkx$$yGav(yOo^+sxWCHzZ&Md>BHeFJYNFxQg~-|C@zx5L}iCXv+6BUJc)
zICVpGExfH~vg`3kn3ct@r#|#<U)kh?S1NeBO^48@fx~S29p0{kw>e;6u_8A0by}%I
zTD9nzRm<URnTgEI^sy>|?7_JWtkieWyxjkf5jJv3Y;sM3(I;$VMA(&QWH@=v;Tkl`
zraW)0TEcS#vX^G2!Ky*byFN&<s_cDO4R2E?z;t@3@ju8yLR04;OZYAx;RdDc@`KGg
zU)$CFiyi*hl^fn(hPSgm*p<etY+oO{!kgIf2wQcA{%b#Lm=*{9mlHw$Ve2r}E*++@
zZ=s6z;dl_Lp94bW*C13`;O)4u5S8f=qLa+oczqAn2}`gl;t6i@Z?K9T57xZR!CEye
zSW9jNksTbQe0Y^!!rOmN1nO8X=5HqiC_DbBY8igYLC#XA33!wS_-PaKsJHPX-z$w*
zD#b_R{eATGzK_zD`Dp4aAN@YUhb(E>+uTQwa{H*_X?DTR_13pYFO3}t2Ul8j<SneJ
z!%XTm59KrC(WQ6T1#KAycT2&&oA9=ZKe@oS=ou~`k8n46M)2;-4UWZM^(+0;64Z|?
zTbS8_w-w;+H+VZ^zBj!XZ%qy)``67&Tj1^fQC@0@rn@$XZ2tkwo{lHWVVS4KZed;&
zR-S;d<-W1g&PwKRD7@$T5_kMSaQ^HyPsML!FTfK#Pd&aW`+oA6ma{+pDSPzL#P9l2
zb3Tms7;lj=g?S$8wl`biFRl-_U}9ynMRx~V*rUfeMo9~^XYg+^z1!jzm6&DHwtGf>
zhPQLzZR#Ww*?lGr&u-EBMHU6Z+94OHA;VDL4i??6Vo|X`v-()fs(9VZxtUp|?phRs
z#!WkH(e?=z4g15QDzN(IZj16h!jBb;=EW~{A0JkS7!M7Ew|&}?OK}fPdI-(Ck{)w+
zdd!nN)D(73dVw#CT5`8n9?ajvPCif1)%^R6Xa092edHk?8ax4yG`t-RZ!hqet}yn{
zNxWXiJ@_7n#lORh?8zG6h&h&P<j1#xGiLJdJ;(`a!q2zK@@<l??58tyAwP_Sw}bw5
zF}LMXG;AzR5BEh2{84S-WW!8No5X&cHyIjpBtr-Cx|E}iOW(V(6LzLcj}A~zhI!%C
zlHIxOCmh_i&84o_$zt+#X>P1bSG;-5Y!`b}TxtwwTc31kA&*-<(4}U5c%0zp&J4B9
zm7%V0(}l@GPkho<vsQ+dbjZ-I=IQFx7@u&%bOl_?P!)Vmk%Q3G@HTC&OI6uL>w>q<
zzq{n`=hENhGxVWBx{k!9=^slP=QU}ZkEdxSybT`9{0=j+7bdg!ym^W?8Sn_*CMSYg
zajgY#^+1vqpGZ<)>dGtW*N(VFeR&TvsXN(eUWYm}9Bn%(L5}QjE+;%K!LGPp67;4@
zf?m*r-7fmDLlV?)DITar{B8!hef(UpDZXQGCpotGaABtJVW)0gaVqY#Q<0aPdUl!o
zp*>FBK*R2;gonsLmco593=fkZcr9M@o#YzzrccUa7s1<Y@U{%Rb=#Yu-EgtbL3*nP
zc&$ggkJ9*`(6~>k!_rpF>$d0T&eWWHCu-pk@-NW3{owF2j(*GVP{HT2oylT&iBHLa
z-}fyU{>R8;>dQ`9_QUy&re=FEM#s`)$j{~Sj~Mo7GEZ6%Jv%T)lZM1-|CAVgfUC9P
zZI>o7WE;e2WjFjr)N*%BjX_t$Xb2vp+8<(+0Sil(VGmA0=4pz@s$el@W0S}ujH2e7
zjeMn|v07b$9pC+8*~f*)3IA{?_2Tjz=h3lGjpPU8i`xDuRu*c;3z`3^%z44BE3x{>
z9Be~ZjNW`DPZ-|5WEb55c9^&5Nd8084^2PM+zr`D)gO?HH0-<D&i<y#Wc*%DWPa5z
zO0nbx?wLp?1H7FKZ@0V9mH2DllZjM%ExurQ`wR8cgwc`e3U4>U+ad7wGQ15ZlQ8gT
zgo?o1fC2buZSaxo``7TcrXgHw@6f-*rxcH0srA4JZTKxxMtY@jZz8pO99e}N*?Dct
zv?wLPTf?tWnz0qe_lZ;&dY3!#3$5*MXP&~YYv{?9WCF*zm^n4t^c&vfHSl%`ysbOd
zCTn$@CX^;KAJ6XM-1K$fqxU&#vS+NSPyh4J<yQTJuASe5y)rGV>PoI}XaRD4qr%kO
zCrqV2hAH7fm_8rCbHvZX%GtEMrcJhwR@q;}QJ(L5dpKP-R0gsZ9^|sJ|G}zZbF69r
zZ=e58U$`%NI?bxlTi}0yP2rX7iuuE?H!tiO19R8YS8i0)roH}Ft#r5Q`(t{sSFD`p
zS+#g0?=6Yk!Z5Tfyv>nJwlMwKKzg$8;B7p-^|}|TQt-AXy!FO=JQ&^vkOtw#PS+H8
z`wrg5ya`s_Yh*{{3D(@OAT{_L#4I;G+l4{83~zt<2FkHDP@mTYipeOM3;OHG1wTDp
z%)IM(`l-MB=^j3(@>6})ryTuTcpKp7qe|<1^sugvUKa4t*cfl!_wgnV##@W1+1|wa
z>%E=KrJ2-e2cU6Rpx-{4RP>NZ#w8}ry<yVw>Uegq;t`^zI=m76$30{RPxj>hA?8Hq
zd+PKGn7G+fbvL0osrSydp+`f%YGnoX1XRLfT#MPB055IL<E=TLywqtkzNiN5?z#6>
zOW<f6wd)V5=&>|>w@L7U%N6)+Kh&mw+0j!~C$eAG8*Lll$y~OlYP}&Vf+IV;{RcJ=
zhqvLq>GSSjuig?g?W3=Hd>Wk#Z_6C;Q~<m+z}vIbRqr$-|F|A?-hVB+2m?3Qwy;;w
zELSm$65pHD{jN#w+f3>>&7`W&3_5b#sB;-c-Rfvkty?Bt3o@&7JWO?&Wz1ty`AQZ|
z2{tS2oLM&xn^l>^eA=u6@U~MvixP&=qit?c%i$K)hqux2Idg|aCm+D}NQ*l8;>*gg
zC=Jinfg<Qq^zNvd9L+scx-*)!D|5WC)QrCEevjiNy<v{T+tiHrvabl<&Vjezxli?h
z9%?oU9Sm>R!rQa(whoL9K--Riw*$C-`a`0gkv*{WK6||<vO~3ND*aS4i!S4zosgoF
z|E9`>clqQO_MmfY^h(ns@__$}a49ODUToeBO=P#+mP+YbxhGwV-eu^=s|+pMpP@;4
zTsmFTrOL3i-=8k>xLi6;KlNu@moC<I=?=Pg<tFX}Z#!KkZ@4>sT|8O)?`Qm<o-Xk?
zL!$*(o3Q(}D&DOM8CqE|LoYsZPT);8)ALjXolaHqs#KjV#NHit?+jtiwJp5;X95}i
z-_z83J$3PG>B<jp4{+@I2X3!J8{_%93~&3u+d1%dPehvLJ!03+`ZSe-wPR+a=@h*E
zmMcxCN2KUSJ!V7wk`!@{`IkA=f9Em3vX$AD-AO6{V|z~~hp8?5cN)^yEJfx7yzTks
ze_pA~OeY?4dYQ+Z(t7%ltDV}jh@ImzoH{;~8gWOb7LH<{_;mWLgPALZxBcL4&3jJG
zUggxPO>py~Q!XAK3~x`sTc6$RX6p%eOL4fxYx=Et{kAxsoZxuf-xaTo8Pu2$ICVP7
zDSJbD$4l^Qtz>ox&Q^lAO}Q+wpVxbrppp>@`nzJH3@sCNqAfjIc)JkZUL3?cE?Rdb
z99~~5QP1i38dBK5(}7;{&3FaanJs<K{x^8rq<@@-XCc%7WQ@E6=y`r{$o|Tq+%Fv3
zkmOLlL}p)pGIQ#3Xm_U=Wt|wKi=D|uY8<0O@b*u5+X3FzhPR3PV$=)XroN9+IK0gb
zZv)`%&OG>sVeFwe^k_tk3c<?gBIxLfvC7&dR!tLQb&*>38d!S;)&`(s89<T=-tL08
zpW*EqyvWCwGVjwiRvp{nCFVK$Uf{fdUTr`8O|9VV*iP&`&Hht0&#^NV-r8?`SHKr`
zw=wfFtvA_7aZzeZz4+|GD0PK<i~eIBysZFlx5L|ISJ_#*89q*kq+U#R1INS+G8f1q
z%wC<DP<%|G>EQ~(^E)iVE(b?%>Y|IH$hgOUTLRuLg11}AymubrXSCv}Y|Ohb=W>2V
zgm$w(eE-!5T_VR}!Vxs_aV`f(>drbcm*DEV2Gow>?Xn7X_T|{AxlpT(X7&e<)Ma=2
zuFuKehqq(k?ceyMs>9owrEH3#j_i}m#$2vdCV0CN-abcz*4S;8EeD>XhE}a=3MXq@
z<xai!+vhOtdQ9f;^)Q`fx7&Gmn`LL1qNkEw)Q{V0k`=fmL|vu_Yd5{leA@zamJDA5
z*+&IW2I|`;GMEyB)u|=<PVlx78u<GL`mKRBWmZJfPPVfX$u1X~_FfOW>ciVz@OG>b
zeY_slCXja+8m5|Vyxt=+4aqYu2XBXwo#f7YbFUnxf8ecwI&#6wp?cImRP*YQ9hE&)
zjr>C-vKohzXYp$>vYvjxTC}ZStzb>cAg{t2q#E${pA$h^ho;>-f|=KDfg1fZKqqzu
zD7q5ah=u%>{i2_?Eb>#9aegY;$4{vVzRLdDSGQjIs>@3<4(IsjXKNoFhqs?nz4hTS
zz1V7K&C1^D@`T(6H0d>JuuIA5b!{bI7~cLf%A}SHO}YnfyIwVE-Zhg>UNPy|Wf-)_
zq@l3*)DW|l%rvX`AyPSSn{~m*tRo-Lib0+l<B3n0{_5S|sP!_p^LCjRa|2$w@euFv
z1NOqL_fjIf?N$A&D#F`tmzaTZdMGfKx+%vi*l2>cZ{h8fwrJ~-<Q{r@(ywMu&vka_
zaQ}iZHs57WB`-mDzxC9;lI+pz#+*=pUK<U(dKJu{MP^fH@{u#pe^zR+^gidp+spt9
zIo1{(gtytys}r)DRSu0C2ydI*Ht9UPy)@dS><<jObH%9B@b*!Ald{0u1W&W>$C@=A
z-df;oU`2Yp{$`bju><*A<ZWZty<g4r4a_pEGP6_AtZ(%#It+KcMp^XNJnG5t)^Nh2
ziJZe!gSS?A`v%^Afw!aJZC7|Z3*J_QxAWj_$>E%f?8H|EZ++3YZ#nGnHrsX&^=XU$
zxHW$-(Z~JWLw|AKuZv(VyuHO^79F97dkpS!%;oxD_wk}OOw`$b5>%5O-^kxm^}SE3
zW}iz@;cVpYG52zGB)lc>x7}j)#&PtT&E-t8_NS&PPhgs&E~aVZR`LmNrt4=?I<xc{
z%Gp0d)8Xw7c$@uuhRV#wcl9DeyQ4Cch~K$Bx;F=!wFdn2KJSw8EF6TlH<)8>csD~g
zH)Uw<^b8H&kPhqqmxs9TdYV#roeOQ#RK_P&6Q89jgBe#7yj?IZRnC><OK?5CQmW<*
zCGU{)gpGD~+2u>e8<{RQYTE_=dp&gSjl?u<?U^dqaWWI}1(*B4E_M7%<Nrxf3$ZWn
z&om8$w^{qADd-Nn(zhpTGIiu?)RC8rp(c=+s66mC5A3Z!Hwi6F-V?s2kxi1Qk0&YW
zEV)Z)(ArJ#<t92c*W%QAH~Ntu=|8?DpP{{zJvB}pq#nGxJH1JG>pnF>2hqL96Q~2<
zccR0n1uu81?qa9P9K<sOpDWV8EL`6yb48~f6m#MUj#s|V@u~>_28@VT&P@D4^g-9G
z#2*BI|JyY)u^#`f@HS#8vpRpGw?`&O#qbfMwLin#wLuB^n&B(FT>@_n9qH4~bSgPM
zo*m@zn!V+JnS!wk*m1KgUXktS9mhs%G#;W`6&zX*qXunYp7e}EjW0Un!k=4&>$}f6
z^zpevAH3PE>VuC64z~$nK4xf)e8w_c18<Y_(>Jv_$cc8Sf|r9HdW;@&|COH|y2kTV
zGI4)D2U%0>LQi(+XmvENot(nZ7+KKJZ%f7M*Z_L7=vH%FtcJkZzYOTvEiiXQESYk#
zS~Zc}!~C)8&;75zic#T1F!n)=inr&oDw$8&Vzev5q2+rV`m+_8668&c>G2a!7(3_M
z{Xi#@Q%F8TpC#-<g}1rLkk|%qyBxr4JC@#Q%_!BX5T$pyqf{{feakMJf)gWU?M;sU
zcYH?=ZSs6!W1hyQksIwg#IbyxUFUMx6+xeJSwXx@GpYG<yEiq{1=LR$S7+8`0UqH4
z^hDW9GjI<%4!fx_v)}DJ``vPKnQtTWI?ViRSslUov0b})zS|ssT(@cN5}RJY%0}yK
z>{7C^bJ->nyp4TI-Iycj2KC<+HZ5&o(-X99&yu`Hc<V+@caT?@mPCi@Kzt~B8^YwX
zEKH}~p<SsX{|0Z}s$1nUh3Nwv>uU+(Jcg`>jY0aJH%PZO2CCKEKs6%|!pBY)?_+-*
zdgaf*t-m(I+Xe7;Ri8lFRs`x1HQQl1Lv)`!iNa{*5>KtV*2Jda7j3Fl-Y&J|++n(1
zJNUb-Y-ZQ!9n_k=tn%><W6ms0-(h@X`medkJWe-;YCKuO$$nw#kA_|fZzqvOT*Vry
z`}Af<Zp3p;o^h9a<QvnYEiw=9F?q*tn63R*AOzhRtm^Q#(_FOf*kIl58!RbU3FJ(4
zg|{~*`YZo-KfONXrxT~yi*wvhV^V$9ueYDBPx8~<L4F!au2U&^J8`YA;wSp5>QNs>
zukcabo<7V3`{-H*ecDIVcqh|a?TtSe-VTJf)8Xwz>cOE4$x6yFDR_ZF)fbrb?;@kB
zz?-eqRqK>AY2|X0Ja?N^8s27}HR<gUlNPLk%i%^cT}-kLGwSch7Uf`;W~jdfe*yZe
z1-h#}e%ikO^VL=j_t1?do;nCer*uQJj)RpwnP<61*3ugfS-#>Mj)xs>U<z8gS5I;h
z=TIlz?x|1Y9>$<)7o1}jbshWMsXe>Rp#}?Y6XEUOjo6`+>#HV3v0rwjr$!GZ_n|s_
z2-4Ab@U|tqy$Nq8@ZKKox8T1p>oANwpUtc_J4~v%$EXe!jJjzxXo9Cf8~-(EXMKYT
zH#X>UPlL8KH)!)wqh8vLvc?(}l3<j76_bj$q~5&7to+N+zAZQpsbp67hGx7q93f@}
z6~)I4Z{Nb(adYr!ZDZ!>xJ93+Nzb2b(cXS=81~OtX_50ZK4|LDV@($23A3m>9DOo`
z%!<+Yoriho-B5f`13cu;^?(udYhi7AQ~urfTb*ggZorQJ^H_aB_ikIy?!aBlEWz3@
zXxrxScJD2+`zqj5rQYZ~m!M{olGrbq1pkv{d!D3?bMX6&!S|C%Pm<nbt8t0ibb)@~
z`ebIvQ{_ble>F0JKWt%N&5d+Lz}qx<+qhMR{HXc9a;B?&M7kbM%Fqfp+7sTk%Hq<!
zisV@J#V?F^Y9=*h7kP;5+R?N1%20vsJT5g|8+pEWYtv<!K^`Kzb-dy2ijCxRQ1hOM
zzFqs29EtPnp<}Pk^|q<9WlfcvBSpL6Z5iI%TzFeFI88%&Y+~7T-B`xH^JDDVIi0Fg
zUGV>QOi^7}Tk}=2O!$I(wMkSjbljr%i7F95&O&xP#%NUwx;3O{l6L-;q<Qqa_Y6%U
zPl|kp)y%+jPo$5Kpf8)470rkCjA2KNJzh0t$Lr>Vc=_In*9aKd1m13ew<#asCA|F|
z-mZkVtKsc7c)Jh=zUrEwVZSD*I=rndr}Fl4YV0z8KgOxV@lHiecPe45Qw88{v0vCj
z2XA-5+u`e&1scHqnaDWKU*c4noWh~>EIr24-@N8z_99+W{(C<h4=2TI+?aUvb2(KD
z-k$tDL4SI&pZ!Pz{dY2edHjwO<RLvJ%YS|>_0(8AwbY7vt_hRlwH1c{Xv@yK8PW1P
z94&KhhYGM;W_uThO5rs!%y*~<jCDKdP&s%TaMK~%J%>)f&dy&PI!5l`E!aDn{QT!m
zhkE1Z%?58R+}4Yo=i_g|Xjnb+wL^{GI+Sd5r~|z14{wjd+dtv0FT9-wZ%@F_&E;Y>
zj10uuL(#4__==|O2XC9g+l%nlo%i39-|w)+;1h~b<T++%sdv8`9HaNGVl)WeuHb!q
z-R{uio#gpfb?AFD&L!5fH+>+xUD<oH2j0$xx8rx=PlC6;@lmS$CQ9vRM#;Y?9w~U+
z4c@+rWk=klNL_tRrqNZjtJ$uoD4VYTYttTB=SSak61Ci39J9IpH$BplKGam3*|ZVf
z)?AO)eP!3fSi6qY4p)cy%#x0wrd$@UF23BlPI{WBm~~lc)%z({jrhx|QSf%_7V5me
zk|TIANMCjYEAC~m-UJ4-UyXcx^rJ0%2>C1_3U3jrx9H8vPsl5r!FyR`lYcg9w`ki5
z)RI#&sr$m)se1yImt*3>5H%sg*DwvO`aDdXsTr?q<*yBuebt6J(r&S2KbU#lYeDiI
z5~QN`KpkHj_<#9^d0P0Z;!=MV-Qln0cm37rgFo2|0V=<UJf@@o4S=_sw*;#Nysa@9
zkJ8Ct6(13-@PuIbkm=ukC_8oFttL`e?rE2g6+I4btC+3yKdtIijy<vG$P|XR7WVE$
z8aP5j)%|;@3ZrRzM25<Z{c=aR?Z#Plx{?z-1m2c}x26{`o>`wbGK|~6+W>f5F^(+7
z`@!=1J6KQu2v#fJ%di&Il(PjXG$~Mlz5SK#i?1@K`Ks~-U+w(ls}0Q9cBY1$Yoecy
zb@EeGD?c^-jm$>;$H#w?B|+WT;O@<Ry|uKokEY=pZa>~ju9^6OrsDJY3qPy}Jy?HF
z&K<~CDuA9!G6}=0O#Ka74R4b#8@2YPQOBvPu7tal9vf9A+NgvxCOw9?j+G`Igts}n
z8TAR??jLT{^>;=sdTP|OPe!>NMk}L>FFrQu)`$PDoAo)=EWJ1D<TJD8?KSJiCbLH8
zH>n5SpV@e7AI+qO3UA+*B^R+ctgS{5bepH#sU?4gw?ohHzTs`hexA%+cuKH)OiePF
z=zk6^=&1vnJmoWrzH%cpa28KxrowuByVc<BkipFNgyL<&*YwLxvs!KC{Y09St&2&u
zcN(>+96eiqgKQQ9Ig|!9t!2>kMg}=L8Dwo{(3@5UReNYuo;aht;jK5kEs|zbX%yzN
z6DBne;kojgRiVCFzQ39EV2oLd=9%?ZxLHja<7uX*e0ZHj?f&^+ulU+XSPO5bjkD;*
zLi)Y%b_Trt;YJ-g8Qlv*|A4oHhQm`X$HLq4ebDPI(Yk2ce7`WWREob*ZR+IQw{8c{
zD@OBg`KO0|!)tYD34Lo=JBuUcFq-&ZxZ;_h1D(kiIEL<lHBsx>BZKBS<VN3*j0M|G
zwD08v>NPMKCbhhSe+EtW%!it3BX+Gahk9#6nx5oJS7w29RsWo><;^qn2;Ta_+e!3S
zf4)gq(R>;DwO59oMrUYyS(n1lvyb5IAMo}TyzL2Z55{KbICDso;O(yC=?a6l@1`@K
zghm~9Eln5UZU1pFH(#1&CzClGn5v(%=sm;Rm+*Ewyv<bxotr&X3nNqXHXubEe3+5F
z$WHUqDfHY^RC;ELykTvHcI?wR%{=Syc$MM1dWApV(Sh(~Jh`{jcE5a0&~s|7jWZMU
z7@yH?cw21-S%~+@7=*W3skNS8geEmQRlPX9*0S;XnKfQ9WFBm3O4iU2cCW$PGxRz~
zc#%<0Z`E@;+=F>p0yyeYTW*XWss;QU?Nq6gPG%ZlSu{0bc>5dm;<0m_TE_3L!`sR5
zHg%L!f8rI|b)47Q9<O(+=+%y9Hl_$!`5x@PnS>|zL9BX|VHe#+GM1XhtL(gZweTXZ
zcQyUanX#(ZBvvPWMbE*<>^;aLUY|fFJKThce}}~=;U;}g?pJ3n8N%>tGQ52cZ#&nC
zXMdVgSz5zMW=Tu0iPopA%%!4l*LC?HZ<mod_>LLX(eSp=IXDJympyXm{$qz)!_$i}
zbSl?xqg$s~cs;8_hdA=1Z9P9Y$a->UFuc8dkjMY)&>nc34c?xPz&o76>%!YhL#Y2A
zj?r8g+A1wZvzn3hG>}}zL-c0Jnm9ZaA97prET}b`xX+z8F)DJNe)6&y<|N?$YKI<f
zh*sTY?4Iiutrq2?H5uOaVy^c3WVkskLhC0pud*saIdgy0*CF54*!`P4$Qr&1Z)2z@
zAHNr+=OSl`I`X}5k@Sxv)UbWH4z*xLr8%Br=0p2=+w~W`{ih-|*%DTLEX9tuI#zA&
zVpZGIT+)Y3K(Fm2d;Z0sU@g7CUbMO78|@8MvE89cV+M3sF}tGS?Vxm+^OGIox9~r~
zpCj=0CTz>KD^SfZpeL&ZX+6CCv@KYz?*)@B60GK4Wb=juvnwS;_u%bPc$>H~MCr@|
z6@<6d;O$X-N`*_K8DEh7JAs<-5O3YB_EmA8$cDABV+B8IYcxv->XIe8u<Mih+St!u
zG$yZ`5>^>hdki{k1mBs(z16CWuhvf@BXE<ik~drQ3eV`zHU9ME{FMjZzJs@U;q6WG
zm?|s@R>aO=^~j{Ad?Hw7S_kV)zhLE$3f7%Vp)ykgZ^JoC^9W{oO4v31sEzr2{+5TR
zYwy7;9nB6p_S5w!#Lo0Lp~~|J4^(idPQlu29Kct1;q5MXyWn)FJbH84JXDqPhN{h9
zc#nSx(Vc$mB7nCK;cX{)d*BMY1iA-nGd1OgUD->}fPFde_6xji)ZJfgs1wg@&AjXv
zW^CU0Y686N1aEKSmAcj5PxIhyaeiOD5Lp%<eB|%RtSh`N`oLS&OQR`w;nf*v)aJHE
zIXfAZr5!!mP9_b^Y1Zt5X8BDuYZ$ye4}TK-8dMYBHix&<;q5$lYrA9A&bvn4fit80
zjLe@J<q9z>s;*IQx^h{^UG+Y?u`9)0WtzHcPE~gqxPR%FMx8eq^`aopJI|zI4^8??
zkGSe*lUjU*?OV*+Uc#ipM~v#-iOj)%^e3xXbf&CXn~Sm^pb{Dvop>A{)9xpDjFa*G
zX26ryFt;83**0idc>Af4Cs`t%Iv0j6_Jh?N1rpdnU?KbA4tw@s`KCtXKEPX_ujueU
z%xZxyZN%rjQ`$u42ff<jM)`OdG-R|vp~Lta*xWTP&|Mzo4eDRRpfzO-Dqhf_3dIfj
z$nQP)n^a9Us&*Tbror1r!6yBd&#d<K%;aC1HI$lrt*6wQU!!>&S#)&*eymw=b*)7`
z@Ll~r41L=g@7Ax(6*aZ!*inn>F>AELU{POq8(0!{!qKkq_6z3<6{$5xax|-l&$=#o
z8s+g@!Q0eYXxvtCnfo?|x830FJN(hZmw4z6ob7ysIy0BKU}*~9xsT@2C!}_{b|5v%
z)qJN?hb(!YeclfeWP-K7TuhMXbMlSq;-R@g{?K$->z$+%4cY(lDOo>@q^kE$>dQ6K
zwfcU#hBSn!1v0eBp02%K>1qaV)AGPlczeY}ZMlR?HJiA|4|ZvE26bjQJN{^fns3aI
z1Ky6vOjr3e>FVAoUH$P6zfUIn>TsG?;b+=1hTbZ?eF1MT2c;?n-fn9{9XU&?_J+}S
z{y`RTLW*X>+dJ?!_nj0Sgtu*WrQpYbxhs=Z{BWXtDmrCs$$ZE<ytAIMcnQfGg11wc
zUnwvT-|A6rD@h-<1m2@8PQ{L)zWNwH4!&F0-B>m22bcYws`8qw{DMxMO@wc7&g(Th
z(yqs{f0SKrusiZayw2TcM(04h{6@s->tE>5A84{0%&_>OL-V3LTgB_?MCM@b!Q)@?
zBDZ(yYghK7j)1R2(89crR@0msjE-$l3f&5~H!mhrYGu4UhQ_N7y#3Qiw$d=>U}nXt
zVs73aoS8n2{DvD2z1|4(;N+?AWG>jF)v&rl6FGK1i&le@=+K%Dec|V0`5ol`M(dB5
zXj$W;mDJ3kZimSCg+~*pv)-Fd-r&BUs`qEKa%4nnIlP@75v`n$qIDrPTEBO9sLO7L
znxAm!ApOv07abaLAK&johkVezUtcn7^T45J_Z+Iq&zJnk*#Crvedf@6_}cXnzuUw<
z*16QFm!g$nbSbp&AM{tf;q54R+h%BtjQiLl$GJgCcsr^_tVUB)4%&q#UlNOdh}!Xg
z|E@*Z3kz?DJYm-jb2q)1yLm@$;cc$x?(st#$k-pwzM9<iqIA7mq|A5ejUI@Q4bAwf
z9GQU3cXlLmu!Y+<{IKka+xbmx;q6hfgO?hjGyvXSqG#LA7^&>=wi>)OHKo@|uk_HL
zR<*&)Td|2%#tK&TXC`zDb1I8mVR}<KRKu4Bsqhqz?tHIT4pJe$=NrM>lJK?}ydBO?
zw_6iKHTp}4VmF1T+~Z*8f`im+0@?IE$@zn|jTZ%K?oRaN-axHc;;-AK{B<lBIYZ2X
zey;D2=Z)FW%H$ta4%CWiL3&X!M30$uS$qKfI6I8JDq;E+*4CUCtgnyR8-}JD^RJt_
zP5Y$r!#`?Xosasv)kj&YeNx*epOiZHi#n|SqDBY4C^?6luB|jE54?Re+$?K9v(7d$
zvmY5vl4#J0MP|*oXkiaK@8Jp@f5%ZSK+o$3FrOcw<mG<yYU!^@W`BL_5u}7x!K%sZ
z&8R0q%<bd1I%U&O>Ht%|u`eCouGwo-SY?|!;lXmzGd3_UoBw&JR>Rq<uh>8LFjR~8
z<AZvQmd_HREa?0Y`o=4}kXKPGSg%V5D{rA--R=>rJ<URt8jp@Hg}%NLpxC_u8oMTd
zy)gbVxs%EG)?aS>{dB&uue{6ps$CIZEkyIh^L^TIt3kiKch^7D@ezAj)QUb}McB|e
zi$MXoVMIBD`cd~B(!`*UpYFO4<gOEA(P*=cnllz%+SjO)ZH#K&(WosWjXDftat9f;
zI*(bG;ca7h+b^7>n?Yd<(X;Tj54>GNUAZF10{D3o-gfdcYP640KVWSaj_&o0YE#Qy
z&JS){_rgt6-n(h&ukO0d{qM9hszp_!wwn#w!t<qmF)F}lR8tG@p^QlxYmFMep2r_D
zYJ-d#*3zg7Q;m9Bf?nt#vrLc8a^gw)P=X#Py!A?^KM8LyR3)Fkyr<Ts;pdH~4h>JI
zMbNVhVb0VSzppo$$FX>?4CF9gCkJ8>EP`>5=)caSx4K|7zNRH+J+k4)re408zt53k
zd=?Lb;#(NxS;(NzA?`}^ch|ZC2GxbPad{0|>U39XF@tUmFlg&(Zcj2Qu9Zo@4#Z~$
zZ`1RlZEK@#+nA}9nDzE5KB|Xiy^iF7xt+&Y$k3#&%zLRa$f8XRsWsq9I|Xn3d*gGa
z{xEmv|K@Y|6sNBZOXnA)7L32@?^JxwRXh|6Yai9YpA2imi{Ot!-`ZPxD7ZN~)`<7(
z7M^G{e90(!#D~bRg0)*YE?r|b1>VN-UA}M>y~Td)D}lSIH}Tp~C%tfs%!W$|T5_H|
zAoOh!M}m%qC+gDFB%On|-I^t<!`ozeze6{kO4Xp5X)1X*U4!an=u_?t6^FO^;qCJm
z>6(wuU5)N7Ys}Cnyvwa>y0ih_b|>pU1m5<9x7pWasKw+AjZIIduE7qwcKD1dq-$AR
zn%>-^PrNrxY4G;U1@`J3Ow}rSw%KN;Y7x8*hPVEq>|BSl|1q}FpcEO)pk?PKYx_86
zYaWs*=u6L%^9D1YXFPg-Hs8zpa>lApD&8IR(2J5}>sE->vfuEt_J*5-V)09{zxx6?
zx-j&QN3j}r1^+8QcRLZQN#3#4e5qv)!=Hn;y7x3zzr*RDSINRpVAsnK=1}Uvg95Rd
znk81{63}Q-4uwV1*US^EF{8+as1v7~i^#se#*WtTIKKDjmwu!c-3HwYZ?6q<>hGfH
z;HggahPV0R<Nj|)`VBloQEo8`xgVo}M`QGLSB%a!ja5n5(=i8H58l?A6QgZIVd}1*
zYC8{39)y$d_E0&8R{X;IIL~!*c`vt#)=hSlmLyN!oqYKny?<!ewjZkh=!X^*|EWu4
z`PQC^j|kpAg16?{Fsw8C+Q=s?wm4dgu10Geyj|GEfnIi~EW9m>PEB2ac3nh&b}{wi
z<qqwA;m{WNdl}vqfVT_$9TL3#1>PDjIka#ezNNn%I)_faGKgL1dwASwhxVbp2U1U7
z4R1S);WNS8Fhh(Yeo|8|&kRz3ax3`vD$>p&>&BmIQTd0qGXMFS{J}S=>;NAcuKxIk
zPQk&)_-g|ylRH$NtRXb#zM@uTO{X8o?r;Y^$}HSB=YQTLYPgr^ujUM6XB)GpQ}FbB
z8E<1}&^O(8k5ZF`^j6{R6KcX2(W*CJSml4is)6jBx%*!hldRf`x2RbixG<R=ZFTUz
z-V0)GB1qpB2gyGl*|@GiHO(5Nj$MN_W>JvJ{YC$9b+AUw3snbryN6!mfT8p}dj_f_
zebHiN@a~jg*II)><$$;0TLM**+V2v0YboYW|IA<4vcs_iKP55qakr#5ob%T6!oIrE
z(NCRCWc|(a*B<K1f8uZY7u}uW5vHKx!DN>)_vz!VuqmHZAKuo1w<X}M0p3=Dw?pCW
zfUF<YrT$wbH+!zg1<&;J$y;ry@lnaeKI*pLM-_#)r4m2L<H-kQ+4WH;&VIy$_Ju4E
zgO*fA6Byyx8ftzw4eaYRD9Z|Zb8`&}=DXH^23n^)c}3s-bu`IeImuCM#Cb-6C7h?+
zvPwz(O)XZk_QNz2-Rk=+RAF#-!xQ$kGw(C;K&YDX_v!&}yP)&0PYqTb_<E>tu)3ZL
zlHVEj<-H1GFCe>MuLr6|Sv2*f0PSt)qwDX<Ej&x_@mKngUgQs&shQrTw|R=ZhW^a6
zG$kXq8JWE84f^A)yGHxDE7<Bz?xaC^hZwXY(_N)9n4TzWkQ;T+TJZJ*ye;MCu4ZVl
z?%eOnEqan!jhfbszAU^Q4{vRwjjG;&`fnk#nx&YuA`uU+o%(NngX&LnSN#n}`Q9*U
zIjp@3E6;HBJ8YC^AEN@#7?c5r4kfzbWpdM0Pd7Q=x#{9_H?@VgX=XQ>xcz*%yDTPy
zio)BD`#CpgYLo@0=9ytsjal$}nNbDdtzo+nf0R)TDjQYkhe1zye-HYYRd^G#HLJ~v
zJWic9!mO3lU%iLH*8qIR@OBBjT?cPl<3Ik-gKNG<M)DgEz4kyG$1qC+Zx3_ZfC=pC
z*<(?+IJgII-A0-<4gLA@qDlW0HK`)6$Ff7^hqo@ySyFNtv>?D;-QjH={vJy?r&$)|
zuD@d4RU^AWE#U2W{sz~=4SH18s1qGbsta#N=Qb<buV(tG_@;QzpDr*T1aEKi-Xq}c
zZ}7G;yj?zt42`}Pr8dAjg`V%-%pw=OtvbV^o}19YA<PCBV}B65T?lVC!rOT<=-&!t
zRq@|)d<}9W(f@;ATU2x|f0Oy>VQST1(a3L3TlDfBSsr-9yjG!uVQpKEm7Uq4GMn%E
zIrQbwR440@161Cr_XF_4&f@zX9#1|6qo}*~f688#pVS`VZ7hD=<6r4zHcpnyEt&kR
z6rDMls#i<#7g76lTqB?A3G+J7(zWwBy;v@LdZ(*M?F>~vMpjdLhO#Wh7quWm^EbfQ
zJ{c+)1DgZVm6v*S89dMpC#LK7ws?)<ZCpZ{9>Cj}LuvYmk0}P9RNVvQbSz0#)YMev
zfwwkz8|Rs#6W<dR!g+yTI^Lsd)W@r`yCx4A4=!e4eaKjXw>Ee?5#F9_fewYY|7FTg
zI~$`XXrd<Y_LgGR1?_YzK1N;f!=8Q1KG9Dxs_ITA?ys=)T&yCFFjE3szw%w*7B0V<
zPiF31ST~lMW+n1?Q({!PESF_snFXM)2yc%$9V!8DU)vn=;<lEfV|Ab|wcds7ii5Z7
z;O)M?>=yqJub%MsKHg<d{w*#P=HKHDnHImq>ugJM9S-3e3P8WY+cgi#Yd8|4`|IgP
zu8UFWc6f!dq4i$J$U2&Q-+buIo<DTD#Sab1_d|P%pexJbak?C>tV_t`?fhNG$mXp{
z=3X{<+m6dk@OCG>tqN}shWyZt<R5yK=O;Uqf2!PG_MoyuW<$GZ9fY^%7QpJ;(b|&|
zttl<hq-F3$m2hYYOkdyHq0bBPC&Swd%N?2vWBc53Xw?ns$?)|U$9VLv4|V6(Tt5JB
z<JX~cXK=rMc!jyX3ccKy`}BB<2IsN1D?e3X$4{-E^ixflCu;N~M)OwTyJ{S*Q}e#-
z{5AHvQFHx+TI!Fjq4f1bRiB!rcV>Vt5AbK!*k9Xo`>Px5`E9+QOd-BXx#6$=&jU56
zWst@`3Dod|ffCyB_lAM0o|`(T)n5;K`RP;->V?gM)wYgJhpF{`PmR<|{IzTF%C?HZ
zpG%$d>QuNuHsFb_eD{wJQJ(M+a=?N$zajPFk3kx~CrG^)2B~zOAdQ2!hqLf=lVCke
z4pi1zK}u!bGka-vyy0)%f5%5(nD@L>k{zqeeCGQWpn=8lB-Lg2+QL9Bhqr6C;pcJl
zRr5AqHM$scne@z7J|%DVg+UG988mPo-$%0y@=7%`D~ayN&hG4M-YUf2w;r5p95MLl
zAobr5Xo+LX-E_RjC+e9WwYc_2-Tm-E)xW$}p~tWFDEnIlr@qwZQBRaM>#0I!J=1_;
z&y_XqjaC-;$ZpAx<jWb9tMdmnoAw+n_EIx$zSN>^FXjLKrJg^0r4{*JtCL=9@8?%)
zd+!xHcVFSpd8JJSU#mio*UG!)o$78jt3+F7Fei{>!27CmjPFSFLbC$Q!R$wC9AoGB
z3ol(P3^(tGs`0f@l?x75($ipBhSGoS8zMh?!*%DwYkQC$-we|2vq3WcR~Ild&+{ly
z>D1u+_a(PtlsEe;n727;RODPZNBwWoa<YC$kwgECe1vGT`qJ;bhxWNbk8eR6gOdMs
zS9^Gy<*K_PkGeC9<1YVbceR59{WikbBscv7^J<|ptJiSXs`l<$)7@Rsr`=Tfr9l@S
z8I%f#XFPV(73XKo(r0zr`dKYceb&elUo^Yh7d4)0Qp3TV`&4k3_e$!iTio^agS)ad
zH0T97Z+SI?T61pHG0aW=A#Pgj<))S%Ze%|5oVVR%IpU^zd#F|Hc9Y*cH|4dsiGU{k
zvDTo1ocsLNh4USF6+f7o?F54=U8S}WY0%@=Mm;QVRCynRa=2h~8Kd?MHtH*Utnimv
zseR2lHQcO2mCd>}44-f|iz44!^t>_irGHRIMa#v));sXF1iTH7M%#VykmqOoQQP4l
z?C#QxddN+aR#nAUJ=m-l7pRfI+lWC%PO0!jO@yP7?%Mp_UAy=@-{o`q<}*+ma@V2=
zcZJ8elRM9>5Vx(YiDsvM@CWC4@6qs=JJ8b|K|lMPQLi>qU)~RcZ<*wOjsERpvzk*|
zo?$g>Y%-o`SX+(bPgq;Bszq7r!b!C52AJ&L72P|_qCGkB6;n@N3vWBZ+kWtNshxQu
zj(^ax*}qVmeqzyh_&t&P7lFHZV0HX<Sp5_|?9YG0K4e~Td50SMD0sVL7MVQ1$I};g
zs!a{2%I?Qc1AotVVU`3RO*VKNax_6ZAEASz$PXINPPe{fGK^1F2KfwM$REsKKSi}0
zrmCZRnsVexX9s7xCO3k+?5G<=UAXq;blvexS5ERF&+X38s^AQrS(2d}bErEPM)PJR
zpPB55tnjwkr*v&Pm#(tg(zUjGy6U#zvDMRc>3N#+9Zu6*c)JJQ4uQ9E^HP<gW~zFY
zrFToMI7bsQ{=X&aM<km33NxlKb-Yi!n(dC)Nj}q|Y4ki{R&#v3^#j=X(mqyCvZ6uh
z&-Ju@)9T1?y1|aKQLAH=<6De=M@Kb19;0i_jNJQ|S<Oc=8oq<RV0|<&k3Eh)e%~fm
zsXbyfiyHI`c)M*(tfs-zLJ2YY)QkHsMdMMgZLp30WV}Nu@OGF1zEmQYy)1JhWDu!n
zly1!buJ<tH?Sh}wu79dJyGW~?BAd4xnzazVsaNrG#F3TMAddV1=5Eq>&SlYRG$mS=
zn{cnVLpkfv-@Hms6TMqtTnu}Q@DPvvsi;oOlfv8Sao<&ueK1cF@z#1q=?q?5-=J@*
z74l6%9lom&ylo3_Bbb9Jcagno_kZYe&<}lux1;jHICz^2AJG&vWZN6;kh7vC2T*&y
z6)msq^kysKLx#6y;O&{x4t1)`Tu>8u+YuksX!^HT9P<6wp;~C)zSNpce7<`eSGcXs
z1&$pK^@Pi-$5B`Q-J#)dHy*t^;i^M7$)lKb*rBC=I>>bYscoZqA9Km0AOkV%{x?kx
zW-gUoHMOopsbD+&zR!a765e{k+e+{@$;GbHKK{x<t+gn5w^b~D3f<<XwGI8WBNw|`
zviYh-HD5KW&90Oa-b(WIVmB}I6SL?=QHwe-_WzM|)^SniUl+HNwYy!ryY<*z*u8dO
zH#QcfAR^L&fTW}nA`-(4(=(KmNDB-M4T1`|K4+ibAK%vu4Fk;kbMHO(92q7@DuMQ2
zFmXmMyt>m9pEcK)e>nr3mW$Hr9Gs)BRhp8EI_8w>khc#HQkUbLhi%6)5xFK4QM)oB
zKbwh1SGjA#x;C~Z8(qlTmPo^vUi7*#OhPu2@cSp{bF7Ib?TZJx@HtPH8cYlJY~S&`
zP7j-JqnIP!nYnA5v+*K@`7_VCTQQd(voszPm#|)Z91kz@w$HnGOf!wgokM&MV;!`y
zSOs}0Z@;UMZIysm`mB$_WZ2jwg6CBs{Ei4Ao*+i%RWUjhOA$%lwhIzMEsw^jwGnXO
zHSe!a*!UwD>Vhz=3y8oV&nWZ^h{BNh(e#59@P8!=H~nL9;#V|1R-(B_6a#zm_Tn1>
zw!atPX^a5Ebh7weEUc1ZIk$*Kw2qMe(?Up1#OOfYE+lX3xAQrQ&s{-|+_MazXYOx4
zpUF70e2_qYGHPzOr6J%!Dzl%N2i}Ews-05N`E3fsi&%qfqYp11r!T>uaSz-J<$mh}
zUK@DLxs;00TT{_0mh&_LJ&Z!ANpj>&wMm5leKjmEC%{TbU2rD*uHa)wjqK8wYVKL6
zcpg$?6M1WOMvcqE)u<S#W-Y76Jn}X!L5&T3-O5&K9EegeXNSFAxeEU}svvV#A&b1-
z&+|<Cb1KeI6wotOFe5{b*=H1Z%lnxD3OpI6M7z~WIGjv`bekGW$y?XKYGjeyQ!bFX
zkJY$9{_f=0yhxUoKUTq_w+aIUO0-B&GB;fX=XWY>;Ma8DuY&bq71GJu>MJVj(^t`>
zNrl~4RXE>Ah4_AQ6p^<l$u<!g)_5Wwo$tjX#48?WQseRdCz;zX0ngZv7U**B{FFRo
zy}No%65bBsKF(_PZRJTQVjY*kKI+yZ{GHga)+KL`k+*AE*Z$%R*XAM5NGI5z?ax^?
zXUNMf6EI;F|5n!}Kvza?9*u{~MV{06HRqGJA?(kma&G%4S&jLrYLq3aamqoBxi)He
zd{HBxHLmY(HTtr*T*<SC4bKaDb@8}F-m;^Kox50{JFrg9O+Y*MMD)JO-CWkV4Xkl3
zKXDE%;hqq`Z|^kj3h_L3^WP*K8p7IkJkJf}ts!|kpS&$PnG9!p_Ln&q`N*GfSs^{G
z$lG1tl9{hbT@_zHBrKVk()6ynkc^eoI$uAR46$)CeUms3=HKnH<YaUrZ@;D|BaHm@
zoWUNZE<N3s)?v&r)>G}7ef^>ip0jJ=DzAmoz8>rD*JGJ6>)OzI{Hv+QeQK01Z2li_
zTavf?GMIVJ+?xbFX5C!lK2h6d_)|Mo$DQ1S&CQ$#Gt=Ipnf}St8Mme{<ig+3`25CI
z{ol-B=gZuyZPV&E9+S6yGn;W$+>Bx5?JDxNdHw%*d*c^#V<WjEYuSWQ_LJL_x67|I
zVo|F`442RYk^4G5^r<!8`vd>dZ_qu5^|*|_y@uR%eM27z@^+X3J&CBX-8j4sQ_0(*
z<ZU#W*@nFBq|{=^H!Uuaw{E=DMOv6vY2naB-*x)n+Vx`I$?Y->>r{yk4=Z6ASIOCX
z6>3>;6**TUU}X)@i`3}st-+=}^krb5)Ady~9M^D%l#IPLnk?F0j_Lzth^s7x^R`kP
zqR;K!y&7D%)4-}$gERIeFq@#of!SKr9oNE(KJ+%sSDD$P441~0u_wvQv)9a#VgA~d
zugruot$=P=1!9g@A|tSp{ymkP=U1bb4Qsuz)Z<2Q?m3Y&+BQ|_VOPd1zH)r$$L`;u
zMOIQNpT<kkSU_*0Ui6fY<vD}AJ#EGeHS%`l^>X$Ym?a*{>>Bbm`N#iwYdnB8<>Ctd
zO>tMopS5m%1xC-R#BcI`R$KOw$=g5V?BTyyw|1{Whv8MoBX55!<~*0Y9cx;Jb}!g7
zcCNw#@-~624d!*5yzNcirtwT-ev8~qtAvu@w_J~2PmkFncB;Y$vi25x(}v`*dwL}{
zvCdxliTgUtSpC5qI`85VlyPr#cVjNv(<{i8yzQWug^L@raJ}<aczwvkj6q+p#D_k!
z?&<V~Plx%rbSTcJ11-qd%^7IhB^{6Mv37D}=I85F=56x%`5OIC*^3%Y-d=vIL@9e|
z{oiq}Q${bh<`neG<yn{K+}S?l!q-GtQ%@23FLg!H8OXT)1<CWN$KCs1ZJdc<>R5*#
zW>1&A)oISg8z=UOYq*!ey27xp3|+Uz!RM0*t51k9r(TG4W<oq1E5xD&LZsx0;5bQ!
zQOOF_@&CSC!x!YFa5r<TntK5<j1@>xXC+0fg%l1(QVhN?Mdf8FME(+V7!ZqfyiS~t
zh2@`E-0ml&FEl?lYtC!GVo;eNgx4h@HXo0HbdvzylLTnFKNbNFLY(<W2&cwydZLG8
z_o@guX20crvoD652jb$l5G3>tMaq~^94-xq-Qy_qaE?OJgb;eVhoI_NG(OE1;HISj
z2iyeMM&2GEZ~LbSknbLgQkPg%d&S~sd@Nc`5TZ&~h^1$RcxxquYP<+>Mk4l;;t=&&
zinu}<md}xL7fy`q8&YKJ$Z1t6$MWF{cx{m5$|5;@+eoqHn+PsP1W0);!A6M^w>Pl2
z`cMVi5EYK9Ralv=f_N~WtHa|_vxxN_Svhbc`@2OdxMZubsyAncgV?_u5|2(}$sY1{
zI(ciwv&xj0DqQ?XZfeNp-fDa!Z&yxMBa*#iFY<N@dAp0e&4^S%hrFFk-hLzdV#(Vs
z9~HQ~LxJ`73QTfSAk|m_yBaxK?^oc4uN(uO$k8rXjto}?bW&Bg#NTVX{`}s^+}Z);
z4DUzvWpDU5dk`M1XV<EzB~-##s6;>VHiKuu{k)o%sPKnh&t$v`gRPY)c&x<DdrExd
zt%BF8<4Rnds~}6|IGLix`aWcLH`a|_D(;Y}(2cdZiwAd-8scH!mwoF62?!%^iwzU#
z`Nr99+eCVHa5tE|4J5aM;*zm<HtX7h)F7L%@A{fPN<7P$D|pr+Z<pStcR$aG35%(R
z-kpePTdDV2Ms|<=@2nG#BRmW0vepi?j)xN;=j;Azcnj3%q2OctnP-i7_6O|M*m9in
z=qqZ(K3B6ps>XEIzRP${Q2tP3DCfC<&T?j96^|8B@pwpf>vl`PKjRaSbe$Sy^0qlN
z363AR9~7AcBQkd`dHb?F2}^z_VI`T{qCN??nvxLPH5n5|@eIM9#<9W4=+lqB%-!gV
z(TRQL<|OWqP>VH`bu4+hH<Uga!Su;uZQFS>@1G%O$=l65Cwh_9FW#_!&i>32&iRM4
z$2^0)eZRgI{qkzD*_0j$3VJ4xw`)h%qB*7(x9-wUj~)!2=|wn)v)l>ftyd8}kjUQ7
zPwMGALl*0>o*c^zSMv7C!bVK-ZiMUQCe%hX;cXf-hc<I3XK@qGuup95)J$)_X6(8D
z8}-+ifk56)9Zp>pd7Dq(27PHpCmHKuL-wOjHRJxIW}Id(c~D6cmXfy$&Tn1G+c56+
zEadF=TZcw;b)#R0-7i?={ea1yAJ}n|8tN(a=z0G;Vz@)PlYQdt7phS;i*;Ws?vXrU
zZ*y`LzLJ3#FO*|_Ng0H4Eso2y^q174AVZ6t`P9XfY4J~k7Oz{8Z{)4R%`)V)t;Bos
zcFTuKeEz43J2#v?UZEz4_19$~=aLCknDD+DA)KxDAa56vf7vz6l3B`~7*z${k+-?y
zOEDn31oFR1aB4#_jQxtZOIHMEk0RXjEMl$_Z@KF~L$3(m&lRB$dE3>e6gwKVFj=65
zEx*^CFFXr8uE2XcdgZsEZfSfa!o`({9?cz7*0qjgm`23h)H&2RMKkZ)s*KsE>>IO=
zwcVygHhmf{bu7iB@KV-2rTqNV-EPz}(@qQU{+T2#2JmN}$Dg~9ANyTg4#hX-#pzUF
z==chH)mOmWoAX@q_Un{NOyi7r$>B=Oc~FVa%u4KRs6-f<dwdx8Y{=U*@-})OwZWY2
zs{dmyc{`O{?X$ZIOE>U5vW&Ix8M65Tdjaer%Tg<`J+=~S^((QotO6&C%W!M47Wq?o
z9vaNE&=BgEMriTJREtgI?d5x=>`&xl=j|Na<{spEWWuN$*>;}$8J1s=*@^irUg>cE
zMUVE&DTuP7mj4BHgHJ!Bz9(l;<I=D_A_HBor_=Kw4Sp?C(NU0slk5-e@0f&uYAHH|
z%dpi+j;(oQZm<egk7O|IEJIX#YKmGZ5OhU}ZLA>=@HwUXFV2Se|JtPk^)%@`KlA_l
z=LhO|os*!x$X&xTi7<c8y~+!699<?yr*?AY{KesLg9u3hBJ|!Pf=7)IeaPGDQ9|}t
zh3xB#u${aenxuf~dO7onrKoN%!KwiQ3~Vnz@}C$i)x=<X9_z{E7#JvH@IV$rFKq!n
z^p3^pezCCQWuc0KxF&+xh~Y?B9fpQ}p)fZOLC2JL(8L5`d1D}2h6lj-K_Ck3g7NiP
zEc$#5ht9oFNPhdHhrkQh$lIKD{#eu2AD>tFW5rQ_d|mt=*E>bxk5@2ev<bm}{}A@k
zgRwg+lo=CYi0%-E6$8Rxtrv#IRWT4c#4>Ly7JW~~Vuw_K7O!IQshbdgk+)k7geWo<
zVo6vuLQce>*DnDMosC6XAD#g|32~QKakLO)dy8NW5whnBP}r5{g!CB9A1pxI%>pcU
zk{~NiiIe2*00Z_{sRetiQZXZizdQ2w1bMrNbJjcL+Lvi6ykI?;{6&SO)hcvnJ=vPP
zEwkY4Fi?eSp(^ARs*q%*LgNR%U#tofTC4GzyiJ^-W{y1TPBPJqEL@(Z!sSmYI2I|P
z8>&FX1qEt0Dj?Xbz`CspwB4lO*Zr@SRp4v29C49ye0G&1*;I~te>uF#TR--H9ml9)
zP8NRW*LcZug6nw|&Ye|3Urz;l)~&W*mB?MMq}R6+YuhR?GF*wB<ZUf^n?~Ml@5hg^
zQX-p-y{NB5j}uDfO(-#A9=~=Q1s?U4<Ktl^9_-}#gFpL`N7PxpQlWz9k_}#BxaG#d
zp|b>aJhu#vk#Gl70!#k=+xx^*SH!t;IW==@)kxZxfNkV$(<JthSzjJ_5|3VcSc{5O
z=-o<%xp$O!vQ&xfCzX(r|3-WqUuMLk-j((2vv>p@j7P;do^e>SCp=~C?xBWXq#Enq
z@f?t(hOUx*``c=4-_QO5dF#yI$%{bNyX-Tk7pZZAyiMp6k0x?^xJ5j5nDH>@=a+Zo
zj_x?>%Jh@4l)Oz2OrrM{bzkf&cNcKBB1=LZX9bJvIG-kS>%McIOy0gDTf0&}Rl(~7
zdAp9BJ=uk|FK_3Pw}L|IymHBGew=4pGM4?6pL=-TUd6rM)9fKXAYaMcHfG7lBxnDd
zhjiy{XY$r)dM%1(F{35322Hipo21rYOmht?2UE8*s|L}Bzq5zL8n=5rtSr9c4ClEz
zx%CJkZ<7t{k+Y0Csyyzptox4hJ-%Z^uLka_|3FRJPx|}+!im^lcuwtd(G>b4vNm0N
zx*1)6(I0Z>Z&Y9Ujj*-9v3Vf<F}ORlj2Q{aq-F>{Hly+pxlG>n8_)Oe*M!7?M&_|M
z;-P6HPMS1AVcduwCmT_>x)GD~nY9!91E)9tz{ZY0Fk~1rS4VM1dz0DZcI=bisl&(Z
z)c2g!V#KaeW(SmFt`U9shB8B}V+pLq)YZgkaYCxaV)FJCdHbGRT2{&(kndVdCvV^O
zEW_vC%;+Su7ECFJ(uT88ol1E5RYFZ(bRut;khd}9ZA2hFfQzeP%=+zJH+ln;w;Rdi
z)s5`!4y-^FcQwBzm7v#m4R)?6#&92IfM2J#ZBZdU$_k;IR|w%adKa85!e{Pf&I_bR
zEc3>e(gSds4s*JBF4#@pCcf8V&<ic5jNrWW1~X__bN|RGLCtICcE4v9O9lP@Ix{1c
zAJe^C8Fcx**JZP2Ca0DjD#Z<Ieg4cZfg8O7OVuUNoyJU;En3*?X)$`Imfm$*JRs}y
zUzEX(yp1Gp?{%uck}>Rg&nI`ERG?E}1xCm#@F1W0=tULyQCac-wf}<emGlDRE)VP5
zW8|&hY-*!8Up*gEfkh4#xb9Seho35;OWt<k<2P_F`vgy@>k?LSpQ;j9J~Nm7BQts0
zF^{LJ464t?Fia_co1}ne1A63c<Sxj$d@Puek5X!6qS+4~8o+#Q){wQcGhv~V30ry(
zc#Qpmnsr~`N8axC=Pc(=B1WF(?!j{ISj^&#hdo$BzXXUUCZgq4p4<OT!1Qf=F4!vP
zV<p2NnGB`ZWH?tPr5CppS7%GPcO`}38|z1{6n%e5@mD(;oXA^i^0sH43ROGlOSg)<
z7d)q5bx`0<fE*j&NbxdTh-flo)m|}<+ljE|sR$$VW9gwK!2X&TEGc9U_JjZ)G66cW
z-x@J67RHlfdG?8gbhr=?jD*~Y65>*`5a0JiBXYR_do~Kt+)n^Uvev5IM}(dYgWiZR
zbUG7;^|!*{{wNIH*!R6PHVi?>L$GN_FswHRW8}JEblMt-u6o}1bjusL*4}8F?~U8-
zeW2Ub7wv@pI6F50!i=}L8t;SWu0HgB_CdFM9?VJbMBX`1xS6|S)=hVW9Ce3$jR&^g
z^}w7*!T3hruJH|qt6vB`CVXLZ(H9Nl12KfWeO2L)F1h~neelP^J;C^UQz)`W^DGc2
zK!;nA=#m(Qgh}r)B=Q}`g}uYnx)1Pv_z^kVV^Du8mitu#3`q({_JohnP9uw@(b!WQ
zjmy8IQA;i_e;<Q!0|ofZ8hsLZyMVl{U$4UKy()BlreKXH$M{q^o|Y@<b)te#M>U+H
zl^8cp#T^vZtou}G%j?@Z6;3Zx!m}s)@0VE9lDFM&sL;zqg(&j&7kOL8I<_}im_Ld$
zMqc);KM%A~BcPI>hg`nUNr|9%1zwW3)5+UWLlhV^N`Wck6u3Ky_sLr$@>W9L8j!c4
z<ZbzVIhG0KSjQUkC3$<5y#3Oadgw*0k@-Eo`lZC*$w~;{DRG{4YUBhZ`t(&|1$kR0
z=HFQYwdB1O@aE2zeP0!>ge&NQ$ur_%IdXsTZ<&1%G*NfnLQW4)Io4;$Ad|}2`<5Z*
zm<oo!l`!#FAYgtRGzMbKUMfKedD}ZuLXT7lIzN}A^r!;DG$nMj<Ue^k(2f0c&M%I&
zVGlPY9$Iqmh9P&7_%l4MR#Mle<a|d-&At-HY^iat<;N_H$FiI8I8QdG52wzZ|0_=Y
zcy>9h#$J9tqhQX+z17SJP$TAsnjB~Eo_+Q2YdF)spvGE1HQuFDuU4eSjRy7<I`izr
z+IDhYJW5-zSFPcUpgIv7wk6>Sd3*UKd&%T&fA*9U$lR=GYQd#^8+qG;-26-pmt8hJ
zAb%v`6?uDuGu<oy@wN^7%H;4#zTJhr=f0dPjUaFJa*{BLuUplcnG4`9@ea-~sJF^z
zed~KQ8GG)s*Y$*ZMK8H;WX0D%qb8bK1jC87)aFvtlU73=EcHHedb&5&z*vV_E!5%U
zZmDM81^p6SzGK59`U*DC`&LkmvgB&a>A|_;SbF(<`_5h&Gf8Xf&`{33&3T;vW-@c8
z73aD)e_><uFI*eQjP@r@5KL*tsYmpp(rHHh3F@TQ{Kg9Mwy5oI%pq?pSOYgkHsjpm
zX2?JN!j7%Xh;6})JI;@nvz{JK#%tN%E~7VLMSFS?(|0Mt<~x3q^`AYcQ4OG0h}xo{
z{>)gvL9bz(I_5T1vUb#Bs$U6xhfBBvT>|}nB{*BH!QGTXn5%P;9je7e_Eg)Gw{ytb
z+vIH>`^t&rZCT4QT-PZ>|DL6|KxS1=FUPe9<rvbX606!);9RS6$XnA<bxk=IzbMBi
zWjPGmkQc1w+?v^IwPW7+>`E+rTY+hO|D%tKF_Qj3o3rQ%TTq0<UPY)cp!d+DLd1*c
zeNa&d1MZYqd}r?H!6InM<4(7zg}q&jy=BEH{6yb;^7aCG>(8CgnODhReR>QEN>CR<
zc9OR%S+`zsVvbCFDdz{JI9#m7B4(ovXe!0EV&<jsyrIn~!ROf}Sii6YbLfdQZ%qkC
z@a0+bY`DFnlsmMgP~0tLKdcnPqsi>+%sOvitvj*;ea2Uyhgk(WlD978?L+c5tdO-X
zd3%}MZ;@XKZEhtL^_3XIbH?8J?7y?8J;0<4^B*z4mA!%$`em5N&lz#03>Pm^=ft}D
zJby1yS*2K<%xpU|*6iym@R_{5Lf%eR(=*UM2O>Y_j*ez6>cU*IHq?`5rNAj61N(9_
z;7!l>MPt%n_IDaS$mri%nutwr6QS_ruA^@vw!BS1+BTkf$=kgCD%|p6zb##Xf@szl
zL2|U@bKk|sQltipaVQ{;=V%eWZHeX1m;`>KrHG!$o^7%e*U4L-Qub@VOJV&*3bRg9
z>=_$}KvNM!79xziCqj=4LU<n+GV@hPKYt;TE(<aF1~ozRgxL91fR%j(m~tf+pUK-H
zRWV4?iN^QxD2TR2V;+02eFw#27J2K>Uh50$igNbGAT~G}<_n`?JvJJeo}8&(i9&!`
zB&xcF;PIybtioH2Z1)y@9sOb8<BwpeKj%39=n(6VoB`e_*Y!rmE^iD7^hV3;-spSG
z8#ixx<JU`XYH7WpZR-PP@^*>9ANq3wU_jpfrSQRPjVIPGa6;^0C&ZjEN2jGP@bRV_
zO#gm^dEejQ*xfg180w1MQEuqx$JecO$JWsv=()rfQIfa#Q0M`5k_ST9d2%-6fh+6$
zA^qDE+H)RgZ{&d%MS<uk2|$qbTlBu^1FtFX@NWA%s8+s16fc+gVbD1fj-}_rQ6>#T
zFZ(cRaz0{CWHh#sw?5=;&r>mo3W`Bn@^&0~Td-3AUte;Uy@8?EM0oREgnhnou=**+
zf_f>Q<jI*at3c6NB{rrgp;u2f{^9v(tP&-&l!#(YHg~NO2lpz`Y_FtlgqkBG_Ag$L
z)8uVYD>Zuq)I*J8uanmu@;0`O8fRH&uK21zSd#+D9t!+;pg{Rr1-g;914k*4&g;Mk
z1;#eWv5LIyBa&l?HUI8R<hVJ9J#teijQO?@*5n<@{%H|P1XBO)*(k$dT{)!t<<L36
zz6ftu&zEB+KPI_AhWZaOG)m?8o+*d*00jzp)pwKQ{Vy5J>)3;-lA-#$4CBb#?0;pL
zktf6AbQxmF+eI&BxY>s@37#u21xg{E5r>wRBJAMb^I`riCmM=y%ti!V_Pk&7@ws8H
zM6dfQNU6_y#GPNCbQRW6CtScj?!j5{aAQ3>ncw@e0O}p|l<4LpgR)40h}|ms2dL>&
z9*++@SSR;kjc%`k;*A>1JFC&WN{z0LYPk8bmu#v=O1K&muB$Pey!B!KVC5Wg_N*Fd
zUTVnF)Uef1D_+Id)$_hxJc>f;i^2N#I&1C_(?q!aOho=g`gxGIy5wzF@-~3H&0v3f
zk(hp%<ZT?eY018F%opk-c>iAeWE8S56~bE9ls)I)tYts(^)svJ7fIgQQ*TvDHcyQs
zZxdJ>lfCA%m`gN`J!tZF9C^F{Cg(luDc__X+n%hw&RSPQ-kup>ixEF-ut&xG?K$*?
zXPx$XPBof>sxg@QAUpOaKSXeT7D{c-xN6k!b-vN`zfI!XnyG(n%^k--)fit%{SLLf
zfAy__<#zhnKCi*Osv2zBO^=5pW~=w-zAQQ0pS=BN7X2YRH9;A|?A@*OR2<ug%RQMz
z)Tt4}=vmR(`zQLdRvq4*-V)qxl^&|YvR(86BKyzppgy>FHPC@NTuaXRsApa1PTwJ&
zYQ#`8+&PfDwB&8)e(ZH$qlaQZ8T^0H6DU}V&Gn^NI+$GX%|hSZS@=Mm;@O!w7)MX?
z&fRjMw<;Gk=W}^h;2xxH9tuP9u<m9)zD4F^tZ@OH=xb0rrWBgk5-hn@f*C<2=+68!
z=bSRmU&~;#x*Ufr%jum`PHidkQT5AlY%8<Gi_1~*l9?{~B^X`9{O?D_7)RdLe`U@|
z9zAZ!+c)Ixiid@0$vwjjIfc+8Z?BNIwbZ~~T~-9w*Yvid@6cN<vr%`JApIUSR>38h
zOWt-m&mQ=tQpjRT5IMU9w$xz#>cZTbBc-sgr}t5MDVlecV%|Kmb51EdXO%+wvxJ^v
zC8%!JU=(@Vc?o^>7nC5E8m|0v^xLm0LCWD$^kARXmc0Ez#y4FnLrr}dwvMPk`H%{@
z^5?tG4D=sesag6#tyfwZ%J!8ZdPW(}khPxd59EH+;#nqp)OWOSH7tQaiUvuM%)@ch
zV9QGlM9(#-;w5L+tAPIf`K22Cb6bND*5*6sGf#bQ3FdDu;QTEcPZwoj_`I)JK$fm*
zPaV^rRGc*CoRz%&DbB#w2lQ|?=05#%{%`zFM5nSutpAV*fe*7r{^34<CI7dnO<8wL
zfd%C1*@p^*dMcsKkzr<k8FmH9p|@O!Th0H?69s5xE5N)BQOLd$gWx9udW?wBDNo4$
zod`1W^EKz1rH6!Y;r)#!LM)si!f8G~wf1JebDa>~HVR=(-X7Y|ey{IG+@JjsyV`t2
zhgsnmbvG1kzl0(DSU7IX3CFYUVYpZSj`dC;RAWMM+CK~x_daqbG8#EAB9LAd4y%_R
z@ZnAvoNYqkx-AqhmW1M=Fa+Ohg0O7AACi)M@Y&8Ad)(Zy&ej9Dqdn1gnJ4@<c|w!x
zg{)TISV-R18oJ|tvpeRi-Kl5x#^VFtJcoNz&+mcdFT60{+Kb+t{+M^$5qsX)Vt0fy
z23a^lxYd@Pbe4FWVg{2{X3#pB;kcg}h7U4Fk+C_?{ub~~dWp}4Hn^8(gFcNmn7q^;
zR&a*iRcG9hc`(!T6{Mk7=;!NzZyR25U(g*lH<G_O?zsEL4Z|+FKquOkwTTlJ%YCqX
z`a9ghJIEK4yAOP!_$v@DMqw!Ab?8ADcOfES`7Vk*{b=;Q9D^3*?VSTr@Xim%ad9~8
zZ%07+`V$)bKH<;PD9(?gkP<0`eVaH~c8tTs$8pRF5n=`B!O@2lkaLzgeWwCLZY!|w
zkpe@ld4Ho4gUH)|cs?t;#<}S&YLL=Y%%)J`AnVW9Bh{Eq-nz9XE4VY&^{WElb+k!=
zL*(uCCkl)vZ|AjFFpEgenVcMpa;O2X=FFDYh*&v<%Vc=Ly((Q@&ZlQdF}j}=Zgmng
zB}njXu>?EG+jnhbnCdM>1m8BNK!O|3rC2&r1`&1SdUs{aT$1s5O@gS_5}4c)<Kta1
z>UjG|pZ%2GVt6hVqx~%EzyB5^K^uqJHE}R4k3-7mIC^_?MqU|*g?u@-Tm<Jv5&tj$
z_xJqvt_U5-TW4DlZm~b`oV@ilQ)0>%C6*Y-;m2#+Q#pq9kt1aeYvAP)L>`o&ot*@p
z{Qf&EmtwPn42F?1Oz*&XwNeIoybSKW<UnsFybmgo#{bRwdKIp&P@|in8V(Me(NY(-
z&qvKO2X}y)_?Yvt4w}jH40+qdon9U?*1^<MjVM*~`Am(IE!k7<MlXyZ@d#~9;64=d
z`?y;bbv+3?$lFu?Ntpa53F+3H?F6xJ9Lb#_O(KR&O~SgPtap{vDi+XNl9x*wYhhkh
zDZHOVM#s~S<Z}}3&Zu$b+h3*ApIMXyeZKxIIUP@Z$Hsp-2kuCpNHX`o-QFzrls{3Y
zTu06>;}yp1+qhc9i@6)hy7psPHCj_Yx{16U;m`e-x79GAMnz!4UCj`B5X|H}(ytoR
zx$pUnKD2Lta3<QK23`NH#v|&3yvf_jPBq*^u0f$CxkrxoU0jD8k2+=tvF7yqiRq7i
z!DB`f%E;T1+@ZD5ZN$NGjaWn8j_uHh^xt*(9>kAns>Sd=%pb3+#@DO#)LlXTFRO;S
zFEiWz<82AGzSPlNAal3BuELoU6*#W0#Nx-bXs4}X2Htn<ILltRK{@swEJwU6GdMft
z;=;!q%p3d_XLn>F_Y`v|cQW_+N(LTrN6(&~1G(IP+!>#XXEu3gB5(c4+tiQTqd#1L
z_2jM10SyM`Yp_Va1f5t1?weE!V+ZP=&S<&o#kyEVt?wl*(%TioPd6Wi^e@oAMxXS~
z`D9fduI1<9d{G{AxbyIy8EKK?BIIyp+e%P~oj2*JN6)%zw#;ZDZ%5TJ3%pAaJ`5<r
zB4r^nmA~OrHvMGD+W@lChP)p}-kO~z3on*p1nb+o(@PNDrvx{2N)blJE@T}$(25>N
zlS{ElOm^{OJZ~_Y=8J|NNg8Bm=%X)SW~qojlT3qz1`WQFxBc4wFZ(9tF7;HloE?uS
zMen~#aT}#@WUu<rAT7Ea<BrkTa%?7Voyc3$bk2Io+g$SYsUiE=vHZ9#ti3bYt0!+=
z$lJx_ZN~@=wmE2!Yo)<kvi1tEh4gajLEcKp+l*VxO3yFF7~c|P+~D_VkOl4BEbNWp
z9?#sbm_y#$cP4W^Q(<y5l{!-9h;fHwrw%hMmnC6nn<PBqxjCegXZdL|dJ{;|U?oAA
zi3GhYB#0R$##6^Q%rfHh?YTH8cf?`)$~dgjiN(lnF?i+>g^V!H11CiR&Jie?5Df`A
zd*_G<8^VNWm_)|ziG=WHIBL_vnTbh$vhMO&{efDk4``?N5&w?<i1BtGkZk+`*G_++
ze`P4_o`#~|?ND?lYoo7(QYR1wC-QbBd7IxY3@6Fk)#5;GI_8fdMSeJb+K;nNKSclI
zgT%Ey$PD(z)}e24vC#+qKYg(JxDO_8^1y*4S7emfp~DAT`qtWE!%iopxH-|+(h15h
zPV^LUf?2i`j*6Y(E_;n%58Yv4<&G$oJG}3Bz;C59<_>X&-5_^Jn#jD$H{2a?Md&6+
z7>sa4;6z*02Ux>vtQl%Do?=_}6HE$viZA5tU%uq-X*1L}Jw-^tQ;c(Xf<19hG3C-r
zl>D+mr!Cf4G0GYf##p0u`ZFva@)DQ#Sz(y;1>D1pkbCVpMxA>B2SXcp1lXX~!WPlw
zZOiF)c-F-k7E3+w`uba#rM-pDF<-nVZ`Foj&;|$K_D_GDmIXmJB$&Gu!OVaUMxiR0
zHEB33$=liF?HTem*dPKyW*_lW{t@*hA2H7+3h!6N;_k6n-1|2c<Mf1>LEh&4qd-f}
zSMQHe;4x>MN3SSw<t2L`8>lz3p$^GXiB?yYxO$x0U(SK;S%049yi<p~9Xgbp_{Cl<
z>td@+1$OW%<F3y{^7bWpYq3TF`?d;d5#(q?-p(U$&CAK2L^(EZmt+578NzyUr)rZF
znwe4*_m^URy#zbdWaml=KK~YDH|G|8$=e|EwmM$|hY&Ghoy6=Rit+8N7`^+7@$N|+
zWIe@DUlF4Zd8@u5M%!~_?;$asEfOPZrWm8h+jXzv(C8$@(Y@@yED>TyUM$-2n%Gwe
zw+BKT+b@EKyq(GO<Ig)HlszVItwoqF5Fx{iEauCy7!kdl#L#aRqk`YxbaH-C4+(lK
zk-(k2y=yDM9`aUysTAK&Na4-7%y~z$R>Hk(HS2Y-4u3-*mHq4oRB-mpxvu?E*0YZ|
zyR~4y;3joa)P(DjwO@X4#@vH%V^3iLd3(%_y=SQ!2ITGUV)mNJ+pgs8K=Rg;yp0?c
zk7({x%_nbvD%jgnQWs68RzIVr`ZhIE)Ghzz#@vT~)IN<$qSp|2f~oylaV!aGeE(}H
z)KY&=#5Su$`i67<ORjEACQsE#Xkxwlhb-P?l7#Ua*jK)ngf?F}&;3o^cIRaNJ+ila
zDw#DJcY@ecE+c0b@Ur1mzP<`i_t0y8c@<}X+~cH{<?{1NoTT37fg5$MR@B{a_UPnS
z#l6F7q<B@M)oEt4T&lu#zJ4BirhoH&`tki|_p8P(GCOH=4TP+3XH;-+<{Ui+r`AEy
z_dCAPC)lzZHPA!q(6g`>wJqw<W?UTtH`4<#@fTJs<}8_drnJ8*>3vm+Mf@58cHAE!
z^Z!~-pLz23D0%yqJ=;-HCG5c!qsF-yac!9Sy@z?F<gF9AJUz1xYWhPA;`cReTOoGQ
zHz59O0hGJ*(7!GB8@WsIQJsn2dKtJz-fCuM;OCGubehI&<RSF^yp;k^CuUnTrJy~t
zJc@RuV`J|OJfFi1(tWJY$S7xWKpkI<DE4oCSi9zvw-)wV=x_#XKdKlv_vS-8Fdx;-
zwf@+Uht+j?aOUk^^7eB^9*&W>-rI^$-hp21^t1iO>%oOWl$#Zz%!YX_<gL#;ddEM^
zLl>WXc=~>W3%wLxGgrGPr3n348~+$xj1w!1;nq@v?SE^SQKvzAp#~>dv-*&m`E?qM
zoj|tUq&LH}5`1s>Uq4-g3L_2Hn)3bWrPTi&`5L3aWAgSUuaVoS-`d08Z9*}%v>II6
zs6oHv5`?#6joY~tRhy}6=6-7pYhL}mrSKb73fyOXJhueyZA);uRs#dE2AguZlS9rM
z?5FqPrcxNvPgp#l80!wuzwlxK{rn3cxt|ZsVD8XxH`V7WGgj%-|L;)^c9FLic<DdV
zz~s6H<#)4iyK@%v(7wXnG!uIFQn7k^DyH5}M1~?BVeB7ScjsQ@C=~{ACYRiUI~eS<
znf@iikU}Zu@_+sCEeUkbOR(^`7#`&9q3*HReIN#oQBhbqC<<{sqA;{M0@CVmO#b}=
zT^qwtIXe``iu_R#6o@H7K^V6z5Cc~QVt_0VN8biQM>ml6rvW&b8vvuk0JO6TfcdBo
z7=JYge_aZqRzHYYK0$C@{tm(9?Dxl^*q|SZyLUoyTsIVtazd~>AsGElf^j_S9o#;8
zVasGsq}IEmq@@#%Pj^MByAzxbIYRNu0VChA4s`Xxm~c-FZ*sx1hF2Kd_XT=eTcd3h
z85LrSTrXQ#ZLmSz3~OWzvxZ;pOT<^d#EPFUaaP|3;uu@*Jln$Xg$2eJTA+HJHEgfi
z;#;m0dZxZ&_PHZ|>~MtUqCM=^Ib!%gzW){*H1;;5X7CAIBu`M6`vm8^K817f6J}#R
zL{AHSl#SEFiYKQbG&u|H_50{;Ziu&GChX~$Le&2$x-^<&?uBO%jDCir!Dcvi?+MnD
zxAkUsk#_wqvfCSAz+d;#U1JDMA4@b<+QKx^9>qg=PFUoGDZQMT72pj28*i}h&Ks0f
zIANpE1%X+7EN*&Zg0VMpID@VfdE?p*A4I45q0`@in6DFr3l{H@QT`qcyMm$e3dP{o
zA0V*`$2xbOkvQ{SR3C}zsZkiged6KNPq$NvQB@+w`Hs}0>C3QRK<4Hs@QL-R5$C1P
zsI3w!IGZGISM$nT!Cl*~N}j#hGk&E&c!mNiUU0^FQ-QgM6{uc9ecN0GT9CI+?AvzV
zEycq#5*TlgPzxb}tXvH9KrvQym*CJW2`)+{^zad*wOovi55*8Wh%q>t{PYy#oS~RK
zIr7aS4h_`REt@FjE{d3EFEN(%5yO_e-F-5S8H;hq?#@dm4()P95T}dq@F#V5B|=#9
z?3hL^!5xthS^b2VYY~gTJY&(28w*puyq$YtA>6r^D1<1aZhg`e5w?-9<}XFK6f8ph
zX%R-<7jf@aggukvP%vB!ovmX0tr5eZlZ3t#5}f!)f)6hxoL@`Ouvm(O<5DymN^#eX
zwJv1Lm6RcRfgBZ_TUYO(o=U64bL!Z-leg-JYM4AzW6w3t#lKRk=B`3zsS1;-sB`;F
zeFXdV+g|Z)<ZUkX)vLd7Ux+&9@$Hy%FpNH5)WBIDh^KCVJI35~-dCA`0aXbQ&SbvO
z5c(~TBWKxbuDq5Ai`)dLT@&!fk-m^?sNq<Vhy!!z<vg7pU(|H;f5EJK*F?<lNW^sk
z=TTB>K)NKsmkho#KM7sP+g2~=8yS^E?*wv~^WHVw<(;vgI!W?&Jb9~NU%AncyW8Y#
z;X3wYSI`%pGqyM1>0fY;`dW)hj9SDxcLHn2uu5PHcVPB&CTB=Lf#^zn?Z&;2Q&r5u
z<=YLa(2-p0ZA~vB)}8<P$hX=^{|F22ppv(<*{{xQ$(%O!e;017#@?y)hS*=r{H$6m
zCU5&ssDr!{^K#y4VVT0*81i;e$4X4sRA9@i3fNdzz<m;Ru;FF6f2IsAjp^;PNQ1>O
z#aJUKMgV!6+PdujHM|w<x%vb%<HVQV@a)TO&&kJ}d-=#X%-qXm%;c)c!3%l^3~QB*
zoVZLB&B#P%*Gy*5e8ChC=0)bvOQc&0dazcqB5!{-ry%n0bUfd~-H{*Z(3O0pk4hF&
zle2MbQvoV&79g6}LVM<sxG{4xp83}0^mgx^kA3~}5#2u@`Q&XX>)J`=ZAcAk<~-K7
zF5JiW&V#zP0G-=23&xB2V_IgP&7yaGIP+L6`TDwiG%m=)?|ymA9>_!anLNbw$j6{0
z*2Ltk19=-w-rk-`Hj0aJZ-54~Ka#QCH25Uf;5B(WuS|m(nHqWs7vpG7F={6>XLXeZ
zbyqc5X`sO#)|ZQYS=YK~nB&f!B)(7nmSRYE7h~=HV!Ruz!G!(%Jn8I%w<yK5cFbJO
zrC$@z7{l0?{<5zWCpVTtXvcmqKX+*>?lY0M6IrVcWUab*IsN<jJ}=4J+w>DYkdz0@
zU3usnm&?4eTr66j2N`)=_ky`G^Rn>WKMUi|(fh%Ko<VxtkGjpT&+n=4E%qV?WuaAk
zI+8=u@RxfU#N+6-W5U{Xlnfo$%g~P3+SM|oJ(Hrgj}!x!iIJEk#9{Xs=AK7j!A%J!
zPnWP~DuzkV7|2&fLbWaeFItD=-m)-wi9_+cG6ddp-eH<=5Pn1iVPlv-T)z1uqhla0
z^$NsWK7Xh748$IxFKk<Q!YK3&%DvwpA=wR2oLmu=@4{@HH}s(LM1Ij*g#LVsRmtA4
z8R>%+^ZYPjjUV*p`$03o50i3z(ek!07WVMP-L7vT&GE#>Tu=0z=!u);?d{*MP^jmG
zg#oXyc$Xt=%pG78?11h=9dW9QBO0gJK|J~e7Il7s8HX+5VEhoPpPRx>VTwa>rpP;G
zik}Be(fyDqcl=E;_@WskHqT(KeTLYz7Fcz`0@ucy!>-5-dRb-|m}W+=N-Km9u)>0`
zmN1C1#K5d)?EM>K)oNpUjTu9E?FkmjpWu|{36i=z#qK0yeEOh|yhSIl@5pgn2s#df
z1U=Tqr*Y}Ybqw;oj(*N};9ql}+1?K@MD!3PwuUe`Y=rMtM$iv7!Y%T4fBqRf-FzCU
zLk+moWr!}VEYU%ieds1Dv=P|Q%h49Q0&USN&lc9^wvfDWWWUD|Q_0)YS)ORV;f=23
zt&qIEOWsE9^+EUZKBzh83#X2Lc>LNA{ZiiI?iyb>>I7iR%^)1<{|;9CT^B51uiE<^
zu51ax-^L+WrwWDg{wLVYkHu=%fd^-B2kesqNo43v^7h}qm3YfNp5)V<!+Oeb!%~i=
z)L*UAQD7~38~B`aTcrXGyq$Q1AHyEC-%{4MvlUQ^Slb%Nuyl(Q&&b=s?ESWy%sRD9
zj7{Y24f3{r7&%N$uzH;sW=F&@G7zJeotXQ_VhkZ~^~hUG^7gQ89OgX{V%`oBqyxy)
zJ|Z}fqu+LNwlO-6HEbL{kheX_TX|a%zqSwu*hkmR5aK!CZ+ec9XKW!d`w20dygm1Z
zd(c_2SjLw}>=&Yd`tNa4>JQ?1_2uV&NVc=qJ@;OO=2Id(WsiA|RD=L}otTlg<JXFr
zk1J+I9{<M{NuW3&LGx1y#<D&iNZtmKx4Pu5h`g<CFN2kUzGHLc$Z?XRc$*T-v$)IT
zr$VQ;D$H1=Lc2^AhHR!jinoj2@I1p=aOFK}u9H-Fo<O}h_px$Xs$r2#9XWZMM&26n
zzR4dln7qxO6OU6D;_-qS=-T9Xc=P$9lL@^lzy0U$#d>#tA{Na~#HsHISmjLbE;IUE
zMkL_Yl0?j#l?YY8L=5Au@R@<kUt*oS^gj3ej1pn@DG~AXXV#fWt@Awgo7X2{26;P(
zz3JPW1q*(1PCS77MEh9hZcWD4o!l8Zkc?Z$nL9+@nqR9#8FeXF-c{lOd3*aL{RYT5
zoxkbFH;#KVtTWU4S7QCXN{o11iN^tzIM{_bp@-?kdyGEvdhF>w<6h+nYD`-)FFciV
z+{HEUxxl<q@-{KN1}}bc=4xDph7I%_d{m89+2rZhO5|m5hjJ-*NCLD-3eqB(%pFpx
zMc2DpoW4yh&=xI5YD@7i>#+ygVnp(C@GOR>cQG{4#dsf744cbEIL|t8O#th{utMA=
zZ?$jfzxu2YSsD2_Jw2b#Pnj@&nt^d!G7vH#114=VU>ci=h^v{nI)L>Oc{_%@E%?fP
z%4MIS?a0ifaPE?Ehpf$4diESoL-5&jn0-u#EGix65}7BqGadJK6<~`V^S>_S!djk#
z&%1Ky+m?g#<n0*p)_|8!V;(N*Wsx6Q7<)Dg!WXP>U*|#bG7qnRGRNgn1_GvLVwyOI
zvy%c$2`xYh^Idwk&qHc&UVZW)9nC&8J=TY~=b_+v9y0iGUzlG~?@6!OWAq06^A%Y=
z$fPzom}s7Z<IXvFv@j38=JfylPz+Pn;qlqU7&JwLrWG2zIL;dKjs`ba>#lUrAeDZ>
zf&DZ%MYctd<v)M2ejdVnnn~nvIcwZ8CHNV^nQ=o2^gP*DaA1!5YG%8Rr>40dUnXNM
zSPM^yD~4G?F?&iHxc6lq-O>`|OfQ6kS3cGS=V9cIJj|KR@0mUirSvkaUPM2jC3#rP
zmotWDAu*Lc_f=mokoscr;w&s+jr-AwyR+A`;6ES>rZV#OJ$uMz?7uA`XXR3Avt_uq
zKMvi(MHu5I!oSBvNak~pt)T!Vnbgo2e8gavC}ir;lVN)#47Pp3$$R1G91@1(EyH*&
z4#kz&P`nm|;y!shkk7#eK0#=4K8U*<{@mO1!bJmDJSug;KTQExm*a=rv2J)i!4(}_
zxZ?X47ew2-ptGF|yvtp%Zleo2ZheD;c5iTp485A@jix|PDEm>XWKOM+i6_<!@`Ujx
zcW4xD=<&=29&4TPWSSkeRoJ04!47L9*h`&m5Bb$sSbougIeZRSwao!W+wIZN-U@$5
zTOjX-1tb?O(A>uYp>Hj*@XA9>ByXPxO)+1OeBEq{hg(eXa=$6QN0`Fsf*JhIJ?2j9
zBe>3Zgnc6n(RG|5Y)TAaRAY)wN;5P$m_aRkg3AsjXmi#CZc)bQ_0|~ORvE*(uQ8mj
z7$JF{Df%@%!M%=8QJ!c_y|q5QUXH`(pW|3V-qs}P!G8B?OmIGl1gDd5esB`n(36<E
z>?+z^zKYgcui^XX>-e_nG`62Tg`?9?;g<eM4C;6i`}-N-O3Gtq02|`(*H7s$Z;GUg
zPnps1oL*Yi*l^MoW7%)sEw+K1t~GiKY$1H$h`$Cq<CNQL;FJrD&%40S&;@H<Twq${
z!u$v~jPP?q(`0wF;qNkOt}iCE4M6?vAh?Wqho-sj@Q}Q9-57#2w@|En6^f~8AFxCo
z0h52@_<u&-ZiE8H<mdJ@1!UxHwypxVJ?K;Mj6T!k?Ih}n%~wkhJw}33Uc1@<-Em9;
zUjqrYx=L^-P=eFEPVv48XPW(%N#Rl*hrG^Wd>Sl<WCpqFFGhNl7?m;H!(m<hbT9v=
zJz2vB#?c3rwXZABk7TE|j`gNSh?}p481^4;slmI!et76n`hTqvV(LmEW(^TSr!f}#
z+~ZnIjr(M00qYh4Hu?#$BT|5|3IX>{1u&`?AbL9cJm+H3<jVefBK^L439+9&=1b><
zh`2>IlefZP?rFP{(cMKz590sVNf8V;i_kZf8vAi^_<SP{XV|l@^<ZD1NsJaVBzQbu
zf}v+5C?{`|y0bTajGklUZ3cP!rZx2vk7aOdBgZ{QIkMTu_gu_<A?nQJZB$5Dpu+Vp
z+`F1W?_}O~B5z%{lds!Ucpa=njYNfiI6Gflp+Zmt{k7OrjwEjfByu;2I;&&xYCiA8
zV=8(3ioA9G9FN646X=&huKvwzUGntex<s@mZzq$tu}%qCI3$64p7it>&x|3~wZ}Rq
z;!-DSEqim%n0)?plX<-4?GVRA=(kM5u?b0NN#4ehx2??CPkv9IY4(}dG*J6IJQ)F7
zxU);%>XNr7$lFUtlhKx*A3Mp>SKM(dKTwI)d6mr9sYL7()|5BOac6Wn-0oLkZZf@u
z@~DfMUy8o1sd;Uu#a$2TU3ZnCb9^~Q{Hj16y$pJ@|2ts;_dDJ*`!ui$*}p58nMQy6
zJC*bYFNN!`Qna0>MbQE7*gVqWiiH;Aoq3VBm%34hbhsEj$%A&hw*M-IOjwLCvbC!h
zbGXUYQ1bT5n__GjO8wCLLY#3agzf7>_y!iDj~#PFK7YfMM0)zzXW`z~3@j~AgUxmB
zQXWgg_*to#%^j&jtCQJtrq2g^?E5y*Kja_gV9+0=RYWr9N%XpmqQ1KovzKPl7t@t_
zO4Q;s&`WjX$rLF6aA#T0$5P5&tY+>~onc-?X$qM90AYM8hA+xMhxHj~aVP`rO*0T~
zk%6<D(@<HTij?$J6m!-%YJDnfvr|!YDh;ZgX($Xzg=|kM61g|(l9q!Z7W8c=Z{M^d
zTYItoBXf<La$rHvUGw}L?3tB^kJ<D}*q;N}ui2P4ARAYz>9O8Di}hy?GthF7<dlP3
zd+BlCi(c-}^YKzrj2qPV-uhGw@NwM2$F$gr{3Rddhcpn+)FACOA6xd4D=e9R#@acO
z_2?Y(b`*Jgk-XhW-o^<z7v9Ag@stuA<?C*aq%Nzt2!BR%Pv>3%?z7*ZPu>okP>7jN
z3Sg&Gz)Zw^+@6rnnJ;U5dMJwAs4aH+itD*wad}-9bZ2BC<l|Q~PtSxWd3%0LCKl;r
z;UWDM?dY}Nxa=#gE%=Hb)?eAb&BUEIspw1IE+B7HUrBLHLOtb*IG8;XAYo@T6z!tn
zu1BB2<WCqx)_PKN<f6kpnM2`-;(pBIeu0RN3_$d}Aao~N|IQ7f??w>zOM>vFHV6|x
z1z|jS>wPYW^Lu~X^6)|neOGKHZ-bj0kxtHT{>K#~dHcSVD<-@lFZa42f36Eshq>VD
zbQiqWeS^1`-{95kH>_b?(7y3Cj%B~ba=X{iJ<R#(FY>jG6ZSoHpq|GTpK~oSh`il!
z&J3N7awcl~5~qGX$MW8muugxD{##9PV3sNQZi+TZhS)1GK{I=`BP@+^wA~}*pLmF!
z7apRr@&Rt--@)MZcVOB1Hf&biLc@}qSZ%0}iMYUA)pLlPdKR}%okOqCyU5A83)J32
zXte=UBktqg-1`tMxsQM~_i<zQ15DI=fcFs(VO3)UuV&6#w;C}!*AP2B4bl491GwIJ
zfH_sy(P`8%B+lB$-rhc}D%pqTbNksN*$4L{`_OUd2}}(<4x<Ohv2F8F{PsJ7mIg=A
zefDYeIC=_ar=G&?^(X1Bb(}d}Cm{TC0+A2SFhl$d{vCK0*WN!s8y91Il|Dg_$4}`I
z@DxoFGw!n(W8UiLIBjo<h(6Zr8`<E3n=RhE+G4PcEpyOqabvG7M*U+87tWz`w>o3l
zV<+UTcmtdH9z3IY;qntN<h}NSzsd`>KRj`2nm0~31)#;J_ZUasCj9=0nB5V`)Y88)
zJQBO(BhjRd#_9<2DwTTi(?ZPQC708;<3lV0jbl;xFcwCcv1q+kh=1-0vCUCP-%$|`
zK4Rb1LQGwt1ntS&+|oE^vc$oCh!|_g+Y>Lvh$3%gv0~g4#o_gyIOH#g!<sR1DEuSB
z=PVI?<RW_K2+^DTyM8bRo5#oC%-k4ef5u?)ofw=oh(We_48pBq@X#d&d&6STpyrO*
zJ^_wd38+gD;NCj{x~c^@ohg8-fefA!i!S7?K6zV|5Q{ACKi^=D+-m~0^fSrXr$UJM
zGdR5vVlQv6Na(?`L<ECPBE$!Zn1vY!i&Js%KQD&G8!_{g*bko|!7TDtq9?&5_LJ{)
zks@%k6t`H{AM}vYqe#ZASs9M}k)fDctA5mxRZZqjyGn`R*3^HKw;9wsuOCJob3DDV
zbm<M<j@mRA_UJPt7_?4~rWOkJ*AzHzq`>=UoR@RXZugqK`v&eiU#2&e2|sok_muWf
zLw=C^%NN<7_Miu*H}f>!a{tJWJAb@?J2)QaKE<PT2>rB#%$2y5fHvj{xJC{)w&PwO
zd7DN)Td*Fsy~`P}1M{26+Zp8Tin&P`vyOAy`_x)-zc)9Px+~VWhv<b-P0r+R<POkE
z?&S_7YsgzEYux>ND{+A}?ww=Y=lDg&R+mF}UIlI+re|*Za+LDE_DBV?-<6~6-xBuX
zsnLz5w*Yw?zLtBLY31BOtY9`YeG1OgUw$6Ff9Ox(8dZS=vNoHyLGP$5qW;A$u@r6g
z(x-sDH9O1wl9^gOZ`45OT*SFKHA9^5-EUEhlt%hL^eo2oDaH79i`?}rM%k1iJUU#2
z4O5xZEo3$qS?fsF>hikJS~hkJHNA5B&?m6}O5V;OZ_~@t@PxcQaU>1JocDQdXU^eL
z*7fA=Q1;m~*YM)m!^j{R&%KgSL!S_1GWWiUefIZcGxwC0?7Js$f9jb7@4rsQ8~TaN
zA#Xc=;r_DTXE^L)UfEu<*gFL)%+nD4G#$fkrlVnLI-b2s!#ABYw5d(SndDT=y`GAa
zb*Y>Qr=s_aRP4TziY;E;XW7ZjY4SFMS)%86F?(}&E_!^)#kU@;jmg`RpE)>5-gdr1
z&vo{)AG5xm6r0T*z-&w$$*j_fugnL^!tOuW*l{ff2CK;KjjSE@^6B}Oj~3*rUve@0
z$=gHZ?INCE#HQRwA|LHHX<$U&z94U(k++kXIXYlw5pK?4F1mdYf~e<qonL}S9wi8k
zDWQf|gN7WkbaDYEMzinxnpxiM3XnR3ee~P{>K^Fp-=+Yq#xZ|%4E^NK<l_=~`;fdH
zrpmx5hYaR*WT5B4FGvje0=ppAwC}%Q?Wjzg%=`kAS6|TcE4}yWaU5R#1>Hubqx0=F
zbX}STld=^2J0TtocjWlBSca8Xq_Aw3;GH+Mkj^n!xZ@LIYNJqiEDFzVe8g7c_n0#E
zJw}aqk3MIEFu)@a_FV(9B{~38tNh_B@Pp%XUwoP1i}ure5pcj4=6ihUx9<amzYqGk
zdU4Ou6=$xya=zh$$3GmQ?%_gRlM5nPN1Z+Cf)>2wdtA_kyp8Jbf@ck{arl!n8u~b+
z{-+bVtg*w-$+lS9#}?g7Y;aX%1JMi{>~3#^DZi|6zq<tt+|4+{rq+nO{cZFVX1Y)C
zuGRt`d(81`ttq`+O>ty~DXM>);NuS?oN9QC4C!O+FMojP%@1(&Ki;mr!Wwx$BF5~+
zKJgyRShxrE>ASGYaW`Ik+JoLMM{wco5#;qg3*&ZY@cY6Q$a-C6Ug}kNJKe@b@-}wP
zeas(u7Z3XCBd3=>{rB{7E$lr0$$fy{<n7okM%=|Vg!ud^Oz3zDzO7H8Ma&8G?0f)8
zReNDOaUbT++=pMQ_Tlr6eauVRhwlUSVcFq*Xu3z{ZX|E79l`Q>NAR-s5$r2J2%{$l
z(Rk`0)NPNUV$}(ZCvOMeI)l^XZOWjt+!MNvpPo06v;01)4DTcU%p-_i8ev;^Bm8rT
zwYSX^T>NE*%(E64^wI)l?JVIl(+WR2*dlVFEuJm0Mf*`U*x76k-$zb3eAfy3QI05!
zb;QsDN9+<i!DO2UgyikSI!`=n@`QN@*1;M-96cL|_hSQb@0veO-u8#?)&Tfr1Y`ET
z5BMDwfx+>gsK5GzD}|qM_uD6Q=IwtkcRvt`b#YOcF*^o-3+d@a-cDx?IQF_24lW|h
zdn-boCJx1%rAElu=iMd7p66n;d?3b_m^duH7YEzjaacExbJ)IdFeh*A=$TdZio0PE
zF$j7Tg9BS)5I;5shsayKlQH;qj<35NgJkmdw?zzGAMm~j|0nEYaO7JI<~Gv%#F85O
zfdA&x0yvYmIpl4{q*$!F7>fh1d8>{^N)@km)YXp_;>&m;Hl3$-iu(H_n}j%R%|0{r
zg3|}HkF=3J=JV83CI25wXBm~{!gXD{yN(Siw%8rh1Vu#*M8U%D?ry}!4zLTb#Q>xm
zmF`B$?(T0r<NbbI14Z;4QJMGNdqoh1Dg<+DoMv>rp7^U<^z<6qCwB<p&h!x89tn{~
zG(@l2La00{lm-{X+(Ka-*InH5c^DDvrI$JyPD<TJ+&%w@c|2x&$0vN;sVt3H`G>=w
za9wk4FY(sFNjnDdyM%a~@lRPD;_U$O_Pu!f;LsyhinlMs+f?zkjC@5OT1msEOpZ_D
z?aC<;)S9YW?@aBU?X-8^9HBjZBooEkYP#_@yRD90KXp2NlP`Fdy!qm7aq-r$oVpY`
zMAK(NwDf1u%sUayIPtcjcw1DwEv6Z_?Zz0&Us7IVytK^O(l{%3>#m-4=Hju1GPs7t
zC@WsOX0d0acxtim4aa&b3s=0|_D+75-><n_`Zb&6ak<q}`2=;}aP`b<-Zs}fQ}{Vm
zA3Y<^Rd>4U&zPd;yu2KT+ezEfU)mPk@rurSOShMA*s!^P`nK|{Y<SKec~=sjKVzAR
zI?u$_yYKQ@dM}@k=kqDML%Hv^`t~gKTyD!}aDeo<A^LWr`n4++;NB`9hqyexUsoTO
zo>$WHt`~3Loz}Nc<}uMCPhAGNn77Mi*?}zL9%!Cjs6K{P%JWpt^u<i+ve(DrDBf=B
zqZzil=3DucQ^Z@ZUh+#3%K*J~RX#{bF?VN}W?u2uM7-U#QZw*w?cK%OLT6)ncUQa)
zlD8^Zd0nTKlPRy&&8h0Z*r%*eM`edP$4P&u{z%<|D^`xDzIrbYj*2I<Mm*=6$FqHh
zZo&o$#I#G`)}RE!_a*RccmkuO0ZN~gOq*6I^sr4~$MRH`+NU!1bSg!@$a5N&&e7W$
zMD0*N%uBJUr8>j<q^ZwddBEA~0GFP~-zfwC#>&HQnTh+*OzKLH?PHkDY;WCn56biV
zUjfM_3)EwnPr@5@sYc|JJ2j8Dw(4Vb%E9Gd4rzJnLvODhop$+**5|$6^E}-k^B8NN
zN12W}ga<z*Lp%D){@LvHeo9p*b)g*<ciZJqLHF`4MRVl&&*6Qur|Qp;HoLI$<P&tS
z{+y0yXfkEQ+X7eZ0i2T=d^nl%fywmLTgw$GEZ(d9;uGTVvt(L#PNv!6B+|<#QRh)K
z&BiOI_QYe&o8kQH5RSnse_Y(XsjVGxv0+{qdp*E2?k*n6aqzUe!;{!s4Di$5@a%P}
z8eC^wQR$w3T_gAPRlY28#pbXxn|ixo+|7l%vs`GC<;?jYXXe+EZ>I7Es%|^S$EoMA
zZE%jV70+OrDSgs(?JiF`64&u0*4Ivw_U5$i>u2%rd6uIc&r-O?S*D1$`^4Ly;_bF?
zr}#PWI2*;=Bh~b^U%c%p-p158N~1rAs9teDtw-)*SM}W-pSqhdUp6w(co#1Q?jpbZ
zE)qI#q4e;zbltX^2{x<A|GA2HM;Fk#=mHiGorm$(IW)UCn-`~MlRDTIuS?eCy|tp!
zT`P{h9?b#YF}x6O-{+2H?M7=Zez)e{uUV{kG?f<*r<1a4hI~&msP%ZB@}cJQWA1$I
zn&&EmXSQ-a=Muba0pIQ|px3AcEFU?GdW&b#W~SciaVj;7aoID88$6!e8RHmO#F`x&
z$1tLk7{APliT$jo;%UwD3gd8VGmbUJ<2aOTLqC5#>&7u&ydCmTv+$j9*cBbefuhsM
z9W;ZO+!+jbK2tl|xy%%AXBMpB?7sEv{=AWhZJN8q+g@|FbNl`pe)Zf;6|=3pUbT%f
zW;<vzc?bC$bZhqBMa?pM7!|jNwibIC5V=p;RSxW0aEOfUj+k6>Wb9Q(M*ekVfVni=
z4bQ8G=REx$UZ%Ww`{|{;!u9m+zL)7U^AhbEU*K1h>)3t0j=|I$%8+v-HRB!=FW=)~
zz&(medw#mkeXg~=&+_*7iIi`vNJ|e^?s>@V%09GO;Kxiu&8PCzOcHPZxCF8?Upl52
znj3@ssbk{L<0bwy)Q+~?gdpWr1d(JGM9VHgbk_~G=qu$~>ed?a#+O$QeCc54%YqKR
zq(1V&Aw*v7Xdk?*`%<cguU;eha;b+eV<-EvME<D6tG@ITZx@KS?fd!BWt1O3=KIO#
zq#g<Jwoes*wvAKP=YD_8#oIw|{n;qq{%Rb+Z-W4;HV<H@aey*00~mTp8Jo8?{}v78
zbNxUb%?qS!Odv)6KW}ri^NxxS=BG4Mp5pBm>kv2^f{S?j^j!$`P305VAIikP%8gkT
zrjDU7`KZE(dn{dcdN|(aG_y+IZ7l87)g8+B)~z-xSa;i*%F+BEYafKuHZTOgBx$n0
z29f_Gh>iYX3|bY&*JfedG?M?gW*GfmO5@!tj7`UtS0jzNZ8hzrM@R5N%<U=O7FrO2
zt+ZCn+9^AFu{2f7)NL>Jns}+lDOVjI&m$?P{qu^NQ8>00d%H)GX%fYxNl`339?jOH
z(c~S`%&Tt?`>jr=Ir1OsDKSX>4U?k@DiX~_@iwNBINMC!yW(xNTH<p}Wm|Q9L%sQL
z81mm6=}=yCd7+*muayV=iZ8j!ExMw--v8x2gzB#P_Zi*aK2s;@Gs?R><M}oDI>g((
z;%)D?&$!)FyT_K&&Xjtl&shNz#aoAs1=R2@z)&;p!Na-{wk)83UF{t|<TF>iy?!>I
zr}OgZDc<fAZ>Nj5mBriW;Cw8^+oj^|+5dgr56zQTI~Q+hq0TPK<=pdZ%%t<ZoSKc}
z;~Zs2<+8kOF0;j3Kk>GgczdT^9HYhCsu}7^S{q9_Wq_RTsk|-C{Ot?LpIkB)m;dor
z@7t#5xOm&^wtAX0zupmVgT>jBlhi>hzw+Uu>Ua=ud+Gi8{>tx)7lXAUsACz&TkVxw
zh_`m?#%Qh`bbxsKLA)(1-iE2K<+oTnt!zAP#oP4l@wBRwz-sZ<NxU5|-cB@1;9`_I
zzvOS(HzJYR`{i}%n#8ObNo=~3%#`D){5+J(iVdm6>-nCWN=5Oug0<$rury3k(^&XR
z``6JK^mvnjZ41r8z0^OmJ(EWU*@V2!rbTu>{^IS0;syK?Z|jJ+p3(!qwaer8$Xu!(
z$sx`;hrh~f^03b0)bMQnZOi7;-fTKKs{4FY4lgvru6zGf9b`{gs+;nlICYAy$Rz%V
z{7!9il%J$~`O>EtRF!Ah@F`A>pK{bEO&Pu^^e>m9Uf^VwEK25cQWE>CB{JZ30vjJD
z(9bmi=gM)EDjLU?FR@f@6H5u#Xf`Mt=9+k0{%azgETi<A{}HtZ1uA>qm-hoanY~K)
zwhnig-sv{3%WiVAiF_l~UFcUuUYQ0i)W~sR#Bo<@+;?S@n=8+6xiaXJ3+XqUacb&J
z^_|YVzwFG+2t9Gm6uW+fap@N*f9@Qyo6m7=%sEDhx2A<1Nfcun+d9(wgd={fPjax(
zNrpL`P#)+B?#@3>Lg^z+TX6`C4`EUM5c`fDln44CWqllozP^t=Z}t+Rul;ow@1e$&
z-JA{G$*MngEOy+ijH^vV<Zi@b(>g}Ft-~!~12yZeSGL9~vQ95&da>2qvtF&P`-SSi
zTtK113;5k*9>XWjA<JYsUGI<OgW*_;cN)#uE~6>mZL~TGtmx-4M%vl2%8(zct|Du7
zFO8+T>jXBvoXXiQGbk7_UD}N)q#90EPx(aFT%AbE2x}&{98K7q@f0{tq+Iw!G8<2z
zP^Ve!aUMqIHFM@)H^-%rHJ?_E<4?D7%ETJSljO0~-7=Kyw}V)BW)Sxg_r=@tTWlD<
z)<zyA8*1IPp|f~9ajn=}c^p-WjidTM8$7?-&~wo^dXJpM%$buo{cI{}OQ&<O@*=vm
zU&J|&MVhgeaNc1FGd3^Bb^B`hs@LFNWfQx@H}hx7HkuCFL6gs0nXqcBdIPsIy73lL
zA8*#*{VjYnw&T)xdk*LCVZX~k9+q_E!dXYW^k@3e>?CEko}|I~ll*SysLpODMs;wK
zZs0W2!%y+9{Ap_KJx$N<XE^ra9N%_1Q_uPuH+7Tf+3_YVhu@_A|DHZKIkxR4Lnq$i
zwaFb`INa5({XSb`9<q0W59w2UaFfQ@GcAx64T4yHA&7f#)zKi{rX7E%{8KM=yn7Sw
z=*@)`Z%SB5=jY=~v4@&xkNOhr;7g57zKqZuTU|f4CdG#+v37|b|41M9eD-0NylnsD
z?QZclLC?3@zEl=(jngy>E9Wz~q-NT>eq5ZVED$?Cs=U<=w~{~8#oNhy{S`~$&#Tw!
zkSHz9vUnS!J8%i{_PBU^zMni(YXjt~44}trWok=9HE>xVLna2X%r%Hlqon1!9jrX&
zVAg$>52-;29ma;xYHkRDd;V_+RmCWjZu>%|Q3@qdckwf$<aw_s->_>Kh0iGOqk**D
za~@OSnes*Kl$9ag2DxiK6>rOy3S)@w!0x^wT#_!!Uia_u@(j=WF8}n`V74!ifBLj;
z!{MRYO@@-Kw_&;wM=g<uTD;vO-X5MDf#VG2Kh9U~vS#4JSJdlNNPT|NG0#%2<kuD2
zGp~+hm@==5iMQr8)O*lcj1_N7nMP4myxk+t))H^)>+N+1@zyw+_$kq>njXzx?V9U|
zxBb2<+v<K4t<OgBxL1_&2&1U;E0TKemGPmRq|v?A19MsVNavN0B;K|i{2HSVuZVVg
zrHt8E^r`wv+2b#1{^ObQS(Qs5#y-=Y_wK?1-Z~Ue>xA}|e+nqm^%>t<JY$!5TT#6I
zt&eH2p+H#(1uPr!oQ2|T?M4Muu2VqP$9#(2)+~EU`^&lc^cbA4zY}@7apy6qQ6Afm
z<x+pAvgb?X%5#;gKIdFYf0l<PQ$1VJSqzO$<CboGdy1#g@V&e}uBklft?c=~i7cNU
zPpSEF+Ks8BFfA5KakI^;SXzs>Jv7_4*357DFNTfcZC;aDI<<(Uy?FaUyj^omI_{g=
zz00Q@w?i|oeoo2+v39;PG#%tuzM*+nAG2h?-s<+-Ry>|JUK+G%nxQodf78blzoKrF
zN3s0)8OsD^*WVhcZsLSEa+<}{aB@7$R>v#*I39DccejJGe5-1<eH_mx@z$kUvU0ss
zXt72)_faYOccgi5YbquG$J?c;H2fn5FH5GnNh*_f=+?PkJvZ@bMEsC1RXwSO;_Zya
znOrE9K^KDz3eCvGS-g#pd5S}Nj`Zre<kren=XWlhHM1_Cn#-`sIs9pm!*9>0gh%K`
zJTr@i*4Z>zlTG=x+3enx&5;$^O#YljNbzhe%VZN=F&m%QEM<yhvhwg#^%g(nP@=S8
zztTA;U06-sWlKp@Y_~m?_2R8%ofJOKPv(`Lm*VXx@%F!N@ye-Gr|1&(oyk`*uh|ot
zCO)Q_=2thPaIR=ie7&i(6FVL=XS?!$a)POy??Y~BPj&U&<D%tF4!gL}ZJdiTmt1&R
z)kS`H7a}Wb7hFy6tDxC6*_j4soGCxw8MQWxF<0n(@DkO=Twv41b8Ojr5xXwV#6>w%
z?a@U}HND6>`B6VjIEnY-lbrTCNsma)rzMV3>d$d3s-57lc-znV1gB;n!}Fzs?t7Y}
zdhX=Z3wvH?*=rYLkLxXaPW{+HUj5Bj&e}|yhTG}UYpYncRecLvI6P(}Ub{C?sn}YI
z`>f%b#|Dx&tiw`Fy+3{-&l_uxJ7gy1VyCg{?lcyjn?l%!QB+wnoXP`-GwHhpg-Q(}
z;O1cE-wozL&9U_BIGVEJ?J7NW#*d--r7`%_8$;)#BRD;67)M;KxMeq%s1_5bzhN9b
zpV>$oZNux@V=0|Hno2WAlf`H<nvTY!)M$QpvgE=%3o70mLWSJE(gL?8v_~txT`*^|
z=_nixY}n9r9InO3vfId<YOBqt=sJiqrw4JTU^FprtXVz7M*c1v3dGyjzpbP#9mA>K
zV;S?>h7see*=1(U%>y=+d@IiP9f$vm2@I(=NxObqjC)RI_h{{%cTOR{#Vq`6=Mo~`
z4*R>587-Ho`)nBnrB@JKu#|098>kV#fhysfSmUI-^P|o9inl++;}2Ul<2Y~?ODAt3
z>&y<kj_e|O!(K|}IuO#}ByQsE-$N(a{QM;K{yRaNb0^4teFFEeQxq?A8vi?|85eb$
z>W9yel6H>OtE7YWa3&?+S$}_>xv|iNy81ikdEz1~#Ns|d*Z66CotNV6dK*u|rg^ge
zqbJ6@v}+by-#9!}wx5T-c6xBTohOS=JY>l?FD~YI5w7R)8!sZ0y(v7zmsZ*zTO9G_
zxAw#p#oKCA#on>Ld{6ShD%eN*1s{UM+nwUA;~yVoRQpoDi7#!$+brc<ehKnrR)R05
z@_h*tZ*SJ{qxu@<hbH@}Q^gOX;{LQ3Z(Hs0$128Ow<&*?iMKrrrDYaxTQ`#4xwY=N
zNJq6afbXXRsQ6U3+)9De6>qO-9zQMKu5TL5PVx3~Tri711hcJ92*)ku!JVUAj{VY&
zUDT}GP@HWm=E@J%XL%@(tu^bO2<6mybu8})=X->*EQ>#;l{8ba;;m806B>%QAH>@R
zCBisdUYV-(L)c&%OzE&7A_Fwz{#Is3N3nfL2%DTV`v!-Qe=U@%{-M+e4&~dwFy#@-
z3oVbJZxi(mbx<a?_RD?6+lZB#asNtdH7}BgC6R0rZ@aCMp4m>l!d0SZCf=@V5`}TQ
zC>oooTVbLYx-yCr(lW<|L@~c)H0PwPN)T_Si?<u4W434#je|7M@pqzFZ52hKK~e1e
zA8&sv*ZEB(e>c76nc-_%c6hC>-PihisScSQ%B6q(ip0jRc(CCm3A(#3x%Zs>PtVkw
z^^6|cPuk2Yz-o2@0~ae_@MwW<{{@Vx|BO;b&lsR@zm+DZ(5eFFiOEhYwO7{Ty}W>R
zh6Tzz&&N6`pEb9%*L2M1!<>BVO!Im6G>=Cs^4J=YOP5W#+|wMp`lIfI&QFyc{gff%
zMm_QNOJo+4#;5C6mWHAFwfl*;YwD(QyqWH^Pqb4%svPO+@-4rK=A@qF*U{>Cjb?Nm
zWp<fquD=+Ar>8XC@-9z$ulJQwKD9wCTly<sev@{_bHrWgx$BC*Z%4%vE#6u#R34Xp
zY`y$c3&q>D;_Y9(-&nkDEZ)AfQRmWfb##iir<|p~(j9oeUn~cD#bIx&{qm1EX@TN7
zZWYfr-F%OWx5dTV7yII=nG(+f?|6b7;|Z9az<tfVzjNX#6dO;jGV_1lkH<w`;QGfB
z*zz`sn$lwh?oVN`Z7OMf)3EeS!{W8_fdkVpu9nW$)!IF$rsMN69bfwl^;U_S=1&RC
z)913dI?=`3wrz6hC;w49<6Me%%3(u|99)&_@!?4(_Xo?POeQV+W>VZNliXREl=zoP
z7xDIjcxx`+HmQ(J?3GLwPst?vR|XC{(pd8<71wBW1?)}5a#Jd1^HVt?-d<^u!stpV
zJeiwJZ#~;yCsAuaBIBeRI~baP;orx6`WeLSr$NL#(k^XS5a%0&tK%b7vwI-RN(7PE
zT=y;u^{9sVP*S|L5O0@@w|3%f%?U1e_i>?c2^ZS7bmqFv71kBLLb(f<ak0Bh>%*6b
zAAX4oeJ>Fyy^P<D^V%z(<3!d4D)}E}wsyPQnjWU|<>Pd%bduS_Pcm1$J+<&8<y}t_
z`uG@4ZXaaK$%7cJJ;=xI2grH8k9_32X|a<J4R#V`u#?T=ZINty)`_<n;_d05JGg1H
zndn9vF(|!39n`D%8L*Px;>4zq1$+rxz`gzAYL^8RX*XXz6YF?2U?FDp7ZRL1n@T1#
znH)Whtn1VGc6JK>??-WR@o-vvvEcqQ3;LHF!p$3Et>s|0l+fO|!)ToJRO~#O7Ntki
z`I!|z#*U`4!zd<q8Akm|!+8G49G@HJblz=FvCkHax@^J4!xjwPZ^7}WLkM;r$QYl#
zY}<zPq&->Jp(n2^^u)ep8?4K;q<QyN*j6>BNdF%4%$iW=wyAUl193MVMCEpa816iX
zX$J;j7cmHrD?_n(HWaIw=B)ZVjDtHZl(At(D|<8Mv>8sm)kr#@wq#dj8-72rrrYGP
z%C(W^YTiT+R+vau$BC3(IT3q@iP(hNvej=Q(Wj>JA$u-)+H-I0u#9~X%Ska=it(Q%
zw4At@t<H<7P<RPXQ<qS@@=_`eSt_0EGR-LBt<NfTaI9inC(XxJYv`B0ovgllD0|C+
zvzHH1@zNnYTOOw7n!^k^e?r+cr)XPJSs(*W6Vv+)uHtRxrL(l&aE?U(b2xoJPm1FO
zzP!I6&xi6u94_+KPgx@WUBW2#Dh2^}*wn>MS%z-7jJ`_)9uR%To!|cMG&S`gNW85Z
z>%p-|4^llnm}Kn9gn^!1y7Q1dJ`af+;>F)lUi3TW#g7;-9A|nfJKLM(wS2fWLTo+g
zLzDZ;kB;+^=TQ6=Z_5<-RbHg-we5W=8=-#0>pobV_2Kb;AEy6r&hGC+wu2uvb-%S$
z7PaYlKgKQdql}e5E5zGNnr~-|w`c20?_5uM=h^`r)*aWgdjM|p1Eg=&4L4Q#=Z^u5
z)r|YIbs)xT)a4kc&WQ7RJscZM@<(aE%7!p)R0xGM<GPB4M|Om;wWR(%w+`ipZo1x&
z<uxu5%H|{D?(tB}-IYbYCzOCWp>(nhrT(~3Zdj`4N;<32bwkxXs;;N}5OqB0=MD(w
zYfunY;%%yUyRfyo6vpU2EZ#oq6w18hP~~)oQlxAc)&GW3zO_8Q%FoQc_L!#ukNL0O
z6W#Tt!CLZ!%RM8Bo2xtSoJh>ABbAY@{t9J5hu%;>REa3;#oL<V?GsZmRJ<)I-Zqt<
zs@}sWyc46C@*|3nxlyd28O0+#c?+W$d|W?In&;8~X+9RSZx%%2peOcKB(-blR;x_h
zB~4#5R=l+t@S1&HUNhmD@@?z9!dbjEfAWGtp31@h@QmMy1sL2eAYwuRokmK_w6uWo
z;%${v@@`7kbW*$Ho1Y7)=c1W+MS;AJ%A?TR$$I9Bw}$ogab5B`-z=ZbRkYVEti9&j
zJiaAMUo2nGOz}1>ESEA>ayW5Uw?9wKXC@iMYi6>tNyngNI<-osQ6w&va#vINJUo?w
ze^S)LmPDvn_D^}1pOx2DU%ZVGYxnAzE#CGLZ#UHzU&ULqrRqKubKi>1)z3-8ofkve
z=NR0|DC<((TsB3z?V-|8brL&As{2s7=AqJcpV+7$)6CmuYb;r-l^-I`hUw!+iOpVm
z`*D_L;A63z6mLt4xAnx^v_5f!uGW1wT{(lg<qp>W<2&)T=}^tMbK_x8y!<5Uw-#$_
zos3uKbUe$i#FHfk$Hl~RNxUs`FP_zZ;>iwD&cl`@c6>>uu$^>9Q&Q>KMYHXVG`?!y
zuj`yf8R@3%7bzb?ymfn#&X`O+vFSJ!q_exMdelE=C^J2iGCQ>kmbbTQ|7_j{J!Rj&
zryReouGp|l0{*v)7H{A5$&?o^lS}<GDKs*ZszWkK8JLNg-q+706BoVzVckqdh_}IZ
zY19{Q*Tkf9N4y;+-o}f!*VIF|N`9>UpHh^6o6KM5L|*ANMWYJxwLRBs$PtftTPm1;
zz4bcTT(6VG+lPA1ytTf1={`v(k{n2f+krU0@TY~h2k+zDu;&iAU#DYJSKa$vI5ozF
z-QsPw_Q*Ct$|}5no}*spsdeBy#lz3xQuiz-El*+c{RH#VPS7$&UYj*1u#7vZ+uRZM
zIvwJo{T}X3+(Yd2J$!$$SJ_Z|b$YTBH+Jz|ewlfBJ6Y|!lfAAx`MP1Jv{gGPFWz2j
zzLTfo?RD{Xmb*QZ`fX+7qD>5l*@(%BjeNA-D81H3d?PmC+;tr*UDwj4)_m3$osX5z
z0$wazz}`#q+0b^r`gi9r!($Rdy~k6grX`JhN3ndwC_a9)AmZ3?b)XF=UA+C(!<^~$
z%~`W%2nE*%QzFw0lWzk#U*Cka)B8~@)`VieEwOrK$u#kHmd_|c>JF3Mav1L7?TE|f
zeEK(peV5FbKU#dPJ`n4nrff(z5es`T)}syXP1?v8)>{26t!Z1RH6_ZnVzXxpX((G!
z;Yur((Sr%!2T;G533s}iP%6fR+4W3`UNneAhe3SWGe|ms{`C1YfHKDhvU7{xSImr`
zmCV?<!;DA9!@1gc1P^*xvMk&hzae9I-ed@^_51mBc?1>QNAPp5C3aThNRF65ts2^O
zi?{9{C(*2<Ee&tZrrV4881Gv|<301a`gjqQrYu%x)na@LEupt~TdC4gn)X@B?J3J>
zBHreBuj1F2RdnvYn$7mBDSKcIgR{ikW_$3|efQPHLv%<vNUL23*xK%>@;~g=e{`Dr
z9i2FR%89X0oM`{ki38=%F-E-YAA6oQH!o1)!v$7!y-3T=7g=2B63J6sDZAkg6MgP5
z<v%yxiMNf#+j~2eC+2>iEz-&_`QeV6t%vd}JxE*QL1Q})+)jI_Yu=Ng;%&ow@~s(o
z@kPA7H`j~r$zF8N_F}hqyRN(sE2{Xge60`H@A<G*ye;(Ahtc9~C-HWtc>BAZFHfHM
z@JhVBB;JOKw^PL1IlX<@Sk{M9VsY79zU1hpJ7KPN%N~C6_WP-W!=HbpbfXn-tBJR1
zwF1~6-j-74cjtKloIItzM0t2ey$|4<Jij-@+g9T3{qR7-2L^E`T6*O(!Myw(Op)#(
z{OO`hkHsN8u?rzfvuUHsA@Zh%5Zfe#Ol69c5|0-;gi_5W6oZkPcg5R^;_Y{RKS=tj
z6J2$Str;pWwscoVL*yX{CaX^{v+f1)ZA%cI7lU!st+<_j%`e(F&pxWzHZhbl&qJxL
zUG@C>+9?kSr-RWWZts7@nrZ4;5N~%+c*1Pym%kp0plqzNAsa>#IXaSOy63JDZ@0@w
zbj4j+(BkcV-EcDpMA1R}<O<^LdGR*iQ~4R;QEc^(VuN_QP^{gf=b9c%eO&Jy+DD7G
z<;B~;yh!%zIVRp_o4lgk=vUl-`;sT^Ua>*n@2BstU-F85HDB>uysZ`{FVC^(*aavn
z;b8%P=M<1Cj;<bFK!$j0b)<mPPV()Dx4jn0dos9y&jYkK?p(m|$^}fhsXO8Ee9G_4
zr$KtYbO8C}HO=RAW#udV%cG5W+f=;m;-AN%4Z0_WYL^`F6u&N6yqu6B??JlqN7D&)
zNh8rVjW3#MtzuINbWNp_Ln_^kQyCwW#P&5wdi@fMxj{7H@?I5MpiD3A<?HCR#eZJW
zOc8I#iMO8GUoPISK8U^YDBCMbGgo~65~EBG?T5STCR<Out*yDXgL;MAYX7Vm_}2?%
ziD-XZYg;T+#M<`S9sg>e8?QLqRJ=72kN<Ovq0)&Mifg|1)GfV<uXN1PY=yh2_bpGo
z%hlppn;S<|TAXI2crvHOvuJlbT@T4`B<3#FW38MGU%jn;PE0-@&jjh7w_lIPR@oc1
zJra~FpGan$JVsNJ$e5ALovJDFA*QG+AcggXQ#t)Mh2HN|*i|Q0Is2&?ZA(?xZ93P|
z(}~cI*;2eMXdtcgwoKms&SG5OY+AR<=Ad>4DVMS+5}HZH9^$E<Vq$J$?@U&RtBr<d
zVxqbCy=LB2u@*h2^!{QMGMJK*PH_EndXGwD^QTmrgs0Nyg#JwCq_V+EeLX$Y{rV||
z1kL4Rq}dK^p}vlv(g<ezvprO|qp|)pF!ty90Us9rRj=G6&9aMq8LT_)5F20Km;_*V
z*Mkb%?{MAiIvK{-D7w>?oeNyZnCQalvC2B^?LwY-J3T<1HP_D*-1IzCDx4*<!zm7K
za-?tEF%p^{RMx8llcV<2J$OGw`t7H9uYI%~yO%WacJYW!82#AB&=uS1y>>e@-*4x`
z+3oUO$iupRCqq{5<jtC$gm2u*aq;$7FZou*+qGitjhWgL_uNFoy-Va#lh^dW6|}Cp
zLYmU~)a^fy=)ZGWzjPkC(eo(wdLBMS=Ci&2Jeo(%q43Q)RBkb!oPDOa<e9NLQA~{+
zPRXIe@o77XPfvz1wdPPxei}mO8$($6b1+{wnlaR32**1P=5j+b3eFB>Z81}I(f4Dz
z*mwVzCD*ep8MxY#vpK^Vk!_Aw@=)@sn{oV*DP3=vaBfR~?9=*aKiWsH98AehGohk*
zYarg5Hf}@syVmMOZ_U3?#<)~AX1Z4^?MGTrG^RP`)ta-bVRLN8H0SxQ=G^_%oISNV
zu=ha^>4FB4zgsulrv2$)+n>xCCgj);;!x-yF10eF{T?$qHych%3o{(c7}MlOQzrdv
z%$L?pxI{<VmhHu%3g)<uAH})@mNYG}WXy4E?BgeqZ8e#=i&HTCE}zlznG~#@$M5p<
znOt@;U&k+|t?OcC9xSFu$`aOyx6UkOmg`bJ<SypF!;AUZe=#->^mf`Z9yeRf{ZE^e
zr)Nh<(cQGZDgSYo-B>NKr|fQfD#zM$H0l&HeU5YR_HiceKhC>6M|u`@qC#^g={cR4
z;q9cX%+s`)c8QE*m$31_L>2M2T-<Fe4en5U(H$0vxzjD=?G|tMiML-TxzjO8%q{Oh
zL?aJ*Fg&o{tgOlX9?TPO+g*RigS!vuD&A%pc~P{n7aw|i(Qca;eZ<=e;%%+x-aN_j
zW=M`V&tko?&hW<Pi#H|z`mjU1Jy^__j_rL2DCxtGH{Mix;7wmwZ;a-7D|gVFyRkmh
z4E0s!jxQIC{W#|?k8Z3VR=VkyZRyXxe$pc<+p&bZ^i*m7{O%CIe-_GqJQ|>^ssIK`
zH}&hGx;JtHs9!#i7UhF@D~;E3zaaIa2h%_~kB7wCr2pxriph&yA%v>pZCTx3WAcL8
zrKdu^vfkcn4}CD0F8zbmm7v$!WrNtJ*Vk3;0@*rNSsv1BENmFa*T#YDQm*9X<AKWa
zmR8Fyh?47q<Sz;)(og!S>LL0Z$%|y8z4biJ(+A~))(*RDo^m9$E9<NIHblHVF-H9p
z>OguiNgc-W_a^LBw@}poI)K$LG)Vc&;_aKIns?7eGDy5FDc%O_hWoOAlrpQdBNlHP
z?Nk5Y?I`TT+i>3~^aX{X%F}39R}5|t#lF5#q%Ko-M)fGFm)0zt5y=Mew&jaRJjGY9
zj<4t+sO}i?_U3?Bgjwo#JL45+s;Faq#!ITCzrgVDb6R^m!{ClQo7y86SgRk+qJZz>
zZQ>z$c#es&+8-}jP{1}b-ENIF$F?hAnDnp{#M?FEZD0F*%qHfu>t`MdKS)<xNxmQP
zwv>1~At{d#zdXLK&!fC}>$fC_s^_yv8=t|B$LS2XoQ~tkRH`~8QD1l3qfL`&^*oXJ
zNpkLulK1t1=2O=w`Iz+jL;LvVmU@k%x9@iAb0*%F`Xzl?(`d@dFV+9NGOQ=Wuvxr~
zT^z%m9Wlgf-mRj$?(({^*jE=%i%3gVNIK*c>Agc@Fiwx5<o6iX{}wAss6)8Cx|IxL
z$rhtKd{VY%xV%^5?e$~QhR3V#Eg+g}^^^ft5XG!uWt%*SqH2n~)kezBR}NvJF49g-
zi09d^cq$&$UirxX_Qml`JsOXP-d6bkylr_go|k9D;;-@CyOzM(J&8EPC(?Ri62Z%p
znO`G?Q(h?q<)*N)P%7izq)=0t5f0TgV~$S6OSj)ev$fytt{<D4%xG)ruMCs9{Wpm-
zi&EL+n~HCAx;h{-_!f{s{@4u4b`xK_XYx;TuA_K+LX15dn8EU0X;^D-{qbEI4?d+4
z@G*@~zco82q|>5CIy<JySNkKC!fv`VUrFVt{<}%lQXx5+Jx7yumrSO8<7C46CGmXW
zV@672IHQ*zuCdaAT+!=vdmnkW<qZ^XH;K26=J@jWy_dXW>O6Jxz-Q|n`q<r62H$lK
zZ*ZpQ+$&V5dxb6~uMjv@xlBG6@z|#f{%hyCDBgxOKFcrpUhmIw;`WuJ*j_t8+rkdy
zMC_-E*M9MCzq+~iu<yhc95!sC>F`bJ7FQP1d3jR%YX^K>o>uoQczA4K|Nbr5Y}`U2
z<s=#I-bBWjP5j-mhOFnS)l0aN8NFBH(P||d_befz>Jm0STEypBi@5u8J|W`myhn3c
zS0JB@qx?88=dmJn9tL&h(%ya!J+IH<N4crwY&PZJWiy`NHDh6m!8B+)l=8tt@D*?8
z{5O=h;%!6!Aw-rOtnL^y^x}~odJGzvDVNrigz&!Htk4geFcaNbOi2?%ou8QUuwbxu
zVdkX955=LH8B<T1VDYdQ7uWaVzfs+kXV`&ugWB?;ZClFawxNP^8!nb^!@Hc;1ZEgZ
zdt=O%O2+K-XoXKfbGEc<PMSe8x^Heua*L*T_%@L)$OxAy4O#NQkV;+#jPGT@YS(tO
z9@>`?2K}*YY|6syK^$yu#)?B`xK|s_0P%KOMPs_uZbsMdjj5ArgxO>x-C~Uxb+r>&
ze|xd+hzUk6mi%_mtv6~CnVoELuRDRLJ>v-NJf7N`dylo5DF4<Z&epJ{+<9Bxx!E%M
zl`Y*r+iI_4%dIQpX#C!qhaNVBcbtrO_*`rcE?|&u<?fqTupnp)BOmM_wz)kWtnA5|
zXHPxx)-%8!n>Z(8-X5jU&EuGhx9`Q<3f~>oDe6ehFGuNrobbHlL`biT91(AOh_@!2
zuF&hBD-pG?(!%O0O|IWy%Do#*op}et2X|<g=|=SucX@yCE}?DiF<{U=I;1_|m~x*h
zzINyQ!G|1nd#Ky}Ln_qvB1Ozy*VBu{HD3HK<U@ElAGQ~G^H994E#8id@@8p(H^#d8
z-qGy3NBd^;dOj3%_o29W`$+n$a;?4jYNd}CZ|(EE`EkJq+ef+yzw@QMG*ox{%ENow
zhqrg6r*7d-$BzCq8SPKvUHNjw+bG>|XQ-!&*#Y=22;k?%0F1=jE1Bx!5N{V{Y1Z|Z
zU-&<9^mH(5#9QBXnn}xr5D*p2-*v%cBm`4AF_;}<?bOU*?ThrWY09KJ5KMDJ&8Z`U
zl+6@G`c-8%OUwOpOdxNY1+r6G=&R!GuXcf?UkJp_Oa7|mK?HRSqUMi4R!PqtC*EfE
z(au`D?IzxC5pSD_w{s)q`<)lglgHtlQl53|;g8AuEd8@QMw3hHe!DM%IpGn!sxHR%
zkK}~7adl@T<;B~X;_cy7&AsC2kET&P>!JIt?y<FP)jebv#pI>Rf}9>j;o(vAZx+R$
zI+}l_X<jAvj?p|^wY~1YKO>ot9?56vo{v9|RCYrS9d*a6`X-O~nd*OQs-Bt^(gr_J
z25#%;T)6X$BKx1=d+<5a4@-CKpge<4;;Ll<=EI~vUZJeOg9VHgZ$FB+tHj%_dzDdm
zMjGSK`PfJoRYkg}+sBl<zbKzio%7`-&cj^1-IFioCgk#0eonXGY^u3u;k_>dE@qJP
zEQ|H$l=;^+gIMu)aQSqcwTrB7^}lq97)I?<x8dw)GAF42Wv5<)EYa`1b`-Aq{Z7(8
zzV&eF%>KvS&0=ksGDaMtun=!8G~XsG4|BSAG>;Rb89z?lhITP*7jM^!x97#$V%1{R
zlPsOAc-t{KhJ_C`3*U>8&L@V>;;-AE80wUjc1mmx7jNf&h~Z6G3?6r4FwzWd78T70
z-)K6Ex1QqdmB1+d9@NR0tfz?dYirau-Z`Fmz2cPz5zjB(aHDpLy_$13io46Cp)%B-
zxw&TFD9yXGoZ@-<El%CJ@wgvMpq}oyt~V3#{iX~`aWim3GPd&BbQqmNE$<X&W~30O
zdD7{Ja<jD8b(xmRgfS^R9g>38w`86eB-8YZ_V4Pmi;7O7Q^{nu)XAW)KG$FWr7Jf$
zlTMv78CgG*79TQ5o|MkR%4x(ePi5FZ{dy*;)X-z_GL5K@%6&3TqnEzAPt?coQ*6JX
zxq5smhu0<(S}%om;_U?UWGYF^wySP3RTgW;pB2W^{`$8nPrl_5Z<=q{Yw<0*N%?wF
zX1y0yojfsE^?+YX?(^ZtU4B2jN%D&86ltc6`gXd>>Sp%O{2cdNouh81a&*1VQrrD3
zkIm0=_27A0H9D_sn{(t=I7dv#G5lvZuw?Z<=8W3Uip~4Ec4r4c>09a9WHZByi=pD}
zmtmV&)@l<M&ut+0^Ll<)UC+G4RnobvrMbgOlI&Mf;@eU(S}alq)B-BDpF!5Q>H0IB
zO7)GCahW-p#+7HPb3hpm1+%dDGmF@Uv#H#BE*pa8(kN#xxhLka(0Lwy&*#u5e-6dP
zTTiozBvlxM{Q^@uYp+{xo+(Pi<>R9v9Q7GO#GN6OKR$#xJBBcI#9&;{X&<}6jN#U1
zWcm(b)DBZl9`8#@=e|sN*H`zqe(F@{%TuGi^smyFiWN+0Bi?$f?9J`uUNm^ugJG+?
zb7Xycx*luCl*er;FW$Cq*OnitZJ2eajdBp%klUsauTC`L+v}!0*xsCOcbn6pY6Hex
zY|8HZrlhXd`^DS#K22D0zY#CY8ZtH1kZq9$yzXH@QI~eS8`zhNb^FtJTz^ha?$3&#
z{yhIWi1(cb6JcXU-Hd^p&gvjvKx6g?7!hM@#7s*gK6Gh9?&{_kYo~qkUmuF=wa5Ix
z;k5fWoJOlgQ09O+=VL4={MbU7z7~{78_Mun!>H#!47=$T?9ls;NUt4}JeGfhtr?hM
z&4RWzgz5E9GZxTOy!9Epko1i!7%ARv7H?Oyu*Yt+J$vTb(?YyGd0TUCpgq$b+w-Hl
zJx?y~r+(MNoXt9+Jdl$-ZR^Oo$xeK@;KUMFCyp3M$1L7%D=#lmh%1d6TxGk-RhEyt
z%F2sZnbGr_X6I|Pd2^jubuDx<yg`C;RgC)H!*ljMUVphS9rXi@C%AKZtvk!syA$K!
zNsC5a%JuT1vb7hfhrO73%A3;S?H2KNM7B2%#M<|JHM{Qe#>G|{(0Z&#dULhCH?~E*
zX<ybGcSCR8vb?!J*Bis9nsvq7o#L&}TlI6)QC5t2+fuxBI3gXjg}Q{Zv~wQhPX{-D
z8t3@Sf34?{`XR;Jfx6=^JQhIEEorC{0`&Ju`{x&d%0&nwO1B|nM|C0;5@U;K_AIUp
zj1gk(=wNcLDBDBKog=<pIvdOj&tP1|->y4?d2SHQ5%Jc#eh_yx-=^&f<lCq~`G5mi
zS4X*4nsJ?4D%;95kRCCC6q_Ez_6|Wfd{>ThN}y)QK-zo@<c@y6k7n%c?}OMNhF2=0
zY^(j@B!#KBH~*0`S00lh-u@Y>%qsN}oi~jjFCc;vRg{C#E|T5iZTf~ttWHT!?4b;1
z@phzmd)g?96!Er-nfAll7x!Nmg`1c&a)xwN7Ex5zk2`CY{iY{C?EPL}@0Z4T^{+@S
z=+`(W-j04I9tGzxMcSG8MR_za%wx*YT)wPShQ98zzqb@%XYq_9i<E<3OL_J&c{s~A
zd%v{UI;4Oi4GTCX-VPIQ%Zs-a)+hr|y#2CU-p-}@Y|`#H=czv4Rh}QswnL>u9?&UY
zw>)*dzD_6bc|0M3@mOf4%d(7DS9m-_H23`}8&A^PI1UuZXQQ3#K|AG1+>&qSuV#2h
zy}qax!MSpJU822ufoAt6@6^k&@Ckkm)gPhPDc!s2_h%7F5pnn0rbyi<BiUoE*ChE-
zm=LWDnrK4A+fea#jj=j}XT;DzbM6~;L@ci)Ur@zZE~Sd;zRKv*bL?geHO1S0;;xUl
zYNp+DkZ!%hG~+%f9Lw`e`J46er`pG`s7?&i_eYast8DDL%Hn(#g@Jf`O}u@Q6or-k
z+b!fD`kfKSsUF%X-;c#pPak9DM~b(-#oKA(?PGi0a?O>YVH!_AeSRFD#j&b%yu1+7
zR;^0FT)bW6sLZNjiR7$EBD)}oX0H<{=a)cSTmtQzCE{KxiP7On^vy`(gZ!=;;_bDP
zDMY<X=F-6=-IbEKm6(KA#T2?|M{gaMhJ7F1e?O-w$1{~?J5xCzUJu-!%AED7)D>_0
ziMPYV+VXn5O;dTBkVel-=@^Q);dRnDDc*Y2Q!lM<&?B6pnX25;^U6ZIP&*o9&H3Ed
zpWCi5&N>HTCjVf)YW`f%*AMF*-b}mY#fXPqeA?*68Dn+5lybxEqO#_P-k@0Z8ysDC
z9gptUaj1EnT5(rM9(RsXzt8d{Ls=B!Z87n7i+Fpt_gO+B&S02-TCZo0F?hWLb9*^3
ztI$4PP2I_+06V%g+D_1f&A2z*Oi9*pY2GGw8m(u^))}<yJcH7=monL76-f!JxVUyH
zea9_gUBnD-q}XzDlr7U9P15dp93Ey?Ol&@y8l5KdZR%t^Vyvk)aSRR4TCvj3iYRL<
zX2x1EX3t!{Kb?!s@p;ns&SQ*N>*X?q9_=TxYO*<}Di6eRrYVVoO<87RN|#~-G0|PM
zRpWu=E*{8@z=1?$4rI@zK}3HTq<<G?{26OTr*5WHozPdAFMTn+APsS3AD(L;+oGT+
z{UUmh=-gF%t+upo+FCjP#<(qLMR>~=`1Wf~;g!u;-mn>iy_?eJeq$Pk81cx^i0^id
zi2dA<4#yiZtw95PUm4QV$`H?9hV&k0h)KByWJNZ{>3kDr6m3Z12L=>zHNg6a0S|u~
zkQ-*e=xzpBTyBT|<lY2!YLDB#rtBQqjO6s@*ju&0ykQHz?rzSU)6Iy`E;+VPW1jjL
zF{z>vsm4Z>5pQp=ZBDV8Em#@Qp5_(%^5Urp14^69Q)a5X0COxIhf#267}IYI!(fjE
z)0PZnn%^*GoLb;^*Mb9Gr4>sbOSpJ@BFUPRHa4_>HJ^S{7BS@L5;nYD!uK;PNDbb?
z1MzlYOM51Yw?oBS(|z_BU$y6mpFP#Y+n1f~>Ggewc3S)K*y2Efg9Af<A7Dh~V|>z#
zJ5#)kseGDwZjMymc8U2fF46s`t6rO4)$59@%sF|Le&X#b@z(IQ{NUnkEAjT`yjv99
zzQLx^cL;uShbz7A((%My=J?#Db&I>~+jy6zwbh3-=00C0-6y%y1DYvs(#T8w3gYd6
zG;gL}_+Q7MH?yXAlRd$kUBmTs_vV0jYcAd<h__{)cyUy_qW^vWZ%=Qs9DTSt#z)!a
zJ}e&WLt!%?4)yb4Xki~7G>|57pfp)Kd?;e+&*B;W?D#I9@L+$sYtKA$v1VT7P7Rx+
zTkdgxHn-Kh+cSWbdzIbc9zes{LE6>HBP8B-%L`(**!S5?Gp~4S<q=G4<zigX3|!1U
zn5DCoxwT%q=Wm)-?Sq(Z7(|7;>K@vnnO2z^#hV7wpq6-R5Xfrrc1JI9^+_P!la(pg
zI*5*6H2;dXb;R4JpQZJv8N`C_LE2*~yQ`ddKU}?Ef%^5osF(0g2*uWg>d!umm~+Zr
zj|}6Tc-z+{f;~SY==3y#mo5KKqinZB^X@@$<4GiCW0b>OJc_n2BXQNN8k(TqUOg+)
zbhFgV+Wbc((T3V7myA;Wzh+>~!V@+7E%jE1itd`z_DcJ_F_M34<?VIL;f6G-Y9GVD
zK^~JQ<<KHgdK~wB297IW^J2}ZUBqAATZabaG14!eD=GOj)|?tyUAig9Jc2jp^4U>(
z*YSBw*UVb*OMagD(jbes&*l4caLMPPczb?kzViF>F%@rllTQ0T@qG4<$9Y`5JW27y
z_m5{{hj^lkh`Vp(p}8ro(X2RDev$X-XavdqBPc1RCRNqv=GhbK{CmXiLXT-*{t0K7
zJz;3z6Z#BQcILwf&JL1yS(>xCdaY6^P&r-VZA5MqmFh<maavudK6)$OHvTV$Ru;;_
zoU9IE@z!9L_^P+(w4YvoL_Ngw)wgsthN!zSbP;dAhsH2hye(E*dMVAg&BfbH@%Em0
zJGNa6N5$KN`_%O`KAJf3w*Cvf_VE*E9!2prLA(8C(fs#J^KV)8EFPEE>TWD!?@8yZ
zx8-#QzM#GIRPC>CYL>pN*?GBk*B1(;=kkzeSiG$#-deOyAZo7iDNiJDM7%Ycl*pio
ziNs4!HR4_ZpB&VkquujwgG3gFCecB>o$xV<zZT+PVj{<0Ci1j)5)&6CNxzfCq=ICY
zolK!jQW^(Zq;vRf8r}M*GD19c9h^#S=M)Y+NnyEm!@+vr_W`N=>YYlRQmH&{p__57
zGz<o$kt*Jf6>n|oB;j*8o?hbZu7Oc}JfQAu=@Zx1R9Dx`AWjwc=8Bsab+i}rx$RA+
zlQ$7@UL<5bWXEMMyu{ntpYGtc;VQK|IMcSSGmX1nA>#Q(-rc#t>WdeYrFely>1UYz
z<qQW0Dd)-KEa?xFtN7TF_|8WN=YX_W4lM8QKy1o>YKXT<Av;+#aSOFPHuL7%X1YGv
zi1*et_^n!o?U$v>cUn&IH8Z%-dIraCEyd}|DjJ$D!oTWl3I|VNrS)WT2T#Vw+?I+}
zZHaC;iAs&fQtQ-E(iRV8(1@Wl&oaj;*@78jt@k=BUXPXsXQ~y8<jd(Xc?@mmk5Qkt
zHIIYFu=@TeHWW6YY%>!kOfVsJq$#D%OxfSslt-ba^cHXHiMO){4CKwgfs`;ENThgc
zv(l8hnt>0C&4nuTr-@Bp%y;!A>`-5>M)Z*$rkCy!J-K+W8-KU9WBA+FER1Z;o3+L)
zI@5wdN1F3Fpc!Z9HRD|4W)#(4xW8Rv3WANO(!+?iyBpEwOGB#6Zph?}23);u$S89|
zPK&n>O$~|cXvn!HhJ2`Fs9TQ#+r-<qR}A>G%Rt#B22@#SK<BOo<W(@>hrI#$k&QUL
zvnjR57;)0Th@zE@XyMoh^OT0T4Qs&o`VE+-UwirkBbxti#F+aH*%jG<^;H^?^|dh(
z(>hY7TX!73_M~QyJ{-F=KppfZY&tnm9`!-2=xD~T3qxqJTv}+~VK~`nwiR!Gz8H?F
z+b9B$T2ev#X!G7Slsh?&{iSSayvmlr%cqePKTY}23phD>1-1`Yaok}8>!UWaqs$hn
zC~su9<rY3zZ=rFD9noKR;42MXg`0LHOts_r?HzpC?7*O%hv_jzv+fB;YKXTppE|On
zj1vQIon!BcOVk|gO42M>zVC8n^<7u}+t<w7=qlBduJUx`HO`*9fsNM<o^HCs)Y@*?
zwshn73^#RR-R0lYyEqlOPZ9C<t$4e&`~&H{9&o#dx4iLQnC|dm_-Zd+zkEpO(}zsT
zme#6{-rwE}i=keW*d=}QJ!!5&z1W-PMPD(udKYi@inr58`cT2#hneE-szE-q6mQMM
z+rJfkIAP+Wzr#K>8mrvL!+vbf^W(dtAOBkTakZTvAC~x0d9UW;EM<9!#ardiJ#bul
z=DPvhO7JH^nybGn1K2k_kQZ;{)78eNj(EGrJ($t^loM?iOx>-)xG0yZ!u(*Winph9
zlie%cj(ni}t<7TXQ0<R3*BaLpZ?#kIC*FSQ7RbFv@+{bDcU>ol8p;UCO43d_PhCag
z?Xn8`b$SF5s9UoC|MXiog86+q2=)eHGdhscCV>>{5y+8L&B|i^pB!}#C#p|Ty!}tS
zO%iYGtcs+fU8Fq9k#xGJ{;|kN2E0@LqwbP{V(fIW_MD#H`t~c$tk%DD_bnl9lz2O)
zP!wKK+9BVGB;-IOW#)^|_L`AbN0P8SlJgfMsTz~5uen*2yCLoCv?OYXw?$`*v%1gL
z6mQ#!w-p=YOAC@m%Zu{x%+MXSeLn8{^2iLyQpZXLFHEG#DV9#Z66xG*kj{-J%0jBA
z-kc%1n2NVqcKNK*&zsOfyK3bCdJK3f?{E^8{&$!B7l%`P9FHf;qxxN1qWn14J&(if
zSR8G3Kc-tmxOz;(xws;X+*Zn|cag8kDvV}vVLW`HOwEdq`9Agur{s_BDc&yB43%G4
zpQkeF9v&a5K7vTviMO|=MKS$$Bx4Unaa;_4lc0Ta{TQspTP35D*IhGiPxS|jw}Wlv
z(Y01?NVOQAyTqt#GKSLE<!LS)&5lyh+>eUp#Yyd%-F44hA47M|$kCc_Hz^CLSbP+H
z+@i429pKH;Nc#7S;?)M}#cxEhU}!Y$#M>eNqUlmoJ;=k=nJC6@l_qO|v{whj+ms;n
z7<<R4S2u<$4%%VQjuA&=SW_;BMyHfvQYM~0vlH-HoWS%o3Dg&Fdx^Jwr2GAILpfG^
z6By7zxmE9CX;&$Zj@9ExtsBR)<#7~#l!%>pTS>f~Ju->aTatPHBAL!5QgG`mpX{MD
zswzvPW@qh-l>-&lB$*A3l6h;by>kbB{Ln<E8YQx~NFoQ{YL3oK0R8zThsUw_OdKIs
zWAQQ8{kcRU9!Za|+7rTwHoojCBrS$sXa85yo8P~^xUMYPmy^A?knX{7H+R-vkzTX+
zefg>H(kAgHzlynV_SOZSJXB`WowKx`f0p{mr<GlJiua*Td>Ma=X5wur+cWIEb&9Sd
zl~J(g1ev`LV^jSA)mrT3@3~$4>%3jP@>{txbt{KnZ(+vVEd;u6rvIQdRC>3JmD0^r
zOIkv0@z&zpLPoq<rkQUBt(Q*W--*c#>~2fDZxfV-H$lGc@szwjj%KH<xSwdw>-U41
zwSBO%C<c?eU?@Z84Iy*uP!da9sMB~j<GWZ=y_pp$Hdeg6HW>3#gSq<EjLL7!DC}oO
z?{+41DP+P5BQbZa301_;MgvW8Ds0LS2NR0_9Kh_21E^~~00Yndcy#F}ZuO(`i+;)t
z?1xcmKOUL)#nW2zt)2GA+xrp{+?CfIy7KD3E<7yWiEsTn>bBaBt{qy^JGlja|7*e8
zTFuGd+l<E6%`j`$j0WqPk~pC;T`i2b+DOd(){y-(8?r;Q>*n+Z>IpEUj(A(Ul_4&@
z4GCB$#)`M=su;3Vyq(vm9<{31$9AUy{q`F0V1)r+IvdckQ*9O{)njBtBUZ-Nk><#N
zNwJ2si7~|4-H?`_>XZDkKFe|q3GQjAU6COjLiI7b4e<?YKoiR*$~)A~dS^T3thVEP
z*Y=oSYOgN7_VOrpWBbSMG_2pBm7b>3SC}$>vpIo2!_>t*jP_M6=o(^9&5Gt68#Riq
zw})~yxj)_`Ou1igBt4p0tApB_{4O@kGM>iiYBQMXF;kwVCA^m>I`GIQ7SG#^k=<ra
zUD!-z@wVuJ9b}xd<I!n5R*bP@mE8_Xmfuf7lS6cy<jDKujx@gQ$RzRhZaF8Mx;Sxv
z*C|F!y2R3HuDlm-qs7|_ldsad(RKNYuXAPUb#83D!5;Cp!p9rRw!cZo$G7+tbBhLf
zx2QYoHaAY*reo<l^j>p^lE?HoJWzK00}3^ASHGG&J$JeDet|n*pSp87+f!Manq|e?
zcH*s>c)Rzmm-3YKZ!X=7BX7OrMe*i(DR0g`^I}z&7q$O+QDu-f;}3Wf7varV>9KCq
z^`XH8Up)6{=UhfP84-SrI_^hyU2(UiA2k;H5n}7d>H&W2oG5McQe{K#_v8PObk=cE
z=il45b9b#(Vum!abrJhAP!X^bYwfPJT@%G_v9LSo?(P&tu)AHm+vooNo<BaXK@f%+
zV6OK$=RWs+_ojTB3zY3A-Zq_|L~Zfy*0W?9%bUV$hU(0@shEhjOIC@$%TsBVmO?@`
z@%McS<7&!t>y>!8KUrspWUe|U)6YWLM%BdZn@P;Lt)1ZC^tN3x`C@O00m=03B~M9x
zd`E{wA{!)Pp<4VuPtBcOl9}8}-k8l(_}y0ZwdTINn`_UZOR~Jc^fk*SlcT-EQ~gt@
zHBLE7`*aq#l*{AqxpW&Vwyw@2Sl7CZd1Uy?3*(MBDb3T*BK1r!O1mruztC-qr+Lf~
zZ~KV1V|2Cnna6GMwv2dd6mQ$y%EkMHYQ=GRSc|vgHY#7Cb2j%+D4$VmTYR(#k5<=6
zIdhE$s(0G8xWOT5Rn1G?;da=4W<Gtu!|u`<S9wV2#s?JMkYDkYTO9b~2G>g7pi&8G
zXPREeL^Y51p=;FMdYR0@s&(dS-=%sXA3NW~QoOD9Odbkr?_l4qNV|E;2gsK<#e@RY
zrsr7@mcyumY|OW16Hz0J1kJjKHqXRMJpiu{>0Fy9|7-DfSaJs5<CN!Iltq*@-ft3f
z_@!QRWv5*2*XOdVOde~8<q<n7k5-FyZqa$Ap{~K&SDtX<JkKLE<F>d!Sb9Dq+e-5!
z-n#16t;Yoh|8+rq_6yvvaDldI`Ly#>4wH0N1>)^S@piGbyhUu3(Iov-zgGE_*0n%y
zdnvzZ!BBZ!2jsCsXBngJ&q6)@qw4qf5pSo6w?`i-1No;iC{$ao*L{tguN`26=Eei%
z73q0_WqN<vY8QCxsJeUE1vW<IGq_OxpYK!~U%0^c#s!45lP9xy+wk8)y0?{2)wN4(
z9#hCX@pe&*0(sBNf8*0dHdazkUcB`aZ~s;wJ2#_<PU5YX$z|G#x0iD+Q(>~Qlf>H%
zt*_F)>@~Jkxkfqh)<nGBF5X@&bp@->m)TfY#Gu3?f{gMMDOSYy@$#}%-y!;T4x2Nx
zm5-Rs{_5Ebjn2?HG>u5@`Mq_CXH)%ng3d+|^;;NQgF}_~63T^Y5gZq9uapiWq^z`?
zKZ4|c89?9#AIyiIp>4`h;>Pb|Mu)w0E4!EY^IlBe?8TVvdni9`525qCSeJj0dFKxE
zE9fAR0}n`JyjSPjourP~#K}G${Aep*MDg~^QV%@E+w}>{m|c6i`slNWO`gK5F7hSR
zwYk|O=G2@-mmw3m-$JalnLzZ;@hmlrWq{>qDv7tF#M@EBhx0vnn7kN<Q2vcGL6ux^
z{l|qCgIt&<-cA_jiiN?M=Np~a5(67DT5;fDE8G{fV(FxII3~6u=Se%J$F#$JS3B0G
zJMi_01NB?BWcK8iw65Kb4hP$DZckU%dvqoEOBePn>%!dOU2uzW#8bSjThxi&(>n3T
zv5sV}>%h`39awAZ#9Xm<^RiC3bnU1NvzE-S(UKp-9hec?g8#1krTp?H3@=xoC7tXz
zU}`5np?Z9DtH%!U*6e3p@?X^DOV3(#5pO*oTM^@Ig}JvSLuMLbyOHrBM(rhAFz1jJ
z_e`y5@ye1imn=D6!Gc5gjeKrs<VvK0H}ecs{lh@)YIANjHfKP-IVBq#ST@_hh;Rdo
z?TlPEwcvD3OU_)e)bnA<_^+0nEM=uy)Qa&5R@k~((RY;k(S7UiY*1ZV_pV2~D-ATO
zXhMaCP3iZmDcJ^l^5a@ic)10qu0Wq1Jy^A|C-cPHE46yzT+oA?H+o?B+yl2S{n%Zm
z2RS8OC}-_RfAQ9#(GX&f48bg3b6w2<&Qu>t$Y=R~OdikJ&U3WWvIrYxK6tfSOtry_
z)t_7;f83R{w(+3ZYY!T>UPFQ9TE+}ehEmDR++jB@$Lz*w{ccPHcJsE37x7EHc)rey
z6@MJj?C}`B<BpNv>;%Q8p5@jZZ+5lyWwLnN>XE!6N2$hr=f|=k{=Ci#pk77*?Y;)k
zV00kP;;o-u5W9N?VJhCbZwcqnvM_$n2xrB$a2!5_({gA8qbEl2Sbt|WrJ{M4AI);t
z7;ME`J8y9}HHKQ3^h^0HhKHuHyjPE@rFgsdS`3fGTW9fhw0N7A6HE0{ar`ikV|b@H
zR;cHEYGfSA`EgwD9#7Th@yvQ4pUC&|%&DC~+fE705^od4+X>=rFV(w|(TSXXkVu7L
zNw|t@tNNyJ;IVe-HOqaVTDH%?RPDT`lHXSQKmjRi5bqAUsUCGm;lG#3B<@YdY=-=+
z#M|%Y@{TN>Ob5-0or|Q2E+dU{?PRR#C*#>!nF(u?Sn8WVUwu!tUM5gZygji)z8&K2
z#<D3~sgS~w@+tI4Q_tB!U#GEhe8t;~UzEL|TD;1tM1HJDB;6yK-l~mzmR4qPE$O6V
zQ}Di}j0v3u##WU-Wked)Kc^|bC!HRDXA+p3hpWF>J3f!Y>Z6)p(yUjt?Xsm}%&1)6
z{w@B@P~A5{p3Rq)+xsd<z6iPauE}AMvXEMOWierGHnAIv2(5WdzrWY0diR=Y*&9rY
zy(Ldb<-m=QKZ10`c3q@>`b{}_(%hT~&ZVa@m%|HkC=$ml%gW2yQ@vKV0&LzYKSwj!
zz2fcniI<pasy=9(`kdwTTK(n{gZ5qGfyX6&6K~%}>Z(@2QSG=Es8_r&C7YnF*%X<G
zrPnj5(Jzfu>B%33sVDzY-q_{R7?F^{jYgTwsxPhC=o~`gbIAQ&c}yL1S-K*ZyQ*=E
z>g7?#LH;50H49e{y|2zJRy(Aj($3VNJ<>5Xkr&1o`C^E*Nv-9DQAfKx(l8GcZ-<Gu
zBg$Q%f_OXKTbFp-<Li0-?wsdahL|emK3guW_!QO46VJ16xU^}*&NEBA4G?dKRn6m;
z&NP8K|2RlDJ#CbD+xI;8G|#=Ob48!a(rmdZFH!aLe$9e^_Ls-T)(fn9ApNpDvj6@w
zAJn6`=aWZ5fjmE-iRIZBaH?Ox{bGguR1NDrqmcIE?fk+^^s&6eO6?oHFfX8!cA0L~
zx<DtRvMU;D{v0RWm+}=X#M?7h#mkjNJR4iY?IoJ?DrdoBw7e+9+e&4w(dm+Qi)^k^
zQ#$ZrRmA1zMbfgXZoaF$0QL9Jo+zZ#$3pIkx1QQxOK+M<#<pbkC(E}+zBiX&#$uWp
zMdg$zMsJM7qkTB@tEeyL5X2PkP@F~u(>@@G`rbieRuCTD{P8yRrA(=F)SjW7mJP=!
zKj{z?UhSiM)4kL$A>Jl?$pgoWy5ep4)IIz@b`O#AcWU>K7hY@k(9eH26JL6=cf)4B
z4_U{hx*jb4<j%G}bJ<#b5%t6W<;{m_9Pct!eh1Sz7C(h<KUL3k8HeweG0LGE!(RCq
z-C8?_3KkRSx@ZW0pK+u2PG^Gub7A%{R|ej7<;f#gnxAlG`a@^FjB(~hH)nG9c9-{U
zckN3x=EPtZ8i=>HdTARf-VQv_ia~>0QPQlHdXO#UWz~{!I|p1-TW~$XLAh!UEO%&0
z?{Y1$8~zt*bN^yc|CSVr!zs27w5s7ik4oZZX$S7yXhCp*7_&xK^A>#l-hzxy;_bD1
zM1<C&;f-1hon2f09<_OOtqx5?>v7_-oqSO2I3%r8DHA)^HnF39bvyZ$)uXp~dsqGB
z(JyVeXK9VeN-HW#S6sG^6)SpM(sO~4`7?~%pKYX;uTf__3p!M>WZ5oD{JvNa8fPKT
zQZe|tQ8}mv+@lPn^)fKIvVqG>%^7B5&YOqk6gD%^YmR|!;_Ynl)?2(iB;LLhZ~ql<
zXNk85#oH3@R-7%eV8vGpY#ptX!6DsMwk1sss+W&hG4@j(7F4fGW6QeO71W`zOI@65
zHek<)2K<U`hEwxy9GT+8zQP_XJKuwJ@izTU4{XHSiyvL7mh8+s*Upsv-k0C3<!g0#
z2w&rdD2Hwcr~e$vG4b}}r16Z_&hEAYi?~&8vGf9qX*X^$oqH_D*kL7Ctv#6e%0qhH
zHF|AXORrh$Dfzc_RJ*sbvBx&vecFcgCr@Q6?9lK34kq>6$?dgAsVd$EG(L$})G2~0
zoT0S*&La1oBWA2OLyP%v-Ora(#r!EH-nI-4AU;99ke>ob92v-ly@9N?4&rg&AX0V&
zaa(^ohRh%yYzkpPBWasQhSBp*7!`U)FhRVXG9j9KMbWhB6T^)4+9wikf2G7QSiJrD
zB!&hxV);kB9rh@OL)z<o`cd_5!&q*tk7a~-yYy!)cdN#6x=);Z*y6O89!I=(e9h!p
z_4Q6XhwjD`E8cq6OrUAo1d@iU?iFvxsJ|R}OK<m0q=EMQyoV++UTh0d@3(Dw3XR0u
zQG-%B&^(p6n#%iaBi+=F6#ZSK5Fy?+icZGPD;dkB%G8~n%tP_Co>?-ZO8qa}J_*k_
zd2f7Hzxrbm%^oB%?XP6EuS;T{cx!z+p6k=(<&mY`r)X&}3*-3hP8<jG;>3?Qw(IS3
z58|-=97iR2c0K%4xrBcw(tdIxBgNZ-k;$C*PR2h@_xDYEQitL>tiGI?e7hQ*iPIjh
zX4K(v_!N^@SFL1X=B1IjBn|(iY4jSL&giz<7p|Aij7_P;tWH&Ck@VMNGTD1E6F-Yg
z%C5^G(<B||!71$Mog&{Q?J2L$Ca0@>r>>{bDm;aYu9{{4nZ~lp(zHq|T+I4D!!vI2
zAnzvAd)(wzyIi`e|2nE<uJTfHsPZb4ik9irDy|%ig*vlzQ>GNkKHi<eY%%vs4b2tz
z708cBSqj0~IG4;K#ywM+A(@Qtp26!T>B^^1XN<l4z5L{Z>6^~?!P-+6Z|8`&_f3>#
zbvTu>uTqualtz2;HcY%7(kzGYy*cWO%m1rfE*nwy!*bPK;_XN)ojcrg2AP?s=SF+S
ztMZg@u3f9?=ZW8Yo+tmvKcs0s-NoB;;%$aJrbDl(-hGo#<q!EhQ5~8Rk<UTRVmper
zHNQyXE#BT3E3XE5cJ+R*`S961LeAvT{9GObyw$H({~=^@9@om}5i>WB86TyWF0Oqn
z)w$)wTjv?dVH9ttxa89%C!aER@_Dx}pX*2RiBSH;rht5YRMzbCh`hySD?7P_zGg@+
z#XWO*wNgECotJtx$^E~_wn_7bo9#3wFxNhl#U(x$6)=BS0ka#Z_7rb}4HuN>DDJ9f
zT~Fs4>+~XgRi}=cCB}-k9cCBFf1rq_o8{N6Id0^Kt8DyyiQ(E2-gL5%*B^>#V<==s
zL;-We+fU-{dYz@}$YX0+WG>0p8Ki0deYJT!)t*Q5`dSou9i#aw)+UBWF?Mq#p6^2$
z{3MvQ9Rv8YOd#uX0$92>h<RtUPg~xH=3CEVHRmk09ZvAD++kLZIK<7``<Zuk5B^nm
zlQw!6@4oD$-ygf^yh+*P;%xk*-CW<Yo2!#{Q@`47>gDfJ-M5{08#Z#&=s~5{?)<f6
z0rA1^Ox!S&jxA<#?5+Io+fAkWA5%GBd;%3-j%3@C5p>-@nv3G?dhym*yluBm`<>b2
zq{kSdoegK!y>%hp%vHQ|rOFjou0^;~?}{_S4|b!?+-_`AKe)@wF6y&7GDdqdAJ#M^
zcAp(bj%f~APJP;6weW3Ui{BG#k~5+vX5DOPm}*U}1Z(cRw#KbOLl(cX=f$wUunuX)
zV(rm{6*X1fxdWbJZRBqbI3I68>qd>Jc%&g^rZi;WqMFj-*-GbRi%0RA>^xSBS_QSV
zH(#6W^J<fMR{YibMjL7=N3o_nUFy=cf*tM+>{!~S7SA@=@@j`I+ahgw_|Y1lG)wM<
zSaRTuC1z_asnpAo{qu|jO*fL)->97%BS#vkZ~WOp`UFeD-&-&&#)6?$EodU%PIzO$
zBgVkh-Ub?ow-1+?<M-2yM?cKyvdk36)u!@_Hf3RyDFY)-X%}Y7L0!c{O*ve~jAiG{
zc+k@z-G-62x{rs=H21A!MNyn3HM&Uad{WHaV$G*4^`fI~2u-TPZRfgr?i=X$r2(;t
z%}DRijS|*Q80I)K-O7=gwvIfi<w%Q>UC7wok@mrzINzl+1xsAG(z*xB#bt}hJ*7YG
zO->ihdjmBG_%?{>$`hIDH=8wK^QgE~UQ{g?bAI(wD*xk7@ty9ZR9wNhMr-K2a}6D&
zTQ&*N4$+=X4E?x?OFcGY*JztwpSN>x(spISc=CPN9(m#H!LQp9CSE<w5%Kn7$1~EM
zoMq3ovve>!NAhiNBChyz{kA`;t^x9B55Pja?K3!#Jvo8=XeupL{~+wd+ZEz%WAS#u
zh7dNIh0=3ED07`6*p(WA;X?$^|BU8jd^B&qM)Rd|41v>Pm^edTk<v^!iMNUOHP5{*
z#;Ud*bvcHa;_Z(Hu_TMP71Cqb_9m7$-(vCqp_<oH{>(k%=p}vyD~ssU?Rf3Y$E*Jv
zFTH32Hh&~wul{nnQvwavh+(R4-TEX_IVq8{;_ccGNz_@JOw~;(#EQ4WJ4%P#O1)$o
z^_ESv$FouUKkt%p$XC7UCvDVj>7HiFt4ce7Axo5_V5(ee@zymZiN~SJ<$WnFw7ji$
z4ou>#`oqtr%X6}A5=U(kc{@N^iJ7rf6mMtf`|X}1Pw8Cg!c+q+x)H~-`*D1hFUZ~>
zam;BI!+=9kxU7xB#yyI!^&;6E7EVDxI6LP?@=s(u2}h+HJ0;ehj^pCVIPAmX*z+@v
z(O=`J`zek?TjeXFYvIOdD!WEAuTL})@1mKc=h@8=OP?9isr(+p_<B0`^ozkn=bLo{
zrJe37FUmG4l=juR;cg-=gS79g8a^&pIxnX*nzXsWk-*!`sd1mBPp@+<?}~i=ig?pC
zm$1Kbcr`Mcb9&x)?2hBEwBzR=rYJ*Nb-DJVwzrqxmd-HYc`5v6m5N)pbh<B1=fxy>
zhBV2f>XJ<2#oPT(8Eg`79d!oiEMB)8oJi?3J)RFbmpqk6idDS4eYEQ=U7N$BR2qKM
z&&)oP>w~h`Q$9yJh#ab3%%QS)`?7v6d*mImsZlOdN9QX4T1*me-NoB55A7Ule*I=k
z9@F;7Cv%r{Wewy#<*LkwRr%bxlFu^nHuSZ2bl>Mw{cS#j{t<iCCvGm@e*1i$l#A!t
zx$=J*#+rl2$^$zpSH3a1I+sdUF0Ixw?Q>1Ik;}zj^2AoZp~YL>pLz`q#M_C|h%an?
zp8P2Jd{tAnM0P$8Z|8GkmwMBl%AQD&{#vY#9g$0>>f=j>JVJGD$WtHTlpfFAV>!g-
zsh7|=S6O*^tP*dBwUBPEV;(2^YQCVe#I@S;&GOT{R}Ahb-gd92{o<Mz`E$iZdHWZz
zb-d=i{}kckR>W_ui}=}9HEt>8CuGS#GFSbGa`KSOksey}So^v1H+H(n;~L6UdVE2;
zYt6Wy<P({x9(sjr9_~`_>5O&{9!7BUMJRPw1T(2p0CoKQnSehI&-_Tv)&60mAFEGk
zHdQ>3xPSnDNN3!vu@Cbncyp!ONq(juV)t=nl+KoSUjIWx9^a|m1W)?*^Tht(cI8WK
zr~3@)pR%{n^!;{9?%S>0nBDv?-kvpAZ7bfE5^wLX+sL2h9$aqa&W?Tyn4UO~E$61Q
zsl!Yh?@Xgs)Of0RjOXvzksKd5n3Yv~VdynbyQ722t>{L;a5p~hbYoVa8}%9vp~@y_
z<z%_=Yo80FTwU0$J)9aXTycHp#HW9|P_}F*wts0w{bQ|2vuMSysg1ZlS^QjV%e<d9
zbcnJcJ<LWLT^p)3w$V<o4KAmx@l3JiS-drij#;z2wXJl|4M^QupK+R{wwhH}ewB5Z
zu(u8uFVrT#U2PuN)~2<8E$j!^WWr@z3>mh(OSI+l5nJX&+hWzkmUjCrS=7anEln-4
zYiP-T<*c}{(^h);nmm-YY40i<4zv<y*V&RK-hTemhUeZ^qzp6CxSf$h&5gt~Gs;WA
zKtU-ZNvdrJPBXGVytTA3Qe&ZkX%`G+<cX=N23AC?cRa<&AYEop4E!Bu;Ou$>J&u~=
zImDd0@6DJZ-VRx0O5bg!gzB2I%ajSnOu6E3%Ao*L?a-OB`n@R&PMUE_tRDKGk&-Vh
znA+4zdGVGsi?HN#pd}4HTPj1sibJuQ8~<%X?$+AO?^1{KDR#;=Y>2UcbKL&4r{_g`
za=cn_=%@p05?k`6No&e$);q0kM~)xpgxSuf*etfE-K|cnDchN0d(@+E;Uuk<6F;8x
z#AR7;ZdZ5X?&H}cM9kyx#f5yRxCqB<i)i$7F|BvFgPZisBUW>u^cqGqUrR%&KfY|*
zNVDx585X~heMdKI_P&K7TenG5vmFN)Ph33Y2YO^Dr&}B)xc@19OPr?G-qYN7J429%
zH}}+E-dNzn+Qq(9z3j`(M}9h|i?@RU2@-G1H4BpNIEa?w?7mjP#7+#Rj(F=^CzK_n
z!ufq}IR1~rxi>t5C+{QpJtCTa#oK4%ZK!xVLA)I+4z#<a-tpZS_TAKsw=jmK7v!Ol
z7o%P9SiGi6&m0=ds@L+36mPx7+b-g5ob*)L%cZ3%Qh(VHPZ>9PMSf83-*56`E}y{u
zI+`0fC-BZALAu`ryri!>n2|t>iK=tIN=LOOSw8K`?K`4cSH0pVi<23n9on(t?FWw(
zK8m;Y;%yW0Hl>61c=yR$a*A>s)Eln%Gm&B*^3+Hax6kTYoy7a8Ntkv^;!;WZ<Ya4x
zoS-`VMJy9_E~r*c`%v3rsNW=-Ie}4RZ;m2AF_O1_;_13b`mc=C3^I}$pTih`FpP4=
zgE6)Zl;5sD#V`1CCfT2+lLN5o7{~`#?MP>b@u+w>KIOs*(bc+qIFBoaGvM!Vo_34C
z<y{nKo2kBViPjEkGz-T`^I1Nc?TvNjd7vIcarp<-jiH!+1|POY>HH;Mu6FVa$xfo9
z9?#iOX}(X%_eZ_$a#{b|aY|!`e>%exGf26qoZrz|JeLlpNu?Y-l~=guLJ}W0B{F=G
zbZa~1F{T}`Q=^sNQX-iupY)uHxgEsY-8m^3E~cm^N#V8n^$qlSt8~_SGezHz<_SH5
z_4B?dPZj-+6gAQNq-XCtQF8{JPe$B~!>hc`RfFaGIV4dToGC2$l1f-mDv9x_yvxhP
z<(fS1BXh7XEx!=ETzaein;|~6^vEToRUVNYG~*U?9hT&AN<6dLteSCuo;><B-)^D)
zuXDb#N2Gzu%O~TivX;c#KU8-vGtVcfem-_JG;fX7KGKTw|EKS^`YVr>mGXEh&i<C2
z&5GgKgf`FC9z!-x8?tfyo`d%zvEDV0hhp<7-A}jcs<Eq{XR>%};VbUWl9uf+`CeBz
zkE?X&2|5d;PRP?PW*+R%!Ms<FJXf-LUNVOls^Ohq=CJx;4)^-#Z1Ga>*T?ccWb^o@
z<^tlaOF!jY49%latn!o912@zCeAZ)_F5Zq6Z_9U*mPy(s`>aAPi?bgG7s+q5hzsS4
zc+$R*+nLHxXmE+#%F>FTFJOpxn|tyiH9O1C(&VDDKh?_~E`9cMoeO@dhpcno-5GIw
z{X@R2!(y4zGlGFJp_t7IrvFEOY}@%WNWA@-<3~b6AIy@xY5T9Y^z-NVHs%zg8l9j+
z?}JzuJ1EbV18k^vfV?~V8FPD={6n^rym>1tq$~ckemig9ZNSrH1KY}M<WP|^Pix5!
zsEuCRc5J|RwcdYsJ*m=<MiqH*@6;07w{+JI;sV^n+dXHeb8z@n8tfcFtC=Hsb$>Vy
zLq{@5ylwnjFP;a<-*<QqntXL-`x95Do^qvpy6W1OPV`&v%$f$yWX$hQ-5Z_g?B>Xs
z%1-iocjC{_ju6wD=$uykbEFmi9!+pJYaovdTMjM~*S_0OcB&2C+u1NsysaSK#)`L-
zPf3fb+SW$AO*v}K?HN`KRXtfe*n(G6jjSwZ)NIAbEQ66aD<h_LjU+a)p{aIU2UM=9
zJZ>Wo!>sU0uw~g{TXN>vV%p!9V~4HebF133u_e~kEQ!svpmb>~n!OXJ#M>PnHu4>`
zCGD~eUyfQRm)wGzr7igK#7K$OM%Feo%0FLyWAQduygeh<n(O*ay!~Wp<hy3Dr}GW;
z5mznZ4b%-YuyMLUUUmlUx0*{+ZO%~4IUixhT4!^4>6&r<t|><rm~wiGDK0Zi`DdOf
z9lcCJyd5vzw%KdS_m?J=vM^<tvnih|7_nb(!OIW}uD`Y*V5TKWewM0LEQ#A^#qCGd
z{PWeC?uTrt)u)dB<{GeWcvAxQH)oH9JzkgW<zLc*Z@V1$ysjm2H(JtbwF3uc*)!pK
z6VkGpVpiFnc>5O8v^mh>b4y~w+h8@lEtkD|N-NWwouU1?S#=n1>=#HkzmOvFHv9S_
zMwDH`^smdgSHYdh_1)<?+FgDx>M5tM<Mi!yyneTyiS;&8Z0#n^D7H{-*)~pZ+D@30
zr#u@yX*@%|+)tHBboS`~@iwl?8A^;iN6I;GdcO9i*>fKrE%arq=DGiD^=FA$Ao=|R
zneZ|Yw<<vl8W5zNFJ%wC3}Q`_VCmOFd3_^{77fBN%noP7lW<m_i%>pK1nu9*8`2?)
zw)3KxDJD!m9?kE1{r3=KcNN6&O4o!->Ltg-Fh4tn`{vpcnj+0qi1u_J#`5HAEUU%b
z!4`20lUDiH3e~u(6W!FuI@=>&JH_&p)DCaHxSMX3Ks#-?tX-MFy1fbV2uNUZS^{;)
zB+8Q~k*5D9^2}ZykUf&5T}Yyyddzns<X=%+b6v+227gJ$G&qT6U6M#dx8iNMdeSw<
zCDOk}5@T(X$UK!q`iUg0SLo|ZPNJXm(2*q)*|sl1IoEOO!^H9FK@6{Z#V{rznkI&5
zZi%;FH$`#o_bB2*<mI(ClJF{#_^yhOXGs`q76fywP}!YMfegLmk4>OIRm9sSsuxOC
z^e6h4A2B!lFwOF#W1Mmr3;l4@wamgFr>_3ob@E~F5FZ+U_NUIlAj;eeV&at`?Zx`?
zUOJRJ*Y%oL%@<ckUuwAe(({K8(PHgiy2g)Fw&BA><}8-B>xA~Lr0rg){pF?=)3B1y
zbaF%nAIM~VochHc@z@=VC)iJ3U?vI78<>E@cRlad<#jSk?3A87<(Par<C94cbBB8+
z;&4BKxfi7;w@aXdenvsBVsSRaP<C<*N3x?;=S8#rg6i^L(G-fe`Q!BaAm4+iEz-8<
z$7r8VyWI0*iJKG4#_=&cDT?B9SR^NO=8Bz}z`(8<^5@OQSIoUqLUpC;zh>g?1L>yz
zXp+Yt;!*V((wu3wJ!OPEsnzH8mws%0k?KdA^YR|keyH?Q%jc;^o{~eUujxD=nuX`^
zELulqVNfl)Qy+g+`8>PD>ahBGc!;+@dgihsOLOL7*-UMctzJbI9(Qu+vPj;X_ta<a
zl*fMYHYrkmCAzO_s&j`8)t<DEcAR>iS0=J{lq$+|Gdqt%`npvcs5btk-u$v0?8IAz
zL*gRd*3*1C=w=Sp`{rm?pUpJQ+JEWs=8n&njwFlDx+d6W^YT$PdsG8w&)0oxHgH(8
z?AJ4MaMUdPM@$w&U9&iDkww)vnbHDfv9hHyfg2YR-mH*>jQ`Ku_u{Ruczb(m0S^Wi
z@cU;y_KsQnZIy-NpPHwiOyl={i9CNCjaP0M#fAkd>okD+rU4Yq^&`8puRQS2@p0fu
z*0w&$X48|*9Cbu_HHT>3?+`wfm9=u`Aa_C!vV7t}Zu~kxnSulOS|6bAz5Uvy+r_c;
z?UbLsl|xQjS>?HjY9%&kr*=KV8+wpy=|ScLcXA#q=h>d+gf?47@vx<AU$m5pp-ZGE
zT|(C{i{-;KlZhqA;_ow9S+#?id|)7dT<y-Nz1>OH47Sm3SNg|x;jbo+_|EUfL9cFf
z&gw?@$xb|rbmB{e&a9r-5bwtIc-_1%m)1&K6jqNn)#`G1cWo?=)TX^xZT{O<8~cH^
zaV%MfHiv4GcHEY>ZmMrj*sxey+^^F2u9#p=TWNgnHn!$`6KlQhSWDxrJsb5&H;vQ|
zjk6`2KUuJAvylek?b517#)!A0#oJn@nxm?&Y;(e(I#@NJ-e*78Nby5PUcRv5+Z<b}
zIonb<(?&a6nx~#JQt5<|YbTAgC}V}we^wN<GEz?rEqG^;51|cx_gmm$srxHtL9;CV
zKU*W0biXI}8u&HMKz(s9q>mAs+TyIak=gowzRWZzo5sM!LURV6GiT~Ra{@1!;r-K;
zGq+5sl4?rNnW|@f&Dh@EoFE@F+B`9(^E6YIWSEfUX~LyVCY(8HLQcAg`t&BO-Dt|x
zCnoGUW5VE;rZn$q#w%M3<%8=iP+-Bm{+4X?u~e_i62m`M9DiU<{t@i~zqBT3rw#qb
z+LEfX%?&SGF00;8({mB?ye3OBYmwKuHZO+P#${^*dK(&YyPydb`!(f26?-O5|BDad
z&5_VSI)}C-PHW3LkDg3k(i2wp<#<M42B-Cx9;rW9<_;wI@L0n9=knzCastKM!QyS1
zA?_H<uEH^TmG<pcvGtEN+Lc~I_0+X^&s--z{dJUYwO;#>>!l0ZLgd=re0*?(x?7d&
z*YOz16;9(h_pHt&XDMZ=+{ka|uyFFGe{mnSyz$ZZ;mhg<fjV0UQo}R|^Cdx~{Vh+b
zb0OTz3}KUf7+KfC2(}By#V4Hi)gq*MiO`Heb=QbUT*ccfi=wc95k=$O(X6;1O_mrt
zNvyr0>vz39E-Ho{$=W9x7{kdnF|3w$%B6iQGe5;LK)n5BjALZeIQ99}XI4%0DNOa$
zwph())TeD6CqD>zNKRE|g1LCRCV@WU?IiJ_yRJc_6PY((K2_pv`|;Y-J*fV0plUPu
zWXzeb{hiCnJeGF2xOiJqyj{>m-c|1t>7mbA-CG`4ofDZPz4OFmecjzjJUuQ(Elr~2
zMAf_E^4I_5|GYqbU)A78_s6hYSA2K%a@9*uHIJr5P!!WvMN$2aDAuP$GO0!+`=l?k
zsuRwW<snqx7tCAn)?`uu!&Te<8Q?EJd}WSx_NUupKW5(YWAb%Byi@$>7ww1h1wZ1t
z`|<lmUyfM$s7L3+RyQ9;B+5rD#+x|ZMkRYw_NM;5QaMqV#avw#A9=HHgEtOCyj4pp
z%g~s_2|d=}n~AI(o5(>uFK;?X%dMSehpO=m)L!);?_>D0u6hms#bNc2GAQ=PvsJv^
z5E{>lz41Jo9Z!g6)ye8f#Fvlbkj_Iswz{nt%R$WxMqkvqqkaq{O=9?MP&9v5kLL2}
zNP>Pwb5N}9psS&H`%wQ55N{pE#*nixh8CM*ICwsWgch-s_$QVZ^}^`dDU9Svk)(;Y
z+n-1B^rZA;y|bw%{#<^W&8|<%DgGa8)mJXJOY9W8+KV}-)!SVy-tKbC<7a*@JC5g)
zvn?0*4Y|ycrtEY^4(r9+Wh3RC@gbdM*QHPQN@2vcWTLZ@7<VR#;__8-R)6EaaY+o@
zn9Q1eX|$}G&USfpE)UA$AMy6`w;Yzt&Slq~Txl<~FD>5g)_iw+i2BOE@)-V1yTOyP
zS?-g`j*l5Q`(<HQE*rDAS*&`IMMZJDUYi_w$>(tEf^^!^+49oKCNd$LI}SNS<YaSZ
zeKt$GXY)!m^&GLb{F^MApU$S!)oiwiw<+T7wqkiiw9n<_MDcffwwRwqLANZttLY5!
zQaypnS*lmFIOmZ?^o=ZPMqeU$=p`;zy`-Gd0&47*4tjuc4ZmrY+#!qE2QukCK{?e?
zku2BG?#%ZPs;KYf@cJ}9&reZ(*C{5CIK`$#Cs{Y~B!T@;Vkh2qH#<r6$Rnyjl(ph~
zh#lhX&f5p!wio`MUToO5gORsAsTc3b+3f9<S-**bnHwqjX#?w?Y+%i|^^85Ro>A4^
z@w&5sNrx6-wRarvR!=4>aXQVr&meF444U?uMnv&xjQ=7pC%dU!D?O9PCC18|Z!qT4
zH3e66=JvhrL@ek|P3!Jd3wL5{b{D2OI1=zeK8QOzQsmxNK67no^<MiXn#1lDW2fz?
zi@SJRy-8iZYc_gkc}?zgtjUj1TX|jA=EvgNH0)m+x73=HJ5ZB1M{T94wPmuG4HwhJ
zhVj;{>tju;UfQP-Z_C?RD>qTvAo)V%ow6dgj3ryM#J)l!Ggn9lT*W|#M03)^&6$^E
zuGc_wlKUEH_}&~3PczCrHsjY&)s^{1ZpImzE?$0&7fbz(%KI_OpUudB$Bg{)GScgZ
zf$)X~k|NaK^)P4MQ)!0_Z73t&ZZfl=pppe=-WWKbe(?5B23j33FnzKCduId20t1yY
z)IYAGucK@F1Ow~U^R<mM=jd^BM)x!4?l&{6<II?O$&`)Zri^KCN}bJ`)7qQy^{g4o
znwc?XqA3<xCcNBk!dUUvRlNN!{yIK0!O%tZ?PC)T%`w5DshI6-COxz{?Oz&b*Gco@
zPL`}WXUTlkxNdINY&c+z*I_-UiPk(ev0;XI+wEUlo~mztKGs${S*rQ(*W}cQS`2DZ
zn_%(w%a#Td`_X{^b~mQ_%O=YGlh5;(dbCciPq7jW<uTus$px)>^1U64=*ZgTot5F$
zi%R0{uJip#YB~_>zHV%~<;H-zLs^wInsVvmX>fQh-`}m^K*TDlmR-%}I%|lTy@od_
zYl)OUhIQOJwzOE!?CzT=b9D!czwMwfZYP7bA0+U{NoB^JVo8_N#LqZQ&26W-cIFJ@
z@1N7T$Oo_gd<b0b%OLUA>1v?*xIx?<9mJXEK{T2a%<qN4G?h2T*Ucg98y}{>1NA#!
zhtq3q1dT!?nD#V+i5ZcMxE)E=j3})AqUE(Ay|OslNY_^Nkss=s7a2p-+472PB<?np
zkJXtNZbj(rWwGoll5dqkb5CpKBHW0jnfi^3CaZ6&x~FQkYO4h86p6QaQ{o8up<G_o
zTvsiXW#3u*xdRe-@^>P;<OykeJf6Eg@%&XcfgS+~WQ<7S&OCWSMar*H`?j5Zllf38
ziPHb69=?!BWAXMvuS6<ZB$D76kIDRa!gs2VJXabi*CZwvS8b~v?EHoa`q{-ZRW*25
zw`i(cMbq|W6mEASDYh+wjBeo^9Tmprr=i@q6UwT%Fsf&SlU+BQ@vTB>xH*8gbNu;d
zl0UzK{h8zAPhB&A<y`oKczax|Ev0()QIa2)VSY4o^P|yYA0ABeA@hef<(hlrZ0pSz
zb8kKec@rq!28*{h!@RL5<IT%A=a~BC9FM=A(^*_IgF*5bTdi}%wM25%Cr><`$RoEz
zvQ*Dot9O3EOx~JHVi^>zpNW1J2_ez?+l?mVY9tSwBWacrLBh;1UjL&#kNM$vxJ2@K
zbSw=XMA2?&G}j`dFg+Sc;;abi-Xqx3K7xJHt#|kmMjy=^4t0;<^e*+(O`<tpCYn9D
z(Og*@PEpASMkGbyE8e=S*Y8GB3@;mq`6jVAHjd@x^-vB?3#CVeP%KJ@)BV3R{(O`r
zPxNehjme>s_KD_;x5sMcVtpWoolA2#Al{aB%;B7B(WigqYQIE%X7RS`R(Vye%4Kc;
zT>kdW(ausf16AAh%~DM-EqIRR(2XZ1(^|ZB((J%R_5UZ$@xxjqQ9@dP4Oz)d(F~&M
z^HfFzW|H(iOS|df;F@frujCMB%;ULu>n`3VX60csNcj`#`d(*d^7|9*GquYkdSxcF
zk7x42GmBB;?YG9+7{%Km;_Y~G*3BoI3um$^<DX4Mohd5y&8B4QY?_=@?v$4@F;e9}
z+AWPe+HE^@HHBekQmOP<dIb5Gw%d}%mz*^AFUrz6K`d{RMTcjZywc3P+~7<O_ZF*H
zscsc-pGRGir-QsBtIN|Xp@22w?FjKUTRXtJ23(-YjSO~-PoVv|NcpjZFgw8yr$5h-
zGwC!(Lr?MR*GXO%os=K+N!6_<Ib%G*-WFao*|Zx&o!v}Zv4>(i_Yl_Ki|OHBG`;4<
z6Y=(?=MLJQ_vCnvC;y)Fl)v(JR$t%9+(zp#99zY{eeTi=yEC?|J0C-)VzObXvKl6n
zzi_nrO=IK-G=?YRMrv<zB(DkwG2_!PN=zThpD6>_a&Q36++C?C-hM8j+3VKsjF{LR
z^WVGEsINSmoE+up?8vqgovFOABN=PkGC{l@_@XsK92%0+t{xTTjg&9emN`?0gzYt{
z-l`_ulJvU%&6=7q(ic^=qGLZRR?e|zTAmG8x7bqH#+DJwZ0Mb4P4OYtL^_HSwdB!c
zq5orI&5V0i?4DwUO$|#Lf3x6ZvIRl2#Y-0>UeX)&i8bebm^m(8%xPZEoT%Mq*bX(L
z(F8L#JT+s~M{~l8i;;dtHfI=lA>K~)F|z82k@8~gC0%by7+E;jz;76&`7@{YN^?ea
zGFPUKsdgeviQZ+RU33$MuQtQ2lJvv+zjpIf-;Ou%dou&?j+-%jtC{L2Gy0S;W4$<E
zzN8uEs$-8HHzmElDXqGiGIEe9HtH8!1nTSBo02EKPCRVN&_~ip>3#41Hs$p#6Kad8
z!>ml0;Z%dEEo<O?!31YVQ`$Z<5erOc*w~caL!_g!FmQXIX2kABo~9V7+**3(85T5Z
zu3e<7R$NK3;`&n6?JKS1Jz>SqhgQ@+XU(wIdhVy%vUo{NUd7kd%(fm2)9W#GOkHdn
z$-^qGHc!=qei>T_iwE`aiK?$$=7u!?;J^m)w)(%0Y)W<{bVnDO-gM=4t3Iqq=uek3
z{goX!nDnkg<V!OIm&_py7&lb==c6&^E~4$XMa&tyl(A*psW);3S!pY{e{Tid8?L08
z!z%1TSFz#4M(QrzsdK>|d}8+KZ~g#(|9yo1RgQCd=V>YrJcE1rvs_(qmi?-qwsrO6
z+62A6@ABh*TYu(@x4k+Cacf18@_U1@2$BzTNqMlDhVZOY2xq&6azMPjH8+B<Arbih
z7r~x&kz9_8Bw=I}YkZ?I&WzzdF}An*g}rpUNmn&L?F$W!VZUDAmsO9!<6sPa;&J*q
z`96xJh1KF%S3`Z|LiLx&i@D-$%3LvbMJ#hx$6~3T=D105oDpxsr0e};n!o_{mV5t|
zz^M!IJU=Q8&SCXFPsDTRvTCxF1d`S7%(<Az?DO);xSFWmLlTZZ68Zcrk=xr7IWQ!V
z*7A}Z^(BGkE99g8E)K^XN%+l9!swF3%}<)`9!{i7^8_~Zj;HbwX^>UJHBn9bR<*p@
zXq_=!BXQ~&u55@f){D2B3qzTc63T$-q1bCKJKsN$g1`Xoinjym`%|NlcstggY!iQF
zPWsWcnICVjX%;rcPd!-mluPS<uYF0s<V(?AZw7Vqra?1r`QUnUxVkq*rrxCKiZPRq
ziQeZ`S|5LUj+Wx>1M$|ixHl=T-h`}5l)pwI#os1U>bSBG<=<84naHNTi41w69W3>4
zONh5`YisU)E}Fg{qo}Cbe4cpgcsxS=qcEON3FEj|xOzyD?C%uIw?sYW-&E6!x8=m!
zO8-P~rCtQ}YegtKGlD~%!*SlLTq<eUue8vPwlu6(ZKCk+7fz;lyH>oNniD~Dz2EGu
zX1eFqi{BY7-<K$CTZeHZFO(z7KYdX?6hp00aVMI6=VGbxR~)tU+`oP+FYbo%L>^8c
zK01LtnbN3hH)~wiM8=4hHv%<-m)3pVea$NF#cDPa%i%m}<fVmgE`3|^#8_#3)i*gQ
z#-C2$-fQiZP0|kcvP9*>Xoj!z*E-8+KBVfhkyfo+BD?o$MlFxd-KsP9s^;vX-Qhzq
zs*@k)&?!gRPx-1l>#6oGlgXL;87z31!RN;r9KMx7%?laIcF17k+jQo*rtx5me7<+d
z!y_h}-^JTCv$flroJIA5EO~}x6DH5?My;|bTRD^5oD2d}Qu(=GzN9~uQFU4Q8rpFi
zyg8NB%v7$rX!iar1MD(+_%;)#<V;4N&QxDM6UPdfTrJ4tp7h@*?`LB9A(Q{UWzwgI
z_I-V%BNuP&#M`=0<S81HL6Z|1v>&Oy(J}2EY>K2&%@7KA`*BgcUB2uTmKRRq(oX&3
zQu1uiI6>}*Bl3AZ$n^)i+1X_WP2cX|dDad_1?^zIJbBVK?ZCd4Crz$+vNF$;Qp>lo
ztnW4+?O9L16{|5{xJ((#?i~MPIroRmqT8#Pyty=!>!T;pe&RS992|wQ*C-}z9!V+j
z_F2P`nokYl;XlfL9XyaR9hJR)U;xF$+qikolrE`knmXMv%XQ*=_b!}x<4C)Xj;!}-
zk9|~A+OBWR2*<`)%&AYGI`tWK*p5S%^;jX^-mg`czLV;(XJKv553YsH{F?k3Y(uLC
zmi!^!W=KyvTOJP!b!~2J#a0{nE6H1_`#CEf^|vOXg|xK>^_I=8X<E&io#JigK<(ee
ztG*my!MK%1O2rzuU~gdEGjrTen)5~d+F7;D=^AAweW@A42AWZ`zZr*?s9*cV3<o>y
zyP6tk{)d6~oeVrzA9%Uy(Q<hPW}G*$x=1tD90OPWG%!)jK3`I-U1rATO{UB=HD#^(
zySKNQaO{o=cRri&rj#iktC;dvD^oJE%-DI~jI_n#ta#g^xS8_j%-H<ZRJk&yY!Pq2
z=>MgBF~RAf31ie7e*VLRFAX)H6>pCpG-cjHu~fV*FWx%eG$DGX?x&;)O~l)e&1w*|
zvIZWfYf#C;MEXDzKGZfPpseoumnj#A7>Qr1d9rvL*vf(rqbzuL#R8`ZmV}D2hs4|V
z;_W9tD{hFlNoVE5VX%>(u_ZkQ8|jj6LAtwY<G!`{VOE>l4Qlf<qc+RM+gCw#Xj`Tp
z+t%4}I<NuV>ey2@$R6Lgt~@pL;A}w;a!>bEPreTuF7@NZ>Hd5cZ`ThWLZ5R(*pfO#
zxlBWGZ8D7BV@K1r<rpf(PvzLBIo$boF$1laFr(NC=3A{`Slbme>$`$O3s%$biL!^P
zZqtlvC!S$@m4keo+|H*c=69Om3(jz}+*xM2pJlIjTcVqv&V_#b?WNcKKLVLvKZr@)
zRO5=bGj;~4P7LCmcGfqGw~ac7u(We13txq6&pASRxCmvGN6?~4q<jP-$%~Apfoho<
zr?nd-#{MPVI_b8(uH54>Tpt`mY4Ns7oL*~ZN$ay9T06g*@y(WI>Wg0gwHvhRl4@J+
z3>8<sQ|J(j$Fx|EEm!SxD3+%~G&}q!ju$bSq1H&?ayw<(tM)uD-u_)To_5;XwOJic
zS@E_~Y678q5;2Om)y3OeR}$&>E|Gw*iM$tYw~DtO;_ZYsV(+H}mbu5X_KkGKL*-Rj
zs9y0T^-pc(9qlY8l@pUXMKkn66yyD)h*=zkhfgGH#M?yGeyR1t)o0P1_H-y~CWm5o
zIau{lFrUnVrTGiw&$IsAbMeR5POPow&n#V+%J|bjysahPHmIt4Vy8Se$NO=hj}Oky
zK9ue1qaLYx#$DwN)>PcB?2U(byU4_w4^_SSUd3B^(cVlFZv#`#vDxPwdm@w@D(3oh
z_QpqNh{TjcI=@o=uG)W~V<Kg=OJ&nTKGo0Tu=_U_Tk&@G{}u`lM)R$EG*7&v<n<TD
zhpQ3Hn-RvgG4dOGsv6x}b#D)4BgaSa@Kyww4I^;2mhZ{X5EdT}X6}z*((^(%u{(^m
z;_WNV9%|U@<9h9mXsxVe?P=DWCC|`*LfG~roJZnq2h9kl=E<k5O|<+b!kO4Tj7j3{
zZ)<{SA>Ota5rVg67!6mbFTW_9t8>EXJtmwrmf^%Xh0?K=yjUj7i^Vm91^XiK-KyWe
z7|qS~yEj3+?Q~Of3iVBvh_}|_ZGMLK&thVgofXU5z1sgin!wk~(%an?XT;mIzY?kW
zR(gP)(!J&A{Gr*w=Ltz<$PakIDb<YP?I`twmu$@Dt7fj<2S~qNE(aS_M~bt@#n{EV
zZT298N7piFEkEtvmC`7fpUQ-7sf-kFcZj!#cV@FFTs7_aY|4wb;YnH4iOr&Iv1}Gz
z%i?2dCPjZ_5_B<x_@q?Q_NC(TLmKc|@;Q?ZfkP>5xs#5kF@uNWGT5bl_m?f1@`C>V
zwe*l#nfzTo6WdGLr;^qzOT4Wh-fsJxiNk+lu4?YzM`q%q`rEof61l6SX&)HL@VE#%
zAB~{mm<T@Ch#;(-H{*|=W={JP1Rgz(#rhMx_;3R6v=i8$(rxk~n*Y@9tj{ibmGs0R
zV#ojGjIO%8m9X%wv?{ZSVaM0Y%Vs_Ed#z)B-8JOZ_Mq1Ke`(m`U-nh`m%@h&xzlhK
z1uthR?{5MZ*J)Sw`Dn)N8AVLbQLKDF5{H!|F-iQJ8<7L~EpQ-1jtszS!T>Dq^rk|I
zo^%jz=U#N8LccDU8#}Y@YbV^UIpRLZ0k>-M=D*dL3vC)xEU6JLr5dr~qaB_{?8tGf
zr=NFS9IWe-I=K$b+Sev-Xf66l^Xpg3idS_kc~aSu)#7c*@0NJJwxsq`ORjE~7sLk(
zX^*T3iM8USbigCjLoV~xiU`eZcZj!%;_cHo3l8?S;QYTv4ABO>nj4rT-cD4%HckE7
z2j$IK^WKaz^UQeB+l(77W^`DpJ)0aeYFU_bY^OP&znLRv2ka*q<RxO@xb|}HBpYZ_
zsDF=Ct=QaLvvxD?Up1rMc{5rzGo^Q+32oM!@P4Za2gKV8;_b20rYx?i8LT)Ol4VBc
z>t?+E*G&G<n#q1M#kZ*$&3!eC6>n#Ww}bTmE~!QxP-ucty#4sYgo^b|$uW@@D%wO@
z6Z-ce&1&zP@aVb;|BAPfziJTHtp@&$YtX=>hCFp^kYi>->47H7Ko`5k+hONThz~H~
zim!?C91SeQNb4$Q6ke}^{X!G^WSeO>$dY^uD~e0A+^kUh!pk+!jk3h`rhzWQj9kvN
zpvE+LavamyCrBFc0X12aRhtw2>R=aAN1Dfalp9c=2Y!wD{;@eF_uG?Y*@n9<+K_p!
zJ$q8yVQJQlGWq@3Si1)o@_R^2(S!MJ%51vXi;PNxaj_oE<I;V3Y~PoNhHm^ja-y_=
zvvIIm!ZIs&+<sfhx#pVh?pVdqyQ_4*Sk00iYjLi;RT+dkNvyq}tP6*!xBn>5njNEF
z*|W4-cvhK}+K<aP$E=)l^62vB#8q#eTKR}UJ}f`%r>s8B6E6oyM<xE&2*IR#2<64w
z>Ei83@%GW02<0|J@ad=On-Y=K5pQ#|q>cGIhBKq&F>@kDeHisD#oH=-W8~2uL)$;)
zM-?5-;2F`JnIFx--O*g=t@+(rd97-0YpvS0u9>nC8pSf%K9+k^)kjv(xs`akSiGGi
z-sVQe5s|FS`-nJ}MJe;(o-|fAs%@R(Stj1*6{*%8CV$Ae30$*JR0$yNjL<xHmwLlu
zZlZX*MZ6t5RIJtO?BQ7L7A=eC)=TZ>h_{Wz+mmCY;j)UQv}*Ok<<)bQ4mjsq6dlfK
zuSZP!_#&Kc%fdNm6OOg!+83IHF<3kQJ;mFn%46DpG?)?p1k2+r5OeYNk$AhKjz8fa
z{21cx$MQv*kBPVU>iLoXhaYn``|_KIFa5<^C-Jtlcsp6VozumeRPna#@7~f{c+=`n
zZ*p|INc=5T(VI5nt!LypY>%G9u26f{pU!dDN**V9Vo9*PLGH;1sFV5t4mx{GR!=~6
zVX@Y6G^wQ>BJ~$uRfu8E{%G#oMN|HS9$(ienqQOle0rF?6Xm7-AVg=iNW#owNL?Mp
z5cQm0GyNzt!jGbnesmk7S@R$ML`dIW{jlZ{Hv_5gP`WT_?k>v<Vdb|V>6t><JUxWR
zGefvPH-r!A;rKm@X1~s6+r6U5cprwhPZ;+JV&)R*xKHcz?u4+qe+WOG1#whYV||@F
zNg+gAh2pwAm<Y|kdpGlE<^X@~`2}F!C4e?F1DH4}fD+CDd^+uqNuD1^Q~hZB)sN5r
zNT+topWWtxbe<W=-Nhknhz%j^QwVJi>uatI=el(yuRNoOP(A)CEl9q#fzm>Ru=QIs
z=5>`f*;hL40jjTWN_(a~v3KL-AvQLdmgVIK`Z9?j!Kw1r$sjmEUS?&LkD+?BmU_WG
z+*9fHFqOd3X*6o7-Q#lFky0Hz#VeCF?X$=|nnjnN%9Ijs-|Wcb=e%^)XWFrjPNmEk
zX|+#CTR%O8Ir2vtz9I$x>S^+zOJkEfFdGJ_rf!g_-hnv%L7($7L!Ke>NePoL<}~eQ
z`KQrvPn!JO)0nh3lRE`T^jxML^8oet<3g!XHHecJ0yrt1{J1JQ8&;6-O$k5z#9QZD
zN16Ea2)`~Kp+wdZHv1l7WYm7f2k*fkzW&&~gQ0ew_?7Ua|Do-eUfD|V-!@UocLQDC
zuBX|t^+XPH$7i^^{&wAo{JKE9qqFEbahm)frZD~1MC@JM=)BgAuU|&<x!f4?#as8x
z(TsK(#qN6})mt3NkD!70h_#L<29o`F02jv$;O~O|ocXUCPGdXq{d6n#d$gj4b1NEj
zX~miMO&Pt@j_OJEIZ&=1f2^rXzxs8l@?1W@b81m<tg_*pYH_YbEm~M9E3VLnTiR81
z8D+&<^)`Q&uwud^3ntoH;1g%WUv;Ia_GB)G8z{cgz^U~H7X4*KXo&p#XUprbjgdAL
zr9F}^>DU|9rmGEjx7Uu1`nTt28hEk8z`RQOoEzrYyPI>cySQ7$oSiq#@R(`Ffu3e;
zbu!~$)to29+t)^OCW*J3KA1DX(ZEje)<wMCB;H;SZ`bMeWQsX^9pyW5Q9C%=!`Zdn
zR6ir>mQI>5aG8m6piSs$BUaXxW~r|Uorah&WsMorx0qoe-rf{z<Ks=~^u?5alT8`x
zV@kh%rkppLQntW^nd0s3izXa?W5PZ0c62>c>eWzP8)c$%q6u&RHDUf$6Z$U^cYBy{
zGPMSUU25>Dehr44s?PlM>N+=?VAs<`xmYHA+9^h#);!nG1fO#z>T#Hm-`xa@;<~QX
zz;2?6G6_v6tYgZ!8G1aKro3-sM%P3$S~oH0`UTCO&D2wO)hv99C3n3nSb5x%3C`9y
z)Ukn78%BB9GPtuc8QR%lY*L?VbDGok<6rcV56RFx2aNsNQE;{^hiZ4gDYy+&qB}6~
zz9UuNJMz3wXKGyO%$i-wi-_#Q+TwkQ5^qn6w}`jJ#oHR<t+jaDN4#~IvWjBISMfM~
z6&-G`!sY!cYPzrGYxix0ecs8JcKhkS@h}G~A0=bbab<xWC#?EON(Y>U1Lyc*@6BxS
z_P%%<E#8h<=ErSI@u8o;W;Fiv>=sOda|o`jLm4UFz8n!jKo9w&>iT~qopoH4d-t{N
z4j33pN>D`fnAm!W-QC^Ydh8DD?nWAB=$-*lvAgT&vAaEXzw3E_f7~BXKxbgMukT)a
z?X?Tmx3DC{!rG7N6zgMQQC|zS-7TCbul@F!7Uhy<Ql~%`zndyo)lM~)HIpTyGkMxM
zQ)dkMkDg?){<&srUT0BAwQX_r^L+Kby_j2RgZjWXGWjD-`O$mTDETy(x=q?*j{K|X
z*>n(ZzlpaUJPu$l-acM(0G~St@aUtY{z=+#zvUn^wDY=nU(MO7j`<p>j3{}mDpfv6
ziTc^{s_iE!Q~G10ELMKBGFp1#kXu$BF3MzGA6@c7y_{?1Ne3&%idrdtUA0-Vg`k3}
zsl;1%V>*ZOrE|Pu8pHoe<w%7TTGUF`e71HH4^L#1e**pGMfRx_FI^;_?0+<~7!yZ_
znQ;^oZ?C$?fq47lQY>z-&8&MNFVqF?o;YtNv$VWV;_JG?W(xc=(LkJCY-eVH&PYQK
zn`oJB;(UUMS-R}^n&|M<Bz~CKy4u8G^=Pv+v-|MAYJOeMt{mWS8R_P#>yF>pFYo7m
zn&~XmG;bCmt1_vfnmzY03!Q^9bRNy%Ok?$M*QVleS>BoF$>iOUOv`uij4u<<&SP;L
z?-j?~{&C7Wi(`G4IGs7-@K_$JdqE7F)L-uR)<mg;NfeDpV(rOT4vvfE;X&=t_-f{g
zo3d9fnmDt-M0|*q%^%XJ)FhRYuaY_GoP=w-yxWm6xV|^jRafB$W?tSh6OnEvWN<V;
zSBL4|5vKY5FzPQ3qgIayx|^eEmJ>~btI>3yW1_$yGtq}5sju(dg~?`~y*1N$P>lK&
z%3A9k$D+A$+)}^%r##`m|BWM7KcBXGosV>h=kDAjw*5)u+|Wc)ED1Ok*Zg@?)$Mbo
z>ttk5OErGpQQD_P7M&xqNQ{w2kgnc{`YGLRs%LFH`2T0vo@J;9vFrd>svgDjq4rdD
zI!2A@$CYDwoPa0Nf>)o!M!fZ2c8WjsPLULRl2P4Ha6`Pk$8n0Pe|urrQJRHl&T5eM
za{M~XJmr~rKFHzc^BfEx<a_&_gM)Tkb?9@1%GwvXUs-9R`e?TNs4~u`>hB_tQD)&W
zk}e*h>mkjTS3XR$gE@r6rDJ%T$Rf?)54#Y9Pk|UBi<)?|BZ@;^qnI5PiEBuh^527T
z7jL%}*u|D(+gZ?6bN)|P(faZlI+t6^v<hpOcXKsIhOHxL@Olc>UBi=-tJxQ{iZ{V4
zsJUSlkLOKiO@-++QZB_E`B7F+?L#+@J{11blTV*CyVF^kUd0|<df%NDf4VU?x|?<&
zbmM2!uAE+`S&lCOw7(WWv2_8;IBH0-KWJ|a%p3>j57(w#NNwgUuFacjwfVQ9YS^su
zj1g~}%q-8mfbtyZQI5vq?a7yAXd~X199V{V;_c=)s%5`<(X6p2P24>>HrzuQydKgF
zJt%h9jSi*V7@p-yv;OLril4>vOWT#-X~_v?KHPAj-Xj;XpS$qg=&Idl>a&Ws<8GAV
zTj^4~Td(?6y77mBE-W15qTY-AQR=h4o$A84<}S3iaiQo{d5Bcsc4(tMY7=MXPIs2q
zT>RC?!m7A%_N5Cxj-_~5PXFd#rPSvvrHqPF{40*f$ZwV5;=;XO&OF`fOr(P||8;O8
zUc5c|uY<hcsx!T%<%+kn#M_V_j(ip47Nt27eAkg)-yNmxOE<PkYtD937MW_>woa5N
z<)kc7M~((LvNGL~)#7cH59-5;l?@Jx!6}aVU3TQfG)HPIl7{T>$a+Ud^==%b(HR*w
z+eq(7BaOt{c9$F|*4hzU@wxaau{P9EJAoZp7cEZz?Z~g@j$HfgK>M>sZihK=O+DaD
zF*j(O*nZxLoc7M#%XDVEmkY6RrAR5@%E$@Ql%wP!+v>&@%_UE%@5wap(%d>#nz9?b
zI4<6P6>rPB_;F;vA9p)e&>qykm~^`eV}IAyS-%bstJK9mu`UkQhT3P;hzeU8X$DDN
zDobM`m$agZc)R>Zd+o{VO5Oh5c{a75GD5~NtD|-&A6q~`-bHkpxP%#DOSmB3e!a7V
zzr@>F3zw1HXFWgWZ06Tu?bJOOM7`%h9R41}qZa#kv|}Hqenm59tO?t$niWblbG5&I
z@5;t9T)g$o6VGA$c(S_3Q=*v|(mI8)C8Q^-#%NhLl^fGjNne;s??<Vul(%%1c>7Ww
znxHhz<*MdQs9<4`JdXtnXEG>P7GuQQEyc9QAVJ#Ys7(5G%H&qnOuE+8{H{5Z=j!Jb
zewWF(vsrw6m8HDX{rHKu_E$1F)Nem6UWmo=D*xP~9Yf;nX7P52YTeV~?Rd@WwrwV^
z3_XCa?s4;1AE07;_4cHpzECdIywrp8cS&OvZ*#=rpEtzi>;p_cp}i&Tv$)qFi<uv+
z40g%nb44q=3R&4|XQlmmE6>E+F&(8P+gkZ~#lqwJdX6s2k}j-_%a>^kXe;gYS1OL>
zQ<>(ILaJt{W*1E2`;`Q~Rn^STv3L$jOJ432Pebwcfq1)davT}8<&$!b<I9v-<`YAK
zXJ+*x#MpUaY(;6v;*zU)8!cvpyfkS~v5A7GP1<cMo{G0yvrIfwUHf;Wi5WZ8yIpJ|
zx31I6^ml)2e$xE_jlFda(fOmr$pd_LJjmV>2Z@%B;H@*wV2kFY-)G`BBa<D|t>hVO
z!EcTFtI`t_Dk+m=LMo@rrm*R^{MFl&d2gG*zAEtyQq5SgU>tAz#xk;2EFQ08h<hBP
zJpt0d3&n6wwfMI;CNeLZ7!jPLJ@B!3J(o^CPWv<Di@&S-J5JBz?OZ*Fa1)odnrJj8
znRM|sTpq9kyCN~1mCtWwIQQj+HGK^uXik{+%7<a997g1-Fj|PWotA|0VNn>}^syVQ
z^l@pQFK&cV>QN{~zlL%;AdDdm!x<17&YSf*zm$yOj7hmi<{0Ah$I`r5ER7suNta*k
z)R#Cq6^ZA2e*LVZ>r|g+=Jsckb{QoQTP%g);_V&rw!L^e<=+hA?q+aoQwF`ZWiWqF
z2B$sry7*{e#8@i>mSys~YZfb>SZFy^uOZEm6@74&g_VvmUAbO=l-oJ6{4r+wA0u6U
zsM~?!?c?Jt`Ef$KAx{u5-X5=hoQs!_(N3RRZTAs^iyYzDkHg9{IZX2gIru%z;njaR
zs>9{04$i@UVh%61XlA<2VXpMn44i!VB_GNwSX+5!Bad-P`7h4mZB6}o&F@TBmQ3f|
zu>@sYCXjMAh6z3~&_K^QF^b4CQFOf#$(A9Jv~<|VxU$;4{dp%%gEwGTW*tSv+m_c>
zG3kmjpRTUr(W6zIS+g3C#jEM+wVM8UR+IW_C9lg*BX#Rc?UkD>5A`^1>>thkvASyZ
z#%R-rn=5*dGph%G8uwuIqb~ACw`YN=9VzwNl1o_>4T4%Tym4zDd=4P>bO5*41hAoG
zLjvzXw_57W&8<xIuZk=QtjO3N6}0EXk1qR4Q)!0>wk<p;y+U4L)jw(bln)W6+}?xg
ztI9k4?5u~h4|#K*dUCM8GAeY>ZY17bo9=;|`m-IrYu453#%otMp2oOh8lc%(ac<Fd
z<vobExsJLJb<Kqn4_(+R-cI`MqP^<Q6q%&?)M?KA%yuTVg$oH?<QpF3!W!}R!vq(d
zFJ0&+{zjK`A?6=v;t$BT)Yn-%!JRo<&6y+3ow+XFI#zPw=Q0-#M5+$Xt$J3xEm^6Q
z?v<q|-B>=Uex;N_=Ypw}3wOoaq;2}zU#bgLgBBKVN2*?YB>q+Ka}<vpF{=J-EnPQo
zvG{htQJDyic#F56<~s2uN<L!E%HGoaj_HY!<3H`$xzV2HjqSA$+KwScduD&NW8nX4
z(`-lXi?=hXI8wE)Bhwc;$^-5QHjWgy?Lhh_Beh2vIWW;k>T@yltopc(92qc7e4nnG
zdyQs>#M&lNj{MVCe|N@6opDC(n=^9S<iH=*!%vS(5AWo})7R3*+d6Yo-=kypE?i1<
z!T4C`9G!a>Y||XoOg9QVmd?D*lisPNl|@*F{T<42X^-{*n#+^d=*RnbKSnjIpq&Mk
zw6CrzYYNrj#+f?&6mRWz)YV?<daUYFpZ@|HC|9<z`oE2-m#+zr6Pj>1xV`3=+vByo
zD?2CmV}N*DMZBGGbO9|cEa3YI&FEfQpsaLxsKOSpt#u&JFD&PI#2Oyg-K4pbEsRat
z%HtD3eEAZ@D!YBk5RO#8Bbtl}CJuHn<2lhx#U?QX_0jyv{dnc|CGuVx;ot3&*w7@I
za$Azwcs7{=(h09A17k#R3SSze^7NtlZX?pTJ0*>RGt&s!nZ}ly=`1O0p}-sqKh{}!
z`^8Fu3Yrt@l1b~~nJ^@iyd5$rT`rTs->m%i%u3o5tGJR$JKrqbtFwsynZ={1OeP=C
zlt!0HcC<9eM16j;=1e{+$Er;>)puu8N&VcC#SdUEcYr^w4p2h8ZK$j3gabUks#%^n
z2dOI7PFBs@AjeAnFx9cEtdwYD)qXw;0lIgdh|D0Wh=t^>7CQd45MR(rxsvh+dB{if
z&cdu?7VM5%ICIXz-w!gxWqnPNbebMXQ+`Ss<9?=c=6nh*tEaHfHJKxMrK2j}ZbIV(
z#%8I$^;V`=a2(U0YS-tkSSD-!`MP}J`F%9g=oE{6!x(H-$2msHTf9y6VM8-b#aoY(
zs%O=cU35!2v1-Cgr%X(nXrg3Ilk^o6N6VYYCEmJ=x6c-7&TFA+M)9`7aPfAM`rR+H
znR82Lkxb=>h_?+M9$=Veo16TShfTcQ{VJ2@-ISrxM}1o?+)<C9N<R6G>!!0^+IXCo
z`lyeR@KCM(Jy!xw72@e=Rc%`!j!FZ>EdN+GSJ87)hRRFvc8Pe~LA-7M%EY;&CW<C2
z(`=j`mtJQn$<fM<iDqDlXl0{C>-RaD<_%3O6mJL4HnC!!i7n32v@IN|oQQC;SA;8b
zHk>A3!stCK3<vRct#25{8e#lhFN^^3)<wJx5N{nChjHm?C<)^2gqxxIxrQ>?GmHfT
z!^!y?p_z<G7L|=+Qu!#hSBm0YjVN_qqZryWg4XI&R(lx3)H^0zYDh=$HDTi~&-Xj^
zSmF}cAC$n6(c-SoPDNc4NNXC;hOzO=2#e>Ma@wXnkLO$d1bUuMAnUDiOL8o1KOt@B
zzJ)5{?OMI=?wrh)#(I=7;%#>^w8nF3%;K$CybTm@8{IsPDc1?UB^>9kUdO3g^*G&@
z9HWA*@<=xyAxn8^W1A`uG<P<2TgY$NMmaL-+iWYQJ(==~1&qz%zl}Nkqg`3=G)MmG
z(IHHa57EQ_FpI`%_ecLs`n1o)GbDpbuF6x<<E(R3=O^(t>tu}P4`Q@aBSsk-;mnE(
zX2Zf@&h`r?J82iyN2*V|bTiqNHt@+~9q!_5mrJV%3{qcq@^bK4%70EvnJ^@fK`$1u
z`M;U;?ln!By3=S7FqI$6rVzGxB0GnV;n*$hOUvC`8Em~6RG<q-Q#v!bdJo3k>c*}v
z?fF!{Esd`zWAVMRpZ2z<Oo4{%@v26We5iTqm7zztH%|t6^V`phU+qh4&t+*k)+^17
zXb;K<Y35Z-9kNN8{CF_sFKNE>Jn-JG-l}-Jr>zIab%lwy-A;N?-Qq#*dLF!Z?~46r
zW%n0wV}h@Gv-0qS6n4Y-PT3TyZ@*7-l@?cuMmt^jC)R~c8LBN;xbRs#9qQ}C7S*Cn
zzR2S{OnmrTy0BTk-AK)xMmVW1bfS{}{A;W;$K@AxKI=?559!6n<%?SPe_pABwVW||
zsBhcMnege(Jm}_(k3PTrME(6t>ApG6RJtmElNi0MsteCrsa789!Y8BVWmUJPXFAYr
zxN6DQM#g^=GgVi<`Nsio@piU&>nz^>?ChuvHnDEABgdo{xBRNwRJ^Sd=7i05Cr$@C
zsRlE0`KvuQR@w8Sygf_b*b()`4qNfI@IyP!^|qG>%}DGo2d4D5r^bFe2JN*|CcGV)
z;_};eM#>L0vb2klMMsS^KjFahvX0!gI>`U3GeBzx>g92O!$$6IG!iqzNc1ox;r)#y
ztKVBVmm_0xRQIY*j}dRbHFhQ^%9$<V?Sv#fZ|Mos=DH9nJ>hYP=CCw7)aREx;iJ9q
z@8Uz?C21Ic`_ew5yyn=-6R^)uy9)fcTBL&Zl~-Wk^a>=etH8ZS{>=K}Pq|lh3F=gz
zaHj^mz0`o+2ODuQR};=PYQofZO(>MN88)YzQL$53mX;a7gB_z89X5@L>C>1#U3yK$
z8N6RGlfTeBn`a<>0+-|DzJfx2o9NqO6P^UoCn*SzA3?;F4PoxrU{+m=;<k9(Uc8+g
zX2y1AjPki+IY9z;`dz#x-u4l18#GMDW|Q_rKTO8@Mzzj-%?OFN8#JH0rm=EVpQK{R
zmxfExG|G#&JH*=x;_aMO7A|F2xgy>c_0_CuJ!K;e%p`cQd@}7aaW0jqSr{uvldS~)
zBaQB<So}Sc`{|iHS(~Yy<npf6&!pfTD<vb9VJO~?+L_7NBbm~sm0|p!GBfr{U(Ry?
zKc54*Hc=f{`v9Zs9H3>u0XFIWG&(7ZhGVn%Hy{(!Nh@zcl>wV2uI$aAN23fT*<|n{
zB7-|cE&SSG;rJ(u&N=enmax*t&C2lO7H+T2z_wimFS9LVoyy><m;ATVkt-fbQyyFz
z8LIJDJxrl?(PVP>OQO|v?S4>BX7P>*+*2=cnrpl?YVBtF6idmK`u<Lcr9|gg{>>N5
z?<FzhX%{0;fSE%HW?HN^vtNDNX5#IAd7`#`6IZX9@RM%5ag&K%(j%<mLObzxn|S*s
z!ldj86NdRFJZGBNG~GnDe>ShiXH)-0HpfmJ;A+eP<_^!|Z)u++KW8Z~Rrl&4>Z|K~
z)3kRc`Bj&-s4nkU2c0{vq?5Nxn)>IdM7ZeZ{4|M!qZ2v&G@ek^=@V6-FHtsGM88;m
zd@z$8q}R=OGZXr1AC-In>!gE!5O0^t$8}2DuKxfNHGW3ZIV_rm{lwg&(VXy$X3*?t
zB05Fms>j@3yj@w}#N6s8D!D|EP#}UD%fs<#8ID2qZQRH(7L^H8Z(p@-g)q8{xvt`E
zz``)x#oM=y<m-77%7)YO;a>=4^Lw$^DUA8z?OgFTp-?2F#qlYwkt`N(vzA2Avup&-
zYerCbegq{lBIvs<Qn^M^R4s0za%B?{0Va;0G;z|wOr<|&Ixbc}*b>JT`S=6H+y3Hh
z`h$3ih__oaleO<MnG1h(CjFekiVW%F|D`j(wfxz)tXw^2C0J*@x#jgUuvPCwesG(v
z2Wj!wDebL0!P5pu$)maN4x0b|EPvmSBFAag?id&Rj#0F@JXZ@2@~~<)4*w|sv{@Fd
z#oO-vvgo%}d196KbIK!!^Q&`cJXQ1KPYx08c92MygUpc*)W7#Zg2da9Pw8a*Ok$jB
z_Waut$XTlWU@a4rYaPe-&}csBXA!!6FW=wnA<c0&ZToLg{^=IFF4ZiL&qhiX*~lZe
zb-3PN#nMZwu)DI7nd0qlhou~vv4E|OXEGvu2FDssq{G#IL~I*C(PbmB^Pj@^MH3l+
za}3vR4piT@7j18}qfnc6^zGOd_x^2YnbaC%otCs)+Cn=F8}j^mJ$kgLN6<oFLX>})
zeyR)|>Xy-dmeLG6qx_~QPvyyY@>R1qM_zjHU!(^w_vjkyLA^E}^l#vy+yr;(bk)_|
zo#yS`84&1BpGodSmUd@oUUz0(bfe}<H*Scx?*F(lPkq=VKUc~&a%I3pSJuCBB{|cT
z2Z632-hSKRf?bRYM^szBndCy7YAzIf<xHGv(xF?N2^}IYQVD6q;;V}|)_Ru{55(K%
z(N1(pl(+Y(6VtwHj<u4Q9OulPQO?+^zTI9?{o7mmoE}cN4w0wH#+lOM?S1ujw~N2m
zra03s-kH7Uoq3_J8{{Sp`ML6ngVn1wsvgXCP}Z&k-CzEXmqz|O=O8va;2_?%YT!t1
z@wTscn?2f*3i_I}C#2cFa^(9+Cr0jbB58{gR`IrNEhDWz*%P(Wo>MONI!oEn{*-}F
z5eA0eu;XSUBR|aQ&AwMHTEm`oS$6CeZ$B@zBkQys)4GV&J&fFRFw%aDQS)&QYOAVd
z&2ZpxFV(-7jGRm~vOQKGS1nv!yd5CkF70LHxvvAO^mwa?w+(8l7aixsvnI}5nd?kY
zUKi=oF8Y4CV3Ag^O}upq^`xqJJ5Rj5l3V#)|7qu7NLhSJ`%-?OFR^{fbAF2-i^SU^
z;_YVfw%M2p*lwtRziUN`epdG5vYMQjSciS$?b2=<Z?4ynuw@MyU%3&HeVQ<4MN_tH
zZN|Nh%F=$)i3`np@ptkdw*44K`ts>i6K`GpX0o8kEb_0Mh5cgb%kl<3ja^RCmF2`t
zU(cM`LFBm*ByUIvh_?mLYCb4K_dZ=;biaF5D~5_2<b&y%K;FL-2zZ}Jd+~OCWRkR!
zWLAi`e~Y&rbEPmt`rq(R$rPLOe?FEkOH<{k6L&VJu{>V0nZ+&qE@RQGf$lTn?N8OR
z_v>UTA5QhDcw4M(CIRa0?zv*6NTQXIN38skH*~|dOwE>x#Va$(nvzNPYGUhc`D!9E
z<<rYlrjGJ&<f(0YG>c>Ey_VmQ&D>lEn57=>WVGWzybTa<b2n7}vUvNyd`SoSo*u5S
zvU85IuqIkn@9CT`-qsLrYZb|$;`R)36|s=A!oot;vuFOO)+=tMy1~k%^*S>Q*2mSW
zuCYKGvLT%s1=ESvjHt)4H0@$aW8;HVeCDO#{aO1Sddu5=Baw`^ngjY1&)95b;N6SE
zaYh__p2zZSd@P?z#M1ts7`lhYD62h&;^OV`+tQB3+pinVr1mq@v$9!_K{fAJ^=Plk
z3l(pomw4M>y!~E9wWFPh#p3PD4$^J4gYky+*-n#9+#4qElxAp}jL~e2Pd3Q^ow-`O
zA@n`-si|y=x>+P>_I1}_`LI>rZs?^vde!>}D_fK`m_ae|wtV$8R#s2d{CP5SwZCG*
z>qKQU#$)x3r?<RxgQX)^sUJgm)qo$w+q~j!a4+fKHO-XvmR9o4M04@Bv}*2aElhk8
zZ}*>yW_{mi?&XiBojHo{K~eZ`i=ywHNPgE*uXVT>7!bh~hY0m*BFI`APEhl3+{D}C
zLsY}Mgt0;lEnYf|pZ;OY%9ID;cqsYLg<`Lowv>2VSiC)QHk8WZZN)#KEPNcw^fRI4
z^3yC!#c<+s!Wd-=qwRCCTUzrQ{|Kgvw?)L;JX0cB@-dR@MNRzUY2uq|?+ob)3(A_A
z6CmBePtUhQ3~nQ0IIVMOtCF!asI6Soq*zMzjn^*1WP-%ols74SbxftCM;hOj==Bw=
zJeZ?aDju*hMQ=Bsv$F7p&X=|II(~VOb^CLeT>mIvLvrvCZx@}@4i0(zj&9P-)AIxP
zn+`Be=gD4Q4>0Y97^~h#Rq=MJ>f3kc<w?xDpUE!r8h6eis<!se_7&4L4{&{ic5N-q
z=Dz0C_mw$7wPP84nxxrnogX@VHc_#MYJ2s<^Y@Z2zkd%MUhZLS!M(ivyoV-!yE!#r
zi!ztC5O`rTJMypP!0lC>l!m-gtbNjcAwfT8<7S*gtyR*DA5Eu?=S1bGjUx2iAO>FS
zN1+b=lsnO#E*ra3cTG3e|J9X|v7PC;qa!!3x1&`+JIXfKu9wDblmpY6Q?*;tdA2eY
z`$_kESDQ*J{YlZDo5A5e@}-ug<Fc|?o|MK$yqy{8N%AyLX4`AG>K)B)insIkNGBdA
zzf)`F1FCK<vca7r-IZZf+no*ucWjEev$u#lJH*=$kKEK#a^ul#WgV%G4cwsYC-K%)
z-c|dPl<{=Rl@m`~>FB4~o4cwdFS&4Niwj4KNhe<6g6%{X9;mMUlj5xX3(mNTw<~>|
z8S>eQxh5wq?Q^2Ccxx7KdrGtQ)ZgWZw+SDWb@-P&K=Yj0mEolQj?#d~I#IQ<>dR`%
zBkrwUX@Vmrd5k0btIsJuuG;NLit5su(q(51QV&*I?CtiB%o{9CHq)LphwWKE-H7)E
zBW=Xn&pjO&Z!2GtyCcoS+v?(NUGX-2q&!Lb_p)ZmV=QfWQWGcnOU2sF>fgFJv3Qd_
zR`NkTyy-x32Ya5^u*2zzfy(0To<Vjz?QTci=XO-~wpT8L9r>!-aqyghx_Rt4B;GFX
zXiwK}Mj}cW$yshBtfi5+g^g@Zve((o9*>?z&FUM;zuul-2ko)BVy`opJ+t20lNxK!
zzWGM#4V2$1-GPcF<$Kn1Y^dJv5%Km#TlLHPh`S?{nbAhxuA`a-cGdZ&Oev}_b5m{S
zPVt2v9Lp+A9eK~)PWbX^vM)oEe3|&lkFN{;7+%wl_4WLCGv5#Mm&*JItcK^CYW&-y
z1}-aVus*Z~`8(97mv|dhp%JeFn^K}n3-V2B!645L{O;U=Q@uOzYGNOpBPVgX+6;F1
z&g7wa7UsK)8Dv^Uf@aH~MlQ#CdKJA&uEkxuIBqXcA2&FNE7?K({WF*wQ4#WLMqqgm
zNp_!ThAxe!?D1$?e2d1=QS+Xo%-U&Xmft{LmmkV$D3u`YBrql`iK@yK9O$7r(`_mE
z=S}6o#8k32rePZ=@3Osx58|!;O8I$po-b#U$z9!Jo~d3fF5cd2rMKek*Ne(P*rwUt
ziqd*rRBOGra_N$lN8)cM)xtF@E3@jR71#b&CYMni)=!zuV#lQ;n(s|gR>tsb!hR_q
zNxXHeb%3(c@*?AvlM|*nmI2cBRA<-!H-nh78SFce!A13#>sT}BTU+^*J>)IktV_MP
z$Rp{L%h24(dHKN#WRN&ebEuXK?M%pEWS<P4pG;?<k9yR$>3l1YPSs6m3~r&BmBdua
zzE$6LW+EN)CUUx2g6>7}{B<i{S=sW?inlZG#!-K|yvdhi`8+0;!QySR8!>px7vwfI
zhPiL_{k>>rSG*b4{vU70RWVb!guds`<%J4Yj^1(;)x_JI;^~-*@=V=}=89^u!|o=M
zDw!B0-cFjNKCeFBd9rdxq}eAt&LrBJ$)Zd8e(w@nEty={s4VLP@{o75Ds#fZOwH#w
zcTv`}o@-X!46LSf3P+{tHKrOZI~hyoBwoBs<Va{d^~1E!@pde$-DBx0Z{6h^>idV7
z`J>m!JR8jdOZ&c|+IDPjX~aEDY}VyoNB#h5$IfS?c{MVc3i))li;SYNyg75mN3rj&
z^3TLuw^<QfYZ#$@<KZ~(4rfDPI4A09r?VIv*<M<5G1avnLdBv`-o}M8RlH3VZ*z;c
zpQR(;z8A_ZQ>Z*Ap`3gd%F$n;v=MLTh{1DzhqB%-j90_Ms3_j{5^qb0x4kMykTx}f
ztA`_$`K5e^JW({f9YxNbXtEoaXw}U`i}GfQ)K?B_DLr55{;oq}I66c6t8|H{Iy>b~
zl^;yL!Zx+z>GM2E*@wxvm6BK8GmRE~EXsYevf-eWsBHQ04qKTg-cGr0#rB$&kD2=O
zZRMk?U(-{*=`8KqDKR^alKXY(bz_*I^ILQZ9vw4qDQx9fl05zAGs)a&Rj$A0?as*O
z_dSck`SxR{%(MZuvgqq59cZ2UI?^cKZ%gH$e^SWZQTl)D817Hha~W;I*dSbU1Htm2
z1k=<%SQ$J!`IM#1#qgcjgl^;fPVMVit1NbTq1;z4BkzD^bZWYcJM))okLn7BiMQj$
z+sNCq*|B*%m%a=q_OHQI%{7p(DLuJ(sx2$)wPyC|c3eEJ*`1h<T==^)M>cmRq(&z?
z74JmavnKr4xDI{K)h2CzZ8i+7%>qj;<ww@w_v$MAm{pExSA6)s)kogjvPA#(qEmo!
z4o_=`d6*}^`+CZYt=#~c8#x*7LGdlpk;U5?;_VY>?OR=^-8jA6+2G~QuZM1QyyV8N
zb#9cK?uKoDH-a0xk*A;=V=t+ejn!;QfbxWWU8z-4S&RS5mo!~_amtsX)EyVz4VE{l
zy$k(vyRco_@j)?QVG$ReOYc0t!5O2oGl#u2+w#eY8e*$i47KcZBJzK{)pbF9OSBb>
zA31Tkyy{x<cBgpTH_?gbqn+4X-id2sp}%-LLmKQG)vX8R`8{>jNbnIOf18Zdh!n?W
z8A%^)#Py9mz0>X4mTZrc-i|wGk9V@P<cCHIc2lh@-d-y$pRsza!&-}>UG?Y2jwJMP
z<e7LIs>f3kC(bT*qL+BP!_A3eVtQTi_JVjj&Y-@nza56h2Hx*A5LMicbXRG|(vWwR
zw#WXu9iyuKudX#{Keqv6H#_{A+HrW4oit-3<Nh`(2T1z1-tR5m&g)~(r2%%z^RwgB
zL_1RK?78P|&zn>BjDBU$w`hBYFE;XiumhgrZ3pqTNGtW9r|68J?@iBEP7EC3#LlHo
z%9zvlRlI#9kJyyqnuBWT$>Qo|7}vWDo7(!&ae^-+vwTUI=&PA<U&dY4%=K_T{*?6N
z`>yhocwe4Y;%$>|)vyt754lz+uyYOF$7*o5eSQ2K8&U6RLy{5#IIWC~s^LvJHn}BZ
zgInWiZO5LqZD~2HFGo}RvM=`l+Kib(|9Xok>AwUw`KZb)Si-3ZYp~6`nJmvOtX>?%
z81Xh_e-N8*1mWWz%=uO!(u+g5We#IgZRvi;<nvU|t&ezX(!DhQGR=C1npFpznZ787
z^tCbMi%R6#%w*0LNFif;3LSE%(tcbjAJfy=ul~@e4(U2Kr}Ix@2FvejZ|)NdhfP-g
zk*1tjBa@&RR?7C1-nQDxUv;J3mb2nJ(8`Q;Rz``pTcyEGmKN>5SAAXGs}}XPD*MgK
z&Ayp*5pR=Yvsk)zKWm0bXA^HTZd)j$%Y00k4LdB9+L%t;RT<h%D3&hD;3(<xcq@CM
zyk>a3)0rsE`&+kkwx~`m?vT#!sC2%jr!)C%Iypa-6J1N3?VdrG6&Wn>(9F{beeM_W
zH9U=HJJR^QUgwITY4Xvhu(^H;Pevr-Bi>pYX~*FkarT-tN7aPG_G)(Jx^j`H#SwN<
z9;30+lMAY6D_`gz)wjWa$FN$woqWN}lQ=WA#%S)Rx0w}H%sf(k8~)g&U56%`1)4}4
zBJQ>|@j#6Iax<Dk;_XB?6XV6(;s5Ikv{c>O*F@n7(gk#nzkS6*>I(~AeU&NEMt<u$
z>8vf8&Q|qq|E*%hTUxpMVha~r%iA_rdDs)w2iTIxtNIDNcoENxGn%83$LqShyeCVl
zPT#CtlpC?Uwvm?X9K#^-)>gdzJyPcd8_kocZkpTG#Pja*Z;7@4d;eW4`Gv*Xf5qF{
zy`rf{G#4VG__8qyllDcP*B+~a6(iMWh~T#L-M?>y<09VnpBv7L*I~T352KGYlxKy)
z81!0tv3UDwOeiI<g>W=Igl(%sw7(>TIK6(|{tcm?IQ&uU?f)qh#M_f%awYNhpm^I^
zye+2Jc)*J=_Sq|+@NPKIW<)3lNHul0ND{@{L9e4&zFgU-N257qFfpZ+{AgY#3N=xt
zNKc(*^n7<0kI`9H{xInhY3VV#H^tIJK69Ix1nsvIGfStkJS3F{b5mInmdf`!>OnNj
zplnDwD|W{*ZmKk-ka+59Hf)S-919z1X4=)vP|bi1j#gj)o6a=l;*^sY&yLqR<Ju+j
zqIeofVVSy@TY0b7c+(p4An0ej@^KbzcF13-v+IOq+U@Zn6F>2`v1bAg_Qmk*ptOTk
zkvy6nshvxaxKxt=>~|2ao%V1v-!APb+QrT<J6Yd<C-vQSQh4JwdaYj1#2#yzU2Qd9
zFP5>P>oO{dx6|@3<CJ+Rx7sZvgb959I-c}tBj~+sAPa|drGCMVR848is|{@l5pQ?j
zYlUTXD^`qe#eY>>XfI7u^44v_l|PLsl2VuY-L)5Bcy0E7uf>L8H7V1zCId!Q*ACDs
z%>3<5NEvVT8NDe}+*?_^UX*QEn&~G!iI<-Etg9ze3QAWNZv(~KyPM^A5^p!P^5DuR
zcWl?XQ%3deK=HQtYd74^YsXxI8z*PD@w=Zg7Mi-@uAc2W)wie5x-!RG+OfO1Tga91
zw@Xo0yglSuie_!Z+d-P;QCGN8E;06#`l#Y<dSMqL#oH+HHp4?%Oj(Y~Y?Wr~tZYLu
zW3=?t@sAz2;pvFqen(RCNV^p$P8M+1InaqeX--U;po|IeHo&M1r+KP7pE>Z<$AQBG
zjCh*u>9kc$-C$2+)r2Re*wcEbJ)1_^6TQb?*$wvpzjnNkW{;0}J4v-=i_=CHRCHi!
zcL!qi7}lwGy0(ct$>QxG@pey7N9}=d<h&m1Gpi#J{pC|Wr@pKDt<9`PzIdsw{b0vr
zTh-0tZJ$R5I!-c>TF{PvUmCDx+VU;LpuJ@V?YFaIhxF#0!v@xTHBi2p9ZBNtnQQj!
z_{*La%j~eRmFMcZfeQZ`xaDrAEL}UcjJ4x|dcxBNi0Q|~+w-c6^*ld(w<B+1J2D;^
zNFFDi_fij8HSsljd8S)xK3Kf%u*gZ?Ts`)BI==+UyLMRnbZdKZRUY#`*UJ($u^f3V
z<!GJhi<fledXvj>>}ok>nX9im)0faq@-DCU<3X?=drtcC<5~rNRH(=V`^qdiUz4r>
z)lp6tJek~(a7#mSH4k8IkH*{|+L+MlEm{7#6?w$lK8Jhj|LB8T9`QD8EKaLtNN1R%
zdDVq9%Ua5Tn@jaOzJkwn*V4JyMm{#$s=XfDu~gZ`+Uh~9n;k;w)DYTy3n2&93Pr-T
z123Gj+aj3#G(z_U)g&iWqvSP_Qcb<PDv1O&PGr*fWbG7AQFe3+dvm37NqKzhywYg$
zUVe=B=|n9~XPkJu>~02In^=?$YoW$S3;W|N<bAIGt~gL;q2@mQtt3CQP^*>hCGC_8
z+1|=F)n5-%t;FrNk~7<?og}Koc3a4~l1~1OnmO8H!8gi6?o7=ioz$Oi>CgY@->_Bo
z(Vi6gW~XSkWhx=H(kQ2zd)ezW3P=Y&=qj#iZnnexG-hOpou8E3`!bDQkJ7l@OSSHv
zbpE@LuAQIhG<u)TFR{12H0Mo6)A{*MxfY>m{MxShAMrM1NE%(m+n%*j*b<<5ThBz^
zsQ2Af_s9|2O<>l(fdP}^$lofCEsk+S<;3En*%;qEvGh3`!?&;)HqX*MQGL+#gJ$g?
zH8W24+!A7K!wP0{8O_)~lNT!4q}hMfwu4RhsIRz2?Am`ZnuBMe@jNfj+uLY{OIHqP
zsyUmT$};iFAaK4#K1>VTuwW5O3yn@ENi}_M@pfP5G|s5@2`HY5cZ#yZny1iobTW~P
zlayPX$Xv~dym+X4y}Z?F@~W2~9?OD!s^PxIFtU0qkEEquwvS=Obu-JPC(j#brligT
zc9l&;jn~Z1T=hunNAW?vq;^%KH0KpXphpw|PVy+(MB%Kr!K!KdzKI}xZv@3_M{wwQ
zIF~1e)80d#oqyzsY7(YBYoR<(2&ISKmd+o>zpp}x=o(5d*H9W=58;(~J7akW$;-rC
z@wW8k5b53_1jU8$GG8c58-<c#QEjWb*rLAgf?;9Y%L?Pnt1udfx8ACs8x08O%e-*>
z>qY80MsfJRC~hu}ChtLc_jR7hu$MngyiISce!0#N_kGN48X+!9kBBP~L+#-)IOzPe
zvRE9y%n9@mZ$C>L7_}>v{SUPZR=tItHxo!a7*El$aX4>Nj+l6BQAT!rS<O_*Gd)rD
za~l^k{iXHh5^o3GilxlQSXOk3=iS#hE>@G5*j-r^zcbiB%1WJ73wz}C^OHuMcu_L|
z`Z?E}te&v=dvb6FL)AzBm0L4Xx1^Kn_xar92xWjra(Z?+HuHAV?$HiTciPG|+bz@y
z+$N9bHrk1|b3bn6Pw}-Bd$WeMsmti|IgqZ;m+?-l?cIAGlOyJ^G;j{pe$6E9{!CI{
zjHkr*t{g4Wm2oZ|==rfK1wE@#wRHd`rZu6_#a6tTtbF4e^%<XCpBdxov#3RVI^U_w
z!*+G4WT?%>KDC)H-Ugnk$;T4@(r2nrqh=ZEj`yaBw>Nvl+tT9gZt*r*-l*_c_3gyl
zKb<|bSICnyS3Gc*AL_GwQknK19R2Ce=a23r&U5EWXXRGdi@h)1sCQCX#x`!0DWpvQ
zd~U>!cjatfSCnSOKQ&!hk=K>Vj;<6h;7X5M+DW%i`|I3Fk?_))<qfr8y@3nff8^hJ
zulbvw&TJBIqpQdx<*&I`>A~?CnvEIa$Q|+aXNm*+jycfffwCXO+mCx4DRbS4X1AT_
zm*1JAs&iYZzTGp%iKVJD8|QUmk;#FH2Xu9I;QTnFyuGSf?}&Mx(rbShv@goQzElI1
z#u!-A&A{>*2G!vP`n<LyR`p?p)<!brXSz^c|Au(GO}t$q-o8(iM`^lxx&e+%7H^-e
z(&OAF{#w-k?dwSLTO(uDv#lUKIQL;Y8b~valTWHZWjpQfGw`jNfhXc@m%TO&(%UcM
z?amtpN|v;vUABR|i3VQ0F)*c}9Ua!#5&6Q75f$wjx!jJQdfZM|4cf<I;LtAve>JgV
zfOxw^Z1q&lJ5yTt$rO90%(urh%Z`%by6b%d_d*RE*Z))bxq%fCcFeJ{XNJxfi9Z~0
z*Yiks)eKTy?Lt_f{fDoWzudnxZ7g0qDD15swzuZ8$}q5+5ADU@mRHI#sfaK6r7tJ`
z@MXKBAK%vaaiXVKd$IyQ#oOP8$_(6JQ@-e$G<K>@(#YEMh_21jdUaU;I)J0bW{iH<
zivE9E(|lE1-fwQp#vYwX_|Ti;mxpk=%mm)Po<du{xi}wP#QO4!@fy0A-)W1rPjI<@
zht{yD;YNx~-@>{}+gY}97d@KqB`J3(A~~7jZDH}Y#pVd|Ka5~TjP5<sFOM!)KK|Ai
zE?MM5O-Z25uw?R=Od)4#3WGvY2!5qpgNAAHC#5l_c)IF4<pEZgCrNtP`XL$kZ&r?t
zc<XscOia<vYI%eQ%(aqFwbtXa77Ask&XN}VvYC~_;_VUf_U%rqb{kl+pP?CFA7$B#
zx4FdI`J0uexX;3=bPK%_#q!U|yi%=QuV)JNccoBewH}Y&I!CInDy=)CUn+|iq*7>Y
zD(jD@;@4LFHA@<6KPW5zseGlkRIBbvV}Ur@ZfzP3R;KZKZW{GxrfFt5P3QMCu1!i~
zs6OYkcsocnMEGD`O;g$HBJFj30*mEib!w5oC)I`vwVNPTyd5vzM(Ex;{HMIU2VyBW
zQM-I{YgY9@3~!_rRQO+w^jWzU2UH)L&0Jh#mVRo+A}#sSPie)6Ov(b&tdDqm789e=
zqd6ZXEm@q+IUP+&z5RA6T3Nu+EZ!GQrg&RR_vWitGDubbaQ2IIg65^M*Dsxg!%~SK
zoXVp1X?#$>-M>mI*TPe9ZI!}_#mO|flSIR5iEONyz$kgt?~4)pro>_FpnmN?s==Q~
z>nIni-A1wWvs3T+s#*D=%9#{{vh?_ds-L{9NHke<qL@}Miss^Nd-3*9g(zIyqNpLx
z_7HDFZbop%96<%~c2!UWcf{Lt@wU;}aIRGkXY6IwqP4=PtlsO`h){N43B@Cy>f1zl
zp}K@pKYu8r_lA(aO$dd(LU>nN+RWGxp3V%xIa&-hhmfkzX(!&!7jN^w52c#?&1=*L
z9+e%&6Y;j5c>BI?IAP*#q<CAocBFinIuE>$qNDOo3vZ2P?9*uJHYTdOYPMOty`sAP
zhBW(9iDrskQg)_hj8=%ZgU-lbE}bH3d<;XTNw?KJ^1RR5H_=>muxE_sCgk0&BTt<4
z=t?z|AEvWb!PDCNq^}t=(@e{^(gO7JuBiF1PPxqFaWYZ+gEag9<tj>B2(A*R+}C9J
z%aZA1QEr!Xh$X6b>%2>)>0Q0H^}A!alc>2y%{__R9i@o`=Zz-hpC}v;M({i^n4LX?
z*;FW)N{e?ZV`@9|=4_#P!7bc4w3(&5HZ%ME23o9J!<g<X8P_I|hLvZL>(NYZubV-S
z4l{6GF^8B{b0|4)4pWBCVO6Kee9jn)*N(1wT{OV&i$A%qR;7V-xqqA1<;A(WoO)gl
zw>R~8XH%bQ<?HkBkors%Z~H}wwLfYRFtY~B=GLIqChb$(;>(`hW!ap+EQ^#y@#vj5
zRb#yQ<KwN2R4>AMdtpARz2s(3&OP^F=K>GLm6AV7Gd?zZ^=E(0&OX(Sw@vQYPjjb>
zcssa<J0JhJVS3@liEK9#7r60ck{hK*xDh(pRX-2)Y3sRiM7$j@-eyQk{@;$+;G3n$
znq7)cXEg&GtQ_JLCjz6CyRX`IrTDwHu`}(zDj(vh6HTSb7H=TG(SNEJRo`A{;K-<%
z4xHRAZpJz=t(a<C@wTjZ+gQAv_FGpa<rJ54B1$^x#`^L}eO51amjib4KK<R!fdT#g
z=OMBuO+C|gwe6(085k_KPOusnB;MZdYGBs{1Fyuf22bqp5U(m=q^5XV;D%B2`9{q<
z%A2GsxwiwkhDjq{?!f78@-XT0=<7hC;SPj4Iq1x!8d9F*GU9DXlO68*_nY(--~STZ
z7uaCQF3GyNHuRWb!$9%Y>Ao%gjSMUdGvJnKz-7CEX(<LwE$nEtT7BMYb`0HUN88s1
z0`xf-DjJmcV#{=Y12^^fU0aE%s$aeI{0eljr*xb>#?N;6w6mks3j+_Q>i@29Af&f}
z8+y*UHmh$e-sWB>-c~YFD9lKP&JcN8JL3COc_W)V>ATm9mUq0U`rV6tpS@Xgt&H;S
zeR$Zp9IoPRWAXOocwf>#`*N>qd0KDtBd>THo~I(G9RH%4cU5kit224KzxeD=NZT4z
zIA4RN9c$y+p#dAyTXD;yHO1pvlfAGlH^tle?K^XzRaf%f?M<#rgNZseh92ksX7(AK
zOWhXpdC+2hino^y%So)WhGsR^;yHL75f3#}7ol9XhPx;hwToOmcXOyp7@>v3*;+Z2
z9gS5Zyw_e!&4oVGnf}SjI0jsfV_LFiOXYPel#{^UcM=%+GlBCCiA>p}JqQ1#knlW>
z(ihbW6mM^K$Y7;->oh+@JGtaNs$h{%SUX_8TBuM!nsP;}JTVq#@5`Xm0A)$k%b>fw
zNn54){?$_YuBXoI8#8FHYfilQaXd{qrKuEND*x;W<q2+*uDvUn2jXq%!6{tWkwU~O
z>AYg?udEc}3#5`P-uj8R6UEzcCsKLXI!(S*We|$D&mX5zOuW4>Z}n?&w4r!gc|{sa
zTBT8;rT#CSAtpJe$sd=-=pAXKiMJ+wPVqizd~BY|PWfY2*d~E%B2C5H4`Nlvq<D;j
z<V`-Iy2nnJG{T$vVyW38mbSmNlUvyk{lwdeZDX+irr95{cDt@k;;oCc#aWtRU3gvI
zoh<2@;_a$ACYt_{w=GRQQ1RABtnG6$ni)r;@zZ(8P8!4nWroG<h$feP1}V}m7gtsW
zq8|6o397@#r%<$>w1_JDp5<uA@zf*^H%KDqL85ZV6KSUT5bxy)nlnqFex(HJU5Y1q
zO1ySH$FaasGfdKspWak2U-Q|a;_Xly^)K~$SUyKJxOn?fye%o-)+s5UYm#PYyGAnE
z62VOA#K*fu(70WMUcV7EnI6IB>(XySBN*R5g23zHbegX|Mzb)AmepL(&rpWF31#+9
zaaDY67app;ZkqA=5z63r&HJ<trF@=H8ia@NcR&bX^8TF{Z=Z;_1Ez-1aaahmLqfRv
zPPOh(aav#BM0IhTM;IFih4FZtax!AWICe3N8h69EnHH{HB=QbPYYi1|Z&;$ZdN~U7
zJ-vM@eOdhdqs;Lb>B<A-2|uHLOS>D<%+u`E8nG?W8pHKd>aR?SVX=7oK>hgc$_M+e
zYz!rSn(-HJ-No2aVs4&#G4!vdJedmeCHk51%425qCzE!C$`4mqJG%_>$^9`&J1~hY
z>c`KMo}i459pdfDy2<nvZ~LmpzGpx>n+j+SK)fv`-oB1arGinp9aHrCc}r)HZ6Vaz
z8bXiMV9lHdvw4yFv>k(a{A&-k>vvIq_cmsJ-pq(1o5&~LHb1_OajRG1^VcezomUa`
zX_0m_&*$o;nL4Y9xAkU_=iD@M?Vm#PPm_5&Y%-m_Cv*Jm2*Q_iWZ&}!-2dcHuS->#
zx1<J}w$>%*xID&>>hb+`JpxPCr~R#3H16ilg{f7UJy$-_wSLT;>c`tre(Wr-dC5HG
zm4ol2Jj${R5N{v8@n(YNfA)KMbM2&d!Or$#e2)Cc(Vm#S<xiccc^;#*<L&N5t5>_Q
zng^A{+k>0kc{SCY$J5>AId$h;E_bT^kGE#|YUGbHsKza&y4GZHrJxwwj8a^+FGV}O
zzqzooD}I%t#!2NFO)bR_^=t!nIV-D8c@U9KtbVR6hWpAX7H^w~x4ZTJe0gcRHj^J&
zy#4sxk&O+-UDcC2b~}(G-X3}7z<POl^L-bC9ULiHPFWJtcb|RG`<dEZ=OHgsQAf6j
zv!|p7HyLQ8>vYwD4eXfVZO77lc8q>%;L>RW{~R^&-)L#LMguF08n90_sJ~&yHtDe|
z#M_zcj98}_DW)0NykhNaT@QOHXK{!FZp$1<7Hdtq>hy77uNYjih=X1)wyb$%OYjd{
zc~xy$C(gaSY(pt88_FFjNmolru3arj-KjRTC~3=<8@Ag0X<&C-1J9FG`>r$aQZ=sc
zEd$k`8Awq5+O@SEU7py|eW@)ORc$q|Y0E9u$=fy>v~SVCX+5_dKK2~#Y)_$G_RNw0
zIZ9tsO8%?JzYQEdVnab61752Q^jl@euo9|+*Ne~MZQN}m*XKx^*UvT2DR=r;(w^aD
zPtIy@)45|_v>5D7Y*ZNn8v9VALpjQcw;%HOYA>BH@tb{_dCV8j>%P?Z=*#00<+<<Y
zN4+l9@D*=+EcGYovOj*+YS8_Fd{kX(Q?Ye@?Wd{F-H}as`?Wc<>$YIYhZYQ-*HXM_
zrSnUBTzYh;)!@NowH{0CiwSh8Hih&jQyI}@4lVlxQfYf2n>sG#O5b(l5pQ3GZ=sTS
z`|qG#<aF81?#M7M7Y}8WcsonHJ=r^y?|s6F&o<$)Lf(=C5k!R~u%vu6PYNji=t3-O
zz2aD{9%bJw{ocP-{=OxVf4!60k(ErL_*6`FGRT;hK_Bt9K`{$X2^L;R``W)Og8@b5
zagyFVG+4dXVPdOl*cJWLc~LDL+kEK^nyML<H5p7;pTRHnStH~NZq-v+8;euedQx@S
zx@67|O2)rOGJ|d<^Zl*r0P)tkS2>g!Da<OEs{T<b16AAg+938GOr=EAG{UB(v0-%@
zU#=>*;zk;I)sJ;mcE`BNY2?eDMz>d~e32$y>v}5FkH|x$4F4kfH~#9IO7XVJ`IT-x
zJ}8-dF-g+klUSfLM+eQ^u25~;GAf=+1LHZbT*16kH2>2smc-rCVw=aZ=cl}`Nio#X
zz4+QAGi&9O>R>XnN37keez|`Ud3~iL_qt(XuWHx3;_C1`(lb?SO;RS<hZg_WUu>$s
zA0^&K21YY^i1hx^(frvK%@{o&w`XZIKc9xr=2Tpdrf|kLg^@i~BYjI^b#@Xz#M}H0
zlW6oiQ8T^D8rq-0%4G?xuA{!6ywv~GHSaEq<#0o3n=$g^UQ-r@>e<58&0Mdiy+uW1
zXnjt8x^dc7^jkh@@u!@6$?YmfF*Plc+?_Swt=j5#U<6+-X@{Gxm-!>KyGJ!`as+-m
z#M&_t?EN>KqMGTc(>jbvn)z9wtcq7Dp_CPGb8QR7bB8qLL!ku77q!H!*JD#@Fw$Wh
zTZHIwgiy#Mgmb!lDupmc{0(XuLTC9AR__YoV%ty*QM$z2DeB+09~j2KQDM}M3S+T&
zn=RhnI1)~%&Hz;lXdXzq_c-zPf_NJ&-qsaw>x;Kxs@Yu~rQg?$X0&)aM7;gBQ~HB?
zxwE9Hj~lEWZa;CfdkneT$8bTK#mh3<@AO6AANk56ykl5cF@|@lako^B(H;h6Z26k;
zE^4NVczZ#-?Iqs+E~d<xi<<4SQGilA_37U#?`>-=W2Wm|wkn3o3-tRX-6%!p!fwrC
z*{d1851p0k@kJRYuVOW`X{N^xJ@>)EoRQAnd`d9cd4u`pvYTN6+xgmg8$qwP^6S!8
z{0?cx=h!-aRb9=&oy!^F7l?fs^<@L+6EJc<*DL-XZ)?t&!N=M&IM;s?t1k^ykEs_e
z_Vr?3>7Lr1(o>#D<vz8mtqkCr3>i_K(Cd|X+|-}Z`~2x(R(|bXe*)L~v*uoPk_T6%
z?-5@v&-7#HLO+U$x2DFvxZcx_wKHY;zN9R+>18NdtqfnDddqX-O_kSP$|&<<@eJ)}
z(+tS3S)Sba?m?{y9=w)*+-{|qn#YB|_o^?d8CM@=-~SrsE<M4W%MQxg&F9X`Yi<-0
zZyU^VV^AJ9ns0Sww5=--^Jp)vc<U$Lo@}8#u-BF2^j$l3)@f&6fD8HSJ92!DvMJ2U
zhS=l8>jzGR-*aM}G}((C<)JOF-E;3$KS~2G^xR0bIQesT80pZ+fqS~9EL5%f!hxCM
z?bL4$#3<9D>U>AWZC2((kaXuo$~hjVyrydoOla%C@e`^^#oM9kgU<Gm=IUify!^k`
zC#rA9*)sT!jo!DR-A5aS4zblNsV&u1TXw5%OCx(RbFmFeI@oZxla2DBOL9Gz4b@FH
ze7kMKl!`VycCaDb-G+mAZFFC@W$#g2E?>9h*h5=FpV`t+yuJF$mef@?{3&OHPgY6o
zG%v}96D3*M#fD4b?KYz=Wy{&pyM%$&;_ad(1|A2Aqc;u8>@l#lg64Il3%Al^%Ol?2
zu3{^Hl8rKwZ0WkkmJ6zphv@(5P|LuT00V~p2JNkqzqzIQ!Q$<h>N-Qj+i;|lEqhk$
z-;l0cr?}>i*4i_$ijngRjYP^fRYd;sMkVw*5N{hi@??m!7hgVm@l^c>_sBBLdF4g;
z<T9*kq}__A%Tlnv5AGv;i0oL7KX-g7DBfo6DaZ3d<*-buNU<B$X*Ay-U-9<XnHroJ
zTnqQ!wWUo;n-Om-3~tKKv(5O@umvm5x8Pv+)^r%&hIH|Efp-UnhV<p)UxT>1da!st
zg!BqS`RqD^Q$FJ`pV#>{&kBxoT|ur1D_G&Rk$%%QQE}%c?in}JEJOD&@wSq9Yu8M3
z7J;Fv_rnO>5{_%n2zFc5bCZAO@-fvd`P6SbuS_as4)qUIPTpGO6Utv1-#Y;>A9;4i
zC+U8jOm%ZApDbzoJ5jq|Y}3eTqdHc+y}dL;_Z{_K>#3g`s~(m3_FHo`4@)I;y^H2i
zZYE=I5-SYSgWIJNGdxvih*YL3Te7&AnkL@4xhwywt@@|p?Z<zT^;%1ozL=ss-W1)h
zQ<!)=g^SWoZw*xD;(E<3iMMkC(kLU{+kJf+t3x%{vsd+OU-h2V?;YWyUaxpt^R*tc
z9?ygGIxp-`rRSVfDh*2IP+#pya*#jlV6t{!Y7V5IczZ07b=k^CXq3R(^YQY@#B(Pi
zjtx6vncGvlm}+YlNW9vTq#4_@y5Ba?43IoWYsK5;ab}wDGjnd5YQ-jI0(FMjctZDG
zX_y@^L}Qg-yiQ~^#e<?LxHp<u7wt3ljb=slXdbkSW_ssnt`CSNPrqo*&`49RrSJK|
zB$|E~Q{SXOP6`9QCo`;nGG)FbkxSn4cKUcnl#MOk`f49zTzUf0%2F&)T6^xr+qRL?
zkeA9A)j-)C<`|0XS1spn=Cyju=fzv6A~A%V*5e$l^TB(~CKWWXp;0vE)=}g;5Q*Pk
zkpzji5nUs=5GFtD0_}A3(_HQ12)6EvAXwf2bBhR4Plc0~5vE<BVT{wPPlsorbWaWC
z&Vo>C&J}m%jWXm9Wp7M~c4&mqZA6InVaX@u9m0F(5Ei(H;3eMv`b*k!gAf`vm!_=i
zx$3)>O+x=4NoO4w<^MH(Ti9MY1Svs45KvJN2RjfEuu#Op!~*?dqJrHDCWxEB(y)Pn
z-G$xV-HKh${O;$EeJv%d%W}OxXU?2CIJ}6jU(Vz%vB=|@k#j}lZ5ro!9S3pFXht4p
z?#ag(@;0-+j{B<$@inCo5%h6wqfgm|yzSMBnru)Z_T1+D5_ua&-i{@28*M9rqndeO
zd|tYWnK|Bz8DZ3my96++DTG;+{9P_NM1LW9n@`?`k+-$1aR-yPkNivVGOh$qyR-Mr
z*(nElx;IhFuq!Xd@<E*UszYDasv;~NT!{HAb?D2?tvzRS=qP3WQ==jT3wEF*&p}z_
z?HKZQ1OI<V$7vydNNruo+V-LfdwOrjz}j3iDagg&ahotbI}5GmWMNU=EbL#I36sZq
zOxwO1?F>ucrI^o*%z0QHJqz`vGtmB6DiYSspoe)Tvn*y|v;8dkFH_)ud;~M)2jlIL
zL8xPhg6GoiXflrTiR5iSlR)VD2cSi8OJ=XP#C4~ZSl8Ykb6ov#T<(vqv;5G^&;or9
zaW~IH59%i#xXnJB{FVo_`#cb6?}6Y39!RQiN5g0CcuU@j$y-PAw(CqcWN`M!bsu-k
zrnw+2gq&@|Ia~7f&Qof`<ZYep%!XKOk8h*7%RJE;f>3AlAa8%ycgC12PH3QU!r`e-
z$o@x9)P^RgY1jld*BtS%EwleE9dVkxokiZ7|KL6z@^%_|>ltE;Cyls2hqdjf4R)O4
z<81AHTl6JwwK2>lY;H?EhFON>?e7=Xa3F8>>#dphV}o67ZMYkqKA-^hF`HW=eLQ=k
zZ!A%7j1^KxS)pASeMe+Tn?h=+ysu5L!07=N7@WsGE$hF!P3aGIu!Qih1r|8TvG#%t
zhi=KRoV?xeQHFJ6W$5G~hp>elNxW?~%22*hhO?1m?oKHh9+fg5Q;OQUGW=&Q!&F-t
zB;;)tdApmu9Yx-rKO@Hx-hUV?MMjhq(>qGBrWG0cMS_3>68s8~B1|oX7kS(Njg(ru
zjL#dHOx`YK-TG25M=0yzig0S&<fQ0}oLM4rq$}kpBrE4mk;CDq43>P{w>wdfCi9Q9
zlEDzoI+;52-w+ElbD}>=O`W-zI`T%=$UYXRw^{~o$nlrF?zCBs1G{*?+6v2_THzvj
zyY(x3>Yff5dWpLR^PC}i%$`1Z>-o(MUN-J1+U15~9d{M8=kM{)4Wq-|c|P-iG1&vy
z;(?c*o)~xDle2f8SajACFPeJ6albFZ8~JlCIS^^qZLrb5Em~>Y!ahF)%g2PF>zi;)
z>d+Mvj2#g1AQ0WU0<|&GIQ2Xlvrolf%%)fz&*~5HLHd@PCm>8Q5^Zmd!q8`l=(}kK
zUJc8@lABBLDs?G$KP<z>@MTyoUB#@fb<oV%jOb4q)c0YZtfv+><ZYmmJ`?tIe@x6r
z)%JY6XWbIqREIX??TNB7*dAwAQgs>b@xIlcGWag9z{DAq$jCB6U1o$md3%q%4H#HO
zk8Tz6D#<tcsp|Xfg7x2B2yL(%xhm?qosH;o+W@h(5&QcZvF@r7RnzJJ6*E6Fx(Zv!
z+s`kma3GZYTVulIMMgL~7@50bL@0S1ltOL!sS)!hvwut8I+M3U$Xjba?xv-tTuYzO
zIr4VC`wk@dQWtK)UZ!#f5?M3s9>iKVnAbA$)`q<8&HDEDttwn4Z|jh^X5{V7ize;{
zGa+G#5noe{{Jj{_m%M%G&+PYqdelFnN1t;QI2B!iQRHpMv1K@&$v$0VDV~wDjmTTC
z)132ZUWzDNUWf0{o2nzr6eT#qYqU?-5(GUgM(1_hY0H_ER^)B?KISuNir`AtCT}l-
zgBg3b&iq)LB5dtV9htnHM;&)}Ykn<xJL$9@p_Y6OBMs138emvr;u*$>UI|9ryJf)T
zFawlM<gKFt`w!?LsaJ`rpB32RT8`1wN7GM~qW_9gM242(<I@tf*}>fitYb@VGY5i?
zWw%!mR$VEk&clqxQ6-#*V(!&;dbwT~;^ZG4*2;7UZ?41mGX>lSUVsnkd~D&q@^;Lh
zFtyIdc=jr4-%>+iAGIc%nlAYl&&;SmF?~|x?GV<t^+#%P%Zl1@j1~hLY3U)*U}%;G
z??;fa0U9hMZ#$8<sb*xWlLk-7+u|_l%iT2i+em}Pp=wMci-)ks+cHXvr{s}fu$I}+
zTIQ|M&wNt@=_TsUtd)O-7hu}Y0(|1l^rrxS>d~w0p~H?)9dZ+xO*K)6#_2kIT*I70
z_NB*;E`&dIhN0B`2O3#Z{-HL^UhPzB5~AE<oaC9rg1^g-m-)Bh#2&qgTs>BT&73K6
zdQZ)G4(ncKTRtIg)5zO?uZ!V+t(dxSF~*a(QRJ=V$|6+66`~z^`+k%TFF&%sE!Clh
z+V<Q!J8**YJ>7qmAmdXJglDxV=Krte6g^{!8U(M{&a9;Ecz2NAC}l1VoZg0o$G0Nz
z(`t;3UyZYUSHte@QdBirggZy(W80GX_>nakE(sISMW2LEBa*QHSQ2I(nt&zy#=`Z<
z2rR7}g8n0hVECHB*f%f^gUbd%`>_*_6|_f4-5~zm2QW7<0O$VsWA_O^e3;S#E6n_H
z!o?pIE7)CJ=8Md|zL1i)Ps!OXdpw}d@W3qAvbTLa(BInwbranon#A4iX6{JUyP!{9
z7u<Z&6sn8#9q(<*ESRR)c$2m3ZYOl!#Xjs|C$u~0gx<DJ$m?g1Mn25ux8)3LPg^Wv
zA2pFa)*A9Ks#uP8yXA-pwuJ9`ON=^giR%w75wzJFvR00;vY>|C%MpSr+%<56v$bm-
z(1>0q`_}Xv_uzh4wH-#%OT5C7dv4CzV)05_eC}%t7xK3DmksU|Fhl7V=U~`B-L}>m
zTKb4fSf84Gr1neJ-r=p1?<c!j;?#W$G&n_ulDB77^!d^oG@ra3Ih0y)a|?Vow?NNN
za@4&oN2-=u>@YcoH)TzjB}2VLF$@hQXzV9LWU3Tur3C9`5-4wrv1hgz4LZp1ptlT_
z<n7%eDN@PG`yZtk-B5-B7BVap$dFJW#k)09{GBDmdVY_`)1{a&Qi}29?Jn~6IC&dX
zE5Vi>67;Jl#r~~QoIN4M&lgg(xh;kCn-tGR%P{4(3^4}Q$K>Nma?^^uP5ml|{Ruhu
zQOMy>-o_@&G3>Pr(R*bW%*WQ)R|fy~GJJsyaXc@4Va+?+$^t#e+t)=FoENh|l(z+5
zh-BRNAVY>Z&k@t)__BlNjYd{*YhaD0LCnpN*x*Mq=3A1t2WQ(MYAO2(*)Di;%M~BP
z-I!<YhHQ-+^kdx+_t_1Ho!n8`+?_d59%x72uHMG;-P&gOp5uuoJ3Ucz!4r7o$-F9h
ztm3^8S=AiBo;8Po{;INvzSyks#RLz3G;SWmS(pF}yBCObVG#7|g0Q7iSNypb1K*5T
zSZ?T#NpS<A$xXoA5yMeEbrSrhOu@hnvw4o4kGHoLU}RuA{5qvG8#A4?=3?l)Gq@*f
zHQM~lM1H4iv}ApecZ+?nIqa<+qweRD4-t8*V^2<&!fVt@UZb`apdJ0DEr;=1JCl2E
z_Lt!nHRelE6(}FeT%u71`0{#xcRBr3&rImIh5S6rY~F=Ckw)z}<i<|ez21o$&f27r
zw|(<in~kOJxzT`Xp#jardQ_+?(XP4@F>kp8#*MQ#WAxlHuSWy=f}fr=z~06Pt%ng;
zr?B4KZA8s0BPRIMD@E4)I%0w^y<AP(Rv~?L6|#-on|GrM9?$7FG2ekY?8kPemOQ%)
ze?EEJjJ&N+-dd8kTc{<s-CKqD9aWe*w+c@$n9zNt33VEpP|I_~%~+lXrZMA!ynWZu
z0Ks8ByvW<V{J;NUu7B(r&TLL(W*+Oo6nbH{hnFCdp59d_xSL=XbNG2JzLZb@4fV-x
zQ_0~DoWUh;EAp7xC@x0+IeMe?tZAneA#gwuJm@=}UatrWW|ia~Da0w>-*{Ds@iO{?
zsqcq1ErMf=o?blzT3Q$|vVc489vTt*#t64WBP8n#DBaCkJcYBu)F0k@RYK5|T7`Kz
z>T#Cz(2P>JH7dm|<~o?QDTC=&33rq*Ct)mgjGN?tc@cV&w|!V&KRUoURn91d@o^Qu
zDMY<}g?xPUNRbQooXCm21-SW(nusbN$tm<f4dOj{>qp+ccuehB#q21)t}4dmp$mB%
zY{Q!Kr54EsW>~~&p^~!JJ<fWS+OUQmsO(G)a~Q}{FAWZnv4?Fn=t8!B5NNQrg@(QW
z4bB8=@a}^e4k2o!khgK<ZKn_|CJ)u3aex*dYc+6qra=etwmMIP4Zk(`l&8hj*X;4$
zXV%mY*57q?C?#)=<ZW~E)+Sj8(*hk{lea5F3gO$Ev%|6UeO)cYyot=)oKlRUK+Zjl
zreBQNBNM3=?@c8OmvbhW=cxWwe4o7axK;vt^7bn~f9N=8ldFo+yd86->3P$#X5PU2
z5_itOThQy&nCGSeg~;~Nq3xOi99Wdk|1))h%be%oOjAAn{SCdEkKavsMmf$|WB&hM
z^Lg?et$}{pcKi}=huQI5B(=#!U2!%7WE(l>vIY;HErgdtIu1^o2aEO7kP|QsvldT=
zN;na}woXJrMiR!~8IK`lqnKAY6jwsITVO;CrW8kF@#07bhDG9ik2v^~w`0F{VlHlb
z%qeS)&7Q5%qofs1Ci!!>n;!z=Ti{h!Uuft}spU*$Z5v*@xM%jKdo#?WpXFAy2ls<{
zU`=0l^z6W%o4q@7db{JuO*izu#NF=8T`}l|6Ekf%BfH57u`8VLZzXHgZB96*AaA!j
z!IL-LGACp_c0eLCB96+Ksl?gUUx8$8HD_{iZBYNJ4WieuMj{)U-eIqDkR|8&EU}u`
z?qG9U&JR)3r8eBmz)VDwH3Fwt!&GUF?+<NoGKiUo<ZY|rc6c9SM-Kq&+f)4B8O$Aw
zpvTFR`AO`{9#pdTO5Vnjx6NB}CTFH4B;;$O{T6U#?e}c51ri5YKoQKjnKl*(Zf$|x
zgP9`{Z-KjgEKrxcQm|k7-GLq`GYiaoFK0#+8Kag{Z<Qn4MUH?C?5PeDqx6%QK1d1f
zmP+8YO@fu=?aiBFl+6_5P#EjpZZa$&TQ!g%FGGU)xe_>Bkie%0Ik{GfrR$}*HC2k_
z5mFR{NYP3vh4!8p>SrQ6s1ae*F%jC#65)xZ1g?K17`{b{e}|>`e3o@C@4tMO(tjnx
z&#N*#DIiz9EU1UbaiX3DI$h)QYcE4QA5+CR3A$xSF+@)dyIO`k>eJc%q?pQ|dF7N8
zNrUNk4whqInjAHosg)0x<9i%`N8xgmm{I#qqV8RXdU9(y-Wqv(ME3Txz=rG81*A6I
z>1l%l{Jnv?L)}JA;l;VI;<GNWzU_(^yIt|=nhT=1&+ij^yn&~jk$cP;dpSdPb%iVb
zoTle}ttT9FJz-1UzGJN$cf%8jwVu!p_QHfUUig#M9EId<ovr`Ng7m_s4DOiHd*P7L
z6LpV!B3l)J|3<WeTGtxR0U>yh+zFr8^numDf#|MIK-{?DFp;<Q`%cDE@2O}Ol!7>$
zS;z~XgSY>sVfK^x++m-NaGwl}7c7L`w1vzuSwa7J78)Jif~&DgtengF#oa2-0jf|o
zL&2Oro}>3~$ACTCVJPJJ`%gX!6N>TAsRFG~fzQ1wsQ+>XmArjU{;hmW9&RyWgb#f?
zr%kX~uoI)>cjCX$o#^kp6O#XFy8oD6r8J;(v;h}a8mP+~(3*T`t*As@^0qg5>q*}B
z8LNk0qsK`NJ+&3g#b0QE<`Mgnqm4ZO8gYa_PfOl<j5hHMZ-UQL6IupUA#)}3C&}CU
z*Q#K6#y;+EdYzi=z#c!=ww-n`Q<!z|=_>rH<-W4lRVX5F8<Drev#JnFJ=%J=36}Fs
zJQtggTf&Te^0qB`YjQDQ?mO;q`^a+veNsIxl*8-}{j2P|ZRtmE5&bg}7mJY^#-7_;
zdQw^Eb|!CA$lDO=f)iFT3z5Cs#ecY8?>=)HdF?*Q8Bh)9Qj*ErqvY+U5N1ZyVVz6n
zPCHkKMdWP_d7Dq(M*b{>JF_INP;2hvqsRP3dMy4*&uTIC_6J5x>1n_tF9ZHjLms?C
zkJLatdYrGs?!-zw^Q^?lG-fqMlmin=arIvb{9csc;253>1~OY|JnLKX_P&YzS=QL6
z$pi;7w<RB&%kd&~iD11Qz_S<qP=0d@VE3y4PWR}CqINtT1(>K{eal&&b24)F1-*E?
z^U!Nq9>&ki<NPM~5R7IPGPqlnyiNF|#gX$`tn8?T&ks&{9@fAkhkmGq8XTNS?vl5g
zSlb#z<ZA=gu7A|%ct(u|=hT>Iqv6g{4WizwvCdr$j}jFcXjMqgRH5lw6{fCHA@-mO
zCth*C++g~L3e_0STDJ{(d+vl5->y-A4$OnbnHg5=^I*uyhf%}b3FQS4rs&{G=5DAa
zgHP+YL%k5o=>HvST?A_Z=b-%Qd*az+`08S2K$k$w-}9bjC2-zef<kHlGpPY|3!-*x
z%ek!y)T5cTmB!oHH1^#2d(OU2FC2NB$LIE(ya?M^$9|kvh`#<hTv=6s3~B?)Ci(C>
zLw|K%9zGn<a@SY^_par^rAP~Vr51b1+qU$Ky-wbal*`+YYPk(=*ow0kwjk!nW^`+^
z64iZ{phxGq2rrwC={r(T>NX8lfhicaeh|{khU1@OJeKs2!ns3zu-YyX4SGi6(b!0g
z{x1@H)9K4hh(z;4y<jo8BkTJ1$VqJn+s3UR{T&2TMJwj%`=i;W7MKy;0)^*%5Wd(G
zlUYYhO83O+86NN)?Sb_hnax<kew)-4lcQV^BczwOA^WW_op7$k3E#_|@VbCIddeM9
z6vsX6oVmHm*_*h3+#$zo-9OLRL#?5I$%(qI+!i4tZBaLwb?0GTgUL|U9UF|{OwMNd
zc<$uM@q2|Ff2e&X-;<-Vfd$0WX`_aall7S&$Xd94A6s-LFKdm~7?ETR4*lZ$W*gik
zw<nUfmV@muKAOD!XN%GMS=TOL?mv0kn7n;P-X0-u_tTr?7s^?hiJW&CM*q(r3tZl3
zflPY~lnE?gAXC2vT3{_%+mg2u@^%z?dyBl4hmynO?Il}2-V|!QlO*W%TZF*JB9uNC
zF~?GjF3w_z6U4|QZzs-?U|@~}@hiz(vQ2YRjL}oYaMFk{!X%_tD#m)b1Xt4~n8vSr
zdRBt_b)@KhLV|BDBF@T-u&7AL{(um*HA197glPjroKF&A%TW<7P8MPQ3^6{=lHfz8
z1SPC_Ls&0wKP1KcCsHgNC`0pcG7Kz~!^f3*kL2xoYQ<An7w?-afqNqfMh=r8w}Ow2
z>~x@>-SfH>u6?A;{*&VLDk)ACu--i`Ba>Jo2g$LFdilQ^8NQOUh4opFPnMy$t{gh{
zrk(h^yK_a3R#q1HRc?WStZ$3x<@Wi@tgClc82*vD8qPL&O&{1@p1G_l?C3wW$H=Sf
zz2D}Q<r8PQzA}UJuRZ)mxuDliSNMJB?!G0SxU|C)cgfpr*FBjx?uog6o-i)=LXVW@
zxI&%zIz7&g|9L?@+Y8-U|GHQ5bsX)9tU8{U`q&fglwKG$HW26camVhFo@kIf0E=`9
zcsX`BE-T_V^V1h~`zPS+u<_`zXCnIEo`@w4lJV3(3GeDmg5tpxeAi}B3s?x(Zi{f{
z#v+_gSdDWNH^TMb7R34F;KBVI?0uO7<Ch$qyO0aR&h029Z_hIO-H!Xbr2h2Vb}7e_
zo#pI@RG>b2YvyG{2<xo;fkymhU)I`az}e-TPn=*tYODd-;q=O~Hq0bjm!8z)^r1>z
zF{j_TsUBT+>LJbKT;uLaq?5N_n&=T6t%ti>kBe&NJC_+aw`0J8L#&J27%^(55j`rY
zVT(-IOI_LLxCxJ*nDDSp6}sOrA^f%pGaGQn0DG}jWa<#|HsgQ1Wly)5HSc0-$Cpd0
z@QOU%Oy0h)<~}6)j@-%H&jlu&BX0#k+y~73rI7(fwCiht{3!XgM33to^xTh7iIKDE
z_r1*={AHzR)0%sQ*}I*5wix$U6r&k=yFP-wx87vhVDgu~8x^n3b;#R|?AvaoUq*A2
zdH$UJ@iI_9%;o(EUX$5B+(6#;{ac8MHHDmYVLlV@H$5vv2lCc}yq!hf&U4l?FI$f-
zpY-@MO%K1`oR#)8V2`r_bN=aZp-K<4AlA5N=_#LIi7CQLo@3ch{?6PA-V*=P8})?V
zE%G*|m^+E+jn0f?9ZQ~ETf-i*M-ft<=pa0=;~wb(<|}c|le{gZ2RWF$wXS7dd#V6E
znN!^Lf4rq0vTJfa7Ld1>p64Nzyxq^-1&3qwL`~J=3Huhy?X}QaXzAIbAAXt^BgxzR
z&l+4fpuw1}8hobj_?x2!xqsEz$-4HukSBQZ_5?H1TAfm3^;tFNh}5i6)M!gjuHIaY
zBjl|?qe29EYeU}lSf#>Zg$h63sPJ@z8UuCYw3Ib+fA)RT$X(XD$JjT%XPXCU?>rpP
z<{_E2_*!O5UZSrm=W#x&9SiVvyAER3xTne6gEs7+?_nl1voXH%`I$!E9;NQyiSx*Q
z<LP4>$?R)pl+0{fj3m~;gXlfp>s*K*gV?(^D}ep;e8fg`7ZUel4ycolcs@VR$yK2r
z|DHG-WzXkoV}3r02Iu3A4YdLKyl$+a&w9TWrtPfVgVZ=4rh>J<66>9}BYH+IG8*UN
z!P#x_kKTsmW3#dE$`)LtPwI{9X1G&BHcLw3nK}uljwPaB`T*`i8-!l72SL$o5Na=S
z*Q{e-OxN_m-+8^DbnMD}{7#5n))BJsj`%eONV?qtu~{AP^LKl+U+&MmD}O|<@<$1K
zwTEVVBk_nY_D8kA>{3tIxOn2^jb_NEXQk6VH=O+7gbzC%V4UWPyw#kY?B|3WCnwAr
zVh`tUoQ-+Mo#zX9y{yk!8}>jSvR{|o%^Ft&tufNk8g=SfqovY{wYC+WR9fMy+6vy0
zRwy4}h1Dyq(RMm>4TmrbU&K7Y@fH|L)`qio^6o9i^lt0}@;0$2Yo&qo1&!dx=tH`|
zI_(qdr@7R0s~<DpLCGxPVdQoxvz{!O=|6z^fzI4t_uLk54lr+V9(SzsdasVRMnyVj
zL#^qxJ#B%;k(T&8$`Ud0mh9_VAb`AmV@IvDt_8Ga^#9b-yYxhcpwq1N7RVs?mO<bu
z!zt>%4|1ij*dT>w9qY|D5?uc*Lfj(}I*_*w$y;yoHa=R+UB4n^s6+^J6r*~+1TQ80
z+7n{jpCaZiED^aYMCu#?yb}f3+eg404*`TO0(7tx;N=^0q%0A@uuFhTcLm7g`+@wr
z<Ld;}Km~BPDnMfkA?7?2B4@Q2B{L+*UM@kLzZ5;vSR3o4=*K$wC;Pjzy2wzalf#;|
zc=BF3s>$1SUnP)?m%x<5T035XB=&&g|42DELhsczDV(~qwrwlLkzUlesZF<Jtz0@u
zj<_Cjhz`pTwnN4oZ+gM_@xD{Z-JLQ7@;U2W&K~mtIUJpMR;r*UYndhf_OwDztrdQ~
zwZh+K*7(5R#TmX<j*epQJkti9>8Uzh#5}5GGPc|v;#zyOA#W=ha|dw`2mC$b2&Zrt
z<Z=dWCp@_a&l7U;_L+w#{(IyBm8BO-$=g5vUWni8iL_!*-1GLt=MJocdwODInForf
zO?Nrej9CyNNL$(-XAC_Noe+VOuOhhLFbWnqQMmsv3ZFv<VvckeQZEeyl1HNcr3v^Z
z=l+9#|H~Ag!VK+Xtb92aKgnCyn~N|!dNI0JujUTSwRmckiEU|Xv2xTVBz??*hP)j(
ze>;3vZAXs|O5~;&p~$ZUx9`)le1yAF8&u*~S|z+bGmlwsz}wbFa@>gI(*`)tAS0<8
zw_`1P?}VN`4LxQWxTBryYUf^wtkg0z$)x`(v;sF4R5BaCl6ktU$;ewllpY7O_0V|J
z8#LYkH`bVU*HeRb=g#t3^epRH<GwWF2W#Y()Tb5oI4?xKS>eR}YAO@@Fz2ChxCtld
ztBRP(IikDVbNbta$>eR|_9`qPZ%1rmF7oCoT;L2*i^)~EXj=tSz6l-3+s5SW6EZ#Q
zD_P%%EIXpdhsAnm+v>58yxl&BGg3E8v3@CcK=_nm^OX|jH8GDrg?Yl`iZQ^0zA5t7
zMBeHWOAxWRgnQwrOY(Z1$7^@f2SsSXojN_J-=;4u!mVV^Ka+Eg<gM#h`gu+j;@7T1
zDEAa%40-$TBWv77)FbVwJ@agGe=&Ek->5`HUp;i)^=L_5KAl>@p>KM~^7T;hyz;0^
z1%msRL;kJ|$F`Q@J~hvQtg|ovWPKLLJw-pr_5$`5$y-a#o*s5B#?;k(EEicz-_fDz
zDIJcpw>jiWK0-YTklu-X1=h6tKGPR<iQeS9)R6b)V;$?n<+03WByR^k%)>-7WH5IQ
zD5-<mPSL{8js01&t`9Zd^ViAIks9`vIPde8Gd|?)Bzm8!MzP;Y4S6_u+l;lX`4}}~
zZm2kKrNT~oHAaxP^_Wrkh%+#n#k@6EGrv!T9V1jQi%?-AZ?h(;FoNtgoF<<dsj-H6
zh97>Zu_cPRROIbX{yg;pEmk$ogJLMTw~n4><l_i=+lsus^(Y^I$=f;P?b>kqr)Csl
zJ?qed-9-@bca}JcIaK`pX}2<?WNb0_h!<mD0N;OF2&>`L+Fc4UU;?#vK38WW^63|2
z?Oeh6BsupAHq67eF?neGAM4c=?)Od2!_pObh@+Qm0&DD9{dit^pNE1AdFZ(;4=cZE
zF+;1tJqI;*leczmN?5UHJD$94-zXPv_1h5FcN_QQZh*qb9E%2vxi5Jl2KG(Df_;fN
zVlxDdpA3Nci~(5ku0N_}1K2McfRlE8abR-<M63H?>Ba8w%nrqpEuo0|5sFE*KqK<D
z-n9-$cL~FsL?3$FeIZ!rgD-uUQTx^#0`j*0J#S3&;jZ^gPuP*SkrEG#o9>EL!=14y
zl{-vj+>>*JxlA$KT{p!B)28xTX|lrcG%I9~o4Ku-scUA1_+yqR;I&mvh7A~Di5$M&
zB+nA<PFNz&nR$Q`<|@fKd)mSZjXPK&@Te8kWUS<w9OK8*uj?wu)Q)o0AzMvkY&2PW
zqKh0udGFpwj(pxqhwx)LtbgCic}-*Qkp0y?8=3phnz<i&dq~Q>#VGa&?U{M_jyvec
z+nJv@%UaI{DZIuv%CJHNdFy(Ley85-)%K^ysVn`t4=m{Ww!or4d@SVcTJqNOs|=w}
zWr#h>S};-uA$hxl{@*zA_6oILbMiJSO^j(P#F+9@gvED7kX5n19VvoKXA#Oeh+uRV
z;c5#J#O+1gNiN1#k%SpfV!WOpMz<K&tK|ad{}bT*U;%8~^WIIseHH?YCvWFIH%HA8
zb96awPH(z76j#krKS6-XRRV;Px8iF8_A-P>+a*N2R*ZAW5=<m-`;)g<=SpF-nX|av
z*caX_Vcxcc&zlV3^Y)9qanEg17`usasjV2z6UCT6Sd9OMu-7b;($_CVfCqcO-coF!
z2kTZ(Dff&^;rE)KdoIP+7%BUhQn)!v@rwQBsgI=C(@}=>Ju>ctr?yOuc?-`(l{~xL
zVy`=Cqa`MFu|lnadlg<<;h{UvJv?JAB5xBp`}D7ZUNz<{FPqP~p>Fo*H^&~~>D&SF
z#~x#<m^-3keszI8hR1M|R5wTbTipbqoA|oC&;+?TPWbcN3BCAQJ@DEY_Xjma*TYRQ
z*xv;O=UuR!*&22DTCVQHnYHyE=-j&*ZYFthCfpOL!M^ZI@kMf4XY?a)XDWK*Wxp7N
zTgT$`lvp%f$$eO>`{C4-;Ye^!!iQ+?rwp5d!+Fzjf60HC`C<VIk{06ZkELi@yAs!y
zt%0Xv4L0Sk#mq%n=qk*_=&xCvho(2fCmSJwIjBCbLZdeM_&|Tm!{B^m-_E1{$a#?a
zMbN(}#?wa3()m$>-*WnKQkb9dx)cpNmq9nG3>(}l@QAm+ttz0Y#~iz)GB}X8Z^_&1
z@s*hTk8>(-DqtK|iLKN_doR?ZBX!&~*3T1?4R|uqfQ2ar#C+DHLw)8u`x)?gnE^A%
z+YXJ4?8zCiZK)9@ye(U0!~mG^wU-Hg|Cw-uo~rFXO|X)3&WCS*jUr<gaOXOC8#<W2
z<Q`S%{K|w*nI@b{GhwSAGprVn;YK~|)0r!kq~~m^9zjnl@g%qs%bY7wYbwQU=Ip-l
zF2$VO5`0ndnmmEI5Mzq5-JN^oeb~G0!I??&_EvfcZXTjXmDlRua?W|uAJvTWJ=b`T
z#YK?MpzqkPh&gwK*#DV6+T--jlDCQE?Rj3$2eN+}^Pv!J?bu&kS_#>rO3oTrV)VjF
z43t*ll$ik&Z}K^NR*4~NSkJa+?qnGAi~5!$?R^=0TBVqHqy#5>lDScw|J+yv+mS`6
zYE_I}zW=&Cch)vxzF1fhQa_Wi?A1SIeYT7}#J$>lL^7+QGa0gbF?ZmtA){7tmtH|W
z`s?!XXkI?<#4yW(yv@5x-Pk}aIg>ko()il&*K)2^3pXXX+D3z<a19pI6E%^%ZSqP3
zF?suBg9djwXS<E8J;{2uY_FQWCpD&DQ{nCj6%Gs3NF_gi@%|L~GrtY%)*mVypng0m
zPKB#IR4D01&W=;z9qU}P<LrktRO1VIowh)Q1uvAac%;No-q(GtM3-765`_G>E-Hjg
zRUw+M=|vW5)OvE~K_I>P<ZWPUKG)=})!stv<MaRh7<c_q+aEx#PGY}$C~Nv#>|J;J
zoX`FeXP<bx#d)X;Mfo`YfoBf-#T2i3?r5Kn3(VoDp^m<ayp2xa&Oz4JyU5*{YxB^w
zG!M4a;1hhAamlkv26gVcOX%h9k;lD0)Q{V%(d0Y(wEeO%%rP5T%B|2f$mMls9maOc
z#H;n|@bg#}Zr59k#Y2)<hbCe~!YIUF9)>wPhoNorf!O%EKk5wV$DHxLc<0xZKDRJ*
zmWSdC=V&8&8#%Nk%tHcT-68;!-}>VHPG5YO;EUd)d~o28H-yFBh+6Iq^M&5%GRqs|
zT)Yw6fwPXsyx7zBLJ_$e%bCdHJ<hngrwQsja>Sea+@Hf9sbUiQtDmglf7lAU=907h
znPpGjet2vNC-OFdjP)PMnlZ={b5=1+p@{iK<ZT#ve1yCmL*52@TOq)anfS*nU`b}0
zk+)06$uUnRM}xOAbp0ShHy?Ur?@O6U%UMz~?#M+csveRZ(X4gJ+iB#+?%@{LbeDb9
z$5wbY)Ec2WYji%%J*~rS;6^R^>09mz*vEbB!)(y3felt46Ju4G7(eu4WN<b{V`a(Q
zF3z^_d)(x0qn`zK{+7depBx(|%W-t63^oU(_^?lcXHUi0T_b{%LWGnzLgs>sFz=iY
zb6rJf<wiaCiU`5{{I(JiBFWnaZA9F^DZ+bxu9o+vAQ3$N6Jf#@5&VipXfj8H(?<ok
zKyFS<5nx}$|7We;MRUA1nWK%$9ILjOW8PtNd^l!~zr39$dt3iCM@~xtmgWfXW2XSR
za{@#x65@QUkbBG|%m|WTS+oQbx=SGHCgJRjn7bV$NPj`jKbA1VP{Ld#3BEU$z_KfA
zZU-@zxboIPOubx!*&-=Av!<Oz?nbz<pKB|{a`N}wFe$Q=q)5r2#vCO@Cs!#J*|Y9V
zkYe*)DXhP-E@sUf&i?O@FgXge)Rk@Nr`l%0{7PnD!3qb++o|O3ds}P#i?hb`Y1~UR
zp0A}F){sYXZ_-v9=$vgaER&g9^tnBrU`L;>9b$enAM!c(ACk8(72IL$>WGnPj<C*X
zf>ZPa)ceJKjN`cP;;l0?LYkuRP*eDkw^J{;KqqkHUI)&qHT8ho0@lG99ymC~6FpqK
z@M^6$0voi!!aW`MH`N8Q&>o0=&<#E7bVrxL-EmRZ8}}~7Lb#(J9##&=QS!ES!z4tj
zr@+lH5AUwegCRT}zbZ1&!+jAx=dHk|sFhfly&CVQuR(8dCfdEvLbLmu@vKEQj<n8!
z*}fdC7^lF2S6c3q&BJu^7H-tXk_(v0$UMCIoHs2jL76r8lqZ++I>r4sVP%LWZ(9s4
zr?<Bpu6`AmUt5lLUrKTKHM8Q)%W$B)91i4d%eKrFJW&o~+e+k2pkIo-jio>M>NPzk
zJl5mK13i{KW9?hVfPF0t7&6rWC!GPFH3sNj7%<4qh;Q?a__f@KW^&fGJ()!*F`>mu
zBP<GxNV#uBE}5zbGvP(CiJnmtju@y*pEF_7750MHoA80W-D}4l?mZ(kE=C-#rPf`{
z`h7zsXJISRxNjwvkhj_7?ObMB6tQ=^Kf464wzEcL-Ls?v_v#fEVt71#uD^76c3g+y
zo<*?P#oa!+#c24x2y@8UWep1vL?2q+Haa{!!P`w8tjN`k<ZUDJwrf=(-t8=8)^H(4
zu(q7z&U}eydFcMB0_VId@trm9ung+*BJO83t3Uwvs~65?jwJiW>D05sdG_&LUW@~*
zx3d%Yb2+~<xI6dTM{+OD|2#Vj3t<YR_c(&S+2CTXl%nUj7x$4r*Rk);%;P%T<8vY(
zi#YG|ZCpOq^vZ{Uyw#JpKj@D-&FrSh<N2}FeAbxx*dV08=XxIe$lH$OZ3uZ=K;HhV
z&wPtTtSy5$;}gtW(J;>Vkhh)Kv#n>+pgDcXIZZWKwSrt_Z*~NETQW!ur<F<=*D7&h
zqzc{0&vWGMl_u1e$=i9~*sD!b;X;2Es>$0rJyjSvR)vJgWN!#-VrLa+*i;DNeee59
z+`Yp+9=xSrQ^M^jGsf$xFo3+xoTj3WPX)O|jr-*7nKlKOXQe~NY#r**J0&1*=aIMW
z<ZZ{)LOAW?`C~IPJjmNi<n8}_#nXB23?Gt*X<hO#iGLTCoZqQEMeUrL@NeGs@MFd7
zLo4!`Q^|hp63!y`pF$rQ>)WD&JmfxMpW7!Ns?&MsL@ilP);{OVK)g(e+AiBLVq-Sv
za_EV&$>y1VD{2O;L*+2e84Oy7!Cx~`vwl8gzms?!9}PwKL|h#>3bEwv;p$=V^csky
zr~O$E^+Td4n!DtCp<3P*KjMSYTOI_}hyV;CZ|Bu%iF>2{kyYx4BsV|y%v<1|$QvIP
zHpjvzUTC@03%6E#p<&zRh<EhH{`Spb$E*QmH!mC<;R%oV?r7xY4DI43kTEYYg;^1m
z<ZWJv4gH1InAzPLM`p85JZuS%i=3SyZ(aEIlWCU78g7XU^6TYEIclHEQ9PabMSocD
z*0I7DX5U}(vVyI>6`C0>@VPrV$3E={YRCm5IgZgEd+MYNPA)Qhza>Qn^7f!g%6TX$
zE;Qlwzpor)vf1N(z`iN_tM6Sor{c)otBE-hsnoGAT0=qJF8X5&`7_SWlDGf*@{GXx
zw*FE2f=k8pSCBJ*S=&W$9*3N8SuI859y0WGWB>LidyspmnUc5F)Q~r`e%q;%;Ak^3
z!p^ftn<GLe^0v<<AtsTx+LJ;AZx^EX5fRJ|h;V472-EwCVB1;*owW#(`XXeSiBQmr
zeb|{I3?^@_$=ju~MOaJT-XL!`CkrsSg8*;ennPY{4*N259LzOGz*ckAIb_Z`W^<e*
zcTe)(i@Z$=5Wtzdl~f8a=B5BASSQ=^_M*2C$GQozcf1fgZwS%2iwNe_t$VeTU@&{f
z=eE*QHHtN?FSX+xVuZXEL1ZV!^`>H&sDBS0Cq|Y7>tptG&3;I*r@j>3tfbf&%=$G_
zim&6PP_S+u$G&c5Q@$;yCu^t_hik~(FH+`q$<Vn{hK>9^dZcoO$kGBSo#;JV&Dr0D
zoClsI#h+x(JdL9ce@KGmC&WnO%upZlHiW&1O*!oGGEcL7nJwBdUu5ApJMNdUL%Sb#
zuw*u8A$hA+ahJW1BSIHBLipAZAxoTa?G62SfzH?*?+mZ^&ggujDQ5e*Kz7N68HR3n
z_sI<_|8s-%r8|2c9@MNo(J{^&*WFs8;mr1|_1SB&4afOIouE{9#ed~pvH3(-G`Zgc
zB|Z_@`!xcK=SLyv_aNw7j6l|!F^Fj~88<^FWA(ih?p|Do*5vKHltpMgbU8G3E8*Cj
z8CBh~;6HUUEbnZ_mbqK-R<{M?-fqWDdj)&cTAbz?z0o=~#?Gb&s8nOYdU{>>@ssoN
zs0HOSUtI^cmpW|DErd5U%C2VINyM4b;q!~IlT4WSsuZSq)HVN<;_lHh#0@P&M9(sG
z^km-u8s-4ps(_Sz$v<%h?i<#l_>mqFSM|)%)nmh3JqA%f?X}!M?ZAK&XAJmA-kx=3
z@3@B%Zx<VJjJ&MbM(ueXcf>E_oNqq!pdT6WJ<x=0RwfL%$d7B7Uq#K>Db#@eUzm0H
zT93jP>@nBq@ofbe&wA1MTqO>Xx7OsXn!J4#TZxi5W)G9MFGJ|}6|y&)RKl8vvz771
z_``XUGpvnv4=u!>Kg^OZ(;;{h^LxqG{S9^4&7SDzlib-<S%9r=bvW2shbiRk74mjE
zb>y!{3o(RVDPQi{3DOrLmb~q)=1dFgps&0(yB)7U!c=Mk<0{a7Yz60Vn7{F@93__2
z5XjpG<Z2t%vD?Vo<>YN4y;)m27UN)LA(H7Y_Q+(m;c(_Z$mj>^tizqv1!zVs^`kK#
znz{LSKAqm3iTOymm5(y=)|0%ANX*A7>KH>0<l)6}^7JqJvjOZslDB?j??N)?qaYvK
zSl>o^<e|(d5BFqwxazCLIP%tPfd*qG8XTqPX(0QweWuepoWZ_rF+a{aE}8Z1(rMhS
z!8+7tw-VK3l~D86n!L3ir^Muo3Jj`LFwa>5Tcd)0L<Jn}ROlY5!op4}yy>RGI5PJ;
z*}Iv%-Ac|@y;CCMiV`m`D%o>XV)j8LGWh<-CrW0>vfgc|f^v)sugKf%*DBnstH$Jx
zYR=7R@!<pe(qnW;R50IzzU)c-owO`rts7Q|XJ_c?S95pKVtTE2aW<Ac;j{C#oRQWd
z{2y~T9;nc{k&5{!+>bd^i6!LzOy+w`CU5JLw@-TUjQU86OZ0C4kGF!y)Rdk0+|A8H
z-7RW-`<#n9k{raBZ^huyE$Dn^Grr_)Ld;L@^~v3e%9wSC<$1WO(`0Ojormg6WAONG
zB7&+CvA#zl>gYzXW*ACO)c{Ch1|a2be~dcV4{zgQ@$GvQDm;6jQ}=MBz3zbW{_U~h
zXKM_4<B#;I{#c~;!$Vg;bRciLnX|5)%Us$=UdXfZ#M={YsK|H2>_pZ8!JNZ5;R>fk
zu829qodFM=u=YALic^}P`WCNivm8*}h`CC<F1D+&Mn605lry(NVkz^923aDYlO;Ah
zW!4aRn@%56#U}Py&&mH!D-LBX`d?k<CD4at;bno*T`h2b9Op@<THr$u&T`&j1`>HY
zWjy(JQ^tKuoP#+krEis5WhLt<g%q~#QpA3gV8wG8{CF)NcTf)7YB_q+x3tHB--k8W
zjK^YpIY4%_7GvxjF@}_I9>*jGSl6yTF2=XRVtgnNV`HHhcgUN7Yzcn0lj1r3KnuEa
z{>H_E`7&~5n8~qcmK;CRWz5)?;$vObhGb>i-XeM*MOgkqh&>mD=vgL&y+(*rYlL{v
zT*RH5BD8cA;Z2kXY2@oCzQ63B5Swlb*-sK;YA+$~g$VJu89h=$A*@adP%uk?R^tS4
z4HIy0yE#UNnIp)`9NR6-ao^h<x`VutvySAgw9p*4UYWziUw{Qu1?W*Bz^Us3WHc2*
zAtn3Wgv{m?!a7L^-%25_9TKAcW)VLA67xJqy}PFf6>UWb881R8Ki=k|2t8$DbRcWX
zc8IW&AKPIsVGav@SM*Yit}DgFH}rU2A%ANmh&{qOzgmLx+a)NtEx}$fJzdkJIC(+}
zQLPl-<n60+&h6flu|F!q=_^t+IK%Typ%k)}?EOxskBfEkw=@Z!yb(ixpE`qp`oMA_
zIz6REKyS0#ZDwKJV(w)kUoV}RBRPgUnV1b0L*B}x9k7q}Z6tYXo8pKMEu3&W-iiBx
zoj5n`jHRud;W*0)O;<aj^)g4!m^)&^4+mU%;ee944%q+0omuD2kQm~HM}xdkMc#H;
z8;Y!;Fyu<ZVe_IR{S=*GzpM*V{JSE(q9<O3N1#>R-ndHMzHAnQTgLvlJR=UV2jVca
zQ9KTg9gn)^<I$EmZWk-kkaut~cTlXx8WA(ie`UfrItyo3WMO+Lv*EsGqu!`obQ;a;
zmsExR-Boa=7Pn)j8iH+V?7E;v=MZ{h$Xns}JjBpn^YkBC!fTfMg+jcqUxWie^x{sS
zH^`=#I|sQxJ*5P56HBr6Nhy}TD21Xqvo$W2;^gWwSQ{!}_q~EMp?Vz3(sP!I9;z#P
zd?kAqf1($t6Z^dE2M=k&9DeqR_dhhSS7<~edE0K05g+$)#y5wWa2jWRBREgX-tyKH
z1_bQYqraaX9}BsgTV09Vjg|k`6Fr_-2~i05uK8EMpEdTTqul!w!yS7MOVFu?p5vn>
zs2IwfIsHmt6Gfi=<9x`+B5Zz1KXC=~C^i<ugIr6vu0ss#q-VX!Ao_!vn{m#wrw$Xy
z+xg_J`f366<ZX<O8N<wtNaFsSW~Z5j_=dT|tfxv27h=VJ?y>oygFl&8m$hISJwiSQ
zs4b-CAVg7tmoXKbldAy8+nEC@(D76mw$<alR?eeDwc}jU8`i6Rxm#CXgeBCVKQCi_
z#(7V}R~<U6<7`(e)^Fc*xJ}+RSi_(DkUTYTCS@k;)gjFHXFc1QwcIcA<g88$yHR}K
zJ|EZjb0+}%k>!>=Z<*5%CFK6R_w@dfuP;6GusneLAa7?j$wPx1^dBG9pey^aE9+}e
z%QKlOP>mkznTzyRjR+6U^JJ;<ihb5!lhkO#+@zfU>Ard;^Xio-7^CETo)TH3lrX=n
z!0s{y8qhZ{AaC!Gw-@)2o9-&=S1L^J!g`gwea+XTT%tl!Jr&-+Q6k!;<c<m@%w{QJ
zn5{%kl@db!j3wl4@A@h{AFYBfd7JY>#a&}+?i^)a6aCO3x7F|(O+WN(4KDNCaN<S*
z*7PB3$=l=PttWZgE<=l()A>B~CU@mpoTpxXx1k2Ra#&ky__O1b7&lLeBy!z;6#L6d
zRVZ@dS%fq1rmw8|$=e$8HjFy*R2ypav+0}OtVSAhEVc-9aL8vX!sFMY9@fM6@M_ew
zT#kf-<@9hZ!J+n}psO9h+3?};^BI8+ABLjE`N3FMJec~<V66BZhgV($uq%21Lhtm4
zEj6ym-D8pYCkl^zdO*=79OKDbrw3tBwh6_jO~FVFY6Hj4t?_nOAd;s9;A3NYqXu|!
zzn>?4Hl7&s&<)BQH=H4FqguJ4%PCiETEu-m{hPw{&l%U(H9_lEP4N7JBhpzb-jXl_
z*Tx3eVTB3wJehs9#MBaI{n6jMs4aPWgE@g&%p3Y7$HF!2zmArpNXnU$t(+bCC&#Kr
z78vGkfw|;uC3*X0vITc2(=RKeC+f0{9$*<VsY#x(VdleLDRz^$CrepNk+)5#T`CO{
zEahxVJ0mljbaF(}m-8l*SxcA2*hTK8X@wX@u0{SO+h&W=Jx`1^)Q>lk!D$D@oFyS!
zG_32$+f8I~tJM-}S5n*~Z^P_mxJfVXpEgolWNkO=ngo-J$QbrR74;-YoXeUtT!i8u
zLT0=OG3$a5-SVg>lePA<g;<*<MBrH=HcIFnCVOuNim-=o*ZU+yGx9c#k1LwIoloAr
zau#9`dHa5&06$X&m^4g)=nw%^?^w@zn8Rpk4sY_-k-VMQ!5qVSnIj?D9P9JV@%^be
zG%W?#nkImcmfopd0<8Za;4Tg!jO6W?cJxk#36ar5h#}-{=|B<skol{rabM|5J$xtY
z@i-A5vj}&LB6McW{_(L0_xbkl`62}HOz|~>+$D=``JC0cDM9m#5;&8$t;t(ep#%vF
zBq$?;zdolInwqoiPARIHkF~iM_4m0lY`Q2zuf;OFyUbelG@09rKWnKFHC=?LcupPs
zz5rdH32^0;0P7O@T=H|x3_>iX4_noq8ploQ7vJSDlDEn`R_OkjyNT$#+qTITmz&!`
zn_!3i+|#r?h5IA^v%|rg+@1K?4j$I_h-gfmxjT13ac9N5&JH*~!U2}t8FjL&16q=|
z5%dBcoZ*S$8D7{j&Kr$qa9%fzdEi-Xu;hLl#I<XSNrT&>*W9+)^t2t$W`v<Sz5_B`
z!*RAtSGX+W&aB=MP}q&aw)jz)Ier|D4IK|Fha{|=JQdd_CS#2<8QZQbgdubpY({0l
zCT#=9NH%a+RW`a#%0<P^?HEknHfCM%uUdsuE!DX6N`ZGK1y=6h4AGPA_~EO--0KQt
z9Z<q=2Cr*z+zD8##TlOCe{e6`#6CK-e@~xhw-OBN%?yYO%*11E@9cZ*`R-yS?e|jT
zP9Sd^Q#YN#`4`r{&RJ#LnZTW-&ACIanmtPLHnms}J8J_L3^s6{p$w~Dmf=}L`ld&g
z!?3{!BY9gu-Y#v=Y+7F<Zji?on+$kyiaE&K_Yy$fZYFQ<FQZ2Ln?JLL`(O%7VL?6o
ziHy7So|Rz0eda=uw*vM*-G`PSzC{W4HD`{&TjoGL;@k{*yPUk$>F9SF#p|+B2Q$`&
zBj#vPN#1tmJm{IF8a$>?=puXHyDt_%MBbKb3xKi$v?wS*>{jMzPNWAZvH;4V{2rSM
z(2|)*&-nG~3TmVcbMafBgUQJ`NSj%K=FiG;brAEddUNL+vztxi?Ll!F4D>?nCvV%F
zV;{L2=U159pe1illDCfs@yt`pY=uQSr1E=D<qT>@6CLMi3g|88<JFP1f%*9PG7mYe
z=;z@)&$Zpmk|@yP+Az+}kO^VFTKsTf_Az-|Ox`|i#yvjt0GG|t-~nevkNV`{;R_7{
z9%*o3uZA<j^eU0JZOB_sp35%ytLY`C*O<Ir@226tay4#}x97>*^~2PN3{oP*UI|mM
z5=A`Mbxl^nHlEDl>+9%T1$K>9Fxy6fEq@gBK`YUWymf5HvmalJ@m?x);q8xDMc<$j
zJ=v=*CA-&Am%W^%L|gK9&?F@;%uyncyd9}fV53@vEj21+eNw@PKC1?-b3Hz&xI<2b
z8>6{9cB~q6`M5p|(n7zLbHlTkJ4xOyDyMEt-e!}x|J%9siFs|O$y;mA3x{n}K|$X3
z9Hhj|`F#D8x97Vn;rL8}Nu8DG&c}E}qs5k_JRHvF&fd$p2-&BGCu?4hi<~tc$G;~&
zp8B&hS)VUQ=t*XxZ(PP1faPd*W(itnWN@}`ES#>7VD3f&D&7r+{J)_Pwd{u-Yq|5i
zMKpT<>5CTg`y#zpBsvE4gzjBOcs&Wnj)ZWW@(IVacfiot_85I96b(E>Av+L^hU9J7
zEFW}e#ai~H2YU2jCe#~u960HY<#JCXeRjj~POiA>=nAhLE(qxAf);<dzvr<t^TM2v
zMc$VBIwJbI1BO@Ip#MD1bsXnjS9*-ckhf9aE%A)~*_zMXtGbW9*cO~$C2zyY+c#6?
znB>km)X!2_vNr80Vt>z^9v)lv1$k}OaK7d<YZ!5Ta<_>b-N@Tt&N2j$x6b74brU@~
z>=(}HEJa=Z`@i2Mcu*oi&_o$77-TS~=eGrUCOl33R3$`nwGa}85aKgJ9Q-Oo*C#@B
zA;XGd1n5ErX~&DPlQr%8ZR9UGEK{)t{wP44f^}tv6vv9B*fO74i1Yw`BX9fMk>CZ{
zQbgXKZz4fzng~nC+bNHQ_(L5z|ELgywL+|0CWK&`kh!EnTo@+AZK(*Wt_$JApFc+|
z!sd7UJhj~!74#uD7h-a-02iqj2QD+mdtY-jmz(3lcQe$zWQOuuGfZ%0kG8QnuFK4E
zvAH=S$=f01ZGZCi5B24`HDoaBTj4zVrO4Z7?BS~32+)qajrI_tdut(%wx<uPyO278
z5XHZQ%#jkpdxQ{S{e;~8z|Re0Pnk2j>16D(vt;Xa5u%oe@R>X=rv`nKeC%>kgc?3y
zno0@tA|>=FQUA@NCcRMtH`b>%mn6u%!C7G+`j*q_%Q7(Y>LvAMcNtp5%a9x-!?pbq
zobDq=bg2+=^M!blAcQ$NyOF$YP$NL_K>?=u32`|^h(qM<1U|2SbRxW{WG-|qdl54B
zf2oNaW-tGc9cO~2%)6rIAlSe<dI$Y&LoE^e-V)x;tT-1)FJ6cho+k3O$+zE)<F3LR
z+>1DidbF<tJvt7UY;eGIkrM)_-OOc%^Sd`LXe4&S#s=<KcEFuI0}qrx@qxFQFSJd4
zQRW-~twU?<>Dn4!`nINrvn^)FwnevtZBVr)7}*p0z`Q4Q=7CXoe`_e`5vE`tc{@at
zjG3nw@_fx)s*za;B5$uLH*iNPGtlEU!fDD@Xl!!n@ybQ;VCFGrGT%nUZ07IV(Ve_K
z^LRTFTPUEsq(EgKC4M|-4KjtZ9^KiOs+-T;>U<!IoViuNtaQ$kvQ~VxvIO5QlweC^
z?k4J8if`2=P+TlQ=B*NF-*A^*|1$i|EQ8mpGS1$YBjiyj_KhpW%&n!ULq6WQT#A}B
z=3qT8gNwWzl_ScrvsVTFOXZC4z)D0jF9QvY_>xP1@h0{??J6;~l-l*GGHjYthL{BA
z)!LWwT3QP072E|A&Fl$Z&c?px3@m4I4zWgU_ObxZc?GDSMLu>bz=MDSC|nCLfSy*N
zTLElY(~jw0fK=xB*E^7pKD;*94%FiDK@IliYTz<VgN@}HtSDvHgDD^KEd}_;uRliC
zt>L`Yw%OEeLpgU7&)WBYKck)w1rGU`@OmpI-`;}z3$|j<lpHLW#+vp?IRd+uV^v5Q
z+`cliw}RRCLCnz?P$Ottg4Jh=$;V>UlDEg2aQ2hDJ<BY{iR7(rmJYK&6`+`z!cDDp
z(2%#M@8{#$B5K2zv}iGn^{j6m&eB7Br<nCCbB=$WrpI_8S?i-kKQk?+ePrL2eOMuD
z)IH3jK&_g6e>D!$JG^s_8WX#+R%I<ao;7EW0ZKIAq{NzKD%|_4!dmjQ<tYWUg$e{O
zQ6PP|0`egWbZVu5d71)$`T3ES{Q8Fq_zYHXpQ8d+-U@W%87{Y{0&ka*ljLpE5G8I_
zDDY~d0zIe^cO}PMdArK}uY1{}mAzEreuWY*>MJolLWxb0N_6U{gg1E`Gfjc<)T%`a
z6;6}0OUSo>r&LIJr9$L;viGwJx4*OQbyLH9D6@*w)ZA4^y?GOJt$4=FDd%firGk<F
zejPRK-?a+d%U2*iLxHt(__!DGd{C``-5~`U9bqkgRDmUrd0yzG1XzE+cjcKhH5YOG
z8=77x6I<tI^0l0a(mNW|j8P+iKRcq?a+v%V;9TH*Y>u9bK(lFhsTzsQ_>qtv7!K{V
z5eQC6z@fK8v0%Yav|ApJg?%DW^r#zxuXI4mncOQdBot$6L!f9Fie|;_q0DTLl*CY!
z1<{Y(r#&jk+c(ys_%%Kl_P%YXRrw%-yUylT({mi@fpqfL>WDigJaFTTlpC(Rbj8Zc
zO%b`%g*$^?(6+iMo+LKKrbkWig<h!+Vtcq+asT;u8$|51W*(6h<ecxBOWrzs<UVuy
zjh>LVF66D6yxseQ^O@9or;xV|L+OR3MqJ_~<rz(i*5vIK8)~;dITz|EgU?VIQWKaT
zI7EhsKAb0UqF0K2z^&C%&TdeDByXp6r5@=j#ch)WIZ+a7Dq>iW4;4uwbSG~|ldqkX
zLYUSHv0^T3q&Y$~xhaGPdDpG208^s{I8z{m(;E@y`it>&kr?g%SNHP!^kaRMzMfuS
zJJwn&1l$)VK+qKGwDkYBSt~$)vMkMCNFR|9jphk4ks9r7e(zi4!cg+ocC!#?r;xqW
zZzsBtzm7t<7YUh1DTEDwzDdMd_pK1Q<nM0&+oV__Lc0mLhluQ4VvccM=D10&CVn-;
z=8I-9BX3LZm|?2X3<9A!d)(&eFx?CVv&~?0&kPrn%sHQKj!jq0@#2m-JYSl_%2I&o
z_XNo0|7GcK0X!R!tvp}MXCGHRTnH8G+2OPJv&niJ^7i}~_H_qPOCCbs7n%EbALo}?
z<0dQ=G252s7xMNY?<aKQc_dp52i|5cV@}6f38t)<z>d7lm?eSJ1)fQAcqSoxBU9*K
zp3L7@suYoKGUl1d@G?*aJ->hOZ9eyW{>_)Lt|xD+o(k~zu7KBN0scE6KvWCX<2FL>
z?-63<VxC7<b8d;}G~1(c1XCk1Ra57uFT4TIZ3}n~vyGI&m?gvh9sIg?G7SC4{OOKz
z)ITLh&i`ZStmC52zqhR-Q-lg4qL_dkh$!Z5cXxMpcO$maJpmY~V7IOvYscEy-MYrE
zuDQ?k{q6I|yhdsq82DW8`<(mS2d~>h=B7_o_SAY$_9nb!$8bbG_C<Ls<s;e`-sXU}
zdzKYaxq@VMrx(EkQCtp-pB%gV*vsdyI)_RsF|m}I*h|UpY-#4*%jl+4Ir3b}$^BS4
z?f$*8G8R_RvT0RS<#~wm)RZ=lYN6GCwU+<6HtIBBwpzG{YPTblU3`mFkF#}~_R%V}
zGg{?ZN9+5iSPe`{l!t$^8dpwM_6f-<)x)YG@HPY9wrXQnc6j@mYh*xAhvJzb%JVsu
z`e>Tw-egZH`YHAN8X4N8YfI0y+7^OF%t7ApGWKy_Ci~DCem2}duZ3MY@b*wvrgp#1
z)X0zIGTqG7Lhe6&WBxag{P;E-)C1lwU%NqlE^g4wm`vR}%;$LE8Je^~S%!_8ablx}
zgl<#sLu4^!Y*WJIZQ9*{EMXsX^Lg@wBiO@m8HV@9)8&z=W_Ie`F=S7a#7peW>?Jt}
zD_XAA_jh<tk8wswPS;HM<OOd_S4!8OqRghk+g9-QLvXrGb$Po$y2fVGzaC4j5qX6>
zkEbd-nx{=WG9BRUrERO^7qvzYnx?A*eC__=KW%w|9_i1|<T6cNt;1heY4dXCPP0-K
z+$UM<P9*5c>_mMSnXF6b{&x>I>3YB>`MYn_qU;+ve{awrc>BY!L4W7VWQRhA`nO-N
zKzioQ=?NbRTc=g<_6^?Sr<2zzV?#P~@67d(33bwYjefsJZL{ziWn92V%k$1ct9D^-
zr6oPbr}QcRfVGqQvtIz-wzz_~*t$}5`sLwA_F~SJn(!h`V-nMp4JMsO6S|B@m3<0Y
zvLiaRBwUP1(P=Ymn#Qc`gA^_M<j`nuhg!j<q42hU7l$qeI8<xBT^EPg@i5p`a}TxS
zu5c^JLEp`xT>cJuz+8JCSWOno8E#cw9D2{^cy&wBgT5&`gpO^tI7M|Hr|9Xk6rJ0c
zqVVh~Y6EZU!rKxr9V*}3p{MBNft?)M^wh54Yj(ANZr3uN!{1)iu)R}MCOAcwDk-YL
z?an>B4wCs|JYv_xV|Fz^YghagyCQDb6?ogOg|F<&DD0r`>0rL!p$70aKGdNroE_?q
z;w&(X@A0TZKHn0RVop@8voV@IEJg!6#3<xpB$+IcXx%7Hn-!%~_ahY>6R9qz!ZjIf
zes|3xO}ai)y*tj(sU>Z7>{u;@)vTq_wwgKvPajpNF8kdOy_-{2^_qrg-_sguyuAiF
z{54dDkEeF6p|<eWXKHoLDqmgZ2GvxiOBFRy5IeXkt9FGza`G!`Nv9HuoLxdI?w3%B
z=K(6Q-Cv;%{j~$$J~-m1eXf3L+njx9FN#SK?4rF#|Lb`nbukoB)uZ|4)t$XLrM(s0
zG@p7F@>0BqmzF;#ui!vl6-pqh?<E<$-#zq*nVP-u_U=A(NO}5h_>Ic%A&(K>W>u#j
z*CDt5px3vWYj)jU?y8mFUF!<FYu`QI9)sU_JhQA}_+f3ibc331JlZ!FZF?t{oI<YS
zFPkuPlAr!yqPx;+Fz@2POABvvw{ul~bmMqe7u^|Xmj6SO#y&QwZ+0`;M`lekTI5yI
zB6~NBuA#3Uzj7rzoxCUJcP?gG)IQUqgvAzREw^Y5ebGhe%L}mXU{SL^`dQF$+~92o
zUwB@JdhrZqZg#q;I=nq|!9{`a_E0Rf<q<F!hJE1kDtp6Pe;4&FMs2sUi*E9DS<PV`
zk5_&VeOXvJ=%7j6QcY@PGpR}vTy-}yYiiQ3Op}hrm^6#$;jz`kK3tOuFE%N2yh%@|
zn)LFSNpFUk)fvxn#f#i-;a`42Z`a*|ch;g4)RaTvZBuw_t>U7AFt<fR7>2HPpX|bZ
zDxMS1@#jG5*xW)mmqfzDDoOO2qsZwP?W*Q=sVj4PBUfE$;HI}RZhDdErql6mdc49-
z!=l_Yom#x}BzJ9?<*v89CZA#N<yyJ)C;fAeH~+n^HSs?maCKMWcg`e_VC`BL<q5&l
z{f@KEBVL!A+{4=<2dFD=vZ!Ssug_o?)edt}&El?_V{_FF2j_r(ZZciLSGJ0?L4SG$
z9o?A2rVmf8zO4_rAABDp;CR0^^ba^6^j+)5nb1vnf4FIY7dxBaZTQlB+IE+0u`D01
zh2&@)^U}IJWQGndqN%P$Rj465ii-K^?nXb|uIjH7uH<N>me5X>)SUUHW$-DZ2H(r7
z*}`(l0dJceuB_d6gH_AK?p=7>esXm+-BD9d?{(Cj8*S8P=xnWY300e%^U0|SQ_0R@
z+M<QZ-8o7-;BJZQ(ONt$Ms;h)X=hrTu13boca2qpI@?rkzfF}}*){W_T@$J}6xW8D
zA7}jLXqsHr)1)!072#km6W$Jox2y3}T`0F+E8uNq^sVU%wcDI%w7TpVT}%G#0CvWP
zl2c1f_u$=3`Ujbs1#fS`+cm}w8qMdW#&1vuI6HoPriMn6k#vYmy<oi2W9SE;Cqw?*
zR^}YG>3#EUTG)pk@6>H%i*8bf`J42m*JhR7!w$OJYt?Zx8T>!e$^IZ)51;ANlFUgK
zVBgFQ{H5?_&9GHkyERoM=~u72n#vh8Rr8jkYwqGF@=8|(-0cW&+g68RKIz&BZ|B!o
zrI0r4!-2Q2?erACrzjE6PGx$Nw*3y(<XSzGpPM7lsyt)`^<n1YgnN=~#}l>ZP9k*&
zG*elAc0H3dU`K)qhb55znWUvZH|fK@P0GEFT|_&WyV}L~J(E1E!elhlQ|@ETRN7c_
z7uv1Y(46?9_mas!f1S#a;V_Dxt?4)F$4Q(QvZX7qDPCLZ#P{HB`Q`Xzr>s=z>Z|nV
z_$sYO+rC)MKKNOy6o*d#(3U=Bsa5iVw>@oSLe*TUM)-<zzf7YJPZovJbOoR9ZC7fc
z=Th_!z0-VruZ`esC3qWYOjXmHFzjFI9>>vplkIBO&#pGz?0NxXpN(Sf%xYK1?sj?7
zllV{v4a>)>`r7fGJ5<`}P)BzMI>n(v-|ZUx&8|9}@hW&ZG`~fP-oo2!@OBV2+}05(
z>br;jX(pQXyhF9&U=MyDeY-p4$vLn?F9+G_u<)*31vBjGI|F`Cv}@EjyLOJ`+(>5P
zZvNeh+vtsUt^Ccde-GRB<0yW?GxSs7?KpT_>b6}TJl8RBKfM`egZ>UxfVb`8ZEbiv
zd8|X%hoafhxxZ1LXbo@I!`orwVss@TTJK`Rwex$VHmqEr+QIWR;QAcJ&6%nO|4vcO
z-BZYto5&7_5o-8!D7)g@=*P}ly5L?*-{#fSut_x}fBt?4hA1qgs`72Bs;cbq>2#!u
zI+U)W$n2H1DO+Xv{2QeA4}w(iI=4MR8X6iTd(9w4WLDCpYXJ%g4^Y*a0s7aI+=}-8
zn&j`VLLUC=p5>=n)RNcl@l#H-pBi}-(cJs&=Q&lF-`&ER(Te?dwS3i<eX{k7vl}+R
zTTx^1Bl~)(|Bk$J)F7u~p{HK@c`6uh)R=GN_UH7}gSQ@PwgX)e%s%taWE~{uVK)gr
zV>E2XD!H`vBL3cmc%7IxF?yj>t22v|#>~n_cZHUPsdI9v54?Rv&9KOO<~-qT5WKC6
zW|~Fj!UIq0y!39b?Q~Nr*Z(KKTvexotF9kI7jeC|-!>_5gGp}ZO?v&&q*Bymv*om~
zOV6S&)RRB%wW!B=-o9gz=K~A-eJqNBr|0>2iCz{BZ(-KZl_v6UO=LBgR6EwB5@Be~
z=_Y+SY|x{+Xk~cY5#DBA|4-kBtai~mn6?Ry3@Z&+^SkIDPuN@BMF-HjJp=Gl!I25|
zT*yPQD9uh?n8zNq#-x_;_Ft<>L5U`9apV1;ObT6MVrIvv202X{Sl^`nyG$Cs$fVZe
zO!|Swwf}A+=gdsTfLRsR)60dqtFz21@WZT7H2SLX7ImT*+>$!;LoaH}Jh$igpC2*P
zy8ul&6Bhq^ZPCUaoD1O8u3;`Z+sH-7VC`FO3rA5yF3s&J-^XVcZLCZ07T&f@a8v6z
z>hH_l<g<{R4*K7-M!Hk4g{jn;@4()*6>};4C$I4{cUd_{RfV@!c<Tvo&)suj#@9s`
zURmV&#G+Yf@ohJlABMNRcUjbobI+GD)VQfDzqm<#{RMuoH)iTlW|&27Vjt%SzSh>!
zMZswJrjIT9`Ng7hxm<LT`hAOq+{4>Cckzl*?{K!dD$hMPvf=47&&sPU>^>ZHDX+fW
z&Z`$LUb?^4ONDc@x44`)y%s#|@OIM%{0p~=>h@~(Id&*0ligR<p84u*Y=Ej9FRy%;
zDricnK#h#8qy`@<Yrx`Qxh7Rnnc^WDRI8?{vdc2;Wk)?5GE3i+<|(^#D6;_b^-rlV
zHG;Q!t1r~7PEpF<H(F8o<F#)?ybMVRs<k~qd1fRjD$%O+);6+G$xjcmD~)TIo$KMm
z77jf^qYS$3(9w_-`WW=d=+#wSx<=OqrR(<v>_wqw`v<%Yg10BHz=<3ibg1kG6}D!|
z){lJuxtaR;AQLT`sX6JH%JqrQg}1q`!`E%-xK-@4nU<--@b(70J#_@_3vaVVkwLMH
zeJz*SeS?o^gy%*zMZ@Mp7mw>nt+?rWnLgs3j%Pnh=2~4k$sX71<k!Mkcl4bXe7!b=
zJ#mBBTUce4w)I`59Pswv+U)b5V^z-!Ry8%-IbSB|`9o$RA0_H4nekH&C$cvqN!O;@
zl@zv8vs&R5Lic%7m%Q+vUgDM%)rLWl^bWIX*i{eS9$1pBn4n}W!Xs1tauRvaiOg`s
zYx384ji1Ql8xr;Rss!Z?P0*M5@nrKbcT$n81Ydq$=-Gs=nOZ%Op9#Eub0I@DOS4<|
z4B3#(Aw_IY*TAo<)q(k(=6GY9JzK2{^o+O0rt5FK$7Vc2cXOp_eNdX5n0amXC{0i3
z&(>sz&$>0M)MOg{Nqo1{I;~P)c-y@6Dz#=0KyG+DksRZlyVF$ed76q$N>eXpW7l!6
zctSqIthFg*$^6IKrzKL@!;+%OOC2f|=+MbjyZq|fwWp<BfA+C!NmsiTH$r1Jvui7?
zJ58;1Ybl%j`JHm9Y)8j%e&cf{F0yMH`nQeEt`g~XUE%HGo9sG&&aR4o+f-<yO?Ua3
zPT+mn{NP>_2fMl*`f=K!Z6h7J(b}O4?HuY>4-IN@XvH17QdZk#UCcc*=EvY|J@~qg
z=TIEp&gI`H$WDAqzC?+iNt%_Mq~j}+bn-!>E@dUkt5%YJR!q|GrxKN%K+S(pqMr9p
zR1Ehk`z5N`UkU0wHeQSKCa6r+1ZAT}kux|!B__q|#gbUPycn(fv!l@fOZ6mWfxZ+7
zRoZ~r>Ki(h{__;=fwxz!6ZO2o2*s5eCR^S%n!3G~=9_EjZI7B7JF$j(xK-D<F(I1%
zqKd|L<L|m~uqGD@*3myJlaCvu#qjp<k05=&6QtMIf;6g8CG|H})bz{cR5!IWHM#(m
zgtr^}1ZX_GP4x8F`+a^ghmk?sr<jhV7gd9~MU}X?s79HJC>&jz2yffmDy)FCLhKmx
z)!Q2d^>boA{EA*^883Bn@=~=Uc@;V#uNH@SYI`MSY9He<M%UhQ^3>1w9y&D6LrL&<
z*tI-54{vSwQ-9pU+e<%m@9)fn!rOfCHshU}`d!8+e9et)W_N9A;;!?{sa1ZWPxv#J
z2E9hhGK15UxszDtMux%L&eWn;SHs&xEx05Zi4DA{DbID)J9@TBV=XFs%cMSOCXGI4
zQY5_f_BN~QNVD2!n6+UC^E7+ST71B)U1ykGy@2=VidkR&<==%Z{H|IwIJZgT`1)2E
zCT)bf-6KpY2yaiqTgQHbI;}J?r({y9t64{?m^p)))oC93*qxd!tawnsMRsp`r#bK)
z=jOW4tvWr*h%~dtgqzv(Zc@d%%+$i$Lh$xgiis>w9<Qs3-(iy~v@q$<UM8hHH)>W6
zlcxEbbSu!LnsrS2+S{aw@V5AI>dN$cqj{WyaJSeBvl5ueUHHYU;&_$2_r<3?l9?iY
z9%J&6nc>P=g7d^SczY1O<(p{HCwOaWMsK@5b@2u+n%{_8?f|qQTK65g{Z>g=4SzsS
z`Y(J}^<ZlxI~CyVw^;mT%eWu!M)rW4<~Lx@cMyH;rSAF!d%p#dQ{~~V`0LDBjdD}h
zudehO(Wzc|xp%l|YgL#CYd74oXdSiW_17%Ad&;5*+n70?PkrGf8lM_@b7qeRqT8?Y
zc6E4rJ_n56W6@ZCmYWSO8oj{6?-YGsczf3cy$hpPO`|Wp4h???UpBq-CGd9BZG7sF
z-E{0PH?>(oek47R5AgN^wdb&od6>D#qx;GD`UZNkcY)dQFJAi8s{s293g|=Yg8J=5
zVNH8iOc&1uXhN|PN-tDW(*gqZE4dQ2upq4o2&OhtMV9wfWn9|~FZyWhO`M@=y=N(Z
z+C1Gq8!GdF`7)w!2j5<#v7?r$B)qLNAWDl1Mr%$&3_k2Ql^hnYA9WMSoJvx+zmt`=
zgY5LScFkYG_45-ox4aH!#2xGtrEZ8<vtiCum10h!7jqKxvcoR2i=0c*qs&ZKa#p$q
z|GAb~Eap-xXRxm)LvMO#Ds&G0+jV69N05n>mZ`q{yY`k$)jXA{6L6*hy!Dz)E!u;g
zr8_;};^f&j$W(F~KH(kfWOrSw$vM~R;@5ON!v8z*8X0?J5f(d?j^`s?Q{n9_yjxA?
zrt3w))jHLArA9Sfsec=;)Qa>pl`i?;4yDVsRQ2p?)1|UW+BP^w>l-G>#gU-DFC|D_
z6I8l+f|l({)X_tU>fR+ur#jhHWcEr8tFuzws0+U=h$okx;tu+We@#r$<wWx4L+!fV
z(WZOwb_u+_2ybKlNYZ&8qYAwBg}1Ro(X{syG#uWZnZth53Gv#IwTk@-tC^)=Czr45
zRA~iufKi!Rx-3IZGw~$*WT@-Zbe*n_|BCwe-EXT^W)X~LE@uy3pVj@+6}}{0?<&(z
z{hF$v+-aHyZ|$Aa^y_Y#Ho@EBc%(w{;GUSWN|7B`DZC&vGWFS+hwsfTg1Il|c?!eZ
zD&#9>zf6|XiZs=Mw{h^ce0%Z~;cZ@c8wP8S2jE+xhx&A}Lo+MUe@(ONH@Mgj-bSKd
z=Z4sI&CjluCG9$Ffw!-%x(07Iz}r77P``DvE4qtax!~)Vb#^_@XV>vUb`9&v?-*})
zxMNeJKWu86VN=LOn|8p^AL!wWjrrX}W8Qsa)64sIm4UZ?;cfZa)P0>DYI_S!4Y&VT
z0Q2GPK6qOd#^&u0+da^|*R87DEJ=HUnSbrWeXk_Vt3!Swye)7&QCpeUS+gop%c=9v
z?wzRV9TV9Bo2aI+wri0@)u88oV|Id0mX6o_n0Te`Nswpx1O;cs>FcUECHxbu$Fri<
zIR8?;$eN^swI<4E#{?~yG)lGLt>xilE$BHyA(cny^xR=`Dmp|yms%?a*-zv8)Z{Ey
zLsgGe)72lr%9R`}j~>A?!`daef^}<hkn+E+B&PwDbiHyV9bF%&zW6yxWeZgO{pD2S
zYk=JD2I#;6au`<xXl{=H^@q3Lv-_)dbwAZ;TU;;t7E@z*JIK4Ju74?_bH*Z?bhEIg
z!`sW73TweOA0?%c<&&0Qr-tT}*CQ{znTWP^_EIps&D)3h+4-KzAMB}R_dQhZqlXH>
zTlWthvX1sp1Lk4fBdH;e!;6fs<+mld^mQ~@NANbSk30Q*`hXj$=U#AA@)<Wdop4jt
zQ*Pwexyy2$tivF3@tGIdpN=NxTHm9st9G|=QK#Wtv)ki0o=D%do11FD+j|q4!*OEn
zWi=k(S|;u2i?*F?((T(OZKe0QAq+3C&8#ik%{mQlm&4m_@HQIW{ta)h!`t<RxNdhO
zOA+3_M|1XH`yX#l&Vjv!jVkoTpryMFnw4hIZFrjqZ$p}xRjr{#11?%L4UIJd-o~K0
z4&ej3^~<6c=;^!kX#eLm&J81rt~cq*M4sDFlbS3rY3C~XvksHaB%^7IniNsjq)Xl=
z-G5_L@#{vFg12q`OtO_WDI2`?tY^}-VJ5AlzPtn7z8l`&hPS@c&H8lQtUmaao6^HA
z^N9SZKg}xto;)gMbOYh-q5#YPWo66_!50RT`=P-PH^eJVo!cMYE|-f6!`oDN>vxkH
zxzm6Cvl6fHTBo8nkJ5u~;wsOP>_u4Ws^I6YssL|mw`0z#g_|1k_EVTvwIV*YJj@T>
zppFY~-%{_7gSS1<n|{>tH&=0?W<*aM-hQ})R)x3Q;jPnVaw<caA4b!T;WhbmiFqh^
z`xkHLfwvXW?#p*uRHz)eAg$2C@OH>UY80O=>TAN2#yKZ-620-Y_`r@)S1#r%fTR4k
zcn<LPJ7>FTu=mXmW|!KN3-%?qmbNE58h=5~B<9F^c<K$gE#EEqlmTzovS%u1n}Yfy
ztEi@Z#!rQBe`Q#4?Qs;>(BF!yZ&E33I1_}Qsj{X_uBP1I>&eZzqeiS8rjT5t^`zxA
zJ-Ikd(;m-N-rn={saTk1PG6)sLl*18&c&K|Vu@b=h$1^SMlY_%sP&;Zg}~eDQ}Hb~
zvf)R!s}Q`MHq5SCkL~)4Yg#=|dTt{eI<f$*!Tf~<-sXq5KYpfY@zAw8RA8-KQ^?^x
zOMU_#*xEhmA@*CtE+uA02V|=Kbn<>3nd<W?LxIonGsD|r@U}H|;3?d<g}3|Ac1AuP
zm5cr_ybY_Bsq7JCCT(9Q6TFSWchnNzzQphJ_H4RZqGi_|OxKZ}>1y10E&ej{74XNL
zXD*~vbSkx>R1NjV*MqjX<7U%q^iAB*B!yN^(m{ATbxe%5b&S!Wt}z<p5v%&!W7KYA
zjCw7OVcsH^neI5fZI`SzuprwKhyJ{Xf2K6wvjdHfZz$U&`kL^zA-r7<Zx1FVtC?St
zJaWYIdljd0_2bFsjM2%{F{)N4RsqLj71=96cPpcj$k{C8i{FPDeT}^2_T%^6Ppy7o
zGkiU*xRD!Nq9^-fT~{fHIi6p=;Uu-)Y<2M>QQs{xj@cMr&JOg27YC*)gu26>6U^1%
z7b?>Z*49qb*`8?{OIC50Z)s|}fJ`VfZ83OTxGw%E&Lh@vGNNMA)Q~#zpgHtx%cW_u
z4;r~5`A)f0^*f&QMQP04z*`Sz@+Iu_7SNlebI{WdN828;F^f#?7^XGY$L~#9>ay^2
z#zU)~I;;wLY2|m%rWtK*vK+MOyOG-REt?iSu(3zYraez=TIb6ijULo=*V@%1)W%*I
zwD3BcwqD?|Cvfw$k=JXL_bsbjn%LB{3jNp84)XkLnsU#kBqxVzWZ6}kub+>e?U5&0
zKX)fF-<qT=G2}RnNm2*u`^IWX8Uk;(6+z49PSV@UiE7z^+C%*Wc7G=*gWJy=3G%Ix
zsE{WKN(zrx-vaT9{vM}F|HQGEfY-7xdbT2eFNN^Dl}*s<yYX7nJX%j{M=O0swAy+v
zl{!t7Tk(ks+CD)$PmI<1N+Z=V_elAFgtxgzXu$Aca`PCX+<6D7XU9$|pVdM$b2L@h
zt?J5sFIe4@f_1E3u-1=8&yEbz<R(Gt^`;U#@G7Z%l}Z|(vyy@f1#14Uiu!GTdCjO&
zUWRHV=!pfW+_C_@xa23_7<6rZcK9%3o3nLsc@HY4GciRKPv2@mauJ<36w&8Hh1K(O
zA^A7+RqO^I`K|Mjr?-!adw8ok`gR_?9f}9J>tavMD&(n(>Fj(#8)RMb;9Bp&OpJ%j
zOFfh@kREQJhw3iOquz1k9#)|?d_9+{(wnW*k$IlB)F3VFy+JRH{tb`t9yhJYV!mbz
znG8RfPdV>S78twa`jMNE$ey{b|M51mxr=n%qN20$^?oy}$8Tnxm}F*d5N?H<ba%T^
zab}b1_b|zMELw1<Nr&^IVeuoKgSU<8&H7hHtHQ9paBJ{FlX5LFY5qU3IHyJ5;a!Hi
zNq!HF8okV@s#A^l(2e+pjPlB7)WqKmIu%Pzd4NGFj}00~t-1Snvtn=KD;{c5U3gm|
z9o<+Kz4p_h-<nz!0w3P~G%IN%I(C*xY1D)#452>@bNkv&WG~YzUSU$5<tDAVYE*=m
zNqb%!<$u8_dv<!aP9_zB!OOW-=lOl*>q_i5sVV1(H9q*PV1BbH%;z2<|Kg!p@rUtT
z_2zrVQ<eP@&o98NdtJzWT#WvO$)hXasTyukuV`xMUoGli(nYUp(Q~f-zqw#I$l0OE
zb+q|q7bR|_C;gmxSci*#rygF#f;NShpWC@GTZ$I#@1~c$Ze^~xDE}o_je)Ue&oj?V
zo&VH8H1Hl*dAOi&-?^yuK5EG;`T10Ekpo>jF3X}R%`6)I)y%9LxmlacGHf-=nV+RM
zy!`{-_J+5AJU46NUz{y?y?b+Z2&Mmh1KwVv&in)I+ZauMa|)V0*`f+)<nRQZFK2`L
z@b<tEW~4bIw8(H(%uZK5%W~5>a<;B@VP^PqZl$N>)|+#A6#Qo%efmxgcM&`XA3dr6
z<RjzGN2jOyY6iUhdZVx=d@Z6WWs2(2jH24OuDJaAm(uARrBq^LkS>)bi~MOrxwh+|
zPd$ckJ{Y2I8ma;_hH6LJP%ZVGtS3jQZGD)H{+NR&b*`Sw3Z)+!rDr>1v~+2#`jm>(
z0sjO|3P?hGTK_NKwO9wc3U8&?b<j@E0M|oThuTzkC~=TOi;to=@P?+{Vn*UC%&1Gg
zPhb2)%tib>h&SitY6W&=-W1*j#FB{wZ)4%@_)unD&$C1HR))^k$k43JObOn0fVW}r
zcEbd6_+PMhXG4ZA4j@OVc!v7=u2bb2tCe{yO)0S}^*31@!+NBtURyl0WE5R<OH;v+
zG<;uZG<Z7zuS>JvlT_=sB*ksuT01jIRkkH+!+<z7>KLbgeBzk9kCn&U81|+_D`iLw
zd*)-<Z$Q4!r)Yf&i_rp?7;QNftLYu$RHrq1zh$y&(0jFwNS4z!D>;~UeT%Y@^%kvt
z`=j;u!Dv0<{@coEy~rCa^@t#gJx2NLF}hqcM&1`<)aXTwUT4K<?$H?eX2vL}ek?QQ
z<RksSiwkcZu&yP%-CiS28=9nPX&;z-AXRJVefFn*J9Qa89hl#@mO~@<^BUk`>Nf>{
zEPd$q#Zz_4k4%UroCWyYdsbebJZXxjt~@t7O@3#{Tnb&uZgBW|kGWF3QDto87)Ow)
z2yd_Ak(vUdU(hG`*&|im%G0NYx4Yr3?@ByMGg1`u4v%h{Ly_?IeRjOM@b=ywo61(B
zM+<L9{ItsLwiUmyRYm$*H6_BTp>X#GyuDn>rZ<CZ%*5LC>5NSy;O)<cHVGYD@&NCr
z#=PLTRhutaHHULc)uvYMDQ8tHAAT3JTQz~58=qQR^}Yf8-GpA=ZPS?BHno7aQ{ZjQ
zn|5}i*!6|F!r;f$X!dfJvM1>xyj?&~pm_TvH4aSDoE?dJ>qN~VCP4+@Z4&1p=fHUN
zI~J$bPvbQ4RUGxWcujz_V`|0gcNjYp-qt=FryIt2#TSYv7e7Wz8b-@wXQcezF4xqU
zrCPFRsm9M(AeXKSw66U`{9_aKV$B5IOBf-KF9Vf&Wq@{!?4?4<J=tm1L!I;X<oe!P
zBk#7<l^ZSe{&N#$hqvFef^{e%SPi~a*7A{+H4xsmsTU+Smmsw@SJd4u<@G9~oVpAv
zr-ReWNOekT>>+k?<MrshHvrFa0QF@*4UA;I-`h{YtBT91WpSPBRZPd@*#Dlhs9q)(
zQ9pRw8{Qs1RY>Ee`>J!Kk0RUp$a_tGg^?qb5y$S*E?)ASn^(OCcxr&RC;cf8b>Q~+
z1lfmAJhT<w?q1-b3h*`${^azZWDLdU*0mbBHR>u^MD%9EnlsbG3~atM@)RS;`5oz|
zlZD;bt%XMi-af<2G@WaG-Tf|#;<`WY7T5MpZuDsX>-`16+rsen2G{8?qs_Wdz^sR8
z#=JL-@>p-w-{*{)o!_LaCPu~2F{t@S1A7__I^W+QpF#!&1sn9fk3k+Y3@VmsAjjLN
z0kzE33ym6&u6*ud(jj=ea=B60CK}0Oq^?`XsC@8tCcKS^GN^O|gA9QNGKmb#&KfkP
zpF!U;jC%Hhe(NzaJEqZIcgS6wNso0eOpP+><2?8`i5VYQo7fhe%Kh)uhkv!CSBoY;
z&GWAfZ!^{!wS0$>omWQPO);ujHj~<+bE|opG&$I$1$|6vzR#qF27FzfX0kcV+C15;
z&j-zNy=PXf&1OA&Pwo4PNu3^=6z+wFZ_b$j-u}*W>sJu3b$5%NuApy9zx3E4i~NsK
z-~EHWF1#%dZ#TWQ=z3XvRP;h;&0&5C&)2-I%<G&)Px86{96+=4by?>uYSEjlj#JED
zonS_n-awgi<Y~a$X-DawF2V<HhQn|9di4LHf@n`3>+O1r_O!96B)rYR^Z9YktZ;by
zWD9v&H_V!M-pouhvruQvGE$2;!L2YJvzjn>>`8h7&!|nn+jZ#q&mUp*WDEXii<02&
zn-~|xA9T?*A6Knv<f`J-#y89$@9QSBO&8JMAKb{wpdUf)yDk3syG?VmYbUoJUcl?z
znca_r$?c|gQ}&#<9#t=-^V16HY)WCxTv9||?M3u(T2cL#&U||HQaW;`r0#48(h+!D
z3f{hG+d<WO4q+GY5dGmfSOKAf6(2KLb#@F>jr`+Or0!%LyF63w<>#nH>`Zx`UZe|q
zqO@XCtU@-%s?mj5-5eIDxxVqrini%Xd%JFJMmHRyCwAVh^tX0RFYk~EuS#urI||<B
zfVVfl-~)xX#$GG+d<dCI>sBjr%W7s~SF2d{HR=$QuD@*LxYK7`R5U|d>Sn63H~re`
z8H&Em4)z=CRju=SEi8sVXitVJp?!~U$&l~R3=J!h!LB;?wc{bKLv6S>>^Lxj+@S94
zmVvj+;q4%NJ>!}Ca3Z6($;ecluv_(`F@3p4>{uxkucGhc*z1fZ4BkG0pCg0fRKItu
zT;c7{Wzo8dckIEIC?!WkDgT5hEe+&aYKl=@kyw3h7^lW-Vl_4{R(XGq)pl<4_QtB-
z_*gX#jZv9<(F%vNAAgJ174GlB+Z<les@o+(a|T8zZ{Zj<{Wn@W;IjDKyO;R9g>ZR2
z*_!CxJ^A<<zDgmZ8BdQZK3PYS>KYs>+1sjqHLZFAuRK%iQch-R@CO}aex_;}>I>T(
z%yhxQ+sv?DV7>+qZ#h$n>YpSpY?4FM;~ffolB&W1{Cwu7F&~qr2Ysj=$F0=m`RqD(
zU#aU>v@{wzR|&FBsF9~PPE)PWRJDS&EiI{9#W^ZzZ;JB6+y9*{`V_FqbB;}C^3dD=
zY~^p&N{%bqHQ25Vd8iZruyS6tsu#SSkI#KqH}VVbTlFxnP3`<`THD8_VLQnSK55g&
z&FI@qo64u#bp5MUrQz(DomOUzt@_x8|D%mn0Y$9Z3U6~eSvBHmvO3kWlA&eOdV0H7
z{@<ZjZHg~L4SyEB4=1~9KFQkYnXE%!lGI{Vk{TsY&+nV0Y-rm^db9c9?d{TuWGyAC
zJ+J@AJ<;mEDq7il^19WI($~sSdi*Ow{mMtGP=!b}3W!wO0+H<a30E)QC7Mujp*A}&
zQ1f?T>USkfni;0qbHh}BM5x^RhsySN9ywDp)b`i}<v%n|>2YIqF=d2y&mO9K?wo_0
z4%U!mJ$2)F4+X91s(QCtldI8E4?8qf>|afEKdZW$!`p=A!SdW%nH;YmjaybpMG98b
z;o}uFetUU!sZd^FvE|flSy?&Y?FW_ACU`r|sf0>b2++cp{wfG>heY^k=fBMT%q*_7
z`o$FzRZJm=ifR+QoxhBI@bGrQK{97g6jHXDzVck|qpLnXGQ!&uHS#GL-d^lX&cL|5
ziqDf*bKkLJJ<dbx<2^JQ=I+Sjp;o?TwfGbLhW=`TPCPZusGqgahQ3B!|8CHseMW7?
zJ8MBx)$@1J>`88#GRRFmi!<8^Z|&<{6~tWY!b4mye{<0acsmo`4uQ8V(NrhUj5)uU
z_&9!{ETi0_jXF8bsGB8>YFE>!hv`NQJ7uINU{o7;TX~#8e(?5pczYb)zV<cfY-NL9
zz}r0VHglCh%j+AJ7;jX}aicz<EAN<0N`AuE#uznnf>DRyZJriJr8^r{0p5m%8??Tz
zL6s^R^cdcj$ZpVD4}(6z`jab78hI3U_)_=0W>$?^^djNy3wV1R-g<E^xHgzsANsV*
zVCWBcTdFDlZfH`M+MF3K!PnJBa*K@Wa+L33h*96$at2szRQAoB3ygTG{xPa93@%id
zx-&X`Ks}SX!rL0@uy!?imp*m+JCoe$Uk|>6-wWOjg2@+_aD%t6eDH*IqvpMsI(P>C
zT)0^bh8~Br)lM)IbphS^5^au_{n42EY)=<e<LCYiFWQH})D>FM8%DSL!S=oI_CQw`
zJ?cpPwkdfT`Q5bnEI#IzoS}}hH*g{ETS)JC32cAP?9(NSF6^>sa&HTNU%dY@=Ky-@
zRp>=~ccu>wZ)?!IJ$Qn%3ePo=`wRQ&zoUES?J{cx@9O|>Z=7cC7#3&3Wm6XQXS`t_
z67Yx7H}3&&JHgvap00{4OU4Mb@k+5|TB5;s(z|&3%1xhJk())nw9jmOUFq&Bj|Z>h
zVfH7*=2o|XdDIQwUOnrr7x30Rxe(c8>~0=jSoX}qD$~B0OlylPKCqO^Eh(WldrImT
zecS>=YAM9lQWNX9SH{e4T6?R9W_$Kl@`2uJ^Rut&ydA7EZ70e7><rzv%+#V8Gc{xV
zOy!EjuUvYC-cJbEr&-}TxHm$tPeke3?O0hiC+Y^geYMXn-*fcV;H@tnnI+6x)P3O4
zS-hhwvQqTyQ;OcyPu2JNsmzwpN5qRXbQe15e5&$(NK<$>{5po!s^5#;o`BUFK4ZO7
z-ow#q8R}VJy=pg1*Njf->i==AHnvzx-#cA3U#wBC=rsyizEUshtyB~}#e18{ScpWw
zQX_qiZY%mfeM{|h!OK+Deo79(8ET9f$?AXxEt)G{TS`-}?GGc-k<$vqDwjD{pN_@o
z$if)aZWE)4zA^e|LNr-R(Q06i(j0i36W)HW7|r=NTD1dW)Dcd8#S3#5FL$1!(Hguq
zTC>-nXSr5CNsm@eYKY;b$?U2GV~a=Xl_gT0mPe?`ln8Y{8qOR_gm&hM)cO38WU@ub
zZ&IYv??&n!*ZH}ZVY4U9#OJ%YB{>P`vbAWkP6Lzm@d^7At64Q;oK>Z=*p0fvrgz?U
zO&w}iD*Cd<T8DxY@d@FX{b#g8{=eas#>e!(tfo}-?{RvbaidaroD{XXou+U3SE@NP
z!|mSTp~C0(7OtAmv)$lrDR^5gA6}{WRIMCL9Uk7!Wv;XtyzN}gq1V*X17O-=i%lQs
zH4HSPQU9{4J$vMK!Ov+hGL+}o?>n_)SX*g;Ro&t3z^+#KVO7q2HhIC@o7HT3zMWb!
zXSo!3dt`)79@L5_ezht)j}b;)Wg#`=&#i45U%{%m_mZ{sI(s#KPu77G$*NV<s+3Mv
zWp%Nt0gqpDxK*bM+w|%+`nYJag5d2{c-wbplHRRM(u{bRn=gr7Qi-yBPSismdbZVK
zb>K#n8Xk_2TVS}p`G=Ef6fPfAxQ^aifv;k@KIK@h>DiXaHGQe7E#nroL|29^)PzY3
zlmc&?!P{Mz!<dbpC-=}fdg(b$QJ<%(`qrs378<9H^+#(;y%8F@W2C089--SchbpDQ
z5Pj=3SPgsz(u?k>(H<RS>)uLpzcp2lp-t5V-Zq4{{aOTTMPYhG%=gT%P)Wt#2dZSF
zih6vsf=a_%Q@iqNnXSBjA6rh_3YJq;KuNh|^H+sWeoDg|rLq1x`-h)8F7sozpr1NT
zF0K;rwtYk~#lI-3x!sED&lN?q3EsNG+j;qY)n;!&jqc>57kPZNy<mQI8qJ>C?A|In
zC7&kr^3>_^=-BHX`myXk=H`v|P%d8&@@~+HT<Z`126y1?4tSddZxad{_2R2R#@~zz
zd}h?Tf6;>Yh=#)5MMdDM2XmnA_;&EUbgbg4N%Y^^w`Gonj|b9c{2MmAQ^$;<Hany^
zvox1Y^1TCNql~IJ9VYr1HQB?c#xSwZKL)*9W6&@!gOZ;*Yvo#Jm1*Uy9j%=;EZtf6
zD!|MR2K_bOpf{-o_5~U>AKrQ%FsjIN&IP}WT5=E8Z87Tec%$4q!?`v_&4sy1(+%<u
z;|6a_^Ei787&PjWv%bD_R<(F%l`8A3TkVW8!24^DjOy{XQ7`cpAD~toG25iL5!9}G
z(UZj!H7C1COV=1x0v$bj3B6kGr#p;9fg1H@n^7qTje6bEsHs;BIz0;Raw}7on)VpD
z4{x0ca4taazDF~cdc$MEtC$pOtI1~dq`&<0j7d4*Z54QXzX2H;3(Yd4eREM4nDz^g
zmm9omVA0|MXw?+_TzHY<s2zB~+p<@w%ko-u#Y0uP4*4BFsE=>7=svtX+!kK*7^a&%
zMvz6$VdPs;YhFT+){@UncbOr2+@E}nofhrjaof#eSKuOxQsC|;d|Frd|NiDYvLS^2
zb_=ui*Q5sSVb;D*oYl~{QE1(zaP=j(AnM-@cfw$v_a?Y}>5y4>e#hGeZ@a_W!F=vs
zbnpe9&&7ro<>3smBNWdUye&b#$Z^hVU*W7{IiB+M)Utb#uZ6enkQ?($c=HAoc4rql
zKCf5oaDcZ};q9d+?4}q=-m}p|KWF4sdOt7C7?fY7x)$VI;HxF2$XjzNs;kBPRFmwO
z)u93G8!4%Bbt>z`yT%$hzL_S~YbD2+_KN%39!=9p_sVtEtmEC3?N|@x9@a~>%rlka
z<y_VHIah@W&eJK))7|C^*~zj@mI5nu*S13Sa);|M^MpPAir4i?_(x9ApTsvc9*@jo
zc)Q}BL+jC7Ssff&`@2Kg8?v8;Yj}l3W-`~M>JztXqnYJkcEq;@e&CkumFc-kl}?b~
zbA6qh=B-!b5*c!Vx7!+!rC$>bOoo4#8f(=)DqTOwbh^S^>yKqC)v+wQ-<VIy?~<nS
zS5h@#T&h;X1E;CXhkSssL#T72mB#tnRN6mDt?0QfY#XOZh2ylpd#q+Oh?U@N4X0Sx
z6eBl0Qs&k%8eS+y`{uGcsbI9iW<{wSto=ALN>?jHYyUpxl;1`x`bIRfThUq#Z(ZSS
z7kJwN-p);n)}OX$-JFKc`&*=jUt$iVSfmzlURjkMp^!Nd`r~l89QVVO^jo;@ontRz
z)d;;B6-oX~q#mP#Yn+MJa=b^K$xXO5F<Gr1CTS}gY!7~`L8p`D6>L?ZF;?=OtSY#{
zrms1u>BHNL1({b#XAUQl*-*Sg-pm@AcQfDlksPqY_{Xa{l<CcDgO4|EB06{l`NV4+
zYPr>+;xPJK9DR8DvnM&Pj3sZ$6r4s*LYgKrPkVsJ_&mU&64c8#9fEs!h&$WtDgbXY
z+10uU-X1gA*f(d@fP+@e?ns{&?w*2|Est2$ahO$msRh@mYgMjBaFiNx`#V-0=4&Uw
z+cVp2^pfE15V%{Gd==iG1gAsRTG?A;RnS$d3Ouk<N1?6^Z#%==JtvdZXM3_Hp?9tD
z_8Pox3vccHt@ORo#s!mA_8Q)&U-%DBC#pp;G;GmCUGPoR$6U-@!Q0kj@kMQj(&nv^
z%ImTWPx(^yi(RUTlb0%A`K8M7?-Kp;TcV237wYAeg&H<=p-!y}Q;%=+^``QCW&07T
zx$)CgXYmMiJu_V0uMAV|^FtI}X0TT0AEdQ`gVgHPAnoS1hz#V-7y7C7$lf|Ks~4KL
zmzJ2@sY7CYb<AFex$)YX(6WZgUaGET8PzqmZgtIGTTQ*11Z&2xAno?7EZqoF(-q})
zZDT2YEL>9i`CG3Q#NYZWe<i2;%c+UK`aJMc)HFYxbn?@zYDKlJP(g(_6_o3r1@!%X
z0d_hR(2NoVG;4kVWnW%E*-{IzTepC2GWXy*&sUBmzPgmp7adwqLpu6s0=zxCA-^mm
zyfu^k?tzAU+VqCp#gQHw*$xl#Vh?)u9?Bl&p+N=NZR%szHF$fs6&_n(xI4$FUNww5
zlbbr}2ZQ><+j8(W9p3(LhGSj{W<Pw$hJd%%-0&B3jX#O5x;&FyhEnJ&^hzFQ7nS_T
z-kjUiAL;YeoyavBe^J~fnBAJkgXP1#jEc&GuFY;_h8PY;7<AdwKvn?!gRv)@Ijd1i
zXBnnBtIZ&1J>>0IJ)LzF-aaTo4H}01IoYUf)N>oZGO9cGt>Nt_c)PZPQG4O-Y<N2W
z-VT_DR<2_p1K*&x`3#!+(OG)otm<*j>RsAdg94p(eZR8?t}yBsy;m=IodL^x!pHoL
z$&G4;UkXiYKhMu4%&72XMhV_FT?*svM*W*+)Mozg`g~1PB{;jyp!lPFKflBI;|8UD
zHprCIsNf1ljp1zYj6QW7z2O@0cKbN9Y!l4N@xY|TaQ>~0IioEm^^2ySy?}Z%JPUzS
zlkrn6a<Rx<-y+BV=j}h}&U{Cy%bvB!aD}W7UW;k?sbT`i4S}6w!{IZ&=RU=mDe|)D
zIj_ayisVC0w<yqR(fsii-R@8Smow57FLEtnESip|>D~m3-r)_aI~<QSHS|G7v)n70
zb+#q1BfK35Z=dF&Hr$c7H~gntXVQzl3~!UD6`Vrzw&r<n-)2@>wC;L7b{OWi#P8M(
z-c~{5Ui*jsZXo{hP_m@a<Td({tA+O~X)b5EOL)ZRxo9!Go!XCmP4Kq2Gy50tP!BKU
zE^m1Ini;j+XL6|wyfxNl-|}#>M_oMhC562g!}4iwM|NE#dTaY_ZzUbhud>lTsyw!s
zM!NZ{YYl(3nIE8w(@W}%p&a{t%IVIua_Y!g_;tYwTAaV4&Rwa?+-NsVcj~E%KE2h=
zd8WR-m@5}}>j`gn)t;wkjTh>}fMqh|TcJWpE417#T#MmtL3ledXQKM8OVYz8HtmbH
ztNQ`FF8poRKzQ4#y#o({Lj~GUd%kSbl{jjKlh|uN4*xGT$n8<oIh(E0f+nljbIeRm
zC^g-1W=T4t>+l*)Aj7ERL-@&U<g-*&y+c0jG4dp?u~&_3jYbvNpK2x#AKn_^Z7X<N
zik@W%yghx5jG~I<`rv<EjL!MxYE!3QR&AM=z^<)0rP3>{LLYXjQ;Y&~#AwxnX!W93
zTO5Y|{veY5$Wa=PXKcg2QTjPEioDlI-TE3q{&+Ndo}=}weY7sqi*@Xy=I$S*FJB^+
znIlr6A0p)L8mYj-kxE$>sUpuJ6##En@V;r&BV-#Jp@GdK@J)xSb^<kn{^1%jFkJH$
zuB*Y}Y8Df&Hi?n!iH_2@Q&#0~VAbCStBiPg!kF0|@Fq#6Hz#XyC97OVS#|d>s~W-E
zS@8D35PGK6n@#^PH~P%3``7LIg?DKAYr7K4*f|r@x8+P^a$&BObI7R)4)vZ(%{hte
zrWCTB;H_&mhsKB5Rc#dBroQwF@C%>BZ~S$RL+x5P*!KpfsS%e;wb5s{>f24L@=dU6
z175S-U#!Z^hM#E{nJ5`nwP;7r_MBBYPV?~{@OM7^gSV~g(0{E>UHKdtTKAK+ntJfH
zU&%6jqQ8p&Jr>?>;QI=A0gLBY<-d%t8_i>fSXI&gzrL;e*JQ05lcdhIk~A?bS+C8>
zD!wyGBc3HPV+&IQ5;duAf^z?gSD!=i%m>Bm7X4V$-*MW#K91d*af)BTEpMFsIB!iI
zvlOrMQdKc7RmYYKbpJ*eJ2u1UQH04dCQPSVgy{}G{sJyxdisI+S<^iIo^7@+d(BYq
z->2$D*l0P0jgZfp;dqRO>gt&xs$F)lGG7nWgvA5(_;p_`xZO+ZFZ5(Sx04#jwbsgc
zt?^N`R%l`~jZUkl@AGP@+Jc(;Rjvj*XW(sEh(`N_=s9&{=LW$l1#jCt50b56kQx=Q
zq*ZzGBF7ffl1e3165c*};jf4_?8_s+q8q-b5!3yY9_XjKIf}_qx{!ANDxl|Y3g|Js
zb-G<ZrQvO(n(Pj(Q9#v-6i|!21(b>>>f=B8)o8x2QjQf=8Rge^7qX<_?aD{q$~DMa
zou}s0X0&F7!Y*1r!UIj}p)Tw3o8Xt}okHDiI=!~S^xdc@FN3#Fsdv_jhFkD<h7-EB
zH*++z;S>GO>ab%udNz198T;w@W$^AkzHidQPuvV<xwJ9M3%_X^ygl_NzS_rlknq<G
zsASR_G-zk)#-+v?wH^JJ58g(?+MVu3-OdIRpW-84WY8rKgMK}5)&zLlzoxU6)^XO%
zKF%86|3ChAg1?6+JL}YGIOd8*hPU4?8}-ji>eFY8sxb*|i>90fZ~w6yR6Ne0Zqx8c
zQHOTUY0yDegRa0^mk-W5xXqc|d1r;X^Z(^=mZ!~GFW)=sG)y$ZTYt_OYkipUp>Es_
z-rBjJz7J0pyxlezp7Vcu@%_A=YSfTLFqiMoX)ijNuW!2Ep!O@NUvoBCaKxaooLe^W
zeU2<|RN?06ZuI@Xu)0qvlV%TxM`M`@;%B*Ovq{&N;;W)p+<<zvSAOPdsQ15wQ@8O|
z1%ET^Y8`svXx12X=y-S=e1sn88H;M7StF{kbI5K{^;dijynQ<qok|_}S26D4?Kn7>
zQpTc2lkjFIk`D!I=X>JAqSrcgjalR3n0to#-8Pu?oLc%<{AmxUr9W}Tw-rdwy$QAC
zQJl?UP2}3*-A3Q~^1Cnxp8ni}e+t%)L(kr)KAvI26UTi7@1MoT+a1I!^&43vXy4ja
zv#!9~)%48$(9b9Mo%y2>xgzlP%53JJ;cs#7^QDlbeSv#_b~Am$2M=#Mz}vo+$j9F1
zuHJb0qRBMdaVEEJ(Zl`FADw%cI!-BO$_;sSCNi%S;H9n4^XXxo{K|2kUaqUZN(cMn
zZTHvbwtgDB$zKIJmDRtc=<oC{PhK%QLTc9~N4cKrpXjDmwR_0CxEH$$25H~Q!8(~W
zl-)AJwEOThg*Th8b*C50RDFrIg)Y_N`>{H^Cmzpbf-d1N3BYq=-jT%oMzUt_W!|!^
zRp+_Z4y9iB;vXx!R;lTc2f4RD^BU9%tHIlrGw=iBnf<frD!Df#)BV~?9qAUQkLa!L
zm(e}xR=v$>)1x@@Bg3uoS;4h8zg71SChN}4WHsE%wf?(J!7#;&ALj0#)Uzky4_@eC
z_JG|oXtpGH`&UUD?4us*V$<h%E5FkTItp(W!rQ8b7};R$o9r=~18<``#pvzHNR^{z
ze&=nJvha-A;O#0h_y)k+&Z{Fd$T<?<K%_c6kJPgBQ3~1{rTOr7?B_`INhF#&f^4uz
z)hHUN#Zx26t))hMF;abQMkwDB`hSxmRNps3hxdkS`ND9ycMsRbCgI8fZ;w_ESC`)5
z%3d^F=SqcZctNZFst3>hr_<23?k-8n{XR+mwn%2q%&O%>t$Kh)EV7Z^<nT7n7`(rq
z?ReDiOOwrT=NdDnMW`oVgr7y2vuS2mTseL|{Jf4=bm(wyht}ky<_u@=z}qJYWIA24
zGuy#jD!lE}$F9%x>6@1)uY(@)i1*Zu@ia}#g_gGCH$KZ*U<h^Gu~waR{;zL4_g|~d
z&}$vO-l`QXtcv7w_Hk}ly$1#_wQ6h?D?KOlX=STAPz%0(BbmJ4Wc7o$As@NH+y31A
z{)MNUGpZJ~GB3ho^82*<5nR3kqq#ZJgWWJPN$>F$?~6#%;q1xuH`v|$Fi{6)CdvmL
z`?F+%#{G!bxBc;Y4{u+<Td$XK+6He+^p4ZHs&Nv${pYV}JfPv~J!diB+hW;v%-5s+
zbJTMe9>Th_6_jtbZa$u++di{&%{-m`X_J+!{&@Y_ag;Xw8mS|*M`%yPFip8Ll;<#n
zooGX}ap)jL^c|?bZVXi9qJiu^>8qx(y;P}gYXulusnFEs%vLwkb*E;^{8?YM-`3L1
z0<|>p-w+Liw`~qomnA$zhYN%#>v0u*3k+7j3qhLyEJzc|1*yWk3hW`SpxyP#D><}?
zf)5oWlPSMO6mXXZwX<V)$u4+F=HOps^ZZQ>?U{>;UL==c3p17TU3K(0^}bMd?LOu%
zUjy0271$x(5D#$MT&mKI9Xx$<G2cnvd<?y;HtzZ*Hyxp`I)4&-<@%7n%eCy1FL@4q
z(WPt2OrqbqYBRfBYT-)>XI68hMSIa1HAA^>qFc6qXLh9yy~rG}p}RqWBMe%z#h^W*
zM)lqdQ(haDgP!DzDa@F{2+zOJe(=_TPTu>%q<QO1s<+l8M>v^Ec!{6S!Q)H)^wVm-
zcC1nPdF&(9k}IR1N5jmhUk3g0#GoJZ4GPHvr{HaKcsrnqvqsFyslOZN)V%vS)g#DR
z*IGC;AL^_ekDPV26E)@&=*T~ex^f$hOAoj%yuCZpsD|+N%m3%?O+GKHra|+w)5C?i
z?WiT6eCw>^Yn;`qpR@Kla%#2JNuA3(>D+iHc49iI$tEW?Kj<XaV@~qE<fKz0owblN
z!sh<eoM)r+VeBscuZes=TlshD0{Y4C=nd~RC?kRA6lqWyc)KRWpql9RZ0KEcmO)>s
zEq`nVZ(()O`$mP8#K%nkIFR~6@LfFC{OqpHr&kPb+peIW3vX+|+n(@n%6+rG<Lm8W
zA@^e-^SbCle|Y<d9_U4MYzPcpS(;q}@Yd}(Jm%j;(YE2e$ZCeyB{Qk(!n_Uebv3%S
zZ4@=-Qq1Y_*xT#DL;jxwoV$ALq?Vpx(vrC*ZKw9VY7ic=#q<KanXl#bb!}kM-$h|9
z=cqqd;VY;9T{xAqK(bl&{@{0l=f7eGXEB(%bQzBu&m8VjJj?W=`*16`2%XIPx54AD
zg?JvE1qvTzzKC<fn7?>#)K9Y4CwpoJ^%4iQl5_MWnv)H^gskXquDZ=`i!TMyxbf~f
zOK$cq@}94s!aEg0b|k#b*3m;3>UnCdF|Qs(=OrH{pQ4(2Q#Z>m2Yc@~<tnOP+ly-W
zlwz9I+E1@C{guB;S$-Df@CB9E<Xh!+VSS)B+xeR=UX9vTE8QsDnz_MFYPPNu*}wxd
z{Ll!MYB&$wvOsMMEmZB3%j8*NIe#b1W!w_M9*+c7$-yk5PptN>Nzji!nX%yd`Db0L
zzH@z?fcCMWd9tV#{?oyxyj{>hXqT^@n7f%~Qx<i?%NvvRs!@Ux^2KXP`#2Tq6t6kV
zUc6WruOXxGW)+Xu@*?rt%B;nn$8qxcEk=RYV)Vf&o*q5ArFpW33{F<~LaY9yKDrFv
zF1E2xVI`S?x2<w&m!zzFiF(Y8r1#kb&4IUGZK4#=AWBKzQRKHqsnrLx=+#IK`#n-C
z;qBBDkt(`7QZLMr+5|&$evi<>pAnk4AVQhEJz_E+N4G9QuXd~zp*Bv`i2n-L%m?9e
zI)-MA<=;6Xbk!EAFdnxQy!~_|LQ9rLD4}nJQqP8K+s<%xZy7FA#c-Xd8%~Z_xbnc;
z{7&JDEE2AH@b-kaRnOq9E4<ywd{U>JoM}EJsU*CuY$ofAnyINYJ|gtuSlG5UnV)B-
zRg2J<2hpXcuaZGX9!YE1JKmG|oaXd*<KS_!ool&W;cx9q^m1tXOJ-!rGCT>NpE9R<
zZ=PM1>Qgi3G0wo-^5uB{5B4Zf?_IFV#@>5)9BI?SdiaPp@f^Ox#$(C6E@XcC@_gFx
znoqH699lN#{$zb`nyjy{lhpDwyIl`6Q-&{m^U@>*b50u`og@eETU91m)t)CSH>~Zz
ztth<hlM~Oek-D%4{om&Ff#K~6SX=8zvSuw$R{jA=Y8yfy7UsS=LLSSOB-QJeq!oP=
zl#iLJ6))m-=T^M_S<mcFQ}SE*9SS{74J9iU?`yPl1wS0M`#1L&lYP8MyMBc#J!h!)
zf0&1-VXnG%o~iHsXK2*>X&U%!iju}m)Xb6-^`Yqmb(uCo*@8x?EWBM0Z(SD+lgTki
zW&6R|QiHUA^gx+w4%DoSfm*Y4ApWwxGQ(TvMy<5S+)`zRG}EjK%~ToQF8Eeoty|U7
zuKraN6&oVQi)xC?UY+bLa^YT7Ra;A-?5P!*nIbPLf;qoyfqGx0qH^r2peHwqX!pD#
z^1fJ5=gGz?^FQAHK_(x(T?TLK;|C4>gum!1d4uQ4`|aeSUidTL;0LZc-d$b?*gp<$
z+rrzS@b+gLcB8@D>+rVkh+G;hGWiZNs|as<=5^D7c4YmHa?y>v%!I((r@hGBgST1h
zT-mwis?NFDsXoR<-_ST^Dv^B{iASdfnLrot>v&Tey>HM6H-mmU(Nl#H(Qw9tnrT6J
zyVHQLYMe=CcsrAPil)>X_rQikuE+afMo}w1TX=CA-p-yyO?D#eg|~BHVbx0XJ9+#t
zi%}!c(AnYb<;Mn1k1?ozZoE=^oK<?Mv+j61Yx|6x@~x0lThN~o)O;V-byjZry7NPw
z)pet@3Z8LR9ADcq%P1qf9S_Ia^MAy)HL8V~_n~i_z}up=@J7Me{eL>E54R6@oK^S)
zwPgCkwUctnmguB{<(y=h;H0D#PMj;8;E<E<9(7XC2`9zm%Bgz3IkkRQPWFKt^a{rA
zgttr4(g(Q@V)p3s8|IGo7|7Q%sK^w9sx>#LXE%fHt}tlXI)kp?#B&{B)Tzewo$*#Z
zqW)bM-eynw-|Q}ZTX<_@&S@<iD^mmB@$X46E)fo{*pKJvI$q#g@XHUsQ%5{gbKokx
z%?5AJ!rNZx#7ep7&31;dUE!BE{6cGP-Hm^EyIH-_nqm9Q8uo#nD*mbOD=d16<~_j2
z9>81A|Hsl<hP8oqT^AxqDORKIUg|E{b$3^4)ZM1;h9p4>Bthynb$554y1TbicZGgy
z-sk&~D^MgMNwe;~&pzjHJFRapKWh``CC&hwCeS}zj7%83t%&9>%$X|bAu~Jb;Df;1
zfRAW<csuP3{ae@I>`ORJjlR@O&R9dpaN|c`<=@N0+r#j7KdcSmk`>-QoQs}@w}<$f
zeq_QgL4yY_N2}wB{1r&wu!YPW-uNZ>I|zN-f;w|WTY8=RkX|<~`KqbEpKea^V~#L$
zR`Leu<k<ja&4`~|KTw<6X3*F{8OdTW3u8(q)p?Lv&*o*-tcf}F_u!m*?7>dmm3g(L
ze?HY~UO;!g7f{{d1$FydL6v+`NPU6|%Xe!DIpOV?zpLoh)2i~>P($;B+bXSVTLu2?
zO0Kf6ZY7P-lRcAkx^#qYU$p95A)9tgu_-t^8LCZD^pC}<0ldBbCti!u7FCZV>Pi<h
zNHD(N3Hm>t@ZO6r*uFxF=JEc#rZpKBc-yaBiei0}mCu-{()iMo3p;eES%S`c#;Mo(
zSpAbFR;`#V>A{TmFK=RW>36g|$%mfW9j6&b<5a^Zo?1HEr7`<M-=L$()wHU|3<PS4
zv;Ixd6M9p8s>Ldd9@=(;;<TWAoaUB@Q;(VR^s4VX<%EM_@OFkztXja^b$sp#Z#$HW
z(LecO)IDd6mV3o$4t(wNJzDGGZNlVel`j&l3klK6O&xa1_-IXnx2@pqt4~oXOg(nt
z-Y9n7N70KPrPWoU^g3OXp8T+DKXvanbL|>1-OipJGOeScl(`X1Eg8kU$|xPeOWsu&
z*8a4!hn?>~YS;HSNz&CsdaM%ZD@fFUg)aKGUCi%rX&ql{4WE4ayObZ^?uNH+2bmZ7
zAz258kw0obUlkWIXQJSI&2*nn2A}y&xw)>$jHU>*uh=)y9G(tMQ6&Du&Za3E!gI8D
zDl--PQTNTqe8XQ()m_W1B)e0mHaJz}IeS^tar?)TH6e=L8=lYB8#oVq=YGK30C*ea
zgYVD=ElXbMTOIhnJ4vVD?Fo3B{bZstZ%LGsoV8~-eORzDPi!K6f{FCAC29OKxcxav
zwc)KFyzLBckDzg5za}Y|*NXjNk}e&Dv(uC0)+te4>$!B!$3_3AOS{+^yLDlr>M*CP
zLNBL=7IwlahgQSeX`J0c8#~k_(4iy86BKcaT0O6`an#*+ESRUPwlHbjOnoRYQ~Pbx
zb$jS^MTJgR!&=kjm1(;A|2vKI)MQP`%<M(xF5ZE+t11lE{jz=KKeCs4*m`LCr|!BJ
z-d$OHc2{U<cUAW3$$meWd#tM_Pw1d4uUjjqXami?U00p+Sk!8MHGJ1<)Jv+VcO%Zb
z9+efDR84bBmr~}6MPzfQm*+uYMSL%;5q*lP$&+HrzM_OS3@M>$y^85(C;E5R<Wc!X
zIW%o{P7QAz#Q*sq_D#T4dNQ8T`*02J=AXcIclZYz^LAzS{^E4XnIoN&D^Rnm#;k+d
z%;;&84&RU2hAq;mJ9V>KOX+Q#=B+uR&!!UErxc!DE_}WB=n{CiqCY*Yw*SpHdX85-
zueUevHSA7b$1YfA88pHZzK^Hd67J=N7whpyJ9v>#rrwzE9vNPkw)6>F^QBn>ygamr
z%<$(v9-4a4lb&&y0B>J^;Juai;*HevZcu}4jV`L%jyZ}aJT)kZJX0r69h9evRU%VV
z7@wBQqD=U{@b)ws+WZ^+cL%n^rTo9lir-<@I(XX1Y*vdRCK=J1fiWfpoiOQRF8=Mm
zI_?a-X6{?kDYF`+p@TiiC*kMLg}3G4?GHTNNz_Zz(71+99{TH^SvB#B6Jc#0E>GZX
zjtgdOr*@kUo)({D)R}@tt+{Vdmz_9RH4V(2H>gy9g9<%0XcWBd0&nwdHfgt~ndio=
zB3sP5*wa%xTHsyh<90nfnXkib!T7l&Jk%7=`bu*TeU0F6TkN5tkH~t#+tCd@RT54g
z`av&IK`-iz_z9d((xWNA4`Oyk8D?x%^-?!7TALPmsqJ+yHJHwvW!PEihPN923qNb6
zQ_XJdxJRSjil-+R-rA_Smi@|XkWIW6;B(&$=@i7zB*NQQ)KO2;$J}WbI`kc0Z&R|)
z6VSb|ZZlt7N*%VE7rPYBb2i$Du7$V#;ce?|^fPs!o)1&6;VFl61~_GbtGt%1?>$w~
z8`h$IOT$?=czg6By~FUf`T#EtAZwQ0$(al$Ti4MGwV#^wZ!%&znCF3)GCCSP4R6!I
z{CTa@$vTBx)<gatG6Dm0FgMG<KEhYb(%_shZUtEZ{GPu(eU+bl{4jckyY}`|_3wTf
zJJ(-NHnX4it-nIV+^L8R8q_?a#{Fi$WckdRpOTs0;p`gUJcq0g$T~#l)`rl$`uH-h
zGIT0HcAR}%!`LhIqoDlZZLTpz^}I)ArIo9qkKLu9#D=O>wVA%WZ>?-Q+R?|~PCLhT
zQ1<8S*BsSdZ@>0b>a~edt1vl($ZBT85B-m~72~w927WC4G@D}+^iQly9XluKGhV0V
zF#VAa$f?5Hz=|oF5kxK~D}9mlj`)`)6SIAuYDCPV-(a40#mrN?!}B;V(et`JR+S6I
zYQZV|*}xc$KxbY@WA=x+JK$~leQ_GSBTm8T;(14yr@?9Sw7zpPeprf3Q&V(@Uf|o*
zC^wglrB^am^I+;XL#(`i$EfGDdHObVo>sn!QQFoRHKN9PI3h-WkyX00KSt5;_BUDL
zm-KqHhqs;oL~9<J_0zp5T@H*^wfxbV;}NYQ{?XcACR+E%IhoA(!|$V*VH2g13!)S^
zI!bq%MyV^jeFbkneX}dWQoHI!+GS{Lm*-o%W~bS8zd@9mQwzWF-L9e^?Apr5@vynZ
zZoYSyUCFnSG^}Hi#`U8QD8QwV=`IDq+rN*wROXvY4qxU}^>e8-9>ZEsmplxKGUY<k
z522R^ZTfaPOvPtz_bG)r=6DM5c2KeT+RWFUR+=vhyd6H8o*A-CK6v=2gHts7X9}4#
zJpWP5F8u9O>19qiY)<uJ?$xt$%&}bUQrYV+9bB9!A9{{+;PrNZkCAWjIgYwCA%s5G
zNr}1uQ`=Qb(r9$(p-M?Q4sR<Q;Oww2k?db0c>s1Sw@TFYnVh$FCMvm3lAfMOQZl?9
z{()L9ybXl6o#E}3S4j#lmZa%?U%PV5sJQK7*Re}SEiRoRKX56dOUu482h-}*yU|WH
z?c`Krc$**I#=u*z75F2S9U4RL?B{*-Ouva&^BJ5GsU_4uZPi~@=jilU`gRJ=R6+Z6
zO&c~{>$*&5j?Z-Y!P}hhHY2<p0&icTZ*ActHN8Bu7atDL=yAQ(Y9V!5M~D`_4N<{-
z-St<A?wVezo9Z9vs@vz=DeI-idjFw;st&0y@9T9{4&HuEsir1lt0^W+RlOfjLB*Gs
z(cxMp^sZnb-6&N+{=W-Tw;>1Bt&pxS&m_;_0DUbNpc2>p)t+9#S7-cW9_Xw2@9=SZ
zr`OL6>@-KyZam_nhWmXquLF7BTwD((``g(^V|Xu($<K_!1?f}>-mVM8FU&($<{k5r
zoOqEb%rPqArS2`g<eA8O5xh0lM909}|Mr_7N2hGY7i<Y{_n}Rqc#r&VR)h5geW}wu
zb-fR44fj9;c<5z4bRTT>hqpfP_8IllU&qZVumLSN*{p(V%xV*7R`cg(&U9wAD(|80
zXttaSJar~BT+PE?oLA@{H*X2vwmL^`uR3!e`1g$H%}wxjAMe-St8$z0_CWzpZN-;;
z7~rYC@UZ`L4~4z*Q0YVu#rjdh{bAN6GQ<0RnN+vAN#(B^b!x3qJ?G&k|HJ3G&1&7i
ztbfp-pH>)|gKyNIl}7gT88!2+QTMovxnflNk$A%KMon&N(jmO)EFVpZu7i)f+N6GX
z)aKqMbuVwygy%*wX+|}hXjEZ0qpm$L$ZL~9HQ{aEW(F02w^uJ4*oA7);%kQg>EHDx
zy@@vKDfM4-H!{i1IeTRBls~-f4{z5}>%Qd%XYZR8*3?5==6Y!F0v<D%JM9~1mfW6N
zTaTY1XVo8O{nge>g*s3pErdpdUvBw%T+8tow!pVtLskifc}zura=tQdC#!TH@18pC
zO!|XD;cXvydkWtE#3QZ@OV4j$rU;B}ht~bC|Mjm0)L7Sf>*sQE!+bucg}1W6*6-e&
zk?7O980W3Y@HX7!ty$U0dQm$tt@P0L9Uk%}r~XfCdYGv{A9&!QfPblhSHnl&NS&J3
zqn8_fO|keH@OCU`uTjhJ#^G%TYQCP6(drJ)T}#M!!`o)?wh_FIhqr~MF!KW5uD{}~
zyIIod5a-7$L(%2%wko`RLJlC~AH4F<+%I^$0p4DMw>{wPH)fRWz3Qujo&99E?5Blu
z{1vl>Iik<~wWwmC^2B7&xkDM0o}Tk(qcZFL=FA$mKP!9i$Tif>r6Nc3s7cqn%8JH)
zSEzuxJ}jW&Nd@KQRY+YMR%G`>CC&M}vc{FEqTeqp^zb$$|JzJOT+Q@&c5`-fwpaOa
z9rWs1C-U;$)#pi1O}sc!`QUAy%yvDDvMbZlDD9{iqa*0U(`hm4-^8Jlg`6sn9$7j$
zS!w94SNG_-e8hX^aq5s;>5=qHW?yr%ym!Ikt?WdYGf!^t_Ahul_|QB};`7d1V>KsR
zteS;IYdAH}erU@n@OJXgXoX(J$9^Bp%yE2QZd=voX=t7}Evp@;MS~MGa!`T-8_!er
z%Jaxe#p)#*HVdAymyv8SysbnQYB>Jx+e0zb-DB`1W8|n8ty1(oZERszBQ)rg7_{X%
zyAIs4D?40lav+NTKT#^;9j#@#$UDK;yzur-iD-H^qm`e}L*Q+heNn2H93?xvO>GjT
z?Ukbx#<}FhFS~{;wkyesZf#(f8@%0|YS*RuQR?T0w+wF^aC>gBw%8ZDMpd(GUnRTN
z4n@N<@A1+Hhbn(}YTZB=y?`#+lUy*4I%QtG)SfP8NjRx}xm2K&i`gv9U`D5&Y(Q4H
z(|j#RL+h4G(adWp`nWrV*#PwIG@VZtAw{>klLw}cXUL``?T=3SKfl+rf1;}6(R$`c
zR2=gqzlAx~z5_ERb2@d*%c)P9o!ZnKeGPBzco`lusJ&8mt-_4M>pAJOL3j3?n5gGY
zW)N3Q(xW*^DqA^856F4NF5~O;-8B5*(wzEaznUbfV0S#=W6YASkwnjEl3GyH?Fes&
z@)`;;CaW>L?F(-`3(*6$JyB08CaTa?mxjUHX9Zl$ad&Ca9nJ##n1LDT)HQg!ueMX}
zybh1T*egF0l*&xf<ZlW3xgtSHU6=<oB3?zO#jApUoa&uqhAF%afw#R2&s4srGc@z+
z3~gLGLuSJaO&Lf|f6Elze>qAib%*NTLj&~eUWgtX4`J?Hhz=x$sP@<pb%M8!hap;)
zt-I>Fch~7$-Iy2FReST-m*HzYYUcIuWh-g<wqR{s8LV+Z<QA6{Q=@@}^_HAd<-Pf|
zIx)A*C9<o(GeBd91gL>GJH>9$n=rs%Czv~Tg4u?>xB00neS$UHv12sVS7E=>Yk!6G
z>h+E}1LQ!yrQjoPqwjD$a}Ck7Sv_FzN6zz0sKXs6`x(HVob_J1gSVKr6+V&4X#=mi
z=JHf^^kJoQ_^8uNviHMt?QYVyb|!5>7ZzG*RuH@#12=b}zjnpotM>QQ<3o7C@OB>`
zKe*tb%>D5@>%&`WrUieK15W3mEdQEy{g_#GH<}e3PA;i7+BVj#o6pSZ|J}^YNwXUN
z!H4z1XNJM)cTqb{@1?<kUMiW&TnGHd<u$w{csuBXr=E;p1_k<WCcGU~3=PO79NymX
zC!-AGYf-oD@sd8B;oOgW{5_w|YMKb!@s69fHR;?XBl8{L=@_HNHZp2u0i!y9Flcoq
zqgJCwHxxChO%<bx)-~$z1fvQ}GO9nky*JpX<Cl%9Xfe@SV8VwqX<=crT<~_lY?IoA
zm^81PN$W16Ys(mw;gx~^n+9FjXwc6PgX-il=vh;PK3;Oyl*8`2x87a*-ngqe_1%>N
zjdHBu`=1-wqiIzB2Sz=sXVReYCWSanszLqP!DBTQP5lPmKAY;H=1vcNBs(=5-sXq5
zJ>l(&WKW%fSt)2mzkJNk%FE0ucP~B4<s~09?q+o4#BQ9c;8zNsh4A`+H6zbNCh7bE
zW=%TD{GwtOci?NByPm`2X4FxCaczdnU(b+f#=pHVAFi%J7cYTvWR<MV*^yvih9t~w
zOx?I0=d_aWb{S`;44i?y){-e(f&b0fp_;=(&J=3uM?BQ-l853Saz=x<$Ctv~yJWP#
zd#Y0mX9cokg%{Bag_m)kJecD!ud#{nHo;467Bf!--i|tmcMfmoz}vRtz4a&x){<xb
zbJkn8;O$~~oBJMn8$8*!K;HfN1^SBLdaF(>XN23__sz@#tCn8l=@Tx4r)(Fy7B;bO
z>Z+ek)c2RiO@Dnp6d*%W>bUTB>C_;Ni3w7=bwTv2W#z1rRc;k>vMVf)deBeT?_(YX
z<jAZ2KKW#{<kS1#1@#TyZiTm%vyjCoSw(l@?Fb)f;efh2wY52W5?UzrSu33y(pH+(
zPU(KMCx17XovAa`cV~odg<DnidZd<23D=Okc6F<c<|_`vX2j^ieQKNJO`XnU{X?zo
z3K`Z4vH#6FzzemzR9H?;iipz=?>HUZJWu0>%~J_3C1LJ{am*Ol6U(lTSPlObqi%Df
z(ecr0gMVB2WVB{Kh}KrFYwRI=WRF(;@Ms-F|9)a_#HR}L^tmD#W3Cql&!g{Uo^rw4
zzL{c`hQ2KbZ<R_d72Z}ni$9H)y*)oh-fwN1e$yuNJ)6>++jV&uelI#S_^e$4H|(;&
z#Y2Om^e!e!C0|D=G#i}#9i`rB>`}`etz?*c{#TTAH%h1Rl1nB<F$XJ3^YMz$;JG{D
zY1yZC<=J7^?j%@V(5^M`H1W4xqv`LQ0$=Z~u`9NSU6F;kttxgct!mdEczcI>_2FF(
zwWUry<%^T^s*Ao$m%`DrMX5!*PP0qs3bk&0-=6rXXBxTm3f{K)$?T+3Nt*FGNlQ4_
zIEtm{@EP_P!`s}?>D4WSpS(0hdGPh0&@bC=J$Yy9!k<ngD(I6-yJw&uJ99h7oO%qy
zsx#B@@9a(;dFP-<z@ZEOIP~zMLvw2-DumfgtH=m{Yva<0tchyVEm67P?J;zy&s(SB
z%`P2i#_u|bHugf>qQT$4b7`Z6*@=z#S<VcNVf3T9c;-q}d1gw#+UL@>JuV&E;Zp6*
zE;U)@QrG1!eH-o43I4`L)$z1zx-=@kiy59SMc(3!x6Y|0uy!-G|3lTuN!@m+VRr80
zrv&YUx4RD~$S^xWG4S@x@Ob^65|7^;r=PQ<WL|Dn*Ap|<Y~f7R&NovleP?RsuNm6C
zV}{mLoUTHtqqL@GsD9q>r6Xl~soaeay*L)4xvN6x7Y@<eF(E3`I7DYsyQz$|o8o77
z(-+@v`qZz1S}tW*&jV&Js+3m0FQIGsvgvg>>OQeSx=|sB{@W}{iOQ(8KQid{7xvhl
zC6_Ui{W<V<&oO^_t@qb-=JTI!?yv9T*(q9@dBV(oU11|vIg$D6)MSl&eRO{@JJ!g7
zoKB{$y55KPY#%kIx2BGXz3cEc$0cS1AL2a;-qu)%*9dRpH<QyDj1SBEU37>CJ2*_5
zy~3ml@V3<elNxrWz6ftE@OH=oa<}mIHg(SBOU%qGGOKi|S-IYsRX!Phjzk|;gO$U~
zDhHEidwVDuj@I2|Rs^{qd#G8fewuXojY*aAne{#j&B^zg_}Z%puom4m2{t#uXI{}6
zPj#?a9UjrM<MPmzLu7y9t&@y%zCq|X@<rp}?X4>G|G?S`T>8P=sPy!u!rN^B)os@?
z?<t1&^1OJzANhUMZ=cGn`#H?YhPK_joGcT(z1_s9%<#4sye+fWpnrE6bY>ge>ugYs
z$p&3aF=!WFaP#j5CFC?RGXf2q!6<Jo4bZN6@P+UFr&A9a^kzG(<a*{(gNmSMOYAV{
zpXCOb;te`g!yscO0~vpVGF*1oQFyyxxx38Y+|_xEL0e866tmc%(cI4LaYpS%Q^(ea
z>*Gus0dH4a!l%B)<Bezi2i`7#x1VNtsKPI!qF)&G;*F6^vq{_Mpnqe`${cUjhE?RT
zs4v%z;2e^jxg({#n0w=;e6c)E)5#;j+bZ8V)11Vs9l`vJPF@Pk;-#||bZTq*a3*={
zt;1W7L-1*bQIGA8&(Q|XHuY9S3bj&rTMQkW;Wz$0o^eq)xgnT(!y{&A{o|=B`>7#p
z<7}4Zp{Hr|Msfd7<wnzn;y=LKT|BRKFEWFKn*MTl+k)qJ?cV?NJlU}nxH}A8TofMz
zoqx)QUMkLZN2cJZ;5DCG<fRR3$;%yJ2G}$9H)Wy@{g<~6pl=s&cIeR<J${8fhFiS!
za{>KO`0I73Bj4ulIEPN3#`gui=UlN7y-ROUu@=6%ao$(ca>Lo#?9X`2Z12wg+JAvv
zRO<p%t7f2%^kv7@!a(i26R5RWGN@#RjOtPoUwK$I1t(<J1A5i{53)NK-Zn3pPfzdV
zQy9E`P`ZLP_o<}(St~1`Rb^cnT~)Q4SafkkL$%*aj-y9Y)d+1NWB%6qFut8CjBT&o
zOFPP5t((UE7^ZLOCTZ-lNDV3#u9rQ+Rp(;3p1%p#C-(?78y_WWX{Rc)PiN)JMC~7#
zq+6~e6}yc7VAs%Qbj53CN-V(3d$^HVO+)6X5B~1%S@ZDb*pYCF+|t`<C6dSLG@Uvr
zyxrzRSDxVGOVO%yG+HI$?Rn-#RGky8ug{|u?-`>Exnp!RPmDtGf)AI8(SS<y;Ixa8
zN9!0h|7cf^GIovQ^wO#p^E<fAtZCPmvUXj0V^iL4Hr>l+BR^neW*5AhYL~}0yCN>y
zWzQR><bm|-Sji3FjAFk4KJkkv1;E=f@U}m^?FxVYy%MDgEAWvWQECHk-;g<ufw%sD
z=vThY?Zew0PCNS^Y+3<tN4&HvZvaewW7m@9c0Dg(*M)p`<*Udw%-sZUk6ces%BuwB
z#W%<fM?7-6R4@bc@xSpno^Yx;yj{TKdI>M}=?|wSqj9^Jcd7jZc=Zq0fr%=M&%VHu
z-kswq%1G8J_z4;pUwPwV=1EYq{cjFMhD7Ewk@c+x`-XmW={CIW)6k`-a4Zau_5{!8
zM|LXogtcc+IrQqZLsPl<!QJRz4m=*G+SYPu{SEd=@w{g_iMM;ose~8I0Ck70@b=ST
zm%`BG&3^MYJa=hxRkV4-L{&!be#IZJLQQ)AUYCORvwH{Lo`kpiw^5r|hhGkFec|n#
zzAk;O$o->U+xQRX4Qk14-!a#i+swa$y%Oa2{+n|gj0UeYhn!tbhl;}6LhyFN{sjG=
zjQ4RcUPt=JE237sjN{_y^^Ve`WmbJSGE@2CZ6dsF0B@(k+bxb6a%~}>{2)~E0io*t
zrk4_Kc2_;vTCqwu4eiuTUH<K+j2%MszEOzI8bkD@WH;ukcGaQ^9n}EdPTo^m+jEwZ
zy+89O-j-0u;B2bVKdS~t2kFq$EUI@ulS;?2pZ$6Uxv~c8-DP%YGt<90yd4E^7q0bZ
zzA<}XoBFE^ye$cDx0ur_EtOu?OU&6j%r5TT%*;XC4u-db;cZoyj~<f6{NzW!<PCZ#
zIme&mJ@pTqjT*xHZboMC;{(4~?Wr($+u@prT96qIfw!~ZZKWBo7Jcj0i5yQYlRP?@
z^m&L$!}ptXvxHeG@U{iK?fu29ekaVTk!V(i5$M&7X8lG7SAw^@;qXgz?I;+%b}gI;
zFjJd1X)wHfN3HTGIbr)yv;G={c6OQ-Lx1dKcMqjQ$K`>yW9g~+g*V-eKG)xSJyZ)n
zd2O7h-ox7k@OCYJYk7EEEw86~q3v40+g<ekbh*p-qqFP&f8OTF=b={AJOfvoRk4Oy
zkKyg#Q;f<w)Tkc=jJn*=s9JfAstIp1b%T#33|dj#AWwK3Jky|#^I_l<gWAKvy6E3`
z8DT7}9pa1EoMzBBzE%a^W{0z(@OB2Ay|e`0&N8TNgh4N1Zze0h1K!s1#7~~;u1a;>
zm6p?8%TwI7{gt~G4>xGx33M>G)#96h-Hv$Iw~boY$|NTYt?ka^Rgaz}&MehB@p$25
zZzb>L_1UO;KaKkOl=H&BMvd|?spVRedcj*S{Q8YNt_A<$Zw{r$2%hEt<DnY;(7<Dv
zEiKr1)I&A#c$cp7(&u4l;P%wyztH3Kg3LA9=)0>p&x~SjRS|sW>ae*JwS9OS(t!CA
z>G+vHo;qh^m%|r)hMrzJ_YEH#-iG#owRjZ6cW@?xx7QnaDENq3zE?T3ZKLLW8l6r(
z{px<R@}K23z;m`0ufjT!n*1Iz?C|ytI=yW#`k&k3qg3})2~YHG0WPz64WQc>Hviw8
z>Clzbp>M<Bubdl-l9S3#Z&3!cYnqqF!Pl1Xb{D*DLY-uBoR^LaAny!sGosNyKH$u-
z8{fGawU_$7x^UK4&Rl-<>G-J|o^qou{<?6K-5Ra{wS>3d`Ufi4yg==_7O0oZP#f=;
zQJFJkQt#qf^=e{P`b4uTv0NTqACX`8b`;Sib20fR7gw*y(t5SOv}XM-t?}Q>$^U5`
zb&IU0D&rfd^OL5`t7#$2gqF(Kw3Q~$YOD2++iFw(_Nsrdmx5l8*VY2lb#CZ*CA<sM
zwmRX;ZVlJ$b>ZrEHA0p!=EVjx2N+K;j^33_@U}VcwXY*wy4%gAgiXwyWe)M55A&1^
zZ(nzrrz*@>dSQ*$)H=+3z_a}mfRBuC+q!SG3c=fb@OItWXkC2=&n87F?QxXSkBBD!
zOy+16eH<gB^mR1eFAN<5LvwAiDd$w1GWWEp|4Ew?(3ICo+o^lo^^NaY%4H2)JNL|{
zQ=6^QMypz!wW=Sy-3M<c!rM`+>`H*Qvv1jTCe5znoFnR=!2jJ5rKZoKR4O$}RWn4>
zTNte>@YW7*A1!Bo1-w1iAxiu3iW_Gm1BE^wa|#dOqm7w)HYIbt3f_hvu`3JAt^6E4
zJ;u(yBY2$0u9g+->QKtA^d;;Xb}d2u{z=ehW=}S4hG)9dsV7`UqG<yU;xAKEZMT6u
z6FM{o23LW%%gRyP#bdtTK2d#}u}`-d8R6t)MJ^(D6iR*8LSBtI!@m}$Xj`Eatyo5X
z4(Fk^)5s5_H>X^o20M#-aDA7Y_xQa|{Nx7?^*QIz-v=Gq2X9xy+HiDtxqS})uQ*g6
z-Zgy&`{AvLkC$*3=zrWvuPDrY<6w4{OZj%vr~Tf=&UNM!Q+LgAj-5OH%*;X)$5Z#M
z2V2X-+idXG<B&r^<DGiQUcx|h>#u*Ds+q~9Y53*OD&S@3bm=00Z+`yX-!OgMMW+@m
z=Qam9wX20wSveb&Bkvw}jJd`=6Ex^_f{N{ASI>k5#T|;rJBU~DplJPm6rq-9rz@e$
zH0F>_*6LZ4^yTJ6dhsSI_QrTwBgU&Evll;C8p;mS9-1B5Tvx-|s2^-?9oShrH+9mb
zeVz2ZcxNpv)?Q7zwIh$;R++zaP;%Lhx;@LHn!e?f_PV&d(-+g_3q@6+TLJCuoK5@S
z?K)GCc6`jF2|Y8a%A-KFfVY=c258v?X32F5P<M|2l^yP{2KU)<w!lw!;q9JMT;Ofq
zG<K|#hboM=y|T`S_ii6$gSS5tnI}IF{=(b&SJJ7QH@y%a$^D+jizQ={8EvpIgO>`z
z+u!iE51O!43lGInKMgN#)|p*oiBe4Rhp+o<n&es8#J)wi%je-sOd7Vxq=XVC{myOD
zhg739i@NAMc-hvhb9kWHd`)UslA35WllD|KX$~Lnt76hrJ}==iF(bpItglQ;JZ4fl
zvPv1gn$!SBpM<k_sZVyfYu4i4X8nvXDXk}8pTO;KKa=pjJ1q88!@kV!@93$g&G@`B
zy}a<Y4ZOVuZ)5P0qYu)jbCw>}U=OV*#C0ARbJDEmlgz3IZ$q+@OQH^&+TEx-y^LDy
zPG8O^@=Nax%G|}E!|--Cyj|bPpjPm91Pt8$+@K2Zc08QD32z4k8rhF%R6%(A5Z=DI
zV9>D5XjpXd5P16%of-gd1Lqo;gJPgQV<7kAj{onjNO-#l-d4)tu4jqvn%Tu&mz%n4
zLOXXILrWjD;5pk2+WeZl6}$}#N8i4HueHrO-wpr0JUQrxCiQ{0%@-Tl!EDrPvRN%(
z8g<%ik_z&@)TM`W_E{K$kB*N&v<ZGc%xXZE*}EWlE6!9GtekxgGcRMghtA^zylUd1
zeOww2^iWqaSc~E9ZFoCnG`ZmX^g)+kCqhGdv(UY!3S^GqZ8*H$^O(Nk>z*p|(^JjA
z(U(M>;cQnvU&eDehYK}=o^Lrj?c?=<M&E+p(fE*Ag*P+f3*H{%xe4LS76EU!kK+u5
zA5k3Nq$8Im|M?vaJal>{=Pl|34S23w&%}3zx8GaSOGHln0kz#3x5$5;Ad^*q`Oxro
zCcJ%zw|#d5d9nA*G(&$gBu-bOz1Zgfr_scXMv&1y@2#!Op^UBazrFP41HSCI^wX3%
z>@<4dr;p9ZOYLITw<$oI)3LYtqrX;>pX%5sKwTaMs9)_s?e3pN4nvSOnzQQduH0Jp
zHn)QI71phtMbx`lF}0apTutEZBY68@Vp%<#U7Ojrb(9_6_Pp3Yn;ebRba`X_9^X>8
z%e7MA&Q`LWYpwV^?NoeEsGb-HYwh4ss^LDKIUQj#*9_O>$Z*|^i%_PE5h|1+QVk=l
za;GPzN1tfbOrkb=ELziVM=RGW@<z;lI7J^uiE+%Ntr4p(OJdbHC`M1wt7~dSlS!pN
z^?H=<rlWVJakS#Ngu>X4b(yQzJxUFSMX3?&8*GcxgOkjP_&Z81!r-0Fu9#rE&Ij0K
zTx7#9v}rfGHFTFvpV5u;CR=s+kyXL>txCOPRrxDileca{{chQMtI904>ex=JF8;8x
zC&2c9yqyMbuik`(Fw>jc{LT!<e%#*Z%TbE@6{U7=(HhC6EW90b9JcO{Qf++aSa@5j
zUX<38lX^$pR+HfUb(<RRvuV);o0`wE@f_ID8#Y-F*|g<>P4{!!HHXV)IJ>o^U5|>}
z)$k&Ej9K`9SscoS#;o|np%L*8%}zqM9d)QdXSCfCr!pLMDjR<DcMrVRV3$_Iw0!NT
zjnX%?s$sJ7#wRO!E;BPGCF=yUDJJ}(Uu!{%;tSHxvzS^seKsX`C+b$+L=A(tt7nsm
z!nZwk4rWF<6@8x?GM@IkEe_>Ka4@@o{86k!ZMHjf5Z`tjjD7!ro-TN6=Hr8~Hi`S&
z`L{#8;q5Uur*2eqYH2N}QjR#~i7w4_hWq%{rSz}=x9c#j2U)O*F6F)F&||oM^Z>hc
zxIWSapZq3zHPEHWZD4U(vd9_nl|OUVfVDx~-Y|A0hWyPA+JR0T|K^Zs19Q)qBi5#L
zf<j{Abzpy-wsIMEC{7dn;^kI0UZGi{H3Ht=KQmng{HLqUn#o!jK8YPC6ICJ0M3uij
zUVq?iqXuJ?-Hkq=?!DD-581(KO;jwfv5HS^s8bcuuczzlrLn%|mTw?en+7TnTTl1%
z)KZ3$<c-IdWad#x-C>8ot(2l_nov|xSM#gmncV8LfH^g#gH*9F^`Ie{HSunsf&&8e
zU`2o~j|foirUASb{bj&QK6Tqq{hWSEYso&7l78AyB)#S_qwX2J%?NL2t;IKnvm5Np
z(s9w7!}Y7$KH5yJYIj|-FE8<t$u@bFr6;5S^)6rPUDV0`U5b}SUF}mHvb<LGCcK@C
zhINOxpC+4BwX8`^x$J<qhQTJCg|+7vnsm3RNyW38)ED0VV}hp{Ok}Z4%>FVm50<`D
z7@HNoW`(z-_<Af1_|^lR9B)$PrzS0b3>Wso*#jo^d}NYaBYro(<2^a1O_z*1@z_WW
zgYWNZ($0}49otPlsi#@@$UFrsq6Ykvz8yT{MZb72eubWWhDQl&llZp^`{|WA;~{%V
zcK7h!T@c=$<o*3lH1GGYDo+-2#Py8w>T1*_w5`PpZ3}N#zBA|vUGVRJxogsScl~Z>
zP=A+!oP$9_;cd6C2CeyQ&^Y*-h@Sm;1F!iPIjGfm!SM!7q((ae-ll`MRpIUW83uKo
zZP1Z8gU<Pql{)RN>qp#`I?<hd^zNFP-Ch6sxa<69H$5%x&Q4c%UHs&(&*Yc=VhxI~
zX;dL<y0$Qr^1V0dPA#*_cEzVJYu1{#JU=_}kQ0r1M6GuPdVNh4y}j|q|Lghv0Ef$G
z!oP>lYvJ7dZzi4CU{XCU$FA@kz}riIarSv=)cG?;70JcN_z@Z5?ZzDBnyh5Z3X^w6
z2dA4#9~b=EQ;h!Bip)+#>;A%r9+=fbPkNcPx+EGE@AxRMr#R}ybvP5<?D)UeR;wb^
zn0?X8ug$83pI?2cS#ewj9zfrAq4r>hy9s9b9A-X<Cz-F`eEtjl4r{;i@hNH&6&jKe
zn?b+!Cp_t#=v@mv(0%By#v|WNEk6U?cMYc2PfhnVe)xrw?1$ntUJbtXMtA3U3&Y{<
zBM&d#;JPz4;Ro>6uN5;ss8PqgVdhmM&Z(_@RewJ-yfgag`wTx-z2`?yH+^#3{gv;D
zzv|ljRcf-o(%z7tsurNWWEZyw2B>BgG8W4MRPIBd(%@~jm$?<*IS;)Vd6nBz1aGsL
zUf6<F+f+(J{*;q#a3ziZP?<S;mDRXWHTCFLUFCk&P_>zLm{r$|J?1S{G^w?A9&D?`
z8Xc9bU8p)&7^Yi^Q#A9*RAucjO~z5v*q1X+o14%}ab~vs@<eLI5c*5sL@8$!I`v4j
zMltVUR>2sxoE@vze(2j6YOA%QWhN`MlKSR4eB1itm|Jxloe0mWdPi$KdiQ#xXyye)
z$x(?q=zQv*J?u(3Z&kJgt9s3}YH53`KHiCx^-`n`9*&gvfk^c^7OAOUBUSjeRg(`z
z>P;7`8p7K-Tz*n}?wn#((mAUJ@_S}%vMPL?Rjn4`DWY}#zgd+4ZwI8=)V&YhE;Zbd
z<eh>ClOsM9rCdv3{+uZKSnP69d(H9{mZM3FqDdo*+4V6yp0STzMQ+;kl0N026uUA`
zw5uQb+HIaqGl$t!WSWihjZJgm?W2P>t=VEz)<-s#^v7d{w=wYc48JSmkp%TUnxNM7
zO1&8GP_bx-O2@Dx&gGDcUZ@=K_Nbj+okTRC!>NDpaDSDc{)z|dHygk9O`?u4$0=*a
zWQ~QjIp~RvU`|u$I(nnuCu{GgWc?k&*V%E@c^kZKkf?<>@qZ_?Q@si`Qh3`0-rj_%
zU(mHlo4FtGzA(JKW2YZ_r9&&PIh5g<Lj@P%cf;EP@b4aa`$s9KuHrRs-{VjtU-Dd)
zoMdG2JmBq9ypLXAm`#@6sZI~5J*>ueUgOl+3H;1=`l@$3v}`y!n!kC@D{ceMcSchO
ziyCnmm%P!s2ViYpF41V)hi9A`?m`RaaOfLaHa;dn?Iy&le}Q=I*cqpZo0%cTuB%XZ
zJ82sIX6>Ssk=Mz~hY<>%6rp<KrmO7=c66+qL~d)MhD;le?i#1aQ{%K~)JP5A&|P1g
zO?7E;W4#{PSf2wLD>k8_cKX#<V?%xUeyOKb=j$n8bv+fERZr{T?Tg^b8Vqkc<SeNi
zhl6#zc(BgxE}_q_bE~pByO#9}QYgIbRVYZ|@OEHIAagM3`GdEmhH;+e4Bg2yKz*k8
z>+uag&5Q8UgI~VvYVuW$v#>Q5?|8V6Y7X&{+hltAZS=xgeV7aFqvjDlYU1Ui;8W>z
z(w8~1<H%ytx0t&M8BgAeey7Jb<oz)hyp4vpHOL3Y!P^N9&`$W5cLtd>5>5KNI2x)n
z92AVb!uvn(&Glg7rMpH={f-~|p8BW3qz$n0Yd(JNj!{jmMuo$HkDHBZM7C(9n@MBf
zZKqEBo;WhSXtV8H`foKUbiPSP;AzHM^q~Ic_f0cu`Vpf>zjl|gBigewc_Cl8zl9tz
z=YY8|(-+>>g}2A>AWKpEOM|nX&(UUFkAb(T`^ey+anJJ}9(va#U-b0&-ORF}ZhCx-
zL3i+Wb5@}jwuO=XY-F0??S+r%TX-7|ZyWt|SL{FT`q|2$+{54~>|6wMbHQ1+HwM*z
zXwa|oaP*o%57ryBiHy_9dHfrU><4>yuQ90H0)r+^HfZEjgXXoMo;!;=ZAW*NxS4A4
zpYNur72MSb-j47>=Z3lIV=Xrw`N+TT{$r_Ra8t%-ZW>=6X2RRrv#F`W+rDsidt?0J
zGW0d^H@)0Jj@o6^u(3uphqnhKd48hFbtM_qcac$5;cZY0_%?_$$_0~>*U@)G26_S?
zpX1zAxHp~!nmF@uJc+z`5#6XGB-0b+PhT-}vr6C@^dAMM(1h9jnO|Ip-2mv?R`9kE
zx#9nO`cNL@fUEpIbmzFeW-UXzW}$Ylu&JkV{>}cSV$80FqdopLYaDv^(Gs(Ia_PN`
z*_!Qnt=Y`_Mx8l~pPj<@t%I@uJ+~jYj^PsJ;h~f^^g@kgUgk%hS9t4-FTT7pJx}3e
z*^iTFCJR=k2|Y&m%f~u;=?uK>%WJ#}y#4T*eAR1ulYen-W@iJub?3F87v{cc<)x5f
z%(jHLxf=WE@Tc_hsNt(%gP)?{ZLK?g3aLRK`Zj-^edw<`@b>w5f0cjfuR`!PJ-i(l
z5Wp<b01fIJpi7qmRDNe}dB4c5O4;-1u<~kSTs}31w~b6C)ni{trK?_EYgVwsI=P}8
zgDdM|!)jXDr#d4%E!thDA+u?k%DJee^7L-42@AWZRP9javGrk}@K7D<G*V4Bk5H$f
zBjku5q2;GXYE733Qq(k^s?Q9_V>8wN<4lb<&1Ki<T(yX@YE(F$ZWOuPcjSfM^D&&=
z_P0%Up4jy%CwZgMQS>EJCp{OXy6|>fDg5Ngb{+0$r#@p>5Vg;_RvYycs}A0zk7I^a
zo0?hm7T%V<5UI=iBW2kcsS|r5W#iu-p<k^BBdMuJGS?zftCvP<{oj$=emYX$PDLvC
zUZggEiPX@1R&^X@)gjnB_p4Rw;B8xYTNU2UzE0iuhg~LkTNK__PhpNd{&I~lGEnH&
zTs7>PRMxI+@HQQpr#JAn2^p+E*K8^dZ(DGGLuQl7>TA>LjyB~U#0B2w472ILB%6-o
zdk<cM4!_5Bdd?nXrb@%xZSeLa8J{V85_IV@*Z85^k`uHgB|%SNN`~_IyLil<;qBvS
zW{pro-Eq>X5%6}_YL^yFPt?@siF%d}w$huscT}>5(W5ni%hff>YWa-W%TMu{$^Oo2
z#SBH}gN-3C{21OA8xL#YZFcnK6y}ed!b^@mPOmeyfZ`+Bsq`2BHVaRgKAVXb*|F}A
z)?9$^90&7ez|QF~uq0V7cpCweTkv<}B)fEPf<vv)%9F}FHTAPYmH4@Oo#>P0`%SC(
zTP8WR2A0Q6bEqS{-8b4H-!}B|aPFuIZwru>E(rf;Ev9zM4%!{$r_<23y>2*kfF4^@
zkV8xV^Nzz3)R27oMr*8Md&DTsi@s{keFy1V`!7@9bC^}-W^pb%8=*U+$^XII=;f2u
zb^ausIWu0nr;JyzkK?%X)3JVCwe@~;UA|aPUOVf_*1w)&qU&n(jJirosH@0Tb@ZrZ
zZ7uj-OA9yG)VjDL)REW|5E(4{^&)z)r3f4^q6!}h>iNck`1$PE`NRz3%0U`fDoCZ^
z?LY8#vMW$56;>H|n~vJ8u`&OLGY6=`XFpXYm;C+vf4vI6$~8Z|+|Mu*9^Uqaw-*L6
z6LviNNyE{z@U|Dcy)oNIbG?039^Q61#=h<v-g1&Nx#q{*h;{UsP|y29jWg2{v>`K+
zewOx7`^C(AfVc19ZJy!urCLmSUDTx3Xw{~LOp1fIp<zZ<g13=y^)Gl^936V_kx@(6
zph<fhl>o=mc;9~y2ll|*-e^@%H)_6l@Ge_YR}3-zU(c)UHk!C6o+mo`LmyaD6Mr)!
zEIvo3Xp%wQAGm8`xVwH0bJxqU?&`1sUHQeJ&h*UA;6B#tXVwTbtS`LndyXvC4YL}$
zs9Ekb=>T5w5wz>sW=8e7jPLr=q+ILCRW~LdOfSv~JHM+GEGukOjY6Cq%tm!vYtY^a
z<fPgfRQ`v%-Uq?Pqi(vi-c604yQ$6(H*GHCuD?UwmGl>ybh5jK^)<+YpS$=FFZj8;
z_QV(z2WN|-iT#t&r^TqnRx#)?-gIb)yK25mwcNgvYSD!>i+#x-OQ9BjER}ntS{_VG
zwY1%qYN<Rc&2n|;AB$THH(jQN`@O7Df5Y1saPc;eQRzVHw~goxE=~RU4UZ*{Z+^}v
zhPp<%!P|**jq-uF!JKurE#RE9j9ixyZyAr^&H%iGX(s(#Wm3ucoNua_G|<n)tTm&$
z(Ek)ujpvTvb&7iLM{4NHhj?i8J#uQLsh9FvIe&!yq`d6H3nFWWUwmjfGo)9t*DyW(
z%H?>iz}sOkbI)r0>jmg)YPFsN%^IGbGut!j<F|O6i<y{(Y1U_I56h|FPlC6b$;35}
z<6N``?F)AgPdBRsyd90k9Sm;|b6+x{xzm#9hdR%!>L5>z#CI;-%~Mxy^E!Uw$ytD&
zpt|hCXw2M7KK^e8=eS$+W$!|tZ(w%?db<&?`4+I&2hV#D9{4D-ce`4$N5hBA;0N|F
zeWxy+p1nvF_&>w>HJ#F{<*oE8k<nMrm-s8=f4rUMuNm;xzbA9oD)?(uoS)Kh9u9%G
z?e_X>^EqZeS~Ba&scicBD3@}N&ZSOW3#h~40&>9H;(JT#F?$SZtgS$OuOhpvDzS&W
zicCp0wDV96y?jwqt7f;<uD-2QcSLI?!`qmKp*nwWu<F9wu#LlHPd7qo4MvcM8-d3*
zQqk~sh<%#6z}wmIw)BUYnroP=Y4^g(P)BHNgGfF23k{mbsxe2csymcjE9Yz~4R43<
zwKG#AN+s~2yPYMsi=UjXP_**E+dOsc`r~I;_g7?^#>2t=Rz=K4*N(Aj7`&bKEmD6i
zi_rez5lX5Sp_tkcY7B4l-i*{>_*)|wtqX5IaXA=DeR*7@TB3XN!P}Yec1>=pR<*IJ
z&IGGEUA1bKmrZpV+O%<sUAwMPSG{c44~JcKsI@+69;L0R<gMVN>3gIq?64|7mjhfr
zaH)wWoq3;C-yfo-bJ?^69qasyzq5x;W#H}BIIG?jx2k`0tDeC3CGgfc*CyXPoF&l3
z$<%RE@RCOsB}Wy(-ql?RS^=;8*O0Ac-tVh$c*5rmGCEXusYAIQQ6EJ=)~M@L8MwQ-
zIQ31u+8l++{vM@<%Pfiy^_Z_Xl$`<nnQ7cBSyf?f#oEjzE<oSy-6V2TFxQ+UE4;l$
z&Z;^&;C6B-Ye9!D(^Hd&oK!yi>wBx1`2`0D4TF6@6O?!>L6`rK_q~noO(G`+Hy3c}
z&t-Elr@GMpdUqn4Gm?4Lj}pigIaCPVe)Yuv<$mv-?$G-8{M$6RnaQcvThZ=`4))PF
z)S)q+aw&)IdpUIDP=d<cV=r-K>az(>MZ(*y@VRG%QzeHu^_<`J4ZYpBA~o8rak^MG
zPTAmXLuwT>;cX@Qw_kN6L+{1e=%ZE7!mR9!id36sk%})Hp}{_L*%#kOGlut8-Zwqi
z-O){9v)XC=mzMg|ys54asiUc3we`>2nhL5}QzgbryQkIFzsG8-=9ZdzSh}_f9;m4-
zgC%cx8&jgNF7z&>LV<;p_@$suEi0(HZuzyaSvGAblvO=H2I*VHAQeL28ct==8hATw
zSfFA*;w3K)(0OBkzR&U3`@(*Dk=0LIPx-3XX<xG6?8sh|Uhhu%C<UfQb%3h_n9(?n
zUZ5#_KAqghR38nS!TZL(bo!f_ikHc_^oO@SCy+mdw;7jtsu1s?ui@=7GNJ{@BaIGZ
z=03WyOk4U;O}zh|MSnIi=?R>SC}C3B9Qcoi=slgmdou6K#bb?Xe39BAtlf8yd@On2
zH|VU69{7{+wjUa6-71)To(%6FqgL^I2Ep4J@HSnHNvCh1S@EQgpkWUs8kFX3P_HZQ
z^nJUl^hkH*UF)vgecknVkh@xsaaYm7)HBhhRp9M@c)Jcg_VuPw*{`9`xcxHZh01VS
zH(|#NvPK=@?Wd;bwvtBO+Dqmbjp<PrJz5|3wITNuY}Cd4WRmFpdCYzOu#UPbynWi$
zpaV;&v3k1e^<g&^hPR`xrCAEEPqUN=ano~n`-b~-D%f3JyK~|Gj@SunFS^s8fKPnS
zUE7|ys}H_#dVexh^#9DwXi%Hl?rQiZ)iQN|s%0!1_rsl3OL$VMWqOBH%hjo=mdu+{
zEyZT0SzJ^9SaSAu)4}g(ZTg8!qd2e3F)H_Ja$VkN!dmnymqvphC(FfSUOkk*m9xX#
z$wm#G3)k#Md2lAW%g=O!x256jf{yeYlcOpFTPwrc)5#{4;|!G^-X6qH?uj2!4z@3a
z+5UL#OR4v+=t(Uf|KMUNPqo9loqdL$=Tvkdx#KaHnV)I()Y6&E_JFsiO7R+5hKA2)
z&^68$r||rLb0#Rn^E=kZq}9)i%yToclZ<{TzV^JgSr<}Ga@8ekhaVkHwyP7I_J_OS
z@b=snYP~-=_dQ1U!rXE#$zsoEM%XEGTpv8t7+*Qui_Eegb4ZHOXOzI)?A*+YM5`Xg
zTkiIp`+MC}Y0TPMzKHo;iS&=H^i&Pb2yL!1Ydeh`SS9uhHzTKQV%FsqascGB>XKD{
zT9*Gq{C}#^0j>K_diiBQ?@mpx4{&)E-gMT#{58CRzq-KN1o`3DqJgXWY4{d@{Y4Mo
z=hB%qxN2s080Ao<A-PnrUIDqo+b-~S=aiBP?@&t5dY96)F{O0gUP|}gl+uqU)z!>e
zLsMSXRLQso+MTPBLNj-isamLROkuvG&k*_K9j2AjhRZk02zq@-FdJ}$YAqR|(gQ|n
z0v_Dj*zua(Z5n+6Gu6s{uKcct>)x4gcE5!)cbGi0cZB{b8KGsDBh=(IKI%=ozV4#u
z)Hg~w`r@TJ?YanW7Z#xA+rXw0<eQG?wsQussTRCF56>dut>A41cpG#Bb{>k*>%kF9
z2XCKbiO|)25gNKDLglwa%2_8;-^19ORytCf`ol<g+h|9m3Y?78Xn1=7-WGzlFUQ00
zTUL#Pw@d5U6n)F4T$PwnQJ+2?c<Xpfruif}tD$(le68r+NbU9~(@g!h3aov{B@?{8
zx7VuT53J1Ax5=wFo-rPB<xVyo>%b)z))ufT6y9DWr*#SzS4Z!z+h$WScwmONFACaK
z60JNMw*I{%LEZN!sKRPABD~!?H$k8He6c^Qg}2#ZP0l~;KQH6d2ejXnB4l~V`d+6W
z=is44B@|84>wL^sWX9r_lF96`Wu`Hi<*S*Q!?-_Dx$TMcj<FNAHodXcxjySsYqWB7
zCWnSTiC2eq37QQDU*ca!qGxwsNzn4M3A%ZSKGvJ;Lw?Q<cC>HCOXysBy*4jM(D7r`
zn_=_}v~R}{bZQ55Id$lD=hzR>41U7gSunZ<ynV!ZqSVv`_PRJVh}&sHZF;r0LnY~#
zJyXP?Ge&+lzq2{#ji=o1kz9P;!aR!={JnGWx(7RT{sFV2x^tV~;OtWR%$OziA8*UR
z+v_1Qnnr#-rX4#F(`@SS&Z<|`A&gHV)$<dq-5aj6L&Mo=7$(EODQb76k4C=fpud{7
zQ{>N<I`Vf5Rj5-}!Eb8m)T^4x25&>hNk6*RRlw1j+Pa~pUIy1x`+a=eUYfeBh;HU9
zqBiihQJws1vL>HCjLyd+lTSA?vol~{c4fO5q-M{v=<mZBRMkn{mh7YbPJqsAVdg(s
zsScCKXq5I>o+r$?nc=JQ+kLh25WBmZ`>I7!dNP3Qft}^Uz8N3Y>&u?B(dgQV^cQwz
zM`<o}Zy<ZwcQ7w%a5|lBN55@4Z{55~zoj?+S+XZJbr1Dxji#OYza2MUm<h2RT~Ywv
z9x|Hs`frokH#V_j7(QM%>H+M0KZKsr>C_0}Z373|@IUPeZ_lA!Ca#0Iyl;EHH)sdf
zDd?+3{LF9|n2b;CM~>FTwY`x^-tabdE;<)4`fhDBsLcRN@M-Ch$qT>Mk3dH*byt<1
z?kZl;T?4nfspeESh1PXb>1xbC<hEwQ*|F&Dj`-El+(y<eFdkp{NO9PXHmp#G?`c9U
zn9pm#f+zJ1npFoL;1?UJqa6zw^)?fH3HUFU<gRop(9_G@RV%Bz@?~*Xl$rYNE;oH?
z{>S3IF3sY-E7kHNI@RKcNVOD9O|{%b`?lx)?TK;MmmPfn1$Q~`xa%Rj9R_bpeRNmJ
z6YjFCaaXkjca=pir~I30iQEcf52RYMPfoSu8<}ePQ!3Rm0p8{;mTD<CBhAwFe43?n
zy)?^mZZAGL)e@93&9dumnkD<dKb9)?KbE3D|5%o!qo%xz$EzE)^7i}<L3qv+sU7z>
zDBmc98uEC)dtlJ-CG;4-ByUY_x-L5WDEX};WVb%=#OK~fFBB|%f1h(vgpqk;%<w2-
z)EAh$Z6S{hU+=b*$CLBRx9sdptVlo6Q)WJq1FlYn=@^-3D}0QBw^1RSd6pWKF_S@u
z;O!A~f5u)0W$kBB3f@Llczcf5P<4|@LxN4(hJW(t8TuUFHtK2CizsG)tz<?Qx_9DK
zddu+5yTbD3m&webncGz6HAzk5LLBw@_tflfF~{U6`uQF`WN6{A-JC(l+r3Vq2F#4j
zn4|ER414%c`mz_WZ(=g}s^;ujq-Xp3Ir@s|Suz;OR+aPmzuN8_&JT;Wp_gCdm4tfe
z!&de>ar-r>m7BNFdxa<d5xyTq2Y;ZJUHo=BosRd_@G5@Fk>5`l;cfJ0e>LnApa-J@
zWT&@aY<PgaR0&kLD~sk^gVd)_3FRIctbpFZ>hmaAQ$|tCg|`o1)3-3cw3e5yp}07U
zDmQOPt)Qi9WbUFvUk5U;YLIS33{so(Lv*d(Fa=H=u16(D$P3;cT0BBCZ;a5tqetob
znDL6~G)>bh(!)4-miAwsqt@@|D%;O6C9e!uYp)2MEFPhaS0eN-K2oiESe4sAP8WtY
z8*S4Z{MGNv@LyNhG_Jl)F0#Z`$@MOJY174VHnn?e)k&BeG~BA^t<a@qtju$cWLIc}
zGE|9>^;@`(!rMJF!}TS9xNbPZ<P{O71TH1Jg=v51Fx5|t)P)m~YX2@$K{>5j#P7K>
z%BtnCGBeB^i}uWW+omT$Xj>2FJ+;QCE*q&;g(K-(h?M<Jg!17_w|o$(l;4rM6kyev
zx>o%?$*T9Otnxc&RlNY4Zlh^`wYF((JDXzQZ9RCq@hp4Z#^PaLwBiTZ*oTYug|`o$
z+En3Rd}xzh5Ac=8%t%n>tqJm-lOT^G3EG|k1~pH>OG?lM_S~$lh_5?|+UZa7xTTpx
z0dGHkqfcl8J92)Idt#o`(`Sj=Ltoa9Pt0@jqfUD}iN56|?PbQq;bty%j!aY&cv~Od
zwycmSU*>BRhm#TT_RiyYZBl}&oQ_wi;^^wW3Chp>;pTAf!3E9$_}sPN<ze`G?*e^6
z2OL@m8>`HT#}AHI?Sb*yx}IIhwduuanxILx1i8c8Svwut0b6%ZA^TMuEjx+b0b>$0
zVkCbb-*b|`?b2sFbZXb;B?%gUkA52;JrDeSGL1f;a@1?_uifU5i-xxc;BED44t+{x
zc2wmAc7MgohunJ;cpKW5c_w7XqhrXtbB_7PV55!|sqvlvd)^{cqsL4|)WeHOnXZ?=
zr)fvrKx#x?^pifgqHe9#DRX_5T~<dPYieog)0(n(si~$Tq~ez?>RHXA+wH4r<cKQT
z7hZ;*^J1!ax2U#dE25+(g*3%fKt4<Islbk0>_*F}qS<rm`@tM))RoyZ#WE}I-we7G
z9VoWVEA%9@(WeD4dm})F$t0W0_><@L(}AVz#75V?nd++vZPKeRS*edX$VzpGt!<ba
z)$M=##uVbCf4Z>O4&Fw?+e^#=@?OdOfl|yd_{`jc)AU4emOqx{DK|59uXfCW%0xei
z8<`aR*dx4OHEG6sXd&h|jQyXV9GnwR8Lk;%+|S-dCBfQZli?)q^Nzgm)6K}9A%j|;
zB!_zdPxggD+x|t@k>Pz$ZS+!S-t*C2-9M50d57<b&f3QNyb0cZZ-yr2XIjrNXrPxt
z4wzLKezn0v-qOq+UE{6~h28aio0|$uaMOUAZi*@ErlaNDG^mW5c4y<?sCAyo;HE5n
z-PGd@e(+^-N-)1@DD~4pWPPcteuM=J@pVnN+_j<vb=Lg!#g?b%2A$Xh4f-42hMjd+
z<IXS=-X>&r*VB9aeutYDH2GuM9`W0<ztlI&feha)6<7bTv?=)0GB)wErK{nK<$})_
zOK6cV7XK1oEKd)9ww%cS#qxKrua>`ZezjyO|J5?R?KjJS&p$1nY`-mKPo`RuR;OB4
z4otOV`T5(@(F}KkQ!NdAQ!Tr5r&<Q={cUj^|81H1`nRQ0T&m@kBh}IYpSi)=G)uq!
ze=H_=yXWB_%TvxE%Z_s&s~WYF4A&k5eY#`eFPz&n68(;^J?$>fM;c!v=M~Lk97D!B
zZJa@W5)8Vs+n`Tke?mp0w){5e3(wi+WP|b)g}YOY8VPTYZr~iW*r<iD_)K*?=gd6k
z+2AlgGoG_d)5c`GTf*56%mcwo_-6!}Y@WmJ@U}Gfqi<J(R-HEJ1;4+0Y5b6r)Z{sD
z#g;IsVqd(DDP*oEGCO1?JU>rfZ3(lj;ceVRv(}fwm&T*7^~hbGT+fHMTbJXX^rYvx
zuu)gxdn4-lMHc?2oyn5@!7Im`wvV7@P=Z=MIXO?}nRVk*gIsgxZ1g3O&(7V6Y&-Sd
znxELi^o%~<2jsiR=$E}jkMBrW`Y-)(@U{T@J`LU;ZA~p2-F+h?e5VII?WTv)PxI3F
z2z=*i={4esuTJLm)0c97$^>s$hXycDGeAelUUUo#P;YqKa$Xj8`eae|q$~=|n_c!*
z*>$W^4i!$xqQjfA=;hQPc2Aeky*3t=TwPbo{%)e*+1t?X)>Us)Lv<!wA32WqRZwU@
zjexg<W(~)y8lfA-MyiWrl+H~bt*ULtvvX;(E+3q(d)a5<zs%C&tFx4Q-CX8uhN;`4
zaBX&jvjroR`6kRth@=OMeWSgs+Fgw6yAjNV!t-@Rs)e^zi|g35xE3|rGO(+pP0vce
zEZ#GZk}po94!0Ig=2^w;jmeQ(**#LH??kBfi*PlL4%dpMVR}3|OeY3}>C>Dr)odT8
zTOGpmn@hD;VameA?Pa7kl(gzTdUJFLoaBByo=lDk%~}lZmb_!tka9Ln{(mH$1yq!K
z*TrXO0YwZ{3<Lzl?m*AB6FadGyAuTi5R{prV;IC%Y`u1OcXxNU3=LD?e%H6|dKO$9
z29WXj{m(x8>_gtJY#0x(k~pY;#j)0o!?B9oH{LNGx5(Ra<n1HS+q{sq(+<{9w|OS0
zkO&QVTTI?I=aoo}`gKYI_R-JvXh8yolD8xHvwdg%JJmN4(I45zWes1Iy7^A_fMZ-T
zP-$-l21SuQrO2Z))R%i?Aaj2P4(Q2SSJt)UR?`Bq%a^>Rj_X)V{}Fk6#=v=34R-^)
zSjAai<}&51!gu@CQ14m|7v>MQW~Sm#@0F;Qz^vgZ)T6s_j@LaCQE%x9COaRHx2_L$
z*eDqowM)l*KOH79cX%CnuV&rdiQHAQ#;tdRI`3{B=51o0HRoDWN9f?TfN#lLNAlK(
zyseh4!(+P)yqlMS=bX{;&s>2{2k5(^A8L2|6+AO#pwyfU^dfIhlDCa>SkwRI&(6o@
zYRT%b47jbxz}LDfxR-)HEAsXf{bAh#*h^+T+n5>NTgRs3O*!sZWVYvYFD>&3Qn1x8
z8HZ-k({?fmH@eg7rlog%GQ9$u;$hdHuQl?v8F_0*-nOa}h4lXO@Y8cN#<v-XWog6l
z(%c`J<+`E_d7HDX19Yd`V|&l`*a0B=U_11szrWX}#>}s5$c*a-_(R@4eN>0_nlGH-
zgDQG2*j=s3J@3`vJ+>MY^{SzgpBu6|S490X6*2Z(Ir_X^`Koc{EF|}l@O*s7$oU^<
z=3MM6jf?b5j=seFs5j0qjdA8a5BieH+vnu%Ue>ZV`a0p{aMrcK+($s}9_r0J*k(?M
z(mCQPvm2B5GYh_MDQ4u+lS$s5CvV4-x97;)pZ)F7k-Yt6A#2aFUq;^kq|TRoz=oOx
zGmTm@D~UDeHR{F>s3RZf!a9ZbWBn3pn**#7*p&4pbx)g@oO`*(tcYV)X!4QlByTH|
zw^e&s<JVwnl`}beGnaK}OZF7WfEncB5Y~PlFDWtYpb{ReD{DouC%S=JZHf|`W+*X&
z^DzVaDlw!EbzSmyD|uRxjGai{{ySljK6}{z+-{MG9E-FlPd1KHAohR)L)mA&(8daN
z$5~-L>nn2)&cU|l%xe?+o(3vWdJ*f_>-;(mt<Zqn-SU|_Z@2=XUo28-xJgvi4dPP1
zM2w}1C30AithrMluA>U%-n@LNQ6pdGoXrz$N}fc8<w>P)d2&gaF9Uq@#nC5U`jNNe
zOXbV1;CzWYoG;E53gme~ffTxy$mMlL*)+l=xfM*Z2L^HeW{?g&jq=ytBr}~&Vq4K9
zVW*6uA#W3Z8D(awNtQ;NWPX%M>YXsl+({OBo@kL7cPz3r!y;z-s-AY@*`_5g{ypY4
zU~hT?SxVkMAaD1-;QM4<1M+s>S^C4s+nE39@jA`~vo@~c#r!O9*0j_De8}6^^m@(X
z8KqhcD}GFgYpl;N_2D_?F13PRWVpc!o64{!PTqDTZ-<e$&&k^g<n0Tdsa*Y;1wx<M
zW6nF-x?5pzPktWHKc4*DU3?uFlIfviPv3VCebVG@Xc*_r$XjdjJaCj1_MK5;Ci@3P
zqo@T?6PWgd`UZI$!an{9>h+mC8(rqtv8OIkz}MI+a@V~Bb(1yp?NJ{eK9D_M^0q#C
z+qW)vPPy2jyaRK(!|DC9x2G+F`6vT8gM68_I{nJt<n5Wa%>1Cfe1WsV&Xw(P{4V{`
zm3Rid$@-T3_w8hhsOFs6X8rBSev8LBdbhVaVAo`3egrt9=w)eU@RdRJDrI2Jv+$vv
z?D-UwW>ymYRb9)%+fbHs<*wXUT>-(X++a`M29UQgO{$W~ZWy<dyL4Z9;!ICJ*o<wB
ztku1+{77He1`oiSW`j}s+z31vJQAj+!C;_0_V=8H(d6wknS)kM=aQWZ&~{5Gjtp6d
zJ#Q9c^oU5Tc*zX-(B*i2Q_VhDELP8pMc|fL_7r20?i7cDcJzi$i-YPxJbQ|9u%8fz
zlbk8?s2h)75|1oiE!)RqeS6lq<n1!@c1ztvn8@1=PZBtzoq%zJsN3#gy|^?Uvj@i`
zfV{pkDi&oM#zIZro}R10+esSksL`NF8x5}4(%_7z2Im7bc+pQo%`+ZfYA2uvc^gFD
z+L5>G$lFNzmA_ISzCnHY#$9qWBmw?5^l1HygYrilI(&=64TpGi^NB}KQ#`g*pf{^E
zKZo`0k?jfid5?8BeOCpnX)lwtb$KO_w^`(I*E6iG=%;S-F&@^e(J!+0ec+Rb9qkg~
zwk{DfwsZdyXN0QckyBrEaQ#JH*D3>Djr81>=+I&vXM{d8H;HxNCvvNzeg!s_<E(5D
z`=aFS!+9&QiuJ8(?kY@p%FGM$c0YOB=hG_Y<F4Yr8_xY)<*cszN}M8Z$C0-_<n0Ff
zg#+kmTKp>mOPA9dO^<JOmJTmtbod#r!`1*DGDhpLJ5+}W8~D9k(qa8w9d{DxkU-u}
zCT}Nr>gZ=qM;`Ybp8lGSsylTU{!fSY<n2*1_1o19gwbzxdH)LZX1&^<J?h=PIrqdm
zdKGo&czT^a@hs8#PX>Zn-$t{p-O1Xw*$f>DPo$$4v$Ol9GjqGV7EVQ4W(20e=20rz
zjZH;(%T&y?ONAe2v$uF9vj38Z_D-C6o<YWTOTgXBaVT{tjyq?T;aQDPtlGE$4kH#I
z;^}<U3!aDh+~0H4iTmH*jKa9g;ZUUY$EInWQM9cSj2}ASAN#XME(sirZDCc?1}A>C
zL=UINNQ`dCoazQ}pXG-UZgtS?whzust%VYk7x$^xfckkgxHEqy`c7qBUss8JIX9es
zR~|=um4gd?$$|XoCtfNGT@3&I7CWO;Q)dKzal+xB&giw5|NoNLJ11oPuV>5N?3b$l
z+xO#$kadpuPEGkSc{`50{nL$m3H~^828CJk9hiSe@8*nu>}S2CmZs%A<tpxVYtCJ6
zWURA~9bSJSW4BVb<IKoq2iCQpnRh|n_77yH6nT3!o%P2RYs^WpMgs3|)&5!GEqObM
z?A+^Vjm}SbAAHWb>8cgnI1A&SWQC(;ct0=A`BT<9<r`b$%XDj0-^H4jJ=7k&cY2X`
z{^ad)a$>+%_Cm?omyzV}9QH%I(N|P~wdOqqUUwxIrzzkvLV?EQt*syH*jE<W(u2%#
zGl|_Cqhzf!$iyxsQr4$LOp{8)?^20;8)T5LPYv?pt3i^^nWg4`7SXn&2Z=Rv3VW}^
z&nt0J!TxAP&ig!9;>k%R%9$)un!L@bRw6gr7fTL#+hAys__r+-3@;GP6!O$MU;d`&
z2~>G<Bp^?=_~pr?-g(ljU!F9clqc^G<w<f$p7iOJF9Y}G%g%BIVqLyami#M}X_3WJ
zr$>pbc}MoPGsu)n2C-xqrN4sxTL+WeDQA+ZON{a~(<p^sjiQS;$!Yqi-YhZ6q3UMI
znq-zwp%!_t(<0$2i`?d!WA-PDeEq?<7s*FI)|Ko_Zz6BA!dcsslP#}X;oB9m^pX`8
zt!0)5YtX78|6kJ|c%{TB>fiHLD-rfniC!AcG!15-n>A`GSv&We5;Lfimrmegb}Lap
zJ-z{d&%^oua}G0yl9w%KrgpJ!9sXF!IZP|u<LBJw`v<5~wC3|{yGY$S&x$n?UngXx
znS33bM)vaBzM4LFp26bC|Kog}MDlB_9;-xio{5sEGd!=aL>$jt71H^;;cKS*5x#Do
zDY1d)rglsCGelT1&z4_<{J*q-XFvnFUCRcas9C4eqvbln24UNot3qE{HNLhh4d<?9
z1@~vr_m%w19(z78Bjq*sO5LaDY6g3{T6&<H+Mx$|tBSV8rZ%>C-^><u$lD$4FOUDo
zT+%(vDH8`g=3L$S1ny=Y?uZp-%fPQv8SG!k{N*jB5%P&OFL}Etw={g0amNAsMKR><
zu#FX9x0*YJepG-<W<}H{Z<nuQ_EdBo)W6yY3CY~W85V#awfmtAd3&YtVEj}M#-@Tm
zENeX$p*tpHy3cF`{G5%<fvj)q&&8nE3-D%KC<@+(p~bev=nxnQ4S72)Y&oKDsS(~c
z7Ao@AlXd3}_Hb2>)O*R>23_MYoLBb%_Fu=vVcT7P{7M|EpQmq>*NwLE+^5VP;PvA1
zymmZl@!CX9cLT5YzxeS130O;AIB;Yf_xZ+Q=R*w~N5s+>9*fZ#8f16YAg&YZYF>q{
zHK<75jwWvds#0I(+qi3LOn;<?O}-j#{C{muEVCry(3ifZdq=4+zo7T&Py$ZOWrihr
zJD02-!`im@XL8*(o_o>aQCyWZ`a9N4ZVA}CC;^wZCE(E21URxbe$$Jb?M*Kfd%5-K
z)f&yZY8THDA>?iM5Am2XkvSWC$mh2S2q16kC$lEroQRR96Y*$ZI*y;#VcJ_A-YGNC
zyHJOz?{yeK-s+k4_<*%<zXRlA<rM(uV@Hv<?#wrI4rC6~+LhQI!d*OfSjUpH5ud3S
zUuRyTidjzq^e=O!W&^#%K?5`Kp7r92D)bnCVBd833jFtjJK2|!xnVj)lB-{+&#E4$
zW3n^%>ac$GByZcS;~9Xw9mQ*FEAq6N4)&aD4X&v}%%^nRyq1njU(zvmmJT(q>JVtl
zJ+na>I78lcVo!Hb{}ous`glC~8t+dZSUcA3Z8LB>T?c)d4zFT$h~QO+oWIMf<8&Ri
z=V_Ts!fT@zx;k2%vt~Bg!!%~(rQugw=8J7jM$;jQ7*H-2Um`TjrO=?%{1}{$=8m82
zaI{l}<9GOCo{bh_h<hl8B`?5^zy;iCHXj4Y+m!fVXxjzj%+4`*m)#$yXZAy65cdl>
zcS4~BNErw0z9`rk*A~^K4LYftW2SRs=vd$W$gPi$m+Rt5tvc|(=Y!-)^hsHJV~n{5
z9@$q%8?8HkAC>XBWhHu%+|X!Qd7Nlj9+fw{q8oV|ALjyR+cL0eTN-;CIdewU2}^yP
z(Jj%L`&GD~^DTW*iPVnSt1UEg=fH99`;K))l_A{eIhcD-c-8AnuT(MT8?@Z{L*BL{
zZ-4!<$B-B7Wv#HsUe0-pBX2W}?7!7wmj5gE%6NZiIGI^Q<n4L#cKSXW_<PgWn?ygd
zJ!d=rqmPI><UA$!%8|EmpO~{x&VFJoH2J#{E2gk#7ssAx3TIe(-)=zORv>SS9N9}O
z#e2Pn71qTnakH%wv#%%+#Cg+4R!VfP!}}}msegG-)rYZ;rH<U6ylv>p`|Mo>X17-&
zXo><ehbz#)mHuFB1-`x{7uTC*R~56=@is`KG6pF_-Y)Shk&~lJ<ox{-S(#yw@=pvh
zhP-wCXOug`Oyasy!8tPKOoS-0=ByH@dH?RPj{Rct_9A(ElYP+_<n5}3B{GAoZPc|$
zhM37(6pGi5e5uwkU(V|D#9?8c3~rw%9~<ULjzgZ5t&k^0ee<LRd3*C<o|L<gCl)gI
zeOkVpvM-REibDDFu~34di)DKE5-I((M2?WRx398}CU2)xS8i!C%KRfn84zZai7Sj!
z@tINd2BTcAX%g)ulkBT%mi9x<GA+w2H(BG}N;FG+q*<O`Gs}S&7Wv1TA2%&B<R52=
zW+-r=4(t2tWD<4cWvp4RP)B~j`ZaZi6*>;Eg1E4c&A)@{WM(F7%7)qGWaR(r!Ri7f
z8teIcAUoTWw{PQ=@cWOv-O2v8R*7!pt#^n;0&6Rn^{GJoaRut}?{t46&re_Z=lp&4
z;qS4sy%kzix5CR@>K}X!rPSbSXA$=|khKN8YVaziE*#9)%yja0&rp6Jeb^7Drs2|^
zwKaL$u%{BQ$>FrUN>~oiGxt=9(6XFOcCte82x=tcZ6@pNJz@0i>8vs43H9~gobNI*
z@0#AcDO=bt)>CWdjIZNh&H!^}OudIW;ZM0!nExJpc?Nag$DEX|^e?AUOFqZ!&|%Cd
zZBO4cS=-NtJ>;hBuaLL5LpbBq)eg(}`5Pv1U&k1Fq79r$?m*4yi!-vimu5ZybGD+L
zd0#D!$_`~v_85JR4_r`qhuZSza`<+$JYx1&z_Y}PoI7(v!7@+yUh%}HpgQPrsR067
zu{ZUw8&<UK!}){0m{QaiPw(}^jT8Oxq1;GxEH{h!L9=k39%YB(IXo-P#qFR_bfsU)
z(|;Cc#Ul_lZW$`9Wv{e3^QXvLck;G8dE1P<9rPs@+x4+H^uJoSOB^osVr@H+`Bc~9
zF!p2|#-EGBd2-MDP8?cL$L&MjJ|b(~c+Dbfr}279-fF&)S@a`6UK~sRZ7dSkY4D6}
z9llkKGiTMf?ykWw{=ccC26b59PH@xU99i3v*D+pAuB!3qp&I-1)TmuUg9?9H7hjLV
ziL?X+M=}c|i}fdYyPdV`R5EtRmpHU0Z}Ya(3+)pNy$7=_3t~~DDu0IPI25y9pP9+?
z!#e8ao8y?f#h-^7bQiaHH0T+R4&?2M7V&spmfydb{lf|LO>Lo$y)^*?M<yV76gBc0
z30T5j@|a$n6+NQE!UyD<A_D<AWZNknlBvJ8We#w}9{PiKa6Z?>9R9!+I85JfhkKcb
z8MYGN*O95rYT9s(eyB2>6Z)5l#oWhUlUWv4?As1xW`rkw!AA$s8`Xq<-<K<Ja3c2!
zbIzz8{Ypov1Alep*PWTpxj8NBrliB<z>ob-$M);#IC_tHoZmQK<FCVf^0pIs`>Zj!
zT3rWs^0qH|dysEi%+ygA)-gvqgS{8_VW|;69L4&UylqQ=^KU=)beprDHR^D*ua5pj
z*3c8k_vJdAUaZ5i`8v$tb2KF98*EEwmM_0%Q5w8$(~$i;1*4o(@Q6OU!YlOL(aY}s
zJPx)YYLsiKM##u$*oQ`8enA9+wnrd!V>p`V!trTA81nS<5x8Ihlr0u;XL1Pk`_6-{
z^CUQY8IO>~L$GvI00z_xK={}0xbldesLAcn=&wHpbn=J$j#lu0-yE0nnqgo7J^pPP
zU_x4b4E<UcmUVTg1^d8U+6zuDo)|~op4nRsi;q-6nVy^l3aErq_1$1wQUMc(RzQi-
z6|FkClEW_OXUAL%e`dInt#!9D$Ko0@8xx%2GS(Rh&zx|3D(hEYN6fA4h$ZzM;W6Kl
zS-PCH9pHr6K~BhS?}U-$?XM_DbREqZ$%dTyAH&%h7v?<}Z1BAv{il6-|2SgHoFV4d
ze&-pU9+j`0-T2p<dH5gclOk`A*)v1AKE1<6EAHy#Tq$|0Sj>HLt*A?qw>@@S;TL&(
zk+s!0-Y=oD!fMvGZff4QZ?Ps)TH!BwYjLo`ZyV0BQXhQ9xtb638Ku8a;MYF|4%@Le
zSCh4F2YQo&s2$E!Vi1`;&YrV4#}sI$P#{ca5oc?Q_)|CTUDG0K$D5@%+9=z98N`Xa
z?dD>TPaR97Tye4deqAhWLW`xSCAl`ZM4B!tk-e!#v7Dh#X*+djKYE$wC^3xp@6WrG
zC|tw)_a?q=rbM*SB-ci>&a6@*^9zb3u1k?zzf~xQ*th+3DNn|>%$ET2HkUebr<Qq=
zQa?{#*yV{Gd3&%n_2p@K@^x>XB%RKa-7oXRaag{X?FwYcg#sxzvrvYtDwOt{3#G?~
z5_u^G8GP9wOZmN3Ei}qtqfzedHOiI=M){Otl#Wl0a)P`)>|v5KrA*S7b*^(?vwYn^
zpY&Xd%u6s!nK@<|+s-VyI%a8GOTnCCi~RT3BG;xXxED@|hveyF{{8<R#9c(>tz)7U
zrjSvmcPsFHAI~j^6&PJziHfW#Z~Rlz!_4zjwgMFtN}T0cXoymYK)!9lt2}wTZZ7|h
z_&e%J-gel|xuvldc^Ifb$$kZ@mE!L#jvv3E#5=NbEEyga&cADF-6hoa&rap{;mBTa
zN6umKwQ{Ex=cIhCxPO9rv?n!gHP1lg><6B;Qo8W#wkBIkv&O!zKwW-M@A>sz$=mV|
z$$Xv%6y$ASTPv*WM_pOZ=OJ%zQkV8vYmEl%1;^;E(Z4@^*qb;D#axxo<*4O!<gO}T
z>dY6J7s8od@7{KJVzR>r4QGd*Fh}DE>)YeD$fQTABj}GNXYW@admGa)rJ`SJHs`R&
z+choO%VmGqVJ!Du-Q!*d;cl)^&h+rwqfs_zkvN~%g#90b4fjf4EQ6!ZU6A0$%;mkV
zoR=*R)xio#UQ-d{!`-lI7<W~3F3_*g54D@tg*mS^yodYa{Y-y!@6nd~Y}zyPza#r1
zow2gpV7xO-K(!mQ@c!^@+&(`WE#J<@)bKC_2hO0sCIknWfl*hn9G*j#V{MTJD?7zv
z*5Fvi`p07N*I4L2$8rx$EG87kVi|ecfxJCIZr+ZLMXO_RIC`4=C2vQ4&|t?q4Wxjz
z>R%1!22&GeA9j6D_F^k)P<4|U53J*`JxYUuAPxSJtJ^Zw*uanfJ;gu!t;X{98Z5HX
zVDf7<?zw6ZPR>S_(x9J%2Jg?P@!wuGqH@$|#9B9|k_MLP8nk#4i!l)iShFet5&vS*
z-Z>75;q3R3w+~o@uX-2@heGPoTQvByT!T|doOv1*hn1X38baP4SsRB9S#h}hUmVUA
z@jO8NxvCA%7p}2bb0-!-<ZX_L`9|ykcb^fD)vx1GhrG@5OJE-%0b16|*Sk|cok-5H
zK8)1saQu%Be*4KT^45{O^<tgtwKW4t)Ry;9TRvKy-YI9Y@@@t`+N?lQ<O=3m&<AXr
ziB1<X@rXKe$a3n)gP2oH-j0aOM1S(O3B6Vq7qU-F?OHjI`fvjsHnWyJnN8NtNJk`j
zyJ?sfR>!pXQzo5WLhgsyosL$2m>)sTo^HwBEm`YYPlpcV?SN<8Eqf&$`P7wt*u$-&
z(V-XX^d9T^7}mQ^e7u&AuU=P&c5a+);(6jjln!nA`K#OMVEZl|&))KDKF}g#Ecf_K
z(_x&I4qo|M99*YG7y6()nU_-L33nlmO+hQ>P`^0B3@~rzQqvQVI8O~{@^%_|J25;8
z8&)nwyM>FPxD|#Cr$Vt=y8y?3%!kZgfOZGC^XJzTjE<jzT3sh#=GXC<P2T2@3BZQ`
z@pev6d|TZaqf$Df%g)w_S=$<gRT|^v??%`>q5(E{u7mQoeDNsL2ij(}Fs#HI*;i^p
zf6oK6lei1Mss|#z*1*}PmGSs?CDd75iSuc0Xi-=Jp93q<qveWk6FFDQ-fY>9&gk6O
z8GBw+FW&5gGn1Gz_rekPS3AP=s1z<;;><#LDI6NYT`j*#q4r!y%;CTLo!U~I(d2HF
z_pE8j+Y{vNud~dGy}*nF-b1qKDbf65&#WT-pEa3d&s+j)&T{r$ZiAoneg;<M{fU_s
zAMRUo{|@_eyl)lKull2eI`DeVq{eg3gZJ&*ca$h=P@v9dW+Jf$X}VDf9cz#UZ&@QI
zaHc293LkGOaV(j7B=6B}hA8<OXO_c91rqNoaQcD*mB`z9<gFcfYw{pxS;xK~&Dv%F
z>#8>7AwRyBz1Us(7WqBhBJ-{OUq3xv!y*Oq%~C+#9=908_G^jw4=s_}bxNdeQL)TA
zRxDQ+6iaHgV)<34Se&RY=c<gd(#s;l;}lp`MG4ge*3@J{g$?W%s@OO7CmUFA*CTHo
zW}BqLRQj4G7s>jdLYbAFFWblDOYMu)kQd}i_aE%pj?a_SW_i-xH&5Oy&Xos4b0sP{
zSK>U#<ZXEpdx{@_mM7u8^JUeRd|7#>Kpe?i<H|xAkyR)a=90Y;g;M`|p&ThGl9=EU
zNywreUe_S?Mj7Rw#wc-iMhWEmw$+TXTuI%0mPwR<O;RV#EN_D>Qoom3ns+x#b}O?S
z@imK!ok`B#Gsuu#Mp0TUqD<oMA)X`BR#{{!xfs;mBGIkc>)uH37wi0&<n4a)_GdLE
znr%>E#z+O6BNXtsr+^oEyO)feWl`X-S%Kx`u0Od~wX#LleKpI4gJyAHoqdUCpZjAJ
z%xYHPEFaf&sS;f}(dX8gI&cf}k$=b6qm{6GM6LP<Ii1U%K|Sgm<ZP6;6%PBb_P)b8
zWwR0v)HV+GVQ--;`v?5|`7O!YTm_C^Wj~%e$L{9Tt@kLIeaPNEYw<w__IEARmdRTk
zc{_|VQ!B_@Z}PV3UHYl$Ik->We$3>|5zmkLoLm0NGo+4n>ptde*xGVmRUbRf;gFlu
zl56bejBgri;A5N%W_^2`j}M|pe>CUfT9>s&bN+i-J&W28xqWN|^San0j;d~lu~+PH
z@s}Mks@r2x8242@u!kkx3FDVJW16xIx?C!QfR`?KU&$2#dtA}}n=1^y<*<vq-J4qt
zZb!Y*a)~b*obbil5?_1@u8nHrn_=}7e>euW#f}y2V3XGY-sJ7mu7lCaKL`()Nc^Pf
zXv`@+hS?2c+2@>!=}}YRLf(!WxCD>(Ey3%)OJU;{jmHJi2<y8HkFGHT!=4)Rjabaf
zCDX`T5AQhKdZ5OP+Vl}miN(OwSb9e_xb>O!<{S24AFvMHpn<zqgZpDOFb~vV7j<Iy
z(&Xk!HBPlv<GsHcO=hU^aj6=2GSmoOug39RYV;v*my)Fp7B%w?)VLd_#?~up)Gt)y
zojpBMWb$b8c8{Jl?r}AS@cl(GYOGzy`gk4r$T_ACe`0Z(OurwY!E-*Q+j%wYs7=S`
zYG6&xdKZ0HvCYZ(shlZceGPuE<;mf)2iPAztU=B(o&zrO%wgbT$=l!8V)0sHVXhO0
zb@_47&Z1`iA|9#qJBKod;rKB2v#ZdXCFyuSAf4GTWb6wa_XLnd?4|bL+{<~^y?2*p
zp!&uPEZmR*^ET!eF3vz?Oa{`j$j;yFd5+-REP2~4Jp(J5`4l>HC0wJZH!~Z#6?yx0
zAz4eGRXROO(Z$rNSwGJjqQl^ZI)vNlkaCGR*sZxkXT25)eYFUT*5aXmI>K4&J|S;y
z{-$$RH@#H#bufGAaIt`UD1N1*G<&zfr_!M!b8E8pbsnWdD{{8Q64uAvbeJOS_ma1l
z{B)>MLx=n0b$H6J=flT;;d9!)<X&Wc&3W9lIPWL9`i=XeUTfj@Ee*S_a1Z0^R9GUJ
zVX>K+DBMZ>-xp?lg|fb_mV^{){PxRZQT=%oYA)n%ANrs|J1@b^-ixrQd?>z{79jJ_
ze0=^oA6tB;z{Pb6YBryWc?sNQJ8eAFy+$(&U=SL44uqb(?LglCA#Xd9w<+YUhP-`3
z-X4C{2>W+5!qYQ#aU|83K42f5J@1W^SG|yN%nO}%dtfkkbR8mZL&@7iPb%Z-jY=4~
zuoAkjEswibUGR1*cYT#`z9+~T<;dI2PR>|a*BSF3vmd+GiF>P^5cts%&&b;m<n03P
zJ^8$tJJ34O1C_%}<ogatk7hP|YAM_#Z)=})K<C?>P29<zYz+6{*s-?#!k*SedV<MT
zA0y{e={Nqk%oe8Z^ds|QdF1T`@>c(YdH3XO!3FwQSZn0*{?~<XUr|T2p-%jkyj{aO
z^?Zo}R-eevCkm|ntH3bwb`f>iN$uE&Ys^f3^0wzn@|V1wN8TpBS77@?1rly?uIC{8
zv3<$g=470^0x!UM8}f9m7iVFd%%XNP%b|F)*tS-nKYNo`I0sXG8}(DpskZNKlE!6C
za(I<NdYvhej;s|2khdFsOJqo1u{_*cEN|u)3(6MDe_qA%QeP}jr9>+Jk7Mlr7Lel$
zhf?oc!g_735-%di64uk0pny_UEXC?#Ic_YH?g2$I*uGG%|I3$BUb#~COpauK$&oF{
z70*$*@}g;;Oe>o!)qCa2$G*9eos=u%$lFVs^JLoz`j#J&$vyJLSC=n04&_VrYx(l-
zRlc~_Ef8HWxl8_TI#ei?Ba39*xDwg3wL}(FG{}t(M)?{*-!gfdMJ;^IcY~<Q8)aG<
zqddQEkkWnz*<@{yV9pC|;#JVnEPZ^;a=*wZb2k`dv%5(K8!R%D-s;n2;!f)B*lL!v
z17@jC-hL#17kE>f=Q(Bo-`a7`st&WRJ_eF)OL-<DJAV&U&<~}+5o*-4SvM~$R-g;d
zNW=P=#MGIyO!RwQPd1B^x^`*y)i3|2Kqk*C9q2>5)Q+zio}K2^=Gmz->ubJt+@h59
zDf7(rTM2V*Y9r*WKY8m+-u~xGeVeuG0iN4#^k%k+FJCup=rgaU;7ka=CVg^#*A(c=
zI{YV}`{@GK(X7Q&ek!qw%$+xc8p=^Cc=^#cA4@M5d0YPuy#~MNq2la#J?3B@Cv!cc
zY_OD`1nUwTY^08(=FG2GX<MAlvqnS~U-vaQQ^cN%%^GSyoKfmW-qs**OXu<P%1{S#
zw}rN)E#{K9cZPB&$0c&RGHc??tar&<AM$qDEPIq3=57X_ah1$j{bVYQVDeV=$OUmZ
zW$}7FYuwMAL%75pV9TrFLI+PwSX2uKY<#iP%NGxPYvUDpTZ_D{`MeEm&$YrMM_}gL
z_INy;xv?t;;=q!@FjO0b=HzW_^0t52SoCf?5&3nd;?D6IcwIaXA1a05UBrAe%VZAi
znuWNyJ`7bI7h>JbMOZd<87gI|5gAV1Po+kA@^-eL1}?=Kd>lZ{xpORY0k~_osfHN=
z8q8Uw#+cb^sDjn(RjaXgfEp`%P%mz$MkrZ(oY#@*YCIutGgH-Y)u@p{-s+yPcl<|<
z?Vr^6&ieI4sG6C6YHVtzMjbyj<{VUG%QiK4!try1)KI6X;S!-{-Yqi_Qq(ZS@oUDC
z$!axPvu3g#N*=Q&?)gE*nuW8)^g1^pZzFt}C2de+e2E$tt;yBR8g$sNK^yY+BYFGd
z6l-zn)h+q6yzIsva+z3k=GU%pcqx`=FN4WxIb_&!JSY=`jrC)&ZCDHzJEU=+MH*b+
zr{T1p7CF0gC<)aeid_41jJ#z&@Rk(%mXk72aT0r@6EiTQYX<kHrlWv4hO=0ohBnv1
zZnzGyZFTUdqJvFCa*#eF<Y+O7Z!h#<jroMV(SKU{YPCqQ(IVA>voyRO*lKy+WFL05
z7F7bcm-s*`tk$Gr#F12FK2C+cQyLEOs?a$Ne$2zxhH3Fj!yOc7wK&S%vh~iTqZ4ac
zduqW)=y%#ntvHai^E@vd_W$5c#;-h2n9?zgJRV5i{<+K@#ud^r@P!uDwrKIb4H^F?
z4U1Rs=TA?=SkE+EKFdAJ<n4&1saWzk1<HykczA>PUkkX?&^`$p6X_8c6N^9BqoIn3
z0(_(J=6D2rt}Vpnt6}iZorm3-Q&45#WF*|2#J<i%`X?sfrN>zKB?ocdcNBNL4Z_5f
zfq3w}7kf**y7qvhT}OD=ZI9z~1e!9!EX;;@evtD%OFfaItBHr(JkZh31I@0~!1Sl?
zh)s3J&MNM>tEqy5SCvucdS&i`;*NrK6_JrrjysfGpxo_3Z>bB$QYWsKMZI{F6KWiA
zLgqs!3|;MnzW-Rij&Q`fBJL5rRSMc<&h|ub_h?66)UWh+m`fMnfb6rJZFpjbYb%*E
z&zko3Mq6wTVlD*dBO8*p(=RZ`Zz}t;^p(_l#Ch6T%u*<AjlQSpY2kga3j3?wdEcwf
z`|I}7+*?CkFNwUJ%$nt9ni5LZ6>4gRw^$#JV(sxC>(wsg?IY$6{%>xz^%W(yA7?h<
zLH6>lE1_LMk1>0-_FEO$$l9rGjDi`e3hsMlJvBptK=L-eiUJ{=W&O!(6<M3?V3v#h
z&5}Qu?;kRY;j~2#1Y2Y}@9#^8m?XQ4Np`Z=+0a}fw^<YVrIyHnA?(?fDUm+ki)GrL
zVyQN_SOzPLrS88XX*8)=Mv%8Zju|9stx2vrDp5RGiQ25Ag>$qg+MDD^zXI{wT_lHl
z6^f#Dp~M>tBu8H$C$kHrL2bTYAy+yc%MqLNIdbHCjwD)gByCr&G%J%U{d?q!OV?ca
z`XyJ|d*w+?R-Sx0OxE7Z6ZcN}GBG7zE*{L6isbEe^7e^efm}A`itCFUIetA)R;<XA
zezWtW1MB682f1=KD@W?J$dPxoawO|ij`T?7_x;8o<1ZQ{jPGaq8f2Zqz!SJxtjODW
z<n0qrvxMl4G9=C*CTh`Z4jZLfYqQMEG|K|=_V8x2blGo~mn|$}Ot;84_QBI~Epmb0
zvwu1THf8ev2PrUjEPog23Tz;6BbY6DyD`0KRr#3P7TIj0K-?XpT!=8rD}IfdQ_PY%
z*di189J{C+?4ni?M;4~AE_D*>ALQnGzDBB4X5CC5S|vLLUVWxUeT%P=8}zI3%yz0O
zz0d6PyHE$2=gl)*PuA4@JRxtl*X7(Zd-dZv|CDn`f$VDR`;)gBe9Sq1jrHue-*@94
zre*Yp{YMWvJ<D<AZ6$iJhF<61hOhM0cVZrN3^TjaZSa0Bb)6;jM3cW&)3|Sg{oRa9
z^d*qDMddjU&a-9^Gdn!TGA}fa`KAACP=-FM+`sItC|Da;p$<gfl*Y*CB6A<<s1a4*
z9)@f5SCO|*%G=|}OwQS{7knYd9*5mH+xyZHKZiKeYs{Rg2xmM#;*6BX%wru>1;?gU
z$9eL0%YGkhujY&FpT6kXr!KyIZVY3KX2>+Pz@f9P5R}H5ez$JuU>Jb$`vzfP_+a#|
zJPI8ejYj0V(Xbgl7MiK!kw~4n$A$5z-EJm6Rt&+8z9AT+55eOuq1Z^?+7~ZE==n&r
znz9Vt=X1w5d7GQ8hKU~L-sG)OufgLFYBaB+VULLQ;xje6)6bMQQ;k2qYP9*GLdD-I
z+~w7#NQGOK*sCLJZ?<M_+d_@`P1MXhQKM`FYQyAhC-S!M2{oScx_pvsC$mjM)NtZs
zp4Miq>qza9^>GvO_Sz^l){)a4mhjJb1_(<~!${Vu=BsfYYBV5kvz*lU(Om`2ITfz0
zQNv8u`ts^b4iDnT|3HPo?Nn&LCI(g7s&MLG3>xwOZ)jC4d)25#-ku+&LC~ovlpeDb
zZTm-~Gud!$UNm>2MAP3Ejebedc(*4SRwm|6Pe?<n+%)zAwfIVH*ZyNV%C#d8H)s)1
zl#YqLGMHPPfxVUJn+j%scOmP`kaRpgMopKz?ZsNvCQyg6<ZaVRI{L_1pOUkhY%MMq
zYT-xz2AI;|rr>TwD=pg4gVc|2Ka;z&c~$h~)liEe1JYpLmx@ldsdzUq8O>HD!+&cE
zZqOHeu0tAqsA;HZNW+@>TA2B~6YbKmmduY(>aepiXNl-{dd2TAmuH1rdd}{Uw=Q4O
zF*r9J2fFK!=&OTeb~<a&bW|p9>+Yo&ioD;Pk_P9@G@K-Fm()zdt26xit5TUi$GwdT
z?qOWY+33M>$bT72y*L&J$lG(|ZNZIbxJGg&ts1ZV#n2sCh!u~*xUXkEs;-!VQ$r`?
z_OeMxEW;ff(}Qt->=+z4IuhH~^+h-DKCq1Gjhl8o;iKw~*8Mx9Z*+U~GqgskvNgUh
zZH~v|eQ@NyHwKcoJBHT8bn<p(We@B;S{?39s^cd8QUBcBxznu*_c>KY-NBU+wy_fT
zIhMy;RXJpq=l<#p7i=BJxq=B!c)igHXUW^Od!68ahWl(IoUr0ASxep?C2QOAI()Dc
zG{M{%@6X+&za5zW!9AnQjSYQhj}-E@_Zd5^V{f(*d7DPwwjpodu!nW)v<+vznCth4
zvz42e55ZZU@6Oz-PTr2EE?1en^&oF2uvSs<UiqE<&dgQ3_Xa93VG4J~t!JG=-R~_K
zY|tyvoAv4yYJJIH6wv2WZ)DATnSIcylh`|CFYZYOxki2R4DaW27V<e*)5;<RF0SO`
z=z+@TJ$24>vuwzvCz!nTDrJ`Q`%KbX#~GM#vy?q-mhD?Da)`610UERHpkL^!zfmTQ
zHOP)i2D!GeM52>Qq!W3&taOPiXYV$Pv$3bi*m{N{ssFu5nvE_N)z@Ol+-#6L)Mia{
z3^J^!M27PF^CWK@k+-kdWA$*!m!Tu`WUh6dG<cjVQ&;E8vPHRaXLzn0&dZTkhjJvG
z*TMK4sT7hUcQbONlw+=RYo05?opL4XbFMt2f2vJlo;cB?T#s`>-G1fD!MeHPvo=Ti
zY|9bP!W`-9o-41fvtE9dBTX;mh*F;;t*YgUZEe=a4Rd96W{xy#k|RzPbEMiUy?8y=
z%lf;y(zG7ESn53a*(_hmZzz@v$4s(#i%FKOWX(U>B;&|i%`AghCmTg|h(4?1^kAJe
z$l@OcsWQMQZo7@r%!ZmWYtq0S7CDw}kvW{D`u)=)o@8nac=l<_EG_<i=99Na+AClr
zZv)Em_jSV}YhIaT9ld0Zi;beCmj67<C=;$4rPgMX%#Pyg<S+Zdy}56pD$i_W<83l>
z=O8{u4Fy^nE#gbwZu!Ve&I9c4&StM*6Z^s?+>1ok7V>rXqbAR5h15%U9*X2^stV6&
zj@QlNK_9j~pWmAO{9@|<m%4Gk5_#+Ymux0)hg76LfI7-y*16Nya{h_Ff@UGiAA3&r
z(mQp4dd;S0Hh4FVIhKoUa8X5$lDY4T=$T((jbr5PdGc1{M!$kL_iBvcKCD=B_aO5}
zxzi!w2OrO?LADL{lfj{{xmSa=@7j3&yK?0o4DxpM4?E79bC2LyW>1Z>NA7R#IdpWu
zimIH`Ep))}=bSO{bVk}%XKcxJ#?BxYv}o*xmYdx0Yl#m|m-oeI@;0w;U9`E|1ePgH
z;rX{YJwq*U`)3;@#dGHHo-h}#7q&eciI_Q~xSwk@qALZX{F-r?VIBwf;PL2vYCM+u
z&&0)YAt+xb1j*k*n4>ix!wd`X8j-l(kon48qA|JMGDJ>NVMLw^aW~Xh@mh_XBG#Vd
z?Im)y=L0o-FR5{mjQmXA{vvM+zN@hRmkO82+t%c5J@WPpx%rbU{8(3w*Y(vXBumX+
zYCI%w|Lj+zKOcL)lp0@H&z9nIwHm2LX&*I$_?XV*?Wb01%zdpwxh{N;31sVHdZ(VK
zpi5I>@jMkW0#yibQDa_d*24xBW|FIl87jCHtMHD`bGtb|&hO!NXB9eiQK23AdU;h0
z4nB=R19J@Kc2Hr@HWjRS2AIU3?Z<Ep7VM2e5O+NczZr>Bbs{l*O(ZIAjHHJn5=(DH
zq7y$JT%3yCw^K0jUJCM?r!p%g9S_+X-BwGBL))~t`-J|XZW(A<I)lDe&b@l1<35>u
zu3H-S2Bly@Df)$}^;!?&Opm_~K|j;clN>8`gtg->vi7qU$!oQ!e3u?yGS{R}!_b?l
z_&}}p;@ni+F{fcKndeUCwqhL{z`FL>u@qdXo{V4QZ5wq8s&X&!?2J?lU{Ccg>u_Cm
z8rp~QImz@E8`DwtRyyh$(%IMHj(4AQ93gKT^BnM!y#2$o!*=#+pHO4|L<XN$@p0w&
zXV10RbAa5QqQ#EMX;{=QjsBT5l)97(Pu91$$lFs9sc<20Uq&)lyI&jvIM@BA-E!{a
zjK;<{tZhBGD?TR*!|kGwcy}pwM1^7C@%i*HhG4R83fcuu#_CAsBM%&p!Q^dh$1zAE
zZ<EjU<!h%8&!)ZM{kaDsqq<|~vrafU+aGPqx4}HC)~KUuj-lOs@OY^gwuE}ZFR&&m
zkhjCid!XWp>Nr}v8m=d~qd_Hiq{LQX_Ht$X99S9I3EWRGy*zHDmV=JGwfMSV;;Pap
zZ+67v1V{XfcI3WaCj{-|PQ7p^jQ-AD*h)vVy;cgNZk9st*is0X%$=R&?TxShzbCwT
zDl_ZqaHrmLJ1j}HgWo~!Vj*vzk7iGbyq!VM@x+rhI2~kzC%gxpBX93DvBpw*SlY2Z
z7_y80B=*`m`6*HVw*n5-2>UMQylWlSH`Eqi^Zr~Zo%dy4Z`GV{eQc2>2NlRVq`+A6
zR#`wjvRHva-gld2a}Kr-`)s^-em_l}ag74qg8yGbZWF4&I_k%%f6a3Dj#&;)qfY$Q
zBzYT6a*935S-nhh&%*xcEu(BaV-)iyvpC-}%fMjzasx~f>1dMr&5bhSb&0&AmRx^X
zi8Lc`^PEeh^FiiNkhd48CvSRNBm++t$(HWLvMRh-j<KiP$C16z5QAjVzdZJNq1boH
zVYZ^4HFLhSdYmWmqu7&m%#)^1b0u|6t{6jeWjX8DYG!hAUXDbQfhUvo@^@Gc&jdN*
z{y0Z`zvjqiGI!j?TnQi({SxxTW-Bu<*vB3FBUi$FbEWrcGL*cXLEdh3qXxY_S4!>9
zk%9|3QuHH7Mpn#~d_K;GyluOX{LR<Plt+45{!%Xs@9Aayom>g3%lbDaPhNG-XXaF~
zsL9)j<n2)MHn^in;*OWdnIQ)0zNSR}D^n~jQi~<Hra>af+d=(}lD5Gpd!6Vxi)9aY
zn@P&@Tr-t98kb%(f5Xo#OIWXN^;gi}z}m8(0-MX4rKJmhADfJFkiU!2l}54oU){XW
zD6i)lWyBhz-277_wW#4Q`cf?Cx*DbZ5sSPh7xUbyBeVBzvA4*PE=KWFm?VMqalc=z
zflINc3+lqb?7t>*W;lr#d3%$r{lYVxJXdg@pB}6&%<j5omh}TIvV1;!&9?j;@^%BY
zYCX>be_k+e>w#I^hM1+?L5o~Bkn7Y~(snbqguUE3WNqRtW}toM?2#Au0S~poAM*AE
z>)mpLZLqDFJ>LG*l9w?<JB8ZwX?mJV+aQtuK0YvqWzq&dhPyE`c~0&A%?5+M)8ou~
zx6gBW9QxQ`9Q}p^%i3X91n26$a~FfFJsu2ae&ir~Jlt=GlWE)sxsyHO_jWj~u*Y<7
z&if`ZPlLYZa(f+ML*5?us*X1|JW%ODO`PlGiCbZwSpUu&-%Z{~@~Ma9lqOKtZG$ZG
zHYiz8>z%;9LPw+=ABx&;qcPAW7#;hJfeZ}djQJS6T)|r5{y6lo8jp7I<KY)I8FB7Y
z5Xx&~-h3?X%~>JxHt0kQ{3oio4@ZqzTI#&iY47H<u3W*ok{tB5SEHg*jeX>*^-~pQ
z-y&z9lU?NP1v1o?AAetwbupQD$&=bHc{_@{eag3OYpOBg6gip6&mp59+Nn{~U5%~-
zs2Pt?BjJn+eUI|F$olPkzU41fnAk}TPx3qcvkKq&bxx(KAmnYsp(+?UlA(`dU~8p9
zqpp1XWY*c_?L~g=DZKiSw{6<SpqF(Fp1xa-xY3KC=KM+QvMBT<Z|`qWA)Q}8gnw@6
z8HK%JQ8+kwDXx>PP3A;kYuj*YY~ir73dgBoi*c<cuY78@*HhqrB?bMgQ_;3YDn3q0
z#k4l*DB7FOzh52ic{(hi_qP$fMib)5s#)|Bb<rXyg*ntkNmy7wt@T|JLd(bE%tsC0
zWofaPEQ>$RI`)JXQ!};rawZMdTX{W8h0mr`lx6MsX?7}FC8uKlHqO~xOU3(^{PTXP
z@Se-Pu@{q=$-~(t^49NS3RZCzXE1sD{sy&gr55KGYB7y^_2mue7{D{Y#VXu~SCbld
z<#Y_{l8)k3o(Xoa|H~bD3G`Ii?qf~Py8J+=bS!gW@0YyQAJRfSPz%SdT9j^^hMk8~
z@I5;jql3BQ;WcNR-4k$_9`PVe9IDaRK4>#N@Uk4a<gNA7Xf%k5!Y79)oOu=w30=gz
z)<sA>wh$rHXJJt0WH^URhEv2O>{~w`UH%&j>u<rB@q7Tj&ufo@e$2F3-wwmax5L7b
zZ82$geN69PAN`xxV-Lv>uA6G1zJoV5>AYa7?g`U<4-}+(;55C+gF01bFSi=n$GhWA
zHFq4CUIm>GRmR6bmHC=?L+2^wk<+t0lu53f-*-W5O(*Qibwp?K_RwNS?(CyK>OUv+
zV9ncigChzorEupm_m6HQYact{!VL!m#yjA~2M2Vz%H7^^%!+Dhj}GMR9P-wUyv<zC
zo!$xDKj6atEqQzP1b6h%LlVt<P}CIqpc*m{(PV`P>cicr=RL2eL`Mg{|3iUmyI8+2
zW8bW{0tViP{m)wDF16jd?EhIjEn;`sA|4CKUNW~O>(yBH0q2}hz~%;h$KROwWTwxC
z_twgMo~?=0k?E_c(m;VF>n!44!6MtTsd2Jyt=Y^huRe0#g}ilQ=F(1oldLCiU)(iH
zSx=+Pq8?i6s#&}{vlj`Iv?gzR=}Tmr6aC1Si{&-_$C}0^GS`kDTgqCtdy&+fQzVa{
z6-k}{ilj>?&frWhmM*`FBzQ%REbd9YwiNrXWa+}rIr93rUL3k+%gqb_WI~^Ra;nZh
zx$g5%%GCZRpJx1HefUp0%*+<QE7?+cnqEHt$d<j+_0syXUK(A?5#8q;sY>3y-JB~g
zpXEvo)}XyM=1KFKoX1^G4^@R+F}K&tqG#E1{cpB3s;ZYJGjgO;XpZc<qL){z^>SvH
zUdCGG$m$9?GG<JU1V`(oJ2_j0yq$JcF9V+B%F>)%nOQnds@2Jpy1nvbk|VjhG+&n2
zDv;0mLg~*lM`Q02**dLAs`wPigOp+!?P-u`^0sSjqbwP0lnrhcab&$1L|y&HUXxs5
zuH=#D)aaX=W%eX`pl&g9BbK?6nVg-9Hi?<M9k|gb&u-E;O%K_hj+|NIyt5B`;V#hz
ziCsbO_vd11`@UGV?kkbS9S!0WZxm;xQLI)Q#515oMjA>a-fR>PC3(u{8*$wz2aHB}
z@R$4#Hi;E=@38UoWwEB+Pu^;uTI5c1evkY4J^VDuu9g;QJJ2i@_p)Do(kvDF@SMlj
z<VNzg<UX^e`Dc4{X8ALmXT7rYXi<BK<c!hegVuOVUD^K=XU5%a&~p&kIg|UPmUAa_
zI~&f#aHkcRl`)&Xs&&j9C2t4W*uZ-P^QzeEUC;b8wXY3UzvZq-_F@vQ+8~Lwad~Rd
zitfzom_W}Jc{@FveyMk4F0&q&khhD-+t7V>_{M*Ok2Z2o74@dMxps*5;r?9{bF5}K
zz?!^`Y2*ek@;2b22mC*KAdlMEtgAI}KgkObNwx5xs}E*Z<0iC;jnHJ0KdvMShI|lw
z{jUSU_H@P+yAiN&F%lQ*1;H^O2xI$>!Cap)m@;DwhA$0<r#2W1w+1uQd;<14PeI17
z`It5^5?<u()BQ1UwN;^;y9!x-)#yd-c1sp(&@|Su<ZW5<cA`avE##`_3-a-T3XVTj
za3LShleZVi%0sMMeXFano4lRxz#8<QiaCTT-p5sVdXn|$3KiO>lXokr=WbNNoxE)}
zoV{D}_Q)X>Uga`tlFt|NKn0I{6^`@yw)68p-BaP-5f!Ecso-FULD2gc*pRmY#W9Fy
z9b2mvzy9zTwBNQIhx;w%-kwP2MMWUwM>yWs3}=4pBHUcM5PdE$#E&hJaOBmTHSFz5
zOA)ejDbfO>u)p0>Jii=?53?e0fxMkjJscr77vnH_d#hp?+B8bV)6*$fYfeUO^7dBs
zRQws9in<k9H+JJZEcw*<L>dwz(x5Mwis|Ih_p$7E8q+X5A{j%-+dAa!&$me!MBdhZ
zsbQvUEIMwB#Ru-)-4T(*`~znA|KM)lEou1g>HoNyilYls@z0+%t2z}u2BhNZ<y0iS
zOhqyCBxV#QW6}8}yk5lp!haH}`Lb`y|BmhAp2Y|3;o57N^Qom~%HHo^Erj~;O7gZs
z*>o(mO-F6=wmHuQH@7hpgV`0@m+6S%+2Ljw&k7w`>y}C99JUr4XQp9R$25#=l7<qu
zG{l@p!P^NbILoZeh!dQxqEEcjPI|@3+f}S@zmvDoJ?S++#xvHy7>wz=7#(jd!j#do
z@o(oW*s5n@ZO@tboHYfF`i$eA$sn|h4Z^<XLzw+I1TE?ff#aG0j5T$ncHbHw>$Sq<
zpDp3_rv;Lh)Wxq5KQ#MU8!m%=F>i)9+FtU)rY>H%u+amG10JZa^}t6{4fJSV9UM4A
zl{j~Nt?G`Fshk5MZ&!_^U&^%-_I)akdV|X&W0Whh3}rBjyzTtk5j`Rtp`Y%EHsoy^
zr4wq=XS{WlBX*kTk)ls(QU-m<u6%zT>)IIZ`K-fi_UaDEP}!q(WA5R8ZHH9}^h@fQ
zwX@b1Uz*zDhk~>Fcd3_c<y-{M`BUD|FBQaGzIW8#dRxJcymhXr#GCTWDkg77lA%3H
zD>!Gaz`jT1Y8(1bsr$AWOfM7fdF{#D8W+i5)*T)9vp-ATs>$0?<Zboa)E_^Q)vQ+*
zvL>-2Z*vnA&<#`IL0#57Yb}z^zU-22W|`Q+EUENH9o%h_bG#StZeo&Wt(b4fdPqS{
z^qj3xmX0%ubtBesA<TXti?vk@l3a@U7KUQ#vBjd9L*8=c)?SLlvtE%DdlgCS>>`<e
ztw>g^ERv3%`QmqnUSVpukC*4j`3}@-$+qtXy_|^GOV5pZ$?1|UM=t%7@?>qDs()qE
z^*<6Z`j7mY`A=qi`6sW+Wy|4<+46NLYt9eZ5-~|HmP>l^yP6|E9<tV@pDLn=?4|C!
z_+E}2zMUfhM{{KJo*X%LOfQG4>*e>8Y>6Rnw^q?h`#pMDJS9h}pV!N|488OrZ>Kx3
zUt1?fHjHE~9H|$@PQAEoB5zmer9wBoXnOGb{GKa&ICpfQ2X${lp8Q^tFJG${$Smt3
z;qoFmxuaN`^JiaUE|8{MsgcuHr6F&#$lEXdjN;nIDC;t)2Ww5@zlLX*|4dRa$0R|2
zjFRxiC|A!iH!{j3o0)qx>xw~UtTV_y*2UZDdG1HvK5J!^g)@y3JklsplMV7PqeNa^
z<DBxsVzFvdEbl^!rDPDZvCIbfNk7}PO4J^Xm58>;D4u+d@cpcT$=kY|b85`|tPuL&
z&ND+RneR9AA^WLqOv~WxlQsJaTTP;QYm$C^jhL32#c`ckv^)9QW6sG_daZ2|OybM$
zJ&@<Ls?<p4Mp&dWHJHuRVj^}@n?A%jH8M2t19vb~A!`S4x8z70TyIIOn6LjqKRGw_
zh;u}@t+2ZlcRcK|#z`yYi_)L3v@kc))&@0e+2Gj=?ks#~!+rDAmP7bHd%Sxp+3<{N
zi`<j8P;Ixx<2#%W`b@t5WhScG4sG_@;aw8zTk>`mdAmx_9ShW&JH~TI2zk5nj3eBO
z98s2Y_P5k!aIdoqjz4z6y?o|i4D`UMFEz2j(G$}*dSXeuH@#gAkQ&n%+xNG}UGjFv
z%+5GGunSsZD9U>W;d+lC%<DA<OT5NlLzgkw_G=6}h6m&As$is&x2x@^p!nH*&U14A
zSqJKVuVOGOHwH~Ala2rDr)oIAlR`hz95ptQx1Vge%a6RROdig=tbz}DyN5Mv+9h)K
zx(cz`DtxqJO<RrLCi3<W`8<}qUG`Chp9@u(Gm}~{S?iIc!ZV!;HtSWW!25Bl(<+?c
z^YkEZEqvSsYQwYctFVsGry*}6$lI&r?IZGbBzgO*BnJ9-F$kU*gX|E#zcL1Kjw<*E
z#$f2W<!FC28bbo3F~34IF8pMsLbY(t$}HsGw}lv$9|nmCL$c>WY@QTB?R_a;uULw{
zta01DjKt<`5xC8dZ|)wByS0{}>&eC3?Y$T^D}<qzRVoe~Pr*~}*$wlGr_Uf2n|7q&
zLEAW<HFz$`<bES&6LxWo!y4{{nAV;<m}bV|4QG^Z?oYyc<|?&|Ohon+&M-fuU;2fH
zy9Q%Xads@uE{w$m&O+b4$N8BwW^t$zahtm!{?1Lr)-}xbAaCnePQ}O?{Ce!4uV|To
z#!2x=UQT~?7-y1GnP+_}0c{^9aPBD?n(=A4nw*A~<Zad>Ee=F#F~D1kt(@E0c9Xhv
zWi8TJS1+gIqxtxB?2Kjh)Wvk{dd@yBd3!E29lJYl&j9<rl{>R{T!}r}Hz{bSN<nNZ
zYS?E}P`e?|KEIZu<H_ZSZM_^de=NiDq-6+5jDq+0DCV?B;_8cVY+15|KJUdCdV3L!
zPP4Hwc?=r=8;wiKAk?ZghJD!}jPM$cn&pRM`^!M=xIPG3XZx~8&<#D_1Lya)LA~!S
z(dtwaob+nIS@YTmjP!y1A|Fhh=z|I5?fgJ5lv?G1oV6Z!waWty6Fis!P#syJRU!06
zOquHrYx35cygh%pG8$E`gmGufq4Ww@d|l2=sYhio*V`F&$l8HV9ia^8%ulc*X6qc$
zg7Z9$sV#RCUWKLb<UIF#_bdhfJq{@D?SM9s4k)SRfI;MK<7M_dmocO9g&p3<anCw=
zJA57IX<IN4p1fUn%Laby=uaMIgC+DPd$+YlCuaJ+46wrH@tnu7&@0oDHHZ`0$lQd_
z-z?JRy+y{xTO^5^vckfw!7P)k2{cKY470R5#Vo}zi&RNfz?`W-@L>h=E>jPDz*$W4
zw%1!~(C@j!mAuU%Z#6@xP5UWuQ*9AHX8%tkZ%6Z9das;W6x27HGB0=l^M&6$G)Ty4
zgCrcF57)@IBd9AkHOY&4Ch@4k`!9Lh+fpJ8zO!FTAJh_hqADd72@|Qs*RM#v{VJ3{
z1BztQsUq=-DUxpFZRRHWh3C=7v@A#FAV+TR)=RTEy;z0oWn!>it_J^;O{M-x_|d;|
zzwBReoA5^#MEsT!R)0kKjP-6=*0vkSTz9=F-)76tse18Ur<a{MddX1d$U5rHDr&{E
z$=e-dt-~$WsYbm#+nysnNA;4Y&z7P`+45SGEqkiz<=A>MbYhM;|Btt!dZ|u^5B`!(
zeVvbalPyEKlF?auDdLrxq?eQA?T)5;Swh}!O39HsF*(wZI`oFIxiX<*zASbxl0$hs
zXHaLpKeb3ScMBvtt3dXDE0Aly3*`2%0+|?IAT{Z4KFchP|MjiUX{ZxWk8VIMx&v9O
z=e+MvvM|faC^_?3yLMq-2z7*D`lA}~bIOf3%9w=)+0S_^-Qi-{dx+k&B_-13e6hTu
zR<CJYEc4$N$(g_+sXe1eB2tQEJ3VZfJP*ZjPI@Nw=oRTk8F_`X(8WecrcUh~$_%al
zBk8>3YX0B<AMHZ65<>RgLf69{rKkv{gpyJzGe!HP!dt`MdvD(MdfRW}ZSTFC=bSoq
zPMu1>$LII`<8=$2*3o(0pVxIg9*?UVd;Ik(<r8PjQHxkN^Ai3ZWH~YS@@bU}r)GUp
z%pcja^^es4r%DPq8|Zq4GlC#yT>6pMyfSD1k@c_t$g}R$WlyCJ>_!b-`kQmzONMy$
zn)%*2hUj^nx+wB@19kT|K5(xUd28~Np5$yYm%MF%l(|=&vw)lpsidAzX^6dcM!2)d
z2o9`q@13OfihAevWbG%`%tH<u;v{R`usHUPj?-)RoO==qxJR%lcWlk)Zp)=6c(B6+
z^U2%A<n5`+%$ZqkivHp3M_sev?pN+csB4Py<T@}QZ&xrABg?lw>NqsUq+N}%qG=PX
zY{k9xcUqzsvz24X+qI_#<6H1h1WX@}MD1`mJ@mwbe|(YjWE#3Cr@^O{AN6W}*!*J_
zLL35d<v}1)t!87rV-WnVQA5=v0R#SthwZyWSlKG@$SN5#J1)oeh1B*g;|w=+IcAf$
z6=dm$D$aLVgEk@m%E-zs<YW|idyl-m{3RJfUhsYVckK!4l-ehwpj|RHPfkWiUozaD
zy<eAP>{-c~FL~>>Et&J$WSnotTDb-1wNZTkV*WiO8BXLa8wXgx{_%hy>aqOEY+vf1
zCMDA=qri#t3TS6?CL5u^LFXj=3QEAOp$VKj#$#MaJklKEQESmse7YG$AAc0HyO!Y9
zgeB}JM&f?TV#J(Sgf;W$lO@rZNZ#Il6348?Slk@H0LR|Whthrl&U-Duv*ZP+L4W9|
zoW=0erDF<vw{5xG-k>>oo1Mn&-8B5&KLs=QF2}0()NnmcfkTs2=;;Uk-7XDvnYpot
z`5L9{jo-PZpzlF}00#vkcy1ZM-a)Nc1*}6AxV(n_Z)(2}kheh>R^ax#6&Pl_5?0RK
zm1MLM4{EMN-xn(|&x+ap+3fYR$G@p?Id!Y_HYIYO&f`=_OXhwgF;AMFrYk#Fq25;R
z;PJ`8*4vqI49~(xyDao1Z`-ra>_@#<*2yeHSWqwZEfeh?W#ZADOoZIZ#G>j<YziP}
z9oAq8wNmB7ve3>n11Ue#ahN&C$1bI!VFvpRa}@ZeFbVChBw_!^B$&rDm(eK^j^u3*
zuXuDmwiMs~T8h<9OOf0m8XL*m;Wxrzdw4E>FY?9a^FGkl^FfsBRCq*9!J@C8s8}-r
z?g!mr?(B{g-3G#VLU#<X=#DpMx?%L3ZrC5t0e8b&!uYlw?$5V_-vB%Gr7r52lO5PB
z$13)bYaXzJ26p%w*bH5^+M+=_8+6a9i|#|XJJY}lIhs01d|L<k%j&?aRvq*zG)M1$
z&9R~;_tUb54OwFWmpRNko?rpVv_O)t1x8RO^~%u#8{eB_$9i*&9Aggm>t+ZbZ})UH
z!|j%4s1w6IpNp9t)zuWPvrRC~(FDWI3=y1P69*j(ac8U{B#>D?y!@$2c6h}6TT4T9
z*hSys56*3eQiohlruJtqn7sX;KIZj%>S)eX%e9nh8M}n~;g3~f!|eR(A!Obb`cP_D
zi!82|73?`@uVv2AZUgisb7ws^K>yF2uYII%lC!v4ta~lf3=qeeTS-IKw`tX6$sbAi
z&H3YM>Z1HPf1FSy5$qq=;eGwaq)PdGmcH5(6*7SAJ#>)Trb~Kh_((6C$$*QEE2PIe
zz5KUYFU`o?bILL?$}E#{b;?A4Un^U7Yvp5>R>VXnIrGZIntJ1X$F$<no%*E+>Yn-(
zOY1E~ayqq0T-X;bKB1OJ<n0VomCO>QXi^Ks+O1F)FDsDF#)VSSQz<*{^8H)YQZPg<
z*+<nf`ITDM=+u(ch)k?oB+1{^vX~q`cTz2~Q7ygF)betHTK3OY%LTr*e4>`E&T2_a
z;$!mF60xR8!iE*gA@+>p=N8F{<7%mSUM=Ud)pG5<TBfq+tVWRxno%TU*jv6tCY$ys
zl0K}j`~NHwU!EHhIX8ABZ@;xEk!KF2^2uB)V;XBE=ZjXBJ<v+~(OPl)O77-q<?~{#
z6b;sjd$>-VM(bqPO0C?z#plVPOlI7#l!oMO<|}gcHqSO&nK3!KLVQ2zC5Q9bC7c1n
zn)<YRdTByE*5Y%VgOjzJdDV*6%Y|L#qID^kN@~+4k0}#Vw=yYVPg_CW?qbj1eX?F^
zT`FhZbh(_H&pMhj;COAN9Lpe6nY+1#Om6eGLeh>`$lH0;z)^?2W+it+@EmaDL6zjt
zS3HwF?Mt2M{kmK(g-gq&5g9*~8nF|;Rgz0~KVuGNl@EI$WmWS1Gj;8pe=M7C!1)_B
z@|@8+-({u-dFy_eyAq7K$0COw-}*Iimo;$Yoto&+bLZyU%yMbZeGEF*!<_AQVO<=}
zK6G@35jrFoaThR|+nbuJGQNh$aHpMoJH`kVE7&jEWdzFx#<=sZF(y*8yksUh%h#dD
z5%!ynO;M)_eFL4y+hwNQacYXR{>-&%V2b_Z?Ye0uIPPhJx*trDy^yor{ie8i)fC=J
zON1oXMsju&9KYFueYB3u2keKc=uvRT7;H6ig3CS^d^qij701UREMq*1L)>ukod=Yk
zrr~tYKpejlh#xa&<3(;TbXOPP<=%LlGD|>{Rf(uPnS{`Z3UrHB;P@`qy3do*W%hC$
z{+x`1<ZU>4n?c?-CU1|EtE<V|tJld$^7b76P0t?jEpj#C9|hdW+h64E6kZ4O6$q@y
z87ph(Nf`?CTg&>iN`X-F_8fUzo4kF&+IIYcWbOgs-1S`&JVz&CQf>lTasJn*P9ip}
zPlOZat=YX42ys`S_jCn1j!+=uEcf~@PQV7%==;_27`QMV7CT~4l@pDD<D)UFRTM6g
zw^v=3;Ni1yH0T_T(fQ$M+G9Redd)?F?_5L|E`q81Vl0T~KBd?5k!HOBW69iZ9T&i)
z^8yT~hdH7!9TUjgR^)9A^QoF&NJBI;D}UKB`+4nhSfna&;1R#K4=HH&Bo%F2rNM*j
zRmV}=&7ZURRh|t3s6ec%!0MPJT$r1L4$T!<Wv_tS3_j=dEv1pS9m(4f=T>0(i4{m^
zW>-P!3S6tP68*{B0P;5N2K`t3dQQKVbN_1!^vhCFgS@Rn-eyOq!RkdCI*eP5$J8r_
zOwEA$UM6DPvk>1bi#yUX(W65qoa$tu!-q`FVXYfSf6vfYnOH#H&LVH$oXW)Y6PY+s
zoQaE%vrt0bmNEx9i+-O*mg(3--i~B%Vhg`#(^?9&(IvsSVG^>baerqRiB$gu2#%bG
zlP=-7?h%N*wf=beV<sA$orzZ+{or5Vi+%HbvF)=D{_f+0h2tjTaVmXnfkQE>>rkvp
z>w&x#-Ed`5C%o_89$T8W!>M0wnFrSjm)5p|vL$s5*O^-pW(QO9*6eBv7$>#BiGD4x
z@Ktjx+|~>Z`)#pen+-;5>cKm$F8o~TVvkUNTwMoN<Zb-DI(The2NBFxebdGg{jQl~
zHMLNlNz_EGw!q<87MSwK9L^Wa(e;cuEO@zpFh{G`=Jbb}W1pQlO0Jn9W~>>K$Xl;Q
zX2@J(imYMWLD1S1;OpF#yd6N^rvGLZ0C~HLyuCb*J6=M#$Ax|5KxWMbQJ37Cn$E;C
z^f6L{QdY=1wY32z)FVIdRLf{;P)@qgqxq;xHnOJO{HjW7leeE8sI}S4j3bL`sp(WL
zJ;_q{zYI8kFhC1>G`qYdS3ekF3wf(~&3-g_TX%&4u8%jsBl31M>s$Ywe?&!%ls)&~
zoSac5;=vsOtb3Z0w>P~iCHsDbjNm;#l=%%M2lUdCyp1Jqe=?KFp1j>lt<*j8cE<p{
zpp?wuz1^lEYg_WgI*lw@r4=iBXMerY$qz@J<g>S0eWO&i%q^A;>;peq&w5c&B&#`(
z9XC@g1~XJrXu?_T1f?8YP$)$%g>uNhK(@Colp{Tq68W!69&A>NM_;vs$EwAdynSD)
zmRk*3yP6kC=WlBHeo8Gi<mToLoY9iE&B)oov(%E#&)+gZEp_{;<t_hhPJvpYlUY{}
zE|z&Ui>S3`J<GcH@Huk!U$yijb3a%Y$+ONy@}V2GQtT~vTg2XV*CN@`vq;W<Dw3O#
z#bWWkSW3xTgOn0El1<*0mI}z*F#hajYqjD+ACvo&Qi);?W$h!SqVKO2jrRY%N6~&I
zlKqcX?1z-ehZE&8_z(5m7pZB!Q7K+qD&!WQ`*9!j;&fIo9@Jj7v@es^X*&7KiaA@H
z84o+l{EwPi`Dv(=Ct95t_AZxJ-^%1U&pbxs%H)B6nJ9H-GTclr%Ud&VguEThx^(ea
z*0LIY{@)dnXk960$C$HKTQ8;6<x;UnF9lV4Su?Og0?FHg*Ol`1TBV#fs**+9DkYrR
z*x3u{r8X;<Y2<BXPi9<i(#zi-6>{Midn5e4G-5CMbnQPfl;_6Lfz<YWW40wVcKXUH
z!3oYVIQMPHS+W0J>f_1V!5f$%No~H>1?F#&x67>AXC_OR-ey+SdqW)M40o8J5vpS8
zS)WaBP){RRQP1o{O=0*-o>jk4*WZfzsij8PW5s;s;l>CKH^!W7<}Q11X99V<QfYz-
zoaua_?&{WSBMf;*E%XQGlGkTub8Qn;cj0_@0{uh@CTQYd3JEgho>^1GD4A2m9&>8f
zrg+_~B_1d`;Lxg`Nbk<fsTz*h^?op7orYq~S7y357=sCBt~fT+4ViB}u;bY@IFYxl
z$lDF%?J4p$&SnmZAI)J#Y&`TU6WRCVeRF~Wx5?WcI}|u-my8(RTbF&Jeu*sYy^#In
zdC9nxl#EWN$Wrq5IC*=VyuJA-8UCzYm+2L_##-*u9tG}@tBpR8t3OziRw^)Ty#g)C
z+k4>(RI&f-Ltc*K+nk#U<gpI^%6hidSaP1c^$$s6eqsW$?j<0tRwDjWBqHd15}e7W
zM&xb3X$p9BR$xqIBFeJj(X%Z*JuBl7{4o}W^%mp6*#+3^JrAqaN5I%W983Fz!%-QA
z-Sxt8r6e5r!y>S0!dy5`or^{ubCK129zOca$0l;PdD%QHKQj;8Pt3!O<MZGWH4n|n
z8?U3QVRqp!fH_pXHZgx=Q3^USqf-B`0trb9tR!#8a6dz>r>XEMOolys;-@oHSZmXV
ze3*Um*d!>@l3*W^1j|lI++&c4t^7IP@$3+?Qh}IZ)Ofok<K@I;C|JvXaOb?7&q>Ql
zJ})&`&l=I!+?g|N*2CUEmQw?qg8l)_pPtWI@XS=K3rfY)jm!$WmWn<m%;4ls;{Ll=
zqVb!RXn&6T8tgOi-7pioA7^sEP$s&Nw@;EY@sV@p@L}mV0qUT#(&2EIJ!3!a8#tAY
zr77u{_V_RK?D-cQ%+s-iwf+20%h7kJ0;#tXVYe?4+g0&cs$GP)%sze?ynwY{1ZLX>
zqjy{&cD4!NzO9+ml=<Ma!Uye!`*2?Eh1os4uxa-cEM7MOzdS}DwPZM2r3|At*a1gp
zcSH5KF37sr0iSK!;js(%`@C(#y^O6e$F3zjZqQpx-sU>k!R1m5^ownQ>s?wv{kl1(
zlehD@hahY(bIDuR$L{5I5#d=Ec9*zQj~=|s4=vHW8@;vEA+P#Cu0AwJ%d4z$E6wpH
zf}WpMtYrh3(Ws#&>YO>Y()VM0oL<|5<~Tr|RI~ZiBsVn0roAS}duW1QKTQzc+Z5`B
zrm!Gyqximn8_c~KW`b<;*6<gz^j6oz3cnf{zmi!LC+MGD#>@)pH_z8$ZfzO&1FWSc
znZCxc$GLZoyiH?$JO4(tETQLhfBqjax>+R^^u5;Lo>ucVRpQ^6jO6|}!ybR+Co?ME
z<uPM0=Z};%sg{lObbg9qRwKO}6CcqF@`T)7N6l3P{TyScPqwGm#KeH!EqZ;PRY_n-
zmF$03DQ<^(Po7#X{cBQpJG5L@vX8lrexG`*ahq+_%e0Gnx$~Bp$h<F~W+ud&t9qGy
zL@)ij@qSI-woELO_j{OA?9QD6iCV!dtyCK8WY~W?dALX?wT@`T@fNw`RxE*6isVXC
zk&Ndxlzrh@8kNi;Z~L*nEncRS2PX@~Ag4eoCKrfV!$O&3sFY!;N=fdllKGtD)|jtm
z{;^t4lDGQ$MN)25B(K;HJ{8BFGj&z-=cr}NG_@R@qL!=wspPYzS~eA^B(|$s{N}2q
zp_2S<$v*UHwT!b<OSjtOFhBMJYvY0BZ6&V{{Ji<>T?g=SVPx%s@FLmPiFI_2TBeb=
znE}Ppi1XpgmBkWFotDvKjm+TQ1YJyt1bCH7%V(t$W2}|N<gMd@Qn^Uo^{mAu(zr$m
zXVN7yyFdND%;sptoeN8<OC*^&ST@{;^sE;>Q=Gf1xx1*C&v#Q>o@ay_v<_vmF;yoi
zb#zi;S1Ote=1AAj$`01fos~S-d?w!uwbK0ov$I&^=1wPbUzEu$BfTgJbn;BY94$BY
zy6fqss=Z!P!z#pQBr~MD=%r2tbzN2E@`=6a6WjFCPsi-)fJ&Lqd2c??0Bfo$CGvN<
z{PNbxbLMK^t16S-AIs%IZ~DMEKQ`;iofI1?CFm{t(02S?<*a{P%_>n+ALdi9N&@&h
zn|rQGzOmnYx{U#|at(0sjsZ54x6{&@^G)84I?uce>g;PV`^4XgyBxc4?@^W^l6BNy
zc~hgXiWwputE6AkD#^F_Bbx`4*J1pa9A;9Ix2qN!VIp~Zc7!qJledStTd1!wcU=B4
z!l^Ty86T&|UAWuh0{Q)h9(eM$7kTSi%LMz{^R?{4Ec1916pk^)s=KCGOC6*Qef3rB
zF~@&n#(GK%=IeDpN5?McpVkFy-*%xcxGTKB_Gf;V1HA=;c5_DIs;d)v{mnXeofi(b
z^~a!kvpCZVWX&}j>92xO`R^QzbPd7J>}agfGvAiHO&OcS+<XP11}o5uj17NCZluIw
zf5tKt9$v<*_*ncE7>BZjam=}i!@}or=r|+}MGkS8H6{+;OX4s$F^)dhIJm?!Ln1m3
zJ0s%pB2xj!x%`}A3hua6Af5N@eFGKns<#yXSuJ38SrD3;1)-qR9JI?0gEnJ2R$7e1
zU#mTEvhD=tpiDu-6TWzDJ`={H{n02k5Y1l&ay~E{ew;^-i4Vf^ze5mzB?LF0gfJs$
z9;TAFk7kBrLdS4=vcj-(Rv6L_g`)0(Q0#J_i=I|<p&36HNv9(a_In--ugyc5{XArD
zjlkE~2&D81M+^O2Z2XXp$2V7_+uwiTki`o8(<~JinKikk4(nN-O~MjbYo{r&m^EzO
z=ShfBQ=j`M35{9*jeDJh$&ZpSC^`ve!jq6QFbVtGCP7`2h_(EC&l5>#$d8>tt#xhk
z_5ykPnAh9!)RmF9X5_6IdFw{r{!Oj-&7sV@3|Wrgz9}f5nSyTQt&SO*J1?dnWO*uD
zUZ-Bkn7f|1XK*7krxSj2KN9E1USpWi`F9#JD>A5G$iVP=8SwF?2F#uF*rDn8-7g(I
z|4PSN@;0<}25X~qxO`rX3FK`LyT9<*f%}!duExUd^l&{(#zr5WAFSdK`X~xt>PBJG
zwI$e{5s5vii&6W(MVPlG2+KwW@C-B)d87G%`N9`nUrgoM(+k<%y^z*qEEaAVgMS;1
zqV9eK%9Dn3Rx7a2?t@U*E(luM0ZFWH`#)+6$D3{NgnplqyY0|)gB{|?+jsr#kaea7
zIu2xJ)cod{#(C}Odp4Za)<O{T7yWu#!py-E_iiw=;;|)8`fz_9c{`H4U0ZFzoG=Sq
zy<?82XIaNqnB!<Ty+A1z_|MM*y+)cNwWm2Udz-_QoK@{L$H{GcY)f-&7gHE=E);Ud
z1P3{*-A&%c(P#f~B>e{DZ6ESBoxB}lWC-I_dU>3grOVo~$CdxvV`s*h?C~kAHx1~4
zY{6Z#omtxsV)o)z1EgD1Cq>?Ns$@0-_p-iVuW>!+s-LN2F`UNwfm)`&71Z2xu96E6
znKkV9N5<2$aqPz*d0I_POYt9>Mh&qo=XJ+9hclu!=NIpFr7!+4*Y}USsKJ>q+1xeq
zkGMZ(9x(M$igOh*%|$PB$Citl+Tx@noorpLlX`n~GW|rE#M|j5fL@pU=d6`@U;eit
zv!d3P%NPZB?Tn(|R#PU|BdLqpq?P>c+W(&?@qMaR2KCd@2g4n-y!IW^%FW<n@uD{Q
z<<27Ua4eEd_o;hgf4GqP<45G}FJqNBuT@F|^7h7;0vTXiATt^jiUWE3+*K(PI;teX
zP%RGOYB9V{W^yL0)>8jet(HjklAWW~k`k(x)4}W`ledS-+pXm7Dq}J<M<p5qwUh>_
zrT;RuEMkqjX(j9BI%GFlyz;h6hI*^z8+okSp_VhXi^Q6|y~SDYf^urMI<xLQqn5V6
z)YAK75t&jfG4ukP8)#&9U$XXyMoOroy3Rcc8{d~mVUto>X275ODR&1lPjWT$ogbgj
zNG@|WjG6JQtWzSbsdaupe^D5D8$;d>99J&&o%K?-oOxD8?Bl=Diz73s`=8ZF0Ux`X
z{-!f|B{CwkL^A&>m2HMvX~8-1t|wZV_JYs%C#|eorIV}tKQ!-FCXb`b#Etoy+B}_{
zBX183E|-DkdU>&-TqYEj$(v^7V$ZtRcpLu@E6G#xHkrJ=RmxrldAsyLrSu|iXRwF;
ziyp6AUSzqwP8$46pBGu~NqyMf3*|DJ^Wgx_h7Ucil<nkeW>;##YOtnlOONmk@|YRg
zG3-N!HP(xdJ9S*l;l4DVSsSLzl}zJKhf6hZw}-+1*9iVaZ~2AM><QIl_S!`5Gg@B*
zPpQ#*991QY+f%E@v*uEI#7~r0O9yHmoEOtOSH;ZeDnop)WsFfnm{}D;U%(|}EUn4h
zDGm1%k+){7na@(c;C!rFj!rbd-Np1K?=?o+En|FdLtWKv=F#Yk@i>?sUh-Do)C@{1
z?x&8WE|O>R?2onJbg(&QG;6^;z7~j*7RW<aIMo`6^)e7Ood#j@PvFd-k!WY=g3BM=
z;I_&G0Xrw+$~Edz=6GTKbuUcsF%=m<7QwkoG)g^};z`s}^xqc^Y>L8%{hYb|io)mC
z(eRzR3^Vt|;N<#c&}>>pjZPF6I7FdrY7~saqOkofA5$8O3FK{JR2*iIw;|+hU-I^w
zJKxTY#fsJ357Q|Iqy8+TwsRRwFD=FE4NDPOYblPBw^z!8kgl4I>b=1jSThJN7pCEk
z!8q71^T7N%6OjFWD#CV8gTKoRxWAl<zK8vp<=_h=S6^%j@#THf7d5W=W8qDI?6aH=
z$B{v3^mG;ms{+v?c@Flz3B&#oVQ8Ngika_1(B^IkCRv4I)0nvk-4KDn$q_g?dM>nu
zAz0oz1k2QOFnmBD(pv?<{Qxs3nQ^(TTM8CmO+lhXDlEubPx7{OvjXioo6U_<;A~kE
zhTP|lVHH_h#k^{564t*@0?6B)3zD!dED5>&lW=iIBIc<QVZ&NCZC4UH?k1yIzh}Kj
z;$9a8n$}a`k83haM^U%ko3m?g_7B+0Z{BD*S`49A*?T$bYI?UCr{K)k6bO0y<veq#
zPo?7Sr&OG*Ooa_ID}U8b#f1UPtTdp0cvKqq;ijRdYX(MX(lLU(?ML25lDB5$t!JNf
z^dfI(u46A+m4-g%{CTN2Klz3kA@9h3^0v+CmAFS;SP%Bo&+9nTe-?|^Ct|RhJ}MXT
zc8F~xhV>6aurd%2=gdINxyiUbX%ddU^28JuPaNvxiTdXzB4g4RY+g7T9fysEYUT)>
zdodih7Y|3zm|=*u>kGAKXQY1ZjEnm^Aj-Z2nq0G`7Rwg@rrF@yU>oKtGH<+HT@*gH
z!lDQ(jObv+-92@vf2@N+HEN+XJv_nPEn(ls5@QEh!j?OG+Q*W++nHxUZgzJgO9z?5
zZWdX4jJ@K&&9U;8Ikc>G>uuuuhnXX%wmCw{+`ZI6>F1boSCa|8Ih$c6c^lQ&1hb5&
ziQ;Ru=Lz<#9-H9bXY}2FWET5BJTsHGmgMcfNz8s4Zh-F1*~{HoEjzQBXaBldmb<d|
zOMTSnTHLE<%?x-xMsG?VtuZs7Ea>SWZ*LmX56k}ISkCoYJtcEnk*|*QktFc%?DGa>
zSBVdMy?v7OG9*ebfxNG!jnT^=-m@*X)0@NHb0N>ETb;@tVT@i*X7O=B6|$8q{y?p<
z{T*sc$?wPiRmlEUdO5@Tr{@UHTbXm15v7x{n{`5AtHe+T^wNWx9L@=seWxym`NLY~
zG`)RLD*G;$(tE3wPTVt_Ke$x>{?4}z=#OFVRM$r<&B)r`yj;oKMj^#=YiO}JZ7!0T
z{fp%OUbTE3sg_H>Rq}+spFQNQ@j9inKU^r@+X^JZyg<gVmwfcKQfAgtNuS9oX;z{V
zk0xpnd)BF~)pCbdw_cq8P9<NVmZzeYo&IWxbz@(dto3FeI_JDfdhAfipJysL7@(GW
zOVyIojs7AsbV~=d?DbK}Z1NW5?cPYW%;x1<i@saxm^<)tc*pm*QOk~<Y8jfQlq>cs
z8PQH9x+2!jRm|TQppm`1>Hj^?w^1c>h1phhsiE?%DwQYXZN(b?nY^8Ol-jC?8u{W_
zBBr%U<hW6ZOxI|{l35yaXOu`Vd(G3DmP+9kYOglv#HgN5+Fhs4>I{8O+cdI*yj`1J
zB2VZszOb5k9&5<nds-=7t`+|tS}}UBl|y7=zh7Eu(y2@uF{ko_lAQZapYuJPG@*Vu
zWJa01R+q`Suhg;qVE?%*ebF1s#Y$Z+xB0u%a1Cd)YQ1>9=I>E=z3jCiE4lZ{`Uq=m
zYn?QEr<L?OW#VwGOd{Tt$=6!tlI^b-W7f9|Ige;d-j=cV(yoa6AK1VC_OMLaaE703
zR3;6mDO<qbr~5pw*=JSCu%8t&bRo~9*O+I?*T=|P)pB5CwYc&15oB(FSDcajPg`#p
zPM<IJd7EFdR!^;v#x?)Q+&9d;V$W#6vTE6vUnTR|tJ&yA?-X_0ahIqGn^+|`*#}SZ
zp-y4ZA8A3o*TRm~@|0O4%?}vhWriVk%`nE?o7{uYg0o)ocG!7iYWkU5)x!kMmb1r9
z-e$O(!)P6M4=t_DjE2T&b+#!s>YHM5d}mCF>yLaB`X`D9pwGd9c(=|G19uIFcFb6)
zIMWSW>Va+_Cg4<GPqc9N!VvPdgY{I@FJFY>NsF-5G!mb(mtg#qDA<v=x_wcY#o8sH
zRy1sF7UIe7K=l1D6kRl-IAswA*Lw7I*n}aK`BfiM!m(^m6b3Ghfm><}eTOj^GCme7
z*T>@BKd}g$7>m&g1s;;ON%{2Y++K!R7nY(ecNzt52*<6+aO{;Z9Q`&M%@<F_;I>oY
zyJ!*?R=Hw8r*Tj^j>Dbt<M7Dd4Yu)a+`&Ex(ZeUfthX0;CV1m!j5mI@pNc#iUxdu|
zh4UR>%v?GRpTlM{kD0q>-UZ{HWf+<c4a4a8P?&8Bf$NPB6j_F1;^Sbn`yGTAS3{7n
zG6cbfA&C6XpIME5SbWJ3(;R%zn7lP3Z^OykGd3xR@~1}VSt9ygOoZF|B)kv>7DcjV
zCvPWm7XEEmA~F^wGUF%_7G;U(Gm_dO@^-s*5<NnR__{d}?O!G0z?DRN|CNYEwn<1i
zn}mhrZC!qja`qP%%v0d$8S1M_6nN1(89&J04b-kJBKz)ByWKF5J?3ia*_^n~aA^wm
z>`Fz~3(R(+UTdQ|6;|YJ+gB<0y9@PHRn$OJ-_>kE8qzuE&5BP$#@{P}F{_a2z6#$L
zu0m||DtH*ILUh9w@aoRTdok0R%wK<%TJ-gt)dwfTxKlFq>YOcbX7JC_1PqCZNAl=+
zJZKV+=FKB;!!r!;3a3HgH5CnpdZTN;7aXsSL&;Sa{AJ*RDR;(Vaq3uH@EXHCKcg{4
zF#_@A?d8YA5feQen;nMXzv4c4M&2I1)DguQ9S}y|PQPT!T)8ILncEo3yNzM^cU}Cp
zu8Umq_R%~mY;A0X$?NK1RDNyDKT`|4{Vfs7-L#G?EqK1Q;910idjT!6jLhA;g^VR{
zTX$k^#W?ni_mi<R%$dt?23tpSC^nd5TO)dH>8ou{{luQ}oYgX~CXu{d7tJ}3w+XyD
zo8Yk__vp1Ufv%S^a(IS*x8E3r@3~)_J*@%EnHWQ!K1wz~Pu@EYF^8ZBd0YK2U-Pf3
zWkUHMSyRgFruNJd$f}l)Ety9_-gXM7ua>p!`(wPX)UIT$$J}1BHkh@;+zMtXXvo+9
z^x|;4LOwBDLEWG8JkCZ_qqWkM_if8f8fj=kFU#Mo#kQy=a5Zxr*l*mzJOp#zKmRb>
z={#$ff0C&^y;>zf1E{~EW;QFBI|c@o%S~#G{X%r|Zl+H9hO)k0#rt+qnS>jbOB#8b
z*R@dU3|ETXHl-|ntCEdvIeT?w{?Z2Oj=nM5X?dxnbKmS|-v969mWu6(63$*VvOuAc
z2-jlzV2h=}+9J_yR!X*;QdEtV;;b!{8OI8wE*aZBvp_EFERbKD3OJw5mm$<jZ6a@%
z)S)M6no8D^o%ZZ0Z)m9&@cJKb_gb>1<-ZraR>`^-Dmn60C10MX<oY9(Y`CYAujlCN
z-KLUd`&F`cp-Ng0P)U=IN?G)ZEdN(2q5SiWA4=(DM(z($NnIDthM%itE$i8RN7WMN
zrjoQrN=clrl-sdNX>&;_tqoLC<C{tj4^qn?&U$A>70I%i8u`4mM7)l0Pmv`vsM?iE
zFYYLc%q)?#lgz(bqmfl)r^yTEW_@IyM}<b77V-b=8UMag!?vSFUd`3W$u`t;uhGf=
zx;i<0O)J;<u`S7UkBh}}g;`kftfQZQ)ksu&iR4}=;r^sj$>wuD`>j^Gc<AK#Wv$rE
z)=AH(atUI-W@&&{?sLb(dVj5~Ajh4%(<{vHZ`<V($&pgI#K+xHl}j6XbwlXq9j#)G
zP2QTaw!UJmm%$zR`}T%2hElDZWj#G^VVN90S|+NU{Jof6CZ)sJf1bg;6yD_$>%-h^
z{&_m{#TxpSi;%Zd!l>QS>%`HRk9B3<m;-C&(-qR;M}-78V6Vo78bO{jU+$@v7wb9i
z{ZK74`I;Hrgc>~d$A^@ciQ$uS`EiH!eqx23<SclF5%&;t&fJQ6gZK1R#jub4E1Ef8
z(^!{Zpnt7JmGowfU5C7_>-B&6q=UZvkx*&{*YWlC`d?;e9W_F2@^;@yWBQAXk?vuP
zN#tz{^7eizcLwtOoathYI(2Kq!LSiBel<p|88+Pg*Ahc-cEs|D-Jy-_j-tSxDADyt
zaT^EjJ#j#jbO&f3I3WG`K=h6pMjf*gy1#eBWCt&7em@_BPck!p^FoAr&xcuY1lAsp
zz?bLYyr0j8XlB9DC=icl1Y$pV+km{aDGJ5SvQV6<6^6)Xq0AVVi^t2uvHjj63@>5^
zz4c-Y^<9k5<%`jxLnPY9Mq<>JNc<NZ3150^1~IGtevf&~!i<2zJOXz{gk#&bx!6M9
z{(d<dOCzV^O_?`x7Ea>Mc30f(Fb>Da+aBa?#YI=tTr&=bqCC)c@FZv#O~Jsv-sG1z
z-0w~0tjz~^+<bBUwl9Wlo`xnx(=n_#5dYN=#^zDM$Qm1rypUitSsj8+heHswBm`mq
z1;bhufK$}+cwg|v-pRg*zdwx`#nUi3&li#XeXxK2aul$xY<+;e@IlP?n3aTWj}jpl
z60tuy35De7nv;p>Z%6*NNyHQKc074IX>THy98bg)UW0knJdy}CXP$wHiP(1|5xPr>
zSglON8#{9NND?-Zw`FGZX7yI!F?;WJr`XpoqPDpq{ZKuV(bbQdZ)&-(agM&kk~-yF
z=43u*Z`o}*?5J&XvP^+<{S>%zhl9f;&UGJ<nVq@U@NFs<_f3QIJZhN9+q$u7m_02G
zsqNFyxi$A3v;W+mHL)Ez?Ltk|$&1|k6wLWFeLpMLQnOCp{s?2%Nf&C#P3h6<r=V^<
z4iP!@ZznEA*Q{t*Sx4ZA@m!c+3+5h$$uRey4D0MkD5~X&K7-uR-k-Is_ehNXJ`fjQ
z3}mmj2lNAbAS<~$9>jOYu0P$7ey=%mTWv9YKvR6&XM+tLZSdo#HJVJZM%V65P@2;i
z3&`8KvCOL2QXih`dKkLMihG=_5W1xf>MyB-o$YJk_IgXSHnT)G?xnp4^7b(2vpdY;
zdcqtZ$Xnyl=9pZEJAUZxS+$wEV-IuWYncDG!3?qO%#dMWjtKJB_M9n_>zHD4Aw32g
zxeI=h3AQ?$z=^XVdjk{PcxlXRPVTO?WhU}J+*vx4IsDA6@E~tn75tx0=>X@YRVUd)
zo>?u2_<GmVPup2Xy%Txs%rkcw^DB&bz8`1AowVHZc8WfXn1dD4(Xvu3PgIKIlS)~U
z&8+|O3RzaH7nkdL*~T2-B@?MzO(7>aE4B9K+_Ve*rq1-jkhj~&+Yi*@j%2+zigjOM
zjbiZ}P%J;aiY2UpM&`}r{hs|zFYb=L!T#o{X{EC3LaE%~jP++7os==p|1Ft&y)HAI
zo)*e=6iUH=1=65yp#=RYlreLZ^qDAS9s9qEtx9>-K_wHuk}FZ%m&ZP5DErDw|0|W*
z2TR14yuB2_o+$gd9X_h$&KH%e*`yRt_J}t$REiaOd+ca|IFh%a=>^g-yg)9^EfAwE
z`J#K8FE{QNh<`JcboEt9$w8HDVJ(_Tc9tlZgFyxx#IX;3S|!P^RI-G;-9+9FC2xJm
zTaybau|2Mmj$2fcu$5mofd1Yd><_<JO2td1oVcfy>K96R`b{ZG<n31SR_Umg-A}1u
zCU3)7gEt$ck~a^P@;Oi`%{-MdYN=8jZYZVUW0lknp-zjMt^Y!cBw`%%s+hOo&Y$_w
z-4eO`p+thn+X!Lp>!^_$Q5qS@I(6SxjVO-sxz{kKl6jMb%xsP!XB$n{h-)ml%--~6
zKBfi#es>nLC_8Gzw2<1a^UTL0Z%Z1pxBNxJemQGh_M(5^E|rrOTDkg4D}k;$>3T{l
z?<eadne*Ox%#9xIPi+;myEKzZM18JAHgjL06Te6ElO;0z4bKPM3%s0j-LdS8oaH?B
z2Is~$tXa48*UR_1)JWB*7X1Y?!#E$#q3(JT&v>(2>SW3QoxGt|>$>R0nT$0gZ(mO>
zm&9r1^7pKAc}e#E@a8<Cflf}laW}+Et!!kLb(i1#ebbak+YP0%o4<4Yp_LwmrQ-gg
zO4`rhYorl%USIh8_f97fcX>Ww?|S{?a{1?)UT${guH^>QdQro@zfXlcU%_)gP`QK~
zv&LSfmt~d}vd*PKEVB6iHkI;naHU+PU-{@j?sWRZ*PACjbez|i)!?4q3g&IJHOB1T
z>_ay;#;j4sSfOQ|+r<Qz6+B08F~QEgrZ7r2L*Pm3B<tERhsg%tKQ%+)qn2n{zZGYv
zZ4lVEJ*IzakKb)Op_7{fBFWpFM-DjsY#^Fc4MfW|j<|Wr35)u9q50%_xZ@cC$8+I`
z{2Ye#%|W>PzdYMO>X<hM;7!piM1@U;=WBPAd~}Cd>nV86ndyh^L1^w5jIc=|XxS(f
zTjImvN8YBAw>FIyqvzzsxI^A9=}5mTdE4}2Bm!qg;<$Mv4p7guhrCU<n~BC}r{YKl
zAG8|cgE4{hheY~NpEQ*_!lvNiF;{f!<Vu!}!}>nsuz&10tUm9Gn_1&<a)}3~Cr?Jt
zB~zf8?S;mcQ*rCeRM;N$K`43q^_DMl**VF|n~uO=fmlc0<~QfO+ASE90)w&izd6Xe
zJO}Fba}Zb@fDhmO(W1~7igUgQC2u=han`up2Y&p%2HX4K>SF4=wkR-kp8|E7C!*Dv
z1pIiDz}-NJFy_4b82vDlk%)n|iLmGUE9}YB0BXE;CSuMp*0N-6cV4RwC*lSfY&|=X
z%;V=dorsAAiCEe^2@hBU-)CR>@c(+y|C^taw;MJpaEp2}&sz#ie#lJ9Ui3c?NJgX=
z{mxs+-G0os26MoeQKAmx-ous5&}5%?V!ae>n3#f;d(4xror=TRshH6)4V5Ff_o*NE
zAa>?Hifqn&leq^zItBYczLL>ezHfKW6f|1O`nfZ`SuK-M{)&1w9|iL|Id5;5fH{^4
z+({FUD;wf5D=r@AHY~;s&0MUCn2QG<o-oRqh`HM*qSZ+c#13)8^ONIo@PZ49uMEeD
z`NMH;Sr5))?a;id9kv{3fh8SUU{`K)c>1<LP>wBHv0r2O*#>b(m?=))o_cGI#kSVm
zGt>|#Mm0ph*#?l5dgyuE3U}kIU}$Xx>qb_X__Q|OCDcMCc{}#01-yzZaHD|*EYEWm
z%No{)yxmK^li^!4xZdK9ymB+F9%GJ9<n3k7X)VPJ_Fqh~^}Q*+)-cD$PiBZaV~QvA
z4xGAdf{Em<_cUf$*f1x3o-r=dfAGx72#JkqB5Owtq-qRsent)SwPY48{SZStu^-mT
z0P%**-8s$-+6k;>W0_s?h<&q?KeF}aAL+%pYmdv-a<RHvX4f!4Za}q|Q&(cozVCU9
zO7S>OpU6Y@$LN)O+oeM6e(NPYUN0Ft**`q@e>xnGGL38-qLCk$)pEPNQu>ZpO7H$k
z$%@LCjuUbvkIa3oWABz64zDhfUzQrN_S49sdc~5mpZ9n2cJo{IL3^<tBX1)zsISc^
z5ZhY?5@N2DYyTCJL4~s8LV<L7m?xo%JV~9CC*h8HVq%;p_6B(}WqY2?ADb(OB64MZ
z!#tUJhdX^9m&(H6LOEW)P+D9plodMGiP`_h%yia?&s~&aR!=FF1%+}@nJ*jn<jdVU
z1(H32+%3zOT?zRT%%343C0~kX=W{P^zRcREl2JERl0@FFj3w{L+rX=NoT=uCYsY+X
za?O_yYx3ofB|p#V0-11DCE42wWmy*ey}Y(?PVCTCC6&=ixt~`k-@g~if-_1PL*Dwb
zW}OArzU??8ZmO0(HR;E#P|2==Dj7RdDb{>${he8t^JDLkw|Cd`d+}#YJX0mZlT>0h
zi#b`Jip7pulHTN61o`IjivC?6W=<X~mh$-;v0K6Co3-q~XpP)tX611DdEb$@U&5Gy
z<)jf+7mXy-OFXcKMy5{ENFSby)^*TGbq;k|)MpK$=l2QENQ3x!7BO4P=sWdioCTYh
zYUL5<#XH%<U;012npvs*)rVRw>Y{)6mC7{svyUt&k&-hd^7>ebbYMSvDYadP29$^$
z`^_u&m&w32WnxF(zGFV;j?8koH-LFrpLDXgl}^&QKcXpp#3wIm<y$f9WLurgA#Z;&
zlQU)@_Z(9*{rgafcwXXMnqNCIsZ>f@Xr=B)tyI<5$<Z_XSbC!LTgcd7B@)UWz&`4_
z-Pt2IpIR#ad@7YQKXh`N{H|-l8koF2M&ACjPA3y*l}S|z`!m!KtmnSO>-8%{;wt2B
zUb!qeP0w?Oa+zOKFBL2FV#2xiO*iVLGwEM$$qbJnm7;TIe`s-q%!}5`c+Qip8*@ii
zU`@;=Z+!xdaITv%+^Mx1-ITejBbn3ujotxUL$uv!0_`DFG=6P{xtuw98(QFPP+j<%
z*T)fW_LY-)pBmqqTHN+9`qUmX>^q?*c{_=`EhKLPH}%K0eCE7!pYEo1z^?O7Xf~aG
zj4PAiJZv(q4V(h~FfZ;$n2dp^Cu5EOG|r(WLpNj?#$*mbubV><^?3x8^*yk$nFrRa
z@<z=(FW8*&!rAY>xUkkA^Nj-Vz%mf`ngqhlY&NXdPshSHeh44yjUIixv8T`rGlzO%
zR4*^Avh>E3jZ?9byfq-#29!=k-~u09GIhf3VNSR>+ZE2=T`=I03(N<&B4)ZP%FnrC
z%sy8vy2>1)n8|oLZwkyOc;Qr$HyV+*Z<BrCw80l8E&Ooynjd}_OvhHU8Ti$926t7>
zfJtHi{D%Z%#<@Ajs5uAD>A?ss2*9Nm{y3uXfqN5Q#JT&Ts*W!%rug6nSsU@+R0J(x
zt_FGgVXp$)>L%i#M*=>zNPw0#ZCbY^*m}of%8q!*hXh<`#$Gacd!ch8W|t%&VNU{1
zh9+QKSOU%^Glydz^D}29V&z~m`cNWz<&(DtNr-HegjJm7b|P={TTzGYpg_=MzKv60
z^I--4x=&5EWiskIFdJk-GQB+XK9jfKh1t{>nco#ejdL=!(nl5W|4onbXV$;0Q@v}Y
zViV`Ooi3-K_HfR4`;e=)DY%%q9EaL6|I(fr*Oko8p26G@>a|Ywr{AYTGFG)sMpZq|
zjr|n3KUIO3jTI;*Z?o$pAm(vA8m(szCi}^I$lHCN=OUK8jUVob;Tt9*x7I{N<$1ur
zyBpeW7>}4$uCV<v8oNtIVTQ6De!jLt?WuNHOWsOb?(w<Q9NT9#$JG(GxF6pX<)3Wu
zj<ecE&23=vj=SOAt<mfrch%Og!Zs^<(+4pN=8P5YMOk6{+B#@Br8aj<*G6~pwqmaZ
z<~(B_)DH_d*0w-n`g;DN-s#A8&SxXcQJXsdyi;a4P{Ld&*0F7k*iWXPw!4EF-oG+M
z;B`~{3^&F1OlG0Go4|QKwG_9F;C9T2dkvWb%N=6+X2$qj$vxU%46*G4b;tc`VE;qz
zcZoK@Uo)7kIKTjpJNzGSr;wwMj#f+dV!gCwRzzp^bgP-&^fvmBTws2{Ek9<?H{ncn
zCv*HgDr8bqdT#ugGhtsL1y<Y-_fRjF<Md)@%iS|Cs7;}UsY?|7o%Dq4qVDwsdy+Y$
znZZ!V8D$%#{QgoX4{sNUZe*^6d*m`>Ay-Z&<cag40;ztLFMX7`a`#%UL~iH%U+2iJ
zt2uJhHBX+WC?z99DJJCX_f-Y*gr6h8fSe(3|E5mZ`4s!T<gGn<8%y3kcgT~^np{~`
zoGZgab7k3}TtQf_v>@TVyz&Hl^W^A?d~sP>ARGG^$hV#aqFhiQJ$*Tkr4G5(NOCx>
zKvq}eN$HI|`4pcgd-vywE+<c{%=6_?biPdY%oiKKe0en^Uxxay-eqsubx6MCROd<k
z&OAwU$deYA^Cabco*c2ym-D0YC6BzVSC}sodlbmofrS!5=3m%fD1XgWNq6%0ZWonQ
zcPy0p{CRHhXR|$`l#Yj#@;F;5@qL-C!5;N8^0s}tN+!-#is3G$Oq`*Vg>A{`!AkkE
zh-U@zc1?*=CJ$A~P4f0ZvP#;K&l&BDq_DI|V(8iJb*)Gg<n5x~WNkvR+$K|-r)eaG
zHS4FL)J`qr&R$QAoEf2!e?!RKF&ddj-X>A=Y^yA09mjq%nLFE#na<42>al}8ir?%}
z__N+ErXPvCHMzsLzSLU{DV13AwjOzVEk-M<wI%ZDHkmz#y?maj2ES)^W<QOz?yZsH
zFU6AnSBadQS0cSlxEtc4P7;GSqwlJhgBx}7+PO^JyK;_yKqqzlby97oV?T!4tlwI3
zrnmYg=fr+bwX&fm`yf|XL-*E+|NK(1AU|6iEs;xwCCt|@mGd`CWr%2HK^%2s>$T$R
ztd;n>+yiA&D$@^_NHw+DIozFit2NJfS*3EEY`;>flbtoo<jMxEM3C)4-t;;p&}*hE
zlT!wIX%(cG!_4L0vq&qC>7}}Jyi6AJx;D2=y0$Enll=3vT=uPpluI;sRIP}mzMlT7
zW5E^jHIf?qf9Pp$L~lBGTIRSIqI(A;?54K49(g;Zp)vA33}NnWh(}Eg;X&qh{END+
zpY-v~GQ!6=?iD0&)iL#9^lxLN*i%a~lpg=%EwQLgE9@H35p^^jF{@=Kd>`2fJL`7A
zm(x9A_o65Lg}t%1O+U`^9C6tQ80j|@CELfqvAHwOH+JTJl;K!0YAD>s42FGsX4tMC
zg3mXG@H{XSbL)(RFPU5U#T|8Dc_5qjS4WLI=f0lE@%KdC8k49`nS>>;reN@z=@>)a
zHgfi64dD%+94}}VOopq%1L`#%s80>oiBaRR*I_(1u5?1}E>2MF9fO*qoUrMbGc$f%
zFzJp9LfX2bue&R3V(D?(=8EIgQ4I^2$c&4LFljXzV}E;N(2l8CPIecw#%=k|2k+EA
zP_^_$gX(DrByW9(20&9Dz|72G7|xi1jc2E$pZRq3_~eH!HK)U<*avqT`l7F`FWkx7
z)B`>+@9Kj=tigST#$jD(9CUW^_<TJcp2O%>YMOvI2?<ztBLNQM<I#(}edwG3pON%4
zpH9GZ_Lb+T=?&hTfQ(rQ=r=O~O+&bMeRl%(z96sPCLq5k0h7q=8K*eMHKwMj1+$uZ
zCBes#9wG8}w2J~t*14Bg(3i|Q_nH~C;_NGbb7r<kW-@%4JJE-a={b~qbyMIoXXH7o
z!9&U0vmfaTK9&N@vne>v-mu!2T57WP1?$(7Bf0A-IvIz_)<&P`)p|+K^?UOCqXNU*
zB*TQfO(k!Cyis5awd980)K{|)K4f|#ZkNSFTNI7sXBX3x7J`w}gYdRi5VEcX;?M9v
z?27e8caw?e)Y=1qpPb-0&;kE;=#Ge{-SDD!SDbFg9UK?A*SW+VT??39+PnpV4>yN1
zdAkd?SZQd3-OOE_6=cn<G;93$rwJC6Hbzgs#&Be2MF;Y>b9XBg3}g?Pb!{!fI+#h`
zW|6n29#|lDg9X}5w!nn`7BH}|;Lc-noZG`psAO|Ao@b5?FU>H1uNg+|<KEgyW>~8-
z#kJ3-7~Ibc$^Xy;f0_DY?$Hd+Fu~pP#<21-M*QDK=y%Wv-kXf@c#094T5#XVW@g+p
z=S~!QdOnttshkOglef0yZGZB1*6eC&=vOWE9hqA(v|2pBl#9-~LjGdysyeKf2=?e|
zP#gRYd23GIF4)a`19_{Vzjgcg3Q2CwEI-o<X>*S|@2GcrSgS%F@IJVlS%~(>OQq9L
zwIo*-O2yj(x$VO%J72ae&X+dda;0@Zu9SVq5r3Opne3h`*1Q_9{#-XaSB~Z6%GIm6
zvOFzUIt|W|`nz+a>TRxk3Cfq)W(Cs5vp~{WW0uUJABVY$`YVNUr@TNqB^Jo&U1aF-
z9C<OCA2U2h%KdUgmgUHkyE(F?buN3dxe{!YD{K4aNl)_jcu2mq3(J=#`}3vhdA?kJ
zm@kr_F9#PD$jcFh(sgTrxRLEk$y?VYd9wRJo=iHFCvEcb#CT@DykXrMm!2mtH|2?$
zGLM;s`SP`AzU(1yLp$ZkKQ;0sYf7%nOk>Ucg*A41zVzku5X$<t@Av}oB5z~-3dzty
zc``sL$IcXTPeGxW<QB*v_OS0vDCFLSLTOIF(W1jjX~KLft3Bi{XVf{pl@j8ilsy~B
z*p^CpW2BVg`btToW;*d{ft3GJ%1ze95iZnqT`!QaWVadn-L39Y)5V;Ot;~fsW`@?H
z@*<fquUIC!Xyhkzuk7jD4JP;6g==Iqc^gUI4jsllfZUgm%_~!1EMHy~Oa6yqS=v)0
z7wt82?*Vr*SQN{;mdxjxRxC%ZGdF};maaF8CC{rwzLRh5$vj*3l{<efl|#-Y^7ev8
z4y`E`Z>wTS%4R)G-qz&TJxJcxA#dmWkGEQCtN+l?)S*-wvtEty(aEYVtasb%WLC0H
zrg5e)%A6VjtyWIo(n^OF%*Z;&EECRgNBp5q>kNC#<n5$+rJ^K1D_FPo&tZ?+r&Nwc
zmx`qwJyFcnI?kSnU02rK>?e;UZ#Nw%ktWQeR+Lk(&7Sz6)#UZIQptM5GbC$ur;%E@
z&abm@gidUe%H*i3Ocw3nPKM{4U9%SxdY*fX9m{0cK%I2l#2hpBt!L`#i(+1P7iF15
z3AOeS<+5&Rg&5Ia^_w~0`8-ctc%>KBS2COR@g_HBaI`c+PJctxx>OT2U)IF_eunta
z$q;v6)I@DESGnF0y>4*7AbH!aVO=ztT_4Ni8{zMUO%U_I1{cZOd0Sgzb6!jAY}*PA
z-n7Dgv)0V{Z4H~dt+Da%Hi(RGhcz2Jz<W;zj62?e`crB_qx*5^%pla(4}w|V5Ezag
zieU0KjJ$2R3e2h+gh5Rl(X!oOTwOF2uImLxABH09FxNz{AB?u!m~+)_D6?&x;9uyB
zic8Lj9yA(T2aJbP#W*zR>x$^=v9NhQ7X4LYxu0bm4mgj++Rx6|c5gHycaKKXy`wP6
zeKhWD8-uBJov>uPGnS^iV4u4S8jo~=W`ql}YPg{N7FSFO8V8jRdGG6vxn10$+3k+0
z@l(;NlMn9q^+C_qK6qN_gG0@HQNZj#d+!-&J~9Bk8qUJ}`98Q9<cr<!d~sg<uy%<b
zuC4S#TMr-Dz4V6fU@zpFc)_y27y8Wf;%wLpaSh|}D2zE&t>cm96OX&O@fejIj{`38
zsGA#yoU!rvvL~K;p9HKN&Kl4!0ZtPW(4N`U#iJ9j(LVt;<ZW<x0!Hsjz?FX!=<6nv
zRS76JPsGojiHI^vf|*ScJVqs9&gLX6=Ztg&S(-zCt^;}d`X2={4d~<TN=7nIM1M(v
zIM%d%`IymzS$~eF9*w*mPu_0IW!@!u+i+(Jo^DG)SSD-QF)0Y_or03X+(Z0>-s(u!
zv#gb0ld+MzxwCK<HP+NmpBk2gJG`ztCn1r%b>ET*18TZ1lee>s67e~Y`+nX>q5FkJ
z=&W7{ujdO;@J|p7t_EV_s6fo@90=Lyh5KhmA++NFEE&Kai>@EaT)SaaudaAb-a3c2
zM-2!1YDe2c!#T*&{mr2~)eLtU+G2PO8*CzPhXh+=R~u`1lDEa`#;D=Z7}L5~;8QO9
z!zy#kR+{5`jyZaAwo>c9ITB`AV1AedPAnj2$=iqhEO5!l0=F~F(WbLGtQ#=<*o>JK
z+%r(N%M6!ynjzoQ43XsRo+s2pv3G39+Sc>DDOLwEi>9jy+SyXS$l6wMf;lz^xW{v&
z5xP$>!Z;K1maoNzZrm?W#-8y-1N=@iK%egXvm-Uh4hEPT$UUdis-<<~YRO@L?sb!L
z*>OZC<Ej71e^@U5<ZTwS=FeU!7pMC=*_2Wy2Va-VFlO#v99$vVn&j;j`Ztd2seRE)
zFKSv=?No{4Ws$hq7t8R$YI#Qe(!o0g(we-@Bxm;Z%a>nIb0xMzu8cmFBQ?L|$kcl|
zvi4Dqc=pScHdVi6*_+?e;Ph{)e(+nix6BdsjvV<x-hO7!wjFt!JAwV&;RWJgTp*WC
z^5ygST=9%02Y-^^IXU9BIY%~-w?V^lBw$*OjE>Hc%8VST_asL)n&irksvP;M%8^lg
z>{Naq+ECW08ToRDES$7FUrMj!%N*ANVYrvHH7S(W-|}Q|R-WX0<Vj&gt{6?plgD54
zWDD!rey{U5%gvKPd-7z~`8=7>IbXi;?e;dTsej~%{o-8FPtKKEqjKd!Os<%&%9Z+G
za;1vrg7IhbWPx?Q)Vxt3vl|ym*04gE-j;PP&kp-|M#*5kKY`5O`@T>V`GsQHh4a}H
ztYLR6Wd)QH@Uc*|W=g5QUMWuOdpGCVrbAt&xbbs1vbXLV$+>d#0<j=(XOOoU-RN1G
zrj(MVDjE1wEpC3?g}@B0t;~u3?prL&J&UE&?PAfqE0()$xM#2x=ejL5;=!6WqL)U}
z$=Wp5uhU+V$GqB-xqg$2#p(^SC&})&oCRca25?~lvm~z;OJr)XjJQxN3m@|1Ta?P@
z@KRZy!rYB}^c#&Tkw47OdKp*r|93l#<gVWq8;WHm``Q~n7E3`R4f_n-J@`W-&#N@j
z*`KrAjV1D#xtrDhmdIDmd+#<ZWk0=C`uwEcioE^Gx$tEjGo#OPhI^MY;ymiJ6s6+4
zqC}pO`P=Djx<cMo{w!gpcBy#Mcf7ZrRz57DFL*WcwaDA*t>me3sg!wATgSdkteQXb
z-V&LVSSll^<8Jbk`T>5;uMuQ!yjF%%qm{r~f6Qg>Y=TC@A5hm#-d;&7m1apLa+$0j
zZ^e&u;!K+T@gsiB6|JN{&9Ph#OlK~~GU~SIcPf2QE;Fc$UiPg_=AC0!3%zt*UsFe|
z`Xi||xm%08t-e<iue&j8CWL#D7S%w3TTM)2?p5@26AWv@Jz?K!A^T`U9NyUo?&NJB
zqb4w2ZH*NttYO5xV`G)p*z9V9CL3&Uc4||ck8X<F%h+>nZ-bUcthtN2H5!+;MM_2&
zyua8NJ6iUGOH4n^{5$~Nv;z<@e;~Gfq_-$(5Z2@kLRei#^lasb-Yw++k#yE^QSM#W
zM;cMV094vS#Y9rw3sg`P5EK+lu<O`LO55Gtt;g;@c6YY|Gs6TOih}sA`+fenJ|j3Y
z3@G#c?Y-Ap8yA-dN*|HA8-S<le4$U?wmRwy^)WxZa_ok^9jJ}*?uEnK{E-?RfWMFX
z;KIWoOnJ#1`j0`Fc|8aw)J?A@Z<lQhz(z-Z^fv8@3sJqWI4%&e7QHcVT_0302}VQD
zV1!fzVR;dGT^WRLQ$nyNC<HE{A$a{b1m{nO!su{6jJAxxSmS|+7&;JBwhZJ>xk%<}
z4no}HAxK?047!m+nbS8Ed!`J<=Vn7O{lg$sZXb;8anaZ?FB-uZ3X_+E(Z**G(#hKs
z?0I*L9s~>WwsKJ-=FCk(9`~hB>63ynL8-7!OhcMy8u!Yk!J2uH=jW$kuuD2#w@b&J
zgK5yUOoRUDG)!ogj``&6zQ}ZBk-4s0xgX+OI;NdWhw-d*RF<WqfV|Bx$b|j>bXTk;
zt2bmqL4T0aiaqRU+<8ylZof7c85Nx4a<00bwdb1)%nLDNUX?TRMcT06P2T<~$wca|
zx%jkmJ{n%1k8|4>pnCoS-141|8Rhd(<MupMPoeH9f_o;~^Ev)kFO#<u$=fAq)IpQC
zEy>z79vN83Ye!fH0?FHwuNi2whZlK!l)O#)kbqC;X5)9|EX>l(geiIZ@bXx^nm86e
zZ;ZjOP2FL0)(fWwdf^*+yQ0Dqx5(RE@^);1D<<A^LDhlg2uo>-nSRbVnQV_m7wmBS
zlr3&J(?|P=d*R93E-no*_?itib+SO&U~_DD<vxMa)DGvHp*wlIiM$P`7V6VSGxYyq
zh6<~iP)sLt2i8Phhng_ZtBEx7wtfpUbU0*+Q><&JY%_)ZCf2v3O;OgLJAMjGph-1B
zdOv0#A7s8Nd0T_N_eo_&$Tl&;2{wSvkhAYX4AHHjA=V@_|Aw<4y_xzb39Ny=N<Ex9
zr^g&t>Z2T(f!u`|QRMCOftAvOT9QdW=);KqCD)o&NbSSSTI%vc{KCuS_AkCi8nVyk
zK|LvZb_dAYHstNP8r+%MwNk9e+g+SlMI55HhrW=XYktU*{>=JcULc>F@coCp^>=0c
zNp8Hb$&+2rIInfr$%n7(>)L6={DfMzvBuSBzxX<t`m(KBzE<YS2t}^UJdrD9cXOq%
zp<4d)QH$zbuEZGTvZtq#f&EodzgwQvA#V#sC$-4iZXp_3o}-p2>(x?&^=zEKT6XfW
zpCWT5H!W9&1gNEQfm*Ce`1Q)wGN(W-r*hSj<Dikcp6n4XAS176<)D>LsxIoJDS7*C
zTAmDX)X9iiI!R&;J7t?j@_%V0K2j?~KWZf@R3~fSkiG0(Pus0!-&ZU1&!}bW0iGRh
ztEClrJIF>OHC;F}X5D_4yd6#6mQ82BoAYFA^7h^ioh&#`wlhDYPq%#8?2<3NbMj;i
zwOPUY+2c;-Eq{Jjvh&5UMS;9yzqmhZV9T}ab(`nQIO77bq(>>OPJvWYH&&ESZI^St
zc<?z}2U0^_qLYeVc~W$W^X{wp(rR>p6xA)1m(0%CO^@-e1;z6HH+TDmQd>+;Zd}K{
z@{<y|N#1p~FO@drtv-2cP2TQl&Kj4yj`ni@!m+0%@|yi(3qGdkU9ns{S4=+)JxaB?
z!-zWXuze*mVjekkm^&6&yPm95E)AQO%XseQ{ofAZLhcZ@@uwb~Ssml)Y5L0NDu^wS
z276iClD*3~AAUEOz2;q|lFPln%7vxUdVQ&!-%QWanKJ42f_c}}Q_bNV`Y~s|F;S)R
zJb-m@PMP?|l!|I|x!gSULlRQxUn2LEVP&Gv`S3vU_V)KOnL^%9-B&KF*gqaL`G@pk
zPdSlw_p?2$gAK~XHKI&*=atH+Le}%2+3)6`8+B(-nKgQY_;MM|=QU-2-1r#tLCM>t
zN$i2QAk$g5H((w8i%fr?S|(dKA700KabzHC^*Q7%XU&}h=sTKGA*su$jV5o;tYg;m
ziC^L!^iv)%(>jfwq>t2s`Fi}81ZIA`s$D5J)Rm%FOAqO-xU+Dg9*o=8z<@w~fV@p0
zZ<8&}k^Zt42KB3r`?0kVw6!*VIM#uCM`r2tt&i5a`lw-Lg|(fmm<M46_XF0rNj>w-
z4mQ|7*9PnMHbn4-CI~QU0mJbv@I9ji3f}W}y9@3fY>R}=9>^zmUuAd368|m;{o(_&
zbMENbsS|AVI$^h2JIv|jif;#9F-pHZ^xAjA_i>$2cdR=u9(PBLEHBtx?TphMJ+X$F
z^Hw?^oSh?>{k03;E4rZaLl^ko?Tpd$JEQ5-&hR-794+y~c86Z*9~OuUroAzfHSUf@
z!B8#gg}^4g(D80h%;n$R`$b;~d0Xfgf`cbQFuqY=&eHm#Tl)ySC2uE24@A3v2jVbi
ziOyXIAc4NmHS-2RZ)X%UIHS-cIf&YpAlxaBM9*RDL#GZVn+KzI=b@;TJ{V7^+m3q{
zg}i9yYYdFS4qipg6H((vBC^QaakEo!+Bp?<jZ+awuT$)q6x2SQ0z2k`bb2}m$IIv7
z#`k0_Y@Ld$S*ciR!CKafeP;63H9Q>)wxr|i-gL}9l#VtdINz<GfjwTF3BO9msPc5Y
z3Fj{2*h~acTjfRlv)9cGW>sgR@G|SrlDQbh+IA0fGk;UNtai$T74K6)y;YOZndtl>
z6U$c4g~K-bbUEJ*yh%;hCf3yB(y^wHkK>++o9Xjl$2n~Td3%t&Z9v{;9ptPyV=jg;
z?`mo1OzuI?Ksb4Osy)9xy*`avrQtHq7T1##;g>iCCCtLqTRa)bNt3BBACDoohoaqr
zAvn`&2zsZ5!nS1)F4=a*;67eha>o--+&$s#>y9yzZK3|`3jHrG(ClxHvec%i=g}17
z>>V-Vp&jNii}9zYE#_Wt4C~<yq4Q~onxR(c__j7aw6BFe|KlzFu^p>fo3d8j(9R4l
zA!g`B50AxvX7Hmv>OOO#9F3TJ%uM7*)n+)o%nXC-m|^}wQ}ozjiU#EE>=n#l<X)dP
z^-M7$%LJ#%TjLNDjK6P;s6Iw8pjTe6#0bv~jj-;bA-b$!E_yHS10!#jP38Wu5e5i$
z)yHS@Hj}(Peu}*1@9NKM+^ts59i=^a&HXJc$lDq8+U!cJkZR`&xx1VFudYAjTu;^x
z>*?i)E0d{Ps9pJ8F7fP-T_Bf*z2+!tFK1W%mWN+{N+jzG_rh}7$vVq{nw@=|eV#g*
zFWatj{>VAxAJ(u>Qq;1yomxiJ;Z>3=)#Tq_vT<;gO5T^M#NY;RkE+CQvr690P|4rM
zDoMRb_Ufr+b~ClKznm*Yzf>~!wMwQHDy7MCg=D8Fq|Q=>+&5B4Xs|{GU*vP<sHAF~
zT85Li)z;+dfLwXsJ6B%p$d&aI$l6i4QfpPNWU_u<@K!CG$U|>;t$ZSH=aRQg$lEF8
zt$UL^iL}a-`M=mNcI4mvX6?O+tmL)cSt}RGU)KuumoIANFlWmO&Z3X5BHww2=)X@b
zXKtybguGQ+^K8*VBX`<sWW^8-=Xe^KS;q6h7;30Cux38bdU_wdMizOZxTTYm_jR&<
zpH9xD>bTc1Pd=vPaqmH%_^I=yiaq1+>^aXqmM{KSsKtIqPWNUl+b3U^u&=)6PQEOn
zS7}AxJW2SjlRBJD7qd5BHjihAGkK!?tdj!P&i@)0%7Y$7(!#w+G-Tas^7c*N61l^(
zl5%y4+#qjNCZ$rxj<ZzGW=FbGyWNDot%jvCpS*o=yM*)h5;@8&k2%yyU)@n83X5VH
zyr26WPIAwK7xh=n4tbqcBHgx?$REyo%h)46!ufE$q4Xs2IbGO;wj*!5@ar$8PHx}z
zLdm6u`^6sipxM(tMxFDh)}=CwbK}kA?ML$V5WQocIWzvo*U(N`A~vNZ;=x^pX^qHQ
zzSagpHeM^0v(!u1*u<<4hZ5OJFVtIRaUIMom7(+%XR=rRwpWR?;u%0gO;$^uDTA3S
z9YF5Z{ah}32g~ILYh+KiGHLOI{q%xT@%&IK4>g=ya|XSdyd4u)F8b8DPh*`MMc?+k
zLhgDRM-5ryGMU9XHu68}%j)yC1NC0rxa-)qOcr(|@7W*!!OYO5>_Kbk`5n$a^godm
zWHfgbaz?yr5cAJ!a3;N-yPU||qQ<{vk3DN%vr1{8VP=}S9<2F0ylFi5h3sMOTMvEM
zhU?@0TYW@5<30sLLzER#)BKA%D{3ge4q#@whY7}|n_$^96Erb2#f+||P&PEf^N^Zw
zT~rgVhMS|NBXf2})I!qMdQf(=M{P6i3WWp4Ja&L}Q%5{jJ7KR;6L=Ro;{2}Wu*z+P
zglt!MaQ@bDYzN?3Tbxz3hW$Jz3{G-|)BAQ9!}(lSrw*7I(Gg1$I-+1+NBlYJ0X<)D
zROFJ;Lp@M&*%RO2cR}LnE{NLM8E&^bW3qHcrospHNBZJ?t{-M_Zv5m$FC-cFM&F4&
zF)}&`PZ#z=E$3cne!3^-`353)KyP>r>jOV}IotGM{TmSsZ+eDWIQK(rP(R#`>xT}8
zoa=TT0OKwLpn4X8iPa$(@F@h#Syy``2I1$JAZ9!bz~zVmcuVi@ssHiThk3qJ2IDIG
z$d8^x;q34z#6(14F?k#2kjUL?%m67$!X@7n1em5`YIO?hx)iJ=YwI0Oq3>!AbCu>`
za@iaVxSxy#m*!wbyJTiCBx9;)It~uxZit$k_m-uhymdN`Q%5y+cskU**?*psj%Io3
z7)joiuF8OhUfnS2m^YozfH!%&l)OE8d@j12nTz~;%(Kd&hUzHuLmFq&Yr=eK@^;2B
zdV<d}C$iC8+(^#A31r}hcLp=KIV+A~-A(4+VSQ~(?UqBYbbMhQ-r-srZX8a-y{I(&
zH)0l)&1Rt;z1GuoGqLZ~O#U0qKvMb)%xF6u<xi$z@VRLimM{ggmQF@w^HH#{j>hE5
z;ZTjCzHC?+jy39ygE;}@S^!3D_s5XJ9;oc#ffySPjQq|2C+%C~^}VKuEU`n+1$G$2
zy#fmt*&$hL2OO}2>WD4Qde}0DsxdlkX^55Q>ceecJ(!TU-o0yL@owq_+nZtXA5*+D
zF~bFCGu)>~?pTl+_Wxsw>W!w*8_T?ka%LizFa!Ce8TPQ2U9i#=3+J1n7c-Aj`8!eY
z#{{o5>>EEcK@e-&)d|%72brMo9<x>hjWFW25qA;NBamx|y(hWbdNAiaWd^u?#sFbc
z4Dk3gGuK_Xi{P~$uCtfyLEc(6;tss~f8-#&{z3HaT+aM0A9ntdHXkbFerkmj(euA{
z%TH+@&AyxyXKkB!UT@6|K)z=bRxlgE_orOAM6cog-_qBuQp|q*mT2Y+G~|25ySHVs
z?r(vde3&mirZT^uyj^}qC(rBXq(K|jj^yp@Hfs4$OD$)$xl&5*UgT`nhP+KW!Fu(i
zO8)FuiF%`onFLB%P2P4Llq*Mm=Su4)YME72CA+&TWvQW3JZ~$cYMDY#B`YN9Ne*|V
z=FrQekbY$K;uk7eF<!;Yf2Aa(lI#6)rH)D^F&A@XsEbNolD93$+mYmL$Jc7<vs^8a
z{wn#Wok}MCR?5Z(tOw6<27FE{Yj0|0>?5rhI&n7pM<Xt4HR3jzk3Etr5%C(aEY^w-
zd0X0CE4TY<<lscLOy8)Mb$iuP>jv4*{`TVroJlw5ytti4KKg3pFnK$jyp1AnU1PN}
zhrQ_k9#Y?|pC?<c>tqvIU%g8wVJSM<OdVJJggmk54EYE%tj4g0b@XOda||;i*;7A7
zz1aQ0d`aM$=G~(_S-&A)vdG&)*8Zy==;YZ9om^4y=T8k<-c9POKT=CQF`qScfeg7^
zAPavLNLS8xKNuIu*QUi1!#xK<i%MiA_ZMxcERjX*D@VR3OKy~i(by8n|6DAMBe}<c
zJBluvFiWziNVbu;&fA!aRmqI!y~Q$xdjpNhxlz1-d18tDCinVs4`H7ZWL!gPmp3!p
zxj+5Tnj*Q$?8%?ZhyGPUPt(Ohv0&y$!7k2wA5dFYR3gn-Tc>v}6&tdCSM5>>%PW!H
z-?&SHymcmTPv(;QHA^Lyyj@A&*5TKf`iebja(s6fbGHVP|72~ql|@oOuh*pwh1@+<
zAd_$A%gV+Da*I9rb#K`x|4t8-S()r0FZZ~z_ssg+vTm7t>_P2VeyN;#LtpZ@QW^7`
z9Ca*{3ihQ3{3;WRoO13${2`S*7fc*eB11XPUd6LT%gfYjlh5JyWz<lY$!tDv>t<zg
zbP9XuthG<@@kt9=-!Cnb@;hbn#jad_F@MCjAM>oKtuFlgQ;yf*-i7jC%ueC&+AsQ_
zG|Y$o{zs<2q+c&a50fMR$S!)+2c~cbCUqEVC+nf&odFDW21p=pBgoq~<n5PECioN1
z3`owSsto8Gf+^>|W;jmXwj*!d-OQ2cR1335)`EQ?wI?bI+<s6Wajtea^Vtsm)Qg5K
zbA<abM{I53jBPKRG5WYOnr(Mxz37a0&72XtgPPKH4hZ#kgyj)OEYdl`;0<*z{hToQ
zu@jb0Z31;y7gWc&;LnCuF#ppIUaQ-~=v#aI?brdPx9JDz)Dgw=z0uv-2R09U@M4iK
zs+xDhrtLjY%RCSc|M}x{kUuV*?2fHXyJOA~UmV`u6IBSrnTXzaMBZv|a~4eAjt&pT
zoz}sql^6_-Di|+nQ`gMeHz76{`^N<1^qAgwz9#@Xje}4c8HAC|`XDSb2wTY8%pn7q
z<H6jcz5_A(4|BI-26JEOAlzaG*STR)coi0fF@2+O>1YBp%+y*kDFJ!&xy$tqudd0k
zot%v0C&<(dDOfXZ4!*rl!pw{${=AaVd;A;>A#ZJmC9_6L!R&tY5M52fX!5pR$8^kH
znU2st>FC#k%q5GJ$?2S9rlWd71|Cn#U<P#t;#lL>)aOoso0%~2=4_U|;`KM^NqRaL
zEjH2%OfS*GHq5@Z&BV$fnW(dabKssCC{~iMN$D`3kq-SS>G(*le{g0^ZOi@d!_$~I
zk%l$HQW3i}4T0-;k+=54XW=Y)yWVCNGf-wS|6(S(JI%n#&(m?Z&2*SNnuZmM5x5vN
z1WApCV5DL&a$oSQvUeaZHjaRWNjR?91VXj7CtjHaAm*_b2G#3`H*U;BR=DD~xhuxJ
zY>vx;PWaoz2_xs)p~YicgpaU7_u+N1+p{jd{<g&Ns}`8^Umes<uY;OrEbw#$cMUL;
z@p@-7z{Lzl9nA2s$`l`r&9Iid4WftkK!PbWlbA~pY>LFzrYN*A#SY$9Fb6eettnbB
zGDSjjQz-sY7xmi&W2i$;J<DAoD@-tJ4*SXRCb&)B)_r1(3~wVu_A)|vFm=h~ZTK<v
zlLr|>L*BMbH9(ia+|7AHAE!8{&9!AV4te{HnK=3NnHSgNk8~VdC6AdIaD$p+mtEYO
zrmB!_%<->e?vj$3Ok=5as&88^-sxphw`rMt@}O^oe;3Dd`NszItJE))=G4AkT~5ud
zzg7m!)yQ+|evPRKy4sPw&V+oarXJV-gjNnvH#BI7R=!86Wi`VvUCh;Do|`Lv<gO|E
z#6DYc#r3pOri@ZZdu@(Pzmp@++AAd8SS91kbH(i_XSst_a<`F6wv)F#jFjSiLLv7)
z=g2mP9I?KcBOcFl#4Sc4W7n(1g1ilVr<A63R1z4YlJDUv$$hSp$9q-Mr<scWR+VTs
zsAMKtKKE^|s7I&;<SoeJ<79MoJ@&iFTQz&Vi^<z-jaVBSuoixrD`z-QZi`$wJ1$pD
z{Bz~GS*}DTsKly?R;mLv%n+t7i#6`KJ!<ZxV|}|!Et99JrR#itudW(-0BW@+Y2@@H
z4KqEo;=;ai^T}Fi?VuAE`ipvQ)Jnoyt-R$-`d6Gzp0Gx(7?&rWTR4NI7HTAUn@Atk
zw4wAc_0N;!Bx<Ke<%vmTo*0m~2W|5uU}T<HZy>LCYDGCsCx5@{xKCCmfk$+5;kZs}
z56YJf)OjzyN1stMYPuy~+A%w8zGJayvWsQOf)Y7!zl2$o+;z_xtrPpk+pbWL%=zyB
z{L@pJ7y0B7`FgKN)?cEx_h05+Z7!1U%&GizzDRty3u&okiM%3zGT9$ilDoY}Q4`gk
z^|PW_hGuXF#DHSCO4g?Hvin;^elj0~wQV{xKjzW@{D(Qxjp;9XX;CUW>XphC&XKp1
zx82EGu-E+VLJ2dPOQZ(5KA-1<d(>sMxK|>LkMgzic4C`i`G>k}?@mSXU>aW=^EGdg
zx68@fp_94a>0rLhE1_ST{_LalIxQk^e~`Ct8kI>l&zI9!1NSZ=^KD9{=xV97+)^r)
z<nrLMQhDP<pYVM4vn$G|8DlnA&<}}cul>u%VsRmF7xCwI|5B;k&MK89<aDzKrLtB{
z4H)~<(^)V3k+%oQTL!XA-Np1LmzK#2{c>r>s~=~=1B$8h3S|A<<fk0sIXKFK-n!er
zCCiI>(b?Py!hEY!<gGrlBa3-<4rg9QC-OFmpR=B#k5(V4Yj)?Eo3p0H=ZxW{HpU|I
zb_jVZHB7OHylq3?29vj;<ZUBn0JwZJhxw6OD12an^n;dYvfc_#x2&-Fg$?2hZD2;;
zK04}%_=Aq9d&rUVGe_!yo#Er|jJxFRDDu`k%>mJ#%#wU*kD?ZiSaZn{Gs)X^3!0$8
z9`62r?SjS{7wo9jmb(nwp?XAnIIL=qe_M3G?5iD7H_ns0^t^Gep%4AHKDhkO8+x?>
zk4t_SG1?yuulb|e&mV>dyW@P5?zsG~FPe?*j=i(U-lyI1;871Wy%2<V&w^05ItZ!c
zZF}mf6w88PyCoP;b_HVuc{_S^Fm{l)*{cK4WFHxz3xKC>A2g%JIr&ZqS}qU6?(GAx
zntE#gEm1IR$GNTLAly7oUvIA{qy$BwA$j}n`UL!nNr3nG1lCCj7<w)NzKS`ho|KFh
z<n7Cp6#N@AhdYLoP%|wFjh7|i%eEvWH%P{sdMTKe&ON)$)8ITk4Hf;<kT@(2Th5Vl
z9@JC$vF;>qFE6JKiM;(gg8k)xGO#Qo17~(+AeK6+;EYW4q9?asU>g6;(y*^64HJXY
z(P?`+mNj7x^PhAC{7%Qq%5>~vuX>@DU+W?@T}#q1WI!6)UFS{*veu4Q0KayLdm4fq
z=wGr=Marr)eBkRE7deZ$)6<bKdo*lYkH+|*QCM9)0(<DET00^JC;COB@ahm84H$wU
z%q7eEHW-?k1G(dT0BqJoV5?C-JRci|=qLW@u)G^C-t<Im^0u+69iklDViS4m8{&vV
z!45du!yduY?cjdX7F#|y!ic%n&{7YzKvN&SJ9!QyZ_N|yV5UiJWc^zUSp#dKCwY6=
zp0nFp<ZGoVdh7EXxx^G!OIf?dn?fd<!rk8#H(QxvFL_&0WP**qsDo-o4Ki7LojDfG
zO-%9irwLxLt}Q)dg5k?eP@9~sy@@#%SLmsoW`tv&%wTP0gjvOgcnm`f=|Yd5oguu+
zxl1g?0Q!*z(C*N~AP@cyHma6!6aUDg4a}^e?johzA8AJ37BlZZntngCq+c@bQ-%0t
zQ=e%}e*?1#e2+2<ky;ehv@%gSa|YFlxqK^%Wb}9XQpnm#7KQRPQ7e(=)a+_B;=Y2u
z)-f7+HcKn!(^-?nb7o84I^5Pu&-SeIN^@l|bx%JUt0l}#ElZW`6Z@#-{3@kntWe6!
zV@g@&t&qE#9OeY)h*>KIxvP{bD^xOPhf3}YQpqj{m2C1;%BYVDiCn3WLDmYH-<W*v
zog@3nXzu`pl=o3clYvTUJz7aUi&F0IRZ7oLmBc+!$pZ5BnGN|nTqPCkTZa$Ml~=EG
zWkRG{LLTyQ*A;R|sgzf5m9qAnQu<6%O0#iFsc}^$OJ1tv={45KV^!it#x5ks4Z7vZ
zK+{~=AFq-(>r}Gpf=cGi&6O*zxzdxh^D@@CRpf2>P_;yuswCb-B~!il_<1V%kN3?k
zQOW3fd@jy@{cG!F&^oeqiB@KXX=RgvPNq%Q$(&=<au3UsR%`OaiX1)3tAtkr>aq`U
zzWgSEJ|=!_Fp$|B5A$TfL7gn-&-_`aR<h|m&huwI`;ar~r&{3_0%@^VC*$cOR-d3x
z`Cgt(CU5h+^Cg6wwrrR$b>HX9Qu=!%sadil@BZggI<q=ou54s>#%=a4tPAB9*%;?o
z#4|>b1TnkoCUYm7tz$MccNi57B_{_L$$OI$85zjO&n}VuQJiZ=Q(Hw=uTmAux%gta
z&upx>WVRW3`)^s1H2uzf!PHGJT~{a>p9^JB2iBv5i{*7^vXisXK4k7c2GnioO5{BI
z(vL5e$QSbVs6BTwkmYOmb?j4lU(Sm2Tb4-O7W6tXU-MugXWp-<`(_`0I?pF|zj>yq
z`F}I<;)eYBQFGU$alUl8Q!I{rZQti`7QLP`<kuylW_?}4UUg?Od<1*q|J(QQzum%p
z=q2_bZ^JcZGM)A5=7+3%Yx8TrD3&{%U)SULq;doG+A~U}YEh|F+%1(!N@iD6_mx83
zSa}9(X!6#P-lX=6SmQE-YXkkp%k!AK(ww=Pt;!|4SGn}0w)#K%igxL^7l|5)KV7P1
zZ&a12W2@x&*DCqg^p9NP{)Mql)pD@kU$N}R{5SG;s!AW7Yz*<f2mJ~g3^9wo{B>a_
z=s?~&k+-YJ+m+<)_u!g{KUou(Uz=mtBF@21*20j)df5B50lu!aVm)DnzP8r5^~465
z?HeM6ytUuyh+Y5K!)lT}Qg}5<w1??Bd({2M0S#t3AlKc2^{+ktY3ay}UT6Fl*c9!e
zn?V0WQ+!|Mf+-hUGB>y-v!+_X!p#*0FI(ftfsXJz-VvY9cwh{9yWZLhG3j1Vz4k`s
zLSNXg_ruc@UGakR-BZ@xSQmC<ua{n=y}tOTS$AwCZ_hvJjuj7k;Ntlp{2^~E|FExY
z6^#4MgAq{{1lJWoP<LV9xpPlEdK&=is{weiG5{{?190?F0LIttgRMJ4Q2IOs4_AgU
zGb4iY+(<N79tFK7^aiJM#%mabGi{^bMBZAFw>MbZjv1DKO`{U<I5Pot5}DsogEenl
z0#21D!oe*Gc8VnK#7n~AlqA$%l!O;slAt&`2d^WN5ydR&^G(y>GbIh5$lJwj({OcD
z8oXHRUb9PwlM^}Xn$BJ+y+u9A;0R_)#By$%N<FYI=d-Dqc+-{oD_=5_{ojhhG<XK3
zV+H+1hmA9^n7ln+!P)PZ6y#h@!H{byXq}USSN>@jNZyX-bx|)J9c|K4RL<u}OT$j~
z!{b#c=(m%5Cmzf~_>ZZ07B&rm<n1r}(VSb3LeP&9i1`tNqRbd94~>TJi($y5-t6)j
zo<n{`Vo9rk2w&6}E~ou5yQ4pn?Rwx=xgU1b_rrN7cZ@#V3LdXoqGB|4@MbRP+{PX+
zE9@|QZev*fV~4ogwixuT5w=HLq1?p^k49NwN%xv)KGzhrSgS^I#(G!J6o*-lp3X7F
zDCR6qVr~0W#XWd6*h407H-wpDdv8-r?!kVsuPJtuw=2lo`5#P>&7N_Orl$BY#*{U;
zDSS^c%YvCP6$eZZbHoG<=W`Cz%LI3gO^|lZ2&+dk>zqE@je17#R2!lReL!a&4e_fH
z_mBKAKpob$4%RiG+Qa=jmeewGcgn?4e<ZrcUzyvyT6S|+!1|Hg<4vFKgmJ%Rdo1&7
zsVzD0!K~pH)LwomldikUBw-?FPceKy;CX(x19N|tb4T2_0%^zhtp8#-2b`;w3*>G0
z|1@%g{u%|f#l}ymA)clc_nVw^9@mI-COtldxpJ>KSBx5{C7-peLtd`<jLDTp-YPjw
z-cBHIw<IV<rN|MBi#hU&^=iJ2LOPST+sWJiwvmslQR5s{QUgi}_^OazD-?3JsY3o(
z=7<-0yLMfU{B=->s+)p-N~Kf;vcB!9lts&x(l><lY+seE4OWR=EtUKvgTHKN&v|~X
zNI&w|L@itID<oitLY|*fNUaA-sdru}PUP*c5lZ&`_%Y7|k6$R|mLIvBtddAF`%E5L
zo~>ktgi6d-t0apIO-;yUjmtXto=R@9r`<b1B|WQ@GKUO*Y@m{qCMsFU`nm)A=Pe6W
z5>_`?Vozyg*f_16nMqGkhE}QqsiFEn9rRQk_usJ2?L?n)be=RKSC8{5;H4W&U3RWc
zR#1O-wpb_i>5u9}f6+=mopise6_37J@#@501OMEJHG1w7>bpR{FYDf!Z|HSulP7E0
zqaJuIPrg&nc95(+Zj;ZQhWRpp8PU_{6-f!RJeHHq7n11zC2!A?w^OLI3Sk~dS9@k#
zIk9%M;a&!^adrsjz!62VWC-^d^<}-v-mr#y6edy&)qDW^tDQ>ZWAhSm(H6^e?i1cg
z-WJ|wM)aK`sq?HzI({#d)7uNhV@aX-FbmqQ4bOJuZ4`MMlf(Syx~vzCxW`vR4-$EM
zk-Uv0Yt;?dTW0Tg6?xm3yd7;-BHP&$A8F5h6ITl4C38yNUCx(}tY`a?w^cbhS=5dH
zf9aXl)XS5v-8ln~&65=Ue9?beEc26Do07NL>}N-FCVP*}y+Ga?gp`Q)IdW4~DoqX9
z6Q-u2*EMEhlC^(rOJ%1yU&CE~P4agBPiB=)FA+V?mRocqudkQNjjz;voo9_q|I>Ze
z$ekKeW9?NY%c$YnGJ$$&>Z{Cj{5@k%<~91Brg^i!{N{&Dpf<vh86WfSR){+@q>aei
zzX??m%RLQe|NdW&sv_u*9J#@K)eAN7-!|^fRO=&*yj{XG_;Bhv1}!#5UXC&9k+;LS
z;~>JD8Fe#Eakhs!jCV1={a<q&wXTET?=6wYEWG{y%j$8rLLqtUPTmeXX@eB%K~wA-
zqAb@2_sbgL*Ox{ZOx`w{Y>zE#?a^U^1NL*~d&<ZG9#8BMam<1H1Dv3_<%|~XnxW3@
z78vTtjH~M|=-s^~z7A}Kc0R2!=5uQp+IB!_s2fJ>x#L|A4{S8|LUNiHyi>j4q4C1`
zRBzPk?}OnpeXx9i4`x5{fjIjj><Mt95py^G>5jMu+&9RaBE|Y17*fk06NhoP%mRNL
zI^qv|3BcJt0jPT}09(EUU<Y}-?r;F2Y6M{Kng9&A5`b<-eefu$54vm(MGXt)F@Fk2
z==*T|C<w>v)8Y81T{r@FM<U8H3I~2ha@H7$h$RX1peA7Q$OJ@^w+Fw)qDNIcmKt)O
z{hxSDR>foV!z9$0m4p_oaW5@QLgJ<*Xz#GUd^!n($>A<uoc~Tv!|rhEk6NT*`o=VL
zDoMko`st`m-u@wPBgxyV0i5TOw;vzyV=HDfw_sKk^;3_!q(NnzhGACp`kqL`I`*P_
zj!DPEgfvt&O@(32RNN+SFOs*9&ZQt>YYJxoY3N7RT7;%iAD6;>?G)U&n+(@0bKn=2
zgl+>8QCu<`4xZDn&}ka#PoIi=`$wa|Yc!7C9))L3N8rBWa4eq`gWTX~{P@7E%c(<=
ze6271zV<}-{oIE!ApjRT1mJ+1KVH@Ef&2&E(6xylTpZnzPTt;jZiR*;TVh`wJJj-O
zfKk>BU|YKZdOfzp_^Xz<e!>zj4J{FnZH}dlYa%()43^|*YLF?`xR_!tc{}}I`eL(9
zV8`sI4La_Mer^goGWo|4W<`;=rR1$OdE1xg%OuvZ`4*;ln8Ml0bQ1(So1&(e;-#f2
z9LU=+@>a8r^IGyYVIDm|5ylv*ZwMFW$_yND0PDGih#5uw@mGDgvwzh5y&k$!CzVd#
zMx3A)d95CXF6KUM?!w!5guj=}uG2Vh56cGbOPRv`;AehGVJl|9PpJ?e=D!{~%)O{&
zt#>2#$C&Z=f}V>sz7M>JW{-{U6L*<EY_^y?(O0p?s?y0j@>cSw1-h-4F*BG)QLL6W
zoiy}PPzO|(@AKz0(*LMN2JX_xZq6)wdFIOZx@vjUn>r|Rw!BO!J1kW4`5pVf|Kx~Y
z=N#D@pCe^wbEM*Wj?}B8kmGN%W&ZSRX+9)dvXCtfKeMFw^(@)(DNBO#vY79eE#I}-
z(%B+M-Yv=z_m~{f@2HSZz6u%MODQHkN->(Fl#&>wWRkbm<n2Q}l?-1({ykHO|6MYc
zyq#^VmK&T0kKd&D|9$G)lNB-`jy!kf{rK3K6IdtnF%M2DWu&`Gez3QFuD?oxbxPTq
zrIcvay0Luzbk@E_>|39=<;<D2@TyFe^dN6vS19E=XU%aL3i*|zkV`!jvb>W*>V3|Z
zl#L2$Oy0UGG}1XjD}UKD*cG6at?#rlYKo4TN~~AO+wtrdr|@H=+3a<1=2_>CPWol*
zm;<Jh{$%2y2<pFCx3$t}WgBPZF+2x(v|_y+PLFXH>atklM%e4*_fnl?4%SNTc&&uh
z<a3zo<SM<;`2)1FragPx&9oxS0;weTLzv;=#abE6t!&5KjsNIRx-^TKV6p7ePAHT^
zeVJ)h%zCsD_Ysk|C&=H0%ykZ7UdN$x%-A4rzsxEY3u>^6Saa6!XP?=D8P4O$(L?!?
zMBdh)R3Pn{(Q$3h|N13`a$!!P?4)NqFsVq~yb9$V>)k)hh<v-8&%=4|@!TSrTUaEz
zqs7vlUSBs}Ygt#WenKDdYtC!Q-<ny)^00BScpNN{i{!1|7W%8G@vc~>ldthQY3WO^
zS1`TSeRcBSgH8?`GINBwI5WRI$?(ug$h!aI?PhX4gtOoynI$r*Co{RMsT=EADlbPf
zD~)qyKWe@%v}FBC{Z>ACYt5a>oA?@%fADr1^<3;nZ|60d{rQ#DfBmH1>l%NTE|!)`
zt=eT`?#*3?eL2S+NzHOHGpt$n#<JJ!%s%x*FM5Eda6dEmM6P`KQ+klMx#Or`o=UC7
zw_nm|81tjK6KO6pX~wfxHKT_fI`z^+bcjBJPU)lSGd;QFZ3=yY_um_#FSV6hlZ;_M
zh&eeOP4R|j`_tD<|9>X?x2NpM_2e9Pm<4ya)yG=$_A0E<px6qn8d;;+8f%OtZ?Ee$
z#Iw6LXn55I^~v3@!yBPdPdhv+w1>H^ErxBiL+@L5Xgbs$=ji3o&u~KQd?y6HamM%L
zX4p^O_WkODs9P;zJEk@EJGf$9dRw&p?1sfAov@4=stOYidKTT0MJ-j!240w^^g{Fz
zZ&dg5!K4{J=t$mnKIp?d=+5|h59rVqh&lxrGk>GEpD$K#@Wlaokan2*p=_c*HjuYr
zT?6PFW<A?G02O!rad2jLynomorc*cz{^(E7Z%=$|)*Dk#hA@jc6zv~|Vz^}(PP`7s
zZACcFoeIYZ&wjY6-yfzC5jguJ5`CDp)n--#mJLim-4^liSQy9r`&caVjKz@tv8b02
zi<&p%VShIXAIy`mBP0>SZ4;6AH33m);&FUoJX|y4`Me3}8<&jFr;_n|X9}toBx7+@
z3SM7IMZ<Hc=(#r)ZMUVOhcy`*nvP@SZ8!3^x_JhM&&tGOuQYTq<eYU)DpqS#xGyRd
zD??L}JUWd#bW+jZh#!AWLHcvDc3%qe)}>%-K?){wH`B2R+}&iu&(}(U3whh_a}wN#
zC*o~*BD_swVSRKqb|g&3A-Bo+w0$({Hy(|yhesjXaRhwZ49D(!!*FtD6b5yO!sNU#
zG-0l_8+mKDGXU?$1z;+9duOjdd?Nj^k-Rl*;))g1S|jv8E4(?>0xd)A@u5q7d~071
z6KmANl&*C#{)Q#~Ua`cjITpCTxDG0k>Y(kU|9NMbtw`QZ?re&k9oRQknqb>u6HNVT
zg5r-R@XIy9Th3(1WSZg;{j*u2tZ{ppBCoqCILQOZ+lz+$Kc8#@5At@F6Yu9{il4Pi
z;dzq$-C~02<n6h+CaB+s41dPnvBm)2Ck$}CNDq1YYXG}z;QHzsxcOQSv)||;)r6n>
zp@$2{_0Vb;GcDTx6_5OCxq>QrGKyL4b$@a981rS>GmhXa_Sn=PvYmR&GxYc0@hunQ
z+%k#ax&158-z71nvZ-yU#0)EwGl}$a9O7O&&cps=t&miUH5PLT8|c)MbVn^U*^gcK
zO)a&lzg2Ff2I^R@OtsIIcdYSZMykYjo>F2wa%ID0m0TciJ;~eWrAoHem6A^OESr=g
zpL}wp>#Q7Evy#1J)~hC!*)nW=wnUS+#|CH19kR7kS(bdcmL;t|XGtJ=JCC0${GKho
zwaD6$IWp%_w!E2~EeUh7rL_rr!X4RHX5HL}HEL=T)~Mub*ifZt7b#@~dAm79AupRN
z#4AD}P6HKUxLzT1CMo3S2!(VYZ|gcK<@j)=G>TBl?<l3L`Jj|bwaM15Dp}Wyvt_MP
zZu4{VL?vhWIb#p%z7El=#2#@S&W{6MDdf#|h1^-75WPj@Y<`aP-kBr2M&-z4zV6~@
z*>Z2Jf<ITKv|PzKHTwkXhmwtc<Q-?nHO5jyb(;P7of@g8x3^mav#Vw@k7E-VaaAXm
zIG3$QP1M3XorIGcC&}V5>_>-@w||Ya($0W>BNbUnEwvox4BeV}SU<JWW)Pp#TPs=i
zT6ua^BioK?<OLa9iyG_$O|^2Le7;B>mxZ@Z%>H9vn!KG%-d-ec&ycs<rWeYW@yx0u
zZ%;QXkVE~LYe`>H5qBuv{hTkwKdI%4DU?Fi!Ic-e6R9nAMC5H{59*{?cP^)A>02pt
zICtbpw>j)(r&3d$P{=-Wp<L%~qhm@wrvZ0A)MF0i!b1Aa$n!GhW9}`G`Ll~8fxPX^
zeTnVOn5V(L5DgY{AL3E!weRN3Rl@@2d>4pz0=>z+|6x@=wMzNYg+2KJ<n0P(tz7Iv
z{do`ab`8I0SN^<J)TC1*XYE9-8okr)spU4-V}G5!a$oj_7qV_1&)0v5^JfEk-1d<1
z6(|-v?lAuNmHD4_i^Pxp^zEmZMarzMsPba@&D>1APsI}WwM0xZ*bgUf>&!2e(9^7M
zSCq<iYNcC}w|*gIGM~JynnjP&c;;9pGAD!D=-tc)(H~PTM)c%es4A1rIozRGqg>v1
zWR}(2U($oSDNh__rloTgwdhrHx1dUVxf8)}t^v01F~B(Tc0GGu&o&z3@)l#%xnqom
zww&LRx3}1j^NnCG>>X1CO)-Q1Gc$bft4ZHOO?)e6=B#^te4f_;6aQGD7J1v+&6>VO
zUIT3Kd9w}9p0~kl^46|vL%fM=goo=IqeDYm6u-4aSSvgHmu&~w*yG>ZPB824jK2?@
zQF**6mIO7!>JZkq<Za~a)@YaF3d7TF5nR{-|NQL$%f@b~W$R84iaY#<dmyo%7q%+A
zaA2Z0CU&3?smup<JvyUmTo<HdcSXW{wx+AQqK={~PCn^^#<#m*@R=^?SJWAu=~)W&
z?1J69x}ueZ;KwJy#Fl{lAE43r?$~#~I~M!<qYZhxrdLl`tO&$(^49PYeY@nXRxb=k
zbfLI$G!&}rP`D+8@pJvLs2}^x!4deccLYApNI>&^?p@duhp@SE+@lzaCmylvBE{m=
ztXTNYk4MDpcuc&Ri1Fm@w8n{8N8bK(EFLMDWUXI3*36B^O7iv&c{_f23TANjD}Kq`
zL6nN|r&HOZOobD9yEK)1gwG_RM^G}JldE3i(_qpm9k$w36i^TCl)$?6N($=MNW}&7
zR4j8(#r_Vd_+prfukBgi+Odw^n}X)=lBt<ZM4$Z$+#`^H@GtQ&yBv>QHFyTN6${61
ziTLTAfat4nICygwG)Jc(`pX0q-sFxZ({WhUZY28HGnb3JZA;#+A#dANM56AAF!+>&
z;N<HN*zf9%r;`G(x!NB`$=lp$e+(sW8(6!dhHo?cdTtM=>-O*twrB1|eHb6F2b(8#
z(Mqo_2A;OWlcko-8?(egse|od^w*x^KKBSSdX(7@4&?t?&R<>cu_oPWf)r*@7;NOn
zD*nID=M0wf*%9Pz<G!YtMmFCoHo?yR%wgo-_XCfNajUm6&e7wu*1;45+K}JorZAjp
z0+TT&Q1N*pxo2-KXSX|_8lfS1o4(Hg&+_%Shv2VV=H8t1P5y|QGn^f)X|wceAb>3X
z!t?H(?Uiz1aiu(2T`Bh7RRZop$<U*x{$#nNb0>}|c^g9BHvdF_E%%hZ?7>~A+0<nA
zV9xE8Qh6O&D*NfV9Col+E??*VxSmC_tB9Eoi5e+3VTKd)lM*h`TRKoJuc&$Q_@ov!
z^*~oLm^VTF(~t(aa*Z`!*d~>9_EgG(-U?>tDdhBEg*+hJiXD`4uT&u({t7Xk%zkj&
z9PxF{VU`|mTjfa2AK9WBmo4M_WXtJ-EKzRGlHYT(WT|_WSZvRd@KV;fjk0CYrEKZ>
zJX^MqwR!imW#YtaNhNRR7_%nLR>)rJm%FS|NcCj}y*3JoxuB3mJr(p<Dx|BKLM(Wj
z--WM%uc_WT-jcWJVG3!)S+GA}qe~CI&i+a%B5xOyt>ehs>&_}^)rkG#T-MI-*(dJ8
z@6%N!Ce2k+j~ePc*1X?pbM|{qAqUCZX!5qnLWNY4x8A#Q<n-tq@hr@iBM<m`7qjjq
zZ-3J}TuSCmpl|mi`DaGX`lK<#V~s{A<dq;Ya&~*2bRMG<^NsWdU)D(%dV5Av|8#n-
zPIf-lahI@8x-KR!$&em;S}El9?2|@3+1p?G54F%+sOe&Vxen{-&8%s+*5k~bbAU^m
zsdeM5_W=1j=!-^%mecn{{ns^W-M(Da%FZISG%HX`7y6cu^XguymN#W;X{V==rp(8*
zW8J(WEl-Tt^Jtk)?X#MBVuK4Mju}=fI7dCwg8L+@iskU!Vp+x+Zw7h0Ig9)wZ|AdS
zKNQN`O7c6ZJMSM`KutII8&Na8m!2idaqMkV^JGz8DAwG=^bfV(spHr?$z-3r4eR}3
z^g8)6)AQea=|tY1B5(D`+Xh_=q+dF-Kj~Q>!<^4*>bz_6?6H0{^?j9EIoVPtFO!*9
zqSQzSokrd_;Ms%Ue|EN3R0rr=uE>)=&59)RP6=~L*;nQ~+T5>5EIDHyp2D*RYxq#N
zLb<oNP;S>Qk}=da+?rl2U6vNhy<F<1+80X?^J3}9Ui4O@5~*T;I-ZQhsZudrMr~Gd
zsdON3dy%(`MsPO*dHbZiRO(W{{Dk_c18%HuBg!SzhaMr$iLdZ?v$(8G0`HZJXQv<X
z;`LAIE)`<Gjr$5se=*1Ew<ON26q~Eu^|w_I_-24!W@PO+L+Cae<0g6A)Y=5LFL;h8
zZ;!LzRuRKmVX`S=$=ejEiT(xV*lojXDrG$kYtjHO=2^k#niT@HR;XWLjb#xw*u2pO
zvrgHdfyxH8FEzx3la27-^2V_3Xp241Y%$x}4wtg*U`F09BX8>ub4Jv8XY3|#htfwh
zX<u{v_GyWq7h0pi6;~W0Z_V;M&@0;kQ|i0nu#G#urMhFmCl7R}>xDY;UWgm-4L?60
z1kta1J+?D$`gTEw%U$un&JDLNy5YlVH;h~FhSnKwsQA|tryh7=riC|R9(6_vdHcmk
zFd$aYFiY^U7#MS_8;tV1VW<y#(C_`>*1ji#)jhHJN-t#0?Tu-xdgC?sCAlyU>*U^0
z^#2wLi;e6v?+=6Nm2mE>i@+xGHrgv5SM%afzB3Mk)8f!!Pb_s@u~<?*8!bA<qQT&J
zR85b^3;Kp$cZrAFm00FH#lpQ&99r~@L$~5MT#Jgw(Mzm#sk_oWNrcm{MD7Yo#FCN(
zw0xI<YuO2CtWH3y*VJS^Prxl|n|+yU)jN)vR_&QZb(Xw6z<QXqYTtv&xOX*~JA+ej
zwS6i+wPW7oNzQC567k`CBCH-JqMGa<O5TRAO27%<1Z=(?k2n1K>&RQri?KN8I2AG6
zHQ`oo5(3>P;-13<gd2{>>09HF_HGRNw;6>E$3|iVc^hRk1m)FXxc@N>Cda}s%%L~-
zl=p;OXYXKiH<W+%Ma{{;<%U3mkhTat?~22J9nm${3hC^hOc~YyTJQR}zo#CC+^Y-I
zzn1vGKJqE@cHb;Z)R|QW$MkFC_DM73yK%qsdFr58uRf?CKku4Q%Vq-YNE5UnC#&g~
z?M~jhvCiH4gn8m?sW~2Of|{Je{`NFMo)h)Q^z}3+Z!NhyuT7{i?!_D9_(Nmt`eh7<
z4aVpfWC-W))IDS~Cu*Dl<~-0xKfM~5qUD}0?ht4>{*O2pRY_;^)<sz*uijM2YVLmb
z(c?wlRxYd*hZU8Q%1rd_%yy0DPLs?N<<f|AS@dBYyR1yEQm>K59#$l0Ne}4{+}^NE
zX6~dvVmfmU$lLjQisd?M6z6{QMwV)2<wT9N{HzwY)%2Ot>-v0vS{8Sp9%uzUwai>J
zNLEYfewEZ_4e9zsA^Xx4;zP!+4OGa?@d~l!pX123qtOb9rKb1`YucjvIot)1Es?f4
z^g!mw<+s^#J0M%0#AHj~z-)QMdUju@EU|aUl6x((q+(l^^e1m`TW3q1wb}A_3G3Qj
ze9XOUDa!lKoi;f#)lnhi)+i)`yfq|m_ncM8Dl#|f6!{9)v^5mcavS^Cx3VRZmtpN3
z_Jb9ooUM>TAJ)Cy6|BX`SMv5Uc{`81EuXHG!lz34PTllo_QE~L-*#`5+$YC*Fm+f5
z$l;scl~OZZDf`I#5`L}ze0}q;XEP^6A-^BxNM>oa)SI6z|MK;!k7di}583jfB1c-K
zsw9J&X&2^VtsTHK4f8RYq%iM!wT82IjWp+ccW`6Q9>eGhPG%MsJwuJ(a~8|q^7&l$
zj;T|dN2d0t)>+HE=F8RW<5zGl{I5nfA7;)KKX161{pG1zSr?)ePkpUCXK&)e7qzq?
zKijaEoX*dcaR#u2dD96FT4~HYjwz?K;zr&s)~ID5Zx@ocXR}qZh`jy8{>A|_=5Y4X
z$aQ9IWmAWq8J919=jll@U{+S9PF$Dg%Ox^+;ktbJ*^Bk>@FKY|Dqm_b`^qz(nk{}m
zR|n@we|n4)f|!dny+AfHQ}a9-Zpr7{PNsHis*!SXHjbGnkNJATqv>O2o$k*b%e3oS
zaX!iK?+i1%*q0Anr;&Ti@^Ys~IflJzgJ=2jne~7By1XyX81tB`Da_Gc%KAQRj#fUh
zkALEwMjF@S&(EJ{5%SiQ-sk)5;Y_?r57uq!pxFPOPoGn>wMBB~P@(i9Z*y6{zid({
zi^~h-G0zAl{R-p_?`JWUH7)1PS8R&K)u>o%I~7Ysev$M)QX)gR*WvYHdX^SbbA6>)
z2GCbkcOZ3Uy^5s6aq7%Ui)GYY?ldf-euBJx$=dcD>)cYZ_IMV3!_-r)Xv%$q(LcrM
z*-zQ($voz*6>_7rg8%Qi*D!`&-7~)>m41ZNo7f8@Z}(3zgi(iT8Jn+<HOyK!<NH)s
zp6w5Jqo+Tb**fIykyKN3>sk|g0_YD4uLFk%brHgS2PduT;rM^`aq^xOs=iv`!X#^q
zBX13l+Hf9hgW=@uj2Dg2Z%JbmH?f7@3tQ}VwByXp4rNyMc>S+EcHE+-D%26-XPcsW
zdNXXO!~BfhF1Ws*bK5Fcn4D)rctd-vJlp~1W^Op}r4t4wyCZ3%JEER;!nYs~Y+!G>
zRk;@u@A{xcbZ1!kbm2~&u5kR&k(s-0*iPOaCU3vQx#7_*Hypdu31?fl!?T+Q43~Of
zpUwm8-+AI&m={tG`_RYJ4GZ(S!NjXOhP?BKBm17KI?>nLwil+F2V&ORK-@3yjn9Yr
zAlM`r>8^cIZ(%6Rr-Wh0-%u>B48;W3Q0CFc<F+ym-sJ6)W^ou*Cl;S<V^N$x8_RCY
zhSTQR=-hTT%7?__Z@XB`u9%I62WP`HJQgoH#GzqT9BPxd3FO`P=mZ=f3l*;tk#3WS
zlq~v;?j&IN;{>dJnt;o^s;?(tRI6kd#Ux{7Dt8HwOhNl)shGVu1&h}uV~|@iex6H4
zHuX+x+fZZ7s~3C7?Wo<_M&4SJw>K}dM%GV+3wisbYXaIN#3NupJoUkG$iEPal)!(O
zu`vPv_L>0mx8qT<a~#T;ff?9sDBgz*g?ZOO*wru!x6&f;cwt{G*&Twk@IILA+z0dA
zdcxT=fV+*mqsG{-un2ZXX)AXu?%W=h%#ccG+6Ke7x5n3F4Kciz6^@a&D=*f=l|%Iq
zLEbj~WQpj*mhe5v-L=(q5Y(g&MqR8)e--yla#zolyQb(j+7wUzu!d!Q+jukQts_kE
zp0_i}+iRTR?q!|Z|FQ{8S8^Wv-WV10xSO`AF>b4k;I1`-No`|{3N}WSsWGOy8ljAu
zr%}BO5cO9dEqWW`rGX)gmm6UJaO$8c_0cy<56(7vc=)AS63JUr^7g{}D!G|cMNLkX
zEGKXOlvK)?b*yj6+X1X`&6#_ArMyB$k++ZN^Vv+^j_zD0{eP6oQF;=R$=hz^?F{nv
znsu2xT}v+`dHZM$ceicjJc={3yL^9I_f{hZ+G<42neBv7wcO5Bi!VL3-<mQvk=o<=
z^t@Dut0lHfCD#@x#agS7C#m$w4pK;OFNN&xppaev`!jhvp0}f|awLW|>AjQL67n@$
z+}K0*_(^VZZtE19B~9|bORQCvbPdcBH}baFE=yiD=j?V`mMqzjCDr9w(r`t#93;Pd
z$*-n&vc-hFjVWPI+F8NuJ~DSA>&gM_Ba^qU&QM=P{*SU#2tzT%FEvLl^v#xAZL(!e
zogA?ot&mLecESI6%h&5$PazFz@_G2W)B17FIz}lU_ECdnq>{?jO1VhZ+TK>mQH4^L
zH(-yP_3jS#k#}?E>rlWRGI{%XO}2F9>srSed@9crU9z%eS0;PjJF}(y0PFYn*;0_3
zBbyp2Wg9hNzK7_IBE$SeBP#a8@7`yfK1waAx73nImfQ|!HpVFK7jn^w?mGW1IAitx
zq!ml@_RJ)$n31cF)>^syQzLmr8VNh7k)-9!q5j1jX=ZS2r$;HrgPvmcsK37<L)WO~
z=Mc3Fr4KoT_uWl4PvYa!8*63hR;|<}Zyj^_cc*hDq?t-uG^H+Vxk@G|RpP|C_rc$4
z>0qFdqM91%vq~fHX6DPqv*c@4o{Xe6`YWGr?@_hPAa9rZ7Rt!>MRJn7J$at_9@JL9
zr+;iDXVbdmJP9CghfQHW_zdf0eoxc$*!O1LUE>|+>&Mm7iM;LhR4o+`)Ux>xJy&(<
zYo5k>yNcOl)aK=qxA)g-<QRDyYog;>Nh|h0=)t0PEMLd-3iC*slea(FOZbNjUo~7K
zlb`doztKpv1vT8ff0dH+>H|7?LoL;8`jSGq-=Qx#IyIOc;b`t;V(-|EpL1+iC{7NA
z(z~EQ&d_Ua5XhZOHq0SuT_C;kxIeLOv3&kR&6zR#L+gtq{8EX$sAjGu^Bz04=AJBP
zW@EG}mPyo@MX`6Zm7b~>1|`yndM(3`%%<inr#t&h<^j}HUF5m=d6|T2S?9Vk&nklZ
z|8D=3p&cv4c|(N=d3%bytr7TJw8yy**rifj(knTWVV^ntj|6qBmU~TVU|L5*1QnWK
z3iT#mx^T8NochdgV``L4QG*%nvl?(esFyhwytjbiNlSdXWQn$SSa+<di&WEk`0}#>
zELK`$Jm<NYBW>t`X$Z3o4RK&`W4yiF7_XSq?Bu}t?L9mAy|croRC`Q1Xpi&v9q^aD
zwMc7*j5958z}y9g`?kiPnQai%uN{7_p)Y7t2N-wnh+6uc@TYGl6ffxn*Jqs&9PEK5
z<ZW)57slN2Vb(=wSU2f{t6w`}$Sdla8+XJ*=4m)Dc7sKr8;ozeVe9Hn*mb58{Cz#p
zWvvJLk+)UnJuzG1iOgx7`!4E?X=z<BP~8PN^ca0?6@W!c0x->_C!#+0!j9pA7-!W7
zp6-3n*C_~RC-z0d?7o;&(if+T`y!%EC_0k2YdXavEISUJ*2JMr(>S;*XTzSH-O3DM
z<K44SuwgdtH{tHzuvq#(*<ZfOy0-zD%a6_5#$f<?JC&^4zL$0IkOVxukcfaGoby_;
z4*iyZdAAbK`Znv{`w84flz?X^cr|C=c__WTQ<CvdzZ6{XOU3<P$;hjj%v{Pj@MVqL
zLy>~sWbH-QRQzQR`F>d<63JV0^46BM_8zZkC>c5t&$^Go(N`l;up|ccFT`NY{%9C)
zipJ3G(OAAC8e8;-LlZd^--dC2u=h|@{S$%9b-6<#w=aUu^hKRkz47sQ4{SH;fqOgr
z@IK3(`)}NF%DfXC#<$1Jq*nO#p(T!#Hb>kxYSi~Q<9wzQo^uCKdP7GHP&nZFM>`C2
ztB+;m?aGgqs61qe@=F%z+}Z*?$y-(P+W(KFvyQ7e;krFtqJWerb_Xdb?xmub*fokV
z7NCRzBG`)E-QDesHFkFgCUA-a9J)mX#JisR{&7A!juHmn-`;x#rbU=zwG(qHm}gNK
zN?#9oJNK(8Zm%#!(@xYI)9;p^$}VW$a@X>Ca#4&4E_UPIxQz*VH!wk48xwS&X^fRe
zjd9_(F*8L?a4W<ZgKXKgWod+7T@CRxTpvx0^iif~gzt9@u{86K^h~RfdY@}#Lajd%
zH@rrskgbQhRZDO3b=sCH`FN#@dZQ|-eq1G<?E3OwR3%*tD<zA0t;d-mXGGqn9R4m6
z!HyK}vc~dGTVzlsm4Ene|2OmKm~U{Ao{Q(3%A{;infwCt=MR(!@`~kz4>dL)xp%9d
zBmdmx*S#OiXd-Vrk+-`7b<(CrCt=H&pE#KMWBOHhF3gnQ*}R9&&XAYn>o^bUh6}ZF
zgDl)wo4TPmtz0K#lkakV-Iy+Bsp%5<I$a{FG;*1IZTVR(r_<DuX{QmJZW__^)-eA<
zBf0<MZ>&bXZr6zBfrh=@8aZvI5wovqnWRw5gIu-L+oh4&B^vqEEnOOux2Du!nUc4!
zF7bv+-WHIfJMuKr;|hOH&vc1xL0<D_T*kX?a0lL#JM!n%;~uy~EAh;&N+56l<&OE|
zh)glspDBYm_nwO6P52}AQ+F~YlzQeXoOkPMscC+jDPuVcCx4<}>0pK^H>XQ9dAp3f
zJ(-^_Q^?n~UDCzhCtWNKq{}$orpHg$%DTl`sq;KTJm>4g{1&y^*E2+8mnr!tGet=)
z_fPWHs7{WoYr!4!CEi>kIJZ8d|0<1h+HrO+wBsI`b6OGSwL?anHOb5ixj7PiEJy0B
z%8@Yg_8fV;iM$ogs=t|Yb(;48Tka`h`(;bSwrpwkH)qIGoDb8uU#^!cgQ<0!*E2_|
zsdszC=YZbiZL|aR*8gP5y1Xno#%HXroOxUQ$(HA2UR!eAkuzzGIlaE4^2C`l=%{@;
za)f-0zNwR@)M7V?qc54SKZ$%cGv;lXd)(mH<ooJuX~~aoC_T*jm#G)~lq;9`+%Son
zfH-D=xstbf*O=+~B3sr!%a-6f?297D4|L_sTQ^&_lef|2t@l0Zy`So2S`W_lv-vzw
z%B<>tvn4K?`I@`3<;j2?vEs*E#NW%nu{qQVQp@*>Gp;$ismS(Loc|j}a9?jzAfwh7
zN_Vn#VlcH|Ud6I=Xt7jSlePS@dhLs(ixYpGJ8aKF_CPRSQ*)1fkko#K7Z-~CeEz&@
zvXr`qkqe9Eo=dS@<i9bO^io-RuT*>n(;pR3BG1{SsAJC8=3~W@XvECu#8N5wS}F&~
z+pf&0ddU0EvJ2(1oxJ_Sy>j)qa=GGFE}xkRZR+ro{|<hND|-^Yk+)ZRR7&MpcI>vS
z5-q#^UcdV#e}4Ut&!Nn+YW7pkF%P55b3Ig48lc?L7**V%y&!M<vje4?H?L&sIxqRu
z!eDo6ycuDQ*a_C?A8Cz@T{bwkg1XV+)G9A`K%LDF=*L@btNeP{8|H*l%bgJYsy>DW
zGIy`GGjgh((R)ZktXR?z-w!lI5qZ1LwK4qvYJ$TRt#H4(4Z5^%k84lb<IGf79Ny}N
zb6eaI+{6P>)t)%u(-F_yTrpMYf<50|FiFoH+Z%Y`$2m{@H_Z+ECb?liEq7Fed7ue-
zTZ{kCy_W~VdU)cmA)c68<OwJ8b}BQBeys7vHG3cYQu*M@DqsBT><2T8P8gmTfL&Pu
zSa`b!Zre~t)h-CPE-|zDYA>`8=?za!AB^tQ7uWyoi}17k(CzC0oGu!GeeB!4?=t|k
znhrq3^hlIWh(UE?4D*hex#Y*5-+S@YO~vE=;&^P5c-$zTi75kTqObW(v~3lOtWz-<
z`FI8zt%<>@#<5rt6^9cy60m$6Jwzksz<Kl>w5T-)t9j=welQ!lCkc4%k_h8xiLhtB
zhTr%^yy`a#O-IMWu}wTi{vD4`brO-enVFH@=VIKYd2pzjjc1)^qqza!W=<mXj>lo*
z-Wh1<I)i%JD9k7ig;j?!&=|A3VR#5K{6cVhZ3vS8u(xpMcm#%w!|}|KaC9Dw$bv!G
za(57>-R84Kp8=TS(g$0Q_QnzJ0%C1?p>E5rI6TfDZR2@wpXG&`7G7xaw+mwSw@00~
zt>AU5C6<>o0xmZ~^&#?gaYMv2)8c%hGX|B^haK$k#f`J>WgCpOsfE?|ETCk+fPSO}
zny>@jDwdt_!_Bdbyj^z542#y8A&J`hPTpqN8)AxBH}00n+h!+C&}+E~#&S=roN5BE
zt|mxs$v*h{CU|a5|9PA-h95M>xN2j3BX2j7!-J<9p>}g4bgwkTslj^Kc8q;_)IA;a
z)5G5hf24IvjqG?`BZ}%8Ip|p<>CdWVw0pIDR8+|k&atn^+5*15lDBQx_5NmlmF&r{
z6g&2k{A8wj8=ne^-N`ID&bL|oH*0*3cO+)v=;@V-E58;bk+)fV26v?X^4ke&O52yn
zwx`A7!t9E6oDm*|b7uO$S%Y8aqSAEIxsFcu-s9KcE;`A1mL=07=!+@Hl=s^>2Tq_~
z=v9W);+%M<d4^1APk#^Zm&dcUvObj_Tk>`lS+w|_M!K&ggFbNfBX46}xi2O&Z7-?i
zda7E~<n6kyoMm~p{o$mM06mSkzEcYfxl`u6Ydt_C<Ikui=_r2<8ND<|EeRjgGOb7>
zZv)dMlRSIcOe<@rYDIlZD}6s}#hDxpzOE51-|pc?yc0W815MtZByYd8<qXcOjpvTM
z{T6G*|5CbCm!!)e2i{$Kk+IxS8z^`O?oIuaTc!x--B#r7M)LMCcgdG0spLdGwS3{{
z{=thg{kn8<HBOf`xf)r?=LWC38o7|Jk%Qjp(we(+l|{NN@=KQsm$lOLK5xnwGTC{+
z899vl?2j2@_&8Gx+UcZ{KBEV<a-?g^9O*KQ{wuPg4;lL1jCsr(a%2zZS>F%Lz9MfY
zagV$&J4bfD%@Nn>Ir5RbjJZy28)v}<ytfCkqrrO^_Yn_uvbqbirh8IjJ||mrL-^}Y
zl2hcZGdXygx8IMFBQu!wVgHaBnfx5huko3Jny~HU?ZaL=Nusywe;FoW_p+tkGS0`m
zjV>Z@M+N7}d(NkMQJf=t@cmuY$ssG9e2ZnKCwY63yj{Ti@2Gfwyqwb;khgi<nVjs(
zkCpmqQ!>1;33<!g#hS<2a)f{9a{uFuOWp>Mx7N4#Hcn^Dia*TME?^c(vQD}@;^%W$
zCkG$XfA&EqCVkl}wUG1hDf*xznfo;@TkLrM82vO`n$+g)*@f97)X^_~&K)y(o9>${
zJHx3nBX6Ud(R($TyKv@D6$KSZWLsutv?~_ho$N*)Q6%SE^PcR;d7ZqSP>?U4%h@Ht
z8**+c{mkr4*2Wddjt@obUtniq1am;>Z@ToE`m$YxVzjnU^j{VVRVZ?0QIQPrr|$Y_
zu@rE=4JL0xZ`1!vpL>;Ou>^f8mfzHv1qN|Wp2h61du39;*8kmnA7?Xzh906R^zWG@
zGw<s4cZp>N^lA1k+}grC);$$+dDIViL*9Bbr*#eI;WnIwYyZP+xdAouHM&Nwbp0bX
zQ~t<PKJVw0a(;Nselqr%DT1sJwZ{q#ORdl)&>AzD^LSv9E$Ys5z@2ah7<F)jwjT4A
z8`Z;vX-;^S>x3J2_0hC>1EiC;{RTBeN6xon_BDj_twyM9-<Wz*W@gN50`G6lke=KQ
zZscvxg|66k&<$JHy2H1z2lm^!<EF+1kF+kRMcz(&?1Gl9T(P*B2i6|&WLAP3JSMt9
zQ`;RK$9bR`vlEJx?pQg*1N$0zV$)zxT&(AX%g4OX&cz#t)_LRM0v}9C_rWUi_N2WZ
z${+gS(-py|b6wHsS`R!i4nkp*AY@z)!kP2EuwZm=IFh%{%$%AN)EB$6`eGRKAs>$m
z#y-B<j1Q*2DHuf)B9TGf4kd4Q{u_&43*umx5sMbOSae8=!wmnK%-ooXj^%N%KN5$5
zcVm!ybq4pO)A7b82J6V%?h$bqql`y6eY{Vpr<(hEHa0y^#BcKU?^E;@k+((d=q>u1
zh-+@MF_XLMdVS;BjS!EtmhreeCLZH>JHBu@0UDW$dSB+EW#w$_Aa8HBn2imK5^;1I
zGZ@L+RwJTOej*H)W{zd1<`{VY4#A}Ee02;#Bl31@*%<trI}YYqBk`{9U|iH2OfA|V
z98nH{0q5C$<gLoB5A}Gx@wx*utrrGiM_^a%4CLN_t1E7*TcDd+OT;y6i8=LJ;>}+T
z(VE$y1+Sdp_m4A{^mB&sYIb7WvPWxb1M2s&$B|C<*xt$>&F$<lcCHn|URq#r4-1Sr
zZVuPA<`}`Qy#0gt{pRMFnqr0~8_f_--Ug7j9S54itAi<0icQe?t_f~0F@b466Wqu%
z#&0hZls7ZM2=aEW6}cO04C`ISkn6_CvN5I}*$8Vp8{tI@BlwfIe&p>b>K@AXFeg>Z
ze-k7Bi0$DT8T6<|0@XF_6#Fgv9<Zm{ty(IdF~^v>imPu|%DxMgGL2b_=Z;j$VCzbe
zb3diM757cUD&!Eg$VrvVpkuz=cHWVe(cAE@lDT)(bj*)rmW~Otg!VC?V0@YMqX*H3
z`>t_ciX?!%y*(>O25^U0iy4Jur_!&=OvKvEej4zAz8LNo{iusFCvTnUk=eI}v+ejy
z@g{Tb`_TK+kow<toG)L~=d)KU4_~FrrWNT@nwT!_-;kSMG&28!M)s`Ki1`wYG%Dqs
zdQvS4GIuWTwGYYQGj19=;;50XRcg+6YUyXHktXErkY*Z*@l%W0R<b!+En%5zIk!hG
zQ3q7wo<mlXx2rd$OQeZbvTtjpJ$dVqrWL0{8u^=lLks5d?b@b`@3wTQXh6QT&X8>`
z<Q{o@I!nu0MI$bG>C%h5eb_WpoP#r@`#9z`Gv{iQOQtmB-dufC%l;RYw8>J6&0i{U
zBX9S#;@fbi4s2?=IFYvx$=f9IcD%KQeYR?u&1Z_5uIbYEl}3JrQ)}(4m7&wQ4~WZ<
z-UB!@kL0cUWrkSXpk}LuP6m^=_sH7?+=HeM&XJ`}$wKaVO$>4*iTr)oJx4yZ%aKLg
zmtHBL21uPFgRk;V$vL<#_2U<!IDe8ieaPGP%)pvVKK;vCaHl`-u$&RCdoz=SyzO&>
zZ%4tmW6SQ~BiVB8Ct3DhC&B!4ZdNBt8qLX+sQb(U%gvG<-I-a!J;&je<QM1Wcij7S
zAlq6LQ)}Ibw^BZ%ZR*Bn0lxhc?4I!dlO?~H`JwJe@9}_anMkH~<@^1@`BamwlYDx#
z4>6l-NpiM4(_~9M&ulprnk`;8`Rn<yY`Bsw>&RMH^70{n%;rqC{KdV>5I*}Y<LA?o
zS=&9v>SW3=J_C@qPdalRo}4Y_C+I_)L_Hrrj!-iFF#iru;_v0fdhCIm%zeSb9O-Gs
z9XxL))5zO>HuSuWXI}(!xMnlUt8-WGjp;W&)P{QHrbY6KJ;cAd6-r-rDBfZwwT8Mc
zzvO(0=};gWz7<F(DHMy1MRMYOk(l%@ma-}A$Z{@bk9?72QGY!lsZdtDD3md5+OV8b
zBrR$cO9Rfsy?EQMeTXx2TQZnF_wM>dGL5|bLEaAR`oEmb5_!v+cV!Lz0t1<~ajjfD
z$=u=;&bybHlS$vzPkMjntt*zw*kWd<7RmbGg<|umP!!DmuxrJ<Ox~P-J}Z{aUc58U
zs*yPI_Aoob9+S6+ip}8r!kl0G=*8%1g)h6TaG{F1KIH9#<<^*NWrLGDY%qQ;{kd%%
zu=p?LPEB{j(&dia8QWuXs1wfdj_W(WK8CwF!9!7x_s|A-7}o$v$qn#6+!>KM&d@Y(
z2)nfnnZ?uyMt7T{bw}=(BU|I)h1U2npE(#u+oEs1c9?sr9et6!`{`WpkCiKq?sG*>
zZ8vP_;)Zrs?zr^a9XD>eqxO4u93yX|Gd*x~H<=vhh1rL^u!y`h-QbOr<gK342V++H
zBBYrgVy64!#r`h1a<VHnZ0&(jKYBn_CkQSbf>3``5at{X!prADXx_dzDg&4incWwU
zHT@7t)-K^|&X{1VXO{D>p^<q1I}$M`qfvb(7ONY@;Z1TZQa;CG`>$C1>=J_`<Y`=4
z9F`i!!RA#Aeyp5<TGS-9ub2V*kum6dBnEf3#6rDl778XNAiv*i7>rHC+;s_vUB$cb
z;{;ffx67>);r?tE+OCO*ZD2g6_KxR_9gmlm@z}XO9)klC5EC#Lqdv`r>db6*+RnzI
zZ;3dalYn-$6TwLfMPW0TEfI>L<3ezUbFIa&5d0k&f_Cjf;8`^q`SGLhmcE`BGY3Iu
zG6+4W%WiUdAf7E{Ugo$UdYl3=kk2V$%>@q~1yPxu@r=K($sujAGolrWl`YWZato+x
zH^ar6rYN7=0N3i(N5yX^oE%yYtLXI^IM^D4e_P>arWGPSTcIk*hC65*7=N;&hs+A8
zuPk8M&;rknnq%G)b69Y`9ZTLO=Fu<y%?$U)m_pCX6enBK3&fXxTXyCBYl80)CV0o0
zb)L=`ogb68Ka7xC-w2+68KNYLzFBJ{ByT5Y|24+<QX{xD*N1T*cEGnGW2tut?5T%a
z?e*}5-L#wh|45?qAF1A3BQZB?B<+9PZBZkhx2nbdKhCo&tK=E?#${J3<=MqbsYl*^
znOP}|>sHG4v(!laW>+X@+tQTpl3l_qIqs4t94nK?O_&AC>^awe*_%e*#&e%_y_~uA
zhf5`rS@haAB~rdAPabhjaiG>!!Tn?6g=|^Od!K1no$xR(uWn^=9?6ouWNyhh?oqjO
zZQ4ImBB(Rc@_uUP$(hlCSyALq(sJ%>`)OqWwMj;tX;aDD>=zn&@sZs8M<Ze6tqFPC
zguJahrk1*6)$*yUT1?5?F!I)odt=o@wM3j(%lnCH**r!qmxrt6kS*ui4QhGAxi&1F
zKh~MN-memci%Py%E1AcjlqG-C<y(nbmSvOGe9a-Z&&^OvWKWH}GfWp3@;0Bm?Z=NZ
zH(DhRV>v_rLvEAVW|10c_dz31PN&P(M$E*ZmTGGY-h*Q@L_w{S6&ZicP9@b}l%gA;
zk{&BnvaGRMoLj3!J1$*JJT(&ELoGdgRC4CJQU>l*N#DC_IXgiuHhgwC6si#)YL&h8
zIsZ@4O09=lSwMYR15?h}uAHaivSflwmYk>GD4AO533IaKHu<@TymhB;*uRSG^2?FZ
zw(L3T&EDXo9C^vvblQ3Dlf$xQ4|n|6Io~bgb5aj}zYF>J<Frm1au#ey-Z}+v&l<}-
ztP1{lPR-<=IbAmPNtX`OIoltmUfe808u2F3nO@-yc^T4@&mEKXGNq#t{mI;AbmH6l
z$yqa>+VqNA+)-<Fl6r|uBp)AKriP9)bQpP?mysnM$y+ynW>p{L=b*_FMVU?<`D+50
zH`Xtj`}siTXisFeNr+Caw_~>U6l%K2a<hYcX5icWm)e0az6|;Q{K#6Pa-C#+;{45z
zKY0Y7OGGEz2J`*D)k(Dn`5MeUIp6=mAoft%WXp7Zo3xL+^@41vbfCU!A~UUz<cQw;
z967#?y^4G2lj_a;@ipqb2NcL@-kN=xd+GF{Q10z7lHhJd63wn*Q}R}S1bs%A@+H?d
zUs^J=`VKW>YfvBu+3!?o%Gc?9X&O{4;e)w<Ru+kdzdno|#*v*0q|~fH+Ou=QH@Q$|
zo6+AiyjZT#yR2Y-^alE^yuTGnrg5PxEB~LCyFii;70B6N1ybN$Dm$1R`H=s97e$l3
z1DQP%&MrpI)nnMr*rTvOel})CCHofZH>Zb+y#4sBQ0`u*M~bi7O^f6g^S(OK%NsqS
zQdEl!5SU?v{#~dYaiNc*A2WO|o58(-IYQc*Gr!gx2LsKa*O9@B&n!_k*an^E*dp<x
zE&esKL&iWm%-wE>2h>#kb%_}nF7yw@IbgGyBThbXL>==6D4);(>+du`?BDD%RXHP*
zy!9Y&tF(<!cDpHD$=iD~TBC4pYxvJ;gK=xxLh0C!{if~EcA^XQuY6UyVBR!W7+APr
zeiJv`t?Lfgr|y_{!<{-HcT6X5duu(gdxa-TyHKZd&<imh-dMWc8&CfB!I!;0n6bwf
z_Zs_Qe1tzrw|2q3q^_ta?~e7~d!UNE{WgRimRZckV6J|xt3h~L&g_N}eX#m$AJ|0n
zL2#=+&@-lfxw<zd>G#13{YcDM9f?-&qhMRkPMo+X?7kO?gw0Xd=n)N{`_X9LG6v%^
zr^B(+bR26j9rKS&$5`{}?8=>iD|2HoVpR-wACH9-_t6g!hd-YZkP?`P=!Luo$0p$P
zL+YN(60n5+9FH#XxD+4HJgIn?lefDY$D@y7Jp7l$qoQ>J^taE!$eD9sadkGbTg`@@
z7iU{t0@hn{A6-5Rzix5wZ9N7Bdc)a|I1C2AhvI;C6h5>KLGZ89NRFTuio6{|-d+tH
zgvsP>ab_<V<#)vZ_AtKq?GL{lUP$@F`)p}PWU4&z*3u1rquRp3&l&2w4WJ$7glo1=
zSW{LHev|6Kd5t5Q4sk?9jRW4W(`U;SYxE*-^_kIq^{EvMe$_!lP#t8}s)KLuYvVwU
zC0;pO;6L`jH}W$_3bUhLon*enGUhKbCn|`4N1b`It!ZwGS1owEC3B<yF~R+5Cg{jn
zwn}4+NukE*zznF>yv6S8ZHR?a4RIn$A2lZWDEh{|@nb!N-q6Ex@^&zJdzZYu5~zpy
z{MK;GA6Y~#a_RWrvXcAb`Il?t0DCb*Gng;)U$uNb#y-5+RkAv>lD#XH68Q6{Y~uc?
zj*k6e#q|5^u8^gt+3iW4h1ECu`Iu+p-KtC$GMBbVBWAEOL#JdXJ4&CH$jsd(Qiu6J
z;mjm(sLGd4^qLG~UeQJcwItkIZrQ}SgI~9=F_&p_Q|1UK^J^wq(6=A&dEBL5*pVrA
z^s0Ui$dqkmoXwKxKjmCIlbRzd`do*Tw>Qb#p5$%6delT!YNYrHXIjp9mriR$pS=A@
z_B1S1%hn_0Q%}x-T{tU}pO*)zrSWPt_nK;1xL7T&WYv1|)`GmfXiQzzEY7|2)pBHx
zT7vlV)k{_4?5dLSRZ8)9Q;HZVCB;c87TeVFTBnwjGPRhrSIgCRO6f=5P9CL_OP(qj
zwpb}U$y=`&wUnMyOZNpD={Q~^_D?nPin^|3PpyPf-*lVYRrV(nsk?f?Jgj(nYCrP*
zIQCS@*5%~sW|c&&<=q+S%#fyrYp6<$pDAV6HKla$Ae&=UGJtb%*#$B`NF$%At@bt0
z()+8GM_0A7^1N2`$?{uG=uM)Qs+74Mi^<z5<ZXOkW`@}E?&`=L=^dTCS*R1caGe}b
zk*7g9Vn*J+`j^_aemSzgDd)Mq+47dG9ZmM0?$29mb85*?>SRN(PU_OL<#;$t&JNPa
z5qg{yRn)SPw<CD_o|2F*n|D)(M&7RCyuD>RvqT<IZ<d-N9sguVsbQv^*qSM|ThMz=
ztz0U1(y`P~Ez^*HYIZSk*BpI5Th4Os9I>4H_OX0XYc|X$OT4(79<q&}8*dFBqv)4f
z!0c@93?jN^Ng?yM?()|Af;|<5WTYEE&x?E5SFxWv=t0~Gx8|(cmYyUF-g22;w(T`%
zT=MoXd!S4Lm}ADz>kj8zW6tqWy?7_~A*1WDm*N|>Xx-VF!N0?+K2dXR#Xd;x3IY!E
z+2l2|(0bE@^)y#XdND_nolU#D(@V@=2zPcb<u#y3iJo3FdX`*xr~PkxfjlQ`wYhl`
zoRlX^x3LFHzd&N2(7Q|jZZiE*CG1d~P*ot00*k5DE0XPGb_H3O=35|{^c+n}DiFmh
z_Ijii$%KK-@aV_AGMQ|(q)1-y=f1Wrl*Qz2J#B%EU0EQu%L?V}*&><F`|;kx^eg*t
z4mPLf>1)0iSmjCPXLc|1X8w$M9z$PoC;lm4I{%^W`&faDBx{rT8lfqa_sK=FcMg5U
zd8JY`szx$<7;=^}#(td%+&MeUaOTdA_p%P;?NajgcUSg!Z>9gSaUIMww#C#e8yI}7
zjg^L0@SI==!yR^LG1wl{_Sqwmyqz8EKz)@X_8xb{QIiJnp4b4B7c^jRJiFCyIipLP
zh8VY*nN;NMfdft9bG#WUid*1BtJX-ZVD{ANwkRZTt-rTJOjZXR{n-KgPq{$p>xxDu
zZt!(-L(e6=;bys^nYTNB-f_p<p&s}r*8}Iqc|upq3nkCJ*dObS@+F*a$=mDueW=Ou
z#fLh6SlW^I#-0AK94qi*=HP8=rhbyQGlmCYTzn9cHwIzxl^|%?XApTO5ch@$!rUPc
zU9!7ju3<NNM7pvIMleSifpI30sML!><BTYn&4^-0UL-6pM4@+n6z;Z<#tZFqI3-WV
zL;vaMP0sFinT{ogqk-gTe7PEp)WtEF<`jcoZDR0%JZ$nQ1{K^Hhb~CK(1-+FTo;XG
zMKl7xOh<{uOz2|c(Vczn4OCR#S;Rxj`>>yPJe(ck5jY?o<EO;KVO~5sK8wft`?GNG
z9{C-UfVpneL&wZQFP90}DI+lFpCM@OKLqoB4rKmPFiQXKhx4`iVL)&nlqYq=MCMZG
zZ5O<nFZfKI+F~1bxF}r_{>T+)hPmQil?#UJwTE3*3q0WS)KO&v-0I8Bv2ISdYT<+m
z=X#h?!EBUSj+i{a5&yV3VnB`qZ2z)`C3)LB*$T?0R){R6$CjFSk8(>WuUTS!XG_>-
zS)kPnbDZpKj?d)nh(pY!*v!rW=2r9~Z@+cs&9*6XpnlTNQ%cUBFu`Qbw}#~Hh%{r2
z<+J0JOXTf2deEsoUUyO-NfY#Oi1`$zshnrIH-0mNH`)Mlx49lJ*Vn^(XLi(nWd^zZ
z9~t$eM((i-WaN`-si7XZIlFyoeW{k87x?XxDhcUVB@5CjWku^s@hbZvGs)YGH5IZj
zm;2+H6=HF+LZ-K=klaP((*Jmw^e1P7YBM{%Av0dHshQkUDsLWB7d4w%du7b@8CfW*
z0fq8}S-a1vk(zmc{p!3M26MMp$sJ#EPiFOBr<V2xy{6>tpoUqZOv#k(WbMAzncS-}
zW1=)eDxxxEXfx(Ua89e(LOt-6ba9@QE)V>vjj~S{-7jjM9`gpv*W^>w9&^8I&bhD2
zUV3_#sHGEU+Y0jbOaOVyo$-ZWwfuKgE#~vo68$%IQRHoh!D?Atp^}hje#@D3z-+bL
z7_E|K^Hmc5J5Bbf(&WhNG}&yZl&=dElAWE#uaT+p&QBrrIs5+eN-1XK?S~O6`G>r%
zKR_w5&MLXDP|42PoS8Y>zUZ%!D!xK5YovK8Gd32{w?sYF3htI0{bGLeCgxYrQybor
zY~P|_w!cDLTk`#rw<q1WGw-Mt|3H<jy`z+&SCx|Jrjq~2^w1jq7-#j|HX0d~ppurq
z$zjgzE4FK;+DR*=<;<w$&3fs5jd*jft##)<JwPLSd$U7{yiFo!&)w27gPM6)(1}(_
zwo)f_Zc&bS>}R(<`w5%Z&Su9=wrmL}vp6d{khh8Ct?MzJyyc(cT>d#m(u?GKH&eb0
z$q>gOTA4V4-Yf$?|L}REBlips`)DL5StA+d^jq=Iu;t<mX+K9RyZ39wn7obqPba-5
zWy_t}+48v#_sJ<b+0-OUj<{vX9A<&kp<cZ@DNEXO53+{&m(w|`@1mZr;YxDWDMMNX
z(5E$qI(C~(>E9$%=KPl_Bee8wbC<o3AO8)9Z0=ICrF<*BQsiyk0nV?SWo<olyp`)@
z0cUk>j7~O^x4C|tm;36(jl6Bpiq984`F!ZjE@XS|uAj4?f;;6P-g`Uq86n7+oE=2n
z7&FoO@diGU`mp;Ks8x2!6*u~fj6LaXX6DDvd3oGn6o|`cb}H2r$jyBPV%mxw!^{9N
zByZpF|68&r+3SA39NwHS`<E~)le|63yv$0na?)#hha3v!Z6z}!pB2az#{wC~|Cbm=
zhF&QYzfnaJ&s-1BO-0gdH#wR_re7_TpUny-G^aq+e1FySHa8%fZ=NU;J$>q=$=l`R
zZH-wWbDQ(!fKi?t_(07Ud0V#+yP()(;YZ#^a^{ZWY}}2kHT|bh+~0C;PM~L)`mJ@`
zJ5OIwB0ZT+_d18!bfb;o*OeV&bN|TSZwzsBI=lUb^3FBU5dEJTz`@86A>79q?X|#O
z^IFKxtb;Rs?O=Y~4li5VBWsU69@ch1%2Nk4wRA+`9)6weQxCfg8^C>A16X%=#v$JE
z?tN>3WwjfksiF~99czlO$C{y-yq!0`HC~stLEW`&p~-8D_r~q9;8+JtsOW%z6E3(%
z-l|8qV$M%jv?On*rMsbHm^-Fiaz~P-2hypT>c7VWzsTF{J6^ck(Hl1xd!zYyZ}z=-
zW72kCEVS`MWAe81Ie*Nm^v5H0cZBZm0WEnuhP?eRE(k-G1z~R-(0pQd-gyJDa4h?X
z8V6!%Xcx5H04(qalE_=v&k+dDjezP?B+NcUaQ>TyZJv>^B5!^2qVTmzG(yQ+Kl1jj
z?{v&In~vx<)3I)OG)8`irk^w#Vc$X#{UsVtF3dpGjTj_<j6uKB7_4qS3;QQ?pS&s>
zcT=NbSssmlPR@XJoj8>4oQY}5nP~E3CQ`IB(I{&sb}gI<3*L0?$=gnIcn>z1g|ipq
z5p*{i!!_YplN^rZ72$|X83AS7NPc|7aQ@gZ+&;;C5B4~$Wgp>tb|n?p>jU+ru9$b2
z^)2fJ<10F2<2P6MoOI>qz+U4wt#NWvE3_<bh+U5w;^7`=tRin4l-9?yRA!SkuZx%Y
zHW+f&26eVtWBNL4YzVi;*s0bS@wYW@QHSTy*cw@LYU5(h+UR9f8#dpV+nr(ooVUQr
zZ``H4Fvsl>_Qiwqt&KSnm~)X!-j=h+XKbn|8q!aDlf1q4!vv#pO<;WX|Mtm#HS82f
zHO9^YLqwnA-(Ti0TIR7MkG#G6w>|<b^zrWrJ<QmwhvaZQoE^%30UJHI*z2MGjX&aZ
ztw!!X`7LcMf6MD}oM-D)OF~|?bZu2FZ+NqvxQd<cbE?FMyzM~W-4D*T3s-#?Pv$Ug
zrlzrX`ge(}Um>>KC42n-F3#j_Q3CZA`RrKX9{DSCZ{9qj2Qack&a)3KWO9l4QX6vM
zbD_*?#=O}mYFnwn(eP_`$pPNjCejB)ZA?;Uot!z(Ti%&0c^aQ3rJRc%Kgg7$oie3Q
zy-e{nrYBXMAtBTTzv3)tO5Ks|POZf8YjrmBpA4x{I$AefZZe;7>P6l)A8O?4VvUrO
zC1&L9*xhQG7^ap9&DCPy$a&O_-}h0=z%k_NTa{dGq!tJ6XX|=#XWK+AS9B_g+^7`u
z+e%609aVTktus$0pR3d4DS2zLHBBBFD0$;jh}+dvDSMnM&6+91_qsw>MJwg#G^IQq
zu9TM!O1WvHl-|wAYbEDh@~>h!XJGE0vw}I}@&<d4JJ$b<c?<qGo!J#?Q5{vu<N$?S
z`kp2c^A*x#jY8g$xeE^~WOs&AHt?o<&WYNufl4{7Q%J9!O7UyQk1<ju6JM%iTrHJE
zk-_?LN-2Is9)D3w0liKp%&3fL`M<lfMy51m&PNm8eaYJuoiy^SN0uDjmnq}+GGp@J
zObOY*XCHf=D9O|X^Rh+Bn`-(EoitjYlZVWdTt6mDZdfwUi+(6=7kZ&8wQ|leQ_7eD
zz3xJmj85R5_cndT^gJaDqNa?zy-ZzN;vTgqxo2?hp^?Bv-1+kkee7tu<nia6U<Ri*
zd0S3j)ZU9axj9EC7np1Lbr?Gv8qhDr8@@HQ+M~%^@8?-^iXU%rU;3N;`S-*=LyW2W
zcBXD_hGm8<V|J%C_tk4^WlCye{u*{g1t^(K#y#=J8qVP4?Ofk%S;|@d4ErFO@_mgP
zsgsW6?F906+TAQEzshHZ+gZ}MjPvt9{Qa5D?9NMB?4-|<e(ah!_?CZPsiQK=(#fiH
zot)w`NB9>$f8@|h-iN-TWz=U~<~^G??U%e?C){R7Vq^9h`sPXK$6WSoFduv!_wZ!?
z1@f;u^QX#~yZqrhHC7YyB#Ieacj(tWwuzo%@^(%Vcgf@PWeIaLj`CjXWnU<_eilg9
z17?9Z6o?T$P(8w_u{u{MI(pl8lDBQh+q|XBtI95rz`qLRWOLqu_vOj4EqRhVE>FxZ
z@TTrg-J)}@L{bxHOg&urZDx<Prf2vd{R8i*>#oR^pUk4^^OHH+>G|Sl&e?WDfz08q
z`Ob2_y*Kn*Ei95lzRbj89(s8u{Z3!l)x@sdQ2|v_C$mbLbgL2D=lbZaXMn6G1_%l^
zK<C>AxM^qzU-I@0y}QLq3v_Q?8_t94VDWA{BskjR!ESqOXzT#rXAYQX;fVY59kH&!
z5qHDr8;WqkqP6ugDxd-KzwpMp&KbYXIOECDrr4X-9FKFG<HCYg5c2lO_O@unE?u|t
z?Xkta18$adz~NXIjM?Ra$v&=76uaV7XE(gfcEk3G?wESQ9pM%p*g)PM*y{mf^0w|5
zFP!e+jgUUxC}39AksvROndb$|HT3nK^FnA7Z+si+jdMzG3@`A(;E}%gR^^Azrv7-j
ztrJr71eUcsBkf}V{Nn=<5x}m5F##y`?~GydI%9c700Q?-#a8EVl#*i_{i(>hITfD{
zMWDMxp*%ARHV)Bfwk8T6hjaFw9fcJY(I^ay#-fGMxac_*m#&23-J4K^8HD2M^eI@_
zW-@&4O+-j$3<lY;^DcP?)Dh7b{4NUX$lJQ)?StBJ=tkaVDP|(NVkTyC&Tag0CYs;p
zeVELi?;D9=J_qbL6M<iYBM=`Djt2>UW5}#=m{%2ox_Ke&tPerGrXw*|H5>=q42SJ+
zW_m0ShS9ISC_T~(hsj$Lr%vekmd_+V+C%faH7es;!>p(!Iwv<n*x_dQ-m4i}|7prT
zoTgayiuZ!b`iOC}W$s5^JTa<^<7aG;xy2f@$=m7V?G3){CtBk#^7ccrB_2<<#G4>X
zWPY@O&ov8NJZFIdHw#>CX91V(=D6-)4$~LRwjgUWwlg=G9ePj6+fed0-<?^G&ZhWL
zK@aU?&bW!xB)4X!%UR}^E;d33c2X`PZ*M#@K-g*nY(An-y_-H9$=d|-)`7eoN#33(
zZ@*aSVMQH1BwqR>&nth+!>3guttzFNd6g6zR7;C`)w0B)S`w*2eimOPr(>&_?OG+i
zE|oHJ&rivvPoM!k1L5@2#vUn`khA5IcAg!a7t6(zKA_RzWt<&KWlC15q=lDCjccj6
zk++NJ`Pn?QMB0(J`risAg1s^8skPa^g_@IT%xfG&O$={gL$B+k?NWBvaR(W7GK+py
zb_vwte8oI}1;3Ws{mPKdnas0zLqAJ6HOWtTkIZDQ@i?t?;(hW%Yx+-}nNv|tPY-XR
ze!Dm~p3z7gbxSF2sa?9SmR*Tz@#4NXsws6v4r<wC!CUGql|*xHb$g<ce&m-;2i|OX
zA05c=k4aI<_O(hWJ<oZyOex_ON;y$qAs>`!61OK!x^+vFw&bnH6@`RcPL;X$Q{`?u
zg*5w5AxomTKPGP{j!?=w^0wzsg|sGbJCV01$Xg+At#7HNCpmXGpZcp<?uYq#)U#L1
z=x!=0IjE2b1BDzdPm`Ae71DH-LgsH$$bh{H@f)X<kXuR_7th^t89A&`$etle(e6>o
z<WLoNz$%$yrjj?8mGWgg=iX$#fATGPD?P^a74M)=Y4HV>m=`Ij*W$e0SS2pL%(Cps
zIh&rNvwNtkB6E$S*fYdi=#A|<sn0#@0`m4ec{_mf?=@y{RJ`LJ!8}vy+{}>coV6c(
zU}m*FeZ<U;{<w}g8?n??ui*~<kwzMG-=C&c%iF_h?ts<u`zwDx)M{zY`}xlzjhNCm
z+?2evByXGV*2-P((hb;wIQcJ~WUXa~gJG78(fglAmpx9@Ry+7+iNQT;r1)`K_26?D
z_tERDnf+nJd;T}AG$wQVFhl!_Mk@<DQnOA^)H7<({<+L9V(y8-eeoyGruuHwKa+(6
zm*~W*S}lDm)KXonmWK3z+3!!MA16z`QA;qWguP4FSrYq;T7pB|TQkqPJ(=tAh8+>)
zX&86im#D3}dzJTU-hmfpQkQ*>I`8H54DtWJrf(@MhTf}Kc5yUfmyioHGs)X%>Z*S5
z$4_jgUYgIITgYmA?h>9cAH%AWIT?kylCYk9C2#N2OKh)~FR$C=iyUDN7X3}V2b0HS
zaufDUe0`oL|FD<n>>X;G4>7xv|K?sa;4Yed3eWno)0ngL6aIXk_w-or&65>t^JLK&
zc2_*0zq)ZQyFqg0G4<!OpXA7N@^*X+W_>c7<oY|lFV43Um{qx#^RQlOKK1M5Eb}fO
z=n7;$_t7oS7fQzf=36f>l7{S3I@75_+79?3f4nLsz_&`8`ByU!hn|OM-n**wak8la
zMv=GOZ}96DdAp0V?l?PB806MMC-OGc%nr|b+2O!SJKVW!hl<|zSh?E)y?L)&^4WoT
z!w#_Q<Os*kPAJ?|AExB(!LJSQZk02%<gG7rsMg$VjO4LxFlcof{BdfFmCV;Q>e(Kv
zcC<&^y4)ofcfjO$7mV8Og2Aq?7_D~2l3`?Rf*Wd`c4OC^JA$&@QKy>+PLQ|9wlX&(
zg}E0EJL1mhj?nwu5udwyVN^daoSWl?Jy*O?sqc*uBbcL+=8Zb}KIqcZ7ms57@aCo;
z{`2XCX^EXtQmZo_ehNU7gaEwl6o5&A0hqNt0AtI^-R}Vy<PeUF_Td;s-X@Z_23My-
zlN5nZ))5#vDFPqjB2jBK^M}dX6A4k!+Zv5wE6C#Qkq8Z%hHK<)-Ct9fjXwn`{*y4Z
z3A+GpOoZXWNx+cFD6$O0`0!|KyBLL%vS_s0G6N-BVo<qdCSE?7iCf?4F?ux<xA)D&
z<#lnWMcz7jN8;||2<%@L0iWIx@a6Nss-fd>WJd_IPeRb|`e-;gjzZSNk+@MZ0t?k6
z@OS=TJX;lvvhu!IKDQgPk~+b~u@lN(`(rnU`BB^2^G4eY`#UvbuTL`!+tC>D)WNBS
z*T*Aj)U&7AqyJ`GylQ2OqGfe?*RPAs18i_%zcno0a(BGa3hDs%JjT_*xxhNmTV5OQ
z$=1c`=E&GU?eRW#GBfY^F7u9;u@i6eIo>qtnj@=_-2*$!V7t_e{kF{fE@JlaRa4l<
zF}Kmf6pxt~HR_fLy2hJeej{UaG%-eTlo3vTHpJEjhWK;N03-Rls9wq0)?Xie%k(hr
zID2VVFrU#y51*Ro!H=wMWub?)%WA|UrbbdmSIOg(zeHVKDMM4L<UU#3--5iQPAYGF
zl`M*=k`*t%%gUSI<r80u!0+;KH2b~W%4G$4dw6!a9AZY?LC(AP=mYvXsZ4HsDwVFx
zoST1#S$ii-B#-{Uq0}tq(cAEq&*Tjn<;$pV?3bBJ-DX6#9Oo;Eymh4Bv;HFX$6aP;
zPGXj%YBKpXAX6N5>;~9L?M$0YIat9tB_rei_wA%5@OFDwE3=j|GkCOChR}mLi{6_+
z^0rT@MtYLBS#vbf?SMus$=egHH8PC6t>i6M)lV&-m#JjI5S6@6RLLXqR>z&}`{hcp
ztECnr^7b%!+l=$??Jp{+m!OpIWLaCzkzU**Z~2ub&&k_8yVE46CRI-BDWwy6JLp2H
zB%M!{txXloOG=a525IsuAWd$LNfV=OY2v0-P{XN`J>+dCQ);S4sb$I)ek|{JN6t{m
zw)vcA+p6U6u`0Pj-ag%>5X-Do=}?d+b%GUQx<VmU8x-t#Rmik4<a=Ary3q=`SE!KX
z7g9vGIaPXYPnA_eQ^kK@nw(Ma^C54S{iT#1+Z5DNDJ771>}{vGQ%~ppnXGMBpp>w>
zDrw}cl99H|wq!8fT|153+O3dwc?t=imL-eMXUMei%%66nuW3rUd_JO)!ZCb1V>R;F
zidrwNN>cZ!B>0p{3^d#aFoz^jl_4+Mq={dbG#S$@P4330$-Huf1ZS$`-z{o6X{VM~
z9~II<qn7M;8nNT^#?bW|X=S4lYmaoXqt@K=se)bf3OU8>5!bfqGIV#Q=zY%=f6mjj
zsj2#hyWzU@h;?6`DYkW)XZ?UT?_@2za`<|smB6oBdBFd+R%_)w_1cpTYb4K_olo<q
zOIw{Gfu}R1n7jJ2M9%e3)Ux0c|4b^?;!fV?lDAD$x$AhUmQ@e<V_kS-XRc@Mb=05J
z(`>J#*DHy6BE!>V@Rcmd`9~*v$Z+L4&cyuw|MtqyJ7&uz^79vYyW)~gz8}@e*WuaX
zd_7k-*UOWJ9rEO5a;`ilZ(a1L0b_4-@WO2I_>?X0xxc@`eRTf%T+z#62h%iWMfPRJ
z){0!&TE$Ef_GrCbLr)WR+Jl%;@`1Bw`Mz9g|Cp)OHeXggFOYi&3uXO31u`IvnUv)1
zM6Y~#!k(w7I`kNkw`nc&#KVBuGDFw_rOlSdZ|Qe_NPpB0-pQ$7Od|JNjiwfSKK;tC
znGM36uiNZDO6UF7gm>Lbzw+f4d0TpfGd1%t=Wpgsn0*RUm}T?)3;joKW#SoFCKEKj
z*h9s9oKDrUa#oFO{aPbuo9H9P(Ezjh@axtJ1B}Tqz`h}d7@%*2$I-^vw~*OX&gS^8
zX9w+aJ9urd!}qgx>;bgL_$T%_=j8yOUJh89ZIAi(4oI5i1V5ku<E?s;Gp?+6MubyC
z<fxir)8v-OT+s%<j<&(pZf$Vfpe;O3wMD!7?U)7C0qyR(VC+j5Sa)#6hDomI=jevz
z@oq>y<%XL(-Qn}v9Sd7{Ap36*bo=dr21h-y?3yQB?$VF@ydz>dd!hSaFN93>!m|YW
zg=%}#>*|evnH#zOs1JsW_kr_sA8M|B(ZstG8hLd>yJh|uoy6VqJAeG;6M(dC0oY95
z?kEjFFEV%StuQ423PZy3aLm`63hN6~vC$#|yR9OaXBdH#!z1zVMkEURMPVd)Tg+M2
z{cRL(?T&=<Oc+wzhvU6&3YG><fkBUnh_ji9qMZ{_HhB^jCrri?`zhEFABLqRk(}A1
z;JY^x1Gh$E@RCUQB}HLC|7h%>{_2AxchJ*fq5XFT49MGS>mu;y$Q0=J3dP#6ad=og
z7Q42G;3;|A{>5mdO#KV<b4H@qrjb~FX)sJS1mo1AV00@bZ|8J_-0_F3^2EW)R+#pt
z1v+>)hwqJMc+sj620rIaV74<(lDE%}*T<J)2V{iUBWkNHURv9tR{gq2d~buvU2QOC
z8@=0atdNvItvh);I=&80khkMz*2aLXmI#<-iKXOi=zcTw7|uLz1LhfDH$}xEQ}i86
z-7#|--)u2M9y<s=*^$NHnQwg26n{}GWyg7T_)X5X+e{EO(*#q<+s;PD2t96u<wFgz
z%H9w!ni;^y%m7D{^wG{&AGW2`M$u!tb(tROHq*m>Gd<+h)`PL09txlT5d()m@`AkG
zxbK(DcvC5guk6pWV(;ajD$yrzLri{&XYvoZ{QSGrW9I78wPY#pvzF}6`+1o8G%L!)
zZ$X)ypH)WhKQ&T|n9xhF&%lxFC1uy=+M6Y!I>r2W@-~<{j!!+gBO`Bf`E|m8x1b&L
zZEhLD+14pr(({;UahG1l^*V7noGJaNX&Ip4?vLIO8x_B{a(_6WPNvL!$E=7`8R9jN
zcRBi8t8Q_hyi6;T{-VE>ynSp(-_0@Zk@M8Dle%7~1kSdzG*U5+%x%G2E;Ytmc$*zv
zu99A}RdP63C2e`z+(O>YCf}B=q~^#@DWQ85($PdAKbk1SY@|v~axWY+SSi<frOCo~
zsWRYDs&t;3DqZ`hN{^~kN#zc?_mxyRcrcYY!l}}dGwz=6smwJ@llA0rlXhv+cvYHw
z{-lu9I@~RLE2Tvwcg;i9l1jZ(Yu<)q$=fAKD&|fo#lYvA^jZB)qCTZa`s-9#k(DMk
zofR^PH)135_VFTx^d)cI8Y!gSScTXg{3fNxQ{>3g6lrdrDzAp5%8$fUsb!TWy<*Ag
zK}s1$E*JJv$#u@Om-#-Q%u&hG|CF+cd*=9brF19f*WXmilSxXsuTsdfoeH@yO(Dv$
zsdD&5iezyoZo?b&=`Jef=Tmn^-WK=aexR3HYCl!U2i}Yy^BG3DS|tbfE9A7TLi&@p
z-y5e%?2A-6OWvmNIpvZ?nq)mrlLvkZ*+`b^@tL9JD20sTGmFUzjkLE^$rtVg2AU{E
zDGCW|&)xZYr92v@k@>GQGJyK(dd#yt$@{&!L8c_!qpo>xhM1*r=W&(Z<|wU19^+ig
zXR*&0v@)BFZ0yB6Y0q?dwOk|45zNINn=TWXNiw@L8BeB$kh5X*h&|ym)|?Y62|KHj
z&KFg(`zoJFZu0qu&uC7KHL{91pEsAJOFin*llXS8FG`o&jhUzUI9+a?VMo<ZKAU)v
zyNz_>pwNoL4z1jwr+RiD?!S-dm{+Ega^8+xwxTy`ZjO9@#k)H5ykD?;s>g^d>6)RF
zR`i{3V7}I5=Dc+N!R(HB?s7L|OY53!8B5-(mvh$r#d)+OPd2Y+ZU{3q4&KbBKRR3D
zpXJEFjnrf{;EgweEKJCk$IN!_M&4fLOnb2-=ikIUX*M-i!cUU%Ce&B|kGH(Z`lnNq
zM&4TRZFS?`$@V(A|B;y=!aOpve%4ubFQ}La#`#u{T^7x$*V3u-We;z-7a|L!wsWC)
zv@Vp`b?j9vDU^r3*$FYIn7JO`rFEAd;>PYj)$S_kXTfKEy+6`#ygs_Cm>=8Q03Vkb
zVDd}Ox<v-~dCL%!ZW-au0^ZcF)WHNZ8%VSr?8w{JkL>W4yuEYT9uw<0;DjGDGRWH`
z^7h(jC)`PNg4z0duxwNxO8SOAIyb_NqUN|=*#bAHQTDjr8ZUpe#>gLSkiE4n7W{6D
zj@ccs@s11jzH)(kGgs_`E5^TfMgKTAydZDOwz)IE*B$SS=*R8jf%te2lty`?m|2f=
z$=lP<I$~5)FB~4^h2$w-xShzX$7No)ci)@)NpDoW^~Ux3KKT5>hkhGhG;Y`l{V)0>
zW(oJHobR@AuWC&Oe?Q3%gq;C!C=5VmK>+6e7luJgID`6!BjaZ{*4_(4<E3FRGY*5g
zDioJrhoMDSI2;yD#W?$^&@~E&dE?1Y&6tehzrye)dJ3HSaHefH5hczO@W5&!F0P-5
zWww)GIDRtB$lDcxVelF;4IP{!uxMW-8j!dB7Dd8jRTLf#jK;*>)1i4b1O8iQAns2j
z)^L|RUloD$*a&iO5)OMzLUr=rn3ci|i$(0s^B#d!{=;yLKA&E-`m-acFTQ+YFYv0K
zSht`jZqM$9D_TG7Yv_j?7Cy)fYKe7ETHrv3#;7TD#>FgW^qT98TrcK{jdVi7^?F!0
z(-H4S+vCa_TU^Ss!IVWdsPDx4?Llk!owTBth%>HX9hiKwMBR(E@N%pL&za`vyu}=u
zw&w6YXa>C@W+>&p_%(Ukm%R1s!Q5k=3Hlx|!P$xYwuuRQkW8@amob|1+k4bT1&ImL
zqWNQC#u(hf7$3@c?}#u&S9ZT2xod!B@dh|+XuzH-efBo#V+^|mb{^J4&Qd+@ZFsXa
z*2BOh{JY+~T9%Efl3UHIWLW7>=`i+}EOw*6=fE!sVy~Vdeu(GM3Q=|Y$=TqCykmE@
z9(y|7**ntYM44#0Gj6`0dZ=gY@TB&5L1Shileb@}jfzWUe<-^=N4zMJS<G(TP9Nay
z3G77k&zE^=dD4}d<g3iUTe6xy32$co@Za}1^7b2fyO#H};kz=$kh{Y531sURYNT#u
z%1G`Fy_RIkfBYJIki9z=Co{zLFKUU%+kO9P<@rkPlt*c02eTU+TT+X|Y$yxPOno`8
z_z$PQb-G4u$=j>VHS+QveYe}Fk1A5hrU;dEBX3Ln`F-wE&-3p2p0i+BU8Nj9s*u34
zG+E-TkdaFilHW}!KL(`96QoJrTk@1#eLp!>8hWM5qW7uXyQYd_XR6%ew+C0G%71rK
z<#}1E=##~5$=&R>Y4U7InhYRszmliPUP>vNs+3PVlww6DCl;oN*P#@-(Ktok$9|K`
z?%%|XytTWYBIlDiGyA7W_bF*oK3O66*K?OWmvbR`d*wr#44nQ|CLRAO)0d=3?Huyf
zC{;rGrb<;ps!SnoqdTWb)m?>jiQ)|0TP6C3$;{upH-A*h{t29WPbwviyuD|vk^@JT
z?A}pIKoRHX7Sv|(^Itb9Rhn-~kw<k>WcoCI?sHRREce%Ur>Nw2kXpjKs^!~5&d2=y
z4B&kJp89F8Lwru*eE-HeRd!mY$e&T)Wa{@Xa^d$ES(cX~Lm#Bc4M&BDj=wK13K{a3
zLROHss%5+Zn2<ryN_NI7<%Y3BMjI(4h<`4H{IgWO)W~^qbiRJ3#H+OQHftrMnO0O}
z@5{q{2BW7aoLcU?hqU6z8TaEq%-(#IE(fVsH%-!rnt%2aHgQ+HLoK@vsZ-<5V{Z!Y
z3dw58+o+a|0Pf%Ss+dR2&xgF-^`A<TZ>U7mSuNwT)Do=L$ofgl3p<f6l{eEx$lJ3I
z=->X$T{v?(&XBkI&8S~?)=AVytxVml6|)W0T+=(A&fJnUeRPtzox1C5>0-~fyZeDk
zcJPipnY>ji`S+eT>UP}SHzIF0cG1ab3qF(Hqdsi~y-?Nk^|DKG;4=0<$MU8d!rmmZ
zxAZA>Se!*ow=;YCdA1ZM=1NzmeCbc#=J5t>u{dAyC*@07f96Pg^4`miqR=V1@{QW9
z-PDnX{-G~f&F9`X^b&8Q|Em*kJR9lrx|S^s`C|)wm|;!cCXlz5%b8_KP26VQe%4XD
z)q^|b+a?7vi(cUm^{LBX)~4g?e9=EZee@jarY{xC(T>!4bp9df)LXu_`7KeUzolVW
zjl7$$hxW;O@O-a_rw#_#`oaLyei~p8^JNRmjWI013=J;UL5Z;q<}r6Ze5x(xamM*>
zYRleSTlglhufBr=9_#G!hrFFK#0jl9<9e^Fhn>Oo(XY2NPV90<`>W1)&m5v<<ZZu8
ztzoBajqT^#AULcY&dqI)k-84pIo<_9b6nv0(*+Z~TroS@6{dl181m8${nohS)fsmj
zEO*D4)*etNc;H#ECo>^Ekrdw%{cm){f5jc?+3AQa6FJ{Tc;PoYi0<$4!UR2U?A`4R
zYi3Y2eB^_Dt9-GF9f0d&{BYl%daAkpP;sxis6{6%S=<S6J3Haso=)hc3qb$2VW`_E
z3~Taue>DxmfL>v69ubDGrJ<<K2*m>O_DN_sa(t)YSMFpCXge9t>?UJ<r^&efWHM46
zr@)n&iu!FPVs(QFaJHC;ZYw9^>hp=1+GrByM@)w6(NM$>orWV-Q}N9=3>Wl6QQtHI
z&7Vi&P9;4^o2TPa*mNBCkAgvJ1e%bywG`8^q*EAvmW86#oKX177(8!024h`^V@EA^
z`r7rwIM*I<_|+Zh*Sn*}wL3Zw4utMYSE$3fz>9HElipCvP2R2uXo;S8T3}|Y#@LnG
z5Mk#UAb>Xk_xkm5iqB7lm+PTpEWKY<_NcSo76)@|5KP`?*0+Ius5K59vcjWbR!GsS
z1AX$g(dAm0G~R+;4d$pOZ<Dz<t~z3d;K63-N#1U|VTz9j*w<rkO7D&d+U+;NN%FP_
zdAppPolf4C7aK!sOHHyjU*RVBHI=($@^&CK$=h1dciYeiIktw#k2An(^0sKXJ}f--
zQKv```}ga?Y>pmExJ%X;=~27(TkcYK{A*LCgkSkZo$pW4SpJj)?dhxKoV$WEZTq+%
zVsfNHT+eZLOh0Ysk>zq}RjKr0Mx66h=Awr(6MYqP8rf|U{D+<3><!ELSt4}cN?&Hl
zB{`MKw)-U#(yCaZjxis8Vxgql@!ve3t4IAWAE<_TyL`6az&_h6++iL3&Wu6sjQ^K=
z^mqq%gv>k`K8sr954<1p$JX(`*LCEMnL4SoYs_lAuazgvPz>knvT(6h2KA$+wF|S6
zssGs&LatJCo5Z`@6W&Eb576_&*Q;G>xjB_PRL-&;Ij_zmV^6nMiGM#8{TwPe)<`9m
z^;P0@MJY3CE2Z;2g}7`?lXbNfa*k|Man_rzO_gC6Q<$}oD$#Ao)sR%_+mT$GpDG*Y
zq)Hn3mbQm%OiGoh#;Kwno+`6`rOJ!!RC(%_CT0uMm=lpIjW~M_{hT5@O;hB!VT#l(
zOqK5BZRNofS>cpI{(qCYufIzBm~ZULOA&2GitKmgE_PI!T#rbTvrAK?{*7-^W%5m)
z9sMdHQ@+Z-yT0-{=d0{F`b|3Je3ME1_&)Gsdbfc$W;e2SSemRH&F2Qrk>^7dVp6W;
zOso{A0ZMVU|03y?pXFt%|3}hU$3?k!U7top5D*cIR4GXVYi$d=j!7ue*m~?P?C$P7
z$F5^{x70Kp5>ko*SiI|gpFidUA`ZhibA5k%@3q!eWB3d;4nO2avayj1Yg}rt-wjNI
zy{8t(40M<oqs5?oI&`HrtaC&rhIHri?3;zo6WAkPLjDg+gV~@|<PFu~Z}QoLysgFG
zap?&DPPuC8ywnI44Hk|hvswQ(<m0pZCl!xZrK5;*=Xswp;A=o;*UiACLLK|F{P`wn
zc+LC$+JI+_qx3pA;aosPzcYIpUCG<0U#YWVMpx_d92|G!?nTb>y#sSO<Dia?J0qqI
z&q2h#Y`o)RJ<k8{82YBpKFmNiKN=NgAYyGMPH+bPoUe)KvJC7XZ_`im^+D#IWG|o*
zYuq-8%o~}R1)DtPfkyC5cZz=MTg(}mk%PDv*_f^6KB<}cxI*p0*vxz!IzT-b`M!QV
zz0schJ8SmenMYzmueXA1H|G6)iOEDLYiZ@CEL5d3SB1QN7)~7*XTd*mn6cH7yk)LO
z+U+8Yd&9gAdV8m5GQU!|n<<|A6zI!Y$vJS<vm&J5D5SQEbL@lyY?{IO>u2VbETkVP
zmf0lq0r&q#Z&O41m&pC%e0t}o9a9eGyn8)0X5|I=&YX|UWa~8QwN|&GU)h1PXWl-(
zrOzv-029X+VEJrjY<bh~)rGz=>azc*pVLr#eeGNkI#++hCjAmD8DD}L-p*B);9>7l
zly@(~0cxm9|Cej?zsyR{N{pRd4fUmJggp6)%UikQkNT26>-FT@RXw?5V<68v8%p+6
zV>ubZ9V?nz64Af5^gUWzrYtrWI|nN<akiF>A@#*@zK!f|ZYwWS8b~i&Te&gIR$6^-
zDC;)Z$<=|4r6GA+|6NmgTBo_R&TT1ICb-JkE3P8s?E>=l?`RKk>EkIrXFTP|NH1y3
zUAae+y(HexTLv8WmTT>OWZpX;saoMHJ2}H${J>Y97Whi43$rf<`pdJ8-0#=5wLHJo
zTKXi=FPht09!9pIN41R<ENvsz8`{XnpKWCF-nP=knzilVcGB%fJ289}Ag?C1m!75V
z#bkB|sr9;pY_#nto)bFCjny4x<(iIC7}8O!kHkr(B2Kogj1$WbaWdYE^{w|<*;N`R
zdAV_N>P@VS4vyvi!&qrk93y?m+poFNQc@T#wh^pxSH?*Dswipc93|=XA|)j+Ql@v0
zqUMWzWxHre@rjYEtz+mrj*)z)Sh<-ME9JjqW!YcjCH=%CnOZSfE^VDEA4g1;v!0X1
zZ1V(ZpUlT}GEQzb8Y{OA#!69IxNIp5mA{9Nl+YK$Wk$b&QuvkG9-iEjSJgx8?o*%b
z+e3D%yGi(`0QsX1kc3+Sa-Ux1XQBQQPTt;l&|I8#_OidSsVq9xME>+^B8Gl;a=y@3
zW|`PZp?3p06=)+v9P7*3LMwT>%t{uMw+~`0#b<6kNjPpHkH}lQ<hs(8yfxI+k?!Q}
zgq^izm|1NpInQ1(=eO~{O=ZzTQ*obfDtY9soHCJ3i%sN07xs<+7|XI#)avCJ%M$X|
z&&EVD{Y~VEA+s98jpenAG3VSyvM|g@_Hf>Dn3<){;|*lXAAPagq%R$t>WkMV&TscK
zM{1s)l>S=-i?M%DGwC-@Ui^jqO7@Cts&F8(3L%x1aNA!=oo*!tzW<Il6PV4|;5!n|
zac?d2WyW)-NW+*4T<Sp`R6zyKe`IFV+%GsukI$|$=Ey}e12*Cl3J!llUR~}Bv#r3|
zfo0GxF2nXiW!PH7-6`yS-6d};sQviLYk?`fiKRBoPsl68-+aHfA#Zn&<c#)U9uhj{
z!7hyYms8xAbLIo$=Y4=P^*0_#^mRnDmwY4_b9n2*8s^U+<|dK1t?1```G@nww%IVW
zV78N%T9+&T=lz*L-WKg)W(0YAgni?GsV_F?wKc_-*IvI&ENhdAc=FcOA`^F)@jCpA
zTBMiha9NXvf9j^A>%4T#+@6l_4^xqQMu$fQT5K<2zt>ua5T8^ShomB$Y-<~rimTUj
z%!g!6_f&_&tTi|Pn~Ki`sW4|hdFr$@q?M-P5IG${-cBKJW67{;-dF1cEyg_2qTs9+
zzBXj$Y7JgZ)#CU89fq<V^zleV8Sl6C9UaC_*3r18g_vsbY`zBijvDmaqG2YH1`CF3
z5d2bukLIb|!;^{?>sdQLN<$a@H1r~Cf3c1XU|sA<-Bl6ivy;B3b8nuSGfNfJEz}5@
zqsG+hYE1l~MyF5>{MKmj!$k`(5B@x@219rTh$C-ZS!Zv%!G3!!)`?~GM6s6rX~E|#
zfb}+cyMVl1SX+mVmOL-$>u^0zgQprbCLL7cz%?}#zg0M9ql4p>RLne)iY`3IG~}%_
zpTF4Ke2(Afkf+h1c|L3E)2Rq1Z)Xf*kDqm}hT1X*_Huu;V4cfZ?2AX-fpmboo{H(Y
zcIEkNKrW`6k+Vmrzi!NV{xH5asGHl@BMbWNdA@p<fo9}wJ@R%m^=2c_kh!OL?zotN
z56SHFt1>X2yluzce)w4Ws!ouJmv~O=l7+)9vrxw0x#t(^>dta5p~}KQp6^UKe{M+~
z{8?%S9H{#WBX6zdao^O6Y%C*h&sAmNGw0%gd~MDmZw->^4eLeC+^I}lSwc<LZ+gB*
zbI$}l(!Fm{cV_zm`rH#`OD*<2X1$Dj$ulYEyOsg0V}}=_s0aJZQiwggjdWq{>r5SV
z1+!t8g>`^=V1uZEZnLt8`l%wgled$|TTAluxhwZ8$K~M~b>s&FIlnI8-i$bEzS!r!
z#(B<|p9MH+O~11}^TPQ3p8piUS(T3^aoi_C-p<;W4;B4g+NgX?4I#U!$zH`Qt?ryh
z2XP*3NuAb*z9o1;9o4a(rEvBy!%5?EG)to&$@>#`6;@#2rAqW_R*jK!e`EOe-&nkr
z^Q|>{a^s$!WYjef2iCYQ+z-08w~5T*%<Ve0q!Y+nlSSs@*UU;*Igz*h>r3Da8_{#J
zl{v``WJg0=W;NK#+b<2}e1k@k<I_Y6dp4C;iA|-Xc5~^K%Dq9IT&2Z%R|zC<--o(M
zoxbi8LEThc^0wVjFDYHh`E8<?d}X$>6?r?r$47de^bwyWzS8Zuuk57{_i=`=G-=`|
zJ6o{MP4ts&Q-4Xh(ppB4x83qu%OUc%{}lF>7q=0+Wo<-Z-$wGJjeMQimj2OpV)wnB
zbhsQKZvETKa`JXgL<br8vV&CD?<glCJ4&0Vj`Aj~qnxirP0^t^@!Jt6)vO`=zKRnk
z*Re9N73)gYwu4^B$)cOF+`AJiOSLibJ}yQ&oQak*&!gmK;|R&oh0D!(V`NLKaGAa}
zT+Zr6O5KD=ar7l~JtJk?rwAFX7bT0y+iB#j+shc)xG`2v{KZ*sfA**y$H~aMvC`^l
zw5+%nEpy26gWfUH3$e1HL#!N1kCv&y5mMDPSdKdcOB<8n5>zuxI%Exyo)de?oa&zP
zFuSL8>C{tNIP{Qv`66Q4PVP`ow$IaFhSm7VlJhO)*{>Eda7YXJ($_^!jc+C=Sl>SE
z-$ZJ^X(Zze8%nEYwzA2if#~<NksY4(rBayt8Du3pvn{#P!cuxLGwRP#&K&eDr2J!D
zdA`<+`xNTPYV!6Gb;#pOIG^>cC8Ju@lA>x;IZEDcoXx$F4Nb+Byj{M;L=KU+xu1<?
z#(wUk<sSIf0mjlppE*$<jHGRmk=zI~mRT*0=?gQG+QCM$Z=Rv#R~xXm%-elxkoW1y
z#C7Z)uho;0d$}i`ybV5GgY)D6U}pgRwZ^}2sj?c^i>h!Uq6#f4DlvF(C1T7fxex6-
z#!UK-FWz5}ypDNL`znCt6>#gy{PWV!a7w7aQTB}|+i;iWC+-?;TaF0w)}Oq+c96U!
za~%&d=dB<8eZFPL&nSTpdE1Jd73v=ErE+h~9_9v`GynZNGx|<a7h1-wJbE;WOXw#d
zZ!_BFA&Bf<z3>D4=(DYyl7}rdd2pc?rZat%1KAr+45W|c2xqdqwu0W7<<zY-jmtsT
zHq`H!XJfa9*Tz#>_;frI=ejZrq6vGS*3{sdWg?TcU2pbJo!CDu_T_c7HGM$j?Fn;!
zPdw*@<n66z=~x(<hF|Gv=sPDJpE)z_Gf0Pt8ZB-_Y7u!}ivwS@7~Yt@*8Zu8snOy0
zZXE`#)!{5s>FFS6)H-b4n~HJCWQubd=5V%pXtj>nhdS&av&UW4LbX$iMZ2{qzpq7S
z@^<GLE!uw6V0F9(&3JoXTZ@S$I#{}=BB2vm{YQ(@;c8^`Rby)>H9oXbBe91X9-B2-
zT1SI+Ei~BGM}vIw_Q|wVOk$0Ffc4}4npD&YOv9{gI+$(I;>`;7$s;-cO;zKIxf)qt
zRk)O<g25{lY*(m}Y@xxv!D@6_szKg<4RTZ(DBfzIw_Ss<dm4OcnT~do(vfnKda<-L
zOnZ_BRI6c~tfsF?jV9!4(^*>NW@?dVs70W?29YW?Di^BpI82Rxzf}0XT!jYTRk%@0
zhaP*`H#niglYeyR!yfv7<ZUv~BnMyWP`I0VZ)(j#6FC2_NQdo-6cqPJ!S753Z2Gbv
zJB|6!<n8?pIk>TqHEu^T_ap1i^~}sBZ|gS5#y!@i8>!Wb<y^U!Ivvlc1AAbTfzaa_
zFgrjE8T;tr<gWfrevP^~8_u=c@pW^SIU;q~fA9Y<^=+l>i_kNjNZtnL@XvgG{lmF$
z?<D4w*;DUD-nx*t=d-ghk@a*}>IOC}&&C$=c34gp4B2yU7?y#eYHHq^WS}2klcwZt
zbI$I)^r@c@%E4y+T;#Kdvy10THv{fqY?_a-+4;yM*YxNEK1C0co2CFse9z>PwVMZ$
zdu#Kd)Mrgvl8<2Wc8OOJl9@%4zJfah$y+n}mu|b0shq_OWsiMB=R7!&w{hg{a^|Yk
zb71z?@B%EP?`i;fJCj@wX4c2RX55i<ssO?CMm5(mf0X_^{U>=yf02jUMS1u}t$2sP
z0^Ex!#D-PO4%u4(N9xvBfxTo~`h@LE5${9akzqMjGIK-e%8clfUon45C5Cq5Zj(Vj
zkhY0A%hZ$18m%YmF6+syp42M48H#e2k#t&VEP?v9q+loa9Y|e?jxd)a-z{ZAZ!7Np
zww9C&^<~Vs2BLY=KtkA4{#jICa=zD>iOp@K;dxs*f2gS>)Ndv~S~iy@x|R~(%2l4S
zz76~BDxXKXNfUo}`BdUA&3kyrzDFJs-?x?YdFmxI$lDhuyrp|9A1OQGBVLPrr8#-K
z@4T;Udd<8k@^&qGdnmzA+{uliQU3Bb^~#5hTFdv9tz|`Q8+k!*Q0swh<bW0H-PeAi
zFKy+We_K(}BNQLkR>q!gFHiH^i(4!9m2KP0MAP;%`2}aYcLQYSVdg>}43LlmaiU%t
zCoWUs<mjU~Ip8#wb>mohmlG#-$=`@$v2v<&thndKh?+Co)~YC}UKJ(N{tlP-uc%|{
zH%2ZShKqVexLEy)kb5x^66g~tw^~F><M$CV;Ae!i%?+2AM<ZledW75wj1-sSk>c|^
zQqB*ej`>KGxWq=waPqb$Gg@|%x4T=!vX2@g^>U(_!5_i>LBX;laHJ^KkC2e|L*?hn
zL9+B@PdPrdr)0(l%Ix?avT9>@85`PNdZc%en<qQUe#16$X|uO%b#Ez`zP6BYvz#S#
zoRbWqzO8jkGkF@+MEaArOUPS?cD6EYZUZS~7V;(PlYe%%mcI<FWS@nloO)1Cf}1l-
zN@XE8j4Y&Yab2lB+f0JV+n}Gd#p_~i`NN#XedMhXdE1h_eYnk3j`uTVJ!~rb3QeTm
z0u#B`z(gX7j3sTju^5{h%ipJsWEXiG%v`hte(tR>l*Nk-#e=-f8DS)AS{X@ca|3#x
z^d$a)9`{%2iO+aF@tdV5n&En~w6~rdJy(NB`fUds{Xu_xHKJ}+!>O<ev-?$HWLhN}
z?);8(>=SQk{RLmO)E6_aqI5C$iHxm4Q6O`V+gEVs*=MvmT7jUv3Iy}dCyYMhWkng9
zt!MV!i874HEyu#P>=`k`=0+s<o6cv36nT4*+Nkc-P<|k9t99H@zMEb{o{1IZ%q2QO
z=Dwhglf1o8ug~jt^iEP|l0{F6D{r-VO~~TxZSf_hWm8+)f<DzFce1dK`|}Qyx30_u
zKEge(hZaybN}tKvcdTP?(9gr1huH_1^K>K=t*Gzm+=%rod3%k#O(SouH5quPN3B%z
zOk}yRuT0*)otTbzYI`%LkX1?S?Ve1<?roeiTIt|sr^B?qI@C$g!v3HJN3W2RpS37$
z#(J|U*)UFrRPwg@4J~e`>+pOQYqW+s4BXF}_<1U>7;vWhOov^obZAQ6cKV>jG~UP7
z^;%@L(qi#f4YQ9la2dm|7ilnkkp`yAG_d?fg8_$FYtB-`YrPu7maEZmq8jhX+s*CO
z_}fzr(*V|tz0_#<R?R&~YTTHmp=V2ryL@clR_ZXFHTKSVIv5A(@Op<9@yp2f%WA9}
ztH%7pY78w?A*MuyJ*g^8d8uL@tHRDI6<!WhW5`V2?*ldLp7HTK)F5W721Vp;N;B5r
zZFSgkJPmCxreZPs&+8J^Fd=WB7O7FYy9O&g$>u2<w4I}YLo*GW->I>9ff_wWtI?G8
zbeDJ)I(=55$JZ2$d!2&Xk5b@PUx6qiC0>0}qE3Vg1+1?P*VEJ7U5&z>N<{GQb55q<
zM3)pCd#S*@3KhKh952dFMc>-#c*@to<94|yE~JK<c^bbvaVD$Cgwq-3V5-vbdUra$
zzDdW$qIBHr&RX9g9qaVT;1BFqkIO&^>u2Tg40zdRFgGFteL2@3y_deM!<m@yhFUiA
z_B?s(n8AK9HE8?HnCCez9Shi(pTfR-J)R5f>ZBvTD}TngY;2~l_zZbFq=55Z9qZ+V
z^k6%t!*^Xe`!Jc<$k*rViS(z1vbWElaeoB!H`Sc^(qF!W^IkO0hx^og=q@r(i`xB-
z+X`^EKl3(Te8AlAAGqHtpF6Gck;hz-N1XXy+se5yd%%nOac6Had%+<EaJZb0H{Ps~
zkI*aig!?N(xo0Dr9;dVPc{ky%X6hh*P;+e^mWxX~zg9)kn@zp{m$^L4-exxHan{1B
zJUpZSE^8h=0nhXBq>#IxIpZ-W$A>a!<Q!+se@!kxG<{HWc#iH*{nbB?r8s`C6!zRp
z>>Eap!f&48*)RL%Rtd9rRoo5n11&E9#M+U+u{8A$?w!+<*~W&Fy4g_lU5#Y_GG6mq
zno2-gEjbuiTkefC7vmC3iSK14v&h?UU42=^>+ruHZRFF9`qG2EP5Dt@Y^<0`b<~y_
z#+==9Pp++NbFtgzB9qfyBq)KI_cgA9er_^&IQ>Ca++@pVcbQ!5Asc$MlJo1lWK<Jx
znZWuskoE2UQ$BL5+DD!Q_=@s?uT<agl^3^|_4vbAYC`?w)n`BX&b_!UAN*xKdAoqT
z%{bOt%p%$_+rEv2nUTAX{iMF5zx?J-zv6@bGW9=yDW}%z++%-f$lLBaWc4L~Ii2Yz
z-=_G<)vkVW4}Ovu&))6daWb`ioaBFq6<hPMvg$YUiSNfrAZt(S&9U;PW2{&_jFD08
zV<h}+loZED$@aKN*?>rysy{~VEgK{6M}^C|>Im5r5h3G_g-L7HyNMwY(&v1*nCFB`
z{GkY$N8Zlt8Y#iCk)pg9DTO7>iE0)l9<fn!XI_*T+>Msa<gF9y;PX-JNB4@95xc{r
za&DL`{}U=#EJ7vxa<I(F7%De+50UZaJ>*>mq+|i)s4e95@=nrbQYTq6vYm|l)J6=?
z`O5ya+zoNTOKt=^NtTI|+}`ORCptSw?}%nH^<fjKU}n_*4z?2avVlmw`f?-9S|U1H
z%hO6raUpO2B5xZy*OOR<h3vDmkb8;doavj%kyCZ#uiv%B|9`wKswEvPYDwD%rn2=i
z^-|>Rl|H62lDyqrU?R;<m`Hu{_S7n4sY~82sB0`!ju^@3*?j#SFcRB?M$9@hl+g<e
zxnIgqeia$WB>HambT<%7^7b%!yNbP;fH*x_HkEpWo_gYTtp*v(YG88f59gyl@Zx<n
zt~ga=_Ngk2>QjZ>CzbSMe#Zgs?HNwq_N9)$;7SGBGn1|DxC$J4@EQJ>KI7W<&&WDb
zf$iM+bv2*@YU9uFE-k}NYM_4hD@GNy8^(_~E7Ipa&@gH$_ZHzScaP_7_{f@vJ*R!_
zZBcKrz?(bEOp0)|jCuI1Z`VJkzJ$EJ^ED5)ZMcJgng3mHanH}99E7#vo;k}L%)LgB
zC;M}2=*hUkK5;Z_mUGM_x@E}z+wmOOGgqoEd0R%_THd7AcX<}tR%aq-2^mWb%*iTV
z=N@I?$Z2YCk8wt;&cMekYJG>KV{RAvcs$cF`B)m>eoTdyyq&_{X3{M!cCqgC{jG%s
zd!r+I=}^YL@3YA&T#r^E<Bke5H>eREtH$ak8g$ZVaF#Xe<Wlxt*+(5{OZ^g=?e&be
zpE{_Y>+pd5t4-d1CvTtstHqWmelN9A9mv}xa&*9AH9m3988=sh<Qy`U3{xIZ;}W?y
za*Y~a<JtS=_bniAU0ZV=>dc?5ug0pLYMec(#!)_&j3XK}AaCQSyDCdl!;j1xLjETV
z&~ZP57PD@$4jr$?lV~-TvcJ16ON9p?RhUlRdc06UrBUJNH`bng)Y!<M`{=eBVI4F$
z#Tt4p`__HQ-4RW-&?&VT$a*@{N&_eI*61<M1?$vktX0FMlLi4E8rWNEkXl=VLKAZH
zjT$X|)v)ZQMt5@3wOE0JZVEj4l7i9XZR>j}=tiE7B5!Y0C^2ue3Xi`iVOEQM_U>xL
zj!~lgX)><0PDaY&WK54w#`o39h$u^eZcG~Y`leGSPwg=G6+ioyjfK;*@x>wwyI(Qq
z`ZVXS@6+MP*+3iCxC?kwb!5-NH63fnTS=jhdTKhpY@^5OTsr(7l9|*;&*nL-avSqW
z{^uEE*2$zN87O8AJm~!YvaIQ|;+%QVU~=3g6(6mr#b)-&Pd<(Vd`(r8sn=eoqZw=R
zO-AIkI+HzA_EzlpdYh0A#Rblh$=IL?)Xej~qF6&O;Q62vc{?eS=g#NM?qDADgi-m}
zagK9Y_Hz?mxzA}bHT&f4RPuHnd3$@%2RsR49w+<4^~l>P`w9`1ScrFp%tu{VfPkU7
z+$Y7XQ1$@EbDz|e9yyTv^e}O@J-jjX5Y+e&UZ0E2!*ik6DHmnrZvL=bD9PK0<ZT9Z
zXlbmChxBL8W@sK(++(H*d0Xtwea!Ui9a@x+OYY3pB5#jgqxYD-YkhO(P_n)q^0E}C
z50~N4xKB8({}pG++i3DOfV>?+-afzd6C2i1Q$pSv{HrJ1m@)N|yzM~V8q7DAFK#9>
zs%tG-+_Sb^cQR)$)lypiu#|(dtfa`wT7nbn%RAPzUgYg$@>WUSKDV?Hzr(g-oX}Ki
z2R7r5)8^85lZzxPT%<R7yNtZu-@{E}-nEpOhHmnmo?IP0xnp~K$hq;%zj)Y6rf0Si
z3+@V<P2PSj_u(ATS5|~GYihrbBzE?drIloG6EA5{%S-P5XeC+SJw<iXllp28u{z=*
z%dWd~kDZ$=?Al5qw)@KN+ddMr)kj7*rsmniN1BfE7UK|avFYk9uRD87Id3Cou!p_M
zS6(&slilQP_cJlFrYo5m9V_bhF>>@*oP6c2Y+syIy^H1k!dThG8EpEKXlbgW-)Bmc
z^j{psT}qMiu6LxEm4->T&SRvcOPF}uhe^iO(Q@_AXnC<ET(%F7kmnP}i2s-|@?ZEE
z+3Fb~mWGkCFep;=&qc~*@-~~iof94<QRHnH`_xMoMoIfQQF8iGq=eRu6kVrq`7k(4
zhK?I0PF|sM(>#dt(V_Cs+95(Kg0wWE?z$*I0^YTgw}actpEGUbkZBw4^YN2ULB3L-
zy!~kJEr%y~$&<m&<?AUY+2!gejmFzc(}Bzg>ue`yHZ_#><ZYHkeYucjEhE-jNkxIB
zROu{a-h2yLz0pDryt9x*1A2k9=Hk1$uBcDfkt56Ni08OEa*4dn_pBwCuA9p63#Rgf
zyzNNdE=V?!)!R(uzfmS)M{QJxQN|LYFk-KOTBsdHa)!KJ6U2Sa)rMla*HAuCm)wxN
zwJ&DI#Ulf`<zpa$)Ft=osxQt<^+e&XCsoXDeBDD&<Px>_i)!G1<OjBH{=w^KHP$#*
zqvg3OxDKzvg|n3i@BbA;O1V?uTm=+|xI^z`1<G<hp;b0BX`YlLW<fbd*>e9Ewf>ur
zQHy9+iob7^AZ%MPMxHLlm-fZ5p-)h8i~C}le8gtfvom*ahg-josI!85(-MoAlgGbX
z&HP>R_F@V1Y3c9pGM+gAPnnaa%!O_Vch)YXPcnh;aqf>h$m_zomdvl<>}~&X`Z&nj
z(oWeJah$VR&ceo%#}6vG17|mVkm2NS-E4HIWPh)WI|w*mE8oi5EcH=)JJJJFnE~xR
zYFw!ivN)810i4qvJ)RDI@^*rx<0G%TsqEQqKgpRaYr)2T)HAVGU35o_^S`vXHB^hw
zW*XE!p~kEj6`n<@kakpszS~ro%0BEP)^GQVHE6_s@GR<;MwziEYOX`@Q_gw^ldGky
z3-dHE$<n|#Lxa7%9jc?nbO$XA$-YhO2{+?M1z9?Sb*fc!{;W(j{&}i~&wln{|5oGS
zN;Up0Q{zH}8vFaIk>;YtA@+hlkzL%x4*lb5WWQ6vx2_7oZB=MJQcX=bYttzz*mYFn
z#TYdrmZ*`rTFpMA8oMLam_n9+JFdcswfvoysL+JpyZ-{2o328)Ks9!fw+XlS7=kp|
zIYEQE?KIfkSA$7|H8?qzGiFEjjmgyUC4Ah}StYL}PdTS8=l4x;)4;Z#1~(0<=c?fS
zkjI7hRWRY>n{q^nl?4izIw`Pga|#T*rr^oh6wK!5KR%`4pSDW;yFrPItaDqIDzMB&
zf%p7gzw60p=$Z_l&}6h<kPHQRyCanU2mBq%SeK6tOULZHoFS99Nz<rNN~3<7=cu2%
zsTXUX1}}Q1+fAcqxd;2{^duRdN<-(>^iwZPgGv8%9EwiI-IMH<Q{&b@GYu8g&=pd{
z^?_ceHRSCu@-~;e9mUx<n~1Qgqr-`oe7#UZr~9OXs<jRWern-An6INT%q1ajbJsBs
zWn~&%hI6i+!++zNhRtK?YbI}JlDG9nXW=B7oUo66@Vrb6Y{E>@?$j2HV+IRrY5N9Q
zcu|*gkLDRr)bJdzI0KFUFjw?2b2@Ff8_WI!-uC1?rt$+;czi%2Gd;R+CgUHFhsNw3
zcN~(71kSsS$lFx<pz_-1z?yp?zi!S&?%`Z`k*RITTSf0&tn5H-H+%7eu5h2@L+*)W
zR%+X$%${D%o|7AQXfTudIde@`bM9;DP90;?2Xq;i4{uBEPzq<CihAdTU-`avEWt2l
zQZ_nKimu1Y@OU=&`PMGS$zA1m7XJyIc6~;_UEgr^!4LSY_>If!YH<53bMW5ji~a(7
za;=Qy=wxG=<ZdEY@0iNuZnfp5gSkYnV8&yLr8vyC;>^v8>|(u>QD25nXD_)+eOY$e
zT3*bvmZ=d|5)o)6q5oRRSZZ!u-J8kYb`Fv;#Yyg;X)YIswGiu5Eu|LcxugBv<RE!_
zxvPiRjcp|l&$W_<NuF{)%T3y}c9)B9-6f{bUDge5CBfWX-}aV=^gHGu<7~V{pS;Z=
zZ#$8<k2`qE#V}XtGul<QczQ@&lB+yoc4XIbcPSn2Db?BTQpEdu^~_!RFL9T4^WEi`
zmAmY;W_@krE|E={G3n+mZ#poOBiKVWu8)yK`gZzN#Ypb87;3L#r7%2B+AN8aFVso}
zGRGp#jWgAg(Nay`ezuI1x2@UZy%{d9x5vnRMVLHlKSnah+apcFq|?;Vax^_u)?1E}
zZCysmwOXTOlSP<VhK!MAuFQzq6(KjABE|h|q--Q_=QWKIzcA{p=0=J8i72@^D@soA
zR_AV{yw8mgKZMH#gt0yyEgy$5=OQaeHj=mF4)>R(mpY4I2zL`^2gtc50rGML_ZE`3
zzqp5>(H%eW9_cF!R{6+3Ce(u8Y9;f2HJ9-hon$Y0d$+&6j5ci|2M5~84Q&ItPTuxh
z#r;qrR&rXbM9;IHEPQ4m7nWJbNh=F!wc1>w_}V!7q^>xxF_YHh?Ml-+68)09Fj8xY
zM<e=f$=X8ZT%0Fw!Jf&cHzqQHybTLAk)qeeGI@}(Y<+G-54VwgoMI%Kcq`>DJ@-OG
zDcx--iQ^3AhpC~&M3A*k20XLri_cVj@$bg0i{;$=e76QR^Z>OjuR*;9+zmhJ2U14;
zz_SfMQ1fp!l3CwwCvUArRAJ2KO0*zvjhMq|MBd6~?xx*cfp6)b(0t`5Xxo>g9d~!#
z)GJ5B?bJWKFM;=p60D%-pwZT1n2n$gYEUu8uKS4a4Mlj>_#=Mv=D&^mN=;es{>|NH
z^aXY%Z>N#BCgg4X0{Z+$(Yuh7hyC<&v`^#vd}c0M@QiK6n-TpMH!Y|iIZy8_wVh7?
zv2JBwE|6N#=+G>zVXkps<_|it-i;rYjZ>lYcKlB#Mc)2zA3;o2CeQMjSWDe$!LJP5
zV7}7TTlBsh<W7P08Mw~gVMl79YQ0KBqo_1&^-IHK)}SvqL-d)i!&fgIZg56Be7P1z
z^uo?wq(M1(>wip*Mf;TW&?<2{M*){uC5{hMA&~W7tr87%oQ<CSk2<GUS{yW{4~8|Y
z_W=5P_;+i`+oDYNh{@VEyoK}6TV84qGER-({%X8Bs6weRYgN{P=C&Hdu{XS*tot7y
z-%eG-c%d3wSmy<@*V{;;Vjao4^dA-SuBnjpP=$|ERnT@&VW61`J;~eltyH+<t3p3A
zv=MvHaZ&6YledeO^2X0gS!@2)Nd@ygDuj`@1IXKBtlgtdvu;<ZaNAGK3@J6vk^gT@
zHHaUl!8&jDiOJhq)IwXaXKci}ellmhnnE?_73_5nSHpX!8Y<Sj&aN7i@$W*(TXh-F
z2jp?)Efs!!P-0)W5)bbv;ANx0e_K*;TT<|9Zwi)eO2NA~WbXh44lGgN->(YHDN$h3
zqhv%|C&Q#T37c9Z<J{0>cm*e8{oO<qha|!Nhze`?d^`@&;mFZ6*hZ#8c<$M|N{ij>
zM;p-t9omMvE=3wTa_&5lyzNciW}Hkz>`dz7SexGNoenGNz-E%S8>w*%BX7TR4iQe?
z_8OImF6{r#<6NNqL+Zu0bG{J8J_Tpb@jQ>sd85USOdT%y>+q~ni+bd3AhqRBJ#|>`
zu7h(ZXYluQSewncLz6W8ORf36&#5ToT;1L{4YNNpi;UW=-`Sa1@|&I{YUi!%(+|zp
z@nmc2qWKzryPAytnU3A`k!hK~aeP}g%5Ayh(dGjlFzaLawGa5yoH?BA_dca|>>jf*
zuYSx%@=KmWZ}2@rZS+~rV_N^kKYwGUnJITg?5CES^W*Ys+-)(3jHTXeU1|=NP=8%{
zlsar?Yi<9OgC4V(XYH7W-RwV2jmg8_L)<yNC=W)z@{ms*<HfEYc-`Qv`9J~lQkkC}
z_z|<(bMN5o5}dlujLAu5m{F@7pLUd^!_rS^!Cd7-HD7V&QWY|(o7(aZdCRPT^>6g$
z%RB?tDu$9Y*jUB|nusRTRC;rV&=NayF<5FLE;ZDnKC34|TP$U}CHMDDw32+k`rI*F
zU;3C?OaFUTGC9~v^1fS&7kQgL*ILGHuP^P6+Q{{Lb~5H>b1~T2LT(;xDLd?4W!ow@
zF`MZw(JmhHw7Z*l-gA|j`fl<f-A%^X^6TF2vj42RoWIsWWSFz`j&znyvz@8AahAEJ
z&a(BBlT2F8+`;uu^7E#XsH~mE*3(%APGyc{hvwq_+f}|<yNP2PH%@Wgq~Rp?u%EaI
zKDfyrGcuOU?MmKylDE&u+YptTG`{R6d23?CqG60wj*ph0{V_87pIAAkj+K*dV<q5T
ztb808D_+jjBv(d?vw5U^*F=bKTh^)c!j{|!llkMqq~cDf;6{k7UJxv)GlHe@$Y5DU
z-ahR(N{-l#l25Zj<=;kQxU()o_U()i@A{EqwJlNtK1a$vvncAyq9iYl{-N<vvV2CA
z_*zEE$A^*9=X!+b3F~C?cFnfYG9xlrT$F)Q%e$MnlC|=xle8jlPvi&4Zl?g57tvOB
zGv~61o~guR{&J9hX5WoIa{rH)6q2_?c)r*SH`(RyDCp8yd_wJ{=k|t@Utud<UN?}x
z)-;gY@zyfnA1nDX*NVILEM@pA3mINwE^6jU{ctpw6OZbOVzrrE+gwMiD{IU2SGAdK
z!JI~`TC!`6sWd%gD&KB%Z!LHI^zbl|cjd++Q;o%H68Gqlx0~-9$qZ(Cf9Pu@F5|fS
zhrAugb6?h0Lpc&_DE&F>Fn?ztKOGDtp1j?=Kws3|^<~gzW=hSb9(m9o9D4Z+UmpL2
zOUw_>T7KZG?++Z<UJX6I&b6pQle?8@KED#pnta7Y>W(MwWj4(E3iOQNzPwA!fpg?c
zc3Bzr^ZgRhlew*(nVHU<n#b)*uz;CzZ<iOtt79<=44J8V;UnCcWv`!Jh|v!UvF0-K
z-`xu7QRU7QdTgI9;=b++%!2nX#AkZxSBDg$Pg)-K%+AB#wdkAl$i<$1-0?$S$9QT>
z8&iAoyCHR-zp}XRgmuc#O!TBbBZ?Udjk#N3NC)bnyqM3_kUhA+Gclqt1D4bmpK8o&
zQ!z7_0_f+=&p=OJtJ*Y6hm!dfD`GRadprZD`!YL%8mNW))6jKt8lKQs+i59tDd?%)
zv{Q#|jdd7#NDJec)I~MY(pRd%>{2Br+*2ZIyAo^aC}DkCfiBTX+$C?rS;y@k!9ME(
zEnLXhpAE>?LIw2pX|aBg7Oyuap<g3%!6pUcHOV-8G#N`@Cu6lD1?34!`1c_5_Nt&P
zS7FI{)}Hk>FlKE#<Dwew>?3bEsl>TVB@UCVbJU!9?pNV?OBLK|t59b)zjj-N5v;pM
zHdSF^EfwnVV_p*#^Y2+7zE)!LeI@$xoY4Q05?u^cI9QLp<9_5T`P^lr3X|5T(2~4u
zLEdJZBA?0Ib#>L~N;Xe;&)Sq{jlP@I(D1Phm`fcqeM1xWQE#lzGX`~1di;78Yuj5H
zD)<ZQ;}Fh+uc|PMXN4pDcOO_=U+hAUQ3Dl9UMVpsSc!(@?L{jE+OJH(O{Cy;0&CT7
z<Su!8kG%c3SOEj?WCYbDF^el178c1cR3~9k-DC_Ik%*bA6X8ML=H?}0ch@8saW*~3
zOO5vY|8cv_9zJ>NIh#FoJ<i<ud9!z^0CnO+oi*sp=eLHx(@OTbAJ|h9R*wy^Q>pNv
z7WxmruR~rc{(6#*S$mkj^)UlWhtr>3Cll3rspv@FW>9x#6`_MI&r~-)v7hl;izD~w
zm*&T&74$xlx8D!(+{Qk~#Zg-BY|_GHs|K4d(tF$}4Hq~+_@UsxAC-!!hRpxqyD?yo
z4p%Fw*=B9+JuCxuyO|w&Aq}fnur~gSX9f14KT>OTU<@-nsk7cbm)YJsvgrk8#wq>C
zUwE6;i8~*;OS%3Y`r}yRPv^eJm*j2rb^6d3X5&@`{c%0B@rrYogE3jyJcXL=znQth
zxpbY~Sy;{<QcP(!%$P&|nD4=d)SP+q{TR-(@B!A+TdAi=k05u;*rWEyLpeVm@h%T%
zxd&oE8f$B2P47F!eMwS`tWU)#C2z|=GWR)(S<cLtYs3tyKx<~YI(|XZQI%+Pu?q1n
zzhSbf1|`k(<kVh$nIF%+WmS4oPTt;*U{?GW)+a{Xp;oJ|6mp;1e}63G<CA(alDyqY
z-g*zVl3R1ErD~tGxc;(|pRcXtc_cM8>^=7&Z@)~omWFmVvL>LRgj(3iE%qsoa<9<y
z{VnA|b64tR-DK(C?y{t*ha4qu&2GBNcuQBY`_@cawz3z?-cFMGmlJ!KPSWp}lgu!5
zmUjBia?r?GmYX=s-Et@CAMYeH$=gY%SwBDJEV7BST!o9&8tTHGr!Jyf>mqyAE^;i{
zMLMLqh)tG@oX>NS-Ni2Qy~0JFe|3=?KV9U4X>-}WuDL{>Z!Q6<Xeq59B|So-WlML~
zxu;@fR!Xcqe;q4dIKLIoSjjpYEuO{H9h0{%&mv^2e}rUd#>kKbWNq4L8M`7>E?x=|
zO-QgrRs_k`2SGBC8mFJJ!7^ZVuq00pm7<ZMGLQA?39{G5AVT7PN61d<riQ<bl)he3
zQaF;2VRV$}4`ntL8JtJnYW|Cm<IlpSwSSmwC2wWnAi1B{SE_G!m+Xt3rI;C&^S^bJ
z*q<Hb?>8N!ggYIs9Bw0f?zfi90j<TZua8WN^_Jw{UShz0LLc&WDd*|a3L43(x{XAC
zl$}`ZZYXoZ8cOkt1`@csft*`rEjGuP<2cVsmMbkeD`&5x$Xphaw+E<|il!GRc9of2
z+Ehoja_63RgWA%JT03_i*0?L_0cysbKT7U~CvPviGS7k<slHRFk+LzCobyJqcc+o)
zCUd9lA49S1ZzT5Q?O)`r4fV*0<m=!{18H;1KoTnT<>xkiahRkp>pJVnmGw24F}()k
zm;b?)tG|%*=qJ|v`vY^Fejs2;HNt+=19Yzn%AHm4qfY9~=<hhL`+|aG>X7$*#xTdv
zusiYzF63=B_gRjg!JS^*d-;o6e+TBu1Sc0G?0GS6<`(0@v|{|%rWkX`TR-JT)X?*H
zF^!%DX2y<TCQe8y_qhF`UgIbCxO`{6yD}erJ`|vCTWUFixwowG2c%ES!&35gmQ5~v
z`_kXhnOVa9nWe|;LHT*sve!9-V-~@>uiS~l8CN*6py|hKWAZkoAQQKJ(lN#_9h(cN
zgPNTIN4E?leWV8UW;*jA*)R4^hlO=In%tl_bYmJ;vZi(5zo|Pi9cJY1KU34V|0WHs
z$y*2Vw(A=m9K3b7vO$Yq>=D=Gvrk*f9&MQtbFM2{^K+(HTZxW)6j0nyprKZU59S&?
znM5yaCoK*Yt1;7nJz?^8*v4dxOHIP;DM@(OAPEnzCnE1$5?&uphWV3Z?r%*(_uI*M
zQa1%dM{tHpR?S<^*(HD0SI$;{&S%fqi5U@xlrUJV#BH4toy}FS@>5})lL}{yRmdHy
zLiHrF?THdI8>`UgrxJh2+Gsl!x{|dHQ<a$1M#=f661&LT-v;a*v!^?;iwc+Ps9?5Q
z1qbqW!D1EKOjKcaxC+kXZtffvX0!I)e}iX%9I~DBT$8hEq|-BWmHq8CE7aKjN)1Qe
zdUjIdI{mvBcqZt*Oo>C}ZE!T}>uxGIbymYIjJ<IokL#=O*EuCRr7BQ#CIzt@Q}7@@
z1$Ww|;Ln?6==@Vqqo0fme2%PNCgDS!WW4;GggzNb*q}&4PEryQViS=yAQ4vaiKy<G
zh-c(&be|-QUdkJ_%NhLtX>do$Ik5`!C#$jcmj=CZwHWgz6%(D(FqXalYHJPVlee+8
zG?+u)#&fpq`5*mTQ+4qBNzEH|((T`+<2{)dnnnMRO(q`H&P2^-_L$iVIm!Cg{)ZM1
zuV|s<EWYMBnSM@-SL|y<&(@&FH$I2#8Ps~O#OQ+x?B`5+Os)#?kEtJ{&+FX>4F+@e
zt@uZS<z#KQ&YU~!W<UBnXY8A)X+OeQM>pnaK1#z{^0p|6`6|)uStrsL_Ctr_#i?ky
zITMD*xZA;xv(?{un8=*4m4Td}<mO=EIPP3_&Vli#Y*^eTcNb^lGq_J`Z5HM_Wy5b|
z7HXnd(^9Kt?~+ArJ9ADKX5l{j@BPZMQ6kLoZpVF1oZY-;Us->CE;`lD!vgX)Weoe&
z%%#qE&qEI9xfySnRsDmW0yX!EF!Ow1OYU#HSAh62%)8o4-%_U{jQ3&A<O*hDbz&w4
z=ThIhe!;M)N;qz>!j4lvp_}>(Vb^}+=MH^2F^M}wzw61=u14bb&PW<eF_lA8%;dlE
zb!GeW+EN)<M}`%eN#2UO63Uu)S2uInd%#?l*RznToav3Wt|!hX>&d69mhz7E(E3K!
zvYy(iTJvnANfvjddfJI_r%?MnEoJRLEv1!@tNizneq1{b`S+{6oJ+Eo*4k$Bt)!W3
zZeTBSI@n9fUygEqh@*V%?I`;PILh=Tjxy?`qfEW*C}ZwBN}Rrvq<A`s%OEG|e%XoV
z1831Ra*_DG&N9i_MPi1#i1|GiSxnxp(z!@xri)A=Z?_bZ$K>r@^7asUyT^ok12rx(
zPF!S#E>e;#qGUvuXz{R&ksli(WIy}80pXFdW>}P%wT~9PfM|K=9Vxb*sE2wTF2;Al
zr7?M1O&!ytzebC3Y>2FB9U_fGgJph6ki0$+Bs04NOCR#qdwsAZAXHWi4V9olp)zn<
zDC^B|sTUX_r?y1Mxx`3$?i?kP`*CjjW{d=k50lNU!(`r)5SjXOq`Y?-CUYkYmitqC
zN&DzPx&5n$IREY<-D5k;%?=&pVZRP?VsCq~ZOLq{R&6BeU2Dl-=`StSKGNu<t1Q{!
zD#PgqN)B+5U4yv8sb^#M)f<cZ6Fcce-i{}4SCF?y4zpL`W5wN#mXcdlPeN|hlgUHt
ziPcF9xszuu4)Nxq$*wDw<gGb*yMeqt*Qkz|k++5$x%<M9du^99<D!G9G+xV`Mn4lt
zw=|Jjsm7ARnZ@xs#&Yn4k<{64B-_bbvm=HQxSqTAe3`p)!9aZM4P*p)`+>dWSyQ<W
zNRL{nea!#9peMcT^~7#L4c58U;O&e*aJl~rJ#PNOsGZEWu={~7m#a~Sy!9k+6}zg?
zaO-!p?D8G^u71HY>XDDuqyE_LGyXdH2^AO1v13UY{<A5A+$hD{ZY9i@FXk??V*KI$
zPFwC4S<c+@?k$V4L%$d)%w4b5yNK^aX2kvTe>>VHBo&||dHaaG^`al&<2f_my}7F`
ztN_{cH>@FVZ<4p>)_DlI$b8%J*~p=naa>dGjTxAQKD)B8VGi^58j-y_GLc8EXtFIc
zf=ub}2<I%TQ#y2Hr-KVK{j1Y3k3F~F7OY2@%iHC0I_9ir)**RY#oUA;)6&RLX8E#@
z_~~;R>JLjt-|zI8khd3FamLq#8sa^?X0iuq6QaevC0g!U<+b>O8UsJ^TFLtMSsx|P
zUWsw~N`&lGAcOjzl+!85Ql_A!GzBB5-PxU)0?R2W=$D-Y*DFaV&?KQ`QWCP2iQM^?
zh+{nxsS!=YL&GGTnwEsaCzDb4Q!;*&<!$UaH;q@~B$Tj?QzC{n;bR9i_RUgaK5y5_
z+Z+oOwzE$>jJ%EbrNq)$Ue`TT%!N?m(i$a}{ZK-ep@fBsY<{W4!Rbn<{FU(GWB&IU
z`?sGJXziiI^leJ4`%SK|<ouUf;dcX6Slv^FO`}xsB5zgXZ4h}|bc>p09eG|%KC`FY
zt&SQcZK;i#qK3r-HImM89^6NTWE&MurSUP(RN_5(tBzowo#%($f2pvF_4Z4R5@qal
zt9L7zJ*mL)e^an{RSKHTO+m7E3hdAGY|=Lw6Tc;42%oR%Pm}PXC<y~u)80r)Vzyus
z)QL%$6O)LIhZ68@Xadgz3AlSH0f~}~b|uO5^`v0x9O|3-Jm|WsaGv@xySkhQ+|?qx
zQ!4&Dl8Rdm*?Zv3xqXcq)5>|4Vb6Y5fA)Oa>)<<-^H*}T4>PeY&>QVY-J8`gYP@Ia
z&~2j*4fuL%$&AX*8(5pt4`x8W^vHDb^O+W3>FcWcTZ3WbXle@mP7~Nic&dc4hXUC_
z3d~MY0QQ{KNAUZvQODMkn)G8lr}2z7-9UqFKh^kGPXq5h%nCV3-4^IsB5(JQw_b0V
z(MjF=qER}mIH$v3^O^HS-X_t5-jaRZ=>PH%`Y(4Mlea&#%>Njb!}A!m4fF!v*q)7G
z&Ruf9GoyQ179KDUENu`qhn<<f=Ec1n<gM$cOgv_;M?HF$s$OTJr7{z>*w>C&Lp?UN
zRbJ%nJAL+`>d-&Lv+-Rm_c{Kg-fBl4MzX%n&E=f<PYyPc!ClDPpz{SV^x&>S=3*^v
z&wQ-;Mc8q+2+d}c@;zFHsnk$anv~<^O6J?0{)zqMt?}{SC^Pwk=ZYHi{-!5Z-Hc=i
zeUbX}Or&K1^`)(9%Zq2V<qCP*jlBK7ysp$VG?y`5%q3!lxdf88<qKK!khd3))srWe
zEM@E&`gN_X<t8;$L+>__c9{*OL3Cp|QKyNF3u`JTUNn_o=bB0P7WR@bz*!o-Y9Vdc
z+sl;~%_OwAnXIv~7l*0#GN#&|eiTRf(8*CAv*)>Co1?hwcNE8qj`H6<&KwII#hSd;
zXC3_Yl9RkNaFz^f7ireONxoG(irqX%=@;ZEhP52UH`77#c>6cQL43wIh--+0j16{>
zAw3+#yM=?;**nNOPX|%njg<X&BBX%%i#r}fiRD_(Oid!hsa}N4v5An^UJ+uoCY+Ba
zLY{d<NW1DW^15h@teZVXtak><ne9Q6`Zh>rO$w6onVhYj3lfVtK@#3JNUpvQ67y~$
zQhYZ=4zS;BlNlnOVeBn`4VOSfNbqH9p(aJj?nY6fwhm=aVVKxh43&xbLs)wbk%!8G
zQu)4@47}J=`t0i|7oq}1e)o{RS>2@IXa|YE+giR*bM3InS0<ZziJ_{M{8{BGt3SAl
z^(uGy)5cxy^=m0(`1~C|VK29O*-NcX{C|07$DMz6d=Biy^j$+4`_4+@4_b-J%SslI
zw_5Vnle~>2Z|m)`5bG!A(s-V^q-WKYs;g$=In_+omDQ1_>9r+mcWv3cp_W{zG?j@l
z+_7h3Dz@aU!q-HSYnw>aNn`OYF_IY0YI{tgMk?1(Ebbc0e|roiw!fid-Z79bw+&=u
zBi6NF^+X%S&&>^_-YR`*SIRvE&$$bTy_hfaYfzB;2O*RGU>APj#?_yAz3K;sFRsSN
zGR|xJvX^|f3T6AMuzkySwD<XrZ!hVmZC!z?x)s>e;4{kp`Gi`mZyo2Cq1d_%C;ltt
zKFkt$S{L)W{}JuledLaoV%#6ho>2p4rQaw*tF`picPv5}v*CUoEyM%*@}4Jf=UH(+
zws9Ye`Bv^^;ogC+mLG7e53do-!QDAK7pr5LgU>#4`X-*I2WMeUB|SS9^hor{!nW<)
z15e&MG3#fLAvK=sxX)%}CQ33n!?Mpn7O#I%<I|AD8Qf6v_E$q@|G81~GAEsTfzq*<
zJ-GIFnA>|O4IM_MVZIMJ+mYOTNiC@}b)_GumFY__NDe(aHq2mT5BVT@J2#j;WcHHV
z25NAGGt*wI|DN0??*iFVc2y!tPl?5A71)uOf|31Ga40ziO~~6;#VIgJ;pdZ6;Fyzy
zp(~SM)h-FDSoV{jCc;5K5sh~wVDB9=!hrqd=}FjgE*WkeQ;<vE4lPh(+X5w`Ybm)C
zL4l1`^v8Ttz&u@n{u33<fKnjOL;+`G1tNzi$Y~`cScPTm4^KCshvtL=K0FspCU2|A
zwZKOT3_i^Jh*V%*e+4rQ6fhxgU$tOwd94!F)c=N(x3|dK?c{AupbGujZ{BcRfdN4&
z7#5I>#dCN@;P0_IT!Ftu39G&QH*M(6o$x>HvkHs#R2Y4Obu%?iv1Y8}cuuHrWuJJM
z0!FtK%sf`$JnQSs5G6D9l{m}i@8lQOw`Ak;v}E)%OM$ax3Ra#;hIjj9+)PZwpk41#
zhrEr<O+s(7_liBA!?Hvy%4c00o=6W)0y_PbfQ#GykGCPUQebYCg3r9q2O$b{Vvk@P
z>su4flbe^SQSZ7I9_?8-c4O8Eb=`Ydo6ltbJC*(JY|gTi`MV6}jC(wF)osscu$cYl
z&E)N83)YfdbnK^6pH9Y_JlDYWwgypGG&pX-*)Y9T=Q%Iv)lZAVE9o~L%{hF&5+ijA
zoPD1H*Yp&$Fjv4TRKfj{3WT!1;Lm>gs-tQI^ET}g_1~P+_hVnaqDqbRKj>fHtigA(
zyBjrohUD#n{yYz!NyW7u^mGR^4`j0rEuV5PMN{sAU=BzCU)$MB(@}269#AluI)%F)
znDHUxZD;bf)6Q)Cp?_}wOYT0Vm-tfv_fawvtS8T~!uMHlQ6|nOWWx4(CU8F!>*#yR
zSFmsNgIOP|xHEAR^I~>#KQw2&mT$8D|9*+*yx-pB{MGLC^6txn-*M_3Z|9(DbRJG|
zkJs(K`8dg0()^W$xW(Mob=<qqJ&?PNic2w``vc<-l)*8g3WF9^p*MLO-R~EQw*AJf
zN59dh;5X)xxfLA^q;H6cY`$qC>)M&htVgwE*)8sMVK%P!pE~l!&P?_VGn3u!=CYpI
zQx_~O<jMkSR=3rYF5DZlWs;>FCT|nTTaS1fnLW@}tadb%$9j#$SJPOwwQnjjZa0-<
zXPe2_+V*6yy(FD$Dm#BS5x<+w<om~FVp!8mCb`+mxJUMKp1iH{C2!3gCHtC#SUh!*
z21(5QUFj%B*^aW?*-64#-yS>QB)!;oUe2C!td*0rUF;}Fx;RR!`i@dz>L`7Zxvz-4
zJx<<wlDF?hlD*6`T-w}0oXOhaW)8BFynRwRMpoa9keK;VlK*#v%$*e>w%^9cD*Aj>
zf5Ie%A6pM%Zn1BK93pSKlC7g+g5+*UkoZpvl8f7dWX#JT?idb|eTgHbS5yAo&>*q&
z4U%W%t#{WD!SN8uy&fV*UW7;&*13%z1<Ui<!J=&vEOCv4q;~!YnOHMaLLT=Mji#5Z
zU(!ol$M%wImjfkfP@wE<-a{7GgV_GvS*HB<m;Q&nrOd@!o^JG#LDj9K)(sCiyw^j{
zwDXXy)Pemu(n8Yqw2;6<4iatYAhnC_xs%9V`jEH7$=k{I$yx4POvq{=o?RM9ZNmnV
zz*)hu$ClDIfb;t5db0XTJy8YLllliO<k@3$>aWe|rK&5p8`kAs0yFBa>qtZPLEdmL
zP&ejEIg+;rhnmW`-zK75V<Owg+g;>sU-H&7$w+pPw}t*j(v-Zte$h~xZ!whWZib?&
zp~sdrZQ3+_F)!mD`6YUCY>U3sCT|1i1=^sa|8}3AG@r-Us^Sm4@CS7t|ANn=UwE5O
z|IgfN#F4juwX4FH+f~@VrV0;!R-*I!3V4KlMq}ne^(SxVwEhGOhjOehVqF_uhBKeJ
z_j+Ur><vrcHK`cACw|1sRv*z8%rJK=Mjm~&u@{SAOD)uZfFeA}r1ze@-5SO`*4G6H
z<bKq5ZS&#SlJCLY+}p-E)9IJo*~W~hhty##jmyP~db!L*Vjd%T`;}P&2JGLZ-(a>L
zJ(I$7{(xL&85v~4V{RsfkKh^pGjoQHa<9PJG>i{p?&26~Qt9akC2u>Dx4&5XPNpw%
z13!<w#Eg7ilcIQSy3T%_|9~`9y-7nt({$X<OT*tU__c}5Tr}cG)-yHi_pKyvKYMA>
zIG$Nc)bSdVx6`dyzx7~^HbQ|-oG+eR#9Hwka~JGW(6cBJ@dFbv*D(==UlMRdlYr&v
z3G^9}7p;@9_(39W-%dcph6GeBNI=wo3AnK{5u1w>;e9s=LFC#mlN5ygNkPyIC8qvh
zuew5kxs?hGA#YPt6bOt~z?C}P)7dF#dN+kRL(~y%QeY5!+TgYP`7$#3hyr__DByoU
zffM{%=T!=HAFsf@feKvdr@)QN3RqksllGHuBb0bk#G0MG>5O(N)Mkykh`c>$sD#r)
z1v-$oY2@F)h(ufoNyPg*Nm$3aJ7FDpN><;Proz`3N|<+3VsJzDtJyyv{XgFNvG?7S
zJ|FV-;Y|fj@*Ggtir+`xI(Ab)Rn5mAoPx4JDcG5mjOII8({4`2Aokc#HRBw%(R;)X
zc@J%50uJv?K*PHU=tAChC{4sf_Kr6OC!#}a0^0XTz|FV>teTsEp&!{xc1=MK*36ob
zoU5~^ZAaePuUEmHzheQ<9}&Jv^x(&>obhfWZ~e*Jt>o=~6KcXfa@Nh-v*C2kroVG8
zU8utGcI*{Uuaz@dfns+Bp7DNjPx86lL46zh(Jz{6VK-KbJ=DXwk+*lrUOV#k%||8k
zRut6Yr65I@f};f~c+3AYLq7#<*<X0Hf_;dgoHLWRp`6cqo}ho~w;G1zZI>@<9NVbD
zib&3X)jI5ZkcQpNIeSQd_y4p2F9+)Iyq^Yp_;@ES<&1qKnai0-B=rd2sc*g#ltt}e
z4m<|sz`!5}2Ty0CGd)HR_@1hW%*0Lhh*qAVAK5Ak?a1dR74)8$Q_KG@6N^0lA4_K)
z7v<hXdle~ZkdDFb#&&I6R7^lovAesw>)73T>~1}_A{`UV&>aSd2!ezHa@TwB{p0zF
z=nOf}_qW$xd+l_@=ci#P2PbFQr(xfTRQxkL14Zob-r%`+7;~pbQjh+U#vRJ^VJ&CA
z$VkrWHh!Fq%ha_uU1Y{4d0XJ|8I1;W=P`FSj86NC4NtjaXdr9bs9cPm#GMBY+!uKE
zCz_rv#{A4u7?Zb#n=0YFs0v=!Dq)&c1)Z)Mh2-t@In?>y8py_uhGO)Ok-RH07PpCY
zB<nQ&r{$dY@FsJ)w{PPzGjV4x@5VNB@gi@NRIG8C_ZR-nLM||qD$&$d&IQ{_E<HG|
zZylv!va`6ly2#&`F0z+<`+oFb7S#{dyw{v%B6+)bxQiSiZ_nDeNxYMr6i#)MtvlRg
z?HM;Q+^vw0FBEd)qe7B23aPQhT`r`#OP40~<mFKIco)@^Pbu|euSI<kvwEVs<SssN
zmq=50Szgm!g5nkO>Y768tW!wi844Lf-rgc_=a9FC<ZY;<LVS6(qb}Uy^kmr;F<C}=
zPnP{JC&_l3iL%geoQxwU+c}JpcHz{9$N9;|1-{ZG+eeNx@D=Oa(IN}jt3AxU@2~u1
zyqBLihmDq{*4)`M*iWpQ`pNxJKMCtOR^DD1D@IRP-#!>Csb|MZ#`7_9f9@E$U^zzK
ze;O^v!$(QG%bsEw(^pE{^%B?I9@6M-4>{}FLn2+eN%pI*5>?bu)_Qi3?y9EZ7uG~<
z|7|R*hBX$Cg8H&zR(-Lb$UiUElh_V!(lgpcW<29=;zLeSUhF6nF4#*jHRC2N?d55K
zy_|eO9hsR@Z<Mz3$H-QanQij4r<F7`x8hC*>J`5%<jooG-djRl`L>xn|Hn+alDAn&
zrsC4nRL)E^;r_lla_b?r<a8skAI4m%QbW-j&fPzi2GWFD$%i%uvhSh3Sd7z`97}!K
zK~Hi;JMOj3(vtx4cK>cY={rDAQh(Kwo0DpZZO@vr`~>&ea@L1Syks}`>&@f5&#-aT
zI4#v!+37EH`K$0V^fxA-_>IG?V-IAMp{2SE)(y&0KDZnwSC+xFTNxHc6k#s&EKGYB
zV)~z-n0EXp{Y-_-*ykRv&H2ddozL7F?$+c!5})bZP2D#S-pq)4k)Mn9<n4?md6-&|
zi_iC&)4Gkikw-2{s3+%L%)!(?tZ|Qj#iw>(*l%S1I``Ht=3JpYcc2*!&4!Z|_qWiO
z@s=|`eTy<-&rFN{<n3CXudiD29L_n}WKAmWvWC6KKbO<@G?6nBpP301Ku^@$9jOT1
zkOE`!wmmQXSt&R^Ed}qUry#hTJe|SE*rl?*r^khv4F9q3wvs)(Ugwh$_LTfJ=Z>8$
z?#atw-Sd`ve#qOQtZ7u_t+5aNruA8`Y@*MH_w3Tzytmzq#jeJ&2qbTf$c;VrWXF3Q
zy1MA_Po@UlhiMQ*-X0)t!-CXU!s}l1XwJp3HVoH5_fm~+i_|b$s7BmTH8Y4bP;ZLH
z1M>DwFD(W->ad4%F^#|I@R*+5T@f+xB|96Fkyq(08rqF@(k>lhzG(S(Ewf5ukkl#$
zFJ{N!@46Ud(9e6H{5{CWC{B~T<nK+gV*HF4j3sZ6cHwh4k%JFdPj1q|VsI>0W>aha
z9*cW*=x3&u{Gdl1#*yz$$-(J<TGSlN-ZQoDvpqFf`CN^2z0~kLp@jcV*3}bu7U)WD
zpU^?Sgnd(fj3(r5%{H;jw~E0&^492C3@-3}U&*e~^hhmx7lU5p?VNNi4%+Gfl@@i6
zX}MQfi(hl-Bcm4Z&|Zb$Au22)Z+noph13F^U7}Izs|F|d{zeB?=-We$8BinAUybqP
z?TMKh9KWOCE(RSg@#DHthgiq+g9Ur2Z*%A&CufuRd7X}r<z6EG937~c({I**=cwuQ
znoar2vx7OaESK<n#a{CO^44%rEVh%Ie|hfMLf$IL+hX=&9W!EJ8_T}1V<Mux6Y;z+
zebA>kdrd!7J%9G7IqP)Oh5E!E9R`rMy3h1~W$5sbyfqEtwK*2I|5tmP#bY4rZ`Xg~
z(J3RIb7JxML@#*ve-co!APMC$tY?pLF6>MKdh_|iM<n81`y`B~UOn|QedRp=eVRbs
zh5g=V%=@alhWnX(GthJ__a6<-!p~~v!`#V)2YGArnK^4_>8LTCp2L&rF!{w?Q}$(^
z<S@heWjZE4Oh-HN)}Efj#i!Ho#*Vuz*o!&M9&c)=G&JJ*be;`gPrcG{VMGSJneS##
z{dhF@ES%%4o;fqE<LJ>c;ko(mPtFc|eSvd>udpI-Z#-lERF7N?=1gExaz1kQF=xu-
zCte*ZhFzUfyc<%6n2D?fihkp7_8-i*sVNIua^8`1xHY+J^yC`@8P|gI9CwW5sdF7k
z;r^Ffm+DC2BNK^TW-6{ZrZRr9nK+i4$@=By^6s#Oe2L=D7<#JSZnqStb5`=*$yz#}
zV4XCzu2|K!lZ_5`;%;m&hH#OmH(X@T9VcqHPSU8Sv-ps=YlgW<%w!kw-{LAi+Ple;
zNp6z0%}v&kx4mm9B+OJH0iFu>gB4=6++7OU&&_IBPmX)k6PsD}WOYD2iKLbs@Yh{-
z-*lJ79_|uSq7XmwwpR-E<jbt>$XjRfwgq`xO5TPvP{>r)whe3*Qe~r%#m^>6)030s
z!Me#Z`QUh&y>GmjvVXe0w!bvHGDbewkCAL|a`c<8ES=^nJtKXjTYFz=aB#F-Q25Ez
zeSR|hv7d}`_mlXFQPN=0Xu0)#wAi|hmUnsHGU~Xun2z?ANv*vlGGU}NXgo|_^%)}b
zioB$}?_jaq+D~R^`-=0<-g0$WXYp9wK_)YcamS;UQj>GVE$6q8S<WpacvUmm|CFBU
z5l!Xxp@uTQZ++>1+f|Zkx{B{47nz`Nk&XwQ#O=1D80>SD+HQ`b``1DAwmL`=d0RQ&
zULL=(lk*4cBrncZR`j!#*Q{&z4X7*g`7`^y&|0PsVGsV1rCjREYoCQ2{QtbYPu@0b
zZzfG&m`Xz@Q@Jw9M7(|Ki05PO2fEIzls$&hb(5hSEHIEOp#~Dk-SN)k?cf*sqT$S~
z)|}ksylvl1>c=U1vY5QJ-=Zhp{q^LRbxk>0T|*l2GGb2SQ}$==$lG>F)mXB!8n?z)
zBfeWT29dXOpH$&k)NedI{2SG+ej_xl3~$5B&}m*7rp_$GhUsPa-hgi#@^`*v5o~%C
zVktRm&ny_LNk4G6@ei0S&gTxFd<@~tPvC?+W<BMhDSiHnSjT34$%SVE^<%R4INu(^
z-b{<dxp>WtxdxT2XSrwmD1A{IyD|%hJINnirEij1aAnb-U`yU=dUKDs5qm=9Z3=a-
zfNW-2yw8A=_kbIG{kM(h&VW?*VDmYr!Tm8sZ`h;FO~cU{=?GYo21Cvv?&o}Ci%lum
zJ1zywsTDt+!FvLEdvbCLJkBS<pFWn782Ua9lCZ5GXJMCeM&oP}8Zmci&i}Z&HW9W<
z6R~z&B6V!;ui>oD<*y0&{DQU5PR@~1N8ZUkWIy_mK1RplxL+)?Q^+!M>rD+EPS9%{
zUth}%BIYjmM&k$d;(5I9>x$Gk6sU&vZ8bWqRHMy$HCAm=V^(`LuCgv$F^;U|``;c^
zW9d5$P8VyiqcLmTxzRY(N{1zC&cnQk!E*BUF4<Z4A({G$Ufzz>fe-8Ofwkb+!Spx3
z(!q<mZF6#W5dTfd`?*u!82Ytia3(>A{~FTYyO5r(x#TYY{Q<S;`sD3@_jDMto_$W9
z8OAbeBA{V3mea@8(N}{f-RN2V#h&xWXe9V*nKeyKobOLeR^#I&HTv~Z!JoXXy*CPb
zS4N?szZUt&bh!OX2RqjMf5`Zyj<Gn$GsD>;9rm*RKioiv#ZKh%=vbWP`DD=$Eu@(i
zLDo9-c&|k!zn&4~?Q8NjsdY4x>ZtH+kP1<|Rd}7Cf>UiZO1r4hiCXflfhwdfP~p^V
z6}EL&!)r2`%wDc(8x0n&)1VeOGjv0X5b|yVc{_mK?TfqUaf@S(-;(vbCKj{E{%z!~
z6M4IlUa{8ob(opK{0zHTq?PM%jo-f|)ES!opjY~j7B+pgIAu*8VF|y_8F9Gf#9R^T
z&NqyygOGs{-I#|mnRzGa@z4#3NBl+B#r*hJ*6_I!saJf`VQZQWZCaAox2cJ2j71^!
zgzHst_}h%LQ>?Qe@v%+GTgM3g&aGvR=ZXYGtfu#yy!AMefZb&9esX=t2+nYqaUOaD
zHHmkzc<jsBua(I#UdJ=$eCjbzsGHYM!o6ke0ZR%j=W+)!J<S=1)6k@LI*zqU$A@j{
zIDQ}<4`!0Jv(izqgqa@nMPxNfN9XJ`RI(2|bq;$i^g<jq<vvMj*RByMsQH02R?K2E
zByY18WMDnLiVY(;8_W#Vi%yxaKEo{S%UNhg-mdlJK260}JV<5U)XN-n;W>P&Iv017
z@?po!*Q9noVRxt)=aWh>=5-13=a*o5Kq=7gH{R{}jZ%f4SaS{|KUPm1$lLEt4CQmV
zft;sz>MyhTHpd!CVwH)+khl6prgC;B=P&=7Nel9}{Xu5fv&WaR-$FXnvy>7uE4c?N
znYP19UVLS}bih_>thbYjUiPBTy7o^`7uo9PB&9E%#Id=v)UbCJbA_`^^>GpH26E$=
ztL%Ep+>Cdwaw*eQepb3l+6XuK+({wdlNGX!`to>c&1cEmG32f1gnHsUwVn(iZ+8T`
zOI`MQ^*g%D{9<x6Ng-u#6teE3f_r4B8Be7qP2SFGqYyK9h1ggrM32`^YlWOTJW<Yj
zOq9o+I6u2(y!dY(FQ=?0NItVAv@^%Ztmb26+0jwb#Kczyp74>Jfj%;~ldo8jw>7B+
zU)${`UXT3b;G)qoFlU6AXZp(AKBMKX<!D*{-CM?;^OiBAyyY2r+x62(S?fJaj17lK
zVValNlec-B`^ijIUup7-wd{%>a`$UzX<)!T4Rcz_(oQYq-Ru@JoV?8o;H(t=W-Es`
zmGpOwW$!J8r2pe86$Y+yz~4nyxcxug?k8{U<~fR66Gv%J-Ug7j4fE{f;8=SZMc#TG
zw3Em0ZDnz9Tk$QgD?7;BP4{f%kvIM2M%MCQl$9(Dvyk5hETpUr_2i4@;@-?$l<hcw
zd)QRoEi{!;Hg)7)(>n5r^{p9sdzZXDxxr9s=Nm|7Ftby*%QoBDK&qeU%QWU9Tb0t2
zOkZ-V5IreX>d7pg@%$(2Nr!%Vl4DU*Hl)^&3;V0FZ*?^~FZqkU->Pu*aux2(uY_u5
z1#T~@z?aBMgx{z_u5lIY{#Ia?q8z_N%TW2C4A<wDp&ogAbaEN?&m&`7a}IrU0U9;o
z?mR^y#<VR!mx(`Mr1-(P(0pucpO0?LjtYDD4byLb!=FCiFo-$mcRzE-Pkb)A#N}e2
zE*IkxbMUT34(@Q)C!V|Ys`q`t_Zsvs($CS8^R#h`KEc&28*NImaFY96?3q(>Cp`;G
z<}g3%|MT|qbDpKCZDlM-#~kKSJU*HVw;w5Z7MF~6Ps!FI_G{?}`QI+ukIXn;x`?$e
z>)SxT6zm(Dg6t{m$xceaLa!8fT};AkW)v&-B%;YhW*Gi&E|Xeu0_P-cLlg0_MiP7{
zCE`1^Ko9cv3wax$jYm=dy-EMEuSecaWd=n9dWJXA^D|FN9g@6_r>}ODZ44S#(qG(G
zi$ZrTOp2m$b_C~CqBQ6{QNx^RvXvU>x;|>W=e4P=8t1>Mp!=jkHCg+~M~zx5)R=QX
z4RwG9E7+HuL*CvyAC3DXwaB0j`GMT3(<p}dSuwb5L`|5z;=Zi0EXmuZd$ia={kdyB
zEi|q8|D9MLw$w5wk~t7P**7IGGTT$XbY*5!8)})XKU<K+%`BOJ_?jB=0UcEHbx1WO
zzxlD24b>uVEo(<_&eqV^w3mM1tiWir{mc4$p$5Z8@^hsA{lHF*kp3!IA5rrA8U^#^
zQFt~n3Ke~#5Sbo@*Pcp@-JwL!x;%fdu1ou=!)Ed}fxMkf-ah2KtZ_drQrl^<rhyg{
zrfIQgHNO^qJu&a25yrEM<GLu6hA2^GszR4-Ds+fZq32%}XUHS5sHPH)Td5E|Lxq0(
zR5%l^LV1n~cgWke<ZVBGu7y3dNWY@RtQT6$59h~<*Wrw5EUKx;|F)){9Z3#PW_|ip
zhoSp)upgpBW+FY%tmj*v;oJZ5`(eRz2zmRSXNFs4T9i|Ja3ybVOpQUkKe6!LOYa<c
z8(EP66aHQdqX+p-U;3VsnAz2jJy_P!!PWF6^LyXwi4K$aJy@HpL&xSZd|mKb8;dcn
zoSkM3{g;^~Ih?&7m`lGoJ@94ksAaE-#|iqu>t2h4CH03L39<N8L0?yI&Nq>_&*I~u
zxWOL(l~^np7RQ|j>^B=R7jy=HpYJAOFY`rbZ(?q!31`ZWQQskN7m&Bzd4_$&8rEqD
zcXO~WbB(<9CvR&krB90Qo6Of&>Yg-o`IL&@?8Q7;nTo&UZDk<m<mPja)#GH;uuO%m
zA7|?BGQV<BI`XfPz08>z&AGjSYcg=|d=}O*6C=OJ7pwt2k=*C&@t%28?KzjfA{P_6
z-{@3N_Mv$1aZ(ha&F&Jcd0K*3<4d?-w-kpe%i*%)H{K+dA!%tT=1wZbsG;S!(6<5(
zA=R+iP(w^t^M1HkU%F2;66H`6F}5?6&oj(qGc)}@a_6||AlA0qxqp1Kg?wvhDJp7_
z_xf7Nt6f$y;fs~TG`17-dv-F0z1$2t2Z_Azz+7A>+06as{*9d_+0I$g$lQ};ueptj
z*bZ}*_V-+6WPq!fCa|yj*H!L#yGd8_b}D&$o4g%J-h#X>Z(mQmsVkdTxJ#>W_I1hI
zfHUs$*MN0xkwW_N|NkOy53+yxeT_nzlDGQg?K1MV4te{AJ<Uwk&i3XCS<C+Hs+<Xe
zB@?7%)_D1}biC})n;>noIMXx5PXbIvOYan4Q5^M^u3LO0uZxdVH6ufhdCRR8J`%It
zPuAV>lOy&0<fd|j98C9>pu|zqOY1A1ZH7qSPF^zBV4$SE_mss?JY{^=0EuId_GPpF
zvcFj$DR${4jlD&xojQwtHxFvp9VF1FgS6SvT8gH(l9aJ6<%B!4Dp=b-d*4*DJ2sUd
z^7cm`cNx9FRm`Xx{2t7+%rR$a73(Auo;%5dVGgV*?ZxApolLo9$8&<6Y+cT}cC4K=
zCU4Wp+b^({Sy?vHY?F;V^R<y|e`|3dZ|l%IwV1qJvdcmm&$p26_06SCxS3quXC}Qi
zo66KSCbH17j(ANnmS;=2lVCZ|W}Lg-G1O3Y=W)+1d26-7Kptn%k9<>K^tfklLrQHa
zd8#MFujomeCwkn0q$lqtQdc2w^T^w+aW&*DdAny-HKz6Zi)-DhFk*HkrjDvaEAn=X
zR|PsuuEe$VRajVF34^B0xQOH)+i7L^=TiwbJ^Y2poyBPTtO%AdKQNwsS{I)Jo=*#5
zom2n^&i_0g{{zML-0#Dkmj7jZ$LX)R2&R|#;<y~n<#LW^T`p$)VD@?cPw?E4jRkG9
zF_QUE)y+6>%ei~YS@bam(JyKK2|xE{qYJ%IjqhjS1b3uuBx|?v8dXXSS;sxyp4{J(
zz};sl>3C9<j(z0q)CuY6vVeJub-9n6UZ!R*li|^tekQ9_#QQVroj$1OGWM1hGrNbu
zVaDX`GtMV2V}6jTMG8j8CL=XD3C)sN+wMw4;Q2(DG)aQ;X(Fn!n3G5kRJ?-sMfShm
zo6(0$FKbWRSnlqN!EWlX$`jOJ`S#=ctYgbrAKEjkse3Hu{p3BH9PC<2e-wG!qM4T3
zYBVB4qOr(}z9=PqJk(xK=c-{$mcHnwhGj=JcL=Gmn7o~zp~8rs)RMi_h+eG5W%_*z
z-PLIKS%bgzS=(NS#zNLosL8qAHM}pgKD|iZR@REaCDul%^<!YeTCye?wU4ZS&bqpz
zjTVvitaY1faftQkhylF6Z{Qh#-srn;w6NwqzjUFF85!g+`?h__+AX_vFqut1SGg8(
zXQS~eLW7|dYPgT!=f9lig<13^b=IJvn+AslYH(?g22<{<;gLcQQ&(!e?UiUWJqpb|
zqOf^L6#RQe;TCyYI!H->m=YTtSi6$XyH2a|>mN0`UR5K<NCUeutjqgrnWI4;vWpfw
zr)cqbjuu(9qhWGGgSX`Ex0zA+y&wupW<<jFLIhqG1jE8E1W!JM!$1>(i1$i7Qz<bk
zQ;A+dD!eIB;jgV4cL%F+Z3=tn*EQ(yO9NwPEl!4Nkr>TBu^W5etThc=>u^)Yn%$p0
zZ1OhmAL{npd(b0T3yXJJ7?HPW<m0gbo>SNh{-;WdtzWemLhtizGI*$83}P!{kx9*?
z9`)Z?)|``+u?V}ynX7lK<M}*|7VEI{cMJ;t#-PNRKdVPNIE3jCO72(W>u`$Cfi<zn
zzrgzUBfW8<^ljHCKgr(Hd2u-Ql)CZ?*1G9&7<-l3AX%)r<6_bJ4>f0ACG+C&t~3@Y
zSJ^YCZvD4e99*xk);3H;9iAUoe@F!ExdV9v^Ej?^PLAi)k3xn{;eLd1X&6aOxuYHT
zcQC8+#YyH@?j~2M8+Y*HUgRHX2rWp(lqIPM=j&|mGVUZ!N=6fUqZiCd!jN&vuxOux
zd!sqiej^RXv$)fNSM}F4D3A_y5@(b-Pt=~g^)vZ`LEIyB#f-ZZdVj^7cHDVABo~`U
z=VJSC?hu^(4cTGbCpae`VUP2X>QI9FGfLRoD#ZhzQe1ypikeGG(PwxmwtJSttA8c-
zZvTs)ZZ)KqriNTyT}#Yw8#4>TSQ^zbl^6ZY<gh=zs@Kir=qL;7pS&;gp8U?<Lf-RQ
zxQZE&u~t&|VO^;?%}ySK+lk{Od+FooAnT4f$U)X#b;;ZM`A(9O<RqOeoaGXETVm-V
zH#^f;wZTOO?{(oWI~Q4Z*;Ot~aHF=XkjnWAxz4`q{m$-UIMZE@E_IiP&F&Ib!kRYN
zUH0<tCEeX6P~Tk^6e#3Gj6yuZ*}pxbkaMdQQZQK|m%SBI-0uI?x0}h?jyK(85@&x6
z*Gv%e&(x1sOpv>_<GBxDyxciGP6jR+EAJfrWKa1h=}6wTKI|)B*ZRncHuMlzjFhtG
z-g0W0kG!(<meoH;%6hwzl6Yf;%y=_Ye4IzhM-?AK-rgGMCCSq~rGGO|LB#+$89P7%
zv;(BnqrV(6=_|WWb(ieyE)uHiB<ijnQj5GTTHitXo3@apE1FA0_vX^Rm>FHn>ALk#
zW0`cgo-FdNCo7U&CG)JS<Xv)+Ik%l<=S*j@CU57Gw~F_UGSbCC4j0?WH1hTbc^kEq
zwJmv@S!gSM-1XGydR_TeU?ckLZRD|!jaW0w@!)GK>9E8~G6!)l(0U6Ab+?dw2lnVg
z&1BMSQ%PILtc!6bGCi8ConkCA$=j&ehGM*odw}SbDzi3_PM_GTjnS89T7CI`Ltn<t
z)0ZaXt@@Fk2s0bkP)knVuO}b4Pu{YOnNeXiq)RvM&`A7?gQxytH~mmGS6AX#Rt2PW
z1$tipgC_kdpqyBVXDzGHJ+cDzJ$No0R))2!OR=(|1iK#mLNC`MjNHV1@Z1F-y!ty{
zEdI{>Vm^At6`*0-4^;H}fgtj><DT!B`Zf<I7SQ)}h8f4BbI^ypZJeG1w=thlwDuF0
zY+?OM-U_p7p0sEFIp^^zxbMq;`6tZSoQ=XZ+0f%Y@K?;M$qmoKWi{s$s2krXO-JjJ
zbR?3o>Ud`ABr->q=jv8k_K4=vuQ8SxdgQIQCpF>`$(UNidB_&r-`*w}ZJ4{b)+QMV
zXOiJb57e3)^f|Fdd&(mP-<UO#-8K>XhbLe(GbchHCcwT8cL1ED=W`8px$roIPK|?k
zR4iiYqcxaDPw;To3B@tg)TlY0qBrJP47Oj0!6Wi^Z&?i1+p^XmZ+HEO!J@V7ZBEyr
zMjRQ(3<k^D(P%b@b3cv9w<HasEHs$(P>li0>6@hndUGE6)LMlsBNbHBRL~bSk_WN(
zNZt+{s>U9&BAYd@@|6ZIXE<Y86^-4jcVGKxFoE3mzoteX>XC|E>cgzpHXLTnwv->6
zIS>|Z?157Its4-HM@=}F%Iu{YJ#{#=SxX%+npyPp5S`SZ#D?DJMbR*#M{4g{9omz(
z%g5+YSs9H#n>E;2s76`28a|w9-7!pqXC3*mn`tn_iC!-9)}6c!Sf<8>t14!SsbEtU
zh1*l2@Z3HMsvc3;DN$IT7K!%1BQb8T3Kb9e*aI4j;u+)J9yJ=1xAhD(=tY*-`>x^n
zFd8j>L}S80*5^DYjA#}P!~a6?*fA7QPa|-lT@(t<BGGMpC<c+Y##SM4&Im;-$1ohc
zr$j=O5?vFNDEdH86?xm$Mh)NTygW2`?#o&{gXaWC=25X1e8@i<J)7|RHAjcJr&-5e
zVpaxg`}4*+tY_`|DOXGHffn^%YN36qMe$4>R*<(9jp#ckZ&#4F9rd(`_(P69<j<LB
zg9GfvkA1DdzPlQDZr9)le|F{cYdx>cd87>*{0ZlIiyFrZBWlR3jT5Jn{V_T?UE&#M
z5c}&ZW6|YQEG7iT!Z{`m4Xxubi2c}a$~a_j)_K#ZIHYeOhnq6z>p5qx@3Hr76A#n0
zIH=aeB9&gST9?_2J{*h7oU;z%%(HE396EL3UdL0M|5YVoCeNCa*h4wb`Jp)S_7C&b
z>JDYTno&ARZPO86gFRXDb~bt1*)0W!=-J;Bl!AVI&291`Yk6&Kl#1aWld+nwuLGP>
z%H~}1qDP6i&vWZ2=Olbla(1c}U!UFiT-TW^9m?4{dLhpDO+i91XZ1MO*WUFjXRyEG
zSP^&ccFjdS&I0b|zM%uHa?pd`sm?BWICU}~Mfiy+V@lxKrxer4+lNz2(Zr_&(ybIp
z-%Fr+QI6-u%za;9P2H`g#Es@$<UwP(pJ*a$IfuKafthq7Z%<#LcP7(JJiC~S?ufZu
z%rKYQHWsq^t%W$5Sjw%YwvsA#((IL;gbcHnDq{!PTIC>^?I_E3J4)FNM;Uv=NycS5
z$#@_7lsC{rRpBgw%Uz@yc{}5<i_DtmDpO23A5@pj>_?4w0%v-r)33Z!Ax;n3i~Xb!
z%j@p)mbGo{683e;;s>?e#p8!UhS8h6H%uWXpOU>R6{00?&y%+|6$)9y+SV+NeykVl
z75{P-$DiY6$*A$N;=8}h)A-Ah7UN}oP3AC0(Brgbl(anNE8aVO#HWq7H0kLr>mt0x
z`proB7ByU!ZyGL-_YD_&i;?oK?nt?kF;p7H4`Gh<P<d1ADJci~i-$!&xqpj(V?$5z
zEgT>Zf(FRS%LAn4t$y-bzo&Ey=q6?Fx`<WJ&T{W{2f2N(gDfR)g}j|l-g@+CE-lI1
z#FdQ&c8z5i`_CU9xk*?TH~BEzRqow#k$=xQ%ikH!6038Pe&lV{B}eg@=_sQ!?PX6K
zd+5P-l1lC_)aMz9b#1F=b>;m$3-)9!<h7NBc$rwp>VwQi_A!_D!_6dPH+%JGOyt5f
z6X_aiB7RLw#eb}+WX?C0#h%8}xP!4&|7Ro>iw)((cthzm&roW0FpyFU_9DLMi!WJQ
zrq&lbdXq1ZxAUB8i`heFnX#s|Wlh_NKB@RldNS`>Ejc=oz1r>7uzg<z&8|wgoT$Xl
z6_x07x&n8bSKwjKKiEF@5AykEFzecsJr!uwp#qmei!k=iPbj>8V&c9$v{2;H)1QxV
z`T6Yoe#g8;-{C;sPJUW|>TJ%&nf}0x7x}0$?mPCr$U~>1Z!qER>&w)LMsp|L@Hp<G
zCTpWAv#>HQ6IUHGxp$74Pc5^N$xQQV=AcLQWnGJGG;f`aNOxwRGvjyyvudty;CwB0
z;}vABC9mt`?WcLHRoAC7n<Ett^hMp8#Y`X1G@NBU`?Vw)Z9S4Pab*%JsrlZzmWUj_
z-kUt+93pue&-(Y^f@Jv71NFFyn%bd6>R_B_ByTO~A+@7EH)RU-x~8llcuxcGWo68B
zP=|4*Wd?H>{5d<*KNdf=F^E3N`5y9i*j~=DER8|_`xsc$b8Gu9hTq2+JfyCxveMx}
z11&1HMZ=6fDBtnX+?_^tOxIv`D`pE;QbT^AM(BJshO|?oe2EG<EmUZ#r$U023eFQ%
z^roqCrH&d7RjhaGs<B}-|3B~9i6^5`pZD$81?2W{-nVo3{)^Obi=xnUV-%XrR$<v1
z6*|ZA|4~0)n#ca|^=RlmL}N}N{m*lwxpS6U^a>3c^dY|+Yq0Yuxj^3Lk~_a2t6_Fh
zjWe65zmBKRNMB7YR0XegDtx%BMtm;~3VP80+lkpt<n28N4Vsc+%|7#SOIWA&V~rl8
zgv<9R9B3ScpLS8q*N8$;hbRn5jl`?cNGxm>g%tXi7WGu2Q<@UjqLsL`RE^cOG}uDk
z-uk9Nd(KAXg+?>$O9cb!(WZwZuyR%iI!A^8%R=BD5rUH95VY$Tif4JjC@>E}fAaRR
zc^E?Og~5BM62Y<5oLS>m-HpPe-mJHcRXC}yVIH^!|0ZZ)W)O|x^g0jb*WLVRG_&%f
zaqV+78qd@s_9nICYAte0wRl6`M%>Us@j{ET$2_OB<197x^>+LoucyY}X+<<_YG|?e
zYcwY9<;8rBOY8xxe5t{M`}D|>udB?c#q+$<;6M7w&Z{vzNsU|I)wsl;&B^1Oy{5)}
zF<ggT8+6bZ$KW=3`~4)fZSvNIy!F@P{4#6VmoK@Skp1R!{2dJ^Z{z>hU*`N!Ahlx`
zGTH3`bEmt-qGA*Cvrf}%y@9+Q7mE$7v3rrX>rCm7<_^R^`<cO!LC<t|?t~hf1Yw3`
zi*KA~B6mx&QgNH-+Ee+QmpjJ#wofv8Fi-Mb*Az75?9nqPdIv8jq3-?^)S(B<i@Ytr
zlZ-*0|1WQCa8d%My<mPPbF-jIf(2{efWyhCdn6fqs9*O?=g#Gt?B^^^<(@$HfbVdg
zZ_8J7FU;ZIq#XQb{1xWozT!#C9At85Nm~1Fs2#<;$L{oL2K_><sl_;@WWJq0_ol@D
z!pz2{&}Eh2=A&{%MEu5tpg*j&Yf1@udu@V&gw;0`+tsG>fq8xt$FP4(-ag4Plh)+z
zo<rtRlxi+NbIiqKv!&QRU~L2&nQCMw<uAB*hP+L8cVJ$SgUlvx7n8SLZqd*4)JfKU
za}vLa&a!!(v#5_d%fU0wvd7j%in&vFqlK$<ig1<l<ZX&5L_JX<b*GWH>lAYF27OQD
z?Hw0)(Xj5lYvnFt=`OR`yM3QW&6vDxN!~_3R!II5g*=%=Z!&qSY^7jUqJkM#ZW8jE
zyVp}(C1&|}snlaO@h5*-@z7spF}q^>Vn5j#K3Wb>9wi^<`%2;yZ%Lzuycyo|Fny%7
zdNERJTo@`BRt%T3ABIU-$01_VZV2-vyd<f{K$*C2s1!W*lxLUw$sn_S@?zrvng6D@
zSWW6Jd+Ya>ht!a>l-<Q>KnF2j&`!)7v=!TlZRAe7HgYharA*!1Lh^?-6Nl!^-@4UU
z7QJdHIma4G$$*A(*P@}kqj&o-_4YMJuF`#oi)@(hEOVw&PgXj~jwg<i&+l{GU<VnR
zU@sy6*-IDlcEdh9X>hKtyxmz>_HVP2%QG!R%Z%iH<ZT!7wvwLn^$*R&c7&Nkk+%c4
znaHrmCbBM;-YD{RiL<dR7Gv=kU@Xm>8%vis&T20*l;3`aV&F?n*^#~}9|IYZqAyRB
z`m&VQ8uo8D+3L$n?%g}ee2d!?^rTN8J!Wp`NlF7fX}P18w3|qOa(qp>xQ{(s&iL#)
zT#3TL?Ax9qQ*A0R)ASEU?k&fbZDml?o7}TrDS~W&;=tbm%otw)-J*QV-_IPVif>5d
z%<QxQc`zVvn~|%xzH_eD|2xWO<U?)r1HNx^kuoP2KWkF|*^q<nO~0bS0PdF#_=NJF
z**N6RymNYrTQbA=&DjjRKE~V{RXUC|&A`s$O!#vrsll==Jmp^6%I8_g*_j0k`X<M0
z%|uQvGo$u%e&HjvsWmBZS22HLHFx2Wx6>k1F}gSX8Y#(m#|+#%LzAIwnGEe~X8&@A
z<2$o>7aH(2|2`g9hOh^#=FYjf$(YOBV+-!%nfN&Yp0k)apovE+AM@W<W(|&H_EB^k
zE~%Ldu{{pU#!-`dMo-W9Slk~Q3xD!<l_my-)9DB1eQ_1*z%A4OH~*s6=n#WhyhnQf
zq%Qe@S|fRTydWBH$=g@I=|g7U*4~gc?lcX)^S*nkKn*|gwgY+Au0(~hEh-!*Z!0U6
zXjrVo{SjnlS2a2stC5+j!dy!wUY(7CG9wCK>M1ezixR2&DlDKM_$K?kQIA#Fl@f&x
z>!Q$(Y??>TT)Y0i*;Exg$lKP`ePe!-nGM+EoEi<gvl{eVrooFw8horzZXRTPN^W|c
zV$DT%KRm(nzy{Ww!_~Os%(KAaC>UBR@#&5lf64v2)hc9aR46J|abKSrW16cmbiE3#
z8mSQTLy3KMN`zdFVm4(Ia$iN_eqtm($46pTKqL&WMqp1tBwA)gft48QrYN!aof6~d
z$BtpmY`l<luTFzD)1%RsGqrL2T087y-}fMS>m7o1Q$tX@T?pns55e|tAvo(4!k%|9
zPLj8J<gK?^8029XK9aZTA(5=5BC-8Axli6^jZ>j9&js<+1-6s7e*BpXyiHA<_2<hy
z><d5R-0q-goTjE7ps&TK)|@q>-q3HC5)rI<Yu$;)8qOF$rq?`<yfrY?;C+=EN7O1@
zjAAB9rW&fVYUI_4fa8mBwDt~1x4&UnJUtAVMp3w5MvwG8ey`53FVFMJ7}nVTkhLqP
z@k~Y?A~;lsvSm6PG-S_$zHXlroRubTJ9ptOCVB+EUWr52v^czJ8OLmhSkB?aVvi^1
zqi^%N!ZTqF&M{LDkEFkO&1z<flGi)<=OB6thEl7EVC~zxGj}0!&!R7LG-@;7y75)^
zb$_zwOWx+^(%0NN4F=@x3G(*Ry;RgwGS{*R_Y!ZR_B<pRP6v`PtxhsBmypwg=__<(
zA2=Zq-VGA5r4c>*{mJu0KJOxCa`BAtB7|9JJL91HCmCks?MZTWYc_W;Z{;2-KE^Ja
zH8AfDZ(C)e?=X6*m=n37`)35mSBz!vX+(`&gf#ueK2ttcwfq6gJwMR!_YbT&!yRP4
z#mLMq#x2)UY)CJ`iPNPBzFLk4pDS@eQH4nj{=)3SUmSb=7ykKwv1L>>{cWsIe(Q5j
zA7?FB7)$Q}6WQ?5RPLski&I-mx&O*iTI$;{civ9SUfRj4PWE!Hp@Z}#ZwJ6pu8eb(
z7OUu?ZQ~@*5}jledAnn!vuxYW>z1=*y10l-Z}ub|T*ZPlobePlxjxTLs!zMg5qtJy
z>3?z}Z?}DA@0GmWepn%!s#wdKyNe}x8=9q%<7$PhV*j=^S=(R{d%0xw7UoygF>#Z0
za_TW>koJ9bmG8{A8a{5k1XugZ=PdH}fxpa28zax@H@-1>v`m~dN_v#}$RrCNsS`3%
zD!z}D-Q?|*n<FJ^(NLLJI)wX-2Fn4@LGm(lpj3qql$PXeojIPellpPh!@iPAFH}|Q
zKJw&3Pw~3kQ-(PAl<Y^H#PCulF__vx3O(A%kS1*<)vt|guxKrwuUg7#bqg`{YbLk3
zOStT2W3hk6eCRC=#ByJKS)Ac6lX|e;uX2^89qE@!a+aWZ&eHxBXMZL-$>DpB(qJy@
zTk`f&ti9wuw-*h0yO_M~@z_=_@3N7%-^^v!M{`M7VlIcW&14ff`>-YVFjkw$qYM)X
zjyI8czv_rgsv|3yA@w=PSX55Nva*%2jOt=67sqkW?GQux&TNb9UWO9)z(7`<Gmr;P
z26Ftmz6AZJFD7C7GMBvFLf&qxr7w-CAqS7q6X&_LWK4W5x$R7Eau|1JoU2C0pGv47
zRASiLO1N#WM7t)HNUN@Z-h>Jas`UpmZ<HfxZ5j0Y{K95^*0oOyQCL*~G$_EhS_N3`
zn}>azpM7SWhg;z6Y^yvNrhUVu@H`x1ew4-heB5U3Tk<*=Z<cfBCp-tk$=f;Ht=XXs
zvlw49Lv>a*_K>%(k(u1%l8HU1I1k^2J7<;Yc=IM5@3N^4n`GclAI{yc&4M%c<@MoC
zlJ?CqaV0YYrynvqj(bBh-com3nSwr*+y}dk{u9n&Easd<^L@+?oREk|laf)tF};x{
z+~3MvfV+vzO#02d1m4XblDEg`F~0hSdGW)Ov2t?~hObM+ZJzNr%w#5knsXJ*9_%!q
z`TcD;6T&{NkuP%+eR!{8R)rmBRBBTrE*i=El^Z=UoDcPRtHYor<SqS0lgL|R@;1+&
zb2hANFBIw!dzbzfW;4{VS7U+;wN>7K+YYCWNR4y>nO4SG&;l|rI)QbgGkN+$g)!vL
z%cn{_WG(jcJO3Qc`}YSW9<kT?qmvR&by%D2i$YFN6g=-KQKP*I^FGsC<@!Gzne{Ju
zo5Fv~{v83IqDTbqS7J(S6?!~S0me$G`S@>cN)%JCPBn?b_S0kxd26nq*SH>aVCum^
z?43?LMi%TOr(Y{^kesbUoqNM)CHEIcqTbsmIQ@vi(-BJMYN;^n6g|O#Dl9Hj!Lb>&
zaq_kfdE0!A5^iLv$-F3ZXKkzM7RfB22&~JFz+j2Qr%MqC*G3|i{F{)g#M@-@lpk+i
zKP9%1>-wx=t-c1ML3$9vHw0m5Y!Id@gV6qG5N^~C!GQ8WyxA6r@y|lg>q`it9YS!l
zLok}!QcI2r#VNfoJn0e!bCXDH4vs{Y4SD-30;A~D8aPFT0xi!!!D?)Mrbbd-4Q?FL
zz=6Ddzf*(MGwhMqBZHgKNA)EdPunw_io7-7sN|mGNIZ;-#51!fEGUk|?k$lxWEPHV
z$HLg}2*#FRGLt-g=A^>dMiJQlHyqQ+Tes3M{J0Q`tLGvx!!;6)6)MiGvS&};PE6s&
zp8lwb(OBJ$n)VFp*Nd1v#@@VrR19@w`jmO*Thp1lrTm!7$vso&*ypIpns)1Y_H`%6
zVj+2(<Qj`hoIy(ZszaC!`^%|16isJl#*tV=HHk$PXP*P_#-f=0-x195sHB!NdSg7k
z`6nWg-u$2+iP+74uHsk{I@I9~LDsZ2?3uI0JdF#N6Y!j~Mc(A?Z1TA3C-+4fC*xp5
zBF-{fV+}LJXTGHW+BFgH-4i(%#9WVr1eCZWAg_o!jz2PQn|1ien)KMQA3bj@_b<@n
z|7;`u5q#Xm>q(dw!TS1K3Y5%~oW_ido2j3W@RNJ$HD56>DF;us<{+cacldE0V!*W@
zyubaR&$$SFeTrd{Ud+#-6m?IOpmEqQ<b?c!X~i!%+$n>N8}kLo+mWxxUHJ>uFzzMg
zeY0k%zVsw-&n-8WZM97$A=p%gtuPatu9k9{ydC6dBc(@eW%V;VSw`NzY2+X&nGQ0&
zzoQ%?Z)a?9l-)lZWzq*HNgl@j?O!K(^2R~t&vufpYG+Am&3<kJ7n$qs!u{_qq9S|i
zWV^{N^0wb9&hVU2$dPUe8Qw)9Tgjf?tbbSaqpymZ^W0R0B&rm$h`cqVjy#tC=hI;J
zZprR_wcW(z6B!lE+VHijc#^ljea3Tdj=zje_Lmlm{bg0^7+FsL)3$(7QgFvt8kI6n
z(aA?Vg4mP2J3<s0!?<H;q?C9JmEcS-$(T4;Vh0V9$c%v!`Ej72+*3+tc}nBvp7P(N
zzH-2zuhi_=M_P{VDO`ys2eZ3La9$S)Ht8glQ##0mvbM6{v8_ZY+Q?+`cKE24Qg*hf
zSmiX4nQt4*qa}^R^;APyvbBK>XNI)$pqu>K;3n-)x=MzZtE6by>z?f_QS^N;U%>ut
zUq`VBcMw%?2N|fb7t8hb@|JVA3&-2ZnORoy%)>%%{xFwc|Cx*T1lF|Vtpj;mlFH2N
z;ihsmkJ%>7YusK$&XTwGe~l%Fd1K*r#`1|7+Y{o9q<E~6M0gp>5$?C$+rmI*v2SY>
zt}jOp>PwG3`ttQLGcH2)#fQ9o$6dEwFY3veZF<s^wQcasTCyazmXw?6iB0Vq@?v~7
z6ztoEU8#f<{m4zG(udr%5_`zorg`P~S67DgUuE3MQih@5ilKMC2<OS$bL8zm3GCHw
z;ZA{8-%$D?7kathP}DsSlWOPT-m7n5stHCp<}qU@4_o`^!yx|)ZpMGX-;3OJz4QwV
zPI1rXxX<YH>Ju(-r<Y2yP<)rO^qd7qJCT8x9vP?|LZ72sI=q6p6O`ADXnKr2)6t7F
z_A9<;qENwp4)^@@xu1bb@^<)Z&KS^tp&)NZElWVBx75U3$w$_ppDN-oJcjxI!?^c-
zRT3uO=e#3*pj*h>l=__U_{G_a{YhALlFVh!-6{Gpcd}1An*FTt)PpVl<@_qo{|y@Q
z{QZPA!Y*pZ)ZT*Wi%PDdzqo|6p$%fu%z-s-eby7?ZP6Rv&*n16mp+=O7j;;;Uxz8|
zxjklW(esB6>N}i!N#$&ahBcEdeX`U^Q-?(3NiO?_w)E<dxAV!{0P;4?fwd85S=x4{
z220-d8>7TFjS}aEs<8hJ>%;*{WOq<9hhK?PlcLb+MFbS(kr+*84r1MFc31_w%y8T{
zkHCQTkyub3fpJce=)+q0iUa$ietd57R;AD9sHKF)JPQ4dSqFzip=}=}w*I7_XfBzW
zNDk4r<-yPOEP1=_g%YF4jTFxHSS2Z;R~U(c$B`J{G7@GZBJg=eIIeCDgXOO<w5Smd
zhs-cM*dK;V5n(9Y8i9sqBXR6rBv$l_#EULu(}M_{?G}MX10x{gA~^dRjxEE(v6L)-
z!yaomzt)-LZDL*oLKjA0m31(}HV5I*+91f)AdC(U!oh+dbZ;1pHl=~+zA}*6gdvpu
zLvVRd2;Md2`waqN`XUrxYKCzyQz$z03PEsCB(ncR;KPkjd_NMx-fJ-YGK1(vrw)BM
z6z6}3V(8m&_Ch0(zg<bMkP2TF(Fn2DqWQgO#Qf4={Ad+?lu>xIJ(8K(k@)c|5);VV
zv$Z2}6JeNH7K(3$!C0OZj7Mp~c;ym;;SEDDs&NQTHwxjra3B^<VEs+r9x-6O&3?i2
zVVt?{t-&4gcDxho+y69J7D{b?05d7s&uzrtC0FwH9sSakWK?%G^EfB-cl2-^a_GbA
z!dli!%RRuS*=ygY!;_k^sIiMP)AVD#*5~gd{nNExaA#8cSST%dcI-;u{21nPGUp2g
z%*F1)`nE>`YA#4XANFxi<g<@kHwl9|AEX<Zj6a^NL#YY%ByX*kaDP)Zv$e?1Wml3A
zR+xl9*23z}<nRgZrMMYS-(x&#jpW~*64BEm5v|@P-~qkTN<L;=U>pkRIlT3kugSBq
z=+rL`JC?<xxFh#KOs8Msb`qA7w<G$eVDn+_TDnLbYz%iB6n?>sdCYR{kb`Dxa!^TM
z$geuN*fs4N^x72Qdi$T4>sSnr<YJ6=EQRG!>dBjap~cW&=&$;PNyE#KNG;D~;vd93
zuE3}^fAPt)maKBGEgrXO%jOaKaw1q?_M00>_kKnaHqBVFnow(`Ul~s<WJ3*0IWdLV
z%5j#myu?yoT3N|wX7p-PtYpeO&govZma0N)d8Vu@Yw4G*yT(DrpLdeKTK02WxyUGE
z7db<I>{{+5+1s3?cdnDvj&hOv3l(yV94+lm&6q45K;F)8Po{PxWBtimD}`*RrI43p
zZW8#@O|0m*DqQ6%*F0S1I_tsg8LpDM)>YcRaFtc@uJl6rOYNWjvX-p9!kzB=7spE9
zU1KH3b(B<G_7$5tzLM+cBd0D5liHrcWb?(La>9G4ten8wcC?qQ9yeI}bs8j>&JL6}
zwLB%^ho>xSJwU=-`pH|LzS60vk92L(OK#WiCB+Rz?7wuAxUjAgceJCpPwpV6ep5>>
zZzH|P+m>rui8E(_mRxKq&Oe$+m$b&x`*<U1v!S8XztljwGuy&-jGG*$&c60E{ZgE3
z{!z<CqNou>&vTNRRb*{1M=1<*kb8X`Bs0QZmaVau$PV^0Z;qWfv!|Zf(LxMrTSx|Z
zJI~)-mXf!z_vk%NF%|o9rqYDGT}j?{C2#dk)seVf#^UI1Ec<I4%b^$}IknkH^ye5!
zKqJ<+l?KvY!F&tywi#z_`;oUM<ZY95J?ZjXPu`zn&z89sJJ;*U>k{TRik^hzaR2h(
zYB>B`jVn8<;eVhS^QKo~(4b0a$k`_Ll~}f+0t*|K<5z4M`dF91>`4ixZ7jhpWijU~
z3UR(oAzVKfz;`oeep++?<gahI`zRM{a=yWtecZ#$TYO00{y^5c?d<Z<pIOMgn&e}Y
z`a8P3`hwl$Y%9kv@Z*llrC=`9lTX+&F#G@R${TzxlY6Bz@rZK(E!t(^#lPGm_B#z;
z=hE=(NE(Vxa#znl?l2>7wXZYL;wSsDDb$hao1DBn6<uGj_iCAfH-8gQH<z<G^k;^W
zm&ac-f0uRX7EJ;i9GIm?eQWEjc=VtT)ZrKR-H^90Gss@@w4F0~NiXrvfz;cY#h`@`
zXHJ>@@NRPq7DmS4dLjKLkLior!)yn;7@Ve-_hB0MuCC-9stx%|_O5Tv*%0n7-$>7I
z{4{3y*5hm~eMj^6G6#vgJyy-xkVEv&wAUh=J;32A4QzOCJ;a{o&`K55tVi{k8?~|%
zwZtqHHeI70yp;N8yb3K(DzSXM5(b0W#|&5E^*AL85+dR7ClXQYecnxpL^<oL%x4j}
zGAjZmCSe$|Bb4{VP{fW4$Iu!P^m0YuYhnbtlef>0MdG|+6#d#NoKusnpCh3Pi^L^9
zXRC-voVgN-RaUImsO@(2iGcO}2proHffJ9SaI?D-_3kLqg1o)V$Cr}-lkZ2O2J5)0
z<)O?3BHv1aakFg*GAzlwTR|w|HM=Af*QlGmYZ-xQha<@=K0o-mbc*D@pa?Yi9nQSu
zaJ;aOz(3^eCi3=d|8VYs4a3gVFwC$05&LU>gbWEl;GO{Vi3x;X(;(EW8Hl_;9}zbs
z5dZcH!r;6>wCPT^4*!UcO+ujF9m2e-V2r9A2%mzFcoh5*?Z~<FWZv_4K}hil!o|9w
znCBLPB?|+wk#%#6%^~=GC=><9Sre1DMZuA1JB~9;tyJiLiaA;Ax%RoBVXuPT=AV)9
z>lK0I6=4{b7{a{-A?T1Cf+@Q~kmMeQ2IZm14G6}u<Y2@`2XhxjFg}F_;`)<7D6R)$
z4nOakM}tuOI|KzsL$Qh4!>f5pY`0P2OAi%-EXj3xs_HOX%(^fA>pQ7ed$5n*F&YKb
zwo}MkH+^PcZlPb=ONU?AnL$#QSz@m^_j^)@pxrv;tkxmflXFV+#9cc=uiPVMY%Zk#
z*^N0X&FH~mPG%Eo%9)%IxBbEWQsiw0vmt%U6Y#EkB0TBOKT$gg2lyJ=L*HR6`?aOd
zlhE*BBD%~-Ks)AI{ckt%WfgO-!Z?Q%9FLLBsYmaN$2w+j)jQ8~F7u^zugK5s2{<~9
z8Z|Sp(ob_|R1o=kh_&rd?!!1kzjgaKY@8L35RU{zbxlAOXQr<DCtzYf^7at3XufCQ
z6XzJd27QKU=dXBU$bFncbMT0~omh)Ah8ejSem$37uv{$KU4-^!MKH-P#*W%07`pBk
zDqsIZ?A&}byITUqlv2c>E#v;73beDTLK~YZTqvz6yO}v}JyTx>t~HPgfrc{D$w&s!
zubSA-SVH<5N$_byG2hC)r`rugn58>$kg@1mTgXoCG5?-sC3$nLsUuoTWAb*=jk;pZ
z{i^}>tmI!CYjOO~TCDomh-E(;i5zGn(VXLbbJR&5=QuG-!A0&e``CqS-N5=bv@`kI
zTp?cM?E}8gjJypZZ#R**eaYLl1#U9syPNdC?kfGs+mS!X@ULWcwToCk<*tNSS9$v0
zUoNilm&ud;<t%x-D%?*F&?D9RrmyV$*H@BNKJqfbTLL-_lUU!OlGuJIGZ}}97kT^K
z*Gn4t4wi~GgCzR&K<SV(Kx%*Wl#8tfa1T;H+0chs$Ax|5_6YWDOL~epiEPR5E{i*K
zVurDYluYa(eHXOnn0#xAUdVjLv8|+g`<8OQV{^H;uBp8L+*sn5H<H8a8p`*J4MdSx
zU#>CxvgiTp@<11fZssDBtDNN)>)YPPoTMboQO5OjlnJjLWH)*1!d(P91AFOaY$J9r
ztYw6cHS;E|rTtiQxzgWU;u@Pv>q0X*b;nE!7n@3wiK%?NVj}ZlB2jsDB(Oyt8BTxt
z{u9RXwT7{@iZ>E-^0wYABN=OMBoB&N*FMmfFb{osYr#D~oa26z##wG&Hvf^a<n0FX
zwzH9*RK(O0-=nqUes&FM*0F{ZJ*>vjm@3p#Rifs!N_6j137tbFS|?ZhU*4XdRe~F>
zOK`cm7@rM_QAK~$$lX8Dg1mjurI7uN0;o3@U{>b>JTCr*giE>fr*i&$P%bv9zT!H)
zP@8(>p!0`sc-b(Ic@Ny@ah&_}INOuReR=&2xo2P&^DA6Gv(NPjbLfRa_bg2O$UQ#X
zIon}5ca7A~z=2EYxKqHLc&F3Q>;Q9WcBf%lcq(e0OU1i|srbyk(V#fa(Vk63>lLYJ
zPu|*@r{FWS<J6pZ?23)SIup)7oTqoQIu=7NahExJvN{ju){?hsdWGu;az0}xbBO4B
zu^?~1QB#{lkJCo-)}tQtg}s@-w>Sn-Pw3C3|JD|<Sg}3^y<5jH<TC~n1~8LvJUvh&
z$YkN$ZDU|EOowjkwb;{8iy2xi){(b`ZFLy8REKj@boBUYvHBk^CXu&})FKBa((hxU
zLh?n{LeyPP^FG`8kP=Nhs*%N+pOIJD`&`I=XS52(m<{poN+t3-D^Yk~i3}UEH$RdZ
zMI^S8R}J=16Xo0P4a1<_6pBtqLUDV0C>9?G#T;s=cjynw4-Z3LbOat3M!=i8<!<tJ
z%0ngc-IeTdu|9oIep3%M35~?+ft*t{j>K&8_6}LHH8~ukTSVaal_>NfZ+|>jqQa5A
zRrZwIQtLfWjd<CRFkBzWkJUN^tI6AwtmAH3hoI?2>b?Dfu)rb^QO^Rf==w*b)(OO>
z;$T#>=1y~nM2A}uuqqEnnQu5kZNu@;=rHtl4Ml0&F!T)yMQCC$8j`oYQv>kID*(sW
z1fb>IKs0F+h+3;YqDgcB?0W^^?CU^eB?iKy3onz8Skw3;8n*d}PuD)8tzIC!etkr@
zz>jeA{0N(20eF9&pPN?@N|yz~b20hXE)ct~1wcJE5Y}X2!OURv*b|7+o`I-YpL%^0
z_PePIL~T=|C1;sFH;BRjo?#X_Mj&8P7`zffa63H&HOX7gr$P8OCkUTz1|jo#FoHh>
z<05(6lDw@VZ!eR#F8mn51F3(Lw`L~va-F8G+)K$^nJCof8LNuD?gK{An08r%nd}{x
zFt@W$XKEqsnCC&>zNL5iAL`8}U$}3OI?BqyI=s2Y>}~dWw|3B?KXr)gc1o<HcJW}U
z5~W=^6TMH18jrLHJ;gb%WG$AM=}<Q>2AxMSE1JFE_vGLI_8c44N<hG$1UPm}gy8|s
z<CY{MZ8RAxNql`L;W@Ql$6?fnYcW6a66^298r)4KCqtQMc{CpD>CyUM?(_rBpc|35
z!R*V_qIY5p`D|G?0rLx4_rBn+<6CjqPTt<`!z?Y<*&QAD@wp!&RFQzEZ4z*tyzRu^
zP75vl%?nbnq$~s4Bb;NPck;L4XKeiO6<J<6c;!v6@<;CPoSREOWG+td{#3rJ2)nC`
z@SAnR#jC~8=&_&s@+YEZ<|F%N33Inf(afX_V^@`<PFV#U+1Ks;tET*3UQ6l})sRg;
zYe)(+<{R?u8HP2b(U3}9Iaz^mkBaf}Fl*<qpLn<LC%mcs6~8XTngH$yZ&*c5r-tN>
z(-Tkq+Ty>bw%j;dTaNQy9$IWH0e_6;P6hQ#zdDj{ZX#*%rZVfQg-o=!l;__#ufm+)
zjg#uit~1n}kGn~ELxpJC(RWPV>h-10%vqo7HRzElb(1Y*Z9*P*5j}C0e}}k=HF<mR
zi;LK2x=6b?7kP4#b#JF}lDf!Gwj}yWLAkfAp>8}fd!%^w^bx&JJ~EHIH70MLB@LC-
z-b2M<-w?6b?<LubyrgJ|m)w{&P~w~hO02o3yoeeg2fp@~W&`@myKj9ZIiZi7ecwm^
z4@qYo7v=wSeF<q5Ol(2L#%|9TC~0?hcej4+?(POVQ9;-)x)p<NQB=f6)aU%3`;UDg
z2+IPy*ZVUwXU;issh4zE-%A>n?<x8^J*0%ZZK&-a4f=B5c1b%K|Gc%-i*GHr0$a(u
zi!G(6skxkTYbw2{NsipvSn`%PlIF)6%0Q#5G&id!+jv$`y{sdPn$oi?s<u2`#C|ff
z<D(*L%BJ2mrT;VfT(F;<$ld3q&NZa7RW<3@sfuLeIZ5Aol_ipXjV|jd$)n~KCH;K`
z>3pMt>>5QMi}DVVagO^wVlNi6?BqCk+a}Rg_8zkpuQD6i@We(&bhVLMhpZ%jla(wF
zDK90r>DRc#QrbIM3X&|OD9u8$7nsYt#by%ZX(nmTX5zc0oQ!E!PL6%~hnefj$&wD`
z#Q)eo`V0L*27N5rPbx$I24zUJDMMuHZ~EW;#I&{4&TGEm*^e*SLEa9FD8Y)YpQ%fs
zhXr{%<l85lG5dhUPv66L#(P{mK%GJ#o*{*v$MiMs&9mmUQTb49$-|&U%&dN!i<b9t
zvGjZ{JfGy^U|cSJ9&*uxSw2@UGhcdeE<MPzk)qAQi%MDO&%Ekz?$9pXl7XsknNh;b
zP{$?dFt3}A&9CSUMZP{<kcNAoQn7=30@WX-;UD+u<BB;en#6q2)31@lbN({cqrF&{
z_T5O2;@|Wbq(<p<D1F<jyulE~OEih4w-b4rLH3Q<l>jlsqvgnWIM$)IcMSEjqqx_i
zh{Mq@@#y`S9*X3xt~<3*tatrb`_{RT1TXG(Ok_4*l?@4)PVc0#%%zRHp8!2`e7ZPL
zcN~(4hn%}+vftXSZY;GOvFJvAZrDR_3yWxMWZ$+m^|)20@fv=f+$<M`idCY~`Ctt7
z&D>Mbadz2&v(F*12<L80y9+V6zAy$A@~KNI9}TC+5xB~FFyl%T2EB+xr4!+ZrxvKe
zEoyP^m@qfkgy!|bV7tPEjEP~0c@&1{hHxB-<eah~65VW~(24!tN~Umn|3zRN`Flha
zg%KW67#%<sv#&Y!U<5Y!g`-d(f%j)3pfp6lu68u~-j7B^Xf%SNqw$6F=dFCKI@a5l
zox_pl8;1R?DWj<eK5a*RQhDl@*6J{&rVgFFwTNI{TJ^LJi>m6;>AMkK+y1YA8iiVm
zB4P100-btA;2rftP5ZGA9b`muCj)NZ*F)`Mz!eJ}qLa0V^3pPAM~i0Tba+s#!;rZ;
zEDKO$VP`cSw^gI(It{jc(qK-e1|H<?)#f_<xT?eaAu0^FRN<1n3cYKpajCBs_1kID
z?~w+ErFtBnt;gL~dS>NlF?6gRZ%^y-eXjvIJN4L0-af3Ur!T$^h1Fy6t1C5Y)OS@F
z9|^ZQ>=~?&z){<9#N_Ic=%>f--g?X^ujljAfZyW`NV#o<tJ(;k9eQjSrh{c?9d10=
zqFkI79sBEWyRHHLJXh52%ir^rNYr(W!Zgl!2R7oY`x&+8C*pC8HSQ|vvir49zzb@q
zoqutM$s!KEeG(8M3CQT1fMxR&vHlwOkm&n$pEHLM)WuC9Zv)I&FHeZU5~nykUYUsH
zcN3A%^IwOB^b}!#c@DLG52#599QlITik#<OegRL;TFU=s?gHyui+$Ac7r(}T%w0IQ
zf^*j$NeJP2aeui4T%sO5YiB&xEuz+%y1VgR<8j_A9-p}%*LPAJO6@p5;okDg4KJ~F
zCbfq=^ByOA59hEBuFhPB8}!?DqL%;dbGWzWJlQfC^$u}XbC~mHzJJ}C?Dl$v6Y=!P
z;GU3ci!6-YpAG#)=IVCN#icX(^bmN96GNy~*1W}~F@+d%w2*tyg=p>d4jaaPLW?7x
zknUNGK^C9U@?0?t(>~zI{&%>mEQXo?H@F}AfrpKM!_vGA$zRMQV41nRA!~cmU%rv~
zUvx41i$#;l$sq4Cyfc(=&Q*e2hrZyF{u9<b{D`uHpKz1Bb<q4q;hKNUNcw~wW4|D-
z`wygU{*B=2W)eQnOlIsilS9R3vhS~%oZV55+41zz^*5KBl`X{-VkuYZuxFWTE$&%1
zV)oubw)JojwYifd4Rw+k6P)DS5hpnt>?8*60B=cl67$xTWoq}z(usB5t9I2SaBOW^
zk;i#(W*xCKx0Rt&OR@F`{fsk43XR@m0eQP=?GUL^dw`U7^ON~o{p4+=pA=5+FFoT2
z$jra}rC@h|>3!5sOyupnIll6`r?>Q-=OtEUo?;c|DHA{Ule;tfim{c4Y)<SW!G3+D
zR`1^8+@+VqS@x9S<n6q6okYmnp&lK?=0!W{qi7?UjoV0zZ>^ZW+fv$WZYjbnz16+k
z==;@J+S)gg<Pi<!Q>d#L*VK~<rn+)pQ%5?Gx0A`+|H#{BS8GX(wx(pUzTN)FS<Z20
zeuh5QlRT=+)_zqa+r6^vGCE4+97lOGmR=SoD+<PNzblWKKWiMM{zwOzL*A-S+KV&m
z+bs?4nG<X;jw<#h_Sj0mapoiItYvr`?%VFSl9WkS;!WO84YCyP#muv{wUidrAXh0d
zmviK8MdlB@?Z=#W^0v^Mxj%K;NB;Z|CmWWNp)>yB_O3s$Ec=bc<9S9KTn0_OGAt=y
z#*B^M*gfnwPM!OK3+n&*w0wbI3(jr7lpw4Mb8M#-V>Wqf{{9mt)c%OlLq$lpEW&_E
z%*Lwy79PLyFqRsq8rAZlXq|_<o%5(Y%tP2a`c&M_MZ~#WX!$mITP|KK%|+B4`Y<k|
zC!=>Rt~Sf&UThXlIcMSZ$V`lXk%6Gi8JNLcp5GhO5z~M>@fDaI`jVa%%<^2ho4K<Y
zskm03SpjvJE#Q>O>p%)DUodOu9DQo3VTuZ8ok}*o&@fMi{<|Ar)1!ei(+_0jOxCQc
zsm&bwA_3c1ao1H9k3FlX3nFhrz1Yj8cf=Xqa+Pt2T)^wd{dn}4l7PgHNx0XMedeam
z;e4L+v&QrmoWmWB74#__$m<=otT6@D)UfwA)}Go`zFk)o3(Iq{2pm9m-ipS>+q{Od
zkE+se&RGzJ-|eE|#vbL0qpU4IM55}cNKCoSUh7ioX0<WQx2IO9HFZ(Dx&K4n-d;rg
zY*sW@m9ckxKY}@<5m?tI0vFDP!<@6xof%;`<QN9i4HGmEOt{KE>dwU`H0JyD=7!;e
zJ{(`kY1tZqn>8Zvg}U97WboYE5l9%xYy8a!Jgv_D?s3jm|3+dic^gSB?{zZb;lK#g
zFCT$N7bEbqPXy|;h`_?2aE$34jw$J3=>8@QXBF&UdxW6{`?NzmjOaDbfJP_u==WF$
z>)%@XDr@ojhz75hY4EIpmN`K>l<YU+Wu+)QTEU(*&kVN@hNHH*9^N%{xW#(5VGq{D
z19jM`;n%a5{5f8Wo4lVoUo`mdnHH}eX;HJi8t#Wx*nD4w@dg!Ik++LJYH;#^2Jh7x
z6qaaE)>MNPgH<piZ;z3;=6NdgYNWyUYy8>=8eCnZg~d=k+&}7Y{HzuUZTR>eXwm+G
z4t=wAX!TTw1s8O%S+B$U8CrON|Cgu1FP<Ud#~bluiV3qcVfYxVN2#wKTgltz<ZXk4
zTAayM!l$PaYuBhSVY?c8v(;GqO$}FT4a(KkU{4htJtXx=Rfl2cr*NcRjzkq}K1b?B
zqK9J)uF%u7{84&dvBv$t8ULTwoZ(u;!@4*YF5Crsl~0YEiauI_vGAvkz49jN;MfcC
zcp3@wCXtvJ5RUeoGaQ;4jc%M-IBZLx=Lye=o8r-BDt$ZtGFOT{Xorj>eB1gQA8lD<
z=5v0ce2F~ve68GHBXsX;6n<i+0{zrBXkMW;XTes(5-==39@A^b!{#tOMLFkRVHS^(
z?%btX$=UvrSRCjchw0tp@px4-Hf1Jb^<-u|P>1#Y@Jkd>OHqF|b0Fr^6NXt;$2gnW
zmzRY63yC;>gP9w<`B<9KKSs+vcMtBr>)t^3Dg}qPaOWs516kBjo^sDc*FJeT%6Zo8
zE^l#F#hhIB+%_C4#F10<hs`d8`|bCLe)a+X-h9F~^Uug_!fRQ-Pl()8h>m3MsTJQa
z#ry|4ulb2i$=_jLnI5&jzTxrLZ*b<$Zs3@22zC65!WEy<R$qjz%nqEutU{~qA7RFs
zneFgggzx@@BJSZkH&1|iKs;(3N<!ThIhgjDHArC*IzRc0slMNlQ}+j!wf=!DuOFDl
zddP~rJw@KO>}M&BewoXuEf!K&VIh-PM+LlNp4$ftS(#7Ygtr!==f@U)VQuxzLSD0G
z_Pt~&%jws9h2GE?6X@0JZzfHaSV`YDw$eYrM%H%yhLbkm@V&rK>RbCs=`~;Zk>V>2
zRld^YJ~dEn{KSjA-9g^YckeGFCiR!SA^qjqj{dUokRS8pd?jqQuk6eAmP>uTWZ(}^
ziJR{!H~RIHoy@7-P2R2<)JICAd&;vt-NkuT7df`1vzRaHB-=C{Wwow@r1$P1-QKj5
zFzy7^8PG=3OIpdUWa_1Sn#kR5)Jx^Ki9LDy#kP@DAJR~a_g&>YwNfs+x{@4GN6cL7
z$ojjr<@K1_(&J1m`C_QajHH?pao1U*f7B4a9`w6d$~@vKRisLX%3{zuN&^c=DW(VK
zQ1bS&R|UD4?I6R~I*5?Bova*W%YJ*&b>Qx;i@g+3o9ueqR;rS>QI0lp;Fh)A?ZLgD
zkIW+|q=&^TOZo4qr5qf`zt3e0sYU(qJLV50k+;)!n@Rp4vi5K}xzV5XZB?G1=!vYl
z@(-^({^0zDQl#(y4R6kEFZM2jqVzZ3v4@gz_BW=C`hl~*ze6+r3mS%$Ab)NNPGo*Y
zA<u*Brxv3Jc{_mp<jLf%j=bG%Mh$X(W(wrzBlUeAHr2|<)^~ZxVej}!C)T$M@(@r+
zuL}EI>^Q}2=^MG2Lf)QRl1u-ZTudf&L&)2~O|#*>FbgxedpmYWCUBg-6`&V-k93@0
zN6wPB`yA5|qNmUC75Z3^x85gGagn@jSvM6wqF>=6J!I<}=w%x~KYR9kCtqbQE9a{%
zIK!&PnQ1ybc?RBOwjKM#5qwQY(o?vtH+yk6;xU7M0^7JFI*@wV|FJRjdn_hBq2Baf
z9R6@eGWU2qhE3v*XIFX~@C;s!UQHSFshD{!5tjqv;lRDj3hejYjEKi!^7iFl>UtZ|
zYv~{NSl_S?`Voccoc(q5rw*2SRULWzh`iN~ib504N}aNz5I3DV+5AYj9E!x@v<M`V
zX@yTX3;at@!)Gy2{fnSRl%5Ub*-y^*=AJe}x7&arYa{Thbp&+tsQ+ojx#4gV@{3KF
z)5U~q?BzBcWr96#Q;(T2b50mm1e41F5zNJnz^s}PaAB|4V}J?O*PHO1bH|PT5m@pu
z0{fP5A7^ePHnit;`kfJjHn7J^o$WHq2=3^F!^JHeYs!aX)8{Y*@HQtc3^wF#Hnm9A
zs)XU9j|mfdP(M7&fSu&+<a0XIV$EnfSBr7mG{{@3L2{`Er(M~@C2yzcjYy~$j=EOi
z=+=+3Pfvc_Ne7GXT5K&~@A#`0cW$r_WG{LXS-7XW7Qf$Wknu!~8%}Ec$C~!z0TupT
zSE1%b6>Jx4ki(iae2oTK^Host<5#|_=)a(XV~PT!7b@WTM1_biDh#`##+r*7dOvHh
z<hvS%vl{$4M^-B}7`RG<MAqZJ5jw>3%#hJohlCm$?3k{luY(rf(=_xQ)xz<;7Td__
zLzndQq|)J(mK^s~!{>+!7S1X>$WY>1XC>N@w<+#w%<QDbjy`HM_^XC}bq!2pa@93P
zyk2WURIhMUNira=m62IeoM)G^pHYFH)7-~u!t>w1^YN%{m4K~f%#PA=PppjE^WL#Y
z^o&Jsa`pso_SB5KleeckM$zLj5|cSs|3&?`)tG4PK1OXEXV8W&%#`BJuPbLCW$B5S
zOx~7IL)D5iiEDg4oZL-roO+4p<gGb-x?62I&o!qu-{u9TP@{Fql3p%_>_bP>PoyS!
zdgy;XXR(;V`ySAU^AGmimoJLN{qfunONz%D^0ozeJ9H9fwkuy^@Tiw4zWfqZRwSdu
zA_=DI@i_V+k+Udzm-!@O&S}n@w=kEgIWrg8Ykt}66^f5?AL)4tW^PY~E`u3Hy|XdP
zEf<^C=b#36_G{Yb(MOWrbm!ip`jNL#*cU?X7oy7SLheBnBIx;BgtGp9a_cQN@^gaI
z-eSP=cL;b<ghSgu;OXyTTtD{_bygLjfmady=F>BPw|x~05pXmQOIb&hBr!ulk%RO`
zxoGk@8!d)s;a=4=I2GqW*`<hCzTEk$9M5?+y|byKEHH3qdEINwNYBBC8M$!M=b_7k
z0@Ul8jR()(qCw_cv~O07nXHXVy6|y27vt;LV)Xy=3DvHCMDK+kxsUh(?v^FkIN40L
z##=}jc^m)LLVku?NGHw_o0GTI$lHq7%*1)vPc&;zuic3yc+>0?YFm7x7p9-&=le<`
z`DsnwE(!LPDI59kp1v|Tz*iPL_mz63e$t3q<i&x0Qgfr9tf}uO3VwdiZocAu$6J<k
z^OE;pJjHChr-XR)ldmh7qxY+iB(?7&YZH6Q^W)v6>n3_yEbc5Vrgai`bw@d9>L9D>
zC$*qbJ6UaPEl=HA%gDE_I8SXUx&F;%fO|7(lhj0tQX9!?*0+%%4dwU+SBbZGk?MD-
zyUt*pPVKhC4el<JxBka#$)UG3rAqgj()*3GbY!0Vz>d{rJojy1au#0DzOpP;I?6fr
zlKsmnO4hB4a)G?PPhGOZaR(VN+CipWvS*LMUK~>Fq-cY!eA;R&l{lAA-eD_h2ODPg
zF_VBZ+m~u92_<j!<ZZxXdLvJ@ls+3QWS@_Pn13<nT-QvR^7q|uikS>KSWXm;{=sek
zUsSXEivzKL7*F^IOHP!cndvvPSj*6Yyj@-L8`;Jm@VfK`BaB}#k#ml?q!RphvjmAI
z`bJgz496JuV1^YVa?1y_Pb@@6MjrhC%fsS`Ja|9Mr~ZZ-sLgrseVd2N`FRMn%0pCc
z4$epp+z;fSjsyJ|!80hCdwX*({?PyU;>29EkX-UE8zFh=@Mqt6_M8k@ULb1^XTX8p
zHL)$zu_1)@Y`t_S$=l`8srY-FJ3l*_dv_!i@yZnRdHDtv=g=RHyd9mJg#Y4LgECtt
zgs)R@{<_AVxv$zJI8l?CeSkgzzZ2oX9myp6`sVgx%{rO;l<(-NFov~2Y&0g;j=`Cs
zF_^d^27?~N;j<_8DaY8u3*oLOd(K&@35X3$M4&PeH+cs4_m4-$@C0gu>BkgK&Su5K
z=W;wga)-9>t~hjl6^rRBqwsfUByu<h9qJN=#^h=x)-FEe?MU-T<PV8NZ09H}yAui3
z%Luf46M=}c5jfr_99zvy=zGnGD|_^aFonVTZWz|>4@149My%Rqz^k^L19DC}`)U~M
z$k&g<O!)bUJ22cmed5d-mv!dD!6u}6hrx477(K1o+ih>cpAIGrZfC-{N+!6Fjd|-#
zu%5urWAC`_QSPUZ#p+h!IF)X|tpC(#_eX`Tb5y96qr}heN{p+hLUKzLirTBNc7h5%
zmsAL{)WGSM36K3usOW3N;`&DHWN$TerXJ16+vrPLgwEkybdd(r7iv(&T#e*#6=thc
zh!0Ys(_<AppQ;dZN`<o~6+Ug(;`1vler9ViKSvAgH7zQv=li3zsKVM)A4r}aQ=`Xw
z75b8$jvG|?a#@8DM^!klP^0B5H53grI5As={^afJk1AXXP{J==f!gyG@Vlcz#S-48
zt6|9+x1x)N`4DREkZ3T4b#b+48ti|o!f7it!pL+r`_E?lneTj1qn@{#-li%Xsj5ab
z^7e187SpWts6|$99;U&{{c0qTx4n+3u>7|YquwadueB1|uL}HVt3>KXB`i9quyKS6
z4T97>|7h^@iUz}2zu%xPS{-ZPp0N=h9oU;_N==u295T7r{AEr&f<DBc`^_i>{0fIv
za4g1hmacrqy<PT=>urohBhCmWHwZ%ow=kS*%AN-GUJ+};v0`W>R&x%q#U%y}IMa2t
zjfHMhJj$~-J#|qcKG)@Q{|0CFm!qHzjKUOZ15W+q+0Zf??Ku0ecZ|kut7tSdiw1?!
z=>0qjw?jC~cp8Phj`Tz!Z!7x7Vgh;FgS`D&5(7tS$=q+nVp?h(B<KZ3q$T6=xR+>7
z-sTOXr(_9zv&>jq52I#pMI2mfBtXOc<(qvti{UK#>2SW5*^3%JgnNCQ;WipYjdCo#
z;WnotDLDhZ|7Bo2{p5B;XJB=1HsF<qYCrODr2JcyCvSzk?R$;>?Citkw0(=oXP7ZW
z#?2vb&r?@YF_8RS{0>vd+jm<(z-RME<kLsMXGIY`%Zl)4-g~4?c#rkBa`0<#Iy9%>
zAk&#M?Es$X-^OBL4Qd#9K3<^Y*>h$%Vt;auvyfhJxp6p7FYcPxm^Bj4eZHm4iiqH2
z<{s6-w$y;0O~<I+%napYsg(W}cghuG#=~M<U<RpMy+S-``<~e*g;>_?Bb*0+#DC=N
z)${M*>+}x!p9_(EsRU_PzCxarV3I#))0Q7_y(AA^d*s2gTOM4U^Rak2{kT>X;9!UM
zI6jY=0rSb*7QS+>fv@QNe5KYhUwKQGF7L<xBY)pF^%F;ppNt@H)7JXQG)q5ON#5RS
z?<=|My(O-RmsI=YDZjlu#aH^t+Q)q*O5aBwJnbzup*`hWbT`Q~c9rh)I!jsqPO`wz
zQ8veR5RVV-<ga5p>1x(S<R^E5GFr)>Pc7x-k>>KNeKYExnn?fbM&kChp*$vUjhkF$
za|;*xe$|&h7InlfwYIFfUR%lr)s{ZHYRTs8n$ojsO?gV*>I}{@>u@!>(5|Y;rpl6k
zz)2c%SF2ZcC2=gOD4N?9<<92{GC$ctRxpp^u%Cm>ZourH_4H{RW-FFsY^C}@TM0gD
zBO}ACr3!ibeTB8?O077Pwvr)E<z?*;OPR|)|JB--vURzI-1K6OywY6M)UoSYo5}8M
zX8!CiCqp`wljH?|;m8cg9;s!hzpE7Oj+SD^h#xrp=POpFm!NL567*)4RpzNL*jb0O
zh!Z7vwy^}!i%YPFydChP7|F54FwXjjaq~YQ{X!uubD8nJBoDUHd8pJiA3^={aba~H
z9%SahpkObVy_OrVaxnT?HYPJ)pgnnezFRI9FmK@NVftGv%*Cv<9Ne+b!BTBDB68DF
zqfI)>tzhm*Tsn%FIkMH7In|um4!@m-n7mZX=5w|4%T#nbn~J(aQxQ5S6*p8V^gm2N
zV@qm-xs$ViynSP01|9uromxM~DC(Fxv9_)8zug?_G;<el59bIy_g*C+rI52O&Sp!;
z#bGAT<ciVM{@kSxMAaDlH;jGTH8B|doBl@?V=;(*Jb4fgOY*iAy$!YGt+|38f9Dd>
zo*I$O?5Qa&5-^V1(N@$O|K-Obrjxgv0Y`I3WWe$$<g-`%y$>}<gIF^TkA%nP2y}iM
zfw8F(Xyg%z*0rN>_d5HjtRb&c&-$z=jMs%QbpC8)y<tRs>Ui9K>+#LOfZ%_66zL4e
z=S*;32O}K$IbAP?A+L%F-6M?{%^q=fYZI1Qm~h9{gc;r@)bI(zugPJU!H;{Bw>jjk
z7kR6*r$(3cWaK;(ipH_7tZzU<lpfkSI@ogd`F*_x_H)!|Q>ub9xuwlg;yZae-I0u}
zLEgBi*bh<R`xX^8tx@C7Z8g5#HDdTy113`+)$RpnwA1xidPoPuN-bWD*C2MD2A8Z@
zo07K&LRFYV-fGF)c@I>W&RdP+DrP{cP;RRhw+veRqCV<ViWa3;wa6lGU50B>p1gg?
z9`NcSe(g0C%2-GKUBy1~c@;c%sIZN^{hG`=(pJOV2R^ptN>mG1U_+4t3(^#rGD(3}
zb_yH_3q{~GHQctSdG=JH?rAlOPiat>Klk=v4ZVj|7-XYne?^V%<gKoW3VtutXzi;;
z74mlL85OLCsxj%l8co?74nC^E;g@Rs<{bGG`_*r^sqpQW5-naUv9GZb-5)4$mb|UA
zR*7U!6*3O8zdTwE&t)2HxTb-B1uc%w*1+1Nfw#L6O;koyW9`2)Jr*Bs#L<H)4nr=*
z;v3I!x<%xuT`b;wXYb^{2y9CUM?KEz>sJWFt9oI`Y8-~+<gMmd7{*Y?b&Kb^IQnU%
z)#3AdPYj|WV{n``?*DXGlR77&gma21<ZYKzQQRHiT>4`ac96Gs$X6ZriAyU*GvAl~
zU+mWuQ;Qx;-9VikQSjbFE>nlF(jyulJYsQ)`nmQ0VvwIqjWu`8ips}gh6Qy98upqC
zVsW?!^PJm|Ne^P_g&YgBZ`|{$%Dpve^jlO<fY~kflDONvZaB4vX3x<;`5d#CGH>C?
zORSjx8vW>jo<AcM6<=gvRZ<3!n1O}lZQ0B$C|z=3QZv`jEEkF9x!7|$2e<C$Apcno
z{Zv^0lDC&f<udCy7Y`ccq36!G7{Q&2`jy#>&Me?{f;$&m-y*wN0jio8KtG;YgxvRc
zqrPj}*yl*4wmXwEk7~0bP?Q>u2b`z04AWs<V?B;6(4((Ok3Mb&>^N>f6nkXO^o400
z7RT(cR|p7C<*rgXf|{pe4S8Ei&#c#JX~3K;6b@kqQhGj~H7UUMx`ou=y~Ww%`52t>
z0iiuVLeE}!_PuxL$$BT}BlAkr3h|ZodhC_AsFPEGIf{HNFU*6H9w4f&c?h)NHF;G&
zwFL#Jyrlr6$=eR>3)?mFmEE;{<(@C=+9AHIJ$=QDyp6f&E1&E5$tjbMge>=w9ix4u
zQ!5{-`Q2M8-S(DN-@L@4nwQ-B;3@6@^^@3C51IIsyiM*S``2}o$#c5OtlM2AEwQuM
zZ|)@H8g-JJF&*VmTnE`&v%Q!qw-f);*5VxEE<>uj%cSs@;?=N;jIGv0=2$k7>?dwg
z_O7AqXyq!68@kH!iZ0T8ICr1O+Y03EwIj7<Q61(#1lJO0*IH6<o3jL;t|3K#tIM!8
zRprslsxpV#<Y`x(M3v|$ex;S9$NP$MpkhV&#XTVXD+g&Y%Rwe`7M|ouJu++Bq5<@`
z_+=wgtJz3T_K|0vv6iNft>x?jYf-y1d+)N9w4%R--7ibAW{&(C^0r-P3mI8zF0Wpj
z%XLq4`8<qse)2YPbvbEIf8>HWe^C_m2d~mu%kC(}irc@CRO1I)eg2An86~WNO0d>M
ztvq?FE&d3*<J_Yi@g6bL3eb#ytEIyW;8-u0eouVfK47N%WBM3-WI_Ej3nR37sNRBq
z)4usQF(VHlxA^|3JalAs^r(tiSoJsyd+TMAkIbY_O~;j9>2U2quXF00dREIs`RbX}
zn`Gcc^>kcQrJ=?S-X1eAa{_Zl$ls6lX*kLpuQ}xHFV?oHs#Ih@NyT`tRH%A0J8wq{
z0`MB&54}VK>WwWp_sKRUVcYK{*yS_Jr9XEKzmT`w!|4>3h|`n*w}+UBPV2Zs%B-F3
z_1L4U&6_Lr!C$!N!TG8+_g4>do@(7O1_R04m(&hVHRp~FXShx$;}E)%UQgWB8Bi?_
zX3THfah6=|k$~RRjI>hHi>WR9ef49}pPmurm&8ELYuHj(&NTZ+!mDQ_Hj%C0J4d3<
zrwEKG;B53YcTBSlxLchyX#jUI2Z!O=GO}vF0rknMPtEl>+>yJP+)<fwP>-|zdMr!O
zBe${vE36IdRZ)AR;cNSw0k>uwVWTtRXblqvw&Wg8ITOw_G(qWU!fx)l{73KL64t=Y
zZP{<8*HUh{5m}{1&KQh%zJdEMGYz=O@3H-f9u|W&c-2aST?f^0n5o9Ca24`er~vjw
z8yS@NSV4s_@^)lx@`QET4)V6*YBl=bQDg9BBi3ZA;JZ|f(=mDs8o`-pAL@@(8uV|;
znz6M8709H6W~`gXsE|L2_3<<n+RaoUg#5bB`qY{n-o8bPD{AVZqBOWmw%$-Gv7o;a
zZ^+A<M-=!tSc#$b<g}X#;k{L8KU#$<hg2x!V^QZT(R+>(rj|<F{-eaC#Y#LilC|XR
zkrxV7o*Ig>wV~+oEEIk3DWLA4ge7@9;iMYfSkFIWAGuAm3N~pfj2)`N3^P8ayGrcj
z`Qomf68>KlC>y6lN7mvET-C7au11nkjk}R#{5I;f$lJNERT!{Vg;hV582Ma@s<oBa
zd{6<cCKT(e6=>2?!R$T-`t?=AoR4o;qym9_%tu*cuV;UI>nt7AJXam#9#QE>Bj)8A
za9}Ncu-`}HC%q@9_6|enc@s_?<6iGe0|MEfs53qcmh4q{@XS>CrJk=R>aLrI;fO~#
z)?J7|6Xrtn;hC}zy|8-oOxct7GvA&*SDX>p*oVX6dKC6BKk0e3NTe<^q9oc#-JA*9
z#wOIC$Xc8CU70&h7UXT=i73d%NNl!>L@V-kK6xA6haM;7?KtwbYg`Nrthrwl#~_Wb
zp>E{uA`^M5WzRd5z9;woulF1i6@#CgoA|SakB*AP)qZh!WydUvYDuWZ^YWKK?!aG3
zMt%{QJM%Sqh0wd|a|(vhGo#A+G(0$(j`^kOm|igx9q5VUw>=xj&eQL;Qx06_GQXa@
zRg$+i$=h!`3(#gLGf*byvNp(P9a4a@wgvd)Sb&cqxp<eCgS~IE>F>>KPkQi#H^{;q
zYPHse^EEz#o+FQW7Cs)1I{e+-s>^x*D-9ZN*WkF79{&H+*D3XQ)YyP+Jhv_l(Ze~A
z=MEG7#OHD+JC}Rhx43&=pSyPSk5e_J-i`hxBip}0EqY3ORZM{=^C`}{F>je!mL_IT
z1XwcD^B6t0LSNxhG;=;rvHsu;+j}8BzGl3^%h_)b$bIPd|GmLJ?q&tq=E2mHnzA;m
zY1-wZ?$~@J^X$-%^Vi;Az2)oyAE|!BN7}mii05rDd411E&iC*Yx68g_!9H`wZ6BF9
z!bj#@^^(kuUb3K(x8z@^E}1^b2U>f|ubKUrN!U-^QanVL)kiAq=HBh(ZnAf2SNU+P
zi?s9WEMc2E$<T(K<k!oNvVTiQnboX=+&$P{-hFE=1C;KPSIu4Koph7uBizLESz}oc
z-cVMoYbevZG?WH)T%}~Qi?p|Lk#UDu-%jH0Ei(#c+B08?yw#Gop+m{sC){HuZwLRY
zE?vmmC*<vfBb8+Wc{}E%qgWX$$<%(8<gF8X%H(YddE0p!vnVX=W%dg@Ils?V&h+Pu
zo4yuiyKH1>1?t96S#zh_TKqh%<#MEz%zIE?x=k)mZ_M)a!LpF`JuKvWdwMjIw{OVX
zse{a=L1p%p+n7mJ^0vjxzvw^nFTO?o!N#C6ynV|oNc&%?rTB)~rZ4zJoqQT+0)qof
zU^TM{#S!nJJ@y`aLBnZk0n!5Uv36iSj<G&nAD08?>p9psEDKrmU2IF19`2TfsK4YX
zeT==Q=3&m(9K1}<W;S*<u9l+*3_re&ynQhw0|Uw1tF6<qEH4ci2Kp>MOM{ZTv%B}C
zp?F&wo-$kH#mQ8xpfBT=5gGWx&+Ydo6^S#cL1qoRX>1Cf(fg^{t`rP9m4fnDQsB&w
z8=2iRnm!gQKP91O3Nz2tNf6%lWGBI0nS{s!zTJ?7KjRazDj|XW$^`0|=|M=|M$t!N
zl{GcUZuAi*Z?i72W}x>%4zC^8D#pOPDX&THF+e%y1k{Sf8fuLHMaCkAyE;|K+a9f$
zf%k>?vnU=b^5RkWkRG@_;&Js3efS4aH_GdxEoYj2pL34Nnq#xpgfa9yXxiHd{HA8r
z#)wW{My!i5;2L@B%j;R|KqID(GSHt-&&&WF26CtLTP;1ZhU+ncEL=<8Mn&skU!=#l
zyVS~(xBJLjk1qy%n_|TH5F^@I(tGKr5$}trhh=^Hm!1v-$=gZfZBz2LL75RVlZ^O!
zk$y*<&8}s?x^%q(X)_Es^Inf0<kw9f6@J*MP`Qr^#eph#hmnWPROoV3Nv~ekYGkfH
zl8pMS#Qk=hm(Ekd(M5@(`AU3xti<JyN}M3)94e|YqZv6ApoT@h8g<#T{YK_GKIi<h
zRE0gG*aIeS-N@Tr4OrKDuzozkK5MWF_jd66j#MJDfdcabL(wNxi8%w6Xq%zH>-`EG
z)hJMf+NZ6J$ycb5Fj$3Oyub0wRmdqO3s);qsSz2+{;}UV1wOec@ndc%{+k$zpS<O-
z4#j{ap%~Ir0k;N96p**8d#X{kPEFrOHGZ*v?YWqbkt~!!N@ToNAfM-pv*fL^SOGio
zwv09Z96t7n`&4+MSD|BlHBuU@nPI4=c8vA*CKcX%RiZL~S1nE{VCk$t;~k+$w^E=1
zdHa;J<jhb7>g-lP&!6Wed7HIIjd2s%mnLsr1{=`1jS)dP2IR5F?9cwiigcc<=9^&q
z(gejE6ZVj|Yf=q(t!CdomNizk9{VHN$JglLbH>2)au|kF$K@BwUP=+Q)HS0}Wm7Z;
z&!Tt8BR<!;mwhoU9L>V%?ZLjvE}qZ+h8qyG$$;iDMjR$@t2SWogXgryeL0`l#D2({
zFlyMs@tnH3fFGRqw~9oE$JCdTw~^%Sd-C?-EuL4OvUg1{v<4Srm~Rn-x762Xb*1m+
z5_)3~pf5|(|8~##dEaBO(!eYZGI=0ho5k-E(2p6BA2uam0=3uC-{>jI+Oqs8?jLZ+
z{y{Z*#&u-g^p7;=0H!0~Kb<}847l~p!s)N{o^#Aby$RV^Gdmj>-)Cd+I_6^s<{;s7
z4!RFuzl}BS=kfV41Qp;N{bifi%E!-odAQMxUhbAz_(6ZkB>G*MR@3{u4s{xj6R`UP
z&(!L0v}(znx6L~I>ZHRHCoR5l7Le0PiykfYNF!sXleZh|@cbIAM|sZtvog3dPA{~b
zg<<R;hvW3M2=w9aB()3uV`e0vUSI-_w&MOXcl2YvKSy=$C62G0%su;L-1(ghhfS~G
z%AMs*W;PYkU&OSEnZ&){;12h)zLB}@7Sr#Gyfq_lubAawXgK$&k0s&Cmk{)s_5}5M
z`y(mZTb8f!mXG^<<ajS1IqTvhUuSvA<ePlo%UAl4xBhXyGB(ghQto-ntSeqp&YIbJ
zMo)2A<S8ySJ>^M5PZ>X^pNvcIE3;pE@YY909PA_Ke7lLSUssuYx{DNebe5|dJBgiZ
zCozz>XUN+*<n0(~FMT()l2N@|$^NSDGM>E6C2ynR8jEk+Msm8Op|tJPke*jA;(W?Q
zhLkjr5A*8Fg6?(X_oUh~{bMceBvG5pjQI(%&hlRcXZd%$hP<$?A@<{{$a-p%XKbh}
z_0BW*=bocnH&v1&<Zafy3Np>5g4pzRkkPIVvYEVHmSiU<4=~f#+g3)tu#q1}Y$UlE
zcYsb<%b`}*GW|CF8ueE4zA<xctCp9IcICxkxP|m1Z=bfNHu<-?1k<Z=ucf(MPBN3q
zJ<Vjo*m6?I@E7+E{lSTwWpEzCov;bNc%A==k&VCO-|rGk*j$2n??2=G{Ss81!C3`o
z{Z|(;OMt9>tY+pc`x#eKb1-#V4(7OK)Aue5i^y9`@^-pK78(|0q7r>x+E>nk+pApI
zkheu?**H5Q8#g~?;{Mt!Eag1*`$&3K^kE-`^VppU)H@ZV!uM+`PEliUrhOWIZb`!|
z*0#P!Q!(vEDpuO1p-WjR4r}OHPPX2g^?$tW;+aA}19Em=3VS3eSk8aX)xJU4pf?yx
z-tHoA-8iRhLEhRXC80cdyO18n4PG*nlwP-%<n2?wZst)BHL^t<)>CKlDJGUVWU;tQ
zZ>5XeU!Bl{8Ya%FlvXh?BX2vCx0~7TT0!21)M0O|aV+l9=gD~&cV+1ZJc+#3led24
zZ4~_v{QI%@%(<26o*uKPH~r|TMRn?KnpuXS&KMKEpP)9VkrB_*4A7l3aA%TdVb&l&
z*n92X$N(QPVEzU@V!!Bc?;ZD8ige7H)gwJc2hChPdXl#TxwmPSug9h1)bP@~VMBR(
zIE-b@`+@aO70z=x_dCHJWKpb<`fU?dfSj#s;yxf}wdB1&YueKWBdU=52YDUXde+Dq
zT#Kt|<l-PTCX$n*SjS#qAGE!d3c+NS`nv)<$Q=JwO04!*;^Y%0`ja6wJ}GhWm=cf2
zvi2mivgaz%ki1R1r9>Rr6k9=!@#O71f7X2IYPA2KHi){Ug+Xe15UFr(<p0&T70E2W
z`qUMXxw4;McZ+jV^0p6o>*_>K?FohJhY<A23dRw)VCL9_U<-LQU!}yqG9~@ZRG2!5
zJ!mHt*34C*EqVKmyshi1L}9!F#<L3MI4jV!i-L3UP@HxlcfW?vcQh2;$lGfzlsL`v
zgffGCRI4y9MTH>)Rrt!E_kRp*$DjRcvH~yCl{~vBv4=HwfBwt?Ggat3nKS3TD)v_R
zy~|ktHe&zXO^w6kt>sb`79Qg7Qm6d?Gvg2D3hbR6ilMDT@wd5x=P?BegB7T@LxGIP
z3JjQ|z$*S+e^#oo{g(!r)L)NnX~e*722A;@$Gt~J)S_p`|9T!N)Ju0+u0#Aap36S#
zQJsAW5B3v0Q|X;?TaOSG_marlEw#8Kb~6l@Lg?{DZ?5m;?E><4I_u?^J^36OphwQ>
za17%)tvtW(a#!v%7wGVbH=P6L7<KhXZpF{@)1#O>%3I?(BiIy%Sw7(yKuvu3+6eZL
zBk}b@B;tohp_B|&NHnfop~jwmXjQXfm=R2-vWGqBWHiprjE2RANMu#!yz?`6zMgSj
zOwY~(>_s<O9ZPSOSX`f#fOIG3M3T2L>>uq;r{?)V0^)0Or<gjbJvF(Pbn*>;cTGXZ
z;k+jJq#^!&I;zJrx4uFqwlK#&cwZ*=jL1T*ENU^gFjMSi4lYvDG`KnSRfBS&Sek{L
zb_MuRz5sg`<-?Y>VM5DX1X^WbG(DsXrzc|{{cHjx5ep4*=yiskqU7!U3DnT@wRFTo
zhdYn8cv?q;Fn<ly?14-<r$i>tXMdnX>_i>9KBUI3o(_>0)v(>bvmnoB(|I<XHdTq`
zo0a%&&s{V6e1hlWADkc6=*(+^=+Ss0|0eO_82us~oBxEt%Y}Wa&L%WkWrFJx6ZB(E
z*m8i{y01K2p9#Z1GwSO0B%lHPyl1e-msF&};~UrTDc~A?So~3R##?G`@|K=+y<|um
zANgqSBlD+wNql=RnP0_A!mj$t%4lDS-|Hi$cHXl6iI<GF@{(tIPno^iQv$1cN=gM!
zSv<0zl#;hy)E<)gucyTK?IAxsyUDU%U1d{X7irX|v-JCqnR}Hw$?cu(C8K;BDR;fK
zq%3fkwCV1m&TT36%34X6;-=F6zMHJQ+n638jpX&QhBB=~L-7uGk<|Gv67;!&M3c9g
z)^%jUl-iQ^vX(3(Z&Tjalq=JmnUUly<H*|z<ZT7=b_01kVIy~<Y#k;1SS1N~QbDe{
zRgkcy4q{oK`2-#gveU&u3VxBbDm(EwZ7Z{w0W>7mMl3kH?P6&oe@|M=Wb*d4!Aee@
zvy^r7ETuDh$jJc~V#E6OcXtbU-ke#stZz$RnaiH?=HeM|CbhdT+t$cj+x7o&e*Ygd
z9Z`n!J%8g`yI<H|^b<)9zT^0>5}evn!oTNdY>)p4YijVU{=Q?rWg%X09^gUV4&BZ3
z-c>#?cT+3C9QVHDZ5nwyxLOt-m1e>thd#&WS-)Cm!ISl_&95vZa@OJK#EgOTOe~}3
zNq?UG;xP7v4|A7iZ5qUv{gC`r*cGQ@BY8W!T^hV5rs3DzG*n%hiX;D+A5uOIZOGe_
zz*OvAnSw<#Q&2!Z)pdPSaEXk)#=2HBBLzv{xsS)Q;IPo=%<O)S?@sI|b5<M7-Pp^9
zB&-WgLOs59dY*`{o{1R5vt&!|P<;Lxi(T~YnM;;VKE_>}#WCnIBpPRyMZuvH=T;8X
zhHQ&Qqu=b?EssH^uQAZJ<L=^7?k+xx!O?o`c`>tQ-z4Uj-;cnG58+7O7>-D0&O8_s
zhV5VIr{rkBE}nlM^`?e3lshATOk_S^8|#cXRENGxPYkfJpe6>?p`J2e6M6ffA!nzo
zTiw^|5qMFD5WZghFX+%Nhp*RQ9iGjnHv((g7wlO@Cvta${=n^612-XW!$)&&q&MJa
zH6z9~HsTli%4z&QmB`s0-A(Ah*K{@3y9Ieh)N5$OHy;Bk^13<C)_@t0^k}<4i;?8*
zvyp1<+b9wCM#;W}5?4pCekE^x$lJPo$x-%RZ`>evZu9+z3XEc3H*^(yw59wQ`=!l)
z1~I!X7_MiOcoH0hfmebMcIg?eAQX|8$tQaaj<Vh>Bd^bOQQ<Q8VKuCew>2Vn>#F!W
z=UlTF=fN-d82m$UY;OoM`vxJSQV>G$3=S89(cU!}x%v=nYpuX}r4rl7+Z5hh*;_tJ
z9oLR1CA!^Fq9tqX|D6Fc&L|MfKJM%-3arQpMe92u%rp;0wT2<s*)tTU-YTi}Qb9%D
zR@tRObEOLDalEhoWT{q(5PPyXHx#YBLUEv0DBALCy<e~=eOO5iD6gO7{J#sFBWtO3
zCU569P~$>**7J$n^-WZwb46;j$k2~I3iy{}O*<@<xq+dusvOEXEELoChoXZQYxSK9
z^c<_e(ojA|Pxhn9+r^HYVY8RME{WG|^3>>VfQj78Z_9s^x2q=eTt?nT_0pr}O)@r`
zyiKOh2f5jKl^#jdut!#>uKh$9jEfEEH_QNscLsPn8M&)%fJUvO4<yffQ;kR>LsIGM
z(X5*u9gpeY9<M_%c{_!?{XpK1_SNImay`cIyw~iR0atmwcj5i+T|@7!wvl|!{x2&c
z3a1nJIwEfismt1aCk7K9#9;ij80>PTmuDsVMB9X;wM96L*5TMz9FEnT1OHUghl8)F
zg)?H2J~S4Ssq_EA{q}p!=%u!j-XE?!7hg%h0Pf3=ByZnPqw#9=YqXAigXTk1Fr-o{
zUhYW6j8++_OWq#xrY|gcn>Q*8y-PFjb3!Ki&dj9mStd?2W$u_?HtO@^cmAg1g(Y)!
zhte0aPY$e4WaGF4XRGw5v;FfNcHBp*zL6SB^0sM63<}qCmQ4<Bp^mZP1U*hp<Bs=4
zH9C*wEdHMo@#i_a*ue7{lz_JqYf3|LqJ|oc$=hmORhZW8F_ahYz-r)A%(V%@`C}o}
z$A`f5D1`GE6*^Ei(K=O!azojR9>%}JIG%SDJnu~6*{GipDLfySlDo6$-?n(83I0VU
z#4ZS9_H`KUR#G8Ug5b9CF($0`$AnSWaR1OXOzP>6z&f6CQ9Q+Nyq6Rtc+scdODv~&
z3CekiGkH5=sh5Nt@RG|~FR4r3&b;qM&uA}MPTtn{@f1e~PpSB;pVS}gAs65Gk(q-%
z<kx1A?#p|~+;-h$fOHijd7Bm9Ntzt$B!6u>Nt+2B<l&q)GNW8;S!`%2<!`o>Z<#G6
z?nO&k;M`RDOmY*WrJKyrH<GM}4W*CPRf0?|(tUx87(X?TKXdAf4?U)jkE$*9^p)z~
zyq3hPYD!^$XSrbJEKTThQOUf96y`A(k(u%jZ&sG7bLp*F*HJD9R*=F72br~mdA8y}
zzfC*YvcgW%m`_lCtDOu>ww2WlZN)jpMpg{v4$w<$3D{*V+s3e#{cb6Fr!1v(rlm})
zWhohf7P4j{ncK}mEGyHyF~?jCI&*1U)l61WCpBe3IawA>Ukma!!u1dC3@k&(DZg>A
z+%G)5{S!k@eMf(9?qqosL&84c)4Y#(HtRiX$lGx0<@?bms!9XSZm;BF#Eo3)sd8|+
zVKxqGvtS@^Jsq>q@k=H&QJF|Por$x2`^6?3<LYH&)zvJ#;r~~E!}Fz82G(y*$IMpg
z%+g`D`<gVAd2rs6m5N7aQsFc?mFKfm#7;;<I&US^LG}K^Jb`+t81^THUWF;lIZeUS
z>D2#^NWm4JWzAaBvsL*TEw8=Cx=-{wUrulHGt6@T#=I0mBG!&bL=X0k%k)W@K7=3R
z+vXRU<=!C?11Bfo3bhL!KV#uJij1XagXT~yO1abe&YQZ+iBXu;JPKdG(d(%qbt2Sq
zzV?d7v}^PeZWV(`N4N(`-cF&%-Y1F0%G2!IHHp9u?n_qrFC53JgySN4yP-Ta#Dfg%
zP3myIk`~#IsiR3!!*o*(ua%savaVgu|9@*{z=$CR4BT$O8xI3&xEf&1Uf|NTdS>Ub
zW*MP_2em;%#%OWLoxTnSbjYFZ=n=2Gwpy~1wMLH+>X2B6xCQADxQTnCqjgB7zr)Z!
zTDWTI51}&Po&!C5>9b&Ul{3IyM%=wk&hwh-5TVCA)<xs#xzu)o7NY{xxEid4*ERNR
z`zdj@KbbY0_3SGJ_UIL?A(iN{QGv0n2h-9*(JUhr?f-^ie49{uNOLwy-ah^jgv1uX
z==mTB+inM;*`*+?v<=4itY?_vNUo-<p(#;fDQBVYW0ZKuYfvzG`+&Ud>dg5m*?zRT
z3e}Q1BOdz<nPm4*^7f`(5GH<j3a7SD5o;Zc&>7E=z`FP1T?P8YDe&fz5|hd5NY0gq
zMkwi}!24OkYabs|_;v-h9#G)$f)K167lPBLLy%UDkGn444+>#_A`}Jf6o{+C8E;!^
zuJ5YYk5#cJ!yY<$s~~T!(?gNfHxv%!?e*%Rz<mWw$CdOtP{N%*Lp#>^k!gI)-<3F(
z#@;r2<-^F^>MfOcV8y<6V+GpfhvMJnQ0_{FVidpco=Tw@N8VQ27mCjwJQMJD8$#|5
zJjvf}TLn(4733QA-?ud=P1GTsUL5l~>S5{4UF2ALeNgK?zC?@Sf%Flnz;oCo&Kj<;
z4o%SG#AQ7$Qm3|^bN&uh!%$@}J*$3b5FfxkH94E$Xhdhe9sEp(Que7|Pd37|!H8Hs
zhg_a$(X}RfrGvDXIa32^sKLFx8urjNNNub|hdVly@r>ATE4?qNefC`)0o#u34O4$_
z?Zw*G#5_er<FZ{0${%B{)4dq9CU4zph9kt0y;$=0msvP^v2L>|iokFO`ieeekD0yZ
zVCtDQ<JnuD&0Hk*qb88I_m{=saVhVIJMMkW6PYhXf6AV(P;1j`T#9;w1>I9%mBM*$
z%TzSBXHHmg3U?)#t67?Y*FSg-;TgWm67Dqg%tW$JCZbK5=t@`KdVeyY`;mrk<(R*@
z_BDOar~|qC0(~02z}!{Lu;#U6#-{|-IS`Nf{o`QgPOtG((eO8mf@3Iq@O5=as>YqR
zQEFTtrozh$3S=(~M)s|z*qrwSW4}Jd=bRw8^R?8v<!uaGeiKg&H?VNbb=X`EK-seZ
zoN~C1gz5Lu@WCUT+8l^Ey8_|6DiD`u2g1WW5W&>~VPE_Rd+i><^87<|9sdxcnmlCP
z@&M`g0#NR?Ke}!Yz}LtCw2277-|g2Cc;q@}Rt%sA!Zo-wzlPlvui;TGYKl8~N~>O;
z@+IC&5=y+J`Uo!>X74FyZ+l9$WnS`tysh}qOB%#@%5c`Ou{mUG^Pb|fqK7<a)I+?R
z_mGdvAf<2n$e}I0q}wu)Pm8)sEt_sK=3f^X9M@SIDmzJ<pSz6t;V$mG+KW~7R`T~)
z3)vdkMAp;e;$Ewkvi@dM2`z3SZzi}&pvFat@?AvL(^ZOmTt#Vgk;|i9r0?hY;<Tl{
z^rpx3Xl6asI$TRORjnoZz05cyZ?}|l7RyyNq@ZgJF}qhy?A}z7q4dRUH_u5NCOOKQ
zQxzm3%t4mVageww4st5nPEwZI$>`5^qMB?cgX`MK{@%8-a5po6?pn!$j#lCkS6;4?
zw;#V+${_N#sDq`{d|@FDXDsB^Bnye@%KUgo3%N_)TI<clyN;QdALZ`!^nci4{ukb9
zW!PNr58@o@i~PP6E?Ga(B=HBFH-E?FD&L?E{)#C#xt}%sBj)~lhpBOeNE}#*#xVt$
zQm+83qNyooo!h>UwJmwOTFJS_#w<*x7HUjvCVEF^!u>=h_DcrlGCLsnNhS`~%R*$w
z3{;=RJla3$xV<SI4_l^VS6})<*`?vKbs9RPrqcT*72`T{kM?g0_McCI!}=6N>`mb;
z@(p_XF*9H;^#Hci4|t`Z>4X$q=Ir)B>l9=*Pr)tnw%%3FZz5lz)x;O*O5RTRoP=rH
z67i@#{SwC|V((klvn}bF(~TJc#}iS#Mj{gEJG)dJhsEq8Hyy#=R<&r@{h>F$O9ae9
z=_zj=0VnE&hkr0(J#`}$UsHog-d<W0jb8_}=)aG;*qgjn*1>5bdqw2!K{xKr<cH&F
zKRs5p;Xa7H9#7~U_>qilc|?N<YE2{SQ<Iy>nIL)Fb}8qp<n25^?($^n(aT8xK=SrH
zYZ9NGdYs811Bd8Qg}mL{lzLwJDQIVCF=h>SVA|<$vORCCo4q|*|FX8ZJ&)JET{>)K
z->NyUl?}b<$Iws*6|ci-yY%#fG~hCMdyaMQCtvRNxbwPPgMG(~^flD!QI0>uIO;xX
zywzaHRnAm{l(4#@L^)m~29UQK*+;GGp~UMu3Pg9N{%F1e<03*~9ZT-ShvK%c0_Nit
zXgN3-{r?1ERz9yUH-j+XVi5CzgRt*l5SsT7g0}WEY^?tb<2yY=?8gxF-%H(-LCM~v
z5*mKbeHYmm=J#3RqQv5@p?Dn_0?XDRm{1yoL7ju>zYv7z!lxM2`YC+Eg6PTc6duQ)
z!tQM_HvJ^4$XjJ8=f#_Of0sD3<^9!U4c_~z0(0gm5Ek?do()3qXKV;s<v+s)^1t5p
zXQ<EbV>K@nw(hLeTPP939(2V_C7%nN$+Aa%`gSM+s)i!7D1`f)tVhhBLDM({C;9tm
z`5$@C$6kSYvBpQ0P?5Jg$lHanDqQ`>-6p=>x{QzM8fUci*cZqS#h(qKXk0fG@1BI<
z-^mbcWo;8UgB<UzK*yzu|6k{NPY=cBr=h4vJ@q4=9W4H`UM4GTr|EDogPt4gV~&Z}
z;w0I3x0w6M1Niqh)6=s{kFMlwn`?Sp;OyZpdy+dNxkI%_iyX4^{8tUK4{6|1roq><
zT3B+oxhDH2#e7aTKE>xtMLw74bLD(k3vtvT+gytu%c-|6rauQcU4#2k`FyT~@!U6R
z1+yWdITtw?hUD+zxKTe6ySO`E!!ru~$=hJ^c92;NHf)Z;m;;;{vw!^6IUEh>os!48
zHl4f;e;)>SY6g?7qVWEH4F1k$A8ASqK0lAf>da`YbBe(qH~N<@jlmBiy}Y=4v#@y#
zzV434*1Ghb<W6h%XKzrgBl}+oDX3p36^|`a;ZNR%<)^@0NAB)nX2JsI9*kjLn9ECS
zDt(E&mMN%rn`d=qF`V4X9i!K8u!~-i-P*iDU7KV$RephM+n=LCOd{O5=Qm|;JUlX4
zbF*Hz_odb;C=#X~k(kf^JUdIccSemV?9=$KQIO$5sN??>@3Nnu>6XU`%6x?Vd+%a)
zr`wo$;wI(=-N3GX*D+v10FF8bU}Wn6xc<F?hzGZ^w%Z+S7;y(9=ik9oxr?GHcagC6
zE?yqGi{B*=QTg0Mv|0F&8s>*^DGb1g>j8N6+#f#VZAbEUk}&|9fB=}U4#1`g0dOK~
z-78;1?e|yVG^C%*we%Ek-o94!l&97DO8PGki7f6XzU1w)C%t4{`Cf8lYESWu@DxAt
z*7(Xp4p;UNrx`uueB~ZuOWr;|-b1Rt6OlH(<&kR-X?(t$l)djN`<MkZGPbkaxZg=;
zkhhh)xXT9nma;+LT!IUl$j^u-(rbNFS^c7^G(X`c|1P-69eP}Z9d(gSIW98gw~OrY
zaFr44CvTqaA{i|j$QR}S*FIO99@DkO>JT-_<n5%LHD#8cv+OLZArUKUNIrQh<ZZ&M
zDxxNDe-Cz&vF$6$7V>t+VFzX>v7h|OPEKvNlLt%eqzARh^$Trf_f#9{>}Mk(WBwmW
zXB}7N{&a05L=mu2z(hp_J5Yf&2H16sV|RBPySuxal9JrTE*e3^qLmU1Y_U7v^?Tkw
z_6J2kHr)HZzO!b{8cR8Jp|bdPVz$K{3-NDdAyu-?<@XtLiS24G#YJXfdCp9d$=fW}
zw=b%ji9K^ZJJ&Jge#S~NeSamXG3^hQleaxfe&Jr^PqZ=pi6$?9U{cXHo(aFA#l<gJ
z$(%9MI?PIG@)lc$zQJ(v_7QpeYS0U8xb=*hDLv-o?QHUP{<BBe;KbSNjEBezqgHJF
z5HD{rckxUao%Cf0b>zPH9%YEx{Q%#oFCXbrhMBX<$g)yQUtfynE~R+Y=RPh}xsO?V
zO{9~rjmgrl`XYKkxnG)p?>nU!H$N8PId|gS=00izd0U0&&dZ~UxX-*8soY&Vr#V@B
zwTK^AfU+e8xXM}b{+zpS`8o$S3CyF}L9SNL;VeoHx)-tM(jXVVXK`oRwj63YIXK9C
zi-TbHPNUnfA8;G%Is1?>fcXwZ^gyVuV`;bRSlN!*NZFhpol8A7DHRQK*)t`>YV1x%
z%U#Lv_f5uv>M1Z?mxzy(sMpcgWX3x2qT^LKb!HxL73zG{fj@FKbW^J&xL9!(^H?Hk
zTw`xEkn=D567lcFM69@<!aQJV$7`6cG@0HDdMn=VzKX@G*t?}}SNT#3b~`6yxd~@y
z>Lg<dXKBt-Up!&S?}0p@+K2V;=oD0EF42$wQZU4X+5PnFd>fR2HtiGO8XAv27g?XO
zMw`vEcc>)b)nU%K^4|H9Ig0J3@~jYhm3h&eE#iIj!87W|>^t?oz`oNCdZK#9Gb4#J
zqxGo~zmA1all7vfr-=Q<CZPr#k0nDV>hSlo7R%mgQ8Gx60c7LOCpwHjpu_JeI=qb3
zV#GlmcDU+L-cJYf7kX^yYQWGWBerd2_5*9f`nTdy?PeUx*T%teRUDeS#?hZm&#=;n
zr<)8=R?{KRO~+hF9TqLr;!ImDW?O1GH>$<G6IvYktV8>^dgd1y;OZ8OyZpM1_K=t6
zd@jCmC^#C2;nR$mM&2IO>$!8)fQIDl`7#44HX3l$%7C21tjSO7u#xrn?-k@Sd(S7U
zCSV#lnxKiro9{-vctTFsrM~@M&z*4w?tO}b>&94oB_{(D<MGeIc&xe|3m58$m2W3t
z*s=tKv*!OFYj5=9&y4=44-aB-X)SsC+z2zW_Sz{Uy!?%b;P<^)IG^Mmi|B8BpLHzu
zgv7!jh1xQC+qo;Vs6KK&H-mW}Ui5f#=1Q|H5z+nGH)IxwKY6>yg!#_Ae*Wb-%JTwy
z`SjO)=t<s|@Os^njKj>E?nT}n+r!+-8%d}jZ?n^qk)q^Ti@P6Ek8pmtQ3{sv9Jl5W
zYi6%x6#ZcBHktl5YBQ(kyB<-<*<=e|gZ#MEt*LN0o`zl2G1F40>oAAYY%k}HTHZw8
zhZ!i^mx0fNGqCH&b?ov<htH67c>QG_3q4*nZZZp{Ryx$r=pp3n+u+0O{|>%^>x0SH
zspM+tEi4G8hl2HSHwXHb$=GoAc6^PuaJ|KCq|vXa>Pg>JX(5hW;trCU#aO<x7@K}F
zbDW-vZKj#bg(e^AS*+kZUv<tSE#?lcAJnPSx8LCmA~P<p-J=hq0I?ngcrm<yet>+u
z9iNA2?p#`YD;H@cIoQKF{l%u)s6*Zgd3%H!*D!i2{m9!H<ZYuf8Qf)h9dWK{c)p0e
zkbFHH-1TraXsJ`h;KL9N?4ZW>>tW2Py@H9~LeZgeD2}%Y0Xhd`NUdOOs2+?b?Sir9
zZ3v8Zp*Y$w6jz)=(Y$3S4)+K}O7l?swhe{-tYCz93_#(K09^78#Hg?!X2@U0K*MFu
ze+8i1Gk?6P<BuxYe&|)=kBBM(@GQ9uk8hWefB!Ph4eKkjNA?x#O8vxZHRpK9+d;3q
zWOj^~jLz*XK2g1-7J0jMWKXeN4SCiO5>^G`|H)l?bFQ|wwWmb2_7szop0Z-Hr?lMT
zA*xOu(sI4KR4wTt=T3E(ko0cyKEO>vhP0DE-P+0H8g1lKCs(=kq@|=SBXj?wKWcY#
znKin(9Au`2vxSp<Khjj@W;K!7VNGOHw<c1`eS>8!8?)AKB&O+((&eC|Z0zYM-wry6
zS4{_TByWe2x6PR=wS9SAF>O^>+B~QuKhD&V%nh}rY2BJKoV;yz-bTvE+u#3MOLg+r
zY^RmHUt}fiOszzDv8uG(P(?;It|Ak-Z?Eya%Cg0S`}WA&DfCG7C2v=<zFpDTT+SAn
zNiQEW>Bo!<_l{<AE80|EFz0iAP5P4yE6KAJmE_{aKRDm+H#RM0c2xKeyn6Q?^LBj4
zy3}u2zve5tkhjZge8h!H?{VSsTQuqS28+ns{A^~8E`5$Z*PmgV7xiRjG}=xrN7)1V
z``SLj4Dz;W_(NQo`2bxx$0JM1c)u&f0P;4SHLaI-8Kygx;Y_D8<^h!9Pwg^j$CaXE
z)O~E|P5*H~31-wQLF&<BvbPxP*A~(GTZrq23o-d%Au5e6!nOm2xEOU8&DY+=#-sw&
z?pFkR@>RuJwmQ$GH=7osxuyVB$lK}(`S?W6jv{Z@F@Hko$6S{m*~q<^gF4PR?4RbK
z#mpRx+mwSI%#00uej7uIIG=O>7INdMB{L^jOWrOlypG73*HJ*<;YI3WHY2WKlcZre
zd3!wXDymH2J!oPwDySU|`<jR^&pD5|h#3Xs*ohele7@{`I$lMRc`}->VoqUFJOchu
zZ=Axp%narq@%~#al^ICnZSEfSM|mDrb>Ti6_S*j1F)M*Jg$q5qVXG3+cjZ-RLg+~*
zOMli*#=Vz`nE#10CZ7_JZ<UBZ>qJ!Fn255eNf;EJjL_I*Y-Zne2=9q&=-E+|x2A0q
zU_npQr?c@`m`7i1n*>}XZ-<b#<F%Y`;l29xr(|@{UBy=3r~k=H#J-R5P%eqbh@J6B
z-N1U*owF~M;$W8+i(^k>@#AzH&bh}y6CaDc_6D@MphIqR9bU(4G3&J!t;pTbj^q~K
z@3M=mouI>iOL^Z(jluLsT1+ja#&%wdHYIv!+ZfSeb38s$tL<}>-V)Biyk-wL|643J
zS;b-3G9}FYm9U|PrX**F_lUugo-r6YAO^$KF=+ZahI=lwSU*e43~wE>_Uo}GMUN+c
z4H!IEkC5b8^vcxY@C+TSBX#)n$$(e97uL|}5frS)FSQ<X;`Dem&VbR=cz-;kL)}w4
zoFH%0YU$Coj-IoiI!vR^eX%N8nGlO6$6}eUt7jiqj~&<axX=2%gsg8pLyvKtSvSW~
zQ;&*;y-zHLU7{DNYCLx_#NjjR>sC$T5NA#gGWGYBJWIG08JQzygf-6rMrzzs+Q+iz
z7>DIu=?SxmMSovv-hQz-m`0`><8iM`B3kD&lhQp2H|r!}5qZ0defGMxoRebxs$uUS
zc^)%CCM3h~FXwkVCu9AfWE@CO!GnBeO>g71%fI`6OG3|eNjUls{j{uGD@G;b`AX(?
z@!wuEC+irm<32r-QGFB7cOCdO(~|I!`pmjpiKt25{x3g76_kM14N_3CB^7U9r(y`}
zcsFY2?W(Zf%>HfVMK_Sh%oZVUmE>)~#_I?;kp@i!dm3Lk`|WuRzizPaOWqcoWlkFV
zxdZlR;Qag>&|bWW;;A>$ke?UJeUKH@o%UQ~*67dcFk4EW#FiUa^@4LXiML_9i*pDL
z+?jcf_3PY1jA9;E!GlcBwsQ7>dD6?L+`*Ls&Ldfqw+nCK@7Gk^=S)Cjp11Sqfwm)W
zRjg-)nN=n90q6h9L+PMA&LS{BE;1L>%5$)jpIbMcUW_p2!O&A^O&@qwd!EBJ%xt-o
z{C^o^yQvAC{$ON|xE^<E>T!tI{f;;-+TD+VNfh;9Q3C;Cn7Zo<p8O2O=SiWMJu3t*
z9fMKjLJ+(+1#$0q5Jp@N#E2Dv7~dlh8~$F#zqXe-n;wX`fFL~F8-(wl{Bg8h0Q||@
zr{@EZ@+JV^J7312fXnDJKY%-f{b5trA30fm)DQg;SMCoF?*O#@dKrycT*l;@eWll9
zZ|U^hTS}&R%g1`&()PKRTnX}$zTbMw9eqzJ7%IXTVM$*A=}+FCwt`F~Z;#$^mlP9E
z>D0<oPBinB3@cB0wbet?db`WZRXt>QVGo&pqPs*~>n0(|UFE22M=6YMBm4htEw)`+
zNzj9q^2?#6^vP`@C&}BeTFoVtye<0ORL1XXD(%VJD&*~*W=&+p@y4=oZeuw&vat-!
z<Gk%|M=>XFU3WRiqGk><VWYh)8d+a<ap$es!n(4_xvoqtttBt7u&!NOQ{1lCklCGV
z<(;36ET*0uJla~OW?M<nJS#DsWF;$8s>-C?DssrLirAC4GnAIHHO^d|m}xPrp#?qI
z7E(9MTw0wsmsRb|<-tv6TkJ8D$*)bNFEcI*eN5#ddAr8OR8H$ENg8=OaQ;ths{a%1
z>oOxd^atENeuvri@6Z)~!>p9|XdCqo0eFYjo!@f){teLS4f3wM!mbVE?IP~9%6$r#
z{!cOY&lBvQTaHBXwmW%iN!~UNe~1gyA7BQvUH;Cd4<WD=F6~QEqj4$RT9x8YE#|$j
zZ~K_cwXegy0|S{I6>*<CPVeL6u@d}i!^{eLpvGCSwmn&hteisV$y?3nLIjr=VgY+H
z^IF}-WA601$L!X4JNn(<7hnQ;J1LNRyvms`xAiV(8SkQBEcfZf<>1g=&gy66!0%Ev
z8gkypqJ9q2nz5d3kc0MqnRvP*6R+Q7(Q|nl#l8NwOD+R{CuGo<%`-W7zB%(XKX2M~
zbfw1CnfU-io^zIl-rd-_oJXCMg5x>K+((~`6|6O%avt@-%>?)iPe5kB1dM&m>;l$l
zCi7z9@6Enkb`p30at`EY9JMcI_&tfonyd8iutw>HMEG;o<PBN9VF3T!hqJBy5;4AV
z3T(;Swd+~4Jmbvf*%VA9Z=X4`-@BD8r4Oe8dApyyO)O^yVeNQkFvjC}Yi0n>PsWp#
z%u%XMEqFg?Xa?|{?ZVtb|9Iq-(4$nHb<_+4&aTko#1v}2ll5pbJ`p3?Bb|PQ-}hg7
zfE?ohK3u5Fs>$4v^_=bLL0y)-ZAo_Bs?Yx7tT-GZZ};nB;nz%s(<fBem!iVT1zNb0
zw*%O#I%%%M&15aE_0#hD(PG*tEyAzHV5CV5dhFNY@i8s(*o&RM!+==+Y?31A(WDNW
zzDkR`J2V*ZCKjgT?dsu5taDT1#eO9U>cn6OdE17(^(SxRlre}XkHPYnF&NP$2EL^l
z^!uZR@~;{b*Xf`nKf8v}&%_!yZki7NM(I%hz5%W7=rQK39z}cg%v;uDEO|R`gaK`*
z>aldO23DU{NYtpX_=*ZYud0x=REN94I<z(Fu`rRWC2!LX>rwfh9t}_HafbibizJ>m
zYOro^Lj7DF3q=GyP~>g*Bh-u6$Kp=&SR5)hV##eCEFugTKbiXb7;5FW$>-fh)E{8P
zoCSQ_E*2(R=)HC@BD9(j9%sp5@^*_h4&9vMP=`O0$FJj<{mmKPw>%rE6LD)OvmmXL
za8Z>EWRkUGlkuCmTiZq_V_L^#^d>{cUZn<-n+(OOWbS}ShMJo6L-u2PZQ|VXF!Gzc
z{kSk0&aTPWP|Un(^7d8NWW=#|af00aluG@1IT_D0V?rRkJ~I<g(~lnP_nccg$Fma8
zZv(C|)8lml4*caz7q4+wheX`$kce1*e_1>idiW+HoNRByJxrrkq+v4iRt{aihVZ)S
za9|EcVtsnA>AgDEoS84=ZQf*NcQD)R3w^f(E$J(WX2#j~>v(G!2ic`Zt;#x7>^C6f
z5ci7Equie7^C`2qPxo0O8t`?{{ag%Q3}H>#Q;kl|)Ob-_jd#y9c>F>Q%?b_fUew@I
z-MjEi;=E1NJ$$e)K*7MfxcVax>x*)^Z#oCR9kcMk{5Fz8ufulxb=dh_LzSt_CZ%tp
z<9+6JC(_GL4WK)JZh>`^acCnoie?7%pRI)xbsa|+6=p3}BI&jYcXTSa`h}yBNf<g#
z3q^~-5IC~F{bCY~*=K^#e_9ZJ_Xxt_tAV(-G!W_C1JUWvWkh8KploOWa=Q6rKofuL
z@%Bfve*Vm@_s3<g033ZDfDqpR{Iw215^LG4^Zuwy-gZ;@q47dL=wJF_Op^edz7c?y
zj+b%7tgo~oZxu>!Ntx;`r^#E(LNEC<-HX26-r~BDb?p$5B|jktXG4xyddigwcWD|<
z-d=Z?3tD$+u*;n@uRTQFyNC3rKWhD)9%5KPE%{q_nV8`wecQOn_?WJ;q*X_WKjSL-
z(_71Tn^rR6nX?=|?<`+5E#%qW<}$fvb2+frNe(=3D&yRn%FE0qQg=-g8JyTyE*xtt
z%gEa^<SoeCy!Fg+^l+5?%?@(0xr6*zXfM`=dXn0|p6qe6la=Ll<@tX#<^A6p@@8-i
z*|4L!)F*G#d~KxqIcvGZ-SSJatmMDxR+44Ux9_S*-Ty4by23)PX)I*i84GC_V=k4S
zm`g$x*0xR-@=IebtLaV7X=5&n$lHti&1C#QQ(4&1RQB~|T|3BB4w;&Yjkc1sUROzG
zF8>MlM?bKq?`P~j@Co1he8PY>pKveyJyO)~QApk%==lySN4>@M=r`mQ^R_jwFk;OM
zY~=pP{^V_4@^<gjCzy2Y5vJ3#7c%i7vH~7p^ymi|I_Uwd%F6J0W*M&e(u2Q_Gd@}O
zv8@B=eJsj&jw{8lyQQdhv=rq7N|Cz%J`T`x?6Ru_c^iwd$g&s<xkvN-$U<zbQHT$p
z3oz(=0sh-p0GIvu(BeZr%)aGg+B#-2TIFHBOD<k9hw%)3PmO-(BB9@1n3K2ZdD)o4
zy#RjsIVdD=TVKqEg1dQoP&=MN-sU^9=8ee2;H{ZhIFWO*^gEcOW#G131`@_(U<T*^
z*9^_T+m6@Kvqm~T-%E$SN;;e#a&KDCG)$P!`v7^Hnw^Yayr0#tlth1IB2L^$z__36
zKklRE$J_+pZmefU#$v8lEIROgQ&`7*9}ouzGG*-Fcr2yor6KQ04JRaG@7hFMnx2TI
zeG}PR=i8TLWC*oK`g9KTPQZVhLwWmv9#*S(Y_dy)<zn6&|4X2kD*<t2=1ip?P>~bt
zuR6R*!gy-V^(OIa9!bEyH0r<u|37cTsbOknaNfn%h=0l3ZOhqF8>452lpfzl8}OL^
z-p=$4O`AY{^KC5Z)rv#jn>frMZ!^cUE^bRLn)kRPWLI(t?<W^^7&t+T*8OA9uBi&!
zrYP|wSczFmCFadiBC?SZaTyBedMi;yAA=P7niB1_a8qk=hP)j)UW<)8wJ12L#UFC-
zgoy#KT?{yutH;+ZtQ%)*pdYA42bBsw{glY-tVFvVN-Qu{<L+rS9xK$$x7EP*SPa%b
zj)8kY4C=ItVK#w=^_?2)zN@kIKQ&C3Xz<G-2E(6da5P0j4ONe2)R2#5khA3Fwt;%I
zAFW5LrFvwh=@H-@1G5x0Dn6+=yP-nwP!(>Hw;h*|wXD?-zS3dABOTV#cdh2<S@`R5
z_LdHAb9sI^%DTHkO~0xd>NiR>`=dmmjS9Cms$iR~LQqc)Vy<gY)=7t5-3-i2X21A|
z0f*ig&~X=a<9<dwp=WDz3nNDG_Y_y(h|%P&?N0;N<gnNMIu@_V+RA*GWE%Ne<nMAO
z>*(*XXzI?~Wq#g)F-D{=GGh2EBTBv+@hqD2O%0MTwLkMNlTuKNecltyxeBh9h$uxO
zCQoOs*ZM>}uxG!6v(T5jGFS6V5*F~Bcr=>lNk{fAmas2O-ge@(m&o~*uBDuj4qzSY
zmVo=`c`b4->UkaZx+9I)xZel|UiYQEj>7moOyM=Vtz9B#^%9v6#+<E_oWCB&`LAio
z7`KAnXl7;wQBO8fhu*R?9W(9HaEM<0QI=_VN?(1E2Qw_)`M#N{$lpMnhjYmD_w(F1
z+lU3*>3cV7F_LFI&ox?Pe9>YhuPOggG3fGBgBb}LG;q-1Tt79AkheXY)adt3g_2$>
z_FN-zXQvwb$lFl%X#cdnhlhvQ&#itBSGe=n?{Oa7$=mTSxSx~S*vztYj2)d0f6f|2
z(aX4Q3TFy5SFvC|Jr8&I+MCALl&b-c9C&?Y@qD>ki|cP<(C2^GWH%M8snOIv9*M#A
zBd{PO40Sq$!Ld&$tjOEm4}<WIKBq~igV4E65NGlOp}rMJKVu+X_y-{FcmVqC4!|v=
zKTeRh7aRK{vX4J5_40=d@rT)Df7ICIkJrwA==|9iZ!G-awA2q~Yy7Z6;Rnk(eh6iq
zd*iS_RwM`DLjB8V;_od(_Ib<F-QMzYinm0QxBJ+i-8a=s>bv)r)3bVt$pGO#B*?UJ
zkg4RY*(-Oc73MCNLfoZuggbLA$lstI(y2=iX+qxWCijrnW!)uqrJLMiUUFAEH<?M^
z7FF#iM|3VyG_94KByaQXILpDB&Js7Mg*;u=TwvEs?DjZGy_!x^(xWN+rA_2@P-DsK
zO)Ys(Bk6Uwp^VzzKob5q$nh4AGSk^nQZ_kAx`TsAS9|%QtH(6Sdh(}*ojfLQx74X2
zS(B>Eh!9&@GtX8!SlfzwF16%J*5Y`on&hWi$v$r@i9A$QDp$6Y(-$hsivkN-qPCFs
z<gM2RbD7@Lj5*P)Wh+&Zj&UZ`olRtl^&jk+{ToX&f5B<mFGSY+h5KWEaJT3;IKBRc
zNY!_29sM1}?>}?j&u2K3w}Z&rl%AhBqx}j0k+(UQ-ZO9i9hzRQ!1J>e7}DhpPG7IU
z9-n9Umh%)_dOyeg15aRh{21vgxX(H55du#=VxHkcnEE_G%BTmZ82^CTV`bPgjoR^+
zQe0nhAC0eb=X=jms7=by;$taJ=9Z%2VeZlMF2&!Q_fc+6-p=IQ?W$s)--<BWjCJd<
zLM%2fgmrcSEYBA}Tl*d~kMH8vn|zEd$i;om(H`Jz&(fN?uzZ?>t6OtWdwvccP0c}r
zui2dM=f3SN+1T7Fn>`fj$TPE1a5)=Z<n4nxIk+2j3zcJUqS0OM0e8LupNtGt?wo;v
z37nsMOs=wTH0B+5>0C|69co?Z!`GNeI%d@6jN6S=?Ap)$-)&Q&y~Q&vdAsK)eJlGD
zF_gT`{m#0U_kp}g%mb_$3u6fHc|VOv*&B<N`LVeD!-x;r%wl>MkMq&YKH^;Gv_b5D
zk-1kkB(M(SeQYqju<hvKCvRVNV0}=7J{z-GG)Ra;JL-{ci(--cjXE!Tc_&%#JXoSf
zzbJZrqj@i`L?5g%2}vuG5I37lX0366b+&;uYAx$HSTv2rD)P3ooe?2J4Dcgwn~@i1
z_-ID%c6hGC{=YhUqN$(u)k9z3fX?&`X;*8oZ9qJ{$=gk=Z@UD>;(8WqKnp$n<66{4
z4E{7w;lyMm4w1Kyqm)RPsYG!-CAMBuAhA$^58oAdxmtyzr7A2xqhZc~2CX#I*S2VJ
zd#@Jic|14tr54-7fZpWod-8U{Obw)g5(`C%7JTffuY^fIvUh<R+m5nMA#Y#q)L<GJ
zXUE>y<*XQ-a*aW8p$1t`)Hp@fURtWgj)@vvj@4j%xdw|8HOxJU!CNPKwaDA485*>C
zr$dz*deq_1awXYm!F#p6QH{8dDqJ0_LV%wNf17L2cCilL50!W}AQ~p*?H&5AkN=}*
z|C9GuYRyL`vv<5N26xETs9`EpdaZ=ZXPyr%S@-jO=H#spd3!2NgOB#CgKZ61F9viu
zU|>CGfD<)!hek#mv^U}rA1%n>z*`1PzhS`Sj|RM>X1^dr50goHnD)}cP{oM6G+t9v
z>FpZI-|ub%?l<S}cC-=8$=eS1jPU(p#EHkTs5vVMd)R+1*vy@O`>BCsCZO-F1cdQX
z`vdzQs}uQuCSfsodxpHd#Ts^3OcJJ$w`1sqUT}z6T4xjSs(m8H(KGH@!ZYA0YBs#)
z=TaA`eTVte^t`<c=B(CUBkGd3QRHp?XGXMPPKZa_L}U(5#N!!>I7GHj9ZRp*684P`
zCgC?V>g9#pe^E6ZPq(JyY(*ND6tE{^o`$PYsaQbXZsQF3foWvix)cn(#2z?bQ?7S)
z7<*5ParDcT%+VrnwHB3m_8Ph;1{Hl`IA25^guLxY-Yz;w#t!27$wi4d6O{-mQDP{0
zdwg*e4sF#Sldq>9XV`Z+a1Wny?sDclAN$#ly-i)aGI<*s7Y~&wz4&H4zc%2!HGeNo
zsqERO8Q>CNK-OLZYLK_BsBes!r^5Jb1&)tXpiOxQYD9<N$%GK>7!!i>o)Kt3-ge9j
z!`S*^*!b!S6uu!SD+xmLf*`n@3gV81AZ&USh=-YhxX4-A=j3fTdAs>w06rP~v8$s$
z8k4sHP5p5I{^;D(A6qB-qw54es2clW?HgYdKKI4<C4QJ==L<uaFGj!c#fF`}sMFRL
z3mf=i?si|?_Vbn}d%b1MM)qzedyB85w-mqel3V?~B+$OMOqtYEjx`b4vBFc{MSI9{
zrH9Nq?IHCpyUU+F?jp>z*xSNgJ~OuwDc$AKzHZV`y2(R>o2*&tCRhJ;lk|06#WJ{y
zZ1~(kI)=DNEAn<Hd7E+7S*}%cmZ|+(NW4>XN&fC6DYKl!HlnH2+}}j9sUbJz?9Ugc
z#<D%7k$il`y|-H&<<1WW`L1%1IqnX!x3hx`wxeIF&`uUku#*!H>q=D{JDIh%hCJ!S
z{d?qXbMkh`99#L-#YUQKwU*Z8?Pl^eC&fypd0NR<W+qP~Z~LCFEN$*t$eu+O(v|(%
zQ}fKlCCE&!bukk!^7ij36UnV+A`4plK}YiT`}JQ4?)?jAD*s{@=nv$*{et4G&lrE~
z6Y|<HH>%YKjA`}(32)vbmHQrZjc+ll(OZOsSD+#Jx_9s!yxm`cPG>6M`}rBF20eq%
z?q?{v{uGDWJjb!!+(q#4F}^VmE%)*x>M#${bP@N{J}SffcV(EnwiGwJa!=ly`)Kj<
zKEkGR|NHL}n4K&}K8kV8mV0})6x095TDE^N_pukFKY4p_Rxze<ZuZreA~dO8gdS}R
zVN2eA$|_(#s{j+f-euNs9$IhU+${I`ygAOiV|t7utL0*JSq@hHn}ZjPbC`RS1M3f*
zhfm1H@u}JP{W=TZBeM{`A`8oQ(%%rry=8gSj=6sznzMkb3vNKy?gpNbxA)s{w<^8J
z7mJury!blGK655v0C(KHxrRy1UHo0jUQJu(Loi<_irGGku3tsb<YZi;{`S9|y=l~9
zjb5CW;f#s>7S3)Cq?d@iJ<5Ae%qMDr3hIL=jYuGS{kKzxWKG~qo(|i^JpZHAitn*b
zUB<spWJW^g1gv1M_4e&}r0^b=@xg$p<gMxqy)`45Lo_QEW3I$wR4{v}|KqI-{XV?+
z?~f!uM^Pu99*>Q#oE0@oMEMTp7^aZztWlp<i{t#M5wCg}S?e3{@hd$(<ZT1;!fU9W
zvjlqlc&fp(pK8pQuE3waN_e@e@mZzDIA;z1X{6!aQVq(<t~!ZITnksC4IeJ6R5*Er
z9L-kZ=?4W=aSG@|6i{zb;M=Qc<U}i&o2{USN<n=@i6d3Wul;0Spav1aWZ#S!)Z0X6
z&(dR~j}FD0d$_nkg}qiP{5h?}$I419wNc=NlM*-Wl=#wDg$v~Erz2|o4Ohc+iw0h-
zV}e;P{~`C+Sy87hAkQAFv0)MG$vhSJHfa!~=l371LIXY@kL(z1*vLB6GlsoD4Mz9T
zz~-d}hx^80**YyAQm1Z5mM@7{Ve=RjCYY<(^HQQG>(wu%N`&@{hD|9MT|wU#Sv@|7
zdiWR}_o-^}pi>N*7}S`smFJG<O1%5T^THD)x{}eZo(e=L707Z@p+$%qU23xaB6k<P
z)zN!wp#EY+wYo+=XCrpKH9%2eK>lq5uJNAU{<Q(o4hDP()#LgEJ?wjO{;9tXe?oOI
z@bho+oOFD-0q$gO+GTP#*MQoE)S^F9lfEB|QPd|ayC&cb^^;BAm^r<WJ^S~ZRr2N;
zjvD@TGwv<?-(zF~HsvRvWIlCi_H>7my^-ybkhYaMBlEbEfx56GKhFOK>pAw>_k89V
zk9ti|27O%gb?u*H#6|LU?LH&YFB%a=&E)tfvezaN=gHe?(^!Kqr|vyG3E7M2YankM
zBqnk9Vlw`cyUpp_`t*nWmn7~sq$d8LEE)U0BqNAEX-Q1PqLvA;`5KFlrUum6%rlOL
zeG41bwl!lgZ<_|IYf<k=RHNlGHTQR@5PnRB;k-T%>`;;SD!7?M;r5D140s=gw$YK;
zRW%BaPDdd<N{g^CJx0@aUaM&wHdkj>$|3H2CT}0n%kkqM=8(=|_Q<d}+-YZk#})&Y
zC-Y|<O0OMxYdR?gK3|mRo}q-ohQIHKP?)-f;$&F}{31f2CvPYI8v@e~5vXMqfpH!o
z|5wlAeGn84f|&&tgoWB5RM`@QuPuV;(FjB`d284gh!uVTI7Hs2fAT|DCx3vvwQt}L
zZ)(b3mHcr!!4FmD`eD`}Ke#mV!(p;kwa*s??R+t$*avr8`C`GpzHnaQix16w;Y!}l
z-{gyJd%VSUo3~6}?=8lE*t`AHM@A<1;qF;4seQYrI9BN?(^)<i(~ImH=^?+OJY*Vq
z+h@1C^kiLYPu})(c9&F#9->X^E-8Dv$r{gY;!VzW+tF3Jjp`!R$99&sy7rR(zJnB<
zb)h%7l}sgXm&ZHHy2{St(zk_-aBMCe9y-x`>?G4Yn~Gva6KOoYvBb=ymfWbZ93gKl
z4>gp^gBr@G6bJf~9Hd7_2boLWdQ^81Tg&=VeW;zRE2+!cysng6))AlJ8scnUU0$Zz
zN|gb&qOEEpPw17}Twx`rIRkWitCfUsr~EkXCb%@nQs&I7ERD$9@pGBe*wsSLZZVfD
z+-aLi-afIZBtEew^ktjKV23}ra_kold;Y?!)Snn{`V-y{zQZ=~J9=FG2ESUL5aIF>
z)5+VFO+T>D@*c_F<ZbX<oFi}dhgBfD^BcSz^aeu@zT|86IgT)+yIJrvr0jTxk1<bi
z?#ok*9?G6=8MD1VJVHKs+u4%&je{Ow0(-V$`Sd}pE5$(a_S?hzsK~sJYNPJs{*Mxr
zjVQ*K2F3Vf#;oJb#fWWE46nY$s7>D1Hf8-gvlv5r7h?>4$-8S6!N#lr)}i^h(>ouI
zyYo<GF=yC2)32~17fSB*QSQw}E!SN9Fk^-){R<gAnD5pw2Nxc34zDB|PO;gT(~%yj
zhMd)_mxV@ive0=)HhyJh;ShIw*)_TY`!neXuuDe}Gw4pQWlv}?vy8bjZq`li%b9x}
z-F4jA>YWa^$Mj7$pr%}Y4PoSMy|h%^8kvfA<n2xNpZ+sXLbF$iSXC<#CFJd|O>u~9
zMb<j-tW4f^CvWGHx1rH|&<iuuJr;GA#UXcMJX!>E9&<D2^H|ruTTk!sZ2ox>=Umwb
zTl+8;gNJe6l0CiV<Za0a-orY?p*<g2^wCV&!rZ_|^z(k#V?23#DNl!Wb96YskE_Rf
z>`3M;MTQ&U#(Qsg68%5aA%i>6Ps{J;a(~X5+8N-Ose>77vO_6a<f?TreHq1^#c0kK
zM?+H>h58;UG;vg8y;6-$zf{N~Kes)OMBKYbYTMDghfpsLQ(}yz0_U$QFe8P3E?2-N
zUV$;>t?LE_EI30LH$y>xmI68C?R9FxeW*1LAh#AZQejDu2Jhy@V9*LJCZ5+|W~v${
z7D~L~eR3Olo4i+vdSl6`%KTbE3MAVqk>g0cTCYS%8T+`LL+rRggBQHd-WjI>ay95n
zrrznRM3_bahxsblWT>!SrNp<9y!Z0o_5`ZY_+$*6Z)vcC_4D1M8r1Bq!JgO59H1_q
zx+4aO{9dMzhm+Q+(3C%i{?vq%c`t3p@B0wzq6DvKyt&W5u$vase`|4WqmFs|<fe<3
zdTtDUM62;^y^^`|N>~<AQ|8;r{i9LdT>)pk0+!w?tUIrU9d*oObz<<_T#w)D_`B+1
zfOUWB`ld#Vf4~`|CkCu0Zy#SZpi`y++v{>(=>Yr1**g5@*~Ewb?vwO`uZ_^*-Y@#P
z78x*~U+2*!de`_nUrk1zv^Jt;X)GKkFcXsA>Q)W8*Qg8o`+-K(qBm<!N6r`X{8mV2
zjwMGIleY)RTekxGpgfbHBX4(-w-1&x7ju3B_Eh5zC=>Q=>7Q%PS>qu3-M90a?|*{x
zP}HtJ^K0(eX~Zh>_OBnA8)QUnYRne=^MxZutTUlz&a>5^44zY%ebQmF4r>&#cvZl2
z=d>gYyq1Ky)i~E=<opqNdxgBU`I?M_Z<7%c&mJ{>5G#(w;^regl2)-#uIBYy7=!cG
zVo;aY0m$2$8vgtjQ~Nm1b0B%^Lf%?9DUi}68i6@cXqp*`uaCp<YezVK4~Rg;*(lT)
zrbE?eJ+82~zVidUY9H8-p2C?ldM@fYaK3m%B7SoodRK`a=1DQQd4x6e0}Z15G*Ir;
zU^8pln<gqO$Wx-pze<cMiC_*)1lqg|!G-V;=5__6d%a++s27SY+d{BlLkQ~7`{enH
zxr%jzVO<o2)zLwiw>b#eRj4UbOI~v_5SPi@1D*Y`q0|o>$=kQ&Yzx-4>#Y6Z=ira-
zzx{BKy!}hP`0xNfL^trmJo2{79$%Ds`eIGK51c>xpi3KH*bMfC+SV6s9DLE1yd5#x
zOU!q8iH!A@153Q*KuI4li|->%TJ@6a(o@ncd&(lt`qUfaDb80sWa%LfsW#hPp4N32
zm$DubdZe3dThmQ$t?VXCn{|_u+ug*sr<>S1x=9=Iwk>)4tYc^C6wpatUviPtyItgD
z@0QZYtfhn}ILnE@Eo94+rZUFFN&H7S$*E;d@^?&AaqZVc+PX9r50l2y>v$tkd}t_p
z);EwFg$|OYbrAPM4swTm+wivJt+|7&|7kBDeC$M(QCC_|sw=y{)RZ}nHKq1sTUkB9
zMlO!Bkyr~GIlHi$sF@|@%^9E><n1>X*0>X^$_Mhc>D<cF?5>5FOsAGi-d<f}E*CRQ
zrO&^nawgG4hB*B}z|21w`QsO}n|>j?&M#a{{E2CEe&8Fm<k4|oQMuC>Sl0W5ajia*
zjqmaD+dG^reTQe{?V)282;eTgF54=w`o$~y;a=hF{+CeJdxD|=mBZ{5_Yq8biVb|E
zZz#v2zUBDu?qmFU_Xu<PJi?hX53qO317zlL2YhrX4s0leb^B7xCU4dH`*>-2A07$J
zVjNYBl5a&=p=PFWP!R@RD8iW&h4{O(5C{2htH&3k)V&yHbBf?tuLxm@g;)}jk4?Su
z5xy%A{<HI-YmtXveRDCEyYt>P$c4@C99$!Be~`ChopSJ$yzSH_6UyUvu!4Cnjgy%L
z)hr7WYGq;dxGem3$l~rC?u=h^8%>woMB#uNaHyXSmyOImd(Eu0_36ltNXKLY8T>jO
zVe8V--Yp#uvgqL;m*Y~|TWWm`&WovMF>B{(P5OvklelX&5iVB};Y{9IjA6d-0?t#i
zhRv{#<$QB2&hwtuhP-V^-dgisQ>O=O(}nDRcH({NFf#?aCEx}#9X7XOFReSX|3l)@
zs$(3!-HOHj!LgWn%ZM%18n>tD*>5tyf%;YbOd}HMj~!Fhh!vyQ=S<~XjUT-?L#a<L
zV?Xh*4*jn4-cPM}$UFnQ$XjpTdv}w!YmOT+OpJKW`?l*H9riBK;h%6V#*l^2bEEKJ
zLKH@wjKuz0QLxw*j(IgAxwA788ooX9IRaS&Bk}ZMBvw_4!b43YBKkyO!LMknVNE+A
zhJDHh3S3TBVB?u+#2%04^Nq$*YXv%HQcLFl(Y&q_z1)>(&bqHQdz!T!S$oE@Zp~F<
z3t85mysgfUtv^ozD5BvQ8IAt=(U6>IjL46|;7(CceT+g_-)Q<X74ROV#Ka9slr4(F
zukVpK9To{mi^epb0V2u#6|EKc^*0*vJRhv*dErDl&jRPz!z)*#nTZA#oXybe(%^J^
z-Z#n9r_2tx#dC!InG(URRXDPc-me~H|2!r9Iw(+6$GUb=G}iu&LjRW0n7ofHk0Co*
zumAkb+<~bIG-Q8y)m#<gJ1Ch6q(GD#|9+om1lMRRDvzet%5y^{CF0wt(8^1NQ4L~H
zm8>+aYQUBT><M^Nm(0_n|4Dkc{Pd_w5A}f`di-5zK;;{}*Pmx@2+vBx*U<OvsN=3l
zE#}_Q!ayy1y_Pv5W;%@gs>S(%TIvf13@@eDIY*CZo(mpS(%~{0?KGP`YM$dfCy|45
z=}$Yty$CUJ_-GP`rL1`?WwCzMB%%)Im*+>tb9RCI9q74t>XCp&)Lt^!&)7MGkCXJS
zpJ7JkKlIP7;F<8Q5nnuv7(m`GCvV4)vCl3u<>HtT_vo+Q#`pj2#GbUm0B2sekM_pU
zCmaJqW1hF}(pxs4em_3UKe3(;NQMvn)sx5b!M@Gxm&pj`_jK<T=dsrtk@QKAk1KST
z6|9BX9p(W)*Pzn?6=v|9wlYwO&rKB=*p%1bk!S=JFgwCrff)YX>Rl9f<wjyzaTrXh
zhGY2KFzn&OToaBpXQHrUwE}i~bcmsMe6~6c9m!i$J+n5u#-iU!o>Ag;NDa`T@;43k
zJ=7rjga)&&N73uf>#tD+!oP>(!rpMaZy1hOL1Fml9EQaiSGbQY7;`x1(~GQ~zLoqX
zZySUM;?s{HY&8u=!nGhQj|zgr#vuB<0}<9N5LapkV)=l}FnJMxy#@3-75ianxgRpB
z`NNRu2h9OLV1Xa}`;*00{80U#FPdKQMe<r-)Sv5%YUHiUEgzV~`CxaZ4+<XpV9{3}
z^s4TQll++8i~ERbx0g)1>m`M=yyW=L-g4Wax0o@rVw$5!NKH@Kvd>d4@a@1^9`gGi
zck%P<A+?Y8kpB!lB$&MYkGyp!Z<Cz5iOm`}Ic?)6-poZF+ODn4D`+FuzOHgI&qWMv
zT*R%Ki>w{kQff_i7N@q&#pAG(w2f~n-#A<QEWC+q9NR?d{nJ<u^lv0z_BWD-5shU0
zlZNtbZbPZPu>toCIY{de2bp!zL7LWf5VsI}X|upy>bcvC(ac_2sOrkO{Mr(`zlIDs
zXe*H&Y-QMZ8yQF5R?D{*tIgHqm5-HNTVf?oJ?NAAUR5?ut|~DXE#-lmrToX)pVLz<
z#G$Q)6lR;srv|2C<x`2-(I#T={RgJ(-6pR4g>uth@U;1bi0(h(=Jf;hFMLC0;#cfE
z`~fur-=VSRJDljrJ}r5>kGu`;`wpj%SK#p03OKH-K+^M9a3XJa?R$x#c26*MVL7fI
zF2@LYf;BcznN3%YSZD5*Aa5_deS~=%A0cJ#1B_f#icMwr@i({>np36tFt!w<O7Ek$
z`aWiUD?wN=_t91`=VfOB>XEnO1B-AOMetlzh+mrvv1~*hX11sQ$&+*HOY$(gZV^gK
z3ej~HeNmqI(C*B`f3vs)zBzX^x6Q?+PPqs)&&3MPoIel9!4c%3KJ&|ireq@{pLvY>
zJ6Nx0wlU{={2Op5PfhOM9Fv97wX-mYygj^%`Nxa6TQ8B>Pu^)bOdsU3d#R}NnOVb6
zxXUAk{Uh#eKjEH^+w^bjUdFsd&L}*My@r>~xo>n)Drz6%93y$#;sIybn<v7`H4*n6
zI19Omz1WGdD4>3J;j9tK=lI&>y=E(Wxc#>oQSjS<WFI}ky6chmkRGMU@mSD-ex_Q?
zKX7I}Q6mBW^oqxD*1uya)BB<{qP(0M;tbwn>4%z6-eyq0a-okmi#^6W3-stwSC0vN
z{f|7SL&_i>;>p`b2X*jd_JbXHYdy$->r)K4`pbY+OC#QQBdcf8lT^ikYAOvL-BF?M
zdIjFJQlQ4wC?u1&$|I3j>=TJ&+ru%QoE=QwUgq29Pa|Mm5rNJYktlGFgp)cF6H+2^
zKRXIb%A&D{z1sHo$(K$FtR-*5$=l&oqH!TK8kd?!<DV~4SXf3bc2{Eb<o{py1yNho
z^-$s*HMlo?R3(SfZn6d(s=&MGXr6(hQ6)1PXD>zLX<ih%TSQ?^OeE5eg=5n22&^0v
z0srx=Bc?^5Mp-1PGY=qhN;Llc8jWEW6&Ow4-Xw2d7e-?OdApLleSJucACJ`VuEd^h
zXX?%kS<CKJBaytnazP1y*6anWM;%+zC&l`BZ*>(`@cZ>&tiZgOXw+H|4gKdRbPb9^
zpEi6>m7|c<ISQehqOj~rG{$UVUyz!2$2LkF<=JD&aq8&QlFb`MqeDXlvdP={PYU$W
zD&f9d#cUDQW*?|eP@k^Jny)j@Q0g>2MxM|^bwLl$1U-u1>G5tXS-g>+vd(&(zscV*
zd&w(bY2kQQ%X6_7&%0^ysRp(1FFa51`E{fhyvp|&c(J!x|Ed;sZfmiWygmO+i>kHB
z41WHCF`ONK&bcmn;Eo2zp;=q{o9eMof13GMWbNEY&McF+YcDVx(}|q5VP3}{o(Z3k
zv0dmPNTpBOIgY!OV{yTmtUYPK_wGjM7t*(IkQrTMY!F`qNjz(Q`>98l96d7b81RL>
z?VrHeD)w2M&W(WueRLChF=uOf9M*+1Z<AjAtf9#$r@yO`AN^Y!lb{`xgi<H&d`M;v
z*%~7Te`NoG-nA}!wD_J(A3y^QQtPYW$8*-YE!6z)MN#99#ANdJAbHzACLG7dhhwRI
zI9`4a<7`+M{<I0h;l)>wdnFXxPlVvpWhJJTQxoZ`!-G8XU%~w9A91)%|Gk0!$C<%8
z*z-&?ypN9i6Y2NcsK%2A^avVwUL6vRUkxL0_hmTR9SKMMM&Zz@!l=_+LA@WrSY8~=
znWSJW{uYeaje_9UBoO{{0#URv5aTv5WASt#woD4dx{S*h{*K<Lrh#yHa2b~xUdC<u
zj;r6J*Qt=2F&{Or`=P~EKhz*^Ta&l5$=f}Z{g86Q7w<!T;kDWqbL@Swx7-KrH+;}1
z+y~uLeK7j254L~s!5V8{+?nIctN_-tdzlxN?<G$LdC6kW-ZI>zw@kX>DVM5wN>O!B
z(LVN&+4M=JOz@C>Be-L4VGsFf+Fibfbd#|6U1Wc87g^A@i!?InB7Zw}k%883GRW9g
z+}E^`g(_F^>gOt3VqIib3m4hx+gk4RX(>U>Z1iZ;T!t)k5-tvq-IJP1+|ecysc0-$
z{Tj;>??z&^qmeWSYb4jo8_JH=4P_^JyFS}N%1=9p&lCEiY#pRimc7iOH`$K7_3mDu
znaQ;z_Gt|X@TnoHowjnPlC1<zww9p#)ui(7Y7$zjn%JJTlIsht<N$fA`c_qHwW%tf
z-dl?2lgjcj%|Z?^OX^@73&|vJi|tKiSO9q&W+KVWO{CkCUr4+26AlSKfG^*Xc;y=o
zslK9M-e<h!zJUhn_c-nQ4#{2KVaUq2Fid=d(8vnB45~m=*0;G^xJ!9;1@_y##K>mP
zaeo*4w&d;KAZ9f#c!cT&%#Awy5YwAH#-zN*+_n7}$;_2{)T<1q9x}6HODS^QN-#%N
zf-mY4p7l#mBe4W;E^)U&Krtfg7bDBH2uHUU;7ymi=xLu%%_)z1XdXN)^6<($4<B0P
z;f_ll>RIL?BRq$DE}6|XJ|8#S@-gM#Jlvj^hjXlN*VW8L+xEFw!@72^Uk(<YU|s9M
z9q(D$uv?vt!J{$}l6eOwb+-_qzlC$bnfOzbiJV;S^qhAa6WF))i@AZk4L6YTGMzaW
zX;@r~d4{i2(fQ+5>Tk?)3Q0!^d3%(+T|?duHNS>i%was6kOoK7G}f=o{y%yZ!B)&>
zr?;XTdHbPtBJ#=G&Gcjr;XTGmYlP-B^<(lj?Gk4^*vmRGmwwTs?864=nWv^lH5;D)
z+Z)lEXZq3gnUUL;bEf}so`<|$v^owqZJBdePS0=?Jsy&&TUqD0&?{qh)_^6{3m>l1
zBZVAR)?x29jy2119ZXDgIK_K!Gy0GAP$O=_d-qUlvbi_4#iwL$6(bhEG~m<>YK-LV
z2c-rj{?v)rvA<eDHY%eqxpoxZ9wtxwL}Kp0;jlzFjB}ZBb0Hju??s^Sbp+ymMBp9y
zdr%n(FZxZstdGLzUJArcQy_u$+kt9P7{dG94%Wc=e|XPp8->V*(b!5xIh91A=a^`;
zVD0KZPXUcWf#v=R7=*lEr^KDV3i=QfIKX;g7I|Blyv?~C4a=R;^sq!RYlZ*!xp3xF
zg`**Pdx^Zw?H7TA6C?0)R|JOs=AY+8qsABVeigqqc{_^Ft&>kQT2)tI!37oCO;ck`
zwi+YZr`}*spEMuV<n6}_&i9^HBJ+w8)`irMo2xKyrV7`7De;K4dF01v)DCAYKb)F7
zpUa37QP}i0673&FGKVUPb0<-FekTfb{i5-NXBae6qSYK~;msAOaEr$8s;u{$MdNz|
z)>vzzagMsXxup^#17q-uTIPt2S}4fH1bSN!#_=q5T#vb|W%ugog}X`rGyUT;Ik&Z!
zyd6T`E@l5<#!LFWPiS#r99iqGh0>bySH)_i2D9fLs)iyr2FJ+TbqQK{b0$l}8h86s
zE$4|jgWQ;Bvww^*eZf5qW^tHuk-HX}#9|!#|HHQP+(+Iz@OpesAJ$QNkJIhA*SI_$
zR-FCq%3ke$=3Kov9mmX>IGBy0KiY|B%oF6Ts}ateXBv2b-l_9Oj3aOF@w|DIHM$+2
z%aTEQJc!a^Ll1h=cgNr~`<^XqnZ3kb%V}_L#35#A^7WL^llzVLC1GprMCgBUSFtla
zR`u!E3TB>vD<f28JcH0{9v7>|<s<aov0vZNmR`J53e@6tV|pnHlX!hTzZ`-4Md9e7
z3dgF6d`sRwBX66ehoPue7+ls~LErPtayk}*tYjsAzg2M#K!?*gv6wuUuOGf%pRz~q
zokriYpB7QewCFNLiw51<>)*^=#fM6$R`L8eCX(KOE0}G21=%K7n18{U+RPBNQiULD
zP6*=5>4EwZgr6OPkn9==oBDwW854-r<Za&iK%86^h?;!^(d5%*>eZa7-5!AGkpUQ`
z_2*f|50~%yVRE1!Ht+Mpd@}aQ6hCBk@I$Ane&|at^6am^Sd`(5Xa2s3TIq{F^~v2_
zA5^~LgY0oWn49c_p5$#e>dTho?RoNcYoA`y^-M3(T<9e`)y$H**Gmq3?j?4QJmrC@
zr+B~ikj)P~<iHINSuo6lHD3?O_w6pXesq&A!QCY0T^HF*-i{}4ohEja1_#^A4(1{s
z@o6U!yV{6HhO2ngca_Q3uJXJ(Go=Q%mTkezxR~uMC05Kw?(Za4)0&Dcd$y}Ro5+wQ
zjb$ur+6#>uiQD^z66@GNR7V_T^KeHw$lZIrZ#&4DZ4RQmXD^06_Hs*MCodE2q$1c(
zP*PiryK76yn;P=bw+4rpY^C*I8)kUeNbnVFvE5ls?EhMcTO}(Q{ZCbSce09X=YCsL
z(<*Z7fTgtiT3M14sU^oUADO)UG{;mthgK4gYL#RWdApCi4e0b2x0OHPr2m17@4mx$
z>KnYmzv4D|dvwzW)b9KqgD$<pybkYRTmLPtkhjgr+p8fJ=y0e4!<JVdYDNWOtY2cw
zw`Xv@@f4=(%W+`W1Nd@hRjsUt7~Je3dft9S&F?XGlDB&fJwjEFGB`cu9{BBKZWqqZ
zlDE&vTQf}w&Mz&&xEg$)Z!sp4w<&H#n6kD2J8bVlU5Ot1*gUk~n1>f3xwz=SomTp6
z<Z$;?b9$bxgyvw`7QQ`@iP<+YIWL!mW*@V#{Af1w;d62Ja}Eao&4Kr!91Jzy!Jc*8
z5s;aUw6?d<vi2?9tj0X_A<TR`b_-`UxA4~FHrmB72f5B|`b%%ZO?w0FHs64024@ad
zF*lg?tr>Y6{GR(`joiJ_@){1bPsOr$=0bUstMp0M{&|hFk=&d0Cl$rilb<S*k!#P~
zfI;+>&{r|xbS!p7Q!n0b#2emM962+Y%3hX%%$-l>M(*N$h<fmTot~La%u+hSULSS7
zz&{3D_MxwoydBqsS-#Yeo6wuOlr?ZqdUch}^yt)7&st4~qKm8x!dM$KQ{nS817`H)
z+^8M3M%G20dEV}Oj5DQg*h?gD8}h#0h`g<xsKe~uT9n?1!N1`#sF|+80`>tF(+yZ?
zW57jc&fr(3=eDXEvsnYIj*7yrSCN>XAAw9G&*opkkn9-_`x)WrO~&QrMsSy31Xhr@
zTWumy$t@B_Z6sz7q7SGm`>XU~UJ8hY#g9mI+7gMiNs&-pWbN0K+H`It^MfMs?PDZH
z{!4bM$s02E6B*dSPl3)J)SC6tILxnAyF)aR(#ULT$&bjlj{Tw#b0HG;tX=CzgmVrw
z97%n{5zZ{UB%g2;-3~`{=H0y}Z)<&zz~7nCNP8KL%z6r(YN<eXdba-I|G9*E;z{y$
z>c46n&QT-iJ9)$z-F3&*Si3=usC@dLO_T`e&7Sc+C2p|ieC0s|6tg4HZeAGnJ>xln
zwfe)^(O6s&g_S3xu$9l>o%R1xPyRf1L}K-!DEvLon!d3DeeIOo>llI<*DDwt8HVW#
zBXP@tKj+*iV55S2GZc7XuEg}_N*JH;K0kqTBmBL^7PGc$t4Hf&dgOCPYB;qGhpY7Q
zP_NMN_dVyG7H>xDFqN4er*1N<OJeZ)h6Z!!2Rm*ZgZ*dK@aL>lE}8r4lo}axG-y6P
z2JvKWL_TMfpJ-vtnHdvX9mXBt*Ye=rhiZECb~Rwfb0fyG-@1l<jK7@ua-=V5LH9WP
zyu>qMH|}Uij6)@QhSr{lMNTF9?d^={Hip;IcmvE?w|eY1V2m|=(H}W;%5y-_7%kp*
z=ghKD_odfhD}D6c$lHKyo)bK@=+64KycK)q5ivMHogwL!8aEEcq58)JXpxL|lbF%W
z{_*VZtmUKGuO_4WZ;V0yRr=eSX%KRb`UHFB`&v_fn8GX+_U#WGQ=l8Ko6Y1cu0$f~
zLIlPpg(Jg19AC&=g%vZ4Uxq=M6o&FHSMa9#6&OlGvG{NZcE1e5tdU{3J2?y;=7(XI
zw-W2=C3r*4-y@07$5)GhMOv8BzuwR*hE}I&bT?(c<4+_$_Kw8+gfM14hQXCzqweEC
z^mGnHJlWbr9SBou#hYIRq4LN;?C1MF*aYIMJ=MOH0QmX(qxCG-uTQ80t9`Nb5VhX5
z0ch^&2k*bWXn)HWhkIT?w@^Q7@P61b+Yd(aRzu!KT9UWseh7K(i~T9|L;3k4Y_Ttv
zSM!B3%Lf}md{F4&!#c|cZYm$7lflo<_#k$p4|3;pm+4C(c~g7I1w}7eq3<DY=DADZ
zcX#Rj$3v`Nc}R!*9^&fjA+<Vs$j7fe<oTKI(%8P6G;P&YDv#(a=~p|6dP+yBk=a3d
zUurKdFWQRRS65lo(^bxwxQOj47a5+}TI%j>En~a4mLq3cO6X8$8DFotv>xmv2d6X@
z_lPEP=1e0w6x%>1ZEhfen;OWaW(`EF8%QH&No^hID8DKk<XVJ-Wbbg0^9A;jJj7lU
zyX+*frk(uzr>@McRZC`<)RcyAZRMPYE%T^s#qh*Nob7DnI(d6OshXS$vywU{R#G~w
zsw5t+BJcby<xxdtnOd{5_%FAR=7TMy<#BTfTWKx>`kPA0;7Zbt-sI)nV;k1mL;}g%
ztjM4A#{58^58v_sSUT&lDA(`nn;;@!q9Tflg|woCuG?;p-Q69>Zjap^zziiZJp)5a
zC}Ps7lA_omDk2stzUzDa-aqC72L%}(p3lAaUTf{b#$WJs9&;EIse{V+gpEx;!*zBQ
zCR$dZzWFD#Kl%aQK9#uS^B#LMD$xI21y+x!K#5TWmQE|j(Ob`Ok@}~nhnV3le}=UY
z&tOZxPs$4BNKuP?z@Z!^UCI&H_bEnG2c_*)3O)0{$7)L8H?#!f2baKO0&}7gnYHLs
zg2c8Z*!8OjWy1>L^zI=(sve@Q=R;)V<YDlpJd~N{Q8SSX4}C5!>a%feO*U$bvtdcz
zZoSVOwmv!Bjg!Nysa&*T?&7`v+4%A|i@B)WFVHIs&tfxiXkZ3n&!yw=-E=G>Z;uYi
zz>!DkxOgKSvD`5-K7!A?q~m^W8vBuH=$XZRJ(sz+%J4p}cD#of6=diK?y+O$UAwm2
z3th^6I}7e0$@LCwOYU*c5qEPJ-A2LM+o<1+zFYrfNXXkhBe)-qyv_CF&eYI&<Xgp~
zj`f1U=6DPtZ`+W!Wvk=iFo!eKhs-PIz4Il1XKycvMHG2EyfJ&bE8@}TJo6mL+n?zP
zc;1mU@-)`J&!VxTF>4+6|EwOxLVfN!2K?{AzM^nvoja6g?grQK^fUMDu(mz<D+YI;
z@E+zBgVoGel14|P>m>epa17QL@EqPW3a>Ur!m*g2t3Lb5)OOn4W<~(}VR<4F$2W(e
zph|}Yhjge3XWy9jrga%)%v3er&eK50`u6I0E!q`oarmPa;dNR}GS@->R*Ta&$-4Sv
zj-`(CS1sDy<b1c64iyp|X0p~j-6j-Y@9WScN=Ltu4p%?w@UAWla!Wn#Qg{5W6?Isw
zM^$8Pod@es$1wO#48@k$YJ9FzV{{h{s^@4?{zZc(WdE|A8hjY8!G^vXoS3h{xNr^L
zO(SdXX%SSTrEinxfy)}K{=$F%OAS?o7TpGiVH5kx6R$_$?bJw^WJY4|57xBw3>i}A
zyokI#OP!0XPLB?(OQ%rNHIuycexilxGA&d~)G+=SiXn<Hd?U}Bx@z%JtHq;nI_?%A
zb9weS!q;xz0rr%i>9A*`k~z-`)*T9Xn5*bXRdc_%hJ6wZ?j<qfz)p+B>_z{p*5kDZ
z$Cel2sAav>ttJ|I4`Ohl<#m)D;LqBNSsc{xEKj%&Q_gtHpT*$4B!)R9(WvSf1>d+B
z_*0YJhrD$qZ?9KVD|9T9I&S9LPm9FHqh!&nD14n8jX}JgCQ;`j>r3tRhbXlD!n*P(
zv!@5gVp~tnIN0;3ToMmkUXx2_a7U`1*Cly-!#xpJmlAQ3*&2Ju+joCCmpBoR`kc>(
zleZGq-IdeX>%K{?*Pb}ktK?kyG8ua&2H&U0z`Q4W#apA$dpPG8iIG@(G7<?tBJg@(
zB#x1lM>zwn<9To?dE2X!XP_%Q`&8aU+cvkbV!;3LHu+O5e*KEzKC1|PZ5e^JoC%C9
zrUrRJ1pJz^?{SVgHu`*Ac*gm@Mh`VPxGy9W@@qPHdurK3(BO6eU;k+uT)eJE?JhN}
zVpMSGtwNfS3V%zK7<gKN6lOc!{UF2Y+X_UTW1qR32Cq3+dG~~RE1p*m52eS5*=WrJ
z!!e{U&q?d4Q>e$@4SDMi9Ws4X7@MJlxr`YWd2&oALoYf9(~lgC!2Q8!{~{R9qeF1j
zCIp^SgQ?>W;_izejFC|Pbod(FssfN27Jzya0#Gm@0B8R5L-Uq?hz{_@EBK(`v>)s}
z{m`=yy|m=*dxvXqZ4&^iT>)_YF90d@`{eWegmixx4fVsfI6s`V^hfuu{y0zGjv;T=
zng00Y>Cb#Ae;i!wCeF`v6Q1kbgs!Q(n1A106p^<Z(>=uR&I3fW*8nkBKR_%M1H?W9
z4`JBaUGx~~#*7U5e5SaF{~GrdgZ-REP!}ihRpTfwdOL`HZ|%h>oxNB$+*U})TaN|3
z#qeYs@p!3?Q19s_=Bay%^CNnQfPGzs<aQUK>DxsZ6k7?G>Q3T%sHLbOYoqsA3a5IO
zV*0ZV!uoUvac@)y@rS&<O5V;TZ%x}<h~E?1i`}!##ZFUmku%Oryznp+<MT~L8%tC1
ztCy*m)x%Wu9A`qkRx>fDzOiUAqp2{cGZM)gjKuzBO+=H8O~mqRhGJ5cff!2Oy8PQ%
ztQ3vK*Yl0UN%Gc>yxqe3cIDrCqTT1eP(JvJX!5p-yxl_Hz9VlpHTr_{<gFWdYfs)j
z&He;qJI-wDSHbn~N6cgYSZDDbJq+GM^|1ms&r**(lo`o)%8@&y9A+up-IMnetA;;A
z<&bBb6F$X|{m*cvstk>&KhCx-V}@ukvi=moA*m3nin$*ko%7lF5^U*Lf=O=NK`TlS
zpe@1ajU||JuNe0ei_p@d5PKgz#Kr9oVd0byWquyC>+<m4BoEhPbD?OFi#a9KHmzk{
zTbG5#;aT_`$K13|*_hKc8%x}|PvBTK=-@*jKfcV28K_;d;7{HzCvOcs)3Mx#8H{IA
zQBRkO)xqg_OU@n)Ooze1beMNeN64l$q%}&z<$kG9k+*&|+)uZH`4w*T{f%dKV*5Kt
z{NMha8_6)b&OF+Vcd)A8|8g6-!>oX;C2#*Wy^Wzsw=kDpfmt=o)aypA@p{gv*gtzl
zhR&mxqk2O;T5!J9h}zTD{pd|F<9()ABv!Q1!{K@iesj+1<fw-iXM6*!BJuqk>)LzF
zy=Q&giJFt$^uZo2j)tXC47yEVHpF3SH_p(v!}{kUH74hnk>Db(LtjguALmK4SS$SH
znR{{wa}f{5pyxztWa~xZGPSioTSTDkAoft7Mc`GlNZ2vYA?3k!93%hrk0LR`Dhx~N
z(+g3kMrMW%(+}!kt<lnZrQvH!UOCe<LEVQV>qvQt29H*;H}_79F~78!)<%aJoat)G
zvn=v<8F||%PK!U}o&7hm*<Ht7Fri49twZ+<I_!w#`Tvy;9a%%~WzA?kN{bhrwRqK1
zk8h)N?3HT}Ia|$|P=gv54Gun6V{APQqF9RrQ`=I}S&g+5HJHC$gM~tam(J9s@GOu-
zoy&MPE#{?gmRqGoD*5ln&($zTjjcb_=pM!%IQ3u2>!`V6U7hlP{pMuWKZ#-3^EC_(
zWd6{FVMv<6`n*DqUSw{3YwE1Ws*pHUfi2N;&JES*<DlWbH4Tg#Yf#x=i)8^?EayCX
z32Xf=Z?t&5o4T_9-TMdVu*5=%)6C|Zk}ij&Qh^8!>z*4L<Rr2lKcPn82Q|VTtFT6@
z!TtF<oa21B*guLH($Scd!2FPr%u4WJzxN}5_W!acqTuYaE*4EQ=$+#E>EcIvjvZLv
zSw|p*b=`#x?0t`lK;Td2Z`}$<z9n_#JXiES5DC8}QApF0cNNS7Ns7jqmh5-rM&ZD*
zC|K`^p+_{9S?1IO{JxGv>ZeS}+x4tP*YF%La!NQoGt9m^kO;rHMD8!wV>|2F-aDBy
zLFNWo#v`j&95W!|uq%?<=n-)^$=x?`7O^mKh{27|<opWej1)#8g1-w*QzH?3g1TYP
z2n=T5HZzcCot@Ma8<M+85$Lcp0)}Mi>Nhtr<VO-~W9DwKzV#(<SChB<`8#v=c{srD
z?dfiwnH(Z8lRXo^OdT$}gkt79&gs|c@kSAf1?27594#V;^XJcV+D0?ZZ+T7K<#qjQ
zE6;H?DlGb}!1qoHOrn<QT&Wy0kIV2fFA!(W2O^$+*6jKr80O15^mYjJkL8HAP{5#<
z0tG!4INM&qTmuD+ZzwRlp%NeGs?b%U!h&!aK9aKoXI?{eziW7w7XW!#0C&O!Qui2)
zgdriw?-hc5lY+6nk{M9Ui<%s8jeDD~VM0Ox%;E#!Gd2MAZ*bn)+aGJXk&ml=(Z$6F
z_sQE4<ZWz}A0*yns$l?@H4VVL_x>38(jQ;;2Vj2R0CeaY0MkK!@QLw5!{dIqN9I~A
z_eUGnwBJ(MW9D<;4*FsIA~#`P-%V7kbrW64+ZB&o#nw_+5tis7IuG>_(~I3jeu}#Y
z80RkTM7W7p(_F;@_kQBT<bLAJSQk<Fu#cz<b{3OdoWuw6Hp9z7Ja}a<s#Lb3+lSu5
zdTDQAdCNvDTxcUgN7;zNExp7AdTp2Vv=)aBb``;K+#z4fyhf+a;&o9cp;B9lZ(A(I
z6jw_zqqKua4(K3O3}#kx&ko|<-}b`n-v9RPwG(CBD__>Gt>`e*T+~Z$Bfh<BE!u1~
z71vsuiuBPY;{FHjBM50G%ttm8)nUfs8#T#I$lJ!`?ZQ<?;`ibvB1~^6{;n_-@wEn`
zy19X{TGdz_?%P;6&uJu@l+_cGCVvr|P>1rpb?6=W7wxnDAhGv%y!c*)UAa{lom_>h
z>sZUq{D_=Jl~~oc66accz!w{8qnzorUHXxkUhE$`y=Tv<0$onM!P#yvAP;zs9nZ^|
zD_D-U@lWBO`4o}vPodM7AvmuL^Orxv?hnjgoKl7sEy}ROqZISL6{7K}LTvp~2*i{?
zA6J4UHYG50F5w=s64>o3f!eAB*5s{CB6kPGk+XLmVu<HM-0GQ+0V#Q$D>4J>Gk4*!
zmu%W97v0I*#{XnvWMvk{`(@#$CJX;G&&I@Z?!V-&ydYc7V)HVw(T%(B{>#MNSLv`?
zo`$|RQW335h0ZgTdeKyrXj5V8la4T1I<5qy<Msg7wpQtAJv$9ALsQYT8S~<BA3sg*
zLHu_Yy<gpC77nxBTi${C6n6@cw-xW{>8A(4lDu6pg?VlIcz8y|qh8w^nC;J9Y_D!$
z)(!4$wV@YpHtW>bI6Qd4zFMnz{7e01H@E9}U>3`rZam}jp4TEI49oV0;a64+rgAoP
z-$9RI)TV}QkHB>9)yf?ojc8Rgu2A1{Rm5O=9JLrXnMrd$8V3iFv8+!%vqtE5iT;t^
z(OBTY%!35hYNI02j9iah%B<o&%x9&yWl&566yGCoYB%r2|DU&ypGLrrdfS=7?B{OP
zVev{GhFGvRQ*h_zQue<dv%glSMnm#sc|Q&6Z`9ytkQ(wnd|t^s^;{)pxT{cFsYbII
z8f;;Y?m{aacVlZ2y;2Jo>Xp`cahA$QwM>gE<gogx7FS<tVbqlx=kw&PmU<@EzB?;)
zxWk(9DtX(Aygm9sgHxUw*bQK>j~trcL5<<AY8Wn7B4e@=pI0d{xK@b{MrxR_*4=2T
zVV<1^g$>zTp2b;QS1kq(X5aTdzb^T=o4mE=|39Cf!+F07JyvV+qd<onYr+s25r%!_
z?S@;_HIcV%;=^DyDil9QgyO;kJ@nLAb^V|~ZIJ@0%amyIkAl5fYDZhC;bX7CT=IXk
zzXElgl~|Fh!PYif>c6yD;Lq2C`Wq?u`kZCHe5MR7Stn_um?wW#!Rwm))(+5%*+>Pu
z9V*<3QX`c2=2d4jI8B|`UG~m9Mnqt;T{Ns>V=#CSGcSL#SJ5mM@3&sZ;{fJf{fR|p
z0{zA5p-9Z8U+YyUV%RHJ&k4tE{%mdOuPS=azAER>j{>MUJE@0!S0p6Mq7X>l-Yums
zSs8)vei3*$Fbd-jMq%OB7#w4Nx`NvFL`U|hBcrhKHv8+r5pZ=0$G^<byg!{jYR(G=
zj?l9%WKX^~uU$zP&VHeutqZ->A<W7wyN<b6sg-8$C7$}dKh$IGwT;0x@^*JEIh)LS
zn%wV4J^1h)5$NR3nfAvB{9YM}?({|3)JEVXb>8m$SZnT|n`IXTZx3GA)L;*79sz@@
zaC~}A9r}ZCv{_DFUPc(+YnUB!F&vKzb(qcG!Zr39PTuDANDodBuj?y3_dWQJy%+wR
z$`)ww=^)QIjd<2Mrh-umC1R@;xK~cCutJU{{p5I(Btza28MC}X;58!%gUQ?d<n7Cj
zAqX8CgtX!ybbJv6zn3BCt(4=3sRHfCDR?GVpz}X6I4EV%k+oNk(Q^~ZT65?%oa}oI
z_oC^4{X+fHV)|s0**_i}g5QgRv3g?=OqVh*ig}D1FI>ag?$<CTIRFh-1>gvE#Y-Z|
zTZtcRCi!8=7$5xO=mSY>DcYM!VeT%)zvOMlM1Q0;3BV=Rt46K;QJ(IH{n37CH`ouJ
z16T(a`Xc8TncLPMy_Qg?yw4xo$=lAI{ZSI>3*)(N;&?qbF?pw}@EzwW{>^t4&o!>%
zM47t?vhWb^i`+#c^0rA|YLQ>{6}K+-6;YLaMbN~4qH(vrqOQ<cIA3uRv4<Q)<FWQ)
zyp_FJ8*V48l59o)vfg6<lHOuMI<uti^$@cnx{HS6x(RcuZsJ@IYw?ivZAxqxapZev
z5i+2&*q+x(%=EGp9mw0Y{Vhc;a~qwhjrzH#y|{1CUgR%o%k1RVV%w#5!iT(lMBaX#
zW-iW1+lX@VHg$`snA*xz%pPST_Ek0$DZ$Oegi+1JI*GApo!nHMFEJ7h4o1S4*~oPf
zhGLT^cM>c!6e+(A#6<G;@u<ea<V+)>xz|t(++9z+@%)Qn@pZWO;5VLs|ALnrzToue
zY8)u4VlF}zTxM0Fx2Qs|NgvVl-g~_2^Bx<UzQ^ZI>=!SsK<Sc?7`2`{sdp84F!C*a
zoq2(U)-Ui<{~R9spJ3XqQk>F1#V?Pi>|JqpkE9HKIb~2@D#N;O%w}{dMHu}${X&Y6
zbfXXfYYSm2D}=+?Vq8fmMqhetGo4D1+@%BuYYJf7`Z3DxK1A>N4`Kf<AAiFiq7!+0
zwtGGXJ;;NVM;`DY7w1BAamykXLyL0IWKuR-z0HE#jx5f~vM|k-wJv$v;cFINv(CMt
z%tS7E+vGXtwT|hyx-bnHcT+Jy!hM+sQn7P$Dn@Q&oqIf;`)Sj0YI{0PFHL2Bb1JU>
zNW~3(Dr%^Ix=sDlOY(O0zPmVO&RyB!HiCcNX0G0CNT`c6$WB5p{Vf<iNy38T%;LA9
zjx&%Nkg=>^`6zRyr_h*wLu!7%^`^%#Dh>xqxZ}o*c|QER8kg%R;(Tn#a_)xX`F<|7
zt-Chx9_14T%Y*c{(!UuyAPjeTo_`Y?fdJ+g8Gm4gkwqk)|Bi-9cr<1oh(;8>5-(U2
zJmieA^Xe$fx)_B`ydU>dk)Q01_U^z8CFT^G=0_redf<OqD^3ihCx;9#j^jSe{BUT7
z(Cfb>0-yKMv#^78<a!;wVf6cR?ihNM_2?hglmRLn=9&K*btHQ{G|(>7pi6)nLkFqx
zZny$nLX_CLRfYE()j05v1{bbqV8t4E<{E0KR%nqlhW+Jv)CjY8H-Ps#_W~`BUD2Wo
zd29NEJ?BO`EF}vas9#C6;{9<e=fKa{A7;JxQlr5l_J2*L^XrrEZ!FaKk9yIE3zT?0
zL5X<swxWW(HBjT;G&Rl~Qln?68YyI72-&ag@_)R2c~=YN3oZJaYa!xQ__A7s7-~<~
zgwf}CPL8Ih<+!SlVcb#~Vn)fZ^=t@Q(*Jn*I?oa+4W>kLe_5RjR;+dQyprS8J~{s5
z9<v_&w|n->v8JOO+nl&F&6jyWnF^e2q{O~GO5{{);49H$ewG@yhcTn?h@6?pGB}pV
zFnyRD@fQ?GByUgeQep>ro8+Xz%a2O9UR5#Mm~}FFdthrAj)X_RtxGf#m;>#^>E-2b
z(a>-XFf$_@t#iX+e30j{V__&xrIw7mjV%v_R|odDAL+54yzNNdo+6Xpo)5=+^0xS}
z9!KV~U!KKmD;?+7TKZ(@Kkoi7&s{s2X~KSVa{EYZzQ{ZXLw@Z>tR1OY&m`}5g+#!Y
z^Y56*FerIWY}{Or$;0WVTFl=6?lAUb!cbTphJF|LeDfH{^Lai9rJjiW?W+FN>Cux^
z&?^QWyq2H+jKZ;7JSRHxEbu~)e=dgcJf%kgdAp%+Bz~3CPZbt{)Dx`FnVoWPEc?W~
zZd>wuUh<f4S+l2oqSuUdZJ)d0SV`X2WKpmFoo6?9>hbb*@R>sWUV9yNf?BL#pFf4?
zzK-m>SnbrpKTCstyk_lq-FW;_!RLb#&zmc;`;!7g^A+6BKz1@mvFer#hHr!EfeJ>@
zEao^F1>#6i0CxZOM`}IhM?BzL7wUUk`D4TkKm6M03#-FEaK0f$_w!QB8Rd^A)B+zR
zZ;NL6ptaVAJs)3Oxa5cY!2$4{Ouy`{U^E^Tgqx~hyf_tv0V{)W*EA4+k6pu#&eu?J
zlUk?c0qE=zfCZQRF^Zqxb)^ryNBh9Pw+}iSOELF}1Us%vVCXA>#U2Sxk++9@rKqp)
zrrypEyU5$t<n6)pKHNX%i{-3|{oDA%dbU3rZTE-$A%AGB{BbJ67j|>pMDyRSV*4&v
z@zmK>Om5I$=r+3wEAqDXh_e{D%}M+|<|NvhJBn3R4#LgDQQRHkAoMoQ!qugZkYqWF
z$wllX4{{Ly46_%1E$qc^?j*RPuoW>eHe&Z^8*$W#du<DQi00#a2=%CLV%=hEvAmbH
zxa;0k%(~J=^ygjz7jzbLc3X)j2Q9_6-xgx;G7E9BhlQYPQ(WEJUOai%PK>c^C+2K!
zEvD8q7e<$u)40-17|$>heTJI}SGB3wv&K}^H#HR>$C!vp@^<5&W?~-e+s@o8zon<K
zX#Bv4y9SKJx*3LI81tm2Zf`6!^#0hsX(XzcB{dq2L{?@4QB_c1<m{*?UQ~RC`}prT
z@wo;jwO`P9-4{HcUyU!9KcIKuJ3M;xmKoshuq>tmfhXVN_tN)RR9%6$9V?L4s)9RK
zEAXb#dn|rjfgeNPBE<9!b$!p!r<ya^Q>FOeR*Hg8r3fmbhc>+otG&wbEUOGV&X*zP
z>l1ojO7XKBbD)$($PFpP6%?XsA^XJgN67E|5N6CU-f90B7R>oJ(ou{2kozK8+eS=#
z2vMGodF%2KW|)ss^L)te=OK+b756f8F@F>1w{3E%Bgw(Bk6HNmA`2S>xJP<;Ccbsc
zg5~o}7>s9qo05fOc_v!*&qSZn47j%BPTIUw=DU%%M^X{7JQaI3rDEXtRP4@Afx)~K
zeEjtQjegw4w&tmr-0MDa58Xx-^Xz7t-$ovL!)xrB=jxHf`_wIXGUG1$@hwDix7j&q
z66`lH!#yMsXDZp}?LqAm;<1l<CbM?TB4{6voTl*@WI%6iLuTllr2dID?uHa<pt>=?
zb{zX=e&HBMElkWEJrva9Hd{-+4rNU+Bn%^5!*Js&eWVjYafy1M_p?G#Ku>4JFU~Lr
zMWO5x>m1W){A$Du-W$wR=Zx`pg9tpI6^V88B2grZfTM#Bn=9Ez<*YP^J;~Exo<e09
z2I=(hpwA}Mll{@n%xLHohB!;!Gq*EGN3KFI)`~WAC2Gzo5%^lkELIif_^a^fnTk6G
zln~@YJiQTRKJ-j*f63s%3eKqH%rKHeuy3YYs9`>o2E(ac9>y7Jg}Vk$<H=|8_WUMt
zlnfu6p+yO6-X-ki9!=9Sk5b1ybDV=_DG_O_g9~|k?kRa0%UNu`3h!2_5KrGkQezc9
zvnFLxjYi~c_fbmh>Z`<g&eyg!RB>l6d1j_Y8GVh0tjUAaYBV^hLE3W-62@_c`;@$O
z*J8vS1x`um@!<EiIfwp>|K#|4N)9ti8KlhZnffLKu{%R>H$H^*kql{nWT>`~V=S|N
z!udAUk^4gT%JGlC9IwgGm3)6AGl1me?UFzROfwYN)>w%LN0m6cjsMS83Z(xlM{k`B
zS1!mfmHS&BK9-@A19{y`j$Q}k7|~RLXImBUp2#|v_xzlFD%@io{)heX`|NWbXsN;1
z(;6t=Mq<rLW^c`l!gRwZ*0bR_70N#GSDvFdhjw+KZ*-v=PYg6DZ_c_HdRVbG-@J^O
zBJ>{_?PSfT)T8Mk=6%r9G<i64^8eRsR`YE|I6@XjU>bY3(x;Jl**20H{}C8Vo=s(4
zoJKFx&(`7SDyTVXqsMdh102cQ4u-6k&6zniS&u)|R?j*chHC1^2a)r7&SzujfvwiW
za9<O9#jIJAc@Av2j&-P0IGTQA@7p;XhhOR8L*A+`gu%mJkFQ145uXi59cQDzdEQ*Q
zoV7RW<<s<2$LH~ULEf73?6B>V4!d;hS4?N!O1;1c_V15((!rePlpW5YFwfRupS2Dq
z?AJ^y(PD#&^Lp|&W;Hd($x37|Q=y2j)i3S_Xl%tEGOzimX6#S=mNQ#ifrCEONxk9B
zHkq?q?u7O25sbLmK}cnu#2#vw>}&n;k&H_A_2bSHKLlowq1$~CTj_(@`BDT0NTEF^
z#bJLb1{(NaDCewS$=g`95Bj(B#r0#ph@S5Y6?Y3b&j`f5<Uoww9f;7U%z4UWwjyiV
z5m#ArdilZdiXS4M`@y$E0CpPs;pPq>bRXh_Xd@|_x=B&iREp&#684%UXh7cH+bZEc
zd2hI{_Qp$@H*$7Mm^&+_2SSP=t9<ahl@F|2`r!R(DYJ8>*c>WBzflsL8zbRdgDh_2
zChDj)KEKyhH0;q|oUPwqJelDl{2cp;__xl&N;rw`#~j6D@^-{q2NAZ=L2MZ8AX>RQ
zi202jg|(rR&`op{DejJ<vWvY~A8#i-!fnN_@ZMs`W*gz{WFwY$?Ji!_brTJ?cN4XS
z-Nc85)?#!gYf=5Rt2nf_i%`fqi|V^p!tAJ(xVOquoGG^u-4;`a%)LL2$y>)Q?S=El
zcEZoUh4^XFLL52RT!fxC6P;I?i3a4Y<1jN}tuz(6+((dQU@9Van}`dRCc>iuwaFKn
ziSM>#FM0cEwUJ08Zwnn*-?E3iFwa1^&u%C>TX3)5(+1)P8j9OH8i=;>^+kbuJ#qZV
zH#nc;KKR@kwCw%`UF@r|A@u`Z?5xB!={sDpdW&V--{8oR3fNvMM=z6dgg)dB0rGV0
z`6n2@sR9FPD{$Vk0+##U!l%tkRBn3)JAS+YdHcE-y*$aqc+;l@@wzf>yi$hqDP`Ey
zv<$5|qn-A#7>TLH7^^D6u7E=5oeI%!06nv>@^PxcLpXiUN2M+wx1{;>sOH1`Vm|ti
z!Asgc#F+#6(9g<8W+n3%m*?Sccj}Bwb8$H=7i(;Dv1vvQT6^Rm?p+qHGP7c@Booz5
zGog4!FYR6Kwsg;e33<Dl&rR!>iPMiW@H{FFrA?S`x+4`cxtD9!vQ!v4roz%9l{)wo
zgl$g23-VUE;y$`Ft7cj;ckmpz4K;cDU{?~_j^?gf+a#=aPeO@2iMrS%$ck>kSTJXH
z6L*??Wab3@JPZBlRi%d}J?=XGoaJt<4%e9p$K7nyD*b#Li+O9~aQt{2XHhZ8<$bcH
zEwhF_nD5K1!$nr%sF<q9EZ&<+=Y)}`p*U|t?T%d-W(S1AW;{JI<n8f^p*TX1NeX#;
zsT=cZzJ{ZK`GIG6|GaTlk5VuCL60*xDM63c7uZwf{pPg24xK8r7-_7-A`j-kI)}k?
zTNrZ4^?hD?to$Eu3+Vkk7Rp=$YL3pbr|G9cOZpW`sORt^Zw+}rDj?%do>bu{AEU_I
zmb(?WxL*Nj6g~PDoMHA=p!}%}{^aezF06x>sBr46n%S}%+;P<4E_pkBv<4kmr^?v7
z9dTHT^gu1#hHCMZHSVNj>YbOV@szXR=_yKl=l-#bwkizYrh!pUHG1T-$GT93rK}0V
z8!FI-^=REOC5mS$5#gc4{`N}bpHd?8ixRb36+Eg{P=~1Dz(-5=ylq-(V4ug=gf;dx
zSI*nI$uK`gf$^j1$GE_nc9$HV&dYF%*?6&+nIDkLzeC<SF69g@IRv_OGM*J>SV;Zo
zc4qwO+RJgBye;V=#}Kmi&Kx;5vp>9o9KK_$z}G1X=-DIwQ>K9JerjXZ@<HBiS}jLM
zU+#9P&wXl9AxPWG`}Z>$l3K8CCU0YE<#=&Yf$i<cTfWZ6JFD>EvJ!JwDv@ZyGmbH}
zIppmQ^0xKiC~PKQ-tsl;T&2Pe8zn;d^PS$8SsTr%-Qt{kI_KQ;%XL`c6~=zBp1Ih1
zxbu3MNM35GHQ%>WkNwR<shtYL-cdYnaXuZshx2L9iv7<;VCO4l{C*_M%XvQI^L}CU
zK9jdY+c5WIuAcdjVdzWV+HI$nj`?1z$ln>{Z9DQ-aV89-FVh!YLX8k-{&n%Zet$8?
z#61%J<n5p4kq8T7t`>Ql*Miq#7ab;ip*Pi-9^`+*(1rT#4Ox23IT?;k><iDo8i5|<
z?WH#Kr}29!X-Z9b4mEe1bli`w#T@dshIQ(JLt5?<(_uVk8L106_ommZ^$q&5+v(7D
zk`_}(Xko^h){K3YqkLW4YLr;KM+J*KHCplQy=@A5g5{_=Du>fEo`d2PD3T~3TPZ{6
zb9!?w(*J73-LaA&%;^&dBV_=0DX|`8?YdCPnz6MXj+3_@M|^Q|q%W*<r5Hfojz1&C
zhuu>2t>=UAem=NG-rkh>Ah5s(izfNv2zk5YhYxdSq&UKQxPCWZOjzlQE1P{Wc#kiN
zfB9mUsUNyr_QOT;c2)ZTTr}{*;^tC_`cecmks^T%T|(Y2E0o}VjW<@7d86WyH}@5L
z^Y_jhZ3cKl9P!4Jvl8fsND;=GcL#Z!Wa@)No&`$D+w0_Q>tPbKvz5TFlLReVxrsA(
z`w7bl{lwT`E@I0*7m;|XuaNZUBUV&6i>pr@#j}Hs!qCA{jLdToHXR&9sE32-;p`wb
zSJ;awCXQlyA4lQiM)vNv6U(b@MPKr^&F<df+D02O)zL;Q`~SQRi0&qaf3_A!=U5BR
zj@IH!bypF&n%Ppntwf(&R^lLe`*EqIc>L5txKn?eliW@$8P{GcnbBU{A#WQ>TZs4M
zZOvyhAwOv*%-WiYolQ-}r5`3@aDb_>a5fcd8k>syJ4{3pdHeW)iP&M(RP0MO68oe^
z;uYt%pUK;c22I2ZYLc67Xe^G-YAAH=8;Vaw4aBB;zn~oc6S>`fFk|)`5_Wt;*^;kF
zyIF&7<ZTb~_9S`hOWt0(@D59Uy+QApZ}9$51sYx}M}EpPoXUTS&!0<?JE9aB%%3=M
z^a&)do?yY=GOWsff?0E!1LatXmTgKgq$|1X#2q>%C3rTLex8eE_|uR3@@|%5-`x@{
zDlEpe?Zr5&;ZE8Kg&4ZF0R5Kc<KdiqbfHJ5)Av059+%H7@O+q^$;YQv`4|(Ji(J;R
zBbwzSVOl=2rsUz1Ssv^f<>3i=TVIunJ-u@AWO)v?D_N*B$VAz<ObpzY3A^tZxL2Hs
zYjK%y;l8}`k_@0xCU<XV&@-7q|2lVL8m3{+id5v8q#}M%D)U`Zv1-756x#nUZ;^R$
zOOjzTIvI7`kyTWfgbkkDHM=Yc<M{Y9HVJoH^Uv1Y5lg;~u;KeV>FqB_#OefkYfBQ~
zyOw#xtGT;{8lx>i%o<pC9bKAITk`FH`^;l;=w&Rt&BD1KEgUcDZJ2i=3M%rp$Ds(k
zXP%#CHM0)swa*NtXJ96^jD?|Saw`=7C_{15pLq$xLowDg6i(!A3;GJIIa@qW|9sff
zaE$pDj%f1rWLGk@n9SsTaM*2Xu6R#2qQ13j1I|NJwJ<!PMf`8}YU!itG&>Xzo9XbA
z9*0p^S>w`sV&)mjy`O5VTgRGlpc1kXO6FKA@rwLSrk1C#Oo6{M$iM$o=uQsE>4|7(
z7lM_MDm2bx{dS)j>*+FdA$L}@PCP{3mQ7ScO)XLdHAKt#Xk$;^C$+a4>ZvXt)S?CF
zyTOCB=zWR2<@=M!+a*KQm>i@;z3;54yQ@&mnc9u+YVJv54N4!P)|xejRE9JE%WHow
zW1TC<{!t3#(^KPeg!j%I1-u+o*i@*(x)K!{8mn0|t0DOugqO^g&FAivn=T<39wWn+
z$3ZB)BF8QEg4NyS_;Q+`m%JUjKLocum@P1h*#e6>M`IRz{nawWImpoHiVWNCGTVoJ
z#W7hj6tjQ4bDJENgQ=t8a}lp)?CZ*qz`FG~&jsZ?lN4si(eobr&E4c^JCc1#*0#g`
zh9Dp)1afBJ9`Ki8d6f*`lKJ<t<mg_;GYhkYG_0|D^Jf%A-i{*|ezNxcafyG292hxW
zfycX*c<Ro+xuY6iKk<A1SBdL&WW`h^EZHj$AE(CmX58o2i?28JKm9j`polu0Tn)8f
zkM(dFtHoy4{N=x?n>AtYc~vO;CG3B&&vKJz!1}yChu!CXi6h}Sl&DA8d}`LH7jO3@
z44Zjod;WtSF7mcU5r%i`!;r?aVj{2Ks@e2e@#8be+vA*Hq(yQzzgLGhk)e3^gSF^n
z*5M<lo#XkhwyzG47gEpum*=z2I!xqskZ42ySXL<FI7{!#THS;>D342-UB=Jxs0si6
zMQZUf*%Ka2eH-<AIpnPsdE1%1eV?I09Cd1^IIH($R$}QavfhekR@S*&dFGloL5qpx
zZBZS2Fc($GUZI2u&s`OI1qR(zAdhVAIaPt(5;^?%Iv=?q$37D|t{j%ZkGvf=Cj|8r
zA&BChy7gNEv29i$j`=Y2P|#Du*{Hs!A3Epy!Y#=c^TU0i$&%n}rWBFnZRjj1wk?pN
z#Yrj3`}m*%YtmO<K6n!3gTG6Ba3+KGrA7(|g%m4drPO9fVev%DS&t8<a5lRq#Rp!(
z7d8ET@LJ=8pArd1lcW8OrRe!bf*VI9Fxw_UZH@%jQY9!PZ^x0hFXO#od(s>I_ITr>
zn>Riz@xmt;Z|Eu|aJ@`5!DcBYO!G#|p5EwI;DyntUMO>uAj(97iOnT=(9%ulleoKQ
zcs~*U(?x8W>msfl>MK?~?<01#aTfoSIErrP9Yt9yN3r6CgV^@hUd$`C7abfOMCD_9
zF*ws+^f_WD(p~JBduS)Fza?v{dW(Q`8*ys2jrd?=Beur%6k>EwF^GF^XI5H^7t^f8
z4f6JIbXU=WzT3v%$lJZ0#Pe!PQNF-Z#Jsc+i!<7b`U&mCjb`S;uV)*PnbK0!CAJWU
zqnZobFJ@w4kg4czZYo?^&z>P~9lx81GZv=8r_My|eq<ts2Ahb+nr33>bYn3s$w-)8
zH4<~C8i^C;M&e^}6Jh?|K*WtU5FbW160Uz5ijz-kG1}@6vh051N7o;Sob?Uk#5e3)
z^cDY+w=F+>VHQj^zN#zHe|sfPT36yk(HkfyzQN}{FYvwBbF`Sv%%@t`qg#t%*Md7{
zf0kmxnJ0*S`2?*{hV9u;(3rekOx`|fRf+)1Qkb*uTvb$p?_)}_@kkkJoS(p%ynXnj
z1YI9<hx4{#xIHaGYwnmlw&5{)EzF0{GwO_A=V5G0E^h85H<PlV(PtxIP7X@;v3_-9
z=He@U-qHDRp3HsC<ZT1;_S~afY?SAsq){%c2W8`yJ`*{AGSDs{12OsxT#;sA^7;&{
z+L*!JFd3Li-fHSH@atO!-uKQxTs&FZI1PU6Bk!zA!KG0t$ozdD348A0I(a*BSu)zs
z<K9ndjYUEd)cPcJS;2WMS=+JvCOm3x;<E{}*Bd3FKH0mpQxYbVx8Z5bDE4NS%oAqE
z?~22gnpjvcW8jT1J+Hgze>LL%n6I(8_8}H#&*-nI)?<)4eSdto4rWd-`*IIfL|_nk
z``DJbGl8MZHw=ZwguYGkwmgX$N#t##i=lWrEEN9K?r!1SaTsgE6YQ}r^d^g`k98m~
zw`^io12x1eskNMAq=RW4wMr8;X#0k9Oe^YjTT|oIgxcB3oPqwK{&hwuzPPENVGWz{
zM~QRKct7X-bL1c;PAymBhN}{92Jra}N^IN5+R>QxZ2;$sA7yw!t%t)Q8BR|N!B+Ye
z-2Mb(D(^uPr6HWn$?$%*9QDcDADwyM=RN-CS87k}G$^%YCSFes)~MMVU97?1*{nlm
zYVg*PGfmzD^{jDggEZJaK*K&VcaKpQDj7wV&<Cj`XKyqn1Nk+d?FzwaRS240q|PW?
zhEL>LQ519A_mZFE<@lILZ_QxVDg`QBdBncpM-?`ZxA`*z@s+vlEw=@s`JNz*N@OoE
zpL;@I@@;xB3?|61t0VJh$=eRQL$H*)M8B{OPF@&-T7Hk6m@_b>JsHm$xQ3b;ce3>m
zxtkalgf&lsFhd=TNh?F(94EuY9CDZcF9-5g#XP>7seC=kxYK2j9B<w-$AG+jSsQ|F
zCqi)jJ{d@s7xCY?k=5_nv+TcH&fRWu<e4ep$KFCpOV+)e<#2pUwqKMphnKx;o)LOX
zmg58YJv>_uAJ)Dv2GPgM-si%>O7x9Y!QE1g6K~Zx#MiKnnMJqfh9DFCS-nuA&lMF;
zyl0P^HGi%Xe+CW6!qvR)YC^GRN*G$a*5hCHlpnCiym=UVD6x7>naAESdE1P<HQ31>
z3o{SzwWhADE%zWT3&T?Kb_;pih>vo9Yz*tlC--#Rv#Z1VojUyLz!^iC3g0GDgEyS|
z0P=PpX9^wI!`0QQpgT;RcmjJMoPAuZ<-Bz|wQA&TvMuM)gXmrD!K{cVEj%u0nS(_}
zJB4CW82iT58q~SzP^{HpbD##-$=fM{eT>aoY@en><zDsxwla^Ufd-SCsd1^ViaKka
zO?D`$>*ICLtO=iooQ1e5@MWX|=F~Sex-5s+PM)D+<kT85&zQXZRu}^7K0!F33q+Tc
zAS95tHS+>7e<}697WB@Lw{r5fSq^<Uw|vo-ycOi_BJ$RkywyyV!j`<fE8`rOyqz(C
z%p`BM-hBV36g$b=rNL74lS|Q#ylqO}K2Mdxl(pw5&Sqyk_d#I?Uq~eE84s7zyGO1j
zdn4ef1ZA%z>{&{1V2cEMLL`{`K!WI--sr-Q2a>mkH+bXV2yfII<^_F`Cyw;<hUquX
zcR5o&v0jQ?@^(}=Z#=r@1y}Mmv!gc_zvZK;1SRhsg-CN04H6uMTbLu6<s^o?JBd{u
zPU1d$$s<!8MOX6HzNMoWMc#h=V=pf6uovA~+h#wq7m_3PB5s$Rh$3%Sy|fjVgKWiW
z^0qa3+kcgfC?;=RD|-rW<|P-Tb{9Y2Sc?VZ?F91n9C<s+ql*Zn7wCZKB%V}R3WHIW
z;?f)DM%`*JCai2HoDI#zN%A(8yw%6F5Dmhbi`r^4u{Y3E^y_IN{?s%RzrQvUe=JPI
znC7OU;14sBADW0^J|<$`Z(}jM%vdC)8i@y&jretq#1Zm#L{Sql{s~!2-rkR6R@C30
zXt1^xcP7{3OWqG$IPe|jJ-)$gAel?v{;Q}#v!P$`_);~JL%DNcP9?68x4Zq`Aoula
zBm^=4>0&vm=9FUSmty={Q;gTtFjZ$3L-Vv4b6>JX<@4>om*76xs{dSqm3}4ImRNx0
zUGk{U;Z9j*Cz|eNmW8H(Oe#eF?IM(?6(fx7ZuqPSwe9GsUH=#j$<WA!dB{GV3uXUY
zd?IhdZe_zjK@ZQ^930t|i$Iq=yyWM6jeNL`&4Vj>JM&B~s{N=%_Q}SvhPl|Vg#YF^
z`^Pohm3J@$PRBCP4H?W*&0xkv1{&N+$KJAZ_(f;n{>KOClJfxV7N&65T`IPdw--L9
za3-CCvKRLu%exCvc^j8m*KY2_eR=C|@$8z2h(SpxUe3qTB%CR`iO!#IBGQm~j17|T
z?e9(e7;+QC$lH0zi5Pq)0f&m?;k}J}beJ(Px)=A=*)sz@jk=P^2w0P|C&}Bd%qB?A
zVQx_;>NZ<(U(O(A&1l21b1?IB8!%(&9J7^pulveb?aum~-DZ%r`cP^cn3Zsv_rW%y
z$l@7(1bI7hDQnb~)T^#h<L`1cu6_$eTk`f(%P_2F?|Ek>eKxFFO6{m8enu}odE17%
z*}TSTWPexV6YIz18qQ>IDY?U!`b+Y5aG4U-SCqH}CDtudVuZ62zt$_D$WmZMwgTBD
z)K~7~?oHlzI#6HPk-T-7%G!@}tn)vDVNTw<khj-uWtbR1O{Hfr^PG5Zzs~xW{?-ZP
z?SIrYccXS`XMhIgOE@Q^X1I*Gv<r>3xW}35N!~*vd9Uo|#2SU{|M`V=9eF#iMvYa>
zulw*jn7jCx0kDZ#vLPXu&%Rzmve3ew{a@Cr2j<DJbdwBf>V$Kt9UXNz7>A@m)URE`
z<c-{+ok0ITc{}(Q?~x0F@R6Kcm=;8@V-OZS2uA#P8NPOq;dNFBDz~zq+%E(>o%lUo
z55{ToHg8x6R<)ytV+ZTe8|*2Y1Y^<tApAbSeJq9aM^f9;vsW<2P6$FHmmnN24rER}
z_jqRn;iLt7k9~qs;S~ZK@^tT6_M@B25O9#M5qs5tm&x$-xD3x&=hl<Su!nW~C$jYx
zdE0?nt4<GPux%&jelK$OxE!Yk%F(Arh9%_iEN0s;YsI`>^7cDt=zH%fF#m%B%29G0
zW#;ar&CJ?;8H|@dgR%H>Fzh3P@a;Jn?ybUgxdz|kG<Zzj_GL~<NUatRcZQ;R67w;2
zoCC9u`(`BfH@JqQXFPK*hcfGHAU!VBjzvAz;Ot3ijp!-<NVaBDbI^nOtl9_kM6=K9
z_K7o*exc0j)<N@&_2g6?_3&zpZ$TZHp$Z4dyTBqPw%=rp&bfsPXXtbPqZWX3_W#Z6
zCAAu4*l6)GgglmNv5URm(D_=_@VSnxWuy83{U?yUWPcsjkhiAvm7OPV-_kqPn%`Fw
zeqUWSP!lyvi-Mi}w;Skb)~o5)p+?R{h4lm3$9bv1y8;Dfy_Tafd8_KJ!0$fHEaq%v
zFm-b-TiH|A%kh$Pg5=T=+_}VFhch+EzJbWQc@3|rUs}%HcIC1DX!hO@@$}Y&S@>Zp
zdHXcV7lT-LD#9hm8ZX7E=~A4YAVn6RQ@xVHjST(Rk@H()AJlJ1j&imexr4lwN^y)l
z9u_0TXfpTgIw`zXNbzkkd(rGWf4A_(UZn(;)Iils@kYa2-WYmM0u5{0_Xj1|zDk0S
zC*H`v>W$DF-pn}g#++H+ct7162gutwL%pb5^MtRn7jqB1aG{eIdYXDMkKGH;7I-0v
zyp2xr!cKE<bT#$HjTcuCN8Xl`se|Jk#W<a#*xS@e7?8K`{T)TCMh@b0M+b4ByMqXC
z;~*lMJBUla?S=6Md$C7jCz2l8i@uHRgj1ER7=6Q5^iHuAP03r8yRAqM=q)a~&<|AH
zOHA+9T?}g7T~ym!3x|Ey;_nn|5!~Kd%&Y7y96NOpsZLfR)1{M`Mcy9wuoQKzJBYE&
zM9x^+POMwqTHOEMQfwR5Qj9#<LhQTUTo`{h6Tjn4L~J(`VN%^p6q2`}J~k8M=>Lgd
z&`dmO&`dOWYAp7@H5PV{jYYcy#$xJmBhj%1bEKHtSd&04vY&w%GS)!&_Nc|+8$Yqg
z`X@3hf8q^!`{3|*bg=%0VgG$)|EmTqrq)1^w@*$}D;4w^8SAU4U#~>(m2XgbqKx(J
z6Bx-$aeONCC>9lC=_+#eZZTvz#b`#}4kT}j-xXs?RWbH3m*O0GyQ&`d%$|OX(afK)
zq*up+OdK~qj~PIZ=z)90y&{jW>+fT%T2c)2iXzOMpNq`PIjGj<V$q;H4Eiq@YYlVJ
z*E0tr$7Z8*Z5GUn(@+<jf=8CAc)dLp{V%cSJS-2F8Z(RWbS^%Sw|%c<<7#j={Uq7w
z=aPkC6&Ywhhkn|n8QA|g9mmF|A$?{lUUf^w#O?Q?yM7-L#Sbt(>j4(8pcnQ9b0-$3
zqTjj)n6T;r_o3d$e&#KX^STXB=E2RO9!j*hiQUf<FnR)akR--qq$hWmkhku4Zlc71
zJ?1>_0A+TuT5$t2WXvp1jmNsN3AkLqoTyE4n8jR(YmMlmwT(nEd22`=Y2lU#h;GaT
zY8Q^>%pV+Ns>cA{%QkcmLwym3AFKl^7cx_>V<@`K3PlvP!DHDI-j^1N?d&6O9TbWY
zi8}h2bSP(iuz|ca;JLl5f}TEG1^q$1R}80)h?=C1EvT(BWk$ps&P2;K)WLAxI-R{T
zYIu`gDY5CRin<UoTBvb^yd9ZDEtH=U|5C>@hq{nSt(4fvn)Y=!KJTE!?|!^zvyR&O
ziaOn<oLh0;`Ot<lNA^}PtPjES_-pt<oo4mQV2DA%XugSjW2WnGdiJ*;3dYUVAy|>d
z?415;q^7B{kRBmZGB>`S3QFE<&Dk&Z^<bvJzx2+u)}SkU$?^mhZZc<SDtmn2SYMcM
z9&FxPhNJUCu<Ua%Hu7`sAa5<nL(?rGSkHQQTrI!u^I(LG4Z*ARGM>p<hcqSE4hEy{
zS`doo(%ZlO8cO20^RxIG7EcJo-u~<>#|9xeHHbO0K`^%o#z4DZ3^opi&lhS=KL$a;
z_ji!DIbp$=M!jfmmk{)6E2Hm(**^<HIMcp{pT7cdwIg?)T@Jzx{~)Y<MSlo+`?xRj
z^~l>7S=76v2jO|QV0exX#*LG~c(|CoY1Zh;{vp`UdUHo2>)XXLMD1ja&YIVH4D<Z%
zk;$uMDB<JQNEzpe>^<94r!_&2p$o_Xb2)eOu@*O@AG8hYGM<A*8OzbNP>!$6U>Lw&
zMEf=L_5PA^UME9)=IdUgzGpCV|7`i1wEP={h!kpH&!`dS#cVLvgWD20bNIm?>}KX}
za*mr$kE-_!EqlPs)aSWt)EhYxo3P(HfHV6v&etDl@Zhu-QgSZYI27I2gklAIGIC~q
z*)`-|DE4cbv;Op+ql5cZ1x9yc4&q9lADA0a^F@hc)Pg1P^Q<LrN9^Tw+E0ac*Hwru
z;<YlEXR`0)c_S?p&h%LK=DgZP!(EUXSny2N|Ca_&Q`t|Zx7#kBwQDGSJUoML?8md(
zUY_mRYS6O{ebCen81Q>I$TMmRS)Z&`qwQH0I$u_zxe<Hr4V1V!i<uRyVHe-#nZ-_y
z;^uPPEoZ;MO9s#Na{gY1z>L?M^|}zWW}l^iS&9bK1$&dXrQ~h<S=V3^#+-<^ez;97
zQ|L?|m_L%@$7(68&7{cFOHf4KP9bl1J(S=JHBF<*+uCM6xJ%wnqju`RN+~v;k-~Kw
zxx7zGPp}lvCUPdbT#6Co?JDxNV5S6NAH1<}w>OTj^M<bj=dvm<TsrHGptaOQEs|g{
zdE0q~1lM~=kk02jC3s^EdD~+qXSZrk%wFON4DiI<Oiwt}dBV}a3lG10;*GHvhW_=$
z)PKA%!`}=2$lE>5JyB<P8C6d%ai(K0md;?UI@MkbO}7_`WU2jZd%*?<GaKzi<Ms}s
z9eI0-yw#YIzcu#a{z`kXZHt{)y~IwaZ0y9*Y+K<QWh?TNZH0ZHtyt=2EABE=%C}E%
zv7Ed;P2N6eLCsV%YXR;6N(t#ACUof{4m_oD_-AKPGPJWWn%7C(uCx@NoGrzj_8mm^
z%J$;L<aQ!$RcoOnZ~s}|QoJH>%gI|srI|?ZH5Khx(+*TO6WwPw6D_HUI$f`s*gB+{
zF#2UI{*tpx?-&b}o3Z$LsHw;qZY1irY$D>x+Z6J4D07k9@2MqcYoXNsgk#sA)Dr#R
zPT240r~Hm+t8eHv;Vbj+YcOY04fF56z~F2(G_ucd8B>L}$#1Zv-3ufgFT;)APw?Hd
z6y*a-@SjUDzBv^mjhsw>T8Np~3bC@Tfb*{c%rxNMyNUvwepdih%|lfF%*E52xwu|0
z7Y*#VzkMFLxH=oIC$nMQ@e$Vcd(2tRLp=YMhh1}6&mPWU7I`jrTIZru7W=wn>eP;@
znDjaYj{8z@y&?rx0V&i1rXqNADjp9`L;nGJXz?`{tJZN}-ncCEeUt^~4cQp=i!<6~
znecwgy*{4lShJIJ*uCjkPu^aior>1vZOz+cT)dowk+;anCO7ds=?2Evzl9M`ld+)n
zZA`p$8)KT>LG6y)@Hn4LUH&a*sU=|;d3*0+0@nVB$Ev7!{J9#>T>5x)$*119bpj5!
zF(>^5GmfuwpC<EO<5S`>ZCC=@c*WzxU}g<7BgZC#yQ&8;+l_vVZ07pUu@6Hqbx>D!
z>$&reJ!9%g&i)gI{+vg3r%rg%G|s!o+r#Fecrhgub9XRj={~didEc7)zqxHEzO~Tc
z33VL9_p0c-RH2G9y~UiTw(6ij&{zew4kJg&TTLe&EE1>%N!H-YLJe#Rl~_5JS|Ik!
zX1rCxlk>eAPc_`={k%`!4rYBTE|9mZXGgPNyqvt9?M|&`b82oHQ<oFVnH72a=?;6G
z2Gr16$zb!0IuQ-~$mFd$>KbN}A3uf!llQ@h`8OD~iNQGVUody024lKcFgCmi=B}6!
z`Z-yHvd27-_gxd}p3jrFM&zx=T7`{6=#S++)|non_evFB^k9w;b7cQ)mxF~5#Jdn&
zJjx#Ot6=QR4u%tXn?=?>-^@NQYwR<u+g!?nF@q0}CBgXGDi~+?vPb+T5Ox1>w&oPT
z8EXLAF9^W&AbS2S1938mJ-{^TKyL(bcWMx|!NDl+8H`(vg1KKb2uZJlu$lbq$@y;M
zW5MXA3C80aJV$g2M)0>FxbNrQ?nT#dg?`D5+H2@MgY~i>>sZ#g^(V3BZ4(45@-~WJ
z!(~Sx>_V>L&wajLob`p31R~O$Ik<g7(6Mg_CQo995OWO@j`K|MGX#f*%Q#DuVJ2Cd
z!$%nZ+>iJCaP|%iC-L=|DaWD&8BECA)JAf&{6ft%`|9_Oa0a}GGj;Zx{d^P%p?|Y*
zKrmL=1|xvyg$VNY>O`J1h6X`i7X-uk!MJ^rXDeP$$6ct`<xKS4G97wz*3iQv3?tbm
zUR}qzry;d*f9Pu>%lFsgT=QRYhkb<&%p1JuK)u&q_NUj;%e6QZTI!*K<3eH4i8^X>
z^+*f$J6g(chaS{uOL~=gO?BqaeKW5=M>i#^3|N!5QegQm*0^h^F;7wA1FwO-zj-eF
zz*&Do*0AL5&qkd6n=vztdOMXr=OI;`(N5Ch6FE9<nih+c%n2iJH5ttOf2qQj6y|i9
zYT)XwL5oq;EMHc`ak3hJ=2ENvp9&W~DUjTRJ)INuZ1K!$8%#eI|8K3x+YX%79w?RJ
z{v7&j>8bVm-z;M-b#pxHEjOk&b)r8uzV%1PQh%&E?2pww{IR)(KdR&XFq^&J@+=?t
zJdmQn8YzkmrMM}TV89FswovbzJ&C&C0tx$267EftAly@e*?en0P0IZu)CF&l!il`K
zAa9lAas$?|LDimU_|_A@zj@-rMQ_yX@`eq0dzHMMMc$e%_J)BUxtixqAF%{oT1dEK
zLjv7FZ!{ur&qjJ<OsXg3mpmbz?}^E7o~XU=iDADzp)B*ni`(RHeJ@nidNSM53wfn{
zUg?Rk<ZUPNwq?R4JR@62k+(tQ?b^lm!g`s#FyM35ruO3OMmtf%odYR)JJGJdPE@?L
z6Ngjn#Kd8CB4~x3*fPORtZZ*5KD@RSL6Np1m>wYKtG2@Xac?0XM`vn<jc9kHr&zMJ
zySVwSo6x`SDq;qA6*I`&=#E{))3nZ_s+!F0XeAOOJBe|>EXA&!9Yk?Q3(@dRTaj?L
zjTpPIwHWuLrFf}pF3g8D7v~l=7gMUtM7a}Zwt3COB<3_KCQ*Ofx|w+O*I3M~HWqWD
zj77`U#ykTw5e1i;h;av+hy(6MVtPme@uEQ;GCur5k72)Hzq=MiU2D-)_Y-UD|HKrx
zA26u-j@_#7Sk>|y0^PpC%eMxvCe*-s%@=&uRCC5#jojtM2pmGrn!jN5L=m1gdV*5&
z_CmiBoGmKC+8)IiZ&!@`=t86`3$c;Bl@^e(uL>~WO#z(A+e4NA$J<@3SIs}?z|b`d
z)m52TmY)gNv`lpTnuAHrA7d7G_;}Xk;o^o|1ea!F?{@BN|3Eg9iB%I)@uLIZwoXB0
z^Az+gOhJ=R+_h73kG{=&*gZ528~f(rM-9C{YjV+qyq&N(3roYYpwG>uem4`%sxmMg
z=}4*~rwt##VcUItSjt@iLCLUwc?+%2B%y-5ZB_3kGo^08Xu%DvElXyOUNVeZB*Sq>
z5;||XjfNMK@p;cJG>za6PxG5Nm&@JnJ>!vnDGt5D;t;WgS;yJT$Z3^;JN>AO`j6h8
zKhap;Hk$jH<B{OR{Pms8&!Gphyd(-!=vUYV?g2SL|18h<hfajT`gRyDtkh%vK|PWh
z>hX(yS||D;7e8YT0qa=xCgu^>qdxQjbxmaM9o7d=dCzJ_-u@e}gBLX(1=Rg4NK@hT
z7bSw8(qF=P-%a*6a}z>fOU;c(Il0q8hZuSZPv6sEDEWC@DDiGI^+UXujmc5sM~M=T
z*}t<Vhp+!XlFm9V>h^p3q9URw5+c}wC}kJMTy}SNd)wXJ1&G9Uvn(xOVu6CNv>=$+
ziQNIHsLy<V&ma2&B{%rIuQO-Pob(uUAX|IW3)PT%+05!O=<3LM75<;PZVb}s+1bN0
z|5q|n$?N@N`l9k1@tV%8SEVm&R&r<mTFwDh4aeE~<PmwBw$g-dCs_-QHX&&w-w!ll
z8F}kJfM3H}?#;yr9F|1m06j_7*~ffIkJV!bYQ}$KaGdp1byGA{|CpJ<p7PYW(O5*@
zZelMqfi-gRK3?zcnlO;ObtP}V@Ya*O?Z#{VHR{<#tl#E7G$E1<@pIzdlpWznBX7Tw
zx1U>apLUv_^Na>$GBY;sOBl2%MvNc}H(v}#VOluGv>=0gd0(gl?<zN<o56?#=H)q%
zy%$-xXYyR($#cXX*6pd~;hg;mM`lVG)|@e*XIU5y)?<xK-UeO|huwg1%pz~U-!Y;%
z*ND{hMl4f?VQ>#4?vS?$2e|+HDBnN9vx66Bj>=h6yHUTUH>>==2-J&Xoo*9}gUuuH
z`W3ZiK6WbB+5dc4x0AQ`Cq^OUB6p9o*K5h1tnC-h@MckiCU3`WiNw<&YNLVNS5Dpv
z^7dO}6N2ZPu-3(dJH5iuplvwfInUW{I{VG+Pqt)@S(R+88XAiYtS{g5j2D+0!TZCS
zm3`o4<We2-Hggia*G(hvXC==G<n6Js^lbC@w<2?0$lyWI5oqBRgC_sc<HqN(+HdB0
zPzwl~6M=2iAuRZ|M^Pj?mqx?Fk-YoHd=QIh9I6)$E7q9Z=kfn|KI}tJp7YXJBu|OO
z$nmjAXP>@C9KC8U>A_~*tCk#tFaIVk(I5BdcoeL2__t8P*OzT1eR(k$_bCQFIos1T
zKNed(S@Sv_gS2lPM$=cn)+rA0oUiI~BND4SM<ShPNf+xVgmsC68#VfGt)uYHCJKh=
zNL;(a{=gp-a+!k|-%1C=WF4AS(WCYh&ey)wW7lnFL5w1IIiuS2mk!~-w3z-%gPk)}
zaHUVF;ENLT&nnq(RnZ%ug1HxO!&TJ(lu*1;B5aikD`YBqGgVm3zV5WG%%+G{;rdBA
zc9E|t^7b_6b}o^(zOUrCdsB`VW98VkNsf4h9H)}x&`4Q_PNHtCRpRG0CEh!!u%NCA
zjX$$zyHkmee-v0TSAjb@a?GqI#}bJQp6%oaVx8N~RgOb^Khj^0S*3F9CC_ivQsBlz
zIqF5pk>nJL@AjccJsXM#6IsKOwX4R_12vj`*)DCwI=41r`%^M@rH@##(MMb&Z>x~E
zO`rIPPMr7gZ{i~!vbKF+<}K<`KMpo{i$h`F;?xmuar=~~aQ);Vew^_T+KsKngtS&-
zm3vDOUeisK6}Auq`m_*<WbCZ0=E5$!xsZQuDi$&0qHlB~F|=bNkusjMw^=Tt=I#a}
z;COwJX6+~f<@O@e&Q7?DsV5A{w&HFxTXCpk9kDshMl_pdBa()2#+JOb{#aYg*3=ZW
z_gIO>ZLGuzM@zB%UUjjdTNUwf*FPMv{)eT5{-UQ7_r&}B#=~bnaY)IVp@P}>?Ay9j
zpy#}DT=Dz{XUDHdTJ{wt!&h`_^%c{WK0yBu_wi@Ref*k#4+9&Qz&D@-1L8|?^X~&@
z>OR1I^7b(Mt~<Nl#li+9h<JA!IcIL74|(fE-llK8jxN-UE&E)je~NqCnf3G~vKX%(
z6k)S>Ay$*Oxn!+HVgc4IEP#`30X&PZBGH_WSMBmKuuVShY{<jyPwb(N%Y~(14&19{
z!@g}MLaJoq{mmTQTbGR?=QGig%#8HPL<^VecsR8X_G=1J*{={!-3wv2?ken7<zva4
zT)cRg3#~J`d?N=xs%ImAYZmvxWy07j6_ZC@g3}xB>Y8;KZ7aCDB!GKLCS8Kvi4-)M
zkpjEPDM*@{!ddxb^iE91!L?~{NlZokRAw)xCu5D}Ma-wyDNezBe!EjpRnU)l{RB2~
zE^sfi>qd?_gUZjGcjV0KoQBM-=RD(zRh)NSdj^X+i(kmu!F8OsSo?z+z|4-bAzMeb
z<s4#{<1nztC1q{f&iOd3U(p|x%#3|zx5gYwKoomX7P}MBGB}Z5QucE-i73C3K;4^u
z7uEz5$=XMpH5s-c67St2@z9R_(jpTMzA-^T-s;0S2TFfyk~OcjIn<e(kgL68(4D-k
ze=P<x$=hpWt@aM*O<CJ6uS4%Evs)K#V`hU}6sk|--Y*3?GR=rlCk%Mjin-|LsXulx
zq0}>i^9ro*j&jCdZNe*Rud;bzSn<yY{}HU23d7+_ZY`n~JEA#1KA6m95BFf035|yF
zTDY0#g43+ymPTMk9It6<QMj8Qjp5Z};WCvvHhaVw;q(=4j7HwlXzXf6pOBjO=Nxxx
z=SE=E9`;eMlHug-N8aZTw@oPI?^Mcbe$%&P@HrEt;jC{bP{(#Op`S4vhsd?Orv{`o
zGC-cnU7mvs2$yk|@k1Dzl9BCMcdE}AQE=Rd6z(ju*vG7%TY7X^tw(BqE%qiD5b)iI
zIm=l)vlm><w>`MWyB>SNb;|U3ZDD}bRA$8Udud9Hh-O`zFqWL{%g4cq+$C>gwlQD+
z3G;*e!!YDf7)BR_;nMvuRBd2H-z471@~z)YI82TydQ&rIfBJt~KuKi;UOLbZ_KuGo
z>)wIn@rV!1_-n|%?$9XoPm085a&|Cl+e@r-JCMI?SliC}FA_akM&d6YyZ!ur*SeeF
zOU~w5b2n-Y6LQWO(cY7fP5*G_!EwfuwfRimpN8zGUJH(efpgbweshnJk-jw6bcTuK
zq<sW;3q)YqI}<)RM&Ky_zbe^j+c^SrTSXv=XR$)+9bKf<8y53yvWk1*g4oASG~qFO
z>kFpRcgE+bGhbhiHj&p?qw$qBt@BUzwwYbH;hPb+>9r0W&Dko};g_rNoVLS+zR4y`
zsY8$WPWr)PSi7HP#>m}Bc#wz3jzq)mJ2Nn?qA=tV&tbPB5OZ0FRa?lv5<NiPc0J2}
z%E&nSI_SlsmqB%qv*Xk@`ttWkoX5Jm13hac^qEtOFKQo!RO-dA3;5Wur{9PAM#K%C
zgU(R@xU0jg-&&06p~c7rT2%X>#dr@L``XNAqCcu8c{}xu4t1|+(f_3e<EN`Ip;U=4
z9~HQFKnc84B5j2d+XpGRZ$*KFf0fJ@QQ~&70%i6Je5CGsVJvlHOEq58Z`_f*y}C$_
zoX&FimCG=RjFe=^;B6(xq|tI*TqlQFE=K`*+c;E#Iqc28*sDaud3uY<+>Z1V|M{iF
zdG=x-QZs%_woc{O=>Eyju&oT|TFK!#Q4Y87Qv5EF;`|{wT9CItCXm0rq3HKL1jX}0
zke?QUzqX;sJQa!ytYiPxZzHCRY$I9?ZzIy1w-J}W`H0zBK4RMnAF*Jzj~HH!Zx4El
z+`Zo7ytlVl-^W{Yq!)Qarne}ie!QBzwHV?p4wrd~O~*Zj;<JY+AaCP$wHBkv+nOKU
z#1o~Pm~_2`a1t#<=(pzLO&T@igU!V1>?Wed^+rMy)kp-lY9y>iH53KRMP9}|1P2lv
z#r5irVwluk<dL`bGwX@a(Y7LonHJM-+KBi>8!@bhjYw}tAM#t)wQ04)kVD)hzuZbB
zF0Ub8G7EVTdE4KkiWt1(AHLN3$C)bbi;w?}cOJjd;L%TfI`R`wmRF#4!*WbsP==$L
zADC4A2T~Hh;LgIYu-^R@^;>_XAMX)V!|r3-f85i~9&Jg(63nc)0~kxtH<TF^3rf(j
zbqVI4x{G_{?Qrt;RyqB|C%7vh;09I<WQH+$>q*{bvHz+pE=K1L#h6s(8nUNe!?>ob
zIlmU*R&D_r#uuPJwchkK+;P|TDrThT!$96zQa65Xn*-&TZ1{D^#;ms4XyllUEDvT9
z-{qdzr)E^1%)*bDOtckdcsFNuBKyME$lHvSg;0>U&i(~xN!|v#<YLeD94xw@i>Y^W
z(e7dnHP$S+HoO9SOhaxX&g*PtR{E-BL{+9B{l_JEk+%oQVCQ2gc;JzY5q~b=usR9F
z(qwF8zU-v!r<h^PjJO78sIeyF<Mt$kZ8?J^X3DKBXExpE6ZF%ZKs)kw;27?Kqd)T=
z^KJI6JBfkp%ho@699~<wV{LLG22AE0g)_5pzf+g{l7Jsi6HuGH-PfGF<*v1%y|`nO
zKH{nLAWP_pDnG)#IGhXoK+bk^<{sIN2}pfIKS(;S4`<?#axVt6$=ikG(jxj(CzMBW
zCotzt>{uh(az2#&38oij=W6orA2rc?%s#KhYnMJ6<2e_yvU3dPZ=>g!HB4jHkhM8G
z<32VTCOhhR<Za91C_LtM?s{SbeCTg_6GB#u=U%V-dU}JIEl2<5n@FDXUFrAfW<osm
z)mr3jn}?ixa0<htJz>ZqU+Se$W96+0uU7{8eS1=4ep?w151kRQL#dH2pto1YzU!h0
zWX42cJZqbjOEJi!_coh7<~sq*cVNat=rYzj)RI3mWsi`|u)ZIOUF7Y?T@gr0HQ^e2
zj<3kuA2;dAYQ#KuYSoKaL&qI4aaO~G$;0VuvZaQ~`_Z4gox>Vj`&SR2QVo8yr}m9}
zY&Mjcc*DYQ`Yb(Er;M1MNZt2&7)rJ>Q<l87AaCoQ&_KP5zR^$(e*M%!@>+|@JG6K=
zNlPtPhoTF5{QRTCb7s`0tkGlSLe{YC+e*mXGkiR{lD92y8PRl~5nA%KFL$HK*z4QE
z{_*y&VOY&Q;C=bC6;@<Rf(aJ9FYS9#CouC&@ST}Kzc_Qk`P^V~*R~h?#?JJCzh@8l
zO9a&GA~_$)9WvzYI^MddsP~gc)~s#Yj*rAl>c&~Dw|_1+-~)HBHD$)HR%XOIJ{SL;
z<XP&X5$-<WkWVne?gjhW>}^Y#A-FNhhz6JGFZ;;nmyhc*vkB|@d;R0N?rW_GWWD74
z3)#DU34b^8b`Q1iP5jvS-6r}gj7a_x&Yq45>&V-d5)-U6Ce&s9Tg316a}<4HQz8(+
z*MZ%@aC|0by{W5zDB{mNVT9{?W+H+$I)CPuGHUsosf#2r(~z9E+r?U*^UC*5@OCE>
zIU6J3%=6b{2YS?f!w@^rfTctA=zK|s4{Q0ByfxG`;Q2BGQeJRh&fi$P{YA~>v<doM
zJR`P_z`WV?*RNIMhqoU2H^b0`XP=D8;n+`I<juWs6jQSa<k@paz7e9G4x0nD=;x-z
z>G@jR&(Pu~dHayO?e$BBtuN(Br`KrdM<tS1YjFArJxZ?>_>rl=Q%fbH$<@Vk6i`=H
zpzt`o!7>GIH&npAP>$`?al4LY9h{&<=?@iVZ&G3Y5;+=nq5jKyH0`#GuWcFf$lJ89
zGL$>Yv9i4!<y+<WeOeAL&h_LUmg67!S^t|1{<$*Pt&y?Lm*H`c9N+l4bq2|i8y<?`
zjU@OlL4r=(WU%QbgNnR8bz6#IcO|G5Awid|65MwSMLY5~%O?bLcZT4WO(@z-55;Ek
zHjlhrF@*bS2e%RTT-u1rM?T_96zkZfKEiXBk9a`dE-LjF3r2X0N-uBmw4JxeRCtSx
z$=>4IPA@TynT_q(t2O`d6m6K3{A{3ySoy(Cv@p7f#Q|<&c4Ie@)ux4bTDyga?AbyL
z{@k40<?g-7oB?XoMEpM6P)NO;#k$M}Lal2c)^1{LE8(8K7)No+%2CW1YcJ$Bb|N>l
zp6IT#6>ar(L|CznxT3KU2YhWruWz+P6g|i5nAvE3*P6QoIqzdtQ{+Cf5`D*8ir?x#
z{Clp%)cKX@+_n+}V}2ve{x{SQexggK?|3Eo21n;|?v5|RZt~Xn;yV_;eF;PBXDD)b
zf{()=BfgAT6oc>M)tvhXoqZ1z9ZKNoeFsa2-^RBPW>L(er&xa*yZ_z9$tO2(XaP0g
zf@18sSOh!&8{BJk14ZeD?5`H0OjC$x1$XU83Q?=H5c%{wxuzGOt1_R9n)0xJY#!%i
zbJ22M9$E$D!jJx>LE+i#tL0#nZw`8l&PE%rY^-RTg{aP%7}OvWudbPqpJ>LBS!UE~
zZH7gAGqg?3=vSHnr78mzhII4^Nyq9^<`wtJz&mF%&M3_YcwxqRc_sqYIfyyJU3rbD
z3%^gpV)C}lvJ`YCZztrPM@%#B4fA1E@$F<tw{u_YtYmDcbq=i}lW=a|DeA%GEP4Az
zbBbBJ+;i%13Keas$&$C$vrnO`H+@bviRhoq-5y_$L&5%P@i+P|S$n3?J3N=nY|cJc
z$QJH7Wwvh0szm7OCBmD$je5vEI7JD#U`da0&qRa_PDFPv`iRda;)~63w9Ze&ydY*d
zO=I5gr8sn_?qyubES$1<bR%z1khksMvOY~>ro-MC)SVE6)(O!Fs=@io*SyA6jX<9V
z;iy8-&zzy`EAp)`|G%=3wS|tJnMd@h)?%(O`@Ea!bIqz?{lHrGty2_kevZb_szzL?
zVCH#0W~H+K)a<Pe*KX+0irK2O<8?^m`>TF%S7y^NRN?h!;W9mj%+ll4Oz!Ps?`n0q
z9+TP{VBIYoJE^7iiVDZ(%y3MP568S_%-<nzUu#)AQcphLgS)b)@!HAjcWd^;B)ndS
zQkTr6S7yO8W)G9M<JRyzK;AZ-9}Q2=J<cA$tfnea`28?~bC?lWz0`!eMlvkRfZu)w
zMDkkb;TDe3>_zru4K8cP^8g<Qzq<5pg@?n29+s9z^~}xS-jmO2Y`w0=TlUdvlDA`-
zS^KXxYtD9hW}5Ln(@Vv42TZTjAaJe**Vn62e}e||0yPNzp=Cc;i;T@$n5$~B(^H2_
zb~;209bCP2=qc0TZm<qFPwO~8M$NH{9wXG$9m&&NjS;<ulIg5%$1MuOTJ8ehMc)1(
zZ?BSnE{S27+SPz-=L~4PjkR}2vX*)B(VxhY{_ML}grnO76Q(^eVQ{($b@*81vNki3
zvG?f5-ovw(RzZJNea>l7vz<IP5+yB6Sm4OVWT62s-x#o+XNhMLBVO|7eP%7M8qJLT
z7$f%$8Znh0yF#7f=wu_k%A9d&6Ao+Ezp~BY@Z3QEHJ^7IdlS0wt<?|{X7Ts<%Rad`
z*+1^DiJpCG7CTM&eb|UYWct*rMi|SDsM?QTOWvO5^WB)e>a`=NHB<X9V6S`^Yip|<
zBP<$HYqvAv?jj>@b>-`d&r`b+13FM6Su)*-X8iqj^D!K9kH0JL<IjzeyvAtoG@AA1
z3*M&j9K+n<VG{1ixvxW~Gde8e49?}jTGXjw;Q7&jA;N&_Ja4TWz^vd1J=SaqL*)u)
z9-dL-BWucIGBqUR?NI)ln3rlaR&WPwoF4P5(v#jqhut20-j8eX>wp%KtvII|qrjoL
za)jH;aaJlrz&aUzzoH)djWuF31v*Vrpp<p(*ir?0k+<gRO1vyk;M@!a#(kGF&qjgr
zU<LM(w@b6-sGB55g|!l2SI98;vlM<t8E%b|qO3Pr$hTXvrI^Q`VPTdblDsXeFUJfr
z)P=kaOpv2CIlkIY&if*V4S9QCC&O;?_Spw=v_=T-To1<87s2Q~Ed-|;N^ttP1noA+
zFseJhXPFdv*QK~~Q-W6}30kg~V7myx==vdO^E?=eJt0t~gy8gp5PWRmBTSX7S?M!A
zIH-+yRjZA#VvpAEkdJU+ZCf(aN362&5qTxv;!+oHQKyEt2x;pr+68)xv<u$i4S9QS
zu9uMZ@)DPxdy1FSJ;jC;55Yw)qWuB-qZYV{<8NKXC+`;G<@e^|9yMfHcyrOPTXXSo
zQd41~UVPobK^QOFi&S$15lr6tlD8AR8VFAKi76F!LjS@}7;D-IZ*@IUW2h~?(Kcd4
zj*SojHewukTYYV9F@n4emDLarOqQZ;banA=c6FhQttJvjSqNeJgG1!)jk%RL(ykIs
z^L}%0#V;JY{u5`qeaD2$Upc?>4jY=ig2#YTX6=_^)4G?4ZT1W<<n4f@hnOD8odAk^
zxZn3KIz^S>w<y5@_dBTSdK*v2+`?VYn^?(QMUN`C;7H!84qr!=)y0TESA-*Tig4rf
zHC)RsM2ViduBs3N$=iP9ZQ%Jrblh49?WF?z(&S@raUR6PJk;5ni!5ml_UC8gSVL;W
zhHMm5^F7-=hdU>;ae=+o8ly5%-aZp;oH9{lHe=^5Gpu@>p>s7eCo%&U_tU?$Aq^ce
zxDQ}wI&2Q5qx5Mytea+_#NLdqL1xVTKt8|Y9)aX^?CYC>9u`;7<$Wr)yQHH1%oNV+
zCqt2Y9wFqdE&H)ohFyfK?gES>FJS+$b0|NN1mmt#sQHZ<cMTI!*E$gsAH|~zXB!r^
z;_O}-Gs}C^9~s0v2KH9F@Z<ejayHhJ8Bbjku~fx5|DB1Pn@mK<ZOow{ho_e^@An)t
z8yYh+t_yeE?4vfkkh6cheVWX@tX>IlZIi%R3wk(R5>Ot-KG%V0d}_zJ%7IZxW^G!R
zJ-O4SNKE3qO|u)+?PQUNI30r@TVtRfP0tfGyD`-1N@A%&=JGnYiMlH10O!cV(S_HH
z1FU5V$lI3Wl*0)6K-hb&+JrsRI_!^`qfkI^E6Cfh4)hL(grVIZJ(N9|zj~E<H3}Wp
z6l$^gf7h86^jYrLVS-MFFfyd|0_v>fh{Xgwc9XX~YjBSUJtJ$lZ)AEG*0$vBO4gKC
z(d-Rs!f~#BIFyZ;`5MeRrUCs(b|!QsZ{M*du^1nLakC=OsvBoYc#bG~8Vz4&RsHdZ
z#%j(#&ZDnJ$t;b%M<elv*Xw)K7r*c<VWjV;Q@9bqDg!o*G*GKG;0g8NA!O3c4kl!~
znedZ3_Ez>L6^9HMz}oZKG95b3SEF_vHP#<eq52>ds*U9wzeI(<<LLivN&ia}cc_q?
z%Tly_+_cP0(!gtm8iSW<u;hpaW2dWO+f;);BQ!X8PJ?g5G*~l73y&bqRz_=)BjL52
zynR64PJGNPzv=Al1?e%1bz2sBdw7Tuwv)o}m%B(G{4!v~KlTGHjmX|jFIOP_(ya~5
z&^4g`7J9l0SvNn?<4_+xugTm+>S92@so~giD;#QSoFT0JwzHm`%RcVL>5(XSp+%=9
z2Gm+=z^4U#%QI81l^%iQSEDKhq`K=-ZI~Y8UmMUz5r%@aF#Nh5h6d!Vc{Xd>r(qaV
zgMKfbpKcu1!#9(C1bWF_mxbf`dK33p(f`KhPEU=#VIS6;)E8pNT>FP6)ThQUWjVF|
z9@LEw8gVMXi1bXdk-Uu_8jjwsCb&Q5@9M~N7J2K;n}eCV(R+r&Im3vSO^mRrOU9G8
zx`R4=uER6bZ3Dg^34?7bBa-?W5zogiiO=)duPS^@P$Qj>{p@iXWYCNBLdyBn_i79u
zqQ&v4I^2%a!DON1+?E!B?=<M4RHF}lNOy+PC!E9A^*Ak7_0iA^s6prB%rb7K#?M3*
z>eAP{gFV!!SIqc#)L?^wS;TiVSmCaPvW^xuUd-|@&|vv|1)7t$+o<z8yq007L<YAi
z<Qsjz^PkBv)LnskI~8~?Q9#P~8<4HGm1J%(>(JqHtRR!WG*)2G00mmmo7CpA95VJ{
zjeTXzbe6$<P0B1RDNc-(Vohf$rmm7Q%R-8KZZZ^ZlEHbOj2U4vT)ra1Ao`wm#mlj9
zvI6z36mWbVf=(ktFz{P2I<F1JGpk^pfr7B|GjGp=F>`VVzPd<oBwm6qsS;FvlEBnT
zitcTsaB`3!D?J2f#suS1*I?9a6O3VJf??wq!rpBNYPAc%ix1wy@wc})@W)$Z4{Rew
z+q4nm$=e7&AJK6%d$w=9#kro`^<(EPHox%_6D_^PLJ#tJKlS5*-Xd&;mneMgAr9T~
z5Y}fsME+6_k&)pc9^7jo>Thl#rUWz-f3`Fer`9(UJH9m$%eOTZiL)AsdZo@{-*6|9
z{>ee~F0&W*<ZY(eUaZNs7lC{$jk6aWd)f=b4Lfn;G3RcXZ=vEo+m&(~p^(`K*DE%n
zWj`CSb9HUew_k062x`d<YltU$OJPc=E^_Bo7xCn+iM-98`xoA8E3srwB|=+QG6U*2
z%AWjUzRgcmR+OQ|^)EPh=pE-YULm7*DRan6Q9i#E;~GB2grkoUaONR$GVf!PlzZ{Y
z?joK$18Qa5K`oCvoX@?9#Xqm(SMYUsNQ)7fT!dk@ilFUt4ej0)AXb%!n~n0(WmF#X
ztaF*SU5JT03lSAmh{u72SbL6pdC1%Iqc{s2laEG4c^ES(4^4LE;!|i23jbq&wn{d7
zsIy^LB?r?uhts%cHr7weLMS=g{+Ah{x6P<?(Trm|%xD2KEUTNz)eO`nZ^P@QV(yky
zR5YbVyy^;$j!Z|YJRPgrW#9rC?Bjo#*??(C+mnup$LYvzaRp00rQ+AWWK<cGf`Ebx
z_<G|U{;W-c=fw+HK;AB`c>#N`ox+?C+#^!;G~BoYpenWE<Dr~i;LMGTynVpAh|8RF
zc)*N+O#?V%Fd`mKRhS8<Nx<u7^f9@!j^#|DO2M68JE$KoPDG0hiJYTgc55*|X9070
z&ocX_8F#p@Pe7A7obe}P&u~tECU4(;63}l1XMIjG1J^MYRXDqm(KZT^U8AsZA$3;H
z-h`01?aABbW@gyT;~c^p6HfB_HkN#iEH)vZ`eWWv6J}m9VFYWCyX5V^K+b?t7uDq(
zFqhY9N6rJfu(la~-o$+0aMa*6?hb30p1s1cp>8+^MzR*+u8|k@45&zEuKFJx-ba!-
z1<WRIrbEG89q#PY!G&3{8(!)V+?g63uif9?>oA?Z5f4i}WO_Y1-q&M)Aa%PqvSb52
zM@D{cC9@yMTN!y<5NO1MhT&+-pS$ab32W-H)~z3bk@Q(P^GwiV1v7rg+xKguA#ct7
zDdg=4dTHiSGi`L9^RDzI=aaWfcpoONqTWYd)MY)>cZ31&$*0@o?Z3C&v%;ASc?0&0
z*+YK2kiI4IwmErwnY>*$hj~7)=<}?mLSSbVDn_cXa4kKe8&$Yz;&pZd?<4bW7yD~5
z|Em()ZgQTnzZyN5E6W!!a(&gv=G?|GcV_RgPA-#cFg=AH)M;AO^V4D~bLKY(=urEM
z7NIq`gLE1*Znx6ITA@eiSOc!`_cf0XgU3YH*?oCNAa576CV%b2Q1hb!P44n6(_IfU
z^Z9<VzDjzm#}xARx1|;{Sc?TRPbgrc4!U#UX!?UY$J?9O+hyHaqQ&CpT4rQxIoC$L
zlUlcdeMGAg9R^?5p(W2!p|`o$lv+h;f*zNC=`kjj``pfjLBZ!fj(btt`h=m$Z!)$>
z&waRhOgnGDp)E$7ToR7ve)M~-p(gFdoO?drhxl_Fo}pe$-pW{W4?aQm(sLEj&4eBN
z8n*>TypA{GG3(K&A>n9Ki@NqBGMTl!vs{m1e66&h&fDy_9?JxCh8IvrS*gK)j#^Zo
zt%LVTdTH6$45ikfY-GgI+8T_qRH3%33MqX43dq|>?KH^e?9152oRLwek#|=OhoQ`t
z;A`>|&xgag>*lnL4)@F&JmSayv4%CaAeY~8#wM4uuD$s^^=f?gQ(={(n%;e87S2*3
zX$HN*ulZaL)Sw-C`|}nxjz(Jip`P*4Ud#DB*2L6qU(kD$a#IG|VKPkrB<0=?DLRpr
ztp~}V-!4Po9T}$WlA{5=M}KxH@PRt>Q>h$2edVzDNDq{Y0=xPuP^TX2<a6{du@CFr
zM~0@wQv3;*V&x1e+W1Og<0-|p`BIEDOL3OGm6ErXyJdK;lwm-+3=e0}Q+!O0iQ^Qw
zWvPJiSqP4iw-#@Mad~zyHd_XxSQUg(Rf4e8BM5tjhoHhq!ab`JtQ;l5idPcUZ6-xE
z^0u;oYz_&=A@a6Pw_rT+35IffFqT*ZV#&P#W|IYAG<o~!8|Qe)+etsD5kK%2w^+-z
zx#2D1IOh`@%(^zuOAO2P65da|#Oo^FBB2#)+r{4EM>lWbPu^yex4X#O&KEpHVMh<~
zCdgGh=+#8zIyV+Kotlb~ditTl8;dUF?V6u1VpvmW;m-X&3tKyhU$q>?&nNa`eXPA`
zL(WbmZy%hs7a1}3;<mz0{Fi4Z2Hm$4>Wy{9F3-B+=^h)=Ps%+62W>>Q;2h8bYf)`R
zP2oU))K}*kVpW2r7<!?)7&fxH*fiHd%;{?(X1w`>VXG?PPTv0Xszku8-w?-t;&fgG
zynmHpZR$5fxBdp7khfS^yA(kaU*N3&bC|y~>v-Q|9E^B~Dp~jGJG_VY#=BT5E5Q)*
zcEEx=`1kxK_Px7~nX2nJ9K-$WsYS3RZ+Ex8hAUn<_&YfVR~+(@HZl*30&}r)dm+Y*
zD8ve@LX2l`HuGFQ;{WDjZ1n<MBX4Ww=V2Z7<ktH*1G|?R@90e2c$9^Crfh5v$;L2;
zOx)g-h2s;lP|YqABPz@odCQC+ax<1Z&w$q`YRA1Ypn8#xwNIJl^y@MPL?<Iq&b<G1
z+_4jW8NSUfBgf$~IvX!z$KlIZ@1BZw>r#>XH<dY)SCI5E6*bC}@l|#n1EbEvwclCX
z+meJA+&len=1I=eoWMf*hwD8#g-DCjXu}NFDOv2L2C{BV<*rrkjq&ErQZr{o)wkoI
z?8f;$&iTp8<B=@mzM7ld6Yh|R4xJJ)l{IP;&Ko{h#QK&R@+<Bj@yOzwA+z^GPsSs?
z3FiWrB%n_*chNkFMHGEXp`5Geup=67ZKJXD2s722aBtfj&fm0R_Tf7+_G<*5@1{?N
zbwkP$>T%`FOgPBiBsJoT<n8%$^Z~ur;F{imP0QHJqkh_M9`)24tP6%xJJjoNCfb08
zE_#eh(jx~3j9kYZm)pW|sY^I_WO6379Wx$m$fHmRuJn)~wYCK9Gvt`}UW)r)=%rAy
z2572-WQvX%QQV)Y(V;7OTbq5Z7tQs!bxa5II2}&$eP1p2hgs;kuhD=xhv^~WZqCS0
zVen!vuu4;At+Pf+@Hb+XH|uWBdxmc`A>%Ez!evHGV*k*VynX#5ihuL;rw(G4hFvr^
zSw~<3y;rqtxs&WPc{Do=?+a8=^FEw=VPwVwXM0ZQQOC=G31p<pC+dB>STB+tGu|=h
z=@LCl3&OF7*U+j73b@j@(sY&@6DQL{@{qZ)&6&s3oAVOq73fUeFP6Nu;%vr$I%cy=
z=~elu#4yfPY$#Ik^{+zHZ3?XU!&#1EB^-*B$a$uOmz4SQ<ZbIIT2xu8#rqH)`W<9_
z^_d>489Fp=qQ@dPJ@+UY@PfU?9)1Q?l4D!PQmcH!{^C^wrf{cPA8YoIi+DfCxrt<C
zHg$xB+nF8opBDdEQw<%?e7{v1tZ%DjF1QZoAF*cZro$Azzo9^f$tTtL%ip)OOo=7*
z<dpAKVK6;6)u%E$_mU2A<?K!Fr8aSm=K-FVPJ}Vr_XC-);cj#4s;B6O(>Do2_n$ly
z<m(Wyo3;J~14d6VFe}o)taE1YujHN^^0v!xdd`}$*L;$%Mec0%<kyIs8st6DU^KPd
zJxh#Olgt|1!+;_k_mkJrqm<`|H9RLwEaYP%<vHSx9(|wlHS<vqNgqAtk++AIYnW@z
z`gEF(yDXLHOWuy%!i@YKD%{wn#K`I@d}*daokvQX3{qn1QY8XLGe2oHeZ1^<MxA2z
z(l9<|9jNaw(xFZ#9ZL9b>3!xOHetqL17<=bt1#oHntomld*K@NXWe^&ymfD@!b1;c
zEpkTZIB$nJv+8(AiFkUOV(A}lN7gFB)yU*b&)4^4GFcnhOpe8t{Cgn-7k`wZP6clb
z*fZ@Q!`XQ<oV+E&kM8vUzTr#_c^kQ0f!(v^*uk3k>;NfV_{y;7wgek&$y=Em(e5(T
z{>vJ+h^####j@d2&U{EY124s)-xBPvBSog26t?xHctt+uwvwUkB^h+&t$CvqJu)OX
zE|s9s%@8C^58;lqV6>SSj7=87cq<Kp_SaD~*>M!7LIM%qA(*)x!C0RggkBLr_<l7A
zC(S`P+xrMA5{_Wc=wR+E3x<rm^==f5+1Wwdj~0LnP64>N*j>CPZ+}ozb|!BXW^WPC
zxgO^NZ!w6x9p?Z4dAlUrOMJQMC2T6aL`Qelw-da@&92_!_q3K`>I^sWxW1eCaMM++
z*yJi^MK=>uJ2erCtVZI+31@+{210YPf!J2*B(5xS63^Q^i3Sz*h4o)Yk=orsyu4;F
z#+mHJF=u-bdEQP;Y|C7WDR$z=VmonNZYNYn?8JruThVEEUE#Q!nHHfoVj15K>t!Q0
zjI$PV#@7_?F*U@M1~tTL@>Z5yT_lgHE>Z?n75UvQ#GceYuye13PkIH)f-10SdIe$=
zf1-#P$per5K*O|eh->)`=a0Tcz`qw5H}C~&E_#j>%@eqjw>AL}vBUKNN`vm<b=$j`
z8&rbTId|}N{vAwvaTDIFiqYuPHD>x3VDQX5Ox()dIi?&8CT|n&XJf{aY+!#D>w#SE
zO61H=5i^-OXCo*%AK#khW6*+ps3P*Q@)Gv}^v*+<A$bU$#GL@Oa`EVq8MXUlVPuai
z#JFdnXsQ`4+n8~VY&F(l?^a<(;Y2gOZ)3mqSqAK*ui)_ERMd1#!`aeQq*X9`B8t6M
z&bF>t$(=icQusPaq5tnJf)}4fwV{_FW1g{LMJis|aOZnb8peNR*5a>=sCoE2R!Pp{
zveQ|lZsHzm^45jCjiO)4Hz5IgXK`LHI}R;4?^$C*0+##zFFP@T^Ge)@dN&rm$=l7G
zcl|`(x_{yB(y#GQ9bnc3Gbct_v487D?U}r-Lf%@=PQ-uIl-II$9b~~w_$SBEi8;lc
z$y?vK36K`WBcx^=dOhOoi$x6XZi~hP&uEw*b4C{232=yWsIJ`M_9l}1AlTca55#{L
zeZA~KJ}`yjS6_NB3k-P3>vlNn)&o3e=NIx^|AiTAEm=F*heMFJKJR(vy|2KGJ39E+
zr{2q6tBo_Y%K-+sv%hF^;59l}i|EE=!b~a7t(0I@H3=TplA!5V`X!!7vAs}%xl%2n
zzS4&|mb<4{>R_wXVN$6M+gtJ4O^&Sp$a>L>wJ&R?mHS!4Wl?vopdQQqZ=G~%sCUUn
z_T2_iFa1Ql@ELcD-O3B22hNCH)D0J!jBr>;7WXm2k$prbX81(-M5E(C>fQ8t#qN({
z78YkT9E|{ZJCt?nz^f{>SsjkUclf<n%jU9H%R8;dU+AHxp6XSFnLcXPNMw0LsgZtI
z?)~hh!UN7G_B%&!ikA{^<|t8*ywqJ&!kM1X-;d=O&iZ1Fnb$)0)0&tSxSK^SkT<VK
z3iy#-C&}AzX6#sBkm17&B|c{<k@SFb77clgo~%aq-P9#VXp!Agi_Rmp$m^)XQ1W)b
zJo>5rr|T|ZU;GQbSFGt)c4Ke0wVre2I<!kM;8^1@EUGkM-zA=t7VDA0uM4K0FowJx
zxR~`CYqo+x8f;t2-7y_BIQ5Tbh5!u=Uo>b+jq}+GHI|blIpl5S0wu<6;C<uo{OY_4
zZ@X|84f~wce&``>rNjQAI?j#h@H&7y$$17-@NBdwLkpM9T3GS-9>5;Ls9YV!uG6EW
z4|M_Zwh4K=g}nW=T#vce^f*oZzWZncu9Lr)`q7)qnp<{UgF37k?_V~cW>5B|FB;&>
zbI^41_6T{~=@EOXF5IcQTaC_@p?G~F1a%*UAj%v<Js^}hz|0<`cCSd#!1=NYQ|KAm
z#>d~UocTtblt^E%#M=Ax+1}+Di07R+@>aS;iCvtnX}D6wbF2#1@oLQYtwEoCTKH2-
zE`tt@d4`JP`Ds=Y70z<@#`U-gTd$~bCYE~u)-g*ll)U9*RLPI0^A<#Y&nItv$@Wky
zB~+Yk{r8lf;i=3*^id<<M}bX8<T$ogj?`x|?C&JS!MW_MviBLz9;kC&dXL(%MxD#o
z$u1c(JIYaoudhvH;+<DAw7Vh0m^{{|<Z^R`1e3_y_>U6o_$8t4D?`dJDRXV5i0VTP
znXJw2B8ACK3dwT`YCib?z18Xl38rf#2qbT>@@sFlCcD>3*>jUHyGVj}Q6cC#Gz8`)
z!Du!h7*p#7BRDt+ub&;o_}hV;r8|P`8%MZX;t0C;2*mO6f#}#L5EBrH1oBqX;|R0I
zkKl=OAa;5LB9EW<qx}*1#RTB*Z+|3Ba2N1&7af|qi>~$EMdZL%qCne9T&-v=t{(Oh
z9@LGSU+@yA4|s`L*SR~s%uCFx#o3?s-omxHw>U!Hj_Btm=H74>i*C6JPxAHybE4X^
zrX6GLBGggN;-jOpa9~fi{FjqhsBjXO`qvl3ed>!x{*FTW&R#4@u@}Fz_CinID#_c<
zK6YaB1Uum|!%i5acH+@wJE1#kD`Lo7UotkK%vv-bUQ?XUu@b-ATZ_#@YKp2wR-&zK
z4bdafQamwN7slz;#k_%4MUsz&*xcwJez;Y_5crd?&k8t9sle9aSNL7+6N=XUfc9D`
zwqGd4x8tP{e_v3?dCrWH=g3t*fnwET4Dox2o!{=G(UE&lmfXhb!8dTx`v!fNx3QMI
zmA5E{mb^_Gm5-CR^N@8im)UpBpJ<hXHyPO|Uz`oGFAHZz<RbY@4t(>oFt8#MHS3tM
zo&D3g^Yc+a%{l2U^C^7u5Z*cur(5LWs%0+vlu#q?oduf?S?EpPUKne}3J)`aN-|)m
zm5Hw8?bOj`s5hB$zC$`r_u(#pZS*?LxeSX``kjWPV3CRSZ41up^x&Sm&gYTjdJZ?a
zH#4l(SyYuLq1qdIi)SU_Ms?25PUL>PhpCu;I2liBo<-^CGbrW0x){qdXlZi_2VNb=
zcwHj%rsDCE^YfbGI2>Ibhnve1(AFs)o0z#U;%yvONMd2(5{K0*<517{7%rzC!?cgw
zTeCC)nG=~E$N8N6L2>w2H;&p~EDm;KKJfHJ^dN85)SVw(K8AE=wI1Ls>|$!kMLF@v
zTM&&VoYB!I)04qj!_5t&5g!zVt{s`z_l&x+Bj0D=b}HvuMn$m?sR~C$9nSbXGa?|K
zUe@2NA3o`jN0vrK^O|v4jhksI?4p)b7$ZloZ&Ea-pW=*{g1L-ZIQ(M|YZmoi`k>55
zbeP88;-tUKmy;=BqP8{tuZ(kB?BP_A;QD`|upw{TT1(-=YuxAA8dRy^^?tMtPnYZP
zatG@u@;0ir9zX2$SV-P}FVn*I1bd^@Oc$Tk;SFo434h3%WIcA|>XAYgo?k>wdpEUT
z`f+EP!!UvTTtMD-qTi|O93yPV+wJrb1(L&WtMdOUIa5=^h*qmO<FTHepx*2oFH+*B
zEoT*TRPg(!z@Xi{-i|V$KtU~A#`8o^J-n!=I?~VdjAsc`B0W;q)sVB_8r_vMh#eIu
zS)xD>UnPFbVh*gGl6jB{_}ek7y+n>M&KMpdZ{6zBE83KG$Vz%zc5`0i9z89*w!RFI
zp+=|-En{WqT|!SMYnNFg<)}*De!Io`cRM-Jj<X@;?ZqKHqm*iq>q@<S3B6`~OoGYW
z2xmR6$LjH+vjNi2I;<~d9ecrm_Re8&W#4oyeNwMj>y2T37xhL1yA~Sk|Ea>ne$4f2
zt-<Db>>;<+p#5JpG_1AKr&ALTRpEl3Gn?dX)Fo!se^TI-wG!X<FgK9A-AI<W^7HpU
z=jYp#x9g}cU)J$#r-y?t{cnx*C_AIYf+JcqAI|eto)&F+Zd&ml&vWgmS6t@4^8ISe
zX{W}T<J35A=#feNL`whBGM*t+i>aCS)xwFd#j<P-7N;7pm}iG%zFp!;ZMts=tZYKK
zhdl(-WT6;!Jd{5AP=uO7Fo!cHZ!_qBJte_v)^J6GH7F!I$9_|w#&iYFUR2;gnF5jy
ztk>5nF`k?aCvWq4PO4PUW8}aZ_#Cs6+N;p(2sLI8<~LbT*I%zi!XGVkpX9JQLSFhY
zFZiAkV{)iZ?^D5TrV4x5LwG!b{-{xE-0sSqczhl=x=_~`ra(eF1(ePTgoG(j^^OAC
zGtBFM!QQA7^P5h{F`HiBXuc-uK9wM*K!TmDJ>TR>@t7W`p_c5cZk94nNQzxgrMSrc
zs>32VE>CB+1pB-r$y+<}HnExvdz(lx;%*4D`y{wU-oCKl$Eiu%QA>6aQZyuQ+qah@
z)m6%!P7++YEWu`-1oN~K*pjyecS6x<2tCFpLviLc`>c0@G5>rp3f1&P?FnY~MKCOT
zu}*ae=C0)+NQ=qaV}Y1KmPYpsz}2z-NciZ7GxhwTUg!_E|NObf*$*A8{5i)Tfcl>T
zF!4(O8XXUSBH17Mv;LTL%OBdB?&4P~cTu&8yJ*b(C(jf&p$~Q!tJ`{rB=UA(U2oxb
z$xG~|FRK4WPcf><gN*bNL+Mf8+00wmd2yfZUssVC?J8c{(G$fvpH2}jqI9r}xNPYn
zW*eQw?-~unrmIe3%s?k`wtamumA>OXT^t3<9K`oG_M%yWz39KsUKH=L6O$b5g!oWT
zr0%XKUQVhf8U@&i-IMIZ!8BWOrKhb}W^W@FeX$n%hSe0+Gp)pNzO|WSC90mb6rbzV
z5SQtb+J2;(c;AM5^=?{-tII6Jy?*a;<jEH-+VcTx=e@&}39s>X*DLJl@Cp}~zr>ut
zQe;R<v1>^we!Dz_`=0060Wx>ZBYGhpVdwD&SpDZN9>(6q=MCH)Fo^R#M~m?=r-*an
zMW}wQ2-6&k(0krh+zHIXGhgb))PjXu4%!ULM#Zx%%-BOO@y1MixJow8Fk^31GiOuj
zS$b%O<D7iBHqJ-bi#)V%k%s|x+zUXB_)DA_gV~o&B~R<oZ`|`$I@XNg*S9p|#MKNm
zU7L>2wyAh8xrFzLmteo-GNON6#>eu@&}%MnuTct;2d8jvV={(RW3E%zi}3P1&%L_m
zFr2&{O`a<KlQ7`v89eKngo(VBnv$@U|F?s?cdmR)f=k9JG^}<CS({Hn!yTVx+}%|W
z%zTNr@vz*_yx!6{NO~}r_edNDZiz!f?w+wCZ<jLzQQII6Q>JhhhxP4)OWbq*j(ReE
zRHyoKZ}qo$G+};|T>$rX9g9U?SLW7uk<%Yyv5Wqv*dK8iUU>|y_QgWs9!>946rOeF
zezsZ6GEf_Ft&DZR7$YiK$IWcb84l(r?G9x9(1JUr*$+E4fHN;`n7j8n49A|(=Sh#t
z0s3&hvF7>Qk~=&?G~9o~zSAStUqS)I%JJlj6vg}-H$Ii&LcSKR<a7I(yr%kcF3?hk
zIo$6xFP~X8e-+Siwy!Z~>U8rY2>ua@bqhjyRtZI53g-d#$<g_?3Oma=YdBnoCrfou
zQV(oH-nOJ}{D2HuX{|$*S6Z}Z%`|$g4%MS{c$A{Ufxpa*Jxf1urXI=vt8eBo3*at2
zyDwPZJfZ)Wyx#td`$30?VdDTJ`q7(8`2Js>vHnjOMsW^wdU+U}<Y5S8O<VOTeNe1r
zKF3Ir?Jvb9);zA{@5yrNj8UA;*-3xT7acA<(^4<i!jrec{p1I0K(UC~J?_+ry?KA>
z3q8+sNZe034%DSrl>hE%!<<;oVjOA29D#VwTF|fgaIXxjId}23NQU~Ja{dkd@A@jk
z_I@%XL~<4*MuyV`GUnrRwxSEaZ?+PXpD@psx^E{pEqao-3CFZ3<Kq@X-rh>&<9<Vn
zt+%Lc*4Lr)c^ww<tTK%rS_?924(E(YD(UM9*J1v39S%;?;Y%`mi!;@D9;?Dk^7dSg
z8vCyC=B+{Z+uT!1-u@+j(!VP3khfuE)WM%TKXBgUc?3PNySd{umAQVE+_P4~%s}$C
zWvmt!e4aY8AHVvs4u14bHKt|}?8BeeikfwI?u*-`W!4vW$#8CSKfnK;c+RvORAVx|
zv+2(C@^Fr&!a;%Az6!J*szBgu=GkYcaczqRD|f2U`Lhzg%Jg`8p1QN60d@L>ps{rb
zGeAPn%sCk8NkORoC<tSR2jeGwugyoZzD<(Aj=cTuErU-V6-InvuF)hpE>L^z%Nlor
zw-Qa(DB;WdXgWb35_<zt<ZY4-&y38F(6v^fds7uQhjNa!of5(PSZaL*j&hFFDq7AA
zQ1&&cS$MK0PMD@*M!O20<n#xgztV^EzOi>Z*-7<({n%gq$yf5WLlp&FS~J_wiP~><
zW;qR4AkSWbS-oV)<(YIY=VqQf<~bo#0w=O^QFp#J*~87*D?$1;35s~1_OV9oJ&EjX
zPWBZBA?`vDZrlsRz@CAyoEC_#^FuJQQwX+eLQwFXS&5ZWENAc5Q^VS}mlO*cNztmE
z6uX_s-Yf|Uzl0*qITU)UP}tTD#oK<NSh*+^kMyBf638<Rzu(|#L1<FyhXrwdkXZWz
zHT>b;>M)KjJ&b_ufp}gVfG#im@p)nZDo6NZ;T=D84faFU*~6T@_Cw>xe)w(SkE7ZE
zI2{eZl~jMoFZrYRqCe`L@#iiNe=IuHL>%kfLNpsqZ<DR7m~h8c)Gl=ut=2Q^IGH;I
zp1X;)qgskzSKP&k`|hHwytSx5&O;bgt;Oe+EyZGomSW8#SJCrL3t?;DLKL5&?>N#$
zEFb71Dyq4NPg-YTf4qUv-gFW{6P?78e)UC7m7}ok>L}dETf-IhX?xj=pP%eR)?PdD
z+TKpAdtXoZled24ZOA@5F}atWz%5&0%C9RP467wha<`toXHD_Xs=DY(f7C4__v%Gh
zigxyvB70_aQF~)Gar$aiaa3y|%;fE{?(cDV^*fy0^cE`yyn!cq+n-+KcgJ4B{=b*7
zZtx8L`<|e==M!XHdWtdQpW)5Xr#z=Tz=EL<pk>~%WhHmey57atx3`cx<_0FA7-Q0k
zu!Ov=mRp2t<n8u@%ur-q`ZhTS<5uTjN|PMU+GgWoTo%fAWuf$HCYI)zkwyLYSVJ>s
zBFy+E%fOF{D{!Q@$?xZ79GH}f+m>eRSZ78PlNr~wW*qKoMrJ$C`f!#fcAyy<Zf49*
z%|KpEI;s?3h9r<Vi%u!Hz9j|scc#EM?=t6^QZQ|IGX87AJph_yJo|eQJ{>M%8MBG&
zSDuA)z*!9WmxPKbN!XBk8nVKZxY8vFk)x8(cSRC5A54P1?-^`9!8xDzC%K#OBzls!
z;oP%*oqgIr%oeUf-fEu3VHS6oO(Ji*k+)`MH!Wel&fitB=*hW%lb$o7<ZXG%F(}B}
z`JEGB?!(<A?c*UD#X~<Z4&N8WVEajC;?#*{50tq)+E^^ejf4CmcLT)6Vc%5d50kgY
zJ8}Moyd9vTU-e5Ee90U2zc4JE5QZ{py|*^#k>wVK@ysh4MOIWgGkg3MIonzd$7*Wk
z=qWg7slsqAcYf4heyUE6wTGyM&SEax8Y!lqWFIF@hE%nb*^E+@kZ;Cm^bvE`<sHxb
z-M?sXtC)L6=*dl^p7qQ+6wc)B0`j)gtWa1@2xTU7C~_w<uZFyx<;UHeO>`JR-Y#0j
zYjrOj8aXqo&WT*DrGpc-+e^@)sxNzl<V8m<??W~EYR_>uikUMVDrVaq)L~Q}Juuf;
zr>5vogS_2H-Yy;z24|in-rH~nG=}xzaT)6UC&PTs3%+VB!{Gg42%Stn&vgSp-lpu7
zqM08(n=R=<;kAChw;JP5u=eV%MhCeLdh+(fWY${a=^@?29QV2!%zUT7cFsSJa%Iib
zN`dhf3i!U2qZ2)(WBy~!K`+aD*0ukuR~m-N5VTH)KD}l5K1GHnJh!xVkz<s-3{@J)
z(2TrY8!p3f@^%4vdx|U`+ev}sXeF+&H~ha{-Md;)PbP1#P}AHi(c+|%HSaa*w5%In
z@%;0XyHKskTd$X79&5^)oYB0vgZKRoH3#z8mXGnNzB=}7G+0Gn^sB8Z?BH9OS&gPw
z)ktf`-K6Zx{^oqAje=(mvR2RAAhPIng#wqzTTd0w5ZhJwdYN_aFM5Go>D#6M$3d>e
zqjTJyO5Rok@%+=6zH&Zp-wU*Oz`DHobJnN5wYXEpyj}9vHkCY6vL7tr9=E!z>+Cp#
zLJnDxxBbZ5T4NMw_<?hw>`}HHA;+8_3TD}}_G}S~F7&uw84v>ht|9Qf8jPVSLD(w`
zf^}#RuG|g67!eF7yI_1D6@qkTC|JZw;Ooir%zPPl_Q^1w-*0Gh&XJP00~;%u|E$FE
z&kAa-3jB-b+zXld<-Gznzvww)ul03J6^vV%_28jIkXnKMtrfV>^JhDg9Ot(ykb6Oa
zF*}sVJfOh$3f7zFsqs!`FG0@CU@dzuQ409*zJ2ne#xa`bP@XrF+?W~6nVaoZ<v2J%
zff#BVugE^xTk1i%^zq&&b1zD;`wMH_u4E#4JCMA+w47{IO5g$ZTPI2}p&9E;@^&|Q
z8%5s6_XtE~Q*yU?Adc}dYDC_;eUl)R{ogeiQk>!6u8uybGtN@{YREYqODV2~OK|9M
zD1yn`#^mjUI-yXLx7G_oF}HpIj+O;r3K{w96Km5;emH0J!`-)sas2#Yq_sGV4?hke
zu;LI(ejY;a4u>(I+F^{^c^J2D`@u`*2bIks=2{)XQS!DmYu@hH{IF2zhjIRXIK0CT
zUC#SMve_TiI{RbR$tGecdFw{rj?ZZ!;<SxKLH|bL)*)BX@uHike$7otd|Qgv3tEb?
zq3+^rS!+=(rL}O_?=Dt)yNTvDEyeb7SD}5<LUi_PE(YCdES876h&2OT#L!wU;$XP5
z=$zO<=w3UC_n}TAv~PXUwSl8(@ykK9+Ug*#QcGS@&t6P_ZzmR#tzRtcgiS#`;jpWo
zI61MNDBf--Vs6$I!?<TJ{8}AR-nEu^sIe9<oob4euGNHP&uYSIu%$TcXDNQot;C(T
zKQZx2IXWoIP_eKKdoFzAUi0_N1$l?Y<n6MtZ_t3e?a=fU(v2^1c)?5P8$ZLOvrjNS
z=`osad4xr258>VHDN;5*#^1FMU@W{3$7Q#0F6K7-Nw<(Zk^5$SilK`u!oZ=|Q1cD<
z>##OWCT}k|<zf=EBTAO$FrO?3AAGZs9g_wBomrTYpNVMJxBpn9W@KmJa?K3X>5&2N
z@++`Bl*&Dym(ihbDh>swV#mK!#D2Vj#~sZWO5J%@%M2XlEYHU7=?K1c1+T|)&+L~p
zR9};dar;@@T9UQzlF|0fMdXQ#m^U#Qi3gH#yM8iizrKLzzzbOBZ~-5-oWt~IXHjea
zSv>rkgxuQXYr+|9%{htt=eS#)`NWy^Paw&38nddLfsq-+RsJ&L|3w0<-f<sTi$r|i
z&E2qR@$kIO-DLDb`EYMe+#E7?S}ZcA$Kupc?lYgs9Joqmy&hne`tB%{50A#nqG&uk
zdkmwHfaBl8@cTg+(ktlmY{V?YGtAE05ruu!htq4v;-P^)D|0Mr6jM80z!^g7shyZD
zRB0K8-QNuK8_@&Qi!-UYVLbD5zJ^@#yrIHr>VSvBRX9NZ#;H5JhLkb`iM;K<NrA@?
z6lixwNx!oSJ!&#DmAnn#CdbltawO2tm^Moa?_?Qsnx*tKOR$C>e$P-|`#7^c$`}k`
zAB;Iwf>Am%1pftvB8pzd4Ur+JN{_|n$)PA6AIiLv5PV;!Lcg}$g<YYBZmI_TSckQE
zphY|K)`h&SN^W$Sp~Wm-!!xc@`=mc;?G_y~KDB79*P`J%ElQS?fBgSb<@~(ETKr(m
zSG<RPyfn^5rmJy(yBbp7uP(fwx4Lon^&RSI3pnG@MuzUx<-*vL>VK4Th`j!Uv2J(T
zE@hSx`^~I9b9tZFu|~aS%l&5bHjNm={JTswGoQ8aZlcA*Mck9Uojb+I+Xw8ywrEa0
zu#o}_6Xh5XE@#h#{Ws1|`q3vEohHX|@-}`ZXF96V^9dQ`lV#ke&KX9D42x{|J0<Y<
z@ZwoOCPPLPa}Dxk*zuVCyKEV@OqQdAEBA%g)8H-Vi+Y8q5Kj+I4F}e|6ExhBK;P{p
zW@D^neS1lbuyZ;rD4`C^$4JHBIe@(0L3Wznao4y)15npoengF~Q`l#`s^s&`&snX;
zJAUqnXzm@Sulkg`63^=>F@X$n%aI|SkKLX|ydMjB?(%1!dKY_)ZRNOG#<|e%a-3sb
zH}I$yu_yVQoS;6Esl}8wtig}!z=3@X%T*)fi3WYabHaD#46@#<HeP{=<n4U&*0GM9
zerp+C{No%8H37A&3<p-p;Lda2O<(2;a>nK4K{;wJ4aG&TPz0Rj>#cVXcBzAKF)0}N
zsjTe=9l?cqN6_t4Anp&LPp2+tRVIdTze6bO_Dc}Uex&U*_BZJP+SZIYNX~K$BySJ0
zZ~l9NlKYKW+a6P3#|oY&db953Ich3-yVX(&3)Z?{x+}4Udb;-wIr>oRk6$N;E&o2e
zLYY@cKT^Antbg|_V82X`J$wy-_b-XOEh2AEM=EfHXNUTG_%q4+sJ+ZcOl4*SnW_Cx
zil+uCGuNaTP2LWqpZ8*p1pkkvvyO}U{erz;2|+**MQrTu4(74DySo!xv58%F*$z4l
zEIK3<P(l%~#Y9X*6b0<oJHLDHANvxO+WNeoGiT16)>jRvdCCAOHRZ%d2Hfvtgj+`g
z_XZm<wzUCe<ZJL}HR>j-(cfQ<DYk0nQL7*ssKP|@Hj+1g@-{wDhx7LIM13=&Sq8c7
zXXFkjC9X75LGXJUv-D`bS&x0edcJ0Kn6q976|*NkT-RaT0R^+I6}a8g8%givXyqxx
zh@CRj=_*5Enirn<c;TLl7w6r)F!GWYUOe|gw1W(P{*%G>qZi&Ld(kuPg);^zMxB%6
z)Gax}p2=|}O^$tja`<`5IeQ_;W^%XRa&Nq+_MCc@*-pkLLiMwWm>_Q|wB+rVPEABa
zr$%CXcta6Tuc5fdeecN@O+@&BCgKx!!JAEND(ZZ264gywi=h^6#2WI}f|<zuzc`5}
z!A_!Ihvs7Di)NzmLqpNK+))%C;9TwX2IBXX1|nTnPh6{CUo;|bpHHtV9*wUnT-m2R
z^V3cY`(!6fKHG^oId)>eHan3w&Q27T+lsbNZNz3TYf;wLN=*4-D%{SSh_%~H#P&KS
z!igT_vb=wAAMy_-`+p;(-w*C5{)R8!U(x7HB}TQcfNsxQd|&?-2k{m^C%s|z`D^YD
zc#XwjWth?V6-M1I#i}LG(0=z*?h$^1>aY?FvwVyxSGeov^CJY8Gf!&$U9|3AME~J!
z%-D1b@vm>fx%Ew4jk^xhimRBPUVv~n?hmM+&wMb}w&d;9mRF!;4%A5UHttm}KBllP
zJ(Y_V<ZaFD9HiICVQ(ZG=fkpaTyg>4S(ymAnTc>Id$8o~n<rU#Qj&u`D|0ZdZ4NF-
zvvIOZHs<GLA$w95KE1z)`&%yH?#K)TPCJLlCTV!OIE}k~($IPa=Xsi?@w|5yk=Ezn
zeK-wvb<=QjPYRYkPe#)EWL$GQi+w+n&^|s1&8Qb=khfo5<IvTb`?Ja1ns1r2ew(|~
ze?(*KL(by7<!lap#lhWUP%<tWx2X}IynhDAYMntUd7Dq(rjoaBy}750nXR9Go@7rY
z0x8_ZvVpw4M&4!)V76Tm{hZ`v%++vs(O0`^C^MWqxUXdz_hWjW;2v$}uS@^8+vGGn
z9GOc<U#>bQ9LIKq<30VH-n{<4TOE#m9XKDsS~i2%zV+t8n0+J=S2-8Dh1czF4&+D?
zb59lYip=xJiy&V-FY&`qH)?C-q2`$%T&Q(s4)VbVa@Lb`@xP`T5lxNDs;&`=|BQ$_
ztK$q9uW8-bCp)Ue&MFPw{?p)VFV3>F1}SC#sQicy(=O|9yQ!X8Q+nKp(Lv3BlXT!r
z#3$~;oXh%)GZkCt+j+=--ie1noZAV4RfAxpj1PtxdAqs`8QPq4M7&=i0qmiYE&a&b
zsnmAQehR`euVAd=eCoq7oK4LPMlyMu#*Emntic<vH{v68;)-G;{?_NL!$~9F?>1s2
zuXDxIc&&S+r|*f^uZ`p_>jmd7ocG_wYib^^z2@w>O$<Q0wSm~2$DH>*oRfUbo+EiH
zdFqdq5+Ahs<AXQsz1_0&!!T-ty?Onee%lu%<9%`C32O-YOyW6@xRNzXF4=BmjnUV_
z2etP2;If)JG8wG*Fk;La_TyN8Ev2Sx$sXMDTlAsE`{2h6U-*_$4>-qOBA+*FZ2+cI
zTh-CCR24!!o7o6E!U8dly6#f)w)fdUOk-U>jh^C3?7z?3Ol`BDKfW&U$D<?cGg`Bk
zc*-AFUFm(w^~ZvIf8;I;#O#ZKh(5vG`!<|oYRtO*iXSY=m%B0_=6o~fugnLjWX3|d
zAI9|d<LnIc{wjUYqn0lWWcfZmA1bo8oPB0@zSes4H8*5*Fs?S_+zYj98~Td}<OiYN
zaesJba)%qwCfRrONS0B%WS!NNdgrhPJ~-aY2S;1@;Q3k~++?=GxJcG!w)DoH^TG6W
zdfqP`23=A?-BJS&e+^!8_Ot0C)^i<|81YSk)Rs!@@1ln6jt1>m_s&|bMI!6MXRPT=
z7SL-HM_r$+&9e2yaew9w7I02xBkOjaHUFNWk8imj*7*A3ksHq%)X>dLna@w&_GDk*
zi#~Jf8`Lbw%z!P--#^UUL(aSo()hxOwXYd__AW{6!H3e@o6X;se|G5Pi(+O*tc~zR
zA<t2|Z$6mU!idX!?TOw-$mmBh52at2yv?Y^{wjGpiu!K8)Brc$vdG@-#hhjJXMfsH
z%b8yTrZhLeN29^}3N@_B+dky2eH}F>j8`$!UWJHeD(Jt_3p|mJNpB@8`tmHoo6}Gw
z%zTtkTu@?gl@d31s}Pr{LVdnBkMps;_F0P~9dyvX*C3#Q8W-Zcaa6@y59-DhatzOr
zV;Ff`JU|A!0xuN#lc$@!;6BX@2R*%DTH=MSHDt*A?L`j(IlI9N=g8t4Tcykom!kuD
z8^jvdgS^ev$Wh=YL-Q7LoFs4Mi@jk%eLCz&6LDxavlM?e7JBE#VtPpf(Ke-l=oQ#N
zY%6XcoZGNA-QXw|1UZU_w;GBe4;zd1Gn<NlwoYPmUneo)9sl%n65S6tiMPL+hzB*A
zh}LF}MeyfFLd*P$*dj-9c!#6N9_S$cM%EKH`|5H3kF_}PlzEH;Y{Zq1cB0j+x<WzT
zwkL1h3T(yoSGMBGOgm9`yq#FgzmJ}4E825s{3h;<4<&C~k+&~hOhoLTn!@P-^<u9-
z5Ffu_j?*VBD5*w`qH1h=R*Ch0D{y{j1tMME(jURPR#A>yL(8%F{%gejrItLq4BcOp
zqM}hL-d=yoY?7xaj(Gxyz!D7lQ;f{KM=)Fc2==cZP#b!H<@bv)q2M;Pom&`i@+OKb
zZ{j$4`(LB0_|m%o^~u|+I{Dl`&HVwDm$5;88LMsb(76-q+Gn{aA#c~vNBm$iHRbYb
zq+iN{&A}{YFJ)m3c{?RN6E@+QxOFiTZ8luM&G;<*2q9-z<RJNPHthCiBZs`TxtYb>
zwJf-ix1O6X;3akBKa<ipFPw%O1JdAJoQf_RQ!t=3nVQ^Li2F%!UYW#QM@cAOk^<wi
zWGq^jj2U}5Q~M(cJ;>W}JLAzdDGmXn;_%^qEO({E!Y(fsb27<S?xF4~<;-hwH2&Ge
zpn^57w_P;mK8iwAbQJD!4~f<6)7UbP`$)*!3}&ON$DO86fjj7mA~1|QLXItEPWoW(
zfqTiEVDff1HP-3~dV>x#%a%EZ=JUCqW$FoRYkUHGnVBeC$DCu&Q&1H#zv5Cjs_KPO
z9}B^{`K)z*aQ9X=>mfC-b@X;lAd?=_Bl9$n^MQT2OUo(%`P9f}++~lByxl<F8dvyY
z>~Ypum3m%#cwL*p^EG*Usu5@OcJux2NG*8^&)}&B^sj5+tcs4AAUeeE(xO`>=kAvB
z+Loi?{D=mFhv@L+oDLnD>abVAId-yl1>e&;^0vll9TcHDETPxsfAe?$O!V0MQxA`F
z`j@HuHhU0+YvgUJ1NF~A!BEpD-MwQldd>>QYE>|NsiD@%U><~&If&%5dlhfwZuVSq
zwmWM*T_9Qv2}JS&dOCSdh?vbD7q4q^J&l+{KV~4g`foF}aB6J5>ESAKqrc|49)nsj
zgRZ_2vGj4e9N?Z5JAaHQr3Z%k`59_`lR`L`LGIk69^IdPwy0y&gAbCIkLeTTb+k^V
z4<>A3cF#{{22hh8(ASsQwnnVUHef_U15){C9rE^Qyb+(Ad=RvQdG6k<wdi9xvCD`i
z<iW?+M&wX;zLwAW`#L?S{25oKQcHctOrg5eAL9&2OXKXudjtJOzTD5unyVwdvDE>1
zeuj+o4a6Y>d+tVNRz-5}X-XhAtn|n0p8n`YUD)y<_p3ek$2l_3cZWY~A%9%&!&>g0
z5#=4Iw=QDs!#R<8)O3v<S-bMzY?k@pBpDdPTHA96>)XThSDRDkeak-dOCOBo8KW6z
z$J(t6#^`8zxpoD?c`09iBYaUsUs5S&UUtx9J$xW_gHV5bKgM49PUicv-_Z93Jvm;?
z@Z)RDc$K~#D{B6Ac-G*#sD>*aud8~Pun+OEhYs&+tFi693iJI{n3k);EK?2N8{`et
zsNGl%X?rD(eNiw2Rf(MbN_6P0!6eS9oI9z(kc%3K4xByl&?1npzhQyQd|*v|m%M#+
zg&9iZZPz>OPqT-*siq%fDqpN5Plt`+zSd>b)0x%(ky`q?=hQK-^L;|z8a8njXOAyD
z_#Rp8LOtBg2km&i`Vr-WCkh|L@jMmE*UU@ub^&W%l_&Ei*hA=h&WJCZw^>%lh_eq3
zctu^e5o^xAeC*qkx0A~ZxI^t|jE4a?_S0j`^X)j!u@1STM-I=jZO*C@a9V|QH_p7q
zs4?19joV?YX&b0;XpD*;b`{@eDs+5Lu69ylNFOC4$=Y7LEpetlc$$*=7fKYq<899W
zn5;r@gc^CT)rhC(sm4Tljt^__(OeBJd28?CjSAMGip_FNSRzNS9x{|(@xpswFHAk;
zg}jYks2J{r)rY;XqtJ_4uU>He=7nK1z2N2Og-o>+y|+j?XD!9Jl~Uxdmtsk{6!VTs
z(S9WBWupwePRj73s~lT?%HibHK=dBvC|dP*6obgw$eRtss^kXZazF!Ncd3E!Y0JDP
z^7adPJFBRnuxr&=JRI9p#I|-4b>=yVyU(1&2M;Ha?d~Y5mpTgN874KpbP$F(2QkIn
zK_rls`BluMFj|St1Fc1FQ!BBzmbH*^rf1dy8}WI*ow$CLeOj%Z81=(eoaH`(2p3z?
zc(b`!>SQju<l2g~dA35f+ER357ILE(CStSPM4TIBBJ}TTicOXO;2i%8b*;X_wecs+
z?pw`_w<_eTDw!+v4kkfwVXJ<Fsbk+@hO8W#A?3Wzmcgl{411@P;nDXOxccfj`d)ns
z`TD1@jeP>2`z84AU@`vqK0?oxk0333fF0!R%90|)*13%@8*X8i@g}5)Zy>V9HD>x1
zV06C%m~e+c+n6h8rn<uH)XP{UyNq#bFJs$S<~TLTMts9;sMqAMHqAu@dFz*)g)0NI
z;L`OXj_WSq-!SfE56grfdHZwj1@tYu2=xH&n=Q=3m(y8j|LGicic~bMN}<m&1>=4s
z<KyIHZ0ehWT1Z2is8rNmm4ZDZQt-V`3hGBEbFXzW#tle>*NP;3JDY^jLz8i`Fd47g
zCZlx6S<J9FOAlo{Bqec}e>x5l^0rN3EOLg$qVpbRJ!P@idW4>0w`fdw7lpg5X>($^
z<Fj5gGtarJIw}fRCon62%4z%|bM>rikCC?n#+*jCmnXPaF9QCLxaVwh1SV)Xb9jLB
zg>~33>=nYyE6&wNhN9{nXJGqtzbLa^M_=IXoYmZ4+J?K%51xXKeE!E-&~=WS2O)3s
z$)z>yqYh+`ZjEOUb{`Ky&F~=H=Un~E5A<$)=Jl_30D2d4UX)tk#|mCEsF5lAv$quO
zgFZ3zXRzkWxvr<Kt4CXEQQvnNux*h6kqZr+eKx?{!hl=UsYa8xvaMPy`lG@8B^r#(
z(%@5)28|xEk4FDW3;G)OQlr}Dp+g_u8hPq4<CqSfhqS1ku0>-DEh@>|&E)MT&K?Ys
zuz$rG=ht9n<NPK=3fNP;L<X~OnJ}FhOOjxO_oAmcEr@$_g0O<#*yZ~;`#6B}spRdT
zM$Bs>Ys|^pBy#ox&j!kaoIx@%;Ab}@UNK9@WF%`zX5QSqz*!0E+pVZko(j{WPHO|6
zn9`F%MtWBoq5Q<JCvT=D_@J77-!nt_*tBK+Nj<s&d7H&+>F2FJoS*hVBld>-xbycO
z!)%@pKG;Ow?x0s}Z4(`4GZSDx&o7_Imw;gA#`5<{CBGc$k$Sz$2g9xyVb41J19Jz4
zvF0(Qw)`lHb@wGB&QM1@I$n>UjXLIq>5vv+z!G2f^Zaz^%g6uFReEB0yU6p*eOn#2
zDmc5t`JKKSm`U%#tRpAx6Q9o=Y0Wr$LT$LtF3x-AP@5+6o^0WswI-}JsZ;)XXvDh*
zM!e)1#gE=sYu4(iyzl*+@I1iBqtcFN3-U6Gua9{Tn9pbCOOLb<ZawqC_%t8-psCGU
z7*WVGQ`Q(G;%8Eqr{3%^mwn3(^zSCq)7+kC4f=CtW%0Z~h6MW1$8?LYvp#xk9IVG0
zqaI5F4H(HBz*RpC_{1}w1$k?|NKf5NhqdHw%sUn5#>kZODx9vVfgO1}Zkig07ASGb
zfwu|;ECZRFz#6xuL5cKHDr}3;AfNLpX_u)vM=)Q3yv-nQ+Z<rl;36MnSo>lOd#j_I
zcpjmaJ&e5V&gZoFFMVs|?e?YAnGM`!bJrJ-S<_Em$C`Em=WjOpVlR2So4mciv&DXD
z!h@-y?26=Tz{>}>*RoFM>*a*Xh~sT}o?^|~i#5D4g&rk|0jv4<lMncOzN7bPI(@=}
zj5uh<9_uU4%%&OeXC-yz?L6D2>Cj*dXI1F~E^n*BKBF2Ar>F(*(7=Zu%YM-xbyCIN
z8|PllRj~i6#0J*+b!IE!_*a44{oWWsPtv4bN{C(x6f9BT$q^;b97>Gm=j>dlgg#RR
z4~ZICd(@bjp~jfTtTQLbF!i<!CLwZcbYo3Q-j*$v<8?<FPA0K__4C4;gI>6JObSDy
z6q)?^Q@R&-SV_>)S_<>pUU-)z#TXa<|7PB1khjaFSVi8xA%C-vNa1Mag|Vl+*i)0?
zb5A+Gvi7ZQO%Kx_2l4KWgE-lyfr#aN>@xlt+RH&iKd&zyy4MpmeCi3?kq*LUVgqqW
z)j*72TTe7)F2zY#d+}+%y}03KFFqGC3+lPO(EYU+H-FoUuZ``+i59lv;2dkAI&39E
zms^RnF;-&8@Y*6|nUzqnR!#n4Ew(kUBl@S?i8m|l#GUW9;?g-=VMpGE95xrz$=iPy
z&BQ&0nHV|Cj6N$<F+ATytT|*NM)fukL!Q?ZSANvs9MLbB|Na862OqJzM>QhqR3U3@
zC4N4Ahv3XN2s_1oZS8WH%zll!n@W+YD1~H1DctLq;!FMuT+=<rZp}0FU;PwwE1%%$
z#u9wkSPUnZN2nP22tm*9A++Ni%<;L6m9=l<)aG08m)(R7cL2mZE~IwBe2Vk=sBq4Q
zsreQ94fC)lHka9(Iq+dF#hA(*Oft*H^x$mtT%7|ya(A467FHh4LRRxEBy_%rdgSfL
zHkq&w&HTUnX5T)~#JjwUi0VMy_#=1DYA&Gdmoy9yPsOX6skk^T1;?tBv2s>2UOq|U
z^)nSWCX&VHl95Q(PV1b43*!^<y+tB2oDxy~C=tUxBw>I5WVAe-j3DxMJkG*yUp&@Q
zOJ186hscmPta!(rvaD-uIZNA&*~P|)XlAHKW5MBQgpjvk=eWm<{->uNoVo3M2H#$s
zMoalgTnS*7e=u{2JD!G-y!|)oG(woG_#rz2pB_eFFf($dEIx|Y^dEh(34z5m&QaKL
zu6`+JJ?ICDa)>~+fjh@mFypst1oqOSG@pLbhRpI?u{oTc?r<Ef8G^rp8N96d{ze4B
z;Sl}8VL{yQ#9EN^AC3EPH}x6nlGKqW_VtIisUO-<tGq=0b3D)cyXf7C;kEyA8$H}P
zXE&yv9^208(FJ-q?qyG>y#eaM)UwPC_#37}OB)>~Zq#BddFwe(LywvUa}rtW-qpa|
zku&cGEe5)9CSf+~U3VQ^$lM3sI?nHC;WAB&H9W63UZFut`W_FD*KpS;zm~PlKlouV
zc{?bEUe6WuEy)65d7fE`{n;NRZ$p`Nc-)P3@7F-2vk%#pISu&VIfK{Fu2ZPzv0i&h
z?OW7iExFf#kE~xW^Y`eY)*<Ybj#>|Wn|6BCpRC8~iF!b-N4Mby>}I}e$Y>+yvW!?t
z-o`g)&y||)K-RS?8}?dnv0m134l=_4J2JZ=Yro~3!Ppw)L+!=~HAfh!Tk4rpOMRdw
zd(|ea<L9xi%G1MaBz3}z<UX&fmHiBuoX6S9>&&kuR|k``)eVf)SPdA;-(gn(nV(KJ
z<uRvcf*$HsI@pl6vDE8bdB&MZ-YyPce~qji&YK%~JJ^CerXMvhRfm0@^(f-GXx3K)
zg4_LXcN%kZ>2s=>%G#2e=*)OOn6md+hkS_YLf=h_9zFQ`Hu<E*E_#r*@%2!~`&_#r
z?^{pSVRv{AAa7+cthdP`U1QFhP@7%PkK6M+(wjBgU~?l{zcAo4dE0dc{jWT8%;nti
ziE^Gpn)+bFE<MK3S8Gn*-edi?af<=%+2i+WrH9RGW+y~x==tSya$b)%-{}j|aRz0s
z7CFhxP0(stPiZjFmt65tp*l^4k*!r27@|ZSUnPQ7O4L><m{X#FsaAoyX$oe(D}eI~
z<lj@m#8r)Y>~&r|uSWTMEjIJ~*vQofyN1!T{KbHd*?P2()1y~^dd8_|MBCCQ?qWpN
zR^~g9iM!eu(M>^*AYT&`SF>l!=Y24FYxcwkzu9k}s-mVb+kpRAt4~?L*OAl*O<MZk
zFCV8Oew{PFeqajy#^3ooryAKmU{983j;Z|KditZL@HH34$9aAYBc^6@=2gS{KZG7H
zdZYgE`8-74I+M4R4_J4GsS$Qmh3#Kiza}bSRZoTS4OLibrh*|uiA0_~hQ3hX@IVDB
z6WDhpZzuNh#=Ayxe0(5-{G|+6TPiS&9%A`tZ}tV`IGrLxc9;wei=`+yFJ<n#7jwD2
zkaJ3g<i1|0W}W)b)(c}^Nb&rz6uE9vYLil!osm-Kk)pp*3fFE@bp0WL+dB!anMqM-
zF2&CgQfyx&#gt9_zdo#Se@RfULW1=lCD8KUUzmC!JIf0fj>s^$o1E7}If`mE5ZYf3
zVztacq)u=U$2vQRzU>@@dp!qn^jUrJ;<3FLqp%mQoRb~&&t5!vWiLAXBx}#wi{I;*
zQ$gN(leZ(r)AzKhuGqZGPB7(MOsi!lJfGQ$b8D@|&%IV++9vMG8)qeQ`_>i{8d`}B
zMoVU1Sc{HJteNk~-L<Rigjs~Sm~zBcWO&$!$(w9ML7cgG<6|Zg4w?z)pC;mAvWXb6
z(?oo}UsFsqr%!5nO%eR#AIi4=L;8U)oPYa>Cgg35O%--euEeYL@364o4RU6c!~Vi+
z^m<hWvyG(~*_Zv=_?P&4{RQT^zQFJ8&#~h0|Kn}Qv?r)NsRT~zim|-$BlKH!A66yz
z&?>kH&B@z8VK=bK?*?XnzK(vljzjye;WO*npC{O_?RNz|{&8o|g3D->mWwB^a-g@m
zggNW7Sp#R0DcLwMISYmCy*{16+IDvqY8*=E_fE&Ed336pagHZAlbI2jaDSSK+}#(^
z%_a+_<ZagTG(;^pi+Xj=qLTBn>pCQ1@vTH`jZVaztR!?zOMv@_ID}@zA>EA3{g8}i
z<n7?*iTKYU5lQMqw6{#42Zes9O>yww8ixvgJZNh?e5fUd>ACxEMI5SL$KorqqTVrk
zV$i2(c!oy9R~L=GuFQD)!^{cmNGzjHth>ftC7j>+@{0S%CUDORch))CpF$1xY)d<F
z7eKv}IFWw>+Be+QIWP>$%baaF9fI**A-qqVMPx7Rm_8Vu&4V%4g8O0w^IQjqVfLJ3
zusRlohev}kvqKPE*>fun4T2AOyFnF(0B_E)P~Y3gy3asw>RrzD7_RxF0evZR=$#QW
z{4w?pb<&p9BCm3uVK&d;Q|Zx>`=H8#wTn)V>8xva)8`cToV_yoiFY`&&OM>W(<TN?
z>(9COU_B(iSy$C%Hu)UppN4CwO=<8vg}t<R*0(n`xLaQfV=FBpo@wAeNrw^B___0R
z7}-dNs@oda@bfa5EAwkP>lE^K4tYD7ymhGSgSmmssIwv$A_E}(FAyKRI1_q-**hb6
zhG-v%-IJJcc!!zq{g?y)i~b&E0752l*EV_UHj4evNY<(2>F?sTG@A9p7QGIuA8XlD
z=CwOXi_lZd#G$wA+z$;7+LJS5^q500rjk8Y_W|s=Ze;dcrV(8nShtSl3>2@sZFn=|
z^-)TOd8M;HdCXdsY<HxV{Cg5>JwASih8eK5pB}z5bokSbk6(@!ms)D^mvw!eGdk>I
z?|MrLJuLLSgyr%)K;BldrukOO2!9jym09=Knxn(Z%{sJuq=Us~J>RQ(WYo~1=YH17
z2^uVH#IwvUHHNKJ!(^5kMGMs!zFLiu%<1XDI?#gG=b2sg^i=4%PnL5eAB{LQfIGj}
z(}&cBp5w*z@X)U#o#}&FH+h{WZ#SRib4=duCfnZr)8gGy_Ed}Lp`o6!iZ$!*tIV{Y
zWQ3Ks0rTb>ah)}q+br&o;eCI~eyEQ>J!hr#&mA`)lUjF0QzLf-@U=xI|Hs<RYYVwf
zzufT(vaXn3rXmf#_-VMSNdv_=Eu<@`gYY#}KUR%echzXibK9`58idW(qFAQIAB7f&
z$=kDJZ5@LOfAuO1C2w;ASi_OGf#hum7v?2MnNQ@VK*n(eYz8Rk?<G^n<n_~3=<`wy
z$9Oe%EYQGY1RuvEYJ`@l;rC9BhV7_1*U+HjYI1ys3bibK@RPcF+$ICS=iH2Sxf$#J
zHA&2iFlTReF?+x|&c3pj@Q}QHyvqlR$n14{_*$~&8FendhG*9~JUf2>&0ICk04I>c
zv;Xk9<k`=90(FZVdIs(rm^W>}H$DeGfBAZ%mVA@1tqAf~&+lVT58{Q18Wb=~Vp=(~
zBrYl8b6SbN=co;T;r&h^Uu!72pHzXtixqggoq92OdxH1-7g?$#Z;$Nu#>r!HJRT~?
zyt^_yCR;VdGCX`LL&pjkMmx#TJzt8`)ObJSND+`G#a!~XaIzG`|4QImM~aWbrOZ~9
zLb6+m8~db4BX83~Bp8w^frOve+FgnXU8FF6mmrlq&HX1qk%<&FLCu(4_1h@LBJ#Ko
zdHemX1am$}=%1Cq<eLOttEEUF+hwM5?0zW6V6z6|<MR6A<?Q;xdj*+G-dd5jd+Is}
zMR9$xy4YSU^0pU!j@k>GSbOp2g}s>d!(OZ)Z=bET7uUMliwy^v{}f>-#<aE*p?hq_
z{g2inAeOrEPS&;H?)P0*oP)Iz&ZsSZ_29m|LQB!>u%)P4YbgRcScvv_&Bd3q=3*^*
zTYtB?NbJN+#<phSX)iNzLS-f#N1KUyUrmG;dE07_iCA#Arl?@=_F%u7VnH$Y5zP69
zYbQQKb>So4bgjnA-1lfQ^F7)%s=&Qo<!~5Nj$Mhbk@md|o-4@Ph?h_@uQ7nUHQoOL
z_uQW2Wn~HOt}H=*BeJ$-F%)x(;S%={XO`T@*T?tJd+jwmaON)fPuDm<aRYUtuH#nE
z>saM>4Zri4O`*PmL%pvsmno0u>s;JukV_8>J;kP3aQ$-;`yOQ>pR8TU+h_9j&EpI_
zImUcq&hHE-Z~x56K!?8>h~1mXURox1<z=E>GX2WY=i&4@4F|~E6Xfj|o3q$-GYRtx
z5@G3`h{utMc-SuqSLY<K&lrbf?qh#7ECIQ>30T=D5xW~F!p$-fhc?H-B`XFolj87V
zeH_wU;$ZeZ77Msf_RfM>X5GhPOP4t28PTI0!%Pe28DFpF&Vj&a3})|k4)cn~g>VPv
z$&={L{WxD2atA=nF~rD@VPp^PSIIevGl!VnyX+($lDA8lDfm6>1Qw=npC@_yDkqd#
z)Xa_N9I5_KFkH_BW6A-}*wBkoa~W&7{`3vCV1D`&_ArgywQ9@#nDneJYZnOj<AG?H
z#5qOkt6yxG%cl)O@cKY(+Zlv(@>WXT7Lu<^T61P)IP2Oj%r~bNHXxIp9qL_^1-*;a
zoR5v7uF6`keM{<l!R)V|)S}5$9m<;NG2}SsdESsqKlIeK>G}Mj!wzQ5*ifrWp?9TG
z8x4l;)4+F+1{PN|;D2D617`!NEsx>6!X@%8qK6hL)_h)^-FR?}HElMpadkCV#Vk0-
zAsYPmfy|xG>(gN)YW$%GmVO{p3B5rY`d{g*9_$r>*@APQoMpPOBmhc#=5cVIb6GV1
zeJtyZ3r6$;v)6em>}NzSuV1mDdRVcxHr&x7@3Iy()73C-tU;(d8UJ0w9l}~*H}%=k
zdQ?*5tLRVtoV=ZMm$Md&jR<AErX*`W%%lFy8dyf1^S@NiX;6=T#_Q=2DYFz<mrmgC
z9m|?T`=1`A({(uRqyuhfpuNRD;|Lx4^LOyl>d=tf-NrM_0w30P%mG<j%xijq0TVaq
zadL+iTUqP19jHU-10DYH-x3OR7)#!o@LW?6t3i50&b^Sg{m9#kGt`(kS&c+KcEhR7
zkFCSceXYaX3Oz!2jhB)YVb%Ye8S%w~70lWw;;ajKo6p{3=OIQUChGBi1$B&*tiSKj
z^TxACVPh?Ol{&m;uYC|1VaA%Z&pg)c4rI<&@*<PxrN8u;^EGpAH2ag($9HI$HOTuB
zJ>Gy4@;1hX+Pxz^$MdN{=jh>cK#%IF<hh#`i)(7}kU4*^LN)OA)?gFQJZF<tXgElX
z`O#|Z&8J>U-E`zD_9)2P{^acq-kOrPon$I(nXbYQ>cmy0O1#-Y#@_eFhQHo8vsQsC
z+sInpUg{NC@1(%=`wDCdVhv2q)a5p7I_4|<o~}W}a1Cm>sBk;W9U1%F@V&Jg8s9vG
zdGlGvC3s;3eX{FXsZr6Lb4FHLD1K_tgc@#{g@Jw`o(Fl3xXg3JSL)`|zj1bF1hwKA
zJzUs3Nv*}q4L)v;wK${B_fl;>AGi5l@*T?_KIh_2mhdcN$H!?D_vf%5pFquH!7=7c
z7!8;|iD$WQdJN5HkAttHMJ`&@cBcREu^N-f+ezf@rxQxtisOBMrNom+CDsn0j?VMN
z2I|D&)(Y$=XHN`K;B2)VZgb@@`A?20vbN0-dYx{`U=<_7oirJy-;lwPynSjZhc($N
z50la>CdK<yDKmAY$eGCclx!UM(*xa$J@9Fp2f|moqsKOPC`L$dHk527Z+A|Y!eg%#
zw$CMKQYk^wYYFzglwb;(dw7r(Tgluh)R05F^7DR5AT5```@IC4@+D|M?!M_SMaL*9
zQkr>TO_3Z{<gM@A`r;9JJD@l76_u=E*Vv2u{`Nxg-Ci_$WG~w9vlplI+$RvuzhB|J
z>_vN#Vz3u-e%^|9_M(_`vwNP{i5K>DM8o4YVp|s*v7NlN8EPet^|BIiZ7sz@&hXsx
zC1X3+7K1KYikj;!Mf=Sb;%UcPV%$@65w*%(2yZiyv&BsGB5x&a%tT>#GjV>GnQ$3p
zCO&(bi0b|(V(;miqL;j;aQRkK{5$&=je|a6QKL^7KeY-IzE#1*x*C0o-Xnj;d-Qwu
zmh)WY%*=d^;@z)M*7_BiEH6dwju+@`c#51~Pw->?6SN=u1h3;tkV)P?nEV(;>PP4`
zs~F`A9-^Jw1B4B^jzn<{-TGgH$&CWsN#@*7Px_;V6>vW13U{vMA>imGTqSRhf6c@B
zxLn-n!c61b9E|;Pk>|dPX#V&D^6W04zd94O%QJ9eS_W1fO^3d9I;^**qhpT@O#PO@
zx-1hjmu0dx%0%DzOe7yUkKS9;*i$@<-RViV5uF4yO~UXCiHMYvr{wKU(<DUCOu)i7
zap*8L4(moHp!1ak4DXnTLFS2QI3oeC0^`_=<xY0yOdS0$8e)4iLVrfXxkoIrX2k+=
zv2bY}hyLWP9eMk2Sq$V=(OBUdjgfolOYY1%*EI@Qe+oV#0{IOi&^h`TCU_i!E4ARH
z@kdcs9ge+m%#`K69Qjl3Y1?%aTV%)K(())u$yyuEn@H#ddUc(gyb_G%Nz50dCsjwr
ztf7ZAioD%JKa|)Mj3&Q>ps`?I_B-eCj|OsHoE}s1b^>{O_CL;GFl(^KP0kk9=gfyQ
zcYN?Heyg7^-)m$9y_5J~?}?tFt9;L^GxV6xYqUH<kJ#ayndN)klGnqn$9b*e&aB(y
z$3*rxH<Z!y^p1O?$y>iqI-Fd@`faNQOPw^()z{!2S^Hsw25$m2_)C^uVJ&kiRf9Zo
z)s}lihIM5Q!A=c&&Cp;XzyB@PF_&yLDEh8O^dB{DyklK6jn|#J^a@8YhlyU?6wd%W
zVg0)2Ff;#|YxK5z0P_-=&s54RhC|Gw)%v53mibK`I4jJW)RLa4L2p%P_LMUqE%f-y
zy6w;#9X)XTJ$tF~<`i$N5%=)-zVbl>$BSCnZDSoz-d<vDVNp%JlQrioTW0voF;Z7&
z&Yg>adqT;^0nFg(O8-g{bFx_fHRRtPk2Inr$bduC6U+GX2K>_@FPSx*L4(@`8cZ0)
z9QQI6o-ohu6nQa|IkJz|>oAZy!ro}=1gv$ZlidgC3u?pbtNA7k46I|PebS&k>&!_v
znVm%5UZDQI>Nw9Qb{fnhZ)=mcl5T3aOjTq1T-LlV_%%nhSiY5+OYFbv;td%7kk|Oq
z|LtYrUK#f3E$I1~w3ggwU->5M^vNE2yk5fRme2bbGIr?+E!5Up&TweqR)_a*49_3b
zv>!KTuaP=$j)uFz`}2AHqvIX`_8Oh3qf-xf$j1!)S!>4`Fl>ndnpNadxE?M2^)L<M
zYm2;X#y<2p{_HfqHs-D8Ipi93(d%mLPEaG29@(h=YV3FCOv?~80y}9?U7|twCmE{R
z$#HfIXGT)w%)?Tm3t8K%o)TkXlsK|pftp#~n6*`oW%K1YA0)>!X5Js8rrfBh0zseU
zxLxjzk_F^!o&v4!DKN^%8~5miZ8O*#$MZZ;RNxL}>0#{mJcPOJ4&%ZEYMwPzxEjqG
z`kxxb_I!=3Rip256+V7ufAk;w>|}0AJ0l!xGZT@I<K@?!(K)S$VGI2X=a}K&K#SBM
z4YqC4!X!hB`Fvg^kC}JH_tLIV>g4h44Uo4#V|g~=^K>(eeQ(yZ=O$BMA#Xw6hVec4
zosabtJ{CvF+r2f^STsNdmA?|D5^Ci1I@NYn;x)Zt-;4^>pGW4pd1K=zZ#WEA;5uhq
zkIj_hJ^jRMSCFeNGHg;Yo5D?o-Oe)P43?qOB`HRdw>ISMucK0&pCE;mT#BsOQgoig
zdi9M2Yu<RE_GS;%^7X(t@>agw9o^@<!#LX=8yCC7KGq$_e|w@t%37CN@*och>p2NZ
zS<4>!|GeF`R*EC`QusfU;7gf=yS^mMWTo~@o=ciau_Hu^q{d!|JK+V(9(Lm95<78r
zwVeo~Pq<_e_us9s7ae_=ugKh|f>q2|<gHL=FQ$dsiw69+-s$#2Lf&SPw}nmZMFM$S
zx5Q2i?omg~3bzqKTdal8I}0%*+(M*(un;YpTZ*2os2@+X6e~K`7PY@yijH=c!s1sg
z(e#hGsJw43KJ_&hCR>>iwb@K;Z(t_+khhu^X5zzkQ&F(kM4b9iQ|u0{DK3W86rZ=(
z6ccmGQ5E|Zp_&TLHoe2%ib`~UQib9LRmjMwz)Q=w*jZ7IE2`IcZuc6)tY2Zo+Lu_b
zea;!t7cl$&gjq0;ahbgRJ?07a#yrHOo%gY1_FZ)SQ-r)#MVLPKA?iH7#{B@-;Qiq$
z-mNc$)tUlKCvQK^%Ez><eDrB}1*f@3;J?sInAVH?Y1z9iF<ge*p&Z<4k%KJvi?A)s
zz>jSgU_$@#tZCGV$=e14GO+S#8pgCrN5A#yXl~A#o(J6J-h%$*rI`ry%f!RZ=_vbk
z9v>E_!Qy%XK9(e){+k4#ECH(530UKvh+iiXnE{=M8)WK)b8*P&8HedL;y5ElZoWvs
zYqEF3q8L1x9s|n_(YRU_1sls~Jm9XKKcAyv(JC5QAEMCrWGwRR<IqtX!<oVu99kZO
z*=5n-lo1MCqVXg=9KFi8kA)cu*I3V9ByYoq9zzOqoZ2lrhVQeELeumJOyk2bSRIav
zJK=crhx@s?uY2h<?mr`MkJI;)md>3aM}u&OHLI^fAexc4#RHhTx;PNem~%UfGcr3l
zdr|JleJ+y%snG>sCTrG>cAV>|<qvb-1{C_?@&nEhX8B|EH_p=cWM2MyAMEJvi(vNK
z-ctMQw3>OUtUqkEdR&m}F>nF*JyY{rP7S$~*Nc0+UYPNg(U!SA>>nLt4X9;pQ@@bc
zPV!dsk@Xk#U`e(bI$o1BW*W{Xt06jaFPBb(z2t4q1J=Ir%z6vpwZKOK?MxNAuF_!i
z9W~m}>wy3@K3ZrnyGV_cDm4@r)u{HQ2V}b%cgVyjvUax%vmOLJwaoTeLtd1Gat87$
zcL0#Lc3T4QPU(-yotaZW-%I=M)cSfE5kOy!<8~D~nJ}M~T4T#z?D>&-R^+W8dwHE4
zsBweW!!&wWExA{W*Foht_Aq-GP)=sd|3QB!dHZ)3y)ERe`xbh|`WWz%8o=F1o=Xx9
z%-N-<dX@&(!?dU~T#GnpQLhoRJRVXzexpGdAD7(OYW#bk#?Vp~vXkUk^gsfq*%GM8
z+hG~(yZxm`_@5eQSa)5jt;Cc(_Tx75{K3cUBF_R(e$ngwO@rMZH3({{<^Egd2$`@R
zbW<aCfts^ByxmdZel9&TFX_cip#OG0_n{5oY)1zJe5q64&!_j*j&nROeK9<U`B3a*
z-=(G>%ejw~ncS;3T8Efa>gpRc*skI87RTq5^_=E|4iD*Ld&ZjeaT9uYSPPH(!#bJ_
zn7oZOuro6gIveq}D|PUD>}A7%sz5#L$lKhN2F#?caLkXdEtL)ttnKAd%sHa&eQulz
zdpWN%io9J+-cBKJ_xLL?BSC>TmP)i>uHU`was+;s;q?dBV{PPkO5T3Dz!^=}ozJE#
z5o)i*xCP$0J<A(Qr+B01S~>pClOu=O3c7aQ%sBMM<a*xd#Tl0DH{R&mMFFcn5{xGY
zx|6rLx7_j6^$<L29m3FxgNWXH2z{3vf?a|uhG)6r%O_W)9pA^SXB7rNSD`U~@0)5B
z>cy#W&Y79O^Hh-W@2WQ{?&MX&;H*V^o=xxe<N1`jO*r2(lX>p!T8C%IAAD|Zs}Ryn
zhZoczeg&~sA5Aae4h>eQG`MO_2HWxcQcs6hAGPSqzQ-dewW9lcjq>~TXrRHd95tH$
z<If+j!ZEcH0kf6ZOHY{HO$8*b^h70k(+BL0LRV%qQ9o{TnD_sxoO|>5cubN}r;*|B
zX8M5l(U){chBpIcnDEjI3As`%Unb>T9sNtxea(hRahDqJ`g93?ZkOQESP5M2yP=oH
z4GtIF@bH@(YAkR^<UDs~Te)KddAlUc9q)?Vv8t5^?g|h31|_J<ks#zE*?OJdgRl9)
zWNcntDISxt#TTe4M@f*S=5s;bZgG;L{wyixw3nhyiG=y^oWtp0Czg}95#((T^7h(e
zJ5fc?uw8Clv2p@?v(ve+r!M!?y4Z`gQTF0sioJMG-d1j^E84uW6Fssx$8*k3bX;I3
zhBmYjqxRPp<H_64VHTo$wS{mXZ|DDyx1%hDRAM1o?z9lW-7UoTpS6Vlb#rm+uc?qf
zHWlVM+*zAuD#q4eecRehJZWquQn#6kQ!7nG_sp7Nd(E%N)|VsVLpeq_EyvBoa*PUo
z3sq$WRF;)EwYv)D8I`ELq5^l!-y%<1jwpWK(VDL@+WZyfSHDDW!*lfO^c+7kOW+;-
z7&SbKk(Kobb(=lJGxB!X^CBEfC}M7U5kf{4A@JIDyzhPuM{it3y^V#)EGxj7<M{|0
zpO4hIJT$PSAF6LIuJ~U<lm3^Oea2k^CE2J9%*K?SnYh84)#6G9N?n*=k(r6RQ!}A`
zmx1hF8E_0t!@47>m~EYo!%Nd~cWVZ29?D?eN+ue*WWsjFIrLSeA$~vtmR(AKg1jv%
zO~3&1b{%=!-i;n8@^-f}0c)>vr%v~1#5IXV3-Wf&C)TJwoaea^16Pk2JR)yx7Dl7`
zVHD=2MsW{V6h5&h8$#YH%c7tRi$&Mm7{m{XL2d|r$K#p3Xd8)J?ILk|Q#3k9aQ1ag
z7&@ee;zQL@jIPC9W}NSN*6|oh7ac=?@^*fMBlKm2GoO-q!}K3Va$k$&GqZfx9>G3l
zyGH)uJmgh>JP!&&K6$%sCHDnTQ`OrAAi_Ej*VxazFJ(`YIRH&60ueKYej%wpLO9p)
z+14MkOPGa6CbiD@!^JE=)KK$d&J`Y^e*B7A4|6$FQp|qaBl5NbdAp0e9Zb!ujTbXz
zrt2|vGWR_%pzhm+bt{?Km)Bo=_SFuPw`bWy%V(`MzCe$DtXl?c*J9%-4bJm=>EA+)
zUtf3~;>V?qyl%^>1@>k=IadXf66UG$disj~jO(0Ra9FNE=_=~SDmA*2w@YhjP(uD%
zCz0Ei)Yx!HjcYFSqLQ}{_Xc1Rc^lx&Sy6gv7x2DS*JsW@Isa#n7GLOhaXdh+e3~EJ
z`EeQhiq_qk$;f*1;Sv>hI58KU{j3*VRd~$nW<KXKw|r6KJAbC%Uo~29<Hy}Jh-N)^
ze<nRnoeelo-ZtB5faOsG_rtJf$J(Q6Gkr6?sr^0B<Fkx2D)cl@XKk3v-kJw>^w+F?
z!ky?}Av4!e2hQ!lvx1!WVVeqp>*T0v=na<yIU4cW(`}Rl<-DHF?<d8UNI70y^2WhJ
zX4{!4=~q*tqm0+jB^s#SQ)^|t|C!v~K;F)2qQy~}25TE}9;5~Pb~e-zGdN$#JfX8W
zDrOe3@9Cz+^kzDw523%UEwyi+8&=7TsKZ`ycY2yW&NQGtd7Edd!;#j!{!inc?_1RK
zPioL@tp+dVXizsmgF0mQW9rj!z1eRi=g;z1Jzay;X&QX*rA2FV>K{Gzxc!70v7-_F
z$=j}jxR*kuN2fjPm8*65^pyHXpbiK5ao`mC<dizxTcgIq(JHu-x7V3ZagX^F>zSh~
zW6xDHO#%NfZ#+FOryp02bn2b2(`A@cBEuOQIU=^n;a;XdJ@U4Zvl906ym4)~Hx^A|
zJvUp9@h$1K@nnVqGkV95@x~3ZxxFuI<ELKeH`ohK<n6V|o|wJc9j1kc*)u(a#&7m>
zm*;*Yc5=nJ=B|)LxMFc8KPGQG`S0bvfWvsd>o6v~QR0qL#aU++wpy|FozLvTT`J6Z
z!7~%jZyVGaczJ2Dp_7(=77dQq(V$g3HD=ROeTBTu;PYeFhCPi6`Uj_InK__=p8e!m
z1K4-r`=yq>1~bfgrlQufX*JKN<n1u>wqFP8416zy@od(Iuce(0Rj}asar6Y9IbJJ}
z)J=i4&%LodhWC}cUCNB&_{Xei)$G?ojut*L?4KaRkwwgWS}wyzW-m_dEkitct0r$(
zlDEUi+3wRM2!G>=FHNKvLC;gvISKl2mY~aM33Beb;ke2TE7RRj_o^Etx7;wW%nieS
zxFLF$J4!>{aqo^hGfX^CqrV5U@~Jg1l)y2Q+$C>~_oyqAx3T2ybH45~`Ms~6l;G%G
z*0`*5T^rGVMO`_u20hK>?a0x#qUmf~vBqF4KCH15Q}@}47JKYOrN&N_o~$dvd)o__
zVfMnpmb>!E+jf!kMcMPud3A+~xyB=&+KJa^?ZlALb|Ug#9nrR-wfK=|Ax;@A#1Qh<
z`GJM_kL*o0w-j$jT8g2qEJS=~3t{YPA-<Se2pwzM@ZYB5V6mylxo9e?kDCfh*0cI<
zzY+fO7k5g(LPYyAG+0)Kj=p7hsVYZuSvi92$}wp}ImYVW;!gW_Sl_%7(l!;yZ1fJ7
zr&i$huQ#aWUJn2E<=B%}hEX3%vH$){BuO4&MC}Ji`Eie)(tF$oT8v}oIA`1RA&z}7
zLbE4D2t8Va6}Rr7*t!VkzFx<Ndsh*cb`=|43gJ^wz?t@ZEE$)NkI8u`ufv?iXE``9
z{}T7sa1U)%4wQp45xge@&}X3E<_w+>(y3Q+FW#|COeJsU@6W)JzUSf6_#7Sury{mf
zD$Zr5awl&JXSNctes==4e~w3;(s+!oipQ^h2{5gnfQkO`7^92Fl9YIem+^4wmB_Pu
z3`P%*L8FJ!ctv)m&5y#edECQe7L8|jxUVNB3i6;Rbcu*U`|K!a?s5;Ub0i|)oWX)I
zk?1fj61F*!m>w7lkrTuH0Wny9BMK+T+xHJ6m=73^SM`oyA@{PbXU?9(<fH6`9>oCi
zHm?6MtmMwvcGI~NHiP?W9FD+@yv^<siji$YFv)|N?aWuK?iPle)d9E~>kl10O}|R~
z(Yp_QP22s^guKo9<Bu+8ob!kZKqh%xT1Ah`2zpiOa(-clKP<@G`>b#4ldZGJq`Ks7
zEAqB{hact}={sWf-1HefcyWX?fl41dkTEl;JN=;(Ih#RlOj@VMPU^@VC$jhF#7y!j
z8eHHtqn`)Q=)bA+6)5p`p#m#s(v#6n!F(pp8?aaOzxsD-vYNe0HL9oyI-aI)$BKKm
zc)hsai}U!b1yY$`ZrNIap`#R760O9!MQY|ls!>3%PmZq|uAJHbeN~NSC)F6i>+o=1
z(<VX9ond+$u;q*dy{>!MSDow5d73!pSFxY9h8o*__JZ%<(_m%~748<%Py3F2RQg$}
z>^MK#S&Qbm+)u)p{%!QFs#!M`vUa?lt^$`TBj^|Vg&kS5@cLgwefIMgGM`#*5w$-X
zoeo8N$ZPU;EIn6?$y>=f)^|${XiDBX^YN-&t4I1yJ<kD3_$^Xk8THG7>=`DSDe-xx
z64pAhgmp+>9~EkT=5-`fiYrH?7_yVBZ6?7UUU$CtqL(B>irSo6lwVX}T}>r6k=I)q
zD9};qjmFcw@wJgRY8J5XHibTtUCh2Cv-3L1(JV~P91m|qomXP)Ko#1gsqlKA8sF&q
ze$2Ynqao)tha1?l<_^+@^wjP%GGl<=>RIf?Zls?4n?B@L^f$A%ea<se7uM<1=Tn!M
zz}lH-g{@XfxU=7REJ%SdR~4v!qQbYY<a!^@SI<?$vN!w5)We&Tw^}~e7aXWxGY@da
zKF%}G(?LhCohi>9&1KAix}wCNr#z2I|Cf8BMuZ9L*#s3n%vU0YyzT#1fe`A#&qgb-
zZ3O+a=j4cIzcJ;S3@xK&*#20CJTo~Q$=t<%sEO8=!?w0JuCTU!-;Mc#6TOi*p5EJE
zG6eM@tDnnJ@<ooj7JU53o#2~ZSl`+U{cB2bqOB+7OWcwFnY_(D#NERCIn%li67qIU
zEpj%%6=UOE@o%mxF8J)F@4*dQc1bW-D#7k0^bvVU(VWchdy5+NX>YVVLC*E(nQ|>}
zE674G6|_8)9J0}1bz9c%tVs<YR5);v{^5S?8IL84hicG)@2TuZ)OS*t2gQDC`9t2H
zRL;8Uz0vd<pYKFDvlRJ%t(nz9pLKAi3LVJXJ>=~Ig%S@oD{$tTH-y$37dWR=+D(S#
z9b|}^#(c#G%#?7KquN=Hg`8_$(@+M#01tRLNf0xD`f*zssx7^+jJHl1Qf9<RF^zig
z!Y!Uis?U1&j0E=V&&HjlMr`Mfo|oJ(Ug3t-=iP83&keiD+cxBFC2QUPrn#eQ2+uE9
z-SPerKkn{<32i*!e#srqhuxr%yCJfbJB}oKV9h8`7|%;kuvUVx)OfzklAv}Y@^~P<
zRqJ^E+#`YSAqk%MuobbhY{fvMt(aiS9L05Z!j&H5c1qT`1MNijMs~vdy{(wU+=(vS
ziRZx2lVs5o^}<#>zG^E5-Lw@im~*_ZyPXIuu@$%SZN<i4R^qRZh46N^5Zjn@9No-9
z4EC%gQs&hXg~Muz5#+7K6?5S`%Un!KHxrkBnTqM0t$mejDkjV^70#z>3Q6ua*gYsm
zVf8CGwkg9w^48=0OIS}V#pJ?Lw3$?f4kycT&w@SNsc*6D<Xfcvearl?w-940P;jRl
zvAfH0j=Z(L_Y5Ma6gLZ>p=<lcC~3v~im&%D@$fy^Exm^&{qMoo>JIKdy^W|xMG&|H
zw-vYHT*6(pk=G%sdljmAg*foL03k;U&@nn6t;gnL^@2;>2a<y|*K$xDn+=^!Ha3sD
z0Cij@4wRfnu{h72nCCDu=R7uDIM4nOwPC;W7-2XM%g*QVsZScZ{!BsE=oAE&r$D_d
z1wO;c(9H?hQ5lc_GU9RkX*@c1P2hcvNBH76ytj@=U-EX|jd*lf9*^$x;@}w<1F!Kh
zSnSVzI#Z)CYYO+qE{S5!UKG-dQSc;hBlv%xc1Pj!t|+*6iG+t4GocbrW5a+*$owL)
zJm54kOHRRg!YQnJdlKynPM~>u7_7Q7Ke2$BI-Ct{vXA>ilY{ZbIt(A;xhpj<94j1-
zqRr@|2x-lo00p6VxjPgWXLF|3KN!`MLgBoG`&#Y=;>3smG$U^}Gm~H&^V1jh_J=ch
zn-$Cqp&$MzXJ%Xu>(e9VoH2OMdB>GTSXeUWhj|9}oa@<@@5@<GKm2BIvsa!U6q$bT
z*~m<s&Aw<(uI5(I-=bkgo+o>^tVQOt4|!$39`Uc}m#wYCtq0_+C;d^6G-yuVT9Jdj
za@Ck>s>a<^CFWdVt{d}T4v@D^$lFYF{=Jg|2gWK8wN`<HlX+bUCMPY~$GuHP8@(~)
zlmhKWC{Y>ijp0yW0eKrXM2+X1*Y8i>b{<N-m2>vpsVzR<tVSH`Q5RlQo?4K}2UXa2
zn{{*#W=JMdza?*N$Xk1|R(_AyAJ9k2{#DccO0*lx*~35;meX^yonD>)&HgXXQesja
zB@P@=pz#d_y-^CdvM#hfp~9|@DvT*+u1;4qXQDK?R?gaDvlczc+m`HoHrYo%GCeG-
zpXqV$5&w<<?i|AU<p8gL!?dvUVo$9rIkQ(u->w2Dc>knDypgv@m@iP%RtX8~kX`k>
zFs)LGp*y8mSxbu8&JrxE^2Ez~o><K5WQ#I6oE7x9Ea&6NS;@(J6>zS>vxygTWLcvf
z=N#nC0y+Lolw<328Gds<Bk#2g?O4ZE_V-4!4fF;LP+=G!&#STY^LQ{@h%Bq9&zwKj
z=qqlpCrDq<kqxYE@3PP6%0A&r9ZHI*I|pj<g#E=KC&<_#oHJo<8^b>fdH#4$4rVcX
zK1ZWK&p=-94c=%J$XxlY^ac+jN7z@*$(O;YFWDy1Ao9Exqp2AOum`T87yZC-`l%(X
z=e#%%#dD65{es#d)ZVFQoP5t&7XFS$sfk8jkz;fe^K@4$(B%ZZGGyi?X4~JqLk;wb
zj60U;>3Jx_Kr%O#%<aejUr64{+RL%`2R%Ra*wg(YLl;|SB~W|*)0VkKe2jXr7B<h2
zql=0@p&Vv0{E*_fofKlK2i8$bp8oMLTsI!#p1=Lr^L8)7$lGSRJ*Zp0hr4w4;6O81
z)L60?<$-(IOLK#kv!i{;+moJB9Q!Rr7tWxLER^GIgg1ts;q%Y4Qm<vqbmD8){yJ;m
z6`VgJXYHx^IA*aHrVedVL&NtBJ=pa*oBl(MNxxONnXE!UxDuYrIKMPU0nhvFLr2Te
zgS_3wIzNQ>;e$>AYe#R`2stu2$J4f$GsdfU_8lTa`XCv1nDTztlwme`JGGJ-6fZr|
z@16&)d%2_T2RGayBikPGfF*g?ZIcAa84|Rx<b80UkEw?gUZbQKNZuY@=ZT-Te4LI-
zP)PPYeeZ#p{vMcX?hY?+HyAs)Vb4T2W<<H+>nk^eRl1@3RCn$mp*F8{NBdlNEalI1
z=tV{syJ5@*H(2j*Lr19_zTNadFn_*jss!%K$=3et3pbRY26-zbZ`YEym$ylv<nNc*
z-Bu*6vlTD(|Btt0$=eAocH)T4PDJ;!6O$X;iL44+@%NIgnCoRL9_Vbv5b`#ST)jcw
z*1lmY8hx`Bj+M4zIvMO-ZY9PnwG?C8S&E;1Ekxn^T4LAATB6g6T4LRTT0-8hmay++
zA*%DtMKjL(>>Fk#YCkm<qt2U(`3FqJJ7-gI_he0Rd(<mbE`N!4zn(+!_67RpKgaov
zFOce7it&!G&`^}2--&X3p>8~iymbtFi#f$_VYlfuGkIP>*6amt%`fLp(`P8S_6+CR
zJ;S`&Pw3k$M$^_0Fl)#?#80}1x&7{8)X+QdCU5JLx2NQ{ad-J`Sm|z}Y{WHKpSjB2
z$%QDLTZnb!ZS7;+8_&6)b{?#4J6?kCr5woi<iIsL8+R+S@bOG08JNkh%fM;!*80Oa
z%qDByQ_s_*&pc!Dwn}#%9mCHe>d0BleV>FKgHo{jL>$gl#qjxy#p~&@IJPGmv)f0(
zDIp5On<ii!>)I!YXE10kckk3^t(q5)AIsvgZ9*LPuEbyi_2MtSQD{3l5(fuH!h)Z>
zdGRR(R-S~a`xzWubO!IvoI&u!C=B_;-2&OCk>qn4(|bpvjUkeLlGB*=_!O+`oPx24
zGd<+(^fO^NP3?D~A300jmTqC*FXwn-MF_rekL<sT?B#OTUG_-sUvG5;+HCGa<IaFn
z<gF8Dea21<Mdthv1Tp(w-XZ|&Vz?7-s6U*nm~F&Npo7+asF&`G@B4h=X<!~N`*0=&
z-19-->>$qTEtIlf$t<83k9<*@<clkhI15<L?1Vk^`Rp(v#D)2AtPxj^rcaf8?$4Td
z^g-qXz2Lma5$5_#WW83(ECE|)ERwfBj%l&*EP1<4gBvf{OXW3TNp~d*m?hxAI_DhE
z%zHZ1&$vU5{bcUQ#(e+t8qj$huK~Zw)i$g(Bbfa|-kzeq_24AGW;l0Y`g;E#mcBYJ
z%JzHOL`1^w?(WvPRZQ&e4nzT!uI;82NhMVj1TjGc6+r}~L2R%H5xY=OjCY>z`}<=*
zJ`b_%-n)CvnK^Ujpn|;Zy^0(#h1%(HZ^W-Wh_~dL!>G-b8hc|7&#MWaz3AJf_PfO$
z6~A<7WUb@poO)tg9mY-YL`Pz0!wgS2hj?Qsd2l?rL8d<YqU>u05N|D3xZ~9Vci8*7
z!|pA)&M;5(wqh-Co?M)Gdub?j*n2!1-Iy<1Lhdo-AiS^B>)OW)T^@Kr%gmhv^fYw}
z;M~YFdWY<p&+)_u*NL~?h_^x8h)?vjzLp(AKjQ7*8(v5q%Q;W#psUDt+CKHbAo^v)
z?L4tE-jlhtTC_N#LR5kpz0ay~ka%1BUWL7PRX9dnt;v0AoJ)wqFEuDVqs1Top3Phx
zjM}gdH_d}HrcBsnZ&LhQhXKT#S-W)bh|-~4$=Qi99^?g{2p`XWD6e^PtT%QMZ|%-`
zqt{023>EYz*)bQ8^Grwhekms2CKGQ5Y-i?mu@98q^ocE^cZVLVmrHo<CVJv~ng>b}
z4&nyS_dl=5qsbkQmorD6I5zpEjypMs`F}MiBIZoJpo5ZltKsiAr_bsZ`};C#=vxK$
z|L1bvi}>r-)(0KqJmE*Yow9*_1bWqy?vf`@Cl1lO6mi)dZ|HGSwcyT}lkUi`)xj*D
z`S<KK=AL5)FQ01?XGVwcxyy*R7WC5^5^t~b@fZ<bds*^~)@o5h-g}der(~NBze369
zW^%@588zt)9mZeOqO?|n`Zk<fS**mG2gHbNGR)a1h2cpNW>mRi^C%bSyE}6)o-<DT
zbf)j$g}y#lTnv-Ibi549sHuvVsyOGMK|Ha0!dQBkGqqS{?hfxs9@Nl0uwtzz7O>{j
zqlPjs#~Ycff5!8D-tHD_zq73Eh_`O^IKYrO6(8w+s;7VMvNw#FJ&{7Zy+*usxbKdl
zWA3;|9b*UaHi&q8H-$W!^D?cdiz_~8v5tDk)nqLqM)Pw)yp1B>4!*;Tr#v+?H;I`t
z72aPXpVcVPl6dQ$F2^osB^>6cFrT;<97azu_1&V58hjq2fmJ_xi%r$shoM3Zy;044
z)Tm<bb{z3`*8kz{Mq*}6Z#ifC<yf2~M^>>M$H{RoHCNzYF9p(gQxKQG9#G)=C<XH<
z<yc)U!xdXO^-?*05^oFn_dRtgY;{q=xt@Hw1vxG;_rh3W`&#;%iNi0*dnXca6N$G@
z%$Ydw)<kIf-9(5Pz!}=*Ed~3}CW8A!6JhBPVlDBuT4ExcXk#MS+%*=C@z&wDvCwd`
ziBSK;M6ftuEI7<>E;JqAT$nkoxnM!Oy*pfAxI?^MzNMLPxkoc$SNCRu$Jl1VwvSB(
z2j((%<9yG~A<SglUW-*ub*OTzLx%-*_+0b@Hv6hkHnaj;KfJ{5E-!JZ`X%=6DTjDf
zC04Y01@l&wI6V6$<POj9*R>Q@lS`2vRmL6ZWhkFkj=63RIiK|qZ67_vlU9$IX<mYM
z%b!9$y%ZMH3vuy9A!ZS8cMW=oS?>37#U%$FCg)(Ot{CT+74x%Lh$l1iQE@zvIkmay
zKRySQQ|`dWE&~&{WT6M~_Q8M*C?jv9kQo-|FQ(yn-!yCpN`YDbo0v+xZJvFdz0qsP
z+It-qN3P?0bM|XLTt%aZt0+5i83T65!zeESo5C*OZP9r+yku75!E=~5@GMTxibG1n
zIPC9q7W=c#!m;NCNQt4kRopkbFcyY0V{u%F#d51yWGiCvP92LUfw4F{iI}>AS;aG>
zp)rldIO6T4-^U>fKFxhdvDg}M2DhqWuqHJI>wIExF60z`cRB&5KM@!``6wDPgZ%ft
zAbecKekuLGi-MW`YtH?39?V_X#htL7L$UA!=U>-{VBHbso!19ppD{CO*^eE{TtZD<
z2=4Iv)xSc}&4WL?lDl{ex#yIb@}16b*EM-<(R=Q~{KDNLqk`a&%=~lWZOjbj$1$rm
zAc%7uoCO@%h5p%X<fpd2*yzff0QS!Ib7tqqEAG&8AQlsEL#(+=x;1C$`*R0of-l<8
z^I~q!oW?9)<dU~$>3QKtLr)C;;(^ja5A^0|dnxrr8{+NTGWtMRqdy=gyxEq#lz6*(
z0(GvQ?l?4%{*+(rn-OpOclX4p(Vonz@x%o~dXmD4(TBa!W19zVb@RX^@|QE@Yc2+!
z+>^-sfn@F)VLxnS0cSbbgR*SojdBCl|5KQm^O3o(FSsv+UQ_pS9pqLz%x%WHop^hn
ze>-F&zxRQ>;wZUvS8sgi=8m8LI72vFhvZE<xbZx;Bi`OT%)j}LJ4o2yTDid!JEA<H
z;B3eiYcEKjlY2UNGk?Jw(ezY3QhVcVZ!Z`TZ{y#3amO<2i#;A#uBC5e24@>uvG+7x
zg+ou(c;HJuOilFw=NRjF#)oxvXU+llkBufSM)SGqYalUTzK$(Dp))+8|IPzDTvcf3
zrbHuhjJ}6>4qjEl@Ro}E9@Y4LPmO~4#DN#g58!ib>(A%L^K$|9=3x3=>;x75^j1NS
zwMVVJ4l7o2@5*=`baynI&Cuc&dCqoyPe@PdP{4ejgIl!l6f-wop~d`4`oHKaHsw5%
zXbN?InLC;?+x9H4m5#4fQaCxkJ2eO1_G$Urux?mEuhN1w9+*%`{^;zEgM*oEOWdmJ
z%bdMyI%Kxvna}Kk4Rsof3fJKbzxUJJ10IRwU^@?Dssp{>=MJF}@iy`_HGJ~dTzY1=
z(Ifle6L-j5U@w)KyL!uY=*Qdz<3+^SVB&27b#gP-SbY876E`Qb&U$#38tG&HEHvmp
zi28e^8pY4F2z#JK{(aUl<8-Lrt;Sbd70x$SV;AxH?@S$9|DsmfN{j7pIDc|igHHT+
z5g&76dkyxiP#|HZ91FL~;67c9oZ})a40J{EP#1g-b;9?D&dgwT#)cwi%<1of4eg{D
zx?BdoEE$edBV87tMr01PiqZTVfA|<q@G+0`K!+P1h!nB*^kZIAqBojn&_{L58`UA+
zIO$GrI`xOWJH25~-MOzmJrUOq!gsPaHnBF+IeVex8$Um;o*4gtnThk=Ik&21jicpx
z!@l!#>LY%vDV}T5{k;}O%u;;K*V?L71KVshV%Z<{CI4Pmp~i++D(DYY;cb!<D|RVx
z<&zvYF0udWq{KPmUG-ljGR`Z|owqT(wdd_J``NWI<gR-qD5_GzwL5V!RmmOFN}T=9
zS*g8pj2tA#;O25P`pABEx(vgDWRN`L{WepeV+}RqA~~)V$}#<=3<tfWxcy3sUga`a
zrAiUmLjfmxsz!>H@KTd!kL2Sk;^*nF3KJ%)u&tX4UOiRF;b%<F&xNe7iO{H@iJ)N*
zcM)?FYphHJKb?v2KA4;}$wX*OtSuO9B5XZpA{aZG2;r?w1bexW(9hXWxV6JjINsb)
z7_V$0RN5N|nVXsmlU4e{Rr1#_4fKWe?9o<T*AotX;fzn$Cc=!|#=`IWfAPpvPgvQl
zsZdbYNN6Xn#kc5>=;&UB`@5^qa9$NIG_GRCU?q+XslXrAOU$o%iZE)y4aSxua!@&*
zIWik+VkKmY%eXt>89J{o#lE4X=-#vp##Uwg9GAm$?L*`fZ@WKyh<$~RU>jHh`#X0r
za%K@$4=+T;=>p8O&WB$@9`=ZHIig5BotT5S)R7IA79-+IK9Zm0;_Sc-9Ewdx1LAG<
z;yZ{)Ove~yI;zvsxK}d`OUE%UYSnF=j7vk`tIT37x`B~N*O}pX9l9Lu3AlI-qCMAP
zN)7q@=p;P(KfGPn@G`_^30S`U5-I~OFw>d40j$m<wGFcqrE!=(Iu389oI&8HSmb7M
z|N7U{NN5p-CFzkkHYXM-rm=9yIL-Z|r{N-v#cpLRs!U>VYC#M>WSm61$tMsQaSS8s
zkKtv56EL515_bl2uieYj_*W4Fh+-hV6%B3dDdtCWZs&Of{tV+@6ZS)6t%7i#c-xNN
zCYPh!%T8a@R8P+0^a;VxmceiseH1g8FJQ8pd2q};4{OW4F7y}AI2z3R4o0;evy2~y
z;5G62D)E*=8_LYMUj*c|3xW{1o|w+8_)7L+b6lCLz9<lPIMcJg0kvV`?K+;J`J9KW
z-ohRt@z$1j+l3zEz(~&Q++rT^3}5crVTRy3?!i0ii?vOdomc0Bo%Ch6()ZDbvx2F_
z+v~*JXLZDuhnzPo@Ss1%17k0HVEZIbEFw28PoQ7c$P00)9_UC-b7gm89x<-AgF9N<
zxMS4~cdRw>#0&Dsc5^*p#`}n<b;lL*l?={Y_z;sOz0twNfqK|Y4>a%1bA{)7_Xi$0
z6y}L%#M@Quv0Bu7!JN8iK(!YVdG@a0S*Bjcj2bb$i?Ledx$E%lCix4`;({jhsuByF
zFS=vD9{-zvBV-o!M>8GzxRE2yV}^PeF^qULh+o_H+XIgxi3j(|$8Eguh}yWGK6|*-
zoC6k+Uu<Q4P{V$7JhOJ#`@6i7x#~Pqi~rF_%(~y4czbrL3L|2ue_f!SNGxlSPYtty
zJNmbGNA^nQtP^if_VmDEOY)JgDinsQG4>GsDQ`VsW~V}?s}j#7O0*{4ev45eY$5-9
zkrMt(l$h611=nYs`A^s4D`zko1go+10<#3XRp?5*ol3kNJW++bDQXxtWllji9pY2j
z`;Abs_o~A2Xcd|&wXChQ%(BozF4p1^ALG4)+|{y4hpP%^_wYIO;kBE@>-uL5b%W2;
zXGgLoQE(=eug7wGPmCbm?m5G(T>6jR6GwJAQNwQQjy&e!ZZ6SKD`yTLeJzblnHfaB
zSf|n<(v<Io10Im_+Mez0jS^}Mo8ma{M4$PUXfJr*;`@cYXA`~-t^dQ@^IFcBGxKj1
zHDlIO&a91IE#}|i`)Vh3<A2X|m~mc*qkN3rEZBQ|q($4Q8VuA^11D8@!>psO^QeCw
z=B%ebXHPaOVdtqr2=TTiPKSY}I()CyQX|nKm^gd5kaH%L8dS_yVDV%*BH2r9OS~O#
zDuewrF^o=&VC3zJ3qxFRqn{(1%y)!asS^T}&QOx?E~s#WzCr{a;_W4W`f~Z&KF-$Q
z@iQelHen6~vkI^3H0W_rhxz2)<?JVXnd^m%o9J1f_A#6q%opk<o7lrxN4%|Lk5{?I
zn;MKat_a@P=jn;Y{4DGwr){{7*Mm7ud)8>Vi%3hajTWXRT72ZSUz4Umz-wkEzSSU)
znTauct*<?y&*-)qhpw^*dWbdQV<qPD{ggFAiAB>D$jFv+KfD|*f{BY(GW6La!@F1s
zE_zAe7AV2Ld<m*7rHB(t(V!2%K2CxT88T+b%FuL!43o{3X!});a|h(yrzpqDFEVCf
z$S~`U3|)w~dM+}g#Yv$Ri|7*+!Fs$H4u%p;Fq5K{krX{`B}g@NLtL6G+Lns2u0V{(
z#&T+`3Vg0m;%TZ1qka>s?@@cMRv?dfyNq8;|6nYX5_8qW+wR2MC7ao!m754{kD3VH
z?j}NmhQ>lonUV0j(n$F5+epw6Z(p=G5<V_76dE=#6b7U-&w_ZHN4(W-X)cJ4uxC3`
zUpQ6UOsF{8Ot^bZPiXYDsW4_=V`12$e~?!EMPRqz@E-dS-S51`#~!cI|8N!aXRDA*
zylr7tg`4@WSZ|hKN{9Ps6G9&PB@e}$3lJ^3k8y_OI5fHpY2MFK<NN}{-Jd}_sT9Ls
zm*C8$Cn)iI44a}yXg~TP9L5#nc<=jo5_+FJ_C8+Ux{KAOMKD{*dD`X$NZX$e{i}I+
zu`3srnjBnfMlX~@HuikW!i+1K*w>kJJxwyuG$tL{dg(YFeH)Lia>j?d1I8zC_j^Vf
zitjU{v3nY<jc(#vMKbEDZ{R!i<DZMK<9O;dlwRgup4N%D+bj`Jn<t|F-BrHUS22iq
z+sTFbiW@Jnf6Kh5_2=pTI*aJP^fOJ1!{5he@TBP(>>%DI5^poUoW?POC>)#_37<=x
zv5}v^iEPfnx<zwuSTwoiX>@fwjr~@qQCJvB56lU8`X0klW;S)*6M->{PhjNwNKDL%
z#Ij?N=qoybf1{7X;1zd?Ej)?_bw^-c9Ryon?xkKGgp+52@Zru8jNq)!mn80LVNSkf
z7WcD!Jc=mp1d!}uuKP~@Oh4wLhjF(`KJ$Dg24l~YU_=sc$6pRX9`SYo@pcaJR<!U4
zrl$sBJa>ULoWT7oPnm0cl6;o=aRTwybW|Xob1&LV&UXxd#Voi#%pRP~8NI>udaj`V
zmAvrQDR&H_E_Wi&15cTguwoNEx14`;_~#4Nd><s!2Ryp9H@Y`r&&ZRt@gEiDHdLcI
zb7Rg0X>hr{7RQJ?*Qe6=$i7r0`C!a3PwXb|UB$C;la%uW5#$}M-0|3+XKj)@3U7O$
zj9PI`8_obyQ#}(-f8;E8IDi@FbvjgS*FnD@d6T^duGxBUe}o5K5i8fFvlmRx+qM(+
z-ck?L4k6wbv4@w;p5|{Y1`uycMgQfRYte>vet!k^uW1?>5$9a3-7%cr(--8wod)ro
zo~}bMdE1Ha+`q|PB`Wg8pS!fEou}jeMS5z8x5w!LuB9&CH=g=k0qY1}hZ*Ml|EXSh
za={Z5`Mm$Vr&ivOdNgx(ehJhLpQ!PqwHnE+!xMSN7st}`bW)3Fg*uEW(V_So^-o^=
zDWlk{Zs);W0Pe89r@|sX`fFBlrlMYn-anNHnxsSvwE{<&uMmDf3HvV!_?}>&OGYn=
zN{Lk+G)SekcW{sfE6Jg{(Kqs!cpEZC1)pB*_0}rk`C5f3^~Alce2t7Xm@`d<EMib>
zlnSqBsF)$D!~veihmNt&ct(TGi>NsgZzDB2oMInTM?5-9Ods5kzUZ$SG?}2o0qVDZ
zSTl{HMlp&Sg$ZB3P`(aMyl(zenK`dyP62&hi$XXr!n`1X*P|sL+ebh8zox3OiRa@k
zFBR&D{WhO@-KeKN>2wG^ljx~^s>N78W-g3(M|ewW5bP~Foa5d$Pc07J=YBTU2)n2c
zoT2_}6r{t+yE<eMZ@Un01BjWX%Qf&}Cg9MgN+kVMaUV8$7cu1?_4!dD)aZGQ$Jnz+
z>8?WabL<1&(=wZ#HCCY(^<A`xE~2;TgbbV1GPHG;W8Wk>>UG46eG)YC6w}u&hCB7-
z@aL{@>EVp;Z5^Q+>WJT!PB4`_<HLGql$N<6TqVLY;?TJ<)Kr%fFHS08O}q^!-kK3_
z?P9dp^n!H)@pjHX`Wfmy;YGY1dX4<v$Q!@9kmoXgv6sNwmR^Q8X5P3;yj@4Ut<bUm
z&Kl=i1#8E(e0})7y-8j8Yg=-?gGwZXC}HVNKF$2Y;p|7OVLcXR%k0GY<hzeF(5DvL
zqK_KOKB&-ZrV>r3u-_V~!1ii6hKI`WJ&t}ae>uX5w-FXHWD;*X$RwQ85aI3+5&gs>
zv{)p<CkyiYeiAI9&aPW3L&aJd%D2dH-A0Db_hiWKB}bo98F#nJu<Ek}rOzZ#e-Pu3
z32T<=BCMCYaj%~U$9=`nzamDOQi5x{CHQ0D2J=`~T;Aq}mi<MzGhK$!e0+n;l^D4}
zg~(_d`cC#^V39Q*-nWKvi8W>tZ~s0v7T!NM7M7Cd&R%9Bh-CCn9WfEI*soPIpicbU
zNGPo^5|r<aga)ogg2zJxVcs<Z;qoE_p=(ZaVOJ3QwVLKa<K@kTW_$F7XzIw@zc&-Q
z`8E?QKj{f2WA%jA;lJ_KxE_Z3U$MIDCzPG3#`POjC>+fFv-g-q5nqAc?_T0_%1cbY
z_!5u)-Nz%R`<NPB2&d{iM7J(LBlo+QrGF2uBg^2kBNvU5bCJ6xkNRyX2JCzSTjFhR
z@5h*)_Xzh|J;e2q#R#doi=2aZ;h`x)Lt^cARUv0*3vi<<A5UZQ@O)7oj(*L-iA_1s
z{L9ABge=bd-N8(gJJ1?s;>Y83%s-LN`SaV%CAf_a@wXBBEe#{rrNY=g6%9w;!r^bJ
zh|f&H-q*~r&|_8<eaCAolhD>Q339I_7?>ntl3pS#n0xG3cookFUqSNJ1Pm8(w#W1m
zjO)%L)|m4<OV8rQnK-=f8;APrGq9+MMeqmuh%d$B&Y)Q4*oGs6p5T)gPGA+YBxYO)
z!*A;_On4uS=krd(aq1~dopl;>@=m~;vpcfo5qQ)l0tx)L#olAs8FLH=J~F>JISlfW
zFwD&jL(I|;j9PjWi|JYNe8G8?!_4Se#F<v^kbb|9bFaxq@Fejlq&D0OdzE=QUAd==
zndV=X1VP^-5I>nSv7C5o9?Jal0Y}h>TJe|f%wRMj$Nk3LcyqWrfOy-6b1vOR1mX3s
z0L*4K%~JMc9nLTleFyJDL~k*BuP3>?%_+^FbHALed`B-)A+zxQu|GPE+Ai@nn0RZ#
zTKORTJ7-_e%VA4TE_E%lmz+Btz~1Z#=FJqb7iOTr@6F^<#M=g+sWY`vV{5bqr`T_N
zIFbI6SIjgWN?psz0~<$qVn6w9F!jD~PTUPbyq)?<2Q{_7o;**Jd0wA8NsiN+^9dmy
zI6{5!4EtJkoTK0WN{6Oxh|#Tx|HRsTymhu_-|-xKlEgztIs0JjzpW=Oln3&hy+bXS
zeXG<eYNDLUD{;}_&oTZCc}kL_hS`!Dw3ll*cgwvaJi`xt)#7>&p3}tJXmVUF&-gjZ
zwdhQ(&5`$yE;_uT_p@=ZCrmRv*;n*J=4j6BO#lC9;Vj~9&J@-Z#Nx@U1G>jh|K!=f
zAzs5?l?Ju<)p$!z<26Ku-o)EV>Zu8d%wZ?qb|K#0dQIJU4Sg%*t*%B&Ot4hq#+CoD
zPlXQQO1xyALa~>U+)xQ4Dd!c-6zIKQ3CT4Dvw@k@vqk|;PbH4hb24Kf_oxtWhmos%
zBZj<GlGpa7KKM-uG4H>76*=}b)*#IBnL0}iy}>H<%Tl1(Fa-{>M|Yq8lpT?BoKKJ=
zkY5{Cz_WR}2ATAvM)YJh{7~{F)+<|Uh)q@+ewMWO@r#;3jyt}yUQ)YwAerx<;-Bv5
zZbFRaj_{S#Df{tvO@FDekN6t2lm1>iHCoa4BBSTUhj{zWLxqpT+g7)^%d~?UovD+L
zzRX$6d}^-KwFvjn;v_Lp_D+lR2<F@K^)AWd>&Smwlf!-VpniOqGojRv?H8)Ch1#(l
z`DruaZ8$mgg9l2)*QpR!qeST=W*mhnG5x6=g^%T^7^6VVPkLk9C~<cjIWP5)MNPGM
zDpg_9XgOBIN^w~#!|dlW+>Vt&rjp@nXBnE?i4oCB413~jAo2ESCuii}a)7nW0mYr|
zVOHzJ9pKIwpXr95dn8DCBIcZ%6k#)^ICV^ckHrdz_9}6k^-x4pzAu7!P1oy?V#3)j
z`UG0=eRVg<3ld}UTjH&=pEuIsg<VgW|8&(8-hr%PdwQaZ`htf2YZotO_7Byf{R7r{
zoLfCWymj%RR=Z3ME0r4dpVbhbR-@};z8)4DEFPf7C_dkTqgc~bDiP_x_t-2Ynh<X{
z5YJv$vfu1QyrrgF(}BLGQdcyd=E7V&dcmgIV(HBNXuoqmmi#(^hmUQb{pQA<^<peG
zmf*!!39?qpU_-o}v|mPzQU=Xa84j1ou+v|Ltl?7ju_YK@DaN}GBK&~}Nj7fyRVPBa
zn;0K2i!r<zIdl^da__mKSEwsyuHtp<C&I^$yuWe@>{>V?|D^-kI@oX*#y)u5vBtf7
z);RUZ8pZF8g>RRPg@I#@h0y~{1QX)zdTVmsGGig+rLoYJcsno8NXRGNN}m}C^Q|~v
z`@4nUZ)+gz8E7Dcq%{|Uh__j4W=0Wj4_N36$vyRjdVPIiq<b@=!)HBVrmLP%m(6^}
zflURi_A{0cZv*105ktJaQu+!@23ErCQ8^+ymSMf^OWZSijs~(4R5vO?!rENEws|=7
zARmVg=3>j29PUWZMG)~eB_J2Ei;8h8st8G~im~zcUA(n?h@Ey1;cf5`-G&sypLpA_
z$6Y9$i(sf%gwO8^uu@k5i!1r~N4zZ}-dYfE<Jab(d1*F!eayn-_&b<4EE7wXW#IF}
zbhOx(j#c$(SR2j#K8w@vt|}EZ9aE9BFcl8#Q!%SL6%Rjgm)^pgxcM^~yWc0lVQ&(e
zbH97EDhXE{ui%z6ft)u1NqrNj(Z*wp_eJ94MdHpy7_B;o7mJuJ5fq2PCUIDM@(lW?
zaQ4+HmUEP+;hPo9{VLRukI?@+E)r7W?MqwE<PdNBT8055!>G|SC-F%*{=GPk!M5DT
zQyq@br4jIH8G*sX+vagW*fAprL6t}FYeOhnl!sutCIt2Sg5e#(%vpZl?jkv>6L(5a
z=e|recXe4Z_y712Tt0skKP;I)93KFE;_VD_(xL^#T<%EOLJ!khX05*E?zL{j+grul
zsrj0FXbreSKtC9x`8D&ILAdxQ5EAA#Y^x)#HswCAG0dRt%IpSTf3%5ah7q&Un|1WZ
zG|uc8a~EDw6?1-CF`GA%`F4XmU~!*Vag_MCfjlF~1A6rySnt3b8un9XRB`TxKGNBo
zdo`n%WIXY<(|Dfi@zf&QtNwTX2M`Zd&iT(@sKt&lEkfBtwIU~7Nex&@yj?rT9bH#Y
zuX7~kR_idYQitC3B}s_4o4q|Sf*y~#M%+cBbcZ@ujVO9D(zdgY#<M$q0(C%V@)c@>
z!_R8rG=h0H{9c6vxuT1f{W8u1x^UKm_ql+x4Q{M`&k`#}F5_O2{>-Hz)_U-Ed<r?B
zj2Sq@+X`|ZgD+ZK=}4R=4w;W(PAhfT7Ya2x@!UFY!ROJAeO+qh&6j%OFg4G4U8(gA
z<D3xjwk3PbPeYkS`<^*OX`c9GMSpRqJI-o3S3wT8KZ&!6Z`fl!Pn>C|LZ=nvspJql
zrc(=^szgmFakGk8Z%!Z6DkZ)eFstac68*NSU`GA-c7PItc@7S8QDRaPCHJc<uxhIU
zmeeKP6BPgReJ)%~Z>O~acKqHn{_X(ca}I0&xXtwD@c-?)sDSUhUt;r}mnw)xP{$?z
zS<m{Yf>_*hAb)nS0#7)z(VKXCW0f2Oh_|;oC@|2N+^-{XtsOOa^5=KN^Fz$TJ<mR0
zLvp0_|MW`g4^FMAQOKD;6ha*$l3t|QTGY+a;sNJHv`fghyAiY3)7QfHOZ9veqIaor
zts8ZAKAvC6D%@Dc_YgU6ENg^la;@JY&RCLPAJ3sSN?oEU@pb^Qx6NzL<plF<j#^kQ
zQR7cG^L5*DH>$A~tEqEOKd!@0;_Z6k?a{L;9GI!%jDQNY7nR7IMsH5G60dtu2OcJ;
zKR}M>59JUVC~(SMjB5#Ej3@qlBR+kc!RJh`t6pd3GaQzp+(Cv3Ph_~zo80W61m*@3
z+<NYYgSTwZ_S*sYTs(ldIvc#TcHr(mJB$|FVbdljoLcCNs#{LX<Z#C39wNwyx09%S
zIxSRS3O!e|chYl9ytUHjys8i9RQWtB=pk-GKi}eA^voaRoGdl<k)6C@aL5}wx_cpm
zeTP}gS;OXd(4W8zN`DW8`B39urp0+)Z&TKi|J5tEM4YkNra%WipUJ%M_NDX^SgJXb
zui{*!io4TP*wBicm%P98X(c{W`%SLr&+Dk0h!wb9E@u{k91r=tpSPA{De?B%AQ!}`
z?J>}CKhEp#N9EA{n6|?X|MZ>E{)Za|`in8*pBM?7B}gLP>XyhbTq&c!S%xOXGAQ$8
z=zdU!;Eqxpv6tX}nHZOIM4a;#;n*5CG<@fZh09%$HpUf`hC1=R;sECuM@-z{3ja}V
z*upvIO-rfG$GE}il|9NN_E;m|hfiMC@Va3QwYfD89k)i@Fk|7L*hnaHFcLi18VM7L
zw-M)zg}3L8g=WOt!hc4>B?}{A+4vU1%@Hkx%q}g2*y9F5xs8Dk+}}WW&7SS3Wz7Yb
z;mw8krp<*wbA2J3c-x}6zR+LSO!#8dOz@r7M3@uTNO<kgNVwdmk?`fzdr0l8QI}E$
z$I4fzO|3w~ymC1AD&r2Ymw3A5IW~(+aH2^GXZ>@rJ~5ZGc{#9f%E6I4IVk>>gPR-g
zK+h-}8+YX6*Nh^>RunN;<}O^W-NnDtccFPvj14`D5q|$JZfovh&afgpFDisZc>z3#
zw`=0^k)D-@-%azdD>DZR*5_bI3Gw!Q7Ah2XxEnJQJr`x5swf>DH>RW5_%=KviMIpO
zFr+*c1KOn`rb{a4(NmH1DwR3j%#R}8&L`fECEgAu-rjFXZu=t<7jIpKN|b=~w(;}{
zUBc|}OX&aM0(<)xQ1zCXP`l3I>)f;O^N54kh;y$FXVAzm8V%D=Bl+}c<ocb)HX#~~
ziMKs&hhqQv6G(9hN6MctJT3}F>WDDt%?QJ*X5m=$_&D{m<LFlxh7*$`pgs|f3GP8?
zw2L_k1A;I!BM?^x24JqAKL&+zcUv*{r*V$eJC=Fxr@6Oga}aus<!q}i5Z{>Hbdd9`
z-L1JZbzuO$Ci)|kIY+f~1F?sAJ7O3!1|I}K#-Hg(y#3jqxea%NFtH>U{u!JBI>^1+
zov9ZGF~6cVkUK58OS3Qln|~8$`_kt^Pn7*|?jeuz$DRN`+~NOEAAb<TIcs6~o4FD#
ze3&i5`Hz3ZP~z=0;_ci~^b~pOARVm5q!4nL&sro8(Bb6)=EgGn?-RA&LHg_yj^NBW
zX8@KHZ`WLBP7Jj<Gve(Tb9xWk$(RAH!iAn{^!co0&c6~fv~rwaU#Fd=4$}+SFO6h9
z&w6{-Jo2Yl6&}ViXMl5&3y9~ViMJ<-xBWM0aiW|a89tVN#L2SB8tkoA!@WP}76RR{
z_7}MsapN4%lsr2fMi3`^N6{bYz<%2u_5W`3yK-)Fm<HqfXrSWnw<U+!Po4AnH*%Rz
z>}9pq;n)W)M%?1A5jpQ$#tf&Y#M~h~=i74^N)|JTmU!ZPi6<7eA-`oVPm2chOuh5O
z(NmsSR>JS!=k2-&E^lT(dI|fT8N9Cm@@meI?7B$)5}`yp;;mvPJx|k=?9b4T(ocyW
z^kiP-{Da?41y)BXP|`<*<kc$dAl_R0@w(Y7ajBOQ+3)x>bBVX}6gbBG*=83M==)dB
znRW%WaRog1y>wpVpTt<38%p$dQzDu;eW{%SoA@(>_<ODU$>4j28r}>E227Tr0X-;2
z1C%Hms2~TCV?BK>4T-n4Q)EbNAw#?&aeN8$e%2}RaUAtt;&}Q!a&`JsE#K2iWKDln
z3OUs>9j_<(>`{8tP7!~};S-6u%0}Fk#u+BNiJX}vmwr>AMCUc+Ps5ox6s&@MxC&Qp
zs9;G>EvNr<T2IzUE9JQTi_a@jiMOn;7O@YxTBpTNsTPw(T99<0i#>JZmTJz?PzNUd
zUm#z*#+;#{pIM{Lr+@pOie3)-$xKyfL(k6|YNNHRB@C=(SocDT<L~L&31bhIc)N{$
zp64ntOt*-!Vx$;9%*1eE4SAK%*GrH5>!FDGhf<hR|IGU)!7$?O;f@lFHlxP5;s6Y@
z_hCAD)yP^KwC?JFHk<4qjI=}15+`(T?~M9XCp<87hUU5iLx?la{pcSmku%Fwj&)`V
zJYwDC*o@qk9CR9=pVu09$X+qiX(M+5T6v)l@wRw8bHkSKwt{<$h_}PpTX{af156Qy
zzYq1ED{_oECdb<Z4PNpyA|l?Nvs7U00QMgiC^<{6#tnWxcGJK6{D=yb7pNsr>o~;T
zeda6bDyNj7F&*Q6E07<^?1?SRgpQTNlDW{<1v2~_EX7Hw2p;T9AA0Wq+eQ2FsL6iJ
z9K0VbeeKbEpCh_V6`{j+F&dM{N7TDv%0)MH%93Is@wPMZ)+<|vA=P4BJuJc7diHvW
zxA%y*SJsPI+p@<!*$wqOIj=m^6<>+BU5U4ohdQ9uIeTn$v%^=h4L<j<hv_<dybrfS
zKn3fYNBdA#Z;2}_tuS$;6?*GgbFSPP0gH@;sl?l1wnoC0)kea!QAR?~0z;wjq@nOd
zW+-ghVI-W|X(Y@g-bM^-AuQ?BLinC&AQ*DbPcswV);AZ{5N}K7G#4J^G!ssYYAVdA
zZ6d7v)kJ8?86a2UZNcKkLTlz8Z(PaD<KB&gs=GDFefI^S7FE!by+WS}mE1{LLCv%b
zvx&EBT0O_9q*C->U5e?8b206H4nAJU!RVT7ln%LrCv}-{_RT`e>MVqID8!Hk)QTS#
zp;7xH1dT6(<H@^dS5l1BI?m4q72)L+^4d3rxK~(+_*R7&o0`wFA|Doy^Ke5i58IAY
zPrjQC@$zi+jmW|~;_aY)8MsaVahrVZ`PrO~z(%*xZhjid+uy>`&8e8vFcm!<QqW~*
z3hHmB;%}D}IFux#rS1mWzez&P_9QqNC*i}EL}oi)#(wr{Q+~yx-;YZe?RyCoCokZa
z#|6x*xPa%U=u5sAgEOzAF}`sWlGa6G)}d(pX%>y7jZtu#6p41b!_e$2^M{GIbGSRF
zW?v}04~4>%v$4;HhhZ)AoHEXYLb^5-6}Dk8t)b4mDGbH|;fUWIj`5!Zs9kfP$!N~1
zQnxiZ!0ZHPe>f=p5p&QVubHRtRTP9}@!auN;g54R+<#INgh#~EzgE<LiO;2e+_PoP
zJs->|erpzpbpHTonJsZ>7k8;}c4tQpcg@iQ_1H8JxW%5XC-<Eh2jat#K=?lnM1(pJ
zhZDKu?I&@zu`gma9cFF`HRHw1t10uwar$Xjxo{_09(RIlr&df&Dv;ds;}6d5Rr4I{
zO-~W$V+zP4Zk1^8ieAw5Z^!|A>5wB~&v7a}Gcm*vo})TyQt4||IMPbR*<o@AF9m)r
zSHN}~y^8uWOg$pwo;w8|Yv^Yr7W~o3(cj9Ayiv?~4Q2&2Q^39xb5yB?YEQAhCQ_nq
zkP>0@mADwke$qR7hNv5spCy+jCtE?C^&UCkiLK11?M}~(uN&GBZ-@7%Pmx--??-Bf
z?CXs!ki(4k?QoIv1tC0_J2U@=z0K~#+fL+ZOX)NHEYYBGwFXYC3A~!<(1dt9EK`eL
z<Td>zYRoBNZrpbIWD+^ce~TP~eaXJ82R`-HLGnP0Bx~mA{8e%GQB4mlXIzQ5kBPUS
z@7jXB%L&BWPVQO+wc+{4bMBc&i9_U!ww~<ajo^F*ds$^2WZXfgK>tw+MDiLKY89AH
z9DR__v$CZUQ+2!@Ah&JJe(h@owsoX8WQqd)h_|aG>;V?bk>6SYU3Ue{iK`Xtw?z<J
z-nZa?6VrFKRbu{P);QNV)6rA`lWx=oBSeVF5@ShY87_w^Am!Xe*c%x(nvly-H{3E$
zj&74=%uk~~CQQscbO{XICFqqbL2U+SJyMBdhlmZIh=u#La7*SM7!_+TzE(!WyheYu
zxN((Q{CUo4wqR|PMjuimHP-1fTc}=!j3B;89w{(~SmVw2#@ktPoF-P>elEwe)x0jt
z<w)m@NPF^fhus=xNYGPzg71Na?1eYdAR~jlX@wdy$%~)+s*oY3x6Ft0qW746aEkk4
z%sBf=EPlt{_0De8<ypsgR#6w;F2}r0G6bHLVilinvv6`-;_c=4oKbls!uKO0?06~S
z9s@D<4ijU?3lTP}L}*gTJc}(7%nFs@?obK-Hg~~O1AEw{+aYOzEz-X1!|(uW3_fH9
z;jKNozq4VF+7=5s*<nhSBjyq}{<U_&Zw)grR1yUF$}#(q99O@}p_#2fBJuX&GZn@-
zsSz4PzwjIOKUiCi?9KUKE&Ds(oLjofoXPH-v8GSkh`sS!Yu(ZDyb3p~Ww2mxxi_`$
zEe+++@-@6Um3fIf=}odwph15H-ZiARm%7u1WX`vCr)C_a!XEzamW#a6C-=BSi73`{
zx_b5!IHzO88tw&e69VLr5pTz)ur52oT4S#Wi+j3Y_)!NmGPA~&zx&|cXdm?NSRt{8
z74B48;9%`;{Ij;ewgr39BV;drl<&uzLObl9X^$Tx955=%1qb|iE#FF@SuH^*b$AtV
zTC>a**|%IV=ASe3P@H+495JfO0g?d@=xS+?Z9i<0WnqIsbq8>=`vI)BrH-9qjlblw
zw#3{0bFI*mc)KOm3iC>>(0rzmQ0ZhO%&;*M9<MSI+7oZT6d4LNfh~jsp)G_?#M^FW
zO@(LHO@)3@dcxe!%>@%v`l3EI7cv_d2m_hXcyFD)(A`*H2r_Ic3^~(K7$t8gya{h0
z7<6walr{N-x^KU+dUHKC_pXP+?HgxnUL)XA6&4y+LE*^Rp2n3(-B|{=O;7Ot^<$jB
z{E)k#=^ZY~#mq^$Sdo*139&hN@{KtYZSSDsTP997$R(%GM+c{TL_`!~c1jU@l|`5)
zEx?e$1!z}Y1dks@_<f{^`<{zX@wO1Z3g}B-MgLVoK5Xss`7?P~ke>^Sqd8bnkd2g8
z+2}kui}MV`^z9j#)jxxK<kInrc<cQu4aZH>aK`W!Vppc(^q&;0S)PKeW0IM3dIODG
zCE<2rBKvQNnA1H8v$iI|y&1pOFA>!SSFp}H0r_9z5j!j%*<+a*^_6)}tIxvrV;n-<
zWAPw42Jb7QF}L;<^ruJRw@VbFN}`y-7zH`;Hf&EA=L$oyhgxveXYL(m#_(=cD4KAe
zPPZ;$IR7>jIRPP9!8x5>tHSW>Vi-QnXMc7<I6D1g)?su2eTxCmaL>!#U%t3J!k?Uv
zd!V)c2qxZUUEuDrU&LANuDSJ`IpwPYv5>s9<TrPuB$9hxpnr+;HrJ}TYn8iDu9!1l
z*DnAmYVJ-Q6@dN@<gLWpby-2EBHmuUOnfI6PYMY@GIJ=pWs+C#4rD$vGbzN(R{VDu
zwQCO}nRr{fm2;x>)cSm7@6pK{`|o(+>NaMmcH=B+GG}dia;~D19P$@?md~{)U_a<|
z6n&lKdvW>fy>YJntsApwf77!<Y}<UFK0xY!ZI`OB@}m;_BbB&HzBGwvelR&o3bSQ{
z!`yJphuY6!H$3R!3c<t`8w6K~<3(^?B}EsW5otX*(|(ft#*uk$oyg0Gw{xhS#u%xv
zgy*HrD9$SMWk0GN@h?+_ur^9Ku^03()s6MMD-7A&*;(R>&+{ahNxZ$u^P%vE4C2pn
zbg3cd>!QYNV(i1gJjZ%7NBxVMdxkV{`#|4{lKs5@e8X#0=$5R)Sn6z-wy6<7%$>x(
ztMMicDlb!?qt{Dy+5-*wT<>r`@XJ2VNmP?B^8EUftVCBU4LTB+#nj3d_jkwfG|qK$
z<{^!B{V!8F{xngb51-qzR@B;wxATd&bBVVb^<?ny7PJ2)Mk(uqP~LYXvDBp(@e=Hz
zv6kr2ObNZlN~BaNU_s21e3qerHO+$63Vfml+>F@TrjY_=#_YEqA-2Ai!$DtxA3V!@
zcXvfAdP4(=w=*utvF8r2zdLzIiWmp3aONUVhEp5J`KstCHIk#J9?xoOr^m+1pf#1k
z$3cuPvn8m`k>GTxgfl3_esVtvwcnO}{qJ5UkL0amTfRTsm}PrYjg`b<Givs`qKUt}
z)+eYH{5&Z`dYuH`Q4(Awf7-ZFfg83md?#Lbb(UlQMmefmlb=nN<I#LM^ENq)=}oUJ
zpGVK03Z#5d!0j8oVV+9(vKRiB_gO=3W<Xsdg71@)hg8_FP@#`pg^T?62|3(a;%yIl
zl;&8-u#i2(Hh*O(X)Z%<q?G%~TwycPi5@6tl&lsrpHYO_fg((NF2ZmFF=8r3%zY8z
z&3+NQ3`KYm=8D@(Tp{i2itQ!N_)NU@@w3DB-Q=}@EU{bP3R*WSBx|jq%dkNUjV*gc
zw*1^W!0w7Ou5@)la<T~R)=DthOon~=e9pxA?|hA49#Wt*O@SrTnd3mb-A<p@1NN#N
z+jIVRl{=OZZ>#753}C-!xWog2{P!Sz{$|9?XufVYy0eDJmY~TD8MYB`qlmXBx61L3
z{e#pV#M?8>Q6%2(A>O8SRiXpywob#87`m9c2E78UVw8A8Ty^Cfj*cFumi=Wod0C3L
zH>EJ8_I_PVo#Q|6lDP=;KRCeH(GK0_SflQU74!lvp-$m*C*G#*x4_7R-N@*_8<mI6
zA@?;$_8u#UtL&iP(;lNf+2i_LC*+XB#t~m@)vl=G=eBN&2xo`5V&!>P?0Vsh*mcg_
z>Fb2VatHJp;sEnK_IN$dmi;*!jC;2q8~C>izgl9~HA|#lw?x_pOZ=T;1%t&_D2ld1
z$t5eUYN%n(>~|PW{)Jf^>+yDGJ+lbwG2&bU;X|h;f;6j%5ILc#AXqjPra5pw9(TX%
z<unzpW#|cAxf6cp;O4?}=2}c1z^}b;BItK)BCH+VK!_*aX8Sh~B!#u;wfYn17d~QU
zuQynKp&BJKnS0#55@Ywjz?aLVm@2u4@)t$ig;Rjgzj+Y*<zdC3JbVt$h4*9X$j`I!
z<a7=)zhuLp)g7q6W+H`n`<-|@&5?WZs13(I&BqUeB3w5vz))i8(MQ|||DXs@jus(c
zViD@g3bAo$AqM>@fJ<UNwORU-wYf+_E+)UuK_BAnx3$?gJ}C=HnRl?eeFpOlGO+yd
zZS)&?o8Hbeq|8V|uhdjLVBdDt#}s_FyNRx*$?P%RfN=FXe6J<4N0Eq^=82GvOGJyY
zi4fAS;(FL+w9-q!+qm=C+v*&9@Mm$>Hx4Vz&Qi~0MujRCE6HW&i=z<y?G&z0<{rC|
zQ5d>53W;}4!L9xz)cwLSryvyL)(~fdLNL%K1U~NEyWT4lowkKwcOzo$!4Pb16T&@y
z!6+96qv&WbW^<<JT@U6U5*M9UG84j^yHGxGU-v-n+1}2%n|<6l<HQ+Oe}9bmL9bFC
zca%}v^*Zc_s3bppNalV};_aYtUs#>>Mdv^tWO5gZC-HXhP-YbO4B&I)EbMM-%Bg|q
zsG|p&S+?a8e>7efz}d$DShK$y<`#gX+{N;t-j|sN+>>_ZFrG21=XYBl_z-U|FvF%@
zxCb^&XYO4VwW$tXSZm}7VLN@jotX2(9@#T`Qhq(t!aA6~n8D0!*{#NF5AK>IKeVP^
zn?gJ`BM!HEz^v3=^p?_pQdzBJube%gOc@&SjFS>?KdzOcpWw#5flg>8c0&9&7tA2u
z4jkl)LwPP}`h}V9^gk$!Wl)pPoOEE{DCWPpcX}!tDA9%I=!UNJxV+?f+KRqPp6TUh
z__b-&$R5j}uXM!(73VNQTycQ7yMC1e3}!iFf4vkRcy7<)c{Xpl9BY5dApDWz?q2pW
z*|S=In|#-VUL1bEMy|wC`j~!Gzim&v9Z0+#u$ftOJ2=0?8pFMYIytj}##2)}_)7%~
zTNPF$$S~bqhOW=;v96Il|F#{XE$O>GLoX5WHc%nK_&ZW$Ig;DYm-1|tA%NK4jd=O)
zAA6(?W#~j*(SDc&m7#99@yHEBWFjPLWZ0fb?lnS!7BXs&byA!s-kxqme{!B2sZ+=|
zzR8gHT#Wv2#5n9DL2eo`>X#h<c)te3+l)Xtx)N{y(@zz<MA$`Kk#%xI;1&_8e!0T#
zHm|)SdE6{&ri<w@<=LuS!^i0(Lp1T$iFm$DL=H+G+<|!8(^ibQDH062FU6%5GAQDx
zJv!6V{G9%{Vii7IW{;GbZeDBZm9x~i)SLbra!<`=H4a;|Z}>pTop9tzyuac!G0rsR
zV;U;Pocm&YWxcg*sf_(?`e&BOai}Hv8EY?JGjcz^X8Wt9xSuP<wr^78nn=*+wFDpI
zWO%Bj*2@}c;B5N6jaBHPufmXhd|m8_uf*GVyrt=wLqyN>7}h)nyLtb6WmrMXJ^xvX
zX-(Xix8j73C60Je?u0({);>8Pf{hnngR3I+|3d%HD-m)GU4T=rDBkD>3Fm8D5pRcf
zamCXO&bWBk5z~EavAu~c-u||PXWU-QCbrDgSR>x!0EX?eLA}BjGNl8yUUSBT2`;$I
zOyyz&dVe(ZGPmIT3;js9KJa;&QCB72P98yx!^H`=S{^{2&pt?2P#f*)2KO0m*qSJY
zWvCQ3M`c*PpB_B={EIF!m*WO~;?$8MZ%8<2A>mFR36AmpZnKuUGl00;j`dS4-*Zb@
zOYyqS_M-m5_u;(3O4OPuG5kN@olcHp!(<r6UPHIlGOT5fdQLp?Hkx}F6jChbT#jX1
z38vj}L%~-E=(gFxYN9m;QN!)N%92^^d)XhiK;d2sOuT50({Fa6t=UdY9=8cSL>u5W
zYXho0Hel%Q4G8SN2koBkW&gw)S?Bg6fHle4fA+Y~GeI=O1ryVpk-EtlI_lORVw^C*
zfoI1RI~<;A3*X`c7()Kq<bowu-mt_5;;kO>c0rLPwv4vI=ml07GSLcN6ReP_t-+Qy
z-|;=^J8RS*2rBu7kB;@QQa2FR^l2zq-eSh$*rr0{Uh-SwZ5(&zJ*4;P`tznjgeh~7
zJ2e*;&2J{uCo~maZEGU9t^A8W$A6$(#~&znu0g+&kGOuT5-HOvFtw^2_wAqIM0zQT
zmzHAlo+tQs<sLdsxr;l`i!da&fEg5=(OHm>M}za=PJdKua@(ZHY#en=#m5e*IMg5=
z7i`n<a(xD-p2&c~y-fBU^RdW3pZg>DZ)`p$8s;InTOK-R6mf2@2#>1^vG92zehw)_
z=W_*UsLDsn!+BUoyzMnD7m37M)v|1yo|lQPyE5RiJ{^ha=@{QL1Ko(XFGk!3UZ>&G
z%rv+pr$XK`6|(gyP*~l>Q%y4Z_PhbzmFo~v6JR?&0VfwGVg~WHKISU6S6o34_shss
zUc~PP=MneyEaJOz&t1qFJYE_LJNwhf+esZ+FB+kiQ5dK^h1q?huxMx$%-Tm`O9JP2
zwuM7MytO31oxU;@r>2HLe?<u9Nkh1&BLuT&Fk9kVFns46!N6O=sM#9K{FkHXxhfdT
zGLE2gw;<S)f5yz?t`p+z^!J>h+3ttU+x#)diu=ax{5hM&y{u`>rOoxjF=DNwo*$YK
z8;^hVg+~#y>cDJYOXiXLFjL2myH7iE4*;<_<{tO0eJ6(VzW*-s$6gEWM_WVfnK=}1
zxa)l^x$n$c?t3M^``q@!1!mDMXvKNj|8{0hKaA08?hHN1eJvK`vY{Thl*@geMIN}v
z*-rmA9{9C|8WrbImtWLkEj7R@dT9<2YsAf%&(N5hMXN?3eXkGLv#VRL;f%iqI}FKN
z%;^DQ?pxoDtdpq|c5TCcOr|TM+q=Q!l^bVm-Izh>1nbi-I9%+4yTsd9<OsS97gSwz
zLEnX9o@HVb$R#jgU4FoiJ;{bj$Ul>-8PKmYT!G&&RG4MVIT}8uvKuOFr~bHjrWlXx
zUD0oqD`xPs+G~OX_Wq{sM7-VkON#l#+i#`JrQ!UBm|S-XdvpoJ+Y#j|lq9JTN=&-^
zj{P&%`|Xxfi*u6UO@Iv5#M0z1DhzAL{?|ZycyjpPbEtC?Un3WjuXf`6;3wAoYwR#-
zo;_k3*yGGWJ1qX^2tO&$<UBVF_K;xH4JrLt<QwnE3trJ<@`zlCXIK_}ttZN)7}ZsZ
zguW649(6<fDL1&Xht-!n^ZE%Xq-`X)&a*RvSRKQ^-Gwth1<cnmA>Inl*b{$1Zx8!m
z&Dcj=HCV>^HW{pQ_<W4WXK%~!j2Lph+6fcZI%BN63k1F%sz0u{SKtbd1+M5OP&1vz
z-+v}1zR_#DjL*$Wh6Hlmlfm5IdQgmsmLgcJ6=9OK7~MQ2=z2<uf4okSdzI)piE}2;
znC-)BbBelv#Gjc%#M@C_*&A#?o&O;FvBW|1KFqyhEoQz_it)?DIKkNwm+x-KrdE3?
zQ;a5aS#!`&BY7>u-UWQUi1o)h5g$g%QO?(FTsE<m?~6ft5`1~VzF(yTBk7wpZ==HQ
ziR}GyerbI(70*&7q)W-s{!ov<#@+?>_~X9pb=Sz?$?H*kj@KzvjNdE8NURkh^|LEF
z8j{;CaD>B32Y5=IsfoIw37_x995<ZU?uucXol(HNinagwV$Iyp)6*4W*{|LH-35&_
z&Yb^p!ixvCcxrk8U-Ndu;__Z(6kEb9bRQ<oJ;2YC4c@!ia*o}Wned#&z2}Uh#9Qfm
zV)%OUI&)X<5s+f@4M&9Cc0@tABYmGv2rO}c2k~~6av#E$Tce?cHNx&&G27D`_rtBZ
zi*FxRkJn&*3;O(4OW|n4>oQr4-h9ml>=9u%=U+4CQ?n*N9m~4zdo*W$`B|{4li@zt
zV`(f$F6+A>^2)=lSPRSPT_P^~6K`+!WsSq0zCL>#?>Q4|JeBWP^5&FHoQM4*#?4V~
zSX1l@ul06V_16jqmsr7Qr6tz8@5MgiZGrc0IG!~}<Exvo+jt{3Yc?Qv%?5m!xdEQa
z4cJh>0eRmo@NM^A1dOvr0`YdUtu5r293UFtibvD{d^4Oekk>x!9(%XM+i&(xFk(IX
z;<F82Tkpf-2rK%>EMZElZJ1<<lK(jFMN23eSTXa(3Khg!4R1>q)}YJP8XT?rjzRBg
zA+D}NU!Py7i~5O<>2)w4REL$lYB9d>2X}{jN5+cpa2fC&<y&iTVOR}Bo4(=g^%~5+
zUxPniYw+atcN`D;L67}+eEm=ciR2~bO?=6E;R(v#F|YXSLsS~xhqm!O*c{A3!tY!(
zG0CStum~+?6ma(q_smYu$LfS!&i3RYK0JqeN^Zl~DiueJQ&CoT8__oDNWGhm$Hz0M
z9cCgzo`;q{@-U5f>lvL7ch_9zu;$_lwd7WPis1jc5U+Cz`JT&1j{*5m(eE6$Di7ap
z<>2~`Yz!Hdh3O6MKu)~<csK(;A9BW4l8$j{>1gPmj=C!;_)?XI5aMl;cPgCuIL4Z$
zAbbe-_^6Z7+2IOApA#{Rczekz9$#{py?F5=-rVQE{V#FWCIRIx7t!zSIn4cX7W>lU
zaAjF6npd2L-tN;Vc0Gk%5vL&ZiNZ~{Q%EPyjvp9>JME&do_ITTYdGHa4acE0?z2;b
z;@a>Kdf`IQ$vFh3#M>A4%txGcgxPCHF#AF<miG;2uaA4hnFFyfj{C^TOIvVOwFfgK
zx@r6nQo;QyBmEF*?}t-jKNz1Te@)?z(s=Ip<bIeP+^3_r(-+;iN9RS&A+#snHcjSk
z6>}f*b06ft<;?$99~|xAha{yRI#v2HubbKTgZ;6AIgFFJqdWVgAD(f4TvOgAPG*0Y
z`!qiiZ_l*$MIYkrl%t1{z-+uxP3bMJWzN%l=Frjy97?=3<?QOS5ayewd0^xw5BPoN
zJPduF|0>mZ+>u#q#M_Q<h&SswGoqn)2pZgRV*hLxwb*`~+e=~pr!zG@>c4fwMU#nA
zoIT)%6!L&q)Qc>NoiS4FjO<cJ)Ydp4aI_;%&Tzy%r8CTU_B7xgmCTM}e7YsZGoGtc
z$gj%UD<B!Bzyz%v#?)6o@xPa_M|Zsuvyu$hza=J)WlwS9YghR2%z7}_f&1+2(b3Kx
z1IpMVsu4pzO@fJ6Wz4H5$L&Q9u}+FlXC$a8lptmZ=MYXyv50uPD_ev?#Kzs^SWd*-
zX}s@+#M`69+ga2sd$m;ITUTmbuWYgH?*ZI=;DljqI7|4E9CV5uIifutcC$nK_cpls
z&Ka4tE|_-24NH9-*nf9KC+cFyKe{jj)CKZ7XDm;XLeY^sQez}|bjSrqEnScp<bt!q
zUGXx*6~;V=XZ>`6p`|Naqs7P~-VP!sJDAN}A{X|(O{MTsNa0GXTJ~CucI3dR<RVYj
z$q?i&q1GfuGS8}=ZcgNePV`edqmRM`dQqI;FmlFS;_yGY3wBK4%p~iT*U4gdQj1)u
zk)qN~hFzR-T$aV}r@A0ws|aPQMKD}0Mm?|5IsV;k<hNO4IQzSdGrCWSb=fNDR;b|T
zM@^UYP+4crny{Z85uw6tY7;GaEvIpQW|~fdU9;SnW9o(~dpA5AB|_u{F@jsL#^Q76
zeo~5J)><PXWcaC&p_d1LCryUU<oZ$<31plLIa0#&o@ahMdy#*s$&VPs_uXg}k~qWq
zkn^;^7ttF=y#12I`4<lbqGm~vmQMYf*zr6_0$n5KB-|I_x04$+Vi)F&IN_tdBSa<+
z$a8T<4?9O>l-gq}wabIG_SjqPh<9>lY%z7kJMJ8~pmoL8*?dpexWKBr3qFK6AT`Gv
z=xL4y6L({~p9PXHSz^S#ee}K1yF1VZqjGKVOlHgc87FwpaK_7J&WLM7ydCQbgA7NQ
zDjjh-$q`X$j<^)&h;mm)3?J!;9+wZGwj*(S^j<VLw+BTwR@9)ZvHb*nw?>Z4zHs8q
z>wkS37u;@7&P=b-Auln)-nwD@ct<Q||Jk)rj8~;n9K0*T07E(KYh|#bm#NoLF)r}u
zo6OmTIpm_}{+Xjc4&ZE8>iVoD<9f0e(^iJLm!xozi7|bW6I}l|V$WIP=^i(Xy3bnL
z+!jwBTfw5v5?dErB15qke=b?z``6uAu-6>gSzF=VY9ritY(Ssd^>F^P9^Kb(fT(r@
zd_V8N+=sgm^mP~XsXq@HxF3Rp9V*s4pxnd-_tJ^Ca~)x@%pQFg+hW=TTYQvKD@d}z
zE>A0{vn=udNIL7dDD(I0+Z}7K>$>*ZovXfGb?w!)r5Rx8?hZQ%!9oygZ4?AS0Rt7q
z1iQuV?p}S)_xJoUFMKKBFx+#W&vmX7A<1rxZtq5LXJPGnH%d0SapQv<tA)42gthK^
zY#aHGZR6e%A-o-W{v#{ye3D1@3w>sN7Vr6)>J>io;LlIA9{B;k`R`dGyiF3`{`mVH
z$Lqi2bR%`WZ+(PS+zn;1U&n5-vK7x>V)ployo<ibFV)V{qp+AU6^nVWtcY(jinv!@
zxOe}g`YV<FKk_&g79Xe2lN0oAa*|3{i+G$;#L;&LN&0X=nV<(qZeF0A%zTPY=W|(j
zd)elYx(N@F;c|%Gn+~z0@?lJd9A@E|LdwlPM)zaKF}rbstNtg5u6KeFC8ubx^VI+4
zN{wn+#BUiVi8VRN`sm|W3vWF;9VM>ML1t_^z_Zf_a9v-(!QO{xRsRtA2lKJop3m@k
z`{-9WU-$of%1q7Y${+dC{^v3Fg1UWb@8r?sY~?y-;r?s~4UT5uD!g5GIfE6=Gno~c
ziFI%Wr<2ksnY<12eOs|9vzde%o2dP51KmB>)8flI_4}<OdCYnaR@uO?#p~$4c#Zrs
ztNHz(mAKv$Z@g&*orJf>k|g>FZ*3fxQ`t9BJ+JD4RW5N|d2h_6QwrU^m}MOoDXU`<
zzyB-j6o!UdC>yb60w<NnwCaj_PTuIxzQROd?)IcbSk_*|w(tb)d6b11rtC!Fvw6#f
z(%vhdKSLesd(>HKzYq)UTk{^Pf4yZKJ#6FoD=MDj|Lc;zo<L4cf?k^hI?PC5-{M6~
z6<6I{-z(E8fw@L?&^AdR`K0>HwyK*nI-cvqGr3kAPnW4<Ti(6J;qzIxXr8)O=27a+
zT*}I8IzxNUTeXz$_DZ=s`)6Q&QXYwCGx@B1gu|L)3jCB4vs&F8|IOfvyr3IR<tY{S
zYCJF<Po0UYCx_9(KSaH^p^Oub-j4}ku~i6lgtuO216eD)oh`hbFTAbcEgW7ELY0&d
zGRK9ouR|1Dd^GE9lc&C{cB1R00jaDxO}oh%@~Mo|96GkLzDD<?drbrR^K$?rX9V#}
z`%w7={G|!ejOpQz-C=*Ozm>kUX(Sc?j=+As_M&Se<=@v?U-x{g(~<1aOfmF_X6ugP
zuX4ljo)SixhVuTEiy->Rbar%@LHqET+GoyG-n({%g<&iZ-nM$`P2Oi`qO6?>D|X^)
zUq42E*V)|YN7f}DGP8AWZRo4+J6{%d^kcm6cI2PI%+ee`wya)<^T9e73CBi;FnMnX
z>k>j(<`hED|9#Y6J6k-jlUb<rY$5b$qqBa4P+GSNqiXkXDx97w@0s>;S0Y)T5{_%X
z2&^Ay-)kGmfM!$aqPa6gc-yBaf<J|^3w{aWNcAALRSIUZu;KpNU>dv@zFZH&As~cC
z;!DR=6Ca%vfr(cnvEu#fH;H1y{7_c*3}UvpykDiK^;jClX7LYWPejmPXQ>yH#LdP^
zOFd!+O`gtR;b85&g}2|Pil_Gz2Rv4~CsXOtrPW!HKUF?4aR!>N`rHlWS(`8#9|<LQ
zQW&R3hLe@3dw+3+a*Z{2Do?1k&O1>n<#E-#^}$43eQWKNy_Mq^6p6j&+q&l>DBKgy
zE3*hDotZ&DaStx?IPE{H%)rhwNRj?%$DFB5sT<6Q>H!Q+^%tiY#-MQ#1a*pJjC9BG
z!b+dop%@JzJV{n2#j*hAr2DhIx-Z6TUp8-$=GCYjqcl&&R|4r48B9m<GH=Dnf1VM_
z)s;5l5p1aY$cDaOZ8)E9$J-1eVO^b+mFSFJ88;?|dg8Xj3)|}6G``|P;~~BrG4mx-
zn%;xTczjVhgiqZ=*({808xhF<H{NW{cIVwd?SaibFrDGS%|jlHsNu=KU0m6R3yV6~
zbM=N3kvE(vBTi*VvjB>h2hvP<yV^XM^A-Itarco9PV?rk9)u0>#CNi|@nYpZiA$L^
zYAXLIQ=*?u1WrpLXmTcuqb(h1CcM2G>!_JIRCyXYOVyWtSlo0eJ$?_1qBtxV<E21q
zR`AAomlNspU055VU3FDYW}kK^@}+daW8C=3%asdyUu;|HM2||2w5njuiE5Ty`f5&#
zC+55n=FT&k({`~b{WHy^PctLApOriXPTcm=d-ttB-+m8d)2jfM5BH}|y11jaK3vvJ
zJLH3pGR}OkiScIV9Cub9cOk2$8<*R;v1pAer`_CWlj25G?Q`!}bZ2WXckb(9HT)fK
zn!MxE%lAC%`jPI-K1qY{8Q1U+Jh=0gCHJ0l<l=MnAw5xs`VCUHm9QtV1pB@vn5JKo
zmhL>0wp^h2(nS`2yGF>GYsw_KOwjF%47a<8cftis4;S-hYB3Akqze|8oAq9^=;M>r
z7Ek=0<77TNPAlPK+3F{`-siZqHb>Rxt<0u<2Y9e&zdDfilU^a8lfu3w9}3thy!~Kx
zh(h7*I^k`d_wvV96VKf05RTuEP*ynZl7E7W-Y1yic$B=dN0d`_n0|{3c)WWrqf+&~
zy6wj#XutZh_KQo|uU!0tRNZ)h;Rg>;->HCpJ%zP(HAn0Jj8D&}w8uWme%Z%cOYLok
z@8z!(yE&Ac$AR-XT&=cKy5MX+<YaNZXBH!8W#BzJgFfdnFjmOqkxwQb9vO@eOJlWs
zwxhmoW@^DE{;ITzO22QUdeVB<eO{-mIPGa0uUEd?2D&z1&-q1bxO96p->#-G>HZ4l
zNq>C$UJ}VgNnA-<PTQ13j`}1j&oGhw`OA1$FHt+xB@EI2w1%=}s|x21tWKbY@NLu-
zb!)#@mV07?IHUyLJc*}puX?gC#k1vGJpXAQYw$|ob)Y)MNsw<lp0Ul9@xMiRy4#iS
zFT8DKP{+*YINIM;?^cUA2B^=wiAg*=A{UV^ymj8G{<4_~T=G+%;-p1-eTCb~VtlFR
z(JVZkXX*-dRxVV-jCkCq#^d`(JOj2bq{5*%xU_)d@`#QaufA+$(Y%v)@kteFkDp7k
z`(rjY#1$Kj+G%c>PMMa{+=z$$F+{pXA9+GsN#m+nYh3135<5k)NAtq)FQEjAtE?mK
z<$CK7dOL(Fmm`G8tHJaO4JNr>Finr?yc+D!?+(7yZ5u#^w?WDk)V*(Ui1bmRxHv^I
za;dUXpH5{%#x(p^XoegSh1K~;q6dZ(I9dBE;bYeofz)lO^ZnBRhOG)<@!|k3PVp!4
zCw~rS`>{ITpZ&}HX>SsMt>(Ax8zPvzGLnHeBN_K^D8p@~EA1XktCU~{==(hlvgh)j
z0c8FhPHJgknD7?eyO+089;<lgcJgfc>)!TT6;FQ8bfKrXFk35U?jCkx#bC`X4}FQV
z^JA3sF>AAYc-_EPzH?tf8u^jGDv%zQ@=gQ<)Bk^2x~qkG{k6v%A4uaT0p$FvJ(z>e
z4F3gFP8g7;`*4GpKzSzuXs12*P~q)0k1+DOgwsAR9E-i76xxRq)mr%9AOd%3MNNh&
z51_?VY6x$uELYy$W4+GeCO?)BQh!LG`b6aae6ATcM*GGGL3sXeKJOpG&%)c+Rh3Jy
zGJ>I=I-9HrU}2Rg{(J7p#|pmm6W*>K5lqh|Vbt#}K3Q7kr%koP6ZhLd`tvq}<S)@#
zWR-UIjl=<jhRu|gLwvXJwybzzr$+jiqgiQt7!&%)Q(2&${gV);YtGLb9mev(;yv5x
z_frtT)mFlM%~p52YNsu{t>mHotho3|K~Z#+ZzNGOR+jWjh7R)TtP1CE`v|6A)10e$
zZQyh1ua(VktE+M@W7VlDyfxR)Nu>Zz#OVBW*H2yuAL_S>q+*W<tn|7*)IQ@cs}O$C
zx$j#`e>PV0rN=QJ77G`9{pH85C_i<rda>-TC!^{FD`z2u&^kK*h$FoeVZ-PXHhg+0
zy!~QB%dK|oSYl*ilp_ucof-AOnJN`rnI`RQ)!SaQ{o98$&1WgQ{U|RSnPMkCB+#E7
zvHrv*2D02z`{Su@(owi`M|iv4Ogqh~9z4$Tph87Ys=RU`=dLsF%Gz^ywi8~$+cD0r
zbgLS`v!Q{Us^US(Gk57HUDUPbMv$jF#-BYfXf{5&#FtW6!dP-6ih1JDO?18*KE;l<
zRqXjqyVaP64*Xuv0kctF%A^aV=MJ3(l0rGIJxAdWU3=0IPx~6k^IL(M1-;po>cnN?
z?V#mu9C_oxs|0t(E_LER{e8=1XDS%<zPRkfi?#MFtZ0p)q9t2}x0SD%E2qJn>7C8l
zw9u5V!rSF#EU99!WLRBmV%K`(Q51k(i$Ko24B)-a6A_!WpDFO6;cFWXr@69lrz>q*
zy0KXE?Ds?$dIh@DaFHuft6eb_-tI`!j4SN-sNhbru=cJV1Du}lcI|VHet)W*;3uTc
zdrXnlBkGiTKxL=9STwoK-|n}FkGo0L#y6CIRKoT}CA|8hgf_z4L3Zca^X(iHC!8hX
zzcYM#b%xOJGdd4k<m<W%Gzlt}KD&tF(lp&SEfU{Y#FM4RNHZMe&Z=V^e0h{Z>BkxN
z=mf7%i9hapoX5e37`WgdE6(iaa>jmMHQujX#Xg2!$RjK>8^f?&(&FS`CqL`(2|M^P
zDx1N%Irv=6=ku>eIH+ujiYpIM`Bed_TMMX~P(W+*0_M~yps(<@_GV=-9@)pgF8k#t
z+0V1Md^R=N&zoWU`RH+gm5cWCxXK|~mpO#L`98e5<l^8g?zq8Ta_j7&N9Db&Ru@4?
z=`4J{?cm??I~i-8%^uq<2DHh-Vp@hcw+vP-%V1XdOhydIP}j$HoI}+wTW=ey12)rZ
z-X=zW-N^d=8>l2cIrGyx+H6^e!yoH7erpxJ=_|=Hq>$ENC8aLNOZ#92!Bdl^y-Z@-
z@g%w!mrJLhEQ+p)G!xzqH(7>d^Cil%)4bY$5lKawXN94iG)EpSQWt0s&9o)){8*a6
z?&1V)PExmbnK-?N<7rhff%5<R-0sp19~Oo-NMOeqb%3h7?%m)xeiq*D6rX&&v^rF0
z3mcUWU-=jHylqx@+i&rNtKV(YxCHEbYku7%p4mKsb)yzhPk4J+d)kefgH6=Yb53|W
zP~ZP&;ceN#czI*hB^n&dogeDL^i`+LMtLu^SF_2U!Tfp30jN1uKB5TTmW#wLbvhQ(
z+y*NHz*HHr;m*>o4$_R-WxD3+P-2D!@v^(P(uo0_z81h_ai}hHLuf57sFzs?&&mdI
zyLlj6HP3bZAKoSe@%Ui?19hJ<ZR^YL9RhfJBZ&O(!k)+wf_@f`3U6BpZ&!<Z+$6ut
zrjhCdsToC`!bqZow@V|!81^lQFK+|*QAh7f`2>gxz*ByA$HD%5ui;P4^?tC;A77o9
z?~e|U&LV^@^>trPQTE)=K}2@)#HymB`f!4YT;Rm5zDDw+jr{26fN{DbUDi0!<ft>*
z8(jG1xT|`<-Q;U>C)3MaIei|~o#sOKcg}2V?TqgMC-!V~VzTac!-sf~=_dcj4j(!{
z^kKm?Umm~krKj-r=`_ul4};h`JeVnN!Td2m-7&hi4NmbV?4BPjO9c|;tIWUmLFyxv
zM{{8yOB%@o5wH2Gn=kz}Bm2|{Vy-+XZl<Bk@eAWj?J$ht!eD89Ts0s1L`qZpRQj8Q
zV3IxtD!V$6No9gm9}~!}M|zEfD{F+E!-PZiW=o6X5lsE)5ax?l*zOj=FHik&Sm8&H
zhCX;5mPTi(5B;SfH4F>pX|%Z4Zpti_&S|1@8dAq-_M9`FulMCG=_Wmv^sF_#wEx>5
zg-c!(dD1&AlNP(T@b<TO`HAAfs3;z}TQhMULxQ;5H$?t1?Es|}YIZP;hdrfP@2rf8
zF5*Y}O(jhGaf7?K%v;)z2So9Yy}UQ#MiVMVuq9tQVa@huwF77+-u#@S^khSo<)M6r
zLO)%bUG>-<tiDM3U!VHpf7_4CzxeWNV;^ST_94qlzMVV%gpLkjQc*C8?g8{^;Lk+u
z0=AfYlT_ZDk(#OI`uTCT(2ENPJmmRyCgc~R@X&@~ORc$`Z%ttjYogOF_}#}K@1!mJ
zV~kY1XT)cv1M#<<`B=@BbKSf+u+@vHn#+FIb>SP~Nv->CxC?K;255d0-q!8w%4qSh
zjr`n56h^(iW>4cP4){!T;Kx=6Lccpmi|fL)_KvLl&&Z)fBePE0@Y5CpGfo?%Nw*;)
z--a^>UD$feg$h+&>F}>G`Mf<3>U(l<pf7WKO4BP`-CWOx_pNPt<Zef0;cd_ed#=~D
zr&1jUZZ2_T=q49>Y!1euP6%B~hhh0ZTH!4rwCL)_o`=#t{qCszPbbQ4bf#XabWnHQ
zX{1@RgHi7%&BskfYEIRBd(qjEQ~Ft&I$KlYi3Oz_n$y+aoEKi^tpDAdN4b+}%@ndS
zO=$I(88u?f@x5Wejo&oup0%RuLTf_*vQn3gC66*KX?NUK+7crXDUNgub)~;J<Z9mb
z#3nmZE83Z&RjwGjxDjP=!{n~?%EH?Y<)u>=w|vm|KG!-tlvnyGjfA&r=Rd~J_=pQH
z@3Vc|J&a>-W9f5?XJu|OzR?Z7Y}DNP$3=2_UZ85~c{Z7yXSVZMo(wpPkMMT$vomb<
zIz!i^r!jLs%@>zqeydbWxy?m97%x8g;Bl6vA7#?`qXY<VM?E+~{cnYoA6CfI&4($>
zImFRf1^hboAddqN@_pKV?3eGQX6L<h?y?)BV>VUZWs=u<2MfPvlGJ?%6@|AY(k2DH
z-plLi`Ra){z@>-ES?qC;u!RMD_AMZ0Tmg+<C?jgsUS_Y~%lwi({I+5rLw?IAOL&|9
z^L|Q3?5E8i2k0NXAD>1C@GWzIDi`+?*gjX<!d&iEl)q>EPGM{&;d#npES1IW7d!ZJ
zJDXABlW!Zc7%jXlnwmk>pBeNL-c~7>$=ANhkTPt?DI|?`b+$3lbu*qFH(_;dBPo^}
z*b%>;uJ70J@5*(UcU_0)o>i1hNuixp3ZE~hQ2p@#$6Ke!WOc42(Qj80w|`nrUE%HQ
zj)}U@ETd!PWpu8%ge!N{!(BsrRqq4>+r{G*r9G)0F2dalnq}9H(41K&UOu@v+LeuC
z<BJ%68ms;{W%p+tR1RR1SbB_#rT&Uon#dPB^3OO%&y1sRjrz87<M^yUJN*$y!*{Vf
zSf@OSj`2i?EaKGI1WbgtY3eRp`7oZq=?Ro8h^La~*dXEUw(7$AxOhrET8Q($h1Bwo
z$8+&Q%!AY!S~r$oDk_Ij*#dU%^5NW*PDL}$^|R8AYKB?cJpuy5sMT0GYul#Mq@DVq
zos=oQUAl=2(o9IlKg2wUt6c+WQ%hQ4&6?Hc2J+~Cc(FW~bAv-Dsvyp?Q6Q_jD;Gw%
z(yLHhkxPI&y#4v9tB<mW{Kz{QL`g+)dg0=qf6|%yxjZV$d^hon;NxEL%2`qDyeR*P
zR|IVuhRKT-!se|(M7<6ms&*i$rviwM6xQ$ZCuyJ$ZfpFwuvi&4=KkE6>c{hrzErB}
z&cs>vT(R`zr&V6kLVL&$=g5LL_S)APaW1f9VFNqH_i^CSXh%k^cas0dnR#(86rOTr
zobdLt`~@e+yQ>FI8caVI>b!KOZ)0cvN!D54;KcGr;@3926Ia)VI!k=mkgiNSqc6Rm
z`>K~p9Qt(aP=&Wugk7F}gUR#P+#Bu7Z{K`aa?6jBqyVzNi@O!q+r>YS+hqdr^YkZj
zs2|}&eCSfim$Lo=?5-5TY3&ZL-p~&5btpwQ!}zs)IJ2ZPTcp=*^u#be`UpqA2Jm64
z&NI0I92o1*xC@#agtxi9gSC?hB>Gw)zi5YhdQ}kZ>jlb_qu-D4Hf4`5by|B9@zDeH
zb5!@x*INgwLpGG5!nYI25wxfs#kKX3ygIJD;JVVcNr&aV(uWW7)}-opHc@z6^JEms
zGsRzO7u;!qe7?hVeKil0E=T9IQGvRLgb@Bg9L2AJq&o!Dt$P^%b&8;e@OG7U5;Mhn
zHgKKF_mU_YI7CsTJ$$*l;-1S2Uw4PINZuTmf^ZI>QHG*X+<K2048EYw80p$GrR{ad
z(=4yET?^sro11=&siVDnO&<meZ$ta|lBLYQ57HZ59p}l&4tibdN?*6rhqs@-nDxPn
z|8l&@iq~Aa)QiRYJs5UfJK5JpzMr>Y&?PHo{A4L#parL9TQKjcH8$UDrNJ=r>lq`h
z$2gGP+Y#d{R|b6Y#ASmQ9lv|A_J9|z!rQOiJou(Lw2YYxSF5`)zmF@;g&~ES*%H^-
zvqd=Bxsn53vG(NkcA%=61GbkP=y%Y89>Sb>Yg;;ww83JtfjK7((*GG~XJ@1S4;#Mb
zxiGSnD?#7v_%1veWNQ@nYov?L0+-g<k@}MjZsiPwx3!`Fznana*z$Wddk&Shr}8ji
zeR+GjKbFp@miE%$q^;AAcG@3C?r%17Bg&qNEgj`~aZs<P192Pd`PkKnRV5=z0O7?r
zN3w@Gu|CLwggfGkCfjk@$r@X03wHl7V^4u8uGdYdS#vVKTTarRehPhJrjWXB3VUPB
z`SH<IJ)-7ZZ*HaRODh`Hw_?SwmK>O9$-wWH@|PQE)WOL6en$2UGZJHOPfm9Sah@(X
zEO%8tiW?6my76tU8_VapF-?zEGu_m!bsy6v4>{N1q0V;?F+9J=<Qexk<9(NvM{d)5
z+-<ysw-&<N>A&3Itnl{AFBh3H@&fmA&#Q<1JjUUtX<Gd>^?DXFv3#-is%IE~T>jU%
z(;Rdx=7%&;ua*}vcVrPUJEW0HJ4)|=j*|c82<A(UQ2kROua_1wYiuDA=MS@HRROM%
z2iYUMb@?CmCG4eE`fgUg%p>-HbL68;%}L7nzbxEqx`UjcY%VzNVoCH~-Mcj7j^EEh
z;qBwXeC3|zGobHY#`N3Ez!7^HDeUexb`N!@%F83X9VWb8MLzp3<?}#zn{|Fai+|3i
ztMIl{=>zl@-VSJ~Y{vYZ)E&N)-siJ<`ze#PW74p-Nf*bT!R3N%`c26uxpo>oOQ#Xz
zlFqVD8GKxlA<x`)_FPYA*PL`ZjY}h|);0q6Z6vtqCXQTCH|;-b`2EXj%G#`^h2=Wt
zcUi~Kg{zcDpTcX46xs-Hn+b12G~XWgOXip1%W3pJkr$^DS)ZJUxA1oI$ECE`y_EE`
z^3;|rBK~^<-%Jy*XdJIuTC=UMFmFLTSNFw}*)5(oe`p6QPI%uB^>&&>^JHu^b(Sj|
zQFv?FNH{CJjZ`<=^Dc25>=lQ*A&#2~>LHsKr;M^VHgr;dcU~;<;+y^2E#$r1BHrFs
z24lYR_eU$&@O>Ohzr@o^v+WzrvB%ylWYYVEY$#sH{D+#6=PYEid_~g|^mkul*{6H-
zPc_5=uaXw@Qv?mg)fELxSCOSX=J!x4n1r&-Qv9%ZotH<%+4=?Xwe?i;Qlq#myscuV
z8Ozj<T08u)+UL)JR)Oq_R3A!@Af{{#;-6zdl(EuY$~J)ft^piM^5=+ls?8<_aIBs`
z!&~}r)=6_muD>!d#fMD}VU6&%h5qmP|Auo)8lhI}qPX8TN?t7O6>o$Q*e{ft!rM6E
z?cx^!lp3QwSg}8Eef+6d+Mi3OwfD;LWK}&MYE1N}o3eeLi5vW7vID99wiw3Qk+#=@
zKVmE>cgTSb$Lz@;W29OeJDor5$jq@(=8A#LX$JYh4cKQH=>C_1s=wOMVx8`@g?5Zw
zWzT=EE`)A%);ZIe1$K`7*y2R)0T(`gbyF9H7j@!%@Z01=v5zkgPWiGr$VdAX;jlQ`
za2rpCl=hV8Q*+i8AMrvy*zNVBQ-=Ki`@}0I1=9R}00rWh2b=n{K1CdLXCF+S3Jae5
z%fGETb!>>Vp}|zvtXT7?cFtv_r7Z{}riM6$`#SRky2}sciP=xSH2UI0(g|<nR(Rt+
z#E-!{12Fdv#A$0FKa~Vwk{m=^om+YcZwD>$qv<0L4z?54p7h}TE^&56o*X_W52;?`
z1aa7dr%BItI9zxV$@Gv2%Crq8YLR9|7oB5L<gXFl_TQ*8(vc`qIz}<%Yb2j%Ny8-_
zoP+kQcce$3q%&CjZ}QHJ@nKg7>2W=SnRp_UiJQVWv^t!|J;kq7h~!3G6g}lVnj(Gd
z*)fqEDhbz)RDA&Qz&dpdr`@4&<@iO&+a^!2cw}?sg3Q!DzPWh&)sg;`)miLRJAazy
zX|L|?!?KOuyw>x)(8_~N!rN)W+qXY^(qyF<?XP*!KSny@J03K+q`ZU?9xM)${&$I&
zC~0>-uW;q)96Pq{x8YK;=B}ESd@iw2m$^ApDwr~Us3ny;2?O>U>11I~zxs|WE$>2o
zb5G^hc=2YMvX@%B6THfemZ>hvEtK}jRA;nGE=0U^<b1dTE&g<1+G_E|?}d-+gexAJ
z59ira*5E+>GY*VAAYGH@=VuPKEbnH+mo)}HUNI=Y&!Aiw8zv98Vckv_Hr;e#?=w64
zW*gb8z4F(&M)q&eOdD#8RTJUtU2DQe8R)phhNxUy+Ele?%{L>Hjv6sOGV<}9X4XUQ
z^c-!(tf7%Xy^V|wG18=sJuj!&Goq;jZ7Mr(Wu-k`g|}wKcIq*(kU!mmwwiPAIXUp^
zs}*w`P5Eh{8BcDR)2*C23s0EROL)7m?POM#OyK5;@w~b-o}1MsQ0kuvm}sYa(Zigb
zugul4Z9yyH^fu|3-%qgQ&Q?p7d)csKnGL=4yldXI<0tL1JGOA*)NSpv|I^-Bcw2vz
zv{KXD=%=0Tl)7%r{-C{g@eK}SYHpo;gZQWthP#*GIIBcm?doSwzs;Fuw=w*4i?SbX
z(&bwT{<SXf<=uJS{&|7<dFQb+*Zx)-;KVJ(1dLIqK)GVR6c+K%tRl+%SH!Ykig|WT
zGp+Ix=SLr7u=1O3jXKJd4@XE6-quezqI{-8&JQZ2O|kl9BM<RvWr4D|v^TwyhkJ1z
z@6+?hACg18*I9h|DT{^qnY3Ax$<^tZSX9}8&x{@9kJ!cXa=VHDw3m*1_Ug>Hm++`P
zTy4LX8p7M1f9&Pnu)X}4yO#r__i*jz9)2uRPi>8Se1x}McI5MEb3QYV<TIo~KFw`+
z6JB+<`2JjeZjnp0O*U4gv$5ZhC4Jgfy1q-*`zehkGq+QCBOR;vX>9!|jn}2p@DScQ
zU)hfDD0RUXq|vAQHoQBnWqiGjoSd*xUb>C+{cRJ~HQ!Ddy@t~tR<p->wYp_i;+vR)
z+p(35Gf%;}yD}QDuApAh3X(&_Cm&BHbM$gD-X*fJAdxpQiCm6Ux82`M@Ex{<c1;%X
z-ggl*-Y4+6r8wb6@tBQ?=G@sB?MPzz7!t>)uG-c1)6O(4n!2-uv-<avzR}8#isqMH
z>We)a&4V>DEDnv8UM!9u<>JWe6Nlw*arl>xqwm{TX<B1(S|6iMl^EVwMdMX*zGn2f
z6kVK6OXd6y@zqXMew@$oGpRUJ8RWThc(g|SIqIk}sS|_6#W)HoL{S&{H5*6=Tp^UE
z-@@pm+}LJsB1n@5bL(~CT|_VsYXu7%Lzp8C(yG$h0|o@p>PG-3!cD6%e_G5}R%vH{
z?*8t_kxqW<#q_8Cf11VQW4P5ch-u<ZZ!ZaCkTefJ_0euJ)Sm-w{pI8L=ipxf`knbR
z<GeQwYWga#-j9wx;#FS;Q~p2g6ZO2VX@}WZJJqk!SY8lE+WMae7B359fM%deUV*gO
zBu-khTSwt-jlJ@}3k&W~^W%D!JA;g#OnoDN!Xal?T8QiQb5t*qc))emnDnva>T>xI
z<o}zSCrxC!-mhoOwZ}8(#40lqhg&mxtU)}mfu`FHJlSTzVX%P*4Q-g&(a3;`_5}Q4
zPlqlJJV<q>)Hf#@SUb{vlM`!%w=?UxbEdm;Was)YevR^PG?#WP^5N%Tb(TEvCUlrL
zV|D-i^h3M0OCFfVdGdXs-U-=$biEkBk=4@0Y8L*!IzT*c0G%g^3tjKagi1cVnd!x_
z$NIV~4>pXDcj~nEz#l>=5f@wMQyAx>v|oMi&a>Tae3<0UbZJIRPr9+Tf){@8yx7{;
zTUviVmag^3UOVm9(!;LU6Tk{9or4<tGSJ6ceRj&jdnkR5_I?v<dvLp|8{3OLX*g47
zmeqb7mrmKENd&iqbIGTKwcGXle)mv+oeMX1xUo_A*hN^|PIK<}$SAz)M^W-wSwv-(
zTevKW(-Wote&^4jfdTTa`m*4W7su}UldqhK_isYkEj`+&r<$RAL@=b3d`I;nIrt`A
z-m(x@UXhk}th}(B!@2Vy46A10BqS+&Vp|07q@f#MN7>HOuQ!{c41GUkQ0O{stFuL}
zv#!f9KkoJQVbnTrD%yKdGu@5m+K2vL%7yW_T)CX$#R~0d7rBT-zTqLAusiq4Nq<wz
zg+JV!xvkuWMq3@3IL$#Gd26<8v|`jXQ=UFEVRuJU?9Q67G|&>)_C^ZQjU=0iH>vK(
zs~^s+s_aSm>7E2>&$+UtyY#@~Og6jFA<Bi7LtS`Z#)Y%O+jkKTtZeQ;!yMt;LL;}=
z8kwMLZgX4VD@H#3V<c36?{q_a@-<s}>zcf?)PQY{fz~$+So|*R9A<-;9xlN)%zti2
z@8(A7b?or(U`yRAHk@p2!xZ7|)@#;`8)2Z^S{tfYvZG^7oi)4I5g^<&eQ2b}VC478
zMjRg7$wzL-h}K3tgN;18X{6>SBiDuB&EMOy!_t;JH*BfBPoL*(fqjxW$8MT&c9$6?
zlT2wa+>`}>n6diT$t*D)%a|Uc8CEogtA=r$-#vjPd&iTrX*`)#C-8Uo3DV9@q}Be3
zv^{9fwLTW=8nmErzXhkKSh6U{k~77YY$(zD(%nWqGPbzBG;+6}Uh683Tom5EXyiuD
zAYJpq*|(K-J$k#c^*2`%cHbaP+Tt#gZs04t?c-8{ab^hxp10^d^)|iA+}8W!7BP=*
zN|R6`F8u;?gtr3+T%cZ+3w$s?O;GdG^q75$$Qi{HXuj=oxQL9QMVwx9f>Ohke=NN9
zzk7^e;Vr`3O`2^tZ#hEn+#@89JHpA`g$x>4$o(;gsW9{)_s<`sjPSOv@OHBB)_hwY
z<Fj|t?6)jD%4X50WhU8mG8ooI*@zW)a5!>@a@%*QpCpgjC40HHb}#95_VS<k9$ru1
z&9RJKnrU}Yd14NSCgt$#a1JlibCi*pLvZqLLi^`aPk4J+c)ROxKC!R&F?joKbqwr=
za=A=R&0y2!H1-8=<K?+jOfyo+e4a|*SKIh|LK=T0s!Ok0Ixj1xF>2p7E(mY`JDAR&
z!sBVS+gRCR8_ip-<xa0P+`PL63+HvD*4sqWU;icEd$o9;l?*Rl$%mK}M)gXey|X?J
z)@)lUg*CfZP)&I2BD}4Yl}z(Lm-EN#M0#f@@_4$s+as1Sy-or}H{$5kco833EFyGM
zy!;^v{PkBHqkQJm@#_D&%VT+1F^=AMqS>z8(?C613vb)hj>f8SG;QOe`6f=;!!KI1
zggV%(#8PpI9%bY3?GeZ8nsK}o-i~_{D?Mr~)z+#<y=Dv^zR@f%tGra<?HlC-k4sV3
zxz!AA*w5ruFM0l_DZ^iQ8`V_Z0AuDeQF!|xCytes<eStUbKH<{=3WiKS{m4?=fYT(
z8%Egu5an5i@T>d~PsM9?C?hStV<79bm;L#WKM(%!XGt0D8ntt}CcV>=j>6mvJ_LA(
ze{1GXK!Oi<HCHY>?9KSC-dIO@lQm5`{(1p44)Nz-W$-vP^{1*n*Ie4&gVVg3)>M3|
zgCAe^1~5e)noh#oYQo!(;vAhPhSB@6IL3f*^+ITGm=mB3V;}5)RW9uTZ}o>sCz-AL
z<PlFM>~ZGMH)$20JMhy^d+y(MrKHfA0TY}E7TylIVx;~uOZMi8*Q##8D_0A~+gMQd
zsX0!&&8gkcjKUhG9P4dO!W?V<n6F2QHRh|W`P{<5z3<lKeYNI*n~|R4$qHIIC<oJl
z5rZA+H{6kpUmbZiU3$?>ce-m|SG}|k^Ywac9^k{bbv{%K@!`@fZ>lWvB4?oorDEL0
z0eJ8zMx5qseQt)b{)D%=tpka9tvp)kzZwZ|T{NQy3vc(__NMMVPn=6U*gwsK*p8m$
zPV{Dl=C?8V%7y<kjMmbL{urR$@)kGDv}->3#O?nw^w#cjWk-n@T}HZ6wY?XaEyTGG
z_hWjA@_KY`sUXfeV7U+1r$`(7LH?DU9@MVmMp1oNW^{LD=YBVut&^w1=!5t45Zd1k
z=VB@CP7C$5?>#B1?9Mc4XMA;Es~POdf*0;2)bykKxe#LOMPdF?9?)0nh?x>ZM7A_H
z&VI_`_hFoa8$GXjvbvc&57xR;ZMM81N4?bT63CQ{Agb1p-_<vmvC8&;KQn~N!rOIu
zI$InFrT?i=`PB61_;3n^w~5cEY5zE#=r+pEo)N+^eO)c#?Tdf3zZm9+YYSg}y*Ja7
zy{Wm#gR#HZleO2COzq%1_jV=sh$|cA8(A{PlZ>m<DFwMwF4%={HC_0})0tV)<1QZM
z#Fl#2gyf3roM*+~S539YoPvkz6pC_%xBV<>Qp3PcjcoDiCf>EWgYp<OuMQCRnQg~}
zcXo^x#~UWi?t}A=bQ71n^)DByJ=V4RhZ9f4(QY|mPy4M#)|WA|;jkTchwQl2%E%Mp
z?M2~jl&-%+ckHM&)Q)u*4b)m}z%$oC1>voe@OMOa8y*gk|HjWoIox*4`e>^>TswL;
z(Z^O+1g*5x?!$_^JFPkUQ8R978~iWXkme$O`d?cP7uk`2!AKu-BiHnNP4?SSK{!15
zp}xlm&D_HJ@$WSMil3hP&Q>0ATgsMb)}F2B<Y$S|!i?F2%(z}+io5XEWta(N+fSjd
z@b*^7Si-xF#<am0B2WIq#)cD=i9Mc0f#dnGV?4ip9#6;76Iii!f;#Uf&}idic?`_>
zGtEMM=@z7ih*v&fiQjH3_9j}h{F=BZ&HA@S>kNF<k@9v<oVen|+byna|HWC^K2A)Z
z?!+CvHy6enp?h%wUz;DM;|FC14?9YQYDZ|iw}7G=ySXv{tbD)M)qPb$?1B=)YFy<n
z=L;OoQZHWgImTHi7wXs3oR2xhKH=>T&9~zmPf&B-G5YmB!r#e-bgET|^S8s4x_Fpn
zFAgi$Q(1}8N0_?2ka7PNDtoPv>6HsvpLl@g(heV}x`(N=cGI!KPR`fO;@SM|_#I8>
zyJH4vC)4mhltxj_45*t)_3}Five-eE-3|`q?%-&LEOKh@#5iK7wDmiQyuOq4*1LE$
zdKY&;?4sQG9AyvZus$t^kDYUwyDFDSD|5LNlgq7tbJ_4wT?7A{htv0P(0><48||Xf
z#tg<5rSY=gHd4=~vU+2x&XK8HdLb-rkj9#M>B@gjm-l-c>G!uX-y@y;1L^#sz3Rld
z%h+&QIsHM)_}F$e>Ee$cocfmqW!F<Pe;w6_t)`3c_BY|}%>^k$oL#~A)+vN_NMVcq
z{I$7!K=qRTuM7U|SLIe@s?+ZGL`*s_BWuVK+FV$O*NZql7Q}H!e%L+D7BN!0+CDWC
zxF)Sr*}C(2r~Kgup2}`YkD=&xG`BXWmwfjEHYwx2YE5-V{~ArBtOexMil&_KHZeDv
z7Y}1-v@n*fS7YhEJ(j_<WAQl}L%kezs?J)#pX%iaJ*uAAgY&T4qAdF!(@ECZe!FsC
z?Q^17n;6ET(;-+ckk`g5fUE+|GShvylOLcSzBq<uM&hNt+uvJ5ICwM!+fkwHT@i|D
zxbhEsgb=Kq%{#|nytQXs^dpG=y8pkc6u@`sj%o;Rw|?<cPKQ4s!rLkB{n>TahiH2*
zCW;>(9OlEdC*B<2<VC{-FOup=$2dcoOPYI!Y5pzqpC1z%`ZK19Kh?w~y147#*Zc52
zN%O7IpZ)=UxNY*KbAm6S(ixRf9*?CozdtSp@u`u#@9X?XEb!q_LmvwAy{R+OoBT{K
z9QJ#Xx=nLtq7DC~7<gzfa7p*Qh1u5p+t!+Ywp&UMY)M*zC4s`*s}(Gm?_xp3-^#a<
zK60Ol8QtESvg@UZbne!e&a~$6JZriOZz~ILTWq!_-rP$0$5ynuXT`ZJ%>{Lh(kD15
zXV^jh6GsY<+EUQgp6VIWmfrGYd#pE+!lzL^eYjQHn}CZR47}?~UX(EDr7M3$YyS3C
zf607Lg6rwu+e@ExDFDkRfm{$L5D*_guh#x}Y45spwC3Oa-gJKH!R|N@D)#hXXj4z}
z7J5-EuSb(p9?G5apz2c(tc16B<v(1y$PMdGZsbb0c~9qqFWNPaL1+E9o_wk9O=4$1
ziVp`Uvp#^{gZ+q+SG0JA8;?r4u~}c!NVCD*8m`Rh<bol`jqAeO-weu!uj9f9X>wM^
zg{rgCpAHK=*gMNjdFIlOMY!M`;HnH)ccuz&4MG0gOb+1caOtksM9NnjMDA|oD`j}`
zhm$A9_AXo~ubHH;v_mPbwB6@Q`+<(kDrMwPai~7eofy8<89xVMcup|&&Pz|OERcJH
zLMboq+2WecV!vtLUKPfjO42?@gmQO(D2b8EDti&gI9;PI1C>=&#!nfezKjy5K5UXV
zfj>0cem1hYzAam%E&kTal|4nS6ifTNXrc=lQ(Uw+lpZJBmI>Q!SUl1|>C4J&&^heb
zFl%a8v8G9B3#Obl<*4NpesP>4U62X4ds%Y6ih%|Lb#2|TCg+kp-PVf-(f;&%vK_62
zU#~t3ce@yH)vjXPT?4tJYzUod!y9>acjnkrVTrx^Z;aXn+kx<QiJgIe)|)Zrq8XX(
z%*l2X=H9o{{@#uThx9!L8d&_<noiHG8KCoEJN;es9s?c9>Af(>K<ID-Cn60ru4K!K
z-d6a<Sdui}N}4Mx0zOz%{KcToTN|t<*b=kGmJ7S=IG1F@wVB%MTH2`p(ZJYhc6@kX
zM~C6U*~@y|7tbxcz4gkLiBoK;{KN*cy2ASDriA}w%KUz2lo#F(DKf#OjS2ZZreLc%
z);eh%rEmR>X_+xROB_vP;y7NlACIN`c<u;q6JLyH_J9ex7A9bScmfD-KkH{N-e$qi
z*%s`&U?D%c1%1=4Xg1WEmS=2e`qYlj&ulqa#*Y0ydJToQHBFtlTFY7b8z<~vNZa+!
z5!<m@^c}vPK3lib`Nno8m&v1sVK?&!@8q@kS?l(@`K!f#Wn&a@aa;i<w@)x`>j|79
zPB1I{1iy7Y!Of!M%*r{=*U01YI~37Hc)N1QF(!69CcQ=>)(Z|})8#O&*BxR(;33K=
zr{dM;!vwZHqCI#ar?wR0S4kh|?dO(99<#3P#Ch6II^NI5x?UFjJ7!X$P8#jh$sW5m
zjpW^F)c&Hq==%)v7Gz=)naRA;nb@T6;CSOK@=EVy@r0d>Pua=jdpmjFVi(Is?xH|=
zJ8*CgPEI-O8I{B5?>TfHnM-y`u5`S)7=*X$-sfrt$R#m5M_lMGx)ko@--bJAo}EU@
zom6&5rBc=@Ri5!w?MqU*&>@wqhiRnMUZ?!)73{v5tjzW0Wc^si=L^eJQM!ybr<B?6
zwv-(&lzn(`3CHU$<&x`i&SWJ~dY?M#$|P|)Cz;hVR#3cm1@{}KaH>rTs}8T=RfiRX
zA4$S7E{VE%>K1sVtmB6B>7)Eo_dl1=RCqgQV;pV=;|RaCkaK<G$(bW<l6Ixu=jO_9
zue{-v^SC{G9_{1j(V_V~eU7^Agtsr{dF@<1T6qBK2gq0;uc|VhG8eF5r}{rC#_+vG
zEWVm$m!-#o@U~oF3~gK&P^#ks#{3h*!^;aOR*p^skLjccZ*7IQD~hL4Zd(+&@@(Xv
z3gN`GKpuNaM<l$hD_p8N!js<jJjA6+d-Ee0hH1|yeUf2>^8BRx|2sGoGwD~(*OczX
zDVVmqwD&6Ur?GJMk@THNOVTRAk3paOSS!!XG*j`%;)f5f@Zr6+7b~?h{kOcQ^20rq
z8STZZ7%#s5CoJ^vrrEzf9QXERqkMD^Yx}dhqCer%toTp%Rv(<Za!TAObxeCL&2aD5
z`Eokmm#V_s2UC6Nn&HcUmA=fH>VwaH;nz8DOgucuF*?)B&56SHcC>q8%kGA@q%N`H
zR|^C6tPL!yX5f*|@QvGB6P0es%XmvdV)ZyCpM7}?rX4Zo+yXPI%rIkWJu|`%nNs?Z
z30?EeSaZ&d7rADvTW3a`j~R_Rn~^)goc~O%c&K}?(;I0*8;ENY-u_uA9m@&_h99w|
zLwormH@ZtJ;6Y_C?LOt3Ow-TUWr#N~q~C0N$Cag#uC!C$U57sItgu%<r{*b(O5S|4
z^TJVF^Y+$0jNa(Sh8Y2D>+a8lTK-h<^5ftC;nOu`;49NFR-E~*YM$&q?@6<Mp3+Zx
zh>!OmLK+#{znz&Pj@vKZ4R7(JrVkv5y)941BWGrpb>+?*PZEW<WyC8V6qmetY5>Ov
z`7uX$Ym{&Cjj(NUO=p@ma%NvM7aFv5p;CKiJx_ZQH|yRt+J*hn<NSTX1tz&M+0~8e
z)7+R~;X-d=)w!;&L?*cqBwcshU!FXXw`l3BK)kX7nW)|U&(gHDarIH3u72k&^u5lx
z@bh0z%2#q=bt8L*cC%yBE8)@td*$LfaBGP(zjoHHQ+RtIN18hOFscm*rDESu!m@&N
zzV{<K*O%Jj%A=&;b9M`5gINgQp9j)c8u&`R{5e}%T=H07h9C9jgE#?;@5)DbqM7ri
z4OQ;iFtwX4ukGx~7DqBnKSQbE&MZ6Sz}$Pn;{`SZ>UZAvoV7HE7F=2=FUu@*zV9%l
zaHlETO{dW1?PPLxPhnJ7OP)GdvHXEKr!HBOS!~2$rPob6(6q0%@>OW=?5nx>yY#oh
z<4!XS>O;0s$Cdt%HV&lsH}Yzg^j0b2Y@O|JcCchyZF9QoIxV_j#-+CAET3XQ@Gc9^
z)UzaVy)~Bat+_6|ZSmNe^B1id6krffD-71m>me*2I@CboBx$Pd>EGH~F?o(9?!&Fb
zQCN}mLNl(dHB|$2zHDgV!Yu<e_Go5I*7MYJ>My)Kq4Vp_&$ev-q*+_M_Eh0*kn|2M
zCfHFr(Uwk=ZApA&!?rp$9Ct7!p_D0;%A3-0k%@9TP2~GF;n$;+DW5%_mj$Cae)MnU
z&HT;Mgwag%7)#wX<EY~}o}pRe@q9X-`F$ra<;w(&0VZ_x6(4=al;ee_TzzHA#+zn5
zPPQUG)r#)Ztnj{JLuVH|BK3SvuMx+uxwqZwi9}Qu$6rHv96N=-PaSm@an$ZVOTX9c
zB&BXAx@0?%_qP*MZ3i9a<&m9{$KS%+EzQ+E*Q)?S-va(}kx#W&0hxvqw4Z&Pa)!eM
z-#pCnE61p|?HDCnk8|pXauPQd(&Nlw4!q5!PgoB3ALVe}Du?80%7<u>D@}Ah6I&l4
z;$0!0r3)GTeLv@&cCz+yHbvT%UNdYZP}tV9Q7XCTw-E7p8)g;KX_lA9PiMAKU3h!_
zlrkh5W-{zY27i9b;I{*r<VZ_3ws;4o^|F;|wUgE7cT%(VE|zM)dq;T7z#L}jdCU~v
z<_T|q?ww2ch+JHva!C{BKCsGV{IFb{cI8lG=q^@yWHVv@4rOztaqxC3v!<movYEQz
zLpE#Iu!$c>Hn6EEh4q(LFf@Aw6K^Dwe|$N$gtt2eEW@hGGCG}D%5j&aY<j+g0eMS^
zj8=!;M0K~DFClmRQh8jGaQG{krdN`cg(lv3&<YM;OlIG#7227sU}-@T$EHOSHfjMi
zZYw8z%seIzoyW0%=5jWFuKKrP`O-lC*z2*3-l08hL@a&ptEYXMG7#I%VMnIAu;po8
zEBtNNNO=;qlw03!9&sJ!QTx~e?lz1jZ=3wDp$k|QyMT{GGoeBZE33!S%r}<6TMM`{
zRUNA>7SOxZ0yb7&fZM0JTvvbU?1Absi%}<x_EkS_PviL}X^gi-ar~sXMR_eMuJ+~Q
zBX78_*|UlI!IVjJ>7WOv(>&N#%agqi#RuC6bB8FGNnXwPK>0Q0QOR_5XM$$0HJTY8
zT+{w*ygx>9!L5p=Yntar=1V`4g|{}s+g~cndt>9nl|gReoSm6E*@HIMoTysfnUh}9
z@UE2>(@yxf)`!wTzQnZkqpFt=E&IqXvqo8T6+Kxfyj@->&)^AXcFuBU$$NV?ONa5H
zsf#i%eWf$>rBl9g0>;RHr*rGM<qlkse<ALRJ^xBiV%JH#C7o|C=9sgiss%YYR+Q6y
z#;K}-x34T&b<UD*7M9||EIB^kl9?wh_*~XP`#;^o+M8jcd*;16CiIOrVbo0%7U!8!
zy2y<C!rPlEW(*o*#w78EY2(Z>X>3j8Faxij3)6+S4aFx1?w4kyqIkeVw)7I-o?YXP
zO(hT99E7z69yFWm%Bg>aTkD(|G+&uJ!m}PzU1<>RM*IckyW6^uo2-4J@b*VPZ|-*R
zrcaI!eD-0u=G#ANh!^lseu41zpzwR{I1lE@OL9QFzy&s*Y;<>LXIteOJk*Zry9?1>
zof+TOiOoG6c>CR+<$8XJ;?he?cY5%>3#OWpM=tW@?lf=lD?U`7ApT>dFMV!Fv(?cR
zR|k6xOO*-q%!x8TyNGvnA-%q`3%(c`Hq@RU(w#lrqg~^luFQ4P*+IOsRfHSeM!4{+
z@OktEXXREnQP$Ck`8yn?zjoxfx#s_$9jTw}Kt+EC8sLC+X9pJ5apqbp7u*{-G0nk&
zmaXj>mu}1V>$aF`CS37U=a>_&B$>L>FE^NF@gb!C6~d64$_#0yERLN)G#lv0wB^!-
znR?UK+?M*{Y<B$`!tf2!f=e&u(!rmiZ@zfz8d@M7TgoAGYSgl(XO#Ao`)w$9Odgt!
zwwP&W>9ok6U%%P2NxJnf$K;!tVas)Y-P18(dD?>QTg<tk{p9^fb7e4?ax24>z-N>B
zS~QulnsW`NW|#{bA8*qR^P45VBpNy2*GSWbMxF|@7WXove+4sag%|UMw@;@V<Ri3Z
zm~>9(wOcin?zW@8reTs%w@Vu)Xm=STylpw(j6>p8>zkT$e2laQ;#S>k#a)M5tE<df
z*ReHOcdewClRwGJz^c^-8fqSI-Pgd#v)YeDTH#pNip)!vWZPS+hgf*)A#OQG8U`;*
z>ib)>tbwp!T7@NfHiX64C@0B4Wu09E6AfIJp83dYgK|I(Y}s$1*%>S0f(3V7CJ-@j
zBF+0uV(i69v<RER*Ka0Rl{KY_@b=#S@K)HD`g%N}>Emh8X|y<wzxlV@-%MCAni&pb
z3Ho=O{50cn5Z)F&8LwA*0`)&nU_zt`Z;qQtYig?AR#Qxbw`X-OPWi`7Iue~3G^3xd
zXp8T1Tjf&OkWgYx`-xWcT{V$g;s5ac^=OVfkT*$q+w+X0I+47Tv9%178_Vdsb~*WB
z$xJ?_`+mn%<-g>S6R?#zC291Zkj|&m=@c)`QifCxWxwXoJ3v~Z#(S9bA)i~l4^SL@
zjM(gBj5o{U>9{;TAIo7`Q7&(Sawwy@GIMkeLxjCw>*ca=WG*l6?xO#}T~wH{i^Z#V
zszWcEs{MD+e!w;+m)lBj;cZBnR3@L>!rbgF{8nKrb*FD5Z00unPj6%E^K`;bY$whp
zgHk0K)O?|jwKAod&gApfOm0-%LI0jRxaFV4*TK8E{%IGd5YEcG+q6p#s~hC-Eqj;F
zwEB6Rb1+-Bi}-T8bf4eJ=FnaA9U{EVNGBs9onxtK?0UF`;>1nzqix0|bhFN38#!*Z
zfso@V>N{P*-m^)Bv|P^WA&FSuTc$42W&BxLy|UtyON>kD{%i?_+n2E0YYBr!EYa+#
ze!S47^7t%g_uypSUP>lv;tDqYxq^|slCbKtOuH%Z!oS7xld_kb$Iju6de=J6o=r9B
zkUT5R<-G9rav$l5rD3XfCzi6Au{8FKrQoVKVV^lTm6}7m@OJIQx!4JBYyTp=tvnA4
z;q6r6?e(<zM0Z<2<J1MXhp3yp`~qSIMpNy3G&vh$cu_f;J0}-#P<UHs^L*)F^gKS!
zCn{|osZVCf_cT*}HFcPEQb+17^-`{#Mw_w9kdF!@DI^d>pf47}+dC)a{rlCEs3z`w
zeWn?wlm{-CJ(%4f0RKkPzX<oN2P%W+oI8K|yK65l{0njC?M8Vo+j*0{%8wVq$>C>}
z4Le(WZ)ZR1i2uF5UR@|<{TXQLL%$v_w9vd++trEDcbqUQ=L}A6d|Ki`?spG@+bK6;
zvF2;(g}Rm3eZ8}CbDZ3<Kjg;1wvJ@n)O}UEtE}htY!vqy^}>N}e|l5m>`M*dZOT3$
zY%TS5;);4qb-+WsZE87tc0VwZF~gGk%guGxFsGoR1?P=cw5@7I!z-4|+Gxq&nrSzi
z2xog*s)NjuZ^agDX=zTCVl(*z%=nRIDh-BqZ&4;xkPmSC!6|sUPvI}mDb$LbLQ3Wo
zZtK3<p|%+oVamd}ZAPbV<}|BgMd^RE_j+r~f6_MFhS{<-+=l<^YR;`H&qW1C?6<fu
zsn7)v?UB>gIO91_9G7-=x1`Yt2z8_E8O_4uT_~I6MF(*NNgMU^r`R+0nmG24KDf5>
z=ehJ~ubbEtQcgPIWDoL$uhu)2$-dQt8G3E=&D{B>{pVEeXg!`fQ#Z_+p$DXw)@xZI
zOmnSvZT*)UnJ10czC%XlE_TGpUmEE~;ygTc9#ih~nm*zn{&J&(y(1=*<O><%z%^mE
zLHSvSbL}Z5E$lIATP=r3b9=;rJ9_?p($#wJQLe&9^{geh(zu%o_RrP7HeT1k7biL#
zcj92WyqteIa$nfh$w&T??&98S=q#mqJJZpgYD1iuRMSZr%TClcI<Ts#{_jn;Bww~=
z)o*s{1F>U}@Yc{=8G%QG`Oq>1=RZQ2s(nf20{Nyj-)@oSZN7^=f1WaOtgkI|Ygw`E
zxh1J{f|%b@yG{oi#;vuc)k-VE<Sl8KU{25#6ISP#(RI4`3R7zqX4<et*U~X@xeveE
zGiieZDY{mEs$@xv6$W`>40QR=oSwSIo~<ya*FRI}p&8EG(^UKU$rSyV#Hja^sr3Iy
zI_tP9%=i5}yW814b2ciVzUi(*cb9;Ph=?snNQnw|7k18=?Tp#o-P74U`*(f5zdz1P
zhfX+f&T~EYbzh;qMES`9*enhtYke?YleO;@-c}Lbo)O;8`0P*gH-Anx37~m)AU+R*
z_-l&hwDRG^#stX=CBAfR5RI>kkNy+{{ep<EDV?tFe2X^uusqR+ePw)T?B_#PyAaM4
zhbU)Ox~CqY*lJ$24-KY{UjIupvra!7%AbD-M{R@I@@p`)3pE#;1Suaim^<bnTyYCw
z)tVsQc?EL0gJ#lO24!No=|1O9<O@&M{~O8-_b@(;597<hFm<Gb;w!GY_t^l3h%fG)
z<3pn;Pi1j<a^r#*`#imQ8s*Ep<L>f57^pSCKu8M%LHFDk+ue<-uCB_=bm2rhXRO;e
zvnkp|=ei5oCI;=3-5Isi1FOd#9Np^4vHv`o(bJ2=$NUH!;g4myA7!igl5yXgQ&~Q^
z8UhIyH~vX`qTxdW@o%Gi$QFSNJ{mxU8}6J~>7q<97g{cNRlaX5{f@;l)m=FlHwG$0
zRJ)O-S&ZD6CC)CJ+`Md79?ZtB^=x&B%wx-r`J@k6K-W);IOVyRw-L%5oVkKRALZ;@
zs>`PLG9s6+AjzSG4qHq38BoH~3d?9YVJVNAETyRR5^mmGjQO$>?$42jHEkP~Zd<s$
zb32QxttPJGYF<oT$%ot$ChOnWSX+HL?Mv9zzl6wfE3sL(nnl9kl16LkVZ4qDL)UR-
z?|KG&)$Y`C6BBQ2V$BPAUTbcqh1+HV7jEW6sV$6g-@=TuTa`bsogAMnJa4;&h(=pT
z5*A1Q+{|LHEhHS>LKWL>bQ`msf}h&`KHG}@#x>+sS;@4=CE{Jw^=`Y8<YTKjCA>}Q
ztgOYh%jk4{0mj`I(y;n`E``ifr_UUgwVX|qm2xQV%_P8LCeOdlV8G59)C`(ItNX<S
zXBX4%auI{Hd;Q&{n6u)D1Fy+<BOP$mKII8(wjKIW^Kn!Gbs836n4OQIYc4lOsJmI*
zFa_#H4a?+odL~oeXl|XI&7NKbbdM>Z*N{B=hpXeX^kg1br}L_<de7fxkYt(7(12`e
zT*;<xb<M^VC)4+PHvXTo3168*r{_6*5}wW&mP=HbTzn0A?EjWWh1h)Btk%a(b2;~W
zE{%3gCVIEJxIX64D_}Ce7)@rUjr!O$-v%jn;^v)+6i=SWQ`hlK(jIG4=Mi{F^E>dQ
zc7A)s>k4m&Dogy=TXA&wsjjT~nsdhu;-}`_?ZR7o;q6EHAEpUs9W15ysW6a(4F^(J
z`1tpzB&yxgjCo`jrANy9pQOF(v7sC}IGDbgg$G?1N1B(w%<nNgH^gvkOAITgE2q3e
zJnO8rQ#>k7W@&wX%fZ}kDXv++=8hT2c>96;Tq6u=5{rxQ)~kLDaq_~p6yCasr|qse
z<K)O_GRn%+NI1b$!m-bbrpmY&E*6J!`;hiHZvu%q9H>mwK=rK!;;~WOTm2Ae`9k<{
zS^4NogDK1nrlGt9<;C^cjR~e;N)Von1E~@3$Eo@OIBwVd`42y3Q}}Y>t1tET`%+DK
z+eUbsG{%=z%YDgm@MX$7KQcz@{k%?dM(aR6*AC%$p7z8$B5?dIlGSA*cpMf^!>2ks
zr$!L^F^bBm(b#>8rqzQOaxeEMRdfH1eR15DkEqk!1T0_3Bk3AXtJ>1;WGZW;m$C^o
zBWKi!=39ArE~K*=r8C=1SkPo?48MGg;rhE+g11P+B#a%|WgxF~Kk8aAfcL_^ZI9zf
z7!^nNgR%U$E#IbeB;&>Zwa=B-byO4;$40ShdMq!V3FA}c0afncC9_yokL%C-jQ#{g
z#j?gcTKV%4>}(f_lW8n&o1#gNir{rO&79I}d+inXD8I2)Ogy!A#nW_Ff;d<0dppOg
zzb=**<0AM)-}iQ_{Ybv;t^N-`{5l5ESu@(o>OoAEZfH_V>2CEk{nD>L>#Ovq`~qo*
zu0-Lloo46P;VjlJ+`LsJ@#0iZ%L_W~T?|_WC>wE??tiNiDcdZSyZcAdd5dy99OS)h
zsn^AUDCH1^D@QAsE``B#UnY#%6;6hS@YXC;xj7-YSO;<GFMnpP)z?%ZfPxaecE$!_
zwpuy9^3X0E6ij$>2p=wmsK+muBNe3oc@#vchT7FOR|cVQ>7!mFYV)GkJ|B7w_Q9l{
z5ARQUbM1pS;jMk>HrI!RqXI~lcjcX4pVRkhCVd@1qZXP`pZR0Z=ierOOu`K1DxK5k
zo)=8hK5p`lx$$0@<r3vh;}maxeI^f#sq+0F1h8tdALH_Ua0>C^(|2#I{e3VM-WIur
zklQPWehz`^S`Eb3O*u`c<O}u<CH-kIN3DXHXcMfxo3MOH5a%BR>iZT%Ns8{F7lZLx
zDetgbAkGcM13xg3o#)174|guV^kn+f0L*@Bo*WuRKd(^AwF$+?N%x3-{#-N(pp)>n
zxwK7={oFCiG*Cu7@QBJ@_zQ3Q3U5!u8Tin|fRpfcpzwC$5Le}Cxv;vNGg%Ftxpv%{
z)TjFW4OgDzy3zHPyLv}Gc(&Jrl0zO0k!Gq*u_uG3cyhB){PIRG%G~p2>r!uu?)uTP
zX9(m4W4<JS8}IyaiuA|+g**MGx$|<q3n9YtnoC?+am-bn(y{mmZ=ahfWBKX;;x-PY
z{om@+(%jS_NBIN7+ll2Tvwz`a7WFB_VwQ4(f@g4U;0)%qoukbEIkcZO2iHFq<09W^
zg`%b61eeHju!QTKmnc7SDILcz<<HC|7|Ji9#r?%>-msV&UW<sRzlfBV3#HkeOW))3
z*tlaJBOc7*uZy#&*<~*8E0=J)T?vnVF5}4j74qw^=1<$z*!{hR8!yyl*K{48fy(_4
zT8qI;v-h=4w0*t_3*}3+thSlRoXuPl-cB{%f>E2zST>VScHl<Vx^Lw1-VLPBQC7x}
z%~Wo?j#*Pyu`zQMonn_!G<PY@U$5b9HTBS@mQb<QbP_KV@u~JoF3c#Vy!Q-_{Fp(G
zX4)Ao)sYuCPa61nH0rD#_*--M^Y_{8Z90pD@R`&!o=Nl1>Sh<-?g*U0&y~egt6NO+
z#v*p~F6L<KV!o~|qG>__n>uP9ZLJQv{ka4QJEI~C@M}<jZEy~DwQ~u!&EcX;7H8Zu
zDQA?)i!$obex1d6>4Gmj&&RxaK4ZJ)aJ_-%&N7pkW|dCYZ<#FLl|c^+^|AY9WBtFM
zY1zr@k;|s*U1iMGQcrc?R7wmZ88uivpDiadZ(};k%rkZ8&f-c$9`9D=5^Iu6pzwCx
z&dChXF4(S%FxN!A=T)Rdt)ZOped_NxG=aIJl{0Iv9@Gh=a66K!enjQM{hq`;Yxypv
z-+f(iFu_j3&x^|YeK(N#U6jc>dJsJe#2xRCrqNAdoA|vkF4FQ!qj^O8xNG$X!d`hN
zt`4L8HD%i!ly}1_5!X~{PEHSzepy-I)dtdEc)KP&f%&&$m@HrZX5sB2;ca^5cp`<j
z93DiyZ-X=+s*759`#DGR^H6<`Uk0LE8%{;an$3vebiEk-gtu+(h||lHA5eOap`XI|
z_EX+`%@=$0`_=cO*kc{Vab*jvIv0$?>p&Lok%puo5KAlFadqz<ud|?mw;yNZ&!{j`
zv#ZXz7Q)DOUvw`&=Fh}2ewcRkCA_h4HQ1L)nrEx@^keuVUpij%rM$R7pSs@KLwOUX
zKflvVxZ75D>neV{sN+Y^FTT<S`LQs}kLvn0ZM*o-1JX*#6Bw~r9-{rBH2Dxl#g|cR
zOi^ynb$KU+w|yJ*=U}SzKU?DnRNh?=;q3)=$V?Is+oQUkw=y>(zDKfYRV1HAMp9=&
zq&On^O60p}cwIBnpcvl#jG>&kvFOFxk2X`ksCeY_nsxo9RXXP#&zy&G(t*b@_pexP
z6h*VXs&v{rrE8VW#YNhi*R`b~mDbc<c)NIlc>iF{mi=Oxpxx|5`8?~5ji8ji_l>Gc
zcdDJ{!T4|<*hNs)Ih-o*!x*nw$|*6Fl{(MUqyuleM{{~W0uI9y7~UkFANjGAukFj&
zcHRWd@F2z4n+`F;i4tFiO%o4zC4h$FAaWMy>~9gq@vG6qy_SFTRWzgbDU)GqC^w`<
zS~@I<Pr|8=y3_3YFN(a=u?+k!U$f51_esj8t}+A{`8<-&39mJazd7u}n6)nKTp7;K
zaXuv0koHZOzxz~}eDE&ZsOw7Y4X%u?>Za}|7cS)aF{qq$a>`fuZ>03<NkPoc7f0Jr
zJ5i@#W@>NR#Xp#_rGsgvj~jjjD9>BDzajplzY}L_<jdUIKHPfb%~5l2M*bz7Jm@VA
zun$dzi@y$0hQoC|?}8vE>3P)K6d><S06XsZV?5B0paYsy&l!}z=7v{UH@uDA*!#O1
zi#I#5c#;$S#yT-&yc6|GoTz%w88>mOwa0i<F+lkc-@N(aC5)FAX;<xF(tZu%mSrHR
zT?1*67s#6AKwfnUQa+%1?<|9fG!N!?-A#uK63!k8#7=sAlRR&XmU)o&#g%u<Tqrlt
znZ6TUl&|1IRa<9H<~wm|jxVvp{irEE`>?Y(>05!c65if8<cjSsS0>GIBTW9C5yK6%
zUE;*E%Wm?nda6@O``9c4+a@~_R?40;m+V+^!4i&Gs*~7;6H)fGY~W1Qwa#pO;7rGS
zXErCf&?D7_(0ML|FLfbuhh|;j?cO3!%`@@{32%D~Z~Nc!=G;>c?X%oT`qxwYUA_Ni
z`*HG=KOKa(x1PI`CcHHgSN&&-3zc$Rwc~YV<pX`qx-ZVlk7T*>N`4g|_n}XJW!%Rw
z`;l~imJ@j^z3Hh-X^gC!&Y78+EC`-V!Q#nuygXSmd=7ut&1KiPT$WwPqu0HB#{XVG
zv*iU;Q;tEi1yiVfR(r*|g~awQ#Bab<EFGuwdf7~}{#(fYhYOi)JeM~E=aABAHn9_D
zksdIch;DQFx8r<eWh|!W!X-?dy@DD3t0`r^ns&Ks__}y4ZQGWRw6lc0PAfUQYb_1F
zE@Ru=Wh@9?Mnl(Sv@lzSjrkhRjbG1@Wt%wJVG|DyY{YE$1{xh%NppSP-UH@I_cxc%
zz6-FbwuEcWOQ=~ihsGV}&}Ymvdc;p-($MJ~xl_c(k?K@0F6NHs48|RvLGhNEymFn#
zxsbVBDL<DP2j<YwdXB#C*)*;-ivxBumCHOsJx?<@acl-|E;Bf~P<`y5iwId&gv-O}
z+_*HIz5|Oesa(KsZSn|hE+1^2TpDL&ljxYmhLu@Z2jnnop)!R(W~z^666w~N<oC!_
z_Ejb>*Cw$cAdRP6)3`r1jZ2@@Ez>fM39Hgr(m9=)y(aN~T?X6iCdeD5E~`iC;`pMB
zy0~<X#%3rVd=yQSQb_HoUXRA|Ej*C_b&)dHC#VD3EQM*0QZT8K%IQwZbu~%lS7q+?
z9+ys^rjsc&&QPXh3b}4$sSrAWpfM8|FTDL<zpI1rHtawu?X|OV5eGcFb0VXE45t02
z1UxUtac)^0BO?Z|S6K$HYKv<-8O5E&;u)_+<8)OshVZtVp+7yICs6aax=rL|%MkB*
z<l->a?$(YkAd%EN12EU0dngMh`9wS~7REE-pE%7ukzDE#L!&=pIQvGkSVA0$^1g2I
z>#w<7*nW2q4?hp$z3}#dnK1WgJVD}HL(BB1=|Ju7mPe~oOT3$}@@F#x_->=!X-SY~
zwqOQS3B@Elj8;>%Q%VS9`L$5$X@|8>J6pfHfixCBd|h)u|M8k5x(Cp~(~oF-KdiMw
z`e&0L4|e-8;!OYxv<G68A1Q5o8NSt<Po};!7vAPcL)qhjFLwKU`R5NGo|W~cd|5B<
zxO(zrqwdxXg~OYCNf__LHu1|lg8XPNysd2Hr}JM}Jyp8Q2H~1@wA(ESA;><M$J;|V
zv^#`_@8#2!-*KmAts0}F_)YWVJn_n(8pYCB^Q5^^Johx?PuFbw{?ACBjE&UKA%a)R
z3hd+=M%%z}c59Dle>#GDMw(UY#i~m(9#?6I{;ixqOsMw2!SOs0whw(TJ*<`Z1?`io
zPZQr>B^nd?HOu~#=2SD|>%Suy>=B{t=x}A_N3!!+B=s&wGSxJaYHuSrF(!fqnGt-b
z5W$qY;cSnU?sT3sGt%T#4GQFJHR;ZV>g(F0?{$X!Q~DXDv<tv)wm(@*v;)`P?M0R^
zRoi$IGt~n>ColDId9&Nihj|M;Sg<&NIBCQC9uA`Ai%`}dQ1+4V_QuO-e(a87$L0{O
z=LIpx%^%;=e%!68yoBZAL5%`fS0#X2cl|j%#+Ro#9_r?CB=5Nmx4u|1Ip2;EX4YK1
zVb11a7ap!~qRCxnEWSA_&(no3w)X5eXv5N9ZKz(#k=ylL`L3C(TuvZYgzbIw==(BI
zK6mlB;&e+|1oNAPyeId9<R254p}WH7$Nub_>&FRckym!|#b%lhQJS}GYI`$mpBGK^
zdg>(4%abS{t{Hs!c}e%SbzYj=Jn5#nFYmjjG*uq-%k*GRxH}t{xlv7cTVt#%cMdu$
z+s>Ji_l|t}%aNU_PSg_KuG91Qw%VBq+IvR5@uIWtFAY9<Q=lv+(^>uu7_0q;xRx;%
zfvoBn$ON-MF8TyAvwaZ%ZdAXlo<mDBJrB+BMZX0}vl7U+r-3v~7jAX+AmX$u)8br+
zY3fWVPe)2yI;ii-9+zAv-j>&#+QFSC7ky~5*&T~7ZWuLoBYnRshXPz_+{1<0U0vu>
z--Xgk9a;U;j)E{JGRHYlWuX%lT^(8PX|1elE54k!B=C?W*$r%n)W5&xwj)8p*-L%x
z`Mae(;qB}(G1%ie$e!Q{_I!9~FW%gNc0(O7IqX37Cr<3NbSCzo2P+<VuuwB~k&hjF
zN^G#ovLP|tmiuRHm3L*wkrj6A-sHrQD~?=j>_Elp4y0FgQ1+d6^i89Pe;p-$FNUM?
zWF1=^!};B#c)NBqiATn8HbA*-%8DI*Aw#|P*(_W+nF_+&+B=jle=>(DLvnd^UY^a{
z`2@EppxYAZTCWsvd9-rwFHDhFypS7R3%O#XY<p$vj%-rI)wV_QkQDLCxrk*2MIgL=
zJz*B-I?TeP)oi}MokQ8k#Y`_;!aecUTijQ3p#5qNI;b~x!Wur>E+=>PGUb3Rqw?xy
zSnCm(xr`V7%lP23j3%bydp|3KaAUr9M#@n#n}YM+De69-%K3nqT<A5IKwn|)hPjM%
zn9Ja1b1=U<n`tfPXa_upIisdgDQX%UgQnyDw1_h)#XMR%m3Bo%(zO(`$Z|GI{pYAR
zY%Vv-&ZYFhIb7&62mji$$@+IDtNO?X^lSzz-p}B|@fnmB-sUb)F2%kgYN-z({N8k)
z+@Hqq66G4Fi8Id6C0%<_*ReT-Ow88Tl*R4tSu8xA#j0)MhlRIK=1(H2VJ73NXVP(u
zI$oEg(Yy8pTCYoEr|`DT>xs;$qVBH+>WX`v#$C@zTpf|YXyI*2l)BR{OknU^^;^$X
z9^(D6R9uom*XAkA?~zO|yHTV!Ov7}FGSuxx5a6A{;eS%FDy81G-_=oPq&aw?@^0IY
zr0{z(|0q+*MSbm~Qsqw+E*~x&NmpgwWp<rN>tEG}s*lxeKb*}i)omgl$!=lm)=LQ-
zPmCi?KEhXf`ctBPTEAMFo#drId{_LR&f?qB!E9Dm+^`j~^gN{QPGzO8tt8)Ny#X}o
znV?+RIDXuUXP5Lpwf{}v$sl<&s%uWFs!S#EnDu8yGGSdbXPaxj9V%}5aepF&w~LO%
zYS!(~)6&9TVa4bS%~Rf*t)!t1n-a%{JF$36n`*vHXYTeWRE~g0lDIbErsYdz_Xuw@
z&dV?88m5fl5b`zi%+=oOiO$IW>jUYc`}ge+0ko2)xTkLblUHj$rn9<&wI6f(`%!7V
zACZ6h@zK|h0qwlRNqce4&YN;8gtsQX>=VYGb@Anw#Xh_m?u-38AF8eNqQXy48g%xg
z<hX~jo;^w3=u6+xKFrwQi;@1VE5h95F4BY~`cWdhz4|>2<GS);w3J>;cx!4U?beMz
znlB8(sbx5WKZbK!I@LIF!#Dnk<-7Dl7nObYp?y4igtrys2X#ygAx^w^<4Ey{nz>3f
z3sYZpIJHMaFn(VI%|f+ftP{)8<#GJeQTfi=r#9>x$qeOY{h{2x&cfUDJ>rqSN2|jy
zl;;bDpKBs%dqkekZSuD0?@e-+wrsI-cdtZpY+NMHy(1~Bhl}>m3>Bsi(mX3}yjWWA
zBKe^#bHp#m4>iFrkbw6=(wT>f*A1YmIMbI_0r&}T%gpiTzr}w4rz>rd;K6FmD{F<X
z-mTo!Q|KmdueeHMcNW(7!NSF#DQAP&`$+!gr;+3y2*+=w{577!-+~Zc>$$)EDGp+{
zcIaJn*E#1&HAjC6g}2w{`IGpM2XB{1TO$tErKgQ%DNB4iSTgdWh4kCzRGMwBo-rrl
z&p2~rmlLyZn=!ej1!P;$;)gZE-#Ku2w~I1tT=_6Y+NI$>bX@Jj;YZp@1Upk<y)z5*
zUC3zXrdip*j=dfn_+47M-O9AkEOjW-m)+t!22J(B{ggLm_r!B%c~N$YH>(=@Fh5ND
z)l@I?jXapx&zp>!o-AMP$u&1m4*lst%ZcJjKN`4_=*A>#H|~#c#eJ{~XH%Rx^U;wW
zMUL7(JCN_;K)j=q_-SXnI=k>>h=D36yqGx9o9PigT(~N|?m*=>{Tj%-KGI7|yS%V-
zAl2Lh8K*t*8tsYSH4kRrQh)L^%V+irz@YcZ9(Ny(_wc5&tru5x4|wbABL2%6OX2Mj
z3kNLQ+B5Q?BbO&QvUiuGdX}8%8tcN%D{izB-fkT2O1yCPP&pTt)N{d8kNR^RN&RL=
z58-XQ;ZFE>u_L*xE%nRU;C$7JQdceLxW`hOej5Th+B4~vBfXQId17L(o&b9?kK6F+
zrVR^g+Ol%AEn&jjm1P|mRL6nHzZ}@H){*@qoq4p}gQ3Sf)Gh6%>{~mk{%TF<GZv)&
zv?B1VEg2Q<IJnx5)w7(4{KtVS4IEfe-ht9z?3sTfTwZDAe5{I~#vGkhrJ@;pCW1YG
zhGDfpXN$79)Q(Q$<blljI*>!F2kCA-7_&vmv^5{W>rctttfX#~p2}JE8NsM8Bbfg#
zMSjbXq&-e$$>vdP>#k0h%j)m=J_+M~8EhPs!R+JNZ0=Qr@#^V(xl+tx@zjMMH3MFm
zN!u2)>G*99<H8mb-eU<5UoWQc?FyzpFX3&EIdr@?m)4F;d9h_7Ht!ZN(|-YL8>zF-
zVjjI`6fmz=0Xusti{V8cPp8T=D!ffSSU~B>DQrzvhM+oWe&1iHzLM!Y=%Jip-2qC*
zD>LHpG<74X(<W#d-VZgCXBP6tv5>2s3OT0i!L!2K!w08xE47%zakD5dKK-w{bNPN}
zHV>-KCc5Hm8rGc6ZTmv@WEV2}@eJ|r#hme)O0Bk2>Euz!zJB737ps%UM;&#IlwmB-
z>+4qHkh5}maxI7I$0ln>G+B6&#m**Kbf}nxM_3lk-(=!Zc@mdir~|ergT2DrF3Og8
zRcQiSG+(9`s6%}GL?-@{#<RI;4BxL`m0g$FbRvQ8#?$cmII0XA$8Wck^LkWss=NIA
zvr_obEQJwGlBs`6o|l`6tm&-2w$jPGRgTcz2Sb_tullB{CyEbO-d$#rvb{#gv!<W<
zi)2PUOlI+Jb;mqNA>eHaeKM7SuWaLa^VGR|GKmpGm1Wo>i3`f{z1Cj4+LH<VCysYU
zbR3OG^=JH%7!KYF;aMwXlIjf3)ZEk7QC`23F?><CiDftGSGH+B{g^=I3(82()SgcI
z<5=PCU!Dp0K98rXc+uZp2`|NA_Inh^zMM#=tyS)-=9SHHF`QF2oI#qHDZ69osu^}+
zss0SB8_T^pu`KT+9`nEcd{2*K@%31yxW`a3J(4JSBbHW^rqVxvwCVwr)p?ScqBC7P
zslmdsxurwd9V?IH(IE98N*h%{XHAtra{mcnsdOA&C;8)?>(8)TzSLOd%gt)uWUlmN
z_<MI;G+R7v?n&2)UVOIFeoT9yT24M}_w*rQr#BOFeONu+3)errN%QigT{Z2VN_#N-
zum?wmdNO&df$?#=*I)EtzOA>s*FI#V=yM3u?UF;4ffz=aQ_94s9D<`VtXB6?KKsi6
zR)|-;@=U({^3q?$`q8De`Vf?Z`?qkjioB*X<g1vUDlF_Bgl#~ex>W=5w$x58F^FeP
zl-1Ws+453tR!fTDqg52w#jl53M{q~;=iT+;d>0-b4b}V`9!=dXAsjOb#tLb5nuPMi
zG?dA=m8bP3l)YUd7<wl_eFgyp$3@^*7>>6d-|9rF7coGcr2*Uyh~V!J;S`pNVC6LB
z5NYNcC$8RYlKh_s1DNq|05x(A7}Yj#{IL`BeVu4%=g6yGj>^4s;`?_e^;d|SOqBL?
zs0;J0T*(mL9(Q%arn5NLUwv`ypgccmmRbp~Zu)z&Yr8MA@B6Yb#t(Phc}&B7vHGF>
z`f0u_|NVbEJa=Ykp2_@XV1mwhLvI%rR&$Xy(wNp8jTxL_!O_KLc>Jr?#pAxbE!7XZ
z(WZ>9+=n}j%{ckWOnCwpq}{Y&-ZyK?JaS;-P8W=Kx>D+Q@u~Jsc&zea&`tTh0)=gp
zo$)E}!lkxu+!EgUiQDXAqU?n;;y9zc=~C90McUOG#d+b-+ml{vy?FD(LpxamsTbTR
zR;I(M@t*WA=fR6c?o_DjPUS_OoEF{|3vY*y)7<~Tz`7xBxC(DO3U7N0Z;uFXFFbdo
zLN`b0X&sp0;-It13BNpN#%DX@d(@fizr9Er<jtToZ_1|mF}IUH;co+Yb~AwZzoetr
zjJsNT`{8^1X>vunge!h5|JNVuU$ie4-i{U@(e1c5Df_(`Z|y~76)%FkT<ENwYT1E~
z*!6Q@(jhxyPukJU+<`S_`nlH9tXtZd-?qB&ptKuT$GP&XuM2~_Iq|iv@c4iOLuNSe
z;)@+WUfJ+tf-UKt><C?B$*Ak*?0slXwJZzvY_`NH%!XkJcBDo1VS}?ch%WYw9?*~G
zX?-=1noxg(3HAOm;f7Hk?U&5y)X$s)Q_WeIV9vf9X43YX(c+^S?{Y1$++mG>Icr89
zv*7s<3(oH{r~f%K)+8W3yAhjb^^kYOf$_rITA%Hib1s}a;)=$`g)`==Gy;*z#*)_H
z?kZueG;yz<>E0?omT?7j96VP>-MT@{davH2ErapfmB_*~%EYTUf{Q&yaMW!C<^CCg
zr*h7R-5p8xd-b2y8OuVKG;SB9D;G^1rtr3n@b<8LDtB&XbGXTLl13D>U3j~G=uFOB
zoJnMzS&Y1*F7-}}`O$TWUaO1QYQ31>#x9^z`8n*)nTx@EDO-iNd%|aP&vhDOn@#0T
z#X>TQ3n(wV9ojaZO;7W9_&!f~m`B4Q`7ElbPMbGVn7FBslluzk`o555j_Pa`zV;h8
zjrfhzSh8y>_lu^IeY21a1%;d$G)0{kh3d5uK4+>!=h}4E3UAMw%@R%*lijnJxRN65
zs}=DUb?wYiMzBpGy+#+ZT|S^SF2yVm-n#xWmA=ycUZ|aiWrYH|PRVE0*F1(A=drqc
zuCjZxIkYF6yVY`t4x5bAubOY`XYr#<mU7ZE<$a&Twepj&ter;FvlCeDI9B;^W7I38
zj`1$qp9*i6#7t!T3GD#hP2}jzG@hrcBm3h7M%-4f`KfWl@6xVy;Q#lw`Z}veRClHm
z$A|MpJ(XUEhv9QA5xeFi825NM$A!18ltUD&Jcpgt5*aVN-Pv>)Gg=R$ovk{}^OESA
zJDjI~DR=%<3O_Ds&K;h@Sf>=uTpmuVkJ9oE7VoP)tCx*3VwFi3Eq`OJ>{wjq%6Bn4
zR(VS?l)V|koohh^x_V>U+JkoTnQu+feYcM<9lG|X!J;^V9wo5#tb92c@w6Wzyj>u?
zt*iOAWjyHv<CM3c4At&2JiVm-pL-;4H5>hsFVA8b?czQ~F<f|CYM=D1((d%SCa(Bi
zH2W9DvbV4BS9ojVE}YKnPj_iva*8785ut23`DU%01IR9`M^M23&AJEv4#DQEa?59i
za9@6gTDyWcWfH_H={ep9>I^9pz>CZNSor$W?6xn@#rc|C^Hg@3hjdYzQ6@`^u}FNT
zuyW@(FP@rtGkUQ%Ihqwx?Y&vF%A4cWq%#fm;!kNnEcN&Mlrhlgr#m$c8b}^%pkkqc
zTVt%T`OAuvX&zL&?}JUUFX`IxUY~5xuGdXi@2X56X<i2hF=f8=Th9ZSde(=^!rN%!
zZKA6$@ut$OXvb||BbJ<M+Qnu^%DX1d<k>*l`3B-x6iB+cGC9J7sMjcr*23F)*1;?u
z9jN?A&7UvBd8F^t)NbL-E(qgBE%_tYgmNo5l&!+*^WB2ECA{Ar7R=pFq4W{n+Aa-b
zv^dBaQ{;2f{bl$T@dvwwuWJKXze9SHkKT-_=S`vZlk?<hG-%g&*CT+5{|IaC0(rbn
z=l6VH{wxfjwD7h=DFaChTsYUsiOu454;VSpa-$QUg}3`GoOvt!@DbidMY#}W;lz;R
zI=A;|#%t+IeR1;_+e=&2#)XsODYNGp_^NsI%K*(VQ#7}=@#Jb%58P*KcPAgou>yAt
z!rR5#y-qRKPA<cl+L_ke5Z<Ph?ai|9y?8jzgf>Ba*)F{O?`B^<q?oF=tS_6s$yU`6
zb@amVduBYmZpnv>4%}X6$I}D{KDTkALOmxszjDOd#z}LcGe?DOrye;o=1*6CRB+>i
zy8*Ko21cINYiX~%Cda)n(Jt{&yqEIdJX!IF?gVc=xZ2FXwRLXzB^z+m>+JnJcX1KQ
zcPQh|rJDx+ur$zmljg$)Zn~?v(z22pi!^7?)BNgr+zFd!nr&M-GUA8>Q-xs`Jsijp
z-i{sa%%MzY>J4|G(nKfz-Ql7h0bd$S)NZw_KiS%oe7&Z-kT`xHz236D{W-hApE<Yu
zlw0mgP0h0doP7v;>P=7GcTx_ElOFCx@HrP6>uz!RhZCd!6$f0)k;BvM=&;_71)4L1
zZdmbjm=)_6iW3O4<Y<~TmY-eO@Q)R1$Juh@nk}A992vIPfl}jbc{<aYviEKHoN7zm
zTq^?BS@KtTb1ttpXMHaVK9yKfDNNi$tQmc~no_s4DecObatl)?Ozg{VcTJeN-UR0Z
zCJf0jrEi`o^>lBFPc)@MgelkbcmIes!{o6UL(5vI@4{L+W0vf&GG|x31=;(|aoS^s
zU4L0m&5fwrvOD!FJ5Ww|`{9#4kEKB$*jl;Q7V;HKJ22{#vTHsl!)9_YC8i-%x))4H
zlMo`*lhD6Q2v$zvv<(QSV?X)9tmM-+ieQ|25Feh4=Jv5TtX9P{*>Mow|0<JJnLA&X
z4rM@f^@~O+-@I}%RVF0U>U*lP+QyKTlFhmJZ1Or}a8G*Z;uiVLf2mHHZ&Q`?JB>DG
z)A`G!NO@&N_&h0M_nu<bte?e(morF?pCRvh5pn&ev;Ol`#;M!QrbZzTXBUw9H6N>H
z+9f|x2LDy%C$*NZbbpRMCLM0M9QG*p;m(3=W+i6x$}XGs9$6F&okqHe`g45McjGpd
zz@LS565eLkQAU$;`g=|(q=8?S{GvJHWu(<Tn#)%4%-wnvbEwgD{?-orYt20R1m*o7
zrnm1<$VlOOiFq-O4%)Zw&0^@kIm|w#9H-IwJj~B$>M!|to8*ylH-|#uZOn>n3fpG0
zx_c%au4ZCYISZRT8GQ5@%c3D;Fg}w+qYvuOwjaga^YZuY9m!A4t`@@EK3~QYXFFP1
zLu2UiZY)7-$B~mXfxVZsvrSef?1^#O4~*r%)Kp~~s|)2sG8ahZ-pS!KFQwd@f?>qT
z|Nr^FA>27ToU|V51KT--l+8ma^Gl-qpNSL*Z}Wt=kCjm{eX{(9V}}zwNBQ!9rO-=w
z+eN$Jc4hRC56E2k|EGUXWLn4|@?Q*4Zf!h8zOjUli6u<hq^<J#|9Me+JNfc^itF6E
z(3@kriyL&#Efn5v)q8U6aqTK!`%?K@g1jSftQ;Cg^nU3Jq)D}njc2`HZ|k!Ab7i#h
z51MKBdPN+j@@_`d5m%F~*QIiK-rtJii14;daEx-IqUf|Ll19=Fy%FAS6y6#MZ!hKd
zr|-yUp6TCswk?=zKH}<5Xbx)NrH)5W`G>sp=F$FWo!%R#Ls%xf-6R}own!YU=CP9r
z{!|SYKl#$1*{A*a=H}1)qrM!v<D+au`3g39;1nTj$abfj@V4xE?a%&JR%=^t0vCAG
zY_K<r#VubG)-)3i*OVXO^Z<9d{A-~6R5$D^YM%Y+O2|+H=L-yMEo;rg9afB;q`j2E
z2HO-H+D*|;s*;`Z%-sm~*WLd=;hmH20*j=9ek6_7z%E!#=|)oP-mDMrOaGF7yxnHT
z=vn66nQFnqXI3o#?ucu!3#U%|@OVHVgN(IP3=UFvV+fZ92V)<k{i1X*-nTq?^iH}h
zov&lMON%Rg^rmuQ94-jusZ$6yU4prxU7nR@+}X8)nb1lapTqwA5KcUtC!dpJfHFG7
zxo`7ji;EAb=lyxKHGsBu9_&8vh4XlS;<kh{-#$Rug#i>Bgvmer`PD{q`B-0;-gn1!
z(*JxM&TQ%6z{_=xe825LM6za~4NjVio!QyTSvgDEwV7%ra&)0hlCZ3^1BcQb*l6m-
z+;2Xb`Sfw|w&r(j_~UO|mY%bvQC)kUraSRhOE)%XzxjNfJ43#?;~>0sJZ#{UPcMQu
znlQLbFXcM*WP@=p_P*`K+V{QKY+_9P2xF3qjkz9gf>(J{(pL5%_)#B%I`rjcbU$(*
z>8##rE#Im=4X4;KINE`;^_>`a%#n;$j*M9%KDCJxCrzA{$>B_*=HJ$L-T2Q%Gwu=t
zz77UvNeAS&+Cbpnx{Eb*=UI?D{q>p&@8UtmJMqKIT)EcN4X2&r)a8X~_Cz@Qi#vh(
z{}&w9ygbv5#tq%r+TWGz|6FMv;lj*BXL@BgF-&)o;bR<F;A&5&Anh?X+DpG`&jam6
zKa{iQT9yOm+Uqv_>BO1&9*qCPS6Rr)+&Cd`>PdOVb$=T1H~_c0>H;|9&H8y>9FeE>
z+e#1Kn0T=Ny7s<zTxeKd_o0;zl+?86Mh!a}cX4EuX6heXwL=|ehrO*GF`6B-u3ItP
z-AY~;OMM<oe(z@`--R_#9$Rrk_}XBfEw5JCvMtY+zm05Zm2XYeLDrmZWsO%A{eOR(
zb5VDkW`oQb^2(g*9_9pm>&u_z`ZB4JDXo7orKT|SMH^F^+%VyMUlS&Jnvm*c!lGys
z!U{}?6#fQpHle+zDYJf<(7S`F&J<Ikf3YB@n^>HgmUOCY&Kn!l)6kCt?tO5oXUt{Y
z!Fn9+rroizaxP8v8nnSx=h;-<XS+TNV3+u?i%SCNJ2#N_leFVb4_59@Fm>het5H1!
zGvRHXqx2SraQgQRXN7q<6|@hkB)qNnTwTGAvD|H<Gt4obGpz@cynYB3l*?T|eHeRl
zhEeONdZs37ruvww*>DVNlgIGenKah7&0uvxwzLfS@{kvDP?#IhOMI(l-?q!Av(>+d
zBae$XbgdY7|5+@0Gy{)e#S}{0>nOanxiwWf{6cDcluz|>K1apT_HCrS?cF?vX|L?j
zLVlV(IefKK&hKOGowFx%TzFe$Xf{E%+4PjJbx%|lheu`MwmyrB^|hlOQ9#zb0*<HT
zbG*tFI`l3ePq-W?ysdm^5-l1@KQ>2Nvbq!15jdXA6Y8RtRw!6I?STuG58Sqp<gd9*
z{+y1J@b-3%Y;L#Cp{@L@W22QJJVE)xos?H>lS6joWU7tM;=;e$&n}$A)=rsR@J-|U
z=Lr~*N-s-g7Agzi*wG|DzfPv4zB1y4w~LpKWSvEd`Ug{Z_g0z1!rLWD>PuNNmO19*
zxS@UOxg8VeIdnXJ$H%ey4|SFaZ|la4;ML`192X=pyIm5Or9;}EIgF9fiSpMC!5}Sh
z>xDyf2T=aI@&LP)QunCvcFTofjD4;Q-PTEr%1FXAMR~X3>Z-0dj0(RFBh4d)8f8;x
zpR9g>G4k(J9KxW8fdp)fXLg<bymE`-#gJGWD=7o^UKDP^TLMB@<f<78FD4vzSBAX@
zTb6sX;HwXt4@n>N(wAeW6Lc5WT&r33!JPhl`%jvho6@kpie;=kB(4`DI4VtOT8eVz
ziX(U_9nPf?c{NKSaE^(@ApfM_m`IG8>2t1##D1{)xCchE<7x!!)LGK~g|cdrqlp?X
zUZ-=2GWUFWRj663q8FBjm0y0ti`M_cTfP1}C>!UEbRWg?ZCsxf#C*+Z&7*{`y3g+}
zCw_0bX4W^ptlux~ad&T?wDF?Uat}s^8t9qjt`15MTA%eGK^W}q<VjiaeAhI4>?|w4
zy|X*>q<f4L78fLFo;n~dIp2*rWnD?$??`xw4ZobUrg>>=iUKS#6Hi(4K{e(hZ1}Ul
zh87EKIP=pMqZ_*Cx3na>tR*vVS&;b59QXcR_;XS>uD9w<K>xlh2<S)gLFl5pex2bK
zJUe1V%S2lap4Peg+Ma+L4xH5I&Gi!>-7$#2=ct!k+SoN7f!a@cvbwb!O?*Omdn1$@
z8DVVwE={j?s-?dM^HexHL3q18KY+m#{8+cgmm7ch@!=0IzU2GLXRSP~A>z2>-5CFu
z8>Y7ItW>7n#8d8cXyZ=A0w3wve3a{?jIHgO$1=S5?&2k1w<r5!eAV^o&MVEeFQhA-
zZ{i@2tvv?et&wy#H@?~6vRQW!BWIcmZxP;lcXz=%&Jl|$4(uQAK)8o3?ezBsZPNLS
z_IHo1iHo*j=4X8$_uKI`+=-F|XJzX;v+P%Q4(n$zJI|QN0lhh4-HT?|d-An(FTR}Y
zNvAr!h`iEMd}c4o32)mv8#8jAF>R`t;PA?XhDZBQ^Ku`?x9`jM$bO8tWX6>Y8+P@u
zBk_w3x83bo@K(B<2M)yi=RgxZkL5L;Sk~2<uUnnzvBni^&9|@KYL8plz#DS|{j!C(
zs|<JuZ%Z|GC(74d{p}tsIpofY{cc##a3xA}>oRBMH#9YHLq3<3`!4dzI<vBi=I6?e
zd_Ckye|uMDZV7LZ)<}EIoui$&BHWJ|;DDd@r?<@Q*|XN3y3!FX`EJL#rFPWrXv@S~
z*1VYEM2oo|+!fBJ_VeT)=@HU&N3Hop8b9#`Bkvg~>+MePM^7q!bK^;G@dDA#%$ey-
zl^@RhE-%jQcn2QtvBT`REhC%SaDTHc-4pCtoFY!RnjIC&Tl33mbN&&Z>u+Pu;8vDQ
zHxeIT%L2!V=2Ylmj!z3S{yLAcoBPR6){mF<pk0VHuXHD^k|Rzx8*!?t^Gq;jafvza
zUCh}y%#7~leYh>Go&Ls{3#Clt5j3Uge<oO(nviI2!f+20?gyCANsl}$&AgT-$_3G%
z3vVlRGQqfuiL`T?b-!5fE!>ju-P#SC!9N!LSkcIY>aM-0)v*UAe|6);>#p>@(-{-<
zPIP<h%a;oQ@(BfyrAO_#0W=*INTmt#=Np2#a$6mM;>_mC^E@{|*(Mc(i2SPe-IpNB
z=zW&{B9xFzVVJCu*2-R;gJ%8d?GVq;n#zvZBE5X8MDjN&Gf(qV#H8Vj7?4atjZ~U0
zN~K4yF%;BK$NKdo-On-@_@#jVE*J9r-c)WS<}o!hpW4l*Yd2ZMt>Z;3SW(QvsF})u
zSN~jMG3CY<vBu<odZwvFwkl-rVP*W5Env#}T<-nMX7}eT#@xu{)x`|TG|Oa3+e|`M
zrt{8d6fG;KVC^-6TD4QTV4Ovc@b=%-EWFlbNkb@K>!mDQyJzeEnUBA|4&UGNDd(1j
z(+bV13#H-RuH9+wL=tOF<k6;a+)E#;KK2RfqnM<(_!QPo%@bcdj<qM#ao;O{O#{uX
z!rLu|9Nsj^Cx4x^$aCfYNuSJY@yF{TvS>JB5)T(lV&tj}N@ZqfZ!%6k^;GevNo?pg
zOmkEs$4(`a={N$L_UbjcpG>v15p=#ej2FKQmCjzALdOBT`F$Yy%1?UJXACVnPN3e9
z@wDDPjtk4jDhD~0SJv9mj+Bqpa~RJ4!}xW>FcMb`#rvf4{&I$pwq^*w$t&wVQGH-p
zgPEy3g8CDcAG}0e<0psV&`?<WdlKEA39ogRs`WmRmM7F3vrL&2!tunq>gySoMDy}P
zaEu;E?1p$&e2!I4S~Np?#qy6)3{6Zz*d3z1?K@?|t<?Nt<wf0X?i4KXBuq2kx)(k;
z9#kIebLof<Cs2J@9LWRX=&Akhzrx$g$0InuK8)M{gs^3_xR=r))L9lxgEC=+j0vZV
zZv^{h2~TYzX&0cZSmAA@Mo~;sN6B_|VP4o6PP%5O7E2=8a85m*!=foxp28#L-<>a3
zX89I(&fRfmbuUllfV(nSI>@DS4D2y=r@>oawyp4INrM0uI(w5?-k-+J<%j<wY!%)%
z(|anom$WP`y*W_Ni;I(`XVlzb+{K38Q8wgXvSHa8TRPV8<gK{AMWuzqdMp#(HoE1(
z??*f^iI%Qsug=pbc?Q3_P<f-HcEC17?-Q?9#|m#}OI(_nQATkD=LTv<J!Q^y@nwyM
zS&%%=lJm91y_K`%g)}1HpPIADt_zP2cV^J&&IC2+%5Ob;GQCu93aXg%PG``eKIVL>
zZH4n#JA4n?<9W$}YDcxV{urqHav%>L1~6c+_+95frkwDg+jm!@r+M=Ah%~%q11T~O
z!e~w~-K5c~D6Lhpd{O<c22g*zAI>*@lpE;9m_$#GIQ#P1EC72y@y8Kv$|o?8Zsks^
zOg;YqcjmTor>gLFfbh1ec>uY;2wR4_k?G{h{SGcP+9nRPraR^DJMgQ!1NRq8)3VK8
z`xATcvd3GPSwZ)Xx6!s-jJBh#Fxd5}BY`1~bgLjPw7VTi;sTa_*U#Y(OROxcd34p9
z{CC#GT(_m>0z1kaaKs?3Y>fbUVZ!>+S2IpZx-r#;^rrf`?&RO@&ZLm;%yaL~gwP%=
z%Qs@@lb&qb-ixjJzP<|U$>mBW{PV8~r@T!#J;)e8e`8_~_htHZ&9|?uDe<=@>x&gr
z9@sH@p*`KUIjBd;fhFf02s`OWgz)wc;qBAat{8;3yRJ$@+{nPv1_lmYccH>by(SL1
zuw0Mkmafd*>%paBS1N~SA3V*KQ}0~yG|~Q|j)6WoZnQt;!Z6{jt+@ARl^iLr``a+_
zCf%31;xo*db^gwbuHcMeql0oZ9OQYlqfUt(-plOiEd1*^RCl=HcI@b2%avkl9_RJp
z=L!>PYdrAmtJguCCzUfjNztSKCU-u67VcMf=fF!Fs?6_4;&fAfUUg!<cGLMSZJ5={
zhB^P)GcU}JJ83p}{j_H5dRxja(SQ5f(ebK&ey=U$4>ISzUYkW0=48~cq+E9k3Wc|c
zW6b%Zqd6T6X7oFto%tQ@Uxl}Rbzs2Ze&qLr7n4<+EWCYmwI36znu(7zlNZQX`g>y<
zP3b|-9sP`68dLVW=1*Z~qfeS|+nI1uSZdMOM43*;^t)$_(OLa{eJp&m2|I+zAuUZf
z+*w#$)|7jnEto94&H2@W4V8`9^RovNF85&j${wt$*MnW(yHYf%Gv?WyrGx6k&#Atw
zc&l8JpW2)J>cev*<z}QRH{*OD)(bV~mI_j~o<IGx_x|P|MDF20`sqD*R2Y1ArgGK~
z>3(}r=~wfk=vXa=+T!T`eW(o6QweO+oODm#$#Y8+si(8rf7)<9CM2^|yWNMbqp-R>
zhLG?yw!fUjyPuP=IVs(3om}F-=W<Bd4eNw+y-yUdW88G@=!^KhNihR0)rWIy2A!1A
zG$~m)Y_5#{zo$~}{S;Q8RG&?`0-6R(>)S;d-=&$Dm}W>%Gl{dM(wXj@M)X1Dm9I)A
zsZI(X4<z$#!AOpGAHzSI3lnaQC85+fHjdZdKc|_uYqtEZ`LvjjhhK-$;%3zGDVz!2
zG!)mQq2!kuimki)R>Z6B4^z(l<}u`Ko5bd!qglRY42y-ge{4?Y;-O3~R~26_%pW{5
zi`otHNf+KenUlxqw8_*fn@x+qGda^ek&!(nVaQIWdRn^r<R)@$+$i>VCGp@l<@Z`^
zUw$%~-=!bk-bP)pe<joLQ8M*}#ruV~-(r-}-gE%*@<aCfCU5Ig<>0&@&$mJ28Lcc`
zhwy=n+ckiOrX%>AGEAM!!x+C-x+u-28y621ehuY{^8EJwrfj}{2QWlg6kUEBOz%AD
zrDi0uwqzJvMya>FeiDxxCF%RC-^V7(udEEBL;Bi;x5fGL#j2mvRXG(SH6QnpmUpA{
zHMQTzu={~_4mD%B-Z+M;-9z~77tCCLFRnLql{VR#KF{4)>F>co?bss!@MebmGF2}7
zvQ9dShFO|PEtN|*s6QRWV{O(fRHRItO~Tu=nIXy*5f>vLpwo?DDjf~ubE^m}4+Ju6
zrF6y5!*FUFLBqz8G!VX?Y8c5Z;cbu}Zqf?Y6efSOR_0)E1kvL|VWtOjbY}c=!CiMZ
z1FxIAlBgN!+95aT0u6*7G0@r5U7d>_G<EQ%WSJMmi#4C`R)3eVG{-NMDX0D5(uUql
zxh_3ux(6;RtO!4C$)sQ#j8599^U0Ps(bnP;t+{D!h3|I*C64Y?5`X!76?fcQ$fptN
zinq?k`k(CCu07Gwo3?aWZ$_-28ON<GIV&yB-ON4|3vZ_gZyWWnpp@`7x3nd{?dwIW
zw?;I*)E$qAu3T)_Mdx>C<`3?yjI%DB*MT&mNpDsPZx=Y5Yu+<M(i~2%)tOZj<?;#(
z<hPjnSUwL|Gpd#|r|b-SQj5I!pt-55=E9_X?xag=-M5N26~_9rRcFFG&mfL|mR38^
zpAGMPI5SZCSz908KJ~)7hX=Q;e7G-N=h9)qQ0<!MY7U$J#+8wkgtyPMdpvJYhPHwI
zH4QX;q>KXjKPpcU-{~PO&st}uievrljC>jA9oW{}fvnl~l={P-DB<lQ?75_UUH&^u
zhC15v(#8(|NA~o5<VeSwj<on<&l8>F@4o8!{0UWWnQ_*_k`c?SDST-S!rLCnc68h4
z$dN=FwkP(Ze5*b*8*ZXKbWi1^88PlmcY3ewhEs4iKECX#xwN}{+D7<Y=t=1_JsG{B
z2UT13Bw842`>{7}^LwkOsTVa<dXwO4LGL?O?3Z4)#MYJ>r>(@}+i`D;a3|k^k*_p+
z?-J%p_cQ3Uo}08o9U3^|KEj1|*)Cl9Rh}JvU#1Ihf1YsRpMx&afx6J3sw*3{gYBa%
zqY}NB2D&+ut6x*aAOEhIxU{_)7dM-;e3Y5EPGOWV-Br)=NqP0riBmo!EZ*eh%rW8Z
zheZxderC_zR^oV%ThU>oEpZQQ+39M_xA`{mvstSH!V1f=eP|J6!U&JvH0p1xIbIm#
z)Q3&FQ{;#*zZYuD%?sAt57mx)h$$TkOu4X3JgYE&#7Zkx7Fw|FtzNTw{k1u1#)H=V
z=<rKlT+f(dKHZkjxyDp#Y>bugEz-mslg{QeH~}3b@lLMYPkPLLEWBi@yr({xo#?}V
zQ~Pozq8~0*;q8`wB-iZ6u1<aV{Ci*QkM-e9Y(FelnlM3lE2lY589g{Ox)*+Ddh!06
z5jpjZShmNQxPOeXdD@#b4NPeD#h72N8MC9DG1f17NF!pz{5M9lsc(W&T@#94m~iw}
zHwM>fOP7Ofn0u%L=G!|^YC>C{FK;Wn?#hr)UCI5{fwfT`Fm2z7L-<ntwKsmk+eyOP
z$LV@MSqfWIb(htgJN#mxcC7yRNI%~u#-D351C?(PNN7*xt<MkQk1e58+b1q*K_s?!
zqDYlLVCl+%bni9@-w#7LE4<w?Tz%d)N&MM=ID6yOz11L<T3(}gd3+2%!_wGbpN^^U
zwr8zOnkm~M>q9OdWAa#0C~Z+#A>%Ekk&&QXZRuh@%$mWnRQXv47h~K<`sU%&8DJ-U
zQ$6XRGP7ygGMDRtnr|bsIeKOiou8yL`Nbsm32*22N#jO_dZec*6V`7KOJC^gSuv7r
zhelz1dNf`ZW7&CTEW->L)EC}nMQ8J}dOl0i{@0nKT+{*ToS8S2Wko~jEW9oJHiT93
zx|j%Wi}ob3?sF2}(~?>3HIk@SV;EpDk)n0!+$x<xLFG*932(avWhqB8pRi(ONtDl_
z#~5j&{>!5G?o9T)8H)3fbUNfnTh?+M6*VjROM6t;BZ;P&!uJM=7(G?S{7B^%j!Gc*
zX96EX;u+mEPCYmBtA-}<=pW_ewi!rHPkBriD(f&MNg2Vy%R>VRyFGvvYX&IuP93i!
zq(>IsF89{1(;=3*BZo3!nx3D}5IWu+z*XM?+TjhNweYsywZWXPIFwFD68Y3?7zIz1
z6I(TjV|l|kC>>SX1&PFN6X&dZ)qL^KYiB3W_GS#zro<?hE0_^QVZ0T8J4JaG_2$Mf
z`fp_nZ1&^b93Rcs(tPM_`1OD@1FlKyf5)J`v?p)2XijV@eeXG69+e6pFH^bc*X0{M
z7t6U4bweL?$KT(b<^O08)I<BlcfoXe9Yl>|!Mr@GJR9*bN9)NeGDSPYpb&0ZD0k@}
z`Hj}Avr^gSIp5`#SmRAzpJ0y73}vjcYwFhsCcT*#J2c;xALq`JGX`u&8Ms+WJtCi-
z@mu7~Wbt8v|2gvQh9m1tJ(+Fh&1~Uql_)RVH+d1Y$BXmw&2|nKx7Jg8#zRh6C0SE3
z+=>b-E&1VL!{oy@)SGF`G0j#TPFS&Isik%(1}X?|_mpTh5bi`&*8b36IdBFi+JCes
zQT~M5gLE!=2_uAmTP-c=T%iwFrQ_&RXu^}<bl#mYX3e|aw9t<B-Kt(h{4l~`)QyHg
zU5RYjg?sBdQ)fVDX4-Yees_0*I|y%&z(etI75kWJ_LMDnwf0Xsv+mDA{RQUQ7n<<}
zGkP6|t4ksJfGs;NC@Wxt4OzivWYv<VQ#^XLWMu<M+kDqUdA6?u$%~h+skJW;Y~%%9
zseGq%Ui^F8oyVEp@-P~hIM|IKKR2oiZ(ENQHsmV{;*u-hmbg;=u^VRP3>>`R#&_`o
zK?C$>2b}PJBwvc=qr6>Cgb8o^3U4py4v@dxj_+TD2d8aWonp!9XBOOfWWkg1=2*5f
zr{WXg%*6jk(s{?%e82A>t=dAYAbaOz?{%KpI}#F-NQ6it0~z+-dvDs>TWRfDwf87W
zjoPiEX6;d0e6P>%_s4m7v{i}sIeA^L`?{|Yrd?&cv;}(Z(w5F~fIR6)mtT$Gk~B6!
zkD{s3Gn(i&@{`<*;`nCio9sgQ)iQ*#z#yvU_%ZIjFFW3Qv0<qfhdX#_Zt&#w6;FnL
z?ZI1LZ)R@xCS$e_4IO;BB+Tun_hMbs07lODW7rbydus(SsvwX{U&gSnek^}W^VXoI
z_LaNhrTtCdo}W1JxCC0ymjCpKd?dOfG-)PWG$b;8Un{x@Z%1mEdoUvv%Z;sQv#J#{
z*2=efuoVf?I@_FRMZ=rPEKSfeDvzW}$8hD2NE@y9cB;8PI~~KQrF+xypA(qXSNmdV
ze%<X8x%7>^Ndpsj)+vDv)e?BVH=d4Nqm*qDh3iJmoVPV=M#fSvDTWS_(Y&g!J403Y
z);<sp`Kz-k5EtQXM7I!S`$#9YDVhbF#F?MeT~c^^E+CSYm&C;`38Ckk5KMkd#J*Jo
zHGd4}o7&+-?Fl3JpHS(vLg;uZh}FLbvRZgMyICNCcY?Ta1cuepXYn+A^_`K<!dtJ>
zP_lOh<MT3@raMBIc_#$dEF&l9NtcjoB*MhVcH!-H;ce-b5UOVy=_I`Mm=wsv+g?20
z>&c>C{+!<Bt{hqy?pfL4x6qoE3D)#H6QG`VKklUaaJRY-@w2_z)!m!s^?X_S%$Ftg
zrKfroNRRM9ychctThCG2Xa_n(*z@+TJzc(bAaA?_z5jGjf3PFz24}kFI7$cSz!qsD
zzrWQ^ckOl@NRUotR9jN#wx#YG-B*RT8?@JLJUvr0e0xlUw^gOVZ2FzF5WBJ|C=r)e
z(usRHnpIwRB70pSDe9u|JlmBSm%3tTQb_~pTgPQe_qx9)mT!78Hd}b)H;{}ugIKw9
z2%bYn@hEmQRv*S-vu_w3cTS+%y9s<BH3_?}(iaa^#-Z>wzxEsse3+xWdu8YMohMFR
znTQs1_`6~@A04EZ8abUQ2@}+tHJ)eM!?qNzISOy<Tpi7(jU$QMHbQ*<a1KvaH@Ial
zZl{+kC#DaNPwE~}U;3ZZgVh~6n7_^rWxQdKv^hgr?>>@K_cYtSs^sPP9z6Q0hq52a
z2^`piTfNlLS*Kk4j0&24Qs0N>y$x@B^Kx2WqSp1~{0`~2(<(IQR$#JYH0f4jSi%@?
zW{zjk&QXk<GlgG0$1r%ays14Z7?fQ>#}z%87h6HgPL&M1S;-M&Uv91HLr|&a)e{vs
zmiM6U({9obmh$|EQq~PAmENtC9~zfpqW!3QW;b4r>_+xSd1GwlxtZNtck|w?c_2Ua
z;4=Jw?9THA-Id$bO})D6wJ9y(NvnLW#uVeCzfU!j9+d4bQ_g59lheC%Q9jwtw!-H>
zDp<R|5)0w&W9^X-jOxYum+F^w7tV(CRCc<&LYk8s3U8+gZ<D_1#@?$%bP(QJ{+7)e
zulCd`E?}Q>*e5SnX5i0xj1%4(g}255=?q90CpAPKb-xtx)bd<3)yyQk{i#7F&Hu|}
zNHyKf=P5Tvc^H|Q3Cbf)pjqQY<sKyShw%31qh#W{rjT~51Eu|Qsr*U2CLQz{YLBbV
z`z=Mv5BStUULEBae4odx^n5=5D^G*=#ETwhVJA(`#eVI1use&<tJ+eus13c!+R))_
zYc6Wm;!-QV&|GybFqtmW6dgE`$miYhn7xp{b51Mjw@Tx4dRxrKw^e^{TRIDG?WGgx
zxwkbl)03oii)Kkq6pt1~Do<aW@WB{v9}?ycfXZfJ-0Bv|$csjPa)cu7R1@FAsu^L*
zrP4y^KO@omjMCZ}c^w?aYW=s*nxz8Q`B3Mq52x1$Nk<>VWpP{=kBakL@58GDK0F`p
zC+&wXPQ5%R7~)P=6%Y2Wb*IZaHzo;hgN3)tECZ<L7tF}p@N-ZYJ2c`w{Mv}=O(Sy*
z;^$n9oEHyQL+6^?mqsougFl8xF`z<cS&os<nPJ?R8beXXScaZdr;&6Y6))2H>Pk9Q
zma6xxB14{{4Bk#@M{f5vr0JQi`cWDmq_e)7!VLLL9SW0~UXa4(DbnFq=zcLtuZGHi
z*qTbmi>XYzmdY8uw+Htnk$E!-k3_u(JCX=plfa6~cnmS|yj&s<_}^GIo{ObPzew(i
zL;tTz1iib5<1f5*6W*rkS<Yx?<dM0NopG9j%whjYX;j~bLuMq~#KEu0i(;fW`|-ky
z9&0sAY0o|Hy+0lo^}HYY(0!see|7fatM{JD5%A>Wcu&s!?!jem5AF(Yy*GNZq{K)5
zbzfe8^u_+52UE6t65Gp*xjNh5)(#-4Qy`vGLV0^W9LpX0dW%9B_8?AO8_Ha$nLynW
z@hk|6rRmBjehtw4yD*xc8j9zh9!swr`D1=hmA6yRWrMt^D_YS+uP#Gcad~|!_HB|^
zW}!Z(=JL#%Msl@NIKOlYBk2VERXdC@w=k+r3ZwehVI1uj#k~*O)vk$U$fFpY*W%ZY
z#E@nY&7yB3IW{>Gmv@oOY$5+@)mTQw$1pe|h8DUbu4}1%#<oy$KL(Jt#~+gpex&^(
z-0}0J#d9BG*80+Fgg>U6{5iSPp9`1$>AyA*uTdeqco~AzuOXZo>c@XW{0aO!jCVW3
z==UyEcaspBj|jq28249CPyFw?bN!sKZMru$%e^>%-<AB?Zk)a9!KtNwR51-;*~ef$
zYzv{!^$-qR)cwfaC@q$e+uH-R4-Dd3?O?`6gmC7ZcnU)Z(<cUE@hO1h9iEs4c+kbq
zmA_j%b27K7I)JQ{9bqNSiWNV!^<!u|AKumQq01C+{LV=C)X#ypvu)Yf(Ux&NZMj>;
ziA8@o;v8pA<xmH%23m7Ocs)1Inm&Ox%5Zn!lb0j&+#LAtggs^+4%&0IBXL<fmK3$)
zSU_9qt(Im*oY=FT(!|fq;JNlFbEI(?FK<H!aeI%CDkG<+d=IYKn0HmKOGrmPBzGi5
z`Dg9KD^GeTf8G!3vsm7pE|xtQB8}=0;jOLkw&OGXn%j#@4SVx$>L9v|9)kNA>1k&V
zrDLz5c&;4AG~w-sw-a!To<vUW6w*h_V-qwB%V^=OtNyuI*@g!u(s%s?GJYISgP+E+
zq4!u8M2=$dweeVdJ)ZmNBUv3d64xOkNtrs5(jQ04w>5%H@v3Ln4N>lyc8m@BFu#Ys
z$K$;*kv@0#pq@;35(hh@Oy05qG!Gi8-no(LaZ;D~@0IjS7k+wFsDrf}htzUTS1qR`
zwTyR5%DJhzG))|9?!w+2c&t9us7fYptB~fgob%tQAMTZM_EE>p$UaPx#<#9^zXy(v
z<k85H?2c7OjqtXfaQ1r3O7%%qP_#g~{%0yUJFFW^PIP5^wK5ztl?y+(lJWJsu|xT_
zVGD(cgG;I7P|ArM`DWxTy*I5yxkSYbdRWQ?SN-!O<>%h*OqbgQlv(D}u~eBzU*&Q3
zTXk})^K0_-68Ty?GxCc9^=XJ(J*>{>y4_g%TDi1e%6lWcU6)wO<%dN)=u|{|iy{W?
zSI7A{&A;2q=u@SVT|wIUe%+JJ-}hwuS!Fh5DeKm=n2)`S*=3W4rS5QVhh!4|Yd){~
zcjmFMHpn+mx+(2ArORD7NSWZm+nuFJ4E#gh3`cR4{iP%NF^0gX7%JXGbLxfWk?T-P
zd|X3iwHVe(BQq@txADo^TPbHwc>6v-1^;;+2wbQv*uBc}->j^m!C4$Sr9bOG=`-RJ
zI5aGQQ`1{veos2#cWoIc?a~jU)Oi`-hDXlY*Xq?lU;kI(!l_beJ%*;z=U57h#NGY3
zHJOXs;<USt<L8WcZc8_ka<LV&XSY%ZXBtOuXzx6%E$ii@aF*t@uOW&0dfmuKBC)&l
zl<lJM>?oddHp~t-vT}C}8AFW9D>1Thb_o93Kef`#HDNVu(%t-UqfqT_Lui;Qo@}p?
z0r!k(CC|JM0rD^T;=9U+t+75do8rlWRXUfZxXBOY&XvwSberL$;loEeHy=J%NS7Sy
zPUrjX#0h6xn(EizU6>x}!K`NDI_C$-hZ2OftM32fjWqqr$lY51(^m%Z!_!cHd=RAE
zg<xriLb<gfl>Ne~TGIWz2{*DRO+G{M<0YFTInzCozz&hTPl%*z%}6qYdw(>D;?0a$
z2D&9EcT`?g<tuy;KmS@D(Z4jS_1Dg@rSR5IeDP17DJ;oWKA$0l7W%#Add1gP_F?N3
z-Z{784`mAev@?k#+GEFzO2lJ%0t<w<7p3JH{!JV+PegFMV+6gv3nxHx>+KYn`c!v>
z`9?nK{GE^+!u;AswzM?zx08{<;>>%L2jlD+LRH~yPvLD=rzq0IFMBu`splNP!C}&@
zZu6(o!%to+AF76WQ^(&+owc6ozw_kw5Kj*O;=%P69yH(P%Aj_x9B<*OpG#LdSNc-z
zfQK;HlbmudW+eEtcbq@<t^{Byyp0|e&cij4SS<))fbe!_oHSEM<9VJK&r#uRphpyK
zJfnD$5ykDZF|3^yi*@5T3bo(d7?8+cy61f-tX;EE-lQ`1%jvaMm|G_+nQu%I*fu|$
zcyS?p^d1dy(;ibitgl5F=}p5JJ}-=pcO#gteZTvTaJGkqWBnNxp4T%;jHIFPc8ET!
z$>QZVE1T)E^zWmkH~OE4_)l@a&3(fd@XAPj^8h~D`KuGtPhLeoLVSF+H}E07nlI*6
zg}>|l2@u}?`lCOGRtNH=W(Xe>L-0Nwf@fc8yD{>(N*Il|g((*r(iZs8`lvUpzVPAU
zWDmwmWAyL`2Nvc!;8VpxpGA9?O>^VzB@Z?WZ`=OikJV59OxD-Qj*#yDWGL_4L-GEg
zd3I|cLwtjn{yvDNPlIW#&-3wBzLaTJy*a~8?~x}nL-aG?ZO*MB%~-$9l!axD*uAkS
z$EsN|_N_U;^s*rKk}a2y+0rA~mgkFYiJxjqyIfmL%5C{?g#~#>&GC&fXSXhJo}aBa
zFT4%Rw&q;}8{Bhk=n`i~+C6(ry4y3?$DW7>ZF#mp-g@C}m)hyP8>}-YHwlv-8N^M=
z;OsGV22RQ3(rWqSmF*h1QoeeBamy{`H|&zl)<4wAxg(bewUo;upI*b6>Qj>WsN{q4
zfVQi1ZB`{_&YF2_deZn>Pj$nqyIXjBSu^B`#)Eh}d<bTPhtMlyDCaNt(=KZiPm@P;
zZ235OwzR`FOu{RC5;G$vv*pHQTBS^;$-{|6{U|Nd?(zJ!YaBLH$MW2N6g_;TVG0__
zFJZ!0;cczh5d;S-U-0)4jOjFjpOT03?tu2HiwAMrvm1}3Z>_$ekRAubQ#w?VD!jFd
zC}XpA7aDaDXM3cQZ{Jrk_<AKxzOA5vSp|RoU507bGIF!axc#6zPNUWR_DdN{7M9`q
zw1V&QE7ch+UG~-r9KJ5+y>kin?TdL{uaxhPR}%18-kN&7`0}u_DeCp5&A*yiz4UwC
zD|q|6X4rZa82|0gqB8Y!H|xgf4P9Aar#@_P#}}t|mDsMByT7TUa-uw=eM?EU)|~25
z!k)+SllIlGn~Q1TsSY&xZO)EYhD4`4D$UjDJX`%$!rP?kod~$uo)nX8rphxn{Cj1c
zWOXL=RcERQZyVcoV5)m({(4`^*3we-EGt9)S`lsX)L+`Nh(%MS<<+iqex`KlpDUP~
zTgjGR%6aj&jLb7-v>BDpefj?eEzw@iB8_v&ZCRX`$*<e;wdYnwk+|gRA<AQVpUsR0
z8LYe`pG8g^+KM;bajX@orkWk)t9u~5?SRM_bvi|Jp*W1!8(_tRaPqE58zhZM)^|y?
zkw>xH@njlaPv*sAb$)zOFGyGU5K5IdAwK3_r!1c9&#%0gh#d(G9Fm~?bIqrRm0Ks?
z@}Z^XsNreqF43-5IJ-u8Yip(4Rn1~S`kw5+Pn9k`Mfq)-p{<h$6yCO$ZqiG-(zQ0R
zd}<oU_9yXdos!0lHOkBTkWR?H1p4GB@RhVC7SilA_!!CZW>Ktli6TBNl3UVn%<(po
zDI9%UY2=^Vp|}s%nSVc+!uP=(G7F)mRS0{;t&N@>LicwejM^!DzH8)DnXp(~S54vV
zoQ6I$TJ6n~X`W1+;KA`pZgeel#BHSm^*cJSy0J61>zwJEXwZ3M;O9TxsU7J~GvRIE
zH5a-EdQfYo2V2_vQsapq4mz(aW(6>JKnPtA2UEE@0I&Og$|?wC-i%;U9|mKVC@wBf
z=a%@n&=4cWT1L`(Xy)A?&G{1HY)&Nmqa*oC9PMvIgt@}o+Bc&3UOegECn>DguE2g(
z63OGl2dBw*Bpq<wZxgvOtd%@o$&3-UdjG8~+ir<mA0+*&^gJC>Q`q%1ncX+yIj6bJ
zJS34-BNMqIjQUGj;%o2Y@SGIKWaT>5ED(=8C0v=6^0+37>MS&Jzhfv#U+4^OA421G
zA@m&+%==gJ^VHOxCNY@5N`k2Q$d4~J`!e9I2RGfk@Gh2)#!0xSbKJJ4KOwGu)DzzJ
zcJ}7uc~_pcbW>NQC+|``N$u@H<B_hszU7J`+La?sT-jftIn3FElsO(WS?0+-4=*O}
z^5gwje|DV;z<alLyUVl_J{H8ot6}tCEY11ZXlgf!VbmyPRq3;D;})eJ$_Vl5k?go3
zPia@}QQybXQ{4HRQ+l31$76dynS-O0nbDz@`gUXaCpezJeUr$qD^KriY0|es&p4Q_
zJJK8RB!?S>;ki`&i9AfMbG+CY?uBu<E6r@&>AKvV(?(C0fA(Umt1lU@+EGl3pz5>;
zG7g6^%O;$5!dp|}ZHHeVP@3$?d;OSX<<BL}g6*34vCG?+lPi7nGv$M6sSjVf_!8Jf
z8iIX(ypdk4^hGeauYx)4@5cl2?GNtxlTt<VzKeFX%|kdm--p0zo;(mQP;H@yICXc@
zXE|`LvjerNIgq=^o@V=PX#B4$?|$*3&vt(rYrgKK_i5P(KepEL<L46HMVITIdD4#&
znr{nk2l37}nETHH$rs-G2yesnv(j1b(R!DbG)`&8?6hW7TW`ux;cY<AMx0vRly$$G
ztBcfvZDp2Z6k5?W+LngG+ksPT;e-Xt@0e4&$eiB0%-I}m&UxYO+5fC`ud*h;y*0ij
zns0@-t0U}ioNq_7diJ;oZ__8FQtFpTBkc-mS|pMkoyd2giQ?pvsB%ZUs^T=<N-}vp
zKa+t!WRg5koffw3$!?~Mmt19Y32$xZ<|=cq6N{&Ipr7<AU%t=f=lG7y{j!*z!k9Aq
z9;|ZdNx~G(ZFdW#$0?-K^FpqCIf&e&1L=CRulCJ-xqP7?W7~{kLh5MleLsfb7u8{7
zI+0$#Or&;$$v8co#L{b%$ap_NUU+5qpB+mb)3F@bJ4ziR$~1ISR{mc3U><a1;8)72
z8(u)g!!DYi#BW<lv+LJWximdEcUU?1(sX^=(S?RH3WZndJ+Udk!J~j4n+w!?D(y^%
z68<?=!uQGDm@%Z1-c>5d__dtor^<*GzuQQ7>-tM~oClWE?YlDC7nWgtLo;xm^hUQT
zq_vSwTUfjCbTQ{r)%8|bh=2V;b%6GucKu$`D2?EcEaghX_n_Oma^^TzFydkb(;o}J
zuL`@QISS7gpL?u?=buW|pP+tD;jQZv;pw>|9O`$$%wOJ=VrBM@&f>4O%Ha%F=KiZ9
z8m}+ZJu9EF)ADHceMi!&cH~*#9J*%bu*NZq?4Ok9Ynjc^S;~<cseG}`9U0@NKCEh)
z9H_7N_>uB~rIVXhr2ZIX_E?DL+T$LlT|q40d=taQjmi9Kl*X}XlyptXEf(HRzFfw~
zv8`w-o^VfbDx+*uv6`Nu{`ofI^fklYOJm!}*6b=2_ZpZ+aLYFQ{<1ZD^cpH``#U!t
z|1Wf=%!s5gSK1uy2FHlEtWv|sp|zp9$H9it($ETX>%COgLrOH3^6gFgF^+w`;@Nw*
z6|+*(=tUYik@DJzCqCRQiKAa9a$$$`tFPjz+dBcXjXK|VOOGkMjs7=<1o;tjo429K
zZ>`mDEboMA8|D|b!A4yp^Owtqp?T|<)yeF&ND@yHhk2Dax<84bsaY)Unm3$##48UV
zp3gqodnG5(W~Oq3<kh?2CH%|=hZFzv=0)+(P}nW+rGt3kbj?riT8AqCBa}8jO7o<9
z|AxPU)ol^XtA-)`YZ1aB;cfRxA$&1cx|C%`9%%mU-yL354WM+gFYf<%b9k{AzSX_t
zdGH{9q=uMYjyzhTIiS4*V}-Y^esI>Gsom5=1GE2d=hM$_)cnan*$Efg{^pFE@V1~>
zm|E_y9B6;q-SOw2O93oB;Lojp<uetocj+C-H^SSO4}#Ti6NvfnAa2$VWu>E$KF>pW
zwl<h%{{|6l=|-}?hyUrZO2;YRAtRDHnvY!0MzJ(Jf!`mbVBNDdr@u{NV$USXyGW}V
zl0^T0i99cEMNW1yJ5!|x9+1QzT@!iuGM<vlahmbsX?S0A-yUTJR8iK^+XPPDP2lU-
zdhdUVQ?E)KYCqzfZ#3!JIWF^8ZlCl}la?FVC%nBUylq`A6bt_l%|*d@Yi133thq;+
z>U1-Z_!s_o{pm;NQeO-YJ&2y_sXLq(dAq$ZS?o{SD0zOqaHp=P8|!{`qvJ2G^5eSF
z>5MDa{&Hn-3pZwTb|t=>f$le481Tu3kU&>?Z#?jyEnV7lPclAulD1Wt-rOIv+5WT_
zkKgB7AQ=bbEBP&quEN`-W6^YW_ffa6vX!UAFyWo{o?Qd<GZD%08Adi&F;f42s5G(>
z^eB|>TztY>XL*|bj=`)Xh7UDk=_GyjxzXZ?g~j!Yv>W{z^=iO}I&l4FD5n#|t9~0o
zkiS1mGrU+H<c0kZS5CONQ9i(3UJp-Zuhw4rloy?}n=p0?p`>jHO%F*2Ki&(|4&F?f
z?2Tccw?1!vG=Jku$HRVnnc_$4H@^Jp@5@zTW>KaO70%M=HuUA&uD;}K^ON=@kP<KL
zFDiT)n&`{FV}02(*hfFRdX@R`<uD%}1bXxOwug9010io*s2XKU(yy8koo$&g*Op$b
z>?n@4A=BFi#{o9<OSR)+XFD=>I}i}yK*2HTnToYzpXI|rYZuPAJ1NWB5pSzt@^p8c
z`J*XQzo|=QksDK^t!RCx1y%1frQ>Z=bwitSW|k@K4w#a7uPO6yv|!p7Eh!7OWKX^o
z-$dFf6U3Glqiw}`SkUsMIcbZ{$gpg|pmYniPqxr5#fr@ntSDP+#f>Z0jE}eF@qSy*
z&9mcBZF>T}><Jnt-S$vrg<Mvia?3=vMJT5wNPI;=5-qBxsuxwhu95OBNTWaNW(L34
z%VK$R<)iy#F)l;8{<<BM9i@ECSnbXZC<{(_YcIS#65EjreL6DFy%YcK>clk<%~v^j
zOnomr6W$&<T&VL--B)i5@w+Wf^xMAtdA=XxmknXz5NVw5k72CGc-$IJq+h#<?3*=-
zD^Dh|V50h3wcm}ES7+?0u^efr{+oTu3^t4+zEnHfuLhGhsT+$v7BT<(d}jV9O>BVt
zp9y8ki0i?Yi9M))vw~L_yHa&a7p_UK+Mq@OJ<SVn7v3(|R6tsP<w*-~9}91n{o9QV
zNtNst-qsi1&Ochlm_JHsZ>xO4lij%^jZ%91?yRV-e(!1B`MySXI{#F{xe+BaaxB6B
zWHBbm>Kne9&!UO)fm#=GzqpVKjYg22*@Nt$3QpZGXOERI`cwrAo|fZvyNr#c>LIZx
zrP>{JWNX&k`?i!{6N{<uR?MHu9d7ZVkmZl`jD@#zrf0FUER()z#WZZMoW4!@^z_T8
zczho78|7l|*@2^vviVJTJGEsND|Tejtw|;i8fWs>sS``I@2sI5m|;~i$$y@qyK)C=
z4QYd$weHRV^1H}`-bFm>i!-sb8y!RKKcd)d6-C`9Q4G}%%CBWKQ-sM^4GA<TiIbl|
zd2nUQRJoKy(yT~MieKz|JB@0>+YyDz3=rQnbzKwznqSZ8E<fz|7>Xn0_X~^`ClpC+
zrabB5!YaEP`9u8XfrX*;{b)qdL7Wt)dU$apc?H^SDc@nTvS|L!iRW5_c#@C)FFQaU
z{NOZJ+|6QmyCfdYQeIA2JkFPN7QI#`c~$NG#4WdaC%@wr<<uNcRR>*bemE>`>~?t@
z*R&>meQV`)v{sjZG68bb>ER(w)~-Ze3=qD)P`2s?aj3gu`0l@0i~;eOkBP@qU!&%`
zI0BCZX)he4XBo_<AA|LC7|Py7p$r!PI`7Xg7C$iJ{9mYYOGB`=3ngz;2xCVDb7r#6
z-tXn#JP^pgR{|+~uAIPt5JKjJkSAW(G(@=yo&2R`6GwMj+Km<7bgu4=SraeZb?$E*
z?!xaq9r0S|z&kq!irzZVFT@FVM`sLc40Q7|P!9ti4j4GH&4u0<oLRQdz_q)s?0YA?
z@ONXF@T5$0-KV}fk0<zYx|J{4N4@A)!;=(Yh=)NtIN@!1jZk(<PulvS_M6Lt1+L2C
zv2Y_a+l~CS9?V}7$d)%osy>Tk@AN1fV-u7=narK$tyL$f9HAb{f3is;yMLncg%jAn
zIhGB2wuc`kFfJj9wrdl3*(#1J$~~BPJ&t*LQD3v$op|!aCv=g<Xx#5{gzt(|_Cp+f
z{)v_UDjJ^|QS>fQ=D>N~(H0oFd@B@}=Aqb*2w{kO2wQdpli4pA8$H{NkwJW?UFw=k
z;)`$j(XqcTck8&(Zixqng)QL+y_C5xzB0m(v9;Z?_HbijP3^w-={Y<Qo*s84TKhy(
z6F2!446N^J;OpxyTs-N*PJdTE{HteMVBpIsuF7cDp7*#1qgMIxaHuf+dH@~&3nXNZ
zyeHD$^sx%$PH!LP3jcon)q`f5F^kN7i3;=M?*czQ4%M9abAbF%q0)4R3SXq-ZDCaY
zl#%1>ftV;}RK)P>jUZN<gb>j=g7bva_?fgtmyB3fm+pLED08J*TF_cry4!xVe(A}o
zdj`C=7#L^c#=XUE(v`b0Jk)~^1HIUB#fwMUSr(iYf4<*~G2^`O&l1<GUFNrYy;+^)
z%k-DNJb2~HVd=A9P4!{+2p{eb5$}?uT}E9WF0}CFUYReCxA@U$q(5oD`{7mYOJc4M
zP4(~lzw&19c2{0##vWV8l?z!0>6!Fd9&F&$YZv97+Dg-Aqpn(8{#b2`{}Nkn)wE%G
zW6h5KcAQGH)8EyOYd<-Vq}^yy6GtW<cVycPXWp8)P!Qov-Wf+uds}04uqIx3Yd@zh
zzva2nA=;YRpIfkEaZ|n)-acJsN^jxq$iAl3Y~P3`H=EL}S~JqUTT=3EOWGc`AnLXS
zf9|!Qa<v76Pna`#kr`J{G-pd!3#Lr7AoMRQrV4Kp##`Zb)|!noY^YY<mJg%sIPsM|
zHJt4!9+Jvh>C+loMlwzslw8fWy?aUz**u0g>lC*3Pi0j_8#V}UV{c?A`z?!$!rOLk
zS=^Lvbj5jfkQsByt}ah}mrQIlAC3PdmnNYdne~nQ8aq0&b8#nLyUG`tnTK;#ah6f?
zL#7wd=~e+I%1xj6XDJWl6Z#+Cwpl!ceS=1`*-?4?569#BX#(xqPUMf76FH$=;e#EO
zEi7+sV4Vpxyf&6}Wkl|ZA5CnCviesluX%MpLIT7G+jb^uS_f`AW^(IJTf&u(cgQZC
zWewA5ot}=}uU+wyXE^2?^;_=FC*WniIwJ}Q%Pe5^@d7seUdW_&$|Z6yX0l_Y`h(;{
zom0-hwaT^ZQ^M&PB_vpM#WYnKC25Q9eWCewYIizS>CVH`>a_l+h&;1mo~|zD3*qhK
z-}L8~7tp_MArFfR#cTIe=BPUAek;fGQaMjPl{4~S1rh(2GpTtw$F>#mLzJ@L*A(+w
zxNknU1Pl9Ob~n;)bXO6b#})FpRVU8pW@^UE#J*TL0^u24`K(-qI$6YA$>Q6m?bTJ0
zOSKm{>^va;SbKnedhc!zS1!J3CL<bV@@KF1^4?36^q0D6gtx&1bgz;|H=#6{(N?W!
zBrT}rmP9)3O~B%E4C~9p^Zpja@ZM3(Js*MPvM8|Ej<qnF!KE?$Q5=W$);P^a+Ho#R
z;=-IrtZPK7TQ`C>o5C<z5YAg7`N<fq^CX6uBZc2rl#MhyhG~A$^gJAiMLYRLlfoFd
zFO(S@LvSn&#`wWV8!s4{Yox6{tH(9REUXj9hM951MaMI$TD<zrVtE-IN4|DoHmB76
z5tGEam6|&>|7I(TeAYzeIs6yT{RHL03U6n(Od-lHl}OLlJU-Z3+HK_t2z%d4kFwA*
zS?@~{i^iy5q-HWD%MzKjIYx7lvf{?a5a|%3jvymbVm06OQVz{$?Wg4H`>&34m`mlA
ze;rJ(`XM9-hH%y~lt-Eg>wOi<;ick^$7+X_7Q(Ul!7M5XrloZ->664^jtt~}X(0W3
z1k&N0xLdCfnt1x*>?5y<{5uO;`?LS94}-3H(@DJ8nkwFOiS**b8#fH;F8tfb8RzQu
zL^gI{#%TvS2Ro5u=PZw<fgaulY8nj~+ZY)AwF}>$ab}S4_MGrGM|d0A*o`0Vx#IJO
zD{+;6{MFBwTe;rkczW{bD^Dz)TuBRd<9;XMwX~oXnr*A!38j3V7q@fVIN8LFjbUyy
zF#6#0i?Vq%!zU((VPPJn?^%AA+yusMkJr8@h8|f-yw%-)RI@~i*C#Mvy!$X|fzIqq
z;9=Ex<w7W@Z+I*hy2r73z5Fs~<H)!fhsiN<_mASSuO&^<tvKF_Z?F*FMn8<@ku<-9
z$3y&kBa0`9L(mR<`kqi8`v_-Fg-}nNLC}pr9_c*_cp1P1t03ie3iF=%)6Cw7>A!kw
z?setXEqCVB7jN*54?R-+=^o%q6%Thh&3C1zp3S!%Tv=0Opzj0&bLJVuaT=rrH?Xv)
zo@rYb?(WrYe6NAtR}Jc^bm2ylD?hGsC1JY<w^H4B-`NGTUCtb;8ps#xg1A>VNWEZz
zR3GW1UI5|pA#o00d$Fv&4}C89QeSxcSa^FZN3*Zy%Hq1dIQH}BnsOE9$rJPEHE|?M
zg5}K(WLrisJH?SC=0(s)^X#8@<YyYF9Ed`F{$>W_e^J@MQGr~2;>*@3Wib40z{bK^
z`VUvyYNt8!i5rF99;~|SL91`{zQ=m$?9#pCCoifu@RE1l3u|e=2IqMTQ+=88r!RA_
z_|mt?m-*B5z6)=g2yY)INL%)$kGdv()&1tnxn=T4NiUXqQ-8-IAC_wVeJZXwFV~G(
zJM?@;i64GtP&bN!1%0g9TW-Z-Gi&-++E8n!jdJU)Y2MnFseNr3GQgU<k!F1Lj~P$>
zo3mn1bB+{RlItLClBFYNlO3@h@66d+F3eiv#F9;p6bNr0+FP@9i>dmm>awl1o93>D
zEPQK1-GxngVc!_nZ%o+}Y|3_nDStOGh0TrFzOV^<3R-Y!PD}p&$$~rLk<ae7AbE)e
zb`P7Wx5A9_EzS8n%7T)mmek#5#fC~N`6#WV2ei^zY^A-QHERdjVfezHjaK&Rl$HkV
zb_5mK5!?xg<i)%gMiPr<gBZ3pjiIc50?kVj7@~RQ{GlYC7*n}ZCmpQSzxb*>XO1YJ
z{kr_RzvmFVMjaQg+wv~49Y?pRN7$kxEqZk1{Gv{{I_EL5Z65W7x9^0vZ-lqW!rMPj
z72@%?bU9(-lYi?+%bX!REg8yg^WltMJOY>XBPk!EEMa9r-D^LQ_#kCNot(hZHRI`@
zF_z6gj-j)>K1;Nloqv2FYhU!^x>Et!2lM$Numks>NH=OFZ$?r&j<)I4shZ9|W78Rb
zrYj%U72!Ttc?T==xw12#-e>Z;*sg$s-xttFcpF*03-vUM*WFe~gSth^|12hZX0h%{
z@)%2NWM<Ztb-TJUZjol#SIWfKZgQ1(2kbW#@md<Ab%rA0j{42^bs=I`KBiYY^RQtd
zISteYyRoM{5#?NPDVOJ^TwP=3*#A($Sk2KZ%*r_=y!|<%3x0*gxE(1Wb503o?ew*U
zw_oopV(e&jyd-sEWl|=lr`qxKnzHg1sr#~OraG+E_4Z2^-#=+j+m{^(Q2uP6cG=uC
zYtJE*EM7!NGaB8FNAixEg=&X7F9UCRHy-IdT~;%L$id=yQ<M34G?6;Zly{RK&%_b2
zY`QJqp!6{>`bQIcRr(mma9S>ru61=7XDgJm)*~8g%_V~-#p9rCy=k44n=me#5b0Gq
z4Aai&k`KcMdGlzP?#q3|NlDK+ezs;PX$iN<7h~KM$!m`YM#|e>*C&i=Ixl}%AHoyu
z*_J&q^2AO$+H50%OSI$c8%mFbp|ssA43n0pdln4NjHP*$a>xtfbx+U^uxS#*5_G0V
zD92{MbhP7@y_6Bp$%eXL{+_@-%~DPqQ<xts?Xzat+sf=I`AMDAebj?_E`hTX#iPr^
znSDa~ti_2q&y3+syBJ(r#YpEDEx(I)vc06|tqj89V<6{#l!u{O5cB5;5hlFtD7+mR
zENs%g^ZMBkF2#r7VjjX}qjs~&A@~Sy=l&0GO@rB@?3&bpfjG4b#N&prZnY0XCrPL2
zsNI(^H*2gPB@Wsl{_4Z_re5ry;mwVYUVQw{lc;ZW*0-^v|7}~`THD~$zyZHg4$KO4
z((cGf_f9AMT)WUP)WB(H11^JIsHWX+k2_8TymO-HvlCaUI}<a=z>im53DY_F_>?;l
zwd8SG;!cwvG;=g?Wp}W!Hp4(fx&hNLc@E8Vo(Z>i*~!P>$c>l!^{eyVR9^8WvW`)l
zlKdyZQS=`sE$kBA&$I_fdLB);`_W{qQx8j-{KOG)OzNfCbVHOn$MqS~&e!3WXx1uU
z!F7nTa&x13vp$yA7vgZsil_XScz$RYt@~#L9e$GLs6{wOYv}W063%0Bs858q7o_|8
zdq^nDrJ=oN2qmtucv#(C-klF*P*EWJSNU`KY#_@!25?e1_teuzy)NPlEM4Wz*7@)1
zO5;dZY~9>gw#p3;H1|$$C9|a~|Ju9qWwrtHV=fHVEZbzEfokIo49sw$Uu$PpmphXf
ztmhl<LcjtSDpwgezQ~pBAEbf(=*+o~PVC(7jIVIj^k)w)e)J&L-IKn3yr}ZYgS|!4
zUoCONc9c865Ar4Mi6@@FxU+h=HT$nvF?Wm=m%p@PhnKtbu<lee^T6YbG{KFOYc(T~
z2X#U@-syiDyHFZmF|xj|QC%>hY&zz}sIwmYd`{k_(;hTR_h4YOGe7D3nU~{&%QF{x
zh-Zzla%ZS^r^(r#JZ~?Tb&)3_vpxAvoOlo6tzUO9`BS}kW9H51f4wDW_hCXOU*1Rf
z(nI`k*dQM^g@{vr?@h4g*SDX&8MWAtTKd`lyj2+!1wP6w@FAtUBL*)^hFV+F%)$~!
zXA4p?&1u-lnnDvR-qy4xJjaF{2OAbFx2A!;IVMpp`R`B@j=GxBbCVgDgtyOjHK)%T
z8#cd}4@Y<#JJFHBi=?kAcHqwy(hpCxW2s|HP8Bz&^~ZYhcGu;<h4q+otR9c<n=s*6
z15B%!@^e>Hp7@($_G1GaUe(9$T78aJHQ~Q36DseTP-kRALf)A&;Lpaw%O*V8(UjA(
z%xLJ=oD*s0?5Jv?J(UHe5478DWrgva6?xj{nh&>Pz$QDst7gxCAME)2E`qMY+xF6`
zeP<94A-w%JFjko`Q7m{HgIjg^!|KJeL>XW1U6VMqQog_q%89w0f%Q9OW^T-8gl62<
zzBx3VlR^CDbjo#(Z&{&El*ZyaE0vp`*NGFSm02E@$H~um9FEiux3GZ2H`N6brXK5k
z@_-)d!lje_DC|Cj7KKB}t2dnf9Y?6!bhJ9W$J4WTqWs?KJ8v|RkRzIJ$Bt*tu(5pq
z<p5mg4Z}nDeD+Iei`}#*&CBKXha65-Z%gaYw)AM$mcPrB`KpO>6#muzaA-PN58Cry
znXJbS<&r-ypQGFJ$y<@n-t>G9%qYO(uL27GE1;P}Aumqn@k?MXbwYEom{zPMS~0^`
zm0;PVE3I9-Vn4r_Ar3|ClUD3Umkt=JN>?N-f1wVR_n+ipY1WykmpkEcMV(?V3%J*(
z3w4(%4?;QqZLP|gTUt)M_O>f~R47MBxO+gn^pB+&EW5C-k2=ZrtM^45@68hVo&7t|
zNBnT7chUltt83ht$@JarNK0(XXldVkj%P4vdnUh0EA+>g$_szifoSb%|C1NhOnCcU
z{VeX0K~3T9ACHyQ`)ykSZ?(o*^X{LP>PTs-9GlMSa8635@1aBjgVZT0yqzJuJ##(g
z|9S9s#75(PJrWZ;Wq>aZWAp|2SFXyNDbBe-__SnVJXbW&9MJh#@hF@Xb-i^5^Wu{<
zhw+U(*yQZazTVzk+2D=IJYT{$1~T}bbo-@|4E-luS#sg{%DYf5J?UQyL&$bgZmGO;
z<rCzE+!eyFy_Aus{bx(@wEHvzj~3o`&V(`M@=<INw!b#2uT);Wq(rs`CNfWWdv06|
z7HKhzI}}U5o7#10mdg7&kwf}B{%fr+mMX2V%#golVJb@k6Yv`&-=Xl}g1r9sl}WHI
zBbK6`F)Ry@VQ8ds6?aB)<%qgRR)>@PM!aO#K<;f3FL+IRs>cE9u?*0jBY@yT0Sx+9
z9N*y(>W3?nXJaroKg(Ye7s4kK<+M5li9-+Kj&wWwq(=&>=EnhfDpvmJ!{uCg<rAb|
z$?{^zWG~F_c`L76{`!TU#E<t@->WCie|ivl*pWWX?XdsTmX--NRQzMlAAKD7(%*?X
z^PDK1;zasP7cK`#H`UjLQwMbaU+%0NVkedgZ=-~_D}}eg!rMQ8bJhOZh5N&#`O#d{
zRPR;HYFDzWxe_A0-5Tb?H*=hn1K~(O9VZM0&OHCgReBUx?pV9gZ-6-En;smyq<cVI
z7~_JXc&xivZeQ*C_bYq;wsLLPDkDpolrL?gIcAXGrB<|ZJfhU|E4-~5MN?C0jTLOr
zcCNIsInk^)kD-dR1F13U4Sz13+O-(v+(cltOS^pG?SIvjnIO!~s1wG`5k}n0LfMlS
zig{Kj3xqwFmxr)U`qbRx>hY2PwN7V$BE;J_-QdrcrqUkYbmP-FH!R*57%sfs>F&z2
z2=VN~+0veFJgVVF)?`=Kbun<?-ayJZXSQf|Z7IC1BD~G3H0VC>LQ%3a-SV879pJ)O
zdQKhn%$o~u!w0z1P<UH)uCv|;XJ!OAGJC&=`mj8xb3wYQUh)?Sb1R?9S0t`sR;&}X
zesko-SX+L4Zp~s>Yij81j=5+>uMt*Ex@yUtC<}59+B2q(2P=Q};mwl(c|G-W^f^R)
zt?r_WmGiJRlm}k~v-r6e*9*Parmxj)mj|uC^PpiH?U$pRHS0T3p6$Y@cP?DhPWo00
zY2BxL@JhV4g}CLL`n|I(J?Z_*le2+d9O~i4$va+L5Z<mE?#-ul!rXA>YpoE+qI=po
z?Sls<yHcc?dEZlS{`tq7t~30o@zjTmZyQtVkH-A<#z8#|j#P8AWYbzRcD6I&M!X5r
z0!(<g*^E1fEEw>_iVy8<sA{kK;|y#0kt{g5wW)H$o6!4y6XmQop~bpJ^2RnL=&cQ<
zbL?1P;mEE;N19D=z`l_%_d9!vKH93+)12bM=Ja=M#Pi?lP%6Al-c(OJ026#yG~m(a
zh8!(2CELxEgn<p1^rAjjeyh)}&-IB)H&Iud3FAjJB<+o6U$e$csNR%0+nXv=+YE1g
zzZ3IX@aA<3V#GD~xog3pWtP16v?A@T71kA2l&rNQ@uVH&U)j-0IUm&@g!8pY1QxHt
zF?W?O>|q3Neu*L|J&KQ2qN)BQT4wiHZVr&o-$VZU+vzm7N#`5KbooxwxxFAA`_<{3
zKbB4n`R@LgPv@JKN^Goj&HdCHm81^Nvz;i4$YZ7OHdA<ex}bnMYqZZiR)E+1eB8Em
z;j&L99u-5F+-0b|`oom(t*rfj^lC1>am$IkuQrj%Hzp`oZUQE;+TT_VqRpB<c(qg>
zq4Y!jx2h+sY9S-c<nR2pgEEBD8E4mqFTYJ8zgsf3w<OWveHzy{WhxUqi^sRy^K^GE
zr3=KFXNqs$sTuK8j<UVg#USiD_NcA$Ydi4e!wxKIoXZs7T%BvVd^NO)q4U%$Q@<-m
z&z7)rMlp`UThmbap$6wLYi<|5Us<GmOs;nLxdd*^V}}0z#{J3<6yD~$bRqb73AM~h
znPFZo?NvGbrRA<!Qo*Cn<(LU?ZEloubZH?!Wal&JU?+~{bs#@kyK>?ADW?t$Ynel3
zZWg8DkKKQ0r~D4_<cHhxS$guvYciP?m&LAX(&=WaA3MH18^W`=A)a|%MFuSblxw2B
zZ0T3&tbf{C9UR(MPD*9KfK+(^m1o;pxw-Mll<iU9N>g=2%JaHvsm?o{bI#%eqhq3Z
zuKCtoT~#LX^EDCPc0MP})!wqOUJy@f2Qp&34=2jKsdveXEvvj(QmR~lp6(=<xRc$@
zmFt!U65e<!J6YO^6@i%D3MOcektSP0N&O;>Tr+gv*L?e32;W!+W4kzrl2OtU_0*hb
zAHu+4K`a(0xO7GkT@!-I(hM2%IELO6jr{uDsLpM8;ut0^S2$glNh6#r9y}&Sd8V<v
z6W%T!mq3|IBKPA%8EB*VZMT68I|F{=Tc-si@NKVH`s>xCIF@Te#rL#|<zM|75r?G9
zSrf+7xl!!Z4sQ1SKw9d)Ft<%0x5ovjThgD4R|5z?9KhoB0p#ot;95oqMY<>CNUw8p
zP%vAiKbp5En03PG6OPKLnH5B98+mA@8!bL5-{uP+W(jZeay-cu-ewDLowVC4l(*!c
zgPzi`da_GrUH#skSnTy+MW`*4bdPTN&Wb7FHq^7RVaZc_WxzPeU*p7(BqxeZ#k0M0
zVE$kS)(z17s~Nb#aC^>4!_j1vJxMR^*(Izg+Ulx|G0i)924(6PxO`Z%Rt;A|pBp&b
z!-a#j{y*NHy6AxWAC7FD?94u+E5lC;r+yVq&2?kfbq{tQ@nqy>&4<-InEu>DT7z)z
zWNT-?TiF`o-<cIm|JLF=M;g_?A#L+|?I@(TIi76J1LgHCcGmp5Fp}Esqqx~32E&3_
zb@axQ_$Zdn>KC|3guI>Mym=YMmEYlxcD7Rkjd*-#BuH5ME>e25%upr@d;F?~Xf_R^
z#y;gIOw!KLOdb~PQk!S^sq0O-dz$xN^m3zTO9SbrweQnju(qQ;jTYGBc-x+ZFAU5T
z-mdl+A3xifx#BC2t#p(>U3&0NPCVM+jCrC9;~bpvOLwN>EN8xaq<OZz3nP{rNa`hB
z)h8#O<vJ---jVCw9q9Ddp4tm+sae&D96jfntz0?vhxA&T#jV~`9+8hTUHaQH=$191
zU94&JuN6h-t=QVzifU~vxn{Ir-(X83$J=SAY^Ur62Y$P$jH9>CWKVISFxHK2y+Y~z
zEP&jnUi7=~$e??UwA5$fRDUORojLLJxRZ7^j-0>ZKo@UE?$&o^X_Nts9{i>`J1)qB
z)56<9W}eEESGI@V(+9dYZ$9TmNmFl9OT6h`CT>~!>$Ae!n(f^f;pC<)G*|t1SDN-P
z(8Sh_GX^)-5A{|qW@C2WZOn&fjrqs27WWs_AbepBs#w=$L2z9bJ*`K^Ml({+SdjXg
zW_e$0+GJR;_GEMJENaf>Rn4?#Z^D5djXC<c5v!&&BKWoqA$m^Xlk6DN(2<=Hj_M$D
zpt<n2#VUJgNbHm$X^nMS6I|EUQKn=atiGr#p0*x;pRCXBVI~|dHKBDE6Lrg&u=Ztr
zW<IOWPxI<9XLlVu(@ZE5-tL&xkRhW?iFskl*so2od)kE1kIi^9)Qsv&%qTK#&JgL9
zU(dB*c6Un-yI2u9-HOH?tZ4Vik_+ip>MXTm=CBB55{1+EFWnXAg;UE#9_hUi_-%{C
zJV_aJ|3ryLjFuNoJYr-V_fxfd@kwW~wXjTh+gNz}rc*i})~KuVP&$uv?z2W+C^4xt
zzMaZ2lXQMdR8Pq-%A>m2foexOaQjpT%7%61tKU20yHvUIwL7!cp_F6iy7B#A<<v+S
zN_+DWY?PnJcliiX!bg$cM0l%R?!Yw@So8G+!qX?PY05bIG#!iM=MuhpR)Cvkm{84D
za}R4~sn?3VH`^#byEX5orO+uMnLVqM`1x}pEf*x>ZjvZnMFQ&%B;YYWmEW}A-F~nw
zPxb%qo7IY!d(*I;uTJo{;>Im=xZWp+?l*H7-XxdhZL_(0SbJeZrsmIVDuuTdvxJQ`
zI`Tzj4(8*Od1<PgBkx?Ue4ow!rP*}apUs}Oo$-58fT?pAmK`qPTj6c#knUtX?9QAO
zWzxSDvbu?~1^x5sx-nl_vW0wqzat45(hS>W@$f-A_8e`C>6vs+ZBV9Rb{2z<X7F=r
zTc$Nuf7_#U`Yp{MU*4@_QPMwY1~zTn9?#nCsUMW3jL<CRHB#SOc3UQWt<3qHGz$N2
zO}n|8ZzrUZsa?HKRb|>nrH~n=?&l_n+;fS=W?Kx8m20rQk2FG_%03N>#^h)uTkD5Y
zIw6c<6XC)OBa7M@DPI*LPra|QhJ83!q#ffXX)h*u5-xA}<4+DW4Yp@zj6G8qIdko(
zJ9XlHl}#PMdu6_@yCJ{2{OgJHf|WTCN?B4UtJa2SzLW<vG*F#Qemsp1pwZC)?yd@;
zwJ<J2d(YhHV7{Fif_noa-{|~4KhDT^E#z|*-sav3XQJ?Ssq{3XUBxLMh~@Nc`S<%J
z@XsD)xE_d>o-vesGo8)V41D^|i*f1VxxbSpsiu54^81gFmZ4%}IB`Eks6R^I%VYIy
zg-1#+CH>F?<=7t<kC&v`R6Ewvxzd7c6c>EUhg#bGZQmKd-hn|xJqRZ5d@%Pag4z9_
z_PvJ!Idnr==}m)ZFi7|#yglOT&wzc(41A{fcC(K<VU!nG=m|4CS?A)#v#TCVZRbIo
zp`QHkz@5AHZfss*Pep%g8f$0NNO-%Vz74B>v*zY~`TvZzbX{Z1rt7xst?fv@=879e
z2kQJ~&-=diOj&HF9!q=mCfT#~g*{)s*1q?5XL9pg@YUU8(K=~>lm}2pc=LL_G%Q_R
zaJVKO`fDeaX+L~&lq065PLw=xA$6tZ;e)Q+m3G%pGul^64Wtzcqd&Ow!%GkC?Uhk$
ztB#kx(5PDw-z-)JQ-2@E8GK0iAfMOY!Cd~^R~iFz4(R@KLb%gK{;867QJfUs4qC3e
zRMSXOT1f*S%pM&Q!Q9Q^G`$l>UORQUOBdwlYve?jFvd276k{k4{|TXeiT3Arf;sq$
zH&5mHSub7GhckLD@S|J0AIB?wIR2+WGqN!Ckc%?%9O;{E$213f4tKI={Ty3Lg-4BC
zlv~i$fTOoF_1`<vc7-EFrH(8XetrJlnV|=ru?%w}GR_&FUe3hlJCm%Lx9=naqjELt
z?Q!B`Q*pNk9mO{~s6W+Sx(pi*n%ZJB*-kzX7hcI%x=ozp%eQvQ5VhlE4_nQg);!DC
z?(wA+N5xSZE3K4mYe7JD3kG#G<Ch-InB;H4n2q*K{=tERGh8s&93COf@qm8}ES%*`
zF-{CR>&TG1j`aHI$l)$dxQ^Ez;h2-O^o}$+>p)^dVeeZfP6s<<R>#26aCeTI$wRhE
z-eTeHLCxT?hvZp}@#6cRz4*GGH#LN}Te=H(oxRxV>CWZ|1IhZnqr=6oNIS5`(ZGu5
zPW&d`;3soo?%0MT?5xYZUDau4^%XyTTSL7cHCWlM2GwJ0(!6OcmQJis#8MOX7#oq$
zzYz^5H<EU{5iw^P(e?L6WZY|nlUrlW=#6k0*hpthBMib@40f0XThnEd6@9n1WT)oM
z;Mf+N+S7!1amb^$*OJ${7Sn_4kou{P=9YR)*ifJB#r65}oBHf8Fu}ir3A==~gI3n(
z#f930FRR1B6cb!Fny`CL12Tr0(qgQs_PM5bcr>M;iWv$2!`q(C`SMc>UUoLee~tx{
zg|}zzt>mS$qRu-@PI)$$@3lE;`&-iACW6hk!>QD{lsz|G{*Ewg)<tk+hcZ{<qv-HX
zo%O<7|KTw-)2=1xizJ@kQWn^zIL)7N+!Nn6ZE!5FebpcFMw-E%ZP>F=*>s+%(zK`2
zbG$km9oncnu`MUUJFxsv2U3@d$KB9gXNmeT7j$CC2j%;XE#}*~rOIn9#}GPHS>i(o
z?LL$<D~5BTmV7=BM=`hKSez$LpyuoeG)<ep1OMS1=~+ZzdV3}d1O6GTowM+^Y+NFj
z4<<8kqI#+(wPxvs6jnwi^LkPee?Lm3YLfaL>m=eQyj}gTyq-N`c(zJD^Rvn=72Yn`
z(Td}j#H+TK|K7eG=GAgISeQeciS*ZRI&iObHamp32X1HZ{jdxUy0_z=MHV&M=WxIy
zM|!jz8rKkyJ5G7)3zTUuygfA~o0CqRS^r5|EcY&?+%4hjZe6M2(VcT+x^rrBR~+XR
zvfES|GU4sW)%nzrj%)hz&Qwd4uXJ)I1O94<Rb^Xcm8-LET^g4)pISPkQ!TnJhc~og
zc%4*k)k)#2lys7Aw<Gdd7VS4@5qq#5B?H^>>m7C1e3`-CUTw+gmd^eKZRqiXc9-o^
znK3REZ{h7E=M+BXN+&f&`4jaL^)nd5V&T#v;cdszF!iCR-zq(VvO~h#pTpRmp^Tmh
z^6LC;l*Zb~Tj`swO!Vc1@V4}&7vrzXqi*BCc-^h5&TwGjCu#Ug?QqqcaM<1h=j&cP
zy(^D|b~o)@EU4q<z=pZf_3jSFJu#Hp!=)SgSzTBUgs;kk?`##Ic`krsnkB0s3?8C9
z_(=IfHTxb`CfpJ6i6@1(!!_r~w#Uh4`b<lo)lpn!!~6&yE>vcQW~Nh}<#&|U`cj~>
zT!ptqZ$r3n-I*v4SB|B4bNd@_2L9{A`DU>kX{Nh|dpLc*i{P105LdfOhdEa>z~8QX
zbK8wM-Lw}=@suV}`)TPkFRk?<Q+T_?S{@Vat&PH4+oQph?+s>_@OJeJ?P?nZaHL}(
z7Bj^|dj#^^G<gXP{y6@iOob0#yvkKZg7h;dI(t%2csuT2?e$XB@iWt1T@;=?eB#c2
zgYNQO?6B=;&9DDh;$&gX`RX>@IbzM`Yc@3Tw<TA2JNkkxSE|`D`l2oEf0X8Ppe-TV
zi_NUC<m62Y{<E=Qx3@iCYu<S>%$b+kMP6-Tz-qIBN>5i>j4|+kBz*-`lxy2Hox{+~
z3@}5tgh+F5x>G>`ySux)y92wsu>-riJ$7Ir5+W(Wzu*5`XR$a71!o>+o_$~Y+A(kw
z=l)I}xS`>Wp5&85^IcGT)P*yJ8~&ub!(=-5yXUF9`Rj%KRbDu~k$6pi%=s;D%(ZjH
zl2akLt_h4FRkkBHgt;=D;|KX8+QT0+e@Akb4#j7e01Q82gDQJ3jM__2Ydn1~N!<Oa
z)9*q(-k=ZR$fsV;(j^>$<l`KMMPmN_2y`35Jmh<Va~3sqf0^q?e^pvD?r!VCpijM;
zwz?mTnv(<FPYv*Pe?)cg#-??i@J_Nr{~{GKPupXjKd*};bN6nz;#IU0I_z*nKYBmr
za#pH6=K@#a?YagJY}I#1-Y;hJ4I$>b@wK|Z3Gv*|`NlZ2SIQL~z1*nfa)ZkSH>gE>
zV8AGzRl>dC)!P-{GhHwx*#(b0dHrP$7;?b@$B4K6xnm6A*Y(w%P+9B5er6|p<9^_D
z2S+@+ZI3%$?79D-2b7OD`I`!pUn?<JQ;Ef~a(1!FkUhr=(uGzS?`4V3<_^f-?10<W
z^d<3}<hR8gRvvD+EL<_9gEQLgbHbRRPWTk>gq#H~_%+srnW8RG`{|0Y)J{bbZ<EwL
zkuUKeC&2yuf6maE=}e8UGam6ZH6Y6Ye+E!{>*Iv#l};$OB(@efqo0!tw4*&ysB*%Q
z7+yc66NX9L(BhRNoL$ZFyqg(<vgo-vA;Cc6?XD|YXdSGBOB%YE<fe=MdvuZKu7|yQ
zh<8WGEk~H3USWbmaVF&2Oi;eo1Ve9{W91@q_>MBi$F=5YrEd;BOLHu1G)MdwJAC8E
z?KO5th_c1?T5EWoRN$pjK@P?Q(?azT)I}ejazlDO456<y!kGb5gc5K6u9l)<oD|9Y
zeCL-G586pF`GEn%0z;${Z=dWoW-f*)K20>k-|1%fZy9;zzvejiPR8CSIga<FZmYo>
zCojvPQLaG8JQd9J?U18whY6J`e3~PNa=HvHb7Yua6$x21?<K_Ceif8KeWQ+YM-*)0
zqESCO8ktXHaLF&82InNGo+ZJ$F&Vq2QL}lBJyN3+5%D3O-hAp*xVzh0oq)5{mMnb5
zY}s+CSWdhxn3)NCehjAfE;p6=+4MPHZQl|x-&*7So*WcS=!EpBPMAy_d)%%U^v5u7
zDQygTyH3Esl@n33V=|Hr*`MBiD#Z2~_#QJ8+T7`;YmCPV%XDltVCKPGYFA3xhuAp<
z{Z}#1oVw~PV(hS1E!owWiJsTeF)ln6rNij8=VNb64Nc*X82FE)Mq^Se@~Nfoe~&xL
zOG)_HgB_X7Eb5@jfZCHZls#dt5Wn{UHLEWFWgu>2CiWJjA)L?eS!psh6t_lIv(^~?
zHXAe6X5lG)H?N1cLz~6zpzhlS&*-_CF)oL_W^I{gPCfD7F1Qxj6^rilhD~f=+_378
zrK9@uj85;YUpFMwc12RNu2632iXkVvBK~|1yqSG}yreazE@PjWS2o71%EH|Zt#F-t
zSH~~BrarvRUM(^IdOG$u(!<jwi|4c~c!e|jes3oE-%Rw&WMAfIo_%g-GM_OM#%tMo
zw}QUcEg3MQUMTT-I@XM27YY5QZCa<{7_a4^p41tgjKR$fkr?A1fmO_5wI|-{y^ds0
z1+!{8u@8=YSh9p5EEpZgzO?|nDQ$tlIn;Ta<9vJ34=wqAeNDX0_(<-Jcq<y5(3tB8
z*#|qG>6O@*YeU_dEyicq;*OdUJ57{`$hAkmNuF4<-v?Q3gOPMN1h-!V@r)IKO6wpT
z+E1PC;TEXvAB+w|LzrP0f~HYnXn0HCM?o0Ap9n*wQsBJ``11{jO^U<<o+Za5MInU#
zeO22?mWJ?NTpxe|C;Z`Y+z;*>eDEmW4TTDOd>Uhi&}2J|KWdM?3ti!RGz60~n70@k
zg>`=9uigA{d6^HAu5bn;-u5Hj8jSWu@5R>8tW}`R78|r!=YZdw7aQqWv>nOp#DDxu
zjo1xl@8oF(;2wRn>I?jk%N=PA@%AwB*4{A?Wrv%ippN_6_T1aX`{J>gFZ4$EKzqCo
zcDebWU9vZN$vhFd%^hc{i3)GWdvt$$SYA<K96b+_P4=*EWslpsN>tpjM)d(}EH;wk
z_6j-TPdVbsAxD_^am0!TN_byW;>l?xv#NNX*lCT(GHYB-=WI-^^Qtg<Nk@6$;5H9b
zC6QYm>;b6{dFEXXh_rP;H{So}EOJ2zXXuH<?^T+-hZ7_1OTDmks2BcGPn39?-19{@
z?0f4<Jx>rOkppm<NiHCpe%%Yy1b6kvD_+aOUCr@oi7oQ~DcJL-fRC3P_sV&HYUjW+
zuP>5Y@^xgwT;lo6r{~_Fmt{Cw5u@GvN8<J?<}VO$&kqg2nH=s3Zw6qKM-XOke(rbE
z53RWGwy5D;cB&c5dfB5JHLh=3d1DFj?;7ugXZRus-DHn7%^WfCh7+1Ea^|(8&t$11
zil@=@Y3#^OMhEtG^RYC!AdWllXWv~Be9Hw{&0NqMPRtmi2b6d_(bg3zKW6EEbiqP5
zS1df|jvE8q;kUw>nSstYLcHCY<bpsKW)oRBV1l_VZU@^yFNpiaR&umzBS(R@HDalU
zN{wbFAn)0;x&NEo#vVSW?J$iq(}W2sY$;YU=U9njQF7`=WT*(Vg0+?vT#2`V%vp%r
z<bXAD4{RxO<?GZHf;_}=FIVWib;iXFPUu0r4b$gi<{9iG`KB=b&M*7g!<4i16Ez2P
z>h6G|<!*SVaKo~F&h#KVU|2Uh+~`1V)JuiJ0`e)|PFTjBX-6|>^k2cV!eS5HeC&ap
zYRq>a-d?pP7GHM6{c0IfsWZB`%K*8n4A64F1ilM(5fe+iwbRBe;%)Xk9n{Xz#W1BF
zW?a>0|EU4C*%%-@z<}Kl2C(mK2+OX<u&6LW-d9uT4mL+^r5Q}L&6&k!j<tcN_)9&L
zeW)?E73<?%iatXA>A)vImw5!bxHDc4vu5kzbD%!64Gpog%n&_vsh4Ul#Q}FI3f!a^
zFkA{x;%#=I0StfX;pR47=CkNv=4>78NYur*Z@T!Ap^H_-+pVol(JsfBJ|QV?Z!yB-
z9txbWv&K+j@9i=LRxY=J+8P^7eQSf@JYxSR8RDnOV3I0B8MP(5^TTm)e;77~(TD#&
z9G=S=q2UpY?}N$TXvd&)Y%EM%l8|+dofS^WeE(3xzb%n_fJB^m7LNxy@hGuLfT3|B
zhVdTV>nZcrYErmwVO~Zt`N<K~yY^>KMMNf^JR@EaZ<9{9#z=CXLnCu>G^i80Covr0
z5<5hRAKCIu@@<*SJZDeJtt?pQw8f67DR^^l24+Rg#6m6RAx-K4>4`Mt_DsX6SL6l%
zi@~TaDR5oDEM0jzJRh)gdweGDv}6C2BpH=H?3t1$<AgQO0`bYn{T74cqhiq4haHR5
zz?}R_J<U7j0j05b#4-hqm8lqUIt@i1QgJOd6{nXc!vA>&>}O;`c9dOMV^X2}oBha+
z$>_E}71mAZ2uR7o!*zL>e7_Awg|>#WMH_@wwt>lb_P3?9g&Ohp!<J6i;L-taE_6qa
z;2wO>_2(|0Sxu6zK(lV>a+e)7imue>uyg5lJ9JEFgB$cfdu*f+c33M6GU48QQ5KAr
zx58Co=$Y2c!80LdT4rFuB<AW*O~GRNV20etL_%0g{63zExxtw@=+8dNahZ5|Efdq8
zXW}7ox6Q>299)(Gd3`#bQK$Q2Pa2+nO+_n73VP<#n^BmEzV!B)o{YqGIeV|PfVsW6
zC;dvT$ryH9RRT|^hQaoI5MtX0;{K@sSdH?BFr~J}-UCUTgC_6gjOFf%2h&}#bDIl}
zPjJGvsSemxMt^v*EqZ>jhKWj!cRggVwvplO1R35XGoR%J_gy2K!C;>+8Xo%Mpo<^I
zjtoGjkO17COFikf7Q8;paEl~YYa4<*t1!%d%A5%D&)R*s^Q{br8+W*17S0Ri4fUrk
z;n>n3WdC4RL25IY4)KH2=@tl@=8tYiskbDTH#pxJbG7VIN4#|mwZrBE_NeCUQrF)d
zo%A?Qe(+;&Eb&mwAIbjAj_GKM=N*l(sI8Petmf<!R-n960s9~u@}~~iH{A{Q1BkOj
zo1>68SNPlo-RqgVV-bL%cYH9D8YrDpekk1Uhd%s#MU5Qmk><#*@xz9W#N4Mo^lVW_
zGt38Xy8Ga-i4T@0d7}k+P0b$e_{CY{J9m5ID{ZlJr3$ulRLr$kF;iNJ*)_I!K+lAl
zj~u%ta%ft}F?x#vJ&P4sF~k~=Ct73O0U1U<mf>ShIb^LBIL)tr<qq<KF?}t59{AkP
zgZ%*>*v)(C#R1gf5|3`>a=zJR564L=m;|YC*TNA-<nxYxam2l8PEf0J!chgY+820W
z&@SGC_qgHoBv(XpcD|iLo-)K6S8V*Sl-Ye_sinEQ*$1nCIU|AGZ4&WT+em@MN;#~u
zWstRXz{gx)Y)hj*r<Fg?K%qFeojJJ7x->Bk#JWmqogTHos5|Us=oEkr#M?a?%)MR5
zTnXZ>ZUZ$!IX3W`WsNaYRaiUL4(oE2xZT4V%ju=e)V4*yD;supbIyIj-M5@vfwL3B
zsq@ZR=!lvqM~GSnbR^yy^IEHOcAKtpL)t^`ZZA5cV=pHh3Uz{HsT20}W}X3e<enz3
z(0}*8xu3hlS?)M~-x=dqI^$fLGt8))*?!Rxno>ucoMDd`;;pHH4YLR3m>ev}T1y!u
zo-zbyDA^UloCQ5QI1}UYJ*XF|v%^$A&g?X5ab!xo%#$&zQidzWGL+j}!D^%>)U7S?
zxy%wBw%d|l<lnvJiWUo8F?oV3mQqWkPp)aedM7O5EU%a2fZ<J?t?T&vBi<(Z*`xS5
z=VzWbRU7Of-(Uy-06S{+Rp{jBh^~A;oFEtQp_3dF&zoX|i3F2w7(ipZ6ni(DBmcHJ
zY~nrOOMc}qA9swD-<M(wBVmncr%h4Z-T>3W4PY=zkGzI93{16PvQ`TV+GxX^csqRx
zXWU=9*h##-K)l_hGQg}B2B?~10G&WX822(J&u)UJZ%tv-+Z>8hW-wZ3Mh}2FEH;>8
z;RR##F)_r67dja7ia7aH2Va`&BAi_Hx$e4{a8(yNA$mBo$N&XD4bkPI1no~ta5B&c
zXImRVHO>f6h8U6CF~X#GdYp-gtq*jlXV!toL>+`Gb=jY#iyMwoEOatO2N3@`8^4`u
zgsobQ;_8_OadjDg7e?A>e@+uKC+m|pvPG<`E%XoZ>r-Ty7AwQ`9|4#!oEl2vZOMx;
z>^~Wf75)PL`r-86M!}4Do5?w_a&R#7#o60%IGCX&q0n7Q9o$9oIQi6ARk7D|Yyxm1
z5h-1moA)dkNq<sM-;@HmZYo+5Z%+_!e_vtma8?GC%`!0dTL$OBHt=A+LQGW~oUPA8
zzwB&GB;L+(&7e+(*#q-4QSy-5t6FB*pHGIy`$TBa4;ks-21{?JljBT+dh=*x{fvSO
z=kOlAQt)R=D(Y>ye<b$!3Fh7sYfGExWztN-=9lz%?4-XwF&3xhMPu2dXlOo;!X`-^
zx_^p8AD$fyh9@9&7(1zmx2x8rVG8%VJ1vs%HHqH+hp|xm#ylgfOw3xIfp&e95N)4~
zD|?umuS&<qaT%zkCOGQ>bEi4;UL@YyKWT&Rlh`rL9qNgjxhPwjgOVb4W;b+Xc1~Bg
z1@^$$QC(1)(gpWR*`ZFnH9ye>+v%Me`LR8AO`)fDN-G?ALw<ZnE6h@`<HRco^Cq&t
z%a`22PUhuHGO)}v1Fl2U5y`pqf7eJqlbs*L(bG2UPz~T2$DU`N;h7jfyiMVmXC(2s
z`MC_7TbO~ipVHxLo{sN3(lG5yDmof)XOO}Em7R(B>=y&go9qkFh(MG-V0({qVHSO&
z%>156KBSoUzl)qb_qPhfPx{=)ai@6C+KU_$_p_el8Oh7_)aE&Rq$|E?^Q>&)MDD`@
zB_&GC`$lcdD{C~qmtk|R4C*qT1INiQ`IQVE?C3S8w>`dvGjhnCde``(z6bBAL7Zcl
z7gs<&a>~mf<V6Rgl6tGNmZ6w3gjt8nxZ5RPYqT{4I;o+!dJE`BynX*B0vE|eXPouI
zz&tN3T;YjRUCDh?>+vAc1MAH^aNw^K=4d&xA5n#ofp*xGrC|TLB~&|AI6lxBigs@B
z8|{ht)R=0%Qo{YN1S_~3D$0}MbEPrd-&i0bNx}Pr4Z1q0kbFXg?@v{D<3a5QufxSY
zoQXJ>rZg}Q@TLzY=CiYBwGSNj(I3SbJD8YTzJ!@St2vVqZ?g|_SKHGYmo2>Uj{33}
z+CFF-N4y&73TtN<Tsh-_nvV9^dD#{#Pb%@lLWvcpY!OZziJz-Lti1xwh_~Gg<@CwO
zkhf5QK5gZwu93sKmo;wWb2hpy!>f+O<k1S$H2c5ofEAh^c<Ai`AL6a%77uio<AH<*
zH+bxK!pUR@?kerjr6v6vzA89fQsTvDB??wKV(}Cw6#jF<j-k%jHlBO>Vkfj!I3xMH
z3;G{(LsY6KX4A)#q3?%#^l3)TY>w?;e4yt@UX~p0&Sr9`pOZtYOolBUGVJxDo_ZsF
zzsz44?&pvEF!sSPhwC@bc#o?B(58s{-iIx4iF$#Ilk_%TaYp?y2lz>?QF2{|q;7J2
z8Z5_|an`t8WDVs?88xF;aBpvoR2R<mMK(D8*oHX?b|_h4hb;;8e)3#Ye%=x7XYqU#
z<A^gg4#*<j?j}!b*4znyR=d)#?~0?u+oqo6eC?h1?^Bc0hi5fQ7fc-If-yZ@kaLon
zGzmQ_BREh0b6_tzXXbn-)LwT)t-d1$QGaVp%#UxbL>ouqXBDybrX>cnRp9;}8+tix
zF{!5tm(%U=ez_8km$+|PDQB-augfC~JWG;c-7Fc3Zdh@dV~N2JEzsZ45(6h$V-t11
zE}nKM)p5nkdDKCTb|pvVik$zb|61q7?ny_?ZtZ{%eEoPl<t$6C!fuob(kyCg&Fqo9
z-X59sw0x>ip_QWwd%BsUdyOtKvg*Zy<T@cMtP_vB{}t!18=!2J9yD&~!RxCYR@d5K
zm#qqC)u|OSSD~f0F`mAZ;^sP2YD;T`+OaAzBfCnp?5B;v&or@hhZb68Yvblz9W0rs
zi}4q95!PM{PIt6$ZXI#>tS&lU&}SEg9=aZtVA&Z7yGYH@dxses_L^ZBy|P8w#!%%O
zVRSbwSZX(k3muz8@V6!rB-O?OZ(=QH(`#pRu%|(Xz7ai)v@oF0+7MGONHDR0Ka=ye
z;*<oT3M1_4XM~l+TPx!2R(_w}9Ub}wv~c#TCi{av2u17%p`HI#3^phehunS(&&Pkn
zA@?e=YD}XjJlP;dn>2`xz768=SxxrKXkp4tBktFwkTy5Rq#`p6J;k~BCB4WWLh!6t
zD0a1H=Gl>OEal!zkKX(j^y%N}5QZ7ExhFf;0$ZA>yBrgY1!tJkG9nBots~Hh8FkZU
zvgi6pEM9Tf=T6=JhE4R5d}o$kH8EEw6^5Rvc+ou-QRg$TxD|K0KRF9trst$(ON`po
z5)JY!dIZ>8?UI25{^?lpB^^#}Qm~x!qyL#?40xRc!=wb9Zj8gbQE|*sibF1Q{|1d@
zwzgk1T7IXe>2wSl`=_9IR4S~L%whH6jH~4Q8^{duO))t1JP}Ux5qH^`L`-0>iAEGg
z4Ws_?4)=Kmap?LW4xQ=wxK2&bl(uoWc0Lx%6JpV{KL!o$V$kPF6t>V8=<y{1cQu%k
zM7%ZVlLWWAB>bF~3T^9jG!0M3ulJeg;oBO`)@5VBuhw{A)CTzza`3BDF6x)ib2FD&
z6T5Tp{#!0qTXjNt+fEo0(-}VkyTJ5xC!Czr9x0yew$bhk=UMGBR-0Pz{AB#>ngY{x
z34HCc&r?dyyC?;5j%nyVGYzljrP1rh%=aGYNaZ}gw<ZaLV%V?MD;b~pIOjK~W^50;
zW$iLB>@s(=-BK~_8nc%y(jg;{Y`QrO9Zl1j&6kFgMM-FTnb}J`pZwX(+$N70?vBVG
z_oKhY8?f3FP9G>Ub&0pjt}r{bf8hVuFI_v&*9Z0gHS~mh(I7V!Wsi?T?BTwNeB50}
zcIQ$Pa>NN4=g4J}uW|`iLZh!WEN3f_LC^U|p6mPad_Rm?YNI|{p;rrjerScGi)8q^
zOG!O4Jr&F7i@fiJOC?^A3~G*Ztz1za9suiU!D#j*7>fo3;q#CHyo}=4a)V(PABw7w
zVAMM^PpE=irfm?0bn-&`S`TJ6xwF5;9ar1AV~xxMCrmt$`5*6{f0@s!slxE)b_k56
zF4EZ&wLa!3<P4ieZDZ>qLp=Op$eoEH#yv1ZuR#(t_K?D|!kFDumY5!;z`MmZm`q>c
zp^e1t!zxVDu|w<!2UHn2QWN8XLB1YnH=B<m${YVUuLin!V|}zYJlave-NzfzW4!Tp
zhBqEkYovCMxllzO2(+M{`=1AnzjVWMT~}D?yTGEKDGtr&^|@`0h16r*TVR7jZEV>$
zVS~fGu8(XL&?Uw!QOJ<oRfg9OtZ?DJB_{5$#KxZ1$X_DErb{v`J<IdWF%!rNjA0gM
zjPb!T{AlKZy)hnG-GyG#sUFb(>xS8fIKN)7L)W27bSLkhn`?{4Y0POSZg`qHU~PdN
z8i)_gHuE!o|6bpjT7BCIleApmw%84W?A>vX^R31>&gmnVLzKu&BVX?Jw^HA^S4N+b
z41cG{F_M^aqtps%jm&YH?Tw3m%z@azv)p=S^*s!LEwhL(y$wM5OL|_fyE5O`6$#uS
zmR8uH^*<YUHqsNhK!%7#R<P)9h4enw)En?j_st59B~}QZp}>Hpws0cf_wk_(-($9%
z0aaK}yxmH?wX=4_rwNWwPv<O8e^rdP143%-@p2t^gSnj5`Oz5SfKn^+_vB<XXA;BO
z6I*}Kf5Ls}C-TQyN1U<G&>7)V*eBEAz+5vYY$1<4xsM&j-BMyskR3MsRia~)4VG=T
zg4c8_jNNC6al0(30hh63T#mFt<`@yboK7iGe~7sXTNRM5<C*lNIkGc2TNO!RzQ+<{
zdzoXS$^z>R`2B;(4e<OUZ$Uk5XD4KmSDHJ5xntBS29QGtA>VYJyn$uB0}c^y<Eh#D
zO}xz{w(p{T;$UwZ1okzB(Q`x8rx>7oq#8q*8^o9mHDdaWIx#f5PN=2Siu6OAPe<!w
z(-J+by{?DHU-Xc(Qv+R_X`p(QIy_seGtaO|SSHnq!P06`|D;;<B;J<lRSBDjI#GXL
z6Q7IJ5aXea+Y{6=d4Mi68MI*Nu7%6mS{UcAi8a48q4=i7j7M#3>!$??wN>MZw`-UM
z^<*<K*Ub!eCr#-2pr6Rq7(wbyVu5p$nEt9s%nDLN(0Db3ZBWDHQ~X@74#U&B*z;MF
z{oNXPt)-3`<{EgrO&ycIsF4p<<C$FzjyIab{j(YfC4N^N(!t}3YN0c~T14x85b5HB
zC{Fq)G;~VDjgksse)x~*=3FH%sv1S}g?~lWtH0vFt-s<yezjQgzFJg7=u+pWi(Mo1
zkr!fseM1fLc6|U=&})&?CKNH%q4c9r-$%T)YzV`7?n`d#hT_!#dZ`bzz+oT0-YuGA
z%tSw&X&#6J_qkge!8zhI{TByf5mCs#CgN?6UCHqJmV!~l+lBm`@0NnQeknLTJq2Ux
zlCj(_oqbvv)Q)Fjr5E+FJ=te!nS~zC83?IoF9<V@+_)zmPp&rBmD%rO6LE}w$X##Z
z(0XJXZ~AeF{K_umLDcN{MB~SYC~PZ;L3U6KW_^i<3G?C{&ZJ?mA`KPPk%p{@!LTEV
zILf@i+3S)J!QA&v)cEWk9>M<cC=}0+#EU+W*sT+ZKh)bU86ohOczb9$d%&0Sdx^Kp
zotPbdG8zfNF<7OFL+NvRJ)M%!znYl{DstB3u@jaNZ@E{U@iH5SR%Am#yzL`t1Id^i
z#GK8+t(G~Mn@ex(dgeh%*#FAxiS>!?VA#|Nt9N%o-r7!BG^IUyd$q@oIqje~if0g?
z6dYQajE!4}pLgP+&*y)*PcoVlZ_SCfN6Duz8<&P|Yk1vU(@~$u{MirOv%g?x^ob;9
z&1YclMsoV(jmL2(efDA+w0%<1^=b-?c&$yy6SSYr?&|+i*~^gz=LY&!|0QDFnMAZ)
zkjUOc=0NR@1UH`W2@c2ohau2#3`WLP_B8JcWN%U+=Gg>b`MDPGm9${Lr8_FsxNnNJ
zN5Meu67wBV%DrJpnIm?TP?NI95lvl*J-$l#^PXSCd!J9L6^3!1)hf1vZm|XSys$vW
zcb2$t#|le{xA%X`n71g$80Pq_xbB6_uguX~<_9V7d$wlGG2h}3$3kYO_vCJOU;w6a
z1}^mFp7gUnlzqrKHV?w#^?@iF69|_CdRzW@;N@p`{M+x&dm;6rYRrR4Wj0I)C$x^V
zN5~K2$vqVuq7}?nvV=K314BPa@b#u4x_>srg>Q!BtPC+}gCRC;*Fn9LAp%OJNSJMj
z{Uc>C_%26~Gk2l8ZID>4L~)D?$Ml&~kZ*^YeaurWc7*G3C%k;@jI-1kO>*|eMCv=o
z=Kdev1`}_qrh8-a0B>}^<%wVS+~H-$eBMM4Y$M*z-)oN4&qg@b(iARjWKdj}V^$Xh
z;^$anpr<WHe6hijISR~@D{$>JbLo0pV!=NPn2oc<pGFJlwxJKQ$P|M*%dlgO3@g47
zgQr;F{!wG}=NwTNEMYf0wW4j@(QPfgEf?KULd}dVpPN)E!|qSEP+YYk?%3jdfGuBl
zO3a!@Pv{*LS`cq1Y~r4;wLQkZcS0O}hn-(IBQl3s1h(|*>M>jCfg8kqH}w7IisY-*
z)19D?k+XK_J{bm$k)!u8KBj%l6Daq>=qWsNB(ZarGwzk{)CH^zK>Z2o)E|1_2>H?u
z#M_74oiOmC6&5M2(fWf8^zU0^lC=ye#+?0#w<E^N;cR9J{aOojikIR<IcMkr*7$VV
z29xgF;I5V}{GQk%*+xZ=7H7jA4roriJ(}jgOc)1d={aC(H+!Vd<h;k%^?B|YR!JSO
z+}s{_UpZpq2q$XBo$;04Av389nseTDcD2VcxeEVo*uwmV16IuE&v;3WwT}wt7g0~V
zT7_O0sasfUgR7gYP)NL8xX2RSw_8H+eNbE>hlCu^2|lLeuSyurHigd%6AZC3fqo-r
zux#pjK1!jp$^=2jO;A@wt?nNSJRNS0ef8|E;dwESvvl$z?op|uUR2L>&3z~2KXt@8
z?gWZs9hg(+fJKKLU~|z9OS`MkV5>xj_BOy)YJlB|Yew|I1garoRD*C`Q6msuBYX$f
zh?=lkac!R#2IZ*X&15y`=Bwf5Kn-*luK_{4eQc*rT^{l4%Rlivu~uCCS0$|OR*OTY
zs>H^dl_Gs<y|54eC%&v!!#G!U1hv+N=2>mb@YUjOLK9DOG;yAI+s{T5@A>}~2Q;xI
zQ;QisT5#X12_si6Xin5Z;{|@tQ6re}za!4Zh|F#jomw@DwQK$fzW_CKoT!FS;;r<!
z8nSEE(YB8owv5!lJDvwl5pNf-Q-k|-H7L5OVKtxUWK%V~UEU<t^;AcT4q8ZFu7h>I
zt3{^;)uL4WgZQuHy;u_bQG8N-72?ki;r{)XI9*dA26U(v<Lm0h+bwnC`+(mf&ag_{
z{I6O}s;m<QrD}-3qK!~;=}xnLic)%!?Is4o_h2BFa~5!H#=Na})IxG+l)=u3cfY9B
z(DB7{J?4y@qn@<E2VX;(U9cb!s`J6vV@EB?5rH??VzI9}7ArmDkmSbxCUUgPwAgjw
z9f!A%nWYQ%CSPO^!sTT4zNO&o#}pjbPi5zMDl=o!F|(*8)+;j5t2Pa-j#H~sos3Wi
z>SGm2|6iA7O}w2!yzNQ6S!2mey)QAS;ePp6XcVrTqh2=u|FX(2evZbph8WCni^X7N
zERHrsp`a=nZ~DfNrzanrMU4;TUzL;TOCa8^&W}RUyhsfE6M?;jp%_kYq}f*XRpf@@
zh<`XDi7~I}=WJ_94}LK-38R_wz<)zUyd9uQ!r#rzVppbM2K`1i$EIUYeJk87$i}#e
zOca%-aCb((vR^XNw<j@YKN(X~Qc?3?E0|2k!u~f|=o{4zqaL?Ir!}3>t4CYB8`&0R
z`}1Hos~u-WW*tZ6;PU=t{7z1S<xS#kM`q;@NnmGqB9=c-ML}X3di7(kPDUE+syXXk
zV2)vY5?Y^QKkGqumhDN#BYthy)nqu9@!IRBu!A)PZ%?P7Oh!-Cd_JxQo?pf=%OWWr
zn?wSBgeD?pDCcO8L_~7NZ@njySx3xFw+n|;N(d_F2jS=NK#cDlNWW?z9CTdJroa)(
z7Y+#M%dGBy^yTxulta8dJI)assF6KdVvk;T?QxVlq4Be+ugquvMjP(chA8l?$`Y+K
zWvHj-Bk_j?7Cg4V+dG!9ea<s~qztQ%$nfpBoVktkwr|i!%P(Gdbi*CT&XV`~<_zg9
zdyJw_v6rnAZrpBx_wR}OJR5F(PaHn!j{)?wjvdP^erme*(rZ$B-vb+8xMOmG6J}Mb
z=-*UfY@8i}sXZ#4p~T4VwwRz}iw%pJWxm7|BgYzJ=0^#(Z`VV7uoj-U=_16H8j77-
zm>aGGmt%&Ia@Ke^#1f6f48;dI&U#xTiO*4^y%J^ND)iQ6z8mp&zON&)zdPdAG3vCx
zP+P)}j(i*|h)W+IkbiwkZD)-W%%3}BXn#+1?dXZ8-Q2K}`kOq?ETh$UO?W>n9BzbB
z|4ERt*#hGZ${~))ae>^a+6K;UU9I6yye*U|*kNadQeM;MdS=KtB*jK;BltZtLc>Q>
z>}W&&;E@%4=yzx{n7#;!6idAIv37+%tm^d9bkqXsaw|L;D#Na$GAtS)<F4C`9h1gz
z8q3U{AR9cpXM^+3+{I5+;lwo+a$N23INlD6p4ve>oV$a|PUNAT@pcV;zr@-RkKNGf
zCV7rE{Q4X2B=?gOy(35RemTAb%3$Bo5{j)R(CcCa4`R@0eH%=pzvHd8JD#?0!5nyY
zvAyTH<1Vu%#&Rb+K?XPC)%i2bAh6>e&zHK^X(sq%A%o3W^0OrtxXE4c9GL}18<-%p
z$OvB^@LKh@W`7s+=r7q|rk*X<?6Za2GbJ_<Z|@m8ke{+f%2)*=mniT@VT(n^^my=0
z@rpVoYw|X8*Vw^*g$n+B{lp1&lz*^C!z*W`HSmnVorx2*;=egF)@-9U<&Xnr+1lgO
ze<~QZ;@-O%Iqd{X7%sO!>P|Bl_c3E$kU6y;7AWyH$H9}_54xEk*Te+*CyZe~%?L%$
zxobQoLEv@?!u}fJR<IQPuSqfOu`xanbH{6%K;tLROx(e5t5f2%H*@o;0kbN0Kqhz4
zCmg6tK1!c1wdxt!^o+H)M_UVf=A$rk(9sT>y;SIC&CJ5R-0}PAL2T5;O@A%KPSV7}
zqm@E(p<1+fQ7vNQ|A^FPwW76NgLtZ^hE}P>+#8KT@A5y<>{hwxo>4AtW|xZ*6U)Vp
zs&X+}_m8;qsY(<@SBY1ODxr6_Qv6%-S1bwnCyIVm3X=;pV(|J}QPTCVxPPrtw5a|k
zwzh2&XD+MZx+ytl;%%QDnpjt>iM6GgSiV~m(~`B=ZA#AhrWRxY+E6%Z<JS*O9QA7y
zX@P&m{`7ibv!YHI&ZrY@ht-Lv0d?Z}t$NY!Yom~G_8dL5N_49)7d?r$cc-dhQ5QA*
z51+kE)ZjYzkGTFq4LYN>5Vu4JDQ%yK8iPk-vf2j`y!(-G9`!~%lD`$xbKZ(Al8+*1
zc!_AUfT@cCzeHc>a?!53LWETQ5wCCZd@;RNgjLmvW??@?P3%uGdiqacO%Iwe@%F}{
zK&)>OjAgFuOMXMW<wzf7Gn-(8u_yZO^kn9O7wV`Nbs^q<{O60ZYF`}d;)jP8{&4QZ
zK1^cn@Qc8s$LwjI8wrPoNPL{iY&PQUlauin9*}?*-pQD7E*XmRDcDZDoy*;>MFjVS
z#H0(?+3_vQfTT0|&Z=bGGGiX!jU*f-rVXYpYr%N-vAp1pbz~eQM)dvciNPE4$>E++
z_?r_&-F!4NBBHUfS2U_m#!xR7i*CeQgVRxH$FH|Dh{21sk?awUfQ1EX?Z(p9y)FXD
z6QZDOLqAk$1Wca<p`R=m*S7>CN;?EAdxRm&otV2vpob@Ws>jn0nn~=w#(&R09)%wh
zsLNw6U{f+~hoz!weJb)#r^4(-D*pS<Ue8y=VLrDN^c_jpv(tn)=a7(!SZ4Ph7@q~t
zm-Ouf(5ri=9j>nE#B+69o=e(d?yfw9U1|&0LAh9Np94GM?Jr)33p^<F%8JA7z<AvE
zVwNnmQ4wD3q3fLnr(k9f-cCj2=47~1+m*F35r<yJA$<;eRCJRuNsn1dt9UK{hua^i
zAva5bcW?3sBa`spRWxGN*?lRZDceoJW4}by4oXBf?yC32F_Zr$J4|LVPe4B$dEp`0
zIVOlW$=r2iAoK8D@S57p0`hQK)W5IFQsLJ-J3OIYW6mD#gs2BC-E9l&?e^Hd*dCXD
zTce|%HLjW2Ao8XHGdo(rWD9kbdo9_cW(lkI7Vs@L$M<8Fyna?FiICy>RAZpI1ixk(
zqWNllbQ-RY(RXd3x!E4i3h8stut)71dj$A9K})#c0{5LU@)p={;f6PkjyOKp0iWva
zvEI{x9%(N`U8R=C-4VG5sWtLqHs1Rhk?vnF2A-`KW+QmUS0mThOM$_hv5FU)GJC-o
z*3FFYbfX?~<20#B*Mz6g!t#7AytLK9$y0{dH^2bLq*4U&o__3z32t~=;T7>|)-hWw
z+o;6uhf1v7ZHMnl_G$i*V;*PH@$?T|oXd>!El$jfaKZ@U?NoBwi+?hM?wK<l&v#>w
zEVC4PyI~0NHn*h>a@?%oyv7_y2hl&+Uc&xm3oIWfr}mCKEN9)v>n(Bc3bW}}o58Xl
zJs8<$$VxI}mVzn7GAZ)4jPPcu5t2NNvB}R2d(NBS#4%$`$tBiy)q}T;o{4$-__an4
zLssge{ce5waH#qGst@z&)S4`jAbf{0UimXy?-Dubvo^RXQQ}g865Wod&{5-m-DP&z
zbHEPz#GLwMM|7IzjC(7c5o^v}CTGXn&y{FxXpM*Gd3~Q)!IL|Lvq!BE;ZF@`1ob|{
zO(3$Y@O_#U&Pl0vGO)!AD3M&@41X^VR1|n%-D*ed>mtV#;%$1qA?~g>gq4K}#u0Dt
zj5R^Q5i7hUemQ)xz@Ojd_-bxJ4X_0+jkQ3^KLa@TFoN|7?zE5Fz?nZke5WmZII|Bu
zr@-bPGCU08xn-0BT^A@&^4c0B*N~UqVuh4N)RqmhhA(-0MRz6m#>74wYPvt$!#{^T
zsjcZraa5t-3pslAl4B7w|F%3*AhB4ETNN_y;AHT%vw~ZKIr<cu;^83^RQ57MFAa0-
z>tuo6#M`g?nH}hC!k!Nk_y!qc)=9qhh_`QdOYmkby`;59XseVWwor<#XN|Fu&o}R{
zF`9>%z}eUYrX8hdeM*XeG!rzmQeq<K@Ir0s*tYRp_skL7LLE@qmYQ~c{fr-Zmw_q_
z?4aU##};|1+)GT*#hh8{xKmy&noa*B?r!)kj*qSur^yL7_Nx?@9<`!qR;`%!yj;Y@
zl#3OP<s$28nV2KW#i8_aF`2kC<awEJ-drJ^9R7&%q$=T2`$v4v{v)CfR*J`cehVM9
zN^wy7M|3Ex5z!-SS?2Utc$@qat9G%NB&SK7I;Vz7%hl<j(ZrZ>ntZR2t0w;XEzpG4
z=4z2vQzb@ssS|IvHHnGSn#A{nMlsy0QB3fy6$z87Mc&9dF}6pYxYD&wcn8*rPx^J@
zX7hhyc3`#WC#w_-EBIKOmx*|TQt`sOOyqSe6H^bB36)v7c-^&BZ2DLtCe42-maDxG
zI~P6@)2=)aa-9dFc;p*#ka+te@2%MR`K@>o`BD7t^hKo3E*2jSmx-Gd6(V(FrC712
zRP;DpEGEA!7PU>qq8>lRfsmhK)^&gCZG!Ne+RYul!N@)9$9;k?F0b~6Ra<ZLBi?o*
z-Y!4FKIPJ8xNYo<ibuZeSZB^njvtbRA3bjVSa&D@X>ru6%w~twXX-$|1z{Jp&ek94
z=N%Y{;hIrU7{%h;t#~YmO~JS}vB(?|gDXDC*ut4Buqc`MOq{4^hRpJ0<{c!WE-4Z1
zOcOBccsvGL#^c#0VofZ&AwS087xhtbqoUElig-%TN8cL}%miXrRY4Ry`!fsrQY4DH
zMWQJ;6mdKcOzXy9>FP-I>Kp;P=ghnv3V5%NfR26?GYTV7_$vZ7kArZ^g1H5~f^q&x
zD19wqc<&O<ys>aRWDZez?{LhH6>#eiMeR^L&b~{)zmi0}{+o=I-l^!EpNfm8QgQ!j
zDi#E$BFrxil^3Jw)8ZL{oVa5$^;~hO>^yA+n^9S~@`AaTHLbDYa~>8i=!BMA+cA^0
zEp{Kw!-gn&hVP|ebe}YciX`S5Qy(>-KIiB-1O_o5iFn(C{-(a%QOEL(GTtr?#fMTb
zd3`dq1@uK9iNktcD+{ncuXP-9mc=8#I1$&)*oFQ+31QTie|rrK501dhZ4szjEKuMO
zh&dNpVE(}tc*$&D`%L;NuKB=rwk?kRW*3Yhd!oCCVshIc4BN)n#0*!2oo83aQ{qVu
z@3ZA9Jgils(^WfW71_aqI-|yRj_{eq@8><M?|yp1xwCaBqG#s40;7Vsr_!~+9r9>5
zx-g@SvrR!e3v7I4jt-ILxR_^#=i^P$d@}Fv5fbFQ(PuZIHqS!(n4^^7aH$E})smkb
zufVsravUb+S5rd*eM5GL8sX#<BUn;va;GizC?k1q9H2tbU}yNyU)tTx5xW1et1_2b
z-Mha<o;9&Hv_d?uE*E$Il#8Q_YsBsCO`@N-8rIHN!yu(5p170uD$`_M6*;a88tB$e
z1KlTRW0(VVBmM?BNWAqR-VP()_95P;aNg{hWsN!BN_r=4pnJv)byvv?Hke@gRXG9=
z%kg5EoH`mA9#zvTpy1!9$0K;A91}ZQ<7q2fyttx(6Zd*f*_N0Q!o0cGCh%8DG5W0p
zW7<g2HPZ-UH~oX3Em7HuShdU?*9^>2K)fByOxVyCW_TEDhBH^BP%5O@^Gkw9=|*Th
zpB_hQJ*QlykN=+`N;~PGbt^q2E!87$sfR9O^pKIOPtTG*l8Lw1iu5sUssS#Kl0ajf
zF};O07+7G9z=xK|-)n<!4{Y(CyXrw3tnq-4b=xirXq1`b2sN`;$hB5jvuk@LbFZAK
z(dli6r;#eO)waT9;%&euEBNiQ!ZIBzjOk;H8$XQMTWyK1IVwm-QK#!peeP-}bnEVn
zQ1+flzgnZ<n<?HFO0XbQ0+~PaiGCPBt&br(-jiU?XcIVnGUeWaIY#HnjmDT`o4*N0
z^0jqolm+#)`dHtEK8{T?oa1A^{mcdi<dipGv4Pue<`Pi%*6xo4`ZjXxSx&BZq6IE?
zw7}XKR#+FvbJ9r8zCGA~8l=P*OBJ^3*+Krz9@hpsV3N9==Uo#_BIl6dEr<CCIXa(L
z;Povz-1OuqrY7*iKTE`N4{31T6ua_GusPTi7Y3SQo~|*Xbd0f%I-`qAjOh<I#v^}Y
z_}r9Y0q0ud4H8TqD?!IgM$kQE1mjnndxsb!rVDYE`hjB}CJ3xEMz8i#JUz<S2%mFv
zM>C|PDG@M${4&qLn+)yIo9E8yyr%x-A-tQ}kt<PR*99fJp7}l<YlgAwb@B3m2J?!m
zMQW=*VpiYZ;&jh)(R)sXIM}UHq$QV$1)k+%TVA;cBF5fyC>LY4lnDn_x#&o&O|B&F
z5Py?bSBQDne~ahMszl7UKcZikQW2q1Dy*`9i-9)3#nG)L;`)i7LTzZRkZ!CKQwtkJ
zm-hdJbIm`|x=;;|=c&U<PXqNEHE`{$22AZWnWvzMPtU7_ZCt6?-QlNbF`-113@8zC
zNhM-g*MH)fcde-IS}oS5REq@{s>C_BDsk}QZ;|n;OdJia7k2xqMZQg?nDL=pT(l(K
z>X(XrK4oGS@iv~%p=VW@a9jLKOo@IaK05s;uKa#3)aE`CyDmKt6NcRvy&Z0g-g_^L
zPo?L@?$f8m37wPT(ex9d)1Z^$>Dp5wq1QRFBkF=^r+!6THvT3wE_@Yp`LQkdyU-2#
zDF$Bg$C-bD__~n!0d~RoI*3}yDZbFBMxo}V7ls?q_fAdbvdvydz0wR$-?&%m;)^xk
ziOFYtIV3Xo<~FmuT)c6H8QUYtLC+8J!SuPz>fFoT;IqwP*Ef<~sxg@Pk34TBy*DS~
zV3rw+2|uD?wIl}5J||%@=Px_rZCzR-_5Aem?n^|}?gSJOZ+{+&N0VVZPW=yWlbM+x
z$bOh<oIAgUq1GY{Z;k=IA4agUD+)8mkoWBzg`ks>$m<e`^QNKnKnir|BA6G#?C$LX
z$7Tw=ZelOy;xIhY3B~IJ%v7*oc6lG3JFW&{>k;}%4+P;WzxO=zNV{>){f=-XcML}a
z@iwAm6#9@;p87oj7rs+3oK9U)I62s9>~Q_by(vE`=%q;}*8Q3ljS~}@8Jo>Ndz6d^
z^hgyjr}xV6EY9m$cvjJx_qjIENNJ72x7l!ER@c13R%nPwM_=Nt3-PveVjTW8F$d{t
zH2j{%Vo6jyG#9ZuR5O*i{p=0Bkji^Q8cdm=_1`+4Tec?P=mGZJou;?gisyykFqn-F
z#e?G^h<hE3P<3YZE(%A~Gp@IqM_}#b2>fKOd2owB+=}+cY;9kZPHV=zJ^E6Kw-+8-
zBZGOxqlmXw#M^fPJd5<EzA?rEXJ(tDgg#Ek!`$J`qz3As5{E`9F{!f>elyu2xyuG4
zsbl%^ig_=`t(lK1g?57!`|l`_>qFga33Fm*m}AjI@@X5)aiA^lhcAd1F7z0*F~jIy
zrq~>2j4mk>eArHovrZdvtF&?Vz9CGATg#8j=x=6jd8!ezbM>GeOU?gq338_!Vg6$y
zYOItH>9&xCaQAqnM0`&EEE1c3ilUz-Vz9<9aml<~xIL~AJvMz7R;x>e{C^zSnjgYR
zycfI4do}o|;rkkDFMKuOzgdg^YE2B=u8FV*8Zc|CfxM2|2qxaT+|)x~&WD5U>e0)t
z2Ynp_wD*wn{6xI$Y=eDX3jRKbeZ<>=uem3UGNR|h5NEv%uzjKotKQ1cjCgx3LXJnA
z+f>vPcjWFhM8+NVeoLrT@Enq7j9=kWgh}<V;;}yV_)5@5kF(A$DM}tOL%okV9&4L1
zqudl1Cz@j7YE!)A{q`PnVOJ7wFF8n2ZYIIqL?i6BH%57iHa!5^I253R+Z}Z<RZkb6
zH1zOrpdS75dRTIhKFLeESp7_wGmk!sybbVZpdprhG{U8Jrr1rQDEYi4LR4lr>_IGO
zXM(~a6FgbWxsrH$^^7@m<>W3rEOGlY|BjOxnmU>A`^kmgHbxKXgl3V?9=yN==VwW=
z=Aaa(vW$^hXbkUtCdj@mg<q5u>&ScRKC^||V_R&U$NZkY=6KL#f}M{H@hrg*k?-l%
z{A7Uh!|4a<DnWN!W1RZV>zrc&^CRZi&OD%ZUnOugl%P^-0A&Y##MEkG>O2|DAVbIv
zYLJR#7)5>b?p+GRRr6=uk)SvA=R1}eV~CP}E71ba|FcBdQ}XjgR!Gj0!~F^O#oYCq
zKd`|Hl^wqDJ@MSc0GSI-@Fmv@-51Dk{J9MK7SdOn!Jn;Rgbqy-XopBK!QTvuUWV8~
zoUHzA#GFn34rfTIQ#D5YHz~6gjPY!|G5n@V(Kyox0ls>ea7zbGi*(>OR39m!hVa?Q
zS(oRNsoBKYZ^jr*4mzyH7$fqeIQ~upx~-Z-7YlWijnl@(YJGBI>}PGXL(~WEi3c)I
zxFu)tliXQHl5g3|=ds8P^+xQhi!g@o@iI{oQ6_F|E)!?tD@4=JGT|yI6@{5)Vsd<$
zF!%c@7F{Y9p?iykUr@1#2>B`k#{Lj%FO&$$lV2kHK$$R|QX#H{{ub+={1Gi){}FH7
zm5P>Czr>Zu-(rSih1jsEM65aVQv`Rf75!G$iQ@GQqBf6w^8N-<x2{oK@>D~zjynE6
zSI5LX8VHDP6lX%~1Q$>u#-TzSZTnN`*p-OMx#V++x9_Y=M19mZQK(iV`X^Njx#Eu)
zYw<_?cmB5s7)UHVRVF4RmW$&9%Ej!~<)TPgCSL3C9$@fO%-Q`_IBfkYG!}dlVt<JU
z82na@ba^TKoSurQeIJR8lMlpzg!@9L?v_}6_p%6YQz)YTJ0-emoD^B3Pl)B6PKvIp
zPKhbq&xz?j`DcSKh||L^2)Bp}VvPC)fu$G4G2Kf-b(9$|#9K)Lb8fBKHGJ3?YU6$J
zk3C#xH!#m@9(@ucJ>fo&8EcQ4VMG^t#npTea@!Znm>IT~d70DuH$yEoAA5<n1)IGw
znRq*jc)NbP58l`L0Hy4sFs8qMG&7kS$?cwDW-N1KmG<m^T*M5k0nD;J&c3d`iRic@
z0h(*6MgAI(@jv2Gt<DZi_Jb5N<EP~%W&ur(!9g$1jKtfbkIaGB3B#GqA*g#7g2a^}
zXgCmpZ7)Lb@>2-B9)-a5b_nK{(Ia|~nrm%=!65=-6mwSp5U&@6!RBuWmgEPcuXzwE
zdIw=V@pdBdR%2ffGiF0E{aYB`vRfpkML6cP4ae&K0(Xe(?Wm#Jcs!O;{5V|YE_HWh
z0y6q1Ah(ctRSwaZpBDx9Ant~HMPYx(SVX;w#Yc9yJ*3xHa~yZvKjPSz9FLom;!&i*
zo<nvjJ!Af{Xa6jmr$%gMs|Z{l&6zzR5@DR<k3ZnvST`El?PKv`UlfvuQ8TQbiuMCj
zaqCPf>UO5$_Tm({%t=C3duD&N0LBHdH<r)66Y+L|V<=AZ_u8J^M9-!GeC9q)r^FRK
zPBN#y&JT8z!%&(Ri1@Y+sHexi@R$<&ZRj&8WgpdDYsk3sl5SQ}x1wT}iao;G*`vms
znJ=70c9YZGKAPDIla;9D-Yx2j0?UcFZ@5cxrXIxQqYdVhD|<s7LdicZTsfwV)Qfu5
z{Yvosi4<SQnqn03)_sXN9v(HteqzN6BXj)M)(lsvl{_|r@Bc&zI&*iMmY@xJkrtv?
z>%gs0AGYM}`VUu7LPrnXQ9an7qZZ$=LOiMdCGO1nCC1KwFAl7DC*t3{7P)S3L|EKs
zan<~rfaIGP<n&#5NXvyCd9gNYK8rz}J_y(DA4FE_C($+cohVkm6)!Km7iT=b3MKKj
zl={it{mEx_q<(R{2L8LmdqFI->Bi__#8f?8kJp0-@m5Z}Em1Q--$pBTZLpVQp91S=
zaJQ;y4w-3{$RGV%{O^2lsXO^`xdGmP(ue75eSD8Jz(vlG(6GXKUCz10+acCg)SFwN
zbq@>VQ9JzQi9WKpGd8%d&$GV-Q?E(Ti8Jd>@{O|HMriqqe82=#{20R>;Br%3zGzBs
zGI`@x64<&)knq<K%4zzrQR-vSLoK|YtBuYQ9Z1{gz~P?`o)LQ&T+~62-MZ*byiGo$
zi<AOg)Sb|UyN3ZJ-3&2nt`YiEtF`ovIp%Xt+sL08R%VP!&fW#rP0?(tK2lZ8$EBZf
zwIw+VEld1tW{yL`4CjfrLpMor`IZs-%`<}Db|ajaYJy1OZHW)(wrmMsd&aQj+-LYw
zioARYhVIozbqr^&L%P^o^iOyzOU0k`Dskphjp(@SpGdaUhj?LtYR=?MLnO=x;#p#b
zDIBQ_i{u$cMt^1B{u1n8XNWUv^|8y0dx0lfSV%4;A)G$nr@C;stb>L0fYxr&h2eEw
z<f!RExm5x+&X-*aOmX{%IaVL1R^8JQM%*hO%a+679?y@|7<InLGo@9H$cbwddw33;
zo@;{Hc~;m*?fBy`dM+nhK<$JndiUe+sa^t$^+p)<z!>wg^`M(-jD6RQ(2Kql8DGD;
z1$wYNqX%gqv9?i)*U!0EKPSa7Q$0izZ-eJ);Xw->#GKPZ<9c2jUt`>l=dSmx6v4~&
z(Is0K@0+xs`%(jjK}{k{p^nnY+BlJCfbOxT&^T;@mFm<dPPBr=RfeqrGPFHwg*EHx
zkD6fyo72X4|E^LjpI0WD6K_K|l!^RFJcHgV6NxpyM8OVXq+gkEC*IZ(Z>{$fi|&EN
zqI2L^AtT<-Al?r6@Jm$i``nL|i^~DO#gRdk;<Qz{(8?(luDid8g6$v0p!Ofd$(T1{
z&(xpd$gwi<!@pL{$!`!RGX9CUbq(TxX`>jQ+$jD@)loG}9lmA_!f*RuvHMA#n7OG+
zH1a(=Hv6YAN&YETefS}4XMGVf;=YJ&pFfN8j1MBS!w1p2!w*q3m}h`5KZJ9AsfgWE
zCh8K(#mbiD;^d_=@u#L#eCbdk2I>72F~r-oTfU0zQD1~s?R(*I;I$Zf=_RFWPsQY3
zkA&Ze2f`=rK5Kt&iTOV+i4<j_$h>|=Sj;~zw5}Ei{puqkdE9<+bHZNnZ2fMrOM90X
z|7EB66TC|tNZKQ+X6+Gy6???tgZ?P_#QYZGZBv~uTpQWfRp5*j$6U~hIVry@nFl(;
z6UDEYXJg0wRO%?bK6&HU2WQwzz0ly?jQLE?xG{(M=lfjHTEi6!7PDvhH@m;0*)>u{
zthMz)@?jtJzsk8;_+zp`3#?kj-t2~8B>!axU{Wk|eAt26g_(*g;$b+DI^<{Zh)ImY
z&P--FS46`_#3F!ryKq_z)|S)nIV1@B)W{Yd3&geoA@rDpAb&*&_U~mjEAiI(BeU$8
z188<O1gAs8P=7THTEyGg0qn{45GcRMtUzLscQW&LIoH2z3}k0h5E2%#(`6!O+pR&&
zh-WXQc`$tGH~!<vT>HpyY-$bcCElK=4{6Ch=Js2};XUW=<`1dadJ=`U?1WuSydBJ)
z^3H%L%yx@HO(OeVpTy#B3A-O`m;+0^)g#`%4UWf$iOl9oi9?nRF|;g_`I~_5bzoql
zKpu}OLthI_ddqIJujIguqHvhFyWlnRt;x~9>m7nW+`FqzvYU+_pXMKvaVIDlop&<t
zeR?>8A8{YQKLDqP`{AdVAHFc}zWkC8_9eI?W-0T{i(PSGf;+mra>s(dURY0^qKOwX
z58Rmbv_^?lo2_9(thkdeM}Q|iC(I2n$hStXdDgf@E_kNY1nX8Z@2sT-0?28OaFODC
zODoioN1I-t!1Qo>Ja<^&eVGY<yiv!QZ)*72poLz<+nAGj_<s~#bzGHc7o<Z$LR6Gc
zX+c0G)j4*owQF~GuHD_;v9@A&ivfysNQ0DgNh2j70^fXp@I&2&`*OMWo#)J%(Cw%t
zdxDs$W~3*ZIWPI%){`N`+s(v^Kg?>YC*Iy#t0%jIwB%J^E&5e8rLU2SM5n6Ap$lq~
znXDmFqan|8TS(e5Etx(;Qylv$Gh^c?x_-^U+ufOX`XK|c32D?3rr=NAcg!954dDq1
z2n^1LX>mTLc;;j5vI6up&cQC{Ow5~=iFlVpjK3F;oW<W^?w5h*4QYt#Sc#2;ROD5e
zk}Q~{BuXok#B+<HDChA0N6q7qC+f0+9>tAC>des9klv#;<?O2#;?B;nPHhckL_{5w
z5C1~Iyh<!?S&B#B$)$BrV$Eg?c~8DJBt=vHqGzCg3VE}2O45tE$qv+S#?Z^4pQIoo
zOQ@}Ep(y@C6y@nO&ZJ?QQm?2*54Dz5lxVR_TuXj!)Rt8@xNqJ<%-f(R-H5k$-w=Bh
z^kw^4Ezu+1#ziQK$yQCdp+=7Qm5Tgbs=^*`RdH#dCTdoya_NdHwbW{IFicI3d{>jc
zv(;tgZFPA-+%-N-Zh2Bmc~+(+<Ic2@|CC$EI(nq4>zI4@T}7^zs!D!_x@Z#5mK$q|
zwKH`+n)DuBHxR2nx^l`^SAI>`m6+45L@`iXV!DynZq57y3v~$%P?aH*TT1314Qaha
zL+&K2%ia^}@?Kp{l5BYm%pB;WRSBu*eX~Ou+`^M_vN##qS8EZc#p~*zN$r;!&#%nt
zJJo_(CQZ4(>uKDeDuah;%Scn6H!d=tL0?lA)~Jil1r-UkZ$@9IKd|{+hk<_e$e<@h
z>z=yk2dYbAoVrxd%Mos(Caq)n9P^y8y?<*-v}q&Kqpj!<Gn89LJIFrjtiF_%Alu~+
z+GsSQv~weBj#Ob11I1T6|HRV)W!SaMKs1<le}*}NANDt30(G(%A5uTNMqN%fQIph~
z8r`n+i%wLPE-y9Y(_nSE*ji2OOw}Zcczcu2@#s%2<+iPgR7ERGqAh)){WPQn`Q^m@
zt;DK`y8-gcSzk0|<pxE$|3yK5_h`bV<~q21EJaFuDIAJQk+t*(GBxwi=6)V>r>e^b
zdIGxGwUdWa+lt4bmeP~YqZP4yXzN-uZY#&cHD!2otQ;#B{ltpka@;L1#X_YL9N18b
zdk=Gwx+52nzjC0Eo(+$^Ine$$6Mx0$ps*naXZq!$Ut~VM-~WM--i7c#P=xP}Wz<vU
zq4IVX`V()Xd!`~cvH%C94EHSSaM-F5X(Jjj)3*UW+_iph^&5)XzcE9j0YOa-)JHeq
z*TfPmyjFujOG?msP%$Pi&Vc!V4E)JU$EUz_Jh+{Xn_V)ISC|IlA!*S3CllsB3$XN1
z8HU@HVep4yeCtsJkLY5=e=SDDi(+c460z_@0-RlwFi!H&DI^CEI)BFm&19I|jmOGu
zF_`r@3fKBaB1JzG4~e(>tv_Sr?~mB>-&?Hqehs(ZFL9c9`)1Q4T%Uax3xn@c3w94~
z#M=df9>Q<?L+tJP5bx5uiS9T@G4QgNZ?kRXByqO<frV`0=PT)!(zTb3bau6sRNj+=
z6zTQYWGyB`%;kEdg?tRNkmCLp)CsdocQ<oEiML~jx2reX%JO7e_PR0aZYp!HJ2=R}
z*AB8al>HKid&oIBOYarrE{o_Z8SX9!qNk{weCeVgqI8yhY0H^EVMDFYwLa7;^^xQk
zeI)BfA9hvtWKM@Wdr(|u+9VelahzSz#M_NC+$1~6O)j5xrykH<rkrq>!KdBj<t2Bi
zS?Mkxt6W8uczZ0pmqhgIE$-Ya*Z7bh4(}x=3W={VuFOPakMjZcc5diSKD@isd$`IM
zH#eEB>n1-ZxrymvcPTo^4z(V=q+JrTL8uQKUC>uD_V$w=w(NEo*-y-)$hlgx-`bJ9
z^I&>snDzSmNFQ++(qCGo^_Qp3{U!cwFY2FqNR?-I@#9=ME6`DfyL6KlmwQPb=jYj1
z=~Z#CmA2oV<wP_2U*-t@w~aaM&p4ZZ?#pf~o=@!iigUZZ(xHPpdstj#*RJj|fxN`!
zMCLHrGBasoKN)|ax0D9B$%?(sqW^>*&s#Q<-NA-l8Ec6PvJ|DI%#q*@^}|5wY{*|e
zVD62vPe+NO7qb^<s}Y^4A?6v_qnR8fdEde)1KAv}BLOd3F>j}p1o7;BKvh>BZJ}P{
zptj6At}R`un=IfSF^71&I<>7F{qqM|MGa8t)Bsl{1<|4oaBR0`tflA9CQ(x^4$u<)
zb@UM`>dMN=dU7#aSN>*Js`*?!QS;Ij=Sc1*dC&Is(UPq;TI`<S_nOg&xt+=DRx3&8
zG$py9pek?4C4SkaCF<YRdFEG?DxS-q_!r=^cPf@^rQ^<rH00GKVZ@j3@M#-|ZShf1
z3yj9<xiLt7!!zx-T$K09#e?z;-1LY+=SR^<lt@hM%X@mCG;FC!gHdcIqBRvIZzr+#
z8TA^=>v5a&dgy1)+74<mX}7w}&sArBfx3(ypdl-UX^NY#lGqMck{Sy|>Hq6DK1G%z
zb8aQ3R+S(zv;<Ah%dk$Ja{#%n*}R___?4q2_eY=jF?bpM3Fj5q@2Mbl2?}z)SV4x?
zC`kPvMRB7ZGne<|y}|0#%V>$^JuMOLEuYTOmVKSHrH_}U#9EX8J*y!PA8E;`59CF;
zuZ=Blz?zI249_UX;Rt1!d0vItH{3^8E6ap^Dx#pMD(<IMWqPKneBVlb`M#PY4pFB@
zn|bog;5$8BQ?}VL@7=7WG;$7a$r=7MG4~GVr~Sz)GAmeBMmwuZ$1Upe*L@B7(nXv3
zXxg%l+N283v4>tz7wxAl7f!a4Md>ZYy-y1nZ%F>#TwTnFw@L4mrH`Asv}j4J`>iJ5
z3&>@ts<C5(8M{-|WK0D0sEK9R(^QHJVab?Sn2hBWsZh<#!RFbjvO8Q&dT~x&!C7~F
zXAN;LQI&$}s?s4(NlMlz$>FYL)X9|M$@Oxm+SlWiDf1EjZ9tuA4eo6(#oHidxv!`q
z8Vl6qTs3_o{5cNkaPEJrED4hprOjP(va`7JcwoRDWL1ecOTS6i-_Z3cf$z;sv^}4R
z;wxELy1EKJ(G^(5-R9VUQfQIu40Zj9qep6BzqbMTYE39`QxqRpbxBoMm9f<d@*<!f
ze|J@oW9o`>bBl`XAl{k~Zw=b>Gp}Lh8dWjyr6=^hvQ%DDl~HlzRT@;~$uu?j+DStc
znBDNsw}r%A)sUXQ=?BfHUifDnW_p$4tXUz<8uKu1N*;_$b78^HJuVdDmwho7Rp+Do
zG-mH_X(^qTYKmu1CHdz?2}0JDqx@7cZXGViR!_bj&C2niv=puC^5A?T9at2Hfm>qH
zMkfYlPSH4Imjjh;8F&?wj?d3BVRs`Nbz%8%KmG$lmIouLI28^SnOJux2P4O4;<`l!
zzPY60Q*|=BTu;GsY8Dr`)WI~Z4ks@+;M3prFv+fl!P!#$9bE$Vj1sKh$>*7PdorjP
zo%&}ZW@k1wY{|gxJ{dS3kPgoe>F7heowq#=x8I~eM=1^AdovNzq6k5Q%g}*4q&p9a
zkz!SZ$!iO-xvl`?lnZg|Od|4ZzY)LVQRESaUBz*De*GKX?2AR!x=3{E91ia>U!fTp
z0_Cm2u<GK6_od!A()1A{Z@z`$(bu?B_Y%)4AHuxiK3ay~LE@7;h)}qPuN&@S)XVz_
zY<~yijUQrQ40GfDa+G<u>_v-nsp@O$J7-x)hNgvF_py+8W@vV(vXX>xHq24CmIC7K
zR2OO~`_bP`&EZ1gZTSOp`N$c!Tcd?IGjCRZ8NbHDMt-m}<ii+hMH}r!=b?j)C*FD;
z?IBZoFpq#8l;20W%Ag0#I&E^7Ip{4Dw5T6mJ6Oz!w_B_RNyGtmjQ8(JA0qPuKT}72
z(MkH>caxM#R|y&IA_ep$O?&MkZd+YNH{X?cFRo%^<;s3USJ`-q82+ETOyb=8NYh#F
z_a`n>2bJ)E{PD%!lEKW5tQKxE=Bul$kL0{;<0O~&cb9I&+j3iWHkv!iKV!Pd5@vji
z9Ox#MdhX&O%nBs8oE^_@wcK9Pn|SMViQU0E{bjHrF<iH|<Qw&t9EaX=u~%=29|TG2
zFZ`L9^_tjEMqFan5%E?zjr}uS9c0H&d)e3BQ7nkJQOx(7>+c|6F0-4()>gEZc9&D!
zDSN8BiT)us(KyyyK0WCxCi9rpu}I|a3%$fC!ByrGZ;wYf%5qEg-Su<io|d`!8r|f@
zF#2GM={f!6K<}ZQ7`?KQHtlWXR#$6g0?~`dJe+a|<_%FZ>DYxDqi1bpoGv>#m`U1W
z1U(vQt>sja9<y%snQy7j3{L7ds<p*+9(RD3T8T>)^B?+=-z4_d9?)i9y0-K$(`FYR
zvtAZcPo%9Vi`O<`RB{7O8a3eGm}cbkQIIuO&G>DlDETWinUAU^M;w_!>YyW=j_b%1
z&Oe64TaORA(r>k{Z1|)tsy(#j%RXg^(W%9c?B57GRgHvQ^e0Fm1|$^Vvr;Ckqxhb4
zr9Y1E^Mk(?$TQSo{hI<Te4mQHJQC?oNPu5_68`Fvh@aP^a3?GZrUB8|O`a(0RX$b|
zZ?lQF)3Y*QdNT%@_oI<&90@0nK;+&IMeU3fq!wo)LXqc$f0e}i$}e;nU60QEyKbDa
z+`X?V^S-DGO)D~%czfmuvswpgib=eNyyE@qkdq>N#1v$-TM;(TEX3PoC9otfxAJls
zz9^Hkn_7;VHRV|Gr5vfd%Q=UaA?Cj#6lCUN-0uR6T~Lm*qpMI?P>Tl?l1JvL$n$~B
z2MAS{vv$n($<Ywc8(MPlXdRr=lqH3peh(wgaVIqB)zy;Q@3iDzYAb0`p)J+!+^5|q
zR~V)&D>$#G-d2{?CB(T6Dl*nzMWz#PSEQ)QXL~hizg$fw9#Ip=!|Gx^NnP^DF&}l+
z5a*%Hr(N7aW)^G8fsZP3@s5g=d{dF-St?S{UR7RjFI;_BO^lqh<nKP(a-7;ERdT3n
zIfGsD)s_(A?g@VlYUI>qf{~hRuU3?A2Nap1q9EPxsmjL&HHjeJ-pf~&amvgdQfCJ4
zgeD9%X~5P~>3A`$6ec0b7@3ufRX<aal#v4+;%&QEoXNK<O8jsI8RXuK>EtBsmNz2Z
z<_}W7R>4EB6c>w2@$5u79@^Gp%ljr+YB!*OTAn9cOL2{OD^1FBbA+1IXRC=0=iT$#
zYGOE3Sw2rvl!5oC)&AI0;(7M?Rm7~5JqnWVUyu957Df9^)L+lUkH>kiwf=!?t%~uH
zn$-61=@T(3fydoqq+ck-iShL~^Qs>CaSiCv^A9E(Du~jAdVH;HhSG&5j8IaPA2rI%
zB~WJ9k`jAn)#Pf7DsvfB<=$8BV%973KB6j_{I`7-s$$FQEB{G;fPC~jYW4<EL*IXU
z6ZVh(gS7lQT-i_!qhr}9E6Rc^IUl`Ixrlw2gSq4Lu(Kf_?|<gQ?moHTZu!s)&%>1D
z0vNR~LCU8x++JP`y}|<6{hNh+qjYFKPo`fd7TpF#;K2Mi_`ZlmWZM`R_lQQNMBw$b
zP)ykrf-?^T@ZCHBv&Z`*GSd&!`bEPcF&P17nFv_U*>PtkVihu=HZ=)J=VCE+aSA5K
z)k3ph9o9ap!?&09sBNml{j@Upj4wk&WC_mo$VNfOY#6#_!)R$X)cRzjM~5uzzY+`A
z`QaEiITZS5!|=9eGK^|cvGOIg!%eBU(<2itjEZpfTPcn`E5Vqj#pq#Jgw!=WZ+tC4
z^Nk-kRF{PhE90?vKs+`QZ<iLtVbr8}*q(?&iCZMDnTJzb9D*4~f}u7k7`j`0anSh_
zhNXJqql+i(*1m?X_XC`n`2h1v??d_S9e8xPgRnJs;l^2aCGKK<yF0kj@($i|2YYHD
z^FYU0%9lp^5eHaEhlv)lQPo0re6)~Ub!M1lSjxad^ama{m;3JKqGwLuLQ*GY26d9B
zrJQxInu|8^cA1H##O<|`xy0Lqj^v+D*~(_uuJT}ry*N8K$nRE;5`Tg?z#fpzvs^@{
zg{xd!<0=a?+@&4yNuAY;3eNQESczOBmuj*L^vU#;+3nrL<1T%|LB!kPP7*iAS+)*$
zk;a`a5=1`L_=t;CF$4bXYFBZyaFsE=U1XbmciF$FyI86?Nv|R&nYo4jm1gc~JNA@^
z#LK(%4o4{Vkav@uWDYTRl3O<^WR`+gJUc^UyNG`a_I<x_ky&%tOQY&0Q5voi6Xzsn
zIX7PoVc%CoFFD2e`0%@4GUP=sW@d6eZPi;0h_{`Hw|DJ&i^BkjiHgW>;;r$we#{mX
zc7(V~U2->R*_WNq+wCQYe0&QhTk$!~{GNmKe{`~u55<;J9bqLoDx6(;Cb+%NP0|iA
zXLA_4<wo|AugQ?^<i+0+Z%6sKF!Rh&R69CS_vs+JH?TX4cstUthwQuLB7I)c*JWTQ
zo-b|0h?#r|#M`667BaV=5qWav4pDP?gFCi%J;@0(i?maXzG(KM2iAdEX&HKQGFnfZ
zoAhMx1Rd(6srMb(N*s^0k_+Xnq{d8#`$5hsJGI50c)OuMTihD7<dcS?bf|4W+hq+{
z*uNHIPc}h0q#50ZG^1<#X8bWyl<}jLIP)vZynR|Smpk0jx!hM->B`$7y6j}um2Jw(
zvhi#I){ZE`RmUo9FZqRS0l#r`BQsreD$pmt0M$qHk@zeF|BlSSS))w+CXdvc_r2LG
zGZFeG6W-5>w_oG2U|}2@Z^s~deH=VZqo~`buHQErJEzA$)h!mm1+jR(JPvMg8Cc^J
zhOHLC@M;?gpN)aIm>i0Lv_u>z%f!YrMQEb#^57|YBYM>XhscjVQ)VZGs@(FSZaPy{
z%$?NbpF<jw^ixBuqRIbiHsQ_Y24tj^;;KUtY^@4$vAht$=Zg`Uk%=}V5|E+Bv-SHp
zd><2s==Sm0_bVQX-@YNYIs+}{7h>XP?#>2R;cjg$3|A>i?HFZwwoXA(=vBPfRYT5d
zGJozrE%~{)4sK5LZn&2qx3(C8)y0T<R}78HBG~I`%NXi-|9DnHiT9pfhE?dsOdVxr
z?I`3aOX*q_`IS#DzKsfhC#vGi`E8^dv-Wzb(yvL}GgX(X{}F%qcw2jF%D|Z|<PbeV
z@h?^6z%3P-!a2W9qzbdY$u~e%B71TNT&6Cj-20lCsmn=4btxj>o6P4Ztx82!?_wtW
z6m?myrzXFu6s6-nMLF-KApc}3O6woWazj%^<`Qq$2B}J&vXXoz-Wp73fEn@j!ow7d
zdYFXSualuvm5BDkn=w8^K^A(bNq|K>o(D7_u4glv?VI4NT8*8Zs-c@#h0<1~@Zh}H
zaeO(9PuFAN;wHTB_#0=Fh_}0nG3@UmRQ^?jkQ1et^G!|W{;tRQE=@2tQIb3FLy_C~
z2?jNvaqJN1{R0Ik`||@6f2Sf~ST=fFXTsEyXEMtiJe-w-Ds9eD9g8u3b1}Y?vwY0o
z_n@KNYtn<Vg_($Zq#mc~OP#rnJLo~=vYl$NO6fQ17b(aUzAvJPx0*FdvN}{r`kf+=
z6QCe(J-J)hr6k*$6{N~cQ5JkuloNa(z06XUezh86>d=CnNsZXV*|xH=0aq_p^P2M!
zy&xO9XR_eAI2-wnx!5!<2g5bv@#=akW_5_hx!d{ZHYFc%A$ch29*#rA($^bG;5DZZ
zuY2+9J7?jdYC4X+O~#zlvG_A76g_{&V#DxQtZ5aCqg|rmcKs`Q4-Un#jUlMHAAoHg
z0`P0FKPII6QGe_Y?VUkz-W80m&%%+mEfc|~Qc>-k46m?IJZS$FDV_nCf5H#fq>g9K
zdL(tIM|65JhK6Tg8}$=wh_~Yoe8<nv-;fs=kIj6KN6t$DR>YurK{&#<g`oe0Fs%I=
z4u|d0Sa~lE-GY;$S)2jwy#?48SPJ8)1WY@ZfaeC(cunDMH>3aoV}GD+Z5CQAjmMEa
zaWGsTgP=#@h@a?>q27L2c=iik7Wv}JcVBoZf5wLTAiR6^1$`cT!p9g-q<naYk^elz
z`6myM)a^d%wC<qj#2sWh+{Wj*cM&%AF8(Fnh8o<#4UIc!lhjdM>HS;(!(84FZ&x{6
zh-)uq#Wv87KHoxgJj~eT)LA+(r}eX2CsF*?Ni^TE|0|w3D~_FHYDOn!R<j#qjD=|A
znn~PY_EP<ShfBO|vyVLVY<54+?<!u74zm6bF_*aDf114KJQqoCc9H(%gSDAe^?8+x
z^!?+`ekR&BNRKUD4DsVG_Srf&`O@Sp4j-N6PNI|ChZ8+5&a&#Qy$s)HFE2PhZ`$i3
z_vjZm!c4iz*UV(wFB=)(j`{XeyUUf}9-^D)BtN#fh?lmzTsY<~D=xT;&Ie*|+ivu(
zG7s(q^H5D}WK@8S{O)HhEq%Ml684cq{OTd9dtBs6t&3!paF%9P>-te{VsXz+UM=q>
z9`kxh<(yvPx~!L2P3<L_YMgBk6MLO{$^7lT#Iqf>%%NZh79`vo;`9zOEXhgw899nM
zwPgp_+sl6TF|RYTm84VbfZlB_r-H3zCV6`2NGrL>KPO(d74s>~Zy??d^%WVS#g5l>
zH|g<!b8QJbP$oOE>&#KkHabYl*$#5v&5_=rZnCnun+(3pToQVwFPd?WPP{$dhux-E
zJBrmm#`1E4kz_n*M@<v862#lQ-t^kn>oc!VU#7?FOYM1mDLAhuqr>&&U=01I4Xxy#
zzvw?0(Mlc@Z!_MpS3^@rc5vQ}FV>Qw#M@=r+H$9!_^H`|m0fBui5kIsJ!?@PUW@S9
zCR}XMjDD}0aEf?4#9c{jDyTs+){?Gkv}HfP?(Y}OiYuu_x^5|MJk3DQ`dp|CEW(1S
zGB~!XK;KUls5Pj>l&_`ec_|084rk(PKq}S^$w13?nfTBq6Z`nFmwTjNk1}DgITfEL
z^8WiM7B`5uf7?VM!!;7UpGINM`)GV5-r9_bg??TvLRZD%NMHtD-w(s<RYCZ;JP2+G
zMAc-T*LS3#{&gOno+(6rC^e07HCTMB4rit+$i47pXsXv^!(auuN$yj1jykhcm?>MT
zAwMz|rNX}v(>F9=%oJi=Z6WGT7El~t2)A=Zh>FalW-|c~iMK!B#KCoN9PXLKqaX1$
zIE@(G_dBlsmxk<5<=8l&3h8(1vHW2pW*=1&$H&a5t7*hZw+85TXu!*Xbr?Cf4vTwK
zqQ<2JFMbuHU4Ai?Z&UwSScLG1Lb&)ApkE;U7h_BDdsihE%x*&OX6~SKm6>bI+&tdr
zU$s*a(?=@uqC3wI#M^7c+qS(_WhQ5-9#-nI`=q)gd{CEthUzl*o|?S4uOh$QRb=O5
z71>U{w=eN_N*8j&eN?1R6gyRjx4OjJz$P_WOT4wuP?ea<|6@Am>O4hx5Xn5fjf&Eb
zyYPok6l6|!&hJ~vsV-^8g?r5y&fR7^MQT%Z>zL<5j`K<e<{$Zv^)?A8(~d%FWg=3p
zCu3lXSoH250aul1Oh5h=X2cAmJq)VJsDhhDI`qFL;-CJx_&0=l9+z^MFni%iNCku*
z(U7tt9NfsW$*>|+_b)=|r&=r~N8-Ay9wpDH$0}8n2#-+QCf-K9^G4&3Pq1`~gY(;O
zu>6yX)~_<5tdfa|rkS{Dk%Ob7b8y=(m+yf*G#MA;J!h(C3&?@?F2(b6^qS<AAg;Ou
zt^TgT<F&QuU0sJ6#QZM_)i^|c`qK_&2_@b}b7$gniTW!8C7DLN-ThWUZhI(7+yMp2
zwo{NNZxrRPzm#MKGay{sDAG642%puBJU9M9UU35kovTJ4oqU`#&BmXNSvYVt3mp!n
zAYw`qE-S_3F*%;+?W5sp7X=OFNc>I<<Jlw}XKJGmtQw78{bLbv?i=2cC-1A54r7mG
z6g`cF<)~1+y&nt{{a|DUgyOVMB*IpHg$nW3;eH5QuLfX=ZUD+x`J+C{5ADYC^V>j7
z*%6G?2jOV&{D#hBQ($SI42|ib*p?ZFi`xQFd(01>ZG6$)DhNjhgyZ|U2u$t~fu}Ly
zDC!rEdu_hrpQ%Bx*&GPBKA)hs=sl7e-eFJ7TUZ?O1XcD}qaT5R<>BZ%BMR?VB%}S!
z3@q!FjM-xn@h&U@KZYm3q+J~1(qfSq8H;&~<1lqY9BMS;VYV|4$*p40abYyV)B~`8
zzaQ*6e8JWsUg-6o7rb4(VKLwnCYygjHu2W@mM4ZEdkBw_50QBG0ls<NhE<PS`0aZO
zE#KV2Mdd4)*Y_@dmfXg_#M|m+C$OD)FqWrG#V*c7R3FnvX=x#e_7<|Ih<UfvpsLL$
zhR$H`Av?a-cJCzK%vTwGqZ2b;J4q)?_FENp;(Lyme9T-t2bsyuL<@;jw32@{nN`2X
zR?4Q>Nt;PsMLp4;{3*ShEgfb1AScNk?ksEKUF1#)J6}dR$*gWp@}aheoO$6SyG@<h
z)#ogAq0Vx5t(!CyILpb_&T@rws$DN9sSoKPUT)o`WVAgysi}#!bCL5F=F(B8v;4KV
zvwWXoA%UFXI$yMsmk-#voaH28@15jRRd@M|dGLDNb-GcHyl|k4*e<t_59Gkx6K{L@
zuye)MntReNa&3a0oS}xSm0vfRQ|2Un|8bEwVNQ}f#7*vQbC%dUPO^sl_78TtjQgAO
zuO|83h3*o&xu+~U-czoR>?I+mT;*R%!NbaW%OfwyqD9VfK-)>a^UM*y+g|Sf>MGN;
z>?Aw2i}X8eEv|E{rR^sx`Q&fGd7Qn|IabVdv=wh=jkMjtEbC(;smjdG^l=qKW(A+9
zb`qbwuG0BkSBWXJm8{_oV%E)3Mn^l!sk&~|vUQVV%Z!<a$P5Wz16klf?b2ccX`nxI
zXNJCXKBOy+mkeY8vv)2iQe!+uUkYROrDd2t`(5=VXpf$l%+{025FNQhPO{5YEqaYw
z%H{Db#DAiigx_l=&9<$?J4Z{loYa=bFPKf_O)tseI^<<lW0vbLnCJe2dUzeQ?li%N
z`KJGRG-A=~-x#%u+M`5ODI`A^xJp|hLbx{!ZG_gMMvSbZ2J&tO;*v6;Z&`rThUIWc
z{)yLHE70BVCpu)8V0BUsbcwfre@KP)&<q5(%ESd??VNc$x0lC4=$Y%*DiM1p#G@n6
z@kw`M;MpY-i<LvskF&~!S6{K%BLbN|e6JDTiapX0OD%J(Dm`{_fw*85h?iXh5n~sO
zO~hNLH+e{#SBQ?@<w!`Y#<!YUG`T8>WoR>E^5|z!t$|}*1+I)L$5@9lBu%6)qi-QS
z@jp<_nftJN9$r-yV*HgKX!y4fL6zjOI5&ITibMH?7`!|bgJ5?)u6ZEFIRs(-#}KUh
zmV!^`({Rq|2Q*A7xpS+>!QM^iaf~yUhk}@rGhNk|SUav3dN*qk;8uq@%<(CSt%A80
zXR4fHtn?^``Hv!e*<FYWcM8x@U4)ilC1|sw66faB;uJIJdysz~(^*A4nF(OY`FGm)
zN^I#@g$tq8a2xj<mZqxGnY+}#I;%@8XS@2h+-tT|mxJVtcl@U!s!l30<gSVYSgXj=
z<w~+*w~|b-ByJB?6-Um{Ev74oVL5YY(TpE6nqj!H8D=dNB$GamiNxDU)cDwPZnxCs
zS*(+yjBsv-^@L`uAl@doRFt{IyZ_^@5%IR`g-iq<{*Lh#@yJ(=!q_1Rn6)MjzY1dT
zXlpo<rhkR+h_8qq9|8Lz5!hQ5j?O=#v32M-l$qtC&8`xBwJ*ft7Zq@MTY*2<D{vr{
zIJ>MEpE=v^$;rhpzgjf-)Zue}Hin-FgKDc#BshoS_>IpvcgGvnM?c}bcMvi@L?Nwf
zH0qtB@Na4c^4FxnV|Ft0m!eQ>k&F2zc~G|cf#H?;sDG1#_Zp=*daxAAk4n(%crgZ^
z%)`Xi<ybbAvn~0Cr*1qS`J`gBT@i*_D~m>@k{nB|M_+>)e6OxT#u(-&?f?JTgwI`b
zGe+K0l%d4iv!|5h63-=mm4ARi<#_q96pN4ZOu9T5`z^BZ;$bFQ`)9)aNCqbJY;wgb
z5nuSayWJ-W-|k1^49kPva*0!ahv95p7+kMJVk+_Wukn7!XncyOJzg-h^oHjfZzL}9
zL!NplR^1B1>@`7fIUEF~z(AbU3&Ms2fw*54fXKxG_^A|t`+fb<XQ>|s5B0}S7eAc3
z%&)!dkHORZFeA|qmjVJ{R272d1;OZV7>xTrzhGf2U+f@vyT*pNd^iHi#9ODRaO^M(
z$CBk<7<TwQ9-F_%!s>T$pY<Mhh_~HNzs1jQZ*er<6Mg6VA<#GidzM9^j~0IyW{H@R
zk$?_=Ct}f`OW0w15m!TEVN1Nd6A_Dx>9I(e8;3t?@i1u<2d&1hu&E2dBBemQ-Q|lF
zoj=2B{zoVmzsIKQUf4sN|1a(nvgSO-qiYW_chy7uq5k;Mu?NVWcN>q|-$JS9EtK2d
zLOk(y;M7Yvrg|3V&Ync)c_&c6-#{k*+eS`#8Zvv+Se9O7p7mz-HQi%BL{3M!#2$&o
z>;Y{XWGWVF9i$?@gIuqmN92L2EUxM(E9f10Hr7}Yv^vYdC1%p?kOegdR<eiLZEZ)}
zh)xl^MM9W+n_w@0U38Gaf9Mq=-gcVhEGsTCd%W0LULJFn3m<Idej6K^G1ON2OlFSV
zM`l_z(1*09yF|n~ONuf3MB4X|rzY$P$#jr?BiN5duVOCs{Cd;Pq<5&9Y^dubyKRi+
zU?=u{sB{u=zpTt&=`LqGihn-+pZ<2z8l0~$I?8A2zus9|O7K4xqBqQ39$&VSH_X%R
zFqN~p67vZgnNPm6n>bW-m&996(!s?=+75M;3*^}!@Y*^aV_)`2_Hjqj<37}k^RT6?
zyk{v>3oYf-&aTql+g_HWc9(s=UFE5lwPbFy7Ux@5Qo5R%ue~|DGPiyB?XJ=sX)C#L
zUE~nCWS5!jpSi?NuRFY^P2}yXJIg+13@%e5Ptet#eRq(~?9ono?J9j%xQZF^b|vw)
z{8U#7ujwM;LpsT<EtYaI&_)zrbQ2ZoksIjCZ~jTHcqhGxmIiX1p2=?Jt*Mn_);DwS
z3RfA33-NZc61@|n^+jQ=zBrLb*8iv{`P41Fqvm+XFKu~rn)|_xT2f8C^&F%o3(u;G
zYk`*Zjiz4bg_itP%PfQlMS1?G5sMFT)=H_y)4soum-7qL-`1fvwGmJA*rP$cljXhN
zc>PjU;?C0VnXW0Ob*l1ukg8~3SC(ts%TE28jddF{;IJSWStAp%q#+qg|4zY;7nvAj
zTF6Yf0vK21Ag(?GSMQ}_<FFLmsV5(_I}U;U;^6*<JDvS8DD_CdrVa5p|6eQ?KZ`=Y
zClT<w9)kHfAuxUV71M~f3B=oD+}&)9O~93ZGtm25Ffv*PqM%(MjDiEuODh=Go=JFn
zhFa&C61a3NhsV*MP@PDOQ)-06T<!slxN~Z(#v7k1Je*Mud+KsV)5o9Ptq@y?w~MJG
zF&&(PjZ?U1JpBWsUuWZ=OIbMnPX-#F#lhPt2EG4|M*PqKB$)<cqiGNt&xhdGjv$of
z24m7&a-erB@yop)e<?IVVIXtqW>Tv|yd7Wp8~q$>&~-o!GdpVVEub1V!>f?ROdCJW
zRqL2(^Ne`1lz7{Aa{*TC7NMeF39OftqIOy>d<RqidHol3mejyEwFb^!6;Rq)fl(VO
zae8bu){>*vI;tXOEveBV=8ie0CQUhNQlrM5;Wahs{Z&c+W;Q}+;_YSPt><DTsoSh1
z{lk=GC-e6f5O1@mDM)m7GtS&-#QBfE@oRK5j%$+3wo(x5NCkO3OObuFit;M09v-2;
zaO=WP#8;MJhC@9D|D!06s7o3aT8m`AES%>o`I(&Y{5sB;FC+2a&`2Cgia^$0&SjUr
zVmt9xJ1Pw8Cx^jwdl>F<w->7sjZRMAFw89n^Z6Wf?OuR?JhO1_uRMzXsZ&=d#GpP!
zIFOf*C26@Bu_Fh)wk5&j`&alM4MU_tC_2svMlkur<ksHsDyQc%I~aN2L(mxb73L06
z_`5L-Q>(+U_e>;aSViIcmne)=i^fy0SZsSA16}hpWIaoQ%l$<1uHW%qHw`DZ=AlfX
z0K=)XJzP<Oc&B2RYNx`Vy!M;8Vzj!zIk<?rI%*ci{j9>|u}#=V9^iUaGvd`7(Ost=
zJ_&VD=wF9rS{2w#PwHsm?OL}|lnp7sV`>{#?M#PyLpo|7XF_p727(GxptC6vAFjut
z%s3jSUPt1MW&~R2M4(gOFc{wS#q^c$aKrEow#B}LTFq1V&-8?o>IW>)_J)VP4>H{R
z@o7>3dL9YHCB+~F_y%I#tw3zp9|)7$0Jx72fc-~*taS26l!_l(P4UBFJ3s7O=!>Kt
zey|zthbiCuuw!BXO8GkfPa_02SN(B)?iVai^}%cVcSv-93%xJTaOLR}C|KS=pS{;$
zT7L;YZtcS8Z@cl(YA>FaUclb;%h1qyjK-I5;T7tM?c@CLXGR1rSw$jpT_Vnn`G#0W
zKHj_AxX-V@YjzP%-WPekzJi*kSJAKZ8Lq3mz?@6o*v`FkK<y_iY2we>)f<m?dSUo|
zFHHOW0d`665zzGmN~gX>$(P4a?)?baR~};U@`q?Q<RNU2JV5)HTll7P3-_FFVqKmG
z8s=Zcma~_U`Rg<cFQ3G*2`A8(czgc2f%IVB&H0hc?0#=72iT=lu%)9|yLA-rV;$w%
zsg82^v#CV=ZZD^Zw=amdvj><;hf$`|SDC%lQ#(jOFtd7Yvy-dQj9E11()X@~B#p9>
zC%Ijyxv-Vq^aR;ob&$nn%wu+O6cz4Z-Trfy2Oiw>rdUh6bmCE^wLH;fruktTQTt;f
z-~VMN*g<>QwZDhlCEhv^Z%6T?pG_C~&^t-79evvLkhx7WlMrTeJ6>d;Wr&fSRO=|2
z#M{?XEak}wYNz*+Lyd1QngcpY$~iOk=sC*bBpd1IZ6P<A38S^mOpY@PaBICeGkUEg
zRDr$HmpjQhosQDX4EE2{I?|V6A-DS4$zl}?d6jHJ?}xE0=9!~?L}z*|&E#b-a|ya@
z$^2hSDSXmZhHBc&9c@S1c&4kgd2cO4w^&Oj;;q6;D{&y+wr59Oc``j$`Zn^4`(o{h
z)?&B9O8V58i}hJ^dAr`6`jgJ`tfsR#R9diK&t7^{GxcvbR|#-)l?#2Wq{VF$NttLQ
z6O0XIe5#F%W#--9&Fp-upzl7QlRTeeD&D<~#g=E1{f+EmUZ^d@o@&d2^R2||vW{re
zFS$BGSL|2m%KJoJ8AczDUwh{N9o3T`^hlP*GUKmzD;ZU&E!M}WPzqvB@OKRv-Aru?
zdB&V%_E<$}$-c6dV!`acgX&81qZ2h!<XSuZSB=9xe?f@1b{jbF2GwJc*)NzhS3tx5
zH!f0Jxr(#WQP(ouE%=Ep#9M3bvDVoZVJ`1`k9Viy+=WCuyOe;(1LNVBo`}0Q$gc)u
zA+uE$qV8lOCOQ??-BQu=uM}uD#N$8W?X13Wh*6J2%UKaTO9tc8s9;PU5rdLE>iG>q
zk=Hc@o{xhe>Hg#+1JK|S02~R2Gx2tD_YAaA3q(T8K)7oJ!hc8rX5J2@CMOu3ZiOMn
zwG>U(<%nW_&RXuJuPv*^uXW6fYgdPurOEj5G8~bs!;x|@0_RPlF|kcFtjDAv+B*d+
zh_`czw<p%;(_4^@gQv6LcQpg;)6?+HECy;lqfyZ+0FU$n5q2;T-XDW-wQm5@QaJ<6
z4aT&~m6(3F5Z5OD!n1x&P$8H7%BmTk>6aMH{GJ<}Yu}P1%|B3x<9#@XR}>+78+W*R
zoO6?luyHJRR>b->Eea5Kjq`Or&khMCa8sb>^l2Gv&Xi&rH9fDX1MT+UC-M!M0pd`N
zs_pb?7}R4H^J)90tH_llYND5?CKlv|wWtqoPw&!gW99%ED9DC(3X-@;NxH9Al0KG-
z^0$Mc=(Jammc5$Md43cAI5tA%#&2BP#Vnz%^*FV@2^XRjWa=nIIkJ$s3`I3qF@brz
zXMe(uGkDM+`Z}0hF!0|zgk=`t^|5^XE=b0yV@bG~6o;LsB5{9MB!&}jpA&CGR(yr(
z_E02^48xn@VXzoS%%!gM#;z#5oEDFFJHO-R#5lZKl>mM9WGwSff#Jb4Ja}AyUxy3u
zer`T=rsTr?do1P;h(#yj?VUYg2sjmpp_;xh*7}708$RGl=sS#8`+$f`q0kx0dB}v1
z8A2}eGx65$D;BPZBqtq(_#aV-;hcE*Z4CO|js?Du&){o%mTDY~*Tmq*%6z;D`+=eX
z#jr^#!MC=>$Qkq<4HNnCq!^XgsK-q#$FubnsPn7BTyhSVy49k4a}!?q|Au=*ExHhI
zv#2x5rRV&rZUs!vP)E$)ZF}c5=qIFO=EpRgDow}AI|<nLAr#58gYb%2v@(qvrqC!1
zm=HxBQ8>P4M!?zH9|ePak)7^|4gRk&f9ng}Jn$65CVOIA!Fz;G|3GZ^!Tttc?l62&
zkl=&c6MT?6&WG6#KG<F31ErWRI1=HD%`f~huZKVW$@axn&dJ}(d=XGVJgxM_z#)Ec
z*}yI3cYo}w4Z>Cff9xDf9G?FLTfX_AwhJ?xlHcOF*HehW6TW|Mz+uTX47R?EWjA-h
z{r7I@Ih}<e@pj$3>*%k456dpx#^q0*uzyKS)07A-v5kb+n*^-s{SA>LU!g+n4(xwj
z!s0dm!GHQ?bU1SbgLYiSt%29@e)mn-5pOqEc|+0lGfarLK@Q$nx5*2)UVVf^?FV%H
z@*b-5-l31nTf}s^hj;gGVSSPZmMVEL+v++-{`A13#%nmw>$#<M0~%AV!*uasbT>GJ
z*p`RTqwju19^Z)pTMWeOkbyWTw3j<^Mv})_b@4)DG5BCY9Zg4>dB|8)rWs2w<`%@1
zv=?7y4XrrG4vBUq($3aI6p6Plq=O_YvvcC5nHW+_l@V+%`PVrIj^GYAtBd^d>mrJ0
zZDi0M_I8Vd4A^TgTVL48Q)aPe)mqEVWNTS|Q%4@T=`q(;U#{(B?mY}7m>v`#{=cP@
z?b-Froh&)m_UCv_YdXrPkEYUlIcK)y*6e-O6Pp-)nfl8}RxLIZ1NzFm+>NDsQzyA>
z*O~nlrZVngdvV#+Mlycuuz%W04sbVGxXwayhIE!A`_1SDG?yC&^c33}%Ov_wmn(LX
zNv+u97(mSG*HH$!no8T5rZmTPlt$*hO<Tu4?-cfoDwxR6Ud#);Y$^)6=CaGioc>Td
znNZVJ#<DjQu{Pqg%UUkZGnd)yKz;9SCRbw3<UX_MQj;tty0@i_TVW~n!>lE0IlC$o
zcpa0?n7v~rr-|iL*iU*f#6n7@JIjsMF4D@H8Z$or4@Kr9`Y=<^*hp?!8%fWZ%mj2W
zk@n2BebALZt2g)S)0i>GjJBRnT1%Ldt{fkyExU=gTaUI9{ok!*7QOaA+;l~4IsG_q
zbeVOhCv8J?C2F;fe6C^^L7s-_=(H3C1!e~wr2n<3g|shMl_D1vDXnL37B#ni)No|E
zw-iOYKQQ%XR_W?`tX))x*SuHv{HFp6-j!HlQ;k*|YLW4%2F2wScss8Y!@6gp-+(N<
zNX&)qsT?>Hi!6(jq<7mQH2=y+mp7?M*hAjSJqp*|<1v|dn;-BUddIR5O1#y-k%_2d
zsc3%l9Tp1`k!Bx{`zIsty<-4&D8*q!QwX|k3BWb(kuti(AnH{xe2KFe*@0O9#2<~J
z{y3@<fJ0pZ*s?|3Z9|M55{Q0v0nkwn#0-8;?-~H*vw=8R7mRkkVd(od2RZ4580=Vr
zE!<0&PWgpDeQPjnLpoG@CL`l%JU&mS)?!H%Hc*pgrX7v`J7e(iWD5M|WuRF<7ZF4A
zvGn0Hv?=_EZ6CdH%PjyWjANi>7lf9)o_=lp;k_sjJDvrR{|-R66>ss<!w(j15-``b
z63f3;qR_I6hU^M_>RW+|$@Qo(&Be`+Ik2GKrR^x@c=Re{7C{y!#Af5O<X|Q}kq%vQ
zVcUh?CzS$dkxwn~rx!E41jotU4^=M3zeh{4<`T1cI+Y`lIR!`BRl;#!6#|I2X2jby
zFO_9t3UylVm1OG~dZMN%$@Em}c%Iha$Mi~=ylTdB6Lz6VGcJx&lJ)!OF`Y?G`haFc
zd}A(<QWLD#Hz4~u^9t6~!F)tL>Qm})quVbSysJb@g-U!4B;Nj0iib7Dm~@a@x!O!9
zdFNoltUQ#yEP~djOy<(2Bau4-E3Y{C5ZAH-LolLc2z<!{1Rf5<u??YkQ_D<(p~Sv1
zVfZ;U3~f7p#cIx&FSs9nFg%Xh=5J^<Gy$4>Q?Ph{8eR?L_pRsibu9<&2j${QN-XO7
z#iD=aSEO$W!`-I=P)quPRhi!Sjt^LR=^YdmK4Q$;kI?<ZS#Lh)GnG*M2?;@A*RS|@
zQ6#L1w`Ve<aOD3pF0})%Jz^2v@D1|<6EOP^Im}5ha3Jp)H8C8+SA|jE9D<yy@wnEM
zf+?5j6Zg+VPql0eV;<VObrslkz6yFhtDr`$-m9(+*n9goOv{*!PjBlUjUNc=o{xc?
zaYwH!fk{m%_nv7O>XDAo)aee%PQ%vZSadN9!Kdj#__CI-@5^7&^A>eR7vJFpg(WZi
zJ>lZ@3=bWjz+mMQwAuR<RU@Ba!H%aWHGYU2OQ^k7xsS!EpHX|~3%vDx(Jjjd(Y<|e
zQ@o+m_XAFEeh-yAAD9qt$8ZMrHT6e34PRL7_=Nn|pJ0;w1%cEZ-5cPC<#YXTEY=^o
zPsvB02*j4<zIe&!_;Y{{^vvJEw*EDic6f?__K(rg<00nkxPS?1XVLZAE;Jq9iU+q=
zWAoju@X7fPs@JX|@b6pr*nT&9glt9e)6JOiVJr47-;I@lyYXu99xP2hh*S4YvcvWW
z3}zm~^~7VCeex6<-OoYe_c@erzk$@t4-nSw3BClrKtUQa8V|g|*mbY*Ap8v$e|nFt
z0gtfY+C98ldKZ26+``IZ9#GqV6TQz~<+=DOGuSVqW!`0cJ9LgZxN~T^=rB%iJBU@;
z2N2@8AKmxvgwJN?oDy@R=p9hAWoPiU7IK^!2gb`xWC-WGF{cdKTf**0W**oTwinX{
z9pvZ{BQa?1AiFMhkS**Yi8SjVKZwO654DrV-)+USuC0u@+eVg5;Rmza3|AP*)=Fl;
zR~yT-XPu<xZAa-lmYP>pJ*nNVD`qk54n4`9iynG1ptHWLU867kKIx0YWCLmO*+4dQ
zvym3$r!zKLvqQ~L?iv`%Wp-gaXU5v?53ObOOzs%pGn+Qcfc;r*WpX(4)>gHX<jw5$
zQZkgF8HN&lq`h<)&hJ&GBS&{yiT@cZcBfd%Tx!8a%`}$qA|shQh%>ycu?)8|mZU;<
zFtgv|1F>|cvx#JenaIiU#MDG1>G8x!W?UnOo@gvxLX2gkhOvC;W-L?4Ex+DqCXt5L
zQuNMRl9zOm?OyCzHL(_r{p`ZpX+|HZnRo}AiCrfPx#+@<>Ry)8awofBtS#u5GMCUg
z6Ip()y;vVLl+Nsw-fPoNl(Wp`(PU@tmYEUKwX;m(xk8VbtYtF|Wj=da#L7sfHt34o
z=nm4|(MYyiGyDB)8@a;lTAOPIlC_6@(GI#YWu&&`u)nI{L@PN<KTpTMbmXRyu1s8}
zE4^Rqih{qc3<%H`$0;r3MSl$$#aZ@bjHWadx0HkXwd5pyu6o4V0nRG)f~v@bp$hWl
zdLxd^{DU_N4fsQEMCy-9L>y!OTSf(<b;?oOAsstSlF-PxQ-4x4$}UIX(YF#*U1yF&
zpDfhOWgg7N9F#lIm$xY!6a?dL(Kmd0nTj5ZzCn$<vEll0Pz;H~m_ad!G>JirYtfi?
zh5p%vQSf>jiQ<)s&_DDIJ0?Wns7?U3yZPZsT_BEb2!QbeA7op_;E?_ocrN*f`2+p&
zm!Ut#QTJl^+#AoSi`dvJ06xUq@Bx7s?;n6m#(dw$dhx8_hy5%0Z@+?({X7cg>+&%4
zNCEWp=(&&k0o9GUIFOcsDSxNqjd>DAhtor_^eYA*`HJPfk(kS~f-~{f^hGkF|4qTs
zuw2Zv%EyXH&){+TIm~x`MB~8#l(h7Q<8v=${Nn@L*8W)J@eUd@-eOVJJ7(X!!GSA&
zIQT6PYYW1G-Vul%7>PbRf8vZu1+t>+uzFbz4tnNbg<md?EMh12>r5>5$)aaC8&9lq
z;PWg8cP(@AL^&Ux8Jw>p=sVn>hn}r+(P2+6&Ku^Vnck3VD}G=lcZZ>yDlnr%CF1r}
z;=(_*$iK>dv#s<GRdF{<-_AfiMKNX0{BVaFbkD8CvC9>3q`qmfVKe5mY(mDSKM0-A
zdG~S?v|XD~9p6N+Nh9(nHy~mkGv-&;A%hy-9@HdV=DB6v#Y&9%U4gN;YM?Wu6uEzj
zVQX7}dGDALRGvgGkz6dfP|c$Mdq|S8gF2g0H$w4<`2(vGx&PA(!D6pqj943jc56b>
z=3FR>K82#URTw5%gyHZ{;xD=7SH7HAcgLdM`y1+dBw(oycfUb#*p!=r>A6{GeK`l7
zw=%H3KX(nALy;Qu6~V;Y3-<$1{pT|lat5lk`+&fi@1T|c0flQm;`JOq?6C_(>AMj0
z^b5fh>VHxOMB?P!DEx_y!fL)w`_bpUkbL&sTihXA#2~zT6nuF$97t_X^-yvv3&XIu
zJ{T>zi}p=Qf%@eX#0RHf=f3agPu~AYn+mL4SA}6F#pqa4i21YYaI?7{TYpsG@5hB$
zsLl*RhkQh~VE#AzYIatYA~P-(W^2>o_=G;x<0<&rA`Cr02VmJWKP>R{Mw0PId<=Py
z$p7BK|I-_6zyBKhQlG-W@Ch`hKY_{4r&v1t8BSh&!i?mH7(l$Wop~1nBR*r!<u4e-
zJ*!WNH{O}O!my4nVdeB3lTDstmi1#8^m>VyTJAk>e}PZx7u*l=g5C8`xLDzXF}Hm&
ztgj!=&-BCa0Dn|h2BP(n04ycmUQYP}LsKt&(RhawyB;In>mK4%?qa6C2V7R1!oAbS
zk(DwR1KQ7lj?YZIzcdw5{ib30sW~t$pNGArn_>KUE1Di{#%8R?m+7l8rf3~T{IdrO
zh_`<Zp2ToDf&r6{VR8I1Y6VUsN8>DdXrD()<PE&|?*Yy|e}pMUPq1eCQ#=~}6i()^
z&@}rAefhVb^6U``#@;}$LpLyUE9cvT9%#4oCLW!<iZi>fVrI9?aBX=Rx0J|PTbzUW
ze<v|&+A(CP9>pDpRhYDSCser8`@@`)2^Xj%Y^x!C>(nLpx`w2g7>W^RqqLKTQpry0
zMa0|5<YX<MwwLfT%p6K+FI`V|ki3N*n3LE+?(Sq4ib6ZN{gAoohuX+nW_}lpY9n5}
z=hrPa5*_00_W4Hg#??sdueXyOJLns~t1H<vbmcqeshuZuWYKDESxmj^!N!&{of;jp
zDF$-#Bk^JryD;~(lM+6LFSFr}F+cS`&SOgK&lnWNOwz~N%<*Q9ok1H(6ZUw0ZYvtZ
zTd)7;m07eC?XK-4ojK!!hZ)FYEj_tps3Q;A;kEaIsXS2bBr|5PH~XfMbk*Vhy1Ik-
zbu(iAlBvAx!p@A(#^TMN(~JM@&^S|xi!~DeV@8r3+fI&_wUbYQ%pw2Do^MBLewqD!
zgEMY<dkc2QSjj`qwmSE1_;1$oo?a@wMP~A$b7xU}+F3TvHWS-23whtuQjTo0kP$W(
za)fw$Ue`<%%-T!iu(q<LeH(eD(N5ZqWri-Z-^=_BWO}EzGO^A`wy`VRY?6T_UTiHN
z*XzjeN@lhywvbcFnlgrcruxL&SmQP_{7P%)u+!tyRaf2((vm(MTZ!S+R+3QJN*cQ}
zC-E!uSC{C@g;%<g;iW74#xUo+QwvFS(2(m<8j?%A{hilRPH$s=@#hxw0nzWFz-$C&
zhp%7qhv(i#tbEgecAe|daWH2+|4Otb-p(@oi4C{uuh^6Zm0IqL`oy7Od^Fmt1;FHY
zIM!#U!`?Ov8y01=UxOYB`y5PNl?|<|RJ_)RgT=LP^dj=U*f|Pi{UcCr5`wuKgYjCA
z_pYna0PkZt3!-4`&U@P>^2XUg=-Ap1-b+8Dv$G#G=K7;E|NQiZ57uk=V8rszSQ-Br
z4k2$bZNe)!RzAZD+b0-0?g5TQ-9sJy=|4;#;1RRfE?@G5r=1rR6@4*aULanT24Z)j
z5A?U>!Ta9=yxNh0XwK)|mSn+=cza_+I!bt^@D7cK&%&>0*OlD!r$}sZio$Q=ZO~!r
zFxw>Km5(QsCcHrY;io9R<4IlDd!B{8ph>({rJ!z5*++P7{m7lrYlPar#Y(-`aBhB$
zlfOS;m-<KiBwu%_*$*}C{Gl}_5EVNEFmxfkC*-RAdZ%D+RVw`08y5U7o!NL<oa3_b
z7tb+;<PW4{F2?@(A4z8!mF3oTVFOVy5d*NhyY;&5?rz2I?v7)Eq)1B((%s#uf+$iV
zWq^U*px=7O_k*El9DKwl_S|c(882D$^cB5gDrF{n$pAa?g$Q3TJzTO0-o4_w@Skm`
zR#Cp`EgL1vY3zy56mI;%>OH@CE!le3GTROqUPqbp^C#u8KS&w)jiqBhFfIHeT2H=k
z##;1MdOvBmss=Cdz#6})q4}R*Oepv)nzI@j3Enmd`bCw_PmC|TWZ>8ryx8}G@&1x8
z^yL%7<&LJBT1~{Zcc`?i!Y8<tcjxknf1S%v;q%R$qG>Ms)7(i(Y`c@rsX7_B<)`tX
zaT<OjQ?PH7iruz!;s&JiFE5367nAwcHJ#X3=?p!ZLF3d+mY&T)*H7MC-JY^sxaA{p
zIsA911f4w577vxXoB0D4P06Hwmvr9GOsChS40ImJ9zr*UEZr#FgTrXJD3rx7!)dG&
zf%UX#`aX=~MDJAm?Pbnhn2uH8119D?<jnmno}SC*&XhPxoii|={D{5dACWBk-gP$~
zprbB$+w4B`E7I9iol5h7$5ciXa{kDD#`n)>^y&gM1#jadlWg(EFHEX=!|9<_a%Yj9
z!0<`}dsVVX_P?rsUr{MK<OLs#I4)X;x)k8sqJZ4_MeGmAV%O{p?1l3_6)K;-j*|WR
zG={)!5j+sQ_3j+XqV-;!@%1Epk*8#-c`#^$JN+vqtD)gNEcSVc{??6^6;6WjciC9i
zl`gTaocQR<0RuN2`vlObhd<uEy}5eZgAQjsh}s?|{d5sjl|->~UL32f!fAg!n&Izb
z`LidUrh^is>mY$*mqd2f5w59wByPUp)HDjGyKfl1KZl^U&XXKVS1zSHiMz~z%hS(s
za`O>Ner$#h8xeFv*L)2J7i^%%%}qQKy#11Mic6Y@ShaF5?H}wXN-!?7|1tJ1yNb!T
z>!@5c!ZBPI=iS$6shpsC<_TOI9;fo{1?CAB8@QXwXWjyhN-K^n^p<=uZ$9*Nq-&);
z8b=&CdB=mbc6M~1ZO_#WPAnPY$mrL0jOu)cJC4?v_p|1Jr4`;=EwK4}i!)7)__kP|
z55M$S=(Z2nV@G);of?<cH&LSXB;V_K17-O4hRXaBH6=^D9Xk2#6jf{Kj%?6I*|<@9
z(R{U(H-=h@SApaWX=^H4I+{xPU@c`s9c?AHZaZbW;O%X}Tb;!kO1~eCmFU-!v9m>T
z#%eW`An}=Wy3k4~7%kdYnWwT_w^TeOpJ-}%3uXGQ7D}+>9dr}C9ok+p{v?kzQ?k#T
z7l}qkM?5CB&6G7(%@lp<S+4orOzHYaaus65N1xhEIVD*Oz5Z*aH2v31QJkA7L+VM!
zopl@OkJnHxzLA_$>A}1#{PLoHno4GphSK9_OJ$vRbLB?IW=c=VB0n78Qu(~HgVMrS
z=G*hyinZhgt9Ng&EYg+!>~@_LGoKDhh~yW>wU927jOI$ec+up3s;BHtmK-$c?V6UW
zsSIhSt-NifEjpu)N~7Ohm3sraDHkfF<20+c;8Sm9<B(p8!TIjWciA&1t9Df$=yz3S
zZWTS3q4dT`r`hZdTFNo$b}rl|*`_}=l{+e0%6~K3D#dbtSaZ6q@@<TU(&$qw<)_T^
zqdT=#+<&xJo<>RTsm%Dtn`$VR<D>&Ur={ZcPkPgG8!LsuYRU`A+iN4avqr(ql>E7(
zD;^`+bT8GEjdPnQGbOjyY<_d4t9A>e#kA&1;^t;b@S-N-DQc`dl0LSs3neE}tC4bO
zK?CLD?*>Y!^d;}_-$0oxkDn%Zn=cxfK>@Yg=v>3J&9!`R{lkXXpQv2;gsaylnb|&}
z`SSzwtV9?4;4!zgAB!G12kVL1eEgZny^wU)J}6|Qcy@F)loPhFj7fTr(EOcEx>Xiu
zH^>ZfQS?Ir56QI5lFz^c@dBiA?rn;k|MA?C?C&S<<Jj3afpNbQ>9{8YU(-~Yo)N#s
ziWmZXV;R~pn&q-@<7_Mq1aCLah@t3E07vErvtUjDc8NYzEcWEcS~u3Fxbpk(UF=@D
zaa=g82`NErZYes6L(zB%Z|HnG7_A+#P$t^Qfzjm6dW`k3Wa0#eTU<zG)RhMediIFw
ztyyf#&!9=y3`|D)GQ`1$SIvEB^2eRiuN+u8*@2PH9t<t;!g-e`p>sX4ndC{TjVBNH
z`cgM0gcE|dKW+sOchr}r*8>>zHITuAw-w*Q2_G9y?c+!ezl*``Up%|cC9!s!@Wm5^
zk8hrZc00)<*d{vbpyw>PUdHPC<y`4ofsJ7WyZ%+Ms`E?s3f_LNUqM|H(fM8~z<OC3
zKJ#;k^Logu;w+LA9<U?+F`o;caBkpd4tV~c`@-MkK7Njg%+56@O4%=S=i2Ts_%-J(
zw_1PTcfd!%i7$M3`ISw}eiGpIgVek~XxyzqQ)XGeyw9SkdCx1?_tc&Io+XLz>2dKD
zYe&5xCFeDHo!*mnyqYd+Kk+R|yk7&}G4|vu&Y2bRSN7^%(sL*o6pOb(6xmH97&#;f
zGnrv$)}(XowfH?7rV&0kg-iERuxOmhppsPf3f`uxC1X;WNX_&F2B^n#ZF4d{e^VIb
zDjrUar<{3`P2qIW*8R+;WM&p#^&S$wDw%<z<LNOrnVzL-EZmeqZgx6rY@!JitZy|s
zjQ2_?8A%~THwdG#=#}p|#8Hx*!rpypvNKM@{!KECE+?^cUm_`|;zY+D$JIfJcz1oo
zs2>S@Tpq`+iw_8rnH!x<Is_#X6P$|1&lFmAzR&A@_nCGmfe$iAo<0=EyxK%m4(HNs
zQvnCIS5i;#_LktSbB{_2|Gj49?^hga`jX|YMLc*}fZ_N8Y#SBuTce29g14tGq_afs
zCCN>b$hjTI+Nv0Ozx1Q-1Rty#_z=<C3wIw+A_Q+WetXa;(T%WjSGEe??%3|d&{=Mr
zdG5sV8Yj^vyI>pZLaoL<JTJO2QNH)dSWh;^c@fdogAMKF&kqR0c~S&Zi=y~&TlS&5
z!ukGRG)>;cQoTE#Rh<$jIvP*kH(?xB31#q}K%7$V(RQdSR>R$J?&g8XcMq0Wxbkbd
z4Qk#NtRJ(VvW#84`ml*t1si!hdV}Z!=kPvqHa+&Pz{7MSF6vvcpRt?Ub@sAXZ$CX|
z?xAGqQo>>vpmJp)vve1W?&%ueXY7zXiW5Zo9%NPVF;2g`z=({Cgnqt=#dcG&Ut4kU
zuQkPcZTLF)4!>s!XKa6w(RFWP>1@q^HFl&sJCGo|+s%T%E$dlfSl5c4Z!Njd-;y0G
zEqF4@6q_k}l&I;lV*d>?Pwc~H=x*^CNN#tziek1|RrzvMRT(9_cuSKKYTgtwr;qr`
zzDoYs+Iq5|Yo~Z`(Ns1>3U1Akxpik7<#kRQWnP@75)`4O{0wWWC_gon)1|GHT5EO1
zLi)Q-43SK^tI{{(&`^21xuK%EzoAlEExH<AHD#{&^9O}BRf;<|Q+~_2e_3YK6o;nD
zV8PoyznUn=-ZfF&q;Jek@K(d7nKIq1nPMZka4iMDjRb#>%J;jZh>quEGv)1a(E$Bv
zrWhZSoYBvs-LY&VJhy0YH)tqzCF}ilhjxnIRO!LGt)WCpuZW#Sb7i+=zy``5^Jo25
zO8-e6ly?)`EAv-L4^E}Fa$kB`vre>7jv2|UC_d4w+&W5Dv!+VF%T1Je;zzgpR#!PB
zxwEfi7kuGtE9I<UyAq?J{COw&s=@)6Nawe|yOy$5^v_lAdMoPEi<#irO?k7ro8(W4
zj!Wih^Gwk;nTyseT3hKac-y6=y0ZIXOJ%}!c}&UcpQYVSvdtxT!nC#WN>fAm`@WTO
z?Welr^@;bjtGZG?R9)%UOI_LASzVcC)kIm6-AD;*QCB+6B}djoO$isgO;1))nh%sr
zq~+pmoYg>aIMh(_JEW!*9Fweg!P}W$a(}3i+#kvRX;r4CXxCFyf=4!1DrU)JwQi)m
z+^?c&%bsq;G}+Ty{$)V>ddl;_x=QV%T8t0>VM2S!$7x*4%#c6WMEqo{<WTg!_6hsa
zY7%|R7$lvMtJkGUCUGK9TBfk0Q3@Mg%G@bFy%XIE*?PQ~&8B5M8ehg`!P^#sw=P*(
zG&}eJON)opcs}IIu1pfX-Dj4i=!9QJ;^-MAJCiu3zl)=HmjvD{PQY6;0@Y<<{GAZa
zkEJo9EsUjci)fy>M>2P{Xdw0lGVO~$7xn~j*U6e$Wmaq-Z%xuHYhp93S>xtP>VmuM
z61?3jc$+*wfJaBHunP9$&Ogy*D*i-w^%TvA13SmL;x^QmqN+%?UQQ-LJcSQeXAmhI
zZ|^z}c^{p@ha28B>*q@Y!Q01~-X#2UplFI{K8D!w@0~jja=du9(^GUHp3KPc;EHf~
zALjUS>Z?B$S$?Pl`|w=w_g;uUr(y#!F7xB?C_f(b3g=R4B#ZrGS!)(Y{MjT1Jx!vy
zXuJ|GKcKBz0Xaiu#`S#8<UeJ2JCw7%PX!nLtKjvp3XE*Zcw$k8s%AOwjU-#|VgUsY
zpK@_-4%&_n+3%K#YwrgP5pQNes%X@1Rny~B9<iqjaf*CSi%F7cd8m}d{*wLC@CCER
zyy0^54+N!r;M1~CoE9%@uIv{rtiBPu;Rg{F)tv8KjcLw1x(nV8b`+0_>=);|N=E(s
zmps>e!M01U>GR?(As?#Is`$v*_SG1+dx!JCSA4cC<Y}je{8ST9=C3eRe?)RqxZL71
zNw}D%6ZKPOwnWKOc$C0`-YGbCN?}_{ig1ppSjev{n<R2ZG_pq~Cva^=91k}pbN62g
zv%95|6rM)!gV~&Y{)qgq*`m42B2Qkwo=Zd<B|7nyq7#2t8jZJJG}HP;^KMoYjWweP
z3l2rSUnri!f!O^HWo?TnDs197em;dd%SBVZIE@V<q7hh<#DU$3Oj!`e^TjdjD34}X
zSTs-n#WH!E@T@Ij=^@^j{T1;{@K0u^ZYmo~QqYpSTHHm+ewd%Y62aS=-%+Sb=ggWu
zvFxbGq(x*DI=!M;S{q5QXCxJWM89Ppj%V)(x^0Z0qfP<(qYIeyC!bDbxj6NHM3dY!
zCYYtt@K_uLvTuDPJJ{JT{1_;BJFKA(8@qc^;pNHKe;$m^^kB;^H`z0|@Y>y(E>Atk
znc>FlA}5N?ohZ$9=Fj)LJgIk&iKpEde_3W+!QH24-B{ewgXmy)jtp=^ZKn$zhq=(*
z(}j^zj=s}6kl$xKSv@|K`)v~VDX(S1%P`!`L)ddWkVCQei0$IaX2ILkXYOeEx$)lv
z2ZATtkkH<O7X2=Av~nY<PCA5Uts=~0h3vKGu*-Nhhj*{wpx~{|%Plmk*hRZ}n;5-w
z1uy?CV?fbDF5Ozd?Ee<>@5W-nb+1uBWF5~so?wvRt<kpwI4A9A(#lQv#&5*kYzFsE
z&KF<x7T$b0Mu3jYzim(9{pJ$={dDQ1a~u7!R;(W(y5q)H+*7mUa)l*6Uo7Rku;kjJ
z+t@ENmHZSFGWOlz^zJLTv^<a9@_nrPy^F@;XUb8Ny}yxoxMlD3ZOcR4<*c)PD*L9P
zAGx?gGUOzOMXjo}^808TW!j~7N{4Z66kX{DYb|(N{8eVKKW&tI!g(Gz+*-M&s-YM>
zQkQ<ly2@4addgDq${!HC&6FH~;WiDFRl6D}3!BM*%ghz^sHyUyjbw>CH&gmsOGlXC
zZFj-j)?b?_2U<!t=v>Lr39O@}Cf8BaH;Om@P+eu~^}0$~k9vy7Bk}kdNrr2P<fqH5
zy4}2);v^Y%bv~;rYsAws{iJlF=(bQ67BpAVyQ&F$ApEQJfvu99z-^x;M_shF>JvnN
zJFk_@gYA_Qsanc-$=`2~Eq>6tl9w)cJ9(7k2@L&CNv7of-jn`fr(eQ%)Y7b`u2M5o
zybr1^m7_AV8?Dn&G`_S{=4fdsgBEEBhpVN8$=rWtyp}RZ=Go!Px+$MLx+zx{c2jCw
zcTxUI2bN*IHp=nOYD)V^$*4c2rtBT2rkHM!oOj7htlva5OCgd4v{yPYO<F7GbXqA5
z52!0MPqk1&IyP5UbW>M4_f}Wh3ErOSpspO%X`*O9Y@|GGR#*Akzm{Oh9XcrfMdf-O
z#kP+0M%UC)x~#3M_&urQ*tt)%kzDiDZJQ|m`<qD@Q!~YKgqmWXD6fxOLq&0yyx1)b
zmESEJDS4Ium?8P=GYfuTCK*w3q2g5iA55M2gC<vh^YZ>5DE%X6qj)?*f6`d}A!80E
zlbxT!!NIBMdOl#&#8h(DC30HuR=r^g%?~Fs-XV>R$&y(*`Wbq|%4jLMJhiHlb8PmM
zPYbhou4M53pk%_G3S`Qs48F~|&vwgn&Xve)Cm1^BRT$UgZv3$@oUv1)$Y~qFe~ZJ=
zZzmjIMmSr(hKrXy3b(&ud}|dbxqktKo%dkn8!Nu*n&Fytiy4BqjYnI{TxyQh%sUL;
zC-bM9Bmd5N^RvRA_dBha>U4{zpAB$0=*QPqp6HFU=l(EPjz99?mXRwu3ru<b-IVb+
z&6r*NfZ0DDi03hbtwX)BoZ~6_6*p$8*zjPiHQ_3ESd(vq<sMHi#CtK&PaZ?~sqOMu
zHBO#f&GtmU$e(&)e$*CwW6{eOA3uM(+zG_0)Q^mhzT6lgeP2mo)Yp!nczX;z_a%{?
zC!VE-$!ylo#8>b(V2JF!?>@(OmShT<l(WBM1^MSInEd29qc0Q_E&Ors9hoc=Ezzf@
z_wgKgpD{A~%$c4>mh3IRai7(Cg1uLTXV8?Kp|U)>>g5x<;W;~w6*8})fR0lO>C;TK
zUyDU|CD^F?xC)bXRcOUmiO1wE#igJ5yYdH%N~(#<k!+1b=}G(ZmM!XU*lJzH#k-Y^
zJY31vj}?-IC;re-(ar5GqmB50Y>vOjsgq<2eitqG@5gvnW?<7fmd9S9ObH8N{qj(P
z4C65IkE3<1@VEiOaY&|ugUqAH+6(3u#)|(elCj@KUm6(4+g6E~jZY8_R~(0hgTE<w
zYc5#qX_UshDfbEcl*QuN+2XBDr?sl=-1TLDFffVM6Jl|(iNf@fU|W}H$qDsln2ML|
zIJ_8tBA6SeLU{g7JfJxd%v~12hy@XtCWg~)dpKYBg=05328*e3Cz=$;<S^O${)(id
zPc(k(W4VzW!#?3${PSb@CtP!4QG#d+6NRf3&F=bSIg67-Rgxfk{wU5DL`k=R+!+T&
zaWF8F8Ox*Cm=MWI*>}XpMsQ5Hj|Y0;ERgqopXCvBnO(qO!CTe)`RoxL#NaNEFlw7j
zsc3-HI>)i+Xeh%c2QaPNkN5w)D6DbEJKvp_GNV5H>p^;&2j7F;m{8<M6IW+6R=Kir
zv?~`<o#;K(3Da+mtmx~+<lpy5I4V2c<8B0=bYq?1?R$H7l=<!q6}+9X!-a8OU3g;e
z!t_&~OiXr1=Y)KQ7KZTJ${jU(U+Ls=<U+I^i)3D0cf_6o(LCu6wV+q38JnM)%DwC&
z1F^x>?FI{@4%0n$m(1APs6TlQhx}LYwf!9GEuKx4-h9bgUcm76D{1Gll1VLAF-BRz
zikfA#C|JnD^$Rc=zK~qag-rUqlK02gf!|ur++59)cT3p2e+m6O7fYw@OpHg)Kxw#u
zZ(57!)P4z5gVwPw<^<bE>X6}cjNuz@v+%eDvv%t-&ei~5GYgueSkm5KuvXuU=3{PA
zFyI#Zo0!ml*A2#OxI)hk3U(!<Y3jX`&E>+cy$I#EzToZ8ObpiE$J+cM^$yA&Ve&@|
z2K=DImcLZAmA<TR&6U<&S}LtWH56;v&jzQpQKAKJzes=b{Hbk~;GpJ;-)iafxFnhy
z@d)&ElaA0R$$Bl7E^EQt7Si4LVuSR$NZ$I|7yqb!T}%4STB^6#5`M9keFn8OJ|@{H
zw`&-fRg0#T<e&=PZc33%T=D1`3Et*!lYW%$^%VDuYRZ2;f>nV{6q}MJN~^KWl=&*s
znPS~gId!ao(z3Rp;w@ZtnRr^<1#k1jH&~IWs<?D%sGJFHtZX~mSm{^tl@O2jxcF6z
z@1&Y3XJ2x4R}-a6ySmDYY{|sSe<Sz%Z`j?IE)V@*n4bQ{2<iVi_OzZ-)=WkDWmQi(
zc(ab;Q@4(y$Q-=qwz|?u_NA(MZIqq9n#w&SpEzDiNtZ4*<8m$KRdGjUZp$u8Y-W4q
zR-HDAhnn02q*KmCyfg*U<5sd>I&&nGcE&z6<?6Dg${O+QBn?(q>Q0oqgGNgw?`v~q
z+DGBke@L%(wBW1!TUB1DD&BpYD07z7Q|5@bBE3-!k8EnOSon`mfuc#eTEvA$a!#g5
zhPSbFO$pw%O>3-to7_Zsu%W3^(oOjI^hU}a$<<R0XrQPG-Zos@P)U)T9}~gbo;_-r
zeDoOsBmPn~?k7(_erNrtA8gw3o1)&*-TdSuGY`FBvray_f%oaPD;aI!(nym|W>wjF
zRmI`+On!eY0qxQ_QhUa8#WqPY;A9WCAqwa1QH&3dq|_^d?j{jZ-XuH3b{YH<ysbSM
z$j;+3FQ%q&U(VR;_aZr|;e~FxJHt8!({4=|&F@6Aqd^4s28J=@a~Q)Gg>dIsFkz>I
zsoEAs>LM#tPMc#IW=7SHTl^k&o8c+9sJFs|!3Ruv^vj%!7py483cnB=s@8gQbDjaq
zM(Lw>?;5Us3}qK-NIyG6mM59w8D&CcA7kd<HetVF%Gr9i3BPTI-yn0Y>e!$=*_E>&
zy~IOoPrnKFg2&dV|F&Yt3mZnc*>HTR2d*<c8R+SOV&g$fwg=97o;am@3LoZ0tMOi_
zzLK+D__cBS{mHu^&q1*tQy&Yz+&herCt>ng4d;%`N+%_o_DW6)A1_EYMqoOx#ryJp
zW)9O2=5T0EIT42CjQCs3>ebI#pY<H$^~I7W_mu71GEkY7L8eNEWDKRFB3XEYMx+rc
zxek+S@1qv~P<8{tueN>6tNVGf+sa2pauaX=&7<v>d@^kEIb&aet;#DBrDG`}tBMP=
ztN3!Uih~+&NvSXSR(&e@P%9lVqS+cc@&!$PzQt1G4Q6JQw7OHtIKkVYaTPf2k^k>g
zM#T=9yRTHCHsw8Ac2u#tOf*vBvArao*5h?!Ic^wARA>k%MZ4-M?}aBmag1`8tiAeC
ztd;%u4D%$KJxOGo%%qh8GW#V*@m{{?OY=l|zb5dYeJm3X$MCU7G+NqGoV_kH-{kw$
z7fsWsA(@nn6yDe{k#0?8Che2NuobcF`W#8y*I~SA?njpf-dqr@o%!06AqN9#v{<<R
zpineKle=<p1X;VoQNJ3F>6UQG=MQJ?iE!Q&hjVOX97VxVXy1;M9ds0)S<!-LG1wKv
zP?Zrw#m89LJ0{R_M*=aS@!Y)<L+<!!vTdWVS`fvo%~1q4jbcJvI6DtUvS(=&U6Lay
z?h{FPR0LgxfBCR9oUA6{^qwjE<gYQ*=8Jb!F#1<U1n2Ju)1syD=x-CK+arcn{X>~C
zHGo~Y|BtuF1#c${-uAuYPO<Ynjz`;zKiL6YA4l}MIkGaq35&MQgn7!Z9i6yRE_=*6
zP6Vqsu|js5U)MP@_ogG~S~|0Kg*(&cy0ct#5ElnKb7k;dlFqx(XM*s;wk}kRyvv+6
zW~}|@z+VkF@iIHmKFE&C?si1Yw@0<clC`nsJV-IaGfj_kCKs{)yNC4#>lhh+m|>+m
zSX8l<<Skn{vT+tl)NGQU&gA~e`E>JNi09;G)Ze*`&haZz>=u*XSa4M4TFbWc*>Ytr
zX8Mbmm$QO78`g<`U@fB`tfGP7?HcbTeDqk%O2OOBX446;p2GY|Q^b=$m2<PEGj`Q_
z^p@zL)H})fmKS*V`7%eQ>k@SOCN)0hcqdrWajgYAoo~_Tr3n$Uj9FM{gs!hHS?jNm
zs;O{0YBX1^w)0}VKSlNY>F*gRe0Dlnbu+1ye3A5Rg_0%x1n(Vbk{2u48^Z0J))bBU
z9Py9%Xed*qU+VO4>4a_~z0tvqm1l!xSKCx_SvpDX+REP~#{FTibWT-1Qc-eC>np?8
zs3<=K^Ck#yJ+?~&C1CeI`p4JuK=8{&r<R)+B=hrnEzw77xOJljyTyNK)=Y9Vhkm13
z$|r_w6c1q=@n(uXFDSi=xssWt^;a^~LJFzu*+6vFb(B?-q4)2-X!d?p)6?(`2QC(~
z<VGno2fv}6Zat-clB(h(Ids>|8!N~7&a`fC<c=;`EWWR~KJYc)M!jU#g)-TF)K@b1
zz9;IioD++`5vu+Z<#Q#|T)$E{Q-030T0*;s-=U$*xdwF<BXgMp4eBcP(wCC(QaJ33
z(&sE4Y?sGrDI1$NRdQ~tDT^A22JA&M#cP0M3|q+e3Gb`%t+6sh@YY1|c9URU!bHjJ
z86dOef`-b)=}ndMCz~sGgfIT$rmp0uOBPh)?_ApYf%^K=Yb9DHk0|MGd9JGT8YSJG
z*J`+}U&ChgKlrZyO`q4LOkG{fY^x%U{wm;jNHLcWzodoC9DTcfX3@cKJabf2BJ$Og
z1!;|xalIQT|31`Ln)ug~yvKUV%rSKpo0K0IjFv7sooCoNy=A6$HJ{#n#aFUkLx25X
z`H?SV<$UB=e6HjYKEgR#&Mmbh{!UK9Q7@jYjUoteilf<*c-*9GSv5Y6HI{K?JdBk7
z5zz(>k#hmT+L4i*^^Ks7B40Nnn5dV=XTjSO5doapna-om{#@T}$Hakl#Qg9iv7tNT
zw%d~BX3zYqk+_D0^X^ZmXq%;r@@_E81dE5f4-)?S4jKI|C_ZIQzgRP}YfRAkX3CkM
zTg;weLjG1`o^CQ?Pe&tucQTS3c|-QyHRMZO18Rrq<MaC}7QYR+E_hpC@HRD2a$;PJ
zacpah?LiZ4x*D^1))8pEpJ}6Yxfo=FnQ&s4vfTI~{P7jzJHmxqlTvKO$Pyds9<kwE
z3v2EywI=GU2lEcz#k|EmX8&}fTe2s{s$SF+yv=;>EtwGacz(l|PzPTM3jDY(^X`_O
zVZ7_&&+;LG;2y%Jk_3ijq%dQ18qU(IkR`cF1v7H6%FiLI@G)Ksa_RoQ7>7mAsac)K
z(PQH0{V$bEt0cE?QaXG8ik8<ug<D^fc(Nu%-s36g%}b@t$~55_(iu>Z!+~9S*w^M!
z_d_0!(<MjxeIA3w{~G95Avt_6`S{`${SvBpuO%L&lT~zS`Iduq-lF0mnI1vqe0@|-
z$N$R75baa0Q3>^Q<ufUs(1iM>xTlxnxvQKBr^|S;sf=D{s(2__4XYX#@k#DkorLq%
zS|+(eY4Mo!4rSc`aAK7BG#UlazBB-x>7n$W5XpE|nPo2}k@z_Qo8~cC4T>ZzAxb(X
zqA7hBkCj_IT9q;69E@Rbl;HEgNHV^rvNI};B(HQvbeCL<DUw%VkSO<)1Ue-rGFvSc
zYpn>u9r2y6Fy`!TBPRFrz*@%>&+TqpzV5-)_5MU|4xodP6)mm>^J`uxzZ^oTSQE~!
z<Ka~1hqI?7obi#7vcHbN-!6iB8zh^`Jx2OtV%TgDiMx1eo>;{4H7=H@S8)tG6ocQW
zXco_rT#bK`tWy&%`D_GHzXNc65kQksQH+**WDD7!9G?{?THrAJN<vZn8Ah}55wd%X
zpeRo`Xt~d7%@hBqc`$vBhjMX4I9Ai5SpGVMD60TMqy5-A+lS33J^57TPU0*_>5;Xk
zgR?zV-5nVBR`?#-Pwv|%GiXC69+^1tI@D2o(T)VXcEoO_17r3$@@R{_Xm15SyF0RA
zwIes~I#N_D-=pHp+@^QA8F`of?ZiWK^DgTIZ@qlY_!oSOwli<>*~E@(P3@>Hy@T;S
z8?G2wq7h-vqhn?a2^BBRxyyXrwwIZo*0Q_-%)c~-n+{{~-8~ia^|N?!c((NK%#{22
zOwnD=W?|>0=uKSC`)MmAe{GRyu@{jbcw2YYTt2Rzi`tfXgufZfin0ATk=BbjuX~E$
zyEppPeOQ^>pU+Rn@nL{M*GZG8Tr>%j?5WHiF&*D)lQ^@oAC`0bFtKq*rrp)1yniP?
z)xXKiC#Kx1F(aj!Ie#N>k?_KV=Sz%fCV1P)M;G;=%LHgE9NaLFkTo4x(8-_6ulz7t
z?oY8+ARnIu5`8L`w5(LV=wy*~BaJfQ04#d@G32T{PA?=3m>~W;8qD<Lg3ar*IsWAl
zCd(hQX^7<9O_ko|UcdR{|C@V~r)B&<Z~4hwP6=k4_~)YXutc%|N|>~(h|D*?C@-qv
z<ic9v_<u1@x`bOCt0A(^cV5i-L)7RW44d{1^(4VtP0_-uy&>nN=qCkleWmkd$NXF-
zdKD7<yuR|*td3$`BR&(^Iq9COCVSNz4jm{a?384}3*L^{Dl=(<bV{nKD60-uv1D`=
zkD5u2weKrDmcGSa@OIFUS4=E<A^MS08W|QaaEfFpbSlPoY&rkwyp*g3!Ewp%%{uUv
zKAr0*D#F!<=+{wRZvMwp3;Fq1e<jlM3w@8ZQBp<c-5|fd(yFWE)$dhPUS9o<wQd!k
zq!T(u@VBk}j4ILx)BN}!8fVvXM!L{?4;3!Ah2-52ZYa6yO_kb7;&-gBA@ILn^pQV*
z+vO{jL*LQwOf{Vreq*a^Ej7V^snepKlDf5q%yP;0%m0l}uUA~UDf#^?i&<t-gla(n
zqg<cjFg=$sflnB`?hEB_KCtcZXPV8Z!a*_yYY)7ke!y$=`qrXmUWsqnbFRv4^k1K1
z?#y|Hf#7X++iJ$8d}QkX{37?ilAe%4KhcV+^-tuRc<RpXk#j>cl0d=RZ@TfU%#q#a
z^hnxSMsi$cj!W`+kGU1i+n7j}i66mEc5%aJ2Gg{CFwKTao=ndWE=lJGalyQ99E6wX
zh5VQJ&}o)8{U1xO<dQr5mt{uZ;Ws(4^d|Y@ch^*xe3_|soIT@D%XfE~nR=IAb6xn`
zS-PHY-oav(8G8ebsdhKwxUn(M*BCQmsWI9^O*lEzm|Y!=Xf0o}<?FMlA-4rr6RU5s
zZTU4W<X@$4`Wa5EUc~suMQ&Ztmy8Ny{8t%CPS8yfRIX8{WGh>W_OaCR5(_WsVZXta
zYY}efC0Y=cZBFKWYqsTC(cjjJ!24$O%e7+Fd~4|vu;OXEyO{O7$JMV6TsY-ScMWIq
zn>cgY#+lFQuH4@1KwInyuJ6d}&c2j3@neOTA5Jd=NEdzc%yE&D`JTeiooS5UoQ{!M
zILlW`Hqd}j(NxMkMl~9rmBJxTkLSkRMA3F6i9ab-bRQW!RnmDZnpnMevL}?Bjq16Q
zS*x4GX6ZjUu`o^a*XiU+?n8R<Q!1=PtFteUZdrMZe3r+9=TFhSmCK^qTwdLhOp?7N
zcqdk}_^^B~r&RF$SSfwo%30!9E*jHv^3Ip@!@HDEw?zk>Q^+%yV%!^)a_Vy_9ZnQ7
zt+I&T{Yu$WS|<LBG7j!9rhif<MPX@NTae1q`;wK=FqDn={aKYM{)`3zeDMn4dI!<5
zor~bTQzY|SMT(DCaIPYZ!R;eiAv>}wCDH8o9ZRU}`@2mKmkhulQZs`YH#Z#pH=&%;
zj^mr$QzreF$Yaqn9q5@z&Eo_v6v~{bE&8l45&ZgQD7lQcx%|_Rucr6df;)y^?(zMD
z6{deJB$LsC?bViOW>{es?Z%LP?)3T~nYx{PxX{{<Di`r`I)%|DB7zM|g>(KK#aFop
z3`vhB{&FM}X2;OsvfKq^C*N&REGM)iE3hVtH<P32`ymp~j6hno7EJ#hK;CqD?Jo)s
z^4&*fB44)64`R`kVES8!67Wg#29v{CY8OoKgCM5938ep$AU^5`(_HeRVrBl$b(Hxy
zI)nqa16U(_i_*0|_;mJQMVT`n{2WjU?OAxhp4BbvaUbhIbh!f|haGtl<4DPMCv2P?
zX|&W)_-i`?X4wh%d52}2?fEjq0hP}VT%O^`c=?|F!bO{hIf0sJB0}zRt)(lkCf((v
zjx9Q)&7|Y<7VG^CnX=xFtKV$-oqdO48*GS)vLejEoSR3@=={b6-<HOFUbGj_O}l8n
zb}a38j-koGF}&C^295c%STkZa-#5;b&ckUOojnV;5lcAmX&G<otia&FB3`Us#4>}0
zoI5j{V|p{lIy{9>ug9{*YXFzEd(rq(cP38ljg3zq@%eY>$m>4DKkrZP0h7p{J%N=g
z#^L>UG)p>+V(FBAY#h=D{gGX1d_bEID>Ydx+;t=E)qJ;HC4R(JEN<GHkvZ+Czp@?0
zM%vOx(gOqk%jD11<kLwl4uossCpfBk*H2~=f63|)Z*OKGmn|d6S@i&|$PC$2q;YbB
zFK<je2$<%M(PR%U3<;u(ypB36BwsV`A!f$m9QBgytCNY0tt}%e<uwhD|0FyhPi9Ez
zWDsp#BjKjRU`55?Vm67M!?aZqV<yPXPxOy>L}M{d_bW*Yzp}OPD|KSNQBm@Z=RH5s
za@j}Ot4pVbWUmzOsbK8yGL|_%W2RXV^}FU$*QJniD$m*3y$YMEcjC=@$1J&HtNp2>
z?zLz9_eXj=hQGmJqh#KEtf%ZStz!OO;jKng(ZNr$ZFiLlHofBLsn?7byk&^=R24~X
zT<%kzxfY6MznFuIN;$daC2kk1#q)e0_ePoQ+Lj8Ps(AV5AJLQkaX_nv63M?SGW^2!
zzn?Hz_M6u8{&A+~UrwZc=g-8iR3EQmKxrk3=ijiPt7tV>e;{AyJ?A7>wcm>O<W2e{
z{OC7s3FaPq&`5EVJ*#cpU&4<6V#VI?T*>~5y5McMHWl10D5uB0T6SOmi(%t>ipIbi
z?nnJ0@BVK_e39N9lNYoaQi|uoXUz63pw70Zbe$mglb1QX_?<)Mw<1iRmNVb`C6^@U
zK-cs&Tjco|wD&cEIu#rjyv^U7%a`lfSjwJNGog~%&8yM;_ZBblnd*C{uxw#6eP#&W
zh9)pJNp@NC-pPm#K)oWAPHn?57azb3<m^xPWB3tYZjKRc>G^2Z{E{83oQE$b%8XVP
zB<E%jBNhe;2NOht)F5uU1QB;r&I-{jZK&%*!!h1mD)x|ltT|Hxj2PSJCf|<fN!Fb?
zo)hKVtM+2#Wgom=+$HnwT{hI(NT;P03skL`U@iQW;O&%oMtt}pb8LTOW)vBbrDntq
zV<UEJ$=uS$h|{f%cy(QmyrLT<1m57<pld7(xk_H*8E*YOLw)zNTsWxD01LtDF8fK?
zyp7Adce7^yR%{1uX6dv;{F-)@B?ZUXwEGkp<ImGB{St4ln$thRigB)%!gHB108@HR
zG9kcGUuH->OzY{>`h{p6SKeV@XInnXIkl?DnagMH@@I|%dj{GI|Kv!lmM?1?`0+=}
z54TqV%okl}hUl3WWCx>BBHVhD2%gJ~zhP-8?rlSf-x!9D<m4=wCfc$l38>a4pcxy_
zf!ZV*2#?jLT>OusCsN)d@a0t^QJN|2bcv<^wODks<Cx=+gmY*PRno8YWmGPEx8!l~
zK^~3s^XT>HDW?wPl575i3x<!x%kz-u?vm{kf1hl<Cpc6T5@ILXXfNp~jVh8p70E_?
zB3|lVIqV;iOaA^R1i#9qvwZ;>M+$k-Pxckxix~G*c;#+ooS0Y4N$pG?pGm`eb}CnX
z3l~}`yW1T;7<_l*#2$YpdI#{MO)yVQ!r8h|aO`meQ*6ViEeK+V-1*Lwgm7JU3eIYA
zj5!jAu}T<Az6A1JG;@Ph!m-T?B`!sD;8OzR=ZNLGae~ad30MfX8z}hRRrE<=)p9rc
zVo2vqL&l#mB)GXT-RipYTh|4RCsvFWymdFUK+D#g(f4F7>+3-2Id?u+yOX}sOEM{a
z$rHR?C3ricj(DwQ4ju9_N?zk={zS`NsHfbQ<oB2U6aA8paORU^@oX4Pfvu19S-JCk
zr8^%RdNOvPyyp0_RWwceGTd1!^JpL8J{}96)>;KqEgsrEkHh$$6poK=Ft&oXaaDn$
z6$%u7A{Zw}e;R80^VvUuMSFs9*cw2Eo1f^zoG3_fU~D5t;x5=RGRl_q_3fDY)lU2Y
z_LMzy;NwN%l+HWIJ<f?`!W+L`;K)0{TVui7;l1uKZ;(CLhda>flLP&xIx^JQk?Ri}
z>FndgwT3d&2Hj<(iYr6HoM~NVPs<Z#tm%D=vVMkm&otngj|I8Occ@-qLqEaWfMIgS
zy?&LTEhb!SVvJFvJ(7*QnO9Z4*uTFgrUS;%Yu*@6%$UeMzp1EAosL1@Y1m$yO-cL`
zOv;vV<kxaC%;(S~dkRPIPeI##GXB$NFyO!x4nH2tl?wwT_qZ1aUv@*Mwkti=yK|^v
zFTDQsqHI=QmN)Frx>QYJ-ZWXbt0iw|v|#^)e&Unp!|?u{+4xhFajF^&Th@wJmFf)l
zXo2^n7EEf~kV*GcF>TzAzK=B7YubVHd0KR8q{*KQO?r81a=w$F%maSBkTda&;O*)G
z-WZn0(7Yyw@cIetQ6!V3+6T8=9<=S}!pkaK$zZpkm766`e_C**hdJ~6SrgjVf+-D*
zDU-9T&(B+=bvL7@tr;8hW%nIxA^QW-y_gh=|GJO?g@ycTQAAOnB7B^SSRnodJ?TY!
zbmj|liobGKdJ6rAg^@BL1mCV9wEPgv)#JgmS9`&zL!}IGdWMP3eQGyK2I$U0-h39W
z^yYK=|9Or2=@MG*FX4w#9-f1rv8S2j*$<ML+Dm+i+Le@xu49*u@K$mkzq<A{ITK5{
z+oz2CM_!YqyyC0qcy_jyUW|TEWshCR_n>De{Ur<E_c@c6J>zhI<OE7S%TdwY9)I-^
z*P4g4P^-bv#FrhDe0gf@!_fmDS-rHHx=SKZ5j|~azj*x8l6a9=LZNRZJ<my=?T$(|
zRpc^iVF3o#1q}7h<H{e|ms@?KVWp~)BD&(_-5<G7E1j-g#FN^;5Vg-=7!>|lc7Nq8
z=v9Kcv*@BiYe}-Lt1K2Bxx3F#45$2{hj4t`tY5IBO#bYC(W~kfa8~XnE8jk1XZvi-
z1aDJh$Jau9H-QC-(i@p5{xX?=(v#ROyPn<I;)!S}Ue^tx4L+64wRJ^U+$^SRqiU9V
z-Djm`2K#TPFmGrg2D+gv-xPvM+h7bQ1d*ubPfU3zA2q|cBe^wKdI`^)=Ev&ywoJU^
zOxBiQI=u<!q|74oRYK4&4I<kn5X*i+jB^emXHyW31aC70Z*}Jekm2r)&y%|h&UZ%t
zr#a^1^%>tnkN1;xxw`WzU(cIR-oRFNwVwFL-ldL;?1;9RqW;f>#9CuaR~vI9+KB#x
zjd(ua2#ZlhROcJ9P3Bm~Lq>d$FqA$QLjoi93EZg1_j{MwIOU3TVqQT@>ni52jxztx
z3E`hlv#66k)6I_J_f?@peH(k6m$S6xYBI_<Fm>`F67`R8rSiD+AspoG&{KT(x<p2v
zndIY|^Kq~xeb1Y5P5kx!-kV^Wu8a44J#-)4WS6f44$JRw@0<;DPuhuoMYy=-cg2t6
zz>NKNG)t42>a-UVx_a|WW~NY;0P^dFa;sB_+{J>idmfBWJ>lU+zth`UastjevnpBc
zEw=JJ-U#EiR-|MRNfwZkWWj$H?qO>>AIl_DuYMwPWS;9iFP1KMV^BR2%kZJGoD#gf
zy;1g|)zVoMAz8)J|FBDZ9gY5GvTpc8{#fVYclt4ndpySU<|77A&&Fe42H}#68PWYA
zrj4F5$gG^nCbI8oU&Qm9`S2%?zn`;ZT7SZzeNX63E@$f$Fl$jBHopq-{8&W#Qt2DE
zkLTVM+1ZwhXLLy#b<L95)Hsp5`hn<n_91h!8?CqdGtW0bygY&YyCPo~!Uc=nNF3!R
zxtuZ!$R1)^tt%&sgQTA@glm$=yIt-gDIWv5)+K=6@;nCY^&+G}ki3sXi*DqE%UlNx
z-x)Hmt|40W4af`9CwZR<t#S;-=cUKlNqU65GNMD8F=L(08ME4o-KTWvaZZm8Crybz
zpvT(2*YK))jkg8Y_@|@GZbuWsTe{;V_%~={C@MojX>1tE?NG^D61;U0&spGtXf{2I
zCT3VHZRBfXMKoI~!s%q`j{hWg%9q^3xW4e$ZXOi5c=9S!_>jLY6m2w<oxGXoYRuWE
zBieyG_sBZxO7&P*rbW155gaI;@PYVfd17?QjSp@7`Q#HIeSbkTZ6mz2tDksgok-CZ
zAE}L<Xnt&2eAiaaTMOnVne$ZU);DYIaCNukp_eURH#!n6``XI!!nK5%(=+%sXP?|=
z(<d|PJhPzZW|>V}+OWjz4vu5(MSJK(*S<EaY;MiT!7fY+moB=g4ovMKJ09QLD2co2
z5pjurt^Y$$^FPkdzeL}5XNdlEku?vFapmq&%pDG}qyGV-1#gQU^un%X4_3wW;_a%D
zgsz{+KD(*R7(QM8zNzH>8bglG7)(cwWrxc+I%`knOV$+D2;P>|j^(JyWU78lCd_^k
z%lA*@zH?8S7Ij03?#kn+&g|>mnKOqw@t~15TSsWqB~g>vJ6fUfss%o|&6!O<F1_x}
zP{XdwtZ74FGj$yAt8-|v8Ve^iB+^2a&*7?cJfn(MkSZTMo5-E61*_J#B<p)S^xJE)
z(M=w+zAwq%zD!*1L#~cDo_)R1=;I^&K{)BzkxYN<!!i$VR+xD3wZ03}>e~yqYt1ej
zOH2fB-MY!X_?R)vS{ahw)j<5jMw~l#OK?M8k2_}U&Xnwqy5<yZPT)~yI;$q-a;~6|
zL#jpeY9KkG<}#-W-dbk9<isR50w#FSFDr!A%R^`-yP7rHl08ryjK_A#t-KOJxu*Ei
znkVy0JA+e&kC@$B_(|DKsQNrZ_jUoN=M~WWRX(5Pp8dbhkVCf~(_gd~y+@RY_Pdme
zm&#a?T1JZj70l@Qntuhw)CtMu@+I*g{u8av@cSI!kj)@D4{9$JGSl-Jx!R>%6U}Ff
z&Ce+Fmfn$V$>gs|W#12LCOP@=HBjdNzdlUt=F6~~K7=Otkn9^m>=T*qCPty(A)e<s
zNti~Jh*q<lN&4kj-F(6b!P|ay#0Tj0lq~76nDOAF=!U)$F1&FAjgPojmvW@<bAHO5
z;ai_Vezq&3PTl9+mYr8cx92Q#k}e&=+n+z`C_!Ik?zR5OOxdmG?@z&I!F}dUl{~>p
z;rS02NM~FwlV3h!q;zdH$b7)n!Zd<sN;dH0L;{TxX)a#D-h#IU^4xkIOQ!SN6k>Zy
zH=WEGMvEn1PVhF>pqRj8vG}c&y=znPgLV(2|82Rm3*HXz63k2=KUzEc@zY*#SGqbZ
zyM?lCO9-pZInW{2mOZPS@mv+mJmWBq?FxgNqPfj?<*JXf<o(-`)y@Hb@dGG^Zd{9Z
zXTk_it{?G6{jzuuZ1r*be1$bSSI~SWXIbr4(t2KH&zNg`R56epm<5A^MWa#gE{z3m
zqXyj~?z=IU1#edi-s%_|a;3Gt+-G%(>vfa0rTPS4GN53+A$1lS@+L&)%yD|srEnR~
zSyveN;|j0pTxH79qYP6!%&7M#2&;1wr-~y4sE*}g?GWl)j-&CZdGzw$fawbzI`rL4
z%jDhUXdlEc{UlxLi-#fplI#U_>Gt?07URswwb2u9@&+H?-{9jMJw^!Lni}g9-Q0-x
ziMJTG*^Y-YPb@m<i0OBG%(3NkPj_lf-AQ@x#iSH(?o_#8|IdZVcU>8K+k=nKJXwFi
zlRg_gMF;GGp_dbZD_!Z+Er{2*MDNor61|_XIF5-IFG?~`T8TF5sbp3Sl-z+e2{fK2
zSnLo(@%~tB2gKr)6~{uYBz_qu(l9+n=ICff%Ko!L?LO&?9x~VL5nZ}IX5Js^HIg2t
zMa#0O+L=K!jWi-!BvC0DUX9YlqZ6M*Xjlq;CGW6_MLOl;zeyhSgvLhqS>OE$Lz?C?
zpidsParyj9$zs^zWK`zG(cdzPf7uc2?;nn@e+UC~g7~~&aMjU^>Y<V=Xe2%r7iWxn
zN}u<87r5oh$1T#&GSUsVxq`950bG3U%Iaii$$`AbLwSsV&4Rhrff(%eC0w!@H1~M1
zx=s*9+5tQit$63H?xejn#JSP{tEc*0xUEm#6$4%g{(5fKV{&6X2GrcdPIeNJM@{Kz
zd7X}7m-%pBkBEh)k{_kd`O8;H2))Yjzt=dlPM5>sH#s7_pz6aQl3s-{c)sjTuZQyY
zQ7C7chtbzDLbAXkIi@SSm4^|ueG*NPWe~wP{i(j_&L1Z?{AS%_<Ewk5x_B`8wkLDi
zT5~1Kf;w|#p4lyTx#s4w(~}%w%X>WD>WZ1*?ZdS$^qYK_<?rk;wQ*-uA2+TfS>pJ{
zjb$fg?)LZNinjQ_Ykirv#F5Bjc7(>*vi5*2=9(64*kVqtp)JRz+u?G~iZ=#U;y;pk
zUdNG#x|Z}=Wsdgo+Z@%s&CFag?DxrD$H79nFl;b1y+euM?bV--_y*ZyBA@GD%?&X-
zx0Y`D3#qejk$7p>vi{^bVuRN+Z0;q7w)~F^i!RA?f13Po9qGVZ&32!oZ0dZJfA1$T
z+<Op4>w2-Ew!7SadNJlkKjuFgk6gXP2RDI|_7fz(c$DnT$IxByHc0Tc?7~De){UlT
z-;umb9m}LuQ1ods9UUfdXWK+NS@%To?TV^ySJYxUqtdT4yA9ftZ>>c}cWu&RHAz{}
z3f;IC><wv7c(j`AtQ*ttTth5nj@A07!n!6Z+#9JvXn6w;Zc(Koq5&Hxt72}h%2~H2
zXn$!c-AC%A3~$TCv#p7|q{-_#zGPba(lO5)yIWqIdg;aEao(hC@}gT~$+X(#gZCvb
z@<s36x8g40X}0|6XG71<mV7A`ygg&aj_byR*BX$aVL*4``$Cr9;`9qsOsj4)V6_=x
zf6bUBbMW;LON##5ka^QZGFm0C^FbJD`^7t!<<5v6?xMkWV`n2bvYWY~uIa&X;bj~}
z>pAq17vsA55NIr%QPVg^$mi0$K?Xxk+(#!Pl@Bpt?1%~C)U5(MW)x8VI-dc$Pl%AS
z;P+AKI-Df==7PsF(w@`!WEryrZ?DPxTXwLJIv&M*oRurPH|bvPl+C7BnLJya&3xGn
zzw1}XtZq3(c9ebAhbX*bW3fD!$S1+hno>(t9WB{^&WpMoed+SkhcBXSJ+$42!J$4l
z?en4KQy(nteYjWVgT-Gz`YtaL&1N3&e&lgH^)VR+4@i9?U06LI()!v{CXdV|Z>w;>
z5A!fj&7t|?Z04@aXQp=nb2=2FtyV<$!_SBf5T3l<a}LYD!*x;xos}0PR|qHEu7Y3x
zrEqO@G8^aLXMo_X$Lq)JjeUalvL|%$6it(E76sQHpqH7(%&#eA43UlweevRNO(eIc
z<T6<%5+<JiMdiudc%RO@XQ>R`n~Y9zJn=_jsV~{&bCoFO^^hFdZef(&2xjhdAHr^W
z(=Xi%)6stPi3q|@v`goI1T#f;uJ>H6nY_=IUp*XQxdS_LgP1nii!P@nqo&lA^Onx6
z_+!V$VtY<|ISTJ{4+Fv5m38g85oe6<&uctcf0gWySMXSTg|RA^h4(te!!f71Q+bIk
zH*}e@QctW&x43xkIy3aH@_WS<4p)iy?e7(A2VdjaC_}cX>yuNZ%VWXYzI*f;cEo_k
zau)17ZGeT$Fh&_yS!#coel0GeRdWfAw#WIo>?qw;5Au1zUYfREDLlY@R^^SPvS<ji
z{YEphVlI`+1}Y!v5IJ}=x@&e&>T`%D@h54id5QSmm+3bAs(4#(;wYNtwO4fc{_+Nv
zF*i6mT8}p`Zel2StG3#N_w#RIe8`ULazAkpok~oEJtqd*Vx`D@ao&b9qdNp9yWm>s
zg13<iUBex*esPcOg13dDdvTlS$+HAoVsvaN+%H(P#D!(z$vCmVmG|F;D;^ck?U-cp
zB2t+obM2YQk^xaSjsdc({4p#>KI^jU{2N1BTpTNRCeqe4klV9ENcb6r;paf%JA)|w
z5rpT~U>uf2ijP-vCB~)F*DsA#De3%CNh5HC+$(1!;vADeXUV}baFZN}D{1UXNT=Jj
zEYgf0@bcJwrVHLWHTge|&O5Bf#r@+}_8!?Iqau}&jOP_m5h{s{%pwg*g{(+>@4eGb
zDapuIME2es9DDC^%-{X}{n6zd*LBXhoX_WZ-uL~!-{Tt1Mqj6C@kQZoPx0q$4P#{w
zZ{l;1tAe+x9rlqrVGrK-Dmdj|!Jr+bEa+Rr`c|@+<>gZUe;mLw@twpJpi&@tNG*zZ
zY%kh%jUs;E&LN{J7ptlgmW<ee#-&Oc$-U#dWjUGR*ZN0t9F7U+s=ZD+a2(>u(UVzg
zXD}r<{IKv1mpL(<{l*cD&j_VXZWrTaPAhy9MqP_g#%2ZyCmq1F3_o7?^`+nS08Xj~
z(!@IeD_`M5wF9J6$DjTd{v7u6r%FEO2D3`|GeG)MefIKD@b<#ny@U?gPj~rq{#n1D
zk?pIPkh7Pzg104x;&pvj#Oo~SRDD~(V4FhWCbD^BpMlx+6uQcL*WJvIabe!P3tPwh
z&dy|Awxg@99WgWK%l*xUx=|~!dYZ`|yF}`>@?&Cg3`aD?-#WdBgtkSz>?pdg@<Q~U
z7tr%kHg2O*XfEHIfl4Bl;%{^C&LC`jCSTX5VBnm>jb$lxj7?&&>{!!HWyks($Ao}*
zeD)*|oDnaXk@2EQjOR{LBEGG&iGP~Se!<(pe=_;=X$rfYEGg5mAZVhc=)vZp=Dvw{
zW9Lz0XiemhnUWthlMV}(VDM`J)mrnI6FE(0Q8N;Z$8bDBhbO;9+oYn#pkwN?gA8Ko
zHHB+_LYquApi8)c_>e~OB4`**JL_^rWhAyCqos3VBq@*e_^C6D=+ipfZm31SLQO`V
zRY&nqV^5wMvwZu|zHe{oobF7=dmT~z+L5^Gjx6roL3FY0uzk};@Ub=RT)rqB1aA`@
zn__#jF|K}%n9{Tnxvd%zdr5`Nxs6$|Q-!Bv8e<$KyRK&=zJF|ief^fyjA|);_ib^q
zYRhKX%Pz}(cEs83>=SKd%Fa?edx;;=K=MMGY!{zj8E!HY1v(ewaw`u@$@_S(lFE*r
z$-==UQP@77HJhX9+B6Jv?J!JZ!pV9S#_~tP>3xsKdQJi}-igMgUm~-cq!D#2jW#*L
zIizQEV%%0<T*~3MN->7Fi-@Z!q+x?1HVEE&cizUB8<JDqPjoV}=R6e5opWawJ>(f&
zU{S@}7}4(EI>qLilPs*O!pCeEuY2yHhiDU5{(FMlb|=~8eu=fF=jgukEFE?07&-AI
z9+OY7sOSP+^>1Kcd7G1Y_h{Wm`j*F>VdTs@j{TBZT;l@ARIj1)Kd)iMDdPXBWmfMh
zF0RX^vuM_Djg92JNi?gb1)v%oK<TDHdKqn#tk~`H8OwgVO?>1wWeh)6#<I{dHWrlg
zTl9dsqM6v<@-QEVSBpOH9BaMLNe}xu?yH}NvDYPc`wH)~FY;12+<5uXF~7)T*XtxV
zy2*kjw`p1>x9Yh!*i&>Dt(dzQeS1i+m5=y+`VkFoJmZb<^wxo=&>B?B?Zkt)Xdk0Z
z<T2Lnxyt%87fI@JmYhZBxT19qbsgd9#Ao;*_81cdZ+D4))zhz<x$9*ok-a2wM=f_s
zPVrp+{|@k~rSY9=$_fvm{ijNBZ8ukh=O5f(yz7p;ncY~lbFQWImiNBp>2k(a?V_H!
zcyeTp+d4g+X5o^_Qku&9Ice<E%w|n(78R>^lD;~hAxpC*3qOmCFEgpGpUIFRSsWSS
zPiAc_jl?HAr9$}NyMa6$xE@D84?NFp#{QWP3o_R7E7y~gM_kcXStT8}G7C4|z)brM
zv=D##vvNPq`uGzic-ufXfba5Sve}>d{(&^s2*c4yGRy5E$ZHabk=&8?dxkQzE`Vqq
zU&79<Lc%e4FkuzOHEyH|ueH<Ah48O)sVta;!<pfXDjq^|*OAg0KUZ=9=To_75nk79
z$$zzsH+|OgXTf@2+4xA-j=f~_c@RHit!N2@=(Q$<+PlHj^9h#hxgd^w4MsCAgsVBC
z-ExZIK@ZVcsU~Anlg9E685jqpF)S{bsK`_fhNQBnO%%-<C$O-4B9E)KGCeAj+1436
zn4Uq7Sq6ihGPuz`osAokdAm7@?gvw;8k)y~k$bRCt)S3NxZ}=8MN3`HuMv_t|KbP-
zz8*$@;9;Ch57FiDVbYrKVMEw%EEkuvXj2vLzbkk=yb>$JT^yGC-<4&%m@9buZ|9R-
z6TIF1v5w`%r&xULB#WNN4t+p4Y5Qt=-H<GjStn$#ku1*rr$j3#-NIYX5xegs!{?sl
z;YsoQ&OghXXBGS&SVF7p0_LUUV<mjX9Jybem77HF{YttDCjO1zLFO;X8<e}Yhj1CK
z{pD7*DUN!zsW|>jWo&Jl<kP2-cO{id;pM|OrPI<mi|7qS;$4+|LfH+EwAs!wgK`?k
zjY4;k?6-e22ojE}%dH54<^9#=X)yCIZ06jmaOSTLquMDHgJ&TeKNW)az7VGN4&~<k
z5H@b}WqOQ1*Hiq+6)rU4N&tE715q6pK=0m?9a!qme}c1r#5dyY?a!_eq1a9<p}yX3
z_Kn-kmuJHD8|`NC!+m6>@25oYw%4e=cn;3Q)+|f<dA71v@`~$67Lt6ZfXBnqIlC~P
zu;!xiyDs~Ta|l(F{3z?`joa^aTsvvcuCsPLFt8)P@qF=9+fY1Wu5<~^;aZt9)%$$;
z*e9GJm3~-`mFMEGBaKv?aS)%a-LC?SVzMOnGKGt=i8Pdbb+c~*_h-h_>UBJNMu|)s
zAuP-C1jcWXZh&ivw9AQS&W|{{ACSE&J`Tgkc!q?>Gs8TFvi^~1Z;)F}K^Tp~Rx{zC
z3sIe?;8$Zv<K|-}$JBuD#*-*JWWm?RbMTI{X8#i_I=r#OYnM6wR7{!NWD@q{r%`#=
zklwAvFwt9wR|x~f)7X=zWBb#0pC;Y%^{DBmFW*0jkb9^nTCrhVJw24uemYd%9zpp9
zeYW%(N$?Flrl<|0VYv?X9u6XVv*2xG4ce?!W6h`T)Z5fYGGlvFJh3y;tveEXp(By!
zJFv842ZD~ZW1jb4W$MjeiZbnsvgmhHGG8>|#P!Ck%73kteQSuuiZ4nl>jv~atAf_}
zpUS1V4aIlcn8T|Y@j~$SY@?P;+24%D+uO20qb)fZaud3|4fnytd>AGEvezZzU*5+0
zgl)q8mNMU7_OtDhf8$gv{k(aYHcG`xBaS;)WBD$6&UGP?sP2d)R=76L{h`<_4&jq$
z7*Y9ATwfi{BVwqzlEAF@3Di2qv!@`Ath9K({S!~~UI`SpPGg0sXgVK?@A;hInRX%T
zPKmzLpqSQ&im_DL#&GdYxFwe`^Md4#J>G-Wo4sg@9;ZdqD$GO&alcg+$-=>SRaT&)
zaex+`q_^0pmX6|4NfbTT&~E3YpFun(V^2~d9)icOcVcR?lil0*a7x}6X?o{LRV4Se
zZyh&m>$v|;vTp~SqnY?44qU3Gx$JkbfA(-9GMDVmvfJMfK6AJ1N#+3@m)ri?iGeH&
z31wl&?MyH%<DGEKs>X81{a1J(-)$UC6t3BS2SYU_ui@w+noW?tD(MD!(dQgjo1ep5
z<2<K-pU3_FMV3l7aEf5<%6S**t0kF+cB0#=f0MJRH~IAa1`}0o@^6P5WPG@ao$L;^
za(jK2a15K{r_nR4=IN#*sLm6Qe1c@p3Em!-e2LlFk}LN`y#1l4x$RqvGF^J1<7?PE
z_&8H$9ix?SiaHiGH1evJuC8jzU5`@~UCXt9B-^~%A=I6#=(27PrlGqSEgGiQ4|k*V
z*hBYYdvFwevCZ-_A{Ufm+qi^FL1p}UQ^Ml(A{JES6E8UX@A6d2W90^6p2oUf+5Bjb
zMSuTPmW)p1c;jq7Bxce1n)FzCM4)pzgieCDW`nn&am<@~Ih*-$Tiz)x*GmqQ6X~Bf
zu%Yh;1{jN@t9UiX#=GLy(OooxtLRa01157f(6{}1x#O&%P-Qj#_g1p4)n;b7c{3|z
z3qK9~a0~XNZsIDQxvZwt-V^P<YnT<hhQEr3Xd>OIYv<1Kx6TxqII$t!1<Of}n0K%v
zNNX0i_FAzfcR1&Lbea2IpBqg_;n8#sqeB*n{>qlC)9sn#vxG&P?O1Zmmd;b{r2pHU
z{km&;q7%fYRUte(7tGgj!RS8=q>JdY222RUqkc4Nro`Z7Cwi*2Ni04nny+_hWUo%6
z>}&*vuYzf8BK?_yx6gM+Grl^U*7brZ8XL~n85vBq$l%k24Ceogr}l3=ZvM$!HWiM&
zLlPU+OVEfb!7#ak*1bhjkaU;@TdIkax$p43Blx|QOfJbk>Ne>RR}$s(uB}49NhQw=
z%J`PJ1D6SMCr}fumfYu8IPSu0!7kAzmN8nMp><z&a_f)ih+|Lj<HSi^1;+<$JHqPY
zl394QhVe$nN#A%v^0-dPcj`RmN6v9LO)&aYEwfAOD1CpN)92)dWm(FF%mM}_<TF+9
z_Fo^#^ZO+^)BP$aU0p#`M!Do77qZhTm;OFU5}}*G&kb=*Iw~{Bk5mG-r14eex!%{}
zXlNXZ|KVtijs`L+ek%@33Mo%1q0xg<Du0%VcXzw!y-GzZmO=8bREDKRP_xIE1I>Lo
zf6)uesV*4!g|c>3Fs`=(L=Pd~m*K%o2oIs@#1LN1_2rk~>BA&H9KU;W^p684y-l$$
zF~;_kKXHq^@&DpUi&|$27O%#ue<&*z$x0k9H<uCr>qg&&d6V62xw22*%T)~erv#s|
z#k|wXr1v7(N4gafyR?YtiG_?0EFe!SUGmD}Y2PkJ^2DOp(IZ6kNEYnMoXqhebIB;1
z%envN^TvH98EMvRm}Eovu(^2a&EcOl&g8%Ir?;zU^CtRA&Xfz9108uB>ByMxPK=$i
z8T7q5X`M`)&52w-l0eOYc>a!!N24^JpN4|BdWkgs9nX!j1agZKF^rPCt>nB|XUaUR
z6^H3Nx!bghl)Km_;lWnYas4tnc-qsyjTujhC(|k11c%P!nCoUh#QV{_Xg`U<0~Snt
zI9vKItyzD?O1hye`F`45<^WR;e4dEvOH-U*jUzWsPx=+LX`DC^Bh{W{j_(hpnzR<Y
zJ=<3w(=2^-uj%1jGmP@0p*)_h!;qUJC>k(=Be#aLsPPay?+#^L=OJX=8AOYvnm9Mp
z0B1Eu_vp^2Jsp@H(2Q*1fM1<gVfW%j9Pw{JUc36D-Kxi};y=pb3%`^XJ-;Z|el}%|
z;O)ZgjR_y{OxZsFxZ<XDPU$%If$}E$lXBSfrxJIfAy<S~HgDHR?&=M&Ep32`aT5mE
zwBhWhHjLR@O2D&iXniPRiR8B*65mDa#BFpo-A2QG(xp4Ml$iJO_wx(rp`XVq;T8OR
zV@SLu-A3V}CAbtJ{8<Epeum<aAId445H5;WV~gPJLfLyg1aI|}cy<^hVC*E_M=o*V
zmx!YKr*OvShconY3@c<dJ$xaP{W7~A(=4PzZ4n(t72|iXnD|E9=(SdE7q)`CqStb~
zlTXl*Ou7liE>)>waQ!N7w5-BuWd)fVw$mfA0FTT(LgbbaKkfp3ZXU%oy^5=noiIqU
z18z>(Mf$6qoD-dd_U1~_Ri8!@z`5slhT(Fze0uyOXH+Gp{NOospG($iN-a}mu4tW-
z&aN*>Y>_)fNpuK*Cd;kHcr7EEFQV0!xlCIW%I`fX=-v`vfO7`-?vyZgN+DV@H#`*#
zKIgfEVS=~A#BVXu@-z)pg-32H8GKjIGNHwJj#i&%+OqTX@Hx*_xpO56mwZI+BAxq5
zZ{@EWtP#8oeRhK_B{%5z>KYrGUS;auTBe5{<I{{|jBq|A8Fc$Fnp4GffAL5hJt$e}
zhp}2$O*g^Y4u<mI;WV*=xAB&;|5C%z-_^YDUPJyM;TCJFd9b3I=w0O`#O%N+LUQut
z#@8<JAVaqACB<_O6P?5dHhL#+W2D=8-)?;V$PPPkH<bfQd3Zv4vvzMIbY&UpBTE_5
zu80Z83us%M%75a2d9*u;ZI3dT+%=mHa^Lw}kiv5PMASZHpt~!bhn-`X;uwKxuMmcv
z@}<|HEfmyx^M2hX#=ls_x^ph{>9Cab&+V`}w1i*VR?BDN&WyqpJWz8-^~frw1gu6o
zcb)VbtiigE2hJzm(Ym{me*TgZK5q?EK5gbuKR;f5UPJ04R}!>6$osgOR%&Z-J?chr
zCwGDbZx`FUGW40V^u{{!Zj>XZUpSC-+LjSRXVE3litw!AIF8rl<gVe|doqGn^YnSu
z*^qgOw%ES15Fgt-oU`XqA$Xg1*;sPh4A^ql2&?|~j4D~i*t<dOeILlN$HI%W3qs{U
zuyEAD|9_V`Xd6fKdPy|wnZ&9$k{w%<g3J1FdOi-O#SLF3%5HTdE|59aLG=9_#2%e+
z>ST9$dOM7->alVwiX-7@Jhy(w(e$=xCr&0xzHbt*Q<8bNJ&}d-EG=5NgC*Y8T-KA%
zyzB^V4jtyt)<fv6-^+_*ReY|g!hDL{ej`iy7P5oRg16m7SLvfF`(d9#_Q*|9CAf@S
zW6_U`ZpmcrZrmhC!Ded}XWty;cgztcb~(znx@x}8l+40!qGLyTSA$ProO*&N$@Whw
ztEE%B-TZU9f>(RXvHnqpQE(2Eh4biau$8O%qM1Eg#@PDh+#FIwm(`-xexFD0QMruH
zjHj`1m=3kV2eb)c+1X&a$PDu*EC?&@CDM6nM~Uwe!hiU)V_ZB_OtY90mP4~^rSy_p
zPpaI&t=7toY;iORpCg$P7r}w;zO-()g(084L}%&3)R+a#*tviWNiMi+dXVulh{bLp
z!aasCLvTy~?-o@1ucy1VH;#h0uLDd`4w~@()Hv$5^P${j1@5WNjBh1+JHgxHr%7_7
zOv24bG96^+ZYR&lg!`f?6ukX;B~v~F=_a3&j{5vG((flpPFoTW---^&zJSDENz(W1
zho{wIM)a`YP&W(u_)g)X&tz_|n#zF<b8+l%!=dSx1jkzQYoHCy)aKHuzj#!Vov1Ni
zM))^t@eWPrc)S%+_w8u@Y#CqvbK+y^8a|8-<gmVQlilOcC`{nZ`Y38r;y7F!k4{Z2
z`#Qz3=xIEqnF&<HC-T5Mp4PkL7`i2f4-PSs^%BLT2_CpNx0Sq4bKbd3!&qk;`e{>n
zRAxk_<#?Q%jKj#zKs<<}_`JRs!>+5*TXz7z^0j$nGJ<iN4bU|+=W6^U-f2!CaE>v@
zRMg3S-H$nE)%a7Y#!Sr~<W=^?J3$@01zPNSuf<Xyeey5svBPmFW?@6wTre0HgZ@}W
z^`!j#aO{8UVmNdl(N{ERtD?b&UjvvXcstTkjbFn=_f*+|$H^^d)T}W^T~#n`-Uu_l
z2Gna&p9MqeF-qf)(k|ksQmgt!sZD#TJac-ic&~V-m_L4`yfr$m1T8$L_!~S>o@+f-
zM&y(zI!R?p_}2=>!0WU!dRhY>`ZVBMrV97Fw!!gX8x+aPy#1t*XVXLn*IQ;x*>#&4
zZ^KY2rrzFSmi8{@z?CA7ivA-1Z3^+B$<kpG!@~<v<R?b3HXt00$D$W~7s{Hgp*YM8
zk<O}MR1HEn>mYwUJc7P&;@B%%5eKh$Za$BtLle;$=!m~!f;Yo<tfQhbjh%wGKX&JH
z>Rlmfhl_~SEhcet67?@9GwMkyZDtgsb5rpCP!2~<X5u_2lTJQcX?ZLkzh2vzB>v&f
z3%3)SRltg4xwMTdAo;lD-D}pdrduVimRB-jQzhx5Ni5x)%F?0?<_;Cj{D0fny-GY4
z3-%B^T{IODGH=h2obpSFe3e`D%jyX6&xEt_PbjxO2jY}3`%z9H0Tuz2+jw&3>jIiA
zo6FNRPK=M*A|A6KoM(ojRxgri-udD=E~M>u`7^1Mb4W7Sv<^yk%+%BDU4E7ZyUx<`
zyxg+vq-QhfoaFY3wo|y}svBonIN>ZoFC;(s-bGCNUF1o$bVL6*PyI{h87KNKEt!qC
zMjj>NOAQ0Z9us}pA(=h*(X!qiS{O;r+mVB86};UMA^nq;(h(3P-6(pZAzfcfo&4S?
zl^O~@SL6Jsn)afPJT80eQsL@%<(2a-Vh5Z0S29pl?p@!@sdOkK^V}|?M0>o>s6y^#
zaszu?!hbTCZ@O2)zt>A}KV2d|64_~0@+i3}{TLRBs0uHocO{v{lanRWJc*~*L)dU6
zoIN2CobR0=y<l?tITuUS!5E5rg)yLQDF1c~p>2c@KBG3`{m&YDjrL&YYd8MxaiP=Y
zC6xVZ$FY<pWEro<c#h;u_^u#swlg;p9dK{$i24jC(M7s(IM1D2!P^&^E4f<W%9g;5
zjOnn2$@(6coONZ>uoYzgBV5-p7o6(mvN~xt<%6uG`%*G!KCPg)kFz|(j?`b~0DqUs
z+`X9QV`i}@%!=~x;oRP>&6C98+<G&Vfe%JtetR56Bc`DF#Dcdc;GVUds%s3;&>F}2
z4N`;r6Z}mL_?kbK79r#4w%VBMKgTnom*^+Aj7PuaB;mrMc|RtKCdHAg7ykQGu0KHy
z{mH!$LiFq)nvC~l@=9M~Vgnf|T(oygC?E1X#LMeT!^2@56ucc`Ejb(^@^0K2N7TxA
zz9%H&JuH!BZA7yue4#@t;Sz+4c;>jD)f*&p!BK9v5r@$WJH)61;?>mL#q>(Ksf*sI
zU7>U<OcM>`z5>4d$`_46HpSxoXuF{h!!2?vDJ`Q}n&dE43O71fJYdqD6L4a`@E1q8
z^G0%d1;_i3JW5jUYT{a-V9H6+1zf15M`kVSA0FprGr2`bhF0j)a-Oy+r*1+H9cSh+
zKxHd+$Fi85n$DeZk{6#-Kr`n|`P_p=yAa6c0m6^<l+1)<kv!ZV%mvv!&8^q4V*P5-
zOfSJ*&5lM}mT-30Dn@+{C)Oy7eqK4GhvwrkDUX4Da;1Yeh2E`WcpV!py*&|<JK)2E
zGj6;wapsTzGU**yfLhuD=ImdNpSL?tdvD>m%0^u8ub}&>6?n~DDtXNFSbEu+eThq%
zv3m*Q^2X3)`B*;B8OtsC(au${_wwM{IN?<eY+-_Sl4MFH;j*BJS_ko0zA7f6I)%4W
zGVs;Q;E_=}hNfv0T~2}(a#!e-#P5$=c-O~;Yo8WzV&qi1&oU>qy#-D-Q_x;8nUnrg
zc$7PZzp17)>ST#ZZ)^6nv0;tJSQfpWME66>@Ev5sdAr$YwV2NPLTjEhw37^~W#R!_
z$+;z~Xmu@sEm<LGH;-d<UIMmC6xIKUXRbjkcXx}1uR|O+AI8&AIAq=II6lscr{aDX
znjfQhqG?0f-&w5cV9M2@6F9Fjo>f!MC<~`tPzrB7P-eV+tk}06&zg^8acMb<%3s4t
z?c9@V@B6T{#{e27YD?bL2u^3~a=5{8dfMtSf2~5o#zFiw>Q7el-h}K^<Er3o;rH&S
zx%FkXnL2ZA2GQn)7B^px;Hc9ua=U7Z7E1$@!>UZ})nD#py?7Pd3H`SndG}KMwHGw#
z*jb&1O9o&ur#tzlJ2UNXM|NbjV}HsYWsk;lW#I5fio54+WyPDTioelS<?Zb&%3y;F
z%FV>5igAN4%D3fD6_aMylqUnODyb(<DC!-LD=X7;m9?swO6|yEWn*lK@*;G*lKZVf
z$@uTIk~sK-^4_^VuG3VQbg#AO1zO|bQiSG~LNvz|GD<Y)9ljOfAsD<%_RBTm|C&9r
znEJ<yh`cWPi!-v5cqU?1Ci=bDNc2T#(#c<T)PF*G_9j&B!NHtfA0YfyFxnkL`0XJY
zQnvs!&Uz8M+Y4Q_Aeze0Xs;l;3Eqww>y1_UIy6KBFe^D(=Db9D7zl6jDVoC9l0T9h
z$DvEfZ2X(b5c6V|To--Bj~o^W-l`{NpgD6ZKWp;2&_pzv`Enz5E~WNsp1f;wWq(g+
z@N(gpEJR~+ssd9d!P<?L?E6|tT~I22k}~jrn9bHzl{{Lx3mdyVTyDGvr?tEAIF!%H
zZLu615{ZYa-1x4CbJsz%saFEgP4{K1_a>sRdUDU%j#q;g^LYCl7Wz3c#@3Jf9fHKe
z5JJ>X;YLN*5!Xtv|5>5jtVMe!c<UrSfF`3(Goslkx!a#4tlKH}-xVFO?Kvt=pJV%F
z!QKk#pccH{)mrj(R4-y8I<98n7YOo@eB!(cdN<t1iW~blGXE&%Lu!~i;4qyZ?P8;`
zXhWvRU7+!P`o{0aQ97Pq`BzhSU2d};XXw~MI;QI%=Z9xCM}Jn+PqKHKU6l-{Wjh(7
zy%U!j$(+e5$MRSid!B4(X3cgiWhZcGR)&-B4p!aSLG_plk|N6q6})v4yqzU@yR)W*
zFVQ6g2NV(9Tr|k3TL~8*%q*V-)XxjAX_8D{?<C1731L)9I5zV6ZkrI#$0foE_KzU6
zQ7|{Z1e5S7kZb=2NcP_rep_s0-HA1rPF%^AH!B!3)P>%)OIUWwjw{3FW4*?T1zyI?
zwp~H5GmaFyI^fjU5&d~i7;3tSf6|>sPHt4juf!(Dl`F5DSYz+V$QoBRo^j<)j=gk-
z*-7@T@Dnw2i7S|m<D2Q2@3$0AQ~sT5&P<uRjIKCfHhT&EqUMm{WJSvlX8a5sPIp^v
zY-SFjpTjWztsB8;y-^hR8OKvwLsBD<JZETHW`J?PSX%WkkY{HMTU+b#wT(XO|H$jq
z&X^C^$D?$afX}+|JaIPRdgK&#wl(KcQ6%M}xlz*mxp&JC-39^NvG=F!A0Nrh^5IA4
zwX}S<hR)AdF=>_ukMsf=ax{$FjU(AFd{UUa4=<O+@n%>&PdsDDJSy38rqNt23`Fff
z5Qoo1k}r2e?Wn`pEImRuhr{d-J;b<@-Q51Kl&VFdZN4YDxzo2vk3ta^g12h~ZzsEF
zb3HFh=DH$0j7nIJWZiGuA>5J7y4&T(+eW&K2JhzK=lzU*eTcn#4~s@mGEk;U{$SYw
z25B9XZrB<;*GNbEpJR0Sx{HQ#r_?J-!&N;2{eHq7S?18H;a1#2v#?B(8Me(TS~pxp
z?qv`33Ra<F;>V84bxgb)&FO2AjC{9+R;HflxUJ#Qy48gC5Y5dOTRiQT(CyhOW(n{4
z@1+F#n#-)PDWBC7^RVch%lO1(@$f~E@XDVEna4Elt>xo+Hy$l;=4-+-!gkK**mDbc
zE|+n-?=n7bSjxj72Xy`}7ca~Ty7gSj7}3ua-CahH40#U}m~#1`o@l5?pkX;ydaA~X
z&q!hOdJpyp-j?l_ewh;49ls<ptU(?zCyH4<OMKaaw|}fsMQfGHC$BWBjnlYUmxR{x
zB-(o=&{%fxwGoaie!YmkI#c-@ZYG_o<M`gyfDfuBM9-g0)I!N~NHArDi3RfnZ^Qr0
zCD3Io=jIyo(AEU~1T!4(nlsdY8j04{+`ener%THSDR$++3J;7Ed~xU#iq)J*=F5BC
z@`&*3g1M`&MR6`Bh7+%2Fzg~X+qtpqY9B|FvvKqmyfu|~{=#3&87+9Lk*-SZ*^Uez
z-js^*XOxv?XOtw<`^xZFkCo#0hK##CmZ=WCC@Snh!!tct^-mva)d#S!kvc6Kh~D_!
zU_xVuk*=dnRU>WoHP=OBZa*g9>_sho$a7L<mbWVBd<M{Ou)6qZ2C?a`7Nv>mcx7nt
zuCpdHmuRr(pejdhbRs&e13g-IWK2>=w(iiNTaE_$s_JyU-=D-G-BG{a8RttKxm?*6
z<M>8g9`IbLYWPUGec_g(ar25&)BUP4{Da`_`*X_s<`<RG53eYl^sg#MzFk#<1aHfa
zpHMQbtCbyBvz3l#Qk8GIsmkv!DN3g|rxefp8%mGX_mmY0&y<4u^(b2YQF(Q>H5VVW
z=JDNp)a?o=>`_R{_5!*eE~H6tk$4-#H#Vl2yK*lHJ(t6z@j3ico+5p|37E;u-O$~i
zdd2}X9UaKj;!w)=g$jonMCb&`2xu0}AlqQReP2)I^YyIK@}kJWOY}&hCw1FEx3?Qf
z6HUqY)~>v4xsn-{E3iH<TuW{uYuUn%uuZ&r62&P`eeOHWz;4iL0<Q$)+areH!Lh7;
z5i9yU$(X;E&cw|bG%wCzr`&(%7-X?&b_OnTXFVu>#z4n3@g1boG^G^BZ3RR{34b{(
zfp><IFMleXr8l#2vdgBwt=w?#XYj`|pBJwy&{WIC?tUy$vqeXvAHnyi5G>4sS(_I?
zVTdob(>Ia!!Ic$GcG8!<7&E!w`RH1*_se)Ths#~xScjn|%h<J8xWB#roZehWuiqt1
zt6#{Qx^3K+T|D=39luVUWZ$bg%yUkoI_DG{<c2)tjASAX-hp{xnfUZhQ2K8<9;v0A
z*q_PJ#?srNU%~R{l`Jrm9*#EBCwc6k+{*WHX>pn4Z^*oKwMcvb`xv`LFjur$t!9XK
z|Nb!&&zurTRxJ}RivD=^Q8Z`l<(5Y!k+nNnJ9j6m!-{A-w45j2<<d`4Mq|_MJTKTz
z!M0MlJ8b8w-wyF#?V$3c<gQ9Ks{XBPOb2J7Yb0LmCoz1Tl+8RJ>FpBzTV?O96bmob
zeuMZ0E+=wYC!Pap!k6|5rf@|RzvVgJ*Ct-x*D*MyL=yEa2(OAj!p#Du7u}VVJ}ZgJ
za>FH9?w%$q(Z0I^*Dfxs+$^2aC+sNdF^8<0nM~_q%rxT_B!@c+Z|T6?b;~4g-VWVQ
z3wiTwA?4c^(q!F2?oV7!-={WAxo?lhc2{yET`8DhFFdIotu`-W%8xlLwV#1;#B_qp
zE#*6BCSBeZ4869LZ5KrM*=I3xPfwEGw*G`>bVK!3C!#L*<=I1B{#iSit;4kO|1^?(
z-_hhw9z)WrF?4OOP!()Io%>j_8W?aVN*}eeL+BqfoJ0S?7cFCE^%zA-;z%sUN@h-1
z6V#KY&~v&O7u%R~In9g~|Cw=pxfOvG{zT)C{b?WaUi&cSr6=d_uA!jeYM$&}!rMZ*
z&pAYLzF#!|sm0JgCypvL@kDi#Ul+%)bZ(SzMFB|51CjfKFiMjgEzw?`ot7?nJ*9l;
zC7K7>Z-2?`q}zWRqihP8lOdX4tzuq(&!g#l=^UG$$-+h1gdUQ+Phk-c8<e1ulaJlP
zTpsq5_d}-~_WY4&Z$Ks1_xJN*WEHV}CzBtN!RGjKe98~-yX8SH)X9Bn?J>fI)9P;^
zc?l!LKT?=RXTjU~DslWyp3kPU(tlRIl-3(usah$YmAM-ZmptTNv=KA2&0NV?C%wL#
znO3=(6Q8|F+UQJL!BWYtk$<Mq5=x)hqCa#A4o+*?TJ4SJg#<zl&R|#f>5MkA<nWy-
z?B8Qb%bQbhJTjfb^<6pSwS-1LN3*8y7|Em<gJpyXr9Y>#yu}C}%^D)!sik!CcEEn{
z80Km9WzfMv1X=20I!GOzW?eaOp&e~SU$bF*KPu8SIDSW$-+dGY37=u0JDX3B=Fz@$
z0xJ(CviWx|I2P0QKr(hSQm7u4Dw#{Ej1au-J4$ruvIB1<iIy7@m>ncqUf=QT{AR+C
zSCa|6H<e~K;yrC-AQ@Q->4(NkSN(W~tu*59J7Y{b8uI0wR7Wn7d?W*7dQUXL$Wk(a
zZ%W=ny=gq!ID=*%W^wiTJkIZx?{=s=AByei@Yeyi)$8cw>cxPS5iAhB*p>Sstg;JZ
z!iz9`zlO0UD;iJHF-Dw;qsyx;#2mMheCDN$%4~^V-WTQNvbV}Fuak;h@F^uE{)`gu
zdRDn+)to`=TjKqt6FsJMrtOo?>|fY}UjO!xzP{eXc?z%lb0E1_2IKI4Ft;9S<NkH9
z+^D;fZ`Os~0ne3n!~5`lnJSK}Ix)$A09nrlFy>Z&_70GI3SV_%pAKMS)8EQQ_4@3p
zXu#wv&AIcWEl=t?FvqVWIomZPJ57Tljr*};a$jr&Z-4daOhwy{lCj-ZeAD&W*!#J1
zvfd-*znWW0@5@(|!0%U-DbM6r>$A$#W*3#yPp&A}I$l(IL{=)c870bKhZ5!O;BCr3
zda25bFG)(iUy5?zPKx3>>Z0;?&JE@4sJjZ*hZWlomlVwzZ<Mhw>od81L&n_@ymc#J
zmS`0g6&6t8P{{g?MR+|aqJ_Kkla=H#xK*BHTEyX1CNpJD7#5%W7&6X}a{+$1&-7>i
zj8HO*LPakhLW6=WY<}QIdYybX|JXBWjXgs>t<X6>5xc)8G-zhRyQpy}^WP|r4>f5q
zVGz5Y|EGKjYD!sFbJh>gMx~jKXfE|pH`3?jDjiNa4dITKKH0zh*(kosm7m4;(KC(8
zM(KpjN@rklI<>O%8r~P}xoQR<tdgm>FP={&k~#k^jp**FytGfod1(@l`zN5&Ac2#Q
zLM6XlGK!YQ;G8Ts#{6`=|4rjf&nOahgc9s6cf=K8WXt~5rAY|4mIc$Q%$G=8U)fhT
zGWM7&sjX}|=d}=1&)J+mV}`4I@7q}nq}QFn)V?qjU8D;+LxO2qEgr$VLSFop8+ds!
znayj6?^(x#ij!=7S;yz>(-;NS5iOW_{mx#Vrj-+Uw2Ys_rO&EuIZd{dVkvlQcP~S_
z$I@sfIvNwbt-RY`O!{5PQ!T5)W!^SwL!=+8ZvlNj7cg*|Wbq`4Z?aD{&n44h{E`|{
zzt-}0+c5_9J;E=uz2v!8(zloNUKMR+sc;&_I^`sJmE-N1&xr|nOpD1Q)W4Mf{%jL}
zaRJ{Bi3ek5F27TAh?QC8Os6cIQqzdJ8^fMfvDg@-@U$+6#WC@c^Ag93f8$ttQDz*$
zTaU)U9I^;vpo4f}Mf0KYK=!Zb80Iz*4{Cf6a|90#lseO<ud8s)E6E6P<L-Aix~Z<@
z<|)C^(gl1vH%B^mrZH^1CacB`q>Hza<Ywq$F<~;^qaDy#ChrZwpKJDWnDA=0%oo-i
zT(gjsr^N5~Of-<b);O$nWr4pd27~R<G!krfTO@a^IcNyp&W@Un_A7Io7MjsT@Yd$=
zQr?x@k<n03wAnp){78K8{@rNyvJ;OdbjB*8KjSA1rcUs-?8!)?yd~Rp<QOi!7{kYg
zWB9s5pA~(E&}I5aoIC6DX7&*7A09@fytC^)(-SP$Coyg$5wAwF#cw3X#KUm;o5Ew^
zBBx~wN4VFF5LGJ{U!P93o(=CexKJ0nTzn4>9DBHo?REBy*|CH--{r=<BvN`lqHycu
zPQ&;$jBd42yxtpmq$YkTcggS=8$iE-f%s(l(`07=OP{7Pd&^c<2TDiG#YlX`e;RnP
zgjY9Yj@8P+S9A~q`V|u<S$}#qnSB41z|G(&HmW7^wnGtSJqy`gFP}rxaycyirB}Tq
zlX6WgDvzx38)lBz`H57un~d#6d;T3}k9B|(U#5Apd$SMsm-(R+AQ_s1wNqau;FhXS
z<Jr2hi)hoTi8ZbRmXa`M3A=@>95`Yz;Sb#zc+x}qf!A|r(pqXP)?r+>maq+8qPz2E
zkDKhIB}=*Fu$1jGALT~blKWdSzYG?0Ct)6!b7wK8Y6c7ETd;7M5gDly`Jrygll3P2
z8V^;DBbdH+6x)7`rnHY>?b|WDoMpm)kEe=mPa6-1K`i~QMP5cvVk7!U$6;UEe3M=3
zgEnzp)wx&Ih2WQMF!5^3jqHA;&(}b8<xmd&9)rWmu{_h5Em@NDQ0<gJp5%=UJrj>*
zg?MO&CG%pEbi0Z+DYYbyD-WfQ$}$$+?UBTJhjFATRD9;9c+MP8Onnndc26cUKynH9
z85635WWXq>2;SPrd#5hIh`a|f!~Gq{lU<|8J_LWZ8%buF3CsJNV5c&bc$wjb4VXdi
zHPiXoWDzH^<B5Sgztioh+wQ<yUk_d!+Qcc_2v)5OqeE;6%~brEQ4x&Fy%4N6M`7JA
z0@Y!D{NwIJBj1hudD@b6*Z-7veSRnb?QSV=yw59%3yvtM-OniDGtVjq<~L`aNh`Dj
zZ?pa$Qxe@TDD6IXr@_M>EN<S12IgukUfv7yo86=juq#fByJBtLnYZs7@Sx#q<>Rww
zikaavMOFWaa(rAT(oWXHJ?W?NQSGkcu;jXO>&i`~ujg0ght6-szJGn}%Np?dVsq*}
zYRk++9q{bjp7mcEOZRURrvKZRob!#*4{pFp&j$4B*@!JR^@U^pPr2Xng>tI>J!Sm9
zTgr#MSCo5Uhn2l5N0dvdXO+|QPAa<$suj%^2bEp^m5O0>x>9o|PPr2ptLW~EQvym8
zlmjk_%Hr+OO8f9A<<;OAC0w^$Q5&#d8EJG*@%DP9m~N`iY3j55Qa=BUD&Uvs;MC;j
zmh_CNUdiWDpFCPF$)a9hD*CR%dAyZbIWC$F#v!bGFVAp^FYkT*Fl_G69KqW!lSK0_
zc-zWgE$jBXv3rFh!}=~j#c@7|#>}O`;c3FPY7<;DSo8_s72|~0ir<81%9Wunl&sK?
zN_5wz{F~5;K=B!u)YD+pOnqEFi<VpPcB%akJPr(Js?Hd!btYn>Hk&^Ik{vK6jmd(y
z<6_fMip2-;Jq2rrWUNe+`S-K<lybwde4obRNa<nwo6JXfAGDklN5%1Y;VnbiY8Hw`
zzCT|EtY?GxIlqX`ep6OBt@1);zmvJyd<)A*ZX&<$Izp>G*l@s|dlqiY_{WXfkIV5l
zU%<)zR!llQ4KJA^Zaf@C^gK<Tt80lbRD9K|P59JBaOPAPX^C59R?nyJjc8)e9AKny
zYdzCWGHGuDWA2K6<H|`M9hXkI+j|Mi%cP{Q_>|=i9e!PUv4(D?v8v>#ZB1u+avBwZ
zX=q-|V&-1CuMaNak;Ya6)MdY%oJCK$Q4M;$l_{c)Hj6w&MEFq({Esq0bWwFC2N`!j
zI!D?c5-v>g&d-+;UB8H5)?2CXkxBX24B7}MFiqaWQw;MsVvvXN@&dBjiBHc`x}cMD
zv7awlt0_6O7R^(jpWNKT(+D^c!}_MNTosNkb%*4iYKag3axB}5qY1E%;EV9Z)^?&1
zS}$GCgQFy$Hj49aqS;gvPTs^Y-nR>2boXUUwO`2BZ_@LjBOO}<Co$-~8yPK^Q}fXQ
zi<YxF+(sLnBNmhzXkztElS<Jp4*oryep&G8rmXkNN7MHB68@Vyf))KnNVkI?=bbc(
zd@S$K8*=M%vF5<-*|=YyOHw;~rd+gRzx^U~%Vu%^zo|sMo{C=wGZuXsM~a>?rE(+t
zGQpZvABU6BwWnzM+A`o+2MT+4V#SC~qVes|#}5P1)7RmHn?6nhMicr^rLwEGT)CiM
zqAZCkRSH#x5a^{#y-xa!A2);=7j3#c)}d|Ka3*%uXK1lHlY8p2++iezPWrSi8G&I3
z`TZd#`0t&{?f27(c`MzW7B;fa${jCtIh7%jAt^j&)uW{x4Ol|Wa7VejI!bnuGuy7a
zVNtw_;(d~drM7{7J$<DA*N;5)0LfkU$1~g?l|O!j$j&_MlK7hE2eJA>B%70jJIsoc
zF4J@x%00kxjPxoMC$sZrE<ZxWM;{kRw%~1(rST*SU-Q?w07JpsXSeMbzuS%%O=h!y
zk~NEtTCi@YIT52wSlG`1eLr)GcL)x?9>L%0%{=k)!E2`vnrXenQ`7^s0B!P1bVZl0
zL$To387&R!FB(AO=aS!-J%|tOMhYK3fyttEojY?54Sua5@{uRgQrDvG;w8JEH+2i0
zSyZ`{gEmW1t+r!ce_Q(0vtvNB#XMU!kHw+0_;${Uh~E}`oN7e0(?pupGv(<)V-}8t
z?1dxfUN&6%X0`BG-;0XI1E}|Q9AP5{p*f--=bH@T%?T|UC5Q&+Q&$|<_hyah04z^v
zv&34Rr^Y>~zTb+Yu5HQF?ZSYD8oW>&id*RzZa9x6&?$s5!=i}n6~~vwqFKm~r@^v#
z9;zpEY>;%5?~kW{k!V~BV;S)-oZA}1X;rL;##4oFf#a~$G-OYzA!k32W4+fzP6*y^
zc%v}ENg?{Pa3#wPWcM-R{Z?bqq!<$BKa|NY^axTjV#O_EI^Wddeb^9=`<Y?mZqB~?
zW~e+g<x7$^Zt6C?bnDOL(}Vfv#5kq~n=rH0Tw-0qm?^$?(>5Df92mg;J0UEwieh5x
z2uURMlN>)ET<2^gbz4j3-2bH9X!lJqy<VXtoi9^{8yr#Uv`#9EEYB);H=S3m`5afm
za*rv^@6{+3|JEsb!5v6&eX8spxmLNkXM^%7w-=M{bd%iCZv0u)m4ixWChzPh853WX
zmS>(Rmxn%6)+~LjIDh!0Y$&V8hPa<f|ABXvb}vpVce5@lCG$Qi7uTOvMsK{P`1(9j
zf-L`2T+-_^saXfs6}3YrtpP1sG@+wbLpJShz+LYKoL$xc7S$(ldVNN%_)l5t_f&aw
z`=;V}<C@Yh>V#q(eMs4Ncc<d4S*SSlDOPUB6e$g}vz6u@k`;~gNG0laoU+p^R{4G0
zP8oMGS_yQCQ6^_bE3H;XDR%A=N@9mt<;dl|%IxXY%Dc5yiow<0%Il-MmGIN~cnvS$
zjz<B*YQ-PFT{5I9@@O|Hhq^DBB)DW^uM^CT5zg%Gy_TOx*P_wNo7M)tqC@ke;qnb^
zd$X3C6`ov7^2A$vEj|m~IQ?`sD{LlF@=D>{)lp1bJ(7<zhA>i5XNF2Qx|Q8iR-byR
zR2M%}nD$tSXxxyzz>d<l+l3J}eevHpgv}MhneU;4+0>4#3hTu18^gGfYQn1l=F-pN
z$(9jmlnOVT|2PKsU(qxdezLPgEK9;fXC*uAk+or@h~MJd)&MGlf_Qy7gz0snrF$pd
z;B0T22zN3;!;2oK>sXxNNzil;7EW;`v~mquH#d`X$D6^r8zkdw9f?!r7P!TYX;a)V
zlHF@;|COZgosY-y=^U}N;3&gskuaF0{WN*munR*CTk=<}J@*pYv$?Yq3spRLX%)ze
zQ_;L~l|7>%MEYr?xlta^*OTd38m3V)IYs)FCG*!Jh5Pl=P>~z-{AOv`ia+Gi-9*{j
z6WJ}^ny%s*JiIm?gSYY0qb*uN^%#7G^NIT!Puj~IhL%W=N3hIM`wK<8P)5!2qof}`
zO0!P~Fg<*bAJW&=!f7v?dRG$MuZX8cTRF8f70bSI)3S<ZiDo(_*K(Mro=5(-TyonM
z;*uuW<59U7$8E*1Msm&Hi+AB&24Cf-qoWbS<L^-%&W+~HkVL9wj=!%R$Ma>eG&7VO
zl;1&w83xIo7l^Zk2j-Qo-01AWYA+XNtX?7AcPr5uWY2};3pki<f#&+j)cl!<ZmbCf
zKa5BnWr%4zU5+>nq}@(U4t>{PY!g-4(Zwfsy*q<9!Medv-g`9a(OSfI=*G4Aolur{
zqMfOx<Z&q&#!Y6_vnlMIJ`J_0GkN!UDe;mMr`dNAEn)<3zfNVo;O*(Q!r`_UM-Rc<
zpAo~Q4@-|#>BBkxL+(>^+7eye0X5Z5v^d$3x^EqMFt<HZlxCu>I<GWVPAEp_E0w*+
z%N1YO5@q53txE49naYD!JCvyP2NauaSCl>Nv@v_BL;CtId<yBp^()<RAE(YX9S!=~
z^`YO&-t_z2jj_l3a(>fP{&TftT#n4DiaF{XC-d5UqGZ&}k^Y={(gU=V-|d(1;^$0$
zFSH?BIHmR8Zj5qV#oO2NEVkQ#b-8qi{`5sF+g~~X{V_b_PvUn!mJA8t_tiiqEeaw{
zE0TVZQFKd;qu<Xol9r@kzaW<B+A;jRFFubGft1Y^zJIzuk>bldYVSvf3F|l)znpnj
z?2zIKR?e8mr{wAMjkBO?h&ijz7)#E-0WMBc`Lt#@4qJLM_Hal3WVPm7WEZ-e?LtG7
z9%RkwL6A`o$t&r>mup>kkUap0F$2Vhr_MkdJ!Zz}v$x0u-{+?MvXY+h18bO<;Yrav
zH)alAfqKqz;+s3s-gX&xYV2q-TQuR@7qhgJE#uxVpyON{27ApSvhM_*9WkU(a{~Hh
z#uQkHUag}sIja=9{+4_Mm4V_5?Js>Zy)b>=o0C)38QMXE4u|@(FR2^qvV+Zu=*h1W
zoyhszg}`O~sc54mxmkS(c+ib)rL8D2Xe)QlE}Zfm!02+(uX{`MRlV_S*Yo76WfW#3
z<7hEIft;*(>O?<odA}3ueOihirY-ePv?Ji4+!OPA^RZrEepC;n)1Be?$LdS2_*fRi
z8)E!n95dcZmeOGZDqko#%}`h+czaazc>Zh0GECQ)9lFEB-#8Q{SeqR0;dD*Y5kI*W
zR)V)-f2Z<eyBVoACIntLWuD+|Ra+apKZz!(vNP^ppt;I~TdU2vsWF}F5p#ICR_=c(
zfxMGF<j15ae(Cn4)vGRY=WkDAwO0K8sX~(76{Y`{T4ho9gUT_@G9_^3HYL=dP)Yut
zudJA{SGf^erPOdpX}6+UX<&O?iSFj5Tz|1anWVl}F(_Z76nEOFj2Z2#4D$C=G^72L
z?Dpx(h0t6ja^rc$Z_hJj@Z2ZLxq*)ry$dguhKWy<tn^1p;Oa-p`W1D`qr}TfT8EFy
z4EM9jbdzg}%c%!Sx4CbXokKq=UhPzfT+|rb!e7dsz$eP1<Nqivr(IGECSFj^D_4{T
z$+r}(u6Goh*LM{Y&!@_&TQ?P@_=+-5<%|;NaY!j%zf+ksI!_rHl&l10MJPVU!<1*v
zQOd%d5z4S1%aymfOO&5FcH$M<q7*a<RwhSBE5nLImFg}LibdO4#cb(bB{ONS;{4AZ
zW%2Ia%1FQ6%B17@q6rhvTb1m2tNe(J$S1;JC4CmJ;LeNX)U$Nv%zH-`4RS_Zw4n9t
zujaMkX70|Gz3+=B{_U)Yudrr~%WTZr*)U#x9?g5tW98l1Jo!8k^Nk8;(?;{sYb3X9
zhoJRBX2uEkmHHX?6tmm+6!Sm#6$jhLN_L-y{1&_o``e9v?RwE{mH3cn4<gQ0hXF%6
z(r;EL9vBZo_n+~cJvRlX{=tkJo+iHRP|4kjrkUXFs$nsbeHSZTExxS0;mbJtKt}8o
zP2-dRUKa$3S2hIoUttUpotCQQCeAr{(dytv{N}9V{){#JYbx3X6IcB1t)_V6W-gBO
zCgi|+@xQoBo}z=?I_!CRV*$I177+ewKDR38@uZ$5Er%Ilv`TI~54A}1*Wi;pKU1vU
zDLXo}=ke6`(lgSYY1Srq4)bApy+DS}l0C0$2rk~yRGy3GspyB6eTbpKMaf(<OktGZ
zZa2}0X%9=MN=YS1yaeTfw|!nmF*Pw1$BU^bE`qy)w{rw<>jp^Y;?ywIMccaWv*fWI
z&cZuLwBCZZZwre^YAkoSgc8hL4^dNlkj(7|>8B%ka-u<Y6^{6#cy#*MZ{efLa$MVq
zUQI0>AK`_&s)>hx;udPH-zhpd8FcNOi`t6}Ce)@-=$TB~!X$am#`3J8cmh60;d(og
zQMpml!z_8S!{WHLC|0@>MGy5Ohz15hl2;#y&Lj^;uX5#Q$a1VUx-hiW3a(eX@>ljv
zz1<7gBDxOa?j~q<HKs?rG1d)@={VhxX3ur-snV3JVNIDcG`RYp6Y70B@baoE%^nWo
z%(C7b9od`np;`<**^M6!+e(*FN7VBN&^2Ws9wFjYJTICO%V}hLO{1UQG-fAGCT+n`
zjE?K@PeOls_US1)UJW*BjHOV~ru6!7j*i!(qFR@ZDT6sPqAi16zbP%uzboN!&3Ku8
zMJac>s<ds~jE67JDJ2b0C>C9JDL3BaD;t~SD7M>^l}hgnB{6un(qz>ErET67W%10r
zikW{4ULWp^(}ph0Xdqo9p1rxTNp>o?9#p7yL-+pwQFPs5HNS7T(%wsnhBRp?G|qcJ
zk<t)JMTCqfp&}s~+RLmEp+QK|&?00-ne`1tgzOoTBpJW=_ecM@x-Ok_KIgsP=eZaE
z`7@G^T`{9aLN;_$(~U;(teqa8sdSIyh9bYq#vb>j=}UYmH{XX0s%Fu8>3MWJVj<00
zw44_Dg%gw9K)!z>=*~*6C;E^;iHS+H?qw3GbG(gsn@9^e-o9n2^t@vm*&6Slj@X^F
zpL>x?(lW>`X$yTS<NHjlOx|0wgEXr-uU?Wwt+Gip?Y|^)aNNr8;~Qy>|0-%+;Lr1u
zh`d@EZ4|bmGoCYOil{pcSu%w#C~?1~EBC8(kEgH$DztdMGA-FHPmJSjSluuZ`KiRS
zo`&)qXH~lQW+?6Dc&o95b93(c<ayhWj;x%(=Xq<oH{OZHYdF&6u%&cj-8!1B!o6Po
zt0~Jegud-rMf#ONbZkK&wKe$Dk-3ZLpv*$LID0X*WcX2NkuQCY^`R+^Uer9<iYf|q
z=<%CpkbQ6!J+(QgoM20}zinttr!h5)%hF7q-B+HkL#=Lxy#L6M#?2l@n&HD}?g(v~
zk*>uv(^aV~aTxt8)1hB2BRO|6l7d=>lg$)4Y96IXBmSsUleGcGI4}ykNTj1Sne;6-
z&^h(()ID9FD)=lRr!<lZjFmaxu0W%g%F@HTinQKTiLO6Wq5T)NsOzCF1*I6$`hGL|
zY-&x@?oFiOo9wBBYv4zGnn+_S?8xsIXu~5$%RcdH#uz*L9XgR7`c0(0n&ZgC$dcMt
zo6+;PI#l>skE(W<P_v^2onJ7Ktk2lfPkSedy*r7r4iI&znb6=L9eOdE{|)D5=x?no
zt)AgdJ(-@g%YHUxZ(l)<4GYL|yDiN~Hl(jpw0Lfr24#;Gqy8gL@XhWX22AeZWqTvk
z_Eh14Wf?+>im`a@G1xpiikR9WNW3q>tylHvZ@YlryarrRjE0RtBx*KpfTid(l(u*x
z_xv1e{<8o^sfqY8?I6gh5FHs!c(L>iKDxd}QT|KZiF=P&uh+=SeTgWJw^wFf#{-VH
zO3L5xclm9^PP>oV&ySI}>pAR}zQ&&)LnvauD5Vb{fZ>+ckf?YHtu3wCC3z9K)9Mfw
z(g?1e#ax|cyzy^B;hF0=Zgmew_%8gBK{a;Al%YWN5Dq5q!<$FjP_SYv%A?XC6%mi|
zmsdendnuZBEWsOxAOufX0j1G9p><#j4*lDLx|}SGTvQ8px6@eLRm<nv8nFIqJmmiW
z1m8rG;F({;R`Z-ZJ{K{Lx4oM}$d>O+SDLS+L>+VbSu&Om^L>~89LD{NQB>y~L-+Yz
zI`qv{o;&T$JD6rtU++wclJz2sP2Tj>mUsU09a6C+qi44T6#dPH8hg62>dOn16u!VJ
z@kf|*@iBZ0o*~Hl6}lbX!3iHQkNchm#tosFUu3u+M3#=b$kUfjMJgFoCI7aeWL!Im
z0`EG}uPCk$G|8YK<48KNERxpEPUAik?qwX8K<}pV3~ZeQvf>=WaGv+4$X}br(FCfF
zk0a*~n|WqD-$9wjP?T9TEvk#6pI#ferz@1APlV7_=U~!OTSpJBMNxFgW?GoAo`%|n
z(aZ@U<iYu&wEKSae7_&9;&?ly(UrWmxl+%2bCSEMNiiGM=!mQaz4@(3`G-HFTw0m@
zJ}S|wQ!?b7yM`1F#L@BIWHLYsDUIO1`Bl7gmCwivUM-<po0gEG{angyUP#rex6#^3
z5mfBDo!H=BIy^ps%nW!|>(3;rS<8J_C->8vmHTPkwq%|In?Op<v1ErWbo}c!l1$pm
zb!xd(p_I$_`#C%>`VdL0@XnGYCukDq4%Tr@FCWFb8nKIf_b;HxQ-L%tKAJqrH_?Mf
zE9v2Ao<ZXl$1{vl$a$hU4fR%};m%)hqf?)LrS7KIceykvavycEedM%0lXCyK(fq>g
z^e}!WdDQZ(_jNnTy(5FqTUk7Bp3k@cZ6mJ_sdNviRFcE9Wkc7Jch^#K@L5HB)&*09
z5ZB5Th0t`qOWxVNfF|>^+QP??{%|eS<|VuXWU(!MyJf<2c1-B;T&{nplB5MivYb;G
z%5%7s=&rmP87|YLj*r^hd!j>MK59|VUS)cHOM#A@SLE3=Ds+Y4#s7-%*)(3B{#uNp
zl@T8FMPnST$upySyUnN+V|g}@D#c4^lTOnp%7roA%`u~6WJV7aP00SN8NJ$TO2$*(
zV{lVH-cI?6y=f98t#}X1mt99*Nj+k<%i$*5h)q-L5vWuONrys=h~9^y$V{~EOhDk<
zOe~i>fXOr4u+vF~OwOrO#9Th-jaH^x;o3AJR+BbP)1)S24GNJuhBD3*R*$hHDXwd_
zSmQ*YInzmL-c<VCIhp*|JJ8TpC(0>wrL7x1=-Na+uS?D&&y@>#CvgDXf62AggX`(x
z_6V*$kE2gL2{bz(iKfKF(<%P^?`}&Zo6ab@m>5SRjS?uC&#SfyJa=$nIvv1P^7)m<
zbC9;u7Vh=S+0JvO`Cg{*XcD<`59qjTuFbl?iN4s)qqS>;Xjqs(?_=k4)?XXy<37l*
zrWWMcV?lS+9Y`&L=-qE!+E6x*c3%{y^cPjwohnZsx7@(N=sURmRf&pS<j6BtfueF1
zXzd$y+8jKLS|;i9js`<2oj!r?RaujIJEQaBj&$ox05!}CrFl|oXz`Mjv~0*S8kRqq
z;%8aWf_KZvA;6ysxZeC&v>$0=F$wthw655fy5@S*=>$6(r#p`OTW&$CrW)ZT6TxQW
z;9K)c$c!IAd!!WCQOS@)s|;x*@tJdl99=Hsf1jT&?fR}xN~N-t=O#t-POH+<IvIK!
zHG<;J)oK56S-LMJORs)NlUjfRUEZKh#Y^;f*QO4Yy;G-a*GH0+zdrprVn9DgpV~Nn
z9~;o2S8~eqSXqHy>B{mh!y#1p`X6-dRp`)Pt`QK?p?G0KS`f}PwSOj%Z#MTBEw`s-
z9g|3xYwad>fGCVn!WBk_GPd+Q%$l-HEy={nl7_ljQp-^b-a*NA1&?&d?x7yJc^lIH
zZ3?7OEK6RNa%A1e^9fxh5tl*I!PqfmH&2GfOp~V@MKbinN{H??eTUS)PZ&sl2Y=`H
zP`39WwKQAGk2fUUgZ{|OJBP`;ub^wgbr{~c3+D|jxZZOGn>DL&*`y3jPmUw}>mfL<
zj78+3AdD+b!}zC%(KO~f)ZSf0z~X$|xADNEBfxu|sTlgt6N|RbL9MPI;v*7q@pT-=
zOpV8$7wH%!x)m3M3K8FV8z#xGAkp*^$~&GQ%e50_WlgaDbsjs<Ux4|A8wk?5523)v
zh_!o;yXRkVU(W~B9~(lsX+IHv<2|}RG(yL%5wCY#Mq$or46&-mw((6+*l-)G@73Wt
zPC~?aD&!X3Wk$cBvK!&MVYa^jeO>{0T)Y%xg_h#`Wq({&Uk2fOffy0t56f4J(P$Zj
z-j_jGYp@eTUZml9`WBcaW}#_n4UXEKMoC;PhMuTKoJciRa=cx1muG-}PN3o$iM%g0
zj`m%TBArS6dG`09prs=z<g7Lw>@w!vwN~WwQb1SZU8(=J3pLie5FK!#u!EDxBW66c
zXj_uJl?ADYSkRNNV`+$w6;;|A^DHd`su$Pi8B^U*=Xjfu`vOlzAHm7wF_K-L;rjS4
z7&yK|$*hk!F!dWeKK(?>bs0MTK$f<SmM2X<FPe0z(*3ZZwC$2S{a!bWLUNtx&9oVG
zrJwf;bN}mauC<eiTS~4Uqv)1P0$m@LKq4V=w97o6Tw4;zW*5(oVe!;*i(_cpW;$*h
z!|&IT+~*xZ!z@Ed^L7aL@C0+M`YJMB70xyBQ9MIqGa2S=AZ8Utuku!rXy7vP(O5uY
zW$skMciJw;Tq#a<GF{0uCr@d8dKjTbuT<1XuCEfl^a(r0okQxua}X+&p=3UXZTV?H
z)H;P^U(cZYU5n`CvPBeGvw$+Vo-BlWQB)_)qTr&1^yKDBsy16o@tZeOP5OG;bC&xq
zbrZ?=NfP%;r*V&LI+ZA-(Z;wW>IzOEHH*zW3t}r}$Zn@P|GktY$LGvd+ex119>h!S
zpyMaEleJ+W{rzt%y$xMN(#vO2%>#E*GvfbuQ~(Wm5J}1_H*vl1GD;ovptE_=bnk5h
zY3YTNxV<V(TKW`HZ}cf?!dP0`XG2dff<#3pQa{;~>JK-bFS?E6?GE0#y@O_~+)1)6
z8MLa7cWLoFp4QjtG`1mydL1`X#@E&KAb%N6=C3<^_$vAo9!$^w1k=T5Tz6K^z42>Z
z$Rc4nRa<i&>5?`n?A0XKTf<4@lsX--RVDo}Ws*E3Nu`%$dDf&d-3?TtTiwDm{^2h~
zuau!T<7H`Y^$6PZS(?6O$&kli1)A@#K!dXs=<Z2b&eO@$hO2TEwn2`HXRDGs8%}H5
zH0Zxc!}zXJfwa=J`0jKR4WDU5LJ4M+=wnJd$&^Ai>XF*2XXs9Oi~A4WBCER_qa_AW
z+|`1=KbkN!_Z%jQmt#cxWz^rOfxBEO<ou6d1!X}dH3LUH5|HhbiHkA^pwnN0FF6OH
zP?C?^+Q)Hj&sp^Qa(!f=Fy;TKf&cOvh@L%yqYG1UUd@sc{2A|Faw46uHWUr-!i+nU
zd0(|X1vNQQXVFyhTINno&E9m=*N5W%`p`6^0NPQpl3q_+P1_vTbDlDe4lRzR#8)vC
z@Fj*4INrW$O{C1qD9W~uqlSm^)XndtS^PdK<dRO`c(%0Dl{8X_Ng?5gMADN?q%G^>
zDP&F@xlW3q4;Q%qZTWnj-7=5f?DC^Giab}MhI5KClgV?u6&=#G;GBj9h1Ob<!dgq(
z*gJw;PH~Od=NdFVs)Auw5frN`(0$-0Ry*Ftr?NYkk=u>zCH%AMQm2=kPjddIN3vJ>
zK6#N9J$+?E!Y&}spAK~8asWM>6haq<t)d8VZ~C`=8hsqjc<;+N>iOY74@{=h<&2q>
zb$$Wuc)N(EW^%pN-C5MY@m6f6CHeY4gw%-B*m<%V;#()eP|O2G4PRh3b^s$cN>S})
z8QyO#Lzg^d$<aZAF8HXB=1CQ@h*76>j<-Dh1U8=kkd!4%G40Yc*GqzQvgPRJ0U4@z
zEJ0RE($wcJPbM4XNq4p^wN-1-QjWLVmgtlFCOvvIMUsZZ3enXVAxdc#rlDd(2;-#b
z=Dr~mTlfnbiqz=I;SsdrmNv~6GNi|ojkpe9lTPv-?(}41N-`ZygMMS^do$>hDx-!O
zj3PTY@6WXZdc!R#&cc%3nORcx`?2&+%$(~Em8joHj#TEzQKF(0N#)9LjU4Z-;d&ue
z`9GK#JcJxuB*+vp6lNsP=M5QBHWZ?#|9W7s<P#>$e}~XX9T@a#hkg2Lyp&!6tptCh
z#4p9p&NbK~I0l!(Run9~gLV6_ps1@7-`<zv@~NY^xaSa-%!<XdEkOwLoP$drt>HFh
z5t<zqp?>B`NcYTu$}&Lss1w$Uc)@Aw94t{^gtwuIs6P~kKZoPsIU)(SCu~8i-(g%x
zyN+SYUm-c=C8B3Nfhfn@Uy@Bw8aR*Oy7QP}bQ)V;UdCfSQ)coV?kC3^2u^6mjj%_^
zS@R6^<}H#wa=u6T8X`?E<MQhoJXlbTsZ~`-?n%Vh>4}gy8i)L4l1O^r%GQXqu;xuS
z*wMN7*@?;Cm|L(EZ|?ge#?>F+;}=2ttsnlY@PmcO0Bal5%M>K^u{YiWZ&WrTMP~~h
zMrEN!wi^HKti>C>N=PV_qei@n<GVb~4OgN^T}o8CIgZ9;#nI}6oD*5Wv0#k{O<p{b
zW^CbI_SSlodee%M;%rFnp$o-vyj@u1LV4LPRK0Z)nTCz0?JAZOX<<PL9B%`5n$zYV
z6DZxym@+f<srR=Y9Tw`wh+EIWO;FG(>cAL@$8cQUfkpS;V7T==JT!fW4StW&?erWe
zXa8WxTWOv>D#LfY{JeakN^?gICGi?rvihMyK__&mb6^yGXmaO$GIQu@g+I-SSV~?F
zk+k(w6it2^MUEF@N!l!)MqcH40NlsZoDomIXT*?q^=7&{B8F}@MACYXwRDJUY#!LV
z(&$KM`ZIeawagEvZ$)9Wi0k!E{0t{6KHJ#Iy6`z+5#1B<<XJQvYjt>z3iot8T4YMD
z6U}IDpB}Y1WueX>9>Ke^kd$;5UM*!XH97?^CPUXIsL`9dx-^Gt8b1HWbGu^}(S40Y
zRJ`1e)+F=Z#(SQmBRq??A6-ZmI|6BG*;0C0zl{Fv45a5ff+<8Ip4J*C^8714$M>X>
zGxuECq_3iUH=Y|TwVA|*$58~=@$cpQ>$89iYA#HsvY;K@-?oE(6>&emejur~2GD7~
z50BNDN1_fcba5Bo`}>Wh6Ll6e@%K79ptOvZOqjuQ?Sv^da1chE*I2nH3CG_)L7Rd;
z*Or-cFQ6@%%;0`>S+2X(w5Qkmr%->$HhK}agQPupZfai;8C?(Jxl>WJ=xrLsHl}kg
zMJg2zT}P=uLwT+b$6KFyq(0GuL`QhgY#k30Si6%H=T`f2iQL{gklhzX<2l~4R84wc
zJ)D}u)Tx~F_w~-ov~xs1bZ^U$MW`~J%2S}TSB0rn^%ph=%TOuD+XjAKh5nGD_6BKE
zoTfkx^A+f$w*q-a%aW+AJe{eLqxAK1)IK0f^|Sc?tV)_Dgv!xRH3iaklckXk(qv>{
zM2EGEsGj33S7cHB20aS&eugyLw}{Gri<m>7p!e+)GOjgab!sCn9IV6pKV>MEFT$tJ
zC^%dR!{*tV_`KX4&o234$neeZPuhxGEtwcSxfFGq4}#6l$E(I7xVY3IQ@aj#ThCzD
z*BUfDtU##QMvQ-u2LGoe@QghL50xh5Dz%_OqYI+C(qvz+O#i8w@eF*%MW|CKG|ihl
zCvmL(<3lG){poG$N*eF6nid`A-pcY2Qu16w_4{M!gz{$I#kZLX?rfqxiBVKGF_It=
zN1`e56u`Y}w<USjv|b8*=g-y;&czg%Y^2fWBX}-n96cSniHybkXh|yP)q}=T+EgEM
zQeQysF|(-4%awEv5?vOzqEoUKWd6c}wD|Xa_#J&3#zt^$jtZTBV}P#fmPqnB3NPzS
zWL0HBe9~<Mpn>PpKF8mY-DtEMLhij%boaLgy)QQ;)m2v1p3eKeCxeQzC({aL3woy+
zLP<5;^Q!4h54TLC-{%~-c5W=qT<b_ykEhYB`7<dXaxP78nMcaA=FsT-vuMngNSqbP
zKveWG3~Q}|kHuXU`%4&^H>|KRg%h|Rc46<~Qg~dHp<N?oXicp+Njpi<lG!Re!%T(l
zD5#P+=UEeEzF_t9ejHgLOgd(gwCSZ7WiFMY#VyiYLm|O6wo-KK)(40*)<etXBKKd`
zAw<3#r`Pmg$JcMT`SCmEWd6pbXdzPJcpErRgdif!dpw0HA$<UUn$$?>nFj6j(WE*5
z47mTxkoVPVlBu~C*K-?_^Muiq$>+?xxwaIiz$n;>>yQ=A>9ZEkaBzEo{mWYLV`K*w
z{%1m8B+SX`oC598lcU34a^#dCLA}L4VcOn@BDG$;{PzK)2Zd=|i3C0VeF&Oyg@{wS
zf(`pFp?UdxXny^Gvav6r)R%|HUp8Uk!qw0p7KH5DC1@JG5Zw#nAg`5(<sT|=-JlZP
zL8oC+Ux}r!OW|;03-*rRj7#HUaP?LYW`CZH6M5DUsN3Ln|3WO}xSIc69l}KnEnfgN
zH!B1TF@)V19dty^M`mdlR&0*L3E?D6y}K2UUH2jVNjZXBYVba(7FpgG@Rd4|x}*_`
zKhMK6?L5j&YVl*(X@tm~Kw#u)l-A#XLgQ_)=tpS!^bB1Op5p!MM%>)jfDPO0QJr0b
z@$;&2`AroBkrjBJn1He)aZp?;i6TKO^KWWk%ZFcO8S?j7PQ`N;FjfS$<s#^_8G>EA
z74Ta_9*-_bAf)yy3;XzuZ4%RmVyp)ggf=6%HXYfK+aZ2>JIwB8z}dPC|H+rbVtXZI
zEad6jXGNM0C7SnJfzk_hL+O$QV#I8a(h>|A{WQowZAa&!AJ{Z<B;8IiqCeicRL1vu
zImadGbd(+S#!ezZy(LwK52ve<S|lw#oJ1$<(2-LUD6v+DdNzsEWx+o<4F7_|H=iR-
z^D*YuG-B4^HRxTvi<#s9!{ozv(AjklkG<RA)BGAjM}+9ecVYT8C`~`#zCb|q5%|5!
z#n{kXGz#s+)3$Zk5UENYjqWredj>t=Sp=_*my(-s2sIB`Pw$$eDD7G-m5asGl)41c
z-M^V<9`hae^NpnM&%F}&BPl$26WMvKCAaFOq{g$aH+z3W_$f*9;k%!wlE&mvU`(<Y
zO>Z)}7-zON)&J6@z1uZ8ud70nLRCm|r#8KwZ%Y5POlY0kX!6nCgZ#^J==95ir`1`6
zIh4Wn$}vnf=KH6OLuv6{Ez*kPxhHwrq;XlBrft!o(1$bVX{;AbRpGj;es40|zmW8|
z1XAVNrSx#eG78{+ignpbNxXP1&Eb3U$}?%CrM`jB?V;5DVkH#@ucPXo7#gn~M<wM6
zq!5@+OM)}#z}5sl=O$6p-UKqA7E24X14+th8HKWiG=zHzx-U4<%5HZGtud$HBh2aA
zQ8W6hqEAOXjcLgRA<|kfh*v*5QMfe`!;kKS-R)hNJgyv3QkU>kR+8Rwo-u2^E$;_$
zrs6@Kk@{dR_4CXf3!5NnEDWL*_nf$w8<dj7d89-yG9F@27Yrwp!bnGoiRAtSc@J`p
zbfa%RPF$zyL}^FHQU4bUI(bPz-O`Lq?^@BQ^}H*qeHgiEDAC0(IT9|Aqt-Xyv3SLI
z3_tuG508q_!$ZQ9xW5mgrh`bS9m0Ff#K@H6?W;OzI$I$_z6Jxh>>@;;a%FkGg*-`|
zkmLQ|lH_V4N*hN=Qqm=9isUmu#~yh)Iz*PvnMzZNiy>tk5GQ?2F|H~9i)@eIIN$jg
z?LVF(V%}T4&-{e9n?E6FSu?(fHKN+R4o`Z@U@lq&U5>Zmm&5QzP80J>b@AutbR27n
z!Oy-#s075~`MD&h^~PhmWj>Vj4r2cBR9v(B#jFR7G3=f_@`pRYZ^t@x&?0F6SAzT5
zr!ZW!0SN;oINWm<PnY!IqrN_=%bU>$1JIGJZuIVsH~IdWM(5)^>CJ9``a3t6Qr4`d
zA8u>uOl1hodlt;Ofgt+O7e;F?uO~&-O*GIRNuB>~qGM4}yw^9LPH6I-vtbA=TpC9E
zHwMsrpZUD|%bPxmuBJ<+aU`O!iPU+%v}2hwdBvEMbLv=H{M(FV0==o&O`H-}h>_M#
z5hBe|yst%v#&K??;gc>c+b)W2$3|g_q5&)$EU|iAA>U_ZLbZ~AM@D91*wAcr{>ekz
z)*9rV>VbFC53HT7MA~o0(b)(qO5I{Zqf|i8<0ez+Uyk!!N2~N!kJ<)?Q}+E4q#A5Q
zGA(1N+-?B7JAOiVpEF5wj^l;z6jJ^nOY3D$KvXUS<(fNrW&Czn|1(A7$5ysA>^I9w
zGQ*F3E3ma93!_R)@o|1FeyLrDdCEPYw-eVxpCant0KBAyXs5C)$y9vCwz6+faTBKF
zo}%PlJcLHY%Fw_05)|YvPB|Z%5t4Edg;D3Q?(79z$*RLPxo+fc?1ALKHwb;~M}Gca
z$n*JjW4{oo&k^B$^+F_LAxsZ-A0lneSLhz@#nA$Jn#iBc|9ehjUW}k;!?dWWe<Zn%
zw4xv4*0gMvEj|5@=R=s9Q%3I#eD8h))6;D{U!n!dMjfbE?S%0NK7;#9(j>nBnfPV^
z_V0foE9WOR*7f1{%twf6Ey96IN%+npP`)r1-RAjNsdNYv9$v(m$it92y9du3_8@3o
zE*|HFp&=v`+Jnn+DQ^kBEOf`xL<?MMw?N&e3YcqFV)V62%pP5dq{dS0%t__%aSW<-
zVsKzh5Kf+%jGjFc;C5IHlFGt3Dk*|VKZZh&;}#r%w>4InGkhc#Nhv{8!vLiX0a)P_
zhoobPcswNqS*ly15wQnHPt?FRu@)OTPNP|)0a+KXVerR!?9Vxefb=SaT|JKY<|23o
z6+zUq2<=&iV7&7X!qy(b4|WKn{fm&gvjLND)`47WFjG*C4+dGtH}He=Q!$v|?O`QZ
z-OMDRg)Mh#U{6x6u&X}zSeeHoHmD%N?@dBz*FMCm1MV^L?0YQkoCIbJTxGw~-mnjT
zXPMfZI;Qv{1~vZaP_NmB?`yW>eB%xT_8vrhcqxKMRN_E_CL0r@%O(cvAgg6OTz`*)
z$V3Z#ykQO7HQvy8lYpg{bMdP3I|@gQq&P_<(s0(L@Cr#P*da;2lci{Tryb9Hv82#j
z8f4k5PX7FUn&_!X+v3$o`l}+{wv!;q)PIm4_XWnk9>F>F21H~UA%SZcx$86><SwA=
z=N;H3x1y)(E}Z^*4ed-Js{AEP#uudN)8`k6nR^6Fp5($ObSHjqOT^WV<rucO0=DD1
z?(O0vT77CJUDjSoGEWxMnf6egNfE^}cw*^j2G7sPi>HB{&178?L!)^%%}N`talaTz
zU!6Bm+Q%YP&)kj8etU3z`5j#Sz&i~)ROn3cFq-$$fS#C)rWLW;)bgMegC?CYgC;ea
zsL=6NWt#Rznw+&iAZPwtxLq!Qw&Whfc17db&rE!uQH7?yQk>B|2FYkKo+~IuZjOrF
ztENpmxEJC`y*4$y(dJzz+LWxMM=`Hw(xZFcbcq&`So>n?dcKs_aK8B7sQ`-SnXru$
zf~dzhfx=?B7`JUbmFk4h24P?F?^!|C+ak!}M=U9pB+!-URNDVEnNB1p@Et7iOsBCV
z>lRBPYJq&;x{S6;E~FR1j1H`IqM#^uDmY<I>y*rC#=I}sf2a#eraSSYREYcd<B>ZU
zgdoQ#sPKPlE!%}#=MUg6e|D=>CCU4N232y-?s1hfeSSTas<+If*Va<x^?d-NcKtzT
zy8?}@GNgg^W)x&2pgztyL_W4BIRytw^K+utpQg}lPiI;;&W_Gp=8@@-cqU>H=OSGA
zd3wiyF0@+Ft(69}(SYlWm6XU*SDq~YkCWlwadp9W2ygff-vi%JuF->xfIdv;8pOP@
zL&$8g7(M8dqSbNIG(ujA=a~J(uOUA%#<vFrxxEm9Fb%F0A%m?V^!Et=Ow4&-w2B;s
zY?UV)&fgjLOOd690xgZ#quUjt)Ux0&e$4odgJ&LNt8Y7o7FJ_lT{-f)DlkFo5?;T(
zhHcM_vGex{gr7N%h9({4EglMQSved?)ImYL7OciZqVQEB_SJ62?Vt_t`WA;Nsi8=F
z7l4=QsfZH(!L~j%$C6fi+*;#|Sf3>*S+)@G`WIl+_W%@M*@ET?B`7Su1ou-9aBsXm
zF&Q(SMaSrrGS^RVytUx^>gX_@-4N$b8)gSn$L7_v<)k-l9=rn44NaIH{Q?)h^Zbyi
zbu_JeBl#<=rJS3aC_6EdE=$JKyOwya5eg)6n`Ja_t{-VD%qMg1N49PwI+<=oUwM9>
zMwbX#aPOY(_YctWH>c6R%;@IRy%?dDhrI^-Au{VA%x>w@&jf9X7~pD*#tIhtSRIM`
zMnO~B0QKRP=q=a>(T)sMhHitcY9{X8$VQ7~K0Jq<!UORec=faonP<i5;dtILx=e!l
z1vXUoALDm40Z9zEAb-wFieVJ_y%<hi*&}FYogtlmYeEw&2XLqECn)JJ*r5Tu4*HBF
zom<#0I0@%R%kbixHd<!MW9d&JJTY%&!;1%4RNryNJdQGzCxxtjNuFR#V-4=9-@w9=
zKA1l%5bL8dq2D%uy_W|OsP_{Eme1jl--9e2?(>`>N{04BNOt;XI6Qg|CtY!>kN6KG
z7FOc;%5uo0T)@+TT8uthi&yPo80s5>FHcsWGB^OvGK-P>ayH5p9pR8Fz?Srp`1++C
zud?pqh*Lk`f&M{Ty$ILVtMTB9;S}{llQfQTPFF&g9G+Wq4#0*yINq-BFr(||Utqwn
z0}oHM!Pf02wqI_>UYTChL<`dh#<>@M?uIY<fmUtqyIgz^TR&WZ>5WQM1nx!buq3Q9
zjld(L^^o&egK&)y6b`M1UT_t15AT6$;vT4YWn)a@Y9u*@BJS35gl%4ee7M6_!ve8e
zETBCj4)y{3v)sdHq5b9Xi7myXZ%L^76Mz@;0`%(uHqMhF@E?c3(_%Q$|Bo#W5XZUC
zG8nEafy7PU*-PhdtZ=jt%x=hGg`)v})p=uI$sD*z`Qi)<z?Pj`V1GUt9iJm`?eI>#
zSE)dUdIOZ#G{D=c9AAq{5t>(l`18kMc#8XXZWdzjGuJtBj6Iuv2#a>)qawZl+kA>J
zV(%p!{G5wF1J#I_l#eTVJF%j4Ev~5up(?b835j(w$L1y$>v4tIy}QH~-8##%l^?OA
zo6Feisby@0)**INH=E^6Phg{#?_qr%4NOw0h>a;rW4cCl%vnJmaT~3%)N(O=e<oqt
z1g^Wj%IEaC+c0TI1)SGwvS6me)bgEK=$Bm9?|Xw?T{RZ3cUz-b$_qj7g3+xQhQMEI
zuv#~SXJw4TOZOrs*K>j0+`AIjC-SrHt{sI1DNxD{MVfk7jZmphi@%Pbxp8V_a8!{#
z=t<B#jXzLT??&7EYOF9nfElZs*yR0Jp)OO(9Cqx(-q$(k-F5(6428Ux?Rc2*7$J=x
zv5fEgZshl4lYI~D=Mz{(Djcg~a8zPB4xFrjo=O`UqrX5*d?pzm_owdN9yB7D=l9jD
zqVYr5QloS<m7R>E)`!v5>K{b5f(V{pS&P!!WoWZKi6zgD;NwDm$7xH4Qx$)bCVxdX
z*KT|9XR`QH8HTMtjd9MGFwL<A22Py_4_76Faz)bb7()8#zp%fl3%4GWqgH)2jD4p<
zS9CGvy*>=rAC*|Rt`xVA<RRgmDCrwMfV}f9Z2H@PnaNzQwyzc!?H^&FQ-j3%Mp53a
znKXx=n-YzFyz_7oeVDd{G*0=G>8WKr1Hqr(o%bQXHXnMxcfkkqW|N1!56@YiOJlrz
z$m&!C#YAkTRaYWu>4Gqtldy*NxcJd2j<<)7SkXSdhe)<7LDbg+@VOd?yaVY_n&yj{
zvpk`Y;)t$TJzNx0gyjqqOjfsoPr*1`p5Tt{#sM%<T7f|Isfbm!L42DP{8E9HFC4Ma
z)sn7vs*|aSBG<&-#X`v%oHTiaF--$VjQaz<B01iZHk?|17?Sp1bE+*CkjD+~X}D?6
z`@C&wj69>dsRD}PdJqN4350!PDbwD9`o0>I=?Mcm;n#pSUVLsma~Aq5m3fA&Jl!vl
zq#wP#*p~VUb5fq*M9y;vZ|TOsjb4mCIS99SVTuo_hqu8w#Cb?l(Jx6-c>5JM_jZ6K
z^kDkV4_xQ*f1Lir^)X*y(C`^1&psi5b5-ufpP<wvL&Kh)#P9hsv@S0h^<NWUIx!KC
zs?+hdp#!q_Z$tR>Nr={PZPd*QIBltidr1TA?-yh9e+SXfyB;pzbPz2*6w^(mV6;#b
z{*k8W$}>Z(j3-<gmP1^8J$zL+VQ+XSvK9rvNyq>je)Y2EIAvr=YxCFRf<bo=XiLn1
z(`v5QtO|g(W-@l?mcu#x5=NA^<H$ID;>EaBGs=mA7rD|U&TB|rnntIOcyMmopFVp9
z^S<Z?oYJ@iyOPW3UD*Vme-B`i(FNDhuaWem8zy=-q$}7!6>9`sb74nfTtg<cZU!BF
z$T_3}PnwrHj~-rhC6B9<=@rNIS1DsCfzM%s*5BZC<u%?c<6OzmJ?L!9hH1$@wASvz
zjc9+YUVfkD{5#CHzdp`<ttwcDng;tQ{!Acv=+1=P?9jZ{8zP$FxJ&%}G023~?rgYb
z9Kh?<6)3OV28){!IQm1Fe8%6y+q(NuOD#wKM=Nq!K7rnD<GhWdDdl|Aq98dn@|wzZ
z=vNHs*Nf5gqV^}+=l{em*}u5IXaEYIKEu-BGD_BFVAUjl1omkoW!6v_o^4}S>uxf?
z^nPYuf1CxI9A(=-7BU+p4>s!Q0>QecbJ_ZG1EjTk<59Q|a4Q7<&26Z;{tPWQpI~-i
z5enuN;ntQSl;1c78O`@dJNq2JmW$DfjE@kxdI}-!WhnoB3PL^AoD<lHeco%(;~j!0
zWh)@-unc<h{9xWP8x!qjpi*fn$~S4j)8IH8NDgBC!86f9{JvT#!ft;a!d{LSW{U4L
zX)E^$O%>9m126T6rkRjolPS4po6%rg4;H6?#n5{%@a;t-7D$}JW!;Y$K6d~Moc?0J
z+5jZm`Y?j)dn5VWZu<5LcDq-=EF%dY*CoQIeI4qm)}cgq4SGKWqi;wxmOJqs>(X;f
zUOF5qior;lW`!>)^YM84B7Cv-z)kI`khyIRM}Iwx?lXhNK6Cub8;gFsGW_Ca-pTeP
zsGJT!Z@Vr2uC#@w<7lkCYJ!*HDzL8o&742{WjAbPFrQ;=&Y#b0{mCBo{^eg5*DH(n
zQ$}LhWD}?s&w)e#Tm(#6hSh^<7_le@v#+he)4HAT)jy61r(#?ZX~1;RO8ncnAD53F
z#^Lk=6n`nel4FHf&T-cIMgjJAaGgYDK7xMcp;%@=rmxAx5stT4URL9rVLlf8OM|(+
z9o8R|h2rop%s{4-IbUyLi<{3gr^^}4@@O_||5eNWM3=FT&Sgx-YA=(Y;mZV?i`nz`
z-As*bWwN_7*&Wj~wr<C1*7vWTWt>$;)fp9R+-!jDey$kqzYv1C0tmYbaA^H`LG~PN
zmaxi^DXCPl|Jvn{^H>!QdczR5%LvGS!m_ruF}d9TSX5FObN*MtM5pAi&PNy6{qXlJ
zQ)f5M8~31*>r#HsQKWnO<tdVL8?Fz<$kJ1ce7}lP$_z15midEGQ(xd>Cg0oa&&7D7
zW~RfR^Iric+4?_utZq*Zwxk?Dazi1ui4-HZ^d^i=-(#2dKWw?s4_S`4*S7-~FIsb*
z$z<491j2Du6rQwh#@zP=5B{9Sv>B1F68D#_45hzWt0<jwKS>L1DLO5V3cI7JDqfKm
z%<o5JNG%@Ol;L9Z3GBFi1nEO};gMoG>b~!VRA&}!q_!gO>P`f0&cVX))425PB3yLS
z*&)|-_AaM|Ng)~1X?I{ijkqUV2G3cWphn&>h;_i392?xcB8-w1r`Yo;qPVS@%5USz
z*grTQ4N3(#{DI#cV{36eqZW}IZw1*JB%;i{b_3=#Pn^$wh8|>^>_mr*T<NIZbdp@{
zNj)8N=-gQ!%1`tmHz6O|D9gj}x#2Yb8lwUw7pi3Lba<UVO+C7ap1)W`tvftvn%X2<
z+RvzTpAnt6EW-8VVx(}q9WiVPYIggeK5{N%BRx_0!4cDj>SD<Od7KxQ!NNsjFsscH
z;SpTQ(Ki*ZiUTlLekwkz*<j7Naj*^=53_fUP?DGq_dL!$H00oqLmoPv3vjLKF#0|g
zBG|nd@B5Xh#*lk#E*es`t}#8a;okf;_EefUiB`?FrDc3?e0Yk0=FH={0cHmDUxXsh
zv*h!Og%OEp{J?LnWt84?8ZS?uMS*!eo^-Wg-21o4Oz(mAn7a^7Xv3cGUFiAq7SU<F
zP}e$(tEqLE_u(uo^v_|!C~30gcw1Zf1@_4gVA}E(6KX%;k$W3-@4rW8-)CIk`x)xZ
zn;<kY3S+w?F=k>UvS)0<v0Y{eJsyH)yG;;PNJQlBH0Vs<j%m3y7}-(+8P2P&Fsr~5
z;d2=CV<kGJy^$gujAicQk=Lz*rhZjqpA*6{EoGE&yxq(3Hb>qQb7y<ue*0P^S$eR?
zeeLYD)KK>Ln?6dbUNg^>Kdf%sFuYH3!S1FR7<1VjN>;wOXvIAT&KohwhR;l*mmpr*
z4Beo=2wEUSQG0}WFQX(WN;uMgY14S7?+jWuUzfb=ZeU7a0~|W*;Zaxz+m+{VO{*PW
z-#x>vo9Cgir<><+a6hb#6)og=>%PX0W@%2RHqN6jyvj5GxR2H(Y#yoJ;9S9c2l`&f
zNZNl4HHiz;w;f;MT=N`fUY^8fsXh4nIRi4yi5Ts(7SmM-VS&wT;+Fy@)Ows<dYR9L
zw|^IS_TCX()N^NITPER~^c<MwC16QTBra@>g~ypF+|YJ`Xyr%@zZ(JfSsVE7${vpk
z9XS7Y66N!`mpQr${|s)xbjM@dJMa;Avbu4tXDHYEenY^W9!T+-xOVbSNL}XIajh4e
zH?2b1;p5OqJ%Hu6cR*pqBG`paf?bd%oDa0I!SEZbKI5S6_o~yj`@~vo^M>}=?hqR<
zc+xaqu(qN?@LA`zpna7PV*0g_Z4`oiuBlke^XK)}d!gI82yeZLP?cT;hyAB;?)W<#
zZSI1Xc{B9iU4Y&66F5Gp46nm0uy%hX!msW{H^0lzGD^Uup)m-Yv;kHwE3s8=F?tGT
zLFCIi*7n|;P1iom#M+m!!(TMm?c<Gt2dBRYKIRW$BASDOB`R|T7sq%papxV(?z<l6
z>BmszWm5`>GA79>g~(|;f{27;SXfhrz`JEQxJD22REHw8RtC#Oe?sljd{~|I!kAh&
zh#c|2%$Y%0u^|B#eQR+de;sVzzG2myo7md=)2u{2n`xgD!J%2_m|>tU#6%5Hb>9*x
zd*<SBryHLKJutxW*44roXUz04^Sc=~B%4F6&K!aN^iaax7xM(+=p4BW#TB+ld}o9S
zm(-wcJrwzx3MhX5gYCWdmrXh=jq-ti>}tm+cCqy{b9EYE<6g*OZjcVXcZ^22j1`i|
z9j{G-F!4e#&uEB7=hWr+*y9PAv%b(2U4Xz1b!Zz1M>NOLem=wg^EiY%ZHF;_MInYJ
zAL4i9JV<#KATc8!%iixn^N<|mEXzgL-+bIze;h);k`Z4VjE2>ASQMoIS-)>A+vy2A
zz3DDn9CMysSeC(dUf9Kaw5piqvLfbEwU>P}-phWeO=sSn<5}mwXg25Z7FH&*gVoH8
zXFJPJF*V7PY|iK!_8))$j+OsrbJ={BY-GW<>r7#%@}{v_QKtn5r>n3D&KAt7PXd*E
zr+((K3jFodQIRi;zOk3sbLD#0G^CJC^(tm7`%2h_#o0_zV-(A2vS(3>8Eg-~r;U5$
ziL97ZzFUx{cRG@^^okhI9v7q04}|Dj+EQ$@nu6M|MS=qVtAeQtChX$V6xLEx$aXC)
zX7?whvFw}nEW0WLKJmFIc+BrFE;sQql54utxkh`H8RA^WpyV{e*7`B%yrhqDyZo?0
zPJrvTj4^Y;Sgcv)f$jIXKYf@T*XIl)&k65v*qi5sG+)F|{WFl1tj2}il{nQ}foE?{
z;YLd-K6sU2O#LAgwr4@K(+tjOi<y}GeYSgu9R9^+K;ia2c(31w9~Z;emU|KG;H+9E
z+uY2)36k+w?lvwfH)2~n|9%Zk#-N=GO#2wtt+K%U)xsRB%UNgZ5EO-^@Lgs!RN8{^
zTiOI2d7QAgI}3+rq+v;F2Oin0Q<2(eddTtf<aApq%$Y>-g$|Uf?@B5B3|wKs^Za;s
zO-+>#eUju{m*FbbJ$@#8Yh;82X9ZL<{U`3u6{0D>v}j4P7L~o%rFSQ$lbgr~j9kTO
zsAIOc9cY5ZIs6XbAdX!PHdw?p_=C?H*~i2drk3)Y#qNF2tnWy{cfT|`k1F6n*)Vk8
zvPR!77hIVbfLT4xuu*16lNg8N(&G_W{D?J7dBtpVcO#hXL!SOYh;GZrZ}CGI5nBjN
zV{sl2t4-FoNAfIn-kWt~EUo5xO5>^av}5Wd`c+{=Gm`~0A<CMzm2eH~VJ*rDR;22)
z^3+i#L2HJ!V`cJL7;QU*&IhG<y6+669zH-;?^~#>{|Ni-w=jdXVba!4=$F03#$L{i
z_159&jgxq%kdN>a76^zv35$#6SoW1`KdRfY@A@5BZM})er>!XQ_zz!B0DJC+U^d%`
zrdg3_c@>E~#YjkB+KB9zCRnXB9&64A!RSE(#_vso)8JNY-kF6{&BvjywhFT(^6;_Z
zEPhB^V%*yvwx&9ZogJl(%};gkyh;`OV}G+>SCtS>60lB^K&p!!&YYczDqUZe9y5*Y
zpD4~oddo1ZN@b3^y-Zr-4;#=>;jUi|Y#HYO`9gP$`R9w55=&vGyB^!+E6^Eq2{|>*
z=nDJ`<9R|9v0Ip$eI@DcFezH-KZ3>;asFefF1_x(gYIp0cqPT(!^}EpSf0c8<C3tx
z`i#}7SFybKcIGImMVsc0r&2kPwu>F@Zg8U-ad-L?F^!fVcBO$`E_Bk>nJR_tDPj@t
z80DIU1inYzZSffkOoL!pmx*;ZcjKGMay<Ip#DoXWv*wy2wtfFqHbp;=ovQ_#W$nx+
z?))y8EPh>3Fl{PRF4V>LMJrL2779`0b$BVW4hgr{B2>!-ZYM@!-YQjy#6~fZ0()Hj
zZjGV)#v!^32wQm=FVv61^jRI!L_fhWqZ>|fy$JUC3g4?e=-bo_Y5gDAKc*iGE^r-?
zV-*rK3LvuQ00M^Y#8^i^jNUg1#T~y{^wbON+QTbMq9@1J!03#vmVS$E&Ae_~CogS5
z-4$Db$n*t*zdNb~{sr=aoAvyAyjBMlm%MS$`zGJT=_00ZIE&6+A&~O_E4XkuoK0H#
z2z@)Ap<a&PF>hbM<v%5;FDS)`hvo3(+PDcjvM^7K^8s9sW8WPOpQ!b4-Ms>>84K~;
z+zz**^fCR47h}OmY-#9J7V~vb@bX53VAXGd;C_~>V6a?KFe%woAU4U1{TAEKOrF)V
zFWq<9kp;sr`oJ7K8sM1`FE?UHM>2Ho?Zwr`64-3l<K9*kbWT*jy}4i6{0d_n4fH}-
zt20Wz=wN@@WV8?Xp<8n;ntrZD$nfuM{)P&+LM5L)d7r|DW|gr%!8w-xY9#FK^uhRj
z{O5N?XpR#FB?}{M#Wz-czMtKb+{%oLY*_mpO?FKGD*Jax4@}trFSNrUwqhQ#%WR<S
zU<CPcHMEKj#kSKDSY^`9W*=x}#eBbT=J{XdHanB~HSc5D`MFG>D~)3xw7GtEG?X01
z;bE!|+<OCITN8{nM-8N^YvWd}Cv2MLA(-Q>@A5h<tqH@s^lbEpa!!Kd?a%v%@!yTZ
zcu>uKJU)4-*4mDxA9=RM#2C!_z5%_>i{ah71eJdaF=llNge6y^>CXfxhACi%|0}kF
zJz=UV+t@{ovn+f>B2$hoVHKNCvV$c@*yM{@?APKgEHu@b1v!mn!wy97--}~z_aoVu
z*-Gr?UwKxf{zEW5?SbIJhV6olOA-Vrjb{bV18xh9OK%B+lWz$YT`#n4J|HFto?y&K
zsfz6=m4e!F6_|^wqdG$pmH)-D(c^?z)05M-v!;a#T9ZZC-czGlcepOQ*(}cNc<p0u
ztUHEzd*ZX9FK!I_qEf;S_tTeSuDTya?pTZ#pJ`xSjwsVF63Dcj7yNh8h)rLZ#Hx=M
zvgtRASnJAU_EpM`NynSB{w8B~;g1+wH*67HOg7++NGx1G_~7?73(TrCz~jYZptw~Z
ztuyD~=5kwDbsOTU{%Fk5e#R6wDf0fLIJ}lW19R@lN_bL_h4<@`Iqo#1d#j+bs}hR)
zDscGuDIDBTin8e?{LXn45&pY*hlMVliY;VEx13`#849?myAFkc`=D@RyI}UwRKe3v
zek`!7Q*dg`GePX{QMM<=U$6z!C2{FC$EXFe(48xfu3ai{&QRt$x-7OhE{lz+c+2kn
zsA4unk*L)9$F4X#Livg#Se+~6xtFW#&sTw#raC2Dybt?d_tE2GOXKTw*-)K_0;9G1
z%tU<}>nongoIBjuwT&L^n64LFb$&KG#_MuEyqUrpE--A__yKlZ!ZbEogj%AtXuili
z^p^b*>>5mEtW+AGod;RmsX>-KO_<MiqS#v?i4GlMY`1J=o$@!?8IPY#cpld<Jdwne
zSZNqV%j4lKb#$&8kNr=^bKDz;f4iKa{&6%$Tpa^z&hNia+RY9~$zaJjU(_CrMcua?
z6g<d-hw2^(a;kA_BHvwAXj1{tq1pdz49O;qC6CYvl;>_s%zq+nd}>XzOAKg=uMSzC
z<$0$;-;m{b6*fw8G`3QL0u|d4u2hXpxA(*PX%tkYqfuO(4)L<>Q0V%AIVY||RILre
zZ#+iE^(RQJIf@c(2UK~QLOjP1Cw?1ao6AW|pIm~I_d1YO_8;C{yn{o(o3N%l1Cp9t
zyAdV8ax<WG{Zu{^O+}}I6}H~d!r7bG*s-H`m_(W-W*^dk(D2V}<i!bWhuIUsHmh(Z
zQ#})UgDc_2Uvs@t4q78lvs*RYtoZC&7O4J*f8J_v*{_P|lYJ~$QxSUOB~Z0a0#3tS
zSki?x>}Sth*3j(6P7I1LlihCwho+meJt3Mb$X|y&Y;k7S{jM|ntCKOj!kKG1`AqS8
z5&jwNXPSo}urC`IpkT;uyf2W(tYj60WX{3c&-3B9C?3McIL~wVC|eVl#^$&<vfBrJ
zn9~z&yfE3xy4?@68v%dVgYBQ#w7<LA60b&PkoSsJ*on}KI}+TN4YKBVE0{l>if_Bo
zn?_g4v~?kuUMIT7y$7pe<mf^!*D8z>rp(jbXo~X2JYygHDffl*T`zQeYGM*~^{lb1
zh-LevGL`dNnCzAbtZJ7PTUgpJP}zD>(5UOolyB)G+F&K-kMzXEyB&f@p^pXMI-d(9
z9&`yDg`NtIOzRNLQd!R~4AEgn8XvP~TH{cgA;8xfci{UGXg@iJ;VS2$n|vAbf_T7^
z_h&dKe8ocPUi@A49pfLq#4^d}c>ky!9&`C!aM^ZDuJy%^)&)quE`eudgKYJZ{cMzK
zroc5}wXJ<YrcKQQS6jauA8ge_h6>Ua+6ZJnGl5s7iJ)iv8QZ|Z34)sA34#d+(gfZ;
z2Lw|_ePan{71-_7akf#i{etDnVJykuAiHy|n5C$`Wxam+5SK2-U5|WRII;^yy{Yg@
zPR6wj>KJxc67Lmnvc9Ssmg(=u#u=pxzA8Pk*)YOh@W=Om?Oo|VmRlDt6e@`jQ9{WO
z4W8%Td#$BNC7LxE&S@e;MI9<iqoE896v<SmC{3ayg(gKAk`$34NfSk!21@VqPrTpG
z{Riw1*KhCjTWejzy$C00?m^?)sd!4M2X_1j2JIoYx$N~p5UM*AHs$F;kLp!!t=Yp1
z+Y9jW`$}w<^u#hJO{pet6vb5!q3)n&R64Mou4J7j--<XqlCl}^#ZAJKqdIVx?K9NW
z_rh1p-a=Tk4%HW%)4iN=<e_#K?Hs-%T6d$%x>VF#c>tRyT!M0IMN~hZB-xXe<G;}_
zaNf3|H0qudmCFCYtk+-hRsWZGw(&aJWrSh-UXesf&<a1~OVKe$a`vyWw5G%umndnV
z_?SR7I^!wlurd|ucH>8fJX~OW3Ok=YLYvLsF}RWAi=~G#J!%^U6~<vyMktOPyA$`W
zcE!g>-SOSMEm)i)(Eu+~q;g{oYLC&RO<M<3xczYICuosEkt0Pt*g-m<qbS@XiGuu-
zNNQOUJ^PhF*MG;8+jdVXlW?NZa~F{5=0)`DgCnVRub{F|%jrqd7~-kYq*&C4pZofe
zvHp2{aH<7wK|SXAR$<AEo9H$-5Q{y}<Go!M@p@nqN<8$K=5YXjhFfDy+h_0>ZNaYx
z_h4d<7iz@FW7CSESk}-5J7O|GYnUI1l>)&$Ef?MxltRy$5-@Wufu7G1e4m{R{ADhK
z?#DB+YP<sF>vZ6;idKB^=q)Z7wi{O&55t?nVIE-U1*6<P!C#@uXs@q<l2ssdtULnq
z&9|ay-cu~jdyQ3#zu?h4Yx0-O{?#K_Q1><mI(FwEmR9e=g)fdlcGWeAT(5^BcMsgz
zFcHU18;gc&`!IR=EJ$Ay06kLGV3jif4=Jo7pH0qGGjuiCN%Z1cPUGld-B@b#Ga&hy
zV`+{Fqo?xYY2H;GDy)>HQ@stidO|wN#)iY7oO+(@F#w+boecBd%t6zXbF|t;;<NJ0
zqLPJav`03b#(7?%dfW39BAlk6M`O|Rqb^KbG#52@UjviOIdIG4JpB0734a$Hf+hNv
zFz@$3a2*!SJA7C0{AwLOx+EX!Hpx=^?LnleCqqU*d-3Mq%Jg>oR}{RD;`-HL_$4j`
zTSsSNg+>|9>FvV_slTw~_-Jx;m`)Z~eev2%MReV95JsuCVaonGc=FFlD3Ga#RhrLW
z`Fa_I$=Vn<6HvW&7A}+UOKRtSz;D;>e6J<L#~RZyy6YWWDaqhh7N|jz=LPuFdm6eR
z46Nq0f~ia=6xRF%)j=Y}(tpu#YJXCak)ewH{b|kZc3kOo2c@n&!EpWqoAjlq?TakE
zYm+DI>qAI!h#EOm7?5t`WMccLP*1@O`q*MfBXdT>)sGip|AlBQe4$60=Nw6X<YvlJ
zkmQ%0JxVGv$EYFSg>I-^z{}``-Yf>S>I(SlYnk+|ZHoD4Nq+yp+0)F63&oP!J<fcS
zB&+7^r;+9fZA~=r^i3L^kx52tV@OWpA_X1IAPvPh$==F~v`_m}pu9g7k9ox7?u~`p
zd-F(Z<O5Q@BKhyv_S>}7@DvSdAexsto>C*mklYt7y1GOj=R|m8jl)$8khy~sFbPs@
zPUE#{BWSCME{Up#QM(S2oP=NMG#W`2cEzZvR*YRki!rnDGUi_>#0989+Ev%l-o+Ph
zz0H6D2YgU|;t32-h=gN(bHS%-I4+Amjk#%&7+vucmTdWj%?)a#8>2!-`cmN4mk#fo
zwn4+=p}4MiB5wOO3EwWWzyRqvc&mRqgk_uuw+a)~sQCuDkL%%Xur=NcD!~X-iEi_=
z9vy!uPvYb(yp<G&8l}%LwVxyX{u4$06Z)XDq6;a)o;>RRq7Y_J?tW9~%fVE9O<}11
zd=f_ZTA+fy34T#Jh=m7E;-stl&@m<ihkQGV7hCURT6z_3xcdc#VN$fQThn}egCui5
zN6&mi-W>ChQnu!K(`K6++bfzoc`BF>mE<6p94w|CS~tn%o-_UFoJLw3EXZf&LP?Lk
zm;_xJx}#l*icd<g>r^0We+WfAPh*_g@*KLq_3@K7UNB^r5*99g4Te26P&W4wtjny1
z=lyHp#Q4`B|DXvD_tb;O>)UYjX#&qL2?cA(zRs>f8=E^tRGB$nl1Z?dy38`^O4}_O
zH~tRYsXIygc4pFs9XIJfP6ySl=pyN*%{09F6^)2|L8s*B(uk$j6u8iWJ|CC#phdFe
z{k{j&+S1|6t=%vsY&BmU(kWCA_TXNvANl2_s!*yk39d@_2WIh+fAY!ULCej6P6a|o
zTr_CPABSmvwR}u_F#OEg!UfxM*ew%^7OshC>yeI9Lt?3@GmcIVO`y=}Nwnd66tzmS
zh>RPI>FYyPdR`<)Up*h=<cxf@YBAs!G?(-8s28kagqXXjz5=Hx2ejQe6~m5oKyOVD
zY}DV&r!6VtdmjAeiVgDcOFNUh=yh`yRtC$ZBT*LK;LsEA@ls$qR&>VUN*0gvzHY;{
zhjnq%8zo5B_rSD+Mp!mC2Qp0#qHlvc8oWIRw*_0wKX(p4+`Wd`%ggY{_=ot_*$&Nj
zOQWs9F>u==hR+2m@H{CAW5UXD*Y79zF0vZ)*)v>Mc?jp)cwo_8J1jkM6<?Nz;fIft
z@K@a{c(ybd@01@$d%q3nKimjEj316{#RwD!O-5t_D%4qEci}bsqMZaaonHW-sN#eg
zjtg(BM*k~+VfD>x@NDK`9C2+MdNi%Y_2KRqQm_SoC&|)g3AbwMHHc>1)SwN!Wl49j
zDrqn+YL<4S7Y;k9|AS!KH7$^SZVVvd-5z>ewub^Y?4}2sT*-auBARTrfNEDRBI5@R
z)Z{UOO4~2u-E|vLIYXKPy_;~v?h?$_K9AeRY2)UF)eyBO8gdq=g8#xvsPgY2)R`NJ
zuIr9qC?3U*qhU-tRfXAAy#%R0o6%_e9vs=}jy+FgallT=y}B9$A@gEj);NSjm6h=7
zcLwyTmcowXC1B%H0{W^4`Os8pI1?Ww?o4hG9&LKe*ZADW<;z=9NLq(_O^b2UcLjVk
z-;@Vud%&B3Pq0@(8SARmQRk~ZEE2`Q`jiFubdNW>pL&hMOgnI5pCZK{8BC8?NW4tz
z+ECiN1Y@de!Q$n1u&Hzg#e3Coc!US~R8Pb<N68!~(MZ1>WeF{*`(XWvQt0+rPV3G%
zQeToI6?_{)FmV968y4c}oU!z+ONX}m=@MHynG}2VX?ma*>B`EIoozi-OuvVVe8M5#
z_dajE^Mwb5$AQ>tGOD{pvnSmS5Y%)Ay(Rs4>(_JCiWlg-T{&*P7=^i(Zg}~nA`D(J
z1YL{o@jWJ1FsJP-%vpUI^7?YvqKaEQOYS8vUK+zyPX5iWEoowphULQztz#g5lLR_?
zswidCh<T($I}IM9)R_QW`Hvw!pSTmZ&A*KHYGwE^st;#vxrhUrlkkmmEPg-ZgR8qU
zc>K^%sIz<lehC%uc0~d>nqTDGeq@5!q#X_%=><~&eAj7-C07_O`u2vO3awxr^;2+F
z%Ro#vDuDQJTY2>{4_N0M0C!BnV9>!(@NQ{=&D|YP@Z=}Rasym;=><Biv4XmMC-4vH
zM_LIVG5*UzKEvb^8@})<WVpSARH-4L6lso4EE|1FM^N1}19Bc<O1AqfDadUG_3dYP
zA!7j8d!1sl%8dmpQ8X|AV2Pf?w@~=?15}?V*~!0f2R+;Q_|9)3s)l+<-pV@2$jag#
zzYOU|sWEwXh$(s6RC@4J&D<<|nEAjr19{q1CqC)sP@X$Bjyw5hGP~0OEXJ86bB`Tm
z75Y=2@nG`p_J_nvNjy6I8jm}p33U!PslOzTFSy|XeU$K3`TLUSgqAkR)G~%ybMez3
zX-a%5jRjR7z<bnDwCc>p*;R><FX71i($wgCmJTWJeu6ik1mio4u{-BJO7-5yx8n+N
zal-@LJ>wkaZ{LOLzKH3x0C<WL-Wz=iN=iJDCmqAdx<7e-qYX5O&2aSRY3S%>ix$ce
zkP_O9)Bg<T`iW}XBR_<v8{J^X*ZP6QPia)XX^fezlkkyyHOy_x=T4$5h!0JK-T)Jn
zySWF_ci8jxzZs5cD#Maf3bZ11Fs1kV1wW<3;q$~zxaMXk9(H>sQTyG7YdiX&AInFz
zp%(b#Lq2pDo`b};nY=}^JF;+i4>{jeu(fC!S}dG^ciR4efCEyvU_Qs&<^g!aCK6Xg
z72vn`RXD${3*8H(D0q&v`Kvm0^GYp={zzHXT;4~;-0JIVx?2)NisB&hxVW1VE#2w$
zCQi@$n~>S9@$_=mL~3e$gmRr1@K417EYzz+pW+e>TN8*T^`Us9bSwsJF9Gk8KE6*n
zk{4Ti;(Kq`f>d`6sI1L^rAp0gklq{d=oJ~FZw(&oY4v6{v-N~<(K3wZ4GIMBD`t4T
zKpW@#WrL|tDhmpCrcWo`=)w5wWc#In77s`y?TwkVXXHh?s+dgX&2oHwf+KItu4jH>
z1EKEKA^6f_Mt5HV6=gb5p_c&-bsL3gDG8t_Vq7*=iBD|5D(I+>WRge9t=3raVIz$}
zJ-9zKK77wVygbRjv}!@o$taOX<1@cI*bj_cr1{kQPH<tv8g3L+!tBkW1fyY}giQ%G
z?AoWTxMI*M^wWuz=#`F>be%d?2Mm^k*~-(^3CeWBPL337$Ks;{Ng&nlGZVQi<Ne-U
zW!4e;{9wluI5c=E>iSN_r&hJ_&B+*kr_W<fgUf|5tsFthdnIdJae_w-@8UT@Q6N|O
z51(?|g@5y$$ZNWu@UWw?sL&gWPZw`RhcZ10SE0;TzgW))4D64e9$$xF`-h|Rq*C~|
z&!10kn}PHC&f<&0t7vUeh82zvv17`=u+&}}RccSc$={pU>wd!lHt!X`Gl>v(eR_f;
zN-A;V%dzZYb}Eir?S?lN$MVlT`e0hG4+(Ax*u6Iet{s03--;BWrRWi#nq!FiW8`o^
zumP`GG#Pgf5pZG!;{M>PXocSVpz1*W(d8*zUnn3OWq=<8?}N{{32<$JId9CpBP6ye
z!w-8EuyX#*6;??>OUrvMmnBD$2Q>&g22fX?Iw^h^(UK=i=}zW$a=05r<-Y=`<<DNa
z{&o+Qmitp-?hZ0lcO?&QLnr6Wq2g+58vkz_YM;J^wLdOnrIr;|of2XEjj4EB%@Ehv
z+=J;G?I0(3EzFo74L%2AVXUSJ?&>~-fpa49@`p(LUK@gARTY`_Q(4w?X%lQ4v<0Wu
z_+yryE9Rb0f#)**5bhQYB}qr&pOpY-`~cj2BsHs~kHwDdX7(l}5Lh0<@2)Q9>bbkb
zzUM21DRrlWh3e;pL$0wHP~Qh{QWRi}{9j^K`&uDX&K4qXr@@OxC9G_yobuehlvRz)
zpHi}MF^+nr!Ixzevk^be39|Nw%ogq*C6fR3LS(tARGfM@U2snE7awu*hR1fZ;hnh`
z&u*N<qgIV$ChJCst)KlAKfPWlJbvsdO1^R*>{Kp5d{{QuZm{Ie_Tz=7d~YH2z#WX2
zRiSyln&c`PLq_2;&>DOO(%}|_ZZyTw=j&Orbq*W)DT}Wx4Pe^W6PaC|5!-jdoQY=!
z2~Dq-K;hvtXmtDm3IlKBisn4L=$nh#DyxKH_l&vkr(437L;2w49>_u+tA*fey=Jo)
zsR$R_R|}2JGX+S#!h;>+_!fObKF=>ud|;ju)7$@3Wc6Scn^Le+NPPMpM=7;q=GQzt
zJa7qmE>*-sQJ%OZBNdMiF2nNhK3qQRBF^8DjE9CDM@wxVtlw~stGrTRuVQD48hhrm
z2+JPc@MtYh?m7+^jN3tCNQc6T0oW&{fX%WBDB}Y#e7`@dSTGKm)DI9V=RjVL9?Sba
zeTx5qShHXL!Qg5T4EE<{h%!d|a3%QxSlr(L+t1Z;qaT)FJKGU9A6~|EZ%r4%T*a(4
zFOA9N&g9<lTKuz-h|iKHv>ts8tN-z1`<>sx)csl*U^^4vIz=Hjna5Ha(uI<u<$~Aj
zCQ<f2Ulvz*9ki$1=3ld<d63H=erBpW1pg|AbqVwEtEDTN)oO$1kH6TKBstN)zKRsw
zE~d~Xfs74R>D(e|ylwN8X*MGF_mt<ur!(dnyiqth(V5kmHL%*huCn`cWa0E!hL$7z
zVU?k)AQiutz3s^7laZ6c$WnCL=*$+Md?{+{ywCpKn8K}|zZ1qpg^SfXoLJL<Qn)wJ
z1=JdwA>?^2EW2XK?~h2quHJAyMtp>Oe3*;!+DpU{VJ70<$&O5Ad<L6QF%Z&5P2#s+
zCrdOnyHME&v2WO1*dsLX#^uBYYfJ&DTn6W(t$AO+-%Pts%$)9T6K>jgiX34K+m;kB
zoXmR5cDXC_cZZz$$yv!Px7!m==SgFBM1y2@tAxP|RCvJm{#^Co6(})Nfy9&Q%&@0g
zyhe2<pB<r&?K+cT-F<t$Wo`_6wqS?wU3xRPzk3K8kBdNG?-?AC)X}WNlbP!F!{SAj
z(q=7FFT=8}S>SvrlI1#NgP#6-2-(yPhpeYz?B`DQp{ECIQ3}6JtrLv4Rl?WYgLuov
zoy+{tz`wtJVOvY3s99zZY0Z<Nm3ep3duSL=UZ{&fMZH|_$pPWU?od|tuN2F3Y&07l
zUSsCId?flxd>h-VW4LV08}V-C>ug4%gwxU$c(MFgNa*J)xi9n4uVys<h+76)^+uxi
ztJJtna*<#m_F^-ZA7TY}hQJJ$S8U??WYO*PSkXVHT$s~4AGWjVx_Dp6Iv!T_Tb#1?
zB0Ktc08hzyD^41HoOwirVXaa%zH~TCsh=etz=;X8)b}#|b>kkmXbls;-7mvciu$pt
z*Bix`Z1!@mzjr}fWv)oRUo1bH*2l{v-Yi2Lh5Yw4?p>t8vra^@mk#lQ)w?v2!HflF
z0Sj8##v`HZ@=XI#$nj==Au5=Mcn%Uic590YilaphC(Fen$Az-QBxzpos*d?u&u7tc
zapKvI%|ctF6-%CffwgN+0cd9Q`pq^pS@l3z75ma`Z*G?8Rr(n-)gy<6ZJBpiL2Lpa
zJGq+WrHtT><wux%M!n!NRb3RHJe-f)IRv_-Y=In=U|Fs)+ZtFVcwNpER&8$<WR;b<
z{iD;Y-LY5Pf2tA7KH<Vm_m1Xqvm)35-&pLkHO0aZSIpU82DbG*Ox0;VpY?Swk4S0Y
zJIlXAO=Jdaj@DtH&tDMh+L&;IzpvmUtBuSjbGCS8LkGm%+W~{z<XD!|7olfSh**3h
zO{A`{P3SzjjwO~Run4()VeLarUR1e*N1l%1(?^|RhGR<DP#rm(yF?epshB`azg@!J
zsjlMAo<X?m*Cz<Nc9sp)iWJw!P7)gwtQ8wFMfRgcm5mq_FTSr_453B>TKMRrU|s=<
zw*47dOyie(wg?}F921_eR)kMlU3|^-J>rtW&&<SaGV}_AnWm@(k8RYT+<RiO^I1xC
zYpcZn8$|nl2GHUbNqw1}Ax@L8!Pizb`2A5fLfT95ma`JRX!k4{FOsLiOo>$G;BwZ{
zrN)Bai1B^)2eH$uxGAPG`-OeRi{PSBAb-DM2oJre3d<UvL-wNt*!uV_JGOnPuqrJ>
z-0~n(Ts8Qz`2PMLp=x@jsOzweIKeeeoGs<dpFH=0!`e|`y=0P5_FIv0WGwtygm^$>
jJIfBgBUC@x&-(rM{fEGR2>ge@e+c}Cz<&t*-x2s9^aFGg

literal 0
HcmV?d00001

diff --git a/tests/testing_data/ultrasound_confidence_map/neck_input.png b/tests/testing_data/ultrasound_confidence_map/neck_input.png
new file mode 100644
index 0000000000000000000000000000000000000000..74a64a9d90402a11b82073ec824a828a854d8876
GIT binary patch
literal 100996
zcmV(>K-j;DP)<h;3K|Lk000e1NJLTq00KAw00LDA00000_H@v$004jhNkl<Zc-qYU
z-I6RxuBC~6094J~W9LtIPd8_-C_=slAAm2<2jB%oNEw-t7WVX+?#j#^?q;e0cmXvJ
z-&t984tYfvsmjcVaCb9PC=^&&SdjngzyBZ62F#>?bYSb73e3lm5t9L#psh*z7-M4h
zz0vf37d%HIy+Izy)__DBOqwbH5$UuJ<a3@mC%@i0V?Yzeq;)1{U`+tn7%>5wh%~Do
z<}(3dBOie_n8$?M1{pyUApz;FGso$u2#J8~fKM6>5M_GMAO;H7j7S7v7t(1;014Cw
z5}BxO8lQ0g_Mf)#U;m49+i!dh5>E#rdCu=N`01t)kEH#_|9FGmRp_!AN_fywVvHv_
zM{?rMe6l0(napNHItB>IK`_z*n2~`pN$Pr=-h>3Xcm8nL8W8|}--YkT;OAeq>pqmv
z|NYZ3GsZkUaGwC41ElYg$C2n}_!-F7$b)|Czwn%nH0{PSfid#=c>dqoum9t}YtqJe
zwAr@$5>)uduSOSWdcXha7ys~0a@+46ci+zEqw#HT4Hy6-`Srp7>|6iIOEl`A`EUAj
z-5Bv#zu5oH-`MXD|3|<3_y-SdeYySrc8~vm_xe9K9E3h4@XQm)WBda1aXgP1M*uir
z+BR}bYvc*-lv_rgGq8ji2<HJLA^_6ca%AoO8$EtUIzC~7AD#r5Z7uG^2z21*I1@mm
zn;?*W%*kzQP08WF2*<XVyD}YDp#wRXkKw@d5r7~KOu$FLe!K4gL3pGib0`p~HMitB
zM+`R6C^s|{xttH4w4eS%cINT>_?z8@d1eF(Bbe#Q+(-Xp<jjHm@&CT7>HhKUThc=v
zz>VRL`T55<o)elHbBuHv%$bmq439*}39S=;4Cv$mcH4|e;uFv=j6|B1Ju2bOwn0o3
z)A(EY={af4KYmYcN+t;NaSQ-vPI_x<po}AkWI|_<$aWW`TO0a_=kW-H^O@=Ywf#-G
zf83p%!#^_}THF3N($8m-i6RE%5vpA<brF|Gf*5!PqOCFM26NyECP-@tkUr?`m)$tr
z!{d24|NFg@Gixm-GeK0O-<V^-1|vP|fJ9V~COPm#Oj-kJoq5gyo5@CwYgbR2a6Zeo
zCz(!pGDAECtnbZWF#2=Q!H~Mf`(_aSm_X}VD?`qic!DUQvT^$_A3MkA=Rh$SwJYX2
zJrbceWkybtely{A`(?ZBx4k!koRMRGevaqy`Gh7I%#mbd0CP|eEfXS=ZG+Rvy1Qh%
zt*4qGot&8trZ;IK-8dY2*|;{`9oxX&Gv_>!I)gXE&k0TL0C{{S@9r>>JCB({|6A5`
zhn!=8KzPhU{xAQB+x^#{Wt<N`$LRw#Q_9+<VG0O&OwfjOlhzO+=Ny>{(&dIHG=-e$
z(7K7x;l%yG*kH`WIzF>3as9uN^L4qNBSafim=x=*@4bO1c}_RI_j-1Zj0}e^2KD?J
z>Ge5aLPn0z1{1`Zc4ACVgbWAM1|Tp7X1$&m^xM{mquk>e;eqtt1r@r`;4wZ&UBC1$
zM1n}!Mas*;I@X!odPn>Hr=PYv^Laje%sGkk_p|=249;UDotboPFwdDdz-uuD5~mr!
zbW9B<zzAeBH8sIcLqH9I%!&Rakbb6zgSj1`B8HMklMd%38J;PIY~^8_5OdN+BOm8@
zJ2IMP9Pw<BpPU~&#&{0*AR?OJXBcf>4`DcoI~*T6d!sYPoPz|JHpv)H?_#)vO_SD6
zNRML>jI$o3LP*cR@dGsuF;GmKDqdcQWD<b8KumgXkU>mv9&m$aI5Fo;kbYw%0;BnL
ztmM2*(#a9^9eB+1tk5@Z?KX)6?F^5o^Gs0vz#9{je4dj^4TcRP8$<f!e2jC>BoWzB
znt)ge;RH@j=)J={9Y*WY?wP*-&hdDhBVguyI@2>JV!|?#slc&M7;{C$%oGHS1Wy=;
zV=CY?d?uzalWBNG#<$K4l!0w((tdB{+DA@g4qtDtZ4(>dj0xcxPJ0^Ac7O??_#fv8
zX4317HxYbf>l=K2emDF%M+T0uuGxf%&miMO6b#~bnlyT2KEemkT~-(rr${|#>0OLD
zB9f3+Z=KAfTTX;zK)gkmGt*5$rCm|BCuGj>10@g<MH>o}F&;pU9FDZ786ZtChr}lM
z3{+$%=a|XNU_cp>0i2#ng3pYBVy;T!{@cgQnKb-7f?W+aadmvebeiGEGZT=^M}=a|
zbe6xG;UiN&ppnOOpuOKe@bup~=P{l+M%{NX<ER`IOeJO3o$pG_lk#C>(#i=0p~@r$
zv@^n|A00F*#$@^c%?S~p>JR!xn~WebXE=l>x~t$!;zS#g-ve6za{X{*<9Uw6nK=ib
zaexwublgANPht*F(g_U>AQaT<l$tru^6Q~>oPdT5fpD*IlpX$*r%WXC6;z_868o_t
zxC}H*BbJ3J@pjEAOhHzBm4E(ZBXTevV{m2aCuqYo6*C8X`fygzIAdmvfc){3#M~>e
z&4Hzaf<k8eA*4uyGe$pl40lo*9_LYa<@)tO4mO623Uz9Mc_1*);pxMU&uzOkAw9>;
zXLu%Ro|SqG5{a%PnG?ygQqD|id`^$i$P9&`NkR@3@DT??%?z(_|M@iSM+kt{y^Y=t
zPaOB~L=y7Ym2^Bj23Edy&ID7ebq_e_e10B@K!yrfKgfXtC*k8g<56r>NTX(L6LN;H
zB~?*Tte%WPl2ksjEX_oAkwPDGu6?APLu)^wu@AZmnT{j28Hns=cOEm!Z<Coq-Jj5!
zhcXq!Xr^K!G==#%AE}sVk4$$gg~$zP%#6yYm=NSK++RkFnQls8WK@upqs%n`MZhsq
zg1~XZFlP*U1#0VK3iJWavi6)`k5W~3PV+N7of+fVdv5|5K4vl<Bg5k~O-7*Rz11gX
zETc4Sl>1*_8l%q@#xV$w1G<KrMy7;69S9j+fHZ3i7-1yxdH|B79G#IsM&v+tgE-L5
zyQ-8;`k3RK+7$?uuCgD&Km@`k?WrF?uk$rqW`Lr$bs;5`x#3YRXkjL*J6>QKHlR0Q
z@0&t^bs?V>^jADLwXQ~U2LTb7sDkuv+HZa0WAlv1%!z?ZbHk)};%O_rS2exd+2edb
zh6%d-Sp7n1LO)RrQy^yY2ImO~Te}GpW-R;%^9oS|Mbym!RSjhu+EhqpUcgi9f;+u;
z#>^<+v+##)j09^K0uSQIR3Ks=*1$ySe4fGd?q}lO5BIXUI%-uBHWRT8l4GAM<5^eh
zBSUM0A}|{sNQMvShC-NtZQA!^ID}-niRx4;HA}eyb5$?@S;ts&4LrwtFkHJyMYw$h
z$w_;BJbFdyjahyoKzkQbdeUZ{m@rXPo6z*ee4H~yrQN{{`mToOr1g!nUs(<(19}4l
z2?q>XxiM{<ZLRgP40F35=gToPHErN`#NMz)PvCJ*5b4mEZMFt9GrH#NB-&^|dYOi*
z4PV}*o8CLwXV~FwT4O{-i<h&|UW*ch(xo@$mEMcA-bC7DSM$!$ak+zTcHO8I1`~dP
zq8bpRbFAP$W2gab+OkJbF%hAQSOX`dAx8VinI7PIz&j+C0V-}Y#WaAPY5{76hG^e)
zw=pSHLWG`#w@H@GDZmct3Ruvscih*J-hmBJ0``@~8=&E^j;eS?5+-_CKa$czHxLy(
zlfId-8Peuxj}f2^)sEBDprP1iY$wu}O$$Po<$EiAG#Z%gx7JO-Imd~MVbJ-gB)PYD
zM?u1)-a8p8fOWJ%!`Kw-0tcCAtJ(t%w;&=Fgg&5~h-z<ZF+<pZt&ooa+SRnztv8q>
zCZ-+cwvjO=^e%`ZgkX;NTxUQx!(i*R@qD(qEGFyAjQIl)im+)uavE5t$!^f3brZ=c
zQ*y;oa%Y^e4sRiiiV?=;=|JlgH8RQ!rM0d~XhSnSaZ*fj3N%I{rkhE{6nQ?8ou9NJ
zwLtXVn4f?um=ZHdL5zW#?Q#Cre%TuG3iV9;&45I#(wZ4k-g>|<-)?IW{b>DB9z~jp
zfmIe@0(sKlE=&^H2924nAR;LOW*un-oXDJy&lB1`8{0<jKUu}#G3F0d1WDwE<8AJ3
zAibHw=*j-gV4hPWQj!RN?M>B*f}~a=2~(^v-10wUfx`vtXloK9XuTD3OWTZRM2xA5
zDJ$j7bTew}CiJaPo%>zLhX>H6hWSh!<*JI*qZPYjHxthLqtOuU+OAKoSImi+nlyP8
zRWzWr2~8X4Ggpa?-euSGF-g&Vn2I#(3ZtWYl4DFn@J6MbozT8cS7h&H#YP-+NUU=c
z`F1?mnlvG%jsQUv!c^==`*#1e!1b8ioA&l`XW_Rwo;kbN-8e@!wx55#HD)@*7XNN*
zo$=ob>+ISXXXTc|C41|gM;%G!K`enrO1l8Mj5<r>af0Fk8?D`b7RdDJ<_`f_MZZ3}
z33f-A4@Vn9&OW4nv-@NHVJ;Y%iPXRDReH-fXh6iMk}HzkRGOd#5uv>`1>KQb@7wZx
zjPE=}WeX2qd9fQV-kSA~%F7|0Cii>q_T-Tn&}8tHX}K}#!KbLUeK(VApIQNxlu1pS
z1r?0PNRigMu5`L8agC;T<hE^`k9iUSiLrM@#ux~@(7;FMWNsahZHr@M;h<`rn1u`h
zup1wJ|1j{=&+{?HOprw7;}<g~478ckH_p#KR7DIL|HHRNK?U;h(WH0b{Cv_lPw)+s
zF6gaWyM63;g{U%;+THNq{`+GAoPHW?o`;ypb~Azpr-#!>bCIO!UeSZ6fEW4z6SVgx
zDYSO`Np1%C9G)8)XIvp!Z>{6pH_r1>){%G<F<>6Uud*@5LRCz=X;Td5;WIC+qT!v1
zHf;o2)~&5KGSK~Y>%HC9qx}wkN)sT*laQ;722y~t7=F${hi>%AIR@IK52?uN3$9Q!
z)MVf7Ct;%d5vZB~lOWsE%47ccQUdAF#YFb)<E}A&JAF(d(Jo2{Yyk=m)5)o$tEwrw
z6JGTdP&%Yfsc*Gidb7@y4TaV+3hnCr{Gs1JJ=<)dwEG92_wPs6@!dwZ{sD5o_x*mi
zU4YNe$8orc(WMdWvbFoRwflB|9jn^0#`*1+Kh6;Yh+EQy-{i~X=HZW!HVqnyw!1(+
zvwKXzE|^!eTNg3uaJx%0n$Y%u7(VCu2y6!&G3jkL#&b__A5%`a&-QqPuUdPfHDSz|
z+!gC;Thl;s&oPRf(Hb*0P}z(~8VH^HT{!39zU?MAs}z&gw?`kn35}t?Zm)^d>0A}G
zi~J$?E85MmB@-p+b<#^~(u4sGy-S0<LEzLLuIiX44#ho?Z^j*-3GPQ!(ap4+;T^XT
ztlFTpEhb6&409kSF;#$2<n|Oq3BG`dbTILg`=fiu2aTL`I)edpfh|9Mf|%St?t9y|
zPq0n-qdh<G8_2EqkK65|N^w7p=ifdbr{+&cXVB#1<G$%`z4zXjIaf&$Z6Ew$&kjnW
zyYcJa#%~7@%+JRaod_VN_pK9jD9Q{Ir5OWhBCYrBroA=w+UnLg)GK`L!e*83@4E&w
z!6MHbb0j`VPia#lp|$OUF(zV6uvHYZU5QZ@X+wcA2J~I9-4L+buAF0xDoslz4gYd`
zoY0%Jy_4JNIbdr|4bL3-z^dK|&SwfEHvk{drWqpANGAPeR#-*GXvRb;p%q{S7Xs|P
z-}ZuRw+8w%Os1VPK*@17V7jMl%uFiax3)DIInFr?VVfAzB}ozPiNFz6Wd8~rHBaC8
z<v2#rd7Si4%NCx4oBQ!OW`*1IjT_eZxNq>En?HW%gT;6H`n+GaZ!;a!I<lmfll${4
z;+PgMW|4q?`+kq{dln_ZK3>qT>8<r9u-_Zc#|6LS{M2FEw%fOlpY{%q6ORhmh`9=4
z{*{pgFussx={e0uc`wPFBHKo5_eq6NZf++wyY_pBX7{b{0y!Rk`|U)3y0zQ3L00TD
zbDocJPVnyxKhys>Xo=bO%Wj;{hX-hveP_<|`^@Bdt6NE#^5_g?+sVg3+TO~Ei0MYR
zt&g!WGM{=^+B8Nzw=Cz7pc~k}-+uNZKj)k#6@Rz8PJ|J}V+eI5wK_B#K+1B3TY`_N
zCW?X^^c!<9=_6LA+6#Bj^c*8U8BY|PkCeuo(_?ZJr%e!A|7fSV^pBr@`9*lrj?WI8
z9MVl9pC3EnOwfI=XvQlp%h4WSgTV;*3d@cdImhQ%5%5XV7~}Ezyl592V`@s-O@XuA
z_gik?P96ih<9{#{xqbU~+wK_L-6ta_2;-bll$h_(6`Z~gcpQ_h;klxLF(aKKDAaWT
z;O>A;F~J^pfM)k^A0NAraR~g5iQTr3JBQa9)7b96KF0Seu}uy5);4MV=WlTSaUbA^
zTt{}x;aM&wrzc{YiQR?Idrs5)&+z-{Z`omH%GzjsjykG9`a0qE_Q4?g<rK!rcB?E|
zw{8!PI0L!9KWUwImwCcugmnD?KOGq=^eDr;ZB%_IP<=bDAYw8meh#FW$fs~J&p9)W
z@7`{ks$kp49RuCApT6BK;6JuM{s8*DKgs+U-9I4)DgAOUG5+@bnQX1^KUuSiY3G@9
zj9EqLV<1~onrS1Se>@&zRSh<>GlhP?GbR&M?&Y}5K{uiOv?rjiC*?86oZ}A?X%imV
zJ0$*C@R{dAY7*dp;!t>al5T*X7@y<FRK0>SX_lW`oA<YhRVwN2szZ;*fqdKxfcn<)
zIFDz8ZQnP*bQeyyu956H^J#+g1CsgKkSnu0tGYm7a=!ze2N8%co)CJ~kvg4xR&_+O
zjXbiI#h>sq%;WJGi9G&bBHM!8;`~$fPVSq~7%S($e<MbF&~_H7D606L^MtCi2V|kW
z9+S+o@Tf7Ho&7g=r?2wL&`r1d?caa92{Ch3UZ3axqwxoa+4k|vuN=S8yXqpYLe9j9
z^LdQt30wPhYj{q=kMo%xv)C{bY4<LDDB)u~aKbz7m(G}jCbzBC9-F8=3Y|WlpLE9U
zh1t3=`Uc0GuOwu`BN*}{bVg1uj+XtIzKUvV7-L+UN_*?OQOu(Alt=&D$H%71<ox50
zXYl+@@AnPnb09rac#(QH8p((5-Pkz-;#Ku}1thXDn}GR8bNqmGd*F#=&*vR_?#~~W
zTG@Eq5s>?B4nT_0Hs{fQCqTc8;Kw;I?PCOJN33xilfsxyj7d}Dv;}y{#>cnD)DBHM
z$Kw&6s4-^M<ZUBofL;}PC;U~BWY{VO4_S1bX50Pa*KfDBl7f)u`T5%)|E2kNXr|k5
z|FW5wh*Zf5tzyVlXXzM^r2p3qaz>EnGe{7TW(vK1-&$`h3Y$96e(4&D^jXG{0JEj@
z`E0`X@5!9^Cxjy!$LP1_!w_k+O@3i`OdbQVcHb9~DRYd+^DSH5erhn!j2!cV!`6a!
zB|&d}do#S`%<-F$b5h6K81s*9F&Eg*-zdKQ*6hzD<a~_J=iygqY0|l4pRGX?RTz_e
zj<}AI2IRTHCbpd9RYgsVF&77@3x;!u_&P5aN=%XxI*3^r#)ZwPRb_&vNEE1Ej&tK|
zM6%!Z)_C;(NMLJNhHv#Z5r8QzJ;H~ly9Zc*yVpQO?!5`y?PlXNVIJpmV)*m;{QTZV
z|1Mqi@!ixGad%ZKzX-dEKjUz$!^i-S1>Pvo5Hpi*o08aWMcnZRd+}GjjggD~IEwVR
zDG$fC6KNnVO=579iFglYJ1ekyjsgAJ50b*R_4^`dp}qgqg!D7Um}3;{4Uyu|j`4(B
z?^>H7irUFhX(2%Ci=@MPXeUTFU_Of)i?z>QxbWkA#z7}P3vw0c^Ukk}ITXk29X=j~
zr;^v#G_q3dzE0IL4bM^bVA9K05@kS8NKOX2S-OJf5Jhb^pl`M{U{db!n6O<Z{Ry9d
z-uBisE{EKk9!wmYXT}JmX@TH#%m?9wn$w%r$|nJG@BJppe8TbHnCD}T2*^o}M=M53
z`Q2TcRQbV|QCw{rKBm9&AOfuzC`J<KrYg<UtfTsI--;jbe3GYElsqSubj2KV(W?_0
z(z@Dq>j4Dx69sZI=kXZl7?akM$Gs?A32fVDuQJVwCvxVjO5_;lpn+jl?HDR5T%i9O
zu%9r7v{jzD-<czeo8M)DF%IrL7bj7Fa6w9kcHf)wIl_HtI|2L2iqRVnd68DWf>9Hi
zfto*NJipJ$zV)qj#&;KU-OeAtKc{umaj{qKTc=6;_~k|($2h*{=}EyC<yN;vXDJ@V
zz`9>m^>h$2Mnaru4+Z>LO&@`2v%PuY?Gx@-hVl6?nUnA_J*di9JlhA~GM;s1f1s#z
zikpSj#iWljVz@&$3}v>k0U4t{Y4%;mYy&3Ex(XSn5l#^miyu>%1y0qH64Z1vvA%Z$
zZ6B?JfsZ+ce?FfN2i^2<+Dw~tgKcZ~jlX=^wM03AxvgAOLH1pwsgaMz18D3VlgFg7
zVe(ae+y$h7{y|0XZ?EGjx^^)S9rU;0E8-lJ_J)y^y__d=U@G=i`Yq@1{QmrYaQnD>
ztH59Hrg+X`Zl7abJwv^A@z39W-g~=m-@bK5oS*-h?*a67|M=FskjKD`9OPs^o5kv6
z7$g07(u6((5n|#YU0<j?YJKlt_p>Vv#%Fj$&N_yUq8P*h{+}Da!ni$wz<opzO|3r-
zhi*F@Io(IhhK_3TkQ59LA6}l-U2Pje#^G06)~>|i0Zr(N62ySsI*f>s5e^-d-g?Ib
zT{L@-NPOf#ki9m4v4*YNzV!_*uDf)Trv2-VW5h9@&>PJZjHz$%lioP4Vpe&Vb8#rf
zZ_u>;DGay8??=ktPJ*c5rRsSMnB((vq@PYg`}k=LavsmeSxK~j=P+$zoW1eP(-Gq_
zp5KoHh)}v{?<RD?q%r5==QHL+180ggY})|4Z>ytC?bnBrZf*Pc_U-3Rjy5N>(ft?#
zbn?0i23TxGCc|M&$C|~gCu#J(Hzx5!8oiUue2%9N08?Z5%s`148ULjf9r$)Q`sKyl
zH;5chOo=z-&o4UiY|L@}<~Uk!Ln+#;sR8;Fe2js1Gq1mkaGur#A08gWnIxjSZbdvp
zZH4A_7%1{dAvl$>D^^_?6)aYLSej|?s~u-1Gtqn7Oj-vgBF?{|-EO_w0v-i=TyEj>
zSf`v|VqIHrUCnH};*sSb_<TH0K&n%9g)1|Y84fcNkgea8bYhM%!L)U;+6^7CG4gqy
zPV2&N<9G&#fZj*feixIz5p=<i2RysHxjc1v5GcvuWcuH}KN5Uy{bR4wYEp|ZCOiY?
z%n{Da9H{ATRFlW^7z6sNA;#!?--v4MV!M-bJjQt(en3~WG2=NGco6=3`?z(&$37y?
zofsgTe+LrRPE3D!w#5?GPI$nK1E?~U>Y}?xj6{0S@s(glI6LL_fXHN^_1+3aAZZOP
zc2xqRu0){LQFtLU!3NWbLD#+a3W2Ti<i9o3-ZsOj2%-1A-SKlC&nm6fE#??wwPhB|
z%y!iJ*67-LVV^T0J?D8I8NmRxbrp`ss@sknfb_m7Ie7S_^}d;ynzoouSNwR+!LBfW
zf6l@F6mE|@{oZc7Zc2GV+6O#aBPSz}*+L(Wr1g8D(Qx|b_YugJ@b4InnCDq1)9)kZ
zm@~7IYp>oIeqVPTJ|>u(8hAieO_*b1(zhGr`8b}B?-33IMVaIIoH5v#4(-=(KS2&6
z;`A8L7|_DOj14vRYpsIkqMt;#gA~wi8dpngF5{TPyt*}1scbPLFY?MKu?zyk-_fiD
zkA#?8J^L8YRYMsbbW>Aw1;!aXV2(2)X#Li^66I7KXmqoiHX^e4McrDpJY-eYNFN!)
z$6CiKlZy$~Q8V7pIMNwO;NgjCFH4Kn*OC#Vn(*kA)+%uZ=uKts)=jKxbD8HDXbL`_
z^Mqr6!0qU?+itfeV}EdZW{!r83F~_u7#`?$J%Z$L$M8x1Z8L%KjETW<JpFJFe0kQ3
zs2}4CrP5vt+rX@E9Q!z*iR=U0;_>_Qd3=6)W$yu;^9&DK5zN}RU;mTgn2?WoK0Y7k
z*Vobd)|)7i>F)6hsNe%1vv_lHMi|{M(<yuY!C7VD1|ApX+N11AF}Ju}{vwk_CQL0>
zEgr!@rHU2SRjs(sXi>>y#2KDw8w={TI>Jp+*{$^=1TJ?{o==U{Oytk$RG(VEl1rr9
zQ}X>bl#4S}8v{iJ7QK%cUX<NR73kF=G6q=xR6cjRSCej}6Fxu&oO5Ob`n)}Y+khAb
zaQO3i7FuK%p9eTH#=M-R&<vlB_QB|%MKaH*ScaHmj5tUlR)0s<6CQZbHx$$nc9q#~
zrk25DJU!?~hsXKbAJ5}Ck7|?5NSJ<AN6q^8*8bDa8zQ%^#PH8QzJJcsXOfO;@7u0T
z$aBm&xZjmH&Ovy@aPZOQg@Gt=6#Z_rikS-%`Vlq#(@R7D;-~3eq^EG)qO8trpt|V=
z6L`##IEh=#=X#{t8G-gqLytYjVx8oh2U{C2V63!oxINdxtEq{C6B*?XwHp_PxGpN{
z5vl^=KBB;{7zq;EgB%g4rvcACTJ<nE>u(@~nPZ?gHunU6=JRRbkMTGrmDbFdVTAZ7
zyE)eJz2_w3$ZD7_xU-1x(J`qAHfrMzk77`(8iUXQTVC6oi#_bahtI0v`16lHMpUhD
zQXRVh?E7~A*7yDY)z~OjMH7YaagOu+e4Izp`^R?2L^ytbB*@Rd;_fH&1kT4}MjmHS
zd)UKst7<`A^s<P!kJTYYFh+Q1h1_D`v;L!w1wlg=S1dy#f3&9{WY*pyGh*mh_!?K;
z@sSDa$8D(#3}9r8)d*%^DL}-ay2ltJVz`@p>6_GLL?SBMRqI$)AdY}sSk}Zdxg1m`
z&e6=65$LwiBYl<=3fv&b4n*X0p3<Lw><~FV8#tL>R>uX!E?+$6@Igid@U#Zghj*(0
zU?|~H9hGWa7;Xms3I}4gMZ#x_LM)R0)5;4EpE=LzamF!X@Z;mxZoU2b^T$uO*3H`Q
z|IhzZX#YYodMCyl#~8@gHWSFB-|~UnIy>Zy889BIC;U{LdB8`6dyG0KFbwxGp&2Wv
zjyU5=#R61#%mLMJkDw$b8=S?S*T4`_Bl3+<51Gl!swCgqATlBl$nD8)R2UwNbFw!3
zI*xF!$W6o`f;odJhUe0XsBfsg0c%kE#n$z{V)SZWC&HsT9Y-G(XIX@bVyZ=w?TF$?
zI$=tjW1eNGM;zy)bu}W3G6`4sr<jr9W5kg$0u@MiH_j0;Rm=N1_2!j@I7T2sD#Q<-
zUB3$0dh85P?Spi`pqDtj(xe6{jv8j|<D={C*N>0=wr#Y3flXV#e{4;*Mr&7w*8Nqm
zW=zcU`3_$p?cV4#Au|ehn>o+%NNn$$iI8IqWZ!rmtR?a4UEUdy!I++veZ}yg;A8`5
zj2Kxx5~i&SNvyC6kCnr1(sv?eAX*q%#7PC&!(l(j!$(A%LEHYGky*BO0z{QfCx6$K
z3XWIv`%KYV-LfYf^t6fyE5C9Wh1RO$hE&@i3?2)}tNtTA4xf&lBJH*hwH<4QXZ3Sb
z`{|BK14cX}+>d~^DIOK2ov{+pJ{I5+Q0+DEhO)xK5$<TWz9Jzs^r#Mx2N_lI95J#w
zGt`>Y9`CL9{;}WscK`N2>9_m6)Bacf)~)y3$F5C{-uqSuuoKt`xD%a;=kK3eDS)I?
zEJL~7`-@hUbNG2~=>G9>(?W~~bmET(ZLupSnD)jT^PFd*#ta{i^WiH%-m9NG26alS
z%3ei6WXBhJSevQ|K1OY`4z8hbQcj0_{;2k(7_FSa&8pt!5#cK&r?S1TNPTVONLj(c
zj}{e?)cUbtzC~rnEu$*EUdiLRdg?zfT)Ci~bg%4wix^vXbW<3y0x}dnQK2(@L^9&=
z&@DP15R6w<8~;iHg*uR|TXi*at0O7$sAjC!=7_7_n6>9Fhb5vJSZ_B|m9EyT+ikml
zeEap6Uw``fV>gvw{)4LCZvTGY8lsA66VoP$A>{&YqBqXRr&pDV7+x{D_I|&c4<^u<
zkbc7U@$0>-itda#ARnLK(c4ZFNSawAVvNV}fW716`S?7Z;|$Qvv@x?vf+G$nu37@L
zjjX3t+go-%pU08RQP^P;Wa%@NKNedVQ4?XJw^nJ95(Ub*M;vpV4X+Y5roA&FBl8s_
zWK=VA@|EUZEDI4S#By~nzMu##J1l%UufDZ-8KM(vy0xwK)iK5JfX=H7ud?ElV0b9H
zL_s2R#UY|r%Pm!%Q-a?2+Y)lIu2If}81zm&bD5^z(B-G+S^b=_TUUgx+PhhAy(4|!
z?jQf|*I)mK+s8+{@xQiDRC@a<KW_bXOIp?LN{L%Pe`)C6naAbpa~y$c+df(|m=g^8
z-ZsTSZuf6$C}GY(_Wj>$b+uz%x;6tJXCFVU&G$GS^PG?80sC&7K)>HR_#8n~V$P8h
z+D0+snNAhA|9CtCMv44Q0+Gf{dbiq<CcPLRhdXvN@>%tL4dT?%@%CA`x|Iq6Rt!^o
z#BjuXc^q5eJJ~SeGN@%MdO4hTmy|2<y?kC=$A99otS6$KggsZiHS;P_%W?)(W43jA
zXOx!;{0j2|jX*6y`|>>6ZNF)2`gT5xD0EZ9gs2ux?S@Zn0v@B<LvMTUn`tw^_PuTG
zZo1!py8ZI)r(f<xOaDJMf2y4J_}kAv_d>z0v{v>D;{Wjf(12f%>MY>%aUsYj=Wj3%
z^v<^b{Og_HKBxS^?D(oj`+EWK>m?@lAjW+B_s>9^blW~Ye)*>5T(4D_IU$eZe4s|#
za9Zbl#3!Ef^h|tt03~2!9y39<-a9!*pP=_no<4mHWrT`Z62`;VyJ!dod%Nw#acT_7
z>R`?l2EE__OX|?%I)qWJUV@V2tRvvcF6*&gb`!6nUc!t}v-9-pfUNA!N;)$m9Fx~a
z1(|r{hy+n3Tt)F_ph}kv)OJBc%_LC*bEOZ=+C{u^fWEa-kO)RH!s-2fzunutcOIXx
zRjbhU@$dhq58JoBTEO^dMPKuI{I}h9?F$Qy>Mm;CI{N(n-}Za|`L-Y_CjDchT^r;w
z3@s#C8cpRBkBoC3=RjKD;c;SW`)sEtO&DG$u2ymi@b7-b@PTdjkHSgf#hufA7vOP@
zGceh91l{p-IL0`SAWNF5L(Gx>Ir=2mB0_Hy*c8<`54B-T-w$qt8uP0E3VOdQsPz`Q
zkEc@7PlQK==S)P+97mKf&T(<yfKc*_$Tg8fPz@qUB#`nx;T}lg+7UsUOE$G0aP@Xy
z&-?0xHUjs``Xb}cU_~>?WA7k_?$V<`;pLdRkoPN_g~Hj@_HEn9kz~XiGn{t2-GAOU
zwJ2^6fPDP&lgR$cz4QOx$k{wV<XOnq0@qM$w|3iaws(K*cK`hK=MB<*CfS<wdwLe_
zo^R5v_o;vTjYL{ENslqo&*$-53EFP03;1s_61p_Inj)30EKu%~8}lUa$Ve9@{rrw@
z*7#zFb4u&oZm2aGTY&aBA1f9`I>I^5$L)E}G15)CIg+wf6k_ITKH4;Yj=O;A@y!vf
z!iL^HjErc4-k#muq0;dP_wYFi`*EQ+`qNrzau6A>R3<_dG*U!SbS8nk{5gb2Bj?|#
z_7E(ICJB(@P)pjiEFK>7b>Iow%|I}8Taqh)TBD<i-R{OSMtLR{1PW;Vx0-F6-F;+6
z#vwf%yDi=|j|Y_aqR#;g<ABKL<MV%MMG%lWo$6vL1yJord)$xRt?|_Fx9{IRZurPK
zR_OX=>&v+g?AjmhbIfzj<G15*uzvrb6CQJp)@H}&pf_BY6t~oJ!Fe77enNNWd4A{E
z50KVF$#kD5e|l+G?7rmBq#@RC&yVfn>0^xNOh6`5DxHZo+y_+gKFPzEMpi9HzrlG_
z!hd_BJy7ckhpKshYQe{#is}atPtTDaf0xIveVC=1^>>24Uwc8&4uqpf&!udlxVeF5
zQh=|=D%~h1+U(Z0V$|UfrOsg;R61zC-R`}e_RcHnn2;qRRx$dUU~m?j^`w9Pjkd&H
zM6?6rllz20&g1*v{$)9zq$m@jx>`;|6^|T`ecze@`~L0MUfN=)w!V!@Kq8qfPdIiN
z34>?f4L=UgdHnVmekA6}M;kDn$)F2!<nz(73gBB;WsbqKG`OG7<Kyni`Ta=o26K+S
zRHJ}7XFIywPW*(8UTZd<&@N)`zJHAIe6GcrtDCKoRkLkttJ<G4KfStRS8uN|=QvIu
z;p29|4Mp3G{ZQ4FCO=z+J9g_Z+IebJq6e=gX>7GRNvclc68KpFBC1TpnjzJFPgFg~
z3qM2%%-3173f)F*&8#b1Zy)<j&0yphqu2;+turD}?cG{+Q=hCe>e3KFnh5BCC;Xm8
z@0v3lw{trlrNXwW2`Ss{e&0A4;Dd<I-~agApZf>i|HGLn<NUm}PB{JJk2|k&p>8H$
zays<Zc9SK8D!1O4pBZt^$1KUL=jXj?_1YQ<<1x<(?fyY~wcF+6eB?Ym_~W<xPalS#
z6{{x9IeoBqI-W4AOJvaceRRg@emoxoTDN|a0y1TfG2y;L&U4JuFE(s%_pM|f&*{)N
zElF{tGTnH59)~~YbFBM~#@v<Q87SD;2XO1Q^X$;7YbBD3k_!uym*1(HrHL#J-4zrW
zSZjc6?beBNR2%-ti3m}wP|du^%8j3Bw_Uoj_3h){O9OH~7XNe58}S@t(9Jr}+v9wm
zQOb_iQ#yW0M>oa@<a-B9x-yp9`V$Abt83WmLmIk2pB#}#jOX+5+n+RBRHg9S0Ky1$
zH!tVbkE2^JbRq`5@D5`JDto^yX?kF7Q({g?ibFqC4h9<{G~2e$IL`yx_ulr_rEQxM
z=kfVW&zvLJ?;jh3e*5N)NM_6vN$WQa@ELi2K2Evy{c-Ee$xEFe&)<>#-rLB>_kpzH
z0kECsb<Sle!Yv2=VfXC=$b-h2rPg>HpO5o%et(SV7(RwqAJh%LI*8O7XgtC7s?N=`
zBys&Kqk1Jt^@S?e7=T2#BB;@CZ4u2fp2hC<<47;H9uZRodhfTscR;s(-|jRva`Nk>
z`iukS_v_Y%%EQkQ$CCP{%Rn(92b{#S2s9m&$(LTU52Rr<Oj<LZ{e(}3AAb0A6dB`(
z^X~!<-ichkit?i!`)#)+A19+`DD1?bx5I6iioVDN+ofx<WVk{e`A25PfZg_8$f#xs
zQ!}+@eZ$W&tB1nr_u?Yxz2D2Lr*k{y6)1du|70?n+&=bRrcFbIJZAde+%_whD6<+C
z5M#?pu3EYAJ)rLwKV|=^-Ff5z%Ji7$7~}JC{Bb--4qM!^?Il6AuMhEae1IRvplCR)
zeBm&^N}F}^EWVxgYHMtYz0z})-nw<W0gthlI_&)P<AmP26q9XN+1a;dU8G5!@&dhg
zRk{+-lX=Q^Y>RCF5Q11Tf?4grxOc)=r%J#Dm?+lQ<CR9=ZvCTgG|2Hgzn)%@S?$7e
zg-FG*?uDMd&Z|Wp$d&ln=8-oKAFn2ysRtg6c5_9V;W>OM=9uAG<pE_=B{)1B;kkfr
z7RKGx_5!;yxrhh*->RKN`_`<K<g{=1)|iuOsGegE+6|bklwekpk`GUO(5&rmil(Ar
zNHxUxDKKas%+dMy++M!-s%UR_Jm&d){_*+8W1RP^7o>6?+bWeX=Ag));|vcUj~l>Z
z_fy`v7uHL_wrxs5%?;rKf|PFE?)_WK=Q&FDGJYC9Vc%~s9Hh72rM0GQ+ko@*KSdYi
zXU_2~P`c-<Q@EDb3?*_DrMAv`3{{j84wGhJrcH1CV^u?w^Y=$JUN;RDf$yz>hKiO#
zO#BdKR}dEqo^F1Q;D9-i{k%9m%n|g%Dl5y(;Z;ZKecdjL_P025UC7U6>=!t{RRMLT
z7t_Juc?AKwdT~_R&AKksSZ!%$^-1!<yvo(?=Qt-qX4~&lviDxvV#-XjXPaO5I3}NT
z`rdE%?c)SuL=BNyo@M(9y+6iGA6{|&%y*Y88?E2_ILADP&+{Wd8KvmKfKt60&xirH
ze%IHq6*$h-qNafLz5&g8E=^aDfPKF+$4GkL?p+mWX1a?XpU*Qnx5+??i+4V!ALBWE
zlHJ)B(!DI+6(julS1BF_YPv6$m^QgYd)AB}KR?;p-6BWK|GC<5mfgUzn~P4C@q<CT
z9oJQO(c)y$q7G9<t?W1o3}4HwrsY`TN_sn-<vw*8j?LosjWJJPt7^*^C8fyf*83M=
zFZ<&9-9JA+?!ODDbP;0sIR?_L=ll`D7Bkf|>7xOgK_?}L=XXKccI*31n^OHC^38<I
z=@|+4jU0sr(k3a9S!fqVYg?C%ZGM~)=U7sch$%A??M?A#JkkTTTfcLOgU{Q3d_e)-
zMs?vLx^G2qE&KrKiL{Tcw~Z5rE{)ws8*mnI2RTNGoX0Z`c-%R6!RIV)&bZ>2_g4dY
zmkN+e8`GMGlM)xed$1#n&+(w`<QOylw|4Yu7zhG&0!dL-@fVmSrPWxG*IY_M%m15m
z1<P00@C9gQK>%Be0vcij5jb|R3X0O>B6F%*hv7a0oyB)w2D3^!xkCR!@vzTjM8Dtq
zed~r`Wus#{W~$oz?O()tzi*fD0Nb|xQR4EgG&F1KN4=g8#H95qmT%t$rn9=7mah-q
z8YGCE+C?)@+W?Vo<a5lj1L+;|3|en6p7HJU7<9W+#2jcXUFRT?U4#_I5bY|~c2v4x
zUMFNj?mreR?r<DFMtaOKM-ImqL7oxEKrQf3DgK&5&e1ElyGg@q<U?UOCekoeOsmOW
zaqy$&apePVcX^D*Z>vo!<JlN>5(Z01VKq2g6;G=Q{wl%)89|yPrE+DZg-N2JaV49@
zNQlI|<}}m+ka%1Tg<q1Xqz_coa3|0};`+uE%>~A15n5C?4PBi&9zna^cKw>YG4U&b
zQB&LW-fw1iq4jFF9wRr~w5zajG8pu7UQEuVepm)O&!RMnZN0&qszkQ^z`k8Vvu-%;
z;TI*N$Fegq1ldNlZ*CutIL@KS9_5Jx>B6kuS<&_;z-?{+;V2Qf(*)+(r{Ocs$Eh5T
z&(ATL!(;WGgwJFU@yc+rGzNt7_=4?QzgO$Cl-B1AL@_lfDQ3B}cHG0WhEd&M{Qf!o
znkVqvbqrhG%L-XF<-&&Joe`vEFeNTs#^Iv2f_8(Y^lom3M@F?-eKGXh!$Y~Gt}`y{
zhFtiqt%(=1vR)TM)VPS+>xuYkRtn~ml)KEFMbn9q^scgNuWC!zt@W+%eQ!HC9~0hH
zE?R5ag^Wke%*b>2sE*p4BT!}{ZM$6q(sHs(N0OO{GCR}lL4x#~44)HCu{ie1E%JH0
zO4hWAGtds(){QvO;_7vQ4P`Id<vfCksEG8oZIFx{--CARay~r&`#8?i&lpDri`Vg1
zz))=Orqn|f$nZJowtsxw_r`ReRY+r-ft#9`lr1@{=QNJ2=?3Fe_<VW{pCoZSzh>lA
z!dm%j#;a9JD?n?!*pf!zQ{+#z?yJ)4fcxkc3xuy38()eQueAr#MZY{x{lb3u+`I-S
zH1Q^S-&FFAM0<krWPhw8vKQH5DE73z8ze_Nx@f23D4$8c-#2=DHI9GX!ZRZOl&8!j
zb50+nF99Ipmh-G`T0aW#kF;HH@O-#qnD%PFvaZiI$kK(WF&u<8ZML^*t)|1RN8j5=
zUmEWAbwI91aemIz`GXke>6MNRAM>n?d*$}h6%>_S=W7w`!A9%ocH6dViY05O&NRYI
zO}d!pnlzHnbIgp3vQPMkn9MT<fBXqKURRhBUKNT(lU(Y~r5K!1X>&=#tH~1Na7YhH
z#bJ0kObJbSDV4G^!dEj4b#e(>zZSg;!Y{&INm!X`SG21UQWOo3Iq;O&;%QoqH?d?w
zn$CfC>k7hdg}YV*8pfl2nN9D#HaVYTMr6*#|5Dl73~wV7=kr*DsgP!Z`$7hzhI^7Z
zj;H<s3^y?6SV4SRoFz3sDHvUF<{Ij>m0;k~JMZ@n{d^{+qf%-b$ANRs&!bo->EkM|
zO+RvaoRy+lx0d75MhJbrbo8KRA??=|>1$BlA}&s3Q(N*@PI!8Zr;nmBFFl9H1Py;i
z{Mm5&%BX9=%PU>JzN{A9rA@Mgbl!17Uv9?`tu!jRPRqQ!hukhT>WYnWB&4Vn>&uF5
zv^3=6vZB|C9bv1L01u|CCpo9;4~$5MkR_;RV(7CQy~?<CHa|_RD1Jz<&fnfTGvY^`
zYTESIFx5H<I&Dl;qo({Y0y^Pb@_H|}o8%bq+=r3J`=<`QiC{BaP^r;*)VM$p@_eRW
z6Df0kPBJ|iW1dlvbl@eR5huki2PG6om1E}sj(v^{Y%kJr<3mX89M;~8)UyUvz5M($
z=NL0)L{9e@hf{ojV?4(Zb*AL`gW+_Q8YG)htOmY%Zf8STFp4j!-*SO&R|n>Di)L<i
z$@2-b6`?n1NSEH+HCK59XC$JgT0xSCf=UwcJ}96*{R~i94J}K3s7;To!006k6%Zn1
z!ZZ@y+Sb=Jhkm_zvtDal^P;-->tVjMFneQ;u^Q35j$K@@Q0xrcH27*PQ&Vw^qIm^A
zY5`YZdX-pfLQ<bT9%^^((wLIcBqz)!=d&h-2s86B4$tZ|j^PvIiuI11{sUanYgCBJ
zGeA=gGD7p%#nhl$1h5w!LQ1@3MbzEE2RY~PBXU%i${6#xJjgiDa}Hl{%)8<A(itZf
z66k<S3r!N4s%9!(tJ5nh(s;XWDII`vF^VZ(Nq(i(MMRa}S}R1yIyzJPKuvY&9_Q=u
z6|1Jc1YFyk<<;Vv{g_Bwx|W2Cx~QQ+fTv&yj+N&eZR+aZTw)CuheBJs#5H=it0^h2
zmB5`)1iP5W{sn|Y%_l<3|CtF=qW5L=tH0HsyzE&eL7Zti`ixU}e8jicbDSrf;CYFw
z<rwD~Yea)j=OC{}93N2)i1Z3^in3FPaC>o7rBl<#tW_W08qStAXY$%CdBDjGd`9_4
zCwv^o@MjQkMnut3G(zJ~F;3{pguj~M>Um97MuF3Pop}prF<Z?1T4%pVM||;X*NBI>
zc)6e{(oq@_-A!t%1T42$Z>lu_h;!K`tB+1W$Oi7EwdI&2FXZ5&`IfuPF)Qhn0@cu_
z)gFeRy>;u)EoK!VtA}qF@`+y!ZYAkr)p=|S)}72c-95-yOi&+%k^J>5-6|RcdT4Tt
z&p&nrn&OWW9{ncaHKD!0jf`=4FqX+k`VnV^6Sjlw=HrW6>`Pq!E@T{5vysj2YXnOs
z`DtB+6EUGN1$7_Q-s*T{=0Nc{#`8EM&RHy=st@UV|L1jA^5>o+X~e5$dcAaskff}k
zl+c1XELp?}JV_(cs<ExcC@I7&Ig^AOJv2bW#H_De_bn7yns;C4U?S$a)Tsa+2s2rz
zx*J*e$ZCR+8i{OVLe|)MA}_A%<uI+8ZmPF_Q7QpAUNTUn^{`5Ug^U?n=gEYF{%Yfd
zSK+pG>p~Wn5s$-OjVYc&b1;A1Yt;!G5!fb!yI!>Q#gg=Vb+!Y`ZeGi=e(F{Ma!OO-
z-VL7>3yay^c7hq<g!UMLNzbgk7(P!Jp)l&A4?oBlV?d0kpwk1FnwtX0yWx!f>f1#L
zYEtNH=2HPgs~U#Ze6X(|OEo0zB_cQ?W%ZnBxV<RbBgP^c;%J+q>j_vOH40g+GuBW=
z1>#D;r6dJ1E53F+t0~1UX~Ju?aK6TZ)g4{J!ZBXR_M3>nBqn;ci}r8VYN)KHf!SUh
z0@ZNa?ug&OC+@3<*@z{YE121~b;&Us5XZS6*T9HUw=nm?yvW2I)zgYoj<p?h=WK;#
z)|6kgk6h2t_Jh^>U1`^-X`x+F)J;{`yLI|ok*89NqdZs#yt2AEMh(xTk<?){^D&;I
zqEfWSI)F6LZcS9+2O(Iwo=mkyhcqtrJhf7UiLuw}Bdw|<`zss`kgsUHhfCsFKw!KM
zpsS0bNN$BRltWI;?W^1?XrQ8`#>;e2^_l@82`*YB%9a`<!K0TcMcLr>+11+QKO!`0
zc$8mCGQ8Ey^sVV-b!t{J5c~VEDebn4U~aFKx%^Fupj<%w>$&bc+v?asI3wJ><P)pi
z&b2SOQR4~Rw4`fuG8_-5HE3}tIG)E5t9x;DmW=M(Dk|(|9gW_R(te&m%%?vwMF(1A
zg^!%$yi~>wd{&dpc+NraDq$LthIO-sBBuIgIC4QXN{||@$^bRksX|7tqID#qXGMFT
zx;lZrrp(1yRw8Mh%V(vFR-t@3lGQY*rQPhShbX}OI;SDZg%ze=HRLrCw+^pBy~a?<
zmhR+fh7?nLt>q8#yVX;;+E5JQ1$-A#N8K(VdNrk)sp(U8S7~@9jk|Ux^o)JYFz;Aj
z3<;@(^~tE}y^*snrW-1Xj43}B8iOUwsndJ<^Rp>FJmLpvK-dpF+~aUhGD)H!E`Y8|
zH*2z^$gbELp_(gZ)?C7nN`?=WK5;KusZ~=heZ9ySF@nU`$uDUOE4ePlz6{6o?dPJo
z+WmFa+yB%ORk$6zbz7^f0);QC7f4DXsjtOLJYKOgMCI9%j9SwHQ3IyyuO(x2bZ4%H
zJ|bt5H8HL_xL>0hZNWRerpwj^M&)Jc*EMv7R#zXGFR54{nMrzq=Bg<XWrJOfp}IL>
zyE+B2S>CkUQEIH#F?*`7$Ks4C|LW!xT!0!Xx*Bu`J72_-3q0=19G`_TEpAE-R!@3z
zc#uv+g&Zk8rY^k|-OUu+RrVE6=GT}w0e9mJmX<JH(HhV3$d?om%y1$Ync&44<h){u
zDu&mn^n4kPsMe0J3XxRb11rceYeHW*TrP-Cd6gg6)uFD?M>0XbZ0%LmWn9Cl>ot1K
zV;`9>FLOnqS5r|a^gguBcok7!;<GDXzuH9-QiilN;YDX$IeR7JqOvb_H?*Lm>jqYG
z8Wp=Or8h(lQ&fv`w9yE-3a{(#!fMpUiTO&386^ihv~ckNR1B1DJHvc_-Nv|_tzUn4
z>#Zx8D~KL>5HDGfDvEd{>6YZyHWIV8Y3alo8g7KXlA6`zer3a}^*r$DN5ue(4>(qf
zp)A}g$m{#+bJD?*{K~3Ud2B6UDF(e}E-Fp-W-2QW^-`tk5QBM!ERCucA~_i;U5xlb
z23PTd&<t$-eCT3B=L^D#B5EkbXvv|AOd)!$_w||egSalU`T>%33$3&7r6Z8*a9m-!
zzDC~E9H(5Vk*w~^68Ksc7utg{0_}BrtO1ntSP$AAt4ZpM-$vycA{V-@F%U^)eZMa^
z8LX8`E{P^np4M(gv+m@4<lzC5gvW?kGewr03~RT3W58ZDBJ0;Vy%Gw&EOMQD5#S?^
zjP)~LJ?t^EQ2OecaV)F$kW@YB@?j0$a2&s)NL4{=xL#=(Dc0GmF4eXQ;_15RH*s#k
zmz-a~SkSUFiClKB%W|tLO|LMhlA$%KR5APu6IfcYxLYVOGUZ)ssXFZ=QC;QtG-C}S
zTivx+iA-2D#(?~MWACk&NhACkCoTDUKkJz}#!}*nMu0v{xfr*Lk~Fyja<xi!O82P<
zap4SFfI21h)n>+VrSe%_5?_*t^RtwjM4%=eCG+%P)?`<U#0+VB-#@xhLY4BPriRnm
z>@29ObR7o;BAx|1OZ7XI<6Eusm)(F=ZEY=7iO{HuM)G$>&h%(1rNLM9Oh9z2N77z3
zah1KkVzgK!u)5^ok)R*U^dMh;W@$;O$`7DWzHsHG51yA?$l5gpJuR0TR~e#|yJP@D
zK)t^uXB8m1BpA(VR9@sNY;ILN&X|3G+k&<)YtpS%uKM(gR@`Z}=E?b5d<RmF7zSQc
zSrqB2w36a3O6xTNLZfgMA#$nN#1YWCU0pO8D<&?%m2rU~FGa0*E|SR*6y`aq4t|Yv
z1knBC+rC(qeN*CmMG)24s&vHRSwVCkN0vg9ul(d#eRO!qk&sBH6H3amo2w3JB}T2z
zUsT6)HjPl!ta^}Cw72SRI(D}=DC)ZSxO`xEW`rQ7#<Y0Z4Vl&L?5L!fzQ6xXK@F~~
zCu22WBxJ~?3FxcLqF;Z_@=P^&-c)S)@)x^&d8p>yvC8fQ>2TISfo{OiLt=m$BC1V^
zamAPEtHJdcrl7~_vbS5`bv6lw>4&4@+BB5#IU9IOk5zj(Je;muP1*+7kC#eiEH1n1
zr9K?t&QKa&iARnIJV5%so9T+a`>tl(N`!&gCQfMmC8HlZePler!(bfw@X|iiHNz-s
z5;9pzKkIP?f#r(!zYvN7F(dwpR*_S6MI7~O>8%^!gMl_YgHk81XeivHq#WgI!9bQ&
z*QL}^|KWP|F<)k9vNUIWu_~&iKi%c)uvQBp-65sVECym;K}Cuz>9z7FA?wJ2<Vm1Z
z*|lM!*W_LhmAAH)wMPv>nyQ8l*OdblKRt0*=}{W4qw|Et(wcraG4LDGoJn$&oYpAZ
zjc^ZzsVaMxySq34RHA<g9bH80nfREBGQiS`0a`%$P4}(cN~+7$z__tBY1)|Qm~9ex
z_{`3@q$y^BMAM(g7*Um-IAi(crSE-t`a(qVqVFoUyn6jGU)6E@0Wr9?XvK_j-3~G4
zxpY;oM9Q?6qHjrVBr*#WY2k530veL<y%V!(Eb-PaT(IoML(R>={Nkp*9A(sb0HQRH
z1(yhpbhi{D97}J(NKv6{by>^=LRMO0av6e*RZ|TJ6$18ySLaV*K{YTauTw+=LdJ!l
zGx7GkRng(7YR=%ME26q~QVf=ASP@e~pHf^MV2@LU@*m?;(PXBFPgdMjQ120dY)aRy
z-R$1YP!ZC{jjeTQCHdZF)3{<ClTajLk|Sf}JV!)$5PD_WTK$c$Dxq<yR0A4{V%j#<
zlA4Xcy%>O^m+1(JsG>q?vDFGcZwJ*XG*_MHD)*>e3_e5Bm*7AdYFsj%%V%7u-E0}E
zEG;xGR1DH%9d|VfK3I5_EWLhL!Z8hdOHDyFH<!qCBGd+su_%ld7nyQmlGXUy8M&(X
zd1+`ZIry4=^3sIH0%{OTTuzusQFEmfrhT^O-!(TC;&iS#(6NN^E^~Z)#k_hKOPb2B
zHU~Y!L$5fm@B>w}<_npM-TSuhAm-<6Fpafb<}F}5yLJpEbe`^b43Bfn<GMmmq7((=
zEd03)K?Jd{F??#Ks;_P#pG4x>zKR6ZSQr3ZuIT!*oC@+PRQgJ>S0lVG{n^(8tinH$
z`TpatG&*1FS;LcGZEkgKV~v<!b$Shz{L;Tt=F(eZCLvkHj&)yUv?g`R%?r0K2bqwb
zogOdY={2cPRZ6v8*A{whxa-UdO9Uy2_)ta7Lk<5@ZUb{Q(=UGi+e-tz*&=QIFgo`N
zi7VGASZ93gy00bd3+-0yW4EpE(jiL=GN0Zz4IhDlMzl4*j5)){^dsXq=kq!atJ2+7
zR72aq&_(7{z_%;9ldf%Bn>5BtErZs-UHWm@4)a`rMuz8`hE^uI+|8c_eO8vcO6G_%
zUSlA0{WgWy<J32ZI$tEgj98IZkV_sr6Oqst8KbuRi*}foQF`z-`=QW_63svJ#b<7^
zOfh(FS4(%}UCj689|=a#t4fCzH-CM5Ap#|Gx{iwF7i!C?>=k@WZijtnB`sHH?JF0u
zW(#p#s@dy0%zRujl=VDALwgP9zAZ7UQgSh9s!U=gWX$vEKgyoj)#p1s&KP4HK6%q+
z))+=8a}*WsB2s2r^u=8gv#taq@`7{#)Anuq8SSFo*p+x5&nO|*F-NwdDE3;Bi+Wax
zDp!>x-iQvNS5@)LY~D9ros4n?>X~sdW^&oY)wAkyF-G-DpW~{+iWcyrA?SMe68f4N
zQfV0L2p80vQX&Wek<TxYVo3qCxT_2TStK9wq6pIFGEqTNtbm$@Z5t?h9eA=(I81x^
zj^TwR=Q_RDgOFxwXhB=%CBPhYt=mh`jVM!GDoD5ctvb7QL_k_hqx2X7I8URpgO$mi
zRfvj_F&HDxnB{3+6(EOCU95!K(Sqp!s}dJiNIX2^T&@0y8m({i-c|O!8;t$fXJtym
z2N6%7W%R^q*juER;F>QbFU(25x)okRp-@O;(RO$(ZCO@?hUR5SmsgQhoRlJ%7I#qp
zEVQm6{Yv&B-Ib;MNdV7R9$pQ6uC0(p#}*u>AwPZ~LC&!*GJIU^e{JZi43o%ewsteD
z6%LB3u!{&6CfLh^Eppy!W1XdOW3rMVaPty**5!<My$JWgbWC+=VDJ0wW*cB;QXdtE
zF&|^`yjKwmmJIUrI13&2DD@owQk-0m+a1y3e~Uu(u9)XlOeVC#lcmoOuA;`){w|%`
z1Uf@Ih99?QI*yibqNGMq-Fj)&L*<HZ=2GgMP?1hvE&Q*}A5us`<11GQl`n6;suvki
zZ9mIyxOh;Io);fFR&nP&Yg>Y_`896p!bK2`EN*2MldAT8hI$rvwdQ54XJP?pYH^IN
z28=hbU9IWOJO;3<ybsh{6d18yBqDP!PNb6sbG@Ef46p9B^%Qg$t^6pSnPkIy6ZUQh
zG}&lGw5ch0JjYp%4-1mvGYMi8Bl!vqi)Ehv7lFpt+DSQ>@MPt|vNCITVwK8G)P!s&
zpQ3xe!Kmf>vEsj?>#woN8KGf{eK{1N*EoS|;a*_T_>#4MxtUd62r?yKmQxE=jX{ge
zQDm8~LuEy)m3x&zp=CL1WUf@Gqd#_hRWd(eAjzM`C1#4Krwb}xvHZ(!UjH1qc1zt=
zkHzv}Sea9YLJ>W~*E#gsNLG1JMZ3m`ftW7X+L<L=%^NR8)aaPQHi<mgx84>X>MiI*
z))1qjduV7deEN|=ct!p58ad**RQv0HSIbIJhnrl4Hx2wtz1buqDYeSgCt%eR&<;+~
zPot(KM?n;AtI-VjbRw5TN363vUZkB1p)Y0VKHhxb+iPC2@rwV}U(}d^tE?th$?@y`
zdOh}Q0jnOmj>f#IhgakvippyUvMlc>BAxKc{t_gu>(yiyR=r`F1R`16qx#yvYW^Yh
z&ozf?5tGBM#Z`?(wRdO*f6TXi0;C9twscUj)I@6VBtRScQ>kI7T%&k=U)$uSM2+gI
zl&i<kVGM_EA~;s_tZI9ERWHMR#1X5NKly@4E;_;WWHsm_`hrUp9hr+nC2E8&D-y%*
zw`*}eUm%32id|{R<?LUt#q#q3h{&7sfK~6**J-gFPBMr!?Bsfcu=3=nwN)({>4nOa
zMb=cbeBIsUBwn4wqSwOIIkg-WeB~`g9my4lc?}JT6*vLaHbxE1T2XxIPh9tq%ZBTr
zNQRtS0P2@WH=!PS?ANj>H3dxC;sTYHV>O#~9m5ip#(uJyY~+m3=Lp!mv}CR=b#W4|
z{PB$q6@>{3!U(l+U+fV#+=q`ia-PR=$={S2M&hAF>4scpD&*pATw#I`^{m!IU6mX&
z+%Vd$-Kix1qwzIPN*W^M6>zTHYz>sw9}}cSrN4^770&kYR@KO*#v5EcRF{W}h5lUQ
zTVqAXUuL@M_@b*kewE8t(XDVqxd56sFZDzDG`%XwSD>AhJV^Q~PrPyS^E&?JZAWlb
z*4N)08F}5y(l2u>_TJ7!abALM=vGR40@`R=xFbr_4feeObNgPb^y%)aQ|#@W)#5CC
zwLPxtrvfQOu*s~`+dTX^9?r}$j*^!xyZ39nq<*Q^nDi?Zr1-gYhvHR7U+z33D`KyA
z_4We&eyCODE4+C<|8+1ZrdVW@G=`aws{`&r?-GG6@wG~;aq9xd<jSt|<&#z$f-Ty}
z3&Vc7Ix~^v3q0^p;*tk@D_*|c0t$?Epj@;iscD~HXLb3zmv>yr#r3m=6I=|Rg^69n
zoNE3vt-;2C+rkWvT?dd-!ep@c1gg@ONK?K{+*Egi4&PP|nI1Xj)3IK%$tyAPO9^k;
zjfQGXnG%wjp;Y^61mii+dB!!BCczPPt{(nCbNT0W_Fgb$754IlWdP}tkg8~Fg*<vV
zF=NC?>HKByIDhwJTcs|*tle}Cq287w3g_a{n{yG1ueFTJZsvu!OM-|rL2@apXi3@E
z8xlm#M6||#UXJKm!xyl9SxvQdy$WBs3Z}61G}OR6(o5_7D}=ea(0??=cTjLi<O@W>
zrVjJ#n?fPH>#^$)1EIZz=A&PHkx~;Wj42y~2F$#p?b;XTIp_^ZpIP(1hc1$E^=Q@x
zUexVeVhn|VR!m&{jAb|LT>T0;s?{IKfE>L5xKL(O$yg(FWU&t}#Zy|-TeqF~JQ=~N
zIsM5(Bte>5H>*<gI<#czju;?qzu)F#pYMg2uWZ;?YhLv-*W92VzCu>7%Ilkrxy~$I
zE4sSrq_ikmKmD3)R(qyh6{}ZeITw#AYJ0_ke$qpu;faTf_5uTBc}Xtr*NY&pUrkFX
zU(KZv3(yUFZ<kPJ*;nWn_8Ar_<iJ&D$xn^I`reqwoWQLkQY^}PZQMuFn=sGw2+t8%
zng?!LVuD>Zvub>*<|DDi8=U?uWuWkQ&egde;D37fD1vANI$rFJK2^bUWT2w$)x9sG
zY_0chyF&AW2Pq<({%AO&hDfv9PrYNfAJsk>B?6|^ie-Ci*8xM-dcJCz@tV<71%?XG
zb*+69k@qT2)Y&REj`o5!+*dR#uay51E=ajPOHCSHVN4wwmq~$HF@`eqlEZX&V`Mm4
zJE^WM6#oigu5bAvtca4i&Scf>k*|`8_N^f;Q1aMoh(jS`%CX0Q7+uFObtc;Pw&8lj
z)AbWNruW-k@kawQX}#AF(?ez-)m4Z<sR(a(C76<R?p1m>+{f`4Nr{-nV-yVig^j-1
z&I(sFAR`wnJYFQc^+!cUo29X>mk*eq$H;19_)|lknjk8hVRWTUBl4JMruv&BKbY?0
z^RgCS;uBt1yACbkRSaDhyB@Z(O8y~*48Qim7yr}4YII7QFDfHJL|aZtP1VdvG*<F#
z*ZI=P=w4-h6XcRScgI7r0`3sya*az<k+;6}*Hf&<jo0Law)UZJt%!h47vu^Vnl-0!
z*<qETV|2NOlWVn<zP0P}Ii3wq+U9p@d)t`OC%zvxrNlV`Y!hi_y%C)G(g(`JeM~=m
zj8Sc35$7D~S6%MyR`NZESa?IO9klAHB8Hl_%~l_xG;G&6GM^FWam-xvqUBxYWMOTg
z=T|q&`j!i9Uc-3(V9qbf#Cim8vbhlN`~ea9@$-lk?qoENMXgwb&vF^-USIX$T(r?u
zA5yz=;0mT+HPcu_1oe$?tnJnIt8r0Z8-G!)O$3o#BSK%*`Y>MR5!7hch3RC7uqIEB
z9knk^_;vCfS534Hn4CitLtm0-lL(!H-S)NtKIRc~g<;kT>g&A$%$QO9KGFyNoF32d
zoJWG-nK07HHEZZ=BjkJI+w1D0Ja1%Z=q7;aO?#_O%-2W6@jSu<JTe9<@N8hJnO}xE
zgV36K5pk{($LgBNL@=~!F1iXy<!Zh{kcIkO5zCj|<VCgrLQCRGSQMnVuCHj)?Oe>5
zIzOu3X|&?B*34DJLM}d5ouKlba8>@7G`F!DgX_Q(s|C<M2VqyTQDk@zw4#v<B2rzX
zL8+OqyNcb;1IljN3R?<*C{j|yic(HMUGHyv%*qFP-!=r!d5rmtk%+;my*f#(3Am4{
zAp|g=KBmX;@aKF|y>>;1swY|Cyw*btpyc;bzW6I^PeumOH|-sCH>4@Dbip{t<M}w}
z<wKSu^Y+DuSSNe<c*+`B5x(HzM97Q6^6KV&ffn%w<Gn*N`l>2ZlY6g7<mHdnjCZ1+
zTWGEcyqA};b!S|4M5PdNpW(|P<w`MvASQZ=4lcsRh5x3c1{ye;Uv-&fASI9MZwLK4
zxzkFtN7krDUC{JWOh;|EW|6iZJ0ha3s_lYkka7Wz`q4RONxAj_r7vo9_0KkPH0Sep
zehzGba=l}1>A&V&!pT7pc|6B)jqDxn)R4tDxg4pHHN`imJyJ}^i)Hls%aV>X?OoN>
z-V#P%Z54Ady>m6BB|)qHpX)Q;N=|=zz46JfP+jtcczC=rtGDVm@rAKPyzZ(bBjwf0
z1%VC?y~^T?q5BH%^lJ-4AQ9mur&Oa;s#vq6bR;ffnIgRiAZwVZ6V<Azfqg-3CT2Uq
z8J-Rzp@H=BT#}dFC_uMbcSX3VQaG`CW5=4bs^nZ^tt*VB+sD?T=c|$7>!Df^a7}Um
z^Z0bp+&(tooQ(wip(Rsv<~(CO=NaM8c~s-QS9#&RDtImNia#}o5Rprw_VwsNS2x!6
z8W;!jn5WN+a(DTW<pRc~I?8Lg|3UTit@`|KN{c!WSKkt#MRHjn!w*FDfA|W7H#O2I
zQQXRxC8A+h&#z|8>NrtGEXTO0voEA@HJ_`|Q>x_c^`5Ibu<)71IeC@Q{ZjT>z+c&o
z^-KQStN>%()LOT<gSRz9skW6Stpfv3A6dY8GuqT3w6c1wY|TWBt~%56{De8|T~`Pa
zU!!JM@IGP;AIIV0an2*>c$I19qMm7O<ycCMp+r_UT_za?8S9tV#7tDX)@ev)3^Heq
zSCuB0NLKCpV*US>ZTy>GnKd}}hwwMctckVf_`5iC4N%WT;9LjVTJM!eErCi%EVPd8
zjA|9}D`SXP6mBJl3R(I=>vQv(EErien(G=D5vBg!{IhPMW<<0qjb61;+d-R>(CG-;
z8?qOp#tp+4Y1_1;sC={zgW;v|s9~5&DPCMz@*L9MJMw3ylY%$$EH`q_SzY43F!+<;
z@ZsY#M!>S0LK&)Rdv(-au3n^`ex@v3Gr9isLTyD&jJCMe#936&n)s(>D*5NI^dBI{
z|H>;~DDPJs+KI)<m#h7y?Ela6y6fq^TT-&3=4Ebw{%c>TP<_t4(EGZBYjCfy5S?h(
zr)OPIB$be<{E%z7myWJpqO~TirVK1qwCdWdsRFC_t1HNEJCzwB1{wA$<+-8q(QefW
zGKMpt56kDroU2?bm@<<b1X$vYoU1gCuv8^xJOgVGo_=w=MvU=1&Ut2WWv1s_WTRJR
zTn}IDs;^w-y0=GOa<FUgwAGT$)2U9GC{}Voz?nJo1q;^e|0~0hS9NinNp<MHxmJy+
zrQmgPm6^Vw_@WwrkqOsQmYpD%BUyrWa~&#WJy$I~zpmm3fB8Z_wJ4~spZ;jPBognr
zjhh=?>09wnRt~79qNaK&5m>YO5lcf-SwmVK@NQK5_7v%>mH)}NPjbG=*Z|2XqaoM{
zU;OJK$zW#sSuEa~_*)NX`t$Mp^c*h=--14^zjzI6c~<eyi?FmzN3OkIPuR3=oe@aJ
zay3?=QpN>z#U&~8M!}SSW;m~KSpIZ~t&Q;~*H9Q&fJikCzYbYeGyY4|>gt#Jr>|Pl
z0GO9gSiUx2mLl(X(b5rkySun}?knd`DL_>Vi2SQt<_g9ImZp`d(QUW(#@ka-TVmqI
zI`x(|*D$KxbBFAs4;VbXkzP|nuyp1s#f~@6{bm2|9+!fIDa4U;(3QiV$ET-XbhCAy
z*y6rj3-m^tU1KBk>*RXbji|P*D?MiBGjq7V*71wubG_qzwez1Cj#TRtZ}sXw9WNfL
z@j5v##U#^Kjv+EuH{8W9dwak9vw*6;`fB7Tqs-Y@0dt_KulL7w`uH^p^R0&PvXU!S
zC!+jJEG9Eje3FI5m#d#bl~}M6my=y(SF>BIrpTcecc#q2u90N|#K}`y$3B4YK71Yx
z!&i$!w~AY=+KAuwkiCFZ7C%`V)5$n;o?!Tx2X+ZvEbM!^Kds}XAhs+eHn)rrC}iL&
zs&L^g;OtQI5Ime*0&_q5#+q*a!}|YY!}-C%|Mfbj``ZU!2CiXcvibH|Z&KPh8}aVO
zR;y9FyyEp??@8lj9#;c;ak=Dz6eS>IUZ?!>GMOOqI*BaGZb(_7;quz;i{f7FQO%2r
zg`%<>4b#3^D+#mo)33vLM7mY6xvVvEw%KNG@WF^@_{=#jLP2rO2JE||Un#B6MUgL!
zx6*V_O3vmet?Gc%*le1sWj8>3k!2NKvUvbH6DDoEJ;r49)GBJ-JcucLLge%jnaN+L
zzpLNv{YU@gsH)G1KQXxe@k>`5JXe)iwOT1(rh1v*T)eQ00CUx=QgxwRSHIjB=6n$)
zF6Z~fwzz~0uT7Hj<sa8KL|kFxMb=u9m=z(uny=EOnyqcM&7iWIN#CqN*k)Q^m+f3v
zxC;55vN|Ot)R~sj<uizMID(xcE82)98Gmlhm}E^BT%5ufS(sOQ%922hJcgIz!1HS5
z7H>0@NUf&kWwJ#%y40scQm!Mua;nTLOw~mZ@{7J6k@1f}6Ztv?|1E}dt<a11^N&v#
z(Fy=ea@0VfCEe+ibSVO<5lLyc>vb6YIbvPc^CN`&Ribz?kA65c=U09F>yqT%<q>Fl
zT!&eESx)8iDQl`k4T><k*c3PI)~%cNc5xnSs$v{5Yq|_@Qz=pM*Mr;Kbu@T9j@Qte
ze!>r)cl4tot{n56sL0!m&jia&XT&+r)qx$Eu^5?2^_<kdE-^_T*V;!l^1Lek$(Jx0
zB7{hg9&shT!aoZh{{DCElDq$B?^<dO@8{l+-=o${w{q^T{3&x25hYSy6xwPRUr$lw
z;?;clrMMEK#iM8|OjdLirhYg(i%lhYNmsxt)Dd}&rnx-dMTuD5wCg#q!EPx<&`eOZ
zkA|sqlUCKs7nm)2&q+KuOUNP87_}9g9zCd{qCY)+`i0{!7O^gzc3eYC3O>zb>Qw;>
zaQ8`dMa>hXsw|ly@HIgPHzzVw`0^RA()cTcShDwdK|5T>J^@qpRZ(82!M~Qh{9DuU
z_`6)_J;7$tVc*!^bx+}$^1@(?Vs!<pAzyioEHFneO5-aeSABK4)n82c%D3MwM?+Nq
z?z+D+j5oV!tMXhIwJN4mk7kt-lou@}a?r-&^+`g;b63EeZPw{?T?Nie=FTioDSXC!
zL4LBTDpLB6FP{!XNef;JXOPo_c-EVTznFKgq`O+4f~ww^@XGscPDhqhjILg`)lyi|
zS-igb>oZ@(N~;*@pBm0TG_7}b<;pp}{`QNzQtP&`_be$}E`^2+NJ)vS3&v}?gt(gM
zF2bU$Q%I1^s}d(orP)PA4lf!0i}vtkyBSKWty0Jj*A}}9jbF8UTAjUBkLQ~4fS6RZ
z52^~A$4AAcIdrz!diBbJ2dy*U7=DJ*e2r=I=>+W>UWSjH5$A>~p?z3aq6A0L_pP^d
zylP~!bX6jr3=d*>D4xY>S^U}>tK;esiD(K-ocgk@Kb<0rby=GN*SHwQFH;N+z1HAT
zDSX@2e`PrT@b!ajtS@~}l-~bzwkyV1e83CBsS7C=%PYoMCr|$HU6SQcrG!PQqd=tD
z#b>AkNmn~o$lJgP^x}(^W4T<u@G9Cbq(m>O$@M;4RCTOc0hKt51A%_-sGy?+;aaZ~
zXU5s)LCo;OgE)uc0mG*g1N3MH#_*gG;}SR0-J0R;)r?gOCBR;@ApJh#+CO^Z#fr_c
zoBCZVF<wpc>kvpmwTeG)Hd7?9<+6cxK{0++Yc8|l@;MqyPxd{`%)iR1|JhOXb{4I?
z<sW;mAlK;Lj5h<(_r98li-ne#mn;(qeZewsoGf3gI{WKyEZ4e}mfqgGU`|k^t!C}l
zeZ1&XYv@|l?=C7wX#_)D#S=n9C!Gi%Y+<_sgK@s1r8Q^g3JT(kaJ}W|7M`A<eN?t2
zyEf~R*EI=}s;$>dpbZ$Ax96xjjR-fj?nCDp>G7qT8HSq3qV&HA>md21-m=zg<zUNb
zWHkaTImWBw<88r3Ll@BSe(L?j<Nr6sIHr=oKR)&3S}fkX5^{-oeGx4ZS^j4kOccl%
zU)+HgoT!lt^Qo~uFVk8F^IOz+X>Cs4bE_<B!&jpqb?RA-eXgi^HMo+ZFS`NB+0at4
z2`#K8IP-Qi!!NHeV7Ip`E4vxMQ}%JhSSVY+!U{zZy%*go7ACFIZCSi!j9U+U7QJAj
z5Xch<F^bTnD#m+!n3wm+KiN$+lOPoLO%~eh9<$b>t}9=?0|{tFPJgl1{RMe+J^xLH
zqyNh91%3I7A9ncPpKM7)gIdixHnh;wm)m$%?1Cf*(uQ7D%J*v$sS2=&Dj`ewc3sic
zMfN86RP_}#crQv{ZCh&SC3VxQeMIWOsyPm}XdJA`e=j1YEy-pBCNT-T+mSOAjYTo7
zVrpOQGYzR!UbfdcvStpp2#*}$FtsXRy@p{qA(&oVEJ}HC(FPWuV7?sT`pzs`x4wuY
zk&F^acta)fb);M{=wGk0tdeMN+vDG4ILk>i{sFM~wV=vBb4=?2EhU>-=62y|e=>}^
z)&z34_v}knFn#?;)#6p>+a6T?Vj@-s7f{+>Z(32Md{K8)HG|@`X<4APv2&z!XnnOr
zjuJ!qBJQ_p&%M~=V<CRSUSXVc)y}vV5ks+7+4!&wOxqQkW1GP7thtm*&1bJORVfKp
zX?TV65nkgmHD8p=!fNay*+@NtdKPsY<OuSWpj8a~wxY9W?*C}zO0KJkZQI`eO@`xs
zxxRm;I#~#PuB)gd;P0<P<=Q10dhtogfY29FKE8V7)}d4dnY_wp^`<1R349?X+q23J
zQZg!+%6C)=)(MyQps-$MCF*!)s`Sy9(%EM~8~4{k3LkTz{eoZU7oW?(<CqCH!$)CS
z>5jE=@12I2ZYCr(gcH-ykY<(f2+~#Xog<2^;WRG^-d=6mbr`jE(3KkAl}4|7X(eB#
zgd)n<tlv;~d}%G@*Y0}l=Gkh+{>i3qND~#cW_xuo|H-cY`pr3N(C%M;<~6SzMQK%v
zY*N19PfXq#B}8UL@Nb`&RNLi`VUg;kmt<yZoYz$X&6P)9$OgFz+h5ex1q>C)TAg~n
zhS0o>=BwL(zS67!23}Or(MO3fjrkl1kJlr5Vr^5EHVh6Q!RjXzDlY^|AV-njtm`W}
z^UNS;Q^SRqn9s(M+6WE?+lfM`#wtdviDO9(MTmNyFHO6uxk=WBxy0=jxhCElb6$m>
zg{0)aa!x?EeXIA}RA0fdT&~3QAD%V-*CA7l&Rh)@ujwSHS<_8np3@gXnP2q73d8=&
zKwq?vT-|SrY@}vQXg-$$7K@b@8ia(#rRpMraFMx|Sm?slS74GZy4s0_MH0OnrM6|3
zF%=l$OT{)AKAtnY@K?LY>;0kDWLLqDHC(l#U3bvZQd)hgb?S5p&lrw?+`0&jC?S|s
zXEQSA2y-*G;{9#@T#rj5ed%rU{{L5^dSR_<Dhr+`HP$GPzkA@pMK1g2cGE?B-*1<z
zj~MM`E-&r}?H^dqU;gwW?WrurQ!_JEKl-iyVC3<Pqx?gXKKvp_l*HSQ)%thZ#|m&>
z|6Sh1>a@PtjJoVFDuI7Z-TESeeSz>r_WUAUyyg#=y=9s-1r8Fyq9mxE=l8|#kEl6O
z)#ud|C3Mikc`g!QwVnyAVZEaG0|s;~=XJd?9Hh6lnbM?fkb@c1kj;h*qq*;>@jNzy
z+mk+GRczPO`SK@x!BurDqJ3eTUfj38UsA6|q&L>gU%#ZObQLYhfP#8-e71`V^$IBh
zZR>lCe*@v_tHhzAW~#S#!}`9jp|5QsXYCe(2Y_eJ0+^-7Ma5MNZ>W&*ew)QkL%}p5
zJ;&u-79BJf)wnYG*|>aPG4lM4uBI=&g334Af^)1C)~bMbCRqv1<i|?s#n4D{aw4W%
zToUtr(@L@xNDsd%VZq~}*ti9lcG(+oBw&4|cWrB5Q<pH=Ia?n+lFZ}IWGLIo9uP7#
zu|=M^bn&kBTP{^izxH3n(ElKO@fQa6kL|{~b>ynS6C4Sj?MlR!--{)$7g^5a$CAZg
z#N00p9#z{^t!wY5`qpisTc%UV=<^Qlb)sgJ24hGrCnu}LSHFCoEIOC!n(G<v@kP2Q
zkqf;*oHxrcQQ}RuO;u{*pFq}lVWpLO>Pk(9l@2lU)yZ|WfSSDWuj;{iC1dAxvZ>v-
z(wOhB0AC>;i%@4+l!@xvxNF9oJ7Z#Sj&sf|KK#1k;w$$x$?d#uV+G!q?Z^e17Vgc9
zgzq?)Ncu9fA54N=kL@2>jj0OFq>G6N5i<(Ljy%>d`AhV?uBctSGRXF?L``3rxNX{h
z>UEmOP{u+?<2XlDZ0p<k)`*&MMH;UL7hVniZ@XeWU%TLoKYw|pny-$E*Q(X{X?s<{
zi_MSV*+_$COVtlT`$*|jU&*4*zhrZ2?%&I!YpxFH<$*FC^VJ}vURjZ@kBP;}v!^Lw
z{fSDWn{I9;NY=-LIL?^}oSPcVaou~*x0{U&wX-^@y!ymowG^8*GdznV=PRsyB~qEh
zk7c{JhxHdPX`*WS3Ic2JVnw>)+5V#cE(eITHN5AYit}Un3YLi1Rf29yFFYd0_$>W=
zCB`U0h9!z77IKC1P2vsiECBgcOaIf2{T-L64uw~T&8uWSdPWI}*=xdLrAzuHisr>`
zcx5?J4O2vlX9*pzBEmYhUY8NCxcp_+0oXZ2;TB&hTP0^Nt{~Pn#Zjc|kT%R#apV+;
z12ts?YRe|wF~b%JHHty5eG7=Enp>_ZRMuqIerp0gJox|6_9sh|B)PgE_5o0snR`TL
zR-L+ig%q|3SIrGK{r->AONunv<UK=W4DMzI1;7PV%{{Vms;q32Gek!Gzq<`ERSe(Y
z@euDdWsWx^4ZaL9el=$5$zF*3dcv7QUjpgz^X@S~VQ>a#cMi~g=F6suFlD0SoC<tb
z-amZ`hVtP6KH?g^=C@Is?$qeFF6ngyBYI)O#+ErENBAE;f*aM`xro5f;oLAjCZc4n
zfdqGVA6V6r%!^B`wvhi5yfPZC1!I=%5H3p)PlQjSE&vEmH1^(#2;xku8)&$896GK)
zU~lEEK{VJLfEv~VB#CVReXLJa?WM7bv`v}-kHakx+n(?;Fr7R5D@pks$9vq&_oHZe
zS+<>RXbAknZ@#3Qk{HZCPYe!z1H9mb%gPAEF`920tdui(X8^0SgHnDW>(1WtH=55h
zbN3l{*$emdlJpDC65<*c04NEVUN#FfbG!OKaZoF}Gpmv1Kqr>0=qPNfM)|2;d`y<Q
zBM>b61`Ny9;wmn$pA{;*TL8oOSHYkGrS<oFY#6@ZeyO_;OZU>nJq}>V3KFUlDGY?H
z4*=M^JG@^uVRcUUw%Yl1)ZaI<S%ren0HY%_t%(BBem-*M?*#+;Q=m|ti(gG`1aKN3
z6J|<I!MT*3^}KNcyN9!bf2jt{pO4V@tNpf5jtgCU?H|-$c>cNv?UI|Qg(BsJNbO*Y
zOm6_f+;h^H=3M`MK3|(uMoi=yjdn!Q3pgas=s;m>m;jhocYwiC2tzI{h>VD%F|E<z
z0x!syaSRH@k+9AvC@Zy&;E^36RPGD`RZS2+iSX`^haHjK++&zm{BrL5Ys~EI(|&rc
zH$&dJ;+t*^MEpkMP#E~Rz_}jO)R3GN)|J`w7+;qwHWX&|Ww+T){>^^n&qfaa()A9(
z^-+{g7L*7{i81kTWv&j8A}SMxXYi07?jAYCno78;yfJ5Y-+M*eOzR14OnD3DK-#!5
zU@-wjgU1&PRsxt>1x`lMWtfJNdLp9I!8WwzQMj!;5!LJ0R9K`^uqg)0!f}QNyl3pk
zWA9-Z9+wO08;0%u0TT1iP69^G+O8rE0MrMddE!T^s^a+d;>mx#om2ZuPVH*u;P=<m
zV0y5wr3zm1Ez?i-4@6G>a&u{bmWCJi2wo!Qg#kiSK{&)Y04OPZAIO{0-SO@oS>7BN
zQ;q2DDtI|!%H&>!LTAniUn~jC-bz5O>aoIFM+D;l7)vUjV~b1>LV|sKpBd~7Xi={c
zK#=l-X@ykW$VL$$0hqqI58nN7+xy<#y-K+N^ZTw^t&i(I#<%}nFImxxD)}8q_REtt
z-nN4lC`A81{|MjDp|_v@w(?xV=h54~9BHwGF9pH3FQwm|6aV6&U}CU91XKp=r7NrY
zVs`@-Ijaa7mb1GI7Y9a!`+))KMkW0b*~ivL5raoMGK;AvQf-ClJ&5R~u{s~A3*1NL
z5QWmbAWRHkkg^y@7F_hO%PAY;j2i0$7-ExGE|=ZG$}qF1n^_zl?w6!Y%>mvc<(pF6
z{^6br4|o%Ukc6n5OvsSd-`9`UXF>hDqvU$f@3JlR|A-2!1QF(CR&2d5Lx5^W*rm1p
z?K1gG3+L}Ygk@O!iLx?Sm;~GiE0LCIW<-J^OJUWy+(pN6E_j$9Z{lQ0#RSN)=FIul
zKQ=*M@SK3SO!o5>Eib!C)4D{3up}bG11r>JrnuRyq+W+|^@Z#o3O~JmY&W1Z{~X-|
z4RP4leq_9?_-5R^CxrRa`}+?MHA)y$UJ*4Mo&)~Ph!ib{_;B#AO)Tgi@77$Z<<cIf
zH)eAEjj{}tXdn->>cz4q=fA$be0!*<jwwTH+E`emEy7rpX<+Bp<DHZc9Og$@IsZ6d
z>3&SZ_5heCOnNZ4_(fWO^=pBIQ?%;eCk3QeYlnpxMANhqT1rB&0!&qN>sT+APNJJ%
zecoaA#3H;kg}~F%ffIHdSh)cNScK&nu*Y*Ij*0w6|1W-Pg8BBCtCRkQC=HanP~=al
z_=bcJIU)LGw9P+s{^tFYCc^QeZO2R(q<i}S(XFYl64IWwA7}Ffcv}<aXhh6Ui}-`F
zM9fUQtlI+gGr&t*SXfq8QOL#(mM2C~g*YN1fS(=_J&8O#DltHUa^`<;EAuF2sf{-B
zKi-bu+IbSMCl3LWg5U5wqY)a%^=J--OH`2B8DZ6lyb$fom$+I%$<TMxed`U0jxb=v
z%{XfSucEc*Z%)4Jn9=(#<G^#4Pzt4!BvIZa^<^{*#$Vnoe)j97iBfWY)7Ri&ieTwq
zKWxGzDl1cV*;#G?5a!lT7Gr<=x_>2VM6@l-cj!+Tv~9|yd{@y6X#+yoQA$z{z;J0y
z1rX80BOKGn`WKE`c_vijf@ofZfuy&_R{eXGR2P|#3P16Vgf$zwg|M?92{Hx1-N4F-
z9SAn+K*muMw8@(&5zqw8`ba>grs6ny{?0o~7jj8>LbzvESf4Q@*O_|*XzCniwAx9D
zFV`vHtp|`lAGXyQ^BxzT6W5mSlKrgQlnB)BqW8Lmc03O5vUYd#_*YUJYN)tjX*Z$;
z4=~TT(X|qi(+zOjpFV{zOSmAe^*B5c;e-73>!%Y57akchDELhZliurrn!;2aUV33A
z0wa?hG@Fqz)q5j$1UO6?M<l|O5qp?apQBHQLeY&iJ#rtNmFHfzW|+jN+FE{p!E@E!
z1=}3fnPo>;P5G%)O)D|Rco-;UI0mC!r>HR@Ghf5;UpNJ4LMbOqv>gKxT2|VIa(UWH
zIb2IKB+pG`?La?{vwL{>AB-JgBI0S)-6(iHNC8r2mIZ{&#Nm6w{qSI3WeQhc<i~ik
z1pck<a`><KK-+^#n&}lPS=BcWKIHf+p!$2BKo&w~z`#ln#z%oQuq|=Y%L^Crc(E*}
z=zxMi$h`-EejH{#6oaYy760Ie`pvxU=Od>Uhq#=Wd>GdjRc>S}#H#>(x!~`GZXS3{
zYD_G3i+=SWikRLcNJ^rz0}AFwM44gH&vZ!EAh;+&Sakic-ZoX{#tZS~<?}P&K)?9y
z!psC>A}LmMI*ABDgjhKn5*erOy+5o6QZ{KR2FbE!>f!0#14wsA5B$y}FNY@@I+s<Q
z7x1DiWJEw5oY|v9bYdu(cz80^{i;=BSUTo3TFT6_$;k$Ub?eR%NAKMw37HB;_}&A^
ze!j)cIK4!a2nRPI0HjR6lIlnsx~xf{5Ljj+i(p)RP@-CcNS6izzkDlcb{0CSiX@8P
z>i?5~;6u1)Ok&6R2@4T{BqOs&W~RG!>kvkIax1bUXHZB|f;fNf8z}LMk;6hN!eC|+
z^mM6~0Iw`71)23;Hco{)k&P8x0yGqXh)nO1k-hgQPXF(XoQOoyWx7O@Y(Pa=7BH3A
z0@Bk{a4lCg9-&OiQgJ6~6B9y2iHy8{G!Q+UKtzTNPzV9pB{}zwME3J=@9|S%*NWuw
z<`WUBRpI9!qcTtHcS4f;8x;*8O!4CcQ4<iMTYKG!a$+8#5*$%M^eYFC3MwY93Zn}T
z23!bKHE-bjv>_3q)#0Cj2M1W=-BSLiI8w&1%vzcVlMcx6_&Wf+>B@kFTZveOo&arN
zrlj7>k=Frk78w~C9=|21`!lhF3r|wz>Hw{dxM9ku*(ZX-bikT(x*^m(cCAjknjogI
zO?`U#Az8t*8$-{)$UP%Iue$~uX6W86uBtl+!{Z;haKE?+wDE<P_aX**4h^7*a{&UG
zbYMxOJzlVioJax`mFXp=F7gYbELYIGhDG^qmZU(zJ=5XPaFzqy7EeLPP{SN|{P3{o
zKTVPA1^hYr2~7SxF^jM?ZfM|zX=3j;`6vp1qfqe1Dh%szn3=g0F0LcJ;^q&GKB%m(
z->gI4>-4<-YUDTtf=vfphF3`VaI^xT;g<<6y*zC)J)j9&aaM*CC5UJNAl$-UL&TxC
z^;)5al6^z|O5BwGL%gUWW4wStMI5)BQ6fTzBO=a!%pm4gc)k3>l)sd`D(KDq_kw*F
zrjvBY%{OHsaE*+>BviAQ<6&RF_U@70f7(aJq(}baN<fWhnEeU~FU*U!jhAU|wQj@y
zq+oPz4pC8#eec=L&5CtznQk5)`A6cWD3Y%n>}3r)F&IEuZeeJLVk*)!o>RNaX1d;U
zci<8Xs;tKy1f^D!Ze~FSu{0LIBVePpNEF`T_2~YIHFS1x`9-K1ow?EhrQ(oA;VKY;
z{0AV{C5m)?s}JMTpF2$dF>a;}YOuMWAz8Kh#|u|K)LRHV4)^^W+vzX#9>4j^xlR!`
zS%xqbWaUj|m37@<y$1jltv$PyP2pxxX^eBjmusRaHlS?wQ>g0iKY}MGg)c>?^p|oi
zs=^=^9jZvgkq}nC(5dCIdf~!`{#gRl0yRHRGPOaAWC2J*SvWi*EF<37>$TtHuR$4p
z?#RhuyZ&^wn_sKP@F*Ko;RgK)3J=qo_x1FCrL?-_{J+^%a#$?8FTJhnrV1EVrlHEw
z0pRf6ytxY`$mWR_g5w{iD%G&5qF-24R~DcrcvHDi+vHaK6AyqLZ@Y$26KTl4Jkb!7
zBdexISXrq4!R{k7idt9@<B=p>=T2Y`D1nq?D(*5t8wbP2fV~t@YfEE)hdywC#LI?$
z{H41=)LRXZ5SAs$Q@8uNihnv`V}21h=6Ao{d>Ui_i6mh3ruXCc=3ATwE9swEzGvLf
zC>dg$L!1ksO<7t|u&vKi^E`S;tQtMMV|<@5%e#8O2qb1=o!k#FFO3%-;P0lQyy%Sr
zAZT{KNK{;X{}Do`JJ9%GiMJfIzU7biAOjrYf=fa9{ofb>$;!-#7KS7iW(EiAnVCkT
zLs&m<!%#IRo2YbW$FOUBwnIBs7@}KxznZ(nSA8WY|I+adY+)Kft8rN^C0e_G5I)9j
zlWHlrLxDkt{1!_0+YwZ_W}eDMII%aO_4<J}SYSiJN1bB=5c(Sm^_znk#L80AB;3Hv
zV-~(LHBnw}H!7kv#}FAa6OGXj4L^ZA_t)ZTDEZimp5NWS`R1seox_I~9I}w;*Nl(}
z3T6gF5zZWjR__=GFJ~g(^3OOXtUTY)I#;&JL>vqUrD@Ol5+9thsFyv0f4hnKbH|uh
z3<>`$@P?>PVi?G+sOScweYbMmLMoC|W#WaS#;eH#v|n3;CWdqxMCHrO%JAu2bYXy(
zV|jh1RUOJvh><ap4VV6n@Z`J#uDGeIVHO38vH(pO<$M<2(ZKdH43!(;5sB^y-whFl
z48X8wMp#aB&Oa78bFTQ_l_dtTkAa~M%cD6$T*pljh+{cpKoz-oZduMO!WXbCUwa2J
zgChljH0_hjJo4bV*8R5X<8Q>zHxQdw+FGqm@kp+VZ3-3U%WN65tV+AYkN~O|MZyDU
za|@Mk!aMN-V$a2`m}A)casVv4hA?V<wm1NRM0ccj_mUy`2HrPG6rk!=)(Hui7vV<A
zd_iq@MxM1!r+9&jrESvC4c<N6#*Um3<~Z`A;QzKM((j3(p)1JcP&u?%;F8`3W`IL#
z)LHXQ$CT#7upvtz!UK{c+B^r`u|6~^Wvw-B<UCQ0SENT#Sbht#{hJr^&F#Ndju>|5
zfK_|2{kX>2U#I_`SG|2kBW7r6B$U?3aNivhl<S9sCv|y3EC`4m0^tC#B@khZ0~rAU
zx9Hv@BJ!s)Duc9De-xpqX~(N5v4TMYs=+cIRW=4YWdjmIP*Ld(@Bz&vaK%lf=qOF-
z{MYbf*THk<n>1y(3qUJCz=dUE4_@zAPFJ^Y^O6acPOpEfo@$N;;04O{NW_#^E=7lL
zm%e}Lg&2J7AvF^m3=g+q*V|4<A%NnSvAn~T84yy?B8cqnfl;$@qncF|=3vO1&}+sg
z5y38MRegqCNd*H?*MiG{uCko?lCLXKWa3r`30c>LdBn}CERc31#F1%Gz7$f&cMB4t
zrZsGZ8-<taZNT%7_i(;FdMlINsc{YfA_l-)wf;)*5$qUnULQcn4A@MJimxR!<`cu-
z_QE(!z-k2zP=0S?GMX;|g89(+k8QwpE{+Di@|BvXIeJD%5(iq3UtQe_oiP)A09ZuU
zIHa392W6DlfCsmjpn}9uIII=4JjEDT{Qx-5kvc9-VQ~yhYtTTe%z&z>7h;rOVJS=K
zrZ+_bB)aLX!O_}+gP(}Rq^xHG$z8<Y*&)Q;inKXzcJ%Jyk-ser^Yb_N+rdzjNDaS)
z(?y_$IZK~q>;MkGfc!bAy4Br{@p@KG?~Gz3zkkzY00y{Zq+jC}0>97<#QX=l43$et
zf{BT#tN}gzU5JyTdB(4`i;C5cQEY)&I}PiogpJkzz5C2WfJbu?YE)!Mpf0w9k0@<5
zkhX1IVCQ=5^^4vCf?TFQ09aZh%y?k1is(%@K?0;VZP&6++jHH8sFl}THC$sKNq^Af
zI`a~6m|s}P_D8aU3rZ_5+G4)!Mo`$}@vPGwa78*nf|*s|BP!<PF8c&ff{>J6w2}0p
z>0MOisR=wns)SQ$4k#(xweRxpxQw4Wl$sLKJ4QAppe_3e%Rum$Y;iOF)fFcgL&s)x
zM|>G9&j<rhTA1kroOGq(^VK79f$$u@5qb+ItWxef;ATt!v-qs8+U$P|NJ=1f84Eox
zU|Dsmlh+%mctrM|o^*TOu?UN+jVs#W!US!E)*B)n(wQMK+H~><7jw+_O!%#>o(M!6
z{IVR~jdUrTv+23iiaa*2s;0a|bKmNx3}lv-ClI|~_jCF#TqRV_f!K)MdgaEf0093J
zISw@$&O)N3lqVvPJ#_G#WcJ+07Sw(t4fmJfK@>skIGCA=^+N=DUrD*Z*^Gc-)Qkca
zg*za+O0DLr+{UyfdOr8P@8KXHal_-cSeZo_!n`ok7-5RuS%p=H>a-0XBG88f8WDrq
zE_vzn!2vB~(u*Jy5I6@%{Gr%+>vLQ+LWPb`Y*9c2QFkGU*fq}iTKT1KN0c&GjCsTm
z_R1kqGQs0>U#Q5WL{N>4F6R8CN4)CM-y1e)Qa-9Vu7-$<Dc(|~1|G9FHM=vvez*@w
z-HZS;A>lTei5CL{2!WIW9n&=*&5}6PTHtsarcTPM0OkO?KR!P#-~k@SUMwWC^1?G_
zdEr$qXD(4`+LqhB0>?h12ZSJs?upi5YTHl;7KI=K=7_V;s$)L2@bEt&H3lXd34>S1
zQ9M!!<S7CqLQF4fjyGyL4_4@#0YYSqT(9Cr7?E$8ZW1%7>Kq_5zs-VQ)1QChooJKk
zOo<@5Y*`Y0=u-_PntCnRe^+EqW8K-;jhGmY0yvDMM<z#&XS)$h+PF5QdJU#kx+4zX
z!`y~-4~V5Dh+9)`+q%hHL-(~`6RR`<=~6^l&szqV8wt@mKH4-CU~3Q-xRE1p8Uo&Z
zKT6`!{ZELSX`6bT-o(Szts14aRBdeHK(!@abtPIarRu@*OE04Z8aEnEO8Sz)h$#Iz
ze(%#!)=FPvvny!+xs|jx7h?!A0(RC*NyyUSnDl0ps~RBo9ymX5KqS{PJpN}0$<}5b
z94rJbe8YUsDgX$cAeKfefWu35jR9^NK*h}g0HV0W1X&kmY`6P@(*nSe)vE9SLnMsg
z=w};Aaeo%RN$5sR6=wSZ<3b;T8uEm-j<ZCkeeX84R{Re$a^{g9L;r%uCQ`KHr0aDb
z5L1Mt`y9PaigA`EHke)_HY456;7FdTLPMp95yfAja(Eoe!fXV*_29<1==U<J62VSM
zQLYpoT5rGS72$>dn3@3J?1ew?s>cjut7vi8;ukWJ$a~T-Lk0pYjhEI2{saxW0lr$1
zk^xAiG1zu-pf!jJCV*~XrY0&IK^Bj+_2Bsi8UpOkuV7iNe^?3DAEy@qaXg@pr?!43
z5Mlu#^u^FyFHS!8ewO%%r2pX|)i=k$rk5BWvGC=$b#3D%-T?qiH5(x?JThm02lNU8
zK}aMGH;yUF^Ah^26spv?#&DB}StzaC)q%MR{uAt2F94`#qBExpMA)w);ML1R{nT$7
z*8h~{z|{aJng7Xfp$N`q*C>-1Koj6#K#N+WFFJrg;11vBXK3jQ%~Mm+MqJ1NLF<_g
zL5Js2u&5^cm-|*bAR)nt-kw-G761)Mptlu`fb%PyT?EmABRpab2mfe;?AznjcY4i(
z$I9|Hb59__NVr5<CSJ{ArfUhRo7Zwk$c-TIv7iNBh3kd*LrpHqV9_E*?>?ISB7pLj
zBWI*&^Q=gsT>E6&9FK7*{f^}Lo2rW%jrlh@&6u7_x29*KOki6#to^bF=tDk@p)}<1
zlpLFZb*X<!D^bR3j_MR6p}~C|izobW0HP>x^hrn^(gNqw!yh1R_aUFqK&>gBz^9gx
zn%oS9{(~3$4)o6ytiOH`O-luOE$Q-Fk*{*W1<br4xu5(>-(*7N2_P9Nv%zHxT+I7#
zzzPCEc8Y;~uO9`fBQT&~_@}TjJ~@KSOpNr|1OR5wG92LyN&4Lmd6nK_s7=3(TQHHW
zox5}afZa^Lit-S4qXXs1Ex^n5gq@+w?M6VJE&!S;fmum_est>yq+8F3y(X^R0;<4@
zF0YAdrpKvP|I8xm3ZT7TTJ&&-gza@_-QP`a{;a?myUDkn6mvFvtth2$wlL)iu1U)j
zk{>*WdV6wib`h2E#L$k)3%&zt!@y!_S$Jd=QX{itdd~muW2}DyOfi3s>Qh}4)e@Wv
z{yi&B5itiv%O<i^z|_WG?wNh`m>WRsV$2H@2~--b{!GlwCS!-NNdY1CMv8F8RF39F
zZ*WU6k#zUtusw<yTfwWt)#tc@fhI)c(E~<6_NrQ>39IlT=VI{l)3JA0h8-8FyG#x8
zcL>q{=FN^*7BfE`7@aIYT!@7I(xD;Jq4-3kwnPoU6EZi>H`UcCbUEe}*JpeKil202
zM@}<_>|?x{|J0Z^H~;{E07*naRNTuk<RPX|6zV$iic_RT4(@9Fj=;IB!Lk~zR5dS+
zbS<)2K<+4HD2uR2XF`HEUJmA9FaUQzh`+7f6yb(0&83pNt^hJTf#B?;sRxhV<FNh9
z*JJN{gk`V@Nse(mA8YurRWFLNmJbhy06T<f<JO+#W7eN`20Y+?Bm`dK^o7g)vo@+%
zXRdCFYt5air#XdfxsY9$_?9t?@=SOQh7*{DIZuu6VW&8$NcD=&0qKZWswWY{<?9cv
zrqghC;O&R>#U)fKml~lE%P+1t1K(Ro-)wA-8Qp|28v;bb&!cW$(TAN_MF2~DRW6qu
zRhRV;WTDt<g*Bi9eq@5OW5KbW&-Adf`)-dfUmw=rrb<{EiwZ!B1K2e~Df{VXrlY8Y
zKu6QuCYLb6i$UQ%GQlo^l4kU8)f)fUF@OMtg&BRU+3unas${q-Bw1hixJyMl51QV{
zgb4|bxy{m8xUS~l&|0SJj=$w-Whe7bjTyeZlJN)!bY3POxmG$a_z(-Y^rFYq2VjuG
z!V;hcS(n+q%Go3%-maF``^iz(j`v#j6LGK(HD?jGk{7~GE5HOndO3#y2<#3~q3qJ=
z*<<g!?N8s2hj*_*FaQ?S+lLX!%AAMGDY_8Sck9Dh`IUPOa&ykOxi%AVMj`b7*tFp%
z2V;^~=X8L;e2t%22*HtwH?@)gxa6%7mlS_GVazHfs!kN<SQ?LB;K<GBEppx-dj8bt
z^q-BJ05~r{(;V^RG;MIeqSdZb{f~wfpmhL*)K?F%uFFkjRE-gSMwkP42D&-0G!&|4
zDAq@YS9MGLi|8RiZ+xtU0dTY$*8}EmW#+E-1?K+ry`L8S^cR;~m$tR7B%p`7QP%jc
z0YJ})!E6g~tvw4MCB^0!i%tHF#?3d!7}&k$lT0+?W`-3IKLt|<uBq?^x&UUgji{&)
zQqHfdptMkCt`T$1c?o~b)A@rLzf5XZ#1Mk!7ay9-544OrmbQwtMYOFyfQG|1fPi7;
zZQZ0nWs`bDR%U;~0y`2VnCce8ff8oL$3{``!d@bJ+G^+&4r)MRHh^B+rg=|`!$$|%
z9OedqNMfP#TitG(+^DG%+yjmdWLPK+28Cs>IA;{^3O0FgAn}0pj9TB$k}^gH@xKJn
z9d~SU*j@+ARU!>;KIIFV%M;f%4@#a}{tU)oX)LP7<!2oLm-0u{=fkoPqwyX$e-O$@
zt(pm3CKfN*Po9xT9AJ@l=S|wu2>8%`_)fW`kLsKO-E>uL%xfzdKhH!Duh-^$1{Mag
z$LamF0Nmn$GJ{%aHaq;NWk`4$I~|Dtw(DK&&1X*Hdv_0qip}g8;AMgE_+@PCCM#3Z
zhV$^<tn~~840=Y1Qo(WqqoHV`wG_r^BYMwn8R2fxy%#;0_z%CL@5k&=a3!(a%6h;9
znv<MElEhQwX*MM4&obm^etIV5Ga=Br2S;{~T|EV04z*zt5qwR?t{n8AyRM3x!NhE(
z7xCUw5Gl@7Yqt+=V`&0UwC~o9X<Ig3O39+B-rN1AqO7vgBCv8}>+TN0@%1Zi3ixUL
z?2mKzFb9~{lJ+LJMbxtJe)=Dag)p=63^Ol1p|BmVcoCg*_@U_wchI`3w#KwIJkKv;
zyY;ccpJ5&YC9-d5v~Aj8hdJQlLk`q4tkA~M`|!B_*Fy#Xg5NZ!n0TrydjKx=zf`SP
ze#Omz*~jn_ZImhBie-dD4aB-i6NTU1`pJ)N_rle3`cEjRIop2xt#8h6gw1R(voK&Z
zs^YgFAe0!LY<Im$A$T2a{WlQlw&_h3Q2p-1qMNQw8%TLo0n87qjCfk2pN|8=jBee|
z^YQt)n*t3@xhaGw``J4t$RvAsPf%+}vttBQ&p2{g9XJ3iO;)CLy|r55SY>I-AVg0n
z-J60StDe9|Jiu%n>rt!$uW5#C`$pTQ-%)QQXE+7`zhD#*=H8?CGCvID^^X1vN9i-5
zN}u7kg11tKP!I{$xAlfq>nR7&S^!3KZ@+-W0cv2qZ480vr~n_^K9JggMbT~t*F5JC
zC4OPWq4p0-ui(q`R+wrpWKyCWoGvb_BxVs_X<e90!Dny(_VEV*s(RC#uFH2HKCDXM
zm6`}QD)KJS$Sk|<KN=0ruxe`yQ`-zveu4zlP*MYn)Xc$rN<{X<?dcDn%5NCZLebl{
z=pss_w+{{3dqBC}R7Dm7v&gy%{Hq5HckgXYE5B18pxhF2*tbI<H<spn;+${pwYa2r
z^T_m=9%lb~J$`9CXO&X}09?}NE{Wd2!$kt97KHNB5Qu5m8y7qo0B31aDS9f_EAN%)
z%A`AoLEw;(Ae5!Em#o!DpnhS_K#lQ@|CnC10<U@s7}a%!Fab<N%B)LUSXLG(ZlSrh
zk3alPkzl=jln)<ni|D$wU*01l{U`!=UCn}i0@HJSoY3o!EQ$_~^fWMf3oRi`x<RF&
zTB1bMpa7Oz`yP~~RKLlpy4=@QwpD<$fBnxd9q09|J*$i7R?bigaZRK(0EwH1+fObd
z%)JApx%dYpaeg6kMs`!-lY|_=4TeiHO+r{$lw#VfqOQ9HsCmiTP|l&Z0z}8HQ_Z2=
z#aXf8&=~Vm4xr5NID(K_EtsFrkw<pkrZ;)9VKEpb*XzZA%iIH*R#suXFWOj?2@KC<
zAe#Q|wlsk7Q9gXQjd^puGbO_zD;#*l`Q<R{Pk+Q{ZquN$X>TN2CcQakr{3{aCX6;@
zxc8b?w8p$}gRrVLAo~u0S?`+y-a51`ysh`|{-&Fj-SXq}2jK7fxu3WnJ;O2#h%`q8
zJo8*!z3)6Fn_4H?3(OESa8d3ZUUlD}6*mPQNdhNj4UOu*i-=GyPq>$bQ53}Oxz+uc
zw+5^zzqDD{BTUiRpBt%9QZXq%TAqk1R&e=+fSCrrs=Ql&CT@5FmWzi}t78dm$lJ1f
zaWJGxV&P5KWfg8R^>7{Vw%zWk%c$$loBL_FA&OXi9M1%=fS-0A=QJrHxOdBaz8DbI
zWoeBDIjq6^(G5VBL`^=k@6p3S5~4M@=BDlb;k!ivJ0Dg{XM+&Z0ok_OcK_k7%0xiA
zZeP%Hf1Vv*9Wgc0-D6HH59|PMdkB2Y>9)Ow8-^J2#bOum=+B!sjAujF@HbzPgQ4V4
zLnsYEM|+H&OWvLsB_hUSOCU@bjD0l?ARM&9aettXDB>LKrFt%;{(|WI!?_J`@{kLm
zHjO)3h>%iT90Ci7q%iBOIvmvG9iTxFINHR_C;C@>_!16%-T;pi3V(jJF93jU*{uUY
zq(TinpZ)LzwFLt3vb1$uim=sKfQWo_KOO+m&puY42GXXjIvlt4hwp#*Ubcq**lmnv
zVz~d2-fz0S%5&JhTcSU{hMmuS??7CTw@NbTgb8eg11`M*l$wU{4nHH#en4C;+3~Fh
z_rH2DiWfA?;&|8T0)!GSgItJ2b{@?7tMhhEUgzOL6)rf|VSc6x#FzWyb{dEUF(9ew
z0ET|DJLNB}q^~O$VIDInA)!PU1Ph3!^^P=FQK&LC1rQwP7^f87us#0w|J#j#^9<<u
zDCexD0dNQbUmwrsz?=Gcc7TMasSx_nI{-6oXaMPY6LCKXyo#>Gh<twfcRY^j*KAqq
zfZjf|;@s%%hadiWS7k&#YppWxd0?$ntu4r;TC)0ae%T-Y{{Q{)^H&QqgLIgOyG4Mn
zjoSf4sYf$VWGX9+Ww?VK5^8UFr9b1Kc1fm=No9K_49K|lA1Tw%1YV%F8C{2mCB6PS
zkqER(pbrR|s)*y`7YT)Pn0NQWSBVdLdGaq2a=$VB5MkD#8N)2dY+Zz91rKgW6IP+C
z4lLBcGSs6o+`aw({`>#yc2l?kqIZS$;u;VVP~ds&0W57nKhGl^OjK@iS&Kv;vLfiu
zkJot$@rnaCY|rnlhaaWOdv>(t{_*3YV~Tf|Hd=htIsjEW*5kU`IKDvO==pru<MI6Q
z@1GuD%n_a*ha22FfQE))Ip^Ul7%Fly1-lP`7#gbKfSUZ7%MG<5u@Dnrffpn&0l*Xx
zkd*J0ILXvz_G^*JEUxO@Y$zD;0}^b&noR~5;D!gl;q3|3$Br1CS_i-IM!Z#1pouvr
z5oI8ZLE09EnVJvFkqjV|oe_YndM=QX`!IjD$A4)PE`W`ymep(hB2iN+chDy{!1^2M
zn^cHewYPLx7HKX23x3)O+<zE-K65}D(4n-6z++F)rm7brt}j5C0lhd}z!A^ci0H%E
z;$iuGJkDo-JieG4I{ff(n1w<4C%4G4@yZBrqotp%3qM(tLyII24*7!&&YyiEuO9uS
z?SiUes1$QxBja?QOvq{QJf5V7!1eWnr+Ta(U}g}gDKk7ka%;yPrj927x<H0S@wWVp
z(QQp(2G*HGlv#ipQ8FN9brp<&_|O_+<cb3f@o*2f2y3}p?e{7xqrriOt93|4Ca?^+
z^$Z*sDr$hY?2Twy*YDR`TNgTE&ts3CKmPrP?{ozKKa2K9q<|{&gPnjY(DyGVXt}K`
zU>zuEtpK)1BD|m8Gkc6nJ-h#yW{;=4Ss=^}m*8Z~QkPY&O++;}ZZb8{EJ%m>DF?uR
z46^d`hrgQs0R+F{tSDbSSVE>LPd-OQSgnr<D9Cdl0YJoo5`jH%R|X05g39YvfL%=x
zwKOv@&`loUzYZ2vmeLVnsl`fIbs%jdh~w-A5S&Y%a99NZ=a$E67Don-1fd<YhP$2M
zy5lk>)2fHPmSRE8pA!TaAFGVOZ9>bsD7U6cv>s-To_IbUKQ62c9B87dMEbgcz2gMV
z{XmoRgrEIPlvlFe8qy9o+s|IDJ8mpWIQwZm4m+*lX6y%XHDm!`T^AtW>;zC#MQA~(
z6u>S>fO`Px_FEdx|6JtoD}$gx0KPOP=HC})f+4D<FG`@4p;Y0lHtd;4j#}fi>Pjmw
zx=aoupncil2B~54C=kN@*T<Z~EV_JyU|-t0y#Cuy1k}djQ%h;#vWNnF0Pr}wQ{!k4
z0B*e}49FS|V74Fuu$1Q%05l{)k^U0dY|Ich0t+1u%~0!rKe<_W&Mu@1;133(pBZtO
zJq>hg@UtJ*D>GKKFG$<7o0;{||7qHY5Z$bwW)?@`Dp*4REm@NUZlgC2bP2=0tttR^
zvkv-pa^vp(&k#3x9dQYV0*DNrw{NiEt9U7{qNt84efnL$txz6G5WL*OiaU?$IJ=`e
zGb}A&qb{syi{FeJVI~!ATTRo-1BEMY;!-gH4^wS+$YpO5oL0oBIKz>>MLQ6TZi(Xg
z)Hh3wdUN1S%JGgB%)n472rybKO+~4^?--|pm=^$OwBQ7=-j}5*q5DuiJJ8|Betzko
zFAcq)-3sPJfk&QUCwit26LSL;r5Cf?-p>GfDi~l<e(}RA@L<sEYCB-v`Y=%F5$*{@
zL`3&Lr}_6LMa|b=&<Lcf-B!$K;=$yI;iQ-tcm)8wobisUfG3E+yfg)H^7Dp~61l@7
z-4W)%tl=~AmtYmtCf-zJk&O$T2_VjbE<|8tz(vqjLJzO?xx%?lYWQG_g9FW`Utj6u
z6*Q4&k~07bB1A0xSVb5~8ojmV0i0k3M3B!X6zIK!7R7Sch8uYO?xU=P$5TpP&Y}Zm
zkLU5|JwQk|vjhn%3Hs5^Yoo6hY2+~iz`LDMxP7a-Ndes6?odZIfNo0KwgoxMK;X2e
zzmy~YwEX5m=8oW)^UWbc`*z4EREm#aa4VHs8C)r3sX>gzeadwRa~PTOzN88Zcy;VO
zN?pJ`;XdS&v;TsA0yT{HOoU`(s0f%Bz?#W8QCbkC+gNMo;;(sWc;yj%E``|?fYI_q
zEvE1y;RB7p_~j$uvSx39=p6~xv2__Ci_j{|ZI!lwH@(*{*#R_Ibj#y-9?xzORujiF
z%pqWr6?-@meqOJ0#*qL)N9B!-1c6DJCMSKMw}#2rqN_9lGR@B24GiFkXP=3u|Et8`
z>zF9?;T=b%ZT%BB!4Xh}kE(=sYEF6uO~5%E>NNV@1Ik`NoHzy44ohuNQr)*NwdC(U
zUv-)ie8s=|=IRE$O>}FbBf!DDup}Qd9bt~(Q22-&4nZJDrbc(6TRQX5sgz4};3sdP
zENv8oXjPv;SXA_#E)M{PN4NX7&d(8Nf~?zZ7132z0+7##?cOuO{5Z{?{q#ioc{(h^
zACKyIEE}PG3a&3<s+Us?aRP%btqHdE;#z}6&=}#++fCO7q@T~zx@V2YD8J_Hk6>Va
z|AfE^#>F$3+g1jWM+Y#=pa)2oV5gwhE2$WYLxUG7z-BRsnVFtEjg%w6sj&kMD%f@)
zd$>jS!G4b<BxQNL0|Nv)b#j=%N=sWWo(uC!Qg<$v@1Vrtlt>H?^zeWRvW)xc*Vm)r
zf5Ki{zChS`6;hHKRDdJUhF`m(<dBeP%*rAYbc0h?-d1fJ!n=D%TW+O>Go$kP+#iwW
z`8<#C3^;nG!z05VkM2M?cw;I66w0@~a4Y5z8|VoS5FuIsE@6s|S$IGKm}R}+Z%c#k
zSSAc75pibr(P|sIL-{XAV!uVsWGIZO4Sn;v=c`QjU|~oSV^d%X%9ZNC^{fP5NnMy+
z<4XYUK`nNoI2Y`@@7>N`KGF3(ubyMgH98X96U4$zmeH?4EzxBZ3Y#_BOR9#&9U{!g
zAPy?2)Wj&~j3ihXpq1D5KmR|m@5kOx2;u~m6@?>hTBDGa>J=Dl0R_-l8*R&`@a)ib
z`}kp7*OjyZ{jwIYf>}55d;fYo<IMAUKA!-99;b~3C8F<@ZR23T5<Zw##VCO}kF+d$
zJqt5E7`p4G`$j^bL<foTbo^5|+VI+lI17?^y->yY_!Bg0uRpj`86ng(gZo872zJ3-
z(FMFJ6d7HHK^$1PmE@Tc(pmuo!w0TugwBZ?1MRS5@8=Px4+%_)#6P+W5oZI4SCIx$
zE&-9spvKQyPEn{T_dwt_6>K;|swfX>arw7$Ik~U0xBvBjxqp5<&!dBYFo=-exIxOR
zOUIY_iGJ)+s0m?h+U_3|`8<KPe%RL8zZKE!9tkH4e7SuPYzmxT&#<RGpVfiM!|f<t
zg2!XG-YP77yge6Xq<4B@j4GN~4cGw`*VlNm9}`%52duH!K1-Ayadw-Mv=MtRjwSJ*
zI_50JlNm{6l<_|2L;+b_lnE5KUL9=K8WJ7jc}ii3ei8@4?rjK3?MFY4IC~CaNla5l
z8mRG@0=M9{w6Q~}&}fy9Vk0aEteLC#a+JUIfZZ_wWl$9#@dyt<fFxyU8*}?#{`>#+
ztDVnoxUD$5c}Kh7H-(TWA@`VOo4$WOhH5sea3hG~?g_M2mBX5ufCbcmNc3)j<@>*F
z00H)}w8ztWEs*9t5ym+DVRfQ$HBM3CITr<5Th)vh?XD=Ti4K5=xh}FK{U7q6Ms3Nn
z_beF}3v;g>JYsjxDdYC1K-J$u=NdP>)THM#$N<=>k3Ra9!l_ClWk83016cg^fal;G
z!C3~maYTB=+2brD)R!T+)rfTt=80Grz3~Qk(3oVsUoc_i0>m?7kJ7*4HDHOXV3HMg
z1*J*ZI06+nL`$RpJLJfJ1J-Rr9_Bl+EjL*!L>}f4<&sAFavr_5awVLJfUbci4FZdP
zF0DiZmnDy*_Xs<`9`=_T9{=O>=l^muizCiHR96Clz1tbUO(o81O%h0GyVL1Wor;Y&
zZ8&xycsVokn6fs;wnX=HAp&vMTqx1eSQ`9%9%rCoeXvhwmlIi$##fi+|BT21-eDC?
zKF#MkPKHA`k~mmY+5!M!5ex=xuu0l^2U#Ex`_>HI_rs38R~bIyhKVu*>CuqM1LQ1A
zyQ?Zk)a2u8c}_Oe%8K$c@N?VWrhbE9n9jKCjgW-rc3@yT6E5*+`1~CjL{|O4v9G|L
z<jyO~G$ihiHbIB$@ty6(%xVf8VZG4%jd2=xv%SRq1i)&D{qV!TeED%f{MY~fpO$8M
z#Cg2<hUSH169K^Bb!Q)O({LgH4&hbXa;|b7j9AVZ;?=Tj_=-C~XhlS<t=N#9j&zab
zWB~UaLoDTm*^EC&8T)#B8Id#`-Dkp)Wl8h`YP|%G!$AVs8tKq=3j~8!KvQK9MB>%K
z=L6Ueg}FWU)7*!KV4~z|IWG=W`IRv(w+~ArHHPQt15^*P+WD3D%ro*hOEvKt>4&?V
z+vxhq&DLHrZ&)pX|9ATopNUubcvk@1<JB&#F4kUe(la6J3SL(Lgc`s<BA;XJYYwyY
z(2yHAfiLG*v;Fb)tO7fZC}ps7?<Y#z4Oz|w05Tkg0is2QDN;NvLj+*Lg5j=72*4!?
z0g1z6-yS!uex-r!xiD}p+2G(6vYcVZ2ACDe3ZPnl{!~t9!G6E_t3gQ~OHUgXuOy33
zDKlPJ6<6F~zR|@Omv7b1I{FjL0_%I69+ATn4WkAp5zW!)S7HHgAC`}d(6c;9g(sK-
zT$AG_)m?ITJN6;yPos$trvtJtvQ-*gRp@ybJ_|BFl|#i4apB10iCY-<?htB<e4bB-
z(6-!;6<aY26%D;H+w=DH6N>=nSF;W>Rob2%etxy*<Ll#V?`EB4WsoTPu|ItfVEv^~
zX4kMdfKGA%fwP~3niLFRA$1YMf$kC10P-3TyKgE4SU0S!DlFxLoV4P=sW_GxJet9X
z31lfU{%N7}>LFjC;mLOa6U$(9@))5Z5{-4OhFU>7CD>NP+2I1_Cc4L^QPE@c^x=8>
zRS|L`heZm{k%N@DuP%#@XMI$MQZ7>uVoVK1gAYfRI1%8Iekr<wpZJ6wz<9f#_AJ2}
zN;J4256XRO=wCm5{e-7V**i25oX?J?st0f{c`{k`Mdfhc?Fklm*zQv)!_W`gkFSr<
zKU(ET!AQjMb$<i`vA-p0QS%mRZEhrnz6&BF1=18JKr7aRgH-qg0%<Hc5GVj#%P_Ya
z+R{{Pb83JP)~dh&3FfH87zhw3`OlqnVpdb&S8s=bMl3=v4ZWxm46Lg_ocO4In@cUj
z!_W(io6YLJM`XfdP+`E41$nXgsr9`HV$vqcV1-9m9%cY*sR`GtofxgGsjBPI0U#U(
zN%WWz$O!nMjD;D9!}lYO>;Q<+?6C!~KW?JH@wEfT*Ph)0ZmV3Qpj9}Zw^8r|-Q0Q}
z{w3_OKhZ7?9bGp@^wTUOA3vHUDgpu9_vfQKA&YxC*XJ_)P6Njj>B#Ow({yD9PVlkT
z<?wQm>@>iXRy40D1e%9;=B1gc2=&&EfFE|&tzeDI?vf*NLQenXW9)lTtR=Z*<zJd7
z#8?;s7<Bm5vT_Nst%pPATQ^5qFGJ|MN15;#@?4|;BA`?siZnweg-M&NEQ4Xr*#Yoc
z0^>l{*n~=M+<Okwc(n!tTxLpc01`W}G_6rXoEE2t1Cr?N{4oGXH@&X_xBD|@Ai_A-
z1tlGzEAk0&2|n81_hS#>>!*H>i)#&k`M7P!XW)GLaSUxh9mAU54?k=g>Ac=j_(+v`
zkbiu(5`v5V&1b_S%FQ>0xmhOCw9Hh)3i!%{gW@=}-&}#kPx#m8+XTHt|G1Pw>Ca1Z
zvTzWi9FoeAdFCJp5SCJT4jYxO2nV}3!0W*QuxyGRJK`)zTWKF60_OlOjB01Tz5q%-
z=bD+3wq@Baq*w}o5Ukr&XGcNlT{^_gB5FB@W$?E9EaeO%LXk#oQvuu%XAjGO3*H5?
z)7J05cl*0NPKE-YMlAPDYs^K7^FKRoxBI;T_B@}D-O>9A!BT#^Tk*2c|Ni3xI6uur
zv;zt|_s19W0ApYYek!a8(84v7ni^Epo8wJz&b{*N1dzsH$#aPV%w_UZr4wt|jhY-7
zuvEf|vO%L#W+Zqb^3O|h-e-MOgr4DlS}#LMSI4D*C}0E!#bn?z=+*#$;%meOgb9m`
zvJ<F?d7IFBf#3xJL5Z1e%ce{#5SGDMSTat~#tp#L&JGZ80zHll0FQA}n@!z)J>L6c
zj}q$fbPvlKc{YLo3e~IeaN-WdxdX_0WUUwo&*#@C{{G!oroPWFpC1Q^q+H_*x9qx`
zKfiuDv*#1oMo|U?Y=0iC&QSW+SJ-BK{gTle9q=s7xvr!zuA20jky*+*89)#PHy-MV
zfo=)X@Wt<$fcuV;l@fU&s{{z?k$)EI_}haT0U#QZ50ex;Bs>cvo<Oi&nGgV4xd9%B
zeSO16#5snQJ|l+bTy%0JL?8eOMAxk?jUed3!&Q$6%L@ry>OqW5fA}j|n75ocTpeL?
zW}w`nfP}}1boYqNjC%Fk{V(751T|pC7Z00!K!jmg`V#;pdK&QjkK>UzSE0rX+rK>f
z2`oZ4U<DXBm#^IpdkjlTR`n_Wd}g&MP(oaesTW~G@@3$4kP$=-h%lHJW!{)t=c6Gw
z23WAF!}7!w!7^$OE4*a^eeXu-=Mi8Xui7AL4{z)I6Tzdes2MoP8U1l37yCBBE<_+7
z!kc3|Z&;Ze8HYRK%-UJ392BlP4IuUsIl8V#gc+c<sYu&aQQ(Ny<I1iW$dL_9(x`YZ
zj?SoN=Xf+3a4NLh)Lt9+YC4mQ97fQ-`{A##8^U~lKF@Tt<+fnAFNQWOZv;3WkDgEe
z`&Yx#7NJHB-JW(P9;D`pl_RWYks<(9P=4nppacl|Cpa|#CjwG&V}&fG#M%_$apCzg
zSfsI(a*qONSPx|M{aS6W!*Rw2gk{#MUaETXxM%EU5EuZnk-H7|R{uXEY%t}kb9zEf
z{1(vmp-&Q=)GcsoE2JF+Xb3^ZQAJmA!^|TFf*;~9nK}ZaR49v5BVcB2jki^l5zQ<D
zfL>q;gC(e5ZfU|S$F^78w9C0>#OE0i9XU9rF^Hn^%@qnq!`~Jx@wxl{<+CI0gfxYP
zl_-iWy9ly>`m$Tt^VxxI(MAx0hZXrF`)R5;&tf?#6LH`+A>0Ar@qHc}fS^RUmA*M(
z3N_O8E`XbHAbJIMnTiYRgeM-hA7ke&gOyhqioRd?#(GY4;(7Z5JEvZ!*`*W4fi!pg
zX*Jc`aq0YJ`7u?$t#TCRj4FZ?J>vCRBjLPnUTOZ>4*+oxqD&|=raH9?k2;yS$?}$;
zvIszflwcOG`9X|!1PUnM3R@vKw!;B5nw+8WGkaXgjg%fp$qoW|nY%GZ^XCHK{v{qe
z0P2p<QNmWlPZhe6B0C;mcPnc>?&u1kMgXBHkg$9xtOF!Vi{N1+DO-b`^=e=FL|6ey
zl4K5H^)6T=!5rXblu3Rrl@i#R9q_|#f7)1*F266TPjKn}QqTa1k*D=Skt_`yZUJOw
zl=n&vmg!HAoT5{r(uhIJCB<AVh#YOo0s)oO*9DmF2!y-!J-}rn^Og#Zjlx09WdXM|
zUAP`B0EZfm75#}s#Z7JBdjTZ~0JC0jW5sd1fLzB<E@Sw$fP*@vds0%Jc=ZujK<)4U
z+y6ruk5A9&)R4^{X4yCiVS!b&0pZ=wZV8bgq`}^RMjvFk1CR3wSU(eMmUaW8o06^t
zk2&!yiGz+f%p*ObIA`@zXj!cWtb+m0j)b4KhHJ2a(LG_|>39v$2|!W}VIl)u;Amjh
z!G)*8=@qtyfTu%J0-W9=;!h*xz2b%n$jRdsAHbAg;nuud7BQ@1sG2HSk%Q8JK0O^W
zFq%{=uK<FWm@m>+^~}PBBNT^#gDrgOtq_<MJ+Vi)U}@9<VO@lg>Be67+{2NO#Hg>(
zZeL#)0}h#)`U~h$$$I<x<$t@~fah}u0Pkie;2r|Y#L+Uvv-h!XFl%FgS8!X}x-HwH
z@Z<U8SpcoJ2(>&I05nc&bMqi<`(xcBUzkk>S5^yU7~J-26Hg#JCF$wF%CCxV*l+_(
zvspNcrC?0L``ZN~($885BGV%g+JUjnh5o6L!%ReYS-R37lS2foE7&rMbm4#yiJ=v4
z$gJ>YMYzJ)oJ4${s#iH*$)Bto0mD*rDu<M3M@$XCQsN%fFnDQq5mo{>po4%UY%5n6
z!69(_;z!M4)b{M7Cd+b*1u*2qsqj7>*}VV#@7I-|{qZ;xlw@`|m^8Zy;2q)$W>wWj
zXiIAVGi{45axmn%oc#oC)X>oad=tPAYdBr<hZXi;=D=>O_DZxCR#7X)>3+T-MD^f1
zCjjZqhKr&D7HB9#_xT3SNrh9ffo2o6UTWz8-U@;`1P(#`NeI%ngWmr{gGmiQWMyzz
z&oO8c7bH0w3tP#|4O(I#LB<dP<PBRP<Fwn-%Jj^z0N=JY5DCd(T4m_EgL!Lr`Ovu0
zm?d;e5Urw98vzIK@yi+pplyCh;AVIC0O%!iNC|XiLhC=?IvU`KwEa9!2LVO2MU4!l
zQ368Lq+b5nn2<aMh@Mv7%sQYzLuVmI0$l=*cDpV1_-p;3?`jAb5(`5`Sq&B$-797e
z_v{43Fe=KD74qms3PESaa3idKPr*QAl;jR#NOACp1=>qLc;a$PseIUp{V9sIucrpi
zZz2a=E1=Li%dH8l4F-V+JqCwI**fH0L4zGq<#^077Xivx8f&qB%(%K5!JLF733j+N
zS*2CAc1wU&KD2cy`Dg2A01Um_+6c6?DRvW(4_^gLf$UasJSM3CMjt7Gw*Qz|I;B}p
zv!jY>?;e&BEbA&du$H8)5g|@xKTmi<7QL;DG;}XbDU^FNz^!gU03x@WLhruZdk2nf
z2VfrH69U+3z8^<iMB2z6FOdgGrs)C18Que}4~A`4m#qwXu0eMO9Fho5W`LDBS{RQu
znoKf3p`}StJ^s|l0n>GJsoW-sQBZO8q4-SzScpYYEeO8~tT(+v!lk-C%qwdY1y(TE
zy4`C&!Ak(D36AEhr6~(Q(31caUKb9c^K{FA!_|P(xRmwXp1x1D0#<woM8Lbb|5Vx(
z69Cjc1L_UY&m-J?Pxq=_G6EI}0dj!=+8P6KVkaT==zxeeZQ7`9Ee*;IL}<Y22`fs@
za_diM&+S8WbI1N_9E9&1@BKUu_bf$+Z$?=@xGVvx#CE!t(EGa`06}UjGe?9~|6TsC
zPBXP4maKRI_2QcGpOxp(SV=HIu3#Z$CNATg=zZXwg^-}i#>)4-DC5J-rnGch1}8D3
z-jBK>QD&$Jqaob_QSE0_hO;mjjW?N?35Pgp5`J#z=dlM$-67bF)!i5or|sSAqrjnJ
zXK;)veX5~O+6@rk;AqG8MTOz%aJQ$M_i73an&Q44uGt`C0s~l2AoGAWK<qS--tSEt
z^n(J-f-g9`h3#3l0Z$WPOM5bd)Fg-)q+Fwn7%!qv(&j6?edHh|s3G!GT36aN)u|SZ
z%*d&?qo&M0yVOKH*@z002sqs6#Gi4DA~CPTMnqdvW?+4z=L%RjBM_1XL`J}~RBoER
z(B{Dm;#N8~LU{s^6%1Y4q5}85yC0Y*&ng1mc%a%!7Z@ZcF3fGu^OZ17t~g71BJ{-J
z<&Y+T=5ct0gI|!PVKW+$fNMlMzI-%zA(s6)`o36;3T6o$E2IOQb~zScl9Xhb>CQue
z+Xr2%xqo{eapvR83Fx^Ta$9eTUiiC~3Lq@d8=g-EqJ!F-bnLfSI#C{i9>NIq%lQcd
zy37slDD2`&N=Sfv17MLcJqT+J5<t-!0s-^GsQ)S3RgGmw-;Y>CS(KSY6mSk|ad<+w
z!5LTgFEOC>3BQ<v3>FXz@l@7sPw1ioh~88M*!js=0ZWiFtB~-zZd>(7OFL>@u8%?B
z3ql_#z4XZPLXN`&AuE#~9Cmnjl<Z}Wckq43GD36z5eKxX0#85ky<%sz3_jL<>m*3u
zTd5Zt0H`e5xM@QHz*{T#DFE^O`(w}kIF6bc-JabV^3$h)dH}h>^r$)b1j?e7seKbc
zl%YEV2~k$qYCb$oUMo^j9*En8lATX45>C)4xv)r2zgDsuP%9V$1+e<|kX`<SfoBHu
zP&i2v6=9~8S;`5XOFh?Ok`Op%r{4p?KB3q&6a{B65}2WkD6<5BRaFH0hS(JfM=Z=r
zq-|Su)7Iq-I6JQCW?<y0`jd<<5gtcb>2VS2c-=jGkWo-I6)t32t9>=murCjXmo4+?
z0!0|wGNxahcY3a@Dv%Z0kYY{1tF*T0?c+^QW*|Z~+S~c%@qF}r0*C{oy_q4O_V@}G
zpl>^e{p11=B%>>-hMHmu2&3hwRbDgDA`8Pn$c(~H7VW;63$xRn2@xLj{75IQ21*$o
zF#~0=0b4(@*-y{tpFf7U{pit3lL8eB4~ydZXA7hekeHEEmD4-8WIzBbw@v^mw*>~}
zV%Lc<V%ve*Ssaaov@P03wkvRPmC1<%f~?YZtv&=;1kx-Lad`P`Kvn=5`|)^mD-d|K
zF_MtORf(oENDDkXxQ8<p9;$@t6qt%TQPAHmB#zz$2Syt}Z}-1`w}1j}ZhL*v3HRso
z(GM?6?S%IdMmjA(YXAqpS~fv}ILkbTXt+_J?rM1;FoF$3&g4a0s^zFUH6_S#1)z@s
zSp<huc^gHd(5Atq&dC^Dcm@u94`J`cAE8!qh5y_V^rkaV@&AgO>h~*P5O}q*a-J-%
z>#5~s{C^pZCL(|qX1Q&5h5^FDix7}}ct6G@u9Y<e1L3_{x;uxGf(v`#!7@rF(d;KK
z+r|v@S}uZFfxzDPV|Rbs{3+kK=-0J2n}{KlL*amr82H=}YRqry8TMHhK^9F^wB7!4
z*Rp;N5G4rM;pf?VKL+A~7|0X-*pI<B+;GqWYoc)Dz;c!{bT#xc2C9|`Q`-fwp{5y(
zp=w;WUl>Rq+t2_yy|oaXUM-v<|Cyu=E@PTS0pyx-OPCmW+1->1MVTr8E){L!*Vj^$
z>LW2>V8~xVP4YUOS~vx70tiD@hXu;uoJuDwy4}|WX8<hHgnQ<3J|B*f`CNgTr>DbZ
zP<ly)GW96(vo|m?H^Olxll_E;A0u*u0UX))W3LY7oCZ!h5eoT|gAcsl$DpVvUU0BL
z42Jbs87mNAR#6Z-V8H-smt)-_=nw5}mdw-uYv+0P6QDtzC=oY5_9K#-;E+J6Gn6z6
zYLA-(!bD^*`KZz6t5M}N=60le<mekqb>qi#d4I(9`pkZ@y%<fkZdKgWLp=iEHZ-zo
z5+XcaNiQPJ&ya<w;Tw~w88}TAzA!E!KH5%aEv}9jmVcSEiWHxsrNEe1C=X2(s<_d0
z-2`j_trsfy<FN<8OQTk;ePsc`-Go~)F@Z#}wi&WEz16UxSZ1Jf>xsZI5z{9j`?NgZ
zm<g)~a>B{tQc-CEDyg*V28cAsDMGhe^krpTgn&*KD7^xl@pXB84}ABd3;@jNOPtT+
ziH_E50V3h0E${WN!?7NX0R%+=5M8-?zJcohG142Zs<C<sL@)OKJX2~Y500zSWV)7|
zt2p;!0I?8J=}3nPV5<W-{5Y|kjX-IB0cpU)qdgLUqrXdPtEklIXf%SeycCm37G|0a
zN>{iG6EYm|@HqThz}?|AtvK~^Q(%Y`VV40o11zg9m0tr0j)>>+*aLEJ+8FLZl@9_y
zcIphUz($8d1-WRu@s{3!eul#X-s7}Mz;PD^qr0e;km;44vjvak{O#^Hn`#B4WhGz+
zH4goL1I&99tq!&mVATT=4~Mq{_Fr2=!ONQ}!_I#8VOV3KbyMbJq&x``djM+nec)i`
zMHw)+Ze=NqfKacA+8h18B2!%C>aR6GoU%$1de<wcMhNf{rC=fAA#n%?l=ldLJA<(p
zD4fzeP`-i{bZr!|={a1ziA9<y1DDFi9A_5Io_ToHB47%*6Hf1zXy`o?HV`26ss#(s
zb|u&O)~#ZZkM{*h$bbPXjKlUlAh)KC*F!KG&)$maQnX-!BdMvVG-7UppYkXVI6saz
zs(!lLDZ)f4z!6S4<;Exl!EdU05(%Ke!Iz&&*%5$2@=-UYYN^Q9R2s66-kIx^Rw3kG
z!a-4};SNB!pyZwoRoBeZPbWoB9FZl(2LKVN5Q6f=S8_iox%yIx^nfNYZxgvi;K&9b
zdxC|E3Xl_G9pg<Oxk5~NDjs-EVs2AKOJ#yr%edw^`_Uqt3~&k{#;D~|#u~SdmeLII
zqiwi0$~Iz$h#0L;0JVb=9_Uu))>KwUQKD_pCOyFg5?A26%(RUF=qiIZ=TW_$8LuS5
z47T#JO-F*@dq1PZW7na8(P$+QV#gUr7-W1#8PWp?cpO+?%Z$WKZYZZyfL!{JBSsON
z2=~I@)Gw~<CMrY?W!*weOS=3F06RQbgc!XsJdr0Z{|O3Z*jbBOm=x%}PSYd9tp{v2
zbG5)&1O)`va893*x0MEMLG+#ga*$@4Lr73yiPw5FUeasPDGrj*MUv|cU>23SaEP)1
z4SDXz>JdA}6bNCG!2%J%aLlNzT?F6=(LsibikP|X5Eir^gb0r4jDkXUFAs5fRcn<-
z$wY^il2HIjE`2n_wQcLYv5Q@#<qXsovu16zjy7y)o&Y>fn+ct)4AJW6!i&}bNB|?H
z#^eA-00&%K4bEyQtkSAa2u}UHSLC#DpUdSap2iD<*L9Pds_M4rf;=CfCC?QA&N!lj
zP@J|Qjppe&x>pH00K|^bE%9e>)Tir4%m?aL6$j0;0b>!@7S#rUTes*S=5iHjjOv0^
zPpo`W6J=XA7|*rj41V-v2EC3%;DoHK(r85hWdR6AF0$<iYjRAC?JWL7DqI*E@K>GW
zafT03y;ou+yzr21f2mbQxJME=-LI+iRDc@|svMf<=>S}^rv0JorYi8t2Sy=Q<|>Ge
zks7UztDQd<Nm|$2jggMFEdaa{3WKJ?(cbTk85qzgB{VB|7mC$z?GE3*m(8t+hpeR!
z5nlU-imKke|FEw1V?`S~H)A@?1I!(G?2C+|X*e{E<*m}<RsvlxQ$m^nHLI&~#o?~R
z$kP1_XMkA|4!y6N!1w3=IEK;k6rR^YITF7uNm9JL*Jd^HBAyt{@{r<yWCn>JYgJhc
z-gK+Yf_E4Lv3>aZupa5t04QC&=D}|ZHNvs?$Iw6`OmQ2KHem*Lj^WmwqlAqdcPN1}
zYM2KNk6_O+;;I%70|@T4th!{td;q}G0!OKFu{*%DR)STZrfa#A350K(0=<K<b@Xt6
zU0jw^8KyRr*&SxivCdD3(K2}XIsNn?BwBygv=zXT9KkNZ(H-3McCWw3iA4D!MoqI0
zU|pFV`0`>`FaW8Z6Dv2p-4*tHIvyPctd(bqD86h_bl3n74xTPtVuRrAyYKHC?D^@(
zPoF(JViNNaX9NZ^zaB~wQUN$JL6EU;@r;}R+`z+z3}6W!Hag6->Lx@SAl#N+4#;h7
z+hd;sE;L?p#<^Yh6+oO;qMvo)ASt8|5zDGta!xVkCQ@<_G4t&J=TJhfR?Q5e+1np_
z$cTjVG}YdVF~&fc57fe71szQw0JpreZhYTX-OB21xe>lTQ6{aOI<oah2*Jg16i2Fz
z1UYIx<^V~8GJv0t8wL{G+@CR?_&gkeX4?4)00+*8pXtbcp6CAMG4|6PP~e1*t={W(
z6gV1%spvv&!Ts_1YsVSDv0@13=EpDh2jJ-D(iu2_z`|fwVcEXF|Lrg6Uq0Nnk9*(u
z1JQkSgW-Vq)tAah5M|C;9EUg}*o86R&zXqcs8o+CUEvx+wt^Y)0O;If%kYpz?|T%V
z6qaS`qj(krZUpRe792B)oveCZb!h?V@BvI1$zCxCFAiM73P_f<(FVtqr1QDlBoekO
zv%vPpVJKSSw=}xzP+F-is`tPC_rHF(t$3aXAo%y&!@X}k-OQd;v4VtIuZA$X%~$Pu
z-6neQmk!QsKs=G)g@HJ8q)dZS@iZbJj}A`|{Q2A;S<@duzy#se9b9DNFsKSijcZ%m
zmR3)z0Le!N%+;B&9vg~Y6Gz?mH~=kWhjshmFLW~ot^59Xd|GxZ{O4?h7n2|_4WKvt
zWC7s=NEulkAtQ|%)z*~)KtxOG-cPj3Qj~;&=ypEMdPeLPZsa8&Zg9`S#W>Q7qA>)g
zNB1QRNLra0>~0pGEFSr0U&a~V@>KzVrQQG<%Z<ZjNb)C;4*)!z;Eb^-*XRylYAS<{
zS^kH&+wH@L?|=AjAMXm}<I^LM|LyO103pSvJ$&fjS%NhKu(KP81_{dn@N$%Wk!QSB
z9Pe8dQM55f3sc|>FR&p1`-()SW;e5Q2a6*1<5-XDR*m^l7F()JMaCy9Z>_1WZ2^w|
zfE+{i(}7Szq7~Uw{B<V{NGt*FWlzpOEZ_Cx%h$i`yFHFaZOG-2@qXlNf(L-9g8imf
zGk&VKfoAs;t4kgm%s>&E3uNJSQElkCA5ZHMvG4oSg!|55(SNiHW`-!&o>1B}UVF4S
z+{41r0EzHOWTo^q10VI$AiEY6-67>G7DRmjEJmyR0T=k=aAF0{@*I$bKurer299O<
z@ZJ4~A3l74|9<;kZwfz!)+6?tKktvC-lYvBeHD1bgs^kmC(B{~%Uptl?$N)gRp+w>
ziyGRpVesmJ@I-%Ltv%NJvp#Ds+pzQCxDm3S4aBg4DSa)Mx+-{K)~3sc+kgKcjQl&$
zIlPBE5TYBya7u}<m4?&;AYkx-uSV^L{{{fJJM4q)ryb8vU-mf8z87f<KyuNKF2>Bo
z#^)h+`MTuRM+=@`z0bBQBj9p#EZlUtFN=T>PwSTbd_E4_?di5dSPK2y7!?!{EmbsG
z`K16qTVxeNcpHbzBL<vy#Qv<BOJ#$E35*aQjg6sm`v?XO4~~Ns&djtD(W=DS2+R<o
zjqZQ>%fJ2Q!^iEWssLmGnmo3LeE7=g8GZ6p#_&-PS{*g@`TmZ!4ijMh^B^T^L;|(8
zsTv<i5*(2Z<P%tM+yGgZjf9YXDyl&MTG|S;p(5BJs^+DvwF07L;;2IO{l~w3__($6
z>knNN+2sJlb-9V){F0-??x6DivMP@r;+*{Nu#K87fc^zQ-#>kNoO$f~<8VVaT+|PU
zV^D7V+NWz3eJdafZ(7fe0Qj-0S|cR3CqPPkU$!cgKQ{C@pHGV;p2uUi0ur~%QXQ7=
zbGS^x!lj`QYvHtE<((A+^qWId7BsLxiHH`jdfbq87eFKo*roF_*UHBWuMuz!R#e=q
zvNEX2LPE=jAO7u!+pP#N%>|y&z8(Nr-Ouzikhoup8Q?LVtI^p2+yqx<_j7SG4auD?
z8sC8`!NQ_b?Iy9|S<G7<dQE`gG$XXNC<Ni@fGB7wh`a!hRc?*jwlq@y+kd+ZZ8dxA
z%_N2I-zm=C1Q{G2=@7;mR?fd)R^1iA|8ZMYZ+cTA+O+K-e#rgwuV23G`_uMnAW;UG
z=4?59cKnG+0e6mX_p4wEfSuim57u=RDJ_6BU9b!ody`GFTLN)>{m;j76qIf>_%h+J
za<p*+v?+S6Fz{V74%itnxslYw!!zDg4@xj9xN_x}+H3*7YPMLH)MgI_U=oo_HlLXq
zz^kmhsccOIZCUSR&n8`r@VTg%-HtE2Lj-5;L+2^b<E-(jv&5h(4&4}KAP#sx{pU0A
zdY*B>8U(Us_eURn1|dHgibpFxtH<+c8E)wr#C>)gfdrLt>YzQ5>izrgm6qG>c4uT_
ziSz8AzIX#VH5lo#wGyv*cGlP)74ALZ0nVM{;0;aH4{Oh+dRy-Ru&!V{e|&uY`rP;D
z<7*!?{y@z;UfoVGFHPJf5+tv&Iz}=TH~_Fn_@NBl279pYcSZs#4CoKUD{y@I`<LT<
zT1O))xpz+_K^DE?c|JR6U699Vi11y-ZG9l(NC%8jh~dFFzEZ#{`0(1v2Ap-CsE8Ns
za59{jY4|#TfwqB~D66czFt^d1U}VSVeOpDR>sr{CeY9P70Leng?p~u=FMG-{NJ+#{
z5s$F+_&K)=D^LIu;q5PfL%@$6dnb`GSKPx*6YvKi-3=Z`7BX<H@k{Z@DMrs1kmW|S
zeD{C(i!y)s{=55%!;ambmZK8^YrrDjDL+5He%XD}NZhJ!28*bP0Q+LJ_%i^y-#+}?
zHU|9s5fHtd&ztUh=sD_+pVW@Bw1t5_Xyw;>1I{F(%M&PFj3s;n+!g@X8gH$mq21}W
z+!^u1UynR80>XfOHOmA6OVhRC{QUXZ(SR<1alBX;`w-BQqzh5Rl@?#QR|362M|}NW
zZyW$~aH+URO-!~{ly7$fHl-mrA&nag;4K2l0W&1zqbQ@>9-vJc3}JpeA4A~3*JKu+
z&y4COf`t(tnH?`Y0R6m8N}81TV*A^N)t~zsU1Ob1c0=^;eyp3cA(DCIxG_g{SPcrU
zai9RQZgN|I+u#2B*G=&K{kxmu^t9&}{g1YAbbCBYbQWjCc^CqLIMZ{iIQ48@90#8!
za$@(r<F?#?yfO14AC}lVu<8B7<Ni2~JRf!rx|CQN2E1%xnP{U6=Qy^iH}s4<X=ozx
zuzjVwpN}8zAMVq0>9*XN1?$IN>=I(c`6Y_vVPLrtJ}>KMKw$@llnUHpkMk(72rd<2
z9u(!P-yQLd!}LT#<0Mv04zUv@u`(+}bj8uSxDbHs0w`5klIaXsx1Eu==|ImogidX+
zZu@`z_=KS$?0CN7ren0_3qCZMCL;^OFCJcmwf5oTeFwhQH85*KPHU*qX?n9GWi?D3
zCk9?o7vZ+tZtHEq?fdWVq__Kbw-18F(Qmh=pZ|LL@%$e+F$R8ps9f673pqToKw}M$
z;|Va#y5ozSUl#$%_HkKo_6(5TY1!SLytiJE-B45K(oSR*oW1gO5a)|wKAb29{y{8L
zl0(95djM`f9G)Au@>&@SLt9vN<wh7Cdb8aQtb#xzg~0c^7(lGY-uKzK8Q!J=9-P67
z<R~a+A8p8Sa={_##Eua;qb_SKTyewe(LiX(V`b+IMj>rYlK>oN0~Frj=8m?MS*D-o
z^YQG=a75pC>mb0M<!T@i4wxb=C!PlZ$4eEgZrI!MW6B7$_7?#06=p!w>I|40hDS}N
zg~Ggb0nl)RMF1;n$tcLy7HYid{fAZVdRw3@+jf)5S;h9f9sc8g{7?S_fWr9bQEdV{
zVh)!v7$p>tz@a<m5Zr}7ijKa0UT^RWNJG2tt)F1^4&>eq7_=Eu7X67YOd=Q@6N+>R
zt3~fDqV5dT<3HV0m!)eu?VCG5%Bm}G1PImyOsyRoqzIt5@FUt1+j&X{!uP}Mcwb5i
zFPd-mV;mQ+?V`WG$c3EsT*8w8ZqU~Rbv;VLAdj^}jC%|2dJzhj`6<@jdUGM<*{vVD
zbxB5IKf8B#R4<mxQE@i6Cl?kL$cYGg8;86{4$vD2Osr2<QxO(Ot2XGm0tWPUdCmx{
zzz;V8sV>R^y{SUL_uGg2w%iq}d|$ES<EU3b3e_(^;NL&OK+^iu2(v!#rMyioBcOgT
zuu$fhQ9Uw%?aRkcfU+VRux0GqaT<EYdK3c15*St)n10+D{UW$sZ61I*fDCdMXL%bG
zo3>^9V8FdSz$?fFjkj-ly(F}d!ld1P+?~vh$I~5onjgmy!$RfZZ(ZU+)(f6b3wSBA
z5t!gYV35SD2DO~cK;S&qz!)fj6bZwj^MNLcR4&xJcY^tD*jpo{J^RCgipv&<!yVHM
zqZNJ2-h>Ux@H30bc$sPa6nSkRwEpAw=h@3LRk(<?46Ilcfu8XdT73*vQo*?%NKiGI
z_;&xH0a$*x-9HMipc^q;?Sv|DtHL0jv#19_(+`;S^D210VQ<2*w+$|p@D~6IyyFXm
zWDPHkF23W#^LYUMzOBdpL}y3fL_29VxDsXH9^>5~EAO={CZR{G!pm2))9%X(^c&7&
zVL<OISamBfT<-yd*-LGM*TWw6`HSsekNvRy>EQv*%=fP7Y&lX+Zs|CMiLu8UQ*MHd
z#~6N)-FxF>MO3#3iWr^j?9!{7j!wWLpfef%e4YTQ5YQj?fGj}3&Mf5Kxa-O&QvPKL
zhvF6G3-kXI>8Y1Xb~AiR+o5CE9U>6YL;+!RAPy)ODH4TOb`%JmZ}%VCZ2)UPl)oWp
zYs>ZF5nuek`r*C;w%ZZN=lOp;b~C8MqQJ3!mI~jCV>;ebmWBv;8J*X63WE0&pqs~*
z27qownk8$n4ZzdIBRhLjmQ27SN7%lj*oj|nQAy5p5U`u>$GRPl`?jh+9zTBgD1rr0
zAiD;Tz4ym%9jxF32i%dz<LmRw^V8#D{j{g;$JqfFc~wrAK`sVfVfc4q3iDSliWrDc
zn#Ah?lL&CJ%h76qsf7rL2<p5_a)UY~$;AL=WCzavumCq+8DIOO11Bi+0YD6L&p5o;
zck2->5rKgn7^mIyO`h`o82n}+tEal{Yb%LHbPSeyMP)}6(puK+?G!LO!*vW70A+v(
z{V7fce1^bwGw5v*xF3P+X3xj5_e?9RUOhG6d%IVSN=n4Qow09YK>aFK1I*8Y@q!m@
z*+>)FYda9}(xmJui5_{iC4PR;TWkYHMsE(dZ@c$>`*iytysr1hXO;VRf0tEF3&+u)
zwp&jMAGRg-cz*u;<EO{-@%8ku{b{>(OE76A;L#J)IV`S@+Q@BUgiY{^ooZTW<ymRf
zZ+8Hvk@$G5#iI@GZPyfORiLF8vnkO~18?m(8~U?XD4>vKBo31GsCFEKx4_AMl@RW6
zVg3fRG2geUu_HzMUO(*Yw_b!nLBb+3_iHFweMBrCnhJoH1eV+V4jAN#L2pT8f=7pp
zPU=z$lWAem&vVa29s#iYRPU3t*o>siB~rUJI*`Fr<ZnzTqVWOlaI`?&t(&9e;rp=%
zFDKQj*e7<ODl7ix>vGms*GiPx87aAM5gL-F>WA5}sN;uz5O8w_!t8Ni$2tZD;g3&$
z|Ifca&gb(yJ)X9YxIqHUH~86`Wehj$sm@c#LdIp7)i9vQnSfK8V#w1~<P3tyfUV3G
z3#N#UZo;}5a?&jlpp8m0W`IvdygC9g#|SZ<r+TSE_X{gISfw-lVrpe3`vSI)H@UU{
zw(V(!i@)jGkY@I@PycZN6~?z2GenmLfY$B1`v=(Lk@<uzTUd7V>|W8`0U@xoWko(M
z47sPn`e_|a@pKDF4EG&qwRl+?FU(L;(Rw|qCzEN#ZfZd8v?AbUy0gI3j)(0&Xly9#
z`qi!>!nb?$w`0eNU9EevZwBBQ-C_g6EeG}MU@Z~`cE>|DI0F9o^!NY#uTT4FhKSvE
z+byOd0&id}=v7hy0CBJ|LL6@ZMZWU8EV0IPIj?h;tG&Vj$VhC;!B?RifU`zYsRyDV
zK`KK>hFC8#Bsjy+hcG%K4`?7zVmVCWHJT5AvVd*D08<k}Mg|aa>1YrkxuMYw3>Lv6
zOH%>jeEyyAbx(xaPAy%SSb?UCP$SU~-~Vt^d;%WG6Obj&H0#d<A^X`OD*>`>iNg*9
z;0f!e1#q75&pS$fz&m5OQV1F^%u1}XuB9K(@=>M6&E6AnB#?V1z|WPbDdMp|jwh%w
zBtvjvs{<&L%#q&x1Ru}{ISI#A{{$W_JTz%dfC<+3j`+0cO+EsBcc8lgKMJ^^AOHDZ
zfB*D+^h~G&S!U2<Aqhcp@fL<6V9n46tso|7@>&OhaAY&RiU=xyB!|{)Ye<Uo<cC%~
zuGU>}_yMRhNDLs(Oov1bhr~l*-8~!;>2?OdR$pr24sJLnlO2R?C3{=u?jSCeUGMLT
zZteN?g92c|LR7Lyy%OCeB0B>QC~9=wjW>Yx!}mX|Dt^zSCggsqb?avjFmUWv%6r_7
zCu|BuSe#Zfn`7TQq{gZUhH14sHZZFQ^Ezmr;G3@hwj$i@@XY;iEAKS~@1XU-emou>
zB&PrY5`gCXw?yUtfo=uV#AOs$!tjWQ&Iz#=E?k(#q?7HrZMxlmI2xYkffaZ<7#$7v
zeEj%j_=~Wp5de0AAku&(2Z6=q<;(TjFkeY$&8+Hm=Olmcx0dmM6$PRI;e(t4DFQ7Y
z$_^an5GWYl3=&y&(y<mr$4<<yWV)Sw$HL4&gu5X^v%zKPXao)QDkfDaeYLpGJ7xvg
z{vH4GDxf4W^+Hks_OJr#gY+mlJ(s?32TZmbr~xi`?mc%f{9HV(pXc){SOM#4djh<k
zEYzqsrPI(8!HTE31F~6laws6i!vm$_X0X~5P+9+NZTF9`<Ew|CpY042-=hWyJHj5<
z78D7PKsdY{D2H}pyvf>QjJSK7WjR+9G6^n`c>o!`pY9!?ADim^-fn+wNGp>DmH_Or
zpQn9(^sDM7Kuh(;80lvyhesBV?^;z#Mu{02ag}#(5fmKkQgv<FvSw777Q+arlX4VU
zr;wl3vG7;{C=rru3U){y3cws5-jmvppLD>pyll$=DweY~_%u6+LDAHuG1vl;jzF1F
z5|`|G`~Tp}U6oa}-9)u*jO^#vPiKdBkKE1cL4lzx>*i>W+X{Q;<LjrVV*&1s64swz
z!*DRFISJBqRVv;6vDr$)Eyiosd+9|+fJG#WE8P|w|4e9lSG4=bf8#u#0KWX!m(Tvd
zvF$OkeKU?b7=>y^X+;oYt~&J7$20g6*&7eFNqNJ70Xfz`n1CUoYRCF`P$Q743N^s@
z$6?QI9e|G9p{bi{xW^hsKo}#CG0=%qV*Ub31ft$+#9-Leyk%}|45_ly6L6syDK!@c
zOCu^UIWgP<vs&jjKOj@K47U-rHNzCy0`Dfp;HceAUjLl{L|R*6QaNr!X7&U%2Ff9?
z+h+lowB1;n>PE<?*{A0@v9}g+lo$b{9}LESJem@Y$8NzV^u*RXfUw~)I!@fX6|1i;
z=w~Gxtfd<N`LlKJTsdM8N&1NfUZ4XfhIP5EcgF4C){n<_1L3~@5BhX}xDyl9$juyU
ziQKqFm^*-$?o#H7K=3!EB+rR?8Z}3tl!-xLuBX$RF*+Z^)&!Qe>r+W^1Bm!YoL@f=
z*=IGhJlxE|f`Ev1U>!KWc<m5xIo4HDA$WkDctm^_XkD(c$e{MhHIxVd=Q5T!z>l}_
zB;a<(0v$NEqWbnimq0wT_a<16`@`l-1die!BoIaJ9m+6})>c5zY2!WOhKAVBTjLIW
z0rE5fx_x?{KR&M-?L-3b>U+nIeHRt<b59EN12_rx_30K!XTrkd;c)jZM6LB5QLq$l
zSnYA_U%sAa^>ld<?~!q`$H5t5E1K|fzp38tn|`GADZvGY{t>_$;LtMy!lK(~;hXKp
z@bR=OQJt+%8ZY*Xs!(HMP-6m@3Me6+x!OC&N+N}0YaI~nBoNPTpPwbMEpUVr%p8#Q
z0014j>VX4!mA}PW03(vexzvyKdSXmdaW#Y82`rU@M2=QBDF6!&1Wc_IMt18ib4`sf
z#pC0U<AuOW4EEd25aMl$#f`&D`#O-|avE|<kWiZ9>>z<e8k_@d_(K7W$AT%|nv7%$
z*<n9EEm}=b$4bgm_~KZP6$Y<yUc})*^yjf#ID~a!2|>EST2r!K&_p85=mGSKn;e}D
z0g$E$hC4gp!cBD3TVXozyd4<`JjeSfAxJ&~oDkib0uY5i@4K1pW4yt$!kh}{*HZlL
z478q`$Z)g`a2NqwYxl~yTZuWh+qO0A_IWR`>EOqv$bp=$e#}<6mBC6mjNUxR>$_=A
zzw7&x1H#1Oz@;|k#AU=rOwtO+1sxjV(E(7mIFg)$StiXy%770E@3l*Yv;@3L>ER(F
z;Yl?LAgRpkwIE7ygwqLL4iY2~Q%h8@w`LLz9Uh?OOv9-q6UXCu_5rr#0G1Qwrd`C0
z40cBB-Jk%RgTuS+W(UBEg+Ph0IIw2vthfijO9SD{-_6hNrQyN=%COs5;=m3CSRuOU
z=(WY;?l?{d@c18(4k&ubB*8^^YYHif?oC=h4;+BE9%YaFjZAMzsEyrg_Di_prj@wJ
zo8h-m6FAy(R~1@r>!#M?3!1e{7^n5d5c7<qM0+Y+!SGDGI`lKUsdlxDCpbH_8DJJ_
zi3uf*2_CEsjb|5hAPPOYz%3GtU<fa~g$I<4M~diKw_F-1dsK3*oV|h8^wGTvkiZyf
z&H*2OeMMsHeT>SIvLLMiWKsq?M??1JK!*d;09cRnu&-ZsMxzyB%xX(?0)eHmGDJiI
zd*2PaDg+9*XAe8Vpko(iq-PwiQ2+@Rn8_3Qcz%8fi~_~eKq5k6Y}6oaTbE)|x53GJ
zK8~j+&i(6iJa$r(;dU!v5rKG_)2T(6or6|2YMMf|At~l0Xs$uEPc$iJUZR7v1I#VE
zK-+D(-<E}>sW8qO@TOb3C5g9_`DzT)TG9;Sp`H#Tm8_4CieNTiBalg;IU6hmM0x|@
z6=Qu&s{joE=bYyiOf3PRF*7f^*0d6NtOu^8cFg4cIp>$Oaz_TlK>>##LE;R_+8Bov
z_vNKVJah6FIJe#y3pQ#1`X1R8-L0Rt9|y<Ce^|l+O47Eio2V{A?BXzo!#%JiU}@gN
zlF)IW9(#`nw^rF1V89~+XODoK9>uO<;>ym5ojm+(vTiE?W@$GdpZQpEJRd(kvjgUK
zAVDC3t<q4}Dgz-S;KefV-1i|UKN1Qg$|O@Ns~9(ch!_GM-NT3gGn+>C{smjAG`+9O
z((VdU&^hT9SvB);>Z8PZBq<mx#!!e6Lt5$^3@e!65o6E71<o;L5H0{JvxrCX5X`X&
zvQFagfTKwbX~4SZ3IUcg5PXa==-dgr^j?@g2lT<Du<(+g#?pR{=R5hSAYw+c!`<So
z`P8^KKh{YQJL9>+JjTq0$uwZYb`b!pZgSH_idt5^Us(840oeKUA-ZxfL=lm3Vu1^w
z#2IRP>=C#~VD**`;j2lM*yN@NHm-FuJswE7Js$f3c|8F_Yt)8l-@*jIjnEoyJ%fS0
zU^mw`MO1)qDl}fWl$;F<3d57|^w1ch?S?O=8x#q&wGlSz1-S!>EN5%z*?R{d4h}tw
zCIEm90Ec>jhn<Z~fj<xreBFpyR6(jp0&yl6HS-N?G6~Oxvvrz$Vo_#clDD6WUf<qG
z=d8N}<O4ck)ZpP}qx#*)Lzwi7Zx0$ujBvMj;hBh0l%{}76Bh6G1q4_>d+n=3n;!tg
zRhP+kp+H$|kEyEzIL<hREn7YT58Jzq_#7+F-p>ek>!&#+wVvBNVc{lQ6Ck=6%dsBy
znBC4}@FeG<!ei=JjP*rtAcWNXCp?a=aM)7@1R#vK0U$9=W6Dc}N?@*za;2`f+}Qyx
zi_ij01Kswf-daOGoAuM4W17f77Ps3EgrT@qf<(tV+hhzR%b^GmA8VM)YqG{nB~^gI
zxB{ru5h2Km#}PLL_*orcr7-9L(Gcc2qv)0c=`KJ=_yEX-XClEQ98L`M7zP?1#p|T{
zVo672PXU8KO#m9(QGx|><asFL@Ng>uq0(!zGs+f=kDG~U<=uY5kEjl_qx%6UOG%$T
zpU4PPl<=-gx|=;S3?l+$O7oOrV+eSWjgiiQ2bkI^vPx6Jc^vgrZz>|luu-nUO&q|Y
zH&w-xXrP%L2Z?%sc<#+3ZZfGDfHbiu!2^sk6iB%AZVhNBw#Ou{?5(N7tQd4{k^et!
zf3zjZks}GB0zlQwBa5u6>Fv2Y=j{J~+wOh1yHizUgqx`X*arY==8-`%S<~11bdkx7
za5p!DssaLm$e6T!JkN2QBZBjQfsg%g9SGOw%ot_`O?bXQ%?(6wvO9;(5Xp;K+H2i*
z8)H-wqNQ)J-qg@l!x_zTM`qcjO}0q6d}_E7+Nh&Z;f72tS(yOZCYf658pIwg+!0b2
z+1Ceym;i8VZx0Q<Sy4X^f38T0c_T!GTJFCehJNM-qEioL@aW+-6Q@pYe_7-<qz4mp
zeCVhdd~uWPAqDeRfKZPxU<iY;9rA{5#yAInhw$6>);9Q<A7@MprHhB4wbp<);N3O$
zO5LBhi!M-9JHgPofLgfowDbqBCou`5V?a9}Zvb?Dr~ve~e*=BfIr<}l?o=Dca~P%;
z*N~q$$8rlNz)KLp(tHDEsR<7!11ly%e#I=xXea?;PgxJq6{~rX*_9iO-3$n$zpx+^
zSBp_SX=sx}5+!9hG2EjrHn`0=D-C93Of1|Y4iGZK6cTm}>qm63D1th0(N>(!l!Twr
z_6okHLJTOyVnB89Gf;&F_UWc(_P_xgU&tEth?|(j$MLk|jKCc>)n9c11SAZD%@D^m
z27JKj0YLD0`}UW={<pU`fuHB|mw&#0ggEH{Ak3IV9mA`J+6NB#3C~%1d3Goi4mg<x
z(Z@LE3o7-<5R+2AHGF$KI{W*#gkrY#m_N5;0A|xa65mrG&WQ-SbJ$gcq|nUz^M$L&
z2V};{?37$8cTlvlD5R}3z{7FOuzYh2_=uWqHti0W4@mGX3bq<L#+Z}(f~$k}JX}*^
zJqEmlB%X&zmATjJP^CV&hvj7um`3~Mr(c*6d;IaNxQP~FDv7wJDzO21inRwf&5cnR
zd7AD55VisIMxlw&aJ&x<37}_I;lol>xpG%yLBlP}3!juOF#dSZw*9pK{2%}EKmOxw
z!+5^`<Nf~vo&dwF1B?d^V4j3I#uL)fu;Uwk`Dq`ACaph6URk2RqL#x|)f$9Yiouh$
z4L^3a!DzoCLU4c3&H%>qRKUkM4Fu7qL(LlBXx*ODJ7Fg2S8;5aRn=ZX$UQRhRnNR5
z&ru7VBVT&F4qK7q;cX^tcs^h!EDzSf#<NZT&@l!8AHp0EfCu20Z{}S3ep5zbQ>i-)
z5BLvE(VG}yVXD@k80}v@^X~zBPE`{{&Yor?KHhl*02Dm?t|q$x;Q~(mWCyT?kpMH8
zTDL=E%yg*fVS(fdWFRDItGbC-lRatMys@;004>}z@YDbH-~ameZ$JO_=f6G(^Y{nT
z|G@j(2f+L8AeavS_$R=Ursr?Mwkz1)56mHREJ-A)odixP(wlRg*^i*?I(ujtHw*y|
zC^}IG{p5^pnEyIY1*XkoTA8yj2m4zuSv4jCm}SI(CSbY0m!jAwvc<Tw!Hqpy{61o>
zBl2pni`Pyku3`ch-jP{O4<<vjddE<pAv_1&0?ZqtOAg~c;9vkcNQgu6HE%~SH0^9j
z0r$yJG(dw!X!aZ$+z*J68~hT!gJ5_{j}dnR%eyebB9a5mc(euqFd5VY+&p|qLgmsl
zW<SB!&RlX7#WV?TEhT7r?@Yb*kj?`2{<r`9AOHEE|M7U+9)0Ht|H*)FKkNAb5zrX#
zPkamm2QfTUF<C=$nn#-e6y&gIF2&8u4v_QBA<wMWkOu`1(MEsD6D*950qTNBY&B27
zryZ%iJF%U3Y=pTVh8Hm%0;7+WiR1IDQr+M<2HApBa(9Md`Es1~qNs_5TPLdoiVpxY
z=CPxX#X<QsFvA*VXrrg*Vi8DPqY1VHQL;Hk2eD|1M*0#fWEZWh#uY$=4LsNr2<8;n
zrOfR(cI`OyQcG|n>uB)|IJ6_PYZW#Kq7xIqv?Jw^I}Fp}pA0mdTShMnvvlqqus&FN
zXF``o;Qo*%+`s+x|MS27=YRb4jbgaT&yQj4uS3HY8Uk(myDBJEtN^#s#{@cq&7RL~
z4u!cx)x?t~PFyVy6D1EpH&Tf-Y0+#fD476+X&(O&{co86{AE5T;A2{*E)CqJbzmI3
z17?K_av;M+;Q-=%zA?MJ<WJOLLAJ<)NOQNhk@gJNUWC#-rs!xV;-^p*LsN#ajngaz
zE|}I48edH<9FbjobOtd<+c3mKlGUei%K3bAoi9DUhGP$hk>?ENyQEXFBRR1FWx#?j
zhCVj0Ss9zt9?y=U12WiaR%pVf>M>LY0&*E<tP9cHPO`D$rsmTTaTER2zW?Q?ZEJ7a
z+kgAZUmqJVwh4U9_xI;Jw!gbBYq&WEIH<J4G<vXh8X)Mfw^QHGVHP%gx#F8$(^WtX
zVdI*miIx>Q8#_%vd~^cd0YKk>`8Aq37Q?0nH0iVpp1^YhOq)U-?nYA1X9(RDPEAWL
zzPyK1xtE&yY+iF8o)0=qZ?b8E;wfh%fJNaOh^=G-j7?N^m@SCBN5aaf+HEnkV-6)&
zCQcY4akn$D#zf*{s&EWltkwq1<{xgu;N}iKddm)(fgrRC{D9`V7^()=2HJS*Zl@W>
z7~e)zoD3z3W3xE`*@um^o>jeeJs2Q_9!6ZfMbq6J(tG_#|LJeve*X5;+un#DJ8%N$
zKPPaW?*kWA32Jkm2RaPqI%bRl93ihufH3rEks~TthKBU0WI_c?W+a#XO*B7V&XOh!
z@_}~j@AeD)e1`bl7<JY_I_*MuJQaYRW{Jc_M2`T)jZGwBxyZ8CSUWCKI*gqv`=P)s
z-L7E|FoCZ0mdqG9w`9!_gIR7iux9CBov}!+rrC5&daJP!Oy|N_14V$=LHa$W3~#f2
z1hZ+awxoawZ%r6yA07yB%4oJOJ8vDt+u5e!<AbB#$NM*#2vu()-tgWuY}mtANFE$?
z5+@O%2^E%2xMy`E+!@9M>fe6)-~W&Q__nps=6F2u)BNQhzkUEDopxaSKmXtN6M*V7
z<V?B^FR6q`YD5IUPB?d9M%S8{W~fk!Xqc36027f^c;;h(Fc>Br?N1gCZ0=XtHcU7+
z#yHLaVbVB{X`ta$;G9u^V>k?|1ccIu<nm|m<cu7UJC=48-+(<%<!Gw-$gSdnc983h
z98NQa0J(F|kv`Cg;PhNeL9>J^>Q@xQisojcbih2oV+8;Y+ptfOGh=#dEiGbUYz+!C
zMiO;=3>zNiGx<FTnCLeiuK{cWcsG_-VnKrbuxG`M#jpn(hQn2OBW;k*578iv_QD?a
zx5u~t{r~unpML(vK(w3vU&rzO;d6wr#XR3X-a(;>1?s~Hu7(yY5lm+0onv%cA}gO5
z7-=VSbBQA9f#kX=6Pr<6v#@B!mq-D0%+}R$060pSK~q&8oyTy~_i5ua8ydVC$6ydU
z^d{oNnT;jsV%^4D+}x<lNq@L5#F}1H=N`q9MC^%!!!iJ-VQ!W#l@43JOcDjTMlQ|?
zsvw*3ZhYpbBq0d{nC9R%Z|vzN_$$VHF!J9xL|Xw2kn)_cbE*YRB&h=qAG=KuhFx}q
z7COcc2nA`b+Ce3nZ<1ZiJ{-b~n4)I@refi#;1KNJ9^d}D^>5$ac&BfFA>jQv<LU>d
zf4q-*j;A{s%TP3enb}ASX3TIK*mjPYCLadDcpNOF*%Ktgt2b4}6&vex!U~faGuDUk
z#yQZZ12YIT6Mzgg1rVoW$CyEXGkXpPxR>m*rz5dlMk}}Nq|U7t!NCAi06CP0$-?^l
zE@=VRD_i~q*Tqu;TRKl^YI4;;Uy`>h%@UKkm%qn|L7!`18(k@(Jha;+dBYpPkEa3h
zwm&XxNW+vd)xk;GK+G`40FIubf$Mz?XPX=FO$F29j64pL>2{eE3bpmkIxq~DDQ-Ht
zgRt-Y@p!y-#_^55{pH&md_T?;_#oia^L>1zG|-u17?`U1Vm;k@@0`7<XW-t-le_7D
z=-`{FM5KV`Ivs9Y_G_f4c0PE>(WR@j0bn{|E?*dT7Q*;YFc}*h^9(y`1B}@O=5Z2I
zEL#FWiZ&iZoJAOlCJ0elJ47*KL~+Ba4qJ2OxIv1<#6s0Vf(?RL5J4+%ia%1BHPts?
ziai->d0V1UTvOma7+wkfEpnPZ!3F|>`^Jxh1AWAK6<;h!pGjaUd_a{tuts$NOv6b4
zjy=st3KTXgBIgPp6@XbnlHVA@ITYi-zCZr*_Lraj{<j@No!{O%#s=U#U>~OE`}r{e
zdV7~s#{?QRn9kLiQvh$<gG;DT_d|R1`5CZtGo0gTGg8SEP{?*Qi%2bAHRZ;W!CkuK
zyG6Tmjt&gR!`MHxQhqx$4llqAS?ASZQ^b{p4BG89D$+!z5~~z8I5h~U)SLO*6)I0&
z`Y=>}1_&*4U?>>IU0=H#uxxHbgGBCiKn}D6#zE^akLX@=cx`P0%`q7#*cd|^D(6ST
z5P~Hg!JDByc8v2m3<E4t9}+pBGZ04C`Ajrvl%nJ`G*I|v+UPk3(O@Rvt^f<ilQ{sw
z+hc!w`{{50`QLwz2>}1-@Tq?O^2@LP_b+3*o&X5`c&0RmAMxC|!=M8V{joo`mWkl_
z*bzsND1bJ>cplI58EeZjRrH*RkbKq-H#1^ko6HX<IA}MPi8*k>40sp#R04Q(47Z`7
zjWb|aB;;$k1QR})0oN&Y4M*PMhGIitLhE@Kfzo@kOz2Co1uu+@q8#ND4)=?T+@mnd
zRuVZ|YWx<BUoU`(AYqA=8_(5!s~yKC;BGKFfXNu<fLZuYWQenY`}U1-jDNk);TS_Y
znBfGEJWxPH15fxPs3|!Eoi|26!OjN<Rc3%Zb@IBrb<8&d<n6IPzWwz-|NGDTKGA_=
z|DWdL<Nf`Y|M`zGpN9(JZMy68Fryt-yfo_4cubm*zIP@-z^&;AD>&$d05AqRzCGXH
z@$9N>38?0+G7i`oSWbcrz}_`_Z5wHjThf%+Fr8Y=)0$zJ!KST{S)C3TLx^ONeu82b
z)UJ?5s;j8<GdaQ1X%l6}yfz88AiYsCXFQn}C%}!3)e@@Z4R)GoW-J82B}EXBWAn8?
z0T>ocixheg5Im>Q5FSJ?fg`Zs);G835%w>>z`u+eu<-V+x4GkIysMg<PN7a>!Dxh;
zOUqmOKpeqrs+&bZ*$~u4=51%pUk!tWxC32!BcNfnNAL2s?G4<6u44n;=Xihr<sS!p
z913k6ZRq>)tmn-j(Nd}BVJtQT2Ikhc^G#;(ttn_jvRbPrpw%m@+5RGA<-)Dx8Amuu
znsf|5p<c;QZ3G7BFtDkqYh{;ORPT^V;(k^ybG_2x26<Smm|G8N>rvj884%@QP2@x-
z2AGF3Ivhh5c<GK4AwvuO#QPBtr!a+=fmT(`akeVwa9grF)XZNzwcA4KQMKyJj>NfN
zu!N-XOVkaH&ua1Zo?Vw?H<)S6uTrq*wF#gy_P+B2#^aeAXqV0=z=J`(^WNKAr;k0j
zi6^%KoBGR#*3Hw7=lO9C2Sp=}d&ii-My<_3stSA@?VEAi<V{2BbO3FXRtI<8j;7$|
z=`|<CewJj;Q?{Xfp_VX<r*ry;vq&%LEZ7XtW5inr{m84PEL!Z?AYnF~nNXoCugX#{
zcQ{>%(G^q!mcA6WO$0Dj^JNMP%$(J@(h>(uheXn%P;+8p33ekildOofYQ92g^a0MA
zj1^|ShM6O9%2t3#dgE+2MBMc&beIT-GBAJrm<nXo9-AgX0)3UecXa3h9dYR$Mcs;F
zWdZmB@T1HtGR2=i@k0Rat^f7y+k+3;&~S)-JTQNGKR?Ft@sAHzKu??Ja}{*G^$y5H
zquxQlq2mdWZ|~m#5ab)8iUBYUjuG8Ya0kJ4TkC`ocCYck%+1}!$LMXP08iU#_=g*$
zOJ_FNfb!JSuu9kqNwsSY8pA#6IZXvmGc=_ooa1nh$)(iBCki(9Y#1+VIo_H9C^<k}
z$~XoGlzM_~1sw)`Bpv4LVgguRiibQ;9+ebIJy0dihI<@WuW^Icc8mZ@S%jg7sl}WJ
za!=6ur~w`Gm`X03tZ{=vH_9o~h#M@wpBHuRxa8Vrsl&~r+(!f(H{tfS{q^Uc88Y9Y
z()&C5(D~~>{_!#2fBhJzo4W>~%}pV~j42FtiXOfXj3)pZn53hH$7EWg%aK;7ItRdg
zz(lw6_IP7Nh6w_758QO1aUjI7(*^Gl5I%MQydOOEG#k<34Q^&uEw<$80JwtH{nVo4
zIWS6GybcOD5v4vgm?_*ipr#6|{W7L!AW*4>79y>1gluVAPeQ|bYzleJ1OyS<SYbX2
z=JVXkZXpjON%a;iuQ|@!BN*QK9k3r=u|@xgx3&T294;wHZU6+rGCeNV0B8fDdA3Lq
z4E7+ljCm>qF?&zocD4gsiHb2=n7+vl2DrWb^!LC1<tNd9;`r$OTYvinz(4-+&-Y`z
zKgU>?Pok!Pf{pwmH{{u&AIHBw9DOu~%ILj4+SGyhFb14I1Y?ASav#{u6e@bj<&PeV
zpCW0E?MFk|T6h5C01U=DIv-i2tD2n-QUD{@aI8~byFUlyXs$C!NM4VIAzZ^tAVM=^
zq;Vc~7%S$mV0gj6MnO6q-5{t5pp7~Bswx1|Y_J`Yk7*T<VWZ<3*{*$#hwS28F7ayl
z2G8ns$9eRh3~Mt!JNDySvL&Q#tv`VAd_Mv!Q{)`5OVB1o1|TAx0~F~B#!v`t8xEfH
zsfgH*{m$?X%*eZU1`xxN(jP3q_S4^gZUSKYct5cH^RW@^`1Sw!=f@Z}%qbQUZ|5N+
zijp+}gn?<#<N4ts6X*&wYz@ts+aQ6Vz4>7n^IeC<z+1Ct(Yq52GiG;+q3b}}Fgi>(
zP#>$wGoR>ubYS?QY9R<Q$=v}iT`rLOV)$uH_*!s)?8+*P6&%b#B5FaK0!w;X8DJwk
z&)3eF{8y=oD8@!XWMdUDF}FZ_`XGUDI##Q_ZW=E`uGvUW6*;$0gVLZf>8ZfMPr(cZ
zCX}~<zCi(EX6j4=VP?=10L0SJazZuv-CL5HlR6+8nc}uF_UAc3POM3i!{{e|`Ph5!
zPNA_a90Wub?%%d1n1dep`1O~8cMMmY^Ek{it4{_zxQ$pI4z!(Mr%fMr=rhI~=1FgW
zV4T=shN(Ho^8heZ!@bvF;Kwr(V39_FRm9tZCGV`8w`ONw!2VF{HXv<4O*IrFmunV~
zlyqcd1iLv-b^}^czN52STLBOwr2vCkbp_)chYxdG*-1sEl};Z(Dr8Uv1kxI{oCJEY
zXM4#}Y#{;l9|~U1{>Bv~!e7l?HGltU{ecaTfQ4%3(hfi#1-)YA9wu>O2q4UzGk-Ia
z3@MB~Ox8icUbQ4E&Y0NP0TO-)$7mHxYn>jzl=olF&yQnH_i+qU595Oz8W17}5{<Sr
z0|PKl8;ha_S5s*+=wrM$=s8`d4`bi~=&`;ZhqNg<L>h6&5{nK~g|o<#Xf`Ch(oY9q
zst-VeT5_erahw=%`3c1-xLOsQf#kxsP{SH@3zmjN<3c0Hv~nDM*#Yi*M}#RkAgyuC
zROQL;R!WF3aTCVH-(}sdbpOHCBZ9xk@&oMy#g)9~9#u!IO|%EPKovWH4(awhB9MD`
z<orYzSpSsy&vroH;|!fzcQGjYWWiSDN8gc)ow)(CpV%3g)8?3CV21yekLt%U0Gv7p
zI03X%Ct>Q)5vwG;Cx<+Ao^R&~bb@05(^RF_V6L?}+*A<~GJ%riO`!XaIh5HZDw|{|
z8$SVq4fB#sav*tiw#FCW<S&AicNm8eJ5W-|R(Cw$f#_{_jjWh1DsxVp^LoDKC>bwd
z;hsg4uNV+as&)b`^)jk<7I}6uroTiDs&sDt3H}4xGyI#d^|rAM=+I-HhF_!u;QjGU
z8m3tXGx)#*#{h9(+a$V#+n9luRtuoO7>AnWSXCC@Yb<$}y$AD8q&?zv7y#2-+urS%
zbIi){LkU7sO>1N`DAGzmPng0Ku8!4%)+d5D67**<L8pN+&CFu`n1x|e-V#ew1XtpS
zwd8qP4(w3?LF@z{kWj9UU$3psVmc(Z2x9;qIv;CN+`=lMLMNtKkO3M05QLg0)G**S
z0$BD6xj<@W(-f~T0OG`rn079fszZrD8-oEpTbNMV!uq1_bqB38>wBxA=xToFND8#8
z-x1L@p7ZHG6gjZB{qzVSl^r174{(DqcV22!2B}854ajQa$KD+xE6ELy`kuM>hwNQo
zzyzdo!%SH6Z=mHX9&c|R!xSn&U~R=x0RU4R11-8k?y82)-291gGSGynb&$iS9_Sc)
zoT?aR!_+hZFa}fJ?fcCWxM8YP6RXZ72g%it(u|ixV<T?j%t~nIwx5`<oHY8n)T9`P
zg>6x4SeTfbo244uN=dHi2m>eYnV1{3)>2Ok7KBvl+&9d@{Sb|ytt~aS|3aj{c)f0w
zp9AOH&oS|0stP;pP&153aTx}A-yh7__G1S~_TvFSB=vX-EuRLk8QbM_kk>cWlmWi>
zntBn84-m~54UL0l`~!bI=NMj)&8fq#?r^+moIVN>gAP~tAW?_w(+mweX7i^gQ;?e7
zy1svm)ASVlq6Q4Jr<s~taB4k;U$`D$k#z`D4MNpRVuFU2wLl}**U_V+{@}JCiCWs<
zza$w*i^4|9^S+4f!-w9ApLKHHD4m)Z9Rp~sRqH$rziH#%0U4pWyq4RNn#$J7r%(8S
zSNpj=j(7*f&^2#(#hQH_O&-90?9w`Z`Vbuf?#0eBAZu`NCY?XsWYJ^OF=mJ`lPjEH
zHpFDz%{K6HGyv2aVgBP+jE|4=1PHF@$Dz|nplFMV6oj0K`J_c*Q~>(4)4Ji<c>?2T
z0v53B58!zW9S%&3F4ukcbztUhQs2ZC9K4q5*2I6J7{LcYeQGg|1uAzLklf<ECX0;=
z9RY|kG6JbVXMU_l7!$N`WA))S+?G7)eF)3RsVyVccuW-*AvKa@3j&k{YKLmwJ_5ts
zcy-3EU}2s=8ab^!a&9MWMLGxu<}M&A3j%#^hTQ}Z-7#R`uT0by2S?8Uu!gUpE@2h~
zK+*w0+en0Z>bN=gj^5BFPR!%^VGb6?`Ee|A{e`8i0u}(ItA(PwjU+&7AS|HH0-(Kj
z=$O&04xWLSD44u>JTr)NT!e>!GN3|El)Jm;CP74SE5&&h6{J#5q(!%8pJ$c>2i%#A
z(buR;cE-|ebDdVqfSu;UHO6lY%;f%py7=;GS{W`BF*0PXdD}xQU>JsJ%pYu0wt6E`
zN(-7|EtwWsu>9UB6@i10UjPiWMITm27)rQB^qRv?J_GH*;rze=&(N#4UIc0z>6OEA
zUNK8FNN!U^L0M~$Z@u$QB<Q7t5UOxPZQ2~q5Yr~uIkgD-8uNaTq=^LDApz8-q^YEU
zv>G(@>2A9U!5tpyaI_n>tJU`E#su&qYW<ojAtJF0Xml;A*G-Q|L#4UNtzu2alMiBC
zlvF@c5Qsn=yhmA#IKwDL!gf$%4h%38(k%4~9K#C-tAL^U!Wz5?42E>J!+bb_g_%iI
z4Sjf}N7vC9aqq-{{6XY|CvO?zFzvCRnVLYSz7I9ELxF*(V#Eq%Zm=Vo9|w?I@zbEZ
zF>s6^oQegV(2$2#aK9K3-nVXooIy-McM#h-hUpwJ%L{ePuM&i4yQC~*pu7XXeuP3{
z-G-U)ujP$|#pI7dMA#Ar5+#>-6KM}-Yc?W-5TJ(<G%ZL&#SQS(FAnohV{sD0cDmhL
zoACo6Yseftnh5|AKN!KYGgljBzHt%OS=lIr7t?GeyL{QyLb&SIr^Q??h$JlU>pwd|
z31Q;E!W;tZInsU^AxRvxL>D0O`-fIDrpGhcsZ2Z2AuPNJ0rM&EL*?A(iO#U6Jc;`b
zCJFCI6NQ~fmurQP5z?KQ7Gr>DHD(FK;jN)83dCri5J3eqhHKqThlY6r0QO0dEzF#t
z5NK`_ilz$TDD$5d>_av%xY_ro={39G(VyrVJU1g30K}Pp<gl`tVR}SSgps*>IH39s
z*ANFts+SX)-`%n%rx_w@@_^w-m{S74u%R<XU&MjVhywh26?qs}aplHZ^HG7#+(sls
zCBn>ApyFm=B^F{rjL*cS24j{XVqOp{vAE$o+~&K$nFM|W3X?tYh#PJ&Jb-s#Q;sn&
zGZxchr$`0Za30T*Cy@__ikFbo^{7BmmBgxd5Hk*Z3t(BmkgW+kRL|y^XY~6^tL2t4
z%%n7BQaEGtXpz{+2sPJOS}4?`73XRo4Xl_tgkg^{9#12O_mLnI-DHJg&#K_dBdQb}
zlCW$BFy~Pf0TDm{nq#^#5py6pEYZSDDPA?TV+>O#&R#^vP*wFzw>jgkWPKJWd11Ux
z#;PplC3lM+2Hy^$kQfjM+$5&2!e1ndN(e|?vL2?OcKlJ2)BO3!_uYx$ZinoGP|LMv
z6_~;;iLWfX2q12Ku*F7u-KFc`Ctk$wx<iQYWZ2*X+<^E>z4tBkHF3aA?fZ+)eq+xr
zivoVyX}~wVX2V!-OupRaAEDO-=v3oSrEdf{v+(f<ZI7PEn}Ncxk~a>0>HtwJo0NXP
zuySR+ouvX}`vMh-5t%EYCGHF-g{e(d8>N|q@DO#Kn%3_1^#eC3Mj`z5EJ@%~G)9n!
z%_&er3!la!J|)?XRbjS>I|LB=SUpqNOJV(G;0${Pd3g`QxeXt!KYa8M_}mXP5IW;&
zVCpOl6pyfd@Y40Zy$B)480DfR(IUungwT<HilT3s7$1Z7&crFTp>`}D>;dd?5&}T)
z0)-BPo$kj8$m98FK|C2q#=<o<An2|4jdONU#s+96m4t6NwirwxxSckE5ORbg$~3Hn
zXK)228l|b}ozO+OAx6w<o8+G%SU0A1mPEE69?ipiV)*T77gK>kbxFcg87*ds?d-w(
zDxA;(BLWOW4sSX6T_aseB_$C@S90E+Ys0TZlM}^6ecJDxR%^AUNTr>?IqcXt(H!|}
z?Hc<8khFWi6!i81vv#VFMVJjWx6*B=m*TXRrBq{hCyKdvJ`~aipkumOxJ%-I19a?6
zk{N&i(D5<UbXew^05pc4K-i$oqAX{0ChC3b0F13e`O}@)!K2{MsHHFFf~SNcmgFRy
zqQHyw8%jb+p}Z4Y*phVjvgJgf;7JI!voVH3UA4kGp?EO^T&G7r;F@K)k3*y^WL4w_
zcd#3~qcz0WTF6oTqAIZ2*bRq4kgmA!oO@oIm;w5dq<tAU3SmowCOv#AAnn*^BhnBz
zis5;Hdz43YQ>Gixo%v~9qi8DRlqB#~$^4l6C59+_+-aG3CZ(%`N8ojb+VH8s;6mf}
zn9Yd&r|US55lRffWWwz$3dL?6$6{o(Qt>b}^la*f`Pk7r=LD!~GmH5UFSAf{0KG-o
zscXuFasXfr3#^kO*%5$oQEb*?j;mPBB^t4vC9^L%fexCp0VomhfDt*GrsPY%g1`og
z!|N`X0&TMa=x6SSaZm|D*@XDq1^`tLoxfNhxx6hYzUGcU1Ogk9%|3R+;m30V?Jw=v
zXHR70j#LAki4}_B3ZnhM20RWXcr2Xs^H8{}lu=Z6Y|RwtyirU*G5XP~Ylv`oi?*X0
zltO$X;DLXI_2YgV9<pHo21aLwn>d9}M=l^c5?(l|H%&r-L8pe#iz&e&(!!~S4UE>N
zJb@^1ky?U%0DXV!z%V;w?8O0!*<y`kNa~?QQtAST!q}a;a<z+ElM|U<WJX(!Twm=3
zjps>3*^l&6%0Y6tDE;J(;m);F5V4RQ`$uP-vwk5Za&EocQNvrr<S=)MWHTux<Z%C^
znhI#Y#NDFy+=d^|X-_=x*pEF}WAjZ)fH6S5oaRYU=snCv57DS-Eru@~@B!;io~0BJ
z@DHOmb?6)pnjBEAKp1?+%1$PWt60EFz>((T?+A$P71s<vy9Cdt9c71*M?7Yb)rSrp
zR=65Vj}poNG(yAU8K=M3oIE;cd;3d=A8O4mqjbsrh_6XEPuhOYsJwJ;uEDuHVU(jE
z8t@wfxZ%QPt$NqYBx(-J;m0}>>o?b)LI6<{5EzF{!VQYFOR-k~8r@><LWrZZCy%(9
zYEH~A%!=O$oCC9UVxMdn&pC$S<2+kG8dGQJFZ%-`sr$oX45&%LYA0{dQF)>$97<NX
zZJ2$*I^ZPC;H`s!Q_s@?IarukY@oMTfET(CR)HTo!;ULV4kt#mrU4&p+Thv*f#F@L
zg@QG7D*}&!!@}1j<^h<9B&h*ic#*0`LNu}6GVPC!{pc|8)>6Hg4DRY6E*A=bU8Hei
z9BNB?4W&w9!3xang^C95AvVGBbC%!{jDvf`L9w|rKGz^Pu$-oiW4DpaA_iYzbgY~x
z=dw{dBFYU_=)0Pm#tdK9=U?@cJ_inP0NvPis9L7nH=8v8@^}-#kNH%vs09fDaKm;o
z(#|UUmLeaLG6jr$d2}Pl2HrXvDfB#ZrpH@kaRU(Oi--Z(Anh5Fb!ne66l}sqtk@59
zRz9%tgs{b7(I)p3ra8|5Y(03{12)qA%seRFU}hNj(b2PoaNg)n@sx*jc%?=iBPQHD
z$OG983Tufbdso|!d!`&y`kx*JXMp6FiizHr!+|fs`FhApiTO5%m0DG{P-#FkP5Fqz
zc);M0$OhKbRMNk-W!T`Q^auVpY=9;p?m}==H5+a$;JtTd+L@ZL$lJG{w+%S;{rpu`
zEy=N&ea#WfDDq@B0Z?(2wr_K|ij$sO^1s#2iNWIJu_XXDy5?LhPU9_dNDP+JDF$-1
zIe|~M9UTfgf(jm|o*E_-LY@-_@aYIWmPKp-yb;?Is<P(fuxeb4V8{+U$5CTu_%vda
znKUuU-lM;gGr)G&cesCOCFqgDFYl`A`Ehsv0E>3_!tngeFt7h+l%2fng-|#~#RjWV
z9`0GKfM7TaK-R!_tg4VdPI1us-zjVXJqIEHbnen<mnJNa_SoOPz4eWK{PKU}SFG=;
z7&`R*d7ME^C~=-zH3|987A<13ffL;@p6^o)CV-NhTd=T?*Xv?*EMLJu{oI14zy>8C
z<pLe^`4}B6Ou;}_zunvt%K+~D#9<D^2+5&DlZ{0I@D|A<92ThnetrHDROFQEhgo_o
z<Rn|MGo>@b7MZsOfBAPF!Y8@fyh=oA!v>3ToWLG9I60MtGgVh4iOLE-MN(k({vwAP
zE5Imv<YJ%S;%3bg{x-4eHE{m(?|r%*I+uh`Oy<xoI}309?dR=V-+RZ9{jV6IKnsPA
zsmC~vIn+Jd45gpJmSnh6rH8-?YNtZyb84+FS6KP+=KtHq=qv!uP(Fy=O)Z4Gn87Bq
z3+QrTwkg6J02Le9(lfyL+I4rNoXQHQT#S88a6AC)FesXt##p(9(d<d^%22hIV{`*O
zo(u^{SYT>TjiQg7|8U3uY&ClggSwaD8WP@(?3jRNww9;edgi8)MdUMJ2TBwbu4KF<
z*<U}>Eme=L(CZ(1wZHejtJ-sX=;#>yQ2^(@KORDmzr6kQ^Iu|@)^iw&^@6F-`Rm6p
z)L3hs!3MC`pmWNxDufw!^cngIxqp*8jL1t>HCJ4CoS{v{NV}(OB`|s@ccQi+kl3;7
zfev6l9DUff1ywh2Xc$@-LH7PZXbJwf;m}kkF>sF+J5R)zo?Myw{Z41tkZ9~HYyB$J
z4xL)htAvey_fZf9!cLC0?DLeZD2ktR%5K9F$cgGc@>$O#D+AkJ{c18q<cb_mriK=o
zl>e};`6+PlcNRP!(kIN1Ck_Q+>p%Vcm$yfQWBd7UfB*Xidw*B>$NN!7%JgIYdg`Y>
zWX4U>EDe5{VLBYD2)SGXD<tWnZSisAX9X~lg*h|z#CKyZwvPPJXJrQbz<%)PjNNdc
zO9ym}k63{Ubbf@dMRo_Iv{=VJ<~h$J+Ec;pA)HZ(m=uBIRbk>TO1e0uuErTRvEy(V
zBCp7y&&zp98r0V4YS+(bI!=^y60i=&mu`Tof|`W~E50v@0!1XZkkhLNCJxtmV?h5n
z`wsw`|Lbo+@Y~sU2!Eiz?LYne(_i0ydVAyE_HTdt_LBh5|Lgz!hk@942+YoBIU&f6
zajDTSUNQ{Hn=}A2fko`5xG`}{luVcFWHOT2Lz_FBVZ=~kE}oL9F(ee-i()mT69XF{
zdl%^W@$=9eXliY54Fjh4C>McfA6=hYwr#)!Z}8EBLmlnEYHbP?5}u$h(u@rN6AoaE
z3qK(#Vo=2m!#?F7KAStm9isL+oKl}7s5u0~h2CU+znmjxvzbA$h@%mb3&M8IUqa=i
zW-*f003ulXqPhG#frEcRpkqul;y3)ILw<gHMBMCeegFB}x4kn4{q^Y|2S0#@IYw7=
zvr(K|T$6%nIai&Vg1v<4mO0-pA=~IhRtn*Z2*wA{0Uy_72mm)53J)p`n6b>@r~$O0
zecKvl-)3;fbb$8<1IN^;G33gqgasNB;W@%ash@kXdDD-IArjH<RXD}jCON`!Kpj$D
zR4d_}>$>sXuHys5OcMZk`*O+WqbB$3%?9}UA9TZ#4~Sigxi-GAG>X1p9^Y}cCGm7w
zll?eZ9RQw~dN?42w>Ja#$4}o##K(@dS25xW^f$qT&*rKjjQ63{=+DL!vxGh%c@X`x
zRx>sx2*Xrm=@ozhxwd|@>ByNa1zK(5Icy!zBm~~U8gdmBL+CN>4ot?|);dJ`bWn$%
zACUdAW3+R$iV$bZK_>xAH^pSzVFMs+TU^`5u%V_p%uIzD#vP_GS%eCga$SqZD^=By
z!|0{Q>zQ7?4i&gY&PHs_zJ7FsyHXo49r~)RvznQUppy&OuUdXNplDJ^@^9L8_VSNM
zPLrK<z|GC<Aw+L|YhAd169s;mgEli&%=7unFYhbjr=nwLV*a=UkVS?_2o|f@oZ&{t
zH$PesQW516Xz-+~7|hfSrYfCneVHU<s-3|xUZ`lmUX-p_(xFWg_Gz&{81^l=r+ou;
zKzup^d%#3ckTHk>Xc)@%(F1#H2$oPZD(s0_MIPYLgHvpn5Uu=?#Lt&!pS|@9`faG3
zht}w1D(<tfeZeIFFR2?`>0;$ab@a)L{CVckLM5KFWRDD7mqV*VM&vOd^dpGz3LFsR
zF~aL(Fu(n!zdiQWBMEu_tNr?qpMH`Z=Ewi<fBy4hj2T7^fU=@HFwauGBaf~j12R)~
z#qQ$}0kNlF3^3f?oZQ`L0*jSwrrU<_Z_K*JRHI*zY8i1jh_sElFXPkd2-~>@5wLIA
z@a=uT#|B@_G6sk`0H2dwf?t#8duMqQrjADY)?hZ~bl3A7|I)Pr=E-w-@q{+UWhNgv
zT9u7J@af}Wg>F_f-Q&=@PN#cNe0s(Dx?xlK53U;qV$kw92b|!v+GtlihU~1acU_HD
za{tjXd<D)>f$$)CD!Tmi*!RcYwje`J0ONoDPi%i_!1(o_|9n4hZy%g90nWhOR(szV
zI-+~#w1#7I;7S0HP%3_%Fh@X0so|6gJG>-a)nPhOLtm^!gl>#3jdrHiB~D2)MT0qE
z%qIY}Xq*tG;`IR^S<c;Gns`Fyvb?7mL2s>#hc@&aA0OMhWoB>y;AVzIB0~@Nc5lqX
zDYZV{BVL^9xdo*mTZ~DA_v%c%)J(L3#$cg!QP6OShh70z1Xc&Zh@E`N7?zg1UGv8Y
z2=OPqlOGJ6=Py6S*8CUHqi?)XURgumHjH6v<DGo?$9x~;!W(>^@n`^u^7cf{a>orw
z{n!C`P}3a5v#;+J@y-CY^ZQek1ZZrd8#|z5miDbfjp4+^V0p9$_j_&ZHDqT_I0mTg
zpnm!jJ~2*W*wYjON-sDjxa(;eFt{O<b*yGD>YojL(+0@M`K|o`pM$fx&Al;{z;gm_
zQD&-nLdSuaR2EcviyVpoF&{r$vgl`SAGdqrdAh_m#JCAe0-q!7`uEsyJo^kv2&TsR
z#oOwSIE<rn_48LL_**+soBB&GOgy^4&Rtngp<cZiP5Ti|GxF8}o8l?}0Du5VL_t(L
z`IzJUc*^qxV8dd(OT%)jjwa!uR9m1MgW@d-m;}St0JM|QK|$9z<EJGCI1X?J#>X&f
zqjQH%OV+rcKiVVZIAN@&QGXGnIap8q75IzVWQLv3sEb1NfdKe4UQB33kZX>*&5cw+
zd=T#&CeXeCyaRl=`*S+me&F#ajz@q)w4DtJ!w87SP0S#;SA0t$d^Wpbwd4M<K0PTf
z*y78-jYrqA>+GpV5S?{Hc?MCmz5MzVA_KEsu#u<eU#-JG0(BY;?+$wF=&W)=CrKyz
z=K53j0|O%1-++Jp5;JDf8-W_!CzLjG8)Ik<1IBP{@V`J}e(ki##00t0=+q8Oat1iG
zL6J--003G*rN6KNVS^oWLKI_9(v(mDdCJBr3dwmwY9o>&?Php?9@u}~!jU*+`E^VM
z7gv@v+Eei;)kDA7smQbjMfc}`0Xw#iVI6E^KAu2`GRH-B25dw#f|ZpWAa*BJ^B2t&
zIA63t|4snndQiWOrbMnw?VfOldpi(45f3YuTE9UuSV#$#?Iz-gd#$2w602Wa)pvmt
zw9yZA8}d|B>ED$0_lN%Lqc>`uw+8{<-{uDZ=!wRF`_QNP`TTevt5<vjSe2OEU;B91
z9y<U-x7s8-8$w+<Hq&4?0z&XPK!DqU&}gSvOXETv37r8bC)9W%7|CNjf*9@_W~Wty
zU<VUoG*pxT8)2{uroGtiquT@nPt`#fX2ji86xd_3M`uUF1oM#;hRjg~&NkeXEW#Wx
zBlC3@{J9rJ`?O?G8u9Wv%YO~D%pFAB>2{16D+N__x&CDW9e@227`H$CF?Cc`l6Z3Q
z)QCXFn3LLDYW_XXfBoZB0InYb^)Zg;INqNh&$XQxTbMAjC5<lmivi20{<;Ih19RR&
z@E-u8|4iK3LTJww(^x|EfnabDp$Cde3_C^)w8=#snlHQ(?qY@m)tU0;b>o&&^c&~w
zjN1|I2|hu`&Kjs61O{mxGuL<#)<XFiJ=o<sxu4K^Tstdu%m%%>xtZkt8(C-!X@WTw
z9l^;wyd1C1BX-<@9I7tbbn?F9E^_*;UNuxrV`Qw=mh%J6hxQH@X7ohRWT!5?b88)d
z4+q}fq0i^X@eFy25T1fgpM|GXi>caF;zkJ5Jp{QsyXNtgRAbcS6bbRfZbK&#mRDZQ
zXGhm!)s=uz7Scg~2m)Wog?NzKkAM%Qb%EO92E8R0F>GcQdg5o3T^cZRULF7@e;a6+
z8ha3UYmbqMCc|_bXlvS}{y8w8$J17MEW~Yt+0azbQ7{{f+<tS*q2KC;O5{i`L;z11
zd+--Z9!0%II-|owrEEh;ntRYo8EjdzCv;prhF;ls)V@EmDK?xB#Ae9C)FE^ie)ISD
z?X5wNkMVJyh7O^7^(Im?I|8+}X_CaXhtzhXL^dMJQ@z1V1O}cID54o#+JqQ+3+41M
z;79^-1PX#-34o<-%xIIM1wjCq8qmAQ0Fc{^8*kuV><`s&shpy`F=IZbS~0mkTGAb+
z>wxs#B&1Op0evdZ2D($9?Rg$Y@<<e7+!-CFY~dp-sS|7T;{r5SKvm2$1KuWEs0A`{
zH0eb-{f-<ILqJs<#rBdjNLISMZXw-Kl~b!mEWv)X)!FzjV8+f}1O#v%9+SL<&;&kw
z!aqL7xS$%aiM5*;MBx3=&<UC>7RJC7Giz?_#-w4y%giDiR;MWK^%H^H5QxD9jpt4w
z!Ic6WW6xE=LpSi)jVD0fxWV<o(Sq6pfVQ=G?8GA$1I~E>+F^?E?1wbiL;>+OL+X{g
zm`$@eEJmYwHeg2a)m3CEJ(wQv&yT!LcL0rRGvXej3<66q=eL$K{RY}k4bf7H$2GZS
zOKLrf&{@e@HHrbe8#Qp*yB7~*0YU3wu3U%l$0VW(oS*-;Bkkb?oud<}0XJ?xGfw~b
za6Lc5uT{7S)YRF)eNy72>#WKmS3*J&OT|jCK~-~xmb1Ldaid`L=F0LU0U$N+p9U~p
zsf&Vijd0ny8J=4>XYQNenI&iX3j#4|cmOcGn?7SZdm11e!$BXNyD-h-J1f%ScsQob
zleZ_{jvjRuY<Q?@@7bUuxE=aD>T_WPp^X<mw2vxXet)z1iVcYDB}Do`MJ<V28ktJ^
zzPei4jxpDua`EViW3ums8*;lCy8RpDM_=u4KmDBcu{%>V6Btk^_5PrT(eZ}&bC?0t
zg<En|Jc&X?BqJKmH)Y7-<~f=Sx6!kcU4wGj`!hFosGnBdP*4g=T%y8{f$E1^IG?(K
zh*Q7P6+Ak)%f2<T$CI4U+2G_uOIR-E4yHdX%wNrbc>o{@vUsJP#(fkj)`6j__RwQL
zfw@XF#poJoj{z{%qvm0P19BJP8o@%na`rz}Rmqx92r0Am7Yw+f)vU~_w^yMFteFvt
z<9<a>uovt$L9?h&KVJ5=pML&X9gW-eXqe-Ch6aSvd}bO-Xl)Q1hn8ZDPr+1&I?xrV
zI4*T#9i`inlZdGvZ^iPefuPYHuJNTk#dM%*OrsmnOS>U)X0{tCfx{uVJVe3Nstl#j
zin5`j3Mn-M=CJ1s>@SHD;z1<tX&nX@v}r-Q*H|CPhPLyziO0ZT<)Fhbr+W;JX*oA%
zAKooBQsGPf1^7XR<{In1Rnrc@C!|^qNb0=SEEn+wKEG<+ShneiVa|nC+y)#%D*T5c
zr#*h&4BUkueV6gEZS$Bz6tnxsoX-YK&<2cVQ1Quux*s|Yea<Jao!&J~9-w7P8QD%^
z>UMND)D~)z{2^u+*oy*0@S}2?Rf!@likHe(_mE}hE`~f@xp(2&s$3rNl!6nQD^eA(
z!+dI~asmd4yxTZIAprw(dWjZ;;eOZvY^cPO43bEZDb$kuZeZRJC&;-W7LQw4*;d@V
zGAdqQ-|s~NW0bZ)1ze=y2Juj$GcT>vwdqjoH*DeZY_Qe!uol{S_Y&@Eb|qo;N4OQO
zy=^xazy?gB=uB3O%LahPIN@$8K^q@>9BLyZ(^7Ry0b3;#Ll}XbI&kKo5L-Bx0C0D+
zOO7Vge$!TgdVy35f`c0~f>)6uXghZyhA>l)xG|ZEDmVe;n3ltashXJrs#6V@c!S9T
zG%}X1N-3az0TW@?c(Z2jLZl_;Y=QVh<<0=wboVC!bBCH+vAE4r?)%FJLCl|Nze)6@
z+b~jX4`KZQza2B#a6rb5q|_1tZZOZ`m>N2{<_YevP!y3LsjOPxzWF>=?L%ivms>Q)
z*0;8Wowb9Tp>;bH&<K{i-BC#?wI~6#c<2O65v1ECl#J)~GA@8@Idy>?)I<8P_&U6`
z4pW>-P5?<L4hr+eOihI0Phu8q0cO|}5*!&Dde{-5Qb~)d;9i&1l>pEb1BR;_xN{@q
zm^c+@&XGQM0=2nh3`#4u<c~AabEH#Vw!eSlnS6!!XCBNF?7nKBGZ;$e)j6lDQFZ%9
zaW?_#8jW89cyiL10ORi{hc$N5zEH&e960zZ8wPOrhj82Qe(JE%Z^R%U@R{Hw;aPFB
ziyjQ+?k4et^d*F6)74NR1`9+qtuYs$bg^hP9|o1U0um2=0@7P2CfF9}qq__=J6Ki~
zp!rNWFev4e<pFHiG0zxLLUgV(v1cbzfm@_Jyf%hu22%xNYfJ!QpbJi-Zea;dPTZ$K
z!<fVvoT=N)Sc)~2iHBnWD!)}+`TQI?R=vH%W2(QU>C%GAT-Yi4JD}JX<)({9Z4sNC
zy(7FcU;?977;#8k!pZ^VkKv|<fBqdhM@p{`pL5v2=sIOO%;u?X`qI9=EeYvskS?sh
zlXPw<CE=QyO83GbU3x4V3GeY@HyNvhfx0*W+`(+$EGK8!*@)f4lxHF&==!<UMlQwk
ziRTbI{r<$6jPW|4s*htKiUo3GHGo4n)<*44A{>y3tHX$*hwGZm!LUphFx9bc3_ESA
zrY~Sk=yIyo8uWM6iu}7AXN{XDpZMjcYltFaI%WeskiM6nR#9+z3Q$NH#BFc~f}mcm
z0Bo3G{*f)G{ZoKBrU9zwbl1>!ID}hc!L*_7m#mEGm^CLle4fJ&bq7>!R8#&Yr$k`k
zY$9lgMsaO65RgkQOV6Fw1qN0zqzP!fh+)>#TG(R1@5F#$!+>O2q*)T!X@QKz2KN|%
zi3T80tI3VY95f*sJ{tXoV~EtaF@@NLyBkx$ewv!D0GzX<5+G$bNN3#C@oVgqX3c-N
zl5fj1hV2$foxB$ZG~z5hrKpBEJ%Y0)87Kz*J?h-6)y1KZJ*o=J{AYb|oWg&obNp*)
z9L(UB&2a1x7T9@Gp3SBj^+RZ184kBZJ8XDdnhm&;Aq|-_erf|J7^mSZIn<oM^l~Q0
z$gBb4rh(f#y5+RMqJ6)V9$O>8#%^O!%TbHK#)7d0omPxfgPQT>1_a}juQpe5S2(c8
zcF0C0Ady`#HBc07nc3%i)RpiBoj5?O2J(7>cooL_{kRDMFnxV{0)a~Gccn>X7)uu`
zIj=d+I1&g-+%V23jr0cQGpTIg6mKVlY@)%e{qco1AHSR+7AS0}4uE#JwA`dAjW-Nx
zVKDys{keNph;<&%+HdkST8-M8K@AznJe5%F5DpYH$JkkoMMp$e!vM_HCN%Uggh+Po
z+!`d>_Y>$mO0HxMU;s1h7z3Cwz~Hvfuyurto>VFdA<;Npxm*;IweyKVQ)D`Y3Yv*D
zegWw0S{4>*i2nezR=6;yMWce^sn;OzY9PG4d{On%lP~i4+0^`&XJR=^g}xU@nCg{B
zo3q{czP}eZ>Su_VL4#W$z)lU{;HSF#t-{Xl)-b=tjy)IzPTw##y0YH5VJ$VpumrzT
z{-jRY`c|2xL_W-gdl)*HoG_^$@dfpC01|QY=#Bc)1A&4?)7f_FLUaI|0P%GIM2CF+
zkvJ@i&Hd=Wj0t9Qk5NigHD*`A;5Wq)h4tCcxDN(GzY)Z2#pB`-BH%xi)mybC(FNef
zVFY%rQ2nk}_SwSas0N6=lxQYaN3C#DgXe%u4B}rBhKL(;vkd*+z*+ycL}G<gQMfbC
zB2Y&=!F<l>Voo+#_ddkBuC>Ah=Xu@SiksEq&C$CIqSCA8_(lR@X#}r+1_r0rY(O|(
zaRXf3k?R$}o{l+^9@&GZtk)-7mUX1rFkrI~(+7Bv#KzNcTGrzzR##*!m#=$rQJok*
z`tk;iCl}ej6*}ZmY`5uOmKe7XAgEs2IQREt%T{k~d^c)x6l8{#je|KP3i4VN6tNW3
zO1Hz7E`-2CTj2&8{chkWNK%ho&>%f<5`U+ixxLLX;D+9%5qz4CL|$hIahPJ{!Jagk
z`ZD=&?M664nznIFgK;DrR4y1yEG;Ziw6s*F6G(zD3!NNX@W4u*+QqjbO3emUOz7bj
zw~SBnd724$ki9m3gKS1&Nfrm1$jN{-+QH37+DdblwuSFsf*@E~D+6XcU}kVvq|kLf
znuvMhwMu^nUULAkY(5=p_9FL`-+}9`{Q7p^k7gCwH~{k72Zd0V+Kuz&g;xD>;5<#Z
zbW06M**{`XtHWXl1RFPESepQx4Z@fq+hZ1K$GV$qL%tnFFCi6QkD|$Gv}6ea>59d+
zl6qC*&;$?V4@SCM_?#?z+s;<Qp43A=J&5Zg?7Vf{M447L$xG}P?m_o7z?6kRg5EpE
z7&E%BuwJtmu4`e37Mc-5AjGi|3=LK7o7T{hTm1sy`>oe^xllC-zI&caw~69X8AKd>
zzfYrp$Q>t@L_%<f0iTIdz0REeao`;P`WfL{3_y6>xxw|E4(c*_M%~N}oOpNFK2-&B
z<aY{j+PE=l3*3#%Lf!BB7V$xOva+Q<b69OhJA$f$jQb5&Npw*z&Rsmm!OF89PWU8D
z0~l&5YyuwuTh3(VDdlTaeNp~bYzvz7-n;Zij&Q;T&(e|*T#J$uLkZbeikP)@(9B<O
zzt}HUzx$_2$MogV;d<X);(l}pa|)oQ@<R%ehU!IvY$aJ5tJgzzuo5Brp;i^zJJ9t}
zLyp_dZykVZkp4%qYg3Na{8|niTD8FA8S<`5_eLx5q8d~X7-6<mJcRBBU6n!<>{ma>
z+=7LkN0sDc_60a7%gS_9swY2P(^!*@%`9M@y9g*}FFt0F+zgB^O&(3U>;gU@n_?V&
zXnkU3a})u+ME05BF~e+^h8^7f)?$d&<9|AAk^}k0OTXX5gqis2z~#nBg*C-*l|{sw
z^-RPljJSS~0@#3G@a_4K4j~A*wHBMg^f>`#Z_x)nVWv-m;+gaPSL9$SYkF*7UB*iw
zF%7kA;Bz&m(%%A+%PT1M(M7F8avk#<iXizqMX|H0#?V(;39~`8C&@-~ZFvbDNbwDN
z6{t)cf(g{QZLRYq_eaj{REz1K(pit4O3l#hO7tBX9>CWC@V`Y(|0;5R>%V`BLLp|k
z%2Lg~c!02r$6dQ@h{k2Q@3De^q}h;x6YSbwgoK5`t$`W~g;SGbU+@L1=`#f$^SLz#
zAR`}to+Qgc>87q5JEl>f;&q^|sfu-b)WwTPiX`aQczs=Qxzzx$NxOo<=q!SvrVy_t
zh)@Ix4<k|H*s<ze1f8VjaYZ0d2+78P4Spz&p$b9B8(Vs?lGg#OW<Yf57?7yn&2Lgz
zF)ey)htm%w5C8Uyn6n<^?_9Iu2?D^y?J)LG-Qkdcy@(1y(f$1stM&d<JO<T|66M*}
z7X_$K34+Vi<3e2lm=1eRecFejJj)0W@uK8hu@QsWk-aa1qGqxY`@9$P&eX4Cog$?m
zPHBKVzn5X1K_D?ij_zg{_-4a&pa*9Vgnn$N5KufYLZi+7N|=(3;DDdez&Hf@jL3h^
zQ!lWV42$8ro(15%v`A~_eepHyg{Ju@ltljIivyLRkREYc#27MYVNmy^XfhbHLQJm-
zn{1N*?u9mdqcm(#VEl?P@omIR*BIxR95iY_)eeLU7Ziq5)lReXIMloXlz^bumErON
z+HP*!rA?ahCmMvP0F10*x=k=&EJzj>K{{1HOwv5IsK=pMK@Gr^_7e+qjQ6AB(`kSo
zfGq$7t#b(BfUT{>CrarPmbU8iGy}m;o!1WT_l6_a<50WG5(=vldmZwQEBXIV#4NZ5
z`HwIy13dW1AcrwiblFNC*FEyq3-7jE!rw8EJH8*8X_u2c0bm`!j(@cdMEAxKl4O3W
zWe*F(p_u102vW!LFo<y##CetJE-jzKRbdEg(U}6#ip@R2*wsPJPErFXU1LHIm3D>z
zZV&@McrUFJK(X`o*nshI5ar~=W%|)Ig|Ve`K}qhQ1_xZg3iynU&|n~pKj7&mOIa>v
z&BS1X5I6V;y9ElM!1J#ySk(53YWHs>$6tiXtz{R&h6WgjU`+F#k>|aq3)o|UPY*#=
z6k3Ud0u!iyL650{{JUbX0V=_4YK^fmCeB7_-;kz=glP`$LV$Z8ta?7Ds;Qx;qMr+l
zbF|4NJs*J?Od~j|!mxy)C=Lj_D%Yi@(AL-Gs<PMDBbGgw*1}-kI`hVyU_yWFon&C5
z$uv7nS&RB(&#8!$4ZxvR$QZF1Fxg%2cNa;94)!b6`Qs68Y>Xz<Srh^D5EBce7}N;<
z{h!qG{L#3%b+4HtiVbj#ZcAU_{_<?qWv??tq+Z_f1<A3Os97IVHzNYavN7MPV3_T;
zr-our2Sz?m^5J8QmF{rw0}6wS&^`T2%Om&v8dluE+zej`YhGhx_g&JHG2lfJGYFVF
zz~IIrEG^t1O`8fLZnA^%reomTk8TRHnT9AV(FGc6>J5NG1njO^rsn1A(wx2Se0`6~
zi6r!xWyVg*X?s-O3w+TE|I1%rym4e(GU5oTb89JhXY(`JJK&Q1z3Ee6!4ylgi#Zud
z^={VEpdD?DUcZtQM1r!k4HFQSjet|n_uBL4urX&8tP<D1hn9bQB`TUlolx~%Q)j^p
zPDws0-a<?O*)k`%Xm1EcTgl#A3G{cc5Hr)hdO`&^VU8gfBsQ6NhdnW8s545=s;k~+
zLol1U+MIAeBI9zi*U0wuKXCyik^+H#Cp?jSSc3SZKPLkFW3Rx8^J!k=QXJ&O`ipQb
zqJRqO^2zS)kYMS2ioTo<E9sQaY0uhm003ji=v(~BZEJ0p=sk533Z>aY<IuP1K<{r;
z!}}e;IMFS3F;3=?UAz?{#FkykQ)P?mw{~GZ$X0g-2yKDcZmt(9qbo*dL1>s#k7>V_
z?#WMrAZGu`)no4V@hj)h1@dXfWjBCaRAoX{XOIvj8u%IYSq1q^TAQjOfU(w`ji&~t
zgeWO3A*|``M@IPnY~;KIQEugomBSNJh~F@*pz`=>4_*G{la;SP+$cs-9n3+(wJHf?
zCdKc@>h&**5so!N=G28YQ3pr}ZMKJh!$;@_dHz-dwli;F9|_=lR;6netS*|3(89U}
z)x3qT!#D|ycj8lFHJo?F98ekE2rn#9AmvI-N548k_B@_pSCWuXdfA(XO~XQTQcB@s
z4Xt6Mb+ZdOie(V+eUQ+DYC&iYVj-c(J@Tjm1oJLK8_G8EC;78l=6?`GceZ=r9MQ|U
zvB9urKiwVkI;PA3V@d>&iNP6yxg~|U*8OtZweP~fW(;N>#1&E68rp0P4?AqlKF=85
zNVjA&n;P+~x3~$(focz+RZT(_-0Lr6D!jh<<clAWQA2Q6m{}j44>X0c%x0rU7N%l4
z-KHsv07J*|u0z-u7Hm<vw%cS>Da3;*O{Mi61mqid#(<U+$BLxLyAn@e;mB15B=@W>
zdv^eN%McL;m4dl2gz?AI1b-re?Cv>ganNwVAhF38M;R{=Y`FYH+#sZ4<0p*43?pmv
zGYlR>0YVT!#R#U@9Kb#s6oT%vq0KFMqQgRihO?Lg2297i`p4BbaujkAR0J+Y>`kRV
znU3X~j{7%C^LU1USaOusR21zP!pxA_)Ir#lDZFd+d9>(>&vWK_8w19M(y3BaAJq<L
zyRLUxWL!~G%oL&!3Ca5GcDrvKdk{=q?{krGv!FGynV0>rH_X^l$@_n0<S4i*YXT-(
zRM=rS^WC?*rNta`dI14Zy;<C$0MzWeX~EVzAg$aD9GDpLvB}n`5rj`^4+3OE9|W^Q
zZFEk&dOt8Qh6SA<^K^%~S@|MWbgXjklGuy~d3#Ecj75D}fs}(A+1h5IkV#_Mt@r65
zU~q_a17=~Dcw{SqYjsrf$)=fW>VLxZ@@Pr54Xk{oA+(1@V<IVBj@Xu&^<^_vf)PGr
zGvm|{S|pn6i2;a=HFC*6(;E7*$WdZmfbn7+64-s6LpT43D~6bfc_Nv$_w+d^B;Sko
zG%!=+?GfCSs*RrFV?&Ec+|f4CbD;I9>NYU2N01#Fabv|mLUgpv!$>wY$QDw}ShqdN
zPPG7}`keu-P}(*yEY0H849v1xy83(^0V=R`OTm@c3zN*ByiGB+mPIDiQX9k9<o9Ag
zaazzv<g|zrT@OAVGZPmW-y@5THft&ZMHn%vEn3gcuLng4^@wD({g=L6!7=r9!FpT>
z0@&En+~6e@SZ!oqRjvqXP$0-lpSPSQ5jU-~X$P2bp6O9b+&W;u902C{I8My4`8u+j
zka6Q{fEh3FN4WguY$X6=jtX&!c<1#9b0V{ZwYm`#&%vCE^qIJ|(vv!PTuN@4YeC~k
zO2?UXkK+GhXDoG?aD?WJwD?S8eW}B<6K@>$o2vn3NOUr8uZr6u-vv6i5OX-PJAVTi
zf(GjUl{rpDVT{xn?pay<J%cR<tP0fgO7~%m>A40_-VWIY`bzH|4b((BrQ-`4%iA2o
zz`ZvK1^MBw)5oyG#;yGXgW5Sw&DAfK+WHplT5{_!z4+esCeeyzzJJZQmS{96ZX6o~
z2bq_<6=2-A2E2SX&LZD3H(vBR4}C{unc&6-4wRe)!R(F7gTiw;FP@3%8uGe%lH^Pr
zd#aXa%#mX~got~pj>Y|J{s{&A-;bFqU9pOs+bcTzrO1_ji#IJvx_&4IaWYFt5wG}O
zpD(rbCS8DRN)id90Ru2|pD}t;2bixXz2asKK1M~%FfR9V=XB2^W?uI?cvxo1!z)9}
z8`My{O<7FxZOX#ts-tsKn7c5jV?>L(zky*mFx1RV^YO>%Q1H2Ai5~fyu}H)&)$1MQ
z#<M02sbe+0IqjUIQbP6CawLwf3YZy~p?CCEi<7?lFaL{i6X(xGLZRfU#7#F^P(l11
zuZ9mw*YdT!$tKPV)A|%REQyszkHwV%$aE-Al@r;pYTD$$6F?u*Q-DU$ZUfI@)n0&y
zslh`LA-9ua$5z+s>X2jAU-_?kp)5CG1qQajM5r<yuy#;q?C~gi7hpX1LrWDe({D4_
zY2;jI9|$lCABYrnDstbo7{Bin*Wau|s~mgk<+S|W2!ISqk?QsR@cRMtZ|BRwqr2x)
zxU<Rb1vRa2mK)GrfF*sS0B56Mdii&I<L8;Z31yhJami}|LX8mCRO;BHk~<iN2Jjpx
z--6r*2o4q3$p$w5eDCRE0mO#$dJzO;r5K$LyY?%KR#6RhPRo;Iu!=;(9H6~FfM?eM
z4A>=9S){litst`+HFOce)Y>u#9P;pVt=@QzjvVPs<(8sWEavZPM|x>+lCFBQjxegz
z;oppqx=%&rTV*S52jOk8e*dd^0qcTSkxN7@Oqu;S-&p#;5jezxX=*x%S<tEPECh91
zD63pjj9}}7!RR)g7_3n9oRDEg$R!FUv2hBPtSypO8VLiTRZ4t)aYM02qD~<Ry{>>0
zGJDHCifnAxw+YSN;7nG&Ty<8?R}mbHmO77=&3D79kEL!$c@@T0jV#*8?%f*-7V8%=
zKn)~-ZBulC(o`n9lz{XXPxc4HDZy8|L3x#=WKWZDE3aNdE>p7nJb!uH9&^1}`?}ee
z&N9v3zX8w3yib$~_x{*hZ(!IM&td)?@OeH<@0FL126qpy7$BN{D+WRsIB|v5U?sp~
z|BGBQdTB3L=1>A#lGEa#du<0hpc0d20w^E5CQ<7FK%d7k1^`pD+#x)*hy?Bs@B5Bn
zmP6W&WT}Ec0$UI;IQhd=02PY-xMCa^?Ed=XodP!$C%YExGWr4j7ZYRRt3r09-N@Az
zk94cn%&Rgt6}WaacBuBJj2p;Y1X=&;q(U2b7qnBJB1F4@Aw;cj+lnOWejdZ$XH1qO
zrwRSqIMgHOL@1OP++*H6E{xXkUZ^(Ke!r%Jy%u5p0({v4#<-yqXjZ}+fNh`vPr<vX
zDR}fl1o}QS<~I!h#(G>N0Ei5a_UL?!!B%|41>rKPEq2}JIt82)8o<4O@%=N+R3<?;
zc2?13CcOlZ{gFlR54>JW*_VH(>?XyoQ6PT)(u_V-bYpKu%ZNHjLxQfjf?otqywRAW
zKLn%qMn0hfObsfA8PoP2kUvNiqYwiyVCV6!QE5t$o<X5!xTf8&FPPF%jv6JS2D^#T
zUQZ;{25Kal)VwC=D4Fu1HzKgpW_|I|pAcZ^?M#Kb1yPuK)=w<GGx{2faKAYa#!Gc9
z=wN|m<XpQ48OPBUqzb?Io9Qje09>nF)UP&t`(I9TZm*j84gFeXf*Z=_vW5av7r!_x
zkxDbO+&I+6Gyu3(;qrB|-a7WUa|4gmwQB&z9Nef8fZ?Zlse76uEbH^x3_J9uA-E#I
zBRO?Aoc*hhZ`0Sr`zL#5qDIN>Ebdwz-`PC59PJ2D?a0Jv91A(l7=iTzfD<98YF${s
zFdpIV;IPoZh;`+HXqrHovi7<oVEHwi46CPTdG-i9tz#&9a}3w1W^r5+1*-wTjZF8$
zuS@#lg+u{>8d2?Z1aKPgL4+C23>sLDEYYRN>~1WC^#w0EsCsjJ;WOKkv@eE%Z+tZ4
zY-4{!8cT%G#L0*qhueg@jc~{H)E$ZG%Qe`tx>a#`k)&~<rVhBb`9k<3B-Rso&}x2G
zYalv}EVvDl2}vGg7m!gZdGQ0`v&jH$sVgch%>WyEe0Zim(XO<W^44IK68JY%qdi=u
zi(|YHA9&`_P{D*iv2+nZpc}UEXJ4{l&jLi3qj@V2GbaE|C^r#lOn|77s{!IEY?X#K
zFl!+Jum(A^f+8e4G}$pN7x;>wW@;c9BwRq|<USjSCmbUN!36|Y*P#l%5i>m1YJeb|
z`!y36Pd-awIInGhvtb)STd$yk_&VEKF!f)e3Rq!7AP*x9W)e14#6c}u>hnPil*}N>
zlNr12yy{`k+$vR}YamAxsC$VPk@7IR)<S*H`j*shtew#Qv1ng~PJ!h)YpXL6o8$K~
zq;&JOy1#B5&l3BJPDV6xJm{g^vY{scv&4kLOQ(rroFD|fW;28@?d2AaqGc{-Qq-Gw
z#%j~|Z_v>pt#1*wiWy@7>JwwC0X-PGYKIMtQiJ=T`-R-D%wJH_8gxn53fJGBEF!Xn
zr@qW%<g*msUVffgAjPR_41s7kwBrb-G<cYSjj!sHMO8;QyA;x3K%{6DTo)Wk7k9As
zmRN*bc@F?c4Af?qlZo96sCNX5f-^x~Y)!6==+iMkK@W&V+C6~g4eyHB3rU$4yK&q>
zY9bQsNpM-;uG8L1I!8g-m>VEw55r1=bGnjWYV+$IUX-EQvwwhSylvY?TWgpG*a$FS
z&eWbbP8|jibm+n`Iu6p>^J4mq*<xiDfSm*=&VdpFK>B6e6rAR?&W5WPizcO1zbLn+
zCcT;2m>%hiu*IaijtLl!-XeeE$d`o!UN9TxDVwf0JGmUpr<n0bV_Cgj;f*9HIkKZU
z)+!K2x%<iew=5LKc%5iG7iPx<h20o}x6VWyw|wI)`?zRJd1|d+U6nTzB|60!>#F8H
zB5yNX4T)uR<Yf{;NFP|hAl2r)zR7(pXAr;j>yi~>2LNcr&8j1BeGbUUeXLVfqaW+c
z!wO(ut1_hb95wn){u!8Zuh!(cZvpj?2N=hBM!`cq3`D$f1LK&kG8?$JRW?eX<ep){
z)qZ3EGu^I3<Bz$f%K<oqdIFdpM!hZ}WKcxW+%iNO2rj`DW7<}p9}FZ3b6bdJ32N1a
zPjYg=0MU?NkyABvMMmVMFCFyHhzcuwbPt6yo=AJ9g=)Unfa$g4Mz1h5N5}oK=GKz-
zMe+GrZW`QPJNfM|j|UV@#E$(K;i5hw#Z$F-NJ%D{ftuGrC$K&^ROi={h4o=&hRsws
zQ-tn1-Wg_OSCrJ`snm^$_PD`7G00hp*=dX-AX-@q1Qb(yLK*}f%lE8ma4bo_c`mK0
zn=?l|QLw~72wwz)n>$WDTVZJ_P^KcezN<$VwCV|f8(alyK`f%|BgN^eLS-_Hh#cZN
z8*cm!fc!TFK&?~<hk#8g-xVPAQDc~0E4w6;6s?$7I+<CXH6EQEIpC3lF}b*+;QYh@
zY2O}i_3sA@-`f9bHq=4FqF^>{GaJCMU~~X5pI6Zk*V66Um?-aO<OXCO1bhPXtdkrZ
z@^}dm12Kc=(Or&#-{?*cgQ$eWb4ttF;Xqn}IKVz&fSJU%B`6p+OoIqI)QU6gu?rci
zLxfp`i{L%Sg+5huQS{+9pz+o=hFT0ob7SE2HD8bo>6q3LC0TebF)XQQDq2JjHfQg3
zQdi_87V7k!a~@y#&x)uL1g8-FkWjR8UGjXe+}~13_q*3@F}+$J!s+z_d>uDv?d^d}
zA0%i{<U~L}RDC?LgF7UL2mNN=DS(921yoi`=*D$JH7M`y7oHT3i&9eqQ4}U0RF$n$
z?f}RxSZqPWs7cc~py~*_MGYAfn_#f@L4MJgbpSHzpkpUrY-5iQ{r2ejSX7xuk>L=n
zK4Kw+_sF)(&IGmd82PZxw$q`pDK1*6w$;>%!7-2vQzu~UB95XqFH(7yFah|`cOXhy
zW`FnGD@im1jqZR$HGCwZlC>(E79zZeG5|P9cX6@21_!Df_)hq;;jwhC)G!8Eg2qYI
z-JVbA(J?xVO)wdzp4=TVfLSIYNo4(kg}^XzgL^XI2p}@nHAR;f&L~l+C1$~ZG;Wvx
zD4gVnYY-OPl`&@u=n8<*M+d>jGyvQ&ID{nx!0=&OO%6$UFwIHXc+ii`ECkY~D9^Ab
zQBR5?@z%I+?IC*y6QKG**{7J3EhdF(n-T?7X7vgQ)S$uFV7g<W<e9lGb(mEjC2m?L
z*)2rAsNmfI$VFxCc@me|Wh9JjeA|MqjJ#m4$?RoX{7U=#zMg4zzFmJ(HCDF{A*9x;
zLeUvu5oP$}sh^T53O=G4^HqhH<sCsBq?g~)!YQp7aK%?kC^T5c4uWKU60L>-cb_US
zm$0BwFR2`H#Slxd9}S(rctlpG&=)B}2qDT0V6~*|WRt>1dOLRTh?+`*HqGs^)4un<
z2~h(~KL9<_g3e9Z^6-4tUXof0K<MO^5)1^f+ohO!soHTeLAVzB?UHY&m4D4PRb1+e
zmW7UA2zu@ttXXjIWypC2hXJTz!|LOPz}Q#4s`u5OglX71iIV~_KaQ#6E>tfw-g7pn
z+X$K;rV?M!ZAxXVi%vDJRoFxq#r&3=TE=)0owT+tr)!?a9W#cZW{~>a9K;SX2x(fe
zax|o_o35@7h%xod*@e_Ee#r=GpUvk9^qhO`=z%FhI5n5rgQ@ktwJrkq(7p^mTQlYL
zJlO~?(AAO<#8AD2?tcoYS<+x(aEE{dm|?6P5UqcyIP#mX>zsMnO%`(-2AJlqf2H1B
zob<T%kX^q%G~?!rS=46u2Y}WpJSHYudU#D82231p6FMMJ2VlceXaqJFU$O9d0?1hd
zniNKISWUuUl!;+E?|7X_#<qyzD65B3WT4uV4C+oA6A$^DPT)8?^VH$F8IE}Ck%TK}
z?`v&QK22PG9}r`pfrm}$Tz?J833W8k&S?EUai;)$etNP?j;vNWv5<h_9@Dh3!!eYF
z`_8aR*rl@#nza+Z5+Q!C7kZ66veD(xx!O6U5tE(nRsY>Kv0{kSA$9-cS3$bVZzgdZ
zB^DFmfoX1o!&=^AWf*_#wL+hV4zm(TMT~BvPSM52AvcGsF<-XfxU6}usZEr5Opdp`
zuquOKk}3<ieSg|F2~~vEB_(s6VMn;zUwz5=N8F2GeC6&;4nMUUCMQlzT`WecV&JR8
zQu%ud5!}Xj9?uzVAJZ`o%~UqlyFfXof;%(JMr}p012b6m&M-wsI4g-g^I2Ut?9=Pk
z<-Be{?&geYRfYdj3QUUt>;kY_F!A%QufT}g#@Ddwq6H)&Vx{N8SOd+i$(w03P66tp
z9d^#xqd5oLV0>U1ml|Vu_36bz=qbcodBu|#lhweV*&Lr!JIkNNnvWr*nw>{m8)=7C
zz5m>=a)?GCzL>lsTnGB0MlqFJk%N)#*ciO@xT`558mdnnPq__Q@HMx|1|8Idg#}O>
z<1kfcSLm2S%|Q|mhCJP81jJGTHYxH(OdQu)dt2|JFHH}CpvjoOqdChKoI(65juqVg
zt97=%Yx=;Vz}S?&LeeTKU^4vHl3wRsLO9W%Zkc&p0~8Gn08LT!Jb%x~Py(h&<`Tnz
zsbQbO`%!KnO0@=ZGnAB|LrPBrs`tT=6*IQJMd6ya9$qNf<r;IcAcn1sTf_q7y?hW+
zn0Y)S7$52hNG}5AaLbHQi+aj@O?epzPJ6Tzm?SIZI8Q28ccvDEcyrT{bYhFIqf&%w
zLUaw~_H7p<LO9Z=$lbvJcVUL_08fQ6jKKHtq8d=3@05~VGm)X&sY2mI#BfWQRWFtc
z{CjQp)m@KSpli25eVL2wTSwJ#0GRrY&U**O05N7T_>#m&5MiWsTyR>dZ?l|uMja+V
zDRkZ@!n}G;t7wRuLTMk6Qn1yb<>;Tp$b{rGk9h|-4nihGh!EU{s>uPs(!6HG#$+B6
zb>``H7)eo*+k^N?9>qC<z2j-QadjdI3?z-q4BY!*-VPB=d(QJ%8$P=cz(M8B5Kgoy
zLdq^oTv{b__fgV_q5S}XT+-yCRM{5?A-WW=&pkpX`kKLHCR$1osf_30G!)FZMYOr0
zYk5d<9eq}!j`_@D!s<ol<iOzojeZ>UJ^eaK!>g2#TKB@e1J>pjU;@(F<k~$-|6o-{
zlwxTaTjesjwi@D>gWbtE>|=Gp-D6~e4b&qwF=WzR_k(0`W`oppP1=^Sx4}4O?F5KP
zbYGwj0L;sHVz5}wps$xI(2vl49(I@$2Z_w0#RWGbVuak>`kn9+OLVtaUTI)RCwv%C
z?j#7~k|bHzX1tz;Dt)E5<FmF5Dk<hVVUyDl_^)L!+^Lxr830=3sLrPA>E@h|vf>3j
zZb;<8w~P~?ssjRa=@`Q_YN_(fRvXL&67s@>1qRVNb*j9&EKtaA{g-I4_^lYItGmUG
z0bzRQWUzGX_PeH7eDS#3cg-YGF(=8l@H9U6z{7|YFs|CD5fLmJ1SoVG0EJrI#XA-b
znT!a%AQ>KNVG5Dc&;`FZoYF(<bxk4cr58-X2B5bFdpmO-n@P@SV5Neei8=lo#KB57
zU;djXURIwOvvc(|*OpWFLsnaQ^*Kumg6%bK(C|w$W5xy+nFcr2<B6?x2&_kfgnN6#
zGSi@I)8WW@Di~7#X1yqGTXy77f?Jt_TzOd*L=j6V00CHZr=vv51DI4i2MJ_xXTsdi
z_-Y|jK$8&x<Uk)KZ-5%w2`wlPhoL8@X;uczjgE7_12*!SB3#{JWFdx4#AIW~IzdIj
zaCb;&7FtR$4D_whDzowxE5~PWJ-@Jc{wi#IHY-w2=;Ud3F+ddC^h}tfZcUMUvO9Sv
zI^GVcj2kzz*RDf*KJqSJzju77&GfETD{XL2ZAd4iipu3ssg{K;5R>dx(cK0#S2JtH
zJBlrrvSSGi)(TP}|9tVhg909ZFDe8!l2rn0aTczJK$pO~!fD|-2Fa{)neK(|0f9e3
z8Du=7A%|Rgur#!kmBRt<H>!xfUu@n|)Rq=B9$mMgZ;kq(?2ORbPg#FEZhqtClS>?j
z^nRZgMS{(5S&h_ZEkTjk5F#(E7-?8>^O<Yk3_uIEoGNVC7*JC+mAymfbP}Kgv_+z*
zbv1-G1R>a|hRC23155UaW!$XHhmq##89W|wvk1G(e8y=d+czvMT&Dt4Fo+3As4;M8
zP_u(@f$Q8EU|68Zp(a>cQw%;hx%u#@bYiqQ?xBTF3OiIV^ret7Qx7BZ%y7<|I%5R3
zhgMzZ`W772RuNzyY#N-wMgUC1d&f;f!HYy~ed#|p5NrlC80BdmV<GuB4sf!KS71>V
zBbGFir5=TKsD^I8f7)<D%I#4n+i9j_d?c;n<C*>1jeF8s(y+wcog&{;$bhq<1r`v!
zN@D`Vc@`(vtL<GhkC@RXzY+w-<u>H-YaXCzXq*6b{JTV7k6HuUKI+L*vL*9tcPLA5
zygVLZhmzkm6&nt)5Tk>!hvm4$j-QKU-fpFZ_$?4<F9U|{)D=1LA!1<Aq6}z3EhXC|
zEQn_5X(N-Gm_&=s%!)$!Cxikm>-Wj*;l3QF{6fme%ag@9E1i%?V3g3|UAnwp{#92M
z4V@oMfQagV={Qb+_JDfa0qnd74sYY$GObf2GMsj0!#S)=jC4l3K0k(AjLqv7rn+tZ
z;w!T!e=ju(YAD^EKx7~`aZqD0Qgh&Ci)LYiqId3(Ry{u^_&kq!qIgG9HwlsPsF2g5
ztXEDDiB}e+*+r_j$?izr**o=mwd*umV$oI+Erf~*@HtM4HDe=+K(ljP8x8-s-mR^%
z(<D7*q^DpfAlL34*qb#aqjh+CP7KhJ5?*iZpRXcBuT$E2e_|X1^k`Sf-kEpl$*i(a
zQ39j3eOgdy9WVJ6;!>$i=L|siq#mcqvQzG*nfRLemHI3+;HoP)QT5B_cD-cFKdTL-
zMIIR4AVA;0^`4v@0QLm9K?lOj0KF7qin$&IB9(bte4op%Y{h3yAQ{-?!tj9_gwAfS
z+NEfsYidzAgCx3)uragsssNh=f{L$Zn!b8f+m_m`XF$%e?2L1kJbQXUh;$$FBBtYM
z2bO)F3u^wpC*J@Z``tzbwD&Gh9qA8YBPVi4QMi0Y|B$j2CJu4yi&kMUK$(DMmtix2
z%T@XM;d_l(U(Er~P0^l8ulX*72*8yMUAPB=CSbD@A#@W5qwL>$@7p<m0fJ*Q5*S65
z9zsGHp(iEyXrZ*gIlG{1Cy5=gxFOZz98M5%%`e%JY%{8oFcwyiwn>BzV>UMn43Pnw
z>RK}|-54{MBYiL2_&YCZ_*~C^!Z7j-1Oui}T`G$aBJ~um1M4gPAJCYoL!H`$!<l#w
z=hm*9>H!-rfHwfspoSgx)a*Nll`fYaQAW<yu(Zm0Zz3)5JOyh(tytrVr|MVwOWdrG
z1j1Pl3@F>^&};uOvw@-u0D$xdC9D#h5Icvf**Og=aEjEdIy2Dn4>E=Z+o7}xmL^Yh
zx{NMR9z=ZvAXJboafcO&AGp+BBBdGBC&p<9VvOgdYhQv>0O0)RokOV7Ot&aEcunnB
zg-c$YjI<@p#Dl<KkED7%hcAq;TSsTWcmQ^U)^+%S2{ZQNaMe+M#=V$G+KV2qkm_h_
zRDRbaomYbXh1!$f{x>f>-4_=wcr$<xzK>sLpo5qMa3#0ihG`80+6l^WA_YUb06OHU
zALEQ47^h*LIt{D>Vs0@u0|R`1E`arZkwmaSP&<GrM#aVAjxIy326(8lo7HW}4L7kY
zX{Q4t3<??;=sqDF<Q)6A#g%*r2YMWY^=x%qaZ|@W?m5PdH|At)yr_{NTu$eAP49+{
z>l=+|5W!BMkq0?6ygURTkP^7p6jflW%Qgy^U}7ChOZf-)S(wtUNV}?vr%$|=i(Hf#
zQ}#*#fQ^ZH%LS=91ygju=rA!DvUagCaJ)Z9Ixm?Gofo8e!7Ym#O`O!KgfEx_N@G}Z
zF}{y~&UOJjw_+KI_{T{U!&;Uwsm|>tWC#Gf1-YJJVwcwJm*O<P_tGG5GhEl~_JU(y
zx!*UpgZo#yxlCi;JLrt7!P)}kgpJtO3~aEo3b&g3Nbudsj?BXtfS6<hr~uE`Jp|Fh
zDQx?q8T-01agkrC|E#$|Dz4adkC|n|lK-)nWVu%$b_z+HV~0$8e?N>lI{v01l^j<|
zam0`T9-|DD05HJj<U~xG89AFO<%iVVT}#NepnZixsT)Tgb(8;3b7wPv>;VSbZibR*
z3>qxpyhi>V?%$b=y$;ZPA35w*fnR6o^_nT=YxbM(5rtVGf!A7!tf&I#UkFp701=W?
zX~U!g%Ch;q&-G+Y??6@xQk;3?`m5J!8pfW*hsW*|3+*<0pY+C;E4SCTfGu@9W-!6d
zZQYUpQ{?KB=U!G0kH@ahV;p0>^Dqn)w8KVNocDOkn~W2i6LI3M#!yvw5i*mb=mhxa
zvFVlg0uL5~-!p4~SeyV;1N(Dp92b=5ew__$Qks;Yuf%BxfAGp4o!|bvQsW6%CBB_b
z--S(-@@&_TY1R+7W4!%}qiqc6n8QJmK1LSvtlW|Vt>UX27&kn?L0@TTL4%1Z!*h-&
z51l$4mn+CB_HJtU_lu9eGq{J<f|bWo9$ROeC&UD&Y;5KHszQ*x6&_Ggj4=S$fU}32
zIt+rLE>^K-iFMq<b*+|c=qg+;+f!a$xFCsXI#~dbe5GFa)v6A54CN0^lU{$Zta(2#
zyx-hce&+zI1KKg`M*6}%Ug(pV48Nvz{SxS?RI5yu?!WtYLT6^UC)tKrIEmT<Jr1AI
zLs&J%>jLm>!6w)gsaWy0!i+m^vR!)~R$7&L{B{4<Bf$!s`gh5FD0VCYLzp@VEo=|3
zzwRIia~H-?n<rAbK{3zrqlKh#<DIbC;<~8d_j0<)6M}hbw+md}km?KiE(NTjreE@A
ziJTJ$Ly$@0m+EASJbKjVFx<bVUHyLCWJzu@u~~3^WtO<nk#DgEaJ1d26><o3o1PV|
zeXi{2?-gEx3!w;wi4F|&xoG5akDDrx!*($~+hZh|iU_-m@A6u69Htlu6*!lx#vRS6
z`?%s}?NN+M_gGk(6j<3LZH`vpz`>n6o7y}Tp>pd`KebAB7NHdQ8tXjI<~XP(_>jeu
zba3xo7R@|Ni^d$WDXpGKHsrj%^3{8_%A|61Ieyx6z&WbPoVN5=OyYKPem`NZXLCi|
z#dEUt1ia$33Ry*R0gJWkD{j6pbERI|HVkn>2_-QwJSY4mo_?e;NnpIzcMk+l%`JSj
z>Msi44&~9?6RpS3chc7j$n?2ivgqxv+Fg0wG2bl{a*<*~$Ii`t8w@*iY6uGf*KkZE
z?n2ygPFp&F`eL`tVPuKU+@<r@9({QmMTtKEl;}%7hqX9znRA}$N$Qfwhgo7rDrJlq
z`MGff-0_3NikBal_bf63xJr<V@j}Qk$y?miSBTG!)5G|SxIx<*noa|ja)2wu<R$Hp
zi&q@?l4A4<c>TH1O!64xTKfZqp@#Wu%6lV*eihQU>6t4|EWSB%zi=~KXVk54>1ns+
z7HaBoiex<HmE{n(XJ2%~wB%%(1Cc&DKoA!04BmR%SoY3a>jItf#MJFHK%nh5_V@BQ
zLizydqG3_sPwUZ-*87(O9R4D;=6@I?zY#fA=EXbZ`b~OU;^e^HJMZf2TySy1X!<4m
z*c$J4QdoOj?^A8;O0#ZCz6sUjyJ@J8xSCt3Jme>5bOhrgoJ&5%%_^n7GBqfDX$p6R
zysj@$L`%~(ggAsdNKECs+a4=EMuMYB$qbqT_t7zK&{Ieu=i7Idy@{~2tx=PHo2yn0
z61$EdvzSZ13)m8Y$c74)OGuLV4K>G6lbrV!Ozgwu81Qd7PBY?`=a>UC%WWFsUZ#iE
z{p-ETxbffK-Wqmho+m=FJo6aAC5s=Q)hHB9UnB(p<g9uTWUn;W^4`;x!hF7l)m#X+
z@asLQ=G3Q~0aQRF?tzr!j-p|qL(n`a<M9YA#KLb`aF}SqT_FSI?^+HGbr~GDg*$JJ
zLe5=i3}Bw)V?2ifFr7AYXTLR9?5<|x1s=NxrftcE@zt*Z<wja})7Q=Uv#(fnce4~0
z`Hd(b+}4<}@xqsXZ@+7*Ne~W_<nj)w1_c?{L<j7&TjfwMyCozy!e6u`<0)E7-Mu2|
ztG!@L_2~Ws!9Y&EPQ#~&H7sKh&QDwEIx)gu=W1DCQ)tu5Zrm8!a)a@ycoCfs6(@y8
z@ngmiCZ-nC^;wXOF`q~H*`8L2Fui0Y1`yw91avb^NMTt+dH@8@U-nRUPuYJXZi3qD
zf;CvRHp(ThdV6X&Jb8ZF7=oJgeF!x>Be6xOjer<Tgfs*ru%t{Q0Nw`qBHe#Z?qBu?
zTUX>Xvq+Cw-cP7^9<QE4fqHv~9M;e%&0ND-?jBo4>7H!7v`B+J>l*1rojuYvvq9(%
zGu6_B@N0&h5&dg(Ca{1yT2{vgfVM~8GzZY2a65Fw04Ks8nI(L#lOs(DW74fcyz;&X
zos1h}hNpG*8b$e{oceQdLp6@XImx$fTBzqZtsIALgD5ZM2c?D1_sot4J)!QSNF>Wl
zHjLz=i^(<?R#jDTg|hP<_O)9dDsN5&nBU-pd)Tb}B|i50m{N~`U!=kH`n)jW99jc#
zwfp*v#n2cW^67$LPms9E?a^3}swoq=XgS2hst<MpX2XQnT!ZP;o-+;}H3gHtb{<}-
zP%pbp{%t)YH*nUqk%eg%K>t>D6L%-$H3g4uIGn)dvckI*2VYlgJQ&J`k)wTIj%ya+
z>-KvbzUx1!ia11LEP)7&KKwS5y&mpYuWNGTR>9lt{?>S7OFe>B$1;p(yXFbY4At|i
zWp0@Vt4a#>iPX5_ssVd;?f?Ki_@)(Tc_mJO)cm7PQ57MUiBMi~GkZQ{0&c0hsyffK
z3(Uqm*9bJ@=3G<{(q7N$C6XWA*erJl0mj?vn*u+~SHFE)?9~j!Tzw#dl~6*FGLnn1
z6-}0uU97OVZ0Q!~^cFss0%`{w!%Ou<jmk69(sPYMRfHsbf@>V<e#uspj^fCLHy>A2
z0`TNryaLz_OpU-Z9n<68sD_B_(kUo0XTpnmYi^XHR$Zudc`AnIBj$KJ=hm9j@0A65
zc2z2pYCXW=#y)_@lRCqO+fW;@W6tPxXD(ua)qGku`8Kt8zMk$aVuio=E=+B%?>a<3
z_F7GmIG?x({0^wO)^umcTcGxoXq>Ux&7N-MtMs`@Z6L8+X!pNyv1pJ+=&)K3f7k$D
z08c=$zo722Xf|KuaN%tka5;rP-O7c(-+-wVTH9un-vKP=5>9Ew{8u1+;#>7u?3Oqf
zJH1dJQeuccfKYrZGKxeEnm?%!g-kMV0@21$4_dmh>hViw?gUkvBN8Q@FH->*FdkTp
zS5d@Fq{VSgQX6y3ccimtiLyp^V}8$E`Ol}ws=@?oaC<59p;j3;2X%uNH{q&%bwzc>
zeGV4fpImD#c-{gy=4qbOeBLKP7LCb5foQELHh8Pmy>xw(G{Y;&m~p{(N8&F^1!5rm
z))#9z+0-|ut(#hOrh9p1#RKMQC)SPkiX0D;pw}#S>GSAiI}UrvO&G2ky={m%06%i_
zD^Q7Ha8Ol?i4wWH7BK5>dG&#GFs^kUpDw)&P3{TCS_rFa{O`O}A#9;r=qcwBxx3dj
ztUG*1ch>GESXw%3z~)8gIOB$a#-#0t?3SUP0Nm0^k8#L)5^V6xM6dozN%htbv&6yN
zmY(=mDPMqNx;iJm3^Hy-a*i{XDgo+NR1cMLMU!I@&>e2Z5Cak0=mr?3Wlu;G&h)i&
zX3*D0q~HYI>%+~F8U&d*rcYtYoQKy3edb?f3%MQv8d&-=`V^%LaTs=;&$oa48?lpk
z2Af_QqP3ANw!#|U`>sS@JP#*)NA74C-4wlcgnm>gn0s&l3nw-RUAPp!92Jw>6=s-N
z?KsB2@k>bO_yx&+P3w=i%D8bitZBE(2I~vc)sHp`lQSh2!0ikgK$KrgxV6AUFDSE~
z-2_`|L<HJgRs5&CgTaJW?SjnCLm=UXU05hbnmJ|zQ%99?5_%tb&b>aPf8UzO4aUnz
zZ(XZ>?zn&V0aBL{N}YKfFPGKa#_2dg%p1cF*JuJ`zzw!z=_~?p;{rw$4)xUA$|<zC
zh4m3$gsM!%)|U8Ngj!Wvvdmw<2<u&Luxz6WnY;~Na<;D@=W%|hIUp*AmE3yjo>+AN
z^nPsRF_;}yt`=c@1XDpJD0^UdZa8I2ffi{DM|+sBj$6Hkhf4t6i8=B5;jyZ(A}42G
z7xxT+oBAIhM8fRfWmcb1z-D~<8Ti?#?%2k64+wkC0Y^D=X9U489XCk#QXuqCCv#Po
zT0iicmmN3v_iW=XWG5zVh|s9et@4eE!QJaV;1<)jGyHl}y=8BPTc_Zj`LkWy&N?rt
z&a2q{Ua)#fuia=!J9#M$x5y1$C1+WOzQFM>!yo^C*j#rt^2eg_{n5{-k3rZ%<Pq~-
zZP(0j0+|TJt@zv`+Ja&!pqjZ0i_0r>5<!@bAa5qjzt{`I35hRm&h78&I|7Tm^F{bD
zgQ;N&>>xT5YgQ}BqRgUkHf6;gWUr_<J<|Dv*n-R;tutVwmKkSG##C&a*l)ruG+O5Q
zRV5IT!S>xcjxGdwL=dIII`hhoVkMamvm(E(n*D!apSfPNfDl_?BYvw&TcQWKT0&JX
zYuxW1MjN?tW|x_SfHTQ^94I@d9ohnsf^igbv2c;vl5^u<8D{xMyxz{c76jWT?9g?&
zU@SNRD6&q*4N1--dh!(=3c_;bv)W?vU}b_k3Akf`jhP!T!~8nBi*0Mb`Lt3$UO7`e
zOfO}@@a>T$_mE~@{e;<PY`FSCbOQw!4G9y(N(iklTrlnRKGjg+KRT~|1Ancr;MJ**
z8*B2_u5*zU5_G!G{x1L&w7VUN-4OfOp9XEfNdK_hwKV@)?RMm^SMicLomju$fjub5
zQLORTRQnW6sekn(Ss8fQHqnKVWz=2{Y{noK?wt)I2FEh)(1&#H9T*Dq1|JI9xB+&w
zqlJjn7X(Cj8Vi)Le8vqC1#7hG!t2ZCBc+5ydGYck2AO8CyBy(xz#O~1-cAL|M(*#A
z0avm2`BnWd)R1*W&KLJw3MZqR>b);4?)$*GKW}n~-qBBE9AJ1cL35DGR=?liVo9tE
z%g~8M6}r227vmyl9ReYX@qyP|nc9`+QkB&Of(5sDE^L?37Djoa?T2b8N1vvHSLg^s
z8}R6%+`?Oj8$g`^jK@C4ILy&3&SJ_*wS^9?K1CRye_<V+NbHS-o5WdSV@o9fONPHi
z)ygs8!_o9G2G8`j^*<&U$A|n@<OH4x2E(1JBmDhOuF2ue3rzFsI)44>*|f<6nz&O3
zujk};5m+yg<ptI{$XLlWtIe2waM;@B(hVp7Kpv3T<->Zh(drwl=KsCwaGNW<3;^rp
z^K&Cwa?amyKG_-#?%^<4pUOSQ3%Eamp~ep!KCJ6=K2C!TRTss5^m>H_pf+Q(6Ah`{
ztNT*Eqe&27i<bZrEk#GR2$m>8$P?gUVu7-<bvFs$WBkbLRS4>yTL)XJnuAPbUizu9
zr53Pd%O;v<)O~u@I&mOd6!&Ww-Gg808f<)J3i<Lnqfp~spI%xHD7S^uZb=t1>%1Z<
z4)CQMmYoI93^(z5;Bw5ZshK;FzS8V;p{r#L3}pa~Asqt#3|h<HrDHf~2kl26`|&th
z*P#Gn)RWfM=2E9?3V}G8F)U+Y)ZEvdlfcBXBnMVS2soQBiN_1ruL;2z46tG#!~duz
z&vDIlpTFCG^2~}rG<-h#O0BkX0HWJ!zK%^^6{G_#rZRKoGXp-l0v<RHyQaBC7zeJU
zeH|6|z;jqMKx(Nnf<A5*<dzp(fQ7xUd6_jg;|nRaTdSd9k&B*T^=k=J+^D5hC$a(D
z1$}^%fo#AIroESC<>*n9H}9hXFi}lmTZUGxkDB_qK%Q}%UkLF6U+|%XSy3GDt1Hvs
zYX1FBi{-u)4+#`q5C{Gk%W+*DrS*oNBIjaovweyyO2sz*^s9F)f<DjGv~jb4&)03-
zz+z&xl2C~gkkEev4T!*qUXNeMpP6H#edZ=<MPPlYP*MLLds=Z*TkUdZ4Rb_>rK>}n
z%@l$#20$Exhd|H=5b~%n96MSR;5yL5RnK(VKrr$QtN~o*{=0OFff7k}g1LsIFr~_l
z`ygi{I6;J^b8LK`SO<w4B(0$=W8_*hAc=*sM&Q0Si}@;e7AAq^4sd&2kKO#%by{;X
zPQ}%{{a*dtN>?0Jnm3G?SIE`ObqcURBlJdJL7*NuKyVYtJOu2Lsb`njcr^n`ddP8W
zK;0ro*R5NT9gCuXW+sA#b!4>3tgR1-8*(`hWa8<78(boORO&G1KDf<9xE+A9=xK)n
zva?|5n5O}z1L~q846B{Oyjo>(bf#o!Ht>msnHcVhuI3CAh&0)`1I><X*&s+TCQ^hk
zS<`-X16uxjTDkpb+<dM9KW9fv?qfa7>#3urau;vYZ&6cmW7Pn60BEx<S<31NFc5x*
zIb#KsulN<b)hbgpl)|o?S<w|UTF7o`{O{n!^|o={Ty{CJ{gU@t?dOX9BpT&S%8e$;
zNSV7Ys~0L0`+i>M{IQQd2Bv=~Aj%EXo^v>WVTXd`#%D|P&#8zIm{$*wMTE)>K_(Jl
zkGOdVfU!T_Q=_KiY=&s+Xz()QQKl^^7#_TeP!RlgS%9B!wxMRuUf46$-;aQ$+ppfH
z`P%xUFPD~vZ1A!yu2`fHMB+v>MoQEirVAA_gtV+qoqkcQ!s}LUZ0HE$A3#ADhkz~@
z6I=wRB<<H_FCz%5Vl(N@SAmh~#WetKi_{!D^0A+Q+d0+1qY*qsXbo8c0bmco>M%y6
zv;+qPGYW5_)>umG)T<zfZ!G+XUF!&FiO@L*<(Lh<&Xg8vm5!8Mx6?pwIeq?b^G9Fe
zhz!R3B_w^zI`d2Tois&!52N~LDcQchC^Ym2BZl*12E=OoCn{a5!%`PZiJ2J9>~)Ip
z<&D3JBDX+{<EJUk9LcTKy-d~}u&@@as+;O)rrWn*a~7rrJ#jUaz}vYFxe<dcP*Xs4
z2s|KW2+)T!37|%8Cj5C}y8s|G7z`$+j2nuLNh};p3^lb~lXy%$Fz4!^a(4qr(kLns
zAA#aw?ij8??fNh2KCCVtzUzOHuXG_1vF~E_`;4}=b?AW5WH8!Q>=yh}=G&G_+Oii|
zpHOKy^O~aiisbRT-|Cv0;!7QJ?Kw<nlL4ztDHjKod&O;jfNxP0-oB4rb_T!=h9k^v
z#xwHrdpDG&Gwht2>bEwbQ4RWvnS6KNBN{_HAT9+Ugl^o3OP)`33>~(av2^CdeGIYj
zv;ofl%NI*{JQ|!;6yP5ekNlzAe1SR{GV2a|7$D$N;TwPT6=vAzEyBhC$C!z7$_O;l
z#uXKx!_Hr?Ai@UOfc%Zd{qNm^3`vFsIvx@t*P#S9xs}2(8W0_4vN{2-w^IsNL*{)C
z<6g8bP?K@uMFW;0!VJRB)G(Q!86FO9urnM9KY^}E-di4)dCc}ellB839J|d)Ft9xU
zFbflt@>QJowYCMC@=D>7Xf9=!*uG-h{(G-#CdhqxSBJ_Y`e%07uHN*$3$Oxb#|9{|
z-fAAzv1a@MJ}G$qeklihLVNi_*#m_6H%G~$ixFv>)%H!!fQLcd^1?GM13_bBl}L}n
zPk}4H`eeh<x`2VbH9L4UoD{-LI{+}X4$<*AS_nvE(<ipGFe!#kMWP}?h2Jd!IO?em
z<HlygX!O}GoJs^>Y5-K_(#=mBd*pvwaHfGsV*BAg&*u6K##~N_tRLV{Ps(Aet9Go&
z`QqJL0uv^y9Ezw9Pi<aiFuI6@@m}E=6!}QeX%}tdx|lnfAdKI^C45=k%!Y;LEK2aD
zOGjx6ACFzBUGUcD!#MbeW`mENF_)e4P3z!jKnnOr+;kHa#~h*1(vP5+$1jqkhmCcj
zlJYDMmd-%<mw^HFf!<ngx2p;@7K<^{b%oboS~xA{!MnshsD|=?JbLK2=yKnQMBk**
z;-ELFA86mD#J9k43>Ju})Ys6@?tR&kjORx=>zwggqr3U`H$Drk9WTF16!`tO5v_HV
z#tTz`BVj|ftGotfKP|AL8>1QPNFjE+J(r_1&uWt2{xjRLe2yh(eVw~a-~>!<U|^+|
zi=KyUC|n=OB|eB8*p^Ykzgz&{!83l;YR2;Acr}<B&V{X5<&cEvQ2dz_>o=$QJZG3!
z2Eqxm)ovZn^N(V*1)oT8;oSn|oM+}TO3WcqgQ0~{Vd&%r##+VFwcImuKE+ME{hvLr
zJyCnkqms}tL!rlcM$wdmtF@f0CW};@GEjztIx{r^z+1k#w2!^6nO46Vpk_9n27`q}
zr3(BeUCY2Vj?Dmf=@?KeBWJc<q;1TFf=~m`b24bhh+1w`jB(jG2T{MicxZrJclE!3
z4Y=OuPKrv;3ge66`G?)r1|`7t8d!Morm9D7Mg#}XXfUq&0`8pTwa;@Lx0avX55XV!
z_E!uT0P;|71SfjOK%1rXx$cF%{1FHgF@WRS3IlE|FF+unR02$))dy8S-!EU0i-220
z)+_{+MLif@gc*H`yPC0>@FV;?g$3ky4)e~~F+woe5Z<r?5c6nin2S9ShKSe9!=K4>
z{5RpB>~mhGDKF>CO7-JUzrD!YtnE@fS^##*Jg`gRiVVPrt<`hC62hW7u3m#hX2j_~
z^>RHZM)B|Mg|Z?iKf;@t`(h|C7LRB~Al-ve1waRZg)j>@4o4woYOzgd=iWPfZqr>c
z+*IP7bbtVTLbOZAFjZTIG3EkGaM3vTGA4#Jf)j^bi;48MrFC`yt2O5{7-NUXhB^7b
zK;JlJP6FLjzd-un%tx{DXOrX5W&46p^C_f~cyU!d<}dEvzZp2La3>EgUn$QOfg_o!
zl-yjC+F(3?=#pYsmuHdX-l)q~J7BKQ31)`<7Ccb}?)43+hL(sH9Wtgkd0AtDuK?o3
zn!H;ab})CyG8XTT4JV-~F#g~6-fg>$U1u6CfNUvqkFnqXO}|(o0sI#LNVY85vf?Cm
zmT<9FrfnUXjV5s@RI#?$6Q>>*4h+Uo0kqam20)pmR6oJNKX?oA5oX%Z$Y5Nc*tSxC
zWIa<#VR7*f)_Oc4o9vZIexMuyoXpd36*X6mjT6@#M*kph1z#zm+*tEG68v+-JA!Y$
zbhJHpY@!-{aa!IJ1<*BER{-pjAU&2Z-wc5Dn!3zM=;$T1ViiPWe5<g<xOTwSWuh;?
z70WtwK;SaYE0=+FC}7>P=v&io2q+s9jwcuco5<$KAk}kgCa(Mj25kW2<Z9-MW=tjP
zy!_)B@{hE+0<VI3D`2LhUxoy51w{)Wj03ctjzI`?jOH4)#wtzv85&4R-)c-E^nc9u
zgo*e3{m4(=ZOYdG^)5dmgO3#xt!$Z##`n^9=OrkPdT5T`cFxy02i|!9>sLcpxHLdW
z?umGSN6HaTZ44oMP)xSKak4XBj$7f4nF<wlJeN03%#4(Z)UpBMzOVfRD&oRL0?_7_
zLGVEUl?oRr^S!)y!sMuCn06ynCT!W!5<{06vlr9T;S76mL9obSJU?U+Z8~@S<jBFQ
zKmI|R6^?iRG7;)`e^9l!*{_TNRN{eo4N|yN`NE2u;5pU8V`sIU8z^``0DqPDFWcyW
zm;zSi88|P8E-t8Z43PMhc19uho0!-ut}eDU0px{>SFy^r9Sg+g8^q2p=hn~KPS^|{
z6~PL*U`|Bfd{`PQtf2y@V6DV4Qcb8p>40_FiZ`_&IKD*U%fW=HY^u$?^qB>mkwkDL
z?T?CreC%bfkb`eJ&PpIYadmZi&pHdKh<f61iqJc9fAxt3yef1WVjOQUVZG*?vfwmv
z5ug<r8kNkDZ;Bkv8=Ipj$dikI*0TKL0+CtFhu=4B8L|rqmCLbiRbq2CIlY|ECp~ZN
z1r0~u3Y3)Dy93$`v{kAF;@dK`WL;S+9W%14g#p#AfnW^<T-en;fT$JvdqEp4N;n`U
zo&}hpiW`NW&uUrSb{l~2trgYS8G?`ni=+54HTs?24ae)v&m?Z(ETmr51N8Vk>p8FE
zz4gD%ygZyylFk?@>X6GH6E-RYlOkwQloJ^=Ob}+}*Z_-Q6EJYxT=bWS%2s(x5|3VJ
zI~=>T{_`Y$E;UE@909>*3a<qfLG2)8Dgsj+L5Hm{pbg+sHi)QfDh6nAr)knU$jZhX
z8tpMJ<{VjDBQmVG`lxg`O7L^q<nOu9ou9G{qr{d&GdBxOI)4OY2RVH06I6|+fbzmw
zZ<l`>JSsQ7(M1LyO?J_`7Ns-=tbirYa<N|GBJOOuX;p2&cBakeYEVSLzz_CwxVT76
z@tTgm(NBO1Z*({Zi$b#kt_Q$ck?tWjTsa4XLK*omZLC6MVA@a$<Qx?f0XM+Zle75N
zTVsWYG8I6vwRp>-n=|y-&^92$N7OkVvS<$8`5BeTN%U!+4;<Q=4&qG&$G(Vp<%YLB
zI%W()uq&5qLT~N6D+?ed$9T|T5`mzsPJZqf2Q8+>#MIm21w~j5%577IHULGKjvQKu
zKAIp*14}z+?6d(D)ubQ)&s*K?M~Nb_woQ4{(bJLI^AXt!6E-k0H`X!gS)8#c6=Yg>
z-p1y69xFQF!)XR=Eg|~Etd<SPIH*7$hvv}z=bH_}gF79EKxW3vL6s;szi&4Fl_lIa
zKOI2KS1<na>P7c_!GUDY7Oe<~*&#TNjbct@SyA==X$h*Q0tqh1(-9{yuoWr)jf*MN
zPy+fZ4F_SDq}OIbRyQ>CA0Sv4?LL|N5}no(<0}wBE$nhfrEL3e@6Er=07N>LF&1l^
z2QS|F(D<{h;S4d5d9k_#xg2NN_~U-_0sGjSCxkc22G_iK0X4IFKl45rY%kpfX~_rR
z)_apHP%s5&m!9Hl51XZj45NTv!7mP->3sb0lQtqK1TTCrJ^t7TTF!t4J->ATn^oFi
zzMFLG4S@BZIAJ?caoG?Cy0D^b3>C$OoN`qISVA~3Xr{Ui213U1WHI^zJ;xT9+-AtB
zrvzTd(5&;)e;f?>`_8t?1ucn&i{o(zn2wiXU!bWOyNdPe&cSYsFy}q<B35|N2d|n^
zd-~#Ul#Q?#JY@D-C9<wbRICEpV%lD`p=db@glbiif`dmY9VS4fL)plHvK2@ZY1)5M
z;nx3CFx9ORLC>fET$+dr;M@V8J^F$%xIk23V*0WN3iw|%2QC$qvjFNDx*32fN=7lB
z1B^Mw53=+By4&0?1JsR}n;OWA9g9;BX-3~C2RdXS2c~6!ovy^gQ^$JT)6Yo`IM8Dj
z)PaLvh6X~bPY_Ij6k+gsoCt8*{*UTLweVKYd5WP(q5n1l7>BYQ(aw6R)@p~z&vOH)
z8~|ubW?k`F8E$HWl}A+&eLa(yH9f|iNYJwrW`O}M4wFY0*Iu`xIhmfTj(b9dKWG&j
z%3<lNE$*F|V?UXFJ;2M&C&x7g8;NJ{Fk7J@7lSeE$K3<Z!kD26hSvban8SOcbfaLO
zV~A}t0W(IaVC<vKuBK#y+Oc&&dzFg9AOjt!@Kykz(y<zF0^#tZlC@Lo205Dog>CBw
zSa(1h6s#^pBQZpb0t7+yfq5&C4mE8CP#a~wpxSKM)aT5YC1^>?Kq|PF6znlUkT>id
z;LYt{`ZiH0X!%DfoOBB7XT)*4+^E?{V`lOaC%qv-V7-o4KVVMvhr~7mFjZ5K#^xWU
zF{WMsvI*w$V`|``7C4@;oZVFl6IYbY$)t`2yV2Lf0l?4NIslbzGxSzFil7^Bg8zT?
z^&0ggbpkp&!BBaHc3`i^ivnm{6R+XNOp%V1)^h&%(;ep{U51_2mGn$>Uqgw`6OSlm
z_Mg{a!)44Uko$%=sN`)O(EHHkRS;3k11#MxHQgPBp_E88I8wYCy>Si}vdS2~R|T;3
z+6&9ipY((N%eRRqf`Lv){IpEcxVHaF?dFwc!d7240QR$MDMY9RvNcuKoMqI4Ewlv%
zRDrGG2q60L^ZC?MhQfwe$y_yG<^Db{@55$~HZyAOh351K9@Yp)O1)i3m@M_%bsr11
zJBB9CQC9$cJoIW=4g0WBkdQ}HDi=EltcP#`wo+y#K=Zhz!t7{Z>)y{S6`3oWR`qV(
zrpKTEv5BDLhyTuX?>2ks(g30Z12{0X^3x9w;D<pvwsHb<;f;WXhQm2`x&o#Rl48o$
z$xWhK);LgiZ3n5P6bMVATqeM3w+FZCH|$YYtinQosf0?4TL+WP8L6rJ6D^?S;}lL7
z;4Mxib4=$_h#Z6M={b-14P}#k@f$=Em(yItrIuGO4zgd=f|nn)JFthL5I;>!K`2}g
zDrfv`K*vGb2=yzWy|i}rE0B!<I^gW@LVy?ueLEk)mFq@012}_<i+FO*|0~(%NJG}Z
zzp@_Pw@;=`n%ANPPz%pV%IIvh{07H))GDOpW+TLy-Ead`E7Yf?E)(IsqbfUjicLs)
zz%3uflB`BG@D)U^KOS`ErcNI2M1TY?9QC-^kjy}SfSyqe$}fE2>~u*-@A3lxm#VIQ
zbw;<24UkF|J-Lm}t#woadT(?rtXyhN2C|reWdoH1RhVwp0BntT3k-WVD!`FodTtx6
zgDMJ_pj(#y|BVj0Nb5LvbKQe6LW(%*e)*`nUQ7O1nZzAOu}nd+KOJPO%4URf=2|9z
zY(QjIF|G<=S0o-m>q3~NL4W<xi@ETdL}GBLk(xKDDuNFcjGU#VL9ikyNi3Y~TxWoe
zIUcOsU$#=TSgDK-DrYgk%*^T;=;sC(4YthCwhgt2oO>Vj7{EYO$cJLhP(VeFF)34g
z#;sIhZlJAgt>%oi_w>0)XZE1sWZD{^L|_FFjswxU1Ct(&NXAmW#-2QAH6BmPggTVo
zCK(n0uo6S9#qUy&b7&M~qmcoM4hs$Wx;yrqw=acPtl4m$T#1@TS8*<)Bzwpg6D7_S
zT()L_yc__aL-plg018OW1kDk$XuZN(LF>H%)?06eN&u)F+M&W;W&^JD)YR^kI$+)C
zfL7O%9}5ivj)D@NKj4ami=Y=6^r(#sLz|g_ixkI{_7%4Oi)MHI=v~L@P&h|~=LBjx
zYdYD&;#ax(_{e~#&NWg7Yfy}Pr2HPl<L<9LHi|JS112tHE{#}hR=9oUXWJ2&$9Ck*
z&ID6E<E3c-4`4olx%BEY#=Ukb@F_$?e=#V|t{Tv5qZ(YBg{tdDJ1bz&;+a@EkSHLc
zisGEV1cii(`Nn`r<4`d6X|@{p+oJ-dF-~f(t0fv*ECl0U+%G;aNUGLzyO}FY90H>Q
zuuuSS1kV?4V9v0ksSe1>VD5uG&%i;{+V6J}3LZ#gYB3rfi95_ZiVJ~?m;$<8>i)88
z-Wi~LW=1`efz^9NiUvSlNJ>2yrj5C9^+xShheDn~%rTwJ5VTYREnNNkS!>N%JD9vA
z20<+*9ja$#+G6)nKeD+H2q+NZ!cd5^Fk}c&9sr;ne~!ygW*yjt)pLmXz9~J*I_(>e
z>Bn3$SVtXDL#?}T1+fJ^C_5&ExhTXBf>&i2Wm8K1vG@i9?&yyuV3;6{_pTC?zrk6c
z=FaZB-%8FR#e0dS+`Lvy1gGsD^<@~tPJnC-pepob(u3Onfe&>n!nE<3U;=JiVN^iO
zI=UJ_&Q>jG#z0}`@v&z8+?#I}J|I!nj#^VFH&m*s={H_?8uLYd9H~UEO<){=hLo+!
zH+ASWEAuE}2F=s^o<lV%_HQowhdgIFXRM*HY0Riv*69-7#o%N(-hD!<2Ahu-%zky^
z0Gv7|xF_aI->hb7(m{IYL@`=4h>ey(JQg4Mni2Y7)yrhT1N9BwpC)5fuoj=Eb7++|
zNZEuq+T2aKG?;WV5ry(rK<I#Oom!!CGy<%3JNI$p8UQpxMMt`N!&VUN+F7dcgI^R{
zv;fTlSCJ#~Ow4!*znO)Hr~^8Xp^szY{c1ARK!V+F_RvLP_7>+h<<nnxH&C>42q9_;
z{>AL?yj2{qdvd?y=>rNfO}@ARb)&d{cOS1F%T9<;oJw<=a!hE=kv}RsrW=9_)>}Nd
z1L&w)J;2twaY2Fn7dNd%X11Y0HxdBdOvc!dIW>JzW?;kVK)s;rNxPwTbhV~90t7VJ
zS%D2r#qVnss5Vm+C<`Q`facelL!%8FD{yj2ihRhj`bB^R(E)1(QaCN9^yf(brLD6c
zxBP~Y98Qy{gl@c>noIU9o@Kpq&ddh?jAwy5LzZ!naGV7OsEZQAfQ~5stlu>&h>P>w
zzb~5S*L(cDUjeHtzc?m6Vh*#I2JGWd@9{K{CqaUeIME`BG-T@)DrXof7$Puv5#>?<
zc0pKiVmBga(Ic5k*RCL-OH)&v;HPC1ND)GJ9EzO@rp*yx<7OSz;W{eemuFQkCF}Wg
z^3BIq-G#tmU*NN-_8@;}<*QptXsW&!4z2q3*r9v(`;P*yyvTjQ(Da&v{N6x_GDYk>
zyyj5FYko@RKrdFnjHzR<E<UKs?sS)-5oGg`8(|%%+f)zVMZv=1DLIZjSVBZ|08gP@
zn0d=8Y*iNSxe+x1n2z8I!BKG9Afior&51wtHatuNy=*`;^lpGbM9jROQW{N62LM2%
z0B9%F0w!IyMq-6Z_aaqL5OxVR!r1v$T!>D=s!1U2a1|s8L)_1WoBv8A@J5Wo`}5!X
zt0DAE%rm97Y$Q25Oppes8mtwb49SYJWN<n*o0Xo!0~ejs<Lh@`d2{8u!|x(}F99Q>
zM}L$Udlcj{TWoOOUYI-BU5CuFiQ>2xA(OJ9Z62{Di#c%eBDoas>F76~AqEwd?i&aL
zqGm)|plY&-cnB_mDG(L(z&tlktGq0nYhKVg(BYr&pDhmmlJ2Sk%_avBc-t5Xm14@(
zf;pI>WENvTb3gFa$}n#r-bVL}0iw4eg%VRL*>a8OmL|st_vIl*I8KO4D%h-Nl_U@h
zsuYtPd|3lf&N)sQNUud!VY#($+fvM=N<+|ibw46lh>Q&$j*d}GjYTObND~FWDb!j#
z*kNtQaxqAH5_PTtuPqaRS|CM51wBXv)#teju@o5e<Pn!?lZ=B;j`FE=^s8@F&>bAz
zgEZBvP0d*VEmioWk2uW}{X>^HNYxq?+5}{xiO<ZZXFNY1>6mdog^BIBuP<13#jgEv
zmVR?ZugdkB!0e#<stOn4TIcjAGXaKb1f33Z@L))FnauS18_j<Mc&mUg_N3w7z7pxy
z@lDJ%-d(;h#X<9J6)@+{_hVojk0IMpyyEALtkrW)(gcM(D5RvyQ5L9J2S~6nU@8sJ
zE#xd*vPPvfbr6<0-=y+Z?7#`Y9b;fpj=nwuqZCH(&BoyhuwyNNhFW9D#LywFZIB|>
zARsGCfp!k*jX?<DMvNmjo?)s>`hq=iB+a78>jaC4*zqTRtuX|{obU-7SmAVx0sk9n
zV5%u<bi+RizpqE&ZbNZw!*+m)^v?9H*_EU5BG;854pwHL?+I~cqvG{`KKdIRwn|6f
zB&~@P8xf;ASAe6g#6$(mTj2t<lK{Mxa7`xfXTxX5t)E-mp{*Pb5fu1>HB;qsJWSL+
zp;AHbHdwv`-AF|&_&R9=LmNz0q;==kH47q|g?tGEo-`+!Zr<(vP02hXmd}=g!8+(P
zj;@_~rBqr7&lq`Vu4=KJQ20*DvG1~K9c02IXT9zdbIsmyA0wA$u~btZ&f5W6M~w&h
z!Q8a{zXieeK*2GPTXshQP8$GJVJ;_#N{i87VgdO77<2?sASR+{h6D2kbOF#i2m51%
zL0SicSz*0t@IMxH0z&}Wd;qlzL~Y2nD*&qsG2g2Os+Flw?TkG#hq-DIRx~41Ef6n&
zWSPL=95ry@;}TuI-B_+`)Mx~=DTG%z^rCftyZ4S0FKtl{U2kkoODe+Yb_4&wm?f}x
z%20a8NU|lyZjd-z$^v^VzMSqksS|_Cfm`>yQ=>{9ZsVw^i;Ic@JW=t0U5etY;KIb)
z>Bz0k6*~hsj7N_ZMispX+86{s!%$F6QLsVV#;VP<d51=TBC<)miJp;U7204hoCzC(
z*sR9fVW1ST0chMyYe&@%go+8r^9{gkjMHm2(BI{^Sp8-sH5#58Cl|+UzdoH64LGkd
z<+ib-#o0OCn8)@#fWcYVdvnNh<WukMH^`nlGW=W$kHd2)Dvz!Myc=U4hh-_VerO1u
ze@cA^-Zp2A9}{zfu%M}UX14*@oTH*Up>6Y9$&G3>td$LN_6^5D{!~3%QZ-46e+Gc<
zOh*Myh7NuKVT(Zr0Bp3elt>7+292W@0Ap(v;0DwlPVU9D5D<&=AmEoETla~pu>&zD
z!FZe_elYPkw_H3)-neJp_1V+f!f4TCLFRdCJ0Vs9ZiZaHYUUl85z}Oh#@XKVpQb!2
z0c*}h29z>@42ytWfssf7`?k=nMQGv-O@)h^c*kqwwt;nP3BG@cazxtz_Q~CbP0>W#
z8Q2U?2j@%ET@po`CuLriI&INjP)m+%Z(x<eMn^GJfdX4&Mdu-Sy0s;{9>Z2Z2@)L`
zxtwjd#Ynzf-R<%YQ9k||ts<6@2BL}<^vrdu89Z~U`yk_9E-<G_UrN@AwB6N(Ki?vj
zxd6rI*IPSm@$a0Tp&ZGbuPVI=%!2Y#e9#Jq25H)5j9U3f46BGLctfdeQ{NPFi3?c4
zZDW_CZH6=8K{B^Z8axefc~CXx-q3^G3ktS#@R@i|dwxe%E)XUh2UNQ|BFcPc^1o{N
zH4@=C!XYyd+2T2q2=^5R(|uzV`3RrMnU!Feeml4bE^1zzqwY5wlgG*-@wLoKT|f1R
z@5r6*G<Ejfky0fn40(NsHE|qStWP#rW0K93=Di6@C`w{jjakZu&3md_iFS(91NbF}
zd~>j#`ZzeZ4S-8RFjS<evIZFn#C!Dv0AlQcAw{U!#DkHh02d~3fta3wtz`T%_5zW@
z+&ma-OEtQn_%|#&PONcbIER9+9aFDw8%@x-FTM-QzvDQuam{vN1$Go}pfdE~@+RE!
zo6oX{iY#ENQ=N+Z4n(k8i7C+tNt!;#eFUWwnL@C1&J+qI@CFsYjN6W5lgbZi3J8=1
zC8G#2q8!%X2EbIMfywy<yl|u3zygW$M-*yM=H-^O5XWoX4?3=3p!EivD<4wyMts2w
zi`OPqR8<tbtB*O6V%9J6$MN4ZkQgz{7oj5<219C@doY}XM?mtfz1NZ+nwtARGz{fZ
zu-*k)G`$m4^f3q0x^A|ShkF9{%265QFbmI_`~*<!D45NFwpkC=Jquh0qeVVoD2fH7
zJXR+krj&OSap8jC(MRtJ{Ai;tqtLS@tRRN~O(M%Z8DN#Qd*<$Ph3C!n{G^EjT+GFc
zn`(;&RRIdrhQc{5wE5{x>mCH`5JE)ouC@)reWrb#?oJRx<>*Xl?$SO>RA=QnSN=Vm
z&-+Z3zhFiqzgoMrETeBWD}@GSl;|QATTj_I{@Ml$y&^M%6e!S2G`gI6=}<;(P5`1(
zGku)0fgb)eplxU>2D}^+SOB1@nvtFlU}5KHSA{B4;I~m8rsnBV(@j|<+7a%m(!Bpk
zu5W|_Xh{$d3#-9)(8q!F+m~SikP#!^mf3e4a0;9G(EBG9>TbjFF)#zrP&9a1V|yR?
zJ>th^>K+1G=Y}}0DZ$J}nM`K=kztw8j=73ZXnOU*Qk1ALLUYeON7+wo0dlerh)lhp
zrsE2_*M`UsvStdAAc>Z<*S?kxxF`<FuC1&B9f7GSiYZU(HmHx1IL&e45W)hh!C*@j
z9Ta*wx>sM_b%V3m;4^<jnN)<|Xq0@|2H?tzz-D)7$X2KRu>D8Iw=)G;EeG`?NB2hJ
z2VJK;I4~YH>uh_Ba`BZ>pePm(keMhrBq|g+PUC;U#e0Fwj5gCojr$DV4%J_$0?|F?
zpsx%eQT(Q6PzonCPEiO9bbVLavZkRryN`xWzV(U1(>3+#*y`gb!elhWI~rhOU^Q>#
z`f)JV!{V_@flErK7~fx`b^$V@k*+J44dkixyEmrJ&0|k5OSz7K!y97c-+Nk&j}g)l
z@W{kC7A!{UeE9p(QWK!OL+VGDW&_Bi$a)i^oQxt;3{n_iEghzab4labi<ukMtZy1Z
z_{AGxY;X)2RDCNoB-2WWvgrorZ39e6*takh_nUxfniCA0L5h_#0U+8&P&eRL$5WI;
zfdKtWQgmfNoRBF4U_KI*hY~30&O63yG)t^wbqu1o7yXWU{l_$a(=%XS?*Z`}gmhP`
zmG&8|I7^lDMnxdn_fSRcKHLx<<N8Q5SCk^<d`P&MX_?LkFwS-8iVh$LJo@*?bp*=V
z3CtMLs`#vhlB2fc7SrZx;2s)z+hju%@lzqwHAk_ZE29Bh=eZMZEo-kPCQ+nY2R(-a
zc^>aF+o>tERQ!P)7`Xy8wOg2}D>peQAojB>R8#M6{khNVF`$?d<99jEyI<p;SL9fN
z-%r{#Hy4O}tV{fL87*F3DwhN<RB(V-U4rv^Vu?yA%${*b3%zZC(A7bOr7%QJ+AP6i
ziXMR#aoNRyY6|ed`}6+CFrp2>$;1T82)#X`H3JMZeZ=HlebT~T5`Q9-JADB$B*0<D
z$n&V$h0(uIb*o9FHT1kK{HAyruJ(Yo=Nyd>cf55Dep)s86&iz=NGTN}G%FaoK6?4v
z0#|6Lw0r2n|0MT9#rcrA()>fF4J`=Se2q+?P!@)ooK7;PcK}dGK(9*g{#FL)`5*}u
zS`t8l!=2#;d@>l+Isg=?FwMd?xqO8gWsJmLN(guG6Iw^>A*%<L^~e0f_3l4jKcPeR
zJ_3N~b6eH<RwWC6yB&c5`>Yc2TUPcRw8^dQtCnD(iRD@XW~Sg@%KW*GbE?Gu`Xd?B
zOAbpq1Wv!8Y|H>{XwsxP!`E<I=E7+G+$WJ`mfs{+QwKI01_3eQ5F>twLvz!Hk$4xM
z9yAkYxB_tjhiZghW+XbM2yz~RTGJ4_xHtf)j);DVgXT0>{^nt3y;_(P+_BG57(j76
z@j5EyN5P+9`HtJZf(XZ#!<q+StVkS^PTRHU7Xj$Z>!Vk|1s0-rv7Z!e$EZ}tgKlS4
z3BM6YVYK-4f}Z3NzIdGeh$tb!I|JfK9f}~ygCV1$+!Wo~yr;@AweTAZUC0<$;)E4Z
zks@S(kw4Zp&{Efop~0SHbg|#W?;mSa*XFpTVgcOux?-;!3FVTMs8X;&{(<c#E3R`v
zEJe?wa})s1Y2K-$i|)tBHkB<$^}te0*mZsK<4m{X#%PM}g*8A$yz;Wv5@g`-$O#RW
z+L>GNJ6p2O8?P&7nY$KcgIf5FUtUN!?jA`a*@21tV;zc5YP-KJ3knz3t8&0kdNAH=
zEGYBdQyv;fuX5s&<ChNFVl`NUE9ksxK>abl@dnVkG+w{g6p%E1omvsdu^8hC*IZ7*
ztT>PnN=gll_2X^h0-O{;E3E`uO~`>4Du8ZH9Sw-Je+n3Ocep@}u)pU~#g0;|h7)Oo
z)@5#e{zE^1nZ$eXi5p|U{CFB*k$~rnzILwOQVNd$^~f|cCt_Y7%<`d@Q4N?9QPizD
z2l@<+m|Y8IE|LBjDGI>p()XXWRI8k(cId9{wXGdWDMr4MugRTEK+V}{%9umXVsAtg
zn*dEZvKxT{R#w?Dew&3Sl*LQjg(56mc94aAMNzWo;+5*03=3Wmn@&J&Q=jlN5JjPC
zywBJ#<HeV)hd!DMuf1a6>wFZH@e!rnnC0SSvz_bgEk;>FoImI{*ZLZ^`z>D%7{#9$
z^%xr`MHGy6p1RSi;30MeLNx5!Htt2@)G)AXQUHAso12Xx9D}eR+?r28_$%ZxMl*^n
ztQbVSFys|F256KU5wo)jA|Y5bDvFK_d%0IMPUDLQE#7Kkf3KSdRlVNb;RF_l-IB{L
zzDX5X5Ab62*P&=eVMm+wIT1m-nhsCifpI-Rq~OF`NFP%=PT7L?GTp!@KP2XX7-nL@
zzZTZU2j+$)I+M_4yqa>zaOsK|BxhY_5(uaVUX^@GsQ^F|1Bi<=@S^3pmf~|>yjd^w
zt=X3Ac4N5ay459!NjdS0J<XO5(O2>th-a$A8e53DvNex38-NvkL30dsxn!78cEhsa
zsNiO_ImRKB<6|~m=2lo_4i7<Xj8P3!STy8$EMZRU$g8_#GP<^A;9iw2q?ELofvaM~
zLUALQRUKyhdP{Mf;^mHadp$pNYiG;B##GiW$X@giuZ?B>F6`*NrUj-0c9eVxW5jvZ
zf{_qp3x4~#oQ4jllN%4I$df0?tm3mx<O16n1_TOa78@}L8lof?oqX5H7^reD7fw|J
zhAx6ns?LEL2X0V&IkhKP`@CB*ITl}*cC^UzkFf4|F+yV>+*)qf%8!#%w|wC_qjY`w
zVRsWk{Wddow9qnjtvtGcKBT3Nwx=k=innNEFy2uWY6=&D7|gnhZmemu`JaN7*fc(8
z{x5($lwxo$M+zb<oSf7K5K2tBxF3v&co{QsztQCq!v>yKZ}twa{LC}uhXrEODlY)7
zYo9X1pR0cJfE3QJ*iKxJLWJ0M+&e4pSIp1s$aYKY;b1`Ab96x6;YUe8@&!OM^;un4
z(SsELR87_+X$-o@O0il{j)O@t>n7}gi<X6EWS<-`M=DcU!+3R3BlEt5Y0;Q+*ik|F
zp0k(ZHrWGZY>Z_GA@wibxRH#`ckzn%MFq#M?kvi9T%q<eY^iC@l}!2!1yU_Um039j
zJ6Wyd$b&Ot0ZiLUuFXfE?-dmtRl23QJ~oF*Shs9B;Wrmcm8Yn})Z%LsoNGYGA59}<
zIF%Hx+BQcdHcUMH8@bR$Y0vf-@k=+3YVIhbAl*HyUt&r!yLlqL8Rh)m5NIi2jgfMS
z!5|B@>!8yO1R9s2A2PV&ec}1QT*?C3h6)NqMVrJJIuV0FRJx*%9cNsjccd&{KKekZ
zad82W_92d%;l@hYK!N;;H}k+p!3eP>Taj(~4eVO`uw2_!yshmsJmukSr$&*ID5v<W
z>OL`;<=SiZ)(6v_ih2*4X(qNj%L&1SPOJ=x1Ay~uicl@1o~n?fkSxVV5tP-hwScRS
z)g{fE8kMB_y7r9Al6(S5IJayGS<(~Gt|n;!s1%#IH7U_J-_pz3i^c)y()6sTlw7|L
z7k4dsv3(gfdicO1_HUFES&SZ3hEW=fg9})s=<l_H-|9FSv{Rk`c64p<7NonUr^KKk
zlVdG>`6&DblvQXbm}>>Aw$@}hK!_75C<7cZ;wf8B0l!C4CIF)5XM-nUsFp-+G5~>t
z0*`3QY`gUO%3`CFaVmQ4ceR3OQKicb@bMidpv*|W@d0H6Jog*iOm)mw+@D72Z}uBs
zjpWbqCG{$}VZ;tAssM;_-*Jay(#Xp>Loqde$|{N?BY#kcIQVn|<>rw``F#+z<;H5f
zoQIYOL$FVyCgy6~#X+5Dpd2X+(Qc8Fz#l)9EObAYdidMh2l4ds1b#f864&UZ!DG8c
z#>_hN?8}6L0-5p$y!7`>mM_K^kwe^r7<~cnAdJ2<8AVwo%ZJ(ibe$0q(56oGui&5|
zcF8&~y^(ksnC+d5<DARPp<s6ezt;(zL(VF&k6R~Z#3+h^Glq)3kgZK|e>c0>{jQo_
zbh3V}j#3xI7@{di?3vv%d10xyo3X=Rw6bon7k(3)&%&l`6&Y#dFMS?jP_~d)eZ9OV
z0$AH3+Bq9p5uZrhJuklP?1CD=g|j{B-JwVOn0uPBO|RZ02pE};lS~B@`9OG;m+XI#
zEYtU#1E2a3SNOijULv;G^XkrR09be6{}MU8YY$+L?|Ro)-EPfYGTkWl`|$;KuULzl
zKGz>}<y;_21_Jbau+apI05_9aZzBT$4V{3AU=<JwR6~1Y$J~oqqA_ynzwSYrjbp@@
zJwkfSk4NwVT}f~Wb~_{gU5V{Z!GXsQEMxrn`KUL(_pLAdyN;=K$GJcWmO1PXV1c1^
zTM7&WTIi(7brfkyf->_q%_P^L@In3(<HcoZ7y;u{&Jf0_p48HDqT(?!uSmo59Hlyb
zwsx0x<<9Cj_uU-AfFAwMDcY;HI8Kl(FTYf^K>miZ$p@mAKh(vdwF8|9ilHHz&f=x1
znhC5A+0S!=Dg|vso@pi9O-SwZIM_neIVO+SYm{uNd--Ui2rBjuT24w1bjPS4J4hr*
zdY3=X%1#8}5>w1OhqFHfty?eT+JI6CaP9yrI49p1YK*T}u!l|9?x$sOoXBVmaQ@yA
zc@R-8#4d&GFPq4T%)!dk_*kLN^OMD8P=;Os$NcVI9+43taMEy0r?2Taug{_drSH_M
zgIWnW<@cxU>Ofv#q!QgiATq7}qwoD^%&z@cdQ%SMspVy@ws;x1P*IdD??l=SloFyM
zT5c)Pdd(&`tHe*{Kq)Ili_~R`6qE-SUtm46WN@Pwu&BMy*XUnLRm!y;2Sz*oWdew@
ztwuW}?bRLYuUi=|k*x^7EMuDyhDD2lgMHGOz|D|hwZe=CD_#OERcEVQgPew<QCbNT
zQnP%J0Ypj4XsdIt-PzCq)E!w;{0bL-GYT^A)dYT*-53Fu1$u3~3|55vP+o!lv0&yS
z7Kg-Q?6`Le-Cv`5egxuSJo}E=WGu`o3Yaxd3v@J20fkBOxdS`g19@s4oN8l&Gj6}v
zC$st&raW4+7Z!cmaoz&<vC;kCVD;87B$RGyfD;lnFj5b1R_Jf9@qOGDQr#GmBn-@b
zOxX@a1h0GMq(Sm*Ccs9+FU({D8M4;yU@9pC+c!r90Wx!a1?&zOouFcW&BBo9R8Lp4
z(e-VdZ=t?TN&efdUP@f$NN|r{!@LQ#XL_3xs04eEz2rBVPt}gXWrk}udU8aT2a-+C
zaTMJcOiWP`orgJV0gUq##z7Ok0nR~za1oy`2$h%h@X6scU&GgR_}6~0l<Pyt()`we
z%<y+&&q@aXV>**K_Tk~tm~l+1I5T{tmMFlwU(GJQlE)zyE<gxYrWHUlq_2R{*!H(l
z6p-1j!L>}j`H>deu{W$^YjL^^iz=k?1khw@5Eq5-oLIhMbD6DC{R&G?N8kCz2_Fcq
zZi*g+$1X0z$sMB>#Sj3|f%~#XgGXx)VMQ5ksmgirFNqs3Q!hrjn6F$n=uD`5NMS+q
zipnxa#S1x8)T6H<Dizl|fjcQ3Uh3qJbdiZa;c>72)TL(cj(UofV$^nLJfnX5$Ry71
zv#aO){#)DpM_jQ>83DG?End5rpoN<*<vJ;Y1l3f{2pWP4`)YgqNB6~yt<P@M>!QE#
zF4v;O7Wd#I`--sk!FQ>=Y&o3KuZ{VlT%f!YOX<O&e>=Fpd0qZp6Re_0@)=ds*h4XC
z&^TC^_B+l(dN}{-s2M-ME6C6@(B|Z;o8TIMRZkkc$j5IkTA4xC#a;SLc*>{b!PR=b
zptzehmoNbW49ZDt`$)g@Pus|sU5rh|nPm*BM!g0)vkaoJjBK$?w2B7QCRHT>wxluf
z0LhUjS%6mWJ^{6?pe(rNbH?FsQ3#Ke>&g{<p67hZJ2`*9tLgeV?<Xgs**iOTuaPfb
zYC3OsQ8DU3`LzW>mLGV=N}R|D?+;Wg6PD399ZClr6P?*GS|t!T2TcOAH-OGMNyS6a
zIEMG$V70q39)+)1EL#;;oE>r~gX*YDE+DmGzdV@p`!0$3%D1Blcq5=Xgml2ZcOxBh
zOc)V3uO%E_kmX5h2Y{#)hM*g4WDarK`#G5-88?^hZ<S;ovZ&gauL2?4I}ETY<>rgE
z?YZJpzMjzndENii@3)s>Sy_<}_`<%O*XTRbtQ^NG$v|JvAyFmP@Ea6~r!D3kR82fh
zcV40VkQ*?{FS?}HY*dbu;hRQ!;Ecdg8eWUB>q-0|TlVj^SG;BbLZ*mEBr(hT!*eKV
zc^U;n%D=p7V+5B08sf>byE@|%CT}qqE$Q@yeaWuhP4gvPGBod(un*1b*=1e_`<0j>
z4E;hIko7&b)sZf*L3o4F+JCCbfxikhk)@YQCB?%sj#;GLqdLLDm|AfNzFLmn);a@q
zr3L;sXBr<giZTwcMlnO*6TygcT^c5OGp=hfsW!jPDrElz%Rv!WlCyI*pd-v>K0$hM
zT8WAdVA!C89hOm`>P7cEmXiW#5sk{+a7JxDGu6MnVE3DkBwcX}UscOY<C;Z8v>QXE
z#~9^)j}5<x+FH0m{%`iI;LxsZ!EXuAc*u=SZB=^Y{IwyrB$Y7|i=0*PHxEqc7}%DJ
z_ol?~d?QO`Z`v16_toO;>dnV7vDD-s%6&!a>nuR`ka*1i@_Ke=M75M;u!f<eHt^Vc
zq|X=fs^>MO5JueR&8TIHV$@fcfN(uL|1fpKXBZKHZHyL9*il@whK%r>8r58-k+|<J
z`b+J%u_Z)tN-(g!!$mU0uO>3Vp8x;~CrLy>RHc+<4P+KnR+aCesGSpdhS-vA#g`C^
z#pV=S+iY50$s`huYF}onzQgLW!tpO37nuc~tO%}^99Fv_P4lI${bJ*^37@0vSOG9l
zEi_XEU7A)p^?{L%iHm20)gU?wrmi%iy7*QWkgrsYX947!rBxLGdeya+8Q~4ljM!Mm
z?{D}1|BQX4Og07_)p;%UaM*k^aw3|g5$Zw60L^G_Q!|ZTNpEa7Jf-X!#>osYJ~ATU
zD{vbbN96Z0hf;PP7?pMXW~_JVC?A;v{@X^l+%?u`dm_G2guu|?ZVWa^DcQjCBB?5`
z>k$l&3-mVwLPu$@L+aNGLUR?r9whEfn|TMlf<2FhxG{7Y?eVevzg718h{M`UeB5|c
z_3{sL!!d;<X9NHyE=0w8fsWm2Mk>31?C5q<btkDnmp$h;uAIR5W>%eOldlA|qfzWc
z2-9zTasE17`YyA9X23-dVaQt<ECf<qxl90tmdgk!7b5=6>rv)ESdK;{(b3=?uz>E|
z{;J<+4+ns901?f_C1M&WRtTP2V4Y?OWc=~p{`#xYl!3nZ)t&o06O01KH*b=Zl7mT;
z5RoyLQ|_enmP=m5Dru{sBiGrB<F=(xNQ)i!xfQug=2gsfOQpGNa^GCqY7FImX*O(o
zo##3K#QFa8eV*?x-{<@D=Xt(&oae)md{tE*5|KBimJJRZPh>>G;2k_aNX6Hch}|xF
z@R?u7=xeXao0F>!xG%2wU5}A*zU?q0`mH{+6~mRm^}~f@u<zOJ%JI?DMeon>+jaqC
zVBKe_mCDDzb2sw}d-5!3g6Jpt7L4aYd)@syk9F@)IayQSR@+h8gaVUKFf8$uu<1)V
zYlnh0k}V|`Bi||n?eWL-(Pr)X#P8aReUq1{Q*{gVzo9gJ<Doc>`(}Y7v+@16TbTG(
z71(MAU+vHyGOKjWeCaGxLu@j)y5sd(<L3~sv<Exdz6v6xWje|86N3(8Wjq&`mt5W$
zgoh2{jv+3TMd;If1q_~2e0A#&8}Hmow7!+sw*po-N8GSMY*)LT?~TYgJS!LB5getN
zfY4DVFrl=}6iBmOt<Mo-U_o{sa2RFRcZeHtF$SCWGkx-p6Spm8Qu@PYU9<uOLAg6(
znaR@%!e&basXT^2Q(m)1TKLd>pINs{PPy3RdH{gnvW)j<{B9G?L1(?@L;~Scmyx^N
zqiG7|j@F25X#~5hg;GA)Yb~)<PEguDW^4o^RODh><n$h9aSQCB1alLY`+gmB6~{YQ
z9KVTu->gT1a)702J-0f~ow>9WXKkiHzGE?3*Lme>^Czeo!)Z={b6!|edFW{u+g^Tf
z-8}_H+eVa1e)dIGdda9zWB%FcbXUvrU7J5cl5*N1eUnR5-Spr$Gx7S08qA`b>KRPT
z^y3%Ej|yILq|W&9Mk78>a0pN$L(Vq6YaFB@k2w{W+$vmea+7_9d>Iu&Rc)zo7#&e9
z%5C-f#7eXxmzlo&8HlHN5=vLbQe!##-JVP}&uUF<aqcSl#u=dDz(8k<+3T&|<G`T#
zb2tqjh(oAJeSnBmINs(6+4)_nu&W0Ck5`zl!qy`bziQS)bH{B&|4}qOP&QI=QNcy-
z9zFr1<DC<a*<KD^Pj*r*@oQwdfI)CVEEnkfsJtM_Y53nFMG$mG)nf&6pb?#x0;8f#
zz>(grs~jcxwwkx=_)|hqDHO{@r`BmEeNd>&_*d<G+KZm9ycpG8fK+lpqJ8T}B;DTp
z(rQlUH;JrDga&NpgcA><o+uaG2Q#T@HtEQh{ITk2rzjf*Xc(<N2=oG%;S3dtv&D)h
zHpM!xp^;gM-*gJ`bCQXE(S$y~Avl>T(;pDV)Po14BU~yCjez!-6(*j;`88(tQb-ma
z|DtJPQ#QG(*I-lI;7J*W^lS4jF|~;Sz%{eG>nWQ$@#vh9@HSE-w;SmYEWNlCr6;+c
zu3lYpN_+6~=NTXTmXLk5pSo<x4nbxO>rJwHRO${oV*RNxb9dSw7|&!+E-(k}Z^37r
zR7KFU+nRxyX*WBiEZr}VDfDQmU=Afsc9)bIK1AFX2_Fw*Q|Ux4x}4V68_!Wia=X@N
zyEgznTQlK=-N86FmWo5=ZW`p#uZ{rhY-_??dShY~$fg#QyLCebkagPhOaP?t;6=$P
z_5+W<c(!D8Z*=YtSaKmXWeM@|wDH1QB{9zW<c)}0aLyxLmy)OZSM_~z=*ELqD3;}E
zM?QLR%W!y$-8LLW|JiVsBtb^GVg{~QXm89u+jyXrsmp}DEgQORQXvWFeVK<>DYHvw
zO4@!<ryb50jsb{r`(h)acIw9r(2$yy@6shh^}@@-G0cYwXvULHO=ZEk-1PyTu21`e
ze!Dzg_}3&j4M1`gC8ZAOptqmH;gzJK<~x}2QB+mzyLs(e_gNOQ*0go^oQj~I2P=t&
z{s|gAwDxDFiq1<=>kxR6rQ<!Vcw+9>K7K_=^!1yhUL0kZY*rX7w?C-ud$n`HXiu=I
zs?Nq2_EuqGQl@wkn6+bAw!wo2gj(vEhra=CLJSv|rN33gMx)9eLynSL-mfGNQhu0I
zGB@(b!u|YSlr>PAPBFx<-6wS(K6FHC8RH9{$J=&c$*UZUC+#X2_Fk-nV1B$@!i8=n
zj4T8e7=>O_&(CPH38SBHea5-BRfS6M(DBo&<%iTBZP35l&&@%m^KoTFEKXr~2U)Sz
zjvEATRKvU1tF}LAydj00yneXNi<@robOkL#QSM^c1^cg#a5f2*)A2z&Hoqp5RfJ=0
zSds&67-eYCSZ6Crh;Fx%8|^v|i5-Ds(^ta$f%imp;}&YSZju&#_`Q8-+J*7+Dp8Cz
zH%=zp{}jJSzAEqwc+PbPjLYs=gL7!bR@YwSPgUXVa_3TP>Xoos?*=6TN71#N`ssZg
z<LG<|ZRri@$@|@}S5V@n?65InJc~VnQqHf$PT2mn&hPVG4nzA<6i+DHSH$CkVg{aa
z^vefm@@TOaRcOd>n$8=dazy3i;s`FzqJK#t6-#G3(-;Ez57jFnJ8?xr55-s>9-$~9
z*u;6o6)wIz{;Z?!oxrXEHx=eayBWe8Pi1?E;2@*8y`$&169{kW(Q}{e)MsrFdnx0B
zJVy?+k-PZA+%2wKOx}OKgdfOH@`@jzK>SwlOR_{ko_m7&GNAbRhWpvgJ2)=(Q?YSB
zZSkbhOHSCAjICFEmocC<crNhB#70%@#6|v}gi0GfH(Jd9fgKyfvG5r|EcE}{;>OV*
tW#b|MwtGhf%m3YDUaiO8c4;9`WOuB{9m<<`wFm6!s^hgwbq<I}e*=n^<$wSH

literal 0
HcmV?d00001

diff --git a/tests/testing_data/ultrasound_confidence_map/neck_result.npy b/tests/testing_data/ultrasound_confidence_map/neck_result.npy
new file mode 100644
index 0000000000000000000000000000000000000000..8bf760182c10b79fd593558f38e7a8d1fec348d3
GIT binary patch
literal 1356512
zcmeF)_19L_6Yzft3q(apMM05N5fzjWE<#ci38j>7X(@sGBBZ;!JEW2B?(XhR3B|zg
zYd^Q&f8hD?Ss&MWuT%TX%$}J&d!KWi8wY*$&6i(Si1Sp-r`l$yS?k9d%`)Ulli{=4
z*)qJBCPVGU&6+o>)~HJ3W;JVt^M$K5__3DrKh~|*q?Y51*>isSUYhKm<jC}1nwT{I
zSN|n=>ZuNq|M5Q`{4WRomjnOHf&b;e|8n4eIq?5d4ix>peI({l`$#%e!yg#tT*p7!
zM~XgfA4&aZ`^Zcb{;PdtG9DrFcl$^KjKNZ@#wb)nGW@2^a&$yKyo{IdG!Bt95B<;z
zU!VY9$3OJl#(5mYA#BHbY{XS$)UE-RBOZC5qcWPKCd%RyypOl>CN63_4HMA}S#XoA
z7-Yf{ea(+u`kfoQ`CuSwqZzIvwev;r39{h_OvgpM?cQ(D7TvKN_fYIV9sO;W;VhEV
z{TV7C^rI2FVFtEh8$#d4V>A{ZoIi(u5c*XElW-QlBMF~}{k_<T75}l>_5}XHGur%N
z`!^EWe-A}b28EFgIq)?~p%C)mOMH*AXp9c%iC*Z9KK~upwGC~)Kqdqo`{^2va(Dq#
z+5a`-;8(t=68wSNe03i8k<4-MTL*-=zprgQjKVI29Lk0e>lT=T(D(fGwZurw#bRv1
z0o+0If7(YrLLL-EI=qCJ@dD!Ek#_fS0@KhMdGUn&^=N~12r;UHke?N>j@_T)np{|h
zDQFe)0^{UP226U`KJqaRK4>3lh(GTePbh_Zzv(lc#mRf^Bgc`#ena#|Bh17b+8)Ly
z?w^CK<gG<1@<YyMpf3hpPyun#!}&eh<U|ekG;#kNvQnWNJ(-aSgHXh|`)IArZ-{Ci
z*L}575_M4;@8YOyV~_zYwcYQ&`pAz9WOc{8`0K7?oW~U$#|HF6A<VkdK9UB9Z}SnV
z;dQLO#a|e7vwfuRjrNga*V{+h;Slm+*fo8+O5Wx6kxm!+@<RJaaod89pL6~edWSYh
zjx^YUYGf59vpb$5{~K(^Q*?ff4{^zHRlJJz+Ev7UB>b=4U$!H#1jn5pi_zq!!bG-3
z@GM3k9fs0diaoDfGM>-1k6b>ZEjpj(&r{-cQmk<dJstNz1$0IgY{fP_L38)~i94u)
zo*0Uhc<^7lTK3Ol79GE0Dc(4%KHv@d%i@Rg?ISxdhu>QA+d6b--+ulpdxZ}%89%T)
z7yIARk8}EWKwsDD_kDhf5w};w@D@Ve&Jz1da$z0L;6=Id5eg$alH!8%v#|@mU?E0e
zC&IbYID+eVPWysrfwM>>?;2w_-jvr3u?wHe^A;G2dH59rQ3%^y{{YS0Q<sd!<i5qu
z&)D0M-3Rm`AJ*#6dw5#ka_HkzID{Q2jcFlo_~0$Rc%M(+;+r}Ab(8;U@MBUw-Nw&F
z#i0Cgd3Hn{IK=k{jj_Z0fn?5Ybgs=lHF<{|+e-gt^>vflvw^(z;=9h6TWfAvZLF?p
zAL+l+b-c7jt<`3WwhP?T4+rg!+o*re?{jZ5`;DEezr}n)cB^gXBxK!gY|yuICz-q0
zw_EP+X&*Vojveev$gTuzK6AoYV1JB$RFx}d@kqZiT`&*JkGJJUFAPN(XOk|;ca*l@
zfj?66O@2N~&R4flmEZF6=ePXYgpU*R`RjbY3@OE7_<y`FE*Zt=59|@E_TrXaJ)MB}
z)Yn1y7h$f9ul`oWJcPB#r+=$s*p5)YKmJGmF&JAgfbI%xD8!b<IB5R^w)7JFHtY!b
zS&!af^tPcl9-V{n4t<@m9AWLz8>O(-wL99i#`9S3zF#o-Kl<6$L&&+0@Hg30Q4wkJ
z7k%+<f3v-VaQp;sI+q36&<^|Xo%=&w--~tFfz=2(9p;tpsEeTIu=d{|)ay`}|K-~#
z_UL0gyvg@3>+@Uso{UeX^TQh4#9NN*U?oB=9)ahzZHF7E>fW_@j-2vXf)r#oN0^T?
z(N_Rre(8=~c$yvWqXGJ04rXIGLjMXP7aC$Pw%{*(uP>p0U-0c#yvE<9P#f{F(Ydw=
zv092-h~xaL2s!f^%3}a-AcdHHi~<PjjXJ1>lE{NMkrXfDf%dQ1euie~i9TqF5c_wL
z2)}8!3zJa{N7+yg2ig5~uwOrlU?M-HL#S!n5$akJ$Bl6k^|eW%{bXcx|4#^g>x)ps
z&Lb`PQH1qTO}aNBjD`1%g%aq3u(sWYu+BY=RS4(WAgrx3APK_SGXcUnbSy$Ycar-h
zM$+{cJHlFG1iONLi4o#lUA$k#WI2!)ZREspIg$rw<V;JvgW+=NtTFZhIvW3n)q+qn
zXQ&G&)XZjh2dmYJ@tBKi806eYZN_R_&V8402XV;BhGm#fc6B<E()kABp^W3j&KGli
zZ_q`?0`fjlkA7FL%HR*RE1m6KJmq{k*Xz;ONZV}g*^MN0y@KoZk7Fc`U<_*GrnZxC
z*L^|$VREjkdtFfx6;K(U<6S(DLP&+J^nQio?5K&|=89|>rLI1MDr)Rl_4c6o`g@E#
zrJkKM#*SO_9g#~1<mDc1b{k(vX1~UM^V=c*aJ|b>^UE>*z<#W7-`nn=L{5M5qGYG0
zqZppT3dgyS9@AXEqun7iL0f!}&k!G1PlykBPsrVbMwo?95p+(Wdobp+Bi~_tVr#d(
z?IS~W@y`zP!ghV!#wT0Ed!u>A994FWxo;I8q3B9+LVo)<@YQPm#@F}+aZ%R!!DwcG
zvU?Y<vwkG!%T4;VRqZAJOS&HK1ihK~s2sM~AB924=y)=}729Xs!gpB?s-^sQhs{&i
z{zM;|=x0lQ8-7~d<)_PHQ%}796uWQ4dylw>`K^n5=q*=1m2=~90#|Sz+p!XraYLI=
z5Z0_YQ3Rb3)`+992BFPvB-g$de!N95%HQTgbip93Ldg4xIF0zm-+9|{cnxt~pXvUO
z$cnP@XSUv#dyCkf3Sa3*YdP0LpUUXlS7IM>ZR;s@8maaB5B;AnrtkB`0DjrWM^V0d
zTRc1RVJCiW!sk8t{<1jS5|e{s_wEYvmdVd0>hU7$p83{zv-owo+AzhsZ<1OvUQHXT
zzKk-~M#{kv`i0kqn?HxB!Duy94!i!cw)0U2okyF~wf|k)1NaSB@QiaiZByA+*5=g-
z;!amSIxEf)lbLG7Y(AQ2pG~Wlv4_1$*j`T`vdG83<j(Bva&V{lLI0BM<|9P#HLju6
z9(~aN^!)HdZZDEE739nIV`{?*G2^@6`0>J7a|$09y9B;ZBnDBj{z5#KVSv~)yTwnp
z)nD=Y2am-tJcGTjzJ~R2<%i@w!td6ye|RQ+EO&4Mx&Gu6g!dWYT}O9>^-qew`S7Z~
z(pjJGGT4CA^0f#X*3#XA&YpCIwN6-%7oevwJ+II+2M-YD&o#CsFu}Du+5~$}Y5x;)
zBO~^(wF(}SH4w?s#C&j#{O_<*zuIFETA(*dJHH*_Iq*|$OJXb@BaQn@;Ai}R(x`>D
z2y4W@@EU#T@GQdfMR-P?<KCx`h|Dk#?BUxmesbw=sMX8(<UNFEu5XYKA7DJvxgMTn
zKGXIPrnxVl`&S_&U)9E5lpuE=E+Qp8L-7dp>D`DI*zqNL;ReDy`2vD(TB0CAf2QCm
z=L;eRTacMAJL3&Lo`~S@MEri<wj0Wz6sn>(reGBgBD8%~+pta=isN{QtYTLh-yuBv
ze~Eku@2twBB(mdsOhR~9@rJm+g(TSHcsyz%C$5q?0%<Xj&Og}uJ$_|>=xbg53hyb(
z<1W8!#Z_c<Tpim{Q=9j+orV(b-Hvo*g=ew3h(mT>gmvscXie{4+{0s}W>-C|L_Fi5
z2+I9OdfT|T=-PDjMOY_wKneWe{;-Zphz4X2p>H4kVO{(MYOpOI8$V}rEipbT<~^`O
z9%M!xc`;Rf+>k3-&|Kbxb;?R(?XmGy4|9#RxTvVcer^0ORb$H|JvOVghwvDC9Ea!Z
zCEAVGK0Gfi!7F5i=cfAPHX*+NJv;F?w&PR0>U>Xap4C2{`x}ziL;ZeP-3jjtW+9<E
zl+5u^=WDoLjIKV~eoR+bo3z4otj0ow=ltpzhJ5%z+urV(#?FZQN0C#Jynf_H@i!7V
zeg${zufkm1#%s=PwatTDY$=UFC#;{2vl)LL)d$o%!p6hu%t7<iemT9@dSaLB+tjPg
z;=hsq@Ra?bTglkreS_=Y?oo5J?X=H)>b}AU%rRsnChKJ~%U}y0;SvU;3y!#c@{sze
z?P2#ML<{$)z&m&YH_6<DvE+}YXZdb2@dAd@eSW8Sp(|VRZRdA(joT!r*R$Ik*kQFi
zS!J!WoNtz@hl|)S-}s*=R&&+*+45?Zx{ZM|%{SP9X}E_ZznK3qM7u27Cv@*`3-w{K
z{*(9q3ipv;d@Wzmc^vmq0LhUDnQ<7qH|Pi3R&Q~gPZw|JbGCo4FS+%vXy`Xzq~n*5
z#bc@1<Po!%#X7tf>T%w)xjbkiA2!RiGjgtqyj+bl2<!KK7>DrQZK~@%wd<t)dw3b)
zUBGkBy^BVef&RFTYT6fd-yw9C(@C$h1<#`n4j`<Te#Ua_!%XaOzMA$k-9InrlBXd@
zd$T(#_e$$SSvj{t&aKj?Eyw88zm9TXy7))+d4|3}CC+b&@fbeY!$0qAVe2O2dmX#Q
zF)?3f<@X^A<eXR(7L#<d%<a?Zp5nfVd^uMBj1u?Z{4vB>8^q@R{M%bzcUAv>mU|uL
zUI#gVXOO?Wc{plrvj1);a=WMzXxZ5uq<z%6vF=HL+1j*1#TfBxtA-()d#`D;n*4{|
z%$YsfN1ArG?(M05;!Z#D7$#@Os0EY6Z-#br^g}M5U21)$U;9@XoBFr`&*CLy$8qFc
zD;N3T>;~hFKN4=0^V>Z$%KsVs6(e^(Gj>k!XPFan`LyTPv-F=gelEn;vj^h)ow$w_
zr{BaYJdX{y#s1rJ@UEJ9k1y_<lOCw?>TugfvG1zGGj66o$Us81{739X!avQIXo*tv
zl%OvTE{S<~m)Xqz73a@glK1o`psyBPYv>AVgz(O=JUv<H>4@<A*2h@j+$(70`eyco
zxoLs+A7UyS_q#tWGTaek@_OmRB(kr|!!Hq@3F{&gLjH!nh8%rGo4z=MWbO&i!S`_u
z3CYQZ=Wx~WoBth$by;}d64rx#a097=uWXYcjceiEat-p=^Xp>ypMjt6^Z65fUx|Sj
zfirl3Y_9D=cpu(?Zx(6a%YD81=zTKgVLy%`CEJG~JNZ`-pRS=O$j0y<{1;?qM}35M
zyrobZgHacqF&$@+pYI0ZEW&RXFQY8hBL^S1!F0^Q01QNb4E~Qnw%srR;ob37ye>xJ
zcaB16j!n3Nd$@t?2y2og;+Y>mVHuti=Yj}pm+-D^IY#0eyn+4Ngx@pHx%V|Rq-zm7
zGol^)!|xO!2R_p8ZhY`Je@sFnyn~nUEkbNZxE|hXCewZaD!D)Wj+21Q*$BV!l&7N>
z!tXWVy;xql$KyJZvga$DMhasgtkIgHKjvaG`k*-8!bR;?AdH{<*h@wPQ^-9?*KN9&
zvmw|rgI(*`H<z6uhX;xA5b@qE_Q&PJ6?yTS{1_uoy2zJi@}{{Q>VA&DjIZ#%V6m~*
z&v;LKiqF*83&#H&wzqB1VJCW{A~xYi*G6f3+C9tN-+;V0Wap$Kta-kr^H1cb|Ca5i
zSnFCEZAQ?0(S5tf>3K}8I$}H@W;4S3>GDVA=y5fHj&0h7nvl~y$B~$x0{9-^B0Fw6
z|1}zES57~=vg;|bSCg}r%$|qj(m}_zOR*1^F&_zB&xYS|8yheI>G7%S$=#P~zvI1X
z?Jl)qr{f*^wN1`!Q71MTBkRR)t+Bd>Z&s->IIvP)tmb3q|J}fj&3wMq{O!J_+vPJE
z?~)hh>@s_d6LiH<JWp3OWI<KOd9_(WZyNXfO~x>KcaT>S%`nSxO6O17jzSIR*0JmA
zMm2=)-Rv8@PF}HPJDU>XBlcBXM%QAoTx9Ga-9o;dXN=62hqH{6ne3dd#!d5lHiiEt
zv3G(oFkbA&sL}X+to4#}@4MDUn|V|D!#!=?yWtmOZjSSF`J3D-<kw#!=jgqM$GDA~
zNV(itz#A*nKDJfn!xijo%I4SD-%($V>eoj8d&CdV^2v{Uvtb{^=wo^F?J;W>c~eK;
z%sFLj%Y~5JedWgW^Ty4E*mX}zIrt$KIZlLfuKl7-$m4SO2x;*?3Zo%5;2mw6;tvdP
zPg(aSBd5S6`FdGSpfloMG56shp1B&k7W&)%Uf1SoU&Os#*_Mfom)Tp2-EG)^NWRt6
zms|(r9qP)v9{Sf#9(2>+opRuw-03UU&u&%=HdqI(m1lgFgYPo)ZO;X25T75OVeKg%
z2PT;B##n2NP#cEo|6tb#h+7|HuBROACO12)BOTb=R*ahRYeUat_00u!<XkQOuBmp^
z_Kv`Q5!;Qp=~^QsMN`)kxK^UN+Hd;`x;b~&aZ2Q=#y3?wd*Ii~blZmWU215rT}tg2
z(Dzbf_cRw5cHHYicYkqb^ZwE9pQ!)x>Nh#HPXC6;sT_FT{;Md1pX6;i=a$aLB6jgd
zLw;GnKQHmq9RA9+g>U80v_0%SC=UGo6aQ~kFLsN^L~(gRd>@N#tIOgpem$;>@l7_}
zV$U7(?_Kps{Vj8!KOV%+OKTp=hez}x)b9}pzb}Pn)cN=a@#x5ZRG#8(abJcv<?2;>
zYss%R7uC}%{7v7#bd{zj5k1xD$wSXLOvblZ;@m#h{<tTPFi_iESjxU_w>{&yzdoBg
z+%PtRJpDLA_7#1416?kuCG@R9Is0jyn~aj$+`(Aw!&>|~_lC8@6?}v6j(il#APov5
zEyDZG&UBTfGpuD^r~d_Hwf~&+;hB0Gx#8LL3%*Uu$NBj>&1vfv{qKkG?5A+98G0hT
zD=*3~;hAN*_6giqh`rB}5uQIz;uJEndl_CPzc%8seLli>17+Fu68#eqeqXqRPuW!p
z;dlM;`+q0?nvJD6j|zM^3F~kh+4(h!i%7uNCvC$!u4MM#LIs5NKu^RGlkl$WH6%eE
z%)ka*z>6W4_zLClCc=B9jsLYdWt-l)F$n9T>i8M+Fb$oM4)<LTzomy9T<4zMbpOMq
z*N~FURrO=4zD?BMe)_(FA42@TM{b<un~g|qe?7)%(@@)-?g`&K9CH8GAcwr0XiEM;
zd_!0GMrAd2A}9S}omYrW!w`Pw%wQ}u#wLU{S_Ju!3E|z>Y3E~bQ`-q>f*WLnHF{Pu
z!?z@_(s_*yCE4;hyZ&I~<PcABeI(Z7#r+q#&|FTGJTIQ|BZE8%zjKt9OJVI&#dwQr
zjO{bt-o=u`d~t~F2l)YU?dL}poOGPhHa_ZMlD2=U4N<ir3pqK+TqGaETB9SKp^pE+
zhG0hyyytiYJH~7C1>LLM`x%){_VN85{n{%(Q5(1L$Nt#0#{hO6VB@;~>cBkqPry`+
zM?Gw(`x#7O$4&N3V_Pn=w~+lc+52~S7mKYs`G1EkzO(<)?%4IiAKHX(06s$0xue>~
z!3OPWZ)e9gvD#$2k&bn?Yvt@J_b%tRrE=|8>%)cg&3A5|e&M|RT#MxIVs%8D7gn0H
z*Raj~wKs?#c?~yP|7@`y-74Raou2Q}6=NJ9bM2%y4Yh6Io^A9lruXs+KBD_2+lMGl
zc6sN!p~!0X*!KL7maZk&CN~>4vnLa~a?Vu~X3M!*#?nmp%}__Co2#aodnOxq=8{?C
z#A&SYJz708H-0zV7#gZ>3|319%5@|h<X!R*x#e23QO3X+x#+$s<J~{OoG{6noxDR+
zj3Kg9()Z{W^C#Uevmq`&<v<LMu&ee0*HMTs^Rjgzzm8sFJS`(vKU(VBa6Sm{cnb4L
zR{p8PR}aN$mbjJON7n)M;-F`AIWXd=efcosggSak4AJs5d3Xtx9d|tGS<m(G8)FH)
zg!%57fF@{%@u-UfsN;G<?f2ky_eaRdE2kr8t(new-jd^mE*Sebi6iKRUCx(qJ)yP}
z==_YFc<g(To!i+wm+j%5Ok8<4SAQDHyKLyLe-q`xPx7F({5z@d!`3;!MnCxE%u@3(
zKmE;r@6R!B@^eG}A2!OoEG|#P>-hoZ_}<n<J;b0pzjaaLJE}o#%$d!tMH`ww>d1$x
z#zrO22H(i*QhfWRXZ_;tE#f}BSJ*fzDt1Wf{JX{Y%K7&Z$8pCm<Y)=+z-?z}GeG+W
zh4dAv(7u3J=eG`v$d`QTsQujq%~79Qe`|ZneczUr+ZBzE`tr1edep&sy(j(pkXQbl
z7-=kxHP>MOIJQn8M}IHcZ*M!-@h&-ccAE1u<%)c3FW<tteVLrfz0CPlY}h1+`Svb<
zU)?K*#o*K-<6BHV5TD2yHjCNM7h~uB^w-F^Va~c`jNhjFPVD^l!Ef^5KL2Cp1LNzV
zx{J3TneTBK1%7uAlI^FHzI^npw!eVBu-0iW7s{Rz)3ffQFUeK4h^{~A3f~q~yvK+1
zoJ05~u@lxi_wQXYkye|mY$~h$f}64L4a0YKO~`3?#rVFgR+D>M9@fTc{YrvW=hRV@
zwm%(zprq>wwLOU9m#jD3yAy-R*@a-+4ciA;it|{4b;##h_)V$Cb@_*yIQgF)1ITDd
zW|Y2X`Sv^?e`E|@Fs71X6F)St-HPJ;u@}8uFQ;7&z6s;3wENodQ5!NcvvmW)JJ9@S
zfe*-^gYf%b7y9zE=|0x5>qmCH$DS~kzreSF90<Q(ufk}A-zTr)93Q^RmmLw_Q-ya{
zN$?`VcNXFMp#B(!X$Zeb|BAxm5Z)1Wz#=?BX7MSG@S9+G|2G0h5!T30Fdj`2zQNjw
z4B|Q$C-96o7eZr1(Eus&yKCWhzr2WtuiW=GUFq47lD!|XKZAa}tFL$TeQodqe=J5v
z>_9f>dLYDas5Z^`r<V5B-S?jR!#XgAtnfSFINU|}9W$)?LQdwVa}v_ge;VO8)dI#q
zC#=Em$YMP7!E{{0eS~j*$KunluTA(3b*+0}!N=r;_f5y>%1HMJwuJ9gKV;u+HkTIH
zb>jS)xQ~?w_vJ(aES4j~<VwhyP`iJYOXm+;vmO+m{nkl)`DnMb+%9YOooXHG+fVLX
zdOU~lJxTbT`|uuhQv0#{$UYzj#%OwLl^*1lqNgo=ndlDRjc&nhe1jCug|&Kac2w5B
zCcQ!DlkL`~Th+)dYCAI9@910}`onJ#W9hg;M+5h)XLl-kh9Z1l@{i-WNQPj;dF^uR
zQ);rmBzwjV=eBw7A~PQHk^k0a>j)IEf62K%be+KjRCfHQYvK2$B^$&JN7nM&YBh0%
z`nS{=T4Me2D;pQeqxo`ku6&qdeEt&q&LU=p+??*6D;{IkEcu{KjRoT8zLD;qyiEVd
ztG|+6WKUgVeX&k%t+#%p^An^*e#bZH&!kOO?TgVrd65{9^()=&5k(xwtI)=^Pw3c*
zev7SVY!5lt+qKlQjaBxHnJnMN%ayVGIof(ZY{!|)oJ(gLZ;ZSiDaOOBZwB#wKl^?7
zs+Yd>RJ(id11j6k(OdqwcBsEtYMW=MxzBy~hO6ylj2LCTMcz_6GLT<lf_gsDSeis8
zo|?i=#ADY?6y>{fScaGQZ5*H0<=cw$_*p+Q^7-fb^cg>_TW*Z-%S-&0NxZ%h&&Fa{
zQr;wzqfhYTzStaXeMmfy7@u-x9m1L=JV%FbZ2xxch@5$RK#tiS#%L^a+!W1m3}L;z
z0F|_FYV4GB?+`L7la<Z5s(nHq)#2o)<qE<xX!v#}16DY`ZEU2{<_qnIyFU(DQ`q+*
zJHxj~1=$|nPlfkwhxO?rEL$6!8=LiW=qj~P4s2VdpG(ZI{1DcG8~7&SY&C}eTJvcP
zpWhy&?|sdG-PM`S)<+$+kI{!VV$s}~Xv~)l*<Y`HBvR8DtEg>B@h_tG<rTM_*2me|
zgD<l2*N4VeR<g1fAMcB?{lR$tBXV4;VE;|qQ;t_+vi*kIrc2Mpv|{@f|6nq{POZOh
zihCM;dWW6wSu1`ZC+QiMTaFh{Z%U|*<>Xpz@ni4DQDd-+obO>RArDLUp{Ji3g4z1n
z6dUDYOXrW;Pon=b<zj=8<^ehW+yvv6ZxT*3&&a7Xi`c-InfW@?R%@x9#{6#YK=v8e
z2j!jE9um`cPqX2yF>=A!Qd`GfF&AGqKi{PHmRR4><}O?Bsb9aTh4*7?dzlAfi^LC&
zQS3(ej%KYqU5UTN`K=x12Rc98Z;Z%?NAhB_{26vGb}drwD!Xpz13j<P^EEv$(vu4P
z?H_b*^=<j#dXznhv~74z|J=9#GT*Xy)J6Hg_B!Wd*VcLUWwv|_??zt0YW?ehCdh@V
zc!Z{|)ztP1Zm?&Q`_GUQe*52v@U7TLglE`roRzMAc!%DN_=%rxV-vPHPUPA;GQJ{j
zxIVsfLarD)UmS`ZX9xJ;nFC^k`S?~0_G7B+6ZvL_c%0KdvzQEY?<!-dGdY>rIuW(m
zyAzAZPt4x<bRED9Hn(P9E&8*tV>>#t=_o$pr{ZXYLdgB!cDe1d{21O(O-J}Xy*a`+
zKp&zA!Z&&M5#9$C!OO^kJP3a;65g4-EiNrF3%d~B7d>!3{LMi4?lr7K-p7|XjquK-
zxOlEWSVx54WX>RqSl2{LR6$Z4cD@HHA^i5X&ApY#48PB%V)MJ~58ohG(cc&KeHcIN
z<&WOMCzy?_&W*=&{Bs&FYWwzoetFaV4-wwe)gtp2@{->cS?T)!*M0No{+tcrTk&V*
z<#gmX4!U7D7Gf)|;1gq`DAFQ)t91h75Pk#u#eGZ2SVLA=Kdz(a2;IMl#X@#9VqbSQ
zXAs*8;@n;Ar^<t>@?o@b6V?Ek@U1+#DQ6;bC++^&=f&(h*}je3&9U=(wN2V$pyLsU
zw?!Ye%5&t<ZXPCWmrFa`x66HdV(WA=GG8Qr8-3U5uFHn=cwYW}V|&fHxvuS}zv*82
zMel90du`;mb?jNA=B&0}Uu!%%H^;T(bcE;fh1%C)?>f1^oSrZT)yH|q;ailVuD!cY
z9N6-?_T9<;m7SlE@s+;p+MvFy7k_$g((?)(E%2jr6<zB>-)j4B;XWJwSt+NMt4quH
zYN>0BwO{Dn$~@z6j@+859!$43o~r&%(TB<Wf!`+@b9gwOA256Z8y%0<=ACK$<Gw$B
z@vddASk0G{3#^S6smJ8sUZRH4_XJf|unSAr(1!j`UEf9jTKW&Te=9j1W*FDDYteX?
z_gKz<FxwVWFwyagn5f+$c6MjqyfM~z!;O)_YTN+*!N`I7;<y^N43c}O>ipimV$n-(
zcC)tWLSJXK@@MBesg<^8P~5pfu3hbFT=p;)w10@a{pjv*?L$s}^8O*a$WVU5=QuXZ
zvpG9XWBDj{n}g~i1^W)7#yItl|Kd)HT?1`k|84&3#pl)Zx0L=3<cs)=%wznvn6EaA
zXSt1hD&{@KKF&@#x63;OIdNg1Iqd-7;jM#Wh7odR?_P7Td|9n+SZhziedKUY3WV>M
zo<Vr0m)Y@~+W&wo<c0AwoV>ea{UoO+A13#xb)+2s8(WOM3MbTM+kYJ2J7UagH`+Zf
zyMF+gwRc!=vbFeTao!;2axb}<f4f+`^{wr%YB?U*AG1(ChkViRhI72L5!*$x*u)=y
zifv&&YZ3g~-<T1zI32yyXv3Zs>O^B}lpoADwaBSv?0sk4mealre|)JQh1I%<m}a9d
zvpMc<bKaZmO=WI+P5vZPPw-|Exsuc~f&F;N`74Ea6t(OhbFDg_e^o3J)A5o%yda<B
zsu%x8BYCj-DLE9+{QR^rl)yNC(Xo5>rnF5XUg@l(K2Rfanupkwi+y>j%6GOOZSI+*
zEuXYk2lVX%QsPXE9CiMSpXhHT7tp6Qnc9u%DptMqeSqu3<dz)!ny(u2-8OzLzDnM%
zXS1Byx}D$TOGEJ;alkt7us(>}X)&C2TFpKu4=$+1myDll^5TYieAE1NTkhVGk9VzU
z@3HYW@yGD{eD}coitw#w__kvI3OTz{T^HxZ;vIfV-nGNrMQ3<t{#gEuI%Tau*K72=
zdDXM%b#uq9SUoS$^9sVevcS2{^xt*;dG@?`#k-P=?mHhl&(A%>XY3B&9GoVzAO@Y}
zGyQoaC%;Al%+ts4uJRb&pWCiRPv;}r{iJ<z_mw82E_qMT`K%g-57>Cv_D|HdACKNG
zSkFgc9TffsGJHGJ)p5&k-Mz1n7uGf5nW8Ws_upfE%;#x#sxR2a7wK&KVT5=LaeT-A
z+}-k;kHWkDbNh`GzRE>Lbuk)FUU;vXfW7^&gN_?)zQE>0Y#W9_2=7+&uqAx+)0|!L
z`1frbvwsv1k=b!&gx|bMT$Yo3x)bHl9fuIsL<doW&+p;|{{PqZG<xDS^m08hGNA=V
zV;92ru3sYwD$r3S=(8WzRtNB>d!LZEQf^JVY3(P*;oGGe=#PE~?~cOX2kmpM9Kzr9
z*3&+Ght!pfwB%hQw}}2M)90l8UW_j)@yRm2X=$6!@qC2$&Z+t44ibr7DZaUg821(-
z<0C%kieO_!vfI${68q+%F`eJjU6&1W*z>cz41WU}e#=T~93)0ug!fCoJNL3}Z!||$
z)J08%^JQEw;QsKp$~DPcLH=y|9<d<@d#16k8+((p|AAPK74xv}4d0r4f+6yuj(mzQ
zH@@G?w%yvvotfLjb(8sUJs+*LE?LcfR9&Mljz3r@SMY}Gov{VEH<)X1aHHC`*&1Z4
znm|@QGJhjGtU>P3xtji;j))bSW4&|9>CdCxPI_z8dyJ0o_onrisrO6dFXG!T<=l39
zb7~X5&;4V2>^NxQ-k!!pm}kO!q(2?U!{D9#!-g*8_Sj;b<G!wJe1(j$tIP!})PUve
zSV}G(scnPK+RiPfE1ougYxBC}@Lgv3hT+;gw#=n}ju_15!&z$Q41Jo$=1IoSc=OO$
zaT-I<Xltbr`ajfsf#-+t&tP-AZ9d0iT`P{rFzcF;a@IZJeMZg6=8dUhG2J!tlhAXQ
z-X-*JM^dz9Ll)<^&^=t+{B(!EZ~24HhPFF!cAPct1odR1x{jpUoOApt9iP+Hiq0~F
zjJ^JB>7zegJ$rQV4#M_l+x*?txNh>;wwP@<=c{&>51p*7I?xeAS39}Z*4)*`Go@{I
z$J<aLDrefO!P=GWD7SuQFPT&E53Z3pqnEtzZQj9K_9x*>`#so_gH5{!ssjT&+u#TL
zFY#TL;l?_<bFhDnKK`YTf9lU<{d%mAkEXDNPs)nVfrVl(rYlzR(ONmSft*e1%ocS@
zP9&2b;rGd~h8c;`_H#Oy*|qB04Ab_L&HR9(TiJ`==-^syOhHq}gS9__@cmS6_n#vx
zyF6UCS03yWqy6f(vG!1|UplP6N65q|`{|uK?D|gazjp6-@|w{1-FkVj+H)`a6EBk6
z^Z8;nKh9)ZV3sk5{Px3|`pi^))b~D!FV1!O<e}WkD!$+I>kIt<zWB^+Vr<s6mab0!
zcj`qsHLJ9BOmX$50Q+*;|5Sc{BA>F@PtV_})xBi=|Dt*3d3p0RAH^{z{1uIqc^r-W
z_<PiMfYHd$wrjB2xo)^%|LB8g<oWy2$ai@2x2R{$Xynfu(Ma#>(a75K(a4u5crEJh
zw4;&Xx1*8juBXs02YIK-Z%NnPMEaY`{GZWWon6fG8Dn3HRYiTSX>M+)&n?YAt*ohR
z59wp7M&`EK^00<_T}_|$w|OnOpzVYf+RM*`-Q-DMeHvm6kLG*6n?FasFA)pA?;r;G
zHmOnKJaN1Fy^9^<*-tFPGhyN5a^R%6oFU`97+j?Ps=T>wEZy`D?UuFUZMk`e&+f|Y
zdwhxTeQ>t>=4#BvnP1IQi`80sm#%QXT&+Uyg{_`b=v;O{jE>5IQ^xB#xkk^5D}Lj;
z#-1DMCtc-n*#1enb2~r9^*)zkpT|?3jeV9}euAxRZh1_tLm4t>>c^X?a+rSonjjY|
zU>4S*JKbyWjB7;?#Xg4?)IO|7UMDL#nVI=uFTP;kb7*XTn`=2;yJ!Cs{^`VqHf%qR
zpOM4yEN!!DH!0Xa{#xT@`VPM0@0axdnDJU;of^N!{JKKCR+?kQWiJ1u+bDL%NqDax
zey3gOo<g>-8iyZ?Re9s-Eq2#nb9VBFvUid`mqmECFGg>6_N8TCw$t*6jd9pj2$N9}
zVeNVtuZV4N{Dg=2<boW)i{cR0Y!}g#uP>wTWzSlC-xZG#)<H#3*7@*vFP|a2*Ba}d
z%VcyyWmG3Oyk~j~;TxnTj?+8eOuLu04}bqzpPY49tvST`HSz9-u*Qvx@NTFT-o`=K
zFUo^Y<-rX1EOmbda`KUxko?*DoS6?Y^UEXtxh3{tZJ7WA#38(o{@wmu*V}1Vm7l`z
zh7ZJRC|@<?tM%lhCa)bEuaW&T`(9&PJ34#PotPcj*i}jXhxdse83V1+9eI%v+3*(L
zM0hVX3~R9uoiNvZQ8GRx>k~4|lOO(mIy_H=@9%$PPeE}U%FZP0?IvG48=phvQI9?R
zu-klrnR4oPxv_M!K5sNG*2z6NHD;CCEU(fo*N3HSz%I;LW_{>f3fD5=8-%}M=p+9&
zy5|S?-XvoSSsS+)8{}8qDL?2siN|z@=iTr<%~i+0xxSeGjvK6-+@F^GP7AG{=bCqB
zvl(sd&vNcr*T&27+3VHqO|k2i!g9K`oK8m0|5?x259^<s`jb?fZ?#>rR^Gd>JsTI2
zQU6!Dx<C%hGndR!59laP#}gErZ;kC*+C}o)_H}J)|Dyg)H=j=B+bPC1!gsTkCaV(@
z=^7_b$H>2t+6-g+VC%mD)=9n1L0!%1QDdhaf8lX^^T^Nq)<u6Xr?dX!v+kag`-$5S
z_Yc<}@@|m3c@kUc9ZCNtyo@&;S8{&0>kmh{kIoeI4eM#2oRWR4fd)AqDsRcEiTU<d
zYtw^{V!f<|w7Ww8yiVfJrbrCmv{Bc85~sF&ic5|UIzHE$UKDSo79w{`Ip0F=G`DW}
z(R;CG#(`}o$CsRsXj4!7&h9-$#&`PN6y-1-qv?Bz-pc$?R^MN@t=Yl6($SjRaSaUR
zw}WWZjSO~9W^*34pBiYbt8cgXc!xe!*SBUP)MdWE#uqL4q|0RMq8Y}Dcy3yx%~CaG
zg*CPuxGWcz%ZVEs)R2w(ivqasT%R@OomFa|wy!QD%XW_A67D<hT1EH1gGcsD+SbM{
zJe2Qi$$NUU8ZRF!A&Y#ximtmn%k9yp{qptzn-0oHtgxTixmm8YGd{vI!3uVyBJaU+
zHZQj3m3w34-iB$$`6RL?$RiBKEBIx+b>%od8mr#mkp92U7lZgGtZV-6Y)#4EtC~Ca
zgE^|2?YDC6E4CF8hXU5CIn|8!#5A3pea$-JIde-wv3$Y%2m4o_l0()S8LU-4xfP8p
zvL0D_*)h(YibmodjYbmfx7{6$d~923Z#44lu4v@h9nr{+ZPCbIj^jE%1>f2~v@;qh
z>--b6KrUR`9*z9up6H%vB;meTIcpC`BdIP#BP*?WlEmTDm(-Hf#$*=nt8$wIimAo=
z@`FAds3Yd}%wcuqU3GQ-JN@{^JovTsfWH1!%KE0fI$PD4sc)R~NBC~KWhd(h`L(*g
z+Ae>RO)*!<nVU=WeWm&<&i&S@%i^+qi#c38FYGpFiQDss<-k$*o#5lse0+|7FB-pB
z*my1WdHe89<M)=h-<J1x#Okj3?Vk5S_~JKf9>k?DkNQ(?0lmM<BYAg8&fS!c<>W-X
zJ>n%_{-$f%DX}^)4=$Ng=(%|{c1|fn-$eVXoD1L8W@pDl_I${$KaZPxjyXO;Kb!w1
ztESwGs}CVB-^YvkG)>>Cs-1(jsps@=Ku08UF8NONN4sSE$ah~lG7=w(mGvH3{qVVS
zVQq9+o1Nqp&_4KS7rhJks4!c?zo|9b_T)+H05+v{-w(z|QhJ&jKb=>rq52=znBg16
zJqyU2!<VRn-NwUQK5DB?TW$9kD_5~soYwMLYv<DOTVAqG8H?e$BL(>b_R__c;`Gh7
zZDre=y@lEK3OjPK?+BYl@NZ2N#5{!W@xJ<xX10A%Q6IwZbT`n7kMAJ-_SG5T8`F3A
zKCbwLZ?(g>N>dznc0T+zc2C<c5!QR*ZxrH?6@I(UM^^><Z{d*r+|GyhOh0Lx%zb}h
z208o5`&^7Oi+A(u^7#gR2)~*24afN1wJTSRYkAPkJ<qy#D|vm%Zm+NF#4W1-1;r^X
zzswe^YvTKn_=jh;^4qOZ?C<5bZ(M&$+YRp7%~#<YkhFX?kgQ;LE;c?*{zm=l&bAb6
zD@K2?XA!%?->dzLs<+fp<VHKxMEG6q0@B;>gfLD<AzVv`sjlrIV-tC!$xT6iGI~nt
z%S8IaH<V#*UWk1^vGea;a%rd9%Kj7bsqYr?+(ga>_pjrpHTt(womX2g$gi*D*Jq2&
zKXUBiJoe(jT>iyEByxVsLV1B4+9g6ix%abs%abu=l{k=jj_l#|bdYaV=#4}F1(Zbv
z{+#9fY;7LWozlJ4$(}PqeVQaTV?2A}rO9%3rvA^j7GT#F^5?7;OZP=KdL}176FH?(
z&wdVls^D5TZAz>#caU3@{fTGE=c&dS*%>Ez=QK`DM2?B_Zi*O9GuEc7rSw0<QTuV7
zPp<7pW98&%&&DI=@hEY$pVoHgaO<R@YRw?B2bd%KYSYu4+r>Q5POWKfZf$4{P@iq}
z*lb(8iTS>n8rzs}ZMU~DZnQ1Z!PxG`4`fUt>yKg9o%F4w`@`|>!5a4V=97ZXpB^Ij
zd&?QRzGzPt8R0qoQ&fzqYn?ntb@q&it@caA@TF`0Tk~y8wU}*{=x)n~D{Oinz1bSC
zk-ThReS{-Fn6r_U-KE)owXS&C)<R+X-B84SqB`oL{f+Mbj+{qLt<is!cX)xG<F;?u
zZo>w;Gq9s0&b3j`e=^2f8Ix>j?D{{hPj%b{#n{?VzdGvQgrD`1e{y#xzlV8O-^(E$
zf0gf}Fa6ck!Sa0sU-Q{(;yP4Jr~SfT@~YH4v0Ny(f5kH6WQBFpD)yqi{n5_-xR}4>
z*`e8(!H@3WEeFz1H+~$yMAmuNcFZw8$ZJC00`i`9f7aFdD(|Yuzc=J%c*e}PRe!c?
zzl;C(SYPcWAAT(PKfis2=ZjR@q|v_68acIGy&<y_`B&+zE?1L?d&<$~1^tR6Cw?00
zS=P2Z)(tV=>2n<Uva!Fpx);Cj(-pDq!H?6$vU5%2th_m@1pgJ_?_6U0k+~y1TVLnD
zSFJ0bwl47Aq@s5z)2&@*SzDAs8Shx;pNK{VABy_-hob&%8SiSaX{+_enrNiX;%KDj
z+-M}}tY~D!)M(_x$<av6_-G{Eq-f;tY0=0M6q^@~Oqy@)W1HfaXk^_C_fC&SinumQ
z`@h{+c?unl?>pby^<>&6aL*d@$8L=J7KPl$(a5S-)cFkJl0(fa;`ykgoUN!PRN|wu
zV)G@x>EjB0%$D18NFF{dXx{$9dgU8)(s$N3wd7MnF>AqRZOoZ+>juB==g*0g<(wE_
zT_B&A$j{~0_VT0aTDiH&nrIuua-~>?@A@(yl{Y7h|I^x^Q!g$$cbQ+Ws;}3~6W5))
zsjedRZSTMD$jiI(2I1eP`&>P$B2UZDG)Bcdf!LQ_YW&j~ep{_4cN*+9UmsF0jvH%d
z)qo4+UGhAB#e9JE^c{3ug6^E>9iOoVJSC>b_~?jdJom0Xzz_S?E3%gFmS6HNe23U+
zi`q!fCjI-A&J4yt`(Mp%i_8(mLZX#wy1d*cuV39@zO&7v-ExHWUt{;};iG13j%)kn
z3T@nT-uct~HHWX320yXo3j60Eyib|Oo^RMNg6?;;f0n-1v*fCrkHLUx`Z|Tr&~q~1
z*fw?kb+NfUSIuT)Wjb%+Ham0TIx@4r1lwA$vn|=LZ??bHvlBl~p)dUV3E?-#@T?to
zpE1CW?+%+k*z^n=FWPpq{TI#I8ve%MHl7vZX|{Q7=bbcvi1#Yw<m<N){;ksRUMl=X
z7uLGpiNn9Pmz;a-+P|mx!@XZ1tT_|fzJMq8UqSer;fl5!oy)3i_;*R-yKf8TkkOf}
z6J%Z&``wqc7x(aecX@>GYhQHjD{Zc5AO20JRLIWGl=_p3{~zn?OgY<7-{bK^96ssC
zKcBDDzt#N9XHyY=ugUFPMc3ZfCaiVByO6A8q~)jW@+m)C!{2x|AU_E`PuLgMU_Y_z
zhFlzeHTJi;@Ndx-yb((r+cpS)PZz#ldKYivfHp~O!}Wgl|8(!S*sNb2$qVboAIOiP
zC#<X1@=ta8=dxwF*fo*Q;rrkwa(U(!xw?rQd76Da*=xjHpXw}&T{rZTW2xs`-_21o
z<k8NVd^*G0X1eh`O->k#IgkY5Z!*GhLv7As{w)5m4c~zjn=Pm3v0Dz-Bzw|IbJJS+
z%4Z2T>mNJj+74vPyX+{&hNANHZTE~J^Yc++GRSjjKepnb{b$CACp#{%?KB&=%H;{m
zjemUze?wGWOt#wQ5}!=ug}*Bf&tXw@D!g}YP43-s;ylutbC|I^P>ty2o}Shtwvz|Q
zli}pi{f2g(#u~r&%a1Ux(?7%YwQT8)1--4mdgw<t^Jy1rmd>tqwHCm;_G`5l<L2zD
zFK22Qt5uB6N^+;X=lBY0&3EGG+^rg9)wW)(r=B)6Zrt;6J7d0+dA7Sg(>Z9c`EIEA
zv#k=Au)Wd{xVDo1TTy*(spd5`u6|I5$a;uTHMOm#pSEe}n8J=n_=C<H)jfYYU%+uO
z+k5!kxzBK=vTOJVDY5K({<2MP`!Tz(>PJr8wqM=3vbOio)&2zAo^-Ax?-HH$>c;l%
zwrx>-j_K&iFHM`sJ4|nEy{zwv*#12IdC8efep9x6NX9$PZ)+fb^)Gh|Hnh?g+c(;X
z8y-YGAGFh+&tp29)5T?`_!Q!wt$de|&+?46rWkL%IXQOC8Y4fG%{TYR&&zTo>r%Na
zpI%wyp1I~LIsDTUGA9_jW5w3}J>)_7&8N8Ix|7sz*Q3+q_6)w8W&F6W8hN+LnzVwC
zSIK?(_~Hinx={{o@%*sWyVh-TYP&fN;hjxb1BLa_b8FbKOg|SIx8xO@s)n-p7M*3p
zKfHTuAotewwC?Ic7dGkd-LB${F+Zz4eDU5-VkM8lx@Jp#`6nK4e@kXj^H=Wnkw{iK
zl8*0_%c}(H!QWB;R&O+N`&!iB97H4gy*GKhEgBiWE*iPBD$1s4Wb#654QrU)bNt3+
zeKOHn=6f`=jyXBpJDtAK$lINwk@#)xx3-RnMkBAck9r3ejr@QeIM*c_$*<kOKGrEc
zy{B<5QA{+_u0=F*wOTaNsA4oyt$Z|6zCtuI=zG6k)w5Rm!Mdh__dm9me#Y{se^Wmi
zsrQQKg?IS$U4F_gR{B&squ9PK$B{UdoPR@3X>%om=btRrm^sAgbMsjV`Si6s!iS~Y
zUyeU($OXC7gl}`I53MHI7uS}*s26HS9W~^rxD=Elxx}`{dO5M({d@WMAYUJ~jy-9v
zKO^?%<oyLT;F212Mc!Q%!)wOFb-x+kFy?N0K0xnV=FdB7MlW?!YzvI?E?{!(Z{GXo
znIGk4208IyleO_qdAi>^i=Hn|Tc@9wLl@moM@qDF{3qQHPN|Q_wK>8!>{_?aT(*a=
z*m_N_<&k^go%-)&4&3Oxe!RAheR4JYJ!km)&hPa(g}!&2Zhp0&=RaHa%gvAVJ+tE<
z+1X%>{um2&jFF*^(@dkoec@Z}R>nd`V`-cAV~v4H^tL}>{2q#ZcW{s2Ztn1$!;Y6%
zs|VW7bnl>v>L;0{_+q>`%(e|R_-Es9ImV;9F%f5qxUg@}O!1~~u^82&^GCMjH3swY
z-HXfhQ@?)Hzdihz$TmiQ`_p;HnA^o4!`M)Mm)f-3nvhMG`719!w`FT@w*GwBnB~hN
zD20kxgG$Hv^@O<*X~q3_+Yz=U`Tt*}5To(751mVoA!3rkJ)^}hHExlS&o=z-_%yaS
z-h;C!=Xi>?Imztg-tcd-MKMNv`=1rFb99M!AKbTJ-1!ySEoX0bwuiNFDlzY*Pd7rp
z^zZ4NuFHX@2>*sjYW;s^L+ta^iWTaE*dAKKW`y<aa_7R|^)A+?rF^;KzVhU5B>!i5
zHIuG!w&CAVdC1;W`kIezgV{7sURJ-v#|Ynlg>RIey~=LeH*xH;v4%hKg5z@-iF}yj
z+$i^lfBU6_KA+aF`~2`Dp2lzd(wV;E^zId-9c+4A|1#=Rg3W3vdtVmkZ`uC)O7V~b
zuNkvZV>i5?`9z+C?^7#Jiydn{Cy+ai?=fer+CSQSg$$$QfbC9{8YAwm=azRLp~*xs
znxb~g*PY~DC;RbI_OGIIE!)^ob0fL<Y`yt&wH#r`>)L(op2_4j>m`=$`1?omP;+xx
zTjQ>?+TF)9#87dfXNLUlFx$LiOrBm4dw(nY3-sfJ^Eqd!_v~27wnHPWw}zP`h8Wia
zjKkjQ4tXQV%}@SRdY)>bF1FOS82Q)6yoma)uXMh0Ep_o*`BK)pr8<At=C69<*FY_6
zV%~6nSM9F3zjZS@Tw9~vlJDhGd3of%Q(y5<2{A!``wvU0ljY=QMRBjvJ`&NsD7n{u
z;1lw?&>g;gslnFw`toT%I{JI&zzBY5&i<0E)D|{8<-QJ8`IEkr72I1!{V2)@5&4r}
zetypGLTV|UBVBv$8{?Sn0&Gic`=qqmjJWK&^@a6{d(+rfEh)}li>>>vxqiU*mNt1%
z2Y<QtM4L&r>0L`@TbAsS<i1`>Zhj}<*>o3k>{oTJxc>f5XC3XAvgI#w-!80X6<{-4
z^Px5Ujp+YOyGr`=#69o3KZcwom93NX^%dI*w%_wxNB$n##QC4p!}j9ci9Ei0p6{~s
z=YzreJwp7)s7n*A!KTKpVG_(&;})48e>K)@=g(vJFLK!R(sF;<Xt_I7F1Y^zIl~61
z-$NXaq<^gEeYusAyw;Pg^QU{xkatNIm<P#ABj<16E{?8bi#+ePM&2OvT6!Fhcm6A+
zUv3@hzEQ^Bv`NM)y*udnm(Feaa714UbXEWK?MfT6v7nhc*x2*3T$!NXw_C@qYwk3X
zYjyd%ns-g*)gu0VP5d)`qDE!327cXKkigt*U6bubG}7PunS$0ejrK<)FYNTL#XFg5
z^P-WQ6QhyR{k&u8VBON(nx$bh^1xc8t93>V>ywP#T(gFGxs~-%qi7^e9cv(KlfK{j
zEy^~v^~#m+yql>Rjnr~2bM0v4RF!CCrfcJzA5$h8S)I>2qR*m{5_zJLyV<?N$>uv9
z+bK9^KZWhTpVFTp8o7}s8i{-mjr>v2yP?L>$XWJ{Tu+ZR*9!YZy$2e;DC(P4vhJ|y
zAMc*x8EXmUNJ`@(wfdOZJeOUK`AmJ6XE6ngi=yU8c{pEgUF6F#e12?{`C**8gcIUi
zT;5ca56`NLdFINYWwGxv+U&3<-DeIzBrlJ7Uv$b`c1F8%Vt(G3yd=LbtEpGbW!J>#
zy77Qx*op8xW`Ty*o8md5sXErd^FuFrI!H|AK=}KFv2tnuI^$}$7#@^=^jx9mWjgZX
z;90s)c-A^#+_B~1UUTUl?<}^<$IY?tGM--{?(*>Q66-R#c2n+!=fGcA8GDXDl>1N5
zQkO@X0|v0YuQkdrwQz#5t<Uj{iP!ZzgEqDJqOhFIC?8k-tX9|_<g@(pHT*umi`^6W
ztmQ=ga^L%N_=JqwWNy>H=heuO&DGVm<_KeJhrXs5Xstu$2i=W_&h%+_mp{I3<Q`+Y
zEFUe5QA_*L@19I_%pY$Y(Q%iL>P%JhW;kZw82U2K*LTcZ$mjZY8(-=BP5mvl#(RYI
z=HiX!=q<+GHfsgGTD61xU3|V<P2|H8`_<(GY7|~O=vf`%?;C3){Cg;W9kw1i;+~^w
zxNSe%lVVdwEEXag${shjI-ke&RN5bOUwqpi5x#}(i#HH{Gn#-)xF4=*|HysG-QR$m
z3MX+&O**Y^$*)0Zfga9<fAcfE+dIYfp8Bv^e{yXVQ+>-L{{L>0_c)-x;qP!O%cUvU
z#}_T-ag$%kT44P@Pj2$rp+)AyCFcC)bn;be_hccf8rk9RI!mJ?w)0UQKAOfKH`qKt
z->b3pF>15xFL~KOe*TH@?N3;fr@0_bIF0WvikI;bzDJsj)cDf*4f?gf{r~7!=-WDS
z-$D3y_S^7FI=Y?;I_V#v&!_ZvhJF^_V6DO4KUc=q)8+aQ`jSH5zMreU%f)rm>6I&=
z%bOBoFv2(=8e7{Z^)tuyaZVl%>uJuFi|==3E8_Q%JA>G#ZC!+KSFexb4|)AE*_-&V
z7k`y8roQ0YmG~M(*)wB>@w=Fh<z=o(Z0>J;(@q{Xa9o3)%G%cy>n3uqjk%ew-v-Nb
z@@tV_YnCy{_PO}L{<r$je2RF=|KX$M-Eh8U+rqxqBHe7G)}<}2)v6m8->{*GdEjI7
zczV~~GEb%OtnrFD>Lv9ep|KQSe%qh!+Kc(jK_#pWDi{Ob>CgB2#lCeF)v|BRv1N=2
zTq&y_d~M$RLJk#{+Xbu_BjS&uWF#&`rt?W%pQc>`?Q4>mikvsejYDSbYVx>>{L}Y+
zb=BJXYF9n>*fz9n>RclJxn4;gl=BQg&OS1Kp=+}K+{xtm2d`(~&vfeZhw|xDv35Oy
zK5wME*O%g5oR8UaD6jK5*q+n8{)us5zu0H=YTxE_eIfg0?PldSr=xE{<JI{J&i~;2
z0n8xxBl6Dj(GOplzu8o(ls=bsk8Q-c@~)?I&sDydlAS*@vn`|N&bP@(qmTHN%?sFY
zRl9g(9cO2qlKMhMGcw!RF1G!UuWR%B$KueRpTgXJxP^RcD-S!UH(kW72fh9DRnBCU
z&xObE;Y8zbs`$>f9-6NPE%08__MYQw@@J0g`$id`L*;vaW4)LB>}j29d!DSX2gxPZ
z-jZAA$vh{|{vB_=By-LzV~U*0^Q=k9Y)E!Nv|1=0i^R#fq;rio?UK(>*XaAheZ$7e
zAF{&Vmp!8AwQioHI+_F9$=}w-SW9i1%IC(`{B`K6Zfwer0=BE=!9o46Sc_cSx%{zH
zPMs|#mSR{ai|4Mltk)CqVI1qFhtWu4Yny*gSZD0<O|bVhQ#X4Dvo;!O>|IT%aegQ2
z8jZ|u=Dm)0D<#XiRxBD>7KujQu--WFiT5gB`W~l}^-H;EWPts9g`$z{)+_UKdq<<q
zMC+64`2KzOq;#Czdg?Xj-ibz@vsU@bx#}sbmy$&z<z909ax{_`Pbc;bvh6kJqA!N;
zmt*|ft}&7Nf5k*zyb$AChnUFSvoVpA$73RQuEj)t`!^=CT>DR~fwGo}M&?;h9f<L+
zslRV`tam2vjz%`<>&EBgT}nCij{d)A{AV>DvzfngnUf2c1LaxP^8Cq{SNXY*9D1|8
zI^I=oihWXXxh~G*CYxi#Z_!d~@D1L3Y&TB!s!<2j)<bI0Q8{<QGtXK1b>2Gol3cv3
z?G<a4tJV+K)MVSXXoDhRmGKMui_70q<~uQsE52#PvQ~Y$CGLsouPNUa$dL~6vhWbk
z2xu`<?y~8S^S36NZ`t<wbbXL-pR4az=gSet^X1n8`Sz}yEG!QX{K~e)a@+p(ImYZ1
zI>~L{$N2ATOv}4+UDUPi)^YkD*SVcB=0CZ&k6lanXeYblS5-4>%6axqX<&>uB=ZM3
z>UffC?Y}jq^eugUdH9jFR1!7w1!Lzq&-L-uhh(v}W6jree&wA>A>&US+9D<eY)9r%
z&ptQTxn7r^lwxwNlUme^Ts|4e2Q~Phn6bA`|8wKO1UA?nb1tv_AM~yCOtnBiujp^H
z<@~ft&0lYx+(g%Ab#bfMZ&Qo#@ebo{r+asMRuR`9Q3ahb1mWA&R!E12*zH`_{py<d
z_d8@R7yn|%j6K^hANO)D^+C1V^>%oIM~=h48yEi0@;mHuZVoP^k>j+koxwYY#UHK7
znsMCRc~adzrEiFj_0ENyOU2g7+tn-m`FEW)<XW=zal;CALVwfYGbC8X=fB22gN1K3
zj`K~UN%}g?yC1CJr)m83Sl&$2E|dF*ui<BTUYU=E(6J72vB2?u{whLu_}kG2eDP!#
zI~;%DIDDV}HoKegMX3{DZ?)5Wb;j5^E5FW(_j%*wg8Nb5e)zWfFMaH<eOmV|b$=L7
z=lL-=p6Kfzd{lz`opc?if3-gUB$l;T>zh7Km-nwOq-&no%{5>BV$C<h_%sI2PShuP
zQD~GgJW%e+qdXlwzqDaj3w><t-O~@&VRhv}CH8%*u9mf~`QG@K+t*r&aW`{%A8X0M
z#xPl>$<EGS3F)1^f?wE?gU^QX|9HpUmdc$4@<zK?$C$U=ACIgr$v#$8|K#_+udP{X
z7>iBG>0mx5KivrHT=K*Bw_PW<k3?{M0=?tplQBPiC?B!m)BfsxZ~0}MHf(1OsU-%5
z<@S60l9K%AjK9BqJ8;%F1$%vepw8D>6OBBz+Pinphxd1>@#cZk+MG{i&6R^4Mfj~0
zzhg*Ab*7lPDnCDcs{Up5tnj`Z$Y!3-tzCAp{Fo0ukaw=7`jBqzrlUt5eJ!XD?0KDx
zn_qhGMqV!cP4ks{r2h}eEvQWqekh7B3V6;Ww-h^`BjZ6%ezRZ2b`)JtGFkVe<p+KH
z{Z;R9l9~5XsDo_3pMgKvRW+g=nPH7Jjl7w;_%w^Un_gXcTV1t1qHSX5#(iWAkQ*1r
z?AK@ie*Gvw&Tx7wxVHe^CGflb*VLs2Y@Sd4PwYL8FYW&s&V42~J|TlY3Xwnl4fnIN
zQ5^OCNi@<Kr{as%i|YJ4=5c<EA+sBMtGPb`x+4`9v%5e0PxISA{>t|qztphqYv{PS
zn%&0S(%ya-IV+Fj4zOMrA{Hae^<%|rqMAJ2+%VgEWv*D+rZm16PBNd3@{BSlHqWMX
z)xMK8OGoP&+uhyNBJ!H{F<$!0ALFDDxqF5i5ANGS?vtsW;mO`iUc*_&xG|QC{J$NC
z@48-4S1PdKMYhay|7YY)!sefiji{Q|Rvl|iRx@)+BX-y4hdOG#{>7^-HWiJ}Z+yQ{
zRxOf)Z`)ROoXVJq@WnfP(kZWJoOeCXihJHf`uG&N)-^Y8d1rIZ-!=Ii>Rs<^_R8_%
ztD=!O%e<4ZwuwxMMw<4EMzXZ@&ZT-ZQoB@?_pN*0vyOSgZ%Ik58>}a~C-Iw;bAcyl
zYu$3y^}g05`4dDV`Qk<+bDzXSQvV$jx$r2)zts}s-+PVmZ?nWi7G1Ty6cZVI#kFwm
zVT^x+KE}Th8xzUp`iz?~krX##BJH$`e=;WW<Bpif_N6hAy(?lOj~B#5#;mr!SsxRb
zyf?<b<q{J~^=C|EM<VZW-trBzwbu(@$s_BWY^|b^)C2u3*RRpY8S9+V`kq#f-FaTT
zU$aJ+W1aXYiJT63`A0!(j3VN04Bjp;pMNke#UioTFY0cdl|v84sB6=#d*+Mha{0I}
zw#F{stj?&lH?|nZyUgbY)WQ?$-&xNO7mW2w-bGwi8?Jb^w;km85Uz<|4&$hqTJS9W
zQ{;B@!p83x;_BMl#pS;-^p9(i0_JGfQ`!&D2)o&Ov4nL~X|{;{a(S_k&6C-jsjK=S
zM-L4(2L8+WH>R>xUe252U5tFZW53ThxgmESYMZpPd8DyE)|Pj2YROk}CYRh*E9bsp
zUQex7Cy9OjTb0OjW(xHsEuEQ+g$&xeUNN~oyEf-(b;WOqsqaT4%`fx+1@Gg|$i3sf
ztvVKse1;d*)$!+jcXc=F|MwY<{4Ph6x$o&$`AJ=Pl*W9W(Omx#KeHpHOgvbKJ$%wj
zf14T$D|^XKWA*j^eAv&}#T8>}(=c=8DD&=kbJ!GejJwVHz3NwEX@y$0+Vk25+l^||
zChLXGYOdHN+p5hr{@Fnm-rZ^3W0L(&+toMM9=Y!(;=4CHD!BKl-P-K6-rCPEVm}@8
zaAO~PoPU0g=T_Uyj$3H+5G%0H@e$|iVHff{e#7=dl*L$WN+06;BXaVXT#-{1Pl|zU
z>%-;)ZIZI>AANXBE^gJYuk>%gQtKssUA2H7{jN4s>?VujXmj^Kwu*P1p8V3o8mpK0
zDg(^%t}W-U^!zn~zxuf6FZa$`MD{ZIOLl)ca;&$`wk?UWYvjOM`Rd#ZzSzW$2K*Sl
z?Y_@%E%f`oKJ7io7y3~0xO&O{@c%){dB!^Rto6+~a&0d<KT5xT3jI9Ff7&NPQT+^m
z^L>^-8<EwQueXau8-Cx&hWabj_a%JD){;~8Uq2J8!_QA*KUO2h1T|X!zR}N0-PqTj
ze_9xC_01tQ_5XWySk6E91)GesR&wnnId|kkvfncXGkd>g|3VJF({4c{>ww<o&taY|
z#_<ndJ<r!K@_P@yO2VGf{D0B8!@pQ-@$0po#$yw*O6m*QOH*2ppk7w*OzCY?)wpRY
z582jGT;}&RM&<KavHBotjo;2Xnyp()SQC&@zqHs?<FmTf&JE2!u0502@!RtHIcq-S
z@56h>+)3XGSjW7%+P4Ot`C810Mv~7{|JB9%YrHF08xtF!b)S)!Y0S%?@ev<{S{Rof
z&g%CI>EuTW??>Xw4RW8hE&Mtgoa_6N+L=WE@m^B-gIuYM755y-U~TZR=M43&E!zf?
zcgBABx70FblaHV9PfpKb{1G3W9Pg!XmT_M<y?CYdju%(hHIU3Ge<h@+MPkpN>dcTA
z`0z!scvVexZ5*BP>5EV1wcOU`pUR^wY;eAEGV22EcCs($8|qwY>jtu?(KD8vRo?MF
zNq>qb$7{x-{vBY)qj%Z+p7|FU<-z())=Ffyb-crVL-(fBrn0vGA4~TgzjOWl51eF$
zQV3-vGez0EXQar;in2pUWybsM{l2d2ZDj8xWy{EjA~GW@TgWJ6B(zBMA!PjS*Vpfl
zb9=q6*Lc33<DAENob!CXUf1i&YMHNJ2<mc4-Qr{L7ktb3oB0kncu1`M159k0#`r%Z
z4p4_f>h_p>muq)c{c@nR^7KuL{q4qL{qx3^eth*BF4DpAx8*Z?!oCD`p%?L7`pr>v
zbDX}Mz@9#aTT|s>r{meP@Z{O-g8oORDc8B_bn0B9@#;>Gx~WHN<45B2VbAo^-ah)l
z`PSs?7~C1?xym1`{0Be6y_9`IdB>HT?o;KHUgXdm<L`No%rM@QjlqYm8AFGL>c4*U
zTzzkN-rhHr*-4r0arfr<;VZ_Zp)n$nm+>C5yAEutt!~wEuc~}q1-7AoraVpm^!exd
zt)lU-cgH-<|IFK*w|)h$>2oM0)THaDHpru_Fjut5+|t67F!D3UJ~1yf5ijVU5=!GX
zcWPNrTw4AkAKsENCG1aSoa9oT$3v##B?D|TWw2%jkI8&F7G6IW^S#+vDEULo_hw_^
zjO}#4_u4+YH|9T6W1-N_Sa@kmES&lVFWDFiS#Xzs_Tx5(VqxLInAkNII@?~{5er{z
zj#=**3r&~B!jjps5Xc!)c780>|H^&JIDzA=k;gfw4*%j^HSw%=<>*k|l<<N!+}|f9
z^cbBIK3ZT7PJ4gaWghHoN=Qt_C$f>!^f2?|^nrezpjU6}&x&;El_vDon1{~&!Z*Fy
zjq#abjDzug$QbThL=V2w_v_&HH}q`-eX>2}cazOxj%|3#cWn6wwth@L<s>Zq8IL;U
zy!}ikJj?!nHRy9m<#=`<I-Q=4SA7;gQIAFePTJpZd|zR!`<#2swkwHa&<*F}od38o
z8rP&JFR0^-Vxd}m3jG+%7TUf`j|bBU7+HjF$LDHm((lgN@&LViwlDjq=NVj^t`S`E
zjF0Ki@A{{|e){i6@5&v(Pe`io;UiwM_qe<mT^fh442GSzU9;y0-!rmJwry_v-eH`X
zuJzeZpUC^<v4e7aq_ML8@%#3^l(1d?u(e}<DsL_OZ$npa+{AU#$Q5MbE3)d>9Q@m(
za44Vn#8}mVL*1X`+u+y^eNnlEe(i`q^uvLM>4TBD0$coLBK(@Fji0bz_Ltw7HCTo}
zti;LI==X2f+D7}Ea0W8mxV^B2ulbg5H+~1Vv(X*yx4q(6SLgm*Z{Ff-erC0D$ROqP
zTML(zQ^hqq-8Y-`Cl%bccPFl~ON?cEgT%j&n`<9DAQec4J^ZHYF1mh^^DSL_j`Sx#
z?1sa8J%7Kpe9tc)G|q?Fx^0F-%5yyat)Mu!KIQov)vdR>+|<5mbhFu){Kn^EwS~^j
z(TDn?ukrtVr1L|x<30RgAf9O3Xt;5wBiEc?GZ|l+&L_`>SMI&KSRIz)bo%a|6=Evo
zpIW6qwLcTtMmo{^dK;9hAM5MKR{ElfcFotO_;-O1Xvag^(CYxdrS7S;V>a7pd_<od
zyE~sZ;3z!z`>`KwozFnVxb|J`z2@G`+Iv&~ZP7Q&^jB5SIpMjf)#ajgp47&@li1WS
z9CI+9WP6`B{Y2O6+s~h39D2&Lyp4Omsk}GXVq5&A8Nb?y9kHFB)s1B(V^EgADT%if
z6tCvQ6Ek?`e{}zr`SC1w_3y75#-uTS(hg7ThJz&Q1ASRo|9<D0llASIUUafG-{jgW
zkK(ugD1*Mgu-5$i*VY-%V>eU90BgK=!Din5(|r6z^X;y?#1@ubN(s{)D{|U<7qDw7
zyv{Wp7nuZ;-Pdgy+rdvhV>=CDec|+MA+@?+@%;_={pY?;?6&wu<=`}*eQFIQoL@E5
zT7miO^~;n{!@0lN^*Hx6dW61~=F9ZKFp}#T+@i3)dWbzJ?+v~q(-v{0`VO|gN!gv)
z@)mad_V4VLpBe7DVt#kuET4`)?NYxq?7RHslrSJIUUWabd6@syw?m5QBY2ak9DLO0
z|0}`996OkY&So<{55Q>s{rrPCaAv+UBOhxlQtoB<WWDk>_<i|4n3omj$;}?~(!Zi~
z_bE18#Qn<JoSn|OKPjDdxo?&G_9-(-yVj8Ao-@+9=K5fiF>a^t<9_p;e(Zi7{^{2o
znbgsJ-L9tiUwdN#r$*C>4@f4@AF7-Uq@Z$FdscU4Z8d(EPl@OBQ6v3O{-}I{F=^oV
zC+_R2j7%h+WBbhSJ#crX_Uu(g5BgHMkXXwYw5|pZYO$S0+R>7py#f2W8m9zpg?qmZ
z71z?c5yN1`0JsrhC%yQXcj$i){-X~Y>}mg9^|H;z25KnhnZ|xAa~Vl*|A{8J#VhQR
zP5j#mZ)^iUI>AlPNm1?-p7GuUm<1!^Yu1*}V4t7pf3lv$e{Y$pk3QyyK7gg8_?toO
zN`3aL$7yB#$sRs#3VZ4sQ*An3UB8CAdAG&3CABqe6&OVSekgD5q7?3k)9f>@T_0pu
zw^O{2$*1erRrIgvZu>jT|Io`5>#UPuBjcv>Rb#DL8E!2w|NrGU>wVv6H-q`54k_Uv
zZd1C5-?w7Ew;BuA<Xhf75%b-wSoml+E<zsrE*2`{C?9?n3-3*gh1{cJA>YVYXfZnG
z9kW>Y4KG==GUmM+*Q|?$QLFI~`IqcC%0ts)VdEq@nK`jA_w!iD>e>ZMV_})|ElI<<
zvCw96%zC+4C_N638e=YMcq}xrzi@mkZ1&8S?g`2^Ar+e9IA^ZoTo0#&0#Bucsn1&f
z(@;)HTYBI)t4G0tIo?BGX}vPLTnrDsxM6)Wduc`YQ}W9{JPG^sYqr{OsG;0rb9mlf
zEJTmqgI%Kr!;be|I}PWVPoKWT(O1*qb+Gztd9$_jku10W>o>UXCb++ceSOa#!1wq%
z&!Jn55nNvEo(_lcmkO|)4(}-fXG<uLomZ<YzOBYa)MbD%{<$LiC(n}i$-ktvYc^HI
zrJsf2&%u2%k)DizqmMn0&%EHdFX97rlt*6iJA<;uJ_YOZDW47XVW*9MvHl4UDM>%}
z(vKrs;VnCy?$|$gNH#bX-KoFGFWc|V_Z@NkF+3*A;gqlwR!^j3qu6gHeC1WwpK<>v
zecY4YEp^`zG7mOycI=RIpOPW=S2%W?6!iN)_it8aiC<H~hZpIix;3E3?eEn;#^^Wr
z^I<k^A|)P?XMv5wwWW&w>Y%+}y~>w%c5RZecu$Uo9d;NCBiLg7$;O!NEm**IzTkU|
zTko&L@GFe@O8(3kHZW%Y_kQ{Y<rudwx7^LkykWfl+6-4VvN2;DU&l1ay|HEbR5{z1
z%A49wB(HvE%<09i^Thv)`8nn1b^UVZvUv7*b$AOVAj{#YIdICg@%7QEmpkscX+87C
zHZcnMnEdD2<;bUwS5Uvku8Y546MuJe@DjdS9g2CzlXPhvS)_iCXiIQxf@e<A)*b3}
z>{B>5^KPE53jNJ6Our4prBe9ysJ7X58-VB1@8P4hL7yGhXYZ(QY4<*%j6t6HptfF9
zZWZ<Th72G>$aT*f<k=P1d%kwWb1F~hv!dVPMB4H2E^XNZKlhm%*v|(YP$$yykUsrE
zJAO1aWD@zz?~;evnRCkzvOU-J{a#<WFV6MuSI(dvFi+Y0J!`+`rZ*mo7wYpVVwRC&
zSZ&D}VT*~L-CG>^uJP^3e!IJ_3mfZT?AmI73;egKxUvC0SxYV<z;-sWvW&bzNj|ZV
z@t|vQ|2g|#YYqA682o(uBRU?T|Lx#QHSdBJr2DWZ7ktbhUb_O9s>7#gpT@ka5DOD$
z!A1Lb;M`U7>ofB5b5G#*aB4sq_LDXx^fmv!4%QyI77NW%<39J{`VV=RA{YMfINNaD
zhn4Y!=W&&~-baOP-!*Y>Yj)Z`CG6;lTfp<{?5U>sB8R%3J7&&_eO}xQ3s<n~&r-q^
zyymx$Q$o5A#UQpr{T{W_I)%9X@v--2<;RQASM6U{0v9QG7az}Gk^7&l+|I78BJW?A
zU(dmxJjTY`S~J-rC4>}qJ&3<1kMt8W_e}}?2Pi*=2l%~hv@&q2%G<Q%H#ws$ate9i
zXHovrvy#dAC+K+rc=jOtO;1l!!(KRW)Hb8-yEwppe4{$--bY$EcG0=)Y3QbL`Y1D>
z{D53UHuq-|!`%ZfFWu!4vs~W?XBqIfn9cM5f_<aU!X&<6pZ$gU>k#hI-LXC-#~FDp
z{W$TYK0Rd}hy9huVc!Y;X#1PvMc`Go3;Mt_vg^O+?p0qR*b0w$Ud7+FA0`f7!`B*v
z@760H-Zk*74xUx?yu2FO{ioQEJm~s2;NPli_^xu|Ij7nAVP0|eQkD%qryuI#67=>2
z9e(5;{YA(7_r>?={4w}f%I~Ic^EsW_aXWVL8a;M>J<_fb?#Q<#lXPSe-_V%u*L@K$
zv)%g=3{ZaC`s}g^-r}C$+~3Btx_VY`&lwFnejdRt$GGMrSjO&(Ph)eit+Q<**b&d&
zq#X-?hKf7-$?YVFVcb7PebT>*V|qrV=h<Wxc}nft_zYWpn$BtGnWwbP?@NBqwco6$
z7(*XieMmdg(wBeuBjeS8-`TtgkNMJ=%rQ3f>r-~{eq$WI8vgQvace_Yy6f9sc((1X
zwrr>tzg5E6u(t-sVxcN7(}6tvb<BGHSf~}xpM1ccTf{=!x-sjwV&Ub6=95V64zX~f
zQ!LEt!H$Q-!sEC|J3QvE>3E9mHvFV#3>Qj}$9V_WX%`FMx$o`z_)KFQrKvfo=CN?8
z4KCC%7G8Zj7WVYUXZkABJu?&GM?Y(FQsjV!#X|27@vfP%@Z5@6=!I|nbR-tSCH2ZG
zuT&E6Zp1IY$=3Swr4#7rLf_xfCq?O3M|$+^9WhTPW0nJ!=jD&sSOXXr&rvUY8UJi4
zAJEx&_cCt$K(3MUmLIdfIsD)katSMq7oSp>oFwm%!E111{%8WJ>imbZ?{e7<{7NI^
z{RFA3oEh$4Z|tt*g!!KRN+EVy3ePEL4M#;fO{$icPbBNfKV(4#dgS=yl{~j9Usui8
zKMViq{Wy}PI$a|fNzCtmNeMc1ioU$5{94+%KOJAke!gT&9e=_v_QEH9axeQU2J3#<
z09&`4d&El`!=h~8(xokUq3zBc`fm>&vd?wof21Ahr;N`W&rZ_eHGjMJ9oNouPk+)!
zIm;bu=Ug+#r;%BXt@fN#Bokcj`#l_k9dG_Bp8X3~f<=2_Sv&_>hF;%-U){36u<ZH{
zF77If=V@1bzC9ybJljEzpbtzlPUH1)n^E}Q`|Rf<_QUob;;T=5#wPXq^3P!@d6~a%
z|CRAGPPL8ODPz{-8=PjNKH6d~XES^Ad%yGFuD~^x8$<WRxkFX=w_L0~B+hBt(uvn+
z%9G46R@2mJin`F1=@anU58({?!~Sd%pC`%q0ee@swDa%gMn0lDHP_-S>*y7U^M}V5
zs6YLAi!7eYue)}<HpDzT{;u=*W$L3|v)mhhkEgbJ_IF=v=Zd)|(f#qY8&%Y&x92{s
zPac9#&7yRzFRqY4Cz7<!?;Zo_JsnM>Ze7&-hI_ir;wL<J)<Sxvy!ba_9#i%?+f{zY
z`*Nf@6xMfp^<5f$m3f0$i^SIg9n^=v?7+?UkiGD1A5OSmjQTx3bpY-i;+GEV>mTSd
zX=y)=W7!X~0q5(uuBPh`yEnehXu>Z2zTNYb+tIV)?|$80B0ki{iXUjFasNVFI=}1w
z9_*@{>$>8q?`Yp!Z2Aq)Y%8bI0<UZ;-`2pjb;Jnl{rO7lwv2pKNt~vb+)ZAXpOw9(
z(GPr2Gk)pq#nvAVmpkc>FTCs<rcbAY2Qtcq<1IgKj)gu8<)1&6OCD=~r1^9)Sw`Dz
zeqV)ubH#0Mej5v|&Alg{kA+o##=_A*V&Sm4`cr3NrQ;hN8~%^{G|p2Z6MvYC{S}rM
zd4_#e@;*onIno#9MQdBzSf5{paVKGV9(<=jL$OISyra3}Bdz(EiQ_C|YjCgRjFb?@
z@u9=m@j&y!L-Dk+u5r9Kn;*q*+<TwgWHzxIdwceX{1V$NIv-XfrGy24$HKhsa`j#0
z<XekvZ5zYi)MCssr}zZ*p2cS*<oCQ+tXX(n-QTh<vYT88|8k}uJ}_Bdtr9c+m=eCi
zRqoAft;Hks1*iI4pPy&bxAL%$^z6p@S}<S)EGVaL6ECKOWq;!cw!O%=@$Dt^63(r_
zB~st!)9%0u+gv!qx2ID=emGxEJ03qQzje~OP8?;7{aX9gX$AZroD!-G;s=fCX7#!Z
zhb|pR35#4aO^lb${&%je0-J*UCVTZ4dD!oTuF0VM+RBWa#p%?o&Q<TW;KK3m#bv!<
zdzAlOXHDv-&do{*FHW~zni3v@eerpepOmxf2N+8ZDeIuRURTdge~-_VhzoCN(?j?O
zUHm}*+*2Ob!1nmM#-Yvdn>OsO18nb1*SpYtI-cMD1bTkkxjXcH6g^MhL=4uDFRI7K
zy@a#Y;^%OhR5jrjAJO_*wx`@J?4XQeU7WALF8*+RR`)GaM$6aPqO$L>r^bo!b+8y<
zG>n@F118DM!n+2O#Uy?og(2%kvd2MUhzLAoi_1ESKU#V|o6GE(zm<c1rCtA&@*dZ|
z0`llN;L`*64;?&D5Aqg-$=ZJUVK_`z78=_`{aQu8F5pjQ>dWR+_^QVA{Hc_%CJjGC
zPdDA`yTO_D-_!igOLVX;Khs(cjlC2sYrO8rxBL<d-L_dDLw~#CHH~IlFC#Zo6}LG0
zEMAc(7TRZxg|(01FNMuP6^(@ls^TX#V_{%Rc1r)d^o)g8WWW7<uge8F-sM%-HL!N(
z`B*4YF&2&&_pD;LPYHNXifxyRg&!)#!uDWYaXon&*DvVk`J}}gq`jEP{;9V}ub6K$
zxDTc*T8vY{m~^maqjvt3-#X~Z+N<pa6X^U1b}&N@XN`D=UQIdUeJuLeIJL3Vckk%i
z?D{efJqqRZb8UFn46iY+XXws^KK!8Zsy!Yj;V)OrgPAZs{#}7qD{v4JBk}hoo+0BG
z<EX}c)fhZuxO;}+ymaE4GVTneo1<|3jPNrZ-<6(SKPt9<f=`C8U5xz(Si8*l*R{Rh
z@1-QEJiR7&D$wI9@;Oz-D%Cujeuf|?W1GizjBPH`+xh11YyBks%kLb#eo}kCzHHoa
z;QzBNHr;u>m~nxaY?S`$2{$_7hiza&SJ=?c*74zUQ^L#o?$T;}Wi5PKml7)2|6_^o
zIew{6m*d2r<0^|_0DCU^5sdjj%$p=<*id{~L>}>e7?IjsR0cW3N8m|$JmwX7BhO3&
zZ?BA%Z=VL2lyhe*9cJ$zp2N3bM|?fc{CnhO^wkRwY9n0iqQ7=l*8lbS!q)ntoA@P}
zjSjrq#~s*T+Yij^PBD(N-M0XLB!?D?M?T}rNVYGG!7^j?mFrfsOEUFK<K)~wi^MZa
z;DhrA9D9}&@O!DT{m=O>%K2$7exU69Gwn~qZ>RE~o)_oAr#-s?Tc0ovCmRcM=)yvE
zYHEAj@5jblAM+vna_xFDnZ68i?-}J3QQogJ#5JU<{gR&X$EWm5{U7lB_tiU|KiWY5
z#;o+4#NUU0MO_O!*DhW+_l<!$>D6HnolD=3PA1_4edt^-b?6~RB5iv23n8MtbTfmx
zEmEgv)MK<~H=m=O^PN*({Qdt+q$R28_fl=YF<tJ;GjgbNA9ZZwTy=HIyaukVgP|MQ
zz$SUmE%0zFo!pKC?!+y3i$}kcTiT1a?}KyujqUe1*a7V$W$jmTtdetQU3b9s#obqS
zn?6w95oP~p%s%nFi=N$kq~{HQwSA3Kce&Gd^!r=d@FrV;t?OSGYqn9(7V>?sXm@?N
z`P%xly8dTR@wxpXPvWqJ{m!pnAA-$kwCNH)aQH3<_+_4Wh)+Dy8J5=e4n!e5{x%-+
zgZ%PpcnBkpi~TZ>iiJr-W1$*s8fTk0A{M@UUpzNC=KFr;=C{Yfu<yl;?tA8NEc{O#
zne${U+$L|GjfG$T5;tCjWB15;WQEg@v9S_jH8QJ|xs7u20oBcC)Z$m+_X}jO`UJmc
z*%qmze$V4TPl*}wnOlBTj)K%FY3)ZP{+-R8Ag$e(5M0Y1vrn~Npq2MHs);$_`<*(m
z@P2daC0~w(y5-d~m;CsnvG9F?SU3iw57rWsc8!Hp+L~IMa=_jk$HlnnJgFeRQ<d$$
zCSJjFX1^mB)z_RE&UFu*S-Aaf?qoYG>Y<-|J;pZ|<nL_P=CY03xtvj<-}w4t@)<kS
z`zRiF5)UAsp0f7fG++0t{QP-1{trBd<!`#@JLA>SxqD#8y#3C*cL6E5M{MS~<u}Ly
zO-c!e--B}r`mm4wNR|Vbh=-`#;4SWVT|Dpn3TZ(mlJ`lTZPqQpr3#zj78$<VGjNmK
zCp`Zb_o+{VKkz}@wT{JO#952Pf|JE+qvaAtz(^SPzGoeDti(!VrMwx+ZKkaFw{H*c
zg_GL4S$R)hl-vGK+?f@o<rRz2(dt#``wR5+WpT%Aa=di<Y6p30I{pC}X@3@7?@rI-
z>qraI_4$p&B=o)POMcgMEx!<}#NSoME9rhZK4o2LvC%UysTA8FOUmN)u37sc>}ZJR
zyefC%S#PqZBJc8VDe|nt+5h`|%155{F|3LU>?k}AwmNs4J(lhZKiJ}TZP{oOxU248
zD?1%qt5ei-^NIDcvgvDdkQ8KdXEU<(``Ihq*qsd?XA$=}Udj0Wsb3!7N3Yk4EoO+3
z+sfzM6c4bYVM}9S$1H1mKZ%+Dj)j%~S}*eu9nXVL7NF-j*>Wo5eU5!uC-b#yANWY#
zRQ&PrFxS%Ob+1~_TL_o95e*mqj)p0hqoKx~XgGR*EId)fywx+Y@Ir-HNKgOI$jLna
z0{dVuE!o89#be%&i-o82;w(90VQDrtPf9!(3x8&`Zbuov*}s@Q7V;Lxt8kul)#Yhk
z6zj0Js(8_~SImbslN(}-bMU5A$+56PjFe8BF0WFD1Gp1x7yuX6J*iDK#au1$;J5MT
z6#6&C{?~l&Px|st``67Q-)l_u;jD-0MIm#9#yT%OIB#6vH?Eb9O>w%jw>{s}*|C<m
zW;=YtSU2rPr+UM9{-N1G`Zk20GPe0g<40ry+)X<Sj~;B^<2|__ve`AA-{u2uipOw`
zSJ~!sY_u)A$Y}iHInzP6#2$~kmQH6YBNkSt<Ii|TNn=g_1~?Yi!I$Xf9`YoaOOBH~
zq^s*+DrOGzQC#c+KIV>GH@p9l%^&%e-O!JuujFoUiPt`nBOQl(MAYL=HrdGkYE+hk
z*MI5h#mT2&M`?KDeEj=G_m9UvMzZawdM3(`b@DD<yOdBDo_+rUY=EhWaDVkCvFVCf
zsI<s<&y&BO3-6a0=Z$i1;=SwX%uPRSE~^d<>YzR_@Ehf}pNVs?=SvRY>hNsQd3=g)
zr`9JA!^FC2@q|qLd2aq!e>}hjC$Z%>__O$T#2#dOx%GEL{XS$2elQV+e!$n;wwS{A
zeZsEhh%@Hh_1l|1bN+KSw-EoCe>d;3Z-F&GelO!cQ+<IiIlr1*T<qDdkAMF&hx_Nw
z#zT}hfBN0I=lC~!{>RSycvi>p{I}<)9fjYJ`1d}tlMCcEx^dpQ58*?L3C0vI6mw5S
z_ohr%HlFm4vR2#1zelmtGm_`R3(xJmh+kHh_%|+YFVoHy>QCNY&R;kd)bST}8$JUL
zDzB&Kv>zmXiolva@TEJfq<`DG$p7fGwO!?Z9IxBox#9Xpz1zF5vu7{x-1s|qwLCY@
z=hl&L?8pC`w!L#l)pad=dqiEA`<+SOT>jGB*D74+Yj*GrUbRvGZ4zUCD}T9-&1}av
zcf!Wq@bWwFLF{$!K5gGG1|W?|ir>Q=|H}Ed_h`#5{pG$j%BZKjwrkm%=gcsU9gXdB
zb~P&je)nLb+EnUIIhjuSw}U#u`{}LV`fISai8#0(8)4sNtEyWG+=pMibw54f8^`|v
zJAdc1*vbQ^?yi^lf&WQiD<!6}js7sTJ%3akm(T5e&A;Shw#C9NKEA<3zJ6#dOz9U3
zLu0Tr1s2-BJs7@?iG>$&pEG#GIb5j7xA=?uF0PM-{%m9pY|d}H`WsliE9RR6Vzvu-
zMk>4pmd(v8z9_=Svc<`z#35C19N0F9M4va`RfGMh(@@)<)%eP1#0&0O^pI;Z^Ql?b
z{(a^@*xa~l^66({<_)c3gqdm8@74C~v!eRt#=p|V!h^S>;j2HRVZ?7y|FIJdx%@7l
z8n)gCe+#H5JYNCd+sLnX8wck<latq;6m?!8j>+=`jH!z2ybLE=!;3E7*@YK<K7uo_
zXZMfV@GpLrNz9f5U(N%s9)-VI_`PfLGPq6oBmCKBF_qY<-AsOemNt?PKf!5M-sL<)
ze>O+w+Se|?h`;2e?63JrtPbZgY&FNU0miI`Cu_`&t&jtnWNv7vyik((_bqFs-!>Q2
zU0x>vkHA5~44lTbXTOGl<RIBWK3FF=w#u4p@|b5|uuW3-r@Q5-J?rWZa?nS`jX%jZ
z*_POj*SN31Y#c_pEeDGCVz4bLt{mi^VR)%}tQ;UuG$bYTQ?C!^spm5F-X!<6OB)XJ
z34dx^YMfmk%_<_MDTC8LXRSeP_TLbu@lQ*d@e#Ji$(PMxk>4#Eh}Y`i?bYEyHJC%!
zFW@u5{cE2xH}M3&Q5;U%-~6QO*ufWm&-RQ9#n_r->7Ir=d`x^@S{Ki&)|j7Z1&_Lk
z38Q!Y*k5DV*@pxMG?~aJvc)lu{X5j$0X(VSiw*Oi#n{|4&%hL8{CQS3oCZ$7^<(FZ
z*J*x_&3*W{e!Ic%Xz$T<=11?7qs=TYZa>dW@v!Sx|0uVn?++NKSaY%YgZ$c0xW#hZ
zn~jx8k}qjz4hT2ML8rSs6$=;Am~UDh4QFOWy%P})js1RFuBG2=VvEk^eO!CB0~@G{
zw-=0slc{3i5qiJqSTxl7i4B~JhEwDV`hPJKu8*_4LcU}ZJ?>$zY2|6z#nkj{N7)zt
z7Y#*RzyJ4WxOP73JM+=-?b&EJ|3@_R`zsn!U5ke0IMZ@m=WIS4=4p8ywwDylS5;Ss
znx4V7^2!NS?jfG&FYdv^ihpkH&sKTfpKbqRZ*04D0Xkk@uCS5#r5oEC3l~0TZ}cbm
z6fSrPAJ=zf)AAKCE>~U{UXs74!e7QQur3Z+A4hoE{HE<k<PR7*ur(a+h)2B5R~f?{
zedXNv)!%TO2}58kKk>&1wrktTSZ{&9g=2CEJ^3uy`wHBB>N1QxsUF7ntZ}P((Ae_n
z6WGf`?B(29Hu0-C;0Sx#NeB4+`AcyKw!V0jciGrWQV;yBBWZ!R)b(y#4Oj>J&S~?a
za$<(k_y-R0bqRJ1^M;faUsQ&74V`Z)u4-kj8Mn9wf8J~_C*2ZPscn9`3_Qr;eKeT(
z(!Jub`<;J4ZZCUE7@HX<hLi84#t+Zq&HB6Bp;)LXChhVa{9a~^;+%L$EOdG=79s<2
zPZ+vitT##w`~6~L2j4g0I7hRw^$N;v3>)6S_t<;qiQ-NAJML@i4feta;}rjYkO9Uj
z*VVh+Zz=n@^N{szY_b`9FJMd(+FO^@UB4#bgKTTKJ|AT)+6^;~Y;GpoTgY$!H5JeJ
zgx{QvFOYnn@{jyvdE4*jD#JF3|86+nnDgc3Ka*=E9mw2;%5*HguH`-Vj8e{5c*J>S
z=H=_3pCrCv>+RTlY^-rM{zJ$)l0v>DTj<5FBVqbz{J}L{-^U@w(?9jQA4YuP{<6wx
zqP$I{jIuM##9cfu{?7ka^~mwLYnRCDd?~)dYs!<M_WyFci8{vHHbxnF#^D}=&Cx~r
zh5WtMOaIL5;@Nb!<s0U{$bwGhYr1MjAGid!KBlL;=yLo!_;ZtCp5s${;3IFtrq1xx
zJq<d*xwq6U0bkLk=hQisepsY_IhVSBg?6ryLt3X#ah%N?<-|6#k8j0cTir+AU$0+R
z;ovLb-D<pW1OBrMf7)$4;$pv|?R487JN$M}d1VxKUuO55Sd5oUxf?%ZRPQU=&`(>2
z(1Yu*s~35=4SRkK=ViayYM4i^KtIapo04)^Mc8gevE{$k+5Rj}|4F`t4VF3N-6Qhu
z_k13lc$3}zx`@B}5Z~t)^T5?(?C7UV?9rT3U9n(fPAtqHiZ^tVziETJ*NcVMU{<|S
z?4y!7e6d+ySoO|`SQz~&u7l4EgUKV=N|#0AGWXw^A%2_%^WfP3Ho)5>u~77?<DT>U
zBkcEa*j0>;v*n4U)PY?;{gh)xahyVU5Ukss56<Pb29vE0&&XG@(^YBZ>+f;RZJg?|
z`kuAs@qo3AtE{cWx5_56(<U)%7saO!SZnZK)c>PK!{?`?q2!5ZSbIF`yVB9{%)im_
zhI5Uwn1{=c%RL(l<vQ{YY<#7<=Ugj?bu<=sU&q&S;sxr~yBhxkPsY89=d?2?G!QrZ
z048tY>whrEkOq%821B#ZHTy@d>38;?bst;*0<VnXM0N2@ZT>(zvzCLAux)rS=QRda
zuC|8dFdRGWS@7~D_x@-M>u%F_`0+iwoHJh^Oyd{F!mPo{c~{N^w>hCc8R6z@ucw43
zI>BEUxHUl>+z%f26>E9+cKd~g@T=qX>&H0TByE^1#+!I|jd%a)#^F<aPd|FE!=1j7
zU)u<?J$uMPar6ZD4uH9R<Yc<jC6dN7mL|zHc}6C6$=K6#dK)v(x;~sP>Ypt$a4U7M
zsvlPU1N+jk?MK*030$WtovkU?QkO2%>jxX)5o7_0e{187I$|`sey6;A>r;*wW)qLY
zEb>f2^MSVU_cWr8mu3TdY=3vWK><0I$HZo~Rh+-z+?nERwUm5oMYgESQl7C6U+LHj
z7ahpPN16MAUE|^1cN1We{i%*!8tfdK%+d{)dew8EXOG1kf53BY;Hqcv$Yc6|oi#RI
zmygS*ozs_><yV|vnHnBw^St!1iCz?=7bl(Hx>q0R*RsicQVqP9j=v}$di@jYbmcnw
zwW6DqaTGkH{n%)Tl#+u<9Sskjh=j4<M?%X~(NMQ})H?vtFdW}Va{+g^onm`P&gRD-
zqG8gGXvnuc8uBc3-;`)*J0luO+h6)cH1zy78lJ~h3gaxPaG3eOM8jORu~@FBy)qtg
zEbs1UD6lLV#xAfwI~wZE!;j{R4_4qxJENf{&h+Ka(a`QnH2i!on|~OODa4kFvq@a%
zC$Ym1cup6XQK=oD*;{T$e(CCzSm*&Ww(f%$zl$;0bM~zCu`qoX<J_jti{7JCbK%uy
z{dj~e{DvdbpY%8R_W#LWU4>a0^r>-fUD{eHI?}0}+)ZUZ<^}wQPDL8?ovp;aaQ;FU
zz7+QVm&l*-3%&V->*VtR)=Lh;x9LDv$KQFE4}OQ=aZl!UaF=v9UX9uKvvl}RIQsQY
z_D&j-i`&^hslP$&w%lCx0&&9>HaW^XateIwf$QQWshh%$+Hkn4b59%pf;dlhoZ&up
z0@H5a!7Xlyv-HLDshvxMAEYzqi?5v1H+gfwbl5l!UOru2>>A)_8C&PxN+0hZ{rrP`
z-EM2A#CvrQ$HGlKCavFtmc+t+;@1Dst<&&r#56cJj&9*HW$D^X<C4A?-Ro%{)Bd^?
zG3oG_|C=#SZ=4SAl#e?jC$7G?<JeUJ7SzMj+QSI;e=ElJ;mG;f@-`dgc=S=v?_iTY
zYx}!?p!;o#;x*Oe9oXIR*2d#a<D;)PCF7U<`0w}FGh6$B%^koa9-0VCC-ZgF<wa*&
zYd{|VgnypJch1Hu=E$YYmA4@iNc>w5gBS3FB>s)KKj*=VPvr^X*SK#64l#jEzoGnx
zhq7_cJ3L(8U?hL<SpC8BCiLJ{vfh6Dzg-6np)>Xq?8mX->KJ9?A<dPO7e;&zE8_Fo
z%gBX~)Kh)?sP7#0%c2f1d`ABk!=^8^AD4*Fo8=^9m%ug0XRBj6b-VXtIhnCUonn3Y
z|L!mcCwZI>?s?rBU^umrEVO@LXZ)t8bzDhu;J8q0*tEnp#eM(NpBv%SG`gO@BR)xY
z|7ioWNEg!Hy@z_q@o3X4Bb2XA?~iu9XLr?}M%t6#wyt*m$sX1yyPERi>ty2J_o~J|
ziaCC57`)cziBsX4Hgs5~UMq2=wc4}+zORSnYs7`ijio+lJ43Eizuhp-jg{B518&#^
zzpJ4xrSaBc+L%Y)D~EA>7#Dohd{rTQDxW#&oazEg|KbmKv%l3lQo_Kk{L~gc(KZ|Z
zIGx<sME_PfK3`10$Bp2pI=y1P^+`5!FPuJw*L*GBn<Q^T$2+$$r~kZo5k_@<0;ex>
z7qc?gl+%G#H~Wb-C*t*NX8S_6!hW7%4>w`g6CcCp>E_%WzqH0a4pJE(*~|`~Wv9jS
znrn*NYH{`8?<?RB{}tok*?b4*(mo=`|1kV#+r!<Hirp9cn+=|`2IxogWc$UK;*Z>7
zi+U^7ac(SxG3xxbcQ<Om)dK7m-fk6R9y}Qhm9|Iyw<zpSj)wk?V0nvZ_-q{9{yrM+
z5x15vA(nhuY}sAxIX)KNz{?K6-sjc12K+g!PFdA)TzTtzVAQR~e0wK33LInlC)VhG
z32Wwf$7Qly`cV9!v$&~&{1goOyrlef0r@o8@K{dm)u!dw@WKmnMB<`P|H2pj&eKdz
z9xr$frq^35riYjDJWH+5Vflw*wb)$_Q=uuJ`Lg_0Yk4p-pt+dN?~=7}kr&)k2OnuD
z@1?xyuZkaA(FxZb$9=|iq!YHE*baG9Ecy0b{Qc23^*h?r%ltmPDiN_p%k#<&kgw66
zeErSE(U<$X;VW->pA4>LXkorx*{ST`dPDt`TSB{2by2SCo=kLXklfBVIyO_?H>lrf
z^IHFTW>#@h0a$`>y!V`#suum`k5bjgE6AmKu*vW0ju&}WF0>3=eL~LVG4}olT*~bn
z8(5Z?y|Rnf*}?11mA604@egvbN7uz=rgKw0EB=26i=Pxru%}a1;FI#Rw!&{<NiBBQ
zguTDP&f|0Nr%5BWpB29S;#`XaI14i}y~6g@XMTRXDUI0TH#xV1e9LYeal5?RW}KJ(
z6+a;7cv9}+viZf^@;djJr*M7dJLWj~=t4i^sBE*>RQ>xdKR~ZirB=o|+=l-3c~hKG
zITpsG_D=SQXt=jrG%UewS{{pp4qGB2^}0wXxH1w(?~a5s8Ss*Z(eQYKXeiq>8Va_-
zKRQRlNIJg5?>FiD=62C=y%{@rJsL8RYyF~O-Q*~|jfN)QMMD|Ves46q%^tEJmFqd|
z{5E;drP1)~N6|2DNHk=QM#CTE)4|a&e!5rz*O|WxuflVNxaVuUXy@f<7?nouMr`qF
zZfz+lw+cUgf*F%vu|Bzrbv{vRkNM6T6TFu`$Ncv;?`NHL9UB<M)^ZlM7K|=Gia*a7
zub)=K2e=lW&wcF=eRhT}pQYFQ=H%<JA`={@2Or|ESDrRbI8GI~dHE$de|i*s83x;a
z^#VMuW-h9{>x}<%&%@i=bfXS&{+kAVH|95+;lVA%w{7SFdATj#bliGB<C@(VU)dtJ
zL}JFe5KJBMjr}!vJq%d8fbP@v4(v5?I80ziiCx)yTY0I*_|OY@OGUO&9A4+4JDE~K
z{(H<b%VAv+|1~5X$Or%07h6@rb*iM2C#4&&q|xrQVw!Z~z<cGpY$v#8@M*XWt5R)=
zh2N)ISFT^zjxbNy-+UpSvV|;#)m_N$7_Jh5+v33FKCy7V7d`9YUbuI<EBqt}>`!&>
zao64}pVMuu`pl5WSt5VGQLgQ{9G~Y8%6)g8K;~*VDEt|%jp^Y?tpsxegXC;y$$2lg
zUW{!-ja^f|>sUoLU_3_a-%6eE8n$;YznPogywaclf3IK2k6*;s#BCTZzsrvn8Y2(z
z0nUj({4x=T;eQ9%#{9lKUS5`e?#5ngx~8JzyQkn#ALACzSLUzJIG=6+9O=b&J@ehy
z<#=0r|Ay^XvERnA5qxgx1Y@o)%gHGFMM+IOqByCFOT5mn{|_FlgcDui#0<}#?U}Es
zPd?A7?AiT1^XybPo>@3Ezkg{V|Ge0H44=b_CAiO09D`i?0_H7J-luZAGv%M$*Izp~
z>4VJrB&T{cX{HSNIE{YxfrCBi^FY5(H<uHJMNRw2U8!RvqP@xbCsDpg8&|!9yL47w
zdz^><&S`<Ky{64%g8feV{<~Mrc{Ro5;n@mhKG4u};BGzgR0Dm|L>%ckMb+b5{quHL
zyh+&yJ+G2yeHewY#;gPR#PfG~_I~Hrd-k1<>f*lpJ*Sp7Om75-m0KI$-eGU?HIZ+@
z(8lGBb*a1g_w{+%-otWbnb}`j&qxK^)A@ai5BpoJbBv#Z0WJ8D=vp?mf-mA5n}5yM
zlD1?a`D>Z@VlI0e%NBd4gs1AuhvXMu!<JgOf7dU4!-75^=|<qutIv4{;u$(!Fc#9}
zw<f0;4p4=DHz#Z)+fZ2ip<Mc8_BjdH|Cmj{=bNKqVTj{XX3LE%kA=4EVm3~a{e*SN
zm%Nw7P8&bK-X9m26yq0*vR~ZhEM9X@4s#kA&G)eTf$TmzDeayK>^SzLT+*&ssDRg$
zW6wz|eJ2&Ky6yR+hL}U^#NMBkbIqdeu)6Csd<M=xRx-+`!uf}yp-Yiy*!^NOJe?2?
z8P-NaM)B*+T;7eXq`u8?C-(o8wyaa<$#6Gv2R>)zn{&yL7Z#V`IZxD(i+D}U&5!TG
z3#PrP&)bX7n#;f7H4j(zo~w2(fdQ?xb6OmiZkf-%uFk((ucRM-KBTV?`gVvm{fArr
z{SPeB=B=)M74EHk(7RgtquAr(yF&bC0drDMT31t2KB=7CuxG5Pf~(p#v7K1PnDUb=
z{Qmwqb*t_D+d8;eU0dV)Q$y#P(vv3q1m5$Xy6$W(ht~>Mi*MV=K{ys~*U}dB$FU5~
zb#d*%X0Xh$wf6hi)>2j%7=1|lP8zEv8RV(8x74%vjP@UJ-=q3vYZOPtkDmJo2VCl1
z-yiX+3vyGbt#8Q19`W|)%CZ-}rXSs(z?L5+JK1tazpK=QZ&l#QGccwIEXV@`*g>Bh
zaEnw{#)_Qcn(XZD0sbqCTrPiC$o|ELT;q7e`THM{CuTzf3yQg(f~n=$Ap1J$Ilb6n
zcuO9<FYW|i(~Xe3hOtN4>HRPu{(a{{_1LOwKdag=<Sz<8*j|Z$#R^ALLfG#86gHOt
z%YR~b4ZmTN-{Okey6e2YyiQ;4;KyvX{|)-^mzcwSb63m3eZY2&&mr8m{@><OKeJ9O
z#<uFv@3i!JMl|egVm>JwzH%-S_HT-W)1OB|v3U_|E+Zj>{hRnsf&WE9mxu9ndj3Hr
zxs>X38z;I=hE~9Rf;>;<Xy{f}j;E^mtylb}|1(EM{YPdr+>)bdf~!>66b<i^iR>bq
zJkPQT=D`x(-wKbZ@0t$wyGO&qSk#<`^B+b-MRmMqeKZu^6SWRD8pd3VhRL^-b07c7
zuI|Ti{;miI>ROxJlAYo>E8iA}z>h`zX^ej@1TzM%u)gP47?fTtK*#T+bF-7_`a-(A
zh0Yz4EBW3!v>({l32Sz6#Ag4K3%Qs6(1~e<VWzPiYmB?ngTeR6<>8hO+>-C0Z|98J
zOvj&e?OXZi27Ovsk{_i9E2<e|y7ndAn*{IU?=iniFY>mgZ)A;eon>4PPci>APh5e^
zl=6GZdhyIUn7azsr{7cO%g@eG&++D(hQNeAFs!q=p%(oA%kTqk%w)TT^WDwKRAB3Q
z;L8D=rWK5P_M&;ixE@^Ackti@ym;)U`8?Qhg2e34`kx#aEc_QQY$A5NA{MI;{}#dA
z8o1+J{n`yb`Ofx2lDIQj9^LVIBo(>lSa0WFeAjdJ{ny4|nDNMJTxNB447aZ1_ZG+V
zxNa7nRBIeO$3+{%+xdHa^GID=z_mC}D_MzOsSUH>#><_IU3c$!_7w;9PYL<oGhXb!
z)lAnJ$3N(Ng8o{|J_hVHhjt96{VMLh==@djfN_dukW0HC2YA%@JjrG%u{~pRfq!n)
zP9A{7zb(<wSRJb&26;i81S{g-L%Pk6ZswbFwGykiZbt_?!H!40!PoQmzq+rsaw@u}
zi~HL>&p)&2X`V5b9#km}M_^4mes>F8`aJOWo;^)@ecYdme1Y$7<-aTA5RbpZr}xGg
z-i0CLfd~#ZP>eNM9%6>P&s_Xq9!|7CT(t;q!E@sOhqQh%9_3iaIr5FJ{bQ`$&`=ok
z7JR`kUe@+2`eiKruAf;MkLimN`i7q7s3M2*gx~bIzkYgEf5qp$&)9Z$?QEF%c7B{9
zueK@Y$ugd;k8<mqj@o{vnR4J-{C{r0r0eJ4TjQ7YmHta6ZCuwICwZ9-ETa1j^j~r9
ztk+YU`{;|l=9#oJMCEA6)g<L5`0f1H>L1TJ)x)cb;YY7~M*N%h8TIx3FFL2MZ&wr_
z>gTE2Q>z#|E5H}z_6(R^HLZNl&6E%>%aQ)ZwvO|E?A`~`!VNY!4}UrEIh&g2-M<gb
zOC*S`+u{&4`M^BZjfw+0T!5ix#D#q79sc`%wl%@A?7Q%WPjQL&teI)cW@^ZF7LhN_
zO`o%i2eO%;&&x(iv7_4Nhg|blU+-$*75Cvp&$5BRqpb(<yYpcDe~kEZitliIp^SCr
zdDz9Uqhf(S*(IC4#xL&U!z;n|Z*6yH!5tj?G?ncYYh&<$9(X`~_m%$E@2%p(&E}le
zyO*8)rrZ?^VxjvqcI;U@+3k)N>Qu(sxHNc_oXvUoTOO8=cnsdBk*~iU33)R`eY@V;
zg?FOi)%o%|KSsm!jA9;b?eZM_ZH<4V$XVfLySMvh$9d1c7e=XH+nnl&haIa7d*Rj6
z*V!bSZrf3g!~T(1y|3>0^orseyre5$G}*niuPXy?_0XmVPl~7Z;ZB>>ku>~H-yU+#
zSj@g=&RgHifqQ8)oA1xYPv(XPo_ogri~6e8BjP?>CvQ>fWS%hA#mwE4*U1paGU7nB
z9FOy`(xt7Lc}_m)1#@Jz>B~#}0)M)-0Z!8dUgIhA)%OkCa%84+yXxU%j*V_)uAnh)
z<#-|dD8aq|IDe~_c9rCFa^OJP)8Ts<O6t&|OAo_tc=wBEJ<wcV(Y?%F%-IZ}qo3eQ
zJMrr?c-s{@>%Z=<5xd{-NpSAsVR}HiJARX2N}E;Q=00}!AB_A%-A}?meBp(E#jW@9
zJL&l(_B=JCJf`dZX5-BrkFTX)%})B?Bfo5$!*zW!Ss$moF1XGKHlGV_{C)~&{T>GI
z*JfCI%yzhKUdM8?sb=aD{~yaWq%bLYP@KB&?)=2g6Jl=H*4?51*T|nO!pmpt_nG>2
zn!cPNzc5DG$@*0Py;t2lR1vzZ&!(ocUV?tV@CQG2&bMTY{lLX^m>>D~4e>-d`BApE
z9$$E_kGWO6<k&sti++fNGz%i(*3^jkg-CeIcIkBcUqr%59OvFYBVn~%&bl1(Ir+>{
z6^w?~wx@I9^Tp*W^G3s?%ITg5|0op=`Q?0qJWne8r5v8~=f~mz_VCDDInPPvq27yz
zEo@|~_@HgIX!r|v`tn))>4j)W|7J8)SGT-T^IHR>VdML3ZE-XlS;Ni_<3YdTQCIPw
zbnGXmxs5{JQ!LA-f;?$;zV`*b6TY;EBQ<fJoqTW)_Li1iRMR&H9y6a?$2@CKIzEBU
ze0DcJTeMYw>-YY~<LjT~oR64;rx%6K$&nh<-B<BT`cmZ*ohD!WgIiuRzVzuP-RKv`
z|Lpu+9`jd);X8~yj_aPNDks3#-F}I`f%mQXfX7?#K`q6ReCIEozwBM}V#Da+N9<q$
zPP>H7t!3-RzvT+~H@{cI-9^*c*Zb-_h|P4j=B|zUzASG^2OeOLd9`gmTh4VKolgZP
zw6`ML`He383LDxV$06X}3|w!#<1v_Uf@FXp@pn~ox$Yiu(Q_Nf61-?CEFTMB28z*e
zn)!Iko}Ti;wqGXTNig~w*!U({>-@kzY}UEodx%53;ySp@%WqmY)lnX;1AL}~sXD=R
zT<3N-@0g1B#*D-}XW&vxW8n}i9(|fGNG&(@D4qgmDm|x9UJ+xp6VJR2GkU`ox}Tw+
z_TxIy1YD1duwMk{88i;2>8Hz|-{pyq?ctYxfh&LGfY-%vVu1|CD=VLUfDix2nElKL
zt}Y}ldD8sFGjeQ|<Zz$GXR7<ncgDXhpPzrO$v;;jU;BNYeCgOuem0%)Jj>^{HMSS|
z?*Yd3*mLq5{A~q(wp>2%9l?l>cjUR@&02c(+F^Em6c+6hpKKG;ZBV~;FnO_<ZzilA
zD_1)Zj--g!wQUTnI{;Tt!oU~vm|xAn=K0r!Y(4zKck#JHjq!f|cgNqx*7)yvkS-MG
ztKn*VUDbGc(*Ty-N7o-whxk0~p+{gAeJxkg8X0x{9~|qaZ%*A*H+@#_H!<eVcWq)S
zo9O(be1Uq;aqN?`=1i{Po-lA2iGO=OEnE49%@k99YIgJ?9<kSTxs+2@+q+(nD`Y$C
z^u^C?<(mF}D#%&kJk#E=?IkWul2^lX`r4ksd*X9yFL$vf5(m803MRX@As*QSce>v=
zd{PsK(bo;x+pwbS5N9g~i=S2g$%n-rZ1XtG&Z(~b{-j&#zU=@U-6AjkmHuC%o(uJd
zzF9X3E(|x9($)K>&Gdo(UdN{D-0&UI-{g&dHh;s8-XUd<(BnP$%2#k_GJBx^bzk9U
z%kx|LtQm&2KV{;($dHG`6UF2ps^U1VdH5atp8nrElno8{UDF}#v>(n9<Hv`2S7N+(
zmlnHD*`KYI$Jt@7{U_^VZh8+ry<EWqe2MX`p9O}|g*>Un0{@Eh*mW6ooOp;mZDwDq
z#C$98u9a}{%e!&z*jaStL*MZjY+jk2Hc+4S6>yt}@hLf(y<*aia_K8xvPM$;oByv!
z7<x7mX8#fi#~+A>98IGki`X-TA8D3~4(3%>862kpzQ*Q%ngpA1sj_GJ?*GiarGvTZ
zc|<>sEMwlOj<};W?$8mgcHsZuSQ>rZ0GG%O$Cg}|4-$uOyd?I%Zrwu(dEwHyi+bka
z%NC5YzEA(Ge9tps(=<G;;6M18zR7lvb&DD4pT5b+Cl1cauiw&lf59-eat_{&NCQXo
z+1#{r!2Y-Dwko?dKe^@b^Rtb@VqbRimb#p($WK?ZKC!y|K~3v`)v*?P8uk)D|0173
znmP8Q^Jl8!ErrYzDC<IIzT!T9-S1<{c&`F{a{kAMaVd2<cbI-`!CPV88}#TA^_uNm
z{&M;W=B=*>)7pA=f^}Ep<tx6Bw}uf3-&w=Mrf$*U_}c5P`(Z!)sN(z}xVK}797jLC
z#CA0d$u%Et@-dV7lU)b-EcceYfWz};0T$*X%gFfKILR$oNxG4)&ZoH!GcNOQm#iIP
zH?>c~DK@?42bjD^98KQXt)EFp`$g0x*)x7UBqlhl-lVDh18gmyb5mS*UOhhM*TyeZ
zrn=3Y;n+m|HB7%n;C+&|!@9Sd;zRnsBt36^hhMrv?~PwacDIQwH2K_EPqSWv?F{H{
z{A=6K?V8_k5qzN}ZgE&{=Je`_`Lal8&^Qu`6pe&m>P5n_c9F1au(_&!kx-&YBy75s
z93DQK9CBPp4hOF$hp#h4!c(~;VSTAcxK=F^ULF|<8S$WA%A3R<E|;VCb<I1qj)wQU
zM)^Q<O#Px^RZ29h#9dl-iH5B8q9GX<`ZcF(@TY=b;a6Wq!VfzmVa>hKunE_h<ho1=
z*7%H(_rZ%gs`qo|wxUO(;pPRr=Snm*gdMGOh;?$Zi2~klEDT@R!_q3SguS(9Z-0;B
zm%rg-f0t{>fg{kLg`MRA2b-Uo%Fldho^w6D-9<l*!#_Lp^;UYgg5IyZJ0A7@?(n^9
zjqj4ve9w70@{j!RP2&nX3uWZ{9=x0PJ1oYH<81Gz<@x!_r{Q-Zj??G`ezk`6mG(RF
zn^$4-M&o;-6T9cLJB*aWoQ6BW+OaVAW6%C>Ie)-c_xsdyoU8LbeS43tcC%ikwffb^
zg|(+^DVUv)?|cw`v)zwx<1uW0H=WoW_kZ*v@u=94Ek(&;lFsk@j>*fAG9(9^E4>>Z
z{F?53CU5f*+Z!RKOc7(^HPLRd&<`HY#$P)1_8w>g>`H`<wpab0+6%_ewVSZ}Y!@5~
zm&w_|n6zcf&0%g+aqny1qkG+)mGgDnR}8L~oC0_0>qi^$rXS%gu2mWCjLgH9=y(fw
z()Lxh`Z^5PS8?C7ixejd{f@7T9@tA<7&D%utvxiRJs05%U$Ze|I`cR`&OSfAVlMC&
z-jo{G6szPHw<Ph)XCC4!<s^m`!B6ng*UQSa8JjFP&UPFpt#P|r>+ZZ{B7ZvHem&!N
z-8P2fyo$HhC-3mLW91Xt7~@rZZR+Cm0k@fhN1cKncmC!>)ur%ec(7ExrptR9^Pz0|
zct>2~W$QXB!@?)wCtR%a2;Q4E{!R$~cfc5|fbFB%aX*|Ui9VC%5jqa@#~Y_qlW_P&
z?4Msd#>ega4~NVtCM>UyUV;0bvuFZ5r`N0Z$&39ePNl=?VOf3sa-8g<%X@79&{yBl
z@l<Ew{E53<Y{w7y^I?5UCfaXtNFD*6t%JFp*-V8^aBV%`v0C}`wv{$spu4?Ee)_y;
zvFFXx2Q!V`G*~u;u1`{@kNIfZVaDx^Ps9_m^bHw6=jX!h#;cS86Mr$DPk!SX&$+x6
zrW@M__TXCTQ^)w`xawXUt5;fcaOvgaau{d4@?ExhUH?CO33t?<_DA9LPCgLU=2NHF
z=9uI4yn5<%SpUpPfdO=GK~eL@8S&^#xc@2KLZ2RHM{C*5{v&YpkoaM{Sl|omK|Yip
zeUl%p$`@a|8;5;xS{!#a8lJcj4b$$kCMAz~;b+((ERMg|v5X!4VEif!fX~U+q&1Sq
z%)(aS-L;zH#{VLr;&+j71uo{85ea+VkA(Z)i-ad4k?_U4k&rMT624s=3F)>+!XP*r
zk^kO0Tg>XY-P0)lMRr7AYA#@#xXnQ}op&bgG0|MrFnLz`Gm3qlsH5H`t%=Tx4~ZZ9
ze-;fnVRdU5{6ku?<jF{Q`bZ>HDjV_MZ&K)cB{6KMkr+nhObolOC4}@OZ>7ZW=-Y{*
z+R(&Mw@PBj_IHABeI<lkV-l<pO9&(OB!s8$PYf@nNen}-B!m{16T(+l6GD@$iJ@4j
z#PDs)#4vAoV(9pDV)(Ig-%xQ)-|$rd&l;K(hGj_(Z;Vb3hjT>2-ma0bW2t!iR3!X^
zYh~b1R*9n*42y<$#Le%Bi4WT@oEY`ZRck@{h<sw;BlYxu3s_`~D!yk9V**aLTwb58
z<P=l&xrgt^M~*yf?(tE!nje=i{zsmsv*qbE8O}DFR>K{FIOSP7Zo8)naqhF+d@+vF
z^q6Pt_l?Hy*{pKU!m&=ud#HqWB$QtTHjn>-uJA{HthEMeoi&PEVe2pAKbW1`wKqKX
zwDL1IW*2YrO9SB+-Rw_K`>=(_x6xT#VcEB03;KI)HoKgN*9?bqaIl$sCUg)hb%Z57
z@T;LX0-v&&4Q*$yyN|2;AB2s!xn|8edE5RLGMZ!~doJ@W7va<I*3|zhA8<lFj^KB=
zN|(KCWIIeyhb!!U&`zAo{vnw1y7HGC(iey2uE=BdH@GL;ejL{|&%=ZY>e63(o|~+l
z<KVpZ^+|#y-OZn<dty`es84U^v$py^9P=iuyUc&Y{r6Au7~A~*lph+y53!#{?fCv`
ze943C>!fuq)2x{>&ouJ5wZ!r^&*3q7Ge*L|L&>4V_T*4+PqJ^xB!`{rlSA^a$>EP(
z$@CyOytyhloc<=+e~>1JTk~+8rODySlH}0xaB`SM_r}(Uge@;c!e<*Ip&7fFU6h@g
zul=Qibu~S$Q6rUcmg2ZoL-x3FYb1>SC=#Z$iG)`>MndOqkuaYf@5FJQ%MuMY<aP?Q
zjfP9zqT%~kG?W^T6MZgj*aANeniD=lum7;V`S)n34^!ITbR6I5`G~ph;&i9194(HM
z8+MrZ3T5?8jZFA}zT4PBY?i=xjJX@TRa}ebZ-Vt3<!#_-sbyj!__}`{th1f`C0vEA
z@&6ZCVyr(oN5_nRpX<i>4n2tDsXX|r@&Rdc%geylhsYxPL-3u^V&q9ridE&$^Od)T
zqk_2uesoS-Jf)y=8nPv0TeY7Yr!oHs&M)=sd5h>ceTe@*`lu=Ljbr6h`-{il<{z5z
z<uB63a@GYDrgLm;|NZnYZQN)1KG8+|<2U#Qo68;L3lEFi$Xe2$jqN54@QJIWKY5xD
z?X^P;xQcI_Chrb+Uz2kw2>+_olox=@BVfcD7?Fg_OvY(u^ofN>$;0o8kKpx>_)k2B
zt1d?~+P&*=oB8f3`HJ;W4aK-k@F&0f(a8zQc?rf&9_wAZdGbDA%jX|7f2GdP-|~&?
zhuFYleB+baT9(~cVJ9`bhxr1ZQ%|40V$52KiSVAsy2$(Xu^wqK{_v65!<gOs6)ZBQ
zh1gbFW7gm|V`t13{SR0A52qEk9LI5T@_|co$=evGbT~-`xr{!f+cU<>c(uY+%Hccv
zE1P?%CI^T2JZ@W|ig|6v`xwhhL0-u5W{w5Fix)J{lM@f36Y=km*8TS`PBc11m*`Pm
zKL2O-{o62O43k!B!?s$^Y3qwQ<ZHD5Tk-FgyYQCfasbBT<KeKGJ?H8vFaMS`X8LcC
z?dSSEeteAcg=X2`Xk3rsLf7cNJ}g#ROw@>;^n`VztkszBnVWF|93}G^@z<Y@|1G|{
zA|})??crUS%Q&U}I<4QHJfi>9yPSUPKyK}oKO^aO;W=C2EPHu^y<A#=`^=Jagk94n
zi6tk(y9wgL_u1QMZ5@g4(BD?}vkn)JM$OH>ORv>y-P`yJTUgmz*|z0KMi_mE4DMvE
znDjHobzFC}E1%IrUMSI6D{tl!_vq9AsavkoxQ{j+J%#V=!`as3Vfs2URsXBovgy{=
zd@R?kPPvl#hej~IIz5A(dHJPr`mYh(yiE7^<L-GzvJ2ehBfPQgeCM{{3LDtY=lUCW
z=tX}gRyPiKLAs-2u6@=feiseDZ8bkn#_qRv(EAi|j*~O9YabIARJN__-SK8}5>?sH
z9rO6EZB)j!a4**dai=wqHOEIn*W^eT@TM5@W&XTMB#bK-34dpY*Xf+QmK=^=Ob(?|
zM?%#r$zjKz$syBk$>G@T<nT{hoM<dgBu-tuDiR9qi-dRb;wo=N!=O$$koa;djGpc}
zWjjYgNBG_N_2lq<?&MG`O|oynCxwG|lEUFfl0(KC$>H**q;RWDQdm;0Z^+#&F}#JB
z%>O<i?3$Jk`VLA6PgP9_Nre)^-Yf~B$b$(X+mi_))$<7<)+8ZJ>W<3{P6+=_N(kv@
zC72sa2-lV+greIM!j1C@;b`{6Q1rRPkRv(Kf7~R7uXFVc<=*QX3jNSGbSRV*{z*(S
zkC+s0oJ;ae>Ey7rWpb#C?;P2b9EO&RgfV=_xx`3#w|gYa#r?8;DE1b&H$UT;HW#%<
z^vwnOS4><f9lvlX8j@0(2QLGQ+TZ1S1>nx2U&RUM<%3e;GFk5CNsAQ0vrF)~CE@fF
za*&VEL-yIO5M6vq?!mF)j=gvjUp#;>ZQ^5=;7n|G+jrK9pM;&t{OLMj({1;8A8xG}
z5Z{dR&QbhQ)vwLr?-0MZ?pSIZ=u!3o&l^6=&%pRIbpDn1VE>2K->|1~<LEEj`s4#^
zlkAsdSDS~hp(Gf|HhQs{4(w+cOi6i*4Y8AAL&X$xakVvKg2U|M7~4CptdnBHU*t;u
z6w_RA?4mf}Z&+bJ&+oR*U2@%cSW$v~UEBt@lyP7-pEnw2rLgZF`o%M!R@a+#@F@5^
zf=v}x|031I6wis1Uc$@Nr7rxr0~-?D_m(<lhvS=Hbl>xA)-{oaV)M$Lt8Y){6?0@%
zKYe>Yev_GvX8j4S$$4dFrz7yx<agz&o11Sf$wu#!&pTv(=>2HugpYss4!bBE2^ZES
zhx=M3hjITVg_rs#g`?e*!o>MWq0@g!)*&Q^-EES?#>8a*qn;f4PfQLS1|^3NdMAfN
zjvefm9PUg@4&_!PhbK-ZhpBi_?GeeLdEeyl`C8XhjfDR3=M0I2%OB9aDUr~wzqzm8
zkx;n=J}wuuu0SLlrl$`#iiEVT&G|_rWIkq1kh$OEHE@?!=7Q;Hl04ALY0;2+B~G(L
ze4#DR95d&20JpI%K!2CuIR`R{g>s683ec&ld^A6haUyO<_Y$x2@$mjxW0Bk)_Z)5v
z@Q6Cg`Ay??5KgbdEynVbqm18b`jmc}vBoi8TS<?1icb&1{-60u_}`fyc<MSmk`q|X
zSEiBMFU6ldE@r%yMb2DqAT^G&_;L9g`TrYF@umD`3u_lT;WtH>@>Ofh$%rGmofb#@
zCAPcb+Zd1I+$H2=D&f|~zPho`(0~q*y>#MdzA#FMc2(ei=)$`Ea67wtq~}M(0i~3Y
z=>lEgBX8p_=h@Zvoz~{*gW_bf_~BuG@FKiEHW|kl<5++5QTRy4+VTKTxIV4@<I(W%
zMt#308qUG*i?whQxO|;X{o7i(q*^OB{2Kqz#5w}oHGWTMr#&R2^Tl4ppX!TED#^<}
zBVH^gzc02-c3r+6a*g!#33a&qxm?m7+cTbV!#9?2wMZUzS5%*ul3UXk=>pzTLtAU(
zarz>oaqQNL-FL)S*x9xiZoqe4GKMvmiXFa|pEYKmvg!D`svmwg-j`qlAO7!ceQT`B
z$l=T~PHl|SpYkocj8*)7%7=^Mq)%8AXUu-ZcOExxZAzJc!g(UbE)U*Q-*&gLEZ|%}
zxu3l5ne&)=FMKJ7GM}IaTRi78&#d$dOxVW{uY)I_vEz@_^F7$r^RBJO=lX{}!QW;R
zd)?Hh`XT;*ZO7?It0_3&Q2sZG9e2myyTHg!u$peS=@tu%`{F}G^yxG?x7Fqhj=29{
zW0Vz7hRsdt!$J5`)ALSGmbX}BJ-T|tzh5@@1Wf!D?>MhcIL_1mlgIcAR_d=CaIK0u
zUvpi-9m?2756R-K`gt>b)W6%8D^q*knvHj}l_mH?X;?gMJU=jwu%nVA=<LvxkRJ|D
z9*AQar~18ckj}=$_&xfX@puKUH^zG!%5ndHUEkJ7ZoY|_)cMmb#Ck2^sqK5M*=jp{
zzN<NnBp5j6E|*#|47T?ZAG`yDl@nth{}iN0e8!}V<{ohV$F9j^@?~qp%*FYnsr0Vi
z2fj}e6C3Kwac!)zeF-NoZ{8`N7%>YTlHR*I=EZm6F&$1u!wfm(s_Ubn+DFz84##Dz
zC)_Y48p6zIX!?owN9S9QJ5QdRTwg55uu9H(oxJtO^3bm-qrLb~+_$QlV~=_#_2)?V
z-&b<j10&(XzL7A`b}x)9!*}0k770UY$*-4=gx>k#X_km@|0ahG_)MPT$=+p54(GQf
z!>8nMPFz}6{JC&^a(F409EQQ@>K&4;^-2y8<V_9(&nJbwxI*0tN#TP|Nnu~Pq%h>(
zq)^~+-|+9^zTr|z-%zxb+{sIQ!`jd{94_8BTtAZ-R*~#UiQ&HHiQ(0vwx<&O*G@vX
z(kCGls+|CD6TFkrC-nQHPZ)f$Pk81Yyr)D$*yihjqnjp#^>5-oJ@B7cLfAbT@0pzt
zmalW}OhRaxIWf$U|M{_JV)$iQVrZPXZ}{)^zTu~(eM9S;_*B)TFayUaygVuF`Y|aa
z-Yf1dnjF&BP7XtGvh}l)!&m#0!;g<d!r5|>P^-DQhXE1aY{q+jkA$KH%vro{%=$({
zZR0d-DGs^~k34G)Q4tte1@C}IPs5;=eCKho!EbEma(Zi!VbBD8=65#v)C2VOKXZzF
z$QAgS`VxL}&N?7=T3`p;!)Xf6FqiU)eAoi<G#)(kJMYQ;=zTQy7+~9HU%6%x4&=II
zvS6mQ<#WyPuQN|CmZ|ucwG^r4GIOzm!fXrH^~O^QB;4f$yVy=0I(}d{>}ONgY?Im5
znEuwL^x+%cvi=!wD9c8Av5^<y;&Ax#BD+61K|K7a+{j8ihCNR`D4v0fQ;*>Hr}@8M
z`NQAEHlzjLmP(v(@F%u<fZc5Ktkv581<dfw((Epgebuatw-oWdj()n9%Dz7NSD&0c
zVO@fD7sMNH>^B#C2!`=Rx!~Fiefhq&eZVH?Y3J>{c;Z9yrTo|twz~WoHt(8j{NcbQ
zxT0}60sD$4<Dl$wm$7>11fIUhdKP2a`VDzfoPAUSdCM1~;S(I=9-QLI63OAYgGr(J
z^}b>Jw|&FY8~TP$pY{#em-G#j-tQY;SkV{Xrtc^FhL1BOg+0$Cg(GhzdABJkOgNqt
zvj2+D=faE1Cx;uAlf!oVt*#{b{!CJsj3XsyObS1x6$2DW3Ky#<g;TF3g+*PGd^bKR
zG`pP?-fxf`A~@6N<;mf@Uy?(Qf^tYz@SZN>i7}BdN8FKs=VVI*JMf$<@F0wpzkwaE
zeW5+e?SExW(1vKZf2TD(r=lUhT+T5#)cr3w!ax5ApB{K#-wuQOE6k7M$OTHu7sAn!
zFz^DrJvx)GTqNd)(TTI1qYH7)oEX;&dQ)TE-I#RtJaOz-c=ZNccDpg&CwD+U9y*Ty
z{6f$E6c1l!3)hVgzH>xQXQR1^7JOYlxq+N~->ZCOExzz3ADNMF?3P)Mqn`0UD-Jmt
z2{p(!=ObYuyWD;$67IhT7G{@od)hk<ct*T!N1kI3^KgrQ<U;93clI~%J{+6v9Xn;-
zo}HXt<NRW4ucqT1<K<#T%8?9@g(nBO7q2Lp+xlPqH?Czg^vk0kZ$(0@gOO18n@FfK
zClY?fLD$08Era2Nd|G4kjwf7u%6h7l71mUVQCI$Gz13lB%Ehc>?RU9p?cqJX-<XYU
zmtw!_S-b}OEyG{riLbBmZk&6ouQry(>1pS$;Xn^X<%>q}+Y`-0t8+I#z3>0bvt*J-
z%Z-n)g~`V3t@7-%s$5Mq{<%8as?APbHa<;_UmJ5BY;JBcn>AjuW{3gUVxzV4tUKf;
z4w)}L#m3p~#fxH@fAOSS_=NeGAvj1!bBjO9x8yKS)e%2gfaCNkWbKVHdmqQyB=`0r
z{!$&!DR15}jWOI#mXSt&_cE^W|2&Sb<M{wbYOSnUS-qQt^IpAaF32<E--~<lJM~#7
zZ!n*4hBxtVe)Z~sleWgcg8CPgt9uZyzF{pB``)>Yjl;vVAK5o{*Awv8uExD{EPV5Z
zvF#*J_KrSxyvSfVv`N0{4MYDQNoN@z)z&Ow-0cH`yK8`hYvJzh5`&OrT=od=?(Xg`
z!GpU8cR0AaL++b<eynFEGn0|M*XpjWs$Tm(F{LxH=5ku*X-W8xCSbe}aDgApHw`V1
zIG2Yw_wQlmE8i(2vF`76FbrcGbGi=4%YTVu&U3Yhh4H<8eGU=F7_AsPW8lpB9X)xE
zQk$8Btc}!5h>>%_A~WHq_>2xyh^3P_?&FA;#Kdf@wNtFE_&v@txZqCv5if%{7hS+}
zoWlyN$qubK7M#<ZjO8t;AGj^qin-0m!${7U%b3LL+Vc8R{O|Zaod00v1@9Hu5k9*)
z9-3OzsKiTKM&jge{AUlrF3hWdX!w%Z;AhrC^hhxI1hDlWYHDzHy-@t_U5KB}(2T0W
zodwXB`JKH>;b3mBfkPg}8jOHz>EMUABVSg81F3{ZrJ-NFx=?FbSra4JH)bR;m+@!_
z+yQZS6<8)icJ}gmO1^!Gm*t*M-_V(w!<+OE@+mLzE)n-zMEDfK{l47xM2GnihF;&B
z*SGU&O?{tkm4vs+PMw(4rw*UID)7jwnHRX-@6`r)^p;?o8q47{z&4Y8Ud?Ok)rfpv
z)qLntsi7W)*7PU_yfEjap<c5M_L?$uD9}*NZ*FP7Tc-xPS*vb!MsIoy_9(K%rI3X#
zO<wEL)M%HoTy$wfEO>_J!sfel*XB}^RxT|r<<k5FV3m7LRln?1z%8eCK6Yv$7^h2a
zms*y0sb@o%p0sr-AjBoV$ECqDTza#e_qgj~|8}=t@VUwRyH#tgTXjCWl^&ib!)SxO
zQ4Fmr;L(bn9(ksCRFd42<(5b7$R7a(y?Ro|tKAkjE9&k7x4}G_sN2aQ+xnxa&GTvB
zIe4%noEzfj%O-e-IDZq+vbvK$j$9{RaeiIItFz>ZhJnOlH#lc9Ylyr&;yHi+pyv95
z7wA2_^lR4Q3vvqh=kf)x3i+fI`S;lhxaq|_M~qF*d@2JUJ!(3&>RQ(0HrDGt`n;C&
zJZrNjF>W;DAkX`W%kkGaI?UtwiPM*3r~}WE!@=DKb9P_=^S3<nzddWt1b!#B1(1){
zu!h==LeCk^yl2c9L7g-Zy~s`t#X3mLnt90lUs)5qrzvYaoVCrm|1=REhqxHE4lJ~r
zdhr<JJT<^&c-I?v0f>*cACbElArE=&9by%6q0~t}o7Y!g@n7BlvOoG#yZ>Tw@nYbu
z%;2#EU?bwqohQ@;9G_{i90#!7O4eM3Q}FqW?Tq2%COcSXH5e$A-}srlRV^WPIscpX
z4Z0I?KbYKg5)3ozsb9U_<obg2W8-1@aG1XCm7LE7J{e0{OVfR-nSu3i&#MP>z48zA
zs!}Jf%7A^dpZ2IwU61yI*=JWWR4Rd?<g478u^N4JgIjeEyOrpKTY-1o8uP?0$8WUJ
zJch2<HdMEZp<ROv&D&w<z!^hD-Wu8mj#&VXiB9g(>qmyJ#2D%iMk+bp(5#_`?oBh4
zlGi(vd30isM>|e<R3|U|YDceL!3Sm9;ngMPXR}{$MS0MZ>w;|-`}Fk)JzT_uPetk7
zs_$2-PSiRk;tg}u1DDg1xML+xmx>{tobv0}S$eg==4XgW@pz}#XSkk!IK~m^&*X{V
z$Lyz75dMtwf3!QXpZSnK0&LP3Zh)AXxio%V@=VIQU>Ra%Egp|>!Vxf^!id?AXHai4
zM{+SoPH$&U9KeUe`Hzc52VqVWc}-sU0**!dxK3{S3y1TYT$c-cbD#*lI^@aq<iqpu
z09WJde!lnq>-6R?^l3JAn~Qj6sRj?7%%=<^&;`qSmGh%VmpNZY*08>pfEhP)?3#Mf
zczilon{`VavMW7)mKw|*#?}JV@?ZTb&9NxF5|8&{a0S|4RtG(39KUMJmqU-hE*?0u
zV_rp!qSk|l-UnU{iXm=H^yt-SaOokBhG+6>(R6A*`4rv3r}6OMt*=uf-tnr+Q?KsA
zzg1h})A}@i8IAp#-wZ4UhVI=0Z&h3P`i}Ja6Z=c@Js;=h{3Yk}x!-y#dj+HbV>KX-
z*jV%An4{e0=R9Ta%6-8*6=P4>?C?xU&}Wi^)v^&c%8=KI4H>qB30W^Ii92U9QRh@=
zZ8t_cG2@%!Tz#KS4km{$A^*NV%-TB#=DA5cU=1&Q!ufwee0WRV{KWeFLw*N~?81v$
z2~TP(&Q+oO;1n><W_X;=@wf%usx%np8{-i;#|_Rga~|6<igQ0+d;AL)Dh0l|14c4<
zubb)r>qE%@n;5}o^<;kd&oXEBp{w!TIy0Z*d%=#Hz{}m>EQm*zB4Doc;11&IE3oRD
zqtujZs4<CyM}`nD?O>U3YQkVRKd|9eZr%L7fydVlrEeA-w2NbNhxl8Uc+n^|*q0cz
zx)Cuf2pmC-tOM@2vy2*$`FjVPLSj<@SY`!txB#QqMPd%eD-v8Y_5}J6*yhAR^!NR6
zEPJRGckw%RQfqBRciK$6T}^Gd7`<>7|I2%Coj}bu5o|IZ{5h7-8O`q-4$c??zT)$T
z@q45D5<~euDOrQ#J5gt}qn2v}{sFrTZOQw#03$K3HYav+n~>3+`w1DZnt{W(?ZWF;
zfpNa@9??Dji@E2T@O&M<Z)NIRez!jZ{NFd`%YAg%<9PA5Qdcs6Iu0YZaLg8m{f861
zz4707rWS8Wz5$bWVGRW(V}F+yey%ZxzCMNih$!l)c7DD1$1mG&a_A?YVi_%q`8B-}
zJrT_J{pdc{!uSujv!<yNdam&)J-1nj$JaJczrfL$h|67q;H_Ju9iclt1Hb$O=TNbS
zPcM6spShigX1|)pnl}dvvG%6HZ;t@ml#25z;X|*MpY>|p9<M%wX&T@a=)Va5f*e1B
zwHyt%o};r@Yls~SE_u`*&&=b0JUaT=Q0w6a`zaWzeZZ|Iez&UDbjwHrwm@qty2_>F
zvt4R6)WvmlT>2lpjJdZ<pMzY=)8C~?ze^wS)ZFsoo8htDjleZ%QYpbPdoMYaWrI_J
zgPk(mPIc)Cj}zw9f~`)Szu?rXCr-Ug>{9A{a5!ziJbcDK%U$%{xOD29OBdmUruBAf
z$pW`N-EgaCZbOMe4RlAes?;8RXyMVU;U4)nc{G{dRPVV*l~a1vw}e+Qt%;L<uO^-H
zY6|(|6CBQHj$^U0J}utk(|`x~mC_S)IB!qreOd3q4>b!fC-wi0XY9NG3lBs}ux=J`
zE$6)}b0GeDVQIFZ9bIP)UZX~4JzYEmegVrgqqgh>2Fg5*d_5N3ejr|_NPLPW^bs?>
zj}?Bw#oq>8kRQ(o&!=PUwj55)N1W^8<2_ing&L9f>w%q`k;{3_0Wi{Bc$^`-;lxjX
zxnJWOO+rk|PL0r>wM0!YaR40YK=>9${62}*7#CSrxBc+6;EqzPg<$fRy$l$q6x>Y}
zJT<J5W1(mPW8g{V@f_<l)ehnjYkJB_>YcOHkzkz1SKwT)vi7bKn;4n6AIEc<4if{m
zQ5Vi7Uh=u|d-lY0qt7a`=E<cCSYvbk5XavQP+{`x<Xiln%kh0o#PL(`Ijp<#jP;DX
z{QZd$?<c%<7YxV$EWSgHdzj-v?uy?_viD@_=34Y&Q|CY69R3E|FHXfdyy(@+kzQ@9
z?9~uByxedzeV7mb^LkXp;Zd?$%!|SvMP&A9C39x*Awvhj@I7C<>DzYGs|1E0?bg@n
zZna<O*4piEO*-z@=u3FNUbt26hg*x&8){M9P}c^A-US;97+`4OBtv%>8_KZ+K5>De
zMhgusoovYNH`x2g;2KeeT8=REdX=Hd4-I{9POhGccPyb-OUk1khj=CM*KG7u-+ixo
zB=YHhMZh-HIQLfjR2dAuDbA;rMc@vIBMtk&GkW22h&izf!RBk(`*;f+&pN+iz~wi{
zo%4xFQ{V~89>eR8j`iamcqaiKhP2GpT<`{kh+`Gt604Cj+Axp6?dj%`FPUGvINtI5
z-$mxbU%>o`@A-QeN$tqoIyVj+Ozx~eetJi2H^C3IAqT8sew28?e0#;~KT}7>Q5WN3
zy@y`a35?VKFWwq*$V%o)=M88k%rlV}ht7njJ^)`ellWx^*L8!Ru0U*qbG~i$s$F5P
z7C&Y!@1O=lJM7kh+9ePynaHCCw+(IG$8m*YyE)&`X87ElFAcSM>fw4#UOfZ9L_Fgd
zUhr|<UZ1Z1^=d^Q-uJYjvO$J=zjdqf2V%*4w;uk%`<Kzs|4I^1yaszDc(^8uM*~NK
zM@M<|cNm`w?l~Xq(V_+(I5CgLG=eK`>CqBm&xkf2z2Ll->EqFs1=NX)JnUtJZqA4~
zg63J=t0OnSP^IA3yOS3N@*6wj2LOLI=6w9G44NAAyK}_>8kCxxeF8ooZf+LXb@yd3
z(sRG=r-G|2%sFd_ra)X7yNH;y0}mzjz**MS>(6*?Gr{L{;W#^4ztdQU>%ci+oOo~5
zWX{tYa`|0G57za4^7gSK;92r`s<&`;oU1fwRafA1Ml)WfV;>g&u7Y1R<qx>pZ{P#Y
zV>-CGKX7uZ;dLUxHA5M-z(94tHdo<(Ch?pbyi_9<JwMEYV_)DgUZTg`CU)?7Zeqp!
zC44Se;0*C1-cOQ_9Dk-d7#V!?IW6-Fd=qq!5koDvlC@7f{Wt*sH}P{y2)tM?;zoDk
z1GoE$l}o{eU%Y7kysjfTd^@$1^%dH5D!j}^snHrRPy2xT4Cdb?>M=h5?bd<Hdw^rX
zak+J#SjYS<#`wVf_`SW$aGdHLr*`B!$NPGxG1o(OQv+~3=YfqnGXLMK;XS}MC1--o
zCQ>hs0cVVWpBe_QPu$Hh2wX($ea7eZ<(zCcgS8@9OJTfUZ*WRCw7ZUAnRdK}nArh*
zlcptD20XK+Ip>7&tQq);@vJHMlo8j29vsFxMgV`0<o0AEJVy1=_XVD*%y|mrbMq6I
z)37GLFh7X7$;nS;HsU{<2oB^}mL(r0;`q&LP3&zB2Ts13B5sKdY5zAsy@?sSh%JG`
z@%naUuhF8!l5f<Dm*F>$`S7p9F(3A6Ab!^?thFH_@U7GUm%${1>w!zk!abuY1#y3N
zXP>6G;{R&n31)pytHRnuyT8zo*Mo76f^nX8;_t3_WxzErd7q)Y&vIUK3f|^bCi3(j
zv?lyD9j|#+k68H}Y%^_+S1vEN-Mz|KmDryVzYMwm-CmEzkM}5FZ;y_ZBKCiV2iZzp
z+S1UiuWr@Y3|G?2t?b#sH;-IgGZhR2en}0_;>+pMnjDPOF1=5M22~6_iE%28i|byx
zl;*cny`MW(3$Et!5~tn`aw;~`sUHoTYCFXt_h5%E&2s46Mu)oYap=V@hl=EOYEBoY
zu03>Wdw!SPd`_w`mmct$<(9hi2_EM{A-C?LQ61RfmY;a^MTWKwGT0NC9QDi4N1>^q
zQ*EQpTXWf?6kk1P|NL(9-JYJ{yh&(wXS{mz+pFtfoDgDZ@Nl1EHseKt7tfM_^I05C
z15WMMDC*<YU{y539n``XANqBG^PT<={s*wm)*N8nA_H_kKWnba0Of5vKrvmJ8+}-N
z-QaY3puu#)FUT6K%N#1g<7L6fvD|OP_!fme49<DzB*%kuzIv#`$T5?_J$E|r|KOUE
z%(DrMsa2Vu?RkvXXBt8*oCWu@nmM`)yl@FmJM;JcZ=TBrcIUS1L-5o|>RGVX+okZ`
z8|Z;O0uFlupY#VmURJow060=`(ahXnqY|tK*2UvCtVh;e(XrGq%fU)J!8WY_r>yZq
zXQ?e%_meKL_AauH7(4iT7|+G;yD@ML9Kci_XAKAS<bPOinJS<cv&Nnhujc=xb^wpH
ze8lm&Ne;cpZwI681GfYog!9_Z-}~4LYBwGQepdj$&rO|?W-3@{6pwK{L-G>mI2Th!
z(^FjuUIGnc>=>W&vStDku*SginIgTqM=t65+N0tt$TgAVp<Eu7zH8_*m?OHmp#nJ#
zy@!+h3r@D<=W^j&Ul|OZ>FU;TyIT{7x>aJDTiL+U>({vTdJlfGQ~2W_xz+S9cqgl&
z{bdckY6VZz*U(}xPTFyXOmH==s1Le;XNvVQ^lt}4r8|OedV|C5hAzy6`?+YyT8`XI
z?tB>Q(SZPXn?~?UKCgDdtu8$YFYpqr3%nK(=F<{%s>z$spW&ybq+_4JLhLIAkMkEC
zo}nL{?=bk_N!-riIuJ|o7sKh4UWVSfh<@`Ga9iBp;-eQn628C*J{k!R1a?0O{#o(H
zuOWG;(OOXxGY@CZ;Fxpl7hUIEB!GX<1xLd<%t~x+)}I*cBX5pDpPGdqiurMV6LW-m
zb|7=0HF;s!9pd&=V(Kfn^f%~~OYt}m&-1p0FRz9!T9tZ}d1h)%9w3gFok3muoA_t)
zY3B@Z9QdXK@hvwou{&7xWH>&i9BAuzJPO|E(MOJ1J#a)sNzN<yrqNA9P3SYI1lKH=
zp~9RGe-T45NjX2y+)8-Ht(5yYU%Ti}*yh$2&ZBvgTTb}7C!E(&+)f6=+7H1AUv(?X
zC%1YPG?coHp(zJAe|(O|<pC#wubdu*_oWsDU;PE!1XKm<pjTD`FIURuQSNkLp)?+?
z%0#V|7aoY`s(^!@^z*1XSf|SakH)s44-YQA6rB3>w_a6-W6w*ywTN>ali8<U`S45@
zX1^tB)lSr`KI+>_@b!+p_|@zB)roW1dN#W2WA>WNjsJug@s~AnjI~pf_2hm4|NDlz
z;4i+xjOb4V;AP2kT`Iv5)FCI63yb%GgJ6wjn@%1jFRz`A-*yP~eonk6tVLrc@rQgj
zCLj1WCvl`CIH3tVQeOtK$GK^MdVxDqkxT#ff;-@SCa&UKf)$?bLstSfS(pn)uQO*D
zGtYr<4uOTYz$Nh7YT%iTQPf|Pz_-k$mE`^2{pjgy&pfLR1}g$bLA-eO9jtu^UkEXw
z`7+`wc&@yGj}B}TK`eOCh1}ek7|f_o-0e!t%pFSnB!)Iw$Ne)r{*?bqN#5r@^MG%z
zfhS71{__~;UIh2e{9Uq_-+vSxiuhKPcy{9y_~jV$9z6IXhVRdBy2eP!>k={7|3>qE
zJE+&dXz^OSXEj>VLa+u{^1?WN>qzD|*e1_VFy&ylQGRa`AIIDcR&j9L&G>)g*HS2P
zv=?!<D|}T4z7w%Gt`+#DC474ev^#K3b;d!~UQ@;?&RurKs3!1nte+-~oD3gv_A$S0
zQ9VApHW-a#*scQDs3f1A7v3&4dgWK<25b5>$8zN=a?cpF1&(D**Z?hQgnn9?oKhXG
zoII1S6!VHW`!O*-i!*)=n8qF$p<HLL9C3vBQ}PKhiTT)Sl~3)dNy<@!Rbc&o$0IYF
zIa{<edxxR@Z^+3y&Oz@4x67H=`KZ_47xU=_SY<KuyCrLUVI}4@_~lDo=6w_Rowlrh
z-e+ZdpR8b^RXkp-GWAXYwCZ$d#6Rf!fHTc@#Y_JUwc$FiJa9JG;cGg>gLe$XBbk)f
z-14Y39PcT=NB=bOXkT*p-c!T{6Y)O>zTJy%ogU@Zm3nS<|KO5k7v7j5F8ZbL#DHrK
ze{?G7hLineoVv8$$=+>F_N#U(c^ao~&vt03)uBI49h#rdp_LgN<X^iE|7+LehjzXA
zZr7#M4!tVnP@PT={Ok^eZgc3vGlwqZaH>^vr;Zs;<%ZX}iBBg(IT!nX;ypg=(w%JN
znn-XL*lP?J%u|(^My-{0Cv_Ye!P?>;_C)n4&m8!FYNi`cJgS-se234XKVAp_V6e?b
zukzo9&-oucxwU+1&ABN%pWgNp^!R=uPiFP2EBx^XFwQP=$HC>~g<XDq#lxBM8J>qP
zetk{=-<KM_iduM78g$EaXh|92kn0d{8Z%c~(lf_;tJMboIk+UIBN&-AxQjK}kG~J{
zTx0$YC1zBx@OpT{LpHEbFy7^+ymvJ`&AGq=*}x~6;it2Mk$7(_>-r_@cY_~{!0WT}
z+Bf~cJ*@FhmFR`c#JuEtdHJ8O0pOz?U~uA8U<vN?o>%!^l}6zmVvet6{hPo$xqt9o
zlA}4Z4q~&SO%*2xRR+(nW+t-UMhC%t4kC6<Ktm^|_1eX{K1fYOO>>&j^Z;>;b^dK8
zF^jc*p4ZPORy`yBwI~Jun~ims0$%7FJuI(SJH+7LH|TA;0GD+fJ^dj58`fRdy=V)&
z(Ve#N`}ltT`5YT!+Ae-?-XJvNTKvs#Gao0n52EJ{t}WqH_Nf^Km&tjY%iO<(#xt-V
z{`X>FhG%$V$Sa9LJW7|9oP!sAH+trey68nI4F&$|){!lE;Nd98^l)n#{eIy!=<RDl
z?(6AB&!?t`%Q*v=QxGoa9J<wuX!NW5ZnY$TZq9As$ASN8Z|FW;KvToO;%+E+grO(h
z;c;rg-_$fTjoUNL@T>JORNZH2`WizkS#NXj;&j^XQByq49_DAq9_UnXPlc#S0%GX}
zB=@CfU6=BrQIq@UZ$pc@N^jK<dIAE7GiBJH0o`K@^Ypw8|A#>@Ir%S^zt5n1uWk$u
zZ28}wI*nSPS+(W5DZSCc;eDRZ;B|Z8D#15tnP=0<XRVooFQ<Sdx1yI`21kA-pJXBz
zaSlf?7aZiplpeT0&hb5PZ0`l&7xG#FHS7dp{&I4_4IUdFNzP(qB$lrykNyDHq$|h!
z6b3JmXWDc7l`*LhJP){M$85BorasllPQAzP@85uaH0rPC<2`bf0ml^c=rbc4tdR%Y
zP_l$ahdG};{u;W(IUTbLf7EoebO-uhYq*!9hSDT6ln#6o4kq0Wrb)Yz^NOy$dIi2f
z#tP2o&gH}p&S?;MXE3jcImhea@!DiF^qKSZkoT<vE<V%MqeD^XWB9HDdU~|HE$0ng
zGoUh<E0FI~0IwBX&#zP-z2&@~Ov{|h2?i?V(Z6;6n}5Bg<Hb7T(YlWwWi3m8UUhi&
zZ1h3ohi}Z|)o^Yb=Z0TSPuwP6l_}&^S!&T~!oOY%k0kY3fx7Tfv*_RZ;8k*J*{kH0
za>S`zoW}zr!6K}gKj55o;G8-~SSLG(KTF`{7Lvoj2e;P|D_C14qNy`jbD8%LujZl?
z4~92t0dDAlr-)i_G?;4ZU1B8Oh6CX2H7U^;sK3L&p3@$Ya|+<8058wj>{rI;e%(nA
zrV2#!B>v>6Og;7wH5~EgVjOeuD|;Ehw-&s}{!d5X#kWx(!UrsaS4?;mzxFw_Jh-W=
z*La_sV7I&Mr$!yw;U)WHzVfrLK7RQu;FSDuy#>k58R4zY<AdV!j|cM{dH0H)zL2T#
zVAs)T`MjjW2x}2=PI=-ezyCb*xEFO?U*>Qj@XadbW$;*d=J`Cfj`+mfeMdY?cmh5B
zB=ubkoD6ZUJ@IV9I5<dfN~-~AM#J$6O$P5UJ`)SwZg_FqKn*g1N9<r5aLa1e#|^N|
zS8fx48UG<}=HP$3dch>%s^`4EUKDj-ID9AT>1K1da`4J<a7tNjSMm2J#vsPNW;_p`
z`5%w3;rUMBm^BT-E&PsfVrxaleK5`9_}|6-f9ir`ga^4Q81WxGk|p4Aa^po#33ej(
zPP|GS-NtcQN}d@;og0q-hI}xQ<650OGnSgYZ9@F+-~7tUd0lsqn6etpbqdEfgx>3t
z)Iurf1-*k;W*7Kl8GC3<foFk-d;<pG2sSBL4NqfHAASAosR3TMgGuIPK!eH(XOkDq
z1MV4A2)qK$=?agTt{nZ};FlD{@8|WwM~!`QGU6JrA6Xsvp{nqlCBQzp@E)gO-_@^P
zReg-77%%ODeZ<F=)P&=``eDMu*xIW$CFm1LLJqj;Q48v!oYX_f8+v3;>d}m2)VFZy
zAE-@(&boDB7;#;04SIw3mtLmtXik@ExwI^!OBJ6vr5HRk)0{fo*Qsq)o%)snjp>j>
zwZ}R%ytzYN3OaN>fkV$8+jaAbUCl1o>3y`T|8u+g#o3iAk3(r2@p_9xiRU=f?Hu>h
zI`s#Rr?=gyJZM=*UN}_|UMLXl>%nH1V$+aYI=I<~jod?is$J618ndCVcqnc>Gc+-u
zhwJ`%RD#%;V7Ev2?|EcF184-t{j;f8CXPdnHFzB0zW>H~6;vK9N{y^(9Mc%;re|<r
zsr>3&iuKzGe>iy~AGL9E>f+P;xzF1A%^KYJgZ>-VZ<<8${Y>yNZ`r?}ag=rW`8BVB
zbE#JtZlWX@n6=-nEZ(kicyKDx1Iqm=#DLlS{kRIar#rO>e9nLl<o>qstNc%g+|0Wa
z>`4TcTFLuHaNG0&zC?2S+LiRlP?MXka7_W`Vh_H9hk9<<Z8#@xSMj>|x_nY%J{O#G
zgZy=c^?ru<Q@$Dg3lrYU@nDOMV679(>zm-A$JA+W@u0o}M-UJ4#KBLcp$;QA&1bER
z>PIYK-PWSk>$8kCN?gdjmHf1VHMsP@+N2%pJfatzcx%>SEuJsRddo>HNza<&{XTpo
z_PwCiypK=%I=}lI-|Z;JA_jjO*rpqCH`jK)>sqw-MbwF-iO=L*3z)72*k%uLuJ{wU
zd^ia^u{dE~zusR5n~r2XRYa#kk2uA-EgZxgFW}V{YW}*jJ-WwyNR!c{R`4bV2OBaf
zqG5b@D;c~^9k9!)zHYr~;8vM@<ha~!{VL6V1Yn#oL2hM$%LyFh)|^T3H&fv;;dJ)E
zANDu9RWZF=jo#rcJAg-iu1gtevJXO4m-bac*KG~=1J|>NwN!|DVFS1I(YZ!kcC+W4
zTkpU>i^+@22axa1p>HrpSChvoGe3J%kL28q#!QZROkMx7x=#xRF;5rKcXgOr2_7fg
zFZArhe(m||)1B|cns*Ga-pJSRAfKp@p0f{6QgHfjyhwk@gWP7z4o+`GpH(=$g3RfG
z(adM&YkX|G%)A(Hp?050-M$9h6P)=4{2AYmHarQOc^-NMsR@5lXBGp$p6Je;Am0AU
z0EW4Sc1WJA$U2yCoxP}ti)G1O9ZoPO&-&Gq+MoycBjs$?CmLNHc&r{h4UO%D@2Mbs
z!h47M)bOg0LH=I|_f-h)#e?6E{)E2t8<eoS^wNap9qiIQmrGL|c>N5#e}i4ReBPza
z#o?I8xY>)xt*hXgdyFoe-_7^jTF5a!2Y2%kY%_ByeGg0Ay2N<QIKS8}6Sw8TI`zOm
z3Bayp@LA2wU??XTcpf}lg^zIAO{oF9c~ldgyI3cWd~h-u%fkugVPA^>c~lP0<|Ur0
zC-6zVeuHllf~7NfG>=%cw+8ovJhDylsO<^-Jk(pi|9X@?wO46Vf?vRKS$X`}PI~pW
zpjlFLmL+ce+3Hc!4IUL;hQ2z(qot$x%%OY_aMA#3(?b2x*{ET&j>V&tkL!MX#~YE3
zIFx~yau`hU*r!9=@K;{*r~+EqW!AtWzHj&+c*W;%-TTp3;k)vovzAR`=nFcXnYGZH
z<6eWgI|Q5<@2%em7M}@r|JvD51i1YdyrdNjpMaXOgki|SdsOB2v)fQ7Zl`nm)M+T2
z#ZW(TNdv|x{%1BEd*<=H2K{y>@40Cy|BD~(2K9z}q@mC;hB{Axi=W5y%MFF^Ff{oq
zIxt$pENb-XN9cRK;nOVUzj-b`)ZX~O$%iI#<KTuqC1^vA<aqdhdR47B_<t7brlDU;
z$AEus5w8QmcsuYl-9dLr23{xsH|>naC!F3)u*g*~`}-MSn0ertS<H3nvjfDhqW=T;
z5D!lfuQKufag*W6h)Kc3>Z1SRfg=X*hU56aoNorU`U*x!0<J2?I$Al6^)QouiCO4p
z^SMuKJI(vl{FnVl-%`^i;(Me5^CU;>B}R@0k2I_arUI+90mt+J(=1@rVeDhAl?2Ch
z;kgQ6n&n`b`*1buz%xB-ur7(yLEK-({rpv^EvP3J7K5)N{?<={-|8zpP?s5SZ^dWe
zZ8Q;&o1){DCNE^gcTb(3H3_-k4_;yP(pciGBgU^LtH3jp!Ja`}2e=TqD3M>SF8kDE
zEj$1@vrvCDC-6w4#OS}Ty=nos@(6w8JvExUFI*-y+$kG#m^r@VC)!Fc@cLMKxOu+!
zOR!1`w5~jGwcwmw;yi+D0vIpBHrZ={f5^|L!Am>AOAWv`$-p<SAEM!(_bM|nvfnbV
z2JwFF;OuL(@M?Q;;z%O)^SBP*x)!d3nDUDnV+L#U;sLnza6?1V8v3-4IB#<+7qLFo
zzj$Zx-K2s;SzOsAp+9-=pgX~@v>N5)T3YByCBPvc9s07<p_U$pZq;=tMGp9xuXeq?
zVb`h?c5ty>4UXA$;HF*sKG=0XBlxG5LzTlF`i*AQ2W)iwlS9=i;ll}cQe!!F|B+Ln
zXjRAY_}K8`6e6EoVDFr~V6UVn-Ab2}wbRGYbJk6_2Zq*B*Oh{gYBLDk8~m2|sz<5l
z@vVl3`XP1UtZ?d*1zz<$N1gPY0atdeD}L?q;Kc3hTXGwHketykzh4dNpf{37%R1@L
z9rqviKV>7H2V(c+({RLBsTHU>?y!EQGy3y)N@|fu)GFJkscP@W^Los$bHw>etbt_l
zbqcRPz`BSaujB@soMt_IPs-koV41ubsd>1cC^c*6C;0(i*X}!9-;w|8OUydF3Eg29
zpE;WSd&bjyF_rs#PV;?W9rDOQbQ|*p)*Pc1Smz}%?-TEzHvpa}KXp$rFby0|?yg`5
z7mrPVr(8l#+Xg3jgnEwoUk7fb=X<cmFX8~WKU-crcjeh90?cwZ7#*M5BPVg7{Csc?
zHO#V6=+WSsx}n62?s$#cqIt2#&X$4e=6?%kX5F%W6aN6ey@MBc3a+_J4!r`;bPCS&
z5WLbp_~)IhyN$$=HE2-tsJVy_t9-1-{=|xQe9wQl&oNp<ows%_p3%;J#bx9665iY4
z)SIP<^YkD6p5s-Lc4(35yefE!yfBeEy9HQ2JzDH3unqAyvKX-z?6GSd7zgeqQCqj-
zN>S(kaOn^p`QVK%^#)5fU=OB%%x>j*=~Ay-<l0Lv^fZ^6Ge3`m!<(gaD&1X&KCO4?
zK02%CH1j==Qzd)AVb*XdJ2`pDB$sMWcCn9(OG&T0^r{j1)^2iZc5(;ntqZww2>GWm
zdG8`~bPoE(XTMkFm!iuYV~-s;oC1Y>atE<b&rrP6E2w)OQUkrm8~@X%U0>mMz&3TC
z;4xwD?!V?!UXDrUv*>oGd`f--ziur3mEfJJ)Lw(|EOie?=OFI4qn0jyAFP~?*jAVR
zy*^-dYWekJ!8;yu0rR}(GUhvT&2A9GBiZ)~j=CaiqgEz-&0wZmZ+%(>t}65f?+p1T
z*CFCiJik1F{)H$sxQXm<Rnei`7c3fd*`(;GNWH8Zp{*zTYw6Pd@}q@^=XR*%6^BNx
zcBlio^{)O7J-Ou28?@?P1*vQDxb&5pstF#xJFlI(3UBN?;Z*v4)K{CFTD%<}-!Z4Q
z{Bp{b7c4}7!sW#-%}YwpCFk`|Ew`SNv-iWvg*SEUFg4rZ77V<6g?jOL8@Hx3cI#DV
zx5k2rk`Vv$R5o<71$np~xQv+i9o!Po+@owQJvvCdyi*Qd2aYB^xaNOP4P88Hu#W>g
z-3)HY|Ca{B5w51MlbS0LuYEj%Ub%f9<pP^@{0#2?<554b&C!#*U$jS&;F+<+t1ral
zX=v*!*Wn{viT*Xuqxa*`+ao=i2oG5uoRe+8NA)v!wbxFM=o2)&mgIQyLtpgS<zL{o
z#xa*S!mF2cYGABG#pgNHI?RC%;DEPu=+Q{K)`!`Zvy@#M-`n(jqfHHG*c33#raD%e
z&T~JQczv!P^SqZ+sk1s2yw9eX`c|dOW>tgt7FD@#(aS3qt>?V9J!eshvleYXX_4c&
zg=;ff<UV52I>vbJ@8LEzkLNyb(StaP%(5!cQLBy|v&s`=W$zm+`PQnn<E%P=-zwik
zn{xfJYr|dq3gFaX3E2MwZmT_O`w{u@aWb?HYWQDh(b>D;RSV=iOa+IAp#%Q|F70Wk
zeFgj$Q`qa_wnv$=dNqvyy?%(A;RW$OJ-vB@nWu0N5nbRa9K_C1Tq|V*Tqikw|4Hfw
zc%}BIi1|CvWr+QWdQ)4L1gC$4qX27_A@`bHa0cX*tZ8}A6nNCZtrJVpKS_=a&5QSs
z^*j&U{<{y~a{xJ+H9d{*P~w?atumn%Ou%!t9;}c6-y682a~%D)f%tPO!V$s2Oai|o
z2ImM|vjjX7zlX*p!Ix4yUIfp?RYoH#i)IBbipqx%fLL1}Jhq83CM$lq0PqmF=~4nb
z3-?&x)RK4CQ?tO)h7ea%)JGpE#6F<pnut&IC_aW4CvOCjvqn=x&xaSQSe0Y3kp4lm
zqPOJBpV456DP9#Zdlg0RPRkhh(k6y#l>+l7LW@7>)@L-~;rCffNetyB&K>LkrZ@w?
zy~3le@0i;hvl4iK3#|3Bw+MaQV560}eab@&uY_jR18kEOtWy^5W>{H1w=g;w+-~O7
za5~_duMg;lI*b0ocWKEn%|FJ=bsN3<(2PCsIQD%w4|`+r!Lg3|+TnjI!{f6K*MYfL
zRmUg!(ydC=I`fDJsqo(9!&iK0CLWkxc!%kOy7AYkU9nD0UgOl00Zxr?O^nIwRJSK^
zHtQTJ5JhY%@6eAV4(7jI3%A%+b){WX*4VXYyIr{t!{J=DYaAR-=K>BT>ElosoKFIL
zIA32oG_o*xgB)>Y0X^imotluHdaaX-K6{s1Q`5Z-bStY9zL`96JGr6TaLF4cGcI$C
z(|WWCZM=blW4YR+Vdp(^5Kqo!$D>Gnyd((BI2~SVKR&<b;5~G~g3WwfJB_+{8Gf7#
zXx*QvgNffq%EI3_<GlB$AJy;I2;zLBIbf3I@FQE{5q9HYI0Pq39{9;^9BaeAm!5Jk
z&<f^9BZg%&9PCbdTEO=|!8y;t9=&hy-f*_VAHhL@lTSSbyF3RoG3J740(h(m&y@n(
ztOwhq;(z^Mo5-2?N=H+x4Ftp3cx(VV#0YquMZ9h`pTTDr;d`Xu`#4#j$M}rIykGDa
zcxSMAV(Rq#IoKOMFL)iia0mQ;xHt7VwM7o*{YvVScn@!CYMkZZkiuW62f-$P(U%q#
zheKsOEop<!)SG&zKROh(&sR9sBwgThJHYX_fVZiG7Da93Wz8Le8-13M`|<S>^}~pl
zXmI!0lj|yY<t&`jF|ZfBO>#K%w`=hUuE2*4wy8ti^Y19MGaI!i-!HrY-kSpa-w)Or
zxN`F}G~o^$g8(?}^Y9wv{i)P!bMJaJb_VksjBqQBSLTcKzK|Q*fOnm#slB)1EyE9d
zAS)atT11MO_-#Vm+E|r+2QIo)7e9NK1l$&KszWMp`w4W`%ML~S>(JJl4vpRC(7T}y
zo&U$79np5ZnQVs>u&ehSa63HZx+V@aW&YO(|0HAXZaL@Fjz3Pd>x`ES&LItRcxN5f
z518jg5xnx`z;r**w3w?+sM$Y~n{J>>bPr}PDc1jx`Sb!%r>r;)w^I#oQ3zTzF>&t>
z);~4W$PYek2Upzv_dma1_jAN-Mv8;%Gq?wgL;PM({4BiAr;Jw_yuMsUbS-M&N^n8>
zh*=}+f|rQZ=Qw_Mv-#BfgjZY11xb3aPXzI(<UK>>j~JS9fcknqbvE_&5o+%T7wHWn
zX4v5aj-@02e!~NEmDtgVm<G>Pk9A$KHhHVPNps5eRp8Vxt(+LDz)>L@)+j`mR|jkU
z!eE^b>94Cl>>82Ip*ATTYLv*K)YLs~z&x#AI#hPNQ@1!LIjO0Ne|M_UzfR7bQzN6n
zH~9JzO>nAUf3PN4s8LI&UNdS@8~sCVbO&wwWObJ!LgCepxHKBB=fNJA7EI&#^Vko1
zHk%P!D&ZLnYe(G{i-!=rTQQ}fJ(Z}Nnu1k`EfEcf&BVp^afU*Qcr>aeJX2Y?6!1-s
zB=j;oqAom$pB&!dCq8o<arBduzD{bclCS7R&EwIcmh25MlD!4sXQ~i;TRsHy-1VsQ
zA$Xqc9?j<Q-(Z>@%i(f3?mdZJ&*5%-vpw1gZ<D(x8W|kS!1lbB8t&u6c)l7wkQhBT
z&LtD^*~V{r1Akb7<8_H1jz?hd$Z-zc=?wnP;?T_(b`9QYSJ|<4b?QyMS>De61$Ncg
zZqvzWHm+e}Q%EEjr;|-HOW5T9YSne{NAa0f74BojuWHqxWLEusV`0w`3%Zg;MZh)p
zPgpb>{Ne)3Y(8vJ>_H2)tVKKbTU2PDMZ5M`RDQ2T&3HV~8H;M)vMApxi&p)#=t)|u
z+7`0vYz?b&cC#wC*Q%{^t;!7Un$9|F$@#l~(uGeKeht2U@+Xe%F8mbL*oO^WuiGPZ
zK%b%JG4#HCaO&)Ehg{tpiu007);eT;;ZQ2{00;Bo1U)SM=-;)xb*j|=T)HXzqIf$i
zF8~W~rO%f-=1B&(qNwQ?fqk>U=T!jf?&W;u;Jn`1!Cq^d4fUC2Xmt_hM@hHJK6Gh4
z_4h{R;wEz7vm{P+%8m9w?@ag(r>arYTv<atjU(37B}c(~6?$!Gxfd@w=R09dum0E2
zuNdO-G5GA&a64_@pc}y%g%Nw#r=w4i_`9Al20ncrkClM;`2=Pu{GIsuh}RLb&%<?x
z*T9PnPK&EgZM>4&qbd8a<L&DH2F>&X^QEhyKY0wUg=DY?Har}2>L_!p@=^AJDdADx
z3(Q>)ylI4s{WzRDUD%<dm+jg;!OmW=c8w&57b0)_zuC3mqn*7>>{{ostH^vio-(_Z
z{vsysb>iJX>nTjkbGziCb}aMWt#uPw3yoPX2jEi@qqEkhr(%JZSOmWLfF~Ag-Y^xt
z;KZPm;F-%+sMjjd7amA`Mt#?U$1kOXrw8BM0^d|5CRW0W5VQoY9!>jsxL0YwH_=7C
zn(>`lg4l1NK8WatW?TRdz)osf>XqgE{&Sb{w~TPBB(=^C^q>sPuZ-iULEE}CBZo_6
zUonpkQ#;LaDkj3IY_*(Ph3{r9wZa#8m_fw(7Wiv2-L|X4HoJZ-<o9!oTF<g8;Uc^4
zZXhQdx2xy_yS^lKXgBltZ%^{ZeEc`}9qh60)Zh+Ied7HaUw7(FX8N?+<6E9e{JG6s
zhmW2X?$#qX=Y$F1Lu(qUN6pmv1pU3lpr=(l`W6P~kN4s;yx1eo@2O01H&wxRJ;5{+
z@nD1f?mcAhlbraw;WvLr`Sfm{Pcg^XUx-?_St`HWMc{D6zD4BHXcGfmu@NkP6^t`w
zDYt9DD_|MR7WR~74B-A*a7HqoTL$Ln&WPvt8h}l<f`uxBJ?c{vw1sDhfX810R^AB)
zDfkeL08ZyF_X|JxPlLAHq-Ka^p8+t@xE=q!Ul{l*mk)j@oSq75aBElQv6&j2?>2%O
zWDc0;6yMJ>gZCc^XEX+WXF7eci`lbd9a`Id_AlUlUw+}cQ?pMj%)VmqIn{cxSC#`$
z4r}BZT<U<8=v3(VNzda2f@|r9W);tGDk0iLYJB=x*ykZX*Ssl($Bp{oE?j68FwJYm
za)!G!eVB#eds&l%S(88czabw7=>9W2KfK@MSZW7!u~<H1A-v7>4b&5>&_foW&&>qO
zj01NKWZ!Nxet{l%a_X|~^RSoZXY`cg@C{SpAvh<Mz@XXSDbj&&(iJ1`<EgqkmEP)>
zUL{Dy9(-p!x{o(Bl(jqNHrgv1Nfz?esf6^#;J27Dz^#)tsWrB+|3HvSD>j3tGl6T+
zOZI_fK2IV(j^=iQLs17DTqnYzw&4zaOb?D>O}8)@BkVfA*RGR~nA63mkq0oBGdPtN
zeKlmcQ%$J@Eac^4SDYICgu(qWjmf9P)R80M7l_%BgXqIrPtR39L#N&Bw{wda_&+$O
z2ypqo9>v!7D)D45TqgS__r{Y)?8>~Knr9Q-@ECeDgTQvxee%Ac#@*^wEjZ%3@U`o%
zz{^f#uO4`mp-vCI*yy%!Q|0l=jT#A0($B5|QC6-oYSFwPv)Z1C(lTG9rk{?`&}<Qk
zo!?*D8m{F#`l`&bFg*(m)AsM7+IJun-8e+4?uKaj{SX!7{@m{&+L|O(rQ<?$ZEJ{b
zI78?&3sL9J!75rHScOstt3|3{4bB*>hIxW@tz@u@R|wY5uEF{Rm)jkl_s3_uEX2)S
z{QZKF|Gk~-{}4lmaUK>r)qb5*@8N9PZg9#*Y?$Vw2Qw7BQ<r0%&#9wXoZ7>&Z%kf)
zl-{Yk#1U&Fr<O#bcMo)`KCxu|Pp77SbL!)Lr!vPlRcJNuy9z#tv6}a763Jc{<JteB
z0=SO4v|3|^-~`HoZSWmOXJ8)zeCfWYZWX=_??<d$dXYW{c%OgYzzIEf!+pXBRmJDU
zeB1lPP+WHU<XXTVf^S}I#6xr*FV1y%n@eb0TZqL|=?NR)(Io?{<HOU#{W0**6Br#C
zwZXr+!8Nt2qHn>od+VSL5&PDg;II3D)9<r4>KvD9z*`*xv*iQ>-lFC#_tzmCbz`e`
zVDJCIlj9Rex)DBo4BSi~yW%R^H9nnP$9LLv1bj0ZeDfjPrm$8v^(bItj~pxeyI3`H
zmQ{&+Sv8`xRp}F3{~w!dxM@);xSOh{EV>1@IdjaS<lvk1;G3r4m{kYBK;WCH;F^oO
zEt<`!&EL7eLbu?3);_f;&3B9Xrnag;e%`aHRcX6exyFT6H6~g0bGemXIIG@7TlMOo
zRTs}$Wxi=u>4Y|w$KNooH+zw+fTv?V-vpyf9_p0)f<yjhc071CuKR9PRDhLUMvG<?
zvMAF<voh~8EBQ*ZX3a3`@DQ`YO=eAMVpih{W<?e^>v$fsY?;mMM`2d!#AZeOHtEt=
zlPZ5SDd#tn?j<*Cdv&w+wKMC5#jKtS%qnBIs0?c;q^?bEhzA|J+7<K8rZIhOaDX-q
zTyNDMzK0uJ*gqwFRXVHU3V`W4Th-md->>+ck8N`N&#uSl9>2(e`JUp}AP!|Bx2E{w
zRLx4{yq+$#ABpdEG2Ajf)B@MIf0w;;!B?ZvR|;_3?<Vy>x%%LEFnl+c8d4LCInRD9
zGvGDBeg0k!Iay2H;SgSaacIOz2R&&HJ#U3iHiJW(JK1$3sa@NzfNl5KWP`Iyd(5U<
zS8eL@!KP}J?HbdTJYLVPhs>Sox$R26*`|V{Y??IIrXlNWI?(`K=AK<m9^y-3y@qps
zO5<@X(h+Qz#iL#8!R&G9q~O!d)S^#!!Sh~(*Lp<{&<{AaB=ktZZ9M?v41&9vTLQ06
ze(+5epYHSAz0dp)_@+Jh=Ib_UzeV^>hO)OE_-0{Ka1Qn1p7$PbH{8uq_8MzQkI8#O
zXNS=jUJ@NOG3)X${Yk|BzjfWJ|C_z`wo~g2bg6y=^ly$?rQ=S`A5IO`oPH=UYkM%s
zO|bF^c$<UO@e0G+q>d%_AK<snwW}C8;Fi~}3Sd3Q5O~L_a5t;%TDHTkC_Kl*Q#mxH
zGRGXu{X5#BCSM)$1>#*coSMz~DZ*#gXoUx8B3K(dys9vMK<aK2x_Rhtw^9>J#@P&Q
z*@wsBBfV6G*wdsn8uw7@oIM_$eL&3vf0&^HTxM5zoDu)+1N4G?7?+Mc5vh&0QX7w$
zh=(D@rz6zE2b18{%#Zi78h(~m^ihSfZw4HFc5uUN_=A7vvv=BZ^hhvH;CeU`ZZE9B
z4+Ex&K|{(3#%RcGr3K&~FizA~_5;`h=XV&-9UR0RG^`a@@O#4Lq`8H*h3;~S@rlP;
zT}8VB19^Dej05l$V3R=JFJv&+x(3g@>IHUc&0b=Bw~dv+F%{vI8lX+}^lM}gdoh(m
zAFIrD0UEJaK{q%?3;t8E(I;?M0^YCEU9{rgXkYL-*Gi+gf*-zhKnH@ejK2nQL->=w
z8^IP}kW4qxXW??ve8P|Ohn{unhgEPg=g@U`6{4nq*I5moNsk_N9&WV(BP+V^8~D=c
z;Gnovc-yEomZEu0d4djhn>}<dfv1jvo8WHRp-DG}yGgZ_9<sT3fA9rm9D-);#8=!G
z&rL_ZQ%&@1zU!RVa9eOwufcF#I)ic2bDab5&5j{H<tT!W?H<1Rv8)$(x}C{5$6(tm
z7LU^Ag*S#<*x)eKusE8-1GhfRfJa4>$c*k(jGVS?4s}16Wa)DUT7yGJ!Rf6-nU~ap
z3(XE)o$gSy&!ONdJci!7#%I^;5WBW8+PhijN9<}xem>b4{(mVrs59?3+Nl}K$iIu7
z>e14vR73F0k98_dX?TXlPE{ZWowDFDTgpgCj7|#|pU17~8Qt_A;DbPieM%fZcNK25
z3cjAxF2$t6JMx!nouHMqqvmL4V}FL(>=6YgG;ugJ$X%!YO%0y`)~Ysx@Atr_;<t&R
z%cE3lV}uUm>90S<!(|=cPg8#P)$CnhmyCUt`C^#DCx<D}6sEcv!gS<MDDyB>DNcr>
zSA{BJp-`2N4pAa^i27CtQO0B;`mqCSGBsFZTeI6hieTk?6r^kSgV2eCRN~(t9eo-k
z=c^z^e+bg$??K9#C|JwV1anQ&V9hNYteJtqs#Q5y$==%Kd1KdycXnmvzQO&rukHHh
znO$|jjltAGX{eRT@z_dmP@xvYWs5^W=-J~qSKrVE7teDj5pnu6=VxkhhqmOTzJedl
z#Q(qKd`-JcT?MCH3f=1~d1LJqC;Ln}6*kVPyxiXd-?5JKUmyg|qrFoNyE#?ZjJL2f
z`D2PpvvRWcBeh?ZREFB(pL+b*t*JM`GMDfw!^dwr?^ZN_AA1PL@XD=NUUN7*{hX~0
zWdgU}rRIMH_wa*S^6^~wW$M`I<M@(Sdh~P*k9p9jz@Sy!aAsigHGVujjP;BM{n4E2
zdh|5|dX)#hA1&bVDtbR6@lAtomV<B7ftMrtxzwj5*SVqg>{E-pNWVra=55Nh4tliU
z8kr9b>f1FMZsgc-n~LPJ>HPq(M0Rp6c*FG7qB(`FN;1u=&?i<6cxctFT~<|^Kt2Yy
zeAtE_#qH?R;016>GIXb4u+46;&2Yv({<eZ^ihyO3@35%oc8jiWwJ2_jMGZEhXYv2d
z;c`Cm|Fa)h)aw)cO;W2S<g%(;1*>{@vg)SY%C$MH=nm*$8?D;E)2eVdq-|%cs(#(7
zVSlY^mewX;Qk!P~v1%aS@$NONUf-}PhUWraTeWx@kGHWZ?=_1`wX~>O0N?wpSv8~0
zs=LChLDS7DISf1#Y1a9!W)&@OR(|kJe(+6XCbQP2G^<l0p8I7|cJPhmy-D9*nxs1>
zJ-K31;02Rjo-(m7k4f)#nUrR`N$xe=pJmdTktTXSOnTA7r0_;2)q5YM*qKrK6A`6k
z<)f50Q<Mh0iPQjhBzyWrYEr&PJ$)Xbi5DVt@nD2D?Tt{%-4Xf`6CwYD2t7&{sr#KG
zmDv%gpJOBC-W;iSmm}pZ8l{WNqBQV9ln$mhq4}Bgi0{`qy;<Gso3&=X84syNGdZsb
zOInp|l~qH?wOv>rXNYH;BOQvV<500Q4tPuA*>pT3;Fr$KjnC(8+6qR@zS{;~wQ)@y
z8`m(gsqanxeq`esHZ~<bNnDAs$;3SRVrMQnz{FmgqObFRHQ;m_I8?$950b^j9$$DC
zv%ypE!k-1kU-S_^9j@WL#j992n96uX>fFI&MP1hZJ>J-)^fagET7K|07Yd-Mz(M@T
zK%D=}Uhq7>mOAhENn-qFuky{K4{<PB8Th6M{LSbea5lB!i0~uW-oa5_G&F)bBU^WP
z=B$QB6Z@xa#a|ikxy%KRa|i#xI+sQT(?69KZF?uUgBb7#O!FJ=(|+2nl*R0-`w+Z;
z%BI*eHf20w(;{%)_W3p$#McJk*?{VHZ3g!ZY+={C&UQt1<9~YLGbT5zId9j4WDf1G
z=TP||G_E1k%QMghmY{DX1mg^X@6YE_!7(oOesgJI7JS7W-FiF6ty$Dk*QvF}Md5+t
z_$8r-%8#$F8Cv;EVnxh$@XbwhYcS&>w7I$My)rl+A*=9UUt=E(IIl5yQyWni8>7%v
z;qtBb&}HNBap%BuP=UQfnzAQ5nBqSCNTNZ0RT=vqb{PehNrOLO52Gt%7+9t-8q&v!
zXt!`D<9OY1xCR%mbFbvR;CBv0<7Yr`Y6dUU^c3Ejli(r7C&o6oiTK{2eaG-A^Y;Px
zo@MYtalCI$-p>Hrtm}vdTpe6f0<4l5jVuwKpU><$_KN+H-qYuh7#x)ik8>%wquTIE
zZNW{EXk4SfSiH~JUHqQQXwG1TGWd^Af*U&Hb8gm<eRX<*W8B~f_>)}A=w&|$|B04V
z8_skx-Wl6ZxRu0kGI&bIX9vF&U_X<h478}9CD~sBtaGmjJPw}G-?`|wqW9)gQhaaU
z=+{J>YW46x5C1rPH^UBqm$uQP1GcHoXH`R&DmE4W4IEP2LG*#y!Bt`C(cqh8@UFS=
zz{S0w*ZBy?kz*0viQkqMAJS>o!XP-ef<9eDla1etdkt&nYXW*u;9U=lrq?A8JuLVs
zt3?><l^3or)~yK+_P0q#T>&S!9Q<JphQDS`@1dsIv&^A{CgwM5G?W-nn|xZB8n{fL
z18vcc|Hm#@C-8f1bQMNGJ8GIPc0ID$bzu;7*d%glIGSvzU6-xsv6t-1RnnnY!=V{F
z9LkT@)%P2EES%@FQmiNHwcd;rN#Rw%MHkw@P2gv14A%eI$Dw4r@0Zgytsh4nJIJEC
zU(I^m#;h@;OiG#Cq}eN@^yHr??Kv2!$-$9ImmyNi&P8bHr3ihT8lk64BV_YNsLSB~
z`rrx2XV_1LYxI+ET3<bF-B-Q8g=zfBF!|<$>6$A{gEEK7em7L66QLTuBUJCbp_-OG
zlzz()Wiy57e2EaH`x&f~>w~%eQLvh22-epZLG--@sRj6EFXJFsXFmAm!pk59z7JC9
zSMbi?AnK!FRm&W#yl^^gO9d-a(;(G<30H%5@*X@hfcpuk&$m2*7X#O10M`UxvnvV5
zaN}8UB=Kg`DPE61BoIt_BnAw`sC>|_UN`B{<h8jk@*X@NvyRv_&#u(O?!26n<m2Gg
zM-zKD!&~tBXwGNqg${j(=Sg10sTMuKJskgQ=!>NSI5(VwgZ1FssiDJ(f#ZoaNztwz
zet<g$$7H__hjM{&idqni^Xm$_#vOXk;dR!5f4bmzo<>~HQp3<2u=TiDdgzExCb+qA
zCE;g?!&^InzeB(`COjr)G*m18L2jpa;JP4<n7jY5k4G_d1pcSoAvkK@r+Oj0CfnfF
zyYoGW(XFWCn^D()4tMEDCgS5<YJcX{54>S9bEwzBH$DH@)o2Ryvb0?*p4hY*u4dr>
zZ0yBvRqI4n<*9B_{ls9G2WH(o$^3%rxUt{NwHW!kJUMb4IujVB1o-7LJWV>V%p!Q2
zI$)TttI(t7TXcJlML`V9TzohDonfIxYZk%fEViiSG7EZ)Mdza}3dE=L`!=8T!J@Dv
zR^7wX<0xm<%eH7>cyay*zHu+IYA)PI2>7M~_~z6pt9o9wf>+^tzFT$pGml@kYCX7T
z_;IVM;^#@l__Nlk)FXMmsa22A&jMTU-7{Kr^t4&&!8aS1fnlbam28Ar7r{3-!8eD?
znKdG>S;MoLl@NSTEQML`5}H*Vd^6&UiR(0&<auV&zFQ`I875snXHv_PCY3#6Qpp`A
z)mdlK>p3R<Wj*%rnp7>!q;qvm8kN?h_9vs{o))FzrJ^)4X_N|GiIjhFq*j?C@liyo
zMbb#sI2oa3tkcala7|C{w~Wx5>WoSeI>`9bJVF_}MM!-j<cN&WwLuY@vLpgOJnQ>>
zgqppGQ2DfxvQ~}MoVJn5=!(>$`H`BMCrT~GM#)mhqzj>Defq)iX=GQA>vl!_fG361
z>-Ld8pRINkgnLMs*{-q7jdh!B%D)Ou2Ho=X0h=C>)5l!4DH5Dho7`^S0=`^i)3Sj!
zbp+eg?t^Emmra?v!t>zYx(f~pbFmKLqc*g4YHT!|;ykx5!%d%HF8?zQ&tWNg^YHLg
zoy~|rr@D@}06pakeli2Sssnj>W=5`c43|?H4#J(#r+V<yvw7VI@X)E9UM)kL8Z`mG
z@&G(IA?*E*E;Xs5hii$mo>*sz&(KG*(9i-a{j3!YJxc-i3eUQ4GuOVfvsS4gQ(kr{
z1$E(pYV<KYbE+jZ=@{0|hw<o|)I>#+(;GDv%ob?VxARs_8U%hutMx}(wQoFHEx0jd
zdK<n6o4$hmYJarJe9xw=ufT)_?W)(yt`i&V`udkxRf*%#k@^>GS=i>#MmKfX1c&le
zbh76d9>PBOBj%#*!O!)fccoV|w+8J4<Di9p?qTTuD*Sxt(lfFV4;s@y3a{(iz#jUs
z{9b<heEd+|=@HNCqaSlA`%rM4-s73dR+@cDyW!nL=coqvZ(M=@|BhZ20DdV0u0YSM
z3}5iJKiI&{b)1L8uYh3|%mLRdB<F%(Cc@`Todwr22`vharzu=Zg#qYC=t<kS%{s!b
z0z7w!|EUh|vkxy@mYwi3d%;X_Dhp!3Dd3r=V4wJFbW8__tjD`E7Eb6Hd{JF5ew$!=
zqMPH{`3DRIw^Sq<{LEK+^yvAzeII=WUCsHIy&utS8We}qsfiv2&tvTe$1?;hz-PDR
zGs|9t!vQO}@U=AvfJ1@9$ySd(_s)3G(1hNOVDFjv1LT8W*>;Y{@yz6Y4bK8jiPwGW
zr5#AWAO8M4^i+Xm0>C(Pz%?b%q&D#PB)FRgJl~kt`QvymxSw@!H)C(pJI4Dhg1f2A
zd)M4d-}-9$%$A@*&0)V4dK0#eqEC4sIx^my+~6C3dwRPnu!my$cwaKdfa9{C<68i}
zDmpb-a|OI*KRhO^h5p0vi+Ay||34bdSA1+S9<Dt=&6}ItfOftul-`U0dN@w7_d#p?
zTl3g=@)$a^;Z)+|%;&9m#zx`!=m%$u#@g*SH9y>AI@WTr-Zrk6Wm6iPO`8VUboqo$
ziQx5r?qaP|=j?&U-OKuKe+?W$Jjg`d^NiP|`ef6Kk$4`+&v{BXG&iF|Lz=-4+41$j
z4R4!7jj+O@=4%}4vCg5r=wdxqJCp<NsP<E8Z9BLr6F!OjHvEcKEq7V;v#*)!`I@vK
zmkG=orJKagl1C%83-7L{K%{P+jZm%72)RD?SAvE8)j6`i@>lM!1#7}J^L0N3NA%P2
z9Q`zBOkb^N*;fNThH386Fx8$NrY%O8QY8x0r_-UTw+BqKJXFoCp<0$TRO9x9Xrd)V
z`-+9A^p{}W#6wfFZLlV!4_4vlLG=9xY3RKmjR)UU1mBbe-&6qKWPcZ=sh@*X>1U9F
z<AU^*=a(!B(wNRc`u?Dg_VYV;#^V|4%QWmC(Gh;TB)BF5ELr^um;?{s7jR7tFiWPR
zcHIWg1n&9IOLAx}yx9usq-E$WOYCa7+^)3Tn!z>er`nZytX+qO;r$zE*HvP85F-ua
z+W_jS$#@A;5@$FEq2SGC<?-|6g1<@Sq`w8f6{B1tIGk4aZ1Ha7K_AHc-Nl~7^kMD5
z!%H1FdX`&#Z0!4prnP1Po^?O_LvEz^6wT@Y7^qApe97?3r@<x9FVQpi63<l{_U$Rc
zo&`0~quRpZ^yE5zeenvy(^Nr|JkyK5DIQ03SNcGF5zOE6iP5k&p<}`Y4a^Ns&1XMo
z4|ff|nSeg=WC_|tUze`Ibw^KtBP{6D<2(4Z(L45n&7;5DHDdxXwkY)_x$0m)o4Ue(
zH0^2CJhw$li@|k#H0$LBvzqTUtJ^xWqS10z9WcvCZ_$AvU<7bXY5X*)51~UbW}-LE
zgrBLu!XocHi~h{UXAHh+2)@Y%zIl!|R0@3ag0YQpk-w8N!uj9PH{p8l;7kBpt;lMn
zm(;2~t*wd#ZzY=ohqKVCqhOUqJK$vwSk)YSQz_P}C3meV^8l^~zUIS4UVGZA$wzp6
zA9x5Xlnq=o5)Q~(&C0c3EJ_e;;Tl*L^*V3Xp`B*!UToH=DQHsQo6_NC_RC-tGphsm
z=5<!HDyKuIMThbxFe}p!lQyGG-G6P;hbL%N;G2_p<oljADbI0}lEj#_d>hzjwMi{z
znsj5BN&nhS>ebbxwiQj<pU9*Y2cmQXjPVS<B_J+R*Uv=i*UU)PV5EjNh}5@Kk;;G`
zY?=&L6BVHkog#Dx|GlMZ1U;b<_*f(Kyh?;hC_??(L}(c6I)f3RLJK1ly#w3>z9|mA
zS(`OdnW{vpD)?qDTu=4&k?a8<sUDUnom?BG?tvzSE;4E91hb}a48~kCD{)HBQ4fnI
z%Bo*)teR8LrVah!BG8)rbHF((Z1k^?*WqgVowF$r{$dxJ)yyR}Ee)~JBZ?;1&_=(y
zO>gVkl%%>%oif=pF%^1SaWIeHM$K*4A7a@`;_DT7<;pvps&R!LA^ZWcWAIIZ+4m=7
z4?b|oJ?4L2a7xQ@jE!8&4{j$D8q`ki=UEN*!LM@)Jkz)%duUYls%SR&pP%#--uLM4
zF^`r-($C2I-Pw$OyNuquMTSOBW3SIXcy9hNR1U9h`8(*!;G3@&>f17Si_dfICm$Sm
zAlJaWjXvEQPyR1(NLPHp3+?)T)TVTKZMp)NxG)4vgGcOdS*!9Fv#KU@HnzJ}7K>F|
z3ftriva$CD1HEnq{LZu&9ILv-Dc<K6>o1Oa%G4a5IRp+Zf}V^1=#VC2|5S1hF+XmD
zLxW>^4czPS4d{CL>1o6d)$b=U1@G4NDeTLIM@rzg_Z8r4h#5^6;C-Zy|Md!QQ+Bwl
z#`q?kU_OpjEjX@3=y|K?Kg-e)Z{|RHaPSz<!ecz~olmvk2UZq^OR0sA6`bJcgLl_T
zf9)Wy;|zzeZz}zVV1}mi{8~O6FAjev<MDxDn#ACsy>|S+Ht>oQt^thW9Zk<RSY<kx
zXw+(Yxi`T@>_D5^?N^e0>{SQW3EqR2#ACnUZL+TQs~4~RgC3KN_i?~kEUoU>(4=4?
zeEcRjn-vf7MxVhKItlJ3np}^MGzUJ?#985ts(^8tu{R;OC^Mh`aR&WY>)>{d!5=-w
z%lro{mJwfX3HFt#{okHEl_J>(1AXZ6EcT$;iq9CXWycNnOL$J7_GfgZzhI)212qM2
z{sB0e+4yjx(5cpfYn<HAfTy@S+|Dbwo20z1$#3?IgZpXs2#yA=t1G_Zb4S>xV>cd~
z4eX`49Q?C@-uY?lUkG>O^3c290bZv(7>8rnxfK1KOYqf}!y6k19z}aCahUnPjO&Vd
zxPE1Cyts9|a+Ignh+&jv-;g}?mT<oFZJ~Dp&VP6=u9eB059<QHxeM<%n!cH2^bD7!
zuY01CYgOX6fyZAs+(BPE`Y5@!tr=fk06iwF(b4;|=2^eFE7;TmO?@`&x+rza;L$d{
zoJ0)_z7Jn+)4y=KZ@@a+=ttSNlC`}a40Oq+34d+MQWj0Rjh(q-*Iw#^p7?Lqvp2#+
z*3m8pyiON1*`|2V%7d-4J9G#>V<mh;M(VAtPpOCC<YP0!yZ5x{VT4&p@dNzKZsM8)
zQM#T#N?mq>Yv68*=Z#d8U2xmP#>27wRRzxGMC1N?{~}x&;BJatgS!dqr)^pLf#v$@
zN4LIuel|?^mV~KY%P@^A7N(dtp-OTrRJ%8YYVwRw_F@a=+UTJgxHp9QF+`IKhHwp%
zV4c7p9N8>bRq)oVdKSdBJAyRmZjfq#U%GPt6nsuBI#mpR59aTSj61RDS6hR$DlABP
z(?>eiN0HzfD>~8)_?m)Xn+af>4PYDNDqIb?aK<^i+Q8Yo2HSLncWJs8|HuxwnoV|f
zL+7plw#l#<J{V6_mRVq*X?7)^2;UA@;{@Ayz&4Y>G`*elH8{XZR%%#o^YHjg@N0>s
z@I%zGM>!u2!JWMsgS(?|)r8;4kEWF#PZfCgXc)fd3*b>ah)4gAq;m|9^L?9eZQH4B
z>({AmGg;ed(x$d;+qP}nwr%%4`~E*X$0nO(v$!$W%v^K-#UJ=9N?+nqYbC-<g?}}{
zV6xe%!)C_AuR4;mXwuVrw+O70g$!#m+~Eee8@yF{;=|E{S=?|liSXrQ=*ZcAa5!&R
zuj<-j)VK@YoTlUh!tcB;ga<h@UV=Z}^<pKs28@0X%#;UwRRWy#CQCPsd=#x6_=^jX
zg<3LqwC1kBTTKo6m9;W4y4{-=;OIxZH@-++Ne=gQHl)Nh#w1!!FSH{U8qV@`>~pXE
z+IPgS6dU~dGvBX8GyQU`@axxIzml~Nz_SLF96!xxIGpl(14;+h>9#hYNnn{P;2IzJ
zrWV%^z8(eMIR=*52#52I>m$Fj4ct?GNI?4s1yqQ~NO}%l=SM)F(gk&yy&@cZ6Ar#f
zgFd!#Hk#_PAeoHZ?g*;zetNgyX}aOVd3PzO9&kAi&j$689;^%CpVr`@w@dIQ55iAY
zGpNZa;dOY9S9%4MyF@_ouJgL!ZyvF?&mZg86!fXXHorEb>?FwVm$GvUzFCvpuY>XZ
z>i^fLQQv%O48AD~z8Uk-r=Qn-T8}ohdMCJLsgHFFeg|ALY_d<w#=-OS^Qo)Pr=HDy
zDpeW|C#g>j4tSLTe3Q4jS6#k(=x_C?5%{JQ{7sS?9!-ts(TR0%t3%w{WpXQ18@z8d
z+?>_!R^zg6)hX>(YB(HwRkwOK0_(u<l!$iAI}7e+qg%<qMh~C5wGDg|AO0q7D~~R8
z=dst)r^x@xWB2OtGB4*hdljpQPbLS?4?Qi@6Q6p(Uv2q@J~swk671R@FI8MPwOZxi
z^M(hN2+a5+Z%E@NhIAD!CmX!YlbIo%>mAaJW+Ba|-pj^%n5kq)OA3cHIT4<4FthtT
zx+3~!HuS`<d`4IKjJFJC9P~uv<jOk^4*C(iUvyAW)4;KOc8|cTXDZ?UVqDgwp4hf3
zN_X+5>`g@vE7)UB%V<^cMXMni&U(1iE{B+3cyVLi<4a;Ke~$*1yAqsf1e__{Ob_Nq
zclgy#&%4PC-`e(VH`(wyo};!cjIX7Ux107Ap_dikQdj_gah_;JUBt8ZjNb3T)H>8t
zb?|Z(1?O4MqwS3YH@65X;!i;Tr4Q;mKC~F#tC~p|e_4aN6%$kg-mg+%@Jqaw3!lN)
znL}!r2hGnLsbU{lqo{AI)W`21PQ8InWq_N%+On(i{L|grP}7+r^>H~oG`vo_%e)6`
zqP4<`{tZXmwiwwOJ-X}A_U>Bvh+GGH8(eTWnc3gs&BSl~ua<v-S3DPcVgww{Sn|`h
z)9?12Osz1^Y;Mlk={@jn&ZCzEk8$niXjbqet#X15zzGxKP0n}5kK+M*bjRa5nEs?u
z@CoC<3usa!$J0v=wn-P=ORGEJ(*YBe<?ot-M`G0?Kc^YGQA97z>_wjx+)smfXirOe
z=_)+V)aCSPEv3&Jyz>G~^9_E-hF4|^x8<hs{|3;DYa;84=hK~jm}GDkYe%EUjG`a6
zA6_AJ%Xjd1^U}lVg_Eg-kLdM5I1PT^Q3+kD4O)~JzGfhOf0OVC!Qo6j!~eU^|4V@W
zlZ&%HE7B*}48IGWmvX()p}-qISHagDKr6b0CkA}7{XJ(PgI5mV!7<Q_=)r?i6s~44
zI@Lh1%_*==g)C(HfqfRzmso_~v!H3+ehprM1KM!~PYv8n-zDfxz3BC>K?VexT>{=O
zJA72%XFaqbIj^NZb!&FG_u8D#LH0n~s^|s<;8@TR9u|z!x0;;kaVts-;o|o(J}X>f
z3_k;#X9UC71=~{l&Y>O{n}WTAvCtOX(nc+P5xy^T1^CEE=nrG@oYsxfy@y@p2iG{+
z%i1RdOHXGHU|iS92Ci5bk_$Y(q%|0axpy~rNab_F%annGEJ6K34fE0$(j9KAR0qdU
zzYJ&2w}Gb}1z$OH8rXkzNcqsNhr|QZ6yb3Lk@#35^^EcE$<bBEOW=(tNycTVu9{nr
zTyO4k$EBa}7(531QRYJUPS}rHqztJk^Qc|xe|0qefh#@*A9^*unpfE$d6ayrM>$)2
zq)Z;2TSRS_6)k_CO9ibiRn6~G*j1+*w|DB~0f#D@9AI~cQueYdEVf;XX4y2bqfLdg
z*!1C@RaNobjN66(X1P@*Mq8EDik~=%6~57;1MMt2deN*)<GGrd$woCR!)KF*J~pX7
z_@(kK6E%`ar*4|`lxy5IJUHh~^y-@wdCH_3hq&!!(wsj=Jv(Yt>a9jK1J|^@3!VY@
z32bv2Y*QMYsr<!A?S{LFk1q8E?q<WmNY#hCX%2UDZ4-L;8hpr0B2^B)X5MVPetdR)
zCZlnUk5t7`kxDr<lAJ?$_+F7Z-VM)RB>rWv&C7r9D_3hU>=D+UHGG{Ff8YjwC&M0e
zoEwokoCW<bDW3z+BPSSXx`D@Q2v^NxouvlLj(&a^uCI8gt5Vv#>O^~bzoNUUG<-?r
z{rEW17^ma6%Dk2yy4O+4xfwrf#%{8MNm7pNMvpS*b)Ykaa-(0t$6SSjo?QuVPi=a2
zxb~sB&%!I#1W%+V1>S);|N7PN+raZ5$9HoiX*b>4MrH?`QUUtrQ#}X6w4-me0(COp
zqv5=cc~f}qyrzr}d|`Pbwa*E5um(A+4o?ev*idlCYY*O;Bw&}1es#FzN9XY?^=iM$
z;lUjT2B;S0R|9yC4+i!@_?(MZ{2B?L<HXyMn7yjk_<+K|GsnR-i>C%Se<`4md>=M5
zz&Sc#C;rC<ZqrWYe}QXiz^811D@v6Ke{s%#zIMhl``?OU_-dL3H5GhQ4}24Zzi}_e
z2ec`uPvDzE;G1x;%t!R7o9I$G!8m@no7CLDW>!$6=JU5>xvyzZ&oTs+;UGF1T<7D$
z{NK}lHIMNt=P>vwuIxSFX#9TFY7L&jTO66!&-vAUtxoS(6yD;a@%*~?$EUR5n@RAb
zU+~_Hec)38@Xf@-KK0)Kb^+UH8lK~cKD`|8Q>9Tp)ergf+syY3ecDsZrxOW%<QjQ3
zX_!|h@ZA(U2Hxo5QB)x?&S$q;flaRFbE|$VaLiv9XCJs#wSilQ(57<3+pH?d?-T>y
z<aeukUicioUXsVHvSDuRP4Cv_WNu|h41W{ft-AdEj-M_q{_WD)L~boYugYZcsD1;l
zN`r4cyznYsL2#In*RaZ`8@!gO=y8h|`!%mUK5st1TD#%c9tO1fen1^A^F7xKJWO5h
zfJ;9FG$MOY$$G%0!I#<LXXcki!^$60=N$Ch#R2Pl4Qlf<d|Tj}X!iP;ML|u3N2`KI
z%X$w^Ft=X~*TZdDd~(n6;(zw)%>)nQoSLk%TQ_+fHFymT;GOe*bfJN}^g4-K?b7hN
zbAbB_f&uHhbppNp1Dam(`{;7t-RjSpx3PjptH6y7(o^p>_Ud47{7z@Q+6u?L75?rU
zYw@mik#eq}C+|HR$_;Wne&V;J9!RhsPKo|YA6}Th)9@y+CReVEhkZNV>0<Cw|1frE
zJYis(-n=K}7{^gZ87D*G{MPxk?lzp@vw&JKr`+}N$5-QhXcg2LG^`T5C*C-$^(7)z
z;uqX99L|_lU9}dBlPn|ibW5a$vPPvE$^J1M&I``@Vm>s4dQmEmFEla!=VtVJJ;MJt
z-b9TvivCozhz3{Dsk6k8`GZb{wq0&Cxe+VL$V!1v8xKx5JlBg?_fT<sO%;=qp(6UI
zBJpa^fy)6K<azwR{Ipit>2bn?v%EgN;O+6XTH#>ea4hg4i-w>vjX-1KwmRQ0^`n=7
zfu{2NEfeF7`9K~uIB7OqOY_WdFO~5ZTks~s>m29-J_74B1`n-7JMw~ITEOol1TX!B
z<JkxolnK1l3=X3b`~O_Dr=2{<OK^2V@R>}(V^#lE4>&DyUcS)R(V)A$P3do2gN{d!
z$gFkIit_{R&qyx~T*piP#)mg+dn_{Is(|5M(0>C)+Y8Pr8yB7?jI->j&?nUtFHSgE
zq#qYqG<o*YM~)x(!5y*+-;j^;oiiumk%NWDxGz|yQW*JDV4ELsIWysH%Ht`n0(Y|m
zZK@mIn{{wLSK)pZ*Tp+rg8#9H{$#M&GCmV0zL?vr?O~hfJ?TrXS|n$T&5t5Cq$~b^
z>H&Du*!bh_Q0E=p8BiZ?=QICqeMaZc!`g@5{s4c0sT~;w_}$mR#e}1u?uY^pHH4qX
z$J=uko-X*m$@?Sq6dvU|b;u3oWg#@C5pXMgs!_W?WbCoOzG==rSR6g$X-LavgH`&#
z!-8$%HwtM&I{xNSP%GC3b#Dy)B&@amd2stHsWE>BwJ#}tGk$KPcBu|WJF_1B2H>1p
z@I1%hD(Bw}N%8R{H;&XT_Tcc#c<q0}Kcs*kV4UAzKOK|--ZFVtEd%=uWZzC#DN>fB
z;H6d}?Ptv&vzdA~sb6_I`_M+b>RXR~r%!0^n>|`S+@l?hz&WgIn}@Onvc`Ry>C)d?
zV4K7)jo9y0!fH<K-sI4ZW)4L@x2tq(yW*owJp$k4?rc-?^fn!O1D-i&m3f<0OXgYC
zei+{3&Q?8*WmUgL7Imy+p_Vpl`arW1)-x+Qo>{#<niL0oa};eVHJB#Bb(4yKYpx%`
zbHi2s7?@@`_b)IhtREUx4wK%THEQl!qpGep$_lP=-eL^^S6IL{`N0&&z&6R@Zmz=J
z455}<d<5)rfPPH&?M~a#nKt4XUJ0kO5WZ&)e?J4S(<JcBm`Ei+ld1tP^Sd9PWpCE0
z7&MtEI2^D|0x(UqCsK{w;3rq4rn2|<w?@im#DBo`l>1F!f-c~mUlzEbHuwk{u;x^y
zwyFT1Ru=CLIBE3|o)`Z+G0%Bh?5-M^2cHxe^(Hl%jjyZyg+GE93M<2Nh>h0=&GHbu
zeiu0Vzwm$IMHyF#=~20l7Q<NDvOcIigM%s>5mfEkL3L!mADj)nC0$U}5~9Pr3g|02
zGwvnuBGke|Z157bnFnRT83p*+Pdp<R*{fJ@^T56Mr-Dfq_+@5qj{>)Beu~}#hADEu
zuRUl_ebAsz!)dJjgiiAtFYQ~urh{F^?eN2A@ff51+6xZIV)UzI4ZlvI-|WJd(()d<
z3_8pvd?qXK<R)n5*CBAwtMz^|VflIdfTrdEufPda3(-G?Z|5)=XwxY2Z=CFvJlB^z
zcI@fw3!6#8$wE)LF`$w&0xG=)UWb}ERidDt=L>3QiJ+R*gOkCNa{|26ay5F|X5Ndv
zXi?yrXMF8zM+U)*fKuNLsNG5YLF_ZxZ_s~U3yx`tPt_lIb)>sjdoFr(9zRN+BVMJ#
z7qcy+Pu<J<G`YS{YvE!xqf6BT-(&~h+>7hi1MtluIMO)ao73Q%75CAns42!CK#Rgl
zJaCpzdGHhe80*8E1g?R*+1bpeepP%LlGCRMuuZAw{J$Mu6&&d$2NVCzaS!~Ihm2#l
zqGr3~smc5-$lPr1Qit9ywSMT*{5SvRD;!Nqx8|gBa}K>*p3H6?&E{6x>~4MJHZ#{W
z?k^9%NzXiQ9M7$kfB1XwK*L`y<&W#8ch05Gaoloda_bU2(aF<Z-7er$VH2-wwNK-C
zEv@0J_A*wAP%}mI`NYTn9l;ngvxg_SKyTc6{8ASK`gtp$^=OIZ7>9Mj=%KU2l`Tc9
zIR>u=XOlNJzT;1LyPtwH!JXO@Br5@I0~Q)eO?C(Uv-?K+*HZJD^O;n`dz~6jQua7r
zT^j4rw<7Raa8;#0yA(ad#Tmcgw8c*KYwJ|u%?@py>`>ikhw`^{kmu{r*sRoN@Xx~&
zIJ72~L-8KlwezxF`)1j7f|~6^I&}9hHf4Q|-u}b}zh>9)X%5ZIg{LXAm$MYSn$2t8
zxC4EZHTenJQ90J&%*^Al%cx=gi&V7)Aw8Z77gz`FiG3(;aX2aVqS$={TEl!QhW1!z
zxK}?~dbMDoU#-#l4yFhwYdrb~nS*tx6Bb6I-Co6Ge;XX1iO1={8a^H_h<?BtANd|l
zU}n`2=YaFRrl)@X#kfWjnD!Yx>=8Z&IHfP>&TVVq&;Hp>9m%U0;Oee^J*j(!kiXBh
zdk}sLGHVa?qJM$@w0Gw?mjZuAVFx)y6X~_tOWq~DX{YgON7&#(rjbv+hqKrp;5GgW
z50VLfrUcp%986d%@CzDMFSwAYF?e4Fp*7*#-OBCIe|W_WUTMhworY`{@R1j-Xe?fy
z!H+l(IX-zd`FR|8ot<3QDuHXjJJ-NXgW-n0)Wus<6+9z&o*MXxd7Q)C|Dzln(rL0G
zZ_*!Ho?hV^WaiTQceotqRm6@~+N|8t+x!hKFKka&O@!|%$)1o2&sZ92-YM&;DSvg<
zqv~Y3;M4wy{`r-9aw2u)(to|l=#&o|#mKvlUi#wX5|G`ruN9dCkI?0&kuSUv4-9?c
zoAJ)nr%$}dH?oD}a#jrdOi}v3r-5t2xZ;9o=7DE&ec(*2=Va<Ug$KGtZeJw$=pvcy
z8Soy_(^L!Hu<ZwQ&yM6w)b6HlCGjn!?xw&_@@T;)Gv}iPfF}xdVvRo%Q2o^b{T)sJ
zDQo%-FM6>fpcCAdS;Ts{hkAqhB0aeJM*=*B@xUDK0_3Iz6waJ{#vI*CuSN=RQFrES
zX6leKf0(;)iG7%x<KS4DjS6vI2^>jA_M)ux@casC7CKXTazK)L!C1Ay)r`IM)QmCs
z$udU4O~R|LcHo5x!bA1~*K_}3`1CWIg3811z52p)0ppB;%W2q_D}c5%CxoYjeuFTw
zB;Wu>!b7;x432!oo5OXKE5%!Sq%YvPS%Xi)$=-?QC(dW`&kFK<Cj@k!y{I<p?u0_{
z<paF(ruJ&WE{`mIJnC*@jew&bzZ+cRMw>d|QaOi9_7X0&`Q$_gbgFg+r&>;RaAtr*
zh3~KqG_-5W8=G>0Z^B#K^dhlM$FAeEId0Xy`Bt5rX;q7!R&rOYnn@pc%4myjrLbt}
zA~U>&S=%_JbxmxucG4de`N*V3a5OIuo750*O_7l%6{tj_YC1TbMMjmLV3c8`QHi*1
zFx^PMxlt2W81-tUQH#I`rNA_iS0gn6+}9gy69C)%ITNW5C&4?%(UuN_Nx-PhcB5Cp
z-6WwFo3Rw%_I$XonfPd?;!&dZ$vztFI1Ic3=TfmRTu=|{qi8sq5I6{IlL6fF&-*<D
zT(i~&|6`6+U$~h8ox!>7!B?##6#|1kt3gdw9{ruVD^Jl#vN7p_$P=mgx!{fX41IOM
zVDLkBo>TQVa66g0>S12I<8VH!KH~#A#b*h=ip$^U`Wce<cu4hUgmeYI?-_oD+SI*{
zSa`SY1Qo`9-4ku)dtd6k*7QWPo_dN>180G2Nk$IdFV;tJ{g6ZWWZ*?Eb_e5kCVxf&
zy(t(_jzsV%cksxvhx|8*zUBdb8Q>=-n!p|4lOv=2`ia(bVy<5|XZdw|qF>+nz8{#R
z7MR2Y&MAT)!@}Q}1^aa4XT^s56)W1WyB+*WToo-Rt6$B&z-OS%q*>*ar@dDP?s>Fz
z2z+4)j|%K}>l>bsX~W#gkC&tfy?Vc#c)}Q`b$5AH0ez%uVV~X^@ZmNFPu%dTKjU}b
zQ9L4pJ^HT=zW78QO}XUO;Aw6pak|x@F^^G!{kavsl-@k{Be!lxp~H;#C;>i}j8(jP
zHVE(8Iy}7ic*o!I>i%-{ApWPds8=J{KU?JUXv9f;yLH@Z+SH@OJa5}vuin9DY)<S`
zbL#tTx#3#!`Sb){p<lF5S&pJZox($W)u)7ae5!KWr{3p%s(Hev6ukbD3(<Hc`N#qO
zAGQe}>XQq6W8-yos_av4@XeY8K9zq7mfGjlrg2`)`p4fa@o4cBx3=7MDKyQcxpFB@
zUKeYlOUnwo6kg7yyxm>Y4=&CoLstgpq=374p200!7Po40RRph$;l7f|-TDK5$pzoB
z6Rx8T*I4jD_U|tFxvvXYX9!s5Mgq6av#(#}Jxn|U4^Lx{3KsFoe$uNAZ^3Jfja2o~
zya&QbtN~vhMSld>6l2Zoa2njh?Za!-mte@LvC$FPr{A^-svr8*+O<KQgiqTJrfG-o
zyDYb12f*pne8YzZIqw<F!<y`YN2{<C%`+8vdy!8IBY1ymFn&gQ^!ppHsh3-i!EO&9
z(ErfOrG#}{sy@f5-i@8Q3%)4|zS$h*(8q=j`S;qjF50fvh3(on$)*-PZ1m~c)bG1h
zQ^~EUHrJ}_^m{Ih#5ZWM>OniW=N4AYXlmswTWYmUR;7ohJ~7Islu7M!rgtjyLYI;c
z$KQ<)D}B;{LSXB@wcu*r1l0cw_<L?Zz5l`oE(oYu5&ZJ-U*~-In(&F|$?H+u<!(K7
zxYenQ8+_0D`@*G5+g&=|*`=r)E@e#R(oBm}8CKe~y^l?2XV^6Ms7=Lk+Ldg+T}gX7
zl=`8Aer6ZG7q3<wrr(ggCFdb<VixM`dm%N78>L;ZIs217&C6s^^v6e3Whv*)#Nvzz
zbgVqf>1{`gD@aem-&OP_fERWo?xEG?@oJL?_wTH`oh^H6Bp&10WTR!j$$1(-$i+zs
zhRKTmrXU(sY5JDH9ckfHqQN+o;aon!f!yTwvX4wx?tfj3ey3M(BVe0M`{`}l*HeWr
z^i;%8as#vBc?m;{fzQdiiF3%-arXNi_!NHk?0WpMWS102L;Kbfp6L&HR%gI(^m_g=
zq6>P+5T%EC4jCSm?t)dwQ~7o(3Xcr=L+ijT_#ii-8zw_zIZ17MwIyq8VSEZps1*-}
zRP1?3VM!y^^CkI-UCCwy<JiE5=2)EZzJuNoG?EB>2D#z3+Z>|5Cm;IWbo`GOd#V-v
zRdI}*)!CPxaeTi<e80O|krCLOzH&6J0(?ECBWEsfyErpi6#UZ~G&3`=$A-^o8D5wV
z7kJO{f7h6YHi|d=H$L$dXsm8K*^rrejrbduKjRFK%<MN4_-w%;Hx5!efMYTY59oAn
zYBF~~xvT+g?f|FVh8$Gp<t6Ggf6ajE)uxA|X+YxsvYi65n5h*9{jZPX1CP;SD|m-_
zdx3fC&r82WKkztu{>&D5!PtA!J_@P#GkPCShjepzNXHxFr-QTE|Acx3ynSdAwMtL;
z$8flExX0!7*puMGQ<%Uy_{$EnH$9ye)QE$41a4A)f=h<tON&OIej;#sH?YlG^yPPW
zAE@cm@EH|kFE4cmug?y62RMZ!1tK+<|G#@Gz8*BMmelzFdSV|@$7dK8(4&{|yWpFx
z)M2TZU+WK1msR(u>|VF3bcCmd%QYT!>CZfujtz3@YfG1kWOb?8bth-%ICc5GL#LTb
zt;#s`?UG#+8rc==h)v~(*wiqmjm!u*nrc>cOlH-0bf~uJENXt&tS86eYBrb^pkFHG
zQnPxZO~rPaHL0Xo?eGsD#6Nr$Kk>5@Caq6p;!IMb5*v-mYzEKxj9M6D)Q3Svy&GZF
z%W+0B>x?=(k880}OTaiQE`deBGtbY#vvB+AEE*KpW*MB#lY{tez%)DHZ0>KSmudrf
zI4j_0mf$;vAK5b-y?Q#{rpeS?<KS@MZpsbeclyJ@^adk!$4A_iUa9~&6kwaja5hbB
z^u58&lm^=@ihxt=g#Wl5x+HU^TT{G1weaRt<afa)I}73YLEHHZ_Whm`z9<<#PsjJU
z!BhOrfN7DkGH#Rp00Vu3=XuCAfX6lR-1(f1i8*WUb%-;t@$^m%;VXp41K+sv;!_9T
zbic(|+!oaNS?m$MpgJ@ODleS=>D=IFwEe7!!91S>GEfJe*~ECK&Q1ls?@#^NpekBI
z@_^EOq~~WXTufiT@|N&xCs^b&JVYxzBwc^9N9O>az_nzTU-?@5>CJ+pU@f;&Pfxte
zp4`W$b_d`BXL$7oKWcv?SA<thJF^Gj*?5EQ;zWbViFbC!dwQMGJ5J^ED(z|yIb|Na
z=Iplz>C>AI#_3C+v)QeDP23t=5PpHZ_z+yeny)U-pL59?+s*k}Zrz7t*t&y#xiNb^
z97c_|PL)6B)W&U2?HcD)BfMe*OinqPI8~~gQ`fUObs(8jXJR|`<gY{ZQ#iG@xRY_{
zRPK3B1$H<UZ;Mk`LQc*Aap-O*hg2J`vYJCvxb4B;#69fLs|fZHdIytdr++Y>L(M<g
zbvdp>mGRqGYU5DEVuy;Ia%k35hvNKns9b!f{3+o*`CoUc@;`Y_6~;UDIy-yf1(%9d
z^JorzP?nq?&8Ba0F&Hcs^U=21ts9Hn`a6qjDjL&7w>FM;Yd`Nz`I|1q0moQ)-9z9h
zI?v#B?sC#6N1nIcr9|^x`u@nJ>1a-mQ=n6!MVY}h&ohGg_&SRFdcxUs2gfY<1WtPI
z(nzi%V4T=rT<Xo$4Xo3R$H>Lc4QM<2Te{U2+!%w76NT5g^>wfIKciNHd)vOpuZM-;
zXz<{iX3ZUPmYNW3vxoZf2=(Cs>cp>b7jJ*U%cY`+6Ha3gb>7wmK~31i+Q`*?E4^7q
zgSrCd89g1}E&KB%aQftr_|2o}mjieIfeSm;%?GXkBWLj91M#Rd+|839?^Rk}=Mk4$
z;1}EgzA3rYi6?|L>x@I$z&G}Q10Kqu6C3S%=eEn1*{-eQZMqp_({Aue`CnGK>G$kM
zO?Gs!RhEEN&x}_16{`YZo3jnA>Q>K6uaH%*xt7%j=Wu`aHddW5Sk=VG{e7%TOCCl{
za+{{tMZb^b(MtT<_1b#1jsDb|ZM<su+QZpXc((_DGkLywTevi5s*{{(r=HGts6Z*k
z*$=x;!6(IDXIIUjT|+<G$UwDGpTLO^NBeJX)zv&!rAT1amzNeeFN*>vEo%78q7q=T
z8Wr$7mbWTPB|HKAzU!n_J0qw=+k*9$I+dJSA~!i@Emq=LKj_oZYVZK~!XMyEDn?K6
z?i_fh>(GBNkc^7$-PDbqn&f!>+L0yN4qx`6`eX#vB@gf!xz4xA5xq?I#Jd<R#@oH1
z3wgO#&IDV8e|rqR9`bg|?&z-3E$G|&OSU_Gq5a9kXt|QJE$Bsl_8*y%;hYPQkqj<;
z9=}<Gud{aNp~kEDoZ6JNx_o`OgAeE%=zR&%37b&Uy+0q+u(SA;9?*~09=`yb%n>sf
zU`|)9I6!{lRXhM_?sJnxslXdD=T7oHy5WG^conEsn-yn0-hj^14<9_<jSROLZ#(E^
z8_W0{5>$x&c++@vRD4r;SzpfuS$B7nWsnJP1N%Z8*6cY8(ew|JIfeKA%uddCM>krF
zK2Yuw>+>0O#qD_gs5?V5!R}!8Eqlq+dVywCp7s1pNI&E8SOp`MTkIuV3k#zMd?G(!
zK}a9E({GfTeVjV=;?#hKE@a)_KvvdX_?4tV<$sGNyemL|5Zrcua!+H>a2@nz7^q!a
z!AFB-GS-3DsS!{=t{vc<7Ti|>e6y`1S{Ax~ya8Ytc<=Q*c9qTWJj~m5%+vZQ$bxSc
zR6%%{p?SzP8_)BAvuU&wZV)_}xOGVXQD0~H3vM|_?${b?k-=nv7~mDF2NkOnIb`T#
z*GmUAvRY7Y8VB_&g7-Iu9t1eet>B^c;F1OGUE=dgn;7l4Oh{Sq<qi6WpR&L+w1EfU
zeNM>!_y$d6b6tG)aoFSe3_cCTzl?^FH5J}IKC=~%0~!a1KZ_cB&qJTCQ*S!zc(rz#
zhinUvR;*+nX^z*B^>93UM#@R-7p$L|s<~9*uTzgUJ2gE!d&3Hcnphq9ksW9Qc8w}z
z*PJOfS>ls<bHj?4nfeEx&3}C?s!_?J2Z=4}eao!Qhs@f&(yY_-%^Eh%td#@IYS!7T
z97W)7&Z0x%74BHeq*$d*`m@QX(IKM_wKJ-9XQRHmjmibS89LCYe&Cz*;G2C@joL8R
zs9NBgE?l`8+aCCi9&j~ZkHceen*i?d5ZETxUig-sk*W-~xwanMu?Ansa=bc=(Us<N
z|17cr;cg~Oz#}snzGoOZ)*#-;zF?T1k;>PN_m=sv+6S-0`#jwN&H>j9fwL*qg?_A#
zV4t>lZ@@N38lp$l#*<tbOj8<6#dz%swu#BcxXplGn4HfGywZhhOCmV79QcEHyu@JB
zoIHL!a8zS3&Hj%e?YTq$>?wSikEtub&l?z5g$Ka{MvxU!1%5IUS!nQlHyK|S&|N+=
z55qf<Uq|g&8Ek)+aoXlSXB>dZP4r<q*?Sg@!Ow!m6IO|_kPhAFD|{ZBQnqd6y3F*e
zZC$_Gqp=K%<JU@hnnxdlhZsX&b5H8!{^Yda8~Fqe@#r}F+HRi;W=Cs6w=k{ssuy~}
zm>#U-W-q>FxQ)tQB`pZXN#Rwpw;pZUM)tsTkM^U3w(HBf-@&8#wLQ!qk1jrNYc^|c
z%gNwuw3odG_TYMCEhVL=`5iUsHJ8eQyLYU0Da$aI76s_*YljwrK9iU|J($#``?s8W
zcfhH9tDM>}(y3WNr>1sr>PT&;w&Zi_c`ESEZ|dB)>~)VFoXzV{>^%<Ep6!su=+FbS
zi(}OHzgs)Bn!V{;X@?dUbtrKThi>9EvplooxwVrQWmkuqcE!tXM<21P+5?+%9=7S-
zDw|d_7ygd6=>k{kr8X7aV$<ura04f8Ds<hZ8}Dq2{l}&R%$u|K*t>4owSB!^mB+B>
zM%k6V3s-Wx_FS=P0QGs&ecbZ*SNVJEYnx1DF}U&cAA^HvnAD*sA%}7<<8|zHaK;-r
z?L2#WCzmo0bt&?=OH;swqu^=Ire^<#!}-V+4i4DL`&tz4W++_Ebi69v-ntaS`#goK
z;7j-$F2^mGCh)%h<m(Uo&5@>VT?gMhEsmeLqF2M{O~}!JzOmkZU0LMU9r(W|vBAw@
zXetfi1zBs;fu*nFrMkF;EC95_abWD7U-1X$!oSsmb#4Uv3G03(E(iF$Gi#)Y^>0Lg
zUjL5d5*2}0N1NFS*BBTC?ndW3)fEmenNJ<Fc-f~tY6>oF0Oo55-_&{(-l3mP)fnwm
z9I)u1Ar5L^hgz(*t50vcTDGt&TLZhQ{zq+=iT8b`O;6g`lrO6d98a$&*d+fnGQ0YM
zTU=J%Vm*6Q*Q!a4ttt(ksawk`drj~QxAFPeL+<|qH$6Rq3=DpML$p;n@f*fmvXZ+(
z23T4fIjDAZukXOC?N9*hdwCXXXjiAo40q~85;(JU;N3j*Y>*>1Y=B+USJ`!%uM;js
z`{!?aw6$v!pZ~L1<hRq0+NYmY<@n#qWF==ATyqOv{IEqOw^`JGw*~&&qS$BY-&k)^
zvyFT|-=cSGEpQoNpPv>D>10*vD|jTA*~#g5>H%}<?Q1wXbl33PUK!vn=P-Zg7xF6g
zEsu&OLmQ3oEBTaw+OCHK*uy^g6z?0gQ~h#5edr8F%-(+xFNWQSmcc$%_d7M!5cmZ&
zw#N9g{-7rpN8cWFowbX#@zje*{kMoLa8r~9eCR4qKd(xq@aVxI7n%w9d<h!kPWGQG
zE+qwPCny3BIhg18(XGP{JfhS_WzTvwmzrr2Yg(Bx)P1bAW2p5#*}ydQx~drd>w<24
zCG_Epjzh-T2XNA3>Pz;6Jat&(udz1{ps%M1ev@SQq`)^l!7-m0gXhtHa%3bI6y7lp
z*kUKV$c)PLGgLrZ%ttO4HDNXM#36}TTUkS|!(U%I=+`p1>+>h+AIR@h0&2L+YrHa_
zAlm^w`{ZDsiY@ag|7M>y;=AjMHl4gL8Z=r;`3gSWN1Lwi@~LeM{RpFcsyGwh9y;*T
zQ{2XaBSqsMYQ^6;9**!i-YwQ}n~l!|j8eP@o`Vqj65r>An_1Tkj8mUn(u!!trMbTx
znPUIG2JgISf|s!^cnN$n6ino#cG$x2)Gi6f&(B+RVV~Ft?ztOKb8tzWH+UJr5?*`}
z^3&G{ce5DYqyoH+;~HEmwaKx`XwF`|5O9{mGSL^19BdPx`@l&DvQW#QA5|$!pCFj2
zEZA|{3bchS_$GF7&JMR3!Kzc^qD^$d?=vf;hzn>HsUkIwv68SUI23G?J|CVDFzyw4
z&fE6|XHw&ri3jFd3$9|HJK2ufYCr1{ocZRoaJ=k2b-Q^~IkQJaCQ#c|cWdNPm&UAg
z$qaY%Uu%~te{*W~1*i7DqXwdHsvz9WL)QKNf0=*9>?%|fjuc*Z!$zydFvo9~rf2E{
z{ZVB2?J8|ihIkg0xk+BlKC?<LHtWq~v(}9;t8sm^DlNl9(-R+YeUnzg*<AQxRBimi
z8G3M=#i+3KMzWWTdTapCbO-BzZ&HJAhL1I>=@g?j&M?XZ&+{cigf5-{e}HT19HU<W
zT+{9d@5w)Gvk#oI3(T;UngG6Q@;dtUSE1i6!&9>eY%>oZ@C<mDN%(`Q8&;3xeMB=_
z1%J||7upcmW-r+0ofjU*g|FKQ7X){c2X3W5*k%uP!x3;z&L-3s4ajW;+f)UIFGiP|
zRvc^upSB`rq%OfJHiNV21+Ixs06vPx-^XXZB?Kp>fhSD@hKdWP^c6jk&+x<DkcNV3
z?(tdfSqM%YhzGNCNbahPuQXt3xV<%C-%@aWJ6FN+vrkk<?>t%;oj*Hx7LIZ8g@C>@
zPIpaY4l<5Mf^W~&WlrP{XdwGg?q?*rvGzO|fzH&Ktd=sYMcMr-^9c=QpHD{?pabEN
zv4;EPsf$mfm`}Mfkk1kitn&^1<_TKPIo6m>Ud=<V88njo$8LB<;BM|TMZ+oQ)lT-H
zhq3T#-oOvLkxZqb^yszl=rVd#&MefmaXtF@93Fu@f++N;_4vLvy4|YR+O577;dH*i
z%inP6*G8A7&7w9R?9x8?iAXq`VI^JiWrlBv184EVsfS0LYRX<QZKjhxYi_-8Hm%`e
z{&V6(KwC}Xl<S*=UM7cf+-E;I=1@uM<>oOCU8v_!nsN?(2Dcl)GL2F>WK8JL+HZDE
ze8Zjt4yn4_uEvAximqeVm@sPp?>2loHfbmL1-)Q(Kbwx2&@C$1G$PEVAAhX+_Qa}z
zXUTqIUn;)Jss;<;1r}KqoNblOVO7auR-L9ESV7$#@0dmBmRS@UX;Jo|MHSmvIJ?<G
z7Mn$tQd=~dIkjfFMYi}>Mdq_AZw)IPG@JsD^Nc+&-4VO&uX%nS;GG`AOVvWBNr?~T
zqf_mlu;1Tzs@i+-Ph5B(pG&Kl;}+_e&of=R%5|%UOS8H$=No|$GP>kG2dDMONp0oQ
z$;<p6I$s6w&HREMh036tW%4Nu@6;IZ%>eLC8~neAVo^)t`7Y2H{)V;o^gzxTWBhHN
z$NlRVhj2O#UeVXj`nR+InT+7^vDC}=;BoF!E1&K~{YRY_Vt+|lIH)$Pu`fo$nU^3>
zpb8!@_Jo}1cc&kCIZKM3^+_Ji%0P3Z?|l<`Tp}}jHC*#OixZvQq5lelO&8nMEu~!v
zV%s$MvPIh)S+w<@Sx4ZLPwX_S$8NLsA2n+Q*P#Q{TRY5ZyV0!t4thG_m}AUleQd?g
zxJoDIGc0V?WB8G^y#GnSH^p1Sm9X|51m9eWvTE8W*242vZT!Obacz2^(k5FPo0_mr
z#^=86Ic)01Su@|`IJC33LkGdo^KRI6X|x?(!LHoZ?VPI$zXXnL3wP65c3u3(k$<%+
z<7TVwcem<u39C}Qx8O5mY#g-c{t7V81dF2Pu;wyG9!<5V>^O^Za$kzM7F7e^<OKW7
ze`w(x0Soh-T(ZjGqZT$LTV_+j(>66IW7nU4b|r$}iau>u;Fn!1m>09gICO0~?<Kc$
zz-H->z~O_XFJyHpX;G(+G0!UDbE;j(sg%sMon@&lN;oyL4ZLDEr*2MgDhJ%^dFq!Q
zXvdl1b7$Q5Dh5s~?R`A#@VAW@(*FncK3K%1oUvUpO>xRf?tPqlj92^-an`$(8~t%T
z`&9BA9^I$ccbSd#W(c`VXy%o=2edDSo;=pvbWvy-OX2X@6KdRG-M$^Ex~%<m7PCIH
z&eso+{qYiC>tg(~QE(&;LuAL0&+-Diwjrp90gS_T?8WKviBKC}0XqzauU_yUYa#2N
zo#$?RiuQ5EtFF7fTE4)mB=k3Ci}EUQSFb{Sz4|`di@(mRk4wF(%<t4b?!`Od)zn8`
z{o-#=z<t(@^s2ev%lV34J=*G3roUdL%);ww<<nBPPXo}d=dC0U6ixe6N_+sd$e*%P
zf3jXD+3i=nWP$(pz4i8>2ZL=sc>{_Bv(y8-G-ba0cLG0TTj~yY`14!P!NFGZXOlm{
z<MzVym$)}xx?$ie?(ak`>naXT!&9*LBDHN9ykINf^>?79QinV$j|Z+joD6$RE$WVH
zSAxm{Z_<4Vb9E`%8iUZCnX?bV_#W(Qz6yWz06*jFfbPQCF8NB2Mk+E9N>Izd@mv{(
zw!pj}43<3Cj~+?9!2up;>xrPIh|gp{o;>!)@1@};>d@z01%4nS{Xh7c7p=xm+?8{H
zDu)yUj;_Sok^3G04-V2X7N5i=UK9J!$cb*vLy!N|&Z!^GoHDqaiiDpZNX=V1(y3(=
zooWM@o^-84ruGhn<#edwQ)<}S)UVMt%_&SzI5ou}YJnDbZJw~!w#fx=%UYc?hIz!=
zoFtA#h3J)9x6Z7B_=)qfUR>pDg*wwsx)Ev8rV!c0@HR)1o8<dy)WF?Fb#G(Tp0q~2
z*=kV5bq4j@#qDAP*~kVp@EP>vLj?Iz5elmsp{U#uI@2VaGu!Zs9Y#0es(FYrYz{Ea
z_k%g$3b*ZGJj2}_guD3-wwbdW-g61QGPs)_v+=>ef8Cq_78-*lJDhni=zqLeF0f4>
z=0Zg{n^OO<4LnQ8iiTyPe<~c@)Da#C4s9FwGFE+R2e8fV|00#BEZ!*QS(<#{Blwv=
z@HSslkgo%_Nes5>4UQT2o80;zA*BY-ta^jD@(CI+n8pgPnGPRuQ9j>+Tj6RJp?i+R
zqmCw(0>61MH=eB6A;o@xkN70x0qlEy20n%!WYRGwW0hgd;aAuNUsvlC+|3fk=U_NX
zGr7Nl*UZWoK#Q7tl0NRG^lwuWm!c;4QiuC;(j$Z()%z}ab9hAZQe(W0^66to_?jBb
z?Sl06q$I2519cvA+rHP!xg&UQ>0`-1)2rdbSZ_jJrRm^Rn_B2p#k}+*diClDnJw2m
zx<j7I{vcXZ6Z&1?0Bq^OIzP!@L63QX9@7UcW-lHYOIvt*`hFhc9l86Oe1g;T!ob^j
z*^4jra;csL9;X4{7sGp!)};sE@mXARl3U_b6k5&4ag5&>_ILKUdG%R)OFH#619{tV
zoGS3zA;(QHI{Hy5u+8W}4!tuvbfY@hJwIQkcIY_R{VH6|+*5X~*@>?djUd-dyV`i{
z+5mqt5Z^}XWOhBqdm5KqgdVGHng?H#DQMHeRyI7xWO$=@l=*8_qX$;ahZ9%`rfG;y
zkRDG*S))}It6G&K%&K0GEs8l}QNL9dtr%^Q+hx%yc+bDNz%ofJy77ft;3@OvoLPO@
zhm)>0tNwJe?hZApN>8)04KnNGM6-r0B`5PG`!(2TYZi+#-Z3kjn&Qh>vtAdl=wStm
z>Nm0I1&?3&Bj=ys_kaD#s@k1wsxi!_>+nzM<Jnca1l|;{U3+=mjS@JNhhE8(rJVXu
zllN{6@9zVr8lnxpyy@1Doaj?Ez1rQ0H6HG+GT8Uj3^KLigQ*)(1A=ej_hr0|qUUV}
zeE6#WWmGp|Y_?>Kwui%6UIos&KG`{}nem6x(;pSo<EGSsaN7rN;M)P)nD7lmckyci
z*gH=})_v;Y9(myx_K<T32l!+q{+mi<7~Wwl!0B19I`zr!qz?sOF}%o#d3NQEYuD)K
zyyg>d8;v>hh4tt|dbIH^CdITs8_!_U>nBE)d2Yl*Z`5yUtpSNmoC#yn`D7*~&WJyd
z?DV_=v%XKIHrr&z<7dXBWzlj!yvQhvZa)HNq_pZ&U$72%w$W4gpC49zfXA7I_Lc9Y
zRdIL@Yw;-FD+XQxSKRMxSNV~6o4Ku)$*vSH;P(>Pbq<bjn$xbA;GLdD(5iOX)SSQF
zxDs3pjvAJezSJicRR$CH0?*_bVWAJ&B0ske#?Zqt9KL3-MXk8pGsufyV^IgPIV^YJ
za==Ajc%BUkSv4MCQJB-J#JzYtFxe=cYZrW!RlAbWiZ^LH&vEr^FgrS4xBfQOcEKfr
z-y3DOsa*z}hJ)QkBm&DPx2Zp`>k}H<C%CW_{N4rl%--PtoV>20XLv1W#3#x))f3*T
zcph{)xT_Vvooa;_qSiE*YQA(T598}C+?EOct<g%jSLXGpEchdO(g%lMej@cxPS%YP
zE5O*`gLm*GO@I58G#_KkNM8l~QNm`N0m2&o2TU`Vec%P_{}|Tf1F6VIdlXU+>c-)F
zL;5x%r0-6?u8p4&kJ7LA=t;-mqrlkfTB3h}ZyJHg7u=xlg6xK+g;@{5HeHT+^>+!q
zgj2{7AHurl#RJ&It0idh#lc*oqP@xp-f29;t8e_Ce<QWu9ysm;URm(^CGX=^C$P`}
z582^dUq<mdwo?QC@oF#l=0;QM1_wBY|9Kkh(PuCH3km(2RohQZ<kzU7^lR*;moNo6
z?#=Ph*aPYdzZ8ToE!Pt5sY*Z#sb3FH^edt;nHFn&yjGt!@VLiI_;k4_k6oS~NNU!W
zcCtp{!0N#ftxZfO9{z)_Yy6s=5YDFt^KLTQ@L@EqFXVru0pFlC<!wZ5f*v((48EFf
z=m*?48NFa4^L5`z_^u7`K&$CD+zzfeN?ycX_MUUh|CiJ@88{!X0X-cx(Hku2ZeXVm
zIl(xY;c?5MPfrIUWrgF*3r}O9x7-Srao}HuH;QN%QVI01<LlA+ngvw`KI$~@Q6f9*
z!5fcKcl2<kA$v!Cw{kpnsU@}U@(ND%xJLc3!!G|VyYe5ks}42n&WaA*V$FY%m3lX~
zLrYofo57<mXKjD8fic^j^_(?433K@KcGmE*)RS<g!&6u}-+_5I5U#YUMR&;X+qK@T
zRII~o3!1g(gNYnKliK7auP>oVwQ88~K$tj_%%qtgj5>_w)VwMF;l~DTi)T=Y%@KMw
zFoLX%2+rG!Q0{mUdip$E1;>T!adbHAb2z$Uxa#4pS;w`3%LcC54+oi>Ybe~!v~B1`
zjP;&t!SyTQFX3)3FM=b5yD`mz*O@|}H{8wd5%`k_gOmEvza7K<U3olso3c*$AaG54
z3)soP??k}$bmsS3<JoD1W?Bz@usZ*z5<C&}$cE>}P>5csobWSQ;DgfQIZukW25d7n
z4q1YK$U*vv?(_wWiq;ySmN@<Zt?6b+xzC1_3T)E@t@9Z;_0$}`A42b}g&w*ZXp<Sx
zsD9zug1fnX82*%TR$?-GR5Y0xc)E5Mq$dy$XB5~uD`V^F0yyFUWXD7VG#q>zCo}vF
z8GP%(_D$(~$$)p_Km5ams`>RbvtLPBD=ysv-*5A&^Az+cw19{0;c}{hdss^%;8~Nu
zVsAO+)k!p|sA*ocV}4#8%v|<)RVf@kwi&*Ns_0dPy_|9FRljeX4RFarrjSP&T=eNR
z;QK;&d-0ki`%Yi-Ew{Gsa;x85>N~Wl(aqf22_L-{pH1P1E_Fv|SqrB)e>{Cca12SP
z(XZ5JFGH8=m)u4Fo>P^t;IY~3RJTRc?PHx99Sx>2vM1MY{{LAb$rIrLg@bf3)^AhS
zCs~a*W`sk-$f+y<cF*=7*gYS&@HTgT+x6fP^Wd0WEzn^5&ZT}HY*z(4JPv-d{Af#w
zli2n4sZFtt+c+D>riXY<Kl*K|2foQ&#-^s3@pQo7<OJU=JZe=P@Xf$s)XYKV3$^|{
zc+PFPtvdYFq8%43noX|vEb8q_krupD7EOb*shq?jF@GvNM#nv4#y<w<00%k_?^=Pr
zoKxa5D>?ZJq0(k`NKBR#=a`f^PQ9?rq}KCI8aUjf#s)ILnPaC*(tDW2q#=Ltvt2c+
zFB))mJo>TWL$~5<JIn3*y++O}HL4N3>ay=9wJ2cLBRGrEUb7NqG;35h<}Uq;FS8o)
zIUCflvq4#k8?-N@K?kQ8H0!xR^TBNydc#lLaca|4_#k+=*IWHyLBBSDtD~*|)2Lqj
z<xCFNwG@@{7Q;CvqBbnd7_Q<EYF1TnI_qFK_&hmW`F8_ZUEk=bFU`7)CUGf<=Ns<X
zGmt&5xkt^xHwFJ`aA;GbE4lRPAf6lW%@^?MRBEwNw{039X_KG#zS41vuGY0E%NDcj
z3C)_bj`gQ8eGRd}H^+^-x7DZ=$BarvX6P%jwcf4bJ(_A%SwCx1F{7&dHmDb`v)Tg#
z+1Ccexol8}Ck9>oXi$wrMwO}z-#h?r`M!}H2-Z2U=!O1P)npt@xexDi%c@sz;O)Q%
z5o2x2`_ZO8vEXf<+4PI^ZTf7251$Ap2wqNz23iE}CLValet_|^3=IrlP>;5pa|4gG
z6Yk~{npCct7ES2RdBa8v{SM$4YRk$!E&2uDm!UuB;0ypO)3^G2yG4y|Sak3Wzn|Eu
zb=0C|8ltzEt?CKybaA{D|AJM?3fPp5a{{fyY@AgM_MO7>WPGpaVN)GEJaG%7v$FP=
zPHR)OWZ-$c3Fg1xI<VYU_^P{^;4$4cEoNS1y=2q5T=b!!yJkj%P0KvnVsNO<eh2<L
zr#@4M3`Zl#&HR185$_x|Nwgc@2L83x0w)=#PCOQPsoLT*Th9FNLa+ZX{C7(|a)K?Y
zWQ2pD5AgsR`}8OH+gTe!;Fm#R_z<W!&()=0wiNpTzJSZDsq-%2Pn?hbNj<of`u}1V
zvY6lt=Are+f5%>Xgz?v%J|8glW-xee{I^>Y&?7Ox$GNmV{XGLey~E3M1J_URa-OzV
zYuxBi9ns6$f%D<66H?=i=DEFOZEOy{30wca`tSW@FK0`_^}z9bXRVy#K*#60Vxh-p
zzE|C!)1ygWL1ZJJnt^XdkKwGAC1@>p0#5CruKW*ODg^$RKyE5}SeN`@5Ok<G8Cd6E
z`E}zqHR=m`KX$`mP_MrE=+nVXU>Mf%Nz|!t!823off!Al`nD|irVbul8(Hb#o40U8
z_2a|A(g)H4EphEF{0ar>N%fJ<L5;lq8vRvZoDb9;Cn`}}+rSqN@+#qCwiXYvwz9t*
zfU})W@9A*z%ZB0g9UsuW!T9Dzl8d#J+^U0MobLfGg2M?1%S@_)w+a3EB-rQVA#zmV
zdeXo3tM3b+Zr>yqV<KyRU#HAXotzWx(14Z>db927M?Da?5;Yk9xwy=kDa#p)%$Gup
z&q-r!np)ScFZ2&YIN%^N*yUZsxlv&4QDF9_e0_f-WB4^$eP!+BeA%^opk0O0g-17I
zohR#N>{NUK)p$&>e#Xma(9FA$16aS|Y#;owsPrl39rG$jb&G1k1J4_2)=TuL56`KE
zHk;J52W#(dqpCkM>RLRL78l@R-93HNs7H8f%H1)jO%;Q-e2SolJ%U`y2)!-{PxC2U
zy)K07=kjpP8XB%eu5evz8LrhAyC@O*Qz@?aT=BM`1#QAd3}5qZ4L)9Q%@25+SIar?
z8s75GVzA9Zdb#Jp&&<S&KAGO|G4L|O;dTasMSAlXa5n=(;2XG`_;!50a5lH$Yc7Im
zem4b6HRSifHO*?#gDxJoD%=@dS_iPrufllT^P^j3LxY05DU*`S`b2n$;cYH}ZR&t+
z_JD1!;S;|A-<BC{bNm*b@XO>m;<qWi2VH7C*CPB@lgLQIbL9fN&aD*EpR{D#!^u^=
zjxXT=oDJGk?+MhxQPdhO(T@w^wPWuv-3zD`V<y=@U7&A3ox-U@@DeY}$QcOiHIc{Z
z^@i6S+8=xqjt8e4eYa$p7{8O5i+(v|6__7vJPCYLxgA)iDm7AGpSGh-C4EAF1~tr=
z<z7u5%AE9&jRL0GT#|X75{?zVCiD!>=DbHc@!X_a<k4Sxcx}D#&l=!yYLdU3AI}6i
z1W(?AX)e1pb1T|9-kWm)w^oC{lNP0~2QOH{d-NzDqrL;*d>)4;730z&Bc74E=qvfr
z<&)u!h5Np85%2C!r(za5H3FWYE;Zn_2&eW`cj{0<aw!wyGk60xgT`86Gn&*`hbHy{
z(>U;e;<rfwmQR(Jyh^mGR6p&?fL^foAf64lo3I&nRqu}n!fe<3`qbHZ?dp}tuGNpJ
zRS&X`Q{(1i{f`dVRHBtl1xj;&Mw>Q*Z{Bi7$+T_kN%N`Shgy{o&uZ3i_#5zz1$=WC
zd~*zZ(;0kY2j5%=-(0I`QR*;@`o_0t<wvvbJm9Rg)8GvDWZOcs%8WJZb}zHeN0_y-
zo>}eju$_imxJo`|YqAtJuQK76H)&#DlX8M@l7eqmP($p6gXr_!s4F{+8t*Zxe|0cT
z2BWh4HE8xbgK{zN2JAHGk=-CuQ-eOqprbB>M(sD~-EM;xtu$zHZLm>1qf#z2=zS-H
z`s6og>bD4OJr;o{E`l7-2$?HIsCEu&ks@5C2t`^9c+r{v=tC!ekWrJ0wVLrbl3H&@
zMf`>J$nk}Hv~>+CAs9V1>!k%QJKtXFv%2V0jp)O>$2nY#zq|+W{jevb%f-IHc&d$e
zxBF_hVkdO#G@hg9@OjCnz%vbTsLM;V9(bv+Mw}G`_c4b)&YaocKZ;w_?h+gr7_QxF
zliG8>ODAyM4e;GtlTn*r7_^7i+hQW`!$@$<aNZNHmt%Q<xHeBP=+!iX{L2mca>$@{
z7Y!<M&!D|udB3vZ1*~IK6Yx;Ex$x2F!9@8?cyCRL8{jNq^vUn=-8RNXwZ7zq9mAgm
z{t~yPSc6l6CC*y#a+%=*TjJj;XjA8#R`yitM>Nft-d3fgjywZKsfV9<8hYm3K^8e$
z!yi#khQrf2Jr<3GQ)=dbv*UI@IHyi+c)|u2y>4aE&lVQW35Qot&sbZ@_g$=78A9tE
z1h(0azXxnSkN<rw2JiJ`^333Fk_-pKus=M4@4K86u7=;)k;<lZa5gi+G`)VI#l5#G
z`nOe^$p}eU$EG0uvs2)k6U>*<@Kfo>+Lg07`c_B$`OE2L!vl9GH<*SRdku4SNhhZ=
zqG@&H_FTAAGuYb(tar*7&!ye??CMUY-vv+Di(2?Z597HHkwf_oUp_dZJsM0FIKo^$
z@?chw!Ll8F^Co;?82K5c(5(u>)$B&Q?}>H~XV~ro-1bm>H97Eg%_HZ6KF?LdeA<`C
zrwiZ^0~q=%<Ln;T<po+t_3>bu)6{W?;HcMuOW<yNV3-~7Fx3rcwJqUqSOY)7UH59u
z@3vu#wk0zcUgs4*54TYZ@>r9>H6Q7}$Oq>$6`$g`Y+l9vO`Uh0vrrz>M|y|fX+Xx5
z)2CIVIG=^9JNTpr&n;a`GP1V8jW74;dIEg)JIHj{<HnCmZ{uUQtru=t-;vvy#-mPf
z86mQqo}>Ro7vZ&l!@hvc8pq~wnSU3-Q7g*vdfU?L$zxB%V?Op5JU#os5IFrh^u!*(
zUvv#0U6$TpmDvFm+=tKSBU&9e=K5o_E9#TI>Bw`yXVd38SO#p9H3lCLHB3npf8!#H
z9)76@Ul-sp_Jfn+-a$vo4NsOYr~=u7O7)5RrjP;hgglZ%yzcdQtI#O3J%URw=}^`C
zb`9P}esEvA#=~VU1n)m6kLNc%+U<L=HN0k`88%(=q9Zl7X>EPxOhcPGQe(|#%r=Qa
zALi?eHK}jl9m67RN<5o+w%?}spKZ#6CY!`+S5~<Bi}}ea<_w0~-E4}<Pp$eIY(rg<
zb{rnO_EvdWuWP`4Z$uA130HdKKeTRm+_!yLgW-Z->@;ar8Iz3EMv<qC{LH9zOQ{E0
zZ`1X|L)_k|apjD<)!U#kIShCgB4k+}p%3N=#pH?5m=EFfordeyqHt9k6t3NdaJ9MK
zMMZXY;fy%=N3LkDVQ@8{!8QxPHT_njAHmr)1>5Xd0;dCSlNnvgg)Zft4(^!*mkG8R
z19#IF?xtN|@&Vy)e34)nxSQ8Dd^%to1H8>Cyf(F4QVTQ&`=Ccnqdq7PZ!-mLuu?hl
zX^Npc<wc9iMBhG`W<@gevBcn>Bw(+2V4uI>6|hY_u+4(^WG%kJCy#g7aTl)?zPl@U
zs(v2@-|R%6S_2n1pWhvg-WWimYQ^tY3@H)#X6I+H%vCVXJ~*6Z_%X+!L9#czZ4Rb}
zzj6OY|70B1WE@VJ11@H4+3{z7sKESBPruYtd^bn&g3cy;uMau9oyglQ?N^SpWYvB3
zDf$Z8H0ZvcM&sQJkTuc<ZWdi2n4Nk84z<+-a!1gn3e5FtdOxq~!w(K>?A1{4&6C7l
zon_4#dBvlF`#dTEkKBK<2VI{YZa4lLyf^pCF^2Jcy=1S8{fsjOF4B{;8Et(MJyCwn
zJi}K!ys%qSez|n|HvSuWOg^pv-;8xB1OAX}U0k|QgMOd9;2${d+OM7JLY+5w8@1qk
z`jSRD6&}KW)5WRP@HZEz6PqW1$H8|~>k`=~8y(t)mtY58vW0ehHm%4)0pHXPbLcgi
z%<~s^wYzFpmVI`)R@pUZvYk8zG+6`l3hzx=Zt78L>NXFVTL<ubE@2NCfnE!L^O}93
zXIY!J!rlMJT0eok>E#yY$z1AW@Xc>8T5dbEDr$kOX_-53Es9HB{cMOuGwl|gZU(<l
z9xf*nno=w<&pUh{x6Pc>ZRQM4vufbaj~ay@6@zc2omp!unRP6YSxx9aY<ro!@9pG8
zEhYPMEP8KmxSR<5`hqJZFKQqhMXBrPE^~~!Z8fT9aidPA0js<+a8|QH*XJ5kGuogL
z@G>!V3>seAKrgI8572|_T#C>SuCK8Tstl$nQ_rAMbq&f3cT@X$gf4H3(CK~=D$^lC
z*T6OHl1J$A*Kp-a6rmSC!c~bYynTemmyJ+|6cMWXEnKl*h12sEuEq%>bTuu1S1m$=
z!ANzFN2u7V2+gK<^4AorO!Py=TE`ga<5pNg{91O8vim$rPhYhiziGZ=ZnZ|6O2+R+
zq`<2+*r~4gt|wl@yS<IvQTEycPp#_R2aX|=RUb>E;ec;8y)h{SS2Y3e-znCk*?o;<
z)W9F#H*lteLG{2fBZnGPdWb<gxjqatC=Gb0BN(R`d{x?|2K_l;(CPE&=Xd|TPhg=e
zMg_o)E$v2an!);Z8h;?T@@Qw1=C(HL_i2lspba@MT67mps|IW3y|`eO>u`GDg4xv>
zOW?7O|G^CwW!+8!7eWnrf1OoFcrUupqct%*{ap7gy1CDyNs}#dfn(m(uxLvjIGhjQ
z;eL#nU(}I|gI$$OS`9D$cb`$oMuAP~)yUAnsL#EP>cN=F&3QM?rtlby!A@Z9B=mYT
z=QVr>_xAp2kTsc61sm|Uyf<sWe+$s+YT+q}T5pn<vG?b)S!UkPdPmUdN?FyPu|6)D
zP35YPg9d)vhc9{!HFU?0cFku#<-!k@r2`z!B8M{FMmtRH)Xc(o=ju9DwvkiC&_YXa
z#YR`|)6J=7^m+}2-&uv%?o*UY&VAgnK1^p%kMwv{lNu_j25ZoLuQKBoy!#6udwueO
zM#3p<g&%lE?T7A^;wkIn+MvqSMN>(@wTZpA2{j;G_S^=nhwuf%u6a4XgKHSIn%}Fz
zV3LVooaOK_HwU7>!m+eKbMkY0YaG~y`!0in9-z;bZ;19pzfR@`UM&LWq-o_<IWSH_
z@J$==jjZt7PH+ymW+spQp0CrC0TZF$7kWUgH`=4FjOz>K$*sub(P6Mk-4*l!&`TOc
zuSHy=Taz4aoup4NFpz92YQfJ-$YP={tZ|gA@~duTyhjGcD>u3bJ<;?yCbfEWYN|&I
z$bUI-n|_YioWlfHbaf?Xy}a@2E_~Eu_?&vI??+kdOTZD;DB#z?-hOS`<yT$an-g$D
z+gicFgrh%s{+F?{1w8T!J|-+7XWZ~!reMFB$6hnA2cG2i0W>T)FSMhZ1p>OA7yQD$
zlduK(5dF#HSx!a`{k-Qh(I5T*Ow^5@uuk-AQCpAiM6H4^FtQzWsm-BS<Wns_Zdchw
z)SG?m^3WS_t0+AIc<IK#otwcZd!|uGdgyIxjeks>5mks<fy+{kGZdId$LrdZj<t9P
zbE_J@x0hXPs?Z(nbq>CQwP>q1_?zYMyOV4x>)`Cnb~Y`m0Z)n7Ej?Uo3w*_$`-5$o
z;HAvS*Doy8UDOKonR9c@T2<Yw?C4QlXPZ>Fqe*twlJ0x(4-Yr$Pq<O<)8Qj#jW{#P
zpfyf|es$ogVNmb%Xj4}rbbm^OY@Mh9(?saz`)~zLhAS*8T=8p$E9s#wy0yED8a3~t
zt{c#rxDs*|1=}0~+gt_PJO|sPSpoL~Z_@+b=EnlGC3u_l=u-Re-LwSX6dg_WIo!<|
zxSLUMHznY1=Ht6b<pLX-c}%pZKdrz=jnJRKHA5SsQ^9*CuNkQV_-%?-1XmU3oT~!#
zY-J@QJvF_4V4Kunn_cn2Q*r3=j)lh<Y||6J%_IE7`|#XU#B(zm?&dt)&EqRz8?>n$
zhtZ_)-85Q>?l=pNF<D)mef)k4bma=1xko>o^#lD<SMV$B!PB*vJg+h2)&<EEX^7_^
ze^=KZ0kWUT9RqJ42j47W+%@Y+<`W+Jlc~Yh;G4Vq$nK+Gs(*}M@9>DG#aosLeDg1#
zrV98b&ss8As2$Gu_#VGo?FwkOc+#tU^{VJ4{KOlWo76DxdQpEcXB^Dy5+$e+;!~TT
zg%(7QDvUp(#ZvmF;BJ1$a7L1ud4@I>iMO~q8v1~*ZY`tkXnX-}VI%$<@Xb(<TSMx)
zwFX~Uim!NVZqUoan$~bB`7Wc;sk-9D>Eu$6>f|z^;g?H<FBWg{@>5P-#E;m2mQ(45
zQS$|z%GvpU_$Ds+=3{JnZ=X7J{WKSTvg4!a5ATjPWnmsQai}A;!g~6<BmUS`{JC9q
zFWa?p7h3ExJG_xyv%A@Kk{Yx|b$WHf@JIZ$smfiOe(tkrG@R#JYK0U&*76qM9q`RE
z=F~Iz`mZ<e0>j^QoNZNM=1Fs#RhyVAM{`*<g5KWI_-{H>LpMgxy<xU!9kumbYVabN
z@lX7P^LWF#Y&Xp6c+{+s_+jJB<-Wo6BRa@esEhw2pBYaVn1^i4+-OrN*1$zfGO61z
z`VpNb{iuxglEtLU@TJMoOp>rirKgriOC9kzKAh<rgKQ_@CfKhJ^<jUu!d)~*b1G@j
zu;d2a2isU~a~+A$^R*G0I444T(2F0!(?s=(;GFshJsJ?9slEu6Mnm3QAVOvTgp)x9
zZsB^gCtTAv!MlzR*Vh)|>}BCvkThK1zje{Gzg-lQ5${}^aP5f+mw#3`XS9d&n!!bi
z&>^r=Wv5X=bfT(@d41nZO1GSBPhQV+?wbaNs0{`>5@}S#c%v?{t~>zq)f`E`HoQ&B
z6!f97{{7<`-m~w9x8U<QXVKUyaAxpc3pwXyAN*G8L?&gUui;lu@<Q{GLvaa=x5j|K
z08GPsazBQzskvGYGN><j=h6s+jN=VDFvGx^*YH?yJ)gli{a?cQ#HRnDEWMrvdO#<V
z&vlYM&=lx~txQ_75DZ$IGll6N{rJSJ%z5Ft!6rpk@SL*Hw}nR3VLmwcF5aE2_;YyP
zVR(|yoPfJoZY8fD{ir%PCn4wh!THtRXi=PDc#Jz(G@}F_V#Z4fywN4lE7ybFCQN}Z
zFUQ%#uZ$YBjSP)pM)@6NYcvMORV6=+*P19PJsfCl-L4rl^QJ*pzrokt=JnwZDDere
zV?y57YDU?>H$Ao+HG}u+zed!l|Hsl<hDEi%UD(9#!tU;Fmz~(j%<M^Gm>JSO>anq~
zyA``Tu)Djv<JjHx-d_Lr!@f`y6qx<n>sj?oaOx3#9!FZbR0zFv1R0V;S@Frk1J#Ac
znGQDV4Bp?%Z2kH*exBl?I?R0FfD?U_%&kDQhMiN~@Gx#=c|{H|I~m<y@TgDR+Vz#+
zSHiCxiT7r9n3~;(<KbTKWFtqp1sqH6VDt%m7T_EES)b0<gv&b*kAi;Q8eM%0=UG?I
zx1{(}9PocL;Ic=f*<a*rERF7Y1+L)^&)Oq?oRQ<`>SCr_kQLtxy3<ndO>KCYy<m{9
z&TuUQr|dy{ngZuC39h;f*k|xScBR0>__rZv2BtYzpNs{1d`kWwLmHE*XhpuFJzW`a
zI=8td@Hq<bzW8S7`Wex#;A_g`J8X%rnxX?=2iWBSytO^XtCt?H&Un%4d0RT@f4=!I
z9qIB4q^k;?Gm}5__M?-0ESQF_F#9^Vn|*lC$Y<mN^X1|^-AiWUS4HrWg`O|E#d0k2
z={>XNj7Rt%{pixF@jp%MBRFT!ANs%G=Eua+)wzW(kk??H81@COqo?&QKA$A?Se1;@
zc)0)VP4Lo@wc0QUUk#bHoLs$mwmbmS6a@QZ?G&YTV42=o@j#?QL%;)cZcU_G9l#rX
z0WXk^j~P!VW>T_JXbivk_~noAORpp|wlzYp-g$KlUOuu;nDXNp=%1F%2blQ$9M0fD
zp~}%FR2%Yzva6a*BDlE|JdD85v%xF{n}ujgc{C_6%sH;i@GtL6kzeErVRneP7p#F`
z7B_Q4cN4G69?U(%@z})i`iu?H)_!DDZ6Q3zLi8DIl<G0v3fy-MS8`8!(4}g?aWa!+
z9vGxVyYNJUYa-a08MVzRD|6&JJj6$tA4kq7(+##s!VIzfid{9C8EZt^7249S1x4+$
zah`VXWK$^I%<MW=&L6A%(4zKnPTu95tR81o*la_~q7A+2V6YFwP?ob6m0b@W;o1nU
zdA^n|KCn${G^ti#n?d+(dcfV(hP!zLcT;CRJ$vX<qbHN6g1ZSC4OWRoZyJPm<{!Ml
zJ-{Oo;2CtOat^wp@er44&3<w8s3Z;OPQYj5tBhY5OjD-<I#(%rb>MCMbHLevZ7zXr
z)}u#_2HUg%+blwpy8M&<s$cO3gKYx9Hpjp=xgN%-D%{Q2tMn$p-8?&iC+7gZestpM
z%lNx9!8Ukr9{S*I+Q73`icvyJ@>6f<cDjrYYZpCAi_nVW@E7A-7y-|}Hy>R8S2DUc
z@PxtdmzdA%sL%hhb9qYBo%#*l@e=!!=ts44Z!gEMc7pjOybQg4blD7hM+e=72(nD%
z81bZ44@Zw`AECT(nZ+{EU-yn(((ttP*OHH#f@b^=vxx(bG57iK(&0Lklv(GwPe+-n
zYVCxt;ohptY_gi1-X63mb2Fc=l%WeB4)rp#x8tE#xzBK3aj#xuR(WIn-%iLHS-J1u
z;w>h(lXDL~#04H598JzAio7!(()9Q&^5^qt)SoaV#gmx$7}=?H_%3F$+dei-HNZEE
zOko;U54@8*O!dE!Q@P7Lz<GKYZafO_P3N9;)0o|=L`LLMS-hy3>8<%1s=JRu)#hxd
zTnXGuWN$u`YdMMEvljg5YPiuExkB~*D}3iIGR4dgnHO<ijpSmE$kh&hv@-Jp^TMRB
zc<s5jsw8mDVICM5tSfjOMv=KWj6bv`IjEz|%(1KS7UPjf1HN&0V1@?Y81OfNJg;Uy
z2fti*>EU5!e>}0z@Dg8RKEKR!d=m5X!b135lalkf&wa)m5y<mDVw6)8T~3A3wJ`7p
z-kOVCYv5@HGk-Hv+^_A>R_60&ct#eVwrkTSG?%IHHT~=w>#*xvbGxz?wCh+ByY4==
z>0hwT<pnl&ecO}=4yGttu|NEa<)~FTXIfQekX7@etz?IoKj2n7@xHl(t=iSpDswff
zym_qJlGdut{#N}=YE{*D1|1-V+P^n+C^;Wr0RJ2xTXBXJ{mIH6WvfoW|9C6f)T*OR
zhtJy7;ipZ%OWAej0`m_(mz4b9NBY~<bfZnXB5j)2#wMe-O$Qs<l&?HI5^qa^k8&^N
z&&_aC;E*|Zhw4N+wWyv;IjX~n@%`Fmc4{u?()EK5)$Qw0Y!<X7@ZP3HXi~%Z{`BP8
z&vmyC?*r=$8)8>gFwRi0PobH1bzN-N<PCP|xLy9>gx25eDqI>I(FGiVr>V>_a3q*u
zA0D6SbDa7J*LL0Jk{?*L$WyqF^mvldj*f6{_9KVh7=C6n=jxa{!Mc1tSRKgR^`1vw
zeH?l9;9zYn5BB+rzO*?=xn~BcPyZnK2jI2I6Qp6=;P`sDG=tYf4A|pYL#NU)Pd>lq
z(D>C3{K4=roVz!gqEA+Ls7P+KxIgev;F?x^U(0b#<7?=8mapX&T;5YVyRaSVS`*z4
z>~|@F94mf;g-v*kjf2AhL%nF_QW^O9zAIe1_YIEzBc25O%8j?72kxOe2yguqu>J9?
z_=Vu2!ni+fO~uRp0dF$?h}wHYb%y6y&Bol1qsY#0b8Gl%et*)f%8%%r8W5&Q%h8(8
zhS9f#Und)xzvUji?#k|Gp5X`RyX;(%PN9?Zd}k+@wIf2Jmn$?fQX}WF4`wI2`#HMS
zZ?gaH9=_T8aMstzi*BXEv?DVep2G_CrR+%0?i6OWg7e6!qvfQ8yLkyGZGwyOhlg3u
z4EL}*SOBeRVhp{E{JtHSr9(rmx@cBa@dp>D6KWJ1(_OEUpdIC!1V0TQ?XK<B#L`}k
zEaX*NI;JvS_h`Xm`ZMvs4y1$1nFEZ%pJ%B=ClS}};$HG+a5Z)4@r0`iYfGNA8`=~2
zCJneJp*NkSBfaW99lmA>c~h>m{CmF+!F^xh<KEM;L4G4S-uod#(6Z)o?s67yfG3Jb
zhL%?e-w=G#lojDh{FZYZT)4eYgr?z1n2!$pFLQ2zcs!bP$yoMB>e`=3_8&4I7bdfk
zkM4LcU7lc>=j3lqd~Z`921_0x+r{6{m@`Vg2Xv?5xx3Vcyh<4FE6Ddg54(wvMrdy!
z-KLN6Fg*u1mLZEvX1)O4&i-lHAsh=Y#g5C9{$aY`-mMfV$nEV9RqQmb0rX-7gsLo_
zn_}Q)_a6L>6Y$XVg5PWde^Uz&WS$W9{0vWf3QWRG`Ew2P0I$Uo|KNMvk1ou8HL4(e
z<Y1wa!68Z#h0lKw^C%dpJD(@CVu-R-2+`EH>|B7i$=-zf@KBI~CkJUMUdYlXTv|GU
zxrB^VpK>nt?m2Y@eY+NXX1WyMeP)Jc%n2e7H6gPd9+bJLD;#e-yNXt`D>diUA-jzp
zcB@Wv23kfL?4z;h!)1&59=2%u0gJL9vnc6liw0e_X!Ko+R-U)$`ZkMRfM-r}MRV;1
z*Q|!OX$ZDi3%2<Iw#fo_lV}k*1n#EuJn~O?Zn{mxKQopbIM`;w5c>801qbz^n<<9R
z;YD)_rV9t|=2AO$*EOZD2Ck+<HF%!#=vyVos};oGgMWAj*Z0iy38tZMkt-qvo*l4F
zYOu{KuuUF#o6h)%|KS=7wi$`%CIM{o0Zpp?u^7GC$IgXaF`5W(Q+P9(MSR6K;^|SE
zNS{Q1dI!h^6lxNqn#JhK_5&M{V|;jlo>%zS40z?<kA}-ZcOKoCY;7?*nec%fVMdv|
z3(pPw&BJK4Q0AC=`RQ)>j4ur=9k@19o5n<Hd@%d2n?))GKKn0e@aDgX(3d0lBi7Iz
z&ui}=4?QFJ-R73Xr=BiC);Hnm#aZIN8f-HOu9N${FrH$Ic+34gp84S1Q+zheRpqyF
zEu`CKv`;B}!sEa{wy(!5QQW7c>3tfGR-BEq;QMB;jM;SS_V6lYfLB*4&^Mfc-9XRj
znLI`2mziz<Tynp$9(Lf76KzYcYejlm{5`7o16}Gay8QvVZ^&C@nI5Jo!@xAmeh<;|
zt5t*R%tmkDC%5+8;(j3y)N2_XefTrpN4nJvFF@NS^v9H;`zD=RPd|ss{19*PiBR^F
zg=+9TG~bc<)V%z@6Fd%j?ar+9u6?1G=4OZ-;G2u_%mT~+gTnCL1aRMQkBrSihKhTw
zB<K1x&iYxi!87P1Nu9we(<)fS(WZ9cp;&UBxqBlS7v}J@;dGcZ<oSTkaGK{{Wu6z4
zc?K4^>Oym;3v-%FKgdiicj9MbZoW~4XJi59h(AtEg};f~>(u?(PB?6*{%J!#u{b)_
z2mG<8c-}5^C>feW2a`j+;1>pFLL)|(Dt^+g2%gWcC!q)Tvg<oo;2!*G@7#7}{PjQX
zW=euhB`4dIHpZs;t!!F`w<hqFRj-*jevAd%gjltyhE=CBTh--1L(ap7dag3m=!Bu>
zyZH58Lk|*LmG7IOIu8w%z%MuYw4qr?3|+lxDCuiM@10hq1{3{W$mikn&Oe8ymB~ux
zi5X_P0fsV^p$FU#{B8SYa7Qih{N%T(MshIBHLJ#ejSh4|yXN2eXR)Ddc0;{N89JE8
z(9(<sdu|NvpNUV8EJ7fe%g$h*UF-Nfnb6`o+LhG>E(t|@^V;>Qm0c+-+jSwYT@(59
zrhFd`g0G4swab^*t}AWeZid>`aR%R`c)U03!90h-Lg0h%a5$ec!YwsHrvfh>J>XDa
zBBz4uJ9Pq2)X`r~cG2RWnh8!l?^4)zc)5~6n&%AChq>TfIK6|M$z%JGZ;Ir#V#YJn
zAeh%Y9q{$=xD*BB9K}boBuH&W;~xeO$7jatcn_WrZf~mLqTj%!x<i~QiWeX+*lO81
zhsH2hChZNU)DE7CbNG63hs@x_fWLSKY?cuY>d+Ce?H>3kuD<)x;jY@{{%qIq$_~B4
zPc(HWUkg0oaK4t$d@Y5*)ipc7MfD|1ACGo=9M3kd+2VMdi{b%Fheu%{d3DDHxX6_F
za_R)DejoTKo<D{0@R*pH{qax?uM4MxmRK_je*cE#81cmYCWl^mFCOIYbTu%~*K?6Q
z-yf#K^c+lQuK7jA;PP&dO!>T;&YaYBD?M9Dc{cBg&^OMqc0D3hW)kxoIA-o~^rtKI
zu;a<qMZE4jj~ns+)M8&uQu@|Ukex_NU(I^-Blw%<;FxykO^xBC-!uo~wC3uDrh{K5
z5?$%BhuokAeg@v=LXmKIZ9K%n??TUo(T*J<>A@W@@I&wSsN7C;^rarMEFQ2w9bIea
zm)p+&pZ9M+Ohy9j{VQkOd$jnHV4b{A=x3mhYD^mR*kWjK<N}A`NBjtfvmS4CgpIf8
zRC_|fF#P&9cxUb)@DO^H4UIqStXDN(F)Nnx=~8FrNP5QR%<`!TXYqu)=($Po6jdO@
zj_zh#NyhbQIG!<bu_5fcn+uP18UIm=NX@H)H`BrQZZUcBtC8ya18*jr%&XK<%7ZT_
zNh11U$(6O)8;QRV&vQbgYV6?|!uPe$GCE+RB6W*7*{>fx?RXnZ%_8&+|J;1A;guIY
z&OGvRbLbJJ7hwJjW?^vZ6?~QZ;NTZ0XNPxdx6=H<uLb^%fIr{cGgME>c8o&n9Q*`N
z+D32)xTFwg?j-bQZ+^7hFTpx@J6H#Hli^?<c;5v-Srz7l?DRb`6TCMCYw;bl;36Sv
z6bP@|nXa<hc#`vyS^SO`bsWDy=U`O_&-4%G-ogt|j+wF$v&c#2Nmr0d1!LfS>Fuji
z%q4&HU1OG0^~yT+Wrjn?I8#m8?fMTL|2um9(Y-cxJZjS$bg0rNZ3^3L)0CWeiYr^$
z<;}TQ$k6g97Jc4nk!OKL-f0&023zQq<Y_v?BF93DhAp>f>N1Pgf@dah)#q9UuGxv-
zW(e3O`!YPfV4Iy_n`3Y{7vOGsqDvixySX=>EEBrar(y6lgTW>J(W!cM*Plo{IAQFR
zbkdg)2%po2eVL8%2jjD;0B;isU(>!2J<;Hr`C0Jqq~p)PG^PE>kArD4CxzQVkD5dV
zJj1sb4gM6PW_XGHpTo=Gw@Ld5FY%2SRX~qQdXOFRyV0Lku-|<i+SMGq#N*JcdXc9x
z<HKndqtsRC!zlollaBuCFLd5KqKklRxNkH5x%u=4jEY8!j#fi5ciX`?Kk<Owy@t<b
zXOvdY#rqx!)^3bOm6tsw@45HSMCur?!A*n7hTAy<8uGrP^xlAPw!UP)0A90k_-`~U
zLbu#_?(w^2D@hlQAHM$k^f>Ir`yUUM8I6V;1^?O=-iAE~afR^L{r2hFLm!!LUc;O5
z&CN$!9!YKrFW~BKc-?CGbgUrwCWTKP?!$IBy_&F|-PNPnlN7?P$A;`dEQb#Vd^7MN
zUHeDjqt}CR$dCRzguRI&<bOp!3m&v4nbE1f(QQChXUHBhr%UNco5GwoI7|_A9W23j
zzZfk)SEev!de3Zl&8^(zHcl*cE9WqHoCx+0;RV>j>{yYVM3<Q(=YAr`dJiu#Il>zo
zLsc1Wv{x*?8?sYA{Po=`;WY-|^!XHmXNvm-eX2b8X7w;~waf&M;WrP#&EHKA$H^Qp
zfEhew3q7;b>Gh8a)(x=Dqmtm9RP1MiyScZW&KsTqAwAJ3&>8mB2-51DL8|?m95~qK
z%0)8a@S$fG!2dHlEaLu}&i!?ynTt+%d~JDLN{ZLw?n5xn4l++@5?OpsHA6!PD(IB(
zUYt1YQ0YYug@JD#btN~2HuW#};xzD0kz?>QtC`a$+SMk;t~JaX_PTaGWB#ZPml}KC
zhMs4Wdz?*|JY*Of@~rnqpL$@`$zAX?qreqGR{a9s6ishcbMQ;RcJKvQBBHOMr%ppF
zCm1R_z|ijQ;Fw8<`V2PY=xUIkGUT>cRI9v2?J`)Dm@BKFMTI__6|vZ?b@R*$7;o0i
zk!HpHYi2jLSqo~LH71W)?Q)p)x~Lib!K}WU%zCrhqSYM@RV@v!;><WT&!UTFi!#-<
z&^v6=jWwJb&n$H18+df!i*|#fQrMKDu8n;kHf353Zo~8A2GfiK+r=iqH_Z8DUSng&
zq)j#Vb4H!AsoO>V4%hX&Hr=>tQ}`{;vNv!`Ihoml?OKQLde(2e*%ci6;X@PJ3Lo?Z
zo}{`{V<tJZ;=PmSIhd6^>4fd{xIe%<l`cqC(er!4;e_mmk9h`<_Y3b8I#e&Nh%Z6p
zxq>v4S@Qm3{5$w|eg&azk@qi-*0$3xNbio&F~;lTOH=ycFF2LGr&Ci(ITiWap*edU
zvQCCi!h7?qk%OLWhjx5Nk9v&W27l+;Y1dLXyl%V>>VwtVuY&K|XjeRVr{go`&w>uT
z?0nn?u<=VWcHo<{SK#9F(MQmZX8`ynEzg2Nc*MHDb?Ij@{FhC^YN2>~rhwtm1b-x<
z_Zf~q0u3(HLGoj0Z%grdo_c}rp$_*N^TZbB^|0#ng2Sn1;r@Jsj{NM#-%ctxTDaJu
z+}r(cGpmr9{8h=L;uidPJv~Yo<k1%Tv7Yo~&Z-=t<b}!pbdS_-dhiNuWS`zqcpPxe
zP%zFyIEL2fC4IKwM=r~(SC?L|ui%!6c#8YLA)qUzLr3z$%M@))28wGwXWQ4z<et)@
zJLRSG5T1HnQ7}&qa!}v!0Kt<yx#QJ;ec)*dfJHunQO<kRj%RqrspKH&wHV9!b&eVB
zEZo!a4IcL3fNc^yntl+?eHvYJ6X+}*%1)8N9y0Ls{opm+I*S|yXW!TV(B{EE-#HUo
z=BG0wGki@Ruiln|%W1%Q3Dzky7=I4u==@{&i@{lIGtmo0cWU?UKJAP1X*NE{8Yl21
zb3VtEpu5M%wTOK$_rN;Ym|t5*C~zDdAe$pJ3V%ZClz94iE<6P%PT+f02>xsCGxAkm
z;rhSOkHfDI!`b9Ve=5J4=LgsHIec&1^BjqYWbX}szgL7>JfasUHMsLxxDMlI__WTa
zB;Pz5!PlI%FCC(z;IJnC&Bf8*nt`s?5n=LiM)#r*^T|ifTD--I!}AzP)(P)pch26$
zr{Qn#0j$Z0hdnvH-uT<ICgbdV0k;Wu{$~TbpnAcPHwxCfoXnW-=}EZCJ+du`-Yxn#
zKj70p4feqkxvn?fGq}&=@H30J7fqwcaSV28BAKX6-{D-JIkg`hs%?Fj_JQS(nVCtL
zEgxp*^?s9`--GZq(kpQVE;LaMyH4RntjW1}?k?xy5AsiBy`SE&(*0yrUe3TloImXv
z8ERI;(8oU(ow;DqHqIu0&Z|w6Eb2DaqW<G7dOVr4a299Re2enLThs?UW8=!lwE$dm
z9b9t?-X<1olL=kQ8_zCHxSOSLH?ybX%bA2H2W_g$Nbm^wCJb$=HQH2{?s$yRrW%C8
z{n+R$?#jN*R^53WvG)O=&1&?hyl^*(3XsP_k6N1%UvX-*EHF*VpX{_lgX#)ja~xb#
z{|jDVIGZrEsQX}=mUwLj+=*d#DE*HY@#35W%fQ|EZKv;(tW+r2X5pk5nTN4oK7uYk
zJjFBX(bZTwMg?-<%}Int^#IQ~*yis}>~Ra&l{SP-xQBfnWT&d5Eic8_o`T$zb4`@`
z&Y+)oSd<3hy~$11Zz$L%KliqEDZMvw?DONbySy8(;kvvYITK!|;&pl_LZ$c9eZPW!
zP57HaF1q|0;ju5wPQ@e<ipSqSbQj%8^YIpsARmAZotL~o`|9Y*IpK7E_;mLkIEH(v
z{049g9{!;N=qSePwmE>^t?;G`v-{9ge9G{gIpiEV)LJ+k_}2I6&0Ff@y8+)UU<P{q
zz@u)=H!;jSeI{{EkPV;Inf}STWTuLHR4J23Z@-fZ2j86BL4O$8DLdIIOMmbV^Innw
zdMV4oX{RAuMK(Mg*{Sh#q8x|6Svi>Qn{ak3qVsNGW=v7ct!TP$X1ov8pWC5y-@?<7
zoeCy9)naI<o{*gy%>0nP0$3*#9O=gp1zZL9!;>ys$j-H)=u$!CgYfV7FAYzcHU#`e
z*FX2sCp6sCqrf>q^uD1%tSruKj;FNalOR=3Am_r|UauE^(Qfq8R1eYyf4VW5>uWNT
zZ$9c$SvqelQ+SRHgctR=<mrO8(h#pjS)L_XnAyMmZy#;;wd|oC>!i;BE~l|mGx9Q9
zymP1v8cCT2XfxoO1>l>$)p+jWD-HwSOd$WWm*-hA@J)Oa&-6~<oLbEA8Noh0x8qN6
zt+c6MtW9OW0Yz%!iAigdhx<2w0-Dn>E8Z?EnzdEWQ{ex<XK4FoLkp(Bm-H}H56n@!
zs3Frgi_#pos2Ov~@?eWv*XAD1YSF)+%-Vj{tQM=x%0I+RUe}DD&8$25&B~YFtYHPs
z^vjraJjSd;UCi2;%dGfOCj9*--L7g<htekX$!F4oj3%W`Y0@~Z>;WdtK5SCi-DcVO
z{GFOuw7_Rk?RyqwDFMFOV5s<TtB!+jCe*^`W7zO*gHfKKb0vqLNnzJY{6x!qHVqAB
z<_fkc<sO@wkr7*d)~0AM-#+Fs`w<(s8r*Z*rgg9IVB@tO6<}A-g?N+EeeMUs-=GTy
zK0*^Ljc*s+IeocPH}5&spYw46=jZ}DjRIeBURDf}pBMkt#2_`@8Ki$-qCI`5(*~S#
zfM5T(&fdN~L8?ECo=39Z_uy?tf^9P82Jd`v>E>2;`So(?E3b*+cmj6gEBDWXr}+kR
z<tDQ2cmkT?35tTdnVbR7BR~6l7wuXNSF-|4lY1rEuzCFK(cNZDw<}?;U1hfMwZOOE
z@^{E#rY9bJljf;IHS5CNfN%1HZyM51FtQJL4g6h%*XlU@VkhAr)8cpj=m6t@#R`(i
zKm8*}hsg}iL4WHApLvL9Q^ROF4)KXb*Jg*|8Zx@y(SDgT%ArlX>_wO0EVp9m;GE4o
zzpFJ}lSAn=N+9>|$6mEkU;uRL_3*igI2Tg#oNNoG`1+LHb6}9&Xe}+#+83qg`CN=1
zjOzG5`l7`TB3nH$QUh9`OTkNrZSkqmaCYCpFXx`^Q_-tF#o?p*gr8;z{7f0)w<${}
zJDG`}=X~_q`{*6;>1{Hf)}gJ>IRd6xif3@5R~fr|6%C)X<by|r;bf-5*X)la+uo07
zcMSOmFYhyfXUMu2<Q&Vz+k`-Hj+4Als7EIv;iLMX!86Cz9m_tNasS&fH;m7-7ff^w
zK4{@Z_#e2TJKw=d8NIqy6|RPJa}u7M-YfCi@1<k>5<Yw|PBH$ES>emmh5Gb;fls!>
z^!Rc<R|A`c!9%5ltD1f_Tr<$-iZ&)oLT~c-C1gq9v)cWjLm3Xo2*SqzP7IqFspfbS
zPJsnW!{b+L7@<l-Beaxf$X}l4^=JIa`TzD)!Q0g1eL3j-L@%6-AN$yB{w=WIOR(7J
zEA(}P<(l(#Hs$O8+d;pLw}(oC?H-2dkAp5V&gWg6!CgC|T^1#?`W7F}L2}*rZ5k(r
zD^Em!fgf5D*rwDsJdK}%wf6>RE7&H@4tB=T#j)KMtR*Fa)rmZJhb{CbFf;ZgFW8%m
zVHfaB7J4^Y+yb-gbg6Nai~c_}|AKUdrN%R#%%x;sm=B-eo7>}59k}siHC&n&z)aHK
zMF)gSO@}f^K8F81Y{ScE)s&wGSr@Aw=Cdj*+;j>!XG?EG74ZgJs~S2|5MDd0p?)b0
zeg0(8&HEPZzGP8LxSR8<EDD}ukpXv;W~4>)N25pa>(rc;e!RaL7^fL{rW#iYuBG6b
z<Xi>7Hh<u4R)KAb!rcVH-E^D_FY_1QpiMnMn_54D-BLsF)u2svA~#h6ZR)m<zZZ-b
z$6yzHCp4*+^bw2Ci#Amq&&`IS@H%<Xp0a{>(x6*`X$r#8oFNz0<^x;|eKuLppq_wf
zI&(Egi<)vD%mTKl18;MX9QY8pn-@pfvjKNA4ZqDW^r%6zz(Nz?dIqrvCKAtayBM;z
z<QmbX(vX|F@q>KhJ^BWYqD`%*lX@oI*aPVg2_`3=gMB1;=$yC6-Qs1xfKUDNgecaf
z;C+X?iNhln{2A>TU;HL?=|{ui9KqT7!P^lPBUL{Gn)EC7S6_lp1jC=7gWrpN42SIz
z3Z>gGalr^xO~ihNTVx9o;BL4#ZVhLK@iOx;@66$zNrpD{9PY8uHT?hkeQLgp`GL-x
z(Rgoi+I*VToZZ*(rvGIm3;Bbdn<sQ;pTPGIzR5HS%{t7hFWdu@nM<yc4c`I2i9F;{
zmsRW+WS)uTp6JN=VAS!zuXDzM_m_RfV{;Q<G1;lE^kwuN1BXNIq8#^0<tBI&OYr{0
zbo;$?YcJWUFFWXZg1?zI2t4C+t6Nt$xqEz$Md`+&`=-)6JZv|~OyPr^y)IODX5eqb
zyFU?cWT)n#?1Bl^x^#5mz^h-qOb&J@-r_lQul);OV-3;H24riC;WbXhub<QBkLH_&
z{)8*B!8-3iduWW0g8Qxz^K3o5#b3$5w4W11?+-X9Fi4%qPBq98ME?xl{JSm<W-edL
z&NeF^&f@)<FYpyFB!}a!hv%V$OY_sY^y!^bPsxRQ@!nJ&?bLu!x}?B2_S{Z2eCtrh
z!{`xn9ctMRe+l~3QuKx-@Ha(x#<e(rFJ*;Y=XtL6jJBg?^E|6!*Prw}=l&z#eUN^J
zc$*5~y=j3j?g*Zk+hn5l-$V<FM=Qd6bF~Xz8+^k7cx$%bH`E*aFkrl)<B{N+)^IrR
zH0eKEv<QuPFx+czlSQQ}!@Fj*sOnp@ik~;DPP|#$`kIxby;(gff?<A}RPG+v9+O^8
zGihVANp-rJ^rN;(c}kg-&)-CT!=$^P1C{=DpvwOjs8UY?_49e4*8B+6r?e(|L``H2
z!8#XA%J#^lrUfi2aLb}QWef!ep_vUgv~Q`Q#77O?MrW(THM6f(d%IgzcCl4S_VQ~y
zMsxOAwe+o30|wa0<KmwN*GxG?Cl|lIx5K8L{JzUWywz##%7M1mc?A79N9{^q5$`j&
zqAL2}Kz<f^?N0rK&n|PxsZaQD7KPI(1n+Sko$9Z~l&mwm7W&gsvzl}GN|4H+Urzmu
z_k3&+J{gxj!j0@O$kmy7tptF1DuIVGkr97^R*5gGIM~-&z=Z~n7FM55_@qv?IqJ}J
zxVu(h+nb#nTEqY8Y;txLKcu(kq+Q`#?dkv?Tgv~r=|H>k{bSeDf6*%W{j~*l)ft0M
z*9{)Wfu8_PZsuA$vx{A~-@;Rs0qdZ56`4&A?-{&KMRZYKgDH6~g!Y72<h45;U)kfe
zaQ$$V0j2P1g7J5Q_nYx-Y6iYpbCX>qwL;Y#pWCme@TLBAoY0Yz1rDd&Sh^fm;W_5s
zJx@2FKVF=KmF#gNE9rjEZun##x;NM##AQo_zVkOj$=^9qfljwO>`3H(zkSuMG5C0H
z7N<w0F1j@{&`7cfMj^79oGX8MkbT$Tj~za>o$k~0<LoO+gQpNbP4jAKP<VTmaJDt2
z@4~##rycz{&x-o=D3MS3@6+vc0N)E5%e*n{c8vtz<o0U7J1`Ac#D)Iyd>FGBJjKT-
zauLDof&<T#2HVW-0G{EB<@ui7&RG@&KJ&r%ke}%GuSbJNf`j6C_D|zl1O|e~*~8~d
z3BJjMZl48wlbso`cNKUXd^xv!(UBU@zCJXn_ZPg{0nTfc81AMz8F4!u(P&iPcfqaW
z!5LMM?0P$Nq9Np|w$o9Vm>v*#uO?(=FVBw9L~>N+&cNZIFO|cO(-@v>|8aWF?BKp~
za96d$wdn*tOhxF|WV(me(hbPp?eRWB|M^9#fj`fei|nHqz`qgiQ*0A2I=WY7!E7)1
zJ{;`C+0OZWs~R1Twc+T?^ZSGJIpQfOyO?>C4uMq1=zgk5MipE$+KkU0PP{Cfcso3e
zVO$S?;_u?^P|nyIc!x{Fli!9HOtd0czh~0N*E(4D$SIztL*c^*?lI=YvOip^%6$~_
z-o?%<wEmMWZCt}W*v+Lz)!}Q%crW_l)F|e~Avee!>>>|PkM7S~<T99NO2FrKTZDh6
zKXcJ7&O3aAE79Wv-xz9xmvGq+gML3la}o?i{XPGp(aKvIdRqydo-=TFCPVXnSrqol
zqUgI8T{>mavIKs=(xUuv7Uk<>QF*ULZJ5P=MOx_Ivyh7h3(d4BCs^jM$L2d+O#)Y4
zu7O~iAz+&}V4D|Uo5pZAxyVf&1>ZcMz!gVU3VgE>ZK~=3Fb&?D&e3>lyma`vz)NPd
ztPXgKo0C1SL-%kMI=|7Tx)$co$Voj(1*Zd7QweNSki56$Rg7Y}0>C!0V44nK8Xwn!
zTQM4cGe%p%HCBAYb?LY{O9$27LvTDh(WKV0e_>$^`Og>y#(`)4g#!ZLWFj-=u1>#S
zQFt4)shIC{A>BofI>Np(_Dvm`7OiIe**Ddi-s^1iLjJ)|48AD@zInQw{S0{3lj6M@
z1HQQoe{=a8S}OYJsaa^HL(opW?DcMj=Z5*F@@+5-zMEU*rmD|~(2_{<;9cNx>XVx)
zhW925{^Ixa^_>aVjdkRv;Bz;_2cJfp^3;XHL7Tds1U~nvPd>6!?RK-Xb@Bi9bhqrq
z{bgZqWE1j_%sySy`DFTv2K4}qnfYhRDz8p~Zz7pP#=_rx%gK51gATtt<faaIq$M7i
znS(M#!e??e9OJ%NT!5~V<aD`n&ZRz0kKZ<Yh~S&y@SAn$zKNkv<9s7L$DC^i6OpZg
zzcF#%9mK!r;~oka!X7Fwy9SvRcUFVHDZ;KIKermZq)+V%eEA+c0ki4wod}+ZCil@T
zR6z|wwX6hlCf$BJUWO>yCHVCO?z?H|QGLlwaUYHly{+iG&zLVxJ`L80qrp1HJaJ@j
zu(~i)tY}Ea5dNk=bNTLvXbuVNVt~Kt0=_8*f3vg-ehQvhjhQdz-f^kkL6;)IHyg&e
z^e;L<;SiUC;7Q-tMw{YU7MciM0&lVNAYHplow_*8snQN7K2)dT@m_>HaA^M?hvH_@
zcZ2^T2L7f9+DTkSvO=%P3Bli-SZ3G$IJ#T#`F{xDs))~+?9|;yHl^Yj-h7TttNSvi
zw6*DHS(}F8d6{v=s{iJqBlow;*V(FE#jScn9_rF<a#3pyHH|Y=-(zS+6GKDt8j9zh
zPQX7mVmzL?E^HhqXVJpc@ULgg<O0la)Mh0QGHZ~`^hTL=`in`6E}3*|gGo6jn3N;j
zL~p%G7-=-AxJj;TCS6Es(x8ulXdHnW1Gd@nC{TBv2I}m)KrK#UVz<0WjskovcqkfN
zly00!t@oR>>4QlX{mkrhGOKekvr42k>m2{4ca2%qs#^5qE7+@o!S}<^k6DI7pBqYY
z(2!?@p;Tr=4LGx&+WGpxM?;?QeQ0Tu?SM^_z=JM4g)Q+g6_^cHxQ*Y~L2eCvIVm&V
zn|4kWhhLj~jF}Kku^i`W=~(>Z33M~!u^RFpdx*(FrJ4`k-9-=I0{kmzPmfau$^OQr
zUeCe2_u%z%;XTh!9y~9vn^bfOa8~#G*QNU9UG&`1zfp~BIJ)Fcc)O4Ac9%!e$IOo6
zU(L9RImoW?bH8cVB(Uvku<es^c18E6qcPmBn-OF^yx=ol3loFv%GnwJaZS5M@VZ#z
zq5lXUdH^4Dd><c^77Y%3)0BS5Yug-Z@|=DTf2Xc>agt*pTe6n_&qunq<KdWYvNHnD
znK>X>Yo?MT$D6j;&b@PpEIgjkCJV{_-lab_1G;fVc0sgp(`Dn<R5+Xs+_!C0u^X3p
zWwAF*&kKa<C7JR*xx%z6Ej^NG&s)&3I^wNc#=Sk1F1TRknBBYFx__D;&xhm)esb^Q
z#S6BR#o9-&c}j2*SRx31xKJCgK~8YMBy=WxgKg2B5`k+fRSH)Hw5V69$cDn}>|DWY
zM#n|#BH$a&uQh1v-aTN6h4^}7(WSgz`Ehn#y^CJ9jr;@HWDHzP4(750<eqc8JX+HY
z-nk>^SZnx~R^YR?XjRN?@iwv&ZgvIp{7=>kj05MB0*uoEZEN&Gj~1;XQvtro4Za!h
z#G?t{*ddEY=Ugo>eh@GHMd(sP@GipP3_A`levS9P_38o~j<d2)2`%vogu$(kqAv)I
z>SPW2S;>XGS<H@!4}5)iePijV51B+i2;A4+?eO;WCWoJ*+w~!P1i^bRvg1$M%K5&{
zr%vchNvq)D*?~vlLb$^D9Lc`pN4kubHyICh9=e%3`Sh_aIIt*tU2b}GWB_^_UuWTZ
zcqucI9mxZaGd4`y@k{>jh3OJMqae;@Z*sS4^rzQ5PpB$vp<}62h#qjJ9!Np&z&kuP
z&w>@q+f)AqtKdy^{)_mDkKiYU8{G1N4yb}~;q)BbKS}4*Tk?aPnLg&j%=4KInFaDQ
zHzv74E@Quoj%YAVOM2j$6OY2Lwwi%YI3Dd4@7nv?PF31s*F1P^Piw1+zu=y~1uuOD
ze&z&thPOr0<A*-Q<A+bM5?Ch*e!{YN4|j4_RPmz4x8TeG-}KIB=teq2(cdjPgie0^
z@0@kQqU2jFS^{5lt}ogYdV22;7X4F?KQC`lwo(@TC~47maLx;G%m=PdTvxe<a=if8
zWZ?P+w%HA~sWZ2`(vq8cHjTI7n{wcrd9m#HMw{w@_hx-hyfk=ktYL6H4s@<=;GuTt
zS55g`wb<ENk)7UXQ-|{7Wd_@9PRz~+_!=+1noICBHOWFfy&9vpm*8Yh^L{W)rM={%
zHnRJjym!nxFb#O7Bc7Y)N7yY-Rw^gjRAT<je?g4mrm`1z3>*&lW;B_pE=}?8lw*%W
z9=dVJP4&Zjllvy!4u|Qrr`Ip#)M)LehqEu*)D?P`*3v^z<TkxG``O7rKS6);ch5uc
z;54EWqG*)Le2ml&a#PQy;T8KAtuzE5SYve3?-8=y;5rzgPixrCJQcrLc!ctI;`NTV
z{ZCPP{m4$$eFe{WC|q;VZDU7)`=i6<*CkvR!SkiTHvxaR@0kzg;=LKL9c(k#r!%qO
z_b9wJ9ek?5YyEX0pSGs*slg|@sqW(aXZ~rr9Bj||FvY={!Tq`_JA0?T;K#d#rp`TG
zGafFNb7Dl8N5z`K&z9rN$Vmo)?9|%(Xivw%{ae8~3&YUV;Id=bfn$NYY0UddqE97b
z@6>bj+*58PqucMlg>>JL)tDYehg2spO%*r%Aze7h@E4N}KYTG%M|ZLN5`0q?Z`(`u
z71Zwxhf^<9N9cfhnd)!XKYG*!I2-Q2-4p3zz)O%2NS{<4`d$l%$jW?L?+HBVVffQU
zbb1fOHDutgB;Wf7kElO<{igf$>Z1v-Kyw)NPY}Hi@TRqSc4Z_##d9G&_{RSLS)WzR
z8)M;adgCj0@w{lqy$N4BF)!RI-r^m1;RN<ORg$^A%V4|_ZOOQxhur<@Q1R<{i#L*g
z!GEEi%-QW7no`!G9O>v~d1=>X{1=z;jMzpqzw=!C(h8q!IdbBDWQFmTyrn<<INqBF
zZoG1h(P;{y&-}DX2hf;j;k)T=Rbg`9mkNP#z8lJa(NOi(2EH~!nL^-g8W?(<-5{IJ
zJqs86ek{6FTZ_sS=l4l1cv{TbwAw7&Soj-<S#9c>wHXbn*(Z~>oi%CRYLosOV<HD(
zQu+2K?W<(cL9op;KNG!Zfja#*P|2SKsvOv6*8M;g<88y2fvWjCP>0}f&g3$wZc&q}
zRWNCC9h0_nHR*hyNu|4(=yo<~A-JhXKa+kfG--EEvvSV@XWcPt)hqrFr_Jhk#H`z!
z&1#K))_93oPgj`L1iw$(yZC@=SoC@W95R@$K})OVJ+SIbN4T6>%y)QnvV#HK_+8$?
z;iUAVd#C~37z3R;2j^WB&h1A7dKrD>j^fE6?R6>Z1@cE|P0yG&lN6^XAW@Ja64QqS
zm-p?FO9!vPp<l+&%b$N`2CWCyNzCv6-3;#S>C!g30;U{w>R?v*lk@m($cshsa}Vkd
zuBEqdEZWq7d=6dxL<a!c)R_c2RHvY+4aQFlb}@rzlH15p8F+EPHPgW~UmN0=E=``Q
z89YuXJp%lF@p>t=7Y{oA<od1P2M4j&0-w>J|Ios+@o(d8cg!JkUjd8*X8YWU9yBzn
zI6P<Zy#5RIz$d*fSZyDmC*-6TIe`31PqHxcL-Yl&TW#`-f7q+h%kEa6K5ji`4$06N
z{89t{6>ev00J^e;-M7K)cm@L{9mp<R?&lHkJ%gvXHEt^R`W!dAW87MNf*o?V`Pi>+
zHObGJfM#tTijU`Lm=gJWbQ@31mTGud$Re#pJFkIf(2IsMqI@_x0r=^X;rf*b@A3to
zQgfbd>_R39eLW6O^Ip8X56LaXPxa~mXIE2ZvzvH?m)roOEN3s?csQ26oLeF2FI||+
zTH-IRMc;Z=kE*qx+q)6(<2?J_-lLsnkN&{njPrWvm1W0hPmdnK;}{d*a^QiAz%{Mk
z<552FP3&WS{naC1YWm&FkSFhiK1Q!b?SJSD8^r%>Al*1KyfSU|Y8M)sBd<@j(WwSF
z$%k_uyYURo1-~UT(bGByed>9*T5-nrYZsx2IIdZA;=svt+z6+00RCz^-UDV<^HjQ7
z(YYM>c_yLd&7b1au77;2y2E?W%BRfCuOC|YB>cird@nQS!!!K%jQ{D;-KG4$`99eB
z+7~dFwuf_%Z5O7;R&pRu@jjk)<4@veYNb2iF}=h=bXRf4E*MRIqkf1==L*p-IGcWW
zZ0g)cdt!Dh2Da(YAy}`O6<v7PjySjv?m2XP2e^AB9<%gLbr{E9l^1wu%5cxrchS2G
zSL$@h(HLK2BbSb!CvytVS#cS=yWt_b;}7;Q-@WxQpKwn^zO`wwjT|WXPrvi%+GHZO
z?lhEx>+1nSEzWYrp~<g?w_ZuM;m{>Rc^4R(KahMk9P!uYhMrX949vrvl*-VU4`3K{
z@=0Ku)*CH4Fv>y~fknjw@ff$WXg~vt8dS5WYY8+d^!rPxnE}&Uv<q*|Wv)+LfByg1
zLA+fGwpj(XnFF>7fV-J9v%CJodt=}&jvhm%9DLJm5c(9}n|b8IkKny&?uN6m!3TBW
z@3v-#RAc-(cyC^yO-1FS>n}6CeM$Lv^r(G!Ys$gNbR^sD0Jkhdb6U&Qa(|4L??O`o
zx0J_MlW_*#;6?0kTOXs|`@lc^-3nlzy=YT8@D{J;&ySCaQR9EWL>{`r(WdIuj?saV
za6bO@i2L#H(0jAy2D{%5M9V?1-xhlPs`MsH6%ehgW!T9;F8t#q`XjfnTWu;ke$b};
zy0G)39Q&Koqgmk*+pw0t;>qmuMT`Cy-88lVx)l8nX|6`dv5#Ch*nQMEJjKE625&>3
zWi2=zeDD3~yxjQ$-Ue^+C%iZR;JvvLNzR~KxDHh39>`C>1Ls1!JIq1He7eawQ4vnG
z$}sjaMEG>DtxvXEoErsv`jQ+zhYVwCc;Te`yy`~2vD;w$c~<6*%5a%ky(;|if8BoN
z!8fxOcoaC$qo)>loW^jzXz$J(_>2?7@7%+W2ENI;nLa<V;nSD}3X%;s@FR9@NJboe
zYKdQ%#y<hO9LK9jw_p5xx89AU%P-W8=Hk}X@@~D(%d7`~bK?nnGtaa0Cm~e+bLqjM
zcj0YVsD5-{w<bQwMCidoQikd$9{#UqLlh6b`9U}MpD4KYmLamkqh6!?ro(UMibr(b
zAEKuid}A6KtgGOgN@QR*7siY86%0e4Li7f*CNqPy8GLg9d@~DtbAfxcEBGe&ZT8Xb
zb7|&cm&{{as=_mDFZXb6@;B+L;qA;#&mVc55#XC$JTsfjrDJM<Q+rxCRllfHW54kH
zx#D2gl0#o6(l^(g3>En1IQZsJDxQ6KN4uXQySUD-4RPoW+|P5sH)YB&uj4~ma1AYH
zHNATS@tp+X!$F%GkN0NbBdcO}TD5ViRqbM|>dCWx0lIVVPll3nzZYF;sBJ9#tb_BQ
zj-lb14V_|MX-&?t(-@1^2f)J?vS{mfv&_fL=wD{#9fr5YYSz&@X6=Df@t`L~z_nbQ
zZBkr+JT?xK+%-*LLzAv&HfcQEO|wsd%8efN;BKH&-V0QZn}ND<D^N4hs=9m&RBOCA
zC9|Vr6*4KbEO@4dN%I?<v=e{M#||bv1>;15aV~*z+72}-;}(;izcuM_Zg5sfvmO^Q
zE3%9k4Zti{eXdS;oV{j^j>9W-z|5XgiyD@-s4e>WiH300=yT)gut<ex=TidTyHxmf
z(W}D17nkFC_H*7QHk@E_a&2fhg$mOf(;S=vb`2&U*9IOZ*&e(&=v5ElYG(2p7zRG*
z`jFp)ZCacL@4~510?$-@jQ<?}K-!u3A#>7W4R>b+Ta;Txf6r-t=3rXaB=$tqaOitc
zI(d?SbC^Z{tf33whMk?FVA|DojUPs5VGlA?V49`v(V*Ip!EX+hsbg2N3UuM*v8$v%
zyK*w&J4dHF1+TMyvRx0sH#btS=eUVOclwZl2TL`5N(Ug=YEWx*IIwp0LijO{p$p*0
zc{C;nk0ttRTD+ZI@qOa$jN<t;1|N>C9$lCr@Hjlnra!0Ga1q_K+=~I+led^bV$rG=
zGzOzoc54t?vb_oK@8(tk2luT9o~JJuXE>f6Fi!j!FwF$AH#6uVU&%e4;8s?iopqkU
z)22hSt`4pVq3d%FoG#hRp<s?Rhv?aIqK_X0AMku`IUJs5H$3w#c0iDoSaOeCHD1D5
zLG;Sh@PV6s>J4s4i?&_~J*eemugo^Drq=c<IrzqP&7*$vJoHp}bPtYb7`UZmC%V_0
zqo>#4->bx4^or;-wb?HN#+eV76Ab?G2jg_YXM2Ia{}{e!P(S9lF<_;c@YUd@B)h-`
z=izf6vZIFiZ%!(&X62<brUv`hnuBN9F+11<kJE=4aT?ss2Co|5!UK>4T!T*ak9-;l
z9+{5Twvu`DZZdia@k}Oa%f75^;mVf_&SeuAFB5x{Tfyn{;C%1@cULH9`gnRi`uXJY
zfNOlrt}Z?YnqK90KIT533ReI#;XT;T+pI;wK%DW5|0CNqn*Y}w{@;V(viO=~nMv1O
zW3MG$9{Fl;PQ%#*pdsej$@$!epIbe*a!d`?lKS`p;K$R1!QH@xUp~jKw3D2pWTP(O
zX{^DSnijt1$+IAR;>;|^eD`F(T}E3nj8p04<n^AkBi)o`xaWGKH*p0_WA5V~^yeOV
zZFQ*U5IlA0?ng)4@T}N0xQR_$qipI=-`-V|O-nsC1rMatq9*&Gz%MgS!p-bAG-|t{
zUt0}j-e&0LUPCi4gI#VIs=(V~a5|@^8JgAyykmvOfxk{%9$m`c(8iw@wRlSQ8y@@T
zYA_92eRnjeByP?RIBgr}qCebC^Ga~tg~?21ASaQ+qVb6>bRDpNo9jDQB`{4ju9y6}
zC6@zi(+6xb0Bo~?T)4nD0psYV8O{5Kvs-gucbWQj*Ll3fZ^G#s4h0t(=vjDgcDG{x
zRD<psTb0iXcN70NGnEGX^P5}~T2z^9F?tPG(+SKHkPxHITVu#^@%trQV4Fa2%i8he
zq~K)=pg*}5lMjc(xeK=O-$Q5LX7b<5;C-f}Wr1)0?!YY`7Ne7G>G7)-qiQ8$R5p8z
zPA82~pSSdi-y$nTZ}IPC?0_2|En83al(Da@FMB+4e}cC;AElM+=^&UCC0jTchkooj
z;G3iw!8v#6E5Pqwcs$sqXQZ4K`u*@04}!lre;)3JY~^=+#TCg-eRqO)@Q^vnMkrqn
zJZC@X*~DAyw}ZSCnE%)?{Ket);Gj=6g1hvCkG%hdd*BAI^}}#A>(QX5`&4isK4ScB
zqg&(0;cO@krk|dK%;IaX%@s7M-ChZP_axrD(PYElm1W+@=EdhrR_Zp_UUX>Aia5B<
zm4VDY^_Yq9q6KH;K2PM)VZ6nS=)S45K1|tXps#~(N}}Vw?#BL_`eZGNF(YsfWqS<w
ziC6LVdUmSK1m7_8T?ui^%DEYU-kUu)_Z4UA?}wqvNB2#!E#R1$p~{T+rg%uG?$Nza
zr7GM_-capJ5~|S;LKK0IzuD#xZRT9=Lzmg$c63RVqW3!+-W;-u`Is{n9U!k57c2)^
zUvqoT^)kWQk(#c#7w|M^=vG*d&M=+sV(^U*|3%FTWM|USY5WPj;RZANF0xV!(WJmP
zCA!mXVndG#z-NpWu`&bi`-H~=t@sdH@oX}?!}>UtHh{chK{!_M&0)MBGuP1LH{PKO
z(fHxe6B<;e2M>JH;0ZJQadZZrc|Fl67WwdEa6Y_cUT+7!sd?F^f0o)*9BpbiyhP>V
zcubOz7a|Yva1A{-WT)x`;k~IxR~0>d-(PdDpE7h{xk0xp83FF^-_;D|gO}ZN&7!B^
zfE~Rp+TPp(e{I1xWY*b(@F8=}N;S}|PXT5X#ZTP$w@EFpnDl!dm}ZbkS%cwjTA5^2
zG^u!glWM0o>E)k5x<mri4P3JxTvPCLpv>n2wenh^693hyJ_M>!GLtM|p_F{gRy3-B
zs$dwI*tu#_6);ZWPA2^{OiJ%F=`<Wp)k!AyY=fVko9JLOsapoKg5h+gmo{s|-`rLk
z{*R$%MGXgs!Q(W(h>n(oyo1Z48s{ylNG{@~VQ9hydesY)p<m7!iDthX%_<7*=Vw3o
zGS1uP(PXw~)19$}y?Z-3bALP8wd2xLe$G4L(7W(6Z^qBQ1)5aBCoV09PwxrlNk#|K
z*-d!Pr@MIW;~gjnueX_xhZCAZhNC?{=N@4W6$8h<=;l!M>SVgnqYmShiNi;{5<GL4
zvp?q;^d)}g&A}-LqR}Poc4cU1SCd9|rKyI;hM#}ZVs=f*O=n?ByEc*uxyoyy;Y<8R
zV78m};P=2p!^l_d_M_{#EB=JF<n-=4v=EFnzXSZue5V#<hi{6*cbtSBYvjFd@H*dc
z7d@>RojB9DFD}#Xoi0RY=q`!1@pc${fi{u##ls(rMtt~HsN&J93O8_TX(6{Z-Jw7K
zIN3(7x6B}KlYn{PT_boV{%Om-4c?gn*7-2ntpk(s)S+#qTLG`bvoqwpTdQ7!e=?$B
zmk(1qx+CgvcI1Xnc)Z%9i%-!_BFLHd@T$ucuTF3#ZRc#dm<CP9O?M2~rfnOazR|g}
zfoJ!dt8gRx@%ApGbFUw}^2lz_FXvSwKX{-k%wto@e>0O^i}onm2`>YV`dyzHtr}RR
zyhp7H;2ElnuUH<H;?E}F#hKli3$1FXg9{!fx;Gly2z)naRNLV{I`78Aa}Ey<c;UoL
zyounORJrKSDGj!&2ZzJ?d7(4=iKEf2Cea<Zmi|9DrVO05ITGOw80wWjR~K?t!^wB7
z?dH=mX3u-@=TrGwlN`evz}HfUv$#E+RGKq%3U)%zGQ-`F!`f=~DIA}{6!K!p$)1mZ
zvk9&3Q<;+NOwJ9z^%qC^<J0DIX8~7LpM&qY0bJ5Hj~q6#Q7OsvUu8e<7G_a=13z<+
z3&AIqaxFfA0d8h-w>ptsZAh2uKczy+@U!=fGj$xVlNRx4QFv?W@-`RxW(jYwlI3Uj
z4Vl86>z%4VW}>>+M!&d?{0$vg+3Y&mll;U)=C}3mky&|NPoeLkjE&ug<Up&C(`H^<
zdB~~}<E*;wwkpwhy10B+vj6OS3bSenm}VwdSv099@G|@G={*6{*w%wp))*SO*3d(K
zofoY0@EqP@u#FGyCf7K;eKCgo=%o160G%7Z-?ePadtWUwU$!W4uSF-9TGS)XqPaaS
zG6h)_*%6O%LyKmWMvKb9nUcn$Yn+euo|*N}6SG=@UnX!R0n0>iE#Vr@^^5mq;@UB*
zyZYn137kTIG2Y_-c#Fr3;_riRc7ShMfN$!8Z*qWd{DRP@OkgFvH$Ph7)2RoygErL%
z-%W{ZbQ1fK1OFVO+vK5kT#Qi?d^EkmEI&8JaE8!xvouDVz&8D6!_AClzrx@c_3Dn#
zrdJGof@G!U$LJ6|P8a@uD4JC!u+U!c&8ZPFngqUayTMH0n_M;M$0<UFDhr){N$AO;
z*Dn-&(~UhUpTIY(;@Hm+6|KYckzXuD|2VnuM(60hSsSIw6WHSlzNrbmsRX_mk%2v%
zw<1+@d89JXd-H^Tf`?t$`B8x^Wg@gHbk+(9?2ubZX94`p87rLyXjA#hM9}+3C+HWr
z+FSHn;YV9Elith0;Y#p?>vd=L=b}@e&B334WEaC#b~@uN9=;k43U6`xzG&KbiwiVk
zR}g1Ibzc9exz}pF@ahLXxRN{Z#|>l7@X*Q03{oAwH4psFlee4=SD9J1;H86~{lzTf
z?&{I`+VH)_(W$ba-~MLL@Z&IY8|0;!2Tn{2Qx~}B05tmWu4Jj|;X9`Lrv7*M><65a
zN8Kv48gKC=JcrB*9S!=!8o8yiZe~<wz@+RIXWxkpZ(DA%Q>mFFcKk!9ye(9Fo03f|
z%kDM*Pz?m%^tea=4IP@M4I$b;0$tdLK1|P44)D#G%yeS1n|s$?{B5haH_3*-2qhy<
zhsl6)^p_-JPXm6-c6i_3t;0(?m9FnT=u>2B*Ov>@Q23i3;G6!;?5&yS%ka$WIvNjL
zG}%N0t)c~9)6)1h($k&t0e<zGQy+GdQ)E8r-ovR|;G33toEnWj)$cgHY|JZ5#^Ni*
zgR!a&oxc^qJZK9u9x;m_B|Ec<EKM9b1o&nk^Sh&>UEcI`z+A9tK)g+}qixF7j6Er6
zQ|o?Pm2lRott*&m$RPdQk@B?`8S!jZ_TLz~j4s`M1<(1xhH~RoDT4O=%8xUFtiaqQ
z%q!f_8JglNPG`}c`(`!RK`v@KnW^q(oe`O-yk^xvmr8xW#QrmrKE;@{0Sq#uk%<l_
zla}W;sYx1>3c=4@fu|{U9i9ekv-239o0EYWaXt`UBT$vmsOly%=?++^Kpr%z5+)6+
zWYSdpJC_@pv>%M~q8*+bGn^3G*Xq9bmhtB-TWrEJY*M?^CY^*s>iLyi7MQD51u#x?
zE*F{f!Dv`$RS9^95`Mz#FmK%mA?H2^F6XpG>+V>19gqX<#yl5gXv%o}hR4WJc<A&=
zW77}3F842ySAy3GKjDyPEk1!%;90o2q$a!o@Hx@P=>$NZyni20&SkhexSR03Tx;kI
z;LjV`*w5jQ{<DFfJDgp$$8a@^|F@g?B>Dd;`QdP~!_&X9qo3Jj*^Jj_As(8s@G`^g
zS~UQl4cyINJ*rZD_Jg8Bm97FO^9%3vb9w<Dz}MlK?)}xK7v$jmcxLP+Czcc3TN*#k
z*eLK398Tw?aDn)i)4}bpK?|J;zPZo_|JMY3IBDVk-T2>+;BmlPwh9g><`R3dYLU^M
zNap22uxi2I{O!^&NvG`A0qnOTi`aiJGdr_`jaj09d7gE6`?K`n?dDK@<-R<#F;td)
zq1uj)aV`hD9?+KuwPOB9vw9ZeRzA3#3$bpko`#QODf$+kyR>KEeV)7Jgu}UjX8kvN
zP}&~Gp5ZX1f59$ZIN)LP*bxyPra<8ZDHx{ig~K!s9qBB5OtBVptK-?-^aU*oULt|6
z;GYRx;Gpk)>7=ocpJFzv3x6XtqyqgtWN^{e--fBun=qZ>?aeP?TFIF;DlMLy+<1x6
z*;jL(Eo<u0<~AN(YR}om88_SN(KkM}Q+K#)bouOPRPXR02JQ1G0(=w1+_&^OS<>X-
zG5kK;i+gprmRFgZfk)tS;>!OIV~xw;Rq-x#in%#sTX|KhrkDL8Tx-dyqjB{t#Vi@&
z)e*2zvhj40kpb%%<5gfsuUeybeZkMu=OUdu_1X0RFEghxvnf8FCjmb7<vj0>-=IWU
z_^!-8-2&63;0g!(ReOxzc@v(dUhGLuiI;LZUk|+ddcF^B@ecf`9j2Lh2Kv3AgN&Ke
zHrK6LZgQ>gb63HxkAp+iJ9DUt#)fDnUYj0km=DK*YvBAM=FmYrDp;@V>`b^2q{*Ct
z<MP25#@TeB0r}3(WFXjiu<Z^zozR+!fYsMbqF=osyJg4;^x`^qksJXU`yQ?|4y&4g
zXHGVt+v19Wr`pgO&agtK4ZK^1s`0+AXzne+F2QJ0o4_;a(BIcH!>#2yv(!)%t`eMi
zedyilHx1tnv)vb;p^TmI=hfp3;4Dx|L!~~^XTuy+eHS^4MHY72p+!Yo<g-}xs=h^W
zRX7vT?MG)J6Ao|l>jgUbBeNPkFyonKKLZ#hHU65dT*bgN6}b-a>$AK+J%4V)cQbJ^
z+z#HG;dpOujKn`Y3|}$$W)<Gzc)Z2K@ZMPP-qZ+$%W02x)dbvAjX#6CNtcuE8@QX1
zf7lE4A%^`F?7BQ2qv3Ed>$j5i#wYv>Y|{d-@JX=E4S1Ua!_b`IW4e3kBhE>G-z#*a
z({vASBIgZ8xfjoFGIHK|=^^gLu7!+C>9FDL@KtnJZ;RI3L+tlB6Rl}v!oO|BXT$rQ
z>|-cEPjOfVa#P9a_WQ}+)^~Vp?nkLHI~h!?>A)Trr7RI(8t}~wyu}IhGNn6D)-RrX
zINsvlcyErk#E(;&y|Z^C6i2sY%`N!J7O`_}4ExvUne0*<4yRCr1ity(UvLy3+0bot
z-HfLzqA#A}P&8_?;kohPoxy|m<|X-pi*Pl2&}5g>3o*f`HF%5T3~&$khF^Jbe-@v{
zedRucyYX&hKAJ&qCwlQ^_~8ysy-LR{QaO`Xo544?z&GE(Hw|aO>Gq~8q8&PRO^;?2
z<-X6z%tbc*`a>`dXVft8&C{uP@P~w{Xc#*+na=`ig(-Pfx^KS0Vc(@Y><Dw-Dl*~|
z-Ky1_K0nUGTHu=prNKJc*g^G&zGCoAsU!Gr)}Te<y?N1#x7JXlYzC)OnsYS=J;sUg
z9N(e;`v_g$bl*%G9HN@=sP9^`%LILDLxvFarC{Z_jZVzWF_L+s3;3pLJN5>EZ+`wk
zm--L><}{iV_$Ch>A^X5Lp2pysvfv*4#hdV9gj@yN>_#`9<I<9m<ZdI$h@0`4Hl=5z
zq)Q3u$ji{D((4L)8Q^cGO+uT38>rpNsRp^>Yd_<;ILb2*E&9e7hxYo=rUKX<P?o-Z
z_?st>;c37(D_7tdWv0LFwX0D8`=lz66{r6u1s;;?^K3ec_oi8WI`Q&xX1uei)G2!S
zmh!9~Lbrq2stOh8ddPwo=P~<IjvF#_ug@BUH@6FZy3))yiSgc@v*_yra@yR_7r2iX
zaQ}X~VAj$NW{sF&)-!aUJ#ZxT{pq^_TLkX|%iz1ok9IU2&&>9EChY;EICGeEliXA~
zw5e|JG_5WM$~YRRbO!_VhAZ8vKz7v!s?+U2y#(WY!>2QZk9mZyRk;EfsT#T$pDSY%
zc%arMedlvOhac+P3(jT~`qg}NFLbI`M@+KcHp%ZDJ|KL^@4z?@(Wn+0W|iu1R^9nv
zv*Y|<znC=xAK*rVp3%`3by#oFhU3g=Z_wb-+^dEdN`n?(mAP+0a%RN5R`PdN#nrXS
zaokFd$;Q5BdWBD*F<y0QFq}^9u1=+X2o4<zC(|2ku@gI0!88Nuw@KTV{R_clvUv>z
zX2ox?k&YwwgBq9M=NgjNs_W2ve}}5E1GUCyGH}=6W+tFH^0Up_4zDo$oZC%baTvI^
z7d#Gl_5ye|dtN#Lz%Icrz%cyWmz^LNcAg&jQ#N*g;<?&o)4P3iBHpyAY-M&*@cEK&
z2Q%^SjD^2(&tTuq1+rS1*~8B3uNyw<w~67Nz+soS;D@jIzi#_QD}wYaFI?t0yf>GE
zHH1veq)Pbm+53|fkABBF=)TPD+mDB6d>TBa?0OrA?`^u9EF*LMpJiyqE9hz1hK2>7
z(=0O{oC<CgZN{9^)veOp(`6#)(fJo&@_6{01z??xc#Ds-?+Tws$JAl+gVUW$mg-Cu
z=A!iIRrl$=y1{(I^%Op#HCoowmz)P=5)OYw14|0ez?^jPOqjaCzf?kJYFq?t(FQC4
z*HaU|qyfLK0$vC~SO49M95=dDxn%Gy%wI{+(3?Kv*-jQ>J(}3T6y&>epgWZ!H;!Jl
z0F5e9Lk~U!cp7+|VO_{nz~OX4quK#)y$y}(+#-IBRxj|)+*=;C2H%_jOAYd8Zy*1c
zl2w@t@&EJ&zf4+8R}Oe7LrS;0{?+wn<LkZ2|LHfKsNkj*!}uI%UU5I*iNHkp`+$c=
zf|I~Dt2uwaRp)IxcpSK_2aCLV@d0lU+)PKXj1LZeC%9%Yyv->vPow<s^5nGEf%)FN
z0rxSt7CJ>veS%l@8}a|cizKk#l5%9H4zL#x|5HD{&o$1`Q_Yon2Ycei;fL%9M$HP&
z*%7M7HRw^@!;FY#nTYJ^8qUJ7OYArKXeFl%M}nW_TL!vh3K(h}PDkDkL+jAV3y^up
zS;7W>wW-%PtHRUTw1P8hDDTSv-|ROXJj0B!j(Kdw7QBM-_<}bZira1I;yyY=(VDjM
z*25LG3+%FooCLqF!}~Lli3;9GU+5C_CqC}VB0~+i-pz;8A%jpPjtoUVLmljJI{|Rk
zGT3v+?+eodmf2A2B=lswgTFqD7PXx|n}y(<5p?@`n2FG)PSoXmtYFd5qVPKD$xyv9
z%Ye6OeV6`>8)h&fm}UlEU~o)I@Js^NC$7a@t9joO{(KwUjX&JYaJZWUxSOePH<$3<
zJnaXr>4i5Zs=Kb?y%}xqu10u^$Kfp=P#<hlg?xFj?(7=tuG1;IYty$F{lRy$`#hQv
z9^rxsG1?Cgb7whTV6e@VY2?8tpegl_(Jd!C7+iRc^PpX&rt9w+c;!^I0yoe*45n#2
z4=h8@yCYuV>{Ft(Xi~JA;;Y#{Hd;&g{nY8vs>JWB27`;(tvsh9m?#T-RFXw&;hQLZ
zV7E$IIGft5q7)d*zU7f*ry}VtZqLrLy6k8G*POyvT#9`nzOj*NZ^d&{CsF|g=_aSo
z@6fvly}L<I1m5Bd8`vScfPI>Hi(k?WQMOWqJmkU;<GpGAA3E%L_GfMhSMt&9W1#mY
zG5zI!mBZB)?xx=#{AAC3^u3@_ukvZ?G;j{uRAIQ?lz5Ae)$yrnaeQv6(6*n^P0T&e
zVhtTPoDaGC!||eD8{7}=OL)~L4V{0lnN==%lo#&U1J`@j1Fk`*zQ%b_pa^|e>6v4`
z{%<!;D`v07>)FcyzM1_mJEt7XNiD)OrUIQA;G5o@QG;*OUwoJhIDB=`1T-n;yv8Q7
z6}8#9S(F(tBVB*=Z=Amws*;D;&A>T&VJ!1wPtH~g98ME_k0sHjvat*0V~D<!Ydx4i
z*ZkZNy%`vy6`}0ZY)YQ0<ll2SL<3%uSv*6x)aqdQkqz$#zHtQvt8=bk#n8!+1O8?%
z`qVM}H)-k1^nq_;8{tnV!;We2jg5PH0v`X@JLvPD?Na+#m&QlXyWfQwz6So&{Oqlz
zry>7;a5u;4x>-d>IN9*;5xm96+87^1=Vy36j?gcMcP?fWIzkv2rv-b2N;njZuJGak
z&$vT&mBKH2Vg%k|o>>Q4f_ci*(+j>?eU3fgGi`8mHg=cM6GN}xz}NKR@mx+#e!Bbs
zdaKAzl`U=6C_3|wKH@9@-yB_L=uvOF``R%3WTD6EH93L9aI-Vemc16);b2#PMdvwX
z*0hyqJfp!5cC&KUFzX&!rkdx`kMP~ZqD@tCo3yqAd=0!y?;>E5EGA7ucPfH5)f3-M
z{u6;}ygN|Owg+nV4)Wn(93NcH4LF^}XjcWvjvx6Fs4Q7c=q~UvmB~_pan{y{BWhvN
z7&NLBF7Qrwe9ObouBOBJpi%7xUmd*;j(Q27^Th=IrDL!fIIOc-DSMd}Iurjf9-MRU
z%^I4^qIaz=8s5voy-&|N=a=sx`ucB+zLnwY_rO_m7KXhhck%XrS&KsN4b=qa<oFIB
zR?DtxJ<twwf>q!1bIA#&xrYWci2N0J;ue}vAU~5JG_D(5Rlz46EZ|W%m~Y^e=luR?
z5Asg)*r@=g(tIEM%5}TWog(i~j_VMgr<lX8ALZ~1rvuNX<m13I5t-OCR*=4CJT|*C
z+Ig1aBfrkicpKUG@pJ;=BZ{T_rx;y8GgsnKSZ~v5IGo%W=m0d?HGhj;Z&JZOfp3~l
zb*TSYyf|st*}(pWqhp*(@r#Z@@J+_mE<LD74?<LsF42Y1<tqKl^skL0_p%%xet22#
zrxtX9<Izt)CPc*+v$u9Xp8Y@UKxB@8S}_#w58buiP%RijzxW*XqcDGT{6CV;I<CpS
zZR29M*xlZC7Zx5Cx81>t4cOSojbRtqirwAat+(CX-QC??ykFnvkLxoShzMT49p_Oe
z_v$z@0nZDdQIWawG-Mw+8pmOLbwRy&?HnDh{j<qVp?{=5fX8D`xK5I%nv7Q6;1BtN
zbP;H7;p)xV@a-DDWUhj|-R?>F|K}0%z9pajHC%0y(=U;QEGD>Ub+rg(YYYEik5CV=
z&L6JMrSaMDm>&Jy6GL}J9Ou;&&Y(N^0xzJOvsdCUb65@di#u>Qw~|L{Qx?2~TvG}}
z>Qy;(cmGILXabIbFS2#ScM}w;reTrV)hkl1M$n4`9;pVt$qT+I48Ga%8oYoO7M>f<
zyEtcQMf{7*h4D`!RTmxTEqvFdi}1<JUk}MxXTL@M6MS_QEM;l|FAkO(P!FFof6t|O
zxTFJ-O889oPQfVtJ)Fg0rJc;2>)~}?-;C17B;JO5kK_z5$ecQ+e6(Kkeu|Xjqxg7^
zW9VC8r7ZXg%7FEbwTGYQdrP(se>8aQ7ys6dYw$$=X2#sW4#4GjsNmRoMd2B5AFj=f
z_}Tw*v2)&~rJ?Mc$G=z_?{JJAKVqU?nb4Y^k7RF)2fXkc@7_&|KJjR-$9eNSNQVtp
z<y=n2p&eaZ%n@g+Sm|0N2f_T6o__Vx=;-3C`q|Q|Ja>b1cL90rL^#&*;FQ&zTi_fY
zFi(@M=<4f&^n3%IHM~ERx#TU6Ilwl7V4Fz(T%WmMDy}QD*?kGN={AW>_z<uV_j#ob
z=vhtqb9KQ(m6+{{1gR5P>^^6~*jE<azCg##R=SGmwCO(t|1jE=p)Ht(na@9`MU#G-
z^^I)9u6t%ZAm?!P8t2Ysvx?xWxyqFX91{V~v4d$YaV_H7$or4sZZd&y*23MC2H%*#
zH|4-LJ;65v!8duqH~)ff7Qx?40pE-dr0cj1*hqM9s_|S2t{m*y#8;e>&c2Y_WT5a6
zpG0r^)GbcWz%f}m#%Wa(`icF~k_2Z%zB|~6cNh;%RXoC@Qjz<{AH4B|S0&bYRhiui
zhru?Brh{p~HzOu_H6MI)4u0m<C_Fa9y$TrU)kHkRMMK%43BIWbw)w^W)x&>c_2d~n
z#q61?c9MRZ9kIH^o~dX@tXj5U4?|_1XBWsI`X*;Sgtytl9)`*6owL!QUDKm61w1O6
z#-j_b={vaYR%fz+_a?h#j&ds#xv75S!Y@>U!zn^8JR@20$K(l*)0;IP?q(RfS?IsK
z-iRDkc{HnRG3t&dfBW5NHAS0Ry9Ce81boGK*{a&{(8FhbuEuT#G}>z1Cll|2Z^%`a
zpNs#01an8%C>=M!=WtK;AWPUT4O%ky@0bhRcPqg%li+Z};dHqtbAfNp6eIVL5xt!5
zn>CyvPmV+=E%+vGa)g$GZ?4)R^r2~lnpKR@r0nb==8SrIJzVko!}V)!xCU`w)rmp(
zWoF1(hrD=E=C}0Wih1K=kF$$RmrH+^xKwGBOAor@wK1`Osv+GsWWzgUr|a*7ldgAs
z#Q&iWGjFbdS3l1^nV#;O9YvfPkphfEcIy05JaOp7WBbr~V}e)p!+Vn(eG0$-3A+7?
z!r!!_8>H4)_9XVCySNcMh|1xWqYI_MC$uMYiOJjS8UTM&eF$@VG`pc&*|nuQd`<!Q
z8|LF}c#G?zZ?vC;rvvV$P^VDx8F=W@v4057V$W{&Gb{qvkm22B57B{UypJcN3jVtX
zcZ2ncjPS{&!HOG!zm$AT6R}4!H9I%aBz8`r^Cp_twN^IGuWZxW?D#NVkq1Y!o-x;|
zKE3eaaGr#ev??^MRgK_pHZ$v_bTh+PnSVHY9;OM>z{eKV-(gW>_>DIvbY;B7nLnCk
z+htba#b&)8Y}R=boy9fizDY@b>J(m?8TexQo3t40@er@fPq-IzE_js`CgwP!29N`v
zaTShczfl)Albu=*Z?g$3v)!m?BaN!ui%y&#MwK6A)Vsk()#z(f&N!p;+Ia4#QESuU
z<!MMK4jR=1n@Q0glS+f99C&aBE+?<G-K0K;$*9BO6nF~O!V{FIv{`Whe9sQE22L`o
z#7;P;H+XciF>f({eFVEr8ACp7twp*Hj{C$}mN!V(TX3Gj<1`xzhr?N#6RvyWvmlk@
z9A3*gJmD7pY^zNh@iYF{jO@L?jb4AN{(~cXbA!2a54+b_b3QM#%KZXQcY2#@Mv&7A
z#_!#dJzd>x=>9e}cn`-0Zs`M_83nFsGnFo*k@ya}qDA2iUXdLi5Ls8>MffO(vR91m
zp10tY{Rhz=&oJx0!CQHpt|R#UOCEmj;M(zY9zD0&*v)5CWpZJaz&K;TM-AZlrY*yx
zlL8NWXENvDoAzjJA5w*?Q+={i1IUQIU_X0DyON=)wFS2yrt|(wcXo}_g%B3u&{Qx_
zp*jw|qsu21{me5zhvD6EswZ6Nv2l2DcA@*eaH>U7m&!MHsW3X#iry~yPQ-_U_rJ{p
zm#+HIW0eo>sUn@^_261~tsE2zu89rTnnCPe8N*JNY3Nd5oO9dBjDvFqpM@^~%Y=ht
z$~+9$(HrbBzY<R85HrtB_RxWGtPkkS0S9$|Pv#&gTzAp6J{FG9Y|fPL|I&%x4sUD=
za+=IhDP82%!y~k#2WQa+u+JgRpsSoaJN~DgPd<m277k}CI#|7r?EPgf>+ge~Fclf8
zTyWE+=)$NGsq*#VY?{Kaw1b;Api`N_J)vYNz#xlKMrvwS@CR6^By(Pq?vXk<8;xlf
z8LaD(@c3jj;B?MhicsqzF0ENX$7mvXPCFfnL1ZB2^7Gv4(&gUlE#-UYUjSYl9%*BK
z&e3Y*(d}qu%#=aNm@RWfsZhx%B{6UA?Ze|?vRUvlT~bHu0X)uqGO#zmHrLbBQU99H
zKLGy&4%%OYE};kHz^>sjN)Hdj_tpfi=qG;3>+krtM}(_u2s~Om-KzK>P0W}nZ2S(I
z!KD{u9|zfvy|<hyjL!d=ym$O%`nxWW3B5!AKy91OXSB+Z3jGHSc;rIvvBegRT4B*t
zxRESijL+Y}DZG6xb&z7Rf<f}pr4E1GpdnX0JvFcJ12b0?0kc#c!DAmj&hzDl;j<YV
zq|?kG^T04(_?ZMAYvQqa$F+ldb^#yz>#dnT1Aj5sQLgOM@!(8hm)>YP)8pWA(4*J1
z1|v0w8>$tgisivP`I+<5v!muao*MAjic1#V+dwvarbV5>H-F%L$2aBu5_B1-<&5M!
znfTDm{LJik$*h^@@%)`Nt3RG$H`-GHFwPCIOldI91+Jr9_rW}Qd9E)XZw|h>hBlQ6
zd~*wJsx<f}7x-pjcl!FshHnJlR0w6a6!@lTYw!@>n>$tDc#4oICpYCv)>YG=gJtj(
zH-xh(m5BZnOXqM9`=eUK=}&ETOwm{Krx1NMCD|L5CQiMRysDNuPKO`UU31l|PshAU
zTupBc+|B$6?85}#v<BZqgKuVnZ~Eb@v4L+^4uiWHKpwn@SMCt6>Va=u+4wWyo2%fP
zJ>Z++;G26VV!4;tmCLUAM0>0zgKxT5k5zi`&2R9{iU;&+gKrZ5;+r6k9#mo%OAdB1
z_;@t?hFfRJ0*<23Z!@}7Z##NaYkIorEgylmI8{Nn(j>79jShtCbUIa`FJU;ne#gj8
zWvNf^K`HvmGtd+9nY;nso88;V94sPhFqRJEp6sj*CL>NJp#zwIRt|Pgp<i!8n+n*%
zy)zR{YAD=pEZ$=9&04%@^_WGH!@JIU#r=I84reYI@&4=zcW|$OZ@T%B8z{<*k^!IH
zCp7e1c#1hsGOXn+nS__PKm8U#bb!^5(9z-%YM+4~oOhgkm+0=pXV@Bi;~h#)JQ}^1
zv+!Om&cQ<AY7duNfbN^tS6%Y#r@weP-1tbkGvi#GHOxZ(F6AtRx6#+7<?uI0u92Mr
z-?-UNT^U~eH{CZk{GFOqfc~2#b_qV<{yazz)%-A}1K%VW!_>VN{eIw^55F9|=F`J}
zoE?cP9Xd1&PqB?ITJ)(2;G6o{+0pP39qJ<f;jMP<nMqb^fL)bc@HTWc+^s^NLtcD1
z_)@3bqUT{XyEG@@TaCeIM|S0BS-LlQoiSabt9B<@fdzCJgKv7$#j>C&x(EEth(Bmj
zx7mBmtlxbhI~wq0T<d}tgBiaTuU{YWTND}3>svVaNW3>S%G)$23(q~rZ$sDk(b-nj
z>P}Z@3tn@JSkdg*L3M}x)Gjz&IOA;AAYEXF`I!=);sITq%q$&;S!4;Mv$HT=<d4nD
zjW%_Cgjs#*CawxEGP|%@d%v4>f451;ro!2v8C9|1{cVh<RNSP5bmX1h8a4WWQE9ds
zRb{17f2JCBe2P(1CK&Z$lu?uI^!-H}^)Sk)j|QVGE#Y#S81>s9p2yFqJXO)N8W~l9
zuaTGx?=qawOFTGZqR_DVnsj6Yy3{xmo_2Jq^(Ki7*BN?vUOa<0!gpL8eyJ^)s=a~a
z+Uf0ydt_E%dW(+KBJ&x-e8!oT8t>q4&fvB$E$nV#9;}GxxMPqUoPmAka7ON@7l@x#
z%b)1>{Opdvm%ku`I&2<ZfWvTN+pX#VwkbWwsylcdkHeESA7j;k!R*-zv(l~4j2cMa
zdq*o?RQ3&nfsz;e>-z?Oz{yl>KrSi)PLAJcwLWk<;MUR|;Ocznyg_3+Fq$0`{4P6v
zvFZst{wsdBQ%c(O3cPf>u1!f`tgKb(^U2B1v06421mA84@1*NVzjJ?^oZzE)Ke`O(
z;nDg*j{tt343qi)1-==PjQ@*%;2V7IHTZv-9voi0uU+q7@#yb>BV_+_X?9QGoi0W9
zLE&X|bHh>Ee$d5ThK||V>_#UW<AWck(PVPL=oC&k)6(33^SUs1&_xwL)Wz;<JjlF`
z&-=rU_*~q-<;h6-;pb?H-Vuam6-i!-*U+tr;dp((H1M{SHiau;FPW-S`0g%;D-Im9
z96WQ0x7%F|*XeWMoD1RlaTPB!&o2Y}Sl%!j{es)!9BEpB+)vpEb+3)y)G$Kh$uBs$
z-ZhTUgANfYW`VDn13$xAG@h=CTJRMI_K<f#KW})O-FVO7e%_GT0$&)pdV+CYWCNd+
zfQzn*PDOTnV-q@!!7qO~oPO{^IpL)1z$1;BL@uHd8XO$+sV{IbshG>qb6%mZ*X_*C
zF7VtR3;tZru37Lkt;T~BLY%rSr()5~d!vn~;4Jz&!9_11IqbX<`omc{a35K+M`&kd
z(X`OcS_Ma{D}T5BJDf`<xOC3tWYh30UyY(mm3|-inko2f5;K#P$49xDv;4p)bSOMX
zZ#lQeT!vdZf-binZVKM%k3Ttf&S_UKd;$NG?XOCABYwvhMYzB^`I#~6irJIdG6yqc
zG?~<>N9;2j9jd%5gSE>=4hcWo1n|s1V1i?K-1_~!zovl|Mp<+@5v)TFbEO}8_hja#
zTNXV;8}`BLk`2DZ=P6kvuthuW!9F4EvKtPbVb0joH%Ps~AUzuJd`qwoT6)<i`i9|$
zhV%d@@$psh?7tibKLeJzJ2yzfr@_sDXNpb0I}9GGI2kS${-<_7xLF%>9QSy|=J;{^
z*&SRBOjQaTlpVf_GjQma|7E9+UbM)$jx53qi&73IBh|&Cz4-S!;<-smGON)G&Ovya
zKUbKc4)ZqGiz8+&C+~d+>{0*>6AH%32&VbKl?!}RkH?-omx_<Q<@08MZ!Y4!Sqi>+
zj`wD7AN)4pn~C6?VDL>H@J%)Y9%Oop`_v^vRT&>nVK9#`nXF&vO%KUB9YLR36Q|FE
z=`F@vV++I&40n^V3OR4IsMA^FR6Q9vDKbwD9*}Rk1co_F_UWis`}X1yUg4$F%&Qhd
zyz0{5tBmy6Bye3H>Q(s>WT6Io*}Lwg&&sO<a5wV<@!K@<>Vz-2ho0i-53xG`01oFu
ztgQ6;J(&}$8r{J-A+dT3z9~~JR#(|m@BhW4jn~PAPxI(|3_Dm#usa05&F%B-2BFWd
z{tmZVEvNf{PQS3>ZXLu|T&RUxwd&AmS=Oyx`P}r3#K`-A3w)E0?wb>2r)Hy1Eu<qV
zq)?2)=`9}j0e{UU_{(iPE}&Ct6q;<;XywCu^9S!u7d-YYvVeQgXtSSzS6z>;i}$A0
zVElP8c-#!+EtwZu7Dc1-p$F$_q_V-8rXV}eq#yGD^F=}MO*-(+AhHLC(xETIx9)`d
zy}~^jyNWFM1aM8y2o>b43Bh}#BJ6HJZ*R`|@P%`$zz%vAW-==b1m{@cZn)n9D~HRT
zJ6x5Mle2i?(y>b}eb@tc15bKy9NakPX?pO@*V<_QCE#%~(QW+RsZHm}e5`S*&?Kk)
z;MG@mruPP)<XEy(UFa@zk!8F`$45%MH{ZZFH_3(@IJ*nx4^!(O^qn!U1{_9@qWf$-
zowZrPH$DLly=2a9mV<7$=XUwghvFeGQ(`*#-*|kb_%5zCvU9K66_krC)vr)E`cN%~
zzqv7vJ<*Y&>W%lIed$mICZ`AG3i~v{H__mmO?WxRvO{PzdEP?B$k_0@l<FqEe!J<S
z<8^9HPq-WW8DGld$w))@F8F3Vy*IWHn-Uw7cP2YEEhE=cE15`kGR%Sxrsp?j6Lju8
zR=Okc)ZIY0W;WW`8~;uFApIzg5086$0_RiVE%;Lg;iH3#{l}LxikT$Sa<gU(VE+Pp
z7s}N(>xmB;CHRtR8^Jf@P5SDkgSd-H?fuDY;}MRb(<YuCzY6={X*Pmkmg2t|f#)W{
zsJ$^peF`<|Nsy6!uSRV)!O?Uws(y3$8t{yzt`UC7s6OSuNB<bru!>P*j~kVduW9z9
z9~m!B$6&eyyVJQt9{ux3@XSo|>0q1*@Hj=`b6Q_Dv1fuDR~B?}{KmP+do|o>)+?~u
zq+Fa=&FKb&!#Os|qAk18;^5HB^Y8Ceg5Ixx!8Z>2GRAOTZlVM6ERQeoVdKkQQOPQ2
zQ@q*+=2I}E-!Sw(uuhjrV4WdW4UD1x9X!(rZau6SeEYvvRp;kE4QyH2MDLh~&%aD}
zbPgN6OXyCef>jCKYKoWq2$}yxw5-D|@iZTUpP9!TOh4maY*Q4iaRArH`&L!E!!EH$
zRwWeX<G-!=pTSsgMWf!q(V<ITNN3Z<`rzCs-iF7Cq5JVRy4H(v@O@XwRDqw9!xj3X
zuie9A{2mX{iDvwN9Ko&#Jn&VVbdKz_tLzs${#u7jS9lFWcO5W^$Ia}_;x#RON_ujF
z`PgP2zq8Y!CHllnr(WN8;tz01&G6tvb1(LB>A*lday$mW=Y%qc^qdE`%Ip6e|Nn6~
z>7DQd@C9!@^yqNE&mI-7QcKAeY=zS~0uDl_ih$GY4VHOx9`4`_T+K;zYcS4pW+4y!
z&MffHc{rRSpV=3XCPJrjbKaEVwYo}#_EnD1a=4tH)$t+wvELk=Q?D7>XLy~o%vYD`
zt9wbF^Y;okpMw!Pdm}>6(Wv@AjnJbP@IT-jpWk32bQ?2%!nc*cI-G3-8b#vWg{$68
ze<5euX*98_d&t7K;(PwZj_Wf{?ZtoXGXl@jX!Iv|uA-blotd+W&vWWDS%>{Y@vXpt
zMOSyK8{VdLK4Hq%kgf{&<1RhJlz(lQ0<$_57vt0&4|rlM{=qctpLq;Ny^&pfW8kFV
zf!e{H7lKcp*CR@U;b`(c1>-QEo_J608xFn**v7LBFEKtH7Sn1u`qrhr_=>jRm!1cw
z<RP1uBT}otXt_-M9^hv@JHe;Sl2g$Ym!GCb><WDxXlFU&*<p;A{2k}u%nj^GO8~pC
z0LQdMgIZ};mqYmU0xa}<!*jrq#xh4dc9WgP!#5v)%1ii<W(n*vIA$R?0cL<Ro^}n*
z=e$J;Pb~7ikA{2~z9a?RHHE<==tlL+u;UUOQccWaaI6tVIGHf|`@lITV&R0kgK7Hn
zarm7;9_x+_QWrF>-$USb1_aS37DT5Yz8hw_*v@b~ZJ1Y@aG%$P!>-H@Vg4SsFZ&q4
zHmlLAN5kEWSxmRhRCsLu9S7mE>wwD^7P6>%MrJ>7TBiGC8P1v6+hf*Y&X(6a#?CWq
zGMFVVneT%5Yto}b<%X}R4X&xl<593p2Cj8*JoCUeUOsm<_+~8lW-s`r5BTPMH@b_#
zH$}iV7w{Ly2J-pM$(+|AKVAuZRER&91>Exk&FNMguf1fXR+5`a0LwVy6xxcn;ckY)
z+4RdyM<1MxANv%tKl7^d9eA4yUhTzqQ|+)<bN8?#bFo*2dV^C+dsSdDx$eQS^bN$S
zq>-+gmh4p^^W<NIOm{YL%P$XnhDR-KdDQo`M_2d5!EC3;X0u1{H?RwrYb|@^Ql0W>
z;Td*;pWyvX?9a`|9=RlTDtvQm?lb!Nz%~8x+Eiml?2o^BDLgl$@f1fccIzS9R2urZ
zPn-CBcD;7?b1SYC`&shP&kcX`>Ryc2AEuX*T_VNEPQAx_vy6_YvGgZgO@X)g4cVyk
z(b}^kT86pkQzOX@ddUx(*y*BZEi8`5I0N3CS5c~h_h#B!d}z!C)dogscT|*WbYX8C
z`cx+Fry}5+h47sF$qV$HNS|L{xSUY9*p}?nMCTp}w>mH-K0MBgd1v5@calw59-)A7
zoHN}c$c#m3ORWe+6k@(YV_)|g-JLV+(uQ#Dn?VkoSz;BlP67CAQ&G-6=9y(^zh|Fv
zzny0n&MwYO^r<Jzi1|4~(*`m>*2LFXl>XxMa5yjFXwIQYt;SD0fjoE|=Sq90s#F2@
z;FsL?5^wP_x{LANyqHLrANXdB4GyOs{5$w&(pU0QHyoP3k2!XhLx1`@RE7I`1^mr8
z_?!0dH?`<JX?&esDC^Lnrr1@tKU|K(u1>O}*VF5lg&hr_LdhbtBZW@<l!;`-gUJ~;
z2-W2xq1v28uiOnXyL4TZ1mAob6r${OKBQ5Iw&L$t`hom6-qSBTf|YDG9kzH)m*X|f
z0KO^7You`#y~PPO6+kzc)WD|8#ktZl_wTdfF|leq=g-=3w3Av^#b-gQ`hezg8Q<SL
z?r+Y9ugo!hz&AxWPs(#|x7}z#GXmGtp#OpUcpu)}C9}<{ftTe`Yd9SIy9@BStUG6-
zm(HXEV@z_o!5Zz!c!NLo<uu8NkLESms0?UOKj#@WeXNn4>_!#z7*#6Rs1;zEet||^
z0N1o?Z&aOjMl}W7ysc-{2DqDRV4901jVe;aC?}6rxSTP3y<qa=>+ts^2ji6G`jdsd
zRz*!}$#aSLbN1oQDKi_t^FfoAyoO&&XV%Wra8C8eH>?2PT<0v}-?X+QyS3nOPNBoq
z+ho!HvliXrY-<Wvp2V3q3jTUE=i(`H7uAjg>BapZP5Xsj^$*=a^{sl1*7wN)p6F>+
zS2&w~u~r2;t;!E)bFz(9DFdwXX@rlGt7&7a{sj|_;rVztobns2|DWBjoMhGW%yiW>
z!$XW`=RFzk@90j}zg!$y|BUc8!C%2qV3&Y<;F{Zf4fvl0SHW24ttxQHs*YRnDJ;Yr
z#P9j<HJkUx!Q0IN=k#uAlm8exkkGc;JjI{&4?Tb3;G9HwBQjcvZ_qh|L(~gjdse*0
z?K-j#jsKe`$?(p)7|LEgvSK;ybimj(o1V(6gXpbXOIP<TvR28^Y0VB<);KiuxkIBG
zg(>G4c2(hJOQ;2plfcds=KH)Uc`YvQlF8%J{chZ!iSQY-U0Qn*ZsRTdMF5$BPVfZG
zFhv~jGIlQboFl`~o~Dzhnij4J>v<iAvndOfF@SYiAI1ZMc6FV%-*Z;Dc|JF~Ro?q_
z3x2?_lM1an8~BDZW!68;PVhLF%QI7fbEboFZZl71X~Y@SlJlqwXHF=-;wbt>XGY)w
zXXp84KKE{fD&K?OIn3vSb25GhN5Df*fWKK?AyNr=3X5{4%{tGHt+)v7NBg?@GF<r<
zg4NK*{W*6`%u4~}FN>ap_r<$+4}JUKGG?Sr>@Pdv(8$Z^ew;1a$uWeLLO;Lg08=>>
zbB$bPKK9}D3e&2k%vo^EIsKT$(5}AsVTZO4eR`MiHqB-J`v}Lio_{acBzruagzzyl
zr_$5BpI$Qbs7H_S5#PW^Z~@=XQu676<gjj|pB;s7S_==fBvRHfoYQULrucVS`FWIv
zZz{1L{}h>x+(vS5{9DT23{xc&=T!wdyy&W#kOKY?Zp79Tp5%f>3DxQPU2oQ<(Pq6{
zM2_ev_ZZ&WhdnLo;X#+ezn5(bof5n8sbobX?#&*91K^&^V4~xE9FAngayT4jt4(_>
z3OHq9k38H7c;`(zyuSs=d&02{twB#$qaa;v2X|wlyNk0cBRZ3BB)ltJP8F`5-GY<{
z{&f`m&Pec&Etq?n&wT;!bEPBa9oWXVDcGofkY@4q9`beH@b#@ea7C}li=s(&aar^t
z(4x~V@a72~hqG|kKX{IF;lD{`QTS)@&1HJG_%{t-3r9Q_oHK~-jBv9mC6eo=udf97
z<s~@gCm6>8rrFNr;%dPAcAoFa$HTxkQ@}Uz;G1ahjRSl$0DP0zgYPDsy;C-Ni#yU6
z*c4B4E%=~vU1ccPRd>^~7o5(UIDEy=c9QL;x5hV~EO&66$_9|_uF7LU_B5m<|4mM6
zAi1X(_r1Eglf2Y8uafAh`Rej2Wka54m%?%KPKS2k4W1mU`hD>ON5$%(f&B^ru}TZR
zIfK5`_9q^h+a7%e*VI8{dPYB=?|kyy6Fs`$4;<r0V=}W}p&1-a4Uc-1@!%KqsCg!j
z94Yu*_MR2PW0U$Z`=r>xP!+Gu+Y9VlV29cU`fkF=gD(WzbdSep90{jmU>`$mw~oQv
zOib<8a&)P5r(!f@MGV{~U4HTGn+xJ?w5i16c#X5rU;HB)KMCEIXYkddPbHttj@l9I
z(R9;|g}2zZPPC2|AxoHszP}gv;Z8>>ANVGoj)mXxbTLFkX=TSK?ez!ukOer1Z|*Ic
z^1<C?3dv4|gKwtNpYUHZdb6sK3+KN3!P)Q(ZR*ugI@+0YG&VvbITIQNGDmUVtcQPf
zrC?Ti2fus^-yt*Gfpzp?Oh${3WtLzb!l<dx!gvuggsZ}57v1;xich)paGOgf=op)V
zzt{!7snZ$WrYf1KLhw7uU5a{6XEFKJ2KXSKF)yWNR?69&d5CQIuXLQh^oHj`KTc0}
zYR!0dhI6KDv*Igee#)39Ou@{ilkr#%q=SFf46;&qi!+3Bw!`0S0pHl+Z#q3C6Ar%d
z$GfnP8Mja$dTPlJ2iGH0MRqD5*{Ov0p{jEnEou(=sXll)yWs6a8`w-XeEV<qI^M)b
z%uM}sE*aj05LGg;Be52F7kc%_ybD&Z3*>9?i1wL6?>_$2ldbWtmI>B}yEYBjfF==-
z$HPLthq?VSbA8vJR^8oU#qSMg6K~ZHmsNKBBi&Qt6L}G&GRM)TW-%Xe2HbCk)?Ey(
zn)|yQ_~zy!@XY|um2%85+{aJ$n-!0Tr7gKBXJcltyk<=xtF!{|O<Azbvpyy*L^B!<
z-dIGg`#}cq4tc1AGhh|GHqWM`M-2nl#2S^Z3mrIZjC$F@C<l*K8-i`h^LB2dzJg_H
z7K7`7vzgCx`AUFie2r?851ywix$*`^<?3u?)-Y=DH1w@gMs|xCbvG3ls|@`+{Qb!y
z*%Lu_yf<7<ulwwWNI^!tm|0o<%<>s*Ry6b0qW5MU%x6(i_^CDUROdKXmmFgL<ZN4)
z8qWYfiyJM2RGhPN_)N~hT|pXmopTUeI3=G|fi>`L<A*#Chttr8HxkV9E7+=Vt~W+<
zS74dd{%C=9ttwoH-vifmFituC%nb6aZAQVX9p}%$$3#Bl_jA~))_ci{{0El7%QN8=
zT+a=w;$HG+xuWl&b6vEm3B1o$`g%Tc1@JkY_}mNpe$NcFvNs8ys1Mz(i_oyHfTw)$
z`BcP{H4?l8{;h@=`Fycp4Q+z9>7e&|NU-Xx4_3ZHbPM(g(SM&ql!`qed(j$F!!ZVu
zf!X{mRO_;l2ZMuLi`Lo;+#a*iu6pn}3x56GJ;GkmMGl>%+j|^(Zrwp}4`{uQnH^5_
z!q;{J-&+z~1+VwpLcug{m(rnAHJRyBg`+NI<Q`s~jBI#Td^5%9C$ECu^e_CaksMVR
z9@ZFo%E3BILc>*?_wRV|;%s2wF52~3c#=W;!8dR@PtNi>e>q%d(5afgU?)}*ybd1Y
zP5C2~t{D6<7-uIK=Sw;CBlw)}H6t|BpB!h?2zs_7G{zXABUUgFIHx9$eN4R0J0sL-
z5FX{d5jw}&)d-EM6`Igm-u^f>LY}^OblmKjAs=zD10KShc$(43{`UI*0e@t@jMm-!
zucr^4uOv8T<NxvpzQ^rqNY?T)Gf`qr@|Qy$3f)1U;tB`)Hd)QRWEkMYs$>gO7JS|F
z!8aH2@9wGSR2HKX%;QwEpZJjBt>@Pva{+JFrz$<q%#G*qI~RjPuMRJBb~Ra$T~Rts
zHYz!D=xA`yYIyb3V4#Uh(V<q8%bJCkh%<Y25SrU{ehy36`}LEb=}VU~fJK)+WuMdw
z{D<_gCk-Uy33mS-LH|pBUd!?Oj~$NR{~`DPzh=$qYt|U<ug_kylE66Y)|u7eu2~;)
z(>ImUqO8m4p_*)#mpQ5bV!Z!*xEJAtkKZyY54lErq(yJJAKx{hQwHAT=PL3_=*E$E
z;COyplr}B=X)^A|9L!+<1gTcuAe}3NZ?FnE@ajQ2SraX*BH4gY<}_xu{ctruz&2l9
zLHYo;c?Gtq3AU*SF4_SmdR!N6s$7un<l!95hL`69oD%#|zKQhqfyw;CEGpf_qGI5i
zboI%KS0f)@#G(pWEXwrGtU;U2TDFwAZv^Kfb6;p2_$`JV!w&EcUf~k4;1}@918~k=
zc$~kO<}}w<uA^vE#lSZn@XadlO%r;H2l2VVJ-Vu`m**UG7YB8f&4|yq3p=LT)8|);
zol}Los&ZPqJ3rZ-3ARbN!|te4V4Usr^DQD5)em0{-kVhZ?8+?5uJ0V^Q$N_}jV85^
zJyGT6p+%uHjdI`{ZiPNYPfd6xu+Q^Y%|}~GP41~9S?*Tgn+D*UTHqT8`xNe#idE|j
z^x3fUJMNB0slhj$c6;<|4ftiQhm5;Nb-_1%!8e=0H`Cc8b($Ty8%ulGDeO@_vQatd
zwXuD2tN2T|d>`T!rlWYwX*WFpcy8$4-o5~xYA{@lm*>!>0?AB01=}>D^QJ1i&5mE}
zV7MNmDLZ4>Rm#4pQ85}#r)5_ISM3;fK;g?F7aoT9<~?1X9(w)muSJiV4#pXRCPj9t
zc{{p4{i4;kP_#zT>zD2c``^Gf!@xJWz&AeqqhxjAXTxv*5RNlPu_#5O<4!@xJ-sth
zTe(+uqfbR!nF-LQci>g41*dxvk6zuIbQd!V?cpA3HXMJkJwoN-h*L0Enc;0}d?Ggm
z2fdittRVAR54f8yec_;)>89anE6LfnF(2JwsmUt7rxW}RXXSCu$Bp<FXW%RDk0#}S
zvuVNem0fz2hv)v_3!wYv2)=-G3!GX67wjU(IE207oyy?9Nly;s3H_S%7Qa~%rWWL2
z2Xm$r3}p9PWi)WQ{l>i`M{*u-<wl2^(|yyZH+mJkeCcNREb%6+fxmx2XKlI@oLS55
zO2v+pXT9irv+`J*Y<Nk!ZZpthORrys9ij54EAB)rnJH%F57n8IbJK|fz9|WRGa3FS
z<3#+$vGjGeK%2sQb1-d)>b^pkI)~3@<NvygXY%@#Sc|MpieUNDeX@EZyRir3Sv8WS
zsbf<RvwMLr@C<9=2_`e+!!N9bYgm@ss(Zibcz6^f!vXeG&fvc09LUBjv%M%<HRs2n
z2kcNG5BUiGxIk$<D%{IocCc$|q*+<WP35SC-t0?W03UBQIFcZ=p@BW|+Ti<jRL2KX
z1pef^QOWKZ)$52+^YGm)n`G4NL13H+o<oDWS&fh9FzU=xgBDyf=+;St&g?gkaWkm>
z7K1A8FlfL&gC1TsDCxa{?pC9silS-N1_#0OobO>&i3vspk~JSdwtQX^SPBm3Vq<tC
zzK^l}O<IE|$8!OG=@0MI(=)pop6Fh9qu0}egGLpf$s+r|cyYSJgOi1?%07zPU_0wC
zi&EqY(lO4>{+y4g@FQMb4+nmp9mC+l!Rf7fR1STqfmQw;tjcEkA7)v=<N7XEE$)QR
zsk&8DYvL{D^C}bu>y)u7s2*HTL#z6>XV+gpt41s#Hv&G{1+F={3@kH;&z(ju(M&#{
zkNx$9R|VG?;BBUWXHp*pzwr50Pg+$7Y-HSlW`(cnS}Uv4)Uj%4AbDG`(;R-!_xQ7q
zz(vLIm|x)kMY1(?A(0=ySA=fkM))K`+1-HWC&dadGrc<<`2SY;ZHTgzq)VGD)v#5e
z`g1Z=x4-h}%U+67b`AbYmL0#?WHRwB@sqvi&5lES^(A?IE(P!Kv0Ru+aqoO_hROFh
z-jPalObv1>D;iDrd@iMd2e3xEbT`4Jd{gNHBo}a;Ox3loE{%Yby~6A=2+b;66MU_m
z*ijWoP70j!m8&0m$r^LG8gUlbdXoqL4;-{3Tr>94lXWm$KAag3z&XvqIGzXiZ+MOF
zjmJ1&z6ia>XB<<U&KxjK$ujtfD@5omcwjc1@vue_>ef0!mQE4+k6EieIA<LEMJ3+e
z2-fM<pU>OO8MO(`ipOPJn8TRQo=yVubj2GSPG6V#Z#Ij+E8Ab*hwkQ7F2yx<s^*q3
zHCyFS=GN>eTW{APc&}W{KNImUEj&ff5E@_UB=l=?m_-KTWt!<wHL#*DIOiA~PosI{
z*4Wo-?MJWTS8|=qV+~q3wP%V`EmNb#uXbrhDbB;OWbB`yAq|1ULHo~&7hnW)<?Oon
zgV4^#o5``pagq6G2S@V}EE7F}y;z*pjk>aH0*-yuE3~`6|3+VQxqjsHSGhDg5x%A)
zJ67=<ekS9v5slW)>-^<^@%E!ZM^vU?GlY8{fByqG+&5=TI{wx~cFUxAC;5T_Cf$Hj
z&0C(`ELAuox{*6DoAtFf-RJNeiD)@3PI9hXL-&4QRxvc6+u(<x@Tw^+V4`94RK<gD
zM!}V=hVS8ew9BHFa6L!y>EvJzd72)-Pk!#pg5aFo><5N}{({c*sUEr%m?sSUlCLbY
zNhN%B@HCZqu5eX+Je8SKIP02l1-`Us)mCOYzW$)u;2HSby<O3#LgAG<qC?>)4)L>S
z7MxQQyiG0Uy<B|X4!Gb66WCLe06vQ~s|;PDkHW~5hH$oo(MvOj+!6Vt`Q()flV8e8
zp2?S7lhMmA@4oa2)1MbgkMIxj+XdlZ@?45jX7)t|KaJDuj9oP>D?CkR^5gXPMJ2_l
z0Q;o|usd_Ydv<!`DV_kfDaVe?2IQt5;4Pj%13bgmo!2Z*m#fDqu~?kGWFSNJft(XQ
z;@jv@2k;KxLx<XL^XgE5m;I7(ILW-cHplAl)>w9c#;Pp%rYrcS1Ni1`NAOL<ScQXc
zDuQoXgKq|dZ@g$vDZn?&!8g6ZH#xyK)xbANF&=#~!R>%=4uNlafNy4^OXURLbp8XT
zA^%<a5t*oKU>i7_fZc8l#ZNpGKk<~2ZsvZviJk1ZXP0KXf6=DUrH+<%>&hQ?X}*im
z({pH2+qm%EgoAG;#^JxA^QN?4jPB4`?1T5F0N$I%*Wqr+PAy$Zcll&GEW5$k;4N+*
z5Up2uZ~XDyH~vLd>VA|GPvRwBL-!P%XRf|vsT}B2t?3o786``0dL-V0|1U(!e=9ol
zR5Yr-_=~|ef4j`b;<*dX6sd1tIZK!izHt8q;N|Nz5H2^E{ow)3OUzUk(?lo<Ub!(k
z`lv(Unzsf%8joI^?sy2{tA8`o4XYfk&Uw)0Q}Q;v_F>M(B1it$?N?(m`xyF?2WLjK
zkln~t!6jQ(mtKEmcD&})^qo#UCX-+U-+T&#-{t(t%X!f@E&XH<(YSZe?}1OU%qX}V
z_U%0D%*+I59FPvo^A?W{_~tD9vJ-qW0(|ojeX46yhtig1&t?X8h;vreVJ;gDzDYTb
zv%d$OekNWQ{Op=k6t6-myV6_@)wlmbb$xoMR+9_Q*FIEX)$rrw2-T$@a5QwHL~Rby
zVER?A&|T~T-;600qR=$(IWO>LoB`vkW#0<g!c6kRJu1?BlN?{_J)1JY-@F-!7SX|`
z(G|$(;JrC;&`N)hRW|TV;l6MV6|K6R75(BHedIUcYIZWyOo6lIH9UVKunzN3=nwQ|
zbnyMmPhFT@Ugc*`!+kn|nYV^ApWW$9e{p$wZ&K4Sa}Nz@Cz&Sr8!MX71`EA6oOz84
zpc%bGYr1aKoV`Z9Ux+4!?`DM2r~t5y6KwOXpi!&S^Y#OSqAnWr@rXfVb{n*9t3ky!
z8Wi%MLE3B3uJZ;h<8$)7G0=U&t}6a)5FX_4?ndnzZB)C%Mw$3~Qhqn8K@OAJ)aAV8
zJpHF0{X2o|GvCj-`hg5rKIX~F^bvaS0WKgXewaDyIXTY4^jOrTPkl5zG@4wcqhPyt
z_;5Hk@018qu!Zc#Kynq!$atI}<MDv*Fduew7Ubs&7q%Nd=5%MPcEaQI1=kqx+x!93
zbl^Hx8cwCSRVnhr-Q=?>Q&DgUf98A>G^Z%5hKxryq~~MIYVgcd{`^QXSfj|FkA)MO
z4o?F%{R)OL?&t5~ci9$fQ+ETH2@mttIIFV4+kE(eXCMQ=+Z1p(%&nvOeuoE<m&KQG
z<L}pffMzui>@>?JA97Zf0i5}8RP(Qs9Ru^^a0V;=D7u>QhF+Xbw;*0;(?RlB=woC0
z*tCxh*^^Je&K-m0j|XhkAv!4WjLkUBuAmy|v1qXwU$U1IFY-{b<dJkKq;1Sj6Yime
zU19o(u9CTdQ+1AV-({xvuNE0c`Zw$Lb19Iy#L0Z}@S{tI;A9&WBqIsF$ppT+)s`~=
zjMKMExGr)H=PF>pgU61mD{wifLeQzFpjE-)WQW7K!DCZ!&h_J*8+YM(zJ{w?270yf
zaK4~X9Vv>&1jb2S43A6+G^f&JpR2(o`!RR5h#+f1e>$GOZeXT!e4QQOru=KsxjN&k
zZSB&kZDD#9%D-hMyITGWqg&9amxt)oy5m%>*X*(*!?5QgIkIPHa?_o7R-9V04bCPW
z&bNU>QJf1?=Am1!z^4t5b8Z)07=4h%zmZjINJfDy{`)atoN4T3LIYpSqXEqGJ{sQG
z$)0-noJwcHR6nCrf6&*<ymhMa0GHfg!oR-kPH1;6Jj{t;l$V^fCGk9_s*W#!Gq^%v
zq~?V%V}fhu6^E~R5TS$)?D6x%Q`4TG=_vjzT|=9}N=-k&fu~}3aj7sJtrDgrX0)$l
z@EWJVUxruzANZQEzxOwq@J}@87JtYCFk6hyXwv)K?2P02q1=aA`JA!5e~q`lXJ?-r
zxq!%KCgqxB((sS;aR%Vw!>?jGZkCbU(cissr@1G)C4qO+;5CNx>C}o|D)2=sW{|dc
zUpmCVn~X=VT5nO@Ze|g<o#Rg|+L`osF9KPBbmXeIs>8+X%ucsg7C4>)_=v$j`FN}X
z|C1X1eGl`?sLvJ^$K%uK3c0BL>|fyCZ@&^P9*wN`aP+Z2i%O!m<>UKk|H_Pi3EmnH
z@oaF-&4cLMV6zzu&Dz4uR~)>S98Ky0*e$x1Sz~#uM20$3uvb4@d)255eRyGbaR-uB
zN~9ByOj0kp^m5UU=iW}A-dC?4{UMW`JWk2d$0<Wyy7{c^NT9#QfM@s-*MCdFGj!NY
z>mR4aa5OuIuuEYvx)i;`Tj=u3#>dNmmkNi(NsZ!kutFSO6m~B>@#+~~n~QjDI&7iG
zhD?+n`KRY*`h^>K)x40G-o{vVaK`HO=2*>Mh(8#7lLmZqFEUoiz&C3fkb5c<E73VT
z?4w6J!8Zp_dem{3M`Krev|tYVzuB>Hh8<ESher+QAx;6lxm?Smi>2r!2H%WM?@_bg
z?2&p$rw!aqKDe9o$K8tB;nv1gZaRU;f{$RoeGj*)+uhpK(XEy2V<<o-d`SVf{;-eX
z3i~w2o?*Ate=+!AWBA#}$RB)jxC6QHsxexD_vR7ao9l1cO^f&DFxph)VmfdpvPUzH
zJtnqjy=+XcIh_e_$WHlvr_=IQl<L!!P=6iWestgT_C{$V-r_6u=}aIy^|y0;*h?~n
zC*f{3M5@Llv?w|t`hai7*XN!B-x$%XW4@q0p=saR1uu-oou0WUhmHNf4e2Mvlh-FT
zy7znTwX1Yua$c<=Kans2?Y$dZH0N0b@XdSpn}IpG50k_3yr$RhIy}uGIyKhPcQc6@
z5N#?K_+}rz#w!(>4Kvcu@E*PTBK*xZ@=|l0c+{N=2j3hGaH?uC^s3)zQn$mjmow`K
zd4*9U!j#+_rjg*Af2*TeWroKA-@H1<u5j>8B>1Ki_-3osq2uUNKZ`MQ;j>Ht-yGR*
z*D}tr^&{D_>A|ZIXjf!S=Cp$3#Z&z4wPn}sR`xJVX1_{IDEWE3oRvehGi#_k-$K;m
zG(9Tp1)9xnhDprRsoRC<S6MJmN^;{*=<Gj6R|>B~^#;;O+nGK;yf=PiZ0xsfI>l?%
zC;T^^+S>G*Y)!h<@Ho6Ko?3}kF%dq2*Ws*X*{hiu@9Rghk(cr7ZKX3}BD-?KISb%+
zZ$GkVAG}Rm9D6mJp*i!~KR%fSZO*LwQ{imfXgp2Ls$LYo?hmrx*YWk@EnYd8EOtk5
z4f;^Z%$#pvjmyvBXfD!EyxpiTXj4D?8x@AGbcJkGKw<cnEa*$W3`)3V(4Nx<Wq`Yx
zvVm*4L1~v5WaDw|W&=G>2DQFnP?h@zO*m&zTX>#p;GV)f-|3z~N6B54gU9iMCwhB{
zz2xXsBR+zu@{_NsO=dj8q(-~x$jN0^rwaT#ID3{(qyvZZDdDDByYtX_gRZ`i^Q_Mt
zX0#pnj`=yH`$K*l-g+`ReMq+;ou3t?vG84+ZsIlg6r_f^@Y{f2HrJxV1AomZJjAEq
zbLv&Ls(neT78awM3_LRjY!k}oF3JfeDZtx>z%=kW=9X|SL%}2q!9U=f4MV^pCaYF9
zWp@T&YcXD){tNITgJ&9VL%#wCweE}m1}wGTpMJv1cpRB)Timh8Tbj8P{;f!3X487?
z2;|In&Z9rxOMiSfi%Kw?KI6}%LzBynb}*dt_-X-O6ZqUyXloC_KPe)u_&VT!M)3E5
zubw7Y<$?n`^w6qXh3Lgb<J$Reu+A=__t-B~P507gz`m+K@pj!f!w%UZ4&CCl+=5PZ
zrzC#;STtVlse0fWGoINs<bzhea<YrgrNa2*a`j>t?i%>k!^|h}u;bH_)hWRJ&OLs$
z5!j+NSff3DyN=))UW-R_6$I~`gv*JA%X!!a4-V%-b`RYV8|enzhQ_r!obMU_hrHF}
zx8!`1|4%de1cy@(jFTK}<Im&aB6x4$ax%ghcep_A9Gq06JDII;zFs{rjSssA@a{fG
z`|h(GeVa4l1suk_3Ct)x+0n)Pa(EH@QPHCdlQXMLmh47bhZ+ZQ7IZ~h8cFZx^#7d|
z+oBvgd6T@se7dfu(igFm*=HU46dX<lo?rTv9g_7On&M*TJh_E;{5@Y1@#c)6e*zsm
znVrtx>To*g96FcWq3>wnznQJ}zjkOLy5IeAE``HokLKqT@&#WadfV#x=>8ufG`u8T
z9RlBUV%K@+NIX&GKj6()f-|p!+4c-~X)gYnlwh-Mb3;`U{k%Y3yIiF>qj<Z?E1Mpl
z3(~Ne?9{HuYxW*;1l8%I@-?X$_xEwGnsBsxo*C8gvk{*na|ieS)Zff5$xXVDmi!{u
z#Ps;(=!#fb(WHI>%o`KwvwA@m;R#tt`o^=ut+s&!$+5|-G~k@NFVU`2fN`>0@JrL1
zQ;DoqBhDrp`Vo9iA{<LP@W`aYbnYFo&?m<{!dzDSC%Y4VTj)wc&jRmMVMckIkq*EN
z>{<Z#OihX23>NzIiF+QNHv0`YApUH?UOdqFZMv=CYfQs~hfbEgxkYo~u4kvDzkL^(
z3b2fOj#-Nrac(RH*YH@J_lKd!RX~^eZD2Pa^Hbfwntd_29Y3$ym7)g^eDgZctDJ7H
zHj%HMi0>tODm{6dy)vEi>Mr;rJ^aeE&tA>=POn~8Ji!I%v+;{lwH9&O1?Q8h3;r1c
z{lm666~a5*I1G-*g(ejP2gF|Q^5nsfh122JfxS`<<5Z$voQ4!<59V*L+TZpn=A>86
z;BI_Zd-a9C|5lt=sSRG8uR(83Hm~-Q>&|&ORw=<Zt_88mGKL*dz1jU89;@&4@;$E)
z)&buH`ttmD`fBd7Ljim<XNN~~SK%9;i5@lFqn2KeeC!^5?Bt>28>~~?qe}R1lE63H
z;cuq=LT`Et7xTc)K6tkt(@|V(J9}YQy6L)g>)TMbF1z7&g55gNj=Yq3Ea%oaw5d1W
zV^r;Bj53^#VJ{-PW*5+3oQPHhzIon025l!st8&Nag-?ub(p$U)@6DR6blfbACd(GB
zgqUbK*->(pp3DD=Mr%<<_U^tTD|J0e-wxryfxp>3oL<Zrb}`_+xk`8Vw8HF9L7NJE
z4t_aKSNR$`H{m@CgKwJu)pPyXJy(i-RPd!Wx!-bMK!e)KE}3ccsrTW1_?uhx;EI{6
zmZyr)y;tNsFXPVx--IsX92?JmP0pRXHu9DM;2rpz`0UJioIg2VF$<CnPr?KD7@l^~
zMD+aLcyL0%Hci<d4!#+Mw>TrSmG1>QIlwnVW;<noFP}vB&3Ava?7Vn!zLEc?ciy`-
zOeN;h={K17qrf!qvVW?wb0-%&ls}V`0^f8(pXvg>dC>z+s*6K)>)-(_h<=`e{y*?d
zUGPm#@J-1?`u$?qGu;UfCitdMe!He6g{m9+RMpLN-Oz=<Gb&UUo3mfBJlG}+*-E?*
z0qmaI&t9NmWAWX1@D;bk^I4f5xg=)n2V|#q1?%w~yv67WyV}y#27lA&5Bc9~Hl<pL
z2WNmywsvG@iqL(MgwN&-I>k&lfe2npdCly>>+vmS{cLZ^JacaSL~~g^j@N55Tn<`H
zi%S+&L#K%X-!y@fH59ZceRA%1a6mA6x&!|8h|te-u%qF%Nt4ghzc&MJB#yp4d@tG1
zhvI*LEw0kr*O$H<8|PeOa86tBM<5sn-*92D&6TW1?eM`{gYK04szGax8nhK%>dh*H
zJc|sfHQ%81iw#-<-YK=mpf7xU`z3=uA2#SY9M4m<tB*XN<BUO*9vW081>8+Va#!8R
zny(>eb(eFObG0IPYgSX{$8MaX)44cP>*B$=5RE??ymsugSrNYQ(G{5Iz;pTW6`X;$
zYP^p=vUe5@BP*U-@Z8|PM-#}F?ttF{7skCNw~@-KgGJ~st7cWNx>hBWV=sFJtD1pn
zHs>Kjm5+`bFio8tXiogh--9b}a@oN;Ik{GU<!25K`Wa@W-_EKsLG(-WJG$nF?gjoy
z<n3`Lt19xbM_`@8J>ix(;};b}AIM1`Aw2$uK^DdAgfG9$c@H0xhO@fDAv~HhnL86X
z^AFI$I1Rreo`7|A%vztB9<n=hB%Wjc`eVAD;ndUg1n=O9Fw8aU*FgNBW6Y`sUl)RB
z^flkd4!2dk;f`Lz3BJXnoWe%7FIT9tv4^W=0lN&m-uoUwoAu@Xz-#<#JG*CdhpBKh
zyQ6o9Y2U9fm90h(AlaV<H=MegA5OIa9h_d5>chES-OKCiEtgV(ZyFYbqXFOas~=7`
zI(aBC&M0`Czq_g~f@jXRAUg@SQ;CoJRU-?&(xnZ9(WhhSMt<&6&~MIxe0*+M^5(wm
zuHt;S%jXQ+O!nZQOI6Wb(mnzQJSML&3%z?dIrW}!Jp<t{9Q<AWbm){I7k!^z^z(EQ
z?xGu%eC14VTkpTV>elQLBM&*R2-!0J?F+fD|MptVD#83x$*$%8aJ9|hJ38P^GNVby
z+4UF=Z`W*kKEXD((XFd<O@PO_Ne0bviRYP(*41>Vu<?I=L)Usbl%<bDgW}M&;B8(8
z(j#4oJsH2*Z+08s(?VvbG3598es-aE^<}=gI)MGY>l|ufbtq*8hdTEs%XYz`^4ICD
zc!#gqmpQF<m^Q?7mXX^maX(bQ+uJqrQK;MtLbc-xouSN5b?B^lc9h>aJmp$&`J9*d
zZJB#&4C4PpDmV;gg~G*68hPI+&kg+lC+TnCdT`dLL)VOWZH(}nbdf(ni+)Mp1CQOn
zKAk^sKc+D$1DwtvFisohj@9T@!?}*OXII@?c;8aYOc7>X#;3dEf>|k^!4dxiGo&WJ
z35QanCV0LB+7uj4`~<j1xRyif*om-<tknr}0=F!>iq7-{?Be*w^Y{n@!82dbu&iL6
z{;9~2^IR+#Xxm%7HMf~*E?VS*yXl43u<;0dKg<G&oB{RWt$qD0dY*y(3(R&!@e!p4
z%M_o9-)27d2H3_A4!GNNvqtpe<4&_G`!oNcgLi+1_wPD;@+yEI@_Kc)rB}znH<`dU
zm%%r?z!x9EH@Cqzt-v?Qo_bXbeDf51GZcLD5BTOF_~ueg`uE^rrnZgKHZV>e^4|Y|
zZyMtrjs)XWbMkR8O;>g;Q~>L^cy56S45T=ftjO2M0tR}+F7J!rtOM-#UJI_mdvj{E
zS36?px@qRslQQg$`W>tNPuL4}k$e>R=IFdwEgA<`(+eKQ3C3ZcLV#k`qAb~|Ozc#k
zzs7i<otVcxnz#jSW-;A-?EmgP&?B)!YHFwlZ5b^Je6t7sreIl*Zh~);gKz%(Om6Bv
zdo#&`2OMx~)p|OK=eXrcq}!&i8y$gN3^qK)?cFNZ(5*d{`Mkn-kJGVN^C_Lg7w{6}
zy@^>$w;$f(TK!{`Jt9T{fALL)7|qKZqdGsMb@m|~&8cVw<Gopcx401AV*lu9UF*o6
zHN3@f1?eoNH^KFWY~iIS6=av&vw2Z^JcPXrk?h&*fbYf+&kfy~OUO?BcmkHei{Br=
z+OF~FQ$6UQf)g!|{=2jo`N`DmYIs3@@f;cz_@?X>a+kfpH9=_JXyCn=r=sC+Qh{&k
zUV@W`8?L#8jtF$+sqi<x;G21zYrQzPl4s)_N+K6YcIwq77k$Al9c0G(If6Nme8vqE
z*{J~h{>+Iv({X;{`*)F@N>BIA&KY2u0dyGCeY2fx_^<46%WvRe&yi8s%$zwL?s)*X
zCY-am9hx`%Y;o|7|2NJV^r`RgH!Hw5JNi1{PuM%?$81)BGlaRS={>uua?YoQe?Bza
zuC&p1<=~vFU!8f4{7DEt&GXko)g13ab<UX8WWyUZrB97)c*TsNYWyxli{WpcuL)5C
z`qZtM5DjAY)QpOB`+;xv-wReo@Qsn|)X@0g|KB?m4!$}51z#~a;_PH?eB*8E)yk$?
z`E6?Z1AKGBs<HSunu2fUp^04K_4Ik#|9Oiip2TOj8cli>9TR~;`j6N06KC-C&4srG
z-|YPt?lwO<_b;*#7vVjI;vck`)u$#oq_k$8xQoudnY_{j_D<Q^pUIh(0H!#11@2~&
zk=<GNhVc<!2h;4YY*aDys1s?yG+*)8a2@z$P|C;fFK{<o;BJzyH_+>2P?tIAP?HSu
z;d(vQpa$~{I<?cFjbNIhTmk4;b2k|@gU7YU4XS#V&-o3{Q;NSQ$f#Q5$&l}-N09Ti
zT~?E_GjGOrhF2O7$8>}qz~p8n<KOTzl<tao@ZxaerF_73jqw<Vkd2sZQTO%a#`sxu
z_=pBy22Y|1FZKvHuf^cPJwe)bl{`Hdu@ailCLg#K-~VwmpR>YwWP``&I>I%c=VQS%
zX~9C}f8l9NhmVHOUw(~T#~E@Q_<FX*(@!>?zSX(tO8kDx^13oEuT?k7Th;9aXYpqI
zK;Rhvy!4ELZJLrtc?5qyB^%H8pnH+IbK(Juk~rrd@_oA&lI!nn(ZBr8qSiB4Cc@P)
zTdpesp8L$MjLRl%-V5gAclQ}xF%i9CSs!+ljxwpvaz1{{r003rMGD`Ogg$o}zgCMj
z;H#-{RO|(6T%SGl{9lZwm#O*eP}x4x!403W`U{!y_Uv-NZ_+wbm}Y}-e!$;M`c9sT
zd+o2M|J7xu+U0U-L49`NM)Np^ZYr_?7r-}r(Whn>50^hP%u6_wr`+$wTXOG%ak_IY
zZjN7==X}976Tv_q3*o<6MwVb8{7QsNk<IB#&cypWxUX~3O@5HPQU5S?s1T-J-@yzw
z@a<l8=q+F04KMVi2i<_ZIAb_pW=C<hzzd~EA9JUnXSpT22FO?DX^odRFE{~=($8jB
z2mIA5;WyqEX18iSyEfz|SC$Km0^YHKaZ1!fZ)$=UyB(PV@Xf(cKJI2uSzpe8q4Wg}
zgBuym$LWvogTwi|r|gXnc`3NBNnn`GZP?cvf(DKzH7m@aw(vn~c`hEz68M2x=`7hy
z{_O1jc6DT4`bwVv8~a;I{R~#}_rYqIK13t&>vf(KtSj}n*Y5_&54@YTKAa@?^KCf2
z0QksbJ;)O9|0M(+rq?DscKOL9+3?+gYyKnyf5T7gMaTU~7sNzf$MbWq%vnU|RAYFX
zhepi>&$QoZRM!9K>D*z|Bi^oX(5O{-;Vyx9CY)h@;T)*XoOAg)9=XRx_NN<lI~`|4
zUX$(?rhA-gbtyEh7VOgH|J{n5cvkp-`UX7igYWqB75LW2<Qcx<{p0`pOaAZw6x@rE
zdjJk+0r%;~<releSTv7$Bm%7}?|tTyR~Bu4&wcz0&Ihh$Lvrv9n$%{nO<H)J9ADr&
zpTSF`!PhwsPt85^WIFfMK=>lCQ!CB^H=NN>JjC_inze90n)WI%4c}KD@=?v00~+%)
zX~Ws^o>|~G+)e%t^!hP7e1A#r7ucr%c@sN>WAQ<GwHa?q4|-vWosHFydS0Dz;^zh5
z#4p8{bByfLUGNP!WG5QccAq$P2j6@yN$v^$W=B)>Dez4q_+~X6PC@WZ7Vynldz_BK
z;nV@&9N~Hc)-i)`4uWrH)MqyXU*jlW>j~JY58cG+(7+r^(4+9)j2J<73g1o1qF#+m
zPBxq!-=Po5gP)Goz)kGJq_?Ix_$H(WS7@wa+Qw=G_@+uFyvF!%65o-P0^iIy!ahtg
z;9KzC{6L$!IKZRI;2RtGrZD&>4fv)u_+}{lO%?pb&CsXrfo-mlmFl+Ft(F_;?px^A
zuPJWjr{^YRPq)fP;>igjH`T_iKK0>vD!SQU;?`aIZt~D~GahYf{`MFZ#ap~;e2i-J
zi&2MgxE%1!aDV#!(#I(CcXou|i&p#z`fb?DFn1<Bf5X^O;=*f;x40JG;<NeKC62fF
z86B9b&yWM(&MvpPQA(iuCOz4yw{6)CS3OG4^V0#IgwN&?yHO64n_5nH=Q#GPc-gVr
zp3bP+kxElIQh8F4qXge1kqy6uNALLL2z~B`R@|B0z%}sZ6`+SJC3+S3rtC$0g}ZoM
zL|$SveOPg1M6L9la^^M68?JRJ(W>y?%*A`t<|r6v9hp<+z$5W4O|YZui|$ms#RZBm
zuYGYU%_XPI`<%M8+^H*Y<j1^ZsoFTzrZOFWWTz&O4L^*oU2Z)&sVQN)%$XQ~KXM!M
z&Y;R+`X>jPHy*$5XPAw^H|@bUeZe=kEo3~IyRv|9CUHjXKudpilwHyAH~Haj?nH1F
zan8-4+plFdI{!Y0s_iwrncG6i(1t2YM5yL~Z?@y-{F9C@<5wZ{^M}Z?8Xq6Kr=CTE
zcUp#MJLgpq@Xf+o%vd||6HgA758XGn(F%^|U|-@hn?7!}X&&B+gF!Yus%ujnyv18S
z<7Z_SyR+D;x8Y=z8(Y=jA96Rm))simYv~buv&(p0#aoiM6S_6?jo(QNeRl8`ZunWw
zk`wR*37juInO9D5*0nR4B{<kkzfIb80X=Fl8KseU2(9$$`Jn@mVTwF&)U+||0*|1>
z4<GRK>PGp3J-*{5ZV1nE<E=sCUK$jLuei>2gX+P_tiezGei^uCra{(m2IWjNXam=U
z@dj1mxgUIN9r&l&Is-m;gJv!<s0Z(tL!&Cf*YE)^btz=jNbuB(VPvp4ce`>9pW-~t
zz`XdZtw|;Onv`yxN#h={_aQs|Tfzr83f^K1Z=>7zaekg}%}h54?wYgHx)IOeBZ~^=
zfafA}@fMA0{iGlzgY%kpJV;9(lB<6Q?!l{l>nFWpf5>2@M4QTpR~WwLFHbfUEZIGc
zRT+N-Y1tb*-Oq!x7~bXu*fQq|x;OB`4C=$4W3Jm=qq~wr$3ru`6&*yM*&V@Lnve5*
z!gsT#Fwd6$W7gIr&Q;#u@C7c0-%bCtXlf<#Cyb$cc_O~4w%{NivbOuk!gCJC_J{Xx
zYgQzf<@$ZJfkWinHt{o9hIeW(T|q7ruVp5^9ZDxOuVJtGcp9|F?EJ2~KQwFHS@4pF
z3~dbAF$=g1930OKyq5e*V|>fG=`C<~;WeL3l`ReX*?3*<e#xQb9m14$7T4=AeJ0mj
zBiyM{@Te7h$vQHB{BhzPK~tU!x03LMo@4M$-4bNHE8!L9UVqX!Txq~GmzttO!QHyh
ztQ<TaRUI9RIp;ubunzcUHTQk@Fqf*;b;;+0Q$65l?_^<@E?h-on=t*y{rl|>^T`f$
zvsvs6o$FAWZtR%Ak7sTVhlGdaIG&!vV5nia$*On78(q~QM>ntv^GQ8$N?jw_NdAo%
zn}T6#gG<0Slk$)!0^77;cDca&u43#~1^c)f+f|{hUGGeG?I%C^w*xB=7-vOydOj2I
zDseUxolH;G0?vxlc4ikip!^QG%b;!5#&gr0{k<mqePEk*oj7wE)0vLP@55d4Yy0dP
zI0~+ezw7yacGf***G3t3(8JrNZDZ5yW;Xot>~Fe;FY6}!MX?~+rqiq2kk{a@JmPg3
zQ~=!0@4n<5c+?qaODl}pgpT|e&+y#I_=S0mHnu~9$wU`sI`BmWvo8LGmql;d{?w>N
z>`|Gr)Tq8(30%9#4n#6jY)&xB24`DsF!{z=Mx9>EeGlH5blj*R=ZxwP4yt|xjtA`X
z7!7Nt551h(z|&x+3G+<qicWRAky&?#n6+l1Sq1ijBMyMGuXFBv#q+{hG^jFor6p%l
zFN>B90i!H{*I9!Hn0vJ&TGhv^U?1+;kMKB$zFJfUj1vyVxyj`Zwn_ONPYp9e*e#xC
zen~utmuDsZnyLIB=WE^x1pn3Hxe}a%snB&_n01!#;|TZB4$g|3!E}Ri?Tz8S0{^8C
z0EboQXA3slM2@uNUe1^eV76^&<B4z}$7AJQ8mpb->Ao8nt9A6<-3fyOv6Ht3J3QpE
zJUq!qu1)Acv-^<C?$6Gq(R|J%umo6RGP+SCu+sUoWS;N@Z^2iyu6CUAH~HVrg+Ovr
zTdKvWa?v=AFBGSxzU0E;f<9k|6QaYfFxbX#Hajrs?b~f(M`l5GMuEH5eTmgbc$@sE
z*cY`0?P+DK2G2*EnjWjWeD90nV)fV>tIFV;So}9Fs>P~wx>!wl<56z#&42qm+6%t9
zKii{G;G1>en*{JpMexn?jvfsR@TesCW*`3I8Spom$xRiz<5rckZmq#{6SmfkhsLcB
zWWh6!!h_R|om1g%9WaA=TDf(qu3Njyy0tSiUB*9RROoJuT9FHHi?_HXoy8y6sX3%C
zT{q4cdf?Ed>ajaK4gAelc9GmBAAU4inO2bhCOehTJzDh~(Q42-T1RTogOeAnDhXcZ
z9{EDLZ)*KV-}fAL(+$K+?2OW)mb_gBZYNKaI(=gY%N_RR((N6-I8wbvMXFvbdl~TQ
zFQWgde!)n(_#)N*d4zmV(FsaVL-GmqN_gp5AeULWDjmjn0K<Lg)_P9w#d$n$JMrJl
z57(!WV4YZcjDyGs)WLU{CtRoS-c%*Kc<>rrH2I4TOI_N6HZ>o=;zhKntl*oCmB2oQ
z>7V#aHVW^}cCu5Gm!M6Jbm}8Kc{Fp`mf}u&hRAQ;3sbujV4RhBi^nqS^<<x>Elf?D
z;>9T+rn6beyMA!Uc9b(_4c^I#@W|ksBsB2XwH&(7eAfkiDj(-hCGbrw_-6A6y9!3~
z2)^;FN{4Y4dT^L?E3sE|HU68OBSTfrNyi`fX8b>RHPeJD`3v@Fo($2ql_4rIGDI`k
zJr&a;M5Q^uj9=;RXFpNPHSA{?3D$`Y)+9eT9QsVAp0erM3Y!9Wt(j`3(-w~TAw9OO
zUg1sMZ&mVnyuRYCXo^Q<L2-Kh;14G9+I)2%-F36sDc3zni}K^6<^C_l{XQ(7oI(dQ
z?80!qAIz$A7+r&zMSacm@0!)Q4EH<!z&^YVmYRj;6Kzt165yIwc!inaBA3C@pc6f|
zGOsn_e5-&T80=9aBWL0VgAP42C>y?+K35Dnf6AcZy9`PNZaF_2Eeg)2^a!xbK!bM2
zgJFgmbZjiSD)g$%_;IFAHpno_pz2c$YJ!fH0H2foqCq{M8`KIv&++<bVlFyzW|OnJ
zz*+kXoKujThuAG1Zc@%^<i!t|)a4!DJA8NpX0K<QrM(Zax8a>xC;2z$<ctepmYci?
z?H}Ez6SLnH@J385uj}wFQ7h@(I7ELcvvGcOs-oY(D?ixR_mkHXFwTP1c!!yROYn1U
z@;*rY?&Dp!M?UQdI|XlpmB2Wu`Pp9_3myW~^k$ac3U-<lMgD^sdP{3EAlbn>%$5&3
z<MD={zxI?f^s-sDoxIjCQ}(~h&QNCCZ_nA~2hIu03;)ytOcn$tE6rKY$MVDT-^cq{
znzPu$>&D^~oXKF)ivQ8?yp$g3+3e}+Zc-sL{97ciZ3#T``+Tqy?DNK?RZYQvD|wCK
z@4i)@Id>hi?GfhUn{>|d8gVx>SS^AcM6mMfY2*L{>>9M)uFQDuD{f@}5WbSLyhdkF
zi;s<2;<<|*wsheHWgsICzL~--GB|<$s`ca?(T<*^XWvROIGc*$%3M2K8Nf5W!8LQ?
zY$m|hB$go?g?`nAx83+)@11th|HnmU(un7-XI1X`=X6`GhoimDZWZq1&fLQ$GS&kx
zqunfGHo@PL>tB4(<Q-c*wyXO&ySCEDxoHZV$W->I?6K<~xSAn>c0Kx+-YWQ-HDHxt
zXgPh_fJqwSrK^GNoZqe*neAEyhf)cj$_B3Q$=EwbSCG#+c6nV2Rja#rS>A@K0`pBn
zc$~RloTt6mNjDhY2Yo5^9CmzdX6HQL!=Y^SOrSG$hQmo#2Mk6ovsy<qt5$rk{tlHX
z>CpO%aAaVj(|nzD^n@O05(-})q6+kS`p4OHvnKsl^=ulBHr;qS++RBU@og*;JZ1*&
zhc<AkCBPreMuI=UBq6=gjl0s78P9z=ntNv>J~r-`M-9-5E*X`?=Ns1Jm1}6!pnr|>
zpNPK9+_G#KTGR;k$4#S~Y9RL*&z*=f>S!eQeJEciitdL2M)`oBRx@Xqhw}G_pkwvq
z&+~lnb|dGpQFhLY>q+DwtFaGvB6u3z)Rfh%`Yp*=_rj+Oo_M-}E?_WOApZx~gX68>
z`1JZemd-M)t*&drbuV>ycXyq-mzDq#EP@kALY=z1yHj_g?(W{F?(Xj1es{0;`*E&>
z5JCv&>^*DN%q;j-Cwxsm@Cta|N-uU87-jQr^f_Sh1-H3go}ig|&GX@EHvI_FQ;sa$
z{|2|y=q^1jaFF9Tw>!ALmXhts>n!QP&l<@rt{0?v)H?4|P&Yh*PufneYz{M}BH=^x
zTC{4SQ6Jn!?F=`nUa*lKvQgdQje0xIsH`)Lnm5y^RrAQ=A_s3Ym?arKp1SotdR7t-
zQXYCgnZPP(JZhcNqb(dax$l($-*s+}iWKnZdNFViSu-{2(kp7_Q8$Z6<I%yS=e1HT
z^62D7j~*QIs1&%z4nAp^1uYMKCa*^ETIPfII|{E9Sf=!xc-80lgr2y{KzdY>%+4g|
zCUvfOCI01op(##w9nOZ=3WB%UvjSiEY>(;>@#q?J6>d1m_OpU%z&A<Yo8~o{`JLLM
zey`mceb22Yd(jYsZ@SNMD?9k+6#c1K@J#{mO#|@F&lX@F{8NeOi*KiPqld@yech$X
z$6d_QajE=Tm+pXX`cGkw=4kLvSMvQLT`~u_G`c;R#r0fTQkvV0@I3F)-Q12-1M-T0
zZjIAOyi-lUH#2+3$r2i;-`;WhMs9Hg-l>6TiyPpbTDvz^PslA^GK`F3a*OvlVwooy
ztCls<7UyP83ErtzcbsZ*(y3z1X83Oo^QZbdRWcO*rVaC_DmyhIn^QMFk$riE87#YE
z^m`Gy;?eY^@F7<9X0B#!dRF<#jQR_Pc>+gsg1q9Ta5e+zONHa5Y)_U1Ud<BO$dmZN
zn)dX6GisOZWQJ}co{3S^Z7y=$1MrGdyQRzzPfb?F>1Sl2UWrsSxa~KK@s~~{2M2Aj
zw=GhAnnmhBrAQ@aXYIse@ROQx!=4D8Bl{-js0cL#-;6+eGa@@{?ibd<JLD7NWu3T|
zc?@``zIO}P>|lCSEyDG_JTv36hO5WpFzq=)AA1!!IFsOOz&B&TH>c=N9nTx4M&O$+
z@Hgd;gfer8{F@QX5C`A%p<kb;Ds^CHYC6`yDHk2ux5l9uGGcp&(yM9-hf~6#JY*NA
zd|+3~Bjn{T!mB&buEppa*4H85FOyyQUQ#2|AHTSad5gnr8jiQOMlzc|J+`VM>s!_3
z_@$`f(g#?ztPVVm7xR!FqG8=fKV%L*Wb$-N3yeem#PBItFX^w;+CsktZSg1<8q>V&
zyVN>GStp88$CQM}*uh?(;}v^9T=5ZVz{Ac!ePW-CD}WB?X@GX^q$fpBBgum1urPk`
z7jV-1n2))bTpQLnlOHps>amU%qbHx4TJamW1#Hs`TvH3pO|D~R?E{~zTVvK#@QVj*
za|zC7@IP$R+pJ2x%o?9ymbkx}`|10lJ01$Q2}GA;Fzd42tWi8x0^GCU2G}SK^<r(Z
zcIZ)gu{JknZM}^apafdTw5*|TOzhj^Sx2d@j-hMpSeX6(pALtfx}6%WCRibX8nJE*
zdOuyr@uJ87aSOR*Z-SKU8#Q7Kc^U96_ZO0HygOK3Z@|I4BqNc}bIEt~3heW?$H5u^
zFE;snu<r5w&tRJacgP~UjyCxkd1GjIesWCTL>@;kX5V--59c2@0=AiE4whFZw9NH_
zRq8idppC4-ZRj1q#az6Dx8yW@{Th5<Xk|{Z=C?ltt~tZ(P~P{7WaM(BWKHI|gQz>x
zo@AEB1T>LwHv_}*yflQrxfP)5c**>z-4~-p%F3RSlHOfMZ!(L(I5&FXhhfIXMEX{z
zsYM%df6&GNwWUsN3x0aypg%AdT>|?^=_{=L^Z?p}b$;EUrsmu+z%`zXu&Xt>$@6pD
z<le!Y^P^T3+iukcdZ!7TpJn6Giox?-r$^ia?xZw&xwDCp@(1Jmx1V|i4b7Ji=;q*Z
z8gs2@u8vj)olQo1O2PDw;&L*ph3o&!Q+$)h(Z(#KCq-Qo$#ri<*VFD1Gn<l_ae$sK
z2#<hA^qw=4oA)2r?3^%N25()4BN>)1OyeKoW1`<<1B;D-PZ<iYlEDeq@P&uvx~<a$
zK89Y?yEga)z&X9SUDuetP))A$eB_an_ptgW{zov(Q?QIb*Y-#}WrsO(f^VF699s0o
zp>i3R)k2N4-Vmza@uAvCuPQV3&+!?ds<1AUe1%ZOgSSit&<ulPysCxiQ)6aSwj_gs
z;|lNh5jE7qJ)x@1`}&8k(xaIfTY-#FvLQTtzB7uGd1VY{ZUDK#)8I?E_CA6)N~7mr
z20pR2hRbYX(XE#7EpVy}LdX!eQj35`g1J`DfqUHX7Nv`Tn{k6tz%Tpp0ZieVs#_X;
z4K>PyPV|4lF{SwaG+sYB*Y;>x6y3n0_0%zqxu2HTugLeN)+e_!6fU&@`QzMg)1GyJ
z>#{mOo09i6W(mg@aMNk#=#~ao4}}Zn+-wgoJh^<J`cg-Yq4vl&on93^*7@hq_QK(;
z%1b^~O)_%8@#*5pLhc)+SzM=G;c@11t@b>~b$fyH@&Wq#=RwSB1P{Fn(y6!j%(?IL
z3_cY6^ZF85Hq>%E>1~bWoX#>HZ5!8oc;_HB<Mog5ds2Vnnz?}YyAXXRc+Y!n;DwJH
zWs5P&^1)xDU-|1Iw^Q;PH58oVH^itrV~i?0)(HP%RG}<j5HE6Mz%$R_Jtl!;%7JIz
z{&lO?Z?}H`aBBiu-QpbixnChUS{I%(BO6-Yq8_EFfG@iqyiPkZ_P`md=oh^j113TL
zTX!3{2i~RKJ@5v+PLTrf>Q*sczuUxXk|kbiz%oP75I^fhFRW)g-t~B;Ku;W1m0nmj
zW@G;F$a%w~JiNxaeIBhz;`c7&wc&1h4(5G>Z=B$p^!(me-X2W`-$a6MHl*?}w}n0w
z{;5Q~;e$4i=QkUSGv2KO;G2GA-%PQ)bsq23>-zYpDxf>g=T`3IXpW!a|Hd;Fe*_KA
zc9)v2CZm|Vn|bJpkKvoT(#NF%_`-cGE|qMDhaBzAgTmzAq;zTcn>aPUiC5})99%zh
z!|_f9j;ANpldQksI0d(Y-vQqg_=|7qC0ZN2Q|q_H>L%RH!6C8gig&8BJyyNiF<-VK
zT#grye{t$Q-l=hTr+RF1>OI-Tr+Yhf%SvW(E3$7YI<?RXyz`z6%uD#fcQBJ}PK<6P
z#3&*(MqQZAkXS25P4dJrgO=G+kHI#3$tlJ=l|*l9jU69VhiFwI`z9MX6=}aQU*-{h
z&$INOwxT1ROLp;4yb<)qFZto8Xh7yV{LO|xV45drZP4D#*uw0-MPxvXhx10SIKdAe
zMkBnOW$_e%Zw7xP#~l9V%x>npEW!`Yy1I$plo4O}kgO4^@-<w`ZiTBEb2Wn&(u*1v
zP7e`(xFuXCn}q8)9?7a+;i~j942?ClANZyi_@-7um`cGNm#a!{ah@=xMSo*O``7U>
zdK>V~64tMb=x;W&9tBil9nDNm$ZO^>oOWpKD&|fNA@9cSQ2AyKO((m!!7t`8+(Bn^
z$gTpj>{^guS2Jcjd#c&BJtG_r-tfyu!81$9z8MJ4vDy@buJ|1IW-It6%~G_*%o$o`
zVuo~0=2vI1D*A4Sa_m8?if-ayIK2{P3WTr^Pr1OpybUeL9A>Tzz$+O_<|q8kt!van
zOW=LE&d0SS>#i95%?mhy-SDu}(Ef!|!@$E1hx_PwHbABgU>j=wvh;pF;*HpG*P@=A
z!4vZ>+8$t0L)N=F6)iI6A>$?`YaxB96?e=kc@obz+Ti!#n(|<i&hR#iCz;iBv{{*k
zm^BnU^CsS`MlohxYzIfv+)QS@S@|oXE3RbLpt@!)YH8MalUc!C%vu6AdUSyv78)E&
zDU0s-qBW+bEJ*D+@iuyZbj-u5NcNu(xytlocC4lU&t9K1cc4y(QJ>LMf3TSxjt68}
zWa4kG!+IFT%!Xln7Axq<-y<7`wXq<5{7xnCosyjq+5{}pn*37(I46ug4L!bwE12Q1
zgC5oaW+R{A0LPfXHZ%6Z-SG9<V|Y|g2W#e8c$_8tAHstb(TY8(Ec1|a2P<1PxSqe@
z6aEhc;Ma3fmtL&Q8cr{2C)~}GgJ=OpQ`^Gd?*rp_HsY<_OU@#=rv&}AdmL}*VddjU
zzn<PgZ+Z*C{O=|OD3QOr3Vh9gY4GL=czQk5=*KMT$p0+3IC>lYhk4;}PFSg>yOH_q
zK#N0m$$fMYkI<1@J<N-5hSp&W`c3o-<<7w0++u$?&n+5`dF&6buJhjEpBlqsr5lGg
zy#m}|F8X>IImbTnIkKl@niizWN8qP&qRHJyzvpbIX5S0dxG$kPhCVsE8ApY1&D;m)
zgZ||iz2!?_nZxw2Qo!5h{mNYON6bB<Fa2^Ob693k=ah;>0|-}(wkP*QG6l;qyXgWm
zabviS>%!fnBBS$kn2rr&o(h~uI<zc>(Db&0Z#{AloU(=<@+5RMgWzXu=zqa7*T6Ls
z>X7#Up4r_DE*5U;6@8~xwdpgVCtjGHS#s&n*!*xXGXtImEOQgyCX~FO>KrF8IF$9a
zLuI(;^QMBkDU2T+J}4!;P~R9d+59d)FxA?Xp?bMLRL<{kM>*ht3&Am!4Ab9=VOk|}
z6TvwR=v!sr``MBlx>MMps~xTK#Y2#Nj8!jBz!mPI2U;4v`+a6SegxaV$IPtCHTcTR
zd_MGTAGt=po0TQMMW2h}XDmmLsJcZ@!8wPxR`S*bzkp>rRsn-_w2--BR&Mebnw&%9
zhNge_XS43*V_phBml-URry)HpUjI-93z^jLH`TyB+;%LFZ?S|$las-jW+11WYd*Ud
zScv0hPVf-FdmR|z8NatL_^DbV^H;W7bTB)ZoAc4ebvf)2+GTpjPg=k?vfh;BT<<<V
zP(7$W4!$JMf$Ot!C2~hwbDi3PRHPd?Wh6Z3OtKxgR#Wc6`+16<GUsATIGh*ojO)2}
zD{<uGno7oFEpIS;;WYi+1Kj6a&M^+&2ds41AFW|IG&!j_pO2$g8yu)2Rngv`!Sm%8
zpcY?^azue&3LDk;y}uG(_$&1re|5-aRFt0)F2SfigN+(Jz$in4Q8~di_Ka{Ya5I_H
zdE^84(J=*mAn?t;Uv8}e(-^=uufMr<n%k6Mo_ye+zx>R4xSWI%__FB<HEZKhXK=>!
zZeS1a&BX=02KZ(M*yK9+CKdQ54frM_dgd_j%^9@JFUan5foVo}i`Rzucufh7*Z9t8
zeaP|~kTG8C;B6M|^(X>8aSwQ#kG$saS!i%3foBFX3zOfqKH8&#7P4@f^L?<*wY(k$
zGdpS<v%jZaa_i&{w<1@-)lA15HInz=%dMO-WZiUTRwlivfn*owEQ)75GyKk1`b|%m
zwQz~qQiog$*ami4;gS}h#R1=RL4R|<w@VA7T?#^5T&@Y3#mv(DP{5_Z%wpK^0>0*Y
zoDz=4DRy%lxQINz;b?L2PMsy&uYZd;wF2Ml`xUFS&tvry@6^ssvHCGRR%-^MxnVBD
zYP{i<o5d=OxeWEvGdKJLb4qSH^%d__b-dy2$@UwR;8c7tI^$+e^{n93;7s5hyy5lF
zlXtU~xg~hRr~UIzF`FS7_+}FM3nO#I(65iy=7-FcBAX#;B$=XJqGhv2>jt@63(7Os
z7oVo@XX>*DQF?TmT#Jo(C1&Gk=!%yL?M*kl6kjWl7f$__{b!^q+-L5|DZCMz$kC$y
zD>i~WRP>6)40yyF($@#y#DH(c;(M$KzA3*Ez2KM#jdJ6i>JXtrWh1nn9{tHr=x(U>
zilV*AjJEg?*{cb0;c^?pS=-=pN`)(t`fn8YrXT&OLrciO8Ou5uMb=-3|E)t!@M~87
z$r^PB{uq2S0(?`1TCfTD<}z#LwDQz})OxL6Iuvn|T5p9z9S5N)wz3X3W^K#|zl}Hi
zC)veY_ml58)2@2G?E36)*G;@rAHX*=pWBq`h)q5V@k^mC&hCxg7<_X9d~=%q)Ny>_
zGsCQ^-V$#(bElT3vMTW=zMXB%4VZ-PrV~0G^s05yeiS`NwiWzAm#OSU-N8A2@DA)l
z7vOB0EI`BI;kG6FSYh_pH~1I8H|8nymm5>ZrzOMcHrhaX7^lWD!!np`KlaB<&)6q-
zQNxX9-gg{6ZPu`~)OW!pE$Wd4U-(b>m?vi4VZB_9mbf~;sPSltZ_Vd1c$-D=HnqSu
zE#PAsIMCGum{r8bjGqN>R825Wd9#8_nRS<AEEuPIE&5jN!9U?<Wu9!-uI*qV)?r&g
za(1X8TlGd$uz>8H<NWSV%;_vhKdLqRcMtaYh5z#p9RT0Fve4sU4Rx;!REJA}YW*uv
zYbsI?!c%YR10O<dHyY0S$U}PlzwnB(SEmIF4n}*jpc49}+F+J;c(du#Y#oJ$W@E4f
zw{z|wbB@=6hgLKHhR3q)q!&d`ss??no8X<HbNKwJw<pwMrd1xev{b?9FR4ALIkO(e
zuLfQ)ux?)jV}8B^U$ZGt*Wl}iB%l%QPCXkBb{RrG8o2YrR(Kqa@_YC@s5i$@>sH|Z
zwLow<{H^8T*oODT;}##FckKgIlV0NKlk^7`TFAs<M*Ad-h9_B6<sf^_YjmPu)p=+>
z#b?!Y0a-%mF%(FTpgdj@u<s7|{F6thbNO6093|hCJ*M;kxc&Y3wF)uc=U*?celR&J
z;JBQ8&1d%V85wEd#A4u^alO!S&&FpzI#~N=aNhBGjkpY_MK8~VKCERR{+&1M*=?<Q
z3g^@Bp^a=CJ9^1TC9}fughb*wrluiV=OMb9zI($oe;&T-=rC>Pn*2il={ZNh$}rvC
zPmaUvFtvya(_{Pqdn&_=6vOBB2M^M9vi8>F378+MVdLRf=p7C8ho5PJ2OOR7?uzg%
zV4SRts84vzT_4V+bg0JU!3&uc?j||-?2|)1pW!7V*J>rW=Is$Y)yEuach*5>r9*T1
zo;N)#U-%w-g;4!%jJGNT?9`X+n~8i)ziIYK=DK_&<1bT~db5_a%pWFS-s326O<wv{
zz22d<X<(NL|D-e0s?9~LY8DcrQfR$CKO~Q-9@=0q=28A<+YZCIz}0lhYSC7>(qj+F
zlzK{*6gX$)2Q#`CvkGuM`cMZ91B(>r`W&5utZ?will#;dhw&tCHmmOzvuf|e!$=)c
z_K;aw>HmiFb5Y<N3)fhEZoSi?-~Ve?N?!N<H?tBynN^q?CW`C44ZqWXCaw@iBd+~h
zsp(<y^S=Dv0yQjJ5<?#hjPnkR^ZpNfkDYmSaEiIFkw2aW&ZcspvKr~ba-RPl!TAht
z84kv;$n}{oA3de&99*NRUF35PLNhl3J=j9@YG|e3b50HcOE-t7IrA9*IOpD0ZufEx
zKZI*62lx2nB>h~@*Wg9;rGn_`UZpPvp2?auKt5o+!hBAV^H_@-vR>6QY5Farn$efq
z;BQolJVs@F>#s*g{q-ZlUwixdYh@RIecI!%;otn#77WzYWmFY7n?2x~YiT{Qz{?bZ
z`*6e2to?&$8hld|eDf4+GmygwzVYKRC-|o)_$D3rW^-}yMs<&(TYIz;9n9jc_?y5t
zf8kkfZ}+I_S&wEu^{7{xc;yG*^n#no0=^0Ohi||$jos*N;Be;Q+j<ldujcsFBTA4{
z{1DCXL62^Mkv6PDU$d0H)Li%?uuYLc9u<!BC^>klj=u*@i%0pZGPj`!9MoI<;AHw8
zfwx)4e5rBE-HM_&m4@C_o*{0vhP&zA%&n1i!ENw1#fzcC$>mmmGH-@`a4GH?a~PO2
zHTZx_bC@yJcRAQ*zDqYJxs)4Uc->wuu3>Hs%+C~92W@dLeBo=!EN=BIPJOS&De*Ac
z;*D{7Fe^@Vm>FI!K2AncoSrp})1N|Ujqy%(WiG>kQ|N1wVwG-MthRt}l6jaJZic(5
z#~i96c&qSE&BhzP_ZsuU_rv4h4fh+#Jhom=?GJEjzMQI8j=3ge-&B3ZJk2vPy1gkz
zLExK5J<%1DCDoyQj51VZK0{XK(0wNJ^H{XBmbu}>=~H?5Iw)GB=ufo--z>p9Rpdhy
zdACvOd4jC*Rrnc3Mrk6wsXxB--J6m-MgKi88<^)4xaMx8Dxbs?N)1<ZZlp>MgPV3T
z_pK8iadd{o!8dMnjBnr3^FJM-?Bo`EMn>qf6O03YQxtsjh@7nVPs5e%2wqltQ>&)a
zqZ$;>thR8i_d|cvkeM;yn+xa%4n7Ri;saso0mq$eK$xC|hG|#ZFr7vBSC}=aJ@}?C
z_23V<=8?<bY(|HwE?oC(_?zM7;JY)hc0Px@IgY1x8EfAF{NRBOZLjN4VlIb@;telz
z-L5ERGfbR=?xu@fotW=ZxsqM0QrlJaEi<3NHwExceeJ_+>DD$qr$2QOeB%J$e1*G7
zVy@<3SvjxBF8+?^_!K#FbMW&dFz=`XYa!n7EMMrcz#&v!Pd{Wru;zy`6T3QDE-A@_
z<GK%lzscGatwqxybVc+zz&C&BgIFe$K@M*m^9zh{2>r%NFwUR=6=_9Aa4N9QC3->2
z>8lT8?sp{XUmG%pt5~#$wXkPe=6=5?qwls^!%mUY2NoF!Z<8<s?aXkqvL%?+nx0fL
zuuau4v;Kf<@`GKvHKXpVgJ%jn^RlQ}dGdjUa`UqV%<5CttjUednrH^=fQ_c0-${B*
z)^T=f$i{f5y0X5{W*ueEp7n~_GZ*?<G&u22^s{Ji=7G;{ed4`$Buf>Z;~@IM*{AS^
zeIUcDAZucqAoT<1eV$B>2amn=BJ1ZX@^Suvb-*J{*}EH8WPTs`WPMX`P7ppcIGot&
z<f)=RF8&YWtYik@a&iIJqDkJ2{uqqYa}S&k9FJidm=gYHBizm6%)w;s2kFuy>P^<>
z21(SML&^EzJv7P;F9KefI3CSY1bBrWVOi?hKj5-<-1hBBjw<?^<=~mEYvJ(0INNsx
zsu38-Q3@=|o^clL|5jJ_zQ_PwuLIA1&Z6`5?kwZ^T>4Q*4?$Bgi#2^Owe};vmnlG2
zG#JOwD}9~|*Td(uy&YcIEcBsl=#u86!#O~P05$N=ePFQAAeD~{lIFqB@Y&?&@3Mn$
z+LjE~L3Bhvz&F|1WAfe)(kl3$K0VMl&|f&)9N%w=|M~Nh&^|qYcNz(YHvBe!+YRyv
zZqYl0KWtTuy{R0ZVS6$Z(^&N-w^fVC>VC=G^}j`II#iw9%jV=lxtOQ5kNIrdcpNM+
zys2G{=i3#X9K3<fEdcG?KU`yafDYy>n&C^_t_Ew&K?^em-JC5{6Wf7d=>N0^$IPJj
zw4*qFa4=3~@J@Pe(-!7?x#&B?-SqzI&|mnO5^y*7;ck)+l5Mraq0by&!8lI(Rk!e8
z*Z;ySmTd4nCBZYiwguhss~+^N`JFfU-7QarD&jSB8d9Ty=IHXA?*-$lO6E{ir%lJ5
zR$c8yW@ijpjE%{nrgpKCQ{Ckk{>LTsO7jKl?^61%{NEOMIOmIk@ww*4-9x{2%dFez
z(~i&+-hQ3yk8AMVA9}tVS>bg8z$PEA!Pk(pkn1$p&qlMBE#z^I981mm#dZ2~jakn&
zkvRo^Ihq1)hMrZXFXU8#g$`~*`!*j8w3=%fzGpv26tB~CKiBkRuF*$ka^1m8V4jz+
z&5B_i=$xAwE$!%)(XYBP0-gy>-Gwz|3HbZ?Vf>drxlW7WX=sLTDTv-HdfwlC(Z+L)
zHoOjQ|H*m5xtLfRkDxa_s5p8~gUL9a#Vm)l^piO!KhmQrPw(^|=iG>gaMARu{+)k=
zz&wY*J_W%yzxU!JpoV!}l-D`PHAers`F*Z~CFq2x&BE!+UE=SJv6|qsjLL<^W>9sw
znJ@lYzs6rr?)jnd_tTA<{(2q-uG!(Qk)QqXXd87a%qTmUrZUGVc$<@8o4;U}X~{hJ
zvE3T+lNn858XvIDwXbe<<F*!$h4FnSe9w=ZV47kc4XB2`2L9xvk$gQDGr5<;0qqBO
z+@!YzwyF6BEzfs2mpt(*RGW;O#_>u*gR=pQQ#O#<3*7#j!BDUsKJjNB4Lj^%_A6Y{
zPTm_msjF}|+voH8^rZ@q@Tf}|?-N|pu07es%{=;23w%|EUKsDa@eO#IlWyHOzzj|L
zQm>Zc0SDKpH#p7hR*b(}K~3DMUK`Ii_$D>@CJcPDiOicX@0p7UzPS&+*|!&bvyooa
zGBR!E!qH4*p5{<+4*n@O_~s-0&7?;3vC!Uh&*B3A#i{WV<}qBxFSRdD71uIn3Vicx
zNSrR=4gcVeHb>0ZEEcDlzhjjPZ}^h^^q=sCf0=?7XApRX?3?|@SXIM2m4G+AMv7Sd
z#T#DyD*42FnMJ#rJ{8$F$-p-z3;ys%_{7Qf8$$L?sRztEInGR7=CoBI`zB|1Fis#G
zPOBKL!WUi;Z1d)Bv>b=g6R(Qa_95tQnAvih9Q1n3W;j}v`7n5=YP`c6afiOtetZ#V
zZ^rgx{!An~LLaysp~FE__?o`_uMcE{fp5MZg{!7-zj-D(sDsfNf^SlJ<E5wvr&9*+
zX%;XJ_@*@Y<{+MmpTi<FkzBvU;G2)d@Qf#C{@;yot;9z<e|5O*c&9!LK#zzw+(Guu
z;`-=rO5(rx&D=oh#Dn|rfKv-j=ohA{;2RV8W;^(%8u+IE4}28#r)t1=-vZxM0^ekI
zhRWZY^{h-NJyN*t2M%RE49A@4P)LG98+=)Jt2xM6cF<d}s}Zxq+cBGA<7_lIWEV#m
zn8{GWu8(B<^?ZX@>X=O<r<1MU%cf_gZQ4NZeG&NPIo|MM!>p=e1>ZEVs&irZ8|vRj
zr<oNn6K`)XaxCC)7MA4rh92Y;9K@Pn<sJi{5JC?H9%2uDk=mE(g{)+M?F+_f%l?yt
zeifW>L%4%?a0kWNFTHOtw>t^%6k3hXA@q9c1?b#Q`t^7u?$4)JKhVP7OHaKSymwh<
zhi8Kmpnl6wJ=cVFHfbMt1^?42YNOofhM$3Hj=IfwOvo$_MMKlUthhGlaGIJ`x<0vn
zRjH#&qe%wa^vG(KcP8?2yv%CH*VFl4X`UbBqRw1Szl!zyC^cnW_;!cOqWGy~?(C*t
z_85Jm7d%tl01ct<@?$i4%Uj?uUXdH$68@Xs%c)t+odCy`L(7twjoh#XWX$~AvquMM
zz&bQJ@D*LC0Yknshcg$SSuti=l&2?E6U;)dDziV2^#|)r1iOH1Hg2U~uryfj7NNhT
zZ{-D-lM0M;eH*+F+?hAna!X_O?6hE@cVs7noo2$HJ*KAo15Wq`#*D8YM172ReJ?$!
z0fG7prWx5B98*6~Gn<mR0X8duwq|Vrc|_E>;fc(g0OQ<(i&{$UyDuI7_o?WP;BKbV
zdrGpww_WFd1<w{WfqnuU*V1Tu3}`f3f^l~5poegt-o|H(veX9)@mZZ>-)#LBk6dkf
zAAClyhXtx@hd`wY!8Z~Us7#y}E_wxnij%9x`)tR#QhY_AqUOSl!Sxh}huSk1U-EEr
z2-$-Qb8dN4bDv~is#=ek(uLrCezCW_1<yR?weB+)>=rt9_OX3n90%Wfn1;F1`N5X_
z+;?!_sDC{!KJy*m$VXuH40kzyJ2J}xJl{Q(d`o7qPoK?o@|b+6>~?KF!F-c~4t*Nx
zP)mBlJND8?f_sTC63XnqP@STGoDpqIBcU5c3!Jt>s4g6GNboNfFwTkc<iyltHbrjg
zmIC;^`CiW`hrSeJ#w6KQMH}IN^miyVe#kVF9qPZ_!5-_7107BU@~*amfxI%|o8o8R
z@cR;4<FE9<-wa33gV%e?NwCukFwJjraUS7CfJgdv-JwEpRt-h3)Fp}>X?&62ja=)c
z(GmZ_Gte$X572n!97es82reiRpe^WG0??n;1V40qYt}1p%ZQWYDjdb12tM)0Z`gM&
z{=~&_I`njp?}m5T%k_AW>?!X1P<I^Q@uTC+a*ZWxYJyp9r<m1$I++Y}=~Z!k<$b`l
zOdauy>o_g1*>Dc}H=bwX`Tu&%nc#j(^Zd4B(8?v4sS)6J2E)a|0rlg(G`P=e{pOl3
zL!XQq=&&EYaV!0<kG%IZ=x|)<-QkCmGNbXU2X3dPI?lNla{_JHcXVRhpK4}a)oQ-S
zxsst2TK6z?Grh?%9>%q~kXmFDSx{%miMmb}@jdVi*rp7(t>JF&^H_VZ&*Ky5#yF2U
z2EzH=!Hd|1&k&3?_bS;vcCyv$;B}`T*NgR}fZL>@kI)Ri_E+~(Mt%J1t5!w)(BS&%
z?_EDNEak6#A^u8v%wNG-jjGwesAZjvN=)g|GLBmuVH|(qZjR8KssX+^3U~7uOfwm5
zlZ;~=eV?tLnP>109WKX0JW634foNTN^Sv;RLLHcI84Rv*z}<jD{OKR9i6M6w+>?%e
zQj6jAm`0OPGZsv;8ZHKIW-Az`5ZdASTRi%}Z7#5ljr%S6UM_f`$zYowlj(iI+jQl1
z((>B=zHl~e@pUz!H&zL4F@3fHy!R!S-8u%oIRU;&g1?yrzIiu@ISdJIGP2xSgD3pF
zz(2L<QGsvX7I7;C&r}UO;o<0t|2yW=S$yGX(H1u*_vYJDm!8jYsTlZX?+_RBeO$7I
zlWXkl(uP_ttt-TQhSV<A{}iW}_v5r3Z@7C;oIbBcLkzw-f_G|kWSo+i&CmjT6UpqU
zSMWFYnLBlFPpk^9;XeG$?E$fR>SUItH(ADbr*amI)sSS|zIJNgWvBk^c52QFX0(lF
zz9!!AA03_Q$XtdJg`Dd5J4WyBGB^A%e(=>X3dI|~wQGzj2QZ74TyXEyF-rH5`7C#$
zHGdDY7?wq=)Zl0pjE&X~Q?zoojMmaB%zpVBrJoPUAcecJEQ!+PiBa0u8;lc1UNOC?
zwD=}|fNku|)a-T-Pv>do5pTnXfxgf@8BO8<`tea<8~B^rb>VhOlY>D<XC?H-n@&Y2
zWH#A1%rc$_zG;GYs$S6uefb-%r}!*}fNxf>3|H)waC#CPc*E1+ZGBV^?xuLSK9YUo
zybWiw2Mx}wFxA9^;cG_+*qr>E3e;Ql%`Z|9_66Ux2j6(npQ?ubW}Y<^btAJH@P)sH
zzuvRkq1Y7;1+oU#?1s0~hh9}>)<Z9HVDV1%KWo>YO?DNVZr8aUcAfBH{!}sWPD*r}
z=!+|$FU~RErmoa<_L4T;O=DAW@J+=dXm7?^WoNFYH{K~r0jn;)BzKnezTC7BdG!d<
z5AaQO@^sgKq9?M7ymHpX*>-$A9ob{cFu&y|niH=3C(FskqNX`!q_+OdjF^k;J?xqL
zsHK_}0t*~rb|rd__3#0?1Ifaujke__dft8LZKg3#su$b|Yt}K=+0oRI*7W!%-kEjs
z7Hb;oTL5)lC2GT|V4D`*%z6-Mmes;~*y4X!#n95MeDF498n7PL#5YA>s#jsw$LwU_
zq(M)d44(b3Nl(+#&ng5);&~=3To7w=244RqYv)lo@KNmB2gad^WX<1!1}B=e^JG=l
z#{hKD@EOkaXiKi*gUwA}DjFT`Sp4rh$+ozQS0XLERTc1_AH8|j&02}%VeA2GoTDG}
z2|rcVV2vvR2L`9`6z=P>xbKUmsSi1cW8q>JvhUK9D!2fx&RjS$Firt5&QLJUht+6N
z1_Ub$*rse2dN_RU4X7#G?FQp)BV%}35VI7>=Zk=QugIF53M_z@rXKHca2qr>@b-gB
zl0Q@yU2!dXRN0w}&g1b;@C4W+4qeW0j>GH={hz?K^#-4SMcetp-#o)7bIPIyE6|bh
ze`-~U&!{b$PxcwFmKH^Fc=MSJxn)sm@Zf5n0QoOQUv)2l&y>0GCN$3AE5CN=e1hqh
zS;>ALgf@cn<m!KXc3_)5HR<sc<o%=%RNXIlv7V8qc!!J~UdPEke1Bem_Her<5Z!YH
z{_X;N*1yTTe*vZe*Z6^Hn$RogOs_D@H8kJgo5a*;NJ~<0*N0Q%f6@{BXZ^yAY5IbV
zgUE6Lr?<v~znKigy(z7FHwFH3EY}HhwI(yy#0HPk1AcUJ53mMvZC`G4=n9<5;~b$X
z2gW&F3a+&(oC+G?FVrFyypH~>xi+aqhH%|xdV;2gK2*08c%ydM_2hzGc^}(#mdvWc
zXy;DXBlp)Ej{ttiv}3?H)Gvp3Ig|koXBiyMp8Ie|{Ju}=(eV@|>$qtsnPlV^55c>&
z;D0k+&OHNfy#;I0Cu@s_E{a@-%+JYj_(-lb^H%bj(PhKkJVZwv@PRzZmgGQg304#S
zuhrq>9`e8Z46bRNnz{mPlkpzc3z(t<_+?{~S$F49Q%p1K>qM@r(Oiqe>FdrmE1EiE
zGMJ_@*X2?8ni^b}<9dOMdYJX8w^>X3(HkB>{V~L>IeU2SS!xY%P4YGT4*1vcaHpyJ
zqkH3LGH}H71^a+&I(J30;{fvn;SDg;N4A?a8k}?C4EnlfX1z^m!P5+`hHHJ?(W1ZM
zyl>W!*-tF`?-O3j)Lf_E(9hLCj{{G<bSJZV-=K>}53|6JHE9XHN6wd%MbO9vQ5!_l
z_jKU#e$7l4yboC(!7o1y&{w`b2j0KBk8^h`=j9ai`RM9T<{)RV1heT_zfw`7O#vS!
z<M(d5f)+Z=q5<HWF6F`b@L)f~z&FF-XqxybX1uQsZS&Qj$G)m+@gpbGPX#mi>x<D}
z(~tOTvX@cD8b*Ek<*%D?G1<Wd+2B4ZaFpkM8t_dFeW+(}H#xyK>%c8(IUY09DIK`R
z1*TboKdA<|rrk?)xg5LRlM&2wa)NIjg#6D}{vyJo3E-Rmz&CYcJu=g~%9Q|*(;NTO
z05A-A<?s?TG^^0qtoNukIA`SsIGVK{J%Y2z%J=rulUhClKlpe!9Db%GeW?s~kLLN{
z6L0O&h{pKUE8yXxSJoYT_8QGi82zcUyWASE)~$k0xBlRjx)|gpV;N0OJ#<85-4rXx
z^Rl}29M9CN=XkzvF&7-&O-Azk(&7t$ltjMGDwmq$ohpT{IO6~~9K2J}Q7+xZ7v7<v
zONYwP&&up#rY!nlw8iCb<B>WQr~h`5aRa{jJu^-oyi+H`(H0w++fX-7t+Ru5KE<lp
zRWgfr$1-1>?3<~vn$SO1lOtkv9c^*B>apamGJ^_l_!Z`c=e^)mrtMCpn~z@#Z}{zS
zrv~5+uT#sZo7tV3{fj=-{TTH)fG2zzGimXLzlcVAV~Ei+yi+q$F}v+P+L|+D-E5E6
z)J4%s4!((x<}oAs;%4OiRbpoCuP7b3AEn$!qf}&36j~WH#R+7InxmLMN$z@)C~9+N
zP{G~IyhHDu-qgIUkqS&iYdjjf(-r-Vm3~!6JXE#uUlhSxL65%7y$E$gTO2z%LW96J
z)5$Gv+6I3G_~uoL2(VJP?xHR3wv3F`31ocs#v>jMcSCN*NbpT;zHp8E%6zHoVdSTU
zDd$W)RK3Ybr<d<(8m5Kd8w>cRH2B5?zR9_tzWh8gFZ#jfn9&*MM~_GinDvT7)pnBS
zx5%Mn<ERTg^rbpFltBGw%!nT637#pu;n7KUU7tigS5MYt^f$|k+tvIJxCXxc0{Y_q
zc*7GSZSpQ<)0T8LMIE$i&}ysFqEoocx>%taYhe!f>}Tk04lz@63K_dyndR6vM9c6_
z<+()72dDjIV6eWkC*Q7!cPCS@2C{EYq~B0`DVbIY=v)2CPeHF5d7k_SYW30e;cosi
zb7m9sD@Wobrq1$bZB1g0n0dxRe-BQ4tVJE*O}?{U?W|1glNtX6nwZ;g)Fm&%TkmJ>
z1KT{O);r1i854po7(8>VI{AL(%-U2E4jnwRq_SD7z&5?Au)da}Hq1*c3Z7~E%cKJz
zP0IP!q}HF{j5D!jR-(u6M=i<vYvlE-^7~HF6WhTW)qFVX{yO*^)~tv>Xa&mQ4dOi?
z>ldJ^^w;)Xz}wE>lm)yH?FiIuaNH5}D_wtp>!=Gu+k<J~AQq!tT(AXgA~j#38`OR8
z;c}7(t0SLVOkVUO6@oPYop9-B<`jc*@`4u!7Qh2l6}_-G^KZg=-VD4{;GToM(U{<q
zn#T8@rKg|u0`JCu^m<mJ7nnmn{8YGeKL0oi{CdG4rTGJI!uq_V6SXRMrXYAGV-~Ou
z_3fGS0ZQ>ZKn3{UB*Ed|?1O#)KBWX$&B*gwp97o3)3c~Wt_XOg@B{q8`{^&tfK#JJ
zEtC@t20Du%^q~O_;NlxmvpU&p*5HwO%R1i{&F6gbg06vea?uNeGrGXhwk@~do8I7?
z4E^Eu;991#ci84JqpUif-U5NTi%zT7XY|bAn?}^*TNkm<@LHw8HygR_2j0;tvR>iQ
zYG2`RU_ZM5gFaXuFcKJUsD*ymT=<-$XsT24KPU^<gD3jFD2REg^z7)TPKHOj0&kYQ
zKRo13G(KbyrWwK<4ET<3cdYo#$cQ>;)A`zVO}a&{RJ=o@XHj2l!m|kWc$pe+6uSTJ
zV2<BqsZZcmu6?5?)DwP({7v&3hj#mcU*J?{dFUDUwW|oTQ`YVK-yGcKKgkwH@B5<;
zeyI+;Za8z~hC4K9hC}gN&;jp<gMq`D4u{hWkJY6V^tq^M3NEwD8xGd|j=31*Qgugj
z<W0}-2N=3cgJ5-zAiLopx>@vCh2AnB8x3)xThuDEnG;_CKP24W+hlNsx#1sASagqG
zO8UZ_@6-|hG&eafn>F_sHNj@;iB<3?iOe;DTL}W!v>Zlu6nLdS*u@Jx@`&3tgTNnP
z4?nQXJ@CmRut*3s!#J?XhG4V4Q-gGk;X35neYnW18Qh+kNiT}mde)PEmJ{z|C_Q4H
z`^}sFu>qdQ$E<lB=!Z4uy5t%^(i#lK<3+em?O-DVYeJLc;Anb4UmC(A`tWyjBU=aD
z{caL|E--k@r|6yY!Ape(D8m|N<bspa<RE*zWgxTB(3-)wn!Mnc%Tpu3^-Rynd6<CC
ztW<yoe!#Z{C$!}~YXG=D1GuRWx8we^C?_0d%A2fH^sSTgynJ6d_tTP*^$tJ$W%x}v
z=<2n=HmoIOO(v!A_Sdw9zRXwfmHPYY;0j-5OZ3%`v%YFK+fSpO`;lGjudhS=H3JUk
z)nkA4Z|kcp=xjQ3^aSUW1KUjCHkhyH(1&^gwkZp5^YjzGBk)Tg$1`wES`II;&Ds~t
zHh6|V={dQ09O=L}@97ym2j6@FgJg(Adjr0?2)-!;zG(@*`Oyol5BR1qSfvB_W(pW4
z5PXw$4Ko(NHRt$xEO_S10`mLF)hRiRTt9MoN{&W@(-;1R*Syz&OrFwsyb8cMW%H<i
z9*<gip<(7d8^Ja^;BJgV!7~Q8ax{W_D$n0XA1YfOx7wy5=jJE#zu&p^6D{#4aLoZ`
zFmyVP$NPj!_m~lW30-j#zVOe9<lsyt^9F2_D#E4lc&0Yv2|q|*s%BMYGUUfsg)e;N
z&p72oTO52NPSz7~nzB7kx$%bQof@ai{o_>18Yf3<Fi-6`rO6tn`5*93T?W(aBD;8b
ztkzAA)v3O*dgzGNq}J$hD#dDdE@n{ucB<M_bT(xB&A=PJ8gKaR0ZzRNbE;W8`cgHV
zdY#p&$v>GHem#bIB?d1$GZ+Tq1&<)huOXV8@-f<wEJo*U<C!`httUzJqwr4s8bG#R
zWV8nQMk}mbv?}MMSM>z%_c@N8WM3>~{>&(JIKA+JlJS)if9S-5<QS)6?%jL%n%iW8
zoxn4JcWMpV8`ntYG~k`eX9WA8y~$RTnKo(2)ObY>2A+!g<aimUMDTxSUO4zBvQ>mm
zlYO&3WrSKiCa)Z?g)jJK?znI=mBY2$PF``xaOJ~$kv<n(&U@zg;xGNY3;)y%vWoFe
zWyXi`w+VS+^ry<83!I7n<Msu7pc~<BCX?&cIaGV#Zpz~mSpmLjcG;nI+o=H;IuuQf
zSGS8p)2R7YS954z7V62n=!lQm^<p*p8)i>=!{4;?Lw5tdx$w)Tsdt&naL}d}%+>UZ
zuqm(>bAZ4%5eKY_VJ<_i{?vB<R%I&-rwxDI<v;4bHT0u;l7rEPyj*f`HveQ_JqPyy
zuQ0h6xwK93Dx(LT_YvLzJ~xpy)&XDf)rUH{8axF(gG^`GZ-UUS)<9znKkSWPxW!;{
zOvCXMcn2sjc>p;Q7Wx>}kAqnkz!kBr$nyGP*8Qi{eXO-vz&5>)ux7HCUYLb$ku~l@
zEd6_M$wc^<Po-Ha3-EYRvvw8-pMZ7dfpMZLz~dC7{)4*-fwRf=+N5sxP3nKkq?vb3
zy8P0lLuuj18^W!Fm$q+b_Bgfc8TRd4A*_x4@y60icZ0u9e8xKt7CVBzHmoP_pZ)gn
zDf;c~sY!*%zJUWtwGeGC`{~Mef%3`+wrNCf-$vhso=V2WL2~bcZ#YB0<yMf+{l>eI
z4xP3a*rga;O^skZ_yhm-1HBHo<vzU3q@uy{ho{-b9_#_v%;hsb(Sw|LKm1pf*`v~c
zci_#-A7{?Z5%%(pff`3$+S`u*6>Y&8v^P~gf^GO-d2qvof@Cej)6eq{z+26F44;1V
zB>f2L;H;x8%yYr7l0QKE(b;s~2lnAO0iK&*5e!N_{OJZ<+E&(mo|~HeB=#G78+CkR
zKKIqB@b2eAyA9^7Fp~cx`#?w|KBMUY>IVMUmK96|4s2dGP)*ri%E94`1cMX+SA~zJ
zS2ly}fI;*Ws^K}|edY(>c;3M+z#%<Y%6mble18rzA!Y|C5sXuMVSvm$zBY;Xc#M4M
zTja{|zp%1L<pAF-=uD4|*Ifa3Rr)pB_PRl2q<{fdqbq&O=L5bRp97DU0bd&U#_^cX
zpPr2su5xaKRpnlzNgHZYMl@?@OMy+`Tz0q|+B_8x_d0kMG;odJa-wEB<Qe5qofv8m
z13jroht4?2uc4=Puc%$SYvEyRZ`aOXyXy3|tHwIJKJB*a!9}|=zNdFg-)lfExS7@t
z#qo0+`;dn-$)R!lj^xxcZ_wNQf%mxrX6iB$KUD*V9wyjS%-5<D^SCbQtF7m}`c#L0
z-9+X!(Gy%YhPq%r2mh-HoV&R$G3$ie5A3@KIj7T)M$@(y9Kdz=us+us7~$-D`c03i
z6RvZu9fzmkT3EFLz3+557C4)i{GSuxS@yxLBz9xoNgUWDit7$+^1VAb;U4%L?$@%w
z#k4bPVSTVkMfy}_(bBY_Z`BDM%}_ARWO`I!r9!;kTVCf)IM~O`Yjx!LP0bpKUM`9w
zxICKw+WdTVa1=+=8t9C<PRD?IBDhxLd5^tsbKi^n8{U6a`q^7JCksV^E8;AA0S4c>
z6K?|N_z{0}(Da%6vEB?@9iSSVQ;W-S{taXX6qsa2GR|!q7`+o2J^cKCd`_$Q4ELa2
ze!H5vpX<ReoA~^Ajr)h_iCttp0OL#p-<$>i?EGR;0>8)q4L?^39bvSe{1+PZaF#(e
zHyKp<l|kpL`f6aHFBy-%T4nLo{77F__wm)(PQEg>g3Af<)pH)-;q9wCZG5%%ia`y)
z1PwXT{KGTglW!m0x&TkphU4b{-@gv7Sp#o#2kq{9aLr2inh9W=6=0ioPuyC7S83V<
z@B@z(hW9Cw%A=*3$k73RI9h^p$fj9jfS*C1W1@F;5$y8{d^3Is8G-OJ?`QM9u^v4g
z1J6VM=_NSkyA|CI&#lmtepfhssEi)fy5*Ml1-BZb9lnTe_<Q32W@ctaI~>~;+!cyu
zs4d?|KRlr#**2xfCN2zDlh>`y+3>b!;Csp8Y`%eE$hnDn>XIFNlM8&ah&;dL$6dMr
zzDWkY=?-_ZYc>bmO|@>!mI`xefyt$2ZC&!OkI$++o^bF@I`GY!AM~Km7WcXqrzLo&
z&Thg3jyJr@6na&I(c;+R$T2{BToXT4W@a_KkJZFWWD~>R+*}%~*on*v?}?8pBvx~p
z$EtS)W-{cARVDDvyT?vlC)=+h-tgSBoI2l!S(*-~4xuexSr%O}+TxL~W0d(~jOOB<
z8Za|Pu{~ncIw(ee>cyx?*%;OP6RkS8qV@A=wCb*p*4-)0wq@?rX+It(w|FMGH${I&
z=`OnBTyQsj_@=7OC);lXUMl8Joj_MylRU4z`Jyy1b(9u^Zz|nl76aOwUdtm@cN$+0
zic}l4H~oF-QK7w=Rya~qQh|G3G8-K4)Ee^I|MjOb^p8*-TZ9bFBUBas=I`%tB?sSR
z|Cg1qh`9`7nRQI|O%WUVoA%(Fa&R~~!l|Lc6nlXgFq^<N6T_6gTbM@p;4cN=+%FWS
z(;32am%jPHV{ka|&91?r>>p&agKv6(Z&H6`-9vwvxRosR1#sCT96IDC#~A+RJosiu
ze%4#OQz@9sd|;(rcbLsEwws;&S9(_Tr%d1JM_sq6>Q0+%6VMcdvt~CS=P!#*Ugxd)
zP4>;f1p4wF&>ohs!s&<b`$Cup9ikbm_Z?f4U0f_gpS}ev&oTCaIb^eR308%=!K#QJ
zG=TkL?J4||^rK6m)f?IpKSD9QFy8|;fIYWEU-}@WnXUH~?qeO9H+=$hHzYtKTf_7I
zVz$f?xLnqfJ_*dDF@q;qukOF0-n+{jDb}$v+rS<hSX-B}myR*xZK95(-n-uxy>Nas
z04eZwz{~u6Zql9KCY8$z$HV%#iM7>Lh_%$qEPXL))IF0D|1;_NF>uZalV)5oso5)&
z9u=lerFM*&Oa1tq+Lbla&HK3818>GW{H%K|I?8^%h;=Q+N3zX|ko(dZZh^hEC>+Dq
zy!7%LlUqnlxO)ry%`Ns+IGid4(91T)A4GruU)G_gGM@{WXD7VQLOAg=@nqYNM<<iS
zJd3^5k=xM)qg(8Jo&2CjWQL)wIf`#L0E`n6MaEEfbUNM0yP+r82Hvb>s$f+^bJGAW
z<V*vyKG@q+Wv1^37F!z!4*0_Rx)-R_htVMLy<L|B<STSf;I#Rd;n3Ota%Lct?JWI1
zxUQtS=qv)vsxuN_@J77!hyORHbS!&W9rl*Ke6IERe0P5{spbolV&9u|GY?rqJhlu@
z>;)L7J9}B@a5x<HfiX4Vwsyg-JSPVS&Zt*0_?nvNui=X(HNiIvk5k0RtcyTAcBAmT
z@ONB*=V|>I{J5XL0~|Al<2ZcHy-D<?MuE%F1NELo?}5kif_ZL&ah4toP%(Oi%{aeI
z)aTW}(gUN&B$|MZ8eYirh>W*-Xp=cVXRQfR;u~^o!IAgT)z$?sF1i}5H)zWSGuz@Z
ze8{V5%sU6)-0W&ogDp1vw>DJ_Wlrr|FbcKB9xzTWuDLT8Y<juZhMv|Y2rxCjLEh9u
zIGPx)Lv%Z>a-i+6f}gRKT|?kf^Dr;v%vADy;BdU)S|7Y3ce4<>n;PI5a;)YC)643G
zAABs?R?8i_&asVLhoZ00%F$0=oyDrXXoEkH-*EmJx@)e%W2x~!q4BB()*kLeV>^Pq
zn4{ECG-TkItxfS#*JYksb-2;0c-zYJzo`J<Q3ZaZGCruXXhlAOZJwCrdlP=;7#M$-
zS@l-YZ<>dOcr0_;z%~DTi7VpKtwn)lIY+wKxi-ya{T1Ft&Yj|1Z~3S#eu7E<R3kT>
zYbr-R@KSd0OM1?&Qf57A$IqGIbOxI>1Z?vsgzMB7Ji|G&o!2n&oFddCl`?W2W<@W@
z>pcTIPvW&oa2-!83Aa-ZKP1;`6d1|tgjqX2l3kUXUTJasTX2ZY!8l*x7MJ(6s4btt
z=RfqI%i+bG13o%qQqxl=9ih$%fVVxJiT)bc?G0<q6F!eSHSnhMv*)?)#|}pq3`QKW
z5N>%LpHUJ$Sny5kK{5zg3%tNLPwsJ!KgXjEe^Z#-yL>(kynL0Tok5K|8&tcyL8liO
z^yiX6VQGA2ujZ@UjeXUzqp!Br@YTj*e4W}?V}2Pl>9s+#@AABB2GzK1P@2;Q>IeEY
za5QPSy^pqL7?`F6*d_{0^9M|mljAr2sAX_AKfyJrz%c$CRpD!bp1ScHFmsz6y+hZ@
z0shaeMW@`_c-*b{V{j!Lz9+~AKFQo}9$&?g4Sr=HM;q=>;_-QBz(!Y?g~@9@L+=xs
z3vQ;UM@93Y+d;ov6P)w{%+!n5Kg0~FPSfFw#-bNSGkiPYf12Ue5&UeRTSGcAV*zbV
z>6&m&;G4SOo3h}W^YAv+!8ZreyU8<k%lh4=5%4#sz&Foty5vGnoCSOn48FMvzOijW
zU$YV&@ig=|Xo{zHapC*LCvJ4%0dUE?u1nv`y0kMN`G4RWH~1!=xl>cG#OX2KsjhgZ
z8siQ3njELg;F|*>c&nPlX>s*9<;%$YaPUnB@Qn%m%}wTpzZny&9o^|!fo~>&Z%&nu
zRfZhQpCa3D>?5akl6}))jZ^u@I5nj^UU7O;gGxD>rN_+hZ!vm!FNPUFF}k}RFZlQv
zeeS}nDSYAOYsBb5ZXRc5c-YNo9X|rTSre_X6QbcD$-F^VoSV$zWhtY{HH=ckE%N;K
zMah3%lxoZ%%d3BstX-ni2Jh5w^4gmfK#!A(%&)g_HP_MD>|ll_+M6fiz%_lCp^5gU
zPG_)9%}9CjMr!Tv2yMNFet2JmCNpzt+(71Q#xS1&?xsOaJmZDIJY@SVxgV~y_)MMP
zo5*4GrpN@_7Zk2kWZx7Z`zA0Ov&7$o>E=1+!QeM7KQ>G|@P^Os4CZM}hC8zv8vjKv
zcn{s+QD#noZ~o~9Uj&osfmfn1+)W+&?^#c?4#HV`!`}oBaws=kb|E<Ja#g`QthHCK
z+BFK^JM9YA&armAcEjZ~BX>QoT`O6i6X4n3ufsDn*rs|$cpK)1?|zS0bT3&q<X}w0
zTT!5uRmqE3weSa;o3qTFUV-MaONbI%g(z8(5IMi#!94`0Fbz$JGg!ejnXO6YO3#=0
zb&m!qVG{mi`q=;GFikJY{jY)aZP~l}Fn1~+nm%-<4_DCV=#JOgOugM0ejNU0*D>bN
z%%-0~-PNHzxik4#qu_2%o@bWAKDe6I%$Aw}hGDIp;$m&2=BvzFS+WB9VX(`MXC^gV
zL*J>RN$%DrB`ax?qn1g*Elk=FVbXRpTuptG8Z2cW-fq&ujV2x2U{d1)CY8Brl8JS0
zO(!#2TDYKx<o~hfk71pR_fV@*2cD#U+px_-Ryum)$9O-|;}35RRv1LC@s<E}p&tBG
zK2XCF$V^>9U*s&AKX2fxk_E{IzL*Wp8<dm!x(u8ayvj#DAG<FeGjPwWD7XY}C(y^5
zJ_0YsJbFq8m`i$uUISdu(APYVJ#1!k{Arzo6%OB!wl(|?{;Rk2rglX$CpkC2Bb1u-
z6YDeeXz~}Vx3vS*a$tb!qD4u?=Y9a++7Gb6A?nOpZ&{<^@z1mN7r%ohA_?qRf_;oN
zzYBYf$3gAh2VcxEvqHP#HEYZs!#?BpPqzfF`F`D`zZXs9VVkt`l1X_kn{*G{_yc@-
zll^CMD?Y2`U@ZR6L*anl(+69U22WN|-V5AH2z@ehBeF@*;yi@+y4)Tt2tMop{<-pp
zbL3`#GJ{`M^`j3}6fJs|09E{kXZbZ5{``HZQ=oS!#r&%f`d|EwPM#aLn_T<r{I1_<
zq<Ae)Z?w$gnJ)p}EPV%j2fn#Fg&raJ#=~cO`xW_Z8R!GnCp!_$Sb7#7bLN<)3}%ig
z_+}isw6SKJ_RO-W)Eaz-=#4K;w&~aoo4&6{n`0qoAqU^%@U3Li-xfCfb~ar*Y*XXc
zHZ4c<RzI_y84z}zE^XI$dR*(7x0MHcV_yrmLk?#;ddh?0Z{Ad<2h|*%u?1eXn?ncS
zgC<9E-NU=yEKUy@@8ZnA%xdSnXxN5x;uSqU&b7)j(QS7lPumS|1y-Fo!~(a%e%u}Y
zf?6S@137$x8?6P71>2;kh;FJF7@#nGT~7S`nRxC`e1*@=x_J}+jqB~rZqEM=^m7x@
z!;I&g7)W0izNAeU=K(dva3lOC*UHtF^m1EKH*n7ED9CkL03TE~^f#%vZa<rpK)uoE
ziAfW{zH{E0bmg&03qF{%v@*CkH9uF2-vv+ewE>(=8F*K)(2@-BKj7qb|FH8{lip>e
z*OiX*E+suPddtm|nek#!e}KdLu@0<R3+{eEZSoymat`ne7$-9rC0kQCl|t~Ze7)!Z
z?`aV4mDlv!4K|;{xjGVjv!3+;zH!eu))%h%Oc|Nu$>;MSgp3M4!<BrNZNN8g7r<Gr
z0T;3cwAutmG=MC^BjEcQCRMm%lzFpJK}r6q{nMZ(b-*TG2Du6vRJ@r%OFae!%{1ue
z5ra0rG02+5SBK$o(k1iNwH&_c4dzMz*Py{)49fl7KsK*I=Pnx1P8ei4VbCrx&1U*f
zf#8@Q+_nSTBm>(F=j-Ggx#4WSgKeh00*ioaj=<Se2a9y#81$Ha6xgQg9k=?N1v`LQ
zCWC2~9Y)W45I^>QH`*$<?(RnKy9*4(F?A<D$MJz9E02GGiz!7vY2Fby8L-e?@XpS2
zyw(Nq5wE!#40MXuIkSnG3JcvTIt@+mX!JpYz+XMdwsC=B!a2Y<P9J!iCd|vMfo2$d
z(++%78+<bbd}Ghb-;$2kNygkw@J&6uQUTA%uYte$jP7RCS(jRXZ?=GM67WsUUO=yE
zqD$AoH?6=oMMA+be#~e9-;AsS)&bwR!8cjKH^0F*et4&H;hoBS2rQGttnftiH_V=T
z1io1n1ny}Pr&(3wSl8n;kL;U2|1m?8?BZA8n`F$@JlBO>zo1xc0^j(+-;B)~tJ6Q6
z;0XAcBq#eV9`SDYq>SVj<D1G_5^p%Ue#O2rX9{g`f4o!4HpFP{I5-@4jK;TSRs+89
zF<#{Ty@{6d8d~B*<ln3W*NkEIRQG7LYZ0yQxxhWlWa#jk84Ne@e(z?kCcf|;Q^@%1
z%j{sZ#kRIl_=D+7r6=3#BY8Ggm=#WzyEnNvyXV8<jE&Ut1pHFWpIMCdrcISd?a3CY
zJ0F;Jb0tDO$-TMQH$usRBDB0^guYb-=P-M(c``DMUy^Tng?Ur^!{u5^X7SK)WpRh=
zUI3mcvTs(F3fC>N{eHd*Q^})fi<dEPY81Kdk<5tc#A6M})PTS70^el0OD~`7;v6eN
z6^Q>K0uA8-bc$QwI8+bqO?!Ci18W_6g7(k?zS&`SXn$)maH>$_F;}z1WxM*}4L`h;
z`vb@^cGz{L2{mM4yM}^qex0_d`U;!6lYP_A&!)oVsSAHtRfp`-ad@W+p+6j10L~jd
zV%6gz+Oa4^sYWuxx<!cEmJ6Y0#Y})3%$P!3d|-UAYKFlhR14N>dMT6Xw{+TpCj~B|
zJi6BYO_<FBUs3Nfv!&?|4DN|OB^~?MX*@Ob6)NDxxkG=XNELch$;i1qOji3aaE+1N
zu!iW1lYt|yfLo5@xq!QQzL+|3DExOM+_pFRLiXSW)Om}3niK~IQ*j6DDShq@4NOXx
z+@x36jm$<iYGglr2u`CON024qYm|R^qrSx%`P+=l88h*}Fex<{ClmGDe<i3h+0Tog
zMhBS&Pk1}l%NQ^~Jet{QyeHPUWxLSeT%qRuOV7UnS;q0~!RrGw;~p4|y?Ya!LA9A=
zZqTE;{eX;_?ab1?jJEeJ7$gH+4xCnfbjSZ@6@G8UOv?`JwI(t&%=E4z*>k(mLm5U+
z`~)<@^imG-nf|$r?veiAeDot1uYeJ{(u491BqtnQlOO(|rh%FT&bpL3Ku@V5gD;YQ
zMqSzR0UrBCWDM`;eLh7a3~v^716=|b&6yQX2lz4B4YRC#zI&;M-?Nw0qF$~AW~{*;
zb`e~&B!(J2klaJ|p}5Ro95hRnubC829ene!Nz?f3mTWbt{VKRH@S=ltKPeg<`hq==
zzSbddN3%U(*|TJVJR&<EZY8v0pnTwP7Qx~47PyS}pBgUbL>;uvdC)C=4^SjrOx`)n
zfne|6UznakCU}}JXnXkkCUflm!CqaI{uJj>vIOQ4(zA-)&WsQGV8gudi}Bgm2ch?x
z#Y|@K&6$gVIs~8A1{_(=Meh_Gz7dSG6yMo>crY7!m4Q>pSiFE{o0*dB@J^2)_o>gp
z5M4aSj1sgf;SM~7ORQRR&#L{etXh|fS=pIv)M++N%4<`G+HgDrY^uG>rkr<d`uW)=
zD;TGLR`kjAh&z)*^;34OnPOKMJ>*K{R2_wL_05LoyaN7!rf@fHn8iSjRVR3$<UBt=
zn*T{7LR6uBu*{$F<jugl%=yy-9G5W)d>0CaG{MDmq{kA*x#~t&%Ta_nA*YodYisHR
zIE|6)`Ijo81useeBtO_DEq$t#^kROZAAZkubsK&JY}18W!W-T$b}9esX?#5x{)XPu
zi4b(Ya5j_us43dRgTUFO2A`NYUrwaqx`6|&mj%8C{Bf0Y<LGUZ&Tw5$I%m?slP1-s
zhB$EAqyp3yCFhzHG0mj>lS~>s!6YZQAz+zH;G)IgphKKbzmAd@bBAjiT(kR&i9V%C
z9Uqv~=OO1CU*`i$`ErgfAI>fK<<U#{My}Z{;PvXi%yPoNcDshQE(|X__-siRIAC6H
z%4QRD6imuI*rZxZOxoiG_e0H88+@~mpFf;}Yn|HaxeFe-Cp;~^sy*;IU--Pd_xs6y
z+E-p{d{idPM<2X>R43R+&D=ifJjO?k<vy|=_tE$l9N7#q*Mhf+G3eVQgOc_b^z<P(
z=7&My_YLxUX;8WE237rS(2Q>ewR&&RiF*e1|IeV5rwrPE%%B#_4XO`clM_sn3LJC!
zB{M1EWfH(HKfx~Z;AUFTdwO>t-_$*L7mhbznP0cix7+~#TyyKeW%B;OBHO?v)4?Jp
z@XZkLO+N6=-+lDHz&HPaV|IgWzJqCobHs4$<JiJueZe<A``!9{$Sn^)+mYAU4^Fzn
zaUCAW!|PPtfktR8ekb}-SErEGH_EMv1L!yPfIo_*r$%3DqM2S)XZWHfa7W;qj^LXR
z@J%J~O>6K?FE2Qwv~IOZ20sP9X%D_R4!)^*mwprcO%3o(Rq#zeeBoW!pe@ETl@5Gk
z2j7%(yHvvAV*VkIw*uePaFNa6(g1RA;xnMnA^Yawqd00XviuHke*?3_=f~+Zvl*=5
zn+Uv9mBBZ)E8`7k_SD5^vEa8@?LmK2c~PvqnX6ecK9;=MSiNo<tC?gM|M%0W5Pabl
zd{e&{JN1R!o69bzviaecYJe^WU-*(tP96OmqfN}6`g=4+@p!|%nadF6B=4pLKB`Le
zrZS;DejTj=SEF^{V6>X8Kwmr}TD8gcE7c@g)6m^~#5Yy^C0xx_cpJ1ghn7d_?F2ZR
zp3J+mkz?E{O0gB9$O(^<f!@>%GK)VPAm1w~Qpe{pb7l;)rg}0*GnB0N4rp%5GLI@V
z*?&(X)QoxI&B!x7+8uqdC4xMR2xTO<*c*KF74K9V@J))V;hJ=Stm5U&Xqbe~*cGl8
z;F~qoz&QoORq!t}r|yw`a|rHcRhaG#XQnUsrb`F<Q+30%7kqQ)cc=#33{~xYp=kKX
zz=;Ue)PJ7S459k;p1I)Qo1Vw;Yv4_-IFZbo1ZL9&q62`}UY`}N{Eb}|F5$=6YS#_2
z{Ve_Lx@xnF1O;^}M7G}#`cvm^3SDl~(7yDje5pUn+H~+c`Ndo4PYt%pg!U#!UMut9
zm=V4k@93BiEpX76V!gNI3qeB{td!^Jhb#@&?w;sCf|(VLE_Mz&&9+y_FyBn3=SXy|
z?A3Y72dR1*_U$>;&z<nfpu0GM#&mj8fJ#iF=8gd4bOhU!MXyVLs`Y-3LHMOQz?-lR
zPk9Zed>EXunhcEf%m|;&*L~P~gUq^1-MFC?>+3g@5?MPx%z`(Fp|-0Hu1Q1R>5Wkp
zFBnw<d^4-BQG=t6IzpC2wl+p(C}33X%tms@jfy01>sn5enuPE=GflenZ!HEt9X!OI
zPCZxzPCqMkV0$Mz!r|-<3oR-NM?DaHaPSkF6*MWkq63t;5|2B7TW$8{$v)tXzdT+9
zzl8~p(qgjy&>tPAUzOn-`#6}dStWW^a0!FpaJJE>`j_W^(TL_X1b+`0XI=t1cEjoQ
zEkUbnfeU(rM(GYZf^qN{)Rhho9Qk1C)UojCa7($DprJU(_u2ou@jMgvzk|WbTxG_^
zdAu9coPC$@cTGndGlH6V4;&Tied<`VW)3uwJHq~Pll=#*xEwy_JazQ%qG-0kE?wy1
zT>(38Wk1Tm9<zh~UO)D)q(kgUi`Z{kao=Uq`Z4feQ`tM%yLz(!HDr$rHK0jG>#%|T
zR9bNDQugF(necy?L1R+`tx#RCOat=PI0FCmtnvk_M*2W4dqFSjB%I9xe09bE=8top
zWMH1>M~f;yLeG4k-UIiS!rwfizZo44&Y^GB6g;#Cj_Ph9@KZzdM&O%uGr%}|0+kv5
zCh<?8URET_1pZ@{761HEcrtiLuX4=%1gp;+gBEIku<BO}(K`PSz1&Kl>HwNE@J+Vc
zA^L*1@WKM-GTkIM=NVq4KUQ5&&x{i<n`UQ5uU`yqG|r~En{8Thg*k6;ZQB2v-WAyl
z0}7(UsX*VjF1Q7Jb9|LunaLacc+IYDslhen99jVn`#l5KIz6!e2HQ0u$*Op`)6L7_
z>S~i6Kp!%T4UZx`*zb98hwun~V3?*&EX;QyOSk}dwgf(DuAvA1JZ3?A!}<HBCVWjf
zG&p>(H~VxiYKIpb13sbIeMMH^U9&2JZM-<|7c3&%8(cGz{^WD6x2<p_73e{o0@oaF
z3ZDnAnNuH31h%PG-mJhJ^rPOGwEv(<!#E$Jz$RbdNQ#4ZU%-!i;oNBsuR4hPHMf{l
zcOkfCtVyALd5mjz0yRlx_|~_Ts86byG;TDnGmi7Fx=H@ejY=XjW?EtR!I~!ZPBbYa
z=iR!?CO!CJQg~VVSKaVKZe~92eXiFxytfb3C)?p<;Fkj6mQI$&U+K#^|G=bY`%UW1
zdol1kGDev6sgX&x4@RAsPTkXl8YnL_pztlZ&-*Lr8NH!x;1ah%MQR(g8~jpeppR-+
z_rWvbqZ9(ywDHlG03QW_Z?cc}QNblX%6r5|ZJ+pPTt@m+RSYr)81$mQLCZ!N6rX5N
zij@XMf?u9KFzC%&g9d$u>-k{NM!1_^=MAc`6P?atgPw8tE-~l>9L-&@O&>U#b1%R!
z@H4B?)D!{VYzEsH9)KUfHA#2>$J1N@+uXT9J|5VnGJU1R7r_kYndbz)$qY7014d~N
zzIhHOa}|7(AAGX~e6s*7a~n(($<c@-j-xt{{lh-KJkJP@DSw35;^+TzG~jp&=Gg)M
z`Lo%rT&tL!FpuX>=D8zykNv<>-Q23^q(A0B_hSO@baE>N_+|t6ru6@@bPmvQy<Zou
zZKP&u+wHEKdhRebCdnkJZ8NpqRH<#dscqY~zNg>szq;<4WExCn?%elz_t|G3_+}V<
z;|AXxbA!(^<JG}iJohi2VEi{PsZY(JKGhb!Nq#XFZluo}w)wD@ESu#p4SbVoLabK7
zH?Lx2)d0SEV_}9kd~>E+Eb|!QA^4^Sd@~2WDf}=-Mc|t@`<bmt_D%GR7}f0)L)LnX
zR-2d|-Xuox<;cNF9iy-}F1;YTI3Ij7n%Ps|nLV{4&ZYNc`&E!jd(m&A$@N=wk&NP<
z(ehc0mzdleyDM6?%w+r3W%g9DXqC<wtuD-++I2fhJD3~ZVs(_-jEGX6h$vN{FMNMl
zv>f=R{i{eFyd0^w`^daWh*Zq5NR@9BsZSN~8Rv-9w_i>*eCpKq^G?m(=2WVMPG!Yg
ze800(YrUMJS}ETWPCZEZ-^{5chnc0hf*DiOBUFe?KcgG9D%i$C?#-jx%%Cb3p<86$
zT)Pmit82q`p3LG7WEL;BlJTc-%^<ry1$@&N{pS6PFfByC`2pYDSRJN2@J(cwFje;o
z)7r{m+M72_?S3<-=5DAOz&8f<sTV_-ediOZG4xKQFB7WbnaN8fn=3ULU1`>aXnkCW
z7L#w$upBdjGm^>vnatv=%)&d!OwFa_`b{7Uj2itXYV{r1^UcXghQ>>-nWymEY@!x5
z0}ms7bK9Rh9QJ^LcmTeUeKYn@ppxbU$}N(-uZD0A`_3(B@Nd_0&Bj~2lpfM}a?i|9
z{k3E#bHm}Ak%9giR}UR0551(H@v^eNt%`p03opnFdYV7E;m2W(__G4dochY@=J;mw
z;RSfdI)9u!bT?`btYynnk@0)pTLstPs~F&|{N8vVE79{y&B2eF_Ivb^+^pNxTCg^=
z$85`fa`PtENc5C_4zvWWr>FDd2dIYsgKMm%2>LnK)iGSN+H<|!?8APnAlJyp^i@8z
ztHl<(B0b4McjuZ#-b6wJyN)-pt4Idsx_q*!$OD@iEwO3+Qk!}%BU9^#O*@lu&5Ltr
z@eF<z*U<tQxNdTNY*g5$=z2Dd9B9+umHgZ<_#Sz@mGk0lZH*qq-_wd3)j-zpi00nf
z%zFMd2fe3wXiew{pU0E`v!5BpuYB~88r5rbn5f$1fi*?5VXZCJ(N}I(W>a~Q8N|n3
zhp^UDw>jxaj!P|OlhT_WmFT51FiqvIUh2-?@+g{8zdo#IJdO$@nGrOU`CJijmmMvA
z6k0m#-Y4{@C*|-Fu=oDX+H{QDy71fBy3E~TpM9!}L$jR@?c#PU&*6F2n_g%#^IF)o
z0v<dbL=Feqm&l1v&l;07)S;5c9eT3Uq3p{Y%wVB@08jR9>QE0~hwcZ#M}7JK(s-j5
zvCg2aRO0onLft6wC~Si!6aL*>J#*o+slfAsk9YuFGn?aE8F(Z&ed3%aCi<nG9U~)p
zK8(ZjyDAU$s=M_5QXkmNQ74HS!b|#8;GVD42pi$2{WSuQ&nz<fe|zf-Z0Qf*^ytse
zp6a80^vImQM>YYx?G0CU_a^gjDEb+Gr?XGdadM&Epx@jZiN||8^CUUP-OwyE4)9mZ
zB7f~aMQ>6DGI6|N8nn&<_|@J7qI;rsy5m>N5<{=yRJ=Mf0?0`W(1KP0+R+w|%v@%=
ziGD@2(@nXUcZ)|kQ}!SQKMGW8_8a?M^gAZ7|45>S#rwC$59X?T<-NN-NDc1zt1fd<
z9z6Gh-_Vrs@DIH}&PqQYwLp)1fhTN4cjg<w;E!^_yt$dR&Cxu&m$v0V1LeLhe7q+Q
z8V-6y5pH{D^-|f?yvNWtrhKLrg=Ud(6J6sN{Yks1|FD-xHVd{LL%nJMzJ0tT1{%pW
zcr~#Be&1UE%{Vz!#Z#db$gs%^FaJWbyGxx1ei;Mbb{OGMweBzt^`8c4qu02t&->{H
z`v=Dq_8+jt9L|wEF6vLbXB)MqR?K<SItO)QW)~cRXG&eRD+hI-P8aRk!^aO^qQw?>
z=yy$r4)cB4267&*bujnEp|p+QF&^`(v*=9^@ayeELyF;J^u|r}4iw77-slP15PrKH
zuvTB{)9(HGoh2OlnAxG`=j>|9K4=|1OI!Pz)c`h04aZzTms!--Q2&O8u2nZwtF)nc
zMbK>u8QQ|_s>+7u@PE@i4YiLkbZMlabBheU-)HFFO+(Ax7<!$`tPwfQXy0aL*PFG*
z$E?G!jsdr{I$&1T(`J1^qe)z0R<1c_6<lm)E|gjR|NJ@}UHSOI8|po<%^$Rzt!OuM
zp5j-5Z5pB7oPcct&~ARgHjD4zdAZG;rknVR@xsJijblz7eqKB`{irAPJcW0M`qI*)
zFa~^+2;a<yTb{xvqhXk?u*@cyrXEMK{~g1*{mK3HV4ws0@dv{*BlzBb`cGay9>Vtz
zLc6K60X=9rTFhL$Jrm)W;c?pDm*3k3UvVV8>47kqjo;HAk2I@s;aYq>d@~%rxml38
z3-C=v_-0gU>Supqwel<eVEE?66Z)s_#Hu`eQyISLy$7aQN9_r|d5^bvbv(H@@XfN$
z^o%?48QWl^=G3Zc#wrB9`I(2WOV6B5`llX~ePh26LoQ*A#x0>XHJ({iy<&6-zNrY`
zR7JljQZ`0!(Qh`wHyKX5G-8KK_28R*)TeewyL1b_Y21MP;<7G%Cf9EbeDib%9^ysO
zDm((-A@?SgDOxXTN7GLkt(AC-^L>m`8u%uO?3*r$cyWeADI;^IUe}LO&m2)Y`HUWL
z`ldcpmwGimQuUZQW%xv@VY5iG1tRs}f8OE?k6;*PPI+v0%6kr3#lxIx9!-v)iCMvw
zn75h3iPw^TZ+ykg(Qf80rGJXdn_`3MrDCq$Wa?69@fEKsNse*G2z7iB&J2=pbx#P_
z{bAv{8y$}KI9!p9=%s>hX6OE|J~bY`DG1+eg>R0*H(@iv^tK0mRE{v^hi^{h2~*(r
z|FUl?z&BIohsr!SR0F(2<-~ilw-mGQ(&N8DyQ#R7oUSR<rJVGJ!#7RHcCQTI42Eyg
zz&CC8kYftpq==`Vv`4Te!8f0qGG8+<nl^nH=TDM#1K+Hj9wa~dP@UBJ_f`zjk*q;F
zhaX@Wy;HUE2NWUurYYXy9cb%4?*~x-A^)o{{jkiVX;Uab`yS)F*+Et@wW%<Fe-*3C
zjE3CI3;#$>3jO9#B6+z%^q|s*nw$RS0+;c{(sR2ujLe&A%%Drpao9(`3+c6_|92t#
z^?S*ERFw69SOOf-m7XKkzUFRZVZEm&wUOQqw28v3&xb{p1^W41u2&nDu@CD_ZQMkk
zcO^9FSNMl_qt^^^sJhi5pGFQ1EWq{ZtzEs&*tKS#UB#GBm#&^&`{0%A+icAGCL?r^
zP5rytl)ke~MuJVhr`zPU-lpm!Y}($9Op3lXHA!vN;#C&4qG$8=L<^pIi!Rr(Fo(mU
z-NP-qcbHmvW}7xtw!`*z{o)#Vl>On-74(0mpzp#)?F7$Zjw<wF^dRqd0~s3c(R3>R
zSDV|on%Tvd@E-HrM3-Vs#e-9}3H7HI%x*x#DQ4i!!H+X;5FT9mobuu~S(TP~>X)fM
zoTN8>A~mN$?7gX%d4{3mgp=FK$GJFKa?~}@ZCa8;*aj`VA-$|c@CYQ~s~*SxwVkIr
z!iV)eV8-Uu9}0NFVyrdo@hi1)C<->|)|P8Neu9;5Xk4}JdRGN~tC3xETG{nF#IDQp
z?W$OaI$&4!!ON&Ou<xGsoxSxPYJH11M%sCg*tG@Lxy>5nm;i&~zfM0AuN8dbf1EtT
z>+lNx-!HIVzY_FH@!T}zZ~e*b&0@^K%8K6g3pS+2)O@wK3QfmP)RgBS2l;}x=u2OP
z{s70N-vG<4<nPN*&u@9oDL=f$eZ7@^3Vktqy%kmxy{x&9;=B6j%SxViddF5hLpRD!
z?lF9`*oO?FA#e^oM~=s2KIfqC{-1s`oLOqi$id<K9*CCn*~NW<emaTg=?68f5gqVS
zh4?7z8D>BiCf6vPzfAbma?)#6V+6c2ihP9*0g7oHpg#D3J;_s;W%k!uZ+}gShL!qI
zXBy(K+yni!xVXQ*!z^ys@SgM9=Rs!*isbx<H(H@bBrGL+atpoIr|_jb!uwsAJSKRy
z1@+N2JJGMEaz4~%K2t%sg<4GGY~)C1^wPde%yQzk4!5H?-sVK#L)#JTvpX}I4(IBW
zSNLhrYs7nOI$B<%9rP#Szq&q`8r2B0Zm8$X?aF&DoSJ1T7^f6<sp{}ec|5~~=<QAk
zBkx7$gBjYP?|GoBc+Z0&cwd^r9r6vNRtYyx?Bq}>Sf^_b{=X+o43}i*^M!D98qT?`
z?HsCyC+=&0hfe+GJmXw?f5NV3+wI!A(aubGW-;xyE6+8wm2Y+hz}QvMbt>5L@Qgz<
zf>)lug3ICQ`hD=hqdm2pjBbP0RQ49kv=0WKjZVZ~B`;d!?v0%DJl0}$*bnuyl96N9
zg0*HP?_owa#sBiu(1c}%w)Qe~(AQ9(R)!AOGZX^LlqzayWdTF$^BJ00*iaUDCkid6
zdP_b=!#UT@(B4UgI;?|@PVjZN3~j$?Xy0qTH?vu<(17xLnDsZ*tn@B3UOQN-KR%n0
zW+~pRo>RDg7Q6%JESrmFG~cY%94TO%s;{W;pxsPGyIG2M^8&VUpxxYpZKBa`{=ha<
zV4Hlm;E5afXV7kZ&~7qaf+46)t;Z*GmU_~6>PdCso0YJM4}4>TPZHo86MRz>j_Lb9
zuGs;z^yXN>@r(O%@ile%+H`zRPIMd(ZUgyzD&MnZ3t2X6s4*>pYi7en@o{)ssCV_I
zUeyU6iy(*4ACI*4UvD~Jhkla+zBvWo<S7Byz&Fct$7vvZlNP>t^gH(d=S+?Mr{6q|
z#W%++aQLRiF*0p-;lEjr$7UHC4t&!9zKI~yFBkrseDF;yd~>fE+75hEvkZ&`-;9NC
zzI~6;zI)W4&N6$7Y`-;g@e+?^9zzdihX=)=E1~1mkI}Z$F><Di(b?D3pH8`SZJSG1
z7ceUvz8Mt7eGZrG%x36K_D$ql{KV9!TERE{7Dg-M@MyIO$8Suo-#2n^t{29Ola3xL
z=4uwZ6@^|FrJGa8zQJ2OkKU;dwaG9}#~kq|a1DJ^{isW|85gNE)TQ2dMrt3msi9>e
z<&}+E)pw^lF>~sAl2gV8YEUzsnmWj-72#yyv|%<w1^TD5!#(8s{kajLSGyzB6yMD#
zw3~FD;2j_4YPO8fQG7Rl^5Dfu2JhSrS9DUi7A^>v=TL5=!d258t~1OGAJ1HdExE$w
z_bW^*9)_s}y;E-ZZ&t4jQ)Bohsdt#N1W>oC%{(gfn-XOERlOOi-tf(%In=2Jq1%w{
z7f_!}<C39T{+$^N*FrRTdx*A^%~cq_c@N(#fp3ziPqqCNtXA;NJhJ_|CE&FgPo6t`
zvmCw=eDj|A)VOy+dIaA@z&H0NQ|IqR4{CGzrYg`=1>Zb*6{zv>P0E?fV@AI*>M?6G
z16js*1JreEfKK-cpjVrjd-R51c|>10x_mzB{vN*ciP!emn;cx1-}!0(QTDg9>EZE5
zGsj!p9N%k+Q@)C%FJ+V;J*t(+h)qfMCwu0p)TYk(;XPr!3HjhnMi6Tjd-8#?tYPfS
zi{~UC>kU4>)#SmCMsElvQv$ER9In@os7=LEld`cFtH3pOh8wr{*&D85ZwOnQWo<4}
zfoo(&hwk3AlTQTOOrkCo$n3f@%y}W>JjFVjOp|PiXSU1MIGb*{Yzpgc(|6{)JRvKz
zUa(DP%*=qvVpYmR7OjPCUe|+Z(2BReVDCkJJhGfcOG7NGyu_k;eBEQire<et%Fi|Y
zC^fPy>>bmfiT7%UzHo#ZRw;7f;kQfFWnMib!;o6Z7YqK6UOq~Jj&q|6d1B}@WvQ8b
zfOnEVr>2G0Rq7?VG()MMoFIRdnqAaSn1!ELjD33Jeq?S?YsiMKwT5--IzEAf?_@9^
zWuMC4yBX`$rYht;SEDXblHOD|Fa2eI?!%sZLJ0l^w3|Neo@&b)H3>~W{d;s5_R=fd
zsVk*I=V#CCMr~k2YKN|8wW~l%yUxLJ<H%4AY!2rz_iqpNy!}}nn(L3=!@m4T9$53K
zT^97L?#wwp6k%8Srgo*SYggSe4uy86M#1m?f`8^b8qPvq`;0eWoS)vxlM`Kqb7D><
zG_ERS-Js>9;CYJ8$aC`9TT?G`uB^njGmY#W^qCo%=`X)Q&n|Vit-a`HX^UP#&rg;7
zct@Y%BYI2U9jyGgDn4VhvxH<YBHGZb2xex?$8&=hDjU!5w;bq-bvReiZwd@1&ut?)
zln==af^Tv)q1M`u44fI{E6yd4=z*_n=QyY5;(O*b^F&L#jV?5&J^ku<rjIQpM}^nZ
zOpkts73Ak#^rueaFGcz*T>*bxwfd_cdx0j@a*y35Ljm9N7V2Kz>cBN*)%@)WJMla2
z!A;evKOWzUpPSd|#B=!YCv}}1=pzluz=KUq-t-oA$Nz+`@`5_!s^ipMUQj!NU5+Hj
z3!4Eg4{o`fisyxM<92HH2;Ao5^K~g<VZ6gX(Ke=_iALYTr*Z+WFy~IvdU}%*@#-f~
ze}ZjFz@**#(zoO!$J+}QMz_gZAMG>`znj{m`67DB9%?%)9rBtFTg<{M)0>_|nD-jT
z7PQX}9UZETCNr`HI*SK&Uo^cl9G=1W`FYQr=jhvke1^s_GUs2qYz`&BHub0#4?Jd9
zN0?^7Y`cn%hpA`SH4&~h9kHv)Rc0D+z6J7rH<gB|+~6d*=>eKi9azb5QCmhAdUVC1
zRp>e$w>vayqC-J^&6m|YCO)qlPc7`VRYsIm1!`GzrULJ8H?$Rez;#Q*6PM6u78?56
z)zE5(p~x17KGrr=uB0I+OmjB3p;$LVRSUv3+~;1=(Cs>gy0tUZ#s>cc8u;lAC671s
zeHFE;GltIK#mSupeknzdxXik3F{@}0K4ciDU{CHFY}U$AW+lKlKVclx46_d6M_vl+
z6r9atL$~?o8?J$N6N`4U>@k`OY!ica^9Hu*aTlh5Z$`p4e_)%o*W=Lo>C;BL$v|yt
zDz&M{r|}5GH&O7-p(An3is5@;lwasF?O~VH@J#|7lZj(8Ow*m?&UU=O99Q}H6&!Px
zuY1AKpYLhLk&f@l$p3fTh^|C!YQ`dbIWypwap*OJ;G3RkIk9ofWd1L^FUo=j)H+U0
z6{oQ^;W7B85qy&czNru29ENYY!8b1WCiV-y;kU8mHBgVb9jn&x&G)15&dykEg>Q1e
zH%H)`TJTN65dNS3@W0`)3dMiZ3%>bYgBexmHxaqHKOKzpEk?J=_G@=4M#-p61<#I=
zM#rd7_ZY1WWR7_I80M+KKSlZcuS>rlyOe_Z)Y{E1Wt;0#^I`ak!(56pT+|$xEuP7x
zN7Sb(!Z%at3-3>^-&b;rTZTmI47oQSt4E`^l8KWBKh8U5Y2Jv^^nJ|N92Uh4SY}Z9
zMJX>cHEWRTH|9~KJm{O6NM^Cem`KfyBjb-uzgCU#;gq5`JPVr6SEp*y8-DPZQ+4T`
z$}r8TrFe@YLhu)RIQ0qd&Eu?2oyA++{91%+?xZ)0%$t(KBJ?0OLN@ORjcE~~XO+k>
zc8ieTU*=8S3D>w2;mpYimyN!uwD3(yW)b&l5U!2x;hM!<%@dzt8~En_`7oW`$DAqn
zrtoa^oBm<y7X;@tB@?GG^BF#es{6H2vir#>W;R1bvWw>ihAKtfP<?bq&ms5b@#PRT
z*h25rSn}23o5%3Y@bYA_!#Be|1Z(DH`onjj-M}|%;G08TgLM+VS%Uw@FAKR}WEZzQ
z9i&SenU^yOU*f;)8~CPbdFuOFf>Z?kW;1=PV<+H;q&NI<oj@&09Vpuk^!3#NI@2XU
z_0igM!8hL?_-pEZGQdXoGZN8Xn`-(iFsr|oKP9jDfS)?g^iyas`MOoe_e)Mb+Hqzr
z#gpa2zPM9)>MK9-zaI3_`-#lY#e-5XKez9w;T&NfKZAbXNb+OYm*>yM3~}^{{ZsHj
z+Ne{pM%Q5P>BjZ=7yHScgSid|{nuZyJUd>2%lHb>mz!|Sn!xolkn7|76!i0)w(I2@
zYAOTR`<dAT7D0DlcAXFPr%R)4+StRUkC8S_Vvb#U7$;>@o0hk-DO-J;3Rku<PYaD^
zjYSn*Fit@-xSpDI^s!l~+245-v1lSWTdzi0H0iTNtH|Oi_sOc>rEQv-z_ofDdiZ_%
zK^l3f(;<4S+{tO7mn9=Q)w#pu!QV&EDM!r+zr%$vIHWS37g%F(KRk_V(N^wqu=jq4
zPhx0qbS&yGg?LUbT%!kMGy84!>B*yD9e(~8o}1`z^trHBg`CC%ffoOZ{kQ)KSm!r+
zvDv9tQH$8d*G=K)Zk<EV8NY8xZt77F9cojIS_A85BI}L8p4kN36o3tD9)byZ{sv#N
zD_c4{^K)#vT){36SmhuNt~(}t5CL{ofpKQB7FGRA&5v3_Wq7h9^NxK_+ccyNHHGYU
zwadiJ<JNXrS#!FKgO|Fnh8D-`QIoo(6+h4dYG^O<Y8N7_uZFkc(RZFxlk$XZ{-E7V
zNAvo9p1JJs&4D2>FwAlbUUPncBVmB0+sHfNd7JXlQwyGvv%>Q+h5x?<3+DZZ&xD#`
zy2jK!Vd!Qj$y`P^+*%t?PaHL>C1f$dl4mlJTTq2LUDN|l52WTZlfJ68u+MI|=OA^%
zoxVzewpObtv+k?GS7;_*lBfZq&sFBJ4_;38;|{czvv^QH^MA0_<Ux42-F!9sB>i#T
zXg>@6l<_{bwiNj7SHO7aPcy#K-$ZYm4?SD!sO#RelLM5)M~g0@pP;L?xrvTOP30(j
zQ<&P%RoJ2zjA4Z}Mn<7c41qajkzv4dP#^uJ!Bu$ZA-OWV)}zp5`hTa_`4`?_&b5F3
zwUaNI_jHp?8Cd4W0eX!%SAOm1``J5m-9-NIO72JVc+YvVU>Xd}&ku;?mfttNBlX`V
zc#ErWZf79dhVy45dd>3n)NWy$v#`U6K@R<)wz<>`4X_=|(*VzKEjSrgj;!p^#VWi<
zD^LgG|IV`?srm1KZ5A|i=mhWGRv8>h&UtkHKAenhQ(=Q$(Rg8;oKtzj>?-84EA42z
zLT16*o9w!Dj@guGrxVZ%?dUrdKjM|ihX)y+8WY3$-HW}>WIV)NBSuZ<v2fpz+7697
zfR=m8N(P~opJmp8YG&#W=rAV@bv|OK#UDdiUmE088ahAGkh!y=#&~PCxEs0%*L+2{
zsgT;xr1Xa3vKnd$<AkE;<Sq}#R54Ty2I|t<kcFC-V56mYaR#FUtwYC2dS+-Ab*hk}
zW|+#Xo$bt;=w;U4Ftgk{o7J|jSph@MdNLZ%@dPv+80W`Svzkpa<9WmP1J_jIcn;gN
zM!Q*vc9Rai8F(L_fo%rEHYsj1fAS{$0o&BPhAxA4(;Dq2C$*`y)TRo<H@%Z!m1D3C
zd~*d2X6Am_2Tr*Tr&!<@8yvG^8}l1Dx^9I<wvd0!?Or|~$88|DW8j%4eBCgPyc}0I
z!!zr6Y|HWXEFi~b8vdQp{QLnhPIp+!MLjA6mhz!5iry)25A>u)uvSfaq~Mzz@XezF
z)VFe>17)ICm5T5Eh34{^*$ePZ&L_-RfN#dZH}&9~CGgF4_{IaisfYh2+xS@hA=__F
zS6B%BW=jXWI89=;0=~Hd-*~$*Q#^I7*1g1k!(2@(+Re_D<lr!uVe+UL9qk&!{yjz=
zTBGk&B?GWPjDCD|sTH%Q2An0Ic&bajz5iny50i^|zszYUfhQ+}OR1O}p647aL+{k<
zdF1)g8=fwR`I;WldQvr7Q3cR+QuFy6dcd!ebwjq_g27SR9Ye++`i&pGQ&mz%so_I(
z8v3S6E{W6+e8sO^<P?+Xx3@kss7gd?Yo<t*VrIC<9j6{1b}I8Kw3|uH4=2~J3*O?j
zc#98|b83DTW{8tpT#()=Co>r`&yG+L=1nb(A>+m?Lfe~Do2rPmlLIdfeN)@<-uxib
zZ_?s$9q%8m29e?TWW&|D7QP(jPMOn@e*@q6!#8u`oBr_4d+Jk{W`@Za4C{o2=}7Z1
zIhoB+IC+@HUZfVaCsc7$;TabnlYMiqN~pSK3zg4H=7L{9zgbGQ`j`;?ji6V&-G4K~
zPf(v4_%>MYE(L24d^2D%nW<!B)PQf^`vxlxzPXVb{|$UomDvo}R>L*%L7Lu)ybUrp
zMwX>#9KJbyGf*wo)8jgrIm(`anufRd1s=)S8|ed~H#{K}k8#5Q-OUuBo%iSs$6FZ+
z-z35}1#8eplEGhl9{DMH4|%y${q!V&?3<E)y7io%pRLSH9OkP>Hf92p#v}WQUa7t0
z<&PyNv;#AB@r@Q^-70g)Tbm}3WgdbDr!M}RH1ziDV>aUe){TzzbCzc&1N+V<ceu8*
zmpn%Oe_<Prs;uQ1$pTA4>z>0kEDWCr*U}}d&+{Hqvq#UiPUD&v%S>L@<nm)|%m}n8
z7OiD_Bb#RBv?)^xo9cYFs@_klVsqHEnOf5-H=9au-|V{<)tX^Zh{K|pIV|dTnGA}n
zW;MZM*o4n}py>=XTQqDaGh>(=^R<;#HP=}6z8$}37<vcS=*wvHMR>e9(OAOHktLRj
z8VjDzD;%q-4>*Rx87=W@rlH670c!y|)0*6Nea~ap%yQJzp1}=n4i$Vt{fPbdP}b`X
z|2!6+c!=A=1(oq%r>A!wU1liHj|aL<)OLF0ci>0j<H@XLRaoDeQ<G@Mp8nx^YEt3U
zACj|%-NREf6wRe6ej9#AF_`e)2l$2cVi>wtz;e4LPqFi!VEuSuW44%0X)d96J+!Gp
zK59<wsmJ-)b)7m@W7eH|@Q{mrb^-F^yJfJeEF4*Y=OXY5968da^7E`(;;^bj2`lp#
z;l)%|?a#&>!M;{CpJ~;4UYGRnX50RJEw#ed{Jf96j;0)5THBd<?ewfHu<&=R;%{s~
zeUjSQ>o%;7tfz+>lG`Yr%USq)7U0Kfir2gZHA2?jQX3q4vzWRGwUW%#B@Uxeoq@4(
zvF4TM^D4_7%CeWWFaqvF7aaPHpU2VsyMt^m>NRcYi-tifvCfa-XMf_k9SJ)+;ez?~
z@TA~HdBpFX4og(!b-pmxOQ~S0K=>vddgN?h^hmUwqp<C%$1sgATnp0`x=h|Kyl@sR
zX0FXsf49Pzyk-q<z-rt-s|cAcXB=vA9UkD^X!rw;Kr_k4?;ObaAMZ)F9h!<8b)!*}
z;Gjh6L_5ifISaS)v#Y}}W1f@c%ehhe1wZ$xr|!Z%^Egk|oy1QA&ji3V7dT&5!A8mX
zyY{b!e-hAzCi3$~kssUxZO7uNV1D0op5yJX&6PSlAI124|2lLP{bPNEL+PnY2Jn0u
zoKK@e*<VD{OV@<DQ3>i&>G7|@ESJ;ax#8%D)^djPEo(VE#dUdoIBvo|1z?+_IjCX&
zvg4P9V=ma4-N!i=Z&xGU!||<Pnc6USN4v&y?j3@!yLaPU#S^w0Pgu1$-YbJ)@$0<r
zzwrIpsFT4m1~ssc6S!T3j)ca${0%>AkX2>kEb2tw%sTkO>k~DnjWEd?L(Qfdx;oC#
zgC2Nm&|aF<qRv#p(5Ac`xeWdIXwuzVCVjhMQjgmv)qh~pzUL;L<NmN@hL*rO3B?UE
zG7S3U4c+oG^eqm32QSWsd4^`}=DvH>u|66aQ^2e~_>FfpHtV%P4^;rn6K&ST?q&tz
z$?+O)R;saP<%4m?({D9%GTIKT)9P8A9>O;3V4D(;(Me#N>}WR=&~CE8H)CO&)VJtA
zx`96!wkdTLP3AIv-)J}P)TS;_oAN$I4e3Oj8pAgS4%4>;-;{)J3d1)Bd@~K*<{IqM
z1%|l{%T(k@y#;oHUk-Bw@o@_7JHQdo*R0?+KllB>cQa#koWhpkx0%ahnoMuhNP4aM
z;dkzePbdnFCzu{BFAfuWQY(0?Azq;xFdBTb${p^4Zx-f^)2EE|g26Yf;G3HtsX@Jr
zmGwcawq3_Va}JH>I6h-~!*6V3?!wwwITn$LgTJ`Q0KOi+Sq<MTgKsXtH(%kK4erdV
zLci&Yx3~elQx(s}=ni~S9KLA--@F@1_Dw9a8tgH8f%m3Cl^Bi9N5(OHv-^QdHQ<}%
z%+-wU1>b~{gVVvKXLVfK>h97I_@?kD<}O^0hWDdY0={X8ezPqgS_fJ(gQ{}0mgYm-
zNy)s;S5Z25DT=vl)T4%z@kj5}QRYr9DiNitDWVkqfL!7Ok$S%<Qk{lJ>M1iBI-8iK
zSvOMo-02z55UF>}46k;J9KXFz^<V1L%yCW~BiHYMpHuysJGHo^6HhdL;;ZO4`y%LN
zi;x%I;>?30)Ye5k$}>Xen{i9uRAVxWi#-chL-?jHe6s_-nbkL3Z6nApZWXRORl~Kc
zV7T(84Oc<<CKA4hf^X`<H-+GvcGRb02I0dAA>R+a2`s|AhLmCKNkhpcCf{#-sD6`u
zbD%AK;$`s`rwmnv>mlT9G6$I4n?!PNyqPQ04!-G2eX2QpbN^MaJm8zB@J$Q&CMSHe
z3BH*E-!!QetiU|M`VHSi91kKBjQ;R3WEV#TsT%rCa`IE#|D{*;QlPRXl2eBF=4gjN
z#Z(H^UA#8~Rs_gnRDfRKmo#bz=rKOZ4)lr?--2E<#$Rp80{Mu3GdH!rO5Vq7ga7i#
z2(sqvep+A5Pr{$++C;xC`{2KLV_TP?HwC`wxr=@2aB5R+$T-Oh<DlOR+lps&IGzmF
zfidihx4uHFVJ~d%#vYh8{BLR2H`eBzSIF34jrr?MPXyQalUe@dsxnJ~diqD!tOaNT
z&Da|b`3WOXn;*B9Uhk3Y>pIwVaSrSA0Gk?{nB&sehR4#TvcIhwfG%D3rB&D7TIKWA
zs;u9w%Js^sFDI?ks4e>3$)cm=Z3U;I|C9cP_ovL%0nDmMt!hGcixR8BIo+vWJz;K4
zL#x`)v8q)in?g?D<>7kUV;Gq`MOnw5vVXNwOJmKr_=mLuR;kQ-)TuWdLhT_iGxdjM
zc0Da?$5)R2{@kYe@9{SOrJh!p`>6>P?o53xjYIPrd#Ne9*05Gy3aCV08+=n79mC@q
zJufS$AxuTn;yAGgO^5ybRv2e1b&7F6sBN&WRUE+H{0-|ox>q1|rA`%D$CKkFX21OX
zBtA~|z@?YMD`V&<<8K(z*RHph$%fx&(?yPj=r{#=TupdfDZJ6gg7Mq1#+-+7{{78q
zXW29#EvrU37_yX&`9(IJUtpEj1gloPwCMK^i(+|Rmyk)H4gKrDGm9GLwQ5gIE7_J-
zP3dpdMtG+`ta#~|RnrT>qs49dTfxSA$c8uAri^RRk`nFA;P|i3;jxZ7Lwe@L_a&EQ
zG|zW3yehBgzd1~;Vhy>D_$%hX!2M^!Gpxhkd*MI-$90@93k$HWqCfqc;~cvWZE69v
z5Cd*Y!#Q-wrslhu+srYM`eWZGc(zL0_1u?QT_JjhD#8-5$Vz_aT^Mv_QS#nKa{f|Z
z&46a)s^_h`Hfm)49I;;7#W_6+UFpbP>OtM;vu=r>`xidj_xM~nym{P1(b9InTXlKt
zZBJ1zyn^TUKDCtBe4L8>-`wQoHKslV6Mclce)Y#c8xJEbg=e?GHAku2^18R-{qpKQ
zo|rS9s(hK}g!@`uz&FF^?N0C>I)n}c52b-^c5Ni1Zw<`I&o}|wOdO8CC)887V8)j4
zOveiNSU8V6mE-rg(Vvtb-kagj$rwDrX7XcrF6T6&76hjp@^)xb9{LaS!am>an){jh
z5!^iP9r`K9q)*h0(fabkIAu%Wx#1X^j~=Pa=u{~k%y@$9&)fBobMRm<yXH{`4XnaB
z<_^;&x9dLVp8o@zO3&kc%<IzOjZG&Cz~M#djm*!_rVj1kT>is#VD%4t@VsYI@mNY0
zq35rtL-raroqWT&2TR=lKz|c@O4?0^j^YD;Imb{#7k(SGm&fQce=?!Bq%bt?n@L69
zn)Kv}NlEv(g>8;sGbx1IJa<hx#Qj;`!#KZ9%9Pn4!_d%VG@Qjv4c(+hRXD~_&Y}Fw
znTAI1Ff`yUJ>>7HW97zYQ^c%wb*Md&{bxbLd5ea#gM6Gdz0Jx<ohl_>oL(?a1huMU
z6VZ@49>X<<IflVDe_)&OuuVR+n+32<;XCjJY?J;bHKpsY32am33T$(U9wxM#2Gpj0
zq20`;HsucAEI&%#419Bk+SH1D^eRzr+O>=Ra<rHka7$H~<qGwtb#O~Jj=da1Im*H~
z3pl=Wcy6Mu#O+bOZaUgcBWhD+65y8E^odX8|A$kL>VrPi8D?_QgB8fnaM0`3fxn{#
z-`9Z0TZ6pi^7LuJH$U^o>2D4+9QbBLN`B@~SO&gX{{nu2Z@R%ZC*hlx@J#{uru_C;
z9ffbA;G1Fa&6@sbIPlF^>Qe_T%$ur@-b4SC7uh#2m^~H$K87AFG#ln>7TS)#IFbCm
znQ+aB7}bYww%THp7QRUi-z>{ZwlUd$7p{?cv%{rt)8Lv|a{S1>8Pe9JC$-4Lfp3z-
zH!b0tY4FXUb$D>*Mym<^;RWEEyshx%R3gVXPqd5_^jE!zQrZhq+PpbR(adG&96(<<
zy;Jqvx$k$RYA{pt!2U?G+5ek4)q<G}_uEA(bFE0`FGuQPYBF%%IO#=aX83ximJN3*
z319L3j!vzw<<!7JcyChSEq+DT4SnGs@!sqpv$*y|dZhZ3RUAowxSif9*d`Y3CKH){
z51)i<==pG+UKy^D<HPlr-thcn-?SyO_(~;wIq*$FnsB}N6sF<uO=I|`Bz!Yx6(3Iz
z)6oIcr2NC^%|yR}Z$|wJ)fr|_*>;3#{De?#2np3ya*O8|V@?Azr}AE6Uigj>xx+V^
z;hWO%&F;1#nqNLd0o12Hza;1D9KEBPnQJqXyy8B=3iYQ~9PiDv(&XV}3f8j^WTcXR
z^O)?Lk|W734hfQfBXZ#j&@=v>+~VVbDix2X*cGU8@J-svWTD;*(CDSiWWXEwifoTu
zwcwo8^n%>-R{{8D?>N{7|4l#in-hQiROF_gGOhQMbqG0V9sRVZD7{ng&9wFSW&1Ke
zvOQXVQD41(%j~+{WSb1eSJILmXZ#i+^u)B<OeWhv4%UGt^u&xp=O{gw%!*E|10HBM
z)G-Esg$Yh`jb%;fU7K~9y{+$O_J!xE@2{nP9*@7(AXlTfL$R-6nlt3tB(M+c&Hk^B
zT`9-HGiU?-n%QJ_rzVxorYvYQ=^j}%=8jdTA6k|7x>aXcQ&+6Es{B-|PJFOvGdbx!
z3s{uqJAIpb(2%Kt-&t=~-Wz7c|1c|<I#pD#MUT!|^o8{~^%SdqX0fT}Vz`9**QvB<
z>d)=+Yl&yF71!CFtkvirUGb(&V{bbCv|YU`!Wr-FYF*H-zqjD9n>Mw4WK*?QHnsU;
zQx&e~nM&EwOVOdY-XCLqK0>|eY6JW?=mpkv_$%JA|GtQCcp+=kNa|9<V4GR2pR9Gq
zS>x(an@IkJKG{92tIzPk+@=<Q?v)D-q)j3A+B|1-*4gzNy(<R2=H@87TCq3o-X0x>
zwf=At9_&?Qm2I`@KJ2#b8(L0An1;vH0lqQdn*blXjz+*Z6KvXb)v9~nth%+;szUJN
z_-L!P;?3@hmucO6i#j^-YDZaA;UJ9ooLe`m;%ZxUtfN&QdRX;fhE>0KEq<K9-~HIC
z<jHMH#~hwj{cP$Hk1m7{eEMQ^rEPX?LObZ$8%BA7{)Lv%r98e-6JAlg=H<N55Tal?
z{22AnY96vCzaEcP#r;X>3rnb%++=P2iDs~iHF3pGyB6@bUwDlt>w=9LZ2aC`HfsE^
z+8NFrxbsFbyB1S-yKoepAd=q!TfO6V&w?Sc;NLEbe)H@py-B&r<)_xRv?V^8!RUA8
z@ksr|8_er8Z8@A!nVjCN%-H^f#_^8+vXSVHqsd-@)z5B*z2Np=$6>LH_@UsN6fg0m
zpy5=2e|nW5OBh{gW)R+A-V>jOQj?qKsY~lQ2Vj=c2dORb+BfDX3(NfTl&^tpZ*gQi
zg@+BDXy<;k7}&-H6TL^bDFY8LLZ6C^=Iaf#DHtlC2KCai=sKBTN6wp02k=i#Lnq~V
z?hD71@IaTT59d^;Kf4mlSQ?!yE9Vr?rwOLH!nyx|=d~A_%8V0sz2OMFZP(f-cHVz@
zXL9lP<fbN?0be?r%YwV;f9I%)uB1Pk_sAvArxmg*41fEDXVi`1&E<>e0hj`3kF#mX
zW?rX@HvM2PaP|(`AM9OgJKXcvs+B=j9U>?1(N?pD@t$?RNKaC3`T;oi`r^^uP{^#>
zcML6GNlp!VO)eN@vXwegb3>jL4eiQ<XNKBS<-aC%`)pEyS0?p&XkvzsNj+hko0m=O
zxl9VaX;Mq>zx>*ygFj4q&G%%1jkZ@XbfXcyRED9GQS^roG<1BLp`Ung4jwSH;-;a^
zzYOKeXV#}GW~FUz)?y3VO(3->G#n2!oWK3dYDb;Q9WTx>7$<%_{y#Wp1YGlgBkU2}
z0NV_OZQRjrHo!Jj&~6f7o2)nB5wx50uuY-MXfkLw&hv5dqc&9$zPU|ps@HM+HSo=f
zLwJ+n8+Z67Cw$`$-!z18?xV>Jfm?>aElpvXW$XDl94%m&f@|?8uYpbY_%#35hOgPl
z|3#wRBrJoI7T`Ia$?uA%AABgD=U)6@9QCGfv?G6X92=a|9zCczkFh@9=IZbld~>uo
z`8n{71-|K(0X|DXPVx`<<pbIcd~^IhK4JJ~K78{p`=&H}Ga0^_&g`iP^J5h~E>=Z*
z;5!aO*YQN_fo*QUHWP}`PepI|kbl|5&tnuoUwDS&F>1XnMk`h@Cmg=%JS;|=Trui~
z_oiO;7?mgj1Cd>vo!L_}?z!}Qi%W}Vx-^Qua1S4Pr&_y|yM{{{3%fKmjZ24^8$RMh
zw360Dld%QMFjsRRa~b}^H-9Qda~?<QLb7PBdmKgPLX_H$j8c|v|9Oj3G>y`V{88FK
zZgKikk*d5tQbpn;wJn@l6l`Oy6R8_$H&@8@`|}pIA=mE{xy3;Vc#DV7JLP0{csnw2
zn5p?Zzf-%DJGJ!%*)|s=)C2F$_a)4n8Xut_-6NF4k6BY_H}=xpW<%roN&olbaE&@1
zu6*!K75F9)z6timZ_}9h;bi+wEkMp6d@}>SxdPwhfNy&5r9MUe%|i5>OZaadkYD^r
zVVZ&e=H-u2^@VR-+e1}|*$l_gZvyF^nw2M1^*@KG(ZvuAg>U*SU@q|R5cyM|x&z<z
zDHozy=|W_NZyui|=LWt>n!y|@_$CSOjR$<w8ot>M-^_t;a*|(sX>*Xu3@7)U-tfBh
zg0whqkT!n^)J`(yPr^5g$>+EP-|VgssI_>D?~+%xjBLLQ@J;QS0h$Qkq(Q$a0pB#5
zK(3#i*$n0B4gUq#kYT@Og`fKNrGKiOpMvQB_>4dE`)cy%dNbFt4Re_b(i4Q2^TQ4w
z1q}0HhAdemS;#xVhZ2CN;!Pku!tUNm{lQBYsll|JNycal-ihYuIP7-|d|;2sUUdpO
zc_Mqr<!R8**`LQHu^(T7_B@6@jaCk|{DW3aO=IE?ySyjC3*qeLve0MIhijgvO>=AG
zBhF^i+6PwczF^g;c&h^2S!K3Thj&^vcbrv?$6NI|*s6>VEgDe4q8iDW2eXNN6~4cI
z1I_9|e`aQKM7vO@YMPn(Fl{ZGzTKiz6{$^;w_XMvd3I~FZlO2tM`O;r1Ko$L^^wgi
zYSQ1LRQ%ljvMLn}RTifDf<Cb+$VOj;O?OyJ7jR5C3g?`&X&$UIhP5`hEga+tNBrXl
z80T^phX%oNNyAysCy@!VlAhNW?DxItm-l2gg%>(eFTBGm(Ys)p;0yGYu{O2dhR<>b
z>(UvA_H|<KTo2ubee=7ncHOnGj#R=Y!^aDP>^y#$hg#WW<{M{$*Jjs&E2ueT<#DcI
zO_}-CM(&zT>zdkCyCdt8H$C=tn^N1X%CgU*lNQ*9+EdweRt>Job2@|?9`(LmbMP+t
zSyVQZ{<E$YeMG;>nA55*wXDn{vZ_NTtNf?`_kFXItjY=B3{S@E*TAO5@Xf>V)UjCC
zJmIjkFxiVccKO2@31|j?VU;q4$x*3^mlaK63FpM8K)9hBHIyjy5!T%Xtg*Kzqdg6`
zD^nysB7eL3p%K(*ZC4U3Y~k-}4qv`LZBq^Osx~m)b&i9lY)X8}`U!8|ECHLgwCj`?
zTEj4y&`t&c+RD+~)X?J5)y|WpOnoYO6>nu3jmM}ynR)0U85Yqi1Sed}2j94(tMNTg
zHsK2yMXhiMzNC?`E3fAy^ro)s9on{^x)}`L1iop_>wDuHx7pB*ilbk8z%~BV9%FgE
z2l4rMw4X%!)3%}|@tSu&0OO+39OXy>&$w|^<^D>1&C(^*b>W$<%c%8nTX`{TG#T~^
zqQ+;ZkE#)jj5am2C=A8#tNj!2%pEfKwo@nTfeyl6;-9ak4s7#)8dT5x@J>m5<~;wc
zQ#;5hAz$hdb;}#{IkDF;t;a*dQH~=k@6Ey7cNe`Z**7#6^pzl9mn_tz7M-`NC;z{6
z9^W^J{R!v%<Fa<N34C+QZ1Nbw`PU8|7XN?Q5Z*6V-Z#^zG4ncQXJ2t9y;b+iGtcRX
zS%s&wC&9Cu8hxZ9Uf$Pl4edlXN&W~O<(r}BPw)tDr<aME(H*pvauEizRbUR7q-h!W
zBnNpmsrdh&Ch3DoKc1O160K&_brUrR6LSMivT<a+YLW-{mw91QtFI<Sry&mqHagDF
zcu${IWP3h$8cNsKkT12bA&K~r_n{SCL*MxV_vA3EMLGJy8&RL?KtGkQS<@rP`0H#|
z9W<O<Tyxy0Q$2ujOfXLGac1R!Yuq0GH$Q3#Y||gMse*QM1h#38cC!JtaYMW5j&^en
zwkdK6uDL+p_Bk{kYExm<rb@y$XQ@qfIzmnje3L|NYUW;8X*WIDJMrOAn+k<*3c@#T
zu*+uH<peB~mg6N{lLW_9;;?b7<MY-Wv8(t$*yi#=>SVKEmvPjqMxieaW<F+jG@Tf9
zqEH@#FTNZr{+hP*uQ!8l>hZg)!Bgd6pkjPYK72gc(U#JqAHg>f@QnfA{Cpk@%c09$
zL#u&r(!w{z;F~q@P0*@X?VcH{?DS3*i;LAUvTs(?7v8G@w-sYG3%)4<-{gXCro5sb
z>QaoZGIuHieB(!Nc-EoJ)a)6fC?}p{yf;J9ZbA#iXb`>O3*eiZcU?-vTWp=dj0X6|
zL9X9Z4`xzTcWDE?Q^QlbwB#OrQm3Qk0pGaKj8^E7Xa(cHY11xRli(Y-?937;yZ9_~
zHODN9BF`#H-JMaI+9pa%$@TM28>PW7Befpw#!l~)JA895G*YR^y%}GL8Jl?`^_pD2
z3FH<BUv?_{cIHYgV4h}wr_zNw^^wftwN>d2&x@Axhg{;P5xRaZLg%(eXvN|P?HwJV
zt6lk=-l?=r(Q-;f=wfEFi@(zwbt_!SkCNZFDqOoKgzINddZx(sJ4f$Sj&k8DlOOE{
z|4lFWW(R!p@N}5Q?F>_1_{IU>#4~^DfnS(D)}ybgSeQQ3KXnIx@wd&PI*5KV*bm;J
zcgkKMRD(W}dxL&c3BF0q%<u+7L-c`cza;p^N_}ci8v3VR1Zy;Wa}d562H*51+wU*6
zsi|b&JT1lN@XZ_g!_!=0HV*lIP4VAMgKz593DOp~Ao)BCq@N&Aj!A*c#0yjz_@*Dd
zQ%i4?#kwp&f^T|IpYmg-&FC}%`h$LRYBP*8iEOd9%o{G_uM9u^l#RKX>*xnL*$Z!E
zYvvPY<MxKHax5WttQY+#^xihk=8I>R*+Of`x9dTlP%}K8WVSt{FQ|2YyqX=zn9WBv
z?Q42r4tOceH1hoLTP&v6bq_w<RWHa=^&wvay?lOQJP2HGyIp1$CTm0dMD~oVXFJim
zn|-9d!oK<!Yu|bFheBK*tFvAeaM@I*jZM9ZQahq{a(Xnn4eRgTHq=5ISyd44%m^Q=
z@^`grUKex&?mtU?YE(LlnkJg{*=c6Jj+uTyG;R3h1bJI$uA6lT9VaOxUl)nS!?pBW
zE3&s*m^F`URo9l(u-JbsxW>MV*)xM%qEpPZDD;m-?;Bfna<NrKZnMXvjuT7GtSGe?
zZ*q6Mw%HWUKD7*b!P2WXy+ktzf^Vksw<KdduH4J6oYan{@HcKaWY>oa==e`LcA>*O
zz(doJIv)O%R@4;=@LW|r$C|&F^>dX&-xBeDayxCYLlt3ySoYh0(eX#HPM@XTRO1zT
z(H5I_55q&#fqgDDxLEY91hlD{g{*oFcNA@Hl^^%7KaW3PDSOVf>@lg^UCv}tSXQ|0
z9Uq5Vlu#ZWte{mnE?Z=tZPD<W?48S6bfK<A6D$_lm>c+=8r7Dk7FEKBQ<|@D58veJ
z#_KSR*9R{0I%!oW_-0fZn+{jE=@;6`@4EDD!C&X#*ZeD}SMc}GdWFA=n#afV<i()}
zxcIvkmO&F?%}tL!<(myZX$lyRHP`DHHH#_up5c-=@M0PGrYvmd(~7!WcJfs|z)}xv
z`h;c`wVUUE1G)`ILq4y3o7d%|O@F9`bc5^0n$ZgTaKyu(pX^!=n-rQx&+$T71dVLn
zJL;tcJXNwhyg3VI;W-cEF+AWrjeupc@>;ZmWlr{jHT$6}!K<w}pX<-VAItychef*|
z<aN3PH}Kj9eIfrKGqpGPreRC86<_o@^p@NM=oKGL&K~upZq%B>(Qx|0GWB-AKIk+B
z(P^r0qArDQG<qI-%mlnS)U*bUCC3--=S&y&3xW8|t@PG4hi|HTs#IxD-G;;drsHcq
zkaNlVZOc@+(E>Ntq7S<u^(c5|+c!MGdwD&0O)8nlM2fYm#AeQ?1$I4|z-tm?SJz;4
zn<(y^O6{2YCvQblJwcXI68`kVFy>CX>hm*9+@@N?{k*n5ynp*O;5CLd8z#cn)R)_N
z@Lu?B)!%zoW<S_8CB02KFIx4ep;h0KEGocx_mOjR*;(>p&^ioypq}N27m~5pz$g3$
z-*D|4hI*pgd``k&gO<_*@9-wP!;3s&nYxC&@DC5CF0~D&xq&W|{g_F{9+P%#H|aJX
zH#ut39k}M(X_HcM<l^Hb>Qq^%QAOdwnMOUUc1}aP@$9_ByYs5Gq3fY!{^92-IoVJk
z9^+1Ua9Uk9n6HdJl*ue-3A6T5qguwk=(L0DMF?KM7_+vZ;cVa<(s2m-%?NHso0%!b
z$7nX=I0nNuePEjbuuU_xn>(<LH`>hs*rpKLO;5C&^RP`xw3`^ToB3zyUpkGxauQDr
zwW-qZ&4t7Gk`L1B4d0}NZ{DEYyx2iD4YjFPTkykBdm4p4lNG*sh-PyRj+p|}_`@=z
zSHeylp?rRm&z~mZm05&uW-h+t>0~93B~xbv3<TfI>B9F%aRkFd-msUM-`R#*R1>oN
z>c-*a#Y<E!j@h$lN$|}h_~ubM`oxm+bHB&xDcQv<;hV<r%})5H5`1%<{_rdNVl@E&
zO&okvAHMlU?^JhJtkU=~#{s_C4Bsq)Z)U(Zx9AN|my%gjc#GrUn;%EWFWyRKaRRx1
zWEP*JH+(@vj7q^b`Qe)y@J$-{<~V$l;v5-%tI=ylGlL<<rJY_bxxqKVm0cQ}$ECa}
zTzU-O#KJdw;hQJ$%^LKZ!hzAs1>bC_5>2*Tw0?bwk_m6|o)u9F>K~;<a&L~bjM9{%
zQF2esY=-CPG&>_T2EO?^E>c0kk=o=DDJQu%^~o(x^#|YZ3#V>haFUzp)Vq0THhrDS
z6y&704<8OQ8EWKaCe^P9&ff^FI}@SNWES6D7@^7|n5o$%Le=S=I@vHnC&<02l_^4p
zztATI-&{S!EX`%%Y7gIx=@TxWfN<@s3)_?q*POg$7^e?cWB6v+oiLS23RB8$VJf|d
znN!T4^63<&n&cO6ug~0Y@{6;5XLk6NP~}@por>P6?v7CPtWICJTd1l~pIV20bCurk
ztaCY-t2sP`&)bBkS=A6FgKyS74^{#A#%~L=is2hCvi(xPH@g}$gAcxG0N=#JH$TWP
zcH0u94fIbfapEbi6{J68`_(2#wKFqw`cGp{4s*kERtnTPdZ&t9q!&CfKt17`N&x|?
z2;baH%RDOhX4WP$>fxJa9{%b?KkLo!ezKhQlkXDxT6_5ELvuf6&E%)L*U0Zkz<<-7
zTo*E3JTlWSa?3|K$Q}8Y?YF8S`E<YOg`r=1QZIU@$-m3Re)r2WJhOY4=Q|M%#|Muw
zz0I%jmrU$IHXNGya@L7xwDRU}S<6qd&qcR5Y-RnbiY9?B-W%<v=q{LMBG*t@p#2{k
zGiA^~{B26c{=ZL7^7yEGbPhvfK)1<^W^f1HrQkP<_7_F738WsvwbDpqRn@Whiyzb5
zz%`|YH+kx9(QfE>FmWA<?qpU;dN#vm(Z9*HCk@8~Km31a%rEc?F0W_S{Dx+Y>-F!t
z%6{vwS<_nJRi>780sZDpJ*zBB(IvP}Tg&4sfp5O_wdon&m5Q722<<0h5I=7izL$QF
zx&Ms~&l}nUf76qmtS9hI^O<%fZlo^6`aYF4b4fM)p4VC1E6_((3Eu?u4_9{PJ0HXI
zxznLFuuZn5^zkmErZR_GM-b~(4xUHYU}IJM5x@D}aNr%MO<M|CbO}#m2mT&6ayt5L
zv?$LAi;5jFYa816__FLHv+?mK_M+@VpRsmSL*E~am-1IldP$}mDzbr(51EyH5Z)z|
zMNgO+7(_kqE}o^8XkIR6B7cSp&G#(&k=)9j*{X)TCRP}88roJr_~!X(tJ=Nd{)}i^
z*LjYz*%gY%^q0e~T095Y7Ngz3vrF#b9iraRKN&s?w5=C#P47~$2A<VS1(=n}xe=Ef
zZG>9MQuxLRj|}dM=LWuU`r$!`Qx3MUYiCya*wLg+_s|E9Q}3YGv2-5n$uWN>o}bk=
z4d%XNXl$_+Va(R7&s|x+`B`rU+I6oEUavBEOyHjr1|D~6r1jydJ=7=PT(hg;4Rj-P
zi$Ax>$mVtkk14n^-V<tU4PvPE^hRrfd%ULLd4_uu;ha<GldHGl&$&hpFz5Y~)cBdH
z&sJ-K_s51d7fQ_ut!^KlnSY+Ex9BsUV4T_1o!mFzgXZ&7=y)|?n@#96AE(i~hSqa!
zB)#8o&5Ce5I)2Q~K)31N-V=41J}R^uuVQFRIXRb7@Vn4TyW#~hjfRz5)5Bc>{SV!x
zIOm-ey~V`eK9SdCym&pTb8eQTwp7rrt3}XfIJagrhihP;V(6*1p=dY@(N|%hKfBQh
zx7pPPrU_Y&$7~9pcjo(eJwEa}U*h>5-rJ^cUC`3d55KsPOGWK-^hD~%)HBEDwyI<X
zs~E<q57a-)a9;IdhSTEx_;y3h8d%q?t9V=veKoXTjiDoGC*xs@ce~JO_Rz;PADw0h
zxqRp=-^sOkh34W7o3u!7$c=i`0DQ#z51G_>he>-kn-s9or0p9_YQr(@h)IKCo#<00
zok!1!zi(1?JUYGppy}W-o<c8GVl{fM__^2cDG%$4W<-zpr^SYzq8p_<Z)hz2Su4?&
z+M_Y$tzlLSd!sMZs2&EekBUSiCJ*NX8qVo~W<?F-8iE(6FxR6J_nFNA*W7|@;yA{_
zHhyR~zhRs1H>g>`HYL$+dZXQ3gl#HafQ8U*4x-&WJ;jkk9}~5yAxG&khi{%9h|`jN
zFb}mUD}2)yzA@mNOz_Q*jr2RgHooXK-(i~sSf(B<a~GC5wSvdN(FM)Mwv>8R0y@nC
z9?Kl`mnmpNqv4$4<ns(7$2g8#C+ri*<MxcxaT8u15BkU(qvO=!aaQ4RmPJ!4iZ+xN
z9S6R#r$ax2Z@R-bqv4xC_$C#6GZ4OcaTXsAd}D`io@|ZPQsy%hhi}e~z-LT-YE>Y8
z;2wB%8qq&q0dH|}W{&4zwq}Zdb*>n7fp0b(ijiwGxqk4?*1<8_4c|N``(|OQ7?rI^
zZ#aC@>$6MFM=sUhPhE;!zYzMut?-QrzWLkArGE5=--2%*B*S;iT+Oq`$+%fduhevA
zZ1#_q3%>bDe|V9y(Td4R&p5qPgKwhQEF=G>7a7L>QCil39`VBTPvI>d_astncfm67
z&0%`OXZlC#ZnH>bEzN!8`VIN%l!M&jyJwxcNv_|x8BUq#n;Py*k9br1#p#>sLSK05
z9}yaPFG3HK@Y}45P{RfEhYyX=O0sXJStDev2j3Kr(7g;1`u#Cnm2QOV?VfO*T!P;w
zK3u)wo2((!s_2~xErCCW?BaRUr(Qn}L%$DGha=Rd=nG%ohxwWzcy1bm>5m(EIBEI#
zOQ=qgU7UUk{+pTfhxZMYn+4xZd1h^9!Gl9@c$+IB^Z<ou;+zn<`k~jrH+$fl>hMjL
z)FJwbe$x=Xc?93ghHvKd4AwDk`otBitnf`X_+}J*vjo05LH^C3i9yVcW~LwYsV&)q
z)PmmdrQ4XzFqK(>2D!!XP2nuC&V>Nx1_scV5}^Jzvf$yHS=pGiNsh;4_$E8`siw>g
zFI&uCbC}K0;f$Y>FDBEYo1gYJ^ixbKKXp9otAg{%5Q$~hKuupwA|uxKB6A29`)HWU
zM}_GP?|#)=ix!ie-^p7ZO~~sf_r~v$m*VKL^%=&z9GjQA;#sIQhrJ}%@Uo3P)r@Oy
zaB}iD*c;cO)>1Z(`a?T<Qt_~S`UBJ4B#$MLHHSTQHEQ<z=i2m{>!ZW4sbod!3e-C;
z&Zh?Ag^#$YRpat;%f9c`HR>jK4!3-!u7bv$0==P13X2@@Nf6hWOt4K(_JPin<YD~5
zf5<f{V?nf=Ht615i`rC2vw(qIP0(x_nAtCwwI0pc)(#y9T{+WU_~)EiQ?gm`O;}W$
z-<jxP)tUoVWiLrhswKVxYG~zHLqDxyP5rliQlol@hEoP@e&-#V+|#js=cBgbi*_*9
zt{*d~3$f>22;Zz>t@LWdI=Y%Qoq9qcbgi8*OeWSz$8P3Sus&W(pbmwuuy`?=2KA6Z
zzgg=aQ{Q2&O=2Bd8BbjXe{Z427Wv&on}KWI9^#t(m;8>`WPRksOU>VMxV2f{?);q@
z@C~5(k76G>c{Tf83qFRYCe_Jq=x<6q2i*+~g7t<LGiy|3rj!5kX#eC|4m$<aw&;${
zqL)4vIMkvmKhb`2SvB<)>r-a>tcsIG5{tfd8SUsTdK;S7Z)$aI;g5Q(d+%)Y{C9`T
zCgZ(c&$N+Ctc|Z(JJAx3r^SO!&Ezk~@?x+K8pA}G#{x^l@b`}5yqF62`oFND*W)GP
zxgNrE-fy{0Uq9Pqrv^10-7G!Ne=GDfgXbmuCVgILN)FDO#?7soRg<2t%IHgI3s;NL
z>(+%no<RIZtmQ@A(ZLF^MzhY}=|Fw76SWz1i<n(@jX6km_9;BK=g}Fj@_)Q;=6B?1
zr-vQN;TbY9&tf1N4;--=E$4ka-ZGeU1h40VGx&>NGmju8`VIcxOjW5lq2W{sX6{-H
zKAs`?%4g9(2EzowG;LN<vw~}qs6m}cz)#FEa6WxhlhAL_ZGr}oU(<=N=e>1<n&w4=
zUaHof`c@BLGTO~ybf<k;$o2b0kJ=OZqn2~t1Uq!RHh!hNaM%yH5MKMa3TAV`iLhR}
zn(!Qa_clM8U?vzeCp9NNzE_-DaS6`VT4<0kPOow3GO*v84KUAYJX8ZX|7yVRh3u?n
z^g;1jSD9?n+c@4EVR)ZPlgXLcrrzhRN{fcrvIbm3euF1<VaGy?7UAXU-N2#^oMQ)$
z!!&qQR-l8FugZHii&>@587jX7U+)6WOAa&ay^GH$qNf~(JKoqhf9ZD$N4uGq*--Xh
zCM7&JsW#lQVV6lx*k;c<ld7&Y>Dn5TI&ph`3;k1jO>&2Eoaap12n(&HMwK5A&W;R*
z+7&e9szioybGXXa(4kIb1dc)5;rD;sOuq3ce(z&?t&*Gdk!wqVD)dot{a9pzV|>ZJ
z2`7gqj(MEa#V7VhgJKU=9u4OqjPn_$>B#XLt{KR&47M4JcH<7;Our7Vz%~`oZhE8L
zT!L-toQFHmZeF3?xWhNq;hTKNVH9dp)2U6hhi@{%H&=JV8$0piY==dtO$Ab$8VcVu
zhHn<4-DHGs`uvY=j1_pB;hK~jrD2<@OW>!)=sEMrJf4bQXIz}V&^Og}7~eaP8RId`
zS_mh<FMxRv4t{?}G$(qe{{5R*8~vv;S%77Dtnf`q_$F^Q`nBMjHOZLi3E$L$Z;HV;
zXYVmb3cgtc-(-Vt*6pRQ3ci^N-+Wj=j^AjqZF)1i!5OO`?U^YK-}J2n^Uyn0Ggqu4
z;hWz0ZzjVxf2d7eSQ?{W%nZK`+XP0(s2^;T0_`Seo*1QPF2jX)F15J@-|R%AneS2r
zd=m!WxIFP2!#Cfml8FP~y!#uiYY&+j4&O{gzbQNgpK+gPO@?nyG{uj@?5RU!-=v0b
z+Q2s}@!!<!5vAXjDCMZb-0%YQQhmi|!_28QoA4G-C#$$mq+)%kNj0KpioWo2<lZEH
zW46>IrzW0usseoDG2N*#J)FAc>D1wdP6d>vpDL46)4xV&`t1lcI?7BbdZ&uah){ZZ
z!|&2N<?29gaUF7S3Ne!^Ej6q6<P={HSCgILT9Xj2Mexne-r;ih4p;Fi)T`(X&zmD$
z^O?=?_Cc5$;k_xgIZWxsg^>jU<5Ue(-%??E2Hyn1H;G?E@yn5Qvy+VCS)nRIcJXUp
zw3`a_P-P0075!#7**6jNPsPGF&-;gHc36mJz&AImh3HcH5c%Q18Hm5Q8hm5HUtA6U
z%?IWTH*FNGeI<kSEn~30!Z!u~`HRWFnKB_rlgWM$XEsBsY(d&{9ld62pb9ek_?roj
zF?=&CYoJb@56}_nQ)jyc=&6-Hs_FrHmpwo}@9CF1!MvS#JjN#GDHkSp^^+enT=5kz
z@>5Q-$#TLsroZGCCoyASmai^2ebtEU;>5qqNKB$GHI11+!9L`s;x9hut(WugU%I?C
ztr30G*~n`n-z8|Rm(ItL$J^9P^-r-U9gQ~<Ux{ZeJd<wZ!my9+gdQ<q0==f;?9163
zXDf?t@fBTzb*^83^l|Dr@7I&x$J&$?O`>T!^p7W24cvpzhWb+iIzdU+sD_&@n!ky)
zcPkzmyspjBR*d7!@-1#r_tj{|c53*Q$=b@wOqc{T3HmyNpBsAg4c#Im^Y2)@4_0G7
zURmlih0$|Z?~}M@WqL<H#&2%fKPGZb8ajvmhZSU6pyiCYXVzUfXW9k!n-?wG+R>`3
zte@*0tgTbf4%o8}!Oxohh)wNJ+cfkFb0g3$s+~uNVy#%l`u@j&-&jw745n{nE$b+2
zdug<aobX;L_WPqb_Ripzb#gb_z{l#;H;S`1q1UuvO)YW7jvh&!<T6^9KlK^b`br7-
zhZ2}!T*adDiE!K|lhTee>Ee1r#y<AD+)mC8*VM;Pgr>fZzt0`+iN0iLD*MyX_$z0i
z)t8Ak>EvOP+E+!>xo@a@KAxYB%-OjO-#w>CB$tI*&KAX6EL!bJZ<^Jj&eZPQ-7ISJ
z09_2Wy77_Bk91b`d0|!Ve)NjNH$`E*vZ>$$G?K?wSg*5PuUIGFEyeG<kNVU_v@KYn
z5qinLjKVVT&Ay`eUErJFc#LAvb}H9LLxGX*;Nuvv-Kz5wtjwE$tEyP(`N1cY7GHEC
z^*7Fwav!aV#HZTrGmqmEUxQccxy`D8hODoZ@I#lh>Qh;(s_<G>NI~BlEHkAty;p@<
zvvbkwR)v17cC6d6tmh-qb(Z~?H@RUyb*v+37%)zbyZjBW@ZqL(Xey7@g)VZkE^{VZ
z<1O~0{|F8}IvZc~di0ai_`+Y30gRr~le$qH8r_v9c<n9JpQ3qRP@}qyF4J=c{oZIb
zlVP0M^waf#Ya$kqkwabT@NB#|)TYv*^>pq-7Pt#vaR4*O(1G@KAmbP9CM%xeg1MQ^
z1k*G<!sFpw$~~6%S|@UU;jc%Pcnn#nPrb!6!{5-B_gbB5e4Gcr4Q#iMzx_fQ-isIT
zDlNbV3^x`%VAC+pUk6;dg0J1fbFyhR{cr3=8c}bYh89){-OTF-8JUNvBjc%w?`%^8
z8}-4w<bzY6?2OiVdO5F&8@&L@&~E}P3ZN!Bk$p%vw84fi%__n^VkT@;j`~hgB|I(E
zqK2W@xTD+bpJk}QRC=POkv)vQ_+$bM(#ufh5c;^=8roP9y~T}rQt3EvKbaH>+vI|2
zk}WbRaj8kgR+{uK(S){blHV$m-Yhd|B8(Hf&!qWDCJn)d^8h9q{Mn>hcyONK!D(5}
zP$m34Cp@W##n2-@0&mY8`owuJ6+cFw_+9$9cppXQ;qg}B+CW|VUVGTai-W&miVOX?
z8?!e1nw4%a+RbpXi{YDBFwHQIG8~Ixn=`P@df3K{cC-5$+yUEEMZ4*Vc5@lFX?hO-
z4BE}Vc~jvh=}9_)CPHmW@Xa@BQycc9wd_Tgfp2odH>u#8?`SudsZAwpfS=akje%_n
zq1{+l!cwr!Hn^q;T=Nsn<_1S;*d`XPDLsiEt$4Bo$1saw1Rmj`{Qf@7sEVdXJQQ8X
zkFT-uINGD(wBT=OgpO2;+FB**Ri*JE7vXQq1M_61w<;}K5`44u8`*sCV)gnNIt_f2
z3%;?!Hw($X$phb%+rkV6>Qf`&o2uieOZDR8=>KEstmC3e+%9fmyLNYXcOHYGCMl&G
zh9T|lR<IROG1l(x?rz;(ySuwxe_!8y|F|DOQkc2-d(Sz~d6M+WhpbNfByERpdc!vh
z;hWm<O(yszE4{_p*Rz9R0W(qq@uS$uaLyx9i(4dWdeuaE7E06^di`cROHl2<5_Dz{
z^WpRsyRcU?hWyR0E@(Cey8YmrWhD}n^WS(azZb7tN135oMbF<9vc&!3HO-%0RP5HY
zRN~)q(2c`v_*VR>t~2Sr=@F-OR%XK+$7u|E8JfIz$a=}4&<zd+OlKAxzM1RK+sz%S
z!rauuJPr+}_onyVSY`!cxsQ$2h6%AMOsC)9%ucm#6|2jYVl|eX;j`JPS>U=|<qq1l
zZ<SqvGsqD$8(t8;d0;}bX+}1uv|XQ_$l<(>QS7A{{kuIzEtm}tW%rbK&lpwd2<Oy|
zQMvLlI`{|w_A^?=;hT=<@u0S%^ro;woZjL~4cHZ4H5z?2TEXzm8Te+|(<mL;!F)LS
zO#}EQ1iop7eiM)<N|ER{7s%fnAb*oGn?67I<~I48Lh#L6^qV5NBUOO?;f>)NmreBi
z!8aENMX)!NZojq>>==yDOZcWb`Qnl2HwRZRCq9<le1YNWQ9oS!^M$L^KVhm5-`vKZ
z+J}B)fp0dW-)x6(rolIldf-XHHw)pL*@f6e1>bBs9in|RLsZ<4K32NqyT1)qS@Ji>
z=df3^Rj|%88{X;z`<D*~Y1WJ&S)$mhSwBerKiS>P-l?FefhrOfDAz)C?w<_6qYThn
zdw>oyNBcG_dl`1)Ma`ri?G^W$C&`X*jy{ezwS{|1C(gO|-Z2}4FZGgp+7}+o#&9p$
zraW^=`Pn1Tk9%RRd(-e)8o~u{{-mpgOv}>J+?QYDdbA3yWQ>nI@TTI)++^YU)Pj4z
z?c`k_#WSzG${Qbo{w(g3N^(Bw9_FPCyr^1S(`)>oU%eai5R>U7<$7L?*@w1p&QPxN
zzvnwgmF1qRjF<L2@nnX>Qv>gDz5hTrI~vDsbn-uDdFlBQ^m1~_0}s>Nb;(Op$sDJ0
ztsjhjk)A@|&RrkfOu<iE$sG0$UyUP!a}|$jBN|SQ%k0DA8ZqgFua4uLjOG96k8W_^
zg*-Kk)0yk~@883+Xs(m!Gog4J2heV23@86#CU4t-9C0x+xWCA6UZeN*JbDAy*<|j$
zm!pLYCU5d{sE_UsV7}=m_lo0eiZa^t!P_dQ99A{DX;HUv^n%2DYCsNVU6*^Rz$dN|
zuv`G=?Qi53x~JJx#>D+=l1-y$Fk8a;zY^#Ej2vVj&~XAcdTG-MbS*O2BVid|bevCc
z<P>r~pPSGF4f|{-({qF0)A#A<L|mr^*?km^U*@ycS1GrdNrNMXb3Hud&HX>u$Zw;V
zAI^YxR(96jQ!vggw3MslQP<*k)bmyb*NarHA!*@U+t8Pqx{>#+>Z#7v$WS!LZ}8@x
zyeIeJrMXY8#_TQk&*^r$WDCK-Fz0~}J{s|aIdycl!{nZedNCu{g1%}xeD=aajY|8d
zWkosxZu@A(RUf@N%$)l@`r=@*3?6?T<e*={p=I%8JQkoEttW%M18)o-D|ooGuA$)^
zeL~I|ZON6t-w!ywL<{m{u4Kmjx{!;9-{I6&YtZJ-;ZHq8$N9y4QVIHNYSWR^maG-J
zOI-(^(*Wk<@SHZLqR+rN&nA;q!;AVoN7n-lv>R+=NaKEi_eY}j{Ees84lWuJ$PASi
zpX1CtS8L|Q^TI=)cz&Ug&OU(lhwkS+idnl@xQ3tWR5dbOFysaNwjXFRFVJW#U&;C2
z<T-~H=7r{U7ET`vqts4>Yv9wCJg!e{L@VQOxD734x(7OCFgYjm)qWq)Dxdi39hyxZ
z9`ikBGNZ=xw!NLdyOHNY7P<jon-Uq!Ho~D#2YKt#4|Ey0ChaSklsEV~uf3FlKKgE{
zmwNT*`tQwSiN|N3^PY5hdFojPn&JdHg~#9trNZj?L@jtcpH1NL?8WoC4IhhkGqeD+
zPt15*&|&(Lt=YcJCe3Cxd^)T%%ck^c%t>*anSx(6*QVR6ZSp3UGYp>31KVF?c6{?2
z-p1bwF3i19HM}V@#%X+yg>htZhIy*xOgt;}&eQ{FM^`-6>Z_-27eSjr?>*8Qwnr<@
zjP~?5=bt+c&J%c45Bu@|!J`_AM>T@`rLQo}7LGuU%dpMwuAAquO$yr0bF`Zh@J(a1
zo33a#7hxOYIdU_wjTP-?1KQ1l<8=5Pg<bHb0^pmX@Xe(?Xgs_49K5OEZOkvhH?82C
zlJHFd_@)wk(*W(J0lrkFWqb~7Qwy%y`6qdu47zY;!Y6oAk0#M~jCM0>EV>QaO+-Js
z{<@LFiK8nplJ6Oqtoq*QGDdzLXXe1#z&cIgr@F9DHQry2e=C-(3h+%beDn5Kl49VS
zP4LZ8_+~tO^WvW*Il(to;hWie*b}}vNe`DZ2R;u5f^RnTq4x&9iRzrBzg*CF8vLi<
zM9_VcClg=eQKGuAbE+1d#fRwhd)PBkH$oEihcQue;G6F7O>y|f3g6^=&c5)!6Vz}o
zY_ljqUB@J-CG+7n-vkwG$37}%!ynOI9P%+<H*UtO??Ey+%j1<YAzrz=<5T&vFT53e
zQ-Pc@d}D)eHqm`EYC)XJvX|kEJDybSIQiv|Q>Ql$T{!Ddg|!ZSpURAQKZkyiEzZ{1
zp^fDo>H*(uqxYux?N}W>5v!FeV%2_FtS*Mestr3C#xXlJyL_xF=ZIB5x^E6M8}3c_
z&7c)_S*F?PR6@UDHheF9^Q*C4>81FXEOzaA7NeeLW8?<kl$nb+H4c{P9V1U>r`FYu
zQ6jV98}r2Ik8kYH#Gk5we$#tLv~07Y)ju&>2iiufVI5|~nVU)?cQX=i>NVO;lR3;w
zu~#!S>UVNBO7+>x&>FtU^C41S==K}Ek#4{ANTm;sR9X6qJ9PYSekud~=Gc=6-NT<+
zO}=>CEcR={H}{zjkM0z~+*gGDX6NwHC+yHX8m@VNqTLJ&SBcK-DsI9qs?6bJ&%!i-
zoyM==n^C@DdQ&w_S3ZYo>giA|NM}c2uTUMwpL&5m<ytsYFW-b{<0(3D(%7%b{M0`D
zDbt%^g}^tFbl-Gu5iF;2!7A~Vom2F^?wihDpK$o5ZjhFL$D5)L|Hi~Xy$GWJmD#Cq
zdW$bj3s9SI_G{J+Acw>bp%&~KD2zvSm;OG^$0f=#zwv>5%{P1o?pycXr4K%dz74Jk
z$wxaYZ86=r6UZ*w$ceVXpQ^y|oc!lLUpY9(eyPX3ANh?H2hm98)2T9mneZ+?x?YF8
z>_fEd@!m=?d8=VV7-t5ZTI2Cv2BHfm(nn$9CCt+8C0;tZjJ*>*?Xlz$EuQLf(o>@^
zF&}|9^ZA*l_Tc4&-{78%++rH+^WrLA4UBUFKdaL3`!LQ;9?WxGSc8v#fE~hI<F_!E
zwWcuo5!d0LcIJuB`zWR$b3t>Mr-E-14x-td#H-@vVKN^pnBj>+)5*iN^W%5=mf@Q&
z2IfrAtCr0|zuATsqb{;rxdt)cQagff67;PsT$6I*edH;C*YOo^1kI+%ahQhdYy#KZ
z{1#Y;>(2dPAKB}A>&PUV#(LOvwUJdF-da@fu|+$NTGVsAMX4<<D$6~pN_%QYB6))y
zRy8-7v@z7A_Q7WD9d1?MmNtEjvuVg6n~HJoyDt;{W^2e(2BC++Eidp7eikKfT?VG%
zzWW~9%?321G7mY&!+viU;^n2{_w`0I<0!w%OUKZDE*0|T-0v;7F?bO-eN<<#uS%Gi
zDMLSK+K^c)vdriDa8D0UUg(5I_JF^~4z6uQZ1ja#c}!Uq$;+e<R=v+{Lw_MV3d6a=
zH{;jaH0vw#zoBSze(+ZYJXXV7_xXNj(3xh?-&i%)M>{P(Du!k=tOOa?yl8+NzY5S>
zUBO3dS27R3k>BASA9f7G(C5&H&>$YK;qQfib!9S)GnbqW?6qV!8V^T{{rFkrf%=?5
zlRJkmgC<g&ELqWxWarV{(t9zpG?~9C8p`<d<csgolk}6EI6CI<oK;V9<Xdd;i``Ej
z;T8Kxm<PS{4gOOCI!)E7%znc(FVS?KPoY-_u1Ork96Kx&N%kfMF7ooHtIx=D3oR%w
z-1H(FS^oFrqp#ull1uM0pZka*uyPnLd3+5n2a7$%7gT4RDoL(7D|*cpIvh{ai@6Tn
zaTYAvkB)>8_$0trL(DJ?jM{rT&qp{V-zXoQ_aXB}&MNnK{LmvlaxaQ@*^IeHn5Z=U
z0^fV`xqSY=?eLWc;|03-&|Bl9lc(v1c?9?Jc*uVdZ}=+N_8W8$qlq4bYl5PAyuham
z`TO^Mgl2NYQ_f^;DsW%2a41?zU!Jcc;20B(<H2(p-KLy@=X7IsE1<XJMRzg0x9KCC
z(&LB?-PNX+Q+XTSzc<FFr6X->FdBbqJi5+IoBmvE<N6Q(;8XdM!<l~_CVI|e34d$O
zALv4r{_E1oVdWeV!+wWB{5;dpc9+rZw~IXU1?ILM(<%6#e!<LWrZDyocl>Q1FVzoW
zKLr1mm0fu(<CEXyWwvf;IPgs|xaJ8=lg?3u!vpQ6F?@3Y?WPKR)ATRqvM<3F7s#@}
zHr{AAuV9<;XgAs68#8?40^j6<Z+78L_1Q-b2fq1$c5@8vX3JLeAiSxCcvIu?rh2XY
zFGFltNzNGU<{)g70NYf8ZA#6@4@<{)N+Z8KHJN!_W~Wlnkw(Km!;`hAFW(=&@rG}<
zM({lYU?ETTgSyj&(~%CH)^zhUfs5*(|5Q!Z?Xvh`@Xdlh_&V^-Ecm7zd@~%rSqI+?
zgKsXtH=EBg6Lkb$fp5BQ;Pqw9hNqJ)o{*%c<Zo&xk~@ZPwz{%o9KLB-1wX48Ib`@|
z_d~p=?TI=}r{9$^iJAi6^hCdz1mA3dZ~Wn#{RP>%`ELR{01}i6-&Ee0p#Jd93i#$;
zLV{j<C+JvfI{o0AF28p((0$VuzIh4Xd{`E*PpR=5)-7Ih;hW;{&4u#u`U`(*-t#y*
z5aP6#d~p@_hL1DEsd{z1DRzd>ec;fH6AsN=>5yTHL!<k`Hm>+p%uX#S>wufd;m~_y
zy&kJyM`QJt-r}_*V)d~L+2T&@qpHn5DrTqJWa0f!?Yexyt_FMTa-sX?E8REk2HI8F
zZr5P^sof3jdR)S;nZII`{SiIIr(?u0x}4_3sL>eqhNIt%?!fF+4c=ceM%S2~diRML
zDfF9Evc+XKMXM6|n^IwTQthKP4}U7S5cwOjH;(&Js(K(wM)op<kBCzH_$Z|_8=e=w
zaffeA%uh9eZ+dKqRF+whYA_^HvG9#k2lj~9j8u-y%!faYkkeUa#L;g)!Z#X8UorFH
zHQ}4n@XeK+aL>bV?K~7Nr;KpD?}286eiKdp=1KN&`OsawbyJw`O$$?#&S5IV-f-Vn
zp}Ko8R99w&%G4)RrMuuu(SK8+NT`~@H~SBV==|^yy!H@Ht`MT(&x6T;2Fru)oB7Ry
zwX#I8s-fQuJrIO06r@&>^wPsOv0nq(aT2IkbQh<wv!_vR-e%wN%CYbbvs3qvvx72?
zUD%E3^rdSz{4nQR&byO2-zL7IcXk_IOEO)+P0%2+GY3u1X6$zE9l3YE#l3MGu5;s>
zka@trIewRWLp1RY9nd=P7b^Zmr^*%|xy|rV^JMP7J(vf7#Lk7T-rCli4Dn0mxzV=A
zP5IB`cEa!eJ-<+;4LuW?@Kg4B>Ofu2eKnYY;GEc`5p$m0pDpD4RQVwHd#AXLp7UfE
zDjq(^IhaSs__s5hD?fUwufa>fgYmhhkv&<B&a)q#=dPE|{o*`bl6z5@VniEei@dnT
z4=2M*1}GmsNl)@Y+x_rO=F+vw98vAJK8ixCckKuRkq@!rgS>@JrVpSOC!Op326P;*
zKcC1W4S{iv+Q@Vm$z9`3{VI>wkqhp5!`#k!=6=u*a>0fvqsb??V`k_h_tNMR#j~Pe
zS<!9=SQS;?s_-=y`OLE@Fvy~1-7Io#V4=Uttc4Fu8dcAv_1R1+n8l>#c}?0n(xl!i
z%o_FGs`b^mpKWSWt^s5*KiFiS%{?pU?VIDc*T$!Fz&EDS@Ja<V7XI%kT&LP`|NRfY
zn_UaIKjl6(7w7V~cz?47dTQPSn{w|b3jur0;cNPZ;{nvSsXp0;-V2zq#c#+w+*4Jq
zTa|Lzswsb4**#%ZuE$p8eQH&8UO#?|f9GYdH&*#)#b2moQ#%9KsP1SvoA?<o*z}(3
zT1WJqY%Z`YT2&7CC_C4k-duOi@w0tGZ+mdeTUSr~=Vu<v3Nteg_bay#k40wT$aa0-
z%C&o&kNi&3;luTOtrhJJjjRJ4UKvm33CFz|cr$2cMfsZ*L&u4MH&3BI-9;1W`U_o(
z8L;#G4a>vw>3Bm$mom46KXv9dp4)f+4K4Ez_~!Q<_Z?=Q@{$p5(vz;`e)QUmCSMHW
z6hp&lh=yZ<arRI5Q!RLB9+{gI*rt9j_GgpXsTIW>6`!B%%-ncoW~j&$&*na&-&;Cv
z@PW&c-EuGwwxtKNT;%w#)F*p`H|0hacrVYlwt4a0;F=<^caHhYIKd}HlgN^x%``=O
z84t7cdCKhnD}E=#;TX84Fn;WZ4nDFud^B+g_Z{%y;oSe-i;Nmahazm$H_BJ3__a;R
z{@)~%_3@>*+HkG^!L{9UHkv66Jbf#k&MtI~V`Nu&Un`!oNAN)xHzDuUm|YA1u;-h<
zU&kpt5BuQ>**z5!?WwhBc=KVJ%sh4)bVRp7Uujp1eF>kKje<+EFb5uS)g~i3n|Ek6
z=U|)0Bk-W$^~yZ<TJUlToU@z9;)V=#9Q>-`+wilF@_3>TC;bte{Mn||T=-pO(L7<N
zWH{<q1Ud~r$0&TUZi~2Q+JeS&ob$&+c8<UCRLvirD&>S1h31>W(IJPIPUT@fAv<}S
z7M`jBe>e9+4=v=SwlGaaj$3d|ILG{J%rc?fRJ}qj_iy_2&~8eh-FTqg?161Mquu0z
zZ%)ECQFv3wj=>*C$i^Jv<KUZXcvES+(Rp^F*KFrw;hX&MO<wrs8`{lPw3`)ZH=$@Z
z@0T)PzL*&;*rp(C(=Y=q2DXWuL1t%4vffSL<v8{)jAFif2y;=r(UlVMnxdH@59Q<i
z_;(8o=El$30S0Tu>y6>B+R2(y1<wn<nO77q48B>Gos15AbK`Ro9Z^YI0pCo8Z_dCs
zk?_s?!%6D2n_SI$`u_gpeKXN>;G4eiO+WZ%41DA7MpsV5BprrtMx)>C&IS*?VRz=K
zM6};TErf3};F}}x%{BPuC46J~?N7ls4}LNy1>anOZ+5^p3*nm#_~smY!#{f^C;-06
z3g0}=lb}BEjpGJ;r{J4i@XZ$Z#tpui1K+s7H#gv$Z17EQy8R4i$QG|+Z#dn>#`bZl
zT8X)-uMRD^#m=c?<cOD|-Ar_-44uU@g{D)7IjYj+jDNv4FPNLUiU)O&d2sZbJLos*
zKCygUto|W;QzBoi`hT#?o$O6Evc>PW+I4FY{l$~mJ=Nc?<S25+X7*08SF=TNw40yI
zeczAavB7-!mKgmt3)W$G_?m7pda4*@sTQNhWQ*TB#i%*6QyF)nRsTe^yf#KFW=^y=
z^o!PErV`R?z&81rtHPVw4&RhH2+N?~T!wGr+0AgiEiWra=}m5QoVV-^zYwXc@J$H*
z)EM|?9Q`-Ro$#wFGbjF?-J18w5}#s*YGZ`LXGiD`e3Kj+p-=G5KKSNzwg~oshbv<j
zd7G)>$_n3X3kcV>wy;j+aMipMrhr}S3ZERNpIyRq4sYt_%TT%yLX|lqR5#hJxx_D2
zrJ9FoAO2MKyCHJ0GjMpH5WOZ}yuM_JDn27`!*0zBL&y}P-~3`H;qBM#?AafLE)*mM
z;ZfBJ(nqu#>!LtC>lUak%uXe;Q>gzo_A9bWXnmOg^*+e{%<=xRGcQ$yT@0HKan22)
zBZTw#qsL^4nUkuC_i~~J*Y3}F4_pgMaD8|+jTz}aojG4JLsOdW5_FAoXdDN)rUVyZ
zHvry3nd97ZuOye4>VtpBdG(IBj?=yJwu-m*ox|tmp0M&se3PDV4jfWFi9ImfKlQ<L
za)w<#5AjrZMdnAjAN%K>O{Y2cO<%&S=sBAnbMC6Ojr*{<cpiBDgE@X*i=S~#AHn<F
z@34=iIrn1d%trX>8Rz32C%DFQ&p82>_?e%qF?>^#T@SIY<ZsanF1d026xYm1Z)F}#
zM-2Yi1^Q$H@IzK}Z8*mL?`^U&t#_akFkh66H*)Ybop?WeRGGYN_fPb))FpER+tk6s
zTf#M|9J9Rpvyr{IO^+bIyRxuh;rZw|L&?69`TR`w;b(py#pUx<XnxM!p;oQ?Y|)rO
z7W(BaI?~>vl#^y1$Yj>vM@(|`G^tx9lWJu)$@iU+?730j+)T<|%B=bD&xf^q+<mi(
zyV<nhH5neRD<ewb-PI!pF%!<=chN2#A8<YA>iXP^qiv;a_EN)5UfRg}^WYJhI&j_k
z(^GlKFLXQ!(|kZ1I%idpT~;k#X_d<eE1mCF&ZkxtIF647*DO0_rL*0tni=RdqpeCy
zv8vWIt4_?ZYT5xSeHT`(=4(X4KPRtP6}#A~C&w-7yTYRTGc1Z7W6`-WRxN4E%y2k=
z(-iha?;)pwK6f|RON03ti+1o<4*piN$Pn#qPj={yx1PZ{y~({^nCYYE>*?-6#~F&J
zwSdRosZPF{=|wJjI9}isGSIN-@A-6J=D*T-3=cd&J{K)An4@DFIwW3OhkxlV>_oR8
zo>1Iq`fpaT@A?FDOSk9;#yiTLk9jBjsejO1=K0cD1E-uwKzqT<Jx&hiJwDa6@ytL?
zByR)be1&a_!#iobZxIYszXx6w-qgMbn8->F8TGhUC1$}3F;kwIUV%64j6cj=7tgy$
z_{ty7`i#$+T8*6w++SRZrz5E*jPndHiu_T2vcPkP(<$Jf578G5hW~R0kIlBp=vXj}
zqYwEqn5JNBA3dqg;~3p5pgz00VfaD3?=}0|a&M+v;I~i7bKyiF*|cbH-Q7bTjOUC6
zUWtK^?}T}&C)}Gl2#qg;`-T<t;-#R8`t!GeZF;xj-k~BsV@{rLu)}=V_Z@o8qBuOL
z&O9H{QNF@7^Lb2l;&0p%wwc}ljqjLEH^|b=<9Xd3&9TNTI*DPM!e}-xhuM^Gpp9-6
zn7x-x1$em|#;H2iMsFlq=0clRq5s+7@5AJb!|vkCa-JxbjlP`{oHOc^vvDQ!gEnaC
zP2L87tm}MqqxEQohq-UM%{<pDevWU<RAq%}vU#Z@$0E3_4Lmfjtfv;dvZ~ZLE1h9H
zZX1wihG`-=>T~SAM#h)p6>O7&cC+s<@-=8TuVI_2Xg4ch8z0!F+9`HB!8fOmGk*==
zw1aQ%9HavfZ>lYP^K}<ofj1Sk4S#A2-q0qp#2c6^hi?SmykAL{c{#m+OW+=~o01FI
zr!W^rng!dS+w4hUW_&bVf5Yg`8I1ljg0}}T3*HOwDS^3h_~uLqJI_1wJxp}sxRCQ{
zPo}sfnde4)|5{{vD#L4K$W0ePo61Ly2flgvZ<6N2H-c|A!#79ao9FONVfdyNe3O3{
zyt9_saQLP&d{Yv>=?&lb!8eEDo5S$U+{Szk^HcTUo0Oa|(ECIUyUngC@;7HE(tFcC
zQI#WM9803cw!^1tl&EF+Q#;_Beeg|%e-d;YzS%WDLF-Zylm_4Qf^Q1JH}fl@@4z=z
z;F~Yk<JAPdnF!yEg>TBkH{;+N1AKD?zL~`Csl`v}D?W*K1K%v5|E6)nICkm9(dEw`
z&Fc<TVy|YEh4@p<POUXN^r4MIy=pnsp(N~+i5_ES!#Dj+&SrnCJm$n|2ff8_yx^WT
zbmCM-yCHj1kL=BeYcR}VJDxZ?&Ro0Zj<@SzAG@+g+SL<(YG^&X1{Sev+qW3iz7s<q
zC@iy{PQP^Kr0BlsZi&&c<}upO?9?LmhL6k|qa|;m+5HvG+#vfk*G20Jd#9>*k5&ix
z#s=Sfhi_KGH<RF-S%;%!fp5NzjMDVzD0QtLrHAFBG&dJ-zlc=d^O3r<K2jgx8>jw}
z+Q$6U2j@toR*KZHZxIT*i)M2Yj|#rYJ10VA;Tt#nscJ3}Y6jn&bc&#JK3wPb!80?N
zmx6E3`iE<88<?j`xaK?$(=c{x4w}gB@WB7Ni+6p8XAXyI?z~Xl=^v^Q0ijw|KUASv
zLv`#5eK$)()T4KZ9(D@Ry8<D4`-EBWBf%Oqk_?VZu=@WIECaiz-oZCd#|NoTdG--L
zW>#uvpqk=Mb@OI+s(hdh{1t#U9iR;ccJky2kd>M6nelYq(B-!Z?IxA;Z=rJZ({VnZ
zwI0o;C-<(k(deHtm%%mRQde}F=3EDEk&EU0Iy9XeVsBr~=!~Y_62HGHeQabBe_cnj
z+2y0U+*hZL@X@5l-U=h{a)@*2x69}>yU7Tzqicfe?f5w6foi}X=rv~oxxN*^KZjEe
zezEEGX3kwV$;|FVkLLQAZ<tj9%dF_5-1Ga<X~8+Pa2K+P^EoGSE_9k;Q@tvD9bdF)
zG9X{k&NoeBPMmY~FZ>x7_@+91QvkkMQrlY#TG08&{P<C>o5k_1Cd8m$aIcofl{t;&
z%yRT2huhy<H_?4AMwnH(h?$*uCjA;@(kT~{T2(jcZ4DDUs>uMXgeO)o^VE}^uphfF
z$rOJsOvd&k{e9$G8{mf&*vRYX2;V}OX{pNYOf;>M#qsOtDU17I)n$WKS)W^!!_%TQ
zH7qJ})vVZKW_`b9(#2GhQaw!yDrZu*Z$`2cM$Nct)a8dp^}c6Rky}PN7Ms*>rCATh
zo7q{+J@8ALb}?rg!S5@Gm%lg$Y^7J^w{F6<pc4HdMVLF;%(Y;;KRM=kcn4d#HgPRz
zx*N^ExJ~7rTa}I9y?F<^419CCpH;IXtXe)C-dT$_1J~q$U$ShrsswCvk$*4E%Ulz!
zs+Mk5%p$9rz%#36T6KNARaN5Qm~2*MI%ZLi#TK=nY@uhwqKwh}zxmpQxgN%i_Eh>R
zvQ}_vxAy1*Fxj76Q;VV<^#4L{Tpe#2nVWzA14f<9?{ErU)dcjx)$E=}mspQ4`f4HB
zk;3#n_a&zS)6|D;zVfnPPcqWNUu@i2**n6dfy{%Emu|j<$2ERc)7)@9`LaYjs@*Hd
zNgpFa#q%KzzB&2FfBBn#+xe-f7d<#3Xf$!~NjE$ze5zZ+@T<mhe~?OE7{*x!+x*Sj
z7Y35A;r#>nxRrQQzrx7wo9HVhcN1M5&niFjR4?g_#0#4S+jN^ozjGh@weibVmF*%Y
z82QwnWQW2!v#*sMWQ*{hX5&50@=;I-jAtVQ+zGZr3(R34&(ATzotb~Qre8T&=O^7q
zaCE+@-pYhtS&qkhHtr=%#hKy0iJrpWJQLUcPyEg&qk(37;;GW;7`u7@!<uAm+Mw^j
zIW>K~bOp~QZ!fy0x}d8Wxkur7cDpJZ3XAsoOxFzhYF6^ZCt-@m0iKFRuek+ZH|Kd>
zr#ZV3nz759dk`bK%NE$?3z^|AJbyRO;Bm_HcL6@sw;}LKf8O7dd~sK}hiq9UUe@c4
zmNS%@Df0KzXWNv9hLe9MJ;tYOs&WH;=LLMjoVcYho>cAszW063DQ~)yF(&tOb&jVJ
z$>0n`!%4V~fA$i#LB|<;jh~hOf3ak`bBcQEXEZ-o0rVjrzc+JP)$x}_cm72~g=bcA
zq;gc|=zkS1fNz}Oo7(WriHq!3fNkExHWOeQBW%;=B)ZE9UcxtDj-u`0P5nf>xwMbW
z%N{--zInQxIq|K@T8VaZVI%WacvBD8vNvHhy@4xXrDf!J&~A31-SnJ?H#Ud9<C)}r
z&~0+UH)}`1DMR^r1}3X{AHHT^ItY96629pX#f(${867J>n+G110o|t^S?K0?TMgMm
zRg2tBCA6Q?%%a0L1>u{sPH+x<69eDWf^ROuH#M0LZ+!*MIg_NehmsV$BT0d4k~DBp
zl4`&=cT(uUfp4}XpzCyD$EG`86&*N3swb&Ad~+4P3I0Tu_#Rms=EJj0O;pI>M9quB
zud>p01K%XTH*eS*zR@XBk6zPN%>2}VT?u+W8y{+Nf~uk6%#S5!%x;E0%ubDQV!!5V
zyeM{uZ`vELYm4Jml-*OlN%3m$&CFB_e5*3?8lH)H@yBFs*sD2wE^|{B_D$hUU1Ki%
z{0nl#R~&Q@IaFh|L+wU8G!?#?(aNFs%!WUq`)1o)vN!BxII%lc`~Hkog)wyd#l@-^
zeAB&UtQwNN`NUp^953z4{EuD#?q@H<YP%ZEw(GzcyXMjD=NHOMxT{@j+0C#lzg=xU
z#3<k47zHekQO&V2bYaD)X&60zMwo`#@c-Kz&g|4}W~WNOqPzGy^WeMjqUMq<?#<q*
zsAxUz6s<nhqxGUdG<!;-$XQ3RCx&eCQnJN#-@NeRWe0kUn?|WJ^WpiQFdu#{QcIRb
zYRxoqH{Bz-pN!O83ws%=MXEgh)SEl((}ZvC;!kbFpE~w?J{-P@b&Jqu^2H6{oBj90
zmH$Avg64&5QSWeF3JTYU*5S&De)E|5@aTPE^y4un9vr4p>}KekElh#`glgm}I&VgV
z>Kppa)mnH|S@5d<3W484^rlyc&a?}W?GI+C*d3n#5E{-f4&PuE%M+}MbQiDPL#N-^
zAl)h+qzaD$6+&-u)>LvgK7si6ftqwXKz&jJl%2Wo5A4(YgPHI{Vdyvc{S~sCo_NmJ
zcIHwia^ITXwTnLHW8Wd!gc3`+9)yw?X~%V$^XoQz8v7<+9pV15Hd%*>Hs*XAki#v>
zOv_gv^_k?ON9f>RnR{LqPnI_UorG)B39=kn_u|<v!}FdFGjOh)R)$<F=b5wn;1swd
zH7ndQ%BF7H*fBHLre3_>(ri;3u8-q8;@`Bw0|~W~akXfAPK%yRG?Rxi>wYJzX0+!T
z&OPBh3*8#rQ!Wi8b3$IW#h>iY+kihqe&Ew5FRi7E><~GR;>=4MnV~M1k9`mDK+eCd
zy1}))-V&>B@jd4JuxQIvi!L2C>zc)^W0y?|_BJW&Uq)_gjLKfrsL1UeD(vQ=l$Y)r
zbi-YbuX-r06nW+&UTQNGj(YE{Dv9*6_;=P;Jgq&Eo%PX5)^(GQ+L5R2RF3~Qn)}dq
zR)yDr1uk23v4Ta{ub9axn)T-qv#y^o>14D?)0|D(U&5sE=r%>J8I``mDCZQT<_<H`
z@nOVcH0f#&lNuV#ns}IO0vXI;IIRUo4epOWFk`hp1ucJpja>qG0@rQo$DYEKC&)VN
z;`a<sW+uDo!tecv3%Q0uyzi-1UT3Y$by)QawtN(ecH?hV+(4`9uZBnVTlMgeRY$jR
zEf{3gu>Mw^O|oiKFRSjPS+!;^f3pm$>P)lp7_ur)6RQSfv8v<|i&`$UXn(4O9KS`w
zhgtLm*4Z}7ss)3|Y~Up{S&L3L+f%Q3+sgH)=^EyU`8)0><NA`k>+MhEi11KK4@ILG
zfFH{B{4LtUH}XYU+R`nM=B*FY@nr((N3_!E(+!Ph8y!DnSAOqR`Ub0a#%oN8p$i7?
zoN|IbA}6v_P4O7fPJZ{`3`W1{1K$+pc`+RyX?IOLDsuPLJeh+E@{^eij&oP?#ArCJ
z;h4)KnES@3>ILI?qvhP=?H+hj`*?qDJ}zr4o>nmVWDk1%TA=Y%g@^L`>D+U4qC;ez
z$^2xUg60Qb<-zxSR-CS>NA%mw!4GcA%s1Jv39jUc;hJhi@Ka$IZ*oct=~+0OnT#9E
z_Y|)+6@9A{Y_s4y9C**0>m%MC^W6W!>Q(TRs#d{&;dwV=1GCid?6S$^V=`b$-Y#{D
z-Ih<8t47-h!LvD0f?QlBW~8|f$=~|FEbe!Br@io_&;hFz<o@IfjIfjJE)0=<Jo#M*
zyBF}4%QZv`DZ}$I6T1-}+O!z1Dad2(E049IJl32Rq0OY*w1&s#XxL^5Y;(0Q_Y2*5
z%qH5{?@BH$#-<Xye8}5>qT@v3Q*D}xSBi$S99F-6#HOtO!1Iq_rO)){<RjBuoor7B
z<_NhD+5oqFfSDJ}`mf9Q$uhJZ_{DjTRY%@im0FaqQ<ZtG#(b?7Xi1IGYQ9<Il-bH&
zMyqneH;3Sx-SEsWjxQX)=fQJv{N6K__iy;%0=ZqZo1$mYLtvYD*rp|HV}NaT!Zrbi
znOoYA{sPzZhiiWArDKMdmv_QO@Xh5d?4ClqX$Id!!8Z>0rXqYZZ$+|hFNJRwq3tYy
zi!zecdKNk08DwrIk@p#o9yF4l2~Vo#0CtD>hE2LAYrli6PXwPE2qXExSQb9do%eNO
zUqoA&ry06Z1HMj8a?F)rs#0iDh4I_+qW{1*9pIaK@J(s><_CQ9H+*A&Z%QA>pW2(G
ztXq@xbrqd`@J+xhdT~avXS@%GJxML#8?y(xPFu3aHIfty->k|G7ky6Db@*oDi9{7Z
z!+A3&QS}mG9=dOe1SHB2pQ>~-W~wSB>M<FdB_9&BpSkdz{n2pxBxp}h_KB0fnZ-PK
z;REri%{;j0f_Mecc{6@uyh@J3m+DXMrt^Qh!)w%!S3~l}H|h3Uc0W!hVVeo;40mtM
zPKFY67ys+f5PJRYpLfV}3%kQ-qSXv_D6fhA;p}DTUe%$4g&fNCCRV%XEl%4Nt11iF
z&%oZPD~?#T^dyVZELNFI$Es;2=Ea}DFc<B*LiWaMC47@^*QZhJ(WKjNNr+v4yV=#M
zmR*T|*x7%>ZmGR7%E8>!i%~I}*fmDM!T3||F<Q}-{Zo}<REOEAUO(6=^(<QXuQDIC
zK3ZAurp)lon(%0qgl{qmMr(7HX#Mks9h&5e&+m)UuenhIfK=TrO5IvBC*BA>2ajsc
zk4P27pNd%=iB1_QD}3V=K_&;j$*3Nwt2yY#c|_g@zB!40Gh-_H4Se$mf6B)rLfz=U
zan2PXEBeiG^qauB>}eRp?y1mlP3aJ>x23|>6aD5F`pu;YVVdI~CV%Fqro%V=uY{^L
z{WtgFn|+-_HNGl-75a_;<q%z78KU`t^xm`y(UyE6>h&-f|29}p2GN7#!)}Hg<Z<XO
z?!FU`YE+Q+mJU)Nd=t1mP|d~#s<D}jahX7kJ<B9&3OkkC2Iyv%0Htj7m)XER&70(j
zXR)8hmEL$}rOGU!r`XX&ep%?Q<DRy03Y{6u+>Wb3XV@=aJviq}#)=v0RJ!|O*n#Lq
z9vAIq^&iYk;ScPnNVn+$I)fv;>8j&gi~hZ|l()>M(6`~5Q&YJY9EhJEik?!$OI;u1
zk?$p!ScCh<Hr&5;WR3!Ty2MzU>V%OktZdVBU#na{T68?rqW4qHN*ij{x*)Tn2bdKK
z15EO=s3F(SOvmv|p4pVppPoDP^Ct0L+6+g<q2=UHBtKh+*>Ubmv(`Zu;@X@4tXaLw
zyXZ%{L0kJ8^vcJe4(<m1YG+WE76wgjWYB1ia;*&tP2~0IE{bX5%B}-fEl6_J%O0+(
z(cM+u`?+dFysH)-byb%<ZgfbwsX%LYo%-gXyDeZL{JS&!-nKB`a|s`B(|Eit&dK46
z(TS`!r96Sz@>w->hee-1o3(hmSu?wsb#=5!mROT^_?VQZwMnN7nAB^bQBKZA`UQ-d
z5@1wDw2_>vQHKw}I~`%6A0~a|`f)3RYYF$cYxo|1Fw3)4&fi=^#JzHf{WiLw(Z{aY
zbosbV?p$-CHZs%1J@JoL<QQt%l*lz;`eV-jXJ8gsCZz-H*V-xykM%toO=h`Om3PBG
zFiy2DR#miE<;>r}i}z)~IHwb>8Vu*08i}UUj|>if!yk>Tdj8p>(Yq~rGT$P@SPPnt
zMdSKgR3U|zODu8>#KW3zrIWy_pkioGT(c6m2A5gR^%o6c-cPQn<fBJi$47j^obpC*
zUFZ7ydyisT7CKgzz`3`0{XHHVnp)45%;Dk5_=Vz=lGn)&pPwv2Hkz5~vo83BXgHJ5
zdG_vwJAW}F=8QLmc4<RP*?PH)Lhx!c{Y#b#zG(vAENlt0nCQLnhfSh+I|=R~Taz1(
zsWAkGVGg`6M{$lgvc)U=`)M`rUq|+)E+6k3NCuhbTKnc?-Ky}l;G5EqxTj!lDldF9
zZ6ewJ?qs^)n;qzY_3p5njI8wUzJjsj!SdlNwf_TchX4E2V|aB(;q+bJdcD9~fq&5_
zMNaAaC7yS@Ug;lpozPbkHpN?Wn4cSkW}3x=%y|L)r$?UZhaZzPk{tRN^p2^p_G0vx
z9iCc$mYhD?POj(ZJD>1&$f=BkcU<72bTqX#=r%*}ucqTmWgPQVk4?;5^LyS5C;bt}
z5sYqwUh@UE3C+#?<4c=5U!rfA$DId{O%K>+18ft`<Fo5z{=RrpU11v=Y?IcF_s8?T
zXn2O_f71{f=TIKIQ8tZ3%Q-!OJlJ^df%x3gYhV+;W}S=w<+ujp%lc%8%POPQw1e3?
z!)``0IWFY78o)Je+4taW)tj~Wz;i7cOV;huPK!36C;olOq5&^0D)GUhXda_|VVZrv
z&~12GmbWiofjwZF-W&xuR`K?ZOLPRoHCxV+<ArU8ou>Z=wsD4SN}piv3AUL8+tfdp
ztTFJ-e7NS@Za8Qcz7{VN;T!*L_*|R$82IK-ys7)^nEzhG>#NX)R*+qWZ*DInLo=U`
znZwtg3F}NlN16zSjDwX%pcxJ3YxIM&dXm9O=Ih6@^D`X1C;*+vn>@3bUL04lI339}
zw;`9)l#i(o^VA?~Q;{qVd{d?{Un>u@VA;q~|76DdQ<Cz*H`Sgp`*e@p;1_s_H??Iq
z`!t!IDgxj1TbRTQ2iarz=5miDZH#7*xL=Zr8Ix4M71^9>Ny;d~{&4uFDtt4DjB(-X
z^xN!B)Z~qccyahsYnh{3jYst*L7g757d$FKUXk<`hbPdvn4sCg<Zc2I<m88*LvQis
zn#_ljEuKTK-;`VEH1JKAG4WbGBwl@b#;YK+;Th%Pm7XVFGas>2lU#AF6>;3_GehMc
zr@x!V>1)wA<@$y%b=Se}7>B;BcBuIb=D-~exw|`5mYtgG{&1+z_gKw+$nH#fZ@%v!
zcSCP+>Cv%rg>N?4*ssa#)XZYBy6}w-KYEK-p0z7_C;YOEj$-zP4;;x{INg2?g3)!H
z$s3nvHx;`XDn4Kb!|oW_=CgZhM2xN{#%L40HxcX&f8UrL4a|lY%}duGvr~nhuy5*e
zv=-4@JOOXY4ZbM_-(2a4KLy{Uz&BUmn>DxSxq)vk&x=w;_D+>>jZ!pxbGcy@87sPR
z{*6?&n=lT1<2)4&2fp!yZ!Wn+YGc(%c8^DD@4X0loItBthYvL^LK}KVXdCn4?TqyP
z)s0XD`pvsL;fmiAE|)psdOk2*jl;r~sZ+S}l?hip_D{V!5T^a3!+4&B=|ICUm7xD-
zF&X2oCqvbFWT@6KKUEFB>6a~3pWvI~^FwqiAVkS+Lv%7P`psqbhR+RFmT>ll*ACVQ
z_-4rEAl2W(e$A2W)~p#M$J;>VSP#?02C9bwRg}G&B@PCtAp62gv!CaoQ-Ibj^Vjz(
z{_LsdUUCYZtM&b~^b{R+W4dS*IfQ=S=r%djS#g8OD72shhPjQ)Z|V7^gYB<{blWgb
zS~Zj%*R6b2?F(}>c>Fav&!&=JIYN%b(bOA_o37k!-a2*<k6{Z(6qye%e0p#016|<`
z?h~7FPAs0E8RuJ8J$PZ&WbUO}1=!Rn)~rH}&3gODq`)sGSsR*_uR2<ax0x;+la}u>
z$sb)M{{gec)U@bpUaKx`LVMZ3Jy#K{hM$Ca&;{yEcaz!EO*x0SD!}NfO%+^~`(GE8
zn(LwtZZ4ct;F!?{bxAVlgqJ~+3<iB~Yf#l@22E*b(B*mt)#Bw#KLfqu1`W>RsziTR
zWp=pI`|2vUWLJ44x{5H53h;07gIqP-?xyLl+;p$1hZ;}u(7GQU+ECu2?PwM?j#*`C
zLT?QW_jh+PyrREsl9gSZcs7mDH7=RiMQ+x)y(Zo2W72(p^2GRIzZ#j8vDT=)u5e8g
zqjI)^Z4CHQu#U3@hJtaX7d9!Wg^B($6WKBPUb#jzElfUgqg4qVt#rGxgL%|{9cfnh
zrY*<qlQyk~Z^r(fNm&kCwxk!fCjJ$CQy0E*Kf!&h4PL2d#Um#t4Cl1s8Zj!2OidlD
ze4AS}w+eqd-Z!hIRiQA@-h602*{m|&#(P?1QRi6}O_*ZQ0vIM=EPt06d?^R7Cs@=V
zz#_R>G^!o5VzQ{d)uKId7PYNv)!!Lpbpmaw%{8)iny0+E_db7zIV-%*_L;n;dE}w-
z*K+c;tG49#XS8Zxx=j~@_&=g0^x&EuHi3Q8M?BS<j=-72d=yItIbS9+#r4o4@u=RC
zLw*v4hQn-IK_2rz-Czs+p|1E-h2Wd(u*X~Gr#_IA34?D&pu0H3Hz9uHX2{^&O+d#X
zXS|AxjR$j556Bbm!mIk1mt6<@sWyCb3-5Oj9~VXTW(3)txpV|>Y{p)3_~v9DK8Fl*
zCiv#-QgX$qa1Q$EMfj$8PC6ksk>6r&FAA@BbsTvkax~X(albGTU9`KmS|od`2)X^-
zsmz_tWX^4^x0Y}9);aX6s$1x-p6{*B2gm|*Z}APrUxNm-o2<!_8|3xj8XxWjYy-*u
z^17Yh@rI@J)odoKe*g{Uw5P_<&9L|%dic)6I{f}K4JF?O!{o!${A8mu56x#Gzw1qW
zjZXZ#!J8R?>b$=g_bfTc^unjk=!6sC=*ioe?SpNO!!}vxaNjZoR^dM8D{NCAw#g6M
zw2ebciG+EAZEEC4mw+$Y3`aH|2YX?hS3G9|hT}<3rE6fRO?P<g-Z;Z!{5oI%4qjIe
zvc@HuDgS{_dY#AGejXPbP8WF`9)eqV4DK0k(TaH%`gbiV0N=bgihjuZ>pie2>8(ZH
zFi!Tb7OjVE>VCH{?`F|iw3>b}%~Fo(f3Yv&BC|*6Hs#=(^5^Kffo+PzHXUG_GO*3F
zW9TAB$iW=KyE;HO4Se$_Ow$*(83Wg}-T{kjC+o5W_SghFZG<`Cn}l`r`>nyN!k0>1
z@?Y+zF?^GG9^Zc!I?xQh_7r*mC(yk!7EcMjc`*phrY{;&4?ZT5`R*9JtT6VK`=jZ2
zqcxfFzg+m(4jiqMwY>@2O+7rR>U?|!xU3|-IfeKf_@)<plLFuDg>N#!H!kqaYxrga
zd{ZC3nX-!>zfH_dtw_?IdAxr{l6H)y`>#im&cin*|0KPEZw@y{@2SFW2KeT57Q8I-
z#kKAysw7-<X9nGVf#hxc@T<t*q$p8e3X?Tno}j;`B&gK{_Ad-c&?fw-Rq##`SjQZ~
z`#L3PT8jkbu9%=W{Hb9uPRiMMy*L!F-FwIsZ;99C74bSV7WNq&FRwu6r@Z3TbZwkQ
zPmR;9esOYb8K=a;amx9cyv^@kzXJ}{VJ3V;xI_Nl?9uGtP{Bs*4bSIL)vvLNzZ0vl
zy|H-Mv5FfTt7{3ddQ7ihbN5*G*~O|x!B{zew5!W4e5jLlyhgjWET*TJ+3+fK-)wSW
zR=la54k_}-ne5vBE=De7Z&n_o^M<+b*+a<N#L?yFNB50^UE&Sd*-$n{KDo)>d}ap&
zv*Ew@YVM)iFK=43e)Nr2HTY(ObF|{hMr(BTXqEmDrMR1DHizi;!=GyFh|*%0DBXo`
z%E33!;G3H8P4<5xRR+F!J2p}idqm>Tqvy1VRMXOts*oj8-Ve#yz&HEX!#MQcG=y)4
zhDWF(eB)Aw{S5F;?g!y=BVSx^R=A1|U^X26W?@HW!^^XW>O&a)p<yaGC`@C0!t@=!
z@q=%!!Z-2f==fU`s^R{jDqI8JaSBxf{Hc_T5G_H!x!F2|?ulSEXLjmPTCm!~H}z@;
z%hxGbyUqmZ#s)kpdytB>XD>tMAkDcMsLxT%Opz_VSs+kJ2Ld!MFhGmy1t=+7fSNAy
zSBDb*YQB~n0Owz~r`$KvLH911?z0M=mH8>V99xpbBr8(26q#7Ospi+2Y2L$bZL&9g
zd!gS{_SNm$>_|c%?>pC9?#zDg@5VXRmwkJ9G%I%F^S`H;<sn)^U3w&%(I?Rczn=TQ
zyOr5z$2rs5oBPHmTq`eFb?~HB>ABEW(#-nO!mQ=l&Fa3}q-y(2y7tPXd1PSLzc#7N
z3}%Pgnbf&4UQL3Bs(8C=;2Af)&*Y{%vs~4qgR7>0a8cJyF0!P!C|`FMjSY2?$>Kte
z)kVeYxhVA|n$1dsT2D79wVy%#LkxOiGH7NegM8Z<bfKw1mIek5t80)e$5!;5866Dz
z7w%bwrW0>9sOU$7p7Qn9M!Kp+Cp4Xkuno-9eX6ULo^@5-%x-dP?xu$!=sA7d^r);^
zH`A<Ii9X`Sxw;_yk@YKV5D6F9EOP#7*7S8|^-M5p_%f6BC2<(Y6`wPzY_d_uJHs>l
zpRblTDxkVi+p-z;pt4bo>l@W6#HgLmjq32&sF6F3>NC`+`6b99pTaZS;-O*fjoRU2
z)S&lPc_ffC;+}Y=H*8avJZCjD`Fh-A^ZE^ct`8;1ALOJH_8}UV$*Ny1$=tvj=Qsv7
zV=k(iRey5+|IE3+bqTAwR^sPxfbY}-?~0>YUUZs=%(?Hf=pOgU`(|6@27`>|+VahA
zQAD&wZD5>>Fpit2Mde`|&!!etZw2T0T9h77=An;8C;D?uO1H3g)}jP&@;c@4B{TDN
z(A~1lG|S%GEOVq;iZZK49*Z(%;d_0tsGl=gAN~((>~sdfKi}Y-JZO1YBkA@br?ZL-
z@~k{`;gzG)pf<DBW5|sK(0lieK493s0s2XM{Hf<#yXf6t%x2S-5Dec~;2Rr!6ARxM
zx{$AlBv(vkcnHs*?|uC=7>}wV8qN;b<|Qu|z&pp_9t-bx!kfxU_NG02^9a7lfNxCj
z%|f!p4R3VO(|zp2LOcC5j!bbPeTg;MUCRu<D?i`5m1L+3Gpmj7bn6f^h<Kl?6TM}R
z!dnf4ZMdI!3+tS<(<wzZX-y0>XSsMk_ZQAQ59iJG(&M4@O9gr9-?C(ZKQVK7j!eoR
zm@&yyOYv~F!#U%|qT7(4*$J<#Ma!9vCwLQWW;MKHT?NZb@<g+w$0n4+{F@o+CT`CR
zBl+RQUHN$a{e%a-!6iJE8!c_+J9a7X+^=@dCV%F+tFO1QXB6InZDQb)A>+_5hr%nc
z&DpN_rw&*@0u2|IafWI9&^G%U(eOBIHk<Nt-*bbHDc7HK3S1O3pFLFD@Ka&*d6&^~
z&?oP;x2fhI<lIkNRGP<GA6WTEl7-!}7KL=R=p{_;GuEO{89e4z!9F{93E!CDn>6_5
z=4*Hd#tz{57{z1mQ)jfEn`W(qY0h7UX*kwjV%{6RxeeP)hiz`dHg>q?8eCJH;|pxF
z_$YqTVX`!^%>s0r18~iLxaK9t*B#79!8Yf&k-vd$n!+||u+9AS%xteEo4f*F3%2RA
zh_@HOAanV?8FUoSgiEI5BTeS}r@}vDn0Xq`_ZZ0M_JM`E^ZgUxsc3j66yK{0b6cKx
zRVK302EMn@Xj-B>HAcs&3rAJsYnCT-Q-Tb0L1xZ#qc3Hp{|~+y4&S_gO@|G9^8>y)
zb1{jXvCMt%WIk$Rk`65+dy|o*I#ZJ9k6`yEe3KEGq>g^<Y%q~QZk(h=wa|b{@o`zn
z@f>Hz)NcHz>G)4!iK+zO4D(5plL;-SUZUDCJJk`s37?vvVEAS?d~*T5SpeVs2un~^
z_{PaOK{s0^Xgm8g#}`S^-Oua=hi|qWX7?0)Qx3jy&xlv{6m*>t^xzDP*Y)3g!>)$X
z%h<~>G)`~PZ^{>p)2S!yT{!E|fC=Ppy3%<Q0`FKIGPQH)G1=nW1?a^26e|~YPR-uK
zF3oxDn@Ww<6SBo8{bQN!p!<eizl!k9c(ONnuiEwfh+Pvl+L__ED=L-jjW50wx=lna
za>nF}|NS>cedsMd0N)%s6{EIWW3;zF`x)#pWTIp6F4!GjpM6wiVq|AFy!J=BZtg|v
z^D)?F1N)_BMr%o*X#4~AYc`42q0-TEa*9^+izsbjuVx<jCi@chX%3CjZ1Okvoyiq9
zkJ5))%vL!?X&U{0C6`396N?VxUUd9<Mk=ahqzHOzPu@uV_(<O7c7$%>QLR}Lp*;N}
zG&LeZ2b?1mT025RGDWDyb9^ZHW)b?$n|@?&;G4PdO>6ij%e64Agm0b=2~#-p;SC#w
zsS$jW3%;oe-)KXq9Fs$pm;BAxT+E20-%OrO?j|@ynd$aB^FCP34hQR8da!1N1#4%W
zVBO6Wtnnv;WL*=aDltL2tsp(f5~N+X*d0!9u?Kv!D_@|NqTgJg*Y7Oe)R<lVGSwkh
zd<Pw!9DzUQ^bMTbJ9CbIfd_SW2YU|F@SX<vD)gPNsz0I&<^X-B7q}*DWhND0Y7Mib
z9mzCw;{Nbi27M)Cyfw22+IXP1d_J=0n|ri~Y@Bn-lliE`>?h}})0|r~G50&A0&^G4
zTjV0^SDcSYyJ_X;<LGSFJaQ+6`<gYY3|ZM1CcF)kG7UHBFLJ%LPMK6^j!9c;nsol8
zQ87b3^lhNKE*y1JlL2npIo*})h^vk}xyt^Riw-Sz(b*9$s+<Vx1h}ZG!9|;M;YZyx
z==2U)WsyP7lMK3tH<iiXARqW<c1MGjv^1z2eDetHCcU;n7fTwnEtf%tf1P#si?jUR
zIjh4PXBj!#{ov&vXg=Ns@~#Hm#K(Hh*UDPjptJ%8?R)L4p)Z{^C%ZwP!wuTd)~qa7
z(Ky^~>}4QJ$?JuCnDrT7srLhu2BTY55^OisqyrHq9cpOOgZ)PB4mWC<r%^3oo4ig&
z&3^2m7f(IppA)|-uThzwd1!EuQT2`+<uR7Gs~P3<(?dD8d1z@54^1!Yp;bHG)!4&>
zT}kfRSjk;m_q*xpWSA_%RZcMAr_wH}8E?|kDJJEbXVU03Ce1%+Qky3xwMaL!e}?RH
zUG%YXcvPP(It@#_FNz0M0l%u2mCn-tazrP}TlIE}MJrEP=w!2~=p2h$!8eopT9hr+
zqGS9&1J@bV>HxW=RqiTt(VZSEcRFm%8fb)nxCRNn@quqNgZpZ9xC&fzju+%>;NQ7j
zH*4G}vwHP1)8%AVNAlnM29m>iYgUE}_uRWJ3V&l!?v?mm{H)o_!#HG1>kp#mk<Nlm
z%vv@0?xO*Pn58B+R=5H?35;arTmRN6+1C)mUX59F&uk)Ve1eSqRd#s4W}d1nJ;QKG
zL-^*;&VI7OH=*!NdHCgbhBy$9DiY0R3~W=IqcVJBg>OE0Cyzr`XId!FHTdR)fw?Na
zj-!m9rqC~yIFG%VL%L{tXEHaH;F_(SHTxBvnf;k@szpx!E_s?o-fB$deRM3(H<;#`
zl}t@pZ@H2mb1TpL(McD(vlHqMZ`Hj>C&C_fvK^zhZX=xtDeNOiX4V^TYFl2i+SllQ
z*+<4Efox0|JG+@%``z{X3C78fU-f1fULdS94W4NY?@XM=Jok8V!rY6@L$}$4=Ti)>
zY1<6#rY;&#ExHJ@(J%9Y%=&$@UQfujGt(4#+oqmyN9ZB+SbV9v%g7Alo34g!&cHT>
z(J&{#HZx%xAGjqBw&}rn!r)`0a{>R!1*YkMw<_kRoS8GQ;A@3(PC=t74)e6--0@?V
zRXvYbb@(>(dOQYu-GR9|I>R(0!(eA0iyp$vt%7(gz&KfBd3`94rID~sYuKj>90WK2
zC<8+mg}Dn`)Ty*Z-QStzbI+^=H_YsgGOH0xlM37X-V5%<>p9??U1#akJ45H+|7$i!
zkCWYnZEnIgNk_=U9Qx0fI=F{DQ}E4pxMt5b{2w%%5%A4uUXO=sig28VZCZ2u&J<T&
zhKIBm9}3;3AbgV#zBx9Feqq>V+EnJECXwk$foVpQ(;0?FG$2_M$==-P%E!eeD<g{R
zP6%HYzWHf`b&O<>o%y(SyxoGG=Z%tOs{^N1WwyK=|EChnYr!}E@J)aCX3vi#b%1Z4
z!#APu%@+7((mAvk_-5=5vc(&c^kiw0jC05hPv+m@o7_FoZ^D@&@0_H4t^S)2_o|kp
zBMa!b>6xf)_CzH{C#rm8BKQ1>iYIe32EN&1PSl!KiFym)9A5_C%ui6w!DMZEqv1p+
zXk%!CO8X_~yg5NX(Qj(NH(SdjXhc?e{$9kZ%sJ-6kFY0vH+h>4@#>HsuXyyEmZRcj
zf^TLe((?!3WL-%P2fk_3Bu*pf^(%DWq4IcBw^JNym4JT3PKM27Z*sMEXmK@%rla3n
zqSx=A^Jq3?i$7<?syuu%lx%Sz-8Yf&O{R*ms+}Dz=K(!8^!lwofcLc4t{-#knl^wA
z93L_`9qlR!-;69o4-S0uh2G+$H)HUD@TK6JBFs%4jAn<pSB&1YkI_B4i<^~-!P8@B
z!@FpGW3T3m!_n-?rpI_X^We$RYT`u(2fp!xZ<LjtRP+|-KNqE?TcZ?1ckw@6nVSlT
z;=Ut_KJO^)EX3PCBlQ5jDYKDTaQrDpw@8J1q2a(ctKl1$e39DlDMAa#7)PCsP>$sh
zYTi9UlOrRv628ffKNSk!9ENX>Y$ba$EnG|Bo4NQ?p7712BH?<<{_s!q-{eRMQwP5=
z)q-y}G9!NJMW`N}Veiy>=BK7HFAm>~$Q`Po7tw9fL$o#|L`J&(?vuTVB3m3z_9iJf
zm_D>%P5F#=b2LbkRs<=n2YVS>1?f$WAWfb_1}8dD1>u`V9|IJzK7bkQ0B!#1uZmOr
zwfG&ndRsqcYUyn~M~^)--}OqfW4J4O^_ux=P&|FTbYe{NL8md%Ar{8&Npi%k^7*R8
zFM3<9d28Bx{CMu8@_2Ed^~I_-7s!tuAUm<us<vA==bW+X%u}lh6+|~K%DJpKxx#;$
z`z}OR=zH$_s?fjw$*T7kIY({4`!HGbe7RXyLd?o}#iVoYCNkpWY6}|m<cbHrtcQ-(
z@{rLTKPHz^KdZQF@+~*^5xdFu32$hItI`|0>g-pp)kj?PewK?|`k>c@xp3}x(NwPA
z8=YKK@^6FMZ^esR2y;v{=<6VZqJ0f2=t1_Ty+J~&DFxr;LANPfgpYwu>cBC9u+5(z
zoXOoetM4mk-GEz0^Lm#|2D&Z{THcBS%_pb|Y?IfZ#Ba{Jf5};UIPO1l*6xqax;Vg~
z+=I+A#+cRD)y$p*vzFX4DfuKUch01`hfMM=U{=a76P>CiUCM7#<&`iCdQF!19;)%u
zLs{UOCUDDz`5xN7-Ccucx@$~=yJD8Pt6+Z*Rc_#+bYFKBs^qGT=Uh~Nr;FkTxT@_Y
zSCxfjE(94=wv4k1r*=}l`yCZHxudGZbkwO=9rfsJM-@BNQQO{hRKpUT^lZGdKBYVB
z9XwU=n^7&|jpRctdIYbOKyQgTXV$wZ=uQ*hp=m}sf{a@I#zUqZZc5o}R;!sNH5g-3
zTjsyoPjpqbeFlZ68PtoH?loNC6&GgKTy<@(D|0Sx@?2zA@*}fygy83OwI~j@*}}Cd
zr2yYAw?&oTo3)+UveeUN9fb=^+xQ;AX6=B9#q3xT*Q9_FWP3VUI0ssEIF)<xR~9uU
zkI@CqV%RD4iuv?J#?!OgnLg;kUMjT~9!Cq=!sEK)A^KqOhK|l)j&K?5v7J5MN4se4
zZ92QNkfEv0eug%56~ie@f|>nhhN=Smvab*O74V|2!8BW896w(E&Kw6O_~{cI)R(-?
z0`#Ae$LarupKh&VKa`Cf3-lm)K4zcmBznJl(6`r^ZeeD>`mgiRrwI1JlA|d#iu}w-
z=Ah7NUUc!+__p3^jsIKzx|gp1O()$6I;fd1b0JsXo_tK=K=S)b;h9a$Cz82&K>q9#
zdAEhkL%n(qvm7S_Jj9c3QhX>+y20_8>UQDx5kcmLY;da>G#h?D#hE>t(+igANfw#6
zrv;L4LZ8VB+c+BGW8p)cDuu@5MEB$~yrO^5R?up!9QV%K*u4bj9EMZSSkA0Kqag=o
znuZ?*+th<?=E63Cu+92dbeM2*VX#dJ*k&edlLpU>Bae5xDOyfr^qCf9befSXn{Smh
zjdKe}cZXF);n>FK@u$&W%y-f6w%~*IBi|LnW7FTFUY-_ZL5IQ2))#o^Mx;d>y)63Z
z49C<*Gc9k?g(4Q6{At#dH=K7ku5eU)YStdu=E!-o8o)ON;F&7t>HCFg9-YQ(I>oL^
zxaJIcO<|4**yb#3^AWZ=2itfZW(NEq+_EoOZt#s0d{Y{}sl1KZa`<N3M!Nmh<4JL3
z<><DK*==;2_pnW^rF<NG^WjgtqWSDjK)ZRD#vD3q^9HslJ%N1jIQEi{Vh_VmI(hoj
zW77+bC>cI+pe04ZO~LR}XLd{3(2YFk=jnug)0Unc_~u4Kyr|k_eX5}2z&G8B(<@kj
zuAW?c&nz$&eDmWYy*2R7nkPx>|5p+m0ka%N!`aTh&2{93m(Wu@3lC}{?;lRC7`}N7
z-(<iyzu=pmb;;ah=56?<KYX(WzKMcw-bW<rGknt({U)DpqE@2c6n7zWTqRNO@+Gn>
zGEte{CFs(L1RXrU9Qg1AJ&7Q1<AN{ME<uS^VVr^q^8XUA{nz5zbA~T<6dww{IkGWc
z=V!;u6}}0DZz{q!P2rmr?&vu1O&9oPa-%pEM!z|6&7qw89eOp%p=YsZH~tR&L-uAb
zdl?#3b|_D7I{%)>YSM*Ro!K6%`ebi*_KMZWs8}7Z#a_*#v6`BdJ)7_7yt!*v<axU$
zku4rdZ?XRj=EHm0b*zhBp&jfRRTKUxVArglF^YT@qq{el3qKC$z&B^<^jizxY_^fb
z5l0>NHk61_ST@+_4Rhevqc!J1wC*pX`zAG7&l98N?isC{@Xgaw(R!CTTCLbSRsSqI
z7}%|O1HP%(A1x;&N+02yE;XX`G#_(S?;`b+{o#c-vtN_>sn_t0*_-~GR*|Ywg?ur5
zWBbhh1!lzW!8di6F+<fYLT}-lip+<vu8*dZBSPn|v0IbfQx5uX68h8g7alIZR^ciI
z-xMNWT#|fo=X82-62nxzDIOJk^Xgfs5>JQf^x9CBn96R=ZlQ|FA1Ximsr0!ax)&8f
z4-I>$==Pg<Fj#2=g7q>mSX1i<EBhBZZ`KCs0@<4_{y{ohH%QCw1**dmjzsds&6pAY
z6d)h=hCgo{psEl3>2P9S_&PEGXzqt&yXfx-bQ^LB*O*0FT?#GyCOfx_`Re>p_8Bhn
zQGI4njcw@IDBw%~20D3jb}gXS96Zha6q@nOpUf}qv#K}e9;fl-N;$Ui`rX~k1K%JM
z^UJDYkI*c>qtBG4OT9Llbx!W>zA-m^%Btv%Rwd1|s#`94d(jC}>u@i=%A{3IxF+^D
zvJ;#crLpch*2bN=Uw4^vx+~8EH(3w4DSnb0eXVX9`^HrRd%G&U6xZ~ZE^4_4-DakX
za`$l2N<SB!YsGaGzDb8~?*4_o0^hj7H}l||TJX(yvNx&D1{EV~)2EU_dy5%#J2!fa
z6ZxE<&ieY*nct(co}=CLfonRwa7JH3$ANPK-#ROpk1JT!ASXEIEKKzHF&uQ!S+&nP
z>*N_{-Tud!o;zoqfRSEjF=$|OgN}bUpv$>vHvAR1jTx!wCf&?1DS*sTD`uhA%cN(&
zjH*1#D1R7l&La;Uyyd};N)KIc?4jj((PX;2t6G@5t~PX+k>5)(FE{r4x$4RZ7rp);
zNoN_}RN8f67#Q3g26uOtt|X1NNmHCsm(&G!cPsAh?(Q}(?(R0g;O_q2eDmYEk|ya%
z3hi_5v-a9+kA!Pl<B6_Bovt~W)P1x3;>|k!)>~8RdZVkF)MkrGEr*%(F4Uw&GHG2=
zlfHd2>c%xAIb}xbx+bo9ytTg^EcDm}JD62+HQs7G)zi_&xJp4!vMS9gJWTkOx|FeK
zR$7Z@!$qn5+w@;!)}Iqu{mgpoW7gygCcXFfma(N-8Shf_yJMBPvW;9itExVy=2yz9
z*_CYMcsf*UGXKxJd{iZeueRLrr5^~!%kI>|f=;cg=~U7PC*D>ky0wdWcP>SYfOq!r
zyK~P)?xRcYWAwc(b*Zk)58pTac8^_Zao<mV7x0YU@YApK^t2CgQ(JIr2fm|*{mI$+
z$((iitj?2dbPCUWzhbQ=2Q^vxsOG&V+c*pP#8umIJq*XtM|F#=zIxQE94Y^C&QG+Q
zRD3xLg_js#a-}|GNbxuJr6k!j>F@xxVK&G}w|ZH4-)8jHH*#x3U9yiVlYy|FetVez
zc~{Qq)6g_vhFLc9czd_hOtODIuqJ+C4l8=iCD`RA+Drj57N$gyvx|4W?{_@ah2b1O
z9*bq}Njq{~+LP;IXZA+*7D_<73I46!P;=S^uRM36+0>vfyec(fdYvkw$0RqVXPWoS
z>!zohp2pX3&o^$p=}YvbCn=KGZ>hxd<>|YIZSG}a{sMKV=o|j}Nu6llL3GIj{(8LM
zU+ebx>)$P84XpH6erieOC!@KHLdSt^hQT(4VVmr*&3rVO`s^c@sh3`XZPHor;K4TA
zYIFTl3Faw>p9&2)c{W)DA$Z)4Tu1%%lh;+)^%z<mOw)4`o|;r<nhf*PfR5-l=!7fK
zZd&^J=?VY;Tev^D9r_Kd(~QTg4bc{>_^B7#%|`Z}HQ!wt@yexhFI_6Zx^>&7VVBTV
zPvHYR>Qa+)Xd`Fnv4(GEz&Ac8;2rp;-Z5ALwpjw(Jb-P!!Zzn&o7QMI5!9xN!#DeP
z|L3_GvK_C>R$jjuuMB)M5v~b<YqC+3TCkd$)N(W%_@>tq=6Wxpp9;RYKl8sC45<^y
z1su=!kA-JO(-Vbuv#~!u=e}q-@Xd-Y@K!SSWBJ|)<~#(^6V)28jvt;KJ3eC*`GL)-
zJvE?4Rg?Ft3@5=iy^8U36+n-IZyLilvGC3IZ+MB}n|3dgHRe__bG4H7<23Uc;F}NI
z$hujZOpZO7#qiClG0fB)n5;zjrfM8)6OgRBZsu+_MDr<|tjX|A9DK77zKMWu9>F)6
z!;&-&{pL5{w6ybhgCw$<n6p_vNiT~gX>jf&v^my0W^2MZFJkBsr#7{%F7q|<-W1P;
z#~8jjcRNAl&m|}vz8M7H42N(2UWIl;4o?3;3CalHY;Hx)pN0PMRq=Y!KVD||rdF<a
z*)PY@Gajc6<C&!y&CKw&aS9~2_-)-deP=Giunf%A#9Q2!zNv|mV|53%nGndFO;fB=
zD&Z^67c0L%n63GmoEyA1uTREk)eh!o;=LI;HbxaY;K}hP+piw8!;3Rl9AELaZ_)Hh
zl2LpEzs(_7XG^r2z&DGcqE*oytpUxV)v+?(oB}Y;ALQe_@aWhTk6!NesL&FR_K);v
zRU$cm{vHj7ZyfN=xilVSe-x#Hr=rOIh*I!0yv6WMZusU2eA5WNISAjpfN!!gKm60S
zNcEW<seABE7Wifwe6t9?>60T;x8R!)_$C{CGh_`q4*E^$C~}Kk5jt22e{nWuP`##i
z>P)x>qTkemZ(h(J-VnYC${S9eG+rC}!{^SVH@qWr!@a|_E<gIs<4~PA9;)nXLN$6^
zs2+C+RWnzp8omorer8Xdhz-#W`lmLg3DMCb!OB7QO<?n2l_(Ibt?z<#AMK|6V0x$8
z1!-ti?mrAv<E4RQu?DJD!$1}MM6TaZ{Kc07<kcoXmLKiZ2Yr1De&T%e%Z}!HF|u(h
z)oI5yZ0(k+mkuAzj~04ewxwzmZ%M6;YsD)q6axEHAR9lFISV^ZkjFLNUt4|HyFQ=`
zuXAhoIJc^Gb8B-bJv77cY^)|D>lj|td+gmy=v%nu)|0ICer9HGdrQB~W4F%jam&6G
z|IJ9Z_OPFn9plpK++@uRa;o`IUv+rwqqmzJx?Y(&Mk$9pU+tQC#V+qH_;;t+H8|3)
z62EncJ~rK}XCuSGrt;6M+OgfLi&Ly>+|jDHZLCV-{F#*3s?5JET5;8)BbzPiJO$5i
zcZ=plSrqFJYnUurTEn9M!6+T_z&x4Ia?;_;L5JD%$*f$|n$o~DZ=Ra<{)w5~A+v5j
zGoyQ$l^<@Io*u8xFS7>2K^?BaKX6UY<7OqDFzYDy-(81?&~V;U=NiGW9lX!}(`YG8
ztuhU<D!#i@FS<|<+KzS=<5Z^xPQ6;=tA#c35>spXvB^hc-aB-DokP7{4$ZFNP^C%^
z-GXg~f41vQH9H<mn<jm*YTruqsuZh+;SY+x2G7B4_wJe%x)0`f<gEah<t}<n*dh~k
zNRuuEo3y2wNmcTiWd3SY^aUe*=0;WDY9xQrh_}W_Rv8-4PLrH_On3rJdO5<Rm8(o@
z&i#(#OzIP8WKM-qZ~Xbbu12jNY1ED*MoqqH)Y;EQdDS#&9<Og%#an;YG0SH!|DOCC
zq_2v8Hq={#M|+d|&M~){WJO0B_R6G61-v!e<gIHnV8uOfXIiU9!Kzn&*mSXtU0u85
zDZ1y>$CmV)jii>a)1@G;xeIe%3Pe8&#miL;U8g3st8-UipCfoFsGEI;ahkm299R>!
zfGeDMkb2;0{mD7^0q5Ry{pl|oM6Cr+$MdE1(qCuJ$cEN(Rwgr$T%#Dg_ak<agNzQ7
zZ6Ug60(02N0XRob)}Le`yvFl09bZcQoc!KTp{`cfPX`yc)W4yx`ZR{wTli`~EHy+v
z|JKpIs{PcZ*G1i0pNtQ!9a;q(;C<X*6Y2eVmfBns{%Wbu{9O*dLl5HO7CJSPEGgKg
zi#OgH>PPq2(5rM2pYXBf3QS9X6Fo#B+36w5h`%}wxpiNekMR*t6_53Z=QZugQRMTj
zgi&%8pw3hd-wfQ6%JWSsl3{~B^ROYk`c23I<~9>wFGR1|T#UYJ*6hE?0_L?}Kj8(Y
z4z-!jI&?RlGI(bBM&@teR}Wmv3~*{vp1H6iY~$c(N(bBc@$Y*f#a~14z|Rlmb^N^Z
z(SSF@HhW>4`OWAZuZyNr(O)Htq2XjeBcA3~XDhvYO<=$*>?>$B>5sEEQcIfdr%mJi
zbZ4X=wHlZu$xpii;U71C^``JmBbc^<pCS$T(9vytjQHE@!A0=R=DcV(nec^C8;yC6
zZurQhdgz1Ce`|#2TpEIQ(~RGl`DgLXz%_&6n*H$0Ncbf;Y%}X<ivB)=cLlcjgOwG&
zdB}4q2l3SGPf@46^il1GDbQ}7quu<2c5`_Xz2zIIAFYR#)}r~qHd$AaN3)z`!#0!W
z(aSvx-_CS2l_~gX&~0{)M=Kf&>x^Mu)hM*0K`AN#-(>DX-c1i?m%}%=lK7rz{K{c)
zOdvW^D?C9iG#MN3Z)65TQ*@>J^rY9sn^Or*rwpIHC=3SQB<Cb^CljA5Eysm#4!%#;
zlo$Af@6ZPh-+Y8`z8)gaW(R%X@XfJB$*Ke21ddMD&H>~TcgBMpo2>PL$;!|Izm1jo
zRQ2ewDg_JWrl+bGI!)&!SvoRfDj|t_Bt28qs-8}vhiY_^di3UgQj)Hr>s0qg^MQFL
zl}Td01vA8{TkRN}sHS*}r<5T7FGr%Le@W1uCkbkLfpvs_sa?#}gl|^==9|<6wS;d1
z;G1FaO=b8dSG@!s?-#G9)TdtMh}VP*)StG-X#)P6XYp|=(}w&T3weKdi`N&UUX_ll
zzk9LrpxvyP2-9>S&#!%~5{+m$@J%E5<^z4<F)w3O^=6FD9b=9r-r`~NV>EUInoUBC
zM!RCj^=2+ZQR-G1WAw*Y^7|e~tJ<Y#UHdy)USp#5uyeHj4aaNjhrhUKw4POrR=WJr
z8lEOvZJ&Ac?&5#hH`5m4#To8V(*%#KZjbKN@n}e4j}HDK>*isUo}Ivhvmr`lr$%X4
zPk1IcN{8W_-0;m(_-5R@NQL1quDB;swMLP9!~Ch6@XdGlrULy_4Rhi({v4r5_$C1V
z&3E`_Pilm^CPk>7GeVCmM^L+oPz>|KZ(ImhmQCdQk$*EOgzOvUPqjn8@%k7h$601G
zEDcj+=P(@<_X~yT{lC<tj)kfz88|Icxu1d;#~G^9X+xD^N{H6MH@l4?`oMgKX)l6x
zZGN!2_@d){3nIHENG0H#^n-%%U<c_@l_0X-nW?!fP}AU>ckoTsPUttq+Uwea0ByGg
z$boiqj(({J^hWKy)LIMow^E);%=U4%)I9o9+CHa;^flgKvhk-;&p-2m>%b-SbCY*5
zYb%+c?1P;SQCDa0f5&xi+gompo#$q54Ou7QaE^m)iN@^fqtR}*k@Lp&Qo?dS{pf>7
z<2m~B>;G=YK5)yhmh2QBKT*<8efFUbC%ZIfs#Ce}fn=oCk#NOF*L)p%Scr4_E4${L
zuq$l0U7BUrv;KDZQJ1PrUFr;Vms+)9o6I)(KgMUX+Nz*oR<%yFD!19HrB$sOl*g)m
zKP?JAWYO~t+|IYCT~~`f!#Dl$3(uEDqp3;tEs742)1r^mr8c13%!gepaLHTPW;aaJ
z`>|P5(PxIS+Vb^p<{3=gYCc>O2kTt9MI8(6<{E4>?}%BS4x8oX{^QGLrGqiDp!fU@
z2dzivNl!hiM50CSid(h3AD=ap8qZ^2rBrs3r+`N|3hidGL!}ovbSS~0e5_6x9h$z*
zCQsYn{pfgSV7P~VR{48b)z(3e*j0`-$ExR%Rz0d|RjZ2@d6lAe#(N)yZ#v%b*5?}D
z@;+wL)tM&6cQvU=JClw#GHH2ElMbWboIYdJ-~&co-DFgjIHP{lFsf_`7$vij86gHW
zzi80Sy#`&}ZcybIqc#P>K`o8C?_*Rumr*U+8TGKAQP&q3mGO#E5ig8-kOgMRZPJX3
zMkUNL(tBZKmZgzejZvY`jQXpgN!L1}@60x-+&+{3d}Pw!*}b*7k+&KJd8=SoZ`FW-
z7VYp>O}m+lFte^6w<&WSr+PSCTGkhzF-+qH+hje+`DueoVZ*o{@6XrxbEX}Db6Ab9
zat?b!&H&7lg055bHxtdp$1|ATypk^F?D(ixDIYCN@1t7J@G2d4XwrTky<FhbgvCyE
zS&9d286L1LPUes}Wnb--?@FiUEO*MubMNLmb#RVTb!RdsYzqE({!e3@_^N0{JXD2!
z<*MYXkk-C>$p7uW4tS<geKnHL)b<zW*hIH>w}G=ppmiLhcML6MCZ7DCBU-5bC^Q?g
z^{&wiRp}e-avK(_h99OI^|s;7wGNL=niS?-l!j9>G^ZyNzwT%HiqKcOJja{+m~2OG
zv%O=k(+9jRJf7`6`BE?Fw?==deGBi)Yii9|@!J$6OB#-`mLT)7D03HJpfB*uKbd*o
z-1w1s%|MQw%rU!ht+RVOEVCLN2NwK*_ioxkcm>U-_)OSlB0U4tq?`lre0S$HuuUP@
z=2{5XIc@P3yZPGJUq@h@TV8Myb*Qpc@zxY4GbblK-S6C5$X=rQZoMyr)=N#w>l%KU
z19)my`l%rNSYj-GcC?#q-C><LKl$Sez6#sahbueg<GJj9y7twj!Duw6xQ?2?3xCZ<
zm~uVYo!iKpfM+7_z`*?O?Bn_9i}Y@Czd71X>KcA$PI2CcTYl#ccV}&9?O@G<ZA!5!
z!!~7Mn{u#E23|J@?dIA(vWxd{e;2$1-?V~nEbxus8y9>t9PQ@!jPReU(0f*(|G+kF
z7IN&lJP+S^!8a!O#vi_UIF20RRE`PTEF4a)Y9QX9ey~n&jtSokg>Qo5@d!oY84lsG
z0D8Jw;<a(|x!@ZYeAA~1?^%!RzZz&c@Xf5!+`>2A^D~z!2Q@MHM(|De*JK@d#|*0H
z<PiS@-&|*Y=GkQBKboxR`{=FONZo3|f4x(EM$u>0KUuXqCF8L{?+K=7yk)Y=I+B&U
zDY-dSll5=@WW_zEHg$)b;>+Y0?@v;(ElC=-JV^~^;W-|Yr2hSw&j8<K3Bz*?-x!VP
zIF*>=kQ3j|szim3OvGo0$GCVRGfEQG`fGwZkcBe<z7c#g5x)5gzIm`RL1y^o)j&MQ
z-4paIIYHOLnJZo&4yur#l;&tR%$>S^K2E-y<5Xu9ImP&I`oK3cyyIl56{p~$ar%)q
zPPRL-TD2LL85gU@o$%q%H#Nu*tK9HS_q?&{^NabJ&zYTxxA@E97@gk8{LI<(hYyBz
zVq)~(L4B%b44yvxH_X&*3ESK_6RpSFqt$BxbEif{s}cH*KYVk~g*T@$+D-*>{_;la
z=udjVpLjI(JXtq8@e|MYsKZc?cEvGk)6b(c@Xh=}9&P_g5BLM-h98f@gMxN5g?XCr
z%?kKtEPQjlLX`T$H|OA+_UJc;y^%T%-*kp=(gj88UoYxX<ogxL#r#e9CJMfJe>6f5
zR-@a%H(%kKcvpm;RE^M#9NdR*X2Lhoo5QuId$>~Y-}J2)uAT+Lb?ZZzZl4R&=`~>r
z>q%Y87^W$O;hTq{avcd(@`_MZ9UUrXQmAs7Lsfh)**B>n$`cu)tp;X^!#Ag%2dfQy
zV{<Wg>PwI&T?o<y_$FpxkdC&6ai~vK_z<Xb=r`@c$t~X1UYSGMtN*J2^`9G{MMVM>
zHVZAiD1Jsy8>L<(w}zZ{&%Tzb$y}oJ_sJ5$pICn%c{TKPN8_>C%zk<?8J_|gw}(A6
zmFqh@I`p-EXwK-zrBBh1J%(Q5cyyZ~ewx|IPjgP;4=v=U0!96lgX^YQ34WTkz)!~e
zZUsK2cK#R!Lf5G`&`s{VpR)Z;eIc()?Y25qkNtiO9+Qc0eYBXG)VwctwYX+i=`D6$
z#%q&)tz9vr?E2n``qTlNdg1FR7Hm^!HJkRPqj&qdRi&2GGe!MmYn)ZrO;%m4VpY%F
zRt@-J(XziSstw<4gl~p)vM4^-qB>5C9#S)!UV-`w+!27j(jDFAD>a?nXUyb$nAPBa
z@WDN^Hp2-Q?wD1P`w!4*8lube<o+DEWz<8n#`9Q3UfXoPS<&3C`5V0k#>oTYq`e69
z{A1S2r)Cx9_*<#x%t1FAo8PMPEvTW@w5cCI<4mVr4{ZFL^qLP}WLK-Rc4b^`SFda~
z9XW@0dZ-o8pH*@6h^>pT>ZZY}<JI^Xi(0j4t5yG|viO<nezR!DMvMGQqFJppYxrSr
z?dsvJ`fx*?r6!G@YEnRFlagDRw7R}YKeCxr=e|+7_8Zl8g;B*|myY2^g_JXDcmbnc
zq@`~4(4eXRGmxiZP>CJ}El4)VHqxLg3k<rnj$0U~IvP%8zHd7Gl5e9?^Eqa>FGhWQ
zY*ZnR+X22=Khdb6LyZdKcF+Odr<hUK;3+e;vC1t>vi)OHtM4Xdg>Npm_140U-l{mt
zTQ|5}vBX;q&v~O$z?my-8WHT!!n^2LDL(Y6`Kokvr;eiO<myEZG(MdH@h)YFc4^iU
z7U!!mcR8>9&!y=rIgc&F)04-g$C+I^O5c5!R3EJ@>ZAED99p%D`X6k$x4%OJk{znt
zz@g{m;mp@|b$o2sjfZv>c}fOBIfrW1<#}|q6|uZ#Fg*Iep{G;P+&1{&LG>Yn&sVwg
z`N|j0=?>r6;G4-QzB)_&G5)rT-gvhfJ@=Du1^VJyrk-dx{Joj*<fpVi4}v{z(Bn3S
z8t<R*!*;ZiV^iRu^Z#Y{wr|SZ1N4}&Y55*>l~TOk{FE6_52*LtrdD*#UoEcFZ*-4-
zrbqbyANtd8@2`wJzV9R~blP7FVVve~@Wgy4C*Th}+4zRHa6Hd9^ewpP1?v~w)d0_M
zql^C3ZJ1%d6@6<7{+n53|4yaPo6q!a5{xr}c~WC|4c9?Y>=li>!B5H5r=!W948e=s
zhFk(X!(*)YbiDBmi+NHF@#@r|C!9Lny!`&k^9Py5x7}(nh<@*q<V)TrbKnl1^$YlF
zHlWKaL>rsvr?sQWiX22AH+#=EJT@)=q7KFWGY7WmmYd!wG@5l^s53popMKn>DQn>z
z{@?xhUAVQ$rRs3+?|NoexcK8@IGN|>uW)JCDjr+S{F4LdF06&DNLB&XDOND67;MuW
zwlTprR@kNi+;eL`{$lv18hn!(zPY&_9@<KO`DSWa)TZ{J-MmM;sRQ5Cf^Vjv-Gst7
zZx+Bp@J(j;=E^iY%#-1r31~a;O+4C7uaS6phH;z$DH_oi%?Q4^)Sa3Xe6u!=nGg|V
z6$jIg-j3Q<3v?P^KBE;q$bdG~7~ZLiFS$C*R1qzy6nTDsQBTW<XBoa(oRQ<EK?C}N
zCi50%dB$u7_{Is}#Gj*=3cl$E-(=cK&-hAmZzd<J?@03g`Z6!PW3tTfO&$2=eT!s0
zurOD%0U8h<<95ZUZ)HTo`I)39Z<6%nZjyGNPLgLATFu%d<)4$J<)f0+2)@Y;-`t09
z+QT=0)k;!td^syKC28%OMD+4RMchr);r)raHkF(l{KvIgB&zB$e8$wLPU0^PT#=yV
zqhXoe3G|dD$Tu!Qy<!p++&n>@;hUvJ@fyQ89Y4pZ;r2LP9vP<-ad>W8#OZJ5GTbX0
zr#H+DfAv0Awb5@btj9w<I+h#|<}lzZrh+F!#aKn<iPeoC<o7*|QR^%8fB!E=FV~TE
zGlN+S{n2b9Vwkx`W^vURX4^5N;a9YJ|DLIN8b9&2Xq}wTY;gEyJbZIKELtY`=4m4|
z9cE8u%@eJfKRoL4$b%QbqpaIKa?kZ>++dID#Nx$q;ybQIFI7R0*8Pam-}lMCIYwXj
zx+qni9Hmt~n5PNfc*8dn;hSdg&3yPK-8JTF!Z$GssZI5cR44kw8yO<?m3+Ulxykwa
z5}{%6jSIesf^YJUjUeBP9&x`2b*_${lQTl|;G6yDVVjNgOpT&e6%kI2D_kQBg{#A-
zFnBXesq4ZterTA2s840d9j3eYLiK7#sGiOWRRh0J&1C-6jF%x=vpz(thKFc#c!)B<
zH+hPag>xfVZO1Y{9Pdp}yf=%^2Wc>TQwF}-2;Wq!9;9u|XIR*sxl<(rm0>}9rEAh&
zGhYWNecu38Skq45_qJ7?%*;e417vNH7J8hY9Fe!o+1*G^Kbhsr@Fx!MN7gU1imER{
z3+K9VV+?t~HMymybpK0cvOJ>>#UA9z!1dKGx4b&Kb-fyOsCf32w)Cz9p#SFbW44Q*
zF4bmEBpUJe-hLW*onGRHZvFd!TGkD>UX`a-ev&;Vr%Q_xol50;_W64sRdDzy+|Qv%
z>MkSB)916zt^(8XpZ2qBDt?=Mod3^1u<61!n;MV6zv04HoW`bXxA5F>{!irmzF-hu
zn`o=*Tdeq(tUCA)IejPS1z(GIcqASXw3|c^KAiUWjhk9@EjN#6v}h;(kt1j^$6%Ii
zSIyc$Z7CT>d9=%{`1R;7E6l342+d{|=hhkcYo?i1cdA*rr<nBvj`_F;pUZBu`mpNm
z#D@dt)M52LXx1g3yNutM<~=QihfdI!)gI00b1RF^mauB{O{*HVp#Q6kU8|bd)tVYi
z$5gul@w+%+pGx21m`PUUh1<5^E9z>us*S^{?$xc@3ERxhWYvnvR%WupJjJXEdxg)3
zpErk*{;s8F)qrnCBzr6L7j-FWQ==!Dln1`?ahdd@wn<+znPj+a)Ph|`O<81ACio^j
zoKp8M^qZVUz5Qg+>dOXw+h@?};RYS(W>DrNgEGb&RCbm@y;d1yJZMlIU!$s_;f(EM
z)GPFvNhgfVOgHN2Dx-2NHtN<)^q~p7W}s1xs~KhBc$;CS9qEi(SfBTVk6xyjwDyEa
zb3Wk>%I2*CCB3!P>dowTZ~XAyIyT;$`734}5w*Z*vv$m*7TAsdOZ2Lrcy&DVfpz)h
z)b28zzrx^`V3&5cB@fc!QgJU>2HxqEo%|#pr&eZls?}*<=DyIMHrq$(@M#x#>QJ`5
z4!v67(BYvDWsGs~ckWQ@JPxgSX4jJY)W=xCPweEg+R49i=uH&|v!)z+f_JFfdTNN3
zd~~9%k8a@ycck&v=sdpi!YlsE+gCkZzPi$$Y)0y9CE@?BrK!sfX70xqry5Oi=>zBB
zLlyCcpjQn0!>zFDWDCqB8+tu`_=V}cMgw`<1IEGYQW+iPEMAxbczw&lHH+Yv8gR^R
zbeS)gV1zSdR#QXzewcow1Ju1>ncVn)d+#Chk;lI8Bx7bTz1gtK*Q?BWy3fo5j(P71
z`U}U}1qao<!tD2>^hNPKtyw2_z%y{oCv=)q<GIeE{`7Ge@535D6c!qcj~LCSOD`Cv
zGaQu2?2{<iDVX_Dt*Je^_}o^w2CvOo)*0AldKGk^GUP(>^R3N}uJfL}$CKn>PhcKM
zI<gc_l9RN|PxD6lsR5czb@ZAIc5-XjV~&?Y*I{jD->C-Aykqa_oeqx<_YZO#0N*Tp
zf=3J8W)qw{ccDw?xg7-KPUQZVrFj0iO@MD^a{tzH7c=Tz>JH;%fMbTTd|A0zsql-7
z6~Icvihyl`VVgMErW?=K;IR>V;g;Qey%X+%Z;HY<={BS9z&GXLn^5>>!s-;>AJ$oh
zezO>jW&wH-e6wRFSvc@bjmc;%@J-dR)Rso$_Zh))hu|;n&-1<b99_uDN#?8z+pLI!
zYr@cP0(ra*If4Eh3%*%o!4u7yIcsCQK6TN8s=+%IIA$rBs0g?Dc)#q_zTlf&Y2c;L
z)R*3n<M%XK;qXlb_-53(WOavcEbz_OEoeHc(QoEcj~YP+U@v;CQmAQpn8`_f>a}08
zhMLI$te>ov6<{LxW>mIhElZoM=AV<4?S7IrpGi``{YeUjZ%)EDStcYYPclpc-#mqH
z{NS7SHSyrUH$CB-@^9&_f^T-iH|5qO%DMy{hyL+TRTFipLL#~S%-wuLp7AnfQcX<I
zg<c7YhHo0dH<RES8+_BCRDy2*h2{g_*uIeE2jBcToLN)&Z?3q>F@|sY!8hsPn=kKT
znY#t!9Ew%_VX+EM#D~*5R$g9saG0r?BRBox-|3Zl6r(2>W5|Rj=Vnce0;iH=Oz+g7
zkQl`p@Eli+AzLCw_rIabJcyR_6m!Dy73YU<u8gEF9KP|uH<O*wdf6}<4<|k2xnQ7g
z9^HKC(Z198hquxrHJc3MfgY`hra#=5Ouw4U*DL_ze5W4u4>hQxQEI#v56+}0<%4hT
z1V!nQk;mYhJK3Ui7yr$+tC32iKDB&4{ou^koEs9Ur6y*FSB{i35Bd%>sBXYFHQ<{d
z_-5eP2qmRNXtJ9;oEr2_<)Ur{-&BKdrmYW`4}5bqGMstQ;Y#`|T$8_qDgUJ~CDJ2)
zW<;1WGzinKjA7byE>yYk6&EJ^rcAR?<t!gck393HMuzB*4k235B1C2Dg{VLQzTO$E
z)!l<NsCuxLQJb30T!!g$nW@<)NT1pS=|}D$Z8#YSX9OyJ)<9+P4N&_x%r`zquhhzR
z<ejubA7l<Bvxx8YrhZ<kxr}%P{m8N?J&1miefR};G*^FW`?KJhm(=-N3}Nn~huN3#
zO|RXs%r#i1B-c`f-Ac_u55qLRZsBH*wV$p#(Tf}Uk%>o7FZ)W~4|qJ$-7CN`dxPMf
z^KNZ|alEK$)#3K$A-oc$*+=jbKhB1q7TxAo8y{U7<j{c(Tt8p1YtS0ItkdiY?QK`t
zFnl-l>>8N|fALM5g4WygrmIa+Wo>%=*{YJ;(LrLYddeD%=jK{^EB)>CexD+rcpHA!
z1r|+)ZQgVw>xRsmitQ{~P|u<t#kr*>)f%7h%r9o;q~`PtjV2p4r6R|uIl(7Qx0`iv
z4c;0!q0D@0P_WGoxaP=YI0z4IwMpnS(|J9wIkL?x`&K+TtbaCB+uCT>&`oBQ;ut|^
zIX=g3f`(J<qghjmSjbbbs98>{&Y?NEGLjeQv}t>GyDAs6%b7^cNj4oTOrGF7s|t0r
z>Hti$ydfM{$*TKBteT(Osz`p;eXlLr?Q4}gk>~ST75fN{3h!@L`rKEKgsa|q>roqT
zjk{)2=3tXPpw~1+FZsNgn$dU|<|nl!d^dNu!6I{v`U;;E{9@3W&+y9|gI=TE{5ogQ
zv$Y1@7;4an&IUD&GboS8p!72gs<qIditx=rFQYcJG-_a&QJn`E<$z`8bG)6f&4~F%
z&6;XdZQgG=n$5$L9Dlb#h1(nSWxMojHDhAeON#;w@`U5r`PZbnU-0>4^wzw4CY@eu
z(jGiNyT`yk?a_DeM*p?gr1jC>s)WZo;iN^ET3eOwvQ<$q;qbRk6==dWc^mvV&0O>t
zQm3d%kJ>k<#&>n90=2+TAAEJ=4_{rl@1t%jeUziG580vA0nu+}&333{ABW}!IWz&k
zQ+R#{{rGmxzh_6!v*RU4WBbT$dR|i%9jA*!K2y+rHaK*;ypR6k_jx55MC0*LP0R1A
z5>?4FiuKj10pt}u_SN?!r>4JfYLnBc+W5}mPdPP}&$nuUOCj0(WQrjd=ons;jAUHL
zx%F!XzO|=#b;{9$)E*BUjI#^Pqz9VIGuWmyTFnAjBrCd(9fmo5ggFQ>%`$kT=0<!i
zYpGQ&qc?!{X8C_LrLIe1omH^K7TAN=4uXXOVWKu?IWEUoaTp%i1-qadoga%vvz^=;
zj`cQ$yuRN4nhC@FuRFeBG@6^Nc<!&^_9R^MZ!-K7i<c=J{V0&n0^1aUZBC)tWb)!T
z|B!n%k6r}!k!K^_+B%e)bzgD_y5hH)jPGU-eF*&Q7v6Ebv<IKfJU>ksz-?>tA${nB
zstU`zW}l(Hl;#ZC3O8Lk_5+RXCpABAi)Dm!*n?8u@tQ|2&Z+p%XSz7YP!AmKQv7&o
znA2RkGn>7LfAbBCT*?gJ)P!%YtU$+sar&ag>|<4eTi$b<hugv2hOjcQHo`LN;F`&-
zLp(Q`$3CFl{0ZOuOKoaBwW;1)=<B976;5qx^g6sXXgA+i(bu#BR)KH2!Z&pn&~r5h
z|1f;>Ybt%-llY$Ta8)Y#!_=m-45uG!Fj`DMj@c7W4SbUT-#m%sbHF#fL-7;0XU<e>
z{K@dmNFP3j8Sk*DUp3-bby8FtzIj-l*Oufx3d2`<@d&{;8#ACE{iHVvzIp#TnJjqx
zHt<aW_~vi;W;T3N6u#LB-#Fo$^YG0`G@K3HlT|1=S$U$8H6tKd)8L!u-edsQV?HN*
z(-6LS0N?C^Z@R!Y{otEv@Xbs3<|=&i6}~Apm<*gwN%DekE;VOal9aXv{o^H*6#WO@
zWB8^keADV)qEc5U>hj`5HGyxgR3!%xzDa~{PQW)k*TXau668H3LCv})XjMvrx<-%z
zSUN#Dilgg{h*!QA@p}6`PNj~K=QkuyOQPab&5zzG_@)DVGbwYN{NBdu<?UD<m>f$T
zJXZh2#p+i}<}oyj)su46sd7@M`ikG?A$rYuvTpXpsL*oq`zE2?^o${ogdVC!F{)M;
zZ%z(sRiC33^-r|Uk$cl0UvZ0h%+rK#%<#?q&}iNAjn<C_(fU#*S`%_cGw;%){trCz
zKIK94^UxROVg4FfejX1qoIRRT!=v8$J(~FqzPT5rbw{G)za~l}Cq}7PclyNPo4)YP
zh4Rc0hi~e`H;3`x#O;pM{CRk721Kd>^QUBq)a@#f{0>B__O}Sd{S%>n@Xb#6rt&y?
z!aL9(?oS3zO;+v*jeQ%gf6s;Ex1}!}zNzdX-_J<Dc=2$x_#URwm&5b~zIh!$ZgEX!
zQ88b$@S#wJC5Ngg+Ra1yrk))o%Wrdtig-eF`CoiCx0$6$?#<oJ!8%ztSR=~>Yuk??
zSP8yiuIB${1!)blr{*4Sud+kht3n56g|}}{pITdmGIwuP;g$-)Lvy4dvyn0|xBEO9
zqaHFZYLZ=F82z~<JrwomrHCZIYYv{nsnnkiHP=<RruZ1<>5&bVHi}&KR$N2+xb;^-
z`WTA2bw4-#P`{YZ@|YRWFY)hfq$bY3F{OZ?8s;Fo?IC-}S(l=o(FdJ}p3VC7`0k*$
z^9p{>8*UZg){E=ZpMN_ww5?M=kJ&XolS3CcM^vxqQ2sl1ZKEc0afF>LU%OI5soR+C
z%KevJPk-7JdC{iqb8X~z*z}XWspV(slNyDOIK--6cy4B+X)H}=)rb3N7{|G$-%Pg6
zTs$u$s6}<KC<Wan!5eRJb&JmYMNd->i>m#Bo`TnA;}dF7x6GOdi>x?oRt~(dY2lN{
z)OZ%5-Mp9$m!Q{VoM6`ev1ZjCXC`+G4>FHc=eb!ccpj!1hu#yv0VY~&R(4iR?srB9
zYJCXp=Zab9;HVVdcNY3eB(<~jZK<KXwd#F7n;ybAo_03nhi!6HpbnSarnPWOk>*xe
z>RFj54(q@$nb2(BQ-><>mY(-3XjG^1NN?xogXu=XHvM3zW3X5MD6>|THLDeT6S&Ew
z5@r+sJ|-1EWmNgOMm6kC{iv2vP4L$ADsI&C^9Hh3sYUHDsMsC@vj7chuo<mpxj|p&
z!Zm{oWY(h9gi)gkGLS7}P|0})84eiax^2)PhfxOtjOr9?R6tkiQ9Y@7^)qVe45JQD
zM!zX&)Zh8hhK}()!3Kp_rM9&Zwpk@rSt~M}8MS>~I+WI+R{0FNZ8RvRFHCjXpe-fP
zd_s)+vfHTF?~OWB7f*5ozMNP5yYqiN4{ht?F6wpsyPdn@REaudn7}xW`Yu(e?ozg5
zF0MtGdC|!!{~}IVs1uEcZ(Q(A{5l_bldaV4wu3pf4w+^-w4;|pJp$;#ui((nd=9OH
zX<A&jt0;BT-Qj48P3_8TC4;GtT?0or^x>F8SspqRNS;B0kB<sb^YRXNsLLvc$_@3=
z`8wn>-h`*V&>OauIbvy?T4Hyq|4^qYo`$=sqBSjW$(r9!!GV7KyU_dd3=OokoB7~w
zS@`>D{l%>i_`oX~KQdZLtI@E;P4tx${^|qMY^R3&4Nc|&Tr&<9F|2?s=AzqBGy0vY
z7e5Ej@kHua!(b2emnYNc6<C0d1LrK^_4l@-&2IBo_swWUYp5^rTot}|X$Kx}j_<k!
z^|LVQRWQpCbeZ3Im4{g?W8j=P>esBZ+@BChe^dxQo3?x=zVC#S&t%4v(+KS*19h(D
zc=e`{8wuO|_Tr_c&`Sx&l?Wm$GMt~cD*X%J{q*w^z2NKVgPTInQ%|m;V4GrfV3&f-
zhq^^>)*f`1gZv+#f@`keBZp^peR8SNTRd4@XDx?;ro%0bhcTCBgiF1L(jPU_r9$J_
zv!=j3v(R(qkxvER3|vATlNwc3_~tDuX6Ju;Oa(Y*F{>5)l9hFn`(NOgYTPb|YtFGM
zuyXKRM;`CL3$B1~3d1*-H`DKgcC%$Ya}d^2$6CV-O!y`RzWKhCK6Ciy&U|{L=AhZk
zMBACl3{CiECbg+pYEwQV$qpP!J!=rPvA(bld~>NY{o_e^Yv3DKB#aV5FL?mRX_cbB
zZn(#RrUTzh^1_eP5GJZk-*{E}#^IY&C8$%uHy85o+H7byf5K4kO$qp>0erLZ3H{&j
zO*Z(Z;(2%mzInbkSp(slEbz_9xy+`T#MfQvmrCS&B4Hu;rUrbI!AL$}ZS)`bCVQb|
zRm+Lr_<NFmzDv@T2h^fYC28n>*ag0MJTXc0;hS=ulQhJH|2TJ&Ze>f7G9>B6clhOF
zqUuqfN_w8CXOEa0j(*cJn#Y>cGu|Xo)nJ>jKgi8_o1g=$6V!Vg{ow=AZ8~v3irN(3
z;ufpoRd#y3Ru7I>Mf97q@XfvBaq2ucP8}oZqjJS5v00oB&^vYZ6xz+sSZ!SrtFGgj
z8Qv#W##nOxS}<p`32aj~R)@03>iMS_o%kn4@n>L|Z87W<F{%XHluV3~uOFOKn;DxW
z@aSZXQRR>LZ0<(OaXeaWHp4h`@e+@S*5%GT4&M}qZ}P%7A4^ATaE@ptfAJ{vKDDWn
z9{Fuz7Q+mWM)jj_D$=7(cIHr3r;jQh{+zE-`rq9s9X$-!td7!|3FH=ci_(L@C}o6i
zLg1ShS)(-VO{D&SZ<g<h)L-yT&Vi9C9~r53wnz=D7OAcIB30@;ec|`vnq$nHS`(oh
z<H_`+NBm^-2+gWR_Fo?SH}9Ao4&OXj8?ILHP08qRy)`qlp(JyueuQb-=`g(+7^W}&
zWc=j{)8Y4_+CM#1Dfo$-*9cXY$JC*=ky{+ZEQVd=+xQ2o&w)S%Fl*tLGf;cJw%1Q`
z`mPpjuN7Yc6x=I7li~u%VWs}$A{!pxaE?-B#glt+vkP<a@FRBrA3d4uRsJiv&RWWT
z!v5vKE7%2(V1I63=c7NO92pl5X0*^NTIZX;>ac%?wWOCLzgzonpzU5DU*ZvaE&I%X
zob*QJ#TyLYlzipK>{9YZMle5U8u=uh>B0SouFhT(#2yp7g==LPr(hcPA+%$Ab@b(P
zaLrX_O0{;W-##Zk8hU(}+O>9rotyzX*HCsfoo841sdnb>+36{=>n%Q%2L5=8E8CU*
zn@uZj*z|fkeafjeeZW&Y^#$JAMOIDfZq-T7ZO59@b6p&NZCd6qP?ySd)S?>rP_E1(
zvv{aQ(Me<zx3MUX!J<=@E$WSiV$W)k_b=*DZ}ANOYgV~yW<5PcFE}2W?bMl$!z!hh
zo3#b)W(C^KPBfdHsb&=)ZPpl8L+-bScRXk~qp4jLTShGk77AN!X0{VL4qx};b`9E2
z(0+Qd&XPexovL^`-W$Fegr+pNs#P<(ST*)fo2nGG=@RGXvrnyheA}v9cy4}#!);zX
zPc3U)W~(}(O}#zDY%F@D%Ad!}v)7{auua%n`rQZ8BUTCT%`>w$!CmDd>20rUR^{W~
zdN2duO#_p9e>ZCFMx*}lHmVLDn^Aulb%Z+6^koM1n{H4O^qAvY3`%89UT;vh*=Q`?
z(P*+6G^>oX{jryxpYzg4FN3UR1FQ|(tTxaC$?Zo2e_QmL7c%OfGDb8`qkL-`Rk8w}
z8~EvKUZc`JFsOS9{M5`K{geu#+xRY&=D;|^_DJ6@N_St2zZHWX)HCQr7lWR{ToHwh
zDjsZ9BCPcAsZkY6npB&*TeY+nvehg)(1!e@OZ3iFwQ1*Nr?OYWLjl`7t>MyvGGv`(
z!FzGbsR7fRdPokzw=HD*WFt3`{`>M9@$mfVBLjW+E%4L)kDAnyY-AI>glTf{*mt`=
zj<wU{W@nz5T@`)oT9eyOZ?R1)`Mqp*-6p@PcD=4+*K(I#i!<4Ei`&_J@0MtXmY?#`
zwx>SKG4|Ewj=t(Whnm$(UsYnZ#tHal|6yiqR76uCE8%B$>Mbt%n-}>hh`+&ZVpds<
zTRG?Q_xTXULBkn_9?}f0<ijAcX5g2hOJE7Oq{JrbMeAXc)o{#WYC$vb?@&WJfZo#u
z4*3R`oQ<cB)CnJOA8JQ1k`sQpI2m4=MTQ=Ar$ko$*&G*rsOm`SUA)i6P&6DnUK#^)
zFVSF1`cS8G^BLfsL#^;9xAWI4j@_F&lsSl5W_*3UCG&F8Y_8at18!lCWizy*>S#*e
z@Z-&JYyMcb*7oAs2e#<~+jNa04-2j-lMinnd(4_s>_0pGm`CrY!>Mpy2QrZa*HrOS
zTKqK+54luy2OcZ-n-{y8F?-IX)5l#h?nJ*Ci#|LE{)u+U7U|N`WIR{>UAjCN4;Gx$
zFBSHk%)d9+W+&&m)OjI(boizOe6tJ<CkejUyaR7BYai<j%u)=V34~vYvBFr1=rn%Z
z&V_3hvKH|CPaaFlYp-nouP?kheDfOZCJlUJhi{fqoBBv?>i3+fh49URMKBF~Q*RFa
zQZuMw!8b`0|LdC?FdFai2=e)cz$OEkH`RwMo*wv#J8?fTMJ?f**?4bW1;bha)TQ8?
zQ+_ZJe6!9Q&8ZpvS`Fc@+IX6)PzNiApBTPbP>8x&9_mzCsc)sH$NUE#;!n(6c!hTc
zz9|6Te7lyc7Vu5G!}yN3CTrL__+>V_%{b;Y^kZgt7yLB|$?6z^Pp2JO$G&9YG{d`7
zGg)WKQ>TJ&{BkBM*N-IafNy#}Ok%EG5?-Dpd{9X`GAT*xhoawfNs<-5vBEcN;G2WL
zs6pLHWcDq6;)fEodKvxU=r*n9Co1ctMD-b#sFVTZ0QRMSyce(SmZ-Hm$oU&jCeDBa
z6@hOSL?&o!tpxqU461`u<K>2LYPE@1PGh{P(L)tAG)~pgZ@N>T`jhP9i^b#A_arkk
zx0Bnqn7Y)sSRLsVt7p-8Z<@!FafXgl8n1Cy?th5Up*t}mfLNV2F>7jC3|bbosQ4Iq
zVq(<3Dmj0D#V}_jMg!kP%lme;_8etq_$IRZ=8*9R-}u5eYeS;d&WFCL`sCukH<PnR
zYxHN2`u&5q_=JaiYmYuoV>Uw{@^8YK&tUWDaaE6s=k;jt7xMh>MCr|;C~aL8rPA<?
z4}22@-wZQQn<^KjIQS+#d=rJgxD<TzXf89S;G3mUk;?9%hpIaL;`w>*M}*2fh)`Af
zr+Tl2b0*+1?nI7p%LtXNjmJ1|gtok=E=7H65qx8ZZ^pnkE3C}dED(+_E=*b9gsRdo
zvTO>5$e1xid)@@A#?4^u*&3n}2SQZwM2Kb#3DJa?!P=NW-c7+EH03~Du?DI-^`?LD
z)3_V8moX|p6PcrPcvfq5NFtYpEWX0m>9u&o{shlt=lZKU+H)A!WWQ(Fb;j$b>Fi&N
z{F&o}b`0Z8IfveSmi`GCry%_jD;qRd;ZN*o^U2ma#vbFs0}RXLzk+9v{bl<_`W3GF
zX+dVU%JcZ9C+xZR=u?=)zQ7)QBCkue{Ky-LqR0C(eFy9#Ke;xl-iJM6xl{iX#Jd-V
z?}6ucJ2}TZ<lI6(^v2P4ekbkfFoPKe+;5j^SIz*tMse<#Eb0JNV4B4?Z5R#LbhD}H
zPMf^W+f;gujUFl<L+j|V)2a<)t@^VAS$-}4t4&1~qVM_(b(!mA%I&vk(=rR$h88XC
z%Q-Nf%)aI@Ph<3r3KlglNPiJtm|j27RbHCuImQEw7r4qXyu+|aHJGF}`iuV(SYeJ?
zu~T6cG@H*O$@v?G=Xf~1Q{1lx<IG1Zc}dOcHfu3l5eWx%hKX$GI;&X+VV{z4&efx4
z#lTSe(Hve-8!JICRqY`b<%7?zQ(tS;!>Zu}tok*N`Wa02W2!}QOVDM|XbQtNqtC%F
zXi;<EwsAY~L~pg|LVAm4l(wk!Qfh5|EIL%4OrED^t>b-{hVfa-n{{g?nSS_*XH+!l
z?RleW#TnJ6HgzaGHapVby*Xo$yT3tpyFo{581UDkpTISHsv892Xm|~&MOLKEFJ5Z&
zua_#G@?s8{mnxMq=#$N$yl6Q2P8rnciQ)hIt>)uP&Qsc`tX0ulSOd%Azu~>czTx=e
z49qN|zEu>KdMx#riEcAn_~WIryQQlaXq$N@b;w~*V?N84Sc8h7AsMNe8R47x6OGFG
zH^-vRHsqPN(p%B$4xodT@KN$DANs|aXK|hWrD&&K_i}32P5PbQlAB0vD#s-!{%WU&
zv~%izFMP?l^i{tdzVdJ7!|yqs?u+DJU$8R2%`P8!<pRHxPA6KTFRKN!H(KKd^Fw1S
zWml!Nc2#{!Kil8bOpn-f@H8_zuG*CIKDEX_;i7kVbgt2d)xbwpNBAhsWiq9+_-d({
zJj7AHdc<dpuHjViVfcU!I&~Tb%gnj>9_Q)SMnCPC#NXLIKLwU2SIQs%@ksI&&bYNR
zC)sx}!^tFn^@UZMcHnQ9zxO#X$yaJgUHE&S3!}W6N-q+8(hasL2iu&3Ln8g~&a@$~
z244BCxeTY))VI68GW4gWHH7S$;q-B_HgbP+UwW_Lpt=$0I(*-irtBwm(QQ~o8gk9V
z*G#g~V0fpE1D%G~EJdpkD;%zQ$YX!<oDr@W;{{vQho1^l9~<LV&Cz&=hqyJXlUwQH
zn61Fix2K9*ABwuw<b@wu<jhdmjPH7ZpOVIME!78}iSa{=^V8O{WcEFWUH7^4ZmmoH
ztI0m!M9$*@mr8AOk(I}OGXQT*f=jnUTq+XkQh)X!XFsw82Es6CIn$|8{g}x88F-%N
z@EZ7L^<tOy^1TlDCKr6u9<FK1%FVh2uf(xB!!89_{;cz?PVh_~*ybc`bBnL<q4}(2
z1@PJ@)TRb)qQ45hF;JVDk9Jc8z8MDJY+R0i2fpbG-#Fo$m$T7yW~Aug6c}X!U#G%7
zqwp6~n<@z3v>(8n2KZ(yzT$BBrY(GPCx$#5_@)|s)4d&iRV~qx;G4B}J|lb+2;cN>
zz-O$*OilRaRarEj;&4wvj*%NbPZl(!bkxeeC#&E`GW%YUJq+J$y@ytF3C|66sZN`d
zHH(_myXoW<quUrq!azNl!_WbZCoWm1L&?f%ovfb@a&j6c>s}2wryTrKgg)_H$+}8?
zst0^?`(ctgokpA4OAplsw43=!sx%z`O&@&39g@^J04*mgkEOwne4aU*+Y>eD3^S(I
zkWXATL7B27Xv#BYWNwVt?4|I^!gzg{8Lv6*ScZ7ra>i@w<ajk}oFEUisX{J#!_jYM
zz&8!xn@ITPfjM4JYR1z$9k2gAkJHtgamw3)S(~NewDl47sBQFqFT!gxHdgt2#_F{v
zRs;QGwWSgBs7l3Zcjj0Hy^B%*n=#C|jZv;OXf|VGR6K#KKXPyU8{o;odvms6jMDrW
zqff8lo14+9eS{j+hG^}a9j)oZVVh3T`V}0l5f0{3)r(e<Qp|41hUfT`N4M|MGj-gf
zunlD2O!cT^Z;uq_QD2KkNmV?$m)oO0pQH5gHoju`COv#JXgs;aU701`K1$z3z8`#Z
zH%pXyzou8}QY85v|M87`P^4~nBGuI=QkI%{j0;4n>dy%IJ&e%M6A`+yE`m8j5$e+=
zLaD7HR1&^%<U_}SZ%UsImupeD(sd74n{d8{Z<?l})m#YG)tC@f2@KX@=0+uM4ALk(
z!GF9C(!!TP%Je8mdscBfA&}W<fr<$W(7T)g8bvPZAGUTnd7jzE*O&)L*2Y_A>}3yb
zsoczu`5xU|Gtgf?qZOA#2kydtm4*9;j$Bu<CoSNb?fN+SAtrIX1n*qq8gI`^`WOze
zx9#CEwC5I^>1W{DtYj*C9oJQZXOk=Zr<)uHKV>?@?4SRUC40zEVWa(2g;`K(>3bbF
z(5VD^VRP<9-{(BB_J~uh|Dv}NcJXfM(meL)op?N3Mmm+H5Bqmk=1j8BAEw^)+Ky+I
zl_|llEIsU6G1#uSA@rj3<hq#iM^TGi$2eEK-E3371vdE(rw-MDnN9&V&2DDXx;%J_
z-{7U?8a-$~ew4*lg$yIBIL=CbKKUp$tXiGNO6IRcAK-x6TP(VYb~6)xsGLH6ZhQK{
zZTN^AQjaQ&_az^UlL5YYgU0d1jLwfA7=5K9wW;%as3TFEI<gAQWg)#(GtDXt!@NVY
z`Fp5YhX!&x7`Ea5{1JRV8crWp4!Gt%x7*;LMy$*5&UARk4jY|=aYk^wYn#z;9+;`4
znsp<uMLu=Wm(W}~R<$rk(xSWmXibUqdci8SQ!Uy&huoYM<SSEm%E|q29y~<@@J3U2
zyU+#giQ{_;;j_Kqty+`4HQk9XH?6mdZZYX*vPpKUNmuBTI#$i3nT1U%(#@zt^gH#!
z<NFm3*$Q)HSb#QzMzhOgP|IQlwYw)3gEwfIB$HCgSxWNHBt7`xrStc_WIo~rb9yOd
zjh6<DlNRI4IXT|I>|FyHRR#s+GHQECqq0{r%2U~>{M43Kp%cAaY*2C@gF^E2`j1k?
zHZQd};ib`2#GWVB+$lw$lE&U;lx7x#W>+-GC)}WQlMK55&cNR=^|Y}@72Ro6`!7b_
z$zxW1n6=GrvKdW2YW%=Qu5G^PaP(t)m_6Flsk~@8^PW5LPcrjkwNv>zkPlsg-`f^w
zH_YquA*VTxjHSEptm>RZE$}2Yz{g|{mbWvzhq|Z(-xqq#8Mr1ljB^0?Zc-9I*ca-h
zdu^IYKVP-OHvKq*Ck(#H_Qs~XckLRx&!GzGe0198qh`~6)c(AWGLykwNWNUp_%g4N
z9G~+14aGU-I`I25ac*tzQrf95rTf!Qi&K~fxz<lbvQQgt;8wNHc=>md9rX*&xyEdQ
zFK9aD;TCk1>iztcW+>M${JsAVuG!CR3-*B9cw~k|!z}(}(U{m5V2#c&M>Ss<1bw9r
zn#+FpWn3KDe92rJv47NI-GiM<!ap<7Vq9>`J-+W1dqj8MFSRV*<Z{e|sz|O?RdW4m
z;rVIIV@6oY3O}*lbHDi?yw`iT);x4;*$TIQ%-|YnJbTPwx9awz=ADdpI-cAJuBoau
zq>rqkTbByCwcsw<j3=3Kxx`QXCNuYPu%B9Yrd9>>eXYi9g<{P9rnmk&+RZogmb)-X
zw^isatjTCFuhC_;cc<pWHC2-!m%j3EI=r2Wxd8Bxo!U9;I`_Q;T`C!mPmq7_YYTWB
zzUj)ba>F+_S%1PcPqvdg%<9Z4&dSZ&$65u$)PZLP@tSO`9Xytcden-oa0FcQdmck)
z)`gAC-$c9VvX-1;*k;_S6qSZ=y2CdH_{Iv~<brQj!#9;@pxsO%#|FOnFa|&HNa|z5
z==&awx2HcnT)k6txEnREj`WNt(EkMAybI?u1mWLl3uCpwd*ez`A^2vn39nC6K7V~U
zs3xDcGG1l)=2|gyB=}}oE_9^Kd=~hoK72Fm0~t5)O;-Fjw=R=kb0k@fb|%ZQ6c5h4
zWPOHjye22pGnuSAJ(86Xz8MsoEEjw;w-r1C-^9W<ZusU>*<`&aOim7b6O)0AoDWIz
zew?KGm+%;)-_(Y0)-OuZzticZ>Yt><j!C-UA@|0aq)Zu;RQzb7o~}=n^CWCT?dfab
z1daR<ufl)hq3IQ`iAnLg=80FOuy_q`8?X6Z@#<VHUS<55#gH1W0Q@!yL*vzFP&~6N
z<8`tRk9CUIn)31L#GK7d*U7!1PBrHgv#8d`Q7d9*LzXz5Jk4wd^qaQmH&uGXl1&(^
z^X^y`gl}e)h*dyFyvJ{1l=T`t;ml<);l0T{B1Ze7@f-WZsByg*<tqpO<YUg}A7tXZ
zWM<~|XstgKO&x&#saf=j52HsMzOleJ>+I1AtQ)PyCF!ezZ-U{QVem~M_~!b0<}yt2
z=x$Fmoe*YgdV6%Gl1FoM;lcTYW&_`J`CpWxR-)sKgLU8=1AK!zr9)+-nAI7jZ?7UX
z5WblW-(;gc^$@<f9v!KLzRYQ;6{&9p>81Kbztq3X3qKXXyyXbpnMBTC*9a|wZ%pMP
z<o`#67T*k4vpwO;HlF!Yf#mxY4cE}|%+V|vru&ma^o*KQ#!~b)F%zo5v-TS73DUl1
zL3(>Eh%7Yvz3J<2k2g4BOMvd*X-7?+SvGERQk`v8J#Smp%hg;_=cq58WKZE*D-tHj
z&wjLmYqEuK&Ani9X}JCJdvPuI1il$HlIy9l=)&xCx7qL3P2jo-j>)l*{b?;ezs2ay
z_3<fwb}L&MdNT8oOToTVomy1PCiHRkkW-IcI=ha(gAY#fNu68^GS|AF6Av|+yJUh0
zUt}sC;j4$83O&!fYi12(9_`e{#_&yZUsZVMqsT~yYM!G8)yb|hoEJ86&d6r9D>c!s
z+HlR6j&>dQ;HhnBmvNU(Tk#H$Uue^UaW)O^Vbd6VHg_G=rK;JOA81n<>Nev}lS#Z4
z&1RNW+xyWwMX&T6`lTW&;bX~WRq2-&wLD{yag#-_W>~an5IKI-rY^NZui;uhQ*E+t
zidZx_J9#*0Hz{Z-pQ#!Bx{i;7+SK&})S`Bxw`?#gE1FBUc`(Yf|CmLvOcmIsXK#3=
zt62j(n^mGa^Qq#@Xr*X6W9W|>%k!|$!cqTyy^OE#!aB`hox{KN9O_+N@f=@#Nyg5f
z_+?<6>u5P!i&^xbibXd}eD+{;nocmxV0wclp*z7f{rFi*cBS{b3EET#i>xVTEusF@
z!-9`FvsrUbd28u7Z|$k?jklkf3rkF@m|#+fH$C96&4w~2RiIaDFW#6(GHO^wqe@dt
zN_`1)jG(p@f|rJx)7(GlkvcD>j+8RGq@lHl9x954^T|umPrTIml9#F<^wRi^Udp@F
zOHp&Y6f(pMoy<#HSw1jP{9CEwMuWx_HL^Dt)whUI^QbXRNrzwgr8L#xrLP^lm`&oP
zD(k&;_OzFRW=J{k8lOEO9lt4!e=oH!ZqV~vlKC6jPJI~1%cx4D@j-7PI|m&q_dTPU
z!IPJ#lVc<w6~(KbV(``7i@q8J!|aW5k`?PzC1!uzya7W@@sT?DFdq?L4c?f{X&w5!
zikecGO^fNfi;1`D=m@J)|FtS@S$;no*foRS*X(e486&!4U38gBuuXPy6}Q4O{LXgT
zXJg)<P5a=RoBzP%Uv0A9K;PNwP%rXywzc%pFy1%(jE@>+^3_H9(Z3J!)sqXp+LPaj
zFVv~5vz*Mvpk~s_rJQ3q-+rQpCz$&0JkIfGrq&|Vr~=%~VMOOxjA!LIeMN8G+V~ee
zQ`C&waBcH1dqeqQ=q;nEJ@NN{6XwYYr`&9XU$_YzRE=w#%JeAJK+kCa6A5oJEK&}B
zN#HdZ1DI<G@BG2q)|OgSbM_G%xqPt8TsUS|3HG3Z@Jn92Jgidr{WZD}wJ#nIse-?z
z9(oO$&N^6VAMbPb6Eo$H;u+rV*0@D(bw{^3*59q9j&4ne!ei6Mt*Z^uavI>DuHjbC
zGH%tNHs0hSeG0IRZ8|ffhI5V8?Y~)M;f<KbS`KgVPjZG&GplZ+ONMzaWrs}`!zS-w
zljn2MY*Jl{OQK&6-6nq<^y3y>b2+F_dBZJo=~**sR86T{@>r@5Jyrc&DmRj2j$w8y
z-`kM;tKpm9Tr-0e!79Ky2)|5a4P<5E>w-JT%Hwt#>))+-ep$z<M|EK>Mz<*o+qhUY
z&~9d--5BAU4Dii~6)+2Y^L7cfB)r9a=E5v9;f`tW&SbQe@i5Y8w3-okXNJNogW#op
z^o{qzuhR{Ffp5maH_M`7uQ2?<f%HMOg}30F_fB4K!?R<ApWvI&`qZ)Do1*Z|?lS06
z@Xh=J96u*K1>d;-fWyf4`{z9~zh9!sz&Ah7By06ydY`t@Ck5YZhHoCjH>buWE36;2
zr*6qyH{lzOfq_EE#c7FG#}5BALBFXEAC;k}3ch)jlm2n|W-fda^&D@_J+l7(W-b+c
zvl_m60^bC~H;uZHlLOy8^uddhH%VJhCTi|Z>Qkrbfr4)~!Z&XCCd&kvrDwddCZgHE
zH`U;qmFPF`eB$-6RJ^8Ni_`oQ<l7u&4)~!sJ=__mr7OtsTOLOZmparS=7js>x50l?
zgU8DshHuFGYt|<Y9WzcJ3&d&Lvsi8Zo7tK%v8qgM>Ic2y59`7`MVTS~C$q(0#VGtz
zjLPka(al9@H00h)h(xooF*h9VO_9<u8kL*un_p-*&&j;G3cvi1T;g@nx;c|s4De0c
zj`(eYn5$`{kE%|zf=b{!&Pu*9eA5@cDGc9igKsuZ_NYM*j}`{uzcJ!9u1F?MPBQ*J
zMk(}Gl+OK4ooYptI>0xJyU-&Z5Tz4dQK|^v{GB;U1L2!m@Xc8G#tq+e8yv|zZhEMk
z%nh$i)?cAWHBEyz=W&FR&oGN&6O1#NdEwNj&efw=yl8|9eMPUi7B17KaINY_rk_u^
z4zCE)U!%h0Qy`4_iXkeO5+n<K+gURNYQ~%PYICu@V#_c)s#uV$8v<22fy|nxWYstV
zG~!S@Js#AKoawfD)ugSCXCq$=wkf=e3{beHbZ6>N$>f^Tn@~Cmy#}U9fNu)3C)udU
zf8n{?eeeX1V824YF~c`LgP0W!3++O4UQIpz$0BB5a?R%Y;8yWHTuY@g4|=AXJahJ)
z=hVFCkO7*P{F)o&?y(mvdFoV=MNTCSajHY26F;R>Gtg)@#xgr;EItMpr|5pC-t+vX
zrcUJ{pY>Q3Uk#;Ju$+v(94qXy)u-1RwwYd>9@|=+Pt0~5Y>##mj)$Q%th3Ih3@h*q
z;~maD!KN(KpccfD-`CQnkS2JFOWN@B*!2FMRcj7Yi&|sVvx#_9y5YBIZ`DuyDLKkm
zwJ?K~`EYOm{WM3{<1L;>4F|1bF14xVZJ42ew|I07yf=lo=Ff_U<*iu(x8WLen+f!C
zcl#T?X0MqU2Ka2&Fk4_L>;sd;qRYgOL??l5cJ`pQgl-ZJ(-cic*WuP4!)sY<qv*Mc
zg@HPmwHF2o97=77I@Ylf)VW}vZ>-E5?+iN588}K;(0%aYH2G;}4gopKcyfxN=h)rM
z`ivzbr#HUn;XDT8?4&OB(TQi8*Oj4O7UzMVnwzz(vRN6@;AKAK4NuZ5^@=>;-&uL|
z7sYooX+jf|ykVOS=}q!nM)RP?Q>+nLd&P`O_k$j&69#R=H=G=3P+jUl)9}O$-%6Qb
zi12KS*%{KS^iuYBUdljiD(7)Dm%U!9wbDy<V4G7Dy|lfXmvX=>DV@Fansts<;hLAm
zf)f`aq}n~CCA;wM+?Lvv^^&oPmkcRh3LoU9^ozVW4|vIa#Y<n7OGS>uGI)$<r_fKe
zQYwufC*Y0rvVZ~I*`Q6G(VaHo89IhObb<c$`$p~k!n_P>dd45HZWa3K{^j>5Ke<T<
z=)t3(`~O%v>-Z+`w+rL$?l#=rlan@W(k5wwwooYbq?VR3T#LI5cXxMp8!%uv4A<fA
z{$6|c`{Vmbovv%2T+cb@KKBarnIcyL6igO47r%4+=}CN<#j3l1(67kvPn<9Lt!B)o
z#aMLhnMH;0nriZUxvpt|=9<Zb83WXXmp{}E&}C-Z(vG4tdZATC@k=Lh!tqbrTwovC
zOFEVh2WY5opc-utlqnNC)ND4{hS?N$*ruK6H**@YcLTohIbzrFbnNIbvQw46sr;KA
z%E0q((;>_>-$QQ*fNN6OS$d3_=r2x{^>ryE)TIei;hTNT7CmAATRwXF>ax436`YZX
zUW;DyCY3#woXcIw>A*c>&}#;A57QPlu@oYQT%0^HuV;rzcJaB}o3zjjSmz8Mcg-JN
zhf~Ry_rofA>ft$6f|vN5xw+}E$xK!_9l4)0EmS8hvsoGNy>l>^RRXSoX=3?$8{wR;
zKV3?C?$R`_FNM~)(4<_dJCtl+5100NT>8}3r8)sF<*AJxRo$gq<z4Dog4|zjmqssT
z26LuU6$d$0pbI&`IHyYSvxunU)V&hy`hE?|?4idIrfD-d2o|8DuP1v~e+wdqKu#F8
zX&!@5Mb>Y3OIXHEw#LGYxDRjRdUn1kKikH5fmS{r#@WF8Q#n8Un6HC@TKm!ii58s<
zw>UX@IJ>qnH^nK$xv`c0K8_DBr{R5%c{zepAH8NcY;zs9`2()GyCFrX>tPMpX2)u{
z1Gedcc9R3XX#v-)fNQq#OWqc?u}#C%Lc6Iy2`2d+{uzY^Gn|g(A<R7uB!>gvRD*9i
zcVo{(XLdp)z*Vu#Siv_#;G6ZqbnLVwlk6b>1K)IoZ}v4~uR{Z}L$$~nSHZ`EZx)qc
zhP)uNRq#zz7WR~<C6DtZSykYhlJHIY$N%kR$PM2FZf7SIeB*;ZwSFr4%+O>l=|lHT
zH}o9%=Cd2ErbDtWz&G{an{v&QwGjQL7<@CaNV0x`ZvyeBuE95F;G1oC>GQizx8D}B
z#Ea>>fp4<FH+Pb~suk&_AB2A6diYhj*zG|6CdaxY=981y`^$dttckMWPc7`1prP>1
zmzV_o)jB~f?Fl;VpP)PFGP^IwD}HsnE-s2!+S&1%)`$7;q<HlvOFXhuyc&=rF322s
zFgIYKywCD$ycV#d;eOkAHEM%b)jVE)tKw9=Mx5TIWmf7H4D%Qr=8;D??t4`Flt<4`
zc=Yg~NB%oKIzFEr4Ff#t9_~@B*`t73WQt3(uOS!R#oyh^{lu+pm)yE|(5=2}-Fh|y
zwi$|U)0s?82j;3QuuWaJx|gB{C!1UO;G2{1O)vPS^akdqCdDeOd#qN5#wuCtW~dme
z{JCN^?|qCmT#u12eA5ZONe|ytf^YJ|H|5}))L&v$4Zg_+-@LvM&3z@=8~7&ukZ5&^
z!=DOb=VrZVO)N^@IBm3+J&sb?IcB9cN3olpuD@1MYJ)dbI$xCPeu|WZ-3;4SM53wi
zGJNxFS%mVyH`9tT7ar9~UuWPyl|!S+&_UfFhw9j$p$cCas-Leq=oLFLQ|RJ*@wB~8
zwrH=R+uLc+eD>bNhG=m45Oz(_QM!ZcP`?&>*p0_7Ei{RHvdQqxq39O!g>CY6Y@v%a
zxi_oBwdtr!1Fqm{-y_3^ew-t+g$_EncJVQD`fz_Y5%w8Q)(<}UhkM&I$qu<<x%P0+
zbn7iy;k{f(%Ao<Lp+n-&ApKL9eh99+EoV5C2_5~fHtd!5avg?c2GSwoj}LV_lAS=@
zhgO4a?vvLXM20dAeQWtH*fc9E`vk}hXQ@SAxHR2pX!p6{n95{p3c@kBIR#*v3v;Y`
zIL)eWW338HwW?GKuZQwJf2&5K-BihM<?qBok2*asd+D{oo9Z@-nW7YSGPK68YHZQD
z5*EEm!`$?J_ADI3=UIbpGljXLesF=uU)@`?FUf>Xf<{uWAlySPwd^xLRlLre)M;j+
z$lOG1rmGLODYVf~lU9(+LA$AkFIAYlO=4d%$8gOPbd&6fe)5B5ydC{iAPjGdm!mkJ
z$h%D9ahsR-qw5qIK>h_LYRVbI=Z_qO)`OPwZ<ePkjFWJJuX}}@%wxU|e$~2c?7sR0
zH}QSmieSeAdQA@WrwOC|RSk{jNo9Ps8~9{s7WMge2l$d#F2#9emUXjPlS0jkdch3!
z9$z(CNIu2pD<eM6>U6#;a-WPZ-qd3!eZypL_Tx=Wc~5rukV#j`-rNddUa5vjzW3mZ
zby89XbQ6PguDsMWjkMsoQ9CXfHT5Vu&JOrxfl<~;oNlnmuSVU8GUB`#Rjz?iy{Z^>
zun^u=CZjT^N88D0R9tqnot#D;D#hpXF}J-&J??MR{>eriTVd3t{YK^<jar(Ed8$VI
zyI9Z5=1Ohxs}?+Cv?9Ani89ek;X{s=&YZ*OPk%A5b=gO!F8C-1|9vS<==cfpRqbeB
z&F<-|elXtXqrPgC&a6g0W~F~(Rtq<MMT_CoZ~Xr(L9cZaf9<pJGv;R<*3hD4JkSS!
zTG3zWTg0ELN*=5Pvr-4pa*hV$lSW(hk$!?7L#;YF1@2z<ALnd7Xw~?z0L@}<ye+*s
z57BR?;ZN1_XJ^<DX2y@(w2!{O-x}F9pu1g;ee5m$PA5(i2QwAqg2=0#|Kw2NPUJy$
zvbUrX`b~;c$5%Mj<O+Y6S=d_}<dR0i1e;tM$=`TTHul$+XU+*7W*L9$<Ir!jk+(U-
zImPQQ^0v^u`uyF)CFg$7s|lCf%0$*C7kL<1XLTvqrvmp!uugh(oy)bkPvTTS>*)m7
z<VKs>T!4G1tn>`?`LjRZE4`-6>bXk?p3#Z*0`29ai#bGmu`=Y0tKe7hb>?#K^yMHu
zHhiC57NQIDJ)iAIr$ij{U%@Ws^||K2HZyX$G^Y|XQzh}<3c8d(2Xj@hP3lnk`x2ds
z=?L4jWN&6gG@Qb8;1qJ|2-l@z_)_a;FxP<A@}w)d8@T31XJ+eRoBJO0FZ`(GPW0yh
zd@DGnKg^Qbh)x6Bq&6adQ=hE4KlAv!?=P4qmUEg@k&pSy#QYT>Ujuf@!ueOXiQ;^K
zVJ2}Jb8c<{zB$c#dm~=I%FC&+jUVR&Y}1P~8ou#^Z#Ka;7h#)2E6M7>HX#eyv56<;
z2j3KeZz{t#U#5};p2F9eNLB{EX^eI=9&f5PeA5BGx!VsG>P5b%E8jy3o)UbM6~1{J
zj-CVG)D0$I4By<h^L<!S^w9((!8h&(@KY^xANXc*Iam$8=}<65qjHgp&J5e6p?C0e
zvO2xRV|otTJjA0q!#>R;@Xp?3Rb5G@crLTx6UpKXgjwL5k6oD+Pr=h7TRacGnFZfO
z1;Rh@O{KcYY7XCAElmGkF4!g=uYd4r<WsMT-ou}QZ+cwv>eVtYJ?36T_h3(h*Q>D+
zUU{rub#0DcRn^O^bdrqElhk8Tl8UiAJY|2P2Kvz7cP~MmdnTxQEPJK!rgos+EP-z<
z@Xh=yunK&$6256Mi(F0bctym=v#XQ6Qg~3luuP7;=r&*CH10v1A}=!!bs<ibm!i!q
zh|`?k<CIV#PL8Z}6>r0LTH?{Ed1Ps(<2OzAsMjcR!-JTc;=D`tsBbKJVvk2X+I!U8
z=ux(69tD+PPeTrm(tUO7{zJF&Tx9RU0p`Tlxb=0qThoTPRS3TM5bD+ef43UfakKBv
ztrYmC4SbVIzS#fISo#EFnPrMq?k?ns+tGPrh*j<K>}bfr{;78{DgoadKS<tYIjl1-
zMw@>_$7vU%I0J7l$IN(^81lT)+6CX7hHs4XqctAB`3JtK>x|Z3_$F_$Xx)NuYCVlo
z&WllcwIE87ztVx@h|<>TQCglWiXB9e`r{ONo8^%z?TOU%fJmK&Z%W}$MRcOmuXKd^
zm+2(W*p6EADO4A3hN{H&P~9LiTzyoiDkg<$M}<&%9=6wx&h5!$w^R1nA-W$KqI7su
z)qXI`u#J1B!FcEJ&2;#t8hkSkz9|CV%tcdP!S&<^*QXA<nce1^a{RhWf1n$Oa&79%
zJykI~_h0SMeYob$?ZrJ@D16ch&#4mHcodJV9rE4bB=EqFmJv?n=Kd&PC9{@OgSgk@
z+OdGWmt)B-_&YS^q+L;~?Hae%&fY(}PXB4wi)A)NB%!}gwrTY(n;sQnhZH%=C1Eyo
z;$HSX_r19*1@LiJ4VYz>gXe^)<E^SclFR}<ef`F<TM@1qM&4#!Klp~s&Czi3!w$M%
z8d>Foc5^S2Ro$LiRP+ozzFRD+GTWk>gDjfnv1psaqTO{YT2_d8s;|sT-6CInn5@l8
zf90IWE~(yh-H};o--7w6=45eTghP4#wTkDrTu;zb@TJ=Q>8Ejf&{x)zr&&Z#&2nB}
z=BLARn4OwH&IT@7fo{{QGu_4s?36+~Ngqx=Cz!lW5dM>y`70B9s$6I_ZT#fuNM|t`
zPBsLPD0G~Gz3{Dg{X8Dl0^ZgXJtsZv)L<L$TTKVxMdr#Mk#9c5*L+B4&QBN%&RTE;
zZ|stv?DzS%`F`ubV=p(L{nTPtfdkE_IPCS_Og4mG-!f*M{l%<`W7sKG%~vOXGDH2|
zM<ZYRX!Ii=EkgIW6zxMk!bdjdnJVODj*6_z;oT-3>PL@_0e_0j%QC#7p35Za9o42G
zky5;=LF9^8KQ!vYS)&3E!6aK@ooTR07o(ab7}-&de$xrv#$be<jNHS)F1d|*0N13X
zGiq8|qyGMvbF%RMyhin|Xw>HBMunl_oF9#^wZKST+Q{A;qe|g}HLr|6#hT0hgQdI7
zjdwmPHUB8pDNlBXoKVpX%u*eoC-5X%)EW8&=lbZ*3?H3c?W2rv&xDtBBKPuDrtx@m
z>wWbslUdpL**t7X*I|FNa?dks;0u0+`N(ZIVuw7rEK^5j%=ulhe?cGYgLYVi`MsL#
z2f>q?!(7xTBb`JRIwkQ;i~q*&<6wR_C*h~g=J#{0RVVPRhD{`+TrE%yq5?H^BV9OY
zY#JP3Q&EJh{3mSM@x!L@`ecilO~1Rrt}gHFdQ{Irmk3#W{?2?~Iy4JE<pI7+hqO)=
z3U%_gM~;6xy9SQ2GaDZGy_!oi1K8=%0FNpVMp*2U0sp8e*MeqT7anph9AVy(`<$Bp
z@P4?Y@B{KX_gyNF=JOst8TW;Ip!D?Ez(>Pipr^3ToMNyII?O7xnk8I6YT-e3%}(YR
zoo4b!GBhwvOTNyxn|MuE(P1vbTl^cHZ@bi(kBKPC=Rb03=^mE`ZpVLG&#b~qbeZWc
zGBhr&P9|$(gL@3@aH;B2-oh@u_(5j)s#8mgk;}=)Ow|PPMkASFU{<P0Tc?UzoEl%1
z8LCX||GvRBX&V_C*k<PhSY;^lQN8e|V4A(~OoL?R91>v@H`(AeFbnK5A3df%+>!}h
zEN?Rys3Ei8^~l!L4pIj6oa4N`5btlx*~}>l-;Ck>jgR#Pb~(=J#mUYYvz1vS&SY4o
zC}$BYlbgpuoP3<zydH*jvmLg%1KVtXZNgU356t-j+q{Hre3p~xUBF(dd3+7{=J_l#
z$MDUrsbqB~lOcw0y1+KwhvRh(#iJUGkCn>o6MPfXi;Qzu_9`Uv7@wk^F?9ODH)})j
zp4y_{v|x|8oxPdx%^>*ZLQ`0)KApxj@y9ANI|bhiF3!wW0Wvu`VYE!-d448q-=}1S
zzG0sN-qiiS*+2DHvYsAd@4_BbgH_B`&1078clK$*H?`rLwCFdz;G1&r%_R6{VT)vq
zv(SZONLIB5>>w{kR=E&4<s8Wx0pC1^Z~lUBy8e%E#=|$Im*Y)Mg?r$eX!xcOe3Q$9
zXVuKBZ}3fS_$Kvbl15!i(ro6YGShR@ZDyi^V-vOE205Ir3G9POQ1=E2@>Wey%C~qO
z$8V~G?=%6fDc_YO3fbZQ@XcoU<}dhW3VgFK54$u!$FVmdPW|r1Y4|PXr7p(dW05JI
z9;eJh;?%!%ocd&<*Kdo5J#)-U&G698?crzOL2vSCT~D&a33L@l;!Sn%s8Cyvk{i&8
zQ_-W9Makb}^QiG>x8^^f*XBI4;rrbRUF}vD_~!axys8wp>cKbT{m^u3v2(M8TW_<%
zKk&^9_~tqK&B#^k4xbpSDV<~4&mOC?@Qt@ztp3a%t8(zol4~*Yz&9J<n?dkR3;5<>
zNQ`DRk5L-<=4$2`<$-TT!Z#(DpUSi#T1AFNt7$x0WB8_NeKN-IjV)cYc0P+zy-QK_
z#E~zKjZ*ypJSu#t@W+wl+&KFpRUE#Fi;C3V29bKOJVJkrj!@Os5vpC23{Lq7wYn6h
zMNwgT_(v$(W~kQ94OPSb^wk81>KEp|e?Qk=75cW<l`SF4Ju*Zc>V)VH{d+Or(24(|
z=W_r)c*1}4oqjF;*Kgj0-E@u7XSio7xz@#XoNL}Eu6?=jpSV7m;hSUJgQX25V~D?Q
z3nnud&d!|r4y7;Y(9&`ax(XbcAILQoPwG3@mGeWKYJQo_%@neGJ<xNrGaogMuGInL
zYZ|e~`7rY)Bk7x;%$#dH_n<4tbVl0b3A8B#TKxTzbimdQlz9UkHc|A1FmLjEh5&t9
zN+$_ilNr6H(P+B$x%V}r)12>bRWOfLc)Zoas_e-yPlQ!(Td)_L=aUOntnA3-^S;vg
zf;V*xZz>bs)Y3^7W$sPiNw`IG$=;N$#@taJ-v6HGs;lI0_p?WH8Ts3B?Dp<K4kr@U
z38F8!F*8PFi$CY$e*FtOGGU5EctpdG;cx7ON7is(4|fcLX)LLJ`qmA71ui)T*Q`LZ
z>BSiV+muJA=@3Fr7=LJ5089kiOmB|2Wx$6*Q%T4BF7Yw1`22s*R+#83{?#-dqmszq
zz)JT=vi|}0DKwk;?HzpFO1O&059m3C7SIQ{mhWMspWfqR9Y2V+0}D-pZF(k?4Ju2|
z;4QlE@T97dDgHaZSx$C09bE3Kmu_DbuH>sHpM2El1+z=A&7Ze@bUl%snH_v&ZtTNO
zUVNR*^vN*qlxr)D)6Jx8jZ8{{Z|345{X{pp&`c^(MskuX9`_bq=9*D`$jSUeu6WmK
zqk?-Ib;4~_=WwG|1R2$+n2}vy23`1I(3z6(4qDC84Db+a^YW)bU(&!ay!-{m$qD~t
z#-q9wM}B88Ul&%o&e!aJ+^DvXV4Wh;@<3^IXXy($XET``!*e_`zCSY=q3vV&z8CP{
zw3!_9d^DV?a1ObgO{;v=5bk;P%114_&<hCb&3o*t;Tg#$RWPf6TeI5rG}EVMmf@~h
zUGtFH#t&@bPbU)oWj=JikiPz!SRH=~=I&ONS@7Bc>RaBb^yr8)%~pC{@V;R1-v+U>
z9X8+3$NiE%K=t996~oDzm1l26Yj(S@4OB+{CiVx})Dyn(9JQ&;7kUJ1+m$f^kBa+*
zQm^cK$qdyCwAAf09rAnVpr4KB*Wc;t`4FV8CjK6Ob1Lh6IyAWlnRu4IsV`1tEKU}u
zCigMEbZ$ntG&upCq&t8AL&+SYmps_w(q6ROoZRaSforP69Nmti{hUX8;cX#sPtBKj
zM;~1}^&OuG)|r8>Q!xv(=4d^$@tr#VL}TH462iwVhil65d5ie`p(p7v=WEyFUh3UG
z{_Vr~PcK{=HIIz&D)?t5nhuW%%h;1X2Ty7|eAAm=i5T|Knb{@Pkjx+7zh73k>=C=7
zPSfMOjJa<(?nggnHj|hc3v((qko^nIocdaZp5pIx^<Bi5T2DXCOfoZrgVdxq8rQFQ
zNu9_F!!vo&W3s_5`D&ooz%WJOmo>0Tk0#`g8^Aqv=(pkhPR^^U+`GXrW8j;G90P2#
zjFXd(zr|xsbe{b%%|uQ^&NjHF9A`Kj^LjI066fF9sB)alyd2Is4%ZBWYo5V1%Q>52
zn#G(moI0Fb@J+jADLM?_JcMu3z&Bmsn_bi49{4715==8HMMn7M(J=HF_~z>%yeIf3
z2EJ(u-{k2^?k5?ICXTNWO~wbl34m{$@J(&_=Cuv}@<$(nZ@lo$#QNlWYSLF+89%Hn
zY*p;P?&7BKO*#1H@sDJ!_z07{VmHHmy85m#4}OB2Fnlv}H|zr6tVh53IxbmHhR|6I
z-~9ZIIq*bu8~El*d-w&udFs#XRde=E)kg<{ZyLfkJK>vZ@Xf}L%ueA?b%Sq$uhUh0
z&a0~MP0K0xQv=a)@TabZdsWV#Jq?Y#N`!ANerI<$d{dlmzt(G$6d0DIu16E;d`r}?
z9TGJoAW`*B$1D0|ysDG8>3uj}f1usmo)WM3a83CHy89yH<$`I3(&M+p60eD9H|H|P
z>-x7ioq5RYH+&NX-;BH(r{~OvhoRw^2gj)~eA5ZO$-UL1^{d#$0N+f4Z!*I-5Bsr~
zp$D&{+sx`n&bXaNCEB3%)FO9W4y~s!o>fLZ?jt+G?=cI0j<b(lnycK(InAvy@J&2?
z)2F>#mCSA>)}+6<IPZsVR(*<9@a<SFKEy2e@>qSG7OQ66VpXdRdo~-zs$|(%J<JC0
zyp2(7_@)GW(+9p83E%wa5~J_IG1}NHhUbPD6@_mGKaXZ-U^Kf1qE&Apxf}SV0erK=
z6|KqeO(FQk3g2{i9)))irF(s%G`}@nfBD(1`HCIkH^>!tiIm9{sd_mgb>KsUj@+QH
zcwK}%-6Aw1C_=x~j?l0Q5!!GuOi$n&$I(y~hi^>XLluHA)t@~IbMl0$(9!nn$qG@8
zHt04l$kkM%XPA!R;IHHbM{>|zmcuo#ce{9=b18Kr*BY)Ht;f1FcnVCjg6r05?w2;W
zR29Ctj@SMH?Rga1^}_P(oa0)V8K0?nGl#rI9Xejfp=G5Ux|)-oy$^QPM+aV8h?&Y0
z4vqMZ_f!HeyCL^Gi^<ybb7*S{yDzJ<OK=m{<V3pWopx<+W>=PQn}VCuNB@m3`>XVu
zb*8KD2=gV$be?5p&%idT3XxMN4%_S)1Gfyra~en{ll$XQ-L1;d)vB*u&~xyk?!~b;
z65Xa&D=Yh%m;tX%k4XtUt4wG-FD+Vf2_I??9u?kH?a{E!uk`n}vv9w~DQ`jJv#9wi
zI`1yg|FYX(trq#KBfd`GuH2V*;soOHG+-X740G4nnaRcn+Wf#zKQ6#5=p;Y4(igLW
z9==)3c#nZy`ZEhfe)uk4(@>Zs5<Yp3w=o%Q=3+ZPRcS@H9~#O6w3{igP2I+RYQ@O|
z<8<R~J{IP%cpSvJ$LCaQ?WcWh(PwzMW*nc7kJTMk$<h}cX)JS9GthV7nycgSyC&j;
z;fp!v^S$AvL@#GfY6*Nb1?`7_cS>$Pmt9P8Xc5UKv-)N;vzN_RM;4$}#IwVxu&*9H
zV@~RUk7mL)xlWK*iNKp`jb>AeTrRWPo*#7fZKIQzzMDnAnv}hsNq4irIPgutG^uqh
zDICt&i#Ij+zEM3c7<J>YQH$}EYPB+I9vPb>?O>9&Ms=@l)Km1D=U)w)^xB|a{}|LT
zt5KFu27P{S&`I7N%SpqT_1U0S@J>r~oJ;srVIlZd1JQVv@wJW^_2`~af8~|1nY1eo
zj|{$fyIHDwUn>1mbV-^N-<Vl1x{hni#k0!gBUdneg8kTOH4lD+Z!XpK)kE}~GNXJo
zZKki5Fi)mxW>s%vR;li0{7|#X{$ti^IJZU}epkqJ=W9ieKnH$zy5ni`yOXbmMUIi|
z!Fa*%S`I6o8(NKtu7dVfZLDwAo_~3#1M}h2@T$lZza}Gpb}2kxJdj*RAbqQW+LzC!
z7UXZX546d+pB&D6cE?s_k3@`Je=TJ%!~;8ej6;8;qduMN(6!Ta{qg*~egtz==XuV?
zm$}uR=l{`8%~;9femXT@I<+Vdx=aQ3HZ*mqXj_*ShcP#fKJuWSOHF6OENkG4z4Sc5
zBV*dRRAmO&2=0OY)p+mx0au)J(N~1td(WlNC$Py&GBIz^V&1v**IVYbUZeRuqwfl?
zc?j2R;_JkpaB2Mk`h|DEH(Y;wx!!DE3fnAZSMCItD)H~WUc~FX&C3}%7bfEErV;2l
ztz1e2+f1s=_fimFEf*a2os7*br#@^(r<zK3IMpdfS9ZL&BzI$QDo0T?<=uEmb732_
znl7E;8*;xVCa^cED}8>^_*B6`$`wS822bgFeR4Q(i>1<kkHf0b#Z?`TstU}5&tAGF
zpU3-mmS?uA6#7jrbTU|{2z)b>lNGKR&*=>h-DO5}I$RUO`2yF(adyBqX*fZf$>?q(
zhXc=K<>cQ$_b@!;$D==wd+?*yu3~Q#Ow*WCfs+c`7?#szvxMx;LNdni&67E>&rEzS
z_{KN|KWa>ha*pEbjevXLn~_7986QMvAbgXf5B-AO$O?C%E60m&<010{-|X)M3$-WT
z+y-67MfTZ--v!^S6M7PS)3`pG5Pb7TC44marZ0RmFF(CGIpCj+d<=Xu>;pd1i)6Kh
zZ!Vu>m*#QyXC9=#W*5w}K3RV(LWdcHCk5Z!g>Q1=PyNd`tKpl+@Xg=wjSqYi**IA*
zDw8`djlTuo)X7OM8NSK>$*aH@UbTa77GLwK-+8aX;hWy@&5NO4GL&A8BY$()&#Q!n
z%#T;07bm+{P2VJ`z~dwxxtXLD9h0=EBz=E$_oY-xQ~*9z%fksWZDu}dOoG~x#YtTg
zuX!`$71A$WAK{ybP&$dNyxxF*;!1S*6^qxptnmthZ%V;8E#RBl@Qnw)>3)ir;TzMS
zINdMBBYZOzzL_-3qx@4n>NL@#8L1xK>P7A*8viNW!)|f*GB`ZU%`it*iCqn)cr4&i
zR0faE)9sgZ*Dcpsx4e7F8LwnUe2SaAp<B0;-Kx{ht>(VWi&uBEcg4-VEpj;UP3OO3
zHF^)bG*`sx2Yhp(YpmL|qVKO^tWwIrI`EAZzNro0+&d5>Cw%h=zBvHj<b!XPHjPom
zGGucy@&0GgN(0{%p})A;qG)C|qZNcdwYf#KX4RtmF9UlvKSb&G%TdZPD@t~6lzxya
zuAVkZo32HY|A|zth)DgY9jOd0BlMEpnV-jn>rDS}l{JSeV?R1<d^%`kd4de&Zfaf)
z)sy4+QjJ12iar~2^-z5(5UMTT+G{ac;wDW)biE~iudT`PGtb&FEpwviw+91Ss7fOH
z@mypGxjx0fH-)&qtQ^by!6>dh6J5HzoO`8JE|p!0N6tM{)i!9z_Z-aJIb=*nKA8K&
zVa1u*=2|*2n}Zn-I0f%N^C7#M_OmN3_l`kyjcxIBDD)Ves<$0#KZbtQUSw)oupgPX
z=PA$q<#(IL+3kwq8ohsvO^ce^m<J|ja5<0;ICk$aD^+=XfchM>GJ{Uuasl_aQ_*SA
zY{Jp=pZ2Ar7p~E-R<%wd-wD%{a??YM=0BLc%>;CtN8BUt;`yWHFIIiZ4fo&|G`f%X
zbdt`yO)$<Zi;fJk=wm$ncP-cv-iVxWNxJ<qShVhmzlxt_4rv>H6nWG2!`buH*<TfS
z-YUrK)Ciu-%9ms=DGNTtdvZ7T*cEk-?mhI8{+pO_f+5OHLoXQt-=M9GLAN;q*VILe
zxfX?|6b5GmvxBn*x(aM_$InkQjO2Uann!iX#nkcB$wp{1ysZ#ib4h->#R*2!S?7;V
z6Npa4=he6IJ{ajuIPZsbvZa#c8Oe@v{@qDK=@lG|w!>o|SnDf{^9=5Zf^F8rT)S)Y
zJ$`~|HkozgSG1eDW_`Ft)`q=J52lm1LBCmk%txhf(X)5mNBgd_3w$MA#b`JE1AUZ$
zH}wc6cy@~2nDol*n{U$lVJ3|tTU;-bNdxhM_B3VwsRA?Gd88tEQ(dnc<=sQ(CfcZ9
zLX9fi!l?3`!4{+T);FptnPGqQnh1EN-wT6Y+%u@eRfEReFev(+LI1op;LjT5;^h&%
z-{*^gT&huyn&?4%ozmzyGy52|ku1)&n?_Cg!Aw;hDJ4?uaU|cok$xPw%l?U3uDobB
zb?I4dN8Xt4J?c8U&A*t)*0QgBCVM*Jn~>-11ch(*k6;(Y0bkuu!*0&XFin72#%^Y<
znQhiTWN}J=<mXJ@e7cE#By^m^Hh(Sd!plow<4?@t1u{pqlfFp!`VFkz*kEOE3O-eK
z{M5?qJ&oe`a|xd7JD9xzeU%I7+-6r*jpl(GJ(JGkTsAefkV6|lci<Wup16(ugLa*Y
zU^m19G#uvGCzjy8A&hy%@#NT;MI2s+{*?jj&iaG>o4Lr~nCN`z&F;u~uno`uqhHd)
z@Eva^550HXyCkD=7!sW7;l?lO?ow~A2{C9XcJ6h06+wFmghP5WpFP>7znAbao6uxn
zpQ6Xe$DBcjxeVjLFk5eN{kX&Z(rvObH(c~IvqRyWOAAk+!N4_>x5Fpv(2HP~DX>fB
zsV-&b8nigor3=5pK79SxTzgK<X6M2rJg6|29#<h#Q<YQ8rBzMvq{_HdBs;8?-lcXA
z$<~}<2k$E8!6(oG%70_6NT<r#otn~sUA<S>ud$M>@L1-dy3yemN%t_<#)HFP71-u9
zY?Ic7uS72R8hXoRSj7*vY4i&_IXDCPIabJq7nK!WDatHWWjrk2zppHeRG8iJ8G^L+
z12dujI23l>p@@BS2&{JK18fsE#i8D?jfryvwkgeNxdl%MwpqRjUkR?M$LR{wEP-hb
zufro+3zMu#QM1)#X;$(6mCSI%HPboIVVg0_;2-#A>H?Sswpl#`?`aBonu%yV<Ir5j
zr0Cj6vNpr<!{D1AgX!KI$V^o~PA@XZ-N*=cNzv9M_A$8W$cZ9f1K&(&k8T9tgupk2
z;G2Vfe6R4$t|oLQ*Mp&Iz%rFmbh-??u;7~_`I+O&0S{#$KlGj4%zO4Ski98>4{z!+
zxf}TA*8^xdJK-4kCg*Y(W;DB};F|`0nUjKV=D;^Q;F}Zh%|<7C#!bAB?9D3prW1Vg
zpdh<A(QgW;hk-t`f9e&PV)*94RSx=1*a~(~&G2g72sow>ea0!|Z`yiwssVc$DtYxj
z7yBE&F+)X%ah=;qN`P-R6;F~IzPVQk#({4xF)tptiEd)}CT)iVZL6Q4J~Nn?VjgO5
zV!Y0`j#n2W+D_egb^VubipMJ;cf1^5<23#sv>N#4?aes(U83s-{bs}rc88Z?Z+Njd
zWj)QVg$4iZmRdU&eP%E@<G$=-=;2YLD36NcOPy?q7gfijB~?6n^9%1U?oq|OWRlat
zKkw)#hHn<0VXxF4w>GYDYxiWgG7WTVR)SmETC$JIAN{5(Ii138J;>}<+(+^^_t-Oa
zfEjW8sa^O}TN23|J7bj%zByDnR@1V^ss?=HfN!?JH#y*&+v8(28@{>QCPv;SG0IgM
zeJ5j#YQi^z&$CApzBxwzW@kdQY6kwtHZ#Z+pGLD8zamN-(QWpVyNM|frQu|59DhVA
za($#i`!Ne1OjmI{JHi#Atnb4$;&Ye`S2}1Zy?o;a1S=*rSXqmOs9C8H`PU0U4+v4$
zOU!(q4pH4*A^KxZ2$}d0y_nZps{;A^L^E~MMHYXP{eS$ubwNLA+TS622N=cTP|sTA
z=J-1{qqh#=UgH?|Ev6OpTa{-HsxsHF(p+2SG6OZ2`KG!K)m&ghCt*gFd`zLTc7>ta
zd?z>al5Vqjw3v22bepwcrlTM1gFdqceJ3mwZHJDKGCk07l9-Y5A&Xqyu5Fvi58Sn>
z*CDRcm2LVZuT6eGn1QX!zMd-qavvnGFf2f(OYq4Ct5z*!w*YKY6y4@4n$5yqR-Hqu
zSq#@ahH3uIG0X_#a6fEnW7TShRr7qUD$$Vn+DiCVg{&H#p8O`yEj7p$mp^Dx&y^Mp
zWo{}KZz@X^`VG%NskQ0!Eo_nP2XjUb$l5SF6|%`+4Q4SDKA2e~uRpsW{MChf^cyws
z9*W{kWu&+G4cVL9FwS3o%D9hizIFWc;fOToCg+CuDYXZB3A)WtxaN2Vy7@ZsdVBO4
zJgLkuN*#EmoQaH16JD;z=T!G&uMwUSoHM2wA7{j~s)>&TkL0R?4g~kiLFZY;83YIQ
z=HrjF;$>LptCP>^&FmHV<;8r>jwSt+)Ds;C{+Z7ETZHgD%=a>bf45^n@;85*RdR`0
zZKKWVMxJ;F+RgPtzWV2PUj+sDDrUD28C7;a?MK5w*O-{%qe^HTiwyWxMaZqZqi2sD
z3AZ=USAzyp3cgui!K8C<rPEZj#laTOi%TQ3ihFsZ{+PiWH-6NyFeBF?I{qAZOqGmE
zi~cegws~2?C?i>$x@a{UpBOX{t{HOKz&<0ioBak|+Gfzcod#JC8{`N3WV>$A+Ls2^
z&0|!{YRrbC2kl8BqchH^DXWaybc+l!ToqqT>V`*Uq93Q|5$4BVNi(w2uT$Hk-JR%E
zhR<qUGO6(=dU6W-C<>pdKiM2F+Ew+8bpHhUYCM|OjrYEKSIVrLP0bpXVAhZ6X8rQQ
ztXcejxy}EBe`nkx_wuvN?61e@eSLP3N&g5NTbLE+_bhLp|Ms_+SV+Dd&K{f=&9SCc
zi{SHTo5>=7urf0h!2KTme;EVSt1`XC-as{-j90ano#Or2L$%bV&U7a1BY)pH)UF|O
z?7DN29-MsiEbupWF4ZCDPUhKj<H2;LcMk2s4#%9mN5{+RAl=)7Z^d)|um^ZD?|AOd
zjJ8u8=BewHG03UM;rLfQ*_l1rsdt;n<lb=VB`ndU5j+!#N7T<n_a+%**ktEQ@-CZQ
z>bJ+Gj)&1#j=Oa3q)UxXbANOS{pL?N=_s%7M~B&t=Ch9Z%B3zv&2cHqWHL3Q=&VWQ
zzUntvC;|_x8d?y{QtcD|61q(70`OCDcFl1<l^}b=b*VNdDkr-bxM#cY%Bf|1f8k_`
zsxD*)??`%@@!?j4vUAJo)X$paZjJ?M!9sRn4i3_eB=R)v$rD?Hw5L9O#J*@ZT>BDu
zxmOitpeo>1m1Fj)G#`UD6IhaLO&0WL_~qMsW+>pCPP{FFkIBl%*m=Fx4~Nn`X66&7
ze7F@|n``ap>F737x%R;~aj;EQ&NSHO#uj`h*ycHG^AoOV${7XQ+=FfEaT4L1MEJ%8
z-+YB_cCSQxfot+CV=oj;Q<k%ODU7igpK1YLXEqwm3_1d*q$u4)m<PUjH5%OqzF9sT
z{beY79^jh{1JR4%n?Ai@oNnkzotei<p!W~HnE~Gvhi`n_<2k`Mubq5;Ai55G^VP`Q
zcN6|?_+}h@Q?e3$%ka&FqVQHezFu~|C-|o4H+uWt!7)$i;k(0*%uCEt;ZJ3RZ#wVb
zeeliX$#nL?H%|CwKwmO9@XZ1ECKbNP(T@J&0Q8<_$vRdC-hpo%@XaRp#tz@iMZb9f
z-~4>-Wj8IeQSi;`vt)`Fku`>IhQT+6{=C1lmkvta?qrukNv}TT^~wg{R6xTCzC(v`
zaFRYxOVkmxn|uWmRqb+uMjlMi##IR#l$xMjWQ|v^BU?QFzuBmC@J&|sgD+zyDy$x}
zQ}E5mGR#N8H_yr6Sm7HC1BG9UBZC{Kg)`%nZ)%+8!Zwx%>{mG9QJ;k#&6&b32KZ(d
zeA5BG@pkv<bp%>X2zg^WUKPGn&dMIel*OAOdowW?C#?q!l=*P@rXhUO622)3-#8|@
z71f`asW|q9w{UAV`J3TY+{yvpG=OiuypPqayX@0E!S3+&=r?0ybv%I_4t$dbzR3pP
zWPxwO-=f#RH{;=(fy-iKoIv+YWQ<0YVlP!T{HhPpc<<3Va+cX}c59woL60%B;f*^*
z>oarUol>IMK^&o_r6ZMzd8oV-BAFlOtd3MCxTfLSNOj@$7o8*Z7|rG{ayBbgchr<i
z9W<GtlcU>$bz*F=maJ(_rw6;AqMe%E%Bj9DnDM-eF8qc(OqG^;hyV1s1p5(wX-STO
zIc)w;Gk(THdPFYf5+2fShbqh_(~E{us~x&a8FDgj>@pmuSAM=-Ng;Lxl%qrMEc+=p
zqJ6iqt5vXFgU{jjZ?LJH)26#kY|6-a7LWIzK(;ZxT~Gh9>lnSQP00wqYGBtwf4la@
zIP^I^dJosznw<LV$tld^jR=Qsk`LI+HM!?_8~xxm`7#SK?$1DVIvc1Fdjph;W^-Xv
zfI3biTls5%DxE^JS&ILJZodrOCJ44U+QX`8cu;eBK5#{mYv_n)6+#9FFRFbIJY%Kb
zURG7)-uNK*#Z8LQg_G5)l#doozQ=RQX$$-PEcE`dLwh98H(l7N8Ook1A96QUEP9^D
zqNdz$4<lRr`l!Fwto7H^ss1`QfZn`#?#o*<H)Uj>K~;afEr_0zj*j1#Xeig|CMH`v
zbQhW4)%Z(u@Q23nwp8?#-^dH2-3)?j2BF#fN^Ur+Et-uTugS_@Dfnd$%yJc0NrG$q
zD)Lx{d2`rg4f9N;zM9oKD?F2(k1fm`c{M&4)+t(xo!-B|D*5oK(lPs#jhE}gTBTqd
zUhjX&tg8#@>I*WfZV@uEZ`kqum#_N&;j6DVeC0r|S-;J!h;XyaPkhNQ`06!mb84Qi
z3bthzxQBVEQS|nawcY(YGt@ok5q9&uAB1+(ki7;ueDvfWdS#B9G=42};j>NZ7l!A9
zF0$H)UIWu)%_rr^FZE3)m0fSt)d~1UosD|m2F|I0PE!^wrwE=^WwOSwO}AVy5ZRiE
zcvJ83rdpuggq}pdIbfg%+MtPR40^THz|IN--IE68I%Cj^hX(a$zG`#@qbA#pS{7^6
zCw!{+ON`2Uh5Ydw^qs<FaFF||j*wQYkWP_(&h<q~-;D3Mqe*$_*hx8W(mQx9O93>i
z=46i6`RMIwA8kqFt3&uy7x3#Q4foYAi+nY4H#*ldU*)WCRvPj*OVh)qaOT|vKb78$
zM|l_jaTGe=DSu7MZc#H|i!Mf3$h?zjA8MhG&7$rHEsDR)9#8a~CHzjCcU!glD|(}g
z+41QCN_x+3sxpD>Rt{9%F@b8oGm!aZo9ZvL>Cjc~4=S^Vif+NxGwmw7mz|rr@rk*w
z$dJta$V!K{p(XTxie`#mY_kQa-EcaMCkAQusvt%4oS*3`9WGB`9oT1XKDsxnIN8(X
zRId=aVmjk9kAZ7ev%}y#xgA(zOa(L-8~TddrRV54OGcsXPNN%Qp-XpHa!<4Yp4sMt
zt?1(0g}%c1eLMM_&CE)zCG)!!uL>r*F&V9AG(OZoxTrgvl0e3|19=*MI&Mnx`d56W
zC-j|vq+d7>Jd_{Zr~q1XA^4|=OPSNbRj;}CdV?o*k4*7JdPMiLcVPzK=OAXk6Pe!(
zcIu4@{^7r=&JMV4I-J*&?!9Put_5C|k=zZQQy;vhah1u=lqW+Bvs6W^$pF_3E5;*Q
z%@(xpAFmxM_niB-XAY&nF%627HRj_UmO!`3jOX>up#%J!+aJasT<2g`#~}-RQx(2B
zj#n54+c;sHKAfdn&}QJ9aQNomPMbln&26}*4d*CqQ<75-zWKg7MLSoqe+s^tfo8J`
zrpW}`q+5jV1lN?G$L=b$oBr^PX3$|X1+VIN*aN=VI+|QDe3Jy<bb@bk4JPN4N>;fq
zJ;y!amEY*_OQ!P=zG)BNM8h|?!{DuU%w4rcAA)ZJ;F}Ea%}V&DVH5l+_@*O#bG0J>
zK73QIC|VSJ^EDfOSb8!;U)d$~mb}fAWbXCoBfbEe9K(-7yZLK1b5rxtU*H@6F?8C%
zH)r6R)9_7c_-1kkGC2-%$$ofRP08@UH$UMU!8cvun=9Fq^&u@^AHK=|#;cw1jp-V@
zrcQYId3bdlzG(p8lz?x#{_0g!v{y5Oy(;2EMhCtL&gxYw_@;Gqk}A$g)aHszPkv9(
z-~$PIx+OvV*Cr_6hy?BFo1kgO<7HYEud3tXwR=Fk+Tu@TYZ<T1P2(x2h*w<2cvYsS
z_-N61eMi5U1>an}O=t1naauD!PUok@=?lKp*B@kO?t8T2?0>s6hrl<7C*VyD@vwiF
z-Qh_dRcr52?Y1864e)464d$oF71t__Csl~vn;ae`rNN_uZ?^sIR`J916mKGTGoQ!N
zZf)r0)}%;y$I0u>*f~|1zCZY8XC^oLNURKZVl@=Lsk4z8s&V*K=r>*9o6`+qRk{>?
z##v%jAHHdNl^O8)WNb#nC{Iv~>enEP^N^jHSEIEBt{HPVT0c)lv+o_<W^%OdWr<e7
zKcaN5b(Dr>h*Y5=;i`NrOe12#n7{0(stwuE-M*vF6$sa&$Z&NW7_Nl-;d;?1O!@0|
z)T`_r(U5}`1=k#j3|7+ut+jn^E4@U6J$Z$Dk3Z;=S<m0<ICSNNAWeX0MswZjn#y&n
z6Z%WzAhqYd?fnOK8^br(FFN!Rz6qN}{<k}OD1*t#lw@}FwO!qh*%gVly|*1)Qp&FT
z`)pd*1)j0nbhM03?F!jcp}I{uYNFfJCLhz-rW*K2m0%ou8oPFJeJ${tU2m4dEH&t<
z`C`)nzD|uTb_E`UZ^#gzXHLp}nmKPe$eN_#@12()B=Ya@H6IxQHCBPD4%2wa5zn9>
zZP@$(bsZF-&7t&+?6a!>Oy)^on~+rIp}NC3UaMNf(n(A<^FCbjxFwoCdd*9VRjqu`
zdYbZabzvUZ=4^3#{c@5k{?4rTQ;S+(v1r#J3)fnU4o_ppP^v{I;(2>3c2706Xlgn3
zYm&XG_J&!hzu8Z4fH~?F>~@;Sj8q?g^@zo<a{23CQ|5mw(vgyneFi_!MV_()^KbfZ
zkKqw*qr+x7+6x(&WuyL^1y7UWr@YaA3PH13g&(yGrrE=p#Chh+JwLilKlGU`Fpa6G
zpH3HJ-n#%Z=5Ni~@Qhh)9$UaTbu*G-&c)~dFl!LJa~QVqLxXA0bL;DO%<<rp+`4F1
zr(I_L=FOU$WQI%4%96vZv}A@0;X6ez1GIFhuWG?Kcko-<)8BV_r?0N^_<1{fwm11I
zdaSSV^8D`G%txIE`l!edb{Qm-x9RFbmz|IN1IXc!z3H0)@8%9nu-$|g1K%W>lr;&B
zq=0l7o*7HmP0S}eD0G{W=Z(s`$fyNF=)P%5?+v|t?U?<3TimGQ<Yx+JH)>HvxFn15
z|MNAW&kZVxKGXadIh@@FrEW5)**XI~;s)JX4C~A^s9rCFMz1yK_$7mWA#>ca3^`=c
z8Q8(7E4_?bIp3(zGk9Xp&~plky~2`}%jU?%Fc)1o5l`sS$z#%W@;B3%YmYl^QVaNI
zVm==&;lDk$myd?Rc1zH3#+UHbtzciZ9_p)5bR5?%Up<BwE9EomN&~aj2Ag$pE&sow
z{PcJvT>$XziLw4lIEKdenJf<cyu1VX8!s$9h<=<o?0Sc>f3U~l1$@(i-|Ng9>BG5d
zRaCnGMKV7%=Rtsa6(d_57AVgUeA}aen%Nm0b)HSHx%W8#i(TIX?YcRQ`-~lS`QaCj
zZOHQ~8RK@79NMvs=iMs~Me`h8ur~8T_%g52Fb2&-v)##a_-Xbi-DcM{8cx=%PW@Jz
z&bvlrSDfT>JWjnD>}0QzlO5u819DGQ1;!ZMkh#j1Fb9m&pa(PAL*Sh8XgSl#^UZ@@
z7Q-#e;gA*RIz0A)Y3eU@$$vI_@MLm1W6)y;ljrH>QgkvMgHh;0ZRzjxBX3g^O$S}(
zPjW%sZlcFrbE?G!ry5;ys?%j=zW!#m{T4pjGrTtV=MI_TW!K>ww3{aD=_Z~C*Yw7t
zig7BX1-mp2bc~kez;nS9&}EX*tI%$?TG(laUb7#6DHT>(#;FRceC4q(ToX_b?#siz
z@LW8;qE86l>Eqw@kMgrTe4qX<xMwOK*MKu8Q;;rvLAQB=zO&z<$@o%U?j@Jdo3WW$
z(YySA7zx){)~2WjEYlgjsRQ4ff^Ej~`XRWcF6Rhb6UMp2$K6J|*$3bBgKyHqH-+Jw
zuZv-kg)j_kb7nsC+jGg~z&0mgn>SPOm?pzO@J;h^@Ctm>0KTaM-xP&!)(v7-Y5=q8
z@J&_trX75<HW@tzz6pbGZ1BxC_~saVGXcIS@5I{*KqoT8NAOK%_$C3qX%62kfo~3$
zhM|h^{=8^U@J(*`rVe~_`c1L|9w+PgO*ESG$?QU7UJC7|4SZ7rzUew8S>6A~H{al!
z4Dd}h4?0a-vd8dEQ~2gQd@~2WDN+U9r&O|T6+q8{Z$jXk7oWUJeeKoY2VUj5=2iT0
zGC1(fL-?jTd^5BkJd@&8hbXTKhIkcZ@oIK;d@J}S626ItZ=U>?$V^BgdS`+X_OXXy
zL4s=fCFswaWNa?Qs}o(s!{^1TSF?B(X$-?KClyhd>`j$;rK!fADflK2d~*lBDFojv
zV?O*cd{g**oPyvR6LaAY3dX6!SC5uIWlriGob#uLUBhIE@u_O{AUBLJ^|Kwb;7)c+
z1=5LQ!o#Z0$CO9oDd|!Af*#$>=FyfP@XCF+#*!^Qalox{tC<6bZ}yFGYfcZhN_S%K
zR1iD^->iafT<}dQeAEA3th$jeUJ2j)wmDY2`^3rzwt30yRE8R{dQvo2bu%+x^)N;?
z_R-@vhy7E(#c0`)Xbl|`t@H8G$`>B3E-l#=-k2Gwa?wiJ7$wu$NR2(tzV89dcGrv4
z!KdtkD$|KRgD{2X>Zs}DXa<rAF54$mmB|E~eaQwh)1BYHz1BSo*20)zRT|$`UCOrA
z&SS0hG_JKuoM1CcRs82ic)!=VpEyMBZ4o*vUh+?TV|O4s;K;l+#|NI3AFr6d)B9+q
zWnMUR=Y|8eAk%xmp@s|arP{NjrZJvTesq)<b|oIQt15gm*J;<gns)6#t2qdJOg7k5
zue41Q3))l|zByZiay9rSlKnTPdNz%A^YS~JPL{B%ZY1}OoR0JipLt}{*Jn00FO43Q
z1O11&RbRR}%Ku^4QFI;y*`1lpfoJoxX>oq;K`RBSfA&CqTM?i;3+YUoL5JEnIEPGe
z6kTZj_gnS*7^`{=u<B4(tNaqtZP0A~4z+Th!G7!@I0wC^GhDN@nN_K1Hj!xke^y{7
z1-9|Yjee8Xs_w7p%D-(<t5X&^cUWX!Y#}R77NVy`6C?2rY~+dS(f3!}qSxu+pJ)E^
zzd(N<+2Xw9C%*AK<?g|5?MS%Cj@QtT`S5b=V<1!d;tMlQkJux1kvZ_gXeivfw_c3D
zGzE=j7;{rS{1nJN{aNxhnOoDHXCp6c_EWMS&w-pdIEh&rz&}NpbIM6qF}!j)t)ISU
z_EU~4W^E(8+Vr(qRo|F(gRIOn_~j6~%xpA+))`=(EM$k_pB{W%m8)jO9yhDzI<ppx
zHOr1JQ;sZb{&Z$FCs(r?ZKexs6En+~ZX{o>uk2db0oQCrkD#NtFxtfI=44FJWo{Pp
zQ4HCdDtJ))^ZVd)vU{6ZqYeX2DlDG!%g__^!=!Xq*q_<eq)_Isr}?s9fjOz4nWVs<
zu*?Ue+Mu1pp`mo7!*AalqXu;&W5YZ*1GIWce&)sngVMq_jnQm2e>X7aXVA=#27P#B
zP)D>GUwo)G+tG1gn|rI#YL+uU1>1b8hCdZ)P{r8>O@M*S%vyEiYnH_4Dug!l4vnZN
z9+hDmeL7c-s`1UJfy{Bu;Wq5*Lh?5!rQiqXO5dfBa?DaWOj?e|whF%K%}$8$FlMI)
z)6KjVFYYF+Q;__Pfqm(N(7@2QR=_vK;G4sF%$g71e1mTeq?+}H|33+X;M%$5tS*qB
z+T+js1|0(pEh>XfxEj8>*%!8nv*^<-i#Fm>#lq&b$&oEs%dYox<j9y6%r}ZXBxeF-
z&L5~{Z3A_yKYKb42WpMSro-r;Wlq@?!i>25>{{^qe|yCLeq`6~QV!K`OSZocyHD^*
zN}OWvCcet(a%4?fkRu^~5<iKKy0v6vxt0vO#LN#GPUWBU&=rIks<X?%560=_)Noi~
z51L5+-Do?vor+BF(tuykVvNiqw05a|48Btrm&W&VX~!@!%45-3CeceXmDi`?F>&_6
zG@~b=zl=hk8I1me?$iBOv>gxm8Z;Zb3+^$IVXg)16yfWAbL#9J^rO>gIIvCgb97Dc
z@`Jye8o|p$VV&x5O=WbOy?k69UO#(){LM;c!AHSkzrsHqof>9?&zi847e6llWc>QR
z%rwO?EARTRqh@3!?g`<T#4OB;!!Z}Qc3$9dE~kAC^2BgU`R`;1(QT}^xG%impeGvr
zhVvHISqkI)tBc*sjJHn?AR4W#*+v)$?dCn&&43B8%@~J@j&i8OTJ#%uWdf%GFAs)o
z0yrDtny)ZTUrv3_C*HS#Qx&c$%_+;7yaF9(8QKqga~-bvh-UNaEIMc4n#%CaLLLvo
zH%H-{0+aBO#*zUZh5j=FeP<XfGz4D4o5~E|%<jVsRSz=GzwvT1x)6NR4!*I&H}l||
z`S4Av)_h+<uoHaK9=^#3-`sBum(_z|;G6#NO;7kH5WX?OH`U;q1?l*@U&s~1H=Q3P
zYx@nl`_3^db(qKPu*)i5o=1-{eA6GkX^wvLCwz1Cw`9F`(@)$6jR(GI0^i(#Z^Gf5
z)95!X;hVPk=`_xftTOOTGJJD@{LQ6*yv%l!D~4~%Z)NA^Os}$zVK+lxua>2F)iDzP
z3jL-Id=p!ZOmaD|hJQ`c(MRmp{GKRd;Y9WtC8!X5vu1vRiorMQ;2Rfw^BTS>1K-%-
zn`Mod0mp}0nLS<|vc&734D4cf6{jc9<J6N}aXi}1*8A)Qhi@jFjZ@ORIDPzuy$tZp
zjxXpluh^A&pLr<w=Fk1~_O11>U(h2<G#nG^(GQnL%V3+WjmZ{QU|zhKM^lT#I{7^^
zXZ5J%r~hWBj-7R@B7Bno-(-ew?x5eaOks~^M?5NrTQi!ny8*uG1K$jRZ$jaljPT7n
z_-4u$`u%#bR};P&3*VHhj$c(cRy%*jXu(~2Z+5dUd@T7JW~2&kW(GWwtZ>0-h0<lS
z7oBF!`Y5$)6QzK9QMy)(`KPDh`fW$J>KntgeNiX6Iy-55w=gBXg>BMwP}EF1hkJ)=
zO50G~Q|SNaf$iUd<=Gyra<|*+eaE(nwYAacC9SpWQcDfw@A$zFdKkFJ+6LofJ^|k>
z3)02WybZp|2j5h!K}L?hwJzMJx$%PQ@i#Y{`-$27&0XR8RuR5=#NXTfVfgYcb^_*g
z==poQ)*V4}Sp<W?Hydi(HD(f8O_WXM`ZgUa#`W_l+R9Bbj(6C9{Em61k~Y0@*|@K=
zso*U-X$sIWM&>4KCuY1|cAdlDx^mm5z+7~VJ-2DXTs!+AZF-T`uAOM;dAa`nn<W@J
zE0Fn^KpFD{>hG5UidaD3*`xr)j0n&)dSSm=*|qS9`R*OedG}*(wu@CC;^?c7u<9_H
zO|uqgIDzcP_T@RDIa%U{RwdV_)4qz8PC>fSidt0(w#iQBra4`HCm+xy%YE`V{DA!H
zm;;}QFMvm|FvX&T_V@<Aa7{IfP8G1QBa`{>fBZH3G>o$i|7kv2%}6|`-{`vu!>6*6
zr>M(4c`1L*%0|ZcBfAmqlOaAw&+vZcz1RE6cRsV(6JQyh=aTTIf@9HT$N&d8@QjSi
zHq}MzX+*x)$B(WDG?^0Ypn^||UMI8qmsz8pn)MP++54wi=QqPHWLTe%=4CV)Ga7;Q
zh*`nRJbj0AF5fq+6|B?$vRQuH&3MnuE8|B^#(UcP!&ko9dEZlC-91Tm1nnjfwkaRZ
z^EBC-Hj{ku?dTuQ%j@l!nJVe4p-<?4$;fj%&+|Xh`>1xRi99^n8hoexDQF#VK&~bx
zH8H~i&rO=K#iZcz>`CZEcDEX1D!JL8iEi`lD|;7Un{^M2DuZ^DcNX4L3Z7IV{80xF
z3Z@x}|1{==K|c=~GzdO1letMEI~+)cxa0%QDfoxXO&_$IarjXe(QO9ZZ?3YRn#-p+
z%wspGcd9{0_ZsAQ!z>j&fb(mi2?fAPF-E-^&cCr9wz@=y`7;bvU78pp)tpP_dB61F
zHeEPh+4Wh9eN;a1&1#eKex}zrr;jd&`^YhlouSN$-#CwclbwAOjrke%L;C{cOz@5I
zzOOc=haKUYO7KlGd{h1mdK(;C0R8ShzY78P=tOEnmtrtGz*7Acwa%a3GK)5aThwL(
zdG}-VOTsrHDX{r+^vRR_9tYBI+$VrpngDip1}M}SsMNlJuogZpf0Oh1+uSpS9aOJu
z+~;zC!u>^q&Fm7tX4kY#4xOyeo~$^BI!z-lf~F9Ur?RgAJC#i2O*}z*G8|88A$rbs
zW`0gGhk%BY{yjZ(+3}{zIW@N_8RItiQJv}M98cfFI(9d~7B$~GH6brq<7&)|!z6`U
zxnzrQ;Rn&H*UhD0`;w0th;B0kZh?394uyeWnz{qvmtOEr7qp&u^2y;YI?~Z{0(oC^
zm%dha=~^)~p$vTP6Q|<%w+8LOWuE8M<k583j3p<`_i^TTr_ykeN6?9W(5Z{po$7hk
zsY$z>YP{H~u`t<7w5PFfO{1IWFk6E(1g+})s37{<nIQ{h-rFxo9<-Y?@JkK2W#~J!
z>~ukTk%9aTtdluwkdiWTZG~;d@V<iRHJ2~bwf+~6+$-`#tPW?;X^xYxK^M=u$)Q&3
z*!8=HxeBzKJM-ySoa)fAap>iv@dt;plNGKpaV#7cr}df?&4g$6aH?>2z%)O3TUU7I
z8a%U-a~G!hvXoxHCG^WIVm^BTS`K^@3E$L#Z#u#?ZKuH_Q<xo}l%gr{&5wz&(pdbj
zQE(D`GZDV21>ekqZ?^Sk-U_}M3g3kOh6e@T97!Pm1K$|nn{2_%N3~&&s};R{LA)Kl
z$qC==H=s4aH}mW9wi@J+;hV<r&4t4B8RsE4l$Go=d}D%dYQi^r9x^+19e?Qzb5DoZ
z5xzB9Z&tEn3ci^S-*kX)J`G0Kfp03nH<QtCo_AnI+=ZTFA){>guiI}Te6t6>d6^F{
zD?6UmPy8qNruM)2;(xq)3*W4SZ%%Kc$8SEJfAG!B0bV-Ey|P4jb+j$}8T{E%Rh3iL
zi^s|5Jx$WD=ackqM3P#*P1M^3iJCSkL9y(Ys^1^ZVJB6UIA*Eb@pAjdt98M6rGsxQ
z%t}qm6tCaY(}e@yjDT+%!Z$zAZhFEuBjKAYE93NjZk)QpH`xo&iSvyvzc=J<$lz4D
zLLc!T@XQ80s3q*vM8CO%Hx&=x&@`ll^*l-}<xzXM<}$j?lr-!T|Ke7K*KS#_)7N*z
zEzf55gD-ZobJ?xeecjrU1n<B%$LwzQJi9dkzL^T&9ENX}!#DeH#cBt9(`^Ge;+|w}
zT=W*hH-A)RmZ}gr9QfwWEp{_(X3tc2I&ko$%tPoRZWgUYIit1Yev~d9jZ(jL^w_kG
z(&ze7bV*05I((C7MYzV74Og})owVy<n5tAoyMb>qoCsB}cA?rfvYna~X{W+HL)5)_
zh|+%vR{GFj4O!V%4Ge9yGN_HF*KMsP-CAnxak6vVTm1>&EPBlS7kpECHuDe(?7eNv
z{YbqaRjY)?j8?jzzsX><(ofuT9Kf6Ea?PQ$C)s^WzPK=aGXlOD4Bw2&!~F<aivM{V
z{p>caSz}|)l+0oz-hK<4&Q!IjCfdz~t$`Yt7^s~Qf%?5oApLga9O+TLl+LDndFiVu
zY}02lIX3b*qX+Z3J8T+DH%H)4IO$J%$hZz0$=J-mYdY%<(8i6h4ELiQI|XXM<p9M!
zL!-IGoXAl2@C*)6bWa$Ej}2}RpaGYx>OBN+DVaVS{HUKDne}dC)iJ!N0sdC4H`0I8
zn0-=pm_ezAj)P|7E(PxtB7cl0RTH+UOy*|XW4wT?%tsx!a4$jU%@X!#{!ZqBd+4Gb
zi`up%bKH#QoAP)EImsG-WN!i4o0s?s?>6|W3fbb7L;SfO`0HjnW@5}d)@0XHF<#EZ
zbJbfvxo<OTeVTdg-SEn4`h#ctX~P)2sDAAGP9`%GLC0__KMf{F{1;sFPbHrF&}ok2
zE7?An_4%1uqwkot5nfpg({zPjrtdSWAv#R2nP#;a!7iv_W<8l`R*pqxZNh(A4FBXh
zXV&<?&DzX!Y7#k`Wu2K@ZckPiw#l1;?CUG$na_|D=DFB}K67fMucG37bq^h4cs(?m
z4`hX3`lu6(^R5`X7yj|lh08v4;nHD4pH1QplM3~NYvN3*>ojSC(WKIqn5V8`QiEzH
z4f;m34lN{8cjTDH*wYoIoH?ZdY5()3-X5c;<^bKp8;q*m%c%8q_brD#KIS(nBmFj^
z=re2Km#Mo9YVZ_IhO>fm^1eZ)D+axReG2X|kd;EGnP|}btIZXEySWO#Xs$CE4LVle
zpl2qsIGi3e4cc!sXx(k*$?GyVMSi&`8Ju<V=+!w)=J+Z9F232H5$ud$KUJw+?EJhU
zC4R<7E6OappGkhpOu9h6_%8ZYNoL?@z&B?2=G-;>xGd;j{M_#Kp%-WhS{K^atn0oi
zz`XeNU+DRRDbo%zGe2S04)nA#Rr!BR2hyffeyW@SW`=RzB(d{r8GFn=@H=L*Xw?{t
zYVo^y!DQ7+{Hc)(*&A})sw^h5^>K7QjSbMF!|WcX2WN9H=EPS9s@g+3|L~2spy32O
zq4N)q>IfO*Tubb-oUyASvy8KAz&2q#52GE7;Q4n9_c{MQpZ}<XFVzA4rXQZnboME&
z!Dl(Zj--oRTb?jKl!l!Jh4H9rJ5>+;qp+Kd?hvQe;8A5i?9>qMwFahfX;)EZy=$V!
znDC>V^zenE$Drjz!#N8((=YKGyPx2l)m>c*glVcIyX1+(w~BOWTsyk?_?$In=AIgn
zFRsFD6<@DlW@f=(v*Y`3GR3e<hBZ!2odVYkL$eu5S13#~dz4doInjJioBET9PGT>3
z7~fkfvd~7leE7b1WQD_CaeX|?fA2<go0)W{4`Pmj>tzhuO;7y$#nm|Q$bNX_>s@+D
zKRQ&JYv+78ryiWMmHWg2pXgUd1J8BaAu}w~`UI@PW12tFaSp;Tu+6{z->VG{WrT0G
z^R;fEk+++}{o`crC&$1({QR>Ip=%MQ5$9jVDaiSU(*d6OcTOsZ^Oe`Pz&77k(oqcC
zwBj6vZ~k5i+o0R5f^TZVH#_E%b3wDYIWt8W;hVhhO<DM+7JTzy0-n~m6m=Ve4+Y;0
zfo~FrF^3M{M8h{(;hU?y@uJ|HzVJ<3_+~qNGoTZ^)1JLmt?|BG?EZA3)dceSe&|U)
zWN+Y`jqpvb`Y=&to{iy~a`4T3_+~16vnMN=Ao%9o=VbO>B&)|Y`uz4L%e#%t@D^sD
zHZdCx-;{uFdcil@;hVAW&CNdS4(|fzz&Fd_oAxcp*!YtxhHviG<?YqU8^br_^D|eT
zC0Wzpn>_H%<CpAcfNyrdH${)pU%bUDH~P)>(OxZrZ!W_(ZNj}e(aNg{%!!Y!;?=Cu
zUiE=*%Dkih=0cM49ZJ%T+4SOEXMSpZA`G0Umm?B2D<Dw|ZYHQ?+XVWk<MpCEoj7PV
zam;}Syk|$|qd4uq8ix;nP6N~IWM-=QjX1qnP4CT;IQiF%lM}vals`@dGQ_FPdyh<b
zREKVO=mPSn<VKI&@J+=C_A<2g$nN8zqaDVnjAm1U+;LvKsVwAiexl2u-F$z^jMOE!
zemP8bnC#7*1#Z0>?^ZxR@;F|%F2FYt@XgUCbmPD`o8X&O@Xfrpv1n7V@~nwf-pR1e
zud!Otg15mpyQ;)$W5HN0{}Q7;*JCsYzNy|ZM(rL)Yg_wh#Wjf5hfH`<_oCG8NR+Cq
zjpE)nN@p4|BQ+^P&0mCT#*}a^%NVZyBf@m$*N)1Ut)u>rq_Yf*>T#Q}-PqlV-JJ^q
zEZeXMh>D0ox82>{irwAux4XN$1O3?@*!teSemK`6EQrXSbDx=c=IKWtUv;`_Zk%bQ
zt6f`Z?I(67z&9@NP5J#T=w@!ABddaRqhXM4HEXUz)dE#}mcPz;abJ`JPx&YM>LWa=
zU1a>Z|9Tz6Yd4Tw9DK8<+&}%M*Ee#7FMKqcBOf~I$u0QtvzSjD&ECoZ?A`0)qx%(L
zo{T;^_uj6F7wqcM+paQ6=-tiPm(2X>DcM!E6uFx#HWip^(`x!t+f-%u(><%~FKlZ4
z#ir|N?8=putZ`1eHsW7(!>_vB*NzWQZf7I88hUJuRrwm<IUjSb&Whf?IM%9{E39&V
zXjRu+>{mF!UO^sn9k6O_AFCR7`L{=^P$Mh3P5SCbSX39yW+$3W+F<fEXf`En7H#tU
zm$CU%7u}}1MQ+ST-7C%9R8cZF`7A1tot~St77d1Nrjog-dJkXfyji98o3(Kr+0Hp;
zZ5e6iHBL`)TRvtpYdYT4;1Xu$NX(i`@6CI%H|~eY)~q$DWHNahvc<p17WZjxQhS-S
ziEMFVL6efn7B?na9Ce4e?^Eo%+<{NE0?wI1XWuAvo!(@9@unWOW<DG3<~7=l3mM`^
zWyt^LAzy<Jb(Cx2jAvvx&~Bp8WL6!cQw6p$!YjvC;!BOAn`W?4Ck7bhfrgU?4d>(v
zIA@PhKAX^PcwQ)vd3rME-GRI?9X9LB8tGc2kM63Mf)0483cAhtxnyky(NpJxR#V+e
zv*@*PP4QF=y9mxQ|82PCsoW>Y4<GYX3$&ZmGjzhN@=#~^W?DFNMHZN+ng@H|JoJUR
zqb?`e0mZ!a=}@Vd2ea0-q}bBz@y;SGXC^!SC%W|x!xo#FlUhU14SrFvHtss(O&9TB
z9}0%~bl6Rux4<=MGxy<=_Yd9BxAB=$+|+_h&FGVEnz)YHs5x#bo5WoBb%Va&FlhOI
z25m|;C@`IyHf46xjlymUszfgiymJ#SGLuuj-vXYYSLfYydW?6wEB-Ow73}qcd9F%h
zr2E^*-dvU*(q-(D*F%-jrm8Gs{wtLZX7rnYHlFOw;^#yDX3HH<=H2;u;@vGEV?1*j
zKhy1AdVLem>bsXtlrb`o#?O2>v*YmT$8Sa*$9Md=ihiRrc#gll*{z3$6Hg!HA`_hr
zc$`g`)0<^h-lzOdHo;qsBOgDTdBN*2I{Rb0`;)OBLf(Ei`^cZu169_hC4n|Y%(Q9E
zReCR5(0{}0*HW79o6hVJUqFuTqywLhd|Npm9SiVL8D8@Tm$UmE&0!ta>$R@*wBS)y
z?Z)+eqA%GR=A_X$N}uN3f`*eW6)o6>ew^~?G-6gfgqiDJWLsw76K!M8`pRE+V6Q<I
zKOHH-yi`p;Jbknm8*}7=a0{ICwH+PA;mkh3J0E$R2-9>7g=bp&DY6+pm4&zY9RE6W
z<&^hRsX~6*p2<(IK03AV9^HIoefn*}51WT31IuiMW7_t|D;q@48Lsi<I5p78d;%U?
zELu}AK9q&-zS>SbE9z8}bbKE7fy2<HMr?pprm_3GFaORC{JV`DMc8}!$)Sx_SY$l&
zq(|Y9C**4I@W#V9HPiSi6b-y7+D-odeDwVyeO>6^LEGrOIP60v+(#b#c?En^55B1m
z-wa+48%<?4bh?jT%;5Yxk(@DXQ(+X^%@F3<;FZ7K!#g>4!7#&lybjZx<rvA4iNnnM
zrt-1;Jhu?XLinZ_d~*=KDZ3c|Y5|@S+KtCtX1L*-tuy)jWLO8?CI#JQD||BpzG*cM
zeFwg20^c-*Z#ECb*BXT0(x1!?eA5`diH2`JC-U{e=-LVWw^Q@FKc5fZ^nq*gdhuMi
zrf_4psV=kPb(jmU!kkt~_@_9z<s#&U^6}>^%vk+`N8p-wPnmtX&%W>@?33D+q(xiU
zAGJP7Ud!;7<|nCUGCVUeNqxt_FhkhU0N;4PH|OA+K=`I2d@~QeiGXh=!8aM;n?dkR
zSivM+f^SB{H+R3#ZSy8kJN`qrxk2a6QM8&RiAsTQ{3emZ8N$x+M6$<i@vGpQhD{Qc
zT8a7b5{YD~*wp~v>_@*@1mCP(o4_8%1nv70uaytubzoAwDh9-BTJd=OZ)==hmyVOQ
zN~|`b*_8ZFPw_`)z2Tdl%z)?JLeI_07!}?YqqAFMl+G<iw;IN%O!XM`DZ`Eiml#cj
zZ+3o(Rx9`>)5B=Ijc9e>8m%r%qSY1crkOcfb~o}jHQ5*LN?&mSI&xgtubC-YZGMv%
z{=yE1H&I%2H;QXflrC*z?i;>&0^ejBz>cY;D0L5w(ocJoTERDWE3&%*zA2fW-oFcU
z`|ZPvS`(?TNpu)@i`2ouNHsQ)#VH@D)LiUq_{yBr4g9H25ejgP(Avx48rwWvj&k98
zks%ykAWSungmL~0Q))<<YT-}KjiQS$g{*J`c12AN)u)zi)Ut3JeGd=OrP9HAk~3Iy
zR<%+|_-4TSmbx0)Qp*drR4?YFlJm6C`YAzbd#AZve+SB0F;JUk`qL-H9u;2iJ=;5V
zI2*6)lgxbLO|^w@+?{AP_3?>0Z<Wu?&R|}n*?Fx_J>kRtS<Y{0rSpd}C)$~PPpz3>
zBzv>6EcYAae?6Elb>K&BW!|)HqFr(LQU}pu>gBaFKVsL@{dh<d=oKT|c%p($rSDi(
z>bXs6Q*CPSlbNSKHvRX?M(2`EhePeUPUiS9`cHneomtOp%D=*<=gdVlVegI`^B@hv
z*nP=dRK9ao_C|2NMtg6+(kkz6R-H6kRkfv623d9KrG?%Vi}Hq(uW5~LA4tDFn$1OT
zvchh3h2TXUtI4cMW%9*%QR9lUe+tdUi68YXGjrpA*a@C$R*6?;Sst(}^D>^(5q2eR
zA#20TN^i8Ao&(4|#F2q$!CbhfSuWM+!YPQRlg6wDFHCxJnY~kcO=`%!cBaYfZ6aIT
zhHP>AX6y`aYEnnC#Rjs)UAS%)BU@~}0mmF8Q$zM<=n`+Wo=T_9P;#c-m=njFiVLFa
z&&;l<M&26FeSHc1sG4MLI+GLL|BgH`9#j4s>@?!u|J?zj&fzl|Hj|xMXw;x#M!o7~
zlxH`do507CnAPbI4@Hy1>1WhPxF$<)_=j2L$ARcNwa{(y!Ay83p?kg5bAy+*quuzS
zIZTZwAM1hUP>g+<zdTjoEM0at$(UTke>qQQ-%3yU*7nq@7an?clzHj-9(o(+p}qC-
zZ|aaEW~YHabKegR(bu<*v5iI2sFv)xtS4oxLca`i;0Mx4p?E>*-@0oiJvY@h&`HBi
zsLs*MY6rRN1^M5S_)wLuxGDWMH%(vWrdi}^`YnKY;Fygt&bNDX7o+V2?{d@3C2l%8
z!cB_>zg#fr(?x?`+=gwQ8+7okK^;FDbS)43Q{GL7>bPk#vsUhT(Tw<eKJa(uo9M2e
zTgm!7aM$T<bQsrUk2(7yruJiZ#5t*8DvvH?itymh(@{A8y@&ddzp32XQ(Xpm(mUd*
zbLcnq^O0R@N=7@3pW_5Cd2jWS?Yft4e<i<Ege+7jGvfTe#AowXW9Hh{MR}{;QucAc
zy)}N4C9lhlvN-g>1t!&dfc{qtPjn2wqmSvEgw=P4(cuK&G`UERWlpOm@tT;_!>Z?7
z*-P@DRX_6cdh}%%)eK&fCvEa|u#0LvbBztqY!VzY%w$H9O#kaQbTt%*XSja%<bLKI
z_cz73CNE*Wr&TuQDzl*bmi5)>e(=mJbl<hi{~X|62@NOzOU^Cu!4f!OW<9i$W-tR9
z$woM1&Pt~yAB8<0py7PQJ91%WyCgZ{>Ku*zRMhCFu6Aat{QUGRfSIZ0blku^Rs8*w
z(U%@TxaOJ2?x5MER72k>L-%lgG@XoiS)ZKpAWKvJ472H5oeEn>ws<@`4P0}q8-M0_
z2hU`Mbs}J!RekC1=?Z5>(&rPzEUD3{$5ox$p9dcM;j1){$RHkP)?y7aP7{1Jv@2P~
z;D3EJR&sz{ACb#Ozs~JT2RpiU#_K+6{LDwq_!;$j&(G@v_ld8WIeo|s8eH=dM!d8d
zZ5)nyz6<Wb^K;!zu90)>m6d#c96sP6<~TS~NAkWAa1UHlmd8fCZ#=wG3U&$Q2;w;T
z_icFQEyv#(sY)E%c#c0z^8-Js6-@IArdfn$(_<mCOR!D+e6l<6&2_w~TJTLQd@~ll
znGfGAgl~qxHyz-cI`B>2F)$K*Qy;!bhHv)5H~Zk5Dez4x_$Iy!pBKko2>9ke8~Te|
zz*+FkWCtIEZ&J~1+}!y)8u2x&(TQ9IU8w@QuX2$m&PsPqCVF;KnRR+c<~N1ME9_f1
zL(k38B<@$yYIY`(F+#gp1FtN_dz!}__;mJ2!8eOXvrDr-%+ifaPF#|*bRdTloFumZ
z_KJJc@7I*))`X2J(_0MR^oDO*<RFIw-_%QGcIs847CcVWfGdd#Jb-7l6fX+C=?&kw
z!Z)|c->eT!R7Rgf-ETxbr(&Wu7U9t)Q7+%;E`C7QAAGZ99mnVdRjinxGxy@v_C&nS
zC&w$Sa=ePZh*RNWaVkTuI67Udihqky<cAnleiWng=VLT4F-Gg##i)yKj9QuKD28v!
zG>B1!YB8D(--N?A`{A3u=r=>*o7ySpGMA&ZWDhe`v+%1T@TrXKWH3a_y#~EEWnmrq
z{QBe|my?m)%`dw9KEo-mqg3kw{lw?k%>dto!8f1bo3`-HV)(`g-~8?N+fy$}D_o-J
zJY`1eF+0M~M{3@#NcCDp?q*`7W_FL%A6ukGHD)e66`na4A>VHF-NZ!borBKebP+1l
zJzOi@!!^nk#`zhhK2O4wc$hxoMQ{!N)Y}>D6_e6dd2)wp(u+1)J+6)V1hmoeHzCTA
zDOg>z2P@0rR{F7|l_mzY(tP$_maW=CDL;Y~FgHk<rUog+7^EDBn`<~e`=0;Vl(Ln+
zyInRVFW`O#9kxSuvUGFla+t!rLo69Vhp&dC-%KsWYZ}iuhS#Ozu8%4oL!U;MemkFR
z;b3w%N%+XY+=IY3`O5IR-R96J_-4;!{3-b64SbUyHfaIh6oFfAZHHYp&?y#Y(-HKW
zAIzv;MZd|n&8qZh=F9Ps+9$9laEDcXd&tpnu00cASDwc<IG1_W?lzT;wrS&Lo7`%%
zXSuFbuel#BPG<5@1Yd{G{~!0JOPyBsypp4=VAb53Ry87H**Jt=z7`ghhikgQHNWwq
z5@pdkyr|cBQNDOlSIgm1;YBqpY|+C!%$}gx^vqyUnqSOEeKM;vy3Kdk#{H^U^~g6o
z+=&mh8oh@5<<Dq0?Ygs*A<V2CPIAVLnVZ6!GUYOBHFGZ49?@;XOz)NLCY|SA`|((l
zD)qplYHy;az{JjGdjH54e{q3-$QJ+igiacAqz4c3NZumj0&g9gfL=2YJ*N}*?s!ut
z{pjE0e*PiaP4#kcO+j`@kqh=^CaUXe_B=f>;uZ2X+Rf5kM!nxihwuXYs$`=Mz%QF)
zjG7G36pmyjDxCcZZHyY<nm@<FICxUEIx;ua-l#Vgqw1A6YR^wE)xeip3)_^O&#r|z
z=nUvK=jkB6*T733(IR&8cH6z4s>9s(k&T{YN0|R+7W};{y9!d^8oa4<!#wDvL%+e3
za%bjyTRysBUXokcFD+gT&y1Ch`jgYGghz$<^EN%Z5WX{uor*^DoILLic0w_uy*z+^
zVP>W-d!p$u6aI-@aYi^~26NsC)7{jCIq%19-SooO4UNyBUn32=(aWHa-UgMMW6;IR
z=r-hT#+@?g=?Q~&z&Kyga=P9%XxKf2Zf8RSf^mXt@pbmm0r=gG%pPpj{$B>ioeWMS
z8K8*Q<eh7ffA%9Tk7bbWfVBRDbc}tQfzcl7w1?UDH?UlJbeoo*+BlFN<K>?0EAv#L
zyk1(@811G5xomvC-~4P(^E00Ih0GFs^O*mWoCC=g^Z)v$I=&=aJDD8T;^XAS`8{!~
zY*O<uJkQzue*I^XUmdeb@_U$c)~tQ-`D-iPl#|gSPt#w_j)v|{tXi39RUFy0qA6CL
z&xM8x-xQd}E~=w8*^TsJ?zO9PZHFF3I^;juLFU=PZViVr<o8iQ`N+l`)m?m&Gh2N$
z;RZWwQ+(vc`Qc7s@_StKbB-s=w+L@)D?Ki!=mEU%%Z?xNx4E2ZQQ4_Q-grM9;0L&(
z(j0a-ZD$AfMW^mQqZ8*BURF-Jk4w;hUfECXb^VmwgxMz;<|~?W9dAGOjWa9di8f>4
z5w7_kde4ap<d2KvZRMs5Cms6EM?9t!X1mXkGu+8c_Y!h76PS<c$KTnJ91R*xmpJ&R
zvs0H~oRP3i5Rb3;*n%*pyzr)YMQC6JC%fh7Fon_Xub{VP0shno?)#$Ib(u)l4Z79%
zU2qL^W;I4RG>O^HUhC*1y@9t!7V&>BKALQSYsh6Mp~XxmL-RLJa}<51&;lQwoxr&i
zW@!U2{)UOVPNv7N4Ss)HAHC=B@6LYU-gGJspr;+CSuxN@EqOZ^M+oe)k>e*vUf5<P
zM_QPsEXQ9@Du$yLAN!lPSqtCPfNxI1H-q4tK={TBz9|jgJVd+c3*X$wn<@j}IN+NO
z@J$4K;|Jf=f^V|GH?Kz1fisdm9r&j5U_7vX%yjp{YwFHScqf=C0gj4>i{P6&A!K#n
zo38N97ayK)!oPxVUK!|1hHo0cH(9HYyFss+?t*57X0z-oJEJ%XzhOVrOZtZ&k_A2s
z&%igQ_b2HN-c<fg9INq_mcS^p|ILQ~84tG%=k0!YQC-=&3Ev!V&)ih&BxP<+7RQk!
zV}m5Mhi_cqo2Mn{y(yfea_~((_-4iDL>b|md+^O>_$F?DqCTuiRC)L&>*Pee?3*ah
zc8MAoNZ+58y$$ur9G4@9Q<%L}ITAJPbAraFBq;5f1P$4ipvDst^w5}~lPU3PcqX1)
zRJ?wSj92TYaqN?c)B6gsiX(ed0{!OsTYRazG3Z}0S{)msU%@eI0pB!(Z^Gc4$Ms{h
zu?in=9j)1wqshRMu|bQu%TboMU5-a<;^t^A+Q`iKcsg-9(dAc${+nX(4}3F;>`kuh
zWRTO5$02)j=p#DKt0;Ya7)7QpN(1*q>Dh8-siywxE`HaQe6fYy;f<o?RX&PN)F@^6
z7OAiF-V{6&sjBeJkrk1e&_7bADn!z=AE^}^BbeLdI86uf;c(5~6fQ&0a6L$1|3YXu
zz1!icUYNP?%VEkhA&i{`VG7P0rt|OcqV9#t;u@;L<AQYcWRUu2YoXP%T9AWkp~J;n
z>QCvGN>jLn{wxj9srmTOoNv-{pLS&re#%-pUB*++F^uasIm0eJEo%KguG?Ga!>DA{
zjd|S5@jBiN-^`gqz8}riGJs5+o&NP&>_jQTPL%Zc%Wui`-}F&;bkxC{nVpzRX9KUZ
zzKQ6pEq%1NDOzkvwAV8ZU1!c|$RxCyfp!hTpQ;jM*JtLVwsY>x0^4k59(8GDo9_PN
zdpu~>8L~5tJ6bitg}h^0&cDC#i_)<Z8E@(DJol>0HjO-MQ?KhbJtnX7)ks&z0Go30
z{Z}rA7gf|Ma~Wn{^OK!y$Q~X8M@g%W7O<*p8mpGfflmS~bQ;rjV<BT{q_dce&1$%2
zOC9!1RbxkH1&eavMcpGmyffdw*>9oQY)-?^=^J}8KbUomyv;D!rr<Slo2Srn_R^uY
z!7T5EX1$$=j)QiSkbtHW%x-5B{l&G-s*N}0mc^_eA50o>*Q6WA>G#`gQir+ZB}Tw8
zotYC4VUM_(d+<8!Whi0N{H!KD{Y+1A3cY*h>2ur5{L~t{-DcD4H-_Dneb`r-;H_#c
z>7=o<uc0wKLvPLb65cwHo98q8<na~lg_-V~x5?R@gJaNc{`()ivdXB8bC~z;LrxgB
zDcTMl$DfRjnJ$~gMm=;lD%_5i6F_fnJe|X^&!1LCW#IjDxYiDQ%T5JmrLr%i<A!VT
zK(fRi(t2q(KGZt22;&D&mAXJb-wIFV>g%a-^F8I--BTGldaCg^uJ6o5&HkGk4)ais
z1|C{m+(SLsXYlm}x#9cl%G^tB^a^PlGgCc%*}Fh@-i#b@0=b(<@0bIB>Mkpp7c;s^
zx`oVX`?zbqgT7+srXE#d9-NF#E#{(($!_#qx#`YpgC3kUDCIa@v(TXUfd*x6Z_t#M
z2F(nCb9Ncjl#lJ-#PhZr^zZ=s%?^XE^Eqi?o**6*UKrHA95du(k#~|c9#@N9<iYIU
z9OABD>&ZHw!?Swgt`TTB*tJ>{FFoGLPOLZNi%WRWTSw2|E)VuvduR)M)3rG}w&0s8
z@J&N9*B^2+SJsd$6Z(zYC@*DP&rJPUFU5Q$<5bwF2>yS<VA1LKjmn$XTSduQRY${_
zJb<3Xwe0!2K<_m9t!WcY@;=61vtMM>;p#iX&FX*MtOeODnuGs3h1dM_ujrTXe#0oM
z(#^H1;2o<n<gn?y4S#9`_Z!?(_~rQ5*D$DvL+RT)v~Y|=)yNokzRymo9CU;?z$fAQ
z`&U;uy~Ib0c5<EF<D)^B$dcsrWv4ot#vpbRB>U>jYIfi5XTQ=FU$uh|9;D|!s)SSD
z8#(2IJ~9=yc#NJ>Z<SNq57AR`lg#fM=9&J`6Aa@tE=o6_D|?};lDDblrzluway|4N
z9^2IRlWTQ9c~)e{RVikq3bLCs2mQo+uFD5}ci3jb1*h`v<?mbp!%Rkpfo*=mG`-rP
z6SaY*c#MW~W})L;fNg$vhiej@%GMsg$`8HC;MDoj<aN3+`;HD1Q^Qxe$Rp<e;-iV%
znX!0Dm&H<c_^xm$IfzaSdeisDIW%}Wx>qvi!g2JHPNECCk3%hOunw~lr(DrqI5K)O
zlhNEq4bf%R!alFt@v{u}(TqUu6*;G_;_b!!oV$19@9o9@;QsWybIv``j}9D|=OLd{
z9d_x#v6v$*ER&t%C~tS;@vok9fVb1Hgl*uPX!z#$60{rmrZs$18NSH^-`qyKnFilf
zfp4b6H}__ciGgn_!8bMFn-cI%TKMM9SaLX{(QDwFobb(?f%sp2*%^gy)1eD~6nx{1
z!y}98sCgamty(kt4d3)_#+()UO;`9PmpfUUhImo6@zg5uy;deKR0#bBzS#`lEQM`K
za}<Sd65tyjG@R}5P1rH?nS)9CjCRv)LlQfw=`UW)F7a99aV9e(HIDt8!;*9jzF807
zl!I?J!Z#l9O-dm9#NnGY_0e+Ro1K+loRaLRDomy)3t#(pqAq?+)Q^{mc-x6;dW&Aa
z<B3YMAyIAdsGh+$d1Djhj&{@0k*GB$ayU(x6UU$0QYcYJvnP^aP0+ZM1YJIzfZv!v
ze?x*&y%KZ+e`?)@cs<x1uj6y$bw4#uHCM%{m@D1IaLus&F|w_RQHcq7PCdxoM8)Vq
zOY$|27^Qm0=u=bjIl<BDaVttrLxjq^MQHZ02;CmR?DZ_VWllt>ZYMG@dDw+HoWA0f
zFwS5+tB!QxxI`<3+)eY0(Ru;f)J)4Bhi|Y7d^6)^l!n4LUoJ=KHhi;+*{K7g*&7bq
zJZ>AMw|G-`sz&K`LHw(qkt)E<)UxA|sz|2zRbZrQ?2b^_wQ#j75w3({;hI||98ZhA
zm>t7azAxF~rD1a26s8P&!qkqe&H7VenzJvAj*~F`>KP`_?d`R`cRP*9+g6o4LbYO8
zkoJuY(oI94_IC@=^3DMo-8VqF1_y925x}`AK(`OzRr4CC-iTe1H!Lc>m+T09(}R1r
zgT4Oc41e^X#}vNt=UmqYJ_sFWRq%Q;D_NZy`i$%wd~>`9nK?5Wb5*?aIpq5nqp{BR
zQP@@P(+-e{+vuYY<oAc!&{Mfb*~pBxX}Lq`rZLO3gPp_C4y`D}y~_u?g3sFZeFU77
z7JupxdeL&5JjfD1N4M$s6aQ#8GusQS`klo0k(JC0Y?J%DMW!FjC-J`Q&8>PLYUO9f
zyvHl6_WCgMU4Xt0SF6h0MT@zK-jWTT;k@m`y=hn8U#%V<RUWIpWnw1$l|}uQTVx?y
zv$-k%=0@DZ*0U(C7BeO2HJRX=7`SF}QHxIIXD>rea+q1rdD8Lo`iU2XA64=#97C3&
z^nJ6=q1z;$W(UK5vl{&my=EEnQpsjTjbOJF_ssLd@ud9Cy5R-~RbgJYfLX@0>}?<y
zQT;j^&OzqE$=<vpdoyydNlg>rot7pw^5VKwog76GJcf)W4Si4U<{mq~Pm{YLd-HZV
zdoQQ6hiVu*pt_^cpxq21SA5e8Pl}nTwP-h?g~;J#V>iPec6EO;s_sFfhHo{>hIZ40
z-0+gMbk;0lMhfj_Y**%S+8Wi^kNpWI{HX?Xtkf{dzXW>$&}kA1k-I_1v7+hp3pDC8
zb3CPTF&mzO2er*h7wEBB)R}%8G#jt)o^rp&4#~s#P!G@}_IPULI8WV=_0-()o+=&h
zsof5;H}BZ{y~abqp=41i&`pdVHF%HI<gJ)zlE#wBZMQ?(I#=@RgD=IX_oSTUZqP;|
z&b#Xh9I=VH?A~MC>EIw!Jjh)?Oz;kDGrbBk;$_&2`G*;)t!{b-+iWBky#1j;rw$p^
zcM%LT!Jr;!H*MM&<lfw%&w&Qz=Is@c1`X_N(8kFIot}rDv(})>OVMxOo%Dwd<UkEN
zkc<2cS)PLUTf_Og;+PYE+z)?hEjz7Fvg0a+xhm$mmerAbn4>DRNxJb&`1AO20Unwj
z>mj$<=v3#K7yscQuP*d9FZ0yu8!#Yz*tsJ4V)UDjBfQjPotNG)lXf4iY*{IO_ONGP
z{KJm3$w2XcnhL+Js>fVhh_~jCqaOhN&H5TGCzRixsU|&qj=!0k?wd(4_bPsGkK$F8
zVLq@AbAR0PJTJ(74W4c64EnsUS>=RpJk8uY47F+44x5VO0XM5iey$7NvI~4OnCtag
zGRD`LgUsNgJ=M_~0?5R2jorz4p&NR`4YY_MCz$QwUZ_$C_eI_5&l*eb3tGpq|M_ai
z2{J49;exO1b;!$3r|M39wK|m>AL!B`=Bef|o3sVJ<g`<*?z4~lom1vN9NGPJ8LufH
zx=fWa^xc%hb1IK-#p6f*yt)M0oI>njK(qOs>E8_avA1+2KcrU&J*ew`=C;<Lzf4EB
z8Hf)B+carKmyMIW569;Kc7vlMU4pZ^^mOV3oYOg;@2@TSV;}N1P53-=Gzr`@jw!&w
zIjzibW*%nS^@&-DuP%0dq03@xCi*GJD;;7d%*_HE{WvF1glR_5z23#4T5!!fGxMN6
z4jqr@=*)hvk@Q>CK*O=ZGp*=P<T$|3^)yG>Aa*kF=QXW;Gy=xS(g7bUoP0nmna9rL
z1G@UCe-~J21*`(Y%;3la$Mog+&f(3Ui*bzL?Fl?5627Sn-<*MOy1_TK;G3VY%|*1E
zIq;1WzDdEG@`rC0!Z%l^cVr(mTFhiPW+HpT$J2EK-%J=yZWz8<I-K0hAfAISH3+^5
zfp2QUH+7@YY9jEf&~H8j!&NQGGyCCp!8g~8WN@0mTy^-ntB}7b$<C^xWN}LH_3|?x
zo|SAfY*XO_o)mmzd4c8v-<&(cUIz4=r(2Wg%0#1ChaUys%vzMBelwGFcoO@?$0q5@
z&?F7(pQNsx$nB85X#wAKgl~Q}qx;wgkE#xP8fqkIT}Aw<66_(*m84ei%^3J*3w-k!
zzIpH{QGaeGDjR&$XltS}EoJ_D3VKdVqKdX5lLOyOgl}5FH@)DS5hdUs_@>^^1pR)J
zphB|}R8|RkRxUwbGbgC@gLu_GAFt25;&pOfyxzb!mX&n<(Q(t`WsK^bLA!x(UQUlu
zclf4MXpEi(qS=sjnTKAoyK1CzRF70}?Fcof6~PYq2(^W8zQQ*(r$=bymIz(%$lOv+
z=DmlK$$@Y3!Z&F;N9%ZoX#FoMGvYkXfNv7hL~GoaD49R-wc(rR_*2)<Mk!@GUB$Da
zbao6`<IZ?i@Xb|gl$KSCQnNzLh@;))-5aU(E|J=vCXyXA5oCcP(D%c&=w^r(CWq+v
ztPplDhiJ*p5c%g1)`yb8@+}*z(`ABH?Q<&?d)G=MU$oNQ6Rk7`pJ{51Aekx$DYQtC
zW_tw5nW4E7Lj(W6v%JG_Kba@_>HBncs?6|{fyWo)nG=9RTI0#g<@`172)p%Qn-U}G
zvE#gEVOHpA9Qy@2b3Gr;BV3RUAO64yG!nA8T`yZTp7YPiE$m1cjmBJ<Ic)m$Qmqbk
zsp(MhArAHF?@)Ji)=;$73*4wi^SaCq-&{WCP(kit#-d}l;rUq`;Xm(VXJ2)@X5WNq
z@WUD|u&HudJh4JHttWGnuLyeh3#(dgu<F`utMZ`LWcqE<i&Trg{Kvk7&*&zZtyGy<
zwV&^ES2e3@U8kFDp;g%)<7J`QOgK-+F#pbb{99_irdRkqvrhS}n*NQ<0jzV2$AY)%
z{a8pJaSe7y!83&`!+bE!;Ihm>;X_3h;bU;kAGl^@R<g$FEvo;U{LELgT+wXi<44UW
z$ME}(S+lON%MiBtae&<LR{Dv#c61?Ad|(_*)8DLkw3|<@&Dw7@>q&jH2A1XCIj32{
z-%Z+&Z}9j6JEeA$Azp4$f3i1z{otBtX2k>G88?%DRHDB)ze&5&u)l%p)vlZPPshAv
z+`=xX#ok&y34I1{>O@EKHf_*!e7&`~Av5D>H|a~V7ZdH~CfB}e-{BQzphDo8W%KE(
znLyrV68^?q=BcJILp7S~B-hK2aLWv?otv8&HL5y!;$m=6Hu9@!jC8elDf$`nJU(P@
z8sbY8H_GcdGvIh9JBNGeS|1)myfnXpmoj~1Pt+N94b1aY=bdy9lbdan<f+L<JeWT0
zUhwu*M+5J_>Y*m|*;FvlZG&%<N)KLPa>Tcj$<=ItZ6@MNk<Im+DY*x;U#gLm<&nFd
z9%8>G9XAKayNn&{t_HA8_EGc+#{fWNVLDfHS6F#>B^0C&C!M>7U363TB5sN!KXdB}
z+RGM$R>Cd!I~mBh7*x;Sptg2{?%E7m8-S+M+Ms2;zjeGpd%H6~H59%{W=?#$LFf3~
zd3adOei;<g-A!AV$C`>qH4VmDbe;LEH0&6!>aHAkWI@B+>4auh9mcu9zaik2)E3{(
z1D$GKqz9cX9=c4=Uv2p2Vkb}bB%pVZPkQphQ}s%+b2ONpTtmohGe4E*GTu}=qgInI
zF3<l_GyK95u&ME=QM>RRJHxc4gZUjGYo2Ehd2)VV4z)Jv&~%gXJ!M8O8^5Cy%zC_>
zockfOHsMp1#g7f2MIR2YiHc3FIy}^>-AAqD9Bqp7uqlClh35Fy#U7!JG6zr=?X?`&
z?T_KiO)YU~+By0*e>tQIbh2=7Gbe_fRQ^6XKAe2)3OpC?eRf`BuBs_>m~mu_2m8ur
zI(x2H(x0^lJ?ApB;;-0QfY(#K3|<l1$Xr<CV^?y<6P&udoUAcgOR0<GjnQ6i!#JmY
zI%Uel9E1xq-+BGip@5&-7o-c2$NT(wC>qazEbL%NgKqtWulLHS_wZ2{zDA~lWM<aW
zV>6o$n_=ieiB2tT!A!al9S3gdF7h`GV65!Ey44#-tA^iR2rb3VT#_F>Gq6SG2zX<P
zj|y+)-<_L#Ikf3MZS1OMwJRH*d&pkSZTRkk;hq07v+wS2f5l7YE3Vkp^E7kz137Ot
zcc>U#Q^kv)B^r(^e3K5oi5Ti&m$8FgCv=B7VJDd8K{IlU{7iqsIBi-me}#tgoX2(O
zIaxw{^eBwJF@Em7$>l`yI39MH#j%&;JBJOPS-^3eV>(9`j%vKWG<@?BZ)zrdV}x(g
z!#C&Pn>p}JJNPCie6tX~$p_y=!#9iIn?vx;S@`A<e6s?+84TY9!#52_^7Y`G3PZ^H
z3?O3z-@Jow-gbmnqS?s_-%N&Y&cioL;hP%p%^oY9<H4L&V>*fJ(sKjfOonfk7a<>%
z5B(@7pPQM#?>kvx*yeAS-^V9O%p@^0g?95`H{7!=NmVzYxva%|T22?v0{p6EvcnVU
zx*1JRAABSD=0``g8TiHu-^9Q-FW{T!@XaXrW(j;Vt~|Rs(QcmQB8!}vO!BWprGH0;
z_(h@;uP18W#Y9~pe{&MP*)%6n8K&YzO-NLDYxXxd6LrX#s3DEn;Q-&9gm0ebW`D!)
z1cl;F6(7QU_(pa#Ama9bZJf7Ynp5$rv?E?uCZXjVjMI_I%!sG513V>0)egqcKg(Y5
z<QTPqZw9uA(ZU>LU|zGw38pz+DnkCSM9+$J%G8X|DEQ|4;0WCr6`{rOjcsQHnn|S6
z_Qtzf8m-%~O+v3|I$H3W(nf1kdUPE4#s|Jh{|8=4rO)qu6k1i3dfkkY?L?G*FQcCr
zz6lu_B@gtQgYeBp-zY`GShG{f)vSq>$NdPUc3>vzNvLX83)TC$HnJRSt!}?ssb0QT
z`tqWs8qz~^CQA$D>k>q#g}-`s@Yf0EwC_yuSIlyM)!pT<J;p$NY!j&L{Q}i-ZXml?
z0=0-taHURxTm<MW@9n2Srhjt*KQ5!CKEVecMcxHp=RVhW<3si+aow-KkL&n4uKV-Y
ze~l+KB%bVS6xV$?ry+bZ;Um2+o>tZGZB;S&rpsMEzY*RK-FS{}cGWqIr<C5I*2Nqe
z!#zoc8V(hy=+Ms?hh~QI+Kl$mo7}v{?$C`v*NttegB=(SHHB}g!Z$tPo5##Q{lZgP
zOI9X+zD<eTC!RY2r);z8NgkVWFuUQp4F5lwd&!<wZG&kxTp;&ypL@&~=qG>hjo`IT
ze4p3g8S5ddLdfWh-9%^KHFoN3V3vCuJ9c<1y^C4ydvF_U^yv-S4?lxR{45f<Uo92O
zzNlh!@^L(eXZ9DeXe&&UJ`X!JbMSujnh3bY0oVM2YX*EIxA5Am^=LK?AL1`DXS?7M
znVVDp*k<20c1*3|zIcIIJINH=hMM)Dt67uKZju7bn(V=zh8i$YQTQkG->-)zbKj(8
zr%dFYnBQGUmUtZ3p`K_u<R-#>O*-7jq@Z$mRJl#6_0wA;pOF*3%KY~s@}nELe$DgN
zg0W<Y@unhT=_79Lt%Ybe6S>Z1EAP!NZf2*lGC%&?s9s$EN^ir@U`FcXEHXFn%<X>Y
zFnx`B)&)H$kscd#o2mFv+uY!m+RPo7GqNL|eDNQ2nb%%w_lOy&1Ly}^*#Eei9Rf?e
z^kNPE$t*9m=*Ems9GZiTU77jO8y<RU26>x#Q#^GD-=!D1n=-zhI$ebw44*wzu@N)T
z%uFqcAcInnIqA!shTxd^IqctNXHy;goHcOHDm<xqVfazZOnt8=O`z-M82x#1WNrRO
z$IWS&CM|4pZZLhq;qXLLayPJry^_27qus1yHoRzBcO{b_p2CcGv7-j%+G)@nIA#1W
zI498{Z#=2YCId6SWN!=xh4c1$bR0hz=N-OPLL8b-PlJw<(^)#5EY4zsY92%L`D9SA
z7xU%C-E=nEP4`y2smNur$#e$pt<220pSyPUch^Na{}$q<Rp#H}zZt*kJG;c$3HqSD
zhiXj2mwO81;Zb!#)5<>EQ>mvsRUSU{%}w?u)=T~Hrw*_6Vom{mgl}?`C0~rbwi3Sa
zSYTBBBmZ`84y#EQKr?SG>&EVOem`EnVAigdNo%K(=ej{go!`5z_@j@b=wX`AURN0X
zzrrwnj78OFa_@sq+5}HGfcuMf<loY`lWQXrcX=WD>J6L9<U(`pY1eM<bI!GNXxU7M
zzL6_A`;{*8(%jp)`RG+P_NAEV*X_ggc7c!Lc5=PFz{e`nz1x=kyBLfQ$DjqzCDXT&
zd#7W*TE{)q$8U7w6mV*7E#|1uNG#Ft&0uD=XF0WL106Moof>q7z8M(j_#3h^U(r$i
zz$Y2}^d=L2Rc1fk;;|%uUdQ{VeS?+W(<}Up{LEdax?SMwz&7tTI#ql=%rqJerVGz&
zjUVQP*Hj%2$`0q?J3Vmql|T1|p`6dop-nAXO>PF=Wp59d&ex&k_)YoIUD^aX^dbiz
zOR;NLuw4($b|uoe-f@OaMe)yDm!+>_o?R~YVVi4q<y=gU&{({A&ZU1jXFmO1<6y-x
ze67aK_<C{dz8J*$vb#f1_<1_<s<u1e8aQVNjFSVtIfQrh1Fxzmj|ni&8(1d^4{I#U
z)t~S8Evz$W1+&#0YdCIm6ohAdIa+g+;h4?)F2Ob_%is(6CLF$Tfp0D^goEIl82F|l
zd~+FZDiXfAH4_a6zUd6#B*Qn0;hS0TO@H{NIeg;^-@G0Hw+w@q29w)?Z_49Q-Rnla
zIDyAVX3t@pz)-Xz*rrDyoyEQ#b;?9Xjys<N-_)rMFO?&2jBb+!z9|Xc7~q?P%;a%?
zpsl>ZS9+YJ@3+YiUx#5%CTZ>-c17)gQ8p%#pGMPJ#x920JWgdEY63G<qv*65O!lTP
ztP{r$&IrCnJ2E!G@Jj$WWca4Anf||O_)`_wHwD{l$(E#%8F|k4MD=)zCUcu!zjKMo
zjCRvwb0Ryr*fBMmJ>l?8wTX$!AIuy$e3J(K<|=%XzH*{gmSnyPzH#}Gphs&GwDM@Y
zCM}LvDtY2AgX48Knp|<Scttji*ZRkC8q0pE8~0;W|5gk;mSZ#*f69F`eK+Vgu{~q-
zXB3&<_sk~wL@3xbLY+&(H>L5TGDYay-Ei%Fga`C2T)SU~tA!&%)4N3|KU$6dVRA6|
zOUu@h8@?2!+;`|B{>cnDT(c0yY0Trt-%+{&_l$*a^eRfJJg>tR_Ao4u($48o+A^43
znq8vQyE>hJ<ZLc32-VV<P&L2V3SGOE%1vseH5pqe+qagQ@VceutZ%6)OIs>sLQAb}
z+fs2OTBvdje-+Q>uba)Av0tN^GW_(D&rv+(Xg__a>!)+&{j|Oak9n94$Vo3%Zv5m*
z?3J^yLoSd%d-^G-E4^1w(Nc%_>N^bIq#hc}T3&BU@Pd-*tr)}XLw~zAb+YS46SBFT
zEGimmQC2H6LNLzg;b<Cr@ncGm&qW_O6oy8Eh7!Jt9+`}Gjo3^-%~QLUr*$Zp`;l9#
z?F!h+Yqu2p_FUPC5KQMD_aqrQIP`;a-!#qxH^<Of%>76;cGh&}`^lf*!G2?QTF$~t
zO0nwEY4+Zbv-z;!s=N4M{&@e^305td%8cqDEBR>7y&LK0I6)5hGJ6vqTiBgUryd^K
zNCTPSELL4zN4_SbRi1^cx_5?I?)hYYS6RrKS#*94KGbQ>-G|5;&1LT;?~k0w=l_9C
ze$fSz6|Tz;J7u#dmbbH!t@#YoG*4rZ7e3S#X1<Gl<alG&<!7+WBea^k_)yo#6Q8FC
z_5|}%=r*@tn=Nb2>_cQeYPwl<Mwyka4}23x4^C@x#ddauH)1bC1@52ov6Ctd&wqt3
z!!@Vl5qe7KU~|qPD=~tcG1;5qcnW>2CVgiHW=ScN24!bY^cQv_Jc4U3cxwdS!-dt}
z>N%4=3nQ2b=6Y5(oO^XYb||=ct7~<3Zlc}HM!VUYj@j{4@-|P5s=0-0;xZnm8<ohM
z)Ya}}Z<v$%9LL+_Ip$l961q&Ysz$k#hItCoVVseVq1z-s^wL7EukUxmGMl}WZ-tk%
z2p?)Lx#HgNPctu-XyT;}MZCxxd#e9-Pi?@5N;lF|eS3PMk<dY0)l<=V_}E=`eQ)vL
z9OR+!Ko9h1Y0zorr`My+OpreHr{_07x{l^y7=Rbmg3jL(>~MNPj}2UNwiCM*vbgL0
z2{)zN?4~{A*y%lt-I&ZiUBWAx|H@5K?30>S*j)}6^2PKMFWt}Xg@Xq6^%zvVpFvS`
z4RRxQGb+}g;VsDGz&6p1464+?pa~q&-gsDOI`4uE?C~-vlIP{>V^FW*^!<%9s5gw%
zjAQE=gM3~a^rp3&ehqMA-ywU*@36C~BL0-GyZ*4-)3SxV4JFZ@5{0iQ^}&OCSjj`9
zqdc^p8S!)M)Up?$-!F>!vYDRD0>g&yU>q_yGh&#T8tJ7G@QvXj{Fs*PH~LKj_{L6F
z>hm<Cx*Vnp=Ns8?G`mtxyva`9?3G||>Qiz#u<;)B8<&0TGkZbCs}Opk7kc6_v&Nn@
z>omXjK~>0^jk4(DS6)BGt>`Xjp*Qh{8{0IL-2PAQLl)5AP&^ZRY0+=)@;ct+bZ8yd
z;ykhJlZ&D&d@Nbo{q$@8CR@@3uE8s5Jd6I8E$mM^&YVv^_6r1(xlLkboE*;8iFENU
zro(q9eZOcpOJ0-j%R=W(IXo&fl#6i3Xndgs<D8nmn0}mXct^*XrMe26Janr63-(39
zFBMarx(VydfoDeYxbZXW!~4JS@$!$z%G`7+@*LU^Y!k5A$?q(^H514(_u^}IptIQO
zRQeiDrRFD7{1&chPM1vqdWGkaCyZo%A|3NgD;=87eOk}1bc&+2d~@eK*A$P==+FT8
zWCwf`(#o!*WU$*e;9S|orb^6d{FmLXn33o;C+M!={CS)C(KP++x|e{i6K7|K4ZU6X
zS_AfzL*()OBN(X!`!5<fRJJ(tXbs2<qlq>5W!Dti(N7rXIjnOJEvFFg+ux9%YdLyV
znc<j8&hf){ha9}!8Gh-<F`eUY7x5#Gba2g29(Qvz<`@XytbuP<EhS?E-&{nySqk5D
zf^QnaHy_b%=D|0AXQx)cHyPj?3w)CV-}HlTy23X>@J%K7=FKQR7ryZrhQ<TmT!n9P
zz&E?#o0NEZjp3U#;mk$BHAZqb-F?u8taJc+!AS5;{U+!>@Xc-4f4<cI!n~cgqn^9K
zUg?>y`ickjk<1Mm&L{XL6u$We-~54Zy1+NTHY91nYVyKM;h<S~O;gZuMklHJK=wBD
zVwSuE`@q{IX<2i+{a~6a-botW01vAuIc4;jnh(hX-%V68Jg5v;6P4~vqDq}$$HEcv
zG-x))<%z1XFi}=AH@(J?Hy(=j)PsCYM|L!{U>4jrQLnwp;WSEAo@$A@lt|X*Nxa6L
zj@Lxkrk8gd_m;7GyObSMlVY`?cdR1AV)ckUQ-A8ms_gR^&46#--oTqWN4{oXj7Fm0
z*pu<3dec`7V=R5a4olc&4%$uQyb-FHIYO2Dglpjl`th2f*MvmKWhB{~h4@sy%rwPD
za^D-JY2%|bdJVami&45jUgzo$=DgvWy6{b2ys3`xPUJIoXx@pE<!Y1`9)^YBoAh&;
zo$5%>AM;VoFGs4-U^;Aux7WMqHoCtdKo#u)+V38q*m40n`qW?hulp<L2)ilQ`m5qt
zf3*$sS4l5_4XNhOZZ&@`c-u@hgPPIJ+)Odw{g|0&<`KTx2H(6Z=cm}h^g6&bhqC#p
zR%SHW8vNPYPu<;NoqBw{EL!Xn?nmm;8OprU@VfMj!Z)D{?8-jVuC&waa$|n!<3QLZ
z#I8~>M&_1eY)iv8oZqJ6)ztUKr$XBp2@@=TMn0z`^P+fFt<Y8~{byAn@`i~$ZR!rA
zG`hgN(o%Z&F45PA*KEPJ>dbj>Z+h<SKH4?%KeU=(?0w??rXL+O#rZyV<KZuz%#MUB
za1Bhee~wkB(P|PW(O1JcwdFK)oMBe2=jgEjt*Ia16HKymJDHlZXfu3Yo|i1jL*_UT
zF7ZIWd5hQf2VH9YTlAO%cvFKc8V1Y!>}$~#JDtSv&X5HDjd*Yig4qfF9*yONnT$63
z5zu9({eV$m84n%@kr_@0)4YIbn)C6%SIj@5*LcD;4eyzC@rIebR<Ow#v#yh0@ZU#2
zA8hj!wz;#|teZ2)*o@`v0cJh#WLEYtd;qkYv!2X~*P_#}B)Q`pX0>2$>L=cm`4T=9
z*PX1ZP3Vf~G-Pk4CZOZAVpoHaIjWlMXDABuWMa4G2lB%Az4h!gdxpr~d|$@Q6m#MB
z!R%+~h+h@zt$}uSQ#E4OLPd5_6((DZcGLQYQORgGKkm_Gvz@u_C3N}WON}1D{1lo^
zN(i&!VeDVPbGqtjRHo|WYtRSIlVOb|TeI;88Unj8ZCBV6wU0iQRbJ|}n7vVCi3>6-
zwRNVKekH*=P3a~s=cO;%m=Py$^9(=AG1?QR=Ium$D0XB==EQrU&#u`;X2I8asKWpc
zS(ukv@K~D0Y*GAjX-GF_rP@f~x2{&C$B#XgKTOh#p6po&ao0dkcl|~iS@OtDzB}lz
zNk%{E$KFeJLp`OBW?EVHExc#e>4rg__TnFrn=vdf$V_%-(h7s_khfWd=hOus>V0nm
zK8r#9@ujlEHm3?2lv==`!!-<AT%XSsdJfNPfM4|sEhisZ&ix27IZ5P>;hg(?O;>cB
z{EulG2&U(+FFVMYKTo;JJe8}v`r6s&*~?uE_R+^vp4qP+($4=$ncqks%X{cx7Y`L)
z;=w!x{#0Qy#{L{r*{``94x}q64?oN7jl5JZ(u-bed@A@Rnx3KF_=3-kMqME1y(<|H
z@dBMh>B*H>qT|noc0<m5DEyoC1$)YJnzVI_Nj-Lv<-(&noZqbb9_$C{4}Y(t^N-*A
z<|Qq1wa|&v9WVBjg$x$2&wW;%C~V_<Vqe2_n~t8hsU<yBSv>5jG#tI+6gmXn%XWi9
zliTA_4R^rL=(g`2%3PQp7Ou~u>f__U2>rIPb07^Kl+{<8nJ-F4!x_qb*w1Oc>cqTw
zQ#72K57`Ct(^p}IVVk<lhC7+rj=`52L=R0eJhRHF&|PRPCwM<PjuA~J<6|<w&&bWZ
zbm{=d-yKoYxySpILWcMb^V~eI+G(fG?I(K!+q7H)J56RLYXJKg_}aCbId!r=>{A5Y
z=)12xuh2707h!T%x_UOU7npp@m%R9EWPzRh8!xvdOXKa(bG-T2oagq!DmiM=!-XE*
zuaiwxTHAEeXp^}u^YnD47x;>&*MsLQva169&kN`t%@WO?()KV;gk3ZF*qRBj&Te#_
zBX-@tWmoQO4i&A%{Ue^(waPy7HjrP!t8zulxq>b>U33>8gsHALbp9F~bQeGC0shuQ
zhgR~wP4^sX55L56^yiq#v6*8($7+r+jz_Rf9gYO}=Iau)oW<xd@J)C4=I`uOX87h1
zd=n4f{Fy~Q2EMsXw)k(ipAUS~8op@;-_(Y0{)~Z{;F}I3;E|ztRq)L*_{OUTT2K=D
zP8>5_QOuc#@iu%@2flHGZ!%ioCirGBd=mrTbbxQ7UGc+;vzMw6AA@hUWg|oM2VV)k
zDFEM`gl}@*VW-qJ=ATY8+r6Ls%+@5d>?GY>NtSpCIpaA=YBxDatH&@iJ|Iam<MFTH
zn}iVN!ku&gy0SMjFInOnOv&y`)Pbz%GY=DV^(M#Z1kK%>ph_zf^mJN+Y7S3ON{<8?
znv$W(%igG+vC6VJR(Ds&YUvWXhNs2q&^UG}c*p90^<ouYJ(gLJSe5%3!_M{?)hR)*
za3ONR<D%4%BMZmvv2_2EBOYOhRow=$8umLz)t|(u);;!tUyG5~S#~hMHxcv`w}Nl>
zb&ZkBJv#GN5#aMjuPK2)1>4jmvup1fE-!0@Qh5LG?)XF#B9z4*sd3Io?T?An;@(mE
zZ+w)R&Wh6T$x+(D-yH)h#r23%k!<9JZ$`40J5q7SBK2bqeKk|@oDw2++Cty(&u|5G
z?V#K(Iw+w@2OX`~LCvqW(V9{%)Xb}eV!j3H?7Kj1trV!w?E*Bw8lYyz04+8Is7$>8
zWx<QuAb<Lt@uCX)E7yx=DmS;8#ujO&{O|qbauIE&ho4%s^V8-`e#-uup78AKjmwJf
zoR&F9?o-~O+w5^;o}dvd1Ot8JUgmx{em1}JPWWaMeB%jMJe&d7aPJTe-*g|uYl+z<
z7q0aqxt@pP#ndT|hf^7LFj{mS4afb)KdmHy8S9#XR&9!5UI}ew$`&%Zu+ER{obz}M
zS4OW+;kEf!pS@KWpUQ=E+*`Z)qP<S!{nvKc)y{|gQm%BL=H;CDo-Pf1{X=i9I*xXe
zVJR7AW}(KyGpCzd6-2IiHksj>ahziv^zFqm<Kd)lk9*5`<e#IN`Fg$EqW&T1D0o4|
ztCAPu{<G^G_C;N^Xf|3;0lc<PXfzADTJ*K9MUTwPLb>BJ@tBvdbqP-S!_gOJF+Mk|
z{VV!uUc((6#dy0K>@y3d>4SFv=mE3dWNdoFHS6G-p6A%TfMzrI5IYujo6+%j{#vuj
zEJ3rG1<OppcN)T;@a|^)isoxGGxf?w_f2CnyT#FqxOdK$(X8kX%<kSNdvlt73_IYM
zrSJ}OE*txr=+`7?96*<!J6W76e4JjtlIfU%dBg6gTXdToXI^THw+bxc+QnRWj(+T4
zh{cm?$-KD9TUF|M%TgB2CNDGMXg8NX(!clEs2*Gc7r{4g=Ng%DA(z^l`~L`bNHwPe
zhim2DrsRlWn?uY)?aD(2haBtE_g>sLc&YSF=A({z>FWQyba6Jj7kYSUUQI7KOT#^A
z466%x>1HLg2Ij!c<cbF!<@$^-RfD{Zdqeb_;=G;BQ>XByMm+FPd*-6ACZpT*^w4(p
zF5ElMbsyhm1lrBb7<&BRnL_wcKFmpVCA(5{8~K^e%q9i8>oB?Dx)sRVkaJncy!Vu`
z^zwCsCpx&P9P`>Qio5ALJvF(n!Y{iF3R`W^=qczm@W}1K20cNuxfF~S1-m?}Vo(#j
zs3UI7g~K)ZGvZshkg=&?&>uc_rItaR`Pf~a)7HnJEqGS{3xak0@V2TkPlacdjV#WO
z+vqrf^dFPOxk|52UV4yc{v>y7cGuzo>=i%kt~8m&o+9!$Q>2AtjA>=p5<327(D~zc
z&qKS=Z`PsTl!9*}c6q8Mj5s8-m&VkBAtRW#ALpf3`{7FXCZZ@AaCf8n<4;+ZlAXG1
z)Cm5sv*1a7^+wN&L*K)n>H*(8CZ|4Tyh+2inAo3W()wIx)rD`a^q`*rz460KI{)&)
z|8DGCYh_Vy?hO{_v?_ii`Y2xUC310PyVHfU+r~VoO*6@tY#U(LzN2={{j242p6T3*
znehQ!m(e*!-gMAgN$xc_yO1i<*@E71eGNLuJ9LFw?BNabg~xq0sw=b9BYd@BK6?w$
za2CN9-QSVT$>G%LiuBERImvE1*@;Nz1~17xk4(&ZG?_gx4mwU>{HE;J@t<y!$ANX+
z?y<`XrujQ-mHR3^edp*FKE`wQ^88I`HA~^0X-<_MO75m3x=t&nW}_25a3x=yk?ubD
zD$^>ME0P^FKj@|5dp=I~<>U|c-Oa$0La$oOIj%eBxRC4)GESUZ(ZP3Swrk)Ln?BLo
z73*SC&*J>7O4;O6mL7Iro94kdw?gS5<^1<9%&tozc4Z8JWqkQL2BL>|v@1P~vz7Do
zvdi>kWOT@0#UYn=4!QSq=m=g_NBsTfLF5^kx0=2dKa2a<?E4(La)O`xd51b(Ml<KA
z0^eM@>X0uC6T}hD(Su_c#}JMV9QiqtVVjq*O=*s`@J%Q9CI-H-!Z!usn_GBOe`lxM
z;hQV)jX!*|4ZcYW-_(F_+~J!#@J(*`<|ceI9=>sfZ)U<bKL+7t!8eP#qU+#I6@zc`
zM3X^=Z>F`O2gjeB4}9~%40pjdZQ+|z@QtH>N15vIx$w;-_@*;_lRF1_<+Oai@Xg3~
z_)!0m|3$yac$K|UC-IYZB;jc%sl*moW?7OBEu`}gzVVxsq~D`qqW<VN@J&wmW<8mk
zK=>vTd~+!eTtud(7lP`6Y~+UFn<BRobpCjPKJHIYN#?lck4R970SUTOgFR5~Vzjn8
z{d#lg^6dg2)Mf8d6?RcRf(P(`KF^Cp8$jdu8li~09M|EDeGzK3i#-H*MHSC7AAK-f
z2e*Xl<k)acofxhGqa#!t9!bA6LihGWXwcONJ<b}*y<nu?^k<*bm`KG;iPYsa%q1O-
zQ0C=4w@HNN6^qc)KjHeCz-)Era3v>(>y<BmM^c1B_<DWeo6hh}p_Y-l*DO*Y=1BFs
z7@=$%!<4acm`YV=uhXXvI`XUoyBON5$%XbRe5AczGP|^MO?yq~(q4}jhbUK6h=x@R
z(W>kr>|Smyqi?X1Yqr!oc2>?V)?9ag1ggl>KpD>ls!cLknbQI4xG+GYh6Ko`QGi~S
z3{W;tw3~9|ZSbakJ!+;r^O~WVG?U{EoegKn2g5fz;hTirPK9vKah%t2CUSG`f0$?d
z&#5ExnFGk}tBQNkC;8nU0k;%j$AM2x&J|}JdW&~74vnSXT(WPo(Nws1xXE*(&~03~
zwr}T}-VZOPWp4UjxQEN`MJMk*W{%*SE)VGHg>M?eH)r6RRQM(ozG(vAxMa0yTNvlM
z<LnxKz-##_`=WTAccWwYEt+g$xF@wNImBjm_I~nu&S2Ne&o+Iq+qA<KotyhY=UMh5
z;PcNwAIjE>42{W3Zk`NuM`oP-neT$NCUDMucN;Aornr!ceaC#?ul`5hZ$q>8c7sRw
zm<g>V-8j1TCX+p0^l#q#qo+l8VIK$cPg$DcQ`IC(UE87$74W96{?lL{^B8}NeF}G(
zhq@2DaNL1=dhq8kzU~f~W)#nPaFN~LXVGs?(rt5uPU3xLE!x2@@J(jTSWV^z-DV_g
zvuz?d95nxc-gM$5aDUv6IjLrF5!%g8G7ovmn01V6%Zp#^UU*5i_y#+rj<Qd4GrOMW
zo9MPTanEB?-Y}D{lf7Bcn0c4-CbiB(hw(3OjeSm^4cVL6L*DF=@s{gc_6&{2kLp4G
zCc;}U{qd^A^YEtH7x&he?B2@s6OZYYQLcCJp-wUzwH3`~j!{$kaeeH_Trt|s1_$qh
zZDv+vM!XV_IpLl3WN+}K+QBv#@8Un5_tNa0?6F3x+1u4i>5N|L2HRB6if8hf`6%X@
zzi};|{l!z0*3omb-cvo$Zi3+(*Xrz+%IhiTFAp7qZ8o3skZpqp%uk+}9Sjb44}IGw
zWnYT_gcsGVHGI>Q&KTHcAeogechPG$x+?`ws+iMVfwkG|M4wF%46$dXo7Un9%{IB|
zqdWQGnsn_IqnGB1K`ohas=%z%yxDNb7&>eQ7}yEKpF<6Dv5+0Ei)P~ryJR=0UlxOE
zXEJE_kEUw%tts<2e11uCITg`rY8!OEG5iA$`Ql+2$mbkuXwWpC)BR6V?SPpgPT^a9
z#={C@2j?92bK;v_{=`gGJ#s<JQ+0-KUT4DkgWFR3{p-KE-_%22nV)(%$3ykV-{fdO
z2M)7kk(0>Xz&Cf<v6+(5OYdqhBOb*LuVjAa=x5*2(_GQk?vatQ@&EX0Bm2Y9=KRu-
zw<^Q$1HS_mhtl=9ggvZP@jQb}nmCHzwT&j#g{gmLh3(<(Tl|jBoNZP+e%G`A=8@Nb
z1@0*tCR=3xZP8ig4;$3LGwxv3S~Q&LBWx;&PEn<vU4G~f`S;t&+_Sf~8ncmp4sv5?
zH}jc=-hgh5rxTWiok=B_v-G4dViDaGC)gEvmdxy3^pNt*hj)Nwn91xkiT=Me^l6>&
zRlUdbP^59{c`<fbHsG9xzT$_@Qk8SxWONtvVlp=yo$A5)u)-0#|4y;b8;&^->*Tx0
ztkrq6m@}{s{F5C&>dy}5xi>hq1>L408qoX^PVI)5+O$W*v9S}gCNo-j;hgvAN2k!e
z(UT%tz*^`xL*Sd4_(2=KkRgO`I^j!QMvFeq>}TnL=q@eoTFtr5bk(NqQ*AJwO*YPp
zaeJ&<m&K-Tf2{JkMi0e#_QAY_6A#hVMJDHYMKX-C>$SzMg*NUzIVUHML{mp&pR>g-
zFBs<$_lR%tz1kOqSIRPHRUNISp+k#DGLJQt`$#yl{a!Sr<LnNGZ{}ZO4hz;Pcbbp$
z=fD0Fiyvj<XwDJL;ovCEaUQk_;8+FQT*aFj4d49T6K;ZUO2IcT=Q9%p-^9W<1>l=?
z@J)62CK<kYGL0TT_@)AUQv|;GI1Y~qzKMWuUJmE)fNv(iH{JR(D+S*iPk@);o2t<;
z6nrxVzBvZpBs<90SjZQ{H;v(&jg331IDBJ<Z$1>G2e1HNFE=_*b~GsXW-r>!h1W?6
zhi{(VN>b$cBo%>gUc)zo;G3fGO+EO=0N>1<fxa^lrWwk-RA2f4JJabG1{1Z$qk?a~
z8Its{5WMpqPwF83HTX>X9wcxrNKgQLQ|~~64l&2w1HO6QKSBNBo2Kwhp~^A3KRrqd
zI+M4p7o`dCO|narray~R%42pDFxRx?eS})WHwDh1wd{>h*6|T)nlnPl@Xhjl;fmWH
zu9ow|bt#V7>ELkfH-)Q^AzWqigv&cM3|0?Q*@Iy!vW8je`C;leD@+4;EHNcaN2iBr
zaqcjk`_@5?(}XF#C0s+p!u1Z$Nf#Ba+s(sO&ls*vwZj!!I$RBNh0Ei2n4Yc=lNVoO
z<?wd;)2*F4M6^@N$56dX3Dum7p(^LmhHFL}wJq94b27D|%Oiw)#1PHP%-qxa)@psL
zwN|qO;b{6+8vH6qH4V-66TaF0Cs01{&5iSc?E4B-M);=F;sB+AZ?Xgi=$31M?)LQ8
zIrt_di$D8Go2l)#X3E@#{LO1W9X{%(GRb6yBm9(w`-~>&GWBzEpHQ0I-)p1B6`-Td
zhgn?w`sLgke_Q0CV#7W32hBfq2>te}$wd4x>cM%FlF$$q2hqENUy`mPbIpUu7G9yh
z1;#nWas4aT=wfIc9^`k?Z%UlR+qp|N=MFg(j?(Z)|9IxH(OX);H#vD7&xCL8!Z*uq
z(nAB|%zbWCD*F`LzO?BI$Bs1Y0%xaXu`f1yj%|8_pY`Yr`Qg(xwK_>}%|SfswN^S5
z*pC25J*|s2l$+e}2WGp=lZVM@QQuc)<>vbwzsJnZG5W~rn`9*``8_S{5NFoPEo5HM
zSNz&rbZ!tC9sD)VT6FoKNB!NQ@RXcQLMDsa@pe34b4@9pbK0!ecuQ$uriZ+p{XCi9
z3;0TZIfml_Z(Dd@PMF5~6j_C1W|cq8&j_x`vJ1Z9?>xMod*GGy!7k)$%`~g<Bs7^3
z_*Z?=`a3a$8*bL}7WCp+@xQp1q*O-NDS{4^g}*b^q!W+HdS0S8f4@oD*0LLXCNnO>
zn4{`s(&aYXXOq2IT90c|Y3{vquvhadGgFVfRs15En!VmK<2{t0!OqNK+@p792DmLd
z!+p?d=)5_CHx)>qU;m8W%Kh2MZY86tUty2tA)_8~ExbFEY;iAqDA;BTeKx^pH|Oh+
zAtrCr0=Ai&fqrAOo5ruuZ<zZYMds%1QS<}mlif!1zBoFG4PHtufmelIvw(e4-eie?
zGb^=mlczS%@>KKTp7I#tsea^T9i=^)q4rQ;dhBX32mO|wn_&yd)snk8YVuH-#_X76
zZn|$W^Wa_Sv$50nTbZaE{Wkx-bk~FH?i$0qb;bl9(L*X%r<0gG&4z<+YMbOHPY3hf
z_35G6iyt+}prSos6*9xW&|K;-BRezEpq)bus@m7U{$e_a$<*Yk_fMmF{kEx|zHX{1
zZ<^{IZwJ0_s?;Y<6_Ohd3q7X_4AY~6K_2*3?(oonItDp-dtW*9oE!$#z1dW+cQ#e)
zO--34HfSpP(YJ7Pq(g3cO$I2suDizLPxaVEz8U@Ip*P*jJ;_dI_K*wnQv=E0WLoc`
ziN8IRzlNtW!8e`Yn+`iX)$O*YKBo85A9wbW;{RoZZ`z)wyZ9SBHhJyOcH%8g!B^z}
zbR+y~{|)n$U=LXnZw(&ett$(?wWTuoE`O6wp$k4(V^ZHcCPin$hlRJjI`A4A&D#gf
z8pnOkyaM>T9{AH}q^n+%$H_z{t~z^Xy_ltppj*OW)1AZ21>wgGZDrR<JgOQ$>}plX
zq4gGrYR55mFpPO3uDx^CJM;|QBXb^Rm+`2|526=_9F8xHQ}-76*pg@^c3<s6D@l(}
zwHOVj)oyZw*L~U5!E9A-e5I;R{U1wb8P?|3b=|t1y0_Hbm6nn~W(Wyx1qzg60a~C)
z-QC^Y-QC??Pu<<!-OfAuJ>QRg1yZ08x%b#}%{A8>E48U;dZ{`FC>A!!K8-w@#R0kk
zvshu5u)XwB9AtjfVZ1Y}QG8r$A6m^$>QkHOlUfD)EC^5&*k<8yvXQ&<c^h=0VE)ZO
zo=pXOH<<$zg1`Cg7JAhAT}Il-G0n{HhcCKZ41Mk&(ZNUf>QGbu#y5C}=lJMYx(~Ao
zVH3`Bt6`R^<H%GGAj2ygS?g<=b4T5M_ckxpL!-`e9G#jq8*RsBpSSEe$pI?{8}c|>
z&3JC;xw?|T*>ezD_*x&u{)>0uwU4_0@=*pgu%>85+iOz)b3~JeZyrn}D+iA3HOp5O
z7E&WygRZmPS1%9pc%R^LKTWL)j;YS7&vIlnWR+q)g=;3UO0e3&H`Cyo4)Bc^d{Yg+
z`HFUP0KORl-?+gy@9^G?gm3b}H@)DS3-C=2_@)SalNG+X3E#|wZ!GZ5Zulk}eB%J$
z-002x%<kwp9npMXoF9p3M=^MD!rRFR-`s?6l0C`1fp1E}Hx=NUp^ag&^7M<tH?85D
zp9RSEgKtj5H>2U3eelhZC-h9gH`Om>DD^1ZvX5Ds@J)`j<lro!2OPdRG8G0Ij)pT3
z27+%U!Z!mG=?M?Te`EWf?<TeezMU`WYKeb%XSa0KN}(s}eVRr*Bzx})I?jnSIUIm(
zws4<InpSq84y9xb%#QDu+D_I0Jh~0Ygn@4+!#CL<x7K9%rqnz9FYrxI_~ytiIA>9!
z9_Ga}b2fq6J|1BB=3n^cTuOp^L?p0ZCFpFU1YNuqudn36tlS;1T^pFAy&zr#rpA*G
z9j_%F<5f2#UdON4HR`-woA%nZ|EgVcrrNc%i(P*b?V1~8*St)2)sBeMnZQ_eb&S=?
zlCe7WD@LjJVtC)iC}LC$*QsdTTpcZAzi9pJ8m;VN=E2h_Y7<dpCPYyaiBgY#QR<f(
zrDNe_1;ahIi4h9<6skw}LiO!*s5Wd5RfWh<#pMXqmA4_<djZZ_AEKfsgY|^0J^${(
za`6t<%re2+2H#wQZ)W}p((#)?>IUDe?jNMQ<{)i55J>+!wW*XqRVdF{r9S6;d=ZW4
zVHwFeI*~K9GyBT=m0SlaqWv#1%Sb&xsyF*bKjzz9;y(5)pAF1Gyuq2BIz=jH`odet
z#HB`3xip^Frg&ddm=!RAS&zrqi<_c#gi?dY&#QmA*1<Xz;f9Fbo|@jvQ#<J?KTC~%
zRd+t72l6${a^M!M0^hWUZ&txKd10HjthiV7fV1v%U+i6PRr$+7oOQaf1|BC%12$?3
z<2)Wgua=jWI#eSU{s*&g@DLxPuCxTs_&a}lARJR_1;1ZrvIxJB%i5dS1@J<)jvk6w
z0C$|gBNNDMOnRWYeumdF@w@-v>#x&~{g~@E*Y1PpQ;DpaFw!;nq~|s~F*~@n?;&q*
zzlVy!FG*-K7g*^mKR*9^-7me5+7oOw7rkZz-}h}D^(i!)>`U+o&n35bnukV?!+Q+d
zxZ`KI#P2kKx|Dw_4;cc`a;O8$V=t*&363i4A+OBT1K!$H>W)pPPjS1$rh@c^@1010
zR6m=#q~ar{p0E_}%?h6HRStNNbJ}$7qg9#jSrvGi++w^pkxTH9PNX)~&#H=P%nPR{
zc##J+D@UvDafYu^kZc_GyTWhqWl*~qaGE}wUG#*na#!3GW-N4Z*RL?<V)~IujCM1%
zmb*@&3FOI1ztj(lE}`3$ykSwU!xnYjZ&5btQibr`6zzp}6N=Zy(V}bAq}*Yfv$x@x
zYi^3%@21ZS>Hlu+CJUO4eH%G_aLw|i^heD#>)k|{X1rO$;P8Xo&V+4RyORInXx8;5
z<lBstqLQ$Io5=7hMqA)7lV*^Q=X@E>W~oV5W*-d1dvlS#sHo4bioWbBJD%Q}MpvE3
z|FUM1i;i{1E8N*AQ@l|hEJp2*HEMj4QN|1x)xk4d7hVY~Nsa26QS%=e<^9me99yGa
z@p;>qMj2q6r8kYbb<U_QXfquOyJ)Hd{u^|ju{F?d&~qXx!be46rN>6}OQS4njM_Qd
zs5YC7I)>M0S2T60G0Yc#z^to!CiSL1)e=wiPU>vSuy2>fgVS)ER3?jA8(N{=TrsQL
z7c<!r%-4*ghklG3uTMU{;imIN;2WDoeLFH+b21smhb(IJ4IfTX{NaJ_+AxV6#&vkB
zF1xGzPiAP8vT8a0s~$;Y0t~T|F^5m<kW~vlTh((Ry?6)68Tf^lt1|C-UZYR@QJ>!8
zp&`^>Kj!mP_d4_#&}R^Rk<7neo_bk_J_Gn>A{x$Eda0_e!LwKUfAiipMWF@L2UN0(
zFa8i;r6>C84>hX!GwC;2MTW!+U(G|uSyPGpmd?y=Si@|pLuev<=ym==KTu<T{SNil
zFzQtP6PUHRmfXKn{_6kCU$e6X=qlH|3(m|Fheev9;cV@NmNS}~{0wSOi(#DA=q?+n
zOKl}vcpF+a>*kgK1#n-}HTZ{@252Dq_Eu_A#i(y}<@=YW()-<#Ty`t|n>zGo6$+5?
zgTDgL`s>Ida(Fu6e{M?M>@(S<8<=5<cW4+o)My`HRirkx<C%{_C;BK7@6219k81w)
zTc2ePaDunO$}(^L6LWoAd8v9|de8=-OLrvWqMw%v<1Kvt9$mVXw@x4N){5KS8ePjr
z#hc+3u#lM($;_JW^p!64VZH`^uOEEW0Cr5F{?{QlnSfR4(?Y{>#``m*6*~PuG6G@A
z_B__}mQX+2M&AbB?n~%7pH4C#43;U#D#j|#`uq87m}UzrhV=-xF|ktMo0jlR6Zj@K
zd~*T5nE~I#z&8%?%~|*+9lm)#4UGoAISk+Ynh4{-Hy_FNI||?Qhi_`ZH&fx8BLndi
z_akQqzPZ&6Hi2&n!8cFgn=tt1Dtr?I-<0-iryic@Nbt={_-4NozGJkT2>51n8S?+&
zn`7|JCirGuCfEwT*#zIbg>UxUW?uMte8xw}>w|BKZsX%McyE?uFdvASn^Q960pDbY
zZ(8)E--`NFq(AdBEzHz(B-^+O^{Jxxkw2#^_6nZjoy^GWNmg;=bX`NoiMpDm#4~An
zb2v>M@!|w6Ow%NfG`;Xi(Mk7Y{i>0SuOSJBrLU+bewwyzbd0*xFxY1B(?oT@N&nKn
ziDY9Zs`e7*EKN+*zBkmS?&Af%nxM^-6Yz{CsC#&VKDi~Rf9V9}xf-vk@J)XBCg;X@
zU0f8eL+CXJN5?CveY^%Xk5@1F#vi`<xZ934X;+D{c6EhqqNDBlTF|cTJ>nE-jnnSo
zF>25|M)B=q)UtJqDz%JJY~~n!{uHei3!-&;I=ADaReV^qexyXJTg7M%+Y_bZE28wz
ztSCA4h|;jC@XmQ=AM}k-#FsGby&9%x$HK_eAwL+_Nu3s|*!rQ`@hgN3s1T*XH$P7V
z>-5rK6;BP;oSMNZ)B}&Pe{=r6=BoNTNW(cJl%{84<Vi1Po-tcs2Dx>^>0{_lkM<kt
zBct8P&~R6|V(!|qj*JoN@`E2TuZHvTiE_;FnMa;RQ9O%1*)Py=O6{OGmwLkz_~smZ
zQ~Wc&3ixIid{c*g_IpWY1+e%2_Tk$0!KUg1Z92^UdJ%1BFWN?NcxM)Sz4<-aV1N6J
zzc4GBpV1y2ya;>fWi!3X*YK8bcCnz(`cCs!VLWx&AEVDuuX@65armYeeB%z^6oGFB
z!!})5z2F-od}D!c7E<RwLv8BA6MVw3O*s0v*LUhe@I|fB%xHoQ8laC<TSlMr1FL3t
zK-Z`WV|3vE9tlst8Wmxom2imTS@aryr$Vdvz2T5w=rH~d@Bza{e)LA&%z{rD-DT4<
zdY|~3X7I|oja=)w##e@Ua>Fs@Snc_^Aim*$V49Q-=w9eGt@(LbmwG720&@Fi!9G)X
zjK+E>0Jhl%+xT`y8*7X1AA>h1*h61!=s?ZL{)5G4!jjwa;L-WPtcB<FLtW)uO>OE6
zwW%ufg=fbfv8;!U`N{OR1oQXdy;+O*<_O-KJ6UYX^wz4zH?7P}rAD=ho?p)9`Nyz_
z^<Xwb8?xL&s7LX9c-FJ(fdih4+_2C$cm4B}+_%fjQ8-MN|7QAZ7L!Ljf%`k+vk8H9
zJekYDYicFj{IejLH-E^Gdqd{RRg3N&M>E(%FVqH$YQQ#!@qe^x$7{~VqJ3yL&FNnb
zy@+3!T1vBnZpyjUO{sWC9(qu(D#`qXP1IiS&IZjkYr-V6`VM2BRDZKpn#~HeGCOmH
z^o)8GM?;nGCshiUzBds)X5{;!bqxDxQu+J%R?aiCl1$=`2`0I?0zXyA@uT)Li$3fd
zPta=ayJ+q#dbiuSXzLWCMyDG!G}5Te-bPidY}D9tMg`t7s8lHzy?6)9+%Re(x{KEp
zqgJr8!!p}17*+O^k?c*Q(#{+4I2rXH{POUdQM=G`rolv(@_2H}xF{W6r!PNe+<v1P
zE;OpfSR)=$e(qfEqi?(^dQ#F5SDjr;-3$ghPUhc0{6keTNhjT;v@X=&HcKN5n3b4n
zRv-8#JA88;|4q>VH|>FM@=@Pg`5Ja~fFr%pVd0yF6D?YC(863{coUs%0{YF^&dk?b
z{=fX2Jv`rY(d``Jn_r1!KvJVBwUF1%0cx2~$$%`2uWOf0kFVqBD(fL*D9qlS+4h_r
zD)WAC{Ek~R&eI0`Y?G+xJ|uIh46{+pUP{dH(xigkitg{N5@-)@^axc!zbHg)<qh7E
zZ3g;@@QbX(E8;qt+`s9*I&sxknLhcdYCdKP7@09Jocx$E^o(!AFZ+^t1La{FEAyw&
zaB>aySGk4E@!bb!+(M`R?yr4C@x@R-n(s+19t~$^2eg@i^l*<0P|=wIx`8fZUB=w1
zmGnQY4$vI5n%*m@Yb}9m<}=d)4%$mi>+3Ll#l5LfWd!I}3}531*ErF0RSxfRRy>Ul
z{WW$k^{Gkz%);>3%NqXL@tipc%luR!&QB@u&39K{l`G_{`-qGC(5=3fg>C33OBzHE
zczti#igF$Mhm7a)WL6lx6b}zNCwnO<fm%L{^JAiyMr@=8U(Z_&sPolYMqk=qX3^v)
zPsr6r3-RM5#QNyNTxRMW^-*c~=G}MvlK<h!!H?4owrW5f@QR18_Th7W*UncHVaZ?k
zo}J;Fi<^9vV=o%cQM5EPoKG;!H`Y(q-x&<ASVvjISq)e_V4D&w5BSCg-&BEb-p<F1
z1K)IjZ=B$pZ`7vN!#6G9o9pmR5PY*2zWEK`{2t#<&)}PF@J$MQQx3lA4c|<JZ_?nK
zJH22f_~t2m;~vSsw?OlOYi<YN)q!n(d6OOJf!^Z|XPNOO!#8j1@$+h+<5cATDuKpB
zeQIGY{{0U=n=j}wAIQ3S1Ebs}x8`DonjXzilYJT5us%aR=r`{cz%jGw_nts5@i1nD
z4`3E}C$fuMX5jaMdqhuFjts?qXJ+Or`nvBj6Z{OlPrK9gWI8#;<I^>&e!6bdNZ0Z5
z>8hJEoqTQPVNOibmD*`)do@)vd#1|RFjd#+dHPr^MGdmDUNIM{bFu<cl9@eCKHk38
zI`k}2*U)i3<E_axB~i!P;H{~gD32V8@_Ipi>T-exQd??|c5|;Hd~*e^IUKLRU9b#3
zndS4SLj~A*ob0;Y+pZ2R?Q#jUYc(Gi8fw>vu6AZm+LhIA*SJ>v8+_AaW}F(1kI^Og
zCO2#|C?iItlVem5zB!gHMzudj>*)MwU4U;|!8hK6qc#8k^Ucz|QK|{w{GNb8uwj(8
zpKhi9Dl^ZCi}LhB;hK6SOnZ)oX?goFX2OQ5^^8#28;3IgFO;)Yh&HYWQR73wn$js)
z4+{kA%IxOkzmluqMxHjh?(k9gSMa6G{=<A{`efF<vMTK{`_y&L;as0yt-xD4mHS<+
zG8X4L_ST}>XgE86x#=r2hc}0ln^BN{i^o#tq0*r~=-pf+2T*&+ztNpK6aG@JcWc>0
zpY6f>@s$0Qb6kfy^a44tpU<+9lZh_`zsH*!R<&C|?S3>K#a?(+Te7E8tGHU2-rvk<
zM&I3(pH(o{qNC$2>Ow7}$OC>S^n*2>teVn=jF>z0Qga^MxSv_zH@$W92Jhz!%(FS^
zt#Z_+PQf<SsYie2z5kr^;U{{fnymewcepv4$ThCFCt!%I?E7U7+0-N)R^fN5^e=T9
z7~pY5dYV$H0d?}=T1OA|0KC9|Ib*elVn=wOfl&{dL#+w@#fLi6{r7OtXZ!|Z&|T15
z(x!Xp<!pXFtg?SGy;Sf{Zulh`=K09l#@f&4rMdQ>fobk7VxH7|W`pxRvHX6|C!h(9
z;{L%Na)xb=z&5>8>HD^OD2BS!Fdw`;uK14|c<>rwPIwXi|Ezp1nKv;H$htXaQ#EQ+
zkEl&~%(SWHFg(VcZ8{KVQ+>QQy`7oKz;kgI?@e}I6ZP<k+OMF|9OOK{)=FOkJTr`}
zo6fL~9j*zWf69nAhyI!K#h9s?m0X<nFw0%`z_Zky_PA@_YU)&T+?6znGk*rz{$6M|
zE@a=-B<Ifo|3+?la@ad3(FcAVw)wCNoo1^=Ch~3kXE8f-h(*EpK>W-W^+CHC|He(x
z$I%Fxk-2`l8~t%)6;qe;s6nm`{m$q9)}ZE)V~1{&Gy?xDx?+w9vkq5)cXkS$fVUEQ
z%|G>+!H}D|lhmZj*OuxQkS=^PDdUAnrkf^uEKPDn)Y%qi(jaG(VvCtH`=zTY?xTnM
zyo+`Zb<z9?{4!tBQl=Ww!;O00(#Sjsqms)T75dm9<2i%YAB1s!8Ps8?QMF*1+%U`!
z7-jAWqgpO8s>K|m3ZTPWK926Q%P2E7ry946`UT@;WOh+Nemu+gbO!u}YtEwg@b7cs
zn(y#b-;PEyYWSXMc$KZL+S$idIhQfB;jAlzvdH8~C)fBI*~*oXMY)7%IC+Jiq{Ge4
z@+aT$7{2T8@Xbko7-uZ=8F=rGLlb*gnjSTaMV~q_n}PS*c=*Pd=cqsdav0HXe!{87
z$;{56_W1rIwdf*N8C=PC<TbIWzg6k;tSY{j{KrRD`s;1-+ste0GTyG-9;%MESTBv7
zCf+AQslCp90h3dsI_wDB;D5`riXMZnp4v~{xbS>xF4;M|wB>zFP3EtLGmE-P>faf1
z1M&QAL5ryFtH$iVcM|XyPaqrTh%fUg$m7jMHcNHp)6w7T)eqnCT4rY7B?mCSznV6o
z78K4*N_;pA$oyNkiFxAZsZ-G_{wOy+PgU^UxCY286b%LqXD;=n8$;-Uf^mYUk!v%X
z`M>kX7KUd|a+^Abd_K6Q2%3$340ERV+DYg(Kie@AJRW~eFuhjh0IjSQpciO2M?cdC
zewn(|dVg{a{qe8+Yf2e<yzY>rHxF+y{^HxIz8cfmS5I@2(?`F?1*;G9fxLB+>)6+W
zWUO;OtK^Ai;5z-LnZ1-mhG>G1mzGd-4-E2B1z5*|ma~#R(fGT}#Vh5lOk=z?b33zx
z(Y?1;@zF(m$_+w&w46Q=&qLh4gCXIY<G-nYW${(l0=`<*i2OfaU-h%2;iRM4^nqnK
z6MMiniJN`Z3_s4DL%v))eRT_lxySn3C-wLCJZljvj+K)&4Yv6P+x(q7)fB!d0pDDM
zZwA9RKJZOp_~tTvGZnsZhi?wSH?Hu_0r)02^{Mah%>(#mEqoIM-&BEb+QB!;@J(a*
zW(s_>1io=bvw7H(uMI`Vfp4zBH<RF->+nr05A+=PCK<k&+ywrr%l{AG+=g$$OY`qV
zV67bZlz&rif^Q<>o9ggQ=jUYHTxJFp`b|0bW+8mj3ceW)-{gmHCc!sOL-FNw&d`k{
zW^0CJXt_5ZJL4&aZyJ52m-_`gLofK6qs+nF#vBIvzcZ6*Grui;;c@AD6`ZbQ_jL7s
zjK_Fbnl^dSTU8=Wt0tr>x=1RsyxMATLR$?9XseGNZ8f!7TU~A3R>@V{s_S}uy^HW2
zw@lR0T8YYAAW?JQQIk5Gpz^B{^m}1~>USrnCMH3l1@IaVkC#&*-kKr!ZF<I|7c+wi
zX1P$_u0|#7s!@pkrQ&v7tY_C_H#nx7T`Obk3W065gxfhk*;OCD**Ysud#1&x{fHR7
zY!{>P@Qoe5iHV3&k?b*g`aD_*=r=Fno1f#Nbzn%eGQ&3s%y<fZ7Ny$D@!_~e>2Ccf
ztvlOFAK)7!eDeanX$jvf-5I99^e{br6siurLN&5ss5;#W(WUJn^0^tTJDZ!M8Ikcx
z?X+o6Z#8V`t@m|3)TD}sf=aQU&ZbT=iEOtn?CCx>l`c<i|7)&I?CDeBo7yuttHU?v
z&~KhT!gmPY^oKLb_GN~l-3{*%dH2`R{K*Wt*x9V^o!t~S!%g?saxKJvm}5BnK^=bJ
z;s5zRlF$<RRNx%fh<U=JneFJoJV$S<g6q<EoR3`Gx9*%{nUNU4*}bis4&V(4q8^q&
z-|CFJ|I6AhFx5?O;hr}5fF_gEG0)ds9zo2W?!f%N46C-`6M40jYb*7Ze}edX(1%>Q
z(+kAkIS*arpdXnn)LE9)#upobmjnhIKbibG*e<p<-bkZOr(m3Op442})5Ez|Px?rY
zP!b&GfzA{~eW)wf=ScpB1bVPxm6a{2F>&2~i?(uP1|A*mOMgi1?G^qRSY+HFYFIE(
zPyYSRM0jQ@K4TcE6xaE;)STYVLl1&wCUEP`=Rq*d8h%b9U)PDRzXl_X9gZhxAY9ao
zS{7{c6Slb;4`)Tt9~FQ<*+NE66Fg5f=*2?2x%LlvYVc&#E9wI`ZJKxzkMwr<hT2s3
z$;_OhFMLymO`TdXGaT<thNDe;@ZNmKdsF*6^(Y>XQFw2T;=L)rW10%<-W+IEd^_q^
zQ9RCG^aMAts(&RIxPVnve&fw}!Q7~8?kaPP%$qIldVzLxdkQtHVf0G1#j64PE^dbR
z=5LltQGBYI-4#iO-~5a8M&Tv?wc4V%6=(+Y|L1Kz*bmJx#=_h}i#p{cZ|<p^GBbO@
zgPPR0k@QE!l64d823JvwI%?KQw3<B3lUm-)tm%CGhuZ17&*(BgrQ7Stxa){jqbf5e
z3z2Pu_clvb%8K}jpP+LTd1+E#w3~I8@Z2mnan?Zp@HgoznKJR@-bC(l(eW+#VEQo|
z3g1f#J=_)E7<G3V+D$6;CHN-56V54P)Q|@TmEC60m1PE1;I>f_qny^D^Q^>2OigL>
zJR|&ORNN#qn{h^c8e>#>*5SoQ{f2jrk231~Fr$LFuf-yxj-u1-;p6E&(QJAdaUG-c
zco?<1ol!kT;O~*E;!|DK^>05Hb*e2pnR#`ZOq?>zqryYvxdTS?lCmC_s?$SNaH^S}
zAhUkga?>wnP+da<`?=mtGq1Yg%O?}Mjzt4`f0b&Dzjz<srz>bVFsXAe#V+{f_ze8x
zN8!|G`0=Pwm4$Qr#o#gS%e7)Q?7Pz{(=%A7Dd&V0=(88dBFMsPvW|x)@m`rX<bQ1c
z{4Vba&Lk!9;#=r53>(2Md-lYp^orvr*_FqeS$@=@PkHNFaWn=uA34zrw2WS=k86BX
zpW4km`WYUCvCrZY$+gv2u{Wq!;ln9Vf$TXiW-E1NX7*xw#xF2O`#<IgRP$F`Z-0GH
z@t17`GhmkaYwZy<lm~qNpTE*dGyk%2fGYXov55`PCOBmyT1@;<>Pqm-G4z=PSjK4z
z{ZHJk;PZ~$|8Q6Ua~{dT=@Fojd{5>?W-^2YXpaZn<A}$&9ITX`Je}v{@Zf8HiFTBu
zi@%n8_-k1qf91SH&dg+fztQ-mVVi2XeD&&rk2cZgox=rf<t1L5(bU@U5-(^^?*%>e
zAIH(VK<&OnGv?P#=JBgU?<smtc^<!q_!rN>K_&KjsWm#-K|k`p&~OTzC%Zit`5kz+
zm-|wiqUL9xj5p^RHLGuUxZ#`TkI-=N<1B56X5;Ov82IK%8hNVl%{M$apI6{7M!$*Q
zO>ftJau(srLom#7)+yEr)*jX%mJ914Y}1!@AGRs?mvP{mg7D1|_$Ca#sRZA=rZ%++
zzUdF&7~q?=@J$W)W<PxM*MH*%-yDH&T<D*w1mBq9n+}8VGxtQ3>53rR4zF=*a{Az#
z;SuOLA=Js>n{)8ZT=*t0eACW@Odc!!UGR;)DGR;{hHrMkH*xSyclahb2W*CZ^8&sJ
zg>SOJHyh!b*OxL>6TaE7m)g{Jbev6a%q+6`rZR7HWQL~nM(07h+0Z6K<3q^Kfp4B0
z;hUV~`r*6j2;T(5H!qJcOJQre7A<356dKOPq0G<hoGx?gbmrrb=VxSQ_<%H-64ErJ
zd7ARkC;l`#mAnVoCa$fb1KX;qCv&4lrRYFpiX03nT2mrLou(&iQgE`IiZlEBSdvC&
zYpsea5_Nb&q9(OVWCjK^vl~;V`j2@`5967IZrAgLcD;mG>U!Aqwx(UyvUa^M3bz!n
z>k_LoYkz4w^GLZr*si`U?CJ*FTx)KpmS@-by>a@ChEsA%jNFFBs9~QNt`9NVQa(ms
zPDInA7_IUPqh+2Et-E8R(IKPtt9!I=_(seAJPQALlzuIYQXlxnS`qKfu2%TcTPe=B
zmG)L^r2{ux%CS$l-V|ZI4O8{^!78|h{F_Q-h>!Er`0_sL*Mt4jKyH6N`fi@tH2a!O
zCDA8Z(AzL%8TE=;Hr*Ucj(Q-uwhh_$a^ej<XH`!4CIYr;VsTgKZ@f2$EwUA2k2#En
z-IKF3eA6MDn{r`2saT4Pw9Y1dX^#&F_SrUz+_2s3XEn%e|6|oZ=mxVM!Vce9B{{Rr
zWxiuin~qU~s#cY{M_zLM=@I@$t)hM>Y7)1|=6K8XHw*h$A!-mN*d_oUX%y%8AZlRz
z-}QE&=UBN8bH+FO<7o|Z*O^H6(HL_4<Ee`z(Z8DNu9NN9Q_<Ap2jdl3z+8Z<<gnzi
zYGN&N<!zijskJ;93ESLa-~NqOwHr@u0PJPJLkssb+`w#|16E}ul6&`txjI%nF6r=!
zJKkVlYDcg|b8dS^!#k|jF?e%e5pNh}Z96yx7K!VIrw1-6+#ii-F#T7sPCb|<WW0yw
z!!Ly=lUvUk!kRG&{z0pmh2}GHw1;l;vpVuUE&4H|kmvGoN4^fWX$f1kg0Y4Nq3KYU
zYRz-(P>-46=uC~#ZkA```#$ryKgMUn`Pl?xo~JhTh1%2(p6}|tZSqO6sa_cMD2q*r
zcyCTP*i<|h{o<dj8gdV>=xJt1ZO3o31kGjw-ikgvW^I{26^5?m4*S%Hfy%%*x#3u9
zi6tMoYu9=Fr~BM>V-4Qo+2r4hWp-vSyfy*M|E}Y%eU;s1MZ0k!!>{xQ*yfH!caG2#
zMg8;ROpA_<qBnV@Mg4|aV0Viei~Hf(zxv(w>l9mf-p|Z%Q!6w?a|`l!1L+07VAgQ5
z`-;Ldr#<k|Rx~Rsc`voj)9X!Nbe%htnAVU>97MnL4|>00nc?(Eb$ATJaQp75Nz?A2
z+n{&cT0;hK3;Myok`?^UMc=Q`({vCo3~V!S75B9xPdJKsmj*J2@y>+7H&417wGY1O
z)C^tan?WIa4T|1ipdZ0NF1A7DGX|ZgX;kt4MlFM15@s58Wr9%-XfcaN7!^IxsP5g2
zs)~<iDEE8!gO~oY%uplqGx%B<sOtcun!_`#ykMG&)Uyg3wSoVyHJ+ciFr${E?;P#P
zEUQm03Jr4AE^>PE7Bs1@6Ma-`nccw5&D~@%het?0=vXx$!g_g`DVj{q-vqOU;9ZKm
zNv|0_XFbZ{(RO9#dLxUTd(neW&G6Yz=4{d<?%CK~+gqXI@Vq`f2is7i>RFmB>qvZ9
zy~q!kiH^tl!VcfuqIv5Z+UIWG|9(Gh3aLsBwmH3ny*-#E2mjw7Zvij<nHuCu`H`uB
z*Y5rS^32h1icKQN0q<;{P;X^LlYRGq{>M5#cw_LQ^d!FyU85HJ@XOM8K<J}(WAB}h
z&$!)LU)B9U{e&!+v1Vo*q>)KS=HJ?*emegiKTb)!E-wD^ie{!Axii0K(L25qzPQFr
zZhXd53)0h#)-uh)>`nNjOA@_QovAnVA?IcgwJA8}4BncMk<1+D))DU60N1=iv#Hkw
zeJA~IUIt#92z)y}0ZKRGL#~MzrwE>9YGA>){Z)LAzdBFzSM4^;{Wiikx%`!S%uoMz
zV7^2*^89Mk51xrS5WblAutzoe<CD;&PiDX-JDG8Z_uAu-r@Xs)ssnutQS>$pZ9*Rd
z_qk*vlbwE$VOBJ-9$t#Wd;IE;m-e}knb?6kJ$m+>&-7l^MGwQzQ!9d84QhVf+v8*I
zO<v%5%=V{z6n@`FgUQ5sD_>>{_-Yy&nIAsmP|nI-R+7nseiMRzlNY|}a)2BhIA#Ou
zuis`9Yc0zWc5!8y;Fmlw%HIs)>hQ@8nBv3V)gQ(ffj0A3yMb@U!Z!)<O@8=h5<GK*
zRdyIPuR|Hyzlfft#q?h<!OOBWLxp$XbJ<1zHhRiB{@r>LedP)FSHM3L#{UoB6oYS$
z!Z!=yn+IO_Z#>Xy&~M7XH@Zqs)a49Sg<E{j&;x!tL;F}iSYdqrhkmJ_tTL?je4cR<
zZ_VKh^@DHz_D}`FH%H-{h49S`_@)MY6Aj<YPR>wn_-3V7hL*xNCn_?7Aqz7#-!W_9
z5wkJjo6B36BL(01!8h~Zo6=p-Zr~dmeA64gdC(80Nla7rQ08LdHFm9sp7SYHT}rgo
zqg-uOdv6N;;VHVGo+5u+ie?l~QGCu6=G-#pyGXJMd`{BIYf0orB<bLYBrWfeq=Jr#
zco`Cuw+wyW&*Sy@OuTmNh*ub1nx4*ft*wq$Q_N0YI66)~yS%c~%bge2DPh<BM%*vE
z*1<LXSWVD$vh#JBKg3b{g>k0FF~>Dd#mB|z&FDDg{Sm7T!(-urShaZ`!!<8jyWpE&
z)Tf^Hh*rIXX!U#*rCjh$twmA#Ha$wIzEP@!mJ@a`T(J|wIlqRgx<{xwHVT#}@2y+B
zulhLqs`*cEEx1n|We0iw?5hvhdtFTQ*kqy>!d_Q^T6mpZHVuVuGA-v84X4z2{J)+y
zP2>Fj30>K~k89Imbm$(`q55#08^)aBQEn<d4o~M8Hy!VWAJNTCTiZ)F@MU&dWl|95
z#0FNAmX2WtPBeP|a%6IHMl9>8J#^ywHV&^1b&<m>nGJw`(DIZ!uUT{%`mTz$HS2Z<
zyc2koSB;~Vu$*h{4zoNCo0WE!ewPnsb%1kP)ghl2uSHCRn;LU|fB%FGH#D65ocWvh
z;8lfh#=$q2;G5p?O&9oP1$<Mi3-`k}RaT)XT>tB7Wo}XxGzgPbqd9k4(Qry$z&}8p
zuKXH!%ZTn%9RCmXGN)zu33j4g*lnuGWA*9;{Y7qY2WQbPjq&IdL^pvgO8DcOhIQ5l
zb4Crtmy?Dj!<xjp)s}h@Oth#2`VZRACRnE=tg{terzbV8Xc#65=9$aN%i1vvO^7q@
zgMnx`a7}%_-puz+Y)?)gKkG?6d1<gsPX4bbFEk-nW->JLQ0HoVO(_q(`v?EcZ+ye>
z*q1vt?LTW%=Dju@U5yuK7P{65Sgor~uM_ZnqH!H?#e;+Q<_6xICRxcberwhGn^t<&
z==a`?wuSd5md9dRcdKsTv#8nJsu9#Ae%FLwd7Nz2rfR>#lX07QnJ4kuY$v;aiMs|<
z$GE`b{*uRk82e%uyjk7L^ELV0H3QG;k(bP?MjLp(-J*?)EIKugnxx&L*y_w<C_%j{
zJ9R4bgtOEr%RO_`u(N1)8+k2`|KIHE3KiYtMP^^aUDTpR(#Mxz)*u(N+}O8`*VxOM
zon3F2=yj0}7^q8CM6-E?#|D4RvMXpbN4Y&=QYHMv_b$*^z0#yDu+6F}u=*F~UA}^8
zE|XzHpVZxTcxe{6s4RJVm%Ut=Cry9UUHX{NR(8QAn<HV7qDC#bU{Ix<1~r^+(EQ2V
zjx}h(CWHEyM~?}?mowg|j(Bz)&|?b1Hd_Ya!9kO0iU!moirSMq+6*hRnR--@|7khK
zkw*HOjXIHNRL?BbxH40(`fO0v>_+BI8kBO1dKhe!U?Lyz1lm#s<~0l@tB0>yeTUyC
z(4-pt-TDd3tu|`MANZ`5tA_S;C6B~a3A;_?)-u*;HSc9+aXrU}+|u1mHng&YiabYy
zs1;H}G;s#0*9g9eM#CA!dx<lQ(<icUIH!zkYE@reA2X+0^?4KCo9E2#Koh)3{j>iT
z^0MFCG@-1AR`7c5+lluD{GW8*L(6{ic_sP}y*+iXD?OCdjDNU#X#yDz&*7Wu=r^ya
ztBk@Q`=taKTAn`qtv))xoVgA6$gL}me&C7crjxItXQMOF(-Zs*A5H=01~~btHnSid
zIqN;yNe1sDKh5~ZUjbxUTn+NqpZ55C$HN=z=o>$S7v}}@8*<X)R0%(^fn4K2`l--x
zg44LZ!Z@vZ2529gGrV7b4zm3EG9R34Z{4owJ?;ND|NFchHiB)M`qMWhJUR^mbP4U|
zc`km&8-G=~09$SJ*R-+paJ66-Rzqf*lS5cz0oo0}>#BHo4e!es_+m2}m1QJzG|_Kn
z;3;-)g$940zEWya!Q^Qq2a{cqpWciwoRi3=SibOoe!>Ct*$374QaMqF$9vpnHhxO_
zMr-5OS=ojD({0RWf8mWTN{$%W9hq8_r_qkN!@YbIKkk2ioPX#UHy-p+fm&p7+~8-q
zFl%@>eQ#f>r7q<)KLfVU=By)2n(7m*Ga8<`g}#!79;b$^cC38QGGs*inEZmh|26q|
z@9?m^C+F^4hW_~ukK`h2uOyj#WyvY7-%i7vsWlnUW6)N5Y=arvrpdY)56Yh;tv{cn
zwa1x5jW$u5HD+#-n#@YlqM1pm$<p*B9hjP=c~g?KaZ-{x3`)||fl2iBCFxSXBrRjD
zWNl?7RbWPTMY8ZJk=4t}VIZ%!cQP5%$@+I(vTjabCezv!*}taf&yN)O{Ffq^l5Mri
zv8}SrOx2>5sk(ECn$e3?C1)nDCU2USPD@j*#q>d~PtyiG!aK91Yiy--^+`@AcY#@$
z_tN!_+vH>E8U)+Sn3Jv|6VvskD|y6C$R&Q4rZ9S`+!s)j>Y1kM@J$)`#th%|fp0Q@
zPL*%*ws^DJ${W5}IX8vewG^FzZ>r(T83y0{3*XEuoXi;~Nr&K@yYrZDIg-5Ko=JG?
znHA-gz^oo-D`e*Oe7qLmz1f4exLH&58FZS4FiTgMX3an7G+FHGpUbX|h4AB`>6~z}
zt1<o@2diC+D{>zU<o+j4ryj-W!m>C$o*$>qgX1)?Wt^_Qk5%`zv3k8Q7DkHIx6HA!
zevhF)K1M@d#%S>4XwK+S%8?MI&8?#732dbw7bA6ad8B?-57d+#f!g%gPc`1b(Cx`9
z|Ky`TcX+?k|JR9L9;bh)W#HMkoI$pVA3ekHO+WZ%5PWkNzFD@DOi}nI3clF@b37D2
zK=!wmbKMn=r?}$3Zk$)}IAoBUQAE0nSD|J${0hJ6J;kGt`JPGX$I$Kfne=9}No^)E
zFWHW+Z{S>vSK;C#S9-c!)qNECu4v}JBV9G+kC<y@){IcHV~4;qoVP!>GiyhNSp(UZ
z-q_8$(wNLxZUYvWRcpIhnGVtKamK6*@2ETc<0iN2aE;APKJZPUfw0aduEFIw?^Dls
zcif_mE#R4!?m8Rgu4MQo6uvnF-)v)l9XOC|+-3MeE|IsA3$Iam`l6huQ=#EBLRWut
z%Bl`2^fxc0HpgR;DGR<y_V0zAsh@3O=0G%$+cbJ!*O3w9OcqWt^pbMq%jAI_tE0K|
z7)P@FsbTG6^@c&(C8O`O<~cyuNs57K&|G$LpDWKxb*}MN__J?MdZ9Sa+WNvNtR<|4
zec&gUrUiOUA-JX>noS^lbSeoQ2)|7RY%>P7x#f+HW2UdGDSn??=sxA(pu*(#pxvzg
zV$-dsHl?H2r5xwuZPeVTO@&OM7BvXnrUP1D46g%Ue8tUdYF(ArKp|cO=xs}1Fl*ro
zY;%a7@O9K0{?_S+@%Z3%nh<N%4qs}3Ja+YA#1ZsOWuZ1T|GB$bUt!L|As*-T_{-<O
zne2hN&~Eytkm+CFU5C+b#uQ~HcqVt*KcWdfz}s=wqNVhKzn*51!$5L#BP?1|*`k{G
zZJPdvZ}^>?93Hvxeub-#;f0+K(}Xiqs;ryZFatbnsaYSUFc-U*SrOrIa3ix0WujJk
zLNYIt?8ES!tRt(qF8$!&O{#!S(*~~jy~(7xcyL?nMY}m*(p73WgYgu<pe7Yrntw;X
ziF;1J_Bj`|+sS-|l`gVR|KA+UO3j#knb$=#FB)Y-U$Ih`s#2DI?e7L<K4PH1#2~Nk
z09AVfdmlPZx<R$p87S=-@p_`!pwGM?PLI`KqdND6bzqYsaYmI3GRhrICVvfB369zP
z*q|<_4eGngpa2-j0?Q28NbTz^HLr^X_2&PnJqW#MszK@ZnrZSKC%oxSde_%Uu~nRy
z!S1BUl}>6K)J!3lo2lW)W-8XpS=o*n^u3@_+Y*fQw;Ac%bk*|)JYHkac#h&x`j1)a
zop~Q}rkTcb@S?Lt>6^(VLSOw_i03YXY@^}sdQ5g>b!t@W@LvsSNEQyhxolJEt%GmI
z{-CDW0M8c;JqVrn!#%Pi^YL1C!Yj{vBVZJ-=`-{Z;!(TqfcM6Rx40u2n~Oa)vI+S&
zXb&B;pf|+g5u~4>Dc+MGdB||VH!?fTM_HEmsNrq;stWtc$wKB|3VC?r$)cmLr|BJE
z?as~|?0SA0+>)N8!G7wvj(+iL%!Z^!b+?kgR$Ad2e8xKmlZCU0+k^hvde2|Ge*0^G
zDSSB8sGfPzGv1P3s@C*$!#G1apx3}T<IrgavCej)P6gLIXd9sJ)T53^Q-ebD*$LYu
zS^|`%Nq~H+!8*nGIn=)N$Y0}*q1!C?SFvIK%HxBNn4Y9N&Cx)p1J$oc{ig`MVEBqZ
z-yrjczp;2S*LC_P9t`tVYrMie;hVL-=r>o%6s=D7)@5{guJ@}?;b|Pi+|py7`tp%p
zQ+(R>VVr#a<YCNUuHGv8M$mD#*QPIHv$txa2OY<2yp|f3Bfa8d(tR|d8~x`n&Q-F8
z^4FveNblMRyv%($7n(M>sl@17>bR<=JXh9WK4&%Z8Ox}_3kTI4-B_l-wWHN|T-PKj
zbOYIMI}$bRP@<CGCh8>1_g$hcvtr-Vzs$<^fw|AzdU4-R{=MN{_TM{X;`ZU|dL=4z
zJUYYncva7B*Jk$75l`Z@?lJr4t~j;a9EX28R%OP->dmNFrLo?QWKU&v8xgA^Xb1H>
zvbUlibZ-|cuc%lpiHudF@K|Pq#*&K}tJ=Y_Y8Vi!Q~t54M{o0j#W9Ls6r=YGV|0M!
zxFm+z5-}>Zl6{%gel54#V>I~#Jwo4NH0BRILrY^-c5STI7mw4!l5uM15T}50Xc-mx
zUHN@WAFwO;6}$Exi&qRhG52yj*H&_6-o)$f2mHOi;{RWdQ>tI0ZUiN&UyDTM&oP_3
zQ=%G7X{|LgTkGyJ<{oTmt+$6-E9=SDT6c;$lPBm6KhjzS4)Xa<a`kewR>_@Y?eX}p
z>ztrtDG6Gclpwbt>QMm+D&Uu(WFNSO{F<^Q$gQDw>K4DxI{3zEbG$Y-w5u=-V|W_J
z`7chhKgB8ce{uSl&8|-PY(5mW>w&?pKr?@jg}PQPyNbd$EqGpXJma}I%HO*vPDS9G
zuJz!Zld;NJ5Uc%TWAQ=6YOgI;#WshiOsx>rDHEdGc|-KRf3T{33)HtK%%nXVsFnu<
z^=OB`Ok_BWH2LfJA99~&lG{za^xjA2kKBcA8sZttPW}2E^Xyi^H0+~S!#L}+I>)iU
zvNz`9+R~Jjm$iq_o1&!*Tg-Vn65npHMHg-O6)3Y;t4vls=iGL^O!`SqeYdW#Oly-)
z{ypFN!8cy)52^St;hNx4CcO<e=^DMMKDpT+_PHwaXs-W*T@?L|-1+a!ME>ETVZYcH
z-ZDEmzpL(-!uMI7JXIf)@+6tmeW;17d}c4Vl;)w~)PryO9bnEe^AKk|pie71wTK!p
zj+HYdb*k2*`TQ3%6ob$K7BiD*H|PAv79B?y`0x+4km_V~P`7Fh->mNnw+)~!vdCTM
zkJ01GK0K}@-e>sbu?sag>S!bQ{C70h=t+2H*_R)^#z%s;Brctd)-`wuTB2>_r{2>I
z4-vJsuK%HfeYeT!46_7biV=-?jNywO=sAUIQnPZ!Ck*GTkD(_EUFQcY0bS<+Z1R}r
zBBl-gWwe-eJV%|-b^PHSTW9nd)&@Rq)R7!w>RS)ce~P8RriplsWAWOwq~9u-pXGy}
z0UfDNGwNmaV5`b_nM-)62-?j>w3|C`Z8DP`n;Gvz-GkKIHsHUZHZ>O=?O`AMP^qvC
zf73h<a&H>r`$oIVpAWAT-kZyh(Xi078tz4#S;>qB_N;(`R{3V&t%xKG2Vb{uBdemy
zac0kB)%GvUj(X^>e*bdInSSSTYE)A=-@}xj+rvI_?uw{IJ*pggVIH&`Y8{^ROuf5I
z&4b$1L$n+3(H8Z>(^?>qKB@}nHPoca{iLtwr5pYs{N?AUOYL*hC~8v;igO<t-^=;*
zM@^<CHJlm?Tx>LPTbX_N6`6Lh&4#gfZaRw`4~};Q80SD;D*2pA2mZ3mGLu@tGvnZ#
z**i@dzku2l8e~lmlP=_k*Ll5`e#%UQGcL-%joFurs8h9Oc9bVMH#P9#oP$BoLjHqO
zVqua-#nDCH8*~a!vCABTUbKUA5)8aQ4Ae~x=u-wIHKoSX#fV00RAP6d&cHdB`M7>O
z+Ko4LDT7h7Y8iDeHy#_bmm#;QZyhse?-qkTt}v+T0t2&p402+9MjzUpjBdlvPS0l`
zGq9P)o^n!=sZLr{$4RY9I_YO&C*3LKq#<3MG-$Y!jE|kPvW2s1M;r8b7TzHE<`P~X
zj}p|h+PUh)G|J~_U0YsLbB?7>wGyw+Za4kg&7wu{&0Ku_<@59W;>9`ojJhJZI00Gk
zSLLH8p$<Lbyf*UueXX3t!}rCisg>!QinXcDaI{Y}#9O>BKUU%O&3mEKK)h<l>GgX>
z-(pc_(1}d*3{Sn7Ll#bTFI~c8lD9EA#?jQLMv-qt-{aTc-WpzyeVF~YIlh!&7+`8a
zUztsKLFlE*F~nEltMRQ|#BcnI9-yjzIuz)qke=k=;K4b38lL#zr_04*m}dU)n7^9!
zz-v6+UqRdb_2P=ZW>TZNT?h|x-2i#J2dGg<fIi1CO995oodWZ~GS^d?jmcWe=i9jd
zG#bw}G#l6803Gn9mc{pdcfx~Q3(hHn-#9O`I=@g0yN<tSx4-@y?XOOsIhT@)H;`VB
ziZI#)yf(wpOxhnK8+C?{&QO;!bRzGk7Ty?YbG!O_(_g}j;D^+l&}D*7d+Iu8+28xn
zbeGVBu$gSE6mrDYFst{jr^Y<>)cLYx;JA4)goK(E{jtIC=uN9Y9u9Mc2VC{m3-ZFY
zG$AiB77tD;*?}FoZR4Z=$jEX1<*l{6fBvLOVO48tS@#+`Fuj^yO)aZU70SpVQyJ}f
z=OC}g4w}yDx-yaTX-f_K6``BYBXoXGgz{4i&DuUft^{Zn9}!6=NF<p9k?J%pQV(ZD
z>h-Kh&7T*kyYnNpYeA$O7e*@6ib!R+Mru=~NbO)V$f}k~%HL8)zD20#rwBE97omR4
zRxHC@MfbMh8t5OcAir=G@D0~N?{IRL!}X7II5U>Wy=_DvPQ!3r&mXRrIm2c67N$y{
z!*uK;>s^@o;OTIE6Gl#dn6BBw)TBkQ<^~39a6mA9CC%k|l78b8^cx><t|dpCYx>6K
zI<~gCdIZs{<JnxMhRyY2SCFy}2*R@wB+ni}8qh6>>?LLbEN`yG2b=5TEqa{qhv{J6
z2=%HFp`F$U)$?ISWLShM--u9!yAgW*FhcfME%oeOOJ)BUshqzf6<)WM>Naepso|~2
z8HiEyzA@TAI!2i%#c0a>7-l-aU;AQo?plnFKZwz|M=`qpIY#H3kV9jP)ppldg_>ja
z-2%J87rQUS>-Kg0#JA$*2;ba<Z#Kg>eiiT$Q<w5Vr<r^czc6~tgV%97{gt^-@XO`=
zc2(ftBKaB=Ev2I4IG3?2e;zV$_?kK}=SG+_>xwu{pib2?Q=ES7iq&E~Gkq_%(8Z=L
zR48W)?fDjpe?F89wNT~09V|yO`2xxZE1w}q(eO<v_{N7iaUy)vS!6cAH<{66Gs8DW
z;G5<25mq|K+_dA&iNz290iE##b!w+TAKiledvqf+rn9%!QG>1A4!>TSxBhMGtyeQ_
zT6Y^wgY)mSWV2cfF)0+L84u57b~VXuFSYy;%oH5XoSTcxF79oVV>){I1y|*uzTYVZ
z-Nt6p9$&boBKaW?Ty<cJt5Sx!Xy_pH{E;r|H<^qw80XJ%`d!&Ge0|7%?e9v~HFJE(
zBpVP$A8sFJ63#Fw<|6wK?H(QAo0|L08graXlS^jVV4eJEIkjOOdp?Vnu@Am(!QL2c
zVJ<E8jlC94yl0WySG=jY+%>)$Su28P&~xVCiCi+AES7nkH`g%-0FOw~LeyR0mj*?d
ze~iYyF+W)@+0bc9@O|m{V+Xmb8nu<)Q#rd(&&oF(p9yEumhbSKcvIt{r`nNoX<qc4
zS@>KwqnB)gZJK0-ZMYp?hU<3&`ng@0&0xf<W1`2&MqUki+PV-A{2H(e?2`{o#|-QA
zN7uOm$N0iLPq<yn=PS7X8vibM=yogi1vqs_5dI!-`lKxM*`p(!;Qu&Qjhq|@JjVRL
zdC+dY^1C#APTlDS`rIjg*WJ{fR^hp!HuZ<vR7!U=D0~rBLSb7sd{_0TQ##nxD3?te
zKe0#Mw=$E2n$!-flJMR{Pk?jq-c&)~`W?y)Mf$>5u%8Wu3ET0QB~zQ~a*H|P$C<ad
z8P5$qr_eEYaC*3FZyR@g4P{<K4SK0cx@%Al;F;&>Df{Eha~9RxWYKJTrJM#>WVBm!
zuo^jj&zbvu+)YQ(Yn)EIY4BERktf{LcQStSxBr`;lsu0t;%R1P37WOFlUe_Up!H2>
zmLwjN3FV~Glc|x~$s|S-T!C)$^N&fj$?fZUkeT1B;hX6uEuTa^YL-b^&}^EurG^t}
zV)hiT;oh#QP7NpchKoAycTr8)=1G6%LPe8J<Km*!=SEdNXw=pP^g9i~W8B23!PKVw
z;G1G=4Elt2Gs4rL(xnYr_s3ZQ?gpjJ;A8y8-SE&nM6bzcOTCJk)J$}g9XavUyuq{6
zz^HX)@EQLwsKs;YPnQk)cfUbb*BVrAJ~|HduCk-~IYSL{?rG4lAUMYbE$4%?hU9Ff
zZ*!egznPOlYdOgV->fa*q@ZF>>gwX8Ekm4?9Mw!mD)ajsal(h;q$6jYH2s2;u3UCf
z26|Lu`q1l-rDuyiu{Beuzb#;97xm%seJsk!^W=PiJ~rNCqn=unkLRwy9`cBI9;cz{
zJcV^<w6JQ{0;@Wow`%uC>bvF0l7ykd;OQ}Vzg&2qeCBnU%@e%=J+jb2ykie}9p}Y&
z<4itgG8yNyJ(&f>{*8uXY2dAM;q){QrZ%&KJ^3RZl1g0TLa5zveY?4bOc?4o4GgeM
zC|MUh$TVI+ulNZ(v3PJEIQZ$4%};sgvFLyYr|kjeb3gS{EoxLH>iTPwzrXgT`|HUV
zf8}3G-Rcb4I4}L>lZP3aHRzi%!6|4lo6vBw!!YOL@E22m`r8-vo!d{`H?viMUWVeu
z;p>)q1jyz_KCv_29r);Nc{H8EWc_95>#0pWJ?pP)v;CD^){l%wdcPM^OY>p|!wVlZ
zJ;F@lb<B;TX4H~i_=;u8rs3lY-|*Zl^;WxJuHU7}E45R*d*Z1EC72<z5$y*3_B(p-
zM)dkO*ZJ6yd3dAoGokP7&*!CnPF`|s<E8bB&~u*PIeyM9b*iO}yp;uQ%uty=@h~5q
zh8Z2uiYmc2Y5!3VoXx!XDz)UWw5EPr%c@g>vT|NkMhVXy)Nz}Ga<6ev@JR<%KJOs!
zyAF6)TdHH8mdaKrLT^|NN=9f}c$kVghiOxx7U(kMm*fl8wY;G^%398<$GZDZsJ`S0
zRb{eiw?7HtP#wbG6|9i8!Rm9Pxk}e<u6AXct3oOAXvs7=@*_z9UJBB@(?Rn45y<&D
zP=UShn`Q*6W_qA{l?WuGGEjwo1Sr>A{NJwvH196Y(JeT1B0c<gzd!p0s9PO+{^8F)
z_@chSvvY2fdws=IR|?v+hE-!IbN0CAIj|o0Ci|-=`$QNsXsp!djp)~$>mo99EyR;r
zGcUeDUhk%CHk~cO9KRx7+BeEe74eYeoypwh9rW%W^^&<+fC^FLTvanrks|_`iNf!G
zBvAJAfvWR1P|3do^<+&D*?*yGcra8KE484%FigJm6HOcxj<+pbjaNpf?ZF5o-;7X9
zj!4z67^&YbkviEjQiHois>IYtS+_*0-=#=qGDgZ+q?I-{YK51fm6mmArCt+SDY8iv
zv!&?Oejg_{_+<v#&3ZJLZI|OT@jkrrB2M`~$Eo*kW++gv+RfLvp`*P%5vN7bajNAV
zr)c*$y{a0gIcQZ!3&g2jt~gn<;l;sw<99k%6^n+^cM_&~MOtV~juv_k-*km<3c)uA
zZUyV&0KCHRO?6|Cx}IWY!dPZ@_Y2h04uKjT5~!y|0(IjmGu}=I==W6m6wycfQ7?9(
zrkpXCnYz@Z&s1V=$R!^Qp@!RYun+Y)&OUDFp1FLqKC6#@erG1zT>51a@G}fB%YBVh
zI1Ft%2e;qQR6dxL>x#2xPjJ@AvCb;d-dS;}&f0@cz89@MFMb05Q7$_8fw_a8CV3S%
zX$3j)_NA^mOb_i;_5=qw=k){^-CB)j;fM<Xv3y<@Z-CKNMW|a1W-n=j-(eH`%jkh7
zJ;VzToQ3^~eJbZZd=*E{YP`*?yL-(lj;|urM!tR!EVdZ0&NB3!05?1+XaV%GuEQg6
zj=j?%H~wNc<$>8<M}o+dPbF_|h`U<NMps`$kIyM)A>DLW!LQurwW{1pW&^y3doGgu
z!S`-%f;ScqPno>b>RNF791Vik`^MGm_nYzBX19^?jYqo|wI1p{&0&bquhBuM|84t7
zW@}FRov2@3sm*m9edP%J@w_gx6{sWqfpg}2pwIimB<L`+!s#c2UG7EWok5?8f@Q3%
zU_KtreV@^L#`87NFi{NW-tk`4q0nuH8t^bTqSjRtt|^ZeR}`+wP3|#gWYasC<$+D{
zmu%{L#HMUpZTh{$rq2AX9S7T#1wT|>{6v2O=-qXJfok&g#hJ;Fm0o_lH?~{k-W<n+
z&EGm2?@hbWRxywke+xapfmX%v{?1Vy{i`q<&L8|uFWhzgGMSDC@YnG8y3cS|Rvzyi
zXg5*R3b%MNhYH_KwW9R&XQqGrBRx7dd8|)S^H|Nismb&tqY+FGA@|0?q5%)xIQ!sH
zUqKJuAo}!t+z26nb#9vVh~A%HGsx(JZ3>JvYhr)1rsCf^yP6zgw8uHsm?u?II{k^)
zFK8GK>o~lo&-;K$SN5YjuHZEW*StqB{5064>dj0l!fSU6jJ?F!q=^$;)tP?lGWc&w
zU1#3+UgmtybHR^|eiP=RPdQvP>J~lRo9JhnY?LQ;r7ks%%9`J(LeC8Pl3~z9XM@^R
zGRW^AgC0F~)~plG`aIuRp=c)E!;R|4`M?9sCXnyhf5o7|_YCr;7Pa7!LCf*hw0>n!
zE_^dRP8*bSC;tW)t)F6$<1m97z)qR@KeojfbkvcWR|V=*yhq>t=cK~$&5ia>@~PpZ
zhDDs@_>YrnWOvfWJWe|Hq^X*oYOH%^M>W6DSPu?0mho0&t$o&5A?}WvlJ2OWBaV92
zvx%l<YbqBvXYKnyzt<7!f7hr3qLFqxgQj+!e4>9XT8o}jgr2wf6L`SUahx;JbBCt)
zfY*!rS+bDc;MFgU-zx;JrWP85_bWflU9K>k&+B&~{Jv^C*$M2)<Fb)sOulBlL{H72
zK2;w7$jExkoMS%^2%_(&H#v42$R(#vGb=BB%r<nJN$3nKm{mYs{wKL~D{bWJ(&Mvl
z3O&vHeAN*TPDAEYR&nvu!`6OyL&yc*>ZiK5_?Q~ix{A!8viQq2-e30z(jQKZ>g6GH
zoCnOi%o3pXRq)>64fgWp8ry=s62{quMzfh!l#f?&yOvti$L3_>_|O9m*PJ!*F}lqV
zJU6}g{)W`0n&u`m@DJzGJ815WnWyRNr}jnt)So`~>iBh?UEr|m<Q8u7(HzdANt`?9
z?WZ<35YLM<UYohhm`S1UHIui}@X!eM8;x%BDwX<lX|5xjKP)$xYs-81H~Q?)s^osI
zpm%}3w3=yTrS9-l@xsjVqj&sXe=n83fbaMRGjGs>PU3A!z>l+@-thuU$k5p7t#NP3
z3EM8Y*RQ29`)g{(qOw}>Um3j&DWeKG%IG~T^JtZWIxKWh-pvlWdD=loZab*?Kg<fU
zhAE#<m^y~CiA03yKSP-I6>p)vokR7xP^eB`AQN#Uv$~R*V}Wn`Icp7^SPe$pkken5
zdGIMd@ew?asj%zG<<uo$y%%U$oeTKrNfz=4S?-zfKBH+py-RjLbLK9i^VO-#e8M{9
zL9!+kU@mJO7;dbW>azO2CWGL)r|v%URNsf_UU%^|c)Ag5Mdnj8HyO)N7wAhr%Qm#n
z{!WVP?WE4VoD{~2?+O3+a8k{7PCDMfNtt7ubiS&Se*J2yR`#YkKHpJ24?5}99ef#k
zIm1!2fAouU^;_!m5AY(6wWu{`x5Evnv!e?~9&uH>co$75;-aVMruE73oL|jF+3LDz
zn-6)Q`Ak~OUhywIa4j19Xsi|P{fB($vcAes*5Z2fzTUlj^?MZgNa!19d98PT;Hy&f
z1D?Ux^#BdA+v-3qJ|3vpOM&YDI*^{o=6Y5rST7ucRTIB%n--z+2nmy!T3x_yw4FL}
zG7pNA%iuWK@5HGZbt?16IE7~=+Yg;)Dr}RtOB{9fINc17lanX)D@&ZpSBp~@_+}4$
zvl+gzXN^;~FR>Z}-_(b1mcutK;hR|4=F7jK`nV@l%^HR%g*w%u>LKdzr8zUTgOvW2
zc`8@@be$f;EA%;5-{z;n8~n(B^OKQGsFzdx)bAi!p1ij=jP#~Q&|4cDaQ>kN8(z{&
zI}3QJN?z)(Kbftxf<A>so?LI4cTM)h+Cp4c$Oh{VYZzgVR}aV)yG<r4XV+F-+h;X1
zsL57mjau!jYr~xt(au>ZZJo6f#wj7A4h3<oo`-Lsy$czqt}4ICRXv!mynuey?7i@1
zqT$?uaYik5LAPT5&Ld{&lyOyQ_$J8DRf%!*wr+4$23(YOpB`8Aoax!{`Q&4+b5-)#
z;F{kXI7jB9N5<Vuy_nn%{1)xm-)g`&EAU{<$B%J@p7N28ENb`zFC;aPUhqv>3%UQz
z(QIIqH+}KEP9t+=Eqz@_nSXrIU77DwKS6(}$nDN&<j@_q$mf<tCD?;E@5kes)m@fw
zce$>h2dfnx<W6`{mf_oh0V>udM+SdE+5q&M{Wh(9M2^fBn~K0V7k}_ET1q>7Ijdou
zeFe~A&~vu&c~C?8^3h%LbB#B1j*UdmX@e${0l)NW0Ry4YoMM$@O$#Nz1_sI%NVcFa
zH7mGiEa$aSu+2r-W=}1CZbg`;B>htP=%LC&-3sky8T*LyEoOsL6G*0arQ~{>3|#xJ
zQJebJ7r#w9^Cnu^^witN`Hp_MDtL47-c0*LPy9>r7_ZVRh4<zp-kYtn@K)fxG4i+H
zim|G#4|AuQ&?jDzIu>l`MgPn`YEv%MrgrS){k@d>(*z#B-qfg);GAH0T{X~Gg?3}6
zzR?2@NA2epmAy(n?ol4&<rdi8qN!=<0?z1gc#2bRP<Pznra{chjG+HJrl^}{EjMd6
znKnNLo0WT*S@&2?hmuK*?szJWnVECw@opxSgJ<l8q%sdovejdL<`4A9W8~dY`}~eq
zY}`1LZo@UJJDRl5(WK(FnPcE)(g3u|nr(Q^;ujlIo@)?%^Y|LC>D?}RG>4g%UD0i7
zxu_9YG=t6?>3cM4l#@}X@1xhiFFp1->-R=yO<v@zJJXzbk2@>tdG7nynY>G9?U?Ay
zya#7yUpcF7d1sw0<gDohot5Pr%yZVDT{o#cT_rE)lR+<Gn^^ocl@A*<3%^Y!benvA
z4cd%m(}bE=gHQv^Y+!b(K{K-%m>uP;#lM}TMNT>k-=tBWYE#HbNw7`-i%m89OjCK;
z8moSCWBog@u?FLp2sz(a^KLd)t#6HWGLNGc)pk^mUXJ8)I_kvCCd%8asb=whFZdFV
z4*KcO7kG}*YYHFYIm8cMdxAxVxfTswMPL6N_@)>+M?c)v7hl`emAroLqIDKy&-KCI
zp2X{s*WLs)oXu#8rHfFLLW8`%hPfkW@uYo%@2hxfezd1rkb#pU6FI!(7`M1WO{Ez#
zl#;!bv55R#_OqqzZ$%w^<UP?x6PNjD^iAe&ex+X3gnAD-bnzpZBY@wv`yH<P<m5#+
zBJZmedBAvZZmjZC+<E%2Kly1SHLAT$$t@0JZg)3&s%El3ppT!ShU7{uX$o3Sk(=l|
z_>KQ#&)83$YGe@e!Z|~JV69-iN3*$yN9T>mxp53oL_M?tbel?OKc@U>2GpSvr@%Gj
zD36;!{}%n<gJ7Gh@5u(<MJ<iYyiyJD7@*x;!B@0#xVPeS<9Au>CC@I@l+ds??xtpk
zkGRKe`Yh*izIj65V<l=*oV`}JMwk9Tk7zf1f|-E(5!C;FlKC(QU*B``w$O6U^~ST!
zwSMY7ve;{QDK~q;?NaoO{~=p(IC<>fsIyb6e#d)y4m#Yl={%RJrR03I$do9n&(F)K
z*Vr=J?pQ`&vcNfW9rU9U8jin%B1$=EUk(S2-|3+KX6i8HiiOaJc8SN(iN~(*Eq^?`
z^p%nSbJI0IcgR}yD&{T6m+;#&FL`r2^98q^shd?pJK4%LWEHc`_EY~nM1R6zwBRFX
zz)j%<mN)0l1V^&Z=~XbXGS$a(P><_&5wZ&2qUlrfc7M%T>9|GTxVF5Y_Fb0MZa4Ms
zT^4npiB8-F4?`!iLfhaa3$myfJqz9I&_e^?wBGKdnX8>N4z6)yHDC$8*#zJGm*K=*
zNp#RuC(aa3WdAnR#c55+MQ*Ak(;XES=%h32opkuBGYsXdL1?PQ$`}+{&M2S7MlINj
zNAEhd=#NId$c;w>zL`_bMcXU6C<DG(&*#Rf^fWlT;8*0F)!wB2)KGHw<{Z|~q<S;(
z4_@XP_l384jGO-JP7gEJM=RIKx>sSRhV+lH2U)1kIFfz)XMwlMb0&y6NcQb@7&#B#
zd>?#Oycd3q36<q%s5-9-SKca-%)yS-`_Zv#2DjKY(gXD%PK&?c#le4*ACGY**7nYE
z>I>hLi=Y<8*L`w}(^~YKnq}fNs&E{c+i^;VZ!-Ui)yo@ri}%MWr6Rghj~1%9mp-UM
zA^%6xS;s}0y=~m?Rt5$GDFY-XnE{526cAXkySuyB?(SGy*Y34zZLzyuySv+UeZRiH
zKh9?W5fPr}JlDDER^YR$y5d<?wTo8OjC@t~a99;MpLA7vlBPp+G|b(TrokK1lw6vb
zaDS$%IcJh@V^a0`W{SR@OHl%R=f<lE`UCBu-#xn64l(z63C|*)_qVgesmsq;<=`5K
zy&0?NoV|AZ5v%t1=x-l@AEuRy*)%RKsPEF9?M@jsvmdixZ(wik#@lr^asC{`e%u9o
z)5gi(60TL-I78wY_?2OO-G=8ucIFJ01t*n3cWUm^&92M{>>era5s~!RM(PuI>By5v
zb@OzokrQ1i#ib8TUHV2hS==0#2HbY3;B%faU+BLt7^Uc(_$GKp6(7hMJtuku_-5$u
z(R8syD{&Uy3zp|>dQR5B<LrdfsUE93ADPXFM<4<JO)&VT3eVT)_&?mb(1Q%GaCI@y
z+zs$Er^!=Y2fx5SEV#@}q&xI2Z~WiA(ey>E#%;jo3J0GRZN>JTdDZA#Gx5|8TTCB4
zdw0jncy+SD;dElZh99}tmTWb8mHAA9o_)aYk_Q}u*PuHXqY`+>Cm%i<BYV0pSu5~)
zc3Va5j=*;V&N+x~Re}9}G8pD3c&C3F9fGXmyuXx>uK?FvLXQfD``H8^)Cz2q1a7L$
z*P6{4u0Ua)7dgmZv9}$1jWz|5@$wS<%n>>jcO>8kq@Q>;-r_L{n%#$wwPx;AO**3E
zdFGhe+q@E#1Mkf>yf>X6!MWnS$+8#U(P}tZ{1glE-bCWP*<A;}F@Eemq4C;VJYL^(
z;LT=#ZG*l!A8qQ?CUlDVoLff2pLIvCYRLO=H*xGkf3SCs<acwQXW@!Rv3hix4kPxQ
z(aZ1|lLz-|#<S8tR(czw3;XFQo*$!EHQ9T-m~9PTQ<MJ4ihan^@p_ZJs3m-37p{wC
zy~)~LkyfWj;{%}w!8TVPM`__%w5H`zdjANY82RWy;2Phmc*oc;26m6q?w>AIeB;uW
zt6cYEU2uH(;j4hNV_mvc(j_18O$+eNaxi(wT=0GpJY^rWprPQAarj-v;-kScobTfQ
z?vLl(C&!lJZCT`0i^)#C=<3wSCQgO1`p<D{^iZe#8u9AjY~aBv04B-n)H7Z;9S+wB
za#w$XbGm|UJis<hkA<t|?r>FJ6|SlLdj-0L!^ws#JQQBYGhDB}J2mPD-0D837BzLM
z>Q0C5tP9gc&X_~2VOsIQt_J_vweyf&Q_?NUh_WcUl|>HlP0Uz}8l1Ig5cp>LSBsv3
zZw3WenSE;2fc;kOoMO|N6E+nN=4-G==W8CJnFFF!B$eJFeEpHbd1jzl&4Fuc$nWgp
zOzyi(=#ks_ed9IWlLps$6_0wB1WmV-;X?<#(K$h#=W!Nj&-1q+D;!?2B)<4f<o*5u
z<C6pWSTj*qrYCC0mqblw4po^GNz8^#QU`0YHaFvZ&OY4|U&!vFm9#&(lF|mDsnGLR
z@O354@@IBHH89S=6n&+i<@bvz+J%O{vRtZiSEUOF56&zyN}tGy*LX&59F3}RAf96M
z{`&3HG>06O+om*~KFK`qt!WyH4|wgFG>u^I?fVi=93JOWBHRrc)MwTc);|e!0rKoF
zW2gV74D%UE(_`$9Hsza2Rt*j!FjYD8G53@2vA#FHfTGNh31{AA5oYNXhikZkzZVa2
zb9kF)>ymZ4U9yJb0bYVvW(xkJVJ5nXH!*+sH!vSu*Y9<iyTZL(Jpw-=&+{+n@B27c
zgeG$K!1GiKp3DyZS>7ED{48gq3e23r(`22JsBI_l>SRySR?kG8Sj}u-__KMVV^!MV
zRI_?v${ilUe4AkM$-&ywrlgj7lvKMqCDeUZapq2=K{<;vJGD5zeWQLmV^q@zMkQn}
zE@dyS8Xm<pHj+-BOZ*Hwz&rQG|Je`Dc?cZ<!`R=(k>OnrUzeRs%WZH8>lrwv5BtPo
z_T*9Xz{Lycgk279nMQvH|G$+E-ez40gJ%cZtOeTyvKZ*7?qC}q)?KiTC+kxYy3zT0
z%;4D;2)-!}zPSdzIRv(O#F;f7d=txF^9T5*Civz)_@*`ZCQ}u%x8R!~@XciQlvwzi
zIIv7%Fiqy(VS3+#`=~n}wr*j%3BDN&zBvuPnE}2D0pAP+-~3h6&biQvhr_DlwXDkG
zv?@Bts!evQMkHI60WW&{hgG}F*>&KIUEzG~PJRxVPdn72ms2-O;f3QIlcy!K=w8ut
z)iF|YKS%1#V3&gO;R&1(r8IA8;Y2ACeADkUy$%gx6xD|w_sMww(eEcG!l$+)n=y$@
z2IuY?mGDu**Ox~}7{;0TLwb^`lSj;wN^exHRNd)aT@#L0R{;3t@y0sLiLR?Z1~kx@
zMU7N`OJg;L!^sc6xd(r<Hb)cH%t96%&+)7KjWx>LKvDhcYkW|B#g?kC#1{3my+%Ei
zs8~<iKGanzT3N<G_`7;Fb>3D}Nrh@Ey-sz_sZ?DL3s%uY=3dUoQAH7~(Yeyqg0n=-
z*)$b=jL+ghMU}tJb$-2~yg2u~J3)@-ct!agC3lR+Eb<R>;T1TSa~>~R8V~4tvMO8$
z5fhjr122+Z8y#t5q>inM)I>6+hnqXKuQF%NH1JFc*Ljdbf0l9RX_SMWK8LQeXU}Cd
z2jBc$?bLbBa|_X^e8D%pz&AC)H-9<lnP;!R0_Rpc6P;o&BK4|}OShB2EX~oS_<h}k
zzuCKx*@J(>2jRo%>gQ5Im`fwmT<XSbKyUIJ^@qA-9>pBwXLOS>>nA_np2;7h^bwuP
z)i+x2bMP#Sjn;&ocyYitCBQa8Jooa0Z+?Jp%7Je>vrI?mcRPk}=QDVS`{NJrjRAZ!
zq<x%L&tgU(_mMC6){+zCsxHDYd|{R~Z*RKF+^hw$ngGU`aSboXQt-_bvdf%V%cFaZ
zNsCj#BXJ7geEKKP`|>NviooN{sGOi$fe8x0N9j44c?XLUw0#0Iy2mD{*b?}X!}QKP
z=4|yZGrGY%YYcd6N;CVSJpDAGWX$d4ufRFQg%=#GGcAr@n|ORYta7{^&c`}~Yfi!0
zRONFs!~fibw|NM*DGnzzl(TS2zV7RvaMR$bWVoAKa5wwGS>N~5>$eFE!}sxXT7n)8
zBg4`?L2=FCY^soT7k(S=2fyOX5XE~_=3~4L;=S=lU#PkhU-2@!fTl2i3Jp3I@695-
zH}^%KQDD5h40Q0LKbL%gzw`=R*a7l1YnZz*ll^uGJt_RWH`hSli;B}fe!S+!lfxb~
z?<#u@`_90fv1$bmP?(?d0QQ?NDY4qa^Kt;%)Xy!<0H4SVaCE7i?s$s9F(rENyKl{0
zNM1*C4IJi`sEK-&B>Qb(<)+)_Eqdw=a((B~N)JZS!4@TR13dEcqco=*o-zwF7x1&E
zUv{Z7eBs|0>H9t6l4YbzufX?re{(51+NI|Pmx8;8qYdMsF`*l!@{ET!8QeKsMS768
z9v-gq6(V$PYPcqycPfGVV-NSqrzK8xM2C99IiM*VXGFSF@1vc{HQK31J)JsN)v4tn
zPF*YFl<k{Cy<R$W`H4dr*`2z(EL>07=k^~7*Mg(uoA!ol+LmyQSsbo}VQ@Rm$zg$Q
z&Ktv3<+)Qy@IygxMeF)FwII={8`B(WXmqGPd-D0PFwJ>vhpV)64cj$ol1T%?EecPz
zs7+0aMmD#Q<F%;R1+dM37L_Ps)#$QTwd`os{X<r8iA~u~+tehlT@Mn%bnhbk5@*b3
zgW$Zn&<nAF4$lJ2KXqfadnq#g39+jEA6(p)I8DRXR=pycFW1fvxWK2r37U|=Jn^D<
zz&ICFc?VDMBA&m?wK@pS5KbqjA3prgc;n6ZM$l9;U!v!)JlykS<|myY<D40<Ai1Ay
zsmXfXkGU(il4SuCMD?$vR5H}v?lT|T9S$E&{%*GvjYa1ee;llXhqsUyJ;wC>{e=f-
z^~_Yg+Rg0EJE`){oF-4siFL_pq&H5}l2K{OyDE)5a+><U^AF@anIFH*ocqk-#FMl1
z3m>;In<|pb6uz7k)-u*L$8UNJt~rzyzYck<=(h9;UPXICKbr>MRPQI*?Bn2y#mq{W
z#B8V*%t-G}#<feL;^8x%-bm81v;3dt(6wGWS*LC%=@$R*(`3Mpg7aoAPgGTStqRLH
zFTv*)8p5o&W@x?eXy@KB+awwt{w=x1X=um|In$iOJJOy!Lk9g5p^38a^Yp(94?rg5
zbCOB7?V;*fGgLYDhG<m<Pd)x@BuwI|Ep<GVRM%6B6FoJ0mQlk-7`1AsQ8B}f8Z^$R
z2GfknvD>J(hmFi`G^)uzM*aNmsUBI1>t10z#_&Re`FC5x2mR#ty!S)Aj(nyg3(w;4
z%w!$1!7YJnK5`cNunL?qihj@O;G8*}g}^s0*q5Kc{l1wBM+3fD#@-#tIsv}<39jkG
zD$WW3-&79e&%iZVE6`_Fo--`ozK*wfwzIc<0o!C^-QORp<8bhvdzsh!JG?x5%vko8
zv|MNvbC?Y`I8tMpMXF*Aw1r!6G)KZzFdI1UyImDN*>(G#T_HE^YBJkScZXenj<9Rr
zFz%&6cAXz!*ZjV`-_@?a!8cpMFULk%6#%{|Sk0=cWvtA(L+?!m_w=^Pa@C>&eXMFX
z5?*w#T~Rq4s^a5-C3DDss6)x`;dP6I>neNA@hahZG$&jaINSWvs9SW4RG(}7irm=$
z@SjdD7^9PL+Wo;de}Zp3FVNw_{dX9AW1EQol;{6^c=-_{@tyG8Sqa}?l?+|a!Qh)T
z=BL(7)%@YrRp@DTg&wCbcyk?X$Fq~aR|AcmPS$LCWBoYWSe;o7ZZy)hhvdq3k|&$o
zP+98W6A!JcXVvTQwd&~7i`r^;s5Uc?YsucPhW2_^*O>{`RBuc*EgDfx1qQK_t7*pG
zbhX)&t{1!0^^tj(uLj`9j!RcY4*H?MTm7Gr3%-LF18ft_S_ZzcUM52fz8Q0zb354h
z&<wO}3(wN>oYTnz`R$<Jt`Tzwx4D=n?^I2&qV|O8F1U2P0d4-dU0zSo<XQQj*)^N@
z11rI4mvg8V__7WdrvMmd0Q+{?HF$59!mD?5su-B)QtSV5IR9?tJR2CHoI@itIyF+O
zSJD-7J5pZH$(bZX>RR1Mooof38HPrMhv8qin~z|iOWrOzLS5QR4_P_zje9?rt_^W%
zF3*{*<9W_7>!&qZ<$v#^G>_+3M{vy^e#cvGgI!~yb%ir~w_n`NTy%<s%+uK#t-U*<
zRRnzVeHYgNo`H22qV?=iwBB;YnFEjTv<*0iXYXF_rHkB0ZrjmFIr~g|MV=fSc;g%#
z*I4Es&83^3W$qiR=Sj>5=pL(mJcsW$;Jg4|TDu@KnBa0F@uY0u9#3Bedx|%_NfMk{
zCveSR_>;bLnt>&ba$Ys<N>C#>wI(m|BH)d=oSXAaF}%Z^U7Mnn|MKV<!7MKLoZsMP
zYJ+21fp><m{^IR#e5@(B<`taHf?)iF;G%wDn^j<&9B%Z8<U?!E%1nr_<hovwk-~Sg
z_A)avkHIJHMk`yJ@c(^^qb7oN2J$`dJyu{JJe``L+!6E{mt(%O0nRTgJtcUjy>G^A
z=rOvaxBSm5T6cWBp5eXutvOzt%4EmG;dA_%7hQ<HpzoZy9>r<*S+t2=amvEawa+BZ
zWPRCxn)5!Ga4q}M56?KIW#VU#FSX)@ST$nr$qlZt^YgymGgjm4@|=uf7C>J3W$xGO
zOX<k&&$a7hUO<)@P3aDQSB~HL7imd`bZS050u9JBgM9-yLe9@6E&C6D%>(k@@HX|f
zMyVNk)I4;kQHfxll5jVT>ALZa(%3S1Z5Fuj`s0^B;ZpoA7d_Z61(E|#Fh|P&AGiu|
z%%D2(DZ$~K7sA!r3C>6lC&L}CII`X~dy<p#h)`Z~NI8!<H4y!F!b+!FgKZkjgu@|&
zeWoya?k$I=`Z?tSe~d_VDmV#!*Wan9d7R2YR(r}LhhCBE?seOtdzs*L;Ar~I2v@gF
z{QKNnGr&@1@!HfM1g@!ycey~g{(A$j3TM;>O!aQOQ!`pP6>UKuW1q~5$2e(kn3_k2
z>HZt`>8f^h@UqJ|*MuL?q(zzGac-OCXl9XnZ;KY=!?}$CdwF524*OeW=xJ5ZKC3Q$
zvMTJXP1YQC<x98g-ch)CEBh?Cu-Y~H0_V^>bt6W7@cH{2V)d{STGbEu|NZ#KbJKSX
z25vkT{gN}l0C3Hm+{~(9%QNo;Tr#@nVKDT(!U?)W7IDFH`n8Y1JHT_=j<dK&TJ1@c
z4;spBJSg$>?{<F)MldAPb4O=OZ*b1SWG%#(Fs&!t$C64sKPssM9)*e#DZ1Y}MG@%q
zTj`w}_?*sNH~cqIsVdVdRYl>uQ#Qe;p!I+Ml&WlQ%%Y-yCcZlJ5e6}zVIdhOd^nHS
zrl}u!s;)=V(8PG9bGQ!YOhb7vjR_uy-1jh6e|$E@_}D8C@>BfXT;Ji;!Bg|_1U)5R
zJ#`13X!w}8znB#|3?AxRqTc3Aq%WIJ!z}P9r4rQ^pT+gBNh*pbW}JVrUR`0P0-lk>
z^z-%jkf;p&NVV%HD(7_Ot)!q!b%|H+>^#TO2QTA&%3A<m;a>KxR`I&ZJw0<Y-B$1L
zbyo%tqf4#<pO(&7UKIzGm%m$}_Phwxf8ZIzWuxA-GOEZ;549_1WHy#jW6K(qFUhDi
z4UMYS)2OWdjOsersPdzYikf89*`-EZ*k{zF<3_r!jPMMeTKdyd$ytl5;R^a%Ke5-<
zXYb(oGwv(*`*Y5bc=KFn|10mr={5Lf3;5;^_@*rQX5~n9Cp?jv*^>vderLU9k9M+$
zWPoj~tR8ST@4+?yfmuA)<8xn&rnDOFcP0P#v(f5gV3sbPxwwfg`ZDP`;2BalK2o{j
z@Oj3Nak0?N*`KVUADvN&oEuyYjR*=;1)nhe4xTxbIZTh>IhXun*VC(ZWxdSv@Vs4V
z8E`GT?Q}WXH7fx;^9OjpvrQE{*vPq9$h%qe^`b@9uUNG9nnn5WvJI`t{R6)F+uN%D
zBCRThug%qt&yBHYhs~tLQK6dP4At5dp$a$<s+lK46?Z>WeV&?B;GIe3znK)B)vT)d
z&3bFFXn1Lh{+MEcFSe>oPjo3X`OV;);b}42gg!L|ed^y!blqly_x2<Y%ULxbp5H(F
zyVp4U2|N$H@Fn$$;eP2u1~0fey=(ZEch%IRv|8Fgrt8hBI{GxEt`;q<uM<TYshVda
zeI8X`Bl^|Tysq`sqES6{=u=lc7uC_cE9BL-)KV3%T6!N)QwJZ_&^~YYOtRZvdy}+d
zE;#B}XEtjCe9_&u^h^dMY2Z|R4_p^F>+*d0r2(7-+oC(J+ru@ol|HQ+aq0yR@hN}|
z5NF{%-dr0c;DfVsCeDV=f={IC7iKMIi_w4NMiw7%>1;ulrVgMl^(Vi(1$K2lX4i>J
z>~;Je?YHb|dKZ7xeY;*hu<IJ$!!F>Pvg9tl4?$Z5-=u<X`p$Hy9r&if67)p)^%4!(
z$6Gq_2BJ^pqgRacWoGiM<H>|<%Z2s?kDrq~?W#NQ`H}b)szz#K<Nx`L^MaXIkqsY#
zhE)}P>T)PJrWV+yI~rAAmzKiE72?`_JdSR%No26Hpd)xh>BXZcX1hvh^YHWEpvUCC
zbT$Ic2F#PEA{i^Z#<iM9Yb=<@4Ua|{Fiw#}(K>oAS}z_(>-G0&b>KcpZxyHV1JJ7A
zLu1gWhHl|(z|XnmM9u_D=_%u9U8h5=YNaqwa6CErVrVCJG9u0K8*o4UoWuPEf72x|
zTt<1G?Jf9y&5TzIbnjl^gU?`v;^=}?z&01!gP{i`;B~~)i}z&_`}W}5XhyHWIqcut
za^V#&f|l+HH-mr1fG=k%IOj9?B@L{z2rToR)q?c^j%Oerckt(Ke4abndkolS2H3`<
z02)?yxM;Yc6YOcePnfX)cQfxSv%(KD`+X~V8T(G}Kfp9&;4S&yezf5pr~{XiNVeS$
zXX8hoG0&RMcyAtY-X4dp@B&@4z+drt&ogQfxiQb~JeL~9t3TeGPGGwTxWUJH(5XI>
zjbhJgeVmyKn>kZ1fH#M84(`U>h5F#@_&D_s<ZZO6eE7bba>lxHGFB(y#11S(j~W%L
z65z*0wb8>P&;Yqliyq|JIh@&MoKez)W9W>JR#<yxK^7+`^-7A}BW<2UXJ1{ajLXt(
zQ<ND}S)@9z@b{6Ansq2j!`DX9AsMB=8ZrYsI7$Wc!rMG>F>ltT&Fj%X(W5$-a%smX
zp4q{XN?p&K%Gwdy1W$2&5jmy0;hGx{Zybn^w_LdXfH(Qd`}SJlcy!RG$bLWR8qQ22
za!eDQY6l<FW}QRJhdVU4jYBiLJJfWxL%A~?8i~iI+kXxX0@GwKhd0O5shL^Od!IVg
z0Q?ht)uH2;9eVVYGhM52of{plopb5S`IW;0+mx>c)=3Iib4$23eSr_UfzGuDPH4VU
z{%|*0;fO{Scgm-iLoxq`$$dwdwssBEhxc~%ooZM8P@95Vo8?x~q!)KWRsW({FEY%^
z=w;FU;})&{XwkTQR`wXH8ceY2!FG7{`{-5a?By+NIvr?N>$!FfK4|}ce_ipcQ7SQt
zZU!=Kb@6Ey1d|>B-)#JbUbT-|DL<G$S0rAmdES0(8?V#ks@|YuUJQo!1S9{km0s)%
zoDclL-?cgG!zbPwg$H>H=K=V}ym#o*Mr+NK!Lz>$9-NuXQM#I>TyQv!kYu{ZlXYh<
zJ$0^1a?*p-aW)#&DfFqIm2@#AMLrEv<c9Y$6FoQu@LeXM56!mWDQ=LeQ+RL+;aT)N
zg+}#~+<2ZeZSzgjPgk0rjAmBDv@~5?KnKUlG_6FVs!WzU0gWp2UFKB1OH&j4!rk%K
z%*AI@3!jY#9^yW}yr!h-dQGs>GX5Rk6nT(a{rn$(-I~d&@o$n`OVFE2C8+`)qnvn+
zwsR)kbf_Xej6`z9=yFL(%>PSLU-C`8wk7Ff-z2pUOQNS29;-*9rtKpG(1!ht^W6T%
z^jv;q?>od<^i!Dj4vkPFa({h)odKrPJ3Yy)_@`!l_`RHZ{WPlf6Qj&2CG<X1Nv&$?
zp}l=P)L@K<dQJ0??|u)BzT%<M?nV{%GHNyW=4+yn=c7^6dKl@2Gb%IqX3{95_D(da
z0QjchE?zT?DtFGPPjBIK{`FKCI@MHgYDe%*9-d<X;F}Mh@!f%M_JMDzfp2DlZ(guh
zJiCSFbcI>1E8){e#K|~}87@7@T{fVzgFT};+>A#)@_7#U`mkuFvSwJw_0ntQGSZ92
z%En$$jJ3v{&NQ&iz&?07%aPA*Z&$;1b~)PEwYa4np36>F!>*DaY#Mmms+h}GJs*mP
zzK&I^23h4X%&LYXt@2rCRh}(Yt=whR*seAuPq*pj9PrF*n-;z0?EyAr?qQ>c9?rA2
zMP(|RwKmu+Z!fb7l{D)buXnh|>K?Ty?=g#J8>|{0Vb!-NtG@eL^wVilF2S3bL*@B3
zg#9f<HI4==XGSoy1cUXZN3hPd4_2veXoFpY)v$iBo~H+^Pjs;UO9<BVlwcWJ2J3p?
zU^N;PtSTGBc$SB$(ep5MgXjI3g?$rzvmbmD3cgtgzG*QBem-lma>99DYnZI`zVL$;
zsw?M!>hc2Lj8A6f0XetTO=~OHpxUb3wzgKZsG*>8_;Q9<QPA({^65@StwXx1HYNjK
zKV6>obmcXt>mfe$gS(UHkNJhO=n5l0^mRM@#15Ap9f6}@tzrFj*ripcBbjX&sYQN~
zs>FKj%llrOBUzX6A5CJ-C{8b_C%TUjk2TB0dhLPlnso>dc8y;?Y(u1$evF{!BuuFf
zY+9GquCcB8-A=RXk0W4E@XbQ-O&#z}U+_&2@J%H6W*7Ko6!<0;e6u$YUT0~TB25nc
z0lw)BzNrSjnE?;}w2D*SoH2j8J9X!)Q;{3Ob=DZ6d4(c0h%AUZ{PoL8k*anW?0f=G
zrir=75oBzupilLQ)aXCx!FfeLS~h+T@c&-H_=@Yew5gj*55PCg!8hMVaIJ!G=7Mj0
zXX3}9D=-AFL`yss`({fs$jEE}-&BdfPZ}Mqk>HzEztI!ejI({8Xnf4ks<xf}lLPQM
zXYixm!K*{xyw@T4yvodI;Mwar1C44L&)XHua^U{z$;Zle#*0j6?pzzs*tapd5fH=7
zmly?<fYYqS?4ww?1~^XdG|mp+$O7R#nT|fQV>DjlPwd(F40092<5Dw08>%JfVsreu
zJ;{!bPtYlJ&~Jak?VL|g+I?mpy#3!9wI15&PkhIT?qtH@a!P|+F2Lt>El0*XfUaX!
zMb<ao&nUy32RNH8UidJ<L?*D!8n8`OzD7B8DKq;@*B8vkx|^WYm-!yRQ(5-lc?Mgp
zU7R2{w5g}V_`Du?A^6^+swb#O9J9YnbQOCsdn`Zq13YrQ$IQvN2$tDLerip;YRrn)
z+oAEQ)+t^C!QQ>_-f#oyA5ZkF>}14W<F~j5FLo$S;p^auXW@H7o4Vc(ovJ!-N5*Nm
z4`;EQXn!wb_2Uv;*x^`h!uz#v9-LUmSfw`s-|+J_@bmA(S!DBKd^h;#ipIu(yYP*_
zjMkG@_-_iKJ3bVcS{gJE%@WD)fge3Bg`|()@e6}(x}Av9&K>j+{~4vV<LMm0Z_|cr
zX!B?ECT4#Rp6!w`ndOVWYZrZe9U?XTI~k$n5$Is_1%uaL;dQAFjwu}zu1kf`hWx|T
z1P{u=fN%|o4cCH3XjN^)$&>T38sR$H3f(C$KH~M@8t~1I=9rv2Iym<^G-<g*+x9!O
z=>^?;pB(z<hl3fm4poA$xpLE?I6C*npu>lqgy(_J85%{$O?|p<(4)SCpW--!H~EdO
zz>2@J<KYUv=hWI`XkaVQiAOnA2JU8Fgj1z+Iq`)#;5ovy1dpQu&p^&nVd}fyt~2M*
zo4Ru*yl+zPqoK^O4ORcOW+h)StKJTa3f;Eo82+1=oIM|mwQA%h&YvHwI$*YGLA*_`
zYS_pW*feUJT?5!dZgFp(!H>PD7IWIs5&NS5uWJh*cOMOW9NztVXboB8=^2gJN%*~H
zpEytOtQ+!$XPghXXEHgf!|_@WiZ;r<asqGbl>P~-2L@jR##zMmn;ER$%Ys+F8~4Fa
za)J14H+^PyWHGwgJJEHAZ!$lbn#T0DteIFzMGsVxgO1bvekodv2dDI4`tsJM=rB5w
zZ`M=|uYk9=I{A!#sfxjaQ~yvZdt@q|?rHjDNYi9EoZR4>y_1>KKz=)!*DZLCr{KvM
zx|$w7JUInlr|Eru^eN7(8^Ac9@Y&SLNK;%%x&p~$U2m4EgnTLLd#RGLF+(Ef6f!+&
z$;wM#%_JphNC~_!nG?0DGr55z^yZ#uO?5b<&}mp^X`=4^mL&fRXj09=bNE)`Ioot9
zNH1b9IMFvzYVpve3SI2_-M~4uzFq$fwX4%8_V?55-ANU6<h!r-{NbZPSv=tbJ(N=1
zL+1lLH1nCewpaF$TQv`P)$||}<e?HHJ+yPGht}-#P$>AOnVV7j;cupZZ@QR_`Wb1I
zsu}eiob#={QR4;~)dPH!4}7y`hfzxo8FlL{f9JJPt-pDy`(sbCmhdd>Gpjz3m8-&x
z<f7~maN~=>IPKtLCZU~l<~*}=B);E4_<ivKoSq`JCfn%toPM<^Jhcuyb9Q{S=*Y=v
z0NdAtTh{X2TOFxTJQb7ZEXYRJ$N{vasRMXk_UAeId!#PBjnH27qJuMR>XOFw%R0xp
z#yZA&`I}Ya)2-@K#j3|CR!u2q<@>Uz<5v9kn=J}umEUC1=XDmPuvV?LsK-)^@}S}V
zIocwhQE)KBEP9APzhP6e3d^hlo@Qk;nl;G7tg^+;npVoJqzJS6rkeHOk(msrMH|65
zMt6&HhM80}JXAv-hUnUe5T#rXmia`m+`%^w_knMA1*=lKV0nXKjy40^)D717F%{IZ
zMFl+@5vWe50`z5HfD--;(BG2-)Uro_2DS*$ma%0y?*yqH`-m^g#JO;RmEP96%v+kx
zS(<)KJJ(?L`pFszzDb--2l)16-Pn?>737d=jeyq~f(N-QxjFn{<yX*`g!eQ)ggq!f
zon1M@wKxYEhwR~+o{fIdM^5H6I`MOor-tj=ywa@yxb9wG2vNSTX4#6u2UoKwwU$Mt
z>R7a-p+!3zTlBo3RfW;OE0(f?3E5{#^Lz2JYPp}4*<e;JL%S<MN!j5qzkXNbH!Isz
zWILF0hfO&Su=ixxROq-(dr#Ta`<zWRzu8nB?Ai}}bA&V8U*MY@$L$K>UKo28&g~qT
z3OKkz7wp=^+J4Qh-Z5lK$zRk6WzU7ft3AUZ+Xkn8UkvYF(W$|B1>TyRD)Ppu*IC%-
z@jo6qLw0o+XH9bAOO`Uv7@tkl9z2<6ISc1x&N7&%b!?>m?gh31^K5+=DYqOhRVnHM
z!;uXSbLl__v?=gSIq;1){7n_~sVD5Awb2ju7@}lvhHrvrn0=P$U*OSkM{1rYTJ^y<
zTM{_Wq(<u@_@-XpX!(q%^9`*keP6V)qE!xB0PZJWmAN%NxaZM3gjX1Sx6k-EH69zM
z4#V-?4CLGZw{xT<J@o&@X!SNaRkAbd3SaAvukZ(W>^p|jZ$Qu7vpjGRyj?LHT{x^h
z_)>n&*$xBaoXVb{1h8{b8hds<bg7Q`4TdHt1{_^vBU)$%9Q+mT|A%;a*}oUUxom)Y
zsh6Mp2p*dI9xQm76JVGna5{a#J2hA?-gfY@H2!P=pL34SO$OUM1>0=SgqH`s`#QW(
z6u!cn|G?kC-8?u1SGgTe^J=`s^YA)PKp#MxD%%;aLt`>kmC-Px(5NcV#|-YDl#4So
z-kV$Zn3;KoeB&-=g)hg81Fx7iFkZ7bhcB(c|2H~b?t$o6Zv5F#&RtLO)?5I;{7p7|
zInS`kV48ktQqB1}qfI4SIlmd<$M9ynxyNj=6L@amZu+l_)q)D-j6b3!AEqaJd5n&s
zCAS$7Lw6c;#UjWvl#0=j9B{EqqScEtT%O$ctNxK5|0xabNUjdKwK}*a=qNL?H_|V#
zC`w205Wg85rE{I46oBuhAsXkF%u$-gvtj){mp*fCl`7}bznnMDtU!BDC#&#*KBw7a
z9O$q43};eoBzy@P_!RzueO%|gx%OYwh9f}>s&dAuN??vca5!ICgVMo2wVe8zgPwY@
zOv^zI4Xf)=c`#0)KCDp=Ro?2LqaJSNgG1JD_}Z>GlpoG!&pC(sop8wSPlx)?cIe$H
zcv8GI<I%&s;A=joqJ8mtw-TNo8$SM^a82Of_uB51V~SH*y3=D*)v1Q%@ip_cU(Dfa
z-p6}0KTIF-K;|hA2Y->X40vT_f3qgtFfk7?R9gmzYVjb>_ovMI{fn8Ndd{QpID2wg
z)gBLl^Nv-26t`);iG3xS^J_JmPPMklZ3KJEK|8a0T{7@2>h)KY-X%t>FuAJ!yyn~?
zCBeUCq7NkMEg2PfQS*DQ3C``s;i8v!rOTT8?M660b#&d8XjB8aR~|N_#{k^z(HpI*
zTY?&J%@*NWKE`#Q+yz{}ucFSQ&qUn9yF%tSFCJV2vl{A^AX8nXk_OUyQDR#qGI^Es
z)5z>_dVrhvq6=pU^RDU73;LF#CHQ@_r!uFZ8(FGZspt-=$^s@y_)NEPzBIb&$xwNv
z>4Xg)Cz0$%LuMy*Pt$>MX?nMuSG>lVo`HSHdVeWS*B`vi(qm~Vg|5{buW}d}R!@33
zHm8GY=pa1D^P^C=WK9cCR&#jz-d=RkEP|iyR8eg><8;nUM?`c*okV-yiM~9RPQ%k%
zz&Utr+L7z7{Ww9lAD2^yx8=0;VL9dH?QxsS>wQ6VLb!-d4m-Ve?BS7il|CJ$9n1Z-
zG|E?#_IT=FgHbsTdFZvfhuXe#m&ZMKwf)UQrm7w)RNVugwTFfd_s|14oY6x(=oa<R
zrv@I{SII-4onW3s{+`t+k19qjXkb)l@Xe9|M)jRw)aPYJRorf52BA>{t{Qa%d^7I3
zr`lfe)DNE9?KUzuw+rVcabK2-(1L&IZDP&*7%o4!(M=b^$*tm@=eaR=cew8V4JKdY
zpxfV}8(kf0RL`MSHF>Q8_OI@c=Wh;mz!O@tf<qU}J9G%|M9nhXOa2a3MPtZ(FHCvw
zhRN#=tG`tTTUgX(mswl3n6-VYS(&z*HDQH?f77Ct;2ayPPzWBg5BSgMM=+$Awa;qS
zoNQ*5L>GFI-K>*2&AOD^tRs2Nit{m}MVeL4Y1S&1d$?IcS;ew|Yxvw#eBNXOe+JfB
z2al6n7HpG@w)D`951V_+VO8$j7In3nbSf-V!|sRB+Z3W%jY8xFzA=Jt{srF@+a0W6
z@J)8`&0_G)F7VACqbn#=^9s7yJy1DM1Sn&7fO1R<(8-B(67>ww*e(IORxv=M2A5U&
zfU<hM&PT3dzWB(>D85`d6=_{gt8bR0bJMI|jm<jP-mIEk%<9qCtZnQEXNH=!a2J^F
zDL)6WUdtaAHO_5SWU@)c+xcsLf{*eq@KOCnzFK|USH;i!%I&JJCf@W_%X_{W_sCa0
zUifNV#R}?vBScm6SrlB^qW#q^`h>^t3_O<we6s_5GY)*?L<3KD2YZ5X4uEe~fN!>f
zZ_eOViU4mWw6Nmcvda3CGZy&f={E2T80Y4Gat`2|PseOp1HP#UzUc?PX#~FU2H(7Z
zuUlIlPgFTHDYW=qrR?hCZdWGOkbJzdCaksTAZzh@o9uj!TeUD1z!y*+Z1XG}4xBUR
zy&$JzI>NR8?v!~U**E-+OMj;yetWp$>0WI`A4tSl`uxa?uZ8dbHia1_k0aDNDN<L#
z-b3&=e0vZ6MxV-G*rktPoSbfOI36yw^>mTba4EJVxhr(kvs3u<=`QBYL@7tpD7C^n
zIt%}J)4BAU;301czL}(GvPYcjS+SAPXkKuc-sn{o=y~ILcqAS!2S4h%USw{L)58j<
z-h`h2u`aSxMdI`)uiu8^#qWcj(lAaAIMHSsnQ!z*v>v1Bw*3cQ2o0>k0&4c*4C3!b
z>;ASFHO(BW?G0mf8;;e=^S$y?{03m06JVU7(O_mUj$c){l;-F|eG_zkQUWvQ$cyYm
zH#$ML-gUZX*uP9*oTcB%YU7=0k%!p~a5(|^a%#fobRuUp2%csNYYlH-;bR_fHr)zW
z)XQJ{A=suedt}r%&b@q{RZq#I!|T7f#CZ?yCSouA(MEg?a7X{m<ljM)-@^A|=d2uB
z56@X8mJ@xd95~1Tzmtvr;<xc+ckmPAk?66FXV0JXRZV1uP+zpD7G!?Y>GO+-*H?e$
zR^YvffXDhZPxJgKw9{?)nHHg=j-eZ|JKAXjd^r{ARmA^iLYo@=f!rS6zcXN)YWQxR
zN5`s*7dfhD<Z;%6XJ*By^Pm{LZwub35u+M`F{)aC`3qpw0kfi2!4|FSS;4YrrA<@u
zVKow67n1J>a@_}_RBZ|waJq+Qq=Ro#qU04$C-GaCtoXP--*hPe-$+|@Dc@meqHu@X
zdNXg~L4@2A;TPXTFnfX?Upi=VtweL`5U%uVPOU$U=dY1dQ2|bL1-vx8j(|Ja0luGz
z52seca21^9)b~KA(!nfK8#-i9amc}*Jf#-;R2%l`P7aOc{RsBxkMRx}c)tvICi8C2
z4@dC>!|j}&h2L!i_ZOPeT>5e@mB4>fF<j$V`@uUE*E-d8g;O5OIajQ9YS1jY{P6db
zYK+zu?No20Qz!4!?bqI+IZwireQFpp5yEubAFt#c?)h<e3O<;1^r%UfriN-;*HGrO
zg(@vIR53L|)eU^JZ;wSK;nEM!<XrUvKP31jC;Ux&@J*>-_@<>zSx4FE#<k(yx9iO|
z`X74GJC5J!(MoA{y7a>(-QpZR3H~#cduewrax$DFaxR4@jmB%<8lOcQt_5(<NVKYN
zTezmUXBy<hFUxf}+Km5{d#1>2@b*o-$Hn0_o6)7P=T~n+qGIT6X@EvGtqHo%1?EHM
zucTb`;3V{^r06x3(E8{^DVU-TO3{-J%sHUj{ZIxOCbHOHJ(<rCN4Eyv-*=Nzm9aTh
zQ!l0}@il&#Z1`)6lKX*&IbQ)TCpt}`wb7e8F&7HYacBHE-wx2H`j9LYe#kn;-xR)v
z%*T>6wTMr{5SXg?(ka@1knf`~JuP^JOLA6@<xK3kE>X*vJ<+sfyqYmL?_kd`_-~s+
z@l>=}WR~@@m410TCeZK3yvL(G8ZF(!Uw!4L$|G4f{Fs~QuQNT$$nINS_7wrDcF3kn
zaW;+ZZ&TUlHl@S{G4I!38ScIcS?j3<`HVWh%0naFz$@3?RbY*~Mz(cV-gplUN$^m^
zP!B~dbXWEM?wYpSpfj@#x;4;1e~N*eFWgQnd`=?|?T9sMcA8Ob!8iT-!R?GTYG5t!
z4V+G%&PE-WiZ^Gsr;Z=<RCED1b?^z+JZp%opF`F8gh?6j8qs#MO2BFKgYQ_vv$bds
zvl<TvcZ}uz92T{&XVHKBEc#8gD0!hpxmge9S>!jH-hw|Y^5)OlPp~jkozEMCPYf<)
zGF-~65f(LrPr1aivp3iFsn_6%muA&Kcil75Om-98-qEa%JkuYvGi!4PGjj{jWLbMz
zuj`mK3!L)vh)L5AnACHvN!yp3^l5=fn`WA{Z=y*b$C)&Iib><Ao78`qNy}H8^!bWO
z+3uON{;f%SKAF@8eDf|poDKM99{6Tw8MCf~Z#sc*zV0?_>=-abRudkBP(Aey)zqaS
zn$j{vzc<9&9T`HuNQj2z4AH21!8&XW*7i?98qaFWN=ggTkIog;x<&<kJ|3X#e+TFj
z*d~LO4u6xoYk)SD3?O?`R^!T-r32JQWtaJ=<P0BlKi=-+qm`{$Eq#<2=A(oDK6+Ev
zTj}+^bz`tM^HsbxWxuzcp7GYRQe0~#{Zy^EpH{g0$r|LNc0GLL8}F;L`+arih_7y+
zV_or8WAKd=e6tvQ;|sp&3%+RvzUc>!8w<WU3%;2FzPSRvDcl6C12<e8d~?4L`7%$d
z%7Jf=fN%cAqjUy*GYx$6m2AWQ=2oSGZ+0Zu^cZ~8pt?=H&<c05UUdP}C)w13KkFN7
z(^A<K=HhcQ*)(wszV6;u&FVtOMQ1C$@m95E`LeR^2V)%o-?AEjY2RJ7X!irNmS^U3
z3pjLi9X^~J4rZr2m8Tt94bF^n1~~PyDcRJ)PJIURbjO?1bfr^0@N6|@|88CzZil|J
z0TapjXE5WKeE6qDk;->1QeFN*56J4$fBDf0z&Ce`ah7%m<A85^gKsK=Z(f3LzJhO-
zf^Tvd$-Xq9?+E{>VZ0QDMm6*znVIa-I)<MdXQf8EqSfp*vlL%Sm+@y@D;O;YUetnQ
z$(`k+=>>>ZS@hfVrsS!Xku5&WGZU=h8xo^gOX;4+E7Ferw_CeddDMx~W%!_|@6tb5
z{O7TKovDS#V-Wt1N4&xft^W|C8TFWD(F8Bfu{hoKr3Y^idGU?$N<L3szyUXt$b2Vo
z#02<~AtUe}&jE9+#|MLsKIRPA0*_4oCvYcdRk2`{RIo}>Fw641cyYiv9l$Z$iqHkf
zs>E7chzvNK%@lBu1#iM9c$-`Ny*KZ{HRw_cI1^U9PHqD3rqW@0ly`um)`D3UqLocy
z&gL+_2hPZiTJkaW)65mYG&Z_Od^tZCX1>Hv=5D>Dzxg`7`Wf*mwkckQ1>_&bFgLR&
z^GBQTu~g=$+wtg>W>!}o&fp*LTHhiQeguBzFZ@ii=@<epcI$w?UK2kPTv<(jJlQ<w
z&OXDRaT4AOe_#``;0^H-o5@Nwe*&Ia8-sT;hK|q}eQkj*l^!EIeOvMWk&nYib(2{g
znZd75$VJ{iijR0SGtsKz#VSqR2hY<&JELSBMVA2n;md*a6yI>E3%c*EzgTFU4Z$?m
zs*~$m<kG<2F8XyNb+idQ;}vFC&WccJx`Yp!ncaOFUgl5wft$d~oFeD6kLw)$=PuZ$
z$|HvsY<6e@=gZ%!Ih2@+*QUNh)q6X1cDzG*$X;aG`M>Ur1#qChc+fX%!&J5b9B7O~
z(Y_ANNAE4rma`{ZO@VC=o&Rjt7&6>>=Y(l$Hh3L)ppV=~gLA+Qod?Hk#>>CnsYZNk
z_%i&%+-HjyaKCZB=*9PyCymby!TbNop>KN~%4Ko<|5-L;beP814pRUb*4*Sja{U31
z?6NB+z@)lmLuCZhBvc92yXa62tOjT9Wl_&979Ad8RnC9dZ{AwzCgi*WzR3Z;xemU$
z2EJ)O#zuD^JTZLn?r(PeJH(-)6Uh|7*>32Dhc1mgD;oUxUGzWiWu6AJZ|mWGy^TIu
zqXZtU1+l7o8eeE4XZ*R~-vD^QEAd)?nOrly1cyp<U&L`P7{^Ss3-lO(>BqF8E8z*(
zJlai*BXq#NPEt8CMqTLi81WOH-NF3Hj?7MCR>R*Hz&hEPJ4Fx9hUWCd(VbUp53`-<
z`5QvdUy%rAb2m*@r_pdN>r%;|rmF6fR5^d7s(60#Q^qvqD^FJq^Sq;~pj)*?uNsOU
zhfUzXUOI73&~Lnqjt(+b(fD)LqGQFw_ju1q)e<_0El254%}b8!a56?vI)m}gj2_Qi
zZ_XpH@PpQV%<ufRT_aXnRqJ}7qCb~cjjd%=-sY#U#lEUtz)x9Y{PZNhpT2$a)yUa?
zO8m!9h4T0-_d$R8w<xD7Jljg(AsH87Qy;rcUr$^0>8~JWP5G<UV;}uI<*8AjV3HXg
zdQ;E?4~n}&Cb?@(lsmJr4f=bgK@CS3w5F#)ZJQa$n;XdM7__UuLHAl2biT1c5Au7c
ztA~dkb@NaGFixvPqb^rQr;0bK8{AOnO%JuG0pGLMO|v??sb?`a<=9tDUf>&=)YR4$
ztlpMjW%vct)fucotkhD$I$IiF$<APU5O}`~SUbd|btaQ`w>Rl<M{vX#la7HYJir!r
z{xPZhE%emPX00g%A7V8tGR&-ea45~dCBJl_JdtLdXZ?&aE4>k%4S42TCo^;P@WONk
z*YN&sJTm@Zj%G#STlhLlUYJx9t*FPRQ2qNRR68DoDhhvZWJah8pxs8?4OQMeChZz%
zQpsT^Z2;d42j83m-{e|j(h2ZQJl|VS@Xhd#CVlyCk{|fyhS98U=u-#z9zQfUE5}Iu
zZg;^#_=nGy4poIhp}M^V4A3D&+1rHZZeoafx<a%#ONc%s2a`oWpE?qxCDVfx+cQWF
zIt3}JW00D^tN`~LprQH8Y0BHOs@Jcqin1<)X_m5LSV>8&zkM`!y^owreKZVw;{@My
z>E)xfZG2QS%}4cYKFnzL#&_hczv_9b%xG`*oA0fktkVm;m2Zi+cJA`l6L%kdE9Il~
z3O;mo`Di)#rtbk?e5byO0pCpg;iFz(OVPz%N>Qx&^Gj**_EJjURz{tsm&0QcpjSfz
zw0Ky6{Idk=P<WvJiVWoX3)K4PK<$hT)W!yZYSkuCliLL{r#nzXrUt6p^gw!z1DQz{
zsHvIZf^vmuK;95N$QPnZ#X^+7Y>3|d7NY%yLsbTxcqbuLr;|c;GbL1RJ<;0-geq`Q
zsD8ptFPsyqT-!Li?FiMhU7^afCsf~S!Q-s~d!98bZ6><?vM}vUWY4_d&@F$wi`79A
zZRsxV%AQ}DZnEo~6*J%p4?9$LhC_=F*>$KYx&G(yJ@DvT$<7|`N9P||lnvmU8f2&5
zT#ZyT_@-1Am->QlI>6y<1mE-r-vktQX=s0!YT^6Lg9pCIR672^H-)^S^b{|}xcriZ
z4uDS7cN-_8Re^7gf^TA?;8b|_eFE2Xc!74oI`owpI=jf(JjJK6Pdc}rP82`9$8a%s
zyOO6`iEag-)9zNZ4it(}>E<!2fHv~3CcScCp5MVrpE;903gv5Di;`a+^e_{%LE()u
z(7Q6`$Nc{c$XBW4gLadNet~~Bo?hcc%-lYSudY0O^pWuGRhWALo~ZFVe9dHV#S*lm
zE%@)iAAQfmrQD(~7>wf%#_95f8HE4Q4FQ*v51ccfj8$o{&g@@!hc$=yO?><&f4<}=
z{oi1l8R*={U*Z*hl%QRlcZXa|&@i~047i)$Hlt0g;0z3|Iyyc<D+co10BbF7g728`
zt4j=+Clu|<D?!8alW+eXuLDo$^Sea1{sEpf>*!~jO;7O%I*Pln@C-Ie<m0Az-Nt*g
zDLZqhUZbsF1>fvPr&@(~5Z$j-e>!hk#;HO&-a-c+!cuT$nd0Pi7d%5|s^B7g8u*yD
zB*v0=r1$nQoXtAsr4EnLy*4pwoJyx3+|5;fFQ&KAYPB1Eu?PKq@N)lLl`7-AnaIqA
z6N!?|P2_yhFBeDY4BSl-IPq|gDAl|Ko*@(UVhJ5fV4ME!T)LVF-(aMV;0hkKsgYWl
zhF%Kx&Od?Kk{y`=1s>l`Ua9&Y^x8PV7hs!pa3#s$h}y|`>A4R!gG-kF;$@gsSsJeV
zT~UYLa0VFhjV_FXVX6wA`H42vhx?!bU0+{nkYid8#=s9_Ho)oN{eA!4uG~-U+Ofv2
zV$0BuGwdq!$xioMnBJz~n@MsqQ^%>!<gu#r?~jGEskzjtz4Ot9;cdQ*C&$<cT$Aq9
zTPH1SS$O^Ry}=!Ig|qj)5T<>D!t|Mc->n3*HMiO2H{7n8HSEgsmNUP1h!$8vG_gXc
z{<endBlu=`VLS)hEIL2gsy*PFT)AvA6t?M>&8FNjHhrjK(~pienaA67V6jd0-g9P(
zu#;=U!#oE4tzV>C!P~a0h;Ob7oa||ruFyNQ7Jv9iygB<g59F<mZ+|*|{X=-%@R?sk
zzsxZYKl++DmA1umri|x#6R&|GWIwuc1~|ku>q{2~nof<KiF#QFFC!VGb<7rMO`q}A
z^~vh~E}1h`CH>iq{+pS2jgK+!`g<j`j1*<mOHr56DO$LN{`Nb}r^-X;2R(m(*CjtS
zBvm!xTuvT^)47Lkg+|pUXPPFtp-K7Sx3Sa7mz*YNOZ+#3(sTju&dx%q%4|>7lic8#
zvdqe^n4%gj;cV#YJ4XgRE&yM161{`DE9oeHn~WLYo80(>H*g<wUYXb3rFMfITKg(Q
zf0gvs_px3Yf5b~q=6P#vTVKtPua+M6)yF5knmW{1qfh%Pwws^wUH8)_lfR-qmeG$x
zWi<ewYNn-l!Hd~s`^5Qlq7^<nL~c#|)G3kYQHG~h7ct7WkB7!*@lfOH?rI<Gt~T!O
zYPQ#)JJSqufNu`=Fest9LHTMMWCGvJNd)J#Fes{_K_jXg<W|^2m5X~YOTj}$m%FR%
zOE=}Z>L$}EH$6V<rYYNssp{5Zy1=ToyO?_JDW(qZgB0W)tnLxPno$8ytUF$koWYz)
zf^>aukOs~UQo|)d`j|af|KtvqPhzl^r3CA1<zN-Ajt6B#F!R8J)o)UWZZ8Va-z!4Y
zd3A`stqoC~4Ivu0F+`I$hp6S&5GCvk(biod`nWqp9m%;j$z@{pgGrMLnzXk9xFpD=
zIl(6N0YlwodAOUjvzSRK1x(tQ!=z_C|332U`@plW^1)E~fp4;cZ=B$fS=mhby{Jjv
zV4R)co9^J7%^OTwc;18#XVR1FCKUwVZ2w>)2WQql1<eYAzo{3B*SH?u?zV78Xk5q1
zEVgkAg`*GAvXlIO#)hbG`w&$F-^^$QwgKPFL7y5Ijfc2Eup01lx;Zq6u9P6n1lzne
z1?f^~kZkWOXg?g@G;nYgH$O#zeWSi`ov;?O+RpM-##CSBsPC%-iN30S-$x8C$G7jJ
zku7}q8T%+;q_;Z2-L$Rktu$W4z&As|HlJAA!8dRJ^wtRQ&3*%aub_{L@aMT$VXM5A
zf2B9`!M!!;duh3@m(qx)rSzC}eS9ee%`K(CRi#vaeJNepR7$1qm(uxlrM2o^X&o!!
zqaLMwblTfTAAEfD#MeiegMG9VtX0(Rqy5o7+S$!V74Q42%zyr>_0wOMvy{=9)@5|p
ztDG8CD5sXq%c<$)a{6oukZn}}{_OxIp9#>DdjVSZAV4c01*ppV0A2kUpxW6371A_N
zm%0V&-TXlEbQQod6?ATP1x@`JtgJoJ1pfy2uA=vgEJdr|t=h=B)yQ7Di+z4c2e|lO
z`zl)1(9`VkV4DL|9SZ7ZSK}Ra4UG#^{4=s4rI>k4X1rbX2-ROg*F5-U^mu0X>}BrM
zUGn4E@Opx8V!=1#z&GW<H__mm<=~qt;G1*f=~e^ZRPrSEvH)%BN0j~o--LMKxk#0c
zPXM!I#$SVmG!l=!Klo+=_@>EovQ-bHlwnf+PUsh7=|n=y2m{}A+$y@*;RCuwYZv$?
z;262`i}-A^#%Mw|IGfNIZ4=Mb=P`1flnNV|e{~mcDmk3ipTJlA+d*BWDLiW*bdFK`
z8nMdbj3YD8GyXK0tSrn*#;0=j4!&a}do$SK1{}%h`t;4PUq2jy|9EzSn&B;YxRY$v
zQGCGQk#e`;U%)t*Uc>i%f_wRnX7n>bS+k(agJoJ~!-L4$&-;VmY^wdhRQna&0=DTz
z_U|M*SK1vkD11E+PA4ck*r_F7KQCO-4}8U%k)VG@@%8&ME2}*-(Hrowv;=(tf0^*!
zyz!*hI5%7lJryG!FdO__yu$a;VYrH(<{9zIJ|tc-9pY8CHeTvjJUJDZ#bTh(A`^XL
zXmjt*qfhMu$1FjoLyL&+iT9=vGc=R&Z(Go<JjjKA#aDPGR;#z*C!QLspB?_EPi45_
z!@0{$sWmbB*qgcVbonKcef;Mp^D^g=nLt;ZM_09b(P(w0zpwdXX?T695&wn9K~h$(
ziP{&K?>Z?;jjJ#h+$T!EeWa%k5C10Sr8R2KHD&@Q!`)1ZaM7p7>lu8BW9VeEN9y{P
z2t6h@YNc1JNNMz`#o?;d55LV7r#>u1E1Ce#Xu_=RM5mT*aL{MxkXu!}0wo+842G!c
z8^(OmFx3J_{2d%7&&K%jtAyzlKDMvSM=-N|rZLxYIr9;Eh3O@o!YA;+%{yS1WwBjP
zezz;RmtD5yc7+aPHWb;#&Y8mS^@CyHNms(v3}a1M1il&KWF{8A{x(i6Xb!jJaOy-J
zc+<0RNMMXU;I2?Fhh`lJQ=w+$r`*Hz_=a7-;V<?aV3#}iX53So%6zB4t$2vCai(<n
zh3eJ_&e5;&7tF@{mw`U@gMP9cHYMhU8z^Mc4tU_eR5%-Qv1`CLAC}?m`v_MIA9P}*
zU6;%b=C3)GmmJzBJGtCB^dnZLCwd99a;HTrd(IfOn!y|~-&pm<TiR_mS#rEN_iK>>
zokA8ET$~&Jb4`_a?ZnqUCNDD_xHcze$HN8}nM2HOFfezMd67>mCn@hmI%ML}f5?Yt
zJYsHx5A$Z~FpF|rC2iY-M)j8Ne)3CJJjU*WQ@DTW{KI2x%aW>#0jWAo&tEG%#%uoM
zZ9K*Um^n26ZK|qdVdhlfG>yiCQ{SA%9P3p1_^0Zh9blzL%+a2o2ycrgw$-oREWAU3
z+&7$AhJ7Y~*n}Rv(-n0foV@o`a^Dvdb!20rLU{k(8oD6J=dA&sehw|Cr9VrlM{zF=
zs_La^HWgcg(i)CFP<5w|vKa6OTYQ;$?JKWizWTd@AD%IP?T##?{fEjZ*Tk|qKDjLY
zFmy8F|8D)1tWO1tDo*j!q7Oc5d&^s24|po^y@%Y}kdu1oE>{FPl)t<37jxIxSqAl(
zWYFc|2K8@mP?4qvdDb%MO&a(n-k_CmH+AcQd8!&zG0LFj{stZJFsNQ$gL>U{Q|}Bn
zO<h||8~-e(zZVzNLtY~{7n9G9Vj8rmm|FJ@($cp<a>3IKg0GqIGe~z%2H{@{(&X_$
zT0J_5e&!&s9UR4kAa$J-q!aKJFXsfwYhjQ&tO}AF8co4H!Rmh?SRZ)ycg3Uf^GvY5
zJq-pM25aW~VD%~#qUUbl1Fqv?;PuIz?RN)(8`DG7{?8Ci$^c7vlE3DhKe2bHk_Uz=
zWCT3Tc=AirLUot({hB{RwE=wd34F7JoOU+w&4(kQ+V?(GUBNdOz&CljnDlglNnOD=
z=D)y8;G3-A8!z~q|6YS}z&H88H^aa;KYZbsLd+^y+pJV_R1vRC>eR-hxj(=&okMi=
zCimQw5Oo9J)CAuc;zM-R8KT~Yz<PheXW4=^_)d`eWPk;G1?gCmAiS<Y$^hRq&x}?z
zr-I_D_-U+%pJu%C)q?B38nNA1FE{#X#adtW1mA40=c_^B8!zxpW$=wJ*de8<k4`P}
zR&z9|d$qtYHNDl67026&;GD({yfvV)xBT0qS@rkUp`PAq&<}2>yEl0-Z*^|rt$)Xs
zX6B2Rp2F!F{5`abb-Anu<2Opl49@XrSW53qr8MJg2{l|*f_#2)bo%1@>|Y#BuDGUL
z^VH4Dp6YYaQ>$3-@oF_HLmy=M;_82+xE>~zRFSljdSAVyJhGKi>NOwLx#6Qu|N7{!
z5MPZ+;s2S6*T?Fsuin0D#%jxI8SJYUp>Rr^4VEPNYGD=52{rsQq*ED{N+_!r)yisC
z6Lh<lWwn1$SvlX8Rj!zFDqp*tPDYehwQm(PB~ys@9u8I6D)2gJnVr!nOTzV};DLU#
z4^R46o0>Ma%fvo9^96kybn6UbKYsd<J=uWwn!cjORq&a$h)~1Mk#Y=+)cToV9X!Y-
zc-HLr7^wmHal+l`!}rG@!5&$gvvy5$lxD^=ll%qU)LEJTOn&M+Jvd(~NiAwp<C6*h
z4fy8IlFY`jlM}>mV}-Bji<We)F=u<Q$mcQA5&ZcB*8Qs3#-oZCXY_3J?d{RpaU@!=
z=uYVTl>US)cv|5M@;RAR{ovOuM`=+GY3a!*MNDIMC^%<rKB-wVM!(g?o6`g>rCF@J
z^5QM%!90R0Jku}I!HK35gMTGG4|u^JPYnBXCh{RYT7zkRC$}__jM5@_oxkY)+l%IY
z0xWYG-vJn>H#o%rUg?6~G@Cth{<j2;`j?z6cqW2n;O&3GH6Ou1+4&qZpF0>_^qIWW
z%<J@pzz4NI24>>xmWLyHz7h;G4<3m<I3L=SS5Nq!R`~vEljll;yFo*{TNdw+feu-o
zB~RYL)qu^~or+iCo#dvL(Z4o@S)=%;AGVIyU)AE}69x7m_xY`GyqbTfZ}b7!<rI0D
zEqIFOGmn%kTBT0(a@FQFj$Vskvcg5_zMz9_<0)odtfIS*?8N#e_>9Bo4J{a}gtIaF
zzAQ#F>!3^d#AwlXIJPUyV3^9xjgE8_<Dr_L7fle~%|W!uj@2bMUw)^znYZvKTn*e!
zHMFS{G4u!&#`B6VeJ?yhP)?Wj!PS%^fAR7y-IA9gRgC#(SMj+UU6FLJMrb3RVsonq
zuDb{=oQ{vIH`%3|PGv$DIu4fT-4p*!eWxnNfhT(41#V29@MD;QT7=OV1TO==2`kB5
zry}@y!6rGXhiN((|A-sidU#YGZ?!9SzFpC*k$Cv4pRudpe!KerjprB4F}a7GED+rd
z1MPIZh3Sbme9IgB|2C&qbVdjMgE?<Az%+yDChq0bHqIAE8^hs%e=<sdZ9Y4+7k_56
zo_KfQl(xMDKP?VZBfgIn`N3J2@!hPnt7Uh)7Cf?PKyncMR>A55CmvXg&JDQUjw4O_
z@B}`8j#cIZR`t7Pg)_HlNKPAEFkU`5oc@V6wVi9zm=Jt=ap0O{yE@kAeus-H66nx<
zF&AqD-Bx7MI+l!5Y{MvR=Um_dn;!kheL94^e<66g0b~(}&|gB2g$3?&WF`DKYnfYC
z1fN%Uyk`IMvxBWuJH{&no-y(ac)u(7z8M-1vmytQp?1$ml8+-<x2J(^ZX|28k-m96
zy=RA1QufW2wDd8ZIG!oWm6D>|ec*7GrfBL}W;Wn4e(nPgQwdDdJyqQnrZUfl-oI=3
zjbEfP*AUIB03F5PofW;%t;(e8<^}$~H(fi%B<<%qJ_zTyG;axg?ND6tcurcD^wd=L
z?qFsy`HwB3jUMIkAOvV!O*|;hP}Kzwu7js*HJzTvuVpmgLMd(dhx2Vgo)s0n^xH@;
z<|dU^wu#;fndhUJY`*GV(icv}S1rE#>Rp7tny>O#Xh<0y8(KzXZkJKf6J_+Mav3eT
z?k}&V{&+3@^aai*`Gk)S{V1jVu_e@HmWQtM{0VE|j^6K1w#lF>GYq;t#i0D-4f=N&
zxTUQ@FTgXS;c70V7?c`^|0dj^w{;B6|Aog1GpLZCLBre)3eRd#(MxV>0G4UKvY6U}
zZ5FbcgKs)6E2gX)i)qrjV%pIzNW;K52RVGqzY(O|*Md}Y3w{!?Oi?h+7qCq^)-rI;
zBi`Qw=6N$YNF}BPX$Tm{499bg)n{Ii3eJbaU_DqGr25N()C%mA_y1Tr>$ocK$L*_}
zLrB-5Qw4+rg3@tcq&?d?YqrzbIb*ZCb#~6~+H|(gp55IwJFsUy?>)coAD`Dj6a?j*
z`?{}O8>>G#uD|o~oaTJ)XPaVWyDe6ax5sMf&RC7!8>>mjVpZ-+Ec$({vWvv&9N6I{
zx$C3}V58M!tB=MhxkkM9u7S4!-wXoZJmVVo7x<>!k$BZAlAw5ZG&3}`vfUCi3J&K!
zT+EDp=uL;=hAt+^LSN0*S7=tBdAoR7J&3`ti)PlVL0NT26RX4TkBLf9GThCYc4ag?
zw~YFqLU-v~Mo+7h(E#wxFPGvJgkCp#QJnrsCNFg_R;P}{YTWKvweJ?I1&Oh$W{Fi1
zn^+yghjYAoNsU-pLPa;0fI}*wqu`r|;F}xZo4>2U`Sl2v70qZ)mk?#`4pGT_q1s*;
zsy_dQ>X+0o+0_kG^*Ue~9wWdvp=ed@z&7*1FmW9?V4AVun`b<J>>8#yynpD)P+5Kr
z(3q<}s+sPh5$Bw>=7zJ@zjM}?a915nbko92Hx(`6u75tb>vK1-avdKvpX8^Bc+q!n
z_tUcjewu3X*T-r8cp3Z^3dUJ{-(Rnv_-m$TfDQ!&D7tTuCXWkJ+^}HPni;C>IiYI*
zDpZa5+Bbg*m6LCnz8e>&WO$}4=fV`ty}%p467mfpyHyg-3hc8MoHMp^33X`_p$<Pr
z;MI%N5vNFX_l;D|l976s9jOE0oQ#mtDzUM&E+$$vr(Ueurj}9JCve(>=^}#j8$KgJ
z?(gUe+nT5my#9SxlRm~J@a%VZz00CY>RgHf;OU2x(f>pj#{)VVj^<|Q;Wv04>Sw4~
zOYlyo40Y^-hBbtpLNo%e8FU0L%wQuCzUR*jdFHX@T+2|G7KWNQO0_(g(2kaD@x=VX
zHEw$`x*&YB)dPGp9egtk{NncjYNI;$^yTS&YYE>ooSu}ia5#8z3Ywrvjl_>kmZ~}U
zX8nO|vR~Qi{(wF5B02gRfe(lGmpnz*wKX1}9}HEY7tXnzp$kXo_<wH*`cXx%lTVJy
z(G2cKU*BcJ0bb%1JWlg@+>e59ZsI4X^BLX)UgY<(=uMTt6W^0t>P99U4re=F-~al|
zE`vdO?_pE<7#t4VOCvB&20GK3$Nae$1#rFKo;L;h6+APAGlRF6@V>qL-dp@O1z?+v
zw|V`)N=M&0u*^}qli~DxY%P%MI(l{%fw5-r`N#0N`?C?=g<rS8^HYaDo_zK|6Tx42
zZ-#il`xGHx^^C3DtL)((LMOoYnYJKb9^=@d>dkehB|X^H^EEpyUt_t~4sqste2XXh
z2A<J_a5!u8_}QhKsXuucJf!u(d$%(46jGXgX#T?r$xYe*nJfEo^oQeds#G4WA|w|t
zS&r(XU!2DS6^b@BFBpvZo{pO{*@_#T&Bh4*#ORN&UQ5+>;VB*n#sS~{Ws)wR1^dpS
zleMd%DR_$O-pW$SpIJHx2Y8TOh1Q4Ym!yxt=7&r<#AK?@hYWVxGIV(jKWo!6RG*)}
z^8ev(bkyU18A=~Tk5d))x_6O%gd>?#8(nBNd8Fa+6V>Qa$8Yl`AHV&R6qRTWo{3G-
zg76f5fV&xLNB3z2+-21i?My=#dY;Vo34Nyvlj(j2-%L){klEY=fHj8lJo$SRIqP1@
za_N|?%ReOR)wlGZ7E4wGKJGU-mqu1{TVR>5<I~g!J*r55a#B6gw5mOv&i85TgQw|U
zQM~?F$&f5e)nhzC2hqvy-~k#pn;c2`6g|G5%vL0tlqFed-@>^Zw`$%Bt8SqSG<t+j
z!XaLbi^c1wi)G|W*5vD>vRX&ZD*8ks{o&*r;0NRfrb$TBAGt}YIwwhGz&A#7I?tx!
zFD8e%{z{70N7L&=roGd^OoiTKuPy+;aeYHhHKhjhRlOUOtzT1f)P#JThaFv+FL|z^
zSFLGICx8pP9iz!s;o;hfr>k|te0Hzs1#HZwO#T0UCh2p!N0(`X3NjzAph*Usn_DZg
z*HlSs##U0zb7Wuf^H!};nI53Z+QSy3*LC{-JgVs1yeg{Dg^s^D_>BLqqLXZiy9dD8
z;4$9cmhCCJ{yu<X>MyRUL%*Y2ZK<kk_MztDKc0|NRhB#CyxD?ULH|ziGIY$<%GEWn
z!-i+<NLmB*Q&C^Vy|ZWu{Kl8!79}THbZC%8eNS4n-_J)WeSI~fQlu6<jgVb*NuACO
zR?f0u?L8W-i#8#O3JalMJ473ULiJ!#sCtgVAF(K04So!lOLhs}d(VApW+`>Wce67Z
z{-%9|_8yGT(KQiT(;$>?9#4%(^i&0R9{2mo%Dr$!Gy<>wzN%E;SBJm&=xBkD=6$f}
z(P9goI2O9^EQ%Thhcnosfq01<RI{i{dGHQrcD6-r_#XWIEo#U2)2XaQTf^|;KxE%G
z;X6L%t(oAM)MZ{eI?qeZz&GLGo2`qy^!-XNwTIU+w~W>5_Oa^RFIFG=$7;pESosa(
zIz2pACwbhA_R?`stV-c6S@dJ9dbN$!jaISnNU@6NV}Af2_2u_wa&4~$hcO(iw3^Rl
z1q0<wi`8u~&aPFlTDvAzJ=eu52#oXHAF;{<-<${E`~<!^craG}hrv>JW91Ic3bBdP
z{g61NmWoqKG#xWwonG|H^g9=)Ggsk*X2fgRO8oXa;^|b6*CQ~_STw5HC*X5J@dqWN
zdC`km@7n~W|C*p?hZ6J*@3bx6po>ov^c0WrZ}@OVlfCMJ_hvJ`<<j7r`}Y$RIz3))
zzm}03S`-Sl-df9OPCywt^~$K^u{b&Ijgx6hoQi<EP6m>nx`^inzN^O`z6Y>d(y~~k
zg71oUvZ~&vk{Zwn&r6daIW!K^;CeyIs~se-f*@_27@!ajZ#g=-DbmwTfooj#^B=An
zv&mHhH~Y!@K2$l5VRU_l$-PRLGOD9X!Qps=Z|uQ0Kf=@acMQ|}E@5apVfqbxbM%KW
z{dzrAedp0jZRevqc`hn5-$6^3+RMMDy}ZA*m$|9EOug(ibfCR%O|@4Wdj~bE>8MI0
zoHgr<E4eZ^%?WkWmJ~Og+~}q{58TwDyQh90;i>!&o_bN#OBGLfsrYp8jVVZHAB1Q~
zi%{;V`T2%3YVkN!J56Eg6C9?R;Iq<W!*ps^m<BEg(~8mIJccWGQ@HXD!&g~LC=&iE
z_jL&sD^`-6VM#p+DyiF7!T*m+%JOeXePdTj1&2!M&@Yj)B*%~|j8!OILq#i;QR8D}
z^t=PUM9&26fv<~%n=8aa)p0+a)1Q(QFpt-}^W3w-zkho)S(}_v<mt=KUpZa_zah)f
zCPiC^;^|qGqM!qGkUe52$1PRvsi|7iBvre)j?_Z~@VJDRfY<I1e)MZ(vhh5UPLnNU
zVICX$8VK&lmCB<{RfoH2<i{hOPY=O22kwD+(5~K9B2xnf`kB+!%C9RLIy;dK4|?EY
zxnBDJ0w&r@PY(FT1$^`HKKWKKP8y%bVK10xt)YZOuuN^b%#-N!2V?!V(NMp8h89HN
z|JlyWLI7K4T$_&!!zX))-JKYGwG-)nS(~q0C+K{D3t8z}AivUNnM6LRF22dObp65M
zv=~>Qo^$CYSVgXC3z>2BBuDO32ZC|t;>YQ77rgQi3<8GP$?N;;=LHG{(^TQH-&5ZA
zh}{jip21)nk85~s&f{0$<JTU<U$YZ_Xd^ghCBOF@8#R;RjfTVX_W{>*pdSpJwY@sN
zpKPuh@pv3U$f>%3wLZ~P@(6r$kv+uy@H6Z46+bs$U+DH}(Ia2Ynv$8S%pL_kiC4k-
z`qmzNQ^@yx319IZwqej1Ht_wY^`@V+DcwS7Q?rujQ}kym?_n-mR=HZYnx5N%x!TZ#
z{vdk%END}$3Ug#SL?7h>JU0gUeZT*@IN1RQgSrmM=0A+gF`DDBgJ9cvQdCEBbLEhh
zKY)1;p+!vvcecWNUe%Cs5`A+?mj1@Ge(eE$89V4;8b?0{{n@X>;0fu+upU7(U6P@&
z@fm7LUz2ZIhVtz*6iA=mj!x;Cc$Hr61!*b`XR}sx=iw`B+AkGf1(~RFsraf=_3i;W
zQeW~(k#H?G=u)?n^`U!`_Ag-L;1a&VooGEH%Ieqs<<#(AlJ<Q}(xWlp7w}B!naT8G
z)3Z4wSy@ezwSs)FCwO5ykKd7lITe7{xIwa>hm%nR%Ul}`XEQ8K3HWgiv`f?F>#3ST
zro0uo##hf%b?q44;-gdbz@Ve*eTuH`hEJLS=JHEX{U5<n<Pv<B$EovkD_f#goinVO
z7GYJXRdHJ1FHRe7^StJsvfOmM4P;6_hL)45M550AM3!@PqITi&coLVSPZ>#Uj+3YA
znWTX|$c#@*R&Dxq4qs1EZZ6&One5P^K@T}fPZxa-TX^n7FE`ZexS>sd@f=Le*2<^S
zhoREd_wcKKX6tl%jzUJ};0w!>^AYysFR(Ra(6zLIpP{mJ<CH1TnAr0A{YrT`ep5j=
zw^h)WvUrQvu{p^;^gg;fE}USi0l!OqRaUXimAU`N`$9(hW^ww9+2lUb7O%`yvf;a{
zs5d>vL-ESkWr1m`(_`Gas*=DrbH?H=p2JQtojCF6R<U?=MsLT{TpQi@A2wkp(6P#k
z&f`{r>fj%!s)Ygc1%quud=!ykQKYv;rEhyHZIibaUiHwq@}6qC+Ed;7aIe}o0Pk~{
zT<n51J0@6evxDWiHCQ8G2CJ=Ch>mUyQB9{%^np-S?jNMPrGm)P1u6X}cAwCx#-x@~
z(!f$`QX)b#xhEYn3XC%?LYd{cCkNjgjP)do>7f$No~rkqH~9@8y^r(NZXaLu{Ktn*
zbRS(mWkHW6BQ+eHGt#07JpR-NuBNs{|H9K8L6@2dt~rPnbr`Hx%@L1|EqJD=MXn_*
z;3tcwn=Kl@%}al+^U@Xgnp)tSJ#)~ec>Ds!N#lL5mU`)63-l(=8P30)GGLYmobKS5
z?qHWjUD2qzz}xWlT<}dnn^^4u+x(0c^WUF8+#y!c_)dm%&Vq0D_2&BiORSv0H{sx$
zGvJ$T;G3o3n@->x7w}Ck_(tHH4d9#7;G075jTwCN7JTCYz9|L1dHz3qlLo#y1HS17
zzWEV+a|nDh7<@AgeDeqR=I?zxf^Tx+aGD>3(>WfmB{%T03J(su7THx3==e*}O3pow
zS1<IZ0q{K65;W{(f_y$DXbBos5cp=<KU^ax#w&0dJ$~e+;xfvpNeuiA_{IR={Da=M
z01d7jd{XfEIE_TlTk|<qQ;)}L_l{WAA0DfAjrn)FYt^EJQgQ*O?2K}j)ze8Q4?C!;
zv#lO`+3HhSTP23Ms$~&3m7VLV*Na?reU+<xz&DS<H;p#=(OnYCj&_(<a*mb>)A`IW
z-2&gZfNzrFXik7{PV;kmxLugaw<S;I7J{cEPz%f9NnhZt%U>N-XS#zf^t9KnjqKT4
zvDcgHcIva=R&H*#e6MDFFlM!VXjY3iX8ruatkR*jI<&z-`|B51%T~qJt*E2=Iyvfa
zB}Z*K@2Fi3oRoCOPfs4xt8_I)MV^P~a)nSWsTHd4+rZg$4pq$wp$fhM#`-5zub+i#
z!Mjj>{2HoX-NV$z5~e%;VG0QWvz36`iGhns#4GK|_i2PF8$B+xPne$X4%5%*wl&X_
zW4~5H(<_%`i#k$k`&l(N1m4h}*Fp5nCdbRD!%y+@a!XLTPUJ9lB<Mq{vKlj}oSx!0
z95$VM)jiyU!prXl<GgT8k;mWoC;#Jh7ANcDB6=j3lDSw5zr8J4_4Xxe9rvnjA0+G0
zD=^U)HUQ!E=T}bA3bN_7kEAFS-)6O<shViVYq%5H75tO;$I%7SG(#KD8|v_y4k*#@
zh96^L1+IM+>FMX~k_Mie`GyWV8>(N5OwE%lxx6Ni#qLngpV7N+(R+(uHL(S`<3;4K
z$mNWVAy0Lb&VDCDOG+EsjZgJ>3Hqke3`ODrnt;!v9y<VWZP@z6V;naEz5%}d<8ro5
z0_d|NH`H`lz9t{!KJ-DpoNe*MfFF(}qeoRmKWfVU4jfLsp>!`zfj?OU*Md&vv7<oc
zz$OpCDA{0~%3u}?*k%1acpW&Jkzk#xoLJ6aaLqs9pGIg>PvC84^5@%~0@L6r9=nIl
zG4N7vxSNVg3e*5!@rLnqi45lLo_xMG<f$6K<5c1A%!K2y^7#V5QD|xR-{<SbKj0Yr
z51n@Nwc{<$<UacIP<s8k@HXDhE#>hZ$LFiPKYhkFc&YJJ??#hbx1AjZyf@vr5AWWU
zu9N!saB|qAvgWB2I>aG3E*mglkJ;=_cgK5y&onw7Y*U;p%|mp)h4e1wv2Rz3y{V!(
zvOzNpoXI`^xA>#n@q**08pk!V(hPVzc(pb$((iW-{l3#s#27;>zJn_O!%p3srR|fm
zRF!OA^tDVi_=9}ZV0^^Y*mw2GR6acg-|o&pJIT=OdUO`l%}~qV+27ty#+&Vf3$4=C
z-~^p{gVS`rCOwu>X_&44Py6YWNUj^6BGWxpS+7(0ze|B%W7}yf8j}ybr#H*#+Wc~w
zaF+gzO=UH;BYD_KW#zGjoX2N!9@CSwi06p46h50@(WwUD%`K9wp3jryc{E9#?<6U6
zWs(Xn(DRVT^TUU}Ht<X_u+5HN*y@C{sTM&#5wDHymQ;ORpQ`d?xmx{^s`cIQE#rZ4
zhF9|5#xCF_^5L848@@~?^A*>P*H-l*NPmT%u+QOEJ$o9XvX^7j@Mo)P1mVR&@4wQc
zjE>?7FbyKBz%JG5MD`ix(qVj-4puT$p5U7m_>6nO<K%*GZgfjh%ii3tTq39Cl&+o`
z<k{iL7VXQ_H$3O^2V`kxq#?)8nR@v<n{;@eZnP##9&5-Q&h2zDvI0Dh5|Y3<zu+Y=
zLY90b%~T)gNaA@E5Sg#>zG${Fa5wAAvq@4xPV~d%udkq4Z_ujHjIu^n(#)gyi*4X<
z$WVRyp|TF+!HGFqSyfC`bb}1lFU_my+PEqjv6Y?8dvyJ|;Jsn*GNmeyEvuqGR@Ls&
zY;Tf(nmm*)<1jjhD_PlyEv<?5OX*{;K#lDRXEQud#gYT@2?pxQ;{fe6`)Y?9_q6dA
zRln!02dli*d##6R{NbUXy`Fk*=cSD%Z@t*(uj$!=dj2p-3%t4a9v7^ROYvbo2v)pp
z2%JfX@(O|#v^YpJYX+%=Pmn%74%G6bAa!XIq)Q#}78u-f-h{`g7ok`<oa;j)R5B%0
z`I|g7biaqXboM}l@l?kyo_LNe+JcVo<g$;xC;$EPM~f0ySk$MJMZYzIi-Du54QDes
z)1rKEO@ji9cFTh2$D&;&EwTmQ*nw+y9Pw8EaBr;)^VaNpUV6FDO9uX$M_`!9`Ci&Q
z*Gq-7z0`fC7kNG}ZR71oes4J#We4X!jPsgP8f<f#Qv-bC4yG9fx3h^ei?_#tZ{ol=
zBfvL7;G2v8eH{2E3T!ioa|e906?_v7zB%`6tR_swmpnIC_VeLu(5p7VZS+|gtMOo+
zpZ|<i=4Ny)&S~DhumKJT>@$u(_xXodc3j|tBEVH8$<e!lvux3>o>_I|533rUvFdUy
zd_xVuR9oXU5)LOGe3K2nxe31M3BGywKYWu3zIo6+K{LTNxBrK4W`S?)z&D;Z(X`O3
z#=lQci8Tr8pOT<W<P}G>p|6<dmR%B^H>JyHzkeAuDo%g#eRSg8=u>O(S;OD#YZj-|
zUt$$-fv<02thx<@&kE=3T~<n;=J@I6SZDQZX;wi4vpV-M>&`^8eCyb$pr@VAjc`@w
z6jxQ8<I0A$t8T1xl?8m`H_cDiZidR$4&CWdsCv7Fk#h^B_cBzidg1}D7|M=Sus*v4
zYfCzLyVCG`ZN0VUw3loxj`BNHoUJ!|WjC~!1Lr;1#;9$tlV9w#;vZY>JYrVseP(Ss
zWLEqEv;O7R{^!j4=9XF8zuKzdSvw`)wA0WJcG_;Tr?bso&q~1aSnXBA(M1OvhiG!g
z5KZeDBH#XCD*I5`Ifv@Hd#Db3hobX@k`V~e=|ymzZ$o5j=I7okR9{N)m>9|&MJRh1
zp_;#ok9iWRgJ7Q|4>+mU{Ju3zBeTLZe@qxV4B=|oB3$2$371``5_+^H0?${ZoT^8u
z-|i@-CPnFd?I?Yi8AbLpN)=8<Y3qR~^`Yaa%H^_}cd(rDCMN5rvB@e0zWEniv#&UQ
z%Y(^k^&cM(w&}nb$J^`Hu>r9?nT<B?doL&J_|s&ye97+B>tt2tV}kHAB*Xjr;n!>r
z&mK~dt=94Ew!*`grMv$6wsg63U%qf-rvCbq9>GnS=wzAfFViLHn58)>2766x$On-b
zen@7)hi*6UP32cvTF&ch<@aPeis7l@_b2k2+Xek<_8YO4$#%&;KK>*9(xJQtf1+F3
z6I>I5p8$V{buYatMI@*VHJQY9w-cEkGU6R)vVr2r203066P!orF7%rla2;q=1AM?C
zv3Ml&z!~-6S$@O=1Bc@}wt%e%xSZu+8u*s^yYN%~%~r@s@>K9Kjjt4_)D3b^;FubB
z$VhR-*~G6ic>gJWzyAd|ozn&Q{qZlu0ZjrY+3~slSjpDU`~vNtinnJpT2?=N&0WDU
zE%_YZvKw9jekc|HPZYWrU*AM<nByCE!tbCxokWY;LAMRw;zoFjJ@6KPhqu^@w|G3>
z;{JGxKYHh@=vTT(@6*?IB2N`J;UUI*lQk3%Qag6@(542a(&-mUMjY;D<BeQ79b~8e
zr(8|^f&5f@F89H?YQ!}y^*UbeIe2bzb2KX~N9Vq>tFePl_K9dxHNZHI_=?d3zuPJm
z!ADge?Xlh~LwWe3x-Nvj>4%0{2_NeQ{OF757vQ@33x2NP)0xVnyU(Ugrp~6*+3u97
z;TOqAkpusLM*8pH>1f_;A52SEU_ZK>^3&C>VY=4r!$;f;Kd%!WdGJF_T&ngyLucKU
zqVGPYsO0Aq#lJ+G!s8P0m>#Lkbb*6!=3a-p0pHl#kU33=S9$UntKmbUSCrMlD`i#r
z8@fC;uvJrxezDj@&1yrBs);NU*kLE$mL*%0$XX`Rx0s{}V4i@dNt$>iNp)k>bgVKw
zQcRj2^Fv*4JN{ufrB`rC|M4dizT!>3#d9i#PR`@_lF@_n@>AG+Pu75zaa#Ee*MON;
z*`wW$txSi8$*Q-PVzhfD_-1&F3Y@H}xZA45*JJgn3HO!V%c#_%GHgB5JE}y@+k+Q}
zuW=IG&3W)m0{AAF-j&+mn>xSXokoYg;+?Lj@$6(}(Bs*g=OjMkbL7+}u|pR@x6r}L
znfS!OJ)_vseUYV7ljvJS>kVi{7Qg|`7!Gy9U+e@n%GICea@D>foC*27$J5{v-{Qlm
zS)i=x1-i4H{wR79)>LI<cp{xpC(7$<INMC0D#)fxMY(jYs2Z&*$>(=A8a`Fl@_6<w
z$x!_|yo$_ZsQ%NN{w`Woi=)6Y=udkZR@GQAPE&TN8ZWN^E~%h*e_54<2wu$z{(pR+
zj2{Cv{cV6Qcn0dPpg<ip2dXIeX2@3`Jul*;)i1oYZKbzze)ZP!z8-R1>7loKJhivD
zmo|iYD|U;&uEzyx#GxQOtHHY4KUg282dmqnV1<x5*$@@1wa<by<F_EKMr)Y1EfBpH
zOf`Ue=MsUkSrViNnI#oAx0JlgM(E#`5t{uAByw!1k~ey)Mg<SGY3QLNKYFNYV^3vH
z=4+Ycqf?80w4lC^cK5TW8+^@e7dV_Wi_*X~w^J;dlWNgKuu$Mfu-O}LmG*(t39+ao
zT-0ga_j$dyO#R@a^1RjZp_jfn;iX}#yfkI87rd~Sihyr2r+CRS*-QS@ymWJsr#{2y
zj02<GSx{1S7nM}&!B&kd4wk8lrx+ab9z0V3wi&?b3f>vO`-*`*{sh}ZHo=Gd9X_4t
zSS>$l)umllrS?HTAO|zNUX0=^#>neZv<9Av*8G%cT|Zq~{STE^Q_iU)rPcaWX$4;{
zt%$3o;r8f=$%s~FZnUC&V>CM|MvG!%^dK%q>3K1_($T7maGB40S~X^CjE*gjL4S!+
zmEtk7O}A>e_`1Kb!rxmp(H&haH&)Yo#_AKft-^a{$k@c|F<GqvUJ1I+J$^QRoJshN
zZ{DC24}8-EeAA9Dop<P3;pDa+w#MfJfAhtIEZ7g8TD99zAL=-%_D@dqCD|+Fp1r0`
zbx;>iN0rWT(Sgpc>NdztP0o9$_dgzL_^X!&Ep=95d9&75H0!`OW*zQtR^3cHwp{E~
z7kqPPq^sh=H?Ff?b!{P!;G3D?n|D*$)wmj}MW^w2{p6=NkLf}>=&CSyox)|V`r6!8
z^NP8uZ;qpuJua?$V3c=_%zE73q(gN~N-l5GiApB@@vTXI-<wozok`U%n-pzh*04=x
zE!|^Q#7?sw?KUftkNx48nR{Ndihr?H>7#ZUe#lNCcJ{jLX0Mn2_BxNIRriKHy{Yze
z0Xpcr%6=Nr&rfUlI;YL|)4p7PeKXRZJhs0MF7%gcy8xZ;6`(W2{pB~;PitoTsl{|Z
zGOm7XveG~Ko1dod^;7?DfogjoR6*y#NB2V&@GunrRG2=%@eBvwtU=S7ycGYDeYj#W
z$!+!Iu^zfsyKuc96i#;<Z%+<aPdspy@cDoGH(bvh@w8QlRNm8gh5bm!03C>hugWRF
zo=nAgJUOG2RVzMOZ<?U}*G*D)ya>VblQk0l#tg2>yvJSy*N869lT`zJ(-?en+kwti
z@Qn}n<{SJCAHX-kzw&xb@B9q<)PClEd;2<eJS%0&mTOn-wrpPX<o<kKrat0V9t%HS
z#Eom+*DRI3l%@LTv$Wb9k1+V=O<|VKz{}hwUz1eS&>=5F!Eie#J7g(!1lw#&*^uV`
zfAUY{)VtFKw<b#s=uuf?Z)i<tLtYyU^3(9`U)bImidXsD9F3^X#@)#O+f?#L6dSM~
zzvM<R!Fh1NYxoY=0=oUlf2YHnR4dTG&DcTc4yQAOw||Al0f%f}jgGV#?P)ig-iP2^
zPO$xb7VYXHU4Y=2D%aRm;0)mSUqzp~#0~=($N<|!93uk-_tX3@v@br#H*hx|OF47s
zCZ0rYY8d^<ec1EtKt>(@sBA4hHjl3X?@dxkwn{wFw~CPwFQm5zOm^ZZnW!yv{NXL0
zI0@dSA8)tKSJ^uG>VtMr1#j_h?rb!CU~B3&Ig-O{Ppzl-c{({VbcW|G^VA5>Xcznn
zkMU#nzlVo-cdkP5-F)i8V>NQYW#Db_n$~|zH_HAT#Z1jn`FdcRkQ^=jf_8`pDq&PM
zTjF3yJN)1m*wOqQ|5XF9r7zb+bjp%@@Z3x{WW)bUH#Eex>$5a$4*TEaz`Y-3>hOU~
z^%w<b(>PO0;^~zv!q3XS49#nwp_3^Y8oehSZYEu42d1O<vO!3vUK6}7^KYf<?jCj%
zHj%Tg!A>Q7VaDwgrLBigTnWy8OmD(cICe{{g2Q6fszj{X_@Mi9zqxEdtSbGNld#0=
z#RfF|T4hy!P+6U)bD^m{omYM6(EC_U+oBTLuS!(qFNqq2_VbzjgWVgGG#8vRYh$9W
zY%ix0^k*c{ZxZDR@AEJH?&r%X;g53o2HAGVOVwKkdQ_IisowlJZKMOD6})Fn1APk2
z@tEh;pyNqu)PSzFjaL0W#;O%?H!E_j3V0Bsru$-aVRnrE`6Why)p>pAb^i=l$89V*
z6?~B2!KXiYP9AqrIc2fQ^GAG=#-`wt1>aNy-_&lIL>50;-DYvm;lYl1Blav7XOKt8
zq_c>8WMU>78GSf==}erNp;hGYZSBeGvz^=RdY1MM#GmsNZU?RLiwXX8AsI$~21aDD
zueyu=M|90Cf72CNpMHP1#OVR}=Ud}D<LCO(XMTp`*e)4LKh#1t*zo2o8dX7~_EgZ7
zNVfehRn$duCH;a2XIu>Z#dX<%`nighjjy7Fl~v?)s*3F3f8Oom9^(SpiRD&R`yLV7
zC0Z}qMC##-K>2W-<^?LGKfFzK@J0;&TsV)P@-4E!e{{-2vjGSEjdnA2gSW;^^j7~U
z4?f;QJLl6-aLH4n`*^AU178I|_{N2RS5^n<#fKodbO}~%IGoZmgY|F2U`;3ytfU7)
zikuy!^%a8DW=Wum#{{a*rU3126QE^yG5=`ouSaeH`uAIZ9f8Y#9pOdKr>AE9>#nvZ
z+|}cbySzd@R5RLBopZg_6#nJUUg%P0AH8U0k<D#yyv<;h*WPrqSTwH``6wPsgKuoX
zGE=}eGatj(oP%#V<P9(Ht?=K`Bo^}PY%){#(WB0LX)ZpSiX*@>{k;@3$V-(+d#TDK
zu+A(`JzMCh!lj-XSvyQ=U_GD7-nw^~Zo%0B>c2Wre|HSl!hXSeell3qPX){6d@!%g
z@K_gu^>?uledI*pNvahRqMj{7w4QUTZHRs=8>&*tcrf6r`lN>Ho3v1c;619kIaHI5
zhN}8;{HG^F<;!(#>Ag^mc@V1EV<Pkm8s6%~5elM{W$vyBwY(6aIxi#8!6P-Wb(D5w
zfrI9j)`Qv6I=7Yu?>*6KbTnGICwK%y`JahqHzis>+>2J7`_a1jh_2ul(fUvrjgDGc
z(~9tGG^EPY;#5eUy%m|n?_9~KSB}?H`W8BtO<;S3EzSD)dTNzX#biHinByp;gRQ=M
zVbbgb_>ByBm}Ik>2Abs%VAk(`W;NmQY)Lc!?&v!e%yQ>+D{rQw*epBF&H}SKR5oj3
zC9`HtHnV+Trw8DhqoZ6k1$<*a%T)&!yUJ}j-kbHVdVb7RJO4tT;@@aiS+Y-0Ty=k+
ztMYzvRcCP2RYzCVs_dwmAB*eTuVx*5W}|8LCRxHwT9RZ^Vy;Oa!8h&d@#_v?n*$~`
zocVnZvmVBn$xEB{;ZMGH@XgKrW|cW^R<$!`z5Z;ANzzVca88#?V4QS&xEOoYA7`&Y
z&JLQD?jWBI4vM_&K#r)ms*1nAa&aYQIqOmlXFaQjhk2E=j%;*RuM^JrbX>?bxTxt5
z_?t($X!V~i%D>~HS5B_lGz5?HHdhtt=p|D-FEmdt-HveAtRe22d&ynb@a$MOgz4aw
zFwL|HmrrE4u4acTxmLJN!08+U-xM7iu83d5HGfvP&izjA>}|Nl+m}#iKnePGOK981
zXvL0;*2K7IO%II58xbLE&j>v}9-}8S=wRj^Fb|L8h#ARBT+O~b_{QZQc5T5oe?EiH
z0pCml-xU8BZR;Z(z9~gt>6V|`m2B;z6gA{tI<0-WMuKnVfp6x3Z`{E*+2EUx;G0#0
zGIeVX{0;af6ns;T{LF_6<bgNQBM;U&`8Z2mxp(Xofrkg2a{-Pf!NpMH=Dg;%2lw=W
z=lP!OaCJ7#`MB0UX6Z54%h^ZSiFlVK2Q-l9iw)hSYbE}Qp$b1`%XT}y<fl2xz$ZBv
zU&%%8J<l}ax(~N80uH3!6Z&Bs$TI~NC^doYg97xa`n*QBBdh&0T*?@HH8bJhm%zWQ
z$3L?b|1cU;0T^f2akv`r%d&H9R>9MxgLPi=_}{KnN3c)!DYPrFjTO$PCLc3r8=jmE
zbQQzhOkKd<)O0dd<LJ5>L{CmnxSY1&n?~R#FxIaEGsWY(358d3<Kw=v5&kq^3$DO9
z9fG_0lYT<H#n<o_|I&x|w_-QE25+acLs1f52XC=0-r@*6q1(W7yH@3?%&%m8@!pjG
zK2PUz(W)Zybi)om(v@62SkK1sa2{Lc(kYm$4-vU^pyw*{7JbEAb5wxN_r7Y5c39~7
zyFq94YWz1pXRCKQ+3*+agl~`n2S^uk@Q&j#Z*h+u@XdxEjbbDG8@M_A9#ht3sc7dc
zO(XxE$*$|oE%-nNlIN+NsZpgeW%o8ir`KkvGd(x`L(n!Kq^mYrsg7OJRaiY;y^F9b
zH8hQTk5mQpW<v@uSXWD`=3Prs(K9J(J1<2&@sQ2#N6ys)p52>V#$~Gpp0mp5v{iO@
ztV*05t6z`BYRsK7I@l{gS^vUglM{8AS5_U0l+&#)Y}~x1gVmkQl~hbbbdari6eHtS
zB-g)4eeM~lsM(P^HYHLwc1Nq{)6#m`DN6sAjlzo(rSU0ox=8M$Az9M4J?R%|6sJx(
za5Q1Ok3Osba!~86WX?0;Zq~)=`tMe~SYTC;0alG`VAcL)tNNdaK}U&E-7zuR(J@97
zvgzF_8KYO+yH;ue)&VbtPANkl4}JkUWH+x()Rf31<)<de2M))+8d!{-!Rh$uE_|kg
z2Ve0XyxHSVk`Z{Hfk`SuP2kCTfOqOAX2_mx-;oCQ9jh`G-6Kog>BSjWjlStyhN?Fw
z2Z=7+YD%`&`QpQg%+>UzY`m4C4`(l(1LXCNY|YnbFErQhxeu5Pr}+w>ekpcm2a?5J
zP@XLx9)GEzGg~X@cT+_<cde)kx4}6E`&9G?C;eSX6RK2JYck_X%ToUq8RXXJEKQ5m
z{YO@{om5(_Cj01`)kp4WKC*VRXklGEH8sf#;km6{0AJ7BX$fSgqrnCx!8w5zg}C9>
zz3$CEo42YCr4z>2L#3;DsO1a~nJ;>(alWrQo%dDW`~YqA4%Fb#Ak}#uq*XuQEgl}M
zjrc@A)aCe+tGyAVlfMS3X$tw+`hjw-8=zlL_>=$h*BtI?hgS1rZ`M~e?}0h+V`{9Y
zmY#Cg-^<<AZi_qjIPUbAduZwq50xzsuIlQoO+R~U>SAxXm+?^3OCCyX?x~xeUTV_Z
zORHOeY1VkF)@5%U<a_Xb<gF#Q_}pi`H5q*Ke7m;_;csSF@#g;*ZU=l*@vxUZf@MYz
z@zQRv+h6^>l+n^teMft0+&d3F;*4M7sdMz;O#k7l5x3FI?t#5>YB?&Sfs5=Pxu|1r
zH}#$4rVVr5^nAXX$}e=&*hOv{uo6G4*<C)i?sD^V*Z5#}rH8s}Om%lU?%=_?;q`$l
zdN<WeOYkw*!pFSeZ!hiQYZ`bFP3*dt>Y|~&x#6vfcf7UhA8*xzkLuzN4h^#C0=!gc
zxP>dIMJcoRzn%@hb;?IiOZw~9kYJq{1D7^0Sb4v3=JWP~VBJ_4jLs9RRd8&rR`F-H
z1Z(rQV3p(1ds8sp?I1agp%<u(ub!0i)!a<Byvj$ZJvgY#^Juxfjn*(1t5PbG1-JMq
zVS%F($JnZtn^_I+n)EW>tR1Oltxhs44}A01AO8yYW`o5nt2a827d{r=u2vSFCdaIy
z;F?eQa4WfH?U5O7%B;Q3(S~N3HDnIhVy>Bf9VgA6>8dn%87J_~kuCUdw!5m#Emv9B
z`?3GzujY?kbzzvRO7?Y?ySuAy*1-!L<*31qw(>h?qwNQ5WPW3##*QYn4KXPw!6dsJ
zlM25x>GU#_ivD9lPdBL<{Lbz!CN=59sS0<q!%W7S?*;xQ8f=pY=DPdTR`0y*RP}(J
zo@Cf77<}V4(O%!c+x%O>L0^7!)VT?cYV6>oHaSjuGsH=$OPuH@al$X*B)ihis?g9`
z@u$H!W*2>kbD_(}Md2e|G#Gr-8+>!o*;PyUe(vzSmc7aMUCvF5o4e`VWH+4y-z+%q
zu63jE6#Wn1^a0<DEJ2rZG9AyNm#2QX{sG@C!mFG&F&vLgxJs|X=l?KVeN81a(TzSK
z*An_dZ{op)^!9*%%Fd?8VMVwm4lSV?-Ad{4yBH<lRs2ksa>kn^ZCsU1M{2T6;G42f
z=#B^9d{anw4)|uvd-k!wH^aWLg=I<6h79hTmw|P_H#1tNt3(gBd3L00*}x3F3(r(N
zUOSuh%2d?UOm#b)snW+Xl@7k~&B;=;JoY+&XCL}AJPi2e;xlm1J^ZIPve^2|(#rZ-
zU{p4w$B-Xh$u;prrjBz>j9kL6k7ugMhfJ*}ul{H-7zy5|-30i85?Q(pXK*_W{$_Eu
z@-Jp<N)bFIQDk1zay4y9t~%YyReV4CVDiXQ+n_DMk<>v8>XJn-O-(v!TEM&XAR96S
zkMSftF!RueR<P#)KB=-D@6Fx<O$N7|M|XPkAD07`*$LK}!fDUjRd`=Izi04g8uMpo
z?;wM|3GB2M{$?rIXD->R$#{lGf^GW28+9d<)e5iAcW^hA(Y7-9TvjkoAYYFY_~|3s
z)FZU0i{Ko*#fyH=SHe7U?xWdz?wPMg&GI#<D)-UJ;2$zBWgY07#ale%5`BJu<!QlE
zyv1YLY3N1=f4w~Q%E;5%An?!UTn#+S-Xgw=yZAHRo6_eew5i}+WqhLZ2Eg)pEjhFS
zIdZ6+qe))uYF+`;;BE557d0n}9MvOsrj|&N?WJp6KBu@y_BRZLtT8m;7qrD(gA7*|
z|7Y~vfJ=+g8|iidO@!;|%dT8gv*?v{%+$9RGVnrXs5sZR>-dS|Zs8|}cPNp_2Do#&
zDx9Lzsc{<JPpRsUXLfZ+icXRJ-ZcS#*p6gX%ui9pv-o1*sjY6YT652;(U<VGoW-jF
zpZ>>TEBy{u%@{|Q$8YHDlM^(U?$U+yrFNT07d#jz>Qh<0H~{Z6$*QN^7vA80;Au{2
z{q7mbJx!!Wr$(w+dZa!EN6IfeQnM2yHMBl>hdkraC6TJ}B~pJriPAAQa1H**B+oeg
zcAI<16?CQ*;G?{0)%?ZiE1VMe3OX#as@D{&mUahwRkx~4Bp7E)j7ran(G9r&eGOt1
zl?4C)l>M4d(JIShT+KLzOf92ueA1_;uqC)YQ7a>o^d9~u5qvYhQ<ARYu{$;|O((2q
zeCN|;*-9?*ScZ~IX6S2Mx)R}SCbDsH*E2&dP0#}u&<(vbQ$3sV|Fb7cZK@brb^+a&
zEr)fpq;_NJX|T^x2s)(=7^e`QmcugkT5It$3D#Z$2Jgbpu5}7Nv<v)fvr%%Wd3nXJ
zqQBdYy~_6NHvCaR7tpE7R;j3zEfuvaqLOBHsHCdvD`|6)N-_@dTJRTMa<U7LH~QfR
z1}|*zl}(<HYWex-LV%Cj{9@6Vh8B8EEsCgU(JgqI134Defw$RT#-bp+3GKK}PBdFI
z2#%!N9&e4GPS;&8IGkc0DuhF+^P7h{-0;z*D!#h9kKM|_ewtU?U3aIt(@pD+uI#SS
zXWUiePp<QQ+_WauMaK>}t7)FIHf(lM@CYY$xaSXl=dZyJ{B%3hMc&t)_2)xpopEwe
zn+b44-X8k>kUMz>cYU|OU2iw?_I-C1Ja$*{Iv#3N0Kd}0OQY_(saJ6i<?Qv)a(ODF
zo12z(cau|3H|^kj;GE*D;{EY=!6oQW^S!-wwwyOPAa6Q|z2y?*t=-_8Kfn_`;A?(^
zyRpB*=X(T}0@F-+!S~It+v2Nk#@n`yzZyZ!KN>0ZZyUEayfoGpd2JkO`PQ&>Dx$(+
zMbz=HB4~g`)cr^?Rg;b8aPCa9k?j&2Rk&dzo5wb22qx{|<olWEz%Xm%bX&Ff+g9^W
z+3E}@?VPQSaSATkY7o~@|7W&pz`4s=;zFj!-Cki{_S(zI^0QX}x`<bRy%q)8YhAFt
z)`s9A!T+&wUvZs(;jZnyJQeV_r`~b89QM=!&N<FpP6(&>eoyZ4Jaw^}r@H2PYWgh?
zeZA<Rvu8cDhEVU1ZTWv}@21^p;9Pz-_JM(C<I8HFVbR7BbVm<#)DM5zsv4Y4*Ow+O
zy=GDz9L{gaW}ZV}f>LIDU1s@#X|8jQa1L^2fOFP@dwxzs^MNO6$LW@B*7!{HAa_1?
zfl1FN7gOHAVmi;sT5F@Gp^i!h<M@Jaj&sd_2)=oJ$5lb=*hO{m*WIhG`aaZ2fB)d9
z%w~=nTh38mTy6CVToX3eMnCX)n$I&Oz@!PKOqx^1q%~<K?aVW&?;w)`e>Uk$3zNdY
zH-o`9mH{T!8)ecrGfdh#%cQe!Y&H8wJ6*qG%WIRZE;-ndv#`@t!(Nj+*z3S3@>XB$
zRkgB%=5*!npW>(kW+zQ)?4)KxoaDdENtTmNTKUdN-SJWvfN!dUZ>pGFWCh>U{+{Q+
z5EuI1TvYEJ9FE0R86#cwE&R<P@J)Jxn?k`ilP9?;Y^=L3piiBlo6x#7OmB{dDdZF0
z>CkZY$ME8mWLGOJTxa`-tL-TI2FJ5UfkstqX}FxgHyK~SFi-e12g0=(-KsU%XXTV|
z%>d(ES`@BX<x8N^!tK<+ALR`foW|zLbvjVNH?_ex8}H$_`5(UN`GPHT{F<%7H`6{O
z>z_~m*Nu8T7;S(qkBcq1uRhOip>vu%1Jg9Vbei^_r{C;Wn)ZCpMh|+#0Psx(TX+~h
zG^KF(86!(YR@1BVItzZ04#9_T<Yb6{x(GirBuj^?qCxe8Qy+krryl)-a6*3g2Yv%9
zx&M`^&W|(Ilq}93uA9Mla9TackpJRLJvnZ$l|^q(CpesgT#LE)9NIKT1FX4<tCFih
zf3Yj*m!}f+-X;EqPGyhhB?RA05*kw_{Kbt5$o|5`{DKc=Oo1*;FHpdu0y+%f-v0#q
zfL9K|$;2JRpL3Y3)ls}Na5Zf>jd;5r?`y{I^#|K*-dP~GEqIH6$D<5)W4DlB!wJ=z
zfW9>pUcV2!Qytk6Z;tL&hrhQx*NIembii9Fer!YVwSA)3Z#>@OS!eUneel|l0sCVX
z-NnQ6mDi0-)%W@IV3V&Zi}5HhUnR`!LO;z@>2vss@fI&$KzH+qJXP(GrzUuB8YRN(
zc;_knHC-ok-gtodocpsg)gV_DQt=jB@GHH_QPS}o1(O+W)rYP2@;Q3zMmGEcJ&iN7
zb)y;Gjd9r;`heXpa#LGd;OkC7n<_3kunoL6_`W+D<d@M7kFW<o7i7_FJXUWrH3N)$
z?>9CB;Pxt(!w*sf?+w@Aev>lP74OZ7H~5Ijf?EpG)xH#dF7`8@>|zVoOlEH%+kW`k
zi@THU+mA=LA-yz#T+d%6X$QK@+4k%cytnG#$5wR(-|WDf*!zH0)i;4}PFdBlJN+FE
z%P4+Sf}&3*Xg!(HO6XMX#S&D1h?Ncn`pHA!ySJ8BM#V_k1j5^ZZ{E_;aJevo?wJU+
zu!~fnH+>E4$@He5>Ry*fO&S%cy@g;NYm{C-k3#E;R)vc6h!nwJg9hc$+p64e@zUTK
z2!gZmMR)P~4ZqDO@K;-_I-yN9ITj-q{(c{{sQ<dr6U)aa!ZAi;ZbZ|k7EPCTv=Zo&
zeGI-yu`9>6UpZx<m-Q$`M%<djYfzG2%uLqJUCHbfq$w$de@FIp2X4qvJiCcWVHx_R
zGn!P1bQ#UkWxa}LJ10X06WF4|D>s4X+bEuov*=vta}w??L5jvx72B7dqfdCQ#^tDt
zN3Keb&Q*6b&XJ??<l>jFF=p-ogg1@nOf`7Qzg^2qHOp(>^77it7Wb-#>~629p!;v>
z)61!-60<96Cp+W&*-Ty%T~Y6A8~S1g)*VIesTJNCbeD&48*{dJ>o7PbbDc$Vhgh_^
zFFxDv$a>eZs2J}bl|@$!r)N6el_<Qp7K>sW=#jCpD87hABmVT(tFhkd+5vCz1Nd<8
zh5tAYb-&}I((%6XSnjK9?c6jv&Q0AnIw^aqlOjhtY5Z`!7vr5&XOE+@8#~Gkec)EC
zqwINYf7jPZ#uq1fK5$Y2nCH9Q;F?R$8h77WTRu4BZ+Fq_F)qqo?gF3ZuKlCj**|nw
zn*;9h1UFgHtMWR6b*{Q;rP)p2_HxG;=D|Luo9w|czKdLR^O%d=&$!@&;2QDRMKQ&>
z=Wpz$!WM2Cick2PBHpT9nttL4xSK$@9PrJL2fehoua^c-@letk4|+g6lz-1dKY?v-
za`M49S2#`$zZ!#lZW}kfZX3qOTgK61w~dt^w~bvfw~h5lw~fx3w~f919~j5mJvNp+
zerW7{@X%;@<Dn71{E;zo*CXT7{zu0C4lj*|on9I(ySy}F4-^^?eP0>dJ3cpJ9bOn6
zP81rmPZt^$&leg6mkNy$R|}0rHwukyHw%sGX9|s1MP3?@gI*ffIE}v+8pjV88q4?`
zTY_I1S3_PID??uyXTx3@wcorp3T@vQl}^D!yahXWIdUK2sI!9|^>~=0R*ZAhg^7+@
zJHt``%yv}BRwr%R<D~uvoYW=KO()<9wxzjgE!WZ;oRj06$;LX%?BcA>ADzfp6xY`*
z2l!iC`u1%V=wYUp9$W&hS$^E4OVdo6KgFaoQD$=M=sLde9{ia~PH;IqHV5C71K%9t
zoZ#H&{1s<b-wd-V$AEwKo78BYNh`(|)1TiKV|S&f&T|4Pm^3onQIn>*s`@{!a^UPA
z?5)aQ>1y2Usy4w+I%-=~wcUzpN>Wj_wu|C}DyFG-iYc_ejpmNB(R<EYxSTO2lcxEb
zv^mD4(U~T>qEl5vr#ju-q{r<{`Vo9nZJ0@|z&AxB&D!#fS$~f+>q%Q%1#Glc%d56(
zXCh<O#ZDb|*{QaJz1m0EqXpUPRC@<mCOPQ*WJi_y;;6b!oV0zYllJl)IC|Jg$DcW=
zvzxQ#`Zz1^v$I}%xhl@zRU49Awa;+X?Fz2yP{oy;rmIfjsh)ipPxT`-xj;9K?8kF(
zjEBAt;Q2`g;7y9J){OI2$Tfd0d>p2E?&0bk2z~+QWE2h8^h0Fhhlgtu-s5$gos+^f
zZWmhBKjB(&BV6%dnX@y(wF7K(3=M1a%y5PAz8!El&&JUwXj@YEJ+11TnWS>d@G7oI
z*6-XCpSel*82F|K{7qHx&2TiT({GY>4UF>|j5EAwipJQ}!3u}doxa$gR;KD7u0eLe
zX)3onO@q<?dy((0N3V%XRlFAPEuN+<6*^_9;#V}Q7Fp^!ogL2yybo-%`F@tnXV9p&
z!RZXo(v?bCy4#AbelSkCYIMJq#}}zg6~GBCSd*##*D`gxNS3OX&QjR{8CnAee~Aoi
z92&{nPg0MD**dl<TMb-tG^!rk0w>vn#6xflJ!XkB9@_(X`fUSP;Zwf$TfiK#V3-{I
zzTk{St-&8X(TKnxKTRUnJqN6@6h3AhJj*7q5E$pL-DpSq(Uib4jp1rKaQgE0MBcX%
z{PP5CV{E0LZv*>PYshM$b3K@g<}{VP%`tF81L0`+JMR97hV?z39JnKE9)EWtII9#}
zRNnkuW-t&wp*Xz7S*O^V+6j(Xi5@jQU)6`84|HOWp#hpzLB9OU<g2qU+)h#Q;Sbqf
zIGLwaoAPvI4qOeHmlZ#-DO4qoOfS7S^qNFJAq##WSDwGoTiiRB{U<!d3Ayrff!}(R
zqiT3>8qCYluRU^9JU>T~{5yJ`$<{f%+h-bN>*rYZ;GcnQX4B`_P{OXO#dd6PoH4Xx
zCLZJZ@M!e;WurmX!ap^sEc)Ujy0Q0Vs@k+nx`Z<|F^YX~w5fvq>}Zb2&`dOwhr83Y
z@<*_ZU%J99;G6H*&-5WPWlGcO-Kn~`mF^}u#Gg0vvsf)zZurT32PA3J5xQc3DkrCG
z`V!jHB^qeeQBF;8#rZXOc#m6kx-xmym2qlch^KH5xeC081(gysgbkT`Xi<+FL~FoM
zI%wNP%8h=_1_6<(>J>?;Q>2=vM#yGzgf6X(&|PyR`<9V1gKs``jFbl-v+PBrw!V$j
zw6#$>{xM2d(1dF*wJNfLRsI%qGDj=@1z@hmR_&S3*U}#^aSN*gv#jjBS@jSv<%~&q
zExY2isUM@laxtp#C7QmLXcb!-t&;Fb9V^kT555^kZv0NQa{Sw}{dhW2eO@JMW<PQR
z%aiqf2;GBVm|1wU`%KADH$S)>aAJwZ>_VHsJ5$NnN2aS1JnKU?FMjTTpL<~@d(>#g
zQ`tJLNA7Ju`eI2W`Zdz5YIIZK!SQRHqrPyVXNTmfO#xkcQ|VLUIh5s-uZC%C!t$K4
z^7DP`GLMzZYv+XW+VPJ4?V1(Xy{Vulrz)tXYei-5CQE*nz1qsz`iQr#-dr->>v(OO
z;iKt^-l}frt)pDOLx1wt{9+dQ{AiIwFN;>TLcajt+^cEP0=S#+Qt{Vt9LrfWKf<DO
zo^;=Fe|`mQ6Y<qsjXI)-*m`RLogR5L+~LUGwUpP^gI9gDBFa}w;b#tIxasp_C$;>;
zNs0BGlxJ~LvaOTWU2s&LeU8do?x=~Q9aS#DQI~xk_4~h$`su2Z4rdot+_j=)k(@RC
zoU@YeIcwB=XRUQ{QH5bH>c7lIHg<Rm>bUFt2zPB>hrYAZT}j*cI#S)0d)-xgY@M~U
zcX34zLc99WO|xf%X%4z*;t3bM<0PGQ(F%_J2N&G~<5X<qrn+zuU-1wp;vpU|-AgmZ
zgK5yh+&n$ldxHxa>!GmI9_oM7LvO%1%gLsn0^i)@l-lB<0S&$y3tQhZ)`M@-z&CaG
z-ZJ8j+%kgC-ZE^j@cxJQjHn;~G1}C;ZzR^bZ=9)n-#B^Up|Ol}Vc8?&@9mF_8ep6n
zCm$I{(jOa7raU#qu6t^<1*6!5Rh}L%G$w#=j-4ws7F;Ye=3Oo{#({BWfN}b8UL67R
z94#~kbFT9C-Ghb3+5Lq^UGU9t@J%ZCW<Td^%q!#bhu20g_czAPW^auVaW*OwZ=>6s
zatStaDr=*uiSSC9Hgc_Lqe{bU==ru$v*9-K^D)VXbMpzm2j7vt&qfJTZRGo_4gE=Q
zAG|&CEZh{lNBt@$jjL|b^c5z3{sQj-e^YLUNyE)1_4?UHr#J=2ZS>708?Ak3QgH`-
zH)gXw!p~gxLW2U=9O2yMI9t%K;Co(xiPrNuJRTJV!xmMsl0}uXxrlZjEdq8eqP-t&
zRWhBt{vY)0ZLxr1EvmK2RTl%DWb>tnGQSp~$Gs@`T17RkLNV0`-+26Fqlf`EavW@<
z`KxX8^}LOC!=d>Gnv@EMQxkmC0erK*wn-(MnREku^QkjloZIj=;IZ~eW?gH5muNis
zt{-gmWUZ}|z&Jar+R<BNr-tC0df=PO;G5%v?6t6)gPevt=rH;5ipL!3zH!piF;1!i
zzG-{hNgGVgn(gkaJD;4@-osTN!8d<{Z<dmS|2Y}&P=Tw)gKsW^Z^nRcMs9W0+=s5(
zRtt`Dpod!HdvA1y&VvEy&SAb<HQrZU@WTyh>MzqrGLZi4b9tdpy<xNB0G!Sw^76yb
zrAE-1{7bkB8lho*AFhDT;qo3Cu5xHsr@=irJf4Hcsjx6yhnI(gW5SiXri5nOlvG0>
ztCFje!&poP0}iLl-()}T(q#s|aRJ}d1>f`q-(-Ss%A-}ap{M@MXZn!gby|@ZZz%Q*
z!8aod$dWu{ulaSFzMoAu@;W?>`I(y6AH51+<2J5utH^6?^+1>6x_63u`)y#H4QNx9
zF2mu#k^hX}<}CcqTJjyIBDfxckGAAyl4qe?8NWvTTA6Y~r+Ts*or-(Ue|u%9%9ae3
zX_%=qU%0NqD;z9Ox9)HB>2U9Pu}+TW9pLrZKUZ1vbG82+JLIkD_{E2F7w_&4mje1F
zxDQQ7H~I#Rs0mw`ozbNF(@zW*ae~7sy?~yY72p%Nn8%yw{y?kxxC=iHSf)2z%>vG0
z-v0M5yuo0bVPKo*8{v1rM4^1#OFsU^Ozv~ROQS}x`Optvb2t8;9}2Xm0ej5+y_>iX
zPEX*^hl7_q;EIZ&Y4P<vxIw?oF?6VH>@F<JSFtJZP6P7Qp*=pLdhCbiu-6ciubp1_
zZ$9U#-aR^sk5Vx72Rg!xJo^8^HqG-?f$NfMB;1W1n_PFnHgH+RXXff%w_F{>Tl_;b
z{Ecm{hU2}NxIIUmrsKisoTJIf>|I=?dkAlF+^}qwCp+~zfQ{7M(w@%r4b_%><LHz&
zNv(GqaES(35grHJI%sK@j`iaII5JD;Zf9!T4t^HKW-6-yFP0M=&hHs|L)T4}tPGuT
z$WZu7wxz%~Ux%jYE;(?wD*Q}&rl|$^#*KVZr`0Li{D=;kUy|9*Ci6EpNy(iOm4e1<
zbY$n44vhuPtg7N?RSx)OZ569_j)c3pidPk1N5^-uIzg}e--8lVyASyaw(<UOi&hUO
zJd0!!2Mvi-s~^BG9ZM-Wx0L*AmC~aDrRX*)rE`rU^mR5EXD3{aBl*P|k@~hnq-sBm
zRR3pe1iy*|t3;{e=_oa$gY$j!(u(BsXW)JO{BN{Yhgubd*QWk#^s_!-oiy?wV4ux}
zG1|E-Mjh}JH)$B7p{_Ax3faKh7Omq$qGew-T1$V1zbPNBip}CwWqlbH`C3L5*l*Z-
zB~i^DCE|4h`=zJh@lMyrHEcG2q*wJ^tQt7R%O{R*z+2@s7Z0*`TC)CVl%~2q_?s@K
zD*=tlZ8SN@SpI*8Wy!8O-hQ;+J#X;$vzyW~5sh~;dAXNxqV(Z(%>fIeZEoV{%4R<~
zJx@9c;^=GQInxj9{?{4$n{vym`RMX$dZfHA-zl$ubMY8=ub^rFl-EDc%d1$IY$YEr
zsiY_$4Rp8Y4)?H^(2<rWd#SYzqh$`>di1Tgj-WGr1lKg~jyBbr-AV9G?HU$^#j#<4
z9yJcnO~+V^9-uiSvbi)FO={t5F9rSSC7)e*8TZmTa?VYS%DAh-Bs8e*a2cb)0$b64
zs*!;y=d7P5JE=x%CzVNZQu~LFx_-k^H|YOp4!)@izF7*ssTJ=?hq|K{b#qji!C#Fp
z7mLVqVNpGZEvADGHd^9^UQ>_k%p+S(<NkJ816PfD=q86eG_RlBl`#V@ewDi{E8G>@
z!%fBrXZ3k$qY`KVfqy!vZyPr~o8=<QaTkpP+q~tJ1>3CVR0iL;nO)V;m3!kXSB=eg
zRdBGIUZlHe7(Rx+6Wle>)q{R~4_&J5Ax|{0x^OkmII%}PxDWFn|L38S_ih^Im$!^V
zt!^0(ZEqR3d)zW^O}S-+ow#M}KYz=ZQvV;LOU`}c<2Uz>=GE^T>ucOMV(Q#C3hUiB
z?wx&T{0hFQvF(wO0=~HbzFAT6vEcx|xwQ7_|L0&5!8aGcH(kIt7r-~`!8a4ZH$%WT
z1Hm?3_7oZy`7@2dHs5i!gKu_&Z`$rFG+J&cG^&mP7xgMMDmN%JN?2bRYja*1Q>wf&
zwuHSge8b-uU*Enmg1@{n&e*>;Vmw|Oqf5RvY@=Tr*H*qZRup|}3@i55xWQxHPw$L<
z``#M^{`z1v+5W+Jw(Wzl7kn^sH2T_j8)boSCc}Strkk`FeA60ya{^7LHQ2_7)8?&>
z{sc=D1KZr}YolSmfS<q<EAQB7p^Zu36@vo-3#|g*>;%^w<lN+R07tkwgKg#%RpzXs
z>iVIG7Dp7()TJ+t)1N;XxqlVY`~Ai6$Dma$fQMOP!_R|_`meN6Fy|@Q=HRCy`r~sE
z^>!$#C5ww{Rrz9S-ME+n`1-f^;m^0S(PF-jb?AqEDRIcSTTJQkCT#%Uln37gfp2!L
zF{$^TCOMozYkO-_1AIe!vhfDBAU6fRv9qz&hLN^v55}2L#ZJAZ+Ns8NJB{+S*MfX|
z?fK3@6Mu40uMrNaz3KmSDt9n{UGPopawlDX?xcNYXEpM4CL8IjlKg!31>YP6-!z~v
zxuVte|FQJeaZ#_|_x4aCHGrg20yDG<qSAYj#<3f_kKKuV?CwHAZ0zpB?#3hqK@379
zrF_=+_j&$!zeXB?nYr%0S6~A9#<`<0W-c{G%iYFUM=iBcoiUc@n_wV${w2BmP$7HN
zCRpJ~V{2vuSYsCT)j{-IEjmbFmX|Fm0&TH#u^lS^rDr~x*@6GDPjoSLH>22x)rVfH
zeok21)Cv2Sa$ZC3sp9^s(>f<y;MJY~t~u<4gtJbty~O#>Ehkh|IiqO@6=r6ua50Kq
zVbmP7XXsuzz2~>Kx#9j}K1Nr1V1ILZJb1=zq4q^Id|}N#^D629RP4;1b|C<^>Fi6+
zA=~7$d(boxFZJn%;rw~fJbHn5Qdjzhx|!TyJgyE#l?ESE=ZC<ocL=tsLeSBQ9&qY&
z#x<s<hUfi^nqcM(Gpj!(2uWI;|B!V)YX>8&mS;|Vvd`FHSoPsNhq}~>`-8AJEEqp8
z^Ze<}%rYnD1I`G=%eSF;?al6_ong@6`LlU^IA;C~N8%99KB$izOl`|W-6$m5Gk?aH
zz1?l-SL#F0Hs>m?G0gN^#yJX^<Z2urzYlU>bvz0m$vB(IIE^@upUYX!Bl1jrvQAsx
zt|#|=xk4SzMe@)Y?ydO#uAJQ{6UZ&wxzk!l%?`CaTV_*#Gm*QlVf0e<Chz>iy;d-p
zs0rWB@5R@cOjH*M`$G1Br_nq0ge-HJ*Fkn@ZH&b2d6CRQh=g{JNP2T3q3=fhjVbjI
ze<Bc*6@eX(=$$%AE%lZN1ja<5jN0mLV3#I+;VTspP-Sp#bA{PATf$*JfxD`<>@IZ<
zr;a3yJEkz4Iv0k_4Pm(7FAU*YVX#YKFT(|PgDvN|+n3oy7NIzM4H!6{by;1QZREjy
z2L1WQH>fYBU*D;zK#|TpBxiJE{taPHKfB9P=#@H6{nT`_YCtgBeF;Jgv;0=}3&Qf2
zK`8nfi0vEMF-6bR7V2$=hXhbpLjU!8f1IYKy&pY$(cFn9D(OKoYK65~>?!Ed0^3@(
zz{17!c(3NHW0NOxIS<+J%M+uN+)s%YJ}jr7l)I|0uJllyqnEUVSsoQ0*pkAmeq#^h
ze{jWwd#*Sh>xygS8!S=d(;zh#H&NqUM>PuPsB!Y68oS%OVEK9%l$~M6T(t|*I=JG*
z6;~`H-<Xka`jc;7Yq}xrsT-75?$|%T9j&&zV|J1|>fXD<_?0`B(&KS?tQV4JdO@>=
z7q;+wT+G=~?RHQ8oVcTcCl+XWVjlNWElzr1@N^F(_>gOUyJ3upI~G~EQ;+VB;+f<p
z`fjV;J#d@sw7wp@ec#hBvfLZfZ+fAu2mh}=&9KMc8&A@jW4=oZv}n~5o%ypfqhDRo
zhgnte{CRUHm`RP?Z%6Kl$)PKC1)ml(L-V^}&H-SYRVaQ<3&otQP+T4nhDTq+&?PDy
zjW)8=qc3y8Z$@B~0dp?-_uGe>GTj%<PzY>|6U$no^^qtPSF^8{Z>xx5_xw0&xF&F~
zxle^FzE-$Ew%PgC44rb!5MOA9*>2|O+tD1SH7)SDrv>U>TR_Lw5}O=2o3W>^n78NE
z)KHLX-tpFmI@LeL)ULiV$M!SqTRP0_7iwIMsc}s^P2KGX6AV9Zg8L0jsjV~NF2V#I
z7MNh}TocaexmV&@wEIRw{KxtGZ?z%%=NYgE*8qj@3^3t=0Yc7`Z338i_Cp^T8})E7
zy;|xms*{gb|BCsmzhY&fz`b21(l+zA+-&n(W<-?Bdfz|Nd6@>Laz^0OMGx<O>r=x`
z?aee(474W4j8UL)`(GJYtia9ZJbUlg!?1KcI0Y#&IaUcjvdslvDzZ(>KlG*W`Qy3v
z|M<q#jv7t!P1i;y=;Kc<!$!_?I3F68sDy9ucj;99S(ev)mSw!1lK)vEe|{FTZ=Yq+
z=g;!%Zk(hRXG#9>EO8x`C8x({$**-;qEEhw`j91^`+k-6<eO*Yn?B^5W#pS4oZ%cM
z-#C(Q=Is0-%AYyXtSCn`UjC4U<eOD}3Z&KG0(tjufdpq2$m2$ZV&hOKT}KznjTwcq
zwr7!KoGO++mx|>#`KG%~v5bHBOIG<5$$rlw@#hs7R3zO(izMNAkyK?A$^7VInW8F|
zixDOAxI>9F8&D#x7L|zEp%R&Rfj{dXB~qqWDvj()<wkL-+$|}UTjpi5xlx&%-c=@^
zd&*>qbEPDIsgy73Kl1O@DtUUdN@kK(4zJZ>x2+bk{?no_S{u1tIOplDja`GNUn$l`
z#cj@f$Ps-_w4hz0iP>eEu&yPSG}FRRzO5Z+LRFU3xl~i5!+DNYr8fJ`xFb^Npf+C{
zfg3c~39i8&t~%MGSt<W!=E=kf6%shDT8?a~5!2<hGLrgkMT`P2*A$4msX+1r1v=&{
zu$+Gn(@GWiLoVv6rGXS%4aA?(z}tSD-%QoQ`j-5jPitbxZC;Nx;r&_@1|K!C-$o0I
zy~tm|ymo2fX(wuo`)OmwA#zWeHeP7!PzRs`^Wi#Zx=siG{nf$3`Sgb$uZNUyUG`$=
zVkh&Bo1Ci8zy1194cEhf7(JA(*2Cayde9$ifXht`F@n3>SH}!7<)b0K7#hLA!U!kS
zM%<Grp{t-z%E1^%oVjC*pl-Sy{qNDn^pJ3V6Kl-83+kvJ88aKpgx=%^+_9OX7w0#c
z!>mx_U`?O2HCB*s{6eXVOKXHWl@0D(wt?qKTO4dcZ(~0PbkKFg+41aI@8X1idN^TI
zGbhYbI$?^I6I$pxVK&(Zob4Q2!g&t$$h$bp>3qhCIpWM5=Nu>W5jjXhh2Urvb7WO0
zk5nUXlMCwce7ZTFth3e?U#T-tvP13ygMtT+cR@s*8os<H==xxHbI$tdt$nxG501=}
zb=k=5R(36`f&#J1iG2<|g0YOxZRcMFV^&Tunpbfy^Ea3|YV7VL=fv{yb&>=1r=}q=
ztRDhnvQF3+X6LjI#$7%J@7)uG9@J|5q5s2(TAM_24hNL<vhektgAh#}!FA4dEVKAH
zO&dc`Lw0-q%kIbGP}I4F;q$66)GG?ZkU5-NaCXs*Ebx@Gkovit;b<^N*or;fUgVKb
z>TbF-+juzlO`NZ|agXG=mOMi?nYo9W;zV|9o{Yj<GR|Z&%(u(z(&VgW53h1^O?SS0
zlXIJ3zWvP!e$K;D)ERRBwTmCWnSGn9sL|o)w42E@?>K6I26HCFomfF@_KXM66Ydd(
zD>m$EFpR>dYI4g@W-Wi>Z02DkT3?7nB)wCY*3kz(oB72fBJsRSB=mwJaYDsin-ROI
z%GpDa&P>;P5qN);yQvKkJUd6=M{nl&1xBDgvkrSyh2!Z5c1@l8Unhg@sBi>?Q+sV2
zj-}<?OL3Ps@@N>QFAu}k?(~M!7vAk2dp1soqK0|?F|MKL!=LZ13-s5JZ^Ak;!<bmm
ztD3W$YwTK`%F9uZ`<y+kOX->F5`zBROD=iMj>m(+?0^c!Ozx#lQd7Eo3AM%DgOJ8u
ziR<M+dP~?b6&;9&?ffz3lpkD~#am?164(Fr#hN$GnYlt=H(5sei!WMfGspLiFYeE8
zj`SApcuZeJ|2v-WP4Pqunc*K@FZA&B!oWFRn190ys~R@J<XOxavtfVjdk=Kb@W8Y4
z^rTWBJeuCrsheG~n0ynrK#j}MYD7A!+2yUqy?$zTfRJ@6)i7-2g6)Ye@V)GU#qC`2
zmin8K<eO^+uAHa3Vij|!mgc&_$jBYR?c8Cp*c~rV^53)F(bmizcc{<F?BE8gDpv&8
zc;a=sC(4d{B7mN$RhK+4<*)~?KjdEOxCi_WdZ57w4_M8i9w<qL>gFoM&Qc+|N`>RI
z_`Psv_Nu}KsY~2&CD;Rc^Z0S~z4-g{!i=z{*u^<dp$GeVDw<=F2m2n~T4F#&OT1!V
z#fC6{tfvRxA~FbDYpBhkKJF)Hs>TH&I6Qz^Qm?7O*#iVC*)`Ufdi~L%_`rRnXHqDJ
z7=__`r!eNAhT#b{UG;WFpo)KcD-V#r{i4tznw&#^cjtNHZe}Qg()f8EGRF1@AA&Bi
zYck6YA%^rqU0~-m`Nrmp8GOh$N>_8l#G1pAnv$SB%&w`jKrr`BmuxICoP5)Od{e+#
zjTdJ(Z9ZARNW%il3eAyzpZ@5R=Ez`HZ)l=9?{ns;qF>s&wF%M(n&6%uXDKkjmuE`o
z+%&@IX+|h;GD7Q@)C{jTL?(ArEw~>zagn*L-3^%2t&f90^>8?e8U*k9m~T}dQ4guh
zonI}dn$(Hk*g9#vtVEhdm&lOUMKW%7kxV>VBq39ZB(YPWZ1c;JzB98WAu&t-L}rP`
zpIW(T_E#=KfeTu-^27hH6!Lkk%QFpJJEx72wt8s!P7h<6ac0v<iMwV>q;c03%Z#<N
z)Gd#$Q=%8=H+yY(9_I5$ijq0{F4UHAzx4LbGg-g?gv?f)5bL`M(%dXTUXR}`FZ1JN
z`TBUtwuqPJPvYe6?KnA0z8N|oOFTzr$%1kG-qvPGr|nr1o}4Ao<Qv7*ukxvFp7bT(
z^e5lccFz;_-S5(C!Viggnj@=^rHaGRR4KfjCKZQ1O98SZqFt6)Ey<CU2Dy?Po-1G5
z=ZZtmT*<rsReEv1wXZx+rYXM3pkd!+*z`i_xUNuItS^+)#|veAO`%wM6-ig}jeWZ!
zi8)^+<FksSpH;C$M3u<#ZY7dLzOh|Qt~ppD*WQ&#n0Be0i!GISvW-r8sr0TXm0<Es
zL-Nhoon^8=p-hUMDy2hCrR*f%n4GB+UGmKf&OrX}>ODpa9mp3Q)+pekseuP-P5LD@
z5#2`%<GBCHd#?$tJ({p0M{K>Ti8UFT$ROVgBj5b^sf8Nyjb)WK{wvppW~ny5bLP`G
zm7ek}E%fW8f#e;3#U`a(u3pQN1%oO?vs<-zBq%WakOJS1aZhzZ!RKuSTAf#L&i_BY
zN&cyT7WrmLi2@!K3RLp@x^1I@xo<SEr4x0<I$G4gYhv#OO_=lhdiz8Ze_v_h#|KT6
zSZd)yICbDdsbAipg^AqBwJp-(SydYebC{7sEs$5OHtu-qV8RF;^mwboj{ADpmQMZm
z6l#st)OW|!$Fc+U@gc1~bbIN+eYPI1B<kVCZ9V9<H^5)=O+D&wY)={DVk*5;4UBMa
zo)LN+q5kGQwfFW)T-%|fzh8-hY$dwTH}%Tj7!HG}yV+(8n?rm#hukxV`Qa_iad#<u
zM3_TaTx<#Zo>tgC*9xiB;p`&c=(V;+6Lx!=t)w3Blr7%7*kM{nJ8b>W4lg|%kTA0`
zZZ~m4wYC#7^qgR<pcbdj5p{pr&lSLZ6*;GzyR65Y;oLgqgkbW`dGd`0`NpQw8Alta
zkTg++HJ?-n?61bWfiCbccEz!OdA=RS$Npp&JowKAH|n|IdyN_^)75ZIRby<X8hyw-
z`&Y9+tp_s&zWSnUMGNfDVMhKgYK*z(U$Qa?r=JBg=O-91KT}UrM{g?m#*uteUo(U|
zg%I4dp$<ov{who6+@=M?rY-Zw+6H0ft{{|LqsNB*MBWpb<KHR>2Z!^_&%Km&;~?A`
z9E4qZ^jCQWL*p!U$Cb=sc@7j6&_`txhI_-pkd+(;zb4`6_$nOz%&AdfF3jP#>^P{2
z#As8p1$R_Nf!rH)<g8|36r3io+j1^vE8Hht-b7u^F8apz(<jAqX;(6if()~Y`kFfO
z%qX%>GH;{!@&nFo`tt2MN69nPCZF3Kg(UJ(*?MYkmh*NVcUfeo%u)2F58zI$3*Rq-
zdY)E%Usv{GSu@{HFN#_xdQ`q~|MZ@{Klj)RdX{?P1ZEYlrrvmFB#MSc;%vuAWcV@H
z*pa&_{Yb?BCex%a-{uy3H4jDL*J}1?PGbH|cj_ftL}0E3=R2k3o7dr3c_JLD72$9k
z8jkEB_D5SV&$y6Y@Q0k=><NSMoG|3I3*)YXS;e<Q;e9d`H|J7cY!`~ricromfCkI>
z+|`l!en!kN{=%M(!-5^u-ZazW9OpK5Rn%C72ZZ3+?_d;NWiL2)h)+U-VUQn$-505C
zjR|5lRuB$R3p?UaAX<$M#Co+qu6*%B`}eKTFs>!N3N7%rjW22sHph+N=IlRhj^k~8
zv2lto3Y+`FVR2K8WOw4+pB@-+lWcR>6L0Q&q9NH~9-r@1>v>^h4{GQ&n!t}5oCXOV
zxSHXPf-_{2{jRvo%Y7q#Q7c`sGsYFsBV6hCRHLhz8dvgFcpjz3m&<A_`K6}z&jlSr
zT-bTRyjp5>a;CV#egpHw&$(hIH8=ywF6%ShutCF}9oFs?V7nuF9%n6M-07`yNA^!Q
zoasf4PK_(~(yrJw+m+rLvQDHM%=xw%5$=ee=>h%y9&p^j8IHCGdaYKW&Pau4dc0rG
zRl&GY1?%bTb>xn$m4zD~G2?S@lm~PkdB9%B3&WZHqs>0Re*2r@2|Jo6D4S#L17E~7
zXvzC4b=WWH4{&RR`;+{TUe6!><NY}=3`V!oAo$n?V?Fo2XD)_dHuX4{MetG=>{|f7
zv5Ui=J4y><o)<ckgW20*&VBFRHq69+POVo@a`=7DMVs@!;uVgy?ddBw<9+a|Ge+KJ
z2j8&9h*sO<KjyuL7_eX9lm#BWF~c|Jvn>5)hDUttdwt6sZ--hid&>e2<eM|p(io6$
zN{cN}kZggqA1z={ZOu7L3wZxEM>(@xPqLdJkok6#cbQ`*87GBo6QOT{7rDmhbJmz=
zae8j4*<C<(*muSV#*>Zk!QF`267;&RGDJeT0qmj-n7d_w)l2m;&RQQSd3tEktv&`c
zWtMA|E}Gw}mSml3342r{9Y<Ho_pv47ewI8FQ6vvX6v^Z8B3W5gC>!xrmK^&c(<fxh
z#pTqt)MUygr_VC@?PqDW;EPPV@<Y}uilk|81>W-c@Zus3G%(S@yjeU`Kcb#>X@yJ*
zDHl(@a(S9y0PhioaBE|P@l%+O%N)6v{D0^)%##g6vL(mvvNUOUQf>sCkTzrYNX5AX
z3Ei+;#u@Tj9WO1s;^l94oIJS|C*Lo{$?l(7GTk6kl6<n{<>t>~KkKt>AN*PFy~>bB
zAJgTvc9w)1lbL2_$(A`;a&1wTd~5wfHa*D^t!p{5gM5>ABvl?AN|lB)QpHX;Ri107
zO8>eP8DEhieachBu`ESa74qc|De|vXsuXoe6{Aky<lyjc5<$KRB;N#(Z|;+CqRBUx
zIKx>_zWI5AbDXP1<dPx@H!YT~%3{eN-|QY(B2AW*NG$ng%ZCziwkVZBJ4?lmd~=aH
z<PFuO<e@UL8D1uPUzf?;9u+c-Y|}>ZM>=e-lE>$&#B{V4HcZz-@i;Ab4<SF)*2<!7
zwUQB6D|tS35^18zJSI&fwbexNa7|p>t%>`$G%=TaQ=fbjPu-8tS1pv8QL|H~jX}I#
zmS|%>uMyw1nZ>A$X52MttyZAL&cCuFsaygs=gIMXRieFHfk_7y%;!>|oYz(IO%w8s
z8~Nt=Z3TPn`MLtW%y8QFzZL9$BkPnYu&<j2`rOdK^Oc&YTce3d3pG)5mb0WgoY`=G
z^MrikP4201(n9@EEqD&pg8zKZaPDhSPpE|>j@099)<)$6ZA_@<UahGPY({aP_DToE
z9`&%oUYF-)T{Lc1AKRwW7k;olW-;5ZC0TpqbUnN|s)q|@dWiKkfJFrvXNn=U{)Y7X
z8zQEm5quYNe!~o$c^`RRAcwD*p~P;^A3vtB2aCBk`sABF1C0^0$QZ5n7^95Y48spI
zW3vnQ$N%x7hD`a6+1sVev}D%hN9I6WA>Yi5vPPM!4Sue-K{IAuS-aZdY<oM{kF<ls
z0XuyB#f<STjhQXQ?utrBmU=r<<Liibwe0L7>n!Cgr!RL_MZByx(r132nSeK)kn@nU
zpD#|>VxmGXM-?2Fs4z8Kh5pOb=(vR4x{X}nYE8YRo(uaEIivrpM&ArIb<cbrc&(-v
zRgL6Q&hRZ=(66}*qI<aD7thI>Jeyu%-fXFVAkrrXu}2{YVaI}T@e#EcpZ>=<cH|p7
zb9Rzj)5Af&xxn3(G0*v{n+78VL2!!*LYt*Q%%ovQ(OP<1d$LcO=l#Vd>|&%|qmIvw
zeY*zX3FiT4OoQ=idkAjl2o|gWb-wKHyusX-+E9$k35DZHo;&Zcb5qIZ{eBThcoKnH
z?x@1`dFFKBPRN&<nKtB(KHMjb=AMXimPL!XV_HXDF&X6pS;d@ts`_M?2b|9gKF!$=
zd1lC2@(OSN<;&H4{Q<sx*dc0+_tFbRF8aT&sa<R6pW<%nD$lm@lQ`!Y!5L29D7+$1
zJ>%TShF|-xiu*1L>NV?;o5~{5lJljM<VY0Wj>NQ6><^ENWZpOX!l$uka}Z}X|3u<-
z3$l-WBwTgb?Oa4Z_<MFs-Czg9{s??q9)YB>>=N(9*^M`MRVM7yEToS385t)r90L}G
z!>2#9i(7_clQJ9*d0|i|h2he+FuWd5)(Hq>Ry6xj`7`{wjr*w?q3Fb&g)TFt!`S0K
zZ7FkaI2v^PN?-96_DxZHlQo(@dmX{NQ}oJpXPyy#`yZ&iNjVw}HTQ{%AnqBnxtrP%
z1W#9T>GwdqzRE89m_TaM0<puwA5CxjVZ~W?By4NRT>KWyar1@Ob07Tc#ZI+V9&qN~
zrw`{E^Ct7(i#%Xi?uNu9?u9r5@ww)SLuWma7UPK#$2_s`4|g}Msd>rqf<=NS_3`f5
z@Yoe;iLMyMxlP?xSKJ|^_;5aXkvTWrMsw#x@AuEGDqMfBLi=WFoG(#9BTo$@{{3e)
zalxb&E_ed!bf&qYlplMDd=t#9oG9jOynN?|w-s*e!*<8<;qI6)-W}7Z#W~%;o&Cpd
z_~q$_l#iVCT%hN?rz>vq-$qT{(EbzWFfHBbnec$!P7gfZ;(<bbuTfi^kuuR4w@cW;
z8>7OXG8ObWN2;6Yib<Akc>KT(8kgMhf%`Ta4bE@8n_%mdrU+iz40r6jF}R)&X1`+x
z`9WXyoV9=(yCL2&^P$}sdLK3Y5y5=UE0zJc-#-X%`8%n+Nw4+@W{&%Wz=4d~zm%EA
zvw>CQo1o5|-<VQ!-Zczu_p*~SHXL2;BcOSK{ZtX`edXU>1pkI~g6Y3ARN+nwb|=qv
zz*1KS%w~Sc**WYa{$ch1yiRkESzsw=HHyycYWix1<er>c-!zBIU<;IzZ*)p5aFLqh
zX-&<M!<qIHvdqzM%u^{b$Kl`R+`Exyn6qMd#T<D@%<*l2F_v(jb4+PW{W86`>*!Hg
z&07Oww4-)6wA=`jw;N$9Sz${@BUI5#b0EqP2fuP&(?=g4#^_;hm>w2V7vs5!Gy4wp
z;Y1H}vjSbbeNSDdR+ac3uau4hD`n85-}3EDkzC-e<q@B=Ex#5@WruIF%`sQ94}Fo{
zgR^B!AM(M=OtJo(DdxvM%NoNMvb^;RITrX_BI6&+m-e^C_WTXmxZ$eU#$J}<%uAA@
z|6JCTyb`PA6zWUz<y`a+S!4N4D)!}x&D0$Env^YfIEM-Sa9)bjPfEMI6Jj}bpL7r0
zBS~EoWU%LMsh<@uO*h5M1Gjh?oJPL6O18PTOSCewWLb;PGD0&`Vl*=4eO-nmRAos2
zOBr(GXS&?#oi0s&rb%;Umc-A_l7kFFGK=~lH^?_z?&rwI13B`IeAD$H=Q!k>I-OMc
zt4NjR<eMSnn+WnvRcVSeBi|&yPm%liDRPW_)4R<#F&gqiD%bpw`Efs_*|8r|e))&!
zeEK1syBEs69Yx~KpQX*0BAHeEOT6`qWvx-M=#z8Km=()1^J1y9=IhRuNatTA;@G}a
z65p3fvmcz_{3#W;_GO~cy<9dfFPHPn{)k^<mHgG>9HTRz|NC>NG*Sya#&T}cvrZJo
zb+Vv)tt{ADD?{RIMaR2N78z-xFZrek`Nn04CJIN>vouZ<P7^g@o2@`BcSiMBYoJTD
z2FiIUYBlhOT$9Jy%^KdeKSJH|S_L*t=KN)RrC5K@ld$u7GAf};D&iGb`2X`w1^LE`
zd^6#!0y^ZIX!6amR|@<`zOgJ)AeW4@h%=mV9~5}iN&|~~a@RFN6HNzeV9ZYq91PaP
z^&Xl~j@QISa?gZ3O*FIDLU0Iu<NSYGP%HG4bEcYaS_pFDjA@HD#y!-=0P;;_6MEbS
zGyC|P4vyN_gZ~*_OiI&*d#n2BH?cn6l5gyC>SKYU9wrWDe^yg{X6fo<d_!vO(+uDf
zZHS|Z)EAc-BDK8{8ZTf!`6(mxOr^%&NQtqtlsLVU8u?F3L~z!q;cblL{kfOoj;c?*
zF&gb+M#04f80KS+<@3zZdNX_NKQJrFka-8|>4V&Fg@d>0F$lLt^egIdzBj^<Xj^QZ
zYm3?&yhhkz&p|sJT5XT(KJ2*K?TCZx9nrnO5uK<*t|IFMkZ~4ka#q8=Rp%wtA^+!u
z57gkaCExsg<OH`2C*}&N(8FE@+h!{0?ogrmQx#fR^W18qMvQ|RJJZydqTzx|<eTEh
zY79B0M)rL*%HF9F@<oj|MlJ}jWs-`k3od)RV8J>!EFSBLyPX4Yyf3|}Q<!nWS<b{!
z)Rl6EV?>q-G$PxWg&=_Q8-qM{6RqIuJh;E&RZQLK3tsjMse@b*1P`9~9|i=WgB5c-
z6wE5y5r}o{g~+)dh{Rbz5bizZQq!}H^L339c2%tgrgjX4@3~Oyq=qEYA`E5^!ttVx
zS*+BgEG6G4a+n{(9aTThO&pri{}E1IFLy?>hI4K+g*&PR)YPmb(`@E!WheQCd#Wlj
z&OOd?#vUh!kY`qNUek!zdA>Z5Y||)_J16d?QsT)!<RY_;)c5dn6f7kBkeRlO=e%er
z^No9R$Mp|iA3`sdH~Gqu89WX6zFOp|l1NO?j)cu?=HMrhWsXN8F_!&P%Oa6IITGpt
zWScgT_~Jtij!h)Ywb?(BAAwQHoZVcBz?1|s&XNcm85seGe<IM&GXkLo5xDR@94GlR
zh}sv9t8>HAt5-O>Hw(vY-EeHkWETI$FuYj9=dK}PNNpMhpEBx-uZQyf5{ixC)NI#h
zzh*kn{~+i`q(;L5=s=!abWu>SRxp};(=CKKi~B-Q=NH0kQO@CRFzazgFr0gmZLNc`
z`3N<|WYXihL7XiFV(~WSwea`obU6SCI{tXK#}7N!vD@HIOWdKhtEvez_%``4<IMxR
z&%0yia(B${?hZ{Yb`iCA$HnJvus`I6mJ8hAeBA?!Ts(1q7_;4XG1ooc6J3Jn&-SFY
zg&wJjU^jF)#>|=nuK2ZwOtYPPDKbi*`K}l@-4(ORH<3eJ;m7$*Mn7kCZR<?koC*V5
zs}ah-_arA5j2_~GkYj47&#Rfks)jpfm`UH&Xj<ut{Y%`?x4;eF<D8JljE53)M;s4#
zLXT~(D7w!X4>`#4jVpS$;k+o+6^(CEoBYfLo?P;Mi*Vza(H*u99!R%i-u+i6#2<G;
zuB$T)GMr&EU4_0A)M#zvf`7<2S5$6z^45(VOZ-_d*Y%OY3j>z0_itHKbQ;<WapvCm
zTjq_r`pr>b?Te|(7Pyi`ztwkkrVnGcg~A`Vj<73>KC8Kx0#G?15O=vR{K4PT6wXi$
zmN8#*lz<P@XrE7>Wo~r<`DTA7cCB*8w=$SoHf?rr4v0Vy=YrkYxw~y0^Ub$gBlW5k
zMl`X)d<*W`|5zezqvikW#?G8b-w0<h@96`dOFeMw3{&{Cw>Y@c3~E01DFd0)%e<18
z%z1fs)(qRdO)yo{1cB#y4ou))e<3{@gN$L&mizt|#`srl%zP&1h}AR3=@NRE=&PRL
zVvO7=+~xG*`+ZcR>t!Xrwo$TM&xmt*BjkD;(aT|onks!v*rty*`g&+Ot6D7csw8$+
zm2~y0k_FUTuH$YhH%}LBQvb;1fIm_jSs@P=m5Xv=nT&KPmZqvA=|8JbKAz8)k(<BD
zPxD-vdgP1Lw$2ti1+~2EGR3R?Yss{JA!|E4levGMNs7fYxl#T=Haxu{t#YqQP~0VP
zS#UwL1JB9x!ZV`v>a^@BJS9IIUPzelM=2SYA}{80J`$TE=gQv8-G!&+NBz^X>gY*X
z>9Sw!Uha_#ianBx-EvrCw<J7_ld`RGqF1|1{`HTOAEmpb_t0JPGd)X|kZ%mgHwyC2
z5c19UKN-^NLWWd+PZx9Y&7vP^GSDbXih{D_)PgK=8j>YP2WLrLKhA!*Yujs+A*bJ^
z%B8WK<&bXzHBx1LO^W1_XZ&BK$b+XTvW{2Z*PP#|QsrUSRC(L#n*<*JAuBHBOM2TP
zImCV5mwD9MBo>J`=QiUww+VPrB+DQDl4W_nB)(p;BpcFWu2(G8M)~4*@2525b=Idq
zdJQU)J7<5(9Pe^@*t%RMwkenKZOg@GcDa0-^+%@ds**cjHL;w#tE%a>;(4G}j1Sk!
z(e`!Hg|nJRUbWOt*Gk2#TA6i>Gn{A*6m_J|g?g7`cH9>>*1%FX4d~O)y*!J&!g)%c
zYVMcFH%-VkUwP%%Xds$g^O2gH{AZe2=9MY?>V1_dGrr1AbI#Ug<w^Tizr~GwlgvAg
z=Y9p29VY*fZ~C87U?};<l6-UT0{P~y0`;g@e@VufM#kCxSb?a;3haMYC&?`|aduIS
zyrh<WVPo#5LNw8)lO`O-YU02OO;lahL^Aa^Qv$V+$oY-cGA#_crUegbp|4PfqqT|O
zBl+er`KHp79`W8f*zi>c1FY*|`~_Vs<KOiRpZe%Nsy+^KKb6W}hF|WS+l=LX!BZa_
z=IAqHKp)ro8o+3}0ounHAbCA=6Z%khbKMA1##yL4wea1QaGS5h+Pz9Rb3gTt=Zm4_
zo6tVYYF<FE_+Dci@M?g|)Vb#!XMQ*{ArEYz_wSQAeldGIdlkK8hpe#mjurYwSYz5t
zYiwf9lyNUxoSbiq<A-gTrEQBhUUsnCW{>xS9C2r%BT`p6Qsc?o<041+mOG-yA4g2$
z4Cj|VcT%=a_|l&G8*)y^!%o}}IN|dr>W+UgLy&yau&E0FbyUG=k_z8fs?h0`3NA$|
z7*wn9(pZfkPiijPtKmO}OcSF<+x2Q}KcGg%Wi^gc7tpAa3p8R~U|F9U<wQ4h@8N-0
z4$S`Z2t-mB=5dS)L=blo#gjPK`7aPvTX-1-Vbvn~Yh3A(a^qf#dgA~5gP3Iyge=b>
zbPEc?u^vItXcUCQnn5Ti45Vh1&x;{}7`ZGEr+EJCNk7g@>U6fxqi#7}(4k6@@`jl>
zYQW`8C_;4T4Z0tWh5Fo24Wm!ofO)^~s4?a)DA$x-Q*O-QBWv90%+Acg%p0D-InEsV
zy_Zrm!~N2$ZS+8qS=`7j|HnAT4zqKUJd?|7DKDQRWEIYBZg7vaKY_j~^3YArZmQQt
zVbC(_E$4A|KaIT%W7z93h}{k3D5ti}(+T>&K4yMx2X<PS(1)TR1O1|wIFo$~FPIa1
zom_J?5`VW*6TO(cGcgil`bI(-70H}VX5Lsc`$i)Y*54x#`HK4Di|nF^i@=115!5h8
zpoZS5UoPx{uE*ZZ9CnW0rf(`P9GDr7*e=u~xibexgZnCKi#u?}<+C^pcX~4a#yJd&
z@<ZW4?Ty>SQ1;e`;;9z(rtg5GyMg6{fftRL<)0<!cvjGMHM!<r_Q^LFRB&!{hx|G*
zi5GK=l+>~Irq{nw7JrB9xK}g`qSq)85my44?;A)hQ~-J%2teU0Kg>#Qg|6dTA?HR*
zSiWt6^PJy|Ywd$zFAwDGbw?8QETw(jajwJ-$+z6lBi;@3V%^|6*$tyP!&ztIiQYXu
zsZI98(Vd>yPao4DLr+|$&Ub>H8;&uHxQ4TvdAnWV!ud@RcT?^QT&W#$#Q^fna`Mfh
zNNUZRI%AoSGtLZl#=mc!QMr>E{CqX8?pMR|G`&^l)cksCoJvun;deDozvagaBHw5=
zLge;_*ph0EW);@xf1?pbe6&H|Aoc+~wMG5wc8F~6fM#zR<Jwav3~iu-&mlYP@M53i
zO?wPv{(a?F2dHm1hSoesdQ_dTkTaa=OI0u%ug0}SoS~A(Ryw(%1^LG8m^)fG^u#-Q
z(ESdulV?^_n1-`cN8!y*CT~<$`e3e(FDBP<4tBZ)T>otehyH%dHTJ`gQ`8#U(D!&S
z03q!IF(oY!xd%8KHU6LGc=Hu9YCCq{o)g6GWIlji7~-SC@GU$HPEF{kGG(_<gQhqz
z*bcW+8nJ7$5xOsGgv2+@+Ivag(^E^7++&Ai9`ks!EqLE!w-VVVA;tpZsgE@pY>Ev&
zrs&ns6lWBs%rWOV&$R)rv}=HDX4EVUF=56r^(UX{&3I$XE;wU$%aLuUV+?O@%uFA?
zjd`yptCcvFr$iF-#U7d%BZq!f<KIfO->$@?xk^~KRid_to~hMF$YVa)0zU)9r0L^I
zL5&o&t&u1S$P_QD<ZJgT+57yD+;03wMlPz9aTb*__-6&@7Zox>Q6WZt<eSH(vh5o^
zJ!6aH%8deP)UZIb`24orB3C+*Z;aftWyreEGXFt_WF1J8`0Y<c*XyAyXm(Fpl;07B
z?rr(!)(tVLxhf%TFUe5F1@WpmD{&cTWJ2<3>0xkM{sy0tm|G_${q+TzclUxcYH>sc
z-8>|An-9s~`}^g=p#ActXpi*Rvs*$k<K^kNSlM+bRu=q-6<d>?(ynf&^n15U0$1#k
z$4__4Rr1Y3@{Q_mhA7E4&#JkXN>7*c+;k}&m@ev^G->cXP2L-1i8J|T2Kh#he6yQ;
zQ$@a+NWSqP-|U~6Dqdq!Mg4E83?G&%8+}uyta++<d{2=uvd#P_DU$ypMYgF^#h-lR
z>&yAg;-B)Ie4|61O(6NElzd~ktVntvD-!c7MbhkLkr+SuC4X{%NmA)AiKC9G(llRm
zJAap+b~%#hkRv}GbL3%EzC^V8B>^4F#VDvkaG+Ack5tN;Q`8d&RLPX1ns{})Sf*0z
z9Fm_S=c@{&>Fh!o^{7ZDQQz#7_(wdN)=C=r=G&24*&42a)~*_OZJ~kFe-*I%slffO
z3S8mb#w1h0-e(2YRVr|ve3P<H13w;V;)(`ygrSYKx5*q0Gv#-5rnuDPirufT;%Sv9
z(_0mb3Hhce`Q~V!YMH%PfsjPbX^tw;gY%n(Xa2XJ+CskBN4_y4-&B25Ac$Yva3J5O
zcbyd4mq>zrwM+=Ek>=Ox<e;M_s#|it)KL??W^2GTP=k-n^l3S0;;xStQvYv$vxqym
z^IAyH&_YrNZOmIwkJux6#cQ<Dg?w|dyAC|FbP%^(7d7{E;ZdlI%Wn1YX=r`8C)S69
zsUD&R>S4kWJ%l+kUvVyTf?n&xj(pRQeA9+}^N4(7+}8*`cetObFv34x)Q0s|;`Kr$
zIvrDD3h#Tnd0!mWjQSh$%|@O@e3?_+YzIAlHyWV#5i^`4-?XLvru?x5W<I0NZlDDg
z4YEML3@bcU@Ue2aH8c$yA~3KK_HMGp#^dakdtwV4J4YPyb;Rl+j_5qg5&nxEVYbc@
zNssyO5=U$;b%X)6I3VBrZRkYJGV}hT=@(CMVyCGS`?r{_^UVoO+??Td!Wl`SD!iDi
z!fEcR9LYC{84Shv#d(e%wKgg>ilWpQIhH=!*?f#&t;WKw)MQ?u-uST^Gb+_MzQKk1
z5f}7o<%-Zr%(17x^}Kr^Gv?UANe!th=Q+Rob05)-I>+9Du=*T`Lp6b@M}A3m2|^A%
zHIu2cIn^`>eYuZnN1n;b4n#!-XEp`AQuw|mf#_q*&)J8wp-^VD#s(q%M-UuFhG5uQ
zX6L?Tz8rhWSu+f+;lPtWp*VRa97m1lEgHl8-%XLQd`^8XSt5%*kd@9+xaQCOQTr%F
z_9xGfL&kI761j+a<JHXg;|yn6EdAWvQ;j6!IFNB3l4mAxUSr2Al`kJ9+pOI~-ic!t
z@pjH>IJ+_A$Ie^K9TZvV=VZ=lM$$(<fS&NK%mR#JZVta@sRz3oZ2A61d|O>43<@LB
zHk~?~r;*UQ8i~b;?3>y`FBQGvJH}C4+&dDZ!l@tfBA;1M+gTR@|F04FhgyjF^oHN2
zzNl_4{osSCzX@j#l@qh|H6vjDIUEPBha+oyIDSuNFGG7~A3Brys@NC)ifnU;UbETs
zgi}B9+d2$uvP04MFuOFzvTxBhlzma5*zyvXz|QcPfj~>{rxv6M)*okX$}&OAUV<ys
zR#{U+e&cH}J~6MT$5Q4yvX3Ffmh-u%)Kj$~vuXvw>JfKSWZe7KWZyURfd}|ww>SA_
zQ!6MOS~2&tC6b@FfWNCRwv>CrR^@>&r<mIx^FO|6l0~-R48w0UJ7qh$L3@}R`ylA8
z{_TO35A5=LLLN%@Kq#+s)E|#Mz`j3YH~MYZhqzpge)=k`=X~a14@WHd+!$TPIHJ!U
z?sBUgabY2Sg$2yav~<Ro5NEW!?Tn`z)NqdHyyhtRp0k|u<eQpX)Z&wG*4S|_LpJ&|
zq!Hd`G{n>q4bg0gHFK`4*?-pv<Br>4n!PPtci3V9_sMOo9H9HSF}v5eBTlo$=1q3U
zZ)J~RN%n9Y?toU=4zSwO7{^;VGHcTbr+PVK*eMk{&*KbrmkWE0Tv2G{hO^vxJ>>l6
z7X4PA>5HEp*950VG)1DS89g7((E1a5c3%5n{;lTlWCxw@QF<MRw8U9vkVkA}f80gp
zCNg6veFra|7iRPRG?r%$<JW=M%nk>`dF-fa%6yzl!HD7SQlGm13E%wjs>&0yO>I&4
z#0CqySR;!%!H?stI7?=?WhHa>ODyr}wI%wIZ{|IqS1R{^`Wc@m7C5=v0y)&tjGAwX
zXB|w@-i28)*7O2t(?eZgg1u=b2smN_hn~E3GQlPC%}Mf2SMp6C^34rCcIGoj>@?4b
zpRA4P_cTUGu@ZYZA5XiZg!VpO)B^l>SBYtplxR|C1Z(<NCOOqgxqhwu^Swqa-`B{7
zJ2mosSdG-wRm%$U&FKeKQrEsp^dJ9`Iktaf#Oz9O)UT8UcPixEkP6vHZN+OTmDwME
z%V6qY^T!p*fo=t|`@#?Tr|i2V{>qiH6LaOIUXEO8@L3-9NRtWYpUAkePozk3UmjZA
zlb;2*B>(*tnRNHE6s24gk2V)%)x7ia@bg(2e)o)wxqMod7oU>NKBpvU=Lvb#`<S%a
zca-y=BcgMCzxdqPEuqf4BxS%(u`7?2=tr?)(>GQI1;&b=cdWPt@HRMBK6Q*0zvP`V
z=GRV%i`gmNax-Nfxn{<~bZH-#E+^yD<qdCL=A_GB?xmu#(&YQMG|?|e6DzXLl&5KO
zxNDjye9~l@eX5MMOO?4JQ{@lYX3*$V**+pwYKNxEa`H_v`NolabMbMCbT<AZ2dY1c
z^Trg3`X^PwwWu#1{Z0D4{3+w}^2L5Yk-Up3lJ2oZa`#A)blg}Z8@Cq8i(ZA|)|2||
z9);4wGGAgAeUSm-*)ncwnix$>lO4P=sAINmoG+fE^5yXKd{NTB6}tR?IXv!{3dE!N
zFX=|MIsZCGJT1S<!<L_=!uYiq+r5?&m)A0J#RuuQ?}L<Y{~*h$(fM+=R^pm#V3|4R
zHRPMQKNKj-QQ#!m=5_}6MZ9fTPCYO=$C<O2yMq-NJxT++{OHj($|QqjO0Zj|od1?9
z1IRaN4f5m=`DR1&-}FpX$&Ruran!Ds<r+0IvO$e338|G2U2DaERIMCa$l21pTDktY
zR^U-5L-@6S^dVFAs*@JB+{N81m8qYr<@Jmj5wluZ(6Cm_q*kuf)QW4zI`O9OYI08v
zb`oe}WN$JK_ftyhfa=n;&?HP7Rjah|@Fg<^b=iqUzWLBuhk0qbcoM6NL*yIJpPb(~
zGE@5B`iRSB|5QzV+~aO))=@oVTj}HKN`3mw^)ZUs2)3;Z@p%-x8a~lKHPQ%4Ka5b{
zS_!>YJWupf;>uzrG{`qy==IZ?V2o%J6HH-Vu`_!;2HiG=A+s%=V;aDU*?v~k;?%^j
z(<6bo+w?omW0pf(=KVcyXMuUOR`l9iLv@P2arZ{B*lr8;31(#du*FF7O;(U23VJ!>
z+(btlS-^c1XE^)suz&rLBTBie^5NrlUp>yd$v2amIAKu;HOL*E;5Ud~^2tt^%AHlK
z%T9={<nD@(=kuqj*cYn8(3jL%rmHZgNQKRYYTS2F<9dV|QGRNac2^^9h8pk5H#H}y
zdwigV$2T?hhq^$ymzl37u5fDaijc)_sMo~<uD;B0<IE+yO(1%7VitdwK&0|{?)Erp
z0QxeUV_pC%(wKovea&v)Agpkv{)W8MkU9;oRL*ReJ3sqVAl{D%MCWq)hxxu|`S}X?
zm>klFeyVdpDAf;!lOOW{`!O5vcrXsL)B0dE;84uG`OV?bHX<XCZ<5Fhb00*a$B#(N
zsTYMjd-hDVh~g}poDogkFge6@8s|6jIcHfx{S9Y0?wsNH?qaVu=Q9teubDxfaXrAS
zKHi=s^K2p8tmKSm3pwcg7Ul!4qvxC-dv76GX%>Ct6X~TL!R+F`)Y)|Aek+{453Q(a
zc4PicBYLe2$V}CdNchQ~sT5}0JdVV+OOdcR$Zl}*Sl{{djq`i&*fSEJ5y@WCNZ6W1
z^88A!`4WNI+_`N!9f31jBXBQ<`>6gAm?-R=vZuzlmcH=Ra40T^Lx(?y_v6B`HHteb
zJ8ExAxsQ4rhW@+55HUFnDPimnr<Qc^r%-k%htk6oir>vbakmC|^n|?(TfrSGyWpLf
z>vw~m@Ll{FPZ#R&LYPAmjA^ONRY_(y$G%`Zm`bh*55`w>>Tm9oZz9;?aW@c676u}a
z&#682sW0{mz%BNDjAMRZg9WYFr%nCM7WNL@Z-MI$?1#F?u1ddV2)*VG@2T!MZ10Zo
zrqt2=rVa&eXcR(j>Fox+1s)i|=lbvWJTU$~XF2TW%cpmI8uMiKGe_q3O*iIgQ72Za
z!e0XwT4y_=*#O>8-0V;|$re%C>~MTx106nZpZ{c!ALaBGS~%kb89s2c8hgn%Q}**d
zK*ni&iF?|IYIgooTc6{CA@MGlFs2a(JZ;EMI&-XH*28;d>AoM<2#1c^AjQ%aH&(M7
zvcMK9>guv?u`~F7LuO&yAZvjgd|TMle`k*^BOH+O$pQKF3w$VQ45xG_=yQ%V>Zl64
z=d1B>vkO*`Z$6UM67Ra9id{`QXFc$;ohS7AHGxUTrf6f-3>i7iP<zE2D<=8iZ+LU0
zh4~^cq6K=h8}t}^KhG}r!`f^9Si_7V`>O%W!si}2E&vuI{84(-AFU^_v%NL|?s4|a
zr?*F&#kO#%u)&-OHrN(rgNDDXuq1(<L7T1cxt|rXds$&aH!Jk7vP1~^MwM@gB`-L?
zxo?S&<eSsnOBp3uz;7M3#Um}CqHp?2fGMsznPRXN8OOjB-`HL7;Td;Qr%j;jL9b8?
z6K0DrXaAlt-tS{>*i6oExOZ9+WQ;C6Cl0kTMm#k(-M%Z);jt3?yD6dI<9WpJzw&sP
zCfZInLc?q$d|Xm1{Z+LxU9VPV=hw)l*EMo(ca02-s1aX2|Aefmmd(sf$&9L!sC$3p
z(D_Q)WnC#Q+bX2tyK;F^RVE|2_xQA(y4OCxCH&hj@p(`ro01FUWcg2tsQ50o4Zn)v
z*j$-aKSw5Pr^ZG1lf?dgEZwpm%C>X&B&5L|`Evc1q~~3c#8a0g>E%VK8+t)D_BtnG
zd`3o`Jta}|Pl`UdCfhGjR+%1>>#O%mh2eg=v@d~q!*Sw+UG#PDl$OCeCAlP4#y^ae
z#s9=gu5YZsD^|omRsw=zWq*fQc`_$fTJn7cy6lu2n?6XNdFj&S-X~E$`6MB`(?pB&
znm=S3C2#izrpW<=G-;XeQR+>6AkWv`mSOt0WV`MyiK|W$(YhtcTDL|2`aM~F`J=4s
zn<@{YQ)N1LXU*Dwk`k-u@~ZHe{4Rbbx7H*}x3TYKkM>8IHYip4(L?1q_M4<s{*(>e
zGevhTl6%ZO?!2f-K6NjWxQ*GeE-ph3t<R9@yfQ4Z<%2p~u3Ud7T6<rMoy8j|vU(#n
z<KN4tULOQ!)5Xg(L;elTklVM(7pa-j{}}Z}m$Kx}r!VqZ<Ezwue<{z-KbJRwuVl!7
zujJK{S9109D>-aN&PjSL2adm$=%`9Ld7xI@nra|UTLanO70eG);OI-va11Ks8h!2R
zNySp$xmZ?FLvC%xy<>8w<dkK|Pn}GuQ)bHB#+jn-kSW7lG9``koTCPL(!@AV>`n9J
zy=$Jd?3*WUeexu|SDsAzlq)q)bEWoPuJpW~OFqgKGu>SAu*wzZ4Mj4NY_%x4R-DaC
zBy~=S6n88YV`l&Ccv`|aRgKIqtC7IcT4~*-PC8mCaM4HuX1g@8k6NQkb9f!12mXT=
znyaWQ=8UOopf=p}bnut=hg<D*5JtY4y-pXCZ|kDnH(l6FtA}+@>LGoe9wL_N;o)UH
z=!Wa#$UJ>^V(6phtv=781~}c)5Rsz|QIcW^eew<Rncv+=iGKb{e2?Z{YPk|Ki<B4`
z$*%Ob#(3G7XOi2@j_zp6?pIU1?9+f<I1TWq#ta(lVQWc!@pgJ0HP6vUMa|tFW;}R%
zS|Fpl1;#0@;d0y>7uAh$I-a|!EL-I99JscjBO(K-yWv^1)i_7Q%;hX+g(J>ibA<9H
z|D8)M4s|$%x=wgazR@J#*alNq+`$PS2l92~8{b7vxVMRYRjJGhG;qe@Zp_75&rF<q
z+)sT}vByM(*Sr?is9@5N+JjbVX!N2_JVuRGThussTn#rq-bdxA(SSOgrktb4eRRPN
zPgk_1X45gi10MZ6(2hJ39T|u@9my%N<dT^I)Po0L5p@8=<_F+ZY5+p(1tN|+trYeI
z{=@#x)$0Ng?oFl{8;H!gf!Oec`6vn0M$QdDf*t2Ke4kP01JPnm5Y{A9E9n@Fe}aQy
z5l6l9J$527Z$7?P7z*aISKN?Uzat}2O1>$)6NwSIk+?$*&VBNQp$~UUk=#A?<o;+l
zU*^2!3+FCoOSv~%L$2Wr$7Bb0Qe>RHyQv)}&nU?`A9;I`%yX1%b8aU)tjIxsH<E4m
zF~|6^zGS4l>CD9$M}8VgmLWeq>p*XLXcVsdl962ab}Ra_^r-#$Lru+hb~AjW#`xj?
zyy18EMKbG=j5C+{H~ijzbc@8S5atxRvX{yv5}tn|Q1F?)@Q1uk@Y=)~&Ws2w?n4i6
zPy_;OB4AP#j%6Rh5qFLmhwH*oG%_5~q2V}T#crX3Fc{sXAADOFN=Jp^tA7~W>V@IK
zt59@~4aKk_p*ZSEUw9>R{T=`#H_{W{59q=DR6qy8^;Puww+_KjWe9ry2*%!L<QZzo
z$4=o4rwhBnn^S}He;&8%WY+K?)Vsys*(`d(nNz&!SO9*M`s2ZFKXx><LgGYvr@psD
z!sM1{@S8nL?Cn0uJ_Q|@X4ray{^yO%m08Tal9>ydaSv+F8O|aXH)fo=p|A;kL}Z-*
z+f$9W#U3f<(7ZX@7`|!taB{TASOYuyE$y({+8GmGIze_cM0A)X{<5?2Dm#hRe{RSg
z=SG+oW`mfYwis&cfPa2C;T-wKne&@I<eQKC)XWuDW5884BAAJ@`>YFMGhA?|BWIS=
z8e!XuhA^Nm_9VNUx1{p^+tmW06K!CeZv+4Sw)muBg_-mYBp$ayBxgK#*#o$(UPH`d
z|56fXJa$7I(CLu_9(HYvlxvOA!`2ZemCn$ePoMZ?HKIqmz-f*v@=e^}bl(lVHoMbT
zM$Jx;CnD$(TgH69nFURuujx(SgEt2B@`0yybF>`ljY5SFmUeA{E+1P!uVpL5-(*H-
zs2>{l@`v7i&d=ugqP?{Wm$?rMKJI|!AMBZRWe@E{TlBZF#hTGJ@QbuTO(AtP2dvO5
z)(VB3**NyKLPWF`Hj-~lsJRK||F0s&5<8w*qA-d1HSCJCePMyrYZkb%!xW*M-5ldy
zYKfaEM%b7l#ljR@$T#E3H_ga5^Uj*!S#J|$(dX^QoG=~g6uZW;$7UBh3#h%x4lstg
zsWFnNz4`k`3A0=!9t@!`%7B^kmo<=3zgF6|ua*mos^zDq5yBD-aix8&bmQxlKWn7b
zn;J=(R6}2Ljbv%oh(qIQ8F;=*y8QklW1IewhzphSL9bGluB#B6bLEovs7xxal*+_T
zCE`)Y9J`~x<i_42`Sq+o9u@qQx0T=J_>>Rw`qz7TfA_ttTK`UFRXmjrpB~FP#Y6F5
za!*#Zxg!TB-V&3T8**pNWf}eOqG-3hAP*Xxle2eE$=HD>Wa`ah@~lUqT<x=0T#^%H
z*vsA0F*;s;J&qIA-(9j=c1foPJ9)cPMi<9Q+PzqL9~mp(J!554gIIZF5GzRyV<o>e
z_f#umrSyEP_}cB1edHUvv!7(r%}=t2SKi%E5_bQSJm9RxE-y_A8l}lU2S3Vz&JW}m
zndY<hEg8=1gT^gc?Vlu{7vGd`3vbGt$2aBi+$1S{drNk_za`aaw}eY#8FnQ}y4*<;
z+mpAX@3sfhCoNfSZh0@;^*+j7j}$TWOcVcOY2vYp95p#jR!>Tkg%i@`X>^+8^m;2B
z9AC(WoTsws^iyeh;;96Vc`8%i-Iw%Jmn3ZMC0TgxlGuO0B&pNxNc5b$ay0XSTsZnj
zcD;Ng3o{?d?lw>5+R3Lf{oPX;Fyxt3H+n8HA70Aw=vVS)1{vqrE7_U-N-kTx7R{j7
z(yrjO<hUnGVq$?@H>r|D{(im6i>Z$;mY>u=&x!gX5tlMWPnjXJH8Ui>GF|2=GNi&X
zLnhzPkYw#lQ5j~+OU`hHJ7&ryr%V}PohjS&Gv(dy42gK0Axmy$Nd67(1F3O3dpbit
zAIp%kqZv|rGDF6m$&k?(Gvvb+>Yz$;B;Gz(`p)Ivu3ojQ^vV?j3P>B;l}JzT64@VL
zBCD6syH`>pAF^v@N<y8)j8<T}u;VIM6N96)pg)1$sqI?Od#r^?)EB1+cT)qj(Q~Rc
z>XUD-wqbA92>Qbp>0;?2UG6jLA%<S@Q(bl0S*44P^K`K$P8Y+|bvbKbUcoAT9QxM)
zHSq@6eUu(U=0ZJq#txIVMu;giLZ+<}LBaG)4d9MywGykF7$dd^GydKi)3?bEH}cIi
z?w?!VHO0=64Pd_54Et-TvE#XK6lXV$$Tw}znqvz2W+^j0KLlGq_qP?9fiv&r4RL+9
zEn219<HoqgXqnv@O|3Y;@pD9RH}VbV?2l$ULXYzs|3i-Wamf+e$vCQaj)*OE#7|A?
zZoHhB*~@(tbvOgbH%q5D;n89zIBce;JdHVl<Qwls&REdH8CfHo;X2b9zy9NXYL^Nd
zUZ|jDpoWPzAMblI3vGs)`SNO<;A8$N@=ZtbO?yoj1TS!*SHuN@e_b#($Q9=Y(__+}
z{*dK7S8wych3sbd{n`)7N`IWOVGsFXe{}5~fGl<ntsvjrCbwJ(rf0ZkAWr@bKpp2b
zu_szGKar2iJm*JP`a>%q07jbRpMHUOzK<VM6bO6nsS@cyRyGKRB(qnC$hrCj5KX?x
zE(_!HR5*rokA&amNR-i=v_6a3c(wE(kuS=qgUJYEE=~6+Bn_cAiW(fx7;26eQCqW;
z`r{4E7T(JDBjcPP;|wFuIFob!@b(*bQiWt2J-*G4vzr;5_dMmt1aLR?l#H}|D)R!l
z^NJnJk0(E^`6mh^BnnH&RK?EZB8w=z;2fz<1+~I?kr?ovyQ%y1iJyz)&Wk#mwe;V{
zL}K9xGEEnDHw1C6qh?=&l08)A5m=HDfkDib{pT2a!#6PFZ)ybA_KHBdA9=@$dYtlb
z>`xA7PjxtIS5kj7EF7JKn19nC9M$x_EW93u-Rr|(GK9Wx?=WnuWq$1=?x!||Vtc<(
zym6tIsuY-Y7bsjy4Nh;M%MiiOPa#NP*3I}Z<}}u4UU5b+{$8g~Dn1yMBj`Ey31&}8
zFb38IVc!MnZbk%Q(Yru2<jgHxHxOUA8~KvL9<!Z(xKrK=H5o0jhr6l%om=8}P7B0d
z^F_LeFZQ<gW}jjcYNE(2n_OY~A2q+>F3@>LZLTS2H)NdFj-1s{rz2VJXilBZwoi^o
zXiP?kwa2_K>=xFx!?0Kz1P!#s_hcJ9nQsBDjTYR~Sz`3mhUk{wkeUBvluC9)zqQ4V
zE_O)Z{bTk3cGRw?j(#h5QDmGeht*ixk9@V(1&%kUHLF%b&&&q*RgG|Ox*0|{GKb|1
zb4<8vj$=Q~(SunnKkY64Kc?O?F6zAt_g1<)1SEzMP*e~>1=d9jcE`56ySuw}n}mf0
z7SgDwV1XjoA}C>O1W`c61dMat=R7Z-7r)Pp!+^2(%zXcAU4gOuy)1`No5J}<i#$tg
zZDon!%Pp}vuL&l3TA{JF4W@s#fyp0Rym5EH#hnhgZJ@v&?x^~7bjA~2LkvtI-!S{8
zzkwU_$TxeE-Lai{IX^ly$Jyqd=w9xLGNV?=`o+HS_|_O?>IJWLtuWwDORR6CK*2Ew
z1RbUJ=a2)=hjTB--{bW<2WTyGK*LlAY-#R*uxxwu>B$+=9y|2EOYcITrtEWT3bjc#
zxG6T6LB6?k$r3XUTcX`mOZ??LY8?4yIr+xFfw?vBEU>YVjC02Vo!P0dyULu*&N=lG
zW85CY17r9#Ge#HgrF=NM5mR;wbDsVCq#;&PlUPKSSxctTrlwe%e51<yVSLP=d^3>O
ziE88<Grp&;F5vYeJCFA{X`?n@3(d+KV^3(E1nvJRV_toi{&&8LuSbIvc<W*LLtT9L
z=JoIUUy}0Vmvo)}OQtD*$+iIv(!{DkdSuZ%>|QT=&FW-W@=y8v_lJz#^+P<~f0OT8
z-y|>Vi#YVJl`r2vN!g)~a&BFX9O7Q2V$BDU2?cV%q(I{0$v3gj#8E9@{v62_oAey{
zc_v$|_TQ4O&2NZ)pKEf-FGHeEUY4z^F3FblDN^E@ECJ8X%X`l_`LyJ;IPN+w2EUKU
zWL<i`+aH$s0}hGdy@Qgcc~BO%KOonVqa?L&lz4xLl$KW`WlT_{Bsg+*qZ7$LXQW)N
zju3@cq<mW!DY}ayWp9%xIk~GqUUKd-j%*Xf^N@VwK)(5$`&<r@amtN~#keS2G`z0K
zO!Cb`vW*u{3zciK+cZ;DmSjl(r5O^Mo+_V`Q>E<%?$ruXW$x=#37e88+n1%u-Y03Y
z!sUXTioGD+KV1+{zsnM~_p;nLcv<$%&X$#Pv&HCCw)Ekd&C}*sw!A)?Ee73g%7#-{
zrS!r@$$NT1j(5BuSsgCOe>NAG|93$aPro1uWE`yz7bGx>bEUsG<k_g3l9rJz1$Eh?
zmV8?#O}!&)50Y`z@}&2$JQ-_rPYUlnkU@hVO47!M;<N0L7_E6E3wAt`wYiVvSIr}7
zHtVrCc6lPo!B3?Br6*#Z^+XKJ3nY5S3yD7cLV~Woki*$8<hXs2WZo{JX{A)2nU#tf
zIi-qxvqbx)oFn60x8U5y{-xaGZ4Mt>Tv#g8$uNQIN=1XmoTu;lQrQz;DhJ6o`}wu!
z<elZ@8;|r-iTqq9er?}MF||ou{?<!?`&)VWr&jLiGS9I27g^7lq1o(TqC+;nHtm;W
z-Tft-t~Q1nb;hm@D$ISP&UmsamhUFpTvbKz7gf01v3H^udnM}Cu-~2C=;WJ!$1#gw
zqdNO*)luG312w^#SU5@(yQga6=1NU$U$03YrY2faV_dmL2Q8-RV*Fv|T#|44Uel##
zOcy`9vpc*@A8j2B@Ggv8GsXbt$TySj8o<6A_s9i?2z4@o2KmNsy)nL=VlT-^6VzTd
zK^FODeHZ#5XPY5$n;H5r(_ucZ{|1fb`$o7WIvupcp0AeJyNZ4(13LuXvqQ;edlcF_
zz;>VmEcm{aR_uVqjp^I)RzPnEyEzvq@brKJyDst?@R0&;pB2!mQ^3>G5#cQzu|I^J
zoBbS-PQE!s-}sAtj)*4TC_d3^MZOtI?}pb7C-^+3{@}9{E`~9OV~8_ul5fV7Z*;kz
zIx^3Ny^$_3Kf}4r9T#Z6bwO`qSENU~qT)Hdw7*?(q-QfMNNtAaKk146-3;1=ZfM@z
z9Y>iF(vanjkF7nR+QA#QN4CS!{q0dsz8S=8`zUI5e%G~!g1rx)-*R`gt1ad=v_=0s
zU%V~!!^kMIQPcL=xT!r}`gK5)*PQP(@ka*tTEn(7JMKmxIEI7U0iZ*lVE9x9qscsa
zJxW6Gezg*tIKLTGs)TyI5_)FLw`tC~47iKx&CZ6AoYipNa&rN-I4jvTwUL=O+sQ4v
z$t?T1w<5<xlV`@0b2{>b@%{)tww?Tw$JvcFpSy(LQ@e^DD>72A*?iulFw7YdhJ5nV
zoX+H?0P1&Iv42?+h9job|EPt*=NtblZ>hg|s$^#c`xp}F4?j$e%~o>FLT25JRbqKh
zvQK*@Djb!t<vc3<YbZJvhr;|OwZx}Fk+Gd#s@de4e#|ds_Dz}@=Qm%NS^SV)>Ssfc
zzn*zFBSNs!m#m`~f}l6d2TBXZrA@)uIha0r<|0m~cPj2$5W66$zZpPXv4Z)<A6cQu
zZiaCB@AKI+{lSc5@kh)nni~jzU(RnD2jbi#=J=gqe&eJ7d><Tu2)h7Gsioc~!yn_<
z_~R`z8BM-XTTUKbq{iLiO3v+m`(gJQKa?Hv!IPst2w2({6PdX?Gv5dMtbOoRuMP5p
zTH|CdPqZKAjt`GrQBK~T$hn5eW@jwY;QXe78I;ro2X&&4`-3xgwNCKhe)jxodmQHH
z^oAl^r2c6N_vJSD$sO-f=EeOsFvqX)=Fl->7s_-i)W5dEmcG{bP-u-HUmN)N+2TgJ
zEk=8g`Nz?BNKJmf%T5?<?2J=wIm6+uR^y2?J_g$0`ebX2v^2r-Ehczb#SY^kW+>#f
z%7m9@2%xvQxRE(-htbdMV1~?Q=5XO`V~edNPH_G)>v|K+b+bb8A8Wj=v_U*`@KS%+
zW8ZiOm>*VPN-}4yjh&%>g|Dw<`okMIFXG?1&0c!WJ^A{f=W3o`a|~i%ZltCsY?rl!
zt$iz;9@GkRcD14gyd{>YDPVWr0YlszuzH6*dTk=ZOtZ(m3HDgo&mJwhb8ght9&ciJ
ztF%YzPdn)Ew8Jf1JG@|)XtgEhS;uTp*~tcN%xrMyp(Q+`En&0S5~{N;5l_B(X=RDQ
zoKI!HwZN%-3na~FXH<9Ui{G;+f$w!KMjB(0AM<=0Q&-G>sH5ua+9um5&QmXZhB@!d
z16zB7>_V0q#j}lUlSQ_v<$ZG=AMy<ye8~4<HfoF>c8pPl8S~mgqwD0oN1c?P`6=P)
zKV)StGi^E7pV0BEOj`6slD!(_??_z?-otCt7plnmtAb(8RAAJ*F}#*ELg7DuWtQ$A
zdD`{2q`&?p&k}yhE9U)ch5wM(kH5*{>aS9>pZ$+bYGvxRPm=x5M;YI-M((e#mhW>v
zNX6I!NmD71(MO+&`o3q97I}|;pIm;9<%n6qExFqLhAc6<MqhS@oK3naeHLDp6}Ti}
zv1ziaWwONgI4@ty&&k5MC#bhLDz#-tWY&+vqTchcZ0&bQdOSQR%J>74XMRAw9g33V
zo>3C}I#PZgjFb-nk#eX>q`dqUAwTa#$Uz>1+z3f>jg+CB>-?7yAq%=^N?2l!yd~p2
zh<`3><QrY`jpm(V`T4F$I_`NQUtVO3p6)e?)3_$X$Tpj}lM3Pdrj1#qOkI>AhFPgH
zW^1a9_?;p(DycF+kt!#qr^+6qH1X}2M$dbiq*|p(a*q`G<Ch{ftx{yv^JG~cl`LKk
z$r4tTB%{ibM5ic8#wH}mlz)>XFEmMd86-)SYLcArNs`;$Q^b8yigfrlMG92YWJv1^
za%S=cvEFt;Zq!_mzY*7EV!N9%m3%XXY_zUETSoT0Ek|eG5k*w4c&X-zO`AN43doZT
z&U1R7&6U9gxw4|^eM#l{z|-dWL+R`NNYYw85iRn~<H1j4+?*$}&?{ehMn9KVXP(Q=
zxaZP4<Atmv-@GH=JiJvR{mD0ZCZ*Ez-%`0tz6sKNDII@vKJ&O#%084zYps{^A0I!p
zt5n*MX<n`^6)oOs^EmT#CgTM1dmqM?N@jei7+xupPnl)Xm)X!e$TvPSKT6ThdI@xS
z%btr`dG)PUOj~`C=+T@l@}xWel2iZulB36d$rf7`WX7oAf~hL1po-*i^or9%6_~Dy
zyOpY_GiQ$RHZ>HKk#p?W4XRW}7Uwq}o7CZ@(11r54crRS#LJPID4MT{8|0h0<eNTa
znm9H|8;$3&cWsX@KX-JwTi3;W=6QGI4CmKB`Y6uShth*S#jXZeGt~h1cQg0qmI3zD
zJLSmN=MwguKG|%9zCVqyCDs@>sJ|)iNsaMJQ@q||iUSsA$aUxLewGD3Y-GOGehaLQ
zW*+%{OHBQ4iIMAU5w2s0MiqAW{LvnM)(&v$>wvCT9B`OA$!jWnKWwJJpsosLgeWj!
zzk+>|3aH=ye|+=n2ieEM5#=uIX9#k{iN20lG07478y#^e(h-ZE(a-vY`brZg^y}-y
zE;lC>oG0s$Z_e8|Biz>+U3xg<(*)+_{BcI3IWAbbgMO;hyuK}P!QPK9&>6?Q!Fg8<
zBi{^b)C`;YHABb+<{*=A^cT3H?v)#6xw<1}hC7^Y(EriK1E=45Aj`ZJvNyHE?)>(+
z#Lqte`u3>($!?vp_RQ7uLHwe&<c_w;Uhj*|=Y8?rw;lhz{LsClA4a`wj~!L*;igX?
z6*(unpFe8N`MFE&>z2LD+og6yg`VdN4k1XqAA-NGq1a1)Xi>ym;P3Qf8?)cioms?z
zoXPZJKj!c-G~vAE+&pTImy>n4XL8+2%?#%<Nqd;%N5;`1&phGD;JF{p&Tp~}__cre
zoT@Eh_=n$<zk)sE3;FoWFa%Cu_VF-!$NNyT+>u(G4rCzCjNaLEkHz2rwhDbzUzD)?
zk2|WzN_@GdM5B26s1B0DHY;&-KJ`MQsWtAd#1CKQ<2Wc`s6$Sx4TU$e{pzpLGZjPc
z`&Md+XL6p?Clr0WL-E3x`kRmR^X7%%TWknEtPa7>At9*r3c)YU5M&pV{St!Vx`x`~
zzRW{(2u64{*@l^Jx7G&XZ_gl{vtf?yd!TP7ec@AqC#AwpQ$a;)AWjbn#2k92zElO^
zSr&CRyO~)%IsjXOc(|9+|IZ)!)JCmd?vIPzxj&@7$}_P8bbqx+FK_nEGp~5ZF&`}7
z?}O>Pe2_cW2M7Q9V8vY@Y&_BiSJ!%>I=K~g&u@Xh`^el&U69tCej;kd?&d1+iZh2n
z%j~d7-xkNHXZbOd-Lx+BCv#7fPyOn@<=kl;r2h1pJ}MLR(e1T9qS(W{`xCQNc&pB=
z@Us)G;PTQ6sy(dn`I<Ev+1em(mkmZ=w8e#%c38UF4)JMr$nmtt!DDtXiL%C;-PV}h
ztqFe9bKEo85Lady;iR7)#%|KX{y-CKC(rcCV?GP>9?X<xX!*euSG9Pbnz~u%EpXky
z5?6X#VtW?<tUsFIhQb=zMK*}K#Vowr_Q(i!z&c*H?2cEUg0s{WF)lc;+!YG~nqd|D
zAe6h^(3^a7k6jc0xiyE0b_@Qyx#Kl6fm*e4N8}oJXkVjV=_<Wgi4M@P<4kC;J@mKP
z!<&4was_u#%j^+6!X6g%bnWYC5BDDS%pI^t5c9PDePoNii<q<ZuqjpyX^N-eHpuH?
z!+n!Ae$Tc>MxGU1ldbSr*NT~XO%Qn3lHN&6Bn-5KXQc&q*A`G6Xo0myjZweS7$;&4
z;d|T=C2x&z`-KsGknB;2C)b>#kMy)5qRBVQ$v1^jhPbC{fOcv6m=dLr@h@~S{E;p_
zEV|HXtBbim>F-EY$A%_#GO*tdIY0HgymI|6$3A|OOSkFY&ipDWv%bhb=3m63>lg9o
zoPXry1~Dvakd&Pb;?2{aXI*rI%sSH`HCG#CdIG<`pg~l+H^}6r^h*VQm-GBtyHC`L
z9&=ZO*V2dHeU`&JKGA<vBQZa!<z_d|Z_+==n~Ch${_;$2?|LR@H&Q1Ocu&3s=gRvd
zIWjNxmKa#xkma8;rIBxj9K3c#w#~gPdwQhHpU5<+ev&L-jhNLJd`6x>I3{L|j!LJJ
zBl6_+VL8_6u>9?GNS^IFD5K*JNN<A!(tlf&><x*MNBNQBx;s*es4cG7i<G045mIy^
zLUON1NNHAtJaggR$}N(<o(S<uO_pn;Gi4MR=V`kKQkYXDj`yh#rr*0P`U!U=*>Y?6
zbqUbs&Pn~6=#Xz#k#Da5&6J}?nUbiRA!=JvCAmID#{WtYZPirqF-aALU8>a2N|neK
zY0{=O&$tw6+$BX?lW#tdZ|*iFr@T#;6<d;Jmr=5`B;UMFNs@QbNiw`=l7MV8sVY%E
zIwr|u^35}TZ`Fbn*+#ypZk#3|ju+(1*bCD1K)RT1xz6mrn=+`wP4OLlQ)XSrmf-d|
zlG!^)I*q%_`AV+bwaygJQCH-1N;-1{GsXDZby>bHM{YXY71cj?sdLJe!|QWp@xrH4
zubVFoOY$Z5RK6sUZ#-J(%ikl<WeWLb9QkHW<_n2eFOoG|OQdmri4<#<%B`WL(y*EP
zsokY=lG#C%@=7J*Ua1t5Z<dgEYFCs>A-Sga%2L_(|Kl4+@{JGq#)$f&Q|I_K&T!_C
zZ}yRI`c#%lB>84v(OY>XmC`-GOuV${IWK1Kht(IU=+Yqlj2h%b(_iwT<1YzV_Dh0n
zRNxh>f^NpD2=!Nm8#%w*8dYq1po(P`s^}W3hJn-BB{5kYTh4N3b59*>s?~8?UjzOv
zG!X5oiR@5KxRP(qP1b}z_fx$x`D-6-M9<a5P4>0l+{Jzd@=bH{&A4P;On=NSclyJ%
zm;?UD)d0_W7$AM70oKx^XqjigPGm!N#PIs0C12Z5jW9pW7~Ag}!<Id0PjyYPYY6i^
zOU$6-ZjRW&7Pz*|0(s<{@#LGgM|cgCYl%%iEK$L{;uuvs40&UR$yN49w{*a(t<)G~
zka6BP!0V3#`Z_CcL8-tbYH?OkhtnfUfw)r&lv0nguv~%Prj8iv<OoM$S7#qbOd;P)
z+v13dMD}hLaDLO^h&T35Xfw$P8rz(hd+vlz`A$$+JEP3U8Ft;Bv4VV~yWJUQEV-xJ
zMoq$b7yKmO#C+!7V5}>W$TxP}PZ_B(EB2pe&}3$7ckU4m&2htJ&To8e-LaI}u4Px<
zF>#g$UM4ih(bg?6{&)-2oc6@DNiFfMuq8IlZq0p>7s?NL!%o8+9xSY$x7Qoddw5;H
zmYJ*pz8JB@7j7?oG4^tM?ES;fOKLmx>A6}+kMgLa0!9wNbu3_$6okBe!SFsGf^$u{
z$C*uUQ9g5m$rPgvssDB39x8x+m_5iN<d7wtx4fOh-4XXo>fAH^*+l>McFt*bh2bYT
z<|uh)3OQ#K&l=v(=3{pJ+8REmiq9R%@A+>T{Z#YAFkm`!cBtvuKZN@)&WFw_sYPxV
zhEtpwnc0Qm8GrvTjrj9ED*^7z7CfZCD^m$Y9OpAp<gblNq|c={d?dRzyRyHbE&Ww?
zO1Nq%apY4dx)g@OAd@qj6XcuCq4+zEUE#f$>DP+Bal=q3szb0NCj|RXgdl`Eq<#M|
z*UvKq{;DC^RuGIO>}4oh8jNGz=mEA0hH7~bh9w4J`7-uub`HWMb7mNq0rly?U}n=Q
zmk8=K1Uh@!7luHzqV{Iyn*h{MQ(U(&fW7pb-53Pm_cwox%cm#&EOQwb`eR8)W&t#2
zrV;rjtCm{w<=jhV`@;8t4-9tr;QTZn9PZ(Ru_iubNOqh(@Wx)|xG2|q!lO(9v%d;#
zIpv7D7f$HX$PrI&^Z$KnK8_4!hJUO%?k+LIUw<>qsx_s5$q<T@d>uA6z|0VREZ`2g
z_a{B}PU$10J2k0Brf892ijjXTae1f}UKd)SA<P<I<E$}O&j#y<TVk>YwI*hK?eVj^
z*1!S}x@sZXS_`?&w5WO0!qPcf82wBO4|{22@e6IdO=W(>D(dn|^bk~Rh-<3k8x?jV
zaXx$Jktx1>HpLcuGfWmU7?qo2r?~}E=UQR`dvq%;t<iUYEzVPCY;ltO(b54%dE6y!
zRp6x!cgl+#kx6}AX_ym!bB^;)-Hm!CcPwI0)pj3hs;{u4>XsXNGB0rUYx)VetNv`N
zfD?5*vBw=y7UaNg2zz|pZqN53avJ$&HTh=SNX~7@Hs6A&H*&PcTOQ57wm6kziveWD
zKJiUaGp;EfY_vgoa~rI9YmIeXtzoBajg=WzI55}>Z&j>t@J18N+|UGu4VE~sVu{T4
z^iJ`=%cp(DI231ygJhezoXI@mbzVlC5xPDzqEC*!k>s0H^34hI&0ap|690SY-qeRm
zmImaE229h{;WkVi^ZWgg32J}j<A0p9u)A<U;7|Ek`-45!KP1%oySRM%CR1|0$?D5r
z$se`ioLVbQuh%m3wpLOP*NT<ZM{)UCBRjs<$d`{b^7&Pb^qNy6Bd64e^XM8Gdahaq
zL|04m^eXW<TP4@VR7rcgD!DWAov5&H;Ec&98K7Pxa}28F_o@#vYw>$I>;GPw&_7i%
zyg-&aJd-KPJejYRCxy>%%YPxa<#vZG@tbm8;+JGf>qA%N`t~d0vi7pP?3*q#T~oy%
zC{a#Ni<4cBr)6-$F-dzDEw3IOk#5fq%VM9yays~sI8Qhz-g6F!S_ysME2G#;5GDS1
zBBl4*NI8f|DQgrd>iH3J??43EEJ6a`g^O=Qxb$5cE{j)(%Y(#ZDHxF{F+6R^IjhHH
z$_Adr(>T{zlPSYbX3CbFOo^_@6t!=evaK#tjt6GS0kTam?F^Z=F;!;OrHCERCLUGt
z&0ND&NwH0p>FractV5~{ADS$IBa&s<$Yhx}olG+~S>h~HWQ2N(wE2=Or@tpj#o{FC
zHz!G&OimIXtt2UVlgNExqEyu;%852f5;P=9!Z#*KixtVD`7%XzO-+(6+Y-fuxj=o_
zB+5V}OW#q+GMhi!;c2o|^i7kogD=X(ap`hmZn|i!Nte*bbXlL7E+*$Q#cE}aSZUss
z(wBE7-z`@xTjoli4Y{)5Os*VU_f*tf@@47v0(m{DK+fDKlsfvqO5eYbROS<IsC^+0
z>Lrqo5{Z6YA`7)k<$YMG9G*}rEjgp9;v8o24L)|WR4P`LO8op%aadX^N69w#mXUXO
zrvLAdbAIv+-cTydx0cG*z2u@(rSfK9sjS*rEEU^}C4O+RoY_(;>;3+d7wkLzva3u^
z<h_*^J3mXZKQpsG*UQxH^>VSaK|C#~H8xj)=4lnQ&{oA^Kj!+4V)y1sc8NY>hv++1
zYza}r+_voeT&@mPr3SReYG6UM1|EFiYw4Q??z?Cr(oYi^!!>c3eA6Uc6RX%g6*QPV
zvWs-FX}vCT$v0ieH>=n+doD>A%Ims#HdG%on16HD$p9yN8lZHJ0amad)_|R9cd7Ls
zyUh^QhDHduV1!4fjqxqW1hyMY*n!Qwhf$_jY0G{F?&;f6llYf>bDDf}a-Rhra()wf
z+Y)Z|eBWT6U-2(nETF$5t->BY(>cFc%iQ7}4roNaY4Y2F*;MqpiUJd7a6Z4D9;yfh
zX2vS;fLfe$j}*wRP+-3*XYQQgjOxOEhAH%*ZX(yDIHHn#Gl_gNO~GvSiB4#=)d_=Q
zoUr4e6aM~of_OS3IfPk&gUC7D4ZNeTBPxW~)7(#$JY`=4=QqjiToJ|``@#9t38*(i
zlYz{>zuXLs-8j3M=|<0p8?Li2P-Q%`>@T`wo4N<iUTBUe@{PgC7DzqkiPl|OqA;r^
zJoA~wN^PXLd%@@cyRP)Tv1h0^ELQp8z?`<Y*UA?YX8B@UkuQ4^+vC=!_NW`fOqIM2
zXk*O&29H1#F)rcoC{P?r-;aGTt|Wz^jS;)H=Y`@aHM%oDD)CvLKB#8o5Asb=5Bju+
zQWrdtzNgvrQ7vJw)ao#paNm^78BV}XdZNfNW5_ced;i}V&HGpRm?yt>oX@f0bNBOm
z{^twNm>b4DCwEZP7QY$9xlK=Mi9^}L;7jceXGWu&GF!)h^PRs+r0~zO;iVE6?<+Cj
zsuCm5Dw!>-MErUsKFwx7!*KFgXC?gGDB*0&OdJjNG}MGLM>-V!u7<+uIJLza$u?7|
zEAA1BLQl?a^ysU4AA(EF_FH^31TU6^ptx@cmbo+M?+<4+kC{t*lG@_=!Dt!AY<-hp
zoG1;#z0=ISnID91!9loT5QH5i%=2S6gR?HMf_|wAW);7h$DI^+Q@ZMb`1XRjZsy(e
z;q2yB>i|qr=RD^v_l39p>9b+a`$T`d^zugsdi-sVcR=%s_SD0-Lx&h&r2b3I>wF)?
zboGHoYai_S*@m-*Ht5{U8!0b55ku|Rmum`axTBy~hPmAQysy~jfN0ewi27s!ZSHn+
zFY!9z181Vts``yILNI44H(T(t^S3_sEzn2uemz{dqlY6Wm@VFg*_+dhu-DlHPTx$>
zCD#n*+pI7!n6n8x?p}IWVj8<36ZsmvvVa{eXZRj)LKDBAYobj9uX`M{peI_mwpI&X
zFSO98t2UaY^7_YB2Zf(?@L;qq^P=?7fjviGnZ4q&%K&fKBfajqDRy&4`@xs`x<%C4
zm6@Zfg(a2^Xo9PpAu*5!g%P%RvBeG@&)K8q5;b;w9rspqMDzZRXuIDLyDI4g=;DN_
zXL#-LpBo<1Bi=c}9S&pN@FT+s+dMdvy`q5rBn2*cDDaj)<DW<enD%qPfHHe5;8ELb
zkF1sU=)ROYD)LQ8b9>mD+QZr29s{iHaq*`u3~t)Ovk^7f%%NyIxhW<rwn37@1~G4}
zA>^AS8rE=4Vdp|uEA)4<!Xffa{l85xqFEDsR_C5-9k0tK7$cwmJ$$%-QtxGmDfD*F
z*v%fk2qV~K7{QHvQ+L)7mE@a;<eSeb2H4B@(3N=_NW7$hAcw~I!wZxwH}(dNtrrt*
z_CkMP=3ZVMJwQLj{>u;9n)5>}72m~(eAC77n=GaGWzbCaH_xw?dCh8N*ScEf)qa$M
zzwCb;S|dgWt7ZBAYDwEvEw-zx#b8Obe4bM+-zHa!9eu}Jhg8ea>M9wvr%IH*Rr0S@
zm1x{ypVHR~xiI0KDD&RSw-xVYWvBOIsrp{d?X6_a?>h;Rd~sLg$(KX9lK<+qI4f^6
z)-+50^}H?y>oa6m=c|%3=ZduJcUhX|r;4L?l4u2;mxNoVBx*v8^f_}(dftnc@3}`r
z<Njf}>2z4w+96|o4@&lc0}^sCN}NYU$zAm*nR6*p_AH2$Gwmbg!H)<Tdo4l=w?#;n
zdW0m$g-h0saGA0)Ts#(q%g>X^GEy@`%H1+$(4Gvrx<5m*lQSfbCnhCBwr$9eMby~T
zEX<I#vohpSm#gwTn7U{By1EZclw}q1qUD+(lL8VXcU^*1Zb=ZotqF3qC|*L^#*2Qd
zvoiAGDQWfTl)U<RN;Wq>Ehc`aMY-~v_)UtJ6Ls;@&^<wZXC%m+d5O~McB14LB}$Li
zi8A0_q8NNil(|0>C8lwbMCm1o_G!*^_9lus*=Bief&}L=6KGYUEFs@qCEvs@Nfv+3
zy=?oW$<QGerN#Jk*+{<0TurvwMYg$?F8{`5O8L+`lAMtvhsie$<Qot2%`@`NM)FO+
z^-ra{Q@*qa%9kh43*^Gj0%=T*_17asVsxTdM&%XDgvKRuaes+aKQCc#CihCsOJzXE
zQaM9bxlBfxmRTy-sG(^)yHs|}BhxG{72D;U?~rpw@&0SxcH~San*6h!Gn_)se&&#W
z*4`<RxX5B@MZNX1WyO-#r&u};EEXldzQN-^Ij>nR4PDE`W$|12?p`f#x_%J93H37A
zyk35OtQS{(=KEctcZzvD27^_ZW5bT~#j3c;8P1qbs@%({;kcJNGJ`cxG*SbF`!q0;
zeAD)<1}=TpK!p=C{+5w(9JP=_kNCN!+T5Y&B7mCWe9mvIH|U~(eB(gA5%SFx^361*
zK8|f>R|tC<s+|p(t8Rc=&Teuu3~->_fVsZRX`nVP+>HM4`9{#A&Tdqg3Er;dZt4QP
z;WJFx=gq!3OZq7LnxlNV1zh>wI*juh_34(_&Ha?i=O!34&=$SF+oI(IJM_-ANAo!j
z%sO<yo$U_DjdH;FS_k|wQQ(Ikz2cnTv|p}30C!YdsWo0mzL~^%e%VU}cKlLcUQ^C%
zIyvIrO!ijoAm602f1LW81oF+grcQ_+=7jNEIKw&VglUhQu<4Hz_Hi%JG?;#>f1I&l
z4mAM>onf8gj1(<uj7?pT<LiPu3+WSI=Ys3hU*_F+h2^kj==i7^ZhA7agLycP>2A1c
z>5iJA>>af8K)-e#NE_gRQ=Hv&{ni50mwCdAd^7)MOPtJW1%n~2xrg#X_DV14z4Bs5
zgEzKL@WGZ!A9g>q#n5kUv0{2V3|!hCU0%1x<qjRNY(ochuA@GgS-H*6Fb{`*=2qmJ
zQj1{pNoBtz`6ht#o8Cpt`5nU<iEbE9bAHp9`zaIh&1TM7l-wr;&tjJGBKp0l!5Phc
z)63252;WYYA;)wl&lpmVW6d*}_h0j|sr=e^K4&SP+raNxN$rh2XFy*kQ+qs`JF9`*
zX?17+crfp`<@5P_-m|7gM~}0d-^%~n8~&k$JF0t1BwS|a)M;vo!<Eon$1aIkO1vA&
zomD3#ws<MA#D+e1W*3K5b2s%g6n8F%V$V@>%?5UcPiCfHH+D0$2t_S34ZFPyf!nPR
z96lU^)P*6K*o#>}E+JT6&#tL^oZTD=#=u#@RDK5Iw{9?I6a>NVSP)iD4}wXDAe3oR
zZ_I4JXUAy~G-h7^LBSv`fz8Z7>}(Z?p3J<N@sOVI_00G07l2Hs09>l`$8pYXE?)FU
z*KPhVXGY^|JLVQwbb$S#4*YCw50ByOLE_BDkZe=HTQ<}q%gG1dr?o*5b6XtlIifrN
zKdt3{)#DQR{JH|6x2U=0?52;IJ$_N+dY9J(%kxYjQ6|WFZ46C4W2_oxgeIwmxYL|H
zC?@QG8?O)N>3aBZSPwHc>O+5%0b&LjVRWJqBB=AfJHZUT>_3|8Xu)1cdaB$kVex?N
zj4I6c+@y&=t2EJKohDAjF+bv}Cj7`bYE_!p(@P7>A8SD^OdDqGrn_ydgDzz{)Di1q
z;%DY#F)O!ph(301F~IbW+`r@+<A{+N%3HA$^0g^KO3iRcgDgcRuy?S+)_d0UquXNI
zY&&Femo)XX11d(6>o~Vs(32gydmPd24ZQ%q^ne|A!m|h56KA<1HHSTMN0>>t*b&S4
z@3mq9d*>!AFtQon1JWEYb-n{!xW{W*O05mKW}U4)c4XON&SvKE?BO*N=QVeqaPLju
zLdbnPT&lFisDJqNI(pjEnj&iw^F-+l|7&G~S*6VJYr^@>A1kOlw89$ljg6}n>gt-{
z!tN$8<NMX7L<=mOM6HdXF-o!xahbEU0Cp!t1sh^(Tj~&bU+=9U7CEy=VJCO_!`Law
zOpVA&vW>qkE@Wz8sBdFLzi0%Tbq$g>xnBB?U`8!H+_$Ui=q=*;{8PG+Z_Jrt8$9Q$
zOj!R>e3*^v+MjuGfuE&$=UOSh|5;A$W)FbpM>+btM*bz=d^u1p%g8rtw^qxjRn_uq
z@&9x*)si@)TIxsgKKrqcwXBvNFRP@+tx9~zH~E(;WKLd%+|qw1_Z;7gm(6<_U0ErK
zk1OT4S*0{ed?z30<x7y|6G_;bEAex4B>w$vG4;49D-5s6hM)|wese|c8eEa>0cp~_
z(>YPeIW38qF=AOCBU`^6m%~SnN!83~>Cyj)1V<m1rR9etzB&7(><-GK&;v5)MwEE;
zj}pJ1k#h4?q|BNSDN)S5Y56rmdY_Mw6Z0b^zEOm%jtv*(x^Q_mJ6y77go{7*%c)f<
z@{Y5ZKf3flk#BZJWysIi3<>9Llh;>e<F+d@$|hYDHJ4<VL9#>_$BT#IdAa*KP7ZX9
zlaBM_WXGB~iCYyXMhoNQZr5|7YJOVUJc*TS^J3*?af}pH#mMi*vD~-C%Gj>4^59CW
zob3@OF6YjRazebcCmR_RCdii~iSkc=qU>O%;o9;<=~kU6w>ZOD_$N{N?dR9}CCaPk
z31WXQLE^I#WZazu@m-iGDLcqU!;)nW`R2sgWZ7v#t&dHL=qpmhac+t<Uy&l8R;S4P
z%PA6ii{7i0i_+s{mVBkBDy#gF9H0GIW)6QUtEWB{y}G9|%`0EdA)i681#;*;J@l4^
zViH#<w|^AL6Eeq3s}dP}phRY!E0J~SCGysdd#TV;iP~5yy5yVM%iK3jU{>OcQqh^q
znasjcNha6y=6TQiu6(TH^ip{@k^HlkdSgD`EvH02L>0@xZN*|uzKPycEO|@lVWUUi
zWk9LCo%Tv@JHC;^C8gr8QYNnq-ZJ~>gY0Auw0ByaY)Yz^w<;>!-KxN)mntGhsN(Ak
zRos}%E~>Svu-dAMUV7{_uT_Iys0J2|(m>Z4d@Yf0RCvv>Q<XEDD;nr{OdS>6RUNsb
z0WCF6Y-+2Ci9>X7Xrm7Gm%8|e8k}xXx+t;KgLZd44D~11?9|8l`}$B1HozY8O$K-V
zZ*!@|VZLUWt06{&8RGs}-p)0`K5Fg4x|`q!=Qr~>Cm%P*6z{KacI|A2p&iZflpS>H
z`W84p&H~N3&;Dnm4W`vJMT_0Gcxh^f1ErkbWZC0%y*;%74u~M%#F1~bYaH;{K!GtW
z6d1tww?#`7Xu4B@Hz)XO^37dd`)(>$;QCJmrkgurKCcOVhjCA}(-Hr2?w|kM5jTD~
z;;W4lJ6WhLUgv}%C-}ZezFGa3Iq~k!*a2sp7(nm%49;(&osoOq8CNr%QFD(O@(N}g
z_jh4Wk_+aaWJV6P#w8}rkae5cu;iQ9)7(%T=Z1^a7svl`hp8(&1<5vA+nB+2)&qw+
zv%5;i6OBfCV!+szu-MZI0exEIt)>?ykMY8c3tr5b^F~HzZ`^t9jV_zpz^DbiHZgvj
z-L%7`we8WfSqF@m!Cd6@4ybG959jUv$oVG#Jub2Ka6k~8H-zAcS}07%hho;9P}H@c
zj^%_B8#znqBViaxzG+RqIWwNSr<u&fSxD_Mbu{m}ZxZgEHgi7nh8#1DJY)GUGl+Q(
z@P3;e?EdE0mhd_H>!`!w_ZX6mV#r8KCxszuB=yDv=%eb!Y~&#NsM>Ij(~R>RD|*n$
zSKhza!%(HfjAA8rGuzK4U5VDG_}D&X^{ip{)C~4b4N+o381*-;l&G~L^QbZh=L5Aw
zPndO+9*VGNYK+%~qSK^MwCEa&4Ia$T*AAtxCIm&-L!fbx`r>&Z@arA|JIDXm8@@Cr
z7}^Ket2s3oJ=zE3pn5O{(;L1qDhQ{>2VuWA8K-d&qVF;9hFYdPeZVqH(9c!ycLw)T
z?t$?87=Si61K>KJo^blXGcDQwP9Iy{O?FhXli~Mz<~8>8r(VpT88IC&rcVd78^FG)
z4(+h96L(T%8}p$&HEnSFE`7=iy>SXo7{<Bc9PV0bs1?pRr@)yj3fy*JX5~`q&lcL?
z@jOce8k-~MhAG_lnINW|p5q@z(Cu%8iPWt=&@x2IaDB9A{^my;J<b8?6|Se(ky`)g
zMn<@}fX^Lgghb|a`rX&TiJ3a=L(;*vX<8U-qlI@1G*LEN6C<Z<GAC3MQRkQ)afLd(
z0!`$VbCwgJg;}q)aIc*<Hty0!kqWaT_;Y&o(S>7`F1{{j_ijgeD(mzyCx!0?g@!0M
zHp1JU#xQ+qg1UdrpscY#zI79N*{o2RZjIhcY@s&X4mT&+qu*e1$xsEJ*K^0*nSF8L
zj;JYNN3OdQF7S0(cc1f9W6p98I-<n^M|>OOh!1NNsGQHNPwuDg4x)$bkps0soCA3{
z!1IMYQWn}{w3R*lJKLhm^QLHdqA9#mY|tsy2D{R1&^N;d-S5~iSEDIT8a72_l?_(D
zvcXR3Zw?#T;OH~v{Fqo{*-tB8_gJAzk`+pwt+42B6O5+s{o+$&>{@3GgO;4z<dbVg
z8DchPH-&uvd+cwBp{)#&^uO$y>jrqQX2|zs-ts-OD|0nUKj@<`uPfJV*T9hA#?YJ8
zAeL+DMRiKOY$e~AcCD8#4Rx}&cfCx%RwwG^-(}e4kJPe!l>Q5ug&R(v?He-8hZ;#d
zRwEJP+1Hr)QRa2~D531PTz|2K-2*k!+Mq@TMpTO{`KF3|^M`z6H?vwQ$v075s^wr+
zm6WSji77MX+#)OFQgnqJ%BYa69~Dx+v{L*pR!Y<4O0hUqDTa+JrE>T?(Pcm7{xMIb
z_nimwd2_Bbak(qDrg!9$%60MCd{xZpn=-LalO31OicCKv*MG#w?3fc0S$ABDHXoDL
z4bkFg7cC<@9+7=J4$GDohvby)Az5sAP%>H{kd5?CMJuCZ;JZlKurE?H`bSE*Q>64Q
zjFA3EBV_-`2&pUxm*_*`;yN!}Ru2!CV*|p)amz^=efvD~Q&Qwj`bByB{;C9x$PgKr
zA!}4K<Xq|%DO`~*XR^*qY3g|iK5|~Bo3Z2SL!4L*IxDs(&Wh%uv+^|Zj0{(vk=t8i
zWaPdW`F=7+&Zoyn<NGn(oyEwY&oOdYgIq(tvFs8nueZmFH~DA6@HqLsEl!@!ix-s(
z38LMQAVaSuidA-^JS|I<U))jM`j{yFcO}aA(TU>pHbFjdCuN(HAggX9$e`o|2^$zM
z(?a8=r7~Ws|B06aW%1(2xloJ73Ct-@l)98exi=(P!p9{`>v_r2k@KA{`YGa1opOsI
zX<}1*S#*D1m2FFJ$Sh_pmU-Ngf97OKG53n0MFsMNe1lzu;u~8i(bsq^pUV~Yin?Da
zk{LX^cz@c7BI!H$g|zScLae*LkQ>D%;ykQW_D!RQoHLwX<daE#O66&<QaNzDSdQE*
zmbE-xct7$N8Hvne`=mrVTxV`{1iks>n{3W+c9U=H=98zm6Laa!8PS+m;@b12+!|dZ
zuV%gx_Z(_+g5S%KS#@OSdMW+U7@wKfe1?3JOTLLG-~1)tOd{XRBHtJosG(oF8p0aX
zVBcQ@^)odvYl8+RmulctvIhDbR)?P&XF35INY~cH)LxpfZJ~pqfjVf<?Bcr?x`_4B
zg<elxdY|?1<d`0|ROvyFj1%3U54w{POun%>YyjK)2Ke%e`zi9xU}iTQn?U_PJ8pgY
z8zZNi3H0`uKw(U+z;aVODl&yZUo*&Yb9lG6fX{OaeC%X}+wZI}q-9fPz1X5C!4_w<
z?U-q72UG6RLqFLg($@iZmpT03>$Fkt9XOv?Ac=f4rH2BS$1=Nky8>MfDX=q*yLis`
zXFpdU_=f_0Oz5K`-<%%mh^T)ZF(J(nLFAjS-yQMJikh1cCybuw#LRyuq!c;9`8&NM
zX3kj9f*s@iosl$|vm5e_9r<Pn`Q{woYoFM;Af&Ae7R_<Nh(qk)eCooE3|GXTZieyN
z)Za{U!{MWD_`9B30bS|_nt9+0I~-1K_CWV@9&jPw=(nIwQaoYMv?YoSTH#{n)|~6K
z#$@L78}0IfX0aDCT6&}RJ#S3h%KZL#?lvy_LMzq}UMlTzg?o=~z8&yqMF+Ud?7-`J
zb_b60$9I1IzI6;j1$8*rw5Ye85Xv5H=HVTrZiVbH^CD*=U>*(kM<(Q($>f`xGdQzZ
zz+DtI#^=_A;n4<i3FkAdw{ho0#t9_TjOF>p`^kK433bKZ+)X{<a|iKza{0Y|$Ve^|
zc{_qRIsLi+>Pmf%uy50wImRwwP-zl|OWNF5)hp4D*?z`FO1RzS9*TdSn=wiVvwDnI
zD-k%Iy;FmfC=aDiyd`HiO_Zo?ti<N`p&0U*Omis|t&gx<b8RSOBK=fdxU+H(MZRVz
z!pqn#nn`VSBt7ACLeRSl_f&S|yxL&&&!Q)Me=zJO1f#MIyM_J+VN6aCVs-~%{_r4N
z_Xxt>dRoeFQC~cnM-yn5DS#7qRO17YMZRfTz)Zg*WSfZr+|dLe<uN<L_pxV+dnq}{
z?8TM-*w}@-8*}cb=5;{hY0NoNv|~qrH~O$UpuRcv#Kt}_Y0?HakGNx`n=?X{+`n*#
zI;D`FzpDyJC^Me&>=8504!+hlaLj0erd=!%l5CEp4^6R-KE#YK#!%U>j}Ei+pwgfV
z%_-E!^`+O*N|&AEx>)p|F19m^bInsjwEIpSYlsmVKh;BRXI(Vb(7}}wZ3L2SR<fhT
zxRoZXJT=j;3uiXnHBrO7h)Y8>p)o;|a|?dGgdPOufX<=*M#oJX7nf*5<+nE4&(=Y`
z2K@*B(vwWaQR%3M)?f9}e?D_R*)1@?-Vi-XjX2XW!JqZKra5PZtD8)5i~Fj5PtDQ6
zp$Wpwta0sVQ*1eGi%q<KDSE|gmH<a&Z>R6^sU!Baa>Adr)LQeJB;~jxuIzI}&&`eq
zrQd2NuWR&E93bNy@N0+z);_hvg6rg^y>|HV-WEF3?NM)T51k-e939jYx31anea{BP
zWSnX;j$yhDa;`J4=d}$KdQI{Ay$yn2^06~Es4uX_gS*rT(QCG;#0qxIC~S0+d$l!u
zpNeUM?et6NA2dcF_k-8jA>4hjA$|`v#6r$%c5!xd%-0ZK+^8|mHo$}=1I*aNZa!wb
zC-6PfhPkLZzx0sK3^Mbn8u+YijA`Q=q(i?3@gd*jkZ<mEs2AsmdKvCqFTRI9iNQm5
zU2<l#d+kT@pZ-xEv!D7iGyYegMzS~6NJiKvdC#oeUhKhqzo<q+HEP7N4>dSmHKHC}
zEp51y>fM>Q%4#{#f_o~PYT3db$oX5|v-7)Lo-8RB#aC(~nALVGwL%uwRLI!3?__~}
zrBtn}l)jrQ<(Ph@*t27^pL>N&{L6g5c~9kgLZ0Yb-xUw19I0x!CW{;|$)krCrKl!V
z4u+qVkl|-UyYi%zZa*Pu)0yK}b5vedL`#lNw7hPAME+TJST5xp65Wi0()r;5DRej>
zdyYoQd9Nrjd>knS)Dz#hA0c*kBjoWF@=a8PEY1#>9Sg(d;@)rx85S-dyM&9?(fu;d
z@T5$MI4_T`otI8I@v<;IUJ6ddOZka-sUj~-Y7;M)qRz`GK1Y>auI^Ung(5T=Y;#Jq
z%VVWsX^e~>6(glXW2DjW7%AJxnbY1F=}f*ENxnIg7bB~l$H>OY7}0JND@(~Y9mzKj
zcf?Ba<5;<06f1jX#Yyu6aq^(|dHMP#LG)b`<tz74{c^dds^lD}B2ngVN|ZjM5=HN2
zf|!zTQg0`S!PNvA5tkr`$v15R&x^~cGlETLM04*MS<Y<d<5$i|+xurE`1V=p|M;w^
z7$k`2ge391O#M$?vXnMTlM>@JneLD#17}<m<=l&sy6mDnFTW%$U*(9g_7nLrvp_7F
z)pKG;p+p`ll)RV%`C;@-_K$xiN10Xlb=EVPwd0voAATlT$Dc{jiDz=_`*W%4|3d!#
zQ6i~RIMW$eDmVI;O0Q=Bo6$U!c3&S#&fkYJT;q{^)_lZF$zoA?R3cNymdXjvaPpIR
z;)=zAe3Q7Cdn)ox=H?<f$Smm4KP9qESt_y4MPkwFjkw+@6UDnqdi<)T=KDA9ryAqy
zNcNof<h*7$JyVmpn_8p_l?|$}_@jzMOEvWMRD)Hf8Z=dzf1|063A*Y~->Qy@5$bT=
zr4HM#>M(Dwfn+UB^zWyMwH`WfCf~TO)j=#ZgqtgvfuqpH68gmZ9MwaQ4|+J?R-avQ
z`dB0ea3<fhJYs+?)ZR?`OMN3>(`odG&z;5@t)>w?x*6m1R(7u(GeP+;6Wmy0ik+pV
z@L-qjoEURBg8E|4$yaT#f;WAmC#<PE>}!jRGq&_m*}}KI9jd2uj&sKzV|2(gOR2kw
z=Wgny1KhQEz1N(cscxLzj3LWxQBVh@Kn8tOo3a(yTfpn#uL{gJaKvgKYBhV%JH`E!
z{b@&--KUO`d=p{qgpvGw=``61)rn5fDt5xTuTIc0afT;%R1e{d3FMm_%bjr~(iw-y
zH$%QSV`Nho1be$M!_@^g2VKx`%LQcvUE!P14EN}1RUhYu5&PY6YLh$K{cwkon+JUQ
zc~F<=fsNT7XuZ5So=j)~`!1e%YS0pG99seO27h?j8Wt_sBR$0nmY2NnqJf=6N4+s{
zLL2;?<pbSlAB?)<gZBUU;akV{&^XZ^R?J^5=-i%}y=^g^eMWk|{0wGqL(TtoD~y2o
zZveHjSAv)+7KGcy>|5e&<Rx{zyU91E<eOdOn-S!jDP)x0%Q%Ojo@Vd*Fzn!b<}^9x
zJ!dt`$uvnk3wg%y*F86LA4LXwz~}bo_oPx=Y{#Eb&7ak3IJG(bn2*ziy5qnw#CmbY
z<4oU_1@~22{F!ys6TfFq_;V!+@6Zc=fw?y)nbWgdiR_i^$(qU@@qzS*hj6yz$=Qu1
z8K{vG6Dv8NdBk1Y#ZU}5OlBkBtRTB}>l}&&Zp<>)2!-c=A<)SP!J+UFyqy`snL-H4
zY(g;kV=!i34@Rrq!R%gQF9SVh59)%j^+pgD&>Mcce-NgU2`_vB<OFLvMgX*TqRR}y
zNLxYKs6cdg2t+5cW756=WQ_>G5vu^Sz2=WTYyF`;${&yWQU^u0dF}5H)zux~9n&6V
zDc*SAwk6!wu!Gg#8K1qF9og9#XOB{|uIhpTD)yM?X%Da2_Ruir{E@TAL)4NuWZR?J
z7&}~KR!rPB@(r~zjW%(o%lXYp&O;wi>rx0_f2`I<Oj~WNSf`Eo3w2<%P6ucH(ZT#%
z?23!j!^PivXy9BYk2==fH|Woa(!tE>+Ss{E3o)F@7==)4Lmgf`&(J{jwRF<Nd0yvK
z4Az9<IQlA?8`1K<CT?kJVZ#kA^w-vg7I##NzuM?KOb2VY&wjN@7oRG1v9hxsGXB*^
z!9@BDTnsUlIao^$7-0alDV^vSn)rvz!(IY~n816Y8LxZjISR5O@0;Vf8#8k*n4=<&
z8RV@Tv2Zr8KW;FCydyiQHaOzQAxAvq+@=HhX4qN&`*1(qu9E|ft8@4BP8)O0v@wu>
zwjVxP=(R@^r%Rc6F<6UxDsA*Q!1uB@oWHcR!RaK<XvjBv$v3Ai+2A3WCz3OpV`@!t
zkD8$R92@k$VvW>I);LYw&DQ5uu)1dj_rL5?s4{}*75bo77~;ZcLuBh2;ydSN3QgYU
ze8#pT|9(nCGz(-OW<C3A{xe`dxdHo344At>=IdsFqQCm^@YF|nsTN*yK6Y!O1}ZzM
zp+`X@#P_AgDX>A@%j)Ix(gulb+#ow!)ywgtpJW&LCidh<8N*(q6(cyO*;6C>Eyy-z
zHS)c2jbyE?VV25A$s18aJxh&r`N^)vsA@4CR4tBQtE90?HM4cAB#-yEhI|l>&F>{?
zcBOchl*{QU<zlnAT%6ytU+8|JJRHd0rI97#{+6th{Z2e5SBf!bIV;FJX@wOMt5Lx|
z?6;Dg^q*khBe|w=SAGmilIPl~QqlCHG=wv=n3~!b)6a^U=^5Ge^Q82qCp@?FaXJ3*
zs3bp)7M*WLWP<w<$(eLmUdJ60n^OnH^6mjyYIZ;(=$#6+h!WR^2w8V8LXM_INXgs?
zv7HnykFM;O#5w!rxz2tW6thptIJc?N3zx0t`(?nO6LQ7!q*(1fPc}I(XP1*P9>q!b
zoH!YKJx+R-#mQsNYhE?s<J#v%t1eDf*`5-6wNv7_Jx01sjuG`?F>-`_Q^DI&>tf{i
zt{Ax+A0z7$V&pVWHIGwjjM!hGj)+}Zeu=S?m>nw_k7C86I94|Oh?QU5Z+*BBC*dUt
zva2pZX7x&x775I5CfhvvFH!znAmfZpl(bUrvB)=5xSMji#7?eL3385nW8!~aJY&zu
z)U{`1)R{Aq(BrJUI(k+*B%YOz*=OY(`DTrNg3KG2Byk^;<#=VPG}lOz1LPZH@=XBw
zW)k`4EBVH&;*ylU$dT_qbL8%JW*r}8MvzURWKS=U8V%+z4|yip2cO9@GR>ywXW~Tf
zf55S4a-4iqO1^POd?r_3JeQ|mN+fJDwZcugyK*g+bJHHm8`V3K`t7#(8{ZMD7I&mv
zEk|<t-jjyo_eJyS1NoZtK)wxsDC2Sq#A7|@H;alThkWy3d67)0cp=TmH*ulN%CRYu
z>t=7HVs#mPVddhn;+?c5$KUe&BKxN|!e?`KiCe2+VPEc~mZ&0*e3PD_innHJnB}a7
zh4yL~;jV^vbICM|)zNj8I#zFC7S0NF)O=9K9d8Y2EY-l88VxLG_Dv&y9cJw5;6C}L
zeYp;X+tIT)UKb1berf$)4<$bO7!|FLd*mBC@=X=_X2BB!=&2e4)B#N5jJDf6Lk!R`
z!l+KhQ0EM{?Ku<lBHuiUGQ|}3=MLcPCjXZ?j!)+P`lSWbcJTe@uQd)fwZ$#wjTNWy
znv2(9@lJLaMZSqWMK9@Hd#n$2faVbgEXj4im+ua^Y0uAt5PD4}D3G~Pfxl4-_$4V|
zpQV8HGX*<P>8mwl4;A@lBKc<XT;~1pT5$RYN8~opJ8tTPR*p_+N4{B}<OBoqjcu(H
zqKxRFBHxsgZ^lnzALlai4*6z44(B;<oMC9?0?XDexYp4Ht%kYKv&?;!vnw7ZH$!YW
zyAA!^uyvaoCREVF`rI9r4jy>lg<V0bJfP1U{Xi-(Mi(`Q-n15I^w1N9R`dn;YYo?X
zt&w5og|{JIcsY|ZoK!Eg<BsawZ*MFa?}O*`-z>Z8gI7L&2tMYA-<%EXO=<_<%Wa{V
z))uSxwZ*Q;wzzeqEoLx7Zc9a5)Vcb?j{4*$^rLSG>VQ}DOAl`d#d$KuUG9+n$2b3D
zlY!)$uha=YUqvqz_fEYzpXpAH8B4~=<FO#uxbTGX{$R3CC>cnNe6xw)^I%>WG^aB|
zXB@L_hB2?WFQ3zykCAUOTT^%A#9nc8{vMiPi1^9OK4urE7b+2bo1K|yN`QN_`@57l
zv_grqQ<Sj&hgpQd><(|vesK#WPW}zWw0F$Ac^HZ<7ud;gC={`4LZM5(dDMydH_bv}
zOujLA!=9-tA-KIS1O?MWFjdLi8><j>stU&KtHJQt5sY5LgR#^j7%RU9;lIm4xU)J4
zbGtG7#)`Z{Z}_-_^q_HeqY^84J5G>$D-ffGF=MDUfckLe6_ahgw+Mj8Z+i64`7>+M
zpWS2rINin{otRlXWwjraH+?YeDs|_@9@wMV9G`v3$Bxc;?&{2}e*XWHPA%;PJF<f{
z_Fl5akIUA`eMXP+Ob0X`P5;j}&L8TmuomWs4l^f*THsDCv%W)3q1dX28==giiq}SL
zls0B7c-ur9-dnUWw7(Aiup75`jW%9U2OPLU8{Nk-n|w6$;h0Y@t+dgvSPTA}wGheq
z%ne@GIP)*Nn<uG*CVc3p{L(`cT7xw4U<|#KJaaFz$EBLHT?;Kd`p!OH&Tt-n)JD&a
zI_&4vMeRzyH~iMc>O*?$fY-+b>hY>{4RK^0cQt98V`v*7jy-geU<mzFoaa;+u}8%S
zBL^9y(%J-j4NalzV}`>CX6R9EhTq=YPfg(KvxqtxYLSMhDzMr@fiP>{x8uzAhXYP?
zX0xTYKK}CWIp7iZ@6;GZw$s2*zL#z1Ec##OUVM({eB>mvO6_zo{sUR8q6x~Qt)X_%
z2E*fQ$dcroOdCAs{3elXGvuia24z^|)Ld&s%xAx^hZSD#wL%N#*!;0HLQyoc*_nUw
zmt1q7f4`|JhM2}P{r|hI<JWYBS||->*!(bn<7)%VO*cS~f%?cC$XuEkI_Q_h{$i!}
z|6%E@<D%Z%r>_D_r?i052nH&P1=mn^AG^D|ySs}LkzAJW=&>7Y!U7Zw&|@GWf{lTq
z&z#@={PB7163Pm=`~J=~*9;#1*2ctB+IT`A<al}n#@=nfzNtTQ@noIs^ZF&Vn|?B1
z?WYVIS}RYke&Bxfdnrp`Hrhh=7;y$Qa``)%s`pNMR#c1o(`vc$xmqT8zmq2+@8rGB
zJNfXXTE1N8Txt~g=3SMvyjn${3TIK7Z)J5|r5Mevl<ZdJk~z3cI!`Q<-*d{O&$UNV
zXZS=K?<^LJ-p^!f+-uoQ9cMW^CZiV?%k}NWlJ=rl?0>UI{A;N!JXIoZTIPtl4QD1-
zkIOasb6ZCrlZ{>{WCq_>7p94aZmP_=%lTAymDq4Ee9(?~@jMhKjZenP*&i`d*D6Lb
z<{Xmw^P|On=Rp}mZ}F2)2c!$T7;K6nW$ZtZa``}{6ikhjN8=-;{)KSqJ0V>D)P%{&
zh%lKtE=&?9hsn~_VREj&O3qm(i6t556pvY)N_z6#>aCWb;1szZlO}nekI1+0N2E#J
z5mCFRNQ_yE6fRK9+1YAoHbpJwWSf3N)H0NO<F`gFp1aj@n|!mLT%+Q7%=7A$THMJu
z-_NV%P^MZOZgMu2mn<8|H-TT0WwIhgrW`mTUr!!kzr-=AXL?-bMABDNdYt_U^x^P(
zQ@}l`U*Pv>!7({?>X_J_J|;CW$K>0pV{+gFXKYQVK`l;`WvOZMzUN_yBHye#c35t#
z`&(8wJSwC5o)8nSQ!+O9v|OfE7E^vk&Ny9@OQo5T_$O1`Y_sI**eq!~Do5<dH_F-^
zxwPZ19IbgI+DVV(3$q&6ek+s}<eLZNo8{!2ZRDHQ<eMP!&06x!xTJ@&FZzj;aJT-M
z)?NA4@|FzWdW$<?xnfGTc{n9YW~XF{?TIWoawALp`(?`}uN;XUd{;(azQ-<(T**3<
zE6?h3<<Y*!a^jyNIeU?NTSZ0E>*q81IHg$J1OJuZrp2=1WU(~bTq5sxmC9$^GC3Pl
zCe;D&r1aEBIUVpvH0U*s^=bfH`iswzZ|<JeK$C}@Q`Km|ot%^A&0X=oG|_X3CZ?&V
z=RBkLc(@jK$(j4d=Y;($6}Zp6s&~|=+jY~y89r~MpXBpo3AOA-`iS|YkDt_`)zoo*
zAJd1Tr2!iDF~pS4MmSEsX;5f{)x4H^(}KGJ1GqbIm8_#{f^Gq(@LFVw%EM+j8P*UF
z=^@z0o++(I7BGKmfjt8(F)4`|=gzj!3Zibj(GCmQHMMMtJvQ9n?j|$-G)7T#PI7?1
zK6fAJ9n_oB1YzM#FpWE<&2+g-%NdnNA4h1<c0?H&M>B<aIM*G~qk!!5#t{i7+%;|E
z1a0z7%gyW+e@xz~a>DysdW?;oVaOe-pX8h8)NhK5oY^zM%rr|E6ewNrxrYmMCozL$
znF}IQn1gf61sh({hx3EEYt3D`hr=wgA+GF9a7B8)D`tormY324Y3PBPY3vKi^1ynR
zrU<rTmT^Z<=q~caikqIyRrJF48D8k2-3&|bc_C$NGv@oS3;C)y5*xHa>qe~*(1RYF
z?X56{KI7=Gt+AdNaITTu7cb+jR$KJm!~VdA?Qrp^5?xO#@i;~aBla0ZhAUwnrR4lW
ziDH!!g@={=nJM{NN`JD6FA6rX8z7BytN38t{T)Id5Ir@z%oHZyEG4V-Am7B&gVTZf
zOe1nkJu*%co(((=$TkCc8$<SqBLgks=YC&CJ|f>}%;21BJhK3Y)4xo<N$o<$@#D@p
z`6k7w6ZV_)YbZGX`W6D83g#3)`d>HJ%Tt_FC9+R*PY4dIWY+T35Ih~gK2U_<k|+1a
zEyzN@*%@9Qj3c?h_<V+WH{_co<eM^ji;~DUkH|OKTJ#*hWS8c}AT$dNg4v89)`th-
zt7Q=8yy1N60=uTR1!BQa?oGJ`;@ErkXr2yW_DKMOxjR*B9DuSHK+qu|#+7Sm?3>#6
z!ym`4`r}c3f20?(?~9p!6X<ao<?08`uiOXU#w@??9Wm&2JItO<P98}wJeg&3EBa>E
zvBT+EQ`q*PFW%l6pHkSbl*cTljZI+Ol{?E8jc_#D7EeNLF>$vo`pvV&=2o`QNVCDS
zC~N4{v%=LI7Rc|(oZyBg$T)1ooiSejf73@&Gkx4T$~@Zxx`;8;#cvH=6#44nV>j*y
zXzIZyTNfEybkUD&)0;a_S)z*%{<=v0rh|re*fmFuV}g-3y8c$stFJ((hT8a1pL3|M
z3iP<Az^yU`)Ry$!4Ah3%0c|`k&_+*ske8j+!Ngb8Mfi2rB<iBYBRzB;rH?E#dYTRS
ze2{2JPl^#PvI}x0=QSsmb5Ht)A+jBek+jDcyVjU6gTVxaoLfDgZwj@k89Y4AV7bK%
zGftQxIEDSYJuI;0panwiHbyb?A_BIUqb5}cZI1Hmo~F<Fi4L}W)<vd0{VUAy3O7;U
z>K84}#<XyZ&t;X&Ev?^xexjz#@|vd&qg46|{q!)+g4baat+9=HKMn`zA12>4Kh_9}
zr`(ah-w0psvp4j#El!QKrN(8AAMLH-_lDWRr>I588>7<L7~y)xWI|)+X|fNBn$6UD
z%xvdr^_v-B<eNv<oXhk5)?baVie1AuxTC)z)c`}NiFeG?ML552_B(Vi>$^7oa98c1
zk2bE%;lJUN7Jk&$!x-|7p7NLIuly-4<eT7`oL5DBm%<$%B!qmknSGa`t=~)jgm<#M
zs#;#0VSn^L)e`uqTFk9DlX80}`o_$y`AGeSeA9C{JvXnbB$m72p9fb-PVcwkSoxoJ
zQ!1-IGjGkWOg0ZFlT*FQWGcHE59B|VaJM4)!H&w}NhRXhqgYZ#vO9ZGvAB_M+GiBY
zw-3d#>wT$Mk1mlXN3x|~|9@oHlOwX9ulHx$tHpnQvV5&hk&;8HqVq9Dmfj~LbWD^Q
zb%N|$9WSm?adK{7thir@k$ztfiTSug(yw2%Oj*d?@Uv0k^FC6Vj))Zf3z4#$yHdwi
zM9Qk65z<a8LN@LQ7c=*88B`i3u`9yFePEbu>l!B2J;P)K`KB>X1{uenJf-NPl8SyR
zX*pFTqc^LhW@3tj4^5H1At{pIE=5LByUCubmXq_<5=XXq&e>GiP--@V)be1MT2{?a
z%UUzF^r=ab4Mj<^?0J$*PgAqwLM<K1H;4G|&AX%)J@QR(ZnE?v-_*WKmIptQ<%eF1
z^f6449lQ-*rIx3Y)#5r+Eh!h(@{6;sSLZnEQmdtV-(ylX`<N7SHdPpPOsv+>mt1jF
zdc+--VN=t@ynmX^?3yNKz0&0K^fWo$=!DGNc0%q}oRSlZPfN@0XT*5@Sy?^oiVW_2
zOKz>mlc3N%aS6|pE6I5>DWO1$uM`N4crqjWk#vc9B!z3a<3He$L^ppVeRmd0`Gi9D
zWfaQb9)*(5+p6h>qTW_0L!Q%<Jo|x|Y2THp@0gpYoh|uwS>jVKTP}6Tk~n4#9#7`m
z<eQ?aS#sfCmZXt!^!d48uiTRt2Xkd0`R4nTT$zxWD{pdg<#leZgkLC<Ba@0{O*`&t
zMHERl?P7V{j~*bG5}DAtL_V_L%&1S9EE-%YZyc*6%;~q-{Hc$!rp&=vq=9K0HBfX;
z11{v7mgJiw4K=Zfd~>%S=T(b&5;ZY0PZK5mso!L4VWXFVy_X8SuTr2F|IeMFZvE<p
z4vx@gye2{qga74oSWA6)_t!_yMVwO|=KgpG15_9p;vjoAEq559EZ+!W<eN^MKWl8_
z3_6P)aeN*d*3lFZYfW)E*$nBUnTNs5FH`bO!9!-olW)ShSmM?mOTgF)^J;i4JID^T
z%wH`SW6wTgW*D0|VD@MS%!qV={S61yRXJeFw#Hah-5C2iHGv&xRN2(?`}cOlYVMah
zY$DHaMz!T4`#19)ai+u(*A1L-l=@Al?oK#CA7@&D6LPqF+KT#33-ZmxCe9e#oJ<qx
zjEnc3v7m~lK6RV6F3d?`*57y+{95dSM*Cb~PW@&Z`R4I=7r1!3Vp@<Z9`$!c`&F(u
z#uvpEWv+<r>xLzR+_<ymj)>(RDEQ9)<0R&jJ#LDSrk=1~;R)w+p6J@k3wpj@=)c$t
z9}atAKlPk-XolD?-YEI%jcz}^QS8+UJM>$lp<ipf@7V^ea@%0(0w3hXD^cd<iv^pN
zh)GqVcE1t{+j)PV64vCKBJz!zeA6;ni4Vm}Tx;NqG96!ZGxo*0w6?hQs4W`Y3c}7+
z)IaP(;P5kq49mT3YAhF5(~Co2O&97jcI23wJmYvOcpS(!5o8`G-uEHjG$r5sO}_bO
zKKW)ky*A_M#o^3rVZToB=+X)QcI0fUWhcCGq=wUw`NvwEVSNq3+VT+8KIF{lCVMka
zvMVcr^Qqn3lUfmi&?zBE>rZV**n8fTEMy+S-ppXUE#qD|`DQcu#*KUvOTKAH{pK3^
z#*_1@I`Ylo7u*@U5QLlim|;9E2+sq8U~IvRzp_ARpAE$IO@a6@kbL7z&UqUEjbj1W
z!riHl$^dr324LbtVDWlJ^E?#HbY$(!d*<F;@yCR(?9jaBhr+#n*x1z%TTT3+Thx&y
z4(y4)+73B}uK2o*-uJik&NAa*=r$Yl_{R#ZRn~Zx#LO3O_BHji!Pf&e=v8Hd(bQ(l
zEN$_}$rh%bw(#g^ix<3geP@H%2zpps+hF`O{#j<@R$VhkrjmP8tGEmG%>X+u=_BQn
z9v*4v;cg3EC_d_7MU@Whf9PN;wVXG>x|noG7b6$w;y&-UZl;UJR=PMv1{z9_u)$&G
zUQw&5DOaFzsRCxt6sW6IAmfDssf7x(zplW@=L(!M(Pmx@J;xihQOa3L&S%bj4(YJ}
zMF;J)*ul#Vm*F0I`1_R}^hX*XPty>e&Ksg?mJuGY=Q{C#5fV<&^W$a+f6i-8%r}ND
z_r1)S8}aa-3C?r3GC<o5-|fv1{oaVZ#rklbtPl5XdN{LL4>ynN0DE+x-9iULY_t(;
zOOGHQubRCw!=@?Fn|l+}G?;fmu5smY<~jJ8_sKW4^|UdDd}BSCdbCO#dffL~dD#Hz
z^Vyd-)*4<j8o_U8BX-wQyD4mhL7|Os?T#%JqpT6*&V9~JCTLbdhB;@9lhmDJ*`bg|
z&1Mbx<~sRiDfy-=Ii@*h@}AtQIQ4@+n|DSiv7mm<j~6aB#OZ;C=+nxO+(oWDqKg%M
zbn%rts_Q<}$Mi)TFR3kMaG%32QGsX|1%8~+!ibRHGKPGUH0!6ls;QNzIkmFY^@kk1
z{!Qk&e~_V**+=d8UOF^-C%N43ezCGz2GckEr&G1um{u+Scvs7{_SN#QN40dctCr3`
ztK|M;?tq`IlH0?oWT4Mm2^m!?x-O;iSHFCj|0iE&{kSg<FA8MGzYpc~#>dj6>WS>#
z!u^<lVtLZ3n7-9w(OFO|M$~mAvsf}LO69yoiMTGvk}J#pmbiU~We1XEYR4qeY^0U}
zD^#*|tV$jzRTB0yQMw;Zl#kyN#C1S|jGh}We*NOac1EnURx{i0`61cY_mGroMT^t0
zgR*t^0Vy~eAxpYOi2m0INuLlYJI_Z*p?`#o*NBkrso~-m6fP}}he`R+FhNk5tZGlz
z=@2G0WSO3SsbqIIm8|C3(OV_;`S#*LD$y9Bk~W-?y-ZihzE>*w{az)5vXbPBwOXz^
zsHM3pJvP*CO#7=Ps;gQa_D>S6?nxrklB94}k_^vG5|7*@dH*_zxx8xWNxo@IzOf|V
zw7!unNAGb)MgPr9^3Ay#&Z)>afwjrfoyUbUE6=KA$s#``@M|1Ya{kp)EoT&J>EuVf
zDBzf^-$dU{-BG#n>8RW;IVu@r)8z7~G-*e!IS`a46M1{M_PC5<7P9yIQ?h;fX_?oV
zUYpZrC4!mJcHQzt>&pY#>RBM~n-|F3wgqxm3S@6Ufqdv)AWdc$NV_EkA~^+8(D#uv
z@AgP~ct4WGn+qjze4#w=U&y`wLNS?GAQw9qNY?8I^0M;-*{*k2TAj?1o_g6bvq82@
zHO-Rt_F3ZGK1=%c%91YBaPrS&iN^IT`SUnS?0RL(){i-onto3{F3puj)Nl@;&Xs1z
zbH%niS1z=_FZ+$jPwk7O&7>k37*iwzeicbz+hW-fR4mU!ie-xGl^o;m*o5rY((Tz>
z*;HF68%pbA;*0tSAEJR}^?%Bgq2Fb~m;ZH(j2Zq-95;WL{_noaDT|*{7psX)XEpKf
zG%e)c(PIBH=Tp>fW{uQ_lA7s}5FPlYQNtOkixQPC?m6mVRhk}_Tj;}co<8=dxI<39
zS!`yAzd29VC*K@TH$n*crUm(CKA#i*@-x8^_Q+X~Z?w9Z;bFQNHgV=&US$qv@{RR&
z3!G(Niw%2Q_MEiD_ifCYs7J3KeSqh`+F{mEdsMt&?v}L!!pA${^8p9Ex#ocO@0qQ-
zu`wD{ayKH7IqcM~`#U*eZ~%M7XVP=H(GkB6I3oHS-*=aLw$Is3rSF7XC3_kAF$ZU>
z6FYjD>sRW8bH6yJGI7S8#?JWW<qX#dXISSuLsj97CFC2ecJvsNZw`%hLFjxJJm1Y3
z)e#reyx<O116N#kb7d<ZuhIK(Uwj4q#>uYC$za#=H@>fp8!q=|Ct-{mW)`^NYk)iU
z^<yve;-;AY#sjS*nY%jA6Wcy<pK7cZj>mbS@gpzfIW$9EYBOj=wLq^Y-ngH`KBA7T
z@YJU@tm#R$zS#zO1ASoGMv2njK2Z58=^s+Ua*h()w^Fkq<3w>rbt*!M<^z>5ouNd7
zHA>7}t3>uXCEPYD=^gcD_fcC+FK+vPS-%RqV9fnRHt-9<?LQ&7Vo$~)w+tY^45fy1
zgJ&UGXErsR!)>`6MZfS{^2%|CPWVPmrjkr_v@!Dn-N-cLoB8A$EAq|5&SW9-O>6Sa
zU;O;#W}InhcEV)xjTZT4BKc;>4SH=(Fn2JXec|LA3-ZmC$@Jd*6@sDu+#B~`$7Vx*
z9{FZG`6lgNFapUpZOAvx=)Z|1-&l1e+mLSx6!agHZ+@Q(!uGw~4WAlBKPSC74Vj7a
znmw8)0}-%35J`UpvMVAG@5%!Zn9A;{5a!=l1mMeCc3fNnl*@tDxy*yM6iln|N7Y4t
zbbm+BcDf(_nc&CQUS{^u(|u=12Q+Qtjm6X5F*ukV-PJbCz_-GX5_;y@SG0(HGi8D`
z4p?!&n0(V<vJE1$Y~V%DYP+vC+$pw&fxRtuQ@h+!&lV-zL%#gW1`YQyYo(11*6g>2
zR){saonik$uZHLzXoih53^0kgdK&rk|8qv<MxGf=UyS`J9lT~YQAsVcYMPU4xaXvp
zNdBNVXQG2HW-4^i?VS#;W$WMu^_c3`+IVnZ0pnx^uE!{_X|Dn^xO=d2E$7txxJ!9h
zfzNvEsG}#RY!11GjPT!{`1#d3ctIWI6M5<l&kA~6iaG0Yc*cAQLtV~wnVsC)5Rs*N
zD9g~tJzr+=95KMlYwU)c#Ot2nCYZ{(bqeQH)f%Qa@WB-6)H)Up(Z>*9&dTU>?$nIC
zot!US;i=*Ikfnf4ssa~>ard*o0wG_u@cNw==6=+IE_ILNJh#X-&w1{=(ZU9@Lhfq?
z+;%Gv8OX;Trwy$})T4FvF<>wEJ_i_K!4mF9u=~8}N(-ErW{E{REivktC5Ak)K*o0q
zoL8El|4YuJ&Kjf7KgM|S|7TOP$v2<LFaG44pY@ILmTcomHak*b1lzaFJ3V5AHd~Eg
z%xk~!&-5$qHNdeo23T~ST3Tma?sMrvk2AiP@3paE79X3tHufG@;I@wfhhA#ovGs5H
z+v%5NO!z6vPqp%3R;?H^|HgemjZ7&1EMHE45Nqy9wNI>;r?aYLR{C4{dH1c@uC9_t
zde!3ZQY}6$tHq9d)0TYm`$v^Dzh5N>BdetCz_&8U;H{*v%lU*=zAW#WFF$VOb2sUM
ztmo_AfH#FQyv1YL9{*Sl((_r*zgXUoaT<_uw)6UW)|NXm$@;cD(7GWd$?R|%$o;4_
zX_C&@?~NHMx!52{);?3QD?%mS!&K6vnMx*pP81(iqU?L0AmchSw|H{A9O@b;UfpBm
z^yU~D+4_)7Qbo($j|b&-mniAcDN-yuM987j;j%10LKb(7l%eVf@%ClDO<lOStHVVC
z!zEi4CONIhHO<0gkw=)E;feE8$%alUkuEA}NWPg*zHuhsG$-G*r8j34PfI?oK4Vl;
zkF2vSjJJnWlAgrR|3@VzZ&dQ$E=k(AO_FW_Nz$@gk~AjY7|%|UjQL4obumfyzf6*V
z3hG1Tn+w!n##4{!%-#3%camkmqhv{>hBM%Evh*h3IFoH!@mTWv<4eAoO1{}ZzFADZ
zxj?=t)#6T;HfLS>YB|wGE!Q_vqnVf{-lNmxRQEJ_voBS;?&5B}f0`KeOq1T^oAiJa
zGJo`G`N|ySq$6i!htU;&EMMfs15vmY$SUsw>C&n|Uh)KyZ$^@D+6D9egaUC}M7~*8
zAno%D=%?oW-v#nUNv7fM_-4OCNh>apL6ZxlpG|>`t;v^s|GP5%=pC`?lp{X+*|J(c
zOQsrS$xN#(Ing#t(#bb3jxyizc9u*m%aW50*<#;2Tej8YNX;ecH{_dfn{s7NRIdDb
zkSlS`?~B^LP-gvkB3`YEWbD8qSrk<yQ|QHcXkN^m#A0dQu~<6s_}zagJyoxzq5E4&
zuc?%G+kc5hO+6evQ4dod*UEv36|yR-Le{LRlDCfE*wOw?Mh*EcHCD9}d-|6+g=)fU
zp%(f-*21|+dfbjFaCEFT+!M5+-CGCoye6D8SQjT#bTO>49(E_`;lnRIgpB!LSHt*P
zeeA!_9OGuZW+C4kyKDq4YP5cQKG5P$!Lr*X*iW`ecw~y3%g8p&T0K02Y?Eq^{qz(J
z-fMwj<eTtLmWZgMhmU--V{s##nrDZjAMN0iY>!<PoKH1!z!1)-79Mm!^XuHFV&>=o
zW|9VIG(j1=8YVHvIEfydGaa}~F_Aq~8<^=w{U-P{^S*K%(dd~YHZU`6VP|#+PNX++
z8~c=1%o;gQwyASM2KAfH<eQ)5o9$uF7@y0GzgJ`(dwM|0H-0@`uwaY}HqLdy(?}Pr
zJM4nt87_GCAKy5;Vy>Sns(QI%^io%RB;VXP>59u=T@m5!hNE5GpqlB1!-w6l-ohO%
z7P#Z&D|hr>$(faHQ}kg!Voi=GB8Pb4Zm1W+^Sv;d-8zRiG(-N*<~aGb1<oE|ci<~;
z{M)QGF66hy!jo-qEStIh?417ZXxJ%H5}-tNCndh}JmY!WR*7By<R0b%7OYUB-YO;5
ztmdqWoTFkEa@;m00w(yP#jUpZ$bzMv59!GzAFTY$9V8x&XJiht$~F3F?vQIt$TwC)
zxiijPsJp(*%5i4LCK;&4gnLv5+=u!Qf^OXR)}wFuUowxcJ2ju?+`($bnO0MFkGpol
z3%<X=lo@~;WFYd*^)mWu$Txx4nK^hY1b5>2XLqriVR;Db$Twa3aWA|hb8y_*KV=qz
z*W|DI<eQP?8(Z?tRPv2*KYZ2XU@RlwO!W-rrG79x$Twfk24U`=AZ(bz`$D!c3&Q7D
zftbv0hHI<Y&Cn+h$&CXs>|p@+s(|10+zizV!1q_chEucxnFCF^7j9_6`P2)4?%eoe
zeT5%J?q+BB)OMKXr9{PrR_L$xK+$Lq<oh?m8bkIuwX?!zFXmP9`af~J6$ZvyVFauZ
zZ()NG^wAp5u)$gG;png*;p-<GG%>J6JLaq0x3fi8@=YT7Ch<SM*<eEtKlPuz%z^4`
zjXk~1F}`<0n71**)pp#w3E@0whaL|6t&0vXbZ~T~4j#^B-Y^+wl^OMw0Os(KJ$`!Y
zp!smVmJZaxAikzL(Rb71unsQmrk}VmuVv0E(Djf4nass{v|WLDYZPd?RDt}h^syvx
zr}CNtdfwVdA3^P8yEb;$r=~kz2L+)zcsp7Lhnv!)JU|B<A8SM1Ob1C5bjWASId0Be
z9O}OH`CKuLoi1Y&^^wE=f#shK5z&s`74CC|ttUS)lV{0dcJOk?Jii?|hOhHI9Tf1f
zXI@ud_Wtr@_ER8nkOIEk|MYcZp2c@+G;g(_uGGR+-oE>wg(>8lRI-gR`QVhL0?VjD
zIqFfvQ7h1tkN^HbW)=?8!H342=Xw5bPM1kt>*qpa<{Ho+yo~wFYxtOMan6)yfd!s?
z{_--xtACAAb($U<^34(E^0kXF#@~U=8L~CTIWx}R$v1zfOWmbzbBx(vJ*Z(%Q&F?(
zZirQ722f>DFWPT_t;_}*H`M@b{m3?#>3#B~{>3hwD1Kk()7M>fLV*%;WrtEN1pKVy
zJm#l7>;6+>7S_s^3SO6T3-MRXTe&gwtz6H3FM0Rg%fe07vOJ(ldXIc7bxYq$)C6Yq
z)vuPlHr0|yw%Kl4EioBY(we@TL*1*yaOYe3W%pL%980CAX{khT*Co+6U+fR%OPu8c
z(O_1}l#D{z8SzjO+CP@H_Qf*4eX(rwE0%E{xpH^-U2z_JORi*Hmz$xNrGDeX()&Z2
zL^`EO{;g!0wOS<?GF9}=sO0%PdS1vFQ`}Xu^nIetiAj`4wF$DZU4pFY5-%0jaq__`
zRyvN3k-=t%#A#o&c$glPZ;cN~;Qa{cZx|ux7lcb>A(>@dgq+zQAzn=*q^2}ndhH39
zjjO`s^Yc)7Y#b&NEyH9rPZf`jW4w5gZ>l@1<bGF`xRP(~_EpJF&a7??R!JPs`GG1i
z8n2Rw?fkrbDp5wOWD)tMIr(PYMHM?Bl4Mk?B<dhZ(yvXD7_?53p5&V_@=d{#B)OWD
zB&W-hq)}y(bSB@_k#jb4UR9i`mfASA^ggJTDUoV<6Rwu2+juyedU!lZMn=#R^pzg#
z8kHD-QAsPFvR`B-4bJZL)bg8iFyFpvX}?S@H`l8rI94q$OVnbxIaL;JOqE=}G?~9J
zO%|}@;X)_&JsdkN-~Txym&iBgsNJlckS{x33naY-^Zdv*cX@`8Z@!ans)EQe<QseT
zHN0P1AP?^qi0PvONm3NbJe@*0u3IQt<eT@83uNjv{vFix)?FgooVz0{mgUHxfE?xt
zWy>v{Ea_m7CF2d*6+b&mE+5R2kV{!ICNE1aR%OWwmuxxlBS-25-I5O%@5#+2xpI+w
zb37$inzXnt1OF6==8Y%v#-K<JH!G3_-HW8znj#6=T_itw^2j);cEu9K-@E6RUNY0*
zwag*gnDu-kHldX=wT>MjQE$cfX1UCMQ7)tF%9$_yMzkv4Nw1#YB+u`=IBC?%Ke_Db
zyk8$3LNu}Vp%(f?Dd74KyTs`?oR^{vgD4%eq=uS2gfnjHsx{PaikIpklA3r=?xB7^
zuh02`Ap*7=qT5+R9N%aJpX)|A8Nk^zpCkH(nqX(139s!<5yz0UR%$c++rb=Ov(3>n
zg?V42E%0oMB?2a~k7k@T5<A<#kzRxw)QM};?a<f9o?K;*(2vZ(v3J1qNe<X>$N{@c
z>1pW5O!X^`F?<mHH#eG~=2sK^U^e?s5AKKe<!&wY8?Pv`4n4+Avm9}Od~=)gsr$Z8
z=u5tdrG9f-?Sy^gn`+Ld6#t)ZR&zdeWxq4ZZqtuLkMUM#7nqQ5RMc;rN4uc)92cx0
z-;5*QTu668gWoQg=E%8JNA_-#Z{o(g;=_7ZtUBb%z5(hw^cwrpi-Q?%%yVUjgRwg%
zjdI82JML)I*8{<yJ@B=#DaOrciYLdL;_(ts=x2H1w4xcdt!##F%>Fz3umw(!_r|?+
zZ{#?%#-#JDaWAF~-mt@H%T4YXzwp7uhDy%Slz84z$&3<uZ<I=u^X)m*Z6X&cah!bf
zex(vK$T!!>H|xnaEvVldp6H9UIc@Rep+Bs$g3yC|)sDgp;D^B&M=$Sf>MK_C&N%mC
z=T!Gj^th05{He!yakt8;6JArZ*+ae=^y_~;;FDj5;2r(KPj0j4ikeK#HEJ}RJx$1B
zh7J8Tl{M^>st92x4}HV@+`sv`mgJlJ{9JeP%~tZw-{hNV<eSTr_-Fc%eLAo=oP4u|
zd~<+&bG<Z}9wKJ>k#Ew-H%6P7TRbHgf0J)Icrn{pmmlvCgp<xexbiCym=cId=K^qU
zK6|CAfQu)A9+9AT7ckihn1=xi<ADL;z$a$jgs1ZT3+cVNB=mg?mg@>;Jod+nWPcoB
zC&Th(?Vz>97f0#mvRmMQxh>fP6>W{V7pT+tuxGl1C3b$afbUBS%o@f{#=TbPVQ!6k
zW;XQN*ubtIyPwoHsCa4v&5xW<DQuB$YKvIzNN0Srfp#uC0FrGmYqSmiX={W1^nmo_
zHNllf4e_*3L-amvinI0fF}qR^UluWYCW3w#YAXG9YU9OXZM<iu)m}GkIQ`*lrjC11
ziDZ(?+Gwuk>w6Q<c=B{GX|E2ferY4bP#Yr_DA-S{!1BS=IXLtA{ukf2Q()~X1qK{c
z!2F^DLmjj+q^mZ*&ez72aDFZJHasHdWOgAtSyMCG%oz}Om0B6;pr^eKGwhgi%-8U)
z&U%=?P8ZwE^w9E+9$J|iVo-n)hW=%YM9zKIOf$jsVJ1+MFLL-gU&rU4fz(}Y@ax~V
zr3c7Vfj+*>uIQ>je0K$EzG=aREYrP83$bLIbvzfyHxv23M&z58_q7mFu7#P-3hsY$
zM)iUpk5E9%g+_7eRX^#wZah*SV`k_hkb9t;><rPi88wzJhB!fAPDv1Z^VG&r{vgwo
zv19A8F}AXIVllI!i;fzjBE=YcsNswaBL~f64<>c$XXKk5<eT~AoAu>JFw5m0IK7B(
zdK$v2)BqdEH^=rFz;T@cu0t0?9_Zi#a}ry5Q3E629Ou8goVs}*W_5J(R$xB8H*G3v
zrPtFhVi{2>$;p-S>rtf?meCLVu2SCfxxVgOr8M04PFz<p&*s@1`DR!tZ`~{9Y4f-8
zoc`fupR1%;Q7r?uS4kVIGJ0QL%Wnlen~h6lX7f_1ZBi=PkMrb9Irmz=<%!*peCbkg
zU-oh)k));<W5Gk|*Nz#t>@2cbdta3L`Es=Oo;+3L$n4EGCH%}~k=2*Q!RD}Jl%>h|
zU#YSwAz4-js3iJ9qNG|-BMC~7<$L4BtS(NLCC15*VR2Gv$5Rn2PM2ckPDHF+ejOvt
ze;*RNkI^#x#z6^b6D84Ek+S|&gbaHgE*{OoWoSZ}c(n<WpHD-@jq|7BO~b@7EL1ui
z36;>hq2lr-RGjLE$#~5$@z)BIJr415AVei2$v3yC-`MpaWAWthH0h<1UDR>x$Tu0|
zRPuV8N?z<$NpX})9LP5#|KpnrD!D?wX>Ok+^WUjtIGHVOy-E^}^6wzu^xdB%t;3Tf
zwk%0(j;PsjP6pw*c9ec(?vOW$QHx!KT9$8B%jE|ui9Df_cOy9$TOTh&d&NtSfOuKi
zF<#F5#!Kt^2@+OI-l<Q2GWDDB=4$b3OixZL-lrdE$06=-J?Bo>npC;HI#oWjOp#|F
zlI5Q#$)Yk&6}_jaQd7vR!n`xm#Po`+qlS|_KVKepzc1?t-<SM}_a%MLeHmo(Ku%Mi
ziyrYnmdt0z=E(>0;%<R-epn!}<ePTnn+fEbgB=TH^t=N3n{4x}<gRS%&JM2;Ibz!(
zN9z0KNZ{CPajTal!xdTo=U+ZCH%m4pX35s{EICEKS@M|<zlqrr)ccmKUUyGCf^x-X
za<0T~&XwfUT$$POzRamBkXM(Vh{CW)hI$l<m(Xv#x=8%UHkCZJCiELODVF-}sN-b4
zkcT5)%cTz`(%t!ue0W_hsd44vzpz~1Y%Ld?gmPKMob53OE2MkMd)e*tP40Srml35u
zWai^vawg`td`_x|Jw^3kH?#q)b~iw!gC<4~*2KXdS{O>c@u&8EIE~jke6HHCkk6;i
z`Uqn_m^+`V#xtw5*K*FM&KaT|{e-`6(qGIy0%f)_4n~;3j(npuWj+MEhKGlmA+N9@
zu54wW=5O|fKV%N;JWIr^vBE4h_ZeQZ-_O(z<*-Bd*>(teWk;WoJu++T@yXEvzo$9i
zeT)NaemG#>KIU~@WtYrf?5rW*Tq57Zk#B~RZ%oNI<?HD;j&#JGe;o1SmLpW}9dTNh
zOyj}a_qI+r#QpI6BTm@<$O&D3IN_s_Ggi}o({+Y3CLExKlg*sJN1RVNxxj$_8&mR4
zw^1%ApY4LTTV1f7^Qj5hJa0I!a&kp?Usvqw&W>gBP3k&V42dS|WRQ72x?*exH}(g(
z;V1Q*e>K=o)!!W-Gu$yE$O8+<d!XWr2i{F*xA^v^XjGrRoVA`fTIGe?o0?%WyQ##r
zC3SFboL%jWs_Ir4%{}oeVQrX|=!55K9|Yw2;OTQ8?0Myb1aImzWSju<%~0}9BPem7
zd^2#V5=Y258PsqZlW&~*`l5WiFANs8Me{g+1U>RczvhAo`9ZkI`Oc3KoK?MMcAh1>
zFMZgJz|5NFUFn|*p-0%8Svk~Z?sL`@M!q?3OusSt<`=b?wq%x-WR_{<o0;cBkPtyl
zhKy3#IRt+630t%dL0+p61k+Pf85n|{K_M^<4#6Cr<nAH(&0{(;1cP}zhqGU^ZwLmp
z4}mWEW(@geIr%1yd}Bwx(IDUCl5gxc2eV@>7+c@aYtxvy!^fCcL+{Ll6@WH5#fH%}
zKbr&dxq6>63GiG5c=PDbp~q$#@N^<@b2PXY&tB5rfSx_@ejj}|TLkI-1r2NcaVFOv
z*37+me$@|?zW6fNhF)^!|8Ku*kDsj?LHCd~q7Pc(FXr^m*0jW%qZSB0V1c?mRyeqU
ze8Z2eHnBk-8K<bB4UBmPPU2opDKk|xY%!g>PZCcinWul5HF?S!i?Xb-wY4RN-!O+&
zJ##EdZiq>*INP|Xg9)4k)fH-EPCj*&Gul{sLxHig6tG>vod07A)O$}(n5K;`JLnbW
z>-v&Cyk=R?Jsr+|8s%`l!+FRq`iO&S*a<K~!TbmX4E?!Ru(ScTFK&R{VVX#{=B%2m
zv+XozRSIofYQz1hG1~0qr5EP{`KgrGKqIvg&$-oPzE02J<1yE#mSao48Kr}mWU>x*
z9PeSeXilGVNs1n>b=C*k8X=&!F?N$3QahPoZ!Z(vv|+9e{Xgnf?2jRvTq586OTNi$
z$j7D2j=XvbZ2d`&A>Ukl&yHX+jmdv(!#Pxc-e1r6S<^%K>8Td1__&`HYNFOv3)=Uz
z@L6AhcD?n`h#42TDf-x@XMnri4G=b&JD~r~%q$Jz!%RYBOZEbDr+f7%V{9hR?9VYq
z!2|9}T%(WXgfS+PZ*Gxq3duKl`<Z{Sfu8A!#%SHm7;`w2|G?a@B+l3Zii}XlUGR$5
zoPRMtqxPyH%y$~%3U@W$C+T1mb++&1LqF;_TSsf-*Z?gwTg2|13@!LysS}-&@A6Xn
zi)0<G6s)L}iTl}&!0Xn7sq9C1TPb=UDkbPkrBnx3N_uR$=(TzyAI+J^cezZ?bSM*(
z<Z@a0tXvu~o3G=PQrY<CwLJZhFUJ$}q(i?vdD=ct&brcf>XRpZ*5}F4r+IRtQN9@3
z=gXI#*|N6P4OuljPexReak2_!N#uPAFv^!`?*2A>kty*3nKEP4H5v3HL%vuXmb5%-
zH;+<f!pdaXtgDh~oI|DBCyIx2f&{IKmphellEPgn=RtAePtB&dELO6U*%>}Omi<dH
z5_<TM^j{q<?za!h@uN|)qaso^$41C2dThQIhl$zvFxjLVCXr`D#d1@qv{(=-{nvyF
zVnby*-`D)#P}x--DruiW<;>?$`TZtN3WJ#=NDb!(`Q|Y_#@)y^X*{_-Ij^FYQ%p@~
zEay}`W~;=VbF^kFRpP{Rk@Kppd|y7#z@;i#Nw!(CiXI*E&DrfLaiO-eZ!JAW|0aok
zIOkoFiE=(EQA&>`ia}J8oDEP*BQnW(dTLBwQpchP$#VwhW#RFXK0RJkmhqyf;rHN6
zoH!WA%lO9el5ChDTYssfkaH??8|pl^YRPV@7MJ#F>AOeG&h%6<nUyLdmZeHj!xZr@
zNESWrp+DB6rc<0MvoBnb7A?|cx^sriVUD5m2KI(FyecCtuF56LtJ2W=s=Q)n#>B2y
z#i0LHDII=Q^0Ti=r23`=KF<>UZ&|YaN0w}Rnk{#~=7<%wm%DGWCG2aqEV9bs?^llK
zx6P5^?Q*0m*`@vOOtG)cl<PM##WXXMIX~2IGO}cGewHlfp7{1L*^<8Smb6@ZPp${%
ziUIki-d6f<4(G}lZ~iXT7Raw1PvipioBei0GK@1Sb7hhAUQ{H{j~B@_^38b8MMIK{
z<-+P0a;w8@IZ#<5Q3qa2z^ZcT+pk>q_AHmNYs%$ZRJpiTmP^UJ3Q10X&!16^9JKo`
z`*eTE+sGeMui%GF|5+}(UX)4LEbd}?m&u=RrIOBh?ZR8{Woh^yN!{51$(-T8HQ)}_
z0(y<P3mU(PdsOrb9w*;yUCL{p6hkPijIiS(=g;JuZtPvTxy2YOn0@nrKEgX~%}`rz
zhI8)bI8|wm%IOvuN4^Q*UU(q+W+M5fyoCD;<Qu&JJ9Hx7Od#L%E3(JJ`VMHtJ(SKf
zIG>7lK%d_Z*w?u+;>kA$el^A&za|*PJoYjL_owK8@Eh-ln&sr2eU1pCezT4FH%rJj
zuGXAQxjVszeA8-<6ZDQ!x4GxUJrw4^w{*ta?#_6%%9(RI?uFlQ#_0lQT(Y3|M(Kio
zy1U@~NEh^`esf}{3#`dE)>$s_dHp~A#(;eDwVNxtj&a4&HLh55&=oajT;W~qik;MO
zwoG<|cA^{B)VLw5n>)5HbH}d7?r74?1G_3cU^|uBI9r<{=Sx#m%=Cn2ju(=qG(+cE
zEwD<nB{D9z#A7#av^?1gYuC4iOIaJ{WcXlgtPcWj_+ZTwYB?``plPf`N^>QqGvhdm
ztTSGeyv9`G>{KN?FqNo_<~gB6moD@q-|%JTZ(A%#@P}iDKl>B~-3JQhz6(Nai(tH8
z!#PhO^8j0LU;8g+?ag4v<?9fP=e%hm=T_xSIw9JIJ>o|66F2BYPc4~-ACLbx1mkXn
zVCp$)Gvu3fWRz6$%_%a?FZzT7?LyGrA_Ps1n0r&leeW9XdUGCD@R8jS>|U6~-S0tn
zxF2<dS;Ps!NGHQA+8d0F-N8_74+geUTjETr_u^nI9}o<K*1_=SY%S_!5F&2|!v8#H
zQ1lSbIUNYU&4IYmk6jg>z!7(#VGFX)NcJvF1-g@KQkZX(J%EhE)pXzX0M~@!TPn!O
z@yFN8ei*;W4|~`7;gXv_dg=J1ROyFimA;tY+ZEQe4p?Sxk0%YegVW9ibD1rypJj=Y
z&n%EcZE|KG3;YhT!pq*AQO)9v>I?aYjPuLb1`6JeaHNhCWP=tPZSZ}k4W=<Krj9$r
z#?0m$P_4ih&QwhQ(#GYn+|Su)3?t5o_dKP(vP7HzMr{nep^dQZ_3-0jJ*<x5|IsXF
znD?aqx`n$?N9h?JqK#G5TTD5JTC__WGgr}HOI=5wdW_LI1-5V=)q~p2&n{#N6Y2-o
z8{pC02Ix1Z0eY{}L|4vvT2nVUPyM6&JNCIW*2cfRxOcFEIuW&^OZ&++^tCjYqK!2A
zcRy!q<JUWFd@`n<(~EPeEqwiEmxHH=F1Ap^nMmz7qq#obdNK!Q1UaF|80tD>#Qf1l
z(Ruo5#&Um?d~=?BbCY~iLcSSAzWG~I0S)ra{;OJ;dPNIIFKb~yjuw`aWt5!dlvZfL
z_l6b@?A3z7Gv;_5V3+VEP3$qGC+CnBR_W;>?y(+jgz3ZK13PCL8X%Cp=LS;@U_wnL
z`>Fw^1T#md9cN3-<BQ<))^^UJtgo<#J(t-w*NqWLzG+FmIYGYJ6K;$#yN&UBwK1Mf
zWBx@O_E3ZxA&$Q9i&KmcpKXYKI}Bm)Kp)ohfcqKh;n{CpG_0bp`>8Ibr0ZZ!XU+#t
zXyalpO`Mpn0pr2-ap!zJSWWyPotu4;K=*2Ceymbr`c_JR@=XZ&=4VQ!m~!W-{q;)u
zq4`?8dcTq2Y30(oKiTGKnWU^KlbsFAq?58t{?0EJ{FN_(KKWwoLhU6tPdpRy<j-Gu
z(vf`Q<&-Bijq>F5lsw75nkSKO@?_iWJc%i}A$>xw$;qgzvOE8ZxL&#@HOp?urMjDP
z>_?_7IhQG;{4?cAzzx}b`ic~rrpvmdG)XN_k+&0)<ylywbeNYY%}*qVZDzbgu8o&o
zFXH4mcct?Gij!UD?9(iXl@@8SGJ0Sv_o-q;%i)l;XdNxn|2in8*P`Up<S1#fEK+<{
zhD*&7&ZZo~B)m9OE|X^-FAkM;lR~9*VW<=x*)L;ehD!GBP^sX@kGu?(k*`8!J^5x_
zdooK9_r%FJ+sHSbWSgEmTgW$8`%q(|j<cUS&ZC*sb{4Co6FFoi&tK%5#bg^B9uFSX
zBF?t>{?F@F(u-Qpj|0?lUL?w%{VKuwMCo#XESHuj?@uR6pXZ6ve}GC#=<6B9zxNCO
z-uax-dBn!cf)(*nt`{#I$a=05;$-HSI62Sn#~$v-fAZn2W`ay5-vp9xQph*gJ=9`f
zMQ^jUTDFjHVrQmGux*MQdyy>WWE+or$?``(RlXLbN(=H$y?}HHb;*#fBQj*tsth?0
znjstCW{8>5Wtri9SuQWVEN9t|ZZPJm3|oCw3No%r@z<-ePw$#|*k6;3p4Y_u*iAW^
zdsD{oXKm9yQygArvyYuKsb)E{$R|hUDs!aIq-@#pGgBrk%oG`xDM@R{y0Mw^?r5gi
zj>!_=FIn==D_fqdza{fm-4hjZrEqkvTt31aoG0Ar8hKw9yeW_^cHGbU=LtPvMPlb(
zBxBkXNqzd+r+#@N<)%eq_=CAPwe+?3c_GOfuVvi7C2}qNwRD$q+0wCGB78ZUT2#(X
zt8!U%k8cN5NWi2DY18ea41V!}zV-LACxBTBE!dxCQZA0UWzuy>nKX4OlTUf2^89?M
zJl#?)-9~<td$a1H*|PdL@?8`Am*}xWh5g~of6sVtfRdr?*d*VqWDnJW1x7f3!3c)r
zo2dSL4%ljpcYJQJ*EB^Ccc-d$o1$zGcRX{<q0!v}4<1{flf5O@G7D5I)dpt{Ho_kI
ziW|mqH=Mnix6a$+6}`p1H5`!A(E&$iIY8?pbBs-x<KDk9e4`s<&5OoJabu78xF*<n
zwh1QqIbsC)#(kM1lwtJVTy{hZ`DW}lN4Qv0uW@z4{C4yeuc2;3{iZmNd4JaIp7JKw
zjCRH&^3C7mn+sQ%hx5Q0HgB9U&C7-C;(}fyxf4#l8M(;?GdZ8Ccaz<lFS%Exa7Cq$
zD=NCuUp#`Ho10uQfcnkw)2^tOOZ}$G74ak8;2PzIe&^iq@?SU19l?I_Tz6ct_Q0c+
zJV8y7ytyfs-ffD;gFR6)*$dNpH$xgb8(Q3Jfq+geG2&WF?EKRTxtveAJ#T}Y0QMM-
z^})#V>}Y)G1AFp~(ntv(SN1nFR$@sD`fWNW;n`OSU-niNamVU1`KI|vB_?(DMd4^)
zyj$dpd&_;{w%QlfYkkptgD-lw^~2L!f)TA~Eg;`Kr;an(l)VUk)N<0O(U4<G>7}V@
z%<fI{%^f}NSp5va@Q>7FID0ZJp>OBO|MZ(_oHsRF5`u8@O^iQ%%O2z-yAWh^7wU?3
z2y#Dg|LH<7GRZLG!-HYCE0~=R!8kxo#*XJTy~Ex;gP8%x$H1e@bG$qKLEVDsWevjN
zq9FFT2VqG9`xv+@<+eKrHCuw{d!e_OeDm-`AYLpBMEyR1rXx6m0R9XCrju<BjRn$&
z0o(bpyJVYD|NVR8{h^-f4|B4N^#wn8E%$?h%F~*Xj;LDO5y2T9pn2LCQHE})^K?Kn
zYkPd(ZUuEy8{`bMLI^t+J*nGFk2Xh<h6U`~Tfx1H6^tiZ!8_gx2kfokN^Pd_jx~C_
zSfQv>Lj+7Sg{!A2Mq6qlsb>SM4XKY|(e=<Wtxnct)yaz=8pxTc3AbPcW|K|+3?-|?
zXya|4dT>}$k28~c2(hS-&0Fikp+^Jueo?Pj&i!cmRo*#iqjsq_;+RD|fxMK|T7ecV
z$v=OycW%7`t2vLF&|bj~OD*hOtAWwu8bCFn0lHDcNh3pqx+*Ya7eD6-{W$a)*YoGT
z53~Nfsn>iZXE|^;Y05`>M*35~>8_2!5^X%J)rMh99o!qIgY#+~Sk~yEhI{FQT=X!d
zu|7sMGD53C>}0ynU8HdasL$?)mgJj}eB2=;=(%aCKsFg?KKaI%^C_?Ye8ktZ@PusR
zPqt~wW5)ZVa<yRoiuw%M#*nAw9xW^;^ZZz;2}z}|utpQkVVqlWzie-PecZju9=-4E
z1vWK6d-Bb-2?jVA#oXr{&W#=z;?{HSZh3RoM3&4+FvftB>{j5J%2P_dY5F%iq@s<n
zmVD#8i(JE<ii4jFF_yEfz21f>Y(Za<IrVIPL*_GZ7krli>g_N<5%=`hG6&IyeA9${
z6GFae7N`T~<D5$+X`uXCef-tCK6IJS_d0=DH8a@>b*@S@PE<;kf2H^tRLZHJZ{*9{
za?z+)A<dqbNY1Si@iu=gOMJ>@|6J~MA1;%{UCJc*Ua71Z`ma2{RV>N<^W`7%%>ZX=
zH$U>&X^<y9r{~FUZ)!6K?}@7u_jDt2Wu$waRHo;NO^ZCyV0Mbnj;nI}V}^LW%aGxY
zm!+A<bs4|zrYty?DZ>tD${wFg8PWHKJlTImDn4D3%BVESyqO}~l*w{qOrnhVnIL{!
z668WgygXgbIm`p*6z}4Es&kw;>c`2P7qJo&6)Q1sW7q`{Eqd*v<-KOKJoh~)))%9s
z*Q6+^zcf;!o^$3@FH9~y43(I_L*?!IP?<21Z1Y#B%qZ9|(+=*JU5EBd>fZe_E_}ZX
z<l7DNL#00XCZU~5o&~C;ANi&u`9{rSNWOW-qvYw}M=ggeGJxKoo}8sMF;>ao<thmv
z-;5^L^q57TF&X930+no9rjlXgn{Jkga=ki1=G@_&DJx#wO~_HoI5F%<Zu5_m2Sehd
z_s}@`xRV>75zO}C$q$c{VLZ8Eandq0PS*9{$2T$4aAU0Gu8)<V?VQ#99V>ChaniP&
z40KN=ett>(xg<#t=U@98vID|fEy@vUnJ_O|CN4@A+t_3oL9dfl%T$^8BvsaCT#&Av
z)1|sFT{L@V$RhI1Q1*&?+~xg-mnEY8Wyy}aEF;cemBIN}rPr%#(p3MtL@m2650kHp
z!uf{G8hS&TMBWg)+#B*+^QKrYyD6{Dx#w+^BY%-^LfYg=-!0kVp`9i7zh}x(ck)c%
zOetBFDWx%)QpDd4gZ|9gS(+uc-Lj>vWJ@(}i;fZZ!pG%GMQN@q<UF)y)O~sMqCk?Z
z3uV(W&Zx*U2OYR4-l$0WzIh_KpP$GzlOoZlpZ#|C=hCXhD_L@~M4IN6Nax$7V%eos
zy8iQ8=9O~yuZm1Rv|MgqE|+N@6_PZfLUcD(Nd3eLS-he`f`?T|W%CNT^RY~N-!GFx
zeaob;bD8{3EfuT!rLt66DtmgC%5=+5GNMs^T&&*!e;a8bvsw?`=jdYxuYb-lU;G|>
zHZ7>_+K_L0Gl#reJ!7=x^Vdc4&Fl<b2a#{;)|fJ<#T4Jho8u9?i_dlCUifnh^qy{s
zC*+&prMArZv_+I%BlKdgT@UVSzCUA+oS*jaV78yxPzPSiJ3zOFJL1lbp*y`Xp2s!D
zzuXUxZcgoHA{j^B1dGae#yKK=3450lnSH~4LbDg_V*cug4h@;(=jep*Q=G7EqZ6XZ
zH~YCewa&&Fsp5=f<eN6+8>=X1bUNjXj@+SYU&dW34;QrQO2!$%L%wm`=z^swE^xZ$
zf}P}>?l!InXyb~aF0N=d%oS}mx?)e1E6$#z@A$SWK9w-{Psa`V-fkGu)(ze1HJ*Ij
z4bj)#5vT8gjk`QhV%QXOdo;z-n5OWZ?TIy=yb!2thTyg>P`#xEyc1d=`!oADKeoV{
zMXh1GvJE~rVZQ$`AIu%+gNRTc9FJg@ToSX5uabGHdEZtE*QU(-Bi~$}s6@BroKJB^
z)r#GY<0t!I8ho(B-v_rlG6(rp8<@Xr!=FhT>>cTY1&`X{;1WN~UF?tO2!DDN{INR4
zA9cpma8A%i{E40z`%XB<xm9QKO+qdGGvu38^3AhXA$a(d{oi@ieaJ(}%h(|`Ed;S4
zA-L&AUygkUQuH}*`WX!C`^>B%-(-_-eD>2L+%Xu9+=DUEDj3B-gD~_*5OZb8F#qW@
zg+cHs2!iDUc1hh0Vkcn`^L~P0dLjrNj<RDanVEj!{G6>p@L$J$@DKFea97HVd}Bt2
zF>L~rnUHPBIFot;=cwJp4F;C=0Q%hF<=9hytcYU&LJxnqX!@h<j33nV{m`zdABI2Z
zh$o{vqOW5I7>-nu``pmPhuw@#sLwcC;oqiK*iRj@jyc@=<^26_XAU#(hUm;4s%Ak}
z&?e_pv-i6Z`(sbGVW#Cba|}Cfj)f6s@R_5D@P!(vyw?C9_tb~_*&iucUMIz<lW*<o
z#5bZ&p1J*|=i#?_sOlq`8cJZpdN|pEJ=0_A!FzW-9C=38nOq;;nGX@jIa3n%i;jmd
z>%oGax@=~5ZqtIIuNHJ{xSMlG6AwOWz^;J;nnSr;-GcheDoyBmHo%hJ)Nlqhz`K#W
ze@zpAsH2q4qz{KPuZzpIaPNy2o=&1qW&yo6+(U5w!$1F0fte4<UfglMVoje<1i6PE
zoKtP7<1FD`_jTse)ajttFXr=T^E!z8)Gx^kodz3V?`wX3H*L65Z(2EnvoP|_8_vX>
z$a3D~o6nrtj6cukoHJTjd6w*Qk)E1#Ev)9z=NX)-g=(_R?A2OW!N>ROi5Bc;YhvXd
zP4u~-iL1rB^dK?M_!;*;Iafbg#Qq5K&0kLRXKyjUN)>Of7?7O|P|F;?NzLgS97YfE
zO8Rf8*L>qncw0W!_CCyJ=e#M=oBLKRjL=-09^zW=j51sBWP3wwBHvsm-)tk_v?AYV
zl5Yl)Z@w~vaSiz<=esVpbH{934;?)2s)Nnt+Gx^I6O9ry@cTx6obUWjY;&q5VsEt!
zYh5i~JF6th>aA>PTPc^my^-%7E2Qv#xpd>TY|F<bvg=HVR8zxAUtA*HMwiL*O=WW4
zyi8)JmdY_^=X-L0dz=&X{nPYq4&$}4qDYSBJdrI4Ph`NZCvvvOJ;~?p?iG4JU)kl!
zh1fiCc$X`VoM}w>ctzTiZ+=#1$h7hdnKzQJlV8)tY|~9?q{<YZR+;j6&<*+cCS6Xq
zyd)F3)6?c~isZN_%WLmMvAmxk4@V|QqkrP%mq)zFadv2K=geqztT^!XJa$p6>=+y?
zfA$=dqejtk*D+d_B^{IzP6uUbS`>STqGbD=NZDGopB)C|nq8r?Vr8fpQ@7dIH&kx;
zhf1YIsA#R<FKu~l>+Tn0`~7mxb-&~;-Y<(Q;^cN)`en#BUCB2NWSjpqoI5<5d7SvZ
zXRYap>A+hzmCPmIgy@oSv{a&BlqhaI3mU1!j2yINno0_&!`v}Rl#yf`)8crUekNX$
z)*TX4=R-1TW3*KEiWWPsXfazFEekYbBxn}>(v4yziD!ddEIXBArIbvQ&$Erkb!jX+
zPGe=>+E}St7b_Dt$I7dOSefxUR`Tw~OH?8CB6@t<K2%BgGipXINutf2`|rEC&vhY5
z7G`pOWt%JwQ<FJoN~H()f_Rc|4u_>nsRebLnf#gU%8+6IWXR=v8KRndS#-x=m45$^
zrnioZdhfn?fdK?WMQlZK05K@Bb!`>9^Vo5$V|RCVP$~%02-u>6A|P#mg-8lWh$yzA
zG<QAsckdsc*9Z(aGsnZcziY3(RzMtr3gZ!HoPa1K;Q9OnWJM()^+N(iw@5_u!-;r%
zH4$5C67k$938njz*bAS8AVo5qEt0W(WHJsHCc_{i3CG$cV=+Gt-;)d-FMhr;_oo(d
zHnsf`ns<7P${vq#aN%S0UGWUlKfZuozn5t9;}!D)(qNTN-%ls*X(eaFqB0u=t#Xjq
zjy~fL*%(2_$>d(uHF}FzO9Ae7E5Z-oV%+s8hKcWc_$)7hFPW#4ZZ0Zr<e=lAQuqda
zz_%7<NS{`QN#vU8N6Jvec~$MCGEC`QhE2c7{clRS;$Mp6%2LFTZ%oU%bAGo3ip&y>
z9{nEudz8cP<}Z95SAn_`HR;(z!CqyB%x$BPC4&_bvPL0;sbP2FEIw^B^H0^Z<er|c
zG`YoVng04RXpg?^p3zcPZ>E>vZA%GoZ6*JhwU!d{&CCVurH^-e{%&qBwvE^eGrxoU
zV)oXs(hibU!CX;=fwXL4AZpHrVq9k^|8aM!=Ncoqan(r7$Ty#R8_Q*8{dxKt%i$Ns
zlE@t6J^z}>6waqCxF3Eon>p^kOyrOry9<p?<@-ug`FPlryIrQts9@KWp_!PHZw$#d
zgSVT>6;Cr66J{p+pO}elfte^2N?F)nDUHZCGuJ8Q?m?xjyr7iVG29ccREjV8X1JZX
z*pYAkn_(_ttJo>N+gvJdGVAY=IfoydW&CAFRWl3u(Ah#TlY8RSaw_hyqe`Qbv?bqo
zYFWu`M=Pm3Y9$jFcNROVE~0AGRWi-2C3>2*yjx+-?f`3<GSpi3eeWve%esnN=Wa5o
zw}W{6;~>ed4(x|;kgm5lucDq~V{b1{?CfRCKzruL*voF-)(o<j>3(+dfSnJkrrOES
zDR%Udvg3+p;8;6pInj>2QFaoUW+#tEJBS-OC@Z43w7J(?u1EJ4AG<zMcW!`e`NjUK
zXnJJ+G80g9DEDdT#i?R9CYfhC`DOw6CY*fJ{0;ZL<A#V9`DOw6W*hm&UpYipl5gJV
z&|6bS@60Ra^F44B7xGP^3v>H8YwF(4QB24+9ls8i!=LCAE@kgGb()^kX*RwbEQ!zQ
z8%`W7a~}+rxNv6ok#A01V*v&Eri^?uoqSV8zPXgg9)>%Em?I!!O`ais3!_=2LLo9)
zL!?z1B;gsP${!Lw7xL^eJ;WylNax`LWG(q-%&q?7vb?{j8FLprrJrcL?<>u<`^w!{
zwz5vx3E!=aZ2Hk!dgiqjld{$__917jKCPq~cLV1awV*c0KG{#)3!1}yLiR)5dBa_&
zIqWpx3}*x9Ln?fKer=>7xvkZuy}d#Xw4?WBrixtruO5XubqMmP#p1MDbZuS-Q<Hj3
zI989b%^I)=4LDNWfa7K=oFA!3F7?&EzAB<;r7B(StI927@<DepOKbXZziY^qNDa~1
zPfi%BA)Qp|&uyzN<_8s`AE73j$V1((amS!YmAw#Za`LUJY$<2PgoPS?bZX*Js4DB5
zs>vJbI;KC=q<oS>T7)RXVxGEOpq3Q(lseIK`g2k_-y-Au)}>Cgl5;U;>22ok0GE+k
z(w2Kx_GFw-%oM!juPbxBo65o=y0T}omN@lqB&QNIW$-pl$spfMU(6kK`kbc@<J^qg
zk`t`K94ie8JE0-NMsX&ksgP4OYSKYN!F()rS*6cxQu=JlxC5H?T0`bFSC{F0Y&*B8
z%d#u#VoJVADrW|em5xL-=&<jjv7}JDDQ8a4hl9GZ{JO4KM(T2|PX90GRPVXl^{I~Y
zcD1Img0rXdWSn+97UY{<<eO{cn=a%VNAgW5`KA^5CV+gi!={OhqgT0>ycB(&x;Fj9
zx>xA|XI^yk-5N6Tib9I7GIRBsLO$M7$kzgebbqUmDl%o>g0Gk{=?lyqzreoz7rZR}
zj6#FY2+}D>XU$L85>|%eeRzGE@&SfbC8&K<jFx`I2um%3%a<Z_i7P~q-+Qe4{T|*e
z-(%nLBG@#0i>d8$5p^dA3&=N)iX4>i{Y2Y08wT_~tZ4ll8s5*D)B6JNwO`_p{&S4A
zO~9M4vB>Tc%RD~zP?B%@HI2a${TP&|MZ<AdG+K6z#^c{nD9VY#3*BfmTXdDzz@eB^
z$UW#3=FBt+K=N{bxL5fhdAlE0Wn9Gab)IN+jQ!tw9(Y#84k=9!9N*}H`XLu!$9Yt4
z)dfUodg4I91=Jrpi^{IlWk$Jj?&69XOE(-H<%V_RIiDKt2F-qM%<FZ--{x-EywDY%
z&t34i!Uc1hx?*rESKM6eiu8P6%u5Zxa`Me+?oX+aX*Tf$^33HqNWSSv|M6{R@|1R=
zKc{UVLbU>MK%Ksv>;Memacc;G2Jg3OHhny6xj#<68Aracd+Uc~?R?Q=+(rC*)C<!l
zc%j6~3)Snra5wZK9(?jfE1nMDy)od2H%_(o!N-n1xVPSiU0FWZztsor$TsTReK2jW
z4|2&jC#mIp+USGABp+m7p-+cC=MS#}ksTd~m~Zq#H_!{D8HB|vf>3jhxjWyu+x6`d
zblU{O&y#t{eC?-A=X`KW2!34%LHj!q*j8{Ku{HOh+b$9vdPU;kG7?RXXt*k3(9Spp
z&sN0XVhZP2-(rx{?jbskdWcbbA7W12L!4_%UG97=w(G=Ut5Y0W*2LrFo&+@dl88P_
zk}&^D5^}4P@I5gJRVR}$cWN>UXC)(cK{9r6KRlm&bGqka9BA?cSskBX|FAT~zet0P
z=4)(jl#YYV-#{%b8{dkuG4yXX{%GW&l~xX>(~t9nGpoQa+0gIC9jeoL2#+m5^n)U(
z4JwAgiTBu?UWk=p1!$v{3+)S8P$;r+(7F@@E`Gr0sr>(BOBs@Pk#km+VcS@Cidc|!
zqCa5efDZ`!R*G!qZL0*7;Lnp1SX?Z@hv*U<`Cfvh%n+<z{T?efy~mco_c*6P%}1*g
zZ~W-n@%;?Vns4}bK@D18sDaVN8k~JmgE6(#x<B$7u85tS25m&CXf2num@lv0S`OB?
z5^s7Z2drr=^QW{IKk`ix`KCg%gS1-PL5{ewZ-;*4OXM39@{O~;frNE4kU^?O%q%gI
zQ5%e8|1~2iCf|Id|K{)xV^KWh9#yuneCy456lYX3&ajgzoOw8$Pj&jqJ*sA=a+Q2D
zfPB+}`pxlkrjkv*DQa)VeP=V7FvCo4ZZi`zFLKT;GkNof^QnC188?_o27NeJCM#tU
z`Q{q==H6LqIJcDYnf{yEZOtXl#$1MyZ~BmLdag8=@Ezuod)!<c6U-$r-CR8A!`ZH3
zA(uK>$m4AKjz5`8<2nl&rQS)dvtxV=_oyy)u@Z~9R#JDzN>oZaOF!oQeOS;{){nN9
z^Mlwee!H7AV(#DY;x6)gHuH|Rka2>$Nf$kKQEhOLsbrkpmmOsNT?Z*5+o<t$?_n=D
z``F7Zo(3N~*|o_|EYI0W;u%{B;j!c1xvi(IO!KytoxZl>bkSBUciPGK96NEo)?IW@
zImi$v2YKr3AYaZp$knQzVieL#R$GHTp4@vnK16O)rwQjgYC|q{op(cI6FtUz$v2VD
zhse}x%;?)TM0%5N_R>S#!fc2vCEq;LA0lStn<iQG2FE*!4*BL9`6krPQPNsF%7x0o
zGLd|<iF~t|e3L-FX+geenl@N;Q^_%LgXI|crks4^Nxq3C-?#@1mffcZ%RKVU7qZR!
z=s{9W?AbxTFxO$^EZJs(j!0`&dXK+AnmvRB?SlL>1JcI|QkFD8jvnW1YUlvjt{5P*
zCiWLy>;5wELqCbw+fQ;5`^ukNw(@;`7r7c@CGKi%Bp|-ETsqrY2LE8z?v~aP*{GFN
zZfPk-E-l39S#!Dloco2Ljpa~=j)Wz#v&ltE9y@ACsh7HJo65b#7R=d<QAm~+=M?Fj
zHJxq1h!=G@mr#qTM{40k#!0N@E%Q)@ZLNpJ`+7|2)PNla8(^5#fbzyF^0t+Vcq~+r
zu-+;%Vl#alx|;HYY~w&*ZktlhcJ6A(Ch8jp2WyB*Ewzm%>Y}?@!5juPW-X}7jg6|Z
zV!x^kKg#}U-cH`BDrc9dO5sISS^H8|id5A^F<nj6>CZXZU%|aIg@g`Nmx?kzhvC#!
zqPd?GM{o0S?vZnMOOJD^B~#hYNk5JyeZvD*Yl%5$YCY1mMDa#js@<E4{R1tT-I?>5
zH10<2(Ui{Qn_BYC1M*G9AWd=d){tK3HRKJsW!WMP$r!`Gr%+AolhtI-O*J{YPEEEA
z;>Sx>#crXhJbqUR&7%sju2aY&BXy~*Rmg1WROg;B=kTL8_pP;M7w1&Pjdf&fOC4Fp
zS=r~|Ix>iS<9e4o!gGj>^YgW?OwnXM;cn_oWSUj`n#pU<rMmFXTgW#N<eN)>n@Czl
z6G^$x**ve`#z0dxS(Eju)g}8Cb)#$Q5=FK;;iN9M$JC|yx;i}D^Aq;hzT<qzcV4T1
z#}eCbn7;1|Hg5fbN9<0R@y{334gQ25n&n91@uPNg^7SV?`~DG|Z<JvPb@s=%KOm^G
z1TA}&&`(i}ZMwxcHm?Yo?D5VjEkMfn0+cl=z}fLd=>Oy$GWB!OjDF46=8u3`kC5^3
zG1hH;ihY|?(c!~0<P<%_IE@$B$A+C9D-*D9cpU5>#3EQb7A_BBFnUl7=NNoH)QUmJ
zglJq^6^&}kXatl;VeRWE{IfC&{cl{s$8p^2c@~WA*T^>s{unjHADvVDP@d|IgC+D0
zFTDW2AMC|!aSmZ)sliydp{CXq+tXZeH^>$HskyWp?TRCoWD^5dtfn5*-NqFo_PX-l
zxGT`u4H0H;ycTyug}EC%+Ph(}jvJ~5yCU_n3v!ZNu=a}!<||zBXR0gQUQjD?^+kSM
z07^^)v19;!!Q>h@o{>EJ$Ss-lA1f-^^T}MEr#}Lq^(p`=xdG^&6@bq?>V*LqDb#w#
z2g08I;8D5(81vm9$}fHxM33|axd@F(UPx7X;mleuT=%(%KjfQ4vQ2BA|DSK>{PxD(
zR@7(6H!H|D2go-q$v10ub4EqJNg>}%B;RO~Zyv_^(4*{!2tJnb3-tX^zgha58dF^$
zT5AMhs#5?C@RT}}zqq5``}ZY0-w}-O<QuDDA=t1i1dZ2*;Q6Q!B<^8ev`Q#WYF$GU
z=WCeD4$)(N>>Rba0Xz1MUS)5&<L_IDunNbu&~V)85CQAG5omQa2IG6jV$Z%<9LS2p
zm&5U}xfhQmIq@(uPk_uw!Y|(>xZh909q!Q$>yU`Xi{i1-gZJgRS`~-#>_jMSA7g5>
zCm2V*agL|wYrrcQFh_FC@N}%7o(|Pj>2M_9q*ibqrJ93C>Wi((ID1cJBRo4BiJV*c
zzsW{kXf}2qrp^;lgfQPCw0v2Bb_)wou9A!S$Fku5K9hT8nOGm5f!I!IXsMQlVe$$o
z6UwlBSQ#Qs%kbyG2hLOz5mT0csizazvGo!M=M*E(kba(?C793a1%noRd{<v0@XAX}
zzxk5g^}J8aOAJ~_57Furbk}=>QHo5|*M7#L2Ige8{)*VnRq&?H?aIuFpWmBE(1Es+
zn%-7&>)Og_FM4jOd6@aJa8`R6cCx*Aer_*TgE~n5nhvs+op(LRH_AU9<guE8G;L)d
zhv~fuYi=a%MzcqJhmm}UFp_BUjpsmPIk}%Rs#s(0avF<fFB6%{8P)WY)NJmUh;@#M
zTqECv>6pqbQ&ahOG5y55OeNaQR8HSDm0q7sWo;8PnP+S!eK(m&Bj)?n+%S_V$>f?G
zGx4i6lZGZr=AbI2YBcw!4v=%4m1GHaK1V7gTgO~pbl{ANe3LZYTsEyRmm}NEMePXn
zn{(!p`i!4@Z!QNb&1L^Xb1_Xam%nM|QsQbZo^O@X@0U{i(=6B{ZYf7~ETv_orPw%I
zN&e){(zQ<)8Sl|q!hTxGX;UkS{A?wo?{*QzjV{u0U00c~vAc9uIFR=o<hqN4L<Ts>
zMD~kU+S-eojlDEw?|5l<dwCFGC#N^t$?yxdGQ!zb-koMY`AJ)LZP>Ce)K-3yaZ<g=
zIpmw(WF5C4TbbI;UNnw(mzm_7bn;Cz&cGBGn3;UVf%|m5q~qy+vg^}8**=$Zsl*{-
zkwcA!N0~81&Xal6sNKY0U^a2DA##Gg;?<V?oGI@|zB$rjh(JNV@ke@Xo;b=EFGu;;
z%~2j5<E}UPW;ppq_s3uvL%!)$JXmVTH{}_F<s0|AtjIUhso#XKUutIrJHl@d7VWEp
zMP)Ygic6W_S2##!^bj$kpXQSZ_q<!sSHrnfg{}zW?Zm7B)K$QJUy$V>A=4mJ%^{~>
z4`9#x0GW28pS+#bPj0d!Ghl3Y>3z&z4EP$=OtcYf?k43QI?KQ0^)hK8E-ji%fnIZ&
z-@Umss&6T$i<wth*n)WhEku1k{b-YPWfL>;v~Oz2)NksNH%pzlIqG81`B1@F&VO{&
z+4V>*)P?(t_9}AHjod=djax<yn(-Rpd~OXAOKXv$sK=f~^|+o`4<l+g$#WZE^r{}a
zi|b)i(10ti8Zhfj1E!}o;H@6dWOebHuP%L!)#YMr1JqpVaLud^7u(d~^R_yavE%4P
zu&SK8sw$oSVdnW<RcX1FTthxG;%(h@RcXIkReBv#l_}p;rD3p|j38J1W37<IM-<er
z)TJg*T~_zkkd13JWDdQ*hevX!d@J*FUTVsMK8>XEY$I7%qbWzOQ%g$Gl+)y&*r(de
z#9==|q?V*;Geh*4rt~A<^dR5#;~XrXKchQ&Mt{0O(sop0zC$GjwXeig-AeR0q9#2;
z)g<Grngnf7lT+!cGA5syFJa8LQ~bu>t5q1l>#JA2YVrNCn!NH=$bw$#qLQsG*Q&^t
z<eLog%?0vJC9}L@JL^bS?t|Vtt}DZCurK(YuAI9>#v$u;pTwP~1x>_gDYJ)pt24ie
zv}YE^kvqEbxudpJwP$}+JpIdLt}n+_#H)#l?7LL~)f*Mqc%=eECRSj?m<kLUUV)T<
zegl)gW5N6q9Gb(vZSK4Fd!Nrx>;eo=FTnYX0zAqoz`tc5aN)mCn6>&7v=!x0rhj5r
z0OwN|%b?UKLukMUocd6TXOl{xI=vVcYQ?ahQH15=-l0iW{{P#<Tu_@2PmKZuIK9Jo
zz6Z2=_6U=?>r#Fz3AY|6;?}YxJQ<XNDce#p`N%W0XS(emy+mBN8;6V@acF)m7CUuf
z5pgyKdc9-VI~9!u@zMA&E*k0d60d6%jkDQNc=;p>n8>|krz^1ddKoV~f}!pfh{xCb
z@z%&6L7jY|dDj~zIbP^}(gPj)p2N}O+<T%HbIH&RQNLXA_>n7;JY2DRxhvlHcZF6P
zSF}~-oau)P4w7-&4RwWzlPmtdbj80luBdFv-6?%H3@7g_Q*pzoc`iuZ@4`%67X;?I
zaCgcD|MGi>rT8Mq$ropMmL3T})$joH8W4a}_5m>N6o5Vp0&ub@03Uzy^E_9`H{V{6
z890|(kr{xw<eRDV3a62aCi3*;xoFMKhdO_lt@g+9OdphMUqr$XFJ{enp??Q299vC}
z^Sy{>+y#67*&D~ddc%hrPO`oaj*)F#$uFAZn%z7*IH$|uS-H;#Z4UY%WVsJE4E4bx
z&bd05`qMKVh<eii<c<wM{6F;ktP4QY{s6Q)8h{a}0`QEC^UN&(V^;>@b<QOWDGEm1
z&=44sZ=&0Tz>KU@Ks{&N$WTn%5{iTVp|DE{#pkb~81Hcv=fkcdM*TVtcD@evdDrnZ
z<u=@!a{HrEI35fS$Bq>b;2iP*_r^U$zx@yKjePSaE1p4M3FNc{_R=R|m2(pOJ||+d
zUp%rF#^b}XIMmT^e(!cHddJ4%hi*L9o#TwF!xIeX#(eDlukgnt9bsdbr934aZ)c@r
z`?+*<r{87?=TRxtaN--~;BZ+M)?duVj{nV3UuL70vsBA{*|1)ljbGD?u%$~8+MdbB
znEbbhW#8x52bq{SI}>+1zrta5iJsMoz{JK87@?Dd&fSynjK|w13I8SDgNM$2%<Xg^
z-Oh2pEaD~7=N2Qvqy&fKi}AhSC022l(01B$#HFO-<C;{w$xFeJr75`8_zkS9Gq7q+
z7M2GWbFZ%$sRkwJ{r)S=3Mvs2sVaw=LtdJ%E5GM3uaw<0GupP2S@z7TZ_!>RjA}2R
zPO)<*oP9kh22$FJ`{1@6rTzSlvi5LCX89XRWrCqp{5F(RZH%PHbe`2lV!h8uOurk+
z9r8_>hq0`F#9VxK?$y|lXE>i~eS(}5X(ByyO(dUu^M!oVZ{Yv&&Hf#xa^bY8Slu!e
zmG`Fdr?HuY8=1+mjb_r#o&DkDo5o3;U1gccS@O-1#`M`(D5ZX=QU=aYO7D|Oxk(R>
z<$dzaBYKa&v7@S1DMv?}%jM<dnJwnB;{bQ6T+Agn+FUZoG*daRvbtq1udka+8*g**
zpJy(+)0GlZs+5FxN_vhg#NfV#*bKIi_MYaFKgwLPme9vDgn5DEc^oYzy^)nndtxPh
zPj?X|{l{mnbe9cpyUVDp4x&T831ueEh+qdXu(FrK&Fm$!z)k`$*-0|_W+M6K82Kid
zOk+d7aXw0|=7_BvqMqYUzPUuc(IwwFl5Z@j<utl#D{CLv%JA2mV`bXO-4Z(q(6r|c
z5NBR#4q~X&Q`RQ+lY`^?%N}+d1pgc;ZR*&)UG6ATzA|&~2fGx0JIaVk?qXHZ7fkQX
z920hB8gM_n`4EX#V=wq0YBjm!nOpQ3pCj8Gca+1M93^fscfb2O%91YZTrhBysL;VO
z*mJPtIkPY86lYc^2D3+!3`DIa`Y`{@bCA2;wu6|<*Nb|My~qt4GLET8Q|9!wB;UNO
zg=pzP^11Krm_I-q?hlaMJp*K^?EtAr=5rX=PZBQn5r<#B#iUtp`BvIf9;e#L2KInu
zXpsS{tmUU;H<>!8jr>T_mlm4MWzlav*;cP7DovV;L505b%WWmgAGDC&XLM!UR9!Kg
z&J4V_nsV|wGw9xPW<sXvo1u{W?`l#rR87Lk7r%8>Wgp)oo$ORZQC^R_D>ayPw+07p
z)xasH2IKr|@wY`Cy@~aZ(0XjHtjEn8^)L*s$Fp1YXvH1Jw9RB3n+71Q9_Q!O<25yd
zeQoP7Q&ETZ^|hG&y%vp%YO%XX9g2t5A!AEDRvK`=^;Jc3>{TUifU5MQuQ{WKs#H+B
zX_c=c+n_2Ii&bUS9aSl@Rg)(HYEr1CkjWPm65c~yu3S=Q=8}fE{;MI?Zk)AsVqev2
za!k0Ulp8h@j}ZEpFVI7CN>jem=VR4NQ?77NYsU_C(cG>jTUTkx!bs{KyEsoH7bTA4
ze)uX4=|`_nMN|#GUH*fYet&Sh;RiP7|G@Y3FPONx9K-d>F+bxIx)puG-(KZd@sYjQ
zCBI-2ScQJB%*5+ni}VMay`5Ev4q0+c2K%`FYD>5pv!2N}@5whg<ePlXtTLD%6igpZ
z{}5e1PhDBYOq}rBx^h38nKf^9C5-;*f!sNJ`dL#99@4`!T|@qBtRa0~Q1kj*kMim&
zcql7yJ@gxT++?qlvK(u8&deyqc6tSZB6Bc|9vHcngKuxMvGi3oa$miH&!bc<mSkL{
zmt`Bh4Drp9v85pi{Te5uL?sz_S0!Q6*(7A~>tC*<VAanL*s|jTdKHwyq)Q1Vw58wj
zZxKFjErRBgLZt4=N8r=97~dlg=E-mA-^jt5k=a<3O+UnpWXwOrpvudM@En?iPp7!s
zyYn${It4B#<1zgqb(`IB=%x~fJsV<iuI3>e7RKOT%NR^6h{l^a(ezJ{Ynn#mmU=Ws
zCP$%>XB0;Fh=SdzE0{vQ!4Y;5qy%B@4u34I^~0zdA9$VhM)?yj+}h!R$F0v{z<xJ$
z9OeeE)^51@iMma+E83rSh20!i4C&^I9@JlAN?f2wb3xi$7xr4a;@3!5n9(=9J<}DD
ze_SzzT=SUPPwxtHP;VF5&vD_`UC`j`g8r{uFs;}Hy5yTxWSar>H1DDBC+vtXrXTc$
z_5oka+2@Ng?sD}y^M7B@;QjRdeKEz(4>QFN86*AB?M(oRc$U(GtmJ9kFAy2sxMRgx
z)=|!=*6s4gkW?Q``s{_-{k?F{&<it~dBN?eC%(mb;(LrI5~Dn^JlYd$9(v;LFmLP*
z^@jFUZ=4|a)Kz+8ah*338v9@m`DcUBU!3KMzW;e)=CBKB!dX~?lG@Uk09+;CI4lmp
z<~;!zd>{ax#{+Phd{gckfQCgun3PH1@%~_3A>TY-8iJP1soUs=uvaw%YtDz_*OO4p
zN(;sEkD)kK6N<fUFJq9+W$YYz8CDxEW7eh1usC}KdS%z}_4hTL>TwqbCq%<|Qw%P7
z$Kd_uhbRw^$AXM_l<Fp6Hj?1EI0@zllaTf#5hGW{V+FlE6QW{Kd?^;YLSmtDB^Id<
zV;LS6k4L?dFj4CXPWO9-Ri^2PBHv6Q-wY(*6p(Lb2Bo95${V!YoPoT4naE9|Ceu6{
zt39)ECO8{0C$e#DS2i-1aG!j7KGx(EQr~&U*F6tGyK`{gITQ2eX0S{B9*(Swz#Y{H
zbgl@;GBQp;{|II*@@TS0{rWu=Hn|UnF86UupA4_y?pUvvh?-P{jBCZnV@AS*+w1}z
z{~Tc+sW@PsipZHMn3nMvGqN*~(=`hXf$U);-z>jf3|0Ey;*Cr2yyPq9=U1X-46jW&
zGhaNL*Lln*4?51BgO07_33dI}%rxu2xxFBwz4RpCbZE`YV!Mv~z1NXhG#zE^AwyAl
zV<@M|H;b9;_imbzd|G8BS$mAwH)ABXCK`*eud$SoZ*KiD7Bvgz7VkEZOCcr_L%wNN
zWFo7oxIfj>RBA?;O7cQeS+Lzy?w>N1X*W!%4{&dse1mCbGH4%jaL$^^@vCOCAi+$k
zGtERvzKPIQiXr*t8u_N<0{(dqcg91N(&wI1GRZgVsKcx)XQr{zT;2^emxmk8#b7UI
zQoGH$>tZfOf#%|TiTOAI)Nnk^<sRo(Pldjn1ZqJ!{2BayRWeS8LZ!5d<owEj-v@KK
z*v4Fr>GNZAb5UJwE<;*a%8dw1xg6A4rj70@CFGlzncZc@AqN>ve{mzulUh-`x!RtZ
zjjp}yC*K5+Z{o=}p}Xj@X>TXr&fCh@Q?}A@&{iDw+ln4@brQGPN{8+2^yFE#)kbtS
z+Q`7EHZpSz`^-n%NGG1nqxkVO8+qYtBe{oc#h|vQ+_}_Cj-2i-;a_^o%AdXEc71QT
zuhmCtf*7YZXP}&4I#3j62TE5JdW8Qv%6RHFx5zi9<eLYoL!^dYp2Osu(wmO*$<tB%
zjyTGi_`%|;K@ZKBL2~N-Aklp}NPHf!Q^9hOd@yA1x8@)jJ7S<rmw~dg(?EH<Ut}P=
z6lO0KNu^(S3;E^+`KB-VhBg_wOkeS6^38+GknC0Dx2b%N>@a;k&{oP|EBgo7%KHBN
zxSy@~^|lqAKDKgpsI83VYd3F;oeVCsm4W3pGAWQA?|W^erCl3QWfx=t*(UiHvk(62
z$<jv6Mg3@V>7J`EwYJQPie+D$kCyZ?`CkS&Gjs#hWooRt1d|IQjwz(~YBkAk)qqpW
zYjH2S1_#Y+V4qfv;Ph(Dep!u98P)ieTFq<9Y79GEi-fRRxYBd{ow>H<N9fPlTo0=r
z_2{~z9t&quuc0W_?PWdG*VN+;ds+s}uEUaMbqK7gh4N=DJgET{rPpF%Wi9$U)Ioh)
zJ#KAKk<m|7<V2o|Z1|xf4yLN|rbtCftyN^RpNdR-p&}R5ROKKU$M3x=J!fi?TS*V}
zDFt;-b#V+*m%GuNzf9DSZTmH3NSKCf;0}}@^JdHyoR5vAH)jiXxcYG4oAaAL-1WBL
zo>t{X4f(J_T}H3vUKMkYCo<>GM#(&6KTTOe4vODSj(hkAeXjn2m+K!a`R@;IH2Z@u
z<rRpU`yC%!mLvZ8Cmeb62|@bhcobNUlzrb2d+ZO+n^&R7%NpERTaQh_3UN2(3~;`-
zjNY%1G1r)Z{7X~bDA}jJPFo&6)RuZnU3s!ySLV|<^O|`$L)f8U_L|;oQ%#vbZ&Tid
z2F&PO3++=iC=04V3;uUHxTFe)Pw{aX|AbD(7aSh>5!wq1@oVBc)KE()8}$y31Kz>M
zFCV2B^00!w0v}$Ro}&h@pbutkWHx$_euJNJshHI#8DX91ThLBMld>e7%ua&oQ@(d_
zSGVQtBs_LWLMwi~wPz}<XMI3c&3jBXC`JdpV#IzgLh0rr6xtM_XhJ@gIpsoqZZ5tx
z<f6sC9L!!vzm0nq9%LtD+^{4x-jIkiizGBmeu#f&gy6$RX3?p0--++ped7I5amF9c
zr~To(Dhg(tMg40KgDo$kQD_#8<yO%+^f?N1sN4Lx9|ixDQRq<r04Dkm@Xx2qs688u
zGx0$%8cBVl$Pb>cec)ns5dmB1sTt{k@!8C&+2=-2j2oUeal?;7S6sbKp4sn;2jg9F
zjB_b{&X`swbJld#1$vx2)g-teQ^gfmmbqfWeOGuDxuX8NEA+m(;%AvFK54nYy1xtF
zEpb6>R~IsX3-&#5!MbE$d?vqWlW*pcZyp}@#X#~+#s2@}8=qai&>{1L^YeZA^{HL_
z(2IPtw6`C&_9y3z_k&4#0FH6TypkT|CA@7$zIoG`e&f&nNQv@CvqS#a!+Fx9XfHJD
z;f0D;Uie<`iKpZnb23ez6`s&jdg79`2fogAho;MUtXk!c<E!1-SKto4EAFt4^MGf*
z2Xq$kdxl>?^DF0ZubmriIl3{PmOL{s0A=I&-)B((hLLYlso|Xe|M(`F`c34*AVePz
zLiwvp=zfMhSM(bnXEsjY*I=lz&vb+*eNp6_`Q)36MWKkR3WcHJWvsNnjJd-u<I$~a
zSQ>o|?g`i6`{5eeR9r(C`Q|J6<{bH^3;D*9d=nKBk0YE<b=sDQW}}kuabXfdhI2-B
zg?^l7@$ia?#p&^}7#sf(MhTH{h>gVT=tx-3jYL@EIP|23(zW>`bZYqu^O!&AyeADx
zn{*r@-~94PhYRy?{#~8{kExmHO1>GXmyInOm~VVG8#6hZGToYuAM>*Db6P&SN&)Ug
zyu+zuZ(%(-2Se9p;zh3voC?1Od#igG_LY4VU&66QGXm>+M&R#}2+VJE56dpy!^Kwj
zvD5xOo>o6Z`$ln44obkNe_leH^W9f(3hA?X36nL?p?5JAji#rvgEj>*w;!YUc?K?!
zZ_3Fx*RTH{-@Lk8j8LNzJfQybY0DRkR{p>ackY_=_txc;)N{va$ca;0V!*7A1ZG9-
zXUBu(a`x)9Z7I{~w|Ptbrt{v8^3VB>;&a4M%*Z#@zYN8?osk3#GLoZvjb!T+BPp9>
zEc<pE%P)UpNqTH72^GeYY;Gbm_n0t?%S1*#Fp;+88-r>SapnG0;z;J+kZ;;izqw4l
zS?6sk^V*onhgoJaeI@%C_L|9V7wR>a+0Q_}xk3GA;}0{LrC<(DTh6Guafj+ZrTn);
zDd&Qz+k`9SC;28XS4l6dQZ87NZ>YC;&!&b$AI{CK>~A>F&IU40=0$Uf<h*JU=T-;!
zwTZlMrz=Y7m8BFNet&Qt`>c3tmCXI~H%dAAK`EOWk!!Tf#nQxFz}eUKW#(cQVIdc;
zbrN;Xu=M(O6|dOtQoq|l(#|<Zixc!0?{SbI^35=9GR_A(+4j^<axdFS+hcZeb*-J~
zR@%y67h4%}*jD29*~*XI?0sN9a+-yWJkle}7+A}eL)~QS^3L*(eW02=<9PZ+T8Y*J
zEBO}FMXF7^F;~V;Jg(WXTGmeXWZ6lNJUi)MXeUP8->RS2OA7M)$duk9V}6RPKQ~ZP
zwdk9187whmn^Llk0sS{`KRHTb96O}WkaJEu%4zb=PV$Wx`Q|tIW<B|)<OMr0$v59E
z2FY#m%`o!KF!Ig#|Br88kZ=Ci74;AKW(s$|84D!U^xW)iEAoC3#9R7^cH=%0>)S)z
z%8jJ5(1_e>Bw@_2S;`)>qAwlgz=@8cvAd(ZTGLUw)EkKIA_FloGmwtQ+R5p5?PSv9
zw$f}yTggA!MqcDMmmTGLay(m)`+Is)Rjemp)AdBNT2JPEqF<IB2B$Z2Z-~1?21_+$
zIX%6jywt`25d9ej^k&>vlTqAxnw?RHd8kDwb(*V1)i`vXeJcM}BV%VZbWT(wJhU3h
z$ZF(x)u8%xEsk(zmG-3uA(LxxWmOG6C)Hx(yE@#bZd3E79@-n~QP!*;Te)vFFR>Q4
z?Q5~5wgv;#Yq5pr$%7jFu&crH{WVzCPDRpJ(6e)0MYc0PC$X>=J(z7iTbo^Ve`~N{
zoQg#Hs7U==6}f*=MSk{Gm7p`K(ypA|7b}Hm&=0+u%vKtyF7t)CGqb4S9A=&{Ju`b!
zHR$Q3Z@P*(e0?>gLrdoKHKIQAQ$x1D(vVNoXL6{69O104E%(VE(695sv5}-u%W)W`
zko~_^#PMVUmJO|EW_um%HvfT#(I4nnRA32bKN-WnLt{TP@1j28H1qC0Rx;O@9R&xM
zenl;>@k00KBPJ~$PrekOd#69R`lbd)$9+NS+3)y!>IV+}_ZvPVD$r)kA4E<3i$eC*
zd~U|<LuMiFdZQ@^ESY(DvkKY~UoflPN9^xZih?PhQ241FyQY4|{_t|Bekfz!*awVX
zRe<=V1$a8Y0ITN|U;vNZtOEKx3NU170ebo5W8TR;+<o^JdOV)$+!di7Kg%W?iU<51
z8l8%py^@h<oQyqRl3+!?QIK!elW(TZNkTVCLQKbG+;mCB8T$_y{P8`Gd@DkA<6_K+
zFT#ZrMR;Oggw}@n=+ZS8<LSFh^2mkv>Ks%Y%0?3%`Zx6-LGLHKqEr(QJo+J?wg@Ey
z1>@U$c3^4-;N>TO41DO1F9-cGbiY3)c+npmM*qx8Ka5H6gYl9mNLmzrg-5~RS`?lJ
zMWNp=a*jkX)Bgc9Z-t{%O)zjX2<A=w(KF2thU>iW+?l=`zK1K*-Jlcg28Zo#2;=V4
zY7IApWV>SKW%_66y_rfMjX`5q^n2ri7eOv?<jm>&Iv4EP?gG2(F31`{4)S-!`PZ&^
zRpN@LWTUT%u2}KH8J!L|qx7jWx=(OMSa%n^8Or;+_~N~ZFQQF-aj*;bx5+k^JW1r6
zx4V4-^36PcoXNB781LKJ5Ao!i-sBri@{MAUA9hdl!;z=}tnNgAZ*S)Gbq&N#Gv@u3
z_#=*d<4wM)WCnAUvlk|MdO^d+3vKE=ah!akM!pGN?ul9Co8Yb<2sn2hC!NouJNae}
z`6ixxV-n(yA(`&j(b@xAH$3pe{T#aRIc}s*bz`s_;*HKC;Otp6jy{W%rswch%M;rU
zF+=b~0QOP88Gn{@w%I{gbR-A`XM>PecnJj;gVFS1FyhOC;r%rPr!J6bl0sqmDinp}
z8?{QZ4*90u?lM+8UPiyLYdAx`DJ0)?X?Y!i_iy6T)-bF-9FD8b;pqG#921v5#LhqT
z93PLxHAzI9iAne}knDXh5rIwPv6J&Ct(W)VJo64m_jk~;^&O1Tzk~kTci=cR67I7j
zF_esBc0Ca@RwScQUNVBMAH%N^^Ke$Yz{b}vQEvPiQ=YxS9_I{%CuZSET^2f?;Wa?F
zEbQW$#&g;(3rSYlNM4<fkNw_3W5ipGVQx+UnP@C`%?}3N<DPst*1rkI<g{=k{$VGV
zO$3%7qJBf}shAamVZJfg-}WIUzki4|dt)(`-lFZT65(w~?~OhEJVklvz2zmwu`{cF
zNh%t*O+}*n6J9$dqQ5~JLanmkM#h;&{l=4gbAWuaihOg_yacnTsSdC%L1)gMFWmo)
zl9d%WlU$2ozv!8)YQVYcYRnsMBqzD2>6EJ{p<7x=+Z`=r`<YfU`BQ6Ac5NebJJDO*
zlf4XMjl}RYyQfAOi}gxl*>=iU`i2_IEb>iMiLtz<ep9m6gmW4b8T!yf9+sKN{W=p#
zGBg$4ZuH*#Ybu6YOy$u3^cycznQCGtd&xI`E6l`jx0$${;i13yeY~0Y(|^<cr<qJu
zSIU|;)LJ@|YZfcz@>Zp6IK$aekW%)=DkZH*DLb^yWe|PFbEw~Rok@?)Qfe?8n2F<T
zE*&qJ%Q&)4@CkZuIJ+88{t2RIr@5O_CX<(Lr7NW=HJlmLa@L+wN*T4GR@q9yZ>2a_
z(W67YDJD0SOy<uXZ6PBETFTUMoy3*B4qdx;m0wZa<;pGxxyTt+CizBhy@ULxZZ8|&
z*@;$)ogAOZy{U0_^3s%gL@PV-;S5TZ`ptjjn{@I`=5)@ZG;HLGv9)YJ)lF*OTF90*
z7IOQEQeGy~TTPwmN*uk^e6BZR*ugcfvsi|8mhrirWu0Re>A`*WH_Yq_>(o_3>aC?%
zg|o4_-DR|IcWHF5hgiMoA?rr=l=DN`3DLf%z_q*Vbm=bLwsx03E4oWukM7LRp%40t
zz4)KCmmBNpd0A{PALiLhj)9%vyDiN$oS&80O8;55ax~3GcKO>#I&+@)MR$>I5nW_Q
zw=PnrXDN=^7BcOXg*fu@KW$<mCmzz{y~<ol4{{FZq0bH=eKA>}FULmfOUPh-vG1cV
zH*NG~fR(;P80*WzR{D~^in#@2TZ?uT_o1%q$>=~m`R=7B0e*UtozK3c0zFA2(=^Z2
zV^^l0)Lm;vPnVutjcX*YxI^^6uBQ$;-1nKIAw@sbC6H{Rt*tKlJiVxc^f^s^<Y^sB
ztZNY+P=h|%)fjTP8cx%zah)e+HQ8o=HR|1~QL?iJfz%HAaaLtDwg!eHYp`=}4I*~b
z;P>oWeC8f`P)RN3bg0M4J9SvzzmB=8wRqa07DLNw@Tg}Ej&-epg;@<owP5FNzZy(y
zts)8ksz@*DG;>eZ!ldFa7T_;F{QQGeZU5p=d(O1TKz)50u!O7;tfeB>S5@Rz7<a2a
zsY>M`HSu;<lg|g#rN6a?bmR5Py?N|-S*sxpr#1L`u;cQzhPaoKeaJdXi!`Kvx`z1O
z*AT;_8e+oN#jq#)qJAovL!gjdOBFJ6hC=WkXJuPd<ZfsKwvMYuvkP??)Upl+JF8IK
z=nqowRv<^~Cz@oGVVdVBBwhQ6lb1hX4&UFm4a-N=lzilE&&PR<d_4ClK*Q$(Xa>K-
zzOo|37r)1@d1YAT@DcOBenbz4PZ%`t6Hbzmdezg*`@emq0aZBU$eGg}?lNV($DBsF
zxS5}ci|2FDVPhWLHsquGf&!Q?D?rob1yKL50OJ-F;12m_J=vxwPsQv4_{}K5?vVw!
z(Wd}ykL9D2TOMl2H*X5wavvre!PMaE{(Fwvh*W3}NycrfWE}sTgi%>Z=$Vj&y=#+j
z%{B>(`X^zlellv$q(W8q18(KMN3VNDXy#soqlb&|cta6J+7-d5JP-cK?C<%NgH@w)
zky$~_=x#Rd@jYSjiwyjCJ`wjV;}IecvA9(z4*d?s@{&vVwd)d&J@&@{4}Top!mfa|
z{+P_$Rm`o~H;V7`_x!MSSQL65h(heuC?o|_+o5)|bA1#t21Vg_eI$Hx!=X|fjC{W!
zta-vd24n6*z4w6qmUBqE?FPp{>NVW+_L}8}f9Mg`sdt42y~QDZWRz8|7}JCA>s2o3
z6y-vV#RZF|xS)Fv7c8`L!N}n*Xhm&jKJ#rhK6S;Rx30*2;EIGKC+s29%xvb2S@zC&
zHQpJ^COgCHhBIc9ZJu(+sfNDYtM7b}*v%J(WSh*rzVIa5?A_ptD;s^$ou9i*#@T4;
zhmW>?sO#qkn*n|}4L_s|^TVduepnnBfUVkrXy9yWpGhFT>u~3se4{+<kMrc4aVx#(
z&GEv|AD*a4@WejyP3H%mSb576U1xZrx1}dGxw^yY%z4CHxuex+cbKhr$H+kHH<i?N
z!~?HRo<jrk3deZ4Vr48n*N@JkS-<m$qc+s!wHF!}Uc@=(<3wy{j<P5J{uuh6KLx_t
zCI~YR24O1sX8GM9)DH;8yog}zrq}!88}^ky2*IT$?B;C7e88rm7@!-91+^iVTpxl)
znxVMcI25CrhvIo_rlTAUh0BL4NN5p;HIMFK_o}<_zkC;kL+?R7G#Xd`OF&!lO`kc5
z=({};J;^o;o;}3X7x$4q>kgWigh55;Hs<NxhH>lLSgm#kr!?<CrSd*za}VoqVgg>=
zN<?aWB1&36!tpDQ@Z91B)@8lGcHLKan)(tBsju<$@oU(<V5fsOXP?v&e>Xfp=gJ2d
zzAzH^>GiFCn1!$P`Iu$<7Rzd~F@f63-LL6TalZ#Gzi^CNdKWu_?xOX%aP&2gz^Yvl
z%rt($ehlUv_{E^K1@jKM2N-fE8dFoEVUZOJmFw|HRgH(!>|8Y2orj*D%u=qW_ul&%
zG80pfyCo6dmh-ujaTZL;#HY`BP_M{C4?|wRY%Yc$`DO(9W*GTqx_t@e(6c<VUkTni
zmSErL61XmS4~<s85dNeds;;~)tY{>c!n8#5j+VH!(2?!k*}tpPOw|2a%eYMj^51QC
zZ#o)D>Ifrw{;#pzJZ&uZID}{*-<*GEEEj(;H-C$XtPdsQB$>$YkL=oPFp(BCV{9eg
zJnLsF<?Br4&N1>1`6f<_I?W8`-z+ziAv?J*b;?W%Z<<M396Q9HnTbC6CWZRVMeb0U
zcT&pf5u7D0SIXjjN{R4MGK)hgpU5U7>N&G&PyL2-r|~1mH}g5C;!J7nN$ydRZJIlo
z%X)S<te^%nls=tMYBT-!DrL}IrPTK4jLM3fM14qi0@-H2QgZJpIa^aQQ-)l@pUdDr
zxyYaWqK~<3xnLpIu9niIODD-VY$XG^Z#AH*vpA=4enc&%d<S>Hm)VKoCR_RQzmLf;
z^-XOh=&y~8YGWry^zGz1=TzmKO}*K`Srfm$MPVb})NY>VTFBJa?9F<r6jy3V&&f9j
z$TuP6n<3<z8+^X0ca##oT`68W>9H<U%HxmZK6<A6cIYIwgV~3^t1~sW&NA#m7b*Ab
zBDSZyO3&S0rHDQ0Dv_4bw5_Gg*R_-r^iwa=vJ_k9I&c1BAxl45h_9)I{4lhT()N5D
z56q>;*Hi|1nu;I&;H`d`$T9Njp70LjpboNrX?rPc+K&BWZRH-j5FEC%2YLbX_1I0n
z)Vh`2Xx>T||7s~8FSL|LFPe+~Gd(%MOg2TiF1sXk<@_?{ueoZAVGs5bF}o#Ir;!--
zV!ljNBWdQve}m7oWb+FxcF^leS{A*h+$CPON>^UJ(GyetEcNGX-JQ=p;>*os^$I=F
zXMa-gdHQUbq34l8?dBg1$t+Trm?i4c<fcM4y;4ZxWi>g{Nmcq^s>9{jYMg#pg_|~2
z%xvR(Ykd`d^{mF!PSuz)yBdABR-^OTYOFm{gGDAasO((>wZ1j5<!wtcOKoyBycbmC
zg-tC^>DMCZ-#VC;*W%H_THO3qgT>A@*v<Rdn$pvwUjs|_8vWZ)4eLpNp)CA^+r9sA
zSLH7vaw>Q|T!G5`3WUG;gUZqxOpj<l;I4X9mDOY1h&pT?U5hi+cHa8eU~8JHjE+!c
z=ZQk>s?{ZpJ7HrcFpq}Z63LyXHO?AhO~x5SuF;RtkmEcS_t>Ejq#?e0HKgNI?nN1r
zgI1`^`{4>%LoZT~9PX-bQk7=ARpjcm22`5VBj1fZllpady0QwJ`tk8bSHSzhFW6t=
z4(R)jNSXH$A9+o;wJmiL+kAM8&Id~JxYt{NML`Ad)P9F<?uDp(RD>qkxzMZ1#kN*&
z@p{r*=8NSaTal011B;P9vKZ$*i*YEZ7;4N#y~vEyxoWxa%*lkNTMm5&d009(A8H%(
zv2bny>ZsYIk#Rhi6kr?qCXjrSK(?{rspsF_g?zJyeAAbF^V*~UdouH2wk;1Kg>P}K
zJ{$j0gD)cAG$Y>_ZheGI@{K`J5?&vtc5@&Jn|dUnx@!`yzfHt)jbvQj!TFSK3C7%g
z&pnSKOs4O~=TH%H78GHbO%ayd%R~IU91O_JLHicD2>6(TZg;Y=X<;^Q<YnM~UwVt<
zxbNHJArjL=QK<;QCW~McY`cWT(f;sr_s6Ma{wNoJTw{J+*S>sz*Z0FMGLElB6q@ag
z!iB&nG!BTu2A3#&Uqf&4;3yRTiNx4J^t3)<cCkDACSQBv=R9}l^gf4HX>K@s*$v&#
zy5ZMWH+*1r%?u^;g)3cQk>ZNm^cFwjj(0;RR~VJKV1mC3n$C5BlbH*C|8~Ydubg3-
z?2IpNPKZ9~1ed)|*t^Avvq2|}3Uor-GAG<Dal&6$C#)jhbolLri{gwdV_$4B_rWGh
z9}GI_gHff-6YTDb&CK6>LAFU;Pp089S>p?z-@dqR>W4Obz8$IE^d;ZyC*P=3!!aA}
zhuryo$k7SFgHZt(#9gdG%{Ze{4a6N!>Qv;Lhs(Thh<wwSeAE7!C-#$X%pOo{A>S<O
z>xl`SJkgzev-re$6qviC`zUvKt#OCdFL&Hhc;JfvIm9n_L+=kRP$9$V#JFOp!+A8h
z<%Qc(Ug+?{3k?<*adk3#EL!`bnrR>9i~R6w204d&<Sln{kBV7<-PzrFx?eC}B#|@O
zVXl!C3{UP=&HoXMrB6Z-`Zxp=l0tBIM+m%jg+PP*Rg+GH;L61iFq9i}+1put=@wqS
z4a3t%*Kr~DI!X(#W0}Pbq^!G#)-9v(jvC&wyeK^Oe24=H4>5pjGwgjBjuwYOmwZ#B
zeH(MgHyy|~-&E;WCf}_8dmr=2H;wskZ{WKGsNPS6YbJSN*CYHGn1W@AFL12ION?Fo
zlG^f1*gt&%?`2WY9~B9vG9ioS-+_?`o)d{<OH!D%lZm6%`N&p%i`li*dE7H_FNt|N
z{UWe+#a)cDxQoEWVQ4u$3`}ytd5>_+z7m1R*)dS~#GpC<?QU&|Le=I-+!-E;dx}V`
zpLZW4-o@h9zp+@<^*NsJO~!vcld<9&AE(nZZ1qlo^Wy|Kwu(cUPaF)A;xV%ieQfO>
zLG9@qG+CE{p#k*pbSr@=`DO_D=0x8T=yoo}>TRXaIm>K8w~zRuR*nPfs_<?@6<*G)
zqlc~@m-su*u}(#b7BaNta3cZdS?`;(|AF~Amj<?$D+`Roh2G*DDaOq9qQ98=e#^)=
zA=^#F>xzkVO)!z{S`%^CGL@YMrZUvpR34LWJ}oen4@XVqFZrfegI!bqa5lBfO#a!y
zop5)0jPIE-yTwd;JvI~7G%`=VnOxRX%2{eS6S^wp(o&_o+f6PZ-(2E*{v|bYc|cy7
z)6`s+cQu#M!_1}KD08u1NVZwQo$&+g5~p@^lD?ba)NVA%C`Lb(VzZXI$RwrA?a4Wp
zIkge;)B8cxaTX{=ZH`j<hACwkb)0RlsXe5Sts<2&k*re_L$9+FGmyP41<NeuYL100
z;w)^;GVZ<8Tiv4GNKV_Di_Ih(x#46ZCMERL<k`rEr#9@&w2{JQcCv!LVo&moIr%2s
z%vMfOzgbJZxj?@0C9^c9c5|A1lS00kPQHmF-^}219ZbIYL9S`MS}ATyrBsvslnrKL
z$=B=nQ0~?bF_XWJX7YmPD$gX&{6_M1>+^|OLTyZCmXEOvecn+F6FZ8jXGe*2?<lEf
zJIV}~j<T#DbE~a8O1xD^>8I={Jq?*(t=mxoH8?xs+-O|6f$YsR5T}<0@<$A$u#bUw
zd9)Q>x3;pis->JG%eb#?DIv?5lf8)fdEZ*d>R3Iw_&`s7hf{m`uZhGl|D>CxHaSXL
zvbn#!;4b?or>TqE7y4pS6mro`A@`{z95qykcQ~_FPAVj#v%18ZtIIuWbvf3D-9--@
z%ZLO$nNZhMvL=&lBH6drpdpq{nsSJ2GnrnKE|1wML|sEN)Fpk8x?rqARt`~<DX$u!
z@wFBkCUH(wSB3p8t8mYt3K}|9cpY7ZIq$1@?Ng19eX3DME$8R@YFv6;jQ|}oN2eN?
zcB#QReez9lHM>))=`F0raLZ~`^{9cHYYhwnYhg&w$l&-II8Wvr>q9jbwyVK?jT)%V
zu7X3izc{?M0*Rw4P{uu<kdqatkE_7k$O^bzuR!q0KhRrLiM#!((N3clVQ*^Ty_fpX
zkXkTw5zo%o!2LiC)*qt>dW5Q!xhte!R}Fco)Q}KU>Nedp<T7Vgbu%=iT%DQ1XEh|k
zQ$tR9Q{SNv`TADQvB*Nn1{$&;R$YwGk#9WIBr%U(pLMFD@m)n0bAC5etp;=U{6+5{
zf6%_@50<O{!OlAsIQ-%l&aW@StS2AwP3<G5zRSas?|F!%hO>pt;$E1C>C+03cayWC
zeuZ#-QHXZWa&hN%F5FgR<K2%elw@aNu6-69449wMIt#<wWMMK-AL}e6&!U&GCI=bG
znb??}iDkQTu)!k-T7UDXm*gXc{PL1~GmJ;5-^`$%<1()RCH#C1ndYZs0kj7f;75PX
zsCpORuu%a%N9AEk>06xUHRqkC*=VCppE3C+<bEoAy&q9~PsWr}Nf=JPalf31gxiUT
zh)cw**NI5aVQ&70RK$KO#!O~&=&UNjpS49evZ4r^Cl{e-lRVV&`KcwOWA)t}j4jW>
z*0(v(rUyr7el~tj&cs@y7}P(Ym*z_hhI|Qy7CjmhjDqpl#ve<j`y+I!KLW=1V|@#M
z1V;G5t0URw2jBm7{4l&_F!%AI@ZL8H2YITUqcCO%y*P%Ku&!G$`aGxa;}Yj2PdxC}
z+Z|OV?(k(N=B`*b6tGjl<%k>Ru5d%6;chVI?$j{uPPIznyvfBCy0ctS#B9ElOnQZn
zx}aw-dT_ow<8GKUbT>PrcV}n3c6NgIQ76pY;e=<aoDjU*3B&g}A@{r!DxI95H`xhi
zlup1|C!}Yd!7K7jq%m`WjeKyXoev^f_~8Fx>Mi4<T-&H`Qc;nVP^7``uDwnRb=%#A
zz1`T|x^00Ox?2UrVCYgw6%&JQR8T=hIz)Ze{XWl!_rv)cM36ycuK%@;wbt=c$sZf5
zd{M?eo=T7r_6FeAKLHrt9Duj{UM<KseaJVH$T#Q6Hy&1j?DYx6)fLooP6c9HOb|Yg
zyN++4ucKYz1|BqD$EmA<_<r0U!>9Y;=l~zg`sj`FG;bu*TWm1do7YHh`o}!ryWa)h
zhdAJvjRUT!I$&MC6iwcam=ocMW8WO{hkDh(a3@UOLe3-K+?nczz#Mn{Nc2HC`KE(>
zQ@F?vs@eg(X0r!*Gj+8UH*h`h2KJC|dbr%gEY1dN`R}x%e+c5AhhRJTrigrV^hXG;
zb|T+cgu(`)7`}px!wmVu)1g>tPaVfQ6nBT-#=PyEO`X4qys$u6rC-O0m)G%|=b-Nm
ztZSlIi@B>SL&M=*5Q$GEkvLHri9zL&`0zdw|B`RQRif~bd~>;56nb2WBEv>uz=(JZ
z+#kX2KkmF@kIujV>Nu~{VI7l!$^SmVrm>lrzWh0E{>p+w&)4vhKF5Wv_p#?T_f3dI
zI++uZ%`=vK^K4Bbx-NN+)c>C0*M-}toSTnf?Rhw!k%#_j*%*G~9a6@}AXX&?3FQ$`
zeH?-Cib!ORip4pTINZ4!hvJBM459}o)HeZ^hY~Pze*)Az60q7k0Wl&0$_o<kt|xa|
zY`l+*-ANcbGXoF!vl!OpIZA$~;)5nT61lhJQ&1}AXQd*xDIITy^8SO_#3|94m~`VA
zx<nNKbIUP{e4|Icxj?>ABj4DPZwx)Zpv3D7?)!Yf)8H>?5BY*JjcNp$F(a|02F}Jc
z2;R}atQ|Y(FS0k4`)ua3Yt+hAM|_B42M+nBbEOG;Q^+|dOvTwdoK4YRJdN5-&#q=-
znxUD%C^IpY9^++O&BSK%&E&V-*+6fx-C}cbo%)TJgudci=3;atS?2+-(~miydSfnD
zk#92T!`Y!^A*7lXqL4GCJG{o9<&4Voj)mx5Wg%X+Q`_jy8I-Q2I5nBGs+pGJ$_7i}
zLLF!ND(0sSS_)<6p;j%l6nFnHk3|h+2stKa2K_eFZB`grh)%}b*EyW*GMoG~*Ft<e
zWg&jh<C9F*=$uTR3bhbQd*}h8H>f(@OgtRHuEOW^8eKFM`}&%SA9qZ|6z+W-aM48c
zIL3ULrLh<vWh7jEjD)ENy(VwSJqbgF=KG;y(5GQSi9Vcn^b}jucY9Tj8ji*gv6Fn`
zJ+hDJ#ILDBzR4ip%pl)nl5ecZH~08`f3u;+^pAzeCf_8HZ<g}!XHcbqIL;knawP_0
zE<25id9bXP@EL0$cH8$B$7*|v8~WTUW^EvzsWSJ!rKeCsPvO_6r|7}6vRz-~wXi>+
zk-Ov@sD{?+i_ahQ#fcPsk({J2Iui875=VXU(N<rST-6s1>>DUQ#a#y{_`aq5oWZ)H
z>4uK*SJM*Pcn**`qD{Cj=s$MPEbc187IhWD+_QL=UJ!>visJ7B><r$iC_>gL@ZU~e
ztjS|Xey+UGr8Y4&gKs~U7hmrvi2vC6l6scR6G#1}pOP>hqa@0>r*TYzhA>J~6Zfd!
z=v`M9Gn<r!WFIp&Z<WQOpY);Ru(!`&Nen)vBwkauN&ciDtl8h!9>Se&)NYpD|B1Ek
zZ7@h^#YOoxy!h4%jZdwdnYY2Jq77fZx1kd`XU|M}bD}t_YHo*tAv5D9KQW{8PYin1
zj!ohGobGK{^{tK9?{*lrQtPR2hvt!X#LsI->sD$(!`qPms~LZinqYRT2?kzGxV*dx
zUxzj!dQcPMRU08IXh3UD1Fo1hV%e_-6e+htJD*w-z0OAQ9WX!8fe8~j;6x8l1GS{k
z{d^8belmCY3jvZ2tl1?ePV?_)IDI)|suZZ-D2PmE3Zz%r&$CNOgpF4gXD=%Yb)Mb_
z$T{S`u?F0W8LcGj$I$z`P(he5?|t^pZzxhDT2|PK+Jmil)u|O@-ZbO=zGisHHKRkb
z3Aaic@ND8|<n{Q7Wu7I-c~yexZ%c5HK9?gwCFrhH2D|ChO*orU)2YCe+#*;<7a^W~
z8fB{su;gn#UO48Xz%(E4dXa6&H~q;r<~;q$H>b%rL)!{*i+q!op9jMOg&5BL#ztO+
zaI7xGzS1J(+ZN-!V=?rf6=TfvV#M(5cv%e7SH<X_RSdi?M&_Gh+<sS#ijT!8t1L!U
zJ3Tk*1?;RZfW1Nin#ecD$TyY&PjHxgGmd=ID>ws#Ze?ISwS4Ud8JPAu1Ih!q`)MY#
z9L#R?@vo#8=shZXmSg!1cG%x5z@&@3kNxu&%jUjA;0g8>zbVAroI>Vz3lOn5pPqvI
zNLZJIDRb^)2-)Ofw=f)J|Lw1+Kvb9pL2Y3Wf|mv%ZAK9Kbq~Vapg`3A<m@J$`p0;7
zw7h4p?lM302=K+&iM|*x&KFC)e4+o*7j^x8@Q9kk{;M9C*~bIR@44eur8~5pU9sRE
z^We_(6VrFIcLVn(zy)LVnVtIXjJ#ydo9yVpS;8Du4`*z9>V)mPod7i_oK7Ow>~%zJ
zZ$}IdbikacQs`SqptRB+iWznox8H$tKnMIG??fNu+x;BS<ChdO?WNciBgIQ{jm{Sj
zG;>d<MU@8}D?QMSzuuVT1^Z_{*qG1fPXF^tYC4_CH{*T>;9&>7ILd*@?-mF@i$FvV
z4#d-Wfq1nh5GmxF#Iu3up%(<VAwh@_3B=T8{+RcTdo?cyq8qc{JO1{<nzQcE?dOhZ
zUEHxR$_-2Ty$04f;{F#YF4al#>W~y>m!zmLcEslcj<^?1j{EG01>>FY@PZRNr|CU<
zLeJ4_?nA%Bom+c7pi;{G)!#nobJh=T8UA=T>N*~VUx$U(O=wT2|K<SaQ+9ki`X)Yq
z3&t$n5Dd5-g5SC1SI!Ja*M^`#EfiafLxFMBY1oY;H<LX%I>C4w$>$%%?w{D}$n;^i
z^Rz&?r+Y#!#TA28-Qld~4r60?^gic9Ch$j6r9ZPp>`<N_!fgEic3_3Vy~{237KOpe
zhuJIol@-Z1)5D|iyd?@L!{d>?F9O4rQZbs^N0KxZjeAp(>5+~D<GJ6%C==O^WSp4i
zIQIBCD&zToGc^fcLK3lf0-0t~B5sjy{-VZm`oE`mcOwx2UqbMYPByZG^6;!I4|U9k
zrw@FG{~X@oks`I5uhB4j9l_q72uxfUfvkHn><p4&k%J767Rj)1x(u(U$q>QEc|42w
zw#Gsk%=*aS@JNO$kur82%TS~hf#?4u@pUsDUu8+y)guLF{ZlyOOGPsoN1kV9?^NvQ
z$(^OWQlVmyirq%cAbLJS&jW8z<(mhc{^gLn%^jf9FOZRMUV3~%Ecxap`6iKk^Mib2
zs#cA^4XV+TY}0!|4b1(kQSh}EUDNB)JD`P}s30ca?IM)c^$^qF>kDa_iP-wpM5r@I
zwI|F}+)6MN>*UNtY&SC@H8m66$C!z#nPy_$0W&e_GG`v{{Jqa+;?x3jF`Rr;M!pFM
zWj37tn=|*#g(frN&RKk433rSCpx0P|I)*w~V+GGu?w<;=5R-WA-=1k9PBA~#Bxfnk
zkZ%mZSrq*;cc|mIb5<2Y)+t_KDK2hej*4%O`C=h7sr~3K<J@P0#s9ab+*{j1Jk|PN
zCW~D>(bSaq{!jBQ#7}aBye;>5-mnloeW;aC*BO18J0y~rcOPUX+#Z_>&vV?5K9M`q
zKbeTAV<sY@!B|YKG!j+ihN3dbkb49T#VO|Poeq<6{EbBX@4=$2bf}p4VVG!NG)xS-
zI7GC`Gn=kDOvtGX5!v*xXpiV4v>x$$B;PFK_h>iMLR46hcc@)`<b2L+9DUK$oURsA
z`(oF00sD%4=vy)Ft}AvMX^X!Nv_;?E+G44Nwn$o{E25X`iGBupBBhs}crr^@G|$v!
zAG0oZZ|MpNPthb@(R-q<sGXoIp7R6`U?1cP9dUU&XHI>%i`!C1Xz|>$;O<Iu9Z_ed
zBlht)j_DzkReOkaXZW?=sf%CS3nJr=L_KMDk*(WRyql>aK3`B4(xu8`*b?q6s^D(v
zQYBGXOg(0;61690pVx9{cFX_eo5Q@$Dv@tSKav-I!3siWB0aC?c-`jp+r6)nSTb5k
z2#0QB`Be?E?v9#Rp`|Pam~em7FlFJ!IgR3R?n*MHe(_F8=(sD1h6F{i(^FBr%Ty4P
zrpXD{Lx0eB({FaW{zS#@HoQ6AitRP682PFd@1C^s8rh18JPMv|IKobs3y)hdZgdOY
zbTlLTVl$r1Z-!0(W(0G#6u74u-F`M>sedaTxwRwwSv&OTQQn;128F0LJlN5U)w7!6
z;>fuZJ#RfvG@)d76V8lnLbhQe5`Q*u&rbvXz2AVju?_5lZGe4t0|M{0U`a*`?zpvL
z(6v^0ZtXzu@D9Ef{lw&><eIfVarkLF4khqw?xmJw_7mSfv8#t(<xV%`gws+vak2al
zS_a68#UJE^(rN|KR-`00jUwBuP!_uVx#z)HS$t8Z264L#^In(IuUm@A^abbiD1&5K
z8TvSuA*k>@lAcx~=o$NM_B7&>Z6hM0DzVeI0)wmGqf}A?scH$PtSf=(mJ+mhmmsgE
z6frZ)a7a>y(8f|k-77+jbs_$J%A6_vwrb>?Oox2#(8-4r`DTR<?*-NKaj;81{^Hw@
z7UsjRwGijYH~Kg8P~uI_VGi~B!94WZnTM9Ux!iYLfK_7(P(88$Ry^iB2K2f_Q<teR
z$-}PRdAMqj2Y3BEe0!6R)%nzk$Tbx_x9OkkM!u2T@e2Q(eF5#zC-A1Xc)5QD)(%NQ
zPHZ}o?quL;YzE|?@*YzE5e`r1y{pp)tXHbU1%q;o41J7Z|Ch+>o&`&<*ZA=34UTHO
z!wF;hGY{rt(YZL}gvO!pK|JctCZc+L5<;#gq5HBs7*)*7a2@wj-wMP>UY`f<4TS2}
zK!kDb5p5HQZ)Z4@;SQy{bGZAl!Vg}je6e+dFRZM5u;8vY*4ud_McxDFhI&ALfCqG^
zc|d`llDA#S5}d<SFK~zI6<6eXxnTAs7kKY>LDEteJR9wTU+k~xQ^O8FdT4fCrUz%P
zGtPEr-us~wVwSPL=Dj02U!-q%s3Y{AIlyy-1APxt)aprae3m^rZ0&Hl)D~xYNnyOz
z0e$B>;J2m&tSY65J0-=(C@Fl&HKR{(kN7xO1Wk9vi7fha$vA(=I7_O$aOtHlZm#yj
zi!K3B;LPWMUI2{P**BUVm(U*E9Yya>AHx7_qz>d~&OAB!#+rLV2dLAVq{`e>=O7GI
z3B;))U!30Vi>EJLu<~zvj2~u?)i3Qab*L24K~nyFNU@?qiV>fs=>1KK>W%#MPY2`}
zk=qVAVpoJCoJ$?CX)rTmr<~C7$O*H_H@2C4<QYlk(Poa!oSi<aCOKoDp(~bKyJ5?5
zH+0ds36Ie?VYUAz!pS#}VyWYt4#tn}As7`Hf`jzo^d;X|Qm31y9*R5qp{NZfldW(=
z!74XA8|R6Mo4laA*$bIG$KCw!AjK1I9<B&2lfWxng4IzH)J03+q~nOEhh3rD?1seS
z9te-{#+Vpy*eeI%&~S1vcW{k;7l8N1^p}T+-~?nSsEb0g2{TRWBM>t#0^3z1IBO5b
zx;bGOx;FwE3)0{}HWd+@*d?g(5Iz$gpoQAa!|){Z_h9y_E&=(I5)j@Tk0s~gaV9hj
zTCOqJ$Ni!i-`_!nKaYmz^DsPyKa;!O!Y?2eyUs>qQA;FN-HyQRJrP(sIsy;w$dKSC
zL*!x^Ud)i;F?pstPk)}OsWOb2BE!<LGL)Lguy2`+Jq|KVjh5j;t_<m2Bk-jo3DSKD
z7@5l%nd~8k{z`!lceB@UFT1Wgcat<F!lp+&0%UR69}<TiUsB-mI1MY-KZowGw~$}Y
z{aRZK(6*}pwucMIu|=2=P=vQBpYX0rHM)~;{wl17Zm$|-&ac6JziOPIuTW814Q=k!
z?fkYHGyAuoj{A7JP3|Tx%u*9Gn$<;Ale*A7!QQn6?Ei}}5&c!kHYVg4@{JeyCWd^I
za><N+{$}EE7`?{KiBBHK8Oum^82gxubADu+By(|$I?10?=Hvs;U6>ocd)r*R<2C)u
zIcCAhH`62N$>FtsG5O|JHFX`%spQPq<u{SO;whHm@ia@J&BsWdC474_y~krXo6_gJ
z=@mb}nap53gkBbUdm5B2#IQ~lyr!CqUH;}`CZETlNfyF^8qZ&cxesCsJ90EQ$Fe1d
z{V)@a^cJn9mwIcPiRf8jB0eUv!}(tmQ8LI_oVsZwW+fO2$JIt+ak`=Sz#Q78o!sU1
zm!YURVkE|HG7@12P5F0XB5o!b3C%bop`&9i(xl9EUFsuZ=U50ievOWS=JYlHU;XAX
zpKBrcra#Xkve!ZKO`aKfr-&R#zBy#0E4FmiWv84r`>M4?e|>H4*U}c}x@wEHU9`o9
zHZ2kIn;E%3TB47fwkXuq7Hh~r2K-nv9%G&k{<@Xt6_2y3wz#`QOZY6-5<7Xq7io$0
zJW4#-8d^fJRa1Psqbc_AScYn{yGK*J((f)@bi0eKI^9KC8#`83u<wLDxru7r0m+lz
zy_0zONLj4q*|&)~9?nu~Rw;?cZ3<%CRRxjWEH9+wo5ShyqMnS?D?mYP7^f&YZYhcu
zJtfhzpOWzNWT#D^Zo(&&eZY!J;&OwM_^Vk-tk$4r)0f$2a!w}aE-#-ci2)B4sh=r|
zXZZ?Z*?Bopt1Kr*uKt65Q@BfttWdb9jsBPx_#bS+)TkEhPisL*VGFG8wnE>&73WX3
zV%N?VEU;*S^5bS$o@vHqt7a_j-Hb<u&B*K3jAtRu(CpcQZq7|uD`}!PvmH}6w*e*m
zb2l|ZI<*-ijyFMhN)z(uuy<}|Bc}JEPSf6im(*;m$Tf~04QTPLhxg?MWChi;C$#}L
zMmOX3xn{t$1;@KL!FqfnZmw&B;?AGQo6I>?L_78|<K+{~N6nu|lKY8i?>NKS_zNfA
z{lXK^Kj?Dx55A52gUMFx)CpA(=htu#s9z<$ejm{MpaN!wW$0U9is{bu!SJ~Jq#pC9
z6xZd;(D^uX)^_Ds_OJr$v_C*S=o5O>eMJ7UO88x_;C=Ob96DEm5kHFY;7Tzf)|TKr
z^Vl2Tmm-04qu4|A|J9YEn_m&OEMfmlW&tLs6rgKOK8{GqHvRGuLLN{g(-f-ZBV8pQ
zgZ|_pvmy_xd2jf^E*DN0bFu$IE~3unLhej1T0|~Zp2)=yzR&z5dp9=m=;vbgiyW+y
z<-j612cC)97$wU_*PGec#UtI#-y4yG2W{D?YhlM(?Q6``=5EW&FR&&02~^281@0Lb
zP@9Ye+>M|oNx^dFI%b4Z`%cNgkj4zWm%KpDs&f3;%f5Q{;=KFkIiA|Tgmb4X{JQcQ
zo!`BI#oV{}{45Xej>h5F={O8~7>AnAaaf%Zhf2pdr0d2bch?=*<=tX$$W5FL3`EsV
z&R@tj$z+=`Jli$};=;;6L@r`?3}-y5aei<-;fr-^eX)GD52hXQ#Go{H7(8>w%X)WA
z?&X2?6Fs=A!2^Akxx@NjcMR+8j@74Kk#)@lVaM24v&jW&bJ)u_gxO~`YB*n<@gR=<
ze&?OxIfJK*Gy29mVXbgND6jcF<~gF`hXekm-)4UqXE{G4*ex$X;$VC1-f4%QX|^!8
zvc->)4lp%!K>SZBuDy~XX}c7+Z%B~AuX(!N1@l)p;e4qRhFQ5H^Q9YN$Ty*tUf4;#
z(OTul?q>RQItL(4D*$(X`(sYEKe{ORa=$cpE@b;a{=E+Zs(cW=*cT4Z18}Wd5JGqk
zl5g&G4n%r^FQT{kB945sZ=pSz*B+6U_NY~qaDRgoTI8FYQYq(>QXH(7B7GGZr`ZAK
z`i?k6{btf#M;I46;wm+(q$N&xaFQM*^36)J&3GPndaZTsoN%S98#H>j;rK8&WNviB
zRPxPJ^_%E6>m~wDk!|R|*_d(@=U)V4ylM#Au7@BjECh+N06eDGTA`Vpg5iGHVByBj
zmrGc(?-I7^Ux78f*a_?y+Sl0`^RKvLK|fbSmq;-8mIN>FN}!uAL6fE<>h`-rR_}&)
zyF5_n<c${|-Z<{*&3#bb_{qItSAxB<|A9ARzj)(wvmY9Eh2Ye`_c-5Wc8j~uckK>E
z_2y7W4T6y~h1$}@FnA4&K>nK;Y(5l&36YT~yBmqv@JOuZtjlm@B90U#Kz%?0u1MlB
zPm>;J-!P<KjX?-?sUsEd5HK_ks=>Lal6wQw<QRBPjz;a@(XgfuXs-t~qe&6?N)2XH
z0NG}>41UyZ#*lBOkZW#`Yi{xU%ac4=hC`MzbQsD|vp|NEw`BN7CPR3d4A$>tI9r*7
zy0}Pe>=cifF^Nz-`Vcn^A7Jy3`^>r}AaHOzRy>V^d_Ww8R~$;+;-KOmhYP-G&|jT_
zp|3I_?er8S>;QVD_7q3QJcYmaTkfNOkI8eZut>88?(DlbJEsPZ=hk2W`Nqbh8Y}Fp
z5k|hz@u}u6i)tA1^?&rHRx|~$_w=@+D3mCRtbe+R*9SDkMRV?|@-Y$ZM@_}yQ>Mc9
zJoizJG!v1`P)&3&6MuV}iScBdOR;9+%y4=|hMJ3u*UZJ3ljb7k7+Hsp3+c;wbktn*
zyJ0TI9cAC}c?+S;xl|!%Mq|h~|CTWW)k>X%eDl+kOhbn8n?MiFBulY}TF!mJ9SuA$
z`0HinmhAQAeCH6oJQL|1HsyS$vxSIkq@S5gHO|>wJiBTx=3L-R%#-^*zncm_YBXz@
z2fOIZ<HBwqeh&jSm<p-3sd!RjBKlo75hvS?g-)Tdc<f~?dfYP-xoeHYr}akS*f=Ax
z{<@(!&Fr&PcSBJU-CNZEG7w)?jD*?^zW)}P$J<C4#~O)<9_GT}GW8~{K0@`gxhM!F
z1I{rOX(KGe{4_EdXKB^s9bfXz7V?c6`Q|KVcm|~gqMEwRX(jI4Am2ojRbuqC#WnKH
zG4jm@^34_UjcFryGHuloy~!<`-I%jh(G=%4YlxE@G}v#iE?$#Ojd+ZQX^6eUHN-q6
z4WY<$PLW+R3K}AxMZ|v`xx0CQx|ph>E+YD=3$-=sqLFjT+|g>nW}q5*hC5#Ss)-V<
z?qcCJ6|rhbSMg$3ClPDIp4_GEp5y+;<xDQra!%!L&E9`9PWTEXasGpXxVA(=_>gZN
zrOAs}@{LScK^Tr!6xV;!JJXZh`|+Jc-30dhpH~$3<(0*{H0~;Rqa^x&r>>*G*_5iX
zNbgV*y1`29W>yfd%jCtpA@X9~M>#R|Fn3E))2L&f+vN?jgHAu8L*ML)@6GH0ZN{y?
zTOixef}F$_h~gH6?%<4ONh>?qTA1N&Mq3QK?gml6`Q3y#)n*njH^C;j33nDU^W;ly
z=2#OH4v}xJHbHw}3nuKR-t(amM?D*%o6rcm!bbKrPzSMS#5k2kRON9Fb*BM0Pt+s!
zYCZh+)ng>j$-(t}J+8xn$|h{(pMOTV3GM?LF@c{~RNR1g=Co_6*JLx#KGcIAA-i^z
zpKnJ^PCG^~_=&xPf8kRKJCdjWf&a`uXnWj<8(xiAw5tgx)^n%J-b$QMC`WI4WNdnu
zVcCDBxVXL)`zoo${4Pa0`KFe9^MrFI+0}A9_9(}M%nH<6eZ;)<kN9Ur1=jDafL(S4
zv$qvEoKuVe+lry}wiqFvCD5k^bd7$P&CG6J{9cODUPZ{TDnw>-0S;3)+4ngghsie~
zC-bq}EFX>f`MmGV2Ri3N<rg)VZ+S3?&cn=|d9WLlhpK*g_+*xcYYDmBk-@yis9fX^
z&BcTvxp-}si*YeIxaXXME$eddYFQ4-V&0+3CmU%&)X}}NamFbd3bxtUav>Wg2eZ+k
zk&RV_ub2aW4!is8xN*&ZR#7rM?>~eq8Arn-1);hb_-}d!T5P!Q@O=i-0-oVmk7VS1
zPREdEnb_k?zs`uKa5?-6M?136cf=b^d6$jeWE;uBIQZ_2LsncI44=ipUJ?h(-EkNs
zi-GY<8TPzjZn$R<c6-oY!Wqo6C)6;=HhJ`Q^d{RRI{Tx4xj#<6<30T@KfFKdi?5YF
z__EjsBUgK(dz3qVaSoMS?2h%{-7&Y52Y&u!KQZWw_2WH$xhp1hb4AW67idwt`Mul)
z`<Zzj+t&pF>>l?2#0+!@JvYal@zlu)L3&Oosd2>X?T%1B=Lkh@NA{V}t0Ei_<ReAj
zM-ud`v`0S;dyJiKM@`WdJ=AUCv-TQR7fSJh?64<Zil8(JHc2Jex6cJ1=u?j7Ikudc
z@e-aHuGme!>G<e{Ye~L%L@i^AJUzyp0^r5#e~7s+dry4uw8I;3Q@wF(wikN+#XVVS
z%z@>2;&ERubU8{N@d@U^&$6@dGV|b1-O&F%dws|^j^vwbBkggkmp!cH_#DPaIQx+R
z<eQb$mUKr-v5uP4l|fQCDm&u6g(FHhgNu3Kh@sVvD4gklt^9r$jp5^92P_}tgoGVV
z=z590%;Xzm^3C689I$w^0}7T%vB8->qp9>;uW*3Z4hOXPJ7PO|WeRnm=oV*i!7KmY
zIl^7yh&$aJv9h}(PPbiv`>sn!dwm(3CS1lr(`z_2`5JEjmSXiYX3>fzh?J0P$Ttu9
zcUe!qS*+&B=j?__TRkvY;tfZ4Z_byzahiN{z{?xj*S#_KhBqFCdP7FW(XU~4u8BJ-
zw)$fs{oKd5GMk_05378CSZoNy`<|Skk#Xh^2!pIthOVA6Jhzvj&)6`?PYJ_HYA*Ws
zBT-kCfJxn{o1BfuBh`4!-5iGJr(-am-s#epZy`OHi$kxn5dJL&-pgYUFftk@2GM9a
z7=^Hi2n3r&;P)39HhRn8u||gISu%KzlHtFhG7S41j`Lr`(W^cjs(-?<NkxX~`ed2@
z)Nweg+8-|CeFx`Q4`rB;FGG(f5pcc|i6w*LaZNjcossc4eTer8YvNH<7KfO}ao9|z
zY4(UiAM#CudmMtO<t#~x!`O!j$jV4SSG^?Y8ziA4KMB77JwY#NE?k4zFK+k+!;Lw+
z+W7_fhrgi7_X}2XcD1o*HIm3T2go;Sx2nN%3TSj|VV7kKJpZ&})_D3UnBnNMv74B9
zNnK<eQx{!qxF?Utl_zYsy6`0T<V-LTfl@Q^E{M7DduGBt%}jLZY%bPMVXxm@&Zg+g
znY@a5@h#?J$szXSY-ir-6nknoFG{1wcPIJgD)}a=gjpi;%};qtVN1U0JeFO><2a`x
z=X{}-qeaHq!Lx#Imm6D(?28uS&1C8})NWqW=TqBiF62Hkrygc5BB&QFKVU8v+~IEX
z{$}C=mkdOwnhLof`iLZ^BA0BFev)T_sfhbYZp$+f1IcglZN_3adtJ_l7>oDOMnbyN
zNZhGp5zGoh;hff6RO|H?+fEyZygkgHao@$a0lkFU*q&myv!VDEY$z6b8i_b9b8%pg
zx!8TqQY`P*N5of}i;d)}nG=|~zt0>Q=Xj}_=Az^qH7|OJH!?&2zZqULJvXi78<207
zcIePMsx4MhpShu?Ey6o#i`Vkn;zzrd*ziV6c&yVB8UdO@kZn3`*ARBr8r)YvPN`KF
z0pIv<RiiGNKB)`;N_7$QR$btgy4d_eU6?#q7e&tMVmo&`FEV0gDPB#S*q|oL)~X2^
zkJ}3F23n>jcJs_%q$V`TIJFww<$Os+Onlf$w8V50rjFbJu)LGlNWOWa%iY-Y0M~L}
z@<0FZ)^_&NELPy2D|yk7Dlg8HZ>oOFi*FW+VyP;7huK5?kRNwvt)j3FR1|Yw(qrPG
zC?rpn#Pllm)ltJ4*`y>qIhR_!OHoXxX49*ftTaHLy~c9FeXpGOI_3{FeAqcm5AdOl
zKViGQ9c7#!#SA33=rptUiajr@S}-)D1zTpcV)&3&#F>!Yi<(gv)Qne_{FsI&_{2BC
zH@S&D_Du+^X~eCAO{iMcgt;r5pml&5qZ>`QI<pBc(i^e&QX{gqH{x4v19Q*~P$t_<
z{oa7-(GB?PYy%!!)#Ih8XWq9S3iImW_N5L3{OjQ9%|F|k%#`1VRrKTxdD?)wga#a}
zVh;`b?bZ~xq2el^|GsuuHn(A3SUX?$+1F#%0kyfmutV+-y7l~n@e>*mpkI$(_Em@)
z$X&ix<+z_vh7``AJi3)3?pP^?t>jGUeF@HzZ<5G2E##YE^3BptWl$#Lqz|dU)Pd~b
z8eW0sX%+ZBq5|p;>@q!5flXEw7*$<_cIG?JKA|rry%atCOR;cW8C()dQSDHK&y&eF
zDFqm%U%=;|&pkW&$S2>}Ezd{CD8AhzAFYjfXw0Vm5uFD|&pZrRm4`Fa*{RBV#oJ^Y
zJH0$CypxO7yL0iBY?Cw~7d{C&I2DqE-p9!`b8~R}*jwamdkfQr?~oh&7UqF(anbWF
z8ZW$sTVN(`Zp(!3o+mJR#4f)B$xyqNh?8TI;2)cae{CKjY)~>Dg{8oX`pvu18R)w<
z1O2GiB+q4Ui17ndcBI1RSq72^q~rLmRBUr%wv}_NJH?M7F2~{gfjC5LiG%XCIH+%r
zLtT6ve*YJTZr$RLe<B81T_P|!@fPG&f^hgEnT9@Fe|l{0xccGCX+NC*#9rM4{)qnN
zhgS}MxVgX&1&27#sqsPS93T9GC;r>%4)5#k$h_(TJI<dRN?nk@!4+eYT#*v#ij0%4
zFm`f*(E%4UZgzomHg`sicR|52XADhaAKz^!ym4{D$%9TXoa=;U@=a6~^HOUZ@tbTj
zsK5bR&N*OQXI{IvNa5!t!OF+>m{@0r&iZ!n+GLAYdDrl9;x)wEUB$f1<Qr2dB6(eo
z+;ExtgbPZ@HnBW=IrF+kzKP<TYGSkp7Ljje`_Qi&?u%S9+$(KgM7{BW$8sO&YWm<v
zA^9fV8_7<dP&?v*A2uFPUEzTRQ#>%BA9LN@4eG^nt(-oa1PSM%Op_m%VC*=1yzgU=
zMLPCyYqw+flmy8~B#7X1U-LwQ)OHC5>PVr~R|@wrQuGs2^w}y!w!H+SW=YWFjvX{+
z@Ou+>xG~ud13yS`Ut0>VsgAgm?tqit4hY`ofKAJ#h<zb}_G){CS5wEzw8g9pTPQxZ
z#fm^%)Ya`rQt2U_Uwa-$cAQ7(nF|<icL9_0FW}~y^LYIBA|{WzjL)`L(QD>4^r^6e
z8lU&POA>l;CD3z};3V1R5ZPw9iX)V)9AQnq8MW69>o$6z-HjcF<eO9EoB8CM<{)p3
zCf_8GZywzD#-D2HI*op)9OI9{+x)R=w?Davyqx0?cQTI3`asOr59XeA?myQH!+KM8
zD3f(8MutIR6^4qLVd$|X44dvn;>ojkl%0&nqOS4yygUq}4#q&ofqv<zx3HR)#U6!c
zkRFYJ;?Nk3w`Au|V-$S%Mj^vL0y2dNOv#`Ib6JLS8)e)jCPOA!X2tVx_9%qo10U0K
z!f}AC6Dmix(UC#54`)+kprlDM<Xg$$O)cjV=UW^9m7yU*hBfCS5j23aC`Uf8eep0@
z8IQH){8@{M!vu1TwloedyqA#Y+sWQ>kR$VGw{ovndmL_((MB4@GuIxEqZ{eB-jIgJ
zp+zv2E5;4wVkD^+V>b6-{d4jo%5^@Yq4WzbkE=$1-)f|iZ$itfvG_+dUg_6xpJ)yI
zZ?&Q$q7~=*(}OUBJ_X)?M)3Z#@6kpSx;J9uBkJ?#RK!7jO)+GOi8wXHRNS^R6Lz-j
z#(85V?!GqTK5u%C7m_<xFjHl1E*$?g7m4JXAkL<g*-?{0zS%^+`74V$PBC|jlW$`F
zFi)hy{o#Gt?KhfxHhHd)ab8llsV3W43}dE=T$8(od8nZl;_)+c(HL$nPC0QNcF<f%
z*O-fK+s%brB4=nD%tRtPdq!7rPiPW-L|&$1<0VrunQW5+Q_(opRP6X-B2+R=#I3_7
zBF@xA%zD87!&Am$O`?%-+G`}tSab&~L-F73-a=`5Z(-|hAT;{)5>ed$b$C}#ky}Ck
zZWW90m{(H0X($3*xSySjGmng;dzx9WPVAC-Z!Uaquq$zwiMX-fSj4R|7RTos3%51o
zsmYv^WzeJjko`dg24V&0Q%A`+DXltUK$DKpE7cM8%G%t~s4bp!Xo(W?jTQMO-d|Jn
zxvwE+Q@2rSR2MBj{x9FylW!`>H>RJ}#bNTz4Eky2l5eJvZ#tf-i?J^1Vm<k075QdD
zf|^*nQBAb0Q4_Ibn=qb3OUXApL&!I2^VGzu$!grwpeD}jQ4xm{xGRA9reiBR3Bz8U
zgpFz^p}1OE)YDTleTAYJMZT$9pdbd4Z;qyLMit3@lgulxZB-EeI4B9r66UCeD2gm7
ziUk7|#r$*yaqAlAR-8r2)6;X|8~ceraDVj{C6T&HQ8e-Q<JtRLr_4Q-oBrUl6Z>iU
zQr9T#z>sqtFr3C7n?>{wtFU+XOA~hsHKC0iFDZXFW4}!^4pPfe>(PR1@-48;XvSWL
zW}Kydv;T7wCWMh?#*lHgG(tI`0l_O95k-#JG`SH^$vA$esp0%-#5zeMH2XE8a$5sl
zMKxeQy}7sQ8!)e^0Wr1>FzZ{-d%JpseXYYUje0ztU5BE@b#Pc(hyE2m(Ad$4yud~*
z%xHkWN&}*u8gP<)Wo+1r4S}uLpV*4)^!e1JwIN5p9UaVE*(-OTdDJgdz4?uqGi#xB
zq!wR3RAKCeDp)m@!-~F|600&C|5A$7)AYxzDTTI0Df%)8WuIAs2=a}N0&~~on=Ry<
zY3It}A5o6eh81{eS^@p=a+H;p<FECcTP^4Nyo<2*Q4y3U72%&p#rWO31V3C#5qpK1
zaq^A-x>6WyDMC2?y{;bYgt5!Ve^c{usWlJxBl6I<Zyvs=(levXUw6)f!qhy-@6W>}
z&i6;n%tP)lvJUy?tZ5#EK^~l@<U;RN4i=MbJR@_k*ER=xw&u{k@dmpiU%>qC3q1Ji
z84}k$g@;Qf4%=qp!sSfl9nOT}@h3>#kj&2MWV}&NK*F{J9Q-R0i~JIy@;ng_`z6EZ
zVG44)XTW6;v$T^lAh-1q;<Xdu#XgWn&$vtBNGgU;X8-DfRIKV2PaceeEG-W4HgVWP
zw%KDH2Wy_z^?dEEiotvJ7}%smLBBE_g-at5GKxJYyMiz^ki9U4zWjekP2xMdc2D}C
zv4=0(V}0?3^PA*NekkqfhpfH6`1I8WkEz{EmUrj=ZFh|L&jowtx!~g-7pzEh!J`4L
z*tnFMjkPPrpKyWkWM-yTyWq|m?!h$V-tXPq2|kCN!zoVaOHWCaqZ5v>!|%~7?o-fr
z!tK+J(8zPZ>H7}Y5$u3liyaV=E5)%<Ql##az%Ild_utvU?V~NWSlZ&)&1+<nYmn`^
z3LB>@Fetf<Uo$V`+~mu+v5Why%GmcuzKO7;UzfQgvnj4<j`Bd@2QPfs?2F4geYvN=
z7kfH<ka6Azw&Q(Z`I|aKDQ7NEIA?M2MAktMB<`eVdYK1zc6q>?{^<$b*gxFtj-`1L
zbf!<`XugE^;u371Awl2%_Nbu7v^v%fKi=A4dYTRXj=P4@%C@+@%pMDjBxs)|!G~c|
z7>$-f#ae>k2X+|CkKez;4hkFWkhI<o8Kdp6*vbyyb0n}HAce4!;>cXixH$8glp(>0
zf9$bul|8D-Hz%IhqTs$Q&WGAUqiR3AZ?9*k^#&B_Y=mW>jj$N85$WSMqU67A@HxE+
z_e!^*rR)g4zdw!7W6!{O)fxDzT|pPkD{#=cg7QYPREq=VD^j=V%x)o7M~q(T$h@-~
z%sCeuW9@-wCp@6%=>hk7-cTan1d?xzBl(!kxz#ssD3fn$$u~OW8<1~CkZ)|Vxoc}&
zAj;RV8@fj@tjRYXI`rb`hM|s(^EW${w~=ou$u|?N!_f0iB*y5oPlt@2WE+dY|I7UC
z<p1NyKs-~<z(>93@a!JL9VXGZJ0TkX#zaB=P85{4L?P8a0<%8Iu=2VLBTvatGFygI
zTGVmghhzQIa7@Su$Kq$<FwYLhh^lZH{0hfy{(iZ+4CW(b$ek#I4rf^^N9e=xpq}GE
z&8J^DO8<+*<Noo8olj2L6OTqc?tQ3;Lt_YMQ<Gy+vpN<A2V&8Y8H=O6<1m!-s=VMh
z%&cd2OF15w`0+Q%H}cctk-sw)KaQrMvm_0t*Jog)6YpiXrzLt&9!89;#EAAvTp9ia
z7xu8*D~;L>`Q}e8JvlvV&@s3MeJ0nycX17_U#`J1*>{}WU5A1f4KU|xePG{4^o(wV
zF<<{LfcLph-Nl(9y5c(7A?2W%h^Oz^HH%#<mgd56zPY%vl)X579JH46q8;qR`Ii|g
zdS|jZW4c7`#wnNcCGyRO7Vh8tZ6WrPZ+<dI)rUSD@8M*cQRExWs+!3<Y9{p1Fe_!F
zVj;SaZH}jyi^rqP#pmAU;<ci=cu24F`RB~8bN}b&dE6nvof4X5rotxDlzY5Q#bUC}
zGv>d~j^?bb!9+L~n}|YY$D+s{#|@bwOEDI==NgMiDMsSSZX=P{YA70ED7w2dyEMGF
z$jmkn3F8ceS#nQtbaziNQ>mwD^VJudc)A!G3Qx|lcFkknN|(JdQ<$AswGhXbnTdft
z&Be&&+*vx*L|ofpEM9U)#!~hy7R@jg)0IrcA?jW)`cYpRs3*=^(C^D0YN>7yF;J(6
z2pOR*CeUy5lYDc8e3QIVOZ1zrDdsKE5Kf#)U2jkq^3-qcQor#g-wggrb|K$PC{!0u
zpQ?-4NU{z4WmW{Liwp^wXB1~xuQ`j_r6!cts|la~$2YhBe|$5Pe3L@H@g(029nN`G
zlZtpizWL5MRh)h&@h^QiS+l9{ky)OSZF-Sy4w7x$^X0{ZDCU&+aOctn?w@?kY*jxw
zk?*f4)aWnPeM)_&r=oamtSIUrFMfxxw}rE)Ri#Se>1XcCyssn{c2g4X7buE#k?g<w
z-=D`m-l<DDr#bZ-KeE_E9MplE?5C+H<o<=IW=zj)!qb~g=pNaG{UuFE8crVH&<rP&
z7If4!qdU(h&Zf-&ZN?z&W^Bo8f~j8<0y;P140BIY7d0TGR|EPeHeko+dNgfq#BDpy
zq`uU{HlQ9BE%jJ6u>sA_4S15qJxx^&_`-~JjdnejwASHDULC%r*5TT{AM9oOfh>zU
zD67^X(d!4!<~5>=O(QZ-Heg+LJ$&BOBg(G@z0F$@b-oon@3tbbr4@Cj+t^{-2BFc8
z7qV6;&uBvWupiLTuf?>7-=UpW1>J8S(Oy-K7(4djjVyz9Rw?2SmEy(XQta<2!9wac
zK~G9><3<S@sNbw7-=wLO;nm@C*xHvPBD5TuH_DOaT8_?Z*=zHs93r&{an8&~tuBJ3
zqYygx3lUYxtdd0uhFvSg2b)q1;hqGWRA$ECF$dL!+3#1ms7=g8cyuoMR^(#Gr(AZ7
z(q|)|hasKvFxxB-`-bGfcwZjY9L+<Yad~({2HHTrsqB@94y`=wwBYBHZLF^6K<;D?
z>J;AK!`tWVL!%BoJqhQhB;neGB!u022<3^%cs(~6lb0l8=c;6kT$zk3iU}xTCUu2o
z6sGQv!o-GXJh`5T(*DV~_$mckOPHh8%0O_>blh8(0IlwcxW?S-BhOUCJ*NM%-$Tqc
zNQDybZ4dvBgDUx^j5DdJ<eHdu)XLY!VR30J63H~Tw}zwt%5c2mLfQV;@8K(7zl-#*
zW6tsb#M5VTo!z@1G(CYU?r^Yj2Xx$V@{JpmkGUa=T8zA}E0$`x!P4Cg)@5!eKjp?g
z9v58Vyh)Axa@fxWCsbVVWsWQQc>QlS<bvX1%tZfo#)wR3Y+!F)zwOTSIymFPV(w75
zM8E1ZM`S8E;_)NqrF|TbB5^=epaX^k(RV}bX4pXX+}xGoX;&%yXG@@V&K_F(sBJ8x
zR`KB)dLOxlz}l-=yznY^9lV0Csh44Gei;u;F2ioC3*5;!(<&UXmAX*x+w45{_dp8y
z#&*9K6yJNnW|J@S2GKtg>w}>?e9+d*2gSd<;aud6BTv~`BlX0hy&j0&=7FgTJpg#1
zgnaYIfVu8HoH_lH;6$wiTRu{!;atb)KM77V<27G}8qQri_PE#}(!~ZJzS`jR&Rvj^
zUwY-)W0&CUXSxIj`bn{P12rA;&6F5BY~$}Mk#ADTH;(3Zn7Y6Yk0K<Hk)iY;McOPW
zs>nFi)K<*4*~3}Y4j0KbFR0yEN0M<~?T5Iq9w(P>z|x-^FmaSMZtk!~?>*LdeZU$^
z_iTb{$R;c)*aE-(C$N%yqc!dfqW&M>{MNXFndF-?@{Kq7W;Xf8qLU-M$v9(fOK^kF
zg=P+T@@R!Q;u(7z=lzd!M|j`>Pdd+;!_2JFpL1K$2fyQ(Ph;N8{V_dFS^lsd!@cOM
z1L*}1#$fhJPUr}RDm#@2r3T~sk6?Tr76$vtVf3_zLGLDKQ8y!@q`|EDXlfvr*`s_X
z6q7c_z&@D1CzCWxni@l0HVXGOBC%2}5`~CF!-z=kW{5ylfegdRC0nT7blWe(@tHE@
zsmtJ05{`#t8%atywxoyS@{4e+Dh)?mBYn@^sMVOtpgdfLnf&=!v7A2SD>9^9lEGL?
z&gl`3r)T(_$Tv!T<8jwA9zoP*n)!35(mR%YII+05JeFC@So9#@=o-dx_C_t0eA72I
z4nB(UxI(^(q5cy_zR6h^k2v1nOyFxk>)8YZB_}}YmX22qkCEt3uiuVJ1j^B$!;aB_
zv`VPTn6aR5(e+p*1`e%(&9oXE-bnAxW_ou5YcQLPv&yCpKY6damEOvbLA)>J{ce(F
z6D+qfHyhAhm`%|YUQVXs`Fb;v{me}4W_G+~wz=pr(_Hl7EXLK?Tzur2ZNWp0rOOd!
zr{b9JCf_uWZ&o!k2SvUyA>a732d7zwJvaj_#cMK5RG_)o9&au#<(rF})%3Oe;#`NC
z$K5z{q11_AqtQ%UFE$f4DP${eGx2K|c>!kPYZvYS%{3L@IU_ptAA9Y{7^ieh#dvy(
z6PXj+pKT&S*xmean~AW{WoGQ5u{gy2>fQW|#OV%0aq<zfQ)3N<)ULNU!<p31ECccH
zJ_B*;X)jT@x2H(f=_!_^>x<V%^+o1?`oigzo_J)YDXit_*Bz`Oj8ChJ4Gucu+;#(T
z=~z!uR;?#;F6s*hWqqOcR8MRT)e~{Sdg9&|J@If4{l!M?pShqZKAzSTCPy_z`Vo46
z4{3@^!?eXvGL8oM<_P&_F8OBkd<`+a3%xbun+fEbYvh{>^381W&A>8saWq?9Xs4))
zl;!H;+k7=KcM84Z<P|4AuAZhQTvw@yv#ZsFG5IE%Z?9gdCajmMiHX#2a_6Xt;ry64
z{P;fPo2~Cv#J1hkZuB~d#Wl(zStyH+B6@$vb|6Z(jjz+}#=G(x<r#m_?oXfWq(2B_
zFHMl4oET87APyW*6pyzk2;B+=aaBiAZ0e;b#_BK&MfUkLN>2RRBquzqxu5ceym)b2
zUV!t_-K*q8|53k@CD(yu({}WTZo|!89q9dU2g1*?bC0>F3M=|&CN?2;a}!1!Zo=JL
zO}Jv#4CR&0=%LMdEB|k*UTT0uwE@`<^#~kUkKI*uaQ&+u+gut@y0-yxHT5w1T93Z2
z_4u^29@bs!5nNn{4?N|k>#;ni9&d~4u>V{gE}yN#?UEm;D*pj_+d6cARR=}CI!qq;
z1J`=}z{XFt`24vRhTm$TQTQE-HQZ@Qj!GNQfS!_i%$-&blQjCEyR>5Au2yXFZpEq3
z%vVj}-&JEP7E@<(;x2{V>+5l~-w!0e{*I$vzhkOT6~vHAEUGTY0P>B>url1G-=;Oa
z2%E=bvt!~Nx;mENE;-~=PzlOxN)ST6(dtr$@jJ>9d!d|jy>iTPFUR#`<uKJL$C=+n
z`2LU`eD+0%nq9;Wx<Y*ThyGz1Gf~6yk=B+6-4A)NURI35vx?BvzXZ+Hgu9bVj1K1D
zX9VX<DfB3&<sz~<7YlyoqCYhpBl63J!PIL;va6Jg(?~7nIC;m<I1lzc^We&lImwUB
ztIx&Quesb;{t73{o})WEIu?wNfLgu`UR8XZG>L%8-Uytsjlj4u$yl{88868=F`we;
z`%FNkEE3zIBC+9FBxYzuL*^R;k8^yzGf6?s&@^=WoQ7J<1lZ3{K(<OE)-Ab@<9|JX
zeP|*s6+gg%>iguS`|M>-!ndP|SZGDvhHSHVRUFjXVzJ$a+WYo!OefpOEe?mp%y5V+
zcVHUM43#f;DAoJnuU9_U#JNl|HHp_J+_ANvJMMRP$A&v@%#pdm!N3*SJzX(X*$v_3
zoq#kqWN&dp{C*dxYq?<RNOBBkQTbKQ_?x-!W@8tG%y5APALrI__cwFk-yOM!VVyJ1
zymP|wc~0Do;K*61BM!>(8k*{WN;d~gq_@P@(E$U=IZwAaK(ibBiEX7QsiHPwDM9FR
zdnjxn19Y~<gurXqI{zAeW?e<lsH<@Jei^M(FT<ko67-&2!a4GdIr(M~`6i^&5o+X{
z&Xz9N9PGjVK@XJad16?$2M)dQK<_MOtzI(!waWuWoJFPo>j9Hv9*`XMz>Pf~7_-d-
zQIp*9+{6k0G?HuXNpYV!sJ?$BxJABMPrivF-z+8HgsouKNYNglA$Is{rVZ3D+2DD(
z4g7cSq91D)l5gz7<A7b5RBQw9@T*8wq@Huj7Uz^~q1$#1$EMpNz}gmmCU)37&JHd2
zIQM!WK_9Y|0{O-!Q-Z6u%xRh0;a1LN{4_p<_YDWJq~`%7hHgaSne_->v;nU#ZbYwf
z*03hu=#y`r9<oNa-J7`kdJ|GFZ9$XmHU#FLMh+Qgm%<s$Uv~zb{y9Uv>I@psoZ%j~
zGw?D#i$!K<v2orxbj!Vr`FpP*zwcG#_PYwZ(yJ)ec0z`#3wrf)!Swzv=rP>|rt8SI
zPne_Q->qJr2Z~BOaCxgI?p`EQj_}4tTOUM*1z~jO>sUJf2Ku(%!25wWas0$hO!2=7
z^LO+~D+got;7I5?MZ!ll3j60rp>skM;?{*g{`Xxhu@8i6r!?qIiovnjQP5M3gl%vH
z4vvVxAI@HfzmYM^F2lPMGPK#yJN%Cfnvg-S6FG--Ddh*@c#ss1FUe&7OtMa4IM&dI
z(~W#HpmR91Cy{k7grjyfvsTn^#*%Mjt@rS=TR4o!H^bDTU_rLg>J<-__tcg$VqmEq
zi~NzXs9YM$o~l?34~S(qC_QhR;;^2KGx#0vIpoPR`tgtsjEAXJJicv?hpJ~h7QdmV
z=X3({?z4aAdjgUVrX#BR6Lt?|VroSWW;N#^OCuMb2Ij&>KM&qI>@VL}38R@coKe=m
zm3*^;`}dYE{Dwoh-;uPn4uN(JnBBh-DHG{QFloX|UPFTXnYSU|xY55mK!u!flb)PR
zGZ9DK;xqLXefn{%$u>KvvrINL7heaNiy~&K&a7fqn|!nJ1M^d5?5T-mFJCZsO7Z=L
zg1Z;)n{zKVJ;LOh=dt|UY;$q1!d%SF;!H|rE-uJ3kJVzv8IhTAq-Jy=idu<;S+0#{
zV*YS5QSsYU418fKn!M??*<&jHn_(*Sx|oVj)NX!$HW6>LOvL;fCSutp6ERxDL{uah
zi^F+F;`S0FG2w@y==Q`=6mn1L9pBz!Sl8ZS_A>*q=#+tI=wcvzyY&)VjC%^@41Lj$
z%bJq@))zxE^hD)YJ+Vw(Q!H)M5LG$qqW2p$(TSWen|bR??8m#A-d!}csEWshsv`A?
zDzzwfz>p88Rd*M|#;S`2^!v8`tuB0KuzzMC_cI;V6g~$vMNMaIA$`F82rIP2lLeaM
z@oerrn9N-V^bdo4v!8sU{asyLB;TY`yBYW|eZ#BBHq>GkPgN8D2sNR@=aERhF(SL<
z(`TbezOmzxkZ%@{Z}yRImht2N{##9CO;;0-{kw@}B`RVi+2#=aHY*2n|C5uvsJ8lr
zzqkCv{h_U}D{020mo1$AbKY{Vm7c=isMy;I+2S_zi0y#-mtR<(^#`(L>~pdBgRlMn
zAd^0t|Mm2BQI`{MyxI|Pjr%Z7+VNLM8@hjEN8i3Sgxa^^WflAMDu1Hra~r<dvByuL
z10x#Sv2z9ez0?x5k{Z!jiSr)ICfIV`WKLhqOL}60-5c=!M*}p-7WYq3qxoHr4=3w!
z%d8&5skfA**WuW?dIXr$AH1cW8UXz*YwK`h37LkoCXM(ylwGOA#*uXhD*XYA&2{LK
z_yb81KQRB%4=gMAf%VJku<k$|7J1fUKyxi_=GS8SlUlf@)na@`Eeu@0L%jHg*G*sX
zm8^8U@CR03|A7mUEtt`x6+Ww4xx=Lue}8OcmuDMXp8l`9r<rpX&eEpV<H6BdTsZR`
zzW@8#J6(l2`jsg5`2fHBWxUpwq2Nz0Gafl8=#`C4DQ^+J{|&;ry(aHvK{qoCom2|3
zCbkSC*q61BO!LFJ97Ck+&LZ!8rGKYo7B!i<%v$bXR{ILQGW7UvTv~t^@%iu>n2#xB
z8~OLtS_<-jsYRG>QUq0(0u-m`p<&KjwCu~l5b8Kz;&QP&AQ!h!<-&4dF5W-P!TG^&
z5dP;i-pjwlugK>ZZS@>mO`oIY&oitGeumVm&+zheIu>qCN9x>k1gbv3zy&hA;!MBx
zZ8*l{hr>=g0_t-k(B+>9oV*Z$-C^;F&tT^neUqwpBGBp?iAVGq#AQaJwo^31uE*dX
z_c**e7!TL~;_0`J$9uB`^d;YPRZfKZ40Z^sJ%HboM7*d>gjv{qyqSL=FFqyV?7}2m
z;%nrbpRo`hV%a?tj=o;u*iG#wenmJgEaKx|;pjB;4)-kFMzLZ5cMH-}+s_Yw{CVx!
zL2vLJchr$<9yYq-j;$*O40grKuiUHf#s#~!y29n9E46+%WSn!sP<m-zK4AB-5p&<A
z&U~$B2B+E?{;lk@8ScVN5O-;Q<SvC+XH;Bq#;v8?!%)rrQnMV8u}=zk+QC|8i&5{d
zAvWt8%Knic&&Gi}I319*+yST97jti~6c>3-f2Sb%KO~)HSQKj8wH3hvL@`jBF}L0A
zx-Co$6fr3k0TBxen?_Q=!k|OCl~Pm`?8d^DZj=_`ThI6YxQ`iz8Nj`n;hgJQfpsH2
zaFsh7%TQPRXgGs`m1j_IdxqcJ1r3FcP`U01mtBsSW$FY4GLd$uH+0AW>o<F0Nwzn4
zh|JP6qHn3;EJPhOx;1BES52;}I*b4QdEvolFRV!SLh4N~<R*EcK7pRBNz~eEdZM6*
zCuGPXi#j|oeHeeHAztjv@PcfDC$0ohTSLu?oUJE@&hkW`4tM+xacB1w_fuh(kaU<~
z$Xqid=$Jugr5SdtGQ+mNX6#zu3ELOWcxvd1K_lH@+T;qexo((f?1tGI?wB;&9S#Q6
z#@um7*#mcSn>$uLaf8)LCm6?`<{oDUHQPI|dgc!10UKgVo*@cP8bC710GEduVX3Pz
zCM+_+4=WQG?>9k?iwPdOY(?a!t@sna9m`Lf!NblRL*ALAsMZ|I=2>FMKTG&J?!&K3
z`=GwY8Ufp^A@$`j+1Lj4e{3;*v>onEw1dnHJL=Cok*RtXj#H?G7H8o;%Nw&cdc)Gh
z8{hYN<JwI6|Cw*pBgqd1Ren(5@7TWU!5G;RjP=7qF?U5M9#)6qwI)4ahr{4*!CW1i
zDE<yc<F|4Q+J+<|iFtvy^pjDMk__*a5x80(%bd|5yr{ej!|AEW7)K9!*CaGtPK4c1
z=J~ZJU@kQ`i^(LdTN04Z=i|AJ39#{w!(qobjNo(OTVOnW8u3UZ<Ltdhu6acrj$RyY
z8OPz8PaG2Z#36l4ES_wNMU8DNcE5^4-oH3R6ecpCF$sMo$?$xf44u^}cp{UE9Vw|$
zP`H5uV{hQlx*O0tcmwM%-@xuqHxRh)COx6_GUjo9Q+X4+2c^MubQ*W_Y3OR2hKR&8
z<dJVacV@<OQaX}<Gk36824{ge@NUgvj{O649rFOLy7v*W>ORVOJt^!x>@|yEyR!&Y
z<eTYxDiEq#g?@h3*w?uREnDlL?eQC?oYh||sfQyoTqmEB7d@H3a@bi*$Ock}Q>!J`
zk}C|(OT<QB_Hg!K-XA$fmwZ#H#90oHZ$D-mQ)ipkd$9P)6UFQ22C}1*_c<|^TAHPU
znSVD}d?f=tr5~z3V6YfVolOa~xN2mZF}(C7BYC%QMpGdXpFeV5#Js*G*Cpb#okV!9
zA@2+!-~7=Manu&i_0kd%o3+Hrep;fqLsLwA$u5c=)WHnU5R>202gMA(80PnByQ_=E
z$EXw5RTuW`jhsflnSDl0yg9Eb3OlKajQc90<(P^%+gC;0>!vIkHxCkf<H$EA14T}|
zlJGmLBqkWJLvpy1a3<fZWxta1RRv*nz880Ey~O3N)btw53!|a(V(CsfF~U+-q&vxp
zJ%{8(dM|l#_=vo4i;@@fzR8OZLwbo{^Ato78Ap{qw%*hkXT~WC?*v6LUQbaRU)e{j
zS=d{6%%WFH(p#iy^cG{(dW)l+=@jo$5Y`3?;xk$02)Ra&$6Hrkc&wEd>Es&|^38to
z&Hv^$%g8s@yfh-;JmyZSGiNtX7WEP(WSeF8x{LW|WQ6f!DKWegb6J>)cbv16aK&aE
z`a!MB^F}Pa-2|O?O-R#hMs{)&mghBbZ_t7@bN(WK<v$Gm*AAuM9cWYUz%Z>27>wwE
z?&J=<IMM-^fDRn?ZAI>M_8xR=1zH+m*hpVBFZDdgP7W<_^J&2zvQA24Gs?!cAbUVF
zKC9J3uHg?7$TDMg|G`#+Klr_zY_s+cuG;)Ty$`RKl5zh0!Az+?)UEx-w3l_<9sEW}
zMjihS)$z5b4(r;uB}=K}>q{NxO|666ggP7;T!)Htb<mtq$IOaalqS{UWO6N10ywid
zSBoPRHMrtli#wyqN!8WRUsVGguNvIEQUk3+H5jy|hPn9FxHP;P#ha^e=wu~*)s+zM
zY7ix<W#&u`-YPWX?BZth+TRTG3(fei2j@3Q&8V8nOe*S}cNNxR|F#<Rn^A>?^ObmN
zSq!;~LdXpHg_|S4p_O@g`wibi^IAS`{C<OM_t&VI^$MHSUZMyu@Ye4+Zk~FHLl0kK
zzQ#9fKJ^W2Q<)9?=o<z;U~i@y^*htPpzERnbT|Hp;1lmrdNCi%f?i=#=m*4WafVa+
z9y0^p<F@X5{3v<{<Mw=5eR_)-d*32)Ex&H!3j`)V!N9>!U}r_|W=tN&y57S|zht!N
z$6@X982IH!W5k$fbhD4bd#~$gW&Xz6Y0T@53?~mq;pxCA+*66dM$IUA8%AMC8t3Z=
z;-I#V8b-cv?Cz5QM`{69cTL3Y!HG!GPDF`iBK}eXx$k@u4n;8gmGhfv?CW_Ql!{5@
zn^XCmd8?(7L7Dr}D;+-!(y?lM2JHMYvHn*U9u3Ms9(PoWc4fh6K6CSbW}<Iv1}yF}
zXMRN@F7Ul)P<%YS9ORm9@rcroN9d|})Vak$AtxF>-qC1z&)vbD%dqX`hc7x8QK@?o
z?b9w|k@E%Y^gE9?Q_mym);WYMJ_nZ%zKBTi#h3tgU_PZ@w~sHn&hf?ZFkf_Lj*U-&
zF9v_{r8mhJe|g^rWSaeCn)Anfv2dp^bJW-Y-q#nCCVRtiE@v}lPRyuxf=|8+<er^@
z!+|q6Y<dQ@zt7<Ea#xfNbESu!UTZTCJS5Y!*SjK;bDACtT`=F>8POWfI38e+qqFR(
zx3|aGQhV5Sb->*>_88N{5z^Gg#)R9#=8P?78ahL2D>=f*1$Vc)u!F({qj$O>kf*nm
z3-<BUxVpgXv<pI%To5bk0t+6)o-XK{;Y>fHGZy`ILR-*b3>jh#LoY}4Bj23sO$Lhh
z!~*h-HTfo#e4{Ya6ZzB@4+?b0!pD|aJKG#3!VG0RhsZa2<eS6fn;orY*s#(ZyEj|H
z^Vb<{wsggdB3JxtaK*#R4w%qnk4aVbSXyb1F4guJR?ABhdqniG!||oZsVCl#>$}ad
z-^v^_RLrRHFhS8a1B^5>z)JE>KTQKjwHe}2XCqAfY=nHya6Z|YP*Y@rm4RDvD{?FT
zCU0fGyeX{CnnGMP$E<r6(AZ|hUQ;WKx3j{R6f4-~Sz*;=YYd%agWJn(;Jw2J=Dm-j
za@%ppo<5F~Mki1jbqWXW+u=gxS!A83X8bb!>`6YDu5b-YKlx)>hd=o^7@OY*V^;-x
zT#G_5ik+on%tEp2XDB>WLvi*|D0J9~IyyK2BUaM0znfnC31p#N%tLlc#ue94ba@tz
zpskVcV-M=VH+Qk>9JTV9Nf_#wgv5i17^|F!;q?jZd`o~k-(P&1nSlB+%((H3!(*>F
z*d339(cU;5<$UIday)eT{AksP$K7>tQ2!f?hq7@{X^uum-x&PVjY08=7<e|sAVM=1
ze>cX_zZs9;yAm<SGZ70f@^V!QCbgwtN)olr@;9(uHW5)h6FKWl#9p%2wDF0^<8v{e
z?`?YVz0HxgH_@QT_bV!C7%@2wJ-4Qj!_si#9iQhR>6n_5jx|l`&{oKR>bMLXG04E6
zKKGGEA7xng`*`1z{c+Ro!%^=(t`u;_?NSC;lM0+1R*5q1ranEdhV<qd9NJ!ooT-2C
z=5sy%+bbnXzw{J}>okO_g_daHZpti3OSIFE{koo=4xFJ#Gb86Rb;rZYBtjxJSU`bu
zoB@NyDDq7SuRoCId`E7uxX;U`UV}wfvd&3<ZVfr-&lJvW=(}2I!=2RS!Qy=^?|YZr
z!%Ur3_XZ1p^}!;ij`_piC8F<hiBL_WPdz{){v4rieIdKe2TH`qGA%LXCbRpvn|i&3
zc{lBv!swKy_*kSNI+1VG6xlcWQeA9{RTnzG>SDMHwKIp*#opEG;(CReP{~yjAM;d&
zE>y)~dcuS5s&E&@?xqT5ak7)La9le`%y>Ldbec6#+~_$_WW_3plpRW9_!K3f$s9hJ
zpX`ogw?U_43L^4EFOjL(OLY8|7sbov#pPBxQE*UBd^MC6BM-=lNxXhWp4l_j^7I$U
zi#ZAMV$x80n#edg|MeC%`}znAYK`ZSaol1Qh5TGa;gLl@TLN`AG7930w1Swy<9nf(
zc;wmZ|IckQ$u)<0oL9;V-&OMBCmE+=jlAgCP0iG@Uc!x<n?lZPI><Vmxx?B<zS+dt
z%`vh~_1*5`uZxV3kZlHZbU>z{1p@{(;gMextiLy6%F{+%f7XcB%qDc_&PCIZf17tT
zVdu?eIMugsf6h*V=yq&-)Q)8nIxuuV2Ue4Dt`F^i7~g@|^&QZ+>_8#?(W`#8!2W+a
zU~)}zeIp|3n7h}?%Z2nn4Qzqds1`VfH{;>NW}M;<$ExlR{Nn#WcRh8x)Bm7d>km$L
z<85ibvANG56ng!EMerY7zgLa2v?^SRtitH+)mYxW8VQ!Q=ySCeUC2Alr)nYBS_{1a
zbqMLry_G^8ruC@9pZZ!XFRO)fJnw5;3loJ}v}D)Nb6A5Hq1ABrtVW`DH3IKe<7L+x
z=vmdE>rP${t>K=j8s{ymu~$?vf1?tw7gxgRaTy*CsligCD%@OBgLy{P_#Rb-F`cUL
z$gPaN=~CE)mZ0`eF*Vc0c$8a&XIqQ-HA~ReuN*xLOHlQIndNf7V3_;`FAF|mzruT%
zoy^CpNpErF#A}=y@Cr^oFL2BBIhuYv!)2SN7+?Gt+QyGCVGT99x_S7zJ{O%2Jb*&W
zeWa!_506?ygPS?P>>OsN=3vo?=jhh*3|9S_;aA9BO!AFn_In&oeTTxH@6fP4A8rY6
z@r4<vH;2B&w(l>{k@Et56Q4tO{1eP^<DA1d9-~S^&~`fnhlYlr+u#sP8ySM!IU!`1
z5U8#R$Lee0&~fH0-8P&I8;*1C;Rp>6N8WXQE|<OBWT0LvuH)ZT`d>dKK%IOOCt=>$
z!bIj7CBpJ*BE7pwQ1(wkS(jvtlS#p=NWNaudy(|=CJfG}Q4gDj?@!avdMF+9A7tR>
znM`&dq+;AV=E=uoz-D$9`!};tcOesdzhpqRoSFR0;_tE|0sk_nn>@fg{r}?ey`8V^
zPS?@&HIkW0(YRz6jT2J-n4NJMldPFtbK?TWD4vJlZmN&t1sq`Z@DJ)Ma;Oo$^THR+
z3GDdp!rfB=d%bV@VE-#0ycy<;Tti=6S>=ljWE#K!F;G64=9w=Nj`~9Ppf8r~W6q5U
zZ>#Y^jUl_G8eDMnl_S#C9r5Ok1D*sqU|Wg<ro4B+;z7>%LJrA4=!_eC$Q|UEj}x3A
zW9o<(o{qS(%o&CwoH1goGnDT-q29?JD+k%5bCewp9JRyIt9JN4(;k~V?XjC0l}cY*
z)VSDUpR+CCWQ(-Zw)k|)7RQg-GQ-3cTdZy2#}m2TmfmAqyxwe!x2tWTu*4SC3%Qfw
z*)-o4bf{qV<I^x3bs9h0kD&U`evBmJ1delsH8nwRPkAt#kGf*|nKmx+z#DlFd_V1u
zK07UuMaQhmw;eDi-vrG!L)LCHsHT}Axvx1I4_hE+xg{LGpJ6xI8T6Mu1DpE}V1gS4
zl4pAUwSz$&o3LKlVQZx=8lz6aarZ$~?y!KRxdk2@Sip#9R5HKTKr`G7GeP=d0}PyK
zfCd!<y#2fdlPV1HT*e5ED~w@7zL|f<1S&i}&fHy5bL@272>(_ZVVbNVwW|i`QDT75
z7Y3O9h<uc308c*y?BBcz-3m8To3;h|8b(<3Wk1R)_v7d9{jg}+kLYk~d>(0wA6p!t
z`1dRxbLaRb>MD+D`QsG3KfFH&^WPGTFPb4pS{Z`1?ICzKHWa1XLLqL2V4!g*yQ2b8
zye9zt;sVgIG!Ux$0`Z(V(w!CtVPi@VZt(e5R&yPe>cN;bB#850dI)#0KS3!8N4Ya9
zGfl+n-sB1X&L6oQgZ41aUPf^j)gu9`$T!xmahP#14im{K+s4Mx;}(Zu3UP4OA^Uua
z#pTvmpfVaWTcV++8iVQkV%V1zgWZa;_>VIqjmkK@sffo}>XZkpNr3as1l;CxdM)3#
z>@H4#R&xSgxAA+Cake~&2hPM}-^qA1osP#F`*^sKw-lJMbN4oX-hw177@dqu(kV!h
zqvuN@1?pNUF#nJa-4@PqdS@Vpd~?|_1M=P(81Xd&t;(4Q9F+-u`YgwBPJ5sK_L4h~
z5jXz{p6_`Aym*BNeC=<VU5De;DrnEE$E?j#LPxu+c-KW$7*Q8%F@gNBMoUO@R$@WE
z*+9P8GF&1C?~w>&H_lP6v5WIPwZ4Ct6Ud#?$G(HbUh>Uya*aLtCYF~M$v4A!+Yior
zTzO1+AInyWSfM*ubZ0i8TEt*cMGel`Y+my7A=DEmk(*Y&qn?=h+_kY1(dnW@DDIGm
z$;=__Do=0HS1mC<R!j6@U+9JzT0)9hH%<pMx#Q9hUfVT<Nly)NEk|9P4ptYL9_r%l
zCi=XU)kJBts#teNRg4W$6$ZwtV(1Whr`VlPAD|+Xs4xB-tt`qb2MO;LgT&F>>_3nl
zC<1;eiM8oUVxp0fcynNY=&Gh9rhFM79xmx81|IJ%eseeEHL#b+`zJ3%p}g4eSYB*>
zDJTB!krT%^%L><h?5nht6Vdc!r;u^#56X+WaCvb$RbG6ZNsb}o`0VZ@E)MN2EZ$HP
z6HAS-mYjGtKu+X+mJ@zF_B;df<;3p)<b~}-c`;*>ycjoGUiA7WC!Pe$2|4yFd|K91
z3|`Vxlxp`Bnkqd-?x3E+rcY1tr;wdeJcEAp5W!@dkL&{YNIkH6dj}SN<bH@=q^$Br
zxRGfVl5N7DaL&TZJ?}ZgxykPA%}p>dZ9-mTGZyY>hNW{G(&%|o4Qz+>$PVo5+ksK7
z?Wp7oXZolP{G7%do0%PG+}?^V`7O9FO->@;>?Gg(u3^S7ySTS&H=)axW{e!$jQkbN
z7+Tka&Bq!sE|6KlL;j#!@87u0ooNMUDIvAB2(qtb?rjaT5UQcxQN<3aDkz3kVQ4@V
z<|I{NdtDX0epllxXDB7q$yAA29NkolH0fH5F0aM3np(8;>s+SB=AK_IW?9tYxNI$M
zq;TJ4S%XB)8WdcrX76n^KJKc9oqIJVJ*dX@|7y_quNp1nsRSQha&FU)`ZZf-Pj%#$
zBG0-EwkJ!mBB2ry>`b|8TZ-s{66AT5pnPQshRT*8`C&04&lcn42>QUYijcjo2pN-$
zV12HTxmLxn82<|&{(i^cr3E;?_9L>S-Xn5PK8}rg3zd7X>3e?(gKIBvXVY`!OnL^x
zyeIgbl7sKk_s}GD3kAr8Z$moFm~Gl4F&*CQk+J!Y{dmVR*o|`=e;n`N*6#c0Hu?#w
zq90<~sXP?r=Aw9VF2*H4z!`-HC_b9cZs~j+PJfH}`fuQS`xOF`Um@YtD|G4c0-yQ*
zasR?Vl+6yrc6LOMq+TFCKNzRWgR!zl2rgQNK>2<M>K2DX=2$o$@<iB&L-tHKc6o+l
z&e?EwDTL$r>Tu>AM__8`b&UCtfQ?-eF<FycJoBj$;QPRbml7fSp6~O#lgQIaQ0>7E
z=CBkTdXa*{)>MR1Ke;L`8A2qZ{%;a4AEcKgI~mg(Q;`^-j#5wRZEmJuVj*+*Eh6yB
zJOb~xM_^DDyP`ftz&bnvyZ1yunLop~n-M6qxsEe;A~CUvoeW?7(Z2o)yr%fU!|)Ow
zmYheriZANubBZ~30l)b1xhwQVQL}O_)fZE3eQ~hb2h;BRATGiOh0Nkx^PAqJ!R+y-
zX7~u_IZkh>wR!0a19M*l&GKa@7q!IH6n_}xi!aZ7@R>7;>Ix_92yui^a^&xd12*|O
zpks#v`?nl0a+3qz=Gfy-ygiaa?6En}9wRurn9=Bf+t!YdPjJElDQ9ec>x3pFC-m8G
zk6(p$DB5p_ev9oebgvzJ1MINa*d94=Y;o)0am0Dq(mQU;3>jO5kZ<D2HyJjzC?eNP
z;F)F0&y#POw%Q__d}B?%dA!&bU&%Jl=GwxaXR{#h%(lhP+|xKq#xeYN1PAyx-kgkc
zh_j|;Hm*>+O1*QGE3|L9q0?4(Je8u?tB*ULJu}0Ir#rAJatAIn?Z8dWZ_<Ly@coWC
zR!z0Q*}0am{Bj1i_s(E%7bgt$az--wrj~P?<t=uwFSUbzxh)PxpG1MxLCoD|fjg!a
z*lS2OBHI}7_~=<+bE-K$hnnE@d;<&_Z-6Dr1~`6a3-0D^!ASCrm}ZQ~xyG2o9TkNS
z)P@-2gSin73^YQD#1KcxHS2Q>@H^W8qvJUPiZ*~&I9W$`6B2K)fULI;to(G~)Mq6E
zWK1z|tSLO!n?k|Plv!9iaPilE49%p+{lH~RICdGnwt?I;vB&BF^OZhQ=hYgFK#dSA
zUJ(MdHoi~zFBG|a4o&q6!7pl$E0+e~!A_pY0QlDhK=)7}&N3hR#k?SBun$d+Y`$e^
zFsh1!;52|)lxBBfYM%*7rzEUA#^=sgX7<&^;N0~XY`hVJqW3W{2#Q0u68fd6r3rJ7
zL+qY7R3i>Tle5+Cap=?(i%I^m)FDR0>_;@pc@{TDW9rNp3<-?E%^BpH_*nG&8H;t?
zWBn_PM>}Uj{%aFpYMg-2WSsby3GA?;#;26Ku(||DNhMGp#l7^AI5fx8^KKD^@m5i2
zbc@3NH&G}&8iP64VsY{zb$JDeX#Jds?O&;fK9Gu_VeH15l|~O*8jzocCm+&qlzh|9
z_borjH_yp8uH+kCdKUssGZA_+6X5$AozzTR$a{>vw;$s&kA#}MOS`h6V|EYE)HFoG
z1NM(U(`5FHrs%Prx?Rq3x{`6^=~?npV7EB=ru7Un0J*E$5hxLLPua~$CYj8+%i?a_
zP00=xGGrAe^37`UjT+Cn?wr-g3>LwExqG6vMvFPX8s!r4xKtwi`0@YdKnMBQhrGW0
zr$qQa;a-Zfn@d3w@%N-eT&KQSb|m{ir6j^CUrRKHXo;7*w8XQ~S|XSI69IcTzxkjc
z<hIbG-Jvd|<J5)ces$qw&3s~J-C&@asClg_io;X|3o}KGwyGGYtSSOOs0iP1&Tmqc
z#Rhgmy_}{jDl`X)3-nIis#Ov*qLsw5)!b1@DG4PpK-j6Ve><ZeyT$v7?ZXwt!4<uQ
zpRc?Kr}n1DTRE}qj+|IMN?u5Hk`oiysjaCmD+<XvUCrgh#y;|*GxPNVnIjy{V;CYY
zhS|yszXmz+O+r@b!MV?JS<&LwQ%Gm^5Pfrdh+*M9L`hf=u`-Nar}N!~a$0vWhi8jk
zPcdR$PccQmr`WSgO1xSsB?j>f(2)`jJimC#R<OV5F1s-ANr@$@^lqP)7T3r&wtqWN
z^SlLz6q~R?stJkY8%OfZAaYGN@=Y$!3+|})RW(A&o_R8>n{eYi`w><#cjjmt96a0M
zzpb5`^bQpC>cH4v?dYA_jxCAp@Z&5e>RUTTR#Kl6+=AumEzsf2CcL^4H9QN+ILl`>
zVa%dtdcvFV;$#!f_iw^~kGT^~uZIz}x|6B-?KiZJ%u<W%+#!v=U4v!4sDs&EjqLg=
zTuG_I^~+Uo^{T>xz$%P<P=)TCxmX;nK~g~ts%2`i>Lurr=WB2vqXuu9Y7n%VzHZf8
z%-vFpE5mD{{-p*XzBPEaq6Xv1Fg|OlG58v}Nm2#>50y|$t;B8aaji@%AmL8eV@m~M
z_Ew;kxtV7Y%5jTabKr3)CNC|8@skqt?o)z2+e&dTw*;XMC2(I<0!5h;JkBY`2=`*l
zA6ATeH;Q0CzX;vm79yWra~8=z5&rBuDta-ecyR$<mVSWgnfI`H#++ZbH|(~44gKrX
zyEwmqeE;Wo<MI?MHsryHd~=hT=x%|T=+6C3(W-PrC8r}IIvws8(_yTUfj)aPaPD#@
zGtaUibKpL<1wX*(pLzH(ClB{Ra<PBU1B4gf$C<tNv9%Yy)<w+0yYv>1&EMkZ%(qzD
z`z=md$5TTRh}BR1nJeLsw-)}$+DMJcCH76;4Mt{JFlQqn_^KF!OMGp1s^MH`N;sY!
z;r!-!IEM0ka|}oB@^I{&6^<J-*~@)097c1l<AqEl*^&9Es_exP%%9<YYBOittF@AF
z$0Z5v<Kpo3a4cT$jfLE+Sd3nffYNK!Ouk6OY1Jfby_*Dssma*po{S6cQ?d3KcUF2S
z_%ZT2hSt(+dp8V!v%}yl9Zt`7IBxZb;Qtq9*zSlx>dpxC*NecXjm+;)4@Q@{f%yBt
z5BFXCpnmQWEbmfJ{Eyn<Cq7UP^2Q)v>L#eWng014b>+SYzv7Lt%y_Ha;=^n~Z|r05
z<>W|jXqWh)tbtwM|9lYBlNy_M%;`&|_xrC8%*WBMr00u4objAxHgVc-SENKaqv)3d
z`rmQDg6@upN_T*xqXX7&aKLq42e?gfKuoqhiX*vm3bMxs|NrreJu?X%5VzJ5qt-ZK
zj+ztp=Q?2RD+gvd+e0bPj(thC+!fiuX`vmA_Sm6fsy!yo)W@h7w)kRw9QHG9p*GJJ
znq(CNvP#Ht&Sc0o|MuJR{+!K_Z)S6T)5XLVs+(-FW~D6($Tv55|6@E0`ItZCn*%AQ
z(P`3A{HQ#Ftp)p$H_j2Sm$_mA`6lIxD|(S{6393Do87VZfICLFxgmb38NRxjV#$Ra
zsA=2*w-sg>cg+l;zs+!Dq6JiUTi~;?C5oS(VRrc${5s```k(yyxFZ_YvKf`+n~i0h
zIgxLMnI6QQtriF;-`J3EqRBQJc-A+V<8H_f%)4fSD{~C+v0e{LYxH2zs0Zm4TQHA%
zsuz8Y==(8-t2$>keT=cq$q27Gi#a*f5N5RoIJ#dCqkqrGmA3gferN%v1h2sIxD{BK
zu>x(Eb>KQz2N!1PFdIe(^G4}lqk;}fWOQ(`Z8<6vb&z>;KaM5%VF)`Lh7G!g=*QR4
zc}ftLSO>u}I0(MY!AMXGL8v<WNiT+Ap(GUB_#B$%69nB`%#EHGfH}+q{C+S1af_Il
z&GYUZ_f<3aHF^d^X25k+4GBhmQ4mghvkzjtKjt0GM09B)*6&F`*OpkAI>jK>IR>WZ
zVqkPHhMf(uNbMBI?D2T`FONs>HF3-UAlE3yp{OAi*WSfq?}b?0C*M4L6OHy4(b)eX
z8hQ5Ar!~evVOuQJvtx1fQ7jfwiyTP4Q6S&Mk#DXW(N|@bfXPV-7<!L6$eiU&_2und
z6WE0uhcvQH&mB?dt<Q5cnpsXU)U3rY2PuYov^YHImWW!;o1**u?=>J12YYZo?Q#Py
zl{b*RGY$Q%({R-}4Ii`9;C_#1P$u+7XW}0DCU!w4-mlNZh!gCMAm1D!-xz=9EK4p6
zGY({-H@%syBktjp#XXpkZ}yUJ=96#6|I!pW^d@<jYl*YxwS;RH=Ox+_apa&x<Zwqd
zfjXQC^fg(BNW_^35;5tvM0_Kc9H+->c7sG5tdxlHg%a_eyt1iUBEqU9!lpzb?(upy
zKNmo@X{UzxEPYb<$UBz5B;wguiP*`<54_7x?Km<JyCTBLQ*)L`#C|R6o13*nI<>_w
zs4bqqo&A{-E%CcRQ_R?<DV(oqh)0t(MCVd^z`2+5&`}pgb!sA6Pfg74r6#gpsEPn*
zRk3uVs@OV7RZN#r6<f$Rp4?5{a#a@h)+&pM<eTS;%0k(FpxE}3z9@esv3ik`7}`ES
zsLUB4iXHlk`zigzsjdA)>3@nMh<x*lUZ?d3xFh-^Cp7NLiA;Yvq01SJ&3g6~DD)5s
zl|4l_L(X4fWQAO$tY{9C73J4t#e)m7BJE#KVR5F17(2APcr?Ac*yr3`ykQ2>p=I3!
z%`&3sgtVwXE-hx+NDH$`U4;Gwdbj^$j*K?Dp=Q#TJ(YQSlcdB7NQv!B*mpq2v0N@C
z^2j&Fi={-D^3KAT8sG}8PC~w|1D9HvE%&Si{gs-a-lYk(+#mfU-?)-*%E>jFj~nrk
zr!V<tcM<z1Pc+~(dE}EPJ<ps>4=tw-XLuX;Lv7f=9J?!DTaZJY&b)0+>?vu7H~m&q
zC$ys^?jQe_v|$)|=CcMfX{^|z@PXfl-)jZWs9frC2R30O=R`NN>gjuIfOK6QlB#N<
zQc%mh@me%wa0X*j3p37TCbw7P=j3Xv<m^V-wu=7fDtgPSkTkjq6PHvW&$9~E<en)X
zs^CYRjOk+X5VbCTb86tyUI|&b3b-t(M8yws&B+>M8`a?b^cq-psllfh&TKTRv0*_K
z&IDJYV|pbPmscR!yaJbIRp8~s3bc=`z~Az6#OIV_USK)yO)kaMt0hRiSAsu3IHM^l
z#)j?1u!|^x?~xJ&4*$Qs)Re?x<XRWw%ZVb`SQcT1ToLwJ7NURUFT8dB3CrZ~Xw3Qo
z$KfB*XW@GoB)-LqiPYT`zQUKZ*O=h(5_?Cyz@X*Npfu$PS{G&D<0<B}OVUs_`6j#Z
zQqg!Y1q!>95#pGRTK9Cw+NI;&+6*Lz+=lMwY`i{pAFrApKr<o_`lIqt(I*#?`yZg?
z6Eh~p-^YyLduX_O2N^2ah@E#EewLX~WhU3f;4qBo6NKsI>>VoeN8&DjOrFdcyLBMu
z#<NqgI2eniL$FC91SVM_cvu&Lld_?>cq<J5$T>?FhGWpOa8z-2(>^^Mv!{|fCWRw=
z;dR({iG-PTBtq{nJGgry|BfbNP(&gUA98Oc7YFU%v3Q>t$=A~eERu~tkt7nwA|u(`
z$Icz*yXw(n<ux)H+nMdrY@7`9C8@B=O2)wv*Rk3<9QNd!Z1T;2sbT1GDU8`kVF>yW
ziozSA2>0R7xG4lvCepK#;t!iEm!Y@Q4;vNzpkjXsd4(4-r|tqKU10uxZ*SaXr&Inh
z_9ZE?D=D4XF6-#s-b&9fwYS4-JTb}16XwjWvkviqse=!`75m`Q2Oq5UB>&iXV~#sB
z|35M__b~JIe)*v4xhvj%azgPcN3`Fu$I}b;NL1xsX{`gw_dB3?kps@iIH2H@J<1Q;
zqvj8HGIu#+d20)$Bs*N2?Ep3M%rbvFEYh~eSycz9MmXRz8ANjHG@dEhVzQ+z<__c@
zXMr73747lLWF7n`>!a@seW*Uzi3@I)XdSx?Bd6@b^gFwdp=*WGW@=1k?8bqZgV4Nt
z5ZyN(!ibND(B66o)7ot?L+U8<cqU38Md!gs(Kg{I94d}r@Z5d4V7?cw|2g6S`9|7_
znwdgZ#6Nd~BeRt^Y;uS0YByYJW`^F&Z5SH29RVFW2t2KWMLTt1zDfsP3zlQ3WI3Kx
zEXB9`OOYeH4C?Ebp~7t$Y;0CQt6>ca<2K{3<7N~@*&^YREqu-%q?W@1$@&)9sAPfh
zy)6*&&m1|yJFxtU30!Y(K>F>qd`_*wq_i~%x~d1+Iz2=@+>8@{4AC^$7+Pw^@KP|w
zL}w$UZ#P82P($>y)}wBBJ_^d`Bhqso4*1Q(k!ADoq<ucp`Ygayw*_dpw*YJ2ErgP?
zHs%^>qaj2ahl;h~dUz3feOZbfx+}0h$Qt#xE+BI!{dirjL2|_(=H!%b0qiwv3PvRP
z<`Vg)w;$&?>Y*?t-=sLQzceubHd6wiGb4aI$N)T=6M$|AfEFM7N-+pMdj!L?H#^Q%
zsoyCK!UXp~cn|eQ26r^o-D05kgPP+`F&Otd8uIEf7)Z`p7E2Ay%2*_5#lq-R43Z|t
zpoM#<tVz_=e2Re-AJdo>i#r~%kmuY+=2bLgpKy+IHyYdC#h{yfERI`nH$}er?_MnX
ztjI`r;&HGo9#6?Poyj*IE(y36nt+w*38?f*K#%ze(A*w}u>Yt<Cg0fTMqvQiru}pb
zE)0)D@UZ_oT{g#IQz-M3d7^lp^L$*Lh+!T{sI^PMir4gE@$5K|h8Hf}QQuC3M@||V
zMr7g~`DO{{H`thoNj90d<;yP86kdMGg!1St%o>-C4b0B!8Oy%Uo%fh$eh-eWInY}0
z2+`jk!I*xP89Md&=|&GW^Yo<1HxlxVk)uR>@FtsFmWY$o$6O1Sh|v)eF_er`@=zi+
z<k8Q~%Xj1z7xGOva!&6eiHPU*%nI(9IPY;G+g#x1Px1DrJc;~VT7g7tAtRl*Nfrtt
z|G1H-%;}w)FA-JB%=fEdANYMO@#7q4HwIeb%|I>oUTTVitJ$47Nkh!~Pec6upe{BZ
zRTsBsa5q(`CQ`iAM9yF}u_a$sL|#%Ax7Vu*?TM;lG5N-d`Fm!8Dk6WNiug{xnKD&b
zH1$^&DM|xHpBqY|#70RZadI=WVSv~)gB_Vp?6!>RCpsJQj8qh>b$W|C^f^5v-~8Jk
zFWhtF#L-AOvDJb*DH%B-zfe{v%JmQ~i@S;4Q@V-laovQ!W;Y>JyNP{mU4{Qg<_9;+
zh=?i~(NW!1TvF&JCa&!&8s>Esy_%&&pGGNReqBl^g-VHe!BQglij?TKTS|D4X&QQP
zX4BAF`26WCYNzvlWS%W#n(4fJtJ7KJoNB|My>0lsqz!LVn()W63D4IzL5*C|n|xEt
z*-aMtMz&KEewQ`EnDd*U+(rauHL_=rdn$Gr3@~JWz;tHwPNgUNMgz5^O*pWo3G%la
zabzQPI%Jmqf4F}l8+Du3h%cQRaert#z8?RFPyF6aH@TO3R1b^64fw9w2!Dr0bm#ZF
zpUSR<Wn>?z1}vRgkNYxpxHq5{JuS&JfwfSKs0AErk@&F&9i}zda+5k64epI1s+d<*
zg?WD~@vg3t{8Wj((p4zuZJE^Lw7jZ9<5gzM#BuUEu^N;8Ig|0JfM#<!IzKB%BC{{$
zX4If{Vhtt^s=>?3YX1GJMrE&R{PQNyOs{}hM+Fpv=`&QXfK-2W*veKQoR6>b;0|hK
zIez^v<4mdqHrgd{<L;@A{Bre95mrqpMxHb0Haj`D(I`P;6|)+=iXnTk2vG-%@bzmU
z7TzyJ)Ve}gzWW7(${)<`{0egu_H9q6AH4o4Jg+=PXigs1lSvfJZecffP~)%ABP5-L
z*?hl}DKhZOIvvj+-hf?k3hdUW;H6zMeq2aGbUyp@w9~P8TRNr;$zU&5Ha-~N#Vn~@
zI9lgo#l2h{$<D>PWx1$2{s2n8^je?FhRU35Jg>dY`Ts3e+uwp+Ksruxj_%$oi0?-O
z@v1BkN*@F9<R4#erTw8+aSg-I1>(z-V5nEo`@tQR&Abpitq(zB&rsxNhQVWUIJ(ad
zM;h5inVfT*Y;#~jIPMJ(M|0P3WWBwPlU9+491;b|-6-7n7zJtXD9kSoLt=Xv9LPbk
z2gC8`aX1cjkHAZnNNNb!f!QMpuXumo7tB>1nv5~@!}={whTeGUF3CGPIoDaI7y<2G
z;n<QKh9Bp`@ZoS6?rjaj<=3H{DTHFLO(<M8gy4%(2!<v2Lo@g?#?iC;ySpE=ubJnP
ze-W>~U%+aI3s`&WENZ))<;=&6d2e3q+4jW3U7qY>pa;p7S-fPL4+b76o9Kb&3U_uZ
z(8K)79kVRmVe!Td3bo9cYjr~fXBT>rXK??j3wAd;q4KdK^E@2Tw#@<aeC?4g$TCaq
zV4P<Un@J8ZZL!CeTzly6vB#7#cGQb-CNs;H^L<;a8q9glcUvst9;#5?9#$Fl(2PET
zH`%B7ygQAIH>cq^*%swwfK`L+@RocOL%z8&NFSGa@F?lyjjleXt<%TUjrth0TOZp_
z>SMp&dPra21dmP@80Kt&M?pKWCw~{Z9<#!f2%b^gLGdi*F&eoa2lK5Ezi>C+jNHqt
zqT@Kf>M#~;K8%w&r?54~0iB;XVOOyeGM75T<B2ooD4IYyZz~Q>F~Ec<9dxzQVV~L(
z^5_ycZ(o8r+m>Lo@e)khwFD6tmcrC#DPX<~RVQ??b?{1jS-KW_KAZ5RR39rtY_Zzg
z7GA9;>=8G?oD0TKJjb1ruQ38jjnIAU2J9?ei@;@T;48ZZ-vicQ(YrMme_RjKuWg2G
zg#q4|8DgudF}&lbxtU>vUgVo^<eS%=-Pn?E{1WFuPIn$A&!2~eE%RVrFdyo}7a;q{
z0^EGD00&<$L}&8N1oBNW`KF;%oBeBxAeXTSb8aj~YQSpj^fJZOAU6z@@kH1dPkh#=
zcYfSO>|SvZBbfP|*B*iuGNJhTDFngCg0Smm0J;POFh?Q)3F8AWlzXZm^35`yab%pa
z@<E(82cbsxI{N&NZ=3_+ZhH+?A^+3hMnc*v0;XG{k$Nl|^Y%yMU|ux49Acn$Fa~RV
zVlebT3`%Cl;KcYCOyKQ(`^BJYeGG!1b2k+ki^wCf2pAL#OU`WuUysK8oBTT8`FQhK
zY$e}(wv0v3%vd;+bNbPbc;aO|+6E^8YZBNUM=efp0%H2HW8!{14x7Z`_Ru)?eMTWX
zj$VY-Q5bzZ2K`4+zdV&QorQ6*S{jFsJi4pmV9V2GHT6I|8&}1{ayIu>&Pg~x#xWw_
zJm*Ov<Cw>!F-tlPv&lCrCS@X>xmsO0zj-wz9Tq;BsEo|Su(nM0U1URg=N-5&xQoXj
zcai0O2fwfJeH^t)e^qiYWpNH<taET6uLO}3E3u8Uh0l6DgxZ9j;=>V5@xoFf49Pdo
z{UxG^EOLa5bD5ePQ|_$JbGH;szRBY5N{!r7z|)=V^5_dQ0?9gloa1Ql^2jHN_|078
zp5z;C`lX`p(-)p85epL}qL5#!%AS676V7yIlX3bp=a0F?12{(-%+4&em0DukR}C?2
zkcQB%R~N118r4R1(fg`8vuW5dHC$ahd9Nl^kEyXUn_b~as^Z2DRk3fbs+gcg&Vh=U
zpsB(glCtoOA0%Go4HVt!mlBWJ=bfe`vO|<a?N%ibs-+}0xC{`_iu;S=SN+6nkA4EJ
zeMKVYH(QspBZ~g*;q+1^kZ<~rkQePw<;03p^iPeIV-LElsL$^yu5{@kKF;bU{!H!0
z?yqj5T&0^ZAm1D!-}EBi*fz)rS!U3*WXlM<AQ>^Di;OU<l@_XBq=mspW&p=Zi|lCT
z0T)S$As3~DoVk<;Qjrqg^__*?@6KXQX=gF*S7(v(lh<!_7Tu1v;7vdi^Kr>ATbkfW
z9ZjE2O$Z|2j3D3iB;N#3dvk?+Q`JHL@?UaCaU<s5YQ*@cMtmi!Jf72ljUza}naIwi
zJ`JccYJj;%12o?-pC+aO%d8vFeSHH$nY*Xs-H4{Hjd&Q!9;UzbSn0@K>%4lTw6XUf
zu^#rR^_cymfqvIUlyWcj>D6znYiU5|et)oDr4~hJYp}hc2CB2jH&(UyaiA8`Yigko
zUxT{&HFz3Sjp@Cs@#Qr4QuH5XlviSUK_%SYRATDAN(`A-1qUtajUQKXK2n8mob^mP
z%bc3MRcJd&z0AvUXtSs7%dv7i%BjQ#*&3*GFO~X%b0NJl%(gE>cu5&PKP$sTCwACU
zhrgti9kzAlDE(Cq*?r`qyfSRjD#L#}OY!wX36}IP!3*_btPw>Bm{)}NkC@eWvIs+r
zsUIF#f|!zG=CTzd{A>{(S`^`JK_Lw96e4|DA*}oU!VZUTP&K6HciVgHGJcHRQyxL{
zeIE8-&BfTU_Ytz>7Rvr+;jmK{N?J1UXE<lC^g*98N=HZ#vs`YcK-x4JgXuXc8^ZTG
zH>lSg$X?>GG+cX~1}XDwblP|q7Sg%6dN3Di3Ayk~$;HKmxft!t{nUeO^ob2aOlm0B
z1XDXqrg=3h2$9n4f$kkdmIy*?YaqIkZ}xIFZpqhDk1}SkuD*uFx>vEX^JOT$3&x7P
zV5FCYz@;$+^9n=B(qZua5r)v3Fy``yF<UhpVI#vK#)M-Bb3H!q4#TenVKA8xhEv1B
zFkT~!zS>YsSsx02&rqya48yBY+-vQkj^`d9N4^=LAA!*)A~92%uaWLiILz7k=3{)X
z+L4GE6ZqdiB^j~bx!1~1LeXtz>(qp!Bq|K)XT$JqPZ-8=7j@xbD2)6=G0-#=Rw*I4
z@F^JA76qa2uWKk~kCJ!tC4BgO5$6LgGH>kybHXm5SHxLN&G5qbEnaAr_ChCjPh2pf
z=Xr@IY)5#aezymDO!q*-01xgR-0{8(y8%|a!e@vp{MNZbbG$1&2e{%x+Zlx3a)Ijz
zXSDusMEo8H{4;Z4f2RXB8aN<;zNg7^?C@2?4rcQ9_?~SKEpm&csXZE$>|jf_Nt5Lc
zDg89waz>+g-xibB*<nwf9a8F!U`EdqSi0;Kde)r65BJj;-`N&+)YSz2=HvKy?{+<m
zC*Qp2t`BYUjVt-4QcoYt$T!wj`j~J^A8u>bqg&)A=FyqM#D#j6z?~?4$E=#8R=8$s
zg?4W%Bqms4ewGzI>Q)Gn*^N%sdvK}XI1*PJh8j<ymJu}jGgI%HA(GE>CUVONwrxfj
ze3-kSVq+W%*^1_=1~|E12OVdYK%0DXW$zNWk#7{pH!Vg>Fw|rT5_wy;=~8sxz6@SR
zbx=EOCH5>^3succcoDfCo!0APmh(Csw$X#4xgJcy^>A^@I=q>=4juve_|jVsABuFL
zTcL}Mt-83`p^Ja7*5LDzwdl+L7T+dr#&H=##PMgXSZN65t%g`O!Vm`z>oHGV8~vN+
z<JJDT*d#v>1(W9C=F)jsuzVhN^jv`6+*f^kyZ}8HX~V-r8}Wy<F)vITOTKF(=j0*?
z@=esL#rW2F3G}+G$JZ=#92s>IpZ}hMN$_c`_I1GQ&X@3PoFA+s{P2C)WsJY-55J)S
zSmzdi&#LVH8Wez*{{kQ_$UCzGAT=uh<ue1&eQ^MLC+WZHcAZ(jf$Xqy!=@vt@Lg~n
zQ`I9NhD0LdZ4|DIh^Fo?8aYnUXbX?VikHzCu!Xy-?J?}3kHPldG1w#>1AB!S^j;E!
z|L$`>L+wog8Rn{bEZhP()7cx%JSo1P_`v<v$XG}%j>VJBvDjr2i!Ge>gz+)W&&eT@
z1avCnzKYiizsI8>B_2xD^hDQ2!Gvt{FNQs!%wW#@6NQ!5oCQsbLn_%OohO^eo_w=!
zWgHHXZ`yd~l5alptR~|eAmfCSaaNFV+{ibf<eSyxn}y^X3+76ktxQ9YhBTObN<+Fv
zI-D1z<92u^3gojOrJs!j+wWka?Hz>K-GR6F9T<e%f&UHW2#rX_MxA8LwN8fZ)nrWW
zp9<-jspz$y-Y<(>MAyAUxl~sndw+ni+{Eq%a!8E>y-qwf<ef_LPB<AyfqY}|jrtj~
z&4%|9(Lsi(eMYAFLA?%HXDn}<!jBg`kca`dB_cnIdLGVhN}?np<s!ev35j^Mk$slK
z$T~6-;lpg3J29G~cbkUryrCg_nQ4eE{WZi_?xgxYRu|ck^n!b;i|P&P!g8RxSpJ9|
z3w!7Z@2w^}$1=-rx~i~zt|B^}QW5JGsEF--RD?w#GjHxFi%n;h#f=svq4I!U@OUL*
zb6!b2r<Qm~`T((O<N%?d)L$&1_U6>Ye&UOVqWGz(DB8(4a^#y%?6gcfM*lZG;{QtI
z#K|jiLS>|!aGWkHYTx!0dfj`7w)*ZudK&pg(oMK@?<Nw8x{CL0GNQCeMocE(Ecz`Y
z3X5e#-*g$FL%wlqlNRGDrN#IPX`x&oEmFxhA1bM}`7R|K$u|My8@GziBE5#a2_>Dy
z(tKtNvjg*vM>Bo}b0*`|gc;<Um*g8S^36c<O`;CxHjCIZIkyS>hH{o8--I`u-^7t`
zJjgfeGaLE4LAFtEfa;(I1aR(RJ+=W+U)kMJSC4Vy8{lQeE(C)HIISkH=rW^baRV+r
zYQWBY4ea@0zf<Q1oIJsPCTi9LWSD8=S&zwK_1KicosV@TK5<r~u=Ec$(VIPs{-uXM
z>AhZGgYS=O;7CuC`+{0TEUkr3Rt?PeSL2;ZHTu1+!U5MRY^o*KJSA%zR$|fkN<4I}
zM3Q|a+U8c`iA5#WrBtGBW)(i|sABJQ6(S-larAoyyQs_2VpYy8mvVeuRE|4k)!ge>
z<KW;j3`QA!7G+p%Q-(*DWzcUc#oVW5n2}pftw1>{_m)%BQU>20rI24x%DqVm44aE#
z5>t$srA6p<w-9@Ceqr3~LMVFv#CiAc7}xO?(@%ZnPT(6B85CmMlp^STFT@}BLfD=C
zNpICRoLu-CA0Iw}<%mbj-FS!wd3F`-%fs43xft>y2OEdqLQZ`a_L^s5;D>a)$w|kN
z*mNvCpN{E$Zo<zr1-kSd6zxlbMz18ugeT(g4bFts*b_7QCR!}h@nG>C?0a&TJ)F7F
zG0(-*Q@PNW&4tm2RBT;AX6PJ@QS@j`%?pI@_dtY_ZNmNq;+RYj>UnyQZ`PIa_BVkj
zjb!KK1@5Z60?~(EJDX;)U#IC3N+W_Xs5}Iz6GL&Ce6x{!Q}{Cs2P?v$M80w95{?Dk
zn8~Ub4yS!#s5J}2US;0zXb7ZEg+RlT8s+}<#Oj6OpGPQ~`-WkbMmWBhgd;dJ97lUb
zz)Lp*4>|9-)Dj7etx?#c8I9JD(YV3aWEYtv?C+k8h;B(3-I9n+(TOOl3di~D+)cUh
z`|k|Hx)otq?GlPni$gKFC4}AQA$ZPAwfH^sqFiM!klHmox^V@?FE8SE*Nd2CegQ=@
zE<o?XS=?o&-X$$B4D|Oz-f2(ln(c}2GnhYPPNo^-fgCvx+{|&uZhtrQS$+m(Q_dh+
z-4)lyxZ)%)6{K9z`uGeKZaKqS*%@X9jxcj@#6Zq(STKsjarWGc+2QIWJA7)TzBj`b
zd#%VcpKUQ`>}l+1JBe*=r=hZkdY9E?8`YDLKg8LLj2#Z2It1xGN8lZH9P^f*<eZS+
zWS!HPdF(V^P%G2&-WKk>->`pr*j%>`QQh?MXn;PH$v3;mHyY%d4cqmROTMw{x*mE_
zo8UOZ0%l(=kh;za(N<P4Bj5C=Kf0KFV;92944&s!nADlOtg<~&`*a-tMjfV}#0WM`
z25`%yuju0zTz5CZrwk+1w;G`z_dt_!jq&`FG3u`vK;!#zEGFM%k#F+%EI~B+W+-Pj
z=gBw0hD#8ru#{T<rI<s$d2~nzH6vG|k$f{uZWD~+*Ta6#daTJ^hf4CyuDyE5+oFfY
z4SLLR)WhEyWRy}}nEcem1741-*F{${4)WKaz-TS%sKx1_u?~AStV8tNb=Wq39aLo3
z;h3EsyvAt5QELID8w52ibMXA_Y>aZ8h2>{v!Oal(p(HS3+XC24)y6JeZ49%}#y#@Q
zoRiv^n5vEQD;MEG_9DzvT!j8l7h&Ip#aLl)2-^TNsJtZKc%G(T!xkf`Bid_YkG6>p
zSl8;rTpw4o|L{P=055D!^nnR;ucr<Rz<ADWHjr&blWp?G1>gyHR9;K{v3y1ZwB9n$
z@ooyTO(IZzJ{*&-Uq{-KNPGy1g0Xru>Mf#SLB?5gH5!^}(YSv<8e<+r<F8r_`(I<A
zL~oRR&lnt*#K3i83>IG`vlzxgS2-3r=VNfnEE?(kqA)xu2G2*wB7Y8bz&f$e+7gSR
z1F<OOo~q+Ea{+(EW9c(?JGjLoV^JI|)Z-vUwz+aM3I*{|ctE}xK)!iEzF9%Ok&ti3
z%>F;V(c=A0$v1b&H+#u8r96wtHa}LxV;%WsANj_Qe4|aiaU|bNB;QOT-<W2m!J9f9
z=e#u7Jxybe7rQ+_r(xDo>Wrf@@l`epIV-YpNAnK;ns7gL<_=^%@1R=>*+!QcTkY{M
zu82p#dwMB3gW51F9)Hfp;dwY`NHgNGKr4g!2U$3EycX3~4N$Lb!{H&E?bvX?<U}?h
z-zbuA>bPqv_{#kh`R3tUa!`&$gpyr)amVE&B;o)yHL;Nr@rk#;C)+e;NW_g)&TOu5
zf8|0Qvz0^~Vt(KBC{1DDt|>zHQX4c^Q=Dnl5Ozr#;<}NBFz%%xh7_oa3ifCEGv`Ln
zL0wp^R2M38>Y^@7jec}BA=6cjZv$1urpc<phIxKJnRl~qK06tDtB8lcl!a-wvM|4<
zEM|8fC?@16ap$HaR-RE3W1bBVpOy>|NsIc6_N)EG4X1u0rb<zC-K{9>%KC`2+)q6&
zQxN-xD+u|U@*+mai<8B2A~5QIT46bHd7Z3~QkE5tnLULn`9^1GHxVh@O<2dth-0@n
zqq#35EFZ~;{^c@a2G1$-jnrj+Ounfi+o<wPAm0on-}sYn_WqI*$z+=rvW*hi#_VNh
zp<C8jbb8fUY-RW5U)N?_j%~s#@=Yh_Cafjjyg${1jR%_ecd`j{O`DLtxCsplnou#J
z2_2eESVxc3$X4!_=yB@j%Dq%b1Fm#!z{36w&{pPrCchqf1!S9=dL(e3b7E5i{Pf5%
z%Ny`yHt##20iX2ik;aeTlZ%qu>M{O6Jru^V=jk*3-P`J6mS2JG0o1YFFN1x1IcG22
zBW<d|^NTeIO{NyOA6Z7w|3s}#ie@b&ziOcG$lRHA%nNR!SL#+3J|3?^m#j)eMO0$U
zyh@xMM9%qLf#aVmaJx?>?&wq^%%>8QpH)I{QWebpBirn+M9<?D?29Rf=InC3&?v`@
z`Z5H+tcKO&YP^yz!;<0LH;pcX`j9f@7M5Z>b;8Y}6pOx>Vvsp!J_E`S9aoC7YbEUW
zD}pdDMnzQ-wvMLe#-|9&9SWh`<rl{E{ehwp-%u9x1?#jv)8k%%BP%{(i%&l4&%DOt
z=g&|f{|p`jo*>u$F}2B$;Cl5TqUrM<RF#KcoZV!v%fq_LY^-a}!qTT%_%|{GXUo#D
z;(j`mZ>Hnt^>mEAl#T|Qbo@V-&O06p{{Q|NNl{uNWwp1YC3W65$tckjP3>fqm7=JG
zrcKD+N)p;TP3^txl@!I@{GHF^`}^bdxP(vL-JjcaJ+JdRuXBpdp2i62)3`jh5G~(M
z!SM+1km%{_N>5)S`<2#27sD#66yepE5q0D$rvAN#T{Ra_`<S}N?MztKCqQ{lBJXSy
zp-_>CJui~r`ahkfJqeYkllcBP5wDjgV&%$2Jid{Hul(Fw@{IxcMm;JHlcvW(r7{{5
z=OlB7n2ZpqR9q$BcyTW@p(+jOjcM>8-;9!^zxYoY)OS&raZH1Rekxpcr$E~)1snWQ
z@FAG@<@f00u}nqtnl$JZrXfN;9s3sYw)_FLpU@0;U}Zw~cP5<JgDJ&qor!$ToccTm
z5h1zI*qaYO>Nrn+okC69DIEB55>dS}0Ht&s4W##mJoUvO4MDxqaM3Uou5Bq8?wf*o
zi4+`IZ~{6piLh%-z(vn^n4OM-|9$o_42r}d`A9efA4cJR_CNkRh<)J)al`E(yWkI^
zc-%qM&ke)6=1`cw4MoD85WIE_M%Nj^upS<aVMf8&r^(EkUhJa&!h7emVEoP9hYs?M
z)2kq8h6mxdN)Rp>>_NlZ-MGwq?sv?J%j0d3#i0OfoXQ)f5B^Xr_D5jzPHY&u6W5mP
z!YOS(^jYeIv_9K$-g_I?NNvZBg`ODLcLzdi{cz&fPArYy1;4Ug*iWs%(~!RxcR9a)
z+u_rFdn_R13}~@Om;dZx&GG5I0CMD;<Ll|okzR;=@=c}DdR(pBfRF{7amST=De}#K
z<eMLdH$y#jGkRX)=bmha^lKLsmhXV^W-rLNZbRoEM<fP0;K@bu%}WQE2RWfR%?aDT
zIpM6ta`ftOf_mU`L>}S2_0u`%?rwopKXW*fZ`#Q={$!h?CFYPL-wf$Bn_9tam~uaL
z$;}dCiY;!;nU80ki?QzHLVCd#qBhSS-L&nI#(mP9rFM`Z-y~St;cYKlTu`2e2W!o#
zTbkqcvU%u%dFb8K7U%oU$LkArP}Z=Amx(>IcsRrCVbExYlR9>g@;1htZvu&?voJb+
z21;+~;Pw?AXcy|>#~B?+2F-w9KY_R8o8|LO(4Tx`zRd&|{7f+5lnLz3OtJTy2^R8W
z^F)DbHz@UAw!qv8%VG4>8Ini*ap$=|?y3f0_?!TAwFto2wXU$jB4kaNi=lsQAl-j1
zJ9Ot^7=7qnmq%jPq(pS3mh*tKczgmxc?xDwlf0xGjQ#D=cs4%;anra1PRxXwbrvoj
zWk2WnY!pn-!9OxeidPQqB$IW{<lyRs987BET~ucdjxqD8iT6=DO8gn3rXyzMVr)z<
zKJ?DRV5eLpwdbJ6k8A|Y&Vp(jxrBUEWljxeK^}hl=i$SPd{mb*L+Ww?OalvGVU`cs
zA^Dgf$=|ny??K2n`Q#fP{;aMx&Bhh-jT!mo4fk9Brju_B@}Xpu58q+=Sfo>c^K%Mt
z*OLAlvQ4;k0mhJR*6%HZ;lv_5*i!_5^34eHP5z-0sE~0~N=u-0z68Pa<4mTOBf}f3
z?&O;hi_T%d19Fb_B^3Hy!c6nactichvyAV9EDLe1-${6oZzhq=KIaxdeQp6RC6KQK
z^M0!Hpj@7h?w$pBo>&SnW6SL1c^J3e#`?Dn@SD(Gj3>M7A>(`_w~W3+jpiZq0LeFR
zp3rA=g?Cq(+)J%tPp0w^;rUHdoG#ZC9jTfkj(Ik#m`5DR%)Xn<wb2@>De5IPh3Rt*
zA)TfnH1=tT+;tjag`S2OM7~*hLS5XPtuE5Ps)?ERnctVFCPvLx6XTgte78hZtog+b
z@H497-C|WC^LwzcNg6B;^cyTbU0|OkJ;mz!gT$SlgGBr%6;XPg-3;+6LhinjP>50z
z9U)4ho1c=<963PLnJJ3(`Td1|-+p5Fh`vI<nYnxxeMD+A^J`4mrNBO@chqkN=Jyi)
znL*=IB`;>|krx%C<wfOxaw1<&PRO3=A(Y!>g*o|VGPRpQ<eSj=uEKL|SK;p5Rd^rh
zDz1cf6&=ysP04f>*(%Z^>z|Yu&%T6;=Tc%3rzt>6$cIzAnJOt>|Bw*9sM$PtC?VAD
zO9&H=aVT$+$Tw~O=(8c;Ogc`TCb)z7_8oY-wF3p@n?B^5x#k_vGUm+^`9?>x0}BST
zU;fVzJa6Kj%8OkBiR6_X?d*5`j@3`wAzRUoi+o&7J*O*q=Y-9746<a#H9HhExVKvJ
zqaEhn?a=Srj#-s$%-U<`e}nC~JG31`^4V|3&ZoLZP1t&&4lhy~aQSy5t_}Z;mzU|K
zS=9pLz!p4y)B-bkb|9?&0`nJN5Xt+er~EjCewz)6&3JmN37Yv$IBZC-@Z4s^4R6Mj
zzw8(p!5nUSW`2I+@4ux94$)0WqLw2ypc%39WShz+wAnS`ai2ys^<dvDy)!*B8xZoX
z9xdnV(fo-VL%k+ap#ihHH6Z^<J?e+lqvl#2u4>m|We9yV&Gk@ltz*|f4V>DmVgIBW
z(GRL&zN!Yc$~7?NE!5w%N^Hq|57|fWurB#6vR0C9w!en!s^`eacm})Ths;2}kCxMS
zVb<*~>NefM82xg5J9ZoIjc%jt)h$$p-GX`jO*pK&gv|Hnaq!SNIR7a{WNj(jsoh*Y
zQ;N44rRWGDvygB6HlN1h%Z1n-atc4Xp2DtBcHwP4iB`TZNqSd+Ux(=1qyB2Ea~jv)
zoP=}PNt7MW#Vz(jhKxyov|j?ol+%|On~1^JxOa+8Le+yLddQRTGdhu({>-}zipMU*
z<GyY@0(!@D-xG((m2rq#9EZ{L;!d&VPA54U(-tISg<~>1^it7bmxi&_Ou~1j@$Z+$
z|IgEK%O?#1ZfUSiOT}}t#k)l*_|Gu~Z#JYrky;Aq-MMI<#+)>2KS#M&>P*8hlXS#<
zNXJI+3_kB<pg15C7k|g1Vow~x&c)%w3i=EpdH1?CAFHPq!2jzhXnj6~!#8<*zL|Hh
zy))>gOoR5KG+ftB!xI_mISW!@)WKW!B`KI3mW)#)lXzc3kAqAic339hetR6Altf}l
zd<3l7#qh{29HWX4;r^S0*t?M#GRx_s7<`cVg<*J67z(+9P%JYGg=%34sx5<|Jt`QQ
zJ%dr98B8zI0eq7>fL&F=%xYsVl+1p1T`{L-(q6=U3&JlM`fn`hovEX~LcTE{v>U0n
z1CZoPpK#v*=Hd7=Tbq8G$el3k^h42pKj<9vLE9iN6vk{r>!58Ao3?T9xebSp_~7PI
zKMeiq2ZKL5(AQxn9E_Rid&?g`|K~X--(=pSR>3LzY>%%E)IF-HUwpKO<~w`%y|u>*
ze$KCrKAij|FrU2vN69z7ZksWIe6x#u<Fa=%w3ly2!phA!<LrW<W*?Mn@xm|XZHT??
z2sJqeOb`y(mFj@IdzZni#u4*3IYQsf5nYct!aKtWa<xwAx?vgaWGz5ii3NO5&1P1L
zIoubU!)T#7PV({gGINZQn@#O;HY_|WF!lT#Tw835j4ul?+;uUO$v4A#E<|}Z>Mp5v
zD0Q@hBiZJQtsRU!Y;n8CJXmX(qG7fX_RAPS|A8THR2gwUJqKwTw)mB4hXwg|=rO|{
z=Q)!9?3gRWUk^2*ua5qjR)I5%XF#%12d_$Xuyo!;v@e>7bq$j+@Qw~tcFe%acSh)C
zXhNOX1V_m?+2otWp(dDWWXc{Fvd>s!?1~Uroh5K`kUsnpXVZ_p0t0S1Lp9171@qRR
z;I}0(!3s;)TcKN;6&{$`V9$3OsP>zSb*}U9IDQ^XE-r@5u@!KCx&j5hD-m)j1%IhU
zYVG&Nrf7fssEx*^St)otfp>z@nV3fw=@F2P870{W)XG7D_5W!&6AtqhihdgB>>SA6
z%)ut^t-h?x!Pk}?wDzJWXGAWJOv}aHMY%{oF6M2bKj&f&ER}QM`!bXH>+}cT$i-iy
zJPfzZLsW1cw8#L(^a4u_FF@Qp=IQ9<<GW@)dXaAql5bLqv$2JIb0IVvWhU8>Fv`Y5
zGL9n|$L@R{Dmd#d<Y8=5E^bWCLtSMaj1=i3G%rMVn?lG?XPKzQj8C#n;-g|5cPYUR
z@{Jq$rXr^V&S!aJQAR!I0{hfCE}Z3LoK*|X!FT03taLbs!0;kmUwaz<IE_~e3!yWH
z{afUlNM`bEzEFVmp#_M&#l09cqtw^*7QG^SZO@0LPXS8fO0ni-DU_#_q3q##Ow%aG
zsNW6f@?BE&A0;oew#o~Ki}J$jYky&#t180T8Ffs7H{YYUqnggXhn_>k!%9stxQw0O
z>;aGQ))co^YKnAA<`Iw66qU^F8_pZ`zw81}N@BNEfQER#iX9BoG=v1R{B$>~iwI?P
zk$*!?*m|gm?nl(bnptXML?d@np7a$r4i?GBIdcY!ZJ!4*6OdVc>jsK2se$5itg?u5
zP!^h6%EIWEl8||(B=qho3FBDaOa&>4+xwKnD_JG6zDiMSpQ$Jo4elo*eftVo^}gc$
z=04({ejo9$dmph|x3?&`p&;h7gTXJYm-u$Gr<n3eUc@bAhRtw!aiLmHR20YwrSrVw
ze$riR-oiU?nQr1qURUucvMYPhy9zDx&8DSYg^F@lG2=fevFEXrP?#qrCU%h$IiDm&
zN1~*7m>?-!k4cK9^cH7IN(z})_IXpcIa4Dc*58v5N*qJ-jZ8=<tb#jHy}uJ%IePm#
zVYr2NPp%!9xSn@Ti#woAt{JV>fzC1iyM^M%<eRJX<M=lIz=E|L-|sN>WCudG@0fOv
z+6(9Tqjq>Jvk$8OcP#qb&K`kw<n{fI<-6K3g1e}{KJ2c%)kbZ$4VQ}AU}8$vk#5JJ
zxNkUe;u|FAd__OrO_xq=;D2L{yvt+`UfySDOSE9b+!ky(+Jd((Ehzr=8K#8|nD)H^
za@`s+<zExBoXIqKO}HP}gogc1*yY#69c>dj==&X6)`*V!M#PS7f~-vw68N<Z(#@#*
z-b78K39AM);ZzO(%nJHz0_n3kQ;#uq^%&<_kCM20-r3aS@$GtyeN>MH*SME5;f;4o
z9rhil!?>ziXf3G2V6}P_p%%xjYM|FpjYD^;G3j(Qz9*1%K2*cUzlwjqN@Om4&)3nn
zydQssk3KI^yz&M5cs_^eiDy{Y{t&vQ_p!0$F20g)3>M#ke)n=@j=9a(!P^*p;}&vm
z++_aiO{COa#M?O+kU!-dmOd!OwCYmqx><^^y#Ka}XZI5MW=aV&N?UkKoyY!OC+4s>
zoy6!B%wYe<-S;Er*zsrg-GnSS9Z3hersBc26R=ZEM9<y{nD>RZs_F??;*o&FO9@b;
zeo?Q>Jh5QzrA|?|xsZgB)Wri2#G@@WjyE~%&MAwA!kK7T=0+n-E1J8MXzZRAjZ3rm
z{2CYy^Mq)=z9(arUNX8&W9FDiGIp*`X3r3JSKi53AHi<!_sK}>k%F@`Q;=!PTPrfg
zVTTm-aZf=dZ_5i!(_rG1hJ*XbNBU_{{1k(6*J3c?Y7F93W6>TIivu5HF*+j_yRXNh
z_DL)@B*(zEB@ZRTcz+>r5;=RAk9_JR67`vt+%E%j_N5`sHVvNBcnjVo4T;=Qg(;_E
zgfj0`nv>WONWM8h&qGTBCRrq4g+x5QoQz^GX9O}NBdB8>gTpWOJ`OmD@pcE1b1jU$
zQ(>5s!@i>(?6>v`#Y=@y_^)CA(5zsb>mSVSv;BB9C>SR?gZbJSjB)HHDwrLN0n~1y
zkL`u__`R6X9E8{&f!H;F5B9v<jps9W!|q)G(zXRaZCL;=clbl%rayNM{umj&6I!W$
z)Gl@)^TT%BKI#ec_a4l~^?>zr52!5OhCJi#c-`!Yi@rW^IKG3rnjhQ@{NZ0lw&C=D
zYll1K_PBq?9((v`d)FRws_aqn!5$~bHVZlD$vG$axzfM|ICOFeGM2B$=i;>(e}-KO
zx@!@oz7`wyI-`52GyWY}$6KlOs2ku4N3u-?{Vba=E`|RL2duv7fLDf&7-H*)=#7q8
zM83&6;)pWdD3yCU;jX45JFpkQNpcPxvu5LIyE!`Mnj_ub9BB*8G5@hSYDUh+QAZ0P
zd=8YzHu1i;nB2Aq5>ppKYMed(kk_ofQHSxh!|s)KC|YQT7kYNM%kTU5(-QlWOfY1Q
z5zLYd5#eWuBs)WVbT))<(QJ5<aq2Sc@F3U@|ET@+a<D^e#XPJYW`Y1qLmVUDTplzX
z3o3QcI&&hf+Dt?R`KD{tB)n>!gj0Sx*nLXClFVdaZi4;fn@QYHMFyK-nlMFvqY38Z
z31&+$mgJ;BuVwmpxz_+mK8EPh(-f_Nrg&v*#@sne>>gu<XX~tRI@JnS^{L&|*r3>c
z9;7$V!=2!Hu<F9xoi)qwy4n%lIvruQa~rU78}6!kL&lJOSg+XUPR(_0FYa<vGBJ5&
z7B+ct7vz@>{b4zHW|D*Vd>^10kb_)1>NYOSv!SN5Q<L6dvd@39Iq?3NgI=<^XrtHo
zAa_-BrsN`2ITteg+Ov*qL`=y<&-Yod=tlj88qRsmJmx;~{X{UIXGRu4o<DmVZS#>n
zDIfmaO_l!1hTr3CoIjAw4teI7bF9fXCk(S8&yi`*W`-fX&w;sk!MQml4>R7-zkHpy
z;feV$VqWKl2WJp@jNJ~A>{f{>#*s}W81GgB^RN;ONaZdnvjpWu^xTw`z=#^oTJEcs
zoGihl{q*NN;f>YubJ!Gq2ImWkAa~{@`tp6>5aW|LGx{W|?()6Q^#a)OIntVZ(`|D;
zDnFCivh$d+k%vX(n+Lx9JfCa*n5)%%sT5=E%CPj)d7y~*7jN#t`QJTi{Pz)Oa34dD
z^bisD{YA?z=I}{t3KeQJYfrQPo3~IGZflASXEcR~(iBs-Yl;^Rn!;$frf5^t6mvgo
z2=#XwqPC2;-_h*N^wSVqoHWFbE*iq<q`Gi4QWp;;)J1iwnh1fK$o@}N%zi?yiDQ4}
zJWlOkaVm1K$m2cu+R8y<b^-6eR}U0ZevxgCD2r_imBp(uWSkoAp^}t@)mCQRj8+o!
ztd+#90!4B51^vVm6-D~@eq!;!e!@PXuXv=|SFGODM{G9eBV^?J2+uLS#auT9p>vfv
z!Vx`1+y?ehWy*^~i<w)@jJ}%Ja-t}N`b|L(;kUWFFm{p^2G+8oPuFfjCao*?NnJ&W
zQ&+KONmudho3!{{Atg4FZR91T#Oq2)@gRvl;&|SAn@fuB`jX<fkQ8%4#`!HF)^QqZ
zC4}!?2{D~BnHe>s$v3g&n-;Q-C&z<)v(km!L;Xf$LkGOcHyY#{b@GkGXlgv<n@`km
zY{)lH>)E5Q<_8Qszr%1-J8oWRN9^TxB;Tfw<{|wzciK_hliE&eJ05*%$N2s2*g(F~
z@oh&O`9{{8-5x=0=<d{pZOLtT*`*yHwtqv6)i-=HV}HVpT3jrxgR)s2j`aJC`smLv
zFJnfZatkglWoGd8&+O4|K-mmxDa#tryrKbNx9RyE(ul;PO?YkGgxl^-2;trK6Wu0M
z);6Md5dFf3_;rOQJQ&}E_Z5xU5z~aU=S^t3!hB(s1`H<8yc}AOLAv!wGp$F{{CYg&
z$0J<p(OO>zlaclCGbGzY)gxS?9<CeeQ23x0;SsesWmJnLOKYLYebnX`)rh}WjWJo(
zSbC%yW}elsnN^ME&PoI=`-JMQ@9E2b3#G(Ys9g9G4~<^n!G`DXJ@O3MJs)C2{XHa=
z-NpB$JNUl09E+}TFQsxDvl4FMc<);nlyws>U2f7Fa*_MB3;5cP84oAuwW%t>L5~t<
zc$|g);zHg}7osY(5SmX45tGP$GJjt0{>jI;ntXU=6rf4I02AM2^4>ZX+I{J7Ih}y2
z?FqOu`8e|a9*0HMaai_CV8&?zawLz#<@a&uJY`O*LlQkJNw{@10eJ`G;TXvs)TL;Q
zBj046r60058W-}TF(xA#KgUF)Vth0@Cq(nsGaA#Qqan}l*>W=)Ih?;D8Olz{n7TO`
zeSMNqM84_wE*ZYe%h@m`1;eaUxc^8&ynPB*uS>z6>J;3NNktF)G_2*V>}TGIG*!l+
z;!+G^md8Mf{G_u!8Zz<G2%H&?ev;vMcj*|~(xZ9P6OE+R(O4Z34KK$6^l~YHl|?2B
zf6~uv$sQjcU+k;##i3<85d3ckqMrNl^?VmzQ5%_3nTX3967iIL^D^={#G)A78ykV4
zQzH;k9}c&3>{j@C2z$SVA#^hPBD2C^+ZhV2giu5~hr-r86rqnoFm_=u9JGTmTp}2?
z<@@n((SAI<zmIt<`;cF;7k0MvSlnYzf@u&`{sqD+j-KM)fhaWI1Ka4`Xpz{BEoXQG
zZpZ9AsQ`3)#@8--i9es&iL2*#;F6^~>ZW)=xz~17M0vn`ya&R*y2EF(2Q#}p@S|)S
z49GqGyZB(E*$$ZH`r$pboQ$`As5tJA*)N&@Vq=E_dR8RKH&@9yk(Ktydr#lV8+)je
zZ*Fo%kZ<&A`1#Qb(5bh8{;MUZ{cj1@4_kuRw#6`gzZhCE_87Ls9)(%_{+)}k)p-eQ
z#x2F9vZc@;=YUUF9bnbzfLZe#(MjL#qwS8Eq(`suRwtyba6+V-6IA*+;u{uWZp3VK
zRGCBFin`5w>NaGYlvC!=-b3cuX@M@5wn+2jzUkB=lqqn3Mz*n_ZqwXhhyG1=^sv}L
zbA=rWt?aN+*AA!m*y6v)dC<OWiofI=*91dc^ftseTSN34V2JRW2KeP*j0zJYOgdnU
zYazx+2{Xp2L&nVcF+^FXK%M+_WModmW4Vc#HGLv<$T$5KPsH;#lhChr5{guHn8l=n
zk7EpRU&a`FmK)>#E)zKJH-YYGQ^eGo&~qc8+DkzH>rAwD&V-RLfEa5C#dKr53Nppr
z`DS?Y!4gs<EHPu3B|h!4LX4U<+@9HBN4*V}ZJdX|Lk_4Kw+wBzZn!qr9n05xAo+_2
zigmpZWbKQ<UjC4~7Y*ynNof9^2KR_eES;Z)XP(*kwIdr+DmhrDn*($5#&m@oOqrE~
zf#jNsjBGsql??@c?L}%1lnSVWzRkfUa?VewT=XU9)N?ZS<#69iy+th(H$G><oP4zM
zdoGqwD8Msg_C4(5jrbVeh!167Rks32l5c9sH?{op3@ftnEI1ph$To%?1@eumW;R~F
z$U+wrdgwSGIFn7Yu+b(9ddIR*ayc8avobMf4f8pJ&tgf~S;+i5i=oo&rRrY7y?hBa
zCY0boN(pi^sMRpbT#0=1kZiMoI?kDb5@dSud;gq;g*b~}o6jJb8KOPOH&6M``fh3g
z9HtfE#9!WPHRQu8BOejW#ro4H2iD2Ckn5Yr_XK%Z8b^Q8`~s{Y-(2H+%D6Ie&V^ED
zywktToz=aArF@@W!cLMBJQ-I4&z5_Tis&J7mi8B~4`~SJftn(4zNQ#Qo_P?&?uSTC
z@p-GJ$YPe@MWHEDhH8qaewv~$yE41J)(~w)8lodyL#%UWU*<Rsv9>{7OxUI_;)bY;
zPFZzf9;PO)j8zja*e!LbSXIOvQ58N`s$y+3@4&+b3#oa7#l2^Pg!76);<4l)A=|Dj
zHU=t-QY&SV(@$C4x=KIsZY8ma{R_*V4-m>>ilV|zQ3Q@q6ya_C#r2kcLM^zTkh|4a
zOjy%bSjF_=9<7g<K8igJ->KUSR1li?dWjq1>}POe7gUnGI5wN!Vg-32dtXjO1<DEA
ztR5n1V|Ou;8N&VM%ZjN+vSR5d8SdPqMV`4dyD+;7m#@;IjC|ukzL6#0)RAvikZs<~
zmlUxElHwED=6`(i^pAu%%~?;rIY++9zAeGtQ!)<u#-DtX!CA+#Ilw$V^364l+}2L|
z%{q|IoSHXe9BI1_3>n)2g;De$59mPkclwLteqhzAADF<qC~s$aZVbpY=i4#$MmsV&
zHoWs5akm}SWE+qB?cCM2<FJ1_B*-_r$TvT|+7amA25UR^TQZAy(cL!Gul<Vk%p*>n
z^_B16YSGV`-jzOe_?KLZYnDySz5k5AHRPP%&3MyB?L@T!-^Z}qU}*yuFK>WXKIdu!
z&U|b@<Mu`@xzmVSPaE;$Nh9x&8?nu#5r+&KF+{o%H5~s9jmVzagp#XGczU89r<qf;
z_<bEtHr2tdPdz#`>yeRHhY?%qFrT;FMi1+7|6eW4zt`f!lUhi{*5ZVD9d`TIqSCS!
z@0Dud_pSz&ypd}1rGL1r8bM?mnSIrm>spNz<!YSx^9fzrKVYiG8^|=h!kGG(Xn*<y
zhGtLj?cHP46hFY{`S)S?`7U}<zj;Kyaq}-n_uShUCVd;DxR+}Da}(!+ZsJYZ4cLyq
zi0_%_k^QX<A&0n^dcnOi-@9zPcNYKb3sLl!JqB?n*^}|V*Ku~hEGmFobv}x><TLXv
z4-5Fb^}#6v-EJnK_pwA+XC`2uO9D2x#3OzEacsJO9OL>ON5#x|q&mc-{336M%HvUa
zEuK5N;}}{W5B0Ek3=5BgF8OBg|L2=>@=YH3rjAVUazr$5cF7u=)N_iXP<bN~tM@Y#
zZ&M^zb&G=APkL8AM<V$<vwx37Aw!)y&uemy6f;^UF!Mt<1#U<|`J5EQIHh3UhZJc3
z<D)_<UUNLVoxr;vF(?m>flWQTtH~}W^CHpPG#uBvhhy#SV^C-3!I1atLSVOL&!y2A
zxSM>FMh4pugYUOv@bq>t!ryq~i@Oi>fBB%;Vh3)t?7(6ARBkBvqy7hVjT6j=2@XW|
zp1tr2+K<<V4&ZBMDAo=ONBOs7cxZYQ6_tnJULJ;UCEiOphQWuP8;N6~xMUlOI+aj7
zJ`h5$eK5jCu)DiuKf31chmO^La?U<%n!|pDhkLPe?q0}T2|}h;5E|P8q2wEgER{e6
z{Me0&*1J*rHGuCm1K5`tfaTO}9u&|^tiB6a<%=QD+;D>Vz`Ez$@T1e6EbW0qUhe3r
z;|`tO?wB^6ec6Mzp-YYj#OrPNRK6XvKW>Bl_idQC&kc!#c0z@Gqei}|xxt+dXYf^f
zyt`lz>0EneFxaD(vz&Zm_Szn;+*{qRvqwaZJ<?O{aX8r?dym^=T)aK@#qmbT$(}za
z_PEq{A@*A@#Nx$^@TA9L91mZL)-euvM!vBj-{?CngDmfpJRdmX_f#h|tad`2l@m_#
zF56IvdW8NW810^o0gucP@W>1jHoT#-GK0)^Ggy?(M(8@~H*>ge^0I|{-XawDSO`P%
zjf1Q`zI58*T(un}ciKU9xgAu<H#rmMqvU`s9>mT=%XL#6Fy#*EgduhhF+l4)eFSRg
z<595z;xY}86JUVanTA*gL#XK+Vw!;=S_cAmt*68L;WTD>Yav*ABGR3;k^Mj$PbW^o
z&$vmL_Hhyd2GEalOb6xfbx>r*J91NFv~D!UyK)m0R-0h@Q=rGYSqMs<38z;2lgT%d
z`UVIf-!vr|W7~35w6C2Fv*I~u8fgh@<^Z+$SRqK+8fN!waK&gIwq`CxkeMT{zI4Tp
z{caE??pRjnftn-Rp)kk?>f?7}_x&i0<<GNhQW7@N>$ZS?{Nv-Yu$mgqlN4&VOS5sg
zJ{w(MWy8Rjz8u3Wlq-{U;<>xx{^<G$_O6j{Cgf4qA>Y`OZ`Oa#!RvoHI7dd=RFsV_
z5*e@>o`EI3vhkjJ&z&#1IH6g9@2c#2ATw<AFMzfgdl!0g-$QN~&fD)nxAJhRGaK4h
zvyn708&J!}?fNXpzRbc4^-S1(OUIPw^u%pTLqSp+G-a8I#e2ywlQVE*IiI)4H~q;s
zOUXBK<QpaO&7SV;ry}2IlW#o9H(kg#PY<1?{&NOqPm9q1ejy&vUvv2dvjuOQL<(~T
zHFEM%RFDs6GS~)shSMkWW?i-blmGE~@+S9D8`3ecIUQru(~+N(j)byw)Lu+SPwvEr
z_Q*jvzc=McK6cTMqi`-CGQIQhxLZCJ(gR)B&g|1$C()wF9T{0io$p=KV|oY+XARNZ
zM^l(hW_}HQHeWn6g}INW7{8gC4VlMw7W0y2G{xR_4KatiDGz#YHl5@?>ad15v|K|}
zD6vPAzMGY^)P=K#y7(ibE;dh66L$uyiSO@JMQw&EeaNapVYsUJ^klG*oi<oB7z`G_
z=qY}^WRS=p->836782gdqEla4e3ela_t?Mi!Ba^%YAK28vjc?d0Y#zgpeRNUQ52rl
z)SYtr3ID9V!v97eQ8BoW7@XN#7*FplR9`EIXL}Sx`UnMK__UYM4`rwK)}F$Q-r~xM
z@?x}%yy$gBPK5bV!%68O-mLE~9$%9cvx{ZLb;ycgd%KBq%esj^i=~ByfwZWd-&Mrj
zlosYUq}T;2B^<ctlct||L7b$}Hj)%?$u=3B$7Gu#&RX(K9`{mZw<Lrz`DX8tPE6&F
zs^Cy3UWawUfqXNBbA+>=yi?B4zvDFVd%n)6-`J`HE6Fz>nZdU-^aqxi|G)}*YYKQH
z)w;YLR$A0+&T=nxwH^M~c?Wf?9X-f6_VgRS=C4)4sn^WrBiSa5+D#StX1{M6`tvp_
zYi}DK*s}*}`d2*DV+Yjt8ibP<%(&;7@~{StJ8N-)T(P^p3B6Y|vwNr>qc79@a=ji8
zB&pxLsfW?v1}N&W2f>$qnN1DYr`L!Dg^k$6yQcqcHzM<JBffASdiGBPUcYRB!Se=)
z4`d<RMl@8?J9DZI-^e!F=h&4%-RORF9sIx5B6wgOnmD>E>)7u|#%bs0>T7Z5Y%NNT
z)FN+BEuvIwF<Pb;y&G%TZ(W1QqiRqis<G^770S0&Lv|Onn+?@i+faqMpFY9&+y@+y
ze1pkPUm@rv`6m4dwhed!gPxD@==FX4Q@)Sk&3BPsco)s@$}v>79Af8f?$K@`E#MZK
zKi-7&mm5%ha24+>FQC->JiV=D=&`I6Sr1DvWNQi18qeaGRUzc*wdrqp64%4CaJV1+
zwtksd%RCIBl27kNE~>fnKE^DKAGwLh(N4sm!wHxX6OW^5@yKqC$C`EVsF)g$4dddm
z(Ig(Gx$!V0<7ixt$D=OCv7<5`O;_U~__f2%ap*iB4ewI!q>7^H>x;&EvdzC9(P&j6
zbCgHH8&P;w776v8kvMD~fp2Z$IN2PIyF((do_zCsSrqp4jYjB=Wb9{F&SrIXR!vL6
z+8HUx=C7B?F&{891^Uba*(RTg3lmN-Z#NeGf5*TtI0j=sMI&)Y6gJEb$Dhz3Jo*rX
zwi$adJNpQhW*>v*A?5-{AA!c7!?+O^g4ne?5V_17MZz1`;=QqGqYsXM^TB{}U(|)}
zKy3LAJnV2stA{%tu6D;tGJ@PCcN`n*f#XHKxcF=r7CIk7T(4v3cla<~j|@ZX@-Un^
z!X72pFq~)&!D_!y-r$mLj)h><>=3+pcmR_%so%WYk0mketJL3*Gv0gg>1`0kTn@sF
zvLFmC2tthldz$J4al|VS$pe{F+^`$@2D=gRg}bS|00anT^+x%l+n^mtyzYmR;lB8N
z$qhRbT%k0~4PG<W!RnVY_HNsRu$LQAdBzR*lHKv#)*WN@Jdi;xXM2A)T+(r6AC@al
zjC94*Ev~!=x5HIyJD8Gfu3fjs(<}B^OupGf4aVh#J^FJ58Rrk@I%nb~d;HC|XSRbK
z$_g#GkF>-7r*@cIV+Y-OTdde%hbfx$&`?u3eR?+B-^|9}<Cb_r-|m<V^x@<<@MhQn
z-WtpB@`fYs|8>NwAx_NCb3$u6eK}E!u*Td1pX6u5_?$U)B{Rg%HA7!>Gu(DH!_J+v
z@xsaiS(`1e&)XK|?F-=fZvkxgGQUUOo<0cPO10V{dlNaQuN{WPFzaUId=$#tBBIj_
zr*D~JUbP_{EetSbj6PDN^r5Ayj{(&J-?tmU&A|X8*BZdST%UQv`WW*;4`Y0F@x)#W
zuKrp`FVKRul{S9(Y9sK3Hr7{Z<G(|bp#EwS>=kv;xkLxwU+Ey;X)-o`os75sQ&2m0
z8m_C(#DdGSuyWQ+wEEA)=k%HQ#){z&wFZdPFoI>W5uEHyacT|sWCJZAGsY67J1z0*
zizO~Cvqri0T>4;_u={K)jQ?%L=WcG8Q|^v)>K@E=-iFV2yzqR859X`v#L62{ShM&z
ztd1w4ZD1-!-b&-o1%3HQM@L{fRDY!7&YTRm+{?he?d+-&S$K3K3%S%$N+Q^O`+$1N
z^=v%YPyL2BR!c7Cu*-mZBgbs)W6o5CdpbTQq@!9R1Ea||X4%>3zni&$Y6X}}jX0r&
zulW`En0b@;<Xd@L{wWLL{>k+ECqc(Q39%(9sPIk2U<+nId`d?_D{mqHOUDwjjmH+=
zNM7XU)!5-OBpq?&n?mx<4)V>z)fre7l!56F$ztm4z~X+*Jd)n%C}t^hXSO;q2T=!d
za5<Rw>)WZl^vgk?X5MHuW+R02Uw_`ltL0#aVJ>{~^ReX$?+@r--l1IpcQT7z7kZgv
z(vWwJ|DF=l@hB)A+YY2-{xNb6r{zXE<Qvn``x1THKeLf5orB@j!lZ`h;EGlbvKH~*
zQF|UlbUxPpI!T{j30kV|!N#kHc)L$s4DGEc#!k@`E66vCsN4MEzUu1=P4UQvtTRAU
zESJ(0Nngn{pEShL$MoIg(}Tm?DK|3>p?yzXxGYr{-ZRuiDYKSOJz{^R5;JeARK-hr
ziMxlZik~K`VzQ>H$SoKwuF+GxW8fg+@<T=F4OI~?y9SDm=gMN5tFo}3q$~vQ!K-)+
zZns`ZY?V+F*YXDlpI}8XdAXvvuc0XXnN^%Nx}ONO>MOQJ_7Pow^%m!od$TjVxA=Hh
zK~x`65PKaJ#P;@HB6D{y(J!i}kji1kpPIZVYL*j~#d5;hl{v-Z<b;P+4>8rcyXZV4
zD{?Z}vv7jDs&(B&l0i37VlOS`=t_&D<<ervRcW#Rij<fzUrOY*ON#Z>anjyPit!gE
zMI-f`Ju@Ul4gJKcIlV7Qi1}F(Vhb7PBKgLObCL|xM!s23u4$*HGyZia3ZHZ$@oXms
z9PY%}gPpJ+)5*-x4y^g~16TI^z>bmZ*5s}Cg5KZpZbmzL_u&3%EbqEclW~gL5p}U0
z1IafR`LP!LI974(7{=XHFY-<1OYWn{H_KPGvG=$Q!r>cwdVRs#iLFp#{_wQ-HM~Eq
z!8ksi;+^zCzZyt$@1tK!4JWpa{ebnzbg75+k$P06)Z@gV8YsS~!uCk+fNW}^D@&h@
zSpz%<HsaH9>NnJFl06&YKBNvZes%aoea9=U0hOGvf(G3CT8A(lb|yG5OW25yi|a6*
zej4XNb@+3s7J5hcXIGGC+-fl=sur~kwbT}B@$n2hqV_V2nCx{ms|LYWm^pm02HP@f
zP%B@9-`%T`W?2PoQH4p1s-dr4jryCPa53rwmWO^oOyz5KHot`R=f`jldV=l`AERVn
z1x~$SZ&dgDsFA&g-|z20_enXTWy-N(`)$P4--3zvEz~W#31w<GcegQfFXjS*EY73G
zxeOn&*?)bv1O*#QVE5oGKD(bnU>A1R@i}PtNoHU$^Wsfw8stx;!hb{x9QU0-E$^G|
zOD17<Xd?5=6L_DJfZcxa2;Co#@Tc+UZXJ)TDb#F6#N)3nb(<_^*<6Un5ALMKWJZyP
zBbW;q!M@-K-i|Ye>|iwB4db2E;ApI<j^aKhik0nAu=pAUY>C2w=~1vMj-=-_0)Gr5
zkVw86*%Xdp8WH%{8i}yYQP>>M%+*WD2(3ycJEmX}|5+=hroit{GA?oYQlqi4i9s86
zmtfskv}nh|F(C%Zve7UukHk*B!^{&1LZVU-7P<#w%eFvttE5NQA_xcB)wH#NebT*r
z@NuCx+9!HLeULX~r+UNwofj<q_?kb{1FpB+Q9jZQ2bQ@ZYnwY79o+HL(H%v8?g%S$
z$K1Xi^wN8vr@bFe7wyDXx%~(T3B`XhVMsF$gZ%0+%vB7<_Tc>(;Jpt85_@s%XE5`e
zLXddz0DdnyfQpg((D-vNhThzZ+%<dA?{yH2E(PICX%M3FgD|);5aKF(r8Wjair(T&
zxx10^z#q$=`cq5vN6{63Brv~b&wC$CNc4kB9UrsYV6f5^;gK$|=)Df5pPVuC=o-xY
zx(dxhHqg`Qju9Dd7*gnlziqC#@W>S}C%7U_-4$mBx#Da;SEw1<!Ez6^8EP<Bsm0vk
zxTM%KE7l&C&&V@RxpU$?BHMhZut&p1d!$^YKPA)x76ldv)w0Hwb2d1>U>+`hvPFM#
z&c!_ou-L^6YxT@vKg5juZHh!|Gi+O6h9hgK$9S3H+&(iLe$Onw*OvHJVToDRmRR3t
zft@ofP_I2354+9A)(>X*QEY}5YB$dP&G4y<8Ai-9ga4k{=()rKJ}wqGm~IQvvH)kf
zkCL8ihm9X?kXvNK>}VToTx0{K2{!m7Ys0;&A!gBMGf09y8ej6uNj;p1)Pv?LJv6@&
z2r3o$7)8(T5(89BA#<z}XfXnQ?w^75sam*cqs6_o76Q_=;9{<gWkK4weq0;5ue4Db
zFbRGCPRH-Ax{&Roi}UKb*gIJlr{?P7ruhuiteJtsEA+7Tj~;qWp9ytuJtQ8OiG|dM
z&L$f|Ptgd!;*C(7VT@<=BcGBoMG*hp+}Ujf^GYjBwzfg%tHsm{U9ebkE3T4hH}`VK
zSATanJoA9VTTfu87Z#22!Dz*u=vNwrD!EwLtWLz(&nbwQn2H4oX}H4YgQF(t<f3$p
zFv&pF=?wH9nF(K?EUYTc!r-p-!f<DFU_mxQms7`C!`E<jLXG#%#>`pS7#f;|-KR6L
zcU?OC=-+!gI0J{jW#J!pOljfS*zC;is9~Ade2}>fdoobuoq>Dq8R+{ck<YJ**m912
z5{mQ$`7&3cFa<MzGy7pNb(%YA>=9x1m0=nxb9h5=Aq}eJo5kdtdz0xA)JR1Q?-N|S
zlJRjn-wX9jLAoOQEA&(FgCk*(f==TU+_GeL6*ZrQ3iKPPW}@GkOgvkgiI#PlXxy5K
zj^~+}pOp={g*gbAm5+y)@?ob%pAcsccWF<_H%j#7ZTXwV&cJl++MkX9{ymqFZBBAl
zho>WojB|_`6Ao$FFe>5SnSAqYDDUgZH@9Xom%=QE&tduakeQFk#>H6t;~c8C^biZk
zH!H|D=gBupGc|>RGkwKdG==+ec9YN16fXVgyOGco+gmimczSQnFz=?7`pu~k8e*WV
zhFI;TE<P<(7xh!sg;$N5*f&&7-0!6(`o2{ascEXhCP0-r!>Z!<AXRZCf3S!r-%Oxp
zGx~&zxTdBe?A!;6vg^vC&0blo9HA`M*C~mK%<*enOyAA70U~?#0P*0BqSz)B#dAeP
z=I8Yj)A#igmE-yftz&&eUR!Um`gm_~plfe2iF^}BzDXqCq?2#-cJvZQDtZc|p*=+)
zb(_XBa>AQ^QWFj3gr1I^I5|*GT=?8wJoD-<yi;Yxw<K9{Hd<B~if+P#e3L(2T1+9|
zoWCF~_Fk3}qc2JcQzt1Q?4-nXvW^bNl)BEbLy|(lT~d^#^Ja={Gl<iXCLvy}ln}Z3
zotRIC`9ik&SKo=Hm7UNb-{g>QW|wp#fnS?NzL~-doGI*G82|PMZv6d@4(spu+vPh_
zjN0*yT(kHe@1VGQ@=tEZxx#jsm9=AMZaaSPV}*=%j3Lj|I=18C^matfX~U$s?E0qf
zW?E+}n!~>!VEPx#9K=UzHbtXrP&lIoMK7x{`c^g6#*lBMsgvAgZ-G@EY>TO<81SCS
znVnA0YM@zMh1i2tDAB7%9NEI(yABHV^+?Taz=d9o*!Zd%CcmrkJf4|(n*87D*JBs;
zorNCs+K`nBU)G|yu@-;X|5Q7n4ohd$L5+_^cgQWlwFq5Oi)X`Y;X`(MTT6D4<@XoW
zVyRy(R&l%!Rb!WN4UPoVAS0p%8w1E$@>SR)U4^kztMIR+5?NP1F*D;6T8lp5vfl@U
z@m^~7tfwej@&sQa9;4^mM>sq55gx3np!WFyNB-PHRLxx+d3A?*Eamvzc^l<BZln3_
zEvUKO!U<jaZv1ZGP2a1Kv$+7B!RIk~MHw8E=xqxtK}aq4&>v4R>w%qhi<xI}JPR%>
zGV!y22KKy8gLz;oJEKyVzjy-fyOQwobt2mNyjLrm0Asg!<gJWHv0*$0PKd|Nx;RuH
zjDyqAcy#TI!-~mqI6X5KrPeXHsul_VqY>Ec6ag2*2vls1z|Z(76huT}#o;In*dGOz
zpeWw9N0A+)P`5q`ONK|Wb2|d>)gsW|Edn>chV#ZU9K8-j;@j0oY~GX1Tgzm;{E&>5
z<eS&E$*9vz!Pb!}7|@!GnPeOJ<|y2h3`3R50URj_#<H;e*zh8dJ)40jE)IlN6}z>6
z2I5kFAYL&S?`~geGI?Go`^e1P98c(P^~A;%o=}?OjnzZFnd?FSWrG)PT=By0bskXY
z<BndFw&Kh$7j~|1#Tz$Q=uP9^$J8CWJ>9WC#T|pXd0^H`54d%B;9a0Eu2%)1*&`T(
zYeS%-6^7+ULy;lNuE>NCm>BOvRxLAZX0rGDU@-3IAHe(h>?Y{RoZhCrXt}T#T~zkq
zv(;`G@gC~Mhn>{&{2-_0gKLUjh*;~7!MXJR7WqT>ygw#g_Qz0;Ja3&O3;oeU*9UKX
z=%2~*!N6))jC;Bj8U9=F`OjKd2RP&K=hb|Fz7oGaE=To&Ww;i%29??_=sVs8vWc#k
zJl2(c@UCc8aK+JO&hRm?V?QqUI^>%RWSi@pvD`a3l5ws*vB&!d_OO3sk0pG3bKf2_
zsox}Bvx9s9Sp>5&|A-m->zbk>-xyu!R~a?I7z2upV6)f=V;ku$sWZR|Yu*p(n7~)X
z1c$yG;-7>OJoQYFJlq^Hy5^8EF^43%W2=@SUN1Got`I|5oijvLxCuJa%+Na93|o7f
zA?}YU<`kRa=3G<gIGJE#r4gQuHiR+rge$%172aWpoB4(aOJ|2dlp&154RQOhA^sjC
z+jtt{?Lb3p`(pr;{RZe7s)yI%dN7OAgL#G?X66eR(DyqS0?~s7wzL96#%rNYRSWrJ
zwD4lK76xt7!i*DI_++Gweg4|ej?u=-3T?b`odlad)6qu0`P5q%?P|L4*U@F3qb}UZ
zH`APFVEjrwyjiA;LB+bb#mt<S2Ml09zOjs@zgdR84yDWmoNElNRmO1g{J;F;XI-ta
zVZAltofkto)dlk&xzH26m9Lpwu&BZXVQ05ue#Lft*zJk!v%GMs+YaarIR@Q+(P(x`
zz`7$RFzp?E`@>S9;F5}^XH#)nH60VUcL^BJZrSt<OzcJvjdvCda<lODcNUJ)=c_V5
z8{d{?!`M6<IlOI38<35=2bmE*ihh{1OxTZ4huuhe*oKmGhOuX5L^`I<WL|@9I=sJS
zkm)lZLB9Frmw_YXo6+nA4I$smCEw&<NraPJ60R=i`z-qUN{6Ol`Xt^m&PYRvK^o4+
zrXlJ~8g{Z<^dWO8`g<m_vySh>VvghK*?9C)j7P%yINoq_N6yh-9fv*FV#$lKxM~rD
z89MP8pA(BS3MVkMYcf4F$=Gl%1J}MXyF4Qk7y0)}d!3C+<YOTD#$a3lZj7flQKkTA
zf9JFNHWmN<;$7!D?xV;!KSR^8jBIm?6Ox?{FY?U|@=Z=OdGAa%Qpq>lhtRJyfxQ)a
zys<ZB|LB$+bo9x`&opKfcASIiiXLL`rXIp$kGeQZjyb2teuXufLYccMlU16+(@Ikm
z(N~<*tRaq&Ypxe+2pDOI1${Kc!Pn|yLz%j8cjwrti*BvlNg1e#6je2`tW#C!K2;Uo
zQOxmMqbkg_RmA~%i`S@ef(MDV?<yiZLPfkCs3Ht|4-~~GltnxFrs5y7ieD%R`#2@h
zVy7g^ng)nR3kC=``ifsHQ533g`-?m5(VRuT@nKeR%Yr__Xi0Bz%cQrM@kc=@o>35m
zt_s3`oPrqjvX?mJ)Jx3Ze(HHPSteXggk;Kzn~)QEWSa&p_B4=h{Oh`lF5c8|l4Zrn
z<GdGN!3<*|E!H<k3CmU~(TDq~Q1XrGIcf3Yl9VVbkrGaYQsTf#De*R+{Zn~TV(VNf
zp-0AP4VMr<G9|>2cnOh^C?VocNC=M`fAD18FSNIHVp{|EROFi-<eRhi={t_@#AWi$
zC4&wOoY8@IlXy$@<_Btjen%4dMv{Cpf_zi*w+*sixntVdj>d#`?tR<gK&G)xZHFJH
zHjFn_+uKpTv>nr@a#tnVuxCUY7W=Z3+v5u^C$_-%UJG_qwJ=_t`M!^;d8b&7CoNSt
zF^w6&&DHq#qz3ZmsG&@+!v^k+Hk8yMkDP*E)i{z&mRMSanKhMow4@4$4^^SzekCsP
z^TxlbuxMR1rqxx$bXN_Y<<~-~s2+}b4ftc&0GzMIn3J_AEw6=qT`kTHtYemD9X$Kg
zA>c$U#?Gtd>t8KCed4~%pau&%sv)tY8n0EVaf-X9W75^|SYOSX9A*q#ki)F1;XROD
z3ePK%*jPzDzY>39KA^MoJ#2K|!}H`jyvTWrcgtU){p}Mh|MD2}^B+U{&LeEEu7Hx{
zL;N{>ACr&qF5LYNqGax1*P3#CB;Snkyp88oH&Nt#4RiFb;_=n%Fw43EuWlFM*LV*7
zn7JO#z6F~z#q_b9#uJk~Y(x&;(mOaJE(>*MGI3ZT10mt5h`x{l^OMYL=bcpEvLt44
z@cCt2A`IxYIpY+M33K8hGb$dvE9n__i$j@K9D=6DqJ&<R+u>2nNQ%S-yGU5Rh=Amw
z2uQd@;N7eUl=qK-Yg`om@q1Puih}08D9qRxMZW@dp0Z=;%;9e6Z8&n2B5-j)1Un+b
zaX2O%$?D<Ir5`8SFcR?xlCds<`VIL;@Bj0SpGFF-sp05fjz%NdruV%tIHd>S_Ke-Q
z`q3YkcJ0Jovz?GnVE@3WKqQcLCN~A5rX&#2c7Z7A?ZZqNFWi6cNspu_?riqN#HF6h
zWc0!TMQ_aL@Iqyo7aB~qLzUXii@RHKR(mU6*KI~*(`MdExk66D4U$jX5M|+xUlHy|
zZE}aAG1=y_2ZBs}ksR-fS<KOUf05b3a-q~YLb)#p#jwN><n7pt!MB6(svC8g@Idy*
z?m=}3yA;;!f%THzNaT*mBVY#{RDIEHgAW#b_a<{vPxAK0Xjv}|A>YW8Z$Q2oa?T&!
z$u<%kZ?aDDcvt#Zy%79^dx)Ms2${2;T}NAytg;0~H`c=Nz-lBPSq;xCD|o}QjD25A
z@z{C^+^FaHy>w=N?rN;5U4cstYw&ZNGd3zL!<ZemFr03OuWRiQA83zx(e}tlw1*BE
zXZ{ItPl7#S=sEuHmOajJhlMx1yIyF4NyBF2V5J#SVTva&$TkJW*uBXZg*}Zi+Sdr{
zyBOl>YXe-rZvdN81N<&DfMUJ@T#g#x5O0Q>b{Qbk)c|Q54KQqt0X{kzU^Az1s{!`9
z8(?g%0iORQr-Yk9-oy-A3TDvwWs2Afrr5s46m1bE2>;I*pSKud;WmB9@a8&mg#d9J
z&;Ogn+X$dIi2UGd2&GU%EZ1abl!_rjS`Ba`*nmD&J-F}H!%99rOwz-L)dH^_1VUyC
zF!~I9%0lfZEoLfAz`D;9@Ti*>&ac;kpM*BH%+$ts?x!pwwAm%9jaAN*@MO{?xJv7y
zrk5@%RCIAHLWkP14qiT=f?4rXnR`7Q!vvWpMi*Tg`8^eSSWk~n02$1Gg%SKN55a*g
zL$UtWFwFfo3?E`QL+<+)Y#Z)^smopPG1djEjJD#K@)r7wwxI653ucjT0?9Wz>pijX
zizjl5k3jM2F&M0fMu7JT+*`#P#K$Qp9hi!4t5R_{ms)PmG}!UAx?)HM`o&PgY0E(L
z_AJax&BDk|_PF)t>;K$rT$-H?TeEEFRA<5JHod=&Sx{Esa9?CkU8Rp=I!x5rlS0OM
zPreySzKI~;yll;Y0r_U=?hIIL&%od_?8~~C$UDnK>}6i)AilqP-XjUxc1gIgBnd7{
zlaQ{RhB#enI9aI(+>y*X!UXvKi$~eOID|i;cKazBQ&OTZ*ddZ#<q;Us5e^y7(<kA$
zJ?tp<H62Dk+#%GdM_|XRXzZNCeR^grl%#me&@CBr$T!-zGVp?oGkGj?m+3ot#oPB@
z8}iXyNxuz!H^U?fu=;8$6q8aheQqj}CFsA|l#U^8)TvIfhqsJ-s*CAZ^eP>}bGh5o
z$%OW`4E&&0Bt0t&MdRsJn!&qzW+$p_&A}m0YC660ai=$L&85!MKh#4wy7Umno7mYD
zqAUhYBAYO8*shtK-&-_<-3bkGu|h-4Jw;D3@1>FqG(_BQb#dvXx{xeU7kymR#pMC&
zqV<rP&?es;=5A_b6Z<pIsfu0n-F#iBDnh8;T)s3|v^ES9B72ZnNBzbsR7L#q7${oV
z3x08%viLAjS@djI61SLl6U3b2>y}EQsAzzYG9Dm?JW>?b7BkQ9eSh))NILJh9{BhB
zQ{AXg(jFo*LQA1=UX`+ER@o$#wrI#s6YXekvdSKjy$PX0(GJ;7Bja~IzwaN{<K3V;
z_k(xO_qon>j<z@bH|%1_j#UtiyqWr|N!=z_PHbj3_zZf86V!T&cUNUa5W6(HjFuId
zTV=&c`ic7o%Zg*mWyR2uvO?&}3T-|PR+VLLGuejv&6cPh;?vgu#H^80V$FCdq4r)<
z1e}%>-#R75&6|9md^6*Ml-PJiN;HQ{ikJ{dac;AuFgBJH%a2Nm&$A_k{rxT?FP+{S
z@=aSxSFwPPPt&@JS!e&S8?uxAD!dW*VJ8DKi|_JIDwbXw-{Idee9U(&)8{>SX$LxP
zvP<)AJI0Z3vi`JT68R>tFLP^{V>7m<6`zyZ5K5-`k=2Go&hO+l91N$QSc<n*<z%3Q
zR?IYRMdphZNG|$@JKkSWz*}$2!Y}xB@(W&(Z_E!iuy3y(wzuo?mW(rPU;{p@H)0Xl
zV%e-lc0V_QLOiw(dXGKd8t^os9?NCuYdKbj#8Y+jL)P(@s}8O7zC3NNXTNd-{5zNz
z+^rFx%--T>{(DT-rZ1U{WE}Gr3&OZ}@_vi@y>Ic$fgK5x-Xc)%Efj{oWrhLwPv07$
zOI@b_z((w&rZmQay$RhKkoKw`vY+ZP#k2t}{GO|7xQ7~1j}gD>AbF_{NvCR%Z1Ec3
zo>k(naTONTyuq5eZ?M$A9DR?MBG&IY_JuxU7W7lZ&wq-)_e&6w_!!qcAE9OULk#tN
zfZjjvW25tZ{BFC4=vnu0Mcl^sXE)&5avd|Guj9V?6|}r8g7@=FxKeNtzkHZqwdx!$
zE-b)Py<^B1N0IA$1lrVZ%C4|0`!DrX-(0l1^DZfo&l%&gFkNIK(KZA54e98<Ck?x}
zk4mH8CX4wslI1DrVwM8seM$JVIT0HQ_c1$UA6EWOK<vH*tlgS`GocAMIywQ51}ETL
z&je`J#3P-2vyyx>!hb&=NX8@heH>&n<8UM*4${SOxG$f8OB{_;@pu^(k52V?9QqcA
z*?fNb%$<B_N;WhOW<!O1Q^h{c3-sWGJ0)US+kUjIk3{+4VEEYWMEF?d+>vk0gSN7x
zXDhaR*@BHmJ7G%?uvfz_+}W@T|AK=MqqYfa9sM!;jvxJ`eptxoeurs(h&S>>uq?ed
z_5QFi^v9s_zWA)S0iQ2>;hdotRy}gZ8fNci3vU#^_r|T|>#<F62N2F3Rf!K4t88GN
z!Upyr2BIk;5OeARF@9_yzO@A~A7>NJ4&4bA4R&~|^81w!hFt9~)Hv>fUR4l2ZVJMX
ze>?Hy@-|eKZo&Efn{h=Uko!@3n_mXt+?@cFfAL3_jUOuN*CSDey_DAO`1Z;Tp*P)-
zQ|yK&Ne?(r@Wh?x{&@8z03SzfLY|TzPX6@3xs|Ty`O*oS9M@oah664XEW@~MOCjHT
z30B>-hgqZ@qAo1r&x+MBu2>G=Udu4&H~X?R=Oa{ujL~9??Y)`xyKMowMRUkEc4V9M
zOx{nCZ(hVO`)|hrZ1_1JPaUQ~XC}SByw^TAZxZ{?Y~Yz<^Z)gmqvV_Qa@P1!Y(*xt
zr2b=xqMMf3e%=!1)L6bGS+e_*j6=TJLB6>|zHuerG_0onLf=aOX9D>qoP6`@pe3d>
zkwr`_@oY4`H%gNc`)d+hr%yuB&`G%P%Lb~h)+l;uiBvsHJbz`5x4q2aY-olD3ny^b
zGXah|fCYC<-xgTmeSa%V)v!X#S4)I$9EbJWm_5w=;rC(VFrR#rOg1_+N1#qmV2PT*
zd<g+lYBzQD`cSOV$EPj^C@wKT<x4~S7-@(F6Aa<a{2O%#LwuzVM`eQ%<j2sLqhW-N
zzYOul(+H`z459SH5Lb<jc&jyrIm2ei+-e4=H)dG5l$y?Zb1ZqFgRf_G;M1;yEV)6r
zDx=FAx53c)Hw6E=4}-U(9;|2T!}ixIOtE)@#V>crJn_I7TW^fY@g^&*hu363n4b5;
zn9LYBzmCKHo{1QcosAIk&F6bL*xQHqhf{M=vy&Q52Q{3@%*j#8!@Fp5$cH@mu0IU3
zOnUa(4<o!6Z?|V3!9wy)O!HxUsyYl8YN{r~a}ej71FN|Qm?wGw&E2@K(L4x`0qlbt
zdJus}sdYR)h=*SeqRl!FcP-OkPVHq`Yz8}HGI&dnftf|j&H9`H6J+An+)U_@Z&r|R
z9LP5VmD%~Co{9s7Nf^dEgy1;|(6@_6om4!0{o`O88H?fFqcA%u5(_gU@hdA5=TAg3
zUp5lTvC-&O8I7xPQS4$!V23U58zPg@*`C3^QgR;oX7cSk%)60?`mj8hayR$f?;v`c
z(%a*50LzCQz}aWnuo#|=8Ns<=YduUlveC0|7UC7NP&y<F@-FP9Ny|d)(JZJaWTR*s
z`+ZZXPvz%frN<#WJbxGpYmed~wVOTON1?UhDCTXW4`@q1a_F&sq+W!}q4Z8Zl@^`X
zdW-xdRUx^7eNfD>k!bHH0^_yCobdj_j{R1>m-H9K%<{YOQ(L@vp)HC}({tmjEus{(
z3E=%i0yBzF_2KVn)Dj(Mw1iZ!me?_sU7GA-m};XbOd2)Bt-~53R8m9C+odiP)~bme
zn|L?1OjY;~Qf2-y`=xHFi1XnpV$&aGp}1FBj8#(>kq?x_OjRWzS*s|n<SUB3N{XUA
zLO~p~mlxUe-I!;}iJs(}lF2<qtaMKib3#@q^^+B!d&-Ixe`Q4TGFj1mDH&%zKi*eX
zq$taZ4n0}1>aMh?4V4zN!+HpJs~#e3vy>P@zFDLvB}RUrE>kZlZs$pg@-#_N7Dv|c
zl@P<+*b%-+LKrWU5C)uQOX)M#mlSFT$vY>yh*0v)53)@wr-c*C8O%v6?8M#m-`ROb
z|M0-?C@k-QET@@ce5nJuB^~g&)`59Xm_Z!Sj(0Zg7)ic)bb^edL=W-PR&=jwMOAzo
zqt}>)L$>)xEoWgu8|>z_Q4eaypjFL?n%9h-X)VwzXu&6&Z?Iba6)RSKfs^_d8149i
zm)uW91T>)6m3l0WsfSEjJ+;Al^iFPIcXR{hSMVOGz5)M<MmXm45%st`p%#8i>Tv8p
z4GJIB;^*-?s2-_<40&f<Mh&xAYcRg37Hb|g;M$i)W<*hE`9Y7*yti<7;ofOAnT9$|
z@VK|I8t@jyfBCgb8!@`D5zDyKTHxP^5^6IJ%;`J2p#g(*dGk$PI(4-kFYnc(=zBfh
zc{M=hL_NY3>+z$e4)+r4@Gz?ee@DH>S-(o0)qjP}%U>cc@&(-YmceIIDSGsJj%zN@
z@TU4H$}OHEt*C^#&W}-MK!0)AL#Vbsz=Zer(QDIv`tR@IpX6;A<lMy4%o{KYWWVju
zYq;-z8P=;WV&Zjncqd(CPcl9Kdke6)e?E@t9)mP>(J34FytMx?Zp0lzQuhPQqRv8(
z?db@zNymie=_qJre&3J`xOGX#*GXxp5ve#eC>5Lkq+nce3Q8ULyuh8)iT-@8bY>s3
zLH6NT&waQMN6p4A0nh2X(KX_(NhbleeYmH(8IO-S@z@<24_R`{F!GHR@0S(@#$m2~
z9CeB~WG{-tE3%Md0kxam@z^#z9yV{|5L+0BT*G**;PcnWp7F@7jz@}Q0-RM7VL^ZL
zyPi?(^9qLd=bbR1?vXQkCtkGdz<BB!&TF@#BzFrkCvU-jH@0AzB0ZIRnB^O?1)Agp
znX%-TvwqC!_CwWbKm0KF!{FY2NP6gtU&0^#0)4T2v@c9+eK3`;fv*npf<Xbhl#aMV
z^OHMzkZ)ElS`X8j_1NX^gSGS+f2sF@`hX2MVikzM*g&}648&8NK=k_*fHA>-<Q?v>
zCb?s3gd5y?x?!=kE5=#vLXYb~s8SB1{<sZ8M{k3I)fVWK2Esrl5EsUCCv{*0yT`pT
z_=gvKyLn-5Ur!9Q@&JCi;asU346eGt@xB`pUi16wbcfnRclbYc!+5`S-1oX5Ey)>@
zZ(U*B?1ZLp2WW?pZ~o`wjah;R3l<}^zdh3bv%}Dz?9P0+5Q8Qxr8i;;f_BeGyv%&e
zBj1FPZ{9tghm`FL;2X67zsNMDWE*o%eF}5`__6bI7ohd<G{m+sLwNcWMA$OlrQQZF
zifoXW%8u-OYdFMN<NGLUOyUm8v60;J#1hx8TB0Ds62~}c`z&!L(h@GAmN>k{5)a8Y
z=hj)`*DCVP5=&&*Q^&EhL@Ry1_Boa)6J(>o<eP66xID%J?iB(pWx&;!6VTs(0<5$q
zV4(_ctR>8_QGG1)%SXesax}YnO>t+I3AC4vLSxTS^zE5q$t5#54jhN}spD|emHVje
z<M1wM9Q32c!S*dZ#1jR6jHL&Me8Z=B+~^|kOw|CAuk~SEtPhU_1NaUx#G!tMn5k!o
zS#J%{`N4pBhz3aTH^fQu%}N6!-1%z=)ld9&`wSspV2Jx2WT0L~Fx_hkUz4%0SZ#(n
zjY0VD{y_YsX0xYI2PsE&p!ZRSzMFxNQ`E(jBZKka&k*!pF$@=n55voknh4pX1$PA}
z{OIR~w0j=-Y3GfiJa6dtU5|LzjeMQthh2ps*v@}jeUta&H#OUr({s>L#QQ_3TuAX}
zw(mxEpPXgRW^*ocO%K3yOD<dwvSTha2iqSoH(WUv7j$y*O*a=s`5BnN=Zg=_i8#dP
zigmx($;R0@kzO8s`cvqWiB!!&fA@6U3r<DIj#R7(NyYG=DUg|*ip0~Ym|er?m!GL<
zGf0DNK6gyqQ~e#qTq#TDozG$31oJ|__|t<ko~%=ojP&FLe7q2i(=Vd%r7eQ28HxBw
z>~&L$gxuyx3<+TeN_-@qaw-xdnc*0T=#!B!so?vO(U|=z8re}%SdtqLh12`6YF;uz
zH)mnhA7-y#$wWwJ227`9;DHuLA_G$oCu5dNGF+UKu|7K)rD563SI)xT9yvIwp9RZq
z?4gn5z2?v?d~nL5Z<)MvnS1`;*|_OOR?497nS8VSXC78xJ&gI}n?2NSBFQ)V*<qSR
zzPS~ck7c*=(X7b6(Q0Px>@PxoxwM#@*IR7mjnbBG8e$GR6qd{I?mJyujLy&&*Ufo9
zHLAau`cGT@e5@_HonzOOy|x%%)KB<*(Gs3nTB0REOYB`srpe(>%12AgV^4TU4=r(N
zhNj4=<8CTjLo9Dn7w0#riz#PRg$I4b!{(}rZmO!H=oRz)uB(V%Ze$!6W$}B2vY08a
zEGAr65{~rV9H~(h9>*2M2~|b0Yny^FXIAlc-h=z>mlNTm**P_h`c03X+#Sh^Q!27z
zbPrkK`CUd#|Hdr8w=%++k7s_!h{W!);^lE^q2enoJmY$Zb<=x@StU|J#ZO8oESC}^
z$Tum2=q=WyJ|izFg1btJ=yD0+79k;Q*YYNOwS>?i+o*9eIm^j7k-Czi!;?9G<Nm>d
znoTpA<})XXBg;`P>O`|0Z=?o)=WDbMIOXu}JGTRNe5?%bKw3-(QV+JnGlm(()7#Oc
z(~j%E+i>zs8>aT52j_AthAOwgC%O%-0rVG>Z!TrDu{*aFTjsYwhM6||rnO+uOfrsj
zGwl9rLF}EcaM=D8hqb<9g!C5#9A!@eIb+YX281V3w^>w=>Z$c`VIJPG;09bj#yh0P
z4H#b9fRXnaP!mQC<-Z1e*inmLE;SesS&c8tsv)(r2JTa8@WZACLs!<|AUWgY_F4q-
zhH0Z4b7@XA!svY?#+%SzW5e4gyoH9*TMVRDbGGMO=K3_k^K2vc^o@vK--z;=jqIf6
z?<;7)iiKn(?xg0Ps)rT#UyIV~@$q;){Ilv2!0&UnR6QK;)<JhxEfiyF;QzJ~ha@Y|
ztosVf4!poonR0Z7mSJS=bM$L@hKiZbU~%;+)Vh|SX7*!DKKTe$ZN=#R^&#}WJwVax
z`{=j%KGI*`M*q@Vc;bH(IsrGBDSn-PimT`_D#AsVi#TI=7VGs+;<G1vQVa?(L@6Ji
zh91L^sYh|ol|3$Vxu1$YgdNLs(RW}L96i!uWSow@z0z@#SvRf|(s3gq4YznB)x@5t
z4T`DEK}<r2O(HbjutRd~ek5w_$E37<STbWD?)*!@kgx=J2zrFC#beJ%at{483*?w_
zcR3#G$Txk1nFlj49{T&@&^{v$lLp5ku^0JfUL3N9#i5uVJ75)uvDW;2*LV<^FbC9k
zsP))>j6)0`$2P@5t}Pl}8iG(UCx{t(LCno&7tiaR&<@^-=J`9(`FRJL@9)6mobA}L
zV+$H9H)F=9K)6V3g2Unf$oB|9#8H2!Y52o8)enuEm}|Gt59<y6;PuBB64!hYcfc3l
ze0=Gx@WtgjK1e#}1!*lWG)A~%(_}ZO&T_?_&Fe7yj|(%^Trg919R?Tp;K&glD8BN+
z%bWbQH@xUW_d?GaFWleijp5t*I=XNJPFpg|*UJrqFS)|5dL3L;)?s^E5L(G-X$!W&
zY{^!PQs082(}B?axe4!|1|aXZ4|ZMfhJ1q;PJi*lynpPb9P5F`CO1??y5aCyH?&@N
z!>ep}Y&qqQ^F!S6ip(=_+dAeIy5M+(Gulo&;T!dvC%O*g`^C^&z63dE76U2v=pJB)
zK3<EkI%XkC`mh&VeIYy|7oa$dH%5)~@KfCqGd|lw_BS=0v-6M@!CNWvjTPDE8rkL+
z_f-qZXCk+G8efx6Mt5NYN!Y-7t__ab*<j)sX1gTXAdUW;>ExT|#@4)Zr_bdD@2wxQ
zUpv7Pl8KynON`^F?6rgr`KCUAOtZm~J<gW!SZPVmC-+rzEOBs_C9^a6IkL@WV@uTb
zvqax_7Fa>P`QayUXCJU4nzz+9=1^E~hHXh>ac}TgwEZ*1utTPp=xGXv`zH9FYyzF=
zQP^5*%<Lv(1b7-Foqph?twwweV}y(?Mo?HY4)T%XFrr}`R#*u*kD&*Ld~>4}c<*U|
z?H&3US*DNMC-l(}ZGb~MhKNw7HuJ`S9d*1De@bu84FlZUV}MJe45+^u;Oj?2jIJ@n
zzMF>FGS(0h^@j9(86oqtDf0}*;3Wp3UuAzBE9?)uMFY^(HUMGaI`~$v1IvFpDE)6B
zTF&d@%FiMAF?kq14AjKWV`>;FtIqeeQA%$9E9D7EddA*acq6332dXnSLW_UT3*-Ia
z^M@Vux0#h-z}rQu9JpNMe)?Yy4jAUb!X=lln{uJ{h8j*B_f;PB9v5XJfP0*jshL1v
zB6B2|r$E*?`8@@hT4a;Ce16a&-waPj3H@>FxjPOw;~nF4@=8!DjJNULlCxlYDx_Ld
z@Vz4iA3{^<p<$*(75y^JsTkNN6+>Ppq4skURJ4<^b!0NUcqcG!OEON!rojJcGPWcp
zAoP4RI?JPQwK)R2tM~HnYb4|qBN65wiFxFkGr^HqN46>9JSE>~k#F2CkZsOHVcy<o
z{CXLUqwAuOXB30Ngm_#!whx}Ol3_X{8B<p#Bh@h(9q!5avpyNxzGR;T$!J-`g(msN
zmHwTyq-5-PM~11QmPG%~i>_G+{!1Ps-^5JI!btK>zv3+Hlh4LOr#$pe&%?fgJa&@h
zG3((leNsnJ>->LwqeHfFjpLoYe?H8vvv2fgK0ealliqZKxh0o*-+BepW3Rw9sU0_`
z^bu`NO2YV;vM5l}5)SkV+ifPl7-)$oDYDN~P0>)KDW3BE%pxsuGF?l!t<w^QYFgq@
z6T34XYl^u6)NQ6}ild5}qCHDPY+?t)hFj{Q|64V&Do<6^P`e4#RTZD8-HbWK+o=W2
zyh&3Qhy9erv%$(@d3R+Yb4p2+4pb7?suhJUJEs)MH^UnGh|y!@#YlEd4dPAs;8Aj7
zvTHA)-nFL~aFX7dKC+@ur;Nz`A|t9BWQ4^F8Btp-BR;&75mU%G1_jch*LrCYO}>$s
z)<d*Ek`n8Er9>wA=JYrzA+I4N$|TvL*&-=av?Rs%8VTW5#7sYSO+BH0lej=av~d2%
zH|Bi5fH}tDDiT780;|U8zcA)*>Mps)Gr6nST-J%n1;3z4zER91(<F5uG_ixXLLJx{
z)q#;}9e8}I9lDw9#*Ajqf_XbOvqS0=^_#Q{Z77vxrp?h-M05A_(2!Yv)7bxQ)rMnj
zt(efS6(+GQ%p+|9IA6E7z>|E_cT_Xx{r3$Il)mEr-!E{i{|u#NpK#yo9lENMUHt1&
zqh1fSA9aYhUx&trby(U^hff|2&_2-sb80th0vk{;ssRty*23AY2G>+8aeGD;28^%9
zZgS4-wbl4Js~Vp+)j)1aEq#j(NSNOUuk1#2ec6a8`hi!|H(aIh7BA)BBK>hA6!`Jj
z^Sp&3+wAmg#Nnxp_^#cEKV+(I<f$HG8gPYssLFNq>>{m4cTe(-OFgdlsfY7&`hZT>
z;dDbS-k8*4>&Mq<*<6X&UteL`sF&<dC`T`;ava)PhV8ATh^u{uO@p6dtHD!zsAnI;
zgva=}`w_<8EJkb<bBlu?z-#+GoLhbu8Sif6?1fu6GVvw`FTa6b=GT#N`zo_Eim<)<
zJgP^Y!oG(mu*UH?t}yf5RyH5Mh8@F4tD^w#J5!m9HQE0V7U}0An6Ev4+oz*ah1yLE
z@4L&>uwrOBy|8I$ZBK!#LJFVrlMz*&NN+$Q{v6%Uz1V*2zqyb3Gy8B&ejogPCctEK
z0+Qavv#Tl|m+j+Gp+UxxO+dGzc)ZPu$E$hqDDF>vCOHl&BV(~cCl()u#^R)2EUc<x
zd8-tQXN|EqKQRtpsOePY$H8rQ96KT6u$*k;`!Wu{H%22*a~B-5gP>;{gjgxwJvHsb
z&kMXG+PM>n^LAp?`yISX+ky0fJ20UC78I}EjNUby5LF%k+eHBg_%8q}v;DE5k3Xch
z`k~m(4<=*$pw!ckKVN*|e%cp%lYEir<qM<U8}OxMJ^H44K|he1jh!1}D_jv_>I$C%
z7er~h;AE8xd_IzMwAW#mvkxqMz47|07Zi)UkbcSwnir`1G<rdM32&}@cw;YjN*&Fv
zSarme*<#MvtnY+VOi~H+-icMzZd!+I#Y(j;_~ICd?<t$mbTR<ns`+(qd1E7eG}FpG
z;aBN_2`U~ax<jtn;D$Me+;HNY8*Y<p#+$kGj?4`<wXQf7yAHVvoq2ENjEL(Fm_B$V
z`i)tMVvEI)+q)QpKG`F@)(-x~i{SBmA;yz!E}dr%@s9<l3*^1j&iTwUo`<dEn`F(Y
zxZ7rnQ{CplY3@AqpE?g=R%D$C^I%YIi)G^{Q;(U14kdDkswHwvZIEhdgJyEh%?UOb
z62tdTS>vq-{Wo2$vE!Q+8lG9=$|dfCqAYQjJE?g3ZuZ7mLTisD+O}EZH2Ef-d{exP
zn#~MLbj-EHv`O?MTUa8_$`W$psqJv@w48in_|O8PRG@W~z;<dj0rK?f=9@uvtr_O)
z(yu#kEJhiP!Qca?7`V<9Q!GsRyk>%3yGLQ>Yhz{-8)M2kV|+|E!dLRmW%9<{SVJ8D
zY5*03aag)#943{FLsdC&w*>Sb(MLSk0Np<fhg#WiEGZkwo5JBx|IZMs<qh%fl>wCR
z8BjYgz|}*%7pLb~$<zSLl?`}%Wq@wnVKrYigr<fevL+hBO4AgIy+`BghCv9p+aIRk
z{n7bc8}ED-(Q~IF=7wqG>>3@k{L#S&seu^KZ7`g_55ck;O>`!yqufaaE=_9CT&NBk
zH&@8glcP4;i+#hq@&CIXw+3!RlBOSq?(>KH&k#6m4u$r{c*HMELb-b?M*rg8cK|&l
zOLH+WJ{LJ&xp?Q718^U_c6BxurDkGJB(oM??1vwJ)>l8G2E*qESq<JC%uYqQLK@E3
zrZbl^8x#9xvG0so=VTbEt*O|~(IDTP<=DSZp}v=btJHB`7o{TDBZW8SDX1z*!s9Us
zSQs9U!Xxom`5_)^E%EH?Nx;qY1Wb!bz=N~V=;gE<`)wnjb$u@?r6O^Pd?QP~xxXM1
z>&Z7wTd3iXae8xXk|WW1HWIVRHs|(4Lz--J#3c$g<eLqR(dcO&gMDP2`I*$Ar&6bx
zmW({k>gmY{B=4M=os8b(n<Vm$D*wzuo0D<dG8xi?(lELt3wqp3-R#K3`5)Anbolm^
zEX>`Wh0FY$jue?@Lmpg`^I)Gw@AAt$Ec<vEi-V5P4}KI^+>at`&rx{j=3@`{Q=J#{
z;rBZq8v+V2+weGsEjW%NKF684cmf(*Pk~9@=-*L_CUz~I`%JBbcTI{dDnk93s`!_x
zCcH*#h_BW3)7;P$)23^QX-Qh5l--$8eaSctnque!O)=h8Q~a~yU3hOzal=SM^d#Ff
z`KpU#O*JtkRaH38Q5DmaRmFdGD&lhl=Qmj@QCZ|{R2IjzmBr?+%AzDmNpu;eB+iv6
z3N2?v(WIy-UYGR|yA0(;Zke1&3zQQw+H&F@y~I7f%Zh`?WQ8hsQ*S@Zi1J1mkyari
z{@s)j$Jxnn__d6f*d!x%-)Db#h_pCuE-j8k_7H)+dWe3zq{M`!Qo?tdl(?bB{F`o4
zVn6w2SS9bJnC~ZjLqZIoe)E<LGn@H0Am2ohZ?woaDdd}1?h;~$vV;g={$GLiU-Xmz
zi(lh^Bc5!t;D0>~WSiOyat--r1^K2I`DXjx4)l4{4#oU-RHwJ&kX1X9P1|8YzRCaB
zhR93IDyHw|cuFfmqFQk}yp_4Ptw^Q6W*X<s-4=M?X<<H63);yydC_E?pcafXXol*`
zZ*bB3islhtFzPgW7Scc9QqTt|*wsO2VjZ@wt;6d>b=-;9K_RpbVWxFZ3#-MoyETy6
zNw#5L-h$B$c;{XNE9y1D2^Gv1tw7I1m8dLujpw>=pkY!C*WxOS-(Q16sD;a~dgLcI
zLO+0aP}F92_HD$YpN&v!ZbZ<rMo2p_yT+~&x5w}<eFh%~G_qf?0aE0t+EoqssMr8$
zk9v5v)Il$z4)PUsaHrqr@Ao>)$f-l0Zgm(oy%x2rtDvy!HIDwM!0fV@XjFcQJ>lh8
z&{>8%8_Q7psT6CS80^*k8SYh;AnorH{55!tbL$@=ZhtZQ+<J(QPw(U6&U-kr>Mk?G
zc?({23q!1LVx8>`w4S~ODW$7;GQ0?~5zOEGauWXzF-v^eaXcMefE!)&(a!Aj$e+CD
zRN=iM^STTy4&jwH`v&B)F@jo6&-PTvR-|GcwdOY77mVQDxA*ZBl(Z+~Wmq!vff7+H
zlZe~t`yp$*pZzWS;5D9ke5VpHS3d#%rSa(ZIUe&2;(22bhq6O)INLn|b>y4(sqwtY
zj>nvS+%J8JL(YI$dJtmK(mxi_WSo1Yv6xvA3tjR}xmg@`Y>C6MHF5aRFAh?-V$n1u
z7Ow5l_~Wt*o!^7-(mx2B<^>^jU=Vsz_ZV@C-kS|O5jt=uBpTR_eRT(dNAAG<0XtwN
z8HhLk1>#zD0B+9@zzc~0cC-3JV}&1FxR*L;<%jpf{jfpG56)G-C~(_|r+FK(FxLlH
zB7E?mo;O6V+%a;Q8)CY;Vd`C1G_<*(+cFm<$-4k8&JaN^xZml5!S-(WXy}cdhrQT$
z=7papys(bi&+X^*eE##o<2EmBr%$HfhASq0bU@{{HT?6g!3_gPL=RYtrhsiIEZ&02
zy*5*~--HC~O%TBwu`$sHHFv!6hs<<4-4#Yz>tMLn1xhLGKJaj*N5~Be^W1Rsp&KSP
zaQBtvhLESO%*<Jb)0)mO+`61UZx*xDa4EKEE=HKtVyKX9bjH|2b(kGo)ED93h=nNd
zUx4yS^O3#87Sre1V*FfNv@EbixSAzis?hJky;C;35bi271E$sj{i`gP9c2N_7Z!-&
zqj5De@aipK*3}X{bS*Kj-Wt`sfeIXLgFkiF5NEA%pM3M@&qR8qtg!T&B|b;e8^c?r
zebj55`L>6rB~FrUwBB0qzh{Bl9sK$~Es#Q<xd&=Bd~6)cUuS6v3-Zn0VU{>SzEON&
zLBE(l(|_Y}M0Fg_Hk%=#zzpr7W-!+=L+<dg*g1F%Mr4|D$7%}GMeK+28HL*g#u(yk
zj9N!y#4k67?;|5TC*K^nVF-s9L(Fd)fhm_pz-!eQXo|5YayDW5m<i;68bIOAaNI2%
zj>K)l5s*6^Yd7fOU{^!@B@>MfH-PR<1N1&;02A`f2=dMS-Uc|PV1Pfoxmq>J0F7jy
z0lnzau`q;>qA79}O!4=*33gxakGTu`qcuzsfj)}pF-H-n$0?$Vhazt@6!BlBB9vz-
z!ECWIY&;bq@mY~RUu7ITpbAM7bv_@u!c*H5hv9|2X5M)6bv;%|Ze*8*FLqtrge~8x
z-$aFS7s$>K>aRM%+&w3!LC%MHneTIOcnEh-Ud+oJk^>#@Y#1`9L4z5bc6;_?hG{ZN
zxUW&amx4v#QqZcBik84s9QwhVq<85U)RBSNOVaU)`=}B8xYri$qsTWmIn&5D{ob<s
zttAD6`1x<MQjn&y9nt2@y>#O)X<;O04duS+Z6xfHqY&>9jU}6+nTr+;qwUcsS-Ts$
zjuB8G-{_KWr0(ry=3XT3c8`RCL?k|X@4;Jo_<L}Iy6wS}fZce)OshjFk?f3%LeSDE
zY$f0P;4CNK$dYfO$TueBo0;UBC7gVYE*VFfj1xOL8Qq3Z&ynSB$~&3&h2)*Jskp0|
zhGV1BuyPjtinlWH>>x9Ql(Ojc<GyNs7MuoB>#E>w=hy?VpUp13HGHHOxzD{kl<=nh
zN#zloP&$ej>tlE;lMlIoeBMIl;|R6(1@tlBS3S<3FU()<a}x8W7edSEBJ77=hU@N!
z$ldcC|GobUqeWeXkBYQNt&|l>a%$q>JvA{sQA32YOX{|crpVOS6gy^XiG|8qVqJx%
zs0`H<ZmTuL^zoWPeW!*PJVITJxS=Lmrl^UL&s4?9&8iH|RuwwzW4L@<Ma-t3xN4WO
z7!^v!@l+NDTFRolQ%T4kQxZ3*-{f6U6qB6TtGTnUxN(483?}lT>5-hU=FQY9=K0ln
z_Yy<eWW~O8`fGm5h@l7Aq1jJHtdNuumxsy-pFL!qS2E%Td1v|qX;Bv>Ekeg|PZirk
zboONS%?>HywpdD3Pn8niRHVd0>NoG3B}G>aNumBgLIf5_h?q4JV#Q1eamq?Ud>kVo
zlE^q}<eSNyALJXi|1nM<dU8U3qYvjm&~IFg_=SGlNokO8qM1oNmwYpfd{e`n)JH3N
zZwlMdZ4Uj!HtjfL)(+{RWE=8Lqf|TmFSkLfq7{E4TOk+Pish!QsC?CezvV6TTrk&$
zeB(g28Ow>f-2y=^$04)@ll7Z1=LK)W)tP(K^AldrYJzK36Xy2&fXKpHcz&#bN@^`0
zyVvpg<P|DypFuJH8CpA@;m~+?Y_F?9tx^qY`q#jIaSg2JltD_Z0#>~$u(G5AHHww+
zomR;_{5QDLtqNT;s(43H1JhU4XxUqhMw=>VEUALdf*PDFY{1a6Mz{nuV)jDj^O0+;
z`*6ND!0t~2-d=9N`T1li!8>#Mo$t({Z~0ao)K}M`YEUg!UaP^qIW;)zSOZ`E8Wf$V
zhPg>4&YY^i)pxJ(@6JnTwY|WGelKAEqYN+H%dq=RDI(^V;-}#=C?6@oaOo1}PCdqI
zuSa;dz8G0Y9-^Y~K30U^L;L)@_<QL#7GJ#uC8wLXlz5%K;cMi&E68&ygr)6iWIZ^E
z;>{;;dn~g;^$M^@GM~9+%-{OT9$v*G+`%12nEU}0E9PR|zbt5*vvc82Dr&Z-U|MD}
zhRCI2E?=9i@=QS@b=s8)Nw9g8h_o(=c+Xq!%|rG>_uxLb4&R5LN%Rp9<Lj%;cx>eE
zsa!D@vCHGoH8u`om=S!uEFKLh@sRe7XQyL4`qafCu}dr_c8kSSnOKx@PjzHlEb~an
zIbGurvNVnvX3Q5}#{PucvFPK>J=TX91hq!P)O8nb{tQA=KoCYP3_`qa5VnzTjtA|;
z7R4Rt(At5Zo$M#rxC48QcHj^BM)$=gsK#u9-!Xr@nZ$0%e}1?a;fKET=`Nk&hYy4N
zpfk`920wf;VS_iKw|U{hWZqJ@xntKSH+HDI(QDv_Mrk)J_~;7hLRV^#E|_iY!dwhz
zSUWo7kDD`dVqFk8WF4OJZly5Ri#;1&_;$n#$4+`7^M)6K_EXbYz`azW3#v^V@HpN9
z(+VBfTj78fWe51@uZDfMHPE~i2)joCkjV8%m9`(GW^cryWFI)4amOzwH}o=b!tu*%
zaapW|%TGrvQ*uIKk}Dqlaz%8$8?N7SgYr~*Z^~SmgTjxMIYD7QbAW9Y^X_CR*2*qM
zuLt({>}n6h*|B?N5uZyJ!g(`qyYI}0!-;vwv9pEqY+JO?vc;J9bKu+C66eV@c4QnK
zYBP6UTae`~@Q7SfSZ;wAoY~}?+yCPo^3CclmRLi+@vXInn;G{~)N=m4`QJ^rHLU%u
z5&UZ+vtO(beb^GaBgjfSEU}onk2AmSk~j-U<XFJ^oCUsKw7|Ga7VIeCe@mJSV`7Oe
zL+MHWVu8S%3AhwK0imH2@HJ`zl#(W3m-={2UpEdRHs+|SHp9~s|Hn56d8;&L!Dw7s
zWlAnG#W}lCSUY+Y^O=mXoSMxb^368hH_cjY3@tljH10IQxur&ki!y}tnGx8SJ_5sc
zjle#S5lH?x94QC%*zu&t>~B4ISn0t`TMu=r!|`jq9&Gc5A-~iB<AV)QI^F=+pX(#)
zfIbei=_5kg0L$bIn2BtFk(LJZ@BQB`dHERwT*)`U;Q|8$^e{pCk$&_c_s808iZB{W
zmg%nunUB(tUnhg5(`Au8v?pqfdSPg+9B!G)BkzeKzN#xhFIN?}l3nn<r#rgu^Z;Lh
zAg#YQGvn98Y4ip-eA@-}>A_fbEg0LfLiuM4W41>)<hXN2eJU1i&Om!@4u-DE!Ht1A
z@N?i@QECFzrS~DJ=YFL0r4L1)dnqUCH+<bQVRs6Se@MX?N9s4V%>E2W!<sqin6)z%
z34W=_2}nf)8OMl><3zrZBH#GFPC>C(DsRD)G3w_wTyhP9x=|2(?*_xoVHaMn4#UYu
zVYuAFYz1aQJz5%pr7I#3vMK^g93n7WHUcv^FQ-Mo@dg=(v!?5AJThg+Nkk}aScc&C
z#1M9|hoE^*2!^DEz~FWm_C5$hG#O{tswk{>h=T09Xw<~VBau7yg*M5Mo1BboQ<8Cn
zQ^h%9!uxSe_S9WW<Q_a3e)7q<KRg*mb5gNlGuez;?Z#$lh*3yKTy!S-sL;naISZj3
zSvYzr8*MM?zZr7?<H<LZ*3uhIzIk=wAief^@ME5!8Fy42e~<Dx<QP1C^WpU@AD@S_
zlgos8^4pKYpw|gB3^|W)GtR@q;XF=@OE{=|1!XDBuxCEnsF7v3C@UeR4w4priLzq2
zj+!`pN=-bwq#@j!HH70|c7H2r3YTS?;*7bbh;z{tcLs1jHAh2aZd4byb=Af5QdP0U
zSygEDR}}-hFwbw1ipV!s5w4}mA~#D}^z=~{4|o&)V6C!n=6>qIYbCLTJF?+XibC>K
zU$Nd<LHv<d5I$Lb#NcXqF_T^4>MnAk-{s!I;czdJ*QJ-3kR&T^x5|h~=Ioe~BHJ`b
z3$12p(aldr6vxSkxl3h42KAf&R!NIA6KQesWDl`|e6!9;N_@AM5}$@jiEe$R#7hY&
zVafZbc@+{u;gp2f<0T;mQoETnLqhaoKSQ>rgs|%(Ax?hiD!#nvD$Wn>Djt3Q3)RR@
z^bhI8iQrBc%<II0Ii2`#N+&y3I&q7P)0cc>NbTm+)DBF()Q&rj?cBk(BW`^=A||&(
zfqe6YeB=D44Ia;0G1apb3d|qAw7C^;j9PJ-eDkWT1&>Qw@PT}@ifm)f8J5|CbJTFo
zkZ=4)HY0iueZ_Y_<Br)!gcvtr>m}yy-DyJioLU?vvqX%kg<afpoN^=EynTwGKPCA4
zpoCp5CFslBqafW<ET2@046jl=d{>HROUtov`b%V9t-wX;N`#HC#KquB_VT>J&vjKu
zPOic|vd4q1)tEk}3JYdcq55DI#6Iq?5^E8x*~q;-@4sI+@HKh^7V=i=>QUa6C$Lvx
zXahXnGG}K4eZ}<G<eSt%dQ2VOf3D%b`5Nfx)-YTDHFp22WS?gx;_53fbXEm(pI#yB
z%S)&{dx2-lFK}O_99MPA&~I`n2J1b?vw_br)u#lS?N5-`<uPV0dxT>%i}BXxA-)#f
zNAAOWuoic*;rDHn-nqpajhpx!bRD6`uj2NsE9`|V#1yU5?C&~>o7N{Vjyk95zyc&H
z=R^7RQJgMhHt}2DcW!5o)U`BBmq@{@>(pGIbEnMwzQ85C|MpA9DrVd~D@#F<NWm5A
zw7Q#<s3Wjf;pcwL*}ESj2kghgxP9z@r_q7jQc8}Q|6d%ggva7{eGHC-#vr_Z4BpL)
z##*&#oOF*y;kIblRz#!gh!`Bmh{1pxF$g8+D5z1pv5IA9M+`>t_wF0WuVEVpcl$Ve
zzY)vlwOCx@uG}~;8WwWV?3CDr2^m4yx-AIyYq+EOwi826?ZCe=TiLI?iQ2~|93SbA
z9zXn$w%s2`_<FT?*CxoN`lE!t;dA`B5<NFH_C5&NwVt`r>tW&OjmIy%5E$;o++Pp0
zPV~S*?xV7r+;Aw}4I>oY=zDQR><co(aTn&KxZtX(Gd9UOBVd3t%JiL)IocVEyj;+u
za2;eF+@YStUD7@;_$7EEKF<^KC5}inT!Wxt4sh3UKtYHDVh_{nd&B{upH^e^lGO-v
zbHLJ74mf=|0PdUpq5r`b3JW*FJcF5SHBRsfcfx}{PSjM{Q`v7Ve)>4_R@spra#vih
zcE#QpH<+w<;~vBfe_y%cqpmBq1v}yQu;p+zSd1Y9mqP82Jr1PV!&S;2qyH{qkJdu?
zWG%poo6Pa;Iv>vs<}s6bE*5T@gY%km=zpApS>&5P<PT%=%?(ZtIpo4C3#=vIOeWhn
za?bH>Cw}}NeK1L6kwx?cuWho1^H>`gsaj*xrHS;(PK3_YiP*&ZshqA>FiEw9F1<Am
zO)XK$EMWIAvJv^FHk({SzIji+F(ThcQongH$P#bNEMR-u0%vL_;84&6=0;3luE}`p
z95@bJOU#j(WR9x&=J?m&9Iu+p@a={fyQs{Np<sqb7NfCrxhdRDOkrrxyKMbYnEKHe
z73szpO}>dG-`JCHR2CT{l=@BW5+kJS;r{5v2<DNFfLhQ9EO+N4_fg6>^_Zoghg}YO
zNFv{;4$?z<`!M>M^>B6UFifc)if6|9ux}m-9VLBC&eBIqu0GT+=;KGFK4vHxV5*q`
zHaT&3wch}{ju^n0d}GO}zHfk~Vf>m#nvn5RgjHQ%{Jt)YF)yTX=#&gP?PM`*Mo;wa
z=#920ITY5*LHmz9?2ap9Ubrg0t#^UjZnCha2kxx&#4;&wjJdxaWnRpGk_d*)%wUY!
z8w`adA-tmtgTt<HG`599@ecLb><sK3k%{4_m=V!61KsCxSJa#L)XZVXbWDOCb&LyI
z$-LE1M&Gt%Y^3h<hI-6*JMN%tQ_*(`b0l1-)p)1k(?)i*_;V*kzBxi3=t;g&o}I$`
z>_lwpmjt8gZ7^RKghvKJ2)h%EgV=?D{XrO97=+5J^xNMILfxGpoc$FF{mxMA;Z(^)
zU=Z16632~vqcb%Ef0e_roourzJQOMxAqXembREOKHl+}B_6b3bYzXfVLh!P82oB1J
zV3I-ztapWA(EM=NB+*}Nl?)y8WE>xtjIasG=w+77yg=$T>dXjjOvIhSM11X%3}3lq
z`b?7{ML+VRDXF+d&rlb7j%LcIW9t%j+303rh+P&eTC?z;nL0y<W#i7YY}{YN%&t!d
z@Tki{w5l`P_}4)sPUhWx<smc{(JwIM7_~I=&7`xqzyB<{+&+u{;?E(W{v09<&ZBn9
zc??~49yg^f;pEsW?9h7%wSZ#w=RU{iMWyIFR6;CfxAwsTS+S#wn)q;CO;lIXJKU)u
zK1pZ_+dkY&P0<vtB^ttNhK8uEQ5Tk;>S7mvt<gPI@!pC1Ee%z%_?L<}ZK)y}wN%8Q
z5@qo<MOmDruh<{TqFzc_T)Cqtf(jMI_@0WQJf*KN)bA_KxhRNe83l2hom2WXa>96{
zoH+8mw<wJ3Eu?q%5*Ji^38P~@g|%@{ad$BDnSV%&82YB&&q@pL)iUB;w2ZiFD<jU|
zkrwamrNt;niybo3Vw-FavDrmRydmHC43QF76{UoGS1D0wBq`iqNr-vJ*%eOza0Yiz
zWz!|ZxRJc0BHx7E?<%y{|Ha%NoydC8NsrBM9Ci7PHQ}8YAKD40Wu16Exf2niI-%CT
z6RzifA*25<{M7h`a|1hYD2e&RVe}JkYll`qJE9i1qi_cGoEqMJXSL!>U@NrRzrlL*
zSJc~n#p@xhD3xr*FzPq@Pg~IRu!Y_T-b-C*f$_c;eBRxHenXm}rrC@|6TTq-S`&_q
ze2<ak@7XWigwq>p5q_~6T~1dcb{spkH$O+B@iXKceu_t5OV~qP0wt>w?3XLSrN9!T
z2Jq4PDPCSEg~6;cY|AajU$+-<{a1l}<4U+iRYG#j8_XYHg#wQ%_D)nYzn~g{npLP6
zTZN?HD*UOf!fB^k40>1(*F^f2A2cAFjMH?n0ZM!6NuJffjNN)<Jg&!^1@*YhJ(Ta?
zTEth?;(TEZLg@W@Qe6$rDX($sS0#>Ytb|Qv1$J9jK>5Zi>{WV+iDzG6jZQh-6Uz`&
zTFR{U=j`KrhH~YnD4A3Osgfr+^zRY8KNcfeyBM3sGymqweW;Y)L*FrXG5at5#dmK(
z{JVigM%VEx<tloNyn?U#MaW*q{!Ez@nC_DgceewuTA70p8?&%wYX&|^vulC>CZ*$3
z@t6Fv<x4UusMnOl(#N-nJ1E^`9Q~Svh@3=}4@^XbZXy;}Cm~xoiCxKwn84iOo4eSL
zt+pRF`x6lNmEPfA@o?Q3hv?2&Xl{?iq$+kD=*A%WO*B?bjz(XFXbkNWjTvUqIOrdZ
zFDIgr^fDR+7h~|GF$U-pi+0Ia7~P9O)s`5%=of=#<rr-39RrnL(OB3KO`R?pX&KQ_
z9v_W$3xc7a8HCkA%(ig}!u20J(J^NmQfF+2+Cx8#`nwSqo^C{xj4#wD`9Xeo02GG?
zpna$xzV+CM2aX%iVCaLUU(6v6U~X>*pZ}M6Blw;dc6fW?M5!klCVAjMmOGBuxWPNt
z4V7KoQ2dzM%?<i&vRq)Z-Wgtlm?cxod!(mMko@6<jyq0JmvY7pJ!dSbb%7jjmefnB
zouoP8^Ib>y<~ZWRkTsC_x*Bs`R^$HF)y#^gCnnPY#n!8Ec<(A)XkLY-G7e~3xEi{D
zR^jNP<yd)e1#~~I#No)*=yk{evNP8pcfKR;t#E|N2uF15Si@{#SExn1Vqbt8j#%=0
zR(Hd3?!auPuEpsHM_8R&21nJ!FbP|XBL(*8$t<{8g?4cETf_jD1#o>aA0<-r;Ycry
zwS+CUzMI3`#yNQXZ#LY9%t3eZiYNJI)oprsIrBJL<e3I?O=6h^PH>9(_Qy){2^mLy
z$wb&}o``Yu*-VjO|K!by_|81P=^V`q6XC;=K0XmkbLkac!K@o<L(h*`z-R6M@lAfb
z1vC?=(QrIdEZFHmy~u{SdS<}Nu7EB1rVsh%^uO^qb!I%aN{_?lr{;)_Fh|Bhb8H`A
zj>{j-@Z`1``WeuJ)7=bG-ACi|VpC)dH^qU=Cg`m<3emrfxeGGJrQO_5k#WARFvcI=
zZ09+1rW;}Yoe@Ya7=hb~Bd~kN2xe=Jz>l}XQFL36`%pcEtkJ`-sd{*<uLtvrVQ}EB
zcGde~?CBf=tKmc0zogIqUn9{JrBBURAE!?0;~l-oMh`|{LeU5~C5&K>AbBR+00v|m
zvttIBe#ZcNJ@n9Xh$a?t_f++)FHROnqvD=4_aHI|TOy12wY{-pkQ`12%V9vJ91?Wp
zQD{jG$5WNJkxobra)+6g2S$W?pkF^Pj9KK3A@1wZ(!7f~&%v<V$(|3zP(1EOuDKEp
z3$+MzzKg)?nX&Y@Qd3<OhnuJPnq?0Cx3{=ozM6<|@=bCN`fTlzP#?S>Va>_7Gdcy|
z_*!V_<W%U)XBXW{cC~PS6Xr%w@rG2WQ^Wbm(Iw-Y=*c^UzKPiQJrNB>+n7nYmA!X6
zF{W=2iVb(cDkcb#7lJUBldy0XPH?1kg4xF!g2v<!tVm&|H%G=R43>l0^G3EQn-YNr
z@{K?FW-j@rQiNc+SqS`;Ll8s0QI-ioEcr(DNHE4f-;S!w+i@d!J3d-($M;Fw@nhv)
zxH|5|-L$><ShN>`cKcC4@3gE&GA{m1WLIk<7O^+a^hF|ba+2^yJ_%o6Bq6~y1;b8~
z<J7sQQqQ0#F9SU^vtU%3g(F{iM<tcboZ@UOolE`8nK$HL4#4r>0rXNih$7>I*g1ta
z;}wUnu<R(Z<MQ!X{VaC5p2hscvpDeREbI%;0YA<`hJ165d=tL>JT#VFflT2QC?!9{
z=c^B46JCtf8l~8~rxau8H%YdU5|78Si#t|DR3BCmw_|wM-9=N_OKFO`EgB+duZFlb
zNkgovQWpcpQoH%9DikiMitp?R-%QWV&dDmmKwm|ikWmp+9w`f@7-cbZg|a9%Ru=kU
zN@B!4MKSAwqL?nPD6T*5D^kYv6}6WX#N-csM8qCBk?kobJegH&ugg5Y(B7geyEOfn
zRXi-Or`TfDQ`Ge0j_RYdxJ{Oc%#apiXG#n6t<oZcc|YbKq(y_Zv^Y@QLtGjxEu18!
zh1HAh;$mKRVQIo!DrNeN$v5&ZCB@brlH$}!2_d^(Lj2@DN^!D;Sl8NBWX|}9`u4xv
z8~(<(_D;Mj?Sw^HCn^^G#=&8~Q6TploBMWRtz0K;WID-<ooHY23wbl?HP-$CH{OfS
zKGhCAGS0!>?U)kW4wdc9EpBCJ=E;vZ_wPN_KEH?SxA&Ma;w$cIf5pskU(qu9D?)U>
z;gsSx_;GaAzVT-C8#3O0!?m8x2-Ijs@b}NiXlz36`X&@AGeh|DdtAK8F3VT%psZYt
zUl*&8`0EAqU7us}zo&3p#r!+w!|1Lr!FKHuOyfT1_5kj6R+nJTqY{kYREm4QOVLZe
z9HR%7!^rCeTIW_^(%(wtYQ8~IpDL72szSU^HSbNUVchlxUTRg4aH>LEQ58OuZz@vi
zaB5=%cAam)IqswyGkGuN*#H-V2Hqdn<Dm_^7U;7n^{9i&wOR~(S%YVLYGA8ZgMO*i
zxZqX=BhA;iFtHMAcU52peZ~3LUZGm@C4C6xi0oC){)I9$3@$}C>*rAQeui;1WVL=J
zh{}AzyX(goSXPYjUml{o+e4&1ypO4e?m@@qE-cA6XQ<z(EVv2pQP*)G<|-EU=HGv_
zB4*c}!q?l!arbW?^`8T{I4%bo4>Dm9kO6nabY%8T!&KgB7M@Fi`d!`)_^{8wmcBY_
zvein|V+SW;sZSzKG9TmJ{ryNvWdCK`et0+R$A#Gacs7dO;;#FkwvwLWeet;aGY$$~
z)NbCzLVi>%UdYAZ-`8kZKZ%BiX*4Q(L}RN=G_?74$Hr)^DTv14vS`$vj)7D)e{GLg
z9R3pn={qsd+7yEc+A(k?-?U4{;PTICG_`a0-VzP{kZ2^k1taBt5ZdT3z7P_G&D6I0
zcMU?2<u)8L^@mxgFJ^`?W6ypAoQyV*cbIM0!w;7={BXR=8yf<>p{(hRFN1tA{}cO|
zeAYwvi#M9)dn4|s7Y^8YVe$n}_{w;)E6oE(d3)`X>V_LnT%mi&6|K~6CJ%AJM{Or~
z{#%Rx&N(5;(h1g6oiHQQiTQ(07$M;d$6HPqW#a_3qiX@#wRmvL5%T*TvBH44dZ$<8
znDuHDUs(kc{Z)8ex)LW9n6)>XnR;7RVcv(8FryFnWc(8Rsb9iB=L(E{$&Y!jMpB#u
zUQAen-(wtceg-=hYS-X_%NpEDaKMjMt|+WnhihxsVI6lAMkZ@<Mb;7LhOflTsAX6k
zYmY`5dqniGhlivca#!0ST7D67f6qsf#C+VGI1ie^wy4-W7n=^wfl;qH=y*CCYOiL)
zu!kl7-m<_-a?NtiOwJlk?K2CUDz(5ZPG*?}I`3LQ+ioJ(m`%iw!|Y2U!?d29$PVU-
z(B(|zEaqH0GZ7p3xpxaD;uyWcGu~L>$RmO4VFIPE0LqiNbrvw>%iHIPd|VFZIRbq(
zfmrfQI&Z(*&r`#pk9gCG@n~K#4w|LrSP*WG)pq7+Rx-!v6f-16nBnwrGc?(crT4@X
z`U_35f_&rL)s!6*CQz?3#<l;))Op9{+=qSNY7vD**<4bTm5}uQWzTG>>=2c#WTm~d
zOG9Lj>`g{wl_Cud(U2{BlQN#;zMtoh=a28}Jj=Lr(RseV$MG5a`><qZZ_#e_MzN8e
zsOP;zyJdpMBV*}(H5MI}G3yeHY3gH4>Oy1brSDFQtZqz_`ORM88;b$ma53)2yWWQM
zi!;Edp#hmXt?7EO3m=lYa_@=>3m%%#F1!a3!Z_Yx#^m)kMqAgIrFq@iT-aUuXvIgq
zNxW8jdorMt5iuv)ijKK8iz1p6f3vyx!8CB6qrqLjhRmPd1gG;&m?eBOz*mdeg<9-y
zqfPUo=FG5d#rLs`$S8J|&g`YKgI&tob4w+cVL6V%@E<=0uwq;wTQ&qTttLpiu0t@s
z7Q)#M()GF^9EWS+w0|!;IN2|o?~i5slUV+9iepp5c#;C+CBq_t^v0X{?zfrJ_gksn
zp2TglZTuU(jS!jPj2FIXFT1MwF5B>w-17@E!#N^+)9J3<p@ePJWXCn~t``9o9$fG5
z!Ch5vG{5=Lqep;bkNJ}sCyaC6pAuo4l8%AQO$=gCTp-Tff;cayTX&hej19s@{9ny>
z1~bxNBSVIU;@VsoP59;)LD&i3*a+V=RSn{6{UG@_6o_My58cwF4|jq$8!UYoGuVgQ
z!Z*3XH>c-E&@d@NIxHeMX)GPM;<ai#F-kmOl7o05hD!BV=~zqTXT2?S7QUH<%ylLu
z;VAQ{HV?Luvuy{5>h6-x*WDBf-z<2)o5`K_FkbkkuJBFkC3{%-Za+tV@8?DHWYSF}
z&w1eiT)Q9S+nQq>N;yk+%X92qa*lYJ-5l~wp?F^kZj~vN{db-fW6xvmn8w}XY4~KP
zkt|uS_0qEWQk#v9a+fo$^3jm@x#zPV;M-pMC+9R!TCHoKluMVv7)=$$et8SUU-n9q
zPU<UR|MW$pqp!U8*HiTWwNefr7seUaO6hk>I>LvvRA#knsf-nWR+rnlN@l38vPSa!
zR%z-gn|xa+@5K||YqgHju(^(MCaSqo*rB;HrLd`TPC6Kttk+T=yJ#s_XKE>W;+>k{
z(L`~c-B?kz(o||B=TAkgp<?%2L$Rum`OSF^Wp1d3(!IIt)vl>4V}7eE0e<Srz5yCa
zt6<R<SF0%(tyGncJ5-e6167nwhAK+W8}${-!}XOf(!pR<TTf}KD!rM@>nN(0b(B!`
zI?9#ge*{?n!)xPj4151Z`*W@2P}Q=0e=T2T$Yb|fn*6QdsAz90y3}x}{x6!HtEOam
zHC;@lo5B4jJtg<Xt=kWhwti=u@XbQuo2@$EiIg2uuNPHpxmCpzIr@cF1V8)0=$0Q@
zD1Ip0Rv$TI^pQ!zJWZax<+krjN<O?~+oo4c+W(5auU|9suXI~Zcuj}sDhz91vGq_D
zc{8dg`uB=D&ng(&<dO6Sl(PL}A$uZ)ZDh`2dn})mQ}f9>n#Y$Dd89nd<JPczdGF5W
zX^$ewB`)G!=VE4t7xVq*ec~KsSM|J%y&uY9Zv|bSR8V}Zg3$O1imp|l_N9XT!aUhK
z9?@vS6Q0WcX`;P!LL7LG-M;5+2!GC@Nzc)~B6^$k;{E>glwqr%GDdb&_dT9)ch_U4
zE)m^L{l{!tR7vM)kHk-3j@pk>Y_&@%9P$v|!4K%~T!Q1e`_#-VW>snt6Dte3VO2;{
zKmo}a`ON#7$BUJD%-NeOJG#3h-pr<fN;cV7Z=-c03kUB^T>IT3;#xYFRBv*+Q!3~6
zFEO^x1(`ve<I9UvI6I$UOYQ-(mHn(V-b2`)UEEr{li7~jx&CJxCFhdpwlIl~;qn?K
zb6M9W;uUDSl@Xn`uxVN%*Tg5)_U&f*x;OK8&1UhvZ^q?B0x|0n&=@Fw%Bpyb9pf?c
zi$mviEJs8O?CrjZIn863ay^<p^P<_`I+~V^#G~*hie=BD&~uNX{6`eaW25QYEe3;q
z@^c@?pk8+q8Lwg_k0?eu<YU-4CWcwUH~my%ST33y!%UgqoQkGUc3c7T0y*DM`m+!E
zQxziGVwrF6P!m1QAa4x%uHlE%YKqpaA|TyOGI!lL(c2A;w~}q9zmldUD>%`51&_tM
zeR=gV>B4kng`O)9Hi#~_=@K0BmPjY)ayA@YMsR~=B;_n6^V$+7#Vz5~dKXN5UD&@z
zbi~u0X!zGrd{vIjj}*VP$|4?0+E?BFi(tC&&H#C?vxrO29Z5Xo$gV#Nd2nqZV<HyP
z+DvrB8ytvg<AB@tdF<Dp$E&P4X!n}KaSQ2>tZPqVo*f3|b|ibvp!&=Vh8xeq$bYWn
zB0BJ2wajob=3}vO0m?O*1sz?0Z{G!!$WCi04%~S<k8Q$ADfu$<`8tOU`g7SFFdGl|
zSy*qdlT0XE@`|R54}Us&EvED7whd?BOd+rH6ju68=Ig^r;-Q*^x6MR?3ny?nVFK0>
z6L9}7J1F6q*TOiX<Z-l||KS_Yd|{kI*-PEC;-?(tv=zsVhf)?Zlov;aa_iVo*&Pn$
z<e8y-5vDN_wlSAeM^1+DO-%Aobm|O6MYKEp&MUm|Q0QC=D}`^=gm2y*0x#j4I>I-1
zyTFFl5cbtlIvFiRUvEjrbHX@32Qh5GAa=DG#IX_!8j7}f=#v4smJQ%~@c=AK`{6gM
zAK$9Xct6>UMd^KI24p6A>V4(=v^V#T^rBP0-qdT*n<nMGa9LqOf~pCnmBze~`Ayg{
zW8TCHi+CAxV2Uw4I?KnNqAL#S#;bYV_;AD!@5zQF_cCO(rXkDjb;V|K7qoBd)8KGt
ziUQm6+sTAqyG_`7&6sD?jIlK}W|p=wMU%U8>O?ouiWo^QyP@<*8KU0Ah;PT+awEGn
zm4VH9JF+<%JvEqiUY$e7G^uCPgrUMWV>)W_&P$6n_q14Sr;Vp$bG{8}MgC_;#uiBa
zQGLlT9=23+HkQ)!`!bRTOW#JkKb@37t_lZ?sgn2qy&-tyhtO3!98+Imo4|0?+C_+#
zJeE}BSh|+QVmBj>f09deQT&lJM2GV!H9>M15+#3hD|_2WFWn&F8Szi~*l%OroNYXr
zzm4n8+dy_y1BGv1$!ko9rgB$$oxs@;VVj;FWWQO@!`)u|edNP4xt~--`O{JO#!SxE
zlm2`XzPT-Y^CUKqDx)CLI|p%0P7C21Q?npS#{}UkjMGFkM0M+kCP(<jMEK^RWe^9d
z18LnLh)_8_j|GaK!-u)4KJ?f3CTD;TeT8q#g>R+{-^7gdk-W7Cd~6~psEFXv$q3Qd
zM-XNq{wpHcViigFwn%a_M8h1tnNAls)A`gEY}!d4*e>z6ci2wx&h2bkxPuM*cCjsT
zHy-bHOV;5Y7Fg^dapE4yRF@v$SNqvey`QU1lbP2cnYuUkVzPQaH}@Q5<+2oBtw`bD
zsubzKNWnfJg`CI~9^FpisoHriOgJxj_vf)%pN3X(ns_*`QvLfTpIp<SB%MEvbNMwq
zpAVbvlicV5m2Ds3-QxihOCNB@N4g<JYvRAJmNiFer5~w*vc+3fx!X-wIWSN%`+_Cg
zrfEwhM&><zg>NovwN!H6=_)xlb(JH-b(Oc~;s^iNLeYqBp^W*dqi71>tnHwqjOyB4
z`BAQ={EpUAY<#tpIjgl4_eEODlc`$DzWQ3qfVY~;Z8uG2O<Y4IwM9c^;#UpjT7ic0
z>5PVA5T>E%PSjAE%8aJ?w7PP0g1TaHK~3rVK~>R3Re4{is%#snqC_2)Y`=DOm1l`{
zl<L8Clq!vX%)aoKFvl7){eSVp<`2`l{K4$oZ_;o6CNKIoy=VT$dcbcg-_>Grx0af{
zwbYqV%TwVS>t8k8eEy4NA7%g4u$tK_(idL08r9^F=;?h#_sa+Vjrqar20uiT_#LNS
z-$`rnhOJ{>(=oS-qc^LVEB)O`@_cyt2U45J>_$$B?nj=q`-pwQOa7Qi4`s(^G(G)<
zt-HzyzfjJeVUJ`E`j{TSpE6l|*{9oAv3YqF?QE(TJfniFj0(2uR-o@%iu#2@{*1hb
zYi>SQHs(`2IG+I?c~l4wG})5JmDhQA&&X$!{ylVDi)fQqgx-K+<{Z9H#KZ@f=apf-
ztDGS}%c<;C!S-zxEb^%!#;=0pn-wTs9*KTN_EXEA5YzG*$A>*<@wVqY+#z$E+0VHv
zd#UKHl2JVF8FR{>60$)2-@l(=GvW!slOJ<^X(iKcK9cXZM|h5w-)~MidFM*G{o*03
zH6Jqdc8TOd-$%Eon7}^87z7p3>R}-V2Nv?%w1Apf(x+LPhoiNet+|r>e3vb!vI+fj
zhhh1*xi}_EX5E?it-poeW#OCXR4Uq}a@X?$I~ShEY{?mH>!0RL_;DWDA7)40{ltIT
z&03FLyd5rm;4)8LD%xR}JxN$jOJb4u1OonU!Q;vn%2OpX&nc0WR*BN9n8?u#@k*@U
z%$rf0Wj~Za+CJ%nTAskPz6msa8PD;OIP6x$5gHQ9>lxC;E%O$~E726~ji%@5XcBay
zrH3qvUwKhXof}2vv?w-@j*+=o3_JJ7aN=YPzpuq`=((_r?5WP(j=^?m4E?9Z;Aave
znL5#=iN~R_@J-z_(fnE-NQ`v=hFSs4f8tNjHGgUs$*x`zFV#;!3Y$tk>nnFwxx2G+
ziJN5lOTV|V8#ymmg5FBWr<55@bGgH&E@zPAGVD*e@=V87I`t$srrr{M9&*7z&X$iZ
zq^CRM`qY`6zAi`_5VK~x@MWwsCzG7$V=X?ZC&D+|9XU7Ek=!arDr+6N-d*-pLl&Xg
zdJ#+N$*!v0k^1`_nf^g$EtiCCf)_Gv@IvvcIH3D=9xF`dQm@%u4xgOEA)A>b)VIgu
zgdGb5?06kyCtgB(E-acs+W6TFy(6CP*}_F99q=#|ZSscs95}c@@*x*+<llUn-E?4i
zTL;O$m;BnbbH(2`hpZlRxO#6k?dQzq_`+Fyva=IUqb(|@r<1KUozAapFuE>&@C%b^
ztV|~9#w6aFP2xf9M8aw(&@6m{<ZDmB#A*UbUxj0YZOU)TUP`!Tn4DJGRutx0aYz{F
ze;zw#S@G(GJmwAI(W0R|`hR>gNcg5o_$EvE=7pTD!ZyF<(M|ZqO8Dke%@BHAl%577
zD?-OB(k~8Cli)))*e>gkKXNX5A#+9~iv<4G2hXdP*j=*Z!cj}{U0c#Kz!Ggg(Hm=7
zivMZ=DvN|^9Q#Qow;36t7e4*757WgrRoU8%Z|D1>OzO*|dVR52V~RsUZ>AN<UPyGn
zu`aSdvNGYRiU|*17?YoAEc|24)fi*iZZKxycw=^jb;r`MJD-xeF|Kttyn_t6Dtz-$
z_{LiJ#<-*_BMc3s->x+mt@YXG-jToBZKXfdnB*X1VkgP(*~6IY;=Q`qxvgYVwI%UW
z8-`vsB!0IcKKl(RJl2-k*IHxY)trUFGOszU&iIY$w4bC-mA^V4vNdo!+>odxjZy!i
z#ed;V<u$K4H+#0CYwQ9#ix<!0rz<;*CBxWZDUE}dNhXiv7IgKa!_YuF3<%<7a0p`$
zhfp^sgq5{nG!(vRDSWd-J%Te6B1CVyiQ#8r37i~9O}?D<lDU~InK|O!E?+0#yAd&>
zkBcCvDgyIHk*pWKv70UP8i#FUEZRowGVz3~CKBbk8G|$NR9Sfu(Zz$lFV+*eQ=V7&
zu>POl|MrsXi1w$YuuX3{jkfxuF+G4x;hTrTH^<~05VkoiM@RT(t?<ndnc=J!zL|C;
zm<w7VxSS0_Wnd7G-^*T0&6l=yd})5tM{;L;7_!-i{KnqgZRyQ7;hQgme7G!p^JlmZ
zfx<TzR?44aT!i?wB5)6jppEd2oA8aQ@XaaVo2Q#2C4Vv!KWoVZyb#5)HPPhAt}bEe
zHhOj24iVc)xU-$830pZTv+(b)lepDbv`2=!Fc!WkRQ9qhbT2MNdr23K@%{dyJ8r#?
zd#4ZaaOOF@9-brQ`8oP3DU3}>;p){Cj=oL7ztwpf4L{HE&sRx|J1_Z*7g#a)g4_Wv
zuxkG$GE3!7yEc_1i(Ar@k&9(Q9+4FV?EX|B83F~Qs}$1iN-3t1l}wGRl>UrLdC#xp
zQ`$GW%=p7Y$NI`#H#KGI6<uYJv#wG;sHO6~UQ4C<BVDEIdR?W)Qde>L*+NMSZJ``l
z*g`4OZK}L_qpiHCtF4TcZmA!e#3$vcr8tV_=9uiJzBy<qUcxu+|20wSJ=9b#%+*vT
zhc;BYYl;6`dNU0&G!(C74MlUdhH_DMQ2MXc75`*)<${&EvMozZdGSJ3$#hp$^7K`d
zX1g0GSANx325zdSY_zVUw9&4k47vG_M%(|gf5u;YkJYk%!7o;u|H7r~FMikiMauVT
z2Ip5()Bm?<(?mm@Rm<k2TFKd~#ZCAou3imBqOnPSFWTXq-+9Y-_SAgk)mYgz3EOmu
z|A2Y34_vzTR(fmRa_r6<;q*7GYb0|QVH?K-RopsS#X4b}pv)?!KKvj#<sT%k@)aFq
zUem?;B~=;ExUKS(fnCb@*;DdqYD)1MDVpDn<vd?o!Q8bExRU#TVKd5jqf(A~MLAOp
zDmXQ|l=}uU2NB))&f0ulhv$>FAfIQQ^XWQLvSOy^;p~;i?i+c`=$21qX})A^7UAbn
zM9lRfoYn91_-rw1SMC$T1G>MIOJcJM$&0Muqhkf({VM1dRY6$2M@;s6gu}l_T>JQh
zwnv_!CAq{}vVW@c^9j0-MN1s~l+Q-bWLEW*-V>g3IqM0UzaI0i_hU3iSIT?ZBc6Y)
z;ESm+nP`kxB$twt_mIPT9&j+V1oa@vy}47&Aj4uVdKOWbU&yZdh3rqe$Ja6WxMbv!
zW0=ROF1b96yUVKNY}RMqVbhD-oYj?cJd?7)8JI?=^CmP^dMqyT?cqtZN=~AkdXhlz
zlceoB$vxx4OiSHI<NVz$9kGj}YCEM*V;ilCk~k8bM3CK9`si#$=c@3{4C#?d6fIrd
zL{yCBvG-<L?cL15F`N0NvswO|3D^%vz~eu8yc5s)$av~Misj?@SRRetME@Z%GG~t#
zy;C%|{G)L<k7h(#6gy>Z!__DXrbaPPFN#aTH&29b0)%f`3g1|z#W3t?40?qz%*==(
zSNNuzu#&sEa8JD$Hp!l=!)=)b?F-<0bO1G?Eq<jPK-5!zrilJ9+R>j6M*j5kUd{G$
zcXZ|m-z;;d$#m&Vc(IZoxyP=a=}Oq!CD?f_AzRobKG20><6Rij#0Arr&eBKeB3)=M
z1bDkhmZCG-XPjwM&zUN7XV%&{OJ<%kX5r4XGjo=lN+;qi#5Xm<k-ILAbn5DeU!4Wq
zo$H8sheh--UBvCWiwOAYh(VDfZS@_AeZG*QGYd%zT!_ukh5YW~K#=ZSKDVAhUX2~8
z5A4LxYR|Yr*&9XKF=l}sZRg8lV|$`&?8UD#o3Ur-5;(_!j~5;IP~;%<E(eCZc3_6n
ze3~qp&!I#Idew2jvj03XoaVACZVqRf&f)Os*;IX<#jc67c#&&M?X>A!j<>=1i48qV
zY#4jVh7B7h5n?}ypDL4BoiLG2I&u;xFl68aYU@wn>BsR5e`>{m8&-U{ZpCLgt!`Oi
zD|}-lT+?07(OfI~3-c5n6fe~lE71@Pk@vWv7|8slrLawzoYlfRhYkwg$gz`?C?{{v
zP&${0Pdrt65nh4e1H?xp*)qey?Xv~R*DTNsw_xuK$=2&)fz`@^wEbz$<g=1zqbBF%
z0MYRcz~n=J=G+p0lzKn*T{mO=RWshEnNb~N#&A!W+r;<g__01bK4r?2BYm+n=*!?Q
zrj!gcB}LfB<414V3HD&BXo2-cn@G2=3Gak&3`HliTli*1v@z$F8q2@i?yT|XPPg6N
z$O-7iblFc;4DQAlAJN)48sgm2kk+5OqUB}4sIo51?5$5sjvkt$+Vk*IYncHW(|?>X
zrG~~l_}!h}I&H<z+!o7<Hk^Ckh90SH#fKrxv#+gmR<*`^b#sQSZ$j=abqZ#wQ`}n}
zzqRV<UDA-8!iI#*ZOobzS}b4Ql<3#ZIMuclcQ3c2i_roi#MjsNg)<#DxMDp=`l&pY
zlOtK&`!rV3T4tQ*qe3J%K7>y9LZkpagj4F_;z<qXNRtSzjEtbIu*>P@v0OMUIS-m~
zY+EedaFX*-?HI#1zZm9Lhq3KRI6Ys7Q`aDphB?AG?V_2xD4I2g#8=RFv*ZkJ#(d}|
zI*jmSek%{WZ>;B5k{2z@eOUHKv^Sx$i*oWuYmGmNGQ+8}$)6$X{OK%wb4&Q9c%$s2
zg>M$h87D_ec2#;?1L-3Eub<}wd8`@2lrurJ?H9yW`S}O$1n{k%FShG_Xmr4fwQ|nC
z_98>m8y{V7el_snP(L5ZY!c2H=0mvf%?tT@?^%hj+$VxDF2XqO5nM}&pl(J4CIchc
z9v#W#ba}Lm;z)56@0N?lVfGeGh2z4yY?of;?Of>*E84qFR0WA&LU^!A$1ReXvJLyG
zyV#nvhcSDkTPAA{I~wg_Y202WTPNdGaESX(=dgTmPF^F<(X;j(9_>?jzA=T;(-%m(
zk}5N(G<Nj3NbCF8I56=h8ys)4YHT{O3(|4DlunZe>FhDREbMuk!!7cdFglMUzdWv9
z7yYJaHkSC^!(dV|M&c*FBEB4Z&jOtE3JLC8LCY22Iet^vVX2xjYN)RAVQvfYnd>UH
z_jHvT@w$rs09|FFbWPps)LgNg*HoE*LR)FPP+R&Gw3U<-TFS;?Ek#4-HVsy4Dbrm4
zulA;9td?@WrioIXtf};xsHrS#(MZ`-tD*dPtf9<~(@?rdcjnf7b>-C&b>+<lbtP_)
zx?<m1UHO`>ru;5bRf3&W6>ZT<58v59`K#4HSvbGG(qd#C<+f%W<>|$LBn#g(vHgp_
z&R>jQ)bg_1FFtDj;=J(9->=nFKdI)oUo|_v)iU!|Ei2+`IX<Em3)xAHS@wf9!Z)j)
zf2X7H&9bs@Xj*(EwUKCrKYpOc)ek&U|G=A~_mtX;hr7=^={I@JxdE@~*X1>N{#Dcs
zsUqWG6<se^2_sdJ-K~m_D_?Ob`K4&Rr0+ZT8O?u{awxTwMjuKgJFApezNPdxE9Cr*
zd%X3`lRl6<)~qXFQdN;KdkLnk9+IXdxiR{s?4MeQy=Fe0ZDbxNvzvBlx#;NT5!OEs
zo!NPekI%!WDi6KG`7Bs}kIJq^nD-Nnd2}%!cNF7(^giVSOGuqvioQ!3YgH=fIiZ4t
z!g3DSRFGL-!3*C<oUM`FlAq{)UR2Urc}!*0W9FniCU@LZW;s43-c<bHnok*0{sgOi
zPq=gEF}HLc)4E+HBZ43CrKEyqZH3n^ma*Hplx4f6k73^f<{c}+TXs|RZWJ@1LouF9
zi)a@loGrZle&;<Bt@4?BHjk!id9r)TrPal|k`0wj@{K!8t-4L$fGp1CWwO~IgNzO7
z(!G+3R_00C-#>}d^b@SOeVi_hj?;GLG432qrscl<w5`7fo#wkxj_x2>bbt0olJI!G
z6&u+j7rxlSYu_y_{VUmhjT4D|C7laPH*>FZ0(}dGWm;|K%cBIIOE%x9_pz+^jOFdk
zXue;HCi6rzB?qE$ofX4d(Z@t?jmBd^G&j3RX2p&unhuRZUCs)hC@d_ZNLP*GqhxT6
za*H6bf%IuRM^JcL_ES${(3d@x=FJ$<5ymilQxqddM^R-HPVR~jN~-1lrRhP`?)9kJ
zt;ed*dRiaw$9$*k>h1jLCH|ss!`-Ey-JOP?-Kep0!@$surr9g`{=k)iHm+P3AJlda
z$%MJ?!Z9Zo>h+R*K2`A+J$5Fg%$dr;E`0jyjLRcuZmK%d>AMrYeViFF!I@|A&YZ8~
z%+!5OOxJK?U+7|b9(5!mdI1eaEnw%P`Hb5kzNeoH`S58Wvt>V!`^pjfY)9goJ949Z
zAtpx`@-<)~c4KGYdd{BeB$>^;w`1lFJDHu?@lDvK{|-AY>WFXqtu1ehZKZ3$PBhi_
z9B4F~o=4`&F4sW@84ef~I<T|Cf!@M6?cO`^s?L1;Ry)w}{X7n~nI|~|b7|XX4i2Yg
zbIxuyFYeDm^SQ81Yw4EyFpUeN<Zu3}4N8^`(@)z-NApD1c}~P}{6sF-OrU$*1j)6X
z!1l-E`Eg>rWLb^pr0|Q1@Qv0r@k?Emp2{m$(#d7TF?rryc*jD{OZl<y!Z&KMR<v-n
z;--agPj@Q{T3a!5qKwdbNG_fk(#Ilauk4);gT!mIL8O#1SHQOd+DLBRk)zP6^B_Fm
zTQE)d#x=}>6?PURJ{!oy&I1|y%bbo2&8b!q@AriPY@t66nf)jc{cw=5&4_De*a+X`
z3EymxlP`?ZM;ND1Y;SJw>_czozRcI{OO}r*HsTRqIK7YLH1?s5Q7`H`_Q2q~2`;Tn
zX#U<9o7>`<lKs?_2xBVlcgK8pcRH?<ozl*3R14o!I&>rNziwQr(+!pTMr;z!=~>5+
ziNOYVRCZyWuJFxDJ+9idA#Oxlbo5%wUc{JNos6;g(Ve9#ZBc*IhR*BSpnkp$?Tgz;
zXMI~nEN{zsOMUbYHDj;6w#=^7`LDn5%`Y|S7gy)sR1Mx<X~-3CHRkQs!oaC1L2sM!
zvVJRipKV1&L^~FbZpY0gPAKDD$&Fveki%~Df8s%IwFleOH?U&64;A}D2q_Zn%AXL)
z91Z3Ckx<c|MG)6D0z>C8nb}E>-rG>>nuRg%V+fDTL-EK8qwBLUUVfHbnaAPmuL?)y
zPq<|4i9R<hiYEF|tj+gf(cAUZJY7!%gAF*ZThG$1>*?CigM<h#{+0PK`L`cdvR^J+
z>`%>V@lCB2zm%M=!Z!%tr2Y@zI0X_B5y)s^o1t=CHU)~8Es#?wfqZWmLiMR2zKAa8
zie?b!<mb;7zA=>7i;suAh}`2vcCr^Xr=@4NkvB){%jbn})YN=fW9h^DQNmai;k4=#
zE?wW@B%8`!PV_k6S4HshTm%*A5u}()7IHu&{Zk{ECa+)FQIdOXC;92JV}C3=yJ^ef
zs9zd`*~A#}RBaO9isW@kw#5<2)CoVa1v}U6%#7Z}N{iiu9^Wl_OM4`LU@wVV_ELUJ
zx=KZJv--|ChLxX_ZozZR9FfAt#0wnExkT=k%k1#HEM9^%PHJ6Y)ybPQ%eqOmPCAh$
z>GE^a*|8;^H8V4~Uw#?GNjXe%%i&o<4*8dJI5Q@X<!AGW`*)AX)&;a1P{9241>`3b
zaL%L<J7Ju5E5BoVy^dn+pr+i|rmghI)=>%@3FE}+Dz`;jygE&CiW@dp+RSaL_#M|)
z)aGj|pId1w_xFoWs-%fBIJ${4Y?_wR%tlKIo}#71jn-1iel$^92;21SsHyCfZm9z@
zyBS`pp(F)pC=suOY0}jd;dN#6Ds`pIOkElINKLu7P)(WMK(xeDRh7!tDvEb<1EofM
zQ-AI2D|?35QC>U#W1-eRdI{fLn*5hVT7OBr@dxMDzv*$NmX(&Zn2oKWw6GedvL9UR
z`;#|nKRGhK2J2BZtlcZQeM^7PRZY62-hY$M>reEK|A^`W@j$76#OTupKD7HlY<=m3
zDt^z@8SlszzOfd*sb~ILJkGDl_pHKrQWeqrt4P^VMZbV5I@NncK+a1d17EUi!*i4y
zmHbL5#pRHArT3Szs<M>JYf2<@y@1p8WF|8)kHyv5{Bh6X&EPEjUS@K)UKYR3-oaTu
zU$`Qd#kcZszn#sU!n=Imn2Y_pT)wr;W5M@a*%#%>T|~IRJfDZ+ugaS&U6n@)NeL;Y
zt7kD6%I|Z2atX&Cl}HZFLsl&>BPFj~`Yg-odZA3-FUnXfd}CJoh=HD!cs-MBoI$d8
z3Vlq?1@T4=f6CKgPl;~v6x~|M?aP%DC$pU`k8%1gEc8z1Kb933J}Jk~yqtwLWmwHB
zr9<dLlpQjc+Frt`E%!;eT#SZJF}2!7<hmEiEV%$P-+L5V<a7E^9xC!RU8>4q)b_hP
z4bA4)<~!&I+-7TF7KKkUaZt}-_MCLw=cTgb%1Qj^oS=T;alGyx=gA%EyRCDaZ7Roj
zGg)+PqjzKcUc6GXw)3FrHhQd0!YO?#<MOu9aIWl<Unf#5K7po><gWEB0WI4EEL9R1
zem7q7{}Tw`lfd+Gag^PS#o*T_RD^G)UW_K<U^HRL(Yz2})A5pMcFc&TU{o~MI!5s`
zB$9RQqKNW{<a<dZ0WG6gQ%CYe$3!qFG@Qn5L-AV`f<eeeW*GPqD)%|ltv)PN^<iwb
zH_u%?X_vQwozpjP`+)~Zojn*jc)fJ>tmWsgHMk5~gMR30w12v@_U9^8pRS_%)hbS}
zc4M8X8%y-vC>Gt#aE%qBjc_HjY6%H$ODI3<!e*J(jQ!7r;(yK@yzk7+LT9WyNZ*1y
z|E4Q5o3Bpnc;>{BF-{bDNtQ^N6Ni2;rqvzkIC!&|9IwUvx6P65!xk_q(}8DeW#8m8
zhsUdD<GN`UQ!dTqXz!WazBz+zn;8r|I0L6mGtgdb&%jW7$qKP&fs;L+!Zh6v+2J|P
zj^I(EjcH}a$9J~exM|CyL|gjwwx#ZTdy<8721L&#<<MNws5o$-K;}PX4s3qzfb$!9
z{@H=F_72RyJCAtRS?CNFo>@Fg^66)jtjuQZGVy}zOeb>CG#)um<8?P1$`fojaKMHU
z1=b|Zoj}5|36cXcfujx+nAUs(w{MK6p6__p*^HMiq45|C-&70VY!;rGl`3BC%T`<!
zzHyU}+X>tBma|-*4?HQEd@(Z1St5F43oG^t-z2ma57c<LWQ^z`VWIHNV&NNY;hQ$X
zH%`Jg*}^wZ@5?^wu4ry#ev{K`5P#oTuqsVDpXORnZeqdD#{==`IFR64bDB7rGxl<S
z7Do3+|7$;<b?b+!N<Rh)&kX<n_-37a{7m@9T=?eOrr!KX60g*PzARAd%dK^$)S8*%
zWz&Z#?fTHCb1&J?_Mq=q6LN2wFn7KQU3VH2u|hOBePo97xH}cwyK~d6J97q$NBmtk
zn$PS;Vef7n6256Bd=nsi^HIwX?TrS^f7XSC>U#Wf(Id>Z4W~P{rQ->GCeCV2v-ZZU
zmrTH=KiWV@8~PNqmhKTf*|q3X(NLdcId;$W$lu+JX~H-D*6IxJrzUeFHO5-0b7!6!
z-G8cbz(5V{0uAvwH09E2U3z?KNvDKX{OYVH`^k3v+SiVV?ygiDxY4N61H0EAk}>R!
z?Q?I8K6rEIix1~c2GHLyR9-tnnRp<SRiDFoY8#4eawv9JL)r68{MeNt1ilSH_hJ|m
za>JwpFpM)F!<hL<c3Q$WoqGF{+}>BbBEIz1^yT`{b<&G0`BJmjGR$#3#S!ae&bFS*
zUA;&azDX6n@g3t$o{nUXFYxEc63HPKwz(unTli**@D0K$mZJi>Bz&_yDv&GDfwUC9
z(G$LjITwiM$socv2H_=ptMkG)Il?#n<=1o-zR?!G8MDocv>je(@AIO`doK!vZ=MR@
zgvtEI`ad6zOD5<wwJ<hVhhy+1oS)qyxHUV1Xt{qmMGE8m7s-s_k#u*7q~LTU|H<oE
z*4IcLY>(pJ&M5KbL{Tapikx@REbA+JLX%kGrdZCG#&S&Vf;pz~99|?ItQkA#YqgX7
z1G~AEx|``u_ewU>Ud)c~WkA$Ey55mq)3S4-2R(<C@J-{DmnaIj%!@mhSzL6P%k8fa
zdo&IAPic7fy-Me`S20Yz%A`M6S+?~iyTxa6Au@v%DVel4$)RjR4!L`BP^rnGJTsqn
zfA7&k_~yBJ0S>}9|27uz`%nSjDvP++_d9kM>L^M|17(k7^5wPFR_rfoDLpfqD07M%
zD;^sgD@AXcC>@?ORVJU;RzhcLE4>?OE8AZ;Q3^{n74<w#W%+4M<?T*Q#U)Hr+2yUN
z{GFhwOdO!87;J8&cz@7PE{R`iqpyat@~iY_UKfAVK6T}v=!oN%$=s%?y7KI*nvyAe
z)2E)AQh$i5@+?I~3ECt2=pGG}#^%zg>0C!~ne~rdqQCLj`j<}Q{?bMLFAXmIVf25$
z>2tOguc5UJRDRLpa5V{)(wABIgA<xRne_4pI$1v?Yxf6JCBw#7{X31nf2QBW&#31}
zpZBznG*bP@Vf*)VH-1kegZGm0^&aE8@A=d99hQq;bEx@ivSlaLBe04|qpFzv>lJS^
zUXdU95=K6zw9{jLH+)Q=f0bOCUrKIBDb-s`slUAxOL_EjD&>Pxz=pq~omrEO`{69|
zS7b@oPA2K{UanS?iOG;GnJs70c-(E)*~%;<DjTh{cWL!8hY=CEl3|?7#o}DrY24wz
zsyh@m$)erzEH(|yX7Zdo={d<`<m6&Tl-wscy@d6TOK2SMkmkh?arstCi>qby^)I7)
zQZXIdi1%tv5&KGuq+d^%Xy$!R34^^<f5@#Ck4W>cMEmJuKFI#+zU-}(@F(<hk=f3x
z3jU3);6#TCR0_(8H7;kzs4`AZE~V?zha3-lAo*$~q@B4>pIgO*K9Si{b)oD*Wp?$t
zfH=K-jOr)ap#ynrs?23wc8=(m@8aQ`&49!^R0Q3|Nai=1m6;UEZmM0{4emOpvg-6n
zE(zc4%sY;!@Xasb8<k_nncqNizs>j4F=!VKXOd7oo<z#2Bn}Qu;$6?J{5m9=HfCEm
zb|#S*Dv6TQyjk?22^g9uaOPP&7xu<;EiazCjtPXmjgy^uEJcZu$tU_@KVchF*+q@n
z9!<wB;)iM(E%|2A<W7&`e5UkyuaCrYN+gz@BN=f#f`ma4lthNIwtpz=w#d%EZV2aM
zHu6JbBX5#|X}Uu+fA~;eUh^)6OFuz^CtZYZ(uHp(6?>3b*Mpyy(ldE-Ev;(SP&RN4
zT0yHh^4^`*Z&s0&vPyIWs~EQ24ecIom<!wd*J?R$>MoakiYvNLrK8Du3A^^mE~=9Y
zfxn%(@I>Y^>zrs>;KV;8XMQ|!!taw4%cXB&pxk?Fx;xSA-eLw`UM%+^;hL+9F*v)J
zK`SKlCeV@QHS?LW&4G~@XHzs`HfM@vv0>OOE=A4cP7~?nJ~V@Xfis9Jvgc`>J$Aw>
zCnwo6b+kP*tL#KOW5?rRc1$p|BSP3_<#$`jzLdGm7F#yIp3Zgm>2k-AzRSPUIqPW0
zfxUCtdS)(fWml!K(t(m#2d1Ss$otkjuC1BJn0vPL>SafosU6$C&tzv`$@2Sa&+rM;
z#g{vch_zFtCwwYv^(V;x-*H$R8YjGLE%)j17<`Z!&A0Jv$Q{qAP2+hvc|4VEgl(#=
zS^UbHin~_$%6#T)niab*NmjA&%?{z4>B2ON<eZhyFP6{!+iFGZ3x$TkmbiOaGF3D&
zD^;MM3EVS+MOG5`A$+r3_$KfG3=x(QStXk?K%)rOX2Z`s3+g<!U~`TI*_$nBKHGxc
zJuH~CY9M>t59C<2IoG9AD)MxH8kzSey`Uc{`l7d~-;ackW=uU|#=%JOM}?U&B*cs^
zUS`}_BALWjdUL9wsqBwTx$xVRVA0;p?r+MxJALq()su=BJ$NL1)9#ZA&90l^yTk;o
z|4d|eY(kNTF^?ULX|tm{+m?36OzBSEKhXqd8nMU72(uf8ltmhnZez&57KX$a8$gAw
z?AuyUu~c8YiL#pt(U&YQeX@?}Q<~YD*sX0y?$m~<9(u%N=#eBmQ}9=hpa%N?d*)T>
zVK`Ke^gm6ht!a#DtqLu>sG*^%COLho41KFgeHAryj%e`4vKiga>2SHCCC36=@kC9J
z9s~9Gw5%O-B<oLC_-5yG50Wdqxhj0)Cwya9<;|dKZ-#Y}EKCpK1=;r;=@^9mH__;p
zh}SwUl%n&Y%o-U=X;}#2FGA=beA8R_W<X&WOWw#F=Svv96klvQ`=aw{75n7-ae&Wi
zR_IC=L)97%JYI*o-+GqCNiLOv7XyTCP8Ir)8tzT;NN<9ddlUQE50@$ak|pZTG2xrK
z!Z+XLMBkR)S>YRFIXi`K5`}M!g>Q6D1oCA<py(upZw!K{Xc9!AL7@2O1DGmo)998L
z4TW!d`+9NJ%L^SJ>D68Dh5H&W>dx`P`nV^PXLw=RC5UnTgXvTgK<SkL7Jm)krF|gr
zk~28hG>8R0;=xJ};?~m$VnQPcd=p8Ztx*&U-y9ddF}e}O^ekbUC((4B5W|^nv3O<2
z(&C3?lZvOb&{Xb(#B;(|?vB+v*tdJPygqEl(OEoMy~HajKGA9Y#ZPfW`o8k^p<8;6
zf5THGv;Gnx{+B7uxy;DI%Q(Hd%w3Bss87B^y~S79=6Qv77p`#5{2EUuUK1bcP0svE
z=hiLh?2-4&VM}u4bvlR8Z#gV{kdIPFG&ypAzM)%4ntK7+kp+@>U4W0gzZqFb&WrF(
zuxkV5*MkO1!ucl3$RkapYpRLT^i*SImUm<6a}tkt^Crr`evOr(A2gMdcQqBOd`%@i
zTT@wbQ&ZV}TvPeDO;fR*r>W%GYAUhTno7HWjg+Yk8!FecHIxV91^3X@Q0!l*E2Zb9
zJ9CS=qP0+6SuK01QGbPLE{YF)l<03lRh8x)RF%)?RFu_YRTM4B_0!X+ue_C<;#$eN
znYy-)a%l5krVss#!M{K3ul`L%uirBJsHMT!T5^qkabbHk&zn|bGxaBX7ysbwyB}By
z-!zqe%pJlv<K4b+==Eo2wf`(yrBBS244cEF-gC9vdundJ<5jG5LWR9!SCGta0^Z@_
z{Z9H~swfSwLf5g1VxuaWeSF1&l9#->Dzlo2m3aS>4yf2i3^gp}=I4iO+)&EPC}E2D
zQsQDuNp&bCyGH?If97#{OtyG7vLy2>i-9GXSmb81|5GNuMp>9gWs%t7HjQS=k7s5P
zbvlE<#TnF@m%)%N8EAC9MRm(e{%jMDeV2*1Z5EUNW^rss4(6gg4k;`W4Z(dbf4VPs
z<_Dsqd5C7mQrw=D^2SIy9Ig~`p-vIL6mqk$5Ua&S;&m@#Ze}qqF8AqXFL`{|N{Ibk
z!twy&7`t+ePM1qJQaN@FD<~XLfu?AMQW{qfaHm|pzA`>PEak=9hh%kl$kOo-m?^)9
z$Nu|Fd{RurkYZZBD8yk(A#+O#c-Q?NcY2HGeZR2S;9TCl%fV>;T`JwPF^Io|`qbNK
z-_68!Vg{}UZ%}XW6}Bu-MdQFp9xge-p7+Nk`~NuA!Z+vrj?2I4qug)5Uv>k#xaX8a
zeg7o%#wKy9NfOs2$HF6M3u=wGu**A<OwkfM>u+X=PXbxZ66l#0&-}IVENBqV&cX5W
z+8W2XWw8`^$c{NpJW;|n*M)8T5~FFdDVl)|<hVu&$I9I%GLn;vBH2A660eq#Oy3c~
zUg4X88$(I%DS2EUH_~zAMjAZ`#z_1RnM$x|If4jt7GK0n;hPLEYBqaPzi0zRb{j}8
z@?dfc57bT9^GJ1_=(5%_N_uom`mg4LnLCN6R$;kUUR&3%!g`|2Zo0WqkhhXLEtg{`
zeDm#^t9a6uP{(r#b053#L-tZX-Z@iU<V@58C*E~%lD-iqyly$6_d>GKek{iG%VNy$
zFUB=%F&$GE^Ze3cCZ3k8Vqu#%ix!I>&4Hxf^BAo*hdn!GPGUM+G#@fIv7CiQz)YMP
z%%o)74CxA(A^Ht_!m-C!#~!zq_FNRr&C^?U__^EByR#kp(rh{Q(Uwz5ws>3HV)WUD
z8Z{dhxJ+Zlq^aDgn9QSzlW=mMNUwPlsp>TazZFwxD7^{}VWJn_Zp;0X(^=BYmSOU_
zlGN#xJK6Ag*)+*7nnu!+X|%AL$`$!F4x+DF(chZ?hK?h6{Wt>JTa#63jg~T=rc=i=
zlkwbYHXfZP)>xmj=6r%RRl(NMZ*E1<AWL3jDP02!0f`EcMwT32JP31}K@{(hj_n8w
zK9>#TWat2_Q_S#5>r3dFzWk}^D|aU_lDGWl^FbtFk_9C5nq<)o3*S5!Z+D&X7P4ow
z5WTMjX@@LGO1GePrv+Ef4ipW~K<4h3yuR%NMUQ08j5KrRuQBI=ZGW8m_Giq<el*hS
z$Ll)%xb{x=P^o5o-fPD72Ysj<-%B!hdhvKtZ+ai?Blk5^Ov+6O{b|aZZKm9v*oWpz
zdJtpVgB25c;3~75I!{a_SJ?!OIVNa#HsM2E6Le%x<#556GRgWgTG5>>3)xluHX`Vx
z5kEbQuqZU7hw#lF(HWol)s@M|b@=Yt0{54Cd~wvr%0-{{{`z=r(U*K<eSA`6ZxyP?
zCOtj0m+IkpPLFMm^)UaX$Ftvh_{n3JS9-Wu=`s9!Q`QLI49r&HPn;?Tzo}yVLKWAC
zs{B`~%Ee#}>1uDrq@~TNZJ@*cK02gGpMzmeOKPiI(Y}RbS$A<GU-(A*fj3FQH~WQe
zW(wcD5WZO}GnaMtzJ#av61CNjafUJv$q7aow-Jjop==fZwWaK_te%9>K=@{9P8cH#
z!gx~}CihFxC0qH@;l?T)t5(rs+bYHkbfa%wH!g{u|IHN#^2BR6D|aOw%-3Mv&71p*
zH&sKt**Myp47tzr62@u#%8w{Re>TfGuvEMQ!Z$yKbv$JrQ#&G%MI(i8ggu&yA1XrF
zW<Y2VZmomJt`?tF{Xl;03cyF$rd_5NO^<kyE?l!um}WwaCzkP^>=@|D&a4d_eZK*(
z5Knr36>UzZAntt$VwwCNEl)|-jxf#y%|Nzx2}J)#AnUxPqwBw5@svhLK4_%O6(d=i
z7)7D*&0gV~F}J0!Q+D(Yl6CQ7d<<oVv3$NKz3jp_U4?IQ+s3iOCXU{IiER9_gPJ9x
z(~&(_X-*`md66t=8pC_>QP}(JlRNi5y7--yUg8v5%)LZQmsI-XN|xD+%Pg93g|>EA
zIPV~QvmuSVtTZydr?FAxD%Xc!qm8F%b8e@L1|Wm0?=$e&a!WMGIh1DPkY1Kgj&lL~
zz7-H5x|@(ug=}A0$Vt~Cj;@pZ<6pmde7=s-)KR=j{?dU^-asjJYoZuwH&K3`ZLBnR
zZmhhU*jPDi(pV|%)>sL@ExO@bno6JRno7hKO(p7*rqblFrZPNHQ)x9zQ+m8L745N_
z%C!!f%8fb=mDc+-l$nFX6W%~WIaH#qgzi;WTDqz$!v@P7r@gu|MZDma$!f}fike~}
z+Ty`AD#}`!-#E@xQ4BhX&;4gTrQmQq<+)2e<)~!%?TwOM+JL{5E%-x?{MjSE*5dcL
z2HPt&yp67*^`>en#TS*j<R@OMe~>!l2cvs@XNmC5>M>vCKKF$QH$O|4(PtX%{e;H&
zj|>^{o_R*^$<lvMl<>_(;hVd{H#dcE?)bi=jxf$a;hT9Is?e%_MZ$rX(z95J{#H4{
zH%qSH=cw#tLcH$Fe)1t->d6czs8qUUWNsrm7`0sm__Z&<_E#R0_T1)&bVw%4jKcL`
zCT(tHGVn<z{~BlEF1atGHfM5u&MjI`yCuGe41Aksut_z8D*3x@kd%Q@?kzG$Wm0fO
zIKVPX`u?)$+$cw~G4n89dynk;MYuh>kIB*!hOd7}x89{JoN}Lc!^EfEP_l!c7to?t
zAy+kv*fyp}^z@RS^t>2XfB70FmN4o}2}}P;Uf|RR>=Qb-EtEgMY8k3;%IVms0-d@Q
zEcsW?&SPc7eJZ7Ma4Esr5BbyfAr__&<i1csy+oNIl@~Kq_$D-?kdGF!p9(I(bI(1=
zuaivX?Rn((%axAd9G)B9Wq6lt4mP<>^D~m|XOcl|uQcAZxkBQdR9?(a<^9%^e04oR
zy*|<ve(gBwyQPm|=5czIOSg&ad^;W5N^$2T=&+SpE4H#^=~j-*_uQ7YiG=Qy9E-FB
zk`KhAR-Qm^-2`-_<7qK0p2iL1S^r%6Fx}(0HA*@btYdMSxQTi)zxlOW^7fLV`6hel
z9bwV@_azG3`;mmm{MKukXbq>z_pzDu+%<|sYjXs#s$pCY43*5lP|Tlhq~F+$9KI`e
z#O=Y%9288@9r1wr3FF9JET_y%=4YM^ekdJGvb&EQ?!mi?^=#e09;Z3(cqr~TEL$nx
zt1G$KV-+UX+&D9T6~6sevEz{&?~L3SQ%~~yo-G$m=u+Mtbj4ftQ&FnY(NrM1;VNf#
zismMzjT5@#oOr1!nSGxYb66PX&ZEV=eY_YSVH?}s!ZFtub6#@#QvD^X*j}E`mi)dU
z4j3Mohoj`xT>Ld#wC=MdJ83opchAD^_e}n|&m`ro?4%rL5Rqw*mzzCr`r6Y`!(O!1
zb~L+W$E?+M#CNsB;i4^Pg-HsJPG|gx>HMs)VeJeX>D8V_gE_(@Lncdq&?H_YO~jvx
zq#c>SuHXsSU6~}gypx&WIfdhgrqcgE8xqIbFs6<T9W|%Xw%Zg&=1yXN#w6Oj7QeTF
z^k2#`5O0+0jd6T$JPsezaa=nyj?fX-wB2EiTdFm7r>*%FYfZDI))<bp=3{Sb7Iv}b
zoUSE9Z<w;W$dt-TQ?wqNGF=|s@=WQSBOkk(vck!fo(o0K(#KS^?V{c7ET@U|I|<{2
zJ?cx27PM~&DZ)3w!Z#-S25}<Vf{k+qFfDih;YS9bmotEZH^N7v9bWLujJoH{MElg2
zA$9w5P|i?KGue}xku$qLKGXV3R(?O0TlS;8bwB3(HKUz)z&i`y%-v(grL;a=8qy1!
zI?~ZSx;I`&`f#JF56PQN*?g}Te%pGo;Cv4}Tzil<y$3Gir9a`b3BkfQH=3GAUyv|P
zHxsV?HD;vfjO)sR?V9k-JlRogG7}As@Qv9PBTl#*@v5H@e^wh(WoAg*p*l1O)S=wH
z1$&D1u+z||jWABRhd!@k^cl2AUo`8ja5~qDH9uO(d`geDyYw)7p-1bFdK`Q&d#WFL
zqW#ljUu9F)FVrIDVSV%$HW1C8Dh{_*xE`)bYKAJFlT;;xMTPERDunu|u;G9TJ8!Ad
zUsH|hFY5fb+Z>l=EoeP%xoCaX@bA3`H!Hn4B7D>Gtv72f`*2X#m%UScnR(ooWy}1?
zzwb}Gt3k|;560O}x)St5=$jnE{Qc5vc{7ZdyJ4)q7e+r}9Mxh!v<Ca)ws{pfk5&=1
zWfk_4E%mr?1v#=y&A&ApzctQ87D)Ew?d1$u<<63|?xd`9XQA+ouJFyn!QNaO=uNVt
zH*u%EnNZ{>^DD{k6uxm4UCtSIe+J3<x5}SW8~wPn$d3;iemKqZ<?0<DZi!Dy&ra?|
z8lLpow}D}{8<-*|Lr(C_4UAs8frpzn5F>o^>aEO#H+W+H(~H46LFAhRk@h`+P`3bj
zJr9uA@IX>!7F2(j%(F%X(Y$LgZzhW7TK4QePet<PpX{q9M=^9?6r+DektO%Mcj?i%
z3ExcW8q0g{Si-)?a#r}}pGll(yb>AlbO#w`N!(8iXK;%Mvc^R)PZ+1cZF#+vEO*t&
zeHh%@Cw&iR`Q0Z4ld~5YA$;@h!DVcQT#-Fv8a0#B804J>p=n${pN3nJXmT1{rC#%^
z9QVJ5fp}@Hzo#=iF@vn@8Qd(s#gzFum|V(dPOWHB_ZINPuaH>><?jBv5a0F1?C<sm
z)n$K}EPT^^Y+Yqmn+D4MxCTnZkp{~5`6^1+Q)<cy)yB#X<HpLcj*XS{0!?L)%xh+?
z)l?MONj;Uh&2iBY@0HzDMvSIX-%eARKUMm|M+xh6)>Iy9YAR1$H55i_DCg>ED51C2
z6+P((Po5~+n?CBwQ$2O1o$$@xd1}gmzG_Oc_|^-4s3;$VZ(a;gQT#X7SIXW<Klt8y
zO1pXWl<u1Kl$CCEl@5LWvNZY+6aW2Y@0VJVifS;tT!Y7<8sfsm13vaAgD?GHkIN6)
zAAUztz8P^#`V{QGvfA_ui|>6##rYGdp`T=i^MRfXC3{bHQ^$mF*2(iB!Z#|x??@NE
z$rZk-@p?yv@J)^AZ(dov!nNu-*GenNTU*IV!xG|C@3SN7KL3{7XS?ft&Mtn)R;yCJ
z&n;!m9huvVEFjV~pGCEK^x1fuw{t{eayFB9F_~POm`MZkO!As$il;7<gFSCy-u)K6
ziZjURm%+Rb>AVb2Cq^fOSu)Ep2)xCc@3-g_k;%`FS*#d+TQsqEap{~#diOl0nBJ3)
z!vDk4S;tkiMq8T{6q{{e2caS;B6_T6*|dU98iWWIjfz+(pny_#cZ)3sC}|)njosMY
zqJHz<@BZ=r77BXyk<H%E8gtAsBoA=yP3ETFq-Jdi_fFoTt?ZY(I^SZLru171=a$zi
zWrTGpb)ripW3Y^yhsx<Ge3PGEfvZ|2J%n%cuU67ML-L4!SJ5o`4t<{8Wk}Us{AAWO
z%l-~+FW#oh-rLe0Sw)A&RXlH4$#kO%>B=c5;8qzf-O70BSAxllV%h~4V<H}#nC$Cx
zA6g{$e1$AdyN1=sYs`Ogg*&w_Gok1#uE#Q19(I;%_UCA}>>U4OXX^NhbgXwBp@;Cz
zk0*y|yY4W{h9Ba|!d+NA-NN0;(!J1f9Y<EL<Kg9Xl)jOiEl>Hm`?7{HvJds^Z5k8Y
z(q#7|jaGxxSWsUyn^tK|dX~y(>1SE&p31eIDfn8Z&`K|vGg-2;?X!wrGMB0)d^0O!
zC2=u{?BABiAg@H0h=+J(+eDJTCdl4O0s}Qg<Gvloqcd?l-5e)7L$P2N%j(lH;und*
zre_Q%^@VYk%Il5pGA@6KBE@Y!&!QtSIXsVUrt{dD7s~hUp)9N$%J6p~gx?O~!q6b5
zHwvPmpUk4RNM4^uFyB&xIMZFST2Ig7yUk3#8_r<$z5tqd1xQ|>KNEBPxpLE=)i?dP
ztSkNBqRR}p=10O^Kaw5@yA=DORpdu;rsNZgk9g2<KZ5^EWo5pc**+9VcFnXolWEd_
zGA)uOQu@i0X7fC8RP!Wnkq6BlxpU0kox$hbur+g&_h(l&O>$*k7gxTDZexDKg?#bc
zy!UY7xt`3NHjF3s&v=?F8;`Hycyv#VBYOTA@ivSWpNBK$`O^K=z?tgBP8gpY$=5a`
z<-KDBp0*B>9VRoT`3^KMaiE)tBjvLlnULkk;07aDbz=m*S367JqB9=~oSE^(nTe8Z
zv-gS<>B2Xe8TRzJV~^cX2M%XA5Tffyn=ZmPa#ETK>-=({-2(?sUUgv9MF*Mfh*#x`
zC9x3}w4ZN5lAOzOa^>`$XCd=Q3moO-%&{O{uDX#HTv}v->Q@gw|LMUFVI1}9o}BtN
zKz3;dkRx4D{p0%cyhksbD@5y2?<HL$y(k?fSJ7eS?YCm{9!p%NTZ;d}LcH8&1g<cn
zaH$!^=4O(wF8O$eEYNo7#p_<ZXm8MqE5bv@cdT$eYb9o8>0xNygHcL%od0$s+d%Sd
z61y{~Wp}=Bm3&~6ZZxPiXZ<P35?*aCJ~?xq>X@S|d=v1*jL!AVsQJ^CzE8v_lhu_M
zMO`I}N^<!QN#38gDLzA`mm#4u?nx?4+o|ZKtD^HieaT5xNiTzn&y!WOUa4Z-B?G$7
zHK3wR3-Q&rAS=EFYa2Jm>p@fDi>8df)12wf23(bSQeZEcQS~-p<wXO@sWu?!q4;mw
zHjsR5b&R%a;xa^w-SxCNx>TFa5pr!=i!<G7k#JZWwJ2>y{U?kwMVptawQ<eV#&(zv
z@BY@3%({mB{v$a$%;rePaQ+UMNAlB17MsuKdYvf7)LX=5n?)SkwTR@&?!=vT$G^4*
zN5*<EZG#7keLQI+KYzLg^Eu=yddQ>2r2kw@{d<df>9mM6pJ3W72;s<<5Xs{Uksgg;
z4%V8&kV~H25w`j9Ke^xcBzqx$Hpap?V}x(!3g4U&jip$098;}Ot{)Gh$H_36an9jq
z<Xn#Ym`~;TC{|yHqDYQ!Rum_jETH49C~ozQqHF*8^gb(JyDmaHRKhWA7(sb>1cOdT
zu;x_+cTDE+u|m#knT^HFBP3%U&vGI;Q60%S+2vYQKZ-W*7E!FTjB`_?>E<mnst3^+
z)mq9plcl%}lH47KWqfEH!zoMg3(gT1Nl#>;+RFds8-#DBew7aLt&+)gQReiHNjx3@
zKl7#(>HJIKq4;o4ZCFEQ!aAbbZs4IsJR`S=$HyvxA<m*@&rFbek3?=s4p^1&%_7ln
zZmv4cwVtQMcaeeF(sPoNe4coNOzfO8(HfOW`_Xb=8=Fat@J**1ndm*vWST0AQ%f(h
zwoNY0Uy0{<d!FQC=W+1441E9OiymFb<)b&T7jLoks$!<@FJaWdGD>^=L7Dr9vuSEd
z<6JdGbA-C`>!7-lGg?b|@2{ow{imhulRa&9BW-2!Fl}Y(v${%gSzV>2Xf+4D>M9#|
z)K!dxYxZretE6tIt4tbFS6MCE&4$r+mC4jqZVTVc`5-eY>6Ef<Ut1}BrK{)!=qhg|
z$1l8_u5v)Q=G%20rQLiT<<ck}Ws$j#lJ>Bc;#04dvPt;nL4~GrbiSry`CLPpu|c}Q
zrDv+OhKAysq^=zKsixG<{KKEGHTZ6=;j3E>Cg;RMy!RJ=Nx%3q?H7L!d_%+g8~5&f
zVceiEe7p8pdI~=C`oB+%e(({sI`7z5>jOne?}c~Y$?pAIZnb{P=Gt%R{qv3NbH9<F
zGtqE_Z~n=i)TtFOcyCmVlKY%?EuPR+cJ1|}Dp=CBf*a?{d9tFMP*3SvGArl9>N5T$
zmtxjWI6}PKzgiXJ;d7H`H8*JKc!icdE@QPxJU1Z))UqyMT$cg{zt5+xoIBe2#Qn`<
z^uavx>*g`eB^OVRTwJ1ZC38Oyz5nttU!RYyZvmRS3UC~Em4$5zdFxz=VY3@Nw7h}M
z$QyK+eFO9ABH|@SM^E;z(&fBbcvFV|w{Vm_y0uG-QA&zgrCTZ;aN@(VC`08?A^Rp3
zXbazr6uxoDswBv@N`&?*YJBeyc;F88F5M+8PxjsayTjKD;-B7kTXM#$SS?>)b3}Hh
z+K9I~MmSISdF7K5zP~GG=9fYqH7ca`rz<k6$rn95k5=>YiC=t$uK!)(xZx$-7M{iA
zZ3d?tGdTbA6duYcE)G6Q_sVqsMW)MRJi@`jM<m1UFbD4(LOp6XzZdV4tp0VZdcOwM
zw6*;0u$H#g>!^EV4atqxVAX3iS%1@bTaw0;zo{&!l2a><M-NlEaUfOviqe<aUGj;u
zQqXZp;YGakMAb?nY33?!O<GBluZeV#pZUzVL{5Yx%5GL74#GAuN+Rt$C!p~)p47(6
z*<T~Or<HMx%ZS6`P#pb|VmV+HD_V98pQpuewMUH18<$ZKExj<2OZg^QeV^oYsAua)
zDtd%7+#rm|8KJBbzPVi|looG7m|qzpS-T;e`xc0fdJqqu2eP~<kg(&k>Eb+_jW)A6
zyJiMPV`p&ZR{(cC1F*{UXWMbPmkjg6OPD3p%}=s=MXNdOhqEwFiEz!7JU?1rkl#P(
z$Ge4o(y8Hx#W%?!K0TGgvwV0QEV+DLC!=<65_xkcil5w*htoWHTkS#86c0j*+-2{@
zox=y+_}$7)G9O)8Xzhw+D_0WjT$t3_g)LiL=*a{gz8}wwh2t4>a~%67k0<8uIC|e4
z!)g05ocQQ0xd+bZS~#(}U?eU4oT&F)G!<bJQ%?sPbaEiSw*$e`9jII6z|Z!Mw4LTi
z(iulm>y8i))(Dx2IrCtNGxc{nlM>-f$RuY<cRAtK%@NfwdoG9C%l^DQ1wF-wGtYr8
zTO6b_)qxA^9oVr<u7M7$m@J&*;=mB$n|NWH+u;`2%IWQEA)17R%%m;oC<i%H<n#@-
zphcJkKZJQs3;Q&Tu|R8k56b`bpj~52(i`>U>(HJ|oM^?WXz9A#Cp>b)ig;n1XV<KV
z5pVET;gjkEmK5)>B*@>A-oi`|g>QP!H{<(sGrWdKr&C)q3M_;>_E~VovKPaf^kT|4
zEAGqR(=Ee_jAd4Y=Jw!5f9d7c?at54ZcGa3CVlDMi0amjRUNu<{fRk$kDJqfjyYRe
zO7@MqId-x;mG)F-QhH|e{?L{5s;;a&ExCTtU2&)}W$KwO(m!FsbFHSdOHq;FrlLt3
z6|1yWOs~=BY8@5dJE>&1MMd`*6@5Q7r^4TW9ATLHty?f>QghNiH|29#Q_fd3#W16}
z=-CEL>uJC};TuEYo5(&go7!!FYeWP5Bh{F>SDjiGnoKg#V%B6WW<J%T{wQry9kgY(
zs!hFl+OmtUO>!4)5}dX1ov%&ft=gEBX-l7OEv8?q#l7iuIq|tRC&Ox!>}1GrEEqO?
zB7X1Z)8@@WT>C6ysgXOw-Q6*7=*}|@cb>FzN6*rok@Y4LUOkZmZ6~oMW*SewPve4W
z7CZOOLU+(?CjJZ-|9c3U>qB_lJ(x44lPS36LFr=grOlW73E`Us(q-?OHk-P|vq=sQ
zMG?Mf!~zEV3*q+x(Rw}wQ|DVSY5#)htPz6qvM>&>2;=FAFir~hG&PPOqkn`vpCVB9
zgfmLK%dHQDQ};|b)Bc3ZTsndk(Gfg96T!Ea5!knz!_cTX(hEG7t#jrv|Li=u=aAQv
z@<<kFM6s`46s?~~XI;N&j!c%Uozc<ksf^~4)=~-@i?>JkCdqjj8TDfLrHZF-km%9<
z6KSKnlIwEJYOh48x00v3SFuv=_pTSM;_|~)td(>8!73*9k-VR<H1ZCvpk2iZhINT&
z;>LL9|Cb;;;|bKAk-(aR34A-B$nt-y$hx?bCyB>}aZjO_k-=2qn=e}DNjrFfhb=PY
z`<=;8IeSIFd37dJvRgA*EZWWG-dPwfyU1aKTz1O*B4S@29fWUAyvU>YU%vE_7r=J$
zsV3f}hg~sagl}f<l^x=de`s>-4{wBTBC^z!7HaBBmx1caq-pBPvT6-wWQLYv*hE`#
zGt^c}duc1tW3`pjv2~TpZgrI}O+~+HBlD?0dP?`tdP=%ZU1f-up3=j!j&kaP<Pjg#
zRlJXjzvh;%(z#D<rDe<7N`t4m%5_g&rM74{0c~`Z^mjVS#S=QpPvM&<BXpEDU38Sl
z-L;g-%d{2EcCtSuJq$5Z#C!8dLm4I+IID(hC|ABqZ{`|xC4Y@P-hh96%&+0prWzUq
z$PSfKLp{-N%4J5i%kLL2w|%3r$2Wp1zp!lR7j9nv%;edhaT)%JFTroQ|Nae6PQ77T
z)AtOweTRq4rRIsJIPljS-U{FRy8niTxo_~we2Ib9OUC!CX2sLzluD*=Z}*2}ov#p2
zNCnTjSJ14eoT549sN0rH?@<|PA4^GnRYFqB5{AbYiwE}>=AJj1@$Uv+1FrB(v<i>)
z1srQrz$3K+=6%R#S$00_g7P@?B#&PA^Z33dk1186$F<6(byg1Dt#i5gDVIF?zE1Cv
zPrnugq(m0*qNIQW)z>)Gr;xm@(!)LDIx}uwmoCuj6i&QGy+PMFn^MT~OGR`O_F1_6
z2E*^(kgvZ*@yTMEk1LT3`BK`gEF~(YOfs1&IR3W+>yeeTKU2wLn<^$3R<Y0NHls6d
z(@iuR$Md&2b-#+Z=~c2%UxntjN}l$tWa0J-)-<W0{;@LN`j@gWvV^zSq)Xy&ArEU8
z5~m2q4a{fSf?SNl<o{{#MSRI*{L^!c_B+e46&W-)$>95^Q$&3`iL=8|$=FEe+N^Y%
ze>y_3%MlEt5A*Bm9?ZOVqcLzTw%yjyS!)dgr>@3jzw}`it)@%yYGUJ7Gel-l-7Dqi
zFgca=M^dR?oyzL~>GziWhW8fInOQHDhovdJN=}h{vJ`swNayzrnKhTJWTu_$Z@*7u
z!+}J!7bH?WF;O0SB0ug+m&=?4Ue-&%!C&rKMlZ)t_~uAO9FEy>wAdfV{EQfagT>bu
z6N9~13^_96*1Eim14&CI4`m@5+vcN@6v<lua0VuZQF%L*)zf58s%t2(9)(aYKH`nX
zL-=oZAca-ZxlkR5Z)G4GvjWi)zIo*&9a7fP0lr}dF^)56FPS$-#s#q9s6UnBpBWM;
zY$Lo9+1U?`xqjT)?T4)}P4-1U!i04;9`a+$RzJx(@e|!#be+(t#C-DQ$4npTW}YlP
zPLpxTn1oB%M8-dod(;V@k}>4L_Te6soOS16H+NK<+$7iCjpoZ-8PdU(&`B<|Ntr;Q
zwRmREjOVB6cwFX=#r4@3Dn5)M#C#n6ZO8IPV>E@~qqud&nU05?xbttMbaG18UWX9`
zk9TCt0SAT-caV+^(QZtIb=(}dUf{sgwvHU~c4U%hIGu7vkn?6Fe?B{rXW@+dKxa0L
zb7sd7XB+~Z@Dq(@$vHd5>)LZ}qCK-u+OzwsJ;x0km~G@By{Ha+YV5!aO$Q!+v!~|=
zd%m2r<VB7-H-v2l%dr%eF_5D#r>&fRawf_77Gl9KVVnqIl3?MRrouPBWoPNz(H`{F
zvqYnd=s=CdkCV|8jm1{f+ik_M>%uwZ!Z`oOH^+o;t_a^4?6o9O_@+>_pw3BVR7ILm
zB7F01uo>1CX4HBh-B9LTsorBDyERt4UnH3~ev(i8+!Es~nT@$yv1V@%*3{}QnOxG>
z9Vq<=mfbjH-i-;uHy4hYb8DYyHtWr?8E(!TEpy^UzxgbDb4=Te-PT>XT4pHzCPSLK
z8cIik0Y61|`Pil@HnUY^jaRW`jEa6b!ZYeB_I=gokoaF3w3bdx3l$CfsTgy(Ip<vr
z=w@xepmqjav~NcAvZka2HWi<KQ(CQVPB-D3evJ)yBz)6W_{K}Ve#Z{+8XGl0nV}|n
zp*k6wn$%6vl>2)v<~-68{Yhq0(ORrX(4yB6ZOp>8WItP*o_5*{3DcI0JmH&CZRVKO
z;^fg<^u4G<GyU3>8P;Y~Ty2Ub8Zz#*1?E)~+4exPe7h`Sv+&Fq8+W{G+_)nCu}fiY
zWJbC3d#49Q8lD(V_rxSe^h(7O*C<amCwr2%)svn}J$Zjw`pvd@NC%+@>Zd#~eJn@#
zre%~rJ+cDOycxiNvH-eN1+e5(0CTjb;}9{OLBj&^Ssf(1`$2R}4&s<xe~1RV>q-za
z5`*|}Taa|D1`*mem|8u8(cc!#*sNeiR|GROHJt7j!)e?kg8Oa}XbIm;Jr=>^#}PPp
zn8V!gIs9ES7j@Yg({zgD$%#nj-iV|`_@<rkO;+_H&NqxE+AW$9gC#TcRy4NKKbqc9
z_O&`K<@my-nCQlk-X)I4{*vv}FA)P_n=5jRgm0{bZ=ML>B&}M>w)$~I*N)>utvD7~
z#z{YLoXo?PQ#5h~t;MfoWD?KL&GDk6$bObf0`~(G*sv!-dKeN}^<fn;+jgQKahy@l
zPZ3{`f!Fi1xScx39^La?@VLO@sTZ(2oXN(cne@udB(6yocKx!*?VCkR{6(61=CEm-
zcr+RmFl0!9=*sy#)hLkHe|i5KT7a%cA#2=aMq(&gYsbo&>G+2QE`RYBKTC_VqKEua
zQx^46SM;Z<E4}WE-$nMR29|3n?S*f4ir2SE!@A14g>{wTqR|``?{L+3J*CSVJ>~u@
zJ!P)D=rl{~C^O4-m1$>ml{sr=?|YoC;vr{qJ6%P$p{_F0OIOJorK_0Q=qion*TzqE
zlx7EH{`EilO%EOA>DF4xTn}v}%|%-Y9iy!*bkbG^Cu=HS?rA8UVl|Y#eWWM-t<2C;
z)Ri6y;uo6zPdbnP($lX-c13Gsm$in^&1z&9`xl-X-}yT88v$RxGIIMD+V%TFdCq5k
z_WMkw(ObMU-%|I*8wMSI!~GxcP_=r8?#wsT@BM}ef8WsH;~RXdWq0cIYYzCoX3L3Z
zM5vz9%k&W*FCU`Tu9CO|6|9|6!4!)M=3g&obVxZFZ_DVgw2ZY@rLv<YUT*Qa3@g7S
z`?EJWKU8|a)ubQOQTnlEZz}U*K2h@h+4MP|#FsKl%*^HCUFloulY{yDY(5rbb4ojh
zo#S%Y9i4+`{aki*%VmzttKMA5!%F^an{fqv6CG%G#~jk+`@dBy7gN#j^egjd9Cewo
zudeXO;2N#F7Gj%r9Sh?dv=_c<E_}0F9#8V`TcRD65H22>{obW)-dxJ$US-_aSw^67
z1&w3}tKGUv$qTMRZ)6qS({Hov+-;&WZqx8m74895%&A?)?@g7ov8ZI%mvY>vmE$m`
zjD4d@`1JD@?M4^LZp(E_=3c|eqyWFc`Fx(3OM#p4O^YnvOZP}&+YIr;WneWsgJwDz
z+;uv`>D?!>8h3&<Ptql)HJwY<N4P)Yh~&N;;9tYNk_96@wXN4;?6I1K@6t%pU(HYP
z4lgKJO~+ixSl^z;-U8`ZIGjeYPbv+CYl_CE^0sFxF-EEA>ZKC?K1K4#Q?QSc?#$KF
z5!El5zv8{wmAR7P7AtXop2&#pi5w10l%Cu~j<iZ7vM>StSqWH+c5}@pp7V~&F<BJL
z^n)?%92&#x)^aRka2^%I_{bPsJYuM46vNUha^KW_5nf%Ru$niYfzQHOay5)PEyD<U
zB%PRlLhvsQ;jm#Kj%I=42@RB-ut1701u{@({k}&7rDrsd+gE2}ws;mAmNS{ULvsD<
z3*S5mpm2ob5AX1oP7FUbzw>3;hN*<N@Z&;&AG_E4@lm|PLr%&pYM&o7H~Wz+o|`}6
zev)-O6^ofuS(5Hc!enpS$4n-s{$z&Bu6AerN%X(#Nu;&p@15~LG4-Hww>uX*y7MZ*
z4V#aynESeN`<3iQjd0;_>;xY4n?U}t@i;aZ&y{IoiOd<pJNGeITp3NT(9w($FLAc>
zC~^ll(_pz1*Gfi`S|htnS|d1X>By)=2dWJn$a`;(Q)>s_G<0B)Tu+R2kZgMg>EU+7
z>C6ZmAC6>Qt`jK+&fIF_Ot!joO8s@>a_tdp^t9z~Z#xXv+F|+Dj>&!Pu@Eii)mnRA
z?y={Wob_AmiAlAm(F%KBEVU<9GGC%kS+eD#IXf?yqqfDI)*H>Gf7u-KRpvw|nA2>S
zh3xamo|5p5kIbk>2;UeA-|R~6!KouXP}Q*{qmy`a8upZZ5GxAj3X5E|WSh*r;%{0p
zy~qlkY%3<5x1vILr944$`EK{5&j?E<iYC-H)r{uD%xI&SQ6c(`ab8zKg>T}Uca@$8
z3+W9Lt*A<LrPZdyt}<nbpDB%hbfHa|^eB8Vqy4aM3^DD7otbDi^7tPd60X@}PPzO)
zG;x+$m8m)VE}5ae(~R<CU8Mg)h0g{R`%_g|rK#AwN`;cCqE3K{91j)dqTM_czHu9=
zV)1=_PP~--;2M2&8>{%)MzkGE6_58e7fsrLg98kB*V=$mn`RiyZA!gKO>x=Nl+3W^
zR6B?lr=bBVV*{kIjH0CmGzw~fwO#|X+|@`PqRyav4bhb~iTS08*KsWd7-~sRpB8&P
zwAlDtlYo(0gz9RuW3V<^v$bivMw`+iZ7z(jMaTWMusfxLGG2Df19fo=sV)899VG+S
zOuRf3IW1W=Z`&{8UL$uB2fNEog&WtWyWu&@ji~jKtG&^k16$pRJ?Ku=4dI`3cZSBe
zOCFAR&FXnzE4t;(6c0StdoX0H2Lq0KpkL*|tVN#8cND!e+Ml38$<q|RDXR>iyYS5_
zi|ITNzIi<)ke|XezgGp}zcPrt6_V*7=Zu`9b>guTzFB=Ph-+<vDG<KV+Z#;X%wXD9
z1@pFH2n{!f(zRAN`u~OVDma|NO%aR`zWHT1M|KS7kn(aaAunVeHzJaT!Z)$kBgsgb
zPmbjxK50hNaj1BUWfy(nl_exRU&8F`(RkEfihZ}GtQ5Za-Xw-`;j!3Xjiu5tPWk{=
z5YT%C>H}7=d_@AW_TtNtj6sj*vCMxK%NNZ!miLU~eh=9b&0J2J^$O`4S%JsEcy9cN
z=TL9SOq`TJPq_zdo1DO`>;&n?N@Uj7oec9oE_rOH`8WAAZG~@My*x|Dw`Z~aaF&$M
zXKC0vi_}h84DOmm?><?m24t~)+(o`GzesnN96}@W<@K|GY4b%Bm{%a4`vMvX-~1RN
z`?nJ$2Y>uc&I;cwIA2b*-5*Ad|I5+IYKp^QHD%xzHRVPxb>+Q}x?*!%yu2s0l(y1e
zwZ~Ua*;JvY%;{ZMsg@q8(`WRQ*Dv*yZ@2W6_Emby$t!wF$cj2j*F|-d>{4B&`7T|h
zOPa3oK-k7!&JpntA6M5^5;Sy`AZuNvks_X(I&$vnDDOAxC;@YXZ|dqO^VZf<)(YQj
z7QP7)zVQ*h`Eo~7Nff@Zh}2NbyJ{$3pQ$Tx3F=D6rD{syvVR<Y{73d9rH5f+4FiYN
z$gHjgTbm!^vHMQ-oR18+^PZ08(g~IDj$zN=GHTLWLiOLW`iIP)_P@bt#T%9z%dS+@
zcVx!Cq1fsTbN;-h{`S{=vU$zbL)G*cSIyex&-gFsE*JLSmMoG=d}qmfx$upDa0LhT
zDo8j|&e&lxqk3H?&%ILa{VkFA`4U!~E2fipYSS+j68h*e-%Q0r+~_g~_KHV%Og_=;
zbLqP{hYlBVWH&vR_`n=K8t1U5B%7Dpvk8~$DT5r09ddZ*mqYf)9QtYGvMx5482ucc
z9M2{?HJh%x<?Cv)d3-AeSJ8Y1R^)Mh{AH${y~2!VSNZR{?2r#Hr0mmm=EvUPgz!!O
z0n%ah<Cb)wln{_v!uWBeWUVjdRF^VZ?<m95r<}$uE4Ut0iRG6{{`9Y+p;i^Wlq#%J
zs_1-DbQ||7R@AK`HK39WgDY5FA^!%$%Nf~HIvIwRFim>1YWf#3vE_9h+g`)7yXZF-
zqAA+uV)x@B{e#YMbHHf~7oH;M@fqsxI>YuaXK-?pd~W}f{IxzI_e{qa+3$$_yB@)`
z_Cadh*vI_FdokDE&1&n_Op#rvCu`DZIXn&5KdB5Foyv?(sn{(}<JY=0PDZC;EHkLf
zT~q1YEET<<;{RWt!q4R?C>6;p*_2G3xyk67%EwKTnX0vlehXHz%y=a$ZYQFVmWa=k
zMDFyGtK{26oln5nD*=Oy6+DZNV@p9So5f2UzBxv+&SQ9}ijmjm7;Nq3svbj}Mhv-w
zq${(05d%+3HpcV${5=$oLsA%BJj1YU8pi!sp)B|i!i)hSv~3$m-3Ec&^a<p}%0M)a
z1af+xaLpe1b>~1PUzyE`g|pD?J(CFi8C)-u-i36TO$GXkuhgGC)}q(!@a414RQ~8q
z<v@SYaw4YE<ht}KxcV`$z3fP-`N?d_2iuF@4BGF7Z+$P0@0mi~5mQiwP3Et`WbC(0
zlFY<ObQ&{}6$3okFZZt6^*#8J>`rr)J3*7(cyiB`&4XQeanyy*2Pgcm$7c1~aYVfw
zOZ4Kg?43TA)oEj>?ly+a2S#(=V6<?7Xg8(KOls#$omo!e_ZTT2xe=7#7jKMsWNJ@x
z;8e0b-ge@h`D{n~e|9Ww=0K3616_rEf=}7ADa4vPZr1cZ>Li)YPUOeQEbE~)XC7O#
zE7At9Hn!Byw#B-$9XgSARA<}K@24GYTiMgk%pPqEd$POQGs4K8>22&;(ZZgc$4zAi
z-ITi_rZkjeFBve)gG^BcnBwAZO6&AanAvuwld<d~O|-yC_~w3;bUStKflYi5%J%ob
zLv$AR01LjK?82XrF0un}%<q4agICLhy8}${$}wS^ttrRtrT@Uul))oRDI0Cdn5kV^
z+h4jak94K`!LHQ1+?D=vqNa4^vQbxt*6oUsW>-FZG8Ns&6is<-6MapYH&OVevnglq
zbiuWt3wrm=_}#V}p+V-%+AQ;^-R8WXZqD&Rt<Z~aiow4o;*D>@koPKHo={=CTZKuS
zN<4-t=?+&hHD1Ms=_&?IRuLzBvqt#luV^^Ugm1o8=%cFAXHK?0ey8;@)>F~4p-M7{
znlo~!0nbI3S>DWmiPp`ekGCn^9h*{Iw<+&>H)TYl=FG2Z#_FPG4877!_F5Wn_-lQ%
zy8IjP_v*g^X^qwRx?Y`c12yPZsm6;9>gdeYV8~ewE_rH7{;(z{W?HzZY03V87Qtm&
zT$lOPwJL4;jFpxFa~)n~>hQUpE~{O1Nfo|v9@CKn8_dW#H<6_mB<F9Q8|}W!uAhS&
z`7%=-J<g5!0d6c7z8NBXW4^_m?fcw`62|Em=#KjncgYs?AR^8K>y;kZtoC5|I^mkl
z9<oy^dD$mDxGjuxBFvMW^0Q~O%3r!CB}4OQ06*mfh^A9Ia=K(y1Tsgs#!F7MFilfA
z|H)Y^=g!(7w1saBg>PsXOqfHk^d$x}OZcW<NigkPgHSdHG5SRi4sOB3$?L=!=Ln8&
z7Eey=IXv{9!}8mbnKf-5PgCZR)@uph`YmDX;3ZT_$JL?k(FBWbb8)<Qm$Rf7qEa%H
z^P(9m9qXCmF@7blRnPW`KhQf?JiD<3SHz<EL0-E)$MB{@tmwtD(q$Km-{n|xizLVE
zNi6?_Zyqm;!>HAA;@2-{Vsptzyed2OXXE)@6VF`HZ{`I^w&|h-x~Ga4CpUo~vTtg%
zX9I`V?Bu)8ajNT|=79TYio8#=a?xqjR><|>Y0A4^;8ueyeznO${c9FW>@MOw{-XHS
zWrxZpNBZ{iSsGD5deCLA480<q5LalfTw}JU<lkQ?B!0+E@fOK0i&h2tR)2WwD10Nm
zPC47vl=Kg3O0OO2%8O!kMf;$pGCW;Nd3>*qGOUiC(rbX8Qo2Y_xw%eH2~E*cwq)ul
zmN|M#hXOstB}-446;nsCimam)m*^_n(<QG@JjAa4x{9xOh^t%cD!XcQl(*k>l&z9g
ze8WgrX(xO$wNOXtlcb}32-H!|*V9q1CD&3S)M_a;soKg*@%$vqe5~r7rqaV(Q~4RF
zp-gC{q4a&AuI%<#SAIPCN0eR-M;iV{$@z(w?l)dd`-)e^7y8O9YF*eT{`2_Ay$$b~
z+WkFV-QRKe=3A1c3fnYzOUDOq=(6h#hoj#R*Xa$5X1=1+su$AHSIuX)YRR>J%J$^P
zGJAeV;=lW}K7EhW)pyzUyNb|~O3}wEu_~yb&xQ(`SKMOV`diE$Djf6eCVP(G<jN?~
zYEImco}(M=Tz8GWORmuE>}CE;74M8qK9R%o|5vN|HaCan!8zzJ%)w`G4qaVynA9?d
z35T=McDu-o_1W}%l#Q1#&T^|9M)b{LL)#pg<7ZP*kxjkWY+Cxt=Yz93aygsCvpKxp
zlgp%^dAwOzz|6*%Q475+@7LG3CmK$T+I9A}E#idCseYQ>BJaa3VIJx9-YnghCZ)Vj
zEybm287Z5}7&=*Yyc<?fw@hYm%PMJdyAs3fN~+gZqAohk#pkk<wYH1~GSe!EEF)rS
z8M?wo<1P!^c$J8DUBti4>yph^NCWW@-yJK@Md6#iX?ZM|p3SSm3_49XL+OcA98f>S
z@T(`;Hua?VA5V~R;RNsZoRCbZ6L|MZ$A}}eI(~?%xC5x>?W5$)ZnnzoDN*zpT}SZ_
z%Xz4ohIwczSKCM?LyJ^8%6auSh00GU(j%P06uHj6og#f;Dd_p9VAWhQIL{`FH#?cM
zw#f{to6Lf5lG`A?3V}W=iEp}+N6VyZSA50p&WTtXC5nGkI(QF=r})`&ruK*<^I9y=
zr^FK9H<rdJF?8x4BfSSP%<x;r?D30e<FE*~cMCBR-KK0z1SvV;#Ak$&vO0{g#bFGZ
z7A6^pVYK)VO2T8=JMAj-eeFPE|4BdQ+u5w@5{UK^>CD_1D4wf8rsPgXd&P9@rc9^z
z9)IlS`Qza3FF9A!n7F{74^w=J=;4Ro5MM6b_QhLcDy3h2d0%@fd*yS<CcYFj^QD(`
zMdcNFbIjXY=1X1-YUm~Xq*IunOyRB1WTrHmOx4y&oY$VjP}hkJ8{>(MkL-qjmdu;E
z?mWtNqvsem>9}+yS7uY)7fVKQ;sn~x8&666I4tju<v`$A-d`WXU%xS2t38H<INACB
zF^Zn<qj+}48IL;7EFb5@#)Od!-!p=_#~ry_>L3i_z+@kL*6P^Ps@e|K3(;WO+lXFa
z&1rXQMvbz@zN<A#BWoPK4X0Gcnk~-OI4-rux6m4YYa3b>+wj@NR-Q+;{H|-q#u0Yt
zF0$j|b~{?0v}29Tu9lp&<BNRU{ix_Whwa#U+>{HUlKCQRQ(MkZVTi@TH+6+?oP}>5
zAL+zxIUQm;)A5rz4u`ujJFq)dUoB*Q*Ms|_uNam}SJRX(1l}^HxX_qsMaHs^V9YEd
zWBRr+rl`h<KOc;^^u~w=FN|pU+=yeJjA(9S%=c*~m`>}2`m#=pUe$@!a@r?#;=Y_p
z`M7nYbX;bcVzb{A6M1ZxCYef(o+(A{!d_~o;%VrDSFssw8=K*qVvgl?+3#*+!mg_J
zv`uTm{z?^>b5x@Li$?QM#nE&X@3yP3h*eP&rNVQ*FwQcWNzGJo!%s!F@J+1njn+67
zY3cf83)AemtdHFVVViyW{QIsivj%-i<@#}db3&|SC+co9%Bq`E9;BkDmx`{|D*Qd0
z(9KarL83l^CmV`KqyZnp8epGVpRYRq1~f_fH{e^>zXSXu)#ct)O*%`}aEn(bzl#P2
zGc_phuPI%`nk@3y#P_2nD`seMqgacrp4yV*A~UR#S`<0!5TC9iJ{cX34bsJPS#1im
zJ0Q!f9Cn$-iQTUF>~!PlAU8~3x?nAN22JGoyF&K=rU~DC*&tl=N#?CGx6)iI`ppbs
zoh}}%SnR=n@g7W)Q<>^P${G)@3geUndPqU42dU>g=y6wcO+V>BTj|e!;hVw2Hb3O(
z$@wIF6DfS-Eo|d1=b11~BRQ?*gvc4RCWz>*L2Ni1gmu$k`iqbG;htayUJAy4pv<!r
z>CIgr8qeh*Vw(iZe=>w-334Cg8zHP4!GVtAKen60`q^{%5EVt{%|+s2T*T<ci@B1#
zn7qoxG;6&C)z~E*-Yngog-e)pCYnQUq6uju88{tcSie6;yoa&umOP7gk3`FPCA$#A
zmoT)G+?!2|q2-}i+|S2SyD*me4`PXJ9LKugICj=wj@z2$oNBfLokYoZu1J%6+D!!1
zY?3U21op%wU~(gY@Nk*yi+|BPMRqcdZNSZAGv>_?(X7QGHpnS&d5Fe^N6Fc6oLJ)%
z<c>PQ#4+c1eLa)Mh8G#+dyzS7F4AB~4)5pYvv160cHO+pldd-zYIc){uHwB|Sjfk=
z($$FU)r_v7;i#{qf0o=Bn}3+}P*chi)fBgvYKl`^b!FQ<UFDFObVW54UO8Mx>04J%
zS!$`L7?0FbrcKgQX2$6$Il?%m$$H9#m3m75*}^-~b(DC^I!f?0U8Pd^=A7`&Iydpy
z7|PMqRW85LQEt4}Q8ql&Q5wj+s)??yvOxN#Mn>x>_x*H~eBql_v9*-u|Fo5wG;O7Y
z%*3okJ8Jk+Q^~omsVw%=P;#EAE6)nlm4WmAN>}kO4xRcf9bLb$KlKx{-(Pv*_mybj
z8?7N<NDcqQ=j$KINO;d3(QnRodq?bzw>+EjmL2!s@WST}lT%+4F!&YWOI{Ld@q&Kd
z)x2|iPJNk0MJ|0xou`j*$$UVm+kHCUzl-L?JB+HXLgz{)ZnAG=5L-so^<w7zyTy;(
zTfD5e#_DY{3z~e5#!s(G9#auLFJ9-;{VQ0=TxRz+VVVi~=vw7dZji^y&>Z<*=aA4n
zhjm?ZNFJ5LddD0ZiZ0V+!$poX7H^8yMJDdaX2qLqvTEf>o}GN|VK#wj*)02di4SA5
zX@2|?(+^#uZN(+p-OOf+l7rswJl4pLxyAB)KBniR<}G_#%?hdfC0TkDiI?am^_t0J
zmig4+$Hmh5AR11qQmSH0MbngCOz$%8TbE;=TP~f?6=b?r@@}hmjZcavr={$`4=mw{
zXE8??++eWmrPux`xynb1@S0r2!TLqAQ*)ih-_OeT;w%dfWw72MgC$A^$*H-R8RoJl
z;2g{RGB{V~H0I`~xH0}DN$cc!J^utll21@;!wIJBI6>0V6TEGdP8)Oa-Q3$x{+K<u
zZQIDi`m5Pnks=zb@Jw7PZDLdLpOA{zpj7^9rBd}M1^1d1Jj+sKr%1S_AVo4>QmELH
z!r{;qZX-JYqTN(%5#45NGMm(siG3&CP!EzwUz@~7`;|oMtt5NCbkDVy9f0GK-Pbgc
ze;4Da?Yx4<JEW7?DVDzav8;O>gIeua>WqkyIZ6ykua?qz(jv~<FCtoh0s6BgPxVj)
zn}Q=my9#G>`*7-{hDjDg7#;n>q~9P6K8B(*K7;^+5dJ=&&7y0wX;(d)pDhD9I!gRE
z$$==9)6qIMoo2z)`E)*j=4S)Anh-#Ni$Bww`7<oXkHlWStXk_M+0{Nw$QQoJ_T|_G
zUp61~#c7N$GfaJnP4~ffnDi(Vddn_~H#u=${A%Du?z|}s|1+89(tVIFUYlsqYrfC(
zz{^o~rX<tP@Vh%PhurCEAm_N7c-q`3N_3TOHdnTJxk$gx1cLm=Q#VwygD;Jx++nQj
zbdF)6-57Sg98Hq{Xl7Q8qK`5PyDiS>eRZPCU?;`|k0g502yQKr9q#oGtSXYZ)I&Sb
zckHBp)0Q7|M2~qdY_r^&<H9yEy{wt6vPSpUa4x<WPTG&*vd3jj$z*Hp@3E$!feor<
zHq82I!|SoO%-Cv6-+Q(+5Y8Fb)sBIR9qX;_q*K*S{6uyn4Uy|0I~;^<majHp+D;Ri
zoHLQULlZh&GokW^iS!nju-dW{)%!c)x3v=!S9Ic)oTgEouvG8Fx*sNV(h)v+Ym8Hc
zF;>Dgf3F#1Kg3wFl#N*{d~;9Bm{K)k3Waehei-rmhVacbBU-GK4B$8;I$btoVr@go
zJvO96X?qN>w8!RTdu|+SkAGuB0!MTra(pN03o)fs_@;@cc!6C_`RZayjGL+Wzq=4K
zvkN{2viqIcmD+=)-*RGW=KpQN=Wb2Ns@H^Tc`CN<5Kr)F6=(LV__A4rgYeBfcNL>;
zRXh@|85*fV-(N+KDJp_I<@4+H3EisCj-&cqJg-lm6Z%YAr%$)J`t*IMkLMMA96rhH
zsze`~OZxnGO`nkaGRL|p`%I7Z>0F`D+mDTj+uexv-5TO6d^1J#o4DBeOnUTpz|~3r
z1}rl9Js{uo?||vTJNc*8xap?O=w=!?kI>+ID^0HN(ID1IlO1*%RDV*(xw$5_WS9KW
z5pAq<Ye|2!4h^>Iu<D@>v%AQQO8BP#_4blS(4UU|W=W=wE6=wI+X&yBxbGr+60U^w
zk>{|F8@uMZaUjKwdR1;zi@vj|jyo@Xgn34a7g_cN`wH8*%h@G-V<hL!Y!5aI-}Dr|
zsSv)ID175u&y($*o&+cOqb^*tFkSj;4+ikqaXQbs1!5y?<1D9Cm_|d+i{(L#kColX
zv>>c@1ktiIh(~(TyV5C`!k$6g|1U_gw*%QN^Xa);X3*o(42q}Bln$?%ggeY)+3i4T
z8wGJXO7bHj#P^&R!X?X44Bmt?=u8+RqQbEpw3rSz7UQU~gjXI*XthRi1?{4!_KYG;
zx;W3MEhVnAcm(If@bOFxgPScS`{6=XEnmd1Z<6~bGj*StG3?(H!{ipREXa-J<ndUF
zwd1HWK8|-ym$R&LITr&qqP=$`qnm7Iq~&HR^|#1A+Gds}CGc4|$MaVLC*>Z|X;q?R
z3QLwoQ4)K4ZKd0!ZA_P}<6qU=*=4+&d1brt3EG48(0z2Ac>;r&bIcJhPRF^Dee*Ay
z@3ZsSIQBBPpGqc*uuM~78I7H{IJ4^(DT8m)z3)wSI94#uyMkXUzjCOohNs5=gzeOn
z&->Jr`nA**r$xHTAlY@&pIKYUnO;X3_qmSJtFfN)wx^yFB^u5r@!7PTp{Fbk(vz$l
zVV+1mWvquVj_}Q0(>h9q_=!&l-#lF@J{)abrEQgta#HrBvW06>9?2Z)o{q9fxTZjS
zH)~hvDBI`hD0e1GwqHXXrDb$2rMyusMcJ>d4C$b))GyOgrhe2^{yfrDob5D}r0eR+
z-sb-#v;H?B{eR&z<`=f&C*HH^CkxBI(%1VdrLVtWXY)m7*`L@}@sZyP-*dp^J+tH9
zG2`@G>D_+I&Wbnm@pyxs{cAJ~UrA5QOA@SJFwwu739iqXc=8#|!k^%s^^oZ+AFyGS
z^eh<OX6`ldoA_5Ty;nKO<z?JnP$qh7F@q&Lal7!%C*hmB!Z+C-qSusO=8kOvW9k=h
z?`}SpJ@Z*9J5kLu^LXMR{_j818zx<1zTdJj=#h<u=rLnEo@egKbKKf<j*o@s@V<DS
zoD*5>@yn+5m29cX%;s&W=r#MYIXo^K%T1TCyL5@K!Iy}gb_teULifDzO}iYXe9hyx
zLq6xd^SKe1Pgz+$&9n+R`Kgf7K1JwT++?QemgqdUaH%RrEkW{dnwFBiP_lF4N-17Z
z%7ZH<lzc5gx2<@O50<j!l4v_WN`&D{q)RPdUVCz7_LD_*?JRVbWwQRq1#%``U|Ys{
zx_F%>e{Tk#9ZvJC_bGP1Kf#sF!Ym)p(7b&HpVCiZn|z9ePfqfz%}M5%o}}F*`TXP)
ztXLrozxD)Awww^H_y{qNj*z3XnJLR<pZk@}oOX(TW^W2=!Yc#yQppgm$#F{MQj1hp
zi(lC3sc1D<Q}8~O!h$`r7quycs?{l6Tb4pCzZACjlKzxZ*#VDACUQzLI!4I^K2E~!
zhV(-fClT*2{^GtXsVq<AeP|*bn<SF7I)Q+ac;3y7m)+$!wr`BZs$VRv{>4yyD~44^
zV%XXxhH%XoBJ-EZz2PEiE=Li(V;<VCBPd=L!Tyj4t_a_B?HG<}au@@IZ~pWPW0B}L
z|4j`cc4P>P4a8@2dN!j^&E}KLp_<p1{IFhuTzfE`&zaLXAoKhtR|3$z5P+KSP4JNb
zZjbdR&cL5OhrEf@@a1;24>slA)Ex9?=U88qv%cu<_a$hQFV@00IeUBv8REmO>)v#j
z<V{nVPpN)SVeX76w7NS9-waRI6v@5o1`o-i^T5wWbop)`=;pc8tiL-;)!k{iOJ-BQ
zU1_n@m8WWw=QqlQXx|C+cN&k6(>Up)A4~VfWBIao4Av%N7*#Tw^bwNRcYYMTJB}hJ
z#+iDLoUpNUqSd64bnukyJts$oxJV{nls!#$*b()>mh<&&`RXH_^2nO`3$3|2*qU{%
ztT`)e6ZB{}TT6#?LatY8Tl2H8HB%z3i72yXMPD1XZL(qUM;m7JvSr~+TOO~oCF+c=
z?AqA!rPLO+N?Xz6Z8=kBOQihTH_e0*i6*el1S{bi8{wNVmrVG4*#yU%CfvDa!l{0p
zSiY_k!(uwISrOi-)rpgTOvLAGLVMwxp~5%e!Z+=NZx#sO+_E<28^*j7zL_I@Q}lm)
zQ!adyBYabH_5bsYhw#mr0z*#f8p`{TA-%FXax$m`i@iF~YG-@Nq-l?P3qz&}V+|f8
zIeQCDIU{W2u&E0lIvBCIj+{@1tQy)t^1IuzxUiY*AT(uQYk3z4Yr<vWn~imvQ0KOa
za$%YU`&F_RqY`~v#km;aocb!!l=K<CtTD%zNFS7UV<P<;Q(~k~c$_}7Hi~w0K%X*U
zn_S@><9K~OdF!)sWn=cw)2Di=K6TFNQ<<gDseFBmish=(CuL3};-VW<cYI?a9ycVn
zb3<BW2;YeJ=G3D4{Ag64pVNO2nA!5z0Edb{1IAtZJD}cLH6q8VvtW}tb49~3s;xn4
zlm`3$sWYyzI*)g%<GxWtdX+R;xK@i6mufM#i4M(EbZB`-2ZN@%{5PjIrPXbztKJ@S
z$#i^E<HGsPlCwR~je*53To=B%X643FFE>sK-#ik&*<ayCBjKB=ZQQYI?@n+}cj`E~
z)5TBbS5rNx<L|+lnI5FeIU|g-T-c^=r3c0LJcz67$<NlpH(s8^6;6~cvx)3+oXm(4
zU!qR<GquYMHuVcaC3;M*WIpVk5yX4ZWek#nDBdYvW3^y<$o<8mt=>$z=FRK`AEIk~
z*yZlaP{|ZmNv=b)dj32!^5<I8EG!queu3z_eUw0E28XcuVhGlqLK*xb6!+6%SS<@D
zcV7gHr_bR`!?`?_`IXJHNPPcB^00C~`uFEE{n>mPH<q5PAyKSxj$-VFg^atikg{cq
z@D#rJ(k_PXfie7dQ1&C5$71O%^QwcwHy>je^j{oSUt~VjXCpsiHsWBpiM@k2QQvB_
zWSVSYRm>Ki)ZPN3-}J0WAjde7Fpor52<KdoeT_8@SMuTDdcq!WV)3_4r0?9!R@bdu
z_1?<Ovs-Zv-A3@36NH)Mv(QX58+re1H10Al1Fw+q^#-?<o8nzBrBF^#zT~&vEM;0J
z@ke-8uuk&}1J`|}%dr~nM*Jmbq?%H*S8{=;YAH>GZ<0FHRvu5Ut<(h6QLYQ$JZ`9`
zm<!*`>aVAKAFrnj5kJlaKbcp#=qZhC^_0P*Wp33@Pf<kcNzkvYEIz5LT$mtUV#y`0
zy<JDi8mpspSf`@|Rp=;Al{(7$JRRllULEDY3LRx%sE#tiMMu$Zt)m>>RZIC4P)nJA
zO<PIdqOGI~-~7?iQmme9D%!tfx9gO;@<sTjUEpuokNAblr=M)E`ibV}AChw}dX(2!
z7QFw0uiqDbi+1BR=_AYLz2|+K_XOC#qiV}r;>Jq<g7}BqPm;Z;VXqm}?iF^sUy^>K
zn%<t((y98K_pZ;-F}+WL&mGAtyF>n=+uZLccl^RPUwo_N-?5z8#nO#Eqy!_;Z`^!~
zu?x9{S<73r-dRZJ9>O_Um&tG_Ahd1)_k}I?wH5#H`#gRe%j4CEJR084mF{fu+!P7B
z)XJvLs`KdHJ;%?3=kVQd4%5BoICAM6tsKr{J2{K}>Dl}{osE@kHV0~Fvug1r2H9OA
zXx~NL%rD{7`Vuw0E=iV{aE|cO;(@}uo3mv;kjrCp$v(Ch_Wg8?FNK9Px^kUv5k;Je
zEaLgvA{youap_kPW;aW)u__fGM=4QjOW1j|gxOC@c&H=&<TAUu(y^2p(<@}nC}3o}
zeA0gHBkI{cJj?eH{BR#OYWrpPem~QO@8|p2{e-UF&#pcDiCc4kh?;cV=N{$7fn&`1
zc${5-PcXvoBuA#4<d5b_I(0lD9VW*~>3N){ej8=ye*@8f*HfsSBs0NP46;gM$haiV
z`X=GwpG?n{;-8VDdXmD~KB>5NkiDqeDP%+o%Unv~M|ui3g=>z6rqE}03d4nK9LA=w
z!!kv(;FI~XKbej0$uww^Op~8U{97p9aR-v5zi1_U9aj=vM>5J~pWxc?L>${B(pGjI
z2N%ZE%PXF5vLAITC6)#qV;S%whJWW{__{TQvL}+=r&@;Nj-|A-Uc^lMDE4ldM?1-?
z*u6A@v7r&-R}AM~h}^dgna!@SS(rAMMf1lqY50C7ZNJVU$#XWZmr73Aw%OwEnT^-i
z*+jPpWV-N8p9|Bm+&G<);w3H?zG<2fz|lPc^zAGCm}b(0Y3EJ(IWNAClh?>tZ$`v>
zQzjp0^!8;>Ctu9;d@+3O!?p!JT=4XvP780e|9Rn~=Oz7oQzVafGOdFrvdPMmF$Esf
z$h~XMBoCGh@xZ^c=<=uC=_9`4(l>7GnCeEzJ6C=!cBTGb7Y2y8c(dmOwpz;GRLgPF
zZ$6f}H^)%_jpX-r7|ovC(frqMG;Ui)u~KIg8-tvAdfkazrcO*9H<J1TMo12{BP(U6
zX>K=rhE1{~ce^e6FKifQZzEr8&G1>)*qK{nreTfGgW=LiGaSEj!|}}<&g-YcsU`EO
zwhp4<thN4Mz30VH8-A>`p|Z$^9vZfE5-sT2a9g}(ZneV8mfWerHGcp9+E_mRv)qK*
z!ZzLIEZu6tdSRS{7frY(d=o2t(_Hu_<hkrkx9-Hn1)Z2Fe6#)k_~xSUP2hVI!v7je
z=ZZ1UWIokh_~wD|&AUOyOz3OOI7?%e*&3la#0YhBBU}uO=%H^U+-%5)K86HX88TP+
zMsB(|klB%~!Z&|rbl|r~2PQ3VPiyD)B(yOkypCu%5vKedXUee+T^MO7f2WoaRUZwx
z=_!BSv;hac)o1#OW^^iQLiNff>^Y_4+C%9!kQvp6wJO>P-?;Bmv17f8WZ|2L+L8x+
zN3<B>8(-m@KVFUT^KC3_s?V)x;gz-evV*73x+D7Zm-ASD@7MqEPfTOXL-biHe3Nuq
zU*>`OY!eMfBSD{zzZ=P(aRW@ve+^i5_xFI+P8$65uP<G^4J0p0_QE6T<G=dXfb7Iy
z1B|tP4rsgh&j6p-e+KNItj5c`YU~VE|9>o<cUX=8|HrkFGNR0c$lja#aT}k8kzLx`
z&dy#LDYH}dDtn}4D-~t$%&5qSgpyQ*#_x4~uiqc<>wNTSxN^>Y9{20@9PF#5O>~A9
zW%}B(_pVg3Gpo=fRZH?Hb;Q@MBfA@2oUHU%_FNy`*#@}97;ty8fq2fEu=7$=Cgoc2
zFKQSSxz3VzFEgB7&LrKIepm7REi-qe5YZSr3D<bI@}ZU+KkB*BtC6sch3IiQy5ZZ(
zjp_27cW&uM;&nGhOZQn<f6?5;xzj4mU3%x-X{zhNuWZpBe|D$K7x{m3JZgEc`hW*@
z`gsb&coSXZ&FL>bjO*dcYq`&u^I|4vZ04|J{v2WTIRweQMcr^e!q5Bhy6XhV9g^2)
z`2>1<PDBx}WAKy7l0`a|8^SjuY-VzF!z>OI%o0z{Y>XyKw)rKQ{nnM7^RDwH!)87X
zwhQ<iyO8U`H(&J@^W9c*0Cz4$FH7=q(w1SIv5fQ&%Q#zmIjjC#&c-pzdA%`^&H;hs
zgiGIev~<bj1WIP~D(YShCfqcHO};X}i3-6iLv~e%LO8c{4>mpba`o0;1`6M-7rv?9
zGlD-0BXQjs$@SZj<Tyu?;JO*F37e&xZZoRsar(_aPR!2Z1mz!R;g*w(zH$<a8PTkp
zAI+l`(d=0lO|a1!&IX<3PINMpE+tc^<uyK6q|)2$h3Ly=cQX0~BfVZQ-Z_o$bJFnD
zq+zJM#Pj)2nyxRVa@#K^t}o-0=x+{6Uu8St8=uEI>a7*x?QT&`4IN%hJ@lfQYFxj%
zS|IuxLrc-x^f6Qijxki<c^j%la#js9RFCv9RKLbmQwN7tQ|C>Q4u)3MRGUD_x#=kS
zVtv_B`5UNRh8w8gLJib6qPH<hGf+bg8>qDo7^n}|8mRiBy?HcNJ||mWT_@gfOX&`u
z)KFhd+OMacc&e-Voz+z<KkBI8^R(0j5i-ANCiAAf|46U<mrfu4ux|e!G){l;w)llv
zWC>^fE0IivpEP#*NnY=tWSuDBY1IO55BM%SR_S0EnJ*oYd6+KCV|Ryd_$Gg)#+fhF
zPX3I}=+AtZkwcL7TP!|gGGTHi9ja&IeJ+Co%`?c|l}@GjiH^vA#!*<{Xs>6Sy!Vuo
zUQe0(?FsEXp0G6gF@1z_CO5pt_<ho$kb6t6r?+TX^A@h+`A!>plfTVx;uLcO$I@#I
zIed-fcT?CJn#k}w37pxVKuSo0+}k8zu{43&BNI3-tP>(y<Fz}k(W050w<)}qJb**f
zlO==dDv!lueYjyVt$QR>F)bObf@IqNO<}$4usX=k@PB@T7VmEooRh@(o{2OvOq6b>
zL_Q8lqVMu6l%Knb^&8ofrQG1k9`R+hxx=|P_c3pJ7r)XwEJ?g0eJ;1zzh8WK@;n?n
z?Ibx}qPVUXMS$>)*0U3A$vQzst0-m&-(*-v(PVrS+S{Wfmn4ew8%}b(K{Ur_MYHi%
zGy(b1*cM0A_g^$RO=1{z>pU5s&f(hpEKO@f5MO&Qu9dr3)jN##&o$hN($MR>=6^jV
z>E1HiJQF6n#BjPeZQ=A};S`y_bY3ifHrqnWp^~lXEHfH!*-wqwLWuPiVh3*_r{5N~
zx7tEAoh{V7A5L;$I7@nlv$-IQoaitLhJ<mxrUt{Vp~M@9Vo|x7eZHHiF?ciE8gJ&P
z%_a(8Z<G$1jSOAAj-xW8wbctD>`^d-j|5{GDjf{zYb7@$2(3M9c-T*}CfWxw*-!kM
zrSm9X7C?(Y=~U1a{ovzSJnlY=?A|k|t(h*}H`Cej*dLPt(|PSVgL@-qvD;@BcMi>>
zN}gog)SXR#;hWZp{&bD;r~W#B+z(CT$&P6(>o=7u-KH{3<}WT)ebL(J%hDoW(uH-3
zy7>`j;K%cZK6EzmBz1=eP5!yjxrNN>7r5|UuJbll&iuOIB>DpJJXMuzc0ajZ4|3;q
zygS<#yVKCcoovOOB$?d=^>AZ>u^Z<^ds9pJ#^8$!!<V^m@|Jv`InFF?=){RcM||E(
zXM*a$M$I^C-5!fY&9TgyJBHFGW9Z^B2D6`|S?)EOj+aMKVKGX2@$9*ANqQ8%4JY;b
zFbZA|mHq8dnok_U=tMhqw6&wkd0R$0*>a<vEv<5F=y1)3;U{b`jIiPLVH=(#+0aKA
z$ECh49xk@bJZ#I5@3O1vY)8NpJ8JH<W4nB>K`-qn5oY?YNSLS8PV(~XcqC_)oD;$~
zRfKIC$~hKi#emCJa_zNZlkiRNr&c_CEo}3{ie+ZPH-4>o=h2$GvYYB6^P4BKpSn<N
zg;AjuFCNK0YnBzx_LlPfThRY^3#M;xA=>2@9GcOBrv5E3p3s7JqJJGfsyXXynltxE
za{>-Dr%PmW>8xu`%G%~k6~2iTzVY>Fj_Y{goT1Hl*RGlP(S>7#rR-+4V@Jz&jO^W#
zR2@;M{jkKKZ%w*b*5rlo&Db1cZoM_eXntMNX9~;6jOM><bqN)p)V~3Bsr+F=(?cdy
zb1~tdFv+tCGLLCt!ho4|(DkV!nVEHHG17$cK-o2IHKDtF?vQBt8kyz%_A#P!M`OAz
zuS3<DCPeO&-P8pWzRC=z>J`!B#F${*vnJj8mngrk7Ajg=g~|@)hw}1FfpT#1cjfGh
z0>$sgUnNRk<~G$!l~KjNmB)ull;7X}DHjsUl%%!gO74vc<@2;EO!n8J^1GH~I#nu<
z?^MC`mX`S7bTG-(;j!?IcWcq$eAC0HuRd)~>0{Q#fayL>coo`|h5nKU^u|d%<jyR!
zcBaa5XP%|HP^jf9If$<KcM~0%@J*Vb8|Ag#cr3bki~7>7CXBPQsT&=HYX%D6WLSz{
zu$>$GQ`~4g#hr`7H&ujh?hD^^%#b~<JYIV2PKR&qjQ`~>{aqdy^!5}_k2hP7dgImB
zi#=(c%>3a=Ry8k@s(G`-QFLzp-ZUEE!^U7gtO7;rU*IQt^$DcsO1JH)N#dEGOoyqG
z!+B>iZFHvcvDs9rFBGrV-dXb85`A9gT#k+nV9?0`-e}L`SGV~%En2{+6ALhZwSczD
zB1}pbvD!{@M8lVIS@Mm?y<19h`(+eQTSma$WiqQ<&Vs%x_$%FbDT@P16~5tQAoE0%
zeXGt&@>Z?lW6coB3JW3dXb4qqg^;l>1drf7jM3i9eC>TW2;am`+J{DsBw%SID~?8T
z`gtU?>+a{s>iyXFJRrNS1N1LGihI*zTt0n_CFUoX<bQ%#>lhloh{aKI&W6-JOWN#n
zvbVlK<G+_Me|(wk@k#7SO{S#s9&;KzqOHskcJvfqPvI-xOvu79E}LEhKXIt<SI#>W
zV%@J4&xL<jQC5a+rwV3nuHf;gD(b45T58{4I_d_$s_N?1!a2LDsZVNGSBELp)#AMB
z>a$LUs{a5(wU(KodP2@F4Ap+-hUy04o4n9!>bS|(R7=UaS#weLQ@yLIuXU@c;i9eS
z-p@dN)<xLHPETDR89JtI^wmB44b+U)25Q1|1NHeR1GUi1K<#~8U+pnPUp?ALU+o<s
zpZi2to$*vhZ7TYkf5JEOowU@29V^l4Ss^}vKdgQ7hgDnuaID84oGVHskE4VI{YubZ
z{8O0xCue&7B;t4hXX+G?y7N0-qw)zKmd|DJerGJnBhTs^ZLWM}PV5)SwE9f=>Kx*F
zz2n=I*R*__$$WpA-w59n#Aa}-sq{!~O(#?MW~OPH<mkWPdXHyB+<r<U=><Rf^a<|X
zPpJL+F|%ttrsk?twtc_D%muf(F!~nyvYYyJ`zD+I6TWG9lch^<NJrx}$@RL1?UoeG
zhNWP0K7qkuGPjwNz$KRi{`F5lUn_ytn=UgxHlER5@zfrWM8`(bTTq;cQNLum#9qaH
z+EpU1T*WKvDtm_{6Y)8jvzaOAtQXyJn(V(EZb*L06@oS<vZiw)4}T=^a<Q<}fOw{K
zjOVaLJjUY5wrLd4gw64k%uYaeUjl7Q6IkdXo!#G0aeDqKo=-W2*TQHP_(bz+Kr~01
zN0IX31WWIqp!w?)G-@41cCRShg>M!vh{Aqr6baSj`D=HA;6KM{IU<^}YtaPdMRTbr
znmd1^X)gVkpPI+;IO8mit4@>KDuT<tyIH;`j4vI+2#_qAmfJOwhoO<od<`}-&#{;m
z#?9wpBwPuT%)oF=Ov9N{EFEq(;WX6SBAw1#Sh_Bp%I;g}U3UwP_2se77FPcaCm>Be
zXKy$o9K-opGn}V4!zh^)M!aDdGq!~?=-g(k%QoX9dbnVZ&7?@)%_-5{q;?dqqwJ>&
z2TR6-x8y(k38C;)F!Rp_Q!gYK{ab7Kle(6pS<)dBzJ}StH{%q^U$<M%xdhSwZ4O}O
z!nw5iH-}c~vk2%ti>Ko6b`G76!QkoGW=~;h>SSC`N)AoyDGZV8*^K^^dDm<bR#_9+
zTHz~N1s_&M`;hGKLu{9+ytNjuRQIX)eDTKMhc|;C`Y>sRFK^EKvbK@TZp{2he(r<E
zX;1p(dC*n#0c9`53v$GrjU!yyw#1omV`s)ENQd`oCwlJ{KG73jNIQ3Wb#dp;On3ep
z<Bs+_@shZ@VbH;i<&~nzjC19BZ&$h2xnR=X1?@`7@(XZAOZcYlDM!LiIbbKdso9R>
zD2^CQy}U7WbRWZ)E2GKzJesoiqeTxqTK1=-cze>GDK+d-4iD#;>u?q=8pga;LrHo(
zglwxJJlJ5zd0m;qY?b{~f6)>Dv7z#|4et)h+-99{&N3Sw1lrJds|}CN+mIo<s*w$C
zS?6pkov89P&ul5vvm>LE9XX@y7(d01AB*fHM?_d>ot<=B*@@=Uj(2j#3f~m(wBnMS
z1;RIdFIv&)oE00NS>cy%MParTw|-kuQM)w*EL%$#t?Z^MtwiT##d<4CKGv|Lvb+VY
z^I8zJqXkDpTaXgmf}*bS-wj%j8r77V8_jWDZO*bTO*l5QzVxq)fBK_nWICDiz|Nc~
zM{}&sm=kfyoDS!j@zkj~C*^b%zFGgQ8Rb2ik>98pSsyH@DSxL~7BVAhTbp0OHCgOl
zleA7XDX$@q!Z-Gr#_~Ee=IR7v)_NH;H_n(}X2LTDCT#mq2bbq{D84827gH15Ere@?
zbJo=tEsgL^caJ)(bdf&GnkHB+H^DF5gysiKI2$SRnJRyjw8&DWX=th9b@z{w(dv(~
zJ6O2pya_p%O<0p?!UWx#G<s5^^j}b<+-aGooFA8~L@&)%V$bI)?#I6<br0q%ok|Ln
z*$;m!trwRlH5>g_JRD1u-p@*u+kStQ;;Us!s#S$z*j(l~W3@>8Qm*XmRH@YcU8#)I
z)snnIZC3u#ksc3S=Gf}-CrgjuF8Z=t)Te3{1J3ntBD>Y5gxqLPwvm%)y`5O2=S<N7
z7ltLfP`ipNE)85o-|EV{@vfNGaHB-{=7R7|xtSZ=gm3Q1p2|q(IQGIh|AcYkBgHQ+
zd=r%6MvjxabgxJjj&M!4SMKZ(?fnW9cbY74M{`qjItd=K=Je)mS1-1=^uqR%yau0p
z@~P4j_r_k_@9o9jv0fN`@+EtK=-@p4*t%Z!%jte}Z#seCyon4JzPYM389mR*JijiF
zrP7IOI+gVCQ;FF>i}c#FMT0RHvq1saMFikj5x^*!-)t4WIUc=$1mT<QEyQOvX%XZ6
z7vr^IDFKrI*XPYr(Q_@saL6*;QkL;@w|F|+NZx4gKwb&ooC^(<Zs9;O69PFTzWU52
zD>2!ziW`eUNRzpZ=5h!D3uH$n+QV;E_R?tKUg5!g41ThYiro?HSrke7=}21UMDld%
ze!On(XHc&LT$y=*%SA_NC4BQd_86Znjx%z_alBokNX(4l^wN`ZEjq)^9cRU#dJgSl
zaa`yS&w|tO*jBxZ|FXOE`FR)J2GSY*A)T40GSD~u#8A<bRJ6~b$*gbE2cJibMLxrK
z<a14@6rYi$(tTP%+g%m>3aOx3TOBp;ppH6MUsv^zoW9I=)zpsttE=^=i2qy9P@P@Z
zQ0<glUG=DEsJb;bR6Rv|+<tp?^|q#(`f*}4^;qv}YNUTv)v;$)^;)HYsx?Vpo!C@g
zoijjBJupyD-PA`<P3oqn*6*OF+85}mW9u5It@RDmxO{ze?+ks_yt}@d>aVB1YOAO2
z679_>;hS#vbyUx*I_fUT9$ehEl1>9ExLojuvg?27yY>$kb4vMAP$E9m66uXE=4o*e
z=@Wk9^YsVSBMZ1$y@1Wzzq7H;cV>*rr^mZIitc~q=A5tmHTuf*leyU2f2LN^dyY1G
zL%QE<=`YEY_mNCut7ghfE`tn{3{GmNb9UcL0%a!Vd+`OgY+kT*|1-v^Pv!mn31-4K
zbH#sOp#PXgJ5o7omWrp$J*Ex0g~PX-9FpDCr4BbK+H-?-=S3qd^Oyzp*D&`J7O@cx
z&A|j71t!qRHGxCj6J&3bK(DyVOi?dWrin*eKVI(b;wbluqtH<P-Gk%s9etIvO|IfH
zUUFpCOP74AocLrO$xLYFvTIDqO5vMXGH<S3p?*LjJO3r%W|u&aXFL=8$4f7e@QqPC
zYyZVz{w<E(!LeMLFFC=pq^sK^mWS_SsJ=diaE}<O43D9_R}9NGN7GbTCBrg`r>9PE
zxt2VSH}8}D(|z>XzmKiAWlp&!5{HCHg0=Uv!GAxc93f=yaoQzE^WkeW{e^Ev3E#}p
ziILt@`QFRVa#Z&;$392$yx(rx-wQ)+97f<R4J!gQjGU%nnXiWH0UD+zNH0{GhS>wc
z@Yp5YCbC!hmJ&u6-Eh&4gyXg{oMv;v@s~67wJ=RVI9>C?Nq7{F&X{mU%Evdh3#Zpt
z;hMejwS&XRou#3}qEK4z+|1mcn?$R=nO!nhjb}3(BQ{YsXFWSMtY^^55X@ad#Jd;5
z#QVXl+$#Gi-(ccn)}m$xF<Wva#;jk%f-ch2du=scCa<F1pr!a62w=@5;hR-+MCUP^
z4&!GrYTHaMo|ulgGM#T(Qy6w`5|=wn;`Qx`#MPdN&cX@&`{u{dVSe0y>PzG1z8EI@
z@YUIenC3pXmw5B&nYZwxHyaJTiG1tB$4S0?jP&JGu`k-?K79J_Nxs}io(}O~^FQGg
zxgKwIb|rq8t6Udd8S3WDp4(0=^>d<NrxP>#y0N6LcttGTSuFRdy$apvv|jQ-9OarX
zv--ZjB(FHiRrH0TU%2PQ!L?4*y6Vi2|C~8l!--2qj<nt4K=~KZ6gL`2*>suT3?9Q3
zlhMST8YTNO>6j`XNsri(q6HZtnss}6l@Dig&~UyD8_vkgp=@d)eyO!X82H(a&f|qi
zGK5oR+p<P<Hytx<Sar+>-<39!r)|SPXB+Xy*wALOjWC*x^pM$5@Jckt#<pl%i(h<!
zE$wBNGw!Y}Z*pwOFSDg>4Li=%v*T++JM5Z>7TMemMb7qME2atCbd+;uhm~}gTamoS
ziUz_rZ(qr->Vp*rOROj}ZcS~Y)(n@~&9e$CPIk1!LHH)c$dV1mn@gUv%wX)BbF^)9
zD$|<bbGsRXj#w}?(1NGF7KA%m&}yg!D-10-_`ELjmeeKbdlL@SH%F_hIsXkXr^b16
zDih4<qTP&H!Z#+uH^!Zs$v&?cQ4O1M^@Rn`pIQ*n(1HRz3tTgrGU8NIdW|=yKr&_O
z3*TtA8FPK6F=x_@_~UF$g78h`2RR?>u;g7Gs=Sc>)7?6Bc~XbljZ8=#VS;T^9iqcZ
zl(?)C<+SxL<#5_R<$mHn<@)M>%8vPel{dkqN{5=oO0Vwk6`LOKmAPFDm4jcuD=Q}b
zR4!%~Dke7ymGc!pl#_1@l?;nq#kButWk=R$<yPxl<-BLE()Dbva(Mq2#V06ViGB24
zx#RLvDg0NYj2T;^lzb~z+9j7L!Bzh%>K^%dN&l29DP>Be<v*o;YPr(wZ-t`$p;Fm3
zy9%W}wRmW!&Bbv#)LN}8JrKGCT-KA`N`1of^!a1fgf#QU3^~(Y^sVDLdf$oJ+D^>q
z<HGi<F2p=_Vb41k-siYrQQMV+)m(8{?<#%l;<1ukr`;M?LWOg#G;!mqochgWN988@
zI9J^0{?ZMTuWs~rbtloyofvO-f@{i5M`oW(E8Lkk#e?zxJlNXEOL{B47`oY$U6(zX
znB__1s$Mv3lwP+oZybDm*i_G#?XJF<UiRgAOFwo#_M@Tk1p0VQ;HvjT>cojZV9X>s
zwwuhW{nCS*JB8&IQwf<fm2)j-<0)F6qIYvR>^zrx2SvkUESVF*l8bz50b4T{u<P|g
znz}7w)Q?4cf3=v}XC(XY<x&!BEMwxJWeg5l%46T9L|Lw2rpF5LGOVEKvK0(o6Ud+V
zKujM6l3r^iS`I5I-X_^yr7LOMbrmz^8fd8<!s}mh4Qw7l!SoQ;Cx!6vtYp3~jHKV)
zNNoN^vU}!!>A~7hh}!|V=R1UX@*%cc9br)CV|;0LTsoYO^UpSlX3wG+xA-IvV$ZNH
z{467!&QbMf9RJ$KvsnBoLk;fYB7D=m=q|TfrZfLzIuYkG@V5Vi&Ye%JIrW7)J@dG-
zHII;5`2?)Xr>Ax)^M;iYB7UTZK0g^>QOJ)V(XGtXQ4`<ksQqf|s#88zQ{VTgu7-P6
zS1q?xS0|jWu6|CE&dh6ani#5qmDN=}`S?uj>Z+gYw604xc-JXa)lLnosvBev<=Q}I
zGB5Ph7pk7xqMx4HOZdj4tDgF;y`Gx&T~{5FqpQAusjKb|6EAp>t~zgtuKIGOu3Ej5
zd@N5#?Jpe-_u_Qa$2GLnc3N8Mh13dm8~?@o^B)YvH`OFn_$;UtSK*uX6H4gxwV13;
zKe^xcCl+rCxg+~2hl=mq+wq;@XY#2u&gY)|4*qlg$jfRUiBHU?UG?{5%z2Ad#2c#U
z$WdPNb!#TR-7}e&kiq)48C;#6&b9fk81&yu?u4b$_xuZ54ST`;BhSSD^pyQ)o-lLX
zV}6c(EV-kPiP@9NF{4z;vb@dBfwx%kT5=cn-Q>fOYfKP7_Xp9)cvhxx;;QIp2Bq-p
zv&?F?Cy+WzW;cUHTU;xFHZq(0yWleO%SET_8&AhKacmBVlbql<CbWp7g<l+=xsq#r
z_X=C(alK?cj<>tYBjpNiNlEg)dPVZ*u2Q}2RWwf%(bP?(f&84&!cOLnG7st%&yMEt
zocbL{^u0K`ZjPmLWvpb|oMK${Q;gmlgS)WJ@wqYF_Kd+&_@-I!7}|wKbGly?JDWvG
zCPk!l((I#;-ZmcYlw7<B4K+7NU&$*CFZu|391F)cKavh(4zP8{5jJZb=f&e_cKnRy
z&7Wvu4P#h;=oISuGx$qq=5gKqvQOE~#v5To=!E?*7yO8@jitMWqH!9!dTCgCKz>fT
zhBr0CXwou_7o)-$xJk4#*TUGkP_#MeVR$=-)7~STFT2C}DEp{J_rm$mH=M^-;T)_k
z8O7S+ytpNwKR=8HX0o5^r{Tw7$?EjljA6Fu{oiiHXXggY@}v*^`Z~reUx!n22&=q9
za4`#^?cHEfgM-mq7))C4V4Ch;OH=c;Z1fDGEO-rxE!WUk@+a1Lt&&cur4Sl`_c-bO
zUOtD1!Z&SK&yqgBnXLFZgCN;Ut=sO;xkHm^-Eb1KCQp<-+C(1Cn84o;exx?`<K9VM
zqFeh?mgGatxjyI!-;Db0P3LyrSR{Ml{ZexH+WWBC)0bj@Up~L|;o&<^I!X3N)z=;z
za`#}l@QtaLE2l<^Hdy#3T=?dK@Qu|CCu(<g5^tLmL)yD>|F7r_8c8S5Tj|h@absej
z8<8X2I4*PhJwIIe`%E;&wVb6-#tDndj!c^9C_Q$LeEsCWjBLrbNf?W^W-N}Q$8z!H
z81dhaX0GH6Ukx9HW1~^jelERHEk@AS-ySzZdt&{E^U{4dtL%sIxzA8$-x2+9zac~%
zvg3nrN1JF{x(&6Zv{d|3muxt(#)i}GHZsGuA)>1dS)FXS+Q)_(BW>6)!-kpLZ6s6B
zhR8zUo+h@W47O#kpDlxy+wwHbmZg!lbUtoN;}}~O$BHKTv@QGPlux$8M%bo>oFd_y
zuG_6J4zr?*i|BHktax_A3R_{F;!jqRM`$HkNUaGM-HbygOP02>WJFC%x(MIwcWKVo
zLCx9lxfw}!nsGKpcqYt(8;dPOlV?G1IX~<yq~FVe@A?)TO|8qUDbn%uqY0tHIEB5<
zDU(@Fiy(7Ooi@k2aWmGpY(_%+W-M&cjJlZ?41I1v%6$vTt+K$<)&je%rd&yAN|oKh
zMMGu&Q*6xf)y9})7@_sthy&w|IVpUz`K~e1!ZzVDyRnz^>}DOTvg(k0wGO?s%9XEI
z3Y56;0_E9(AByh2LS?|-A|*4YNLkmqSlQ@TtkfG;tc*UHr6i5ZQg*b<QqJpTDO>J+
zP`safS7yKat_=SBT~Rw^E4R0ORN7>GQq1%|E1$1_R{A%}RUSF#Do2EGj_vrOMD@;B
zE<XFN#5Db>=&UPJx{oMUjD3rhw(+IPoYQ}l4*UNqLj(UQV}!MmX8u*I<I0tsrxl9*
zi%P|5Miq8tYhl?@TY4$AWsfPje_6Wpkc{Jk5<Sjl>2pNWh-0RW@Qr9M84PV1Yb9CQ
zSDbM58c*pJM@~0)B)i{0>a`lkg(d@8aB%?rFAN|degG*614!>bkg(5=<VVPSGQgEr
zuCB~F>muGTC$^t;B4oJ}#<QH6yv|wt`|=u^<A!yqJ11nf`|XEt{eB0svmF>X+m)g+
z4}wp6((HjJ?R_OXbDcLP^4h)k$cMJI<vzs8mrWOadEeAeW{G}szwSrZ;S*@`aRP;r
zl4CV~66UQZlOcSw?9mkQr%t8JW-2dEPG#>f(ekCuky+4O7VVu&M*aCjOq$Qock?mm
zxsdJ87K%S~5rda3rXqbYa}F#;=h;#|3|PvJ;Y*pGw1j6~OK7llvFHO9bE{$r>5Ese
z_xuVLM+EZjuFQ0_S7Pri-4?S~QnYs^c0E^ND^0keXM^$259Vy+5Jvlluu|qa&&P`P
zW=|xGUqsTq#(s9r-p{h!1N>fokjG69F*@ZCHyjQ##_}kwza7J}$_bLYMX@3^ilMt=
z`E~O&9d@2$*!gqB{=P&`-8d||#BqCA9Ao7DZl_+lXf4y36Pk`?MFzUgpLm}2iBY1V
zZPP0cz0G+9O_5%LSjq0~nlCf`eA%ymmps~h$t}yr#QcZofC@25)mF1AbX3(yS8X6Z
z@jvd>)fbX)V;5Upa*eC2H)MuWCY%#kUS0kANp@2Us;Q6Oi|(dxRka|~Kz)D7Kuzke
zub#gxT?;9C>h{ID>UQbNJkdi>eb+@#eJp(AB7D>Llk|g&j<{>MuDV9}CRX^SbcW>a
zbktS%h&TM{3mtV(FD>;)W)(Hxr;6I=VFmsse|eYvN4zP2$iG?2o@u2xJp4_I`6WDf
zTg<kNKPeQx`Ic44>SG1urF<u{^eg`^<<MbZ4vkiSB&8%<_L|wGeEPuEg10PFvm_(#
z4PSNNP*|DC&Mld|?wH9wj|_&dPRGbGo#$=SC2vl;CpV<AvR@j-;<kT#{252uJY{#_
z6Q<1-|NW@P{OBaw?(M0>R>-dK#ck=5xy69wo9Kkxq)cWu564~Orr|Z|)lNYTlvzu!
zMCM*j;Nivu`uHT^&>?}1-!Jod|7HBSOoMCjT<;psp;K`<{ud{^xJ!(9bBUKpmzWzW
zGoP3w-uhqRmvof<{+Go0@FZMLCrQ876%wPQ!^0>Ey_1Q=mB{Yv!DZerj;G<sc=S3+
zMs{T!X2QwSgqNCcjKw?g6t9|{;z6}j4Bi*R9pRgD;hVo+F>GubBU#2Wf3c5}&Y&nR
z=|(a&XOCnLYZzxO+<tU3WA#GG+z`t8hq6yH4`Y_F#Jzf3NZq%c4+jp&PX8!V{~bf~
zC7SxeH@zF4VzW*x2Hj%V+T|ca7160(746Lx4TEnBC%H<`rHe)~pEaZp(lCCo<a{^L
zP!g=+%x4XEg-w!Vj&ode$6w{=b<WW+HeNj5qDRhN6vj2*aOu+uXU5_%wjT}S;E^zP
z<b+WrD~t)p!YCUX#@uWT-|A?%+)?uU8cN<x@kY@IZD6+H2G%*Q=l%)N6ZeoT@(Urv
zIE2tuCj@(WZLgmmOuHVE6;UUc*nqW6HeJhe>0hY+YYp?-tl`<#)l{@tMVD$zIV>KB
zQF0A>Reui6#q+sz-7JFl&6Hin4DOGePM-vSPFwhs`(vW?Z%w4hl?n2DnZR-n(a^l}
zrL*jv{M-A|;-(MvXZv8*z=!jjylH;di>@cVcyh#xipD-1wDMu$3Ljp7@WkbzCsQIl
zMT_Bq>nIQ2%DrW#=qQ^D-&_;EDYtQDbw5`;Qe*~W>cp`yCqf^%(j(iAzZb<b5-U2(
zL*gZwCVp9WHx`??G5(7yE$+L}uhfY<b;c7D<0$Vdj!dlW$ivSLta#)gbF;C$8#7ii
z2FEb(`)FR<k7n27QJBph#mJ*0$u$_k-$Z+!cd#dOzUXcw@8*{4FgjKphAO(7eG7(=
zDf*cEu6A_Fwv`NEnZq=-#r3K9rNV7w_GrViUN+QfWW(BOHXPIuj*&;p1~xc!wc*DY
z8#*r(<`Hd9sBq3I(dEpkV#^0pTV7e&a;=*!Q~KF*U$v!Zpe?C`Y%vxFvJ}2K*VBr|
z|IarC;-}IQzDXNr#Zw0>ws~7IY?;h&=2$Xwq9v!sS<=qVk~iJO*WJ>RwEC9ZlO5Ii
znk_g!v^l22H@9y$BUt=T_D3z)D|{0rY|}wbmher_-WEI*w()kTOU0zRv=F{Yt7*<{
zVI1>8=KOXtrzG5*huh8NwPh|nQ%zY)Q-aeh2z+LNvGC1&;hPTfy;i?zO3A^d^6z9W
z-b`b1H;Z=nwGo%njJWsMh@u-t*zXgs(jD1Z4X;U&%xofV)nRx_9nPlLVZntulK)$#
zOgL1ajEnuMeBP3+bg1)Qsp0TW8Grk&vfJ~m;$`qw$%@NTToz_2U(_tczHXM%?)Mwz
zb+atxZ)BEYo%~ig-SNFrwL!LGvi74=_tqz+UP+G9`l8Hl>gFn^eRGw^Ww}a&3tyDt
zE_sT3^#bMhk3uD@f05!+S)`O)DN@{J?^SK`ALXI-Po;Edk#aw~Scy>nDjzSED^rsz
zluzbW=)6RWhyuwwZmrEF4{em~+IU>lW=W9_P7!+Se5Fsn>5aHMp`r8+HbmLhmQH4E
zaoglb#$?I59@d1?7mXPZ(wJHs%@{l0O!B(TxR+_lt8i2HdYZE4MkB`Lw<ajAD{c0U
z;N6cA>@OZcsg2}j|2Kln86%i9dxUgMjUaH@2t3-4#bp0ju5BJie=`SqHggo6+jz3I
zJh|uTiAl64+wOZ(&C{DdYrJ{*-J5`?KI|9G+?<iVEI#ck9uz;0i6<cCtse&mO`sxU
zf@BU(<gLta_8ptdgyvJ&b8!mcRi@IW&s4<dNsqB}_!B4hPCo-^V=+&zU-Nn6GoOI>
z^SOF?0aIr$<hAw^jJqyjRN@lcwl8Jy(Zw7YyO{D?4fwU*jB~DL?Cxg9s>X{MlCem#
zt|bRKc7=3x1hVpaAP0Yo=2-aV{mhkQ2;Zz%R?(#YYDx#LrEZ4Y=hY1%c72H4uZ7Ua
zEs~_jNbbCsIsBCUJRNa>BiRSIy6PZqjStZ^`VhVC4|81W2&cV{F=**=er`V@zhmhX
zd>@O_^&DrGCo(AgDg%CA<#>%`wzo;<m*_6i79~?EJ%!<EDbfpZos_;eB(Ls0*@JU1
z&;Cq@OSy!YeZe_X^t6uOB)=|?u)I7zOWxhp?)j9?6m81dd~6*H7~b;-3lD0m$+D-~
zQcG9$e<0o8_p7O9nR2qKsR4JYsTon#)Qv&aBp;`mx=yq-!$%va>JNSO?m2yR++uyT
zW^aAf^O|H3`-(3sP*-)UudB9|+`zk%MVzp(irT7v6}2#_l9t9*)ZQnAb%bxSgm1Px
z=&CDCbk%p$bkvktTI%@fT589+Rn%n%DtT9@jML_S*(rRpF6@uImP;AtR?5n|zfp>R
zQ7yMvavY18JoP8fT?$!ZRKS#z-x&J+E2$T9&{K0by6Pjhe`eFlFPmBKKk%dQEkDVk
z-`+Qpeei~Qqh2$@G?S4xq;DxM9bMrYKZkUlwoeyd?Mq6QrBS0^8gY_k_w1zTj9WhC
z<J>2l43O>%`^UlokIB)b^6mR0CilM2j9|$retZ*;v*MXreNB2gu3?dzLh`B<wsc5l
zb<afZo=)KAngpU861Y}Bf#B53?4Ng8a@j6(W_>(eg+nZth_~7(PBLdNaeDhDman*k
zNxwMe4NfAsN)oO;lf+M!gvGKXmii|V)GbN;#!1r0l}Kf~MATH_o1>Shw;-MqHo{A~
z@i;w?<7(G9l6IVC!Qxm>Up*yW)>C{CzKOgML-!LgOlTG(_s&rawuxfyyaQZV6~Tk8
zdpV=Kmlabr(hm?y<;>08iP+4-s-a8`3T4hU4XbK}QT#fLpHISxDBDU+%?{pmKfsc{
z$B1+|!9u?%7C(uhU|BR{>K<V5;eC9o*hR)l4Sy$!rdYUU;AjoT!!(lpt|76bbh}wd
zk8NcrcxtFFyQ|j5!Yz>+T7_t6|4~EUISqZ+YnU#7UieZ&%2ti&Lp4mR62_k9VK^_B
z{#$=xADIss%gm=%uQ0h+(MX?`Xo^cWb3;dRsB$)9lDvT<snRReW<3Q**U{W+9j8u&
z@Yf~;lLFBb?+T{d2$|b74*s9+rl*he)TIWIJ}L;Eyfs{!v4;FBt1&KKNx-}%lpYoR
z->SJRtuu!Y`=xtf(=7V$oQdZ54BmX3PT+li(WLv6nK6-bohGtm?F3G|5ii9+Kej*d
zrH0I9+GP50#=(cCi@dp2*PFmyUbv@7zHqn~y%&4&sf+CNwY|_U@Z?{rC%IQVS)uWy
zXQ2nf`+D%X#GO;_qN^0P2^iwaWLsBCWQG$L<wQWqcw7veId$2UU8CJ-9_2>WXg3}l
z5FNr)H)1{9<lotih&!(M-IV*-T~3@f8PCvJjx_A;NXr_I)RoNQoW~Bhop)eK_!u58
z9K(S4(L698P0+DXTpTLBQ0qrBvc#SNPWIBZW6!#P;TUNRm(Jl~{H!vJ<`ahC&}N8q
zMc7Fuyd4K+x0GjNOT<qbn#t~|{xlnU_qO3fP095uw5CUvHN9R~vo+0HG+Nf2Ew*N(
zu?<Jt%jXRj@Ay<3`mMDgcApKJ(>A=gV#BLDve!zrA?K+LC!gEk{KAHHat>9r;!k(U
z817@mzP?s`w6-E<sPN7xD@x>XxrY^=ftIYBWr^ZtNuHx6Mxr79+0T;cmclnCmgx6t
zN@?q+#J*~bb%Er~l$)|l_-1HmGh&v>{H1>r7WZnxr;GI%_oOaEN7SY91nG5}DA{`R
z>f-ym2`hzhreA2nyO<^<?`p!ZZB6LgvH=HbG@!3x1Ipz4&V40*?-%uG6<d$d2kY@f
zc2aeJ3fIVtrsBL2$ybapyl8~JoNl*`m~p{~ocTtix*HMTYJ}qeBd*KL=BK6(N{>3?
zC9T8dv@)e@M1hjC>#H(2G+S9{EUaSxPWdL>GScm>azOX35^*j|SvNIHDQTOfT=@G&
znfUsRvPBptD=15udG4*UqSbrFtWCC}JK&?@Z}CwvnVGHhZ1Yi>ci@vU-#<t3TJ%}j
z_$XJ|uYOY+t<6^=)_hmW+7~Dt=M^e<dlV^cClx9i#}_Kq3k#K9?|v$`I~Oa}W6P9b
z`Q=J}*Ggqc&nlt^(89c2@{O&uDIclLFyWit7qnUYO-J(A^%&k<pUP>qaoy4Y9~XHK
z-%_6smGzl6tr3wm8sS*jkdxGB%?~pS#A9}Elo?xUnsFh+l+W&_vcoc^_W)D&&orge
z0aIq1m{Vud5XwTO%WU9qhJLYUzwZd9O&>wy2O|hQGMW(gvFv*{mfefT5qZr)@)sR(
zvhiSTO;7$hdy>Aui@$%oFn00gN1$knMKhNy^PBqGzQkAR6F<~|b`b^y3g7s?@Z(O;
z34Dr}z<`{I#NV03))SNIZ!v}3qf@w5GKCf$M0eMB4kfFkr}Jh2agPF+(sUj{_2%>G
zz&th<O%vVpRM`zo#r)6|b{Y)9{g)k^zSz;Mb33{BF(mZ4A-D4k$sBCNn|-Eu|1f7r
z{32$Xt-!%GkVhK>2^$)S!HPhto35ls+Dhs#TE(=qRXk`PB=@Fk>3DE0AB;m-voVCw
z!y!0&3ft_8WJq}=56|x>?!N;pOh3TR$b+0}e@JvUhX_^<6IXbc(Zxp?bNncdWihy{
zJq4MP>-y*tyH;GLNGFM>|0U7wyLe)>u8{6|1^30$OO>2V=<pPpFHJ$Yf1MoH8?+zw
zfn9HN=opvFxY}RP{rFY1@ZX5h<jH<NkK3}d*(kG|drtYZpPkRI+(PWW7V@lV5xvuj
zI9KNvhyMFTm8GS;oLnj%^HMZErJR~xDtf9?EY6qGD?{|Yo%PkK=KAW1Q~K)o1^Q~o
z9{OtXLtXXef6~*rQAceRtF8LvXsLTTXsO$!(^GpuCA$w*V9>pShEK`~Us}$nA?1wj
zST0$R<@9P*L5;*pjtc+WXk0~&SzSg#`9JbI{UuoTQxQRb2>DUU)FGwJnD?7$TZ(a7
zR>a<IKe^GdkXWl9qWR4u^}sju*GbR9^&Bcqa=0D*3G%YZ8Iw))tq-hmdC$E!Zy6Ar
zg=WMXj#PckYn@Df$7L|}bUHQHNVnwJbp8t8ypp^-?a^tp_I|-V@v04N_MG2so?*1<
z3BS)grtc8R&AXo}@2{y8zJ0_h$?e^}P%`~GO3vQq8+c2$U2damIPOW|N%s`FXShnC
z=whZvCGdM`0zZc(Fhf6qE9WlDef(vHrpI&7A)aQkKbkf;jwz{^=)L3;7WS8z)8`T!
zMqi@p=*wJ}uXTBw$b8d8qOK)!vsx14S|l-}G?9Jt5~*7)k@FD=?B6VU^ds@y*%rsx
zEMcZwahz=($Byl%$=n^wp+2!BH;?6Ntym^6J4NS8*-QP|PtubJ2KYs=bE$MT1#QK+
zMkr&BY-T~n&8%s(nf~_DLn--9$9sfQygHPkGa3S%G}y(9ex^?tFCT<ab<1|VTI{4?
z$1eU%J3!k_2k7h?!Kv4ge2v(LPTk$KaThM?uaW#r4bOya66Eoqu#Jwe&Bq2BIv)<D
z`?^qS^^twm{ZP(oYuMIYL*`5kYY)q8CrBf`1{#L^ry*JPSyd-#FcW^+mZhObx%{0B
z!^pG?<5T@GI$DR}`9(uVHw_^>LWP;dD|L4hdjq9Qcl8FgB(5j6k?_r-b?Cec;dg*!
znAZ%E`A{%5Rs}Pyjd-T22<u!5;==JD9_$U0_pTrsXRTq@t~K;ET7y^jmEzA`%%FV%
zY+oR~GL58jX3T8X)|$=lkF#*vI)fW}Gq^Zrx@hL6@$B(L?iEfzf4=OdZ2Yk6<;Uu?
zzAUuxrO62&(E$3e^o2KVr+TB=<i(s-UU<y+k`8e%d}QB$P1t76drzjxUMlXQ+&2cx
zj;cuf-oiIFFWiZBb7kyMSFD9^W>~w@daz^=A9s>&r15fJ=*&XV9+XCjKk9@VVZt~T
z!Z$0|iY7<qHz#*Xm!_{P(c7Ge**l)PrsFC1aAZ^mM_hCqX`kbOm+(!aiw<1M9WD1C
zqp_Eqn>o3o=ss^0O^%I3ztafO?%UI4jJ<f;?Wue-j9x{<@HHJq^}mv}#}Lw5+u<OZ
z;-ENNM#x^twze&^WDav-nGG`r+K?;!Qt`=}-M6gCJZ;V6!`5=0wWjBBYwlgJ#`&%_
z&)-?|<&QPf>)4Rk+D80rHdqW3hH;d=m50oMd~I-_Xv2<4HvE%QL-aeNgm2CZ-vkKX
zY!kk@BYYDyL}oU_t+4#hiW{#ixg&g&HQAEQ!Z%OHTk=Hs=6rukYM5Kn%(kg`ADU7w
zd~?UNDUHfRNBmzs;=0$P(I)9TkWAkO!ZKx&rFUgyUGCY}#ksrux1V%8$;XW)Pj85s
zWWpF2^XiuowjYfscp-C_heo1_F=BO~k=*kbaU#M9FUiKczRsAS4aUT6GM4N#V}@yr
zBtO}ZO<9I$3Js}QwFXa3YGB*62Ju~L;FMPbGdJ-)3E$LGWu_zN*_Jv)4zD8`^_p19
z&)7IkpN39vl$U>BD_sx1R_6A3r+kfjs~mKEtGp=7Qf6tg6z#!T%8z<k%AYrH6vvBi
zl!u9Llv}!4%Eb|Hl;GO$l^uTXmHHpvD_T=OC{5;NE7r3=Dg(BCQZ{$YQ98MNRvJvm
zRXUY_RgO)`S0=9ft`uZ^SH5p7PzD<OP$KhwC?m8BmA@+sl}%nhm9*VO%7(LL%9Xrw
zWlon$<zCk+{D+okh_y*-rOi}(ZOOLMX8Cz-x_;KdVY?n*kLxqDs$>z)X~2^7`l7|J
zPwQ#*dDyuDT~{~YzZVU;cdsD@*=7XE{NkAK&F6onR7*A`L$VN0ey@j{YKl6-RA#NF
zTz+OMSqKejSlW=QgPZZEwmp_z?8$JpC+N662j7WqET9FQ+BW9k?8Z1BYC@nq58P-d
z9{z@C&uYk2;hc?Myx7!0{8GKW8RVfSJ+jgdx?CSu*{2O?FWH2<4XBoBK=>^`$?KlL
zZt;kBygyMgfG1&ccrstRPvOq&DJ<MO1;ZB7>6tNyz)J!2623_kzBy_>kG_ZJ@wSCO
zx0hGL?7bfKD)d+ppig2&Rnb3J$98^o@po6}PG3VZ&Kq(p-;e>;MjQ+=CAMKx+G-Z@
zyXFc)$BVAGMj-bG2GVa{AjdAOmfZ6-43PUat#jg^s23#quOKvY)-vekTK?s%<(p`Z
z3qOU>fAcyvZCy{ltNSHB`T(0(9Ax98gDh!xh-BfL!MzUSKO~xd{xS6b6+>{xQ{+E5
z&yU-e&}wm+k)IM-@ja2G--*(nmc;VENup)D!XDw9!RxQk{bw>(i&F3jx{hDR8(eq%
zK<5uR7+uUozvdSlv%gY&?3;KK@>uvak7-8IncXFyV#$U{oSx5X;hW*YH^cG@xnNO*
z+oK}Ue*R*4$6s_a|3#DPzlfGj?#CaBWiDL8%FV()ze?Fq^)DYy_0?Tp^wi^r^whE9
zm)iGUS1su#JE)a9>dlkdYA4AcUT>wP?vT9Xetj!(*jhoob`{KiRnGO%<;-qd&VQ!m
zoGdEi@ar;ipO+DvT1K<PGQR97<3Z6snfd)=Yv;ezx$%eWPo)eTTuT1T-+T&@T)%lm
zT+sZalkC{M>;GV#TOOKbxwISdnH6z4ENPa*+clq<^zS42%d(kTH5<>U_tgFMjvTjl
zoK1Mc`R|z=4-h@^(G2pU(<LuGoq_g}ePj8O(EVxjKKX+E&eA`rD?LrEpX0ez`q(|6
z3fDa5OtZ%fyP1lKODgN0JYwJE``nswi)O8Fab)fd5=?Fg16<?vniLB5CUY_63Y}jk
z@>#USQx_&M%qD?Jzb^B1&t>MczAUc+(f*1?#$v1NA*%|9NKePUj+ZE`xJavS7nyG?
zvz(dI<B>1jWOrqk^)-Q<V+o>Fljnx$ZK{7r#9(A1Zt`*K$q5Af7ms7*CHhvm#Gi+k
zsNXk^s@ib`rd?t{ozuAZ$8z2?mYa1?F!S9p=`%mZ)rJvN80};6=P-s93BQDdQg_E@
zo=P^%wx~_CJ-CUy{hM%<#|M2v30W1&ovOkc?>12$yaDHs4OHE`0l)pC>&@MP&9Ghk
zTC-bv0{4hEb02f!_Hp*(E@lgp%&jkc(^BRy?KK>&u3=*fnbUODpvn$xzOIJtG9Oww
zER@jAp_0QWGnX>SscER8k+4jKu*`9p`RsWg%HqGFd~T~@buSH0do=i56`y&g@YXkZ
z{Gg%Jec5*%ky((;itJPkqvnK)7JU;g0UOzwxq-=3Hn8vDdT!KM&()pl_?jBR7hjpd
zmIw3TkaRGN52owCwG7M%l8iKA8{Z)5)Rx^;ogkK^t)Xq?8k~o(;j3gw{M@sMN!tRr
z??0Cx8)lPyY8LHJ%;L;}S!@rSg`UX_I&7TIKgsQLzB!TK1rxZiQo0j_ZBAwi+syUl
zPAy-i?eGzQun*lHdy}xrn^3tY%wFzAJvT3o_VD6?mKTm`o`k*hWb02?G)KqNvCNU9
zVUEnM??GrscdCqbp{?*uuFP&+gm2QCIx}IJ6Juk?<Ir$CR}94qb;y;aq8m6~!<8yG
zoF&K2nR8X0X?{`k3yYmZ4=h|0Hl9s29qBdNkz@-;e*JbJSA5}zWj|#V<v`!G(TrU<
zngb0-)BW@)nUjr@=jlkMG#Wvld-nX2+0B`>;dHqty-yd1(m-n%`$S`-Bf6N7&$gU-
zBz+4$wtQ_No+;53p9{BP;V2tE*R$bfwlza8TT9NKHAh!kOO}o`7XH?(nr=<vJZmPb
zw&ucCYc3wMW{Z5@oBP%r%CKfyjx|jStSK$FW@f21=6|g@DJMowxSVOlt#B2-aTmVX
zFQ=9;j+b!Fkik|A>?eCF;hPD<H{R1MY3pq%UGtJDY}XR20WD=e)Qp;576gfQ_@ljO
zf7>?YX}LL9^34hBR*!Gxbr~SJzzM=KTZM1F4i`<aT9@_R#IJ2BJF3|=c~`S0H^Yos
zw!)aD)s5-?%!rA}Ml$y`B66(}RTmqvZmJQB#~I1{tPvVxBVx6TBp<B?p+z+~5obv0
zDMM!LGsJI&+?VUtKvSy*&6?C;W9J%}f31O$vk`~-%lE_xBbnitht<Kut_}m<*5;G!
zwU$O#<;x0v7S?*B)ZYGD$q0O{IGxW@p6tw0j=juM`mW1TdUeZEuD^JrIL&*bjJ0^9
z47vSUDQogZ>E8E^vbWlMCB^-{a^dZJ<<j^M%9)^S<yOo`<v__tWuRG(Qq<<NVlnHp
z@~OcWrMvGpMQizY<^F^3%E2k$m7$#q6yK3Ql$pDJD6=YmDBs=`%1lVOr_L8;=Z7!K
zY?FNDR_t%3dD>s)#;t$KZ{ePX`^uH=yDOBzhbxs=o3w~(r_ZsnT0Fd3Up(Rs*!8qN
z`E}~Eq-uS-MAXN8K?82@Z@`d;4Jo|RkaA%g>$Ya})HkEbkE+bfsY-R<syvuzKx{Kp
zH1kbq_tKQ(mS*ToGb1h3j9GQX_g23TIgf@gO8U**f*P`Ma6|U4Z9vGk5=CS1OWB@M
zpoGsUPz(wRl)wo;ls5)tinC3H!juZ7!LbVE{mDw@z))R|&DCe$6Mbs7GQec70j0D2
z*cR_cPVEV}kD0*udy;=sFbV(9lQHcvMRH=MNY};`j+jp+S9dCNKhKd_bO2hF;-OkS
zkCn2s(py)Jsju`{^F)sd`F;Gesz$ZD)p*yfI=*wO^YKr0+VwYN>r6vJei@=~XN2}_
zQ_i+(D!m+wm=U^&6O&hvt`$g!p0b~s8pzZ5)hubahI%vCFfnEg-+!+mV)<GE53Lp5
z*IHh@lSivy*$sw>hie^^ve!|3eLqb)9Ar!EK|bamWLk$qEZA^}nRO4-Zdf!mrpIv8
z;1qM(pQ3ow1)jW*V@8w991*_xlAp+_Ux}Q3mPF$}NepRog`rEY5PvV3RpU}*4=<fY
zTd(s>=L1uAe4tCwXErA0O21Yv$;ZFQ4*MIQN9XbBSRU;@=5bOO$JZ*K^Zxm0&(Eii
z@XdDNn{mGiY5BX5(sxBP{$9kw$|BKqO9#T#Vp0|s^K)>qWXlxu(ykb<(Z%$bQ%tw8
zV*K<<80%0%RfLObR#7|ktfDr4S1Ee;O1><sq)1aqx2Bcsa;ack(+XD0{N~^Aax7X(
zUuM;E>VGW5A)|~y;hSp0H(L|RFm6>w^{RjHEGcEP%#17#mQqFLNfqaQbFTVtsyr`d
z%g7?GE&NH2N1^N-@+G@A7v0*qjQa004`n}P*EC16wLeLR<44l0KVtLy1Bb%i^U3f%
zHo`Xr=iYMcekS;5@E|;c-@-SWLee?4>lGnwUovrL8ttQA@P90wcRben8~@E@wluUQ
z?R7fsx?T-i+e{*%MMFEMLS^1=w098=BBktERw6X)(X>}KW&N(-@B91X{kR?7=!pCC
z{9M;-h#Pa$MSC79jXCr8D)(sq`iyMpPjDLb5vyEs;WaxK>I*(#{i!VEA56zJ#dP+{
zr9w-BbBp_vQTq1{+P5V^*(8zu`tjHz#h$4HG2A^84cWG6)Z4{kj!Y~RoLD>F9F2Ih
zXf(W!!rQg1xhjf;Vn`&eT#1C|>PYlCM_?yE|E7B^*1YCyK(9EcTE?O7Bzv(=#X;&K
z@1=M?q?Hqk@rz^dr<OlYMI^E_A|N?599Nl@$o4gUOne2g@|RfiHH33F*{e6enlEu*
z`06~s&5q0X*mVgvkNIFuu{Rd8&#3#I7iZafVe<qpP_qyoaY7tiDMU=MCuYv`gya@a
ztZx>ep-q4d`#kaMy(cce^2Ac}TbONg3$Z4*5UzO()_<;YCgUX>?(o5MQP%k?u-<0$
zB|IzlL0=5_c&GZHm}f%yQ+-g#U)%fwcP21j>G3S*>J=Xto%F_FeSTgI&tQ1Rl*?M6
z(NR9w^wbA;9{Rw5^*|%ea;}Z552n24ufe;kgvd+ybmJ1fpSXmPgD#=S*a!L&-muvs
z#I_Vq?4Rk0_lE@Vv-d!MKhIuW+_7P<JC?Gp_<_(3wX5B*GQ|~x-nn40lM7Dh@?L6+
z3tq{&@Ltmy`|dgey3P=MaKhK6_Gn*y951Yov2K~O*qLu02HN3_A7}TSw}a^^JN$Mz
zh&lgpN9NW&nD%Ek-XG=N)GTY%<XAz#o+$IboNel1iJWnkc%QTjfi4!TxwgPk9Sc~k
zw?Nt$3*`OUiHi$O@$Tz3$m?xG_XcC=4>rauPa}*rFv7==ylb-Dfv=;tW9ugq%sa=O
zm=CsL)JtQC?=ptzAY)94V%_C#Bdp+^Q&fu~M8XWw#973!HN?V2h8V%xf}gU6xHXpN
z?}Lqy${Nba{)QM@Z;1KjhTLys$ge4exZSb^8}hc`oy1nSIc!05&t`OqZANd-Mkq&Z
z<et_IP}N({IxJm;`Rd~PLM?QS;O?mD>~s2uyECV7x6~Ld$W`*pB|sC!7Me&_)<kNJ
z26`iTHY4O&jg1E0ZqdLyXyDQ!4LJUzfk*t9xJm=++PtSS)WBdX4V*cw0r~$lu<3#Z
z|J(EPJ$awyrGZi28YtmMD?Y>djA6c^$Z6QYY~#yEn)#++-E_3BosP4t%Q0`^JYvVG
zcy)X#6b|s&#5sM;H>U@yplgQ;>c^;H@kkYD3|E1D#1u?XorDEV6S3{rL^R2b#P|k9
z_%h#|HC4oK=9^OH8&&3;3o6Vw3%Ex+M1fgU0i(|<;L|<@oUh|K3-gT$^GzG`O}f22
zyqIsc>B~cWojh`e$U~p`rlnmDi!x<#@0BbJ{A5{gA&ZwfvUs~!7Uptth#o74aew6S
z?=X1`TP}|^%r}!r9zHz3IlWdM6HW|=m;`HaLWZF^Kmjj@55`B6K}eWWMi1VWQg~=7
zbssFHK$BA1vc8mpjLT`{@N!zcy_`xr%1FzqjB+QG(Zea_6x8vZPHg-^M$9+Qw*I6^
z8-7vy;tDdCtt9*YO7=%p(Z*f*<StoDrS-M6Z=jAIY-^y?9~)`Zu_hA#@SE&rHdFfH
zW=b92NU<F?w0Lz5@l~0!23OE$i3-{#UqP0qD=7JR6&*8aqXFMGiv7e|ox5U4XEysU
zZzLRzM#0B-6lW2RMrqGz^l6X6*Hc3=a`;e$?v&$siX3>?z`1>_MQ)P9l13?L+e_i6
ztrU8;OJQOCAc#gwV1~O0j+yn-qnW)_XWB#d>?70cSH$bjtlQZ#7U%cyu8U_h`Obf7
z=Y)2032CM(x!+VW^EWx(X`;-`CYosYo6P%LNmH(!k}v(C@+UpiBOIVTY7%hS!+i5*
z5caUXPGhh&j`2NBlZ`dj#ag4g%o@$xL^vNu1ezY)v3^qo1NTH=lP!W7?IIZQV?TWQ
z_amxzKSr=`{NfoKC}cCsWJ==fc}eaI8N>-LgHXkMQ<5r;ig_}4*DZrT%r~cdWYB&{
z7XBOLV9ql^bK_BXv|<wU|2++vv-aqeb>v)9M=YM@1nWIca1C*Sh_W+mY@Ipp%NYeF
z&bWNpl|5Uo=!;;s`QeJ?s%{uv?hac64`f{Q!-98yoGamvuo{2(%LZV<zX8ZN7Jx<B
z0nl6i2&Ok4q3_aj$nyL~q9+{1+ap;29Rtt3u~=~>mi2V;oVS&Pm(!9Eb?^;NdcWZg
zz_$pgNk)ZP3f>g{fX}|4hzhR8^p!PuUQ&%`Wz~4MtQN=HYO!fr9X>9sgR@~BF7B+u
zRLurRaX*v!iALBvHlnS$1=7qnlbCNlGv6F;Y{RI|HVoR)4oCBL=xDVgmigvJAM0KI
zwBes$ZMg8I4F@gS;hV(solzY)Sk3bt{Z1IZ>BO?PJ?x_!fT;Qaf^`RQMR9;T4EoWi
z(2v@*KGy2>!D&Jt&UE)eH>;PupuLFy(u=SUy%3$<i+z!Qp<K~|=*9JrTTstk9rfJd
zRELN?b*ML}gEDJuqB81`IlK;g-_)YhtQHQJYp`3i1~G=!2oJ5o0@W%kv#NwxLj|;L
zD{%bJPh1TAfs^t-m^Z#d%(o2FAD1Few*<`{>_KC`S>sWRi8Y0IdAR@%vkH(K%DQ4B
z&QGk*!$XriJPG}R)WXlWEW!PnDY;Nsk%JSvvw5fTfisb^u>3OXiwCFUTYW12HYVfv
z-8VS1C=vbD3Apku0WZViu-hjV5&AJWe<vERSwk%R9)+2b(b!-fgNB4?yq_M8%N|kO
z>lcalC%KE#FcKCz5%~Hy0<S0nIs-Adp&5%etl@EIuBqC{`892^C<%<k`m7kVH^p!-
zOAPKWh(?`%Bw}kLVEZ}(qcy`}>>7q3g%_B0`8kZVf{`#X2pbkXg~iE-_|@Qxu~)9L
zul@=&dA4$ElMj0Gy`jc@b0xtG9;>|YaF`byUJ0>bkdSlCJ@IFRCw3GIkoQJ_nF#_Y
z^W&Ke0n$nZ&=mE=T4K&|^u)o*Lfk(uM9>Z|=&#|dnLKaUzw|+8w+|dDe6aeX53+_c
zlSui%e3uU%ba><YD{rh>?2WWp-0Lmx&1~d_!SlJ-g!g0fIkU#{qc<GJ`oPWJ2Nq|1
z(0zhG-(eq&8s>wpkv<U7^TBK8ovGriUuLFK*>DNb$-L(h^?{X(5UF#7nEIIerG|RK
z*NOSY+5;0BSvP0KxmFEsn8&**i}h|usC30_A!nEexuA$=H-dRC&>YJ1T1oD733SGK
z_E6dWx`>b3_M9zn8ZH-3U{mT5j1xVEPEYO&X13YmV}~~88-w)+VWne>+IhB|`M(Db
z#rW7-^DNLB>8Vz5HnPICI!oS}St3-y682AaVTiW{j8|FUrltjksafD@wkafeccgUL
z1fIUz5V4ZA!pt}6QpVitWQ4hzM(jc5euYiki5y{yWfH8zQQZ!mq1$lgsxe#*jWM&y
z2o4X7;ID0jkl{vHm1>B?Jd3HiVTf-R46$sEAspll(J0NG;DZg(Fu@SZnR)j67$Sk$
z=2ng&KEE;KOrtH>)3gc4q&Gubm%B3uZ^pWqjrjO&1128YfH4j0G4%0zxGM1R(}lab
z4kpX6#+Nm~YsIv1c|a4xns|p)sEMnAtg|_w3AK5e_{Mr0#A+ah8AZ~Zd1i$M^d@TH
zgf!1*depI@T^+Yu)nV76&R#Kf-kGr;N0$G5ga&GqH1L$4drw&dzo%$$#-;`q@=@gf
z9x+)1=O<}kncq|-{+@>Rmz*<vZW`q6r@>!h8kP-F#To@w95S1RNv6~A?8H<A8LOgU
zqbg@8si44Gg>#!!@NJk1lq{4XG*gC?hzjh&Ij@iJiESz-BD`cG{JMB2mB-yqoUyn1
zHTONORmAQU+>Nk|`@0t_;zqv$#F=l}{!_rbL+s-=Qh-B>JZ1;TBhXDAjm$UWZRL5t
zA&-?a<Wa)Bk~tuUDQR+$V!oLfB8#h;GEmErL1CW^u8ft%l|gbS9w~=G%s2lGmB)&a
ztjS@%Y2?$(=PA!{RGF3fwB^}*ISk`34#Tt43f!?Ig>B;o;XrIDHD4*EgVv?wrdLYZ
zTBVe!R!ZmoDJ6;hWpuQ#l<e9{>2Yf*<uK#Kttuz`;4+FR`$6-z{G`p2zv%t63c4mz
zN$<W_Qh`AZsb9$@+1<JHq^Oo+57*HXtp>`x)JPYde$dymKj>%nclufOo74_A&}Hiy
zI>p}dPO2b3<{K693OdM-*NrQv;bR3geyt?=h-%uou8$^;9H0wUBQa-!BGOlkf=$3E
z(1X#~8#@Z$k8<zT+Bejdnn;`XCDQT{i8OvwA(csfp%+#k>D;0mx;-_A{Kw{yyjBkV
zXv-lN*(zFip^B=8wNp=5H@OXTlhOGedQsX#4QqSp<EUQRWYa^j_1(0$tb+pmev{Vu
z-!#dWnZ}BDZU<Rg6x&3S=D!JTtt9ogjp}SW$V{z=1k?Ly1M|)7HG^=Tch4CKR#0NT
z5npJHX*;d4;yKT5eMOM1Ge9a&2B>tU2m&C2PUal3=OUavA%f+lB50~-<`L&Uu47{O
z^-coG3X*uUSrT^tNy7HJB!ZaVjHXE=^Q|-#r^{e~`9|cg3@$O>ys?u-pS~Q1@Xysd
z{(eQNO+usMX|z8%4ZUIqjLdaJ>SQO(9_fVn%r|fAozQ3LjIjI8IGo`Ovuo@Z^>xMQ
z5Leu7cSS4j?|-zrqo4dZN1lD*t$rBT?vH!*tSwdu!2BfvSh72Sduao({9ORP9DamH
zUmwBd(Q}BOVcqAX2<E8>oLLwFKfc%3^@>66w?sHLBx2xiB1kp~6DA}<&|Qq3)#X@p
zrUK_Bs<1w%8gZYiF^bRboz<x0YyG7y)kx#B#i$xD%&I|etMNdohG&|!7?|`MYKwlu
z^=vE3gW9-Hy$!M7+VHob4UYY7*l60$JKT0m*K9|qNIO>bwqZH*&5yP=6qUCjHLDGA
zgL$qqtOGlac0f<76A$7$u`;a(@4EUCH*x?IMh#%^!G4@v+0X32x|`@eY*O#TeZ@W$
zSNCFFDr;>X_M+Op7aQmI!Zi6Wx~f~yda)j#YwFndREL}2>(KYH7GLhva+hH(qDR)^
z_Rv}k8eWU5-nIB=X)R`5t3gU<HB9uX;Tu?m7vou9Y*L921r_Mu{tJ1XKOubm121KM
z;GFSySllm#aBeB0R+nJ-$6|P1D@N9gB7Bl6#L=q-SURTwzOVA(wmly+>hh4fEf1O@
zU(l2G87{X!;r>7_#OCH8d0#f-N;9CNnu@-bx5$5(gl*dru$%o*D}OO>{7gXF`*`fp
zj7RaVILO$>a;JR^E?kVp9@hI7e~H48W9*NLj6&m;Xv}DfLWfoqq#s0bwr>RFS4W^G
zo_ChQa9pVg$It6Au#o4^Z59W+y>ZxxI6SF{MVv@1<R3+2!jDKK8Am{39(Pwpy~5hY
zmxy2a0vme5;5RW08daeXhOk$vFA)EjJw_59VVLA?)~8*8zBl(Wv3{j$?Iny_?SqG3
zm@Ry~apsj5UN7}Rd9@JpZG_y3<ca52p4iU5qY>ExbiNcY9|~~&g#dQZ0?f!4psrAW
zWgi4+JK)K>2~VU<6ry*k7Zm>W#<(<Z7~J#0fe0U*De=Mf`93gX?peh8o}y9S2;RZj
zdq=#nfIHigWV|uxgcsh$d7+*C-0M5Npt{=|O6=ike$Lr^mELHo^2Uar-q@+^1Bp6s
z=<?}W?8EusKHOdIgZZ7@!@w+67VCw1XLygc+Y<*Sd%~Y}#oe6(=<gFiMBn58<&mqZ
zxnpHHck3Q=L%<R@oXc{>Cv#WK6u3Z=nPz{oGydWGn^py9ob9p4Xw%c^{&O02A5TNB
z_7wWI9Yrzw!uJ>1p=#MdoE~)mjyE`i_!Rd@S=l18n>(b;_Tjwd9uy4b{PqLZ_%Y5J
zNVLLA4J#CWv4q?yOI#7Rg!{E!aE{<yjVAVe-`|07JN9fxY)6#BcIY2BK_Ba5{FV7Q
z8zZvK2wCUY%h72FDHlV$(%1o&6jLZqGllbFLv$H%59Tp__%GMz?sR>)I`OV)f)Q4J
zGvs_PL(cy;#B845Tx7m^J<AX$nQiup8scN00hTBka<8HRI-Cqq&Cm7AGDM;IX56aa
z?(eU<kP6X3-60+L-Pgea|Mh4I*??cYyoY+Z9y<E#;gG6}P<PhcsOoSo9_xK;H8J+9
zCagbd;%$s3=j8HE>X0UD8CWCSqk*Asc_t&!z$_CDq|DXe?0*fMty9N`&+4#$s}9LX
zb%ec8hs$$yScIx$Xc)ggoZlCv&RwMHFo{#gE<Q&5wp+A1dv4UB!+##~N*$m1xmvzc
z(Zy%MI#oPfs0uMzRs5Dw#V=+XGdZ5s3|EEsK~<b^R)yCwRpcnDK<%gs&yQ3v*-IIE
zyOp74&V86WlwoMBj2<OrTuPmc2hkHSsAwE=A1lFSgAz<vD`C}AC9F|XMCWpz<1pWJ
zEM?x|v%hNy=YS|+HuFuC7VnD|DnNRo0)Bm!$L(kG`0OqZEMfi4bmo!~^3cteL&Rt~
z+^&^H^ix^Z9?8HqTZT2@GDz)};d!DgE=$Rwfwef3n02;^%i(LaEE<2xVlwM=q}pYn
z!ZVyOW}GkU<*{YyFqA!4!2Z2b7&&GTciffI%pIjvzO9t3HkQ(_wWTzCSt%*1lv4fZ
zQu2CPN?(tZ(tC?ia;Pk&7e7jAb!iFxT~I{dp6AjOW`$*Yv&n3BHo+vDhEL9+Q>$_*
zPc4@Y3UX<MRRtY=R6z@+c-M8UoRkal$VT%M1&+z3jb*j8prnf445_A8_F{clUO|#g
zKWXKmUsScZf}X#xAY(C};T)2Lg#-KGAIQVTcnE%a4S~*CMHsfRrY2<+Is`KC|By(9
zixcRFasurinLyh031m^7K##Zb`xVn@tW7$3{K%qr-5+SadJYvi=1{AD4*AW@CE?>*
z5}&}@psl|tbZHYUtZyVcyGGLA)<|Om8mPdtf%aW%AnE9O3RuxVVv8E7ZBP^G>~5la
zSL;bep@BR*TiC|cM$NlAX!43qdY0WwD$F<a;|C#S&>)o9TXBCfXWz`T#<>mF&=V0w
z{-^=6Svf!}9uClb=9{z3H*Rhs@O~<ScW*=x%Y5U(d@~2)7$_3wj1viT%SfVcy(BtM
zN^-`5B=kZh@lTi(EUrkSvs{|HXk^gOCzSc-qKGU8uE}BfMS1MpGaRz(C&A$FDLC4m
z#-X>T;2z=#^L|IX8qU29cX`iV?F2pcOI*J04A~@SoWJf0+54^-!HjdO%MD^j-La|F
z9a>BMFzkXKBwGA9f5IQb8vJ2!GXR@!FykB!z)^m*iDxa&&j8$!c!cDKfrv1DhHG*m
z@Lj<9_*x<O{5k|f%R|tylk?ZkKgW>=&+)M1IbO?#LL@m9erg%me<%lXdU+W6IS-2_
z<m1Gle9n8w$1d>#yt6I9+>8R`%`d{MUqw(fFXR5%GA!F#j#r`OFueE!|8d^$e<_Xd
zt!YBE*l&!Tz&UwK*>An|H{|@<Fmrx8v_*MG)x$HJ?lxrdd?$?gW+L;=9cCdVxenY3
z@8IibC&o4OVSyrN*Suw|&F_9p^6G~v&u;FF>_=&7ALr)wp?Y;6Gg==!^LzO<fps@_
zy_go!gABVKY*J{2sdNLL)YYMa`9_ubCYSlfF|!W)-qvzQO)cIq-%Mq`nJZn3ogdjN
zmQaIln$<|WP=&Eet6*$gg)zS>5o}P2`kfV=BlHX6FMdKg_y_!de8=vla@;iF+#A*w
zpITIcZ04I?pNjZ<TZpF$g}4w@fYvPqhzRHJ<&Jzbr{-}_d>%w6bN^(-XYN${1eF^(
zIC?n~A|~&!=vO+=ak%>+o;}(x5|Q*Wntenu@GXnMhO`*;sKsO9pm@HX#Nne(EHY-s
zAnZsqhO$@t-P<V4FpYv_LnQ0Sqw$HgS00j45Ydjr<?|6ZzcU;<|Ar&w%qxh`4#Rx+
za8N}IRJu5kZUcM9Rk-6JEf(`TqH!-X3OD)V1PS4oKJ_(TR=mW$=fO~V7=$P3oFn<<
z5xSKFpmF9Q`_}z9v-b|%)p&oj@+$P1Z<?1}#_3DEo2q7?6!#XaG-q~b@WS;cURZY6
z3$Jp8@G%#{I?NM?%stV(R)EXwJCb<7`jtBZW-<ZZJr-aD^G*@Zam2C&XczIsVJ}bE
zDhRRAT8OKEgeV&Ajp)ff2;lCgs54$@PV|E1AaB%;^o9qY_F>$0$zL;B-wTh;yf_cR
z3ueceYc8<{DANnOHN8=No@YDfym2nxoBQv)@g>R|EmwJ;mcSp+vmaG`_OJ`Q@z6qu
ze`X7z_rViUcRjIqyC-H%@<i%)0oL=(NlQxrLw65I&G5kQ2k!VVoqun6H??mUGtLiJ
z#6`NI3a-Fj7mzcv%><s+6gXk(7<;VgI>BAvcI>0JL;hRlnFu?i&pL{f-A8aH+75Tu
z9KdBmTWq|_^O{4p@Lp?+TW|J3V#8iU80>+Z^d6r3S!3lGYfO%|!o_Y&xbf`ffj0ZT
zi*{kL!Y+*M*@>kv!%02%ZSOV37tUq1?J$AcHWR$L#j_jPZSb-;#<Lb96mK$uEANp;
zpJ4ws&t0zGHNdhR2FRJh{ZQBSxdTxjZIkqI_8)z;KGx@KMMKQ}Zitl$hS<b>^B?n#
z-g(vx&tzSVgdxs%7~nyz9$vO@Mfwgs-tBJ%%GYDlR_=}Zql=BVbx{_igK#$;?AWV=
z6z*A=oTh`BJg-s8VlA=tde-x;$B<B6_U-84<OA+qXl9-7C)WJNYC@j(QMcLiRAI$_
zZ)TfR5l#Gg$86)Hfv?OqCuV3svsWF3`RcHaP)FSzbu4gIN9sv+^x3L&uQb2ULLHCI
z)uC&q&iW*ET-eV1vP~VPhU)maRUOkeG2`f}W3QGvCTpmpQ%xP3%r^CW9`YO~8$6p?
zpo*};s<2|tS-@vKv(7D1RVYPIh5qcRSnsX^se>x`y;=p*_S~JYUm2^JZxWbqrZL~-
zFyCDNF&Uc7H&un>kjH%UfcfSM^NktvjXd+sIOdy|%r|ZSpKneu-!#}OK+Z$~OEqMn
z^;!z4F;Xbv^Nmj>pWYlP#N<g~Cv(fT`EuAXN)DD8vYf9a1BW{@ki0B|`w22wn=S)|
zFEU8XmBDJB>B#WoIX;<u_UFl9X@v}yFym;l24f!2ar9Yd`e5-exHI3xDM;Zg&t#&m
zGS}#o(#I{Ol+Ux9Ql8tSjxD9+;ia@xu9VCl6jRTf61urJpXO$NA)_N-=$6PAQgHc5
zhu-Fr5#`WW!)z)VmrafHv+4E5Y??7Ela?pGr>ySx^ui&Ne&5U_MfWUPl$S*rx3kE1
z=?B_#h#5nWLs!Fp(3aO_bmw~s`HGj&nvx=l`0$f<YW$*>H>~SfAPL=_@;JtPGw&4h
z;A;g)x(`9(Y4)XE9fcqr8Kmq>qIvTY=+xW<+RAJ*RWgB+Cnb;-vyQ7`BAq&tNLypl
zD2Mqb1Ru!lQ8tZg&ZZ9`IdtrI4k<}~q@(_I^q4(gJ0zP(w5X9@o3p>&I-maZ=9B%k
z0y?ZyNIM1>lP&M@mR|lwZ)1Lu*BSPT?W&`iUG?O_?6a9UXQL@+IY0kRhum7oZ&V+p
zxk<piRT4^vC1Kym9(N-#9LyBO;&=Ttf1sbP&mExn;{&uldw>F`ia_L$2+I9Lke4I^
z!2r*3{;<Y+H+QM~h-1rE33x?IAZ>#r&jomXbBi_1K|IUZC<T!l(il=Ajp_<%{ARw1
zWWLF;mqol3>(`l;ldOhf?aN^(2$+QN%r<j0PeC>0B<9sRKz4`|Zq9Xr#wRD7T7Mop
z{4O9e&Kc><H&^buvZmV=rW;+cVW&HaL);;{#}i8D{m{|iha^9L9Pwb!RA2xK9t1$;
zZUFZq2f#Qi015Iz2$&m$iTXi^I2pwH=MZeZ6oQeyA-EJBf^5a-xNG&CwcyYBI`$m;
zp2?8Cmy8q2DOmjSBjyZzVGm9o%1ZJuM<E}Dhw~8@pU>P@fCn)JP&F<@e<NSJPnPgK
zNjYXFmqSVXEB0%DMS=HMe17r+H`92|QCW@Dxiy%;d!U8)>rl439sz32NQ-ZU<n(r)
z0kq*o2WyadKeZ>b4XTlCIL?nx#X2Ai>OjP=J}lb9dnm_#Y<}F2Yj^r_`EWlruI<M@
zxqiI(*vEYVeHb>U5BoZM;ryl-=0UyKaj+LDuenR}U=JoT-&D6ZU|Dk=#B2Y@H;b5W
zK7Xj=`}JDxw5Y{f=9>oAPuC5uMR9HoLYZ&=vS(~R>#EgSD{&*b5*7B9@ZDaC9a<Im
z<Maywem}9-^9Q<=zM}P6DHOCyq5HK2ak3@&ytWwGK1Fc$DCC_@AxfSX;Iwf8JR|b)
zuXH|l-sItyM;@A4W85G78R5>KSfBkKoBw-<q0Q;sMU)7=m_#Hn+pK<`h}Vy}>wP5W
z*X)bIili7g_s5|-Ee;XYvB;Ydi-zQAXj(?YYjiZ4UPgfx5O|ptiHhBksCJ9O!punI
zw?^QkYy@ZHy~c*rS3F+`<NeJ`%*P8<aIWyWgU>kcD-b8QJ%O0xBdmDmk3jZbYaaH+
zjDq{HGQ5Z4X?HmrnS0NEJ>~1h6D-YpgehwRV6x{XXRci1pP{Qr&bkVrJa-#Ryn<oO
zH|jo@aH`A)#=QI4b%19ld0xo8=f!gyFHA_|osPE<{vn<)GhpA*JZ74a0*n&~aNkXU
zI8SDU%L2R!5@7#30S>+wAiP6>KRho94EKbUx)AeBgm{%Kg!f=Cw9fa!PXjOHgn8k1
zuNUX6@ZL$$oApTEIL!JO1MW%qa94<XTfHD<?uFINSi`P);V#dU0+@-;v99_1T5rr{
zKi8`(%sLL<>~-`;haLC4P4+^ZP{`LdA)d(zk<XeS%gekoW37-p&z;_73J~ET!2Ev&
zID60osk~2KamgJ4!`(3^iuY3+-B8K1n=}<yjKAlCuTn0!tLco}H5b^A>x6MLPou*8
zINlW4agV7T9wgX7`K29v?%6?e;Ssd#IE34;ZE^jB4N7HfaM^S}c0Jh#ONo7aKfjkV
zjrZV+>>ivtyPLD)dH??03ZK4PLM_h%vK3~SX<!DU6`ZG<Zps<-rbyM<ft8k~*zdg^
zE54gRc9jVXFKol7Z=Bh>k+nCUdFP~M1eHQVXqp+KsLKGyj~YOq4Dc&fAI{hG(RETE
zdq(PG{$S3nDc9p(bbVxH>7nDgA+|Xh;^_fH{4h4e8lKk#6>f!6?H2fAEBwE1h4$L5
zSWwS9rczzbcGAVVXS$pzq6@nxI%sv!!K3Z$OV#791)kwZebdFM`|IJYxgH^Px@chT
zQDQInf%EK-`pKEY(VDn%gLO71c`s$4iGsPBNF30Bf0hRPu4%x@Oao@jHg7xC@#4KY
z&v(@^%~2iAyVSW$O&z{V)ZsIo8E2e2lt!qt$4?!KgVf<Aq7JzpHJEg$@x7HA&ef{n
z?JqT+F{yFxuNuq?)li<V1_@@HVm^<VZ(NA`xcjC;sedZ=h;t97lqx1M>#QxDimbg;
z(G;M99T!!g$GV%7ZpxeirwlnO&hp!-i~;u8w60NxcjIKtN}Pb2igBEiri8V3mGDGM
z3G0||+*xz;YK$VX$19?dwY`u39fmcbL*e1UzNlT?jj1LJnP*b)DHg+rE>YO^iDDNY
zJwAupM3K@iio%*cvbrUL#m3SY%-64JH)Qz!LI%C3W$?vW2I_BR5W)MZ?O8Io`#}cz
zd?shhz=io{1M^M9FB$li%VHh#4F{c}PfZ@n9}LC#emRsT4}$fz|M5%?&uj+Pl#(8m
z(y}?F6xLZn5n`oOvZR>4Uo9fdqXo2oNj|Oo_=VOU{zBTFpJ{@}N6L@NrN;?5<hnGQ
zf<Am83%3t6b#xYe`177p#oyD3>=b$(nMAFxlBl*gi7pO(L+32ANZCD;>g01sdSnin
zoGzedqcR%&vxFvf71O$%#gsIui0&2qpls%wxAP=%hxx|jzC1phRDg1r0_zcnK;J+S
z(@jP}>tGU<1;mrdkOYcwi>1TOvGlVko`$r?(^i=T>inBP#kz@f@4rMk_dJX27k{9i
zkF#k)bv9)^&LI}SQO%iLYWeb!?yjgOubcI>U06d|(RnmXJD*;(<&)H<0@7buNdAk8
z$v3f>+TzNo<7+vc68%aFpKB?=q@E^(*OR|Y11T`yYz+BL2gh>8vsMd5Nw$(<bsI;U
zbkfzwJ@iVumu%B|sm8sJd?xpk$(4TEe0YEixm(m{iU{0oMIe7)1P{wapjRP^hs|QR
zVj+$QPjP(Wc~C;T1RR-f0?#npFyFinmPExeDb8M%Mr)8X)>TTQuTvUlN6Fwx6z`?d
z<gt$V=8wTp6nz=WxlO~-T{;mZMkmoe|0I?dIKUym0WQ;=Fr8R$?BIwSzs@1e#sS~8
z9bwnwi2RcmpgN9qIm=wpdD|VIq&;!M!4HMievk|B$DV5e*dqQ2b6z}wchXZ#ZGMU)
zs+?V18Hg7OLD;u02*Qg&_}4QC)iXj6!!sP0r6JfiI2bfN7;7zpk&}{ys{ACLg(PFm
zxD?zC{s`~>FDTN^gI;MKoMrQ&d4xL!p5?ROJRjE*3UJh@5F&MjSbeI5a~R8^!hG{s
z>?<}h-^lxXg~8JwSe#ygnN`(rSx|!^yX(;Vpbj!Mbr@S%50&lBxbeLi2OhMs|F;dD
zd}>155O2et^1wE}$7+Z4%??~;zRBI&&t8#!*xq9=`1O8x+V<n-bnb*`?!(__eHdib
zheOl)kXP3W4Q3n9o4v4k`Um%jedpu;!1D25kW4Flxf}Ce{`zXnHW`I=Xn0i%s~NR8
zC|3jJ?rL2AR?VH>tTjsK!+bM4rUv%RH`AIcAsb$a>Qj{%&3rR)Tm^2O_=OiP{QvsA
z4DL#0=z3BL$u*_0OkrPmS2514<g7$t5pG^8<Xv_l?|BL^lKJM4PCoKjTO5{{hu>a#
z2%eh<_4Lo^o|21@CmB%rl+L@4L|DYJ=9~G(Dk>2M!K@=@zKQOMf*<Q_#8aYieQ6wY
zT;mXz5DQ4gLi<%T{%(lIK4BE*Y>UF|nNgfS5{dPCk+`K8iIgpoC<%(d`Ul~x;dqTb
zdaoeY`4Z|6U*J*yb3ADehN;{$G|mWwhSn3FF+akaJN|h3i)S)?MY`|L9U+SM@F(aF
z)~&sb-qM>GmU{z^J=c+`dmTZ~uc1KV8d{vLLT2C!>=s=?eb8mhwz!P08JE#;oi)CL
ze6TE$H83e&yj$~v%YYCw6NEV7A;hkSo~Tyy<g7#wELi9P-xLqX91~!gy#T@I1lSrN
z!0L1XeuWFrdryEnbpk9n;R(fILhRfi#F|7Q&PedwMwRtEBfZd{?+JHiqK$_Hkn$2h
zk2NvYzdW(QP6)emLhc?EVmALBoOQi0#Ka3DA9*1q)C*Vqyl`@iH>7!ARe#hQl~$}5
z;*&drGkQJ=S<@@TC}ye*DItbrc%s1FlfAv{@#39SP?!LsX9PGso4?*x4_Li-$D==P
z*x%^J-V!%FBR4z^=d7DSuF&>$fmFFON(wI`Y}y5Ex$J}<DSH$RK8B^tH`eKPsEJ_@
z_%l16>Di&f(hiOn4k3>BQt#qzuyn8utaj{&d%!-}_Uy%H>%Ht#--8zAJy3tn*%x1}
z@zB=_SKnL0Bgg`g-^`#r+6+|-cfdKy6eHTU^PG@-GW|_p)nEeiI1_A_G2#8>Hne6K
zW8Gq7yf|jW{Z2;s5y4tvbwliWYJdaQ29TU*fDcjnICMoHIaBp9UPd2MMS2*Up@&`f
z^&q}ok2zWoDgo^AUc8lc_*?KJcMF!i+rk-s?1u{31nYAf@u^Z5El+h}a8Vb>F6m+)
z&u(5G)rI;W9dz>SCSFH}d(yZkbFL1240Z5ylP-35>A?ItGmowgz9nnn_9X7Z<k?Nw
zeNF5?s)<%zP3%?C1pU##)=Uk^-PFMPT^blRTZ4bE)G;?z9TRS-<IVwf{Nnk`A7+{t
z%rYU&FDHMh!S1sfey6HIEk+HQp=ywPtOmckYB+I44FV4}$emZi8hbUIIjV*c2i0J(
zpFeJ|8g|{C3ac3I#LS)wW9FE`?x`^OI~5~)cwRF|6$yOwa;IY6{;ACDDwyf4f<??X
zJ!h0L{FpL=EtRo{`DUrQGNRWi!^UzV?!2A=i2^0Oe4vDaO-h^(r^Fq(iddnf2+ACe
z2@S*W{PHkJtsVxihcev#!TTjA3Cudfp58(+{#X$xcd$<=@Gm{^?xJFwF1o-+-M))7
zZ2nMgWe?3`-KmO^G`8Q7hS?UL)9~joIwynHI2m+h$e`-I430C~JmC|}tRu?IbAkEh
zYmO{b>t!J(Cyxf6-`u-36jQq8knvy;hU%5mUXM~5F|Cv~%qpcsW}7K1OKJMGBHC73
zM9Za%>G#z_%3M)I4XOpSTrQu^ZO9|j^Iyn+-e*#{{*k@}=F)9p4m}@}P4Q1Z(2y-3
zsKPXhUPryBkCN$Br}BpS9wyPNn@Kd}O%kCeiLTCnLt>@xNkKc49y4RS{<nb4dFN&O
zvV=xt7SoNb#Z>>JkV3{5(7qMl=-I3$I=Dy@8h7Opp2xX95(=28sR&usQJ7JkNX{nl
zWcWRnqL0UtgL5o3y^f{2s(8BfJDxPe6G$kUKq`Y1$i*;`d@K^_T3Z&a`<X>LDH)Vd
znn6c4X3|7voaKHW$#wi^YIprY!L#!y=2;&3F3u;>ihTOwUO;PR^Y>y-F`c+yOm56K
zwahn0%s16ZwX}D8JsnA_rziFGw6mm<cD!$*-EW)e<DwQCcAfcVUK>TqbWm|v7bULl
zC9l`LwAr?gO69nZ!mXeFNDt7_g9BtCA%gcZocrS_iti)D(BmP7FiCOLn~Gz9yf`MV
zkbtPDB);iN;_X>UEWE>=T)PJ0=wvBq-Ij)ZpfvYb^3TQ;IfRMIW9?&kDC;Re=?(Xw
zY7fQj`^>h6!+DR%IZLS%5cT;ul)R6l&cgu#lbo<K#t}<jokPi~a|p9Phxqz)n9lpF
ztJ+R@AafB@9y#;9iYulJ_QXx*n|!|R><;pWIPaq#%RPeO%O@BX{}j=`pJMmSKv?|<
zM5kO3;?#l=c|8cbzXu^>&NH|k<L>6LXZTq0405W$_+}XlG3J}=%r{d?l5wmu8KLV_
zP;@Yr_tmL5^f(od5>qjvAQi(|<9fF*73-4H(4U<K8d?VJ)-ue~F2_W(ava=Wj)pDe
zsMaax{E;GTURs1Ri;J-9pCZox;0)x4buj*0hf~8F@HzN5{9iWXOG7h8*|*@r<TgxL
z(gtM<_F4(r@UW{LEz>%nqtOA){&xJA*$&lL9jwh}zDelA7qdQ`Xz7J5=iMX><-OFz
zKFCD&;@qWPR1Nvd-kd+ktm;O1em8Q*{6TS0H>A_rxu>)ZGbEZ=H{68&H+9(jxfVa_
zYp^o71{ML;+=KhStQ!sQDs<LX;#X!RPQ0mvcW@=jHdP{Ud<8B!eMRt_a@3%lbN#q)
zYIGUA{7dn6d?`YrOE9vw82e`xBYjyByssBRSF8}f9~Q8Nr2xrG^D(|H4{~pLHpYB2
zFh7qu@iTljrJ?Qx=Mzs(f^~f&YZ4N%JvI>=nQzRVB;s2@BASEQyYw&`8sX8%`V$LJ
zojAxV$71ib7)*$YM(XTnJacBY;e4B|qoWXaERr?#k$5~P5(RbPxHBaJ`zRbsR9_>i
zIt+_%zQm5DFW_?TIq&|0f!{%hniPl$>z?4K*&|pA{ITz|A6C!z#l_3_;n{N+P7Zf)
zNbxq-q~Byly8-v6>+BW1j>8YHVS?B-_C#LA-i|A<-*g3&o?OOU^UDxU=S*PM<qkA@
zBR<9(Tf)5%w48M?yjQvvDa1SGjhdUDcrxD;9kaPxd#neNpLw9zjCI9(Sc7vwfMXT{
z_TURJ`>23BkOgQ>5g>7hC(JFl2SH7Uk=vPbT7=j>PYBn=LM(4&Z#K_;jv5Q_;fMfx
z`1!g~o}5w7OvT#dDt<h?RESMH_bHj?g%h1ZERgfUIxC*}w0L36DsS#0^yc4P-cyNt
zL%YBWpW}txX)1&=3GrH-bNtdh@xzIKA6fI9cw2zYTLtKP=8o5VAM)Ot`&HNzzPZp1
zndWZL&vS)zs4I*Qxq@n4@c4uaB6m2$@5x1U=AK92Ehii(I*VQVj$)N4Yl)B8;Z~F#
z{GZrC)Wwdw01v@t`ym7|-~4%Q10!)8oVD1GWB2#rP0e23YwX3jTYE53We*Hq?Z&t7
z)(E^}g|JZ8-f+&%lqxfvH#EbkncR!>&=k_#os&F!JAU404|nf2tTAOjlqcs9&*nU0
z_HVx&XN(wQBiN+#{Kec5uB@Lax@>?<BLirUH-JH+K9qLoqkOzR9)H!t>`XoU8>t6b
zA3Yq|qKEs^dKl`o6*rb_MIrYq6lZRMVd572P2I%4)s4{5SdX);p-C6$@@pRZWLNWU
z>K|P^wdJ>4bl5M=^O>bOXq>`6S~-53v&F7+F0s>dZFrh#^UhlvEf!k*+Ngzp5;ZZK
zJEdGUa__<<&a7$G!1XltLS5HDvn4amJPlZMtK)czI*Kka%ka#`i{~}rL)4+%sD{8C
zHHe1uT;`q{_PeX$;7K)L56@<d)wusn4YO9M!DYT0d{or%-xxJ`DX3xmAT`(ztV849
zb$H&r4m~fXz&Kh28>Pfx@=X#e|L@*_JvK@v+?|=gekn;+q-9OTH|Co>feKDB-|RG1
z!BXa%OZ${z&V19t`zc8iWyG#ihAH!n2lLJ6d?g63C}HbHB`lesh%jaw<3+<UZ1iyC
z-x-F+{ll;?Kn6!lWH9!S1Y}}GkYOZ>q~QbP_NSM~<}axRb&=ACPV&j>q~QHswB~3R
zjh@bWSMxvg@H=-zN%T>+0efWBWw1+E24bgVF#om;@Kgp4fjrCcl|j-?)*kb9ZdiXO
zT|V4ODf{`_I)ZcgGHXb-XMh~{vlgUV5>Y|rl(wscI);=`O<4)4#xk#n<x^inJ~_QB
zpf9q8G*qjQlI;p<pgEsJEb}S1K93%)&m(*08{?)=)YkKn^45GL-(@-EoAQCKUEnkE
z1FgLAo@|0MXjVrWJugb4Z~rAx{<I{@-j+nFHA(C@e?y9s-qTW<_tX)ZP1DSBs5~#9
zd{2~+=j~#0omEWfnT51dqJTu(3#fBg6UAjU(FxYwwmcER?7w`Cci>L1M~URJHJ(<y
zizSi$u~cw4mcBd0l2=_U$$g8bXZ7)v))!Axd;a&feM3CWN#o~tCeXUvL^7P8M5ehZ
z^dKXJ&IV+2hv-Mzp!bQKy}poOY#v2i%A=@h`IMfLPa|g)(AXyhl-*oNs+WrC+|^<d
zRWG4+n@VXDm6P@1Z}g8XYoy=R(#l!&6uP>Bj&Zl>*a<C^bGwynf?BC}Xa^m>(?w(F
z_EOLNUJA?TrKH_`)K=L?@!|udV>LjOD*_)qQM@V?McYU**1m{A{+k#?jKxv=Q5=^S
zN?_P5NyIbXwAe}_>n`hW(s@@kUJ8R2NMXueX(WAP4qhnB`3<soFDs8yeFgMJ48fx{
zLs4~mD5jh7&!h-vNrp~<VZd>`e{l@QC5}RAj1$CN&f)fib4XgiKF(E^Fg<RGrg#Sw
zuH#P5J$!HXn03ca>?@J<gsPDcefRwEV3|J>tNih`EC4Ank1&4EQ*4QR3W>Rai17}D
zQ(Yk1CI-R&b`aM72*UNb&v5tnGn{|T-9JB{VPJAFmRqxz*E$UH#;>sOMI8RMj7N<}
zBD{G&vSa!iIEB2$&6T{j(oIIIX)=b|CF8v-cUSu*<9S>%&K4zO<D68Cy_kjwzck!R
zOhZOl8bUSFF?@GACb^{Jb=NzDuJ{0{OWgHvq6yY7n(&V`@2Flk<3)8d_Ko4poZl^w
zo8Ja!y*6YWVI8vKXFQAijN?@$$a5&gp~@DxEpEm5v`);u*To##4c*1;jWXzkv2-6?
zcy`mkvzzjxz1Sf87oV2>L1=k5d%3zX?MD~>tnNa%elz@in~<}o5rL`oc)6k;?WwiU
ztEyq01<!DJkLI$n8aCgmuqwV1yN!4kB~b}+?O(jF`+>Ng?=U|6jrA8_;h$WNJ}L*@
zEkpc>G927j3hfc4*mtW0UUtPuEG$Cgv?Ao6E5urnLd@dqoBs6$IH#A71I#yO%r}ot
z<{@zJ7Zk{RLCcI(Odk~wm$pP$rYAy@bvLVceq+b88^?!<IA9-%g@KWf=UxT9FVPU=
zEF+WAai|#*%lCdUoEyhkG&0OSSEEqFK5#{O&b3jCMAX<ww6{bcJ1rby?cw;_@fwc2
zo4W24hPx|XVsUXO611Kp*)tf9pMr2+DG=P6gxbxVgMHi|LlgYaGSQc3`S%f1ei!#P
z-$BmLTUhRY6VU-Tu%z@l)-Ak_%Invl+;bJtHdi5C$vu>Vub^t+5^7em2Wf&2_OQn&
zq}&@*L%pFB=!LR{tdHR?gUzA5pJIKn6z_R#N3yqioCk8nJdpXu9W#!4V3`5)iMas6
zZ30~5Q*9%F@d*KDt!Ir+lmL(KGE=Z_M{cJlu8sDDO`d@Hm)R$d{Z<QE=etON8)^dl
zU^ZGc$`j5!!wF?JS<2d-1!IKBWLAm4Bjo*=5Ko$fm^{}De;m9pjrUZWm}@3Vd!yi!
z7aolBf={9+d@gum@^m4@x;>fmJ@KCy|6L+IaLmpFJ->K=e8(MG%r<+SxL4(q8~R?j
z;hT&b9AaE?!QK_l)~+bebAf{yXWne!48LO+5h^^7E^8<DU!B409o#QfW5;_4I}Fi1
zjGTx=m>hNpboLM|qYj{Lk1cm-*`T6#KNOg6hQ;p1&GmaRpuZOxVSBi@l)I-M?Z(VS
z)==1Hg_GYcFg?*6GaJqD(9{g;ChfppA5*yHY{$_y6AY_iz0GH1q-3&=*oM2Hcy=?w
z!5ICWM(nXO!s%E;-Vw1bcm#X79SsnwZGc;H2KXmN9|K1EoZp}i<Ij5hTBV2SfqJ;$
zqK9+ZdYn6`$2p-}5ybPGRrXsj@4yyVEM@)h_l?+Ww;rCW)+5DN7lU|qqqt8OrpsAR
zGhG*-hUwzwc^w>I&0I502a{xVVBMgNFX`F{yrhlVP1^WZLL2jfxgTKz_d#`O;@VS9
z)LU})-gHe!wQ(+SvIb6G(cml!4Xl{2fsViG+-JfX8D^W}9qKqShi5heYKSOMgJYN)
zYJAiXdPEHlMm)PAHC&mlhQGtrkUp>uVU6o>qih||FxJ5)VI5w+ScfYC>#+OwI{bHe
z9Yzb+;jqU#oC*{{U5_YQuX4s5>q-9({Y|s~G?9a63vImfhngR<KgvM`XB4LL?tLoO
z_NZ{q^F*BTodOM>-z;Um*~onJuO91+c}L~VeDg2!O&s&h9Oj$sk?faRH3Dxp48gOR
zGFUlV2JUlYux73dKB&oXMuRxSV|WKKx{n6E`9sV3dRLm+Nh|rpzV9SkKA$o=3F)1*
zfceHlv6~+5_(KlmJ>(?PM-7{#AsEIPZf)XNTrQ3x?#Aq6zmNH>9@_ZnFU4*AOO2OW
z=x9wd-ME)U^WJ4r=fSr$;&>*VwaTL1c3Gr&CW~h$IrQdE4y`zxOWt80X<7Uyik_KA
ze^%wuw5@rRwIH7sY|N*j5&3j9A)iJp%_oQAJgU;nqssT6Y5A*9bZg-!I>IxYO?BCH
z>CFc^tdvELi!w?7cLt3#$)M}qY2@LQL=Tb^DZf9FrY}h%cfTZR8})|jZ8J#sS_VD7
zlu65$<WTCz9Gb!$RjXG_y@`e7v8RAcTk<LSd_GC%=TmuGBT4>l<nKizt*}p|SuF`<
zelwnKX2g=^s#tQ>jirX|v9xJ-EOj-;l1V{4&8Uqh)z<&ZGx3!4KAxUlkEiHE@zh}!
zPZ`<?RP;50s(7}vdvpqg1ZI=fpj_I#`4e^D_(JLmc~pEVk1p!w)0;S+TTLvWBHsdf
zx3Q2$@@(ns;Ue-sT})%L%4iJpP4K~QRKK}~(#7h?dr3W6YBZ4e;70maubHmhZ>5QW
zt)v##O6E8C-&U@dR<aj;_COB_zxL9|ZG9xpj-=%61JwCN1lKl;qBvg^GQ+q({h$~Q
z{}6+gAMd94-}_jq1Pr<*xMN%r$^uDDOOr%iJ@2WQZ=7y%w$A*)c-t@-l`~{9?zb#=
zjL0MOyaMw74uRa#q3q!uifW_bFzFeC`QGCp+B}J8p4<r>dIY(G!!Yo6M2O2-NY1i^
z#5C@RIbeyQl2(|%$`P}|9ijEi8I@OE;Ki@$X@)`^I>y(Ai$d)6=KIfM_pr?29u_{n
zhZPI&qhjI{^rSq+lbwN>XB!Ba<NR0{gz>ANL2TzU{0!&L&F|a;q!^46zOJ<;Fz1Zr
zzQXY#pd}$Zk70d|ZU`>eJcY@Yr<k?jDe5(z!WK^<oct6cm7d~cNC5UE2JoCR09x~c
z5wwD5K^noR-W-g#+k#PXG#EC~!K~E`hWhDu7_&J8;#GO5?=8U8<G)dLv>C4>o1s<T
zj73T<SZ>gQa}lj5(P+c$xt~$pngiK^41_slz+qDc>eq1}lOTt?XDXni(}m`)E}Z<`
zjZs^A@y?93Ha~h{d8ZeH`R&ptJ*?mRgYe>R?q%=Bi*H@v1PwHAZbm(8oVRalg3skf
zWKHFKV*NTe{jEXd;Tm++SL58yYKR0?;p->fNA*^qig!`12EVYO^gFT_eaH9H-=Kf+
zE9XA4$81$O?jJ0JM@=c*l}mB<K?x)WmEgM#ca+X7hWnTzG~OwMjd&rV?-np87ND5(
z{rXvdv;SNk)C#{~=V9iX316`COA1ETvlps45shyW5s{FHv8=s`dz^^>xbNLCBm$x@
zBRP*d5@r{ov3Xi7?hfZnKHV73<Bh@JuxLE$h(bEgZ#-E`JeG55CMiWCwJ!pLOL%VM
zAC4P_;W!`t8ZpOSVf&ac?E3WrZqA{2&=Z0lgJ5L5Vy)KLK!~q?0&_gVM`M5Z`uo9f
zh%d|!-pBa(yU-hZ2hD-EF!{($7(3iR<J;?)KIJ+L1lJ(idKEfmS8?P6`@x$pqpI~1
zES9i0dkp)T%Dj1x#h&cHUO0Hf3%Or~xSquG8z&)7>kFap=E>RN0t`{`z=J4vTs`cL
zChjvZJ?w!c%s7UdIn#2B08@7hn4x%2rR9#$Lv9#V>dM(!7jR(-_tNs7C-ROn#8$En
zS(~|NngDO73$UFTr(n4N|NSSxjuoDWe&h+QEKgK+v5$K!GY$VP-8{k?=l4R0UKL_D
zYmft(afUZ};h$nY$zF(N{;|5|i5p`*@hOStN9UQh#tAUyg$HYXJTR)q9l9CrnCIw@
zL`8S*vvFg8wi`sWSaZ|niYLLYaNOmJ<5O5u*Ufo1TU=0K<c!d?3wUza2~o=IF$n$-
z-D{3uRW@ha%;(w7=EI1-dI(aV4nlV9LA0Ga0K-%p%-g*m*K7CT;jVq?j<m+0QQSK-
z)(RWXT4KSjU0B&>fjMU^Fk8d|*Us&PbC@}bTX{caWQKnx@4!9>Q_#lkNV;nR=}+6R
zH`JK#*Nt#b$_Ufi4e@ZX5uVlmuRB5Pxgo}qA<DZA;C;vdTUHt%UDN>9yq}u8K_9xk
zdKe$dO!JawICu5%>VO`sR_VcdU@Kz&+lr)=Er@ut5xWHIu`PZ*PG4D%pR5s%c&!V|
zo4OcxR+nEdct15-7e~kI;>$A~IPTU#<Psgs7|lB<p4pUTYh&kQZ8#p)#_2`cNU7%@
zs2f_ix>Ad~V|ZTU&7BMDG;u^k6LZtq7j;<!o;x)#j^{RIJ?wwVP{;Et>PTd^QJ=xH
zn@%-;pBmpIs$uR4HC))DhKzY?_&+S2by(E<*R@eWQLz)dyN|7Vky0w4fPgI}1{Pu$
zF{FwG28xVQf(XJ4F@#tkp^_33B3+Ue<-4Ej_x|y@1{|1!$6@CFuD$kJu|rfapiLPj
z`N}x)N*TWcl#$E~lXF2Ciw`J6#!#7e>Xor(wK8YMD&yZYWyDTUMu5CB7Vuum2+w9(
zcdMB$-e@KjK50!qNT=-w8Et5yzua9gJ)xNbvzp1YyO}&jx6snR{o#C@cew9N!+GYK
z;oGL7dkgn_vwl;_d=pY4&rBxI95@2$=6#XL`!2eQU9{&@H_gA$LmMyk&@AU3vUcmC
zc;>5Oeos3b`G@M4{iNXRR%+{Lp?7@V{B5Bd3au2&?}3-z|Drj}H~W}xX5H$fTZi~(
z`zH&Tm;Ldk-vBf=_u&rCKDd_GNe`HBTIMq2eD0vyRjoAEs)h8L)2K?6N<YnB)0D6`
zbi68+q%G1&_h=dgo=GE9Zs={|xzf-ZpXslE27L_6B=MXaDqWXDB|16udulH2Q^=*C
zZn;!>JC`nu$))otIizHnO&-g#=&1OAeAD=y`6%`y#g9&>Q)Q`iPAZl5fBZl<2Y#T@
zgOkZ~!Yit_PN1<436v&Ipj76aC2EN@;B_(`9rA&64O7WZ?h{>({X_<=;q)6=K;C+J
zw8-EKolDQ9=svk5oS03!Cuh<;w+t%#9#843V<;djn&jR^)3ReRbY`T4dUeFms-Y6v
zuuDQK4oj#~C6+p?<H)Evj?S}gQ}|yj?VTP=9z$ZuzC%JwA4y2WjH5X*f!gxl(9s(o
zX<tDGJu%6o!8fxhUM7e3nCDQlPYw+X%B81rUufF{o}=mJ(W8<)T5%ws1mNtTgd%#p
zvY0xil+e<)N-FGIM+FV_<h-kadZskeYK><4^SG5>2egti^UYAZUu0C?Ng8W9$@qCE
zox9LQZR@&8c&nS{8uA`ht`yIs*n72@Jy<7L-;?Q$50m;JcUK?yJ>gxfd}(C<l*XDN
zGPuPaoQupizKuMq8p8Wi9{r#)O%?;IWnnR40CV&}i0{dv@zP)@_6&h&CHrqCum^1F
z2+kKBi~Z*&AU$I;?|g8t2tS`!KDR=8jum9Mqbe_O2Ojq_!LV=>_7iY^g}Etf0%kCd
zzXa_n2h0%*a5YK*^N#||*v#GHI?hnmbB6sFXPlhG`?hm9xAl-W#t9xnD(ErV`OFM_
zjA?vE1wY2TxW~B5XKMUoEcEiha(@4Ma$d-}c|y!TEyT-HLQJ$4BJ;2iYc~nGYg!1+
z*+Q)6tYjx8Ay#h?Vwr{zHfr1rF<J;q_MgrhF2r+LA%67}VkDnOGC~x$`(Op1!jFDX
zv-ijF_D}@A3CBmhR9;d1$k)<U+>_mmb&bq8<64lQ-GU2EtthI_#80o!&`L=`jn4;c
zTlE1$dy=u}eKN#PlhNq@nRjd|VI=sCmZg92W^)&2@94sqlrE@;GuxPVLGx)Rkl(>M
z#(&U!;}6=c+acnet-q12ShcJP@3#HGhdYf>I$h8E|Mie*t>NtG8r0TR<228w_J&lU
zcUC2Cc2;29<O-ZKD~DY=Yd1^3Vfm#}tP+=Cd~p$uSr@^hZxPZ=zM^+tA?Jn_Vy#Re
zh8-xtuUQ4q8<LMro_V-EI}d6PzF^40FMw|@s(Cj&-XsSTin1~8TsFccXQQC>4ZJ$U
z@Xg^YY37^dJezV0NI-^HI1-sfx<*H0(FHNWjzr?RKW9-7iGlU{82GG>!MzDFC=HH=
zR6owAIUR+~oO4sTArgu|#5nOy44aoCjBym<iX!iXmqZ}8>Lun!hGD~<Q1o8?0vE*3
zQ9B_RH*P(H<A^{ArTsCg{V9y5Kj9u^AMP{b=hB`>xOvwLr@!9EBC~s(1$P&xwLPKm
z)B`VudO-H_9gMHJjlDXzaUt{;<TGyK$ITlETgo}X!`vam`8293*I{4B9+!Q*vy{(%
zU}p0d%r{ZeJfC8}utk<D-b%Y*V4yShY~vovublHcp7)Nr1nhSaK(f^ViI40tc8@)t
zDB45zpB)0j?6AVv4#BhSkU!K8RzvKtu%8{u#@J!^X*<5&wnI>l9fp}X<KZD^$USl8
ztQa>;4tGO*o*P#Gb%WQIs~FGT<0iGMNZG-BGw&L%?7fC%vDfh9GtaJK*poB$DsoJD
z7RLJN`WRQ<g?7a%DOW6uc7e@g7pT=aL;5A}N$GRe%^?1rY;fXT87ItJ=7fzUjwpEG
zh;e$37|<=?UMB&LP8Fcxg##+Gc=oJz1<oTbp#P7vSkFG<k-NBG>iAKJH(TLW<Y6pW
zei#ow9%66mA*>9rMCB|CxOtd!m#-Q8#&aIKwJBx>m>{=%A3mAx!}|8UIB<3^M1S`n
z_!M_Yz1R)A=3V%vy$cyeJ5hXS2Q%4r)>e5Y_0EVppA6A?ma~ep44}2c0EW&6*z!gn
z<Njmq##SGywfYF~*5^5uKBO1x<7$T<3?lXLWhZ-XhHS;r_v{l^*upHZm7gbDL36g^
zTj>^@TFJc%vYSwLXCsP+a2{X(jo7nfBl3SR*Rb!#lRY<qr?k*#lNKfp=Ghecif^#j
zrkwdEeUT=jWi>IQSOcM<8jy0-z}amY=r=(F$+_ydZqHdabJfxJlNwf@WZh;edvNkL
zK>6MVDC}jvS+D`Z4ppQu+uZa}MdKb-G%(xDY-2BQq6&&wv+*-gfzoQ8Hw{(6o_b|C
zd{D+)W|~rGWjsBkj5-}<W<6zOGs8TQQRci!CH}ov!l85}cqS;JH9`sRo+|;*lu-Xv
z3i-V6JAwJ;GP8{`pR6B0$iMXmRWE9y+(%7hDdGG!=9{SIW*RHoLYmV#W4F%$oXr~u
zxuw$(r!$rPol`O5z$6?=mB*oL^2qTVf#qg>p-?TwuLn(Z^GrFlDwNVj?snQd^gDIz
z{Z7HFYiYh-9fhPek?oUadQ{p(QwKMZ;rIr6pxi+7RT@Zx&#(OrRKB!@2Cn`|YRorH
z?wyop)<nupvKSuLA8Ug8V}%auIQ6oaBx$1`U9BX~U6x0W)RQ*<U1n@=p_5H%6sed>
zGnsD$&)-mCO)4E>U++$<G%7inM*AJo=(lAWt-Ah+%nxPIR%BA8S0;`Al}WdCvuH<1
z7MaY+rkvtzDjAwXhCMknry!e5e6y(k$}G~4$)xTp8RVbM9%0@K|2sdGL?=EFvXbbh
zLK20}Nv7@HNn{$CKvMDv6x|+Ao%0gt^~qOcGdGd^J|^<~ESUz4XT7H}jjHuNQo`vB
z8tnLm9<_a;a+_>=oBD}0p;QW4@Sf&Kza{6g7)lVwkauo0$v%muZ&A^-*eZqwI>peg
zmKZuVP(t$h66$|YLTBwHWN}?WVfQ5@?I$6vHxiOWOXxzNgo^J;$myDdh9yd9_{(@Y
zU(CAF)%TS7GlTwo%%ZXEzq!&YhgR**q0Bouv@|1!+UDibk;q)q-;zgyqCAQ_m`@c-
zMfB*@SGu14m0qb9(W0A`G^w$cgy$M)>iY)r(`ckX&N0mMX{GzjH^oESXw=i6^mkqd
zC5-5#Gj}@4iuq>FVxBNqGYg0I(9VW`)VEJB+@0PFEgHSBIHNblcJ{_`Wu8k#NF(iw
zG+f%GVL4a^f{pAgekX(Bi~GX;lnnlx(Hny=$zZxz1|ykoMCJoAc&8lFHV=a9;h|VR
zb{O1x4adFqG3cloi(7{$KtFjhk{3@!f9CdUTdlCu(F$_wEU-?;7@ITq!PMUbp_5H<
z+LS$la=hCvYYxw@iwJPL2$!=bu<oWAVwavkb>EYm+j0`KRBU+0V~c`U?r6z$$Eh3a
zf#B=tZy#@A+2>pEK6e|{=Wk=p#oIVxdmE?hZe#4X+wh-!2WeaGVQlL?JRJG}86zHG
z=S&|Ap67!>Nj{jz+%j>H5LcE7VZ?khjQK`GU5Nkq){gmRFY`_EFlHM5x*}$qUwwsG
z_{RtBHOw}ZK8X141E)40o@)w`uJ;^&kA^~{PZ)Z>r$Hb{$EOYHIIEh@J=y8lAeWAv
z{nOEBR|+olNx{|!AK<+^8CmT~_|=$%F|(8L_Cqpu_Q}AmK{bf?{tfvJf3WRrC$``0
z!f^L4v@zdY2<^m`(;b|}_lG&)H<lb~hjlRXO-C!7PPQQES`+X2G-B@2dQ6t9$Dhg?
z)E%zDe&(B52RZ9!QWb8xS7M7(DY~|o0-OgV_g^u7cND?kO%b~7i!f572nC^Eag%q#
z`wrv1srQBOIai2%4F#w;RDg(S1-y%wj~w<FYt7BWdG;3%9ry+3&*vh&G6zTZ<=|~;
zHm+aF=J%8=tdhKjZ>ku}7K`A@bE#`V3Ap4Dj<%&?@Ej3_>?vZb-XTW*`ba3;k3v`f
z80asF!C*eYaxwV+gu64#qp<LIBs`3`dtq@T=f;SU6ClEtM<V<=E`o4A^G#+1#*KK1
zVzV&J)D1;+YY5V;pCgCom4&;X;r-<R^w;o%_C#Mi8}I~4jhsUu`xqMyA7SKiFN9vd
z55wV{9bDlFpAnwuy6gd!e|J!P@(!#@ZsWc3Z8Uq_0-|oR7xe~QQ#pf%`=**xuEU-A
z=H{zwaN5j!OCDF@=f)nG%j^s0`O=(CZV)kh9Dm?~=Eu&ERdHqpXSV6*gcZL8Sd-;|
z5d*nf`>s9qZnMX~nfCZmZij96?O>r}hvSp&Fp9b6sgxbY|Fgw6W}T4=c1SyGhgQ}|
z5?0tFZw2>2onig+g)3SNxZgX{4XR%`6S>um_1CM|!dgy$_T`+p?S`?x+}P*rhS>|*
z6SSNeXtNtu^5>cLu6V)UpMROzrppDVm~R^TyPy|eUu<<`-DbKo_rN&e$qOfJzvjfd
zTTVFlfpe_f9O0<KndD6ZOt~sR9`lVh^Ub1BydQpq{~oi?qkPjTK>q|(*Br&uWGl$}
zABKbcVN82+2z_}j)$7K7d<(U}huId;yK0Wr`J7Wc(hTDdn!?W01jQZuU}v!p{=fIa
z;mlsVY~voO6MImXy9*1R?&5ifF^*UnLtoPv`z{*ew}vsw9vDH{%n;MHwqaWsXVLWK
z9AQg+6c5wKUe;;SH|Rq&TOTn_`Y_ejhs$(*{Qji}&2T-=9p4HGcVO=I+=A|YTQDhw
z^Z0t$JA6+Y&m%dLZ|5e=>$M4;f{p0?MT>jlw4j@yg)ebhD7mMFYwW%8I-v!}DOz~u
zrit~NG*L50llv7ku--%i?cEx9%yTIZo=auz=iBuf_#&ghy9}IBd|VwFW7Hw!y(tSL
z?!4^g+`f<vXt3n$J=Sej{8h#DWL0!9+c=u?OlqMj3fom+AJ6kB_Sg((Z_PWNLuIpn
zW@3gicX%qJ=87_=>{P~pHOg=s!&=R6B`hycf_<D4?^`Khj*AjzS}S4rUL|bNRl<5z
zCCIN=!lczo+<Pj8DCV1E?hUl;dIM=k|Df4<Kge`s6Ww!bq9ut<>>F(+i;8B_Z)g7L
z+d|XiTPSHt3k41y0QdZX7-PyE0DP{!AB!}XP8#pe*U#+n9A3q_CB98`-nyKIt}CVM
z8sBJb@9$JN=sTU2l#{o11$myVAY1Ne@~^ESWq01w@vEniJ@xdT6!QrmP39ao=9|0a
zwWN5EI}PUkB%3Q$bc8cmlH+AD!$TJJma^!jD2rl;YU;dFO`^p$WLQ>5!|&JAx}-X~
zz}?*zSF364y)?SE=L7XNNTfqQQc2A?jUJh&kq4i7>;ay;FpZp~`E%ai-n=1|bf%_K
zVZT(G#~REJ&(HKZCX4pH%ck!?v*~DE4lPK{rilSrv}IiuU09S!?Gc}8V^2ES{7$31
z&=gXSN~U{0?<spy5>2m2qTWS(YnVWtl6bP`6Ve?|qwHUiyi_7>P)(%Wrio<1pZD^S
z(1`yeq?i>$r7AJ>_nVluC5q|p4>8S{6iEekF(k_xjao?zHHF7e(Whu?@QkKBVKnI;
zjG<F}fA#ek0yB=hHv4ZZB~)=mLIun)%FH)Ed?j>3BB7X236*(E=sk1KqBRmKx)wvf
zug8;|=PPP7dr$rYGD$f<ix#?O(~gL2I%}LmU9LHFHz9|Xv+wBY_*~k?T2?-LlnQuH
z{QiPMN-Zp;%0XYrM6rmJZ7NBnp_cBPZ=i^5es1M9(%%n1XllP!3ZKzNPqo_#$N!S!
zVCI{u4qBn!NuMmc=;y3%QajsCw!gbcqpXKAZc1TRrxYHl_rfXWn_qvqS5!qBqnK|t
z<Vr)aMVk8$*jH@Lp6hZM40+lIXX?A?!R~)lSl=5F8q$!l?28Voff%_%4nYQ-jh#6d
zw`_-UM&3xw`Zor@*?)5^XB_0xcH!fs-8lB(5XQ+{;gqEnH0D{LW3DltdhbJ=hY3#3
z;oP2urWhzO#n=UA`1sHq3LWN%zG4RHv1S-~&JrIlAH<V82XX%DQ7n)-hcH`vl<#$f
zzON%@1@ZM_m?JK{b%ZOw#<qTOM4FKkbpHFFX1|d&`&ho#wEB;G$_McE=Xhr<p6iUq
z@eg1<*c(}^A48?o2W7p5*guN(o3%nnG1t_n3DIYQ5T}`MG?;H5G5;LvFT`bOA&&GC
zaz`?IHkfZNG2bXN-{kXSB(NWP<X<1ycKAS`Aw+_eFOqq;*f+=*`=0tD{yukty8A-%
z#}}%yeu#RP3`;>W%Eu?;X>AfVR3_m=M-skmPR7<-$=s!tfjOx){P9mby7Ci?On+kf
ztajc5ZpXUjHfR*IVK8S4PK#)R=xH10>9^s+RnFF7k55u&Eq4R6pIEO3)7RI)cYZa5
zvehswu0p@#mAGJ53BT%Z*ppR+0r5rL5mwBZtHrSAjGv%aMVx6>gtnHiyvOnt^L07z
zM)oU)v#)reRUt0daA)}80?3aq;J)E6xHjSoEU)wa8T*Rg-N?mQ)m(mU%z~d%CgxPU
zg@yPH%3r?5y_iHC?<+>d1QA}TMqusTaD46+hHZVrF!FaOuKjz7qw6@EZ+s-?xrnhr
zAclp57^Qc_7#hSgs01+<N=2cMEpv@hBp&hSE02jW&PBxj1@4Uxjlitk5tvXAjuDMv
zSh^$(d*nl57ZJkm{XCoW4?@#~XOKM-faMOH(Rbk~g1-szo&UXJtscWg=Ml!ki*tAG
z^DN>n#15WlN%w%3F>5!?cQC=~4ld{3#<VrJVeWVf_5n9>cl`~#;9Qws_m~;H-JsXU
z4K(s9$||lxclK2nymjLaarWG7<-9&-h-WukG04pY@fFS(sN{@m%m}mo@-B6q0GmY)
zxY%TmtgH6ur^Spj#2#j;yjy5z2U6s@)Nnf_{k6r`7F$eevxQn;)^8Tu!Dzo7ZoIT(
z54b&FM+h)p+ZEkzJiCG$y4dq$|Jn`z3YcYb-B3~OhO`rISiX;Ut@!@$^WE@t7W;yj
zah@o+!M4X0r5Ua;6}rOsyeo7Q1=!NSy_n;iF|pbS-z%Ij=#LZjeRP6fAbV~c`RDKB
z#Q(;QSbWhD+AAF~xKhCHuL6vjA;3@On*w?6o6@kuT)A^t_{j#N)*r{O0Y~w<mAU53
zA<n4g4D}LA&R^S)UQ_o&N@#&yi!D%QX^x)Z=5U{2h6D#w_P?26`0st#v~M5um~Unc
z<_tg0ofx=qCsao5z(|QPO0OHkcZV^1f{n55v=Ps=jBtCkAx@ZY!-faE!@Yrf5?HH=
z)6<87j6Mu^@m|zTJ*=$ZT$^k?bn#yJwz1rC+|D~wwp&nqK^Hpbb<xnQ4N0&z_r_?W
z#Znt{dTGNaWHXfaZG!ygjfhm-2))-@_;pi@cQ3V|!+dkZNefE_S}2IqL@#emJU^%j
z*EO0LTA+a!VHy~*ggd|&X<(p(1|}TWzyK``^vvM?;#Tei4^zhg6Ls|Kr;bxk)!@BO
zjdzx~d*L$oF0A4#Kb}eHv)|_IZB-m(wpq1|Gi*AUZCJP2=A;6(Eh<QxsDh&nJcEi;
zhMGHTGp5SiKdOxRLzPimqlCF{mGH+~2{sp$khen#2S^DnQ<d;oRtaO;6tTTb5%04V
z;g_n2MM;WCdMpJ$=9_iQH><BTklD~ma+9khi^27jrQ1M`OPcBRie~Cu-AvYv&Ge;z
z3r%3Y5i#GqncqTr8(YXis*SEsX``<u?ezVj42JIMi`okI^;~PBgHM}C%D9}K%r2#m
z>fdPD$TG@hzB$f(V`Wf5wepqZy0D6F*VItub!MCY&o|4NZ4UBT!+c}Ge4|lTOS748
z#xdXY?W&|rD~qV*N-@Q{6jS!8ax%SLPRWhsq|1D>+^L!dYS)m_rh3|uP{&$DEuGm^
zOU9R~=<Crm3Sawy0(oyKqhA`0VGZY{Q5v1rU>#?C8r}VtN?kUobY3ZyDkr7VF6mSn
z(vd>lgP3{NWl*#ro6O3xNv$}C{N7}ff>t(pvsY)!ozJ9o^b?6|lITcj5?x{4CTVpt
zosoS{>9^if%+MrSZyQf155&`-?eTQ?WITyZ$J1!L1bUjyo|m_;Xg}*F8S7$6e2ev!
z*%Dft7DE#b#nAnFF@3ic)A?XA<;q1;=<FEk7%ZlW%MrAGMFbURhtusQF^zSPCR4{~
z%KZ~f>Mk*)5g9{~%s6+Mh1{8Mu3VLn&0S_3e+dOeN~l?xxo5tF1X2>(AsbEO&0|Q1
z=ToWXuSiFg^_wx7^tULBjGeM+ZCEzdvJa`rK8N;F4*htVMYpZvX=06pt~bU|S#}I%
z4=JSnS%sw8Sx9r#izr2}lCD(N(naSw3P0UQPF0PxX)61mBAUsW`_e69epA%ZKV-fA
zFEw=irKQ~W;Q6D2YG!m((&=uRFqvnVxjm$LM+!Pad*Q&!Uf6Bi3l~rH!kPrmz6q2@
z^J{4=W5%hKmBHGF(r9q%gOp)y6xZHPDr0}sN0(lhTGa>fyQR@JN(S46GWhK>5bFxL
zn^k56o~w>V@U*d*HI4mZQag~i$qcW0EKs-VAnuJl4DE1pO!{DkuYHU$=@9G8S50tE
z-V_z%O}R(d6pMbDawisNT8uHn8W}Tw->}5Iy$5l{{t%|S9ztdCAxyw=EVy(Wf)^*z
z9d{B--kpSL^+~t|+aZ~Ce3N%}I2I#-Yy|JB@@ap|{l$G8q0{4tZRcI^?+Wkc=DKjt
zCU+2O-oh0#FYFKT!opi#m~z7ld3+4+cp)~?3wz#q;klfUb54ZV&F8>CW+3JpKjxcW
ze9b?w)(4YHe9*U$d4_jzg3H*)!GCtXl^KV><~Ps1Y%PS`3BWv5AcSr?=d5h-h0_sV
zO#15!@oYbMZuCQ>wLcCOCSgi`5=OFy)3q=evmIE|d6W#<kYx1U{sCwAf53k04+y#V
z0lx+O@y8Ds`#T@^<qI%+LIJj{;12KkMX)(q#Cm%XT3CZ=8D5Nxsxp*}`;H-dzH>L}
zckE6nhe3NO&TT0{)zcy*yB4AANfGyZ7oowX2+jM8V7-m!QcH^PXM7RAR)0nL<*#V$
zEJQEmLTtTOfTF<#cx;&uBfmTxbN>QW&O>y+$}_1Ixfq|!yY}(fu&>R;K7~&RpBRTa
zCkfm?MYB#9iPL2w$n+6m+L8#Iy%G*Kr$U?`3e&1kxN}C$&-|C%`4PeWArWXDA>y4A
z5&E4G;Y+9p8^4GU9w5eo3eFiG6A5!W&Z&+Pp)Dl>pPeFbz%K&YdJ!1&DI9AG!Vo4G
z27}fYn0O`x=nO{Xfgo(n2!zzS0BpSIhj*W!;!-^GYPS#H_Ym8sJ!Bor3)^3Fhxf+2
zaFX+c`dJTrTjGKBRd;aWz#aTczYWQM+y`~}7VbW|3A}R$4Z4oer?29h8~a^~TzUVW
zbBWixLOI11_wTUZ*TM~-HJCAGvUk^q*}=z!bLgBgWtubEm7Ore*%5!@1z4rxz%xsG
zOt@f=MeFS`yweWnm|L<A*g<Nc9s2aPgI29A@XZz}<+j-J+ZKN(*ui0o9p?quA?BAI
z{;m+9Ro)d2M_e(ES?e;h%$xIW*zC#P<Y@NRL~u96IyZcs`aiAa!*n++n&gJ?*X&PD
zW^Z#IXYYtzA#``eQlWq|jRbh`oqx9Zj(Ggu5k2fDHZS1cE!LfO`>=NN(-E3{O|c`;
z5!J^XxhI6XR*LxRodj4vR{+H*&hZ@XfYT=~p?<+xR4UmZuHYCdo*cmy*TYCDIf!pB
z4{+YKCE9%ULsou2&OWrj1HKMwJYdc~L35l|Hba~%XB3|>!GVT-P~={zFWYy+BxF0*
z+}Mt*>$juVtnIJ~H)f65m^D;m{26YH7rTv6)z1hqtGMfOH|N|OFaXUo;Qc>+)U9M6
zag!drRrIjyj2`-A=<z&85BiVw@a&ErT&i?2EKC>rkvbT%PY34p+8F&*8}GMh<5wSM
zo8Oz+ud@jQ%r@c5n~fN>a3dDK;CvhAnFkh}Z?j7a%eL`Ms$3J>Vt5zaQxg~WX|f)v
z372$cm>><_1Jl5ucHWO_QpbXU?7dm2foVfEFt1D<`cKqxLth==*jqf;T@8yTsiEY>
z26XFez-j3X@J>>NguTQzJd+x*R2BVutKvke3KHFT7R6qhS@J5FT%pXFnaX&7Nf{Hg
zm613}8KFFXYGAIpbWaKAk0_y^ni7mBD<QE%5nlPsG%<>>d8Ei)9E$AM;*U)fvB^jg
z+w>HnZ6k&ClbK^CHqga&4K$1Srl@}<iI{KdW^h*k^Uc+j&2*faNmsR*M*VE2F81IQ
zG2i^0+d`2sm2}{IC9TY=;taSNiYxv>k^@b&+qH>OIlKSU_Huf;rkvJ|E~WYNO6lB|
zZ**fs87=N9qld%FNpEWf4Q9TX&U|CWd=u|e&;E>hGUGGvLOtnzs;9(~TGCxuLpryr
z$eVi#!aKS1U<G$Soi3)q%r~1^AKG@SocwCb>G9=ia+R*3WsNly^`?&IN7s?j<2o9^
zj59p0hAJi$lG4{y+T)f)AD_h1{E}2!rJP3VRMRMJU>enhq>_VWDy?0cN;y0WdQ_i6
z?YSwG9hXAs7gFet`3EYG&ZaLO+4PJvZ}j$MQR=e{+FqVSS+A35dqff~2v4FJpOPpz
zK8cd&y`xVtZzyG3JpG&$Ptok@eaJa9MK*DCO)HMxS|w2b0}0f!`W5BOjU`blvxu#P
z=EyU9#Kcf)a5Q<<ifR5aF{!(Y$*qkwoR878;+}|#FGWzr+z1*SA5Oy}!pSX9Oo;;`
z>G-KgN;Zk0iKY?c?H@tc-$&4{ZxK}6Dx!1BVtTn%OuCk0D)$!CKz{6gBciDBTNL$I
zj-fFLu~eY>ithb>N3$kolIyoD`lXRgg$dcTaCHtX8JNv;!gr)QGm&bFV=1bN=U;mw
zY0z6S{Sk_3eeY=cySk8$Y8TP(Rpqody@s~js-tfXjdY~f4=NwqOwYoaNp?pIeUJD}
zqCJ1;l-yr>QuUWcFX*5_!VcQHgFCNS#~E;>n;iRcj`0F1++ZFax4VLhuU1g9q=Md7
zSCF}&lFI7%xjUwjPAWFihFI?RG;5|H%NDYBX{Ez&+9<@YgNFC%gF5FvaF&*V-!txS
zxG)g5Uk77o=t$^t9^#6T8&Ub*2=l6U^KP;!@{P@Swrz&xM&>ArG{XY15uSfHLiP6@
zSmI!U4)zzHVZOON$rK^XIN3_37{AFB@A{a*OvMrnM-M^5e6yGNCh-0td}qF?H#m;5
zX2;>;cpTx^j$^<hYXn-_K+Vbq<DS_-C(RB`83GI~5kSn>k`-qh;da0UJD6|0ZCwz`
z*UtiG9IdDB@Kd<Sv+SFg63H`&Di7%V^nlz>Pq<ikVurmZc3<_xr5B#;4f4d}G*65w
ze8inmk8tMaBix?jjq;t|SZe8w$L`*UYw<>8mpA$`-z0VU;Kx`YHY{Zw$cATHo<ekf
z7vla8A?ED%Mbl$ney#OGf99L2J^py=;g1i=NxbtNfY2KOsLKh!w@HC$UlNFP1<v}}
z$mi#Coa}s#+3cUSv5H1>U*7YY`UcI@-@t$U8_rLEkK`ZkF);4~U+br^e<+P#BhxT<
zZyM&E%EjC>xrpDJhraI$@o+JFh1Yz=-XlejH!Q-6)kPRPvIyq2Ur{Xnit+YeQCd`p
z@X$iob>!lFCihJZDulIl0r&qDKvp&%pNzjC|4uGOSLATUT@H?QWnuIAEWAn0#FtYU
zcr_vx!<?gFTNa6@!y-99EE2KFBCM#2fX=80G#P|LJv|h1a-rC1_5ySNdx5zAFYxm9
zOSGTh8I!CCp({lAkLONvpNnuUTZ9O2F^=Sk;XgPMP6xy=QWNoBNCYgmhhwquC6-r*
zVHNX|zaSJLiJ>qmc>&T6LH65Vge~HXYx_VP>GWq0uODjPa$oBeAqEybgqQh4SWJ2d
zJ?5L3Yj<I_(G$lmdEotM4}|vjz>Kmx+^2H~nV)aM%EujFwz=~jJ$Gt%T!$os{lqGq
zsmK1Bs|l_+Xu!Qp<2aLewJWk&yBT5VhLcL1|2vU=Eg`Nr>gs}nKe&r>s535kIdLzw
z6B5G&D4*Z}iMKt{E$!jE$R5HPJIuakheJG@QX9>4s770S%(cbWbX%BZ+9IdI7Fj&6
zI<ndhK4<LkDT8NNmju{a>Vkas8!K11B5VbFn)kbLrk5M8d$}Re(G7xLyo<&6o#&r*
zW!V36lZQt0j=F?*&0~2t-HU(j)2>jo7GRdI0L{e$oMO+g74Ln^e|Lo38z(e{aL1H~
z6DHSk_BeCTSfL}r4>+Q0iX$>|1^DGCfX6Zc_9i;Opy?tkY|g^v=o#prw8qR+$B<=m
z1lx8T=B}87*x-DC`=2av{_cKM4c^b$Ef(mRVu88N<_Pa^4tu^%Dsnf)E(;Uf@YxIH
z?%fFP+Kwl;w{w3ScP>oXj$ft5XmB^itCPk!wb2-}WsQ+zU<CPgLwHRy<llP(bSfLL
zzQ%cchWc1LQXh9bnQ3O~Aw|3um*w;j?!Y>Zi5_H=wjdF@2>Pvq)wVi__)iD%U$l|y
zqz!WwZ8R|3?7qGk^M7u_#x<KT_u@vF7inRbmln>R)I#`PEyQZGk9dI==G1deakM6m
z-qb`l?}8udr-`V$JcC-R0hcivP*zljRkS*CYSgiFqy_?d)Uh&~*@nHve>B;9Q?Ca5
z3u>@sU-3A5o=Z*G0GCg^8^vs+vO^W$=BYy5qk>1-oWb`%1q*mTsxPyR$2Vn64^l?j
zNo8zRR7Qjx>o`1zS`?^+!>5(-SzQTs^2{|2iYR!kh-LQ`p=7NHXI(|ySgMGU(TZr6
zQpD0`1^lj1z`0Td96l+9h)E4(#C)@LLIZWrYoH^{H_gcv6joJFTNksZnEB>B^UVq7
zo8}$NH=|g?nc6~|qAN)_yM*=*E1|NauN3(<pXOcW^O1WB@=IvS&k~CJS3<ppl+yE2
zrL<#gDaDT|B_loq_@w_Tp?UfRbd9-3es&qnXFf6cTTZiTxoexx+F#Wa*-=fQ3u<Vo
zdlmKTS3xh=f2Z4y1+<ep5<E5((%!yBbZ1@>IV)C^f?+k839ISU=Nj6-zLw%xySXH)
zBiSK!)Wma=pzs>{uYVzVdwrpsH&f_@c@n)<jHSFiaiqH_mGt@OK1ih<J5#COl2kf1
zES2U~rqIW<6y8-&p%jA@irDypE__X<9T$>m(W+$PvRN8a&pii8Nt7y1qEcp?l$S|V
z9-2h+Vv?wG(OdG%d`;o1@#G?jqZNj6B$X9Qf4yR9X_$o8?B=e7GUl8I64Lk}p^zBf
zf3jlV%m4``K8&HEzR|R~N=&`BiK*p?nBv^TbkLT)#MeZ$$}NI?D#B^WO+Mb?Bn%9v
zF)`s(njB7RR)kab%y3%tHk>l}OiKzUtNkLHb5lfrL?Wt-6VW6qG3gbEsbp^?nT14=
zXF)VoE3p^(oP^XLN@#T~`<7YL(Hj#(hQFi9B|Vxv6Qarcc{F{W8ABO!W2l#LBt_VZ
z>EbMAr8W@_8z`cSbs}nS`%E9=i^!{AIZZdKqYrWo^p^M9Cv9Q|v2CV)_ANB$XDh8Q
zYNz~<ycc)jFRhx{K|MWxNpDsM8EE~Y{tL^<@lqP8n<UYG%S37`driA{y`g6rZ%AX+
zTl#bDE$uISPo^(^P+{f|_GUGaw4jMRIZH5cK@-W?w9&Ij=9}NW+1J+xJEqBCg3JJ{
znJ^gY14m*yXH#u$--xpRw2|>r7wUlqNK5Cgmd<Su`Wqr-l@Z9%2v_}$Fge8t{-HZC
z^tcH+J54Zp5bHL~I8y(aqEyioT^mhtxt;euXIf(UQ%k(IJA^UJH*xciLt5oH%$aXC
zG2b+vKaPcWts!r20~6+(|Bl$e$;Jk4K{n`NU4JR_&65iDH89_tn&FK9m~VbC-<V%>
zftR-n#tGdq88<PA`9@#l0pCgwjQj0@ciTPT#C$XAf+v2sdZOu`C$cyP<sbVoyju9<
zZcjYze-~p%-NpWuce(%gA^sV9b1t(tKBal{wJi67GT+3_V!zH3AuLD;4|^dp0(st5
zF9aF+!u+8xK2P%F3{^k;(eOuKH-9`&@#pLw?gJ16V8nRN`sw43#2!D$@$KvT0VsbE
zfZ|?($m|=4=^?@JO$<g=docF?;|#+C&+&jWo38|kF!oj?l4nL^$f9WYERROzp=d<P
z#o}&@1O;be@OIHFgvh+YI_~8fUzvba#R<q`&+n_W1jrdAAXh5^YZMb8Sdjo{zVCKV
zCXzXaxMeZ-RSYk{x)J%vIG2a3#km-_IR`fyvN7@oYc~I~AaTe-ZDl5`FG!I0DjF|E
zQM`l4eU+;tv9M5tMt>1bW<;RyS2(Uu2}g+}6fS>WV5QCrg!Y7>*N6~2a(j+<mtUgg
zcLXXIh%kJs2<eU@>}UOEz*X+O%oIa=0DEu@#qdsG4pNDL%&Kti#(IfM<zd*rCln+8
za(1=;bA*`%!!9uh4x@wdM*Iw0R06T^l|PKg4^7{nVD3~Q?mT*gTIQRA(GT&W+zVMZ
zJutJ62L$zZ;OxWRE!J)lx7|STrW;86>5gTO-Lanerq_7RyZOnT+F!3D!k;;U=Q}pH
z*|)-MVcCauoQ2%ew8stFTix)BHIsiIdDn@3EibHHpj^g2-G5H_^2iBw@=loM!T#b0
z_85K69v=<tVLRO(2lMT){GuIxF16$5lPwN^vBmLMwrGvC#f&&x7<}eARkJPZS;N_-
zZ-+IZcDTt*vd5qKshW45EO`H^!4(hZaHb$LR+x?(v_^4GUy>_^`MYw~q$@goIg8lO
z6%^`<!OUKV_|HX`Tsafj1s9A3oE;~?kURkl$MX!BXTl8`j(E||*9+cGc+kyRH>HmF
z9M3&dcN}rp#1ZM^SSS4~KpN{e(}>S22dL$q<$R#C(CD$jkW1ERntlv(rX0cBafi{q
z=O7~04<LVvC5(6Phk1ns{4Q8v-WUsvbTWr!lNqiXn87~U6ncgxu-DjwQ`+2pS+^Y-
zoJl-p=XTBzU>&E(7|-mD5qi)Vdh2*jC2fo~$Bp2YXNa=lhIn|&06J3)aPyl!DtQLw
z_e&4%r}c1gAkU%hY{hWihe{9Iie9Jq<Kwz;lh?(oR2{_F=)iuC4i<gV#>WfVI66ri
z8<ID})qXS1CTzl><c*lUX(OuMYvDd?H~D+Cu*z7A^ANRo?x%&Hb(&CQ{ib!JCY}!A
zJiqHan=;eDnE!ZZiZz>xN$Ri;Qs-+Y?gQ`5+D(f(>^`U?<r?pXE30!ani_5#R%5Tg
z1|)9XfaB$=c*1_0WyY!~ox$^|pDK__XQl~ML57tIawe%j$ZYeBy*6z}Ik$%AQBl2=
zF*!{MQXWb;w?_%$c}n>9OOdrlMNGb}h<Z~+Oj)G}%|VLjsZ!uBWd$4$P{3w)1(aP-
zfXguj*dA1XhgT2vFx$-K6T-e;$1x4$+pmFipH+~nVmZycRYB8cG}9uLW_qC3Ow-Go
zY2}0#TC7`1BWIS-j+|mz$ZTWqI-e43IM-K_PhFxC8c|t7!>dc^U`+`L8%rp?xr9!%
zl~6EqjV_;mtmnit-<Z!Spu=+u=;yow^5d*H>6{|6JXuBxab?ua=U03gbujaU$yCtc
zb>C^TT>%Afo^bzBh2+Y7qt&a3j?5~eDa<#S+p1~zqiRxRzF9)GbRn*e%EDPYkgKEO
zueCHHsD_@k7m%&b7iw`%q2WW4NXA}5t)pY<#ECeXz}{YsvJ}!-mrAuWQt7d5DotYT
z=9DOfJRYUcHO&-qo637o>B)5NbTXwZOD3;g$yChR&A+!vbRsN?wlUurG2c8CB$2gB
zBCV1nvhMPl2Cj-Hw_9-(IxUW>PR7z3cL^=ID4}ll1J8LEL;sdbsO_kPl-MJ)ly{!i
zcE`}K^D#7)S>kfJn0hZ0(@-rjU1BDg!8%O_Yb9rHMUc=ooVH&ICw13wdgZ}7&dG4<
zSrtx)W`&dIsBoIhY;%=QFXo%rv~UXf6i!<+!fC&kh>qP6(UODAB}*ddfpruOoEJm+
z#xW$vKa(f#VL9;af9&U(-x^Kf-}o@!3}?O(eTkyzxF}M)A4M8gQKWiUOku;sB;6<?
zNQr22frw%+Cz9*(JW3k;m9vQJXdrv528$c1X(H=-2bxK=tA%p6x6-wTZ4|?MRN3su
z;e;Fd6y8FozLimqY7(tbN+PX|Ni<PDk>0*Zr10d|WUcd-(m4Zq+oN}sWb>XHZYA;j
z>jRazr&7z;blNPRNnYEt=~TcEItZSpe(s=ACA~3gT|byq^@q-g5h&k10!90@(Dr>J
z665&y{g5t}x^2PC90R=XvkiYN4YBv4A@8slVUn#8R(Kl0&&3F<dT+;~(0#bK+l2d!
zOfhkQDOz5b;B6FV;$)iO^ifmveqzRtV}`3mW-x6uL+L<sc+WA%MJ3L@$T@~@Jg1Ul
zz7d+)aAw&5_{M?xCWiTD7|*H7`MOh!`DXVI0lL`xcy_)sj=8wt?R6KpI=kZ&Ux$BM
z%h%&uJ#gZ^2a*ar@Z*mM&X4ni(Hc*LGvAam-?T8_=)UrVVx=cDf+y0LZ@igrsz%<$
zK<1l8t2y)5z#G#ad1F(GH{9xcK=QnUHJg1r%s1}lLiE4ECr^miwL&<k_;Q{i=M~EP
z!O6o9M*sOERp1YWNPpC+1)%R{e+(GlkHcMl_)o?k%Xyws^)LWAF#+(13xL7BV8q4*
z<5*oV?(j~~v#KC${~iRzk|2yL3_`%uAV>@O&rE|@Uk*ZTNDwv|K0`~PAHMeWN7j-6
zbdL?h%<cf#yyu<xgaE`_1fb0%06R<rkaj2lZ)^e}H825d|4V>QHFMCYES&q8jfU)O
z^efBe-t=rNYstpw%xpZX&c+nyY~1b6!c>ne?!k=3i5&I@OTzH@PY7%!A^3MLg!gm8
z&{Dy@2}KbIEDJ|fIWtUfDEgGWz@%v}VBZ*m+?wa`(R_~ATVLW3f9=*$BAn3?q0oVQ
zGH;4-T}p&cU->a!MPN-t1dI+vKq)93&zY58dA{WA+%WXcc!6_CA;`CX&KgrNmYWA5
zKJyvu+ybHfBLES%{n0+skNsv(a52^g_maHfe&Hb|IeH=dpa=JpatDL|9Z2uLgXM$n
zAaLp}C~xIG@Fh2JfO{?Fg55Ds-yKHW#Zc9B9oO@@Z)&d_p0~TAc@OV-_jhGYf;B?k
z(>615gP*n=9;|l5mULG*F-OEMaKY94&Is;y!rjNL-*gBNT5k`pGxqQ|Wbcg%?@dYB
zV_}RPv`y_$GsF(JN^P-}nMT*&7RS78vD()bYoqv<na8ci7Lyn7?zpub7VH#2WaGj;
z0WSC}&Dl6bF1WFS^M>QN3$wo)7XRbCWM-2+0w=igUendn_ON8Hk9{C}b{;d|c)H?a
zl`~Z09gx0GfVVE}5oTTq8sUfoMvmyX?}+bJj@-A*T27H8wx+QMF3J%e?yTGFVei{y
zN9d*rpw8d-aDxC%uN@%geHKYS&S0gZ4JK$<V^PXc?5R17dsc^VS>_;gdn~c-_kK)M
z-H*&%3n<%IV8L(;O!PL#@MbgE8SzZp&J@jxCNNXlgVjrS;9~xE4C2|8k?wZR$zkor
z*BHys7-KQdrYaX3L!-_J2^WpP3IJ?m4YBr~0Zw#quhc$$bj;Dm=~6w++^2^I?pt_u
zh`DCcR(>vT!Tv~HthLg`{xTgL+M$Ee!8(wNW8ckoZJ2PbO~{_jNSEG>`hL7K70kO(
zW~|RhX`+Gk8Nc0{7`avx<Mg#)rK*MLQ#iM{NE561zPpN=XsgsfUt10C^3uSa`Rpz3
zREOl38ZO^e$MJA=Nc~pFaMp2FzE($zlR7T4wi8&WhPQjwxL=L?z{hXE1YhpQWVTs9
zg}cBTRnVHCg50|*7;38mg$*jWUakyNW*g@toNL2-QCGT@(4Xf~*6bUu*HOZg(MrfD
zRz$R)BJU|O*Q`<G-Zn*C%29xyzXF1oXUI?i(yJ73dx8RIz$>7t1JG^<?0hO{h)N~-
zv34_^`DPCDO&ar!U0?;x<hha@&yZGhe5V6Hztfxc@04crjpC9^Y0bA{l1?oq<xgL!
z`yKaBGT&J64$W+zd>SGyp|Q+2g;gc=jTuKFi!=K(ODMOoh)#JI(lbdRdGOhDj=Pnn
z7tl85o7M9R=;Oiynz6Wm1WOAjV_5-BTV6o3?-h^_XWWH46p-@;{_`lFFa0f`1ONE@
zlnd$Ju|j%iQb>mUSXEaF>DcW;I$l)7Jx0~!Kd6>wGT$UK-#jnm8P}5<5|$Pa?@Um^
zl@wYmmqdowV(1+2NGIvU(E#R;2VYVsc0nqg;Mvp-=V&^V98EJe#gGa6O9a7h=u+z&
z(tMLlO-GXH!MtRe{5Ofd6eLl|t0WqIKZ#mQlW1*m0&UpviYB~zML#AalK;b3G{88K
z+`V`&`(Z3y)RoZfB@()!5ySJUXtFyIL*4xNIg=#hCm%zvL!&vLIhuMNM$@DF(R8CD
zioVH+>GEVTEn?rz6_J>p^K9wJ)d;$^Bb>r5S)V!0+_F8KEclqQb~AZeI9=!)P6L>2
zy5EFT);s>XlyKUV&Yyn_r`hb;nS5MCrFIciWF{u5CCn<U$4u6Xp`Dg7H1{a$Hvh+7
z%s11UqiH4cO*`|=k+^6I$&RAYu~C$9H;Qf_i=st5@2c|@QOTMxQaBhv8kr(;_y0_>
zk~}izE{kb(1>}0Pp1M9Z(B~<<kH<UWs#{w~eQqn|l(o>2?e*0BC!cQ0$5Fj(9Nq7Z
zr8s7rx0FOueCs?okv_(84|>9D3ebH^E?eKx&8T-YN$EW;U=8m2j%3m-{6OYSX=D-f
zk%BK|(}v1MTE#nmQy2Ea3)cZ?kdcEG&!AHlYr}el4)ip2kj^t|gLoa#5nTk^Z{a;y
z1I~CeK-~dD)>jSD%=(SaWg{3|HG+wb5k{66V;;X2js3h2<yTDb<+cg(9-1IO!UXM~
zOwiZJ6z$$-$mIP{W9FNf^`;oiGx&&GrcjMD#flFHF*@=Xo+loMOu03JnQx?7znOE&
z29sISSAWBvgs%cj{3JkDg8+R8vwxu54UwFaAL?}-3B2nvG42L-OKzi+Pi?_1%>2qb
zSk|80r{jqL*}G_3aTnY7-^JXC_mGu;52{1%W2V}D$X~n<-|-J&T>OwbxgTLT&#Uh5
z^~Rp3-pFJB&4PYH$W9WXYk?3I%0lenUG$)LLR5X>o%0>s)2Qjoy-U6jsrn(`%@6i7
zZeg~qKj^tX!j%G`py!YBVg9(^>4zd|f2?7?>0!PpV!m-?zA<LL$@ULMdUY@~nQw&5
zHzv$C8<=m#Gv7R5zWHqy1Yc$wp+_M6ZgU2rb0DJp0`Mu%4>ha@P1f;8OgaB|cvhpe
zEdZn+fVJAJ4;cku;@$w*G2a}uW$t0dv0j(}`DF=6TapF)&)KNtbN*{K#(vMnq1<f5
z@&0#<Og7AH(@<^q8oD`g$nOfnNv9AP@;uG8^*PRs4nb7@bIe>4itC9Xn4}p3o0BgQ
z>ktat&=;^M3c-E;I(w7nxG(!0!*hdSTo;DTLY`&ujH>dq2wD2<!Rd%VadrgzW+EUJ
z!gDD^;Eo_1r9;E9&HW|o8DR*0`vN@yA>0M_9D8;JW3f&UKK%(q_Syg>uL;0UU4NWo
zueVl%5FtB#Ami(eF6D<vWp-`f<$<N5JJ`DM4nEf2hRBq?Hr_XptbGIXIhP<J&mED>
zH-~iHar)176xUqG8J;`6THuC$Ctb0ddkdaqx?tE|7g$9)qg0RQK^wVudl@rMmn&Qz
zbBC<Z852yMaiqryV@lXZ{Ky{NcJ|0Lw#TZa_OM{KiG6N|t_M74oneQGb+$N|Xp4Q1
zZDHrix{bXpa|h3+9@@f5VhaampG(qq7|t`S&&))=A06@KlM57nxnT1T=9vEW*v?*M
z^%?fq#XNID-vNtF9WeH)J?<%Rw(wAUj486il416E%X?dAtC@H1IKX#`0F!v$6_w07
z=s-tQ@a}iUO-D?A?Fdiy+&t&J;wYX?bv|=Mjy-p38aQI&a7WI(7vPMW06yvh1hRfJ
z!tpHp-=1M!vcaC~#}Q+36we+X#<d}ba6jw-;>DKO9k(AVChW)T_ZEmeVu5|!F{SXx
z97co8k)&pZCyW1Y|6nhQjCbHBdx^)Va$ly)c1&Bo9b%qMDLppE@x#WL!TVBDB}Pz5
zFhYv25r)7B`%1Rq-em)xBkJQ+kskbc1{DULO-<3m+}B&tq`DP#9b4cSy#>odS-Y{+
z#c3ZMp6%*zcd-tlMcNpqugyJVn{hyEGX~W1&Q$+Rm>HtUJ9?V9`cnhN0uAnh(ZI;h
z>bTmYhN1*Dq&uiVUqg*wOVsc=YXfH+Z$My@I_BtWVBZW47*?rc&?hzY5vrl~oEn6z
z-3ZFnaUfnD1rF*kR^SYrST&4P=lgH6c2lOx%%uwRah!Kk#X5~x1&i5B{9?BX++<bI
zSfY%*zRHNSQpVx`W1CJTm?ZLXP(l*VqtXT_LG)1($u|_SQ(qB<V-@kHgqeo9M(MZ$
zR1_3&W|#t=)&LjY1O9<P5U&Zg+5>Md00Z_{kk_(GYF$@J{@Uf#q{*6%YB@byUrxX0
zl~b@>ISu>8GoiNc6l462L_bQ&Td|aknQziFis`_IBI;tk2}<XglyW{@nwL+8L-J{=
zLpF(*X48S!S(IE>LPZ%R^pyEVm-*%iv(0rrmzhcKO)DS`)^K*Pc9Y0_<H&rI#(a~&
zd^42!W)Sm@4D-z`W*hUWd@}Lpp6*`-6vT{E)mcEx*iSS3E3?mue7ZI_pD;6@W=_qg
z$yf3SC10p!%oiHIoik$^vS|08JPI6>N10Li^!aQGb;%}CwIGIyW!Tee6-P@o-;mNJ
zo<$8#rHM;<24fLTzFyJvgKw(}qv-^{?{~~ipsCa1X{Xz3isvI*{hDGu63Ju>duA@j
zQWfXuO`efJ#zS6F$u;)W_`Twp-Ycr_oj@IS@zkCq;S8D>avS}BT*bTGh$d-We*6c#
z|1^jFHtY?y<(X9H)o60Cj;2|2qUl*>6glUJsP`oiJ^9AB0WV4S#Y<X}|B^<n3#a<6
ztj|z5UD9NIMm3x!MZKgs{lmy>Unqs@hSHarq2$cmbBFn6E%Qw$^G$tLIGMEa_ibdp
z=?bUX77;nsi0IW@5y`I+(YN=k3mpxoM{izIz~EqV>mN)}{enqpuRr~3@+ZGD0W>Ky
zkV3ycBe{LSl-?RbE#IDz_i4^Z{Q8NEetjWPSRN%G&8PH+eEQ*+Pa!(_G$A3MHWxS3
z@HH)D^{9nH4%Cx)YBc%r|L;HhSjXy!A;;IT)V(l{a#ttO!r4g_`29U4GcOK2|C-J%
zeMdX5zoUmS?`Sc-r?+R`lhWd3I+F2$r1H~f%dU^~>K=P_`mrY`N(wUP2B4&iJ2}2@
z#^E{I7|>e>3d?mcocZRZsSaoT>L9dM2gM7vpn-P}Y98sMU0)yma=IuEU`_O~5@!5U
zLZiR{@=pvQV`B&dH$(VtGQ=k%LpX6~j2UO*B)&Gmy*3k^wKhYaQ|wW$FhRAJDS|I@
zw)hzp)E`wrbge0pms#P7<QVeJu3(6fEtd1OuGD!u&Nj2d-sh~V%N<8i*lq~UbMKz_
zE?lYEg&g+xw2m;t|6}Q`<D$;r?{B+1*IXMr)}pRMgBY}gfGwc{q5@(Pf{Fzepp?K2
z-7thGf?_8Y*dpex-F=_W_jmvCdJHqJzzQ?-e4p!F=lp){g2zW(ag1}W!=qNB$D5U~
zbMe8SX?{3=)DLrpu0cWI8XV@eXZ_R&45{6Nm6``Ikv`N>*)b^o7=!!Tu^6*A79Xo)
zabrUq8dt_)NlhF(q~dXjnLJ;v$0M{2eaQ6S{8@AuRpSz|GB^p*h2iiT5P`uS5eQ>O
zTv?Ajn73YuXDLejD=GL&zH$DhK+abMKE738<7)-nUnns2u>!+t6?lA-zT^r8a8!Zc
z%at&EqeN>SMZ6Ms<CHiXt3*ymYBA&+Kk`j9`R305w2Q!0%uP&z`E><)-cummRsk<E
zmTCfbrAPDg+bXfziu^P$1!tGA<BNP#ydwqv9gktpfMaOZg1I$Mk5ZpMih-#|v2YYQ
z=D7+lPOF%)uYy5{3LP%y;$>eli)$8s8fKyA7qZRkOniNjiTw{U(f1HL-alm_r#=&Q
z-Lr7EPc{q=WuxncES$c}?y$NH^uEGAjWjY`at6LH$-{Zi0z9-Sz@E8<$m?H-RkZ~$
zC@+9+S^@skD!@Y5eE2lYhryOSoOqXuT9aJpZ=+_kBNM&<$)JZT4V%|83la+SKbeT#
zpW>0cB@SWCPnt400UbBRVcLg7i2Zy3(>3-(ebqj=yxEK8tM)?c{2nB`N8k<ho8?2p
zq4Rt<M(^8=W+uDwa>OpYx7?1R7h&j^&RLLU7zX{`2CwLC*zj#D#xN)5F7=t<wp(Bn
z9*VU?LeVPr|Lqm&rEErFv(5Nk!K|CzWE=Nj6w*`d(k2)qi-YhqB?xx(3@5Y+!i}mx
z{%on$1O%e%@<7Bb2}A>L=kort*g#As`wXrRM8>y3<o4wLc=I5XJq<)R`jIspgD~w&
z5Q@o6mK~U>GdCEs&j%y-STLHe3&!M;!Q2N7M%bkw1YQfmmaf70SQHEk*UjvP2*Gck
z5d2da0?#I)aIg$T;jU0DE)PY^E1|f1ITQwD9UEmRwgiTvnQbWcl5f(;LtS@-a34Jc
zmwT**=LUb2ZScj!kd^4Sf?b+3JegOyob0+3)33W>*mGC*$GYMpXH$doT=2l$g*$Ca
z5VCIx0(&pPK5Dxmcb&2Ny(8)j91;GVU72Czn-Oy`nVp$2Hyw}~#(gOx2mI7<z}VaN
z7*uYL#!2?D)3C?obXz>rwZS)kYxsK8V`F6v-SK3bX7m+v9u?eRiI#PiFi*5ZW|9Rw
zrdps!lsQIQnq$=;Gn8;|s>u{He8`*ytI4zI?VX9HhBIMZI|COxPRHn%)9`-DH1^g`
z<8#UcYrY#}S-COngN@OVY?9sF7*-dJFd)GQ3)V2R#>NQqmNI)d&lr+wjI7y4m@~=<
zQ7<`rx@d@w73_^FVsCeWA$RHw(S^6iv)SRzzJ*?G3^BUS0Mj!JV8z$NG&ANCb0+oi
z8T0s#=_8ptQfnvaBZ0cjwn{yCl5OgY^f0o89`dSnadfjTQVeu)y@@U=t8~z0vkpv6
zbeKh<gC4iIy3d!D4{Lx5D`4qFAgm9tt1VF36xiX?z`fol(r(fdnclxazVvR8Q@t9b
z^Ir{8oKY_|tsYCmn|kgkJz`e>Ln$_TD89_GF&}(iyxDW0?srGD@^47zMmNMg<(laC
zu8~3AYNSo)8p(TEBbqI*$v4XyS$d!IDe_H6@=eb}x8=)$+j6a6oqXl#!{g0U!qYvr
zR_>ESDtU5v)IDxVo5(s@v!+gtJgF5Q{{8bG>co<BrhcF5#QsU0H1?~Ns!6p{Je+JZ
zw3eNWwPc4{F<p5}{yB0}?!3DpMs+ok_@YK?if@V?Jwo5cT$1LGtHqxh#R~36zwKHj
z!^Y5)lX8^1W5wd-rIOYWD)FsSiI;D&9Nb$hlWr8tu|6d-m@E?Cy+k6~l*sg9WD#YF
zxIZe9$9GDlX+x=W`&cT;SC~1os6y;?E4cS{Ojf=;D#v`eJJq91#uahalUpM8|CGpy
z+r={SeX-c9m&lc{5_y?fB73OWv<xZ{{Uy{(x|N9CiDEgKSSY`B3MGg0rR7O^(mpj$
zUXVRP-sQ=wcKNb(6!jaj&ZAEGQeKoN&c^8y+dW+Z^Rs2k>1@&Ml_O_H<Vq#^rr>hE
zoFd;ue9V)Nh56!0fAXuce7V`DK#bQF$VL8HRC>N#9F!p={?3qx?K7ki`DP>e=9{fj
z)E+72=BgAq<eMtFx6)*lUYhLOm?lF@)8x(aOLCRj%L_g+k1*+$j6$s#bS{?;+sno0
z?seI4{Fa=ZMSk_-&u}ICLqDnHKk{3XF~w3EUo6YMmdg0zGWl_u%o$NG4~EivbH7?<
zjA5UJ>2a|sJ0XoWpA^#<C&f;yS|YqpiE7$uDZ6w=j(MJ!?7J7F$+f3aXY*W)3mTzk
z(>73R(GHtyW+8io8S_m!j~;CfpHb$_(KLr<k~tg>QqMKl#)oPx{4Ujiwvz_t^OWt^
zz?_SkxK6&=z~^8j=itRIb_gH9-Oc}OnJsLGgJpKG803H{#}^`~dLi!LUkKv|3-JNW
z8|8C8W)i#Mc!o|jKzrs^d9OEx|3VK`?eM_jQV;eGF)OU6C+<!2gvEDHXue|}S7&?F
za}Iy|ggs8Sbijrow#@Iag4u0z{9|bWQ}*HYcwq~hC7hw@>_(TM-I#cOHx~8g-mh0U
z2At=0`{)S1m+V20<9m=*6@?pr#UPJ-Q<V~nnf0+~sU3%2hvU$2JPvKI$Kl#RvP~NE
zRIbJ&kzV6w;|}8k`DW?z1k|guXSysLSNlaEdRYY4oQgnRn?2C+<s3?>L?bsPcG}SY
zO1^pbMS(!_O(^+hVFQ`wp@K6g1?ulAP))XJSFWHBLxIMn3b>XiFh55@AB6&!d6sTf
zG9M)c1`cTewHwFKRCWoaU_yKf1_Ue7|AGP+ixilZuD~}91-g=NzKkUAXt6JQv=RlQ
zl;|@?39mz(Ta~3?C_T=FCaL((EESbDsj#z8h01|Bz#ggS9YK!Z$=sRB3{-Yky6|z=
zRQ3j?V)%zllswJE%V(L)2gyW_1DR<4EDP$FGm*A26UW|W;oqxSSgMwVEA-3Uotl9z
ze`i4Vb2{z~VAn=e8V=n^g&G+xa##WO1{dJ%nF5?iETB)K0BJ)CFv>9>%5QnNzc~-D
zxEHQ%kqaN+ER5-&i9h$!5jQFgTZg7#T0t@fmn9-$NCF0~jKk)RvB+L}2*dRcqGMe&
zvJ)Z^`e7et4%>(Jd-md2GwL@p_FyD`{%&7)<8tLL^u4eXP1$jIat^yIFR@cwm-%^j
z!_aPL7^)_Oq5t)5?1JBhY1gUUOx=nb^x(d2xrJ{Mp%^eQ6mR1~V9+T9KQ?Seb>q$G
zSQ3n+0P=)YFdPR2qv<oUNpujsQ$aAeABdPEftZmHhzA=2ac^!Q8qWxXkx?MBXOVRl
z1~N;Sj~xy~uc|=oxf=-6_tbZ&_guOih-se!ktdv6)d#`#eGq=V2x4w5XJ47Yn0F``
ze=le6WnmDkp9Y{jC<rskn6E=ma}~9u<S)UvG1~_}w)^0Wt}j|%2!@AQDBrU~kyjRq
z`P7Jx@r<enMR;;3f&)TvjeK)2W+RUJY{dM$_2?e67Aw!Jf!dyxxNf-;9-X~$^ROqp
zzj&Ziv*qyGNni1DH{`{;;%kWuPI7O`YQ774HgZA26K7b|IpKSj6Pm1ZLMKZnylcZ;
z;xUf!e>4X-*Uo`c|2cSco%5)T4shw@fWhJR7}M7Q&2QLa`XPH9@UX{ATRVKNn~mH`
zYn-;W#wL4f=3!Xl_dsj3`C<jD##XR6Vu@|;mI$e^fXgZi92jAN>PU0!nPHCJb!M2h
z*$it(n&EWHEbQ+x3x$fAn5sP!4{yzYxXi%6>{%$*ody$XE=`2zBQ=>)c6<jJ<GZmj
zzBV$($`eMovX@NbZiL}xMtEszgy+0JJk%IPnnrNzY6N?FYkbeK53}45$z&SqJVT7*
zxtVJSRi+_syy3ogfdMi(j|y2~05#5|OnXqjq2H!QF!ODU^>J#XKDLl;+)wJ^;}$(s
zPSZnZOL}hTwaE_9#U(9Wto^Km;dwf6T~2?^a2?El2Arl=)6NqZsSfmP1r*#DoG52K
zbfVy3gkZ!rL3{Fz<Io0aGO$5($v4gVG)O7=W)u15UbhBmqa@=zsh1IXkEF)*kwjTP
zlzOv=vSjK*sU7i9hX3TO>c#^}IB{Q`xGVkPQ;n39BaU{gk@IAnH*RDbb8?PmjSME|
z3~zc%Oq8|KD!*1{Jg5~%GL22oI+^o~y3e;;(wOHqxu%dOgXgN#EeT(GOFBE$$@_*{
z`9Llj@wQHWzpfMID{4Fs$v)I?9(dHsqS3X|o_v!>zDXzF{P4ad$IEVtbI^5JdXjt2
zYc5JN_Ffm*)yfyLN5F>?IkP!m>bGS}=IlI)?^Ga{D+*=blOp*~U&U@pl~m@dWVC0o
ztlw2EZk)?ZeO4@khL(s`3o=S0o&hCtcQ5BR)g^NIcBwf0FLRGO+23-@q|J*mNjrH&
z;!YivmsgI;oWGCC)yJjcVqYS=4NIi8QL$LL7mMTeVtIeHSWNi1e+<3D;oOf}QzBKf
zOQiR&VtLE={nIN8<<;B*8JLhSP2=*!C^k<r)AG0rmM7Jp^JHFM`fd8<OW(HSmB=(X
zQ=TRx*QH7KqEtEdBt>R~XNy&8wp8}dl_2+AS@a-RUXqbcTjWdd`g}?Ho=06RPmb$n
zNJotfNmk2{%z+s)x_^euZKITeJqpp^tdQ$tl+x*fQg(T#NKAODc(hNGj^onAXnvZk
zj7k%?%rr4fPm_OL(q!|}G?}?DO@g<k$%GE+@?&kfn9zf~EWSvtT`H1et5n=?DwY>(
zRkAisC1(4ICA=|r*9)1YeBg*g4=$Hjv!n9<pJURb_L$U-KQ6*KRcY6g^61k^v7cQn
zV~3xTp;o7*P4HPcW^i6gPF|2;#S__(^F(seo=W7iMhH`5=c-x<)E#24YoqQsY1ADS
zVcoft+nw+GJ?L5KfemAO!f<*|+)-1<tR5OTIYxumLK>JFs)6(z4J@nBMEh!en2#|<
z|JUrXqjucJj{PswXJfI(Y&=+MgRXOI5WK@0XU<t5S(~}y<eE^PWjrl;woN48SQ;RD
z1H0%dr?I!t6g}fiakBMvj9ob$({D~k{{$-xzh{FL+zI{5b~cWB%tlj9E8I;pN9$MS
zXr^v~<*>!9xwg2l&=$`(*up8y7E2;+VV7fz$@R=T<?P;RupK-{>_PeXJ+P_VgHs>(
z;C1u8I8_;g?iXY5-7}8)WN|n{{buZ~IO@!CIKV9BJxAhkow<L%M;^v?=IT_g-HpNI
zo0N=jylfN!i+>`}vLXWVJ_7T+*%z=&iLZ<3!J%%mipPe0<4C?aM80WDzFGS~fiC2m
z`{bKH^wPX9qaH!72_>hzEl?mLO92a>zC0gzzgrh2-VRE^KJH{q=6RZwf^G^Wd;*kc
zT21edip-R&fMucrH-;(r+NeNF&cp0RD$&5xmV7f)lip`-B?e7bqS;J(p3Rjo*`vfw
z?t9&COr57gDmt`Jg=^bXgm<S0x(DZ5cB$~TW&i17@(wkl&yO>)l_&RFCT@ghVhCqZ
zoog~+!_2+!=d+-5Gz$i^sq3g^vfDA8y^Qn@8M7y~Dg8uxN>uAA(XF1dyH6=tV3Eq~
z$9&kW&i_B%M&orpzD49i(U6B9)Njt`<)VIM4#up=K+=^&q_<3@r|B?mWpjr=CJ_d9
zi9mb$qvpn8_=6bum><I5Lk{AsM5FviB>d^S*?n;@+$ZiudfXnIsE@$M`fz5g?nci7
z?n+hf#GBtc;69r@+sw8Jpl^8O<uF99B;#}mL$9oD2+-UHRqj^&?7J0FTeiTf(H1=1
zM!xAA%B;B%%xD{e$v&GoYoq6etg(0nc|warfabyY?{pB(QG;w?mS4&BK>SezqL4bx
z3`cT{7PT6+K>RZ(5Ti#2qOX1+TFei`tH40`Bn9F_1^?_?AWmKhg!$<}W`GBx`?&SU
zi}A<qkN()wI{-t+QM=h6jC*RqI6fp0Ya0b3`ceSKFQ+$|Ts56@u{ABJBmMDV#?xvv
z$zF+}YgeM>nU!dNVI`{TdAhAak?|^o_^yIYht)Vfd^NKRSEFypYWB^q#^n+2X#Cg>
z?e@B%pQRhx>$@VXnJen<F#Bqz8~%Q@3=8s>K{Lx0F>PHjtI`FX=D5JkZ3%K(aBs@Q
z89kaiBk!&g+LbsVdW{qQo#}+3Eu3(#iz9mo=HQq99Bh2+fcpvuw6Jl&=__^^re}}G
z^>)x3ZjX7~3%_^L4&zVE#(w7X<vLj7?ig##=blt0I~Ht$te6jBg(lA|Ig7J|LfaCZ
zxfe0i-2(UfTVOx0)gJSj?d}ybXm2#bwgF~XzkL>7Uz>?H+h^j}KQrNceFn@tW}vQk
z8q{J;Fm;g$%+yR^@X{Ek3ypD&e4}Z~PKEDAC|4OFH<TTi_C~1EHv-$O(V2{M+Kn2{
zAR|QlGDOo`hL}=q2+cA>yvZ?y8c#3kHsg4Ev@Lg|!~mAA2DoWwfPg97nQG6@%m?~d
zM!(HzM}79nkZr!`!RVA8e(qo%pP3%kwAVwxSzW9P*2N97&8PSD4X5c~;XEB&?M~h1
zBJd&<@G$_oHv_l=gpd0LTJC~9h62TK!RPLRx^{xo%%JH!tU=n5Z=(ABKfZ}2-|Qsc
z_>*tGCDcoY<a*}M)k`zuNAktyq1cme_K<I$l5fV5Z>ExO3dlF5n%AW4BeQ<VH;c$O
z-MZAs^tm<6ajOwG4R$oC*GScr8ks-2M!cPBq&Me8r#scjj$ia8lW%PQf4-?B*Cg{a
zC);#c$Zi6A&Ze%{ij&`MdDP>!ocgy;@?O-5{xkME(ogfIrcMr>B<G~vlBx0Rq13)9
z%3(KUf9^#|K6gq6PO6fpUD(}y?y|HGyDDcsaZkv%MBX3E6E&-BS+*rx0<807f6D^t
zpIazDZx>1b$tnq5qZ0R2m6+U6iPnx{sZJ`EKxz#UmBr#i&1LJ^Vp&8+>9B{|&iP`g
zx?C*DBROB;%&Q6ako+!|@pE!NE3ZuYHz|`~!%_*pq>^y~#j<%+vD|y7l8h%RY1FS+
zH2jL?az4LC1oa#GeqMT($nBmba<rjX>NJWaZ(M;)otH1}S-CPTFIPf#=gGP~d2%<L
zOp=}_N69B8&C+D)ja0dS6q&7&B8JUVq{S(vfHS3``;;;~I9qzG$QGAZ*>Wa3N3?Bn
z*%!gjqn|H2sytbhmnXf+H_ga5I^>&96H;YFpi&0SOO_!0WLY&*A%FE&N~V)ij5?)A
zk3K1~b5x4F>zXDl4AW$6YN{AKOO^et)8uWdG<n&ODs8m*@uoB}ewr=r(fQJec|B*J
z6v~IUg>r}-IbTmDd44Ln8>5mo1;tWM57H?9dwu&<CYW+Wg6@~g1&gCHwB(pfxmzK(
z?jM&l`zkSG2k6b~Cne)dwJckFT3#<bE1Jw`?ECqGl&d|FM2{!3hkJl}Ya7Jt>QmAA
zyEPWHYll9&+rz6v2V8UPfD79@VC;s@aM{oUasTv$-He|2I#?Zj+iT#*Fb(`M*1(jF
z8c@&IKu?t>tj22N<!A$B3@|`Ad1V1Lnb&;I9-3?bU*3=AIl_-M`FIr{_vU?1o+sp*
zf&UmFf0zMkM;M^jcxI3r8(_c}AYr5q;^^~-Vir1_odvIE)Ry1Q#8R(WP<uQJx~68B
zzto)BIOdpi!yIe7*`k$>E&e#$;yd}~>pEMU*=CFB<eOv3w&*s@4u<p(ZQN~#rMw>O
zT56A$Bj=!4kNc=Lb1-LQB*t8g#HbYq@ggV=AFJY!csmX!-p662A|5+S<8he!&6EL$
zQOx(o1G{$OSPnZFCxqkXZ+cb7Mqs>a1XA8aU{&)yP<SYD!cd7NWC*+eG0tb|H}B{_
z{!amyIt7w8#xmFI5Uz2Sa^}?`?0s<v-oc0PGUyO0nBQ|{{~_ob#o%;N930Y<@Ud8d
zQZmrjOA6fPDd6dRUV$se6&P2cKn=O*9krf;G4wH$Z~VzO$xW2dRa4^C2qo@KQ9_@5
z(}a98MTdSJ18Pf4lrS#jEUlCrc1($muh>iRTM7TBDF|$nf{mS1pxY+}37#pq%x<u@
zt5UJ-cPg5YZ#1rQMs+;{Th656bM0Y_y-%J)8tP7^<5gh>wtwPWN-G^pW~F0Tk97Rl
zO`WJOvw`@%9G`y}%UqbTNR7+6PKouml}LT7<n=c_Nn|~Xf0L1WE*V1;xtF%+Fk19w
z??!k6Hao|obTe7zcPus^i9?*_VT9PmL$4?nTmL<Ts2>NR@jV)snnq*Nqe#r~+lOC^
z_hQ2JJ$RwB2WPfKpnqIA0%CUK7=2qGi+95ABl`_z?0`@4c4$r5j;0l180`>-Rv)$@
zoxbAjUAN)QuC3VCgnc5uTafxG6d}|-Z2B<YH=6s_Ejc^o{AcHjV1Djks5mpSXI8QM
zzs#^<j??S~%=7EO+v-3Z-4zJi1=MS%1|qOKeKzC~t)Bsy-zpG^0|L=UHxMrta&J5&
z5bx+YPAlPKJa_(H2S=wh2<hmLHbMTh7x+_KWxn1Hf6RUsfL9TL*x4)qy*F(_czOV)
z%n8DxH9>g68Sn7h?7Ds6gGYa_!iN1TvEaxG3^}<RV|Oh_3(MtjVot-RP0O&l(=w=S
zamOZCcUW)-%5IoDcm3Q^c+ZXBgB#{`cf+9sSJ;XxtbV$n;;IYU=eponQ#btX<Aw(R
zB{0i$L9?|kD6C(C<Zp`+7UYa=W1TUni8D^$b3!@!X0X2#DyKW4hmIrMPtC!CjdNi4
z&m63K<bXeW9WdO*9$}T-(d2WrUsF4H*w03{KQ<UX!Uk<S*r49m8i#eQv6%CyMD|G8
zEV4rLIqd#UvBau_mI&`_$?gd1Hz5{ye%BnSndTVEy>N#K=2&yt4E{c5>|URR_tvwp
zxN0WW`Ok#m$eFluZ3a@<Gxc&LyT4bOV4;x-$~&9j#eHL($}#3F$QUzb8sqsTvdktU
z{v3_a)zApNCmCVdI3rZ@R)3KZirO1t{BuKQ=o#Y7F+=PvHbnn)Lk!O{L_nlIiVo6G
zbDpQQ0cWHJ=+K9r;$O@pzNHV7G-leY)W@yy`piGlW9Pjdc1AF>&sGmBdgvj&hP&R|
zbrGSj3-vGb*re-VtdkDHIFD*k0r-0}PH8CMbWboSTHrZHuwsxv^Sw5T&S@hqQ=3^i
z+Q{CejqYj<vf8{sGDbJZTk?$>^_vj#&0g|NYic+4<eOdt>%}DKk!;d`BtEte#fTjX
zJM<n(y9p2FN$3M<vEYH|k!>{TaT)ZmM*6j{kzMU-<R&$nRr)n@i(GS!=i$^EIY_=4
zL%y+@$Sh#z8gW<E%8?Fra^op|sE=#uTdtL1QMIDaZUj%W%ks(Rvh2vKmE(77<$3)r
zSrmLtF4+GsL%2pTyGHIAo)&L<etVy<lJ6;1vLdjGoex#=Nv%q9hgM0#NZxN-C5Kxw
zU$69%SoOFn&0dtqNskio%jJB>GFwVRvSp!Jp7bx`oF}P3{C*V3(0PS2w5Cwn%q<f0
zokbEfzepBxPw4PqYAq4P^0A^=&c9X3Kf?KvQ?Y0t;A5O4^}AOrv&NB4hVwnzqfDHa
zmx<SkGC9q?q*CW%v7N4xH4jxX<+)0(>{Ch9Y`(u|a6fyYN>-~C%WclWF2okg-9N=r
z@NcpFOHHGDzhc?WyqoDUxuUq7BfTbPiy3mnsaLk-1!jqT+bnrrk|v|omGYvgLT=qn
zme%)@<-q-9d2~BjKEF(s&BK+lFrIUzlj&mdB3o?d<wz;{MyqL_3>cRu{hs8?-%E2v
zrI#Ut$v01<Qsv%5rF1=?Ec50j%NLDgF=(a`-$;dYZIdFw<eT@sQ{=$N6mfZ%DgpMX
zvcw@(oC{LL^LMJ~G)@zbo2k-vRGJL*OJgT^w%m=PKS;Mgw*D*-Q|@h5UM-TQYRt}A
zsS;Dp0?%A3l_P(bNzZg<|NK`bIg^fv_w{m_YjjlnwjL9!q~mg}b)|gYR4F&FoRB+*
zPD;qSYAJhvN`8(wE2X86#niV!`t*JxtxTTCvxg1xYkq^QwP}!0J7ypiw8p&$?chkh
ziJR8}wYxhYbya6<A>XXnF%fs2C!_Yl6gVuO3iTUP(J@jTHa#`4=adG#6`B|}RU0}N
znK|`cA9Z9FBQi|)Kg^>d!%X6_;t3(&9OvVUcv8C4KX*qD%{J)a(px<Y{HljpFWCP!
z(g4aUz>@dC&@npbQmq54eY!BunU1A=?HT@LCcgWbVZv!MG#y6G#)JEwoS`kfY>tE7
zZQ)M7Ib>%GO?O)uvY%$sR$Fu?-^@B}ixVbx@Nu<cewQ5_I@n|D5__E6Y>yR*%tXCt
z5A84Zcz@O%rVl+~+4K+^?TkSmop?<5ipQPAcr2@8uJP*x__aNZvvE5RM-S&nz4Zu+
zUJo0Oov@GH$=sct*j~c?l=7WubY~X^zu1Khi<8jzcmmGnG8=hpG#Um)!|xXJ<o<+X
zn_oEl2?MZqN&qg74Zt1k0K}UH;EZ(ubJ7AZy>B$XMhp&bO2EaFNq9Ca8CyprW6GFh
z{5v5TmE7IxPk(de!el&roQ$uplCgMz0^9p5@coT~J$T8uo6nqOGDRghXAJq~F=tj^
zpu}bJO$7PoWPuWW%l;qV1fEf%_Nx*T{>K?DQ|Jdxfm2sz_>4(G<Z|Zdk#8b@rQ%3V
zCW?YH@X0y_ek1wsb2}bBotX`E_b|FYOvLh*3Y_jnZW+UQ*9Ur`k8#FzCl!zBIZElS
zq)+59VxKTOkr~lx4-)XXGy%~G37F6+9#^QvJU$<b#bvQrl^=`moEtVAj76J0u`u`<
zi!*+4So|awDa<F<ycC0RnK9TRv6#B^5K?|eW9`sr*p@}1<as2H%wpEjn>`rqy9Z;Q
z(^ott0_U8=vB7gUwrr%|CTS<Km`!ZUPVV2S+wo!0cC5(c-jojWZaABo!P%7R%~m9O
zZbi4}%<fyX1wWsKLe_<%NpH?})`eio>&*z(-%KA=F#C6?rOe>2RL5Yx7ttFW!JNuz
zLD=*?5E}W+@$(79d_8hY7kX@7vBT+R0D4>tfMI<A&ix9&?!Vd7V8D5l8|PI!sPQBQ
z!m-&pcCM`9b;BC+$QqpLzXra){jmOuADpVz;AAQvd$0yWBG#f&@3rXXz5!{|{qeRF
z_q5gp;%WIN{G;;0hzBbX-h?xy9V>9XdO6&-F2{zm%h+GB459PfG3uBb#;$flri&YQ
zhTNFF>4xUR+<=a52)W{l;k#Y2rj0Ai&$09S2Io%cE@%|uf}#JpqJEbv8uAvS%x4L5
zH?n8SYYFCtEJn`<&UiJB+D&6;=#p<fb3SFT)(IoM9dWmtBbussUEw(g`#78Wb&y`0
z#SSoULf=iA9Xb!NMbbm|O6kr<{RHlS|FlLoeQWHSWexSOR;Zn8h2Pu--(g{eWwn;L
zLch%~H%t8e$pR}AsNMW+0dKx8JWDi(+eUMCO_<}>eKYKyZU&n-vyd}=CW?Q|V2<@n
zlnt7Rw3-=+Y&;#8*zx_2=g4sGbpN1kQ)A5U$Cx{j#;D#%wuv*sqh*{&SsOt)#t4>!
zjj*b}5vB|@!lL0uc*su7u<M3!rcP5(Vu*NpZ@kp0<?PUh>o(@s?A6EBeD+4w>BEs7
zQU#Cn@${5FULDrQ1Rs64>#`fOnLcXI>EQ!+!2g=7hdTrGIAhYqiYQ$andzb}{WdT1
zb#P{Z4n}p<p;wYJCVx~@x&!(r1uK081IGyRUT9-QmNvruwb7nFgFIe`cI>5%;11g8
zM!xwlg7YZy&41*Zq2!x2<Qr%5ja6*D%s5ytG5zc1pxGlyi+m_vlOM{fo)0Bcbxj&#
zuSpx`y)}PRBc_j-XM4Fu3I>ol$Tt0Xo>H?}qE{nl$Tn+uqRBT4$v0(_YPgeJBen0Z
zNYdLY;<>z9Hf^}b9hr;rAI~|}MH&B%`%8nWWz^l%@?vn6C@)n=8GX7&bC~t(R3Y9Y
zk4txr<Kpyhm7Hl{U*-jNLMBzozphpC&a6s4kZ)!VA>)jwlKZWzWN?#H5?pji=C!#h
zz3!H9msllw!*k@0QMPDo%9c5LIg(6GB(yM3?%mB3ZM%Fa<?hkD=J_(D3q85Eg;Gq`
z`J7iIU6`?Ve1u9i?NQ0Hp(>e8{b9jLl`N;Ov$(ccx?Q5?;#w?b{>3tYex0frCGvS-
zsVK)4%c)i>d0?TEazmAbk5)->evzz@BGK!pl4;v>#Gyrw^lg$OL*A%l%LkQcZz`6;
z_Qf*hV4>Xe%$3`^Ibs@=DR+)$N@7x`Y%|W3M$0oK&08Vk8<OR6U9u$INtT;DraU)p
zCW{t1=M8n2MzKoynU*eocj?Ko$&uZga%IZrT-m%bS8|hbrJa4Q>^_|%zusrc+Tav%
z(pJcyv&oXTC|QhKCQAmnX?R$&eEO}FIP%S^UMaG^CPmVQD8=WKLVBx|GKLz?tuLu^
z?MJHo-8EG-&!tMoRcTU6?{RaN47toa!i(gaIQJr1f2BwiLsX)gtCHVaRdRtlWV@TQ
z7i(RaoOn<sQIn5IL+TOsvL6+Vn@8nM`f<72rc$(gD`j`>2|2&_B(tZhW$}emB75p(
zyk3J;_&3O`LjEpaZ4m4726;WEUfLu-l74L;iS>Zja^Oj8lvTAyWA_e-i0Xt#-FhJS
z%s=>EGYMC(PKJN=6tpp)iiERMad@gaj<nOj)hZ1HAJD|~WcHkt@wxh5A04U33?##}
zC*P!zYgUajz?^{w*l<(e$kUFq;Gf}ISRz^|RBEB`BQ1Ok)x(7D%pB))zVHH&br1A}
z0^?ffqM#LhbnSF;u9GgVcGHFP5M6d|>cZ-cE?zA&XMUJDd@q{g`bTq^XxgIK-WKI<
zwpidt{bq|TS{&fV@wU*jvO_Gr&)R)vqgj{PI6KcC|B`RUC6IBR+T)gv19xH_(3X1o
zjm_SOj9rOCtK#wVXFQDZ5};k3fPKFbaERF`QQ;wIRIm{S75=cf=84W5JrOkB6Pt8B
z@ri!R0V_PQaJMI}rFf!RkDg|Y0Ekfl`db7bnLKk=0<epZl}!vl%qX(WqV@RWzaI6?
z)??T8b?|z+4s*V3fLbH&BoB$gMD9DcD2zpwYdlQ8#>4M>JVHh%p!2u{SbHbpPP-&r
z|0@Yc)RLGPpM*)=S0A`C85*0DF=KBsJ{;u7y%cCeU1#kCC5De@HV--H%xonFaK2?u
zzG=)HnWz6Laf^IoLB5$ozIoFw1(xKSSn^GC?wSAePldZ}1}1P1eLZ<fliqCgFLCUR
zipPU@3D|M^F#fxgh$ydQ%nMMUa(xPR?<OD7yYp!q^_$<x_|$`X%<qK%>%ve!MW*4u
z>x<$9=E)|&GbjPQJrcO&5sSO0W1%`uueUN5P02SV5%hj<iN)o^vCKn@1;)f8{CEs@
zEr>x(^BAlgcL-W{qaoE%eD8=v=DA25qlbLw>OI_z-2;=z2*kV!N7cY^#Odxv3+G*k
z58DZ^D?5-nY6qS~ZbwY#?RXg(hM-~GyDr{_jYiw>{o+<?a9im&+k(<rTgaxNXzCq`
zGVWOK_h9ZM`R0xKW+c#Cd|+KLTB`@++t(mmp#None-IW-pwIV7Aj-KPWzD&i8+Dq0
z{|&&4$^hI=4S=OG09Q)+J5d*aoxcNcgmb2s^ySp~1|o7X^Kcrk!Hjlm;P~ASnZ4Ga
zs@WPex#Pz@Wj}bR`C_}m2a}ki&^yuxt&@DPXV4nxP4valDqsAx48)^De%RW=7vnTm
zV))J#aI{;2s0+(c_Fx%uE0$r{%w>q2>yG%XZYcG3Lx8IrOsw2sNxnH)?+Wj;uG9@(
z5zyBak;h!n*4G6Ccy4cWK@Ppc2SzX9*INXq=ZoO)u^0^p7Q_D6VjS<b7!SSX!HfD$
zbW>-{zv+Yu@{Rj5N9-lv-09$m&&hKT!pxf{6%MG`;(*(;Ih*S202TKXVwp#=Gj2B4
z^SQck6gx7*thps+&3-v+JUMNJjcW7~U$#Wp5Gy#|u*A1GOS~_&;I)ngJy{lLcFG*f
ztIRQ<-W%I>=1?%#Z`2NUO$D05Ps0>9I-4Tn^fa8BHVwv)O;Eks1k>lRM}d1z*T$J(
zX?yNIJuybzNn_04#(k*GM(9qy8R$<xF+If7`xv2ZFC$oWHsbDs5nQR;wEAR-U9$|)
zVXPtUGPAGK9euP@uqP@;A2-Q1CiK;O4`l~tls?vF>mygCk6jsLnFulscfb|;d=2fW
zkJ~r(@F!M}d`+$yrH6^k?W;(kb~9HOeY@!5SEUYqdg|anFCAD{0k_<N*PVdK5<%OA
zf~*dLCdak0&rch*Q?$|dj}}&6)k0~u7J?$RFmZ<#T99v^l50%=Zjfnx8{|!|2Jzvk
zimaEP2kK>Y-+Bo+^++<)ABuu}^K;}waqs?69u;4cGw#=9S<`DW?{$srsI8H($QoJJ
zzecu^ZBp9S$ZZ~vAvMx#G-p)gnj1W6)NW3ZZ?2D{rbE8jO1@Di-(<N|i-Tpg4DDVm
z_jzjiS4+v>6Jo#}q|QsYi)2&5K52S&9V*0&d{eJ>T%M{Qm&C_avj0&PGvcb`LSmKp
zyHv^L8C9Y+s!HkyRY_!@DzSWC$qvUVS)X@a&aS^gt*Jy798poH%8@$5Y`N!^EyKwT
z-v0En(9>dlAy4wY<VnD9dUu=BhfB|H=9qlBm0Tz-(u$aATqM8ys$}!qA}Q>x63?0R
z>;y7TCV{(6m4z~iv!_AR>6uwvEOGSf{5PybDi@WA=1-LvR~1R?Get6>yGnwN6-m$E
zg>oaSP*U&bOF#1*`NHo-t7*2JXvmWL|7OX6A$j6BFi-k?%@udcTxrpXOcRkI*~2nq
zheM|HotP<y^2j#z$zmU#ELK^`vhY^29O3y?n=Co|lBLlqGR|FvEHp@!b7fhwVRN=b
z6l6<cj~t1T9JyheD{I%~%1Tqty()9WjlP|@!Afz2LJpE|LY<Rkc~O#NS0>3^y=2kN
zNRlT3NwVi<k|g(1NCCB-3D1*d?LLJp_^lA9BT5l9`j6kGNX;&KbgEM2$-y+a)Hz*j
z#-xiuTDIsv$l=T`PZo_;$)p7;xfnt(PEG0m&G8y}mq`L=dp|gz0&@`E+Ly~K-*OqY
zw_H^0KtIs$sQkTwe(n0B^7!^i`8Dg5WPdv?-HjXM8u`XMw?S}`dx2nX_K{n%iF%-V
z$2#c|`dYrudMif{zLSTp9WZ`iN0?pfj&QG@h`uukkE|zSWcU<ppE4C|j!eZ;&Z>fo
z)bX%F1OI;4#jPYgMEmMMo6or>dOGN6C@7pP@ZF@1`?s~Q@4gl;JtfCH*FtqGEu5I5
zg+o2HxcjUHom?$EeWity!Fq73p)N}=^UfP!_AKzcxh_1(H%ju2ihSeQRTmlqbn)L2
zT^L^B^PFGXzXkUS$TyMXn;ROonC-yqcJfV~uPuIt+G5mxTiiFJcB4Vh^G!?i>1u`6
z8dmJav_>Yio8&xuT>oZ|3;pI$Z<xb;Ir<#!=0M5o>$wkuQ2Fm>EFQKUBgSpVLT1`N
zIURx#J44X9a3lS*UfdD&MA<-3_z&{L7&T9{8%JZGqbG(qdtu`jFU)P`jkoIFFk9gb
zWtcZCa=GL6*c<aaR^oi0bucB{jOn}{&wi~#)tPnJdT|~6=(Ct@L2W2tJv!3!nm#Ow
zX2B@9&x%93O&r?HkHc!OI86E;hi1d$VW>>Rr;J3L&q+k%-AVYbOER)nQ_oREA|NFa
z<#~~KQ4<MOO)`>JDNuM!0iS=A@TC`~%X}r`&MJ||Inj_8ynV04M)Hj@`6i8g^So~g
z*4a_VA=@;P4D8~pY81cD6^$5Z#>e3I9cBc@vD@7sfxTG?%wc1Hdpl+xiUOW~)O@z4
zuun7<-7k@AJ|-ccD|;@!C!l3*0>+bXMv-r-$Tzn9yZgyETI<-4>6Ad9Wh_pejm3Cs
zIA!D;cQQ@d=2%Q#8;jLlVo`fE2KRzvFwiCjTR$IyZ;L~C@F@xb0a5UHbO5Ke9l(xP
z5oq3Z54N3)fa$CV^r2Vk_KV%jG~A7D>bsb=wv(ONJ8-ne4j3^LxaL<F&IX2|q+J-k
zL~p~*q1%v}wUzUNtr(lX1^cFKL8p_UxH_NR+IK?OAIhGoH=FT^-W!|TU|4Z)%6M=v
zxgZE}NkO>A+0>2RL3qdPzLT6eCF=yjkPI{VL;zOo4ZwBZ0BHIIpvi8|m<q`{_jsBI
z;^#Pea;W2cS?Y(BbAD(?o_Sp5hj>3fRQ>ivXo4>)%>6L>y$?DV`asvg2m5yVU~#Mu
zN)G$rx`hv(xAnn}TdU#R+mAgN<RAXdP10KdlYT2u^=dheUSEdmyXl=VVaNAEcf1X8
zgNl5^z)-Xz-|*E1#pIjScU=)rzL`aT&DVMt%nfpZjiC$PpICyjE-vW*c`@czFT|vQ
zi_qVC5v&x8@S%tqpSg=*;5rX`eVpOl-5KX@H~~|gaP5sF?wL4Z^=L<!GPBsqcn&5i
z9k9@sS$;+iSoO&sj$U?H`)W2??wpOJX0y?jv#Dz*tm&7x#-m2o7;w}I9<MB6Uto!o
zi!9+DZ-Mon*}YJ04u8&~24qpAVK%8j(kwX1EZno637h?<@ZHRw?`6y%X7;dJGgElb
zYtw4nG}IQGK+Vep^JbZFMs9-VJxuWT2V*S1XpF@L#^~^`5&iB)s9(=Mg=t2(+`$N$
zt&Q-kxe?C%rg!)&{lYw2=7zA+VZQ>ihYb=9;9tuw??io^JwR<bL?0UKm_h8vJev*r
z_{#gYc*d^LN6K7%?hdjKyfg1V)PrxD9`3Es!_=vIFyw38h7w&Qv0tI}FkKA4se=tc
zItUr1!x<rKVSRuef6+r+DoAt`sQ=MMQ>8ZYEw%BdsW!GA)1u#83#Tplx;k15d0n-z
zzoizU$u#3O-VytZd$K*}o|Hu17Z1OB*}&`J=>7GgM!s35dL(<(9*T{|Lzzv!+1l*7
zMDM*O?I&K7gWqeU$@3bKyc(Igw}$&gHKNzLMrO3CkrQod#Ff0WbtFAEWSgEm#m#EO
z?BQ9Nz}!6l%V*_VH+CA(8}qhrwG8f6EhD;d=eKFKL~T7Gt9Mk$#!VHXORh;L-;~)_
zNUeQ^IM1(;P39Hyg?~PL+Hu*R%xsyg<I-SKDWfAR<;>qz<c})(LbhqryGl+sR7&8h
zN@;YxQq+!JlIphf_8L$_X_F&02HE0imo4@ubLH&PJn3tcCzt3&Imz9k{`9o8Z<H^3
z?YURYEMOP<R0=t>n(~Z!y$6e=b{FSZFAKz03T4)sLJ5y6lz#LlxBO8k>HVp>^yMDb
z(PEj`njiCPzT?bmFMFhVC$S@8U6E8d6-kPIq3md0C>LV#r8oV*w#b$)-?C&~B6XQn
zncR~~lcwHj^3^{@p1Y@rvMxg&OwW*)8`5P*aH_;@q~CXjLZXY4C1DeJ#-5C$mMj~-
zCrO+8NmA1-S-P4ei|09o*#Dg>{ztN8ze~1E&&`%^U2-H{i*u?ex#G7dM>fsLmh)9{
zk~$$;EF7ZcS6Z}Wj5;W<tPjcC?s4qSOcceuBzf6CNrtvfk~3OKayu+Z?lei3UuTlV
zF;pSy-xYFYuTqi+q=?$G6v^3=Dxd#JW7bfPOfbrk=jqg^I_FAipImtzN4;oBiELD-
zUx~ZnHv3EIo96v3CGtR<-JlbX$is*u(&)z#nYE}~s>92<^H$D&kfZXK_fdI7O~-HT
zF?n<6nDn|?AqUo1%Y5fkQjl?4qSP;mh1yk7jkqpq-ET?{&OZI(UyH@ux3aqGt(e8V
z6P+;~aW|?vcW`<@v9%}Kwww%)x0B)6dMZXKrg8^b9cL4#ATVGGGBl>32c{x>=2VQN
zzVl<VI&<^Y;lde~ZKgV$E7Z~BhB|(}Q^&8q8aN!NiOlDkm_-d|82RQyjutxK(BkLP
zL&^nq#Z_ow2z_$B9#U(4KpzgziVwiVPe4j%UFO5;LdZ9_E>p*Os*5+w?)=zX4}A;t
z;Brb2FO5v`Lcxxhcc!rBj>PzPvoK84jMr0^P*YiA4!wxu<E)@wXoU>!miA7u$L}<I
z#2#?KjyMNsD;!XE+X45#IY7U)E2b5>Vu2>JNserW^SuyQoMayleGc|5J@KK|1N(Tp
z>n~6I-QN>4hj?P&XivOz@T6bS3!3DcZ%tP)x6d1!9lbF%&>Q!X$u{S`VLWjq|6f<4
zclA1`eOQMJ$JgP(iFNq4d>x*Iufe&={xI}ej|V%~<EF`WIE;wGyRlJdVi||EPI1Ut
z5r@is?2{;r<GvERvSuW}%8Ohu<uJ<gBhfV}5<kNJmsc2x$hZ4o&-`AErcsFM7Y%jh
z^emx&XW>l+VhZUC%21$vvH}ld6=;Z}?h?kVo-GQj3#8v?4SmYoSqdMT0WY%7Hr;r5
zZBN9CAu(_YkHJCugL)5(N9>h&%nM6E>CprvPdtouW=UvXkOapj$;^mJ;B_g#uV&1D
z{vX?XNx+<&ocEnhK+=%}Ov~Zyj(jsOnET!2n_<ogIBJ=I9~WZbSQU#caj~3%P}2#F
zg_B<_Uj2>%qGGUbMGQ>HS>v7`g8KV|IL^7$=fP3fzvTdY*Bn5<(-9cN^ENO7+HE4R
z&odl-Rl9NG=`Q@}$R178o%Ad1K;u8#G0$^5nm=a;CHtlV-fu&P4cqXe1-s4Yy-DxF
zEXe~~P}pw^PG!-+IWrVl&qFZ5Gz9kKo1t%lQJ%!ZyqSgVgYooy5ck=GxH}z$uFaWS
zTt;T`2t=pBWR}|jSQHfis|B1#=>#Aj^vCeyJTgwJ@&NRF9)Kqo{IRaJA12N6gWGsN
zT-Eo3%U(Y$neK-pk9|=T=!>-G<eVKo=yH8EMmw%X*cbYNqkRw)=Y!0bt1<B4YM2zR
z!maHq@r`T~5bBL{%o?_e@Icg=<!FC?8TJJ%L!SCFq<Fa_W3wA{R=J_+5;xRYyJ4Z0
z8+$k0FyRt&Y{)lu*{-<ph-~BIf{fWqFx7eqg04}!Ikphq(-$JlYa#a~7visgg|I2%
z{fYAtc-I;2RyiZ2hcgyayJ@ZEgy~-$am2`x`v{JZ&2#YZo&#R(V9w1V=J_>sz>EKw
zjXv5INAAzYrhwT9{$T?z&ZoA{vxZ%w6-v8UVa)?e45+Y#KcCZQ_gUb=Gz+LbGe_YT
zb8P&Ze!RVA%quWMQur*S44TFF`k5F@zVQq;#UA!z-k40j`8^G$S=?vnHx0R|Cg`!m
z1ox@k3>shp+eRj^uj4!_&lnS3jq&Kd5wb#zaB-~>P8qNh^M@gxeqjgZ2l{IMBhPSG
zYD2yuObwXVx6c4IYYh;&mHERr$v5$w*;Ajn9iWd3-kjmP=;O`;eT<o>k5oH-=HcjL
z;u!jG+Va}&kseC&^sson9v&O$;dCoKRMSsf8KMg>ZCx~ft%Ls#>LAvD+RbyoE{uEM
zBbaY<O0bl=&D`JGIDbeROSQEzpk52LyR<OhPzy7fX>tEw6D?CU;k{Fn|E`)?w&9Kx
z@4YS~XI$cL)FpXZcu8U}T#~Lemt@k-OEUK1B{}=yl2~)s`_r=;F?>=ZP7O8Evc5)=
zIg|3`?UV~O@-eqYD(Sb`@mG!f)q+g(<*K}IULzs&;>_(yUk=%()t{@P7<ECiADor7
zac4z%EDO#`fAURY=TlNg-DXQy_EGLSA;XVW%7Janj#*p5?k4V8lW)9uwEI=ag&q~+
zu3aJBMsSxV;<!YUZ_bf#Zk8UGL0*-zrJ_<o>A7(u+g!U@DXp37zxGU}=w7ar_MA~g
z{VtaA8Y)S=ku93K+4A~pmMp)LCEj}6C(_Q7GsE*F&L&S>Khe8O#))j1FA4wKO;RW0
zIOj|D&V1?9v_RT!D3B#}1(G|aP-0dUN^oo;J0S{14=V8~R>^+#A~7q>my=<6QrVC%
z24f3l75QlDq#~JftB{`ILg^V(AO}41<=|X0)r?%}rNw#K%Pet-&Xnk$=~Ass6HW5X
zkCxm`x~UY;4@#NRBtrrlr%GDm6dBE(ro}qRGL&<pX0FN7`JZGtTAn0NRY|h_Xp%TR
zO_FO`oL7}7<gT+)4rOM^lRsH9{ZO{7zL_mLGjpW;Y?gG`pDj_2(GuVjCI9+HNz<4p
z8F?p4YyzUiw(B9;XAmPE^W)^^+C*8iJW-slCrWO+B>BfSNv78)391rg-jz7nayCvh
zZ^wxZzqi1dDU!G$RW46VlRM;_&xSejiF|XiQ?BeI-#EwT$^-hPlh>={aF|N$V^wnd
zyh=3b{dVkKDks#Ah-v5%Y4+xbn71jHvYq9kI9e|Ct&fVa%~5%meN_5RIwr&Oj>#x$
zd~^34m(@+GWd5mYajici)23gMQ(v#il9e~*$w+ozeP`F}(zh~^d=nD&P97O`#6RP^
zVcx6mNG0Exed<ZAeKHmqPC?b2saUpY3Z8mQLGY9*XsSIG(MD5|IF0N>zKOP+ioQYQ
zn_cQ~Bj31Hs^iZ!bzE<$iBern96P6p*-tg`cBd9n6M1b>Kuw3a4a+y`p<<;L)U7n}
zpcr7iBzJ15;qbVUZ-$a@T2PPm_@)CBUe9cxexpmi*+sqy3fIGcW4zWGY>N2~rr65f
zn3B0O;FmoEeVE;qG{b^=FJA|nS)sirJ&D(?;Ihyf<$T?&>&R=w&$BWAwJlZ@+abr*
z9%ac6D8J)?Z()vDyL%q4Q=@---yQP99ewJTLuKZHZ8JPDvdn|~Q66~TiJqMo9@Hm1
z@Op$N+%-L6I-i|9AHCqzYz3P4@<yE@^O@Ir<HP}P<lJ0|R?O&$YPt%oJFkQ0&~+H6
zxemSmTEi^bHMk$V1}8GsV0F4bo^4x?K>9*pu>=2*Z>~>?!bpjN8qdlj;kZVh&yy>=
zu&{a;rk3u4edaFiE$>3k?p^Si#66(LThS^Y6q62zV%fRvnA&C+Rt=4!Pl_Iuf1>bb
zLKNcpv9(?lyrxBAUu6{LU5dhjJ5kv6JPPCXM5FWK7;L|+MB0lu=DNjWlK)|3c8@{d
z%`xc8yqS>3@hH!Ur*9_#i?!HI8k&ICZrr24m;m?X2{=Ih%C}9x@}}(V`pW<2V*)lj
zO@PIj1jHXnz!>t)k9`SvNWNLMECG7t8;jal^g0)d-Q=54vQ5jNSS$#L#oswGc;p#_
zRvTgv>l}ltBiS=`?GV=AJ&2%9Q8;%hlDpgo*!{U5+YUvbU04Jz=tuB-2}j7(aNG%H
z-_(&^@N3wKU4wS=^Xy>fF?0KD+39p83`NYnslBxg#?16nJl~2{UR!ba;}*1Cy9L9W
zZo!d#?9ZGSii(Hq_SOu6KJ~rf4Z-YN3Wm1<{WqW34PF?8C%!?LIVA|=-UdQNukicX
zf%rsiCPfv1rON^^nsdoFjRSDMQ2-`zPMN(l00FT9cogW5M_v86=j{upcfQDP=7*eK
zeq>8O>>cKZE~kC*Y=JNK-6Y%8t;U{ttFd$3YNYpBji-HABh!>UwQ;Lq?!6kTH?6|u
zJ}c2F!W%sXd2{z`1(tYva9*_>1)NPa_gRLCYRk}j`%(n!xHG%b4N;5Tu!MZ`PSXu9
zdax(+y({i_bY%xUGp1&`Agucm#J^aKjPXm5rC5TzoeOc{%mQdI+p|~h0=zO=0QowP
z9+CMNd(jzFy`0gji!=6RJE7%dC*1h%h*!Z5*sAA%^?e*TlVdKiw*z~-*vUY)sbo%Z
ztNz@9yv1%QpV{zwV*~sD={M{KSBtSiXbUU6&a%YgNJ|Xl9{Ai)3v_v8j=m?%ab~?a
zj&x);DZ3Ybe4WML(^=gApM?+Hk19?!#n5}xph+K1;(H_Xk29i3Oy5k25e%3)9BpHQ
z#0e%C*vW+cc4K_0;A|?~7^&vQ*ni6ikr74+wJ|~|XHc777-Gm%LmYWzh);E#NnJ2R
zzgYH16&awO8GUK<3^1yd0p?%P$C_Au?B1!5>C|rGUG=fbMjs_c`pi^k#?2^wob0I&
zvtP`!p&#)dUZa_A*26Pq6EEz^-tRlSmODt@hWUMysbfD?>0sl09qupaz(@)Hz5%S#
zMlT=wiQ5Wl60}i8FXU~`qSU>$(DW}Y$Qex}1ZyH}h9<rZ&_rZ2O`QI$f$_VqOMjh9
z;`QW|+|%bCis@-tG2^r-9-Wr;FZpq|GqN=3l%&r;Ck`FXi+#uQ(#rU}eExo37M;5w
z6MtTj?}1n4aOhRh`Ts~d%djXA_Uqf-9cy>TO3Hm`P((Tf>{dDq1Zgm6hh_jtX*&QF
z5$x{n?v6G7=Xu``bLpzMiopEl+$Z*3VE5bwKF3^?5uYze_S*}xkQ$DHcd8(AP6BnD
z;Fshc^358ujVd2scbt&BWE)RD?8rB3GLFjJx;jy<u9agO4@sXj?8qbEgi)_qGUJfs
zcR3`+eAJI3-?Tq0+D3;ZhJ4e${IKlXN5*lhm7S+*Wk!o5(zy2#F+Nf&;bpaQmfdid
zYwN^?-k!-f_e$3Zd)T#CF1uQmNeitqQFkbly`##wDYadW^kD<T-%5GIyU>Xrm1197
zDcib|X~ypmYs(#?6|zI_yxJkMY^St1zf<<yWarGh9bz<%+RmiilG|gCl(J)Zt7EyS
zU!o7Vwo<-T(aY0$ml%_OUNR$HvY$Imw|C0h&O5nxuwAmLA9>RAYx<qq%F{CG{j*fm
zvrFV6`DVTQHfgw3AQQV4$n^6CqP?R~a_tMG_<p`9m*k5_n>-n8mnV9^bLF3txw3qJ
zuEg)k6@#)|*?A_HI$NGREY6eW{g{2K$|oa~iG56&9CG8|y|z>ajn0;-ebboZPLcT8
zDe}{ee6uA*Os}L!3$;{v-X~4kolTd%6Emf7mr@FEC?##vR=Iz6tE{7jQ!qSRPO2%z
zz(*;2b}D7ZRi!9iDP;tE0QC+Q%B6=z^2LSuD$NQhnpGju<Qre|%?t94O9lVleE%e!
zt(195d&IDAkGvecPqw$%FK2pG$@Q^S;<LR<jGI@BlPWuO_EgKnQ`Hj28`U1i8adYc
zpu{T=N*}XBVjawn0eY><kMQH@G<P)qJ0}+>UY0Wt-|)}>o|+{4j4!{ZPv?VNBHwgh
z)Dd06yJMhb4``Eb8ea6oQ6CjVPEm!fz8c1^RYkF>D&kaBF;`Cwx#XMG<eOgPn}=3v
z7!|0F%FWbn$TwfB)Ulj=bBuhmgM4GwMgzgU*`xEX27ZYK0+kvlnXZX7A2l(ymnQPZ
zX<|WVb~OA$G@K&%&etW|&IvLCfYgcFc)3^yy$<Q%unxZlSWv@Vt_%AwdZ_+94I#`v
zT%K=&qkU&!*Q^;B%5I<U5$1T%!4m#=E%B$j74i(NpzyT9`5RW4xy2gauUjK|uno7;
zZ4pJj(VR~|&J70yr#WIn=UJFEd^T3v&Bcno^RVsmGDL(e$G80}@pZH-R^__l=N?yl
zJMW79k6q#L-W9ry-4M{h4W-@Oun%tBeDuJ+E}m#x?+M3tWE$pk?^aQ}`Q?SkiQbsJ
z#2Yu$z0u~ZH}!SCKJU94D(_dJw$W<zyt^95Z?47h<aPK>#(A=6Gye15j9LDh5f!=_
zHypyC$O%Jv0sCo7!{EP@J+=qK@Q|8<#i(%H_X)?)58-Iy5`p{K>}q=x!LHWLSZWo6
zm7imfFf1O8tK;GIARe#u6Y$G10j|drFs)@GHd!U&UV0*|A0)!tBOP0K5BvLAiPTHk
zSTv9S{VN@Vcpqv!QHhTCl(<6;$1^1h`AsvSGdL5Y4rW1<zG3zE%sjD!Z~w4tST^DA
zI{D_mkJ;RG&c<c_UIWQDXY;vVO1`o5rO%0c6ZL?)&6P~NJ(P*~&6#*luIWMT#=d<9
zu4`xDEIl{RmuEo7Gy{eY)6w*B8eYv#hG}IImYqz(($5MM9#LQvJ=`P7HDBoC{?H;0
zk?UizpB=^e=Ip*X6wM8W&B%1xjGxChp_6RF{=FM9j60f#t2c0WV*_$dM&X-96cVpS
z;+u0M^T*sYSk2s2?+9Ew9}b(z;q=yoq2@*?o@@@q+3}&MzZHUy86lX>+f@I-A+TeQ
z&5c<4i4ly0)Mhf`$v0X-Sn*&z125}w!(%-TH4VZI`pP%6U*;nHGR`;s&}+9Jf-d_Z
z`kEj2+x%er*ALzJ{oo2etT6C{`yC(L{LdS!K6pd@mpA(weW2T&Y@_Kz51kK!r}<!D
zu@{``JYlWo3Fk@=RPJ?W|J+LKyR!oRdsd*u$`!Efw*oJAF2{e<mt%XAW%zIIQkd~J
z)uR7Wls8_A^C?T<Vz?0HZVM2&dI7$*Scu*G7Ghw(MVRr3Oj9)%t*6byShachI&L0r
zT$&5p19OnkZ#DvIUC?cs3;c|nVK&7X1}&-Ku)`0zjyR#sZSZc6NTByEhrXNMt?4bk
z<^X3u2OQ~Rhq5cS=;vjNVb5$ZqQ4FObNoC_vckBxmbjl}i9^1Y=suO7*J~|s{}%HF
z34B=4i`Uc~4a?2Y`u0rJJCbjLO;Oxq1|CkIj;U>o(RYUt#(y=0Pr4yO><y7XfADSQ
zp(1sSp*e&d!_;oFt{K6okiC8$%uDrVpWj|XT(dKTT1P`%e#33=YX*p>hdBAX0X`jP
z$IWg7JcuyB=^TBuU9XQ`Ci>8NMJ<}X;V!YvS!~q9TY85VxadJ$m;2wNsOfao!!!1W
zoq3^)<7ah|T&9bs;e6PWbB5|7>6H!)%XLt_LI;b6=%B?FZPW&7<Ku8`bUV&I;)Q(8
z_gm00LC}1t;MGnowAI(bfJ;*`ZShn*`J;)sS(@DS)I@wEP2?VB<~3OZUsh{CXF0n)
zsP{J}+eB4yr*Ln*^x9o70ZH}pq-VX%>{l<_ht|u(wkITB^@P|2oRH3ZBrl?KaLx%a
zo_#`;eNV{jeaA)qoE2Cz?-a*;QuPJVd3QnTUo!*6M@#aJJv})e$T~W;7i3ELdGYDQ
z{ch?ubNJXow()7!AlG=CicRCLkQqG{_0(`GY9%C?OvBgaS0@~j$Ab^a@=k|jV!J~!
zzRw|<Gv<)gbUQ2;=rjK9cUb<}dsqz8Yo+h$TG`I$xw80Lao<`i-}Jd{dGN3da6BWc
z7O}@>>u$LgQ!Xd=l}j-BCfT!8t}dmIg-r3jXPM|MWRB=drL=llDRq3_-B4L6@94X&
z8^{hk-l(EHcgV*l74j;rQi4BIiwN5;0iEgLxm7HBhl|Cfs#s>57t58m#bOj$CeA}D
zWP=I4Eg#EePRj~8GQ2`2cCC<@{2j7y>vlP*v0W~#sFXibD#YhgnS?sgKTPge-;sRN
zrbtZ52~JZB#Aa8%v^<n2L);7Hy<UNsoXD33hkRLbBUi?B-73xxa%DPwF~1{oWp8}0
z^mNY^&!3qR9Fs0(Ch6kxCS8V@Dy3&#o^0&Nj+}}j`KOOkR+^_vlbI=U(I!RiSf|LW
z`6*%%ks?=4r^v0bsS@2iO*WRL%j3zJlCxDQE9#Zf(T8mFP$}2HE2ZF5rg)E3%5P7l
z3@cX3spCp^J1XVY*#hZYQ7Cof8#nUJI<;c?m{cOyu9Ql}(K6|>h2CbqZ(kg)6vH!>
zVzXh7T=d@~?;7uuEx-23)#=<C9a$y&c2|jln+<12S4#=EP@nFrmRZzrTI$ruyx1Bs
z4>%}mMjw(vE{El_ZJqeOsh5PSCnZXonq#YX@?h<Ixf|aWM^ZW<q(evinbHw^rgg-+
zR^8$GqC4bO4}>Y14}a1V7xh(e=U-LK<YO4QMrDX95+<p_v#S~^4b<>tx*F`rH`Y#S
zXz)|VH1bVki8>0&H;pf=)2pYB)<4t{(2Ad9eKe4!t-*~U4YV4g3H1O?dOkF<gs&^=
z-fQ49{c)Ck1Un`ODs2T%`~{f_z~?%8=yvH~qN6Sn?fErkDR+CLrop;&8hW&zj;j+)
z@N~8bzIB*^nGQ2i^3x0lHk#wyM{@)uSmJ7PD>Tx#0#;ffh&hZtan?9+%^FcdZE$Oy
z4SHJJVio%Xchx%}dX6Jjw4R04lV;=N#yMCxY993CmO?XdIpP+qKvrWi4YmI0N>}VS
z<%-n%d|mv;6{T(H!y(@ok#7dlmowGC4WUX8yg2EJ)B#?YxylP~Uw9!?)f)q>y;17!
z4V4maxbn83ddnA9t^E-7ZWS`<EA4r0H6kCZMMXD%9Pn9()4u~SuzMhej}64Hk?T?C
z!{<xt0W($vqeF5q7Vr+za4iG|&%<!DPdILxhC_=twN~x9Nji-?iop?xVyDZ9#2B=C
z6oXUj?!Dk1kL->Kury}hA$t$A8WOPaZvy{+Td*!t0mmgtI6E^FgWhGrJtPH>o}{4t
zku>DpWFGr=20GFwQ)r=tcZ!nvpbX3|%D|Lk8L(x($@^Rup4Vj|u^|h$nXmsAm4$BP
zo6>jWp2yi}b1fU&$u#HNFoWMn$y-1sHJMCqZIWl^WMT&YJC<6F@<KZPyPA%&Rv8FT
z&48zU2ENYW_uFOQbXFStk0zn{D+Pjja!;&z5_AtxyGd2hXB>yMnQ<`Hk3-tUSj;zz
zg+@&bW{#n^I5Zk6^}J1KY({O#COD8EPh@h7vfoB%=WM{P;Tv$QGzxi>qcDG8B&KLb
zV%~`eteO>p>YwzNSA}ES*l-jkgdv>SsXO81o1WZ5sA1PlSO~V5hCsa|`)^JJ<3Imk
zv}V`MgHu7!UP-N{ExqLn*5mr^KqS`$V(aepQ2DhQg~_Yw2U~?73O@{6?+0T)Kji28
zp*qkH=S#^qVSZ>vFU`P5KDg50jiz0^Q6K9Kz2n}9e&>yWt$nb4gb!}%`@nEN`)uxc
zVr7sg&fN6CFdYwccXnt0)Jpa?FNdWa^WaNXpt#!#469g<5k||=?&LBgy<38YBTEpU
zxCHrS3lY&_A$B}nfPbRqWB=`WSpIb$It-YP5q;<51~r-u_H*#P{aj4iITzQ<=3>a4
zxzJoSha1_magF)#bw19BgEP#BIV0<jBNU&xE%nL~UHJLcqN^kRIpqM0c>3TTIUxMF
z1FYN}&}5%Ij@@T>=QQ>@Jha6tg*CRHx56FXmd9MM#6wF<L=Lj#j+P~Qu4I4Td2<{f
z+jzAzN7xZF_#~SlzF{VN5!m7PeFmm)Gr>h)V{VDjSHrA#o{<sOkZ(Sl8q%vqf3UkD
zgua{L?#9^h(Fiv6M%WQ+gnL#-c-_PZ$1~X7r)>xo-lTe*GQc(V+Wb>zfbNG4Fo^v&
zJGUBO?K}g_IIj<DW-1ok)5Afs&C)P>ZuIG|nXiWj%k(hHmOh-ZdicRxRF}89_;Hy#
z3)Q;t%F)FE-sjKr>ss1a_WAwLLD^v)c!%h~9?VL8(#DUi++LW$j+?hY0_EIuL%@!B
z!31N0-6bvDcGg1ACsXm<bt=k!Xrjv|O`IR9iKQnr@P4fZzH4eAmai>cUaDjNO?7na
zUM)vv)ywctN98;DB8*(qs2Bac{Pth+O$Pa<j(ihNzOf?PbmSwBeDlregd~t}>|Dq^
z<eNFa&q|j0Ice#0K~(QukXzJg9Qo)?zBxm#$>76`kCEh?w&WWd^37GUjV2$<$Tuh3
zG)P{f21#y5#)&&B9gUAl^#67gee0xVOs!Z`uSx5DNF0A2lzl%A%6CtCE^b%Ll<qa+
z)3-*}ygVQ)K6A6OQ<c0QQYCG)tL0tWZQR~2l0jp4OEGrKD9zoX+G4k)GdGp8rd%#L
zl*^muW%BJWdwQ8enoOoR)woPRJtr}ceDknUj84-hQeG*$3M*t(2EDi`6*5;@As+s`
zYYi)rCF_et5yqb2_+nYNsaS^j6pI1bru5%p`NJPy*N^$_rWMkkT9NX8xm2+S#+bgB
zB;IRmnpcX$aBeLWRLZFimEwK5oP90is}ChIcuTRIYEmeL8M%@lw^eT5-6|Va<w|8~
zo;>E=$K0quChX0Z3A*`Wx;;;J?#z*$@!9nMWC@4IrCCG<dlfUdEt(-qmZXctr!?6X
zmL^l$r^$lRsq$h&iWqC8h}XX<(m_8(40TdO(>O)G8Ky|BWr{d>rbtX(ip&|GDpQW7
zNnKpJ=;|ruL84NI?^cQi?@(&TloEDEDO+!3N=ZL5jlEJNS1FfHDW&a>e2GqBo+`Ic
zUb2(trNcIfB;VXm<!$Y8sob7cE`e1Q(vG?GCiFIG^xnzc^IalNdt@MeOY!?kWOrSu
z{CHg^H=FQ%$M@mwVbyZciJit<t7YcfYG&7J#B6zuoVZ;h+o<7eUrqhy?NMnxw}Bem
z1<4G3D_Wc1iPMa>IQO$XO!GT%`>-SSd3VI`vE8uLvpf3O_24FBPdEflKyvpf2-Q--
z+NUb88LEnPeN<7<UlnE}RbfiL>DEJyd1qBTo1liH-fBpBt_D}~O|t@Z966|tS>&4n
z@{Q|vb$oBZ&olDPSY7rZlW&^wb8!0)4eb1+fxhoF(DtPUcDG^QW;emev4W)Of(|-b
zsLj;EzE#>-mahjzGXtnCGUT?W5zOeb8Q<Il$9kLKlA#HF7Moz_4pW58F~cZ!dW~hq
z>d||1^whDydiI-M(cyh+kri&3T4Q3QHG;2jmya8LAN_5B!#3C!<bYcZ4)7U5zPU{=
z&eYjB8aW4^hV$S_{XWTSIhwSlzW;b7es6N+c7rQiD_rsVkSprQIa$0{6}NYTA^GOU
z82WJZ+%Pi513E`N;V{$-*F3$j<~8>t)V<NxnYztfAM%zjrZW4Yal#j7E&Xu8VKr(a
znfq<Mh8?zR(Q2YU)_JXiB6=MLZ45xxus~eZ|Nq`p*WE!-=W}O{A{ee!d@g;%8%XaE
zoVN{uV^#=q{)BKtH56v-)g4g~ioLaA=&ZXL4<5we96O$SbdST*m+{D%o`9d^8!Pfn
zYR4_uagcl4s}*SZaSI+8C&9L9GCI+tyfiZfqZ*~7{zy8U4rM^2h&?`SGErhjzpq6G
zR{l?0@ytMz>lqj_JrhqmWT7@S3!AN(MPvSZqMs74e`lg!MJArD&%`WSvJg3?`{i_e
zZ=f!;EEy{8lF%ntf%*Sz!Dw!L8}J#@!Xz1P$;sGKnT(IWlF@ua3Q8?gVX!?3@naRx
z^;bYAP=Rz;1-_HZ{0tMZdq2DUzQrObF&3`}aep;G2CoLku>U3+S{<YDK4mk!=`B7I
zy@_3PoA5nkBMvp$$ZRv&rZuxu^ximii^AX3NSq%Si8aL$aMy@{<;8ILTZJQmTG^y7
zVK93gim5K4*nK_(v#8w!uH^e)oxOg)*<G_Q7^{{AV_;|QFjNGguW=B<cCN?iJAv3g
zj+}Zs0AbVDpyJAET=w+C_NjiTnC6FcBR^Q#`=RN5Uo7kDhd-<QV0OzJPJP*9v(yXk
zmwRFQ0WS>Y$I%e((*EVoDX8Zp_3*);$KE)^kGGmVp7?0tiN<R^V0P6V6Gyr8CPbfD
z^W|8Wwiq=-R^WWza?EVJ9L4>X;rZ)@NUm4_mjm<B?Smt}?Q%r%1bcKyu*Xa%2dv=M
zWtWqVDDa<!&hbuIvYXlO3$t<e_-q(WoP(hg=OW?79AvA`LGELEZ_=D`d8so!YR*XK
zYXukb%{%gq`fEpAdB{!ho{soLe}Bn32h1GkfPPAQ4D4->$IL+<G_&RQtqpQd+TdDC
z8|)y@c)qm6t8`0f{cDM->?&_nYQe2v3+%~f57r+u<XtvHT#Xq*6lRE-%^tsRrU=qA
zg}uTA1@qWvL;dDikr6)X8o^_?Aq=L_PxC)M@i0I^TVqV;&*wfhq9=_T-)s15G>o96
zxA;N~`}|Z45k#+zZ<PV&@8a9Ai}>6BGT`L~c)f)^H+Ba6v+AQiJHRfS)1$Xg59Ra{
z@94se6g|UM*}7;uQxETY>9JEl7iPzFad(?8mh-kg!dDj^ZFTW>JXxoSE-L6PJe$PK
zls)~#U35@VuZ<yI+Su1q8<+P0%`AXRkC^}V68!f|3!_4r|8Aj$R*_RNnE8#rWShe7
zniy9^o!m%+n{n*(C|Ae%wd%-l;Me78>ac#kPda@(Djy>1<OV&xn=|Vqw@ba`lW%;<
zIof^e<@)e?`Tg~{qz*VCrL#_mDcR-&xn?9EmVAWlJuVZ;H}&kRIlA(Kw4ZxH%A78U
z**tPi4D(T{XXR$s6H>D3xJ;hbAdCAn$m3oOa*9mzk{eLpTQtZF>Nn^A*2_Bbjj`gW
z42(R=OirEjI9?~mkJQP!=|?1ZBE7=?2c_$$gK~NWeHZM0yOviieOnbt<MV~G>3N|%
zX;~z<Qj5f{vPedrC=$opMe=^wZdpsdIj_1~u6*Am2YxfZ6HqQArj*OoCS_9ji+<fv
zrPAV2iCFztD&IPm$+$12^5;7}zC+3+U%yNYBIzI5SRn(WDrAIxDYuP^WvhL$^jKXi
zT|$aw1X-q>{k${g7t0<gmZQyzCADd>%t_xSOQPwisiME-Zn>mAE*F&!)NeMIOWtwz
zCZDYkg=VE#*HuViNx8I~S0)KZxrY!`EW4i;$cHVtawKG{lpfqF)zvwo@XwL4HrcXz
zbB^rVlrIm*<;$mvJh@<zBe&;d%i6>&8D5hi|Au799>WY-`YK&E&r0XLC`~S|PLnRJ
z)8u-5syKzENSEFzvOPapmW3pX_oNg#GnF?fl@tkKs>;kZMQ$EQ5&c1_(lI8Ld8Kse
zI!!6+o0Q^Oq?9Lnm9lx8Qr;IR<&!q~riW6V+9<`kP$_}*I_E3%WpY%3gp4SX6?M#v
zPvlP3!)@{=uSBLiFBRPx<?^V$LM*70eQmNs?)KX$otaN-%zmK7^i6+ZcTQA~G8s9h
zO!lRg$@`(@a$z9<4aQbWmkrgDaF%<h7pvtPIcL>K>Nw5^rIX!ZS^V>;T<+N*)yz$;
z@qQ~!8o!gd-`+{npKWoegq=47I>IZX3p#%3hWMx5q0xCFnq^MJ5Wh+2`)3k*Wvjsc
zlnVZBr;3$5Rq?BzD)tUnrME#9^Q~18TcwJd<JIUVRKw&DH54yV!_sAHnB=C0i|%T8
z?Wu-v{@jhVY8aBHhHH8n7|NXok4g;`k!cifHL&EF25vvpfK78jTt~sI?z~O)pl8lT
z3k%L_;l@NkNH-%~TxJB!H^$Om+~++yjU6%55tc*mO*a$tw={vqHdB~6n$g5#j(YM<
z$aZrqgaz7#SYQ<S#%aD442+ncT5gR_2d(j&d^0e_2K|rN;9{@??i}N8=s%9ozR8V0
z&DpR?n1e}k=V7niQk=9~j#~%$xVjQk)Lii}$`${Wx?)p}D?Xoa#Rc+BCT~>7I=ew*
zv>W#=+|X~G2maabi7skhXzt~O^>4h8Y~YOqH6N_}*B1vpe6f-FmnBDic_UwqZ|hfM
z@Zz;NacC`ixA4a=V}EY?`6FQGI($wDfZ^TsxIk~}b^16y91OzIIl-8p$lP%?pHE-2
z@3t2^{2fEE;b;if{1?L81G{zChhi-GX4Lag^vw=KQ2J)v=^l%{HL;i#7l-yH3CIpi
zK%d=-+-%r_H9NOJfBzQrc2XcwlJLAyGTxs|hJzpZ=0h6B7p0>jF9UW{(y;d@y9tqj
z?YtL#cFI8f$_$)1mw`${cGMhKqD!wVYIAA0_9`7=6EZN@nx5b#8F2E>z*Tl6?p4HM
z?(G=Z7sTLfs~Gfe8UwQ)F?h5*2Ilk@b&yzi?qVn3KXGWaEe?Y}a~o4X9?eqXv2#lT
zb}UNd#!({hU<=%g6wscXh~Q^&IL4i+Y3wRaTF9-Ak1>3{iGlX_Xe9YYWAXdVaPZxX
z{Lh=P#GU@)mm9Hm3A_ECZD8;12JC(vh2fr2X!$h~i~S?<q;Vt?H%1`6M+A=Nhof<q
za103v!>@K>c>F9B%bn>fW+wdS+7L7)Te=Sq;TA_QUT<S}pF=R-{tAM#A_xx$2jPp^
zdc3m;#6H&mn4VdOgOk^wW}_e6jQJqTy!hjb#5R6-%g*7O!+oK6?gQ6O-Z<QneDm20
z0W-W%`kG#KCoi<-$5hf1PgKqG#Mewu4D|8DNaH0i8ZZ}S9p_+0li4uC3N)sVIIfvH
z7SbzrZSYDQHRk5+bQkoe_oeAn_9V=+N6WVM+<s)Q!s!3?e>0o?KgO74i$+82;Qh=V
z;jOrxKp*}O6?S3{cA|ZS_ol1NMJ;hbK+SBZ@-CI*H3t{J&qlWf7tD)yhOe_TM!k1J
zv$ot|Am0rB;)omMn<kGP@%*L(taBXjdYJ>pcXGh8WP29x+GAmZEk3gU{O@NQG(W)I
zyI;IXxmlvXihJOXEwJ_)`)%^bHntWR`pz8wGuRJNYzBWFGnjZ$w@olZJ@+$hPM9L&
z$_$L@WP%&qlG@eM7|lBv@n&g&#p?|)tFIw6sJ*1OGJqdDEIKeVHSnGhGxtUaSzv?{
zgULAb-n0)g#DG!kx4C10Cb<U4=GzR(Ho%fB1N1MV&p5^a7R*k0EYioPMtWGhl3DM~
z^xG`e!;=<z`0tu7$}`wO9IK1lt@N<w9P_41T{I0O-z?;fYPv2qkEVZ+If->QbkL7k
z@P2MOFd5IsGi@}aXhW&TZ0Zxf6bb_TyMue9f~xU?c;2P5{^f%?jn5;eax;T^ct84W
zc=tCOr9mA?9V3^hBdEVRjy_XEP@NjyZdb$Ir~Bl~v^o)UKCJ6x{>C~PN4_y9-z+8H
z+;?NYh=09g=G4oMA>7+`YLKj~2D#*XLWYoS-en$_`h>HxdeS*D)xRL?%r3~#`qT2p
z{(|TmUXX3%n+a3SN(}jC1^H$y`NoNS)2(-dB#>{ekZ+!lZ_1iB$k~qd;!eKVwdJU2
z8XOg~hjsFzp-vu>ZxYBiUgVn%wg;u(%7ZeTxtW|+2PL&bk=X7jl+M=*WzEw<+2>y*
z$?T@v$=l2Ph9dbx?O@#HB5`V0C=vJa<whcVbnlc)1iO4I#+S>YN2RjrTB)42=6>X{
z5^*|IB5vdhhqEQzfGU+G1IxtBfH$Va74rRNsc5B?N{6Cic{r204RxACzhW^DE|%_O
zoQ`B2%{kO<REkC8%r>b_<VJ$tHfh3q^ySQQ8R*Fy7B!!_&Lwi2&+$={%VoGvg-rck
zAuEzA<mM)3xs1xhV{@^n?_sANeKRfew#wauS#tP7w$xEKdGjb!>Vq;xryx`M<!8u%
zzDh~nnJJF1GsSjwri5?EkX`$zw_Hz^u{x=8_D8CGS&%CGW6~tgI!z*4q{;b*DYD)t
zMSNV6#pz6vG})FUD`zFkRcblTgHl9mc#8a0O_7u%NiyzNl2~_2maf;6q_}sgZ1}*p
z+nXWZ<1@r&RHis|&Xlw^nbLhnraU6!ENz-69jGyBMKQ1CS0Lm6DUv0HMY3<;Hn}ow
zn;b1Jlz$2f<xozc{7f&Dxy`oAs%AT+iaw+}*1P1|&?5P?oxRE5N|~oClXe>1?Q|)V
zt>tC^k4u-)4}D3Ow<=5KsdiS&kosyl{G?jOudflEHwR_x+*%oUxn9(+*UO$qH{|Ev
zx6(P{opf*YUUvQbA~@C*uMIk(@>e(Lz2t4G+eCa)PDCT`Nf`Qk5;|n4V9ZGsT)nD-
zMP!_A-BkHGsS1C6Raly{8z)~C6UM1QZJ`>b@)5IK4HhfZuwbPcv(RejLe2>$-`w{h
z1CejW8EW9HqXwp=YT#}$b(?w(-T^e|GuFV#zgkc~tpz{k%4>RSp>wSkUVYGlZc8IH
z>1c#6e0_B!*ccTpr(tB>G>l?q>LxQ&pPJEEJi`Q2!c1`DyeXC}HN$Xs-lo`v;nLLt
z=jU0_!(ahl1AeV!2BUwt6-@S6V@(_8Gd8m$^`i~M%nmiR4$$cCh^c)2+S|^B*`hhT
z>Cc1iEN+8-SPF+1%iyT790y17F6HEkG1PGWRJo$rN&1e-H(wi3zv1&p9J5sVwr*H>
z+=HGHPxvq5ZK~7@7e;yGOIshrbn#`@$QM!0zPvN|;^mE12<W#46MC<OojV^jYvI(=
z9|@-ZD0%A78`e5zM*^|JWj*#Xce8y-5IcB+pliX+qWEA0)&|oXNiAnU2(#0?kL(G7
z;r9^0F_d|tP%POQip?*W@v#WQ)z@L@tsjomM%<02KP7Qv9E#S*^L7=FhU<wqJbDZ4
zytZKA4BmQLCc%AK5?WR!VeZdl*iBAF;OSKO4obz&!c=U3la3~u?4mKwKtJj@<G5S;
zoi{4q>HKf{f}cq;rcF%1O50eB?H`M_onq0sUo2MA3v=*qG`m-$;d3Gyjn7A;$E|4W
zzaNd_&-{L`7>rmOgT)_WaN8gj<9cxu)1RAP+!32PG9HV<;^7vVfM%Z(P<|s3^D+|g
zyF3wRI>ckl(Kz&@?$dfJwHviq?5K-Ds7VY)pN)p2Id4<vHuE0IZ1|;3^w@0T7UD*%
zo3RliPi=tGbOWD*qUb-3Lguwdq|kd4`;dE@%OlX~Q#h>GhGY8sFa+3#Vb-}&d<qT4
zHNKC>6;QKrA>$wfXS;_W>p?I)GlOwpI`31jg79)(5ckA`(EW8Fx*ZC@`0RBUyUQOZ
z;#Wf!`eA&3KV(1lrMBe@%_+Y4SnY$RPq~$F-Upc-yfLkrH+qtdKk&|J?&yhiq{>YT
zsS6f)prPM<G<r=BOzb@9HJgVg52>M)%|`Zc7i5lfLf9fl96Dr=m>N4YbhkrRxh+ap
z+oSg#2Za8z$7pMNRP!dJr*4O<`P^7>vV|8J!$-{qZkueF3A0D2A%CqVy}-R@;q<pz
z<RNER6gXo>8yB3J$*y9%+35LmHq;N!M&oTRxSdZ=v6(aOUUGuN-dQjw-;5;RbbrG~
z8%N&59MC((0SzwXoEGfZq)wF9#h&|2ws6t4#lajK1heO6KKaJ^vL#GZ&3VHyM|_+)
ztomC({~PyzcblWy$Q&PY%y8%9Ox)6($=4{R@EK@|dh;0=;9`O^gQuZ(wh@9Kv0M0v
z0j7~p?s#!Kfx1nL5(BuN*2fZdQ`pceb0?X5mLW!1WMPEvWTC_LhU8pBZ0chOUwUnJ
z^fN#iHJAC;%zIAd-lek+wye>CYCdzEpSTfqR~J|I>f#miQrffiaQiWHP{q1fldOwB
zn{@H<EB%7ZNX_?R2AuxEVs%|O4`9blV_jUkt%DQWb#QLI4)z-9;NKQHxK*PKdsl5l
zbkhcB+MqE5XnULb_F}=oSNz(zLJM~uvrEizDt<D%(PgS8%*!;8JxqgnDP~vI)e(JN
z4Z54u@UNvBE{|4&K`%8}KH*MeO07&J-;5>S%qHLXlQXuAtCt+Bdg<v{FSDKNrOD=c
z$vQ?aj#-0DP&CMYH4WlrcT75apOxW#&xz;Q3t~;aDQ!3{zICT%E5D62z94H)pO+oX
zUm5i0P3z}z*=>AWrmG#7lDK2+hdU<Ez3ZfR@J@N|zEiGy?389}cS>f}PH|7Fl2do9
z<RCp4rfaH2Kf6Y5e5#Q<I4IWKt9-lqp!6W$ykL%K;`Tz>ez}nScHCK7QzY$@i=;dG
z=KIkiS$U~Q!ueR*vQW75C*JhkykS<zU}Tw08e1Z7Z<k8OWipPG$cQ4c$DR^-SY0BE
z$U4t`OQqnSGMT|nnwPw(d<x`-(kSXRn#Hnm3hzg6*jcx6n=D}ddCQj~IkUe=zHKQI
z2kI4Vv$l!-$ZeASsz`cdvfFokk(55-bN4y!P`)ab53@?SXI>_`Y31@efZiVd|DXNX
zCR6hZrR|=4u?@?U5vDovAdB~=lq^~HLMcPDl`@cgv(KA87`;qs5<r%*&X790Ou49;
zDJSbQWQJXa%+F7gsi#w!&q|dcxhYaJB1MXiB}<F=WN93wkf(tP*%Yb}m!xEAv@41G
z=Si~EElIqFCP_SbX6yST>2o_teiSSCcUMS$l0u%8D5P7vBstNIzlTw>SS?JKUgtC9
z_S+2Ud?rKs=4Hy%&Pp+Ck|ky7d9rJ2KKIrO<Y<c`@m3UZld6cD5QS1qzR}Ao6rT*<
z*tQhPy?Fkcg%!&EtU~$Qu}C!fQFG(JUGJ~d(<YS3dG#`JHz|{a>1EQF_dQQD-Uqu@
ziPe$=lBQ8Dx00*nby~G757;3W*z0rMwp?6P%H=^sxtNYDmx=1UL*|yt*H7j0Eq8~U
zU;9<wFa9k96rGT`sXOi+=z$UQCt}Q>iE!ID371k-@awn=Hl0?%-wP_tEvjJG5f!X6
zP(?WTCVI0f-Lz_`nX87Y^VG1Gd}B<$aUtJ0lW#hcZ|0J3D#<q^$Tx}Pn^WYQw{tY$
z#XHrsM6yk`2JY~6M@Y6hvc_q_{>D_yqW}N%W!|R9Hz!*d;Zt)XSeF=~eVj4c|Hqs8
zoas;+O~=TJ>FCTm)wO9R_z=P#(o?4Jbuh!G)$BMvX^u+rO^*d+o1+$pox%?H8cTRs
zTA`?^H4e3~!2kuf(VN=hjIJHJAG3oRU(ar?u;;Cx-K%@%;I7wP*v*^|o#qQ*W;q`Z
zFU-f-Sqtz+i+%Kg^cgctb?CV(dXjI-|GL7YqZ`ihKK1IhI}WFNz`MH_ns|Gm{F@i{
z(N{dKoe!q9_r*E#&DH;WQJ}sGRuk#L=)VS=yR4-*axE5BuH}u*pS@!KnD)aTx26W7
zcU>Tcj$aSI`1P>r6@-J!_<Y$pm|5FkI35kg>X*TY>K}q1z9FdC8G;U)p~&KMuPN_S
zw|9i%B5x^6&B7pO!(eL{j$L=dF*ha>ONU0`>-{LKJ+=X3ZfxW}!xrqd-GcGc6=*sj
z2`A<!;YwK&a($99_YN~k0V#OWJ{2#+QxQyErtPP6xarb=!yDDY#mrUlhV{8$IvFS#
zuUjM_FeVmm*0ET;D+XbHF*r6f2K#qKqdjj$&9+8k>b7VMX^2KH`6KjhG_HP#hDL`N
z^j;hTz4g>~wD`7Ru~^~}$34<GDE^5@=f&}u>z07l2NJlcl?c!9M3h{PL({i$(9e&<
zjr+0CSx3)J`&g(Y#o%4P7)0kq!(&u5I+blk{G`o@*trRrlQ*HLl6&D3HsW;o23(%7
z0iAb5VX8_LdhCxxx<umbkqC^M9)a9*;dss4)XtM(IHwtg^O>OtY8Hx{Y24CW7=k2y
zcHWF8Teb{g4vY6f>T1Wef-(Ab5T=kvd)2OoO-vyCM+e{?`;KBau0_GD)i7x6i+{iR
zB6^oEmM-AOL?*LP{ka*%kCV0geQ>{>H{u$5qtP`l1hw|Sw9_7NyY7Ld`#jKjwg>u<
zZ>nF;L;IL{yx+~mhih|Ct(=X~pPkY9;4FBs7kD=BKTWpTVP#`Gl!Vyg*hX7Sh-a7H
z8+)`fwnvxS?8emK9s)B<d&+E(NwzpguiBTTHr%yj@3)!*`wbm1k^0Lo_gR?j>qJe~
z8D4vxvGScW&NZi>xY`Bo{G4$!a>2Pm-lc{*qun+qtUomijYiEv&R$3CnB|DH){Z!T
z!T~`E4p?CCfPb4fAUD<?Pg>i<P+^O}vDA(tZJ^WG8ma$T^7D)L<#FctQEG;um!_EV
z$sAjmo0?Z<j&}y!1I{*s8+Ti-_nL{#W~TVyJcFALCa8*@hL#(R(6TFan;Z1paPRU{
z8ohgS=nI}t4^090E}isobUE`(f$X5!qK8hkdRX4h00;jV!eWmhZp}C3>kLEmrswa}
z9bJ4`t4sFL;ny&FfO$iD<Ew)oxjMK<KDl^U7va1`opsT{<4Mds{nWuPYBxLM+1DP)
zw_T-+%2e`=Ht$k{bup#AE_<8UNle{l$8H@g-K2v%j`S4|)Ir)k_V_8Z5vWi94g0{h
zhXR&8fqKSruZ$90-lm0|AzJ9f&OZNUQ=#RmiD#EIpsl5W<HhP2I#eCbrD|AZqz2p1
zs<^*L74u?Lae1vOx`q@mL%d7=rjc>7YQ?EVy{L|^moBFDVr^3|_73&(p8lL$eE6Hx
z%X9wxh<OLd1_z|Y&;1e;#x13C>JEP85;JhCoM^dKo?Xn5^3WW1spUw0<1BghNJ$?q
zzu%!yepwdE($j?!&j-v4q@a6&bZf4ZSAR35e5g`PrYq$NJ*p4hD5Yz+EO8v0B~KS+
zN%iF{8EBO)5ofZc-Y7>d*YHjgR45Bm3#IY=Y}sDP%+4*PG*iivYuQ<HyLYy9-k2?m
zK4puV9XY|09lZQAKBsO|m0l`cI+TdX0cNhtOXZ^!NeBKu(<}<*Uv5V}I8z`-j)m-G
z&K1*fx$<Xxo_y?5AQ$Iz-++6P3DkuK4J(!}kGDx_1iM|vb5rtTk>uwUNla{!>~$}a
zRpGp`>@Je3H0l>ic%O0Q-oaV^SoJn}bgEbe6qZQ*;xcif&*ba+9I=|5Eu&Sl<;=Tm
z3A4$O^szZIeMGjnC1i>5(k$_~sFXZvHD=YB@@i40^a#%o#k>reJ3K@5{-jI#rgSk$
z%a9u53|V(3h5MB$l3|)6b@T~eZJQ$d_9x5YLWS%LSIA#4GET5U{01b+%!DNN=Ojt*
ziR7BLNm6@PA)W^n@*-U!I~^6W*gzrQOce6iULpIF74ps_Nj`5+lF`qT<;B-j*&Rx+
z&#!cuI6p(0+{}=@Klx`Y$`aKn`QpnyoPIY8<XJ?K)J!N6b>6Ci$Tuy?H?HKH*NNPB
zAm0?SD>*2oP_$YXNpMg8n$jZi{9G#2$v2J2H!6l@5;=!AswHGo%L*|rt`yrb2iVn9
zCB<K>By(D|RH^QeLG)A^nwQH>-W@&Y*^w^gqM$}*w3t6m-_uz7xdYf8U3CAe+{$l>
zvL>A|BBTd2yqF6Qn}{E?CSi5jB+QRfK~M6HCGS*Uj;UZzjS9XRs$$Sm-m1d6-$1_6
z(NV+UIrQOB!@0MVnJV&)5!oi!gWC?|o5jA=a8xz0W+rtS^3755P5lB5d`i%OZIlMC
z2Wg->Rvm->(PF-4D*Hn<@tl0q_&sk_jRbdW4KVn*ArhJ2tYGhG#!6;B-%ZEvUEGan
z%{@-)H;z3`v0^;+X;U+5_2y7HVUDj|EMT<A0@bw^C>&!6)$NwZGPQ!wLn}N8w1y(W
z28QizF>9+WHjT8y%`iK}52sfC+8)0uXXC2%TzKxE3tyRsk_U6K<?>uy{m!gN>qWR&
zv=~DZTygy&J;vNquKMGOrtRIR7rU_o!yQATJYd(EehD8hd?MeR{g<}}`Xvr_<eqb9
zKRD$2!F2K}G_P0<Rq8iyGuLo8W-YcQtVP2Fe@LP~)_xDbSffDnsSe}@_IkXDTaO=I
zf^f+;2(|1^xps*?D%rv4M*YUUUkDaEvzzdJ2y<Sc7)HK%MZW12ABtOh$v9)#xw||J
z5hub>&?N%H^&(KRJ_5~pMMC#kB+S{_`w=mST9$xI3$|d$KS{VfD+$}Glele~jM}PX
z?ro;9*Di%yl&QSsr-L4S)RS?3cF({7YCGd5reWT}6fEzNjJ}WJk<MK9))ui)kX3F?
zj)65<W+}Pma9T8X2)N;06iwYc8eL9B<MhpFSiXtIqsB3i`7wCr%Pj?F-X90VG7B9C
z&obVqn#QBrg!;|8c>M87KzH>-T))pPFzPk~Qn`7m6Ngu)V-eyKiyC^VUoMS7ESYJV
zKY#zG)Q2{1Mnb2}c$l~ee%&_Fqq7mw+|(SKumNYgY(T)4D0J`1-kY>Y`1Oy(fZPbE
zjO5OCX*h0;3&%HR*5>ynBhptq>_-S5#f4y+RS3q83c<fUL$LNjFb*FKhH_IdcBllS
z%LVrN*#yBMYCZQ^17V<8ht@CFqOtoLD702Fi{yjzi+td>(HCb&u(NNT4_;e%W9lnU
zv^4eN?l`kg8Ez=K=Y|Q#+|iv(bIHpCB?cboG|2<c)!oqR+X~crEJsZGLbR@Dw_YZ-
zlpoF*%6+v{(GD28gg2-NJ3PE+iy;eb5#PiXiPS9zFn`_YUppwjus^}o3M&U$VcR?A
zub*0?Y=#wv(yx|q#~$iq9pL_v8K+~8+;g0TUw3C=K^1RL3)oAX>Wq)uobjZN8&kWT
z*{S1<_H&%^PakJY4s$}lo>?%Z?o+$n5lQ5mKAqUXz&miFuLFC19B``09%ftY5c1QO
z9SOGB+r<`Pyct&wWNYzhOH3#?$BJKOcu8-~i+ocw{yY<7E}KJ}`?_Z?nL%^28BQIZ
ziBT8$b025Ge)<fkx1Em0*+#fF)ClM36ZUS)em;6<cE8t0=wt)r^kFXh6nlc1(YB;F
zrgViK)=|UhQKN^wJ@s*lH>cEOG7<ZVC*&KTr7C^I`*ktyl{U)XXk%(G9XP47dz6gh
zox>d&c2=m;Q`0I;7Xzs47>?AT9;<^sS9DOZSr<)0bTO1(n?Xt)Xf<PJ`$ruZJ>=V-
z(ZS#yI(QIEzF|JpSw#oNzqHw_sEu>(+L+x}8+~e6yJX94@W<><UCI6Lw_514j9oW(
zrecN3RE)0V*U65Wa9gQ?Mkm#AZydWhGS&F?gq<Cgs_=7EWlxtX-oH`7vKuOhp_f84
zjJ}QwyX5_mUGi|(F0mcDLwtJdkgk?RGJSfHjNM->-wqVZF6M$3<<UcXbDLCO%9k;Y
zdD1v=tC)Cg70V+z?8eQJ8+&s^r#MHRx#mciNsd&JZ@%AF%53^HIt3|Zo|#ha{Hv5R
z?Ui!-SEd~Pk}2)JWy(PE&9-4mSv^O|eFLSaG|iGbU9u#tf0igDOZKQ`$&JZb(vJLc
zQ9Vo4>a)b^NtWmi&X&tcW`g+lIPhC3wH>o$YG{cB#nAgp4d&KqZWmlCkv1Boa!<EZ
z^i4|T@az%^Y+E80>$v@BR3ulZF-(r(rqbU6DXA%t#d*0huV^c`HMdHlL9Q6`uC=TK
zy(HwD-K7Px=WT&Btt}9pyaF*!FOci}*6U4?G*GAU+*Bl^-xSK-+Cu)nsOOj!iT}7F
z8S<l0p6V1y`Go)NRTjt&MWGB_m?t;1w~C`=%LZn&FH9thHDt?_zuD4pX0}|qmnHx5
z?|OF{nXOJKK_4>3=3Rz3mS>3ZstozhBt!DMW=PcWbh+u6E(^Rf<n5Rgd19L^I-{xA
zj7*l+<dD3V$@1GSS)M8tvXXj?o|{6<)+=OMpCmE#N@AyUl8hgiB$FB?iA@99r$8YV
z{P`hdo8a~eS){L!^A=>IO$zZnt&oX%Ns`bgSzZRFFk_u2#^vd<;a<9|zMU$Qo~Fo_
zcgb>l{uT*ul_;town!><s7Q-^dHkV3TC}H@ww0N2@}zzOHJeR^GMjm@+2ouq)OWnO
z>2RAGoXU+-sT^7+E!E0oEcqsBewmCS-)yz25bdH$S@C$k%(dW7Rc)0xHm(*+_Av&N
zW9Au@%l!%E()YhIDfmqv6~F&?0q>FYJzXi<AzpXC$mhL{FsQU8<}~k&zL7mJV_{DW
zOYDt-4wLB9orHzOlhAbcB>G?{;Z4RA)U8s%IU`lLl5f1(_1BVovqYQsDQ3u*Gf&l(
ze4|FbF(co2kZ*FyH^=*HU^+A~%t{0LHX3+DzHuSnq>^v6J;*!v)G_R#8kW^hM(1N{
z@ZYM*onNxeCoObn!d%rf1JtrtE3?c9S56q?*5~QSk2k^J^Cq}Fb_TMeX5iXvQ#`Yu
zi53UUuoLF!z;2A|o%s5Xd{a<u0atGDITu?(;bny$<Qv-s*7&D0weWu2Cf{X?NrUb1
zF4zwBpY6~r#~$$~X2ZjV9vqpAF$?BGBW^AxGy7Wi#|4Y+c@x;Y2(L_*z>*p9?2oSW
zmAWFhts7PkcjLB`J2v|8KDB`T#cp1hbkhr;d7t|Hz#D!3_@HF0AL8=-5Ho2NrY~8I
zv46NP6vckGUu&SET#FhNfA-7-pyFEqCg=ykby6TSj00iiABevX1Gxhn#B2+_9SuRa
z&@>nuRk?9AkDgYQ5Twir!NL3xOnl0%fRUk?w3Hqk-lu9t(>Jm*3>(NdZ8}FF01+@*
z6M_DS#D?-nZV^V|g4#w{pNWMLxnb3vEihi7z{^Sn^`>OJOGt*dc?#S<vpeTJ+pwRc
z<NmXB455y*?Rq-CRi?5NC=IIzrXaj^GW-t4V|ZvReo}+E;TnU?PtjOj&+Zv=O>qi2
zCO#VN$Tx{QqA`culX5j0qn}43j9!^Erx+~c+jtC$LFuI!?vV2Lm>q{PFXC`lB_5ru
z*x|?9SIDqLT)Ps7<h^k?9?m-!IY_03-X0z5J_mT;(u~2cqtP(5iblZQ%~-LN_o?@r
zaNK7T6kj)Dlix;M`?3L!z8j$U5`}8ND4hNtiEZ@WO#T}I&Cm$gwT!^(jp10_CLF87
z!Z7zoC}LKGBA@TytU&IOX@#Iu>kt&52}aYO%z#f1#FH-pXt^~2Z%hKwd}bi7EMJcR
zvg+u@>#&Re#wm+eBO=@n<(cdUnC!)lX>a^|?1f&#J<;#6JKP_-Vf+UV%<y!_qq(kp
zo_59BdN*8H>5iTY-C>dIhNo9uq4s<wT-ZlEs1Lov>Ps<Pu?UM27vkCJIqb7>L{U2j
ztR8HSz4ms<+hdFHskYqSv_Xci4f_1BMvFz(SaQw^&U37Ap@S7J-m=7jvzD;s9rw1j
z6~0lwiS0<O<tlZSC`U|eGz+(?X2C4l39+rXDK*0xK~Bzi@y7`+O`Wm7w==rFc4Ci{
z6OR6#h0&+!N$xfaDrJsX>g<S&9^7Kcrsrm*13r&&z>;0u$0gr*dQ&^yYzv3KHrz+G
zK`1pO{|pQGv6ndSy&3w~m_o7G6u$?VB6GnESVvDs0Xz9VF&A~upWd0O8F0U3g7B}?
z@wAki-vih~+};Rt*m-lVxgmZX)5ot|eH6duPGvi0mg;oj`CAw6|FS!nT%vT=MgM*D
z`TpRIX(4~?s~*;#Wv7k805^DhvbwB|fhV-Fvw=Na&$aQQs}3e<>EM$C*<>es_%7-o
zqF4uane)8gQwP`E>R{?`ZQddIcJ$la_v9UFiw@Qu(ZN_gZ1(A3T7eGrZ@w+F;5YPi
zu%L?$d~axDL9#a1%;4L7;l9)sAbupbYjz4AstQ)^)I!lHEu^MQ#kjwkXzHQ~)t&tM
z*^K!OJ9Qi=r$?YG?@$|6k<){(Jr7WS_f^4+=_<JRuL=yw1qWYdODi%?$3NLJpj(c3
zt{|giZxvlW-#=*2=VjMIF?K7I`q_mt>RO(>P0JJCS$SgMNlnH*M`pU^NIrQciG1Tv
zzIj5vSv4Vt&#zf>vqhE^K49-#zEU;@Dn)5df9gbbq;%usXC`;JGbM$L^ZHDtbm*fL
z_cG>>KP$zNcaDbkSu&w7pW{`tWC*!t<B%*F&@W5ksPDX_k2NM!DfQ7x_7f`QD|0)m
z$Txb;vScOsW)=CymV8rBz6m7XJWwx{{^T2Hy;7NK#~*K1A_J(!Y+6|)Rv!vwqb{Fs
z*`If!ut3UVbEV1Ft#Tx3tC;r4l@V9-<#)>hIYut|lT{$AE*FUZz5+RwQ6TEv<xa>c
zkob}Ush~DhzlOV$)A@W&Z%O<ozVE0>Jo-{7i|YTEBPtNfHU;vs4||4p=E<O`*-|zs
zn;VnVr26ywFSBInl5DZ6%9cBQv!$9kRMTErk`qQ<tCdnx_GHTFYZ>x0DMPNV%#aY(
z47t&mdQD}zJnxn+9SoBtPAggR)stoQfMjXtm`vYEvP@i%EMqP!<fgYmRF){jW2r)#
zuOq+YDWr!zZ&8Diq<iNi`E^_&U$!fxCXTnOg$gNYqmWnMw}{J`Ez;}W7BLM}$fX*E
z<jhNwLqC$l)ip)tu1u4T3)01R6a77xQe{O+iuCQo{v6XJng2dPI;~6KZfk;E+L9nA
zTjWS)WVS@pA3fJPN4gE#Di$$Yg=)A2?ah^I?D6SUQz*A)7Rg=S^YF4%5~$zIF{1}+
zQJEZ#<L>za?t}$bO8e~na&y6c*^*f$lXsF^R+Y<YYB!a<O}SK;$tIgJF|I2Uy~gYS
zILbSf+AC3UdM&&6H^-nhEm7^%2~p>Jpzu`>EbP@AIX}lE&3+QDE}Mj$qDi>3i(caf
zvd+Cp@Vz<(wKG*vYp05@t5wm*gMOR=yg!XpL#-cuY#Y?@fv+KChZ-(5u>bF_I*j@G
z*HWLisOcIQWuZZRSA)-28aTFC0~c?rW5p2m=X9Bb!ph00p%*Spp@})vZ^TTXU8;}q
zQw;E@ogw_vjL@^l82`DO;CKpkXnsw*r#1ss+yB>nV~V$J&7fUnhI8uXDBEL>CST0)
zFa0+I_gG+dTTA>N$JeF4R=9uC3IpghHhyP~u}5tXv)vZ=d)uKykR4XMx5M3y_81)E
zfV1SAnpbn6W;qw1Pdg*R)CI#bT#&n9Hcky$h^G6NAe<TTtK=Jf@=XlyQ`ZN$VMMMw
zJllI9sJADYnlbnC)(eX)>7#h)joC|m&_KS~mF<TgHmmS;<Z6t1xf&MXYnc6Ai_6h#
zvH15oxElmu+u8sex)^{#tpjnAoO5suv&A<8akgeXH_wC6a4QHolY+^d!R%WL#>}oE
z7-JWLf3}98EuVkWM)0|qe3Q;Dx{D*jus<pc>rb<rt}UOt)tL$Mq7O%%9WSMkXg??l
zt*Kx5eA<W?w>P3?LmZ0p;-FM2@FZM;HgA)lnwf$XchfNN-*o62r6VXc9TV@TL!UV)
z6r^Iq*Hi@j<qp)BBs5~rUXFh(8f((a?8rXCx6v3+ohB<Q8t2wTBWek6RO!(eTuCjb
zHX8q2B;P!umP5VejCBlF@J{up2fGW0$6#LTSS;uhhn_Rzu;NP`qB_OnO|t~td=`gx
zXX9{<+Rmjl)NTgHL5tqyw!>o)&#uE8ZDMdOIU4JEo66b08A(R$OFX{`!d}0QyvJob
zZbbJR8!+Et1IAyE!dVAqr*1}~mfoAI_ab1nfS-*|!*OP5IR1Ga#$Ki{G`Sgy$rhmq
zJR5=+yiH9X6oO`M!3gLSjMSGw7{h<3S8oHbD>(osoCBDJUx)MM%))i`$AAQX^!mJ-
z8JJZ#zTFoy$9m(VwkIyRxx=Wn8}jD4V!O^tr2JTpo-LQbGlKl=x)lFM(_6+xweZox
zDt0HwV;6RSfCc|GD4~E#huDP)27*d=&M-xnv?z+*-N)|k?(V+pdEa|K?B5J9gpABy
zdu5jv^U;#Iy>Go2pk4O`$cgfY@gYBioSToO2j(H8@m&7Dn2m>jW}>`w2G(_(j?fF<
zFz5f{f~%hV89iYb?}1(9Dj#ojN6|!ggg$XYf~OmtkGazKO#l5C7liG1!Ss#12H59<
zx$RtGciszsu6kjMi8m%U8V92{<KUCy1G8T~SiE&SJH#iTd8Y{|*-OpJVmuUjAGB=f
zgDoCD++FfPe!35$y?xMb^*GcazKLwjO@<tABur;kv7<MZ-|@m}y(c1wy*jk>!24_N
zaH>X}73PBYo@1HM8iS$3N8=ktW0T2f1k@Of`gI(!c*jV5?Klz>*jKZ>l^rs(ZPCrj
z4!1gRGwL61EBvxWm+!XNeb*L=HEnU+-x`BW=(Blbg~nH{@NV;P`oqcLj2(`H^!aA5
zvO;s}LI%t%9Y|8(ST6-y5NFifuRz!q1=?+(R#e8_-+Tq4xRvnfrvl%MxSv43O!0K~
za;>+5b&eI*&m|92liM=i$iduEU{Iw3Q?nKH(<@-ZuD<A4E8Nf%7m>5kM_OU>Y%9ze
zWrdTS*md)skKL|7ogf9e@cX~v=J4P7Kr#DnTE$Y{;dNfCU`ux7SYVom1q!wdgI9xL
z@Z)P~%lsjDdY@hL^g`Oi4}#-Ib8bMGqtehEPtTZPXrUSIB$;tPgt^}5h4PIUXY^Nc
z2Tuzn<w>C&d|W8oZxqT19=Kj8JI)nK!r4MO{+nHHJ&I)Qz#<7VEt2U{B%dc2N!FAi
zQ4-_q-dZHfDvRWgDe*{;6|yma*<>Co7v-`aKVLE(3q;+&KqfY3Kgv6LNv`F~PxE|<
zFe{J&JDI<^TOie5708XB)SDWtkc59$NIl{kL*g3`;+tl~Hxd8SugC{g&!ZldNF6Jb
zc<4NHK$A12M^9#Xw&%(14!QESQ;xXT$dTdIa>S}mj+i#jks@j%-xHXrdP)yW#%ifW
zua5bza+w#vPMH?vqIaMtHnc>%vx=olOtDC9?hu^krc?EDdF@y(Wz0P9I94uAI+sh2
z*5z{XcbQla^Ry!_+IqZP_I)dtDJCmrCo$oeO8QNx!}UG9QaY+v%4X)FUhwN(c*AZQ
zk3v~P?rC&`LYdLDP#hE4op(219%U8CU*ftuo2kdmT_FYU3S{!^0x|iNFOjSA#eY+t
z#3klQ`2=E|u6gqIL#~tz$(6f(GiCWdnPNxYW@)EP>Ge57<`RQ+tC7KtO7b=d>2lXM
zU5-sim!WRy@-UDXW<|O*t)C&!8fVBBKF)7vx?CW~^I&|MZ0L|C>f3s$s7#ZV>NK%V
zOOs!A=~6?@kK1R+1XZSdpzgJ1a*hm@Tyg)CBVDg#i>^3J<~GlgAm<FZd_*UTDLT1o
zpp#EKw9-_sl|d0&Y4v`ExEB`6o}Wb`s$!YR-^qv#CE`}MR2oH<%B^P9wW#sMMODaT
zeT7UYtB`W_DyheOTiwrV<>R>ZQru^&>{z>6UVf~UBKGO*DqJhc>}5_M*K;a=jU2UF
zE4Rwm%G}je(rwcVX;%4C3^VHB?!uPvHT@Tn?*GEj-k8_m#(31p1oPZYaC5E+rpB5e
zZ-oiUH_?mpp9y;0Hi7=K85%t{!w&~?o+;)y+LU?mI)jkF!~KdmqK}$mVwpM2%FXd_
zg*jqYo8ymq5O(u%_1>7{`FHwwS`5M{A@?(v`SL{OunbJ#)Wihua|fc|{=xXhJqL9k
zLF@lUplh}@R6T95`!n;Lb8L}i#?H}24)~bO%+w_ZxLdO8VzVPg*^h>O^|84Ab}SMH
zI^ibqP4XM|{FyrQb;g<A5Eo?ny5eP`D;luNcx*p6)ZfQ093yuaPVqp}eh(ae?*V&P
zPvp3I;o}T%to%I=(G$i)X-GfiQD#DhO~l89iI{nFBF>x6K$GLMFd%g{KA-hN#A84F
zt?3Wf&D0Z;7ogw|dyMlIqNK}W)PA=ZKBEKBl)bapD(Eln9fa+>m%_j-nEB;kc=ro|
zlWz!WZ4Uu{v)lM<D1t1*U>_ESX=jOVzK0>Ce>fiag=5gEaMV~5fpG!s{J#{5mjj~U
z6+{l_cofz*i{@r<G;(vJF^~AB9(AfDW~jE`<4)3$So(=$asP5GR?%Oq92k#Ux$$`1
zBLUj71g!6nh-11$eCm`$JtqlgYbN7co(6~gHCW@F4$txoOe@Mkt=hTtuI6H-Z7z1Z
z<zhlgE-n@3;%Ig@y06W~yFHm`a5Dpj%e1)StVX4o8Y%r!F`fRJxSc9oP2hgFzX}!8
zRNO#P;cX>3o82lzpH^WPA2a<Wy~ZC@>;p)JNyAj&r;43OsrX~8#&>Tus<<b0`i_R)
zYYnE}*WmJh8tho9!TBW`v}mV+PnjCS8>{hb3iDW3Rd_Lnn5nu7Gt-rrVWgz)m;$#U
z^cU|+#tX}2jAO5#n+5&FI}@>KSR(fCO2Bq{Z+!R1W43iXdK`_z+L3YCcr+G6NAUNz
zGX@h)V^D2lG(IY#v1(lu*sRAc=SUdbjYJ!(aEyK)hPcQuY^oN9DL$dd;`7?6K7BXs
zgAsmdDf{w+FpPT&<IXSUmeWF*3||0;8-BQ`nvXGd^APyYT$KHojm48@A-2XW9N#(%
z!Q<z{uot@t$m1NG$E@C3KTJ3@9|v~NLuf;G*%05<-#Qb`s%D^gr575^^TLGAUTClM
z#Ina8*fZ7xt;pTDCcER<Xddjiyg$+nCOci>BCfc2#RXIIU9dsJJ(Cr@ztsgzKDuzn
z%L^Gf^rZQDp{kTVnZDyN@*2H8LEb1>>5WUj$74j;c;r<Zk9z*xaif<wGIJd2FBu2h
z_2aO%gEw9WdqMHY9W^hw<D||L#`&H&LCjUz*#j2r6VFX`gYP(Af0&L!#6<S)`PmXj
z*`oD1cGU3ie6E_=fp7M>_s$W=YB=K8nvr-*{$^aR9a>hmgC?1M#IVKux^_5C&Zc94
zEo^^`z>Iiqc7Gd=S#7v&`Hh^*4eCNi=#e>3JoAOy2Q7vpI-gkMhXRQ=xq+}t0R{0$
zS$_rAeF7|bYpmpbwSpTj3iMp0z?Xdrl)O`5cV{culXp2W#R^?Btl+KU);IMj?-mNo
zex!imH3j^Mhx%$1xW7_?Az%18PF8qIOtX286_WVa3VM02cIVfrM$PNI0^Rw&*Lf=N
zv91C?s{mT0QE`d7l&9b$v!r*2TjE!h1!}akKr8=YNZd6PkADq;)_e$i8+lz>F^FDn
zb8PZ5$K^|AXyI>$dR@&B_rVmCZ<=Bm@r^O@jh2VzOQG~2z6l||DI~tp65I6Q@r~J?
z!NfKOX9^{c_+}sRO~wDkH@?I-&xmgxPA`%{<Z&KtDUzSWH@4K0+L6EcfA}Vd_+~kA
z&XvD;vbb8ltZB*~+_(8s^HRPX@18GbJqx7kN%qDO-{@YjV}(ABU-hU*(N{a+P60i_
z1@ieg^``>`@^dD0$u0#_6`L#7I%dnO%~_K8HeIsmExFnvT^c@16R#;5yv>$Rm$GH`
zhiuV)rB|d`j%0Vpk)4AoWPc*}t>adUIk}Z?#3V;|m&=&%Wn$Q~Tvp935ff#x#Q7D=
z^ex45c}R&2Nh*`+)yn0AIdv)5Qu(J3IhgQLsUu~w*Q!kB7nIAu{qz$(FPDs#D`kg*
z+e@=o%B;+lV#uw@Pu$y1rMD-QywHP^LV0afD2?kDN<(T;nf7_|tAD;^9A+kp99P3%
z1#*GfT;o>xa_(cE^rLTBpOq)8=&y<6ZA4(6IGE>2<Cb|c@ky=}x6YLw*E1w$S-Lc^
z%Mgp&88YmBy3Bf)F2S#;F;Q0<<eo0c^w0#Eq)V|Q-{(h<O=P;f{+uq~{-(=G-X5Zd
zxGafW&B!!Coiy3FUoX0Jy#xiP$;z-aF`v!*ebdA!Kqr2?(qz*HotRo{q^`Y28o6j>
zWq^kL(;6upt(7+uwernQCx0JmW$P|2ebZXGkffEV!CL0=w9+$HCnoz>NMKm8yt!H|
zFEUERiumSGLa9VIDih17GPyH?o~K^yYR;&T4Enhj-KiA)hP9%ozEMU^SS4X6*faF1
zQhYbAk+Q(G65D!>#C2LD4OPsao3E9td28i7@y!G47ZOqNQap%nIxlL8nZ!4bJpM%!
z2V;8tjZs8=6GVJ-i}<Dk@y(xn6LclMSw?&lL40F$(F{3{%<$g_{{2idN8D~=oQviN
zyk(Buv+T-Q!%X>NGvc8EI5)o^;xqc;ra^za>DeEZv-%@`Lx1+F3_#)10mz>-2s@$%
z;mmJiENp0kr3nL3yVqcpZ5Rs8M!@0V2<*{WBcqQE8ho-r{b76^`<H#Aa~<#{(*dn7
zIN*@2BcAPcL|FaN_$Pcc&a>a|_yG2{ZzQ+<%n8r?I^#;9GxjuL_vvg`^di2ganBWl
zjNEXuzzvOCxZ?<WaL(=Wz>1X~n7hgY*7rQ{r?Dqadyd1^#k?+UHUWlD=`H9oksUe{
z@$~RSbok63`j)=v96THQ-_OBL>ICn*_@Nbd8ma;pU{0-t><(hj@k>7hW-`+|DgbTi
zwdi=1dx$nm(O$otp4?zO4G%{D-XX|v;||r<5Xi3(T=)`-Tf{eK!^5z0PZ)yVhjDK+
z953dE!=QcyKC}Ne;dTV3UX4VPeo^?YWY=Fs6wWk?#-vg7yXvFy>v=S^5`*($F_`zB
zIUmzlG$6jY_<$V5ggBfdhhwdehg<suG>=Nal<NstV4n!fj!6*eHOa4%pj)1dY#$Ah
zf9R<HW#B+T4)>YZ@ke~q!#Wo|U2-uaHWxdya#2T_jjnmws8yNC{^|_u*`UTtJ2h7J
zRU?d^8s%jbj&4)ocYzA-lT{e#pu&?eD)b=!@S{Jbi&}-oh19T~s8EeQ9E0z~Kh&!>
zllR&7RfPj)sVKKqV{?WE+n5)hGeLuHw>0QYo~apq#oYolI8mKBE3KM6EUD;anu_#%
z6=qqeAlKRHw^)gKwUo$YQCf$#+)Z7PjPOp$*t{YM13M()X<i~`w@W0ZOMr^rn@5H5
zNbO4e$6R>59&y|)jpg&t-(PMFX10n!o-P_S45QJU+}X6&<ZtMQI<X)eI(mpN@;SV-
zAQa8cg~Hq;6qm{0%>5pWAvJ?hxG4zgoq=d;xdchG7Nc!BdA^zpFgAl(JnQ-Nam_`y
zYjZIE{%m|RoQ263eG%s|3)9EWLbHZ`XvoKn>_Uz($`2M5^KoI`JoIcf7ag^}h+IDt
zwJg2h-_r|@ZRjIA=?Ob)PXv}T@7>G;gL0`&vD@+$aenz2dWZ+QLATx&=0^12RMCIG
z$ORkc(0>!|0xx2nV)8f^{aq1f?E&9K9;mgR{+&+TiE?nq`RDF<Xz77>wjMC9=gp2f
zZ*qs;*gei0L+rg#kJk;0ce-J+yBo@8v!~{O1C-Mp5Nc(QuH8mq{`rwOoo<Knfy_k)
z+hUQsE%!xj5i!^nFQ#+LQfZ5NhiqYGZ--e~cKFMlng<ScF!@C6!rs00d_9=ZjXoJ>
zxGx9V^10*ZldB0h&E2Rm);Rub1Wr?Ts?mQqn%5f+j|bF*Dy+DNV1?0pt+14D2Qk0Y
zR%3-X#48<-DKMD)i5oG-vAcktTL_KbQyZE}u7<c~-DCx>D-?LWN`Y%@6ll=V3b6_+
zY??(aieE#QV1>*4_|qT2%$NK;=DIz91Meafc$7<QbDw@1ay~x9M2*S&xDF#8>PAm-
zZ7WRT#{xGgsI@5&I#hwS{Cw7KW1|r}{a8U)LSONwJWK4U&rE561$JE;hMzr$F~2bs
zO-hF#>daugq?h=3XKv|CHb=E>W@y*f3?}<c(P6PE{wPe*x2GuvQ{yRoRw(Blu~X$<
zp<E?bqr6xs4~T0P@o2^4I`K_&;u{a*n@y+4?Hn(ZqsRFEQt~~;MY6Y~NbapEl2Pla
zKd@)6l=!9%vCU!fH>10+kg?rXh#B!sG<6IM`a~3AdifKemu?Hv<SaRZ75lTKT~wYd
zswj~8#4--VG%@Y-WkZt;X-_Zgn73N_uu&_jM6E3K(8`igTG`9nguPnnSfrCD2X&Hp
zTqo<A@cg_xO+F;2Nle{zxxOMp;?88t1o}c8=!Jd$Gh5nmhoF~Ph3q4~F(0^E{{FjC
z)bxc^oG+2|9i=k;e3>lrDUqEMie-~$v3$=img2j`^7emNC#YC*^NS_=O|k6tERkx(
zB{HpfnIxi2ic-rZaZ9;4(@(qiE4%laujIddr7UA^>&}mIdGeYcFDaK=1BxYTU!ioT
z*QW643b7OZJKN{cbDS%l8M$(~bFNsXGWY#3Pkdh!)11wdmgRX8Oux<8ggm(!n<uZQ
z=gH+ZdD5YNo*3N9mCWz#wON-g*G8wyfIn%XyO<`9x9G3AP0Vs6O^(k=6BoBM?u}A=
z@=lYB^Z7d>#+ee5E*6i{Wy_~@Sxar|?}0SQNYl%w@p@_5S}!|S=%s0}UTV7N<+7Px
z#?{ox_77U=r_#xifm-R~s1ZYG#KA!$Sz|P^$5kUEQZ!O$i$+?E*UJ5wTG_ToOYfCd
z{v~(wBTOq}7Hg$;fL4-HwDM15y`)ly%jwMjH#Li7aF1fiN-CB5hGo)JS0<Y+%O!xn
z!!Yt~2m7p+tN&HXux{(bbyB%Js<%=MI<1yf#5`N*7k`<;-O=jo1LW~AW{nv2UMpGQ
z%w64GD<|nmE_Z(+N4CC{8NX^nj<iJfW<&h9^<P-V8DrF2W7Kdq!6|A}f1^$CJD2>;
z268KhP2h3I1WjrT#9rc?#>6+nYMR5mzBx`cWJgZ08J_Z}?rVlxL;K>xgTAPm*^hcV
zF;4aV_}jHV3MTi*u%iC#rR@*v1p}bkF#unyjPdQOG0Ya2pu*ae*W>iVJsbj4YFZ~Q
z3_;%pL-2|{q2ZfspyX@cra#=kQ`#dy<A5RLZ)Vy#!t$Ua%4?0r7i#o1sym^Li4$s8
zI?<QugrPm1@p+LmtnNGG)(lsyXFlA5_$G3&8|o}|L!_ZQ7A<zikK4SqoZx}&#5e1n
za|5TNCwDJB@ogphR4;nsYc(%q^BUIly%*{^jKeN=&|m9435WYm<$l#vbT~T|9s6*b
z=P>hB^XR|%&krjv&?7-!XTwuJ*uM3{G2=xjX~*uie}hm?zm4JSVCchwQ9}M^xifK1
zSqK(i2*xfRKko%2IVue2wuRx!6W(?VN3Yy)`a>ho!#RR}g9vUUaEH2A6h^9|aBNi+
zc70{%%Ft*Wj;3z1JsNEh16>F^{<sA>xqB=Ml48-38Rr$uP;D}zzc?<Q%mdHo5ec}=
zJ%U=ciFj(5#Oq&f6&NKW#4-h)EH&`HsDquM9^>ZcaAQ0dtvctT-EiiloOAJXdM=9Q
z=E5Q@oBL$hSTs8m8e)(}IyDZ^dvlr@rZwgC*_=@!kX{>8W~Z+EsL;$xg*Ky9=rmo0
zIYBDiPgda{er(}I6($kae0!zBmzyfwyHDKnTt$zQ3I)AVF(6Wd?)2J(I%#-brRQe1
z8nu?F*$YDK^Enl}u5yEm{W(6nRnQPeb!w!-{EbR%by8yI;}rA^Od*Cy#_X76_>rSB
zj!8nD>O3M6@%(oJeupK%g5Dc9_C6mph-XJ}9M0E{!>PDfc-COXJ~Re9zAzKMh+Q|Y
zqA<}j3LfNcDqcq5rDr%g{2PWDw?eRGQ!rj`4uMx;Fd8)uhP+u0zfy8{*8;Ja9<1;c
z0oe3y5p0|nV$vReOdsQiEuZHhV8dLj)69Wg)okP(_C=dCUle@x#ijF8@nXwl{EeE3
z*0Uzy(#-KVm_iLmI}R`3dSh$2H$qdr(8tsZ)oXcSC3UGh>aF<Wf$uXtpgHZ1K4aW5
zG~W%keaK_3azW}+7j$!V!KiyKXd3B)*TY=M<+#9f0ym!GT(Fl~)re0nsJ!5cMSXZJ
z5a)p|58d#huRGqZr$@QA2fmtl;Qm2ksjptR9_oo>$2~E6h&S4FcE_XbZurZMh4xd&
zV$5y_fVu6-VvkZ16TQ!k#7eClCK}n{@KRf3k7i%6sVy9cSLX4yl>Xj0?mwMuXNPc4
zJK`<&Uk$Uvq&K#t^!V5}Hn4eQgMxQ9DCYL118*-qv%$Ajwy+;$jV{c7FR~ks743$j
z)@Lhv%dOBVllo90HKlSZv?b1&9>=|2YDRa?a^vox0ujU*2C3ZWX8mUKaX|7RpvQ0p
z-uuvZL+n!7Oo0c5tPltVru?+TeENM2UI?D?M|(1r9m59zt2@AlE1ZtI1$6z$`SiI8
zG+j<U=RXC`*RX<`TGVLboN8_8Kc<$|>KV5e4l6K=dRWC21@s*h?2!UE=YlV_0QV%p
z(?6D&$V_<1B?~MXW`Vnz!*J%wP{eipUuI(nhST$y9y$n~ndXQ%Vnz?W8J2{ZBCnw-
z?(Z5%Z{I+4TS#vFwOU$`t2z9zP@Jz7N>!F#u4}1F#OWo6IAk8rZ!LKgh3e&nMlWk`
z>BZx|US_sP6Q3bzGSn-J8LMo0Kb!brPPT+5XUkCHnG9l^)jZ6IZyfmc&xN`2WlFAm
zQtIVIlwRf%-+Y~?my!<Y;$1US(ju~CbhkWeE7X@d<V(l78FFbO{U5z_;zAA}t3)fV
z={#4uYsJS_D@8nNPuEK5HLX;Y@G+ZoGIx)TS#O<K`RHWRJzmLG>exM)A<ftYv+Pc`
ztoo5H{afcql1YWsBfe2~SS=~{x#P5nJb{0SG;=JK-F-?V^<c3CjVYG<QY;Bei)F>a
zVkw<lEHxJuOE^6_K@aK8aVU|^*(FlzpE3!jPSrJn{0%)g$>eA}xD{nYEvn@M`iSqA
zOFVZ9ychF(gb@4OED}#*#nnM8<Zm5%t?3D#^gdUtDsp9=Z>~7i&XpZ8dDOJnF>@+U
zHm=Q+{me>T8JHtC3bSR%`fTxCoh@sdW=j+^-={n>Wa<`re&?o3yLRdF{brh+-jXIt
z@+<wxV@>c+lPY4H=8`4_J=3Hcb*Gv1{`#*^6H{VGKjNY}*Qj0fqOWGBUhL!b@_e*j
zl;3qSiqF%@33_=zZ1aAFR@7BmIaEU{d%kEyUso%=$lXl0&`1;F8$;q7Gvb@{WPWU;
zhWiT|nL`Y^gSf^=qZMrwanVApJej4HN4{G5LJp@rIi6jnT4~^*lPNPY#9UJ(UFcP}
zY*Qxc)H1m@yj<MYmCFO_hv{akWb?$;vSAOs-o!mxK5v0vR?1S}RWf+bYB39|lsj!J
z<@BvexxR9Zyy?j7J@L)dhxFhqsgghC%u-pelPjaSy)oy3TwVWGo<8~_w<i6PRUQ6{
z*?-lT2d;((@2bJ@PIc`3T^%89YOs&o0Be>Qpyw(B?(wuizs_x-=+Opse|N)&W=2?=
zZiKUIdZ60Wo@f+egx`rq7?fdzQ9t_O7V*usw*6qxrXRd}_rs&-{gHcU0ABwd0OcuT
z#1AsY$<^j)vU?Cdo@95<O=ipKIbK(vc}G7RWHq(Ljed5h++>Fp%k43SdGHg}iESo2
zqRn<kZXAq8hSzB1{AE`=^{MI6PME{2#o*q~cxvs8o_C#5M83Umj4OO@yQ0<*H*6%n
zIr<N;GZ(qz*H!iyJOBS)<D1t!5WwryHWNJIPkeLo0`bjnPi$N7#qL%w?nHaz;%9HX
z%^8Qb)qU{x_!LyK3(%u{8umENqW^dSB92i@i1&ltdOvLH=8u2rJ08w!^=i)+Be@YX
zR3(9Udngz`t^~v4PB0RPXPQ0;My(^k7|L_w4Zi*GZ7>e43B&%|VMsM(hN?6iM!&-G
z&M^YdPe)+=gGg>>L~&P?9XLf%)bQE&M~`vm2>M;OMx*D~Xbk50_yfD>9vH@Qb1xQi
zzHq<FH4YcL#pBY_cm$t}M;rQSP0l9ZSW+SmHB7?lNN#FBPe#_^RLrFB#r?GwO|I!s
zbY6?UGjfoal7sa(bNGDZ!puJ%vY*|0&gpP>$-qYs4QdFxXNIehRG5meMXBsXQ{hz%
zJvKAQ*?3Z)>ZC%)K`I1~P$8UH$9@sJX{cNMo1;R|ar$)FiKCzwx$jwebuKY;b%VOq
zJ#soNsaH{>5<lV_W~Exz)L`K*HLL^F*hiiy`yKoIE~etpS?*AzsNmLF1@%28I;)g8
z+h2(_$5L?BI|ZX2G8aBO87H45;ks`UE^v?Z*OWx`zng#)lM=9>-kbGP<5A~P92)t?
z;pdZBwDFBa?T0b=K9PG<+`}+%iAKv~)C7&AFi;tZ`I{o3QiQ|m8$HCkL(nWe7@r-M
zLsfkl^4|o(*dPd%fr0pD7Ko8P?C@(Efas(}u>P_DiUsTxpxNSk={&?uoQn{dgWj`e
z!*rZ4mfe_%T|Uz>V96BXp-J!vpNROW%q~qCkB3n{m{KqfYnb02(VzX4N@5J=|LWiO
zL^3s?N$|v!JsyZM@<3;11z)^&!~2QE^vn>36uaQeVi$ZEOFgQu3+CT)M(#IfY@tu>
z`Cu2Uo9}}CrTo}a7hLDQ$<)8D*hSCUW^*?rCA;Amy=!}^Hx+jGzz}jZ0sq@Mq2}b2
z=z(c-$?v>!MLSb^{rit0r!X2ThdE$=$te6>HWG2W>=5T~2ghc1SW;mNm-)6B*xQzV
za$9KsvBmZ-wkRB7i%uc7Sk=%Df#hwjB(MjoCqM7LEvCM-LEQ&7oOHK=#Q__POtQf$
zZYNyjjzau#_VzWmhQ)>vIAhGt;44;WO5Vo5g%vIlzx+t0KEyosb7r6RhghM}2L-Q1
z6lk<pfqLXwvWaV6eG**j&#ACOf`!B{!^S8O*;#=muYlknAl62(_LC*zo>^k^Gr|9=
zN9z#Zxa|URb^}xQ12u?I4)gK*S}1U2Cckf$0?VEdzZlS0T!$OMKNYa&+ZRqKu(*Q$
zoG^axk;FQ+6&S#+;UC_B-E%?AEbdHQrtfALcQgwukY2+Ar|B#1TETn<y^B}B4~D(L
zVAM7kgzojsvBt;@Wz$Vz`QJc1bsdN%)du3&NfY!V2S2S#q8!|&lX(vcWg)qn*~B+U
z)yvUXz06vomqRl&a%GZ6zIgHcMGayt&tt}J8ad^ok%+$<**8`vo!oSy>y##cx}?eE
z-f8lU=hTkmWVW==mQw0R13G2Pzg@EB?n1qUUD3(WIeJl0>#*Q$OP-TIkY_P&m?`~z
zvgD@?^FDXzO>LPWk4Ng{T2GxAQQtUFtd;mQtz5F#%6c=NBdxX4aGF-4u4|>qW39Y|
zPMZDDO6fDLbR*`ec3CTm)5Jt;bu#FKPAsox%N%yyIde<NmRaTT<OIgHUM+>`<<haL
zTn^1C5rcc&Z5m!IhU95lGS`&Hyt|J_F?S(~<;cupiQrK^p;*pdFXjd}@3YHp)PXYb
zZCoxszVzTEm&-wNG^_X0r%b)ero3GKc+iv6y<GlrE|>3drE+d#5w|r9rG^djVN3F5
z^{_k{xGz`Q$K>)py-Ec+@=TvC-UV3_e=$>@7O{uMB|~=Fr-^>9M#iM6Wqf$5tpBUx
zUXx1FOH^WzpDqLYq>Cl-%%jcZVCa{Lnw}<EK6*)a)Qep|J$D)OlG`LrE;UM%@;!Qa
zT%ION=^1`Zz3S3=jZ9mk5$Ua!{K;Bzo}iUl)Q^r^>%`zJ^(5*_xec^({fS0OuW6*j
zNh7b`lCv46k*lpW68%*zo;@{^X{C{S=^FXIUL((M5a&GCh#PsH5#-z4M{4D~1@Y)`
zt+Y1Q$}wlHn0jg1N2HM*85%jbRxi#Q^|Fll^J(M}J9jD)RW!BF-Z>J)KA*X^+0yU{
z_suiOmDOWT|46x5(=V0XlltQH3dugSO6Jw9<c4acbS_yVj&0U*LvyWkugl&7_5xU~
zsgj1oH+2WEmv^nVh)MZdx%}>vI5_^2pu@l9dY9i)lJHyBe)}z5%Kpgr6Mx06W_6hP
zR>%4i)v=|Y0r5sn*mkIetn6B_GN{cz#@aYHr8T>UJ7P*+M;Ny4gx8<D;dMtNG%_;6
zeT5N<^NnzOKu=t1){hvaAA*AVbJwmv?v@Xr|CHAl^y}1FHwgcnAB4OkgJDLTlN(}#
zCC;|Ud2fq%v+dA?+0X5>>~Ua<J*v@bIA;<u&K5^F)E<q-&Z7}_Yb?8Aolq6$MBGH5
zaSvydjBv*HN6t7so!6=KO~&tcWqy?SW|JGbkiS_niyi=a6P%npc+E<k>aquZ65n{w
z^@N`IMoD}#?uRG)eZANT<^>C#Hw?abV-I}>P4}>y_23kYn!t{=e|%B0&KE=0&&P;p
zKb)gSVo7Iz#P#vVp|J}v#+e!4&;UGUw|<Ws0odGh3G8PF;qYSi^YewXX-E*<VuQFT
z6wDm@a`b#3jBPn#*uI6maE-%}n$6v!kKuSid{b{%1bo^=;z}28Gfa%aEq2P86W<uH
z=WorlXn2snx$hW*mh|7;ycGjgU3Sw&#^S^JSnPHtf771(R7>KaYD>;yc>-Qgn~94_
z#KbK6QRYz-+CY8id@ACZN$PY-iy_@~^!{iuu|SJji?q1Ek)1RhHTb$kjZy5Vv0qJI
zhMdVO=D=T*Lzzf_Z}ZGl^kqi7Dw`hPhU8~z(hF0UJvXiCx9Lakjj0NbmMZx3w&r*h
zp6bZ+9N^a=-{ZTJKI4Nb>^!Ez%9G?^$p7p*rNX0{si>|Zd9PH{tHFGEUlmrhXO~VF
z70xqL{>wiV1`m~(|5FM58zl_4v9oy&?>AAR6*s5Oa#Qo^f61sAlZ;iTld#nx328@}
zof?ry?^goc6bWca@6C_l@fd$N4gq#?#H!?N?6_lkI0o*D81fR)C^e2ohteo)YZQg!
z?36wn9s#Eg;W%<R6z8);U`K9la?j<c`(`P|T@QlOt3cF7AR2C5g8C=v#Y$Za%XW)!
zGJ65Ny#C0kn$PED9=aRQTYQdLDAQTk_IxJn+-D%`?=;LZn}RJ@CSlU@iCFGQue<wr
zY@pwzp>7=hJLQd$+&a+4dErM3FO=-|guxtgDK$LtJd(KMi8}^~J9;bK@aTX1<t!KG
z;^-yzcY)DJ7d+nQjN`e^__5lVI-)b$H+O-`jXvWvX0)%n;0g7sx9r4uK`pBWyJ_;r
zyJ0%LYn@xNBZt|mFHha*A9I3H>oMs0+5vXg?cq7X9y(Kdv_Ccq8wQO+?2nOn@!v>P
zk$;`GgI#(f?XcCt4sDBVF^k$#$4<mGw`}ll4O?_*OudSDCxm+}fv0RyQr`~Uy}2Rf
zV~Z&Uwph%*J%0=CK~1n>Z?p||2HD_z0lS3P+n~n|b`uX9fq3dt!PKA(h#%CY3QWmS
zU>f}}Cd4vp{jIQ9$87c@D>OW*fE~G;pVX$>m@1G&n#pzz_qnqK2FxqDlv<+WB1`(B
z1+QiZ%v=O^e0#HtCH#FYacZ$8=ESkfVja6lA6jBZ9`&gTX1ePEgBAkwz5qj=6^Ntm
zWka4Q<R!hwueiBD|8dRZ3fv*?2`6XNg&63Ru>zgnQJX5FJ~jf>H`(boN^rHx636RV
z;?HagOg=RXf4UFDdOv10a)zM!>cQYR1~-w-Q50mxj%HIhtQ&|O4F;l6XMzP_PNbd*
zhP^jN!RC1B)IL$dh;dfCQiEt&DA$N@^yF_2lVgb>z9}QVSwVc0?W~bj#5bLYZ$|&o
zi1`?uSUc(D;Cr38f7VIW7o9ZzrV}S#`#<`nlVyf_$-k_VBNueyeO@O)XLXWsOee#s
zC;86O3;V{!;oo$rXq7G+@+?i)@Ze)t@n@+bW-=P0lip{vQoNJrM|wKiji7I%7kQZ;
zS_vY)37D#tcf>ch`srk}=%f+xO*HXMe`1>&2Y3!WN)2nZPRtII-)YHz%bsjmL?6zo
zhB@+uT`XqoW%+ZI8&&4?%n+ZfX;mUee=*lHqgb@WG=F)&3y5=S7t8wWBJs{BlIR&l
z;==d;x4lF%h;RPa#rLIVxu~d_Oo=8(lT$7;i@AN7SuUdkn3w9oE}fU`>$_hnRY}ES
zPaW^cD(-cgtdIlN`O^A%uB?d6mB-|^meNPFJT6BbhGx^_k|nj+FPyg|Lvjo==;=<A
z#Mv6LXs(v~0jctGp-OsBR!K*Hl`LALl3CBw#BW2IRMJ1=MPG2g81gPodbv78FF*ew
zC-+e&uTSXYj#4L!r|YG2PrcOrq?0|x8kw!ph;6J!&Og-1gPPQh`cgldq7`%E#^tW`
z(KIA~c1a^Quk+ZWkv|PIaxO*9jViSqRjQ?fm0F5As>P~_S{78LO4sM95`0@NCy1F0
z+iPXPAgy#Bq7~tDw|l5oRyNT{^&qu0@1PRr2P%nkNtOD<Y&-aK93wWa3{96W^)lq$
zz6>cjo+<XhS+bcpc^9?Df(YuIC(8tL%OvhcnV5bo7xNJna%L6x$v>`^F*_=yX2=>j
z*JZ6dWGA4eQI$N3sFE&Qt3>~vIq|}EqUU~l(&o4F`_p^LeEdUhRQ{4Joqx+H^>69c
z;E(LC{a5n${}spT)$wI!bxb~4ojF_sJbz?>x{)<8y|5YbUp7NStLEre+75*mIwH1i
zCvqd5knPzS``o(Vs!LbIj_roiIY!91XoP^4{ZJ6v50%dS(eh@0n4BL7!?y!bvceou
zuju1BGzfMZ24UAaUYlHC_bB_e+<WnL(U-jzIy)R%X~$d=yNlaUr{ZS#hfR*?Uvo4<
zh;OD`8Vg%?f^CVVM)lr_ypc2Z+EAl<>J0CRF8C7ff*N72u-?cV75Q_^6Z9k8a)TrJ
zn}wr1;HUS%k7FLF%{-OX(G$0Lo(PKbM8`Xx@Y=~d$pUZeSiua)v~kGF7>Dsar=ZQ1
zDJW<(6HVQH(UTt00?m9F(;qRXy+3xd&$!u$1vol#0a}k*fB@ng*N6bLqCW92&vVny
z1Yl#`C2(o51iwx%!C&H>Pbop%u2_!MGeQu=ocOAeFx;#ij*%(hIQlXitw%>N6B~gI
z?D}gI6p6jHqR@+-D<zMjVAqQK4BW8}?ZS@TF)>)Hi@~>ZG3fR;2HL^IHrwdG@rz^6
za6Crj$HVm>cJ-3GncO-NsZohInv{gDOPJxVRAKyqR1EK^MOKa$#Wi$zze|hS><>P%
zlYKOy^z>G#G2Tgyn@ZvkUllH{Vivn|D*D-{!X-8p3nHl%QDb`EOvQ68^(bPF?th6n
zno?`}SA{;rBqm)|XhDpkx1|S|I@Q)R73!C$aFDqwL;7`26X&$tqk=1s1v^x*siOaw
z+{|Tv6&_IE8q$aSPg{PS=Hz+msqo)NB|hF%!u|&F%n>DGv`W;mSHku~3VLgpo$8iC
zFK{vp&3J4|!m0jAuw6ypQMW`~D@=g0V*&=yd(*cIH#JM+(2(96uL@?Tdc-2NECzo&
z#lVfdehuj5{x>EHUtdLH(U?fwBIdluJip<FQ0SMiml46ZUvoL`U08~zyMw5^QkSGI
zHF3leETY$Z2D=~MZdiz6mJ8Um?FTEL`G{t2>Tk&$ob;NFGwfH2)6PU4!x@Oln2N#m
zxY@mV5<1SFhy>dS^y-WUe0<O}ieBD5-k8c9_xK?Cihp?`F4Gg{Lp+gtirk952P{{+
zW7=moG%$7p+r^M>=K|w7&iKh5;1+wid9ckHwS1gm>fwwIdE7#H>5KwuRYzl7@a{hs
z9Bb%`#rCeK9qo!KM_rNE+YNuXfBE(m_i8&5`}A}|WWpF&U30|L3VP%2*&|qCkLw#p
zVW1T`8}&$PcOwzrbtL9}w?pW5JJck;u_M0On8UrxMz)yo!3O>I&|~+4TTfqXu$dTV
zcn4cF|6_x^a9f;Xj<Rp5HN3u9!=?edBKq0jtphbFTN~;+HfR{Z9n4JbC-fMB6#8xA
z&sw4Ga|Hsnu`iDv;c812m=QxBg&NRMD=XBYM$~hk0(*&VB6kySY-DF}Uvek*zyqbA
z1&=uH-%azfM3Su~riT3QR#4AUU?Y~y99Ux87Yo$-VZq*aOIXdfMBr`qcQq7zohGPD
zUWd^#*xVo=B=p}T@aN$7*+Ad%E9T8rhZKmWkLRB>1x5tXbK|6dcROaSZc*zZP5EvB
zaC@h)^HfkH*%Du`TEKds1-q7q(Tg|~BZmydu!%$9G<h&;SrOmRBk=2vDKhw4I_`rB
zT8=bfXPGh11{ouBgfSd?8$(BYlh-a$ench8qUc0f6Ots~Qj?_mlvMHeO%=QKsj`IU
zthC9g@?~_YR3DZqwTN$QiErFT>%^Hy0Qr%)H#(WfBkqMxhVXd%Tqk8b#`CDYMJH`H
z>%@jf0`Er?Z(P`}lbgnR8S$OIkgsX-<ubLCS=3XAfx7t<Yt7{Oc#=-^U34;u_-3R7
z{i&_B@}QnpmfhD#=g+CK^lz$cs;!o!V70soQ;SoKT2_;LxEDx_vq39o>Cd@%S1U8$
z@P2i@WKnmSkxm}z9rd*OIbuW}r)P47R3xpE^*hRBLf>+^OYHJ>dZDx?o@qfX$x&A%
z!Q@<CQ9Bv7tVn#Uie%Q4B1zj-BJFFGN)i3A^XNN{LAiXMS1!HcnZ+WH<Hl`*-=ml*
z|6C@1++tdMwp1b_ilxS!A{p{-g>?AC-k~2kGBB9uYtvjQ9LbKnTI`Tnmcv|GmOMF~
zDISk9r0Y;-pl+tiZ>My5LyWii2mj42Q{~<am24TPlAPWu2`^H~m5MZZu_R5lJl4yN
zm3k?sru522FP6kHGY#}obxkJ?`TOr2p_2jR#aa<>sQG(*tka0e5)E+_Ju{azvhBS_
z?zSWUHI}-Pi&hGWbpl^$BsO0oTkEOi!~9g~xGGh;^Y^{pMlBE8t404XRmPX3(nFOh
z3(l#e#o$yK@j6v{_^ahjBdr({*X-=0l`YTJQW26Or%$KIintW{*h?u@>y@&*o=OZ4
zs>Hg7TDJA0204}<=r`$Ny*)z$c4f+E=FrpYWJ})KY-!Vr{MnsS_9#)qI#(w9Hd6cR
zTp{nru9j{lmDCa|Wg~k5jM)vCI(@A)jbAIT*;$=;fc(vuD%q2}PWHapECF}lNXVG?
z(!Ki+DcJi{R=WI>o~6IU&+(6Rss2}5CH)nzn$_T2mpI4ruXJiz9m^gXAUV7ywh-S~
zy=sOsi{|KG(hm18lE10b3A#m{P-k@~e0$azr7t@p<V|M;kjF7TL(|;9{qW>=KTwH4
zORoV4J3SB|UJb;gL8hn~WQrEdp7&lc2#pl%N4{c%1>0?r)W{BLQ|x#RYKQ47?BKK9
z4y!NO!tdQE=Dr+Z_{S0N#*Id9{V`b8iruDJPB7rc=G{)t(Ahb28`&9UyjHzTPQGuT
z3$m`dplcU5e6@1JhpBGpk>^g&kq2^1Jed3Tz|q$pm`{9@GmrV&B|LB3VvmU9c$`j}
zfQ*L|@TC7lbXznD=C>!I#iFVBl`#!<=g-IR*!j>d@`Dlin~}sfy?ZRcAmW=n#5eyE
z-;5-_sYZO$g!ra5@y%@Fo0D~xU?}lT;ISpRdp;0-h;JHp=DBe}2s}TK5112%YJbDn
z=N66y@!|M5DjYfC;kdOq9P8N~mvt%vKkTDW?M@V?Hi*Vd?gfl9ioqQ37}R3F-}7TJ
zNP8WFHyvZ~{x`en*z@OpI2P;f#$xlmcsviF$GAx%%*fv~2u?y|AhAUext1O59PXfn
z-3(@-s8J0~(&FBC?m=zU;M5{^=s}I6<5c)T{IbqLh0<&lR4b@M-Q@-X@y4X;DxCPG
zgaa|h7`|`KB_-ZmRbo8fZa^$@oj7Mbv5AH_XQ?&!FZq4HYuU#~5AvZ370wW2btay<
z$D`g_dUVL^Eamsu6+t{hJk`-gg`Pe6v;M<d`fN@c(5p<&#^JmY7JHN^EmG3w%cHjv
z?XS{Tyfg)48m8b)CcAIiCZk(+656#$!tInq%&nD(HuPEh{*6byNMgSl@#szO&9ypl
zh@@{RvR*9w64;k$5ChB5XgGg}LgjRBFr0}*r7^QpZ6Z)~CJf7yL-BMB_caWI@%`>H
z^xe1=&2rcUkQWHwT}x1MCjeUZA)#y$YB(;0=8ivB`m_JqU_QDP&c*2wb8z)Ocf8$v
zkyt(huii|<Xs@a8do&s0>Phe)I}z&s6Ohn%JQ`a2;K#gi_*p?N@rM^`PW8f_yPos{
zd*XaOPc+T-z%xS+45YuT*LpWBqAuI+D?7Z&&-~)wgrU|6z1le;slF4IhdIIaj}u-_
zamMSz&Un?u1$zQqaDF>Ep3mfPy1U{BbL7p*|6Kd(iY7sB7;wM|XAU~UId%*hvtKhj
z*a0;LbL;lLJzk~Q<8vhgKxU)ZF~d#=`;j=?dL&#w+u;p0=6q^SWzU%7u4_w9h@4Fn
z^`=}Kr0B`Z5U0#sVS_=lZSe9RZYpe`-n7aZld`NaW~DXe?_gJ7J!(-38>}?8p&!l$
z&8OMmexx<|J?=Wax56#<${fALe0Mqd8Tx~}QDgevp1JQ43OJL~dG<fg%TD@kh&di?
zRq%WBZTfrLhLUrcCwLbvFmxbCV{M7YgDhdkZuNa#1Zli~w6-OdpR>T5n-;k6*#aKT
zE%9WQCEo6$N9MC7Qirf>_@toBiawshfSSC_E_!g1*D;@UiaHf>mD?%>j5N$;1uJlI
zvI1i)_;b})u%jA?i2zic0Zty_jG@4Mwk2kja7Xj|Fj&nThJD<%xnFlEd7L3=+;%V?
zJv7J9R5NrMZi;(t24Zp_6L>B+hVs_{G+i}-*FXcXVA22_+7vI1h;JSe-%KOENhH4c
zv^+_c@b>TCB$;(6NzNTllBPG3B)e6zJnfh)Uz?@M7J4o!Z)?Sx_+|p}%`{@0#xIF+
zh;8hjQM2OFibn$;pPuSuJ^eAadD!tBTSSbr=K!%wj*i_W^vXP;A9i1wxXw(I>u30R
z#5XgEZw?XPeDAE2x|P(YsByU6(ny<~)I*4Eo_$P}_e<20Lmt3txmxaqsHH}vTDm2u
zCBK?R(hM}zA?S4_z8Ue7yg^O9Bs!B%I+-SCPG-vqW~y9?Z_J5ruDq)dt94~k?Gv>R
z=D!j~6w0T_A_+gttosuBW#$%%I-WR(ULtdL`HT=|yBxRy<;QJ=l3cNGn<wG5sXtLK
zc@xF{zNO{zb6mM>>P{c*p)z^nQzp&LO2whHSkiiu`}xTYgGUA8vNK0^|H_e57jxu^
zDSw`3*|LD!Pz77Eq;^iG1Ro~0sh%N+iqd7?m~;szM>6w7Dz_q4k~dBz6S}D6@mZzZ
z_EkxI1D*JrrAh6*dMTt|%CSAQza2VJP1Q*v^VwkuTJfb%sRn<y#>P6?F+eLvuWH0D
zkyw$u(Bdf?8A|@7`f2Kq&9rhvv~r4?l4TvObUCS!q#J7a_BEA0n^gIAF<JWGQi`#;
zTK@b@l{tG;Wn^Kh)TyT8ew30uj7qUKRLZJmN;%VCEk?F#`A#e}GblwW7pBPj7Aex2
z_dSi2@@SP(d<;}_GfKryfmB&YZmxVNHOqVH(t>>6)ZLlVkUh+kDzaop&1}hTT`ZFi
zbHB4^nbah1?3i6HbG*6z#mx(K4|X1TS8@}iQmVMiS(lwT--&O&<#E^HRF%~DTP0<n
z*Z9ghS)RT{u1t6@sjoiDlCeLf%K4XEEBqx%-Tz3)wLh}+(=VCV?6>@x@<U>Fev)IG
zKgog_^{{Pg1DxB@5N@u`kTJU%riV1g?)>IBTC)>QF6;!aRh{sL_@?0h;v2)BMyNx4
zv$(t;+)nmG-nV{8{@st;%L6fZ$3T3!I}m;Am?EZ$DZcs+LhY7=FzAL2=KiojvyZk|
zQqK;jOsKhkw#MB<)|j%Koj5_Z*hWv{Ip)WYKCyxQV;d}f&Cb!ke7$T--pAS&(<a%X
zZGkho(qqt;IjY~5W4N(91|PPJ!MCAfvFDdF7O!-J_bYb<Wq6>HS@FNEJyBznC#)aS
zXWY;m!PC6qG;$m|TKNzcv5TI4^jT9T;m_+y=stNWZoABf!>suTikXjb#5abaemIis
zhY@D}7)yN9)OG>J*)O0D6@cRt18{y#06ISmK$k}W(38KhAimjiatS-c0<ouf8N%N!
z!<Eq?XvDpuPzQQCs)u6w;!w=!6vj+!IQDoD+pG>pM70R4iH_j)b|hRTL}9G~b3Nm^
zA>J(pn@7hWF**k7tugcp#Gu2q82B5+0xM$Ci<{z+#5YUp#v^VC`)_!DZx@@0X8$EZ
z`6&@D?~*b7Rx*74RioNOcJ<BCBH3MwuX8krEK|dFtQzP4Wj7yr5|>9xG}^3$*%l=%
zsRwB<E8+25iNHTfBz-5Yd8Nepi%L8_ro_w>)QiZkOyI}bHlUxlCApk7?89lN!uHWB
z?2T0+Aw`87#8>Hs>^SCacOII2>Qw0}OrcMw<6>eSI~9)fr`E-v!HHPvMjdKu)%f`@
z*yDFtiB-fltD}{8<f244wJE#3Dd;>U1twppOGPE4JH5q^B9gH3Ya$%|sZBjez>O&h
zIDID`X_MnI<Ut%tX2#(PdvAUc5Bk51!LB(m$bKA+9h1nBU5Y}cZ4`3WN1}79Nc?&h
zj_c&vrcVonY4;Ggkbeu<u?$01OVN<}qkdi>dM#ao2hrTi<ld^8<sw8p<+gVeH$7U>
zi(X0ZjqO~lXKw0l&@A?I&%^;2;+u8T@TJ>SZZb?p-mFRJL0tWWc|NaZ<FTZt59+hW
z?~}$G!ykCzmYo-x?DxcbdrzFY<$*#!<^W%~<AK5*cl^k=q`G1Rb*Y`#oiU=dGr|@-
zp-XLg@ZXHZyTMMlveF4-jhvCNf?oVa<T__^|7kUQ74EvAdTm#jb#TQU3s(ei<7E6g
zUORY?#rw$77{u&i#UXZV65sq8<bcHw>|vN}kIY8)sO&fjeU^^IhDmlX3a~@QRqkBw
zq7OLR29C>YuqT+_x&=1qwU~J*axhWLc^hT}TfTq2lMN2FAeXb=8Z9cU;ht!XFU(VZ
z$hXGl<JK^G!HtDBHqe^d;OR&kd^Tpc%?c|hwpw9oTkbhsq{oF@2hpz7ecCYF#5_}R
zCu&p;m{Z!%>^8B_|M;CpB?{ck<X(3eIi5X&(PIP$yR%cz)Dp%fmU!CT5@y`QOZa91
zH{y)Cr!264hXoFvwZOZ#7W5r+4|oPSp3RmR^uQ7(^#tYoe04ie;{s~+rq5=vf;=BU
zxf*fH6u3-mbDi%?o2r0^qXLh6v3rNvtAxFPYXH!sHNaXK^r$5m<7kOGb1X1^*D$m;
zABH~pL%C1RZSZgGD!w}i`&8x_OHXk)`!&kPnIP2L7_*B8;G@w1^t;?2?xp>4M%5o3
zBjV+3M545qnI!w>C&`Y$B=*N9NdWOpjeSWnhxjJ?NRk{pnj{ZSB*|MImpUcOfGU-|
zA4M<5O|4WolEWds`Se;RKEyYph-v<ZZ(8#BfB2@7IOYV8|8Y2l#5dW*H>zx%L>22~
z5j`?itJ5U?0lQL86E87ivzy1YP_2{{XvMxlOTIuOEdz+D{-nw$Vw;B_cuRb<iuk50
z@y$Zwn+XwW$%|FXqGGiuKT#iPtPw}zo3GEcQon~@?$TGgvm{%}Z)QsnJB1?xDrDp7
z3OV$%RLqW%W9eQbH3zdtxND*4a#x7uwn8~Gp-86ADU|7nD`Z966{0*;Aa8mWNF8oU
z&S8#e<jh=&n~)=3=dz^tp)}Ea)JsZ5wjB9VDpyL`gXl#ai(3T+ol3>Bs93u7FOoU6
z`Ssf7%0}ND*`J(4pLC9#>6<MB>t&1IjVu|pj(Z9-x#`^@L(UPStS!>X`KD^wPQI!l
zC{^0rQ^`?hmF#Gt5~EE@Iq@w;G{@EdAJ3ILQZEaMGoO<m^I4{ot^;($b6Oc<pp~&(
zHPStYx8!c_5c}OZmnu#ZRnlk>pU2;7nM!Z4?=g)GrbeYP(aJM&IVZ^7bo0>2po3~@
zd?Qu*m=L3^N*1e=O4+beDJM&mav(q{E?eWNGsa4je({n}H$nQGOO)NZL^;wbNw)VR
zX5Nq_seO`Vt2$Ytz9)<Sq!gKcD@AfvDCIbD(eX5uboEb_$gWy>-Y{KOpGX(=UG78t
zPLuIFGi6#*mOMDc-OtwC*(ffRwbjbx<is-h#C+NZ?zlgEzFO9CyL6g+rR-g{MtT`m
zNw8&=d<d*!k71QGxLPH;y6a@a@O6U!c<49Cdc2o@&L8BR_fH8YzDf4`Ev55*OJ?RT
z>3`;zI9I-vCr0l?oAOSMUw$X2>b{r2&F`h=ybq!a`6zbkkMbm|0a_O}z}X!QaIaBA
zoNCz`Zzs1#uW)wkkh^(Je6uvOFFZ>7;q1YF=rg__>^}9we1G=(Z5;^3&4H-?dmw7o
zGeysKrZ{`Y9M`!Car>hU25<{z=u{h2QG0jf>slRN!}X<RUw^1A9J|;e<DCu0&~sSh
zz6~?kd`<jmgZ8a#(f+0#KKvO4vqEy_T^zA+#AwW|7>zS+$H4f}7|8E2C^<S7Yy4bL
z=e!&Ko9zLAo*Sl8pX%f3i6yt`F|O&&?iz2T{~ZVYW*;mXFaceTPDJL!NeFyC2{|@X
z(V%oHvOMOaP2ha6C=>70^AYGyoO9O?<7WBe_QZvlIBOw{sxL-|4ok3?Aq{0%Al9V>
zqGe7X;w!i#zBLdxbCzMxwq=;02}X;c5d66kg3r`DlK+OF?fg(&Xv6NfpJB*z4ad_}
z%<EK-fbpaVSPhQEIi70|Cr9CrcQocxM{%&HCKVoocy3TNJ|4sF6?Pf_B)gg)i<P@$
zaf(^-)j#8~PM?52%nlV(pW1OS5j#I7GHaFu_s}FHo=?Vh3pK_y&_Z)d1CxOoWYf=k
z{Js(;+~Qt$K*^k&5^F1zc+QR)gEmUECWfg`-X@<%DY=>?;+a>4<WMw9eBPnNHS#Ni
ze<%@JONEo<UHG>Pd*YP-U5Q`3Rd^q#!Y!2w`>0i&CZ_qVSK(PI`60gl9yP3hY2<kn
z^d^(L$s;}rC9e6)ZE)j5)T8z)u_=fAQM?i(XOTA=ti<atDR{gx1v4B|(C$Gp+WRG=
z;hQA*&Pu}UTioArPQ=Wk3HV0uO~syg90I+?2jkF#-kW+SVsT=0ERLM!<`nT_`G3*4
zK0F#R+oG_cXB7O?B4PSD0$aC-!(kbB-iL-_)$d@WA6X6;ZfJg)!JZ?#AS4Y6q%Z7$
z-HrkOhotY0t9kMNw`K3WN-8U(l-BF1iS{s(BC=AFB$ZTWG|sG)5!tf$yvW{rbFa;{
zxhZ?|dwzc3Ki-dXBsrscKi|)ZtR~ahFE|bEx$&54F%=~Trr>HwEceVOV~;W#7n@8%
zv-}D8{A(P1rbOYd|HhzcZX~8TkA{uOD7>u*$EtVi1+F_1vG!qb;@&}rQ=!<!T(}8+
z#RusfYtAh9P3F2=oF0mx7DI8%nq7XbA$Ujc&A*F-F#l;Fo(>Mgi-!U5JQRS<ZvyZr
zAP`et2O?~C5c>b3XDvJ!`AdUw??Nz|y$pu(buhlYp<m~2AnJGpV85Rq>i6|Si?zNO
zbJGXA(|xe9uMdL0dm}H~8==;Naq+|;bQ#ZGm9AdszMs36!Jeq%!3_j<c3!dbWIvE6
z{&gq+!u$KJJTan$Cn`I8!iIdzPF};5=6Zl;L2ll8V4JT8R2d#vH`@cnM?En3fd{N=
zd*V??PjrC?_B3FpO%Z)H>)A8Yls&=c2f!+uxv0Jau!4DQehVS;Ah<IRbmFbif|#P}
z0DM_a56paWJ2`^Leu9y<g5|^66WqZCMW!wo-^v9&YP)cw%bEKU%rGT8qURzf)ST;#
z(mT#L)WQXI;>h7_azWw^7wo4WxF5G=?q>+Da3^?lC_Ov;+K(>?G(U)a_&BzC4#1xw
z%$j>Mv*pO{n|AEoDF@aQ4^@ew^w=Cor{^c|n+sO;cR@>YXK0d~xJB-Sy=jg}IO2fL
zH~PbA3pbBl?QkRC7RTQ7MJq!eSSI&In?t?eH>el9epqAqMQbeGYmJbM94T<rN(u2z
z_fc9IJ6$W<e67?VcN26}D^-YX9EfeM@_0mSQ_jP(QY)Lxbb=hC{5)*rxnzzw5#!_&
z-w2QTI|`*KkFGpIc!0P@!=p2?O)9aCE3r-Qd4=-Qt57<Trx`Jg-kB}bw)l5#FA)Ds
zp#~LCeI%7Ro74F+{aU^_laDFo`Mr$z<}2~dDB_#XFY{#nC8N0RHA)MOQEJg|dt*Sp
zJR45FU|GJ5Je)6m;__tt)_hrzO@2d)#C*+Kxl(f-eKhODve#PC|0$8<RhG+o?jdaL
z$4oqXgpbr+E)VAXU+(Kev*l7sPm;NU{esPy3EQ+(dL3UPV}2}_Ly?PR6Bf!9YCOZI
z(^H#XD50$iC7Qk^apm*>MJ^(3uH0cRX?|&uoTi5+np&J6xrsX87RZ*|V!3=}zVx`x
zUOnz$#vkRbfh>?w+&-9EF<%a73+1n|`EqRqvB!}-(XP#tyRGtMzse||e2ijJ-zeXg
z5+5AXOSN`-DH@e8Lp|rn(v!r3#Et7G7RrGxh0>p1B<D3o@i(aCbT5_USgN?`q>|+^
zDhUh9l9w0M;>!2ZjQ-_y2MZ*-${cC>ra%IDl+r`<pI3qW*-fv_biGU+uaP>Ja-~yQ
zj#S%Xkg^2^X@5^G@t4$cBU&x>;<M#dO1AXqnIp%&b0x8Tt{l0QE3!%}(^~4}*G!#c
z-`2^gL3*imRWBK{3{t&@QARY#la%CqS?E$IH{FZmDEkCc*qfZm{^a}AtLl#}7WMpM
z@vm7b%~hrJIhV>J)77%F{%UDIe1n`@vq8STSTCi0HpqyL+^zE3CZAHb$?UXkvMO<#
z?E1Au`de<5=VP|Y!Y5nhZ{>Dr;rvb#+}=yiiXXDD>MybI{3TNhe#$h93h6$hLhfFw
zkYoKSWk}t3(y8e?soVOUd~f?sgo|Y+!`{ou!as7gW;HyUSB>3jHK2&Dfx}B`;@Px@
z$h_ME?@#r>s03?7<yd3YB5NFvw8m;*Yv83dR>s?)*Df1Wziq=@f(`skY?0vAkGqHL
z)!AT&A^YqQdC`u&G<NK!<yPDs@;B7tn^U*9Aa`Tv<b{>YhI-%SKHeR2+`Rw$8?|}=
zfrzymh%n|$^O!Fk(a{&N>_Sy<qz}2LKbHRMkE=iZQB}b$8)jmMRt<sm{t(=HJq&ZL
z*&RP;I3C^^j<@BZ@D3h<CiO;QN0o4xe+<Xq)KS<QF&Z;3kA_cLB<!eB^?o)MQ7xlz
zL&lSH9*?i9#>1=E1S~U4#A9_LLJf(iIzJJ0s7>9a_oDW#=~&@06KAt#qRNzHRIE?N
zhl~{bzLSdVX|u>*&EihQEHodThLd~K&~x%^gfvZOzZg4K0@;VNE(4QmWFmNMCQ4f=
z@WQA-)ogBUG9#X}PX+(lYV-?G<MdW?In=0XTvMa6QjIeUve0T@7N(8J#_UEp-07sp
zxS9sHGc+hWqCx%78vGlnMKXJEj+ST<@k5KGb$agK=rMVy0hOx_cqs-{j5XjS@yZkO
zB`=rJ=R&@u&qF<2>v02t`J{_|bg1K?!*6<d2kqD6W(Kj1!hj)b4fKN;kV{;$jJ(T>
z`rO!T!Hp<0;+J7Yyr=&3F@t^@tr1Z=BNE8j)KgN+BDPUaGU7*w5m)-M7m&Kv%!bTe
z5i1qHG_d=g9e&iM_U0K76K8-`paFL}8DQ~PkEV0=XyvGf=M^0@TRKc+Z}EXZVz{Fk
z_}efS&yK^v9dmJ)ork&(IoP}+8~ZG?F@0keDtmF)Vw)N#daL2SNrlxGDnyqk@ppYC
zbpQqK9?gV}HUlw!>?&$B8#8Xqg7xB5cB7=AtrI({Th4^VAMSFzP2~5}bd2AVfGe}6
zp=tMcY&sXmUYIG^Lhp@D{$$iQkA`mBM4aq90ihelp^-%t;<t{$(V$3dsXH3=?niL*
zHXKD4M`FOsFnF{KL&M=Cuy{=<%xj0DVe)Vc{=`i+ZfSmKKMaRA4@D`xHyhrDpm%vN
z0y+dEGa?8p*`KhOI&Q_r036Q<z`A+dZF?Pni`xTnC@Kgw=)oD<KbU?r;-Q1V@V*?3
zGIsuj9SOp?oq<qx4?tCOKXiNKixPt`PG0fh=C%(y+OV_Xw>M5Gyb<ww5N_uW#9a$7
z#9XJ&toMW^IhmEN#4_Yz$D45%ie8!tO+C5c=7GJ}JaAx@2W;=V<DYbPUq!p4ejRsM
z&0|MjJUfRYxrOQIfp%j&m{<2eIS<qB9_ZH10|nH5W>s)+`^f;P7r7#AB)6bW4ZwZ+
zZWcGBHna`wZ3cF7&&g^esL25hi8qeU5=1l?{CMYrf@{nu@qSo67c||@{Yn!TYI!c`
zNN?{>dTn&ionbiY%zV5fmL7M&IdUl`XPwc6_~uv&w}98XU|S-+$Mo&Iq~G`(`5f$~
z=a`=4$4XvGo)W0O3MOU%i>;}P+43>-uqStu(r5s#{0ltV$X>s3z_mtz?`})@HMhdX
z5-Z&6)|0soC)ho4MBj0an7`hE`wRWiUT+WUs&<%H-xk5W`ttj<H@2zS!*R(PrDLqI
zt{tx--}mItWlt#5bHsu8X6Q(*G$+0>CB8XJd{cT*EAxnNS`*)#KTWK|<Ng_~%p$&N
zPkd9gvrbNX8)f8Rqf9?)ly9f<#JL?gffQzhnRV*5YK}bQv4-6zaqH*Ett)e+-*s{a
zkLSoA@+p6JER;v?g|aNFP=*uVd`P7ySzjop&r-vh%nTGUPSk~b`N;EX^IiFJ@KU~<
zAig<4d=vhQnB{Ap_|wPQ`#JH>eV+HJ*DUZg%8TtrNg70aGdEwt#^uQd^E|nKV}Z2)
zvOs>21Ni$-v8<W1P^w&BBo`e@<N!Ajl(Jk>eU?i*cG5g-vs_kEKiXNaOkQ!{^Ak6-
zPU@D)hmp&~n_a@gMlY3VJC;b(f)e)jF|So#AS1|olo1<zrp_}bsz6+si9W*4o|xD4
zI}=|d(GRm+TO_sG%#{|E)Zyr-*<iLn+=GjyIJ-b%wFMFvnlGBDJo$9WD5lhEW}G!j
zKDD3YF?q7}oKZF}{~wky%3s8D`E?9pzb!{9Y;$D6mTb8&StB>A=tXr!D`V#9<eIle
z1~1MLr;XVXmZB21sN_ttN@fmF$s;qB9B8MKSl(|vQ6=^AR3h_La^jO(^sZ_d{az&>
z6I5c6qmro`Rbu%`C9TwIdE%1GZ&HmE575Z{b-5C?Do396%$A2Wv!wf6wd|~_mhnT?
zaxyMkY7pO8n&*gnQVy|8u5_K8E8Dwjr1ZB|oJQ-|ajlbxetKzaXB6)dMzOwVq&Fx}
zhCj)d7R1TpM$VNv1#>0$^jvws{i>o>^QE+Rv3N(b$C3R>A>&FVZbvD1JxV3uda1l#
zS1Rk;(dWHnwaj+jCXeE_iQAfOGWqT{X_&c9b|!BV?_t|yit|<(Ien|dZrCO%<YcPN
zdneh?e@Ne>KV=>L#di!pWybHfa=Ax^T+FBtOI@X`tNl(=8^4o%E#FB`;+xmRH>ySN
zWKfg$a%W~0y#7=ZZD*KZWReMnC7a;x5fcnKR-YYS4KeO%3xwb5f$@&kh={jFy9jH1
zw6n&KJJz^3%LZ9%ZP4~_TePP~ZrwXuyd=I^r?$h1QafzjYllh1H~GXjpS~0Gz#iB1
z9%#Y5#XMr1Yqf}F$lIh|<uxJw$k(V*?dPp?OZH}X4}@kIJKQI+N35kUJlp!gaf=@=
zSo$M%t3R662*9(I0rdUy_r6I8HXjPXll&ogo;wuPb`Qtus8DX1hth960?Su0k6SYw
zPRZezkUR>j!$;%dmC?vekA&~0F<5mr3dLUIux;yjgf^N0b>am2D<-n&D;j%x#KLP#
zELO~og~iaR82V!x4tGhyy9YDzw0$y);*$||DH*pV6_*vMs8>4`8k1SLd}tO{P^Y@h
zj<`b;XXDZ3+1Rrz9qoKFFl=oG7T04J&b&-?X{LY!bISQ9O7uR+PL?N1yrS25#XuFR
za<Ab~RW-S9HI}?kV+VU&2JGQEdw4d2G&!g{F&AZ@bMYyO{DmqP+w*gg=BGu%6fGR*
zYf<fu7RB76L<c>5XETFTpL|RY1Fyvm=(J0Z#f$Yg#a@}!&vmd^L|v<=4j!F#Sle9(
ze>)v!_~~HE-G}D<oR$g1L$eL|N<Qb<58kiJ3>A5t&-Cz2?7@zl(ex81u$M2Dd<}8V
zIv)BoBgzw*lZvJ%IgI<9^xbUfW<>2q#5`XNxc!LvZ+eM`l^Ad#l|47n)T6=-=;~x(
zhdueEOXP1<dL(q??GYX7`RFj|losC|h~3s`P|{3;RARgZKXTxinuFWqw7M#@xgVX4
zHpVQ>Xp)6vMa-ZzQzMC+nr3ym?WItn;++D0BNeD!kcs4x88A{i40x7?eH&(>>GV{b
za!G+hgJf)fmc(=I4D{WY2r*2@smKJZGo6O5m!}f*bBDg}6fC3n#<$;O6rP-fL%|ba
zcWXRG4j+f(cgA9O+!*X{5ee^`qws4}1codLhu`XvSb8jsUC$%1-ImxpD-<jK9geX9
z>@qkt43;6oF#7#av>7rKMF&DqmzlxddBN-kVpl>N?#;&0dw)9sL#G6wsZ9V}`f;Os
zW&m<K1)^uQ0Mx1%fCc<mgC2qC<q?Q{Z7?E#GiQ|)ghi(UaVOm$tFwI(PHZ!Cwl6m|
zeUX?;?+x+Imk9RvbsLPOj|Z}E+zUPUb6I+b8K(?tPyRgMiNIdmx@_l(DK$KC`K1R;
z4tk)U(gUY#JW!Rr<r3!(%jWLvpW<%i6gQNFxM5ei8;*Fm!>NG>EdO;!$3N~Uxy?Rd
z`e&}}pcX~%?%3ZzFB8y5B(UG-!Y(Mmsj-3~3NSlKPs>3UZo@evgc}FPf*mp8j{~}G
zaX@W_11@DdqU}q1TQ)i&X*aP9^{8sZILm)JW85=m)O+lV+<DG$f9rs^g$_`eaW^ly
zKYlIh4+|eh9R2A?zp)cy?>nP+!vFSxopF6N_b^YoAfi2ca+t|-YsoA*Gv>`6Q7daR
z082Zv@3{Q{<kcR4_qVA@mC}DhSQI|N65ZBY;=sR_$Skm8$7ENymG{7si;fsR%n=tB
zJ3zOoKm1bdVf#-%cxKtac~l>I)tT2wvBsAxJ&`)SCmz@n7d7pP=T&=R?d%*m@1T`6
zgS65iLMwY>v~rpFCU>4zJ`mr`B)<7fY}4huRu=I5`Kz@~=6BFZs2FA70QxNk8s(h7
zQJP#fit&z74)4j6)~)G>@y-`7`Z>OCre~F&jvd4S77gdf$6j;fR|)l!YV7Xkxqscb
zLa`#gIY4|fihRYn<AqXZ0`;hHdRkBA%i%5ga(7d{)GEoR9zqSOd%isB#5{5vW_Fk*
zdRC1d))#qF{{lVGeT-7fOi`QzeIYxHvbJTOyuZxd>|YCHO>D80o-LL`o(m;)21&3+
zB@$;{BHevT<Qg%~r@_nRiS=^PkgMtWV418^E|VwR_^ck!-K;g-kf^y#X0TUzN&q)O
z_jCK{EIE#L%yL~~rpU8EdI_KBzygU3E08ee&1y1>)Gd~Mgg@zJVIR-6GxWV2E0RAK
z=E@arSZ;qaU+zf0yf7<}1Zr&4Yvjo>dWPTsV~~11jj}()DE~Y%$lX-pH4mdi%`wV&
zS9)4nX+%3DSMF@jk)6ah^%iH#rbAgWd}x+fcFmECo3dnBRF>@6s*>PGN_nzT$y}RK
zmISIKwKFkJLmqWivZ1j`rV;1NB9_^gp^{NLmGnyE55_bV{k$p}so?z;Dv7$KlB(Z%
zf1Fy5T*#3x-E(E9F-N+;&JweaDk+_(lG@oS>HAV8w>{O8d`vC13bN!OagOxPk=D<0
zq{i@Ed4DcfvXol6<*t*EBAv9Zrk5qeH^=!r&;2n<P2$z&i}@atzsVmqR~~A(&v}A=
ztpDap8htn|e=LyWAim*yHrcgQ4$UbQdrhgtuV<GuJ>E6hA)UBrzcen}FBebmmuvO*
z%jAeXQghOFS+s##RKu(CxWhd$zP%@V?!1&k&ECq8(?4V#y~UTO|K$F|PdRY^tyDgE
zD_?9X<g<OH_!%l?OD*y?P2NeDR_|m?r*~rI_D+J^zL)R$weg~^3D)*C!Cy8e=-l4~
zqr6NoFWrRM+&Xx0tv<ZJ)W<fHmT-B{5^i5x;=MyFJUnZLU)RmxkZgljdK)w*zKOnX
zi|5a6k+-QI5+>TAA#;z}o9z&Dl$$D#?a-x_J=`7Kaa!#Li>)45aMc5EzR`Es#S?o@
zP|H6@&7NBP?pj_r*K8n`4jzcFLkFVV&j;>4zIff-51T6eVEw@#ZDRs3;sH0S=JJ~L
zdk~7~F$m^1Row$axEDMWZoh`%?#NJFJ{pRR%%^l3Is)IU!(n|P0;!8e;fzlt`n-t3
zAh&T?w{09c2aLz6vhg^kn25Zi6OmJU5@xuvcXs<EY-=2grJ=F#SsIH5+v5@YClR9t
zC*kVDBn<GH3H$pqxoeS(!$m1*+A{^guTpUIPbxe$voNS>8qzkTq1LF`2t73$%HipV
z?UI3p%QDc5zSc#nGT9TNK#G@wJ30z9k5qEoNr|+pO0+OhVRAneMpRS7ochX}`)X{V
zMin$F3y*@cF{2;{>2-2(+%^{<$K~QWJ+>>1+$f-CwJThUv^iS1RS@eCx6~x}(Wa9g
zDYx`!+lu_i6+IRV;*Nup4sGez{X$-(=}clAZgLNAufsnbbcpS%!v^As)%<tABK7pZ
z8nAY{0gJN@D4=I&#!myr@c8`EfY$@K6GePeGmd#H;+bB_#4kxk1WYqRN#15!ICEOQ
z%#u6M`)q1NCApvCcLprKW56itQdz|Yq|Y?KJ<@>o%uKcE%DwP!dh9r;M;tdbceK~T
zW0wvMU3Bo?t%ZX*H7a_G?LX$?#@Jje|2v1irW};s&PM8(Y&56$re=H=!au4pGnpMi
z>=PQ8L~n705~UNAc=@*i&-t1!h|EM`&kXja%*LPnX;`7)=Ek5@9BrI}Q^c>gb|gV%
zn1N~0#5b<fakhQ}KAerm?c}K_Y7@tP@L0SW6oYHrSZF+X64F0TfHrMB3agI8YU5bE
zvKoUv%!d1xjDkZ-1T5mX{gg5iF)VBOcAi_@4M*VG=uj*!8;)S};TSh>7(&{RyRjUK
zJ?)2J)y@zUjG}L*E43z75SCpJL?7ao3-bf;qBHl}Zu=wlls~3?^+%eAKWbd`M^ViH
zl=Tfj2)l<hx&W9I1fWB!K&*2Nf)#x_#|M)8-0p+^A--r}!k?|yzIe6A2hHyd#_P&K
zh@zf+@}VatlS@t5>xs_HaG$5g=B2ME++CQF>Oozpr6)3fdLV;3(<J^ZSBvw&^$s4G
zv&)@bR_=K8-3@UI+_1{a4UNs*U^~bSy=J<>>4zJ_$8(dLeq7(jK+0|)v2XzPwqVBl
zBw)cv=>hU3QF~pmD2MpML@<1v3tpWSq?fp0Z-O%tL!8mv&k?!b9N^3j;Lfq^5uNFX
z#G8)zSm=b7JDd?neA2HPcM<3%zBQ*m;(FKv9qlofw{^H*`RY}FB*$}4#@HV_T^%vx
zog+HEb7J0{oDMxR8&5jJyUZCsTDzcDp$kliaRTQG+|7BZKp*ioVwxO2htW;h0r(Es
zd(Hw){<c7yR~CqBYl)NuOKiAliR{@{DEnrGEnT}}!5J&e*<poo$E?ufp%wbvv%;{m
zR@kNOgEp?#==IDR*@JuHKl0_3={>OBzX#HL_kf*Q51dWQkyUnD`EIY37X!6oJ6S7F
za<$@4eAD}wR&>NP86^g}@WLPu>+*WLCG$F6jZ$h)e=Cd<;zeJmpHZIkJcYYP`M4=h
zS`mBX`Z6nHD3B+_H=BuXGKg==nWgDYUu((YIkK|~bL39UNfF=Z;|uBep>IcBC_j%C
z%G`tW?sS?XnZykb*5ymt!hBIF^Q9-Xrxo4u<rncyIRB5=#5Yw<^6Bv*w%KbCyKY7?
zWaLSQy9U|D$NJxr1=95&b_WvQte;#g8_pDqsoO$nK7En&BEA`AMQy5UiA)<&BAGlK
zw$k^zd6|T*UM6kgm&qn>d}fVbDj(-8k!bEcZLwX%o;h+0>k6cmvOtEpvGZkkfz0Sh
zEY_<)27_J|cEk7*<GejuAi4Cp1d_9|n8j?nksWsBM%hs{PwMo_6DJM5FD>#V)RxaL
zHBT~@7{$GVQ3B{?xw3`lew|(_+Zeb*We};Yr$1XGg?{vazs_ZMOs?G9og>*cIdbTG
zmMqE0lJSxyqZ?((f~9KtnW~o6MJjpzmr~lzQ_6p<l+y6MQkwo&iu(_xM1NGu#d<2y
zwN%O0Au4GWO*}J0B`age<4jOVA$gs5Q&h5y+|HOoD%txVv68z=^vhLp<%mM2y;exS
zr%F-2RLXRIor~jC;&M(UoBONf@IkeB=(8kvLbmkmlp{H3bL7ClT;`W@W&AIVY&)Tq
zjx%)ZLej}<TcflcWRx&^ilWQuE%wisD*y8fO)iuNPYXr;yh!G9zkGeS`SNho0(m!Q
zfo!`}EO(}Imm{TACUSrMlU1o)?NTZ;=k1f`3--&4L&P^eWfGORU);m@i%IH!@f~|W
zRxdav@$*kh>w9M;`qeoJsyHtjT3wI>)5;~`e7PJ7c`IYKzm<=Fzm>(`-pbyz3TcsF
zDU%mf%17dx#^i3w9o~t$`FrVCSQ}I8nIO843EmRloFKk2CBFH~)dU}jZ*<XhpuS!o
z|GcV?=G7XYcufPmF*d@ID=mo&TEXY1DVqK<#i1M<99v|AoyE4$-LQr0!hY}`WQUg-
zc3A$!iJe!@=uS=i-Z;U$ngh@+)eV*FJ>YlF1D)P_U;@2~R{QC}iS$H^{`49q@Osh3
z3;DeVA|k;X$_yVoCB9ifufVEXzPMAJI}ULH81*gyDVc$owm%r1=`ncnCIo(qhN6sK
z<M#uHp?|wjnC)V3@x&2m!@ZrlRl|rQM&a<}(b$y}1(zpL80In#Lr#oGRow)nwx0;U
z^og)PHW9%MCSi~7B&6m|g8h|A$ZZ=7k5{qS!mhuEZ{rd7aT;7cCZP3~>1g{OJ8?Xc
zm_3?_rfpL2=1dCee@Vr=lvzkJNkf%VcB%x;hRw0rXcd{x?Vbz_UrEluB@;E)XW~FT
z1-^JF*gLDh=W`1BT9mLlp@gcMidj<?p4U*rX#sby|7NFg7Bf`uvtZ{!|Lmt6I5yA4
zVR~yX5#Mx=&cy@bn?=Mo>)o{|Tf{D7_Whlkr^C?=I?Sn|$CYz>bg0H0l(P;yY;;g|
z)nQu;9VUNgH{eYz{9bFZ{Hqqhjdi?^*TG}}aZf8fCI#y8J5P_E|LIY;Ik_L|R0sAN
z@SYf@7IRNGL4R<B5uYX-ad0Ykr{ay6F@?V23A|5i^D2Ow<HSH#J*i8z<lj{>a&ORp
zNjD9sLY?f<Tmzm@GazpmH7REUR<|}_#A`kG4fWU=t%r3pJy!5J-0!19_!=$dG}OXp
zng-9V=Yk9=>h56IVc#4)*^`ZR1F}(ZhCEkr7Pj*>+Z?W@XII6pJr(-hRI+1Hi5-U&
z7;mjWa-U4(G2gcT>};GUNW;+Kv(UDED(*Z<Mz^gq@i2p(ej&`qStcUxa{|sDn1((x
z={M)j!tmo$&^IC$`zt2nR%SGO_&sA<JOORZ#?x~gg>y0%o)021xNtOgKS$vn*^dPd
z;n*;EBu-8VL%)?H5dS<Bj{QTCUC5rM&%;n}>@XZ>_g??cL-3Nle8uZR*sT+cr>B{j
z>K}x|hCn=f8h|YS00dq2hcek8TfP0!L+OuQP5hB~$R7v)_~TaZ0K6I#fG}kM`O*Mv
z$q&FBes1q>{%GFJAL`jY=ziCm?}0b!>>P{_`aw7|f|<ruo=~uR?|)oPin}LVhkC+{
zewvJEc90X__&D)n-IxWZE;aSG2jZj2-<WzJl3LR>YEyp_-C-Q)4*w@^EQ8_xc3(H-
zKVTo9k~;{t+_KzBTo6akN)>v9uK;J4vNL!R;D3s|$8mtQCFr0Ka!d5jD2Q>KUC_bI
z1wZJWS@*~Zo5cwijya-){LPgw4sc+POlLm_n3+4Ept=K=b7$rTuc-{%?Vu-b{I<G1
z;<=YMqN_a~G_^;IKX%YOx5J|{J5>LR{$5{ubo|Q>X?AuvPye8OQ(N{!+ai9913r~F
zqiZ7<R4#BqJhiXs`OIduVXyH%;OY~=RO^I>d%Gd5wgp;svA}RI3sl^-fMX3yINMsH
zc#|c1jJ1M}Sf<fWD=2nbp=6yEirAxC>n|(p-D`ynN3787t`%Z>^n}Co9!M(hj;g!5
z<5^aBn2qfYH}CFPCf#w%H%r?7(8xert-Rwg(Mv0@Cuzl%_{N3!rr}j;Rm3-$FAZW&
z%uuHl`&P)4j3&N0?P8Ss<Q6Vbr>Z2rsY3q7?Uzw1h&_C|@x1?-8qpHHBrem--}j9&
zhR4L)yk=*%=6@gaw9?B<YGYA-wbHJaRzCI6idWklS(=+If5zm9_C7Ju#(e3_u9d`z
z`NBl74DOmQKZ$Rcage&r*uk<ePu6%EByx^HCeicy@V#CN9~z`u9ix1RD`M_ufvkGH
zKmwbvPiA7V^bRVP30)V+!)0^jx$9h6b+Sm-vj5JRejxYX?A3XnE!NkwWq->Y@%dOF
zN@n4I6&1*v4djyeMtS5>Gh31`^;hT1o8IK4zU50KeayEz@cDKp7efxmu8jQAcV?+i
z6v&CR0y*zSjg9=%x`#%2{fWMrj?6$MGGnb{);g1qe?XoLF&JfaBVxyq^y{YTCB<Ja
z=U(gON|{b<Dl~F-wnjWXG_vJot{irvrzK7yZ{ii=kf4@u*DU$fE=vNIsHIY?ma=AQ
z*+uTgCrQbk8Kv0gE5&~aIh-3x34NuM7yP@G#6JEHh)<kU;!KQVYpar*eO2N;QY9Iq
zRLoYZqz!RTm%aS6y-Gf;P)OZ$g*0EHkmwf*@$9UW{(7afxucXl&MGOPmX&RxmL|K@
z(y%0p8|~Th@5^kNO#RbXCs$sc%$1=xHIhxAan=}WXybJ9p5I509E~!Ey4z`LjKkdd
zn)0=~+VB5#&c>0<!`Cd7XUalp^lzcGY&utB`Tni0w@^AyDwVHNDhFGYiW`r$#rvds
z!G1X(Rz`13ne^2jl&JE9vSZs(nS1Ai3<*6c^){cC3iUb3T7O<Dew~+xZOi4t)N%=a
zTQ0tx-pY;5Z@D%9Rz5W)e>1B>9?q$hjSDNKVY7Fl@AXdF(p&tB_~u1XZDbSQR3X0k
zhxn#~_-4JG3AQLq(BzQ`zI-x)HlYp<y{M1QRT|)2lLj!jG{UR%Ez!-s6-qV{-zZG+
zYKbY@&Ss`O&lG1D+Tz?%8)6VUG(PTx_)X5Zdea#*m{BMg%G@RM1rrn9aI=Oxc5HP=
z<5r$f#du<Nq9=a*<AHDHp7_N~M(gHYIO;VJ@BU(zgWktxYG0gK?Ta+xn<?bq-;lpa
zVRpoKVjwd1@%KOd30;4Lpd@1mOv;DiEw6*)ehx=hOLnIzMxd7CNVGo2Y<SElOtXpP
z{YYG8|IL&p<I!N+c%;x1s_#T?A#)<iPEN!R=3rjdkH*<8lkwI)7DHdgqE_h?Tr^Ha
z4^2FtyqJc16$#k?X*xQx?{7-~By7JilY18_sCzI4FaBje+|*fE(=v_wMrkPRG8?VS
zX2U%s9mCu+F#dQ3awQWl)@5R4T?L9g71*azAmWSy<@AQw9a3V;GbObm6}%pE50csD
znowhKrW#)4CVqA&Phnu!OBJ5m$zL2KzFEY5g4@)migGkqYRTOK>Qk49ZxV=a+7TO=
zRMo?Ve8|&E=8<f5i0Q?TbIYMOIh$dh=+imE&p)QcV!aj*s_O8_T!%~iyf%T<x8`ya
zfqvzl|1g(DpYLGa&%LF?zx2n<Kc>g5p45+sVcJuN8W&@P331M&aonCFwwV+}-;F!H
zIX29do6>Jg&(6-T23&k*K=f7S!FL<rG{=C1$;@l{8gRas0cq5x67K6Ux>S!1<c`KP
z(8H}%hnK`-znKlU{-MFKp&G>R$b~(7fX3@`(5D{rQ$^W0(lHx9*R$8pmYeHkY9zR-
z5qyl^VpkO|l_~MdMhVX~3M^-KYGJiZq+L&k=`#AiC(;w%V;0hxfAiRrjI9|n(aR?Z
z4_nQ^z-QBuvLyjN@ze0VOFVvFio@!uQ{Y-F7QKrnW29v?YVDW^E0+nFcxoK{heo0I
zn=yD;5Q(Nfqp_pSDEz1rffIj5BCq*K+;9oQGJ1<!90+CqQYbD>8jdz+h9RuSFjUhI
z#h(>J(D6kGmV6IJ+gCwos0>2mia>aK1!8JR0ICzibWQa~K?8qe?BzyxCqGOf4|Ctk
zkN$6e%(M=`-G~6JpB;eR8v1R><NWU95B*0!gqQfhHqZwXgT2`~I2iv78-#<yy|92f
z@nM4}Oh<X52C<DPeZ(7>Z)#89jXyDtKA3rKM^6OPOPv1I12@UrbRXz}8n4{pmh6rn
z-P|#=gFDuhxxsFw8+uQ2!=QF<C_UqfG1Isi#on6Q%pRGwW-stX;6o9k_I^NE2vBj4
znWw7s2h;2OKt)}Nhj}VJG(BC=om`Bn$QiyXoN%qb6DA*VB;IyJ<a-BHndiVBb_a~8
z>40eBmkp=vux<@?@J+-y%s15gVTai^_PAqVkLuhJF1*eSVQS{J3%D`N4Zikbhv)11
zac9#OmTzp}N{o}Z*cNLx_k(+Z13YTdZyZm2^~MD?`u4`U8okNs5|hlbz{M|Ju&|;F
zF8tjEWi2hBbF_f(26n!0vVgO%1txe}phqP6oU2yweP)HExmN5=x5ClYR+zlY3ZwQ|
z;Re6f*Be%-e!n~O8(nd%pc~pR>xOOGZt$Dl4O_=_!^Y5Vc;k~LwTN#{+GwRe@l8wO
zn~I6_Ns!0!h|`Mo2A+>Dk>642q|HwQGdu=4&GV8F-yC<Q=f&M9d+5O_A7YfUP3%75
zb-05`f$Y6S&1JD(x=ql_$f0`i_tFy+=_RR~UcAipQqfs2ajxWtsEsvkNpEvQt%TLm
zO5HCSsee%;i<H??9-bpf%L~Mtc)}toUrNc<<WPSK`lOd8<T$47qd)brUcUD=$l+cF
z@gdhS;E`VXJTOR4XZkznNwL4Ilab`5TGiKbkW?dP*>SSaH%^-Q$4TUnIQcduPWmp1
zlYJF&GR}RftgbyvS{%w2{k?3-^~n+WrV*2_%*cNwHx-g6eZjm|7<1|DwtJqz+!mkb
zjaT{d;AOtFwJ4Ade)-bvGyTuRWsipBi|5`v=}f)u)BgYSnS6b2lyH+g>E@azC8P5s
zcVwP~G|rRqX!>f52I=o%kPg;*Iq+5|(ab3|RcNHee0CE}*GNZe4LhS1>_JpW+C+tn
z;1Lw2W@nFDLKmrJW`<gRk5<dpA1b-EODP>XDCJs+Qf6f+#bk(5ZUrjkQGKP9e^W5u
zrVvAQrHpy2l$lLc(udkrdmf_?k>h!ylvd<<LQ9k~;#-C!F3FT8_6ix4!2f%$Lbg6t
zNE;u8JhN3wOsP^TsC5n2tE62cwWP08%d;42WI@^Da3@<bV{*jmM~)1=mMhbY8nK<J
zl|vh~vS^r2f~V-DN_X}Y8mT2-qrbUNo_s%@M@}VQD#sT{`!IGVQ{StXNPX{op`7T;
z-s_kmIkKo&p6n@>;a|9^VY5nlC9jg-m$}K_YQLPGR3=Al4{{InptSNnA~uaqNKuWG
z+}%1U`@_zOW6n8=Uv{3I0&nDcWAZke%jMX&a%t1yE#H&3QZoImbe&r%X*^avt(5k}
zHc=hlQCEE@$GqN2w@PNpf4`S^`)gxlBNOZ<zA+Hr_+2)^&VNj}|73!|k0vlDzG?Kb
zJ{tV4kD|s6m=kS;<O?lfbZ&*?GflB6-W18>OtCA}6n&CRVZFi>U#M68;6_{nc0S~<
zbEXdEjOQzzvAR0*n!N6E81D$T4%E8e^Z3UZ2hXuv>pH6rcd}lGnTluAJaKF-f8Smo
zgto)Hk)Gv)hUva&M*hY+)1RD301i(IfcL8a42%dwd0`;hu;=5FC%fy_Ly+t>6kpjN
za**EQRhc8`<r#_4S`nCil-iWbXnIc~p?VUDW7EdK?b8@UT8&5FwiD2z#YA|XnuwAf
zlhAPGBn0n`MrKhAt~kbG#<N%)+c*U!DN|7=As!ZUr$KnWbFP>U<%vWbk4l2sg(Qq>
zl7b<xQ!r~!DvI08LQm>bd$ZG+>6*<B+1Xe&FdgQe8Mu0y{u}2^JSDz4Wukx!@y&Ck
z0#h$5P}5$CyfsQhR#m~pT=oCChp68w?6y<mQ3^Y8xHYh+b2c)_Q+&FYgYms`p_@%?
zb2=9%0yGF^mg-U?E&37TTpr85-vS+a(YvxBNRQ&>dUTwuM@e^Pk$UJb{jnCa8t8C@
znWXs}wJ-*0VU?mq><8|2*3t3YLG6khPUdtSDz^~htk<E49eY_jbog(l4hL><$2&!j
z=JX|xxU0kGqr^nSG<}I}{^spe`iW<GF(cmJ2q!Be-n5`sr#k&P%uKn`d-MI00e^NF
zkeJ8Z6t$`KZp=`bGbdiffR)$uXjj4`Opng>^%$^%kFyIiVFl!FUNHynpurB}xnJcu
zXg($f|2)aY$@pyC{*?uPeHQMt%)-@WYMirBBXu3U#a8s+tWx59TP5b@DPaC96JKxA
z=e;hS+SP1CNE&xTW?|`x6j*7JvDlwoMlF({yq}1trPE;+O>N428fxB}ipA-1NN+a<
zQQKn><~A9j7bl@{=tT5-G#<5L$6>&au{b<$49W&XqVU%!bhs9Ql0)G*aBL*X?uFq%
zoiI!rI0DP^Lou4(n>WtG;kIxX+J7F3v4e+li*X3LagU@<VKAy(4MN1=AlT7YT=pvf
z5rF~dvdJGWm?P}J%nuP3e(31n%bsuMpUiyF$H^PM+*7z-Fqr*4-1Xiw7*C(_&+oiZ
zZ@(WdC(&!u)dwjy-bi7`jhF2p<a&F-lYLchl%5C+^2AwUn{)3y@L;b8VnRGn%ftgM
zuehV=p*zaTJh)@w0mC8>Bpq_+&$~N%#JVG~g*(Q+cf*<uZtzyPA-1_2>YQ~&e5xyE
zSh?c)odK{_un+hb5YG(MDiv^&xaRs4p3>Uz_Foq$BVEvodj{#vT%fMY-kMBja*oub
z*!}Cx4ALfBC-gr^eB<E=Y2bjm<^ADCY~%N=KURLVLr3NrZZ5Dx&?;h_BX(G`!Vd0J
z?XY?VwQ~NsbhRD5(suaucRy%P+T!Ck8{9Z<L$8w!dQ7&#QnA5_Ok1?m_e1y_{yQ&w
z__^ET(52q^G@utdBP`JDV;7_x?}87tyTJLGIigQ>LGsNmF!Z*-`K1;(t+&7hC+^!=
zS>W6xa|GP%jBaB(WA%?N*qv{Mzl~PdOsu0P);U9t=lkR?c(vIQA68hg|Eeph=6A*F
zysijQbw$$5uIz#8icj8I^7Xq$1`*%H@VMehzGi|}8WG<NA-=hvNN>&o=EGm|HbW=Z
zvxyhzVYOVNlZpa^%r7S1A;!rWW|Z~{@;U$I$#Z6%9(>nJuc>-D6r~phvoo%l)VoG#
z#fRfr4T80@-<Lgt&`P+eR-`Hap7^FQvBWOon>TNH9=xEDpsCriHY`U*PAHHcL-WPA
z2K5|AgM^&YOWzH8srr{*cHQB9Q-h3QS6t-(@;uDsG%aC<*+nmte(PlR1!j?rI(bZP
z=v6hXJpWfC|MicP#r@(W&L&O_-f>blK29o&;$)!bR4K2QDpi7J$@4|o67f7+PQTK~
z#-3Uk7jBe3FUeiK%aiK^^W{rmzDy*a^Dliy4%&R#7hsU`g9dps-6)-nd2(%ko>ZKt
zA9`7y<V6`}<9za4%z4efWfX_^%nSW6ica!m$#<hzB^za1RioIb4Dz>=L2ljDNvcjK
zK85r^1!%;5xJIh^Yh>{lr5yXIknal>QZGy)X2TTnfEv=DX=-j;sio5xwcHFMe^OT|
z_YP!A_p1sybyp!BpObH?trXXCh2(x#$bd41j3gg(_>e+cd{oGJD+;&X)SE^rWf8GZ
z#A&6R9jBCf#tf<QJVQ>*&Xk886=FG5Ar;;kVtqJ6?pP?~3^CH%sY)40JamLyPx>zv
zc^b96sFEe0x@L2$B3n9{<Ve=r9LYG7D-F{$;vKJ*>!n(`=BX3ESe+dItrr(^fTva(
zrL}3EEI&#N8Ivz-sq^hS!=7Yj)ce^o>;AtS9y=Os*~$Ee{mk)8i)Bzjv8;QvQYQMY
zlxBlg$&r@(Wko`nygE=O0h)v2IPi!BG&mue+9xGs>`D0<aZa`v&WToeUc55iNK*B3
z=|FrF_Pty@iEoMryp`OnN~tlwQf%*5%KYb*(uTPyFXEfk)TTP6yqAq>?<J!Ay-aid
zAUism;N)K>?4>b5L*kp>#5Z3am_Yf#gnj;XP@DMXO3en?p415Htu3+Zdq=1jcEXbj
zov=RA6cS*{>lITN15J^=(iAg?*&w)MKiE*4y5EmI6H(5XUf_)O=N%ET$DY^H_Lw!n
z9@{qBqeJulxRTKyg+rWhWi6QLp!b+P?hn@wMCL_iI7YFTgI<B+_I?<;(GQwL`VHtW
zxJCYE$;$vd92SV5>^ghS>)e8Q?3};BJwl!@-X99ZNyia9e~rMKi6fC%9*%?RQTUxY
z8hwXEqUd=f#wCwIn;&D)cg#3=?;MZB1{2YEFSncrPl99LXk6PDjg;&dyt9mj_x)Jt
zcTd5`ia0EDh=(d^8oCrGVEOCmm~to)G2uzjA5X$t?-XookP5}tRMc%W3tMZa;p~Jo
z9C(w4@%q{HwWMRSNd|77We|h-W<K>P58|7}feJJ!RlxX$yPHD)%@rjc5#O|$PF>}?
z3fA<G+;2zzW{eue=b0Jz%EGI~S-4A`>bVOy#OX7BP=j5*J{sZ^X2h?s^Ntv&*=Q{`
za!Y&zc^A{6#0Y-$(R|ioWlQQj54AAd*W%D!E!JPvqI7{4E7%FFW&Wz$Q!OliYKhB;
zWytfiny5p>eC~Fx(II7%j@#8bxDwxt-lN0TGdg-@i91f|pgyd_4f<q`1k-mjl)cE_
z+^lk9{|$AijqTV2SkH)B-|4e?VSw#vb|PDA;oeq@fqblW?r5=^*rs_q1BQPgZ*!Kp
zts*_{^KsXxOCR(t4NSXfP`X%y8GjSobk;y(={q`-g8-WxR4L0wv2Qk7+|R=ADOrg4
zuEt-~^G?)L!#`gI%SI~nVpsIA>Pjq6R$%OvOcbz(Xktb>Lj7hlubzgdf2U$mG2g?m
zWLUJHiQNxpVA!fe_)VA&d-DX$z7vn;+No&WD-QS0#$xrT7<j#j#_<`G;P7(-o~!BO
zYB&xq%g17bjKTTWqp@$(D169{z#e+rbGSP-Xhj&-ULS!+%|{@Vy?)o|h4*VT9MfZl
zL3eB@3=O%fHed+1;zO`IH5fsAg3z{k5EjG)!e>ta+E)!g;V^$3Ay!H1?T4Ssec^TA
z2X;+;;5Wq^Z=MZC>f*r|p&yLR^9Cbk$6)k*Fc?+3d!yu(H(bJfu$?)n+wHv(*?BNb
z>kmSzqZc00182)?3-tiv9P+5kOFZC845M=Pz@M5PSbf<Ych9+#TXe_59qxFr${qI$
z+_56t9o_u+@o#Q0e@ajBd>-Uz%YM3|=`L419nY<WX0EV2Hvk7M2cYsLpjZc#3p?}<
z2=3SAp28h=2-BRq>6J6qHlUuJ?F=hoo2bstSUiq-?ygSgGt3DMv<`SP(*gTCIbhWf
z?!X+jM?$teI`!x8lqYuZA$~cU$!tS0`QtP@)R|<5;wj8;CX%yZ=HXxBoGuUhA-_pK
zG!<K%D7V3!zj!_7WCOR4ebM(%Uo3jl2mRTrm!<59<3T-;Kz#AFsuczlSir>If_|GW
zn18Sf9)2~)qRZyUSZNObqg}A1y9F_{1-y7aJFW{7BF%Bn-yA!NI>V%GXS8(fjKuwB
z%%+>+Vu=}o{;)gserF_i?t&+~E%17cCA)yEkeN#zYn~N)7FwaN-U_>OtnhhomQ)ho
zgz@|z!k&_mp~M~Iv{DvLzJvJY0`X04;+vzdwQ@H>C;19qbF;(5eJOLkn}~7V>1FpB
zgKP~Twn;IH=i@xlnNh=e&JG%OxIG}g8Bcrz;+r<aH$R5(YX@m%KJm>#;+y%@r?Q%8
z#e|RH+gIWm9tFfV8(wI{w@tRZr;nr6xE#3>nkPm4y6(M<GWRa|jpgKk$l*lY*UMjT
z_41fqaSt{cr1YF#))U`+Y^0SWVx~^SPnFCh$930Ai}qSs^<E=7t2kLwKTcA5#z_OO
zIJp}eCu0kUcf6;{jngSomOM*ljmwr^wKZZvzsmtry{x3q=q$ChN3WUBZcn|5eRa*5
z%{qFAoI{bGojQ71euaI!BaJf4Do_5SrzV14q}RU<QsG37BFQL`#6DN&QNyE7mH3Jo
z?<Yn{In48Vh*26<7$hdaAe*`xq@`9ThplxoAv0SZuzT^+AdM8AQp#oGo0az!5?-K?
z-ku7X>7F6Gqg1jYL@g=w|EAtk$%0UY_{L_)Oy^A5W2TV!T!n<4Qi#{z{QT<*QEeqI
zQ7L3XjzUJQS4giX3emSwN{+Kq29V#$r0(U|DMQ|N&yZO$88VevrskMT`QkZSqT0@8
z|6RJw%gSIsb*60DsF1ZTO6j&;DVzLMvbwrjYSUjl<(pdGOvsXn1=;enGMl(CM=l)6
z6?Ke8u92H_FV>2}Q6~+i>SXjAy_j_|N;>`1PX8FC@H{hl$;5{P=wsT&ZSw!P8_}Gb
zo_U3`gxh6JS{KTG=I=JoVLx_iu`E;;%fpk)C9mr$Icl*=%FXsmd0Ck}swfkq>mj){
z`H(CMIWA)oPfC*WDRC)0CpX`ollGS{NUX<yG9$cP8htF6b#30t2J$y^RF!ggL8ZLh
zT`6l`R7!&%m6DnIUZyYkAl+VlkknQmWls1<S!(`4*5v&V>p{O|Tm2g7tu(>vd=vC2
zHo?1f^>8n^K7ThiK$XzO$WLvLvDqE5@o`5Kd33_`qn(g^wiCv=n4-F?DVjN$!ta+U
z`tI$GMJ@WF;+zxi^l`@6k<PI3aYm0bj>uVKkD~7OJTKT|;c9zSH0h7`S^e?Z(+NNM
zJ8$9+cAT;kwR_1x^!nkAbCEu%zRd^9ZhmOK#t)y~^7^y4KORgCfSUZxm*7CmSR9P?
zMMH42<#1w*;V_E`g@5Z27@j@?TC<VtARLKyC&Q7&TzGx!(I^Xy#PrvZD9;=N?*?PZ
zlaGVBemq{3O+X#u8~3D1XlWOX^M#YKAtnYd8x!ALibeJPQ?Thy96Y<nqy6-0=x$74
z&LR<4j?;e{!Y-VL$ygShg7H05@oRM|p46Fzlh0>i{m3+&zLW;rX!gf7N=M+8blfS*
zz}dl>`16BXn}Zejz^#UR8x+VSzG<gX;u`(MA3Rj-p;ciQbIofXsPL$v8ZX1t+^b|i
zj*8f3c@{MsdT|<Xhu~~BJukUvX01WjI1Rp(YS43o1{dPA7}P`u=WaU44)Pw?w0OZI
z;f5A*_vp3ZF@d;Z(@8Bi=(R92;qJj8=F6{ZF_Iikjruxl@2JBjW~r_Yr<Z4y4u{C$
zJR_HLj~vc~{W^3arWwuMOLKBJb%<@25c@c-*5UngJtp)vVqhP3<1i2YUn6!R(|7Yv
zg@Jp7%xHNr+tq`(ir8jubuIIBTFm^9kDc1ovv+zlqVBbno*$n8Jqr2pW9}MQSZdg5
zmx}^&RnBYK{S!oe%9sO_Dmge=kPSbpY^*$xg`R<YZ62s0F>1uTSK-!76~=O-{ad0E
zi~l3G8LPm@(o94sGN2rmj!f!@H~yQ2j4i4Beo4Xe?#Y<>ItgJLW*~WTB2p};qwwK0
zY%7e%cbSUtM^kVtH5M)E#o+tAXv}Rt2_sfa;3mX)eA*U;Iv!(j?QtY3=ZwbWQ0|7?
zMZmsiI6AtFgeUhw4lf!(&wVH~eM6ap<6hdo!_d=($DE<)b!P}_HX4GH{X$^D9-AF=
zgK+vWvr#<*Q7b+Gzqa|~uOEI$9PWn$hkfz5t1psLeSmY`Sn1#mk8^`jFOS$pLG8&n
z81ps`#-{6oaizL9Lh5-V`=U33A9~|vLt>rxgMg<4;lz%cHk--agfhq2-4iDo(O;ut
zHy?9RF%}*e_}U$<h;KfhbjQWL?ugmqj%TIx+ZeeQMbAw?cXveo?S`1`ZWyR?L#U%0
zW^CmyW;A!Ay1U|86;}l9=AL)c0WkRoSXu}??8{klCuY3MU2x=#3&IvU!-v_ZX3d?@
z?T#bL>9grY&+w8rj+pV;5&v0omr~<^ghU7Ale-ChZO^?ad-RU8M_4C&bfx#^zQGP<
z;dW5;w?ngFJJ^l0gKvx-o+a4f@cDlD`$Inr=+O`VMA>4V89Ub>+2Gn58yq&b!QJ$}
zIMuc<E;R1LPH}6L<@7}B_g%4@_@+Pc&3&B(n%h~R-jpt=?9v4$5$0%LW{zVjbL8vI
zQ75P~ez)$7ExfI+?2Lj(ow@Bt?nd1i@xRTmYmXVyQ_L_g#tfhH&7fj0{Nr_<F=dB2
z{GN1y-98J{9cT%2ttGZCv&7<+mWbwU$zn^C5BfiRGlJ*;PUI;XITMF4Cmuk2)13IG
z5B(B_Wm<{<kH;5w;1J(z(C8#(KKTV=oGB}bch>5}dX-LYjb)x{Ha7!^aVkyoxwFmg
z7FWIOt*4g(#4IcR(~$Sk$b;3)7+)raxx|kV-_$3*Ia!C?3b9Rl9$!9d<lQrk$YG7x
z)X$ceqq8L^g#B@y^Te7xEmha?Yo@U0ze}#v8J{Zy`s7Me=6qH?RLjL5xzcC|eJbf1
z3Hd`Vs5N<;-o!Z6m^zthWj66mbMrXqRWpvbC60RwakA+06dB|aC-b)?%dvX1Bs48c
z5}9dsa88hKtCA)8qDrFg<jN*t2jXg-EbTzA54}f$^b0o_tCOPpI@x!gdfQQU3A-EF
z^Gv;oS?^n^d_8U&#E~7n4H~kWhuT^rKK`c5jIx(lsqs~Mm$vftvN6hpdj{#op1uyR
zb&?L9^z+OX=kD1uW^lHwh@j8+j8Y~|P|5>hl!D<3xmKAjtE*+muSpql=8H-ibX3de
zu?o2mlP)<8h)>oKgWSlJGgB1u&q0N3x~Py@`xSCDL?NxHHNBajkgtUbY0_6A_DKp+
zP<I+Z++=B=AzN;z%QAiqd(RAc)G|XN_NPfgWSTtqktQ*fv&G9cL!RekO4>|%avCe8
zNhbLo`g%rJQ_EHQZ+t$|XS_X2q7~UP=SjAdJ<5?nzJ8sexc?EY5s&7~;l0tyic>mS
z$>)5uK0QhC23dW~Ag3;o+fvdmPJenZ^}Rk93S?}pIWmbo0)O$=WbYgq$X%?zn7x}^
zK3~!%6iZpd5-D7<RH`;#B}(^Ia*G=o2ajx*E!B3(ANCY4kKQ9?$Bs)ir&Ds|^C>y;
z`kcf#pO+ycE=p3`McI1!qV#-tk@>c_^3RvIQqlgsEWY?&VjF*uTZ2A`{qzrF`tJwn
z@#}-UYxz;SyL^<Z`j2wt<VVR3`6HS3+}|v%3hzx-QPHXz+$UE@_lphDuX7VTb#4#m
z*p3*W=?E?HO`$_4>>Sq#;rlyb#EVWyYG{g`b<8ll*c`i)dLsb+pwD;2!A{PoKiC;l
z6Ui-;v*~}=4q0vJ>kHuZ)&gc4TJ*=HEe`nkg?qbhZg6<*f!>Y1H~{R8TW7t|b+8Zq
z-pO5@-oB{$$CtZ-et2x<k3f11TBZl$>%?GuWw&d$KErTw^Kj%g3dQD|p%`>`1TJh2
zLvst_oBH9Hw<?@nbP+h-h&^k~BJt6B3~tvMiwO59q|F<LtxLwE#?c9=yM7{WEt-T)
zJZFh^GQtC5Q1Cg1J6f^Wv||chT!}-Ueeq~#J{@cSnT~{|i3mD711*Oq;pu~9yp2jh
zjc%zpKQ9%l-H31gnT06#G(7%)99?x>RO#PU?2fe&7{XRs5%7CpK&2E^LXj}(&Vd<X
z$e97fUK15_?T)q9?(WWAdu?6Y_x#>Jp3lWum0G#yo;;#~Co|$%*V!S*t)oVv1yK1E
zVl4AhB=cdKqlJO^rhQj#imxj|!(T;+7+8!B#5Y4$6r=P?F-)jYEl(_gMPw;<7nLGm
zQ7OK>EQLecGHjGG^n69{$ui#Wo%Og@qlaM$HJn5PjCIT!EhaCsfga)m224C^z+c;`
zU6HqWaKnIP@;2XB8nE;aJwmz1J;Tlb#UXAVJT}0+Rt0u9ufP*(SwX`q@P!)DhYEUf
z=J86MYVHp5I6Oo5RG=;S94lg*&x<S2w~9DNMQ&$2u@HUDi;X5@blu5V`*9LN{VFhu
zxa$Tz(CY36_!HM`|E0&t7kbQmtjC&jdU&kR!)sL~f(yyx@V|Y2$k$!<sA-``^)qU5
zr*v4!&-Kaz9ZE;Dv-nsUcCoYAs}8T^wZ7VwvNN!R`>yN^x={?{F~x|uUxf0ABGg>v
zwyB>M7u1EkpBLa*pM1Rhp+Uq^X8Ox=;pv`(N8hsP|IC7A7I(#LGBEyW8h)!zMbwZK
z{Qe;cBWEU}Z?6O#_%jYg%2@0>OmF(&D8!zafL+7KBmU7?d@mk@pN^5p{Sl6cdtvzR
z)@WqD3&kO`P*}5ba^9MeD5^CQwxdH3zbzOing?^MAPARN1!CTV5y)yb0-O5}N8^-X
zaNZbz;Me~6*v+4wj-lAHkUpE&ez+UphpyXwvAiw0n>-&R|MbS`W!}il_J(nqHy#yx
z<G~VMk9s5eu{Z8L^yap#HxAD7LC1PNn8Li&tpS74x9K31R&md}AA2}k4aCiDp3p~o
z;x9)}tow~R)B_I$O!GkTAP=<J=7Fv2Jn)40W(c*ZsE+*gMkVrgE3t1bvytR(9Dd_j
z;*MJ*+_A2mJI>wae$xVOU>b4n`vmBd;})hVGu7tYpgcAJbMCpKc_sVz{9Li|r!!P%
zoN;igGj{B8M)GNA{BwYsb|+^vr=Qqf(VrVAPFQn?e&K3*Yew?gupfTx=nEI(n6?hg
zZl0&7=9MG+cKYHovz?dR9I;`qBLWsX@;So+8wNNao?WDC&)DNafIWWuW{Vd=b_l;|
zi#K;{@NQ5aY^?0f|5v(WK~i_TE9-*A)?H9^!U|{iSz<t`B~k(`(YC84W{<K&gBlAQ
z7-GTg5({`;G{;#}3(PuXPJY)MEAE)%j|b+^Pd3MIW6d#>+kOjF=2(Bq9PcMuAkUe9
zAMw(+>sDAlq6_S1bitT)U6_sF``gk5Pd9bJ-w(A?Xkn1>9tN31j5E@Q{lqzX2}sgQ
z#!9`6-maIu2lb-(mmN5ts9_aV$O__{J~Jz1*L-%-R#!;OHG{-nG>GQ3K{j60%Sk??
ze}1oFpMH*be^<*bXSEFRRm=XH8d?5XBQACG<aSuT)Y+FWE;kEg!`lM!q4qJE9M0qy
zda)*V(~J1#{!P7H-%Y;deW@HuE|tbFm_MQh_F-A2L=#uFdR#8oYG+HWx++mV&6ZiO
z;^fY9cC!TOC6@SRZna)4iErGj$Psm8Hw?L)5aOHn#93`0Bubx_Nut#xO2z3!dGI1t
zimkIGv?f~`_$SHAf0E>+bCPU0pCEs?%@fUo0&)9PD0`TF{#{chHebu7Hb2HqBMdU;
zP=z@EO|Q}RNir>Bvc%P{q9<gsl)6rm@~iZj45gQLY^8KmPm=u0lVri#Ns{#kd7OEZ
z<xD1fpifMaL1!vuN=c=Bzf~b8Qp?2eL8;6+TPpn<QEy_VY3zw2X*rP?$EQ#ds1dc>
znlC@L<x9VY1!7Z_D}RQmrI}xzkjIw^EAr(t`4^KU;-Aq4GP+NJ{JdHqOZyi}=+#1L
zW2}`;gS8?(3Z>d8U)t5lmuf}6G##EVt>@)SLw;X8|IHU8%L3UY8u_AAOC_%-&S)gG
zTb}fOmnY}z6v#uTLNTTn#}HB^ahXMubf`#XFE5r)TS_D=tW*~4FO{a3*l9eoRO+=a
zlN~1IQZGU$!R!bOSi<iIKiA9+)TjDSlF)IJq=a3_zWuAjxvWZRe5Y_@l^a+6xnKWf
zs{FKP2jofS)>g6eksV52S=?Y{ukl&t{$2i?Cwlcl*&ex294!}$|Fdd#9#@mQtCr);
z^GCkhB1^At5#5@Da_+%V`TF{plogzkvuUSgxcPapP+X9mRu?69`~&%%_OD#O`9U_f
ze=oP)-piw`_i||adpS>h6HR<`iulI0<b!-X^g#-ne3UOkKFXv%pQS_4XDLtmOy2IZ
zENN?uVLD^fJ8q22Hk~mw-V`x9Q=GnN$_-{SZhD)c?rJj}eq@H<nkjIlu{jEeZ(g|A
zz-X^E2JUf$Yp5M_%=S1?&k-TajDMy7&xk(17p}}Rl=j0M_E7X8M{T!?p2IFmyg1CQ
znr?$(C}4-`F+PhL`=TH5P47{DsP)kgjjCu!>^>Bo-Tbkz7JFK=xVgC`h`LHJo)F)-
zbAvhk<S0}Igd*T+DAs3<#?Crn*r{i)TD@?*yBdyhQ4!2{jYV#&aj<Y5kFA)1cl#%z
zxGD;sYocLP%^s7<v8bz#<A88H9IwXX=#m6<u1LkOFR4h=r{Qh$bf}M|v!gWwzx!o!
zpFN8kR9RTT-W1iD9IUF$MYz2h<JPEQ)=$G+hX#iy=iy3hK2|)=N0n~@o_#66T=w7W
zepiSMd0LD=rbUklMJU+J`|z70JW&>-I*lC8;$kd0!_L3g#VGhvjN4;NF;B}qao)#k
zzh+jveHnN9=&7w_=U-ELY`5uP+E9;%)x-kq{i{!2X5vBvUeaIFeXjxg_Ym`}r}k83
zK&>FYUSU9brXDRT^*C{enQIIBZ|L9MyuyG^)UtZABXBi+$WStGHKqb%=vjVQPyyd5
z73i?A0?pRYlSAI-Iys$2<aet0x&yIIB7g7NAZD|i>B(tJzj8Tu75JKV5<d=hEStF*
z*zIaS^9BZt`pBNX2YU29tw)_{++4`hqja(!t@&{Ur0U@osz*dWJ+3y_WAZ;b>^rVQ
z@FE?S&DX*2Jo|1c%J{u3qc6S`i}On{xh*%k{wSgEo1E8yVs!JS_vT~~u6q?>;sI_l
z^wZ+)@In;W6u`AXJ}z9>V111m#*w+$^;-@WpXMgKE{l7P#5bSPQEN>a=7yz0)gXoY
z97$;C%e}I@@!Wlg!;^nw*!dETPtPafX#521d@~MRipFwJfF7?)5m>V@9HpgUI9NU!
ziOWK9`o<`9F&~A5%#oOWHUx)zh2XND+SJ1!^y(XgrFns<+&%(ckJ&l=+i-OC8U`EY
z1g~%LNB;*yaiGai1Y-z}my)}=%Ut&$UrgHQgSXwu&+PTaf5bLdLg=fBpe~i=jqJJJ
zSiXy|o%7~Cl{bv^-mtOrMw=^MsEYG~jm;qLn=t>)jMTw~^!C?yV&w!+?DJ-CpOq&}
z8+bx<!vps!Jn+uV1BP!(-s6?<UZBMFnM(9p$PD;0CEBp>CVaXQ*M`u8^S~V|<K3aJ
z?M@8mj!o0uaKXh5QGWpk<^oe0ud>=IIQ44)nywfC#c@~sQ{alp^bbFM>5Rp@$=@t;
z#>OSisI}P{s~0$<@q_-j!k-r>HBRW;-U-Q<`=O|+AC3*|hud%Z;`ze9SRF|J4fiT9
z@n_Ha!H$@D%mLp<JK#=R2h?LvSIByM*!Oe5q)ra7eq#^YRD0l)EutUVV#ZZlJXvCk
zemiW)Cs{+Q?TrzKyW^kn-4Uzpf(_K4*6g;TFV7N28cQruT4HJ|OI)?LM2iI$sPVFZ
zTOA7|9WjUB5_3Mgm_s?*9G~sX5xIewXN);^yOYClF~_E0bNEoBN@{C?ox3eikZOrK
zb*+#UZ-t6wR#>yo3cu*<*>}teH+bFiKr0uBZyb3Fy0F_Xo!(yR2%fHGa^ifJB&f4w
zwmwUq_sy1N;W;uRr$Y4k74nOo7B}LXISVSJ2l0&^@lAi?o9@InEy>F)JfW8ht@N^T
zlS<C5Qc2a#94SdwOV~$^6cgWECcX*cef0AUasl*s%==IvL)&Ym8~KI5uj*y<Bfb22
zST9Ncl!`|Nd*+C5X0)%A*_uiTDAq}rDn9$GN~PpYmP8%TmiYSd68JJsM&6E-q0{2T
zw16DYTzYtjZ#MFC(-Y$`n8xZ^y+lPM%DXy=Vr-l!QCW%7>sYFIAIy}$Rar9RX`(d!
zm?-;KC(#p=D1#;^NJ5)DX?UtYZg-}ymtU`s>*=f2mPn8HWzwLOn%X42-1=^i4dk>U
z$-fjWrfzq$LTry$$PxCv%xqX8#!>wJ?Am+9?%*xgilkF{nf%Fq8(f<teVC#CIY%o2
z2emTYwn&P}g^bGL=bcb0?|v3bCUe_ek89=3^8&e4Ss<}Jc*f?-<%oPamzyhFZso|8
z&|LZNmqu1y%#(x2mk)pCi%I7Kxnf-)O?u_YL#QQ9lPCJEd7}QDCmV@n>aNKb^VfM|
z`fr~6)g)hnUGl|{nJ@FI^JVYZd`aDyE9<j!rBgGtv{|W^8w!o2Ki0_i#d-30b)GZ}
z&X>*|wW4A};>+kF8P81CWgYY4>_g7>FO?I^N@Nf>8=Baa%KF!(vf)da)b6g6;|_ZH
z_&+}vxi1y9MAfQFarK@go9VCqms!0`U6ri*S|z#6oE<M@mk>FeGG^{)%$zFwv{Pm8
zd-f_C&ld9?v&D&<UAxy+%k|iWG9z@M7@IGY6Hlt;@2%Bx+oW2IyHv~a_-gT4xka{P
z3p=XU5p%B<U-Pw+^?r@4v{);3!`4csVT06Ou}fOrd?nTRB+o;?O3~zR^6TSwF)jTe
zBVB*W*RH?B$gUPH%rrv22Ytio&FFn?j-mw}P&cL{j$P=8%pIN3-q;u=Wyb96GKQw6
zGlI=caBYk!5^GH1Ne|A$Ff-I!V1|!(%<!V60>`&2fK3)S*`E8Gi>%>y!Wzr2*`T5=
zyT`Nb;Cj^#JNr7K>^(dGtoor%c0a5i>x#Dd-0I!!hH3mc*)3uaY&!a2d5#Zk4*8(j
zZ}dS{u%jc`4;Ih3lQWI|e*KwoH69LccHU%H1|pdF#@#U(H}?nQ?>6ibIXeoggG15!
zbto*-M<ek4Xq3|X5MUgRpxxmp_KJXA{Ycc^I2M7S<8ZIn1XTVr5qD-pA#!vyd}hYr
zS$Ql>Gvjc%XFOi)h=<qV6s%dBiaBO!csD5xcW0-g>(O+~sgr^9fJ|J?%z~WDLRIY?
z6fey|X<;tnn8mqXO?+dk;j^=b`dc3UA-=)i`8eWJfC1;ZLG^nf>^>BtmXaODU)l3F
zk-Lxdh?K0RPo!rtUMCe}!~9|#J64QVkBc$(eKBgiDu(~BVoaP+!uqRHoHHxKfe_}R
z^<{`!SBA^o%29hQHJm1T>};$@`#3#<s|{$q+Q2<i;+aJTj3qvByP(6i4>~M*PR_@I
zx=@TBx7M)(*-ekwf4N6KLyxd=di*ijfT0!@Xys0RW?Ti7nHA`q&n}v271*(WIA&!9
z)|1O=L7dZ-+|9vCYF23#><p^FTek`X^r=9d78S_;mmJa#15TD0a3<V<NN4I&oq07f
z;Qbptj?+^!j2zO|NqSsT>0vg3`%+`{Nan}CFi?*X_IiBd=Tmt{2dhIm6wT40_B<U9
z1eQTDf*e-sGPv$5Wv5Lk8qs@GcS;HD%}Q{6busGN7i0U@B7CxE_sv=@E?H=~%gk(k
zLwcbe<YDwS4GIg%w>jk^>ahx~tFy5#I15L)EmpQK9m?c1)HP1Uu|xD0M<*e(b|St_
zkB5nI99-weph>%ExK5eKzCYrd$>U(uc`Wq1BJnjT0^41};c68I-`=BfV|XZHxjA+7
z{z$xa9f?*8LNJD1P4~ltv1~^W5*i1=ctjxbCethb7qe5ZhQY;j7;bQX;a09cI?f-8
zOWTLgui}T&c7B+h;0yapKKRYY2iE7k@iWsK_m$rCx_Tpgus6O;^u}LPy-~i=8=-%B
zqimBmMylvJ?m~ag1uvw;d*R5p!3YTVMAblMrmlM8AUyHZ+!NESJz@Cl0n1<Pw7JY~
zzVXzg+S5z?j}pb(l~_Y81A`K+*ijtIuA6c6-Q>6`QP^F{y;65H2_tXwoO_s?+%QJt
zhEjTO!q>BgW(xcIE(+on2&NwzfU=?isNUiV_rosq`nzB^wJF1LXWS@t#*cDmJS*;x
zO6UHFr-yjL6eoy}6MnX4r_Cw$*=YOWl|w(6-{_0IrG2r`t}lF29GT~I;C6xo%9^p4
zYpp$Y{K4&*rQCR0XOHkY4%kay@q#j9o1eB&+_Odbaa+V~v&95EVx1{As5xYf1u4C;
zb8~l08EuIU?=29qksg|p7I?Yc5)+Lrq3vtQ4s{DmR9g^>TcB4v3;gxg95htm=miDt
zTvotop8_cs=6Ggfj-l4(*w)z`4LX=(S1)rcA(!K_*Bq~0EHL+w1r9k`;(CQ8eo^!C
zzhnvJEuPz!n89oE0<DCYu#?1$y*I=)8Ikly^eB@<Z?mN2V}|^4N|P&%)1=+aRJrsb
zMe5W^l?6tra*S7x-l_6tR;mn~pDMNA6v{|;)~>t2txobcGl*}l65ll2pppvWo0S)G
zB$@c8L#;gN`zucxY$=dgy9?yx;R4xAeB=C{KGt?xxivs5KfHLKoKYmZ?iGncPiBGG
z?Q(!wo8L1lrPsnT>9WxvtJKVMrKqIV;4Hb+AzohmijxNS<K*_fIO-E|GQ=rPdftzf
z^Rr{+`i~fvQOC%e39*vVFHYh%$BXmy1R2yWQ4}jvWvEA{{B}E2G&PBGdTyfJxsfO<
zHYLd82k~-cP=d_L%afOD3S=NXwm&awr4BzfU*?47d6bH0L79yEO(&;_ZxY=qq(N?l
ze3--QsS2sHxkBDrRY*+@x4Xu%lZG9_)|TuHo?Rq%^uye6(aQ2h%rf=Vim!)Og5tDN
z+PX+~?4=LpCp&{p$Q6Z^NYd>>nV>0@PyZCiEp36c-JLH!#3c8d=ZMV8k-Bwq<bGe`
z$LTpz-!E6%6l&xuvB&pY`Es!_Ju}27{T<b^mmZqw#5iGl@}y<cd|7ZQPrR-Z@0`z*
z)*teuwt2qX9FZ?h1^MzmI!7Wq<Ve@cD*4zpM>3}7$ObF5j6bKArKK9tj?jqtbd9_S
z(#V~|YOyDORM3+cDl1P`ugsT6sfBX2N-N3zi=<Y1ksP4!Il-q)V%QDXps7KwAEf7s
z_~scgU_$Lm3Ef^P3$CzBz;Ti!k>~15Y%`<QRH;9knKWkc3;Ru#`w9H@gsGDMd779v
zo-OfPXUmIy^JHq^d@-FdUvz8d%SOiqvUlVHIncgZPIasn_ioiPF|Jy&&o30WgvH{&
zZ?RmhRU_lM12&HN@{y)%r3>**VZ{ddwtAO*xkG)*>yy+E{wm=C-{ey!@lC5A>_qw@
zy={I<-tF2b@vnyqQ|qDUy{5=N+Z?ZlwnO>8j<7ZBggIL~A+Mt`GL9Lui?lP2synkc
z!vqtLn4r#VQ+$0zd=q4bdB2<CAo0z-b_zJ`P~iP+3w(av8xOLqadDG1oY>3Rr;Qy>
zq}yTe6+8Up#-B6O9N|w6XU!WI+_);R*vZ*Q?(p7ar^bba-uOu0V?r9WseL~9*}xaK
zN`3LZpC3{l_(5@T2!`wqz!e#e7cbd4KXU}EItRhSB^VcoZzhF@pwO0`_eRtJ-iG2*
z@@Tw&HX2QG!{Bckj&2*+c4ZfV{m&wBzT+4~`HsVqXXBWUW(UrqC`7S;Hq1W;hC#77
zn-qr)R`GnkjEDc06wD*Oxnq@vd$ZH<YhF5%h;K&i%78=XEVRnaLf>o55^u|9K3Ijg
zm>hJqQp0DSnw?-8EPblMA!c+2t<T5FMg@q8DZtPp1=!h8i@>2;4E{rlV;z|bC%$ns
z6v2IE5t^SWf`2gg1hmEY?Qk*XUu0JNQ86rU7Goy49DRM_8*K?5Utv%0$TBn`zR|BK
z!_m`aXz5mtlA-08@~|AQTy@CqPd&$7hamb@EU4`$59?4(?qlsP9TLcggtjFH8c5F$
zvCRYOTO-JowAYY33DBbhHLhHH?ml^wr-`E;mCoFhrUK{5)BK`NRn6S?F8YoqEUJJJ
zIU1#!p5#Pk&i&an+>e>6HvGK}DsbqL0jX>0BaSfO7Yy(;V-}s*W+Zi~!W(*+>|mcw
zjh?-Tdbp<Ru`z;sq279Y@2$twcEnz_^oV<*!}setd|FN3rb>rNR%O`Jw+s#FyV>_|
zDF!brrS4UVs(ajH$S%Rv1|{sBrboLozix}kiJ2C`ZmyP@L2iE4Da5yj`N-It$Bt(W
z-i=iw_P1R2HK|~il#O#%+!ec)f$H*fw6afw?X46z7bWABc@kFd;-;ogJer)1#Vnr~
z^!h6bbsQ$*=Bn{1?LH32wv0jAut-GIi6D*)L;HQB(fxEN)L*Gfd5%KW!jUL98i~+2
z_WGR*#znhe46F*m=?8)M+9MFk=n)t;cR22{_omCo0L)`I@l_>rQz1i9HFgN*r~08}
zsV`1_@WF;iA7s!!yf@7ohpoKX3GR(!zq~NhnmW{QZ$|sQF=jP!&2(=pF61V7kT*O{
zz2SP=i<+?)_I((PPiCH2QO6UT*LtE?6Hn%qJh=bnL4S(}I`^O^)xra+y3Bq*RbtQ?
zcK6X|vvRT$bxM`w4V6gbSyHZKKd2HDt(5TH$*oL3cQm@@hM&}>wnuVL>H*+S?xt03
z?kt=V(B6on?Cx9Dg}&klSJd4`Y}3+(-A~RqTj`9v@$3~gb;bnxZPG3JL&wb~zY-^`
zcPD>S-wCdJ`mqzTA8K>=r8Tor3&{^(F*qW7uLD%%YbsyTJIt<%#U<>uSz?chh4zSb
z;A_M+C-~Zc_qK3&W{dj7HoMo_B5!~#^uO8SGrh$T>#ebUOmF0@w8U#4ODunBf%~g0
za6Q)oi>fTpImZHP*IA%$mIa)nEiiJoIYu@ljv1>UXQaUX{R;RRE6{wZ0;gM>V?sA`
zjQOcR`%en?Gbj-MO#u&6bCgG#^ZCgf_c~dy_sat3_<R_5o$u$l1<t&&K(n_Nu;kTj
zzE<{iq8B%Y9E4IYRxfq3;)qJ@{>$Q~PlinDmL@kErOAM|sq*Yls(e|WA`O?O$g4#u
zk}@+z)H71VXi|!-s7R5XsVVY%aH^CHQ%V04mAw94CG{7pq<nsk9I{Z$?HY}YI+!Op
zIR!F__+}OHO&#K!RpdrGQ={tGEKh1Z)QHC=jr?;&BhSe*Sp3YBd6`;?K3^(RRAsVf
z3U?!j7fwAb6~}Ou1UqEO(w6bEl=!CqtT-`Cjgve(>Qq-_<?qZ`ax}4`yc{F0Z(`&q
zF-|z~&V%#uQl(6gC!JHImm*D84M~$`C5bX%SfaeoOO((3;>G;8c<ERkFU{*ENC$FH
z{u%j_--_G~IToiJcFN2zk?uXovBZ|idu|Yzex=6eQo$;6cDqbxKkpwE@`e1CTkQ&2
zOAV_rd5fP-*-yg`nj>lK*E__VbvO3cd@K}wGp+3I&Wuw4an3*V!@ShW+dafJ?@Qzx
z*eP?dP!`1%O8&h93CJjr2`|;mNvUKazrLN$sASRsmCX7_C3Sss<Zi26iT<UQ#D{rO
zy`8$#@?2T?cdnf1rj}O3IvtdG(w-RS<byo<xjj!z)Vz|5>2NDgCe+WD_`dm4wpt|?
zPTBINkxG^(s>J7&N>bv<5w*^hkS)1#F;y)|U&+PUXvDX%MtaXzON_OeeSf*K;*mx+
z{Fg7qLG=DSER<Ojv{HLmk@&ECFsebh)UsnQA-j#VGuVs6UdhJP_Tv7o5Oe0?b{*!%
z1o@j1BYKP-Crd8<OaJrZ7*n%)S<a2}z3e!4Cy(=v+hi`k%g=`M<a+J-(s9Im**=ZC
z4a7G)ug#Z(K?}sTWPyCVut3h3R?BPhH|tL<5=1SQ$o-2Yq)rVzPc<^F$1>@iy+(Y(
z*UHfs2c+mT`)(rsW3RwxF)@4*zYSl-GwZu#xA-B0mi`c3pP%A&vo^|x*2DgadT394
z^NaZA(uj8Gcc>$Jnsj0pg)wf8F~;II#&B%k8T*KDo;NkYuiEtBRGC6Q%M|0SOmVlw
z40mpuVReB5$vgR6GSU)O2YX}UU~AOTTO*;H4XRq&L6vNW+1KshZf%c^g^u*>(EoSc
z1?5eFv-5xl<gWJ`u?MHX8^?`(;F#!xM}LxU|KfvP`+U)-*bo04_d_we%}la}A<K6-
z^D5M)h;ROA8;Ca5K^Sm^_lcPy*y=nI``VAfstutS=P?>P&W*;f3TAWXgrQc0aI{<y
zjt>qIXm=(OIcLY<+~aX*d43`WFOI@r#5dWoF?eDd%N~zd9Bat@@A7!WWGCT@J{6N(
z(y;1z8ZPFgL)$w8zQi{PowD#;m4(IZGp@eJ=l-Gerk+zl*DnW=PjWD`G?yhGYBb(k
zi0(%VxkFKik1q<LtHZohD{3W{T4D+<bJSY;lC-cYWk1WDB8)v+1lP|+h_fq3Sx7OQ
z3y5>J7PHg67{TOlPSq>HVDl3AEu-JKaTx~b%Fupo84jH+!@H+ta3j7sf3zH@9Ozf+
zr9&Qcm>!_xHl+>?m?tm!ot-)KuvG5VA*Q~b*(E*p^w%TakvNB3NmLi&2|GRB`x>An
zzG*@3W;^js6uFtM#4v&S3i!>cz?y3I;E=azJCj>gr4=}yQ-MX%?A;ke&ZakU&36NK
z>@#q0hdHSn0}hTb;7wn8n}6u>lzyAe7ujR8UysC%dTy4I$Dt-=AFfBeLF903h<Q5d
zQO1w&+<SHvKh)tid8c<ZIviQg%oMq+t|r8Fb<1$`bScWH+s*1$3df_&gh!TO+*@`Q
z7Z&s0Q;f_>McC4WU)yqK!@qEs;cfxU_T}T++&m1;(BPGq8r|Qqx41+FN6E&W51DAr
z!Y2QKbPV~FidXYfu-+{h$L}WMSXKg>8pUI4aV*+?iN=!HD7ai@Zpv#sp6?%v0p4S{
z2O5F-Gs1CaTo`5!AB{^<+zYQBh1mDhrP!@K=XeOt^$WqjHNkk%AQ*F^gAlnp5Z&qq
z;=KC^e9st;3pK+q`E&qYu*-5zQ-Ate*j3z!Ui(^p*xl0?qjP=m;~#JKM0>N7(i^o;
zd12!*FWm3uh3zZZC+y~lwkxO=b#`ZelN;voIbyRrbs{G><c7Lo;So2uOrqxW#silg
zdmxrR8*6SXoFJ}=;_Ipi`iadw*puynl|SgMxuZnwV@i~+qXtDh6O^Z9Cxa5hwMyz@
z<a8XB*#4J0SUQBs4cyU?y40@#?ntrMGj#>f#K7&SAA*062<GqQR+OIm-G^ME4s=DM
z<1T3Z))}=|JF|0!UBu#yHXr)KIhOr>jr${}+zDrfI`Qkk^SK}QWBOs<gnsD#qc2*G
z>WhL0<c1G9Fk3}_=9@j1Ubjc#M(*RO?eW6f9@%T{@OG3Pd>YxI&P!VyqaHQ-x-G6w
zutlb;E&bUx`1ywoQd-%d_}|_*d7&qE)mWmv8~GY~g&$CZ>X>YSCm!?+kD{(*XMwFr
z7R)m6?X~7O|6YMn{tA4)XNJ@>X6X0B3>$kWFs`mS%r`5rketq`vkGiKN=}ET+f`zn
zuL{(5Fh^ykIW})J$GSV_?1nPOfw~rmZD4_>_36*y+gi&UxjI2F|Mu0(Ct{n>gF3O;
zuM(FpSrWc9L)_`HH6edf_9j)TkETlWniLsQlOiwXrpRq#n<JA`<Rh`oQl93z6xk7<
zB3DPGN<^HB{kQbREK*6=j2s!-QY}%{8tFiMGsYrc))C*VCce3Gs6hTcSs=>S1rlMD
zC(EyEB<GMu48%9*iEoUEZ^jVc1e`0CFY(NaS24#ml{xdrr83!1C8JEUWK5fQ(btOS
z-eH`4N{o}*)^RfMN~{b^i<KkoVkPQOjNCa8BaffNh@xq{1YC=k*dNIfW1Aw8b<<>E
zO@eskBuMeDcxe$7FQG5u#Q9E~+)~8LptJEZcYlJke48U7p*gaW_xavR@;9H0WOE#M
zD(jcY$X4ajXk@wcomtL)<O=y7S|Nx3cWk+rebyj;<j3kY)`^E{spJ=FW%by68JChL
zokkT(`-6pY`(B}}ZLSsjr^EwM`Lb|OjyyV<DaSTuO8;_dgl)2<@=B)s)gVjW<Y!6e
zuxw(WY`N4$B~99^BzkMMRL#kj*$q`9)3RkuQm*{0mn)wZ<w(blxzclUuKaGP7VSxm
zY^|Rs3mx-ha4`9r>3LGwJWt-1<ca;EJbC&uPZr&!mh~`Gwx(ur?>t-T4_3(u;-_Jr
zIrQ%2$hrx+a=(XKx~^5rx4UXNa$GGNzUN9nEwxOakt;pcYorsuw=?|;<-s9hH#e>P
zcTy{vm+8N$S1z#S-r1OP$)8>>*IyIk<QOF3WQF)V;@3@8DYFhwl3BGT%K<y?QqaRR
z_~c}{YR*oageqxP!>*vtQ)SAFY0`r`7mJf;Nx`YP+{WT&gKoauT0LJrUY;-Qhb@pl
zRSU%P_yQ?!S1kegizJfxruDoUId`T;8d@xop}&?$2a9ErI&iK0Id7*VjyNurpV;F!
z@;~YD=s%es@?8w!-(^w$cX|HgyX^7)A#37(%Ey;KC86IhS=XQzel@Ly*j_Erp-&6^
z>Dz*xTP+YYu?5cXJo{*jY~#*&7SkCWV@yze)C5KYOyFN)0-J9p?Bg;;@>2ya_?x42
zi#Zn6W%u~h-Y9HtjbGu`co<~^*Y!5&dcYQAZ!mk=)gB)*9MN(XkDm)Bo)HWy1vZ;-
zi>A3dhO|~9^_LQFJ9{Bxiw|mEa?g0LFHDO3nDOz$hKK$bpFRu?qlTm2kKwQ{8iDg~
zMqu>ZAnxV`V^vKEzLoRdac3k}@LAbvQz%M)P^a<@Lx;*RO!^UqE7QW!zHtODABe!~
z*fH2&H5U1X30RyH1@wtVNBbD;=@X0fJ7Qr|J06kLAu6>An6)PXtC$ln`H_Y}nsiw9
z$Uw(^85o?G$;<}vjY~G_+|R}vZxybeW<Om29P(hy8XIzPY;z$V?O|Sr_~sn(O~JcD
zG_0#dR7>tMm{Kco(K1iO&N=49fAcNEt?5M=y{QOMcZ)EfH#d`liqSEbdetAr_;|S(
z*IpK52l0)D`b*j35;WXPoa0f(-rO=QAilZ&qztQwZ)}*I-t(Xwzq&A6WvxR~e;p#m
z>tLV7Ou1f%J@n`1vm;MUPG=<bEcbWRwXF51`;i!;l^$!U2leYi|BfB=PQ*6`;+u!5
z6)=xyu4(~&IkT9(;tqtCKAe6t=o>C!S8{d*{-QP&B=p<VqgHi?Tc7kchn5;JeYgR0
zdmC`Pg#oqcxrx4~$Ip}0oOY3;S*6D`9rr;~n2`#hMh0S@?(`J5(Zkb74~I9zWB1r!
zcvgp-`P8TumLszR_of<^!SN}1n+>Hfiz>yFW~FGhz64vmN^tL1G5(D&#-k6!Hn~NZ
z@P!_yO!h=yE`;N*0vOH9$Fl4^ObFEAfRh?SujatXi`~V|v+2Xmgm(t{n-1wXv?~?0
zB2w6^mjt`&MCx1#IKPLze$KIYOiy}l>nNPenSlQuj>GnlvFLsy65c5hh#+U!_T^}_
zyc>#{pGV=C>nL1bG!h|gxf@ayg8OwtP?Qynn0rCQG(o`ZK&-nz0`?{&urgpchUN{!
z`(**>d(<Ccw}zth(hwZj?1z=)H2v4mpZ3`ssu*v)`ss!5ONePEc;VBP!ANg15bdXU
z;>2)z+E%(_Z+mwH5Nk9T=7wGNZb-gK|I2gWb`!omn^@$kJ6wrtde-+qwcZ00h;7DZ
zd7!YZ2W-D7armVYBOWQS<FXPa`<2)-jeJd^5-NI$gJYEN7)?IMldnB<M<Day{?6P0
zXCAz3i5q4PaKrnTz!h5Q7j|Oz?_PHK{U$iaUf^(hZdq<}#ri?6s1xggIp>^_HQ5;@
z!<=!**ctQg_D8#s{c&Tt6Ru`BVZ1mYlUz-Mo{pIF!2u13VQ%cR$Mj+cOr~GBJ&zT!
zP1qxQY&v2Oi^cZT%*flUu|u@89lo}<L)9Bwd^l%|-0ikl-h;Z7ku4I4ZN~EF)Wm5v
zsQbej<=1<|e`inZSZj&t^w&&2Yk^ml78o(c0>S+)(4h%=Y9s1Wi-|?-%`s)Q0+&1$
z80e_LjNN8fyOFulTW0K~RKVx60+%Zl_|D%G@P`6F=P3|b#hx$&yTfKE;KH8N;pY^H
zd#6C7X3SI>n<GPEjzgyAXwJ6_sb?Miq!Mpyy`)&`<rVSGK;oMccRss+&XQ3z8RAEL
zbGK!hw7iun3y)Ke`XfdB7N*EaayMIuZ%#}}k(b0aOL!iYr^xoW6iFhVqaLA>CB!%D
z7pmAtqLOtxdG9-`re>s;7ArOKY?VfKt<^~3294Nk*2u8ke4Ci&_F)Zs@ifvuguK9L
z@;#Y(l0iMC>&pVs<Z$ySvP`CH=@))fDhfJ3x;Dy^!KX50uTi|b{}3m^?K9+yQHIoh
zn=Z=&V&r(O7%5*6Ee)E*aN{~g8gGdavygZ>KPOS1)Fby)ohpYGCCER?2{MJcm10P|
zT)Gk`lV`_?>+(1`J11TemM2KSszkZZ+*OB%iK6Y6C{Fux<mLzJH<OvqUdj!oEyZHG
zuUOt4D3-J*+-<sCEZ5g@=U_mwOmQ!lNlC>rhP`}`O_{mwmoEtp%p{d*q&>MS!-7Il
zQQtdmrJ+wLTQ)z*62JACk~=L^yw+t()P_vS*_$cWA2QiXlO@BOWsAp@Y<X}pTXxRN
zmS<7fGPqB+oUM~BZ%wnM*U)S^{D(?x(y3p4&5`oVTyd?Zk#<vwZ_aAO>b*ur9_Rb2
zt&uu+G{h`<V!I(v&R)wCyK{M>n3*ZV`)5h);ViM~sN$BlN<67mtsxILzGbddq~&tQ
zG*_10&Xv7y>Cri#%dVVU`KM>DIH;&w8Ea${y~ky<3#EAnt<>e`{pdoOTs10})z;;5
zCA?fri3JbPXZ3si3NhM5&g({n>>g7ommIi9{&kY{?8Q!O=IOT`o-7f~x!n*{C0p6|
z+<^OLw#TPPV989`%(KIb80XF$`B*teYTL~b|1-1Yy=pe|XtO2#&jsS!p;~(9ERwdw
zH@D~3$gs0D($;c`bSqyX|Jp63k9n=Un6q6Faa__rzZJ_t|H+4;-(^e0cS)W2UCv~D
zm%oa?OTB5|<;%S9@?pt$IlJq-960t}K8-VC-%EW=t*8%WWql0hc`&aL!XsOtWo!$$
zm~}=A7ZZFKYl4CGOkmyH1ewiEP<NmS9_UQa)I))KXB3z^vnvuEd!aI;H(s3VgLC%Q
zI6A=wBRAM!Q!866N}$et$reU@zKW^lzFge_$mk&$5f9K_j5BjU|EL=lUsU2wZBG;x
zc|o@MpdT~hGmrXWpOqi3TMt3}9|8FAX&5{HM_~Wl5m<3I5YH@vF=j&u&a@v13&Th_
z5#N*yAB9<)L(%<fD9#{^_q8zKTNra1;pkjH0*)0CP>qVjhkqlXnLQRwcTdDUPxkAZ
z#31Hv3^v`3MeoOP=)Nr;Q&b7Kwle{JZziB|-E<`6q@yYzgI#SIFe}T153@Ee*_RSs
zmd$;GY-qG9P!EOgiX8k-{$?ieO%U<Tha-iULVS}?eDk%A7S~#6(aS`Oe~53425a$S
zxE9R^7s08b2!q!YL3O<dF0G1TKdczNvy0)*EY+;@?EQODjN!yLNyIlT2bSQ%&Jq~i
zDZ$;UGB^<5NTYHX&nU;zJ>}TJ4!|N);+7sd6#MG%XM_$}DfA#0>DbM|ZBFK_J;>oq
zCvLd<l%8F3B=uer``p#BE1A01F>ZKj*`qgF2Q9IYLZ?GRb7r&r=#_~k$3wp5SH2#~
zT=wJ;`#cP-K!0jkOS@3N`azG)6$37>HsEQz0TG=H=v#+)=so>D^!UuUMxN#nvr}8?
zwOOi%T_Lf}1U<}$)BEGZ*Z6+gFf;Z2hYrK}F&(?5!~2suyiDgF1#{ur-^wuUb1D0R
zN-_HPQtlEF=X@-I+3XVhXIFxdW5p;QS&W2-Mff*{+g?xUa~iLO1HT{Ne=k5uc0O_g
z^I&1GfrX_S&!*)dxv2_G9y0GgFB2a;GEnj`4Lu4|@tbK1TK<(p%{39J@8Z!-9f#rX
zV&EAbjRSioVnL?~xRb|C&1Yi}936=*|AoV6X&A<(jK&dW!ZjtMAeTnsr}s!?oeROl
z;1JZi9E@<^U=;2Qg1Jc$7H0)w^Is!S`fWI(dk;tM$YD585P<3h{&=^6eZ_Nza0|>2
zFFyHTSQj50A#d~7eJ>p4zCxvg7b0H`M%C28D0=3J_1uKIHAIPx>)hec${q95-O#H)
zH@ewBvy0mZ=J$c1UqG$CZfLjG9X&0HW1c8+B-{hhsUFxH>4C$wJ-FGW#H7E;(cGc0
zc!v_RmvKvqdX!I&5-sADSRAQDZwDnRj<`cN$sN~+xx=o7JAK@47^HH;)t+wn{R;bT
z`0xDn1Hsz?g0u4nU_k8wP$sxy?mZV=jdsC@`s8z&nMxveV`uA(<DdJZNeaC;?C)zi
z!-+lJPVlMM4~v$v<A(Sq?Ue&AUb4r|?bM)ne%`Ui;M?4)ykU>xllE9yV2{sZ?Xm8)
z9TKRIf3>y4$J%yqxnYa#n{81u#TE@LY!Uj>1}j$C;PrGH=EZEFsO$r;yFJl&C;N%F
zT4L>O3+z`};DlHp#FBbZYYY5MY_sMRIhO_I(Ak>f_7Vl6h;Pg_W_Vt1hK7}97@;<U
z?=ds%-J^h3tw3gl0y|0+Fizn$h<zHa3N$0X6KAP_NlyjN^<@SX3N%qFFj5rA??+r@
zLmbpgfvWc^sn<s@{uX+9Mtq|ozVRK9Eh~v{mJ;81x}-^9>ooa9|IJq7n>8!R)yzu~
z^JytEi`-2pv5YP^MJ6Vth-YGo6o#cpIk}zB*HgrDib^IE-yCtzk$KcFIuYM=B)<8$
zLL-leZ?+TPsEKd-65k9ZzKJHbaS7JQW_OLmHC4;L7HXMF-e=ZZjr>O6>`i8v=8P$o
z$Lz-4uv9No`m5wu?<~QQ49P3c5TlR`39`?S{wLA}+tX#xsC21kmoDneX!&(LO0MjP
zmc=zOaxO4lhD63o_dSWysA-C1`6kPa>xnWhAx>ugij_uhVx>GPRxW*tl|EzRW!QoQ
zDgQlDIyFv`(*23@cl|`Uxidj7gd~Xf(|A$diI;mF5~LzGL3}b3xZjx|`dtZP^iP6p
z3QCr%o0BE}aI%aTl_u%*&L|Q!vXEbkE3O(Da9S+`lgV9`<jPKqY?(VgTe=34qp8dk
z$Cykp&&!k!<(blLR;IXL&6HM>MXzy|%y7$=^`+S|IhA>@cG=>#FH2fK$&}LcEbd}t
z$zF|0UWJecI+!DFPUKg<sYMm3ksI?gvS=SM<Q-y?9cszo_oBlvV#uX=V!4tSr&R{?
zP?@5vlO>bM71c18HZWR653-8+tsEKkSB{jq<cc~ySI*3(7IrXKHm=B(<>a@Lw&qCC
zpj<i3%$)j`TKwBl^CBk|$sBr{9wid<vrMM+E|*rsH;H$2((w<2oas^_E7>!ULY*q7
zaiv_0sFY)>Niu5TB>9JX8)tSiGsiAOXZAh+%Bhm1rB%{;=M*{jYMT6Xn<)b~&5&9~
zGi2I~>GC>%x|F`*mP4bNvT4$6sZ0NHtzp&T9K}8Ovy0>${WpDD*2tB;HS$=sMC$ch
zDv4c}Nw2-zC8p|_^q==uX7vA0?83gwL*kpcs_)WB^Iaxuzf0k)@3Mcvcd1zRU2YyC
z&apJYmRj{OWng`5<GIRHi}>d7r24#<G{J_b7WhDXljdv!%lphw{pyUce>-DAeG{xF
z=J|Hp1p9WH!Z^_qcNcd>phZu7<3`+-sy>KrYmIKxtkGnh4bC>TMTZz$bS5YKsDT4U
zO?1N3)BzY;SMY0wpw%<M!cD-~hHh{_&klEg=26#q;$$!SaRPkt2Q{j>t^LrN8x6-!
z`@!zOFf{(kJ%+$Q=zbr;UY7vOzV8R4)FG&QZU}N?Lojp1DAa5XMFIJnhhC#GD>w`#
zE5h)%Ivn4LZ?@|raNjBtulGbE@ZK2q>5W6>(TQ-aM~ssmg-!RPa95(y$SMY1ViJIl
ziMUafj4>vu=t4in%is)PTP8F%Svb8t3*Crs)WzAj`#u}nb5&^6I0x%$a&Wph7pLo}
zF?e?&-W@K)mkWhRdRd5(#5cQ}Y0<$~i{T5kSawv4K8=g8YEThA7Z%~zvLbk$Eka0B
z?raVz#&P1C<<-SlcB&X>o)zP1-4gf^-?%DEP@KsgyCWso)}oYKUEIu$C`UQ>H&5*>
zNAAsXRJGNiJvp2f&cp~qb$BsOhb1}Gr7CpjSj~OPjXE6PPi%99xQ6Em{l|xicNU%I
zpWC2AB5_SK{<*Nh)QD=zaeP%d?(8i`o%7`we5)KS?w6yzN)IjjkKZ_Q_raA|$CBGt
zdE7BiFkss#1AclMaD-W@x&P^r^nyCnEj<d(>#^idJ=U++W1g2DM%2gB$=P@-^k~eJ
z*@>J|BR%eZ(P03wR=aCDRHYJU5vO&^E5`+TioKtg!k?Wt9}0QxTZ*WAC5SC5!9bG|
zoY=yQ{h(reJ<o0Sk=)k2rbXol;>oFnIGkC4vcP<_watTHa}Dgj<zjMS4$eNz#;aXf
zn3bBz4&DrWT$_d#o~hV!H<`QWN%&-%h_bEm(7ME-(fSx<w2X!!W+G<p9*;e3#$i+L
z7)*N|!Fyde>RN<xry&$4?~a0Ry;1O+FcNiMhv0%C1hY&+nCIu-lp+`tW(48Tw?Lc>
z2}DlK2)w&B98QggW1zz@OdK76$yxqLV|RCdy&wFi`Qmgax1;8HGjrvIi$lrXwDm$<
zEH_OWakq^*sQMR~c^dAC|K3nfy6cWs<J}SO#m%P<ZWwYAcyf?B(Ip`MExUxx-LUGN
z8!F1&iEET-#y@kbod=rGbK~`|5`RD7F6J#I)*t8A!bT<Z3zbMuQlfph5|_p*F)dgL
z=blP#DY@foIlKIZy5mYCcSQUzGmzkh^_|>so$qgKFwpp{pukSBclrSMeR4&QaB5R$
zUGOD{-kUeh*gKC~m*j3XH*v<Fq5bjemlH-EbV9Q-C%pG|!UgU}1x)LU{RxhkKGXqQ
zh-0SGJA9f)n2++I9;Fy(k4EYCh)A?Ysg*ryZLq_&I6KsIv%`wl%t?_;o;lMNP58Vu
zw4E)&Hrv3Z$Og|9Ht?-ujq%NTK(Fq?{wOQvvMsQH9W^fGZ8TOEs3x{)(9nWA7v_jQ
zWDXl<zC*3ekvo~)ro<<0BF$hw$PDMlo1t5v8RArC+@DaOd#nP7lNI=L96Q1WD)6e4
z0^6ROq5mE;be&^{Pt^HO=aTnHGecC28Lo{sLoa_bG#_Y&B`|~HjY^vJ){Cn-v)IHp
zONnocoU`RI{WoRQr$QalWL>i~$^AW50%oR4ih>yCbh7vp*PJC^voRw@dh?7+OA%E9
zuRPApQ>3_eid^wek%u8E(oLT(mP>QRkI(!ksAXI?qDQr^T1Mok#j!{&Kbg(hGejdT
z{WY?;wOYP3QOj1Ijl?&V#5cvnH-m_8+BTsFnf_db2Rm_Q>Sdg*O8)GcCAnKNWNKN4
zoDa&7dbSy|>twog+?Fn>i_&GaZMql-q)N-MD7l&)C9^cqqMQ&dFC(MnQAD&X9~Uiy
z3Swka`(!aamMDY2#!69>SUFrTR?K<r8xSp(&&P|)y>W7K&NxwD9xJ|=sb%FSNMDl#
znR+2!wx5fa`CH?eiI10mYbD5wF$r>ooKWDS1PN`CDEU2-rPjt|QGH65UguJ!zJpq3
zhUdzVS~>ErKrQ-Dx#B-JOFkdTVtzeSHr^mkd7mLKmS;%o3mGz?eWo;y&6H2ntClp+
zlAgLO38a1)I4oNXC$ePDl`J`7l_||9WXSQC88X>1O>SFf%HE4AaoUh0|JA1l=Yd-4
z_S49aB#nF<M{Gi0&fyQa()52C)1QeF&@EYxK1!93f251W!3@bYWJ;)0mXvSGmS26T
zd(o3M#x6%%X>w%qj~r=B%v3WnSB^=p+@voj=oDWYkt1gHRC2aguH4(4BmH{v`?gmj
zJLsVf8DAv3^~}I+X0P$Cav3(!AaMr_@*n-zzltl^%~K(s-zp@F+Z5_il~V5__sQ5N
z^y=0mdCnfh!`AG@j^hsJtSVWvi}~><)8y96Y0`>boQvFW$mz>HPA_iKhfWvArZePp
z)0xt8_-siIsuuqj3#IP4MUs%ZSb~~RpITQV?V`!y3|lJohAfji0XyV<)Gm3v=&fAq
z^Pf}@-~1rHDI>nQr~b|!pYJk*_~s<>&0*r3?wX&{`)fV)u&B@MM15@Kxz2;w`e;vl
zQ%!ucHNFL!C_1BosR_2+?2I9oJHzj3XVfOnF(Z$&`J@T@^51HBtR=d)>57xfy2H6o
zZ~m<9gMS-Xqt<6@IP%%3M<ZK2qs~3&r5$$daNzEM6M{1aVBD(#Fe(!4q^)S_Mxc`T
zrsG*R4EFNC>;eyJ-UBgq(jasgHyAr+560iIUQqY-Mz?5hJekGJ>tSz9%k%+^d@(WE
z2MdS#Ag;g%y+;p)7dH!&JBDIvt<l&_AI=u;hu@eQ#y*d5RM7XCZ-_uAMI;=yMPj^a
z3}WkyL+r%y*mh_F2A||6Rag|#4o9KauV_>*We?8B1RNQV2&;lbe5{v-;87WPusai@
zZL<))H49P3*@(-}#=p<AQJJdZ-<N|9cXF_`C>KFZ)rk45hEcKx9Up1XJ1-B32MdtX
zqYzEz6q5VaB7pd20K0YfyR*AEst7sMs1_3Av^ZIWC2xs)RP+_kWsmW(V)|=}v5@%2
zg7~JlTM2%}veWn(x2Xn@$EaX`@8eP&;KsO#RXMEhl;c)29drtMi~BNn<fB7*xDGbN
z2`|XYG?-3qhd!L8+lYPk@O5%Iqxb3X@Gl*{ZYL(9Pv_z&Zj##)=Tw)&oA_q+3SuUD
zd)A&T$HsH?G2bgk<LBkbCg0*=ribw?J@Q{N?^N4Bo{hPwMh0|jX~4j@)TD@Qq92iy
zxk63qFFigmv$bRxeLhNhaU7WWvee^#YwA-?^>|C)&8<4*az5&)jp(p~oYS{p9oj1C
zl};$<K2aH#-epF9btwW9OR-*23f0LH++a`8j7Hogt)Y(RP>jX<i%{uRgrMVEbn(=}
zB#r*3zyf@*%|~`q?t(p4WB!p`<h$i?haelNbF&cVo5>yWblg&<;dH%JI4?{_nR^oE
zJWRlX+<3(Qj796D7+gNe-s091F(r(9t*gf(v;G*Y%ZxzemoN;NHyTeOL)kUQuHroU
ziJyo3|Nob}2eHfGX)rs2=)L(I1jp1MyuBSrTo{P)<s-0f-*EhVIt;yAaF@Y30MiHi
zqs{msxUt0xBZAqPu-zMuE$A8c_d?Cm!RS6=5Ukz|M1H|QgsD97W0wbV{X9^0Uy0xD
zxx+Ek9W&dy!R;L|W*=Z#Ol@i&uzm|*b{cR=bVI+RZm65$j*UCq(f_SG+)bH>nyW-d
zYEr@ERFe-YQL;*jN2N;Ck77nDkT}Pkoqhq-l)EW${FFOhl#svib;pJJ?nv9_hB{-p
z4c^2J6So5Q2Lh}qg?m>)Z^HmId*O-|?EOsm(*>{Lf@{~D@uq;gnZ2EHA*MePdiKYf
z$4*fELGC8c37Y;+*fydcZZ+tO|FjPHs<+3-uXgx-HZ`XwcG#9@k9V!<BX+k(C3Wy)
z_V)O*1Nobkb~re}4z)YmA>@`Vp0Bb+aF#8C{P;7ezAcg$+2Gn?YqZn#!3z^B9Gzl?
z2W_m-{(=QAFxRwYu?1c?vfy)%1(w#bK%-yg@O{nf_g-_fGMHm_mICJ^sXN^?#iH(}
z=&Gkj@s24>FPh>aJ2Y1KDX?s$0!BPH$#ow!R-o{q8MRR}L=H8>_C{vZQB9G)*A$(W
zo5GR)#xXgj2#YesxRIu?8D@%MM^%#EH9~q(pE}CC)24geZ``Gm2`#h5g!j%pyjR8&
z-;5=`@g=@#Z<ZqWk0#3(Vw+|>Yl&@o@QfzDi6yq#cPv?|wj|5&=p<2iC5ef5s+jf5
z5X)g1>@Cig-sBsC(z0c1R<^`uXUhjowk+M9EeDsY<ZxiFysWR5&OD16sbv&<SAIUz
z$Y@2LBs=HJ@8^mnq+6N%L0|4;mpp0GMkN!7Z(b7z^di3LJUl~M_T)Xjc7{CLk}e%;
z(xsVAx`^pSnf`U6Tpt@HVe!%OC$UX7vCSf$T)w@Oc&C>xM(%!3k|EkynYTS!mhFy~
zXH8<Hi85N!pN*FdcgM+xnd4;1@v*WkdaRh;87tby@shVXUZz*X%fq$t(&Syd-2W9X
z?xPZ9_dI%b?kC9hW{Fa1lFY9k@y*9%SxEo&m*>Qe!})pd$`<3M<Y1^d%^bq+9b%Z=
ziVR5}mmww8*J?*)$f^Yy680iPHacg@YF#FCY?+caG)ro?%#za`GUc*OhRk`AF6MXA
z<=X3XIkYBSo=|i1_&=7;GAs(j{o1x`t}SUJT@q5y5k$g5LXht6ZjhWo1iQPtySrVx
zySobv)c^Ut*ZW~Ej9o-j=AQGM*qb6oi!#KQ9-PBZbLCOXJbBk6PX?;z$>BM<a&J$r
z{M^YOo0K5l>k_5^w<Kw9lp>`AQe|Rnnz&HIx%eYfI%Q-_VAC8qxhzLio9BxDpj_#-
zH&>L6^W=TkJXz8!Pr`YpQ(Z@`h&@EpcW24Nx0y2QdKULLb7fWmH9OwG`o1od3B%Y+
z$Q_A+^cq*YRmz?FmGUTEAtN){N03v^U5;vr)EX~Wy|`0WGhTFh*NDlv8mT+PJy&{g
zb_Pw58s4a^H_>nWc(M%ULycX=<@J+gq~R3V?>I$*e7WUtW{S-IHB~m5%#_Yrvm_vV
zfjm+yl<O&r<bJcoa%;(Ad0EH%RNPY0sa__X19wWJXl{@1{3LJG>*UJ&I_VMIAm!l=
zay+U*>RlToIG|o?{Oe^JA3+-$q)YK132a85v1oxzrxxsGXn}U+Ef9OX4Q$CahgCH2
z@*hn+Ii?QhY7N9+;^UbH^w(%0-&6w%KTX_g!_4EGKF|wemT`s-p3c?7v9$&m{>cz}
zyk+lPXAC?s=DzU&Tsg+qRBO#K<EbUS*zrDfjr+iz1dihcVf=c#EoBh8TpGk(N_%`-
zLQVXKJ=TR<z|+AR_eNS{e1ju*Tb!`@o)fn0a>fUG4L*!wuSX{PJXW~k!~<7+=`@u6
zXhU(ZnH%0`jDqfEKMeQv$EC&r7@W@=6}PC8>H;CkApBAfhU&Ut?92+moz-DzP=>>O
zas+OYZ_Wls!GL_z^;0x@?M*=QuL;PiO2oz$i5Pt$5g(o<qor3SewAjTW^ooOyJh1~
zRyJy%WMg>@Jvd&u7;q^U9z*lcWKACYS{GnRY5^3@3sF;Ah<Ce+u~}y<M$8+FA#J%m
zb%6cFjk*75TnY&(#kkB;Ok7xsr^ib9^|TaovdW+{y$li5Z%(~npK;T2zV<H1cqvD8
zSviiTRiK{TFS!#d&~Hfvo^PsP|10k|b(JXp%Kb?)PM7{<7-~4Lsl&K>D6uPm`<&6#
zMzWQNo1}!r0&Z1qQ)2d3CBD*^6S#tY-eM&t4j}jN|9v%Af!*}Ij9aE)evF#US-z!)
z<8ecQaAvjk{ZQanKeC5v6<*Ps{Qg%Jl+<gsG-H2YyK$IF&8C8$o0_|1p9@tuu&|1E
zgesKs)@3DC7(wsPxE}Nqw_|SWrxK4o@z?*X#HL3|ysuDVPqGq++?i(|$UBukZ*K!C
z;eCU9Q_RUb1yrC@`wG0<Q4W8<a@?pZ!-|Pzh~|@dJiSa^tV^M@r-a)|CHU+)7ItRE
zIMBWbr<iR!cQ_xqlk+fCHy78BWh1MQ_eSkZ<S`rG!!ZrPCsJ@IAQ>r*lMu0rcgn%>
z^rFQ=-y{aN$424Bfe85i3`2l~;>g5cD82>4F+Bh)+xX+>X<uaT8iUN|KDZg>gZXAY
zux##w6+67qcceGoaxeUs+h{yH<b|$%z3?sB6C;;-p!k$K#=joL?etOT#!kzhn_bxN
z=z>_`o<e`-pQ>Eow9^@_Z#%*K=MelX7=pRYhF}vt#T%^$<2^mW!|po3E|Xn*_VzgT
z9QaF?$(js=MF2l|FQQ!xA@hM&)OMauAA|;SklS*4b;v{q+dE+O5ay*8I$+XS2h2Il
z+tpGB_!m219<`hDq5LswKCKqm)5~m+x@Gp59AOX5?)KCj2XRZ6TMW*8t<rc9rmta+
zzzT5LBDkt9*fZ7+%7?b-W@(H23z?bHvOy8M{9XlFW3-Zcnf6xr@_QhfOc{td*DWz)
zge6WFTHxa|eHg6M$Lb<O+`nN6(`+N;95SXKoVv_46IcbB;@o9Zq;?sABHaPF_SqD(
z$C=`D8&h-$FoANBKB6=AdHXg(G4q|*b{k>WA|t3T(8oc;epvdpFO2{8#qCFZaPMLt
zgp+S(tYog~S#KPANKXuzX84QVh`-w#m-qC>lRM0gCTQXlzaFk_tqCW--h1KS19eF~
z;OEhUo+oNJWSTQWdr<>qwo1Df4%D$PY&|uZVbpo<_Q36Ia*#<6G-^bSxycTWZJG#~
zu8C;{nuw0mgd=?c-#nA$u2ztA;N$D5K>2neP&N+_6wg;d^0X#c!q!JgyD!mV!{_?6
z4l$C@Ek+y<C(GK6$?~fpMTX|2$b|G1*^`zm>*JDTN<y-H2~Uy$$0SM7W0uMwLngcM
zW<<VuLcXa_$(Bs=%{=l=9Qo!O`6iuwqtEBhN3ppQ9grv2|Kv%_vH8+KpYi6W1){}1
z-A3%2dzxL$KAExNzmu6NdSs(o=SZGi7N4&(<O}&`(X({9eLh{L&kdAMLxbh2d#EIb
zhf8gTNcnLyQXIU<H%Fr+{a2K{4WN!gzKJH^<dAO+$T!jCo7R(~<)KS7`v`f{a*USE
zuOj9Bst68!M93iZP?@(YMAY>{WzAZCJ2p&;$HvPz@Ub*r3{S;NlO_q$ZA5~E%}fyI
zdkNCHI!StdOcK;2%dn{_(ql-f>|2{9g}1Zha|jt_Ql=!e&Xl!XGo{PBbZS%Sa=lr)
zJb08QEqbKOg5(VPf-)qrZKmV|WlHJNOqsMQQ>IyEib+g{oNS)KTy(m$P^Qb2)N~mW
zm?i_erpheO6uCVlS*9^RzKq(++PCD7#kultS*|n=%awIm^h?v1)7Cvf2AoNhnD)u?
zpIVA+Y?LaQPHED!N2c6Ak;%P+EK#avi%NBlq&;C)D?3M?Hsr8Fkz95nOYUoCirZ`6
z0m)XK`SGlX%9N?QGv(M4`o05lWqbR4d9N&xJC}I>Dk_uu_S`$0P$8SVD`oz@N;wp%
z5EXl+Xm%YZH5=F!q%mG1Mva%r>^dCZg+0-0xmV#|EBn`SSJh&Iq>*ukuBLu-VS=b!
zpDa`EPnHO3INRB0?8isjs40?jkllYTrpg4(X;SY!L#k)Y630gi<htQPSrxZXj4T$)
zyM0T<ZSyi&vu=g_xxGoEI_{FXse5EY%Q|s0s*}GR>*R>bSJ}VmomjPhE7hA{%IuRb
z<(Jbd@!eT3=IRaNf4BwKYPUqh-j>*Upe5d{;TDxvJNkT8;JQZ*gTARDYLF@da=T-d
zo(i^nR>Ajks<>aN0k3|VsI2dW7FYYAk5OM#&NIOFX9m=*4Y6vX5%O(~v3D;$dnqPR
zW&e51t^xFZnnNYi63Z;CG4HZ1uC)>LriOEF0T9suthhV~ZM)c`tCl4WSy~|JlLaPf
zup7|U0*yRu!I!5fx#)z0ADwWd+8MfLF5D+}#puPZ7<|hW1?|{BPyPPs6!yAabHn0~
zqhQ_OhgP2c=-W8}hwB4y{%IfrD}wNkYB1(54#rx$FzneBhDoa7sLc(B)s{$1j*LR`
zsVKz1iAJ(fJXAg<a1SOC|E@~JgY${VA>XWU%f#ohOmv)?1@*64+~CT_#5>vOawi7^
zM&!bRe3R^&hcoN)5T;&$<m>`uJT8EJc_B{Q7QwjJSd^|9i*v0@Fwm_8-i=DJMZXjs
zY)X-zR0_@6%!wZ^#Wic@o72nCVM-Y`oGHU<@=YcAW?wI6sO;GDl2VR?=@p16=6!~I
zle4&jTVWMYC*Nd#t;D&v3Jmy1iNAe#f3jph9`z3=>Nr-O%vSj^LlvvU!zv}rX3={*
zO^H>jlo-l~-drVW6Ui9enY;R}z>nDqZW1YQaw)S}#}yb&zWGJIDZNhp=Ai<;-;j6s
zYixe1gsCBYIrdf9@tVEF)N%fqQbk_`Gv_<1ux32{H)E^dRagac`gex%HdVvh)Cyhh
zOOcI^wxs8nd8th|mDswS-sqY11W~(r!H*@7x2a2I+!ggu;KTh&T%msRdqM?v(t8v1
zyBsr?@h;_Djz3S!(306HgDz!ATw2PWBl?=wmteAX392l|;(F&|oO)Y`f$Wx^I*;8q
zv3YpiF&DmDv*8+^Ma?M_S+mlyxqTW;%Tv&#TQYhcNyM?#1iWe=&+J|-Mt+Y*ce^M|
zNr=FTnPG@J7lP2P>{dz-#NKEA=#=e;xBBeTX7A(;dmnB|d*duUH-2ZmVO8yo3tHaT
zw1NGU#-nj<r5B>O%kU(?6GLh}FnE<a{Ev^qm4|N3)r`Q<a2Gu9<AP(qopEr33({J<
zK)sJM9&~iV#|lSOZsLA-#9(Au3`RbE!5w}$;MxstT;@1nm8Ly%b^*Dof$28PG1&pH
z0)ea106%>0TMoh)YB(~;9*t-6wse7<(unye3kS?kV3*$>2YlP^0Fy=3bb`q{)NZzq
zx5o{|APk(tZMkK(=(60FuMuofwbTx?zuDpDM4&OX8+-nmy`0HW-)!L?VT<b9+}#ba
z!OrjW+3+!!o0)BPSfevvzqG%^yi~9ibX2U^D?AW-+^5@fn)?gA3=uQT0F%b*!(@p*
z_pkKPbH6@X1RLRsgE2Ijdm6o&UK(>#oLp#v)PCfd1x9%IT?ZSd=%6532Mbbk(4w0z
z&gto*Mne~!Tk68-u?~Eu_Q&Pi{`gM53A)X$8FuZxdC&)U$Tw4u^g;8)-WU+fTU1VO
z<m79j_e2etq-o$;x&}(2fwE)`xXok!;=Tr!F4V+;x0=}2j;x~D10Ub=US-`2yXdRw
zZ`KQ=*%Ow3y(iWe_JnOi5Bz0kYV1AU_)9hM+FlbU$w>RDlMmah0l!Ha*p#P%Jih-U
zqmrd7+2(S;Aok=1N)q{IJo)C#kU()Y2xP7|P*QLEOY5IxfPZ6TU)LDfF*-)Ju1=PR
zH<Bc>f1<qalPC)f62%}qS*AxOOTxw^v7=w*(BLF_pO-4}b{XP8zG+R(A~~2jAM%Yu
zX14Un$rhi@+2TgN+07k<P2PF(%Rf){@Lm(eyUqN41>(7|K!%fV&b=*?KIEHMbBiUF
zeACUOP-ZvIk<hu>%nW77Hu`T8YJ8>DC4cd+4U{Em!E(+uR4%22%ZX-@a`{B0Xe33+
zxH(bssXj{X`A5sQz-W0N63soO82MVCB=-kK%e9YDQu02EJt$GqdtsD>b&r(N!U(Z#
z5+SQy!(~`y2=}u?#3Vmd5{HJ#x9Q>HJ3LZe{E3sL`{SkUuXuT7pCGxn<K;w5qC9s<
zk^xVW<O7-T<D?WZACoF~va;C8k|k~Ejow5JMs;q63~HGnbBd@reM%GUm1#0|TAJ+K
z$Xx#0bon<sLo6R<h_gkel%=w(=WB+<Od`)YWr))QGEHQ<w9!kGg7>NN+AdXEGUI0Z
zB3U%~=QFBRqFkXbreZ#`U*);-ogIx!dgMwL{oi$G<7Ic_1esY!%_)X@ThkQT?#j%3
z_Y`U7l_rV$>C!Nn+R)x~QBI@|HZ5H)|4f(J><xH(GDD86&X66k^ghw+q<TDEGEI0-
zYn~~?5;A2qe~;tNXGtzKxguwFI?gMQ8Pm%ouS11=*j6Eb=@YNJ%Y7<$g}iC0l&AFL
z*fRTn>}9oN59elC1iNs)kC!I}HIk>z4bmyKlG0>?tQj>yPSs41&6g&~k&}~U&gIE6
zowq9E2i&C6oFePOrpQ~{sbY9@s_3gulkOd6i29hBa@23O9I2lpGfU>lgdPicUz{fk
zS1uFVqbsDb^F}G}xI@NjAC!+>ugL7yH)ZO7cV(d2ec3VdfsCB=P<9u+l=~KM<WR_a
znb+xyXng)DADT77#eW-di>?U{q&C5Yye60*+X27LJHq)?C!F2b1x;sk!{x;8xUH>%
zxgS+<ew`|MrfA^KPYvvR!`CP6sn_zg$rbXAYjZ=auQo)6r4cr`adVgY&Bibj#3q^|
zWWfO3v*YcY`tT2P6HK%)K@aN08~OcvyG>Ac!4whqOwq5Jo4zf~k@Vbx&vh0!VrYTm
z<eTl?Eir$!9pZO5q4Osvpwb!REnJ`*>58hk?4`friY={&A}nzzqUI07h0kud`Gegd
zEq&3fy+2~U`D6XF0L0!5L~wZ!yS9R%wJ;bDhK1qT?l2gug(EyE95(cemPAFt<7yQC
zv5w{bQXK4l$KgS33WheMV7W~?hNxtq`ST38p2<Yy1m37V@-@FxHUdL(pmHq-w${0b
zzM6~f9(f4eo)0(n;J6eQ;3xT}+@lcZEQ`3gQw00<W1-!?1ULQpygQ9voNFa$rdo>G
zBTEreSc<m$OA-B=cPe)M9iLQ&-}}li_IVlJHeqIpd~=O_^M+l2N5|1~vZWl~Z<oVY
zvjT&7uelyjiSPR>G42I<pqUagd84x9o#`td#q=9*byWf*sORwhq!Y{^i&5g;I3-Tg
zljFvI;Dh{c26j^-_oD*O_;^24f!niqpIXRF)+z-m$vDHwHwyC2H{PiJCEv^=-*{1P
zvC^)B53^7PY4jrdGT)s~U1nYtrcbTHKm|LA^Vm~Nwz(J2o0MY}Zu2%ZkePIiu2m>v
zUaILkC90SOkK4reHHAJP>PY>_Hy79P*QEDGZ@2<}`jyBz&flw|0*md*b??ieGnJnM
z^KyK+!d!R?GxaUle>0;LEqZZ#YH<mcs+XXP##nND5$>NU#DT>Hu*=KGS(iM#{gs1F
z3$n3xXcm5bV0U9qI-hS+vDZ2Uwr8o`lqDj?E&<^W<8UN~{Q)PU;n)y~t11!r*DMT=
z{6f%vW)O5<2jG;KKf-RYr?_Ga)Wdx+fE~s4{k-YVVGrdzZ{9+^dB0<)!T8a**KstS
zPv)j(V=pvw^W^8sgZC15Xe=3p;@NI!Tr&dsZK>N_b;gqI&e-Pfg2-9U+_87UHc$2{
zd>V|o+Jo`?3vWsE&**(--su&4_%1o%5&5Q~1EYb{S-;sE>`M}?S<47$nc#M{AXtxn
z98aF52lGC4a1a8>EeGhOnfApVC#cJux?}@QdSgtR+TgFFEoR)dhv9Ylhv^;OG0PUN
zHMUs(zyi*PE#R?-{p7!_;XBF(vd5ZzJeGKDYR*k6GtA&?o7ekIF<#pg*%#@3S#E-J
zf6WnZOl{WQ8uK4pK|j?BMP^p0Pa243ngda5WdZ$rhIquUDN!Q~a3Vt=vPBQ=8g#kU
zqL1e*^igeQ4A)X>#N$nH_mwevCmO?<zP}OndhpZILvpPSeegQaP1K==#l3}Ax>(s<
z7YpC(P}}T}s`CB_)alPXmwvc&vM=6;_C=dr+{ApzT?JQaG|YWJINu9CzcldW6mRq+
zG|<bIETgS~Dsv5}1!&+NU+X>Fsez5uwp*NHPuOQoJb$H$9v^z5#<Ukc(JOqEUmv}h
z8F#1Wrp@S{$llz8w@J<a=WX16P6O}9YM`&32BQC}W9C`*jxANk8KpX+^VQ*z9U%kB
zHV*^HGkhHK4U}c%n;^$P>1q}z4Jv_B@i#z%{_~gc-_a6CzIiYpMg}iSmbX`uq&N9y
z2>Ir>UZNxnP83tWM0rrm&4Ve4vVCTvd{~w!9d!Ae=#Va&t>}^MoGBJZvSrC7`c|)#
zZSG`C!b9@S=sXGb$&)6#^JHNvHJn2FLYT8EFE0?qwIVUNS0szy6iGfeB}+aP$+p}=
z?jdE%su|gGu_{AG1&xskO<!rI>MMGGeC1?<pR_vc$Jgp!vb)}0HvMrI@8%w2(%M5R
z+IdJ>2M_5wWwhK~K3eRXN66YAQS$3@lzgNQdHby>Nhsjkf1>2*(+C-!7$K{^g-f=5
zxEPs)$k~M<Vt+13ejW^wX8~anv@2XvyGP5&Au&>x5i9k~UmY2jAj1@ia)cWXKhk5x
z<4~M<_D&XSb|OBp;jJz{Qyv_q4wIBFA5W!8<I(I$?2|58mr`YY-&7eNmMSMkripFS
zba_J0wEf-;G3l5o<A-Gmj?!zCnjv-6#t!dKm#b~lWyq>j@l>Qr?LVp1tx|Z$NtPK|
zNuswcK{hzXiz1l5U}oDrY;tATQ+k+dQl)4`yl5SamrLj3WnhN{$t+Be1EUgU)4F7_
z1-;{c(<CK^KBn4qF-=dGE-C3UVrjZmP0Zj9Lx!y4Elt%sL)MY6PVm23O#Z2Epx3H6
zLq72L^6!pJnN9!t05Z;WdaE|CER+34<??Mxg+#ir%lLGq{373kHdTt^l2T$lt3=J7
z-OtCkXJIp5@}7;CvC%d1t#hsHt>G^F^IA!Fnjnv+aFhMk1a7%bl9R_L%k$dF(v@0H
zaN{X*TRBA>&8A9l!BiQvYMRKY>0&Z)w)CAdTio^L$i_2sWF<4?8w%!e*M6Do+q_bu
zjW)`#=pEAFyicAUI3PQhUzPWB?uf7beHq{Nq3o!BDvIjo^0UQD8K3)JS}yu1_x$Q4
zfA4RZ_un50E^mxmO`5{qtQoG2ZV#XB|Dv!>N7S|K4BLus^oVvxQ6Ck&_@IIhi&fDl
zKn-!D)v@`7Iu7XY-}eShJ|Faehl(Mljx)qT=BJED7-2<QW3+NL!OCH#ND3T)uu>D0
zxtU-FckDvcOkiVdf=3=E=%_SBpOXWyv7;GZW0_-Li8<;&o73xJfv2h#xJ<rTyOG<%
z^X+i(Aa7J=&L|$}g5g(Ou*}aDF_T<TdBhd^r`ZQGb0``axuIRZQK)I^gV{R1C_e3n
zT^rf6wk`nwL<eE!ksxT;2D6Kw8LCO4C_5Mib<J?RAm1Fh8wq{#jkZq=V*O%q`X@V#
z%eb@IEfM;G>@wC*#oSIA&_A95>#6KXsm#Lc4_T;VM%-spHufFM!Qg(mn0hG>4OaQs
zx-TEUbPDjQoI3#QG9EXokX`ijw$>KGXwz6+(=S1vz!E&4QiAW7N)X$z6l>`-UYpH~
z_{vhaz9_}vUS${`Qid+%8|V9FNNQ9L3;J)m(0?<Cd~<?)Q@yzy%dVB<fL;YsM^?al
z8n+6zS76HJ3XFSI0ZVRB4w+qv`%TCWJ$Uocp~u)xi6`Wo|9JP=7sM_=a>UN5^y^qD
zv4O0zi+8O}2Nl?3p}>Xd3V4%oj#0y@T+Sccu7K@%1yp#S+VF@Pkp{ID`jMNnhwnk1
z5-A#0&>P0>2d^r;CFg_`Q@5dJ(<{3Qs{DRCAKs(*^E1gpmyMaFqJO7TNAk{Zb`QRw
zFNnGI@oSi!8qXV99yu&Ufg0YWv=>w&Be;@#jg_cAR>9Z56}VwZ{pL|QpSR1gsc$)I
zj+S9uD0QzNrN|mziWOZ;kv6jgMjiQiZaNmjZg4AuoziW{oOxsNxqq35{8!u%Q)c6Y
zX%?n38~0yOI_g%YVjH!(R(VOV8=QzqZ4=mK83!N7Sj5hchQr}VB%cUJ)Bi$I|0o#K
zx&)zPZ~$ue`C*oYFZ{3g;N=o;l<gjkr|$F-8+v0=A2%G?G7_CkM&cs<@M(v=kX7l0
z58pju<?M;UDIUlQcZYKUv*Djc;HQBLes6JxN2xPrM?1q{n==w^oe@p#=F41m4`VP+
zmN>wrke-+|^!}1-l=K8=Zg#*f4Z-|8figtUcDi8VGy&(1(K1oMvjO;02WwOTM$j;I
z#UM;6We#e)J$6&axzx`J&N)_i^u!7;Uvbm&xD~vwTjL4!mqkXlc-+?-wifKzn`eP;
zAI;bYYQ}9VGq~_|%JUi%tWGq6^&BJUe>TLrU_*MY4UxRU5KW&NqAuJ7UV0{Q=GO-2
z4Q41Ww?c)M6+V>>MEBujn@5)T*xLeD4-Mg_X^40|1OA)T$Hi%S=zKvJ1*W?AoTQIp
z17j3~aPPaN2_7Cd#s(i_+?`;AQ1Z<aZ9TlL)PYf&HbO0Q5V~InjvsX}=CckqT-Slh
zOX@Zm{c-J?7KU*5yUlU#=Sg4oPxZmITfJdx#{Bn^Uf5*X0|OsxP#4y~l%X2<$4Ub#
zni|ONrvYCV4J=C1z_n=_NZzY~QzJEzKSdKK4r$`JgC<l|df~|c@=b3t()*s6u(~I_
z2lhk~wVUX1nusXZz|<}pkp1egNK{8qA9d__qQ-s<HP}p5!|ZZ349kd+sF#6qk8IP0
zk8R|ex8$3%9)WUYK%m5F2g<YG0W$MjfPC5JFQ2L7Sn9>dbjKKRIvgi{^wIWYU+`<+
zILWt9WcHeVlcGeqH!)H6kZ;PFtFGOYDBTVuO6ny(>XKwqK(chFo-y)LrWA!{%V|Cb
zKI1Lp<gr}&PbW`q+2_mQqI{X1TOfzZ3M7YoV@<w^Bi~d?v7CQhB=>R(rHf;>7&fGd
z*<&Ahsx?NgZuF5RjXh;)H&5x<%tMm)yGw$$yJ%i<k`b$%B-GDVirUyoikF>?*=;8$
zzdOjXPD8}_j*~pRGm>{EH~Bl>O`cqFlcxrrGQTEL2H8c(Rs9J0Q9V-nY#KqGVWeoe
zdq_~bG2;8gUphqw$-V!A<&8&__*q9w<?U!G_z)+r9O9+#;CT6O7c=V9V&vt#7};1E
zE8FYiWzXv*8Jw0X57*G!#2b`KQkoQfOp$u#(EcW+%IEnhGPopJezr`L?_JVl!K*Z8
z(%2=<KIwP7zkQ&OdTLMx_t42nE7QqSX`;!`XUC86^15-HSiXy6-(s9-*2YP+IXjq{
zoo|>GCy!Lwam=i~xnHU%m&A)lM!XCuiI)%a<E0UEV0}ZF2jgzVfoaKdR69kEolld;
zBhsZ+t8_W`Ax&=c<249Om-v3^;zjPtdyyvF@1)6poyj%p(xl;Zn)J9$-ADLw9!rzX
z^glHolqFqn@Fu2HAOkk@KJ~g>+ODsVtK1LmOuji2tdNjD3VAh$`=xeO;_;RHU>C<p
zDm#s5-WV?izBOWby+(?CYw5eLl^=E!qz(C|_Rd7HwVN!S11HOY(#djzTF$nklVxWA
zNm8(MvJ|CGC0|dIy|vTjzWHn^nLb;h`puE1$LGkz&GV$ol|{0)<0?_stP%alEwXpu
zE?MTkPhPh_C<Q|<OWLuk6884K+-&n$j5D5!WA=0T`s#&zUGZLi7<`q1o9pBmJO4_q
z{E>94M)ZxehHhvZs7!8)_6z^Towpq@r$tw0pu1t8Q+HfZSHY35Dj2;;6`I4;U~)qZ
z{TkHpz*QZ$254Z=eCo9C)bS+45DRY^!q?FVhL6ZO_n8szZH&fo#`wve36&GZ%!L}m
z__HzYHDYdj5dDA5K#YAo0L2H)@M@(w9IluHpUlyQeDlr1f_ETWG>T!@=@cjU_?zKa
z8w<F(48+H~199Bf3W-gv5x>@&`4(GDrbp+NwmWvO@<b?o4pW&I|J2?OKZmo2gn99$
zJ^`>O3Bbc`+}*SbLiW}WEV&wnBfTQ<)G!i*?nh!v`)DK`ih*WqEV|XjV~jEZ&V3V6
z;hw_ngH+5o&po0Z8Te;f276+&(3<+qFY|2bQ#qKtIS0v|nZ>!Chi`WIczP%wcbP$6
zuH=1+evgS`3US`F2t8|xuyh@F9*wvm?pK1X6H8EVz668X{!jTXg<(c1)~_h#c5^8z
zdQh{WhBIsy8Hdl^|2->1JKm^9*q7sKNI6zlmt*?Ya&)_2j(eT>WMx!=h|v|O%B{e6
zZgI|~=eS{FC5*^7U;iptBuidsqr?mYdT_`%$%E)C9!(EU2zM(TmF)SW4_Tee<E}uw
zngUDo6iDvD`x6<bG5My#MFB@@C~c2%yPIB|?aYs>(yx4vOp<;<iM4OpjohIM|2ory
zGqMV^V!1CBUWK`Qs})fNUr&D9jarRK6^?gj7Mz}&mzvzs<lERE%t<|=59ovvW0xtB
z$@hD@1M};t%uM+z;4+WDhrmiClW$IuZ;taWr)FG%!PnX6oLY`UoyxIoYZ)eba>wFb
zDUOtu;t@ZmXC{;&?;mcNzAwh&-9?yLS%_S}0_0fc!>Vf@mfp+3pk&^sRI`wODFfc@
z_KVI<MfuodL}?_UXOl!&K8Z*1)HtluiN&7cXqePQVqZo$u7!odCo&ipY6H>fygxj8
z`QcXe7(7<-!NlvMap8&=ZaaFR+fX-rA36eU&ke`&cEfSC*)TLvdt9380xWjM;kM4`
zzuAeL?aaPDC*0rX%sU5J#@(5H3eE`9aK?%YPUx24gnjFVpwq*_c$&aY;cfO%Z?K1d
zv;%^6IN<XZ-lmq@q3?Y=oNq30brDc%2kS|Y7R6o$H^C_t!60+~*j45i{eh#?*f(5Y
zkG^y5P;p}*OuJce+kGJJ475PjNDG|kYXv>A!uAPPsBC79mE3<?G{zk6|C(dVLNf%X
zo56U18IHCvL+%n2s53L!xr+&o+%jfv(h#S58lo`D5aGKF@#l{rZ^K5&BmV@jPvX6U
z0TfU4F|N)K`?<L=^pGW<oiNA6`6d`K!vN_=$TN?0xKpW%vGa8CwoV6~|IvZ(QED*G
z#_;%N#BBy+EN^a%e=C{YJY|RhU9|9aZ$G@e)enocbg*K+4lX^^!SyCO_%l%(S0D99
z=h*&;-l@fo%6_<9-WS;eng7<Kc9Y!;>BpFHI@1HXEqmY!y@z)lHLxsG9S7^w@Z5+x
z3vc(S@6~YBQv;uQvl=p0gMFHsxRR%dOHs^Um}}zh406#Gb+nwMj;tFR_*kj|YrdcF
z5t{f;4aj;oyHv-hL-|7upXR8+*ij9gf2d-^Aytf>tBU#KR1xA9A-=pzeLfgSUvZ!u
zpFqAT43uP_K)LM_D1*s2{=Wj`!)kv?NcNYy`e=6X#)vbu3NxP=(Fuu>-xp)#WcN7P
z!d^Xl&qSF}m?+a~6Xh*$RC~xb8^|}S_9Tki3Ffi?B#3DpvqdWs=;KKc>t6{n+#*p9
zCzFLP=gCZN33O(@*xiQDhwM@rGrmBUk#9D>;3h$jV%c?t+tzADGN_pQM|_^me3&Ng
zH++~?_GW%*w6ys`_IT<cbDDTa?H+eQKX=)C*-6yNH+kN+GO>l7tgg2c%Se$U-(6*w
z*+^MfJW@Je87ZTz-Q-B6n>4!MCIg>)i26Kt`E_oj6e)*`TJCW9oG?Phg^v^`YY$oc
z#8<Lxf@S-vV6pEKB5}zfGN4(gnC%Re4O_ybW^cIE9Ey;3`y$!@5-lBKW8|_&tZZDA
zAnt1tCGL5W6kke~zGHX`ieg`4c#2FiPmy1pQ{>jWMCqXtBcJJo9zc(fc5I@IOG=bm
zn^I%|d86V&y7btS&gcJF>9mo52KRYa>li29$SRubXYR%BW>bSW`Qa2N6V7puVr!gS
z%%Xne7bgoQ#)*pwv+={@WvoTKXb*}PwWxRr*%>b<Efe^@6C}cejJhL9zS^-L=r}z?
zs;P3zB~|iwr^@W?G+Ed(UAB^K4xM4PZd;nXr2cf1z8|}-yp1KMN!!9y>Ep*u4>jsN
zEi$<`ks~X(f!>VTjn12L?k!Y^!JtaHa;{P?u<I}S2R9q0DaFLHO6I+-lJjTA$;CI+
zaaN3%D(f1VwW~&s8Pv*^?X_axe}d@unJ60vPm;~trFzzDGB-FTNY%I+S@pG69J@~v
z>&BBMcI_1Dsy|I?Q>M!%-5GMI<4n=$KTC2w=15EWjkAs|6C1<Ta#3xq%y_z9M*Q3$
z3vG5V6Mj(Q^p46D#R<`Nx-FZ&Jra$?rxH8mrHokej@zCeWZT?2@vW<q3pamA&FMd)
zFlvMq(XA05+y+7AZE>`^9a=qahkY+QF$2^MD{Q-?O|A-7y;H&I1*%9|u7**s)UfZD
z8oi_H=%l8B)d?C{@I)PnDTc6FZwO;^BkXoGg!K?Z9I!V==|p36CEqmT>#X2=#%S@8
zxy(Og9vf5qzRFDJn*pfbZ3eAX=J<To97DdC<8r+Pd?#C>>mT-qO>~6!AybU$V+Myk
z=CE69iMQLyH>0eu@v{}0GhaGj_h9bTxFP!ND4f;v;!dv@oHlwRd58~o&-a7&Z$AWh
z1>jr}^DVph7!ZVSlY;PiO&D%3iGZVLB=S2%@oQQ%&Yh0IoU~XReHVw;`x6kmED4u~
zr{GPyRCGF*iuT<zpfWK7pGS~ysNXyukd0f$IcT;d2OgTah}X}>-+{S!xrx~+y8<k(
zDuC150%$ENM73TK%x4yHmvAi74NCCay9A4?OJH!e1jVh`vuj<7OovkJCEwipuM|_%
zxjmIiFA4c(EBWRP`KE*!sw(O?s`1Q3lW+FkDo6O|a`ft0fi?ye&|`U`PId*J>>}fk
zZz|HrC%h}2exg9tGX-+VIrY?Wn$bUejC#!o-l}$xaYoRebJ$eLJBb2&>$&grsS^8p
zEAU8Ffg_#CHf<Ex+*bk5p$hcRQ82f!z_n`%%;lZSY=RO#hnT;je`Z+ADx7RoMIBxV
z&k<GF>(AUbx#j`C{fD~EwgFYhxJ8b+Nbm4pC4%`l`IUDxYCo4AGShX0-(SFcS{6I~
zs0{`3*C{9CG~(TD5I58Fdr`;XZK_8qJBw8-u;xTLnp4Z#+n8ChWo7Vlq7VBq`x5g?
z@whoP9AycfHZH-tyTw>Ew+MBih3IHjfJ<HSF}-OX&YjIc;*2Z=?@Gd?cQLT-9YZ}J
z7USy@@tNM6)<>w_9OGsOx5<*!Vv(K9ZolA2`1Pdj_ay`;?+4*=lR%uZ@h1~8Gj+=c
zf!^Lo_~V6%w>|Ns&I9kaj6f%|5$LgFIEuauL%%0O(X!H&ABzi0XE;N<i8KAhPI#K`
z1cO8;6sJ2QX`B<Z?m8joKPSZTt<xqajInV-&)!3Dk9R4b414Y(4Z@vP>;X@*$G^;B
zC#2BBn{9_3lk715xgD;)v_opJ;5WI(?Xw*woVMeRf*n?5+M$S=PW5FwOy^GJz(@;B
zzHNz-;oOJY&Ml`_7Ko(}PIClv*{K%zu*L$_P3WHsHpgLeb1d&|j@MRZ{CjSOyOw5%
z(=~%$b2IE-Y=Uj$Ofa8(^YxA~Uj8z`lJDH)_A&%E8DjenLv*V(!t1F<?7w6d!GN8?
zC-pJ!H2uT+hM0d_ANvFJ;E=5g$00gk*(3H}BhTdMz$#N0Q}1iTrkysj$MlC^^Zv+t
zVFVYxPsi7K2pOu4(yv;GvDHHOI`Yl(e$Z9X!NO(QxHnoGmh{^kw(rkPcK-Z)Ex1={
z!L6hpUPSdp3#~qIROyYMslD)ae@`r(-UCfHYhq!c20lDh$1w|agpz4~9Z<vjduph=
zq=sY*cA&axz$Q)uU3+Pwmx(4?cVho(HFchc>aa3XM|nSWq$R84{2g`dG1kCjzJjsY
zqK>%_)le~4jh)_V2tGxY2~)*CZB!v=RM2&i3L4d@pq)#Etoa)%zJ~(k&fY-TcqLG7
zt_qdLvD9$v1LZ*fKp9@|D|Ze3M0b3&oO}>1^NnI;BKfA5cZ_r)-&kCVk%22>={bp&
zTRL$vo}1iPd7sjye$%itQ36&b$^vGP{WTKBsDb`lW_HH0Gw=GDI62uaRvh)BMbV6Y
z-p0}L=x>x5GGi53k}c&s@}-XXsRj0hGI48>>^e)vu`iM{*9zsGb((x8+pG%lm3s5h
zlHoa8_FVUnpIbb{G}J>*d~}zVyWJ%{z+D_SjFQPKoaBj<tsHK!6(#v5CqU#vy{kMS
z-vo{wDaIE@ii@tB+~A{s7gy=H-c|0a4wW>+k#ctIaM@-$TuvH~5QoDfWZzYHdHvW|
z{H%hdVs)^5?iM1S5<{f1YN%+h4i%k>FmYQNF2~MBNW_*%(U}-6x1D0-Sld|oW)fu8
z1oF$tB#GUSEIn*eq-Q{qOu3gVDL<3s=G}Pd;g~4jRz}nJ9W8#nqeW$Hw0u1hEf(|+
z^_dhcGcJ>7t}**{m#q0DT85m8mMHINS@$MN(r(1a;l{C2eJNJb8)9Wq3UxJZRtz1=
zUT2AuTz<bhH%O2Eij#RC<K)lxIGNg)x=#^1p#Mn_v(*VwXGxZtmn5B3Q)Kj(6xsYI
zMXZccxo64@-5l<9%}$l!=cv!Dr>0YpCcY=B2U(@cZ^u+Q{5w@@n3;R~B2}D~WS!IW
zUOg_A3Gv(!>sv0F56b1yoC>kGV@{l%f7axi_^%4?hAKs0ze>`%9pCQ5IQh&zW6Q<k
zC5(Gi!?)B(RsULf`><AA_D+!ELlfl+H%S{}l;V=a+@G>q7G&4Rp8MQnZ#+q!Kb<6@
zi>An)_R}QMW4f&AJ41pR&6Ek<XUQ6ed9q{{`)_8il9IU9qQH8oXt_cDt=}M#`a2{i
z`j8B>JSr7EPe>L!|C)chFNIkz<np~YQo8V+1aJN(oz~S!nP<KHyz)b42RF!)bAM$}
zPGhW})dVL`x52l<cKEfa9UR*Hi}iQ^<wkY~+*scMCXZE66Qc^{8CASEsfHZ#O-#Hx
zo@b~-<BA&QUscDOX$Baw$`GIY4AHNR0UGQKk@m+3eR7S_bgnUKx3UBMtTEQ#HpcPC
z#^~171SKg0@QHkLny>wuwCBF>UGB(yV6X8Qd!&A}$I@aG{8Tf=I==zX;n&melg;qv
zvKek|9>_bG73T7OUs}#xzLkSfZs&%-b?$iD!waPsyeMUOV=rGv9}e}wqo)2?IXnO>
z3IZ@RIFS8WL3lDb2+zqkpI1a+hF2t-wvK}GTr?)w#^VbT@GCSC*_BE7I3O7>Po<z&
zMJfg_O+(1*bZBvdv13^lE>XXkav&Re>dc0-`=%NB#+-aJn|$+iV?Nk6jAP>p(6vh;
zY*rLvk8Tl$(FgL7{^A+^OXwjifi|Ciz3!Ia_z&)hYm_2zP${l5Bi{a2DSmLLVGDb3
z!otch;&d50F(dw8hjQ%TdoBtuM=<&39r>oor*gP6N94sDRhP@<$f+&IGkSEI(UZJ=
zHG7fQGAn(p6qbK^r=lLS)?0xb`fN^`u=8&yxn!gg{jJ&SH<=96x)MdyN*vEnCpo}=
z<a7Lgo>byi2mXKLorGuwc4a9rW|0CJ-{?Ii-?XSzqRj>+T*xLWjma{<lsLf*RN`y)
z1lqIfhFPgq)O7yx9(7W`3dUF2OH9pY;{heE)H8eaf&G1tm54gYj-E;MH@ndf{aJys
z3<b8frl<HsB?9Alo3gA#61$80x>lgNIky?;yBV8Rj<r7J2>n`y>(k0`*@~HZe(p3g
zso&7MeWA1jjT^?|%Ee-QDJa5I>q1O#Re(mH@^IiyF1~fl!M+Zun64Lx_pf6xsa*^V
zJ)-g99yeFVC&2PRJdVW2!!S3L8(4w(c8-6K{tJYIUl@+h4}s0_AY5XPq~&ry9K1aS
z!+QI`WZY=X*6_maYwRPQ;*K{@M_`uT2#j7g9ELB4Vf5vpNDFht&_OQv!TXcJJ11Ni
z?}R3PPUv!B2s*kuVIEoMzONH*v*V_Er4!U+9nt^&VCH=sVKjw1-cIc3qa<e%1dJ-S
zLukGoo(!@>3oARc&$44@ryT}e<W{$bz<QAIo=E0dW{24SWf<?<p!8pAqhy;v`eBBg
zHOG$(b9|g+j@t{&F=xIxZqh^3fj8q7|Cz(;hB*#2Gsm%3<`~-99Gkdf`I7%0Uz2Z+
zl5b4NH<Ox}p$qxud87&S_;35rU1K=t8qnlpfT|9L?7lX{&IUtlurosLG_p>#J|-vX
z<HTHj<j_+iJM@ueqYvYD`pAmX#Wh161oqZJqZT^2>Z5}T)jB9%t&Ksa=$$d>k4vkx
zu-;S)0q^_4|9(I6v=;un(!wgy!W6%LNQmu+vE|xmnV^kfEBhlOy*~`K`(w-}Eo9Eo
zLW7PL=AQ3|scwC-Hohk|SoFl-dJXPXXuvW~6K0E;ks3+9>7<4e?^WT){b2{*ja-ka
zVa#@J5tpgqvc4K_+*8HE&8kq_rHUIDRblp06+@b-VZ$vo9HQ1UVmR68k~+q9)j;ni
ze2sWQ6?VPVF#4$~E@!Bsh}yVUp$e|4s37M;cjPYW4!7#=xHvsjqJD?U*T<n!aXVDX
z_JxWDpQB@zgi3XvK+!wwEAvCg$kEhj@vmSP%tOAlqR(Un`DSH|G%|~owqCI^s5Dk)
z@;S9H`(u{&kCR@N)MMuH)<M4cy&_RA?xJSVF;OOEC&*$I`aYh=$xroIY1ckhqIIJs
zlYDcReA9w_GwXAd{5c;bGn+<>*`gR(y)8x-Z)gATju?4;C`L{{q+Z02L9u4E%>3mg
z)*VO7(21UM=#+=JFQD$?<RMQQd5Gb4cS+poF6R43$qkoLQsL<&o{ep##X}nzwB1fR
zo7qX9zBcS%v5{m`TWRv%R{ovnC_VnTh()cdXtfwBQ<6r=?$g7?;pQ;O_&Q7y97c%i
ziQ!T<c9^vJHB<_i)7sxXME;U*bi0Sjf#adFsVGcLe8c7Qod~(oAxd6Rhl%bPBULwJ
z<Z<gb>75-f1|D(Jr#@EJSddY!#7L_Tu@cso8VrB#05e?2?W38$=ll8@CZ@LG^5B12
zF9rS1TcgG3JoD{WqownmXz9p5kIVE*cib2*nGRvpZldH6KSzp5vGRIrth}eLmdH)f
z9(+5p3p<$y$BEkPIJvrw8>B0_x4te;4&RKE$2Rfu^jf@3R!~>dPL#sZBsu;!S=#!i
z$i;Q!oLltcNUG$-rAkt1sszNQO2-E3Nz`mcO-YgU%$6@Yl_I`Z=r{kCB9{57GKYOS
z1$h~={Y|O-i7b=o-(~W2W4Vm1tdPwrmGbZcw@HH(V#u9_O{Gfd#*TvGcHC=tJ5Ju}
z&|^GjyzFgXBfBGOB&0_z`>ZF>A2UHLEfmswze45*C}oo)H(>*-#VEK&TCykS$h3*F
zXX_+6Xf#DG?3*f8M$?&zpCLZqW=NUcELmZ&kR8@b#m#0FchFYz@ANvkbZot}>bqZd
z4?ZY?uMTpf^pM;zI>CLelXCa!L)kLxg*>_SMs|1jD)Zle6VF|B{Qa{BQ2)C$zxQ3P
zaf>r4tTCQsHAdXFCRn@eA6zeJhiy&%Me+l783%Mg@#_wFqU?x)$tt+BM-@kZt3oGG
z4S%nzVR)??zVA>&7MbVIXKr-1SBH9U>aNT=OjPP4H(MVz<Mg@3YlIn#j1jTJ7=5T)
z>ph~5Ltmf+JC2XFGl3d6$NkxP+}PF#y?3!c$HSQ3OJmG|317pSVu`~5#5tH@XO$T&
zPMaaBtvOl_wZhYIE9~2Eh4aj0=+Tqlo;C`H*LmX87cZnodLzEE584L%pks4?++FRD
z`kVla8ytw2UjniDS1>{%!_g;)9s914(4~IU!zLNaQ<Je|M>4jv+iiBQWUMMn#>M6-
z=oXNI;`&r}7IJU%P#V4;&O$5SY}oC~hI*SEy!waU8;x9qk#DNVH{RqMP0IqTT3CQ?
z-3p=UOz#J8By?t>)!wnt>s5k1LrdVar3CNEHx2ddXi+PLXE-+@nf<wRuM~?D%g`;e
zjGg^uFs&@Zj6G!t;qGPz?^BQYoL;8n9>F#~$6qMN?C0f}wYMCj=r^wEN(O3Ij>4;@
zXq{4ulr(10nK}2<DZ`v0Wtb^tDBP&P!@~+pAE$t?9eZWych=!v#LnLp2;{Dn$FfRv
z+{(>|la&a1P>E>rP3xWtNDMn~$Q{czQ46|G|4kM9X=ag;)^hJ^7rEuM5;<g+JM{Ft
zE#m#_4&PF<i8ZVOn3am5pSUS+S?|-8*dM`uWlJTuwRq>EKRJhd)1Uo>-*Oc&R8c_9
zS%Hr~D)D%GCAEu6gfSm>Z4!M&WfgF=sla~b!t2>{SU;K%_8&GKSB8}a>@L1k$~|*t
zr~c5tU08xj@{RSrV(eL0h$Ax!&}>>h($C~U!!?&%WZ6i&pMt%e<M8NN3~XLSV`I~3
z)K84T-p~B@n<(_Z9L%1BAgtLKh=2nD*tXOklS2K_D?J2bcLu_Txv4w6h5k2S49<sp
zqiUNMJj^_C@Vq;BsptnAIRb+khGIj7E44TmjQH$~o~N8)sP2M-D0+s^J0UyX30p0l
z(3Sd(|0hRmcN~Js=R>fzVF<?a@3zlyM=X89%_uK!M$rfCd{?l4fZ${mZ%||#b>4AZ
zjO?&RlaCJkcDfzjowh~ML1vg1+rlTDzTn9=P}yq@)#lbnyljPsN4W`i-kiG><}gV$
zN0&Tvw5>6R;b3!2*EOea+#E_xb5yID!{)CUIyGSqvW+=rlW*RUZw$yb6S;{nW|Aok
z&vB2Xiv2QUxP3rY8NZ3S>?8Eb{M5zEeYz<Au8Yv-dg$<15A(<{?tc1c9jXuC2z_K$
z@Z0=av3Q_9zPHu~oORK^s}6p))q&?nZTM2R*>Y4HCvvqhu(&^7f7614QVZVwwUGb3
zAKLFD&)n#TEt9q2!5=$k*AIGq`r(+LHat)CN5NuxZoGKY>f0adK5Aj=MlDR%B;OqG
zhmW26LY-`LKvfgnK5Jms9S!U_uYsV$8u00<0gKP7*m_+RclWBoccm&iIjPbsr;2Oc
zRMDoL8m4bhr5A~2CjMMCxBRT`sv`K0DjfH!Vcc6aEVorhb*?%Zid0~Chg!IYDvlJY
zVz8wu!nUcPe@_+2)b5zwyE~R$=!RbNx}hh%k7vj?JsU!0DfuRYd~=L^Gm?C>phKW+
zUFj!v`+UWd-re1a?EWo`mO?(C`u%^t(IDUK8W1aMN5#r1=BqceCvVk{SP507KET^Z
zFWyADapPbo{kBJ2#7j&H`(X6r<)4o6yq|C%ihR?SeA87sT0+P-yU8~*>GK`V4AVOD
zO*Z-F{=yg;PrfN5-!vuPWFLr;=Eq~ChiQ!1EgLO~U%kZQlNYrGPr0_+L(Y!$kOWP>
zt#g+)3*9ARhPzyw;4V9iN691)Ct2{#Ub0ebWa&F=K4)7?-!3-do@*m+#<tSE&Q`wH
zI*QLl7x|m-D(Z`e%8{?UO}!l|#>K;=!{T8Qyl$8zKN}`>8N=j=`cPTE-&vFsf<={l
zv;JO)R8)sbSD!Fx=@l*)G$W+=UW9atkCbOyBSqyxq)gfqF1>z_k?DE9vSEV1Ty%_*
z$|cb<JUd1v?2DC-%yunyh!*3Xd_SeZl5*3R8EapuZ01joOoSMcZ_fW0Ep^mzOm9>7
z;%&;~Rfv3y43?mrP>DVpE?s8Ch$%m}kFsMW|3{2`No61NL-Nie{yDWFS4G6hu<ST`
zWVuJ`%PorXILUY$Cmk2YOL+{td(;xS&6FhlUnYr{bF#RUGH1@)SXjGM3E#t=G3!)0
z(?6BF4=K|6agwwcoGhQNCCeB;_66i7O7{sV64W78Hq&F&omnh5twb4}n;?JLU$pII
zsaz?n5Fhf*!0DB8(@P;OUMu85fl?ZgRTSi#c?YXy>4j=ZTv9D|ORMF=+-iC0&s-Jv
z9gcKRh|9A|NeflT+8YX4^<F7AGpc3RmTFm1UoGb2Cd%(QlO&T_@x4o@%8WMC<@>Ga
z)VpTNs!dCzZ{kwv_jx5fGpi(i{3_{by+%%U*e_>p&{ynrP_)h+lp*AseajEYp&cir
z{?S7@Wb#^m)_xHqyKiEh@l7rcs}sw@dPy<+E}I^Im;IhUMV)-pJGU`pQxo<SH%BM?
z)~Nm19vhl;z*+AOSnl2lXR11(P20{eo7)8;ZTN50i1+8g)NyjupuJEHEBJaQg_`Xs
z-mKEeH~K01FmlyLIK6@0$u}218{yFcV;E7t`9zlaMITPY3u8q8XN(ggj4@=VA@2P(
zM6nJ3yKP2TIMNuCYmA}4)dcM>nqo71yX^Q{QB%zfUrfxn(QFPbp}^JK3VVvI(E1EL
z0!h|*dBY8TJUlSrn<s*uMx%)i?*v=Dp&vU2^_BjZ9T5Pj2*mPBfw*xe4$Z&EVR45<
zSY#yO+xjG&HcCcASTZK;PKMjVWZ1S&fx1fyZfs4#G>uf;PU7C?m^2t4VLwWzbOiaZ
zH)UrwbpOf0K>HkQpOJ_47xOU6CLe3&=c6fg0zKbC#Ag&@>6#*p_*#s=<QxB8WAT8Q
zo~7i8%;*wqxl_WunG&SimZBwdKQAwrlD)_x3G@_amEkmZsNyTi(7-HJ_zbenbw2o<
zzL%YLA2*fb?WuA&94SWt`DR;nDMr<DGuyiqu>(rsYs5XP-lZsP$6Qu<DOAWiHHM|!
z2`WLO4P#;4WGrk<#-e2lehd!0<E$ycZI?={E33qo<(0@gQV9!cIEp)!*i6pxyI+a<
z{i*Yupq6t*fm-rS$2rt%*6`M}lfL8=ymOu6wpBK}1o<%K!;QD7>F1e8?^A`yo>kaS
z2GU4Y!jQh7d1RQQ0puoTz9(&|K-=Z?53Q`g)^*&`yj6*zE4U$=Sc$9jOXn}EzytD4
z9=*j2Jb0fX-}Ilztx>OXT+uGa&9XA&_2Uld=~8?k-%KIjjLN3}hJ4emX)%oR3UNEW
z0Gf&U7-o@&YVM(?Y)ZpCg9PaR6Njv8G5GIrG{&xo!v4<@Xr&zv)yz=*JQj@A>w_?P
z3%8}z129a>AK}CNusYlqkDLS1Z?r#}=KEq=l@CtLAB`{9J)u6t18)C~!ZLj~7+oKZ
zwl{|&AcK7d3Kw`hcgBnb&Nye~gvlvG*lp~HC#@Zk{B1DmCOD$@s3W?o4#AIcL$Ifs
zj8n-Qt(7Cz-5d;OtHDU5cV@Pu9crK1!hVAdcD1sGcc2{(KeFZCnl0~=wpcKqUBORn
z@NcFKv@C6KnoLt%WP@>&tZ_Nnin)P-xUtXzJE_;4IcyF^oH-_vah{TGJgD7_*EYx5
zp62lDNY?3O4(|jr<SpYZYK0lDPB25Uu^BUi1F&SKA!3~jF#DrEp5D>N!xQ>2+@_C+
z4f;?#*Tv{_x@fUR7Y=p0$ZVnqmp^*!&(g<EetpQ;u7^{``gk3yPj8hzPV4Dogt{*N
zy`qgdf3^RQq_d2Qvj4um-I%}-2GT8^?kydX5<_>1ibyD-B8}bMaogQ3c6WDocgJ&n
z|L4Vb4YL?P!kN!?&JH%)V<etEBi0MsEM8|s|2and;3#{hod%qdX23+z+Kj%gPcQL7
zZ5yReb73B>#s*|peX87)4yh0$+8i|Gn>mK;9cIY+x`w>@&VZlS8PKnz0n>u?=o8kJ
z2Y+>@>W|J0y4aZy>Lugrpm?R0b!O-XZO%HU$)9^Psk2g;WrncKD-8}7Kk3>(8hqbH
zld*-Gd{{2Co5`B|B{Q64LrpfiqRC@}w5V*T#jxuN4$;!!>J$zBU9Q1NuQjMIeht?t
z8Z7^=;A39}8!T04FMV}3JgLU<lhuSX)%fg9GA0dIV%pSLxXq5mif3_X_$ChA7z2+!
zG2#u0!dl^*W9CtCmKj3c<qUj$n1*|E(?u(siQR@-n46P@_SIQfT$~I=^JFakk_63H
zNie*XgyhRfP~DS+n&ydkq?Ldim3XxJ5r=e{4eb&gv%~8Qpm`>Igm3bMZ~A=7z`%DI
za6cw<2w|M6wwXApCK<!sGNC>)3x4CWP_;S>OHE`iHa`N3uZ833#c)JS4MY6kJ}7qS
zgWnzc!17%visy%7>$FgG5Wb1s5Q6rF!I+@xj_I@AaN(sh9CMwqtIQcq(p+$1xhry-
zy5Vg@Z)hC%hLMvm8o&2J+9W^t@Abz(+urgz)*CI|d*eZoA6mTi#i?pvyy@(NJ@>rP
zdXccm)g(AMBqQ~4vUDD#VuxB9+|H-LQz<(tyL8kQ#iB(*7;+uMQNAb=uTCn_y;c^x
zF_~!CEDJw8(-FTp0|sqGce6C<|McDewnd}pdKB#LDe?4U6x3yAWxG2P>t06UgF`g-
z#m3;YXulg=NPzjFB%Hh$hnG3gP(2up_4WzqE<EI=$dJsMER-~oS(#>EY&s_UF8P_?
zjZR0#xiqAlNJDN(I*P>`w|qx7nj7@Tp(mmezMX?jP6P2mvW?10MJFVkWB0@_w>LQ%
zZ^a)UACiLBdVLX|))xg|`a;>aA0Al^fQq(sbIctGH~l2|H%-FuX-V+TO-76K6l|6G
z(U)J92-dE_*@-pCR7f{zSuM6(R^t5Sa=6VMg2)s7v6=&r<UJ4%!(~<!T8zu*iV>1j
zik6c~VHPR*N215tzDXFTybd8bqY<@gG{(FijkvrC2o`Nl;#0}TIW++@TTjORD^n1V
zvKUV^m*CF$C7AkVIX>F1K=08j;9{@}A3E*Dv@`n<7FZ9pW%algu^-7-_oLJNqqwd9
z0Cof4<J_Om81DK71%tjoIpYf&D!;<m`a6Dn`HmRPpD6nG6PMB(@$1?~ywtTBe_U+N
zjB9PF*}Oe>^=Z$s{X6p6qK@4Bp(7hl>?EDjYP>j4jdKj-|8;^E3&&{j))FnAKdVK{
z*7AHH*%NvSQ&u%JV~bQXR*p2Ix6E(m&+EZJ>9p%0d=n*n^WTLY{QAsX`j{;EL%ikl
zWsjcYY{?wa3I5fWc}`^yPS|e6qIzrTSg>K%XX(%rpVb#b>C}^J$tF%t{Nw4wrNf*=
z*Wg4WJ7=af@6C;Befao%7<cGKuuvtEgH}o&K9mgDpk$vWu{^g@w8Ywdxy-2_ldnjx
zP_rB=+;bSaMsoN+<j`DeAh*f<#^$MHpLEHkOLDGw=5raMF^H|R2GJ}!j~x!=@kRT5
zz6l#5orgv2_*C@8JBk@Np@jFYm++@uDfMQT(!JqOW(ePy=MUpT;hSk+${64>f`*4B
zOIIPex*p|{qhHST#nROvIdqRrE7;&z1y8G2GT*<FS9;5SN%m7qf`x6w_Z&1t^d+)`
zv71pz_i>fft`L8f%y$lXR`Q;z<l3cF$hB()7l@8$m6K#3TUT(jWWcC)so;hd;ywRZ
zPAk#+G@4k>?CT@wyRD3~jLNv|co{o99Kqf4?@_6%mRU;;7p2v(#~fjr^)+lP^Aqf-
z;o-eC{CcNGuBU4_{;kY*#)`K}_{M!|9ak=uo-5%SJL!6;n=k&X0is)u7q7RR?KzT(
zEX?B~^BH~FOI<l!NArGljE<<Kx2JSw%KmAUc%h7~s#$Me&9Td>rK_WwKA+|Oq8jnw
z%I@u~ykGd{hw#l8(H4&mt>y#an^@tSlH4k;GZatR-AZA%O71;U!72HoF@7!GQ~QQ;
z!_#86uqoz->=N#pE&IcYB5DOnCxiJQn)(c6RjhPNo$AY#!?IYBmdU^b$$6YB`F;P>
zpF1YA`KCk$PEO#z1@XMtB#zy_$8bX97+w&sRG+9Q2HcF{f6t?6b|;d1E{G@UZ6Edy
z38nk_U=DW-V$*;CW>53y_5hjRB>HmiVIR8A@Zl46Z>B_e@nwW3J74qQo>d;KZ|%t+
z<2;%1%#+?;UTo*+#e%OMjJELP*oz+Qe4e!LAU%{@U0Actg%<Cex&NvYyR~rQfln^H
zvA~6gi(ENusSC|kyKr_>7y7()=Jj*VJZta7f|+t%@X(PD_KDW!t{vOSPHMtlJLaa@
z@tM+&TZL(2jb$%oV8^4vHfB+h8~ogc&*s=rzqutx%eBk3MxqHmYr=`WO?X|hWoE1~
zqUJ;+h7U4gBVmf{K1M7lG~$MQ6Yg1N!kO_VRPA9xg^P)N4HJgUF`-GADX)ny>YjYA
z>R2=063y?Q>4tQCVkliMMs#m)#JUtC*<~5=p0G_nVH>|0M%+_mK-*FSexLz|{MBdm
zF?~AB(WhsjK3`kuv->DLKB6A4+UYT0Pclef8w!IN@@tGC>$@4U>Z1X_r0CP9n;td3
zbm5LA;)S{?K5faedAXr8^JjNvqGX)9F4tz*3Qdlc`Ap3cO}?Be`rm4q!K{+&yHgsR
z{8)pBCyHK%nq2)t!M{B;SQe~7_cI#2cvgcMC9;27sbKqy3c5DekSrt(=~U5>JO~Yj
z$*tRO1$VhAcqK=jZ8xZKNH;a^+N;XoDpkpBS7lYHQoOyfcrzdhTjaRP`QO{u@}iJ4
zGz$I7qmURMg^MOpc&rwMSEf<8-Ai7RuO*?cn(SjFV`hb%Hu|F3>6#2>Q<?t=-?V<2
zgi7I?K;fJ6y-6?*Nx;^3@o0N9PPiosVY`%AxmAgn%Svb~qwu|Z66UI>VDQ@v_@2zb
z(W4n?eJ}&_<ea{l0Vm;`)w0Km8J&fr6SB}(Hw#%aBd~OPI2OtoG9?V!8hzj)KJd0L
zL$T{rD9UDqVrBnOTyhOXPRn3)o#_F+_U?Ey)eU!rZ<>yFgpa2)TFthD%6muASvcY1
zJr}$SBF^3SfZ8{2yiD}Pp~?O@`O_a4+V)2634g3f^h1NYzIc-F3!m5C7`4(HcYM5Y
zxY`HnT>bI7!XM3x{P9EPK4Fh~BR4D%aeBd`fenS*jXtQ*4#&%?2zbs^;#X86%-W0g
zNM<+(wx;1h?{r+fkPau4B!umYgWaNN`FG3jV3HECh{W~PVNeMUgW-^H_$5YQcnkSi
zdc?w}TLN@bqR~N^#~9K0krs{j!7=#0G6u!_(y?1u=7!9XTGXWD<=-UeK2MZf8_8wv
zm5TJvvP-K=#T4<$HHwTy^V(Pp6Hc)amih2a_K9r<z(G45$)l2CT_E|*;mLS1C>fq-
zlJP@76|=jiBB4VnE(_1;NSDPSZOI4fo{0CS6JWSB0R^E+@adI|z^D{N{Y}Nr$~0)r
zOoQFRG!*Bg;*(@0E?=4q-FHd&??MuqdMCqRTQZt!rXc5B9>%@R!!R}R(R=12wzLFy
zpOr#2Zy30`9J~M3z)hzP@EeJa`D38iH4e*i$3e?}0^Vv&lzHSl=(V1Yzq=MA!E^~Q
zatUmoEXNJ271-#r5`#AE!HRBs@qN!e%<EMTt2y<^I<z0#&L75<t4Cp_@&K0A?_nf-
zbJXz*ZsdG{rSOf)ZF#%NH|+cI2Z7Sp&@ieY%_lYDuHB7UHn}M?r#0tg;}-O6)t+a9
z+w)LWM^0bTk!j~Ta%V~>E^e;IY1L|+_gjr4|I^^i5-p}r*P{JuEy;E<Wa335256YD
z#|RS+vNsdYk{K5Y-^2>vOjsk_Qf<z-Y|$NVHfQcZx!q^ZX4388a9TVTl+5)0mW&YJ
zmCZ2G>YT8m+FEPoN`}rQ(IS5ouhq0Jwp{SrmS#sC#jD`NQV%DdtaW0`pH7_I!<oO#
zojG%@lk~he$?K^H|GETn&%|KrYzyYKo57+D>cc~Z;hg9j&c&JG+))+I0?$ac?~o!|
z=N#I%AIQ8NvZH!7kTcD4Sv)A0Uk^*qqG&Wm3g1i|lE;2~^4M5+uyhX<@zTm7x_v0(
z7U7$S>JnbNRKi@#QmQU2Wz+YioUwMOXnuxqa_iw7vuU{aPRh6-poE`$m(W4Zf^o(4
zsx1~DND<pe$HO|wn{!nymL8;He(xslle52TF|TVCbCm3>{zzxT0AU(E;hjFMD(RY4
z%;b4RJT<w9S#mZ?PN13a%|C|<`U~F_3p3ft-fLbX>9u-U&O7_dIXS$XYBxu)!`d>g
zXi+A9=rX4KC*1Y0l1-*mGeAwWyfVYFtEu7S6*auMrG{$4H`jL8&{P=5Sr{iv80X`2
z$)eG!<<Fp6s>IjQZdxtB9uog}yE=9W5}$dbbacx386oqV)H<F_5WSBu&ho-K@!-qH
z_=|3NYK`=GS25mSdeIZBIMG(vh1EQ^vYPYORa0@PnvQ3x*<p|LPYB<9m{85<In~lX
zQO%ccWG^?biV(i(Bp=`DKqXrlNbkCQU)pBFnRZM%!mUdg{<v876~(l$D`HWHA)H=c
z$nPWb*<P4+Xia}U*`3WDow8VWTk?;Vq%$=$jk`Ce(r1F?TYpUwKT9Iha}#*DBA(Y>
z<M`u93@vU%Q+Y+^f?FfG^-VM<7DUn0Dw1#hNEY#rP&SAN;hXD0)KLbq)0W<hgFi=g
z^y6tGUz*kUuy&Gke#d$7E4}EW>Phvf9-QgtA<W>)LK`P)l{m8dH3!Kna^UTw_B`ir
zPuF;Rp6u!%IwMD}$Z%%<T4y@DJM&VFlVl4za`<3Ja)b-xPdW2Ti3>X{cA;GZ7jAgz
zO!W)SH16occB>t!S?b8JE)Mh(ZB4!GoJt<pN`{kY_k%@StZm0@A4Lx*p2n_ewtOPm
znlF-Lb7P_<;}4k8?T{%K?G~=7HQ^TFn|&rG?6y{zW}*?x2N`i=kP$Cx8?olIA<M5A
z(MOmj&Qr4W^i24~!Gt*;CbSKcw_lp@O|~gpwKCz+8Urp9J<LW&13s!Zppl>S_HH%g
zf@Z=ten$M7ZzR`1MjSU;n8(L}8(d_^q;0^FvYX1>qR+o|`h1bB&-JpCDz4LGl7=4J
zY}Db%kvfvk*`3~d#qV9K!+{<;Tt7*h0m<6DBez$rB$r*hP51XpHv6g0(hc5Obj8||
z{i)4<ahi0m)#R#)np{6h<~OMt%+A)JiOgN>WDavcIOd$(|F}fKt;ZFd_eH^_?iw`q
z)L_G-8oXYoprgER%uxlEpA_uUU34~H8f@mT!3(W4*e_qf>~ZQGHbITfn^d`^gDStT
z?8JXLotP5di4z=@xGg+$@V*jz<dipvf|2lsZ*&wU$!&db6h0yfJ&mK_)iDZg|CCtO
zQ06(JIUam832HL0u&_<WcG;((V={spi-+iI5<;IRq21jiC}gJ5{&*7R??}Qu;hSf5
zv1lZpqi~fHW)6K|`_=;kn|fmDE_n@<#~qEAkAIMkW6#rZZ(RnGg>MWck7ml_3``fk
znIwBK$5C05qo0L2wUS@6IvkBE!caXW1TS8Ppxx_G^t%^|=tH3xHX{^2`-P&Mp{QsW
zj8{{ACD+CsYx}xMuO~3b&JnM=*u$;X4kw>G!Xn2B9tsz%5*_mGWlrcH?<pQ*Pqe-4
z0h17a6uJ2e)A-@<H8BjzUTVGgvc4biMo5r19PPZ(v9mY2HW06QCvP+;^hP^NAL#n{
z;@WjT_<iV&-X`L`njegW|3Xk35CJJ`z>`qP>TyXC-A^iFjngDcIThbB6S1p00ar3&
zF{3&fpQ560&rON!dl49x9fl3IVK|o%j?ifN{*NiKcw#hA6${_TQSh&bz$MdYocE8$
zebpFfEQx`^M&S%$Oz$hnDBhne*?x)G;gE=D>5}JI7l(B-V-c|?7T2H0;97kQE?tem
z>$$Nw{Wlh2nu%~aCaf|%8KEs@29ul!cL<9lCS&4e(QB!u;M<-Q(MKoYR!s`b8YQ9S
zxkLm#On`H~WMW$+pkL!eM3_tNbAafsH>P5Qe6LBD6H)J%h^N025O660#T^pyW=0}n
zDStbylhNc(GTavBVT^dxhrP)|W7T|w`Q;;|K_PAjm*RfUVF;I?t;<uyhc&kry@roO
zvSb4OFc}LC(IU6Znt&3~7q_mThfmGsW8b#Ls4-e1UEoVFRC_5LNAJPWS$kmGVlS>W
z*@rzd_hC+tdQ?xS$M${)Fty<k^qYMTbDKOsubJ=h?e0hPu=|3X{$J2`))(Bi`-<-0
zzM@f!zxe6WfX{*&vcrgmOq|q&;kixu!?c;`Tw8FjW(&sqwWoS@NB;h=BefhlQF*fy
z&kNsd5x)8IP3Aa*HK^NHi$P<xxJ|rOpSO!=VYo4?>-FiXV!$eI1FD}iqRBfW&iHP`
zgW}N~X)In%9oes!ny}=83A;O*v-=%$>MWOhKub%i_O;{|4e@*~>cJ-$tQacYGLGYH
z`1OPhr#!G>3sqbG_-V_T^^W`^dG1TZzp9ex#3zU4HQ}ohW46itaZc=P=f?hZ0W9e*
z{hNA$Ofm>$ph+OZ?E>kT63EsA1Nm^B<c!P=lDv;#?(3nX`HVy^Ih(_Qp#%AB<3PFy
z-<;0L<+%D>zH2dvLDL4Yf$+`C-FZB$7|a=(MLa&dh_&B~_-SV`>xPwZ{FxG#9V+2L
z@r}cJ2pxJzHjds9_UJx@+y4%x%4+F#$d}JIAeUy{2Qu#W0Om`N=jJ>8xmCK#%Wm|i
zik$SV{kdUPe|ELW=hd>ovePKyO%2g4iwFI-N-+;MD`xlFA{NWne_U8Z@0udI2;V%N
zBzjp#>3R^pF%`afBW$x>y+Zo9B~S2??8GLOGd)l;cW#LeNTZCyuMTII+A?-hj$ohr
zm7FNL<Yf(N_`|Y>O=U-Qp{#~eg>SkF-$V=Ftl3k;$ER!PaI1z#pU4d7jpXZm5&dr?
znZt-?W|d{F%mT#k)Ks*_-y|<5L^Q;~;;{;-V}Vk-AS`4jHNTd<>}xqfw~7O0?()aH
zisS4ggE&rlwfa}lTKMLZ?5SR@s^)l^+k7})&80i#R`}*Zbv1LdC9_%hM)|Ia#V4!i
zC$p>CFBMD`U2(0<!4Amx_`+crbC!q?{9+01p9&lPD&o&hMZEBKuykG&@~Qj|f1JtT
zv}o~1mG@=+uS`BVn8DJa=|qDx?t75JZq1S<&o7DL!HN9jmB3w5@w~h)mdbH4oO3IR
z?gmkORTIhR?cscOSjl;_BY3Y{7<)*5_Q}v-Mm!3n*02EfRP9Z-tA30-<I9_meK=It
zhw-sqqHFh}CA@g<ss}r|d&n%sl@2F_OB#wd(8H02^BkC`;y{J;OU}2kr<I32RV7z1
zU;Nu?;ZE#y)QNSAo!Fv>WdCk*WQ#CIuKwu2u*uGJ@9o0zBV9x<C;Izm&Kx0pqiF8L
zX%ii}<%GR-!`U+^$BsvYWzGiKGQZS@YkY-ugm0$35Z+m3%UGo?XNzt)VyPt)-OQ<X
zz*OcdCc>H~>^Rbd79p~GYHh;2rN-3q7A^2MBc4n);$d|omOeM6=3zrl+HAy@FN~Sq
z)`X2MWOn0b!kA#$S6vb|+Hb_+RR;Xe#DL-5C67iJXSAv8p5(S)oB?00HK4o#@OG^s
zf6kNH%V2$G?bnxb2?I9!t<QI7glp#Mvzw(phwAGy`Jo=?SLyNnyzVSclMLa%(&wGt
zjgM}0<?*{+$WD^W7o#oNjoQ=`-EYti@kT|72XvF<*ofXHM<M?2O5qpTG35n_SG%t!
zeTQh$b&w{P`)V*LNcg5ygImir__M9BimL{#$`s=9Q_$j$g4JskbhnmSjj+qB3I(%f
zDQLe_IOm>dh8t^er=|u63fp|!q~Pjr>U?!qjdiP3IbcO6-ud2<Z^m_GR#-<?J9XrB
zdnINH*ObaBlG9Gkxw}fJ-cds3mJ&KQgn7h&U3OIomn%xN_^w3OA0=!qMZ!;ZBylQA
z_{@$L28qQ5(SgkPn}qM5CHL-W686Yzo|W*8@!2GNIU)0zKQi~ZDtUo2kBC1JioXv%
z@cFfe_<TL^-NO?PWRJ0{Suz5SC1btt&FnQ92-=l_x5qOuUic<VW;^dGT9v{qEZLWd
zJCza0ZxD{XS9)ViMF=)-48fkpq41~=m6<>&j3x-%^bN%U?@(O)90J8$U+f;>jw>;4
z7{7p6GTsK4XW2k^pdB=CI^qPKpz+BWnVC-bIoAnxt)0=+RJ1vU9!Op0hO3v|(6qz@
zj}kqw`+^7hnR-K~r?8BJH-71OV`39;$rAFywr^fI9p#Nf9ev<u<cq8wet3GjH%_aF
zSF1P}Hu)i#C-bFi;$uIP8in1fqhPxt0mn|HVwxfqtJ@^PQgWlc^JC#RFdCDDEAPFJ
zg#P|8Tv7|e>tB8FaaI@%QX<e|vJwq-qS5AwXmpg(IC?x>0vf{c{!A3si^oePHx^4G
z6EHh95vz+5;rTZKU?TF&5@Gi`4hG%h5dS>}!?k1a(lrJPcf}y<QVePv%M7P?EIw_N
zTt|8AH?;()oKJuk5^;P^B7F8HV)U&<<UdNr#jIrHHA%vq1BtLWn}~C=a|>LbfYt31
zaCcHX+O-l^s*1yC$sWjW5e?m{XzYIzjkN`_7*i69(I*p-E&I#9lSIeloCHVXWY|4P
z#?OT*FgTTmmoM{>-7z1=%?j|J<Rib?BR%ETWmvqv5`#Ka<3VLD%o9iBd)yc(Ta1<2
z&sa2YpMaO0C*tScd6?Z~KBDI@z_#T75MH$uo!hL%`0};b{$Uez$~NQN?kzaxwheAt
z^;kN-9?hp6z~k0O@Vfpe`WfGc_wWb!dFlg7mwiG&(ibEi_=2}#U-7>AUySb4fLg6Z
zR~*n#UZWfGW?d8R$Zx{vnN8_4yQw_?HshSyW-^0l&sA=nXjRaO%@3<GTU(8BW7Q}M
zCH{-j;NnOvn%8Qv-cpk*=1LZ5cYOxQb5ECC*@GW6V0S}9n#pW<*my&Fr5drpEb(ie
zGh+G&Bfb;9IjnBVtM=x+E_*7c*A{$NV9Cq&J=p7h59WQc=HUn%9vo}K^m-cx-jq4c
zTO0a)w`I&OM>-#oF22u>Eb(&U@##*?IwYFOo=!}D){FgJ+~_enfMvork97m3?<SDf
zjRP4Q5XdiM0(pLt@XgFXIxh_5;QSyS=@d-0C&6rE5l_X6Bo3TCkoxxq@>KU+YL*G-
zY|G{B-MP#Q%;TcHd7NUF&&QI5GbyToGshLM`cMHge@He?k3!+FLb{a<=F7$Xc;Zwx
z$L!6fp7?Q=%HE@^c?R3RPUE1*X<Sj3%4VJ^wDCw`qD2b77^JXVFNJwJDGYK>=H%r`
z43T}>D2IHuewNSE{R+8G&OkY4a%T50<obbyT$(vVI&F&hRc1h`)kTtBC_6LZn>)fc
z`-N}D3*QX+C9nC?HMw}4<O+(e`HswQ3SP?2O#JGRt;9RsVT5F+2=nxiozI(U>WUBh
zqOoXjhSzZ1&>D8$Swm~#o13d@n0=y#&#u()*25YOcvZs<pQQ8huRK-^J)6X1mA|)=
z(K{=t+o^`WZ)<7xS@LVVMMK=Pj?TuCr@TmJJ^?k{s!>HZ{VJK)i4V-8icv9D(hVS4
zg~B(ni>i5BbU5=*Rdeo9*;k3*eU9+WyivkE8P(k8EDWSl#e_|je6+TLO9zWj%)5fW
zR4aH~Sgum$R2PN~rLNIX-hWoicB1`W^=1e!l@Dg+vjUn<8pLzQWGC6IANK_H<(DU!
zqWQ@XFIhTYZ%Sp0MJb$fK8XqQ5_!TPfx6@4nX){Nk7mWPRZ28hFNosI4@$1mQqr#|
zoZIh*(PW_Pra~o$_*gJIcn9(A>i|Ak+?!wf`*T*hAA48&(qXRT^)2#d4|7j;4E3VI
z)r;P1JQ&xmC+8cxiYC>GUrsr)UvtUxO>z*Ak3APF?YUbrp)Rzs=hSL@cFc01$sk95
z`zZ|6NphfWIC6x{WFie5+5D6PFH1jVsi*LNjuVw3F3fuB%zeT))2_<L2rF%Jv}2ZN
zZeGHU4c6Fls+%o`%Kj;}jdXKs+Hu=cTZYUQy-kEIm)y7H++6WR_B5x;7E?ZqGUbXt
za(yyYW;ntr2W}d(!zyFC4Kd~b;hULTjp(dz#4R@snY-SQd-@r2e7q66Z8hd0;hZaw
zYZ@yPCbl$Ui%~{=RbaqV`Ubq9FknqP11=EG$?9Oho$9iCYNXGuV`SGfPmdkX>2bV)
zKDB1+bC}F+4m{MC{6l>%v(=|nC&?i$)uUOM9#0C_jJDNbg0T)C-|R+HWjDUL)|ErV
z8#QZ%Hpc{Nb5dh%`Fqo5+Z=7a4bkG0HnNAhCG2urg9~IP^UqK8F;1Fn6{5+RIvUi}
z)8KkD4O&=faEQ!tOyqg5_g4i+tIMt`P{BbX<+*Ubf|JH8xa5F>IeitPp;54t+@G^b
z?w?Zd<1+=#<zrUgP)OyGu+Ri`-n3U^`tD8)9N&?jFLvOkz8zR*)qzDC9fbLn7<p8Q
z65*O5ayE$0qh9#t-)$vc+*IQ0btOWCZ#;!>3ePK1`&J3ZAEF(Z9f3P{!x8l{5~p-T
zzhe{w#z{ub6S@9ApM>7el8_{P(?<AaqVP@0$t09rkB5!$P0I8r3=#&hec*wkD?Knx
z_$K6zhjc2+tmH-#2A@sF1mT-QD>GobHv@ytWkCPDbVq&3fWPQ(KY3(g-n>jSD$Ybo
zOazkt>y6|gA!sUdk<nK|@JxK+smnvLS#&p^S#pGL?h4<0S>Ox1cz4VQa6|nHBC?ka
z{tU5!>mOTOIqZOc2OQC`zayUCm$&Pjpm|T`IWwKnp^*!8Pq@N$xf{~V+_Bf%1FmO0
za7Q?1mCS4&cK1e?cHS8G#tVm!d11)~FKnI+eEi%Ky-a$-d}L2p`T8RCM<AAD2cuVL
z2-YqR#bWzD=((>Ch93@xs%ZpP{EI^K9Vu{BBx3R7IK&Kyg_%k;LSFWP-8K21cF50c
zQYZq?hQfJXA3R73!>{MzXqg=a?ai{s3Xg`_nJ{E2!twBq=&tw4Y;H?5)CR=h*l+Rr
zCnchMb_~o<N1!G=63r$kG0Hj$v%1CL%*+_1j*rIA4Kc_*8zX%jF)(x$z0jOkj5rX7
zyhaH)D9lo!lZXrngBY<;eDT6K-L53!)P+REw3j}N`*AQ_m54>biMS}8A5T`tLI1Q&
zlo!OHo9yDgWGUe;`?Gc8$;)~ljX%O~1B7qFb|j$V(FEKKNJJm&By7ExgpSXXabi^p
z+Nh?%$Ug(KS7oDm?Lg!d7T~*RmbE=3M`>cEc+;!!a9p+URW0=TjKM<ZF|d~zPSw(}
z7<hLA{<WTn9Q%o|^Phz#hiBo#qczZsS_hl`>v7j?BYLgfgv+Nk<KB)faOt)kFDBK)
zY{mg});^3MSC1es_ZZraxR3Orr|59>18$Xl#CGSu`19*86i5DH$o>Yr*SH}wg>M>~
zHRkVv#{4?1DI<k%w1jUeOq%gkW;2eS(TroNThc+N6^A`*Mdu~0Im1q!FH98D(V$?o
zxh8YvdS;Vo?JBo-V`Ei)wiNANbK#pE2MnmHXUNylhSV3nX(oKLW~LF_pE2U;H%1&S
z*Kl(}&Dh+_oavIsu32osjscc@Y}A8ueXaStxebeYi?+DVhI6;uu$|}`Q$!1p^1+7h
zB?D*62FW42<H!g@$t07U_YrfPcxZ+rmDvvb2R8=Q2C%>I&3@sVg<S(hgBr*g!Z)vl
zZ=5Cv(tB1Q2MXWlEth<f&4G+=A1s}qAxvl(&p+4VIIkjytFv+#`a}B1zs3tIB`~En
zlN%~B8Fx3Gm#Wk0;F>Oe@O0+4NvHMOG@cW7>2xQJ1q0H!er_V$X(vdAOafbN2&elo
z$$OJro_^X&zUURhFVkYETp2^B?J@M!isymO@l5CzPbcGeCXW-X$M_`bE=i)^vUFOx
zW$?}D0^Xilz)`abI771N{I3-7&h-LbFDYdEM=~3_Ihdgr2NRu(_{UjJW)X8n6tTbP
zO<ZKwb9$7_fW{SZw#<0mPAOvBy~Fv@YdF1c3}@F{Wo&(4_Bwi1>>~Y@!`jz~pSFhe
zZZ+H$T_eBG8upZZ)q*88Jhi`uV=l=K>%MT!3-Ob`t6|v38vguLN$<;*^xs{{k=rU6
zak7$=O@)huams{oM!XX5m1vT$3*U4^4V$To-|KfJLmF2}Z&4M`MOJaPQ8hcKR&&kV
zYDUWpr${_gzqiOPYP@t#4Xu`3Lt&m1RWun=MeSDNg}+(B#`7yUJ*|T8-75H3Lww;4
zrI+DqIrr`qJ<*DCE?->EypSTQw=R-?ghJ_x%BR(wL3GH^qT}UE$#l!$v6dNZT`3)8
zSJUVekxC3o;oq7hW=u(->-Bj4>L1V96XLk2JeI0D(X7ml;>2A_uKOFw7JA{_Rv*T%
zyF;bhBZRH1f*4^FNc+#yaevgG>bv|{@=A2YPQHwZ@?n^XxAdfWa7%Z|usI7(R1>zj
z?7~Y+otZPiiFakja`>o&<od}DX^iCgstMm5v*RLRlE(7(-2?U<zubYRyd61eiZIJ)
zNAB$9$Oc~>xO}q%t@GtJQszEw?YZ)xt@L5oa6pI+d#{py?K)d}HM8T-Fwy8Y6~38b
z%T?O8?6Aa!<!`O|Ezyep+C4aLzole<TJouyC5LKRu#5Pb3|5<RrRZk1{Wjsno+fPF
z-h?Y}8*{AijqzY(uI?dO!8?q2MaPJd7Yun^biFAdhU{`(=D7c4cYVp2-7XsQv-p`3
z-x;!Pwd|rY3^+=|K>C|R7xP`8E?4zg`B{1|8|X{#h8}wgw<yQ!@!DQJK5L@Sw=%=Y
zxh$UWY5F|lrq6IKeKz={N6SJzhWhJKk)p!~>XIG&yBkmM>_(%DT^P__vTTHJV&-X6
zVWZ7d@k9JuuSJt^ErwfY@v({~L(j?Fre1@keS~N1#H%fS?M^K<XwX8mG_5t*rM(6t
zJ87`3%wP_iYX~bUq`ys_b2=-i@KCU0y*gE7Zqq75<~6Ab?iYsfl^IUwIST0kQ?Q+Q
zzt`TBzMJg|juYKYb2oJcOi|^iJsmkWp@ZyM+jFN&dyZ_;o^wC7qwnBQMCi$GN*JfD
zaLgFto4vv}FJ)G9`jQf-&dF<}=wS7ZDB-j~i4~SgeA*O&`;HMP{@w?tNA$s!7JZ;d
zQR3D17~I?%3ya`5<j8fq%hY%z%!o&WIq_I4xjK`t#iRAGD43^&z_wj5mIQi0WsC=g
zjQ2pA@J+vw!Z-htFy>k^zCKGw=&}r)*^_}`!Z$j?H%o<Ya)obJHOs`0%1nIA7kzV7
z1bqAV7Hw+?63jwiS`q^9n<21X8j733H}kVXup}lJ)%vm{igUw&VmJKh-wWkEh_PnY
z*r#KS>nE(S_oofq8#uto))Duj9MCM@0qxcb<IHoyi=)mMSL2FR1KlwDp__C>xx?1e
z3r)^>p{%jDWV(3az*R5ozu^VU^TODC4>%449-QcjXQ7VLo#_ajosPJ_)gOw~VBByD
z!L-St_|UcwjD=I)-4Da`vEeW(iok0x(GWYzY{Wbc$)2L2nHG+s_)uiHhM;-RP^jpI
z;{3o6>Cg^Ejb|8EFO=^$+yl|~fbNdK`DN}nbUF-mb>aA+D!OFxX{{CZ86O#q`TrvE
zO)C<*;yb%)DSQ*CgrV#s16oF-l~W8R$42ALRr#8SV=(7+3`QEn!n7h5S8QTn_ee4z
z=O*BeN+KExzXVi^UVL#P`t209k>eo0ca2jqnEN#bzhB1TVRkIM<nPISM>N7!VsO7g
z2}CK;wMR6XZ-~a(57C(B7Xxcyr<VWX@jN~O>2DLTv{e#f7A9fN(PV5ol7clhX~;U2
z36C$?xV<n38KQY^c4HvkNN?8cnFCOGCI^?ZtMU3^4JOa2!;BH5(E8sLBnMAL%e_<4
z#cCQXM$Lu6fcZEbzYLw9t$=6CRe0;U4jT+N!og<~dK7NP{AF8kI(sVy?Awl)bqA2=
zb{KQ*j^JGWQP>zCM}L!(sDF0~DtphMW9d1VhhBii`D?JPx&@za4QSP|5$EeQ=E1ne
z{40F(VTR0Zgm0V;o3Uk5Gd>c&$trBgky@?TQ23_bqSjnus?Om9)#>nAon3Y*I6p^&
z3EG<C`R&HtKf7^HTLb#^Ghnwf23%j+ofaK*`L0w~^8WOhE!SO_M;q{-%yq{1Hs+4k
zCX9?RW$-6+E?H#3{%NwK(y*ehyEUgZvEfKB8yZyD@YQA;PP$;jIgf2PR%SRl-{t-!
zN5&qMTz8qfKM0c!ViPCkk96cKI|rJJ&GcE88#|ql-p<Ev43u8-sB7M|+8#j01+re|
zH*17%nlB0zzfK^(O9q>1lOWE#6U55v5o~!qimwmF(Qvi+fm5Ry<R8u1C&R^O9?pPI
z!Cbj8i1C90xlxXVcOcz-1DUE8B-eC7ynQo}J)Hx&>0}rii<f5o+b|y7D?QO`B3YrX
zWOtVs>JN>fwn`lLdc@I4atH0pM7KI3j#n$BD@!k)J%w-1UQ47?MG`HAo4O0%oDse$
z6TVp?e3LGGqc42(RQTqP^a{+mU%<aV3;5cnkiNeP`BSrq6M~AUQ(DAZ<wg80`>S%{
zo7KWM(}i!EPA%f%Z$r87?l2Zz9>#RDGCGUi`0>p$uInW`qgFMnF{xn-yBeko-?Zyj
zLj&1Y&6r=qw7oUbb5z6kcZGirh{s7-XQHBtX74I_X03Fci<kXeLM5HnS8|6iPN!tq
zb3GTIlpM>8k~i30{N7fgSN>PYHsb5nd0)wK4Xe1lkK`~5-xTzcesG!76wj-siDdcZ
zzOLkW`wEtr3%lsaTuNK=d}K}&y{4S^mx}LeK{;*Y?QFR<6<=6=dO01QjbO1?0k8hd
zXRD3*Y?+YGrL`G6l9IuYm`on{mBEZtGEbBHM;ByJ?{T_lzhpOMp2~ASQt0QF%w92x
zoG>h&(Np7iVs#Xc8bt}KDA}$of^khFxM5Z}pO}Ty<8By_jt%4WaiR1t3FZ*{Ko060
zz&YRjx$%)7=l}BMi~+tp^i6cdIzD`P!Hdrqc`(1F8!znyv);S%U7ZWJ20JrR=8uhg
zJ2JD}f!{C79!YjmVWN}yG*o=s8InmeRc0^k?HTdbp0^i>Mp)mG@BF2cvaurzjySO6
zOb0q<IPimx1N(*AvAfJY@8wwwXWB^5gAHH5v0<OHwj3yoV|mGz6$Q2&+gkb+%51o$
zvkmtZ_Mo|RbPr0mU}q0=-mNubYeO@BcxKA)YfO0{z?5O45q1z=O_K1<i@V1BwAh%N
zN{nTOYb=>RMm(M<I^&0id^f?6qwEdYaf@UMztX2+mOhOt^m$44Q4?YfXrE;uIX(uW
zsWzb7J7Jo~GW$KD&lcD9Iqj7m4T|*WmZHZaC3;j{p~u_z_4u-lK0V|bYH(+Lmg?!V
zO=IDlt9l$8uE+TEx{Qq0<+|Hl=(?*5yGSq8->qG!klZ~-;hT%15gcx%Mb&IAdX3a#
zv7hABwA12*9nw9sM>5)%N>;nGXlS~Kex{)Y>pqCL>79ZZqMv#4S-~2azkI)>pl&mH
z><KkCPFCmadFoWp6<_xZb<UMp&d^8&wPO`bNK>$Bf$+>|1z#>x@S<>y%5eo>FIDiY
zx#)E|sZ;+)Cps?eKtGlCyr*o(k&W9??^at*JK2`0g`wD~D>E5YB^q>6;%o;cwzXH{
zL<=P%<hETuCH!(_H&reBD>*-eZ?G-`105t!?^_?V?B|d7s{S~C-w#O-{ZRQU1UC=#
z!F|IpY&Hu+QI9Za#e|_jd>Gyx3&UGskVUtGvA3B!23>T=rqLd<x9~vd5}Aj{t}3?B
z6IG(Uc(E@9?bR|cytCxtbj!di-3<Kvnt_AAGB8%WKQ{)-$&u_BWdxpB_Qu$4!MIv$
zi~ZBBab~GCPCxI74XVH>H{jN7Ana=|%u{#6^6$Mc^)PWZ-3r!ktT48vH6AyyMsR{P
zHg~Z>*i&1$U9rQRh4wgX<AB@wPFUN^35|W7pzqofsq@@my2>3E!Z!ytd&%s}3-0H=
z5PiZ6lQ()nzrYJGKYC(L7@(Ze6K6g;;8`y#EE#11(^2O5{m2rLl9l}4ITZE7WpB1J
z6fX{jVOL?e<j;h|-y#C97KWpyYdA6+2jb@N0JI+M3yYDy7#t}1gRa3C`_>Pqtpbp<
zz!Q)Dxub<YMW+q4_}de$HvmKb^u*|v;gV0IMD76Li8;~e<{K@21Ch8kJp}rZp-36k
z2hp9vF)uI@E?;B^IU)+V!XD|TqLC)MqffVEaQI?0Ziv1)TrCE#en%p>GmzX5*r}xC
z{R)Q_MPtnCXk_Qcp!8D+ItpvG8s;T^+g@-y?}_=Nz0lAv3~mLXsOu7rYided>!p;r
zjS`<<DslNx6dE0lMznbKCdb8L)q^-(eU*q6rAe4&nT$(gQ?TiM8kV%qLhE~y8#KK?
zZb*0Qbnkw!-Ik5kulpiyW`BJ17=$U3*W6}CKjhmEMonEc-a1Ucx%U&W`NTxj9G`^i
z8dF4DGZl;4uYiTxN=!Yp5+2{yBQ$#>6wNl_TlN+_wA>1BuWk5m(E*GLKa3jZBN#U1
zC`Kh7#|6ieNN;o+rWR*mnS2gSQqRNV>?Mrqa0>%EJ;oW0Mhr-iGo=ytKWapsk&Rhi
z*_3a7H|3{3&A4iIGp0qhWL{xAUh3JN^G<bQWm7foc30=2SL*C1c~p9F8hn$lDSAC^
z&X=wgjb74EH(7`8Kk4w?B<bN*(Pj4%T}C<UbNK^(J|1JhYds{#B3}0PFH9IKd{cPE
zoP(BFaH{kJA6+3ksyZv$N&n`yS60+jSo589Im|pOT~!Zl*jJvT+kUfQj|^d%WsaOa
zRjzkMi!<PgBgY$xCezY^!@OjMBNcpO*SoRuni~fQ-?VG%&OOhhds+IJe@oZ+`3Zsi
zyf~1qs{*Mb^PBs^H+2?4JUAwZ;jWP^{v5@^k5P>N5yj<A!nyl-7%fUd_^U!@KKX%c
zCrp#06UZOU0@=GsAm_FS<g>Pc3^0^k*ugL^5FgjoD`BE{4rk{T(j$E>lCOks8VKLq
zD~+L*@Xayd8*||sE8!b;;hVj}HyOe=D#ABKuS5@2p2X>I(r74rvwxIu&D;Y1UQ|E_
z;hUeA3#fOafLnxbE(qT|5WYF$T}ZW`g^X25F0b&-WZ|1t!Z!znZ$gD{>LwI%lT9&0
zpBHoH?^2!^J4`a_hw*&tGHx7I#&P@0_}aLN*-dIVOQVJ>Y-_|rQ^W0jYq+qyh9~FN
z(0qsJjHk=o=0O$p+E+7TkMu`uucT>P(e^f!J_+$JneDFNUC9=V6z^A>uuj^rO6d@n
z{JmA;dza7q{!1k<|ET1xC!(2tRmsYqm7M>llIOE!E+c$%F}zy3e}s1=uUX}~bXIK3
z=i)ATbZV5x-66Tmu*_xj!GY4hRYsFY*}2Kytzc9+|BNW-+6N=(=U+mbC0YFSGLsQq
zGFhyj#!)Kie4&xSq}U8LJCwm`#TgvaAw%+T(z$wbIxCA(`Mgak=UhsreUl`P?wug|
z^>})Ai{tl2QR0)1qSHbpui8ZL^yzSJ7JYiQ=!s`73}a|u7{f%%6(Ya;&C<aT)2ufW
ze)@6WcVF%?_2u<vK3sarhc<t`xo?FRcUO9d#-|sbj{=wOaFx&LLcRA+9Q;W1zrP%q
zER23}q&;)K*wJH{9TnndO8jNZ1Cp}`e>?v5lMWL%JKmdO&-yNo+;>=XH<KLLq|AY#
zQ4Sm;vmcYIwvt^V*9hU(|JMcg_klIzPTMe2{_mVWW6O3cZFwZpmTI4E*hI~S8U3vJ
zJFEwTCRj3Gx)k(>n6tD>GWayj*!{jK8?Q8Fw1+82%3f-Ov*?0X8*`hLG5?qtv$l(5
z3`<^(S@HkR3!bB9B)R2=G&Yeez9|N5xmcf;iNY{LMK>e4Gq2rce;sPTvQ7qE)ll+;
zg>76f>GQ`f@p@P3i+56wV_NEQaxXo)3gh@q)MI~{;Us;~W25%^)Rg^|N>hCv6b^bX
zvzsOcdc1gCmsT;loV}q7J;rz8Ea}HPzn}}(1a{#mT`f*Jp~=VET3j2d#f|;M6Ct+^
zzG-s*W=&2LwplY)yiK|qR26<{{6ay8YYLWLmgl$Iat$YYr2%sCgj0+H)j3sm^u8}d
zKO@(A0kP`bl_9g7-U>GCqu}jO1&=A^<AiZ$mk86$kh4HMR(lnU8lm9iA1YjZR)u%M
zJJP+Tz3hzJvhJ@6CmmK{^hy<a&6gv56D)jVE`0N^qi~Jz%?ja~#8ygJG*jZOs}dBx
z8CM|tsZ1q2!j)KJt%Py*|M(^xR+;|D*Ym>*6+e7z;0M?CesIzA#|Jxq<of!f%q&=X
z3WE_mGZ=b@gYoHdFye0pL$#4R{Pf+ixXuF`CU~Gvy$5t;=W{B@6Bi0RQ8P9bf1jrU
zk<#0JG!5T+reoo@beypjzS)u?IXaoh@rZ!_1V0=f;fa7Bws72PBVX4FK5vb1GTsR7
z>x>Zg+z9cJ#@KMy82|M#!Hx-LX!6h;2H!1kYnv5##0sVtt?+246$aE>VZPEDI~}bt
z<GwX?3hnVQ*#VEIIbc8wC#?VDijaP?b4qtdrjr+%P4&W+9bR~K$P0n{yil^#3qBEE
zX!p|->$(6}{CgsAv=xj?EU`4+0*_41vE-B~Hn%WEv#mXFW{oeZ0|U`gt2e4M!$ngR
z4wrVK8J-=Evs&S(z8HZ1{Q_{{wGUj>d{H~Y2g^&m@$Zla{Da)F`w?+xFEN~y`5N$H
zLQhQoDVc?`5Bu4sCsx?FV#{7v1g8<%v)%Dk^g<90H7gCkB<ZAZG7iDrA(HdlS&7#(
z!_i81Vc*18p7kgiE5FJfX+$)($=A;+cE!l^E?7I?1?6zYjxo*{-ophM%ZYYl-EiE%
z69;yC<MknLEDG@g8++kWy(b*4y>MtsAFN*=f{=yc`T8rK_U1}V5N1|+ti+L;DD>YJ
zjqBp^Dm)a6+(U7wc#<egmIQb4;|(8{f@2<OII}$i?eF%5yYm42H!cUhhW*gQvoG|e
zi=t&;>7OX*i*1r!m>iUYC#Hj;Qd5mfb`!Aa^#s@)nF#Z}lhC!3^leU=iasq@z^}tf
zbUm;V7n*Ot$kdH^_ID#PVz!`V>{fhhwH<N451_ZtVf5^I1k0Wt#s0y^am??eWO|=Q
ziS1d8OFxHgndfmQ@Dd^|Z=rW#19ng}Vnlo+X4W*K#qCCX+^8|*|7#*WQBAqlrx|y6
ziC?^`C2RHDN`GTJ&fM8ay75#w%}JdLU#oLfgk)2BYtUId25Wz7F=a|O4(O>Pnn)eq
zd9A~F5#9N+fi6c6(WQ~0J_Chs`izr&EgwTRHkWz!7!&sXVoIYk=Dcv#f@(>YtY0G?
z58o_l5Z;3!((N$!M-Lvhku1Pl*4%kdv^c^y`@Y(6poJqBCOYy`14q{Nu;+lj_Oip5
z=XGOY96E4)oC8}&IWzE`8_%|JXM~kIA4a>g$8~q=^zh){MZG!abRgT`31qj=fov-C
zmm>2ZmJJSKkDMSn=LJjtM>yNbHTq?l^*o&xNZpZvj2aq9<M2T7QH!=mC6JSaZ>$>!
z@|Ez-H{qKui^JGj_~z;HFa`?WgiBfTe_O*@zbTSVPa`?oK*`;{F}xfq9^;ZYCeDqc
z-aeTX9hHv8$8j7P6whIA6S=b}Nv>(r*ajKAR9e7G(+ea^tbligZ~j|RAU&4_r0~s1
z;Tw<F1+=p%WL;_@6*onf)3T5kR*6REdm-lq4(6o2LpZWS5r+%s3~N=)6<Q^HyRno$
z%ZKuf^)U82C*4e`Wh`G=#_+FYymVE(=)xgY_L5^9S;KYd;-6|MyrUdW#rt8*E|QF&
zd!=l1LcCsEN@T`WD)W(2{&pM6yRU|_!}JPXtgRr0bu`55zA;L=v&8e|y{M9luT|3b
zO(o-nZKgk{<h5Cn=d-es>(^BB*X&Bp=_38;qRH7Id#l#B<#SxlmxjGOdbZ5t*H44k
zB7YDU#^!SR3-Rzz9w@nKWt@>$&aBdM_6QllCLabcNOu4;YO-j3P5LmGXR_QjgH0c&
z((QX1*9g<B(#wzzq;%SRPv^bdboLmY&i9$AtbLinl~a=0VQ(TQnkA5$@f^4*mbRj|
zSy2$llZ6o+Ecub)Q^L94E}W*P!`Lr1jF!Lqa7|_q+wBiv<G21i``(X3fBJHN6JI{)
z>BCkrKKwe-ogGTt+0)*gwGE{Yb9XQKzX$1CcjfQ%(lb)$#Oqa#+`QC*&X4S=>txR+
zOGE=KY%~0b^smWoDNAx_CZ^kRbzfnfk-{jjq02sN)~>hb8tGB6Fm>R(5C^I%9N2E2
z?0DZ>^Soq$4{@}nO{6?7$ePP!?>nrUEl<c9^iuY}lWf>a9^2@+70(A)vDb1-c5P@$
zKS<v2c{6ToX~tPs<Qk;Jlnql&=qveqI%|!&-^rM7g>UW&-`GhP!fD}~**}ap*3*dl
zE*kRDY(wUMGT^^S2D}8hu93OR?t%KO-=fcTA7t(#bDOjl2Gswm&!sZ6aXG60f87e_
zL?awiu1mwmx(rg&W2%oHCl%{SUz8pf-_qlugEFg;(|4U7i>B&HCZ^1JwDmakpf2_M
z>oT#d3r9-EUVxsaXnHkyd4qVip~*MfG+FpwlbJ3u!-<mTssJs{eW=ME>%{XkQD!!#
z8q{v1K~I_4JUye}vz-cV6P|H6Ci|sa1%Jr7@l%cWZ>uuPUyXr>)#!0ZjgEiSc)FL&
zvmdH4O?FQSJ~F%Um;IEmPHd)vr^gEuEpE?`{o69BcU$VdQ(>PIDs-q<Vex-$c<Z{H
z8Et4--i9*@+i-tw8@}-iMXs>T^tX}N^ez&|KSkouw@7^X6N&c1H|bWwKEgNu^;b$p
zs}k44l-T4fx?|y=rqYRU=TRVD-0;P{U%p6969%~Mgq)X7=rGk8?OVIx?|oOOTq8!w
z;|?7R#y;t;?0DJ@Dd*jg(M_~0wH_EId=n^qqZA!Zw$c+id7j7|=85zQPi(LBgiWm{
zeoOA&;hCQJveQ$#qdl?mKX1&4^uy99PkeH82V2@h&)gPQ$C_aOEh9`fFhbG~L*zLb
zi4NBYp1X_?rEUy|;l}8A-WZRrS%}ln5)aC)P;<gcy34K5af1~a`dCY@uoYVOvqseg
zYmEM952FYNG_4n1PDdwfx9AD?AU8BI@xrf>q9b1Cg|oZ7&}y3(ZsdC*-OCH(zkA}z
zx}F$N>3~Hx!dpEoajd%q-u5wrv8gFeeJ}>HjdA>(5k{99Vd6tal-}?`-Zme^)b)g_
zny}F@;>9L6jC|sWaix+WxYGwyY<$p4zUI?p4=nHKj?i<&frUg%TjJd)SN!SR6Mx>g
zAml$6sE%^M_0KL?)6W%C#qaJu${h#Vil=L?<QBRG;H!TSPJa%@dSB7>EQo|rurIQk
zdSGW5F=tLs9I@$%nyJ7gYohXwGmb2B!XJGn^js$zrPfXuyTS>X?R(<kCSc1{54>*X
zh2GU(&}!v{b)P*kZHqe$e!HQidLTBA3c?ekF!X#LiE;Ax^~_eZII_FCQ!JSTCec_f
ze6#O*EK;_|q5tVb84@Mow{J35i9f5`xfHbAo`&vusTkQU9Vtsv@VTq(E^lPQRUx@c
zlJ$7ot}hNg>I?6-{V=!rU^FbR#&D|%=<sj?y6u`MdB&6Q@$wYWi!H_-&6TitwhDU;
z*P`8+4LC7;Bf9^$3C|L@VC;~sm_Bkpp2_^CZ1*87%RU0@#ACReb{y`jPvYFP)3_6P
z7CpwFgZ1_Eh~9ZielPbhGru91K5WQO(T#YwxDhv8m7a!Qjd)>x6P`p<e*WE**=xiv
zp3#a$tJ=^nTZQ{uwd2@%s&q|MXCvX87;gpFJ8Dq<j3)D6Xz_B9?5C`B=rmr3U7zdl
zQ$Tm#+~1vMLv@*&XUL2zhAeV4;{RjmtmC3gptdbwqjZQMF~Bf%4|#4nq!dLgLQy~g
zm6Q+z>{z?jc2(?d*Dj2;yW2I^?!xbWzxVy)`OPrYC@|0UoO8#ITKY(JRaivt=(%3Z
z!C>BT(g9~YOrw{4hZ{D0b;Ca$-0^KP^8i1%!_V6TMl(Gyr;R5zx_Cij95n>aYV2(3
zgV^~#7(uPUcKT;8e)Yw-PreBCQe#@48vm|VWBdkxTzts9W2XSD;&t-0*<t)T;W$k{
z?wSoT(CQe6MW5s7>q<mkOd=ff60xLzB5pKEgxT2yn5<7g1@~DkCvmUUm$RC*1T0hY
zF|!2pC*NFb!p9pWAeVe&LcX~}zA@X-375$?)#o~4CwElmZ+5~f>(2P*-5FYay5Z2r
z?nun;fd=n19NmzHWur0>IVl5P(=y=KEfb^isk_<T3;*uxivvAM*egyAGj+_LOG<E}
zv;?taoE1AunD<QP;gQj&1nKU@FkMuPi?2)Y$AD7Yq)+Ji$5OPiEW?yhWjLZ&j(2J0
z2zyx$yTD3>=a5l`41x2rAsBB(oiT=C_$>AqJQ)TX>W!Dr9EJ`d!_c?YFhq``PA9br
zfAy?FT6Ps)*H*F^paK?QmFVSFg$cc@u!DRv=tva~Y+!FX^O%jQYLK|U2KJ9?5Yo98
zw+Gh3YEdnGnZrExTrECaspacFd8dkA_ifb7T&}^>$Lv(NP=j^F<v6;w6jR?7<7P?`
zrYedswq+6Cy(`525_X+-EX4Q(YQ6^z#FrrY&>jq+4upLg{}y1VZvn<m&c(%NImo%1
zgMRfn=)W)r+4k9(vNQ{;`}RRs4bE?l^u>G2zA!iLi{5?upx>b^%sAc~gL?F0=3WL)
zH%-UqMLjXLeRtq-D!wIk!PVFlc3F0YclXYm_jN|2`D9nSPPjHd8KKdM*j*csA6sJa
z_pKPb_#TabmeKfX8O5DUBs43+aHlF1H-bZ{?FvC$%Mhgf5@ZJkp+#B%4(s`&g0mQ-
zcs1(&pug!6^MVt|Hz&Q}pXrTtt-W#oj2FCydm*L56E&N?@XeH-!s(uPJjD~<uAb=A
z!xcT*jbgvp6}RYn%Kqw3&zJ}7?8r9(bf$T7S7qUee|~yk37O||mM13l@kES|Ck7nx
z!00G;fIo6WuDKh&nz>*&XE)2ga5pvA2|Wfn;Zqmp*)ad1keZtT<Qrp86|x;vcxp=j
zRBIJ}kZ&%9IN<7DCFahh#->EcPI!C#eVkc(pKW2*)sAx@JF=}EDjPV@`E7@B`|RMi
z*$#Jjo3-AST_6hd%u?V%tpaWTQ^3rbd_#u0H-Y-$`L<ZT$QH8L7OvDCj~Qi)xKvxz
zkb`<`Rp37;P;tx*jkW1_IB!Z_Hna5#s0V)82LF|tVbnb{Jaue~V6x4p`({{viTO2)
z+Ta0S!z_*3-~)G1%Q>fcw}FfkY)bz=^)+KmkuuO2Tjv?0af%WC+h&B72FCbJ#?e}1
zjA8SQvG=|)aNQUsrN&s-)rh|jhH&U+fNpE_5fq{e*Oxl@kg0<sA6g^pNo$?}TjSxb
z*6<6{#=OrhnFFbfnE~3E3vEP&YGViaW($2#GyacnM)Umh<Wm<}!pGJ(PL<SF)ZuU^
zr6@?1(V3}|L%y-}N|gx*y2!mQDbnq6g1p=pBlEe#SyagL-c^68`IkE#mjLNd6(C_Q
zL83_}7^4(fYzfgZ36a>Q%+b3WD!m4T%iGc6+#8T<me40ozPZ;WLXt8g#9~N<Xpi7Q
z&e=AS2l>WyWVk4T!^E&>m>e?*XU0mnC|ib0i=)1xZ|Exxb9|(GS5MI@Q^{9d2f6=2
zN#C}Ej7oQq?0N^O)K*DFXO%pjs*<1Dj&do%S(b*oOWHnn>FDPn$Fe-6*345(IoG-V
z&_i~t_LR%deB`x*S_TjBlZRLR#3RLDE^i4Iou47x7e&hKRn*;VBhxI16vs^wGF>NJ
zBI^94T+>sYj`ftuC*8#VqpR#b;w+_;oTS@PM`}h?^eypuL^(*+XQi~gsFY8sN_i5i
zlm~M-cZzeA1NzRgE7?=-yLwC51s}Qk#aDK2<gUv;P|Ud_@{J0R_Qrwo=f?oq8xtt!
z=LX821%Xo4KTry+0_AJdK=C^eC|*50C9S1WrVLifuysmVovD=VW%g1z&|bE_w3nAJ
zmD1^|lXP}>m%Vyk^r8F6n}NPEo!=||tFJsVP)m)eTB08LNkR`lS;EX9*B5^Bf%@xJ
z=Ypg=b<~e?!$o(0ghaNF6zx$W(*uI!)AL{%%KxMFlj0@yY`kpQl_0(!JIbh}j-sC4
zK|U~Vv2#vW>0{hY;+A%o$%A`{`}&>|vpG%jD>I~$c>$f4_LAxzY4Uq{H@WFWua!%R
ztlY#o;|T7v4AaD#bDrDDxnl7!SKR;ZC%aGem!3WH#V@Ep9u6#*)Q7`l?&sms=h;x%
zTrxu3<0i{0pJ~!#%?z2?W4&B?vq8qR-z+VEZ;=Oy+hpkQ?b2iPL1{GckbICM(sc1L
z*-=z4|0bT08JAAUkUbY=#r8{*5OzcM-nc1!hTj(NWA|kHIt{eCuYvd$nlLHUMEEI9
z%>1Yc%U~@$2yDW=MiaQ-Y>vD=EumA_3a)64Jr7#LI#U-(8O8|u*93`t4OqeJo@|vF
z3O=+&{X<JkJZA-yan^`<V2!9K8+6!d!|Ng?DvFf|?8l6*Z4MZd%w7FrN3>Nsq0uUO
zsHk5Yv6Hj#yUtJ@aKoQJ+)(1@4$tw-H)fYZZ}QFisUG-j%-88Vo-keNg{!*WNF3>n
zOYF{Rcfki0@WuYwz8HDa7xNv}@Tbml_F*-wIOoxu<cHIH*vI)t0Iu|8HrnVwM8<_<
zPh~ia=7l49XE^p<3P=C>k%*Zak8N)gsD(|yD$aNMtxdolYLe4?C!kGY0xClj@DDW>
z)hg;N$Tw>4rq*AG$B{GfpgIS)qvB!HAs(;0#^ZM;^**`mjUK?GJ|4P8%<8A!d8}r4
zoN3#g+NbUa8QBA6r>ONv%7lGNCbpzy;?3q>c)5#S9`cPh`9@8ynLx(5@rl`jXNwSd
znO#^Pi=fe{7&^=v>||36J9^2}t9buv3HLmu*gB&W*6&Mk(W(sh=`o+XqYU}c+*4gH
z$Eo4W*^w$-4jh6#Ylbibj{R5%hGIqgVNm1_!}H={Xcj&UUd+cypEm^mPO8G>x=KtL
zS&5QX6^K$+z%sf5Jwq$uM83J0S%v50s&Mmn6?U$vM%&jTu%cfLmhG%T<5x9ENUlY%
z(pr4te5Po7Ee@Zo#q|rd7(Sbs!{=%cQ9TSEcS=!nuLLJIm0+Mb_if*b@pNbrB(w;R
zZHw?hy9heV3(>T+5Pjy*U!ODxA$j!iH>aO}(*P6|48Z;H0@$p`#k=1*sJ@+prsNy{
z^EpuO%t47~4py0F<5XG}dOq%j+8dcT`6QFwlUZ<W*M~X3S?G78H(U;8;=x#E-7M*W
z$`j1LKG2!E*3R&_odP3+6l%{pGl!`&_TT7)76sf-Jx>PWlc6)P12(RWLsG981fGh<
zJno{J=tiOba|Cw03d72@FuY|K=C-#XNUjWFe`*L04+A#N7QB1nk9lMLFlDnEfBff*
z!9KpYI>86(cisr-|Gzt=&r`f`)YA(qc6cKFq6d2JbVn3(qdxU_#lcKZxZLo-<3v|1
zaCOCGdY<&|xMJH1SFCy9hSv9ZeW37UuYf16Jzzi7X%B2M_r#G7p4gP?i4jho&^zP-
zfBF`e1iN9ot{b#1ouOmkj1!xk&}5<$&W&_LRRZTXjU3rqPnMykIF5WXgY%o-<QsGH
z%{%f<MliF6S1WOP2L0jYN@$aBo=4iEeir8~;q*k!wZr`rb~t&>4igvHVSG8~H>>Pm
z^4J=>-Zp48)&>V3+hA}e`KDfhcY3zy)4>*%Rkm0|t?~bSowc)Vv4K0QEu8C|V-9hl
zfh{ydft!!o;25=l!?oHVd^7X&{^OpAey0<c+n}0$@5SVq#V456N6yK<WrlC>+Tc=0
zdbVw;z4>K|-4Ez}I%bL`cBbfL#?OCd#2hFC82cEa>rq40yBNV{tP$e=H9|KpV~p%%
zjHCc#oSbNcH`Ru)JY;}g(fatdNe|ca^x#&fiyxwk1&^4S8mfZ{uUjK_d21~EyES&L
zZ-S2L%^=||FzRRvO#IjaUM*UpoqkL7D36reE0W~)%Pw-5^P6b$jrw^PnR2U()DDOg
z8y@;Dk&+)5DJFj<$a!jSN;joQv(+i`Mk7&PER2@M%VK1zv%kD<&+GeH{$euCUkbDR
zrOLx!B3}E+(xGbc;hrbc%vX9}aF?XN-K9^cn+$Vulkd&kWLrBoS^lS+v@md&-)`<w
zQJ|KV*SJc#5F-2Dg~&YeP&suk)9BAG<GFlyWq^#}`SoMDzueCAll_~0<uUoD=>i{F
z?(Hc*@3_cc?Umx$NhyD~aFD`O2bprrK^&T^WI=nC<c?BF!%p76sgm;DDp|B&B^RHl
zWOjm^yxr<1TXfuIUxbJB&hZi1<0Iq4)ne3+y5ogp99w^J*yb<FM<QnrgfcfLRFW@*
ziutcVnX$x6cJFnU%z-X)`;nt~esPrLgPcTXmy_f?auVD7j`GRRQEKO^<aI9RHbxF|
z=O3lqou`!7P|7h^rS!|@^X%>@K1ZF%gs##n%v~;h^pGuKUXo?vExY>pNVJBUeVsm{
z7v(2umjh(BI#BZa2FmI1KsoSVfIOKJAoKM-q|IYnvDvGncSs@cLu{nb+(u5fvJn$)
zg#@jz6SHpia&w1L=G5@{eRq>q=RM`f0WZ<6^p^JjdP$Y5kNhm~Vb-9pyngE|!*u-Q
zM^8V|ulEzLI<;sYQOl$A-rUXk%By65(Ow<MF4h3~vuO~sqJw4B8<B7AxUU);C5xis
zWoTf$)Qygpxz7?Lp(IgCTXhs?YLNcvMo)U*9^$FdQ}W0yfA`A}yLHUVS&||te%;t{
z)=82FB+7!W$<pX~XG!mqB7eW_Df$!A#c@ieWYw@IB)OlMA7Fod%Y3<QM!sP`yV;sb
zDfF+F6Sqgm0)vTC5jk0wWlWd(t7po>*gvIo*&;a|xLI0%-y-Lox5<cc+hxM#U7QL1
zEqw>Ev!P_a?60jCtF9;H-sw|PedMBe?Y|_5z9DtDZp!FIx23A|zC4`#LyYS+uuE4H
zBl>9K;a=t#ztBXon-=B;YT<d8CKwdi6bT1gB5VQYH-W8Dc&9bGXX#+Z1zn_+8RNwR
z6D;RUw_3{#O&gozeYyn#ORdnq(i(}^tzj2ogYk9th-j`v|7%Kkk8?nBjta+4t1vj<
z5ij>SGDFS@Mch$+qnE0@*crKp-Ed!n894Nbrww60Cp$T(xO?EnA0Bv1t>haGFI0^8
zLX)3fxYNfQitFBpKk5S?O<&v^<O{+FJT%mpYpzC)J2OOz)L1f6jX!yPyk;7E96AM}
zZd4%Zw=(}VEg1TPgK=kUFs951#zqOji$P(S`6vwg%fm6HS2SKB8a;}l5%n=1+mjP;
zHzWbs<eMSnn<#?>EWFwt{WD{+tS}Z2?ApVueS2I!!VbkG>M%CMqqQD)RwfBJo|}Lb
z^AZrK=m@<z9Wms1N3?s<5vOmY;Y#attT4_%U~(o}b<JdFUthf1(-#lB@LWHr1Sy=^
zxLztki~1rwJ643GBSmO<R)o?YysyKoI`WNqXfe8%7sGo}3F62&k0()UOlBzbD1+V5
zG93I~hV^R)W7RJD$a5<2VNWHD$s>0Qhal3d8qqF8aBtcW=ze9#wiA2K4TmCnS2d<!
z2%3JW#A?ocyuO#Cc?;&{Syo_QKn32<sz9Tcm6+Y73ijKpaQjCUj&N5Lbeme=m>SGF
zSOc4<%->0(pQ@x5gXdC%Lr?b~hihR&ElykVO{XEl5aLpbwJq6OO1=q*D?x9I5`6hk
zjE6NvaLeKmQG{U{ML2o95RJVHv0?5Y_BRhgs~ZEcQ*$7+)(*hZya70;F2F_d&5(v1
z7~IXl@+&!be<%k($v5}ha-hMzlrd*Fd(5&haBwfglPO<6=Pv3(I?A<spxe^!^wD+0
z6`yV}E=;A@rVBP@cgD3qc1ktlUMiieW6>GPd!4XnXD7@G?Sz?IlOau$(W_TH3Qk4g
z4mHGM4@cp1^C)<>jYQWM;W(iT!^J~D-(H~@_bmixibD{i9|C>yjh-8Db)p|Ozf+?{
zgc@_`nHqWB2Py78_%huat(q{0X0#V>>Ukk&o+liectWN0fKGFF+!*7EyjT~!Z{^I{
zi4%H;xngW<&Tcrzc^&GCm3v&E&+Y`>o7^!SafaSCXB7YE0-epw5!WQ^M0uh%*^@p$
zPtHL+aQ%fF^6tApS?7dh+Z^%ekR!@ZJ7VGnM{F8GzKNkX{GAHwr7GN|rg&O=6{gs#
zP~YGH-N(!=KCWaBffDXKTtfKwu0`+U7b~p$X@laSWRze#7|*7DX15)t?xo&__lJ#B
z;L;vzw7O%B*_Jj)=wpMJJvO)#pg{I21-|@HU?UkPtiToyoY(yR!xrIFxTj*4UrnJc
zYGZ9NOrJX{V+E>Dw_y%(8+_f*OubraeNUTWgV5i3xD5tnn_=o-X6R682D?*j&|_yC
z+~&u;A7_dq^hSMqOs3(-t4dArrh^euUmGz?(10E321wpc54S6G_kJ2;KsO`i5E|i*
z9s3lbIn!}AMjd@njj9arZI?dsf9oM;1^rN_dU!oS7r!{ax&BZG@p&z=)TjmQT57<q
zu_iRM8e^F>#pc7!pvrH~8D?{|yV@LM9yNz1`DPpW<`IwZx+HcjkYjjg@My`S84t^X
zNSUvUl+Z%%UuH$hAAd*6{fG2G_lOdgifEZ%5+hf*yEqc-FPP~sTZj6~VeYD4?(vgr
z+to6geDfhgEi1`4737<F<eOIHn?mwU)8(%6vz@Ix9jFlhK?-SIrjYj43YoA-L7lKd
zR;*RX-JJ@#vr8d2#w#Q#)kgXz*hn>O#M+AI>igExGfgS~T6xHWOYUOm>@HCrP7=`6
zQ9j<algZvnY46}5duKaHOHGw1!&Fi~L?y9xDyjZkCBExaqQ6xoORlLTG0aUq&v26_
z)ByLM?Jh>CJ`y~{N7|aHrSoNKaQgd6%mP0#(d6txQ!Q$LUv_W$$m~bn5~cK(zh60v
zMxe8VFLRQ8yBsC(l%te;I?1%rPO|!tljyE+lr_eV^5#zmiC&=;`|e6fGEhqHEqm#^
z$zI%>DP=#8Y5tDPf*}`ub&>wM?qYG&gW6m#Ise*Al3M#na$6tieA8R@IeE*oVznG&
zMo)bx^)askq^dRf#NR_^*4s(<#tJEZXe0MZY-EqpM!sm;i0vC|@i(%OU;Z|7>86dm
z8*NWrioK+KahK)PA$_Rz6vsF(=Gl446_t<Fk*8`beZ}{TuUuWOmhtX>;t;Hs%E!La
zYL%DhZ1NV%Pu`ME#!PJLFE>{Q$ij7ja*VsH14Be|KSYQ@Sd>J*h?7Oe@p3*tUdjy<
zq-lplx%DGad^0-A%cI<pWzmQJy@zzX-9=U<B+KKS?1HE7zOo=tn)C{iRbjzW**91o
zJq(t)6T&3zXM_|!=q4Ej>5^)gD;>1^$>~Y`#p_2NvjW-ixnO{}=8Y6n%~7)M#Ax}X
zVE*5jsd949G<hpC<frvq3Cvh1n@25?51+S4rs+0mJz$3zKHVu(R{ky81NX_>;{D<_
zsa_5ho{+1XPD*39)6#v{KeDCMIZ1nWNs9BYOIh>-sc-aD`p$kOPQNvn??!F$HchzQ
z*Tik=aFPSGV4`S(UZ<MiP~WEXvuGpmdn;_Z*BWzr>EQNtU2N@X%#33b=A|(I;lDPh
zV83RXza_3LV5U|&vqLXiV<kOP8H?=cxmMyXJ)=KkRLo~n!E=%dcbI=VnOUf=D)u>i
zbfUh8uh9dY;WN;MS@7=A@o~qUB6n;)<IWsZ515?w!2J|YoVe<Vu2EiCbeG!67+%}b
z1KTo<nesd*lu?IpklKaLP1RV|mL3%H&BZ}#oF(7fF!0Cv83Fh&BoOH(fjFZdgi}3(
z;Z+`tW#fXef_(Fnd}Cc0hEVd&*}>dl%?gM1w+Q^(G#a+r(YVz=8oP%^W5(=gwAdVt
zJBOoD5fuxs^jJ(E6pMc3n^xqT&*YoF<eNzHO>^>1EcvE)zXWWfhG*J_1ok#2peOfX
zhdB#an~{QL>;ny4(FG>#lQ_1n3leX3fku251}&rpb~UvkS9_!QZEv)qhQ#G;A6(1s
z3*CD~i1}PZZ+bD(?27TBcQMYC6(f~Q6m^E4DzZYxky7ltRtnv*W%&5K47<k-MufT?
zC;OLU!t@Gwc(5nSZwO{`uXNLD2zCcoA>w^G7QZb=!|QTnUMXjmVmWqR<ZkOxIeLC5
zM|-UbXc$-Un!f@jrrb-lr%t(+eH7erHOn53u>29&dSwJc{A;i^wT3xDHE_Jb&MS*r
zSf<hoRa=V#+)L@Ns70d&_HxCQ!rrbFgEdPLOfAvJ_!7Kjwvacw+L!DsMnys~%3OKB
zc@h3SNwzU5#Pb_@*jJH<wcYZNkv@>#>H+vCV*pM&6yV*JT=?kcqRXuul%CCj(UBZ}
zogCb7&cQG8P1Wzd7<IBQER3_zcy}gzJ7nNUbUNDH>yB14x?{<Wu1MM0702Az&pNpa
zUbpTHlb@ZiqhBYuaBdTSyc0^em%6vO6UM4KVECU&Xy}=UuN9otR75gIFbbKcBGF(G
zi76iJjrtXa#Gaw3$^^U<g7Ir!AnscPpy@gC4%d9{htwEO2LI8A`_wDG*yZSpx+))}
zU+~6zPj475^1^o`vivkp=1qHIMwSOYj$)VFAnIgnT=D3YGg@4B!b$c!Wj|&=%0m}S
z{NjSzjV>rJbAjGdXP8{%QRj>f+yU3KCu-wxXJqVl!u4J%OuylPao5>l`N)Ak0~KQP
zRaiZXIW-&qw;OZUVMhe4bi|&)j@T0Eh|bLLd(R9$??QGXXj!22v^jDPn&atVbG+GR
zj?BI0c<IW#nIH?yZ`lqZn#_&xVkfu45_OT5XurS`#ZN8iJG8>CA66JL!y2cUH$3`;
zH7;q{U`?<M9*(iWM^goiW+?F2B?UA!s4>p4#jI*u946Nc7*Abs4f!VD7A4f@JZVk+
z%?V3%-Pi^bXSBiIGE;P+7AAD5DSKV%?_6L)J&g(8oHfB|b5o3^Zg2s$G3)!Xv%A_9
zBSun3NPp9d5&F2lQx`3l>mg~s9u`g1N0FHUp3gFXogU{ioZ<9vGs5gu28i6D$2qGW
zst@U*SA!mguF=Ev#d^5*Sr0X{^sv6E9?lQc#aQyqMZMpW@bR}?%5H=%ZyF))ng%@k
zYhj$S2}+8nb5t~g`;ca+uWN?cyP9FfVD4DRHv4(RkYk+4Gw*nNk;h#g4|w>P&=X`E
zDVO=aSn|zy^3AEcky6zpN+usy%Yu7qS$3VX1zz*-=6T**;Uv}FoaEdxwTzw3*-fQd
zrY-a3?t*&V1Yg<C%;3e`BSn#KGA&)DRk1>Pk#F90R)|xaLOMh%q>(7(AAg0Yr`gDq
zF*b5$pp883$8&O)4fh2$(yG0U6xiCx0Sg<s+TB`4msm^kV{7p_XCp1Q*vicBwxZ!`
zC*IC>(r&w*)XcP(W0?-*36%^gQ8BYuCAG^{^5RdGV5v&7PN}3N*<)*GM`^LaO>Td5
zlk}g|1P9Ygb;?%?7n0NX@7!qZCzFS(W$|&cO#t^e|M<w;WnL2Shl@;T>mu$$ou&B<
zC+S%2D4!=dO49~MG28DTTOTUr#eSt6U8I!8tCaGxinAq0r5t}?FFV)R%Rgi7WnQ+u
zjNfNZeXE_Ed0;1-$!9x5?4_uKgVa|#$&fjoa_5DY%-!KFjk3I{A@Y)`r#;z?5h(w;
z1<DdfAF*rZC;?mT=_ymltTG!pe#=^}U$B;@oZCG3*IJ$(w-()-)-wL1wH!&ekqd@)
z@=WX{>W#ZhKj9(kDm-P$bWgEQbdo(6ouu(l7fB!GE_S!P<<A~!<|O-y=?7ot{dmcP
zC@;=$JZ1R-Z*ea173W)ia=kP_DuR7w!V6y+%b$BjAop8()cxp3i{E2*cg4m_r|0o<
zVoicHdz{D|#w3YP>Og%^2hsXWZ(Ve<tXiBT>vIDo;dFo;4-S;U4S_OfRFEw350>-S
z=t+ASA_dG>T)@3m?p=16)#b>Jjk)4Bw!g#-$rIyq`P`WekfLrQrMdPf$=N(wk{*wh
z$MjP5+dfT{JEzNokeM>2%R-r3x<r=V-9oP1CgumXOL_S&`RRO6zIz>!b5D=Si81x^
zc;X4sT6a=X>`u#xo&QLe>F1?Un~QSd+cnvA`KI{mK9$kuUrKe}dubi@RYq;ngx3vC
z{PB+_`YzIf#{0(Dy}Jo|=QhQG=w?{mtOe5TwDFCv0T;d*Vy2HV9{ZWVsK696?aer6
zwZQafD^wk{!VG$-ntx|5g)OtX=GvqFCUd;HC~+`Rh2;e*Y?z?Jn!{wBHIBH$`Hg>T
z_Ep8Pcf63@d)Y3CZoxircXw>bbw_BV2MVe@P<zt@)0vBNb+sp)OxRm|h#F%zZ!|gK
zjXr@s_~(od3O4v6?IH6)8gqVQrbc;?8ts1gVfj3N9AHP4*&Jq_g$1H{Ss=C;2H_(4
zX2AGh{B9&rJqks6^DrE!4#SF=aBMCM$AG!vF#X7WLv<uHTJX9&I~qRZ8#VdHZA&yJ
z9F0czpjaGj5sR4vW6^hTELOgVg|a9Pt8c{dIx`;n+;`pXm4Mk52^cXm0Uy>S;4FPw
z1CtXmjaue=6q;q!Ct~`|L=-$r#ERr3TxQ-%{-ShfMrUBpvJA|zWRC8kOq_U>g<A5B
zzCU%bobh;2H=8y-2RFauW6F^NI9C^9if$2{>PyHN)Z`p4#iQ$`NG&ac?)ft0R+r&s
z=wNKsD97Leb}<dAgz8%r_Ajc!Kc^~DLuPRy-^?Q4d?Me3lW)v-(|>ic97VUNP5!SO
zZokUWMY{rROe=7feB(^M(HUKZ`#M9hG=n*Tq3nRTGy;>{Yw$d|2Jc7J;J~gL<|daS
zu|p|(rIaFKbSdh_mZC#VDZC<?S=g==xj#x^&$*Lp9ocL&cX79iku|g!If=zkl5a+u
z6j6gd2tR%D@is0WR)P62?VN`a|2!mE^h30J0p6Mx;3fIyhi)!1Z{%P*`6l;h4(ey*
zVE6NEe7%{CH?6X9_;z3TY4k<+b(v@xn*mjup2%F&9X^%aF|xiZbMLz1jb}2mR660;
zn`GFxqCZOA3DM6x;Zsp3^j?vO&kfPoIXVir&$1)RClV&j*@dYQiHn|*<j@FA4vxV6
z&f$2)*-gW;5X9aUtk({PH9HinUCG(n{upUW?-O<K-Cd}MfA5Qs0lo-6!K@o|9}Fw<
z##VX;8~J*{W|=3R7<!_6wFf@2$4&X#4X%&5f7<2(^QoN4)H-4Ba`MPV6}GbT<k~N1
z<~X|G9b91F)0w_^XN;ccj8kNs>1&<wUjqNuM$-?~L&YvK2lT$`0OxxSSkH_;NA8Zd
zOm;*y|2BhmJ0kIbBU*(!q6zgj!|8>*N?pw8{T46|u|V|`bNHS#N6H3s+~!ff)f}UD
zo5O^BQ|D~~&pQ@4wATWa4=pg{LOVQVpQY6tOKNc~p&;L^J!S<9SflM&YqZ>J4TFEJ
zG1<@t8Jzo0(@@~=VG8Wyj>`3e0```+7)Q@|0`<osqivx*oO`SOw%8YCi|fx6IQN$&
zu1{!#F{Ru$Ih!KxDPNl^>Di{=Y1c**G}>hX?eivZXiKlRy(yNuQESta`Fzw3rwyTw
zc(o})$Rk$sbaBX=oI+Ny`O2PQGETM$XEO&3Fm#3?2K6?;<AZuA-pKjPT0K1F$30xH
zhbz<&|NK)A?&I{J9HWQ)kGi<iRTt%j4Px`YLE={ZmaTq`P`0}f^8hrk{kSI1DjVY(
zns9E`1c`A?p{{NU&uLB3mVDDFH&V`%V`lb?l=$pO`H~kY3;6a99__UwWvNl5{G{i%
zt?U2lXOB{AVtkFf^Naa*2i5Y0{-P<Dy`-73x2*e1DWkj<a?n>Hv&jkf`>JJTyjt%6
z<tw>4zG9Qk>wNM}!GAt-In7ggY;}>9oZI}9u8=w88xQhL0{N!LM;mry+DPaW8#zM0
zIYqv?L%w-VzIjW&=|;ZsR@g`<@=aYgYjzu1%fd0%a?8U;nm19%wFkB`u!Wt>wdRpy
zCo4AEiQ7+mDZb($4KdW=OjgOb=_)xlQYAgdsl<e{oO@)0Z844#*2Gy#y1B6*l<f7x
zTkfa%O6Lf*nEkDm>?>*+JyI?E4*N=Y5a&^6e8hgXmsm`2k=^8*jlG;C&)!io!X2e$
zyh_~1JIG*j2Wpy?qMFW~S&>qPlq;o652alCVlP@7?ByNxFxS)VWrn}K*tf73EVL8*
z;dYX>)lRe<>|~#gQr75F8%<5lReIm6|L~I6`rZ=D{N+VCo?<@0Ns5Y8lJ4gqQFrX5
z&t8Sx=l@%iht~3aytRaUw34Bnt)!KwmE8PgC0`d=OYKr?dArwIdWT!f;W%q4K4~Qj
zuULwvhP7;Lr;xww?c}SEJ@ahrMfF6<jA`!j#=401Rwp@f&sow&x^h<JF4v~H$wigB
z9MGXYdxnd2E^(D^C)}jX4R;A3FTMZhE2kXQ(*IhNoLm(rF=p}Nb2wgpO-hi1(-LLZ
zup~Kb)<Kf(n0siEBxku_iv)L_<^eK#T7Vci2Fk`wfifm3NV*&(<52tT(j`PvnnuV+
z`maX%vePU$M{3e?<<-#svZGU;^!u78IVl6=Qs+SupfgIU8jqG=zsJanUQ=c7l4(*&
zzF9qOhBO_wKrCl2majLL$%gYAB(-{*Y^mNOzYZM`2Ztll{lzf}9#=2MOHRnJ^(SS#
z_F0+v&sn))aY43cUXZw_mt?`*o6@!9v7EJgDP1z(OUS^_G9vQ31a)o@y(t<z?`R?V
zpT@Y#9r3{dP4PRd8O~j5jtM8VadjK@aX$>v#)xy-WE1RuZHl^n%>LB3Kt+lb_gPj*
zb+X3M23xdm#7xqe_IN<P(e13n=;cZ%2CA@>eDk!6BYs6Yao^$uJ@QRbtTP<Soe`7a
zf}$qwc%^bj6MCqog?nIacj|ApdY~0G1XG5QZ@zkBHFcIrFPTd=pS{LvAM_pRi&3Y2
z5pmxaKfm~58qXQ2^jBGb_QTNg{>VxQz>s?Z(2NU2o6$ksEd`@#?_k843d(nfqV<DN
ze9;QSano@8Am6x`hU4i%_8z{9z|qZ7%u|U*H}0!W4v(g`B^pLUVlZ%Z4C3ekUf7)7
ziT}qpJIFVK$v4F}<IrJpJhhqeXp@lupK@w=rc<X(J<m?JM0R}Br!_bcS7s+dZ$~11
z&53x?I|+e-Y1E3)yQ7(b?_tc(J)S{LK_*=P$i$<2Sr~u14+aJHMJoB`qGL8*>KCA3
zWC2=s9)xw>3pukWM4ev|EYBC=bV)I`{VK)pv@$f*mttvF8MIq7@A$!B4E(bk=hjwW
zx<eKE7**jM^);ocN;r~lCXjDlk#BU!H!UWX;qdM<#J?_sm*HTH`dp5;8WlLzjC{n8
zy%<r2AvIO-2^oTAKd99?Fr2=V5omC%!RHP&7~a1II%Jw{fz&%|l_IRH1X<J+y-zJc
zA#b-Bl%jc?QfPfD!IX0)Fj`4Z*n|=sK3|O43yYDTU5scg>WZ!n!tSO8<l=m4Q}Xe6
zU_LIX@=<D#hlfXUu}RerllA){t$;n5_j7RiI%hWIo2%rTqtkQPtIZt0TiLK6-z>S+
z7wz-tpIV%WAE6m&+q@@g26cz^gs${`{ZCgW`=gq6#D`Bw==UoT?LQ_$@hAy9s5zgR
znTQ8NW1;RGjWyX(SiB+<L7I`+x;g?F8iD6|5h&4#fOo%eG>i;`LGMs{;el6~g1`0!
zp-F5YIyMWyovVJxzoAC+P&M>lvR{hX!qU<gE&BPuWUn_o^t{;#>V-b1Jn6&s#K1XZ
zn~(0e66B7iz1*Pc#r;#13q0JMG2FokbwTVuppU7eg9D~McgB&KPWXP?5q<gh<<5Q6
z?lNceo5ua`6lcs}Z+DN*)C{v%!F(^fE7=<ta?Jssn7jE#Lxt<?wOs$g5}UVLV!gK|
z9<{Z^z`gA-d3HPGU9^DTC=1+hv4F;X=H)FX&#X3w@ho%np3d3LBy-MI%<;t50=;Hg
zAg751KJwOhJ-ZO%Eb;CSOSHVf84mfT4PPr<=|?E3w#MBJ%#%22jl-|3(e;rP(2_YZ
z^f0X2VFI@z6LyW7;P(?_%-d^>wx^A8Uc&@;Czzm^?^jGR#Z>0#HM_~#Oo<7a2C|cT
zs0r5knqX0+35IVl!S`(@s6A~0W9n*}+M41A8OR}-dBVLpyRqbJSTAx;gf99G(1pVS
zUFfi5xZ<21n&s=G?R$N!sUzbU8DIr>Pi<G~p--J2a|!ftl(U*Ii}WyKo*rfn)5Dy0
zdPtnCi@IMroIN&3N@9b|JJ`TZ>EB}Bn4LKEtuC?Az~pwC*ypRo%<sk+ai}q#+ccry
zvk86;Xo3a7BG%kH?dlULtNKPt5c%dB`6iQWqu{Z-MWncEQ9s*^zGoLc9vLYX^CBgk
z8NDH2m;tj_ElD$|MLp{!hP{-M<)n~-77D53Ir6Ksf|+}^lG$4=UexG5ndK`%<eTdj
zzOwqQH}m>D#Kq4=o@SA0x+~;Xib8Y~6jDjP+4jLkR+4Xq@I3vc)`s0i)-t`!TF&&h
z7N>sJvNqRR{^kA3ENf{&wmFhu&5i~#j*X2}a^^FkKp`$uZKdTeTe+HSC;!#iiN+5*
zDdH~cBX>iiTc~6TbvSqFV;YyOlKpE{vVEunxlSc-xUc$)d#0is4{`hGEnmoD{oASK
z?MiaZ2DKQ3&@(>FSK6=hkyeMjWqen6X<g?em649}cXLPi(b7>K{%bF8nhsLl$U&^N
z9OTJ9rL3T~S0hL%TN0GA#7HTPj@irm!S-SsXfL&X_Ohb6y_Ce+$zUHoU(Qu7RNIMS
ziJh$aWH07gDjB(6B`an*%RfP`vhTT<yvp{Hb-z6%B*jT~r>JDzQ>7%%x0BPY6|&=r
zwXE~AmdOEDQu@(Snrya|@a2}G<z*##+pOeNw6zrQ{mq_P$>Ubm61cgYT&=Q{l=fEQ
z6k;u_Co1Ge8++M%&0bOuDaGR%pTlcMu^ZwfOC6kLU!<$F<jkt0hnrMCbdoE3$b=<i
zye_U%ljkPys@$bTm4|rHZ*R_7jwfe7R@>?S+8HAz?b^$LG10P+%)7V;`@bIgNCR21
z`5Rw(LQT<Jc7bgR^q0Sk1Ek?nfMib&l-V{xGICy!bWIBuv+t0ROCzOUL#nhlOOtnZ
zvSmnWj=W{2wfCG{iFuVPzyHdW5uN(UsbzVxsis8yJB*QA|BaQtho(yFBhw{u><r0z
zGD}L$7E2?)zr;9hm4qH!D`&lTN!;;0q91oe`cFM31E<%^;O!@5>GqTI^XDnaI&fA-
zSzi#Fo)_3#aYK@N-jXwB&tz51OL<ZCUS{_HEFSnSMqUjvJ68j-xmt+d*BDN#n&9-n
zrf>*vhAHD(;@&xJC?2+gc2grHml$JhkSR{TG=+6{>htt0aG|FaULWAx+146^*=e_n
z9Wl#F?GbR>9``-|cQ1d^-vO;tRcPOr*NFYN=RfX*=i1KjZ|{s&?BldL=8WXoF3f6o
z$GQ;is0L7HyvrT2qdZ{Rm|DnuPi(&FiLrd`pT5=$ovOT%*~J%EHu*xAd~@$E^$V@k
zaKFrqSA`#J>A9VF$RC$X0`Mw05KFHxOO5B5wI;#1_bM3k>;%mdLJ@x}6oEfOQPMUX
zeWJtBvLqb;t_;WZClTmQzVV5O#^SVSoEgT9Hu6ncbqv<5jDg>Q81{0-;@jz19K08c
zmY#9gH!BVeMe%4yzUktefM?yAm0Xs9O;Zzag?T$G=*w!+EfMwPo0jtv(Pw8O+eMRb
z<?o*OZ&Dh}+|qI8Cp96#8Q5_&1LfqK_1AhM?|T-m*Y|-@P+y$h)EC3sWi$Vz0G}rn
z;LGwsSl+b|#{U$ebwCkHwiIC~bvaoNigCr9xi~9J(M+cdrq9X{T|XE;)#c0<q)wPK
znDd%dn6b1HgR2IkKW9FVyObifls%LinH5NFvi_zrxbX8dbqB-xO*v}5mBWa09C=0-
z+QppYb>%p_oU<Kf2R=^WeeS6AN=Bf9I-Tu#)B&|G#W2%S%pY5Vc#jhJA1TJh1AJa4
z^t<<FW?}PExWA+C{bUKot|-B+iS)mpDn`iy?%r~Wv6OS04F&~RcQGIHj^^XcoP3xK
z%EwdAa&~LyVT5HrTuIA?M?@}~KFPr)YHt>hZ&shnp*AZAE2iZjoP4u}^BZVq<HOCq
zFr_d2HGOJM4jG7{Pk+6#JJe58F>pZ#<WVc^`a1~|8nKu4??h}}Kn~SNgyrJ|?C1Wm
zQC$pf_(kJ;OcadDBeDEo1R^6MF!g;n7HlOS^Y^D}aTtCc2*rd|Az0EK==LZW{c?hk
z(ku|`&e1Ei#Sh`kt+BVHr+bYrd;>UpT<3#7TQFNV%^T;|dm%#03#&7zx!F&~xA4Fd
z?rTf8GUNNYD{3CN;N%r&7@Z`?9CyU&3o3-2<?iU75^7#g%$er^PhCf595G+Gg$jA;
z4p_<Dy=rz<E@LNL%2X9Js0E(x&-}fS?Ckb*K<+1U$0$o2DzwD!L6&IK%@R+@Hvy)W
zFeKmn9Mz7xaBA#^T41811&*I3+srjb#!_?C%`nIAspc@BXpZH>%~8|K9P8$>&+?Qx
zRwh^=@=ptV`rZ!fNAT;Ov_zu_D-7Rjh3Sr**$lSE@H%Vwt+R$3XTkf&n!&KSDfFM1
zAb-6Hru8x5Y|(@sZ*mQc(WJo$ho>9iPwIN-F$XxZxiKQX7$a=1G2C_<WB6xdB-@%G
zG}RdI>1!I}ZGv;;oxpu22w!c2xXHXOJ7|J(FH=0@*FW@)-{+_fW}9;F9<Gb9B3%rw
z*M*Q{Rwn7=#aVr*67{gDLJv14>R~*&X4FVMbmsAEf*u}C*MoK?XE_bJ%<G{xv5PKN
zKGT7E|1arg-XM?0Hi%ct-!f^=Z#iz)2+?aBG2>VR%MvxQ+)xWi%eAmhqcO(&HKvxj
zF*JijUJZ(raPmzF`DRtmNKy8WltuZ>t|QkxB-gxZ6e;x@k<yX5E~cj3XI0S)d^=JO
zG^k}zm0Au>_mUW0XSvOpgs!<lp7U_DQb-(+lXeQZ{KiIRf3^{~8aG)N%M89aZz*iT
zOuZ3qGGH9FHsqQkofOiMY|}hUA(zw&8Oi<1e-3=xRw0Ha3K<q>NS%(M{4z6?V@8G&
zZEPsJ<{3)E+_qv`U@5DOtVCOF%^8P{+^^=|MVH_Ixk6gKww1fVcCvPzoxHnZCmYEV
z#^aRoYb3qnPaVYXlY=}}t7KG#gCtPfyZIp5qKQ&&U2+odw_eiWg12CWj~JIZNu!f(
z<@S41S>DZ57VR^UN(U41U1%)ZZyCwuQEjBkyN&b@C!dk;%0?N=!Z2ewonS2bVJ7l6
z)l_azQHYn0Li)HV#qzwp44iE*`+}*vd1oi1%<N@<h@Ffyv6B}I6|yIr&;3ub;TT)l
zKiyWsW-00UaS$CndaKeM<<HwHxw=s$#gWe9eOJYt76)m!Rw>r}oT&xWCMm3C)c`B8
zdtxby9hNd^1leY&rRdMKlp#wkrN@6(a^#Segq4zYeA-E>NjuqRVJYThpGB}1J<fMb
zPAKG6p{=y${A<7odwH|ZQO0vVRjB1GJHnmhUwZ8i&*bwt;wX>*aFV(jXSt|w6-^yC
zx#{9AnseMGu8pT${mvcQ6)%~;i*p=?EXO?emNRM}c~R>lwcEK<yX_-`tbJwkOJB*P
z4*mEm>U<9RNf&z33%Ui!hOYra9jiRO6exC`f+S-F|6DXeWX0dHvVUI>S$?mVDBom>
z^OsCXD`mgurVN?yl`j7O_K@ih`%BY31#-QlRF;&FmANL9{y*PbAm41BHB&rZ&XN`8
zf69XNS)#RNmTb8_OYEA@mRAM`#3J&T)a2C5f+Hs+_vlGE+w?TM!_SD%?+Y@w?NxC<
zb3;aKcr5!`K9l`*|H<-JAH+ZRvm_dPm(qe?V&~fk8kgrwH`|4>$7!)_DOoBlE-#a$
zq&hicyNdaVtuU&=5V18z*mBz#x1O3}V165fr8D!~m)B&&Im4;9LToE*G)c5Z;ACqo
zxo3xn&-R!VOugY~b_?%tfUOq0Hal>hKAZZ)R?cX{8G8H}XBZxJ#wC3hd_CxbQgKIm
zzB`t#cZXG(2V(bnK#zLJ6ZDX1%%}e&&kKVl`{IuUzHnLXOWgzc=Cv>A3&4u|YMfza
zed%*Q^zIUjFPzo<E(k{Yl^`5!6^#6m!OSKN#&*i@+C+q6dNDJOH;3Zid-S*%g<+d{
zI080?<B(|tW(|l$w+~VH#T=8xWzlHH{nUnI(fIvW47;_NO{*D;TJ822ZrUDQ$u<Fx
z<8UD*9&IngBhD!SuTq)+SCW7jW(pROZ$6N3T*x<LDiSfObrOnGIzZvr6T?rj7h+x-
zW-d=h$-8vC^3Op2a?WtZ^unUwy>ahd7W!{zf57f6oZgp(cZai3aw!WY?F%qzDE&m8
z3XyZV5NXVd`Fm>-x{+^sk`W3gm*BobDd!iZII*(~6Soh>m?E+WXE53$%P}xuFn&BK
z#fzrJ$n8u1a4f~;4DONEl;XpnGOSuv1_TYp`=f)=IKLdrtIDC|=j|N9uB+AMNLb3Q
z(YfVl&By932Lt!Y@McjNg4dCcBB&cym7>K|dfC+_ct&pdcDNWX_H!mPzZkn(6l221
zB3$dk=bTZ3%S9!4!7kG$?t{=mYY?n52VyN5Zo|WTY`H|Hsn5rG^34^_Z>D+VbB@vv
z7P+|?kdTYkUhDxU*SxyJOh4XQ(L1$zHFd=wvZ-^<rca96;@8w|-(aUkabIKxW+J>z
z1}0BT$JDFc5Kz$>br~IS_dybRTPESli9{5ZCn8+M?7rvBa@@<GrF(l^JR1WI`)CB)
zM!`HX5|c_I;Bz?~ANp|K*CZUzCWN6E=YfAU4MqKTu$xlQwqGz5ErM`ve*h|H_``EF
zbBC9x5zW1nW}YwZKV|n5dlSsod&BM*^DN`Mu!g_m7Vnw;6z_o*%a{fA-VNhC*yY{T
z6%TS;(2(tn+r6Do(Z>;4-1|D0IUsL^60bJeqb%0}1Fx#^GFOH2r|f&8Z`<<>`<$9t
z;n;mk<lV8vw+cHv-C~2G^Q=(DnM!h{CC0>BqF0V3mgZYxK&mB1`dA{x*b-V-+99~Q
z9XuXbAftl?&VMt9Cix|4v^g67G{-W|Ze~m|$Gh?7SnXquOQ+jn?4Gu;;@dVR7BKnC
z9QPx6E0%bn!EWxOCb&>)f_0p=xHU6D<!fUkGDmpW_cj<I%&%!?$~;hJ1Ftec$4(}g
zWM+cKRmKqJi*{dSgu9cC@N>Ko?oXo@c!Lo}sf=+T*BGrw8N=kA5uCX{Kf=ecz!2@-
z4Nynz@1OvEZ1>Q|(YgHi`9?Uxw;wOihfT2#&TP=Z)Hgb)CF9(KF7v0T3tpy+8{2fz
z?SL-kMd~3nOApbddN^06hx%T62qNz+<zqeb^^jZ0y;Lec&OsNA*XTgc{1>xWe#xkZ
zzqk`=5UpJe5^K#J6&Yuxsu70nX@o!XG+?i%iAK{j@%ou2Hn-ElOwM$!28vuH-wf_b
zy>M!zym=iiKRElZHHnb%&eU+ei;(`mBV^vM2uc0KES3ima#fpi1L|E*bn=o+&Jc9C
z&#|CR_WcDLIowtuVdNVZvP~?H-Q=5M^3B~>HuNLgNcjsJ$>9BGKWx}rppYiqW3`M@
z$oyn#i{lmY_?Mnk*y~HoP6ILc!+;qi2I7`uAoqJ4$e?rs8JlJxvnCiw;%)=!v&%r*
zy)%%|U2P?DfVpJmScq9<J1N((mf1&bWX>Fgc(=Bdaqnzp&{{ire!))WlW%@@P>NTz
zQZ`IgGPA`&&Mt6Z50ite-)ARFv+bm{wH^DSY-PebFBwApZ2MI_@5fj$qqvPMyksh-
zD@??s+C=KRnn=(RV+m|*EOx_<B>1eM<X9St&KEw<W`<(vU?gUI-|!E{(#YLZ92%L)
zgAW$s+{9eM_u9*?0emiEC%Z=5N)7cf15CMt<83QV?%&o}%YxrlvT?7KyeYPp`VKat
zuV*WTeU#!Rj&k#;N+vB<$@AINNvp^=@k$AJYb)#LD+D(<-wL*pUdt@S@~?JMR&6P}
zA}uAqlcl7#wG^EmmeO~FrL;}8lJOrbrErXeZ0}(q7us5gG2b^n$Xfc$w3a(F74j?G
zR*K#1<o*pB{1?#^7V9MadZ=XjE9$nZ9pqRwXJ18*;uq*7KU29syY3>_wz<lv(Qe|>
z-Cgp0J*1zWr#RPn$sgX{(y+!`B5rug9D5%bSL!2u*Z8o9-AC>k`O1JtzTy<CmcG<E
zwa!<|`_I&LSNVzedOuld#4fOzf%3>CNPJcWNp^vVp<@VlPPDytXTD+A5GkD*N{v{!
z*oIK=!&I#lMT+dT=qa-z^Ep=<C@20Zlizd4OOo$I39XzVuIV$y_QNa*Hkc*IoFy%W
z&5|FBXGxsaY}xEJTY`cPNfk41CV8F}9nbS(HRYlt_qZZ|&blrZ-nXT|_rCm^`dI4D
zKNiQd`BKqozSKv|m)&mjrLoa`8P{sQ+&?f+2AV7oyM2p9({HJq8L~{44_P7qT(1+O
zKUd4R@vSj$m=SJ2HbRmneFXG7_A}F=zf%X*J9MyXmkxsd*1`9EI`}kN7qz=}F=4MR
zlKDCj^Ng<-FLcqri5|k(L$fGUh40+=mnS=+U3X{XaemWwv<vd~xxhJ?GaPDgES9^&
zG|dAKHhEw!vpPPm@Pzw7W*m=T#||@G3&}TESNfvwHDBy(r^cid<`-We+x(}-`?z5E
zkZ)R+1jD>47z=BI5xG4WYxW0Y)|+6Iw-+3ID7fNHukIRlI@}FKPlGT-52xpaJBDH8
zn<nI&3)dnM?ih`g8PUj}7L8*EqG4bcgWJ)}I{7OW=I>(B@VPzW;^Xk^U>ph};&Enc
zJobK%M_oz+J{PgC{10XTu4dO2^)~~$CgQ;9MEp3FgmvG#Baj-96f@@G%uhoHYH!Xy
zOvkW*49x7EiP*QC-$Z9&(y%Q2yD$sO{>nno$}Ak)o5g&QETmu0LUOADOff1%zvG3t
zPfzv59Yv_OEyn9v#h7PE&-joM)I8;kq){0bYL;NvoMMDdEyk8{#rV~n9K;<^+Pflb
zZ(9ueU}hzApY(?M7_(kw*nFuBO9BVu-r>Orel{3WwCFQ8FURW2ayZQ?$E8_ho)%@;
z_qG&=4F)6odKvS2Ik#C`N*!q_n%I)9$XJtoO3>wWF;0+g9QHA{k9;$b-@9&o5f0NQ
zH7ThGS|bbb?B+mpJTMUVyb3T+Re-M_^U?EBJ`%66WBp`4(zfTLPfR}cruM^s!*XGe
zlgn<!99;LyL5GLgcuHTm8;{h^IZ*A%#?-&F@lhumJLsMIgM2f1WnZ+rodNB|8F*-v
z0naJvWb1A)?b#8J4LYDdHBg)1u&;yKsA|tdJlW3-$ko(~SH>ZtV|%#1=g(Y4eYsW?
zVs#=h*CGNpdWK`=g)r<&55voAp;+(2BQ*q%x&X)B1RAe{uw!K))P(`~l<bcW9sDqk
z83zCE_r-LTFAV20<KU+^rg!p2)FLk|f9i=}UY>Z#x#fl%?&#>^jydd3FxldY^qbTU
zzu^w*vlGg{Iig>K3bUH2u${TVMe$0Y#2)tIfZI9__!gnUY<8mT>t~11h1ADwvBbjd
zmS}s*5>vle!ekmXFx6I=N4^Qkv&3x9Yo^X?hb#FO=;&{OC2khzue3lPGRjqL3yk~9
znGL(Ri>WD2;x1}OsX6r3=16(m7S21{V*lE<=(3*~Go@|eP}>$!!`s5}i#a^VKKr}1
z#g%Y&Ii;K8Jm)93*=_LE#{^FAjPde;F*drHVEf57*nHI(bK4o=c~?W^&N09wZ)5yy
z!Z}MfV+8y(Lc7IA_`%uDqf8@w4l|;TimYO3j50D#?Fl0++h>TmYYd=4Z+BFb0YYza
zzf`3E|5$qKu&5LF`x{gwXXt@}8A?j&5D?}Z5R@=5z(P?F6HyVcb?v||K(SEwW5-%+
zcXxM-u3ado2>RUL=lcEezF^UH7i(Ue`vh;ywP0!^3pSo-&P|TyH18vwG+s*i{i&3j
zmXap2jWF)LlAVm@?<yyA#w;*r&*kRqyHa|;!Yp_^*@D-4Sjd*1{N0seK?~7z;yYRJ
zVy5K5Yt7lAr8#pSE9GXc6mG3SMzdNhSyzh$#cw$F{f(Jxf5ZLWZ|n;D1D&28&lKwM
zy|q4<PtoU#+xl!LT^+?Pb+}ac#<P7hgtU_mUfH{RQ4@>C_2Q(@I1X;YH_B#lxb6@K
z8|l&8@LqVQXD}`q2V>9AAh;(4<K5Fn7<ShMepg(vL-?jd_~xuUW%b?ASbQ+E-pjp6
zxF%oTPLj8Egm2mi-#n{v#eFN$WrEzWQ#6l3sTK%Gvp~`e3&hT~z~hdV$mwQ@Tj`b<
z+rv^i`YdrE+Y<eSZycst;lNBQl>Tdl%O9<fxWGX+0~|4_n+78ywa^L|99t_I#ss<3
z=r_R0vIc0;-(9vt-En1{2W)&i(L+2i6*I->E8Ty)hk4<ecw{7dhd@Jjj6cx;4R5-k
zw`f7h{hcvvq7yPls^L1o9%GK$i62{q{81|WN>#yUjV*kB*<jdl8=T!`4OdfZtS`4h
zc0Vh8-C~JmO>E%t*cPp_?T{$?j_EXej81ky)BX-vktp69g*!GIHNaQ#`yP4Y0zP!X
z^R;qwOAs%xXen1mIAh5eXW2w@fxd?uIvjGtHt{Q!7kk2NrZ?&g_eO2C7gD--$sSe%
zT<qtDtp?7R9PEVU_qF&m!4cI%9PzM`7WKAh5%@)mNMW4k1Ds^PU5mA&oX{#sgH7uk
zakZx-`WHLkXS^fS@^iUwgcI^_IAi7y7i2pL_x<!nL?gMA9rHle8xNcg^g@Y`H@a&Z
z;-+NZi&A}XYN{^`2l(M!xIg}J3_v@>Kn#>FA*ZpySZf%9tD{5k<!}i6Y^1AbKq%I%
zkqrLHP+a^RigkAzV^xn(4D1^K-N^uqY#xaJ?FxiyY&goRB%hB+DBnh+d1@5;eA2-<
zAL0=QEVt;;rH33^zXGkLU+JNu87A5#!hS?6)NRuNcU$zrvfDY(>z<4GMTPh@d<KrC
zEX0f53z4#55wyZLo4OStd~Ok{RusW@PZ1_26l3P{V!Z!YjI_uS+#gpWyYwaK`lAF-
z8!g5>qor8WXep*NTME0pr8v=fDIT?6itiN0X}A=1g>8=fSBf+9N>TQ$6t5FjU~b_G
z{L`rnX;aJaW8MGI)qFkn{N9L_p_^obelx;a96?j<ZLFMFi5^vz2)C@ltVUI6A6tc}
z<yDCJ`W<HsexRl52SQAKq91=^X3|fT4Ec%M5!I+yqffiIy6pPSfWfDI*!7AJ9~=1!
z>G?9Y$d6Cj_)AB=<i(xj>)0iM+|wXP`uPGF6&1+d;~Mjx@QtbP&9QTh8St<%r#XhP
zPTMd#3=L!J7hw#{4Hqp%$3DV0+l6mNif%J{myR=qZ(^_O_}ml}Ig8`{NTc~=g)W-;
z8u3i@6Fo;yGUAuy=1>+V+rQ1|*uOa&NscOIRy<djCeU$t0;`!wqcw@LM<uzbZ%J&N
z*n+=mTX3H6jj{B-<^OESQKI3LrHk)Ix{qDMTJzJOWL}%zkqs7g;$f>UvR9GDD0Nqk
zEa^r^=XB;QO6THR>5{$f&fZ&OgQ{Uq-ssbl?j=2WNcg5$_{LrMruVj<3^>`7^F_mH
zRWpElUk%{up}CwLJ%rs44&h44RT(bK<Kvn<E^D37!;3{fG0ms<?>x>IzIiz}kD0<Z
z9`*8g?#~c5I4#{eiacIzn9qQ~0%|%J@W3v)E7=X@y_%ssm@$lPpAX}<nqeGjH=LPy
z;^CPhH!AsdFX5)oFA6yOYyoFp9?GS1=en}KfF@Jq=_Nkm&iOp4kzLSUdD7#W$5SDq
z+t}rC+Q}jGnl^+>Glg#?2Oc^hm&%8O_{nA<rwHF%GR@(1;Tz|d1GwV$0H&Og&vVBB
zzIN?L-*cIaJDSOVZe%j-aZk>uo5dt4K7Zt!#Yz8UvWHH-{hu^;_@^r~FLdKnt8{v$
zrL+0!bRHevohtF(nC?sGmiDPEXwsH*!`d=TcI2Iq%#|lvb8V(<c6?~XRLS<2Oliqa
z9b3>_8diJAJ<`)Gfp<PN<M(p$-0Y9#*1JtvFRCfC#LER?n}G4r{IFH(kM@FnR_XY&
zAc~#hBDqttQw_vhyvsVA?z%Af`D;wAXg3R5hSF$52ra6EX&MvEdlRIy?_3}o*#+{K
zF#de$z09r^wh5P>?@V7F80Eu^>5aH(UPF#w?9K4i;-A^($z|6(INig8kDR?Y;H&Js
z3GYM=@Zi*E&TO*ZiIv4pEZyM5n@62^sM?9`hdHxXqB9LLop?nlKAVSHp53eAn<R~N
zG-zldd~;s-#>Y{^9JxVPR68<B*k<@{N4_8B$jV-hj1X>#ukXNrr5||tRW+TjscD=h
z`%bMKxFE)XtEF$xNHW?%W%gXGCwtuOr87@_Gp2o2oZ+ov*EhDjd`j5FTE#<`rCZC(
zR(7Cl_^hKfug<aJq5{!aQZ3jkOzu;XUwU}XnjhL(bAWha%-dVBjxdb9i8ale%fFLm
z#mmK3w45kUUn@2av|^{bmVB9Q$@6n0>*OJ9GTWRxB(vScRms1eDHv(0<VOc3=Z7hI
z_NbDBZYud$o@S!!3=Wd7)6A4~3NUB3Xfyc{=8}yxXI_RmpAR)>oh7ne^VFOpf12}8
zk&?%^DY-nf2K_T?@V=%7fue~Yy;h4wa;Lf!@Eb)|zoD)F8?(p$MoqgvSftV8l<|5z
zeqN92YCR@f2<HgjToJYz)K+rfPh&B%HWt(C#KG1w4tk>D9I_MLMjeN0Z5--G#bN)<
zI9xCYMuRUwC|%tMFT@iw>Y^)tUzR-6RaaC=hG~koD{eJ(m8~gP%#w^wZiy=vy>`W+
zSFYGDk5ZnrFOu#4BmZ325~tf)U{11ybnaTfYnBD-r&^+-izPM+->enBsUv(7EPOLf
z_$Ea7=HnzQG+bx}ul-i|bz2Qxo*mA<w8o1KO4zz7v3s90ZuNJCm*lb%ez{?|a|7H|
zHGpXyx$|VWA%3hIKHYai&rcpWTjPoK8ZW%9=Y^H>d6cbdfc^4)qei>IO|nPfZJaSJ
z(FywosZlf19;-IlA?ChHa#kvAue8OUb++Q^w!z9n@+`DQlS(VhiL%6mPnNJ5t&rDG
z3%Q4wBmbi<-uc=i`7e81)d-Ktb1}{Vs-F#{ch3#qjGdABRtq0vC-m#?jK$xaQ7!sP
zRzqhj2yw>J>&|#o<&2@?yLP(iio~%EVBXvd*C&YXB%bLBa=ThuD)+1%T8!AIK_874
zZsiUr_~d|-zK+QF&k;G3w3u{Gi{~%2_<lhP%h6gGKG$GfeGMjM3)5_KfU=$g7GF?9
zFVF#%Yqcml>x6c9ToBSrGUoZt_!Z)UA~QFPw{nNgZBJZy=!FZ{g|P}7;%fsRXk&eG
zNcxBVd+vwrss8xTLGtaoKpZ?GIjg?GIQTRezY;<aa6|HT6GAZgT?jf#r=US@5LP?;
zBXo{Gdf5pVtqs6N|3G~BUmzMF9Dmw{W2m9@88&T#e{MxWJyVB;YKe0xfF9;LOpDiH
zz<nJ|FA+nJ#$m~m1S}eyf<3;{S6P*Xg%fh%*d`afXBWc0?F=Mo79z6$LY$kr2wjA4
zsyY=Rd$wpf!Z)Tzi%=3>jH}~|F>OsTdYuw2=f7ess#k(5wk4S1Sb`o~OR)Z42{QDT
zBD1Cxn$)HEZMYN<|C)(&6K5iE(@gX_Sc?8}rLu=!io@MkVEl{~I6ZhJcxffN>?y;b
z!ZPTMDTB+TGOS&@3UkXh;>EQ)l99WITbf7skoE`{iXUNz*<%z8e1aoK$}!<oxp>{m
z@$hjuR(&eR=Rf5THyB<F|B2I))u{NT&vCcwaJOY$dV1C6^Otow@IN1RkUo#q(f+JZ
z1TgkN089G^(s+hHzes*O!zz$<ng;UsjX+w84`=klP}$FtJC%iG!jr@JRx-E63&Yss
zf!wMfTqC!sI^%WBD%J7wP94_@-+cO}6Tby`?LV-quT+rA>rrl-Xt_B>Gc7%a-PXr2
zu5T<$TEy}C_&BC)YtDR?WWt4SE}A6Jxj2Ef{)udQB#{Nnk|cYU#I_4tFs65Fb{Q)^
z)YX!+x!Ia#a)U~5kt{xpHf(K~%H(UQ3^MG<@6$T*g}Mt1Lepe3u`3TQ>BfU@>GYYC
z&RZALdGJ$rUSH9JH;hHM5x!X=d{gxQ_$FNV=JEgOHy3*{^=41*@|F9|)d9R>l*4%G
zh_N{`gt5Xmw?w0v{w|N7UGll5X+Dn_<nzp?JZ>q@<80xZ0O6Y^zlSj8wq&nX4q;D&
zJlY6{^l>X-hxX$86%Wqp>H=Pn{+yWl!?<tMFb*po#-v5VSn4vI(N;tGZg~MG6&G;a
ze#vT`DBzeAL%H&RFwNQmo|-C~-F?JY+$NvS_W9f=xo^Fed0glwT|J+N(0Tt5c9<$$
zwIP>72Icbd{#<&S3}#)6fqW==s5IdlkE#J&@MZvy-;?{);Q@T_-j55Vi|@+rO#VKU
z$-?WItY4i;8^fNQGcA*k2V_#cH-lS0bY$#}R9-uj%99SMY*;sywck^?`C|&7y-A_l
zJL&DKkRD>wRJl{O<tq2KY_Kz#LE~CcZ*B`dZ{C8}Yg@|Bka(vuTXN#)B<|dq$e8;H
zY*{yflRL=v!ksv_?<`*8_@;DQ8N<nTF)W=L%?J0%4oWh}RW=yxb^KTp$%_x0P;)AR
z-Hyo4z37W^e;TuGVq@-K9m<`0q2kF3;ej>cxp^MM=h8`|?k>Gg>jSvE%Ac)cWw%87
zZcZQZ<-51yQ8V&kUB^b;?bVPB_GWYO(u|Dv<hM2+qWv`B{Wv#vG;^ha%7sf)oM|gJ
zsh$&@xG2wwvE!Y%ah#Txx3!EP?!*~Kw8Tv<*X`AEyX;cl*eq-#e3LJHbJI`5*}^wU
zCk-bF-;5K#u!(qWw)b=7#4JZD20Cz$qXYf)MZ0;d=Kk|)p1Yu?=Pxx=RSvwAre<Ba
zIc4p%XZ~P&4!k7&c|GiSa<z(+yQ)}er{X+en=#S@e4?c-Z*8|_ccm?Boou+ctu@;&
zvSRO!mON-9d8LNtY-MgO`F(31Y+_BV@XOXhOLmO4Wc&<EKL2XTHU3sqcD3SC;hZ1I
zRy?G&;+!j%T#ze1;S(0Zed3vEYc86w=*<cxhn-dM&I$$J{!}o;Mz$?&MT1_YWQ1ri
z`%{#>Vy9$-FA8qEr(nQy1^;I(%tIxQ_EpO5UASn2l1ut2+0Iu<(;-SmKdXl0%wPE3
zum*a2YtSjU7JZi1V)gx6#Ga^?F1uPZ*<XXXuZ4YP*Wz;9-*~F`2lJ=>LB}V5FhxAb
zVr{@I;hRIU8|2=y8FpV3eP1sQ&n)8b)HV(ag>MXmZ^jDWv=5VKN*w0C55oA@K~ODg
zg!Dl!m@ZsUNBAatg|l?LI>EPCjdwZrNa$w|T}OLtbhk%?wpLhC&kA*aSz_01OX&Jr
z!rIFce){sh@QpO>qgP`K)J(TPWr`(U3*T%Oz9|*HQ48OwdRvO`#R_#MSYeRx%>;S7
zb+i>cuc_hC$qpN~TBBs65{hvOv~xB?`4Tf&Pd9+=1p_=vtdB>>>SI$oL&S74MV6@o
zmdlkG=5B|rC3d)^^uj*5w^WFqd8TBKj_(vdv$g0@yIpYA#2MkoG%)+_Ae)}{aG7q0
zLzh)3Z>fS(?oP^$wn%+v1HJ7wh#hVX+lN+oQ)7Y8P716EQ(&jR0?zirIlIhJ-r5}7
zZ`$GGPc;@9I-oYp0paZ(pdRChrAIY5c}atLceK){=Y&6&&UpXI3FcRw@aBONsy{hl
zVWktkJ#~`qKPSwc;3V3C6FygJF{`;2!-Q!vn`-bK8tk&r;D?(AP3t;fRc8k@l%J#d
zD;zO9LkokATKMmiXN?wf?6i0^O9SKMj>vX(gxh*GmN!?!sj(U-#;Z~4;(%HHj#xNc
zgN>s7bm%R2tp(0lQzgD-@k(8`@xb><UUIMT#(K$%QThyRX8B<2MqfO7?2Fs;{cuS#
zbZwUh;A2%F_BWS}cKsmnwglnwAb(`7_l4dRUz|SXi^DPESH9<mFGKv{`qLi;g#mc=
zI{=hz_l$$VINz}`rikBbOdHvWxZVVd8b{&44ISF=(&5LgC^$Ro&}z93P27O#gW}L{
zNCJvVQ{Wlj0lEcQm}#AZsDfOCj4njKpcyFsumH`+FTjVC`RFQqGe=v5Hz`HfwYdnj
z?!`EmUW|1wisAaQ8298kSW%4lZ^cMam0+9j&1AO{_+2Q$_VcAsi#}5#Y||~e5Vt26
zLhnc+e3#CIwZ&qbu2=%qg;E6lw;a=FufP!Dn+@kyVwXu7rZz4^rygafSGo$C-5XJI
z@eY<hxra`-9^zGphrou1sDJVy`rmkjPLYpMv*`(DpDf4COXX<#yd0<AmgAb<3pfhj
z)EV&tlLD*p<%je^%&5avck1w<@J;otx?F$8hldmWs6SG2u)XE$)&xH;9O%dXx&Evx
z`&6xq1Gq<e&Q5s+3EPD5t8Dd{#x`d6{f(*LDvT=Oo2J4yUABpT^H~@fBRWpsaIP;4
z=YNBAlEc+;ez4RmECFkGfa@jSQu<uS$JLQMS{uo2qVssn&@ozg=wv_Ac3LzuFGth!
zyLgbLn|6&Np6=H1yeEEAJHJ@|6CEox{-X8#PT*X(M5eY+Wcb`f=3h-@e5)jW`;{ae
z!!2oQ)QWFUx8hpy-sm4s=Gcmk{P8i3ejmDW<zmqoeABsSUOKZbrE@|>ckT)9!9lNk
z(B8HuZwcQ_7VW0h7V+Wi6%Wpdo?Ik+<C2@riFF3Bvc~{=3FFx2<#3eoK&Jd2#NorG
zL*|ieH~h%se%U*)-;>A98}j%`be-}!!Ws6ta=**vzCF1#OB=#}r7v*&)_mSk74Wg-
zo{s-pz{x*k+v@vJUe_B&H}x=vv>3*tj|+G%PjcRW70`8N0r$-otw%bOhlpn5bWr*P
zcNcKUngX_1mCwbc`OH`#n-Z4!G<Yp}sEv7?E4ir3PogtPNAY;&5ZS_#UgGRrW_%w^
zb?qRwQw-$u>KtwmzDW>ms=x5fsfz<-XQ3axHuvPCvP`a@m&tWw#Fw-%lmGjeNzM06
z$r@*JhdNXI&7H)%+JVb2rt(NL*~HYQ@<yFhZWX?%`jEof*Rp&1UkY8Xr|`r56!|r1
z&k2&58nGamRjrcw>7OKO+9y#bzADc)N#b8k<mWw!Y#5SA-M9pf-5Sre(eXUKFiu``
zV|jg<Y%kPCbKCG}*%=d7jw0(7fL-Rw4RnEY5lCnD;({hrW=GI;a5xtgipF@TF^3vT
z=S}xedhd}wZsQPsZWGLrQ-k<h@`7!@1hAq>0G~_|?dFOf!>#=IxP>qK4fo;ql14lu
zy?afLc+>Kt7n5&$a+~=0_uq79^z;Tir*Y%uk1m}3!<ktj(lfKxiSwhJc;T3qJ@d3|
zI9ym}wpO~Fw0y8u%g#Hrys|<o`>9%1mTBnOLN>RBZzc)fG!niE5WcaJyVTTr8ZMsW
z$T8`X4gc=I=i<BDB0Jo_e@JieBQ*~llaAn1YW}rCwh+2Vr(U3%cx6wkRrb=EWzS}N
z>{#AOe7&<(Y}ZD`OZ8R!zSCCvqa=^{%7)etY&f>R4HwR{pl=81l8Lh5{wj0E{gAu1
zxw&k22$xhSIAOJdQ->=UYA!ja`R0twGv~mImQ?6l(X*j+@`YNlgjV#iv|`vPOWw|r
zt?nBZe4}T<uOa4qJzB|LYvspirJ%Wkg1%V_POOq{;a3Xw`A@-&rwYEjBAYZ@6#RKz
z!F`7nv^cMzf2D%i0ZOh-Q*x_xfpMggvxI+6>5Jymwi=das}bGh7p{H$g$cuJ(DPRf
z4v2<Ry0I3cvup8WY7HFYf58&JWH<2_3Ug{?7rYj0Ce|YEMlC9S*1{|V$V+a9zgss$
z?+dXgIv<O{XJhg5N-Q!j#bT)NP1(0tj58A*Com4{#>K(>ZV+-L7vnRr5j=l5quF6+
zoL=dS|DAKf(kp7XjJ3x+OFO*$sluaAD$Kd2LiZ9CroXerj(xVW+ii&ldo0ndfhC46
z`=D#<^+A`sN*_bR<VF=@0q=1ZIBRVIl{{bNNmE&1hRzD*W36N_$_fL=2;+>f!tUW#
z(%~jvm+n^5&t`=andVr~Rsol<rtq3+faBv0VE4oTGh^yw>EZfd8$*~nnWA+YGlULN
zVD3FDgeBXcm8A-cF1f&Yfb<3qcg0WX!n1DVf*)nh*f&&*72;3%>x%>a+G>xmTsy2e
zpu*y&DvbMJi=S(4ak1P6zs3K&&BYqKZdu?&7c(@yVTLJHW=QB}h9SMp(6`hKJL)NM
z?vE17$B9qc-v*)n4p`aB0hc>DqSjs`nQ-yu{Lo<ZORaSLh!1$ElXNONAzt1de<t}S
z;hhU*PFT6r2@S+kJ>As_c9%qhN!LQpNQ)tk8u-0&M0B_sb5Gb~jkOv<UiO%kE6lRa
z0mG*{B0{+5tdka1W?JkzrNM$|N9>UM-onqqGg)eUG*&}hM~#^+)$n+rM)UfP__)Xs
zcb020Ia++mqAji6;R5w{SDe}6j<n+*_|-$Q^=8ubbJ`nbo4m1jS3}H^oY#pieu$st
z57Q|DDC#epUONMk_%aZ|^#d^VyDv;6Z?$8zFQWJOBGK0m0}lCNdOLr-c;b&aodfVe
zd^?@~4MNA*VA)e`j2jojWM?b_r%IaOkwFw1trZRDQzQyYqM&*og}%LY2)>|0<s%)|
z>LrM_mVySw-SDnS7Iw<^_^vm@Fea-|^vD^&o*5{;I3LgN%|p9_c_?=%!lyPxq7N0p
z!mSvsg>SYC-*^k#Je22_@J%D(n?J%gPla!;xt74?SqY-HmEzitQlt-_j)3q&jQFb%
zp@#}#y<{dHPb@<Dx+VB@c8Peq<og7yK%ZtS@$b=<DAX@QL`WG%jbDXsKh`06&>d{Q
zd`Grt@4+?YAzuGKzBzN^A<~3zb~ky9nACEVpD4!=;hT%XH^IU;lU-gQJM9Hpmi<7y
zXtMg%`t)5^hhH8@|IVDcZ06mFFOM}~iGwd^@9|~P>xK-B@nf3|>7PI5#~;$Onm;&z
z5s|V7xH5=a!a})aY$)%X45ia#*`l%zWB1rF?iRlB`X`LJm&5q*Ll~Q>!a1c=I2#v-
z^YO88_V*H>XI=y|(jz%+UL^PIj^vt4k>W3nWQ1Q7|8&yvPG22Q6zEvDAe!rhaTY$0
zX57RWb}o+Li>)#AZQGP@OPflTT6zI*$xig!c>3riFw`c2N#f6Okk`ch5u)pu%buri
zOJ1nkiaU<BVot8?Th%1<B--%((lnk}+m-vpd!y+sxv8D$Ob+YL^<%p8FWGCz)nzb7
zG@2{%S={z2i<L`z@m#mw^zrY*@Oskql9o-sZ`oWgo)pEV0n~gPK(hlme3dhZOEL!2
zRCvbumS{LD#oN0pkMGt?HpwZE1$A=S!7^8}Ke-&cAeTjQ*DBS|W84PWyb`_{8zUOd
zp`q+}WGK&73}xQ0p>(=eKwdB4!*K=DUsk{>$yrSvUcmIp;;|_#;Nh|YmdV~#zkdpN
zM{-y<g>N<r-%J#~*>9fDb}!{tCjb4;!Z-DWZ(@aSZj|TpcTp~fi|*8Z!(cwt45q=p
zft(<Gv+HFJ3(x1U#50GQu32<F-;+NIGx&6R24{-zdRbm3do9T1h4M@~3gav}oXLWW
zOzA4<#6|8M*!w~%E1IYBxA2Wl_(rudg%9VaaQ~DPJ|C6B4ck-Ldr%6Ou5M5J%67~y
zN#^;)WZvkIL~~p5+T2SNZ6T4HItky{CQ@}wyv3aps9PG(gb3-iE@;N>-D3IvUNld4
zj%KvzHrrcJUVlM6(b2b#ju*|NSYjSYTT|&LFpJ<4hj7-)ZEDQ0#&nk7vojn*d2f(x
zW*!V?+Mgilk`Cg=#6YG^3}F6we@?XZ=k;_yZdmEdmUn#^R^NwX8#dxm*>CIH$(!kY
zyr?(ClOH9kc)hPXrv*wE-fLIxKP?+@*PQuQJjM6>Ix+C6mhUHuK9j4ZML*GQMr!FW
zRm=Tzv^*55rTa83$N$!FeUV0PS{f!sX~Zk0;U^Cb-)S{$t<-Sg1IdJkIC8~J2Npkf
z;6{4~eyiue6`$1nbXv`&C2H<SS2Ikf=J8N9t<2Tzb;F*iGwr$Tn;qR(+wn$yJKi0q
z;!4SG+y1cS^cAAtw6K+qZX5n>Zo_GdgheEyv`+jj-@@gd<Zi*C-Q=e9TFIfJsXTwH
zVAL`N2X#`giL-)9^7gBnf~jHV+~{u3?Lp@B-(*g=Y9lW1Z6tk_hLV9WWO|w*XAU;x
zr8kCb7-z%>(WX3@V#=03%;<Z}jK_zXQO{Gse|%+&#z(=wcL+btS4h8tf;*!X^pO8f
z=TZf|Y85OHuc4x`lJDv&sTC%={rxAN$k%GkS69Q`>lX(7`wJ_=YjFHb4H6P+p<Gyt
zuR*o2c<>9}bAIBOXxFPNe!_KTH3r%J!rxPVA>_(0{9Ex0W9mgAPWYz&xmfrK*Hp_B
zb|zL{A7jx~c&EXQSS%3F@ixOa#7D>B?x!HAuLmLGVj#w5`e61GPwYG7EWISoq7Mn<
z$bG|Xls(>9*r8#yO70^nm_1d=Hn_asO@%|RZDF_F7A?P7qR;scx|x|DbS?h8(G4E(
zRyU>Xd!1AF3f)ij4_(B-pSm^&e(L-Us&(61RO^anRO{ZJsMdMbSfcqbD|{EWIXKh`
z|BBCLqWF%VsI0JPrYWuuGec>lDXQk1VE<qPT)$}mrGI_w*i#>$+8W}qsVPEcnqgF_
z0$m3yG0?;u3oqNE?iM>#jgvf4Z}GYOalzhNXY85d4A&4Xax*okb3}=X9!mJUwZN6h
z7FcrF8bb|i@cNH6%+6Y2QlS|-^fSTAp(e2VW`d4KOmW)Q3^(1)urI_6zyC7BtC<Sa
zTcCuwgC&x0%AGA<Zdy7O0(+=&V~!&_hKc5qp+SLoq(jwC5E#QNGr14hJAsv2OjJ6-
zMSL^4qn+T_S@KXS@dBULqG=Z`GV5yb_KPFJ);J<<h&|q4w8N`zwiq+Y4*J45hBF-y
zSm1~mmKvP?TZ79b8oXcch$a;dxYNu5LpG?<Pou`VL-xo$Vvi=mIZdXjG32cSPET=!
z!b9|-Z(4M6bw<M>E~wn&ilJ@|P`8UauEu&`eLL}MPm>#;<m=C6NIpK%2dzi=z~!(n
z6zBYKJzYFF%LB1^jId4vKeYPK7y22#u<GZF<W;^{WaWpiWqxpJB7C#YA3bHaYh%3t
z<V)9J@#7$DuLy<LqQ+ui4MX<T2zXdV;`6geL??yI?t3`?jS}t|5`kM2bVwVh!_oK6
zuw_Gv{Ju^{{V%;xK4<{S`b~#L=R(xhn}NH!8F<`m2KG#yfz6|5!*1<N`8*b)5sKjB
zQ-puM7r`mA7~SPg)o_0?Tv{$cg<UDmdzIoqR4E>IEJgXqQVf_}ia&j);%?@2T-Gnd
z*nx#u-DW2KJGc;v!9_Tqxdbf_E`er4sr+78imne<V9@@R*z;>8VvAPc@R@ZOowy#U
z!Z+=&+`;urcTtsdADNvWVnxq~7<~GX^a9+G+u#%AFa0Ke{=Q*D<u^pte8cZLm3UxW
ziEQ!f3_YmF^t$>Sbx)t0v&C1JXh644DjL4mGTgw83bzK5J8r<dmknv*Aw3?V;rK20
z<D;woT(CrPRmHNaGb@NzpM!YdTM)ZBNMDFXHV7kwnXtYwGc&_@O8Cb6N*EJ`Z#LP4
z(;+y5Yhoh!=0X!L365k&t4P-G8_EC8iR5_Un|$G$J;FDwg>Oa*-;5N#Ss;A#J3pFI
zMk?$RO&j|dwjLBCo1!ti6Wf$}i<@%k{ibrWkLU6?@p5;F=P2t0b`ZWvS(ZrGkR;j|
zw4nNS3tBt3<N?Xy7#v7uuOG>5s%s<Js5Y$Q+?G3n+p^{2blxdXXS1m8+%~a0V+QwN
zTv`VGK4$Q3LQi^J%%bnyUhI}CTLW$T(B(oOUYXyQ2M%PjNo6)KO%%r2D_wW{bLe*@
zhYhaf@MF>-J`Wkf#GbNOX)%PCa&j5upUdvQ1~Whyr>0RZ<MVU5Uv5;#s)lgc%6$Ii
zTfkeC+_vzIi||eMhoQ7SFMi=$1?;n@fCER1FQ;7roBbcdbT43JFUg>1iRQCJ{+)^9
z)e-IH-+bA?l3trU**jmBkx#qtd7Sk;Pxci=uNf(OQf-GY%Y6vj{m$i@_i1cZrz`8%
z$>u1#!Q2!*h+ntn@Y8@ChBeEP4#X_+2KVIYJw5qN_-48Ajj`~}wt*S4>zT=q_cHnV
zWF|Wt%w*l^nOxa6li#ij&%8@z@sd>Wi_80KQ}}2}3YBwHcx-YC9Y&^bMV}P@JuHQj
zr?nT|tvy|XlQ}LXnU+RLY;-%3*?-Gk=D<YG4ou{XF9|f0&LerjVxtM7-FZq6Qf4#m
zZ5GSZM$vq~Q|fYK$agy-Hx@9bjgIysr6+rCBm+vDaB^7$J8lYR=i^~~FP%5<eHzng
zTqqyj4&f265XR>P(|V`uCfo{?ew+Xvvk2hOG=KghJ5nP*_{zqFWWxK(-hya12M;!6
z^*z}+`Rv64eJ_@qdD73qga51R&a8)S9K6|;zlvNKv&NYrH>JzQ-HC?6HmBwbw@h{9
ziG>;-^3-xnftLDBv@|r(a@rvc3x#c(&ekwCPs6zO8V(E4@RIONnuGM&R7e)e&XKn-
zI#9jcff1tHOgC}h2)RYQc&TRA8qsfL=gHJx%`1}grb<n{3VV(|Y|p(8_S8PLV_~H1
zA8b_7c!Y{_;h^fh^h`{%<z5e4UcYa}V$oc@R$4N7z6EtL7K{iH4M+H9gm_+NjaAYo
zSjk0?6}&l5!DUHuTT&_*`^b#;rDpu!3NzV@khcp=*=C<H=RPwc!;QFkhauB?88R-#
zkfXB=`D~*hwGl>Ka?^;$VWvzDFy)bDX57=-j4KDqmbY-uwLCNSZfM3O|C(~MyD1O9
zG~q&;^1>Zcj+9$fyh_PHxo7GBQLsbyPfVREtYT0Nj;zL^ifSyI@CzNj{z6b$4K|fm
z!Fx$1N({bB4@ou74*QP!rQeaG{egCyf8dMHPi&tn-o)!av0IquOWi0q3Ck=KrnxB3
zD|s567QQ(ii;TUoxV9%2eM6dJQ+QJh^^$(#7eUyQ69k(RfoLu}OP6Iosr0Bb8Z2~%
zD%Tl_&7D!?>VU)qd&KD3;m$`D_S{h+QTS#_R~5dysZh4n7KJ}7QGWb`ZoCXJc7O9m
z_kPJ|oxS3VZtAozx<dO3ol)UW-Qf1sx<>n|b@m^B=@zuD(d{p()dgJA$D8;%n6a)7
z#zxmgzcyCLq!n@>T4K#S6Y=GmA?t$)E)Fw6^J{hS^DhI;xM+aQN&{GD8=z*U5ir6C
zx-tdUW-HOMMk!oq4wt_zuwcC<Zv0YXaxV?4-)WI7UX_-5a{p;$4%>N3bPP~p^>+n&
zuU5e9nKAmTHpbQE#+XxXjBy=IFvrUT2a-)-zQhEFpBv%7-A3p%-UwmCjc{_jF>;zI
za63tfx2MdJ^GZ~;Tx%@Nmb=zuTk*Q8;n>s>FB^$gv{<~oSz45fW>R;Z7TS|q<UQ0P
zMZCeSVw})P?nKrtobX!MM|n((RiekdF_GKW4M(I8cf^QgcDOA+wuinA{Dm)lzu3UN
zo+Zx6pMp!P#K&Z6h00M@uxo6GIX-syr&f*D!dQDR+T(7zz3j!?<H|vM*<w(mcBUFr
z$2efaJx9cqXmRn86Z!<X;Fxq0Sv+vVSn)?UwU8{B))PwU7mRT6#-nfEn0L1kR=1G-
z(tW<r7zn#){jh$XXfnC7Woj-u&2?YgYUYcEEq$?RoZPaieG%K#4`H=_(pT+|y$AjA
z?X^Ez)eXR``9WAGy3egt**yIyor0hIvGQOLQW^(ip&|sQ9))0$<ji|Cls&TR5$N|e
z3XAOHpy<&7X(PL1*VAdxnNCNqvgugauMjW3%N@yc1|~I}f#a6KI#&y^dqE-0)rE+;
zJr$?_nSym0Q!ud26qF}S!KD4tgPb=BN7E+ZYO_h`KXww{9-oAZp_5@J-GPT)rlQ7g
z8u~Api-vmhq;GNoR!V<hm(wNa9I^z5UoXLl#ge01yaJg|R!BZ?C8i4Bd|M`-n^Wu1
zKWRNC3*USfz9~C;7ZoS(BL4VY>5#mO`bl?iDEthjeLaq=>uNFJh3K>KztLLm%d=nn
z!Q%IS<o{8RZhQ3jrMW&|U(sjX)p8$yZ9^yN6;0^p%5>Qfe&;A1$g3J~q415Hdn1m~
z_|a^#AA25?&A|Bq?6W42H)hBl@+Y}b3E!j%-}Df^X)b(Y@pof(JSJWDl2Ms2d{a~z
z#)FRGTp1X_>Cq8fEPOLb_@;;O&8|L?)Xt7%=DkQ-y@=$fJjwB{j$)OMj;%I>2A{!+
zq2&GaXr@1o<{`Co9nXm2%$G5or)$d1B~2OnAePfznlW)vbGB&|&y1haZCsK-d#6M?
z2PSdzJIPi#x8Po@meLs`TMeID(eG0-x14K3i{EW{N!ykt!Z(vPr*qDmbY5V0zFR2y
zDdC$HEi>r)E`vKiWOCZ7EOwaLi>5<+^IZEr^t;rDgZuU6k@ea9_9~kfhe+?;o*bSR
zzPTnE&VrNDjdLT1N6rl-Y^57VIL57YF1z~VN~UM9Y%~w%4~tx?owM1qZ8r0|4&mhY
zL%3*0KHsB&L#224%FqHHle^PlxkX({7Eg}wjKwm!VV%xr<0JWu?<}m-P5j7d1>D+9
zzJ0!M&j{JeERns4E_w7(=h66{Y)RDgkbG@d_PdtGYrE2TWM&%M<frjh+cZ{p6OVOT
z8h6E}v0HW*rp@Zhj^2G(IH(s3je7CE^cn^V-|Sl2lXYi^ZX<jXKP7|fh5i0_On62-
zIE#gI+8oH_7r9ZH=VprLm&xm@RL(h(LX#aST)jMn_KQ+jXLbr3j+GnLfE3Q@mBPlY
zQiS>1^Q3)yrrRX*Qfw029~IrLIFVaqSFByrMCqkTq}P=MKA)e!yt~q4-9Mfk);FhH
z%Vx~<j%82+*~RAnGI*fGZ0RsgN7czFZulBWPpe4I_Gm&?;|QiChI3jU*<>h|UV}%W
zY|%uvMHWfl&9h+kZz!4Y9NF`p7RcDn0bIAtpUvO;Np?qe!T<8*o+CbV{n?1eM7t?%
zC;d@Fyg5tyY!)u}lq|modu(-QrF=c$GQ*9k0j^B$E*(?@o!PR~iMJnX`M0;0sVy}$
z%66pQE(ewe3*#)%aMy6@-@C2hckvSMh}ZDcbny}A%E#JjxY|p4h~+MI+EByt3gH+x
z2i|_A=H^;857%|z%in6M<>L<f)wCI*=Bp_AHdi&9Sjzjq?5VnIPw(~iob|_!D+=vs
zaz@2YON47A8-A*ZiY6~?c_`PGMO8LbUb14Xhn4)<w&aM$lDiJDV3<5T8;CDBNy)d3
zlx!sam{V6}r*g7_W1A>=@245PkD77Ka5JVGnQ?orDRrAnxNnp2%^f3l^)#Y&nIX+O
z7_w6{LndYzGIFIMe}6Y*{q9C=GTE3Lm8Q&yH05h6(QgvW_(z`9HfDTgYsNM!O&R~)
zguUg*9Oq)nxUW@6+ERt%2dm`9^c}8}pXhV$2XyH_apcxd{M)7)NA_03IPDh>T&%|D
z#Z_23rxF_$RHEPBN=(~Oh5J9h;l|cV43DqEqN`PS(EK~>%DyA$;ddnb{0_@svN!!M
z1`~yA`p6S0&!AJWxP2@Zb*983-A4NDUI*i8aWFP&f}xulgq}YF@j&())$;ljx6c#C
z%U!T<x-*{VIK#KSGj{knW1;8;ZXV*ttyN*@Z50{_-|Xt5LcInmoZMmyRdXu@ZTg^V
z{q~K{vF{t*==q;@!<1iiXD5BpP1IKCu0OBPwM?znB_FER-Fo{=_b{bK7b|@8^`btE
z-ReNQwhpqJ)D`b-J<QdskEh*@@Xr<#Tyii$+ubG@)!77_x79_x|LOws>tRkE15EF4
zfRSem(Eqm~_P#N~U%i!R@=1wb2h6c%kp(<YS>Sf8C9M1{B(H6Ox3A4*TiqNU2IiO~
z+D)qp1umab;DY?v&+izcXrwXLj4_7WHe=}hGDe)G2}U}ZpcE!}x5o&#CmEqI&It27
zj9~J?5aTKh(b>QVdN#sX$BoeHp%MDf6m#p@;8L6&M$|f>&{%`gG2+EJufZ8FVJv82
zHbje5(P;`MN=9s>BP`Rz2NNr|s`{ew?A4;3l@?oS9I^L^BWltevESSVPD;7GwX}l4
zU<>U3Yyl^E_5@m@ro<A1^sR6**9tA9Pssg?3b(JQ5Fy&lvSIePU2cbTBRiC~w3D7<
z(SiD^VUXznzYXHgX`)5H5-0ps;f#JQU6Hj^GW{1D;L3e>RQ>0H|2}(S^&KzVp5~3;
zM;aks`j(^r^@Vqh^g1j2aKP3NWqN*ic2&4$m@jIbeKEyD^qmgA7%LykaP>oxc#MCK
z6OZv$f6Tw`k3q5pX13HHvsVSeT`v$@2Fq<rK7M^;5Jor$BlSu!3d%xop(qqh?}VeR
zqila(3WxV=pwaZsIDT~+)WSD;4%2b{;dJCZn~n~53$g0ILR=QMX&|g~`Dh`|&nN^R
zO_$w?Y0&nYiX{W4KwtP~Q~VT|9h!u#!Z)jhZ+b*eg4@tZ$dHc>@|=uib*4Zuc@B2o
znFHI@xw1(+4}BsRVEcfDs69}E7pY6|_k$(K>#_ph`>a5o@XZsA@J+v!7<97?lcN5{
zn|tf<@8I?5IsOh}-rvDC;hK|^@8Eu-XgZ;1P_gkC>KWdJS#T95i03ZH>^G9%{>Gou
zKd83Qqe%xnb~>%cLwWkLM_otu`>g3Ke#1?6cAS&t%AD4+2b|G>hi6I0aXlXv?)PC(
zr7u5^@Z+NEe$0L7&-p2VY@rHb=-42xhz_B?co?n(gtFs?P@1f6OoQsi>?A&%YT=u^
z-^2J+_~vrgaBf*I+1%&h9M&p=Lsmy{U{DiY+TMgGM7R0UD2fRqqS)(b6fMI+*FE6B
zpFo+>qS-R?)23*46A$Uu+!$I(mOM1PDO25I>Crrng(u?pSDWTElk9NVgLp=aP2jA$
ziL@_HVu!a$e5q<7on9^IzOxl;-IIBHN*fxTZzH~qHjHp-%Lxn9`CxZCx4lj0n%M3<
z`@im7KCK7cLo%dKuqRhO@5w!q4O%`Vi}@R~_*k|Y45XvC?vcLyIX{~jZ?pNo%mExA
ze6vCLW~J~=?x7q$7rwc;V<5Xe7|d_UxxCOUmn*$<Iqm*nmfq^e^OO3sdT%dU&hI5(
z3uj56M;4t#R~gkHk5k^~OKwUsQ6uFJHMoEg<<eu*PngCxpV4=_OZRGb9y~RW{W=X~
zqoN$%${)ZxZ~JpkdOwc7*N1I;^k)5!S@OB|<h`PFo><zIQ-p8++nL6Wv(xA^AdOI^
z(Xm?>KK1Ux)DN9`;Atm%ZtBFOvwi3!x$X}pz1Tu@tC35ym@>a7E4uXLw!4|^oSZ?c
zgbc14n8__yGP&t!COuY*jw5{YIyRF-qchpwB$dmL$maK^6fO|HDOxBynNw3(lAppO
zqTNhtpTcudDfD^Jj@$Ixb8%H`*4}C%+f9iq?U=}Sl8>9MN@V+T(QY;-uwHNigJgHC
z(TnDM?%$j<r!?g#skew&1N!|XTi_0o1uxPu?^_ff=%P4rNF-MmH{r(Zk_kT>E<b-^
z(qAk)n!3jPFe8-7-$JNsB^#PcgZS}xAWPB$>9#*WG)I3Xc=@y6C_gqk>&p=qzO-%c
z!-n$b!S%3of4`FbZWC{w_VD8KNKfjLBop4oomZ0@@Ir(eZ9QG-?dl>PGuc7t>clxE
zTAsWkzTx^BIyZ5o?*j48SUYguNJmzMYWSCQ(_~(DWb-kOER?&GWgiD#uM$n>fNWBZ
zQuEmyxjpSx^Zs=;yF6F((Kj`#f2#Rj-rBuZ(`lia83WZEBR8oe8#Nc!+H>p+*@8N1
z&xm>U^m=Yb#c0`m+N@&ObQNdjs_6RElFOG_Qh%5Y%^un?#?G3hoh>+To9r5RT5xui
zcv~(gm@l3eJA27HIV$Nc_n@yQ6dXESL8X_1<IBycTWcmaDl@rJno`r!lvRZ$T)xVf
zv8RnVQeng|6AU>t(vW(7hO}*C$VLkddG@V5@kR_EYs}GaP3W3v!hjd1!cb;BJ<*gS
zW}1i|XCz-U2pc(@@JxFXe%SLJH!fEptyFTMv#P{H{2e>XzvJDiA9xh>6J!7P6BPl~
za60%C5Ptc6vJ#JHSK`8mN+cClV*T51m{;);*_$iS`}9}*81fCbetg4^UX^J2ZzcYF
zSBX;wRjAX^8vDOSLj5WRPlRb&%aeR07FSY3VAonUj*bMQ-fr0g7#M^hU4!5s8-$0$
z0`a(hAaa6zB?By7dSx!~{mU6I`#Q_lyv|UBN&d&&8LRUh@y5duD&d-z%~a?XuENwk
zwn*AwizDl-u{l<OziZ#=+NHeFE$Z+}_uk}-E^x#b-3|8&UCze}UF&w$x@Bjpb*-QN
z(lt)6(OEB&OY2#E{Lt2sZsR&AV_obqsfQ1v3=p@zKDN3W<DH=i`ih2fE6N0oCf9{g
z+d6m_To+f%>cJ?-00pAuEWdB~|FeblU&biBqr})_=7^Ad`lMYJNa$>VTRs*zX>5UM
zFU-+*n>l(311*(Y*6F_$FyCN;`Ti!DA$m;b0mkrcVT|S7jgj?Qv>aia&Tov-?xQi9
zE;K^lR3pUc8^Lk6A$s31gvT>OT>of@_rf{#4;$g>T_fo>HbqBQGxYbCTatK_O1?Yd
zdA3G2vo!d+Sp$P>8eEdjozeG&M?_yLFSme$z6Hj1Gl!e}zxT0H!T<`K>1~35btbrw
zVu}lLe;e4}0=f<saQ~ltpJNsncgaF}iR3OeM>6YGlD*He!p<ry^c-f5r+cko?k>MJ
z-u5U?v_r}=6%M==5A`fNOh~oIj)o3!mV8y!e;QnG>x3UWr8mdc6_uH?!7*Ptd1Q;+
zSoT4$Z}-Hd5-*fRd!sbP8%0lKzv>@f92afI+SCtMEc~$RmoF;L3d{8Jg~NLv`2X-h
zx-d?+1HKsk#}|RtesCS(j|G4G<87%Q+R4r8M6N%UFY-gy&H#)Pwoz^L$C1Jy#MB01
z+Tvi0k!?`(HbEGtl)j<~vH`Lz4DTla)%8;`#IzH-KbnSa>ghNwe4`S+DHgpZ>3$*N
zjuj%Hv=AEc%p|84;y~Fn=|q}}dp)M&aQ`W|-FylTc}~F%y(yUaelprUoh&!QNjTGI
zl6<a{u-I%e`mLXZ<S}!w^2QvTZ#5Ur{&UgUV;;I3orl&}ityxh2~sYLh9i8lQ1~YN
z&2l__vkc1?EJso6705Zd3D4I2jU@f`*uHu_ZqB-cDTQ~?E_ge}IPXHq%wxE_`W9}^
z{0f^NU(q!F2QGWp;_Z#!SQ!5YABArY57*<~<9b{-Sf8I0%=s+Ong+2d-iUPOu51@Z
zv~***_$RIAHln+}bpCDhVRO-NzDY*J$yxTu*Ld>!A8*#PYshtN8uGF1<?N{mq^IaP
z`oY26G$&Yghr*a8{^D7;!}!<NFs`!?=Xl|pz8k{j@78b*9}vOSJ0o~Cq6w#eZNg<w
zBKb)8X6NWAZafjiCP83_-O~Rj`c0D5iu_kh7TlDH^4HOPJ~)Q`P6+2nZf2Ho9E)4V
zvFf<=^8eM0J)btGw|tHNGB1JSD-&4wLcGSWk{D{;f(?>eFur?h4sO$iq0`#1&!#qP
zqTiMt8{0Bf_{LiD%JoFQY1O+sXRPhcmeYH%mum)(56hsB>}&e?347#aF>6y67hKKa
zr*(a(yV;i}BeOZ?L$>Ua4xo*00P6|k43axlb735B$AK&!H;_kH4Cdb8T<O-&Ww|Ps
zr``JVXjyNDoa)6jfA^C7Oct9D&*F%JEFN~u;t$zqdYqI;Z|i&x2$kK=vH4srJvIxh
zbLjG>8!Ps8W2;!{Th8vvXBBB|elU&s^V2v!Cyf!wY4mEG#v11|4ppXcj8PgxlhPQO
z(UmuTr19{IH0BOUldbwRTJ7i}d7sXV8sCXydUfK{iyb-daYs%p>O`}{z1eJXFaGa*
z76-1#;;N}V*(9|mlO-$FvULV03;VeU3yo^gi7hsCWZRz|m@}~huLX2qws>yNJWFBT
z!4&BoN?~8&n=cDeXe8d77Q#1cgl{?u-*gN};r(mvnb)WtizMe3J+c)yI43gKHc@Uw
ziM;tCf$GBv{CjEwvm6rGbz?klbZt&or{=622gZGpOrE3I?7r%_biPh<IXe0eiejIA
zkxZ=+PqDTM$1#HJ7S8A~VZ0^U;w{l`lp&#X-V{Qw_d#@Q8zlbaK%TJ=6kkvP|BCac
z<4!+LHT9#pg!}HU@sS%)BOZ2XBzqSPd1IV6v;OvC=y6YuxaT3B5_hU98qnyA8y~!K
zW!J|p-0;MiH!7UiSuNdtZ6x<SLo(jSB-5?u$hR2|H2tFH=rl)aT{Nuw$5FClj`T`!
z<gzOcY|_m^yjE&zQq>&XTFsnsYIfYErr#Ac@7$NWllX14qT5V)FYiBAb3~z<mc7+{
z?x|+e`tp6>+B4vaJv}$ubL%jBx}US-Z`rkUU!dahJQbhXSn=FrOR6$0X;H_LX`7`F
zc)U5S|CsW#+<=niC}?&_JT<2j94UUAA4<tGStx1$Q^B756r46d!TXltleu8VV{^^u
zMKj?UQ#SQ8<$t**oV&=F*lI+-_lE4z(U5m-3^~WlkiEkV*>92|n?5vTtlX$HFN|e3
z#+ZN38}OumeMSZv@@=%-wEXIGS>yVgIK4g#V8{!h#`1TBA)QxNVUt4@2KN#D#^gI@
z-uRBglYYS0>L+Sn|G?ulmGIbIiTQKo{w_Co)jr{uLtkNR`wlr4?_nDH5jXdIhR3Hb
zc)F$ncAj4mT=*5cu6@O8y>FOm_YEayuj*>sTSNW876$q%1TVEm#5#K%T5pd58|`uN
zA3NN1GsVv1^%1n_i7qVaiLRaN6WuxMr@B+Ut+4c84URZyabtutv{}w**U}jb6}B;R
zhDW&*w#(~cM3n=y)wY;&%NB=r+oGprs>dsAu<4)@##Er+<hQ!bHUH^iet*^->hVRl
z-a|Y%hF^6)ncs9{E>`P&*8bAn?Np=l9$Krj?D|``%i)hM?C>Amst$VcI-!s4qTx*0
zQXi{s8X@(uF*e4SfZOXxl1Lxr?scGYuZwL-^>9q|8@Cnp@#l~s3iFL{W1KNoHC1An
zr#U{1H^-;d!ZqzJu(_cH4j5YC*AsK}*<_B7a-Rx(u0VK<0?%idU}3E>`t>%3T6oII
z#u$T*jWM#)2-m(FVa9JGEPiT)FBV4VUTTO>74_lV*$@N28=$md9Sq#151mdQuLkI2
z@K}A^*=T~ie@&%NP>E*iEO2Op1*EqUCi)t@i_l=&5$V?vpHE(b72@YvB3@W!?>BRt
z?qH5Tmz8)$B{notU`4D6if5W&kiRL0!3;r#O59(i#HM>@cxtJ{S@Bah{3+S{>6XZP
zYl)YwtWY9ZuPgFvAHTpFv1Qhhqqh~-Qz3P%4X%$>p)W;Ka<G>TQMHVCIAZ-04Z4XZ
zsQ*x5*>V@SNXPU)-5Q{Hh&yJFk}e<V6uQ>a3z7ETC=T*QZv$`q)7MKnWIVBSttZ^2
zH@a%JCz@~bM1gF51s-n%XBQu=sFW=X(E=NbCnvq0AFAY@)zQljZmr~2C3_Fa6aAoU
z?uTI6rSP=$$9~}-+p}`NVgS;M0<dg<0E)weyT=9LU0x78`!z;QTo{`5p!6knKu%OA
z9CPf9CI25wZygnd{(XPL3^TN#fD9?!9a8VJMI{6!1ZfomMZoS3Q0(sR{MhZaJ3&w^
zL~KN`C8d7nyPoHdS%8-dTv-=hv(FB@zO*B?FPTJgMm&Y@QF9iOcJ4xY7r%gxZJSS}
zY4a&*V?O03vY*BwpWH&`(F@yol*<~;TjrZR%r}3w<&yJ~T)OfomliF}qw3R3Xw&Sa
zR2#dDnr|;7{r&lL^+rCGtyw|UPuEhPv+L-`_Vu)U#|FCnY$HupY^EBQEmVGe3w>q2
z39;W!ug@MPdVP%c+&Do`jvXiETgPc-#VI<Hbe+84-sS&4p3%Yue<{1_FO8UAL#APz
z4S3Q@cY@pKGV_gMq7oV%l(~Oa8;QC)2pOn{`=j{z<QH?eRChxE2HtoGvxe_dC!A__
zLKia|)H&MV_*ENxmf69{%?@o(>=3Wyh;IqJ|HmwJvc?0y<X*Ut=LOGq?8D)%w7ti?
zG4_u)3Oe}WXe4*3P2wHHLw@*P=7+N*{b6?2pZ&W5u&D{afY*VT>=*<u_DuZa+rr~y
z0VU3lOKS-^-hga1aQz9te8WED;lT*L5)2#juE^8uj<A^SFnQ1&8P9o(=w~R5tHUs!
zeW-tKhGTtU58gS7z=YEg_|E+}1_vYI7aD~$=9~BfF=#&}7RL|AB6U?inElre9l!L$
z$*lg^&-zWl>U3O}4&Z+10Z3Xr5bdT9#%5^-cbH_Lb?y+<$7f=|i7a#)nGNkf*;s0j
z1DSgcl)`c_Kb|v3e#6n+G#uH7b8xR>IC^dy3gfB_$nRyKn0tH`l^G~JnE?rVdLoYv
zhDGpTxN=W&`*#Bod4O|BY6FovW&n==O~;x|>9{c|ojXSQA=bAq=Kf7X;(uva#LP18
zS1LANNX50~sW_I+ccxyf(Riogt9dG>YNfJoHU*KgR4fTdMg8AYEIpBmznQ6Ut4_h!
zg~?Dqodn}aN!Zjs2{{iEp;DTNxWYvAyOW4oy(Ad1x5g=K5b^>Bq3O>+<jom~cdXwG
zF&lth$CJ3jISHQn%tWUXu;qLoy!Y+{eB-;*!+6X*8V|z*@wm4o9;29V5~g!*YIHob
znQvyq#v{ck9^;?J;Z6JA*boqn%e<wr{a!eZJPC)}*Kj1>4M)nRaF}KA+n2)dWN{e2
zcMruV)lhVeBy9Pgz?l64nPUZa$OsrL3WCA6K#UGxUY-zuu6zBV{>TreKRe@qUS~vf
z@x`twKImWKjkSHfv23R&=4yIk;aCqiKjyA(XE&5Q;+s@AE@j&00(0*2o8|9}4%3`)
z`>Z2IR69V;h_#z8_IQ_WhrUy6QL)$t&(~Sw<3=l}uD8V8H5NF*y~GK}%(;);3=+8+
zw7QsL=TsAnyI~Aldt>%+8(|WAWkTEWM$}IeG;J}#W?vKBe_)L1iF|t+XM~5=d~b>{
zf=8GFTYD<-HBo_)_D0xHZG@NnsA1pDkEcdZ@TNjS8heU6vA$EHKp@|x#$Hr#SFi$!
zG6g1{GKBS$j<`0FADi{DjCZ+1nQywLwnyL}UEc4~h0|0W$j)dYX@w>xF3?7o<J!2-
z-Juhhai(%E`~P}sWHs8DmamPaq>Yy6GW_>ahJ-#^DD%+5`lFf%>Zl3({u;QwP#tHN
zsX=m66<3T@aq5#CE+6C=#IOGtt%~1I_(ru-iihu{IOd>&uG>^lFkXg;b22!#lcUC1
zjzTXv{*9KyP+1jC`f8X^sEVWg>d3jGmQtPTNRNFrVfX4OazX=XHP(_sxsI;St)<oT
zYRPhVEkA#)rO4h@yk%5HJ(gBc+16?*4ERn9+Wn+(w_kL#{1;iI{HBQ~e^cY{-?Y*A
z4}Ea`LmTcCiI|&3V&#n@;kde3eEafLltw)ln;c$>$e@?v{g9WU+nASPf7VM8`0}N=
z?fpvJSy3t;j4u-n*=1tQ+A=Y0Pq~;I`9n1BF{XOvntL6&6OUgiGvD}MHK#QvP3S87
zXD9j@Qt>(iIxyCNCN%5uhLJY?++HQd&#4kmmRE?IpMHr8_$>~7W$k9*U*VEfBQ(a<
zih-M(#KGid;W@5F$YNSWh;Eyhc&ts-_EMsqdaa@=w^bx0wTO%zEyA}Wzx|+96f2ad
z`kxYwu280Jrc&A*u0q4ue=~KXoKlXd5@oAVmb*G7jL@Y2{cT5Yrs$A-t1fRW>XMlQ
z-*n`<WLd65vI9C4cT1ZF9@nNdjauBH#(K>obqbcLlfhjzvQ}fwh}kCXBlA+F8fo29
zqkbpU$lFzw>|5nDy;w$N%OqqnQ<*a2l*ltfiE_^=QS4?V_NXe;nIH*$3(%zD^R#K_
z^mb&btwXOz>d?BUI`mm?KqEa3DD8(H?{IgZwt4O834e{(`RjTVsY7FKw4?BVc9f>7
zO<Am4?dh*Ulaw`S@mLMI^+Su^25OTfXSpt~(V|pWZIbcdw{2c~T2tPhR(f|J`OXfs
zr@R9#*VQBY*?P31j~=;b>(No(mUy;KpCsS7dwHlK?HXW2){l*O<I9x(D>tL5H%w{c
z6EpTunNv^)3#w)<XP1O`C!Sc5+gNK#mOD@vBL^B?X;16=+S8CP+@rYNj!bg-B{NRr
z5j&dD<UoBgo#^Rr_UT-6rgyF`<h_jjTu)tSEbBP`S-6svTxlBbT*-R6lKp!Zdc_{(
zPF}9`Mz~S<b~maucBhyT+yyw(ou;?<q-!dkq~q#Ie(9XK+T%m+!JJ1|_oU=29)zzR
z+@Z)_%BfxHd2ASEZw;s8i3{nl#X@ooT|nkr=F@@r`80|7=EPj~*IXLI{5ggWUmQcd
zgJ#j%h&j}DbPj#>%cVI}a%tD4T*`LHqgUPY_$HM{TQ=s=)O&fPnww9fw&qiZllk=V
zbw2eBolngV=TmE+1@!UkLVDx6h*rBSrqss8)H!rB^<TK1G>7e=StUnkc>Zx}RXRmW
zXP%<2_pVX)wVQPR_Feiq<{7OX{)c8A{7XCL)zE~*7BbOqrKK-hX_|)wf6r>cR$Ch>
zTX?5M#(d+bhZ!f#p!e7uU0G8%er$o(TuZF6w}S6PE3AHLjjc0m@L{_Rrj4}ae3BjR
z7T96UcL$7#cf{jkjyRU;1gp`WSdj09<PTo>)5jZEPI_bgFK^y;<bIs5K6vQli|ABe
zyr1t2|3RJcYIJ7|>k)u+Z2|B&69@<UAO!UfVoz!i4#x^SEE8aP8A)Ei{<FX)TfR?S
zXZ<EJ7zL+;*+<$H5i7bPWq5bEvH!-PAOvQ7yD+E>LuGI70OaTWLyLPro%yEdXaw$?
zaE@wkBo<Ug;#p%PRQmVCh^;-*t|T4T+%j;>Cxd&NGw?lxnIk3xsi_&*Gc*GW#%18}
z><qjZ%6d!mP^?KCin?K}-Ao*cIeNUo<jEOv_7^|Y<1DpR4sUJcV8iGfl&R(5qk0ZX
z`R!O{A!WlM=vkD3nK!w^_kIQ{?_}W4whYV|pMe<GgiQPgV{SETBz*swFlrzyJ`I3L
z%mAFf#rx%<=_vE=&l?SW(WKHB9WSQAdt4fJP#P+f)9~qBD%AMqG@iM|H!>ApJo$CA
zR1~mw^XOj+!fI16Lpc>Y|L@N}reJ4o3f3E>puU;)mla7^H$Dlym{S%p)1<yg#6Z?@
z@-Hy698W}EyCm$>N<wH;65{(NLvP;ztce|f#F}LI1|`8hFA=&@e&1CI*#DC?nG5ky
zUCV4Ui9N*3Fu8&8czGudeZR!vrD{Cx7{p@<b4`vL=fyw7!LTF_NBLfO<w7q6n}lQC
zH0~%~5{}&Q;rQkr4qKgYsD0wi_}yXXs2qkr-k})&DFlmIx7n92aGE=oPX7zy&f6gF
zj0{5M2=;Yd3c#l(e;B*_BQn(wr{;9VW!@1{f8oPfG9Mfp;*HT-UTBx@iLARGh_>fW
z?qP25f8~nn8Lm)z?t(vFE?6_y8KVlFu%f*aY@;1vGnaigr|dDj%nk*uwpeX!i*fEY
z+)HT9++_uaFiW)dut0TOCtMt6jt)!Akbcn=Dt}F|-qQpbGmR0<TykKf5w@__a#q6J
z^TY((mz!X=qY1`eGlpAtBb4wzYu|x0-oXld3sFFi8Rs4|j$Mud4So6kw8RJ_ml|>3
zj}c7x@gS7-o+k?2ou@#%D~7Ct8sbW#Av_;-#N_;r(Cg9>j_L-mp01Ar?6bN0hW8Q%
zwTJ%*4V+vrN3?|;6F$qKKSvJpttz;`Pzt|}Qf$60fn=uy^EOE^<*)=g+?6$sw+t+I
zNTGU5%G>2q;Hnf}Yh-wGO@{MtWN7e{@pB|q+;vgqt_3;f|4)tvew=EQ<GZga9*`6l
zbENoMAjKvP72Zfw!Jt7hblE23OqUF<n*6?PGWZN+h8iWuSUWjV-qq5WIkmKTZ!JA~
zUrQsIb>bh?kjdv7Dv~kZXxCCnaxHa8tEK(7s;FvI6)kY6qCFZ_<kPi^-YHj8KkskU
zX~}o8&;3Djt$$LlRX=IdyPv$l^NUjbf009vU(|BDNaQlx+`nEVE->fl^6T}i=?r^T
zB=%PniH3hgVvY4nu{7zW*!7fIC+L;v;a@6NZFwz{zP=XEoXf<m%rY@@dYO1Sp-i+4
zDHFj@O2vR9rQ)V?sj&R}N*wt9O341a5^A$GXznBp`Z`*J&c%HbTmGvOl^d$Wpyi*%
z4zCK4IkZCL82%R5D*uQH*?)zaVU2h(x>iKy*NKp|^`c=)i`Wp_D#l9N#Ks+MV#%f!
zvAv~5j92iZSBpqq*CGNnTE(v;tzzk^HZksu5>04ip0H{W!`qt0&1cQx{mEu=W^1#E
z-`Ff3?`#&^_B9LXk!JBST1K~v<+P|tPHEM0vUnw@t#{;Pcv?<N6{?i_M@AO{RjA%i
zg?yi@QY(9B`bMcyLY5j8?Ny^G`D)ZoM~&ReRH@H#IZd_XzMaR)v|yex8Rjd|J$EJQ
z=b=QWW0mOYMkTV2R-#rNW%~16nSD|kwCjQfE$_?OarQg!9jrqW9<gr3jPX%Jn?if3
z(arHH^!t#6B6i5AZJZ9BG0>qT&SK%OHl6OPPEohj$?><G`nmAe^HZIo_i0kLz7~z&
zqDjSeTGaik7Uz+)Y4wcuq*l_NzL|C)r)BKJxz&ME|8}6EnR;aQQkM#1bx3zi2U={a
zPe0ch&@^>JTG*(d7S52LTBe{E_qAzBXB{$Krb`#jwx|3(9q3oO9<}|_qXDV<MEo@m
zU8+x>+6FXmumN>z;Z6VxL;AB%L6sWDWc=KO9%$ImvP?Vb|IvXq4ss&b)y_obe6Pyo
zEY*D%+TGwnUyWTUfmue`$CaACJJN4I&SWidBAYob)bzuZ_Hs|opf)#J9p_G3;qEld
z*_|9_dT>6^gSt6+(wZ<&`k?1Y7jJvemvbIuf7*l6`K4Ex2l>mqDaSIHN_R!lg5hc8
zc6tGA+Bcsh(er6?&s@scJD3)|?nw)0M^O6o9;{=9Gk1nl^|aA+`N<r5V3<qO4)Ocx
z<Z<_49@*#QQTe(&-r-`-NzJE}X?)*Wm`~x4^GO-=>Ezw{<dD38?jB!An$C-8N9!UQ
zv1SQ9{kD{rc4nVW+zztbwv!%hKT4CCZwhL!()u+w$k^&O?X54NxPiZDwSEl=-3Dg!
zW-76iVt<$>S~%NqrdbmQ7inP=XAq8TYKL-5T|~6-M$Aex%;K!aA(c+})~gdf@^jE`
zj#da~zPbF^8ncaTaHG%$CD&~+wuc=w>+Essr~~JW9r4QD3CHuD@b{e)9FMsnJ<uIx
z3GT=^>kZW(tmCZlLC#wr=yc*fwPfC?;@v98yR7HPI|KJSV*qEXYH|Yca7Q3^l(G*;
zF9=mhLAc0U4!0@<ih2V22Y`vayWoi>uwWF>^Eyzk-W8TPU2&^_ci#C9f$QB6WGxIu
z(A6-k4h+Z6H{1dEwg<d<zo8)@0_EY{XWWCCCax#qyrZyxU<|4>SyxDlN8zk~Sjsml
zANHw)UhIcMm-}Jp=YAM6a0s-FhQf&RQc{ylRMusp%fC##Zpp-D<zaY{I1FtYvJh0C
z1wZ{9s9WYhEiea7%sMNiIWShvL7PSnETS{f(3Szu2fR^mF#``SWT5Rx2J#m17E@XV
zb~X%#b>v{Q|1t;%iU;Cj7Wd119Do<>Z;t#g9fr;QF;1x;tiGnf^*nEZ&P+pQbQ<mi
zrD36d8iHSOkI!`0Y%)@@D=d{eUsK^}n#!5@RD{)}a9<(wO<fAIDpC+?oD2@&!*^B^
zin5b<1B&@&R1&&>PK52vL>yhmy+GX(@lqoZn~o%6;;uxP-$+Ew+eDn)n}nUKk}#O>
zSsBBVU}2dA8~a2&Ea`(ud~b>v91r*P@o1gK{$Xd<XkNr2CL#)lgLzBpU{4&}*%KQM
z_e9mBDCE75!t(=Bc;Fn3P2+l@=ofpY+Bo;k-@CY;tiRcYW4>xQ`u+%m*}X7WI&)@f
zdMNJOg`#wFSMZO19Mu(gSIV38tlRWA34-IQKn&)4=bvr?@SW?A>Hqnm``6BJ(dXL~
zxW8by54!L4M(tZKTzc$@VHTX3n$G>)*W9qk$PJSgxI$aviUlKF@ZWuB%rtSvz9CNN
zb<h#FD;!{E?tqvW?!1|5hvsdzFu81lfhE@X?~@hY{jg-8ngxHaI^l&{C(P|+j%QuD
z*JhL{?(Q`~#b@46@HNKzEv&(q8^Lmcg1ZP5IP}m2H`t4ldC3^<&a<z^(+J1jac@nV
zg1znv<TB$3W*mbE1xlh77}QIF`#lu+(uH&1o(d$J^KO)?0$z277_q|;141|}Wp9YX
zQbTmP%o~{#Izr2=BVvacpv!Q5-qF%S_6P2zVV_Lz$!d`0s{n7LSjStFaf7AUp(MpX
zW{`*q3F6lCZUOg4{8%HwyL}S;dL+S?9};K;GTZc%;?FKAyf5(27l{gnEmFai8Wm&(
z$Z&s$9BxUh#mtaHHe3#!EIF1iM;((((bbJVKL5K**YoH6B*h{%72M|Dt<(RLqT@#i
z($7jzvrz*7rxI+OAjNyvIx>&1C8e-hvKv%OGnduS$qhA>e7}awKGx9YrW*R!T0?s*
zYRRRcif-?$qPe}Q=tlEbs($d5ZoT?S?_#QGxJ@;kGXF+V72oLCnD4Z>_&X)&{UGfw
zKPWHd2kjp6gH&%7i8z->!n)@pVa{wb_F9o}zh5MR9~X(*S4E<%yh!*r6p8aAOT?tP
zCE`o)E0O&Em8jhIT0H;sTCB1u6Map~*gNw?s09{_F!y3nVN)#TS{IA>!NtOH))Vn=
z<Wo`Bu0$N%SRx)BelE0oF)uwU7wvwQi)xLx;>^&`;zs3XVbi5T5dQ%5_WUhQkN7M8
zS=5L}9<`#wf;!<oqh73wY!Kb}PPOw^vsmQSBKpTQi&6WVg{o4ExZ9yc>_~1Ar+2l8
zYvWr*sZE<GZ)+2}ot0>;O^dkS(k!Zro5fycnKkR1Ma1f6F??II7`(Sx)c=oPzHJtH
zmCa(LT8l{0Z4oz(T10?ti#Wg^^S6}}-O5y^A77Md>_iD2`A>ze{U;}7BQ=UnQ6tBZ
zYNR$wjl$)u;p|qW|F+91t4K!6uJC<qgfdmOv<aKNZ6eX7O~|ra#m4=u;^u#CV$E44
znw_gmu|68~eWnI2Owc4nE#I6pb?DtqEmAdZM{_%B(fJ}3TJ%tvjQc53_Y?`u^^{P>
zI0>zIFQLYKDV<oMLNiy$C}5wQlG*3eGe({LD;nfys!1)ye6zZtPWxD^dbC50zvrqn
zkZ*at2FS_johBVUrAb$UG|BwFI(b!V(hc_O6fgtbzNSaBf(_{U(vB2+o%@XR_+z%y
zqzV&l@=#`7Yp5=5THc;kbI0NjAJ&&{>Cp!leUc8>r>%4JsnZ30iqSHlZ_G$HMt7tY
zOBJ;Fj}ZmjH6g8L-u8;JqeJ@~D9hi86jPij&A^35a2L+ln|!CLaiQNGU1=P5;>=+$
z&(7KQwC#@r^>B2e$tEroxPU#%J>AGN+>K_B;(nbD?(~4O<hNe9lj4{MskgHK-Gu!)
zmpnL2=s}YXd(e>M9%Rb@EU(0aRG59j418$z$Y8SD5=Cbw#M12xvAl;KOPlzyxG{!$
zxCc|emBE}z>q763ccZia!YJo_IOpBQkWb4TQd^fx7s_&JifSHjYUa_W)p^wYS{_CF
z@Lj8aK9%;#CxaXLq!W=(ne6A$osduS`Y)tAR*Ps)uf?>hU<oC4T0t}7S5oxjHMBx`
zCuK$Mrp>#r(4pnm=wk6r`mX+t=KcFY%N#yYX}>1=^0tLWMJS<Dtpo!<s^au>b)4L$
zi6uL=knullm}_yrxsfi4O#e^cHpBA*-WRSj$I|dlP+4k$Nz6ArCR?H2j62JU_+B-~
z7I$yi^0v4gBq!`K`hp{Ze(+8*XG*Fpol#)r%-e+A1^CkqyZX4pV46E;G<dUcqYv83
zeel-Q7r&TqI<N4>id%fMQtyn<F6=v==?A@+e)!ZFfKj~zG3#?6v|WNwdX9T=)C87A
z3EZ6_c$=Jazr#85znNfnmA4ZmU9sjzSIn1nN9<ho-&lq~(;)<9PP|#=6Nb5Qyu+Ez
zw<`7*yP8Mx-ZTGxE$oRQbE1%bEed;&#<E`_4&U3wA=D%eKWciT_ut<9d$u>{mihj(
zGYz}z`e8HQrPf4d;FL!ucRKUlQb;Cj!ZR@>A`@RPWg?sXykbihmNVmww8+6Un;e{O
zm%|K{jZLiUEGW;$`Y7fbzDMC$2E13WmpC^A$2Vm_Yclgqcm}%E3`Sh^V3@`a!j`iG
zF`F~ipBe@rg}Z=vA4ta(?i*ZC*O&9_X;`y0jk7sv*cg%qUuK&>#i{7NEESQg*$9hN
zSm>vs+9egcjak!~lY$>(QeZVB1=V+wn8lLd!QH~@rxTI9l5bGc6LBde36@z&KvN<-
zt|j7iVj==$iP*6_0n553;N{8$1YJ(V<p<0=4-=7nD-mU96M1(s5n^#7?(T`jboMZ5
zUW&zL{yV)nGafXEGgA)nus$6H<2}8wF02<mjgQ2R<_L`V7lE`$*3Xh7ku%``w6t(!
zm~kHbWEk$3hCxdbj>(+y|IRxcYGq+KQy7N2=r9DH4dp)OP?+k6LcN|aQ&Zp*-!jFF
zAe`<LggFNTvCuRSLDK_J_0S*N+xa6Z+>f^`JLAMY?kj%E9ZkwUIOySxfvH}&sN;zk
zT@M)hyJOG<H=HVT#Rnf(<m_^Rh8pio4RXfwQ%+E{I-+woN35LffNmG;5&X>#LyYW@
z6Ksn=18lH+sx=lZwZhs>may1k0iC0r@bZ#5wD~qQt=1I6(iD#pc_(VUF%Ew*!klO$
z%qZa8b{_?XHyUEqdqYTy4UzfW5MEmpShrSz<(zx^-ktAJg6~YsIIUe37{(gTQD&S}
zW}8;#owgta_PHuhYs&hLrh<1Mm}k}*;=7F@O4=L3f_F5}Y~c;foQ`}~?TFQV4bTv-
z&pWUk5R}dNr!Q)_;49;O6cx1bbD+=<5=c);P&!S5*lrRiY<PDvT7t<!f>k3VxR5J>
z-hK&oFn_##Bf+9jDcV-?F69v^hJKMkGD8K{-&AnooeUpt$e?O2M|OZ5eH`Vm=bO{+
zNpd8%@W1cK+D(KM?dM4`owc8HFQgcJUJChADQ3TrU>kpqU6vBGSxI13%lXevbyVO|
zONaPaePnPAN!TYd@BUwkxLd=WVl~vYu7-YU)sRw9HNAUXMLp+M(RceQ8vXDq)z17%
zs{OxG&&6M<Si6d}E>%(I#A;I9tETZCzY#@#qr2JPC}YMqS~ZU!s~!rE8TUl#!F%F<
zxBDW}xKQN&S0w5Ri^MqQ8)xR5aMp2dG2@JAEfN8hMI!NGk@$GGNIZI3B=&tT5@RIA
zVv0+#=)qj$$dBR7HiqWK!l7TW2>M+tZZY3v>z0Tn=9>k_o{QUwFGX!hxj6QtT)a|$
zD`r@K7RL^K77+#&;`8AO5jFRVn0WiQ*y&s&cIDR!>ApG<Kd)Z2<TMETIgKKI&OgyT
zrAb`=(<HJ5vko&$Z`Woqu6wgc&0>AVu|-^7-zrqQvesj)MBDbYikdUcqUXFOv7i0Q
zJ5v6M-**4Rtu4*sIX|)vHVdutW)a4Gqsn~qQl~|@G2dKdE~@3b*DW*Ff%+)ZfZNJ6
z<E=8~oR!kIH*(tcO_j_z3+0}nMjiX9QJjSu<xFI6&n7tqte4U0hf4HtVXLU@*(|;j
zf1XY)V$Y!#p_|hx7R0uRrHM*3-%gp9$FdLlrHquSWfVSIg@RnARPawio6Myo_f(>H
z>@~8|RHkGHB`PXv6~l}q?9Y{uPO*fpkCf8MC>7FxjGl(d$vIn<I&m)Dr%0U!MQCtd
zTZ3*+R;RCj)c7u@N=r3VX;Frpa^!NlV<)G&MRF=kR-<c6)ams=O*(u>oBR*!Qr#Rq
zO4Kx<OAgxPY@kXvtJKKvr3S5drbXS@_wyt|m->%yPeFY<(DU|sWVl_Ax~S^Yo=AQA
zFj}9+?a`+dKlQ0}lmR)$bR;?VS(lV6sBJ87p$;-78}9BIpT&JiW4I&Q&Wz6SR)mau
z4cnHQQ{OqAD2w;FiWS_i6KG55Y|q|q_VS!_q_h29$ey*GhYwsyzYBYihq_U4vNPG~
zaS!rES5h-}W47o0@)K_4HpHE7yLgc86c4hR;X&)Sc~FN39u&*`v#QaP{@Qubxcfd7
zY~oAB`#Mwio&jVU7)T+iL6ll0X!|!%;_EK-V{12ZImMd~N5aY3bqu9^m_xbqa%sf#
zTncZ_CFuaxa8~BgA<k5VRp*gz&>V_pZ}JDbd|DooPb*o|`8Fz_^X&^sS8ow@8L*hn
zJ1nPI<{Rg8D@m@inmn#=r{!KdX?^uBQrWnN7AP0e{r4}Z{9P$EU2ULwvSzxZ+Defd
z+vp@o5ZOTr<!9>98Kemt?wdfdHqLd>!R&e+7@sx9Lk}|?zG{x4pUhF8+lildTcBo*
zB}S@RLz->P{VUd}Ic3W~i)``A-wwx@+QW|d<{0+@CR#Y7`v2#fWE*Gb_IJg@W;aZo
z<&IESFGMc%fhv1CElhl2$@i&StN1>3!xtkzcSf_99|Gt2;dq4~ma`Ak;9~&l9t7ep
z|GrwkBnZbp2jNpNdo1R27cJkb4i6^uT21KY9*jKZn@HY^u6oi9v$}?$PyY}!rt{|F
z^bq`SQW#oN!?Bioj4uw4!i}*}cr=ZBi)*8>XiYTh2+`odMzoFK-rOy**s(Vj7y04G
zzo#p&$HMttEOMu$Vb_5)?$}`V2+hF2XwFLWPIe{pjUDq%Eb~p~rc4~WmWgw-v+($M
z7A`X5+^o)KeIOg-eq>|et8836o(=Q3Ot_Y0z-MO$j^y!<Hs8FK49>vjtPDI3%s}n;
z!Q5v&7^%Gm!I1f;rPn}s)(^l0=9^=y(y{epf9%rki`Iu}P|Qw4a%vh3JUQ>(mWpk+
zQeny78+G>F3~Qf?j_Rq9mZsp!rWEK5OhHkP6yEGghRW?EjC4)HEcOpeP9;K>`Nm~Z
zBF+v?#OgHm2Wux`{kBACCM059RRV06B;cEE0vwn3!5H?^bp06zZR0q+3hRxLXJQdF
zC>D#&V=?$y3~pz{K<RokLPDc?pCJm%(xNcim|s@)g7e5;SlqoARQg8ZUQ`4e`TO!e
zML2ZNg`qqo49NyzIH?f^c|;f{oe0BP=9#na!Z7X+>o&*3;Ltk^HSa@FvO5(2`fzsY
zZwR`7Aq*-Gf*rHXC2PJ(v7cB;1Y-T60CZ9dKw6SNesA)_xM$ok$sC;>;EQFM>?vOB
z&EGFCSd@FB=!ORdJ#fdo-)`vR=7yKEUGeBGb4`c~uIzEfclMs;N3jQOy(9d~_!ev8
zfSC!b-7K=hz6-Wce6)d~+y*|}6*a@p3XT7Fn~Jl5BE1v*N10<A_xIWF=RM3~6Wo=V
z;8v(Hj;%F99%r{tPE+7lM+JP#46){_A(Ad}x9}6@n<WawEmok*K)y3EzetS~_~FAo
z8rEulGw<YuD){SDV27WA^JEJAbW)&-`)#(WbH~ja{uuKNc`L&ZdgBcc>#2p~2AUYl
z8Kgns?a}Pq9x11Ep(N48s9k(R3YTH?V->jfRl%xH5`>&&y@tEYThk;M;x2)y0e70W
zlOWVvf|#xn^h}nZ9kWj`^GEGH37TD{I6Fs*md#SkyDr7J(<)fY+RbwI(&(I!;m9|>
zS+&RzRw2XOTQb-lP(kBQ{v1Y9tn}s^)kG=gY?Gqoq!fcTNzuE#6iNJX=k`>_an5#r
ziEE*v7j@K4wT|{N+t{A3A*p)}E#CW=w!HmGYrFs8e)#VcAM}kXcU98^t7`govx??t
zR#CpRioDl<r6Q-Vr1`XxUY@C>)juk^SM4hq7k;G}+bZ6vs-pYMLN*(#2)nDuZhsXi
zb-yP9r`!_(_K!rE(<2d^`am3NDiG&C6^M%^MMCvWk=Xl=?^ZvG#Q$pecJ;PMWH8&j
zV%?`R^Np1GX0lSTnCet4e3@-b`0?ASShQ#si<I8QqVRUH*!%0LxHqsw{5)SG`WZeG
z2XDU;_b-=<Yp+-XGJYfemAn;6H{Xko>pzI~78PRcq%UH*SEaBj`6F(})ryyg>crOk
zdLbXvAm+?$6iqY!iN8rr{53X-#}>`vhIO-uHf<LDJetL|)Mjy>^_`}d%|a=Yxy7kT
zOw#-(F1I#_p`99p(d2ruwzW=--d!io^{x}`I@O7lBkM(cVY5hj-z@H_wumXrH-3Ec
z8hwJ9gzs1JGh2kluNLv)Lz@UxSK@BjHgUa!l>V_+vWzvGp#EyqA%S07s*!I$RjOGj
zr>jeN8=}7wJt%AvBC|>Sa%mEmzBh}=tQOIsQ>$3>s8#H)Zxct{+Qj51EuvhtRg^7m
z5i?6$gr-WXn9aAW`Qb`5N2WyDtJ*}5^fqBMt5xJXw~7;gl&N!!gnZ9&PCi0P-fAi&
z|Dr<H?_?C&Ag43dYLsEhd9JnU)V4sKY&xmalDn$pk*`Yc=g28S!CA8pGMcR+r!|u~
zTi>Kgg=XsHeUa~E&04gyRz?#nne$SW$m*$tR$Y<PIo5vO?9iZL%d}|6nRYZ;=+Zad
z%orHffzmjK^@9C7$10hL-1W({pFZtatWW;;c(;NzplQJ!De*h^@%-l33l%i1(U|Wu
zrWBsXJ6P=H8PA=D=SQ$loqL#fJ~pGU@J@8v(}I3AS<#7KHgu`dj@D*7(35CK>JiNQ
zowr<QLZA!vIqXPDaH4-^&Llm_eaS(tH0zfuEwpyy+Y;ZdSl4OM^`K<V`;D0DLBHmC
z(C@7tr1y<|IA3`q!q|(VPxz3Ep)YNC&->*IJJZMKohj>XAhmxkNaI}>I;rSN2B*4F
zm*w5b>|i)Q?-)&qiDQ^sa_PqDTsp2lo|beNPpcdzP@wHZ`r3I4Z?w-L&!4k-LoAP)
zD4!lD<Wup8e6lxLNGCcjqVmYabjxZvy_2t?4WCxhPwUmx|MYgc<FJ#;t9H@w8++*1
zg#sGReK=n8N-6q512rW7qrt7s)NdDe-h}YaB}0CmWvPM#_L^9`K?_ctg*c$81370i
z=IV29<BS=u9W}?Kr{>r(y%R3<wnX(5E8Om4%}isBRL+OQat_wwlPxZ6w8PsG_L#NX
z5h<ol=*qd-cg#1XtlxalbitAfu9(r_#y{WPu|(GkUzhptb4DMWm-(Vk3VU$Y`(ixv
zjYl=#46ONg&|p8Tuk=GY^Gz=6H@Ww?^UorPJ3E3f=~oa`A#j;D8!Da&_P`Q?c6LFa
zMKBUHx!dn(Hw-A?EzX(TfAc5=58j4g?$J=}d>x9hQ=>3`T@)^@kAm6VC}jVQ!sVsW
z*t~`N{jNnLu~#f_p~qqj^UZc<n+yE#V!nxCzL7BBxG~?*(KNi9JQ%0LGH_dwiP_9I
zx}lkHV7^IYzIlBm6CDk*&~H{293N!CmHFmZWi~3GX2bDuHvV16=B!pW4uUh}oR#ug
zmVtTfw^13A!8!d5yaDStiVV&y4d(0*`)@)9p$F%!D<TI%<@W#-9vy(0v(n)_t{>(~
z`=az-8XAYDLAgg7qB^A^>_;lzvG-=kBJQqvmxAuoQxIIA%=z?WSguUQ^Yh85pOK7~
zt4YXPkc6i8NoXlc<Q-_{ngNLz*^@Pys6;%oOT>&j3Fvh)0Xef05H=?Pu3ic7;SH(S
z)IONOI|^B|<1pVS4hEZhV>|QB#PnG7?G%frMSSOb7R|a~G`8?QrvAMsNSJT7bc}+}
z`(8Laq!${3cqg-8B=?|1Ah=&R&Kk1MW?>j?yM|#^_fV{4P3BKy2x57EW4tU3-P(sC
za!V+3<=kgf6^go|P)uDPia9+(aq)Htdw>A#b)1vx9E1r4foM)*-R4OEcfbVT@;ZOW
zKKr4~!4EGoI&+7!FQz>5!Mi4J40H6x^EfY@9O;SsBRnu}raNl(xuNxgD-L_OV#Pui
zjQQdWjqa@7Y;fXz4o3{~Vx4l51ES8^<NO~xWIEbGo@R@-d>bq~Yz>2lR#^VQ5*jrY
zC{?yVs!k_(S(szEpBd^BO>twE309soh8Jh1W_MxV%x(qhZ4@XkGQ{I^hP=<fJaX0$
zg@uMF-fW1QFAQ<dS%GVRc}t;#0+uQY)Oh@V8qNn_1+4j@%#Zi(eBa`a6QHNSlvYD@
zOfbZh10C^ZqycWFYva1T7Wy`;p)FYr4|b~}h<(0$%;e}WK!#%#D(JISh5bA#*zc!;
zGaL8@HCqC^!4gylN$^M^L4mRa|FtM1Rz-pqeF>r*nN65;y7grp=L|piew67hh0%N|
z($_NIw3lH4Wc)E@kZzVCj+v$hXQMXkm7y?0#`_y8n9RD5rLGj?%~``4#Qq!RBhz(K
zoMio|s9u7^-pXk2p$yYA{IiRDZp6|$ijL+y)ax3WKcR+Nmi^@}{NJ?v)=wJm{eyoe
zeWy?l=9|UUWT;h5KL4wtG`>w0z57ZQgTGSalS+ChDk<pk7dm|F3*Bq`LOMMwscC*C
zojY1d18!8(==+rv{GgI%@yka+_ryWlM`D@dBN4=W^TYR%(4X;091JfIvtJa5F;j}f
z9OjrCpNhnDW}MUPM?RTaBrYcv3BR6&;)F|~s537VN!EoT@5y8F;rU~6BDP4BUn&yz
zUd7`5cGh+ZipBWBCF1dg60y?YnHXE}N_@IdDt?x{7EYRP#JdOO!o>Nl2+(6K=TNyY
z4JsEwY2{+%l5)`_;jOql_bq#ZzKOAmzl+Z7V@_{w6hHbjiDz${gtu<9xNxvZBp&-G
z`fvLuLQXe{MWdQU(DY_;T7&&HnoYu|k(sEffjOr^)C{Q?K|kt5;o3U!E3!`9)2S1i
z|JDkx@H%nvS)Iu3RxgGxsuycSqqxMM$L-`lapF*uC}XeBIo5#=_HPpwrLE#cN{a|-
zX%zdYK}<K`+gVVZn7X@Ge4bM)Ixngfj%RAcbB%xE%kl<sDDR(W7xYgI+SDLs^==Sv
zzBh=Mfz4vPe~b8M)GWRfHHp$?&BEzyi)buq5#3r_#B8Y&rP(M^Aiu0nX%o{OnnlC#
zX3;*UMYLQ~roa9Ys$9bvd1EPEza%BA#VV9EUPgbIUp{YErB`#*Xhbu2@TjQMq|0hF
zf3hkmO69bnpPcrulu?1LjE1w{+Qfl<LMv3Mcn$aQ7-*37O(|v8G>R6!+u5>b=xZMd
zg>F`%#&lKEV~<Z|KTVpetWLjdb!lWKcj*N1ed;~$WMt{lydpi`Dr2UJ&?i6EaJFvJ
zr`@ylNqHS_ay@7M>8ejFWgW@qx*`2qVniuljVb1i32k&WCBLD(gB4{)!S~E4z`7Hi
zkXz9H<5m=Nz=l4Zw4)Bu4)k2tkwg$PjgceOhdWaCMn~#r;Y1rhIZ@eQXX?z`S8WPs
z(z)$K%4?nI>>(#=AHi>@vnTnzD_uY7Miz@b=-n$1+W6X&G~4s;!%Eg~ynJcsC|_!s
z-I<i0bf%4m{prtf(8KC36gssl&E3mAmlJu1ZFD%r)QzH(V@K1hpQGu_p)s^?#aLQ-
zZ5+XD0v&UnNDE!3P`ec~=+n;G)K;EHE7IpuhR!@P(49|R6bosc&myu5Sxm<omQnZT
z%c;<6BdsvmN*0~Alj7wL>N;r`g+JUw`;Q)^=;|YM!ljh9t}dhP=PPIxKff9HpAt5L
zZ|4?LRJp65fn+#4wIklo;-81B6d1+58CBXQID6I%k=xi0`M?}iqnK|3t@svXg^6dZ
z5y*Vg>#sF#um?xHv&EOCb~vkUkDj^quzci*63&q?f9(XLvCb&u9f$ErZZMeOj`wRl
zQB&*59sWM}^uPzEQeRvN^2LTNzR0-e3vcF|2rECh4)lX|5N|aM41h&!AgpimgP)T>
zSr&vJe}b^XU%+^dfb}l{>?HhL*9C<-!7%(1jC$sq#z)<lS3;O+Lh$TzC}x(hM}yfW
zZfz9j%A@f3b`<QHZ<_x^p)xrdm8{?N-^UtGcr4_s-z=TX+D%?8?>fdJZ4>j&&sg;1
z+;!EFG<b1ldsi3EOpO|f?GuKg)_NHBIS=C<ona^u!%(M^g<Y9h@V}ddrSG#a@fzzj
zPqI;eFB_=`vN46Vn|qnr$oJ1?7USF0qznvY-DYTF239gNRr+QiR*&yf?+4?T)?obL
zjMCYy1JMvR5LbVt<Kp&o{Blmm2D84XXh_4^TkI*$NW*QuO$C{zA-XCRiZiLmTbPPc
z)@-EnQs5y^K_K%@;<9A??ah26NruvuB!uy%Xq!$FdhmTIcupeb^<xbuED`B{5@7u$
z0rtxha9~0LQezTOr;&h-XZj%Z|NO%HIp-7|kNMScSjl{I>w0f=@#u~CezBOu`c0`;
z3=$tlqsP{0jOx$q$ba9XnQ!{2M<MrFFN6;0g(t48;Vh0s%GU^7Pw0U|%r|Mdtg$T$
zMY3Ng&Pqek;avz81@i5QAM%J$%$*sE8_Pqnmsz#%-B76V>!-Vf;-hOYHXavPlpX}z
z-+@>=B@mtYwz)4S0NO?V+(+z>J>&hb`Fv*#to6ly7hg1|`(WM*Zw$KPh0$L;aZtwH
zl=>cc<H}oV$!-W*<BIfh7xZOMam6BMO#9%3-@Z=Bn(GL)LI>FEIY6tAJzlM_!-V^`
znA&0k3l|&o?_-VL6Rdb2+Y(Q9T5xA~C)8gt$M45x*j-@?W!}Veb1{KcKVux+$lZIN
z`QF4DOxPVmT%2hL%N2%LyxR~5m}?SJ3~_f6-<EC|;+huUslFIu7W-)uloYu0%MfQ+
z>&av0$^B!9-fs=C;k^M8nz?`2sv}M&cf^=&9eHov0G40bS9gY!#y09WeO(QT2sM;#
zRz*>n99H^rcty)#{#*r1SEyk0P!-7CR8Srx!ID@Bw**U2XDq?{Hf8LrV$P^mhEjtv
z8q_6NYbJp?YdJR}S>GAL59Xo#R}!ptks>NviUA#EkOj){slN<{`7*S!Mr68Oh8Ni~
z>}ylO0DhnHV!l7|^Q|xq)^9YW*uc7Sqg(>9S{bRsm9dN+h+lZe`(Q*fl_u1Y!RlJ>
z8>uCqGc^>^wT2#y`%97cepC6bpXAu?2i^YpjnqxpH$1YMY<^Zz%&ICn>s>`#uYaX8
zUSH|r^h%O%|3ZsODkyMi1vPD}Ahm}TB>h=ICzQTWf5{j6`mcf({;r@y)fFTWcf@_m
zM`A1UjbOgf?EFZ$G2g_y6$smh1;TzZbIoD);9M;d&qDZa)w@XK>?#zMGYZAx?uDY1
zxyJQhfiO8-AgucqivA6S;{DUdV%pFm(ZrtRL9>g+zk9{v%KayNbT1KB<?O%t@=`cB
zmI{Bh*W%2{GO@_wjhNW}t!St&7X^#UMfZnqgspv<Pz@{<S4>L9TeVU#Q&uW$C8eV9
z+bgkWL4iovS|B{<J{3osOT@xW&qPG~Ut-MiMzL~hlL+u`7XAHL$9dQ!l7IdaODh`1
zqk;yZ_pM&EM%IgaFY1IYdzL*=Ct_6UMEJ*A;V`sLoMewr`Nw*3y<dYE{Ha02);5a3
zUH^p7lqNBlxgs>BRdlT2eAd8LG10e498hf(gRazzxQ}&WM+kd_Hr9%LlWIlt>{?NJ
zq*g?qY82fA>czK;I??Y&od{T3C-$GH6L(fMh~X<5g%WR0j9?vVV|0`7d($XVPqm23
zf)){9)gof}zd>^X>rlI!#82&JF^Ac0S4oqQoog1|4=9l;=ibj}NyzVugyM%uX{D12
zSt!fsjGCNm%vEV}vnr)bQKNY)xJNlnjiTICDQ*b+r_agAqf~{meyPwtZ)Tj&GD>h(
zrTePvA?mM8eg!q6xM!{SUuJ{ocfL(r_@PXiu_|=)l8kb6`S!+rh2cNc=&FehJs8uT
z#`$!h%TGJdx+Fb{=FN&+zES0M)+bpXeOkg^oQ0nHbY@@&5*m6GBiE;EIvwdQck8(9
zQP7WO1^Ha&{Vg+7itKAjop}o*h_}rb=bF>V#oRH_Xhz=4JCTNq75R2|pvoEt8fofC
zO)ng1o4z9zOmrlTT4tRKPNd<$`i*cV?=MbdwcLr;^k&8x;6%la&J;S-nf4xYrDMI^
z$o{z-jlJVRLmzw6Ck-#M8|6dyTzpAqs4qQf@g*M>KU(AEPaU^`PW2Dw-l(orzPTHn
z8{eH=C8Ox=y-{>5cr<-@KAP_B;9FJJSn9fC91XRbKv|O}(!k0oG*^EXJ?=Z3dQO-_
z-f45`dB8j>JDX40^$X|(=cl^Km(tY-%V^k^<@9*TMp9|Fl~kqMDf{jY8arYaZBp!^
zzA=Y*zwHq3h#aGbi4Q1r&I7Vv^@KNPiz)skcR?&@rZ-T*(IDPonAQ;&W*cH7>);tf
zjqtal310nY2F*Nker{}zlY_XsW2hxI53z#BIcA$z*4X^Z8f0Pv)sI}ZH`WeICH8nS
z+Mc(d?XmQtBckkhbCC1o>(pH^@uml4ww}Dx=Lu`xV}7~P2TPf6q|7((m~V7<uzpkG
zi-d~KxajQ%!(D#(qwvSYLH?{G1R^{x5EfBE+z%4O_xB)}cNVZ;EAVPMp^$fm>{fI^
z52averUzs2!EW$ozKLVL>8%LGwZc%GW4^gDFN(K(qOkTB-=_HYtqt?d%==MLeHewM
z16XUwi$?vYX!v!H!GIny$cl<VbR6GV_QjyLa&P2KO+j|fVDt~-J%$lO@yvKAlG%e3
zo07qqlMKun#W}07?44m=Y5#o)BvzT&@0A6Q>sk0*nT3@bvN2?DHpH52XvJm2!!aBF
zJ2(&CHv=0DGoaduGvb;Va7rEoadiM@EgOKR!v{b=ksqB0pjpBj2zS%*p?5k4tm}tr
zoxX_t!J5sLG#s0k27k^?g|heN;+ItXKFNJI3sSM*c?$N#r(le73YwU2&QD85O>i>u
z*h?IIobyj(lJNUyA_g5u#H1OCIG&OSg-FEH-wCMwoPgcS6EJo>-=ks^(AzWt@9y-$
z;D|o(TM>`dq49Y2Ck`uiat~ip97dGK;^2{3%=3@I$Hr*f<IRgdE2Cl7mv>8zqwz91
z3PY8n5WTM#();$pSqs*2cK3v%Ya~{wM8NC-?}V)h!|5-fIL=*tLw<&!-Hj0ZSrmeQ
zvQX%E2!)X(6dPxD#|6t^WRB>9cXt7{ia~u8Z!oMDF!BvT#bNHjw&&Z_)&SgA4Zx0}
z{&@VK9~xx5qZ!c|_4&-wSAC#T!#;OgZv^)7g2_xzeB17U0~g%!;FTMCs=8rMH&@=7
zbwNa_GgiAf<5I2@BqffBvvkDbYzMqQW{*|B?Qq-Oj<-l{@pp#}raa-TsTM1=SXiOE
zutY|z1zzxNN^_(+*3LA;fHkH_KVgEa&y6uxX3W`i?yH%vfb$1KIQL_{#>kL;Fs#{(
zF~q@+hP;Ddi1m~CM#X=&+Kjn}`!*k@b2p#N5G&4b{%KrC$d4L8f3*P~tT(`&T?QzQ
zFo59z19)c}U|p&Krnl2YZ3_40T~I^+9_$@HtcqTz<+xhFoYAbpdm$?DDpkSj*(!K3
zM+NGGR8VtCf(jXHCpHpnV79rEuMDpZ%J845%+C>&p~IYWsZ|*>+DkBk-$x!O!P$5T
zhK*#uPQC<p4W!8J!knTbL#a2jQyjC+SQ(TT$PhV0hQZ7{efTz|mBc>eZOSm;sDy{7
zmGOD8GL|e+#@IE=aGs@%1u4qt?xGC+-%8kC-b4nRnbPI{y;WMZ^m|DSP3^=RGg*Jh
zHRK(&`}dYSKEEaNLEpHCxSBj-tLgNEDr!HqiawZgUTQ!4ZPdTg6Sqnl(ON-=uY9Kc
zA)je^z-M~d>oXlq`Al~cK9e#3*+rMnWUly3o3uWY`;FV8wCT1Gc8|mb_eWw}&LeT`
z)g#g7ULfkO6o_w=iiFF~B2g7rBtBg)6hF=sih;`u#opnCBE4&&xUW_yj$SAbeuoOg
zyWIsscTRy&I#M9M+$a!Q2Mfht)_kN-`EJ$Yu{ia%Kx}1A$oE-+*#Ege^r$NkUH3f^
zXU@D9A3Bu@HTLBE8C)*b?|&m^nv{tcn^GZkO2rN4n+)cgADqXEYI!BJ%3g_vF8pyO
z6o}o-R6Bk=5ig#;5F6WZZw}w1MllbRsWyv2%m%Wuf1><eqxki+UL<#A)_7hg=3T23
znd9q3fm@w8R8uQ<-mDc})9ZwoRxds<-=Htw!AcrL%iBg_zVM&$V68}#Z(Th&qxF0v
zdvYE&3;XB)M8ACvV(Q9zarH54H=XN5pLw;yXGpCW&Y3Uc1GVCUPpz<TUni=XYQ<*$
zJX<7nBEMUmFuhqPy8mtvzseiMvxomgaHoG_`<6yg^S^pwrl=D~ITOFs<BzcI{zvTE
z{6~x!`d2iy)eC3$CSh;IcerQ&#GYUO#Nc_HiQm#H-t18(gUeD%o~uGBd>4!QpPYOa
zsZs|GHF9!RBeNFv3DwJKfjf8cJW-+bdsN8kqzc{Vn^x^?8Lj><r~GHFVs*k#am3@7
z2-c_(Q%Bc}LE%lpm;KKxd{yWd`<H`$$|<^&8vT03ACosJhD~Tsv3?zB#e)vCwx=Fl
zKcPq3O?ouZUY~UQ^{IsSwZcB>(Sap8q}oH5CI{%z>^1{h(n&!xjw|SDg@XRfG9vev
z#`ML{lw9jfY5F%aTHf27qHdbfPTr;nea*Y3DJFEOz=TYHn$nvQW^6=sAm7&x^oj3P
zKDQicbRQ==T<1jZrOvePv=go5w<}s5sYUL@oyg8~j<ub&%U$VCfEyXzaijGy+?yly
zq=^qa$^N<*Ra*PdkQiSI{Oe0+zILY3<$hFrHh}6lV>K_WEB)*hLY93;QkHZS_1Za#
zlGH|%^QqBvaKjiH=s%W*>>fv#BPY;4Or&;uXHc(iGpW7fZ0b-qo6cF~l27(rdUa?X
z$>z_eSyLBMe9Iy-NLfM!7naeE3mYjSehZDdx{ZuB?x4G37Y)$eLr;4jBEPny<p1*w
zz36t9RHvP#w&!PQ`hO2;+lm4z>s3ID>k7zi<uh_W@{%58s$%&_eU#-H;V$P=S4=g7
zNv=7*um@*+LMJ#4vBa+aRv2;C8lN~HURTH6^9yZ|UTup7(RR31?}S_4F7O=gg2fMA
zFfY~>dp^2iM5Y_QT<}1Vl_%5}dSXhcC#LFn^QN;m6kKHUku&11%YCuvHFxAs>Ws`D
ze$d$Khh=4ccv8=OCCoS9vjXw+TOiybf)KY?;Jgd(Cawg0wYxwmtqUfu?7|tDE~shk
z0%?a}DE4-P#_euMxf%lR$D#O|ABAblq7Zm93Tv2e)-&I<W4`fVzL|e23QlL3an3U9
zuuo*6YYa}h#bCnh7%ZF<gBs2$?^zdvWcJ-QwkG0x#{tNCkbzM?>@ViN7>D&Ks9|5r
z;9V&=b|eM+ho|DWRvPap_l23PKNODr@v}#N)LtHlg|m3GfgiyI%rTEL(204*^8ayk
zmQh(ITo)z{6c9uV8U&;}#B(+XSfomW-K}Gf9oXGnWB1sNjxD9AbSi=hg0w|^=ly=%
z>v3Et&KgFpXP+G!qSYu$9(?L(WGRluNxxypcNl^>f8yYq8V9FW(%1Js4l7IJV5l$z
zyM=9Bdkld>t0D0I6o=YdqT4KqL$X&Klz+t{_jW9H&W%N9>AJanB?iNVZ*Db5W8|%9
z+?gAV5xt_J^F0cy&qtwHa^9ZVkw`chiGB+t{~aEQAg4$i`YBq?;|QFWe7Di82m}b<
z1PkB9$Uf%1!QpU|jfH`|!z5=j7^<TNqy4NwPzW7_xZr_URuPIC#ZYX#6@smD`->kL
zf@OLknDaVVHpqjpaCZ>8#{^+yw;((|8HkaR%g@;r05b27fVTbMsNNSA!ZGTiy_FsJ
zM^c18cH8^IsGC2=P4vUt<-Jjr;f?JP-q?H{c=wxOE0t;HtqEJ(_d?!yFBoO?M1^lp
zIGpkXx5%wC)C2eSx+6ip#+G(<L)Ktd99Zju$a~UTQ09av>D{jE<p{^2;(=c&nQ-ay
zyIyIFk*>CwwA==PZzT_IW{u35R?<)114*rWASl>Uv?L1*e{GIV?aeWwPj|EzUvb|v
zrs$Pw0>g4+6toz@LT*zR^bO%|Yk>Pc`Y;=!hc!#3kNA971bmcyx4GP%{>Lz1bdkPS
z7tLy&@n4-T7Ix^2bdvWRKJw4PI#oro5%p0QLC<yZ?`~Z*4Aq6suTE$kse?{Kbr9ZL
z2hXo*VDoYftPfO2xwbl9A5}wQm>MF6x5e%IswkMMihh4okUv}nI&wc6w?qk}g)6d8
zDWmgv6?7BU(Hy1%*G{6*yimmKO^RsPs)(Re1q|t<fN>29m?w<mazPP=HH!F47|2~b
zIkS2Q1IeArW0(@U&r!ltxmDTgD#Q4_3f9`G;)uU0^hT*dVU{X-jT7yriz>EGR6z&1
zW%b^th}&Mx9B-?DOX9Ifn5~FV`5eQhD<Y$xB3um>u_r||=UIRF+|YV{)4QG<tLpgV
z^;*f_*7EnJTyE7pkFU4na`lm1URjyL4=&2nx}4ifZpv^>8BfzJ<LSG;^3-o%II`w5
z{~lG!FP41b;LpYUIki~4rp0{iMKN!@Sj-ixi&=44F@L9Gwz82Q-+M;U@z3a)`g59K
z_MF;OCX%~<GKK%0LYkxB(4>WL=*8|g<nkny`aejeEYWVPHm8!2@Xb8$RQjrvN^dTu
zP{HXGve}zL3G-8E|Jf8W5x)6rRVwMdc|%ptQfX+V{G9O3kc<?1@-BrMg>(Kb5T8$J
z3Y~L!M^o!^>EQQVTI4EyI)CTWKIc4I;FT*r-yC`%d=n~sb5r<cjquIkTzN=-?f4Wb
zl^a>V4zEeM@*TOZ&7$`<e`xcMKXkn8kL1q((CozDbhAiqQ#XImMU5ZS?OFphjc=f&
z?hQ1&q@HST*OTY5ddeJEPqtz8RKByGyp!wcf@HQvEcikF=lrB=4RTi+@r!Ey`9)nW
z{-R6baUK%#i&|NU#u4#@u3l`Qf+O{GLbzsjKt27}P)qkm)zJ^pLR7Tt$-G@X#hBI8
zv#IrDkWx=Z5u)|<{Yi7>*UoQMPiuD8QU5&AeU{gej<C$k-Zhlbs)nj`YiP~18X9I%
zOQyqxo1%YGto?5?nJr9o_cxWEkX|6kfSn!JLY+4&vTv0#-)^VMv&3iow{>f_zu1~{
z#3ST#r!^lM-HOK>iLYk63SS?q!jq?{@LTz@p}}t|ocD=b4}Yb=#bsoAp^}DOsiEAh
z4Yd38ZwfP1WIyru+>$+vMNKN~|GX9N+S!^r&uGma=h||dr#2fulkIkY>A^nNk#*}j
zvZ0|4@70sd3E`Z&1nGVrugU(kvhk(Sk*{3p$oUps_}1kvT=c06>nxDX@)zA$^|GF9
zKIyYzp#iHK8FK$w`fL^_ALrJMOZ(~Zb;(_=I;hVZr-`N`{W_XQq(jiciqG%1;@Do+
z;;pylV+GbcE#8{jb+zUPd#o6jtT;>4nj`1h@ccP;9A#_ImlN#ym*mN(Dmd|;+fKax
zo-+^8aOI|MZhS4vjXS2gv*l$EE}qkqb%k#pEA{5eD1WYvoW?f4r?LCy>0DMmoWCiL
z;G^6AVr!L|tg>kqAM~EXXg!w;Pb}t{kCw2K$s)eEWeI0bTE<HTtl-)-VVuQlIC9E5
zp73!!7kO>u%HNx~#&;_lUfa*zJ{;t{DTn#SildzEb%L`a&T(JGo7~m@CjS(^35vYQ
z!?)k$_bRve$?Zh$G&70kEJ~J*i&XZLEgA2>HDLO(6E>DgpJ`H8@NIpxzHWet{}{qD
z%n1J^n_!l$Db}QzqVGF1c<OdX(eds$S743`TMM*pV}o%HwlJGui?h#d@g~F$jU{%{
z`EL*1W_x@zaYCG%Gyd#%#>`*hF?Ml5%QzR9TDn4Owkz(*dus0=ZrE?+4%cfQIGg5y
zISoCf$J7g&JG}6zQ!k8NEq#Bbz2M|TIJ!i#;tt+eH{KgFg>N=VkMY<)-Z;?S2Y+|>
z!R5QY$V%{sy=XVx)&%2(@XftL!B`}GGhO)R<HcYc`B#2k_{LWF=JKsz^u8X73AaPB
z>3%5mQbXbNHdJ()fv6GRtbOM}xZgJ#jZ38eE+|GaFfrI?AA?Bu7<7=Qol6Y5_K(7_
z5mES_Be$TNF*wpQ1~2!<;F8G@96K}=rTVgC&}9_vjTw!UjL{h0G8#@zk^vVlaoNSu
zXlfWG+gihrC48e;9fvE)l81T`C)*Ek==w1ZyBp)s-gpT7x(>mOrZ`N>i$m;{IA|`2
z!>yii82CeUo12mcpC5~7MzM&y9D^45vmVxnf#Hp4ET0;Ury9|CC3mTiK*@5qkHYbc
zNNAmiMCzPKq=iJ{qGY=#)kR>~6Y&%8iooib5qK+n)43oV*;m8yEifG3&a#ntRXT=c
z2XoV~!SE2iIT<(zx&sH|f%-uFJv0;>u7yBheF%I-8_P8bLElWd3oaF=j1Q76&>+m7
z9EgmVK<G$5fBxP8*p3Xqckh0(T`zkY-~6$ElX$24`9oQ_rhS7Sre^yIm-u0cjV}z-
zyiptGjUm^>68@8ATbyyBQaT_DyhI0+p5j%q?@{I{_dZXo+~*;CjP5vV>y9te+)!}A
z6^*4XcxK>&gM*wgXN41PU3NramIGe3*khQrJ?e(oA$6}UEc0#f#?l70bFA^|sTKNZ
zTOn>}4_r7Yx7czEES6r&RnyHe<x+QCDl-$mnHe^Ro5E|c2{s%s#yGi6CCY8;bD{x?
zvLq8;u7{E0D?Vb;4Ne24k8ed6*gxuwU}2hy@zNz+*9rG)b)olF7Z$=htL!=>PO&pC
zCh5Xl`uI*~>*D)EU0mL#i(N6gqJQaNPJ9QPlC8@*(zkc~lsfi@3)jddCAzdnymSF?
zJ=hkfk10cMrZNhB<R-17jLIq{e14;ZN5Ua$hZV75jUrY|m%Ea?%KvlBf^Q0VY^sRr
zgUviPvW2%B$aagh0-8oBpp&NX&{0MBK2(J6FS#KJ<DBfI1YHLu7|2cQ^f)Cr9#eu*
zh4kdCl^d0XDh^8~>r$91#*9|QzBpCbwpGRFekw3*EnehEMNHbz#Ayvp(%aw6>T$w1
zQ-x{9D5Bm?5iK1RvEEj8yi@AL$56-X<yp75jt6liYfsH%yYgI~Iyjey8su`Hk{tHV
z%Hdi2bGYl&9L~O*&%>k2_-(5)z94z2%KR_<>fL8H&oANNhL4;)>m%!q_{jf7d}Qh=
zudP3_Q{o3s*z|#uhJ9eoz90C`nr9R@_8E<z^o%@PJtu{SDU@dPhRXe<180=5iQK2U
zY<olc4^zqEj`(bjrIM29Hv=c8Qb&(e3NK5cy<RCaMLC%q>k{e6fkf&%FohhFQ|QO!
zRBFYq>B5y%TJkN0x(LtY3)`5drO?9nDYPI<UgbF<pVM_~3T^C{N;6e+shd|8%`(oQ
z%yyZiI3u0r@O#?bCyiD;O(jcCCH;V8T0J6}`iG>DL2(LwoSsUddtcMepmdrSl1b<O
zlfL8pKeTDhAFA>DP3JX#(jf8H+<Y(0BHUsS(m)lWt3;&LQ^SdRdM}<Cr%-wATTin#
z)l=EadU|9g{+zQ7^i#B)2brShc>SV&bAQpMMZf5D*e@F9`imZ&`aw=wKWONJ2Fjc(
z{4uniCRqxnWY>~nU>(h#P)E<q>Zo(aI?C!-NB^bQ(UlP4pz?b1IP!yTU9O|G*Xqc*
zrb7J6-zf5NExE?j(7&=T8kSZ~z4NOn+M$MyB-YT_X?2u-sDbis{-PHhev|PO+2z6?
zTH)MC5zCq>{Eq_P2vp|x!am;`TIl#ZMcykr773$ei&S{T<+=Ex?G$+0wiX&;(L&DW
z<kuKBQS!24O1S%h_Sb%*Ss7obrRf{_kF2Ii@9Jsb?%y=7{5SPG-a_p!wor+m0+U-a
z-S-u5bciCq{YQx}+bQz;iwfLZL6LJB6?y6=WsXc&;pG?9xXaP@9JEw}d7K7^Uex4U
zV>_^(c}FfT>&W^CyYRMh*@LL+D%)q>IBvh_GrszKR5nfLD;V<a4*G2JxHFgLcHsqH
zdfefEx$<aJzME*yVX^`7XlxJmQnHfGEGy1Sw_^7z)*PK^%`Go2*sjrnZ+DhER<IT4
z{<h+WzpQxfTx(u6#)fA;v*oZHJD%3XiLEZnHo3M7_fK@;;4D`@ch!y0-gM{G%^tjP
zk0-Ak-jj#7_;5&aZ?5guk7oq-=Q+MXJYisn+|NRI?&V;fWIK#E+s)$j`)9GnkXf7|
z9-Npi^SLs90sD1a#2q#-;k8{>Fzzkmbq&keYvf8!zZTEmUasPEGgeD~bv(cFTgBUw
zSMe47!`##2D32d^f^}A%<14B+d8+A6?&EipU(LA5+1qaN1mT;oQHiV>lEj`<lewt3
z2GZ7Q;O7$!^j6e`*&bcg_2`PeKc#=;jy@(h7-Ofg2|m9v!Q39ExFCE}8e@hFKg{59
zwmUSF%@HBl*j_raH#o=!i=u6z^`9+*$qpYxzww=Ik0#-pH+s(a;Nc9*L(XV0b-{Zl
z7Z{qjVz#p@V&=M{Y?<`ojB>+r;hST=?$YV&f#588M87c<PsK;^-?ZH7g%_XXRrsdo
z7{XpBZ^VjzlOTM<g~B+BK2R3Exo+wM`zx}u_`o0D8GZ1pu@CyMm%Nnl&0qfnL+e~H
z9tq#12;a<=yt1L_I@bS&!s}Y7cws}KY&sAfyAQ;|fPt9wbs$D<7=(iKaA>uP!hP}Q
zcKQ^J`jiMbl}8|_RU|e$N8;m@NZDx+eP@NRgLX8$w}@8qQf@)5V-P<u76bbZ#omfx
zSk-A16tza<_l402DjW^29t}%*9dvdyj6}n^a%d!$7D{JdQ5@Djj>EY-afp8|w<lql
zpW?R}-%(geV+cGO;;=L$4z}V6IzKNCzdhqHqCOU<<u;YGAQlbM<LCb|25P>N18*-3
zmK=>9!=o`#B^t9=Ne7>7C=6^Dg@|{e*BleRnH~wv0ixv?N8;7@2;6-Xf$GU}lgbOn
zk1OGremWfPGs1BoC>+0?!V!3RFix%=jAK^^!BP0;l=!f92Mt7ubtqbuhT!zY5DZ-y
z0)5F(Rp^Gm?^iG?HU;C!j3BJ*6oj3V0#O(ph~?5jbo5LBK8+4QPpf{Yl0B*Lv;L?X
z>5mb%{^*hG2bFvBobf|lZ$JFK#}^CKd*iE~4_<|O!{7jr*vJ_9Qo3#~d7*u3PuRNk
z#G0L+IQ_)~8g3paS>%odvhVsy$qm1JT=CyR7c99Zoc+ZKr@J}f^8iN#&UL_<WA^Cs
z#ttPacIeYvHZ<dHurWopWV>3U?>H+sT<rmBw1k$wXgBLDa4f|fTCx+~dr)^g+-io8
z$<lqPWC~Av6C?*4!%ptH1#+8uHs1hy;`Om+mmaJxc7t(pSJ`UlBAMqdNQv%@!J^xA
zE9!(5_MOmsfN;%KU92h4MRvX}5>s?B^_DKCJ?Vt@PTFw#rG<(YS}0np1-;A;sO_nV
zuOl^3r7Z7HHmT!BYJ1H8s)E61Rd7pq#ivjibVst<!<7-<OBwF!(q~hygzOZ_KwVS9
z&-seD^-KZleU#vr+{Bue3b;N^0aj?{wZYB2>iZwwKH!gJFBLG}M|NEPQNZBcidcP9
zxahgus_GQcPEoQ`a>vRN4W~=25*BY!!iGh1x3UyHr<E$cmZ>1pL>1AZ2c4>GVH()N
zV-%a&OgzKg=eKa1C5>#fzKJ)VZRS{SMKq5Twi>309%e1vNuiEkO0Ufn$$<yS?WtSG
zkG!Ixh`V3UXAO-!-aIpxkDKK3%r7~tbUTMH&C6k*_#6(5mfWab4p*+tV*53jy!LYj
zo1M?#eHj^i@Jcqf>hgi7Ru;*7dbw>C7V+mxMLcLs5wm&`7v3%8VT%g6Ag+)*3f~wE
zeMaraJR^6d=QLDqQZIyWEQN0dkCg5|xldiVol58L3tQYwr7@!4Ox>PJ13D*DmPRt&
zZ=X!h)RL)rQxe_znn(vFe>Ki5k=A!iq^cu{^vp4p;@hQ?m*_VUc`4Ldbe&jv20Trn
zt8Y{2SxO33$UW=Lx)i!$okBsi$<*p=GI<|OrU7qK$nazmMFuBQ+WA+MsrHIS-+D=|
z6JF7~&}2HEnoQ*zL^Bc%r{sMq?dkQJ25wKMxeGGr{<ut1Kaxq&|74L>K{hFiwo-bq
zf#$_D&<9t!B`Gw}3DIsMx7X9luJxp!Tt`=%>*%oDn2O|H)n2WE>gF}j?Eycj$Ah2L
zO8*y`go|e;@)tcb{zc7kKd8|A2gPUA(+lz2R0|vZ=p@=rQ61T()ly%*8p@V_!cA7S
zq@z(wK{2(I;#Wt3uj|NSMID*!t)^<Oqtmm#Q!CTY)bgm5dU}-7$${T!L}N7_SyxR<
zcFV_ntfr9>(&zlEmTIQg(E+)iB`o+!nz=toYt=9MF8k`|<qmgL?n?@3KWT04M|$d;
zM@#19kjvC8`qwFga`WC%+Xe4v&e3<&bJ2SWSIVH!&RL}VB!@ah7f^?1MYLgSF|Gez
zN{{`@X|#V8Wt7&@m&zYxcjLEw4K$KlXJMv0jdW9<h--~>Woskp1t_qiYBNne+$@~}
z3LHCBkq0bO=F#~o{3l*=T5Dyi;<$9|$OcBiBH6zP>%j3=(g~>AkxdG9IPYC2p8u>1
zI}hu|O$+om$yuL|pEuxBH3mH8dM6&<sWTUg&a=YWfcLF3W$h$0-nhe@kN8^hAJ-n7
z{HX`e8*jza9ffzqr?dBpF|Ua+<wGeJd{J&#M~ixJUVs%}xo*YZCR+1uUu(IuS@Tmr
z8$L0}j^})G;8DTOTz%Sw<HTody3&ok&$#pZWgdKKrw6y%;>m5hd-3WlA1>bF%Vsrx
ze95s7JNWhG84msU<c?tOlpM?%{-ONOsDbP_F`C;3&0!iokG;bF<^v52d2;YXUeRG9
zn^er?16uRgXx&17;JTD``YxAkt`(fUH=ZF`@_t#eQ|`Qx(;biU#kk|#dCqAzS#pU7
znBC-~{crLR;Tx@#n|!U`OMcn>iZz^*qz6$Q@6$A3Rj+|}hMKq~e}A4z*Tp2Iu4u2_
z4PS@q!Cu~XdM`G{w;Lvy=qCNeuS{`ys2Kvjm?5vBJ5Ky(j%nH!u<0hfuAw%l?j@eg
zH@09`JM4dG2d#awyBz3%chOFm+S3_#S2-iY%mr1#H@$>!{uRD)o9BwU<*ryT(hXC;
zyTM-gMkB!k&2K$`9+HEp@xn97PgURRh3CRIQ+p6>BM7~mz0rB1Xg1ruk^9ISt4h4_
zyPXe?NS3O?-3Raf^~IC>;^WBb13iVlm@Jv&+7-cYI~9!b(~_aO9E?^6L#3l66pOBe
zqMht(?r98#k?KI%kRAw^j|1WVO|nyy2IHen7<8(_aQaR-zDP#;kZ%NzNUrFGu)r^a
zNSqoSiA@$!SoJCj`VrCa*bt4pN72$JAA`4XF<5dU28x=}RU=-S*r}t?=gep%<&4Jk
z^3ll48jY_1jK*2Xoj=<=5`k$$ut_wVA@@X=xf+K;dDTvjlkSu_SV|_m;G5j7a^sNt
zBn~5w$6@%aIK1_U!`wf@H2(_QER2P@Z!9)QFYy}DkdCRv;FqwLo_LFEe@DS^OBAm5
zi9)h?i~mWB1Rs&^oAHsj>?e<2BsvPWoxK%-M-wA(LHMRv_(n_kX7+T+PU(eVj(Qm4
zUJSyf%c9#@$c{qQK(xIo`Ki&NIMNgXx6g8qIv0ZV^FnaNSMuRHA^7@3?tdGDr5`Z}
zhgE{0JS7nCqomu<KM>lN0?=hb05%x(L*4DZ=(SIL#p?d}dd(01$Ng|^fgk3!_QSAp
z$^OskjYp+EXq@JQ5o$h|{I9ocQ32=t807ASA@h5pwA2&Jd_3X0(F1>_$qs{#J3d9a
zVdWlI+(>tUo|X$93~)xuQYTEl=7`gu98lBI0pES?v2(T^Mx3|B@Gmy-wy}|1vNa0s
zTcNtO6=o0afprHgQTWjU+j>|)ZK^r4{_T!&b!M3EVTSI~4Yl^FiTKK8vrWlZHnWWI
zw{-gXSQsG0K_4xwhYexfU^SyFzU=6N+X><w{w~|y-F0z2QoJ*hb<ldTHU{N&!mM|~
zIhnc`AonQWBjS&{)d{ccv@xSj_Mx6=;mlwy?0=w%QBImr8m)n2&FYvWo7={64>A_M
z`AhsXp&e9ETA+-jw<VJlt1SQjDNEP2GA@)TVZY?AHy%{L!L|zU+p2)FC8E_FZDQkZ
z&8(!_%$tQj#@00PwX{Fnz3eyZ9{SC>MZdZA4h3vaRDk6HMHJsx#PD;9_;Nxt9pRm4
zZc6wr8qQDg8PDr1oqTFa_$x&b2BH~Jg#xCm5?!dag>C#=c<zN}P8F`XXxhjv{(rd7
zho9VKUNbA-Yv#O{wfw@UmbYuxa-~Tvo9?XRe)}Zvol(p!OAC47)dCJbme1Wj=kh<H
zx%{LmhqqqI;T0Ql*fA=HA3Nl5P-V94MrZNh0hw&{GlMl#Gx)=S432-F!S7aOvqEJ7
zCygoOS@DJZ*R(>uYgNc0PYT4hSiqK_^0~*(d{&;E&#P8HqZ6Lb=%UGEdYt!+=5}~a
z+nk@1(WIADyzV8f>iCMnU0zY_=~ViDHI;Hjzj?YNm0EO@DM9$=u<*?};hQ!alju)o
zBCYXFq%+!yw6Nh7HBJ}4Ih{zmTcy%O(Qi_OXP%r&A;p!#B;wOqd_RRYi+3mcu-vqk
zq|g$B6k1-MOxll<Y3%M~iX56u0ai(rBfqEZkyrG}^c9tyeo3mqujpw|B6*(|AM~0O
z(y<ce5?|1B+t)PwhUB;Y$7VTi$lN`HM(StLpy%0?qMb|QYwAhQqJgX>6W+g8vQB&I
z>1}*H{Zy@?Gqu(9pJok}ED?R>K|P)MTu;<jyvJ@o>9qK4TB-h`74(a?_7eT1_$TT3
z{GgHI8$PwVo<{iAQ=+<PKXQi}mR3jWuh!Co57p#9w3-@sSJK-Pm9*ISJFQ$_PWN5E
zQ9H{Ddeyy}(oYvtvR(-_osxdTD<#4lr4-hylpb6wA#JO2YVob6u+7!9uDqI-xz<p-
zQ8m=ovW6~HR@03u)zo%&H5DJJrkOK;(7x$4<mXaAs#*C|os>&e_1Sc-D3g{RN~iB@
z-_ee@Z)su8Tl$**jv~jU)5G{oI&PFhBQ5f2&%Q$H*6AbFzx_m?kAJ0Nol4pxxw5qV
zKPXH6H!aX=B=^OQ#L^*jOP<I}jdXW|<h>>|Qp&?7Dtg&ODSFKm)K_*Vt}3xlkqT?X
z$URJYbw;VH$&IKzuQOHWZ*FQFcu9>T^)<Qc2pw)hCl36k%a%G_rPrq$>saXXBp)r#
zZr0*BkB*E$U5>Bp##!@>d7bc$=S4GSKXV?{*^&qA_28Q3F3i>4c(a3n^z0e&iCAOa
zJKluV4x90Xcb437P7hX*j^z4%R{U7JIPJWJZ5GMBD$AOk!t7Xcj~y$WapI8KE}RzQ
z%F!9l9KF?<iypf0t_5z~VB^7GmUyyQ=6T-|UpB1s<0S@txWu|IS6TMsb*=%NyEd5P
z{tn^Ikpo%x=^!q<5Y6{J=E<JMFdh~&jBhoM=J)jzIDYm_UYawPFDzZiC%P`>t60vx
zFRkF)KUT7y=r<leR`Iw$8@O%cQBL~zAHHzyB<~8l%-%n)vCG<9oLPUD)2Ao!j`lBP
z*X$L4ey@&#Ms>{1(15{j4e0gI#HRcX=zLilE~C02SA52{^7rwyxB3WKV*uwFM);Xz
zgpf(5upVLt^&&I0{nZ^~EUi%g+*<OlHt^qO14Scw-}}rKB@TA*^svXcarQVo%mERO
zj_54;sX23<;kU;b2Svj<v(E*GuDRfTiQEiy-LUGS8z$;W_u6<547%xo5pQGz(MJ5m
z|9PQKI*gqXdZDSL7f#s|Jj)@T32&UA;f+$^8>^e%Se)mLeR88x6ut>?@WGN^KG+fL
zBexJAT&(wpv3ehD>D~voXAi)XzX#yxf&s|M93XmxY*hCNz&{fMF?(ewT(*azuXt|O
zd<m6qn}M)S8H9@+!%($147a|9VYfp#hSZ0n&M*SgKSW@Wj`Z7!mt|dS6mCd1DL*V4
zUE`zCTRbn@e@EkfWDFjijKMw8Q*5O>cz|@$B=(HO@A0u%yihjE<#si5|48U98Y!Q1
z2u#Iub6<SJQ|0zFcU>HItcgSKAnEVx6Nk3r;;?3F9CW6~;kEc~j+n>d#kv?IFOeSN
z<QNn#ibcz`Sj4-@uBLPppK^~u+qUxTkA`ReXgE|w;fma+qP(K8uT2!DOZQEx_=~NF
zM#9fqG^j4JtsvQ{Bj?3m94!3yTRg?WH^;|><AH2M?f5+yr;ZOorujgeejEz5%b`$}
z3|d&<5Dat;L3mm)c8e}{IY9buBr|0a6^sPA|Lr^(jPc`xkP{IE_l7|Hb2$*lA_K9k
zJV1W`0F0jzfP%Nub91Q=BBK3aaorC|d;DNC(+~eS`XOMpFP3HZ#?^7X@$I7z9#8Rs
zk?_sUv))+qKrq-zyvE92_-!sX&4ZpW{4O4KN6F$%bH}a=ZaDnS75i=F>uiz>noc=m
zL6H-VbaujtAV(xEb-;xy(&s07qgEF?oE>P3=*>2A!?8waXKUz=u)>|QJ)l@?3Hx4_
zXkKoCLodv+v%NXaigt5kj~O-;NS|MKQ(PKjf^&O}fjh!Bl0Cfo#sL2B^iiIr2lJ9{
z*!i<7GGu4rAD1rJ726q|R?A)LN+(=M(}DGqj<{*1jj7=sF!ZVh{8~ET+}%!C-cuJ}
zrE56aSO@E?wXo`umh6gY!FPKHWZTKU_bv^%)Tv|nk9N51tSUEL73^7{f&=O*I3au!
zbzK?u;gY$wkiNV|>F1M7_DCBAJhyJ)MDrH*tZd@Cp-tRoP7^!LY36x3O+58^6aRaz
zi7jgzS%2nlKIHnFyC3{5InGMnx}}2Khg7gdWescfQ^4ZyEqt!9h5wg#n)aU}%)Tlj
zRatW14T`XTtO(l~a)Xkam+5K+46AJ6UOiekOKx8$92@!U*&nR<qk$h<HE{f=Iu7kx
z$Gd0Na)fRz|M^(Mdc`$t+fu}19)IAigNis`nB?uz0#;a`&#$lK@>9`n`rXap(X(<m
z=501VxRuR!o@BAt*eo9XIg?*IWb)mz4A#Ar!Fr1`_=jgE-@ls8-PH5h<V(KvY!q->
zs{+<KpU)-T^ZD79Jho|*$Ma9*^6kIG&-3*u?V6K7+F=Q_J3N7wMkUaJ*aYe3OP~%T
z63Az20_jgrpc}Ii$bMY{P2QM57eXG9&Wk7Xbk!5e-};39KKg`~Dm^8wu}`R^_6fP2
zkl(ZGD|%p;NRtjH(vX5gD$W-#u;iYmu1}%ioswzg;uM;DOFYQpOOD>2Lf!sKp*6a4
zvnoucvg^sTenT?l3{Rq`mPyjPlSntVyrNk;uV`}oOL}keie{WmB9D#9G<sSJ?Nm;s
z`TJ9;#`HD4I`^8aNID2h-q3+tZ|SXfCS|H-lgY;%$}h{ML*i?3(rzF#xe*mq3nzpX
z(@FDBbm815y5At*--S2wwv~`mb`9N1uOao|8v3)UhIVbOrGcWioZMDN8yEhh6>$xe
zEFP6JPWAM?sgB<KRYfjQ<rI9OiXN}1roOkUsr>ji8a}at`me1ZCC_r=x-X=a^p$3f
z_(mr+i>d9#Vxr1VR4}51?#wQsRZ%5m-=~DG%U*?XVj2DQv5a;*R#U|KYFe6FO;3cu
zlI3poV{0`f1XNS=uPW-hxrz$UextIm659MApZcohQ^#Ssl(iz8?hehQ$I9uHSpJsA
zs=TH4MsI1QhUhq#-_!k^45}WTO<#5MXw#$uir-R1W0!oSn8Hs~q*F$ga-$pI@q=8y
z{Gx~Fep7MPFG`g@pxASbq;s&5PONDp*MI(qZq-ERw>Hu1k4<D(-a>}5>ptg!3jcO$
z#UrKnFh=&$m(P@(y1i_z|7gp5&ZzOhvF%u4TzmEu@A0i!;zj<Y!Q)qGvW=T;nyP8B
zX}%V#-q7X&ojS5bur{xA)8}6EOgU}48DIWtESu$>SpA1C8y@b$Q7OVVx(3|2&XA)*
zjJbD+342d76RpIO&HXI-@@2_f3FC|xzcIU7bIc5Dewkp+O(C*VzSWLvmO1f&&W`+Z
zhm&laI`P!MomumY3mXk_<Am-WY&*}Bb=rIJ<^XWlDnHK9?8AB{eL3xXKaT7X$oB$*
zdH$#nKHxf#Y5O2P-64#J=0$LkW(+GA4CV84Mzi|s34CG4bRK<sHjkgRkn`It<*$Cr
zS^3Edt`WW|_!-Z-rmMN@%ME;S>`_kJdW?g;PO?&`v+OkQI&a^2i|s1za?q&+_B;8Q
zD|)?Rq^qM&MFV;1qTRG;z{#is#%6VZ$w_S-?c5P9!@DBROAiwZ^s!*A0X|PMLd+A{
z&>3cokAuxn)4n_Qz3zc~(bfomYK_kZHt4<DM)Z1H#9x(-<!5#<cD09=Z1}97?tr-d
zj#$)3GPpmSq&wdkO}&M0^jwjA)fL-XyJ1g=8%7+GKAaQoP^@#ueCe4jc<q5ITQAYq
zy^t;5iUOrx_~(#pia(dFhApz2Sjp(v8At_W>l%3e8E6u|>EI~-n|`9>NS;b}fe-qB
z_rd)(y`gE`8}|<T;%{w##EtjI+y(xa`qCe-MBDi)y>d%d_QivZ!aVH*5V|k`{o(_V
zEx8|+WAeS+W)M_k!sJdDhA811_4sg<m4?GAJOcmSj=-h2;>{fuh4l0&6b_8W*hSH}
zdP#0Vd!jM(rtC7Ejgj}|F>tAj!BEXuOcB0`niGrk&9TtC9*ds}ak9@GhbeM%Qobv?
z%)fGTIvIx(i{tR`;5eK%k?eOD;hWZSld6lw5y^o+u86_NBjT?SzTpL;(M*m(rf5Y$
zfials6@!E>;tQ&YhVQCqXpM-*pAOMbdl!Yp%cHQvB??DXqtGue63cGOU2SkA&QK)u
zI!8iLvQt*aBCy*p0$Rd1!!L#-@^Tmoj0YppDHQWfLh+wu>;EbXmLA4nToWDa&4pl$
zmb+hr=wc^!iqA<rPXqLWk=hc3`9(oU*%pM7(LvZB6okeK@fFJ^gKqCYID8F&pWLSs
z=LU#2A{i<1gu4y!hpXr|8<xlxd|yAjF84*Po3DJm^+x)T-Z0DaLGn1s%Bc#w4)R9&
z9YAdDu*>X;BPE_tkMTs>Q4cK1aK~Xaci8#6N%mawdXHUjN5KWL?2L2soRBTP8?RCa
zIO<7Xai~31;_Yz%t}PnsY%tQr2Jggovp}+d-IcAdGf*}(cUYoxt_7k^ERZ-(`ur|*
zM~L)59rrSmohMV6yfwklt|qV<B-+g^BOF_9h=VH(aAv7!pv(2}b$vGsI@lHd*So;(
zZD(w(*M*P1uH1b(VO+Ql?)U5n|HoRW)$4#~J2jxxMjZ~z+u_J4HJqN@4sT;M5hGi1
z$KJJp!nf9#qNIxb3#A|UzB0B1w#Uo(cIdH94YNL~NQbft)U{M#nxzbH(QP_Op6O$E
zW&C*8!aIvwIC*h1XB}?lXK$K$O4lYf>EFc3LGp-pQ)J)74jY^J^MWSs@S%~@N`LXa
z)GB@|+XaC;Dmb%Oh2&T(xTs?VcRVY7GdKRQ*YQ7Wn<gDwXBDxiQ0`E|Hr~QH%SG2o
z9j%D5qTe(N-~1KO#CtwVc5`MUUs&G2D{|`Dw7c}$Y_I1-7bNptU(5EfwcK5;mUGV6
zu-nIc_DU~e2M_VZ^e^IJv4uQ#M*-Vy&Ety;b9rfbj@-9$_?mSNk3E*no@Uv+^WQAq
z5}Cy--(~V%%}kC+&)|&UbnekdcKAAEa<qKRu@$-8W=$S<PRrvV=ks`{YaYj~%jGpG
zIo!)UhvUv?v)Z(5o*MXoZVXDG(!mK7EqpUs_~wZ4jh*n#q_M&{!Z+WhB~a$|1bY4~
zfvR66&`Yg{)S~u?!tt2ST0N%RkjK<-<zpHc{DkUqp3sopPig$xXJjV#E~l}HbiXi>
z-bW<Sq8CYY|7Q|;X(dy(_>lFlr_lKG@+!=;acm0Jc21#p?~*A)Zc_%Ulj(L$5=EFN
zkyE+cy=J_k1;1ZX+@O~<{nJYtb|{HVXC~9a;1qHZkFq0DDMjlw%{cs;96jF9&A2qO
z-S?LG{5xtf&7y6RD>o?3rRSFUbZTV*byY8-AdQdo-mjRJDt{v7)xtPwpXhG;5?Zsq
zgxc&YrKD+}Y2cmDvTOC3-X1Ka*&|Cy!M>DMrIk?m_)1dp{zjcH*V57URa71Djm}Ux
z^&eD5?c>TQaAz6q3@;~#Vc#TM{*B_i%BbG>3swF2Lbo(OQ0UAeIy1Id`kOzIqE!jC
z_9>yk79})BKL4PSPo#3~6UDDABb9Y!RCuh6_D`&)Jv*z(M%bysrJ8z`RFUq6DjMSQ
zovc=TBS-mny5*lw*JtL@%=#R<Y?w`rB^fku_<LH~;Vs3grjfl_8nO9X>QV5H-Wq1m
z*TY#9qn}I5{PHPIx`(#-77@G(X;eiaP5M$nI>p~fKj;_D7q*#x=qKromTXm2C3zgI
zpen~onzrXR9jb1m-Z4#7_PUWON3_sF7bSK$uEI6@rE4&)6=xr9%|nK?;c3zz==ilQ
zZ#XS`>0{clid%aw?yAmnD%H8<t_Cldr^!h+9oVV514m!f;^m>*?6z8)b#=74HbIMx
zT4{5G>|!}i?Z`7acH*u%y1a057akSZjbl&fb6^ic?t0va7p*qoZSBnY<1h=J`OAU>
zruE=689jJ!wH5ahuW^={Y<kYL=9PaWKi<Zc-z~G_%qMo-Z=eI;)|O1wKxgi=)rC)(
zyYc&0?yR-QljTc_E8M}8KKgOJNgqDp-Ix33_2-do0(n7c5RV@m!a6@fdDP57Y+Evz
z)sBbrGu>#O<vvvY45N5Sn~7XIa|-LEP3JR9M8~-?mm{Vu=Q5?09N4mwjT_^6Q^0B-
zdwwmqGhENPDLW-gu#*SOKhNuXUgjqgu5r|a+g$MZF8_WYe53h{QypKjQ(FzmNNC{B
z3{A`yzIoG18=a48<6Kx5gr#&ve-#7Vh&8~B9|qDtXC&QqqTw7gMhE%(+~21Mj^<n8
z+E8oEe`t*ZU2LGbzy|3yw&--i7F!DK@VA{kjAz)xInWWmCpbdCuM?)6cEao-&hq`*
z1(y<Ca9#LjR9iQw4VH}tPj}hwbH|)2clmznfd$v4<L{d%93gp_)jbjQyeIDdkxm>l
zFDyUU3o1p7J2I8C#tTpx4*azV*z#LmxA?&HybqjS$mXWxsOI$X!<tE=1#I<$#SK4%
z^!LZrEq$@<a$j6d?u!AZ`(uX90C<cUfJFxeV3lhCs=ox_WNHvv1%%??5d&psau7EA
zgrQ4(7zUe*rZF}g9Z!a1?STj^sfxf+@vi*i7lpU4qYxY-T8U(pQs+mbXlpbmG#Vko
z7irm|-+YO|WQ|z7^pu^;S+O{@B^JZa#G>lISlrBt#TM&0oSYvA_kTp6nIm_ozv3{~
zQ1&uAiXW#_9R6!1d8ta-mMV@##kUx2J{*IbwK4L!V(?;849*GP^b@{`7rtq46ocma
zXxz+=hHHX!`(;R*Ws+=zi+;1uDhk?4QJ9n&iT`dy!d-g(ZhJ+-q9Oth7RoN>gb4iI
zD?&Uy;W%+tHofx)!PYAjX%3;#6p!}CgkXG?yxB4FFI8QVe-^&!BYdMJ`%%MA1mn8!
zjar9b{MQ(S#f3rmqY#8B(;&=|r=Tnl{s#l`(IF7^nE_}md}A5W2mb&0<KuOI=qvi8
zjc7NQC-~u^i61Up_C=qrzBoRuH`=}SL17=+i%<22^E7Y77<!}mDKK3+1IAtNDR&Z2
zL>qXbFv0_#;=8Fh>4y9qSG?=&ibGK@DBSG~>-SFREn5szgt>K>IpE$+VVep&RGQi0
z&JbJ7-fM&7nbt7rW)0urR?s-r1B)vxanscjZVN5s?kHV;s^<7BushamH$&%qQ{1sN
z#lB@GXqP6rDGg(+b}|ybiy{2n3}9%Zk0wh!I6HJh8oFXjco&?V-WeHNbdh$o6SUsy
z;C5k0^m(EUoiSQ?ov(?2VH(IUY!8iD?cf)xhRY6Z(Z;C_LcLqVe@ZJ1G-`$RQ<RZ#
zUI}MsiEgq*Za=;X=-XWu>rP45Nm&t77Aatfg945lE1)>GnX|2%xzVke5A|;5Us1v}
ze>L;Zwapyu)W~a`nz);J6K_A##PJP{{Eu`6$B$^@GrN9C2U!(AI9SR5lvZ$?nH5}Y
zQo%~Uzj4>5Z+zRPLiTSec>D7Tp7OhbhcBq)weqo{Un=>-fbZPn@prBZsA6ZuYPOc1
znhUpUIC)VmZ*i>Syp}r78e7i`7s&>uT0Q?av5vJ<YWa#&Ew^p1;Zw`2dHI}tezCWR
zZ)O*<LGL0C>{iId7sNMQm&*x%%N?pVn@_FDX0P0A_MM;2dHJH<9M9qd0a?7+C6hBN
zGq~HQ_k5#AI-k@?=Zd2le0+EoQ=c4O+%K1Xcjj`;@LYCynIpNZY@XFRn*(}g@xD`;
z{C;RAyE;CgMSt$oG2t7}kOW$xltAk8*a+X89G*a_lM|?o@XdGOn^(d&HNrO^gm0>Z
zZ|3Seq6emr$tdVCH7<NiZU#?i)}1G0xZx=|{{4)${eDIp=Rc=KR}!h~;3T^DG>IJQ
zljxP)wSF2)9{p+xMG4<@Tb)9!h6~?xNTFc4ITd;*Q};_rbTKT6_Uk2)Uuhz}QcEQ3
z$1h2D;0v03;w5=5Ns`{aBvKU*j`jW&>eD@yF3XMT?)cY~WcP;lho#YgRd1=!xp(AS
zo=Ky0vT0IkE*<!pM|<WM&_4}@q+R@hT-}SQyrP%_rU*~``-zr)|3rJ|l!#7MDxC6}
z&Y%2D9cIew_EO4^D5a-{rS#!h3EfL9q2<Tr^}}cSoLNCf+I*uAYUT9FzKr^yO!{NW
zXyT_b%2`-W+a{NjH+`kyO7gMRU&wey5q)<oB9&1^<d9xO-h)3;^U6=8Ci>ADxe*;U
z_(YBJa~*nrqLr3qG~cs~=FTpokFu@)N1hEC()IIq6)9}_PKN50WUlg!%oOwK_`5uM
z+nh@y59ZMBlq_1hB!dF4zN104Z)jr5YqD*8O<lXD(J{GGHAJS9T23a-cFCa#+dT5T
zl1~F>7090_pLAmLsCYpR?Fh)GF}hi_dT=I%<Y!Qr@Sx|RZ&dWPoXQo-X@z)&mR_o$
z?vlnEb)lL{Xa6LdtBPEms?6!E$_Ey;V(q1^IlWgK_Ec-j{R`T%!6`L<GqD{Xb#2e(
zZPXc=>iqba1{aReWPKyqe8}#=dzWkR3b|3O(9>qs6fMRkEmql}#hGoj`AWBr{JC7R
z=PPu%!;{YZW1JqJT&d3%!Z-W>HR4n2%{ZjCJ13T!^R8FsJbj-fE2diV67k~nYHQ6a
zw)EipidH=Kixn@9wc!#STRtp%R)f_X_<6n~AN3MHj=w9PsB-1^iEjKr+k>|(BHpGC
zcAV?WT66umcx7K6oY0^5HwW;SmqC2VE`)s_hjQZ2f&B8?U{?AS#x0*B`L8xZIO5(g
zj_W;~m5vXWoc#zk^%%+Lyhm~JhDE&0bS3v_kX_C<@oXI}d*gG~at5#GIo=!DLvb^g
zkK4k|#=E%x>2rKm<u-rJyTh3(Pk4v_Q?4v{#%rg)WW$PAY!vmHiwC^tO@*IX#q&2S
zTKwT(X&vD<yc1OP4A3ZDe@;n;xVYX3i^5EBXSa9|*a}Artgv*XwRl{tp`c|0U+MFU
z>S2r2YFoHv+d<LF9y9OQ<58F+UaxS(L_a6cDJQJ|%Ng3@!FepVsUe44;T|B`%_}$P
zdATF%h&vqW-7)B#2cC&P^nY2H0YyC#?(Bun3%#)Rz88AdN{3q?hW2jhJrTa?E_>qr
zXGmm3Ha9i0#Ba*I5!Xk0j5qm0^Mx-=L;NtM#SdFXOJ4kvKL&jE$F$16azE-PIk5iH
zAJQM&v<Beqc)45M8GyDS0eGAjh#h}P=4V_eY$QMRm+(z5&oJ~_6$W3^aC{gYj)%v>
zaei+EqJ?kP_KCz1?<n+mBN|R%G@4?fF?4n``npHsj65fWc_ti*#<bCrr;^=n8!ORn
zhQvZ+TP)h2mQKEhu`te%{F8AU{3Hw2eR3SWOq3gyw|Hx`<IrDp8<X-_?0FQ6i-%&-
zajNWs%bhFjPz<)MjzQAA7(AOOcQE0bSm7I2;Tv^rVV}Ba{K|>OEXhxe74639iEL-a
zN1?TG6uvf!FZy02zFdmLQ~yY)szt){p!E68iGYE4i#zs@z{`qoERuc9ld6MY*)tRi
zB~$-R{MZxkif$u!r~X%i@l3qh9>O<Qgl~R}elte+#!GzMJ=B9?{3i&bO!}kc@Ba8B
z{+eGu`eOf;z8JEjFOG~DfZ{;|Fkpdn`K8Ezn;UXN?COuIYCr537PW~I-;JsthA;HR
z`TE{iA==Hmr#|Q@cgH!8z0qcjH$HXs#!nIOy6MW^ylqd|?eRoJnFj`Ic*xeLI~w2y
zg&D4Ralr-A70#F-`ydl0JHhj`Bl>@EK=*DAh#hQ?wd?J$=!q@Nn{6<lmwYWRv_{1P
zE8JDGLQa3#V%TDd<aEhQ=~-aG2y<AT?2d(!nbLPQ!=@#scrMz_VFwco*(myEnh~bf
z8KUz~1N{3%AKlaSBn#IKMQL4S=c5a<YCEGxd)d6S(uGn$C)AA7f$FS|=oPDtz3N&x
zdPox$mKykay*;)>wnIlNHRx-#MVxUPTpZjQvFlr*V~*@=8>!;4_=Tsgmpjlg1^NF?
z0j=67NVk!&&I|?Yscw;7-WJ}H)y$6~n)#e@Gw(8N=3Mh;=?QISfA41Q7u3udb$|HH
z+dr)Or;%&58o9MzBNvE<Ge>r{2lcMx;ANHE!LyRLT&v)e0Tt}4AdHjyjqQrQv3>gr
z9ve}?fA&{!+xHbbzfC3E1jsh!flAhDuH*@Sf9D_Ft9VRZ759Hw&BG7Yu>Xi!c2yU@
z@Wnc|{$0mItLr#oLmg{c)bX?fwOn?hfV;}JjI~ZC*Qw_7TF*RoZ^-5G2Xp!I;atu<
zl*`M5g-xWpwBwCzp42a!hn>peW3w{({0O;29Zly$YUzCN+Izlp@I5y^c+Z9r@A$;9
zcdS{M&JDXVd7MW!hl%D>I6a5;EOOXlLN*%=$>Kk~GC8Gp1|Qjx&QFof>bdvnzpwYH
zs`);(F27Cf-`*C_>TT(}xlJ#7-KM!eZjn*dEt;Tto9352q`S(Gs7Cb>?bCcjj_n>(
zef@vb$MqqdvwuXE?;cak!6!7b?Nhpb_#r*q@sPS~cu2q2zM?)mUy<G0SJWvaiS&jh
zN!B}=)P!}MKPQpTzv3ZYoJ2cYr_h;W$uvavK3k3@(GB@nwXR7N^C^)=Hov0mt1oF+
z*>f6j<^_$}@{)R_C()WkvI}aKLN66l$yvNXj#00vqy8HTwofBj%b+tK(rMO%OwzT@
zru|2A$?5O6G_f>I`gz_{$AEMyb1J64i;HQ;uus%_>nCcJ@`((T#h=T0;)RiIbmu&}
zzagKdspiutt5WJO+RfqXB{cM=>}4Gko##O@wVCsYs$##=_Mu-WX4+R;aPKR<o%@x(
z{Q5%gyuMJ#%FndpMhTtyw}f&%im1=KLfW2FNUNNSsO(}9#qb9*k_=Oki+uj<V)8F4
zrj0GdbgZ<P*3K@bZ)ZQ!medcFFtMDjYm0s}r;0i$R*`SqcUoswK`JPxgu%H~b0(K;
zuH@2)$Q<(9l|_>sGw4dIcQk+bYg!d9KJPP<XZ!k^Qp?jQaov0BXOKxthGx?WYtgkt
z_c{J8k1|&0(bajm^mn%$idM-cHH9on@|7H0Vg}7zpF!J3X3+danUt(kOu6dcNOgV%
zHC9&A!JI0(II)hVtyJU>&y{(NwJH}&CjH5<);zs)8_uh5!yPl)^3s!Pvc=GjZJgV4
z+^_b0`jI+otk+<Za7}Kj-GQs0cHk>Rwb<v87Q4LA;$~r;jiFk+WQZ19rE2k()7m`D
zt`pA~p)1{Yo%v&LJ^mT0&*Qfk^NZ^y{A0NpkIU)K1J0Q9s<q~P<5G8idDDy^DVy_O
zHRkLPVZlcvUzKgxgKtW%Dj~~;&&f{ty|?y!U3zkg?QQv6H+#0-;=nTto%!q@H&!g~
z#WqvG?<#upFv*JV%<jXrHvM@}VF3SoHi!eXLb&uyD8KqRkd2oP;?Oo>e0t+x_DLAZ
zw{8yOD9hozcjj<DQZt;Z8b|OW-BG-G^CFJzzmnfHtz=4$=LJ_+asLHtIm~xGPjK1D
zISre**Z3{0Wx9*oUOLB$X1CdC)E(|-_=MeiJ>><<p7GW`FZk!RWN!ZWo?RFG;3LPy
zJ2+VxW<DyI>Y<AI$@1PZOc$>uCmue?5DRMzVbk6c`BQt~dxRD2FI(aIcxy!6v&Ka=
z8>|t}!fe@Dw(ekuv?4oXRodZVqCI>f9bvJ{5eY_4_;bVw)9qZ)uazsp0wg!}%N55*
zxWOyQ4cB|SW5ZFoPc8Jok5)ZlmEeW-EnfJAUdY|n3qe)A@ZUhfwEKiz;wSyGi!oYu
z6)y$>BbER|t^(t~04q$rp&sQ8ol0L!vGl{EX?{R2$r-Qphk;2S)CTuK%85QW_@XbS
z%lF(Zllx<q<^b_H48TC)n|6T#h|LZ}{1oZMF$+bE=r`|g4#H*CFibuihKEMsP!Yc2
zqv7~EECSn=gb!>ZalI@Gf$gK=?-7l!X3@~FjmF(>(a5ro#%*DqnxoMekSRZIjzL71
zSZoc5g`)7ykdv`kbYC*v?_x2wCKiphaX91|hcm)C|7nRQxlJ6VH%jiiG!}*8A6^zG
zJDE<gSRkDA?5_0Ni1%jxm>8Ihi^2XOF_?Bb8fDVuH&!hg0j09Ra4QOT3ZpPXw52Rz
zoBXBnbd@Z7P9zpyk3{?y*$0=~l&@-}<o6@sFgpTQCdr0on+Pl~4o7CsF#LHx5c<M5
zlM;imU-%~BPB89@cC-6RFm?#vm<r!q623_jzL_d~Q?DfYj*4u1wGno_&<|SC{qV%F
zA9|JdMc-?Eal%3Nq-y(M(6v4Y{_T$qx8yFhL^`KC`D4XrKkQuXhgG_M=q_7g(LvHl
zo!%S22lU4CTRvD|=Yz3VywNSv8<VuWaZ~P>cmL^$orXQ3u+kIjA9+AE(H$3_xWV+j
zE50g8Hc#%Mt}C4p`_u_q%1(&)cEqp+4hXw$k3XW_e73a1tx>k<a>NGP@~tt%&>DkA
zim&)&4{Ry7#7rkkD9^J%**$agP%wwHzidsdle^R#Q>-vFMTdDNm{VX3zaZ%>K4664
z2ZmUE+W=(;_3>em9&SzRhV+?Tv3^+>@y&F`>C?K{^SBd|KIowK|2R4es45Vq4TFe)
zMTmgXDJ5VcGBY$ttDp#Wtliz+-JMt{sI+vKblpoziGmV>0t%@7^M7a0J?^fEEQ{Rd
zzVpVvQM6*jC_2<HOX3MK+#@4RW}PEw>dE2MYC4Rz3`o(|7D;OS$2TcnCrLd#gjD-E
zbIm<6u5Sj@PtIv?IwC@C((K{g#h#kS1FVG%;H!Us@#g3O{Pp@D2AKWBEoFbPlkZP+
zB>$m?{6AD0%0DVG>(uvQ*Rp>6wB<M6JKBd{8-8JEdOLQ@wxgR|JEm&1qt&N2{Jy6R
zPb;_Ksj^o59np$6Vp}nRx3?#=hLgqn+kPQ!xV60v*G+E6Ij7pOfi;{TOTS~`unyc?
z(}AI(oxHpE1MOz~M8%e$_<UU#UVO-z?>$|(XJ{Ap*8jx6%{A!qr5vNBSU))^#7sXS
zntZOp3iiXK-m1a_%sBD<cw^7EBmpzZjtb0^tiXS(%2Cv#3@1zw;>F7X6#Od1@KdE2
zx~&v*Rf@6Kqy+mm6{EVO0KXRqvDd8}x0_U8%*qPX6RSWw#c~{#BSfh+0z7V6iU&58
z;O3Dfn6e=jl1k%XU2haDZ;pa}pF=>hI|P>LhJwh&PzcuygQoEJu%q}rOb(2Jy~|@E
zt1%Y-?T&*zt#Od`gms;X%sqNZuwiowDE1}7^w323b0rbpIweB5Z6Y*OCBTG|1ZXx*
zfR#E4@JuNIWJe~ztB!bZXpe_^t?^(|7Y}N&F`!nO2&dYab@(=AHX{e_DdfN`zHgaq
z$busy(_zb_45(X^32}?Fz%nfhP7UP1GC?l*v%g3BG;f326+rp$Ldar&@hY8C*itQk
z`jcfKWmy4(SF`sxxDaNZE`mnaVmO;z3>9_7koaE-h)9;gG1e^NnG^OjF$V$n4Nk9y
z2D55#A5sl-pH#u29aZ4;tp>(@Vm2|Whc&#Hb?W9PP*Z4x;?PEL&EpKxU-r_JHo=du
zX4rY7876OUg1j|fU?ShXo<bu`&98;kdbL2?YeD)?Eo7am19`Q2$h%(;P9hEP>)Iz6
z{OA*uWHo^0fd&X`sE0k@>p<^zE%cgwWIe4Co-2I?f8HQ1xXrqgem$JftAb~x6|g|A
z3_6{vV1P9l+kNF=DkX$dJBp#<5__<(=Ypm|9t^&o59iYhVNg;D2sR7h{eyDQvZ;hE
z3##Dcl`7Ektpd~Mm2mV`1?-J0hiO%1d^anD?iWJvSS$oTcOhuM5rUXx1+41)1d2v&
z%&+Y*L9zp`{c4B72mZnO_hP87A<nwO5Da(YZO*R}c(Ps+Hx)^tHtRR5whqIQro*w~
z>u`MZY6K47IuhF*q_Jd(4EjdM;14@lRK3pn%~4r=Fi{p$#AVTYtSrtuD2sP><?!j)
z(dcHSfZK|dP;E5xjj;-Ty{3U7p_-_8UmI6k)Ir(Hx}0Ou#qE1_@WMH53~<oKEe6^+
zGD;g&=IfxzcwJ0=tdC=k8sYt~#`rGD5Cb<G;hXQq_)*#fZ;vv?KX1)(r*ImUot}=Z
zlL6iR5o_G-@kEFN9$Vyu$CtWrubvw=wYj6ankQ~L<B2Q1XW_9vPkdeFj-d^nnD%ZF
z9#L3~|I8Po;nl_HlduFEm6qY@6Z_F*?lH6xJ&wJ3$8m@M37p)08nuJYpz*1*C^j`5
zMJvKF_(lY-kBP(#jKVaXIQ$}-iVGkOcOOs3e&=jN%)|YEO7X|eUs!xygrffwqqi1=
zX_lHe9X&IemgMmF=l7~~aSnIP7pjqtB5ODQaZZ-`X6!2il3H&_4`U4Jv6K-VI?tYk
zWFz)GjHhqRH+KBI@v+5}t})+~KQN<AjR~~&@dVPF%-ti*H<xBkCe8lIw4XPql(MH#
z^~9;9&AN@>2fj&-;EZq7bn+LqBt%Qv&V1wBX-QGcH|0@4`(ATirW|Rt1kn|DqEE~>
zZ<%jum~XynTJb)!6+L0T`L1V2B7B>2NV21s=j>@Z^G(T22eLivKwBm{l9fMq=H7Lp
zoDWW<<>yQeiO#f^y~gUVT}bbV8$Z9e)2KD>G;p0ec;9+ZUA_kmba~JN<{pQ3Pxb@x
zUO~MV&0`jr=rM=-uFWCQ{JFf@%^4*_A9~527e$?U)b28mPCc1Nm$)|v`nVTxtRKbC
z@}mzI{b>4AKWd2fBby@b9%f(76ZYbqA3mQtN6e>T%r-s)%reY9j|=>$^tK;;vhky2
z_7&SQCk3Q)&(3CF`v1RPD)>@z9s8S~&ZFkJ%s2x+^sd;4Ry_5gD!xf|o%SJT_TJ3n
zkIRU?H{<vwm0&QJ#)`}(KkmM<W4_4_pF_3mn|>+79XJWzG)C8pT6MhWSI$hnA9D74
zuLsRyzFEe%DGT-ySMj5R*(N>Eoo4cF%7XdEP>gd@iR`;kn?YVFPLz1XiK=~^=neCY
zce5h}ymh3q@s1?f$-7{09BA=hd)gjiPr|MC)Fy9F_UU$<iL#^jL+tqL$(E+*+LB+4
z4LzA=LrSl#X_Bfnb>Fe#oeV3gFGf1T`(#pgE$EptcNXuMM$4X<Q_|_Fw3G9AR?9h?
zca^<fr4z|*%tXqZKY@(znUNaz-8|;Z)MXD7-k2Is%It}ME^16qt&C{ZAwwFG%sW$q
z4d^#(H|7_|(XkReTA-vyyBFxvr$;)Z@JX9Y$MFW+MlD(ur^&u-O$t1xL6&kF%rffS
zZKy_b4ycmMY!&*dp-g%s#?m166;IJnA`vr18fmLQ8GfUwagRI&-x)<BZ{=vtb6L{f
zA;aBD(xetMg06Y-9;WCp_9#n{eu5-P3M8mgLxS}AJ~gb5cU4C49%d+Kq}b=uut9{L
z$cT{po`1Nq{STg<(U1O;{n+C7hqq$-@m5A3rkVc1tq=eHFN;0>-e25S@fS1t`_S%e
zABxHKV!Z2bd??e0hGxIeBD@_-I+;~!+i+)58;ag)!ys%!<K|YJ5YURpPPbzCrB?Kg
zYsCWAa}-_La60>r9~bl6sPN9^=62j&(2nue-*L;g?^u-8f#Y9vV(^w9czx<mY{>tK
zGc3AL=+%Xz<#_LV2mkZqT3qp<0_WT*#~TAe9N{HInKl6~Z4_YREdlOzEyYuECEUwV
zjDyCMU|@GC-d<gfRb$H0^HmwX>ky(<w*Y4x5#TpLDVqN)z+~$pEVatRc%=gT^Qj00
zREnaCLNq<Xxhk)6^xjvFYx~O3v_i<PECEKGE5S)x#aJ@62-jy9vSuFx(^R7&qlfjH
zrVvPL41w6L5D=(_f?!uDtP>3b-KAmh@@F^<ERKM#HL;K}5DO!=#c>y49K2w^amb29
z$jDEErhz0l(whj!nQ!he-<)B-xyO8Sg!x93`Q|S3jXU$rX6BoR!`Y|Ie51^KBh7p>
zi23Hfig<t*F<?572;Q8L3Z29r;t@IUuqhMTEmEQ7V;Wqw$biYyGhxZ?OgMNa3u3K4
zfbWcK$i9&S24U<;=3X8D^Z9V!p#c8z{&};14)1N|!ow@PL1CB=(uRf5ZB_*RHN1=E
zQ4II56oZFP3GCy2^HhGGmj-5qPZhAfxB^-VD?pCB1Z9SE&)|9PADmak9edR<MXLrL
z%hkbM)q2pH*#Mp;pJ4QU_U>difQ<^@aDIP+QJ3n$Zu2K7_HBfF(>{Y<-zT^j-T-BX
zYGJQ!4a^p=g$rr5(7d`1W)7-{^0W2O*;&uoZ2q<P8sN>z25_F-04J{0L)Y~>_&K!}
zJX)$@?ez+HJBD?qsfE0oo&%HSWkS!73>bN=94zWGq5oVuydISX0fSN@_IEOf9*l>K
z_!KbHPJ?$-Ga!9I7FUhtz^%)L;L=b6)_kXW?^gl2J1e1gMkNffuY|I?3fMHc0{XX>
z!;Lp(aC1NillBPVmLc<vo)Dhz7eath8FU<}fOO7#>E7kdchM$z8O;8q+9p`+Erd<2
zLb!Ro9QxN*!ro=gpynot(=(*7n*GIIdxxQ!G3z%K!}0g!5xCH2B$irAW79urd>kl)
zeN$xd>^|0QTx9WD72mAh$zW8L3`%OsVyV9@{vF_Lt3wKS#X=Fsx^o9|of7M$W6&Z&
z6-Am=ao0aJG#pUF&RH5bQ&toAEZ0IaeQk_a)W+8bw9(N_2S14FVz9G5#`1oH4)0Y(
zvhUbxkTFjAXpE<%Ot7=V1f2rR@o4TeblW){C#wSfJdL>9#va%2aX``kCg55!Lfs5Q
z{8em-qbC`n+ciU6yK_7~?wNoqFU&yo20|$bu4I^N#osqv(DvUVe7<`L8fh=XMvYDQ
z)N4C_ao&SsyANR3wZph}I_Ig@9pm3S7>_lEqkCc`I?6|(&dzAG3W~u!?TKh!pYlKb
zX5eT#Hiu>6@;SM<Ylr|tlgdzZ&|lOUAwt9diI9r2GTpnTOb0qu==@bxs`62zgfulu
zR2fGh{QZ8Fa2&17*P|bWdjI>;xYdwmZZjf{L?aqrWki{ujHqL@Debo6y+T`4y6$93
z-S154BHyRP)=i|#+>if#?_?VJcQOSnm_qr?H)AJHCCy;=-AtRtuV*>u%buG}>?sbl
zw4~NjOL{N}=<qZkiAhAtkwm*VLv_~7iuNAmjCiya4YjtWk6Wy1TB0=_7;Hl!(``s~
zi!JpR+S0}qcBEisPg93G&<!sK(&H^7@dgKaXz0khflhQk-HDp#IMdZsXUbJ|p~3fE
zXp)F4J^aZx7+v1r^zooUhdgNa0}o0H^`M{S9&|+4ljLGOX!1`F%4N^(e)guU+2lzl
zj(O70GoECk=|gskJ`^^39w~75%*!V3*n8<q4LzKtV*TduY(Ls|mNl5iel&?UF%|Ru
zctf3an%{nOL~K4a4W3VG%r<pBe3Sa>M`M_8uHW?|Yx1KmExv7i^(Cz&UpltUm%Na5
zo6){xAecwHZ_cA94)f>%^GzXZHpX{+Xu%pEQk>0sDe$2kN}LU6?@hJiTuS7=n}*?Y
zNTq-^sYTv2%-x%gy!N7<YrVLm$cyyVyvRw$i##G`(s9;ru#fZJ+}UTgl=m(sxYLP$
zZgiM8GZTNhlFSEJdVATKjup?KM!Oj_Sz`umh;yQ}%TAQx?L_7(PGnKbey59$RKR?b
z{oR2+zjC0vV;tz^dG0E<wWl3Y_Ow3Hjv9{Ik!H6onJ=~__hGhl)!&8`4Q=Si9nQiH
zv!>g7d6PrMiaaxrZ$8th`Mw2Js#}ommTBaA*_@nQrc!dxWHS58U4H75sA$zhs(i;A
z;XlnNagrJR+iXgIgG?x=c{~l(9Z#yP-MBwCqOw{;s@E{2waX1?>Pvm9W$mWMY#asc
z(4)#|T{8TuLxEO0+@q{bQE9xRIaG^6d^AZXLWBE`HAqmWPNnYZRK8M;T%A<uiX`{?
zC5)xa7h~w?J0*ITs7MRB(?+9;GgF=NlqoAu=ckS$b8k8NG*_0UO_HG>O(W^Y(Gk?F
zGMxHyhm!AoDH=E}NimNlsH}eoZH(l7gIw+lw&0GKUhbrc6`|6lA~aJ>gnF0!!>-?d
z(1-OH3ucuWQNOY5*)M#s`8Rfz{=($4e$1TMj{#!+__n1FH#qch2WlV6PVdFk{l9QS
zU@uA!{)IhP+i^l}8@dFx;qh~A_<3d<Zf1^o8r+JnwzlGVuU0&}pcUONw4%R&IY+w<
zk8Ehe8xd_-A=-{kKJBOz#`@2w?<mOrjt4h(;Lt0bIAie-Y#q-xsF<I)QK1V#nLRh}
zf1>8>4%}8yiCvp2aJ^?aW~~%r!VLlLIx4{I>jJ!bq7=)MOYl){F}`|GjQKx`QS41A
z?==W8egfZ=EXuIDScuC)1o&~g06oqZVuExQ`b_?S6aVGlsL*`8X;XxDS4*&ZivWxH
zZT)VSAq1EG&vUFECcvbK5^Q@^gyTC3Q203?k8jRL3!PBdz9tNs<wId9^Nqr%5E%ZA
z^&9C>_~{=Cb%Vm-%d#+VY6*iAE5qTaM<nF`iG`|laiFCi5B6sh;T<M{_^~AT8j}P?
zFB4(S2-a`P<DuqZJox{J1J|@Ts1D;_!&*_-lNhj^83R{(<KfEBcu@Ej57~|J@S!%I
zy~*)l%0KULKL$z+li>c&B$(5e1o`>NVA!7mK?l+xt0Wy1{WIXpicAP(?)k@F9Pw)(
zV7OKe9CXWtt=zHm;3)4#2tR;aN;V{xW<y0_4rK4ng(8z&SP_*2`kA?~xx4@-O)Lh%
zzG9I7QVb!3g)mg60-_dI!0M+J@ZenqqgDlMOsRm*_6l%ZT?t#qRKYsFU4>6#ugmZ{
zc*k7gs^0)pS#Mdnt^rQpZh*X^dN`<C55C*$U>VlH)aH6vQdS2K-Ri-?p&oSoYhcho
zH55Ilfm2s%A#+w8T&%4Hm80B2<5&;()f=FGc>@H`YJk^$_256N9`ar4pje_7M!v0v
ze;yV5v6q6dhW#|0FEfkDfT8Stx*eMWnGTsCH#Z%W#nPZ$Bo%zik|9VV9^$eR!DdVf
zEWeuu2{{>1^7{i=@~zKmZV4QDUJBX!1hDvg8Eow;huL-&kZ4!|6)EMg!H~1{`^vzx
zNC;)tLWpV-z??P#EU*=VI{&-%+sojiPB~1`t$~sUwcxA3`Sxc*Fk<iW-vl9SJXH>(
z7goZrs~_QQY7MkG4Z^%*LvixKVW?&}90wH+$BW)0P{&RjQ`moGcjyy%e{KM|3#?nU
zHNaMFSq$4DgQ>1E*s@jzU&b=$sL7%RZ+Bj2|IJP%Mcju<s3=gvQ}<OdG)@)udsWd^
zMGd!JSHtlJ>R2~g0}q*M;v0EQyt_aX&&X-vp>A#b-Kc{d^5gJ}zX2Xg8;5(;jWDg%
z2!Es+W7YHVIOdoMMwyu6<gsQbG06hO2c~0wkPS-q*x{F8L%g-X7-u~&z`M}~=-Fa`
z@ti*=KSR{EH$s1V6I}Sw1jYLh-QEN5<yv9yWf!cOya+w@7owAJ18N#?$Jp{6X#C#+
zyxDjNQ<fjWR+D3RH6;Mu7ldHAdl){w9EpQRM`85dXe@GxLl;QIQCE_&;NE={5qW^?
zPp4y^XEsh-or5AX%5X|{8TNmx#M{vyv2o#G`fu}Kx_E3bJ<?a@Edw>$+@eN9Myb<1
zP$#if>hwoTiyEA?NVQ3aV7eY{u+<|mYdxAhO^==}<X!O*+H_h?o8I!HjUU5p^yrw3
z8U5MDejN5g9156B1)QB4ws{IQeP)mGmZ|i6oH-4eK8;2v@OJn{3o83-K`VY((tZGP
z<cy5j8s?e@L@6aad!=MWqn2BdbC4Cy`E5me_?f?f`KC@_O&9vvUu<bZw^!KEWWI6S
zTV_YY?CrVB*MY9Hm$fCvfo9k7R-=?7U7P7hO6#3yuBS8om&TqD&QImVxzfTk?w{kF
zkaQpWN<=-#Zz%6m$#Zv|h6hdMEVSHm4@z(0ea>GVq&wP^EYx_<fi(`zL!K1NeA9W+
zlP0rHa*#XqwC=Lkn7j6@(|qYnH~VEY{77@D9~qwX;~i7ZQ*nkWE8CCW*ZGkHYc;d@
z9_93xZ&=JWzTf@md;@!Ka`<g-`H{yoKiaM6N7tHsNrmrII}Z4gjkPZ+bI;8f?zsuM
zF^~M5xr4`_JvRG%Xv8A!yjkf(pB;VZkckgfD)>-4Z%r&>zNs*tLq`YCp}D-haA&(W
z=_+_rVW<}gXM0ihc-Czcyhu#Mi;99~Q_Z}Yw09Bb=T~@8DEo_@$8ryJsT+Z}8+mb8
z-$1uBeH-J<n#2rxV?TrZS-*)*aH8?2oancg6QyUf=ZQb(x!WD6!{45JU+n3?2Ro{l
zv!mnR?8wgCo^<&(6&h_vCfrr*_}G?~d)kt)*M@Q~*pSy4_7{({rp%qZ&*00wH$6oC
z4n%teK;1WhGk(CF3=|O#loT+9L@!MyT*6x*uO^au{{))uG=ZL+HzPr=DUBV<nJGsT
z`hIjgEl4t^Z|u7XB_m2cWJq!;26S57fa;v}DS}!4B5OB475Qs-zAlMB)S(l#+O%Fn
zo1&L$(eCG(^yj+<ZFAKis}glOAE8cyJT(&4RpV|-Rf<Yd=KZCyq&j^J<+>_S_F_fK
zUaLTBwvMKEC*>(LU=%$qm7}l$S*n(nrJ=pj6vjIXU9KajqJ9|ZUL8uemPyerH%V%m
zBSF36hS2=I;+%67C8w#Pw5wZ$f+Ix8ZGi~Yd>z16&P2U#{DZ~w`q8Vl7u_!X#`1N)
zFsY;m4_fu$CcZ<r9PY!5OZ#xXOCN4k?8EZ5UNqU>i#_(g@Sc7TKBI1=wXEMfZNrG2
zZ8*-g4fQ!M^*4ptW(VJ+%v-Tkf$vkAt(fH7%DsH8xQM;R<kp6<|F83WZo^(^$IzSY
zm{-+~>(7735A!;3)~HU@s_jHm{hv51m>+|?aQ3gC*t6y*`ds~v<EB=ja8U*R<lY)x
z1tHpc3GjrR0FBp`V%y6Sbe>XzDzU|AvAr0-2Nh%g%~C9I6riGb8Q#+=!&JUaxkwb^
z>VQOSeVKw6^)m46=PV38n}fk7`S{<PLcG^s%zMhEXlcv0uYW?+lq^FJS0VNV@f~YI
zF@AkifVUsy;p*MFIBFmVMHNCJ$T$@8<w9Zimk?;I4S@yBINSR|xSubSHJMO&vLXzI
zH-*8uMd2X+EgZ(~d=DNg<6x^!JfxrI+mvMztk}r+sWVBu$&?5~2I67)jd-xOkB3{o
z<3My&9E2ptLi*lV=xmDtFTWUgF*t_%dZVG~YdpMSzCq?2p=UI#-x>{-*J5Dlg?L!B
zItik_C&A-8$uKJ-1=L=qf;*&xlv@V(7-vGvtV}TO&V*HOvmj+*ER1Q21>3YF*sGZa
zpRQ#>D|64mx7nbuAqV!FX2IQ!S<uXSkCky2Oltf9l~D!IwxSrMB8y>)Pyhi#D<EM(
z1)P7(&+l(4;8S!3G$dC*4`;91T`EB#q7n)-t6;k1N9f@ERU_Yx7EZ5$jMh2`<~!1o
zAq{ZsW<7NOtOMnlb#OiABRstQ5l&Jyc<=uRvlQ#t+fxHK`G(Z_q6&7&RYM}wK;M>H
zh+S33Jwx>n(pnD}EgImZK?A%ftcS%q^>A`%9n}BeJ@v>cu*xon6;lO(K?U%Pz1I;%
z>9F8O5?Jq#gR%K3uwyK5ZPcg2s^2M)dMp-l?cT$HVmy3Phz5nu6fhJ?hv+8uF6r^-
zV9*DcDP9T=Cre>^jsT=X%OG<uZ(OhzwArs5@Jku|b}9qcheA;OCjbfVPMq>w0Psh^
z_b(x=5if%`HDyr!<Rkbktc9+fpLugk0D?FHT;CypM^gN;^5=Ag5_@@4D#4|_5@d|4
zpjD@eGgnpc{$Ujqv(M=peFTZ^9|697gb8PBpw6NWZq+q_^y*JAn>8KdN@=_*CWAq<
zWia5O42F%D#qTTSQKC%&51J~WZ;}!|>QKVsC{_I1r;3A=S--iWhHI47aelr!n$6I_
z|GsJ9;7OXeH%=3`Won_0rw+zMau49hacFsB9Cq?{Ra=%3HisLd;+^sMdW#8eRyM`j
zE8|gaxgnm(GenDfdT6jm598a$;jT=5JTS=s_ii&lhY$n&*JXf~Zibj!Z-{nzMp$CU
zI}n!u=Qdj4#%2dx=spLN9DH$@%6}-Iy#Y5L-iTr@n{em$O_-#27)LuD#=yH<aNrwn
zb^hCqeVIG)d;VE`cK93~(mRhGN$1hS^8(H@zloBU?%@51J9zle1MGhI0B;=3!R*U9
z7(X-!J-!5@;;9(C6q1VuvAO7&${P;Df8xrLZroGVjrSY6@#JU;l1Lp&ap^;8e*RE;
zBq>9S7s-&{Vj0qSDnlm@ji!x!+e#R%K#mFu^p78k++imDRh15>>Qd>!Npv=3GJRNY
zPT%gD)BBg^G@Y|j60y947-3Fs&n&2L{&ZTZV@X5)S@I46kitErrm;kEtB7tsB$8*o
zQ5nm8v(kzdzO|y#Usg05tjT!0HPu&G)2=_<kHUN-&7RaJ-)w1<qCFLq*;6U^-&|Y8
z`_oa(IUgNp=MYB<o8(BX8=a`{!VH>|;!H<}vA>x4Mv3`mH}j1s^UYW08y)7G(or6C
zf%)czr3cAYc~D`S2b~k)ytKS0xsUOrv8>@JA#=_tPg=};^YZ&#a_#b=UCQk7EtyBY
zd{4SPh<(D$H?@oWc*B$#=e{3Z4dKo2Ozs}8=3Rw0KWb&Jsp7{y)@+=L{b(t(O+%O;
zC0u8%$DF;z%r}oe`qI%LUmCxU^&2~1`Z_R=beL}f5Awrp9;NvBP`ow!Z+v}-m~EDE
zzU!>94=IoKq5TJVC!9UegSe~EeZZSaUVBp}Z)Z+!@S=M56Spt+q7p+dIwb8y)x4wG
zc4Ic(keN*@&dwwUWzJB=xs#rbJ3WtbBh7eMnrq@pR)3u-iTTFi$_xU>8T43;b7`?o
z<jQ;_^M`k(Y#nJVYi8FtLD7TkC!S$PmM?ATua_;2sI{T733jxv-j0kT?Z}lq;h{He
zsn5=qN?L75VV@0kir7%Y5NjH-&We8U|9{^TL^YN~=iVYcvqEa<0>WIN56m}tcYwZl
zOrpx1i8L^ABAvW6fm&+Kcna2x_N+9e<xfr6lQf>bsEnsw^NnfWJtO|y8`1`KLmI}~
z&3@KywziI=v+TQ>vr~^&#PG%xYd3PN-5ffkO}`Sf$oZcp>04=%)d>x%u284s1?qeQ
z<D6l(3UxkE;hP?3rhbp5kDteoSDF$jH!D)Es3HxL;w>qO(IhoSo}S|<-d&QT<H54z
zS|~$(nbPENawIJpJ%YA98Ad~94JCcv{k9$|Nd^N$Xi2O%nVX4Ig2^DNVZQm)A;SA<
zA|&lALb_!Gc>4A~40`YfyUzDveco?;{pA-1jrfI=PW7M|=dc&O?nb|WUuZp@8OM;{
zrnwgfh4dmV?#0{QJt*4IjjKz$@Kbyn%It5$gEnnwI<gIe3tKVuL@NrJZA5xnFs-Ns
zUwmxAcxIf+{jIpKycOTD4`(BLbgpK$;RVfh?h$Us+=EqUt{_0mU}llCrP#?j%+1B6
ze3vi9d*6SeQqxbA8}|ddFW2DSKb5G+zTcB}<@k#|!V<qr@$R)!bk{CL&qF16SfvCv
z`WK_i(jxx;T7(DI731E2B`A78fJ%}=6y!(a?mKb#du<}?L@5~MpN=zkW}(BF9DEg;
zhh>_DxNCPYj@?>{OA7_)drF7_CzzvH^LboSf)7Fpard!2G*!>RHQzqqi5(yCj%+CO
zYlgzbk)cq}x2dwq5Li+k0*b8Vd>j@G7OwALeZf1pusxXdiV#>lBn&FvML?l?6g2;h
zgZh+sIQKUJX6q)wc8Mg`J`w@?;vwKtJPfwuyHigb=qbd(_q15B-W3bKnQz)R#z4)8
z7}${%4QDq;gVES%xHCB#^xUJtd`C2VI2Hq?B{2|`7Z2CVlEB|386-}oz_l}}Q1Unp
z0<+Vh<6{O~#uza2je(P|V&Kc<_powB6g+qp&6}ICU=z(B!;)lJ`#g>Ruw+2ovrK68
z%!J7DOqj{f?X^3yz_^xK=S?=m&n$)uSOR;`2_R@@1uWge-jj>0-$YeF2J1H;TPnc7
zq8z-!D?njK8GQaEWDTkeG$)ip)8TaZdnOas=X`*vYjfb|+!~NQ{*nE?)lla05k#db
zK(veV+N-J|NbVyv4gU!K&K2-#Qx)`8SHTg58o0Wo7FI2&gX5a@P=CB0ZoRLEs+aYU
z=w1(JZr8zij}NS^q=1)vDhw4%g(RmGh)GO_g-=p=bT<`XVIs62je$E>k?`znGF&{H
z1QQ)Xz-Lw%Xq!hub8tAMUyB6Q(pWHROo7w8)4}mYCM>*?1(jJ@P<E~aV)hop@#8`W
zJW>YYZDnwmeZ>;5%ix(p8N{y<f>fvgw6p|}FjD|4-U{HOwh-Q35yGkmLf~QwFm+Ca
z9d~l!(jLxd@!e~kx&X?$%V1JZIo#{10QCozAXZuli<|i|whAh|tHAhH70hCPlR`%|
zET7A{t&)$hXip6cK35APhSx*bCiXM!lSUC=Y1F$RjoZFT<C2Loc*Q{mV}-K#%3mIB
zVij=YQYD<#ri3TF$KW>JtD2&zjJGeU;nWXm*m^=8*XU~C<WLQ?{-c4qD>d<|g%)Ow
z(!tI%x~MMD!z8D1Sd}{tbB7q<p<;c!$NN?L#~5Kth!F<Vjl=Kq`uJ*?9_Ihm#SeaZ
z_(xnH|HkNJvbq6EE$7_%BLiGr#T|m&AGoK>5LI~x!tCjI>~glmvX|3vcb^AV+k4}9
z?YWqzxf1(lZ^YE28&TR}6I$%pgrdotP}z7h_ULXyozL4aL-`bnH=e=}tJA1>?=()(
zzJ%rbF5&i)OSoSDGWM-|fNz&RM%~cIxbDsqJRuQ??h1kUd|V)Yi}pviBQLQ2cmRIR
z3&3#kSEx5362E$~hv&aYboJr=syP+t%ijk(#CmYutR8%|`xg#9`3s+2{)O{Gf1!-I
z1l8C|(0vC9lAIwy1I`jO*JuR2+dh(t$0$=miaOcd&>_)4eR}uToX+Aj?w6ZJQX8ky
zs?*cxNZWK88DU8#qkx`Uv!8f6(z=~UKFUOSuKaqP^_!9W{Pyk;XJ;N+QCczIDrBt*
z`L(_3a%-xnv!>C*Y)H$(h9c(KkbjdccktTN%uajK<mXng{SH*kzShx|4t&RQq-E^G
z=~(GV+SyLD@5Ky~`|ixTjtd<Ob)~5pyphCSW2Y8Z5?Hv=$gA$8#C-FV`9_NQW_-H`
zy=T73V7_r-z6qSd`wq-Eeatt&e51tq>%z}Iq|P_0ZN>BGLaHxWG2e{Q^dk#vKYDnW
zeK>q`3Vz3oli){%e53MY#>wJ4)&29#Htf^MTJA@$5BX8aUS=Hrdj-~S(&YRo=&mm*
zo${ssw)@f|zE7q7<1BdUJbJG(k30T+2)%viGV3*`mDy{<Y!hYbL*eYb`5Ql%RvB?8
zA<dy>a&yS4(VG@<TgMJ@Z#w_ciypA2DALV~j_P?)rIZ)FZJJH?muJ(r5xk{&Y!<b!
z@2Dlwofa7J#>IHvTX^qE?uMN8j&LE5JZDmTI)n5*XVCRtCko^HRI9lYeZJ>N;><S%
zlN>mMWltV_hZ<nE$zEYglCx~-%Q_pvGdASJO#0=O9o=4FM`I#*CzUt@_nEn7tqoaJ
zTGQWlD-xPoQF8>*Y1Z<3rHGycB3+!1>=kDQ&IjrY1G2jd^uA^y&Ed_n?FkdeQEvhr
z+G|FKlT9g1#FR3?gc^5`r#q3x<kDkAIp#)`vDc8kBp6VYm;udY?dBhAH=9cJ$eQ;@
zhOr*GpM5urxz{p4o%{S2@)pBgO=>RFpdLjHdcH)R;xbsb=~tomfhz1<Q|6q!GKHp(
zrNO(EDB4|#R-NR|8}=33z8XyjZpl;C%TctUK#s1Dl%u1oWNGSc8JgnH+3;l}$%(V!
zS-!)lRdOh~G)NMrOHkeOA>_YDoUVBA2BxGKd%JmuvYm5M%r|Q<4B(;9y=edKH`ZM2
z#n}gc<Dy8uNeO?U_Ol)glkdjr^lrR+tQ&8R>&D`P{O6Rlo8-t|oVKqQ-7R}DTF`|`
zKRCy<whK#HyE(?WsI5ae^Ig)4Kd-l<fpaUy|7t<wcP+T_SPO2r*n%fNx1jmVR`xx%
z;zZ6;<@>eaGrnnEk88s(!vx%YEWjVfOZjzTF;*28<0`!pTpe11PdZC*?dKAHJyn7+
z%>tC%D8!>#0$ih1fx$-QsGu&uSr<#uWo#)Pb}7M8e~NKTd@+BO7vfjbLeyz1K=0U0
z?A;cQ<9CGPm1U7=)E<RRvT=CfcLJ`JNWpaTbnY9=!j}Qrc#b_d;dctqSfUu4RZG$N
zpa6$V5@Nrh5a&7za2b}M;=}^<waUT4E3)v>v`my1WT5egP?#kV3K73T*vrcNQW^qh
zvO~Z;E*SO(zk_uDH*jWaAOsEygvEbegPTYoY}E~fkkf%+?HvRQ*9JipdyFHdCxTR0
zA{d@XgnW7a@k~5;+Q!4*wm5jL9tRUMV?l0nECh<i!b*o|Soh;S4Bi$AuHz!$)6@tU
z$=+s?jE390qhZhb7+4t-14jenVPs?yENJA+mS_Tm?um!{>}j5B9tV-G@8P!Cd-&24
z2}?%3=Pb{AsFRC=U5D69<iR@5DfaqErb25)DjbYRgYG36u+uIRmc7h`TV2dJ>)4xo
zBMZ#8WW#k}kIsgCXi_SI@Hs`G8qUwrcT2!qz7z~El)^$AAsiph-raD1PVdNo{!!d*
zIG^A5lr*r^N{7G$8E{mKH6)D>5O^#H>ht;bwMGD;--U2+XE`kTQUO!Bb1!a9HGjNS
z@N`cZ6!!@rmv_^v*bDPPJPjT@q`<k`Nf5$!xv~9;kZYX;%bk*;(>eu?uom<9a|#?V
zN`bQ#$&mRX8Qi0iVKnnwXm2t|6~%z-vM4yXJ_1|<*cV<B40<xbFyAE@b{d4h=Ch%k
z%@2cv`cd$CS~6ViO#=(X6yCi`hN;VwAznTm8u`Eb=xhPp$`QiV|H|NBY#EsI$65DP
z2v5z0FnkIB9A~clx=T5?C4lp#0=Vs83ZI>_At^c;TzgZ%Atwjot9iHVd<nQ#momo+
zA;F;xyf#$8!IPEXkX#9~Dk~vGjPqKa?A5+d1>2Q5pH@%}A51@jao9(gv!3r*G2EwY
z-T)iJIY%Bo0{5qm#C$Dj6z2}W%O%p7Jy-^xX31i}X?bjTtbkMc6|iBQBF_7%h$?H8
zu*XUn4eqJnz%w-*Wv$M;mm1izR|EefXrQB{CXT4o#5Of8jQvj+XT<2C8g~IM9L;>Q
zUKe*b=wZ2`E)HJ9>=Ubt#}xJOvWXrp9ioQ^ICH*KpoRYowb1>wJ}w+(fH8A<`)Y{+
z{_r-y3w3&^Z>NvGv4)thGad^pCZNB>Bph~h5*}#xK>fbi>>FOjJ6antFK;7eT5Uoj
z<{OKoO}I;GGjBm`!_|!^am%?=nAd#@Km53fs!q4it@0Mi-+zEk5)X0O<%ejf{Rpj3
zKEhjxk8shXC%EC$6FmO&3D&&!N1e$p(75F#9zL7JeP22Fl5bZ<hjUQC`p)*pIT&Z3
zixOvRaN45JsAkrL2NRpn{B$!;Xlh1%(XY5p=_?u?`HDi7ZXB)KjbFa@<H_Is==o#_
zwdM?=0~3Z(@1bE-YB`qv`#P3h6{}O4vo?uM(WW7jw8?t8E`9JYr>3XV>G)qudS}ai
z;w3<(yO4C$h%}u!|9yn$xq!QF*mooO)QXmMT2U(V%~*YFa$vrRZL}s2=9@?6He@v4
zhI|`sDVDRx-aqZx@#H{StluQEueD0xK!Ku;^oRMTe2pW$32>xE!<<Oo(TSe3XXI13
z6K?=HQ3~H#e5~0^vVR6u@m|uLMeIAyb>q!I4{~O{>EONW9g3dhX2e}N%r}NdJjt2w
zQ<s@<)^Yw?vz#|KM{=Ht`KJA)FDY=x-!<l&3Ft?s`9|fqnEU-W?>+PkdyM)1wC<)K
zon;N?*^7BJqj4Tt|Mekg@u73sK6L4d59P2Xv(es%nm^2=VM**;UNeuvt@t57k4hSO
zS1Oj@X1Ncwux9g|d8S2-`)-owk|(o{4sU9{W-oEO(j3wgox^)e-lTWbn`A7!Np!%A
zOk%xg2=_EMI&tR>cQz;V%%+9)vnlKFY$_c#n`&3jq$&^I&t#ABQg=7{C*ej{!@2jF
zcQS>{H?@{7wBemI+1NVMG|o<CKXxJ;O()vUnW>)34s^l9fdYE$=w6&HnY^{7>5FXX
zE!xtFr8eX_(S`!_ZD?D$HOWO=)8T3x8i?ghhL3!cy3EY#V@(Sma%QTAsDXE<g4~Ey
zS$Z&gh2+1OD=RgS1{DL1V!nyF2{duXM0&%Xn_$lg6qaO0w-n7ta*-(=xo1L}!ts<S
z%RYEdWA+CbQE`bO?Nc)3?2-XpeWp+P?7O+n+RawZOf8JlrJn;j^u<Pp`q+1)l%z$j
ze>ACNswSCi)*!=I{59O5MzUkos6$SLVos@0NUAaoJW!@9VPk3W4@EliS&`=IE3)@X
zfhyfc({}D6-Zya+#VwE{Z}t`MQ<0^-v(j|ZP@2v;kEAgP!zompyZxf1sQIBJT{<j5
z2WJkUWBTIsOk^-6UK~V@BSfj=&H&Ec^Os*E_2aj`UL4Mv&86gDcs{5HZ;1awan~OF
zE7^nB8@lk@*)FVY?!pqzWQT-x;Rb#zi0ej+f^O`w?!n2gyD&Ve3zY|c;@iD#m_DHm
z!}D9w^KL7Co8O9&BU*8AWDB+}Yr(gMEqKGU1xLMT!6Kbj)V<M)((SFd*QyPF&1=Ih
zPXVqxREn=JmEfK&#puaOdz5Q2ZpbRe5mWLoMJEqE74k5sIv1z>%ESAA3-HR)LhK%2
zgd_Kr;=<HYjAyoy2r9vff5q4|zL>S1Jp3r1$NEeTs&Yr^otv4MIU^h+Ji>9dR3vWM
z5``x>$71QB1f2OY88d6supuH7Q>JF4QCTi7o>IUYg9#|{ClULnrJ!SL3SR%6j5_NR
z(Oy0QBa%Phr{7sP$2}eGm#5(l%`{w_7Xq$>LP3c=#j16z+wecXjR}E_gkV^I{~ffR
zegmsF2ExyOuR*HsHFyoYhIh<2%4Y)MS#KZ&EDM4SyMsVGIstaRPJrySi6E|!2&$*z
z!O}V&%-_YqM!h(A`aTxoCEkMN)Yl+d5&*r|0-$eS0Oa0!1HGyda7Q!}&c#MS{I+QL
z+Y=2_=f^<Ww-}i6kv%+@lAte%yLUFm!`Iw6kbf5ix>MdmUO^-zghqnwf=KvY771IM
z--G4!XjpJ74(=UE2I=9cu%SE^0$!!T`C}Onq?ZXA_n2|MXTtYoSup*47VJ^T2H9yj
z@O5)OX#Zu;?yMr%8d3yTZkK>G^G)ofQg~s?d=nysdUxJfADIes*$d>-$9|xaRH$o9
zgOe9BV6RFR?B>qq$jjNV<SY02OfP{P);)4_%b-oM0$dkV!iTmB_|4sW&#DA)Xn_F6
zbh6)<_c0`fq=BJ+3JCWnLEEq-NH0l*f5Vet)1+irGl4lIDFymjzmif*fr8RxNO+hG
zi?=7kaJyvCut^3n+h`a!KN9XP4u?X6P<Vbk80=oWgS$uGL9_53oLL?WH%|pan_MIq
zJV*eGQ^|n*eonqjhPr<K{bZ70(1B<WaVdbtN$d-57DDYcW}HMJbXy9+mTyfa!%89a
zM<Fb~Qwks@fT@B~xNO9mV6&1S-Z~j}`!JskWI)KWROq;r4w-|ppzUx5nCG%rs<Z%N
zmhi2Q?|t?ItcPh7gBxVSiA&iKJ~t1RI23XhVF4^#Py~{}l@OOw3r*rf(eTYM3>X-W
zQznc=+2D~Vr^*@f{nF@{AdP(@a;PS$fc2*oP_;#Y_t6z`X{RE-SgnL|)!h4cS%qIu
zsA9xlHT<tb9Xt3&m3l}6R|IR|!uguGv|kg;r|RNve?4q-)Wg%ebZ~*M4&E!`AGLMS
zYMCxNgzI8~gdUz(*Tatkx>)R@jSuYkeMPCG!!Rwp+o6wt%neY287GW)Jy&yP%Ueeu
z9dix%`;!rBt~SHqITJ8N$pllhCbCb?6W^rH!iar#m^RuBzqgrikN9}Zaxg{<xlI@}
zW*Z)AIEfpJPN7WrO|0p-i5DK-!o<ki==kw2KDlxa=O#YDPK`%+_{<|TO?!m?)1ROo
z^No1d6Py_6kE=~y;KIh2SaB{3k9+3e(KR_}d^87pujHV?gj{^Fwh~*kKjW6mUr<p0
z1^0aVg4y$$@p(-%+KPR}Hz&U0yiwfQ#eAcr*NrdCx^c_rel(x>2d9P)q2o0}XfbDA
zgU)g86~|JC&sd6(RVLMGDzqX(jb6HGQu;Gp8oR=r`q*0><OuA0MY`3%e4|SA(2;2J
zA%5*%Otf>R6=^@`jfQS3N|CdsbVJ^qT47D0pR8%=C>shKZ$m1JZD{aC8@h4Rp8jy}
zO@}FaZ<afd!4(J6O?IG9KRCmq!=BgmjwBt(@0{-_+f1Cu@2V5+j&vg5YA322aAF>t
zL6<krpgnxUYHFH6r#HFLtGjL_Z{|)Wb3G|;i6`gDnQwM`QpHhETF!i<_?CNQc-!gW
zr@8d0j+ujdjm<24X_qg1WfuC<LLXl`z<Kh$BJ*hdS05Vs(TC2L`OwyHK6H{jI~O^7
z)y9vP{XW#qKi?C~`wugH$kBp#zrXSO<=a=R**xO%BRVS1ojcLIyKumV{#)!raSlE-
zXq*p4cCvr?7T?G$=hCS}{&+a|onp>?Jqn!l`sYnU6TK<*pf?#$_NM+XUbHpbi=<Y2
z(E?}Y8}=1TN6e-_m9waB=qxh-ZzgBxJZU)h{AC|=<NKByrHi>y-(6RlB<4!bR=AKV
z>o?Mt&a^sm23e1v!JA=DwARRpZpS#1^?A<bi`kJ+jV-mraz9a!EzO<7JE)wEy2U!%
zO!f~a6k1dABHn<1&DtDmbkC%T!gwqG#T+6#3FcXUq7|7$6UGx2G$EC?AnkaL^n$-e
z{*ypD&t9DNcf8**Z369@GJ%9Nfo6VXKk;HS%6o50wtq~hcq(tAZ5mHXuX#&^cczLB
zjp*DaL;4e8Kx_W$6FKk}!-a9=$lA@bv3eA<SeJ~S>d@;(ZR*k1rq4^XXwf}QaxdW?
zKS>Sto~cvO3N_llLzU85w>eonmR=uMrtdAx?@!0lgWZbsBS(Su2o)$zQh|2&%hRNL
zqi9;R9CbCyQiq`|^`uDCa$9LCdN_hEdyb&w_TjYStrRWVBt^0FC8^$uwa{L1iYXgR
zRqw>;`ust(@~;SmvTk#G6X(FM^<zR;FLs3dM%6RF(EL&l{@c=xNoGBGtFIfs-Rr`}
z)m^xCS{I&l?!wT+oLhR^h06|g<GO3z_^YZ5x9{#ko1<O$?AcE|Huxvb&TU0&_76W=
z-iqRSt!T)f?Bm;7Fif@ugEPM2hp2D3U}X#1{At16dwBbjea5#tTk&0=0INH=dnT&{
zd)Bd5<6Vs9GR62@yBHNc@=$Yf9<I^O!vmx8(5obu`#|#W#MXTLsaA-pvkLK9QX%&S
z7GcK|{`IaUXcJP5LO~Jxq4RK~L>}&(o`WCRS7W$76MY-Q@YRYi9KJ3BYs%kaUtSD4
z3`)THw#j(-ej4ho%fzsn4;avsgT=}5=zSs@%`>C%^phyOX2{u#U+;0ug(!3@h{D@?
zad<W&iT}Hkv2aT=o?V)ZpFadca%u=%O$>p6v=DfZ!P%*|AuwrW2+S7^fvD~`u&Oi=
z+Uj0I(%08u-1{1CZ4Lzg_&{jy3WQZlgCKiD5Z{!7ApT(j5ceHF7Eb`-^>`5f5(n0S
zabP$q2wZ;$g0Dv)xQu%R$-e{O!qot<pC14V90Q=zIsjx_!eQ2uNbsB;4PKG_P>O-T
zycj5B&1YRwJXrmX2j6w^@KZe=#vG3U4cYhb>_Q|w*b)hjBO@UuA`*0h--B{v6dWsz
zf}W=dpdytDCxxuv+)D$cGa1mVkqOh-m(wwV?=lOsz~@#b++z)A_O1f37*hx@8VjM`
zr3iXoGv8b<f$gJ8!RB@;oR;UFyzN3Lm&*WGK?;a$NP)((6fp5m0iR`Q&~S{qojEV|
zfc@1<g&B}IE*DJMvn(l93Wp^G&^ccSq26UMR=Wat!3Dx?O5rZEfoo9)1Y4zpR~`2V
zDWt&VnMq*vDG_$wO@uuvF>pXR6n?J@fe8M6dc9I$i%1HrN=k-xH<{h8B*A}%$#6p@
z9QIxdhmzEASnnGKiWft8`ym+SD7}NoCvV~E?zfQW`4%>wc>_nbzlQm;32@de8O}XV
zh8er~?sqK-*03jh#@qK`JUI*oM!$up_uoKs@l*IZ^cnP)K7%RaUcjjjFX7wZSibc|
z!BLl3C^C+M>YMS9pOgsN;YsjrYcdojaEG-_Dr7dM!SCF3C^ydl`->S6Ud@^QQJDY+
znXvM27TDbU03rR{iR6<FKmW^tUE=v5cB%lx{R=_sVJqBrmcZ3ALs4z@Fm#U@j`5Kr
zu#-k&Wa3D4A2$*&pC5_dMY1?Viv2em6mUt60%{v6VlQhr%FH)rACyq%unL~dRz;&C
zHFPUd$5KrVJbXYMJ9ns~dZ#*$U8jz|+?Qi{Srf<o(ZrH)ZQNj|gG*C&@N9+-M!wO(
zoBqnU_ntDI3|Gd-nkrbCtAexUtKsu0>iAbm1FLsx;G-Ry*pjb>D<<mTFArUG>==j9
zR}65^WFyovG{XozGhAvk9<>%4;*qy{SRcs!gA26K@|y|@MO3iBOc_aREFKb$!8x}!
zqDcK#jG1-{-QV2AqII`$^~GCw{pn>?PPvPRPTWJ&E%))4{6nm~_6VhNALG0CPw)lv
zO;`65lziilUeBJRhx-fMFL;TnvvW{?eGW#I<lq#CT(slP&qiS*j)2c7SNjFuGvAzH
zzR8>4j60ZbmJj}lkIsC>1ALFyV7^(!eDiohH|}un#|xAHpegfBb?p$c$(Nw#D~HkI
z4<o3Fy_Z!R#**PEWm>pSg<4;z(Vgpj2hf~Mw<Ih{*9Pg_U8JrWB<(Sr1-EDI<_VE^
z1$#@~tVkrwip(ZjlYzW7$=mZLLoC0}=Y1j@=9?M+&o|4NbNcxClli9pi#>TV-wg70
zp#7Yms(kH0hd()xueu`%=Q+|X&h;cP-$e2AvE&UWIv(vrcWRvIqA2@Fm~ZNsZw_2!
zzIi)??hIoevWGJj4Y<+kK_2w5U^Y$E=KjC)US#*%i$+~#P2~C<?uDC6bv1KISBJZU
zIZL&V_c(8~%%Q(O=8(D@Z(COIp5@oMl+L=!I|JTJ$nc>8)_a2f`B0w}_W=%_N8@^Z
zNMixtndY)Ln01>TQ6GxUoI`4t=aBr~In=*w4xRt&O_6+O^2+okk*2xSTQ`?p56q_K
zp<c8*%A2fTd6U!)Z`yduo0c8-Cci7*Bze!9<~cIA=*;4+%~{-A=S9ndy~vZjHyOj(
zTbwkTc5j`{{d}|OxY#UOKA-p0d53g4cRGvhb)#e(W<2gCUVew~c$_uUT;@VXInJau
z&6&#hbF@pEd;Q)x(L~+_cl^cMUn<;5JlKvF7T8iij4e$Ku%(IYAFhe8A=9}wv`W{8
z7Jjg%ka^ZLc!L$~TE$%Z5J>;4B{i{zcT|&oR|kobs*t8i5FN-wn*IUl3iluxI3w|_
z3q9d;w@ZT=W#{m2=65riTyI7j*i$ScZbonB^G?Wp6FOWxo?0ZvbKjCNl^r&sWxOxa
zIn<CmybS2feSK>GJdS>vj3aH<Zbm2Q(i$;cN_W)ZjJ-Ck%+Mku-Ux4*u1T(}-5h(O
zPA78JsP&sFz5lF2dc4EnCNq{s-c+Kf!-|xWra)uHD3a|?1quxvO}cgRG)zyP1j9yA
zFUrxrHL|qnrVNcbBu#c-NAeE&2#T;DLA6c8>EdE3TJlDcMvRjrQ%MPWbX%N!P7kJI
ztHtPm;UIGQEJCxkvVP<H7gP54<MZlXbUFSTGyHzxc%L4OGwjCovOU=5*M*wAzkLC^
zP-S=*zT-UhRO>F3`r3uIF5MVDzZ<PvyYMpqyr^dv-kZzb-`Ea}Flxmhk5<evYsHaW
zEqLZZ3oe?_f|KIDq5r&ZXgTH^9x!1or>q6Xt!c&mcUtlDS|L8+jP_yn@E-LqLHV3w
zw4G9nvr3DwHX<KCyXRr*#5`2f%EOKkdAKJd7a#7=Lw&7$9As30>GFkWd$|y=h!tV0
zA!|3i#VE!4Px0O&+>lv-=S}l5Kt2~Ip2|XF!%RHt5Qc+dL-D}VaD1Tm9tG+#Xn8mu
z9mJAR@@^W=-kgC7lcVtn>pS7&qwtASG^!npM&EN$=sxfshvdFT4Xr2~dOivtJH=wW
zb^;!3k4N=g37GgW0SA403GQ!$q5B}e#%InM!w<PvA@Fil2q?`72HUtmsE!SU<GHNg
ze0dGZJ+HxhTOgPvu%^=y2)H;1%$RTP9|?lJCxhU?|FLx5aWyaCA2*_;p`|^Qit41&
z(DOX0G=!$TH`m@|mRV?1iS{_{bJ~N9Bs7$fmAz6~iEMt)@Avxs@qL|pxz~*@=iTS|
zjQ#mAad$pE+nf(6|K;=6Wj=%}=EMB2dC=#X2UR0;;5hTm<xAOceo8ioY_egRZZ?EJ
zDh0KuQpnz33J)~PVEfq$7(2ZZ(w4GkC!q%3)$@DltQxT4JkZ+QbD*`9xdzI?ZcI72
zC6~d+_Hx*|y8<@xzh`l^0(M*|0VVZPIO|yj16NeT_i^Xo-o*1@IG;6=ZME>{UmZ+)
zP!C<6*T7&;3lwU!LPuLGn7ZABRVg<?Df~7JRJa3~xp%;{kG&_{`|{}76|g+Pd}CV+
zH#b}a?}%FPR;z)^P3J+p;Q}0-cM*1E)<S2$tB}8;5zKbq2Ib{<po4D1HLErd?6?nu
z_uhg%HMd|!*EP76&HEKyb+Avw8yD{{!UN-r5L0*oTuiG$ds7MgN@2#-E`p9nVwiGL
z4A<%}!G?*fyZpEa$DP^t6kiOv*NY)4ss!EzmB6|s#W3TA7)At%VEV>FxKUgHR_X=d
zH8UR;J<WxEB|=DAlL`|*o`v$TOOP}F61X>3z*^%9=m;r?Rr`v-?FE0%He|r!f^=AQ
zCml{EXRy~i6EX^f;8B+Y3)^x*>P9}C-om#`<x&{EsSIA+EQ1m3`T5Aa;}la1S$FH;
z%sKwGGOdTO;Ce{m%vSfGdXSL63j2GnL0|B7=<dD_sXh(xdwwGXd}#*nt{bq`w*@R_
z^hdM3Qh1g18-d4AG}$79lcvey%-gcq7A%X(&9b=JUk*3#ki(n1<S=o%GQL|s9F@&j
zzhS;PJeT*z&#Iw<mpcA8NE_!@=wL<>^UX>f9K2Nr56{%WtlK*15u}4#eRMF4Ggcc@
z^l@CWK0eiDU(RPubRVyYqa8JI!aVNQS)+;LGnsFOYvGS%Ep+hJ#`PPu@m`!R#*fy+
z=^ym)M7aTO9c+j}!;H{+%t#bCnWD^}QJ7<AhF!Y^m=|S++I+)$6+9B>UKoLYD-2LO
zTMrL~>Eg<fnwWY{9UIQ8qjRP@9-FiY+dDR)ZrM>BD~LsVW}Nq_k?5TgiFwCPqyCmN
zI94_Z{~b%g_YYHW<>*wL^ePpLe()no!|H@gl)RFOFOsrwQ@#)jH#DNpy+-uqY?YPk
zLp*Zx6>jK$h26fdar?bD*fjnvKHl&am#*u;7yN9R(uE2WyYbWRPk6=fCoawVi4LZ}
zQDxh2?6@(Qc0@?hhCFFna)kNDU4y0tXwc5}8Z@4L3a8g;k$tZ=>4X^4Qxo13U*Smp
z<B^8WMEdVB(#~3Bk1|^!S;yhi73qv|qSi_$Dsf<*;XR^kzG2K{zWMolJe`*2_e@)7
zI(o^OIt*P1Y9>&_>j|8}a-)1d-q;Luqna!?dhyha#%sHC|C~Ebjd7={ckc9bo(FwP
z^q}8$9%S3;K^@AToW=4a+r^%AAi|S|h&-w9i6{GNC(;v1FIqIui)7?j(>Ux)LELW~
z_c?$tY$_cH<c_>T&WtzuFvt0l>|b9h9OXw=zJ3(2%a5d<`Ozd-e^P%to%%j=7tKi4
zZ~SJE{vFOby=BJfok8yg&!h`4xNn%ZI|tQGqpyDg==TBERTlZu{#C3s@yFh)>dt%b
zoO9<5g>Ih!>i^20K3?{x?r;92&_94`IOi=d;B5(~0KPAAKjG>C8d>5`zF+<5@<Cr}
zPVl8cKYXcts2`nv?n_3anbo8M$cgzzk@+T@`DS^nAHg4A3Kjd%LqG1p*)o|@ohFka
z^Nn8gB$5lBNNMg9>BKkwb;;mc-d0bVGt`rgi9D!R$DOWya-&8yS90J^o1`WeT3O{n
zw~~1?bG-|x4RoPp^PH)_u`}JTCkhH8nkOLgS?5T$7LHWP-2k%F9BI5Nkn9<xk{YCH
z*7J_VBh8&WhT=5FP<@vr&AehsKXNU}KFg9q@-4}!*pd>Xc-PgzlA^={svanyjbuTt
z>&&Sz*^DOK=RO0q(KMeoM*3%&()LfhIpSzS;={%ycXK3{B9ElIi;QTy(2$zBtN6(1
z5j5{V1Cq$lC+&7Ul9$tCAC@jjEz;qBKW$27K2I~$B<W^#I<!-b|6Y}rXAP%uO2esl
z#Bdt6LWNpSDU;r7CF<U-Nal446!Cr-Eg3M3q(kJG7iDQ*=1^L{Lz+f#CcJ#)5K`PI
zMbnN-QhNGeS~YeMseuHoHs&p<pZ#gzwSH9L^$$<m{lSFUyur<0n#bPXQCa;P<{9+i
zf%9L`llx@?oxWhF%@^D`ggNJNFAjXzi_gCI;@^8;aFyOye8FFXs7~m`)BwIq4d%_v
zj*lp8?8d<o?{K&l-<$@%!~R#h(O`8q>i2XZF6qJzFFSG2+BZ1n*)tr&y9pjg?%@RP
ztl6@XGfD5-aM!~-_~*!NlncFue>84k-u{~y5qATp2lH*pvKf!-HRHy`O=u`+;;%WI
zQQ{u^di%9t`kfX$xv&*YKD1)P<y&Y|c?<L1Z=tc-O>{Wlf-z%nU>M(})|!d%o|_1_
zO&8;*<|6!fu>_lDm*d&cD*R%44i)7KG1OOt{{o8e*o_jjY$`+Nz6u;*ScQ^vtFUNJ
zCC(pGfpRa(@!*9Dy!N0J&mAv8^Y!JJu%Q~2kDkTEmuK<gb0KK73&Hxe5LVU;VVhV8
zovA{&9V_JbT_L3I$bmnja>4XmE_`{J3xB`lLeQ2xh>gmFz=wIDuaFP+%r{E|^CA6U
zK0IR1so#|kz<eXIG9Ow@^I@`74s`U-f#R-Q@O_j6$<5i|yD1wgFdK^Xvti+|Z16dg
z4eskp;pECv*i~H$hbET+`If^v_e%J8>MSf{FOR-I`+2n4w=7Ws2^-2_V0Rg;!7?zn
zD1!m1rOY4|;B%k?Vj4<7IkyB>X_vxdC<B#-GB}47V7#mfww7K1VeKW5)4B{<3H9*1
z>H<iyU+5w4QD_Kj;3D_Myjsu<YdTs%bNNk>mv7^Jt2R*RYk@t(YT@IlOHjx9l%dI4
zxc;abR>{{u_tNt)Y4rt2p2d7~x)wq&)kE;WtB@pj4VEmu4%?hsVA$6NXq?yzp~G*%
z^cUP8Y}p7y57mR$@jBSPwH8)3T!aOE7odG-HK?&Bqu^c)R)?53Swou`E{3Q{?5SBJ
z20SDN)k-n+_b!59_JNlzD2Do?V%|n51~KzVl|T$i&k7;3x&VUC<-?(id0@Oh7w#X=
zhFNx*kZ7C&L$VV<>CH(H>&8IqqbP9di-d-vNQhYy3j^6>HJ5dp%EjqWu{Rz3yffhR
z*-X}SgwV@uD$>q{X>(bt`^W#CVeIMQY~4WaM+&@O3SV1GAiSrFZ-9JjRH}jVpRd3&
z-+Bnm;k(uQdT0o!gRSH1AhD_ra@+Wo{po6L=ivC}3lLe%8F#rGP@B>WnlV!Nfwwr{
zm=47`AcIMAviN6=3_77K&Rrpk`)jxp&`%D>&6eXGHF>P|QN_HYsyM%06?b^3;l+G4
zW_@-1^H?2iJ2mmlLv8e#u7l4|2fqx}!MZRVyken)bG~U~)@L2=D%ZiZB|5nAq!vzl
zsfoJ8{$zVi?D5pZU5lA_!Zgu0UlYw1YvFh&ZTxRGXV61g8#=9xwRL*fR;G`2_ebEV
zM}}z9Y=lwi#`wV01ZO-n!Rh9saFC1{!etZGR~d<gn@6B@F8g-|8elo|jkUBc{+gza
z=`m_pwNM=^+&AHh7n{(x<S4$?i^ZRwaj3#<GdVC4J5EQU;`B2pJt+~#dnd8Bn}S_-
zsi?%>n{SKKaFs<S+Qnp|LLKu>N*2ba2yxc7Mm+wY5!c+fhYjEEq3pI7DDHfP6+N%;
z^_Q1e(EAeA58lHAZ7(q4>kGWD_!46R`D?mIKQUST7yg>?8%?}_;{(=i!q!TX${}g0
zJus98x@u7W%^EaouLj+WW)1(k2KBDdrpyH+NM)BHHE7sT?=DBuxXt;gX-IE&BgLFW
z`h5>cfjf(v#_^t_AJObszGJX%ldy_!8d1DE#lDTbe4px;ai)>c&Xj(`ncQ7mNuzoK
zCA^<NgG}7$vbP&)Z*!wtLN{`{>qe_MBi^LvPFtMa$$N=A_xrljvSjWx=FUHt4iCDo
z;z{-6JSl_u<{00no|bx2@|cOVckV<QC*?&JPF{3ym=6{0^(Eh}0d(V802${6(C{oD
z&JOud$yOgy3Gtz<Bp(XT@u8_teMoDFFFmgHqi+I#(uwe=p!wXfH)IBXY%{1Ja0cDx
zY`4}&_Su}7L1XuGe@*c;DjztN`fB{>U;=jxDtXWbSr5{m>&RV(j`ZD!vu48`DD9Iy
z=?3_dWFYG<^Zn`bB!4PLe=@7|rz6){hxzAAFP`~QShFu#AMmBkYketwpD)ET=Un;f
z%l&}9thxA7k)ki(Y}iMfGMSA3OrnI@ljuSM=e{`;b;xlNrNAWWy5vZ=${k5j3Ft`_
z-w_*tma)gTa=R04Eq12$mM)Z-=0dSbu9V3=i*LA(crWvfQKJhfR=UuhL>Jn%gjwgI
zGc6m$IXM1&#<3RT#{QWV_Z%tmKSz3N%)Lh!$I)UNM>>`^j`br)3b_hoFq}QaLmg;4
zduuvX$I#>sOX|C1N$SitVX>B!&YI1(-ImmUiY0ygAfVri1+-0ULGdrlxr@@ABy7!S
z?ON7u3P(}JSKbF7Z%TW^OsM9eF`YCvrX`z4QdW%-_3_SBAMZ>}jTk|fS`6rui~${W
z*C(~zdUPjWm)!5^(2Vcev|@-hz51g`j~g^-N2oeE`KeL4C+DSDx0%<VLX-X|(+1XU
z-p*1YI;BXrdF!p+TY=tf8Ah9T%F{!E90|K+C^}>)?OZNR<%fpQmTD>Tbd)0h*^*?x
zb1=0&l;FJ$-pcGAK;cdOX@{^MEmr@B;Y0smk@+u_Irsx{*mu0#)Q8yk6%RVI&!+n`
z-;=-K5#B@C+1`s6U3+o6dN0=UWA*Mc=9PcO&S#%dIqNe{oc9^kZ}i|HgKkVH>c*u{
zyKz--H{N~JjUCCX-%vN6D(k{+|2i@6QwMtU*En06p7EXO8O}MxTb94?;=ljy;?UP^
z*w}mrV`AA)?AU_xi<&TDY7@?MX~ZKP4HyyMz@4KFD8fc8?%%{am`ynBcr!Llx`9FA
zEhuf*iY<Ry@&3wNSa<XmYP0vou($yi_lWV&8xi(+i17BELfmz?5a%Bf;omYbu02<b
zopEL8#kZzDvjSBAP=FgvL?{|lgg)^lI6PmBu{V?W7MFwu^2w+e%sa#iMd&n6j0aa0
zVeGkl-03C8?^KFI7gb<RO(kYKRAKl%A!OedLPDF6&j&*G6f@H>-^`5`LTI`Ws;A_@
zV2xa8%*%y??7J~>$%ActpL&*`2hI2Mpx>~3IKX_fWoJG}?#qW8tli{n&j*KLxo}A`
z7gBFzgNz+(G^4X2aYQzhJ`sXsybxsf3Srt0A*{Y81f_BzOz+S4FE7?^_!*L13a8D=
zAkCu?*2Wfsl1nyx$jgQ(qe76j5W~yW#h^Z|6vjR&h1`K<@aASIRC6~^ux2T&QYnRj
ztY_HmE`fv<rGPeNoPjC>Y0g(&Sycr&`P`43e+kx}sDtM{ydRNO$=x(X;H^{y7YnPn
zckdkJYn+3tX6Ha5@f^UM8u%N29vXcv!rn8N;8VysuwPRN!=6_{YhE=NuQ~^=s^`IL
z;RTS1V{fq46-YY6eS5d+Aw%jaJlS{+KEA7GFK!KZaMtULT{T>=u7-gU=Rnj{1#+_I
zL2pD29RG6`2K=ao;`&nPd0q@hIk!KQZ-lqoIm5PI3`JIA_+}&qD}OO86^cPHit}x0
zMX>Taf9xNMpgF7vbPB|<YPSeB*%iX0U-_`{N*=u3oD1gsIY{4=1(UVXA(#c@t=CV1
zbIM8Rd=(8z3!*@7O(ZN*iG<@hkx&{N17@2OL8kdM$nQ7}cYdCMRnduXjz3?PIoaT_
zCkH|s^I+m0F*H6ahNK_GAbYDA>KclnoBv?jv#gkVtBb(?W-+)Zmve7YIUMX?3H`OI
z;3NB~P<$3X$Z+P3?{9B&t6@RpStyRK0Y!^iSjD%i;oi;g@zo9Z(li7=R7+zvcL9#|
z{0%{!{cxhJEPnkegXRgcc#|3DvW6U<9jl7JW~kztFjX|WtBRM+)zCgp4cm3qQ9Wun
zZe6N`1|>=uI7b<CFKFONu{H(==-}RVZ9H^P8*MJ>V9`|_>^D&d8}qf$=C&pVNN8fv
zcMUYM)x={HH1Ym6O{|X6#QI`prRU5!)tXqquhVyOmmq3mp`SMH$uPivDTXLmZG<D+
zN8)-H6I`{x1n;@Bmb23oH})E1IN!il%8WpXFauoiQy=ey>Y@YtgGB1;sJ}~%yGYe=
zqst~7wSN<)MjS<b#aMh`9EYL1Be7&#B%bHz->FEf@=xRpRua0uPQmdpsW`VU745Uq
zFvlPhH%Db+67!AQ$SnL|FGQL9+4!rrk+*}JQ2y&Zl>5s)fZX@_<HtP=ak`5JoxESi
zJzdpF-%&R0JNjpS=RYTY;KKc%QI7j9aE=7IE|#F0Uqi@n3bV~DW%|sAruxqsRQO$!
zE;wjWLXZ~eN@>&O9|n|sbp+L{vm+ySdm6Uek;+RQ$x9Px8+$8Qh9+?X*K9E3%#$OU
z$?t~}XPl_gYdop1A5Xo^H^Z23W?4E@%VKBt9y{|b%$Yte<h^j_o3Y&!s6*3@=1p>=
zJ3HJ+{wy!Oy>p`;=9`_YF_`!>+bnP=<F)Q&5bI9w8$D<o>o@b6Z>pU=X-Tvv`4_V;
z(c?)YnQt!6nMhZdZ%VzqXv$3=I<?uKL`HrzILwD)Px?^h9v_;&!iR#l`Ou7L)^`$p
zNSgU3{)rC_knp9S8ou<F`DRtLKPj``;>x!yxuof|KZkpS6K2r+JnpMGJ%b!`rqShQ
z)>B6KQ^z`Q`ljMR0SX=z+U-aSZa7kXqa%%MaHJs~%s7$`r1sFBEN1vqJwHxf{uJuy
zPe$hcG~UsVZuk09avgUOC;HO${l290(uUSwv7umQv&N0S+@0u4Ye)Ff9Oj#;4U=dC
zcjL7gT9dhzH65Q~N6wt%Z(C(YNj`Su<7!7MZ0+d$2U|MBdrxzh*pcebabz6AyBVyv
zb&Pf)zqKyh|HD1R2Cno|#g#@d-(0=!LQP^9N*n1yB{D8#Xv3TD(oQ6rN#rd;I<pO^
z^Ohr>T<J*R29DIoJ=R*a4wQX)97RVs(2{%`diR|Fzfg6c+O)A`#C$W0?@&+AS<<;A
zOR77>TOe!sbuw?(D_Bx|fq<H|1hjLp1w9Qhr`zegA<{6KF7z8s>s?3jPMIm|F($Nr
z6z`4%8dKpl?r|E<eNL?19LY7LzE2})mCXq52Qr{BMf!B7Q;&XY>(RBzx|F>|hmyJf
zQjn@eW8yWbmYIF|1a;b~qejCuRY`LEaGJYUh5m_@si$3uV&s(A7p+K@DGIdo=`bqp
zKa3s>l&5ufWa((43|*Nwlsr78Y5Srf)DkX5i(@6}d<f^lg9g#EI0@QZGLSA651@m|
z{pr!cel+{pU-Y~C8(aE*;%1K@n0D<OPTksvg+5=gnt3Ki_A}a;ea5=P9t^qAi=}eC
z_<s6lOmhE>^-lac_%q&|_8GUl@4?ul9?XdA!S4^cxGTLIM_=ql-P_%`rKTH2`?_(1
zdN)4W--WN!I&tHmw+J4uuyGCVUe0-h1Cs9J1p77|y}T7?wcNlr=bKTR?@%L8HsN)>
zCX7ARh^03hFl=oDh97Of8Tk!3Lah-$)-|HqA^vClo3Ul!4Xp5P!G7$+dGNgza}{sl
z6TVH!9xcWCBgMFudrKQ`iqN#b2(8u?;?N0&cs{WZPiz#SkGKf!kCfo9^n6s3D8RA8
z0@V3gh%I}Qu;qIq_L(Q)ZkHt1z>{$Jpk!Pqm4Z%DsaUW&3!7APu&yr;({B`Fu5>ZB
z6_w&a^>Vx`Eaxo>A=jA*q2!hjgsk73EfPXOrVxBXh2TugBKblX=$-@TPv<~gb`Dtc
zGcqFwVov0M9BTrnj%GvEzHCqoWIxT?Z1}r68(jEl_%9obS7yVKMcJTflnvr4A&dzV
zLi8dbsJaPZ#!UXPy_l`|W-^xlThaY;;juy<jMFHExZqMSQWk+ib}p=I%7)w?A?(<c
z4ik!EVB?_}cs(c|?pd){XqgyR-z(+)Y-St1QrPjV1l0eOfVFoac#Y)uc(qcPZ&U^c
zE6d>QhjMszxCVlgxsUL3EmX#pKrwUG7Q1+Oc{v^w=AQ!fj!Y0-tAGm2N_g|K5}ZF*
zf#H>Ekn=qYzqfH`b7Bd&bd<rerZTw3%oHUohp?&aS6NjEtD{*f=_>$<Ix#%1DTc%Y
zmtp7kD==>ARY<q52aT=OFjbO0HQiM({Z$p5d{+gDC#t}ab7SjdDk0`;Ih@HV1H%*i
z#gC{MV!Mmr1Qo%eA~Dpti{U$aYG%rcp&G=nJwyzqLyEv_a}lh{D1z1FieTMEF}&&&
zK}=I2oH<?qZI1cidOH_>*yh0AaYC4ro&o<HQ(=utBG@#>!$gIXkhnG)jIT$+p1KHF
z<P!l8mPEiet7ypFat5Ap3)~>La7fQQ1B<j0;g`RVH@UK5QC~KkoR$N<RyoWe*|6_P
z9#jN~SpzGAnXDh#?-qkfrU<sEilA48@1DIy@a_oTvCfynr{$I4zNi!`N0fp0%PKg3
zvKpEupN0DlHK23(63pT*qiGMCVD+aPux$P<aOIwzI_?H+RDK85tv#@``6C3)9f$=x
zWbw6~EP6M};(yFHFk2OqkE`POW>wsys)o^~s(A9EBIgU0uqr?aOOuq)Y>_hNL~3y1
zo;C{Ov~k%H)?P+)Ct!^x$}ZHv{#oi6wN@Rg#;Kvh9t}KkNdtfO*Te(Tnz(AACjMf^
z*~g5tB0>|7pVCCjMosRe(!$dI+W1e7IcK2(PMBecVrkA;RgFX$=9>v~O;BopDXM1~
z@z$IX+RBc=$0-IFDQSQsuIS@}(fauBkq%n5tKpdiYAAT6h6iX9--b3}NZ3(aJ|GrX
zmBiwk&5^i|p95PWQ8GRfm(ES(`*{+slTJawnN%$NlZv|HG_)O+j!$AUQLR3cJ9n}$
zSy71L`-OOZeh#kYuB$tuCN%oqgy#0mxWw!RhA6e-`18G}S^XW$SN*{c2mYZ~)<69F
zz8@)F97sLe6=`&-A}zbDNGPd9bIg=!_8@h-bx?y;do<|VH%*dZzOmk?MWHXW>HTK|
znov7}?6%mEyq!ILJ7Z6VC61J81az788+(55(__9dyo<D@11Www_ZM?MJovm5UD!CD
ze#ML@!y(Qz-<tbU_BqoJ<{LEuYY59H(2)e*HDdpbo1z<;Om?Fq@ou#5o*T6>-~5!~
zo0Kc>KKi-S3+9_*=9^)Ac()<bo&BpGv|HMfYMF0<cQ~h)c+$Y%o^*@%#y3|?q^ZhY
zB$wtx`>(S0@!FRb9O7=6!#*@j*M}lJxEE&$-*dM6@Xi$H$Imd=FyFjqzFF4KmnNw*
zLojzp#rRXo1%HC+0aOwaKxHWb^g1<wGsFSZlomkm%L3@Yl>izr%Acleok%{)ymP=?
zT#Ifx(#~c_deY=brp!0zXFAfE@s4!mo;@W^^QU8e{?tbPr0c*LC}V$$Q}?5cC%zPZ
z*_XEG+E7EbHAQ<_Q9n)IJI%JDF>|fxg0~f^23V2994m@lXGLp@t*A=Hnmi4xDSx6h
z+56d%?@E4{b();)=(V;Tb$8m*sa{(eE3l*A+)Lfa+h8`#IGKD;Jj_*bJ$giv?DHNb
z??UYsE;Pc!g<QN`C@M@qM*0Fe<tL!?`}vjot%VPG6G|7TG;thVb{R+01MR8xr7eB=
zVn_X=$I^}!HgxVe_Yse^rfCX}WO!*TIZIiSYP%)vEw`k(tld;^v7{}&mZWcBNp~I#
zsAG$OPCv1r7!3<r$j0PEM>EozGn#6nMv?A4QwlKTuHvmGH2bPCd8iwc%&L*pkGqP6
ztlcE|7}Am05v2dnfOIts$a=ay{Rq>emlt(O`hyO=R??v+0rxtg7X7o+q{}+2MgCFa
z92@UvzUG^h#Bf?WR)uaYQ6`y4C9=7s$X$a9<m9A4b9jqk<_UR<jgzC&Kw0|iCqpo5
zC|%K$<}L#6^qVh5yOvARkeP$Y%X<(>Es!AR?E~rWo&ofpH^UVc@s?)lU+jwijpFj3
zc%=Uiyu)1$UMhW<aQzGJlJ3Rf_j>T7#wUC`^8;?V^bVbuaMzypXLQ-jTTmN%P)e%@
zSFibm&jdX<Gq4BGEaaW<BVCxeu^VS`HtJ$^Hx_4h<M0*T81SbH`_1pdE3-Rs_Tvt0
z-t-ddO`c(`{v&i=dLKWw-p2o?v|?n<4ZND)jLz&KmJ8<_)cHpC1abcIR0D2vZNUDz
z4frOk0h?bp;Piuy_zIhFP8IiU-RFNN^#*oIx8kI?tyuWH34g3D!<T<baL%S;-o6pz
zw0ses5EXLPwE*wO72xZX0yOzufTleKcuG4TR~*Sl-D~+c(XRl%>Lg=?MiMH$O~fg`
z5;00O3DXuNVcUl!JoPRadlsdlT~Rs;$7P|FV-D_6E5Ma~B2;uOMzy38)UztZeatt%
zZVO>+s}LG$gh0$UBeR6yuulkHzq4S*2_fXU<v<|wjRUhyv@i$6X*qB;CI@txZ%UbO
z_A=kx+>s65%r`B}HtSbq!*S-DNnQLkKR>@Zh3s7s!V!MjmI~p<WFhRa7ea)&5S~Y7
z!5D8LjQ)@Vuh>KK;YA5($BJN<j0iG!<U-lSZ03~=7^is>PDMt6#QZ489~KSn+oPdz
z%PH`gmJb(S7sC3Rh47NKo{O7eVYNdnsN0={tl$`!B1nY!0}{bYIufonhr|Aga4;7{
z!r3D+5Pd2RWc1=8xgs78%sR!p0|{X9{xp1POoHOKDRAa;It)LU$^H}})aT^GFaDUU
zMi)cin-VxYPz)E(vv0Z~2R6IqL6C9|*l6d2)L_n?#TP*@=eyqD<UNTa^$-$%1-cun
zAS=5H`k$$St5H=@$2oXQ_6E&Jse~!D6>v4R4Elmf;nR{5coI|$x6T&9?_os{6(j}=
zWij}?6hTLq2=olZaAT<$CbNcfu!c313TC|HVt6n}3|=-OP#;kU`LFUpH7E~ix^lo$
zhCeTLnUK(v2ET%nVL<}_{KRqGYjgrG)<ylFZgabT1gu;W4o#!N!Dw3;Ycb(qQ62`@
z8jr$C+oP~_K^QD}m<s)ynN5#mL&`(Wz1_`*F^Sorl9U4}M++c4M+`P5V)&~lhPC5F
zFy<`Z$P`3S(<XxI)68Zc#c*qS5qwA~hFz-cEt<!@Jlun%`I!5xADx3KnwQ`?_ZoRC
zH3F41f|k`SP^xVMAMOgWIrtDR-FXaowJ%`rie3nF9*7J7%HZ2*SzPr>7LT)jb3TOi
zn`^45KR^|?xGUm_^@{9sQ^ZSSm2mQICCoEa#eVD&PX3{VVH(<)@1~6zN3}3JR1<}6
z8W<L-j=P-HajYEY%L6p<`xy;<a!Uhu{nEf-)^m3J*1!%WO?0u*M3W7g*jTQKbAM_g
zp3}nn>^okYrNv$q1Kh>=tBb1*(fzwF)>vs_V38Ic8=#NDe)@Q2f&uCo8(@5wK5nek
zM~kWYxV1<JPn=Rir%}UkoP-+QbY%Tz$0po<>?oG^oJ65zEVeF<#GM-=al)oZ%x2E{
zoDhlATN3e%OfnWvNx_2-X}F>^4d>~m<A&r+&TwX;Sx*)Qs0wk-b|KnU3Gs#>_ha2?
zM(c?;u=U~%RExfc{|4R1FNgcEPWBH582rQEtN!7hdhSpi-j9~Q8c0QVCFle9nMt^)
z(u#R%q?@irH~Onn(Lr^p%F&=(FZm{QR+Hb;G-<+9O=`-}p(8FM=+DxT^!<Sy$?<#J
z{TO?4dTmb$rS^2a!k*@Eud&hwq(#g(0sQs;w9Q0;%r|-+PITbJc-nY%JWaQ8rY*tF
zG(zZ1OPO!pui(2>Ja^yRoj{j{aVDJma|cwq(HrKQ>@RM#hWX|g^UVnEF_v8DPOsPT
zU+;CNBXREZ^@2MsWWEs^deSJ~97sGkk$O*0q~^{^boJ0=l4o7UpYJ?zxxO@fF>?*G
zjKm<$hjRzst%;nY=5JHk?L(FkKIDGdhep==P<6WxY5w)07plJGHJ-Q07jic+cg*Df
z@F#%+`+-#h=$usmEnm(X3Cub+?D-AN4<P$7?5FYbq~**v5zIILv39eO`KFWkW(Yr5
znQwM6-}G;@Czq-I<ml{Ae#|!)m~VUx{prC^U($54r2$PgRPJL#Jy}*1GRcy<T`ft)
z-jaIL#&BmbXR0S!(bm~k6wQ2dwaAKcm8|Iq^UVO>MSZm1nlfFOX(rl{m!lnTJlav~
z1zR%bAMdt>4b9}f;908HguKJ?-=Hy6v(=IozZQ_}8UZDBSkUj=7IeMQf>yLx(BiKa
zbVgA?L%4767k|G;nSi|S2}tS_^Nyt@Da>X+O_(L^%C@8kKO4#wS<@)q0>Ay%nl^2=
zCdZ^Plvig%rqY&V)NDy##O%EZw<N8loC6=rH>zI(ax4<iw+Z~3V?kdZnA4g^X7u(3
zZ^3qqqQBhbH|sxBI#q2#D)J`eyVRJs%|_DYp(APiTq9b-+Kuzq5mXJ_!yRNm%P#8E
z;NN;Q#9EJ@FV&@jNF9<W)27@OEjs*Ild9S^$iGpYT*}mFRF*1f=MJaYb-dyBo;^2e
z$~4+liA=f6&snHIqG!XXUVa#%nLNo^$x*VNEaeZDp_ot7H2d=qs*xW;_BK+a<tRyu
zMh>P^!v~R>kpx|&ffPJ>0Lgmwr$*>UOXvN?(Nllpi?u&dw&Xi@=zl|Z@mJjM@dfwi
ze#ZF6?1R(&h+CuIqxJ;ein`0)HSa%Qv|SHAKKhAoG@mfU`V(p@_Mq{S9@Oc0hZ#m)
zDC^ga{bRauac(!Zo$SW6>6{Di>Oz>@g_ahbD2VOA{#q~b&7fy^@L@Y1=)Q;ct#?q(
zuoWv}Zs7GD&FHtU2}f5nqF_@aPLpfI0dpI$LbCx6{JD+^udk!=i|d#;zX7+7ZA4Go
z>nPgUfVxgs@Y?+<l)qPr?^jo##*Q)^@Vf*hHWXv&GcnEy6yXGa)>)Pnp!EHGOpMLP
zsdMx3U~xWbm*?TK`T3Z3I3Hh132}yQG8PO@!rF#Jw7HguYOfMe$}$OcFDK#5D$Z9J
zrlR@ibhJO1i3i4KqyCCK9Q?NcJtV}q-LnXPJt)FC{GJun!rBezW4@mgf;#h!4D(G`
zpb)ICWI@i4ObE5l0F_JWFhD6C5~|anggraOp*h^YmIE`Av*G=zY_Meg#)>^Pk<2%i
z%r}0_H)EM?bouG{f4&K0zR_DEgfC`7&=CkBgI_HbvS5I;5JoX4_#G{Q+Nu(;=e>y2
zgG6B9p9>1b+3=rZEO=~-h8fGEpy6;NG`B~9XHz5?EQx_#i(_ClYcR3O@!+sH4g|Af
zA^l_&q+~{dt#Ty%&WM0Bt0Q2tY6SR-!y)QOIFuDe!Dodys68GBC*<N`a9TXXOgRPP
zmnLwB?r9h!nFKnrsc`LFI`m_{>2ECriw*g3`E4PzabM5|<}WAfJW$%517|K|fk#3%
z2$^r<`tyyhUIg=(aR!a`B8}~rVYgmA{57irMUyJXolpf1ldE8fa}_*Ms)B#qSu^`k
z1@z+{pw$b?z;1FW?6YMJYH~4{9w~wyPsE@$Q4G(R5uW6Uc=Ldnk2z+Tix?J0iXkgr
z3}ySp(C}IWwzEa>^k^Yy>@R@p?CB0{&IQF`%wAu!pk`$TteKz6-PDP2S2qE)_QZn!
ztQgj(B4P1_2pDP{&Nsy{C>;<68)h7bxjx6BH0UT4$sdJ=>!Gmj;Bknm$^ieJ%%h{S
zVde(b=t5by)6RyW8aYt0Cl8LA<-@9W5$q8a!r_KO$T2O11--0m<%+;NMhq9a#gH?L
zw_EoY!&ct>FpOZ1{mmY6*J=n{at`|Kya=z^YhCw&_cX#9A)@vslm)ke6K_RynmmLW
zheuFo@(echJ%j8yeK7XiVANhNiyh22S_|d)7R?<w?h1J1sRB;3Q$&+_iWv5vA`S~v
z#QEMz*eI=vk#bu2rdA90RcqnQ5-t9kMGGa^N4!v519vZC?Z#3a4KJwTHWLlJ`$8QT
zyjMr5f&6`+HE?{p2G;j!ppu#<Iy!5jvrrR9s%qh<eOkCYMhg??Fw2No-%-)TYlAiM
z1b_cbc`ZC1!Wz#fJ^X8;kBPqeoN>~}Lv8x_H&-7+a<ws=ceEl`4ad&EDtN3<6`zk~
z|ION?c+Ku4E<Jq$y{B=XpIamb^0VGE67Ts%;t!ul96yP_-8~Zbev826y%BhmJ;jsT
zGg0(06I+@yQP`J-Dj>v-fkNiM`}kb&27@<t;_l>5bZ_azoPJ%n#kvaz4f=o|9{j}!
z?SE*q@*mE*@(&eM`%%Bw1IdlEtm~L>+?`eF_gppln5;&A*VX9zE;TCY9!@=vCCQ~w
zlH&3tY1VZ~GB{#H+YXJSW3NVXUS||pHJVfJ6LUKM%bXPTEhy2|f*hAvP^ifmx=}Tb
zjM#^hzK&>BJkdDreB+-L^&J>b2l_kHh0V@%HHW)zI6GCd-Gw^2!}ty}&h=jt_y)z>
z;UYIu<^JN{uWs~<x2(Ond+xS3dyJR4)6vcDH1mKv1tz%Dk6L#c%llNToq3~bJ8wAT
za917gjX&3zM5l@O8@i^@tw3**XaCLM**@&g<$b39J~VE)56xq~89IwQ|2FziI^V3a
zSigC3-iJLz{QAj<#>o4U_DJ4!GWI1I-lAyT=0L#**h3TPKzT9&)MCfo!Si_Ea%uqI
zj<|bf95agtZys>3j6m9hG<dgS1M^K6v(5Oqj#T36NMrOJ=vulx-8^JZcfp^y0iUE;
z_j&fgmrk6vCFSj7>F5?4`aHyv%$WgNTbVPS3uxSX0ri{4jBs%b`7qy9&0@|uZAA;1
zZ|pBw(e4q<6U;ZY8?8xN&xSUG9SIEVXj8r|)tA^(T!S@b@#e<cdMn<~wxYkLV@QJY
zOP9A<Qh1GkT4xGq(GLsqId4I26&5u4q6LlkX+fR*?H0^8kC|_9E3;0$fJDqUgZf$0
zSZ_-jx{H0n>=|CJX-zYP*5t}uv$?{WUazqx=e=Vn|K=FF<0GIfISV?JV@ckjoSB+x
zNhdWeiS7%?;Fy3cbp&*Ej|By$nbXUoW~4ZKG~HY~iq!H<N$;l#dH9;p?QCOO)!&$^
zW{#xJ0%jZD2(Mu6CgSu6N_}ZSB4YyzT%pf9<$9#TUBwd8x>RYSL*Hj;(~pf>bTUYj
zrtfD@S)e+_tX8ACC934hI}8_sRme1xy*EutbnvSpb?YnAB7X%E92`cC1@g4>njDS3
zDNCByWvHQgC>a+^)BN*8NcNQ!jTk0HOSmh%w0964xGzCgodf9t_xd^L3?MIq{>+yB
z=#k1_ERgz*`C6>s%>9m!>iY13=U0@z)r)5E87o6Sq0I9Ss5kFDqTD;Y($s}P$2!sS
z-#b*{em=k5T{z@?C%(S<1~XT6;H9%~ac@&6np$#JDx@11rFF9g(~a@o-RSb73mx3L
zFgx%qI*xgRx{qGqj?Q*ez;=AM@E$gv;?A14Ex0`323m}6Mxj{~9=_jzPl0ooB@MXD
zxB<ocuH%n~*HQiKbvzw;9V0aw@ZYAZs5|63o?CeiUwx^@tQS>y?QtdEU>{7)kuua_
z{pQlTVtjC2jA927?&>N)jVt+xL-MgDBoB|g$;HOM+4z08khiz8(UM=ca(=?HJ_!Rm
z`4*qbJ?#;RxL25nUp^<|w1ezD+>ne3^(pu+HVr$gGqBoRh`;XSpru|uZtO0=FUlg+
zO%~ynMj;4fw}W)Z4$unR1r2leLRDlCY&v`hK1FB1%v=2a?wA2O=Q%H|oepc$(%_hU
z8vL1=4EL*&;Gbp^JeUv-zk3hEj~$1>dHf-Gq<sXg*l>&Y=VbUkFdcLS>Cl~(1s3<R
z;6EwO;fxl7EkC}I>99I9lRJvDAl@+t)*LT^w(JrxW6$c+ej?Dd&V`+^eB)}0gPNfy
z;He3l;3h<JS8xOzt%!te#xW4*76WCSad6>T9Q<*J17pKjC|(-{r)ES##B=7Fm<ZUj
zI09r?leqajjPusvFgGI#Iwa!Y*!DOu_!$R1A@MN6>=Y~-n*bMHo`$wRiJ-GA6=Y(W
z>B2MN7BgTr=Y}Nt=RdYJ3Eo^w1~Y|BxZcL>7n21p3x%+DT@JMQ7r>-Lg)mrN1cg`2
zAbw{#gfy1I^r{MoJ6Q>JeDiAl$JzBh_WWF`1pQH!kjn!_(Kh9<a&#G_sFs3OzY?$-
z#{QxyoY~J7!yfLCeikBv?v)}4=GO(?oQ)bKhPyUmsFfCjPL2qM&k@0~>Oxrbt^gdn
z^I=6n9{8E$f~5!Z=9f&EZIlij&y(Sa*BQ8w91j_me4BEKh5^4L;OC5RFtiJUmTkx3
z)th4=;dBgsu09H*&WD0{YA9HY4h5O)V-O=a0~M}9IP+2nn^_Ywn#K%^3GCNN;EY-x
zNHg1wWc{l3V?KN~Du8LH3gE`keDI!J2*=nv^!F8itn3lmwzwGP_ZEY9U<uqeD}`6#
z+`Ds&cRRTAXQ5RcEU##Q`#T$<TJ|Qie7+6m2XJQ)_eM+EKY}4o9)ZQgM<A`!4!?K4
zg5*9aeDX#XCo$jLd@YA+ygUB)mI7~gDx%0&5f?cr;@Rbjm=>#uEu)oiQUG@x=W5}e
z94%aBqlNG1YhcYBbu39&#~=-L^e<Dz*p;d{C`KJk`0pK_s^hb->R6+tfqz9BIOYNW
z^FLYF;hR;z64r2}v~ZG-7K&DK7F|aRH?dde+A9tG{Zs=@A~mt=mnIso)<TasElg?E
zLC>vv_}W7sT~=tL{Y3Wc^c#-eAu2dDS{av~RL034H{y`rp*Z->3G}-Wji$blxRM#>
zD!=nj9v6w5Arh6-k7J=?Fvcth!g|*rH161gmf?GF%7ZX8`<#rEUT5NByDYS*%0fk3
zA&%Z7#5r;IaXa&k2lLHV<{MzXNn*Z<WxiR-e6x%B=C0~LbXfckEhGQo=zsmFl%IWe
zoMBa#plu^0Xfkh-%cBG-Uj2z7cE8xq^b4D({>24rc`xUPBn?fFBw4W}?Ky5lqKJ{a
z!8DS#N12k-F;kkg$CNT%O{tztNq(*={hDJ+>Z}pWdSXLby*8BUIF`JgI?+I<@nkOT
zLM=t(sfO=UA(_rJ>bf(ned|mIMXuD_;L2V3t~BkhEA2evMvgz+=z}`%SX;W2IeUv;
zm~Un>-vn=QCqL$!6{p!}%zP6%(}T1FJ?MA@XPr4KHN3)$8dWAyEBD{b4CCGGyHjYD
zt~a&Cd6VW7ZyNK5H#e1hsL|7hrp@$W0P`U|_To^y59Jp6&>7AkwOrwDV`h;JMh=w2
z+Eg+#$>3^xqDFg~e8PdOQyqxp18Ai_?~}h|Uk~4#MzD`&iMc<m=;ujZlKipA@Fo{~
za6<PvlGO@FvQ%><1^zKFU1aYP^9{_hC&zy5H%_yoPJ27{zmBD!k8P-Pxh3Uk3P|Fb
z1^bRHX!mDx-sKcf;$TZU#(Ik!dsifaElEDXivBR)&;`C(u?Odlg*8bsU%2b?jfnGY
z$CGVoW2P;o<k*t*HU50?y(_QQiY(KuXdUy-o)7GOnZaGblLVwVO+c#^1ms?2K?-FS
zlyTOAByL*Jx1r2c%r`%oZ?>NmP%-n3M2~>n{t2j%y)zN3xwj_IlGe0X(mSyg{aj&9
zKa;HKFu#tPH-?r*T9R#sB^l2!C%qMBR2VawoF<u(xs)ZTl?%vhHFH&e0maU<ptsx1
zNx{jC*4!OMbKdd)M^>hECe(yNI1e7}VoY^O?7R74#Cv2$)RSRI<*ePDU_Ej2F$2nM
z(Wlk&`ZUi=k4%GfX}Va41P`^T;I|fS8?HrqBQ?ojv<3wkt5XYagoo*<l8y0jY6TVE
zhEk@@LELvERHR3(3S{zU7<rivqY`g<vRWxe@tb9-VUrARfDfhA_0mM_ExuGPMG9Xe
zsqouiD!DL-45K7yPwqf^b!PxA>FrN8fBI2v7ki5v|KR2FU--WE2U`C9hU*viq3yFT
zIABRHUcTLfO{~qBoc@5?I=r25r5h{ubYa2dPV7JHEn3g%K!HOSe)a4`vG@(n^y<Lg
z9dEHXz7uO_JjdKik5OXPV>JKz2=DEAgoX}}(DX|?eky54-;M29)%pl`i`sG9$aciO
zyZGt%O*E}+!TZ0O@k2us)_-h7o6H8h%{vD<a~g2wuIqSt?=^h6_bLWGug8w6tJrwu
z3btx;M-6w@#9uj!#=oj@!cXp}d0UCiS{3MdvJ3}FmEsS+Px+RMF@e1|u{-(pB$tnc
z7J2w|UoMVL&cXG+gxFq{iSHo;$3I9#->E71_fHbG*Ct}!x<u6WO+>{-i5Qoih*ma9
zXr`3RT{|i0#@UJmFVfLYJ_`p;&&I^rx%j>(4?F+m<Jh1AoHuwqy!*BdydMWa`|BV8
ziD39}YCm)d4#KgxLr^s(0+JVGz>vmt_@SN-xk+iTTQ&{mGT$UI-y|^KjB$;I_a6^~
z@y5fDu{#)4)q-LC#UN;SzaRdsKL+VfkHJ{)<J^gS95PH<BZv)y7aLE2QsGG$?ZRwP
zlK?-hv*EzrY=8^wXLb<5{(pt=yd(#v%*cQ^o0BliDhhr*jR3PjkuY#-1gN&MANET$
zDC)<+x|}!&J$e$HwNJwK^C!S_Wi$lzMZn<f2uRo!0m`!@z`QgZu9by>(fn{YmmCE_
zFJd8bVH~V}83&4M;-P!sDTthU3jD;e;65}K+V{r6T$OmRW$rntl>k2Nr$OTT8JO>s
z1kd&*LvD8l`(-j=zH<gVpOeWs>`d@_ngzpNWP{Y_0+^mz2tlk-6s#_VDV*zS8eR!6
zj#a|g>Pqk_tptw)m0<I?0*v%3;AT%5^!z9V=T{|=a;cb^s0fD17D3}i)@ZJaAkbI@
z=^7%4autE)1LhoUF%%n%;m2nYL~j#;*PB8JF(`!O83pi;^ZHv2xnKH!{ZV{d%o&;q
z|9ww|vaLznua^MI%5h-b7z5jfM!}Xh;Sk{;2E{hVVa}Rk;L~~(##$bQ+MrO7_z(hT
zB0@mCGz4byol1@U#@^;a@T*|Y`G^$Ic^m?+UypKc-!b^}EEMiO3gu4fJSaPy2hr>D
zpgW24b@rS~H!p-S3k%_LnHc7>Kc~xs^_$nlFl&7YtQl7dFRxUBsnJ<zdvOkC@4gJ9
z5_$hzu@OFGw8GewTi_pc2M#QK0EONUV3h7d-ac;!pNxmlsrV4Wxq~p!`X0!X+=KOX
z4<NVp891uHhIuWWFevH+9M|cGYe*5d?oh<@GZneNMgbky$m0**dyp@d$LFuumpnxs
z+cTJJEY$d}t%@OIRB@823g$|w<NabaoH<Y(1K69hn;EB`d1&7e4eYzEfkhD-IG3~J
zy{1|yH&Fu<=4qfolLj`YYT$Ebp$=xCiLBwI8fjq&YdkG7+PEe|7Z-81+-|BiT38Rq
zzfV;###04b<}0IclQN3BH}anip*Z#V2|SSzjSZ~RJYwCZ03)%+HWJl~kE4-UFlKHJ
zLdS(c80{2<`m#Z2d1w#rJ+cS$F9)NXQW{1S3-PsX4vuTe#ok+GczVZW{JOOc_pHBy
zSC-b}jcHf0$mJUMFki>_k`3r%{0Wx@e!>aMdNDew7mYsk;wbkonD_DrM*rr{c#WT^
z4L@;P@K1cA{0q~ZeqpiAU!2ZbPwy?xx326*CKCseY`P>3D3hcoM~x`?gc0rSG@|nL
zBk78*DOrv+r6HqDxg$iMHh$Bmp`2qm#9gw_T5Kp$do1Y(un*(86P*s`od({VzH`Wh
z)+V`78+ZKQ<h)L!sVkY=xw2=}mHd~x(w9fBG?+ccmzZzPv3~PvygP}QZ+6dhr#H+u
z5BBqY>Z&{0v5z<K&i~%0Ceoet6G_F|i>|m%qA=kk(&b!Hh|v^UvSkXLub4v0TfOP{
zLvJ$v#gC#7?PLFq4D-!~|9EGUZ&Y4!K2)3QLm5XLsKCU5c60vcaFRXQoVKSR3400|
zV^8r8_H@|Ko=T$Z>C%pIv?s!mqL%=fY5LP})^COz_|xuQKRODmCnY(P%sbw(V7~EW
zFK(X?bI#3ir1XG2E2;MMakV|2;4OBwA@(%osU4NF?h^UNf`rd4$d+G^kG7yqpUg?p
zeKcvjF{l4Z*?Ze;P75RjWVO$dwx;n@Vnw~@tw@{sX615gs$#xTV!qKfwV@&Lwlq7*
zmgG}ydEd*1=BHZIJLa1@p%sOjbFbba0g0yy$Zn#5Cb$a7X$a?-L>83CIi~8B7BnQn
zf|}$6^kj;F*6tIKWT}7#v<pa!^_$AS0%~Kv$@tHbikZ3A@%?EA-_wY7q-Zy5x>ms-
z(|7g|b0+*tqJZvon~{u}85xC)rr|J}3Ifb%o*Daln5*JHT9A~N1x=Y_PIZ4rQ@8CX
z+PB!0<SI?bRLz99$c<_Dqmi@^Msj}Jh#vMDl7*KcZ9X%CmUbCXqqPC;+M!P#m3s8+
zvo86V@W#{(9g+^#rlILt^y-`@oom*hls0urxT{8E+Ei)mz2Ox8T!nsoRi=}Q%JkP-
ziRRB#qzU^Js8h&n(;`ouJ#v&KEl2CrWXV%ohH7+&lDnxiZFV0*JB~<EPMaj{${9={
zrw5T9e|@0(-#`jYA3$lGor-AhM^?grXut0d?wtP%dw4(UW861X`Slf}wtc~K-#%m8
zI`-B)`iRecK4ASlzB_TQQPr%Ae}BG3pP&w$IqeNfe|(Kq_!@b52Rr%Rv|;Z{9Q@-6
z1}uGo-_Je9sAZ3F_WMUzx$_Zzw0(qHJ?;1>uN||Ow&P39cD%m69S7?0*FRV8;-d7M
z`0GduK0Dluljk;J(TPT!xs?4jGp}P8>ongdUgzJM*YNk0tC&<+kMkw!G2QJ7E=;P$
zJ(n)x+)HQir_@<|tXYj)w5oC7u_}!FUWrSBD=}E96316m;Go7rtXD6<?+$s`>y?W;
z_T^w;a5fJ5&3q!CiN&we5am<()}DgKR!L~Vy3Ot}i8xO(5vLAIL{Gm&)V-6457PNg
zbuAgSoKvywPa1yg%D@$`vhZIH@6ssd@`g?x2JFhiPrd8lz`Sh`|0oD5UIanapkUB9
z4~90w1K{d@2r&FGe02+hP`NOeTy-2m{yPqznhwL#<A-73#KVv>;4r+bI0Q>)90ET&
z49Z)BVVX)XNLK|x>8c>8Td*HChwcZ%y8RHOZ~(S#JOD~=$Cyh_z`3LoFupY$0)qC#
zfA@po?#y7weGm-iW%onqj{P7sIRKJR4#1@2M__te2$WZb!pz^SFN|Q#!9Nlb%%j1z
zFB*<+i{pUCNf`I<1bj<70bAK4dv<>Wj6E98{e0{lX*&)D{lZ{gU>L*}vDc(82A&Ly
z1=opju;5G_1U+F5XI4BcpA!q_#V6t40KSLqj)O$mc<|hF3OpAk@UG-(P;WT{Q-&nL
zbC+Z|F)9_-k4lH^&8aZtbsFry#W`wG76dQnjM3wK=79qE=u!e&LrY=Z{c=#xt$@|e
zm9TYhC8P#c!aHy7A!P1pc+9=SMP(3kqZB@raKBz)F?8)Jg11-1Fp<QN$eK+Hv(l&f
zLioxzrR$%Lg3^#<a9ZmaRI%2w(nka@cNKz&Juu=%?xlIoor|1x6AsS->0eopfEgh9
zJO!$h6Cuv<6fEFhC!a~tu(vz{hA{7)wmlB_^^d`fSx4b}bttG9g+eWJO_b9AkaXVR
zSpVN2N5jfap^WTJMpWmV3Z+8hQ&xLVDk<67dsdQ}B!#3+O9LfZG)Pv3N>ORj`kvqQ
z`{Q-#=H|}ry6*e&KIb`N;O&iQShp`4PR)vD9*qL2gHfP&i5a$<|K^LLp-evp)KA60
z2magqm=z11tKwjbMhTdGD25J^5;)ENx{cK(pf&RpEWE|Ly-Q`Vko{n8BhSKXt#h1Z
zcplt1Cvb}WWk`#=3TY{~plMwVDD~ZipQrD`%6IImKlvDD-+uzz4VZBrKZlQ5E%0_s
z3mAnxheop|FqUtErFl*8%kViooYD%fkG+KQ)^2!r=RIit`U3No3gX3$vbfJu7L6~;
zV8UD}{A?qI8YiXj_;P6+=_ZX;F4CwdE{775(%5oK8ts-y<Ed0R9IvE+s^=8&!8S!4
z`mKm@tkVhKriA_{l(5xX3G2D*YG8yCM(Zh|m%S3soTY?Di<MA#w-O%f=G%wH;WW<8
zS-)~TTII6V=cNiZfinKQIR<BGjKPhU<uO%Q9^bu?<6d$Z6fN_@_PepT!Y>&eE%xD#
zKdDH(hl=32&2rNeR1!<UvfT;jBN2)=t3z-z^Uc*!As7}CgkBCo_&qrY#Vo_nV_O*B
z$*aNRdbd&eLM{GVco$oK+{LMdb*Q<#9<#^a<BsBc7}wB>vAoCMa&7=cq`qMGtS@*a
z>kHl${EGG~2GOH)5P$s|#7xB@d^=?bC$%%*82my@<G(nw{2y*%zVT3Dt&WiZt(+=A
z)w)9DY(I?qB1I^Py(8ZHL}>d2ak3b#O0rv2sWMWP&XrD}*Y_sSC<%4SXTGuO(Ig2~
zOUhboNpxo#)id9Oa;H&KkPU^g|5)jw4LLoqAzS8~R2N&yXTI6Vd=vcKmQ2L$NJ?N9
zX-%9(0W^z#G2cvNzA<OMab>>A*~g#f9A2Nk*%YihhnC))LuyuYDPP5b;>{hY>JWD{
z$T-pIHBMAm=tLjhJ5lpR?%#Yrk6Ophrx7!FX5+#d<Mph&*~8xOi&N;qlF6iHKbaz#
zZ^o5c({I*L|7*1(MHy@QU}{ZYnQv5>Z`Q1^rrxvGtYe)*DHAx8m-`0tR5*ub?0@vG
zX%@|`n?dDDK)?FT>94CU`7Y9>F5a`$^GS@+rLZJjx|gm?5!^L!?Yu69)asHzr!H9x
z>QcP7F{REireqCc>VIuS5i<;E;yPxW?}l`S`Q{6=fWL$(b>*3o*ag-f-!P-$%r~Aq
zGs)OxPB&F7XbI16yk=X_3^xn9GR=bKD{@zNhB=k<ek!h*`NnJ_S*$Q2-~X5|SpPD>
zJT>|sGu1KPGp=GEQ;QM3(9);F68dz7IY`degj!Pgek~?6_apDim~RBx|CGPUl!90<
zoX32l5o|_-17=jhUanI-%Q@c3U%$+qQlTa^SYSxGw++Z@ngK;<GsAq+Cy^3EdZB7U
zlXy;bK#g;6RQNHE8PF6zJ*sbDJuzq9EGpEYJ5oAS!yTHHEn0+AwWujwlg!vNHOx+f
zWcRC+S=$7fu0DaruTi7iY*k8XP@!}I6|&P)rk``illb~^B)dn6_Jk^uVUz;Biy2E(
zlX(A{Ax{aVqv_WjIr`HhOTwbE^q&#;F#IP?CVQml`$0*XQYJyd*GG|EtvJnjBu0}u
zL}@@klnNY0=#G*wZO9!-6K{;5Uk$@(y7n;68W5skCEPnzDL`(#58g5H59a<GLiK^4
zxLWTAj!XWAt>Ry~KW_lvihstQ@Q;`&(2w>BAMmL@YlfTH+nvC(n~86*XhaXD4R+y(
zjxHQ}@*1TpUSXYVC!XkQM{}okEIiYO1xwm+_4`)5;n&Kq4XyaN?*-m3c!A!FU*My!
zE$9j_aCFae_I5tQnBWEsU-%HO@N2`8wtAFPsmHk{b!Z}Y7ZpnHpvpiEuDex@?MioW
z_O9DlbF2p6mQ^E(R^j7@8~8fwI_4Qy;`f(V@I>-ue3Ntu8@^XySa=0%rYksGxExok
zDo3%{lQ_k`7(cBl!sM6}D7*JKdde1J(V_x8xh)^h+{{By^TYUVST5SRu;!*W8wby4
z<1ap*ec8BVJ@2SAbFs?!Fn8bNVW(6+zNtEj!-I~YwfJ$2oqGaRHH&b5QV|CDY=bk?
zec-E=4|hcdz@w3&U}P8yWnQ6h@K7k!l!ZfeSU9}(3y1Qp;UMZ74l5uGp4)_hr9&75
zEe?aM6=CpqX*j6w4~Ks~Veo297`!|k!hHrI5WhSOw4%cx^JW;lkPL^69ekhZ@i3D2
zO>Nr_fW*}#ND2yrZI46YicKhp28Dv=%}~zP3xi#T;c&1q9CjRthV$hyP~#E{zP@p=
zY*9SKXUD@h#Z*vynF3MnJezrv0Au&=1Gyimu&pu)`c#r(6(oW3xJ0PDo&bVP36M24
z5k_Yu!&j{oaQl%49^VdvX-YbbxRws*n$qCTq=PVNUm6$<rNP^vbl4+yh|2*pV9laT
z=#I^TsYkQn!j>FRSepxece7z}U@knlng<rw^5Oom0$8`B5M)@JRJ)4(W2U80ZB+s-
z$Iijaqvtv6ryMSdR>0c%od0WE0e6DRVY~t7>eXF@A3+zOuk<{mvi_)1>nu!EIRhQs
zXBy4^qjvi;7{_^vFG3H1w;}J3TI2YAJ`REg;-G(0Jov@O!w0QXP|<x7#OLsC$g2cI
z?TR6obxV^%3t>mnQLybe44*G&!L+b+7<+v`#GgumIqZ#!+mZmjBjaJCKpe~_X2*gU
zn4=T}%^A@!U>Xhgd!r!fW)y44qhOOl6kKbM1ixWX(7q!IHhhSJ5eK7Twpt9xB*efZ
zy;#`&Fc#Xi3SmJZ=Qi#xhEM**(4NLUWy4Q`-DPGz&Kt~KRR*nAXF&VaS?FM&9BwLy
zJ@%I%eBdf)``v_O%Np2cRR<gH*THqYdRS5U2o^ni%$oKmaB+PzC?0Eu>{Ctf-Jl7m
z<{6kieFos!1pmoD2N&N~F!Oo|hQGUEU41Vo>3sy1$S)A{=O-MvB#Qge#IU1C92cGz
zN5x_(>{=j=6P%@S-a=`Nu#!f{n^HJGRth6g3gz!h;ik)Sn3KwU!`htfU5dEOObL6I
zDB+QKCDe3Q!g~rzSovHL{lt~9N?H-so+x6zpc1y2Dq)$70-jyKc{V#0QK3^2*DLe>
zs(m~Tuok&CQVB;XkHKG&@~oR1jdpkC@L!$`+Me`6!#lBfmHB3o{65?|kcv}UQc>MC
z1yv@dpp<k9>h>h#Xp;mqe-(@sYl6}8RS+ItAA~)ILHO!=Aj;+kqOYqz>XrKA51#-G
z!$6Gv8i)f~L3rIf7(ZMI#_bk+5pwpT2JgLP%|7C;if_16;S1*A7mVrriPzN!(ev0K
zK6){TW0-Hm6o#;i`DS+qdsw~yu+Hf(x?cE)*{mP%QWc=}%r_#;H>aixkVC&93H=hJ
zu}MOFdjvI>iIA@uX8~?hrAf>;^(7N%&c_KP<)A@->v(tIYC%R5*dxq7D*0;m7&G5|
z+h{`@qHSnomJO|AzOi_0L)SmB&e(~4%FH*vnQz9m*iy2H9qsuylWdu9!kBN|ow$E$
z<t!@NIg9>dzPZYL<6AMCB7U;wW#=5Ke>R7n+s&m#+_x0e$uk^DM-rXtNV%&V=|Yer
zS)@9WafTy}<4z{uzYcWos{?7TbD-E2oE?+Pdg)9n3O-;(2a>HQAjyifk6BUt4J%S^
zwj%eBR%B#gO;^C0+80<;;&Sd9-E2)7H?67enl+u~my+?~oNX9_wDjjRy5VWT-fZ>}
zE!8D)Yh6-S`hoG|zTwKjulW7JSN^+yM#WQ{WfJ`vJ-+dN;#@zj+1QUGmT8k0?`6JE
zH0H`^BkC_Qq;5;@&R%Um{&x(>+QEo>w2aw!&)SzGJU?N+xy5{=*lI?4E6nL&D*Kks
znv>)oa{^%tx-M$L`#y7uU|q2-Yh;w~no&-P87;AxNE?>%TxEd?r$O<Kh-XIU3XSPz
zhA|xxG^Tm$xTm{KkA`09QG~ue$-Ok73+{&Wqrs507Z{Pz6C<+Zoze^DqUByD<a%!+
zr3IQ%)kQO!e$b4zGS?7i6uX=<CE+b5^fS|tVzUjXMA(4l-_xf}yo0iA;7+Nd`t){(
zK8<tIrxiB(B$lQ}>L+yR5$RH3r4HS*(4ig2v?)(Un~wWw(WW*{ItrR3m##s9AJq8@
zQJs>bC(y9_+@m>4jTYLg(uV*QiY->A#>Vlq<m))PCOM8?O;DoE#)_mqNrBXWJHqY8
z(7grn^pJg15rJ~Fl{0T_F3He@7t-|lj}#@UO3@yaq}|RE)VpC6MR<!-@?J42OBJOr
z=R|1nUtya5btIKN9YIRvBbdpD(dI585^@rv;VFXDT_`~Iro5ke^&8)34B;N1pD1|t
zJ3dzVhNJU1t516XRf|7ip#Dc(f2xl)G9Pd~`@R!H-f>pWTRh&}gWHdG<B9Ap9Mk?9
zITRBkC%?jyzD~@0*N#z^?RY!A4dv#xVc6?doEzARTg+Rrq4x#;%zc3y>|fyEgBE;o
z@j2FDGw#!9!rxOLW5R-m=*RiOEB>p;^87k<NW6;;g|+OXxPx0B)}Xz2HCElJ!Y$jX
zuz6_>s%clF^z&QzV8Ttzx4({`wqL_}U03iz_+?bOcL{sGR$xU$1qO|-z?zfgXzR><
z3QJC+;q)Rrwd({bW*tZ0ltR2_eGKjMj^cw4M^MK-51ZbzSL$942Jy_s{7^Q=x@BYE
zyle~%%0{7g*|_u|&vc?$dt81P1C%)1G%Fu}DHq_;fMa+$tq>2m9>)(CS;ynD4ceM^
zL$$aM<RA0_8C5^%jR}Ty=9?1cn=8y6wgaK;iwT4C%3;uN6$TDd!@!sMrhjf2AoGna
z^NqN77@RT)hjy<p$e14n;e4O<b$-y8;miK-Jz##w2kzh64N7yo!AO5M2*vIO&n3H|
z?uIw~+p+`BPVs^r0sgGv35AL)p^!WxjQb9F=2H<4Drcf#X;uvHZ(`uzqBzhw76(P4
zaj<$;9Go5%2U;odQ2H(oI*k+HLvj*W1}DM1Zf2Qpe0mdLYE}Z6WG6tqR3dESS=2C>
zL^$>%0b(~Mg4)d__#m7HpLuTMT6h4ocBDa?)IqqoDINYBeh5lh55c#d3|Oj@1p>y|
z;PxjIUZrM1)xK;Ha?XWY{&_HAZ$6lZ6oA0=LhwlEEFqH<@F({K2=n{<f?dV1)!;mw
zSGoW<Uzfu#lM48@yc|TAmP4adIqc^CtVJ6xz>4hi(8N2N)N7pE$9d8p<xa!$Uu7WQ
zUkW?!q{FORaqzB@_dyrq;OWCS*xC~Z^R?q4DJ&kUz8!&Zlany<aS05{a!z!A5$Lj)
z?Yv<jhz~ys(qD7o<oZnTtxba$U8!*Aa}o?xB*0dl18u5~g?|k(yl3P&Ok6a4kcftj
zhod0OJ_;^KM#1<mkx<wg$ytw)@T(va#@9tcm3$O*_(y?<NHlmd3+<W|1HWZrp`s=h
zc8w^6Z1(dvb``<H@x>s&yaZ&sN<bm~Bxe$pLM!vyU&}M_tokgJjW`e8-!DSlT;5Sl
zyarcIxPQL08q^oo!NFT~kUOCs0>kP-arz_f@_G!9E;Pal$41y|%pK+>&mh+B8Q2Fr
zgS4^SSNg0OKIpW<|8=?>Y-k4YFTL<)SwEyj4Zwln-$BOoA2iwsqFSyH3RMZ?1Z82o
zWW;PUR~pasN#X0K5?HfV3eRkl!p9wwnEp=^H&{sF?{YaD>#m4J+@CN;TM6UmDWSfT
z67HX@gm?ca;+%BeUqvWj40m2h`Z5QdR>U$Ld5pWNfM!h!n6{a<JeL&F&~F@8r?Lk5
z824grP(qb|iuhrLJZd$_;YQBqskM{gK07ZANQ=c7ofMq^GZh2grJ{3FDxSNLic))1
zarVwsR9%ybx2uD2b$k%M^bEo+o0w}J1!9S65RNVj#JHS5%$?_tOV|73B$oi}TN!`{
zwg;e8cmNh12*57|0oZ#k03&Y(VB@G1d}y1CPdGbckY8KHkN?0UNqilNgLtiA5C@wF
zF=2>P=*JG>685jQejLJmyZ_*z<6q46=DieqtpwEsXdCm*>nQ?M%Y2jZL6EL6-#9Vf
zyf7O<R~)!6c@$sUN~n^juo{)Hwp5PuPySr8pkK|FRQT4Ce2!buCC<T-V~@CD7<X&#
zup#>x8+w**L$2jEG^^Q$`k8OW@%-k3Kj+IGwWYzAww&u{N5_B9q~EN+898khEn&W?
zT``M$7-o?u>u>HnXYI`;&b#4EUg;fkXxoQ5B$(tt8s{7+gtKABeRiOkqaEqdL`U+r
zcclLR9BH$hBaIo!nZCTcnh@+jhUJs!jfo|F5i+L~NuGZMvQ{YAl6q%Zk$N8ckgi%$
z;}g#N;Tg^pMQchhv!=7tSs%R2ni@Co&dS%CW-#B#@?2*HbHrrMtjm9_O*Z9yIR0TD
z%Dw2vYhRdM_V=L_dt~HO`_L(+4|l77Lg~Jbc&+gx1}yxDhd#EW_&_^;d((p&J3rv=
z1$uOgI|{a|8qhs;19D)$^lwXDvYw<%Iy~FCs9{1UJWZ&J=O((1ru6QaDJc)1NTYJi
zX^{D55$kPMHJH;wKDYSxK4kB>jyZ+0AIkNT5e0C+(~KM3?R43IUP>EL+z1nL+i6VS
z6O3u-pAikS*C*F!dSqXxM*^eu>4PHctJdn#O`hk3+|j4w;RbZW){vYzi*FQnVd@p}
z|4C~iNv&kPbB-wutu-aHG}cFVvag(XQ1A8_P_3^4?}PLyyi}i`AARbt)Te8C`Xu9_
zPm=t2?bGz>hqVsRl2~t3rb7#Cb!c(5HuX={rdg~hUZto-J3=&R$tw-&B@OzLrcR6A
zOrURO6R5#gjb2r#lJXZ7+NPsIr=69lC}=z>9v(+eE-BF<XBN+CRiJG>V`;E|3_blR
zPhKPCX^4GOGfd^k(Lt6Jxr?DHMw$$Zq^O`)l2p1S==<+cR3$Zv-l&MvMI$k?m?26|
zt3*hYb8kHEjwJc}BWU|M_D+o&##$#KI%^<AX+eT?@}L0S7W{|q`+nnLqan0-^8-^S
ze#Zygk-6Ob3og0-8HdmLgc~09qx_OSG<yFYUq<#~l<qsc!NpxC{CaTy<Zj%p(}g~X
zuQ4zCC0=fLi8EewVohZ``b})bV=6CDep(v_zih={tiQR=pIi0z1y0R+fitGRz_N-K
zRL*^l5=zauR<Q|PPd~*(zXmMVsK;?Nby&}{n+Bf!iiXr;Mfn|E?sOYxh*aZ`s4DEU
zsKQN=Rakzm3VV;-LPv?4m|}Y!kNaFh`Hm}SdFC>HeRv722P-fuvI2!<EAZKoa(rf8
zj+qNiVlke;z_{Z$ro0dp&K$#&TMKaIKt7ILcLblmJB+Gnxp+M^2gg@tW69KPeDOI8
zE!wm2gJ3p}S(S}jer03r;~aeWIv4eLm!<E`9elU*QEKogwtg<ao}6R&Yv33r;5L}w
z><QQJ?SjP@c60We52T$BfE|W`FoJc)`x`?c@>3{;^E@Sq`6k06474YQLH7)v-7w!g
zUK9pa%r}Ap;oNr_2Ju_NKxkeV>|wrHyWbBU#QK73;2wZ@AJ)C@hT~e^U_5R&Yjt)*
zBkhKW6W(AwZwK7Dx&tKMdcoy|J0atsH_V*1mv=g0;N}z#PnyEPE+rf`wS|N6p(xlF
z6A4#%juSN_4!VcM!Ody0@O?rYOkNlVc^~6p;_O6-%1MN!mlGkMcT;vA{9ZnveL*YY
z!I5`kV;~-GI>p13%mhe2nFMP`ro!=+`$0tE08IXR0B+7X2oX-{ApJWX#@#;zdL<cf
zQ;sz_8#7@|V<vdb&w^bV*|46qM{8!}!KCT=uy11l<f|9Lvfanwk=P0Nyqog~Q;J}2
zTM;~m67GID36}X~5ZreLHtTVY&W(N0ADsea!O1XbRT7l44rV3mV5THT!fLNbIP^Cf
zQpZKZbe$NGzrhS*8wbW=ad0gu4!i>5;7C~<Od1sroA<;+PtH-;mskR`ICs(dPBGXP
z7r{Tr6VN`q5VVfx!*jJ<$Qhjh3*POAO)rw6OePUZW#d8qKrH0$i~;RK(Qx;36uk3{
zf}y|6GkK9P(k&9I4I?2{BNFs1BEe}|By_Eg1lQO|u(=os&jh02Av4aPbTr&(jE4O}
zv7q=Y7MugvXBBW9#&Is>l*%G7G%E(n^CghHmGi94N}(!;J3l|O4=n2}yel~ecb8p+
z6@KM#KIbw72VLi#W+ilp)`8yXI*3rMhtHw)@cQUIusr?{j&Es%9S<7d){RC`t9k+s
z$De|g<TJQ`?ioy9)(lgMpTmL)&Ct&OrkZ9rv%DWRUj7Vaz2Bf$@)xYoXU6FiK&Nej
z+*=@sXU_=Ye{+S<Q+hbs8w#UNgBVtBmB8eKlGyZE5;OWFF;`d$C$&i85<w}P!`hv!
zCMkSlDvi~@Wl&L379W|&;#a{jX#13X=<Ra&#(p%y+0pp#iUQg`R=~rYN&Gl#45kT;
z!*Mtsqx0E&&Wt4eOA#McDB^xEIh1yl#V@a<(J06Z52eQ9v$-iaqCFMgzfQ&Hk5h4P
zek!H}rJ|5mDvoBp*;N&U!<la$FyGiP-)KJw#9qlDd~`SvhfnrLoaT=sR{LYw;s8`z
z6@YEQ0l5B30H%rr;)*eWI8h@I_ZSCaXMG?Z;N4fjr%a4f?89WEKKwPa4=1hZ!>+x3
z*uUim=0*QN<D)-tN6irSn*YIfNq;b0_#e(^zBy9;4~GvIpd)hx$oZ}S2{{N-W3?dV
zC<)QNU&Cq8dIZgp7bmyNDm1-al?*!7Nb90G2{PaGGT(e=zS;H8k`^-Ge9V|e0Upz7
zD)Y^yNj5Z}`KB|?hTNHN=mq<y`fTWzt1V3mu%*{WY^k;0mV6}bsJff`HCca?#e8$h
zaTX0{zR_pCnR}Ca53)F~_x5bk_&%G~dCcKX&^h#9fdef(>p%ytFyAygP{20_vg0{U
ziK-*{X*<#?Ax9DsaHKSxMAz?I(vm<6TF7j3`Xl>_hRi5J&z#onwxpS<mZZ7MiVj`o
zxy>{7BDGs_R+|;ceYK+5%r{A#Z>YG`nifb{ldY-|C4V!Z{ZCj|lF*GJy4^UR&uf`3
zw7JuP7Zf^JSHJ~b(2iIBwc)+qHuRX&jy^lu@rYMDdil2Fs)%-!N^HlL-|grf--Fwt
zKj4C`dK50ka}z#GnQP1}m~Z%uWX{Rq+abG*IHSdgZpItYswu`)%KIQe(TQ}Wl>Op&
z%<1eya}v4DzNVdK<Wn(`GYXBUsGRjX#YU8_X+Up`4Ct$c0ey-!ph!OhTD;Pby4s9L
zbeb`}vC^lHrF!I8tw)n@>5}4YUGm`l*)(w-Dx0iBUTHdX<dY7qnX60DMY>ccqepLo
zdG7R(=S;gyNQO-57WaW0pXJU>cOz<9Z$QSs^*KjLpU%eV)AED*^yj=jDIL})`CL6(
z6~^6C2Y7aMMVkWV=#a3U4*At<Q<}3jWjxlRcXPBz^0FrF($pmRSPky(RHwbu)k$*y
z1lrx9Mpv}d=-ql%63tT~g(u1+BBV^$^;lQzFpi8iD3M8kBK?R{AhGnZlyYPY^`Dfd
z30Fqb)cbNY>$NO%v<w|k;M|)@(sX386wTi$NhcE}$otqRa=yg#$vQFm+agNI{UWqd
zM1(rEg=y}G5fprT1ewSUqnO)5v}}_Qb&eJy4Ie>Toy;D^mw$2mwBM+IZ4j*{|3K%W
zZz#n*3{tlTFx~AlZhHF>U+wP4|AzPBrgZjsoAzSf!?$?I>kayj;ck^XUAQ&nHM)wu
z!d%UlXno`*s$cKKV^z;_XkIfmbv#4spO0~GVIz)P*MQ=a9^t!j5Amb?1GL~yg?!f7
zIC3|g%<QMQN4yCm{TfmI0Pms})}i0@I$U_~F1qt>%4J?HdLO!jlPqpy<$)@6penR}
zbqi-IR^iabDtsA!3*&@tqQ>;=IDgkQoO0m`{!Y7$|CL|D?}C?ba#RIAmaM>S2g_Ll
zP>!!X%kU*0$1R10Skrn8)!NzrbmAyx&(24Sraav5eHiad%*B~HIk<UcHhMhC!k&;U
z{IVtsCkJGqe|;9-UYw1g@;MkIoQo3dBfmE{4~=$l7FZ~E)GRxSZC{V#hUo=ZY_kOR
zAKC;@D%|15a(Aft>jpc5-JtP{8!T!GgfCY@U<30_rDzx|Q{%mpSr`~@4h3b|t+29s
z3v_SU0(;}uK!EaUxK`;7&X3%|;*UGL*Ix!oNz38K-W4!q#7cOfw-Po;t%L;qb<mWx
z9o(kuhSbhoVAZ}IeroLii_jeqp6&%sT00>j%$s`ze4uh(DCE0@!H6qi@R4_9@!!JW
z{@XBk8yE>vH4$+3Sp=9~i-Ds$JU4g}3kNLXAmx5Moc)pjohuTd^JfC=S(5-QGZLVK
zx##Mfcu>`ghgZt+Ai%pGp^NcwvNs+s{7M9kVaXs+kOD@-_Cb~9LCCgdKi0Q&nEUz=
z)E{NO8I}o`XJ&$6ZYDJU&II9uS>W(H3l5lPgSK!EY%67d`M-Rayq<d~zpy91y%6?@
zX2G*tnNY{wPe1z$px7c8eo-deP|JX%>jz-n`V=^`iT&B^Ve(p-2tCpX&@T`PCKn^X
z#xM$A{ELDWKceBHK5JrXW8vh~I7sw}gUT&&Ab2PaX1!-E&Kl;mGpv!`%eh3lJiAFP
zhI2cLK#fj-%c^7G%G#hY1vy|>n-2PqQo(3nB3u!Vhpp_5(w2w;7dh70ERKSmHzPsV
zDH48vh=BT>2za-KpIf^KxH>Na@)9GU`)&jnsxa5s^Zgb^0t7{Z-P=f*78(VkoTH&s
zJ{D#<WrNMvqwt%v1*YHR$6r?j8$TC;=>B4;F+T~<-k#+AAD&6QDdVn&v#@5{Iq*Mv
z0j%pU!VB>Vm^SMY=x@CW8+KlUC#g3fT(BOVG2g5nbq{XG-GdMQ4?vE6RTJDBq4vKg
zFu(T+e9PgCoT*QF?$ZQC^{j3YZRY2)89r7v^UUZuTu^%fOKjS}Gq@9GU49KajC)|(
z__t88g89Z%5dGQ&(Rt-CY%Cao>k@^~f5ULpjFrFuGf6yGCW+Zkxd-C5B)<D3iQCpo
z;p#>y+;1+8%`GzMB_xZJ#>=9ro-C%Xk;9m1IfU<WxO*Y<O}r$&u93vS)67Jx<S?jH
z0tcl<5H|?p7As-&`l^VduPU+^OAa+YN@GTlG;VS8!r2RAvDTh*if^Z4AMcZ5rw8Nt
zWx+Uhdn!t9NX2KXQ?YJe5H8pqghE?`F!f|0-aZn@nSg<uT^@)Jjr}p-)E||W`(xKm
zf86U8fa!Sw=+zp66MI8Zt1uQtN@8(MeJu9oC1ZBwK^!yZ9JW=RLqFg1D9^7|Z;EfB
z(A94^+u%E{lKO#l_kN+M!C!p6>@UV-{l)bP|9C(C58=)~+%-&qmaXEsjkp+nlMo|4
zB{51;79$BoG1|;}%d8<~GT>hE&6Rv!s?m`FH5%!vMh)pIB$%l}QtXL&_1%b+Z<*3Y
zW*;xkM-iT6Ng97G>8zG5?U`oFd3CllYnd(SY_X-3U|YJDY)hKAY^h4tj<Oco(S@Fw
z)X97^iTS2{Ewhf@Y#JZUeN%<A=|t^p`t)`-^(fCFIsZB2c+`RBpL8J8gAR0Sm>tdT
zuqB%kTl&L0ub8_)mYhLTr#gv}hAe4s9{aF6CQ=RWcaHX$QRg6QQ+UR)mghO=!z}48
z=j83bXGIMyoD0l+W5;|G-DO42%r{YdBw|==l4?wwm-pabi5_%M>BdsN=ANR~jc+Zx
z@P=#`7SwcLRdhT4r`nD~Biga|QyV%xX+xRDHf(y?hLhK~<KC_9Xw01BzlYi8KsyeM
z?7$wu9&BMB+D}hC`mLZxNjkbTmf1#(kD9SA`D^Kt8y|B%$!fYZ37J1wiyYI<njFcA
zG<emVimKUDm1IU|Yz@efXFwUB^{KjFpXLuUpiv_X=%fhiddv+dWU2uv_!yAeVnYhO
zYecJ=l`Jgu>Cj<4%Bj<({Bye0D5}HzE*;Xku1)j4X;W5$4qdyZLw9?1D9S{aoKtk!
zf2BupteqM6K!-x&bt#g+-}aV1Ipwf+s6dZKk{<2!(j$u>dgPI+OPVLO$#0D|NiWc*
z4$>wKM{SCY(k5l*n(3vS;m2NZDb}!`^U$ITZ#3!2QcX&^t3jVlHRwgEI$3_;+0C>G
zG=cN{9z0eh1=fL$pT}LAtSe3}W8F>DI9l;jiOS@ZXrrDYEuW%5U+l-ypGDjizDk~?
zwvVPu?3*e|lcmT~8M=H&n$llOk(GcHd5o82owEd;nKz1j){0aAZti4=7NygfA~dm#
zy@Xds();ugv{H6B)d&xxG4Vn)K|_dU*9em2)<5{s_z&I{_>KP_456^sPkgfHJKh`5
zzVNGG@T}Vat{wP<vtvG@>9~F@Ealyl4QKVWyh9D2w>Vtx4R)XIM&19qu(babUgMd~
zfluvtE{x|i(JdJEwHd=UH=#h~Q`|B57*FLl;+>TZSUm9&t{eLhze+#A+uR?udj37^
z+*^-OREH-X-o@*ycviFME_NTji;HVnORQFl#lvgy<d!>}mwFquMpvPybrn}f+``_d
zTj-T@6HgS}z(Jogc-6ZMPtQ4p-3v<b&GHJY+;tgEi!NcG$R!-Xe3LHD-I&ScDEH(d
zc5o-dk(sO~zQ^;GQOB@az5vJe=i`+8BN$_!hdtkO@lI6^wqMUi^>0~toVliYL>8`Y
z%fyy<nfTKn3nP#6=bW-}lx7YJkITg-^}~2thdozvM{xU-Be-`-K61GTMmD&^)OvSt
z2y=%;Ki#0;&kYoX+#ucG6)w(Z^^v42>>lO{{T8lJw#XIAU0q>er88{GaE90WogpjN
z87A`WoAX^j^{@*J2)M#eVK<l_;0}7sIW>JtAT@C*m{c!?OYtjU-=3AA?YI&K{MJGG
zmvvB3wGCFS+YY7<J3w@e7xWM9fGkB1xZ>skxk(=Is@DV7uk-|TW{pr`5JWu<1=Zgn
z@W3Ms_Dzp~{4G%s*%J+gEivF&8Vf4z@o>B-0iGOA04HXfV{;QAL_GmE9*u|cAF=Sa
zCI$jd#DE0E!LR-CP*%ZO=gvf!(~$%(LsOvRMJhC@9E9nPyq^*{1ouTUz&twxu6@e@
z4bG0)x;YcPnQ^xN;w-=x?lAb81+619;i!EEyl6NCrxXu?&E0h9PEQ99*6-|X%787?
zj=?DAjG4x{P;8qCLW=CwV!nAeGX-p1*~9LT1P|vXf<4b}rg}ucfA7QLk83D=xWzi|
z%h8Z@kvU{hEIb(z2g$7cy%!S$Dl1~3khL<VcUdd*CjmD9je(*C#h}T%sr=ywz^Z01
zEZ^w|<>P$8<gO3Yw(N&fCCL!|Ar9_$#=x%7XgIPb3OY|jf~`m-ERK%=H{A&6<XM?&
zdN_n^3WvLk!l8FtIP|;?hn>thr(Z|FEQv_i*A)R$eG%ZyJap(yBvh71!HwoPu&c;}
z{!K^W7w09bu{ZsaL=j9nUj$}Lis7<)32c==35SGBLB#tMxD}rUm&<2i|EKeCT<0P@
zPArGmzZDP}dKqF*T?N_1N|<z!yTR7g!;ahaVCi`eT#X-q&7X%*nbH8y*F1(!qbGoh
zPvCsa6BujV1Ro2VpyEC6s;C(PPd9^k+H<%$`~^5)YGp562dEZxfd1NUxUKpYuHE<t
zPv#3^#yLUU|40D6_X%RD`d{!77C_u9jN-#2@pF(QN}ZC#&EDLB)jkUQtwgXpMFf{=
z%b?2*85HT4L8(!)xI$GHd&^}pn?2}17f52<eo3_SlSIvGNu0G@iaS*#u_#LdttCY8
z!ct+JGhP^LlSNS0Mht)KkieuJQs}0@%;e&Q)sC^KGd%@u&ZlCVK`>@m2jizD!MHXp
z7+>EH#-r>7kJ}N1HzEQtC^-;!tPaFi%L1{@I}kT(`{M--fBZVZA1}H4qtkYOJi8+R
zPrVAk<hP+HT^WOJ-7y%)M`v;@YOq(!;!HB0`jm{DPanV==MLcXiUTOwa}(|DYcNov
z7Vo!x$7zRtpi}h^%pLfF;To(HaQTUi!9Vfo`yt%;<{$bp-()i1{AIo|VZJ%3B1Ve3
zVziUB#kW2yQwMu-M_pAR({n2H?Vt+f@%fjkLaWkLXlK3(ZT+E2HN(|NMMjOLj#uOS
z6K!gfGo|tulPD0l6W~8v+Pc)1)~>fDH&50WGvB<+wWXPbwq(zIvwXiDsY}cx3Y|&E
z0_`dBgFS@~pG~>$vnl`KY)WB$ank};>&G~d$6-D{=g^E&&ct15Li?7R&=EfM{O{Ct
zCPb_i*|OJ|Cd-(R!+KLvbx@%-%T;I^^G(GUGy2Ya6KQP8{hF3^W+V3+1X<DKMk}&n
zzOiP$0b-3YX98R8Gb5E^)~T%Q!F7{+FutlAw<LAr4d$CUO5HfXd~<1B7b;eD;1Bx_
zJTBLc3%c7-_yuc1Z?~a)c^mFxzKMR*hKXC+QJH5s=X~06?%sBc73#ot^$xtE)P|~4
z^ytGeW*0?W_W0=1dox`c!F;oaxn?0BO+Hf0H`6ERQjjk5&PyZm;eN_S4<oYM%JZEr
zL#lM-?^~!#H=K3p7HgfgTJ-5}t3KtAG@#3-tcRIwKu%r;G;)<8rLsqBtbj3%W2Ukm
zuSe^z=u(3ibIowx86DK7g@xL*(^s3EH)zw)4sCM%p-pOIbvS26hYmOL{g`c>gmma~
zoDQX`>(JNPJnQk*p&KTg)2E_GzV3Q-)IgVh{L-e{Xl;sH#XVB9v}x-^&h6vl<e*Ih
ztF`GQ@1(4mZ3OpnKe(_KJ&4kzM}IZw^>z&s;jCg^R40>+38dW5d47Dgrx316{#7cp
z;*T=*n<|t2it!X4IgV}?DluazQcSY~ZSEb*UGHOP@yIchB_~ho)kjmhxf})B%hIRi
zGL-BiO?4?!^s+>f+HXtH+0IcE^;4XAQJnH7h>@v<DD9XfLT_Az$!^(5>ckP8eKU*>
z4hRwd(mdl4q#ODIB)j7ePBq{?@4MglllkU@z)wuQ_znG1zv3?L$?SU1nZ*g8@Vf3t
z=88Uya^W8E54|`#;vKFU{}vt3_u&4y-FUC{H7dBhLMyLM-sQAordul>-PVE<_05=M
z-GtJaPtj%QF*av5qQ7ecM(9046{&}KPxt{|9DSd2B<|s;RrQFmb@;0CE=Etii)R07
z+4EG33JSH@diD;!P_D*>{#BSQUxkm_ZlT2ITWECZ7S30`g7J4R;FOKr4W)GkukR|u
z@6%7AJCtJYl#^&8U4q7kFQJ|2CG=$dO}0n{&WSH)AJ}PB2`WRy)tuG!w*aeV6<~wQ
zQS_OTkL7>!u>HbeOxu-<WsW)M$^4?qbDJ|gnRqND6DyrEk>>FU%jDOyOgs>jg|nt)
z<8`?l+{XFFZ#r`E%BjQX?3#ze@8@BK@DYqEcZZT<cktNm4l4%Sz}(jjoPW7O)J|8>
zn&t|L%quDNE}&uK0)NC^VBvFT@GWzO>4%)bFVz`#W;w&HE6(6(?*d~ExPZZD7w8?~
z4k}yR;r@U->=IfEcF{{A=fYAD_^_P2Cs)Axi7TOQ-8wk+aUJXo+Xgc|w!s78?O>w3
z6~<L=g(bhX!V~g<tC1csvdaTHmw5tcZi7oAKCt0bAY88vfm`9Ba4j<&)}=>){`zRJ
zJQ4#p_ryYeZ9GJCmW#o5&ZM(RfY<B~@*kA|yLLxGTM6%7en&uTK?EdDj0KshSm-T}
z1E*Nlv}}k2LyK70h_SGf=R74+37q$n2+dQH;IwTrBursVuJb|Y=8T+Ie1g0)A^t)Z
zOeoHRls6gRG&KX3Upxd+_tK$ab~?;omJY_=>7d<o5L8^!!0Kr}JYv0$rFj-;_w9$8
zf_-puPXcV)>I>3Sc7goB4$u$uguT0VfZ*wEknZORmB)f%nMX9tbc+Gcmoe}pAO?bc
zVql|Q3_ND0JGeI*0y=p|^ga@zdLzL8ek6oh^6W<|1o~eGz?`*y@Kx9sthVfd2~NAg
z@pUxVeT;@%=~2*_8wmnU5m3nYce@Y{P7A_eBJUWVT@M3^#4wn@ISgu-gh7-)-*yOx
z4Y$K#&5a1o@QHvep8Qx25pel<1kACE1T&E+h+3TnYEkSPznl+mpL5q$`*HYq^#u6%
z6v1uHV&<S?nDmbQR9{bm{p?e)llwwTHl2k_?&sm{o(r&1q8z4HSHPG5F2e@)QZ=ur
z1euEKu%7!&<0svPwVL&iw)-9w4n2UClXyoZ^B4@cpZ!_MWB7gIF@(xB!TR7PXyX}9
zx^^@7tv7@E)#niKtpz++w}Ys72W-sk0O1{-kT~@%#}Ei$pROPd3l~JYYyteV{151S
z{{_Mu|3K_7LDU>Bh6{I#V~hGI{Jd8b{W^uQ>6<Wi$qD1st-`n|R}7bI5y!91qfkDY
zcUY5T@$Ethlo=_Bi<U^@spXRR&p{I7KTBe9nFNNemcXuVaTIqE#<Tq+(S|ukdyg<a
z{wIR!P2xD-Mg*6qd*O`N3211Uf=_qzK59ZRF5w-N^87#)T^fjD*8_2S-~Z2QoS1Lk
zObNvA`hhsxJrMO({PBRQKOR-^$HlV#81LkdFSq)mCu?s8Eke;^aSTpnzFEa=lgj5I
zpU)<-D5)5S!<^#q$J7*z=6<VBB`K)Yl!D?nb8zP130%0M0doTyFh9KkH<v!<nayLo
zdg&>S)Om(e1)EW`j`fKNL+Ef$fS$C9QrbIFic=M%0247%*A%C@mEzRG{e`u;D)ju2
z3RN)M6eg=s67$V(KBW^CX+7&YO$-!iuup+D^eWH{_Py+_)FMq~Q&Rdfi4wRc=@;u=
zZ0A~0N{khGL|M_rbX(G8zB$CR7w=kIx?RfJn}hbWo%J@K>t>Th$80j@y;Q5z9C|s+
zfmUtcsXF&$AJ{ybitLPOqLC3b@!7+tmXFF@BT|YqqPw|9G_t{nCVo|>pij!AG>mgn
zmh$ZB4eNA&n2|bX<gJ=*NmiV*x8aZ_$s|}(HnYxjBUAdRVoG+b#a(yUnELh`)1Igv
zJh!3;i{5wRg_>?0@7;}k%r_Ep-KhSk3!SsNuxj8n{<o<EKMS|xi@$AH)zXGpHElTU
zd>j5NYs2#PHq;tyLuLN@cFA@WVZNEj*Xmz+hNJPf9kaYz(R5=QUK7$K<J&sa$9xmQ
z^O7nJ*5Rmfmo}dde8w=}81v`4xuZ$LN|*X4>yjt%3}qH)|1#g~nWszFm~Xx?-<;T@
zOA?QHj>DW|$U7=E4Fm3qFr=F)hNRPEM8-M%v&8l2@^xLB&mQgXBXnqQ0?$-93m}}C
z#+pBFUaw8tW^2>LOxDJn(xweJwMloYHVNiy)1;5u)WowM?Fl-Rz<Vb7-8}0#uR|}t
z=+dTE9ZHndrDy%xgb~`5vz+%&piQS#`De*$lb*CT`Haw}kVY+f60Akr#%j^SqnhNX
zqe;p~H7IJV2Khv&(?iZ72(zC+**R)7;{*3hne+UHyQZF&DwE2~@uVxsyD7_Ygf2?d
z?4?MdF$yG|HI|NZW^rMKJbB+9O*5a$(Z_CCYWyif0}?VcT1%Q5P>NDrB`MZRf?DH7
z(fxdJ`d1-F5%)#uT&D=_7!ali0b$xaY$OTv52y1r!{|w!5EY9F(aYt6-0k`gy{`Sm
z=h}ZT_x*2tbae<1dGob{!Z*|!^%XlF44_@eXY|+ogum|gqw9t~bpQ7rPaouLJ|or^
z-+F^x><e$}>cZz+USs96mnb^26O+{1@w!?oYTLKqx!h)S7i&WEJx?*H?=k8nHKLkB
z10Gj?gfQ$OJ{-D_(}v&2bt=4{nqH5o8|&~2&uk8f+(qppwWv3@7LQil!S@z-@aa$$
zT45Dhbl<`mS-0?PdL{Z>Uqji6mr(S?1zc@+4i}9-gPl9eaIMKHyf~(mALArW$}dK}
z@kOX~|2T%7EX06|$1wGK0X~Q?!0`D67@Jps`Mf_0h(C%;^YZak!V$c@JP$AP9A;Br
z4%Xby##IejXfBwApI2pK-RBIPcRm9jp2)yA4>GVwGZXI~%fz9%Sy&;OjfbCP<IL<F
zytg11|F9<K+1$fu8hjX^9(D)ARCib|=nj{{-N3iQ6)cNfA;QxYo=kCt<$qnE{EiFQ
z*tvi<^NnP)GrTBuhBax<&>Zg!*Lj9>uG|^$hcifrxxm<V7cd#-4z7#cL7{UA==Lmu
z-R4VSSl&`tT)G@|H?4qw3M-+-ZVmSktb?ES8^OkEGhAD}1!{-3z~88?kn?OS{MPgU
zZC?)<-sAx`Ydv9Eq&Ix7-VGmT1j4cnAuw-KD1`ZhgJg09Xqxc*{O4#eyB!0e`LUpu
z5eK0+;=#}`0pfL;b^gV}v+1lY*&Yc8D<fe1h6q?f5#V$t7P{JFLANp%l09Q##pqaA
zzBdXMOpNEA%>)QjN`%9ck|1zeGB}&1K>PbtkiK&e+OK9nyae+N&uo6L$pXKp>>;<#
zfCrI>z^)`6QsmMhMkO6=2M>bx<byEN^8maE*#|J^04Qtjha|&&@Q`&(dhNlWUg*nO
zByZ4k+72SWJwU;PXIEQ2VXCVK$b8rW^}~bVn?MYt`Nn|UrWlaqxy@ys+lV$qL+iR|
z_<S)6KHuf}%gG4X@G}(VgonVag~9OjYygBT^#iZBdmw!=>z~AS!-m<cpSctPE&RLd
zo(qSuCam4Q8U~Zu<L1cvDmk8w%{Uec`XT&rLnsI@35DCdi~J)R3T>PlP`@<<5|Tqe
z{{ws4lEb0&dpHOj;>X<_3966NAW1!s87v=`ZzzQ1VD6&#J^^p0aOe2@BKVnG4F9f{
zz{tjv;4F6v%Ey<%l*H38wdoxEy?Gug^)A4-2j$Q-^D^8?xe8O|E1~b&b?~`%6J(OA
zVO@AFoEUo#&dzuMo^6ldXGsHiK5vBd8IPg3x)F{HJ%a|9CU7chf;_QiX!34`8$-=-
zxT}Sqhj#e=s~rsYbwJqSPI#2@2Iknjhm_)9@b2*+sM7xh74m(M-2M^P?Hz!WM_-};
zq5wKB5JJJ<!|>*_;V3+DB&NR@i8@Uqxm-XP;|)cySy>cIM~PuzmlzK8iDPB|C|pv_
zjKjOF20NbJ81nq*g#?zWNnk{$I8Lb+=Z+|G-280>zL_zSeb>S$TPK2<&qiW+`v$CW
zibB<bMBML@j2^#(aLLjjtSbt{^AiGb=_fzzZS%vyAHFzP;){zaebFV}7xy&#a*w7T
zw#NBkq!MdyB>XXbm_IuA`SC8qALX_M;4|}3?&OQXIm|cnm}~a)vF(XLUFBG8#8@n|
zjl)RKI4u4ckL(A;K+Xl7*_ek<U60}9+;SXVQjSY6mZScSa$Hecj!z$yqn!a?k7-;*
z=cSE!qT@O0JQ>2_9Rk!`&z%lDmv6i-LRUEJY>+e04!865t*8nm@J?zH^G#Qh3Ux5!
zq)k$!gS;0IAEQWe5{fkUwE~S|ji>q>1xlEwLC@c4krMAHM273r9@ee;kI|=d_w-5X
zkr8=qG^JHrOsUj^b5*u+=52^6{r2XpIQ9)a;^Qk|N)F#mX!bV~D*9qVE}u=vx}SBd
z@x~OEH=aH$98X3|$5a1$?gs4{Pm}+Sr{BItRGwo*>%S;dc84-;>Q|-}f-0oFgf*$X
zX7p0doZgS;oE}F@8ot4jCPlOE_!aMax=culbMOXdaE}1K!Q|&XC{xjcBV&5-4(lWw
zKXjvaZ8vW9>PE|lUARl48>e6BLc;@HxNdhBzPIbbXIdS6{o00sb!~Y3Y#Vm5x3{Xa
z4bNO?!~NIV@MvutzPs0kYWy*Vzb;+U#-15wo#u9IILbQGw>qSpp+m_LI`ny`4%vC?
zkQ&c)%sC%s7W2(3p5e^mSx!6ijTke~Vdk4U-dXj{=G)9S+RQiJJi~Ez(<LwFn{kiy
zNxxa2eEBFVv;IcZkjfejXdmDASRS*`SzTIxmAUDHE>#HV&=*f_+QD-bYyPOjM}=oD
zLo<1Q#EesVLYuZM(I)v#+BEVI&tEQR(?g!&*m76YT3w#wY}cVM2^~7b`z(_iI<)bR
zHodv3O&deCX_vb;4f5QkPE4DoebS<UXIf-&MvJnxYmusg7HOQ-q|O<dM7)=JJX?d(
zZ>W>2sXFh5*r)eHjiTqN(eoTtO6+Ey)>5IUmE1M8kGnLlkE51uCAvLQiB_pAlHFtl
zvUeOyKFh|Cz*c!0>o=P2M$1u4x-4BTlA+XV(&X@1ijMb6lI1W-T0CBY=30&-4<~V&
zwN8vq_=%EFB4_#?7AC`!BWZ2f2-=@NoE9YwBlC12ie+6|#~04@)B1<$IluAE{XwiX
z`-wwF-*HR8Hw;bxin_^P&~@1Wx{UdZ4mBUq%%dN#jOxR{6YsHkdM~CvdyCsPzd@Uy
zoOu({g{gmD;ixq)Q8>2)eVB<KcXFPMd<&LrYsML^&+yHEPqDL!uSo(Mv3yDc8cROH
z{x1)3bMJi&{B#e84d;v-)p{JVspAgeyJ&N-76+&nyHD~yD(W_lx>|$XLe&`h^%mxz
zyM@2!+(Ms-N-Q(FhBZ=`ap1@WTrv6_IuD)Zy!bNwB7X|=KAps2btO2>rx-OQicqZP
zIEEAyqI4<0mb@*%m~{o1DO-RS1q)DM-%*^`l8*;IAHlq?JREtQy--QHD7z*H$IQ;g
zr;b^u;+Kg}8#2(>Jp)~3GVtBtA>1XKfj({-D1RpdYh5yN>7PvWJe9?nJ=thGn2p+g
zIT+rUgF63mP-%ucjJ0%!yYJj!(*`$K)$9rq1+MUFlPjbexkAn-7w9T+0R^7VNQ%0E
z`BP^oJ>d*jlAU2#xHBw>at6sW&hX@eGu&9~0@5ump!d}c5@))@kB%j<q<aahP+STF
zF-svcemNYSwgT#`Rzl|fH8675I-Ut_1Z$T~;Jtn`xXEpSM3=2_`OH?BE#?6ln>^rA
ztq0ub-v%Y~I8i*<hdZ7EVWV&eEZ-Fh_HL{>=H1G7-s3!Y7!7aVM!|CbXfRQYftO`5
zpre)mFd+ff{$!8Q_$Ux%Uzne1FdVNAf)w!}_`D$iZeEUpY1^V;Tp0U<6j|@$90jkK
zaf&<R!D?(GEMmS9X1-~)NCE2u2cWe1Ae>r}0mjTUNm?0j>HI-BcO)5li{e30G7eOY
z5<!!_SPs1h;YZFv5WbcMD`gIV#V+=;zet23g*a#zje=~aFt}C~3@eZLLftrTI8(n3
zDr`OAfZ$f>o8ke>Ot(UP)n*Xx-3S8lA@B%e-~sD#t~#(DSSkiIYNDajB^vf+M!{_M
zHGSC=0nKOmIa$nlDal~i;1U2Hqx@jc^*xZe&l^T=@Pc<Y!XWoxC~SKh3SBj!@SJC0
z<rE5g1Vh2;VF(;P9s(y~Lcqu?1d3LLz;x#j_)8&B92EkwZXxiL^Nil_4FRia_SBsU
zgH#9y><EYce&!{!gP?!;Fx-Bc2Vx@%pk*xgm|2{FL&J(-+1VnfaVmy64xIldbrN1$
zmxAq`Q;^7A@SaD`fVjwc@KHJoJB7+YbKzyUvg<0d=QaMm>oB3`COlYL4NFaG;rg9=
zSpV!kB+h;a4IT|pX5R<~!i`{4+W-@fK8E9lO>j573C8p_f!3mCaA|CYQ=hoIYHB--
z`rHmHqdH*9PzOw_ehDQx@8Dd<02Czs05IzX|I!a&=kO5@2@b%c!(U;o`5<&m5n#Wf
z5Jq1cj&HJu<KuV3|HsmK$5Y`z@4u|fq{LN7HqB3aoO9Ycp)%UTh|0{kS5~EHmpvmH
zrP9!nHc_%F;u<Y0X%8gvJKyj3_s4nNxYX<8<K@0R=XsukMujq1E0)2N3bL5pPYxSC
z$zf!wJZ?IsfYo~yu{Kr-&kY}lw<Zq6AA<+txJ=GvePj(NMFF$I70_kA6rTMnh0;3G
zI5AEJn?j^8OvMpPj)q{MaVW;C9LKlgk7KHR0IFNC$0o}ki&XqkuTg|0aw1&NA;cNy
zgt#M6h$BOUxUNTt(xxID@=u7KdqtSo-w(YzL|D}-!Zpl8qnURW3wGoA@DTo8FbsR|
zhhYlyjT!UJ=Y9v-vv?5KuQ|w`#V{1E4Z~%Mu~@q;mh+#neA6KwW5dtmhBN2!PsVwi
zpLHJT;(2VjbRK<!FJK_|SmgQK!m@o&v2UIjpAKn4GlN!?ywt~ev_3p<KY+f6^Ums&
z0=<=0q-;OdZw_j6w}>|P-f2@XYc_@2+!s4kgVI9<Bw&B%r8of%O%RYoxq!~cYtX+0
z4ceWqONAM_)Obmke6HwH?sHwrYv*hX=YJ!3RKDm^<u_fv%dJP6`g-(j)Np#TX*g+p
z9!~Fm45vk%!|7Y^a2hSINzU9=>#D0si^lS6+*(a?V!o*pX_5{1*m|jG(Ys77axB#1
zP8%)S+pb0RBeZGt{t0B&e<E#RzA3mmi8e2oOxm^oQR1@yD4K5~><Mkfg`TaLu&5P(
zD74~Y=9`7NEm&dEf~CqW`0ZCS9(mb>A<vs|W=RwFK5Rmh%T1h>YQhGuCX`y(g#Dk_
z;+Z?OSiOpGInQR6S=@kv<OU329p?nk{@V?h#gBnJ?mQ(I8nFFr13rAl-kg9sv_54-
zeZ2EQ?z+qIFrq=sH|?s%)T++<40FywEn}+FHKs(~QT-ilOc~5KvCKDebByWtGS+LD
zZ(>-#k+$P|2f3!?T5L+yPfY1L=d(wvkD`Hnqi9RJDXm<}9K=kN<;^-zyfNLo!k^2#
zBr|72T5M@ZshbSxg|#91&o`t=^9(7G=hz}cYFTMWVtan>82|a4AstCGq|Yse^fS|t
z0tyVt;<F)*<(&7PWFuPhi`nM3A<Yjqq=D=o4mM)%%})awTgE=(QwDT?s{y?k#B6g-
zpNdxKQ^-3#%G#+%GW<O-c84w%ao<fGYd1+}b?A3L9cr-{!M8oMX@XdbgoCwc$O28;
z>@l3OV}{Y{TN>=m;d>0co09KUqo{#u<fb*0l18bL+zifzFX3GHS`~V^a}fQvN15jF
zEzNt`N@P@|NblaTKH0^$G?nE^U?@i^Gh``b`2f1IhVM;nl%_wnQq<2<k~G&#&_2#g
zrNyuh|5-2ocI(0WzyIL%`<>{a#(XpKH+CBT!b**wxQw%mp4s0p)$S{vR{nx=*Z8L9
z%61HDZ$tOs5BO<NE0!lT<2CIjtciS&vA^HqwdrqggbOn;=b~mei1Gf68qB=#3Kd6J
zVMpFe&QiWWYs2RVf6LM1T^Y(&m!fuU3F>`#h7bQ0<66EcwRF)_lsLp}Q(uJHBa5(N
zL?JfYKIYpG`IwUSi0?!^!g(r>@aNzMeE0Vr?ux#HTld_+U!SjH)ZNQCc<LoI>$r#!
z#TU>r{yaWhk%fkxXVK|;CKg0y;IQ~~>{ptGE|b#mbm1Ai;(G?0zonwhgjCE~avBA5
zQgA10H|~Ep5B`$9!*>$!+5^r#e2Bxs>2dhuW-O{Y#G-gkELxbwVn9eND!z}!2^-_^
z&zCq9hs9&~2xgqj1iU;n5g%?%<h@%W`bn*U^n6?Jeqjqk9Bg4rxed4^*+7ClbIvFm
z*!aa7HZk9vn_>;~yRBedIq#!RTfyB>E531Q1@)d*Fn~3jnhGmOn{5rX!);-8i!Id6
zS_AsG*Fdw^T6iG84!#DggPV>UVEmwsVDxPxl#m@b&$Wl+Zycd@tP@<B=?oV>I76J*
zcJK<@0i|DdzzEhbZa&z_`DIrS%kKj3;XdHxw;TEn1j6{&L2v~*UnR94-ew(u3j_V&
z+A=@b9Oef*GyGtGk{_&E;Rojz`+>8_k9(PS!@C>1A^Gxd*lNEUOj`Wl+jKv0D-wZH
zln8A0`$Ot)&Si}~0NbV<fV?rG5Vz_isD9y_o~hxy1B-x7$0I?dneTQwMu8wV0%rdW
z2N~ILh&Xf{QcfO*x!mEqqA?7Tt{nxTWi)8naYwRq6fAiZ0Rf{=!jlI_L3+R;_@Q+G
z28s5<l-xiF3J}7b!=7;9s|&;~-vNntoZ(vgHrPMc5x&^jgWJa~P&|}9+L`O1La+xm
zS%gC2Zq8N(ghJ=6P`I^>^&6x8aNcqs{9s<Wt`iLIR|8<sC?V87@`5>5yP)%=3v9jX
z26u~mAT%fddX@%()}0`bx*7!TL7Y=t6a>*~K_L4*5SCU2!lj3SAbBkimZb)QU`-%A
ze(euY`+{K7rT_>#&d<*e2653ISQy0Jjl=eV4)3=@LzCdg`ebN6k^;Bi@okG|X>hDF
z6S`lVg-X8DrO|T^oJO1nDen9eG+uz;cG>Xr=Ow7~<QrOd&x4j;8k~C_3mdHCVBq5f
zSi<+wQv&Y8N2y1AU$g*9J{7{AsAsS$u@sWE%i;E{=l{z!Qw~-@zxNeTGO+@DV_(9$
z?w6oZTn%SN*MM_Q4QMTD02=xlhRVEw!`t4#_1AA;Rr*_4((xYbyqn-xQw#Lwv~lMB
z6BJ+n0;5B}!+;~dz<R<T_MG>}gIoKf`{Mq1Wp;lo_}LGqt(V5Xmt-*T*#LaeB8z!q
zIqX~|kE6yY;ExfCcxI&%_H0!`87C#|+@^$dR0&7ND`KFf0^Z-DfDh%Q@PmvLetaQ?
zbMvGyf0-0ENIK%Nj1bhg9E!mT$8nm$al|bFIAl=(%BTil@W0(S=c@>{j*Bq-rw~=s
zgm}YWh+}<)c#hxhWyVotzLDpdX&}PiA4Pcetq8NkBJ3y6TF%hjI6fi-mu(Bf>xaWI
zt%Q$5w}qko-!Qb?8is+qyHZ;jhR@|<QHfvmUN^;K!}&xUXTe>ewmfH3@hVRnPZ7^j
z9_P$dj5?i)f|k>KA3Y0w?Q+rA^fA_ttH9J9Z*kk9xA<fZcg)N7;P9#a=+pxR8lk30
z9bww^^N2RNFx#Bb)1d6Z0!rH;pe;KD6wNyV^&<k(V!pZlNI*Mz%<~0=MFJY{HjKhN
zhw(0B82#`WMuXyp(a30B!c)4mjAvYoE=dS9Na+{%vvmom*Hu7j#{`rUEg+hzL3y(^
zNOiFWEf}gvnL3)3XrxIYlexQgCcoDCXwsetO&a`OlWt^it|niL;<>y3UYiyT=N;96
z0~4sJd;%#j-$+SMq{gZVlp5EDJ`?$lPj)LFI@pREx3=Pv8LgNxxE0rbX~BT<7M%OL
z342yGq4T;X{I$Lbvo|*3;nXHn%I5cnG~t}%O{l<p^PgrDW)|1t$=kKq|9mY5EUU*u
ziyClcd;>PKPNUB}vz^D2N5tdGvxDD%i}_~Alh^2ORfqSF8&R3ai1zVpaWtaFLF~U7
z%3XlWHz!%cvC=fAC^PoDEHff!ezad@L}Koc88zFOdU$%BOexBhyJk+AlGk-pvb%3e
zX`J&WM^m~p)|4hK<vrDFexBK826GXz4%GbKh%}Uph?RZvXHHQ@-X*Owq*7)Q8Ri<*
z*@pCteZhn07?NNqdvRFb*&{NfbIFFRbMT$0I78}Wrcpa#NZ!mndn^oT%Sb~y+Qb}m
zgWnfuNE_zz&g-868SXcrwU-U(kTVbKHba?ho?7Zt`zJk$->XOADth!JR+pUAnS+n8
zueeKxV%O<V&o%blNROZ+X4>@ds1{`xGS7Amr|bH|DQf93db3@FG`Q2RD^Z;+FRPL4
z!=co#LY35C4<U_DgXwy=3i-&Z(B0vKs9>xz#myT?8f%rP-9wQM9#Wu{XXNQjo*dP`
zkflW*22f4643)~t(065Nk{l#OrOJ|Irzt@br}ii1<^4$L+Kcy$dvM9~KbU&HllzB%
zW8afsSbdK(;e1c?>>j?!u=pF^ko<~!u6@Rk^`Ee-=Og}#ZbQRSA5h5LTsONJ2i<SP
zN#oz+>tk>EM$~I;9m;#nm325iS&Tm=YB0&~72g`JLfM3u_-Dxr96j(k_BEB`fRZv)
zd02`W4@<DV_!(}lEk@bzPjMpWq_)j}g8%%BFg2qPPhDXD-J1docgjaet4G);_W-L8
z-^Xe1?xB{>Jv82a7h9g+!DWYXu{-!ON=aNo^_P4PA^iecxu3^UlPo+}e-<5*Gx1eG
z2Kol3V`yd?V$T^&+jRzG{;>C#J;1i%r?D+F1&dE7V`=Cqd~KJ6Yvv^4WA+d~7!!{Q
z7ICN>7mMCXv6y=)244rqU`0?2jyV^D#+@;kwS|8ldmW2!?BX!DISwbRjmM;0@o3i_
z&))0=l>TE2c+D0{Hrc`r1zVW#$ObO#vjKH;8*pIlM%}|2{8+o$uW1cazFWb`XI8L>
zJvR>ntRTRd_fl)DAV0(kj@+^0dyA~yT(g0LwYG43<{B`{UBfq`*TVI_weWZ0dgva!
z0s8dTLw?*wP*<^o(F5(l!`A`iq@BRR(3$TtI>YIe+u^;{4lsPV1Kus$2|2v0G0JuY
z=U@+rzu*O%c<0jL;s^g73;>s#K`^~f2&o%HF!a0#_Nn{9{FQ!ii=RKT%@3-{52Se%
zoc-X$Q|{IozZ*hLnQt2XV9YK*$Y>XV>j4oYn2A7Tfe7+v3t>F#9Uo<O!<qh}u!8wU
z?ej?xFy90?Mesf>5^^k}K=)iE_?(M?lpo>njdxZXLXX2?r^B%B;6YH24g+WTP?&l>
z1U9KfL*YE`oJom*ampv*T+UJ0(RUDn^w|^5orfEkA)3vEur=2ME??XUi$`vU2Odsf
zS?LJ66CB{u0{$93*bEg;8(?rP`=gSr!SnVi&TOxQWt%LaV}vCruC;{KO`G69<zUbl
z8UTjbzL3hfwB^-a@HW~DHeA{cO_2`Z|8_e({N)L&IImyOe>ZHG@P`mPe=y(c59bd0
zgU1nn_;tb`ZpQe-i)4RL$@GW7RDXEs=noT;cEgityCL3eH|UqM|0ysCM*j_lsKPz4
z%W^M_v)c!4K}ldCp9~($QlRU|Y4}``3VSA>fynkuSh|os&$;ISw6Z{aH46@pKMzMc
zGN90(@0vUE^BF1N85##`3SvRHAP#PN#Dh}&Dfl*xdlA(h!Iane+$C5DCWoIwx>+$;
z-7N);+A?tcS<XCN0Y8i@V8PpRxKvjT9Xb`@IIR*wd#mB*?i!ei4N%wB0BZ)kfeSWo
zVBqyPAjMfK->CO6etZ*qe_G&`W*Znpw}VE+C(gisg_@W@uvn@a<S+b%2%R2CozVlg
zyn5iBe?RQq&>t-yNMrV~0q8bb1|J5?pna<xJ}#8U@l^_V*<1<x)+*tUjm$XS%suAv
zIA*;p&MK9|$|VYDT_}l~Ws>L@CWV#KlBlL4g*!VOa8PClI&|&F?#yHO@82=p6C8l+
z<_6%DK>=9Zx*L}d*^Rlret2QB2p?P(qRbK@s<;So!E_-?ln8MEvyCI~u9Ve9_<5uV
z%jS#F#8-qttA(iN5`fe1hv3N_tlco*{CFCMO0UDvr!x%aYz)J5%r<vc@V-hT7MHz>
z$D8Si=zJ{^BVHxq*SbU$G$x`&XCkf|$$3mep7BX|aYhm*%umAaxk;$-IvEXPvv9A~
z1@t?972THR;e&to(28?dLyw8kp{@zLCv{`ofgyDJwJJfADtkFqNll<mz5CSZ_7XK3
z&wXkdA_1K`ETARKH*=Y7#_`O~5Kw-SfF52}r)3wpcP&euY)`7u+oNif#LqXB2*|fx
zK>OIc6Y*U@H-8Ff@^%3|2@+7zQ32I6-=wB-?w9$dZ;=LdS!mEa=9@S}P1-g_lgg%R
z(tXmTJY)7_+|{6V)^K(kYSGQJT6{}Fi`2eqkqPgqMzEH1g?Covo$_Q~+lESg@6g$l
zZ~E{~pgy(^f0fqanyGc@F`4iCOsK=HziZKWVl5uks>K7`4||YDSGyJ$eiq~U6fvG&
zEJmHNV*EQ(jFAPkSkCkMMlHskt;JjY>hb%62K*b<fWs5nL(DughR2e}i|1x0`*AWF
zFlcWB8VPIh>F7E<w#|q_9gJx7Oy2V-8Pg#ZV^U%L<~1`;6*ErlaAOK!t}$Z1`EJf$
zm+M9}E6<4J-WX9W`*O0KjH&u8b5Nr(4azd5{oFOPgga;Ct+|Wlj|o|GNAiR!Q!<R@
zuA7<0^iIu~a(E|Zy~K!mdFQn9xdF{)ep$n}5q`2id;oJvJaf;=>4r3s=LWM*6MJc5
zx3Z4o&AIMrhEykDZ6=>JnIr>}SZP47f9TWSL;6(uNRRrY`F@AVND4lp!+vSb+&$K%
zw;K#-@pJ>SeW6dm*80@_Rgbie=usDIcE4^i<IdBiEssZ139_&Ff)3dX(xIE&o1K-X
zO$F?u%9*D{WBoO$`XXz2;$d{GSA({&emH}7Q(|*<8fK?P<-3Mb-d?`RaAF8|1r4U6
z^D4CQ_8_t>R;J#%ft2)Bi6%-bQK+UO8BbOq&lU2Nu}zLTg1E<$J8u-SWN2otH03_x
zdzwX(BwfNim*W1E|DF4a<@%AnS}#6s{)-xC{$Sj`P7FEw8)J6<!Y=Ed*g`*0P2)Ry
zHGjqH>0eN3-DmWW`NY3he8j!;+OX?YE3UF@!PM3!Y~0d_TIKI>`RKQJUGy3o@71G=
zOdUq7=kKqgYR)IU!ta%pSn2%|Lrh<w@J9t+d{&O%bIP#gY$+beEWw=@o?-TNz6W0L
z6o=P8!Abv$uwg_I7Edh1mhq2q<MIN&#hs7eBp%_Gj{E57aUb=n?x7;{&11K_=<tf~
zX+F$D?;)4*NLe;IUb=`OdoEz-lJhuECJX!SokfL^Omy6wfg0=5F;bL<)_2d~Hmx%V
z`%+Q0<TTcIrJ%Dy3Yy6zqsG@Hlq*X_>5B<?AUYlsBjZr-axBjL7Q@^UgU;WhapcWt
zoR}4j4TaHICKH3lx5OZo#^ByrvA8Qc7FY5PYrvK`Tpb&SH92vp^34{!cG|+66|CLJ
z+Cs%W8#v@|1G^U3z&zG&LYZ%FF0_Vc0&5U6-yC{i1>@O!qrA%sde&Nj8d<?wBla23
zwSsoeQuVH3|IKb|AX{s&cD4q|QT&=*Yz<fan4>D!!oJh%z*=D=1YO+<m#gjI<}3%0
z>T!eyO=pm}?F{kP`E9LjU|-}6ebPH2bE6A5&v1b&>`zHc-UZi`Ji*_|4~)kMLA*f-
zGh&4>k@qn62SwoaK?Fv+ez0MtA1q;Q=CYw5OdsV3`DT9Lf5Z>gKk@^!%YLA_-4A9-
z`+-KZ2xg)PTDpW_cU1^~zWKrc7hm{e;0p;tZ&+Hh3ufhcLZ6W*RN6;?y>ukl%;x(B
zMUhaF8UZ?A!lAzN1YFpC95&874Ev|Bzj;L%Y|lLazI#I8gVbIaGbsp~i<l3tir^s%
zVYa#t44LT#JH?*h@x%kf3q4`-C0D34*#RlVPB7GP8}z^H0M+B{VHbOpRuyalQS^GS
zG+6^9hgie&eXAh0-%8M5ZV6IDEn(<zONgIu3C=u<vsc2CxjR6RzYU}k9btLM8YocS
z$X}1G5R&HvrnBAP!E`TJ+v5Xw9fUA+qzE?h*X>6Me+~MxmNniF-mmcknLt12m*@vG
zFZjXNNIz&}MqJL%y*uLvzXbl!vMvxtEeZw|){~C0U+2`bJ$!R27XB-agKd(WC(}xV
zcm0yMKQRe{HYdS_|Iu<Lr-SCnbO=6?4hu%7!<7-~uz}y#-Eay9^(BMszgVc|JE!_1
z<3N5!JQx=zfVeObKB^~!`=flA)lvY`?oYu+rx+S9l)}y%WgvO69F}F5!`pr3Q0`w2
zUMb};=0`bb%U434PbCCA<@*gqV(v(6fCkAo;J%DCoZ{E;pz{q>X}<^ACymhXyBQoV
zeE^lDk1+L5I}8u{46~VUI%T?{`Yii$m~U=R?*XkXJ@9B<KlCh<zzMUZ(L+iG4~&w*
zRxcUUx0S;e+H#n|KIMV&|6qBsA}U!a;TCs!T(wLdWxEI9dVLvOP$-34I{V;iZy!ij
z{Q-yV66kK%2fOqgaQ<z3^sfv->!$s9<^D0Wjz5M^mmbAv<p93F;E&Hn@5W<Z+-vh(
zgu3sASSJ$V^Q*piVVn?W=?L-b5FuU}EX32YBEAtNLcX?)g$snJt}jG|?E!dDJq%qu
z!!Y_r7``fI?WR5q#jNL;@J{OVf-rPl7KR4}u^3v#+Rcqb{+W@8A8L4*Z?u?iMls(^
z?@q)kJ&CwbirI#zl;;p@JcsWjp;vV>u8H8C`{oOn5`7g1?a9M=UU%@0N<Lb}=A)>q
z36~{&#s#O@n~*9)!DnP>V}}g=n9lddXAGb;9?KbObceZQM!bOPP7COGx`0Hf?E5TK
zrwxCG&>o&6-9yM$N|nfDDAg<(N?XI#Xx$qDX|Q)^i>-k6ZWd6P13v}`$UQ_rro8h4
z<{RZptO+sSR59P|8KFs42AX`|NRv+dr%6T=Si`B)pxN0PbnCbVRk&%8x}^ps>$4~5
zpA6jrDXO{u2j_+U!KWWPu%+i0{vBL{LriP%1Jz)6NiC|pujT%=T3j-}7W-i>daKl8
z&+uAIX0CDN>0c-2oq!l;iN)9yF2?K1V(c6!MgtWw8m<zf*~MCXc&Zlfr`KXk4|mLN
zuE!Hr4Hy;IfXWFC*vMLrIzL`aYQUaU_TB`sKgY5b&Ae*yrDPpCZQ{KYbIYj7M&!wS
zW5Im$dx$Z4Gvj>YQDeqg!h0%f=9|j7Ms$n$W^%3(U28L<{6yAvs*UOE7tWR|no!mR
zQ~FD+-z?zHVD>!i?l7VEzc`CtX-d(@P3g=@6B6^^E!%2DQVUrRiejDRi2;4yZa_Oz
z^lAJKefp=PPm{hH5F1X&W}gB1x*O2%6^2x}(U6Roe-=+PB)RVfwBexvMa3A9)B*!C
zf1^*hNuO@U>QV1%U2;<#Np6lhl-)goiX?UD&e9PyI#!<~qd5ouQ;)tL)uWD4dgS{=
zmtJntB@6Dwp6)%8QorfY#4S41SvZ17cLW{WrA>$Ov}n*@P5Q(CuVw9rQ_-<u6rH0%
z)h`5O{8^ngO7f_w(PzV<lr>qE4CW1C4jfFE)~k^8_Ca(~s7$qC18Giz5_Miuq^3dz
z%6ubF(SPKqYKR=OvMfzmFo1@vm7#*|(saN_iq-{6vd3J4sv`T-f(!j<V00f&dGe37
z+Q0b9<qt-fcjAfJzj31cFMR#=2TpwT9Y^JSL+6mMShVB|21|a%L08++*7_rE|Na4W
zLt63bz!tPP(S#EQG~x)4cUW2S27@%YH(*;mZn;^@-8N#B535EurB~>ZR*4nMU*g6A
zFHrQdf^TV-qf1;FzB^Wm3y+lGxTDYT=ka1(9?hLMXP#i(^&(7sT!>1AkMY#y0<4J1
z$5|1N@ZZaa7%}DnF0i|g=S%P5N9}v~#r-ZWth|H$SLdQf;w3Ch&qkSJ7tv+Y1srU6
z9s?WC;m1>Fagu!|nqmevj88|0`Dyr8cn04*PQ{OdQ}OrQ(`av>f<D}Lv&{7rE@Q2D
z;<QA(q?Uj-|CmpH$MTKwShSiIgG*0E<7Crlyww!Ny~a_T>x@Dxm1vx?F&ektj>c!h
zV=zz@gCj~}P<vo3M$e2z?Uk{Zw}$&}=CIx}-xiWvZQx&?4P^V+fcA77NNlr)&QsRl
zZe|U|Jyx))-U^;3TR|s#Z=zYdNieg5w>rF+lCy$hJu8q|y9)XpT?tbKtH6bIor#53
zaM!>Z7D!n`kGD10b=pGCk+qPzZ!;9F-U{6=o8kP;Es*E874%K*;D(Gn43=?#Ih79Z
zsbL#@pXmgf25pDo20P%qfh!!J>JF--J>UWJ$Ma!As8}R~yS_qzADj`|CIa<`BKWUM
z1ao*-mB5oG=Lcs8`$5)dKghQ5gVe2l@MD%Aoa+#Qe;ju!&lEw=S0Q{!6~dXBoTHBL
z<<~V|@bUA3?Z)1q$=%95XFS37pa%@paEA$3U19iI7dYFn15B#jAYI6rCk0P<DDVRN
zmA=qE@F3W8A79(y12CQMlbZ4C{!@t{P>kFSt+65)Fii-b=Xk^UZck`A<OxS3J;2hy
z1HQJo!%?z_de?0*ztjPyU9^XQS$0shX$v^CZ-m<w>%fJ*QCD59z)XD==-jY`6(cQS
zc%KCvkhKI8JxjPd$r3h+Sj+is3EP60M|sEiLURpt*sX)*8JpqvY)5$hhS_JGJKUSP
z3ud-?!Ozd$pdaG{^@Dw3kI)z1RQQ6jmJqC%8&^gM!Tzog7C#q4F3*fXBDhj0f?3md
z!vPEc@w_0;q_Jm5We<ql_CUg~J>XW%w>o%NrIH;2A)HlnICB8L1c$;i)?z+&q`_Ic
zba3C04m}lV@bqdLs9sEhQOi!jn~oUHKE=Xd<5+mbeM1`m;=pNr0!#=^1Ox3;@a29A
z%nG~<`bP?(=usg|eO3T2p^w1(P(JV8iXbzt1P*7ELa%K({3$61Y3T}ZvaW<&r%JGN
zc?l6gFQ7Q+6>L2F3S{rpKw!%&$P9hQ-}8;&Dr|;0`~d!|KElkc?J&x>9gZLQ19EEJ
zP%8Hq?lRvfK@aRJ>;)60et3PbBz9eu!gxa|+&)MK`vRq~xNiXFU6(+a<?li6(R-M`
zs1YJl+d*NsELt6s#hgzAu&-Vkk2^5mH1|QqxgIcG@(2FR_zmBadtlLU2lVu`$BZz0
z^yv!0-L3oaSN(pBo3$5jt_Z}rwH~NH%MBft`C&-92=8jKcQ{Ij`f0wnmHEX}Q-}wd
zZN@6ICugJ(KTC*Etx1T#=JWgb?I(@_II}wx$1&f0yB&u4kHav@A&h(cc-DmBrTJkP
zzcdVGm~YNi#^cqS?9<EVn=i~a{p%CajQOUpBN2x)--IyVEMTsQ<tgTo=buS~3zFG)
z&7E8)vv6``7G}G$j+1u*yE-r6>XyssDt!-y{5zi}|D2BX`i4P%-!Ql58_quT9p|6?
zjt64CqjRwgX~6&rzN13D*Hq{as_<<EWvcU2CPk4lY4Phz1wY27DAR)es`O4-mEO-+
zrMfSwG`ClkjQBO`oVtKwW(mk`7w@>3Z$>fSR2&r0r{e;O%NEesYy2E@OSp{&#p-L)
z`%#*dJ6@9#8Z>BBi3UYn&>;Pz8no4geb43^6g5nSifZ~3tnR_5Q~%)H8(p~ib2TpP
zSA&rwYp~0t24_sDLDfk$c!ytGMl+-6N!FrVpBQ6*ig8`57*GC>Zc{JDkIXj1cxE%-
zv@+jVsEauR#lLGSigBQY7^iI#qj_{K?oO@6@7HV5S(f|s!|L&Fcs(}8*7My{zK?jO
z9vjZq<EYDgOK4Utwk)Yd_kFck+Q^@?z=+<X5gF_AEha@HN@T`yWX3tn9I}%6W+*>~
zb1tb=neSThbH~RTah8X*9_E`Q=9}DC#`NC@-Z}L*p*wn<)do`n_H-A>n^OHx6WYRj
zqs%kJ!<5Q?n2@Hs3EgZoqDW>N{cz4lB^uBzM+5Q-XZ^)SpGM2+Qv%=LxG-Ix967uF
zZ?8U`a$z=cVZDa69VIQ!Q++ld=?4aMnQyVCGOsKz(jy5geadA`=AVTw^$Z$Gb^1C~
zeu{5mnvbAIoPj#CcO-fJ(xYdG^eB+oW>|?X{oKi2H*&hPIB6seHX2EX(sgK!fcH~D
zBj`dM=ikO?(-$xHR^@2YkoMuUWB72|w_q4ubk?9*yc2zMTAhM#vVK@Rl<aC%srthZ
zs`)*b#Qg`8h;vi!8iQEVQzo;?1IcTl60KOveZ}qy^y#2Hollb^g9oxSwr&9J|0P3h
z@-j4EOPVxCNm0OLNxC>of~KzMPvac=(Pzy*T<P|YU)#Ge)2R!;w05B5<6o$H<|kH#
z{lHCb-*K-wdv6TB;*f9LZJqrIGd8v3LFtd!n(_f(O=!h}yk?Y}$@|Uh_xN_`JIr-^
zgHs-GUy^b?`a9HOFlWODEU3n|=T(^KUWt}EFR|h6bM(Bzd*c)3c+I~ISMMstP3|R_
z=JE`WITvG@<5O&LdV-FwMYzqi5G&U|#>(*psP+C4ez@@fpX%MijV*cjDmWMKa7W*p
zvDb0*y&Tk9c@?d`UdF_POQ>)n8;84J#1*qHaOdZFG<|pu+x^dC)a*<wSIxlv{nGJ;
zEa#@iv3A2ZrXHL<jVn3J;Pa9_H=OGZ6DOe?|IA5GO~7Y?@!Z7|hu<t?xlbbolMh7W
z6LAy@)<$89TojIOjYQKgktn4eg*`S=crYUh%e$iR!oq0OJ`|05_oERSqp|l_G+Iot
zg~`)xA#a>5TzqE(c^TYabHIjwm$!j6@2w#-iFZz8tzil8plX7wK%IRzYlT)2Zf6A@
z^Q>S+-)hbsuZEseE8)WvOPIHBC4_6Ng2Ioh;KSY3Aau8a&A+Tb&)OP(6|(Pm?pj#x
zuo*tfZGw_Xn?XHp3ryLzl{v`{p8T-`iFfw!h4(3|Dz-thh7&YuZ-@5W?Qk%{1=bvM
zh0e9^5aHnocQ5;b!FVD3SSW;rwL-95DuRF%5u{g&z`9uky-gyx-X?+v%raY;b0!G<
zAX&)|I$w!EB2)yg#)#l_gAjf(+oV_uVd+a>*fiD`Y)|=se783U>b$@tb{8lHdV=>R
z4+xcahneSGL3g<e*j(KKGQ3OMld}!}_IH4fzqf!v%tmlfTnC$1+rlL72^x2HH+(Vj
zgFI%2BK8gK*(-u(10gJs_J;Ybo{$yf39o}ZV1kMVK$AN>>(~O<S8N5(?e=i^q8&W4
z-wL@go55$|Mz}S69p9O<fnyg|!%)AKFp9OB_irsA@SO#m_+bGeDNDGeWeG9MEI~HK
z5*9|T1o4Aaps8UEuSTqe1@|_BRgoPG$#sIHvYnuL!3_o;^?>SUo?v-#7nH5{f+OF(
zAZ3>~9DMH$C&&kikN7~-3m@3U?6cU`7b3;JFr`2UuHXkQBm5yzF%ZIU2f_YB!4Nov
zZ))W1f$6++yWzPHs&;UX&i4I~;&=d_nTJB6MHu`jKLrOCrNb*qhb*52_%<gIW=%~3
z#d}HcWgBPB9!5jcwHP=O84GkJ4*dVdL(aTJ*ei1iHbkXBMDZC=`I!#3yE9>s&3U*P
zk_+R^^C5QTV>n^Ly3WE<7~fI~U02HC(>vZ(m6yO8rxLg{sf4pqB_MiK0(y-lu;^Gh
zJWH;G$J;9*QKJgNyPM#yM+-cZ`v^ZI+d+O!JCw<Og-0X1VR2D6ILG~kC(-}IBfT*9
zwgjFSAc@zON@1-u`({cdagml3hK!TM&vEZyUfVk`_jwOJ)Ci_AO(5Lc0uwKO0NI94
z(AwW0gYNc$yy8FjY0ZD%;x0&D{tNp2|H4hJF1{0DkF8(rFi_48r|WLT<L;I?;j9IA
z3M_E#j(#}t(KncJZ4SE4oy)J8LfmtLGgOhjICD7frZj}u$ZXSJmbDyJel!u{yEGwI
zvwmaA+Kz9BFDfh!Ksgs?8RnZ+ZtTNhzHwb2hV?dK7(S2noJC=1DvrSmtn+$R#N(!i
zi8#0@5xbagPJK?qK<1m+-b6H)Ov2f+Nw`ok3CFEY!iM-{yqcek4Wm=g<6su5u!hsY
zyDI+e0<-6&;j-epX!hbRzPnqB`4wg8^`;CHwad|YR5>bcDo4rFuhBU3J)TT%#2oH#
z9nU==BR~Jf*<G9|k?6qjoG+Qu)PZihlnHqj2Pl)caR{AZz8O7Sm2x(yQqxXVD)3UJ
zKF++n=Ev*as&vmom3GZip-0nHNcqhmGOHa#O(O=;JLa1<YYp<TWlv7{U`o|gq5Tc=
zbfQ$AH1Erk#~&FwRKnWQ9vSksl%S`<y|{TsHTrC?M!URfocXDm{ZQ2y$9(gX`DP6B
z&HITpC}t*6VZPbWzZRqaiE#<@&D3vV3}wFYZx&-CvyD2>9Oj#9=9}X~#5m<=4L)JU
znYK)ft2VNpbFdcGlWQ?7rxtZz)Z(3*TD(~!{@=;rM>n2fuf#b0ff&o4iLqao7~6bn
z@ezOQ<MBq6HI6e=nnpB@v)5*y3~4=c3^LQW@_gb6=BZ%5(d1dg&&8=4QMJDjt^dt_
z9rg)lJ~1Y32@^Us%!Cq_nNZ4U6S8?{Lbl8^9n3flJncN<ex{Uv%!ItyGjo!2;n$N5
zNJeBp-J1+(SCBqku+XPyNqt&Xs7JP0dgQ`yPvd=(^m={r;YVdtecHTAk8%|BXfx-X
zo}JgFz9L<E>ZMEf500dsG99{?%ALc5b?9gdcMcyMK@u}YkgnVa8oFsD!4f_4Gt{Gp
ztlJnc+gJ_OrBT;L(loP?q*kXx7uV}h&hrtpjz346GkaG)uwQ(f7L9S$B)_!b)c0~2
zZTZLAjgAJzk$~3N@V)SzYBWD^C`~-3O3M?6P~O?WRF$Jb6YmY8j3>&ZTse@;-YJpp
zcSX{cQlwqO6sTdMJSCWOAHq)d-W(i2c4;zn;Ept<S4ffFTV|V25~R@8pX6owQ@`Q;
z$lzEnPG0>NR}}w27n3fGyV!xHOMW5E_=#sn{=f>k@A!moatJsV{xJ44%C7r_LgjYM
zyU~VYSAM`*jodY`y&2nD8@UtWJ@+2J#a*gzaE^TgcLvvC_#oEv0&38r_Z9Y!sKOJ|
zD>32Q3tWDa`yGx{Fk6-5j}2uQYR&nn)g|b(;u(JA?3C4%r}%5k6P#>ZgwewaF{i5l
zzu(BmwZwU;DGyNY<6RtgAP==C<YJb_O-!A99k1u*;6JmgsPyGB9y@&rWdpO(%jzQP
zjJ$xJjag`rd=8z~oW&;^nb^^qj#rA)cyD$FZQpV4o_Z=8FFK9dlTxs9elmVqaf*B2
zlkm)V)`yi6u#9~)vmeHy4SQ^AGNW-?VHBn*M&Xa(NX#{e#G*eDIIt@MH3X5o|BuA}
z;gNXtRV3C8jl%a!qtIbj6xtn&!oY+mygbGh3@6&cWfNOi{MrUO&TxO<ejAYH{FDY~
zp`rt=;oul+P%d8y$)V;jd7C-p&o_rj6U?FiXmfC?v;-w{OUS=x0TRiUFv4dgTphj&
zvRYPw;`P-q#lZ^pHd%p^t~D&)VFRzmuYu~l>*4*AjqpKYGkE!I0UdMR!>}L6r_B!H
zAKL@DIlyQYM>xtF&V?aPe4A+d|1w3lgbQ?Lx<J``7nsz>kGAem)$9Y#`9jz`;s3le
zI=4h{>%It#Z;N2$6A@e!i{NaF2u?EZgt5MJ+fxJv8Y1w1AcRmqAw;ZY?MF@s!*=<?
ze+52}KFtTNMDT5_NH55D+y$R)Jz?e?4`}@71~tiiYkjf{c*N`g&0Ef#8`}maKiGp}
z*%r96ZX*mXS__Zazcl@=6@2)!8hYhcLsiF0IP=gFN;g}8ELp%Wjjhl%%L7j9dO)eB
z2L$xFgZ%roV57eol(+->jiMdQDBcS9Q@6nUPn+QKsSQAT)`H0)YnbqC74&9V!t`?%
zP<qq?(jHpC#d-@6Hd(-934Xu2CHPIYgr<9z@cqt8c-65QrXI58{PcSGWU&RV@!soX
zkTbMd>;!*vSD2XR28AW=aQ?gp$h&%i(y(1{@%%1u*Y$#Xp<eKaHJ*Pfy<yW!Zzw(H
z16Aq1;BCsCfqj0U_1+(L#s|VMg<#0I`hR)-JG1rzjM)d?+!I)4x*vv&;a(h*P*^xV
z3^sHAKKC}~?zbO>HJ&G6;=*t^_#grnj*f!Hu~D#oRy4TI<R0efv7og+4ib{$A+I?B
z-nAydzT^}*bL|W~>X!la$1~x%{CUXs=6t#GZTR-80A^Sf!hcppuzcnd=)Um;(iS|0
zEnQFHy{H)cA3Ou~u_cgnqXcF&lz_i;IsCd*4j)!jfYy>q==Ey?pHVGPRn!JgpL~Sr
z^I60B`~|%Ax?xRmHyn!i3$Eckpt?Z<3olEcT9*WxN=UJPMG}LvB=O%ZNz|8=#DJmi
zz%b|?Y~NQ884<N`*Q*Ya2i8N}hI;5TZs(3h)@xq$K-{4oaQ*TZnw$TCN8v}v82Op^
zl$~&R<ZsTCeFEjGCvdr?5N>o8faSM*aJ-oh^|v2@){pyeIiVby#?0X^<GDDY+XJVE
z@P6u-FUq$1;;d$0%<Az)PiCA?65cq1glM27M3=L>@ZV)$Y**uJfc<yl`||x*e=ZCs
z@@`7Jj{U`)8*dvQhEEoT;jTGhXjB`6mJ4I?H~VmwOi$n}X(BHAmWXGVZyX0EVYf!o
z|N3y2LK4=TNX8w>$$0EhG6sxE!O*FE_t!H8(@&@1oeL?b!TC^BKZl`n&fyujhAQ)}
zp~d2BsDAi1jy`@Hmqgsg-{T)(*X##4JF@`w&0gR*%NKat>IG`8eS!C#UtmLh752J)
z!mGR&^7l}t?f%LX=A}%72dR>enktp>gedW!aWEbHB1=DFda>%%KOAoS5j|6zvDdK~
zZ}2UI%j(P_Rn_QuX&^=BE79!RO7v7#fpQjC<MSr&r?0BQa*G;#Huy8vwtPo>!4G`Y
z_zJt__%6?+YMkd-jf<XD<Gt!?tb1FHkDID7zO|Zlm>P`PRfA5vYZ6Ul4aQoGGnqTy
zw2P6LXKD+?d>4%UH*u`v%n;)iLos>{<Gs|E8oZiagOUTpc$vL8X{xnYv!@o@5^FK9
zjG5@U7=vH(Zi^@Wz8Htj72^b+Wjq=gV%+#xj0Mb13l(bdoH6ggl<IhQ$ak%Vuy2_A
zWqLmtk|s0EDxLw%HL9%ToMzAQb7q@I{5}uPXK&|^eKg;QOx76Dj$R{bljaP06YsK^
zZ;lE1MmO_~d4>tu-!Y*F%r&QYUh`OVbGKiZ30*&ANXhjEq|N@BMlS<;Zp)oTCI(c(
zzTd)sdbIwm9yzDz5gyi~@dxxckEBPr8+7SKe_h(0GLo*hb2rUp9l9N=L-9Lx=!b<4
z%|jhh%hVxnPaT@A$ocL_)?kSFM!@>b8}`v$)20)5b?L|pT}rmsr6dJiin=wDDmRZL
zvpya2I;lfb)O2Y3(Gj$@Q=6`>)+Q77jF)}Vq)xuaV9L3=yGJ<_euI6*VgY^ptxhME
z)VYh2Z%$1ZO6%vU(qM}r^mg4~`t6`XA6*C0MZOijbk9I4KcYla5)^64MFrZHFHdV5
z<jAsHmbwJ8bY<!Qs$9+bjjJ@h36Y{T(UPQ;CPD3&`qSw<e4j1851U{9!(@fOs1))C
zm!0Uqt$a(<y7?#Gc>V+7_IDhd{ta&({ff_Azi=M$GoF$Egyj!E;+XAitbu+&zo-`U
z8_xYnF^xD@<~_DKzQv5&uW_hs1KO>r!=<;x_}{1+tjc+XITls8rSB!S-F<=I4n4;O
z4i%WUtQ_ZI8J?b2io2(m;G*%*c(+xIT~bfc;YSe$Hx^<>@nbw5Ux1zq^HKBGLwr-f
zp5n;6e4~N+Mv#lsRc@m5tm}CAMh>o6bQRaNUB)Y?E@7-oHtOI-%$4HXnFU$s8gLHt
zCY?o_Zy7l5W;!+>Ps2sQXVB$nDmFYgjm5vX%P%z<m2PoY@k74Dkjp;5lmyJ-Jotah
z<8aN0SQPa{WAo=IH13VWn^Ph&<WvL}OpCx}QW3cOZ#X_xjldD}BhV-y0?%KM!2Ycf
zeE%U5(KHgrLnLO+kHpE&Ht=BkOb{f@<PKc!yt!Zl0`?bwc3=&n${I>ytwCj?HRL3j
zL&_d=7-(q*=K{@O{|z&E+`R%mY%zyEM{~~LtN@X%891I>4lcKr^Db*SG^?6{>vI0M
zJ!Wvr*8&Ox%=x~8Ehr#&{@vLC6}g)r*?bF}g{_d<vlYIwALmt$Jv`s&z<nu>u<*t<
z2yk<RS)7v=@Qy0$!wyK$ae?hYF0lKx3#86*g~3Yh@NtzZoDi^1vr7bmG!a}(6hZ4D
z5y)hSAo#8buJT@G;Z64F1d8Ag?^(RKbLPWE9~f}P2XwWC5LxC6-_%*7u=IiZd~0;T
zY;RDr@q(B!yP#pRCrI&5%&fx=2E@98j)@EGIk*EZU37-8%r_~&?7)xz8$McXgkw9`
z!ouY?@L`)3j5xj;Fntv~VlBnN!4guZSipubbBNz*4v{A;U_I}c&TLu*_x@v#()_hh
zSGpEvIjx5t`%N%=;wGp(v=L0^Yyhzff2|zWz*!M<V(Cg49ccl60{)t|uHai*7Mx?V
zfHl)C;Km0FnEl=Y{$8_yiLY4O@w0@s<EvoVD{JVNw-&VDZiLdtt^98~!hCIKcs6DS
z9LU=Vmwvl&%|7SvzqrA%$L=uL&jWJgJXwchZKux@9FFe-lX+gSeS$Z1dHBF7Ss`3#
z5P{Z%-7x+@0A%X~!TS}#ywBMKTD+q&X=V+laUb0K7y?q9N1N9#6mkS%5W4apfcOwB
zh&TcRf{sDns1uNQ@g&?D9RYXaB0=L?6uf&B4P{j^(AF;wPR)u3(e4DkWts%Tx1_?A
zvQ#LG$bhFy&qDB-EC{`K6H@bXVdItp5Wam3T8c&7v;71{raXbp(NE!D?Njh{Dh91{
z&)|Vt3ACIof##YLP?%i~{esFt>q8m%SyaFdsaK%&wwYPz11PZv$6-o4<hQj$!JDrz
ziuuOr>0fAU;BMp961Y800#({2(3}0ikSmFbLP>O8BZ-agB=C*XTab~eg_lNR7@$-K
zCfao{d}1AF&8vr1^V-4x&L4;}?g2I4cX=N83kt5?P;T}Kj{T?t#X&D1XI2F~v?u}b
zrYG>9QXwo*%m**djzwL{hk&dH5c2gtgjJNmYgsWY@SKaiH6A$6-v{+R`Qnx)Urc-N
zixa;1Vz9h78tHjq%1JkbAUDpAx?)U`8~(oGhRp&W{BGoh$IJGk+wD-CbS)Hp-iKnh
zau`-LhNAm`P+ZlqAAbzpkLQN$N2#(Xly8W^nLlE%Y(Nq!DJP*R^NmP13HOF3<D28j
z_~J}5rt75OLFSvO%r|S}Qn+V51rN2H#dWvOV*d1WjGCE_cjlzyk9q01VIlWmoJ+^f
ze`#32x^;+J20oC?z!vVtz--QpW~AW?=^OZN>J7Yq={APizQ94Rt8k5XJF4q-Vn>ub
zt^YHS-t&FKw<nZn_dX?h8L3F4w=2@+r4qDyWF<Cf7T~?+d^~b4A7>}#WA<$BD0^Ch
z+Ojn`%C{Ob*H+`Z`>!xY@fFTZe1Y3MUa%*o5)b89;@Tlq=rpYg=Q&s5^wU)cTCZ@Z
zDeE`)tI@Tr8ZW%A#<CyPxS_Wiw@cOFIfWXe=4#Z=smAt8)%g2-HGayh#wF*f@xRN}
zSagm5E#(^2Rjt8C+BF#6&E2vcH5k6P1|Qhh;J$rge4Q!A13k<yx$MQ}*C?B8F-FZ1
z<31h_o;sev|DSJ`^5eubG4AG%^R5tMI_pIP^=i>or4CiT7*fUmu+2yAz~c<k8%1MM
zWR{u9W5BcPvLWRa8B$}jA*J`{&S37Ixd`krwl<;`d)`yIai*N-P8;_bGvBx}-)J-6
ze7kHyQTO?6u?dZDGNHNb$NAUE9Xg?gwCN`2w2$(Qgk1*om_5Sl^$bYyvp(;DSf^nh
z%_1K?iVoDHBwIb2m#fRY(7H5RR+kp^j-<{>BPk|Jho0?WmSLaGfTcP#W~vTd`A>&#
zY}TQ7Cj95X5p+y%1b0wt)6l!x)E25u(--Ry9@nMf?vbRE#eQPebhavwq=z{=eD6w!
ztX_<u%w?PjzrmeLs;u8QXwlB|nzXQWIBD{3Dt_@W3U=3^fN%j#%jVrwff^Z952XjK
ztp9Zk;SQxCv`%?2nF&<*F4-VD#Jefuu>+}KsuD%dSEP$p3Y4@%p5`5pqibohwDs`-
za%z#GI%yeF*OjKu=~5(bCP`1%Nl@yx{uJQJ`N6-vs5<Z;x-aj>+frTV2pwo$`U@Kl
z{zNBd?(|#v9q-KkhT7x4Vxb!6!hd|iDR<lPug^yuq0@%tw_7oONej*^Yr^A;8@cEA
z9ZneX76n^h<EZoX__3RJqwFpIP+84;s#hqXP=(_jy+p6D7pS=YImXYdKv~0bJU^@q
z_iL15yiN(uR(*yqJD%df7f<jHdvC6#6rxekV;noL0JGn69(>_L{IKypI?cU{zC-fx
z)~{RG*?R+*&bW@lF6Q74yoxICE@N=?C0w>48@Cx=#OAj1IE?d@TI<gtDWAoM1sT}8
zm%YUnY4~vB84Te(xM0I+43|s6dWB^CHs%x>O-sVDV-rzja005e#-ZqHES}gMgLjri
zqw?4&+&VTA@7YA)&a2_Lac($z%7o(<zCXNBDIDib<+q)}Q8_vs4R42|*~@U0YYfL{
z?csRrYd9W1J_W8dPla)v(_qi(nGnTXQ^&o<W_{LR_R<;-9IyuSmF8eF*c`gF&0yOi
zGdLhL;~b0`)G4n3wT=~Ve406&8nOax{w{~u)aCH>%5spcS`Iq~nL!JG{CocRYd`t(
zzO8_g&E~L*J8>{-1Dr_N1f!;IfuIRn!Tal0(21~v38i*mV7LM5|7?J5X`8{3eYOdO
z_HZJ~39`m=zQ}wB92)Ne%GEBAJ=GNky1RnCiYuJ!a^YK)BACj)nvUHfh_MvG?END6
zcZRdyoUN)qDS}4}MWB}{gp42`m^{V@uIuiEF$;X5{}f*sRO|!KOT1xUfHx%ec!AK!
z3#iQ#w(@KJi3oSdemWnjjOIbq;W=>DbT+i~609FU(EAlZ;}^oo;{*#9EQTwr+n7FI
z4R=gdgX-c{aBJF17_VRnr$3m(#X08ihBdG9%jW#wX#ttjS?7AW8WQI7E--x!MCz`C
zVU_D3$A2U2`ndr-@2v;5p0#kZ{~9RSw;H_iEMbzaIk<N&gF`Ep!R}FupiN;BBu6Y`
zcCvtPH!NU|z6I>9Hiylh&7t?RB}hv19x;vYZFsMT?W_@<wzq=~N{&#a?*!@3oFU3+
z2Z(Jsr?$(58PpYSn!CaB-)=BD&>duXcQ)d&2Sj9gLU`dW&>Z6p5w$)rE|<OC>@`mE
z^M{H51;Vu#L11)07~ZDwE!OzGuqJ6A2(m)p_nrNaRe1n1|Ag{2jDv9O`XTl&AA!Nc
zkHL-G$02a~NoY714oZ@d5ITzQw)0)nc$XONABY9H_k0gSkO1rUB|+=PR9Mgb&u{i+
zzyzk@hH+Ui<JC=2Ov(k@oP73jKZcNJg>dV25ojKM0)^^NA@VWnIC)QD(dT0Ln#vlF
z)?Judb{B4qyARIC9>O8NQi$~_hl$f)KwEh&7(0FdGwF|@HKZLn&$q*#=60A{)ef+;
z1)i8T!K+=3U>5TpqINce{h-fa_w_q`N$G{1y%IQoYJUt(djl>`Vz7T#17{WM;Fe|`
zr2kh38nf!5-?Tqqe7T!%W&H)S7k^>JsK2nyxf@o<eFC}Sm#{SQIlTE-0#z?c_<Q3C
z1T_?b-u*{#>f%Fq;r0lQL_dHdAMb-xb18T)uYl+A_Bid92deJ$K_zCJf_J{S?~N~R
z;>qaejYUJfaBH|5PGr7$z19u)bh)7WQ8#>$=Y}Irdg9dnUg-RF7fw9qg-MM8*xVI>
zCk_W=QhhLHb_8RO)PCH=eA7N?KbF?VU@Y^^a^{;ItlxYVa2GoBO%L<U)?>+dfca+P
zg=AdLd{cRp`$ywb*gwRb7lL%|KTStR=9|6DH$P{kqariTZv9le#QAdDI@WrkQ_zd&
z{0Yv7wx@9KXbNupeF{syoWe=pPhpLI7Cwr+gKox;apAbfcyh&K>{cwou#mS{#QkQu
z*WRMm)wg)@;{Rjmyu)(f-}i6ty=fVdm02mB=i44l4P=X~RI-y9vJwrI(9klPh=h_9
zr9w%|-h1!)_+6jx@%!UC8g8DW`;q(Ab)Dk@D*_&H%IJHH^^D<d_ZZf3j$s$47)Hs*
zbtG5buX_$Vj?3fwwt0N?{4#xuFLQ9|WzP6^nNADy`Sx)U`yao_npHQM-@1$qoXU7(
zXBq$8F5}XX<utz{o|N(mUbtVu^&cweB78GW_$Hu1B_GHx=kJLMUKGALBYdMHd=rpZ
zLC@q0_LbM0Vwr6<E7?S&k_U7unJ#y&i9;(ne_|y~HdS#`q-c+YSKg)x*Bq;&PE-{`
zglRU(ILlm=8SY%g^)8Za;x0KuYpQrYp^87Zh|Wj$Jb7B;%Zj?qt`EAY*)N50Bv)qo
zy>2R2_@+?iez^FtW4ft1C%UOGS;9S+Mcdm<PYn{j87zI=X}03Mnj(37GxXFU;hWC=
zyQ`L-J=B1Z9?DO;E$=0Z-sVgXHKeqM+W4r4D!(e(H?#EAgjCtv$ZcwyuiT}CZyJiX
zsj6XDHQ|M>8ZBLyN4IoQ3!J;Cm*}F(u60(6wL7b@W}Vdz;Ts*{n@;^ZDU0BaYVq2R
zN>_Mh#ITO)y?#gK*Q=xIsMk?_))ei{3?0>_fsQhfZUxwBtIElexA>oy>iwyc8nIp2
zMsjY3)a#_$U+AbR7D~rQ9bu&a9ksWSjw)ZJtqRK}r%_K!y>-!4nuj#h(8><VyH*Ev
zLGlS&TeVX`^V_P9YuhN3U9HuJFu83ev{ZTNE!4=I=ITp9Gqt5yGH=S7sMuSLmByn+
zO80F;)%kA&x%Wx`NT2%ZsC7M+xumXg3l`1%>Do%Gyq5a=zNYf8Ra3Q+F3lY})s;#2
zYRYTcf70#rhxMQSWWO`t`QY7GUSIl!zIvZ|=-)^7dHI10?!4!<l6Tb2e#>4lZ`f(s
zYdZ9O#haBcM4$JZG5?<O&!#8z`SVEhI}d4p@jg=<+~ZfvJ8ZJ=7TtbVvf=6qc5PYC
z@7Xu0v#FFGHaGa7X9@2%E#|!cuCepCt6cG;i1!*6(f4-&x81nHMG^UY;gLrN$IDDN
zzQk6abGUkoWEDTYK>eCo)cTUaIc2hMy^_vZkJH$x=P8;+r1DeW6I@t!j0HX^JYt#5
zo6VBAu2?d%<W4=)^eAm#N_O$yL%e4Z$1aUy+2Bz$XB0;<?A}3sE{UZ6>Ilv}63(e7
z!svELIyATL=fsKoXxv`BP~{=qy*HR%3xarY;vTuF2U0sDfIo)?@cKW02EFs=`RV~|
z-Xnl(#s|=LRREp+0yuqt0K10=&><{<1_uJDv(X6MUl=3#i!tKrnP5~`6Lc5`=@J6Q
z2aLswo@Tf<+7ivmEpe}f71o+q!Evz_j8d)O7HW-AkF4>q<v6suWQ|u_E%7MV0+S9|
zA}wEjPh%@w8EJ(#i>$Co_CC%Nt<kI08dJh;P$+wy8(%HaB+~*<Y%EaH*8=;m*y4tk
zJ-)|HK(c6&-|nA^SDCX>E;)e7whQ2U&l!L67b0xD3pz=UgQ{}D8Xq?dOLN0c&xL4X
zvJhu=7Gi+xSuXtYK->AA__$l{R3YO3Ug?Ptxl6ruSc(6vT+!gO3vLIv!1Ks-Ol&a?
z9sA8jyZXyed*BNEys{j6|1A@(#1iR0pN;_?#$sQ)k#buZg7%LIuV`RH4<OSRi0%OV
zG$OL6C_Fkb6rrchu&}@AWJ>I@{j(j?AKBt_iVd#F?#uU{HCB~d<GK4dY~E)BKiXla
z<9KvfHxbk2juhT}3jVg9iuj9Du`P8f@(ZS5yY6IUWlFx{EqnBgvO)E+R_L?F93QTa
zL0^qA`1fZtZnYVM0Ut*r^W<2(nr(&4T`geo!wP5G$xf$(J(9Of#4h=>{Z~&%>urwk
zXy=5v{paG_%X#pewE!Pa$v)@LLaZL(it&Bj5d6#yjfHj0G~LnglRI3hEx|X-rSQ15
z3_qe*U}c~OqISw2Y{eSX7`+amwKt&MTrULKZNv=8Eo?nuv*b&Aql5PrL>=0ShGpB3
zp|Jz;(t|O)gD+gq??O)-KV;?kqkAvO4_dhgL9s#5xE_qIHTI(W&`{LbvJdaC?#Bkp
zFpS(9h0P6P;cs>bbG45ksCGPDuAIh=UYYP7n}d;o*=YGH8~s0LA?{^1T!eWF>*nHO
z(pj8;au%v~2BsEfz&RxoCNIw8;m-@`S#%k<mR-Z)k`iQoNkf`dI>r>7fu+Vd40g=L
zxzt>QHp#<plPidST!?EK*WhvVHjIn!z{j=%p$?Tez2z~Ug#CwEMb%jS)HBrUE}G`P
zH?i~gHRR?N;cDjs4C`BfR>ywAZ~reG5w004o~)@aKfz_;HJFDLqs-w3S{yCLoF7Hl
zDw?4|5&3wrD-VraE<$I0HqL*Q4Cl%$m_5ycf5lat=~s*ok#qP`x*T3jT}g)m4+d6w
z(DslA%e$}ONTcO^aoL^zvcK81UKmH%rsHQfuG#L+y6NsL7_o$lyq9v~yJgIKy^Q}7
zmvi-(b*$HTJ-3H>G5EO`vogIo`;RxbHQmC8;d}W&_-13BaJDWMow0bw|Hg!I+H&!7
z=7#fRemD;oh0|DeI8FK<q-jVLhy9G=+-e6ot92B|w~OM_4pD5a6~#B&QS4rRkO!&`
z@^tV)PL>gE6`#u7_LG~I@XdlR5$ydYf;-<waPX%Hu4^926{S(^RT;&uRZ)!H7s;AU
zk~q;Yoij$J(;Df#ZJf^R8EI@E9m6I&Vi@ce!_fILd_OaWHEd-k^iOu;Z=$(E?<kj+
zo#uz<r@8gUX~x8!X7=jS?7Jh6V?X7y`mZAH(7Q?ZR%P5`QO3yaWn6!wj7|QQ{U6`V
zDXL)5?Fz0CzF8@JbMQw6y_XeJ&%Kzlou!-kLb3QoN?6xh9wU5XBYe|a*ye`JQ@M?4
zNG{1@+2LgARWfRDC5tU8nQc``2(PR;E&H33a;K78)+zCicRWxfxviCq99hX4A(!}8
zJT1OWFH`?P0RwEx*<1X-p>9>2a!T|$A8zrkbRVqD7j7x<rrzI{&ZbJ~qRf$enRw|m
z2$M{jgWbfp+)aHu-%UAP>ZUq1l`aSAM`$rnPdyzhY$JS=KV6tb_-3<szuO7lybJ1~
zOts|RrJ<*43*X$59nRZ}qCK|OQ&SDPsos(!{7!CDZo)T%hIUmuI(Ah@>U32D@93&=
zl222%s*9RFy^H!II^e$9omIo_ot3xnO{(xs$5EZsx{{7+bx=q3cvVLgJ*A`SZPZa|
zbrM~$bbSZk)lrLdbyS}bI;!bsZM8m5TbcdSQijs6u)bI_`cCVpiL$c^f7(&aIw4;0
zWs>73TI5f;I;yX|j+$Srt>hI^1s~8-i(X5PqvScRo~NNUgme(kR(sX)g=9$9m%OFU
zZPk2(HqrsoT3Oh&QV$$OTkO_CrF%42wbnON+a%NP?&c;cV{2oTyR(rxAJkB-mdu+@
zXX>jnW%X3EA9dAFt-5mOtD}mYYpaGkYpHsNYpUj%HI&}f>gr2HHC2%Dk5ey5pJwuJ
zzFYK@+opV{ec)FbH~qr-=RYw$@FNA9dD!DU<E1aW)uOk|cX-3f-mhscxqkjZFSxk(
zb8b2Pl-o_7uvzLO*3)^&2g~nsyW|CwzPZf@18&hYxRM<@$c{0ij4>;2a>kHS_Gxp2
zC%zZc_K{@YlwD)Y^{X6vvxxgY6mpSW0Rz3S@QO!1CmZH*PM6E9FWQ@3@i{axkp9eF
z7r6e|c~%|CV4&QR7lfn}S!rzD{1iX#NM)Y(31(hC#(it0FVi%cz5XQ9Ej@t=Zt+~E
zag@_?4zs=UA)2+2-R-3quG<++{RL4RGwC38pGL4(jR^X559cA#-8gqXz-s^Yu||F<
zTW;LTgZ)A{^Fa`&gzsT}k3c?I8^C*K{aJ40&%F13EV}E*nV<Zq*TJ6+hWoSCbbr>J
z@6W%p{rTC!pKmSwxqq}jkGdLR>oa5Iel~_xT@(D)HNm;bCb&I#Fc!KDMiZK1hL#xu
ztt_#imKExBmsx0qC<ALu-)jx;3Tv4BXN^VK*4W?N61`=YpuJwk$pZa{TOe|T1#+ia
zihkG%3yQ4ZJ<uA>u38~;xD`%3vqaDZ3#7|qR<xDZ<XBr2eX_%=B@<A;coO=qpNe;)
zxjn2m2Zx>Jp;3f0&W&D(OQtT!IOBr5<u2%6-xaowZm=D<5Q#k(BBu31<aS+%iez`Z
zez8h=B}9v3?TLlLHWgPF!sw?93@TkvV}%P&i1&Jd?2(FIOhxVYlM$isgm(JNv83yA
zEFZHB?i1%AqOa_!#*V<dYeUi9aR_RS0H*H&@>>BDpP1tNS5t`X9jkv4eg-4(GQ=G1
zL?`oZw>=sjv4i^_(fdxY!8v)3ZcVpFyIIya)qEUY&a;8t8(U-)+oQ!9(QErik7@hK
zn6Y>YK3GmgKl`Z|<S_-wWs{(_aw1kNut#g(arh#8o4=Pv;kL(cEUG>R(;tsUn>VA;
z>fvZSYCHxbwZ@9h$PDdwSm2PRHR3enzb_0mQgVqBMd#z1JstUj9kIY>HsV{&!RM&C
zn9zDYj&5H7_aDx<sOf?u;<H-o;)<r)Zs^%+5uB{taqseCELgh~B~Hun)p;ehu3QC&
z6{|6L##(p`T91-?UP%Azh0MPjvA@A)Jkj+=WAiO=St&i}36d%Oc{`dr_#m)+C(`Eb
z!lhTckviEQKaT~V$A>_)&<?_LyI^bz3W4q2y*S=&AAb7m$IU)rxVk4wbigr~s&fdn
zA0Eb!S}B-ll7^JRbkv=99=A?iknZ(tjEG1_<98?U{8=hCW}U*i!)d5?_%xcnIEOcf
zGB9UFCZ>6u$EO|_u=8~;hD2p!W$Y>JEJ{OI*E9I~<qS-8GJuum@wYSw;VUoW)4zNS
z+h2%~UsrK%L@8#TFT=xu75F!$5*CJ!v1j;yFxpd%MoG`$+_w~at4c(3eof9-MYz_s
z0B<!45Z>w=^tb#%<D5^ZSFH%MWlvLWtgzMnYiPBk1Z!fiW6QE4xc4f+%J6*b-kyj3
zzq8THDq9#d3mQdP`1vpkd+!zDbBki^i<`qXtCrBlawT0Zu3(#h6_VYzLVBW>(_3>f
z=ViKcm9IM+taE3~7dPg=abtXoMeMQ7oo!0p`Fi>iE{|Tyo6nZ<u;+5RRjgx3z4eSc
z;6<-&FV@(#neTsl^U7au8vpX9i||cVZSjI%592=Jn;hYr%~4@I=M%;OUSSND(e@AH
zlB5W>Pl@2rAn{XK9%PE_bCPx+WK-GC_=+ZHw{$*iE<4CqK?gZOm}Zj96d5}iOL^Qn
z;hWo^Blzl71fR$br=IZ52;rLv!Z+20Z(hiJ3yoxXcqH!~jHJeqNG?f@WRG*QL(&zN
zT5yyOX)#>YK8CI>VpzR?43qwg=A0MN-2Ob8NgtzmSw5Fz7Q?oZG3Hm2z%h$X^U1$7
z8q~?>F`p|Oed-F|w79`3)o!r8XBl4?l(BYN86Dr1adb`vH<wg!xoCto47tGw$rQ;v
zUCfIci`mJpn5$<M^Gi`NLp)1poh_NMt`(e?RKc6VH-5rHcdJ)&zeXifyHv8jVI^-1
zyWBFX<ihcynHW?>Xv%KpsO)ees+d!|fX{zlVXYaLSZ(1YT3x&(orRaVKIpP!`4w{L
zh$7Z47k!ZIaoRt-#S06&sZqi+x;G^2_f|KhRn|@Qy4Xz}6ut@A+fCJ$JiR^Xk~fpt
zO}&uE^^wQLb<|V$`s%6Qrh2Nmc&}WQyziu^BChKx_YFOi)}9{fr>36r6u!AD9hu(3
zH=ldzsSeV;;C`#CqU7yW2;bD*(N&p2yx_t&xnCrA@0_lhe^ggJUeZPVvg@L%KkBSr
z?CY$Sjqa>sC0lr=PbYPzZYSlD(@}Mm{mp|F9o6XZ9o6vaq8hs+ovVwq)x3Y&s=1c<
zyY+R{>*v~P{|s$a<gKOp){y+g3Qd)JT1&MGlYNhON0l?9qw<kkmdzt!n{7I3T00#z
zT6Dz&YHO=jPFm{qaZUA0a;#@}*HEbwI;d~!#rrSW11ltV;N9IeYV_OI>gCT?>S*0o
zYEqk)%DYPob+31G71>`h{RT8u(*`tAE(02?B!fmunKo4G#x_v@rq@?b*3?s14oc>Y
zc>mnK)K>R2YfDyQEoD5jrb=5`Lt$HW)jyz`y4>?0KkNNv{no#w1LX(rzxl>h-M>;d
z`ZIf2f8yd!9~sx|18cT;&jTIa@wV1m&T1yPH?LmNGT|ko>|gNCi)U;r{T`7YA5+`)
z5i_eEu&UR6R?d{1q#d_8Ft3Vs-76*kv7Fy#m9a~=o7DYvgZHaSc;%w>PNiOFpLp?3
z9llD(#3I_|6mp(v0Ug|~&}&gXH+{d%UazH(Av2dn&N(!ElPx#p3+yrPJZ&5^SZaBe
zy{4se@SZfT{&14ki&I&n!3nyaI>t!X6gu=w=JXemXC9Tno%Zoe{BwjA;fL91@FAv^
z$I@<13>Wu{X4&t998wl39`^|T2$7$elEbK*dVou!_w(-ReauFvWH5yAx=%1uy9M!F
zQ6O*c4&bp3{ydZ9$LbyZxT|b8!!PaTr2D&RDS4>f2l%nyC_hdc=EqOH{g~Fmk0F0{
z^Vo;o;;%A7pIL)&=YuhR)HK1+E+#lO!2~Bv2BU%S&3~q*xY*JR8+%(K{+A_gwH}S#
zeSwD_gRx|yDf$e8WbrGk|2!0fDn{a}8V!%77I3z-!1bXP$dq}kYl-%+EHT8+3c0(j
zFcem3Im8N{_bd_o&H{rTn8Wn0IgY)uz&OcudDC({a<)uDHG^qzoau=1^_;No=^PB*
zxBwc@o#7<eI5r;^La)RHt^adD(-v;RWeYLC{X$f$wGf}=b=q&6GnTek0P}Y9(PZpA
z40RDt)`2-N`sai#HJ$Kc*(@wyH4|I6Pe*3TRG7`244WeHh|RUb#2wbSIC-r2<VWIG
zkCEupbU1os48{F8z@ryndDaxug=-e%nPPsnsdRFgBEyR)ZA0W24ache_K0+|7mt}8
zy35zop@t1kO%N75Z-vbTR#+EkjU)OtxW3gEyN21LmfW-^h|cL=&q+A>P;wEgPsK{@
zsrWK>3RJ=*lvq!MMMry-Z@0#d$!2)ba3p%o9D+6WBe6DTG%7BSMz&1rnq$!C?l6?D
z9*MlRW*FMQ63I=+;p-qfjF~Y3##1I?X49##Sv(!JHq69BvsrL>J{$97KZaJa%eguq
zK3dNB{MZ>`4hs?0!UgMkxT3>u*^wE#Bcs(4oY!84F``fEH^&1Zo}TdYT!YTe>+pBN
z2JzmB);M7!wq|a^T<L8{ZL$S>Ms9`qhHYplz3r`Zd~nilC$yXH!r2|W(er~JJWK;%
z;TDKL!F!;6GYEcNL$JzoFY?PmQFou{jO965dV31~$lhsOtP3Xg6(8JlFN}E-f)+mx
zz|c7kHH!|z+A{(DhaAVc&Z(G?auO|%pTfYsY2qC}EqAJOxayUGr~#R{@gx&nG_xc#
zN%XrKC!w!(3K?zEk*RYAW51q3FWqxk=zShq{yF$#cNtgT=VP5;A=cKq2AiKZ(A&LC
zSgZo;`rpCXnm?gy{tp*ty~Ks-C8)Rk1|IsBU}WA^yr*b?Z1Zt3JrB8C3#7-S2=&~G
zu&kmGXM+lmpCcptn&8OmNG-gEJI#vFO1}W50irM7br~A2X*j8O7QdHdVfMu=Jo%P|
zb~nWXCYgrU66SF4q$OPYZaG_IuHcy6EBL)?IgdSG%<r$=nRC#c3pTlPrKdY@e{`eX
zGdFfCbE994MYNA~XY)9B_AYa0&5WgNU9pUJmMrJf*XuY>cRiORdNJvY7YBX!W`E(E
zJlW$IRd}<`8E<;&dGm?IRxTX7l^F-am@RzMT-e51=BZy8-Gy%+Cr7YPNF?XZ6rHi~
z&2ZVvJd!yXaFDjbHA`hW%lwaNCd&Meab^qKKxWs*gPix^AX~nS;D4_p=p;8Pn|~2J
z+#-_ax=H?*+{7IAM$$+4##8v_{-H?zcQlfHGa?yuK9Z5$53<reidSAoal~86eD4*_
zdSj#6|EXk&J&tCh_tD&1FNV)m3_q`n;rGHAmW+>OSGzR+O-ZHxtP{*@e4G_OQn+z(
z3X`vv@T7kkf5ers^^!8~*DhmGTq)nWh^C>}4c5=TOx?mvPC1du{S!}eaO0EwI3k5#
z8>R4_UJ4u8mD1@@DSO?x$pN=2cux3cZTm{jI9tKh`4zk`T4l|L6|7?~I~u!6_7tD_
zIQf~o)x=E66v*Je%^9?!+~*u~**q$jr-g6MPbg$Wpzuw3A(vJZvYSphM~HrBV5i#*
zO6jJw<@Pj1^hE7P++b24$#tW=zoJt1ece>LFpk54ZmOBQ^*<yX-*<(H#3x=uG&t3S
zZ;r@>y6UM_*YwnW;Ts>xx%n)7)AoP+8_^>V{42lzhvf4~SHkN{UDdyouBxr@&Drh3
zI>ud<d3*Wzwys)~CjO}dx+==0i{u`6QS&Q0tAJgd)%t#&)uo%ARPee^YMbzlz1)=Q
z3g1M@?xmN7aLPw*wMz1>mkP_Q5pA$h8{w5sI_iOVyn9D$sCq71Dr$wMnzKwpJ&Bf^
zliaetbnB>oH}0tVK9-E)7<s&vj;i@kTg{rKtsY+2QcLu;RKqozs^E-<Dtp~Q#kTIC
z{-V7ynbS^XZ)vM$9%!Rt4z*S_j<r&y87-A&Q496Hs=0dosF`~Ew5hUu(nRGvXslB1
zHd2@FG*l}eG*B;J*H=TT*H@=I)>H9g>#FA~>WD9^wz?p`sU|;as?>Hh)v^9HRN(OH
zYO6&xRl5H_h6MiMz74<V?)-y4mVV=r6JNNl*Jm!0%$wWsAGjy-J%7c&;~wb?zhC@@
z;a6X?M)E7ZbAQQ6ZC)_z@H4jV{gm19kLj)Vh~@hp@b-&)?AG`$$9K8S;PF-bBRA6Z
zW6Swda*Gd_mh!}j8@zF#gv-P`^>)*B_F8|9``2CNx{XCF^eJSpWdW^iu24i-Je+fx
zchfJ?cxNu(_sikg#B6&1%VKxk^GxcH!5jabp?!yRzMh#zqbnzQ!8VmoJ|5@2sAHTx
zHH9_WC)2YmksEvy*w!?jOYR?`h38>@XmCh&uCd&zAH#ogo9gL%kk(ToxwmHoJN1y<
ztU+PS7av1aH_0mgE%|(>_R`29gym0z*wtkZFMSK(xp04;Tj|F&dv<fgpIvNqY!?^A
z?c%NcUEK0_7YBCR%}f1v^QHD~@f7Z2T+uELi{B*~f4lf_q#?Y&8Nze&AXJVYgdu^0
zVD;Z1EZby^>J1gf4H}8V^+zLpkr~d|DD<^5K@($RIG7p9p4|Yq_Zgs3lmP}EG{Bzy
z257h40G{&<FeYdi3{r<lufj+)ZZryd8J1{!-x3F92b9>!3R%C+;U-+u%gY?cH=ALw
zxf!(TnB$x5f%@v$Lu1J#Sag_%XQCM?6J6}&b8~QS<pR+;%MM4hz0+jeelNtCwJy-U
z?~LkyoN@1(GY%YahT%A8;g<!HcQPM2Rr3(oW*$No%!Q%d95jn}Lh+H=c+!0q{+P|g
z=B3kd$$u)^-kJp02jgL9Xa~p6mKa(%8tz?3z&mp|c0V76as7#+Er8V?6U<+2f`Bqp
zoI7ZWv|Xln+sYI<!a9eG<z}_s6dU^uLZ4v<=ow@Tw+v$>Y&Ah*q&dpvYqdCRj?-2a
zsPl0gG#ZJY{-hmlb3Eo8od8$y<d<%lj7OGsu=2ITx}|p5aK{$sFWaE(oi%<HTVUm>
zF>w4K&uvTjf1^7Xe!oUw+S3te|6~L<{WgI|Uxk~(SV;w=5%$pxK7TCnUsD@wX=abb
z_a%SYZ4werO~K)_({OX!41DYBi0JFH@Hf;6e^1YapUHeI`#B$j3>IMO*9B<SZy~y0
zbiut)H$07XNB6QNFle(Jnd4WY`KDFq7`hrZ2i9Up=z9FM@<MBejTrB?38VLJMz0&*
zc;0*~j!LJ*=fLfVf3yR4M(xDLGroxGy&KPV`r&VZKW4uPz|{JCu*)_GLrw)FQ+F?(
zXokZ3|N74y5EL;LT7PE3S$0Y9Tdu~GD_+>YHw0%dMPayQocx_Rj8EPPC~A8Gy=+o3
zHtZy9LQkQ(e;T?sISWUha~Lu+11mmcAbeyd+PyoA1z(cz+WQ2y7N11S-_z*c<_zwB
zID-XE&tZ<r!lhLgp<8eX3q`N|a&aM!zr2R4SvT-s!cFx1RDsl&rTFT19i_9*qsH$`
zl4+EWriKM*GOGYTopRuQC<pD&<cOy}2c1vkz(U@?Q*;S22QH(wUOsHgFJpH9%aVVS
ziyqFGfcQduY*>IpmU&2J3N9}@j#K;3VA$L&Bot;LOYT&y8;Dj~|0?uP%;B0AOZYT-
zF;CxF&aBBR_+ai5&QfkHY2Ze?7p`=%b?1g!i+J|E8-wn<aeuKJV~)5{$H0v)8LnI}
zjMFqr^f}j;a^A#c>}RozQ}(Q-cZ2nOYP6nbKCI=l<h9iITEnSaL(dLtcq(@d<8E!_
zlh0ciGiECbLc;iXLl|evtXLby(v@M{wls{poDcA|>~IDM-+U2nnI-$17~z`%!ZlN6
z{@WwoDw%=8J>l}}Gh~cqmJ8qb$^_gx$T0;G-101fM*iX4&@YgkW%~3F<SqoVY-AwU
zY!Yqr|L2=U!Z+!{H>WN{^4a4^W|>Cu{F^A6y^G@E@@U%147(G}4KJg)t4<8tTgLFO
zcMSVnh~Wu{E+9LW&PU=nqau!Kaw4Y{r10#t6wW-E#jh{27_n8jM))RW;7x|g&%__a
z=N{MY8prvi&}2ah4^L0w1Ka=g?<DhKYASQOq_EwGWZH~M;qQ5+oLN#z$Csr{(!I$Z
zF*o^PVi}iRFXK(oN!V$W^ZX0xc|KmoMmi<5?^VpE?nzA8l*FPtC%C+wWW@H*pmuOB
z4>!Kdps9syv8Rx6R}1M}Qpg7xg>)HQ#+ChV@=@DTb}YO>FO3`g_N#=y%B52=tAsyN
zOL+h8HLe(ajmOsKbNQD%zK!UnGNt#@N%CtJY?iJlPq|%5Z<OQYZmQ98J!M{`r><=7
zp%S<CP&Yd0sqvym{wn?~C43Y7sjITf6&<ncZ<50#cX(S@)!eYFs%+C$wW-imjivLl
zv%lo@QCE!}-bFpT+gbhI+*wt1?W{CzbW&^9cTzT|I!ZrhM^zr!QDrTbJC%lxI-aPl
zCW&U|rG>Uy+)P{fZPQYvmRhRmS50*)Ohef_YN+%<nksL$bk65!slaR<wY^A3?Yy9)
ziVy0jZq7PtdRy7wBnjK}(N+cFTGE%GscMYTRE;-ls0YV7sG8;N)rVj0)Uz(_l-roL
z>cNyY%FeO1nmM<XTC=33>aj_5#R1J#m;KFD=L1bu-;gG%g<oT}YkMR0Z*xP{&%1$g
z-Bn-tM%Po%a_Xw>@9U_|o$9Dx<7=zXEw$AAV>MN5X$@uauDV+EubMKbTTOLq{2wR&
z`%S&qKl%I4cWxBzO-27N9F_2i1rt9~yUz!9>-nBvh6&%yf6L@;Z#dEaHGhe=xXR=u
zTi<<79p`7%e({7Fa~?C{$wNMNd?5O+dmNW}hXW*g$*M~9>jNt2BR;^&$eUcfzLb;2
ze>g?7#nmSi^R(@C&al142KHC!F`<YX?4*0DcLBRS&KG9SqqS&@C;DEZ#jIS`sg}c^
z%d^?_au!Fu&ZMe3$7*qB7*~9nrCrnb>A*>*_L2KkxilpC9^(wN6uzvR%;=m%K3tW+
zFwxBo$vwi!6A#n%P8`Fh#`2te|2=e-@5gor`CRgDKAsKd?DQ~B$vHru-2JR`Y@h6c
zLm50^FK6ThGi`W~<Pit*phf_z$N2Gr<8H<-+(n%NUshl5OY^n9Y@XuFzMp*gM`sr;
zba!!5HQ|{YUlw`$^4xe|n!%R??)F8s*M=~xVTAJmMw0zvgiv9e!wm-`wE~E_+z+#Q
z_Q&z%26!;m2tN~yFm$c3O#?%DT{S?pO$P80rm4BX07dQw=rq*;W;O<hnPmi>vqm^2
zb?Kq!hrr>7?04pmK*$<P(Nl;X_K!J+-8V<A_2wvaHxob6SbW`MhHY|F>2Tc^Z_VX(
ztz;5TeX+;9+2fJaR{A?57hr{}>=%NaQAaq(D$W@ljyt1Usxt;xIK%9MGyLtGalO4W
zP7PTAqk#FCUoa1w>d%AO<hj_>bPjapIibJg?L7P7h%+5#qM4eG1j%oHk>`NzZ|q?%
ze)rIx=D7TBB+S!>!`o{(bmc_etjS>1KVgFJ?k3nW#RQ@HgE4HKDGYxO#?l9aQNw63
zHtse?;!h*&X>W)H`vxL0YXByW>W@rA6SUef7^fPWqU8ir1PBxDU15smABFSU*~u-*
z9#sY7p&?B3S-KZq_O-*#KwF%eV~b0|Ku1GtpwY<&7mr!MYQYE$Jv9U~{DBG=6EuBj
zgoHXqXj*p=e72dOOC2Jv`Y<>rkHpE_W6<WgIig;RR;tJbI!o+fP$fCjbtj=#^(m5D
zJQb@%lhe7&Oc>Umh436F%v>`UZoTHA!)M9$xiAj{x-EeIOJ`Vol0L9Li!gKhV#L2$
zisZ2?pylU*x#Gh=er*jpR;|PJ=NshyvJu1EZ$gp9W)uZ^WB2_nSk!47l=PmvNAJLi
zA3pfu<coQwyYNxp56d0><@pal;<7+&%HM-6<AbqfVF+wQbJX?jM0}V!5eIdq;@-Cz
zIN51F#-5YxqHA8r?YtewHi-sVBMyJJAHky!DHx)80_lTNalrc|KKh-)4ZU>St8o@r
z=bjVaRR*e`&A`Rv3}Lr3IAWB9v;oJ^HYyc51!>~LJA=XR&Y<vP23idimROgG`I_<A
zIwb+N6Oz&D?Fj@}oyNOgIdIIpfx-K(W7PCBm~bZpu5+_-D<}uA-E&|(T(m#NIrum|
z2kkcIU}b0yZXd`&;PD*zPP+tM)61BWl83xAml06sGHizB;>Pt{bV-xULizRlfE2t6
zNQPzaBz$^t8k!DSh`lVlm7a?;NrkA@>MFwGixAX$E<YV#%rfys+4(Kv;$%0jG<9R|
zf39?`a%INvMYMY1#tl#1WS8y6Mn!I{+1r&f`#AF4tc85P-<7tq%Sk-sPObY(`K8)2
zMz}AP9KodwUb2L}r!Apf<Pt_TTt@B5o-}E@hBLC)aL%fY^jf`<E+0gLEPI@lb9OS%
zX(zkQ-bwRa+gQtIn{>kL;u*~Y-2FU^J#4}`PG+9WTibBHm&}|y4&e-%63#}`!>MM9
zCu{jZ`8VqzTYDX3R;lcE_C#=GX#_QUhtceP05v`YFk5Ee#{f=~c`M^EB#?LJL~#A-
zNR|oToEN^)$&RGmqeymr8p#5~C}zqIC#@`+cVtf9h-MwpH=eB-!_MN<y7e}Oum8m`
zvt=y9-o<iNd>j`)h-3Acq8A8Eq={h)$J|cl>RVZS@F<H7pJcIGP8OqDon>LuvmCZB
zjhgLJSkNSeyBnu)^nc0RCp>fIQ!;hurP84+nQ!Fdg)>T7*Rzx^(WNv#QA*8sH@I>{
z0nMcIXU3cp)V!O_G2Y3NmzK!w4vEw}AJ444@iYmH=j_i(yt*Zop%c=War_K-JW1n(
zW@#L6a*7`uPjXkpQCh4x%E+aKv>bbrzAWX#f*UOFaD%_UmhjA}5>5&);f7H88U|$X
zjJ&Q5b13BX$jcnH?GlHL%9ZZVi>xu|5WD5Y^5v)))(VZ_g}VQ7@Xx<|^XM;IG_Ij$
zN(Qe@kYod0Y@m+I551FXHdOa&H&mP3cT#yBJ1KVPs<dQxlP20?9pRg2$-2@>s;h1q
ziN^R>7d8Bsc&7G9FRE5&Rj;6vI{ryKQz^1{iIP0xIXbF9PxdaarBgmiTV0)}t(rWO
zE{uy>>bKmZZdz!mgP%3kkKG!o*XR!Fm*n&9y3<CL%Z|rql8y>nsiU%G8aPSDvE={t
zeI}Zlb=vBcXmvE)wUkGprfS+oQ#Bi>p&~rQ7ar1Hy*n*=k5Aev`$lb*fmR#!w|#5r
zGH<0!y0%nXja#S}X3f=8yJpJEwy9b*wuu@dt!&K)Hd2#%HdMX3HBhm=>#KFc>M4!6
z(iJYc;;N$Bs(zi?%HOn>8o02gQvNlRZ(?=T;e0jq-*)Mm-T0R?-G1}JgrA&c{+*7F
zUm14jGuKJxP0_UvoDu(?Gg98swEQh&o4#e^5t8dS`898je#OfAFFE4KbN=rAjHhCs
z&`DFg;n5G-v;PBL_qxZkOYZR8@LLR*9bj=xIjyFa@u%@k*3c`Zr}hmNYm~4^vsk+E
z<UXZ)jh5Z7N*`|#13DM-K=lH?&(CL%kUY-YAp0AqOPn}7mtV>+vT8&&b%L`vC^M6_
z51wN$qceQ8;k0x|oMJ!clMJbw%2sEN^XiIYY%nN=)4wONL24pf&rjg8X3{;CaD?X#
z4|7CH97k)#vgM9wPOK@uaLJQsJzKPOBgKz7O0v*x575hgKbNR|Y}_!E+fRjX34$f(
zZx4gz`+SF~zudWYQv<s=y}z$?RPE#e?&ROG@^4PyPIi91lb<{JGG5!4+n?>^0RNr*
zWwMh|zkE34whs-i_C@~7et7lM5Sy%x@cp_GzCJQSX_h|fS?D9nS|8W1=_AUk54LI!
zM86pWVe{Sq^Q#-8U1tOA?qz@<oeaclG*J4K24Y4Z1JURhAjDqS=A;oMRs$bW6;>7x
z!J)mw@ag?9)D4$=R(%W9D3$$9h8dD(%3bCADEyi{6zu|spsLtRvUknoHQo%#kL>Z*
zW<09ZoPaL}7r@)z86P$~BgEGkuY#P>ML6f;dS}U9mRuLn$p(&aMvtb>$ZNa+-y6?I
z*yefoba^hC-kyVg;y+(BzzMpXjhlszF#IqBukxqiwcR94NN~XL#`f@#KBsVJYc#JK
zDZZ!SFzlwFrzdRl+5|qeOt9^WF^&qa=xPqeiCTjZS~eJkH3nnwEo0R19E68DMp$*g
z0H3=K#NaRe(fn#Zuw_5Ye`<nn4uf%a|6sV^9gGQeOp)1{*!jg89?mw%x3`7%B3l?Y
zwu5C;J1pO63$N+6_)^Cf@!|=0?`DIWYb?<5*D#D+D-6@Z6tA3(VLRCfP2vqPaEB30
z(~O}hdnCItLog|Exa^=tVepNy=ylBknnBiZYGRAa<Lt3)&IB~Kalp^ElQHq~6lA$h
zL+Y3rI8k>tp1*TKp}%DMcsSw5;yJk8U>@$&Sb)aErQ7|uD^8lYBPVwWZVp|JWl<~9
z<KZfNtGNcv+pojV{u?Am*h@OIHlpbFCeXnfty8vO)aR|x?7kfdGj?FqVITN3_Jzs%
zUASMd8~%U&Q2NdvBZ~rXe#{;iu8{7Fp$=&CKh4o-$!oqh0U6R0(&^}AjM8>MjX_gT
zuxKqVZ<X%!gS*kGaRlrO595Mc0$RxarlLox^pT!|zEK)pJvt5VgQ6LJeiki;Wx()2
z2KM-##YwLq%#H|x=9d&S8E``K3e#ZUA{_%aoW+U>nb=U0iDc=;I#&{kZ3X*a_+J<j
zUqxZ9RSZ)5C*ktVBy>M;9A%%xqb~cNktOGFBQOI7+BxX@{32@A%0W)M9Ju*p;g;70
zEIpBpKecnv`g<-i=H^Q0SuX0W&Bbon@kD<RFW2V`1Sg$?v*?Dh7N_9R_aw}`kcda2
z33!~G2KP=`n30f++~h*cT3iIRzX<6oil9^P3N{wZ=jSV~+<nQF^^Ul5?G0C|99KrZ
zTqM0%ZcKXO#%VX**v`?FA0|7pM=wX-(R7sj&zWq~!HJFJmeoYG$HoiX`DMC0Bj&lY
zR<$LpoV<kpmM@`Kqh&lHeDk2j8vYW#spYYe-8^N7vwR~Pz23s#Pq%QlWa_lLvzAqr
zYnh+5m8rhlXyvqv8#VVc-S+_B1|49YUN|4hOvpaK@@EHlZ@g%gMb87t4vg9u&b-)g
zrWMG(`g1tH$Zg(K9y3>%rol;n+MV!cBYD4u{Ch{4Qkl5V0nGUtz*Q!pJZu?Cv*bul
z6~37vIm&~CZ+7WLaqpQZemyNW|0~hlcuC$?MRV+D$xb~U!@PGfbgUN3=PhFS_f#x5
zCWzPTa~xZF9-@9wBGW!5QCB*329_st<fCLp{7L2zmlR$Smia8Rpjirs$_#6oLJ#?P
zsobRMeo5wd;hXNWQn~4BGS3L#JR6t7m8(iw?q5osai!eT^aQ)&1Y?GtAgoh(Ks$xU
zvlEyqoy$F!C(th~o^^ubCF>`i*OtZeUXNt@1fS&m;YT^5<Ouza9HE_Dl2+;-;q)B&
zdKMpLANjQo9SV6nzkpS8o9fj52De-(;RVecE~|Hu{l*^Soo<PYKP$RV>2?0$eUxTv
zj&iwp@&}!bWglP3CUK1Aq7_m6xGaqQB;Rph&3(*E+RFpa_VR9G0aLFRaPQng7L*lo
zT2&$SZ_Dl|{Tek}m9YDdhin%7l+|P$CDYI8nDlx}&fmO#x@yn2E~?SKF3MQC-~U_C
zS!G}Bq%O|tq^hJJ>gj$R_1;nPZA3@BMl?3zL6S54LrV>jyV9aFT57ED&3WOQF7fTv
z`o<m9k4Ej))d#JW>%=zdgr&B!uco6iyXq*PemW|&r;Zv?OZeujw#u>9RzWwlRPB*k
zYDJ``@_eD8%(OL>)zA*A=fw7E)l%_=A8M<r%G#*t$8wu0ZKZY;v{YLvS}2S6%~kJO
z&DHV7&D7NTO;y+5jfK4$sfl+b&+oeQgkP?&T3)HALhj04s%Bj^*+epLJZh_w)3wyy
zPc>Cs*P7~tbq)2-wYqw<zM68W`;VqS|FG$uUu=K=2OW-o;~VMI9NqjgTkZbH-~7Pi
z+V9z*&pY0l_g0?MH*8Sr4XgEf%>wOLeE8r6r>%O<ai5=Z@~kIJD0#%F84o%C`+dIK
za+e;xZi|P!lDT&(cv^Ju2XEbE=j)}sa`lF2<4YKNtC;#vuJh;HYpnkLD!>0NVxMY7
z%=liw$+xd?Yf?VvdgbxKOyL_`Vy;duJ6C4Y&GQ2LUpvo^4Kq1?**VEbPUoK<r#U3@
z6fOFn<W1Qv4%>H}LuMS~3*8i+dM@7Zm_*t*B=FCFM>!|t2q(2W%t_vHlIa=4bn%5Z
zm7Q&<P9)9lg>!*qhi<=qfCdlu^Y^oTOuP}wC1HDMGbDtir-RtB?jBw<4B*p=e(bVq
z7nADv()FnijR*S3ZgvNw>ibY{sSiKA^5NvZJ89EpC->a&;k;=+jC{3&Q}*wW@1GqE
zSs~nVr7t4h^%E^)KOFno2W8Is$gtJNn~}mcBZPDE_0eEWA7~5T6ipk5Aj5&EJ7pk_
z+#87V9|q#q-GMME9EfY12cmG^K$QO-h~Y9{!i`{ceGv2`6wcmP@VyHd)i6cMS!29(
zH^QQC2I!q*h!L8`Flk_liBpN^i3+Pvjm6JqW^i#dL-R5-I4&|rzlY|yTVjqMX3p^S
zbjFD_&Zz6{jBEbRu)HgKhria?BAGC*ffG^N)&X00$Q|an4ZgLO>=|K}b;oT{*1!={
z+B>17=WKL2=ZLdqGjM41G<=`#0Hdzsu}|(nt8;9S@N_hK&l?G!mcvk8yx%wf$$hE7
z1o!S5W8eW}+zmB`r}*g$<ZbWcCa8K~f-51$;<XtBQ{lvV3uT9uF#v0F`oY7pA1Y_{
zMaF2^;Ut>C`>_d58|mYadmk)V+86gd6o#B0C%YjV{Cy%GHSu44H5`LGw?^W{&EXi-
zco@Eq9|E!6!A+u${Z|98j+^3qn0VbR2BC+qA^KM{6qYfB+nGUFy>u{^?GY_d`cT|E
zKLV%EkB03zGhAF~3CA>RWM8vEMZO(=$BoCw`4gepY7)ApOvWv<slpA@#8WgKhbtv}
zC(sdfgm-SsbNi#oJnR+yv2{NeY=7k@9UY5dp_W1atoX$nufk)~)%fSI77f;}$7^#h
zSU5_zc-$r&?%<96o3@~T{#G3PxDD=YcOY+^55At<iM9H>(A;-7>csgWbA>++v<g7O
zv;f>{J{gM}iWWM3JceH#k9-@+%YHLKn0Eq>brv0xWJ!K`KN=|wN2APj6s9<hg27SQ
z1^zI{`1)3&9UF%gtL%|;U_36@oQPNx2jn_TM!h!EaKzCGeIn<fI5rS%E%v~z+#i!V
z1WMlGaa^f+0@f{)k^D0lAKXG<_csI<+xDWPK`2b-M&gBZfJS^tz_83jMD0mNflex1
zAD%|ypR@RqeGc1$Gmw`je(p6Fv7q!K#%Y|#pix<fx4D4qkZkBZyogt++0dPP5f6^$
zpkI8Bbh785YF{?0b<F_3rsIa_f!b|9j%)LhFlR_2vL+>9?Dlwc8+0DOkK|&j@Xg-^
zMOZ9a9<{Ux;nl8SWJNwakLIIKa6YEF<YUGLSKgiO%HI`>SYPf^jyK#mQf^b#X1dB=
z+L6tgI@0gZ45mDv$=U@od0oSay(1UVtc^P-2;WpracA7W#r$Qqgw4h+;fbN1RH2^i
z`gb*hKdokR{u*Al+DJFyo0o1I+2w|(=x;okvDK4XedPV}we(Eg$~%7BIC0)CUTM2u
zJW>aEe*XcM96G?3=?D0zU_Td@Zspm%Uc4fc6znChk6v79=*3n8y;%NcC9lS<WXq!~
znU=Vc`UW0cR%111)?Ur@`m4FV>1rNpvzqO7R#V5zhufElf89QmTU*P0>P{pZY8>PP
zMDf}|={V1dX86WvHcyY{RoUVE6TS&ZjbX3%G2AD7Qz?8ir%4<O6609$K92F54@nkZ
zB1>K-@$t@N>E=mhc~!F9caphB7-p}qOtH*NVH?x`F^=%f3;DdU%vWI>pYzFb6G-Lm
zv&p<Fzjv}#3P)BZaiL2R(<R@v-Q`4%Zk@;-4HLPpdZJ{=Cosb=fyXzCF4H4HI^z?>
zgBH(UPVvlkmK@kg@oe<wDA%7l%I$tfIdlC{J{xj`n>!z2`WESP502(H<D-1o<|u<p
z3Ya{&K(tR+X#OsTy*lL3VObL24M?KCZX$!e$j{ZWN9noy2uoiaX2Bcr&x_Bj!3EiE
z4UFcz=~49V9>Fv94sg_zeasEs%ZgjU96Tk66AA*^IxdjaS4ghNus{ZC1k!TQMg|vn
zvCzVc*OI+?^4K2HgFNLk$-8;`vWuD_d{h3ti&_=jMY%NUqE=3pzIXl3YO-YUHLc!B
z4c)7wK1|n9DWb92{6t%M_R>~69%`xgC$-d!NG;`LAz6F#TB!vgtrYJ|=h(+qYIAyP
zwf#|RRkTS<nV-^Do2s-`Kgq_~^g~-66`f8!VWhks+Uj+pmb#)T8l9DzYUg>;7JuxZ
zmNx95&bMl>-s-he;S1U-b*PP+c&xS3@@b`9JX)&r+oW&mkYwImXr_z`o2qU((iNWE
zSjB`kQeoQ~s_f+r)HbL3N^5pKHEd;Fl^-cxQ@3lY49UFlTUblwB-B)&9@S8WO>3yb
zeXGlDu9{lr^B>nQ`@>J>zu33S51Kds#%A5W@ZpkAeEs1A@5j7nd!KhqjD5=$l6kYz
z=?yiCUemVDYmTY*iXAV!;7hCLY?JYnN41`?$EHU#t^1H|Oz&~ll3Se6;Fjz)#mjfO
zoR3D8(RAWXE^?B6;WF`sZz*AieZ}0GaGhN;uF<gID%Xp5%C1saxVV6k=~wt7AfM&)
z^Vn|KWxi{DiFq$`ST8f1QyngdC-Xd4+!t-}*mGPwNwRN>(>Tfg6xV%8<(=dcEcQ6g
zIisapvt<hHs^sr@NFsC1<UaM`C`WEN!U6vd@z^xk-^55p!>uUR>>z%F;0VsL6R*O^
zF!rA$yPIYExoN{bE^rIwFvGnxDG8>YX%Igy31snFe>yJO&6V!HJe29f-rhUdykI*o
zZQst>=eF~`^$rgEx`RQMJ{)7{!>5;aFjQ{`!-J%2X25p()Y{Gn$L?4(qYrxj?2DkR
zzUXsE9~<WBqp$34LOSW=ld(PmLiF+WpFZ}hKIm)TSF{2B@TS)Q^lUZ|#;*p#tZX23
zCJ%(g)&aQRcOd4J4@5%MKwMaEgvY0hF#CcL#>pMUP~QZxl_q$$cn~@qGsK>Jxo_<>
zMA+Ryco{Gl3&gro_=echcr@Z_n4y<wWc6DQh34j=n6}p(wngU9xM&W~W9Fzm-5ff*
z&4E$kKks6Vd=G0}O|gci#yI?pkgOgZ8-z(F%qGdYyRU7F-jcD`<*+SMq`#nhqAgBl
z*dn5?9XfeV#;QgRNRd6!%Llf&;%<f2S4N}Th+(j(J`{a^nZWg~2@W4M!J2c%s58SD
zEuI?U=1fD$3^G9cdIJoWUCp<KhWJo_5RxAoqK53WzE2r|QtN)mE9`?MQGK9UO&<lN
zz3@1r7rM>sjnD`Bxc#^fY?t)KLVtzp7B=u0U<o}BGucCoL5Hg&@w#9*cGMq+4Yotj
zJDh0O4lrFRd!IqFFY0Fu$7V*jCHKMSEe+sYCR(UqBmAjvg24NxXjP;zOY~1?<45Ar
z=`ol)!5pdc<u+(<jobk?=v>_nO){j*O7f*GzDz_jUFnY3ko=vd(_ndZG9K)mj?F$Z
zu_baA^cv5B!)0Nd1ZT_&bA{0cceo8-igUM?WB4!+xOjNte9#)C<gY{fI~#E8n->m`
z+yv8yoAGDb7NkdSMc%n>qP^aZdhd6@W719>d+3YS*1NHOiXU3k^~b0Q{ut458@|__
zj3ycraBrD(h1%Ik&!-)pT^x_Z=i{;c(|8QNZv*2WqN)CG6e?zp!sgrZ^(BqOvRb2X
zM?NmK8-rK2=BOcBr!b#!81-u+e)pV=(`p(rn>xY2<2*#4_D8$Af!O-l4{JvHLeI|^
z)yMC~@<Kl(G5{L40$|=dP`vVzHCVY9iPIzT@OC`nekNdAX(E#5CL{O338>d6U_R$8
zrhA>k=vU|QO8nfPM8h+~BMsWSPQ(0RI#$)lfQ_Tvm$b5Rd_^`6KFr3C!#TJpjFY!o
zc0BfHG0)))vYgWJ+3Y0l%{~UxM~RXlm4NSG;xS)#Qz`w=BlJ)%UL+P`{HH=#8yCo4
z^a}Rf%17zdd<==nhpXt4bx-9XW?dfU#LnZil!aU&zHr;YZY)c8<+|Im_*VF))zuk1
z`(ZlAIh3Qf)p&MjHj`;xow(+d?0ed}%dW<qKEgL;hD%tjnPemn^JHSMCq45$dF8`u
zhL^79o2)fl^Tv~{3OqUduqUszaOG>q`P?{qK9_sU=g~Gxc=(euANWgNpxZ8{UfD(c
zXg?m2JNncqoB3*w7pn?ZQ`wB?G|TY}G8@l4{qbDf#GFsFhSKz^5wBg>W9yfCT>49o
zBlZmAgWAKnOnW#N84TyOF~fNxXgDJ}jb)<MdUo2ejdvVFncyhBSx#Y0atUL|6Y*_5
zi(;gK<cY10=Gr9LbzF<)QQ@0P;hSHwzd0&=^G^8Yv2H9QlHz#hRUD19g*AeuBTzS)
ziE=j@5R}YP;hS*bn<ICGSA=acWERV`m-!`J<0-?}$$a`gneWaebKA*ew$V<d#>Qlt
zv`*%ngGpTLoJ0@djJKB(*<bi(o$!sB@Xh*f33L&@IV^kw;hVCN349+H&vN^CYLANN
zi>~o3DU<x?@S~z-l6#c!4TW#&wL8LJtD?DcWi($p#SyFGcz;P89mdCTi+&szevjoI
z+0QJfh~>8xv79hDmWSo@cH-5o`#F}=ykhxO_~z#Wc|HQ8`B6o)>i?1S-ElSk|NAIf
zw9!t3_Oc4sbt7eKkiGZYDA{{!Xi+4xDx;yYOT!FhL<8A-m%aD*`T70+cs)*?G|thZ
z`#RSQ6JPdA$+@{DQ+cwP3*_~5LSz!>zTChyjpaqtwFI6FP2m3F34COfz)jMx7Am~j
z63Z2Ax@H+)w-lVyIiBH$&$!U_IXfAfD7~&5D-WZMmAUhbm5URMl_8Ccl~1>2^HjVs
zF$2t%XTc`QI%^Z<^f6;4X0x$kUv91xuQyjR1l#QGVW<Rn7%HE88Y-2tw~>F!P$_tC
zs0^CbRPmH;g`2V|H6+eL`IBX#th{2O^p~CRq&XIfiR7kUMVc$;f0!vZ0|nn~HdTgS
zFi})*8!M@mMoL6=b0x*8xnjDqnbInvsUnKB;%083c(&75tVhcx!wOwxT#SygYPq(e
zHnxe<+_$kZs<oD~#8^|Q{-dt^{Hmr@X{sqry&5SxOByO6<*LdK3sog~l8Q1Qt%35n
zvcA&bOFbn=y`IwC;4fQ$_{F?}A2eL^O?m{raO~1g>{s@Ik9)lruJj#O)xP0qy*J{w
zspB!7I`-IJOZT_ZdsFv{I+-uI)Zqo&M?B-ATTeLB^)VCHK4jhHdn^qS-_1wq%l50`
z-6z-Ryx|%>s>Fl!t%?ao(gEI~l4l23&};TpmPK8m#g1}zl<d^zLzmb+=OWK<zQ70b
z$~e=nlszOHe&yX+=1DfZp!~G(F(<k7?{Uc?6>-n+0{ZJ5V}m()9991)s}3Ee?Xp7*
z8hDVEt_PT@vY(qS=kRP)HrFfrnDH!=K?^guzIHDsNdMK~sk``jxp1lC$#eg=o$kfk
z=vA~;cBhhAsh-Tc=36*LXEUcu=S|`CL<W^@p!YM;q083MZFMYfc}KJ3$hBOi8Oaa5
zA{eh7!N)@*c*ic1YjoB!Zuwev7_yd=&O~z9`v`t-8^MLbBR7u;=ec8QXfw+mOFB8i
zYK0?mrZ{2Q7AItGb;g{gf^(c*U=u25j|*0sxnkZBSIoKNjMxN6B*r-6a63oT^>M`G
zZB8(w2Rs|L#!1nsR_||v2VU**KD|8(kG03rJEGe(_eB2&Uf9~s6Rl^rlReJ1(7(_I
z(_LD_+@!U1{I*8q0}nJ)+@a&z5%WESr??5(9w#0OmYG-98)HuN#_nvvH(iCxncf?P
z?fRk9-+rha;fHtI{7~D_9|qC>Sg#d;(0KtEb0Glw`hgfYFc3du0`V<55K)<d;u{Hs
zL1iG+Ziz0G6A16k1CVMGh^|$A1xIy9baEG5E>_@iRr2zuJD_=F2juSZ!ci?RysvBr
zm%w(|`BJ>pGuvXYeH&=VV;Ix3EtcE3WBgqY<o)wN@+VhVYP(>*sw<|@8DG?$kpI>Z
zCNrJH3*;=lz0T0M>57K^T=78kz$-)B;#%KcIQOwTR1f;#)nRX3Q0<Cy{W{}y1koa<
z6TU`u#FiIcP%UeZ{snC@psVm^b**tEw++7ex5M8!PrMdA?bkz~V0UM%Tkj1o(W(4L
z^h8g~-YB@$2MtsDp|SKVE;I;~4XOV4+FiPqC0}LoItX?zL=$Q}1S4(_!w8j;$nhVI
zqTge2;MWA4l}^Cl0n>1DLI^JS&BPs5*+E}6Pqr!+VE3g($TwRG+Ahb6<twBcawVqi
zUyUza!(bI24kaQ2!<I+lzvXLD^d$<|8iPSLacE_|4!K$Duy5Bo?0giC@lS>zaz_vX
zp9bPrK_GO$iyta=0G59jZ<Op^JQj@Vbg(yWzVStxuP;uV^1-idKKQKdi}>E6aX#_I
z^H)9aq-}4wNMFH#h#<_oDUWCOC^X(C9oYvb;?Bi5bkmN<<e|}M>Jfvj-(xWCP%Mtz
zk3~j@IE*Qe$F=wI=yZM)4COf2OTwI_opAlN4_-U6VX!_2!?JU6>P`;IUhk7#&wW^S
zd>;-T$i=1phcL9=Q3STj$Esz=(BgU_=EoPKllYMRY)+%?($m5}ibiI13|7aFVc)r9
zqIc%weUm%{`5nQ+mIsivDjV0;_u=fCEG$@_j!5SdaNB(rW0qV(dG19(I(57S-!!^b
zhH1ykFn)6xdfJ!a){av2Yg&pAV@mOE>?HcxPoeI-3C!p)o=W^!*1t50A;}}zoj0&0
z=mzRvx*;B^8eDZ4!Oq=+=yyDXE6gN!CEjH1mNU6PZzki<htgtYC}#|q&Ieznuw%p|
zHjA6gzU#;Ha@2V4TsNMPY7>~}FoB=jOkhE)2|U_wJXacy;LWgs4C^(JYbOV=_s{^Y
z>=nRIUIDzL-kUpb_Tta7UW_>3i@)-Fu})5sV4LZYc6^#+$H;O!UZ}I<gim%1`eMh*
zT07o;WzEY&ZCIRV!$-3nsddenkB9ZAWwUW?H$%EKX6|Ixi4@){O<}Lisk|lrn?sIi
zY&a{8#nLGlP@Tp$;w!!__$K=JE}Ea%#c6V`3cj%x{^nBE9-gt5{?)`B9%`P;UxzZN
zb|izp4-1|s%wTav4s}oG(nIvBhfi{O;aM)vzsP0Pt6ZLWol8UEZ05@GD9GjXgSmX;
zafppwbGbDkhmnJF_|9H>dmd(U-kEGBIc2kb<~|PJzK>ZOgj1Q4$@;?G6nERlU%RuI
z?~}zg%;LMBnKZA;<g~0zPM;$jPIv}uT{5^NaxV`}Ph*piG@f0O#{Pm)3frgAUwqZp
zPg6PkQ7R4Zr*d(Pa59%tdES3Fn|0jH_Ue12=UDtkj)Li|cd^TD`8??*iIeAP^4e5Z
zh_092Ern%PvH`SrGs^_u1kXz1_rD3Osz~4#!8hZFiSN@effE`_Pi#HGIHDEKu8QZM
zH%s}%Yz04USjm_Zsq*-rF?;+oo|R4Z6BCV<jiZc}&_6~>Ou3Qr_LPw_^O%uheb7ib
zbJ$4f?ry9sm2FesfyT=BapuZ_?uN=WM?*!;$xxXrd)^<Un`USu(UF`Dl@aPqm7EGg
z<)E|phrP^|;}a|t)dLnv#cttpMp`Jt9-AxC|CuYfcg&O*Ze~i#JX0kn-9*`+W32eZ
z8Y#o3H&>><Yo^qf%_;pEO_dGL3?y%^ugra*r_3?dQ!aMXRTlQwQU2?!t@JgJj_h|@
z%DVfSirPsH<=F;xB_~`>S)1HQ`F^RPa>BTw(qVzBvg(?OQemT_)SJ*i(MuBU=6F42
z%ASARm-a_``hKyM{STHt_{z{@pJ`e4k$nt5@WJMHY|{TNE4#j7(abvPeh}W~L@fs`
zs%7WruX+2~D`sqe$=zBn_+QX7rfqz}yjzbrso_J`N^Yua=1unKQNvNf-3*(4jf4MI
zb9Z1h`$t#t_VG&Q)m5-b^9p8mx=OEMqEF2(=hU#vG?spWy)hRVv-kqFMhd=>?i*FL
z^SpKWtY}ds{FQf_;gu&@d+#_;mlg5)<pQ?8n$L#Xd5p+7LdWTc*`d!NhPFOPQ=<cP
z5?}q}(>c^zna#p>``Gt>CST6ZV1oyHxx{o2U$mDFs#d8S`*#Pg582Lc#-dF-Ze`iT
zWZI-{p+SlGZVHlEv0@Wf8YFVv*bUq~KA!&O>$oB*hViOV)O{GiYjeVxaUhI&i^6Di
zKa2st!fELpN#BY{c3vOJdlw@ZcqE*IOT*~XC5*!~!nm`q8d^SAlRgx6<W{I-<}eMo
z?AJhvuO^-xGeqAiC#)an0<|<3bTf8EimNN?ZaHI?;2TfDH~-o=;&)F+bXn$vs;}-a
z`{$0t01s3}w!u%~a#s7cgMW5=T&WgskMJA6EIe^zRC_FG-41<rh{iR#4IbX}z|U+C
zRG#sG&v7^C+PL9aBX^vB*jo0P+TzEXb_ieJ4hEWSQSzn@{yuDjXI?!pNixw^a&2q>
z(;Kh4`XOqiAIxk0&})G|X8-bsrubTPPXwT~cv$><1mecxKrG)VTGd|JXvz=7R@nm`
zdNB~Ha?SiB8{ENj0?_JuAAE4_A%F9`z`;Oto~7+^dVdFG&g_6UNnZF->WMGf?QvOh
zXBEPk>=-ZpYT>J9FZaL=RS)?bcL+I&q?T@|6-}?En-k_XbHF$WmLF;0i03MfXf@Rl
z!vqf=HFkzqxig}|T`(k0yiU(tuxyEocsSZ)*Ua`v4Qq!iYkQRMv_s^4JG>t31-q`E
z@Rpz9=rG}neA~eKya$x^9vCpPHGb8$!Ef=2Z<^|b<)1sEDvoG1z6&1K_rbB)?x?=r
z0|_I0AzfSghEDf|QLt=07>Zscd{$Cze;D`;z>nUt8KyA^Bf1a4lUc(MS2qIF?~aCs
z!+0#;J_%zqry{U_F!UydVtvQi&@Y>dp!y5Y>yz|FuUvwKbC+SG>I%uYEyuhrGtlax
zY*~zzjCfE4G&)D($@mBa>#fCNm1tb~7Yob!>kvF;9i|zsLxoNR`fL~?ykHO%B~UP6
zAgY`AqnmiEhxrUZx8(!yAT$tP-u1)YG#~M`c;j+fS4>~xgBOy~s}y|m=cX_Id3Hy?
z&ApHq(Fa!=2BG<0$;HQyLVt&Gh#x)?W6NT2MJoc0wWHDSW;9g93-s(rEIjYVqDDIo
zM@Gg;k8nJ4?`_0W%Z>Q{VI%f338g2J;J+;!YWg`?9+53PRu=9($-vC1nRx5H58>l-
zvAE$O7`!+FhgW%UvpNRrtD;}6FNWpt6L@gu6i$sQLb&BI%&E?Yibp<fPRPgnM|sG6
ze+0vR9LAux`%%{~8y?NGU@<xaTh;brN8K)5np%LX=SrkY>XLMYU&IGqfcKR$SO|Bs
zE4B<Lhm}EB@XZLdQq;e29^D;Ek<w)n6N|^Qf3NYfVLgHt%1Ew$ehmlO+`zB?H(+()
z239^0jAK{>I@G|jOCMUF^5ti%_H24ZoiPI%h!#>$GU5NQYrP}uv>j<M-JU-;+4ItV
zdv?)x=T}{Kn%1~+;caVvudt?hsWq<`S&Qb=oGC~3crRLyW>fSSHe8QKC+P9_JUzi!
zdeon$&%f`QvY@OPEsL5{?QL^j9dE~#v3BgW)r!7zTX5DmOIA*@<h}??YM-{`m=Y^K
z9%#d<MK;WKwB_wAC#t-2VaG%LsnL7_P12Uhws{H<lt_<GjNqHssqEe&jYUXfug+;y
z3U@K8Xct=-@1m*T8zaFs|H+AHxtrs~f0L}Yk8?(3vw`4=Gkr36_INrM%4v8eooS`%
z9A%urRgE(_qemwD2FV$l$)yu=IVLTaq5E^i?<0O4!8g<P<<eL1O_98>otI1REZJ7R
zn8O?9Io$amn{C%*vunvd-uIHNQ{mA|=L(;`E0SO2b?&wAnbbAT;)xF8zp=_<^_NT@
zFVE!S^h|!0Eie1Q=`4S?SNs-fygE9KtH-BtlW;hd9%;PtCzU#aZ#)Fwq&-Zf>Vs55
zI)i>zrP8b{m78A3$8V%EH!YR>6H>W1K9!^7<j1B;4knesD^mHyDwTI@QaIOe3zesv
z88CJ;m(5J#!j}m=vM+)EY!Z0ZIDtd-5*Vf_JkFnZHW43`)$@1`sfcH!+fo+kuHe$7
zmFy<pd%Em@^_%gGIiro0wX)mp`^89cyJ)1$K5nGk|KFbbez_HVbIjgY+3jMitm$m5
zT=NtUaA!m1h`OPYV{NDmw=`6iUp7z*ZyG3yV4PE#hRQzK!8Cngrc_FIpW8<>rGAQq
zlDox1=@?+43_fA5d>7x&vl27KLDx(wHcXZIaVCmYl(Ayc$w--e)=)8QR!@1g;XXqi
z-;q4hEA|<vuXG91Q><%rm4l6S6(>s_WxA@ia-gKKvO7^rX&k4i1WuMc1{-yyp{|;e
zY1>E%9M({=J))|_+NdhOVpNo`PZ}up_L2)9RbMjm^^^hL|M;ilAI|;xlg&4Mr(wIV
zy!87MJF0zT)WG+Ab?+_9w!Go5=sH%Mspb1&wNz%-^7`mnwz&41ku&abS&w_1X(2r~
zKb~^?#3$5<dPLXR4>-5JbXIr0$*zrSxM}S*CVsBw!+F(Axn9M?Evk5Fd?g)ID>$=8
zHm4e1<tfK2Y}TusUB+DIo8U{l7jjWJSLu54DHGkRl(*~7(O~~s9t@COll0RZcKjr(
z&K~E3^+jC1vVhY<^JugGC^vRL!rm<oiynGNHhmAWq`?9HSC`A>1vwnJFq^I1_OajX
zOx~TB!K6ETxk_aZ8#dm>glj3Z=quXP&TUMdw3X8nlBrs^g}1wG;hNCROq-rW-FBO}
zuyiA*YiwZgyY=k1ERKWsM>FX_Bu6OW?00evkN;jx{fn!)32T^rE{vy>BRIKNB$Y=I
zeDEZk%Vvbp({~MP4z6bL;?>;ZqK3j)HAH%-OFybQtOjaed5#98e;O(CG{q|`IOl{W
zyuWHff3Y*hXE@`^L<ih1ZHWg@TcVzWBkubOkJHx)=UX{L{gOMnXSt)fV4YU~wMMT)
zt<hP(9W*YqM}5f_`PK8p{`c)M*r6>l&$dRliyjyq;{jWD54o<n<7#j5q!hX0+j3|8
z_s$+E#dat-Z-)gZ?6C8eJ*?*0!&dmCW=q^qsN;b~jlIxAGFtb)dZT?=KcK`91OM~K
zl6U@CE1Q;<M+30tm*k^632!6UM(uTiadrh_z)|6H&PZRfY+v3;4TOelQ*IjMhv%L8
zAgNnV__-)B`zD#Pg{|RK<cNgemhjnRjl7zcsBCD9umSC`{bE~m9?%xs&$dCDj|XCI
zxFi0j8?LW)Lz<x*CSP~N#MzGcvD_YGvh1OM!XDQfIbi8@2OKbUf_kbGD*8L&MwL9L
z{ammx-vwT(E>M~8jNZQ;5i-sJ3kBaiSZ9Yz)9pm3alpXWj+nku^sukuo4V$Kp@IXe
z8o0yeg*((jTBC4sTj&U9wzgph@i<BUU_%84v%PVrfiEl<`Qq5E?ijVSC&t+J#*9+o
za-@rTgKR#G%?iLud2YT&_Qzq#k^MY25SOaN=PMY;B5Ncb`HjWP=F)k5@jqDio{k^k
zGw^BYEa+IvR@cz^*lV;9Tgn&V#<Il-NtRBW8?*55QwWwyuG{zgWK1ZS3>V3n53Zbx
zIN9X<gH`CM6^pGO;^gbCL&>EWe11O@??fj%U>S%pF#+%!Dx0K6z44`=KN@QJ!7|$q
zGdD>OQ&bQ9aPxuVVsFHc>xzH*U9jl24-!WB;%b#IQv7@|>uyi1?$8Ip?FS&uei*v8
z9fi4)gV%7Li1Fv6&~#Ne+MSHT+cVM7NfRC?B^D<y#iBvOI5e9Whoyt!Fj02{2K?TD
z$0s+U<C#R9@lAqx`DXMVy#;&kZ-HF(v14rpE?Z_|#=9)+67A|wm3VWmN-nF<A?WrO
z3{#trPSUSaG@}3k5rxPamXFspd1!qq59<%)A$oBhO!gi{z=<QUxp5ewU-u)wVGe>D
zWg=c@FMf~TgQh2ULG7dXo28@2sjvh~BG1A*y9`N1WeD9~hWQa?m@v2ud({Q&)GtNv
z!t)r_^E^W5o<rx6=Wu?}1q?h;0eAaq$XEc3TyEg1;2XX88^YJ!fQ3N~W?I$Y&Y~JL
zSXu-7AvLmTQYibZg}7FE7SG+!!Rf_4bV|C88-1^#Pn-8xWd9m7eP80stmm+5^&J1z
zJ;h?>F}CVF#<;+z(o^&tZQbhd)%d;a?0&$<{13=<Qj=agHHP@8G4{49KlN_Nqwx*-
z{cB_HYS4s}*O~L(ZVR?~X2B_+Etp%+l1e8_y3DrZ#S@k+F={~@#hN82Y-nz4%a1#3
zsolkeRg>KK#i&0El4N(?YX!HLtYE5nDz#swa$%D+s&-1Fm*ATRf^XIczHt+L<0$wh
zSYtPLIqzn^^&YC;&!SJ;eVo@ki`xau92ZQ}TCmNg;&c|BPUktf&!-}tukNQa@=H4B
zYh>_~Ne16JWbj#2;byFJIMY3cZkuyCJ0q9Z1>f8_mdlc(x$GzS#$52tFTETNzMsu)
z;o00{6Up6Hvf=I_M?x4kH$-yc!$@x2vx{vEv$(ZI76aa7a?|rnT3*Pcs(U7fWTms^
zuypY_rEz|xa5jNyJUA$gCt9a5#98z=!8hK*S&bHavtICxh2Wd-f@dyB@5roD!8G!I
zd5Ped)Ksn$e6vTejia3J>r**d@J(oRDt|RgrR$v(_S2N>j%a>`%aYhJL$Hiwb(c9M
zP_;z@2Q^EepLPQE)e~s?GoCL#$FrY!zR%s4Tw8fOEhGT(Z2xkWT}|b;>Cc!i_@;3W
zV<oxPNI7!TNZDFoq}0onw^NLi+c`!`?O!7$$;McD>R_xavNe{y1!JX4?`BHg9RsDs
z66q5@Y@mFS-GqQW2Fmmm2Fk3#(labOn16JeE2ZKe9@@F7GDdoY^S3ur7XEIiL`k-L
zYbSH%%n>tX?0Zw?F-?`TD@~Lu!XeqKG*jNcRaFKii%%%$E=^Y5X6eN{bZ`HPN8I%k
z6ZzbZcXgE9&)Ukk$|g#a(8fxeHd@MJA5F!+nTFCLQ%xy}Z=_^oG*luVsVbd&t12BU
zR21`GD$42f2Fg>(g%7Z=uPho|Pw{&Fmy7QIW=-Tzj#T;1Wvjl>yzM7;>+pf+qu;Ts
z@muz-sbl90wOm)fmLD@;)8D0*o~pInm3xo3L+<fQmwTLOdyg>>@A8xBeWne4$dwQ7
z@}caw&XN8kx6B(nt8<M@msd0IYZbeXtdcJ5N|tF>^2eYG&fjvC^QF%q<##z(v?%9e
z$PS7BC4L!jk$-z$;Ct6H?oun|w`=F<kaU(mQNo|QPjhSbN!BDE=ds!XdLGQ@V3WN6
z=W{Q7d6?DZhdAl+LD@(+z}Ij0Gx??9`@=b0Fgu%@9rtnVP0>=<WU$-Qy-a<-n_b_e
zF=|r^UryQ1*vhT4qnXTEAGUDZ>@9rod^5k=ZDxW)61#m$WWDH({Cp-sejnHK(1%$1
zi)T2!O9VZquc6WNRrGLO#bv)%GGpv2kqXwZ&OL(PzDsViUj&;p3+GdlHB8vOicR!a
zal@099RIHoJm;yQ#!(%kZ>ozQQUh&sHLyxs6Ky7G;@~Dt>=Ask<&7pTSZQH<KP@a@
zWq=n`tYOyI7KfhM;+v`?rs+GOM#UM)&0Ub2=>ofM?#L@~!^9hIa{UwSDXJYFy%Zkj
zxAY|ozVZ3m2D@K);C2@eJo0o$c90v^eRqZbQ&%i);ff)C&awq<kKuxEDs$`vlh|SI
zHaplR+2KY%dz={TisTRO(EsQO{}>{(+#Bon^+R9LUaANALreCa3PwsUDK`MmKgj;1
zM<5<e3B)YPB|Q@lPUzu4v^*6Ej}vm12EwSLU?Sl)29@-Nwrfvx5wCNj%kAJb!W}<s
zY!PT-g{l$@#M+soc$GPJ^f1Bd(Pk)p?S@mWJP>T+fzspd*zwQ}6(Mfe^uZO!cRJ$e
z3<r#_v_tmCme@7T4rVI$f=}#Wec2v|6CBX)tfTY;J44#Z(etho%)^{8HpmI58#!T!
z;()lJ_L94|!>qw}SX^h1L9-kYJ<SO@tzFP|nlmo_bU{pSSGe?dgV$|$^f)QLpL?Qn
z8G52wpALAspcA&9A`Du0M(>B-i0<P9^*UcDkv(v*g|v=d=#6eW`%11z_E9hSBlTba
zUKRyH(_#RMjRxZAt-&}sZa5l*jK+ga<FP$)GK&4CK`$=^?<;0PD}D}Ml+MHA{tK{h
z;v&5LAUf6YMTnCvcGc}6(AYj21z(3DB77)ZlZL``<OsOU9E!=EhQL&`qd``(ubv)&
z<tqX)qvH^~I_QV%BmD8Y!XKNn{ISifClup8&^y-$Rx!P?&%QTQq@#IEem7(;?*K1-
z$*?^U%}xXOeo(=K#a&T%(FafZ_Qda!ekha;jKcRrVRS;aN>_|Uw&z4V(pif^ufx!O
zb`;jdMPW!%G<xiaM(cmkFch5CK|4nHk{Dd?9t*FU4QN@g5$)e4VwzVH=1cG3no(Qe
zAo#|j-7Z{vn~w9@8TeX13l;D8A+kpf_Ds*gSKnN`dX<Z?cX_yxmyf0m3UIkz9(;G@
z;rh8elpe{0O0wXb!Fl)<bp#nLk6?Md!)Ws@7hk?-<CS3s`i1Vn$78#(#5)x>{f=Rh
zNdYPs6u|a)33?@-Ma$M@*tw+)ZfnZ$u3wq#dx}n#TZ-?Ql0m<H7G|<NS`v5;`diOo
z+oB5eEG$On8^!3;p&0J&#n7}Z#)GEC_^exluogAwBKW4wk{aomu7O%sAzBm{Vw6b{
zjy^bxo|66A^R^W8I#uDyjY_=TTZ!=jRd9)WjDR(dp)&6=mUVuNyoEK0n)pOI8J=OP
z!wcjY*P?t=9qM)cfE(vN!Zzp=)bD&oA-*Hz<4;U6`h$gE{vxtZ6Fwbjz$dW={FiLN
zb;k_&;i>_{@0imfz>*P(f^jrkaH5wrD~zn@)5wY@4_okbYzy`ea^>a~9!&4qj{9e{
zWBaDvST?pBb#M4ESw(o84ry#D_@-9y%~HWPBLv^%{}BG>;BLMapNjeIEUIhlWB)!`
zd~KA)Qo%A41=D;gN@r+cItR$xFXg_kSJPQjoWXXdGdQ(OG%Lw7x!ud4@0$!hQOV?O
z?MxbKW-~}5n~O}c*mF`At%|bP)nK1^7DSW!l1-!JY@UzEW`FZYX36RNCY((_g|mme
z2A%pSQnaaE4ARNsB+1wwsm<gR@!kZ9xA?ri>@gomXOqLJH2#=MbGtO|?Up9js5ItE
zFKp9Ra{h;JzNgYgcpCdB(g7rRCPa9fkLLvIZAxYDBEd9ysr)ZIm2(B(Xvnb<eDhQA
zO&`HGkKd;-<zEVePDv-@;>~QAp2X!#lB7>XIGdCN{;^G9a&<gSuE$GuKAyW9CNN9*
zn{$G190cFAeh|+?w<Y8DU@^P6FJt%z(YfY4WB76-#rKAhQeUvmu@WO?U7nG0CEZ9d
z+$>y5g^}{?v60f{nUT`yosp8FZLGAtYoxgLYNnV~87M=h87R;F43zoC21=Z~ePWP-
zVx@1W#MK!n->*pba6vO=@pnz7tgWgtm-Urn;(6JR@0`8s1IzN?vT?Zd=D134&LR^<
z%g;!u(9=+6O|Rozi-+<&-lFSR;Xnr7V#fYkG<u_}+#0W|gdf&X4$J1$)_qNsGlq>7
zoYYjBXlN>ptJRedQ#GZtRwE^%RYS#fw5l@aw&+sJRg@4-6~!X7fzs+&edXn+ddgIX
zdWuoXUrt>An+IF`WcrM6Y<2!K{|)=d=_B6rao$_5>GOsrT6G-u<2BDay{2}>D@M+L
z&4@eq7@c#Eb;0*o+x{N)_3jC#y~}3n?sB~IT@GA&n`L`%a$dh0j_Y$xy!+L(Rj-yk
zrz-lqkUjwaN*W)o;4bY7=8e3{lI>ThQ&G--KQ1%a^fGnYTw+sP6g~NZ?2(tT^wW8M
zKXH!p)|};~wk2$H>@**To#dhOMSSCUkn0*Akl*)Qu6&Zi(u0TCbon9r?LWx8s|T2T
zN4lThO7Bx{4*h0i^P$~7j;qO}e{6<i%f)|laVz&nCG$|DWHvpznN2jdve)q~Jknze
ztN(1~*-e{yzBY-@A2zXLVInzpBVWWPaE0W;uP=+Gl};3AWrZ<!&??3}U%~!8R`8GB
z3i0}_;H;BN8TM~JuMSwi4s~m|AY?T|&aGlx^$I49S;2MQD`<M65!S1y;rwGYSl&^`
z&p-`a5dLPCmL^(`(uC;-P4qdb36r;)m}jYlJAJj#bFCH}A8O(KOKaq8wZ)@(Eivtq
zBNit(;nsF1Op9?skGU>L`sj>22f5adaKoIxlF8aAIjjh|?wh(Hafmx+_Vhqfo;yDH
zxglNffL@LZhIVp6?Pn)!z2b<6sg6i4x0Ag?JE2+a&~vLD#z)zqk)}OrEyc5v>kho~
z#Qn9>Pt&9`dJO9V=aI5E`MNtw5Bj6uhya-F6EAID045j(ioYcgi&jbQIw=r`?RsOY
zO>dY8KalU$6a77V!nuA=917?RTVLY!rVcovA-t2lXifKRu%@~NvWHus=!g-TJZT2A
zOHHwRMN>S9YzDhM?nv4$Iq>%G=vC#0_VRe*qa0EAq9ty2Xoc+0wm7HS5?-BJp<{hJ
zRPD9HpicIP`)iLXTSuhXIw5F=6MO{Qge-7EU>_%_)^kFVo&&Btw!_nocF?Z1MY%lg
zGj`&4>fnUgEu64f{5)A%PU6>g#>?F<$ZqQ<-NqhJ7vJ|t+jcnQ;EBBJl3U*f999uN
z>SJeI%I}Kr!@Hr$8z1P-?+$h2p15V$8^#;^BFD%NBdYw6zQ`XJCyIATwn2ND41`OA
zp?DH85;H8uVSn%>IFFx-g`a|PuI)@LGns>@1Lq;&;(VNzzQ1+179erp0$66vLsaXT
z(!n?Z!MBG(^Ux41UOWW-Hx0qaMnjM<JFd<p1JEc<JT^!DktbSNq~Ozm#*%~k?T2$3
z{#c;ykHD9{=z7*0Du?=Df0%g7g?F+t?Fl>MZkYPi3lolXz>4&aSo2*n{4oj|J9ou-
zFCU~V?Sb4;vemUC5c@h06<Tj3B7?`GO;D(OuT`jZSOX1@C@km^h4l$h7<4oWYt*Ar
zu_YSUUqqur^B6oWibm9uXlyo(#hHom==Wp?%%1MV@lmPhHfk3v7w*BFVd=QzmLYkz
zOgu~8C;KBgP;Z+9;}bbB*_w;LqxWOzumk8Fe-QPGkK&MF9xg=5wuDX|0=pi?;b}*t
zw;&e>eyqpDy)m+35{=8!@f^P;9T)rW#^6?|aA}r`-hRiVSFr$gf^X`C!&xZ!#$C4*
z&+eYbidAKJ+P4hb|CK^ZwG<7zor8vS>ZJLcMaAb5eBN~$|I)?hJiHj+-HNf%wHU!y
zi?Fl22!@hNPu)-iHKQVwRTW}UX(6KX3o$Ug5Os3<`dlHN&Mv~8ykglIIfvM>Wrzs7
ziWbYS!ZfuKm{o<;Rgd91=dt9yuA|aQPOoa@w0eqIuNRo|{smN{Yw>H!8$|DTgZWMB
zke=}db3VUE<+b{Zf2z%|8ak|LW5E3(1~iB>VA3`NE*E@rRPfD+YvycgXUUxDmQ<Z-
z$+#|-Jp0^~ahpszCDN45<IK46q!}FpEopVzhM^j^RPSudd&bUu9@CAt9{I4UUK*{N
zrisrWjaiku*!79{Z=_STtMvM*T+U+M>nu+8%3{4%Sv)S<&@$mqdY(#WSwT80ex%83
zVaZ{AN#)f=iCmy0@~cN8`(4Q3t}7Wle?5cwwHX{;Ka*Q^GP$o!CihOyq>FGbj@6Nz
ze=nSM<>9nB5l-Kw+4NZ@IOjtIgPumPLs2+))rHF@b2znIMDR_wNY)vw<;xqp_@qe|
zb-xOKb1jpT2WL`mPzDz+P35%=+2NYDgKZ|u2C1iTIN}RV7H#UgrR2io=ZyWD%8}x|
zsVEokQA8?dif8zj=uTY($Gnxi{J{gMjFtDNTMNIFBwJu(rF&>n3e6X#u*)&Y@0Lps
z<H*fiD!689gmlR+OyUXIW}dlGxEhND4!sgj<<ksS_77#`>v(z#zPTm%M)hSpg9P7<
z9=C|8GnR6=)-t|-p2Di7!rv^CyprIUw&#tM4yTQj@uExF>^4$*Mj0uS#ETR7$Vho3
z$4maroKnF$^6`P9MYSw5kbISavayqaqViQ=nX_A8>AF^5IiMx?9W+p`UD8w3OVpK`
z(uT@T@xJtOsjpnU`Hzp)zw=Dh7X~K0W%DC-bei*;DcLV+n)HHJRZqET@qN}^x<l8Y
zw?u=fVf3~eJapwIrw!6osypf`X$d+?&UtNRbwm@z>!yZM)KXn}Zzekm|9zm3Xi*cE
zsVJLj8z_~xR21*AD$1mGD$3`o28yrJK-s*yzOwpMJ>}r9e~kL?FTVx-X2Q82^t}C*
z&;NX4@97`7D&!r_Yu>PWZXG+dujNs**RrGfiY>}s^0C<qhFrf#=j?mzCHO|e?H-3U
zxX17dce!}UT@JCh%fu14siJnB%_FX{TC(6J_SIanu8ODrN}uDbN)CNl!S^F7WE<`(
z@7rEwLGTs2>?!Aj>dXB0>k<o%FEPySBGs%f@QtQyGdz~fkIZvCJohZSw3BVkn$w&a
zeVkFd4)WBt42F%*psQO3x5ee~+U;z5O*+UoYY%ePkptXyem`4CZjfbZ>=L+>wdp%J
z&SVGkrSrj2ws;Ms!*6&<GE<*zVbt;@o*%uL{zaSFMQ}~0xXpYcxkuZR;$c6#iMrv5
zRP*1+z{Lr&A-ax*F0s^Iy_UD`tY%5^a(-LAl!ua+FvocbkAGfF-ye(E@xXjusJDOx
zo(p+w`66n3Sj27dOF7PaDOJ3dGPqAeM5{K!WY0$MKdBDG0UEfGt$~r6niw!b6Ke6A
z=qva}P4JDHg%$=0zA;#?g(b;a*mhD2I*vBjS!E0TTw81wTyt%`6YRx5Yc<&kCKH?x
zHpvNlM>*j_f(wEqmu0on1t&^e@cExB3Ld!Qv9Itv?_Ci;$^}l1oYDW46Bev+g5x^}
ztT`h*%@#Y1inc@HTsyQ0wL{4gJA~)kA>pnI)`q&`<vmaIT1T8!>5LZxx}f7CZ}C!e
zL)t%YoKW?~B#*8ritK{pl7-4y;R?vls9ZOPXP_STT-8Q<e{D20)<$e|ZMZ(s#_vx$
z*!oNt$Memgd)^%9_3V+~sRh>25-Mq?uo`WIvZKw=^@kzKHW~^a)fA`4Hb?0mBbbY~
z_+P0TwzqM@<2+leYhsHg!)-8pmNkz4wLyh#OLS9hg{~E?a6s_d&H?r)+U|hw;>$_i
z?1<$Zo$!6I69%+%k}hEf^qyjmjaTiUUfcpj6K!GN-wpw<9dJU=5qgFW=(WWG5nc|^
zTI~qK=T4Zh(gk1q-C*0<154Vs!7syhD35KA293n8<K6+54#Ze{=^?t>1)&4HG3Y`!
zJRR%{ooT}5RQ85hTwh%L+!t!D{jg_wKisqNmyLk_&|f_W-aCh5<&iPCGiV|vL{Gue
z&cQIx42A2t*_fj-53^^^M?<#-_`G@n4*#2vYf<yC>{_V&{!GB7)v{lHXfV3z4TkxN
z!Kl4F7(pD26DtQ`_e$Z@Ci~-fu55NZ77zLQe&{!+ANGy(gWsfXXc6dxt18|YCmQ65
zM;*{3-3xacdf|em7p56`Vc^&f=ylT*y=6n~s|isup)*{jdgFCscVunrE8JQDzC_8!
zh3QBvc`w^!mqWmXp}1ha3P<h3@Lrb9`+f;SOxrN@_^=ue21X;oAsU13L}Ts9XteAU
zjkZtXVA*X4THM}&t^anSb(>T)x}Szx-#zH^a4*_iPsbFE4E$S>j-EZ!C6AwuhlSbb
z6_G1C(SDe9K7i~62N4{56p_}+=<B%!-fxqz=6Vv`o^QhE35l@p7>j29F>pK^jgVE*
zFb|E!vG#EYp1K|SuTo$k_{PkjKs@sWC>c|N1*6Vl$MLgxbNf6_rkzKO-+45cd>E}a
z9EPjK5gZRV0@EA$_<F4bW=^NEYHTr14=jd;@HP_!--HXkNfUf?>Ov8&Z7;&8;3C|;
zC0-%HH`faakuUfrd|DyC#uZ}Tjzaht7DMM_F^qd0M@@rENYTE8p~K5@bJZmrA0hqD
z9WJ3)!bL2YcM%^4TtU&kO4NG3fLYlKjDGtPR{5_mb-`-{nAG9h^*7K8R_Aolq$b?a
zro~Mi>elJ;aIgV2)(ghjX+ZA+13K(8XLh+ckC|9<S$|7D8)HemFf&fmH{t!}ChQw-
zO3x@$8U&f~`hR8&&M;$UkvYGPv>=bz(qN4}lj<B<&)SC$jnn8Udkrtev!WmNzkPAZ
z7oEyttC}pn7q81|=PVivzIh|~=8NE)wI|bA?^hbDgb$g0U=xp|CGz^jM9~)#*~&bT
zdbWu?Ws=CH%@X<EFp*nblDHu-oh^dX*=|WX)q|sCOFf*4o5Km0HEeDjPR}mkycZfl
z>%0iIiU{Y!{BT;_4rk*(@-d4D=Jbt_-swncO<&7ztJd<-p|w<s*Ye)&we)w8ZV%TO
zhCGVliThhPwQ>vR%-q3xf^m+z3Xfx&#tL=GLJ7W61m6r1Pw{%eHztB_)C3nz5YFbR
zUn(E;Nu^aM+4K66!hf<6w!k2T%MT^<Vr(+2&TV10DM{QSr~3lQ+^<fe*2E+Zj!9sL
z{}QMc8BYzP5axFd=BME43?Dk3JMyM6X<i7w#f7rEau$DOETQ$qrJU)T!qRuoXi=rF
zI0(M^9ATuyoiS3r?lV$a#R$%sX{1C>G*S}k$^9pdl#u&^fi4&+PK8E_eYBA>;<|yd
zP4*H(g|8`gGf?VR=qsTM^_5Wv^pv+I+RDRVEv3~L4dru^nv%7zq4GLKMS18@U-=sP
zkH*#Cn6>o_lS1CoeSV$n(7qO4@Fg=lzF@aePx)E<KI4DiX5!(SoZhg8U(a9XqLS-;
zlX{&_vdQpqg^uEXQd@aDrLkhO<vU+L{LI$tKG5^!duH_fAiF3Hltk6{>@u&uvhGJc
zCGtlD<@)Xh%6o$b%8-Ei%9EIS%3<*qx9jtleLbZ6X6X;wb^XfovK4+}^n11pe#=R3
z>ZqDr%OS&Fb9~!ZeBJFOA7#Jb=7ML8y(;`o#yz@C75|OnJr;hs%a*6_vhja+`E$iB
zs;FP%lyTR1`%g7H46Ejx!&UTdTEz{kDjD%xG%49bo3CEMvaqW>|LzK#cfG=TOUgMT
z_c9aiU*cSqOFZ1{B4-+0VCJ7vW?wrmo!I9XFy$;ycPgRsL-OG*j`NE1L5>n!Q%|(m
z*SFJI8ka6vzif_~vX4>Xy>F$PPU~iS`D>?ih0jl+RlrW_rS0Hw(;e)-a6514Z>Lt#
zHs0x-%#zlL{4#J8cdyzcUgRVi-%H|i`_0^CEm(SQ68-jU;vv68@y=}I5Akxf>@<VP
zi>0^1BZPVb=5uBJMf_NQA&;+LKzpYJtQaw$@7xyfd(Zhix^6ymlNaz-{aO6ee}VYd
z7V>G)LbiRikOp^Daj;YsJ>~o^Qbl-A4f)%qfgRs85G44f(|S!<3%<z^eX6;+7KZiK
zLYu`}l1J4-cBB@@Zj{?rHki=S7FwQG*s|FI+r>9i-%mJ+Kqq{%aKgO4a=kw4gmXsD
zxFZ<n!c57X?{$H17iauw=#CA->AZO9j7-5Ek?E4*G7v6kjRPu|+T&!b9kvLj={w5~
zHDd(d2*!CTx8dXL;IHC}`CcvYywnD6PS$Y$(*n27TB5pv1)heQ!LHgA<uJwXEEBlv
znINToQ+Vd+$_|k(emBxVUYa%<Uu%L18BH)=Pa7v%YGZh%4hCG%#fLfi=n!F!^6eI|
zootCdGcAy)VupX(Mp!?;86sm1vD@7M*ZUZtTaNHK$wsg;HbL8tX2>6Bg>AiU5L9A?
zi3u&REXxW-zpOFU+!m>tEfM{#CGP9lA#9asZ~hK&s&T-*0^w?=spGK!?@j}4>!Dj?
z6{v4*go|r65f-b1a1VXh-e`p*FC3t$DSoFntq{1$1{dzPl5ClHlFS^@`>YcRVqLIv
zx*J~f^}ukwHaHp478*(IuqoA3w4#oJ4}fdE6znVRj04TP;_h;9G@RuFm*8G_*S!yV
zoAp8Q!`@gO)Eg0(`=H>1AGX{d0EdF1_%(JkW&}-uTls(Zx_LT$97Cb9bvCv<pNo^-
z=i~C^G3dQzEOtmY;1SvUN{OBfo85y^Z}4E$?H+_v2L|Fq2k{PH2twR}AiQ!JfO}s9
zFr=s-tc071*6)s$f4bpffj637^2UZ#Z`2IyjFOp(a4JA$p<uKPUO4#76EC7Xp>OPo
zP|-W<k~}eNp%=XKfxV48!}?oS{Fve^yPth=(AFRGevZYw9^-H#Wh#E`4o2OM8JOcU
z3)<N;P~|&aIJzM?_HPKz*^GyF;|UmjYc3jHSc0<&Yq8-{6f(-=;LvLa>Mrj<+Q*%6
zwn@db&@{CAu?y)BcjL~cbWFXJjuBncp>lmMhNkXCfL1!brtFdbx7`?@nuehkJMsSO
zRvgz&hP%rav^={8`;%9~`SB{OYq%PNbz@=JJ_e_bM5E`FXf*pB1r>YoUw4hc%d<(6
zn~%YSgfR3>TaPK*H{zmSqWI!AA!yn`OgwxLR+8I#_UaJy#~en5(GlG3e*}|4kK%0i
zeApd2g$%(qJpzhx)~6VHf^XK^6k}9%5km8d@NP#D%tMM~8><l81mA2Cd^2=PA#O@`
zZMBszx+VL-xP=eyFZO}YZXc|aujhQ~6yE(kCI2_4(Q?LVT=zST4MR^uZ^}gs%)f|Z
z-^-D+p#nLD&r!1ZIZg&%!}j>gNH)BL4AXOX6L=P@6HcM<h!)$=(%}PlUCtDJ>f%NN
z+NBv#Pv4Mk>E<jdG3Qbn3+bFQqDgaOW=}Qdk5psE3C2-dZA#M!;c^_zn4@jVa!XTs
z95CnJCKg;3VZo1nEqVT-J$tlqVrIAx&C`9vQ{%&v&(^Y^ZWQ;MM$x_dMh3UqDVS&{
zBWCPmVfapNX(AkkdkVvBQ`qhHCceL$$bk4n_6bO2wN4`cHA$pH{Y2h+zL76CMY7(q
zNUFHTQpF&asqr!FIy;8+T*I~2tJ%b8H4hG0&4zc^P;+D$vyX={wpBRatPJPxyl@)a
z3a8nRaIQ0p;BfH}tk@gDf%cJn8MBsCcdzBKqigv|@Qs0VL<dccW#`$kv|Y7<rI&@<
z-?xP$MVA^Uxb3R&H_?sKXepkTl*^)FT@*~SGF9}BRA$OaTqAuzf^V(~zWL=TKAhJn
z?7BCF_F5^_IGoI<+Q}Ry*O~41>-j^p0cGNP?%S}Q(23_d>v;BRAl=c%A@mClW{<YP
z)SNM$tp-nL^1*3*acUaBUI?SLY&jbit!K)Xja=|sGU74Mr58|N*|E?_$qYAAa@QFt
zeyfd?j8URhSr{oDk2P14zc*7pX*5%+hc{N##ZNO?yv90njg&nfn<@SN=_twZO%!*n
zCQ9cJ9c6Vr9py!#hH^h%I)uAxDQamNN>rGd@^E8AMTu5X4!YJ?^3DHn#?o&L(EUP<
zDet&vXdQ!Ezh=?Em*RJQ&MB!+IoV!%f9-Db^P3u;@wma$HP_f~Xf>BTsOFbm*LlDG
z4eoj@+ozB2G1=}h%NIOm$-VbH-~K&aeBM*N-h1Bp_m*3wucq+sOOC30%(iC2=lrU#
z+&8VSM2PohaWCoh>-d-ZgMLwO;t#rC{=&IkK60wvJ1!XXhL?ZUva;|ss~5has<-S*
zwS7U~RnJ+S@Qf2K+~e}S_n0*09^LKk@%+oXoLzL6YA0_o?YZEZyVbOKT+I{9#ZUaS
zit<;C)tf8%MWvDl)>P2mq=NU4US(4Et6XsM3ZEEVVNg&x^VVMGDB0DFdw-FAH7_zi
z{Q}>9D&?rl=UF`O9CJpV6}_`WFw$vGsTED?U^?f@2KdAM>FmEYo!)H(&-kYEO^>~t
zd386pG}}$b+BBNxq;h^}3N?G~<j{>fXk)yCF3q+I&P<@DXq(elCGb{*4a^)Qyl3tP
z_AcK*H^U^px|YQI>YMo_If*lWBuWS8OeP)~%kB@yvbgVfUeOF;hx}k3_Mgq{J+qlV
zeU4~obNH*n9KLBWn?rtvuw_9otu{>KuExRix)I6+J3{H#bS~fc&SUdk^H?xN6`BRA
z7@e<*HyNrZI;@J%k5%FHQxzX{8scf|hS>VNA+)0#AuOj6KGtdA=qxQ5$7-STGA&pK
zXd&KE3w>MZz^tYjF5WW6#+_#9=i-1?_Z@KOsUxnp7Eeir6W&QCdU}X6?6RD3ug)2k
z#~iWwz8#i1*yG1aXEc862p>;J9NOc69MNvNxY*;s8qsrvyGakWlYV16ygb}Wwu0<X
zY%F?JSSthzC*oLOg=^kc@W^k0+jcDwlw&F0Tr)JjVhS}cQ^f2rfktX`Om5v2N%^|y
zWv7cIEgihvqK#MOO)w;-2^LOkf(Or<ppkGohvRh-rszX8z+Co0O)w|K0{#D)A@HaP
zzMpH3K7P&M^vM8^6ZNsy-T>AY4WK*DP<R^?tiNT7ZI3OWG1C&WT`XZ!Zi#VaEznQZ
z8hcvUpr@X#oR-)pk7HYn9ztS1b<$Qob@EU9+{sb%Tc;I;KRd12`LC11u?A@G)Cfy6
zHK2B}F&<TEV^T|9^o!KTp6X`sy=IQZHCE^;JW;Eit+2GOJ+{?zgkPx>inmA}>pwRn
zn0er9a%*(_)CPt3+QG7?7aY7g;{4@KNOUAZCMdWO-x*&tyJB}{cZ`VZ4)2BDn17%r
zMih5PqgFjJd2Ap2)e6MA`-5>;Ji1#akC$DA|6tc~ItJGcf^EqE_bsb#^+nc5UtBcp
zj^JTEP%1v(fR8;eXM7*o-|UZ)?14}j8HB}#128<a7e-h2#H5^_X#2ef9w^dV^VbJs
zp83FSYBzjs<c(hayrK70!Ji?rvsFf9r2$jsbinD!UhqF6J88{4QFpx^j-G5UUCN#q
z`_~iMO*&#w5mC3KD=Js{z-vVx_?Y=2^xIhM>M{<s@1~%C%t(yZ8U?k_gK_20Agp>d
z2$sDEqvqid^iUfPpFbm@g#3r_dJAQ%d?n^=kA!x99NKi;fdiLzV1n;X>`hI<tr=<X
zI<pITD|h2U>~4gs?1BHwJ?N&t7hT5fMc<Bladgcd(LZ-XyK^cQjM;(j-?!r7gEjaq
zxMtAImGIxV5;`|mVx`(D1UC`x$S4MB$<eU#iH5<MDBS5DBN}f6E=PnRO=}JAC9J~j
zZY$9&egQ65Z-f$`h$GvQkl6R2@I;5u>E$6jA9WaoZw}*j=n=erbObYR=Sk-M6#fPk
zV^W`DjO|v8|GbK2FSi)uEQ%4*s2K6JMd*5QB$7{#!2ENAP_$qmCfNjGXq)bs-O2}v
z<_faiyJ1XQAACsi!HOauQ1H#p-=|R8?lh)(pT=X^I=?mSG@jlp!;O6xG4544EN+$~
zB;*Q)PQC)&>dVq2CEg%M+3Fg97IxBmxIgX`+;g>P{8@*)*6Y({hyg=Ho4QhGz>Nlm
zoVv@LUbh6xtTAF-xe=E%FlN#?V=mue%-_e1IcB*j_pULOT?bR?Gc{pkvN0PEF=iW8
z3$}2z;Qav>9Jt1UkEU62yI%`>FKt2F%odz7!iuKPU6>v3#lI&z@~m4YcJ}SW=0iI1
zXFhoC9JtoID-WA`b4{c#_2YecqohB5-uI_>?Evcj7(k2B!+3DcFqXfW$VEPrI9z=a
zQ_f6e%l{_uQi};}lQL2^Ax5&Kb_LaClRf9q3f7)l!8TjNWP2x!H=cyCq3|~WH)Y4`
zYdANlM9^+v1ot0};79Qu#}r5M&aSnz5quN3BZ{j(MsePhSlK;~WyP9Ub}85(+sd2y
zd&U+%vPolGi!`1TuhFk7qCE+|slJfP9>UusODC~`St?6KkE$_E<wis4DAtnu?xpaY
za4e%uxAH=-WHzWsVt>Q+{9?VH!|c{m_wG9CJYC1qA?rDQ!FnESAHqXFf@zW(%-OEN
zY`$+g^}kP}$L?v2OOor0WOC0Mt!JBp_4MAbk-r7s9Bc8M|3}h!$K}BP-@nkNg$8NS
z9#U4l&#R#jWt8304%vHCGNM6cMA_MU#*IRxz0<U!A+kfL@9XpX{c%0q?#Ck<*L`2-
zI<Iq{XN72@Cy7n`zP*i7E8A$aWjm#8Z>L#@y6CNYANwfzKK2>6RP7@x<n2!#{zb=r
zw$Y^QRw@iuw4Xmy$=-FIy!{aFEuFd1)Be~v1$)Iya`tJHWbL0Vkh34ZQO-U`PS$>P
zq?G*{w{G^z=OpZ>{{2VxJ%7@Uh2Mz|ekK#n+Sr|Ip(%RJbZ2KHJzVgX`2zJceO@(f
z<nEEJTLoFZDkJ@rQfAhb&}jJ*YWiD3OFov7Y|lD+vFbIwKKq)URlcHgte2u)KaqEF
zElpZqL+P(8$<L>Z`fVtpf2Fy!yfKF}SLO7pqn!E&RnUKNg|uu}5$Uq#P^1zXv*`<2
zochRYfDa_Y{63kw7OLLgOgG0i(bhhV^wjnp*(`odwNLALH>IA`r?JlAY;oVZIx@Xo
zM>|edGq0eWrtxNL1T(RR_bekz=2frCDxp}-64GbiCY^noqgRV)en=5X_AH`NdkZ;t
zTtLSy3aB6`pKhMZqe-uGX>s>lN|4AQ?~mEkT=<k`xIdvAo>?R_@i8@)@E&;aDO%=o
znw0!bQ`@Z5lwyCHrs<^7kLM?7QQwo~euy94PtxN0<0OBVZ=J7F$zysd=`oYsjWtYg
zUuGtqO`!Uj2~@<qq>G%XnYSi^u5xzx*@pz0VVOvwdYedX{Tyny2qmS?AS!qeNZH#0
z=tf)sZRLM<3Jjrn(*x*JUMQ{ZTtL2`7LiBvVltb*m>w{v%VHyQ+x6#CV0IXVuA4(|
zpU<P$GZvER?j>~e+fqtjxQwR%>W-pwJ>YV>2Xs&Lz|5OHu&K2N_W$mIl|7{}db|`q
z`%58nnlutaq_KFbG#(w7=FX2a{DKwG!&w2zG71>{OCGg9mC#hK0eYeVpMwLC{bUdv
zuW^S+ox4K5X0UQH!=MpnkT)>HRnFGz-D=GHltxg~HsT(LA<|B=pK;X$J#0-Nxy~3x
zW$fL^86jt(5&C!;VdgbMj7~H}X0;*CG#J9$%Mjr|2jT2lJ)E`E!`2(R_-vvJ@7p@q
zG+B$e%>8lQra$ub_T%j!O~{?>i^+3)Lt94~KSwITb-W^`KI(<HiM=rKsRE>~_rl?=
zGQ6`Wi@?3om^@txL!PK1zn>PqXlXG=SQBNeld1-*A@+g<?i}k9qhp_oXA>($(Txfr
zT~Q%E+dUWOvZ}?6*gD~PR2j~DdvSiF7b-6Gg7*qVnD$e`qy<V?(e_nbo!%}?B)^EX
z>tBRf>^CuRW4ll@>kwmox<u>lZjexwM&MRC?Dp=7?m`i+c}kc&UKJi5YM9zf3y0Kn
zx&Nb&%jtvAzSa<yw*SlYm|!z&oVz=?k7Yd=-)>u>(8?N?oi@10z4lQxBk*aKEkcgi
zq5lnmf5||#3*iT|ewz9^K$&@phg%)-gm<?-TDm}M%2@pBnt)+C)3Lz87x(f5G0Y?k
z7bKZKv(p6*RnAcLb;itSXC!`fhRJys9OFJZ13d7{cnqp5$D`Mo3D9-&gc4_#-)!?h
zD0@ZgN?kGRf-C2d$D&7t3%cKMLH!A5xIGyK=<(;n`}L333Z#Y*lC21?*LdTr&>G73
zZD2OQ28*k$kXteol7|PwqiQ(*d5yplUt6eyFzmY{Ce=A(|GzPK8#w{WCKEB*!4noK
zv(TC0g{>F7SO-mmYtl5_KQ#>}<!4~_n;Dq7dlpR3`QS*pA5@ml!x7#y+a<XL>k6X~
zz*`ar*uVK<bp*Ru<8=2r4*LTqVC;AjPy9}zb?QkB4?KwjHE9SOehOo>Pa$k|8h+q7
zmMlNYS)jwPEr^2b-OU*GkN3}ABM`D9f;UPcFsv&AC$%DR+#(V`G$Y}zu?th8cfxmd
zG_ELbM}!)0TwL0Mp0SY#u#Lc2`?b&wSc%bF_aa*IFgiP~;Nkgem~4F=-Q%xwe(gH8
z&%S}4={NAJ>K05M@xBA=8*;sea^9)%W_{yhaSs+coWp5OLe}#=_&aMgGG$ky=pySI
zuW5*ic0pH&3qrpVrZA3e7;9lWYZrXl%dDZhE(m6QvrhUUXRIH>+36wT+#h1VnrArs
zK8JgOxsXoHg?&K*UOy^8{rMc+4}XT9oEdr_nT0bo88|vN6Q(hlSgfT$Hp^8>N~<?*
zp4yvSPxPj4oV&5>*N4vD=tEk=)G0P!o&HnlL(kOu(Ax2u)G$|*VsC4bVxN9w&K^$7
z#{Lvvs!1pIYVvE822F9%U`B`r<>skVzt!rLY_Cq2HPyMNrAFr>)#!b%zJzrJI3Sk|
zmyV~HF!>1Xt~mm`>qnsS_Xu32jv?9n7!-dU$GPqK_^Mrmf{r3|m=trqs2ER`i*e(3
z7UC~#fT7wtOtA5$<!;_|W{)==tMn$fF}@@h>q~bEeM#qnA8pGDB)8L>N&Eh08uxKC
zNvz&NxtUw2PuUh)`DY8QafqU&`#2x@Xb$VBIi#~=J7wS8L3+cYNuQamhI@9>`n;Xg
zEi9HMERChr%sgAUD3Q$iC(-dwta0M_Ud0{nx~^k%vhEmlyMK&E9zI5|^^Vcp#G}-$
z<_KLnmO`h#BvV1}gJdz{Af5B(yyxbFG%@KQoo(1pNk8K0fZ}fY!;f#QZSt$)X=imj
zMNQpJm*?%K#X}cT7r&oYV||l2VgZ#cnNQbOhSNNydGteN4&9s@O8rVVQ+RAVRm8>f
z4r>Ap`<p<8oA#2yk-g+`k8gVwd+Buae)>|tee+)xG@+$}{EExz`~7mtE-a_c<&D(-
zs(}t|dqYXZjWpoeJE}S)ZNJw|&R&Ci*}_QL-fmG3`<IJk?C+*Z+wb@)YhNrcYky&d
zl>KkLZuW=bB<yFk|D$h_?eyc#HySeMGo_9DKu^4yDfi7=st$QgdfG2Y?5w3-%~iBV
zs*<XwmQyHuAd%Ng=tOHF*{?35-uH{?-Mdn<`oW*m<Z9{_!G4e8E83><g7$0I(&T?N
z<ivU8<GoABqN<RNwB}QNa4xOCnnRs#xio}5l5n2_da$gJzArDLycNZ?)wP7~CDf7Y
zi&h%^?LC#Kyr<LMT4-H*6G^!?Qr6qIbmLJ2y&m(LDzfUSmi3Lsq<UH~kZ)Bbb<E_e
zquRJ?3XCYHk@H!PX_e9CW2N*<qLg&qODKI`G3(7D>btFodW<cigX;WsJm)>trG?~L
zUqBhs1@zB0pN{U$qdOJ3WYHs+&i{Ev!&`Wpf%T0o>))_V2l)1wOk+2v()!d?IyLn$
zNh%y6M?2oYJ#>tuB~MV;#S;|Ex@H>Rcl-W5L48_|kd$r`85$(f@a{>J@n0hC2uY;r
z@`)sMIDw2Qfs#yE-wfgY9W#=AkME)BstM#P7fF`v{pGEh!`eHPo?Z*0=_Y{`YZ5>M
z`vj2t?HSC<oJK33dQ+))5G@}tmozSf)9p+1DJ*wBZT}ulFWTl(@tH8%c`1}RKw-4V
zdM+t^45uH97gA&HA{xROF_-P#;mMDM$$Y2U%r~mzJuu-z58k8dfeq$T*y1CF1@oow
zbd?mmMoXh`wKSY}v(9-cjqX3CQDZKPfyd>MnJ<svtMXuAG}Po&QL51oIa{^yJ&^Mz
zoRv-VGQ%+kGw2Ru#+^RrXb+o0Gt3xny^Jugko^n?L#(eggr1WT41XA*yCi3bId@aC
z!4OJ(i}_h*2-S;*2#L|ho0a;=%{7GDI78HR^@m)h2DZAW!*8cD+#kx}i-#;~7D_^;
zvP;Z8|5v;V{4IWzd=bY^eimPgzKO<+?_ys1FX7`Wg90ZRocEN)GIx2-ZAwGYPX_0v
zN@1%_cT}!qKWnKh9Q!nj3%yH(wPC3kf3{56hE|FMuNom*tA)AKa}g$0DW+~H7wX0p
zqBf~g*xafTd%|i(-`+1prRqnKapRMSYWpIVJ@_U<etj2d4c|rl`!*5R^+^Oy`Yb+=
z_#*B~eiJ*+{t%NV{uVsoA)c)2hO;}RafakDd6ztP9qNT%&PoV>sDghB`eOa*ewe|x
zz;9o4@u@~1`DX{gcCjJkhZ*Biy9o|lWWLTM3oNa&#69kSSt@fE|E&$=nHR16bvWh@
z7>T_-d2gIIBMg!`!{R<1z14Uh^d_N(c|X$!+QYfs0hQ6CkdQVWvcIQdY?C*-R=MMZ
z%V=zDWv|8C1-l<P<Cv#2bOM|)AlVsvZt<3RrZe}>{`Ve#G>q?##naEc$9}{EhRU9h
z?mZdOL)p{e4EXIZN4)C?1P20d`E|6l)CQi_L$Oh2Br^ZA#qSzBs4NBo&f6h#f;HZT
zo8oXIck4NuCC&Z$7S2?k-p0A~-xf#>utB)iFiZ>>iZs4k&9`@Awv96yCC4E6m@DL0
zxnVBnd@AicFxZkaMYg=bMbp^Vm<IW0(;)6lN7%<1$RFSh1#Mpxy9A&tJP-!k=0fuZ
z@0A;E#!1c^wYmD?=(v?o)?W&@pNsJM*$F)IKY>Q~lTiM00!Ne6&|m))+P<ZsPw6J~
zJFx<v3RdIGn)U1%ZGzLb%}^-Yj10X9Xt_jiUpoT({zO1tJrah}k-W#nn&(#po_>tx
z?)i2+;Y`r|*eKi!-NGBBk?4?(z>~gfvCwoSoLBFKmFZsGk~)GTN3UYW(ChfX`sO3+
zoAq8d@ao_VW<D?%*)0R5Q|@8rn0wq;yoaN#Z<P5)HI_d%m=Ym7>;Rgb#zC5M+v5&9
zBX)@sCaxo7@^eA1me8VNkD*Bpu$t$B->h$D-ps@_);GUd-|Vq@2zSSa&}4nHaOE?E
zvA(I0$wjxsTx?pCi%g?jc>H;WoL^b^yCw?{Z!?QIB?EiaGjTFD6Rsh06sM#>0k>4?
zs%LL9J>8okSl=|L_o3BheJCSbiHtpDNOOY>%}tS^X)3DB@X#a&_GzpOG)d)}Chggy
zN%I$J(q%(UI&f5jy%!B~lGLCZ%+--xqfQ=X>dcN&C+qcUlzKyzq<iO~=1&Z^*YWMI
zAqJajV(_LU2G<_NU|`Zr<gT2FlVfKh>XIj{Moxn9BnKom+jFnZ0evi(nR(O^YvwxP
zw2CV#ioFnEz6x2lyeKr*n_QlI(-t=$l8W-BK~H>1`t)ql{WP1*<NZl(K>&3n@O#y-
zp>z;o^l@VtE%`5u?7oMQi^d!>)}KQi+H>fz;vAaNErfQBiYAx5X!?73C!H(jy`3P|
zH;ZGbH6oT8Psh@)gR%6mC6U6XC6VfogESonY0vOOq{a8Ov+765U-u|^_B%?IT`8m?
zyO*LaL{rnPXnMmNKApuoDbq5J=xZD;ycSP$KgZL1$=&2Ex0{x*uGwGC?AYRXD$0zf
z3+LnM;%5GL4)N6VA&x$uh@*LDqv_tNXo_V<e$J#Yvi%uC3ZH^0IU|U&LjuWLGmt#K
z29T$`KNUZiO#%9TWctvT^U2<1yv&<&e*4kP&wli*C4`2q%py(JIJK;ChDtTksfafu
z*Yt`$xxJ&1`Ubi;<^{#vs-wYa|7h8aztr#9Pr5B?sK70cKCXO7EAKz1$mDnQc-~)n
zdZ&YWP&@OKzmd@SOq(S?Q1ZVfDu{YZ)4g6Y!<Y9cb!%vWRTcGmRZf0W%4pGx5;}Lh
znEul$q{G@p1gktcaVC$>9V_P^RwbQlsHM&~bu^a$&Z_7Ns#Gi>!|wSsz%P$pUdyEg
z<MJt<warGC99r`@hm@w|Q9>8@l*SiQkJ&|Za#1m5E-0bYrFFEuv4vLu*Fv`pTj<l!
zW-58bH@wmB=>6FSvK{z_susK;K8CT+Q%{QeHDvjKd*FvF$X2z2q77IFUM^!McNvX+
zR?41ADJ3VB(6_Q;+ND%XQ$mZ#{C6R3s4gVCf<o$bwvaq#6;eV;0e>I!NmVAF+PCF#
zrY)E9y5*8V;xn3bk~Ixyec%5%z}^gV9H;N6sZs|im^W3@7o^ai-l^0&?l4VvWTrss
zahe~``TUP3NY~>e%??kZr@W=`QX`2bs3y_2&xy3;bRzlqBvO-1A|>rkAafPYrE|CL
z*<0?|1;o*ggw>QKx0%i#Tu9q&!pUdV96F~PN<D`M(U<iB^zEQODHTnnd)+3{Lz5}A
zA#E1j;4PV%?}KU2<uKChoI?fE=8|i{9FkucMiW{?XwKLWk|+(KOF?0DDP}Gi#fMWy
z&U~6Oc>%fql!Q>~hOI}tBV<o^sL$vQ<*5I!ZyKeg(BUA3TdZ&Pv%cx?B8^!d((v3M
zjqFrue0amZ4v~SRkt{s#%D`o#JgP0_ai>ro!>6dC-|7L(($j`&y8)KR86bbhK#VsV
zh!3#?5qMw_KFAot#ohp)78zhv?jU$turG7U5S3Sru!OxFt22i9d&L0VP7cC@G9~Ps
zq=e2kMT{HT6CwRsml#UH<hc}He3gdjg!jT~#5=Lc;H3z+uNOz{YlYW^YOz67i@k?y
zgqrM2arV(iabakikQn=2*nMjk{kwIDw~{?D>{kzj8%kr`erb$<-yI4ylCXRsiSXj?
z*xywxzG@YV*V@G*U#&zG_?L=ZrDa0qXO-x>TqSJoRSMI(GU2OJF7`H+i`W^J;`W>>
z@j|gygf4#}(k3;C+4md8tRXF8sdlRfiu)ie7k(5nH6MlXlQuDD^(RrV`?I*?{#B$N
z`z}_6{1Pv&{S%H0JB4&+rwIMsDOz;pQC!v&e-adt(WQ*{)_w3eS`&fu2Vj(q4wg#l
z!H6}^>eGXuIG^)4M#k7sX#$O<W^nG{4*5n)sP(tPtj9xO=W2t^dxzrssbN^MY&hDP
zEB$ruaHQDUV3UFkLjMd!$>b4OWMYR22L)?WLWv5X$@%T<Gouk&I~w1Ijz&_U3re`B
zB*(sujG{A^TRY={n=`@-M`5nNGhS72M{Ch&^nEfK`7_-i+rj(ioF!iWZX!M~gWC4X
z80haK6kWH$(=FDRx5OI9<ZO^q#XBD=BQVv$7RSSRqdjCK+(+Bt^NS%+{bd4`9253!
zOfc()DHNDRkgsNeTh4rs(;5uBQA42q&<16<n7^3dhzw_ENTrO%tQoFovUG!}nutx$
zC&Tfv2b8{hqTwL3aehq2`W4e~dj51Yo|%E>(_Uz7_d(%rf0S4TqEA&I{Nn<#8-Z~9
z>w-IJV>ov*2^+6XLbb$Xw5UymS!Xb`B@Q4W^&nm?nTOubf_b|=1S>S=qBAudJG1Ab
zp#OT_h}eib6E-9D;$~bPx&dRihVph<C^UlSU~3KU!@h~Y=a+o1s*k{{+z7Z9L?G!~
zBr?Mzu|_r$Zb6ZVw2Q!iwd<jyuo5fYFNITK67VGvTdwYb_1$<ZGl;{McQF_q&OLXz
z>u9XFh5er%z=pe5dq&^GJk~cktZ$C*+6#+`Nm#Hh88m((N{^Vs!NVM#cWf}@&M?Fq
zFa!H2p)P~atYnX;e8>En?f{2vE*N*-1q!ThwiIXbf6otb*zO_jvc7q-^cj+v#Z%ZV
z7pk(k$o`xI%cLB<V-C;gBHl?4%s`)s4<Pj^9o^ohLvd~f=7~obkt|0!!xd<3Q!g@q
z)|-Ym_NKS8eQ05yp7cC@4^AiXj>g7#ObKKT<c-}hv{0fSO1&v)dvDqk(VGr$?oA_(
zXz-q}21&2bAV((+YW=Rx8&K+0J5QZ9Emxy==BcQ@c@S%xVi8>*gEZDO+58B96@&Jw
z7z`_l!H$BN*tlmVZiUR`T%jkFp9^&Cv&WBX_Gsr@X#jULTg~~Nd&Ch}<~ZTqawlx4
z8;cXgz8K0}r>qTLw5-aTjw<?6^IBiJd()RbT$)Y#E&LemPXT{tlk801Z@=PCP4)cy
zm;O}v+@C%#Zh?<lGY(BJLC?>(;lLfdbf^88x_lNHp4~}H?(C#u(^!)7jHMx=u~fP8
zf4!1e&YH%O%!*h#I3<>rTJg>5AMf;hjv<Ey3Dl5~K$MX}!wXX=(=L&I6+|=lCYp}u
zL{rzx?KJw@c3OFCJ6T_hrT16jsdsZcIdsL7o78TKtc)i;);8mrkH6ziJcXW$r_@dH
z^indOUR;i&o+>d^UlL7KmpRAeA5G;0qxmH{gvRd*Cfgk5SsMhB{rv#)sPm_<CceyV
z_N6zye0jf%Z+2gOC~2ZM&1Jt!^N2Ui-|a)wjr~c=cnK}Lwvx`*o};a8DZEYal9_>Z
zRNwrZ1_tnk=dwJ$H{GR>^;c>8*RyoH`V8G#ew`Y&KA;!l)5+%aO>((*gFGT`lBUZI
zlJxsZf89RO!UON=cH29eEBl81trv9JiL<(y)wJr(bDDXvjJzh6QcGbmJ)d7hhuGuT
zb2pcky~?E@B9CV370}SOLQ=h3KtWl><bAGyywVG4@7R2LJ0p*Fn&;A~v>ZASm_u)5
za=1_bjMTg5kp9bDx^X+7PBByO6?-Y^>x)Td{d1~_c|&a)&9r`BGZ{{3CjYffG+w=t
zK971!l5^kCr~|L)HuL;c&eTxWuPWXacurf7RZ?qO1$md3(?iX2GKeXoJ6)w@&Ry`j
z*Cmu@UP7mQifQ<kBI<g^_b{JA+B%1CQrin@_7?7iyA@JaegXZ7yiS)(uhCbDL-bzd
zAieTVrj(BdC_DK8{n>JW!q`(=(3|rxefQIyE6MzQIz%zOQYbJ!g~okOr4JQH>Bi1u
zWIp`_#Wyi0^JgMGW8Y?_Q4)RYnM4LsNhH;nNR<hRWIipCc1tHxj!gosj9gCQ>oQWE
zzMOPtt{@lLm9&TNN4xqgq~A~G(R9B#bia2feLX*!{&h~K7-diL{yT=*#O~yD*Mk~q
z2C3ijq2;Clr1Lw7%+|4vQ3xf6k`Pk+5lpWog2^r|h@O}PQ?FGa^us2MS}W$z-}iIb
zW12??E=Zzyh9t&RNnqs?2|Vg1fm2Fd!jd|Litazw1ycAlT?%UHQqZxIM(G4;yj~-X
zqBGJs${x-xRT+%1k-_NgGO*2%f&LpgII!QabE^Wj59x`K?Y*!rQwi?-lyNjo9jhV+
zU~Qu|g8tKi<vv~J`3*$S&4DNyHwgD04TRud&lKxp!*xAKYUrW3RF6AIJyEe<5?ym7
zP}#pr+!mc8y+jh_`4X5CDS>s%TSRq2lejqaotSF(Qsg<*ix~4-5wN*hBxqNQ>$_{j
z_scKD-vb{-!N4{V__a;+o7gU*9Dj@8Z=FKJSrXGvcEd^k?)X*T4JQ&M@pGIc^FLaJ
zL7y6NGO<YL<`fCf(qb{MPpP;vzD!t;Di^Eu%0>M0GI4HfnNUnC6It8J#hgDC;!)Rg
zaWcL}?9ZwfQFmSozfW&PYH*YIGQCA;RlXM=-?R$fv=2h-&_}T;rcEU0d=hzUzliYl
z-^9_(A42lWKhZg_Q=Iy*Q>^~pDbAdcL%#wAOwH<r9U97T$W}#tsTvNx)j+>#d<Qhr
z!MmP%nD$B^S6K@tcpE~6IXfW-Oz=a?426fxAqH9^>Fi((RUU%rF4o9~4K{w@9{G@=
z7|UB4#tlOdc6kUke7DBO%e)U7IUF6Mxp&3-=T+25th!{6wA9fExH}p-I=uCC$pxid
z&iJ{)8RmhbkjH(j!|Kjhk}wJpAx=nA9)_JAmWX+4g}$6if6^+4_b+AP^TGkMR*%Qq
z#xY2J0~nh!gSmAGKI9C+d3$TTIADSFqc-T#(;Az2??Q6(2&~96#*sJ^zGa#q&C&#I
z#wJLyFyhRRDLM<yKx558&X)MZK9yUt6_zv8dEt(sSRKW;xShP$)y)ah501mpB6s{*
zHyL^}m|vaZi2<%tpc%=&6mxzCDNX}bPlqbMwoTUZMOBeMbEgAAHv@5HT_7G?b2omG
z3l7)1pk~D+_>Z21p~;h=+&%@nb_Zkr?J&f*%!7(`Fvgt=<_=;A7M9F~{n~Kk+~&=$
z&+BmV-Ue)Zx)#R&)?t8WD4Z9CB6D&G{QHICY5{M1Z3~C-qX;BFjKI_<5zx37!GCT9
zMmBK{`BVgs>|PJ=P3y76eGL}(UybssOOQWdF<!jfg3W)nu%3xRka85v`bD9SPYfKh
z&cW~64eaHvRQ8v2*m4I;@zDR6$>(wXz)2jlImFs(BCJoCW8-*p%#bz5AnxHkzRkOG
ztUFJ149A)Q0>@?Txi@EzU)$_4Rml;>oJW6}?1B}RW1(O-7S_whqFnAFbi*H_KHwp=
zu0F%09=Wjln1l9jPaw;^a@+YC&?&l$!76tU)y({6zIC~+9);dNoN$i$h%Oh-;qR~W
z7`U_@zc#l+ZGSs9vt}E5Egm}(<MCxfJk)&RQDGW~{LWa+tBJ+T@>pg?$3jvy4%5mG
zAhz%T(mox4l~giLgr(w@Vk&0;O~Lc2shE1=AUqpm(UBX2+mB<gg*8nde%!bj1D{JV
zh}jo|6Rd9*%$te%OFS`_pEtg-0xJ{k(T$(`m3{2dX5fIZHxBHFI${Rvn}$D5klE&p
zb#MI8kg@{X=6KQi5^vft+=sp_<Q<GlzVw^*jdSB{YU=J!J*=fjX=o3+KmRu-7XL!a
zvQ|W|dXG~<Ey$3p#Ptl$P1#(>ewl07wEQG`YVL(tb{Li|F`Rkj+|2;?ZVY2-6=!f-
z*YS;tA7_@tQeMAU+M*Oon}5bob8`%ponxJ|I)Mf(<?MrB0!66oq}GgRax{;o0qW5d
z^lCdb=^D}*IYWvmGN2p{L(=y$B-O%LI#Cf%SJ}HM&WoqF596tmy`8?i8#Y8ep00e3
zqiCxbIw})GufFdh$=m#A9f+nqW1?xZ9Q#56!Bp~J5NXK<Qjco^BwOK6Ev#=|v%aZg
zeUrfYW<Tqj{iD69l^+tTyvd-C56R8+q4}G9=z@VSb%iV>S>AbdRX<2?{Pt7%$o(X{
zKaYAzJR_U?cgXkeIr>_2hElno|7Ip{s4u-r_3G)oi*=Kx$y_Iu<STTg=n_eWT_V9u
z-MDKFG;{bXa?Rn})cqF}SyjtkL^Zwje@=OkrBpwofGQ00$#;DrP4{PQvib>GM&?p@
zdM*v}$)lT}^5}AF9t~l=bE$s;DGkiw+{!b0_vjg=Hb0~DOP^6y*)wVhc}B{O+0^hk
zo8~XhrOi=%3+i7;XHi7?sGy0;uSn<L8{W}uq&-PZBoW(0C7gF&9{i5992-b};A@&P
z;w4F6<^Fd>4b5z=qIVaclfuDD`gnprp1rM!isj@rr;L<xO6m1r?t|}ShF?xGEo&;G
zKc5Tf7yD0_P8HDk?*+7KTp_LfR6v(sr_!&!sT7=-LN?t}$glDs)xS=rF#Tk@zViS@
zPCGz}y${ePW&w<RyN^EH-$&C7_fvI`Lp0~fA(FY2OhLiPw0=S|^@~a%#f^z{{C*<E
zeoZ9#zDcCiFNym6<y+K0-i&&kNDi*ssnsTc_n~G}N0mR_njJ=K^22D#qh(|syNqI^
zmQk%eZ~iui(oQ`;?g$K{)f-2UAn=xp1NBKCMddF>Q|di8Qfiq*E_GAs{H<B!^T3xH
zBK#?awX<jrq$xiF$t5C?{`?Fe^XCDy#Ve2^mIu-4SHbjoWGD?c38SJzVU)B;5=K^%
z%#W2ozi<ge{Nik@T$fO@?G(+*|HSRsKO%BVhwxedTV%}mBVNY#zz-v796!t6&0}fQ
z%gNxMsSHLs@?GaNb9Cm&A;wi6S)K~apYI9l@4ZlWK?yGBl#zWv9r4K;m{OySk(YFE
zZHF$NZq`Fsh92TF^)SP2ASOl+fJcfJDzy4zw~{70C&}ZkUw4e!D2d}H5^#FlDPHyH
z6!m?(M0<<`UXGPO*t!-`a-~Vg^m`|cu6!wexYP?xxmuwhUo9>uSBWEw>&1sjFNDpi
zkK(IJn|O4!O{|vxBG%OX6#2*hiO<Orcsi&XX0Wz7`dbqDS`yH|D1rBH-wV5eZ^fQ-
zg`(2ENDPQ7W-U@8*431Xd#lUX87dP~FPDll2TR2bl`?T%u3Xe@sSpR(KNq22szsmo
zbz;MWS0ZstgD|`DPFSTii9SPH#L%$!!Zf&5NG<;$w(t2UX0*16_yeDXq3>7m!~TcR
z`uj&*4DA#<bh^Zf=I%&YC<Fhy@<?jv31b~4<jhcklDZlms%W5i7=NE8vXA4e0|hNT
z)V$Qk-~{fH+Zdw9BO??zn&9OvQ%o>3hYatew!OAQMCV|vdu4@@Yk1q?{1B{T#^M?q
zD|E_Q;dZqZLS?Kmrne1B^=zQDYzX4&OpxQEgZ+IqcsrmsK3!A9mr@0!S;=9bn+yyu
z4#dy(PB4mfM2&<!-2WSiaW+HHgBgsU)@#CHm;wgfkb~NNX)M|<jla>-aNohtk?H+#
zNXGzW|LNiRYdvhq(nrmOLAbQd5c(cQ*tgdRA2>%I6=;CfWP<zFCW!4Y#xCX@j*B%!
ziH9ltImcz9WQGk{rtsWtjwOzk$lS#|qB>?g_Zxv<4~D|b!V-U%567v1(WtK)2gifF
z;W6D4-4v$aYxk+}Pn?1T-lp(Nn1+4V#$mUf9gZZ~BA)DUC0O7o9I@Ye6f%E0LwSn}
z0`$k=<im+D`ZN)n$9dw~rJ2~C>Vx9p0f=201ofz3*!YKFtYRpxZwkZeev6Rjv;t$E
zu0~NyAm-HubLJ-qLsWt>b!Z6A_6)`3mN^)-Bph1p^KtLv0>o7<f?xbf);60Vy&?j$
ztRqmjb3HQHyRqzA4d>!jSmm(<>1vB%yf=a~H(OxZJqmX6QRu4@g{uA0h)+L<TNiJ@
z;QW7hT$m1j+jL}~U_P?NDfH@p1k?DnUM<)LPN&SF<!p{`9cCDjX9g4Q0*p<uLElHb
zpR&LP+Uds7ooj>xqnXjrXn^z#U0AHqL8H75<{#6+3#l>qkUAE6imY!YJ;cR+4{@Bm
zou_<%I^~&(HSg}?^o_evxp^0zGIu!Z&bwO!U2uH;C@lEk1hR5MM2aK!NI1gfJ#(zA
zcH?PVJZ{y+!|HN8ToU3jgZ-Pk3*vCdI1c3<vG6O3g?Cvj>Yv3z;zBG2XvT5>fcMPW
z4q#08Wc(hUjOf-B6n#m?RGWj4VO?})4)1XA*ZeUv1|F3$Xf2OH+>IFAycmO}yJKKz
zybD>rGa(u7iLEOI#%&f@@lRk_0dR%+QU9^l+4R~02mBmSM!a3y=mht}{ut1AHopE?
zh;L;ZkUPtZhS~Vg=xsife~sUFc00yD!{Z2kcLDmDoM&G?13vsQ-D~|>*ZXsq!ylO!
zrlP`W0v5MSz?a(-@kb>H!S3@g?AUtb`)tF-A=~i&N(}YWilrX|V<}}(EZt=fXXL6_
zavd2<b05W!S4Au=-p9c}eGfFh9mJw5JL%eqon-H@lQ-9Q(p-urcivW(4BJ5kLky`y
z*^mb37*O0Z16nU-NG7ulDOkdYI**x>;&e0W!G29{csw1s$Ue@Icxsi7Ct21vTeM?H
zPnz`;>l?@VT_k&&|IC%q)OQejL!*Kz{%{c0HuIe!HGmSo`O$^`e&o*jMz7qPzRmF_
z{ZZbeL*8T~ys0tDn`UqICbJ#hRI|&Q9?SdES|eX_eHcVJCqk+1(|jsz-bXJV?xSbM
z`)R=2{bVA0miH`9lWITi@0*;V?YK(2k6)qXHaE#S_ZnHQzCzW!m*M{P0{Kqht;ZK<
zDY5+;nLfEgwS1Fucw9j;*;RDyN+q3|TtX#A`P5eOltwf@CXGW`lobDnF8z5(F_sT0
zpeUP^YxBsbGLKT~a;Z);mp*OHCjIKCw3{Cn?qt&z(`?TBJ|%;~Y^v&+O>;IrrP&vs
z(j>1OYOyS#U2AHnL%yDNw7n$7>u>1dzIV*hYoyv%zDor)QepX9(zJU+Ek!SBV@Ex8
zN!QU>)X>y|D%y4EIjO}~(zZht6p~d=QQQYV=v+qayGyB=Ies#{pZQ}!F&$c2#5?na
zB<){7txxkwV_N}zTUbC*>kre4<EeC2Ih8h_OyU06A^Mknke(JL(-M_rst-9p?&A*7
zE*ai`a^jvKcbuL}$5G6eoixmR2hFYDL6cVOq$Z1avh2B=3g+yg<?9kDg0t;&KC;eH
z;{4b5MB4CtC;i$RN=x%+(<9a)Gx%q&f7g%3EtyTpZ}_gH97YbOd2jdh5{mh~gl5Oj
zr!UJwsk&kWy>7Io{x^ow_DO;|7ur)xnG@~#?Lv}u<CwiUfr2)9lGCf{B%9zxkDPod
zvcr$o@V)AhwLfL5_|wFi+4Qw}HZ?5pr&j?1<g+J`QmzEirA@)K>s>Itagf9XB}r&M
zkU(>Q1Qd67iQdmUg|2$1a5DcVK1BZ!(yVXlR{a*WCw_`QZ+-|p%O9dh@rU?w`iDr1
z|0#M*{Ui2ml0ZO3H;ky0h5bJ{3@}xIf^|>kw)cYM5ha{B-WRfo>bSXG1DBWsEPqr7
zMVoZ7(p3*CetIbRrVA?vZR~RyfZR|mJZ<WS)TJ6YIa?m<XLm;f=V{i<N+4u?r}%e}
z{f(_%LU)P;%==5g@O+CHR@EdLEgD4U-Irpmd%d_-TO*Q!s>J&I3h^(bUVM>zA(Zv<
zh05Pt@xr%BnEz`M-Pg1W`^4X3?c+Z}wfk!^My5eLo7yZCsZD%SZ4)k&--{T}LZP|1
zK&ag+6kRsO;?<54F=kV#2(K#@7NMo$esPI-T3RAHLP|x(vNAFBd%5`ex>7v0trjbO
z)r!f#UWn)H;oMevE6SxC#hVjN!mGVmSZKc&#%`@5W9J7k<?lz~e(95_Tlqy?nea^v
z9@#Fwr*w#ti4ssc(jEP?r19TfIo$cDz_|@Y<Xb4S22+KTVqch=XkyJe?!a#xfScYr
zn4ql(&3t{_4;{p>5eAqP%51<=V<_?t>Vh3+$hc>Y0_G33?_uV#=3tC*9*jl2_24nZ
z65FjT(RPuULP3L>&B7cXx521=&E8LnAvR6ZM!%=(=-s_H{4ObCoq+;EgLuz_J6^Rz
zdSIV*H`IUQ&v%&wEbF@COj>^!w`g+yLjwVEs`wEm2QLe01hq?HnX5E>vt@C-ToEy>
z&s;9+B3V@rzhCL$e6BvujWfi@Mnjlu7~ylYA(kl_p#9`P{CRB*anBfy*Nri&-T<**
zO;Emwd-Zor5K?9W@5`J$3N^>4hQZk4XoG#Xc=vvQCCvL+V%}s+q;4JwrHx~8=e8S0
z8caZj<0K5L<xN#BPmId*M9mN8A^)(0ySpv)*V!V7cS=u&2uR2}B2#7*QcgN!u&oPk
znT|#?CgK*q9?cp!1*d#wBGuXlDsBEa>KX(wGZ=SFLh!aZ1fIn{=(A-C678lji)IEM
zU-N~8eE`IZK$O=9A-!)1th$9_z^O0{e>4v#Yvx1s`T~62ybw1IEM|V-Qpo2m#Ra{k
z=>KvFZ&WP7t(xV$alZ)ReHY_v#Uga9jDQ^HbT)Ku;htU;dh=ua)$Ldjd;^QV-$i7}
z9X!oUN5kIRsHwS(KE03Q`1m7`Puqps4a^fbYmU<4=J0-D28CN@=!!DKgh=+#%uS)o
zoCayunR@e#5HiXLf$t1(C_@+WYuSI5)<LR*4w8H6LV^4E8+_bRdV31ic1?i;a{$tM
zPD7>5G<?pQ2FXzV9IWo3S@I5C%F}V&pSRU-jl$UHPDs;t!ly(>JomK6v;u+PNP*?m
z0=qqc5zGbXu_qp7tZ#DW#34c29-A)_26O`d4FaBg6!`8Mi<b|0Kg@x*HKxbm&XqXq
z9GZ-k2a-|LoQ$MO-T*OUmeA)Il;^}?{XNz*w`1U38iVE6V^DV{22N9UVN>1=<TBGT
zU?Om_)DHUd1&(!Ngy9at-|qyk5PO_^#;gJlN65N4At=uY<J$efK>)lyvk?C}7vikg
z0Heyapli!;;1RPV6TEppZU!!UOo!D;evX+8!<G+2xjV^h^xq?qF%OtF#SvW+V<81s
z^r>~ny91N)`+6`YHqVEn$3{ra-HI*4x1nb4Hhk5KrD^)H)NB$<g?7w$y|4>P>3mx&
zj77n!ZIE%^hI5sBao_S3u7{mPWh*nX2i?KCwjHFKyMttg8`4fAL)!J&fUH*xqT%I(
z$T3xyf`{nQmkW9n*ri8q?@Vad2}?>?X-!5;<Ei{mJXJ-;lR{e@*$<4N%~CO>^=%g=
z7w)22=c4J%@@T5ikEV&2c2JM?L3Af0fchuTCd)=YX1e*&%H03$o_UkjbZ^@0;7wIS
zz3GY>YZ}%!R;+J!@ARfItZ$Ozy=jmx=a$Hqob!T6?^7ss4iBe=Ui)a$vwie#^nM!n
zeLr2TIZ0thPLkg3G+Hz36s?ZBLKiD9ljW7Gq}6qqHqO69mw3Z;hxvJmVD_{5lryww
z_bD2@J&lf@DIkq1&gOoopp9CbTWQXvd9R+(FqOxY6!DM-dOV<Yz8Tb8E`x?n%Oumu
z8RV{$OVKI0^uC?hcg;^|*1;!aYMe!;Js*>Y`$Nhd{DAa~@6%fGlnRb#QJKqAQsB(X
zF1{VTm{UVvcv~U*#7pw>;~QDyTUxK)NG-vQyifL)CN;mIz04~P-|&JSXVp>j)LQa=
zTTQtqswh43Ih|iyN$s%}wCOf?*x#2?@_;hRnO;gij+W4)55;t#Uoj0)EF!IR{`+?3
z)94-9)O++<S~2tpcgPM?LwPD)V|_DicM8pAKjwAXL9$g(rdR0)XszyMdYt1)J9|y1
zy*=1RYw@S7!U)>d8cE(~w$p+=JIM9FXzKnajy|x5=dKq|i<zOf-fa~rBrm7gtLD>p
zi7<NehO<|Re)O9&XPKQoWY3=6{GERMoS032b%V)Zawy#!KabujEvBz#CUhj*ly>M@
z(0?&jbX~`q*2oT}H6Mpll&u{Ne?-i$8O2+`V`=nPH)=PTOf_axssDi))N_azIktP#
zSt)*gZ1ttXr+n!ad!!p0d}&X(A02X@O%wSJtU5A)R=j8b&?S&;`$^*44+-pLzv8;8
z1T<H5i8DEW#p=ndXX<{5xqW_#!<<zKoYXEtw0?>#{~sdz<9D$j@4MJ;`a}4iZx_Bp
zJH)gLT|zIV8+Q_<aq}8`I1=*cmLw0&DGGp4M!<vK*yN**ksbZ8*;pIC0XoPxVkWSu
z9_Fj*LBU)XLVExrGzLIvq!v7V)o{e2FPd!?q5ZNaJV+7?Z+3|c)lOmB^+&kw=n&4<
z5_s{uODM{{7sVf&L;+`AD%)O)#nbA=)2lVY@Az{Oq+B5!my`%Sy+X0<Se~$M%n_EW
zo{33z*&=HAQ_*VoMD+djSh)1c5()ZQB0DHcEU(TI{bQbp|0<q}KRa?nxK+MrP%RXT
zKNJZCuM(l)R4T5Amx=}ZO2kFJSy_B47X43@i1zGK@nBN9Ft(`_dJC(B&Tq~jU#S<>
zPOpUewl`w*nzuq<zmYw#CULcrb(9YKLu)ycBk@r<J!lg)XFrSCFTRS=-P(onf({Ya
zQvw!--7#;FH0)l>;&OilJTUBqTJGX^2o?P5=*|9#I%k44QTwnzMjsu3^Zq(GC8vjm
zRDImm<~$DPp0e8vanZ?`J5eUEJ!*>giDsyq%(<Lg^Z&<~-k4#I{xi){H^LlFtITmO
z-yCbQ%^{i3z5ag88O$08J*fdWqOFbt3sm7bRS~nB<q<GQ77yyBK(D$ZTe%y?R!CqY
z`)MoxlYq-@1zh>20okjnP%KixCVd4=a_NpElcZ2w#F;5=Ik+zEgYi}a5FMezo7lRr
z=d85w<hx`Ndjpa>m^D)yr_|ZcTftf>r5`S=(8tNu+!1pg1W8u|ObIZ6rJ6C!R~ci~
zjX_8{V1)j&&EUz5!?zCyW65C)e0ysF&o&E$cUoX)lLc$JF?bj?4n75L@KfZydHc!C
z!Q&nIrK7R#fi1%N+v1qAEjEp|MRhkj$n*R6OG);4<KTcX`3_h$m|u%boMHW849xlU
z+-1He;x%XD<qvP(Z<vka1A|~YG8iu1S>Jr{Ms|k>)S2^>YB3cB=V##hV_zh|4;?9f
zs5$S4KHq%tEZPsn;j<CsGaE9I^YAKUKF0N2h?}z)q3y&H+!afq7rX@1(--5<v&H!N
zbP;6J7U7coBJ?=A2uGhRMAO6hxOQy|dY|5cg{*aalD0$N{sw2EdHW&gE;O&F!(r5Q
zxOg8&z?x&2d1n_)Sm*Ziw!jR&PlX$p0}st$`_dFj!Di^2V+x5sM$ms^gg9n~XS2R}
zvd9p5ll351r-Q|_wQ=pAHdcRT-c^G(#>~>j!f0(ETN@i+4n(G%9ZZ+oK}o^`zE`GU
z$A;-h(c=tIM>>Z5myUq}>9G2K8}&s_SU1242RAuF%^rw%6|mr&*t~@TjG91%BY20M
z_g796^dA!vBLq~(2n_BiFyoUzM4`Zh4uMTmVo_Nii!DDnU$Z+7$cuy5i#VvfibGgg
z9A>?X#e`2W*oW~jy*VDSe3u$~IR@iXV<6w<j?<eb;v4gRW_t1euEbzm39!Pl^H%Uu
zBWztt0P7psY4(_N(H?8xadyq$33FKAxSIu_p=|+%&RvM#6$>GGY7sme7Gur$C2&*n
zf}Qd#?m5oDpD&(p-Q|IRjG-`iF%(kOyg$6DK}5RhpvP1f1oapT>4~oB?LChDwTY+?
z^gzSM`IyIhwPVI_#V6x!n2c?x-Mo$S>Z~tHV(7lnF0?1|KJ4P1$lkUcx7(vokrD;v
zjxET`*n(Swqu_h}4>lG2!_u4qw5vvk{x#^))2RbVZ{i>_Ow=b`+kUjyy&uiY)u1(Z
zHE91S4I0Osp^gAkq6=2ML1jaW?S_&A>ziZq<7vUiSh{eXvp}u8NS`w}dYoC_WEV{q
zx}3wgvV%^Y+(8<n0+_kzPh+0@k^Dy1H%mEF5jBPKO}x09Jc}IFyeXldH$7r~Q_hc6
z);CjF-|T07Gn_kUuKj&!RYf4R$Oe-}Y#6QF8%|rt?4w_A_VHdNGY1!+q<?NF$v85N
zto2S&%s<|!E=r@dA9=IA<RZP8c!~O+yg(PF&Xe)tvs6)hnj)Fma>OT%T2<1hUve3D
zuS@BiRRNvy%%+H)kIA*iBQiXiNxM7l(SX7CNGb9GMGnlMjTat}&Z+w}r0qWCjd@G~
z=d+kolf^mU$F#}l5m`E9GBfEu-RtomecOMB68k<T!~gD4gU4gK-K&IVQw_~KQ^&gu
zFX?QrH#9HlEh#^{L9t_R(j%{Pw9@Q6rG0xs(`u{e@7`LnK2bydk=3+pc@;B=p3|4W
zN>W;1LCt5%>04zP?Q1V3Da}&aG^>P0ohqhrrA1V7tB|TW&n#7(O&dR6pvh~`(9hvV
zDBSxn=cH1pL5I8Bp(!*Z{t(&mhSTig3ADz}opwl0q^ysVIjb_2E_!>>s6YM`uNOgw
zE=AJP``fs0wS$^C<KEyLO=j8ABtj!d=j$3uR$56%*DfV<_qo*79zc2<XOs4IU)nd{
zmxBNE;eC1^N(lF*5k-DP+k;5=QV7jlHkTHq8j{NtW0F)hC3|0U^0FLEdgeoDu7?d>
z(Hu^9$JtVT0chD5N3zKtP2Nr8DE`F+3Qh7LWKSi9aWm*vpIH>-;6;C0yl4uKC=FZd
zO(zOiH_!4R^8vnO{nwXt^Zm#_bT*aWm`y|fN}%JV1hnlW5IUe!Bq{zA1@0XpeB5uY
z2mKOT!hVW_ra#3DSI(<+{1B6jeu${b@1pYjck#XHyJ)d&7fsiGiv-m!5x1lpY7L}e
zYbl2%U**t$w>+wv74R*o7hKOP;f#h7rkg2Y<2@xTQ16XM_uj}c(&G2TtZS6DaVb-a
zyEx37G0?y{Q#GXh?Soszz2L@o8CxSs#D{c=th9eZVJ+*MO~1tUsPDpN+h<W^_CXZi
zY8IEJ--@H-Ux`Z_>V<!=TCuLJQoNpAE=r|KgxZG!F>-01h&_}eLUf;rWiOwK&TUVG
z*}5zd{_?T7m;G4mE`BU-^vDvs=4S~j^(Ugi`Kd^o`Al^Cl`FO_DiDY77KueJC1MEo
zsP<Zwidm0}h3u&!vGP-q=*le?A#$bSWpbG~xvWC8Bs>>)KUa%4i|T}8^$XEM{<Zk_
z`;D-<@K%_bHwxAGCjPi)5g71ZtpD9Al%IYS9z~zTQpc}CI`)T%{`Xsyt?d#suXn?t
z-BR2+lf@e+-UGAfiE9HCadITz`<|*|$<e;hkzlr9FaEy0?~fIy2Vjt!4jxqLLef(o
zyVD1vm!bi>JPcv5$p}g@##p-01RIzKs1;+1yz8cTm}rV$6HIa6jDHoTFdAcu4(15n
zJ<R>KS_AwyR1f!}`lEBC8ou9G!R@ME$lECo?*bX5MRNX}`*%Az4@LVVxRcT)RNA}5
zZ&O8RWU9hALKSkCRp5GsH$KeyYw47P$F**_rq>Hv8Qj}%XJ%rIHY`Tz;HI)J4h`4E
zwr|>Ky`&9)S#5l|qlNh`%x7eFqLo}f*q+hH<ko?>S~Uo|#s*lo-4NxIjL@bw5MdI7
z@Ymavw+77dvYQ1IHt`PHT?>Rfv4HIj3p@$7fREA`VDva_j$oEyA@7xSn+&H2Tlk!|
z#j6Fjcz$gp#^j7d#ov*<t7MBB?#>^HvPXly1J0drfDd~)vV2#W250n68^ij147&PH
zfY!nZnA6Cf&^Rwx9reOx9dBe#@_^qH56E_T;_t&LSUPS7ba)3#4DiG2ReroB<_F~>
zUyKR#Ma&&vIQ3wC^K3R+e$U0mS>Z_YTF9T<Li`sOjsx8nVvYY|bb2htjLb#&p0Eft
z4GXc9Utcq_eKGyIH|AWQf%pJV=$x8}3yNFdX2$RN#SPS2+(4u94VbH(hm4FHHr*SC
z)pe|Q#~2`k@0PE8%+M5J3Ukh+(?k=z<>$w9Ph<S6H$ui`BmUVKB7GJAjHYTMksrCN
zWpa3{<fflCmX~XDUqBnDr}u~Z5@tgvtHJG8Uwn@0i^wxt81c#u_ntW6iPdyO>&{>u
zn2x*WS>yPnBQ1OsS~H!ndbcBb{3XmW1wONOIoVsF&sRGfFSNtDYpgMk*&+L&9iAVx
zL&_k5ASHo9Nr9%{{QHk~xRcEujwEvq!eUYNB^HzW#^KV<IPA-b!-#y|NgWY~_VKZB
zZga&}yYX;8Gah$T-SB3q8(M4KVD_DVe!B?U+XedjS|TpO67vTQMoW?v<}I><(PCyr
z-6IIz%(Jx{hF`o(XUF&0s<CePrxJ>unG4Wo*g|}@VO?sp0J8IZpsnhSNitqYSw9nL
zlcwOE6>pu4@Icgi8?+v`#vSJdkv+dbe6)EhwiLe;SE3q4N{A~|1IA&3=|n7Y^1%Dg
z^U-;1BZ6oPM#yf3&fsn6n7s||J+~uK?7+C4JFucD8WIn7AoEugX6I}~bM_|8PuT*^
z^eu3|z6B1;w<2^3GnD@BfK;;@nV(aqZGjrJ_Wvk4@30&gHi|b$MM-<_CEC0D+!>*i
zrctu9_ugA%yt1+*rD0@m8I_PtsLT{X$Sh^$yT3oWjMr6op4W4pbD#4&&PF`U%$RGu
zj9JKu{P3nBKM{R1EYpw|ZZhQUOC+zIZpq^gd+>+v9-O}0kM&CedHTl=+|I0{boN(r
zi}aPOJ!vJkS6az&PgZd9QSpefSMZ#u!91q(U_NnWAlq+@;heFu@9q`D`U|6Z>a{4g
zkzS0hRUD^Q#PR35aXfl#U(S^7*;dgv*?ao)4Wj{k{%!*QP#MOSYex$6aV+=lu$is$
zH}l&yTlj-QI@?{z<i(EJ+~#{W*HmTm8}V{#dS>&Fy;;0)<N>w~&1F^54~Ja$a>Af&
zPF^4z@kUvkdn}9HJ7jTy(tZAJc8i}@U*YNVE^+h50uJ4g&;DuWcva9j$q1k0-dX24
z_TYJ*vg|w;ti8a~BJx?|Xc4d6TEw@`T;eqeg?wpnKEFG3p6!G4c**tC+)wuy2Uncq
z#HA-#Px}yC1{U(1ltLbMrjS)f6mql19De?FFPF&OtgHEcUh``|>+Czg4qVRTC3o}d
z=VRVt@|fRueZ-+%%lSyxGVU^>l#RDP<f^L=*yHYfe*dt9`?oG(tr7P)VC!A!7~bK8
zN!K{y*AX@jKfv2}@8PLY=}cQwS=lCrzkH11r^%6Q8y3OMMu&6X_C0t=O&F)gh4Y=B
zQCxba7snU(W4*9RJfXu}UZJ#@^Ts8z(c8tW=Dma`S}ft<$_1RVa6S)<o5QO$M{@az
z;oLWV7?0mFnC~kN=Kt0W;NvX@@Vw~$oSWO9RRRX^-bDj>Y+3>j)J|ZtJ2t#^j~&|#
zcHlo~%iHF(<K~JkeAmT|D`$Dgy~&$7%#RcENcf%|cy(<MPp%4P|6O6+PCt^<4P&^?
z(q8;Kx)1MJAIl#G#_^OFalE=?U(PttS8`7A+~;dNH$Nr${xSWz=1YG*>OFwly;i`a
zTMB3^*_P4tj~qV#q2j*6(8W)(-`_~vhcwW_whiQF@q-d;>nTL~%}Y*;u9;m&*QeEy
zL&th5x!p*!jyBO&LuE{{QpI0)H3YwtkKgLJQ`ZvRVl-r<K@%rVXd?ZZCYD&VM%Cli
z=-5IF5tFrWDN6@KcIadHV|{4b8RMglAx?hQL(nQ+oS35l_0x)|*drcOz&}a}|4U{+
zevzd@BVCEDqqxOiC@kqC?QBs)yOvaw^|(q>ob#M|);=NoVR9!r@qq5Fzej3YZqq00
zo21(DI{my@OiHt_QdQ?G^r^>XN<Usirt^!)ZAKB5r54e$??u#h)n)3Fc!j#E7gNs<
z*Qt58oAhYLHHz6^Ok=hd)8*~Ol&El<Uf>SxEWJZ<m3OHdm5@={eR`7fkQU^Y(XOgT
zq~i6I!sb4wIon^*?4&Ac?C^@T@4cq7;5U?f{w=w5s-d~>-c#PCkF@{gC$gONl@wpp
zky78ElvLbAX<e0I*|Is3=eEFs$!gf>*%IGcY2cGtYdl-r1~avEp&GA`4nquaBET3b
z-%apfov{B5EbvtFO%ZRbkm_cGopP_*n`kF_U3=8UIG|ab0|NUwLNCz~*C#q67><%7
za72IU<y#7O$imwJ7tY!t<&U-ShRkuK)(A7&>Y@0X7EC*8Afb&qMhtF&G{fdluu;Z{
zIf@7-1<b5&qAUx^L@)eD6E%eoEq>3MM=kMVWfN6h{7vHn{?PVKipam*3L`R&5q(}X
z?F}<5>@HqSv;}f|i#981iI`dDc=gB(gXfr{rM)r!O*BGEsu5~~jPY!UaE)B#&v;T;
zjzM-XD6&SR+~p31SfiJZEe8B-hru7)qjsn>0)%zgknW83vz%dH-W~-()~M9zj*+op
zsK^UL%p-qn8t#v;uKuWa?~9;tKj_Z$!{&W{2ow*e(o}ZHCI+IirQ|M>I^dvsAUtI6
z!z{Kl?hCWWY;{+pwh70^k>Qy4G8}E6L|}fU>}DlL;jTk8x^0R^=<8@)H;KW4xzTvg
zMRx6%Mj~^_Kx76E#Nf>XWOqZDgQ9Qd)d+Xz%pmA_kH)mhF>-qwkJmY4<@IhX@{f(d
zoMxl3y2nU7a~TdQ917dqA<%Ldgu|HwFnv`&VIuXysH7;|`5capt^^g$CAjC<N4i3C
zgF6_B)2+Ir;baIFXi2|rhCNoywZlQl^Xsj&fx>hr%sV4{udZ$3E_v=6FGox`<$z<o
z>~KRKiz$7~Q6FQDsXfiHOuqjn)*P|ouXT~{Upy>*ffy4MPBg^**7_JwuLr+GJ*Xw?
z;o3}7T<bxw>)8os#G`2%&;i=gL+WHn*w=~By9Z-U8RKUKLr$4EQBN>$PWX2%0C|N0
z=q&o?&8z^7S{#7QqHm_{50J-zq*oSzg`#ajt3*HD3P9;il6)8-xfIEkC!=ZD8sUGg
z!TvdGa4kuGUcEDpF6;`Wi@~^D9}K0|VKDLx!?U5Gcq;x(@(02i;SA*Ma7MPNGs+EQ
zH+Qoux;%D8%VBPK^vn(8{<tG<yC+_E^@34KCv=JoLx0Ob7&d$~44TFuY2ZL?H4ehH
zp}>FngtdpG5x+^ePx&zzsUCx)r^7M2)Dv+NJ<zV*E1Gtxk}N$cspE^+)P2Ys8c-38
zR*Bt#c9IX1=Z>@ZH~%&-#37qRT&-J*t=h{F%geB6;WAABy%b*6%Mg$=93xT(!m%s?
z7Exp1ma+g}k{2W5#1i-vCc-Lz94sHtM7`%6Xco`Io=59szpoh=J=N!?#YVj9s&Irn
zjd?^XWA^=J#7@#f`@YbS_s8mTLZ}KyhO2N?PZfT(&WEcqz}<y^IILRUTmS0LYr2JU
z#I8`@;1I^;`u%x%^Z+*07{qzbgE{KKKt}2S9zH6DReHy;f<_E)85PYJmqoF|{Rp=7
zh+wB1(Y$hTPfqF4ms6ADIs0QjuG-z7hbkxVti?llxy49sG9Am4p35G_<jveIZZmi4
zyM>p`*u~GcWpa1@Y#t}RO~#FE*`3ekg&nea!G<iJ?w!f&FAAgiQ@U)6rt{~#EZ(k>
z#dI-~=MK)|O`>x&8usy?(3{-(+GRdla#1$K&-2Sod3-J;k9FJT@vcGVSZ(!r&e?jN
z8{N-yc<p)KnR$^1X%ui}*(Dxoc!`@X7I5L7i~Q=!IZmE@mX~Ut=ArkFaAmhc?D6{$
zTLt8DTcf=kzjzPl{><V0@ErEtoXvD7her?J$Jfv8;{zvU+qEi}eWZ8Z_sC<`AM}_V
zk4ksl>Jh)SE$6dVWt`iil-I0&$U84S;IzH>Ic-@9n-$#S1=jcY>%u#%`r{~v?LWaG
zTaK{v*Ib_cE{Df2+QmMvw{x+14EK2x#V!jX`I&13yVUjI+o!`=uWJ~m4DP{(`y%;X
zd2e1L%z`s3r|=%bIb8p7Atwee<|~sHab=ZkTJBxMod?d~h<~HF`+p<2<?>;??wPRo
zW)0y1{|)Be4TJc5%m8*BAJ6Sy_T%bd{rU5e{yh2M0IuCKgjWu+5;n9otL(64!%_A;
zu#Y1*E_dRLckOv{J6Arm(p|hFFTM|7KD0J~<2aD396Iyo*Il{Q`0m0B?7@!9BYE)i
zXkNU(Czt>1#T#Gt=A@;4cw@6zp1m}d7aGQK)TKE7G^H;aX~%P$74e*NIi7#rRe;?D
z1vDK0OG7>WlETeDWFY$Hn$u6(y}OY#Mc-VtYM^dc>Zx#C9jz{x4)c~enjI(_=jnIK
z`~00IwyPsIVaT|SQN`F}s<5zBLteZ(>OZQ>zI02;K4_rz08Okru8HzuO-vSjbNW_m
z=qYO<Z-ne6B<WyEs6Ga|>A`e^E)*kmg+r%{?(cQbNp2ZUM-`#JQvqA8{*i&&Upl+)
z7rACPP?5W6tYu$F|Jp|y<6J|dS5(vB=ql23d`<(0Jti~lG8*FYfJW)wqiznjiB8`j
zt^3!=Jf@hQYh5MRZ<nd0)n#fwt%z)$ifFJ-5iM<7M0X|>(S-L!^lQ&$da&sV1?FC*
z@GHesBD(0ozG8CUSWH>lifNY4Eppv(i;DK%rfF~PQ0pQ0$X@k6x!rp}BlAk><*jm>
z`0FtZw|hnpy(;L8X(gR|T1C0DUXi_GHSI`xL$Sv1=-TZXntkj8`B&D`#^qmV$nWpe
zZfPU+k{tD{=Za9;+zd{&DwvR~ilJWW$okq6YrbjVvZofp)@tKyfi4;^=)+^ZA$H2^
z|Jx^$X^9t4Oma>K3@q`tj}^Wzw8qdiHh4VO7L6V4u%*ln8+`2Xd4N59V(f9z!5)^s
z>=1L)4*JvWuv&7ym$hxMYM9)Oa!g?tZixT>>fnf`7J}Eb!o!7X82Ca(*vZYHe?$qN
zG!@af;U9G>`b)MW{?OpEpLFB;57OvePY3jV(DaFo^hxCxeOxYgx7okw$&e;GtD=R+
zsirV+FvI0QGn_Bbfoq8dY!hFRYh4As8T*|4te%r<`{r1iV~DqwMmSYzg!6BVQF_M|
z_501C{MHh_=gl$0&jO~2meAf|f;MCAkp8kg9`u$j)*@$&U*-&r9?r-bV1XxVEHOFB
z9Hy3*@X2+=ke=;?1MQ6JdLLX~;fst<zIg5IhgL)U@GHs>)~)=oYML(&2~#KGfG^6T
zfR3qwSf1Jm(<?f|zD-w@2=`}JuW)2O49C;k5eRq?DLtGh{MQr(g-OwvuwS;(??t1-
z4`6};U{fR6orrK4Bt_uMv?%PdiN)&^{SmoH_|OLv;JSMhu6!Ad{Oe<2VKNqx1!J(y
zcnn<R=R15Ffw66d<HNC`C^1UFj8%iM?o~h73!hN$btHCIc1N51PAGB#rZxA$2~QW;
zlt<vi<sKNlDF_{(IKoEL5f7HxBhbYTBZSj1VW|y#URq;;?1hxKY>ShJ9HDOJh}JVD
zb7O6X7vkN-ijG+?`etLaJf}qCv=bkvi~M*2nIp<t_E3J9Ay%~YM{_;6PS=C8Y|r)=
zeG|1?A0_WhQ6{hJCNKTr^u`};KS;jGfFKKdh!h4v!7KStmn@BZpn>r1LI5l;2B2dr
z*-iBffWa8iFpC1vO7zVW@pa5q2t(ckpx?^?EGi8^wEX&p9Rc|HoiMtNV6=wd_*Aq|
zf-oY-`6GLnY<JxDhnu@Mw%=)wmS@__@68+CCJ6`OzAwIN`Qcf%Gd6^@MZ?>6(9m(l
zfPKzbo#Kl24_$FgviF<sxuN2;8?@!e;o`w8v+zRJ-OhMnF$m+I4noS4LGT;g4L?Ex
z@hH>-j}6@MbfzofmboCHo8;Q`BazU<3r|ek5idQ##>17A_~!-1HN7BH{YsiUubM7t
zhT_N#VJR$)g1=cbsCpb+&MlP9+(e{TEyYFUWti=<Omb99ap(O~#J(DVI$>@`6n4j-
zqEICD?tz#2Q?M^)F#=MSNFQq~@)k|S(2(ik<ygVwqZPXCUyiksV`^u-26gQ>qiU{^
zaP5rvWlLjD`fbEFQw{mTx&4?mEg3V{Bx1w-71%j+EiNk5pzrb)+`dtj6&k>oUUcPm
z*}=RqG?Xh#Liv3_7>`N~<2NQfcyD<R&UxB{Co4ws{aKORdszqvb_?OJZ9{m%{t)&`
z>cPM5BY07(2<~?+no~PQbIX@e+`BQHS5X8{FdoZwfg(BsS4*E}gS<!I%x*n4a~t6z
zk1N~E>RP+l+iDl9sAO~Bm~7dt&gOub*<3BRr+wa8?0RK4-^$M5p3-SPS&+_0{$}!{
zRhhiIS0)!kX7Y-cnOvok#m_tEu(#}GKiP1B%Qv28x55)VNbLkyygAP0D^GAk`;(lk
zb&4~W=W&S9c~&yN$dxnl`Jq|?x80D>ow{G-J>u1LIe3~wuAGtGvQzwS-(j|jJIH~_
z`?=}q9zJ^@n+KchX5*<D{6Y3HG>2wN4kwE*hR8m>+dd8+wx1Wr<#I(>E-Pt2=6<D*
z*d_20|5ka#i@KHZ(siZm7*NVngzM*g`T_r!QNj^t_wf~-L;O1XDEI3xJ7PKV`GQvS
z!#L^5txw_(%M<y-<fZ(=WD%c>o5>gVPv!NiBiYS5g2Na0;4;%N{uUj|tt`X1!-a6J
zm?%8GN#poM_GIq*-)#P-yMTW@U&y9Li&)=rHan@zWM7Bzyr^*$ADA+N-wztbI>!=N
zJ#Gj;n>v`sU+>R)Gy1ZLX*};;7tc=|BvYj%d7f92v3KkxTW^-^f5(bjt+SDBGdm7U
zaNvdiwdLpK?f9&<3!AKR;|1=XJjB9>TlMtkJ2%07$93dx--5VZNHC9z59NoN;oKrx
zJSdZBJ~TasZ%*yWDsH`a?zLXrDXusF{M?&guJ6Mxp0T{NFqXSI#c_+lar|qx_%{uI
zsd>g<dSv>S799OUIo<xyH;11zc55U3m)t-jTz^owWQkA2)lt^vI@&q6jzU!G=&Q?j
zG6?@hVQSxLv@rEfjF5fK1FEQLsfOCN>d4t4-KX2?m?rPX%f@J6qGV$$lQm)7p|$i+
zT4RWo>?91(!u;Vnh<dDpGx9p!a;7e32I`_mk&f&s=m=9x1;gW&Fjw@=%I5#5{i8pm
z+36Q;9^OD@dUf<8`3t2f)>3_J4aH5XrrTYrXqW19Qg?k!r-Uc`_Sk*$S#y_4(r!tP
z<_48UU8B*DuTtsmD>QHOWvaSRM7`{a=*;a)WW4thbv<;6?!CA~_gOrgOGTvVe3_E{
zi)c;vBJw_0ObbQd9M8K-n_pk16Xw@wZqp4)nR1KPRNkhk3+|GkR|)BTzfTE;52^TA
z8I3A@L^The(Braa^fa%6v=>y8kKIdBJ@<<0!mDX|?HelE_l{CGz9;SCkJL2rGo7pY
zMqlSQ&}-knR6SP_!#$fpYk>;-e{Lb!4>d$wSI2_9R`99R#83S;*yyf<1J-(YQ>%|y
z>8B;NHAYZ|3EC=|p{a*CwoASxGR;z$#a2iVoulDrgK<SRxM6CGlbvmm8eoh0I<|OG
zVS~6GHdx!<2JStq(K|!>`-<jxvd$R!`}EQIqBa)j$zvz1XZOdd_!%Y)YyD<e=b|J#
z4+?m*s!4KDe<<(vFG^X_C^3c~)LBI`=u7KqnOy@-oYF`GJ~z_J^NrMVUjwE6`9b@f
zf6?HzrZDMgirMn}UMrlRy9=wxIIM!wOhl*dct-vv&7gMOKp2$b>rFC(N^fJ_S!06c
zS4~kkzzo}NnV~*M^xS)66l|2=r#z>w-)e~f3pK<FkJ_<H1-*y0z_}Ubxbn#yJir`F
zwJlMz(-9is?a=qUGae@RV6N<~dv5i`-AZ3Pv-1-n;s@0VUu5<5McsQJj3r;}`U2ds
z3WS?}N2E;d1Y5g~coQFl>+YRlLbB`W(i!!)BH<G(J@jW$Xx1kh)22t`-$%)k+yO35
z_D1CaPv~09Zq}-BROv;+V@?dzEP7)$;&E%-5KOx}40|F*!8k%ZoP%Q|t0vyfVPWZn
zjKaX}BXFa97<Q!%6=q5T?kpOF(oOMD@6Zc#<m>(&XD{AKC!9SCM2c1mF?GjM>Ci7d
z>m)Z9CxpFm#H*H$c&Flk^$vEZUvDcJ6&oBDCPU-^YuK!^NAV6jw2{4fO$|FtUMRVy
z9BXV!FvlQ!bDXs^M+?z7KcdXxFVC%hzLL3VV-Al7Gd#a(hNt4g_CKc!-?@5twMGv~
zYxRLDee~GsiO7fkXngID^!NUlu0;6kN$4XTY){G1)QPV7r~pX34|T$GoGrh`NG$-#
z9syW6G5~G=3&6dM0K{Giz)`td%@F-k{zw>2;@zy3U+?nC4=J<!U^`Yme_LTE<@sTd
zub=Fr%Ht?H^_+`0=05ijpQk<QH#wp_zyTxMI3Tsr9`ow$acaARbP*l!BqEo#P0pqI
zy^`5l;|lA$uBdWxgUve&Tz-9klC1a248~qkeZ7x5h}QWrN&b5;50s~Ppp~!{x^?!%
zohfcmPj<yj6&J*|b;e;MXS8@J`ud40w%v1ujnzvkGpM9#<xlCu@u!r0K^PQ~)%4ya
z6m3g;z;8(uK8k;{aP=tZjn77n=3+R7CgS3aL|pTh?tkJ^1f7|K<zGg~-+y;}^p|au
z>@M)J4#C9GP-wOe#}n}oT@O!&@n{o-2!pI(wFMOVTcZC+8}xXcgzfc`JzBOJj#jI&
zM===<?#W15mW<8T(!bYBM#S#Ll4YBZ`+;+@D1I*bg>MyKu^vmSTJYVAYP@>{vF*5E
zZb%K`&No8Y$uo?X{ujo1x;;2!V-Mbz;Lp#;dGXcd-dtPj%_|dpczJ3E&N&*wsn5cA
zcS#SnXbNY877^^)Jc3s$MDUR9VSF?pgfDn?=c{R<thH?|mv39gsx#N~_#+!Q_+Sdp
zlugq)C28FIXBr32NMnQQojhajPF^pYU<P+KanaH({E@b@lWcCy*_qBQ&ZhILc+oVn
z13$NFH@9q^$u$|7!hg%;oB7$iyE2;#SDxizKTh&Y|Koga&oMTicud$eC-|_=NiO+)
zlEYi)ajohZHtT$nXN@_@({PfV-<)8RQ73rqrQ_Vt>KN}UKg`kpq{nsY5XZ(J;Qdqf
z@mIq=?EWT;FF($Zebo&9^DmuG74PP+zcbloihS){*@QRU&r6%-^4Gol`1ao%p68jx
zgI8v7quL6Nw^_zlk0kQaUrTsH_eAa``yC6**RrD7R=#Agk)M2${{Mqb{ITyE4$)6y
zrCCck==T!#7_^utbYIMA5es<0ni+g!=2SjrF^OG%MR2u#4=x!V$`WSbh=bj^{8kUP
z?m2=#=#1xfEhh6(ui4Da7x0~-;^E9+$eR>r^RXK<_{ZQ$T(?NJ6dDtF^w&Xb;ysX$
z1P$PAC;M~X`*B?A+?RJ=>dPVC@tpT0p1lgBD>t)058Bp~cWFuP>#!vUEwbkD;kG<^
znDqCSJ96)9PP{|cnZL|;<pm}l(rfn;?v)=q#t^SL708!VJF~u9SDs@S!qX0f@&(b%
z?J^=bKskzcwu$D*m(g649K(H`dUE`go;+_zFII2p#ZI$&b4Xclo}ki)gKV1U=8E6s
z-QqVzKlnxGr7IQk;U^t+|4IFqHPThrdO9dsn;R~5q`jn$biBTkk@z^<Uw)^e#ouV$
znJ+Xc{*(Ow`GHI~)zHuH?<oG}JBoV$j&7b*!!F4C{3caw_iZJ-e9<_QG%>@aH6HAd
z`$|hKyz8rlA02g&Ia3FB2k0Wdm#*B?b?`G&8$15f#?d(G6!#N<z(NsMl>Slnxj$qm
z*_3>4pefBnV{Q6E1s!WCHCa016RW8QRZ&#wGkW*p5jm_arEvfIWYlzr>i^uNyyWY2
zT{Oz-URTM*`3ik=yG&aXim3YfC3-vZ5+!?GqGG2@6c%!c`mel1x1V34mCcH1@9;}B
zWPBkNA1|OwtqUn@{3Tky_A-sLDW)?UuG1R2NgMCoq9t4IP~yOQ6r^*XdKEsPiVdYS
zZ+SVrTmG1iEO<)220f=z;SU;@R??j5FX_GZYce@kP4)BMQvcC4bZ5&48td_iz9tLn
zn14{u@qcKyl_E~PR)!B$u$Wtjx2%eh>(zuI(-Iw5YG8U+Yb5V)gS0g|$coZKV4Xg0
z&ND>nM<XbAHNlD`Q{*2p!}~&WoH%8Hv8ya`&DRR|WwYx|D{BmPv&IicdHqrpF7y>E
zBu}@3(@#sN>R4itWJvcH%fF8s43Vayhn&f6@K0S6apKEaSgPX1p5}PEKpBtI6cN!}
z0n>;7rCI!&xR*S4IyKPK*m}A=w2rFI)RE(7(Ksy{=-kW(Dzp^u&EyAl82N*W6@SrJ
z1yiUWH^Cx3xvwp2iC+g_P>R`ey4B$+?b-d5EMGjQR;kMPs$>ZIXNYO}MhG=D!POX3
z<fNHmz!y_2SYd!sq=!u=+UOppiP7pU(e;!nwn=~eoRJE~H8jU@8#QD|uKCs=1K3=!
zfMuu^jAaw-X09F9uh5blj4s^Y>7&8R5Iws2!g8*!JpX+0uCp(iZ}maJ3?D4M0(_LL
z#;oNXFl=T=>^Uu&oc4h@)<iHH?1w1<zQ}p#E%}j-Sn1vgr@BW&OR}lU5`AEA<&BgS
zFQ^Rh#LyS+_#WkrJtg5_mk5|NL?Wgl23h+hcPgErK31JESF%01TSLUR>kgmo-7$Di
z2-f)rW8Z<U_|V@DarZ=nJhR2~Ig+EfV~5sG_V65Sj~83)aUmEmn<xCAt!}`UcKCD1
z32U{TkpI~cXH*=qS-}Ah-r8ZvDqAGnkzKTD);QS08hbJ&Yc*dw`=z$hHLykUaT`25
zW+l%bODtX^d$VKB5N2+U+z4~j1e+t@*&K^nn?qmpP2pz~WF6Oq#vWa4{-g_sb8@Fj
z(#M=xp7@*KC!18h(0b(yy+A)<IQt<l#2>DU2yquAOC$H8MUoS?P6|LT@nTLr^~bXY
zf7I&*;IK;oCLWX=&Vc~wtOwxC4f*y#0NhIgFiZ5#O+EQM&-|d6<bm-Xp4cW#rEXfD
z2>IfJKF@qm^S}r3*50^O>4AIa9MN02=->Pupx)X6heg}8fjvh45S^o$P2b}8kn@<m
z6h7bpjo9ykXZo&C^EpUXU-pxSS_TCa?55*YnFOC~ntnEyzAtEx(x;N^Ds{&5VrLj0
zaz<r|9bDH6(_n2ogluv`iLeWLEoq0*vLU>`#an86SV=1vKck7#fw&O<m=2merj&Ls
zsm`&Q{+5Ly_-8m0J)&`MSPa@4%|=Ar9M}z549zE^H>&3$^733vyEGaqvSVw%xC>0G
zgAn_}AHBW{3#!QpcQ%Ph*r1R0g$B?%X@W}W9cUStW0bK4-1k{Q=e8qGl(a*dohx|P
zDtKnCMz4>;Z&pi&i$OA!{;r0>qQ$u0b}lY7&cT?bIoPW)7gJmO!271=Jmj1z&vb9e
zrt^)sWxo*45ADvKT*G)l`>vekz&xnJpLOnf@*rhz{<y@Ox7T>Hmgt+SYdf&FL#T9`
z!uaEXaQ6Hb&W}XnwEYv#s^b6Lunb{K@ppQ@?9PTRYq_X$Ehmp$$Hpz!OV@n^Z=IIH
zO&vD!dB=^s&TJzO)8EKJ!p(hbzLEXAZ{kMTpYE8vnI~s#Vg2=6c}HkES3gYWm-lva
ze5~k-LA!a={oTB8U?!JrlC6P#ne2Ksn{UX5yupA&+<nMVF7`Ug%kpx$S@eEZ+nLKx
zA01>T`MKR@C-~c=6MRVNB)=^>!RsfU;HSTj^S-{KZ+;))xqgTFf&W3CP`sZ<zTC@G
z2JK~^h#WrDDvRH5Oy}L5ckv674Og|Tto&;y&oav9*ZG;e^-C5%tI6TRX?wY++^=-c
zW^!X;I{)mpf-Ccv^5n^hoLIVqBYrRDwz7X=6}pJqoL|N(nv!|JNnrxlt>E^vR`83A
zW&CLPQtl#JXzt?$?Amt$k4T!$pX{cy+qWrvPA`EsI7jk-cf$Bly6DG&!Msl^h;3H|
z^XhBEdFzQW95H_qS6!UN$3D#GqXr8(%61_i51z~F|4ru!(f#<#g*bkjCtbI7eR<_<
z$-iui<C!z!_(@6}hdznp=$tt2_N*_jsPD@Y4f^q9zy5r8qA~BuFymh<EcnYfE7tFC
z!!rlj@%;r3tbDpHt17nV?IT^d<u5nhf6<c@PWZ6a1Aku8jCq<vM{cwU;-hc6@~8nJ
ze0feN2TCsbW_maWoQvSysgZ2eCyF=ziQ-|~qdCYUhHqSq;T7Jp5BFbBK5?oi-~RQR
zs)zlil}*3sX2CDgp7x7gRQ)7h|DV)NJevlidK$aEj?T8IqlJ^}Xrje;DtY*g4sQ5H
z$4tJ^2LI1AJ))LI{drFpsD@q*en-M)q^Mi(s5-rd?yY`LUuQQ^&6XyL&Ta;4i&mKJ
z)Eb8At?|CO790m_VT_>;wukA!HdqI%Pm0bdYlEbjZIBfqTx5A&ZxsKcMT!!BZ~sUA
z#M4lG_LDj~H&B<q->Gcd7kWFSmOk&QA*ClT=*{a2N_z90{v3TodxJ~qNM4C_qT~%?
z;7#gw>l!6%7t@WRE0lKlGMyHGM&a)zsu_BTI{Yjoi?fAvZC4>x>?@?4dxf+|;S%*y
zxkLq<3d!S30p<T!KpW-T<93(Gp-FC7?pJAQ^fgLRk^J$=o1}B}HWjYEOKrVN=>6II
z^l$7#>K9f<kMW3}I6a{ks?W&q{&T8a{(_dcRZ&^#OM033nk+)z&_(xmG^y8n`g89i
zjcxo)GY{0!?UTQ0pLDJ#tx`t3rUHs@DI@T-3Ti@BF+w=eVQ#J9>!XRbPFgTf)Q0m}
z9dz%dhszK3Ayi!Wtv14<%f`rlZG!jjP38Tw8H#q8<5~v`c%QX^or<OGqFQ2{nI%Sj
zx4_yX7C79;0`Zc!X`y6}5uv7-mTQF0S^D@pOb2BqT1frY3aq4#Uwv9&PhB(2eXN8+
zMMd-|Yob(#KlH2mC;3%3Q04b}GLEVv??vCKeZhBXTTn*|7wgH({s#>{SWi#8)zO(L
z-)TV)MZDf@0^@-usJLQ;-m!Ail*ez)$)}X_;Ss$W`GnT{J|p+-N?6&?0Q1Ge=@4Or
zoxP1P)6y8H%4I)Q`gZFg+koG<#*^<F@O;}6?Uo2bJ5d$Cgs*-5VRNk9)*MC$nxm&p
zb1dqqf^|B|*e1+@UCR|Pw66kU*DFeP>@WF8{GsF@zv<Dl-}GS7Zz?bpcEGjYbfjE#
zdT>kZYGsDWMiMO;43w30K%Y(>;FJhVqyV^e^uu3cUsz1=Mx3H2;{Up1c$zz0%{`$%
z+#5;;-nh8H3&k<w-&}NqhqG*j<$GX-yDJL6x*+(mJ7$^LqiKkPycc&u#iaIFJHi!*
zM!R8UH&^&IwnM*IN6DetLH)ihmI(8EZn-V0#)-bUW`_f|vd=cs9zVC+!=l^?wXxnP
zGIkStr5&txJE6^Ld#v1Khm2)*&~X(`c$FQdT$NpqfzrM2Z4EuiY6Y&g!WG%7?R3%>
zr-#|%jgoAeWmus@pgBIblsn=@6I>rIJ-xeTC=*??poR3Oa?KDk&jc5$baDHTE}kvZ
z#kqI7P*D+H(QFT7eegj=urJ2X_l5p*UtIC>!_bR<s0s8(rwN3bV}#u90k}Uu06jnZ
zW7`?IVcqnHX`28j>jz+r_%ppl-%L0vf5&pSI(9Vx&&9VHcQXKA3j)xtxj!b(_rO$P
zVD}fULEvjwcu#dh)=m#(U-yAtfe%_6dQ0xl6UGzUqe61i55>RvqUwN0(k=8iut)Wo
zOe*e}MMFAdP}^nclyhqrCDR_7;l7{7DCE*}N~f*;)2TRiH#JYoB;~v;I(Tpo?J_t-
zzTa-q_uhACr1Cuqn^a0C$COjgkB@2L)EYYW#}02KqrJYpBW^f5;`6UJR8w9_563*C
zXY$+*wtr0L&OD;&>5pi#@Ya7iRg+~^57haEqx&J*G-)3LW%D_>I)5&5i|1m`p*ff&
zee^s2!?3M?INru~L4xSL2)V<iice=~YJ|WU1{nWY9~X`oplY89EOwZoR#zUYQ|2(c
zDerOmSYptsc4%|Z8OhQSTg1__m$DiQnkS1cOh)v<Rd_yVF$N!*g9m%(;8n&P%ubwx
zeph~A=IiFX?Y|cMvrLT(eKojoy#`<O?7%&q1@fu!f$V5Ud?>_|ublMcP7YqIQ{*KZ
zR^A+OybEuB9m3CDLs@ryIIlev&c{l^dHc(74*ed^O>M)t#w>(we+F~w=iOx=dM($>
z2G)$Abv*XKI)2u5Jy(vBJuB0V93{M5SJ^u95~kdUpDFCun8FdFZ$^v0vDmYj&HkT3
zUM_52rCq$WB2D&?cJsn585}9wqlb)k^Xayk>^3%&v)^R!+c8<J)s(|W`X6BDzK6N~
z!G2yMH>2U}_Hk0jTpqmhAP1Ekkw5cs?!DjyCmuP$DiPw_{5#I+>yC4f@OQ@gNbY6n
zA>NvBfNwqD$Lr)Sbwn6F`YW@!T|_3kU*5&F$I{rN*9JZ_dkg0X7cV|9l~YZ3^6?{S
zd@6Gn=kMRelZ&(2AUcyj|K7#sZ?|!j!BS3BOJwCGi#ccFA|7^d0e4xqkUPIz#y@{8
z<Kw1D{I8>Iv#KZY#q?!-reP`HyS9Y?&703NJI`lLlPNrA!&D9_NZ><@d-2%Lk*xDI
zjBT5B=S-KboI9Wsx4IC(_0m)Nw|ERY=T6|lLzDTK&17LxPvn)2CS31k%mpy!VIyPM
zv_6V=n1=JPrP1tptq-eBh~pZ+I5v-rW7Pq%oTm}XDw})q=z=&NTOZHI9rRgEGG2bc
zraW+nIWLa2WTi-JZr0nDD<=u3V7DV{y>sI0A<q2mjw|~w^bjtzHwQ-f@%Y}vN#T;i
zvF^mvih_8dMK`|PE`<BN?#^F&h(4a!gB|*Zv!UGF`reD+8I$C8)jWz1uZrT1Kci&R
zE1KU;jOMMAYsu$S138ucpxJY4Y1h(PDlV?2Wb02fZpJ6NvhNe!D)~gE7e3R&xbJkM
z^>-S2;~TwP^^G>J|4d;oYpG5AN3xjmo{Yb|qlFIdXqMSKQn>JrMyA%#?$q~m;$bZv
z>i?S*Zws>~N*S!t96qw~bf=&>0y5j+WP~=N=4(R-+GzE%4IUM@L4<?4uwJBpY^s7?
z2D15?q=doNig+yV`|F>7r3Y!RDA==#9^}8E4G9(WwWOSiCYDjpe-9}=^8p<icb|mu
zNcH1OXzzvFbhqCP+WV%MhW5Tn{*G7ZuG3}G3ooLlYc7#ueIYGeQb-X#g`}uiNTtoi
ztI;bY^LB-_!?KVT-7BE)BMN8>6_DM&0_uISkUqs1QBaWduN;bLg=Bi{R^A}X`?pBv
z${o5d9?sFdCFG3<^sHqmrF<(R$0v_y>ZvDWzT_Esas|16d_m@0s%TosEAnYoO(Q<P
zq45eev?2NfebM|xd-cB4FY_;?(7AzH33q7c*MDUESqTp2&9U>l3Lq@t&1JGnb+Z-5
z9TpzVY%Q>ZHb#iXndz>Fv+MNnrrZFUEsQYT%owALOk@Mq1RYM9;#{N|j0?^1TV9)t
zY|Y_iC|TYbGtAs&1_Q~#EWB<ion2FS3^hhoh5`P{-O%Go8=TG81n#!Pb9FUTEEMjI
z=+0mVB?QGO;PlGBl>YD+H7;(X!f8Kf{nk2C7A<Tv@*54x_(oke$nMwrI<ix!r%q$)
z$Ya(wdM!NZ(k_bFpJ0OA029po*9sSgRgkXUGwPb~kZw$QKnsruA9~gky0KLePU-sC
z7i9=<1tT<TW`w;i;^n+GfSE!YRM=^u`iUm2bTwcR))K9|s$;UhDvUH$;6A%KV%jyw
z2w`z=4sIryXJuU1R6?C(>u-JfM<#hq)V^aA>16$(pew)W>e%1pm-w3&g#D(nlfNin
z^KY8jwuyX%ztp?G26P6Rp_h(5OixNKbs9s{3uvw6hD`&V;A`C$o##8C!o(Tf3?u}z
zR`M?wU6A-q9q9?GD63G#%HA6Io}z&-ik4WLYJ=@^GidY58vY@|pjNQK`N?*eblMJn
zMRw>v#twbQ+hNx{>BGy{JS3XL^Q>r+eYWtuWQ+4>Y|%E!R<hl8XfE2tO!UqB<qmka
z*%SSjdtrFJ@Qs3;FlL7^9#6{WO|e5#7tuIAb_lJoMX!stsFwWAqxROQ*0O@`5KHla
zgxRpo7Ew{Qm{M<pVe_nzB+r$bji#_aVFIVYCfFff&d|+r6Wm~ijVi(cDAGe)=wagx
zT`Y~z#bwR5_@L&6B6V+A%jTKRVQ-xE@xi^hKCm|SgVhN?^of%jQ;}?<e)PxUp8lxu
z@kg<r+^`b-QE=8DWm*CFt{Z?b(Kni+Z_XW-yH#EQw2K1ZDf#2k#R1qLdS^zx9}>rS
zAZdjgzT9@j^rx<}f$D~Y>++aHdShsqw`{n0VQ@Ry{+Qeztsgod=BNX{i-)!Lwdldm
zl1HoFO%tzWP@G~0UCv9RXu~wR<+YQVV+R@kNu{wZ(rE0GRC;B(llqKHqiHjC(W0^G
zv}@mS>N_Nl=H#5CGl%Zbkl0ds9a>IvONH-QRY8v&Ka;~w2MpDAM1$m>E<6%Gg!HDA
z_r9QyJ08>ZkVj-Z=n?gq@`#%7nC6~;LgCA+>74yLdf6C`-t!|Mup$h%4@K>yaO}Sw
zj$30Q@Ukfa8cx9|-_BU)<q98Txu>PJlO3(Luy8kk%X57!+hu@`>rJq2qZxk54KYe`
z;QXHj{>b~8Te~bVE6WC*<o-C%OZ>%*ffzq?1{{~pMXA>!xbziWqdggcD$}ra;#7Q5
zngeyrz}ox&uso#&A9$|Dm&a+ccc>2M4K?I1hs}6v4<|0KY|r*K@-N&DPp&fe;!k;A
zoIk*u^G|f)l$#-Z%_3B?E8)EUU^tiG4(C%9;auz(#ygF}Bsbih8)|}OH>f-3d|1P~
zWgo)&)mk2~X&t|7wVu1ISkKw-*UMcdg?-<p@UB-WT>nD!O?V2YA4=iXrsCo3*v!ty
zw{Z2KoxE*S8Xqc7<I$lzdE9|?9%8qPV+(e%`s-ahJ0zXIavIl&r=!L>ymM9#YlQFP
z`nmhKlk+|<+Pjx`blNNX;W_*)ejl3*I>5;j4)N)@!@NfI7~fJn#*6kJ<?`wyyy3sY
zJhx>o`xflwDDi_%p2%V4R@r>;Z59U}+RbnEw{h-h*=1K+&%1+C_=ap499W&oT9rHa
zOp7#rDV&<dDB;zNPU8t`JJ_vaD<Ar?jW-4)vewKc?7U+kU(sK{ecsMz@6!2_)m_Hb
z^~*R(FNuG;Byq35%Q!!68T%hx!kTWgS!v)b&gnmqdxVM~Gja-F>72mt)(MMfbR@T2
z6UKecgz&+Co!L8<Wn;@<IK%GT$<v-kzck{(LAq?VuQiYB*^2jtsI!{Bn%tOF**-yy
z`Mnxvv}JZZ5y?weNw&&4oTJO4xo=bqTlS6QZ#O%#{?5)^e65XamTI%{Zau~vLq4rz
z!X<%boZ88P%{o|dL^m7GA85y;S2=Lq^|oAO-JT0lU0BJ_U51rC`Q;5CuH5U-odhHG
zqFo2>eXtW(d<o*}k6roZ&R`y{(w((MC)e4?b4D0B=W}{+W=J?cdlJsy;v-nID1ryJ
zisWwYksLZ+USsCfl0`x-S^Lz|Rq<l>oUNsFvrp7^%qO~*DtiV+pD0flyZHw{k#@o-
zTG`?gJ>OkRt47q4;p2}qa>fTzdQd~{7QLe_*WXf)TW{${hG?7}HPmAK2U@bZmTpb`
zO>I>F(z>e(@Smv&TMKzV{y+(~y_?~an<|dPXkvCk8<>>0K|PdFqofSw4@xL;RL0@c
zO7I(~B;Ni{Ds%Wj71!TUiD?yW{_h1@TzE=vHkVOpzcTt;`jFNgdq}>$9?<VM_o>4w
z=`$U@M~aJXQtr&_bo$;^N|H|ax5Ue2mtI7PC6}ns=n}12P)OsO6;kGb0%|>@fOG~F
zP-fo(+T5>zI(9FhomvI-xg?)LZs(IvrvmX~3aLr5I_PtmeAKSex}C)|iLO)I7PrV=
z^A5e)c9(u8$zGMheR8?>fI96hrQ`pVQ_#rA)X)1Vy{LOehxZCkIku9PtG%S&H(pWy
ztZGU~e@oT*HB=#Hy>59eb*}$Rs{<P-E46{%mj9+;Cj~rTp^QP|MLbGwftdMf@RAMp
zL;e~FZ_q@_b}h`c(Z<USI>@Mz`+%c9q9Y8@qn{zBh8rQw+88IxjM00f2`o!YWS`s=
zYucI$zr_?Mt4(leg9&as%1x`#7;`2Xqr*`{)Na>9=U3Vo?VyFlgEY`PQysHZRdIAj
zb9@`2jJVN?DCzc(V#T{TIpHVW(rTa=zw1c1o#f81eWm&4U#b7VuXJw6H_~77ozz>^
z(UDQ#sZEQoRMPw-{rUKk+ykG`LzL2$2M<VP$wS(5`~j`Jai8MM@6$l1`!w;$ebQU?
zh&-MtK+o0?(;r9=xx^4<l?G_65j_{Jg>`nVG4YiK+~f{=W~MrnX9_>CSqlUmX$H#+
z%5eF=HawstTWZR9Ql<p^JBrZGS3ttHe-!21L?1`|rB#Q1()G$ls!4974zC(%%e6*o
ztMQ96JbqKhS$`?n^B<{LD8RRm^um@}Vb@!0Ja(|bmJ!xCIaCLFKB~f$Rzt!~Lwva0
z4&8;FSETHMm>T)^K?{r+s0v4IHB|0XLyg=PlYVQ!?4JpIK3U>~u%M>Dw#4d=)(Dl(
z(wEhCk~Oju4^wVMQFhQsu)~hKcIft6Zi$ArSbke{QI<6(o7mu>jvcDa?BsKZ29nJ1
zoB`rXNVcR?jVD%icg1N7C)j^=#Ij^NER?%RQ)fF2_m*zGy&cTY+hU@HEm}2L;jMiA
zib@N#US@#^$!ebpw1sP|6%JlE$F`ql$a`vvS&1g-HOm-(78#*Wg&~T@7~<3!eXNPo
zhnKlNxJVae%DU+G&K1`uc_F-ow=f#LVROJ6;dVY~eCv%-w|o&f)E_y=C1Z2JA4$vn
zp%~?l5YacU<ikMjR-<$Mu}DKQ#oBVOG77+Ss{nXU55Uve0k|PP&)~}eXd|~N_X`22
z>FojgztZK_cf-;1;)N->pi37wWUO|>pkQyPJ@CT&YA<-id11uV_LzBFIE6<X5G>i7
zfa7+s)v&>5joqZsFM|eXWzgNjX;j)OjXdpkQg59dlpUH%qh4$yHyLCzys@1|zuQh-
z6;jDfE0s>1&!CFiXK4SgJo-2M0v$ekiTbV-&P3d0>VgUyocfB+h))}=>WDzuOnH@D
zO>-|+(!!h<6r}u!W_&6q?>A*+(7l`<+dL-wNl$71xeEGT`-zMq8b~Kc-apK40b8v=
zER}rt(~d##+0hw&vpZqPZqcBIZ7}wRHX7V?F|JQLgh`ijVySQ(3iahN)yLHHhB*4b
z2xp%dqhh8BN+($0pWL2Trdy)aLwvddEBsBghI@W6sw~3rb$kS>Rr^70-XLt+l7JU|
zM`F?YvB>cnkKA6<abfXPsP&$Pmb%k1UwsDpb)1PQTV`W;`-N!M)rOD0XwU5@xbpMu
z?tIMJlie-6c-VO_R*Ujw_if$S#~_$Pmxc20bD@0eLn!C!hVcd4Fs^#gofn1$^T`*%
z@><oM53lXc&SA30cw;U9H*X!!S6|PM<JYs*g!OzYaXp_txSlJD*K_qN*`wYf+thny
z59xsHmd{$xs?WCbU-O;p-!qNpN=N$5!tMOw<4(34n8vr#ck%taT^u-Q7suu8<iy+z
z?z$(FKOV{AyRGG}M0@%4&po`pSr$)yk;N6?vU%0Iz5LTFmtPJ&$Q$(!^OawRc}T`l
zelzeW)1D)&rg4BfPu|a~4ED02^wZ1pGq}sREPlFoH!Gak#KD#ucvtLNxzDZT;wQ<h
zZ<Zpj?Hl;n;T@9Gmi&<DtS6dj{8V=*ugXv5lo4AwTy&bkuq8Z8czcR=^SQ&wxy**5
zZ<fzv#S1ezB6li336>qFLlZgPY9jl!naIO8PU63(rt|MTQ~2NX3B0oF1U@r!1m}Mp
z!nq3v^60(&*teB*ry?W+<=BaB>xnf_dU3?3cDyv)niF5hR;`{cADh;iPnxyj_sZ&g
zuT+(l2dT2sP*n~&qsp<4Y8?AhjX$?g=dP9=ST!Yz6YV7<zM~W0?W4in&S>(Cfm&QL
zTbp;(=(1~~0cZX&;vqJsJg>buhdNs_ds*|>5L;e5!JaktIP$t5PJDTYGf#f+%B9I3
zTp8reg-yP!doX}?0)_XN*@3^^?ZiXR2eIz(uB>;f8=E!;v-<ZCt}gD*s~3gxMbSF{
zj)(CT*B+d|p$8w8jJwA#>0OCNc`>$@^df8Nx>+rG`qq--@mhLs@QF?i{X|RFexeg6
zMdOHv(`WZ5Qt$hTG@5EDZ(S{|ZdOYJnth~zwD)wxriMOzeM?2bZz-YQThb1BM}6eA
zI%ksff+TO+t^Xg=`PW1<&nm!CHX*VMlu&e733Gy)VdyYr{1~Z>{ezTo+ejJTnkyr^
zP6>-;FX72?B_xL{Nf$r>7b*%E5d4EE<sHpetfC*CUyzgN8|y7)w5)d-WjuUHe-1sQ
zeL)YXQ8K>WyzWy)_C2~Dc#TdDze+#WU!jq?m&xi$5p8NyMCs9&$V2W?MT&)VV_^ZM
znHJF4NBLxaB%k7T<kO$E`Sf&lK6Q)9r)Tc@G&ne)eq79_tn31s_r8#Zj4C2k>0o(}
zx=Q=+7L!5Rb!z?U26b`0MbG^1lH!?rWVPr%jU4ch?gp08QR7E6UN*-|4m~CR5zlFZ
z{tJ3kTuB*;FX?j3YYO&%Lu1>$qwf!E=*XEGI-L1I^x{W)nDUiM3njZUu!;1vmGG-`
zGuX2VGVN61D;&j@*IL45jt0JJx5l^)T2S~Yi2}Lxsm<5LAkim*Y0|r2Zh%cOhWPi_
z5b;SyXm}@GYa3&v_!#4ol`-bMHNu`%M&c0|Vew%@%x`6cvTg?WI9msHs&e~TtqDcp
zCpS$}L-W6~XL`69BG)S6R=NVlU;9g8M}N_{h6Z}RzMgi@_)e<(zS4k*FBG)*GmWnM
zOk3N1rI@wfXj$}knx6NK?wEa{@?-DGu(pb(=Rc;2c@OD#<$W?cc%LT!Eup)T)6Up^
zj|OhKN4Nh7XXsKHEmtTbg*T<tdH5q*HvB2Ix%r%WJN_e$iA_`+sEIa;;)x6puJUFz
zn99ce$4c2ro81g;<CT$|paeG)CA3NvP2E=sNy*YtPf<Wsd*L?C|4Vx}{Gs~hKPld~
zk=%|nklxKk(#~$A4LUz*b;>V#RsDxbBtL!6;~)Jx@sG@{TO;X=HO?s8pu2R}LW))3
ztgnhiKUAS|Kpj1|v_O?nb9jDr#%qNZ;2ha68KsIO12tG5Rl~4Nvd{KR1MaaVNEu{_
z++Yj*jj@zXH!ECSVvV~hHi#Gfky{{lq6@a7C2Vm+`Y+qITVul`D`@q%M&}LIxFLRz
zyRIEhiN0|Wee=UvnAIH}fDiUC39^@rxhqy1v`0g^+_mO8$TpaGL4E9GYurvWs~zs0
zmiO<XZ?5QBVW__)yo)RlXlsFljpkVEZi~B~R`_y7y89#KHWgq7|86GYyBK5QbR*2U
zXNXBX4bUr3SXJ`6YHg#B&0Te{{IU*=t&NaA#S6z)c_Apt3w^S^v23I_KArc5$u(b8
zh)3gn%pb9Giz-^-k11jPl6Cb*Gj)HwZSRlH8UFatDgayMM%B$A0F7n=2y~TO)r<h>
z&XLT`;s6}4a>v`29@ri0j$v2app);6nOB9!{KgrF#J}0QUNp}oFZ{agCA!`V=LULV
z>x}krye{`YR|kY>*&{0523aAy>A~g<n(#fHj&De(me+UDDF0oQ({d->Qr;mvhwU`-
z<Te`Xznv}@ZKs`;+lhZ{r+><+q|`c<7CWVqZ(SZGrk$tUs~0I~L_T@;lz%TY^2s{t
zIw^}6IobTBbO&Bi);#%teQpJH81sm7yO+_O`2SIK)?rchPZy`VQvpE~#lQv?&WvD-
z@!0*d8x*@x?3NG}0a39FyMwevQd+^HM5No|o%fHu)<mw|+r7_s=A6&TgwnxG6#u=A
zE`P2daoSbbWnRTX^J`FO(}<Z@f5H8vGDY=Qp^bM{==4ZoVTq1@t*lFjX6jQ;nE{pg
z8qp1H*#}B?ZLfuyaL2^+)W(c{Of{o|a5KsftrNY>oHj{#Pra)J^)s-b=l`{)^@g^j
zA0ql@bsO3vOsO50+LOL_C$jE6oyLxyMZsg{(&8zL=<_yDy4|#tmTIk_0|Qr+Q_*TV
z+PsFw{##3i{npbt+l@4{^JeKH-bPP#?D_1AZhU-QcV3=yj8qUT?5SYtR^Ep%I``!(
ztDHG@yBp6gbmQ73{aJhX0CspWkRNp(BL3Z>lCc@aZ3nw^n=p6&{&pk>=K1p9zX8nI
zhgi!okX<GQ^0cX<b%Fx<)(z1%QGvYiN+26chd^zo050+g;$iVYd}iA*esDXO$6N~G
zA$6hrSuc#g#GT|*M^CU>K)5hq!r7(MNv=6{n)PbW@-eG(>~jAs|J-?2<|${{HUBIx
zG(XSn`2sIDyu`mpTxCC<>)f&E1~*6C<a4&SxUJhQwpO{tTQ)`T9D}R;-?d9Tc<4Fa
zV1Jso@1M${yT<USDV^E;_aW~1*pEkU@aJIZ)<~RjfPdHQ;iZ3e^3U!2`NfTsysYFT
zfBtZSjZcU1o5*09Z5`nGN45*Ua6A8;vzdJkdh?6WjT~dQfk*hRW!>+qIO4?$eyp^d
zdtUb99SvR_G1rS{w_3(orOP;J?=t?pd?_1vEMkpsbGi4_nfy3vI`=p_oV9-s;+-|F
z9C+A;fB5v`!jw+@@TDzhJ#We5t{d||1$_>=sx3M}leewX;AOfRys3=_k6xp}Mb9-j
z_<;tWxS_${Q#5$aQ4O}3sL8t>wfWy`9bVE?j|YF!XVY(n{J4_|e?DTyHN`FXu%4y#
zoV603unljsvE{;UZMpsEcKmXyJ?p&e$ksB$c@!)>oHjkU_-QZJo-91$!oGZa3V7c+
z=Kno)<qskKxvK9#);T$dcgGLrS#O8%sk1WAnK_IXE4i~mpm?S=hqJ@t;T(2lICmD`
z`|TNJ=qz&^d!;grQu&0q9(lNxnu+q)8JPVm9o7@e;k#UPj&xnmkXg=u<IC~8p$s#3
zmO<~yPkbF%3gh?^7%%*eJojRlo-RU{v?5Hm`3}d#diYm2z~NaFlmmaGV%J}6ZP^T!
zm}Z1)Dbal40d^2gQ!q!g&3I+%K0ujZu1s#fmFP*a5*ef_k=t3B;XsKZ?<&$gGwHot
z@f%N;f5D1{*%+0a4OjDA?C$abod##1U1~a#ggrBb)A06(?57;wi+3gkZklhfX30w&
zb`f@8;&VLi`W&mbKSPh?IE2go>5gcdeswXpye9@8PBCz3jz&;kG^WXCmE(<Q>F$ii
zroGXaBR_Vqk3qjbF;KLP!@3R6@axlaoa-$!p~<hH{_hQ%`^LlM;5#TqB%;^+WQ>vf
zbhdO;sn(=pL`o)1&u62B=LgtHX8c`79_)90MtA40@cr`*<CBGp6J7*gyYJ}r{5zgz
zmEh|!@#E-Lq5ERVS_?bI;kj^%>Xc|pg$nIT5mt1#2E|O*qW6kAWOqoHtkm@B<var#
zcEylFGL2|ZsWD9|lJoQ{Q~I+{ytAFn>Bu#6nxoNz{I~_R86h2=Jz7v<g*i2d1{(Xv
zjE=4{lTINMx~gVOO2SFD?V(SabvpDRREz$|&PR+>l=?{aj5ic%`N?KX=<^3lJ2pXg
zMLo*I^ZoBz71kJ5;*<1zSt^&qae6sQepcYz(<;<vS7W_bHB4mR`N>$gKk;8M=~OnN
zL((ueEESE7DUzK^!H1*C_)jMpmrIh6%E^eeNWs>IG+dvS0o~5o*f;tk?4RXfn_N>~
zm48qY)r2==Yq8T-bk(Rbl-reH4;NzD<_hST*5U5Of9UvGzSncj&>5#dv)3w6oo_Q{
z*8M@ki6%UH(1=lsB;R$b7S0!>Bcp3I6e?=b%DfSo&zs=B@elr){)2PRf7t5(58Xrl
zBJ4vQa$-smA6$tWZZ$9(U4@gzHAq=qhufPQFhjV`GY6~C9eJO6d8<=Z3k^Cc^PAOe
zwJ0r7hg9C{lK(9Wa&fRE$A7J)_d;@RJ8UUg?l-R++Q{oHJz~Cgbo`_pr9QKzIp=LD
zeva&lB=7IBwl!TWw;_X(He@}=hF-_W^FW!oyl+iO%WP@pA-P}uZA+@2c2qFGGj)&a
zM8V_jY5hCN3fZ<J^>jO0c-W3EEwrQGaki8v+KP%>k&^72yxL11O;_~IC`&pk_dtt8
z;S|qnMa|TThPJY#Sxy$DJfkIfK9=r{kLILOV@8$AX5^e`L<u&=)HTGICbVryRhMM{
zY1oro_x2*=21iO5>qMtdI?<Ln(%%~BNJj4+sZWg~ee!pr7|}FAt(-~ipLmaroGCKQ
znP!W==_C54NIW=mMBlV)Cw?388fQDW(2?afG(@;OeMehUdaVUTAF`wm{cUMfqb*gx
z=|ZK$Men5dqLe3&6c_JEW|tkQX=pFe2kofMj&|g{@-$ZYoq<M&Gq5o_gCnAC)~B9^
z(dyIKmvsVLkDtIm$1qsgAI0OQBZ#XA#ObLaXmSh2wCv+3EIW?RjS_BG3`3Ms82q|~
zLF>dBnBR$p)5sW<U%rfu|6PLJwewhh>MVXgIRkf#vnV)s4j-K_;#%|-*k)WuALUzE
zsDB&#4_}4lzAGq6xr%C~2>4Eoz+%<w_&f3!?n+0_HBzCu*Hy?|^o?mB4f^7&Nfpo{
zBk|E^zSgFK24i|zXF_9!NN(+kDOImFB^Dm%@=0dYwA7p;MBjLbzWKG>oYHT!Bt0_=
znq=LYTo%~M{;&-_mU-Zq-)%^JjqEWc&+fIiJE^ajOJ+kB(x8YXbWm7>*FLYHhrd?Q
z#`4va5wM07T-MSmtf%`uHqsB{&Gfg|b~+Ngo3=FUr*G^}F>?-6;<q4~D~8f))f03k
z{2OKc?90=~JM)Z_Zah87jWwqB=PUmW;8U#z@uh&lT(f8>@9pQ#`YlFqTJ}h;)tJur
zd;)l4!Xd6Q4dfwz5ApxD9^!2u0(kqeKz@EOkQaFdayRJ?=%^IHtJVkcn#Vyrvu!Yc
z4GrewvqM;CMJR9ZJI>eR!??6n7{AmH<tMJEWL9v3t&fCrw;$oWHv9~?+H;0=$DH9~
z4re&x@ENX>Jo)IL^L*_61y(zMg|FL6E-LZ{KVB_6s(hK@7~bOf-)^#rMFb!Fe1#t^
zzQpfOtm4A2OIZ7H2X=_<z>{)1@pU!f{?0!rdccp{8TzqN!XDnec^4~O*ug&CcJjf|
z+qh!?R(2k;hqYGhWrghh{2*)_XBKQ@&rRF-p4TRR5x0@w=5F9F^VV~f@mj8nSjCCE
zR&WoOW!zH9ll}U6vZvxw>4#X#@uG3+A1&qLHlAFdw1_9)p38xu(vk6SGB*onu_9|Q
z7ms%3{ac(lC!{BT5PgxQY0II5TXMCcF)NMK<JP~l_;|GjuPRmNGw0QLZl*e;yN2|*
zYw&~>8hmP|2KSk(!8S`YII+DZt6kKRo>CpoJ*~@kM(VQx8}hsb#==`P<&EFWxodMv
zj!?7W>1x(|Ttj>~CT)17gB|-%XwOal9XK|x6X&>fVT1eKczds&9CO=|yR_}Y>YJST
z#dUDaZDy10t~|6we_p#`0H<9T$axnBaoLW+9OWk3xNrz#)lj}!Je2zk9L7o8hlvk)
z7{BjTh6dwYEP3$}Te^J2N0l7(eUpjUs0=ulr6W8)4a0L&vHx8%OyZL;{$>&k9Fwsi
zKN$uY*=Q86O+U2~ScVp(gI^IABow0Kqym{=)Z?8_1MD;!5k0XU*0bt`^;?g<-~ZvW
zc{8-dFFVRniH692<BDjUL1UGvo0~HE{E?0a;S=}XE*+CHn^j#c-2b*pwC}Q_bPp@i
z&PBg*!Q~6?56VW${w!oWWn%xV4D6Ybfi->7G4??kTDzp-(w_IQdXa*Eeks^5I!4jv
z8NyAT<5m4L<nMllLDg|E85xJ=;j#FyN#->RVxV6djlO=-a2+0vG^=Rw9!FtieiRO;
zL}5!>6n3aaV~<ZX)(?!q>rt_Y-5rO-q-Q7)hEA!;OF%N`;f=3gsP$I5#S<_<I|=^^
zlM((h70s8@@N{nmde6#2LB|{v7Jk6Spj>oy{e;TW&**XPD>^R7M;F^dI9x2kwCBax
zK2l~ldOu+=KAfJvYH_stFI<+_&ftzBjk~B!cebcf)&R*r*JzNUzm~96b!hn(;m4%u
zk-B8>a~usxv%e9A_A(}IIp3{#X+m3On^J0yDRu5<Cg)i*a$963IT$m#-)u^weNAbd
z_(IPM=f+6Gm<H$>(TZw4nz}%jV&-Yno<5qC-mFH)Q&mXqsbtz-DA0)Se=$Gk7nTGz
z;B`(ddU#Z0o?9jM%rD2^nxD8VY-!saKXKqpIb3Y2aBqGt$R!gIt1@u<Q96zf&BGg+
z%^PQ?VxXz?dyPoO!{g!y7v8kt+(evip9txS!_=oqm~c21TkoeMe}5*f2|r|6yL{nE
z3oqc=Z%iE9h;U&QXdbIT+lW$F`xYZkyxZ?|DiG#T33KPa(0TU{RZV}fx3(GQi$xDz
zZo=m!O$Z2Y#Ol-rtgWxZhOad!$r1m2Mir_ptMTVm4Z?>@_DfH`)`NTunO%gzs>KN0
zSd5BS#jrOm!McnxxSXuO;jBvRy<d$3PId4d(SWhDl_;sJ3a$53qp{P}silzyxlhz0
z%X8Y|pVgs3;{P;-g)nN&DeRRM*{(B|T(=Q@KdUd99s{x*VJGuVJG$7r4gEN7OJ(BE
z+57(-!ym0>Uu8osZqi}V&xVr36R!8vhA#J)9al<g8YuTHyJc-@cT*eE8)_%4U)ftk
zbfT4Q?P=l->Ge8lM>Zqv=uLvK08ZP|cJZ2<RkRXDlK6)0T2aw`E9xa&qPDB8sIyN?
zY8)p!4P|TUB;Kt(K9&@0WkKEgwxrlIEogDNIcb)d(Vbc&+I7!}HkTUFxn0JRe{V@K
z$E<0y=$n%prKhvNkt*1UN`;O0V1gsnopU7P^N!Rg`bKTM6QwHlCEdcllwI1FRF(VC
zw%A6reAR@NCX&b7{s;RD|02I{GbWYVP`r3#8!D|yXP_kw>1IimODxDx_GU3XTTw->
z^hYdiLtFcb_a?I!1sX^ff^^9r^y)=k4zlkXwhsn<x54U%4_v&qV9klG7`Jmbx=lHa
zFDp*t`25p|et!awkDWk$k1%v<bre4<k09_=0B&FON8})X9KLxBon%k7T{KSny5rDp
zI*yH2VK{a>49O*-GJ_07+^JAZwhP6rf5D=!Ltt||1T!;3;qDWLm2J+0cVEQgEtj$P
z*cHT8UqMgrE0{O+3Oanbiigctg)bI?ePt2I{d8S2tiLh0O3rmbDs)Qp4fhm=YAa3J
zq^w0&Nm?{j*x>%YTJ*z1n=UOdp--BobZ?j`>6x3+6E`y|6@7D1^v(Z7-#lMoPCbHJ
zQel?lD?ixKfo`@mVR{?d;ATtzd)tOG#B)_@+Lc;}J{%o2mmG{2(yfBUWNo;dCQ3g@
zJI7U|(PB0E&tF3?tk%-wU+d^vhmAC_+a~eo`B2I3g|xwVF=fB>pwjbRbf;-GO=y&E
z(Y?3m!tfZ1-;_o_1{c!4$@SFrv@3_4aphvw0sKR65P#h=m}g8K$_cLSd{=J-r~Dkr
zrz1x5Kg$4KwCoUfzIljuFAiW6=``Q)TxK;{0jzxD5I1)X<iQ5wM?T}vCHMSz#n~WM
zH4Em)9^%8%4&m6M5H7Jl&SN))a+6mm+a`x_^oL+RSRBSrx1Hi1L&JHs^=VF?ahlt-
zJk5tP!a4HDaX$L&D7V}a#JzL2bEg@T*fDV=@44Z{b^{|gw0i{SjF`$7_e|jzkB!;<
zo-x1sYr*-m+wr8)_PoQpJs<aL&+hX(apmNdtY@;5&!1h!cTJY@m*rl3VDEB1xOOF9
zt6Ie$Tde2KlecoQIK8`{*~WUpohh^0#>?F|v&q0s96HCF%S<-1x};NelGbtT={2l8
zbQRxJ_hPML5B~ScgU2*`@PbXAyu@xPTlg&HK%1r9Fw28)hA(92iF4U)$_)11J((>Z
zj^(Q*;_r?h%Gdw)<=Lh^x!WD#?X_yn8#cFKTP-8rHeOdEBwCzgtHEbo)%l5?I_K|D
z=hKzy+`5O%ZX7iDq^1Ucu2bhbEj0L@%yCLvYVrNO+B`&6mv3Fy;~stn9Ch4?onlQm
z?VA}F*0kV%brxJ-WyL=$tU0`{HOHv6<?MFtSYe_)ZwTth5k;MO!r-oK6VaUu^?LEs
z1x~#Fd>{TE>&&<BfR)xTw^MTEpc!r)vcEsiKRAGYEgr~4_Ji0tVGvhO9?bt_uJgL>
z5Z0bOgkyZfliWQQ?Zb1x=W?*#CkF=hIp~p?iHiFfc#@NjHHm3>)szCA|0QCfbs`>&
zlY7pWBrMKO#vth=nD*@h{z>1!t+z$k_q-6D-W5Q5V?IKA)?<lQ1G+2`ujz<-`0c5H
zrFiX9@0MegWQgytYR2E}W;~g$NPk8P_##(6zwZ1-NPH6t)->YR!E$)#mST}}A+*J#
z5vTM8hZg0*(&Hog`{w|b+0qG~g^LE6D3saFk^{1rYEHvX>Gj^9_#QXDrNC77Ljz=|
z)TebUG^WQvztuC0?D`Co^5YP^Fb+=fvG^-HsYT0Uuq7oLf8C?;@pBZq1x8`>)F}A%
zj6#TclrW4%+n7bcYIGEmUqoT6Xq-1IV~`RMi_q9ObonWBoI2T6WxPOu<118meuIN|
z;t_W@0SAvIA!kDhyl1>eoLf3ltTM5$Gz&&oaxi+*M}#%y;>3wh2pReX4}O2et84ji
zb}dBAjUu@0{0_sxr5K}80W0qste4$M$2Wh_YJ&m|m3_;_?ke=OM3t`itCPRvo+pH8
zk^66L(jBTx(>CeR+hBdV8D>DLdkksnL?imBZcMAgjOnMY33Zk%{{QxxP`xmsw=OiH
zhvp`f6lzS3pN+`%ks)=@&=cO7E_GGXCFN_{H0XmS4bGA*(sS|Y-czQ(4-~2Xax-k~
z{$OrFBQk!}A<3x*?$0XGU3Qwg&i%wwvr;rmkErL+A9&_f291-s=-DR=$&`t9UKxnI
zCSIFo;+4`!gYKbZWY0~)gA<9E|04mfDFHc`-=Xu!cX+no9hCPc;$fmN6o;kYL_!`$
z1bx9wYw1JXAf2W6zvK1462uo*pxuOWcp3e`HMtgBXyqefQ#rPrsKDDtzj0V}^S{=A
z&{}3a>%y9_VNnx)xin#yPXqMI>(R`0*rr>9Mx$y7(HQ3cRpI6DYCx|L<74viXL3GH
zxD?`yVi7LvF2Z)jVnp}+j@6Mrg&SE8^DdRp>McGa>7H$HYe2k;5@|ZCkoIE9yydFW
zZWT3JSRnk%ZF<rNsZVPTi<dsim@GwS`a6j(C^8~@=`bm2)~9PO24ug(fOJ*cQkjDd
zReArvzgK*r%B^XVvke_;Z9{iWZK$%d4Rw?5_zlu&(Wq@h!zN0Pe1i=c>f2I_+%^=q
zxg*Vq?L@<!I#HdbJ?+@ro@PV~ms;jRR}ZwMg*_$rEu9|6eOu8UC&@ucM@k#<WZ6%$
zls#KZ>DM-)hK+@&vHXlb`*IN@NuqWaEa;S8OL7ZtL0K=%scMZedA*nKe~%Fzk1(QD
z*)4>1)shrqtSMNyFl(OlqQ#buBzYRLa*~<Ne~wfu`o=)?&9XX2`fsC?T&sO)G4-V-
zhwI=JBsy8x0>kdtBBOI10>p<h*`X2f+m-3;BTX8&P>cEtQ)`T_FhhlJIi%W*)Yi46
z#lJ0Rsih4)5{}Q5)NZt3Mt4eD+ns!dw4$B!+LC|5UR)cq3Fe(Q!>NxCQfqwB=js-u
zZQhEx5>!7jVmn%2*#|d!UknWO#e=T?GK=#^)nR|RvjiZ2x<Bgd{jow_G)}o6LSurU
zrEm<bFCN3fd%+lZI1K;G4@KV#p=jC~ipmWkIQ}9SuYUxi_x=!Exh2|Xei#PSoWo78
zi|9D(GScQ>!M5|Faonz8Oxr8y|M@DW{k;mMK@k}C{TgcL-azf3o7gWIzOD)iG<BFN
zO}EmZ6Uv%2QT|WXx@ghaNor&#yI7+gvIn)$pp+%TNA(mRl<cMp<Q!uzof!*6-%JsG
zb0oh7&EMXV<~s;$D%YBZ$F-)dk=9fw*OqBW8~W8*JUt!T(@xPhcBOM^{Pg)WZtG%7
zD)XZGUMuLd^(s>Qv`TV+tLg2`HMHl!e2UPVPv>7up`66&)JJJ1O;Mdqxm)K_%eD(>
z+j|ch_P;HZwfz<?E4@ehuU}JQ(+9fcSx+xinn)+KH;;PLpVJiwaoe?n`9{2WLpHgy
z*}&m^&u|1+><}+&;TTrm=g)cyGP4;Y`bK;#*CM13OEk`kf&gyPImGTa{kdwSbgPf?
z<E~D=9M*J<FFQzA)wN)DnH<9T8ljvuFO>79hw@pE5MHq_m`}TevH7@@Tu^a>m4vC2
zH7<-F&N<FU#)h){yd(TT{iw`qcJZNy9&8yrj=wH+=e+a8P0n4pxkp<*lx@j@#iqP)
ztucR{Ys`i|raZK~1+NIRVEbj3+~Z|ij*Il<kl)MrvhoJ5(^<yX<}YQ}cFVYOm=}+_
zzJeD<ui_50mXBG8rzUY7Pjy|-cBeM+=I~8y-DeYDZt2aF`)%Z3dp5Asz4d%|?mFoQ
zTf=R1m-Ea2q_<*<2k+eM!4`Ktcx<T$+dYt3&M!|cGVtIZ-4^p<P04<n&Eaif|KpQ`
zr?A?GF>JAZ6z|>a&Zh?rmaKh8>7Z!Gwd-xzYn3^7ermvH%yguWUW0>dWOlPmji>sl
zu}&9t)_J1NMt{{gu3Vk{Z>zJyHg(>5Rh=Ks(BPK)G`Vw;77w4J!#|bvcz(J*uYYIA
z9kPu%rpT1{mzwj&qL$n-&yr&@TCq-s4bT5%%Z9aW`JTDVaNIj^!TwI{nC`$9ZM*Rf
zuO6I#trxFNcVeg9K72pYnIl(#@7EE(9MO+G*1B@)Iye3}p+8%g3=p5w0G>Kvpyc!h
z@*B-TY&~d@bY~3W_=h=?tIo#Yml=px&OrXhbZonwj`eHP(f)4|3iA_TTq3>OQ3<eV
zPQa^vi5Q!jgwfK))aF<U?yUR(bzyHVF)WgvrUH1M%ZJLrZ+O+Q9zA|b_vZO}>~R%O
z@U{vBwXDF@2-$BbmSel*15P-W;=->#_+$MK^9!UCqU&#1*i~a(QU%sk{zQua@h9sR
z!hG)+98Uj;!0b%CUz(0(a$kICmySVJ={T&Hf!dgKOj@3fe~Rg__eq0M=zB!U{c%dW
zXdK%ejkq&0aF;pFz~ONST^9!hc`o@9E1zevIDR|^=e|YbXYXjNITr=zE>U=t7Kw(#
zktkXaiDB-M=;j&;?Xi)l@Qsvi&q(N8i^40J^Nc(egK@&d`IQ=n%&*U2RrVZ5$G?O}
zS789oeT(z6-=TC!A`W#<Mt9>>bol)qe{$2Y^HwI7F3*NZ%MUQV_YpTn<e|lnPdIz{
z3l{bJhTJdt__n7IP6vz8()|alQ_J99T7~oCJ-X^AU8cVOu&k{jg%&B%ilZvjrK=i^
zzpGBF!gulCtVPRXgl$(U{Rhpu<oH{Ug7Wmq?t%ez9%D$4at$eZm=V!oBbxlsh(<jy
zqE>!Jbh@t*&AMYqBYzlB(G7iS3e+WcOC8dYe$;^)l4YAJz7lUW%HN|x505C(Tj^tc
z_4F^I#{EK*u=~38sl}|_RdCX)5Ejr+yh`~21?v)oeiF~Le=!dIC_(?aEF84Wk~6UQ
z!57JV&^H~UDw5D!F98F4B%(4b0kJ>cA=LRDw(pC_(89N{o%I&ml;2|7r?*)7>m5dk
z4q9+L58Y;e!AF~XwEIzroIl0z*j9qY{Uyt{s0=34qfsz83fpzQqMZ~>->)u*PMcr2
zcHtLR$#dKFO?cN$v{+CLIuzGp=!JUZb+1S7lv=dxEc2IPRR~xs+{i^gF(|DNm75A+
zJ~1ELlD<J*OZvn7zv6#?zGB4QZ^&sXf_ML)P&6yYxUCgf+^rbD50|4wzb2d<uSgk(
zl<1zm%qB;QkA9>Yv9&I3Td7CeuIN+4U_)BbTe?Yhn9{RALn`=cKpMhq%xu)B*L@6V
z={5r@_+&s=+P9!cS8ICp#7calHZ)&!L+)p5`cq&{7vvt5mS9brg<IIHB>nM2Y-o9b
z4V7lKrt$~E@tfa~jJ9;7K9V^vl)TUY;p3ED70>cJJF@?5MgdP+P;2?R5uyPUyU1>;
zjiKzM4C%)M19D5#r^Cg1bY-;ggvv57Y*H>>AI?SKn@@23(L%0+7WAQ;3B8+XEWUFi
z+9=tzRikvtC0vV+wiZUwD{Jbww<mSE(u?jJI?}h6qHWqaQn9lmd9WjW_I0F1!UWpp
z)0ZN@_MvyO8ys9%gQ#uN-zz@!9d~PSq)R=<cr@XCNE1}jn^2;uL9<-6>F{lBvJDNw
zpMF6wEIW!F^+(aC-68l6K7@7?gYb285cV7lN9V)gh{)NC?5bT*3Gzmh-6ptNY({zq
z9~5T!pu@^7=&809-~M>vyuorLob(j-wkM{|TZ;7GOOgC%9fs%lAzU=gDhq$;sEFpN
z^n+%xA7YyQaM0Qx&-?jfn`ob9v;48rGYE~|p$MNBiVZ75(7ZSp!B>OvXix~A9Sp^o
zOJSJ&_$2NQ3P;|uaIDR^h>E#a;G}s4x+Rwp({vf{cU-{`&#U;~n(I(mbp!RYZ(^n0
zFZ6l*7eh@I>9P2Ye}qbY(o>UyYZPg5A0@i+ONp+<iBBg?g__Q2(U;-+R3?1IYvWAm
zqqZ4U|1+gkl5?~D*n)aZF{NG4#eXNy7rTq^?zR=p+iOFruZ0ajZRvt~dphFLot9M3
zrL8odzUMBasA-Go@U!LA=lDwckiL?dLKl+y;`vluJCEEpOs0#~Q>p3qbb6aIgHC-D
zZ}Nh<^w??v^%}gG+PCl`d!3z>*nE$UX1}Hi%a1f{PCf0I(?sbOy}7$ZA0B2ph;OzY
z#)+xJc(3T2M}0-#sE**kBH`-Pj^X_Qt9V4QKS%Zs;M=PMc*)TK-hU&2H6>$}bw7Y7
z><wT8XFtAo)t4uAImo+*9^*E-$0QFF%;w3#TqN9?1EWJ(>rODo*B;}N;6u{4eVq4e
zo#6E?POyU6aX#sLj91@0%FB`u^Ra7zZ1s5;SG-%nXNP<8%1z_gUvfLm{fH0T?7|*r
z+Xx@VlGU0`*rd>iCq^6ba90zyI$_4U&00v8Knp&7v^BR{?8z<sJ@`SyS}t*2${rq`
zd^OpVV`G<cbE_4+YUwKWt6jx!@2=$=W7lxR^L4ysz<PcV>CH>Vd2?gl27YL}fnOe4
z$5mt2bDiEgP9L^{oqRp{#zzl69K3{mdU&wu5)V%6>%qkfJo#gwC-2Q%!aqtDNe}CM
z?(}^&FIp|VP=-@jQDY2yeiMewcz0H5C3~u&ZrrKBoU=O^^TISk_Ufd?(OFvDJW8EU
zmZ@^L)vA0ZSCtKN)wtzhb=JyJ=a!;#f+wqU+eS5BZ>!F3Y3giTrolU1WrlNKoBNH`
z<yI#8tkr11ITc2H>!%5~E;Qp8SuOb2YYQIo#ESjySo8I}t@%_;8xGB~<DB2^*;2a2
z9p-lCr(s=KIGDV+v4^}rj;z}3#7OSLYu7n*ziJm=F@U&qG4r3r{rK1rSLw5NlU{f?
zKGLH<Upn2N?d$q;UY7xU(tQAr42wf&*)vr=&yZbk2LAhyj(%b3$PiClul}+>xRZvA
zn+Z5@BmohL3Gi}AM0?p$eczLe$J0}=>%n`R8I%LFgnVotRDi`hzTxuP&(O}y!~RM2
z2)|d49p3fGSt!0!qYBIskFs@H8A6P{V@NOYN5p=`qwLQ}v;2hirg=ztn~QzBb8*Be
z7i;rALS@HC<aYdsVIM!>+k_0f>6wlwg>-!SnTFbmG>jgaj;E*6;Wr^22J-8MqHj#1
zBEbcbnCTIPFsEn?w~WE4+89I}ip8a5ndL;r;`%(%JMUw#w`B}gjgQ8~kSGjmh=kGF
zNbEI?#G^M)(e>C<>|6g7nO;xfyYngji+u{0u95IciNv_GQTP%PjhG8D`1T|gNwNcb
zTm1|_roVup^((xwd4m|kcr0jmhtqk9NO+Tsdk<1k9G(WBjT!jioP~Rzve9+<2h3>t
zhyw@m@S($Jn7#jkFY+4dk19Z(R}t>D6K<i;Pdpf1iDUn(#i}`t@OSzH<+Nr5o>rt<
z@hqK?eNlJ0cUIn3r*=x3athX>&{;C8_SB(6l6mUyB7NV#^l11ged=LnK<y7mmOM$=
z(dE*$P+~wePYvkDTm!Ps(x-l!`t&DDm-ctmq4jUH=vka5t&%L##BXXeqqQoPxhvD!
z6^eAzO}a8({6_lPMm(Bchobe>c=NUb5TE)uwNivhcc`5~F=~V<Jz;zSLht0l>4(hQ
zT4v$e$V@bi&p@T@IY*v+gUz$!(a`Q4v@_!IQv6GQrL#RW^$l#S-e8u;Yb*+U1s$6g
zm^Sk*hU6x}R%T)!mA}AI^i7v-g-}>m483m!D36RnUY~ndt$H7}yYC_M<YPE3%t57i
z$nI*?V4`p<8@(H`M0!OGURR^kp&IVO;8uKBgHdy8pgXu4wsWg+*IV{$K@~XbUWzkg
z3URnGA8~Qtkh0}7roH(Ly^t@MrS}yHZ9d`OxUcYj{s}vdN>)a8W=~toZt;960%YzG
zswbW^WkoU(zx*>771HRTMoI>{)Mc_B_1dOSM~e-}<*5<fa?}&nojw)%8j$}PJ$hZF
zPhQf!;VJrN&qo7t9&bqII-)z|&!}G3r7$A{x-U%HPPe2JO1v=_p9qI>gf;DZ)QVI!
zrRV*HbX{I*O*ed6lfj^lRNb$mc-cCTWl4M5F~N>Pl`N@^t^v(N39g^}fx~aL>E<)J
zRvzop%PV@MFZ|8Jv-C-SvOWd6=*yl=k0N(wz<6dF5)XdD+^gpF=BGN1{HjhfiquJO
zg3SFU>CoT=9da3?O11-J#<xNAw#)_UKReLN@Lp6hvKLjV^`c9{*b9m3PQ~pzQrQF9
zDU}-1nl-|I>|G1JPu2JvTZN=Y)%f02gEx_NSRnnq)v`m(I^KlHG~s$`i-)PFHf4Kj
z)6^qDnA;@?c1cHJqJ9Y16%S$U)XiwTy9s;SZbEjbH_|k`vFFi7%*)w`ezxAw(AxxG
z&;S4Z&A8aj2e-Oy!Q+!#Q1IFdTGu@B;)o|2UwXp4_fiBmF2(a7>oDq?AJkv@A^(dX
z{(SPoiEKaI6wTAQ!ViD7{SnkzG>^MKrcU!m&MtrK3h>9H6Ngb`7liTZ!8kA_7~?8~
z5vg_@iXmaJFb>D9{^9tyG#sfn!*N^nBJvC`qvX<Mm=;~e>@`<Jw_SyX=$qO#H;{b(
zCej<C@uTrKW=!~pmWLGRt&<}8pHQS}Qx&P{ejctXe8EyV8>K22;D22TG0sz)+E-|k
zLyG8AV-xaSWJ2kchV)hT&a}Xk9HttR=WtWH{;dVQ9AibrBgEhSz?SaXwxtNw_SCq)
zJKcLTm#hZPr>ln+P^{@98nSUA&E2<vj=IdFyS?Yp$|LhA^XLS!FPlW4CrzU}Z=?gw
zX9l@noh7~Rb1A;f0;=q@n0B99Mhidhpbf+CQ_}a>^k>XRniEh%?m<n|ah3viKh>L^
zhY#Wy+hKepWf&`OaOX9hhO?j22v*A<!EZ~)aD?Y7HhbsKUxka)b7lZLZwcUnkN~#4
z7r;|42Jo$20X!hqmunCEa<F)F4L1a__GZyB^MZI#NDyC_eboE1V4kfX%&sN@tg=#e
zTwg<Z->c*NtxXu8RSo4_$vTbSaD=}v2xR-)0bJ&=i)XxC!O?M^y!O&8ZazPnBhL-w
z&x<?q>%OhnQ##kH0*!fjgCQ?{ZOEU}jM!nb39H^SWy2X}tUuL^zXhA|$=i#hmueRK
zRV`vmc<|^29_&`+!B2*-<bTUn^0NJsgOW$5z}1|zW(}X7xQ@p+uH#lN>v_k%_1w*F
zJtt+a<*GF+`KQSW-t=}Uk9_9Ap1VBw(i>s+J*g#Kg*rO5c09jv@nn^mo*eXM2`^Z*
zh~q}j=YLCQ^Oor8?7wXyTe?badd^7xZ>+m?bq`{%ZEpO0n;DP4X~cgA>$AFx2KUQU
z<2gR6taez12aQzWp1P|1B20~Ax2p5~H|qRUbdK6TH6ArVjoWWg;}}Po^9&VVj`)!k
zl(o6xm=3QUrN>FF4fvv(5eL>7^Ow&u!+C4Y3+}Y!izhAlfBvnwaE}dt+GoplL2dbX
zL_1E7x95u$9l4*WgUsN%vca(KJZwl$e$l~^Lxd6Ov#SqVHS}eZ{w^Fl4>D^b?kM@H
zZhu&MQTy?ucCKu2#Ff(uUAavwH*Ve6jhETS;jZ{ddP%or$MvbGF%|6)l#WxPZ-$D#
zQ4xK!WTEsA%uT>O{{(C_PDI7_B)ExB;)r<)BC0ZwEcz%SEeDO}-{2Ga6)KBA;gWtX
zW_A65qy5Bdax5F^le5vjK)fg8vd~>M3vJRfak}?cT)p!Zim9KWtndj=jk3Rp%|*PA
zXq>LO7%Keo#?2q0-Tou)Wq&|n-wcdMO2eegG#HDY#p-DqcFF#%#r|}hAC``we`#3Z
z6$#^6k<fY=iR}(i2+fE>ad0%6C&XY~Neou^jfIVHY!+0-z<+iOejkrULtGTR>LSr{
za3nMzJ_QedivE)KeDdrGBF;QPS;!MCx$y+_;|W~GKgEU8r^vb)iC)1`xO6NUEzif`
z?o*iuNv7(>xaUY!cnO2Auh9SH8(h2;kL&WfrTQena&-z)7re(b(PRmlnQ*_H1@F!|
zICJ3x=5)wKLu4L~&Hs#xt-oTgQa*f~3Q_*87~00A$husPg{!O4(61g-OPbJW@n5W(
zsX%?}6zTH@Wl}3qp@u<fRCiFF#z$$8N46&I$d!H!$yWUh)1jsby3|HGt72t#p!8mk
z^bGY$rHej&YO7DiHG1^sm>#)k=#c`XuiZ+AuH4k3mUHEN=&C{e>eOh~YT2*EE7K-j
zC7L-xfhw)V$0EMl3jcbv*js~{H!3kxr5x#dO5xwLM6xBtI9gYLqYd9s_Ua2##^qv^
z;|I95%EF%iBn!1O1&&Ftp=$aXZ(qHE#l*L0W$_k92i_nv=e3-%U*p@@SBTvB5}Rkd
zfVS5&{0xajhjub^^Nhkm>HPYiiRhb=g}}Y9NKL(q*_ZFZZpvNk9eEdt_wHa!&~5my
zegOaIGRdNpV^mQ&j-*t<q^bg4mR2G!xk}i`)#y372F~BAF<1Q0JuX#ZT7=}iBs192
z^#>ND$^BwQK2E8B!|X|)@bKv;JiYxHCsp#G(jgB$<=*n?XC5v*&O^}VTpXWWfU_1g
z$WEw5T8}@tCA}h-7bwwR@flm`tI-)7b-K|*k50OazVSDpt*u3Wie~wGN{>1$(W9%g
z^+@}jKCP1Z&5gAN)NsduF54N>q2C51Ux03tE<MoHrMqF;Wd27}GFlq+%u|g9#;A}X
zD^qP3$%(xc-%n37vKNj~Nn7#a^z29$k2_F@e0x$lU4Vm+3*jvchtu`np+55mj)*_i
zEl#eH2cmDn^r%|0VKJg@)F$XtSJJ1w{q<;W=M)?nm55)h6zD^qI-ST?r%KsZ8p`tl
zdBpzEraQ^%6fr{jR#}xUk~$4us79W74&-0wK<j?>pc`*{Q0n3ybU>*y-FFeLwtp=`
zjmy!dRXHB(RAZs+Hs?xKF#lLJUi7R(y678w$9h~AHWyj_!Q}CBZLiUw>B8?GHC3DH
z#9OOt86@YdLs;DyfDzv}3pd>xW755G;-NQ;PI}|@L2q=qzEL#nM(k7fMqPt9u8!V>
z5vH55UDF4Xzxm+Cq%8>9=Y<#FJrL;XiB3;F@n6@aNPN8%-><HN-ldH=x!n($*Zpwd
zi669L{Gj&6599Ov@V(IwJ6iZ-0c4Id)*m)L{wVa7S<WSYl$8dew?Yth{yqk0k7HP$
z6pH3gr(kF(>>LWmFOP8ihzLhv@&&wjb_q38F3V@mWw?nCXO_iPB+j{x7t3#8e(p_N
zb-as;pE20l;~!?)D@b-pSWMFt>2}!%WGu`8Nj7tk{YRWVD4zSAPiVR8JK{bI2daxM
zDSGJB>751?bk>l52OCkFb$S%u(U@|DRpd9tgo@{xlBa143cY7bu@=Iz)@(;V<2usW
zeZ6V($obTF;{w_+eIX6CTSzxLEugcnMp5taV`%!#apW**g5=02Q9qYyv~10E%IiIY
zmOc5OFo$N+k#4hTEf&+-ou1@+V+Xa{exDXMyrv5sKho3vU&&0Pf$lge@ZWqT;jj(j
zL-~W*=JhakUggfM+YIN~w?^=g^P{-jVLN|6zn_z*_;PWuA4mW26VHerJ0J05F(tEM
zSO6av53*@rU+!UffZf~dXH{YS&4@h8T6IUIvnYt`>kjgS5eK<-{6YTx%a_9k2k^^b
zp&Wc6loc<BaBJ!C{Ic^1-y3(By-NewdbJ<V%iYP{es1Qr7G9k9XCBXy{!k;)H#hFc
zJ@%9(7v-7r!&F1=`_h0P9yZ`s?uKl=)`*8JHRc@(CY<hQ!XH9Sxqmpz{AC8uGo3BG
zy7?UXb1`pty_hc#S;^M(S8}}1O4ixCk`-h|wJLrU_w`@PNrTt%hOl+q!+afwj#$T|
zkFVsDypk7}Eoc3Z$z1)viJUgFp3JH%>8yJty}nXO0e)4~r=o&RPOYHv1{G9QR!(D;
zD(KvV3M%?rK~rYc(!mnu)vZSILD4tM&3cQM=NA>;Z>Hp*3hZjE%$K{WaoI~%cHN-N
z+7`+@phAhycTwTrwTk?vN`WurEAUi31>SgHjq7%(^6gkv>BLaux54Uc@kz2(?pmyq
zDKnhyx~w@wpW9m*^7&sz+~uPQdp|YfX(wccv(<tl7F+RuQ>?k&nAZGcOdGyF)s8)v
zwC9G+9XKtZ6T6;p;H$#KnYgz**G%cjy|o?L`M7YDl>4yr(7r64$ec0Th4%p5tr7ew
zggDEZdH(_CCm)%U4EwR2LqGoBE)I^aadJ+L!>M_37$|IBtLe{Sul53=2``a5^Bs0}
zPr&6R?=b1WJ4_gmh|NQikZ+KTj91ARJ3IxCTc=`^cy}t*Kj81R&oJnjhX)%!VyfB)
z>~hF~!|`l%8X!Dj#cb?~%|gxPOmtoJ9<BAIXIr`qRBcn`+?|7k@3SzwRSNunCnHiZ
z1*L5zA3Qe&T#|y0*HZzX_gJwd4R2bep?G1M%%GFtWB(Fu<$jm4S~QK!R^C}hq8~@X
zdRrux%M9oFiYTlwjmD0sXuKUAgO4|3kp3nHX}&QqZ;FOZ>u40pPU`W<NR0G-iqq9k
zu+HZRI@&*hR_SB7ync*fQIAoR`WV}_pTK6h{QkcuIQQTwY7RxBaaWXZc>aI46$AUM
zSUC|t!|9yoa1gICZG4TUNpE2#?BrvzfBD%h8Oq&K;bxhJyI<1LV|^wzmt?_yOpdUW
zKVbE!T+FY@L*T>D*mwCWrWfSnopf+sJt{e>hM&UosYI)`wFtV<fa$8g(NdUn))5Mn
zCI0IN`^6LUQH2Vmi{QF=j!#b1pzmWf=|XQUda9&NU+!sB*a#hRf2l(~<(}$4O_#Q>
z(j|q3y0osRE_F!Pp>Bh9Xv=18Qdp@)+q!BJRjZToD>eG(qe?m2l26^BMD?i()N1%Y
zJZ$)dolhI!9bSudk_9jMD1NElKN0Yt1ji<e*W5_DwX(jU>$Wdw+?j{k(V5t4E$p&o
z*)XCvI6L<h@>;w??($bCp7a{(*Ir|}*K3%GF5KDW6_(8xcA|8+k39Mu9#e%O9U2SY
zN1`RaKgHINNa%{@yqE9*_N906ZSP&&?R6KOo$kWb<&JQlZ=<R57Ic$u!o=wYVx;S*
z!>BSjdza$ejGu@}D~Cct1tw}&qT!@?iH&QZx3?OjCByFdy#jTK<;WiO1MT7pku|RX
ze|~&}X{WDnZq7yOzC0K#$c3t9F2>8_h;beYN^+5QU6@8=KFaw!LuNQN2(zul!dw4f
zF+f2&yA<i`I2H0TQYGsGHPYTJy%E~_^xi<&jLABb*G}FCbv>FsQHT62_2|2FmkgEp
zP3UR^io0Pz^Ch3(ztDiPEA*%~Rfin?bZEv>ZL-hMq%>7wS<hCZ!IG^BwN)lZMd2=l
zDw4r+1#&$42a1osp~^ZJi)6QJe=Y?VtWr=or4TPB7NhRscWkLnlB|6qR6ZvN15-Tl
zN$*haBzslq98uZ#4lC>4A>TO>TUN*8qW=r@(pI4HAJs`qv`wZw%H=URQ+_VJbMqQR
zzkO1n&!H+*5~xCom8#S&NS6lJI?$B1U4%2-h4kk+&@16_C#Kinq@!dr4McZJ#;(Ar
z0{+J;u~=A1`Y)g0f1+<zKdM6@>XDLLk6x1Z?>ADNo-NU!^9MC4r%_Yp4BE2i@DuJp
z04|042%~B<2A9Z;Ce0fo9(bcD%p1)Iyz%g;=#vK<vAJ?1w&!?bVxLXIliGx7_c!DE
zNgpf?TZSQXJ@D+d2X@(dVq%0RdUaolzf0GkY|1)RFZIQbA$};|=7&AOemHdA4+(eu
zFe%;-7KMKB`2U?%4}a)~?L*oIe>m><hxIvsw0szd`P+`-(|<uwu{;J(_YmyLJB4Gq
z;dt3Q9FG=<!_4HI^!Z*uH=Ro`(icBY&}Fo7x`I)^uE2q=!)5vnSc$&LXmuBQA0D81
z#1o`sJVT!oFHxlQ77K)zeo8$9Yi~-{;^zm<6p!M{3!ktcQ5fl5iW0fD24zc6ooHp1
z?S`bdz=&o~*P|^KMzltH{=y`8U~|=&PS<GBE8!aaIH5&rA8FChIw?EtW<^~ev?Uw!
z?$p<?H_>Zn@sGPw%7wu+Y^}TW&W<Ca`tg)KViK)ao+ccn>2x-AI_d2AA2q(5K}Q{C
zk@l-O^lh;xo&Ua?>;~;76~$<pUXw)odZ*Fv)1^dxD@o~9Gu?MpW?$)x9@Kv*U-&+Z
zcl935AKnb-wT?clJ!PBdkrn(@W*2Va)iu4mmEW80l5R#nzVGSFtN$Hj-_nEZcXB_E
zX}6DmChZeHaS#vq?<gDWJ<7YU9_8n+4)Wk`2YFf0L2e;BDNgzxG7Uobc47#(SCouZ
z=20%ac$nXK2}f|5KU+;d$d8hC@z4Sv9{FxLs|}ye?_;HJDQ7&#)%4-@8(Q+IzXsAt
zVZib8_1UGnK6kpL&)+r*`=yT|e|c!gw%SHK&clRFKY_oiPUl6_r?Q^u4C%9)%MWZ8
zaZ8oCeB|0pp1XUd%xUJcV^1$`TECdD_FKeH|69ba1xxsM%5whocNNbvnZh4Ol+&ew
z6?7)PijJMGqHji(bYVjU9dN85-D?%}U|c0lxl$o-Qw5!`DW{>y<#aQ@oKEzrp!PWx
zw6{+c9bZ{Z({ERip-VOGa<8R`$8|I>rGYxcHc@DwKQw!f3eOEvVCTz<>>aGgD<>#&
z;dTYCG*aa93I*wWQs9A=&9qftfs59t@RCv${v~>7;wLpOl?-|5eoZEQZMM3l!}W{w
zq(91lXDb@<p-f{Qf8CVhcboI1IWoiPZ^`XDwBi_38@{Jy%ge=UJY2mUFVM2*aP5v9
zuiBY4YaDobVpnb*(47M%Q)Tn87hf`W;<Xccv$a<rVcYcO<Ncg@XOlCBhPZIO6<B{S
zI3W}KLYr7H?R=)K>{VQZ<uW=BS_|SZ!S)%(PJ4!l#n0gz@&a8Jh{maSgNhdMSTX1w
zI<-l}G06q_<Oq-afaIUmQ!qFw1xiZq<qVqx6}Qj0Df^aoSsy?#IryNOgLq-HEpo}m
zh`(7d6@8<+GYj2XWWjA|Dh64nATCV)zFrD8K1sn;trWyoCc~vs_%W6#Xfr+q+PNvn
zxs-}?W8S0c?tA!te~)z8ksWA$h46>b`1Ugr3ztVC#4ZvW)gtlGC=!+fBeCSDXdU$^
z3^^PHn@-{hj)_LTqp)m*PXTAqBjM4|yBdYQ$0Bj#<WndmJV9U4HK|u0i<kH@Ug$kW
zVeuoRe|&^h<&W^d?lI2nc#PGWPf+|oK4*46MT-2pmCK``w<j9CpGYpWWgLvJN=N7H
z7cg>oh3zWBHZFaOZu#%v{WB5oM7!OpPn9l=G;|)7fjd_+F{m*M6Q<-KMrKt1KLhA6
z<P+T7zF_WF@pqUOB2D!>isiFP>2x`ww5#F2xDIEoG-62nZz#Y0hehudsI7Ed7Os;n
zs0bD6k)TSgv(zZzojMJ;BC{MXO;WVbq7zrO=$5rM1+USTA8S)(wDgeQ)E57!HodUW
zCL2F3s*Tqq;Tq7Y2z3gTPEUt%s-&;4LievK3D;JUT)Q<x_0MlOB}hl6^ejYP7C-F^
z@$X8<smC1Ym-t+aYwHW4D&F7ts4qCz_7mnk`GAM^85lG^8O_P@@O}Rp9s9h(%=0g0
zzxNW}>aVch<`sfcU&6%oCG2;<!2R3LG4Zvq7oWz#YGo|ey_NUshwKkeKY`!MM>rzd
zW`Om5^m%_*_5gQbwd)Ro{@g~pzqg>&^%nZL-9)X{4Rl+14b#qCLA3oBe3lLohngQS
z5U;a!moj+WtiV{&W~p1Nu`#9!?=MzjrELYAa?4OUt_11T1z0YA8;|whr04w$danA2
z$!@tYc=8dWI((Gb+6OG__7Sx@;t_u@Jy{2GFyM$}_{2-DXZsC@P5+_TRDl9JDAJJe
z!WpbnAwMTo(hx6`f2}69%h0BSdvs`Pj1IkTt3!Xq>(nYuIza<<MAvAO=Mn?yb26ag
zzhsX!RgaEp>rkPO4!vO=Dj6ppl*gLXzC@kg^j9Od=gPG8h7y^G?>k>S`&S0l!)tjy
zZg>2O(%2l-zRW~pUMiZ>lVKfIfalT$dq#X=3-2UB$ukina}$6G3D~;s9pbu(zKM&+
z!q9k3-Vl%c$?-5$j7QSC7pSgo#vH9?1iezHE(z*XCf8W&N9uG|U5(yfP@|%&D&*#;
zLV;c?v@KOtxW$^3RIf|VyXaB3o_f^(RV@}QtCh1uB_ey5qnB<urj-`sMSVH0t&!K?
zwF=Rvo?x2n_paQR{5{p<!-sl2Irj&PUDe4|^o@tiZ;JLvUa-|cob%WQ=l5GM`Gk*T
z{5QjX)h5ge_r}@38}aGDMifll2#bvy(fy?Cr({R<0N$v2<}KNPO$Z9!Ec2VCFbH0P
z*4LKc)LIX$RrN%_y`K0zY$-+rEW@tSwRpA27n9rg!E%}(`mFOq$bLWQo|5MWewdTu
zhX&bGS&ZF-qdI$0-+DjV4fDk(4S$?Cd>C8Rk3z3bGFUl5Sg0BT=QYO>r6GKsp5bt~
za~d1loI_5>3vf@rh-qn;U@`A9?r2;=<L4{F3%iE8-ZzkT?k4`uxd+QG53$PR2@>S<
z#eK(1s1?1zX{!Weg{0&4^eoso<-mGM9<J^Agu%X_F+tc`Qa}O6|KsQ^qpIA#E>1}+
z*c~V~U}Ma+?e>Zab~mA7V4#8`2nvdb4cH<Ef^?^J8#G9ZN;fE?@A|(V&fwlL?sYg0
z&pCVTx#s-M%pUY`5$`!q_N2!1-6^S5nUY+3(%RXow3VM9g%U-I;9k3hRdN)!RGu`W
z<>^Qd_E7y$q+cn$$gN}yQOk7l&YMNPm*&z-T0&(@R*=D09V$uDBaJB=$v1EdohsW(
zZ(ePq@>$zxq}vY4RoF$xZtSLoOZL%Z*JBiN_7r`dc9j-1#?#=2WO84aOmDX&(Kyp@
z)OSQ1i9S7qx<emvth29B=lze4+aj@3^N`qk^02UUJRn9bJ|G6a*eCk`-Yr63Sc*x$
zR^rNXYtggXS}2{h5p(uj5ffXkh}c_K#lGXWgiii#(R0Whv1QX8G0OIu*wSt%*6Umo
znE}^?^x<n_b>4lEzwm)rxca`B9&=mF_~9rP{&5ho-R;E)XIs&K{$;Uc^98YoXRBGf
z=ZW^;AqKT;i;Ro?MOLb+n0Z=7Y`WV+_<ZXwa@@Of&qQ}|&#DJ=0#!tUsfswBt0Jzn
z_Y$s?=ZJQdE#i{rCZW({qZqt;t61x@M>GuHB?h0|EP9!46M-6g#GBsw!gacVfEjm0
zUS2O0R%{S$*ER^1ZcU`AT}R&=YsrW;nR4?6^3iOd-R^ZX`*$s+w%5{;XLV$>k9C@L
ztl#vir}~d|bn0atsf5&#HfuQ7p48KG)_5$gG}1)lM(S#8q(f_(N%2+-`D?dPV`n?f
z@Bf?X?*E}zH~!LlB?+-$h=dq?>>u^c?IhFSUsPHvA*e(`tPlN1zHMDJW{$MjD<dOZ
zI%S025Lq#QkDM@zl^3HIDT-V9N+Q9gyI8$YMHtBS6ds>?iK54SgsX8sk+fidI5BXL
zc-}TxG~^8xdt!!*={{;A$aR#^craSTTpcS?j*n*zZldTlKvR5A))F)JP8NarQ^ek>
z)5Jl8=|a_fhPZTGTdY|&Q;d?BB@7?U5`9O^7S2~^i;nEsqDFC!7_B}>4CWpJn}R4z
z=PrW9+`BYr*a!UB`~joC#NZs?pVw}U!`FaBRE8y?O(g{>Gg9HAl!iLbG{mpXzzDw#
z&aiw$bK)liKmLLZm-2A7G9P~?=i~nVJh*&ekBJF$Y9{5Pjs2yPVL6<G$idl#IVf_<
z#t@#5?%m74o9YboOUOXmsSKPg&A_0a8F<n2Bj!|OVDC)UZ=QUD<dn})wf@ZKMg~Ul
zz9%pv9*b=w@Vz(;xd+3rg6A5g%1~@94Mo|%P!zI`Gr%bfFO<X4d@mgJ?8QlDrp=1}
zk<1&7fZC=Ae3~5&-;Pjde-DA2bO`#a4`#n`5Wevl`d%{#X3{~hYYD_?nILS`4uZ~&
zASiYZMxs|RY)^&2Vq++#uMLCrzHk)YiGXb#=gdu_;Wao07KO3c_MW@S?<8WIV=~^q
zO2zTibUZ5H-1@uEu-3@L)x%lHe8_CV?>ShtBM;tx^YJ335J|~Ja8xM8!`tN;U|7jL
z;y<C_SA%Ws^{ATMj1lYGP`&jROcwq@p8P*(W=qhWzRU?)AWg|zWT<p|HyW~%J2^+l
zQAd#+Exjzyy}}BVZLUD784A=-Qjs)zDbil%1MW*wAZ;TB8dxk(3)JQ5@N79U8ZS#{
z6q%D;CQYw6r!TjLHGznKSQ7pRb3A^*_)#1BIya-(y#b?gnW3as4P%=hxFb`E+&kq+
zW4&XzXAyG{3*gS0O&xb;bhxFXw?i_fyC>jp4d>~5$H69nd5ej$nEEOfJ_}-Tz%>Ro
z?A6wm{(#4d(eU8+^SMg|)TP3aK0Ora(}K}$Payuu1)xuoAFeSUzbMKFz0<t$rq~NI
zzVA`^&J&uCJTPqQJFJU+3nh=&IH&Xq1$&Fp`l=jH+N&`YHMl2H3oFicZCzf4aa*e}
zmvd@k*VG{@h5OTbe!~Zi3atEG4$Em}_#9k>Coj1Rkvl(cE9BwOq<l!3<S`c^AEvu`
zo@bBP+|8^}F?VVA=TC61WSvg-6Aai>{vjY6Hwp@Hy{!a8bURRWpaUhU@>KLxiQ3jH
z(=wh}X7a4rkM}KZXB6lSd#JXS%9H#*c{;yUfx`ZAF24uw8B3IC^D<=$W!)!wsS<7b
zDNijc<S3VC&{2-k6n9FBdTx>=YljYe8}}U@x65!~c_|7^i{Q2*8+-b(M|%WoX}TFG
zUz33x?wy^(pQBN3X`Dq)#m%Lu=yfQCb+r_14NHdH{bX!4;^%f@GLoW`aN<!sN<2Q`
zoh9pcQ~zPz2MN;dl3=Y*hV$Ru=*4l?Z%)gQ7Vk@^ujhT~O=$|8Cq*B8B&hgcCl>Hr
ze&b*h#_wx{4ez21L>+RJ>X7dE9hz&u!(Ap6`uzDaaSgz*J^rxh?}5Spx#NXCvtrC&
zq2}x@gkE;UMKya=ud>6gvlfs}xrpysykGJ$#pIc$@Y=(=4Et^r+Rmb%#aSe;ILqGO
zv&gt`7RHgR;S4+n-N)y!`RsX~@AR?Zi~)Q;8=!~2AruxHp*-FQlS~fbGY&KR>j;KF
zFh=@>)38#p$M2c;xV({d8)JKzT(QU9=d9<1*`wPQ8z@e<!PO2M6tB68DRI~E($F5q
z_uNAGz1t{oxr6u5@1jQjK0oIVa6tYcl$Sh)UGY;a=U#}{b*@<K?go?huW*0DYmEQ=
zntMFopkW*LUUhh&WYc>DANIq4Jp-B36O6{gACPUp`sSiU#BWK#)t#Bpr5rpP&mIBC
zd?*@m2jAx+1aDz2y|5TMUR6-Cs)zgXRyeE3Q!gl!|ArP+9G0fkVp(cgB}b-DnMaw=
zy04HYy)b#!XB221&ppP0N~9>;mzGW%Mg5PDrKH(&X|enwDw5;A*Bd%?XMrBMhi#;H
z_FE|Y*H)V9u##fESCdY}6p9%-jbi^!qh-sd(<}!)N}9fgmg#JyU7C8-qh62X$8Dgo
z%N_W8>jwEO=%A=Kf62x~N?gC$O*Frf7mefcX|aZikkRbPewd!ZVNEX~smDH!|Fng|
z!M$SZYb$Y~=CV+8u;xx|8}U=?ig5J5A{KJrWYvOOVjpLqo_gIDN@{o5n_(xe1lS3E
z$!nr=_BHWZ=9-Xta#OsJcp$oL?ukj|x5VFL4x+})K{QBS7sKXV6=lv=;!VN@an5z8
zu<x^8#BLoZoE~w;=%K3kp`{{vRdyE%t;%A@8fDSiU77m;l*PLn%3@}{vLNa%rr+-_
zDp&Rt65+GNFa6mfMt-9x*t1?3?$s5#nrnpBY#rgJv{6(CZ4nOd*l(irou(};r*w^P
z^g^+c@~3^L-8nx=ZD}>>1k{kkhFa==vW^rU)YIc%b@ZoqExq4XLn|ND(CeGEH1~NO
zsY<ebQ(Z@^-`0_fG4H3Y)X~kTIx5hqr+*_GXh&89HOe*8WV1%<x4Vh1tZ(KH-WHmz
z&fTdp9khGIZ%Y5xNxvs_(Fy%73Viv8JeU8bhcWFGKD(V7)7rVm@HegV>!Mq&k|L*%
zw3s(uMl|Ym6RA$JBI}o&&^@6bR3w#z$8%-juiHbMS5_5^QhSOzhu)%KV_&gLt-qMu
zz}}^pL89g15FvYhm{_=DglJhXQkacV7aQfq2vIUtoO?H3%+{YMnv^s}*gY+APiC^P
z)0-mvEvItF$28$%IbGOpm?5N;wZ)28>{A~%leM0iLX-1W8Y;8IkFm4F>swK9V6W}c
zh$wh1ipJei_T%_`fVF1~!p_9v_(r~8A4){Zo+SJ$O-9|>6g>Qt3O5t(bXt@SndKSG
z1^kFj??0gjvuI}K<l=^EJ{Dik!;P+7_G0E@;>KKz8kviq%pxBgl!K=ia*)mQ&4(M=
zX#JjnEhl+K8lHi|igdgWNylaVbetNJj!j5I?3grW#ie0(dMc7VQjr*&jv}cH?8!>R
zlsS<&c_p0hCt*n17>1=hrzl2*V*Z;@_VR|po9CRJW5VEN8HNM@!f@X+9C8OD5K<U{
zh1opkjE%rP_6mPr8;Wh?LhyD{Fm{^;A*d-3ha3XoxhxP>1NqJ*7YNZi5TWw};czPu
zHxz<67Z!xdUBR%K6@uuAp%^%aXDt12<cIJMWkwVfKS$#>bK3p+{TnhT0mtVg;nS`Z
zjCqy@htv!#dGiTd2YkVD_Dgfh26v6Kd2gPB2c~&=v8(`EE52fhbum6lmEmSo1tNXE
zLo%MbN-JvdxT^uZnWJ~Lk#&1!Ctfr817qobs14)}?*d8M%e@w<mC_WS%8cQM-Dur<
zS#sjesw`7EdQv7w=O%Fv(hhm*epsGXZI-8#+yyT8L5^n3;j_|)`<mW#BN@JjN8Xnv
z>oZd1yhxG;H2lLdtuAQl{Kn>;?V!Ugh_`G+9ru%YR90gYZ|2@Gx98Pp=EQiF;^UlR
z%y2K{&IO(Y4rar3{AX-PO@l$NWazBoebkva?9-1$cV<b?x*3B@6a%xc5BQ+^0p0jH
zbmo4Hi|m=&q!EGM{QF(AI237i!LW=7#24)VNX_@hIAcGkp7Mq6ZXf9N_J;AN_gK8k
z6S10nF3P;a&Y-tArTqqZR<Ce<fg2_VzJ%fx_Iur`!a=2Ke5Wd`U0Q{eJmWO*jPvze
z6-uqDVE>FgPg5$G#h-!D1*!1ROy%yxG+fE~h^DqQoLQO+JEd&M{>wqhh8)-y=HiKG
zE)Fn9Vfn;Q7{hZ-t1fq}y0EAGd?Kpkk|DJ@6^Hq(xwYy$bVI6P8^+z8x}6wnz<r)g
zf6-N<K#MLaP%kwFW+BVdp-|>FGS?`3wE|76RUrQgMe>Z{eN2iHr8p?jX&HGMmM%+q
z5v<qQN|8^41T7yXMO&IBD0Ec^tWzs7#<UC)lS=W<qzL-$S%{d#d%>_moFAHjcAhmO
z4|3oAM*f`fGr7My6(`zKaK44T-Vw=sHYa1q+a!ED5s&eE<M1mw2Agc7aqB`1E^Ld!
z%mH0!Un;?S2}$l8=Y7*(X*w-rs5|q>=WxGQqPjFyb0@|JlA=HC8D8Mrh-miKw6ljh
zrmX=FZ#Ll2a%M%S)M10xcVxDNAyY2|y~6$QK*IxtiZAhPr4^K0EbunsBHX@Oz}(mZ
z#d|Ga$Ul!?wm@nx3&_8{2#;B3SxYoU;6zi5pL~Wt6Q}sUJBh!R$Kkcx7~fcnDOr68
zlh|L=*mxG#VhthFM;}Y%^|@og0DJEn0DTRyrH*Hu*G8~gcL<6@52Jj{5j@j3<_?!*
z*sgXGg{LoIF3(izHJ8x8*$UqOStCu(2JHi^p)F~Hnk*a4Sa%g;ysjZ;tv%ZM-oo9f
zw_&#64k`@q!nNccdcJ>vMRv}7pLvA$Kc2$Q|2bBkeTLG3PvO7eDO|0dqkYQ@6pnDk
zuzv1{w|<8Ub375f&I`Aid||-*@!PJ^IJ-Iy2f0r#_Y?n~4t&A*S2=jZ`~CjEav&8|
zfa&*(;Hg^7vs^K2tKab6vJx8;e&8kN{7!CZK<swrQL^vdEr}W5>T={_AxAr-<Vc^t
z8>)Tzo^wQjIx`gM?Nue(`<J^PYDUt!ic!>${m0JT)##{aFPiJ7!u_&6X`OXHy1aY{
zcdDt;i+ST|v(rTKHP)nOAGPR>>!knpmlqGwCLL`ZQgK{K+23@zPePB%r>rM`qYX6N
zV*`2j-$2*TY@pntO|-B3R!Xw7pfb7BH29$vP5E}6ycavrlJ>{c?PqV{rOjvK#Kqh{
zzE`-{Ul#vevKC1iHX^^>Ms%LMA||YI5T0`Eqp7<kzV@>dI<8m6hflU*wz-{{%`;AY
zrk&XJh>aSucA|2PtvIytx~Oh;5EoMH#F~vZ!uN!g@K}9CoG4)ap6dm1#Ac^RzdTYD
zuIVctaktW5_Lamu>MmRtDvRpnN@CwjMX_l+cQx%-WR9Mqko%%2YP%Fg+dL)l(?~_6
zT%Re%kDMi<*3A?x(Yj)H(pur)wOUN+7%v9*X(EYr->EY28}0Qjr^WGQwEjssl`pBF
zr*pqiKb;?RC-Wz*8CFTMH9sk<R}Fo)tD&9!s>tMV1y$d#qMyC0Y5%zzs*tTC#MhC0
zS{<ob)sgRL{^zxI6n><R0_y81@OeFj&TF85w;E`4eFO6w8_7|&iH6iQ@waaaz5dop
zr6KK<s`rZwO@7n8nBSB#?Kdeu?w|)vZFH+g8)b6GszPH2eKhz(&KVLScC?h(r7JB4
z^ImI9S~qcdf}EK9NM6*BQWTk?%){B!T^OsX2(@@svCpEHI54-57%kCHbjI};^KK3l
z61xVA!YM;Vw%l+rvv`EK=s!{%x}q)yt{Wqk^cW}Zgp3yr>n4iikG!jzuEibVlh^|_
zS&a3VBCa`26%!1n3Du#~#pD1kQJXPC++&Z<=E@o3=m2fuH(6VZyBY=Qbx}C!7X?f1
zRa%r6jU7)u;Fxs`ym`lRa$OwUJmaC+EeV+)laaTb?{>DSC|jJyp5HXMd`?FrYc59*
zWFj&p8#C_ZK$g$w_HXPbzL5)?4Y_dB&P8XRTsZ&8fu~;%R+!~r&fFY)yp)Y&t1}?a
zdyC!1=@{srj_ywBC|;kAFH-5~!=9D3@@a5vO@(U&XSgOM;KS2M)cZwXz@Z5EsYJkp
zXNu0aFnG)lgHBE;a&Ls<iBTxN>xaVCA{0yBha$2m6nl1r;l^mrNb&zu$}>pSjR?f5
zMxgmp7^3flz<f_IR;>xbg4=<R?;eQgR{@CG9)R%^18}cT0DcS(z@)_i@OKDc9|r$)
z4CI`15UlzJW2zMIwUk2fep(nFT@1(hh6v_6N5N(42YmY#!@RgSEGbOjZiQsDo@K8w
zYdGqmAJNt4Gg^LsM*in7oS)7_Ja>(oupehYNFEHr3ZUQe6|WDJps71|iI#sua@G%|
zeW^wP`$N72HNwrl1+~lC@wnm_7Vr53&(Oc9Vm3~hg(Rh)WKRw=p`C`v(4KS|D${2z
z=6g4);hdG<U0Di@VirrPEcNC-@)F)dg{jHXt4G{F*~mVT;WE7MlBR3io%vful9ng`
z!>*NGSn2v3J<GUXN~IM$W;bDpK|OMqbMv-=IfQ)w9vxGGEsM*Lm0ApwHD9>{JRg2M
zvsJSHJJ|mtq<Jp3IhcfdVe!~f5R0_uF*sfP0r9*K_0DIGH2c7A^7|I>CK6iy5$JRc
z2g7Bsw|^*1Zv=C8I1oc90LSY5V0GM&`FnmCu*4T~%04)-!3$~`o_O--9UMH|k#hVk
zhN``Rs^2SkFLs0CikEO2@Dk7~22;o|x9letTXF7cNfj3Gj8ibX3J;iZ@VulFO#{jx
z{p<_xBGPeedMdPsr(*tuR0N($$3~en6zHVlFZW4AaW6<P^Y!=1e!&y=i2Vut2q)GS
zw~YUY-C7x_FHV31&lER5C82RvD#EvYg`0Z?##~`0;6(QS=<{CeSPK@$wPN52&cWzN
zQGe!H&bcH{o>$~)aJf9iO;w;7?6X;YR*^2(C{U9!>u&6gdMY7L{yX_>`o{b~=71k(
zU)13homiyt7cC<?u$j3)F?-7Jv%VNv>xwYJF$*i^CUCcDA&!)$<A7~CCNAO6Ssi!Y
zTcp9bPZ~VGrJyJ^8J-<{&YpRSV2h`amwt#(gC0U%_6hc+@aIK662DC%(c|+U{P(L1
z$(a&lbX%J4)$o37Jo|1&$<P7T5Y4%ty|`40jNUe4m1`pc?lofdjYiapdSpd5z-4Cx
zX4cem-nkAN9m9|j!2bT)V2pS6#Dg!dF;CqFT8qrFxx*X}v&?bB_X5w!7FfUD0$Qsq
z&@|Hm2{IPwb@L)lce{uSfu?94Yl=7OXVC0>3avX&;$Hb7sNXX}!eT@88+QmNBaL7)
z!U)FBhFETU5Edy1A@@@s$2S>ZUc3QvdKp6MoDq&rIE19eL(r%_3>O%qTh&qC4Vhre
za#N@(oJaT&bGVpVVAkSG*gwq@&1IJn8*dG@XEvzVXN&4B*C4UwI(jd^iM!FaP|<W7
zitNLwJa88q!X7}|-Wk_|9wR8{DN_8NVQ<`1e5-$g)LBn)O!^td_I`mi<u9;M_9Zq(
zxMAm`*SNIyEl#g`2T}DNZ~pioNZlXWXM?fmUNTPh`-0XdIoKAT1A}ilm?)cz*sem{
zjO0w}s$xVXSKz5`C6t1HK>k@RnqJmp{|oNCjcmav?k;()B1f}Q_&Y#?S)u*q=`QcI
zog)-T$4`-tvCr#Absy^XXe3QIHHtE-MpEg~-ZX9j=PpmF(0tDPoi5}~`Pv~=%U!Iu
zL&nlIy9s2sQ<K)jYtgi~+(YIymF&i9Q_`%N^mOeq+L5=E?CW%CLc1Ql{I;G>jov`<
zP3!4Oe_i@$x{mfXZ=?Fs{p2;_B(18oBuQCoT6^;b{VlphTYfz!)6SkEPJW4a@^G(c
zZn!L_np%r118l_Oc<$4leMNX}b`b8#j$-a$TVXl(s<^lHs@OL7vWPOa5ne`iV!oxF
zxG~R8JPfuKKgQV#McL~jXRe(XDrYC|xmb&C=P!vdgDk|9S(f5~Fc(7x?GiB`1_?uh
zJ|g3$syOn!yNK>m5;LzWitpDH#Gjw?LdHd2blA%a30HZsI7D9fg~@XkntyKVE&}dr
z3)lWLMGEJfS5BHLRQ<HXEa{0tUGg7!f2bzCj!KG8sGu3uWwgJwloZs<=)OrAW&bXt
zq~<c3d7+$^9;@K)jB>j5w~QKH%INjlGP<$8j0(a_$z=UcTCu&F);_DH9h>TCg+U!n
z%Bv-}l{I8Mq>kCqbu@%^oDXB_d0$meR)ZVJY-<Cxn>En=*9~-bQX?&1&_uh!o9V`$
zRyvy7Mu#ifDRM;zxwW^GbVnN{)V7k0Z7ZGM)<#ALJGc|1ljc4CM-zmkh&(GLO#GyU
z&mS4lXOpbh8Y?HP7b%DXpE+lKL|KgK&3ng46*1~qPceIRZ!xN<kEnO<C%m=|5DP~P
z5(9q>7NdNI3J<g4%=c3hy%a`?&9UlYnaLQ@&~u!4^l-dTQJpA!3^hc=D@`#yNlSc8
zn<UEoCyPFoQ^ZX6=JZIODm0f(6V-302}ADad^BddI6il}5Er8Gavb{`T%(XbB^pl?
zqhY{&-z&Rf`1>yw^Owd!{cJq;KTW_en?yLYCqsW(Dt0pu=fI0JC>==0={&v*?$3l=
zbSBp7<=~QU4$NofqG~`c!n^Zcs(&sd6mt<;o&#y099W&t!QSaPP&Z}GoHFpfE*(*7
z>F}^g$8PovE3!{zPB^m!dA500nF^h<R9L-E#e2?o`%H+0mv;nIHb%hr2hS2Z;W+2b
zXW(ewO?idl&E`<t8qNEu-k~@%I21;_(~#cIuUDZs^*jt~OT*B5H3aFagJC}=7=JGX
zBRoAA`gemcXGb80vo<c&0x&w%ABso(v1BIy9O93a?DhD;n!#^tf6S`$M~-O#`=0}$
zT^oocUxJuT8jP@x5G?1N*p`>!SUZV(tk`dC{@??an#5wz$#@h6C&G6l^Ib~PpjXAd
zz;U1Oo7u*$+dpIF<}a+VXTonS`#?wK;^63f+}K=*n(!ica0Yn}^K*^LD`6V*6SB{0
zkae7A3+@nD{iOxz+uHFg?iUXC_ygJLe=%c_1bvj{uJs~GTKSB37OSO6p;VfFZ<C>i
zVKU?_*^L~=cO!*q-N<4{H`-9a`>d-n^g&*R-1bS+OBX4c%RY_^Ardra^FN;9y715a
zH@t^;;IC;b-;bMcq+0`YSFwl1xeD!V-{HCU8{!MgaBoiuN`HRECc^@(NX<pxwk+&R
z{EV4PGH@Y01=H>(LY@6Kt2pmn#QvJExzR{hkA@yUhntcjA@e5!N7)blt8X|yck<ql
zGi>Mf1*5Ad5EteLpqMjK3(ooCk&7>8e)Gk>G3+n??2QYy@A10C1F~lBkeK)uspYTn
z*x?oSPIJSodRLqs<BGgvuFzEYia8U1Kx$(pF1)V5lzn9wJ+h3Mish^$RlrBH0&f<S
zBIzpgym@D{K0O)h%#v~6C<VbMQt>-C1w(JA03*I&+{P?q|Cfa<HSXRx${ks!KcY2-
zJ62m#(VCNhfHMgwzn=g$4`GUGDh^Knid~&0ke&S#?vrb=&%6Qky<4#DNGq0%Xh(-M
zcR`Gnq76J(X3vqL1olC7>!CpXUn-Diu>$Q&P#}x*typ)U6$7;R_wcs`JJVX=bEgdp
z*0x}xM?1zapYZO=7R>jpU`@0PSA2Qatj<QPTo(4Mk3i;<NTjiMZOycFq>fI97T-}u
zo=ZcwfoV8YnSw-@=g=Ab5Pxf(k-5kTa#x&i|GNuH9zTS~!wA^%e%Fv9vG&*>SkITB
zwah^pa7dC|o=Q;@cgIE4NKqE=NypYol5eUMY4Loc$@9(6yFBAuZ$x<CdJH?(fb_Wy
z_`y8Jia~XV9KlQ}zJu)Lo$rvzZ_(y%4)1y9cskY``DBhCA?7&ZbO9P$EKsjw!R&Gi
z<Y`&Jy8R;5^e-aT;4IW4P4RieSsXoWig=qdFgH4hUr~pkcE|`(3WjL#(#PnUgE(n<
z5aDYM;?uT+2(&*4-vE8=?9cO!l>v5?8DPaWL-c4f!tC%v*!AKtBsz~^Ld;R5*B(c;
znkkOvpF>K!8SM60V1f1}eAlu>*Nw~Ab<-LXkJ?~Mmo3ugU&Dae*KsVw0heYt!vD)H
zxNf-v|EUhx#o2O=**74O<bX44Z=&G)O&oG@#QLrW=q<<mWS(_apMD1M=mnlcxg!3w
z8_L;}^M&(NVNUMsgZF~&I&aup_`tu+AMxyw+MoOx?yQMN#WH)LA_q>K;~0Cd5SF`(
za96PywW}(yp-&}}zx`x?VJ!xgaE@?gBQ9KEKEk|a)P8Km_-aWybBOzJ4#?3Ep0A^s
zqj-e-SajLvCFQ0}0pADE@Fk-tCv7CDmG&aVl%BMMSrUpXRcS-kVER)wgoeK#MStza
z($Z}cDD0I6^?#^Eb4N}l$J(jXZK^hvJOka-TuQ$6OX!V_E-n0^M}4f<(|n)xlw7@r
zcK_0$y-(KDp86fMY0qJLw)rA?<zAwxotNq7ej6%2@5npc7xeass)$iqDn@WGwr|m8
zG4GJIaF?(VgP+-mGrg~ft@|Cs(p!$A`{S!3JMD^4k-93j$5@GI?s_^mk@r*!xc{ln
zR{S|?E527>70-5D6_&i$Dz?8Y4p&?j%MV+K+2_o}ajSE}JoUWz$URRz<_r|`9`zQt
z`m2b4`;~>yGDXoDDKF&0IWs*>PBf>=iizH`!qHP!guIj$g=U<e9wsY34_6Wf<uk;-
zZQA0L)C^I4YO+|SqAARqB*dAMO{DEvMGosK>A&=HifAvTF+WP^yjux%IhByl=@N>K
zE2XLZ%jl|28NFRxP808!(M!!TvS=(N>HEc0;8jfCswE_u|Bc2iET+!h`K0kVk1|s7
z==halnttIcB^p=L{a$tS#G{U!=hoA{xApWkrJj<m@-Ax@XRS8W)99%UG@0{PvwAku
zf#WTdaH^H$n_J1FsFjq~w{mA&3r&x2q5u3_skd7@xtRT?{pDTsal3?2dnPFclt_td
z<79-aMK_`FLskshEiY^<6~s?dC2^-uccC5FLk!=gD%P{-dS7U7p?0*d&>z!Zl$Q(;
zKOYTZ{cDIYA2du1$Q~{>Tvro~vqy>j1?pnxkuf5;daPKme!N)YIzedWPZT*V8ltLM
zQ(Vo_5@ViD5_@zei_c}8v)VC5^o*J!Rw+ys&u32+^6RFGK5o3@X^lk3$|$%G<(^5e
zXm}pr4uhF7c%~kU$&=#nZ9_b&P9$KJQ6gt3lkkCez89{gVzMFoaOBdVygVI7dowXG
zn0;>DbMW;5vt7({@c1cv-lB7G<8Ka<q;t_+k^}O}f&V$)QI%$6(ur*7s%2ox^fXi^
zr(w-b?tYL@XB{RD9v*3U#~j-wk<5I{NyXaYB$VBW#QtHC_~8)&v6x>e;dnPA9Cz7c
zliV!~s?4Z4!81+&{17OIa{kFT1TujkaES~-U3LgQ^S}2$5P}Pi!F;b_pD_D{2eQ{@
z!?pm(6|!cd695-?e`rkhhitMRd`<nZajqXqhWg>cAU`NGTgB#-A6{nrA$f~Gws!gB
zRAd0M9tYx<3u`}t!5H1bxpKWQ+~Qn%`f2|APK}0ZH{SLAjm4%l2^cgpnLDY|5VwFi
zb=yDUx#LH?`@|iRzMt?~`3r_|Kd7cvHu_lSqQ7rG`gZ?{c86l_3gv#jx(dAZ{f;-M
ztKh0ti_pw^$n0-IZf*<yo6Ncj^LW>pcjAh17tCk>!$4#1z+g}Cs*jS?x>JfqWJ%Gj
zDbgfwB2C&4cz^Xonl^DCx%FDkm$yqXBUg$H3MI*Gs3hIw+)wedzX-qo2h$CHL#Iy%
z#yoAspI*&)#q7JbkF~h3QH|p_eqhV*Z}@S%9J(zf7<{1!m*op#e>D$9wb`&7l?m+u
zpOBW7hWYI0`0+9UPo~7da#;+1@oXcP5QVcfk!WFEAcXx;tIsl5$T|$4&a?iuIRttG
zf>9F?2=D0u80q52+#6r?jPizr)q5B{eg_lNcX+soy9g(^WA2-`c=46lXx#l{wfPlV
z|Gh-Jrz?h9y}-%T=a6)M24xBEtqRIzKK3`P9#D?+_e)Xryc7vf%J8*!ImYUjVPpn-
zk)0B8(K-c>HYQ`*q-3bGcG>@SGD@8Jz8shgQ}a*wtd)t;n>g!Vm;qDXWywXRz)UY0
zRpXP8e2euui3GGh<1E;pL>%%<!HY>>G10RGr&uGZD*XZdggQ)Ez`1nh4C+O<;e>5F
zn%}i!#i({{sN}s5`#eKf!#R3OfgJJ_Xb#`0|CqI6k6|n7*nd0yPYYC%TCm8f1rL|E
zV1hyeHdQvj((otRBrD+Zs097)=VEhcHYB=#;IlOxrN$9R3y4NmpLCpwPD4&2_oSUn
z!|;A-kmjt^EI$|M&2iyf@&jBFPT0cp&G38|^t|;Djwak)q7(q7wGnt)(+M?5P>qcQ
z$;3&}KyB`0pUHV;eobK}%j>Ta)Ok{ZHu8M)^JODa?lj_;E$cUz8?ef$23LaX(Cc9x
z*7ps<ujQdIOA5vUr8fvHHOI@P=Exatj+Sob7;a{U+p^}&uCaiDwgpP3SYV*01tR)c
zV0Fnw++BMS*0$!bvOS08BhTX38B@&eeg>LDP9y8D2_6L-Kq|w4?>PF13OI;kx(6{*
z?I0w^uy&(!5ank27@wsN%|!;-;$r~K(T4DRX^2Iphp>FZVGPPW49!+!EO~t#m+H>o
zd&)UxTAN|V;fuJDWdX%amUt>IWAJWkD6FwT{>m$uJMtQaIbFw-VGg)bc@vM>Yuv>C
zo#wyxxLt6KbL;j<818_K`3{)lbQ3*pJ0e=#LbJ^S=-WQ!d(Sf@OFieE>kAx}dWp*O
zuTXpT4ODd8SsV9)(IRie?e>MWo<I7yCS$?*G=vW2I|OGpF7q97_=-FXI$D4aLyPdW
zqzD1_<tR1yhRcONk&;^jml1U^s%gNa;Z2ZHWfqiq3x0cdK&9a~n(L&fhqMg)ahQjG
zRgvA@O4Ko0nUt0*)9qFz+Q;1Je}|Y;{h}w`Gwng{Z9T|rEpsTjSLBXqU#gy_MtZxo
zXq*)Hx4TTGzD_eJgNa<p5p(F!*m<;Q(;~VybO~Mjw1ob<rAML5*3(MWRg`SMg613P
z(mkxF&t^O6%*Ml{mt#S?DVNAE-jX7;Y-sC2N4lT*oVq7x(YErX;<>M$*rRt)sI9ja
z=9SiB%VitkCUr%eK6X=V=;0`iFR~FUPnips7nek{rj?j(XDh}_+6lM5cH)kkt=KWv
zR_MIGDn^XCDkSe(iQ+z&g?EdkC{n&4?vFSxjJKPLT*H$hU+$P_d2>XpdDT<MobE1U
z5)?%qcTLXE=X1QEn<!k|O<1+ah^rYg;$R5B?#qaCOJ#(Oy|l0%A}Lf9ev@kNHkv)A
zm8369i?TWyp;p{Q?}j(h9`Td#-#1DPDx;rYOK6{WG0h)QOn%Wt^jfos*7YkS-C+e(
zy)~c8GIJ<eHJ7rl=FsjRxs>(aR~qK>l}ulKCE4f2wC71aab}rnqw**=D33I|<x}te
zUujVMSF$@-O%Ju|=yh@(on2N>r`U&l>3$uR_|`HLtCsQ<*|WoQ&fkOv>KNQa|EV>T
zX;3qr`r1tEyqam1T{HE`Y^G~pTj+3b8@*ipix#i=Lp$XD(SJ!2;<JL3Fk2@rQeVml
zH+fmncvep6cgTw#mWslwkFr?d*<Gw!&AsFes-p3JFX6VZkGRy(SI9r>FHH0Xiu<yI
zMPkqpQKCOgP|p#fz*|jNE*-^uMRn1A-WW0b(O6MkK2D_doFF1bPZSG>YlsbhHN^ZF
zO>xRtOBhN_5*N-+60LcYM2*^HVX$tp*lRFZgj|cnsVMFzn9SMcJ5d-vCK|f@{pV=-
z0jnm)z)PAtm1e|avQ7dHElxySQWCxlOTpx=saVPW7~_aE><&)DgC(4I9-M`m*IBr(
z!EBf8Yz*I?_5Yrn;5(d;VjqKxOe(y}QgHfn3f_H8!3@^J*BP+Zub76t)@hIx=}2UM
z!_^q}I=C=vLpv28Es_xG5s6f0%se?0iJ|>i7kSNnm+Tb|@C|1k2YYHb3*J^1iWc_K
zgz_Bo=Sm2&c7$Np@(>iUKjy0%XAW5VRvaAygLNTLDGSEImBH{*3x=Ip06L!gW45zD
zf`|K~w}&4l!w+qlzW8_17pC($cRa`!&Z@rfXRYAWPG8&&^5q@|KXiU%-pX}<W~>LG
zd}|<%9Sy>6SJs6(d6r^M;KiO1=t_vhJjZDCKO2LP*f^XXoCIg~VvJ=5w0;C<#I|N&
z#^?-u;=GvlmQN^_{DKf>EN(B%hIY3+=&vt8PW)FWZY+UUM;WHLeuJ^j50uw&R_Q`5
z5`WdhY+Dl)pS2(}p^bS!9jJf*8*)efAV%sh1|0o|_DvFGUn)Tw%+r16E=dJHxyMUc
ziVkrX`Ig_3G%`k#;`k08z}}5r+azey3f@P#bzyLJC-xNm!q@0_q+2p;Z8U!udNttU
zggP90SPeJjpRly3gjr`f6fTzH2;UV?ITxbW_<W?i%fZb7S$M%&qDdyq5;%~GcJ^>>
zeH@Qt>?7{n^uO~q3T|)N>+KbR?9_0W{|v*!-=TQS?4cu-!H@|JLdr?@J*fsj*1->%
ztv>M9^~N5b_xNk$iSIYw;g7*Pge-rDdCWu8YJCHm{u&Y*uMn6164FOq(L?<OdhuDI
zHTxO78lNI$YaZ`kbFg4Fdw~aM<Jz};JU#Oj3(blUem)8p<bu%Fjc2Oq326Ej45jK2
zI0q+T+jjQadMBd)mL$ymoP<Rw8E6^u1qCKwI0Ke}Z;LpKbc1!r4++q@6NecG5^%yQ
z9%B%Po4cd2ds-A4*n{5dML0gGM__ML1g08(#?`z3>#yM6(P6E~iEZT`6y{(Cv|!Eg
z7OeHIMd}I}8n;$~-Z?4I;4lT6JXe9@9&}>kgjTfv<}T3K7F;%MLGpqY_WjmlsbL)!
z4*!nVU(3<0dohYKc#ioShrgU3w4EQ0n*4CgEDJ#wcaz9}jKs~CY4F{}`b|wLQYSjY
z-@_Syrar*A&kt~FrW0b{IYXhtg?EWz7@zYVp7%mvyg!0{PMxrwAVIVD@O%?4LG8?&
ztK%8xCeJrN*GiICyaaKhFU>INWX5G9zbB12ZqtZ4R~qnca}69js<6vB07G{A!0zcQ
z3}<hx!mA5d?P8AWlg;7(&kSFR&9GnF4DU<Lpfub9mXj@@$DSKSo_}<t%`wrzj6F5y
z(Yxn)w8@;qi)>T)d7Q!1{%2s)?=+?~o8bCu1Ke9{0MCYlFgSG(`2!B(^Y;T-{PO_5
zc0Y(OD-YtJtv+5z8({h+1N3Y*z<_;*aFsHGQ>PIQ96F3w*Nw4%*Ku6cKLwwNb3iY1
z6zg7ufv*LwO|?XEmlb}^vc?<OV3W84%hs!yYIhyA^*6BZnLS32wa512>o8Wnj=|yA
z5OB*LuLe6nZGi(mSTJ+h$`J`YZlOr+0ovw2#ejmRkbM6PhXbGEKi&_T&v=FJ3*R72
z%^mWaJn-|p7dj{V;+Z^uXWdFhwrncKjr$10y3ep0pN-ssc{n!m6ZEV<A;c;Z-*rnM
zd<yZXD*<)GQ!spI5xlq8A&7lZj=4=p;a=L9y=`#1{~KSIa-Um47j%1cqaqV#U}<o^
zQbn2Mr?C&dLWz0?vEP{a=)rq?(ylaS>oB+JYClyP$2|1xF@331ON}DuXpu|lBr?7+
zm84$HpkE7S(fNQmH0s|xig~+`W}I708#XMV;+u;p%Y6wAxW0@|c3(lwvvf%>TaQdn
zY^N*d4|8w1IgPTlpaqMp$UxJEb!bOA<@21D=z7q)k1;gp;aZXOcC#?JZ!Df1IVr5}
zoDo;nIWSM<rcgSyN7T$eDLl%JMcRntLfzd`kd3XVtmd6mm#tWN%~DMFuoMekTME@o
zOYwWPmFS;pB@7eKh(l)Q#T}=!BJm{uc@IoPe6q19*?vTPEL0T}3YA1jKLufIE-QSq
zWyGOx(jt9_v`AxiOhJ{D_?jpsoE)V@w~<of&kRXnKJPD8=yuQspB73z*Gz}ZB}L0y
z3DLLPZ_?k_Kv8<%DWO{hO|mQ{Po7b}Y%8K?H^0(_#zL|jUdXJ6d{TOqM+X{nDdcbt
zrT5CFBA-lV)n}4IpG<lk@r9<m%%_0)1+?U1K0R&7qt7q$=*fdTT9lbb3Wp0RYspvA
zZK$FN_p0f-ULF1TT}SEvI9KIaOK<wtQ2n54Do?1U8U1SM+vPe6%HlkESp)UE(@3s0
zjbx<MM42<1$Stgiw0Pc;ebY*w3hYG=XK&7~E_zw<kA_T@6dh-zL}H?}xG<)hSmY=x
zM#{^J3!E1}-&;v+a8(x1=kyTUvs6U3Nl$UBZ*MUmg!g^B`-yz{0Ych)pm@7!u!#CS
zM5NyzCc0~k5Lf)v!~(5R;=nz1akOf*STSy_*sC*6yx%rnyj(Lu44*JjEU2C+Hs96|
z;UhIgh?AzU?$8vo=4pxiBU-}fvX+pL31iPmB;GkiVnIVBc1T6zfPFN2h!2?2`2mvO
zV&K;jgU+q7c=042zn&!_>j!5mkP3@oX&7-U4T)-LxG^^qAxc?zv?mL4`B~VsG8_8g
znHbmn3FhvpcqdX({Ez3E;uKuWNWsmd6!iL(f<pdtWQ$X=UpEc;W73gR#r}qsoL7xz
z4#&$x+~fJ+{+mc--;Klu1Lj+)aCi5M2)rL3ftgRkkt`PuJ<d*bXJ4`LAodg6h9E;L
z1ipWQaknrSL&Jh``)V*&4h@EGd?2K&{IT@8AEt2brmUwgB>(tgpMpPD{O>;0>WitK
zzNlT~iwA{1m~Q2R6ybwTRUf!Z`0yRf2buGI(BqB|#whr5hn_EH>G?s7_J;v`bWGF&
z(OV}7ZypB2RV5Tkufi~IYXsi(kHWdKXw2^7Ox5&67_Z>G?a?%VXPcD%>F`{ThCb@)
zuzA3lvHw2d_Jl9c(#XO;?$_$LlZR;?%vCvA1V4#V{N-#?zj560=))Nw&1!7CQwxjo
zdhXq9!a%JS98+(@sg8E2zW#-oQ#w)N_y?T{T{!gVFS@<^hh#pJY7R)yO1^i;+ey%W
zCnTtZ@8MQy|FHJkUo4C2LL+;5W`6vQFFk)jPNN+oN48>kdlP29Zvd^WLrY0DCK)ip
zueuUbk5{1WcPU=lG4pOX&tZZ2c)cnYx3jX~xAF`6KKqDStldmv-|FlU?6tLwL+6|r
z3|Y@JUn2LlUyeZdv2aA44CC%n-bY;w!3-1D);0vea%3R>X8U8&F+bGQ`oM~3n`2Mj
zqcy+-KZd<UPdhivP<n}%>_uJtpBubixZ%o2-baPAH+8Wq`sP2!TgPYkb?hl}E<C{w
z&8IM0$XcICHku8xV45BYmo^VrPxC_R9G+ie1JJ;GFv(v*Ft{Iwxbh(Mo)7|@2uJq(
z4;VN+0c+I~ahY@MgYPHe!>Dxjb$*87nG7`Xj#d4?G<dOJ&A&Dp4a_1)?jM0${Mp$(
zIvk^2BC%929D^Q*<8V$m=CKb<!8Qt$cI4vhlzRBI)?@oz&c>-Xa~DVx+&6IFx@0v@
z^eRUe?;M<ExDTZx48yyEaeady!k@CoS+yMU_22N|MIEd-U%h;IJ^a{HWVWgjttDkx
zHn$j#+cROyzk>wM)(`dL{VdOpD}q9>>V-c#&b&oU!4ug1bcW76CnVl;f^DDsc=qo;
zlp-JCn3OZ9;2|1bK7;GVH}HZdWDP5Da07R-By*?gZsr#*W1WwmsgvU*$a%8_J#l3J
zSCRyM50#+zw%moQ)QJYQcJOy46vs7UpJ@%0nUnM<HW<^!d%}L?D^&D%gZ0dd%oe@?
z`C4=C*)~T*o*5QQG=oOfdBlX7A%BPkbT#;;Y=Iw(%n|*^42iqUVE^+x6y(k$`90^N
zl+STC?+lLiIs=^#r_oSg0(EBtoG>_uCYysu*F1>Q>;t&sdVral2Vk6ffIDyxLR(88
zI<fk2(lx;1Xaj5;ZivwLhOo6a!iLyG*gwS>20e^$_NWn*`X0uFOfwjdz6dLO3#{?H
zghH}{gxh6w6j@`b>=o3ly^6gPuA^W0br_Df=gwk#d~~~xGgq(S)ZJ?+QLu;Fu^U)$
z-2u;!-9+IoN9Z*=GLZE?HeYxE6YHm_igCsqHSQ$8?SfT*T+rP78c=eF&1?@u>3LzV
zgfEuYvZnkx5I;_a!hCBa_AHFyy?g@a`nj`BJ_D2VKVkMr&N%1=;DO=?eCB8R6W>kE
z_&z!&jAv)vMy#`Mg3bue>ua{*!TjIYB=H9w=3VHx@(+g09FiZWL_;`3b*5d34D*!8
z?zR#wJf%d{yOrqXxbCD7qD;eedXTgn_r56hrGioY>D#igw0_tGDp@m;o^xOPw6(J+
zCu|P=$(Y9+goTvgvXBPaEuvuE#q@9Q5_;Eb85N&hPR7yeXo8I%IW68!?zV@?2FJ;T
zNr*}FtXPY<K?YkKDRTL9Y7Td!x4%3nAvuE9Br6Lvs*0bi3tZ4w5xYAj#ZL(>5i?_f
zm|VJ7#CsnS$#aj0O1BfD^Vbnk5o{!S&pIN^Kb{iLB+iQc?xy1XSyNFHcShWIJ0jW@
zPl!p^P6&x@Cq&tR6XI3!QK6o5M68NEEY8eV74oi%;_VhWVYyyL-0_kUvW1dj${|Vd
zQA1LEk&zT@3nav}dlKStZwcWu{Vy$h{hOW)<?MS=6R9m}q$!pSH2t!KNb1{0ua1AG
z3oE~o5AU4fHA-mI$|91jE2JwO1tc}0fR4?{C#}w0+N+jJMnO50RFO?*T(Zdd?-$zK
z=L_l4XPWxz6S?mENH~&CFOKBXhSWT=IhRMv4fE)NdmfFyS3tjV3aQPgiYoh6)7SKB
z3X7|wZmlKE^!rJc$ExVel^^74@RQzss-mc6HRS!UmaP5iXwap4(v@qVWBVKEaz+E~
z;cU5YXcI-AZ=r4eZ8Syu7ab4lq_~-XY4R2cvGJj#Sn*Rz<SdX8OWnGOuYKjjN(c5O
zsVa)CPD*0rSk8+F_YgzYsEXF_J;h?H-eUcTzG8MnKM}-UoZ0OIMcj?SqWg%U;==1;
zV#nwaqVBeuxYIaNbV6NBK0aD_JB$&h?~N4|m&b|lt>cCKfC-{8Y=Q_}F;U!#m?+!_
zYlwk+G(`Cg4WaN-Lu}g=gtZBw+=U&ALww#W5m8X`jlwhL8sE<NfYrBSFex(z8Pj7i
zv4)v0<Kr+eC=Rvzm@g$6kHaN#&<%~l%$e-XaZkp+N@mwB$iy9;Ow>qbBDUlc)E!e1
z%UaAF*1vPvzYvm~0{5sCeD+L1!=Oa&?oL8PRuV@T_@2zWDp}q~X(Y!()jJZOf+D$p
zE)q4YGnMU$#CGLKXg`gByjldFJA`9la~OK)gkj->P=u5)mu7MZ`)`9W{XsB_js@e>
z!eG>P1YywOAWT^th=f>wRN4Ds@k3v%`0IlS;Xc?B<b&*3AFN<~pxM_4WvhHRqwJ0Q
zmfkp`?Tu|x-tevRLTr;4T!(n0r=d6Oa=h{Spby^lVfIRiFXkuuaTlpS)-(nngLCAE
zp0ihSW+*cFT+w9T!nS!)2x(@YNn$+Gc0{4Od=&1Ch{o@R1WZ*<!4iId{+BOucF2II
z3D1P5zTm{2EQDup)=WJgbvFvJNUj(PHl;Yj40Z?hkD3<$z{zRVcz3K8^|$MBfio)K
zFEyiPSu2iKwQ)YF1F>P8uln&DZ+>(_DfAES?dZbf^e)6H{)JA@zxYwy#T^QKU(5J|
z%(_m*W^z92GIwfAbwHMT-|a105O$Th$;TRC!g*fn=4xm={lwkr-_bLr0{?cGVa2av
z%)a{-I@1fFmYItM$9V53#k)$&Pgq!;j+;Egs<RGP_az=38)NZF?E{kLMxpHupYa>Q
zu}g!qXM;oWt5*n2WP>rKJ`k4^1K@Ro=SJQS4$Sg_6MGJhaQ1s)iwB<naEDg=YdDQ}
zLrSVE9wxXV<k?HKG`{3~%1bota>WY1`<|Nn4C(DpppyL<D-s^z;oye|;`?!ZNhl_0
zgu`n2Tg-R$fW(dWkh<*+Ns|Bsc?ZIbpQVOoKXj!BVh?BJ%#Vk`-jn~XKcjH@-3M5>
z$D)z-nRnxo@YOL9<{hjrE)Pda&ltEm#2~ShvtPW!QIQYF-~r5xoE3(CdO;Yvgc$_a
zm<Jgb4yp2RM1+PTZ7km}G(zCr6@>0DOYz*W5i34(Pv`qu4F6t<8Li9{H2Q+SD|p{E
zBM`Z-Ut;&yTUgz28TNlH@!!l$-1hr{wxl0O=jXwD^>_Ts`GyXQGVJS7jG#W5=<odz
zxyO8QX5TGzes{$DPmVY_`WBKB?&8^nd$?457ZaGz5}|ev{yF#XsN^<I_}k<0Iv2!@
zdxDwqFVM8%73_9+z~>fs%$?>gT}OWIbpPPv>n`XU{e|&h?v3B<4C&s^SXt}@mrN&6
zPwtebcAP%fj$Z2A?H<+$JEO+`pKt2Sg3&MJEsu^bP;vYsq-V2s!}HBeo^M9To1>e%
z8LE}c_-=C^w#>8n_2w+@>^KYM#ES^`ya4@K+^ulR3|lMDW0}->$nt!%j^`UQo^PDw
z&)_<1E$xve7~^1oQA-ZOPWvGAD-Iyq{s7uH9zgom0~l*_01Z6js74-y#}$2;PB6f}
z+XmS4$pBHS4WXi91P#eU82!MIGhc=npl5_o`9oL~Y=+pc=D3`75g*Q8Le3y7Jhi`!
z@HA`Y_}HMW$5njlcO7j!(|jqqj(-i;(Qug=fyUR+q-M)ZFFVZoZj0764p7)~6J66C
z;Z*5}6}XR>Sr4#9*NHuf&IpNiMuxfz?pnBDQICg6Zhec=xVMo0=K*JFZ>SgeV)oJi
z<aY(4@$OA{9dyL)H4h+j&<XD^eZW+Acksm+Jw^n;@@6zF_Qb)VZwhufromEzy>ltt
zLo&Gu=j)oGc(jdoj=#`9z7s*z1>0%df4oD2J}@Ww{2FO$`YcU$n`Fp0OpaDgRwCO}
zW=k-$pl*>et-qv99?VA{uB}36ReIBX?k`WcIfzaz<i7janzW1cn}Ofv&;*US<n(Gj
zdCgczgGVnSz2A%InCcQbYPgvG4qZ<FN6}S<WwmWlQS8L-E<ASYn9J@KI{-nDP;4>j
z?i2+HMNmQM1_^1zK+2%IK|nE)JMY8(^e9K=uRYdCmEZg5*MS4H+x{fAYn`K|^Ka36
z4;}6pu%wHOZ@$w5N{f6%ce=c3TfPrz_75bNc_CC@9YL#$;wV6*)9&YG)N^t(t?mAY
zy00EA(r-(Nbu}l1;i4nrb@!vft59Bi(l{l?x1SJE3yummIx7}VI47+3s0($o6GG{#
zteBswCZ<e3E6(jXC(iEI5SyoHh^B96MG^Pe46p3V9d<p%({j#6pYA52b-U<n@jqH(
z@Q<dh`A4Nyopf_oC*5EEmqPdcrhu(KXsjn|JrQ4NUrsXx&TOKCb80E6vW6a;RMB4j
z3Yt8sj0#>BlN;lZ&F2cJ=I9q{Q~gZEDxauf%SS5L&!_44^QoU=J}p5Gd92MMSE+1@
zIgmvwAHOGylke!TMkaZ-WYB~&c@)~7OE(&FXy=`La{KaytZo&O#*JdSU0y<m@0Rl8
zTT0(wmy+JHGU_y}Acsp;B>TOZwp!HkPJBJxxX?h??>18I?Ix0*!@WD)p`&!1wX0QM
zX^~DV)h=(R)d#+lkN*$C_}?_p;xDcF_mAGibrW{IyNmkcJ;baiNf9}vmoRkbEfx&y
zD~?<79{JD#!rFD9cqctrltm2@V|MX2Rmm{%{K^RNw$~_OdT+G&Gi|K+<HLUASrbI_
z{fQ#)?<8?y-xT3sH&vXAn<mWiri+Ta86r1srs!`wOI+n$&gl)a#rkW~qOC?+9NIWX
z+%cIW&WFws|Gl0gS|&x}L!T&UvbM8cCI&LY5>R8rp5vwjSbHX-ODl<cags1_FlVVM
z*^@IM8Oy&UV-e@8Zfs4&Ti)Z|k)Mo*hDlIi+@oW_9@1|)u=<{ZS=V#W|8o|u>GL)x
z=T8Q5$6X<FH?vbRF)@m}bhOi8()9{C8VT6fKLJgTSSOnsk8V=&-0cvLtR3+fVHl5j
zjBOk<V-RJ|-j-9*$l4u^aMq(9-iXDc+Lw62_YdF8G0=SyjU(w%IIhG##YIuL$oERm
zcad0jEfV*>a4sr09EWYfkiuOO?yNmM$qmNPi^05S9gNA7f-!4QFl5-<(eq#smRANs
zK|2sv=LBMJa{&C_1z<Aw;TE(8z-dJw!aM@`{uTtYtRR@&4TimD2)wR@a%L<HK^fs#
zGA|NRuh`#nCk8pw_}tDNE+doT@v$r(ZtOQn>WW7RKgOGGr{XAQh^MY(-v)10sWZlz
z74aTxqCVhOP7dsPa;9qcCk%Z01@{LO;n;Qdi<XwbZha-z+^t4Kel6Z~Z$Qw9Cgcui
z!ON<znC;$%jwRnP&igx@S~_q}>L)z<{KA5qUzl)#HzvRQ#)e^k5H;!#mQ*vI(fSS9
z!e2Pj;}<;Jf8f8s4y+d6nXhZdyMk6+WNu;n*CuqlZ$J}coAU>1P{Wzi)xPC8xUUq|
z9YvV$QGonypHbbKk3^?jTv(h9?fCatx|+Sz>=D@9Hy!5MDVRGU3AsNLm}_HRj<aT~
zCcMOz3hwOV$7-HWB$7FoKlyw(ur>_RT_Jej8;sU9LD&-=fQ@7Q5q0PVI(PcAUxUy6
zTHaW>%L~U3vk!BmH#(1bV?>G<z7Ktj#EFlv$k_vJhutBw+zrDgyP~<#1>gN$aK_*v
z+O52CaHkJs<e9%w3&88cK^Q0#4%;kW%()kUJ^w-=n-`9!rtCXsjx+fiZ^(a$hu_Rt
z1hmECXGJ2sty9o>KN8RPMq+t!C~oLSpj;^yu_~O^^NWR{>r2#XMMH}F1`a%rK(PYr
zp0?q55zHQ=_*fh~5etXH7(B9#!V|9uTs;(q!LIyzD)$4H_+svZC+OVAzM+S1n7Pmu
zrpYd7I^+zgVV3Ao$JyyhJveZVCrMTZE9)&W!a0?<<nmB6<RfJI@c!q_A}D7k!hxPc
zYOe#1SK8r%jvX}O?C^AwJyf&p;gabHdETjf>T?$*^>?vA(jF#%ELg8|M)k`F$cpyF
z1jEM=`@AvevkwO3F}D!s4#~rA_`ts7h?g#K{^*RWW=`0|_{e#yDGHaG!phA8PfXn~
zWyB9?Uv9(xsoW2JmAAbZzg13YhUa|F!`*a-u7d%h_ws&sk~UfgYr|Vh3tPKs@z&QJ
z_`BbR>x)~6^0<k?8*kukmL`UET*X?e%h-JX3Le<r#s=5hD3iH`1XIR8);I7W^g2px
zu0b-Lxtr6f_`Y5l3!W>(^Sd(kKTyVvHOkQMRKoo}%J{QN8FL(!(QE!0bU(xz)j(B@
zxuA-_*Hp2ySQX<;)Np0BI`(v_;!~~~^q2E~gSk2uS>Hj3l@@}R>!5J19{NcdV1te!
z-X<7v--R(cTTLL{#|%|4hsRQLejl2m#M1<62RK8{Tc7>SjiH`w#63zD7`cYg|4=(z
z-{}C4PfiHwyodEt&X84gLh?Zuct$dZHsAsF`PjlH#Tr^|7Ff917}mdx5j55m(_PK5
zcq+dhu31C6zb#Un?T|nEE?f&9Aoi~>Jii6NCN>PN9&s2rB^CSs%S01r=^h4r#=iGW
za9jBm8_%_4ooK`0l6HPSeCJ){zepI>h22fvC}*ey<*t(;y9*LDe~u(+_U%apDcm8!
z+aRXr`%#~x+*{DAKe-O&-Pi?#$-F~~2G1By*YrlyEcuCKb$vQ%OD>@D`qeaK<{Ek@
zzm7b=tfw!@8>rQ5Bi+ncOSxl+qP;d#Qruo@cHB=Ufm>*PurxWO&!q(qwJGAHC23Yz
zk<aLR<W%vH9$0(RNDUvlA@4_xK|z#V5>DoBv9$J8I<0wFN>MYKDDztbd8@ROWWY~K
zl8_KZ2e*p5iQ9xsvaFC%loNlmPKZ74)`&B_SrmI_qu4!Wix{`>xJc1DDK7p|5o=zk
ziD-$lqO<Uv7~*wKe4KSwXq;CSYE6B`!goEyxb{wJEBiyg{`*6UpMTS1=4qB|{-)DY
ze$!WjU+jVYNyc6sG_B7!TG#cJv`#cr<-SImwWgko^Qx)Stcor@s-UA;W%R?Iy)(;-
z=;H1IGWq<OjDkOr+4GOI!7ZQG=;hIfuw1e;$s>>1xny?XJ<0MNt0mwam9BqBL*w64
zsO=jnQGZQyrleEM^;9Y?Ng=oW=|qDvC}iwwy4A0cUN0&pXPpvyzM_=QDV5Szew-F-
zexc@hMRa**IgQe(q#@g?X+(Jqjon>G5&HF{sn<ZoGL7UhsEKMmHBsxQW-^=mm0XKk
z=|^Qd4dl+8K6ieSWzla+-r7k+;=3qonuMsn)SdH8J;aCqc;~^Xmyl+i>ZopC;lr8v
z={f_%kN@wBTMQO2hD(WG?nA{^>EYsN@CfmA<tTABdbGH|Y%KR5jT18lPY|C}CyK1#
zNuqzlWbtnBR1r04nn)cpUF_&&oRcv_sOZcT`4eV|4c@av;)vPe+=ba<SMY41Ts&Ju
zwX>F$9RdBIXcSJ0f&9c6+@27_@52PNhbQoRmV0p2k`Tk&R0H&qV9uR5a>>lM|42mg
ztwcm^O2o1M5>Y)o5qd7KFm+D~W-iFY(~umrIb@^R?E@12y@y{kYg3%Dy}-C8pRerr
zOy1qjMC#jD`1zUdb=-9l-N9#I>v-In5|6QIak$mNcwkXH3<{$0=xsFWKd|n9D;fr~
zcpG69|2U3+<h;cTlPJu2#$5M_7>vFY4du;I5N(lgN{@uGJ9pWfii97ZZ}R0Lpms4F
zTV%p;Wql|%9uC1Kw_w~E6AZoBAbh{Z=b@!RIG!DdXD0*EQW=0)%>cBH2|!k^Ka9CE
zuGNdTU=sYfH`$+cxB$H8uPHPRg!|?o*fBm@!(2|>bUybV41<F|^Eew=W08tRSt##S
zaX;OnY|c9eMIxT@+RZWX=<5`R$dXssHz*Z-nE%q9^cw%zqco2_NT=D4(>jnngM2=J
z&OYplm!I%;NCEpz*l!$Hf>{I0p}B{*gtV&><Xnq5*LwDDHX?a%Gdh2_;Fck8Db=@O
z_MC6Hu;DvKE#iIbZyl&*59{HtKM+0RCoazZiT*!+;Ii!xjN{+iF_k%=@$3QN4QEN+
zHkh=2g|}$IK-DHFTy0>!stz|c)*xYU754^}Bh#!DHfxJ9_geu9o_&U<>_=Q1k_U<S
zY@ATeLTVTLvn(0!4$8nmn^Y{Fk_@+x>_tiB=eRBwDmxgL7Dc0n6LTcWk#Jud0l%@~
zaA^qzqC#MKB^WEmF<0gn0O{HONHTeWnj~LnWjsTd9p_4%y<lSX7)9}qQToISf9k!U
zH^K|GA09DA^1$U9H+%|kMNdcO_zhejOr0_8uro??-H~tLg>83v%ka{3Y_EIFTSs1S
zKlKDD>b|_i=+9hC2-c4Z$Jk27CXAIU^`jARD+Y_3VsL5^@5^`NO)||W{(B;^@d}@x
z*^f2iAV2q0BB632217lfA%|$RH*ofAdIWB<AH9&z+w?RLRytwWb}AHk+}Si#G6tG*
zQIJ)M!1RVtM4S(T(MHxFY<*bwc!KvQ9z$!N8|uG3z~)33I2>av;co%kz6LlIqlYFx
zT}TyZW7a_}<aX1B%r0$wU8Tc)hQ`=>-4x@~cq{3eJ;Gqmn~rw)`knW+pV%Q_qCE^o
zI$+B+M@-&!7dyjj;jL<m`THy(n`8l(%g*>|;ewuwZw?-RjMht!F=&M+cQijl4?}wd
zS=nO1eH$FBw1(a~Yb4(=gZq3_Tvj*1o4KZNQnUaJ+_0kCH$2?h1}V-?Szc<zik-Z_
zu(%m_8CS_gn&7Z9<C~TGa8J{Q-b`&6=xAZ2gcc4f-odZiw=wJXEtp=qiGoEpu>7Sa
zPWHWssh7^<vWy15-_Nn9_Z$qDoa62)?i)6}%KZXYp{jfpIjdE0#Zn3Vmnp;2Nf{SY
zl+n6h8K_sn5=SLm3{Zk`oe~T(ln`#A40Gu-P~UY1cIVG<N2Mz6E2<*!w<?0%)UYg9
z74l=$pvu1E_ea&SZucEL=ZyHj%i7G<=;BqUE@lqb!>aXq$iJ+IMR)ZOJKO-llMFC#
zwgKE08sO{-18Dv-;=HI4RE`?KPS*$nYK^e%nI-)8*}(q0HLQYdQMTV6it&#8S?UBA
z=D;Q$cEPU#J18%=h1G9M%+54K;dUdOPBzA#5vEA_Xa?Ok7AP>VLhEa5eoSm|q{AMz
zx%cp&nhze%4d8x`G}OLK16ikIxIs4Jt@E&C_-BlhE#^L_N<{Xkg}r?PN~D`{&!PoV
zo!?Qs<{xHor&0{#n{^~X7tXP_YN{j^50RvQy?T<VY;S6<=}pGUeQ4ILek2vbI@O%P
z)Z?!d)$-<p;mMKo>*9Dizj-?KJ~We@jxMCWGuF_dHR~w<`FbkZ0lJe+oVQv*>1r!!
zj_)Gc?jTL)T&Ge}znNrpMVj0W%%vsG^XbE(g_M_NM5<bL^rhZ~?)LSj@hg3(>emZ;
zx-N(;&xX^reX*osmQFX>n;RruO>c*DPfTSS4b=ZhY9;?@_RH>~My97Qxzbz6w)PWq
zh7S>@lctI^>Dgj(_l;s>w=E(``>42WazgAobVj(XR~0jw)WnA|XGO8?SuxQ_UHrF3
zRh08a<bPd%Xwi=jl1=NNZ0>_u`mTe{@|6+VLGJ1uv@QEP4N3b(N2aw?N-_7$Y;C63
zB@J}>O&xjr)X;8;D!Ox`g60O55iKvJ@)5;!>}~<QdGVQsulz)#dVQpdmON6-&!wn<
z9LjXcrlN5lXn@gs+P<Fg&EL25q2@L9TK9(b{mJC*<k!@&Cxf(&(#U*4DoH;|p+WCc
zXxY3}+VCWmp8v_9i9d_z{OS@?>nNeJrKL1z!3WwiI-jtkfYun7(1p?Ew0B@7wZ~M^
z@1-?ld#jdeZR)5{r=Hy88))$EMv7BuqK9eRFR1yI^3S$WqvJPP@}+~W%>PAitpCum
z&Q6-Yrkm(6mJpLacNZI$N(#$++|?-2Tb#VyM?}^16{8jU3(ukf!s_H8F|cT`SSQDu
zT_r<>svLJ4e;FYr$cz$|iK9j3oU!7o$v9E)alE)YdZJjhZIYOyJXs7sKSlgdm?|c(
zn<l38nJyBdri=Q6GsL6MGepzYnWE5Rrnp=-Q-luSorfv2MEATf=6j=YUm^xu>Y}+b
zA{tw2IDf=m%Nov6Z8*u?&9)?DbC!44E!KB6lJLAV5so*xJAVUj9?nQa^Pog@{!2g}
zb3ZA|5}Dh{!DOXubRGVHW$eQ&I?jF?+Z6alreRTBCX%BwQF-kZUNR=q3QoZ3O$k_5
z!Ch`z@h}_|k77^uLb3+6Uz_vapQ5opESfvoqA{JdrxA>6oO?%OUt1J9n5(fij=~7e
z5Bu=FW_KUvbvz?+>}Dk78P_PTjzr$C2$<PLpt&s^n?`c(duSLw4GINK3xVkg#w0O8
zP+c7anan`6o(jaW+5p^F4&Zz--%ro_qgK)%Sz+9TYv_j|SN$;7)DJlcepoueA44Db
zBXUIm%)SOdJ1G#30YNw!#@g1W5Ip97infSwj1G*%`NPpD)s8{nvX^j}#rN@<+z+S_
z1qtr$nQ|@xvx<^%e{Cw3ZB55oyhh`qH^}Dv?xRgvP*l#wVBS0Zn3WGB{!Hq1q5!|$
zi?F1i1i$)~W6qpPELg@o&g@!vbgyUZ*nstNO?Y3>jGN0jt8%Utr*5~QK%pJg6TTtb
z?;9MrzouY4W1fxQ(P!v)yyagTIPx17?rev{o;G+*ZAIAY7MSm7=Il!&R7&f)Q?M47
zC2BC|YbCb5D~EVkiUI1y=$c;$_14cY3FgfI<ve^@kb@7kS+I9`hu95okW-q0oj3Ti
zb!Z9-xWDhW2|wSuaf}yV;z|D)?#PY8Lhcbd-#Y>g{8~(Z5{flfLLj|77}^bisIv)x
z*JOW;fA9ije|%w1K2SRN6a)8qqr&7dj@Wo&u;gQ$IOc_OL0)hO;=cR44^d#{hK;jb
z@utTEyl!zuY~_8t`+E-`Bkw_9&lPTJ4=`o2JF@z@!z;!E)dxI5uf4e^&Ic#i7b0;w
z7`>drFkdYKKbTiK|0xP4kJy_Y7mf8j`Ss@?g@7?pIJuCq^!i{tm1W<SV+aP%j==Dp
z5m-|jh49!YsGp0%1;cRsiwcFc)f2dJK4!48Gxm>g<Xr$`{0T8YRB0rX*7NuOG!(0L
z1>xFI=9CnDU@`6qzA8S(rDRu3W*^3<7cQ9jL?0fJ`pDJLhv7v%{L#{7Y^jYgzwco0
zo;$FdstrRa9ZdbIi!1*6kUMDr`7iqTdB_e{H*Dds!4^+Aud{NgJtnu<Lq5g<0lBtN
zW1nj`H*3&9OU%_}53GtaTr8ZCX6S+-MHeWzyJGde2lzM90q4Hhan9Tp<wtGcaOWDj
z9;o8t2~|v2P{k}+6Re6h#;ZXV%qcZMv3naPur}4o*{N0MTalps6;W%NS&Os80vjV3
zNf=<;41Js#pofyx+9<fKg_Q2>AHI19ebsN{&h}dvzwag<55Iwd6D}cm)&<-ed>#eA
z&SB)sv(W#hjy%qQS3l*9xAJLN%{`4UOGT`jb{ebOPec8?5_GRAqlf!x-i_hQSqbBZ
zDq+NaB}ls}LGrQ^_uD8TnSI9_SDnF|V`nh)mI}5Os-QxRvsbg!AmgfvmrbfTc|(o2
zQ`KS3x|uF>ID>t(;PF@oEkAUj-IuWq=f%6L>EVI79xe<sz%|A<S7#Z(m+{T$l?M1r
zhPd9>2$$*&(SM>5&UqMNYOM(t<})V~XN^nOY_W-XH90v4gpPLN_oM?hCE26X-45rS
zt>Adi0B>#>;#N;1RDL!_vy%y|ADJVo(E_fU*~4sMg>DKq*nEqz>w<guES}=d(^NcJ
zk%og|X;|iwj-H#d;k70Y?QcF|sZ1fZakeV`cMX=BH$Y#VI}7ZZaefE)hqkt0PDU%<
z3~s|U^-esx`w#!`)jE^bopQJ<d+e;96m`8PE#pqW`(OJ~)llvSHRTPmPASrr7*3_j
zN78-y@f3D(D$STRll#(V)9SmkNwY?pvR=-m;p-OB@fk~L?~!FBC$)@L%P%7Jqtf(m
z^i(ppok5F-Nz;=tbIH_sJ}uw8kQN_WOxuJl1+8+V7cLJ;zxo+v-g!ZDJp!q!Ba9YJ
zjw8v%ugPTC2Qsv-Ak#w)v}GrIMGSt@1do3-Xj^wNFI!SfMQ;%j%H4CxgN5$#u_BST
z0K6PGhzII>MM2tOq3d;AsD#UjQ^{w<FiADBw}yAVubdT2m#B-vzN*4KjyE+8YN)-i
zg?f53k0IMePRVT~^{b7H|F+Sy#5U4e-bSBAD{qvxkYs8T?KW&61(`Y$J8I~{@+!(`
zEvGe0%E&aNg#6=Km+~tl!Cv6S9UrN`c|P~q<k941xui2Rhf)eY&>M{``Xb)){%IyD
z*}SF^$9U5sJA)ETUQ@==*L2P}gM3$}lT=PB^-M^i-fL4xb9)NCv`eA+zf!2LMk>ia
zN~Pe}X>|H!2Azz3Lr(XzXm3s~S?&EyVH1kz>YEZ;SyDz7+%5QgCVPTzSCgxE4Gjpa
zrJnwEq!(FF^XmTZPi~@6_9j=?ekJQ}?c}}eJB`!g{i>W_G;`%&vh({#aOoy)S4)T)
z=pl4COLqTvPjT^dZ_)chA8};^XV1bpGe3{>RNjNc{Ru<FhPzT?K>uOFOpCX=8b*ks
z{i8%<z-WOUV@1mLapHmAc=7w$1aT~GqPUVUN$m2UEEbwf5nel{3dNsO#a7#CV)M}H
z;-vO;VVyf&6bzpsHm#i@(hkfJa<{_Z-YbkdN<z`YF$`Z-!_aq57>@AyQ!6$a6U#Y+
zB*A*sqgR;bl!!0$cprOOBIaF6#Ae1dlFqMicKIu8)=fZ|6YDOoVsX>%CFJC@5wx7U
zT%Wy%?t-^yXyt7_-6Zs5zUPR;EB1!GVh=$A=H6o+?7swj<}63LMm(1NiG#$AIMyX%
z@o^z{59db1!zCI|b)!)}FB-Q;a|TK>8gsvLwk4bSn@3RyI~j!&TcfaNN)+anvkvet
z0`b2iu)81vIxZ1-zaRqZ9KunS6b82sq0Fa;U}Ie{rVS0oSJoxkL)bYvDG+aM19-<g
z0LRb!Bcs$0`s_D2-0%X6>|S6y>mobo1>SR3XYcJ7P|JS7oU$KQRro>M$sbP@12AYy
zAg-?q;@;h09I^^Q%1C}3PetI{kSO@~kLDisXq>sk8b%=Z=9EOjJ0b=q`Ek7I_zEYE
zBx5pn#THyl$E~yMV><T+>u<b6pluc=hh<}1MJ{YdeT3C+-oMZ-z@z8f{r9>Ak3aJc
zLS6-LB2}@6ss?e(I0Kqik2k9tk!se&*sK|YU$!9J<trrjwPMScR+O)3!x8S5JIGi1
z@-}qla<}aIRyb;XWlp9Ak5ro>HGzH0zMTIaUXLSkwFtGS#-@jr==G5E(>A;-b*Y%o
z;)ST0$h+>%A93Mj9^y@MFm~SubRYg6ULUyk*MzsGmZ#%SSqlEzC1I-!XXUvU=-Sd)
zNS}$pIrf;gZjZ#AX5Nx|8ivfXp@`?5%!r>s2nq|t^D_ZB*~cG$?!7?cP}W#9eIOmn
zS&}Mmly>vN@V%b+w)qj-7J6d2k}IZOa7D=&SLPNu<9z1<p31qfhsYTVcihLPb@z}c
z<Af6SnZ#y0!zSGYhbmlgu)qypu6p2m`6I+tdtnTB%dIr_gGEgcMw|@8fWC}Xrm+UK
zJ_<#3QOFC6g2B{CY)c7&zb11_bAxarHwd?Vf)SV=jOvLYC{zoFg-$rku18?LVg%=z
z!ZCSjDCE{X!3oKSa2Ub8GI|bJy~qS}FB<R;h(5xP>cOL37w*&au<fBGds3}gkF>$R
z7#sYVZwucR8)($(L)h!1x4J&mI`xn>MGrH%=WfgqZS;4%gO_h_L-W%OJX~@UPs(q>
z!S4?4eAI*Ie~fR2>0yks4d$xZU_NV9b?m>H&G$}EV+REL*}}li1|!HCikTKTHq-=x
z>$ub9f&o4#>SIf_Gse5!N7ffd{5@xnh=X?UU2KbB-)nfki!~f4RoER+h2~0CBytz9
zy{$2h`<Y{>WFy**xbKGVkuLXJc|Woh0iDh07;b^vtX~<uGQjy~`uKNCAF}+t$H?nq
zlfD*q^wz@ht9PJ~b{p;!Zox+SCe{wS4EvE6ai#wSB>mJtbp1Kld!2<<qdKyb)zRCN
z|K8K5Q9k=Lv~Mb+=N5U~+I9+08|BeQ;uNeVo<f<UB6{hZ#zFPdNdA5r7AB|BJU|Kc
zr<7pR;|xa6JOj(^XK=je48r!Ppes`aiwafIuwRwCA5^h-iyG1^)vz#I4Ov?>5N4<W
z<$Nt@^S^bnx(=LH>moZ^7vS$D*G3PDjBf@qzWIFD0Ee;+puE@+YB`47-((0=Z$sQ<
zjI+nf1T|h3uy(Y@^24@}dUglu4ZJBa?i%Kv*MvK6VhD3{DW~<2$$#F%IflI5V94(&
zL+D&F!DKIUBr?9ax6uki4XxmEgrDR0c6j)}2@eN*qfX-yOy{IwF84Cn=D)#_-n?&P
zlEwTF_pufgpzBOImM*MC_V+qOuWLX=egj4*HsQtP7F;>Uc>=3e{EcnLV6PuAeej1j
zSN>s1ZwYd;>_O9cN45K`-rSATgGAp(Sb5fAqj3!^{#7G)<RF@qF_yislX>HMDk)Bz
zL06~Dq+FkwG;rxG%6qzi-pDN`_Z3U1q+l`C%$P-fIpgUp=gSw5nL$4@XLGOIe>C*c
zd|I(;A+7GQnDLJ_xgT|)?h_u;r6bR1dDV0BlnA7t@!@21EROt7rqQnVAIV`^Ik}&#
zr|+zzj4=I4dP@IjdV_@cdR$UCb5Bjz^nT)k>|k-Fg7=SpE*CMe+r+4sdxYZYLt;<a
zF<~4kE2NIgi!;x9i-yYybYos3?VFWKr?$SKyX*65mP<LgCe+ap-$pW$Y^Ej2%~UtH
zg+A?Vp}lij=uAK}b-Zn&2R9q3J-VJUV{7SWOf|{8uB4Lp<rJS>O05zlWa>~v-;Ik%
z^I`$r|MHRU|CdjNA9Crjdk&4iolQzIA4q<17D?#5qnbW%sNrS?1+Pk{rLyUC$~&Dt
zcFQ2!+;l3MlTJNnrIGxQRC>Q8g^s^UrrgM6>i0F7#_Ua@w$v1=-<e9wbkb;$Mh3|*
zeM2w(-jkO>4yj%LNV{6T&=R*Inw!9Q%esv6rc}^)qe@bJRYgJN)imgL4fn*>(JlFU
zO3!W}cdaHGF55yrXIjbMs-5~|aGq-TPx^l0H(f0FOOKCq5xC#SZe4dV+o^|`*(NDm
z5A+f)QN0CC>nonv^b<Y)^%t5N1I6H?K_X%25FryOCFG|J6Gx1O3!U;2LSn@z(V;t9
zXvK{YsZC?WoPp!T(s2`n|Hz5L^WQ{qID3-VX*XHS#}twEeu@~icdGai$sV17)5OJH
z(}aZPG%?<M8h0ayL9Z?pw}L}?w<Q#5r$h0QdGdaImNfQ_#)Z6SeAI}+BBull=+E7H
z+^=`<WgLF6hV;!o4u`_xkoPto-?<~O$Ei5@U5Q2QjF)J*9gghs5avx-17Z(D559lj
zUY-Pb$wWL__lnQ+uOR&;0apwY7$+s*EN9f$D8%DeeH@bZ$D#M7SR{Ro!4T;v-lt+*
zpdAf2_F?Leh{jO~#$BBEUX{Rj#+tc(-i%rn5Q)2$VF+Bo8!cvxXBgL5vL9ml#t1zB
z7>*kv9NTY%;j(=w`f~0mlYJfglNj4H1#)L#Ao^Yj!1P!C=snsWM>YMBUGM^3J6~Y_
zTkgU=^PKxCp5sWTFE(_4j(78)W4hjRNOnAjj@b)DtnkB`PCxES^GAAW0N!N;Vnl5a
zR_x$>`QmW&{ThMDN0C^2ly`N!!=XGX0vUX5SB~b-%veN^;GDE}BK|!}#uz?tr9Mta
zuIFo9_u)K!)H@u^$%5rK#yDg1;JfZ4uActP-XQKbwkkrNWeHB+EJLq-6__)q3a5gq
zadAa0wz$`!xuPECy&CyG(}XJpO*mrCTZV&M(8;~SrFUD9>)3)ur53pKY(b_bpQ)qx
zbM0*-!onL+z#TiQrRp%ktOk)~RWP1di9s97G2}og6l97Kv#Aie^S(fD@F(Oo<{=?9
z2fnOFF22M$>2+^0zaQ^Uze~qo%T#>YkqmFf!c)uR5%oTn_rhYZmNiB{_C<Adgkv7#
z>c&l>=r0k1-LHZWtrG}C=EagK{jlEf1=bIHj`cP^aQprg=^LNG_m&sbeLT_U*F(g(
zd7!VpJG6GYVwAKiy7zF!_Wdr%-s6lH_V*F^<Q{CioN)6#dlDR(_YZZzvVHe(xYik0
zp1HzE%MFWrdZ1F~5lUIxYV3X=^AtRgbMG<Aw}j#EC-zd+@y?;;3oP02oINp55gi!L
zy_3PvtO?{E+Cap!_gRhg$^LCYyi*<wXau9gB^W*dVQ^Xyj_QA*SjikuS{A>Sw!5Pw
z$O(wC!j{{Huw?#XSeqV>F4e={0$ud)r-yLXYt#~W@2JKa+wR%mQj!houG&DcP#^bi
z>hpU-AJe_`AeE<!9UpaISgM67qi#c~5BpSZ-oV-98%Vi(i+7Rk;E9VKdOy%Zr-UvP
zqOGCqVT}<vR@k{p9Wte=xU^0WJsxmI`IQas_O!;mG;`=A86!%I^^^m;xUZ_s?=vmP
z+|xpar8d0Q+vEBQJ9rPdj_tkAqKASS_p7Ml#X40ytW<%Cw<<;)HAaDrF?^a#(5TUb
zp!8-mSv8}__(ph7Gsm3Q=G<Lu4$nV&h~K4;Czti%_<=c)vD(;V%GgFq3j@yGK@H=Z
zo9#F8fvzB@@FG+sE+W74JeHPf;1KuHw6RunbBhuV{^cuG2>~}wVf_44Sov5U8-L5e
z^`kr@>gAbRmWSimQz%%W2!Az2$TH4pNmhi@%F_t<J&o3{r=eMV8V`3VfhH<Lnla8o
z_8cpWS3zS8U)@xZ8lr-TZK{x*s)j!<YREpQj<J{2arli27WkY&@W}IU?mEw#<ht0<
ztjl{JdN5*KS<#@6P)7rVMesG)5IY|k;z@5qD2!p8a?B8Wvvir0yo{&OXCc-9EF1=2
zgh$aO)Vr~N=+iZn1>S)5n>%Q&*2QB#eWbS=VB=jwOv_*nXs!`%emCNtLKAF=Hp5w8
z3p~4EjpqyO5ZKQNNvzM(kB8_k;f{$nT=~7=io@%iU>0;AGyAw9Bg6vrCoQmrxwdk7
z3(W1}-kwY=#D_N`LW6T|O)WSr)r$Xywj<KD1GC@##?FQRuv4iUC6DYuE;>DEP@nGP
z*V=(8O%1FQ*1<}x8WVM^u)cx)c{^&6WKajEhykSbbvo6(ok7QY&ZOe|GsyPdOnS6-
z4)s{Nh-N=qObXXVk-_J2G-%djYVey*@$QUoe*H(<XXi5)vXI_xT})FRSWuRYJr!Ac
z($}X?srJls8Xe_NlQ)Ocm^U%JXaAZO+vZbEKq(cP)X{C%R(cuwllss<QgM?I2RO_9
zdvP!EctKxrxMPsG6+cd#__SR7ez#SeuiY&!Zjcd?a}J3=y<|oB4mr^?<Tdr}jHmdL
zS5)^pg?nEz>DlI765orcw@D?*f2$#{)H?c<Q%~6%4OAG|Knpw@s0s~)tM$~ixsLwD
z)KKt`Dk_{(NgH;RQ=>*HshJj&TuLEXOfI06gFn+N{g345l}9@c<dPP1Heaef(741b
z>V5A$rR;r2E=g~wBkVOb1*Fp<i!{<Zkw)>C(}=^xB&W&_{?b(HGbWYRZ%CmG`(*k$
zKbek>Po|PB$u#3(GBr+4p^oendOjqL4z#CJp=l<C2EC*Dx!F`1kw;tVKG8zc0{R$O
zL>{IkG-hTQ{kvOE>g5%*byO8~Y^<i6w`%Cpn_AkpwVu>k8)$h(6D><?p}MkGs*?If
zYY%jgzt>Ni*y|5HzTQbKKfCC`CJEu}&|Ub{_7II5dJ3&ay@Ym;K4OGwU!nG{pKx9_
zK!mX0_(}J{V#ApsVqU70sG2-Xq+J~@MkkCAQWDJLEE_F;og5>4u8tKCZjTdQ7srcS
znF-?3l!>CFV4~=wKS`V&Hd%DqOcu|6OctkhOcCQvr-=N(DZ(syiqQPUybW_V4;@1>
zM?Dl1=Z3;JEd+VXx$|XPFm#QBm}?J2l}{kX_m0A-1H6mS9t*$yv54fn>r~k|?Ayee
zl5{+*mvY8r3GYFD<J{02?!L<mfnyPOmhf4)U{(}<^-RLPr@T|X;1x#qe1)1f2`Iji
zfRi=xkPeMU=h1jHl*VBO;xJ-SEHs(3+5R;WeZZWD25V-FZ&FwrsAg?yPhk{R_(dU@
zv&8e|Bavaoeg>Z~<Ogt0dr}0JG2g#uQ3U^6!eL+<j_)mD=(&Wi!=ae36oNkL!T5QR
zGdX&J5S%$ao8phy7C$_h?}wgRFHn&C92-_VhpewJYu~=uT;anzqdt6p^udS%AMB9k
zK2aB6_;at^ZT{LBjxS(!#t&Pz`$Mc|Uk-ba6Ltr2{v;Tsd@pQftQXoD1PkL}WPA<5
z1l}3&=*}7TD8`OQ<FF_s0ZJbe@u4^wRt2ee{V^RmpI)P^ls6B*@&<+^cjJuD!S$ti
zXxaJ^DhEDe@__<8-dKc`DJAf!FGa+Ia@Z}dL|?xu4C`BilufnRAy<d81NA7L)&TD}
z4G7uQh=-nyn8P?`LuMn!yEbC$+D5z%Z2)!h2Gf{&-a)Rzmyb0tV{gxk9#x!=slczW
zG9=cOU~AuEj2>NxCu6=qO6n6jyXEtFGY6g@IExzk9upnjB2Oa|B^xs!Ga?Os70GZ7
zPQ*de1azH?WB%YJdtbPpD1q_b<OoE&he30BC@z);!|-kpR_+Z%&prV-5#xtD$}hP8
z<2h#9`|!Qt8Q$G|f*$X^uvh9avUfg0mc0i8zPO<}n0s4Rxnf%71IW}rK*xM%EL(Gr
zd#RnE`S&hryWK@(vjdXf*+a#Kb1G6cnDNO9Az_X<>R^ww;r6(-z@Bk8>jDFGu)pjQ
z=giOJu>M&DzEi`Aay8zdRKqDxH5klQhi|MKQiEKOJT4GpCI&(-F%VPVus?ax0|e%H
zp#Lo1GKjs8=?)64?a892`Uz;cAA^(UQQW<C6gu-RqFaCos&WjV;i!)SZ^j_(;W*~Q
zy>*$~%gfwMW{NTEOvacu(HJ>=#-G{G25ss3h&rjyIX*q^P11wAr7n)|(7~S#S_tvJ
z1+9}e(CzPaEPZqh)f=v&?AkTtwco(?zc=B@z81BEw=p^ECPK^4VY{jtrn;%H&*BX3
z>T^!oNR`hS#;7Vbgzp+7&Mh(DBgwqb0BuCG4pU;Gg)2+7q0HWx;<}s2_Bw~4Ky_%W
zQA6r_?)`nKf;piocso!7ucjGeN2?LEi;OV*xG^?AGUDgn1W5}`am>yXf8Q~t^4CSO
znJ(V<*T>9EZA5Bn<IxapG}>yxnY}G-WAEUq<83s_v8P7z3MwKmqN?cvEFLM~m!$%N
zlNI3lOA$&Z<Z<P`96sEYL%}^c2>yB213A?5wd|8DB&+2i-6;<fyHhY|Rlu3mio7SU
z06z^yoUByDwEd?!yMCHqqo=WUz7lGhS-X0z42Pv>a9;ilQXi?nEr>CUq6)6{RfTE~
zHGDs$hB2SjFh%ba#uzbAbX*Pz&a#}@mW9XU%lN2y1v@r#)@y}6k_z;>^U(mFE(Uyd
zV(-iyL-cYmK&!q1wwyJ<COuu0Zn}&Vv$JUVpbndy3n(+bgjXxB;Fry1=Buw^__dob
zou-YECAvshs0XJ~J?xUv$M^_+c;06(PNXqTJ~4x)l?5c0Si>^h7GFNuqrC5Zj5K$`
z_*O?~mN{ZY%3b^_W1PR{K6bMwIWWiqL%x~g^KWxF?XrLpdzY^b=k3tb);Ozbhvb!w
z=uB?Jg<Y-a)zpp+GQZ*fq7&x1T_~^bM(_2yQ@Qg$<OK3|i+mFXrq^=ryb3RSR`EVD
ze+I=?Bfkgl0;@5XT+jml5AC>k^EX`kk0Xa$V@Utk7%I~rNMQqqkZ2o0lg^E$9m6M4
zp~7@3wwz6#Gv|^OcOZ{ByNH@z=+XXiQ@Xa!l6r6Sq<1TxQdFKVMa=gnFFSu4njK5u
zc4g3ux?IZBETOdATJkh)rOGA0C_?%lwH=obzc2I<n{W0MgR6TBv%^xNJ!*lN>9$TR
zxv@pqZ`&=-Kin@0We$pWdMCw}G+B}2#`*KU@npI171a!4PmA|!O6Z+UNk2Z*;f50G
zZC*jqFRJJx?<@3+uc3WYYAKAfQ4$3;B<9!9f77dJU3w*XFRh>xCY)7zUqWpIiz#hW
zAw?T}p?Z%`B>N+m^cLmNo-3U9zMf4J_I@Cl|FY<6zxR|``j)=9z9HLbnG~RyL5CyL
zDC&7Cjop(<BSb3kss%X>OeO!vDbzfW@z=Rz`ca-lp_WOs_i7TAI405ix+Ds|oJ{I#
zQs|j}D%Brme3QVq?Fs8$*;zESUoP$F_K{M}K2t<s0ohm;k>rvRdgfP39lgux+U^SK
zvS9CVP8At1s-cA7TGC){tl#`b+BcPVuDJJD>j>{xS#xGQzk^=R`$Z2O|BzL`e{^_j
zH&J0DAr=&O7wU^RPi5LuY^dlZ7Od+dyxseX1oj(SpBx~%q6UhZ;e*A`vqMBetdt1q
zGfYVD94>Afj1d0<Mv8IoM~UMlqs6(BF=Fw%vEr5QIH7QLyfB?JLFBxdAi9oE6eYP6
z#gWyM#6gQmA|q{*c+oOR{Omqi6laFw0DB%xu7+aW+E4^_gh0oV^_!o;_-}47+|P68
z2=fsI(t*&)48-TL(J;4)##ZBK%(@c|&)&R~@Q-_FR>$H9dwjR`VZEU-4B`A|YIuY|
zHzpXVcFg174?sx&7Z|i43})YW%kysn2B+}W(4_>(#B*_Xi#ZPMGe_7!bNI!Zp|I~u
z%sRoo=g3HG{}T@L3t_l8nf;r2j1v+<n6C*z+Q?vx-ye*t3pl%UA{2X0g`s#GYaL19
zkmH=TYi&3dvd_0`eK__o_W8k?UkC0YF8Pmr64JqV#aV9}8i)l$10XZXAEU<l;ot_&
z=P)<Zn(2#^lYEh>?}Ny$XE^Th3~H*+&`<6e3U!_#{Pi<<ukyjSR3D7i@WsJd&+(!4
zIrquFz@BG*Xf^c5?*#$4ydek<)j=rdj83o6KuF!^Er9{7ZOjV8(x(ydW>1m)Gv1GE
zi-R<CTDM0hL3UUQ-t|vIcCQQs_2ONC{%?`Mnvd}$-Y}JB@AkYrJfFk+q~kv0Tjv+h
zyFz?2E5^q8rI;I4h6}SR(EnB?CcUU)Uvf1b`_>@wN-ZKr)?v7P9n313+Zj-g{r&3E
zy|50xdUa6zQ;W3iwHSE52EX|Xmb<o!ecG&h8J1)7&r*EfUxEXUMes{3z~R!*(Ej=n
zgFEtI(4K?W^&hab=soUazD0gmCjN8Hz%}hOtd&i{j{kSvjCqCsy5ew>v-~N4qcNX#
z#GKO+C`}H-Bldc4$qUA(n?ab&uQ8b-f1I@Q<DRn@u&edOVj~|M9sUfZ4o}dtyEnee
zK8B;uBRDsBz(?90!`8V%!NwU`1Dvp)bIs3JJHSlV5j#{k`#kP0!jC!PmVyK1w%Vir
zKs!`h*)ZqNJJ2a+ki20EZGR*7wA$e!Uk_Bc5BJzLd|Y%1-zJ?$yf}+|eKiDiSL3f!
zMavUa_BE)%G5#9Xr)i>Ji!;s#d0=MvBShzEV6p2}_#2(Y@h)|IFg}aUno~%+BMZNg
zC$M14F|6Bn6w5jeGmbotyWOrLV1NNiM(g9jF4m^_-2YgCcO#W`;WFL`g<p+e@y7@f
zJ&jR1+8Cqzu#OR>kA^k+=pU!YnQa{mY}Ur8Ds7zStdVQWbzHQ%4&4LS5L9;+uhRJZ
zKl2)HM%_Td%bT!ixrLuLH(67>j5cfj9xf@O&z@8GaYqiO#V3(zbR0$LhhbuN5GHBY
z;k8*4uiCgj&qWK<4YgpN#X0PN%lK~}-@otOM4CDG-kns(k?*`uIh?(?cAP&_Q-Shy
z4FqYPXD`1IWOa-%@Q)#eR2iTmOCQ&6=^|KG7dqiOs0-rVs6M*rF-iy1ZM3oQj5cJ4
zYQxS+3;U*N!E@*x<Z(7ib>~&)dDO8vTN!VdcR74n0ZY{t;Bij@pT8>L`fhp7^UFcs
zQ4X{D+RfKo7de=1m&2xTS={d}i{q0|Kx5`fL>^+S*(?XQ{8QXDst5@OMbvdEV(hxp
zaKC&SPrECjAyNsZr<L(9L>W2>$~eP(&fP<-TlH4K@-d3gijc#db_KWxpF+&6Q~0)6
z9_8h-@b7aB&Gm=zUhe=Rl4UR?d_Pv(?MGFdK6=R*pwyoAsey)g-mH(i(gtX1(#IMv
zeGJ{gzMUyIu<XcLT<)fZ9-LACxLpk&&Cl{C-+6dFx`@9&uOXu87WOFdmgsOjB)-wZ
z>9xEILHbbj=hu&_0cJeo?wto_7{Iu_2Y2d}oU_O3?T)Yrcf_pojyQbIk$cha!b18k
z?ku*)x%rm7jcb8E`R367WR8^C7SJfL;Lj6F*qpM)@b$JZlXJlJO`kDr8EcDj?bzqr
z0cOLnX<QfgN_8Q1NH;RF=!9JTZ&()m#NsD>o(iwT)Pt4SUdmf^C#tcPG0&7yjaU`e
zf(u3M7$NryFTZu+#*f~VDAAv0tQkm!{(~u|U^xAW9ZjQWPo&B<Gk8mUHXXQdjHV`^
zpg|+9(U&K>B=ypmB$ipwvkSJAT=;}q?)sAa&`0#D^a)u9#L(mCX{1<}MXlYxP*O(~
z;du+i75<_w#y2~sN{Fb=?&5;Kq}X@7moPrfo8<nh#DyK}1TEYw4%u%PkIVLn$>9ft
z()5!e-#n7;&v{7-p|RvSKY={ElE`9d295H4Pd!iP)0mV3TA)%wkD|-y?eYrhJ*|=)
zJS%BVawSzV9+>yLf~E{*4>5aMcJt1HO>;5(q>IQjpn(4V`%F!nKhd9q`4saohdy+E
zpgCqAsM#Znw9Mbrl#B03{oq@=xZn*<{Pvps?K3E-e>zQL4~_6ip@hXLv_vw6iW-th
zH7=RHolmCmJ(H>0E{PUOC(-TVM7sDYk=i~bQXK#H&d4OX#=K6i&ne{SmPS%#85C&w
zhJIXoPvg|GDQa#WZFm1jkK#VlFV_MZx~+(GGmA-gQ7Qd3ETc2o<s?0ew>)jCNM&RV
zrN-CNTFZJ`c%zZFYBy7m^H*As)<(U1ekWxa-mUWgMc2ptB_F$gG~A+_NXeEEQzrEg
zM=nST`^282YgBLX>U<yZBc-o+KEA)mximm*OCBgvh7J~WCx(bs?ovYf(@-&^&v4;3
zcZ3+WX{4C4dz3KWGFq%&Fh=a^HCFV=94jjBj1yPKjTe`mj2GS$I8UxSK`hCdAQbyg
z6c-my<Sna-Vn5@M5aUp!9t_36(V<9T?&iN^+|ip7j2^sOxK}0!V>|;9`YQlE_XS|f
zW`8s&`JvtP1(vz<+1kbzSsp%k9RCdEKc8aHfKVidgt3<-6tcHMu>XEAB+UYFF4T`Z
zKA+=H<1@@V@(^cx+(XJ}dyF4tgWl~Hh`wuvUyHdD!pfArs;0;hW;kbK4ijs01crsf
zn|GiRkB49(_tZQ&7K|P2ossMa!V|_RrN4qPe_ALcc_+%`XBZA#3rG6%aL!b6pUI_g
zOp)X)lS&xmVnU%*AA-~l?zyQ8!o2iA6h91L-P#{#)cv5t{+b8-pJTR>FZRCj!4*j#
z$R2)%Gl@_6zVQ?qQct1X^#pInKgD|Gr%27_UuV2iCE>$eqdxGu#roCO=WreQ0;@i~
zfa88YoR8%V0oJb?cKE}UHLy$z?&j*vne?s@yx9?sKfaL&XK(lYRk6Gg9FGm$d*{3}
z3A@&&pkzrJ^#03$^Gxo|pZ*r=Q{TgM;s+Rw;OxC*9#E5yfS6D4*8YO4^9tdYQiOq<
zOK``BH;!7%kUORV(o-v;(5DKGudBEpu^Q`3s-Zc*2Gs{^5VpStd#2ZL&sjCZj%u86
zuR=>kC1>g@VB=qo+2_hIre7%*TNLATTOq<W6kwyyXUu&55yvy~aIY{2%WFPhM8kX7
z)V+l^=l^9sXJC6~8lt08VC<cQd$zBzPCFh4Shw`O7K0@gQHYM@-o9?(SgaHZY36N;
zm`7V76^QpS{<wbD4>sdoU~IZC;xv5V+xHoaoq1z$oHt@QD>eM<BLr=H$e9;+G^M!0
z=7S4DWbb2?-Ccb0bbx)8J?EGlFfh}BbMFrLoM4YV>3kOTwnfxo8>mNGp;*%bX~w+s
z%vqLj4}HWkhmv$z7gg_g^XTstXxzAn9_kuMj#Wohhbm@F;&b~m74)}NK@OkccjT+V
z+mP?Mu2&#8Lj|YysW7jnii7<1B}wY2QB%i;S1K4^CJ(#%lUQSOoO5o6VYT8g9$6j6
z^-D)MpU?mA5`DaM(!;nqUEF4D6VO)|hZ(oz{W8Ra*GA}5WrQzvM);U(#C(ep_Lk~n
z)f|2FG|)riI34K9-a*DN)^L{J!2CVe@he&r(P>w)dcjpR^t^(gH&<~mSQEPMHR0iQ
z1M9|JguRpoO6MxWcbXzje3R#1NjZ3ip2XW7C-CUh5m+c5#KPVO`Mhx%32L|Dyi5zd
zFW<x`o6E3OybP_DOPo=<gxt?pu(gYKDh*W;{Q3+YTAV>i_cJITqJ;88jBgfRK(}c|
zu>WCzgOl|Uk*15eTbQG2*1=Xo9qirA_$HV0UITUT%1j%Tr?fF~xHe2k3->$kpz!G(
z^ijNn)0V2ZGC>vFR26XRwjxfARp6~K1)Nt@z~Si%IJ8wB-(SihGM-bjZ70#|+)0$t
zQOw<Q4102qVe|3h2<UkNmaQi+W0Wk)nB#e}^AuK?Dj;RMB75KzncG#wzKN&NBkMFj
zuS)nhn7fVFC?lM&9|cOhho_9<1+07d$syH84z?|F__+5JG7IJLY1A?1kY%xW)(Pyk
zKa9SNZ{A1C;KAqpNPD;sIZOA#sPA6x6Vk`|7WNvw)kn-5eJnlAShXi}cfIrw*M1!`
z>Q}K{T@@kr%on|9EIo$zK!Y_Pr*naK)~~`Y=Q<*XX(8gNE{?eBA!mp_zy288=4j*a
zXKh#vHAL@$CeSr7N91f<+?imHihKvWRC2_cNxc8C!X6vh7d}VJ5rxsVs0p^gVkZln
z2r$Qjr{<X7W{%zd7Fb(giTx+6;lX(=2OE3zTY4APb6mLBpbfuIbRa7AFZ(S2;TL0^
zud}<@OaB+It$)M#=}$b(uYkk(3Ov-Tz^Ey_mzP<Eck;Cu8qt7*Cfr}y&<;D<Ux*Fv
z!u*aNG}os$b!+cOhvNs4h2b#jD>;hVJjPQQZ&k%8$kBu!3iPHym1HfiP@Rf4d8{#_
zN4~e|V!1JCds@@UmG>xN{yok%KOtGgD4K4ON)2nXXi8x^DLB2R<B2WQjdzmnbI+b}
zdp9x6vAZy;=ppR$B!zzQSfSKox`5VN(SPeEQLu8mSZ2x`j_p+LAr2;UuQ2LgA4Q|)
zz9jehIGXVyktP_jzQaD>H^XwMoxQ;Q3=62~YY`>cl+c|ArR2_iGGXVr%Vu{Oy~r;m
ztwE)v+^vM7VvDHcXdzMY7aD!wGi{0HPMiO6bk<>AC0-L(#8wmwySuyQTnrQtknU~~
zSxK=Kr9r?1m2LzTNs;c5kS-Anu)Ev${N6u&+<hK*7j^yKxpQVdGrE;_ZEU9OQH><!
zQ%~n}?@_<{Ix;V>rLh?`G$EpzcAvgW4XRbtJhYO|C707Y$1+;rS4!POOKDa`2{nb6
z(3b-xWU!@#ru{7@ql{vT-&;%$3yR5Gw3sBu6w@HXVk&x9Oi>Lbykk{H#nt7s>P;mb
z+fz;Y5_NQ<p`NzwZ6b%_7Mk4AMpH98D9QXjePhnxWy5Yt&g&ry=JV`5_?Rv~e?q#B
zePlkBZ&kg|Nuuf{nU%k$!`vG`^_yADD?ia{r!O?C{2Sey_mdRQ|0d^Q|L9<OKVjMk
z5y4brfN&^MRPcK@P^gp~EJ*tg5ym|jDlA?+T(EW;Ap~WQ6pVj~3A*b>3#0dl3nBqy
zggZ%Ng}XW9gx<XILi?==Lf8cfLBnFAAUAK4@Var5(D2`6A>qqpVWagFL9=>_&^K|a
z(4ab17=C!F;Nv@0u<=ZXBj2UE_(tIWnSBM$ro3FrELYAC9DKzr8176zJ$wrd1<Ckt
zP%>{F+(f4=Gs11I<GE`R<h&9g6><$@a}pr=lDC>g;&EvZ`y<=aA+M4KLrcyj@U3CY
z^6NOnI>=-|9O}13ac}q%>J869vY1)rmwgc8<c*^>Uhr`9Lf-)|SaP>@(>pJ;9`S*Z
zz7MRQrQ_n-G`Qa3JdbKBz79=AT_5+k|E0h%E)}c4@h$9rI@)*_(`zkrXt@hM#))$h
zTQcEwF9Q(^GuW$5M_(TEipx{+_ht$boo-?EvSbYV!Z)ds>yTAWf|Xt(-g#WZfs6!{
zKIILgx$*dUJPu<2n6;7;i|0YHaN+NBzQ<y|WgPrJ#Bp{x9xrz!AdYoTOX)S{W+Y+@
z-?nTLk`NG@ghw-3<4n4N$W1pfe;>1SUfjaZid6PJ(y?a2ZLBf5gHfJ2IC3eEdC<&g
zc4Lj>Py#jMGQ=oV;Lp}7Tv$_$lzFxI%|1@@-+FZPG~#4NGm4J2qHlFOj9WU;Y;qss
ztz8(gl6PnR>%kqrhxp+42(N7(W67c?kSltE{%d=&*rgYBF}+w2*^4x%UhZu4VnpN<
zB>j90TJ{)<n;)Td&O=On-Hj;b_Klv?h2=q===$7_fm_;e{!j}PVwzA=+Q9kCd)V{3
z78PHsA@jQmXP5y${#O}Xzm(wH8_u3TEx_MLdHCF$1I5pGFnD|x7Ra-wb~zoD%va1l
zlER%5<|fA7gxBsH@cxkmySPN0Q@)0Z&+)K78wcZAv1p2ohWFGc3}v?Kx1KQGmki~c
zDD&r-)AGOU@Res55x4jP^wiE{z1(@;D?N{WpUz?Ansa!@pVd1#t6O{1kGnZ%(8`_K
zeak$Npx_3%=2Pgqe*!a(GD9WW8Ku1ru#dFG6u!Ucj<&`QSYgSMJ$QJQIk$$EIGAFI
zrx|-7H_r-VIg>Grw~Vw*Eb+x-50XajMfM15OdDnmoibA#&|)uPwmzx?H8HhA16j!$
zkQCE|S&JsNDC?oe`xGq4pTZ07K+P8A9ES2q486r2qF3y340nfZgFDtYxnp0bJNN$h
z`?YR}is!6~6#p}ZCphzP6!Nvq>@_@yH}TGh?mmEWeiqspJ7W4s2iTo*LR_#jW@I>H
z!D(ldPv4J2X?uCE+8pDmO)${Q2r*d(SbkU^<7N1Uwowb)eyMT>SPkPI+GB9XVVq?@
zXqUMI@3%PMaI-zOM%yD}i9HIV?4W;RFTxjD;FE$W{_(v?$;AY#)|=qs6BE9T8ROVl
zBc$!%Z{>W;F}1=%Nf)FQoWvT%lbG@71dbQ*%}nzsCLB0~@Hi*Fp*SE8c2MR!RIQIK
z(l^>-Oet?Au)ehC4TW?6jBwP=7+n%3aF#UTUbZnrXBy-7LPHoh>tW{x9UNutI5AfX
zsealBAI5+0r-KV~b&=Vsi%rRTcvYp3f=WZ=u}+c@H$}l=Q{0F$#r{lFd>m;8W%hAC
zR+wW_qdA&t%n=!`k6nC!DtTapd{0B(U^T?DE<=oaYy>w?ZKPIf<AS~pWS{H8<}7<(
zC)FUGq>8j!6&ydUjQ4L9QMo`72ZML>KCco6Cz!*2{wX~E&G{+rS$U+mAm;~n`dL?B
zv$H_OMN7QfVFfYHF+K19AEqzZkARDguva*U7CT3{sW?Hr#0e`j4xp&_AWVu6VW#s5
zRHe8gX`DM=Px8d6Z(ax<%Ub2*88lY;!{G2)WDPrqPwZbR4?Tw*)<3h(2f+4(FI-Of
zpvl4qiLpM|vd<TLuAM>gc7KdmbCy3R&g19vi+Ejo1+PxmVv*Tb=*j+u#3N>e=KqDN
z@?X5!^$Rw_Plzx2j-f}Mz&NuDyN*1@nH5iPkMn?$4bQN9+bblldxxkJ=4mbe0XqB-
z{~QKT`ryGdnHf64GsUR?+A%cVWIVN0Orf0**3-BRh7|bRgw}4eB57%R3haN31{VKE
z^JCQMPPhiOn;MgTf-|Yv2awvLSPI;In{E#)q8&12)IPnEWX!8bPUSrr7V{3|&c9T8
zub=Q`lZbG7$p9gdpR=oZ@A|mK9AV(XMS^R=GQpubhrS+9r~KGd3erub7_W3vXKrAh
zXEqIX%A>~FMf5JNjGk28rCnd{(X)NcG`_u!7Atm=y=WJiE_y)2?mr;y#oZ)7vzu-d
zKcFoWA5g-QE(#pZ9qft@s@U63>wmP8;h`4t?royJ<&EU3Ur+nr*U_XOwUjWhmOKX3
zPzd+E&EHg!PkSW=WL8k6TRHXZE~BYyOX-?H3EgTgrk8HT<f&dv<5w0_y?8PCeJ>)n
zjv}&6C?aF)A~KLHBB?D!G~B(2#!oD!^gqQkYHKO|=q;nw&nxKnhPzaeSwm4<?~!O~
z1KECRqQJi`G=YCkOTC!EBXOTP{kzEe+XGsz*Fy%JiPodXbnNaE8pRrC$p-GZkAFdD
zhP|SWv2R#szN1SzA4u}TC+h3`LK!Q*Q})@Pbno|Xigox$R^R&xy_-dZKPLwWz2%}p
ziP#|Fw&q}=A#{i^^2ty^b>499Sd9>7a~@|@otV)0ZIm!zqPQTrY>cpE{aC?w)i^<z
zK3<UeHC`yLm>>*wWgg(ti9$okM8QvXl5jG6l5l9!WI@wnvLF)1Tkj>41=p6zg3tbR
z=H92{{FHPQHK*b8ku<zxeG+yc729g~y?R0l?x^rx#hp1(<u|eH*9{z9b^}Ig*HQ3a
z61tD`R&(GrG^8cKr6C?&jd3tljzz$nR2Xh#ox?th;n^Fg7P*cA`>w(JVJwW6M?uQ+
z5)REe$G)c@M)mUn<h=3nycf=~w&}KHjkC`Sh5=r9HQF0Xw))_>FYB8LX^`Vw@R6OV
zyuFixA$Qro;XJAr`!EMo6VNG@hUMNFF#5+k-&-@GwJQ@pCbEwcmjQ{V>AV4&j;JeX
zn84hdDb^`?;(8O_!VUOw@4IC2b#%!jV!bi@HeLz5-w=-*<#E{mHWodrVzDYR1|Owk
zaB+SNjtVh&<rag+PcblciACV7IP9*E!yeCg9NwOQoSzBEh`fgCDT$CYOvLF=iCFL`
z2}k~1$7bBbygSM0V>a>(k2L&cPJTw|ZA|XCgJ5R8y=%zBvit&kiY<cEnG%>fmSL7&
z1w=T{a(YoUTt?L5=!-h!=GEi0M<XP*HRI{a7MwW6yTKwIkg(~*ublg^e&2=o!QBw+
z*MkY|J=pEfx!_rkF!}5wSam#t`R_+C`p(;uHIMN4@FTeWdk9;@he!$PLDBVYIE6pJ
z8~ZMJ%)gJKf(~rm+K!yWR($NwY%tj-n4fAu{LOptZREVo+iJWPy^Egl%<`RGjw4G-
zk+Q0odv%3i7Y3R;a<NgJdvLZ{DB;bap&k5t!5Vrd=WR0Bw^0tii4B`?FjFlF_N)!N
zR$as5&UiSouG~2(7NJSec(FPPVmBjjX=*q^PlO`2DHzTZnAf-AGEOfE#O>)95#@d!
z$wB8hugU$9KWFju$XQt33c$jf{#bg<4?g306No>{f2nvOtj8S_WnIzS=Yr8v#}PmI
zFl5XQ;O$d;q-5{Mpz3|_5%%J(_8z3LhM8h%fxTlb`2Ph9C}&zg>jg7Yn499k&%h8F
z3w+#RiB~Rr@WXvC0#o%NQ=*U0ZTc9xP!9ze8i<*tfvF=ku=Ox&94SrA;d@YTKfX;|
zalxh$F5K5YiO-%V@aX1oG<qM!ax)j);``3H3U@>wbZ0HVnN)E%1V{6oLvY2gtw-_p
zj3WXg9Pm2Y5tj}+V!NXQew=hb<6Q@+o!$@sr2i28pA%+<IpIRN6V|%eV5zqi_DnFt
zy$eP-J=XvwbM>HXrHQH^s_@;biot%W$ey8v1AlB_Z*vGMSZ~xO+oH9IeVZnG9KT?X
z=fQR;_qD^g2JTX?H%I<bQ{=jvV1|kb4v#lMP$K72^o)7?)CgNeIpZs01RV!s{2jx-
z3G;)8CR<>4E;EF?jzMNPGy2M%Vb<k{4K)tP-)@Kccear7;;hX&Ta2yQi{-2(_kJ*h
z#5Wyi&Njxf0VeQUZi1K3jgdIf7@wx=<NOv4c<8C&O^Ygm9;;%pr8Xpcw4r9CgIynV
zc;7`A@3i%>#!4RtZfU}Dlop)(OyDhR3TItYtRqvzbedw6V1_TTW+<>U$1P)X<{{`H
zsg}JV1w+g><4li@A*V+*aZ^c)-(j?|^_VuE&(VhTU`<@pRfqF2HPnWw!XQHh10N{C
zYL+6Df)$uupaA1XyO~F<j(L~%!ny1ehI0SSo$qyvT2An;izDBQ%ptMCf^#&MxGQc2
zDb9X<E40CIZwExhI>F7{5%;B?a3YUyo2+kI{SPu*`4Bvqdo*!1`(!$9P&DLjeu5X2
zntYhGeFjBN{>&r|fL#-7nm*P!?PsB45P<!hSKqVW7a2-ENFL*Z_mVysB<>4O%QL8-
z<d1&S_&#NI9tw3AvBf<Ig`caTaq}xyH~hl7`rkNn<PQRQdqdRbCrm||4?4ORz0qx`
zw!4oPryij8%Ofmb*o$+spD_>rB`hl5BHiv2KDK_xIFWyd`YuAPc7vE_H<V6VkE9B5
zacX96&cSL4Qq$W&F_8xJQp<$)iSD6qDz@~c*@S*s?54`$iqyVYoy>THGiHS&H7om(
zn|eG=cDq9!rG+%RrIZ$XRgmtqO0p__PY>?=q@iYi>1b9z;o8^!f^3k8uwjvi@ch9a
zvR(0k+C?5xV0kskozEeQs5FvSO`#n56lOH@UyEl_Lt_@LY0IGjb_KK}k#9Zfm2~AN
z-*(0|kfk1HeeShV$i;S=SldBUE_Bk5ubq@6a-R(II_diIPRcdzASL5=(%RfcZ@#yX
zQB*TM+tfr=`3=;zw4PR8tfOGbT2feFLuQiI^kCavdb+iWW>6(5ud1NK)4229zl`#q
zme8YT#WZk6G5KFCqFO4Vxr2*H;zc1XeNsr`p9@L3e-RCRSxA-F3MtICkR}=xl1gYH
z<#T>$Bx@W|%@PuOT}p?V%gOOyB}u4PQ($^6%@wWxf30<4Su=g^-%2l2+9+X12fZum
zB=OZ<^p?9-hs1j5f$u}|nfjQVN}tecw?3L;@Qj|yy`V?(uec}whAy9cM-j;%Xu+FL
zv}?mx8s_(%<lp?H3e7(>x$+-9ozY)dY$zg(2pb^W=@u1KW(*RZ8x9s+LWT$f+lC5i
zBZmtsBu5CURwD%~AKr0@9wkgk87<6C6Bqu)j}apL#|mPW<Ag)Y#tX}O#|vTJ69k=U
z62jnc3Bg=!q9D0%qF|RXQRsd>QE(QYBrKUTNifq(M<eHMrVik~R8|^f^f}A&G!<(!
zQn8(RF4Dhl!Dz`XJd$SKo%c-yHQzw|xEuI~>uAtP!e0ACL@<kN(DejV<;U}TKpd8g
zkHyYiQTU-4hT?5EP{z;nSCf;F91;(U#nG5DJCrwZE}??oA<|8K_zl?;JL=rw;p7F8
zX<n#P_d=q97o6n1uw23$X)}E=i~BX9|5BlFB^6d{SmX4j;KsxhSgPK_)v$O-izlF9
zQzDv}-?yYC10SYj0*lzYVXYJCnt|rI>`Q%(#jF{z7?BhM&ELsTn{yKvO0OeBHwoV*
z6Y+57HN=`FAl^3~388Vg857I7L;kT3qIst?8s{6LkdzpO6@^i_G&Gu79nnx69s}bW
z+$+{)ZvKck1Qf=>**+fioEiBMoPd=huc4ST$+{7V@Xtv??ceLDJj~pj)wghXDsMx~
z<on>_OgJsi!om63I5ajFso(S9*-?PwSw;99SOPjwhICc-ah6qK@xW?)XsJO%XdO1|
z*7L?*BhJM&A#+s=dg5F0cThX#@9u#2p-wn^-iP*~E~M^yfZ0EpW$V=q%}?F<w5bRC
z^m;H|s|VMY_Mm@PHw-PhVN&&g`_K<iG_MOAXWfVWpAJOjwqvnL8$8~$K-#<+11lS$
zKBWOk#`mCpwH8{uwPE+M3M0l<a<`%!Gh|C)rC*H2`wMaRL_Va?=0Y`=_l}CP@a|bA
zTKR_f$TSUwIVrfl;1>4B-bBRe8yHcO1d08Lm_Gg*Qn&-2Z5{`|!R(hsN8>wl<y49z
z@Ml{%Tr)$FJR<}?$FAaC<z>YD4#Z}OOZ@qK0iz1e;ch<jU>BXm^ppT>9nCC|jsDPC
z?uVCOc(>Wd2N7LfDDL)v-4!=x7xAs2_as7sj^b+cLEM?>gc%)nSXcEQ(t4~Bzl*uG
z>Xw++-vSv6%<(?e4B8=P%*HjtpDHuVIc5gEHJnS8Fvrtx=J0;PjNDqzviRvE(Nmw9
zbb6>R(7=%@b$nA+N05p-ni|yc=!OQCKh?pr0QPL=xxjkfNoMArz`K&;;PGUb&pV2~
zbG(18X9wxk`#F2&j48UVsDJE=kU&@T?s7%d5LZY<Iq>b(0bQdVv5xOaCB^pm)ohDj
zrT@Wu$X*mbwZ`lP)`+~a553j<@m6Uc5;gZQ|JD=_DhyEQql=?YwRrDD9fuoL@n@PU
z-#t~aWS$mAf3V^%jy0y}*dkui2HI)+urkOFCVab#KV^r}2kh`D*%EKn%=lfx6t?W!
z?BRRSvhT*2f01>^b0e6j7-4dcAs#+5M8sMnB>XbMeM4hh4`S_e!~#J(j^QciqK6)H
z=8mHyYGWLbde|OM_i>iEzb%F>vf&)mURd<*#;HJ6{M(|0kw4^-_go%_y5#YTzj;Y2
z^8d?<%)e5^w+IzT6{%oCv?^pbY2#O_HvE_AATC=6c8a=i8=!|pF|03jH1ShO1B0wp
z5&usK!={*GTcHU?N1Nc@AT#_uYzF&Z+#mmKhD22zoM2+?Iu(7yZ85}7OI^Ggp^2C`
zn(TFHp?9%19`<SB%3}@I{pvXFtcI|Qs;G-r!Mw%FxF@2BEOQ0CTPlx)gYtMWUj<8V
zYvW3b7Ct)I!<+X)dpaHQlz;AJIcDg{Hp8MmGe}S6{EyTgeC0e>pt(Kj@*Hql)e*;D
zI%4!0C(aWbz)snNn6czA^A3(+^6b;l{^$yWEAG7g;K^A@Z%niD#ou#hup}w~`h5Wi
zSaKHIlKh~Z<d0Q9&#<2L#g`e(&q?%#&TVg0h5BI1m@~+G<j0>u0T?0A9LJK2P}v`Z
z&gMw&-F-#coS!&2>o*o0_zfk~7jTj4!<_jwIK#Vn7TK*Zv+jg)K^KnjyHfe|Cm5sk
z6w^k$gt+h)3O_y~Rp&c@=KqEANfCPXU?3UNP>Q}Wg2s;-P4+Lx(BP92WSzc&yn^*<
zJ9{*T@-3)2Q<J`S?xdH=(iHVoo~~7^(57R$bW(ml_09+;$3amvzdnYvS4LAoaWoCs
z8bkAc#*y8WC~}$lowhh~e=513Ftwq-aCxnWu-UY~P%ZbHRGz)361yIXUr|km3$y7>
zSPBIOCzEqmG7XAPp#+(9-jBFVG%cIbnU8!ZsEB6vl+nH^cS-wMElnTXKwll2$b3r+
zxp=fvtpV?P^Tzk_&uuimyp5Ktw2@tSE3*Y#D2z8T$DC=Rr;3gA?r%MXhTfy+OY3Mu
zeho<+ucn`NcWLXPDjIgGk{)_iaHhAMI(_*b<xxsQkC%}7kz$&8u87cFNEvd4B>lF4
z_GA>$iO>S-NG_li?+eIgWg*R9TgdFc0vd3ofKJ&KkaR5nm`#P$GpmTE>T%EgNeO*S
z<Gy%)1uf(~hbK#FXum-n^*Geib@N85Slvu`)Ixm++Q{=8bCh}0s-ovUP0@TnNp0OU
z$NC}76@N_g>Yh+_L?2!8d`5{bF9<%b=t0yQvafnaB7Z(mEPW<j&#yGF>pNL*`b8_k
z{*a?+KVb;<7sL*T2x&J52-BX63Z^p$30~~ucz6#HB(sJJZjXiu(?v%JRntcbTULq*
z+cu06f>)0gmdp|tp7$RkJZ~B!7zB<Lf@Q}E<6e#v)Q*f7{(T=WG+0a!E*4G@rVo`6
z)V4?nQ_UoV{Z0}>+3s}ItVzeG*J+3dPs6G0oKLCd&eO(JL|;t7^FHnY@%^CV-%VVX
zzlpK8Zs2o2=Eu*!4k^x>%rxU{t?xCQh)uxVw0P9;J>k=f7(AL01*Rckk5~{cL<GR*
zlrPrv{i#dF6M+^UsNh^@x12jN8ohDI!~^Mv+!3qpj>lKsVYSBt(|UQwrPLGc`@Hd7
zjP=bZ?%2FfWkzW#v?nsxRz3wE{@#KNzvHK_i-+G0)+F53_Fj~M4VnB5?VpJYV>5B!
zeFi#n)1gH1$SjXV<Aw-qFp6Lggtr!x;&})7I)3FPqL?!_9{=L8)+U}=YH?7w7>j+U
zm~m?#jlCzMc(*DN?MjhYJtq=BH%7wABNEzwBY6ur3QIRcqw6t0htI}f?Uq<%yo|+Z
zX1VFkibwaYctmk7IZ5*x)EyELcAGm{+|ioRaT8B(+=72_Dn@c=eYyv09EU6%XBOb_
zwYhK_nUAZz+)2wW!ij(q)SH%JE9Y@W{Hw&I(z}@8QiJf7bx>@&henGAyz6U(k4iHV
z;#y$yq!r^?=ky!dftaryxLDMQV>b6O|Lc96FzCX?$S%Ydbzyd17w%l>g5l0CXgA)+
zcHGC<Q=OO;+kuY>?Kt7zh9<36EFav07hz4XpVbJQt4AvLxt49OMVn_eVoRzZ`>g^i
z=a%EFd@1bq7o*9q5M7D+5UI%Jor`Sz7QF+lrMKZ>z*;jn4bwUMWG8bAb(uG@Zv71e
zmn4Do6Y=3Udw@~#XxtTtN$+Fu<a{*dFxzH%c?3$;!Xefkip#1Y7+rD|t_zuK;}V46
z@=N&I?;=`vpNDPFStQp7pg}wUe{B4ramx?kRcD}Z+ZP&#eK2mgH$IwrBIvL?o@l#b
z0lz<={&*Zi?jPojM`!dFIADLVEv7Z_bCo$K-`4TwBKtMRS<7g(nnHn}<AZ;2w)C+H
z_q+L<tqF!`n;`EMXLZxfP^)H$jJ0~WV6KlZyY&&2se?+^4vLLxaIjK?{BbpiO;pFH
zNg7Zy)4{>jb_hLn5~C-aK;!k}c=q=gj$|D|=vGJW8#^F#jvXex`wxzs**)?75L&-8
z$41H(SLSj@c%UmjuHoln^nZAo#aoctoPqwdA5ObC_hR-Rnz{4y#BUEaPBh1eZRVKm
zYmVR1mI%q`9<&~NAK&!h_*n;!B(-?&LLDLzs?bPL!JDHh%s)|qm4hyJ-?7Bz?)}Kb
zKJ4d?^pA;qk!NLxMr%8acxs8o-n_pZYKl0{>Yi*f<{YpwBHfKqWMGWTl}4Dk$_T*`
zhFFqeh~a+?nWJfh>6~L3Xl9IX_H%+$&C$5z2#U!WCALmDc95Up`|TlXYlq<r_d__t
zx=c+S>5FzlwpS73N)_S%T^>uH%A>bk9wu$_$Q2Z!s;dam1B&Q3P#IHPR4`Ug6@{&;
zkU6Oh!|&QSby^1#hU>z<T^%h=YDj;oj{ZrkGc;8Yc})p!r<K5KEbx$HjmDYfeNUN_
z6K{qgtY!biY2$Z?K2j&^;{xY+#+YbeUXVJDd{jqZKh{^`a+nsSfhV6d&`(nxFyMc`
zjVj!<Rj{c>35g23(QlzVen!h7O>7rpSI9CqQx=Yj8feheL&R-ITyS$lm!~ld`R4WM
z1bbmC%^<diZ-}OrF#WrSH8pbu&+^@Liz9No9l^jI$Z>Z3{#<9QOmxPBCx@YxbP6|)
zvHpGThEVR?**kmVl)Mi-p8DYS0AF}m`odT13?5tfVfF;(IKE@PBKw)K{e5s|p*O@t
zz0q$p->I(q@?M@FdMg9)cK&(vUB8GJ+aQdN4d(xBIEFdJ!qOuN<785BKR**sdaDp&
zQH=u*^_VBwgsD4QG459fobtQj`{xm=9Q#nS`vtTvzhNfzM~uz=hQG0YU@<_1^~pf`
zb9D%{eHc!j%SVyH3vv3WBSHK8Hqb^JNvhP4q5*AkbmNL79qrgc8L2yHwf8Q1d|#3F
zEz+RGBt5FJilFa3ku<_NirQ);X^vAA?@>k3{@iHV7#zv$oCwm|7*0`Y{e;rO{=%H$
z?A?s(FUY0+qPtJu)87Zo48BrEWrNeHfBa2yJAQ*oXWgX7pOb0H;Z$0;ErU`NvM4z%
zhqfLlpl|z1Xq9ROt+;=e4z$#g*V=k=9neT~G@7V;b2Du!Zl;}InrUlSGbx^Irod-S
zv||Qqpp}i}GN*xl{k=y!E9&U%$y)lkw1&p?+@-g-s%S$(CG}6Ops>Pn(r5PYt*%l!
z{;Y%oIXm<2TM?}vT0|wQ3hBhj0^0mNpGIBDCqILHs@s)M5{~&a=XE}FoD1loe*xW?
zS3n6l`SivnpB4t^lj7<EQk+^ylU0f+xu%#ld6kmDI}Z0fE6I{OVQX@0C_Ar?!lUXb
zf%C>k#y3-GVhd@nXrt_F?bJTAlY)Zp6J|c3nzU{T(|JfV_A$jgd_s+deWaZFj3U#x
zk6-eNDj&X~bHm=#(=8vV?Br+8=zOKv$v<ee<1bp+@P{HM^%FkGbJqKWh%hT<fFSi)
zR9GTDNNA(M!dmko!b}g|W{(^uyh<M~m=uf<A`3<e?&)H}kC0J9|3jmNJ3GV$J&`ej
zaBGavxO=P+-#Jz=(i|t;DjX-6P8u)B8;=*{E{zvtv&IV>s>Ta{q|z}*EFDYurt&i?
z0nY4SjM|cbW3N-7?3RMp?YFp(oy`8|P3R8g{Aa)oSbe<?ck$~eUBo>j&YKi0)raaI
zecWO1wxORf7F;ob@_aLx^5<8GS|DDqPvg4V7XzZaP_v5nbyYo(kmHVlf;+r>+#sys
z=gn_7<Z~WzgNHi~ZS#O}sRwT^dBVq%_vi-s;N2@9JZ0S_$KFl9if}xc$SiaF3wT;}
z1w+1u;K7kl%v%wMUeOFF$7G=5V+IBc$;7yq8Mv36j*@^d?p%dKYh5_TtqO;7R|Iz4
zj)e8{c)S-$M9b)SG!2W#Mwd9;iHt>ye++j$qL~jIg{BFS*uj}rp)ec+0>UvXJ{<Hb
z9Q%((V4_4M;tF`{XkQe($45gWKN?poVz7NE`$P$`u;X3pN1RJG5KVx_tZVpS$s2j?
zN!WjopX=NspFKVW?|-C1xhEZBg_$V2nuTR{*{Io;%d@!o(CRM0$iyNXbS#0zrZQ~&
zS<cK`-f}X(i}<139gnQVrG@t(9##*Re+~G?o83t#n(^{-3#MOgMa>b;%WP@K(!O@i
zS#`j;t^?ynccNo?Csr=*L__~hWM}Y3xisf@(%RwsuZ{Q1+u*Ri6|d*E;K#Qnq^C4O
zTfG4*AK!zYK^<}`Yw&JfHE^&BF8LL(`cnq8b)`t&!#Al*g-9*PNA9y+3>lS!qZ{sE
z4|k<bU&=sieHxBRFjvurbKUoDqG$UJD3vBLYd?|urreS8iHAS`9xiB&VQ(ZFa;%Z_
z^O)bE8IHeCL-C9GeSz<<!o%tcCbk8kd2=AvT(}6^XXl}%bPi%~0?^&)k9VK_(74PG
zk=m?n%zSZ5)(5=<y)ie)6EBz-vTcSN8mFGds=+Qu`f?Q7Vuv7qkTbdJ_V_PxKPuYy
z!LWx}C;d2E^PY7J=UNRNIrEXvj5yAw4)1S_o9RaI*E7O~HAbxC4SBbTGp=(CV9=zC
z3+x^2Z08%)es$L4YB*S~io`vt7#5(4OS{yt!c!ful{B&am_Ftoa)A4zV=(@DjCV$l
zLEwBv)*A=(R@r0kd|UL`|A(d!8+e_!!8);{@ZcQE3@_&C+U~>ZjW+1h+RHmd+*y9i
z+X5CAI2><|120W+S;~_06z15*e!xclIVdkQ$AHgf7%pyxlkBC88K;kF$~q`KrHPUT
zH7u@FLBt{ztofkK+bGHqmZ+hO_1TeZ6TB}offVPCjP96Y|0HG<EA7GN7E9!`n&8k#
zQzXZkV6m++Zy*^%%H9|URgLjJ!U%`f89_$c5T{NUVso7#HjLvtm5vc)J&e%mXM|s(
z#(0rohM=zpAg=9%N%{^*Gq#7+9y_#8+>hU0`(QOs9cR)NINv9abM}fDW21mw3Ub`*
z+y#a2a#*G&4`(RAqjfi2w<|$AUm1tysN!ynDn!)OkjX!{G#Pd1994(J5jBj_RzqEd
z3KpJGf&MlXoSUbFN$d+fe5i(j?=;bqV1jccrl@x0uPaFxdV{s$GE$3kZ<@$Cpn<&;
z)G_;w0-Uxh;LLM5guL8^?z6J^*d@pQw+xPWNF!?JPFOsa!su<=k?Ow{hc3ur_I_FB
z3(4SirVRJyWFcX>3n`+C*rcWb8F4L`4AsY)=Y|+Eo`0Q`jCgm!2<5gGn4V|}`^|gt
zOwIw9S2^N(qa)h(JE7#06FeRtfN-BPSHBM7S=%ubi@2iQ$PKHWx^uU|8)qkTFDlvx
z&nkVOFxQuV@6MoXh9A<i_*T2k4==u-LGx-KRIl;C%duWqYUPbb8+;)q?}u9f%%=Wz
z4oiG5Vv=?cDyo>3IXfJS_Q%5BhxuPSQ!v#p6B^$u(fO|m-<9rRopmF^3|i3jqaDHP
z9-v?SLo7euiwQ@bL;wD3R40GneC9X&YWNLu?N6_6i&7=`s5WjMPLE8)=+Sa<+SG3x
z`EQ>=2fYNo&}^q~f7X*j@FtojCP^|!wo}s(8Ty$iPx4Jl6g5zhzKJVRa9Sjt{S!gE
z?}u~7HHsD`N72|H5p@531Wo=BP7}w4Q)EaO8Q6r8h*TIwZ3?4~En)QQL>QF`5ftYf
zL0yU|RP;5Gc8*A*!^^JIMAkUlnT>2_kwzxgne^@U9XchON1eY4X<TqACBLkogrd6?
zBVJ3Vdg@4aV?8}w)Iie<8^}+zkxc&au6I@g{oB|;8Q%5Om3EKz7SvICYAszqUqe}D
z)ugucE?xe?cc=#yG@z@TzPv1>wf)Lyh6J-~mX**AsbU(T!`jBakS<>+ppDh}<S;v*
zCWhtF0wIrd{^pYB%UpUqFOLGB=23Q5K8+h(Ko3mvNv$)F8hEo}_VGM&pOR1BLkj2|
z`#3e%iYQ98gk~(_JkRtBlA2jXQ)g7ubg^1edRj+?vCQ3JkE!)BGY0pzkk8XrT5r<M
zJutpkS=^^DKe|XMtebq)AJTKaRTX}HLLVRYkyPh1@_F=v3ckFex8vW^+->j4-1#E~
zrGKV?pI>>S;s-@v{zYHj{Gt5W{e(uv{({*N5#emi0Kuk4RQUCNps+@Ku<&=u5CL0<
z3S|nz1ZU;p!V<X=LfyuZg4Pr<q4~X-V4XEe(6Jva{G2Q<l%$IbTO`K_qi&B8UQHh>
zTskmTc$_^}NPat3=o>vwaGEtv_`rSXWFvh{dZ&++{sxeW(uZi99z+M|;^ky@jO(Wc
zQ72WH@_S;x*6Ub!>;`N9>xeZ>Lin;v2$C?ux;JWkXVyTdp*GIX(8DMzzU%X6zm>fS
zDv$1gQTATwO!t8U-<)jzd7yR)-v?scF<~w<G7{Zjoy!b!CpV=1FJIv5hIU<deD89{
z&q1EJGsYXozImfK(FecJp24*oKREMqEBlx~q`J<b<7N<^=Y>FYO(^~`-|fG}@p!~c
z=J}i{9`PXqVw<iYzAYF-6+=<KG8hq#!I;t%itrDim~<hM`5?@|mx*JaJr236<4_PC
zi?!J?*bop6kIhkd$U5j*ByT9EhT%Q){A*u_a)%}iA$!80_$iDx4a0d~Cjxb!B4Bwn
z5_bf?Up<P#c#mkjS{4JBM={V(<?Y97akx<#j|#q5-Q}KDir95@7vDhbnPgO$q`-a?
zGl#@7Fyvt-2H(g+sB<>13c0Ws$;a@r0_Nlw!F6j1a=w-#JhmLm<SVheuL|Av)u?(~
zgF(7=h)ubNd!KmsX?i1`uWZ7Nh0TcPp4I7+77Q|P#gx8QRPJm8xwqk7R2xQywc(C)
z8`_t)A+@j-J*!)>#jOSZIJfLy)r7#pMtlru;9gBV@<-l-Q(P^+EUm$R(RcA>cojUY
zD)2eGj9I&-xGqzSn{I^=&&$WZ54k9sn}c>u<|X;uMqFVAzWh#u=Qj2Q&$8$C`X-Dt
zZeSVTr2Z%+LZv4GlMcn>l|&r2@D5Z6ZxUn=iNfXU5%5w7=bl(73Xg{1)X-qy>=hJ`
zx{N~4KxBWrh-%{tP`rB%=F(?z_O(AGtNf7FbOx6`_+sr~U(6Nt!OllsXu0Z%`%68r
z{(>70aEHaY*9GU8J!|{!Fb=GC#?nIeOX_X0Hsn9VJhw(f>mE!FH%H4kQ`Ai|!3Eas
z-7AdnF3=EjW*R~<#Q;6q3~;JiAGW&;nA5|ol~wvEzN&@NX_`28MICcbt6>8F*|`*_
zg6X?eVDF}aKgOy^d8>wveH!@lRS!0E9dOm{C=TpB2D#XSSio5t*)Rtf9k9o0eOr8M
zvc`zdd{>%fgSGk=p!;T+yUH9}qAehk#r~3~6*6=7V8~HRn13+GUQshlH8Vw|lQHI`
zo8Ty!;Z3+1F1DE=#m5ZGu9`q*t0B78^>98(8?wb32$`yerR&(wIH!yc2Fh5erHr((
zTFBm|0|(B$$V(bS{Tp|Mm=`nbjuE=F*=IP;T$wIo+)6S*;401*4={!{vx1khH&ZWT
z4DX{xkeq1%uO|kWEoF#U_HkS|LsP|iaFu}(4wxFjX}u9NrHruRx+$cFJ7L^82dJvr
z<3oroo@@Sxv7PETwL=l#4HXc0SRP+i%AxMFEZX18Abqe5;3y5pMbeO&FO8AIWUxwY
z7w6IBu~Tz5*7NP@({eRzj!?sWM>RwqQp3*mYWR?&ioeM!$hYPG_IO3!iBp72w<6wt
zRbuYCCYEtVIf;8$!wPheBgy{TJ}pcy(u6MmI*lHsj<1_{qa{IsJxv7^ES7`jo?Xx~
z;;$)G8riZtp{Kcn?`%@M6(@znfbA%b+KNxxCGq^HH1AkSVb)|BWZslPv*<3&>Mzgr
zo!w|>%@pUOg{!u@ynmsK$2oeCkkv!b4^2p#v$jq!MxeYE0_QnGlY9RsOr4O|<%EJ(
zTLe`*;mDjrNNxDv&ZrB{KXyfWygLjhcq5)aH&vW{u*u5@^JQG2`GzxRnVztU^nxw?
zp!oX?3OIWc@5nxSn+H0Ryx^1RgUj#F@E&UbZdaYdGNX%_H8lw9{$0VHE1~GszlCsq
zAKN@T6Ax;#;9i~suc?(d6JLiPM;g$g(TqJUZTQT7aArsk{&1IS?S*I9JLENf_r8bT
zs;|&|`3p;Vmt2YWteVvZla%K$S`asq76gqZwV<)|CSW?PQJ+T-H!da#$+fg+^hVYl
zM4h}{HRRL|k`0liPdDUg*b;dvxGPH|X2}o@l%qlIA=FbEOykCflc;9|T{_CzB|My>
z8^b8mG>qODgwd<5VHCJ7jI@O?I%*$ADvKgW`@aavw75YDBJp&{DuM2gPNYw(uhYJ)
zn{@AZ3JpDyP79N7Q|q`Kau3KS<4;9&TD*)JDk>;;@Lj5JsHV{?YDsEw9bKxeqx-Y&
zk;saB)bhKIj$PnyziMe7_paU=)zD<?YWiq=moDt8qAunRCylBgebI8t8&O7;vrEa9
zyHIbq^R)G35&aG?q`}1noU_iS*OT(;l3^a1=I7EnA(y&3b4Vd5ht3_#p=Y@{Wc?tQ
z7R<;e!?b)laUqX>|IVcyV!1RyE|-RX&7}v=^62~c0unz~$c&+4TGCQN-6>^sFSvrH
zxL46D(`qVOUQ4p9advZ`D|uQYosVgv$8%cfV?-;7&*ELIYaL8VyiXT;y2#{wH??Uz
zq_c}2Q~HQr68`qloIkujG2|t!n(>;BY<)|L`*@2p>?2XfXZkql8(ACupe+f%=-bCX
zBs!y?@ItD;ptMIsQ1Tiee2NtnP%u!K+B8UTc{o^5emX>O;M{Rz>oCExV7Rb4WQ3q-
z!<?MuV$9GH6Y{-A3HPRt7DQu53%^<G*tm!b6I;ZEdlSb9cJgC{hc>Kxj*k)6pV32~
zvOd=7^83R&_SGBp;JHH&!{_KCYLYs-dQ{;d!?*cgDv*m*g83*#*fsDye2X$p>{o(`
zoifa3s^Yn%8f<23Aa}0@?@(&sc)12<@}J8I&rF!lvX^-w)~J5(fv`4ry#48p>1l2#
zS?Y@KYNv5fk~ej@FR5whipffD*nQazw-VeC>vI}&jF=Pgzy-yhPat9CNyJ}2iTdrX
z?8~}h{x&y!J@1JHKRpp3@b;XXKe7${&^CrQ=J-zWEsU9Mr!T@1ft+`}j48japhr3w
zk0V1+R2#}!X7+DN;-KQjeoa^`s_bIn*&c%#J<&)Aio)7Sk=%n1hu-us-q;Fd4?P59
z5rPHAA=p$Mf-RPzyptJ*1@&R@Kh7IQGa@jW{e~0zk+2hq!u6CW^w_bdBOilBva!gu
zio=iV>=ALsq$E5MzooCkLWJ|iRovm?TcDj`8fMIAKG4fdyh`Ft0PAdgo1cq|5A)Cw
zSb&?dyl?Qi7{jiV;_sSrc$HKjZf6zrvhL!}q#8WlQ;V^YbqFoFhc~tL@Xl|*sEdu*
zyt@hA&ztbUq8T0~&FJjc0+ng3dnUGE<Hu&!n9T_1obr;aCj1s_!ZMjgOg3$Rrb#^_
zr0=0?Y#jzy*I?!TYV6~8wF;L?NPRA64O@o6cS;a9vKXWE3K77&VMcwq$YO2tf;(Xj
zVYe~1D}(#=%w9B2g~hE~@En|s=Djztojt66YKb`5kpNTcczowx-ISOZ_{&CP`nyOJ
z@`hC0=5VNYg(BuK??{OU!#L&&B>DGoe^wx8h%P}i^8)ObpGW4kvyffKH&D(EBxasL
z9rs*E-S)v%X1Y$`&+L0Hp4h#?1Mly;Vb~^DxST!(C&QDlWxmSFM+Xrr>x9$q?Vvkn
zKWvh0Fo*k3_8pd3{nr#}6HOo%XM}++hPV`HfPMElH!SF5=S@A_p09^fXLPZ$RU3o$
zXk%VK?(B$aK`%uEC+4WbC03O=Y$`C~KO>JPDC4lMGLqF)kgcnRHc<_Hj@t`AH5=BF
z4tVp^0Y|teb99m;W~y?g2KG>$V2gSYYaCx}g^NS}cVpshCH7`svQ{bEWdYTXmJofs
z2O-<ITg3V1m(J!WcQ%FDQPvkTj4<J~0TibiV}!aHca_aBP{s@^uN!io-GF(II?NQ(
zKq+Tcx-3+2{D?Bj3U}jg1!r82DnWw3=AB=;XPu*i$DBFw=j=($U?W6Y8ex)Pgs33y
zBKhj!P&{k2KYX|8(#Kv!&ZvCR!->y&xT2$vju-kk>1%*(gAH-(fFT?k4Dox65$1rq
zNRmeA9$|!o9)?)-#Q^6y12gQ4G5lxQ!DzEBZ*^#L|4tD{7Vz_bt^)Gb%cJl0E?l`Q
zi=Gx4+<PpI*GAG<BO#5SzjoqFximE9?n2xwc~q`d!j``(5YkjJjq}DnN@}>NtcHjG
zRME6v6&>T1aZ*i@Gsg-TuuTzu+I;t752w6n3*Ua$q2<^*v>sgtny?;!8f75qrO7;E
zHN=ik#Gbzj7-*${eh=iJGei#7yLTak{TMx6X+*ByiG#a#pgL3v+eb?Aruud~?!O(8
z!z2-)NSG(Joj2vBn1wC_l`vVP9hbxO*9w@uT^Z~3)o^LGCYJ2h!DRmY{Wnn+_VSwC
zGf=_&p9aV(=L~hD4t{>s#k0>w$nxRsh{?97sdQwf=RwGgJc^g{Cow;TKS$qlmhqz(
zdY*ct-q;5hhPk2d*d9zWG)K)ZOTfw+31<7@n&Sw&;v>-LJ_Rk#1sL>tV%bvGLjh;7
zXei&O&Ywf(l8dlv31na93Jz`xMILYTiFK!<guQWJW^`V?kb~==$}lRa7J29Eadb%&
z{sgw7_;@EOIMesi>@lQ+pCWD6E98m3#|-%|yxaQ|e;oVKDxU#lkUWScw-2TLpGQ#f
zuTf;aXe`}OpGc{db4fC9A+2AwhA!@2PZc7YX=y)6s{gx<=HJ*!Gj_?6#u-@(KOjTB
zMbadhxQp}}y{U4lC+&M4Mn(z|6ki%db|-`AmSzwMW<g}27RFhSFq%Crj8^o8lJ$`b
z<k;p*hEf+OZd)i7Y>cMn^jO;16;D~K5~-~0I=%5trrRe{>3&ECeecSmlM8ZbrcMFr
zD==H8m$S^{DtKGFlH51mB@<NBl$X_%sZc|=t!v0vzJ?xrttNk`YEr1VOEU-FrIqul
z=;NwNdbF~FiWZjB)!Aj_xuBG~HkXi*PBCeoD5CRmg><gIfX4jJr&Wvd>AgW71%%{M
z_UjxnP|qQU)@-VG$fg$yv#Iv)9ojV~o8~;sp=Dok=#}biG8lP>#;DyP`=D%68jwSQ
zB{@`*lS|30aSC@7(89Vx>N{FYQ>9ADO`@E1-c-<<9QJXJ^4Ge!mP)JY=#6eY9eLD1
z2_{W6>0vWDn6%Q57j3lhcn9T8yHCSvyC})Go9y^jb!XjUx;~?qOeQ>~_Y<F!<NTNO
zS9ncV4c`)ZzNezhkM!l$XY!x-jh34JAk&Cnl+^u)vPSk3To?BjT6TyCg?a-7`Te4T
z^U;CA#?ym@)29Xt@`r{9UKT?IL3WsMc*bx+|K)IDO2P<1M0=!Q|6`=^*jG$=BsNMI
zeR`BIt8bLhx?!~N@9=2hVA5z|Zp~=n>;2J!409XaZqmp4?fO_cULPZG>7i^c^RV{m
zf<|d`PE!?5{Zyg7guQENC7f;Bjl+JsVSjozVg@MT{32z%s8K=Ad^K2eu3$gwn}M+!
z&|`fwB)}N`Z*mXw<6b<!Z;iUO{Jwn79RZ^5_>X;>(bldIHE_X`QWsdNoy5mEytTKL
zpFv06uzrsl{N0#MqIDWQUCce}J^^{oVOpo2L_gjOG+gV7(kxe;Jne~BpFJ_V*$W|0
zeR1*38N4rH?=>#~Av4aSZQMoVjR{28zaSjAeFfJR1S2v!1d2JK>{Um?LM|GGMKSpO
zEe4S#F?cZ}2A{`7qbM;F?tFu4T^5Fc-XVCS5sc-DSF!QXRm`|{6*uMuBcwSP%2z@V
zq#MfKLKv*#SO*~-KCOHMJ01bY)sgu7GZMERM8TCclL+(a<aWkEoV!_?ORllzOGM{^
z>nK@q6QysHagjUg9X6bU;x7B0N4yJgISYT~vr+gv2gTR&@Rj*E|32}qyFc%lPA`R8
zOc{)(RzPsC#P|nQm@&H=395Xv`mYupwsi>5xrc`f>v5s09=?tZsC?akeXLz_Pd8#|
zSR-_<G-9GtBMvTaL{TMkb)*^)5MGaW5ANaZzd8&VT!-GzHIT2a#;h}Up}naJ&zmby
zZd{Je?oxbIE<qt@pc^F$anY80QboCZ>&d}=tvgs9ejDw5892Nw9Ul*-VrK<siKiuF
zGy8wyeMz{hoyeX-0zMjZX6sWdg8XA}Xhk#*a*i>K?|xI}ghQ<=6yIz@Fmz-vUM5^&
zw)|yiH3jmf(<Q8bbAdCi=MnwuEW|tl&@;`SGwNp`%ezL^9-KQk?#=8Y_T05Rp*Y_I
z`#RmQ(ax1~N~bVQaDnuXV|ZnJnECn#c$>iiw!>`kXPgZzFY!*~cMC{_nPC^_Wz&86
zIXuS@??36oiJ#k&ll364!_WI`I=DVU2OS5s*(1>6=b;v^DQKbjj3#P0yL#uE8o%?a
zz+|K{{K|HrGh`RejFdwGKc_SMt6*cII?^=u;*-HXcsn^TcZF|KUmP$pns=7QIv{zL
zJtm#9!MkO95kF=RO0JoqeTo@2u-7qey%|3DnInSx>b1Wu5O&xSd#75!-@p_u+l=tT
z!~hQ62Jro;j~st}G`Q%a|4udh)YgE=Sk`AsdUz6`jbT09_u;$Lq?JlYO5KgJn+h<q
z+6}qAirj%zMlokkdi`{uBBKk<0t0MpW{uKmh(2l7F$SESR@8&JnI5Kj>aka)k6H8e
zarU7emUXi?v{)ZMWAzci*<6!a1E?r6Hzv~%kG~pXz)ZgF%`?LCH-<Q1!kVXD9|=>m
zVR}Lp&A*gzd$bZ7ycE$fSP>_>WML8^gCP^7QJc2|vDajAHAe=&W2G@>yEG>M-H8vk
zSXb;J^jiR|7_u4HMiOg>4cHaC0f+u$EwVukk{i`<hu_sU-dABBixO6LE1-Op0uERx
zVBD?UFgmmiom%U0@yA*?NUp`-&ug(+VIBNjHeh<|dMr@_rf`-<>#hP4CMsZiz8oI#
zZEEQX&eT*%BV~>>E==4Bs}(!&=A0CQ2TL)#f%VMiZK$mh;MEC~-y>!wZo#bSQv4Ys
zi<b{}L6bGspL8V{aBpq>L=E)b&}N>e0`Jc&z-Xot)~YKb`hgN^=O}YuO9fw7>tGB2
z8aBK$hyVM%_*3nG7F}m}h#!XTYGzEoK84y#>S$Q3i9-!q7&BBC{}~(L1!u(uH=1F2
zfF({3v1To`AF-#IVYKWBivGG_%3;piPw|9wh&PTbIs?^0f9UF*!{yHx(0MixmV>Tf
zfo3poo7_Uk{Z!Z%^R`|HZw4RXeV$RJXxv(XX;<!HPHO{ib+kZpaR=|-JV5czM_6&O
z4?7mUgoF4y*r|So=8GRNuKEYzoe0ezHi-0>3?)~Y5&TRYMW+7T6Tcxr)l245P0|8-
z{dg5Q*sdj?R~u*p&-$G5*+M_xZKnpqom9F?npO{&rqh$9>4%XFO%j!(rDHVd?qyZ_
zprl5QPFnOOT90N8H72oma|)E(OIEvV>7|1+rF=L+HYsj&Qp}gMb<WegwO46wb2tqd
z8BPA7v9x_|0^O=gq$}rc(2%3I=!I(<{fNq>Ri$_6;oV$Hzg$2~<BMt1x>Az+#B5--
z3i8meq?zNZDAuit#-~=1Q%n`TXYO$4=StEvsidLF74+!|Ynp-Obaql1NzX4O{WZ)P
zmMNxI%OWcBFQkCn0($x;pTcJ4(=Dw$QaYDQk_|c3F@?2_b2jbhyF*FVcj(=~J5-aN
zMGFpO(KG2RGMkb`cTe1=jhk*$_TVfM56Ysb7g_Z1<{gUhU_IoXLn60w>4RuKjXhOB
zMbnC?qoJ67Gxu+SRyi%7P)X5uS$o-6)0^S7wB~9Z1<$CblJEw)HNA-(Bbuph;s23z
z)=^RQUl$h;Td^Oz1(cR%pG$X3cM2xhfn9`+0@4j)(A_N{b}NcvcPj?Q@2vOzgQYHU
z9A}(+&v&1_KbyNRs&CU+*Spl#{~jgP-={-i4`_}@59Mh;CQrGi<SzA`-@BJ|PV+VS
z*uJHM!S887)JK|K_nD^gjB|1FcRFYFlVW%D(dmLe<a7TYiHM2_exsR>GfhkwGPl3r
zK380Dm^MK8He#Tl*E>)!x;{vlls#B@vuud)W%f{E@8zL_NYF6hMc*)?Wy^3u?d@>k
ztK|rxHhF~L`Cx=l%rj4#yo4}LTS9ntMjif*d?xtejys|r*b?myANJA7EpkKMP|hEO
z*&(rZAs)us;=>hdNC#RWVJOcy8dlg_V~yl1Ht1YqhYe%waVX3l!9zT;KgN@@R$iEM
zl)X4Rn4|t}DIULLF36Z5xPRD&{#W;6gTYR``nv;N|Lw%p`t2xvvklWlx3ZTY1iP+o
z#Oul1@b&j*teUtPr`19*CxpB9CT_z<=9z~b=J}?0Hza5ey!`j#_uFv1&ff=>O$ShE
za}fWf9m0Z~BQP0z3^6@X*nTwzGw#N64|hB|Z4=R$kc7Oesc2|O$F1ub_&X^J*A``=
zM<xraI1>ez3}l>7!_cQG=<7^^b9VwtcgHhpBn}Z7amaZRhix0;(W9M!cM}s)`7#k=
zJm<K`^B$!t8HSu2d;K*9DeU)|vNH{%mZc+Y6|-AnGP#2x3uDwz;BQe5dJRrN=EZ3Q
z$K=ApEFXja6+oi22wID|KVm`|wA;#&u)30UH12h;tHBD7I^6hL4~tOl93R$<6X7lR
z{jC*Vj%^s2)efVc4y+&4$$60TI5_?Sl7C*nx8{pjy!sNJi1B_Y<TA{zUPj4~E0CLe
zg|DCN3H*H-hbu4RrQ>B(-?)VHs+TZj^F=(#xPbjd=aHV%i92EE@YA*f-2>a<pLZ6E
z4O&rkzL_<2?p&>IfZyzTp0R4t{}FqM)vK`NKm}4*w~^r4#$kIgWUm!s$@~JoXXQck
z!x?CDW?q$Nn>lVdNPeD;k}X-xmdwP6>U8{I@3+LDRFoAaW2Z|J%>E`oNRP);o^9rI
z9|tzX;PkL)Oh`Wlf0d(<W*_83@54MZ9D-EY0n~2a&tBzy2s4a8iA*?iXT$L0%3e6_
z*@F|4nKKv1ok~Br10ZfQGUo8Pw|Fhy+OLA{xF8&Vz7&`cfZOf9uxRi`gSj`Ze`O}z
zaSt5#a%c80du|<Ev2(Qx=BGR39{V-kt#d+#o)a!c@mIMM;$)q$oVA(aS@x(t!S_&I
zGfcEL#cndi@-kCwPB(+kKWpp~v4`k+XAI6=#vGmH`2A=l#15{4&4yJNGCl}DkNe=a
zn<v_NZfW5B^3Wb%9INs}i<K|r<Gryf+5-oMx?`HF9!iGG!l_pV<%XUp<QZeediKLU
zaKWUt&S)rc!KUkGkT_t5m_Bm|YHpAUcf+<SXIPK}><et6PCQ5SvZn+VNGq|xn>kh(
zFXM!7K`vOqIcm}Qu86zgfsKQ_P{SU(6^)*-8|97-{Pz!-;*OZ{?ii=z0d>BH7}vW)
zy3!pjpWSiN!vo;G_fgipbY#3R^Ca)~zIkw8wFe?5u%|`J5j{by4ZGT6eugy+ocMXX
zv%w%fH`%SSMkIT81By)1yW1G@Vn+DA&;aQwx)_wKjr#FAD4%PLq$frw`U5CV*5c2Z
zCW8K|A$YkO(if<~Z=@Ou*QmlXTLpHd%2>8u8ScT#$Xd&}?c)}hS#E(Dk>-$P|FXw#
z6}(VX#SS|aG)t?%Sx|xD6jjJatK#GtHMoseNA3tM^y!(Q{iX%?%9u0%Sp#;11$Z6R
zgO<Djs^=RZcB~<8?lQnnQ3JeLtdEt`weeuT0FIQwDxEJaBXqE8x;}i<4N>*Q6oX_f
zk)&>ev`Y&)qi>HuRnA!bwtz^8B~;Qldv($pUTQY@Yi5J`LR)yualnosFX$a*Kh+HG
z4QU8MwA@+@__YBc+MBWHDr@PR9UwKr3A<{T1Jvyb=Z7AsV^5Tsgg-j}1>i{p`_Vgs
zaC^X79CHo9bhVxMv3C#FIENv4@IG`m9e~Wn!*FLV;Dg#IOtCu-u^Vwv)=5O?<zz_v
z6ysJ<8KPKc7LTt((m}o!NVOp)uoKpNO}TgTIxbzmjhff@P&>5;hju*0+ZnGgefWFM
z)qcUW>pyTN?Jr~&ic-11IOXgZMCD0CNx6Ik)wGSGBl9Lusmly{oFz$JI}}OOX8}pK
zs?wFcnzUS6AQN9*TGOsi+Xop^N3<dRG&Ca1YsMrQU`9EvR#bh%hU(tg(bg6x3i#+o
z54^nT&!|OIF=i>nO<hU0`Wt9K-gbi5UP?CKPtngJ$uujL(qAXh-<@f+$vczwjXXgj
z?I-C`<QXzvolgmYMYMiNDZLM<paUk=bmVI-y>@A!<RwiMKfQ%QcC=FO!Lu}qyAx#2
zwNZKhcG4Z#PP=cnk;&FJ!tb-pt~pEoCtB&zy%stl)<RXYo9Tpp6U`6gEcej{8qiWt
zGk(=kr%D|OYicRGxQ2ATS5ugNHBF3T*5L0-y1TlPN`F<5Lrewv3KgXNvYa9V%gHpL
zlvWNeCE30b8vd<>>NLw}lVb&qcdVrG`c;(1&sl0!4W-r9QnMWUk!u=gvR4y5AKpSg
zxsOwAZ5x$N>7eJV<2c!#C%JnUsLtyWO@45h>OHQ~f(O^AE#L;3^xmXfM{m<*&AW8(
zeHTd--KYO{J|Hjm9#YbOOy`uJlD)!n`lQTV6}qn}-Qg{z@r-lu&_|Lf{7ecrzmis8
zFWsH-llB?*QOTk|bY{mtx_hplu>G=#aQ(8V@cf*ZP+!|$Sezp+e25qz1p9FJxZEJ&
z&AUN@XU1Tm%xs9@_GF0A89Y>odBeJn>oDPJ#W10Kz;NNV(QwXZarP>DxZs*OT)640
zj#0_#aIoWjl)fu2^O@}PC!YE3-C*;}8AY4za4viyW=q>b!ps^qHI}I2&)z{k|F{@i
z<HBJZY~#E_#31I9Z)f(yAod}!CVgd_7fhD>AUMz$gYPVb_@ia~Sy_Sb!eAJV;Y{T8
zo!nKh0~1(}5udsh^OCmW^U*En9u|U-;Ei}&7Kkm{0XRBk5w;xh$0|o3)bhS-33I(_
z`}^Tw<qoK?-HBAAz1;o8ex5`7@K62#e*WE$;p-2gR_8E|3^|6o^-=Jh9)meMV<69+
zM5S-y@E5#ms!YWn<#cF8XCNy+6E|LEqU&@fYv373FG`2p-&CBrm5jkYiP$kV9-EHD
z;wS60?z|g`8xaeUx>)vH$MLl`9-ld@d`&J9DwT;SQBA_S<|KSrl8jT*De&x<ik}}j
zcknw6ZnHA*#6J_=4O!4uIl&!8IatKGsv*BlBRew}88-Q>M-;&9L=kRUm+&mZyu#RW
zn5$P}5qG6eT2ceK@3ol6x$A9x4cHpQjGSxidy#I%w&1hS$!KHV7yErLcHmVLGXYaN
zG2ibzB8Of;ef$Mn7<v(3d@kZh+(pbSzKBJ67a_a<B956}M9<3$NLqdYd#;{G1kXn|
z%sWx*a}L#B9av`Ej%^d$P<*!)*CSexDcy{o>PD#OH9)1V4(^JzkVvXViU{|D`BlKb
zp$v+XI1{$67$uhqIhR@h&x3iW`g{hXeV7My^CYV5bI{Y3jjUztbrQ|Qs<d=GV-L5)
zn-l~_C-Zxs1c{dkkYvBjXtg-(y?Y#oHplR+7mbLDW6Uo&iW6UW2N`u3?dpdh^O$Fw
z*!{5K*(OdQ0!>51@%Tk3Hq`9Ju$_BAO1oiuf%}fqwj*@=7BpcK{=HZSujth%whqRt
zNz37Sb1@7b`D1CP4~**AGm__td+r`^7Uemj*%e2UT#(2<39m3Gi1KbHw8H^90S=hi
z=797vN3=25X7L+GxSVx>%v&2gZ?S;*VH5UK7{j&67~f1yIG1Gtqkp_pxoU!^oWnl)
z)eZf8mLq#%5RPsShJkl5?3xxqX{jyt?XgA94p+1;=DgEJFR1!6o7bK5)|^$&a`wO=
zevF0$IiyR<;Ok2nq<hIi3T}8dnE5qg%%1z@i0mE<w1$~s(mzuu{x!rcCj&?*7@+x!
z9zGPAVclkPWXGFwpOz^`)th13C=2#hTcE$ABeVuMq2F!hEhoC5Fw_(KK6zq0&nd~d
zo;bnY+u5AE{(Re&XA?K<9^rupZSFXC)E(BL?if?-j%7nVa3#qDqk=uLD99fFy|l$#
z_Vr3hTfwo;606)Tap0vjW;j@3#uf`Cu`b+~Vg(`I5>eYMFkQqP2Lnx@IY5tD&^pL>
z(nhST0MV`5ILR|Z)?h+csTNj>Yr#HT6B-3-xV%XXRjz8xuv5c;f2yc8QAKyA3Vaik
zA>*rz9UjWqcuX0ejDU$EX1ICFjD6S2IKNs2L0YWcG$`X|k}^E^C?jsP3jTdi#!Giq
z#D7!6FHd!3>S>^&TLr<b%24Z3#n>5R(Bn6TbM0fHKYu(tUkgy*Xn-fb^zk}K8`dfU
zA{8}wH>LuoG<CcZ1S}QRhNQMWzI`%5=miVpT(!oDSqrglfIXgC8^hwQ1u{ok!0w7E
zXrUEe@jh$@XT%a-+2FU1Ju^dHV5PAL3nP|5Vnz@W>sKQ)bsh3*1K_mF4tlZ<kiF-~
zc}{2C+3Si-BlfZTc*C6Ye<5}Okbl8l=yE}jbzFnVUYn5pWjhK_@5Y|EP&BOLy_VYn
z6e@BT>x)QcY(>HTzvK9j8HXZMJ{yXqU}oNFoC_<&!13icYgdbQrzY07&%$q42NsxJ
z!gtSWkP5kl*z7Jey?uasuP4wM@)BEqzJ;>VXS^@?4)NeWs4ftpEj8>hzCMsnd>%p?
zqesy95u->n<v%J^nn9;6=2HFjd352S0(nL%lZv%EnRaQ>y<Y-_F4LvYqWU!JKLgrf
zU`U%r8q=nHQ)-`JNppW&Q>2z1nGbLxFGV*B&hesStNiJA4zrMJgURW}dJ6e(I~{(o
zhfXcpM{fg<&=}ts3Up2&{qgKOmQE-4zD(NFc!I7SI7MsxbLo<C0hP!X)3;$|WO<2q
zKwi~^+*+ELS5Nyr8cDRKiN2k0CU@?9x-_wsRQy^gb1myL4z0``<j%{a7Wy}`g|@F~
zruNb%QvJ|Kelyu$v#^0g_taCr>N;A;KAC~)wKRWA4cRnT)74?s^xnOSx{A0fVpJt9
zVST3MS2@{*bD!v}a+-6tjJkZv=w)LmDSDSs+|Xj~dMKh%6HD0NQ%Yq|%Sif9IYs=a
zpaRh<Qkz#zxhrev&W&0MaNv9u=g7q)n#hB_%#BxDsd-f!y&c#=$B&<*gHyTlD&+#j
zPPs&S374sL(p4IievS4k+#rpPo8-FUHg}5Or5Sg-s5JdPb+3OwKiP9^$2!gorKfa2
z_BrJ#yre^{<4iYuOM0H~sb$?q@;>~T8cu$tkmg<*+x3IQ-v6Q#%mB3R|BsS)^%E{_
z5fQGg6cs8w#DrMA{zAl5aqf~57ZfT62<B@C3MR7$3A@_{2}?W%3!9z}77V<G2tntD
z2sz9LELt;Ea4j4vY<e(M`1qTDJ#?6mYNrkhXLW@8sl)oGI(yL!ai+%!0}WlUeUCGi
zZ?fZlNDH_wG>6SKE7)~d;9<7~gbkLMd&UZe{cNGHz<X`xQ@HfA=lqX7q$cxcD#VL(
zo8B10d6Q8&i!gL+Acp-4gs#zIC|_TKVMR+}<husu`Rj1#K?qLAZNYVuEyxI2hWkxR
zFm`7k`UV6rpP2I>oBZ+ir#F5F`tbhQ2j6miapIjHw!UY*wShTYic1hCwh~3(H(~Q9
z?q7Zy3c1DzxD@YapXVWD{yxh4ix~Jlh{f{>Nhleef_+J;IM|<c6}t=!(auE7KAvy%
zGGVcSXONlcP%%lx*1Jhq`Ys-gvt#jKT?{5PMZ>r<8fDTk5Prnq#Nb$T%!$LnN%4^F
zjK^U^&MoIBAVDb+v+pEACOHX@w<P1nniS~mPDRM+G&sIV$8Q7nE;CoBP5A`kTXGQY
z$@%&}%mPfy#TJ8nG(0Q7jIbj7o?U{bdhT?0EQiL+3ZC1lFtV>24>#3v&qW<BJ2s%U
zs1Y;&HDM98pm%XA9Cn_?sJ(5ty1E^<rXBF+4y)Iv&!Ja^^H%H&*6r<tr0jV-H#`q@
z-ShZ2l{*z4bYk1CPW&C%i34lSVR&-~k~#CeZ7j2hC$-`A;IoK-)`D~S&5&Q*gbgDb
zVVPSGQ(`W0Qw^H5t1-N^5|XnkU~{O9z1G|t?_3Ot8s-vDD}eIeJQ%z?0}J=lh`D?c
zGc5Sud?_10T(cniE`vEe>9{VHhCvroP`^GI2{V%5b|C?;HpXM+tT_0eKaQj|F>sy`
z4a2%)SmSq;{Q>MJE<TLu-iNS$#6fIrWRB9teGpTRfQCpo`Z4#nGJ7xd*X+UV1-s$%
zWGAzWcffPPR+J72fh=cFtj_QrXzfZA7_C70%%zBW7r^s8v-%S__Y>xcj#<24X>mi+
z5?5@U;Q|Hj7g|#92*WrB9N?_?u)}t+k>}5ctUa#1aX=Q&6V}rm*-vSY*bfHqD=~nE
zuMzr-8)3pg9rWDO!Lwz?{Ms1f`a%=jW$#A$Itv_=^5uQoBB)0NqGeSepX+VlU}no~
z1Y11YZi_!>SkK7u#I3vTcst)6D#4b_xt)ix%jQ90>wE;hpU<<641PMwLc2;9-G^8Q
zXtKbxCk~kS-VBE9OS!CSioh-dWSJX4y`KTjX6RuA`$K}&jqt|M1a=uFQ1dm#d@FPO
zX*9<$6-V6c=C5$p0R$JgZSzFceNR*m<-GS|R|L;=WoDl%PDQ%H(||p(x~@1G!~8tn
zTTL}_XEvNW@>s`dzvzIaKK3a7Zy_fCvc{iFR@`%Ig$3D`cqVCy-QE@$ci0@QX67)}
zGR5`-CKxDV31w!*jPuol`&K=Cw<0`r2A)Q1p?H9R<r0JjD*;>|i|^rD=uX!}<6{lH
z2vdRFFg55~sKI);8m^pC<-TB51n*bj&!95C>MLW2fiic-E90rJ3eFEz#o@Cm*w?Ad
z+-zm`D67CGQyB_7lwrYh&|xKIJld~})x%Yh+^vTHD%3dJrvi~fDsWfl8EK6Ir2dS?
z{e@$Yy<seNj2sUW<p~%oAq8h!0R<NX<VvgK&lPn9$*bUzff{n6v~a7F;HRbsIrc;^
ziRD@5pf$#1+hVt#3HP8F;o?I*-XR%6UD_172hDJ0p()g@t++$X8mH52;4kTj(SC0D
zIyL|=6qZAhy;?W_Ta4WSwlE#G5YJofp?k^^uic%oP{b8}3GP_9m9;aTlkYJnpmp{V
ztPfm))Q#-%GTsR9ux(hz9-O%Qd$G1V9P;1yBlzPXd??|bPP-^v>W+c;ia7YKOhopN
zG`?nWAK&i+&IXp@lw1uat!E8qbTj@fZpYDx3*1F~1y+-9Vzv8S#FupA?SjWhc>WyU
zZofhJ&`(I&*^6kU-#9fygt|xer##7lWN12sYE}=Y0g)rAS92`A$(};`|IMNQnj}fV
zSdMPFDbeJas<b9egEp7*aYaL$8N0fa8m&jdECV_qX+#<4O=$2hbJCTtCd=guDPG-?
zKKQ!Qu^vxytMQ|eqn2<VQ!u5<t*0ho8;Q{#k_n8UfiDly1iL8eCmP571&K7|ZwlGJ
zOsC??EOOqPL)P}ENkTG@DtZfOTU9Y_@hYS48x>?ZqMGK5)Y6xeb>t}CK%EmBY4M9j
z8n>Z|<T{(k`c)H|GgEkQS`!5rG?D1JM$%VnB>Tt)a=Bkmq3j3##Gc^?DYf+IQ4O_8
z*O1xTYWmq&MNOlsXl7s~IW$y|&GZV2izui1KV@`eXBjDuDkG)5QhI4sN=8pg$kVul
z-WwECx?d5sdlu1mxndfrS3<9kmQu>8GP-r9oHQ#dXcqg2ZcVPHM;mJB=BrwAUs+F5
zB8~JVu8EwKTF9}jm0}&+$nHTq4e>okI-Fgf?tX!^ZeJvAr^{5)#W}FRYqae5bvl-G
zlm409rl)`J(6pv5GG^BC<$wnyVbVipN{=a@`?x%(KBs`GFR5YHYqFDlOW!o!({dBe
zRXKg8-HX1G{i<Ga+Vq3IZ~jG58-CN~rGM$WbU)$0|3n1MzaqkxZc*W1iJ0(dXMbUj
zfw=JVm$;C2Vt{bZXrM6h&OqUT#~`7WcULYnSnxYOSV(%re&ne`gl_X8Li&;+!jO$a
zgnWH<3^G&4ba!>Ek5$LzPwF^gr-5t_P1ujp!hg}gp$mli>3Vp@dhz=$=6Gpgfpihh
zY>u?Tn&p<bJ>3p_qU_-I)sD4jd%QW}00Ub`sEfHHWHDdk9Nn3r6Nsxl%t?C?z@5R1
zaf)-TVx6l|zhW&O4%>tT-bIB<ZGleTQp5)?!CQks<Wwy}a)Up}*qb{}yrDnc2mY+#
z_*nSjDl_1JUt+HMq(!*Ab}_E-8C)iE6MJ=cappJ_#W@j3E97kCjKgpk8wp3BSlpLP
zV18W^roBr>O;9QtyVB4#I|FY;GU0EOiM&r4cr`HtA!pJ!yOV+iy@`lh9|wh>(cF6%
zg)s+tC&GJ?(~40zAQ_F`<uN#4!hVppSUg!3hiR<gPMH;tYw__osF#2n!xFg@ClLca
zCZUecJx}F0S00>(cP*?N$?>i#hdDv=C-As|v+1nkw0=B|&;z-Mnw!r(5(TKXFM|K`
zV$51!imCs~V6~?Ld-_#j<@#!*_0%9%w+<aq^~kx?fHT9J;H1>dKB^YH)o#T+$+IZ$
zJ&V-^ZLqazN5IW?bnA9N{3vty&UV1Cy90T5I<T^$gJ;nWL{9HO>8W<qPH0DHP#Y$l
zVk+Z>Rvf?E0)wl~Kv@&UhBsoFX#=``)!|oaEgUs#5YSPDXzNO7UF1G=BkpIfD&gD~
zcS^++;<ZQtRD<&%bN>v~tWG1hnR`iv98g0xE}CQ^=5_|}3(|3LXd2JCDcI?l46%Mm
z$jnWEp<6t@4~&C&*>Mc<jlrLx(Xc2#hI>9oF>-h$?@SNl!Ky=OVy{5XQ_g=U@55rp
z2xLfv!}Dq=Mkee<)1p1_klBs$mpd_I!46oi+lrFoAsCso0nP{3qCA*6b*!7L)>?*9
zGJJm+w+PSJ<GM`W3*j;z%*As<Ht%Le@gB)<ixY0CI>O+SJ<LzqA=76e{?oEWSDp>N
zaW3l0a64wvnqY%EXJ=Fl(6gH72uB0tUD8Lt!-P3be6O?8hRq8dM9Av%>tu-NDJHl!
zh-VyEbM{YJpn`Khs%Ndy(qRJ$dEQ0v-bO{;78h4rBd5Uvc}vXDpR+^m<D@aqT^dcJ
z=Ar7{JfL(wx@~1J(@GYuC9?Q1O%6`^M)>Edg_<5iWQgm+$HoNb><l2QX282+eZ17v
z!~2=Km_5r7xur(1*lCP66HRbpj~T94n&I?mYn(mJ`VV_tXLdMYe1#K^esO~25@$5a
zx<KTT3&It-Q*xmz-i~p__6X+q&T__x_l_8w=YVln>>)kX0ly9Hk@VXZBTa449c_)-
z7g>)`u|ij}C5}$E<lU$_R;@ATY?C?YfeCYhOz?cQE_QC$!I?;H{9Y`;Z64wB3?Rc_
z3!D8lVY)^GD^s-)yix;wVruX(&}3GO1~#*<QhI0s4oy^JubC?DN2ub(6BRsYREC}@
z&n%ObF?O0V*7+#ofbs&!3{=7Tr^?WXR)+ZzW$c$zM(qJ*_LnJRjjA#ReqR8O5&Zkj
zDo99Bg~4D&w1vwfYNI?VQx~AgTAp?0(I~h!8u6>g@{BhgAzl;s>@XPv{P_JjB7+kx
z^7!;nk@E-2$a$cGQ5l*@{jG&>Cb~E&se?#I9TZ5K!)2-&Ry;Mv9e*RVh3VnxQl5_|
z8Kc3+n3>2XJR6%}>OC`PuCPXJ6!Riw+@N~U2LY1}(ZO?Hvau2TSud>FZ_6HV&cW|;
z#A8`!3^Q}V_$S;=5$B2TyM16@;Exy`?zq~xoVlUQkDRapFDGxs@(H_G3)%}~-Ee3m
z?8mF>L+}regv#tFC}qT;N<R)I^~pHinTFs0PGIM}eD(trqkMY>+}G7|o~;RcN3>zf
z_fA|{ahWrH*HJU<4!*6qkFkGxAXV`Uji+A2=I#gFH~NM(Z+@ZLw;!4Ni;>gD0VE_1
zrtQ~;(d5}9>BxgIRQi1a_cBZ;BkQ^3_jo=XR8ru3u`(UrtVTtLHR<YHqP05Oba8|(
zCA`uji>ZdR^@1_Icw$CR##qs#P+O98cA%4cTxj_qPqO&n%berIq`NGLF4(W5y!l&c
z*!JC|?-WkYRve`HOODdet{57Y5>LK8N%Y|^XUdB*XwKGb^3p#^+2Uu&zA2CLmlsmZ
z_hK6CTt+?HD(JX>6}7O3_n2obHLj_nEaQ3#@2)3pYM|c54KyaOfhO_K_kO6SQ=#>g
zFUH)O;5wRrzLq?v)l%Jx8k$yJO=5$q>3~}mea^2W)d7_h>Qh1c>dR^M%yQ~?pp5MQ
zl+t?6aTSY|((SksGFLC5BX^4FeQ+^VJSw7zvx{iNp+eG+DkK+;B6>W(m~1TBGvr!I
zKRwDwerY+C?5m*0=1OuJTTRuh5v}DNn{-G$bN4uZ&g?+(sV%f8tCiHH+Guh1|7XXO
z<2xyP%mvDgyhtBLUnak#D|DQDC9*rOlj>^b8qT^+s~_B<(yT7(Tz8+wTR)(u3Oy7w
z@iE;Q{FL+l&*^C23%dO46?ZkfA+LV#Nu~cs>KpKxEC+t2La|;-`SqQ4zy3*;SNkX?
z{|^Q7z6$C6gu$%km|Kbpietru?^nfymwWpQF0$f6Uc0#9;WR*ab9;dB-eRCIxOkxO
zaqu8vk^LY+kvpyAs+kpddyug38EZXC>L}At$1H2sZuY9<(;an8ve3YUnVPViq6OEz
zj4`X>>+^M8bdNAW>o7AMax#PN0SkmASz_B%OKf02;!i%y#&_8v`+z-ElbIK0!Phx&
zC(PgNj6G{y00Yij-t%VfO8|B`2jJf9Korei4et=1ajtE^UK3v^GIu?Z{W>!0zF1$y
z9Cqe+*g10cL)05K6TC6&trrf9^JDP5quc0%Waf=(GtWc)7x&=B1YzmDtvFE2ow`%P
zFo$#HZ<zNmaqeNxT}C5aCJt{UnBnj&5z;Nm7-x`*zVb9&8O-y|6V6Ia%fy6g<`Z|N
z<Bd%k#;T^kk`m$DFBV_c9mCe_NOY=3qGoL*hRr>Sw}!_cb14eFmN6&}I}V?@{C$gg
zhjlO(g+F4UlNE;$)*qY$67YFRB0ld-;_H4gM!rkIG+pk8%u2^b_MCXcWnt>f6NoFw
z;XK$Wlw3cJ+JIbU@#SIc-U76YD8k~HVod$76pxOTv7e*@ch^>8>8&bU=Ce~IYdI=q
zb<lfU56iy|_{7(l2Y;Gy<3TfWPPIVSvz1xst-O;z3$rI@F<Ge%xBS~MZc7^yHnhRc
zsSVDo(R^w?i&)#Uc;C{B3sYMm>)it7ea)Di+yu`g?s$)AK&W3mw#(L`{Y4F)$5g{v
zqY4pM*hlPLj<^@47`C(oYo2junO7miZ{%}cHV+%>&fu`}Y1lE}x=%U>38%BsZvnG~
z8ZvOtJ{`92QxSbQ1<#d};o6;uC;JkxRfF}gm$A&AKh9a>7(DNd!me!INjV(lo+f_%
zD-L7U%0tkTKZrkH_Cu{^9}2fb;K_n;X7+|+dfHyjdF(;VoZS$f?}U{84tQm5#rroQ
zNSm|~xk~HMskRzs?6DM;TaHY*B}h>XzzjV<Y;tEd3unF;mAm2R5LbLS?99w<C)nI@
zfZHz4d(UN^vXW;HZDzM!u)?1vON5U$hm@xQ3S12Eo;?`n<P2c)P9H``HDR?=6GM21
z^6?lDP)67or;T$KdWg_6#Pm92_6?gNlJ9S~&RB4#l@+=pt?@M127xnd(J{pqEB5mJ
z`KTJ6)T!fXmMJDZlj8nxDa;m;#+n>yoLD&zJA39K&{75-CbGC%EQ>8u<j|nP=ac!G
zm}04c3`>ITQ%xjE>%pN~A3sucu=Tbs&O7m}m8HkKNJG5f`|8SXMp*sG2st-Qv98Yq
z59_T_KiUZ$@lM#B$iC?ZPB_3G>POF<F?X*E@)cO)HF3fE@h*7o>4MsiPPkOz2rG97
z%y+RzeXAWd|FYxfXNw*q8}u)>Mjh)n5<K5*T5XJncZ?xkXo8<VO(6Ni1fEAt&^t*V
zN{e;5qeutSw`k**qk!-cgxI@UxIUYAQNuKFdzCt*^VINkmMVMGRiHXs2~$QX!e*-i
z>sCBV^jE~r4h8&*Q-ISp1?CvaV_$<j@6zS5AW<ILj}`E?O9>Gc%B=fJ;p-%M)>D*`
zcVYp~|58G`A?rLNmGM24fBr-fYYwa6w7dcmv*j^!vpl{om4~FWJR0NV;BPq^i`I{Z
zSKnBKTpo|^;}hVYF$vB&Q_!w6AM=*WGiy(gb!H{pE9BkRCJpH1Xy8@37QRYpqj`-E
zT*Y+J$1~lfGkTEBFhoeTE=2C>LkKXy@Sz6yU1p3;^~N}P(F7O9Td}`T9bpdY@H@uZ
zP`w5=%4#9T&=6xrSm5+B8x+adaUR?Prg|3mkYR^8137no#se#kd1HQ~A3kXX!qJbr
zW9?Sr<n49n&fkm;_B%1mdk;RvgkhT*_vIWvgjWAZi2XW-h3+x9VH$_N_GAbVX_&m=
z6eMHvFsr=?@;l3sY*d4-&l|9$rWN_y&tbryi~KpdhL2UZuvX?CM$|vV<E>Bec;zds
zNPG{KeqT|V_!It1|3S4_lrFc5)5o`iXo<ow(oB`0S-W|@ai2i{#HP`Z#MzYDAx%f7
z%F(D#iqySWg~WEKlenxF{a8;_9Hvd%Qg!KJpFTxa8qqz@i{}rur0zpD<Qia4mC??0
zeyRtB4E3dH8v@B<_X^6eTuVMLHq+wMyXcc<7@g_ePrmU-$lyj4&D$4CZdVeh^=dMG
z%uS=*b(!>5=>(-dKS>LYoT1#=`6L%#NJE|$)0>~AWPQDy`md;@SIt#4?`}2iE~=pd
zn_3!FSxeR*YH8Om_Pac&CAIWgdSzToO*d+2hfNK&*HqKAY1LG|xr#>KtfX{>N@_V!
zK_g$6Q@43J#pRcg*m&NB?JlJO?5Fv=vV`t_<=)S|#gsd(n3@}lXm&sm1&b8X)9^xi
z@vDFq#1)YHsRFv7Qb;q07EvVct5yyxAsWW}D}Jnl3z*}xsDe`RD#?0CH5qK=ycIvk
zWY%$%KQ~awx+e1a+Dv_`xR;z+JW;;w)OWjsw%c^l<j(WdZg!D~J3xE9u25y~Rq~3v
zPGb#k(k1THGOW5o;=8+OyZwC{AoqZ}hqE8&+asF%_z7iSe@16JUXWMgD>A5hL+Zuv
z=-sIgB%b<-#zuaj<SpN5g3ovQh@TWOypM$IziGvRzoa#(pWyYppWt~)MA+&lDhwJW
zCU|pi*ei?v!pz&;)9ENKw6%!~tL6_7baxIAdfNvGJ|Y7JN7;cwrRhLnBXf~IPg6(N
zTy>;ssY7PDI$EoFzR}X)tcxb%M{40^qb73B@ww?Oa4^gm{o75D&Y6b3ITpAkX#wSn
z7TEcivyyYoQN@1$)ebfo5$}Lcx{jF6zRQQ+>`x4I;S8%Q=TzMwBJYj!z24YT=?69T
zPSzC#BV*l4bZPiP%F72QWVxre%NtW;neE==4Uuzxcv-|9$H%>J=9Cvsa=vxmJukcs
z_C}bW4@#J;-q`C8^Ixm5qh}9B=!N1F`%0HA+lPiH%;HQvjQqbbh?pIRwkPqBYv7!5
zZZdRPOKFWt!{#L^xc534pPwedJS_vwwag_Rkp}x0$&hVLKn9;(dfbj+5oeFjH5@_>
z@1<AuKY~=NNZ2VKL&4`L&d<g`<xLDmiXKPT%NXcb9A{46arVQ;qV0Md{&dG<;jaX2
zk>*S}`*M0Kd3KtZ22I|}NcGEv{;n)^4m<(ZqdE9J`4rT+U$RF&7sJc)@KU`1N#%vm
zQRL3?%o50sEMs3wIUFukVCB>*4DqQ(Rdfx;SJ$HDQXPV>*5gWJ11b|6F~GM8`4Y{j
zKhX^Lxh*)mzXcw*T97)h74p;AOEZyuJHJ|B+}MKk%UiJRH|Mg}^7mhBLh(f2shc$-
zYf%GcFR#Z1uR4BhYvDbn2I{w~F#K>OtTij((N%^+Yq$qotb{#KMMxN3h<j1_STUNr
zGLM|Wx?${}-hUE>LvrAHI2((`XW?ij>o#iXcz-Ds^Mg{5Ju;a)#S@XRD1je89y3~F
zF=)ea`12mHsw)cjSvM2aIf}=hj-czrVGQ*-1j`u*VgH($k9GSnGb#ce?%^0VDh&0t
zdl3}02OW~TaqH<$^ce2I)$*-yVW0Z`B^xm*a~*0}Hyr;o7=7QC<LtntaGM^;?0wE$
zFY|$Pp(mUKcWjn*#U=LVzF|G$eV7AMc?VU>9#DA?Tg*6bjd7E$uw)^#bbZWV`oaVy
z>}%=MG{BVc1~8H~fb9=e3=L96V6Hj_-_t;GizfPg*TT8!0(N-oGB4VIJtxM1uPJVR
zFvC6epKKXyg<N%OJV{W*qGBbK4^u{_z6vT^RAJPoh94>#aEO*f<mI`z_Hi!GY>>q7
z2~se4B86Y~rLjX_76v@uTwwhs?+W`wRMp|Z86wZY8j${|inL@kIDXf`yK(9`AHsVb
z_OZlf>0;gl_U(EZLgXuFe})@D)}8a#JZnfFGr{c_)+pJ<eNC%)wrO@kfRr<4oM7K_
zm@}x#p83SgzEa_AQj+ukd36H0Ct>O{)@AP4!Qh`Q;%e*=a@7|9hT6h?m<<|3wc$<L
z2tQ3Y^M~+Xj(}G&+U!r)!9;sQRI3;vlINS58+5RR{VPr80_J=IG&gGD^?glD>rh9E
zo;toqs$tU)73jWHL2;}i5{eY?=!-meA1UJRCk2F`RltiV1?DO#;8dGDY>&!A#7!PA
zMo8mPjTG;eq~P>Nk~{1rG5M$*VqVMQQOE+6KT^V<4~lrcUjY$e3eYW=L;FiPlxNE0
z#36Ys;d$k`yF9L$%j0;sEaV@Kg2$^-XmuY4-3#Lp%k#~Xq)7-%pTg|G>4*uIh1(r@
z^eHI8;o1Tm<ga(_?0J^a!bLG{^bgX3dan+5?&;#|T0OjMF@VlN9r*mzf$&8ad%QSf
z9%2Bk6eH|>%bH&{VLI#VX{9{hu^-4-UIX7FG;y9aF0qNmSar#m^Jb>#|HBj-Q!PNU
z3vuSF3-^b*<42SiN?Mr7$a}5axl5rsHW;DdYavS^$o+2zZXe$blgpu4EU_Pb`wt=B
zED}!lj=^jqcLfF~AT2E!{bbTmJ@OP%nT6KeSIC`aWw<+?a|M;mz2a+#*4PfnMqPkJ
z-xX~1zlntIyEwMt0n#;|V1dj_=8wO_#KO-A)c=8!0e><3j|gFeIGxoPMAmzUQosNS
zTJv=@wa5HNGuo$+M&C>tdQXzvX3CJvXL+(+ynxQ`QKbV$8dOxFMSZ`CrkUzcQoA0#
zjxwb7QzrDU*PJ{;t?81V9bJibqWLr2Npz$S-PjaBzt=CP=&5UH*On04b#^D^jtwQV
z(0%mp!67Prd6eAu#n8@+aa4OHk#3wxq3J7`jXXPxy3U=TZl6<h_|X}9t<N(KXR0E0
z6_cM0>nS~D6suA}eioIqNvet_UZ^5@p_)eQs3!O0)pS3+n*3d=sf_(Jn@&~H0%gv0
zvrf}Ht&%3PrqcVNoIX32bBA~tmCY%m?}u4)VV`i+3f?<CDy9(6VruOvqIp3@<SkZ2
zUo#4+&a9ABz7&uF&osvt6wtNKeDXe+PwSY$tT(%WJE;oD{Cpv4HxyAQ@6^&8N=WK<
zDZLUcr_lx#)DcrjcfMDVOF#_?UA!N&tEYeq%my^%ZqAx!>Qdldh|IGzXM8&~?C+px
z(N1#Mah^g&FVfOOm*~w5?wF{%N{#;4$#%?5lDKe-jvu;1Ud~<ADt(_ui9Dc?hYx9L
z%Oe_g<_Q@be@1zGUXbLPSG388bsVdAv{(HDg-!iL-~N0i$J<|N>d9Ve2>wB*=KUh$
z+kGUyf%8@D%L&r$Cy0ORCs?M62q&~fg&o&K1t&K#q4cJhplsM**uv}_-Jksh108X}
zZlkzhm@Y2-t`!%SbczcjhpMxmLY?!dd_J^L$I1kC#?q?eH=hmMxANH_UlXqywb<Ve
zB(xaux!V}y98B@@wmB9(WdGk@W^P?JgQ|lWYo-<$!ZXjKax)nGvBT9F_HaGHU31L+
zd&&CEE?qZVx#@*F_MUK;_k-l&MW|W59P4bBGpEa!osmAMdGC!Ye9pLI>y2+Ie3q8+
z!y^wbOrGO~)&8Eaz2bqqiyqiM)e94bdn18aVoL1C`KHBvf4zHPW)_Ocvf;1@+=rC=
z2OznWS>zJOv2Sb~E}UnELsTNn4Fg~kwFng-1K9Vm481+ea5-cXvfl5&4E8K{4++Cd
z@i6q=3PqCmG1%-sz~|cioL@eGf_n$>_ToVdRz1w;v`DPYI)<4c(b(D=&6%xejGiBj
zpu^EHvyXv{_Hl%1$0ETd4i>B85t^2O1wDzl#yKnV11Wg?B^3dI>G0{vz;f>_xIW5;
z`=T8Dd43XOmz>7Q=Vx%&HxC_mxChj{5L3#Fuw+sRv^SK(qoWLKB`PrAyb|*_R$(IR
zDW@}P5SLQR-A8q(@u)}nlm;ZVHDHfxBNA^kLY@Pb89SS>qp%6f+M2MSxd|PaO)y>F
zgx(2F(9df`i*zHj4mRN6oq9Bn;A@sb9s4h85vNgu^jXyq`BRBm=PIx?qMUO#+=KC~
z1f^lc_&B`?mx~MdJ?74Z&Rk4dcm{sAPC>{0Bp&u~_vn&rJo=mo<y{$&7@Lk)XHxOb
zGzI<NB;nT4MBJnV*u06u=A>90wmFVVVlmup9fka$W4J!+D17f6;cP2=iA@ioXy8G<
z$FsL8i+57t5y%b-N27Kavv>C*_2?e_(b$c5Z+61hd<RB1Y{go6?l_9t$h{EjpsBD1
z`vO<uGV_4*tCqt5Q6K^&7NLXn!&^tZ@S>NwHi@oSdej-et}(~P%mFug?2s19+<(?o
zRO4A+d18rEznOi~Ws2jf#t5#^<qlKc;d<!exu!0%XDFfP*aDbau(t4BmG`*nIP+W`
z^9E^xEyS3#R0sCa2Ds#DjP=^4u;v-5i|3dPAqt?OirBVKk>>y<Tz#+r!|o}=W}qtS
zOw?d<Lmk?dbKx({#p*3{kt!~Ub>WiOSuMqS;5@9~IUhI2$a0uS9`$FG@bQ2ubR^Uf
zKTU;wo+?;zN*TA)7obr_1#83A(b5PcrR!jShc+UsIk)Vr%br<XB%Rg65f>wD;%w6H
zex}S3Hf3Kwb0CtPa8k?}+F4FKFS5sDpgpeNw!`>&cBnA0MN5P=XpI$O#4MpyY>Hch
z^su-^3-bnO;72Jx_s{Cs#(Sjpj~aN|u8H1tT6jDc_*BdK@O=UKF4{Q$RUcNl`si4s
zhs)~)DAy6P^a)3PYeG1r3Gw~v=ormjLFUx_ZIp**+guzLBvBbBi5Y9;A^uhtX?tWb
zS5zK~ovfYI$Ri<39**Af=;obK%Q-2$SS*ED_Q))ZltfbJTu3O-!I9mvh;UngDMd=K
z_fmw3AkVv0Im|MdhI#Q*uqIU=P5b2Wjb|hWD|tv?mc*poDVV=x6b@RALbvodJjfo8
zh659j5<3aAPfo$?FVpZVaTX?Z$lz1FBI{_1FsoBSjHEKMI~1|{gE|8L)8Wsu4uZUN
zd55QsmFF~|RG|g0U~Qar(S>Eb9_oJS!)u!%lD6w%?ss4jUz-bJ)NtdPDqO@>@xVY8
zKlsnQ6g5U)xG@eNHAU1bQ%rbkj<QwE^)a=BSF9Uuv#0D?iw~B!`@_C&F=A_1V3o`o
zoNC^PW81c&yLA_a4-3Wlwtbkm`Vi)5M#A^PG2HmbCIvowq^?cIuV1NHyyGMu4a-IS
zltM&qDTT@JN`!^hp=nw(tc%-FFLxdr6D}iiFtfJy-GQc5HzvK|Uf=uA5%lj3UORrm
zv4_1FnD`qx1|sypu0L(s&K%>*L&(;31gRQ~ra#5wDQ5p<+P8cL%^x<GY%q^|y5%Tw
zqayu?SEd8OYBXO|lP0YJZGSG1_g-E0j2O_7?Z))+rWxf1S<xKZh19#pk@iTq(eXcC
zq~^Vd#+fan*z>Ds^}CI<J$nb8f4PTyz9Yy;<sfB_iX`{MDEfKpI0?7oX?{Txtyqyt
zN#iqUcwrVbtL1Q?!71AD^$h2=^Xc=LLK3-ML~DIZ=tWT}UFs+!(Uaxm!VDVw^A+@V
zd?hW_tRx-vO8PXmlK$SRpv$}1>pQW6+Bs`JjCW6y;#j8{T1K<BmeTllCG^#+gbrOR
zrZc9+gytgh)8)r+Eu;%Jg>;-}mVliFbZc$_$z98*$-DE(PAQ)zJ;<Za$M`Eak52QC
zZT<5pyd<B->?`12t3o=ssE8^8is|9T68e%<N<VItk=?Wk(%n!=UUyk{GOVE`CCoye
zQ%{XC4K!y!6CGLKOnaZUkhsHHT6DIJrf7B0iIQ{VtbCrVt1pnJ*(H*AbeYn2U8PYn
z*GaycyC`CB(RtT9v|x4@P5yF^nmfA5D)S*-+x3W?{GU*#@iYG37xZt$D{}tynl4^>
zOM`RYQ_ilBq-y(_G{=7><!j$4W%qabHkbER)xRi6=QrgP|Dhpc{?Q4Ae!{f#{RAI(
z5#h~4W&^s23R112!lc<^LeDxeL9<Fs`1*nMoeBMgU&{T3@$65Q?p0$>w;BfaS4Ykq
zb!hBRM^d3W4y@$spSuQz9nr)Z{_H%u3BJ}ELWVW$J9mxnca{-?a}4pzh_&s{2H4cb
z`{HZ^nB6wujs|1&J86oAx6SZuttAF3TEX>}HG)?yg#92V4EJ|Ohp{hWghiNWxCH-#
z0}(mS7wyx0ILGOYyZYWJjQ2zsXJDHpys&aKpF3JTF+#!<_q;ujFv$bsIj{PTJ(o?K
zbFsbajlB!KU~!$ZUtO;F7VZKYDbDaVI^)G6XGpubz&+3vMSX5qr^kK6-~I4)82ggX
zEx}glW$@uH#c2H%*e|<@=b@cQ*4c%<y}KZt7K)UzP}ax8VN%7NE{7u^v3VcF_v}Y0
z=XJiz9>EFcqrAh9LJE!`JRl0&G>_p_<1y%^MFEM?eAZ<>r~Wu>pTr`09B0fO6EGwr
z5kG$=A;&8PYUfk2Pdgnu$}=FPl!Z&V*%--toR;*Hcs1cP5)N_)Mqe&O0`r-pQ2;&3
zB4}<d#*Wq!ToWt9E46ZL_NV}@<^IZ$D!8qz<}6$dT4vTF>3%KV?yAEz?s?c3Q4imj
z_2^V+z(TJEWUg&M<Vx;-uxo(Y<ObNBuSXN_pQb&o!wmB}q$buP_I3@X{i}xmm}+E=
ztAgUdN<OQWV^>`ndiQdlgkA~8eJO%fQX$?81(<)Ay*4XzF+t=ECPbcs?bMTa#@yzw
zYTPG(Hk0!&8JPPf4GDWvv6)%WUz(EG6P*aH(FrhVjN{yOEVSkxM{`#+vt*;tZF&qT
z{f?rtl3&}^hw)8@+48RraIR@TY`5;i4fhC)u?<InWf=6OL(zMKyC)azhL-3q)Oqhf
zW#?8{YHh~6lN%8#!5sS)YoJ`e5+VJA5T(BiYTFj0lRaLW{`g{;6?ZKZc%WLz4gRm4
zaq@*Drb+W`lWd0x8Vix!V}rLFt)c$V64M7*aNf)mH#Qlg$Jc=SwY4$zod&etDP!^?
zC77|_Vob9VZjDyPRX-I>uv3NkA64eMa&9?I6Mol#FFCS!mn{p^QdztiEr;S1dG-w}
zV8Jg1%nVk9ONbI@eiq>6GG&}GX5TCOKcZ!3;kDiz42hkCTfK9T6*Ct}5|UW8RSK(*
z%|qyt`AF<9i%`C|-x;igS%M0RF0l^rP7yMyN{|y*#<f`skY}Zg_{(Y#BVb4-VemTv
zS9b#qJnu{}BE0*qi*O?YjNN4f#TwqB?J-3c^L;mLV5Z6u3*>W7x%`?P>^W0;?>gUk
zSeMydYK0%Y7TA5p4Ba1$khx1AF{cISX=vc_RCPF=V9n(fYs2jA%Y3c@+fYpee%8bt
zJ1r=U1>|J_F~;i^vZQs_l36j#o8h=2l<xr0<NeecO*|n@Xq%~He6K1Vu2;az4N{mB
zJ{L2)=VGF+BwlWp#O-)ljIoi$`F`@qn<<ZM)>#}5$YV<ldzLh%F=)3GHja|QzY<B5
zDN16m(Ois5osE-WGx3b|7P)s*vE|TIh>o0#q0eXFaQ;-ha-9OL#7U5UDv32ZQW*Y0
z5?20`VNpL4`QJvuAbTX<*NnzD=kZv(Wdfp)P2#SbDcD#%4WEK$p({ulhX2XqOM(J6
z-{jp{uOi(06wvxW9v$!HQ29-cS%$1Hx$}F&b8P+$_Cj70xD!tYyHxc6_E<+|@t$fN
z>up<r-ETC|7_5ec;p_<l=e`#5?0R1XY3V9>>tc)mp1GG+>7YW!nETJI@IrDSRMv5a
z>JZ)~O1eX4njdu8A5d?*9Gf~;;l2I_bl%v4v5$6gcj{i0vX|}9?t_qYKZ5NA$1wk4
zG{gt8zYxig{GE!zupHPa@cBiaJ%X!BaOidgqO5Ci;#wozIY*^({v7-?FTp+c8X^>L
zW6za)coN-%se7LxIO8>be)<4|wcqfcWFJ<9_M>^n_<v&E09qq4gv1kuQ`E*$bgX6^
z9k-lBo-e0Sjs0w@S|UYXBxT4TR-UFcDACL;6^gc3C#UC{)Vh`Ey^0PM%+#mdrbcw?
zq$w4dSW=&yEv@x)p!siINcpBG#mM+`|I-p0;l7gAaYlU0scqa5v76rJhf!<He(Ewm
zOp9udQq!Ag@_H6ayDJjNGANmBho@0{d<LnF$|l{FIdnSj6dgKyhISO@(W<2dH2Ym4
zEzl~ad$uK%v4DH59+%P#zcNZIFQau&%BbmW8MWMEZ6>>nnmx<tk7yY=#+TC0Ii>U@
zr-T+umXK>kF$GQK86>WVPL1PvB)X6W{8va7$pticegSFJ<&&99KDG4ak$+Aey>ZPW
zxlwuaw=<V~B62CyK9|A_aw#J?m&&_xY1h^~I$)4b@iPjj-|#{@G>Co2V~fdCxr7e;
zm(tj*G79@#P9Irw67s4jc1#VO+*(T)x#wXDXS&>L8tM7mW;$@Rg#!PzQZ;9*&OdLb
zHs<-9c-BcWt1i%G(MxohdqU;(u2SXOYcw(K2HkPHMgCLnki^rwG_~j+)$Z=5fi4ee
zh~gt^82p4rJbg+x>Ymg6n3wc)*=tJ0TPhmxp2|BvkWSbqx~KkyMnCvUK^uE%i|7x!
zvh61gfBB0Z^PDqkFL!j7um}0VKf1QKpJ3D7PdH*MA_$oxg2WdQp+l9s;g^dFDp8_B
zMxLm!vRYJV?oq?NhiX{-SPehFs^Jvxb)T|lFP61&t#K+Ck)wwHk@VehIsWh4Mn$C3
zo@no>eVl14EumBtkxgWeLMoBcAW7PL?~-I?6B${VAA3g;na}xset+DrmvO7xU0v7l
zK4xG$bL%%JVVEI%1{S+u_zD-Cc<2oIh0Zu5;EbqxCwT33!md3|SaZq=6T4ktx4{ij
z*O=4cK8Ih&JejNahVX75ESNHbeSO^d$~u(G;5=B)T8N_W%-w7YLHOtp46d7t=j(#u
zBE=lfs37)Za+l8CVEo$>42h#baDO-(E+(__IDR%n&4chrHyEMY=kU&7Fz!~*hDBo_
zY<&WtCCT`v%pccf{Sh!D0AuF{;$-J+ydd5WdNQB=|Dnjdw*cWP%Qzdf41J>93;*9n
z#Ck@-Mr{W+JlTPOg}fngWi3QEaEJ1qScJaa2|vkQu+`lSd7pUR&x?ncCwrLR#pB1@
zy_i>=g!Kc-=zE!hC2vwO>)SrqvM)AV_W-WXNQd_BgLru%10VT*euzCe?H96e!Y~Kd
z59dO1Og_#<6=3E!&Pp#WM(=|X-0~=cQENGrCRCy-x(Y8ISEF@eEyhIEp|ztPaX%YC
zn!Ky-+>BMeE!gIJ1o=*_tP>u^k#|Rtlh=kF4#$wxd5n8n+hNPxPW|(C^obuwg9iI(
zRE|Sa@HjqpwPSX4JMJpAqc`^$d_<0+bAB796&=Of8?9*ka0E}kwV>uxGeYk*!M%w+
z#4!!H;aZO$L+jYn${ozX)z~?t3RyW7NU$nLSXU|LEGWUcA4NEkSP0^q)!n14`}yX=
zt1laRX<1ORISlz1hwwTz1IkVZVg3C8@~hKuf^nk2_<az#pMt3U$=K?hgpg5tp?om`
zo8$N3z_fVW(b&!VR=aRrCJwbGJE5Z(i^LPr5Ha6@tts2l{vr~S+_&Iq^F~ZiTaRyh
z*WgIsD$JZ3hChYNFuZ>;LhKjfP73d4zMYH5-rR?GaTYY@2jC)izw{gXLT>GJsJ{2X
zuEeQ$Hpv@`eB-S!@c_)ZTc5q4(q_(xdu@-!DTLRhmN>9>B0R-S;Lqp(myh^9#vYHI
z&rI<le<I98&5#ykhDOFeJ6iQoSfmf}0)3R1>vLbFJ|3>;`|&&@_<u6O1V3XmSejtH
zyD6duO`)}8B8I%u!mSP!ID}2W4HFdzGS>MRpazdJ4g4W3+z8OYQXO6NJlEr#wgLLK
zvZut^2$OV;5H-aZS*+VEkLHhSZ-O@(rl>zX5w6jeFjlfg#W))X7u)czE}vtp!9F{C
z*x0d0+Q<c;Swj<Y@x;5;glE^Baqy@cYjso5c-jd+z1W+~yu#m5XPg?r{K9-E*f%*M
z*Ub@<YaDU&-z3OzZh_8NpqTlhFFls9w`cCA*9vXxtr6VAe3qdNZ<T=e>VN|_fHLEo
zYQ{HxjBj@40X5}3$#EB$_Q8s^AWI}NcXRcUIs9ibR%_M86+sPL+^WHOcMU8x)Z`9G
zeY_FUhs0k4q>VGggkVDq@O#mistrK}ZSINEK<#u5nEg{ncY->u)TzNmS`ChWCm`(G
zI1HRohMbHtwkaxOXuC3ADJUawP!WsQi(}q*5iH*$g1p&dVC5%<HE!Z)P7=rCK`~Sc
zh_eS*5_4uSzDZGlT%Ho_rjNtj-)fltK^?nGIp6kJ4?`y#V2pzioE3~PZ-pVoUNAuA
zbpxz8!P?q$BV4RCgN2U;j{PB&&9Ot;3OlU$WzX886{6Sjdy#8}hi9#D-^PM<BXjID
zn}~bf6Y=)ZM0}iKhMeE#IH+pPnOIvK{>nM5?8$ieXDXcM&p^yje<<3{;++0m<jq)s
z1n%vSU@kN9<T`}QZh`G^?j#$!lY6aqV)N=m=nJM`*v@_M>`24N+XrD(aTuw&dHAuV
z1lA4}7&Wm5(sLW|tF{H-rFf4e?j)pto#Abai<p166ZbuC;KhyGxVxqYkrq$Tt@;AO
z^cD%RpRn^!KPpNGu_*Q*+OG_w!molf)<u|lE{o88-f8G-lA>}Ed8$~cO!ieObf!du
z>;iP?*jas&|6)XAUzk!-g*n{|<UExK$n$|MncTIf{1MKydX+0FOL@>o&WX1h&!CnT
ze;S$*L^4n3QG4cMy4n>+ueYru!NU=x7qf$!+;&pmka*HhO(cVN$<#S&KmGi9fF_>H
zAeS|Vsc}ROMeyE3<mUn!XID&yD@*Cl#&XJ-SxKXXs>vayhMsoT(%)xw^yGX!Rc&b?
z+o6s0BDj&9(;E2;*+?g{8|mVvM)I(2B-P#q8n&;2R_ZiRMr%DycC4qP7wf3ri#4yC
zwG<FgOBzpW=-09u8XT-9pQLJ1)v2bCD^--TzKXK6s;J{>C0)s_B!vZ)q-R`7v&K|X
zCU0jAoy5N`siZ}ds_4@oW1p+ll-^K7x@EPLT~kMM`Sl6%=Qq&e{P+syn~t~89O+ha
zTFV%xyN&K}K3yT}I4KTgZ}j3*RC?_+ow7PZ;oRGEUG6-+NV!0J#4geD1DENCY9}pf
zy-G*@I6waLI+>T=q{FkiX{pp5`gM!*;~Dp9)4T`NYVe4{em|!4&ZnfA{fz#uctKxn
zUePtdH#DX5ElKTtPq!z3B)uP>s4n+2{X`$-UFoMuUf*c_$pQK{;RkIFAEawlKk07w
zFWU3^H~o12hukjxrMC2cbW3B1?eVokY@ap_u^s((h^@5vP}}D^+!<&w)b`L73+|D#
zK<{zZbK5QO#oZFe-<lx&p(%R)m?GuEM9wp?UcAT>x)Ij+IUcwjL@@KP#SDo_=)K^C
z*X)A}^=53dzzLIc*@tl0h5a9sF^hX!o%pP~i?b3{Zr;du@xksz+yz({h?lY<2n?Br
zKZD#i|C0B8!a~sbXD+mRgCQ&%jPpB!5Row(IvaTRcXS}cBK@)IpC2ScW+P2(7RpVz
zldx$Pf+KiOA~zUU#|2{+XSF={^M=kP?prtKE>ku34l!SQadH6m&JRS(6Yg0x2;pr{
z)}RMNF?f+PxyP2E?JDPkH>}1J@vVsBuB+N$4~S&V!0g&!T)+yfxfhA*jXRLz6$9nE
zSQv0VXLM>Dw$<#0a?fr&yULu-+C;QO?16s;Z|{ysMxi15k-bupzH%Sx()PoWuMu&a
zDW7S@nerVOSkZk5Q;iN|d}<aveq_TqG#A#_^YG2K023+;aag7p%eI#w`$Z|dT*{G{
zSAo!PmAK}}7-v@v1kThVzP}C%at(+xYDA$`6WmOj5vbULjy~pCjxg3)(h5hxqd1s&
z6d(Q^<-N%^m}RwL(#1AxeAtGjdu^-@w81r=cOY!ru;9T_&Y>Jd&CON}8noj6x+9zw
zYT>TYW<FFmVa)wTh+J;K#_D>+$JQazr<T37HMo1F3J11U;+b{@?q4gzzVK4sdoRI(
z$|4L-E`-;!e8g<yK4+y|+-uK<V@MXHM;vBce2DYG?E4l!h{fj)U`J#cTFmz&=5s2-
zs!|}cG?{mjlOX#dk^2S{kQvGwWLojC>u1j9N*wgscS8PtEXJzDqF*rvi|cp5=+AcE
zPTGpPMO#pFaU(QL)*~fn4fc--2W<)CJp;zG%o$!RT!_A5q5ORvg3_D8c;`7AJr9`Y
zi=T<+dA^8SIUOe&eX!hkDq<da;ds6$<k+*Bnl=Tq*16*LXJ<?mam2{NNeD=`!FxLk
zh-I5%V2Lp{6d2(idsv<Gj4^z*3ABcp!du%EM{7)wl468C><xCv(}!M`K2~PyqjsDs
zw9};!d`%MT|1*N1kum4dO^`f<|4p(fY|fbCj)xYyd&c2n_INaIo4`7&3Nrqx!fG08
zRlhay{kt~i+|+?-kscOw>qC?8?<d#?`*WiqV;3Xb>|yUMa|ZuO8>5&xgcJWv@yF7V
zbHi3hTwsk`-PU+<-WpRhY%u+Y4ZaIFVOg3p?po{NX1E!$)66)hZ;y6EM<j3t>rg5C
zssA})s5bMr63(b*yfe(v3E?G<*k$90`pXUo31$t+gY!*}7VHzV#1{6>OfhDyD%A>>
za~a#5vqqGx4Pxioz_Z;3Js)fk;SF>w12Pf-mM|gN3Y7d|FYQJvOgzfm&0`BhrJBLl
zW+J?zb@7F@rPpd2=xxw|^hXVRjn%+;eRa&5tqQvdiU<nly!JLZ7`&0gMimA2lPDq}
zLJ9kll~6fE84KnpWAZf>^d(QgnS<l7d4V#<e^-LXNM+2=Rz~;=C3qbZN4&T=?!Fhn
zzfB_O2ok|t*)dpaAd0guM6o$Y41SBn@XApfC)SR|&tLKoPE*2z-^wuhs)p?Q>eyMY
zf#$cmcrsHTpI#e4#>Wt1g9g|UVSu;<1B9(Mz}MkM*kH(AJxk2t6J>>}FNP33V1f<B
zW)M7PiRnkJS(~#$m!&xbrcK1JI8(%ZH^B%CQ%Jq#YxT2<n8w&I;<goPItX404#?#5
zs9MexG~M)q(eUZW9p;Y(3Bj!6&&TI^eC-omiLOa&5o5FoV>-972Pzidm+i##fH*vg
zPeQEDKA3GzLvG_iybn8!-FA5hO)thF*K*9%ug12?^=OT4#@%~I*~57P@AErQDSQFL
z)?LBR*IiIr*^LCz`|!H{2!pN9klpzjPse=3ifw%;82<xh{eLj;-%v_d7o@cCktFbG
zG;JyqqiY3{bp5g{Dg0KX$73hZ?LIYXNY|p3e|1UD(2x%5n~;Ez88voVP;0t1ecMEo
zA8kjr?T%EcJDC!1O`-JTUbLWh8ogLGli)g=cNRiu?D$2bX|RIkm8>C$t($4OM-+*?
zjG>oncaiC{J#<4gi4-QL(pA|sT6`~^nj;R8flwB0+nhsZy7TCtR3Y6mFQ%!crF2ZN
zoF-LOkd;~$&0ktgt=nqIbwMrtP^_cCvN~ETQcuMm^%S$Tp3d?^)2E&mOsJ;`j~NFg
z)={q!Kksrab<C-yklq^ly|sphiPTU@W;OlOttO2VRb=H?MYFzCQo;U8>ZeM2__2bv
zlvGf}q6%7TP(e3;l~ZSTIqf@DPCu`d)0trv)U%|5_6%0g?aE5(-cUtSe$|xXTtnZS
zYe{Qn9qru6c&fC4?!0KEMEz!38`(ncH;<5>{!yyk-A3PE9iy7b$LVY7333~Kik7cF
zO->Iw=;Rdc$T@zF##mh-i=!7Q&*3sf^7dBelB=|ML>K+7x=tE_H)-bRZtCl}O|IMT
zQnKBB?kRde@fRP`?BvH3GV>|L@-FB9w#Sq5Uy|^g*K|$lEhV3QM+3`0P`U6Y^2qAt
zT-F!*oY_Yn0$<4^_#3qr4bZat-^t_W4+<9gNzK21Qq;p=G`#pXW%~W0BfQ)Be8pcn
ze)BK=8268Yrv0NcOaD<4cREbrEP`f+1-3<5V1JYehUb~U)Y_EuP9~VoJrCGuhOdvz
zF*%7fEPHD>*7No60r37DbJ_KipkL#FW$*aG*S(-MP7qad#ycnW7L-ka%{g~iZS{nY
zq&L?5n2Mv^g{+w5i=j*WVC)o(uio>Z_$LUR+h<|mLm=|g1EFy~2)SLeam|Ojlqa)Z
z@LvFK+OgKOhdb0Ja3@Dy0PbB0z$(t}PRR;HvpZ)t9DR|zcslla`rw<jH(zsrnZK-1
z-o)O_QASYMsfSJcF~3-{o@Eq*$p_{kk@tmuO<sT|qsho&Uy()_aW5Wsj{3N>4|6)E
z6$J5fm*T0x21uH1MZoze6a_?MMO+N*G-A<a!#>snyWm;Qy`SB2SgRV3HQ5OmXR;T9
z8<Oy-I2muRry%xED!v$UzI<^SzOY{P>{B`_CuJa&`;;U4G9lrcg;%ZFNF1Mw`nWva
zC(g%}X@#h7EW+!tCD^sR6vZdYxYMHoF*7P*lUjv2=c>{DsRo-x>Y%4wkIUnk%aLrv
z%7I4AzSxA3+)a^a)&dF6ZFl<|fyPPpBg?kJJg^lqv8}jvuoV_*t#IGm3U$|32>d+)
z*BpM1_7Ny$w_wztW=KzIhJJJt{K^~o^EbfmJZn)W>hQO$7BP&mM$fLsDfKE`cv%7a
z+;TWiFT<;UB`_~922Cx((EbA64bNvTA`k0(a*(ktoBM*Ykb0H9Hk%K@OgjTEJ?RKe
zJ^)kKG=vW2uIb~c7#op-2J2*u8Qja)#6-kwPQWN|zs=v>us*j7ar@%<?7I_fA+ZQ&
zzfJOl7^EBXhFN12-(R+3AYlvaUu{I6%X+M-T!T9KaFp*2!;xXjab*1xJbk$kDZZg-
zJsScun>oCT&HG8NftdStCgzs;B6aU{c;vI6<SlpVuJA&$jwkOocwox;DYz`+2KVzW
zh!Jx_;~zT=L=o%`TOqT>43gV8Z?nY+;!TEFdEF3)<{Lp%#2Ei^enupieKo;GoSo4_
z`gARbWRAxbD@9<g9DZGq!R}*{Ft?Y4X}2V_=4zw&k1_sTFo7=Ly9yZ_iRNm;Q(+ud
zo*IWar^h4h=L7^TRmF~JYRJ&gf_|bl5;y6<HBt}D)AV7VZh)tS26*$>0H=2Aqo7(J
z!h8?y>NSM-K_d)mnj*ZLbJU+Lp?{k51QV>Ww2U=2F>6$>u|R;8BfO<`AeXC;_#jiL
zg#OPbVh{Dt4%mCs5i8d5KK5lN?EmD1l1EOkPGN7gr4vebGk>G*h+_}U+1G7`mwd0j
zSZfYbLra9WSaK%XiuW9>nB%m<?3LDNyUhN+7Au@&AGBS)CB!ZHT4ZE_rumFx&M}V>
zU<uoImZ;O`*SOUZPIt{w9byJ61${iwR7IJa8ake+^Ut>il7-kyvsxW<>Qte(Uk(@I
z<&d;g4tG|`VZm`ZEUuBqllcme=vRQ2o)Yh9D`AJWGCqVVBW3@1{Lvi;%?C>8vsS|S
zFG_f~OBpk6E5db#IFg&hka$-FGBZUmeS`?)4~pRa^)Wb<EQ+|BqL3*OMdoQSJfAF%
zp-)7)??-_>xcoVOj)U_hb?6_`z+jFJcFohnqpSKDU1ET_^^8T6^f))DhlPFmSZ!p;
zTMev7mK$T5fdTf%8p3}FUxU9)M9eJ<3|zM24n)@YPMhIHALrAxO!2zF7<NJ?=;t$O
zAou$u2buE*rWGU<2$FW}V_)rzjstFJZ}CLOXCGvr@I#a1Y~1Xfiw8oBaiVJlZd_W8
zdmbB*A{Gr(n^;_N-3c4>INoO8h0lkQ(Z6aR%<~Swg!?X<I<xWUR1rcQ%P>u`3X)oN
z2=;4&VkUQxe&mb;cbYepox^p}%P5Yz1~0)|{C#*A_46O%{pA0!)%O)lliovY;4}L7
z4q#KjZwMO<C1WoEs@*q&)(eiN(kfAE-z7n&yJV<0NrBGp7)NF9s#Ne)gL)iwXxBP@
za)>k{g@vZ{dXhQ0Dp`@B80e+?B<ft`K<z!wlp5#8c*&C_a(yUK&W}`I1kyC&xwIr{
z0rjt6M)!_|Q_1v=^ipLTZHtK}t&clNjN++dK_Z<9O{Q(wN4j6q=sxGEMha(=cz6~W
zp2+$Co8(Ib3MuDn5v@L6Lb0>UXh&x`S%_6qGUJ?8qpB(HTs5it*U*>__WTOhQn+y~
zwc6Cu3B6hxK7u_lT{ZMIzJ{c&YRL9cHGPYyrlm5~w4kPn=DJi-{=-U|zP^(7iC0oo
zZ3S6#|IVe~<usOcvDyXY^jWQ(Uc4)#wxeZ~mRd${*O$@hurm4^T}F#el#vnty<1H=
zNd{I>xl$!fWxvpgN4$xCubS>Xt0Ch*wbZ3iPyMqRXm@5KEnz(^$+(4is3SD-ek-Mz
zw9!BU?_9lUr+Ka?Xny%gnkaUf@;7%-*xNJYF!vngJUCB}=3FGLx0h(^mMhdOag}z}
zU!y%UuM>veq~Z0q$ZR3|j+O7y>AUx6X<84-On*oh#2(Yi&L{La?myC+^qii4enC&O
zU(qwCH+1-a`{V21)6c&jXh`HIih18l_MTtpcUB+mdDu@^g&F6l4Un(ScRC^egS5Z?
zpod2W$$A0fogY7`XWcJyyZ4KlRDaW}8Ncc1a>hQz7Kli<fd5+d;E)9_Z8E_GM-vQX
zf8!_i07Uki!okT5pJL5<XPPy1)_$)o;x6<o*2~ifN4MC*?TsBwG99pIk|VyFI3eVP
z6Y6W6kr(6w>+NnhJ(c%bxc5S;ZYlztr=dB(7kB3R;cra<Ze#}`G?KgB^SKjHE&vq3
zJk2A2Y_yw=NALY{X*c($+WJBLg)gG{QGK3w3ny?!HPD|sDExWrg)=S^etb6JP6$ih
zYS!S+kQOf-WPfA#Nh@gHHo=}C14#VRVfV5Y#(vSjvZor{F{sJsWo?Mv)5F+d#<(iR
zdwQXKb~NMv!=fn&tDDAsaI?89eIYt7ha;tLBmQfSWSka-QNBB%E*Q<bFgq|sEgBbf
zV{mfYPE5|`{g!i#aTX`A_c0Nt{_e$V(_}1Nn8F=fsnG1+hapmF*gER~l1tN3{gX50
z(+>f)-1j1yg-a`W^Rp`lkw$rl-kXoAHwBQeXHU+)VhlbhL6K$|?kp@vQdR{zE>&Xg
zhbs6Cufg^)wYVc(hbiCd(A8a!mH7?4*VhO)0q#G}Xo9AEGgA4CwVSg}X$ma}bZ$Yq
ze+zV3(-OC7L8vfuG`uG>e_=D8_cdYL@+Q3Q;*P!vjo8fmb5vqIM9b@7*-{Jn<{I29
zsYZHY6?Yp|axS6*+Cw=n+EI$C$P%10DaNK}g*czgS$ms&D7?tU&%_+4TVx}Ez12u$
zY-4i>z3&dfJ0~6MeGkBOOd9va@5Aezsrcudf-u2kbf4dg_q!93=#hZaqI>Y*=5FlE
z*afF~aoBFO6R9J)m-uHihM7fUIifK%WCz*{w&PIM7R>m$5vvxihf>EH3{?+D)BZ4|
zj9HG{geCaHJxuF&g+f|z9wK(mfrUsA{IUZv!`2_Gzpy5CiuV}VUo!l%H|nIkke=m%
zgq7|P-ZBO0L2gLSn~Z1!X9(Cipp^43j_i>=&l;{Q>$%%hjUhFaeKeT{xKe6>y%P+v
zz0nY50!Db=W{BOZ46tgiE`-fAkvK38@go#rqbCRdP#F{_Ng`fE68d$L@RHHyO(SDS
z6q=y@s|owYP4Mvr>nd^b`0s);tVfN9Yxp=k)>lQpuo~<hYvPuMHXcpX;hj`n?tatb
z&O&{74=|4-uMe9tJ$(M52l=D=NO^7mn;nMOsB8#(Cw=(z=_0mJ2ah)CVzaF&nqs);
zZj}k*E!ekwR0nRib@9Q(6t8C6VPB^`WVbsaj{UzDZyoVn$q5pi8N2W11V1k)7?`qW
zdYvOQM>}Gl)kNqIGsB@AGb~py$I(xmcdD>N&3G#;*kOes-HfI9TJruO^D^TtAt+^u
zQ8W2D78ZD<X@T=W7T9^x0%ttAx8ecoQk)C-VNYNY`)+Q1GDQ1xRjf@>;q8ENa2J+k
zomd75CuJ~tj4Y-`%i+x?IUL*|2iaiGYqMswI93i{V&$PWUIF&C3Yc7}2-?ILhB+KD
zCuL;KP=;Zg67N?iVQar4f6qw3?T83&Rf-{o1vG6zQQT@2!F(+d1mz2}*H9SlABC_)
ze+2H&6~(UdF<5eM45lZE!s~`8{>F-7_;SvRPgaI*zY4-OXrSCl8$nySi}8UTVrJ=K
zf{h-YsxaOu&_SoLF1B)4<BT$0=6e|LMCkKwnE{3b81n}FL_A{Lmsn(kz!ydcnraGj
zPgAreX`|NP2yVw1Q{OSh;Yj8WJIv5|(h_OGfWdq_=x=j^7I$dR<{rU-(^DZm%@;;S
zf%v#I7_Yf|<1G8@ACxcSF7q&4mS2THxmbMA+lhs;aR{MZP&k)_^uMX__B((r4>B-(
zT{cSV3L#-sil!lzyn$Ye%PNh?4mg6coMX85=_E2HpT&~Wi#R5I6^RFLKy>0AeE!&j
zpPf(8f9nN~3cf?~f?gE&eZ{SYpE&M2gql|lC!H!Gx+yV=&hnP{{)J=duY)w**N~^K
zK_zmnnm`d|>Xe+KMMKZ%(xV#&BzwV_meoyUkBSBP<yceu38L;1_GA#{L?;YfX_t%#
zZML6E$@Men+o}M{+!0J~-p!|qswH&x^h(lGUQh4KBB)$#2c<8MCC}_#bhKs<y*Rv=
zsu!lvO0oTPBkKTd6wjdG<(bq~l|{w8Mc#Tpk5->3AfK&8v`CV(<J(H<`SCK6yirbH
zS}W-92F4U)swjM071>{{qMqMXv{k&Cwv4W(@n5Q_=4=&t##hl7hyUYd1`k!zd)rF-
z^|*p+wpEb-xC+ufSx)Zr%jxmRa{AR+#y!hrbXl>CM0-l9H@B3!!b-`<xs--!lv1E%
zDNUCyrAp?MB-WIY!JSh29aKiSL(8f0L^<iFR8Z5}N>W_H`qqkSQi-aenEYCj=&Ga7
zLabG}HqyY}CThFiOwKw-=zK&gJ-B|9Hgn(LtLS#heR`an98XeY8Fv9nc96x+Gj!xP
zb2(ei)0Tf1=+T}_)S+~Ro;7vS{pr_8XP}F0b8b+$`z;du*-a)@cR2rjkFHDgP{^eR
zbSm-@JvMkkSr494SL`!#)qX*j&b}n+z}NKl`5XEd`i`XTy{9*mK9b|XPt^aUmwIHs
z&{Nw!5}ejg{ytx6(B>OyiVu)Z*8nZv_?`Sm{-9kueo)4<A7rODNRwyt{==F<s$^^v
z9mD+^(=8DE)f`bLCgQl530xcuv02FkGvkdBBVdLC8FSoc9V?^Q3avUeIJg*?A5J(m
z%NF9rc6_aLz_CsTIKFm7d8HG&S2;6(>jEX6$%yG-A7+dPM$egw_wTr4bNmeajpiQM
zf83)X8;Id6XQ3vTeZimn;it)u-TwI3!+#g*kBFCk7$WS4^i!-E$ot|@s4s-#{c!(<
zA4F7U;?h$;_?(`OFX}!R=im*yc2D$h)>df08#b3)Az-@+ByQ`&C{+i^zqsl1oCXw*
zX`r!66Z5z~WOlV4V)^<|++c>J3L8ANcEC>$H(aykPBQL4+0e-xkM0WmQd^7V%A3%A
zAp%1-ZNqQIIcJtdq3w4Ra}v=A;_HHebQ}_ecf+(Y9=+lTyib=1-2;1Zvoi^=1X3`@
zJ{6a@@5Al0`{6Ej0N3WEqqKp$k%bRoR%j*$j~~Wixok}2zT<P3auG5vA1bQ~pn0qi
z2}6sq$e{!mHuH6*ybOBR%c1_E0tbIrV#APXv~gd~+{ZOo${6I$);bJKtY_a+JvJ|C
z!1|XBkoIoGwfsiTZ#Cle$R;!>Heroi6Eyxc;w595k9!(%z_<~<ryI~_+kg!v_3-{(
z2W9&@+*nh~9XvI7Q&x>7HC6akQpp`-6*#k=agBEw29--O<yA4pl@%d!K_TWT72rH;
zZQ0z}eAP4u=FhT_#~G>f4w-Q5&p=AaK?nz><Ei8Us9xL8on-s4WlAdKMx}7Zmivm5
z_u}@9MAXV9VEE&B+%DXWkY&4IWE#g=jaY2I9D@yc(co`rIF&_#o<~B$If7sFW;m5>
z<bJdD82)ArWNgE6;ZPXPjb4uW#3hK9Vr-KWiV>XElWv}a(Jn!-dl-n#+x&6X$Pb&i
zpK?BDnVUHOz4$Hf9!}(rC*CZWQOdo(UtHlU;R^Qv7mVin`_f1|4COO&@H*C7D%t;Q
zXM!^+hS>C8A8Tgo<JK~L%)YOW@WlprTx)<3Hv|0T`&PS~4mw_{KyLCl49!(Q`WacM
zv`NEpo+P}_OF(jx1g!3A;b@*V8WK&Abe;V=0_y1bErSV3a#+=@#NX#iNa;|9`(G7!
zKUT$@nVLA=rG+7XIQuN33tiTf<TmPI*Lr>Y2-3l(Bf2=p2HvS}_3-kmK3vk6J6Nv^
zsna^}e8%_OE^Q3IqJ!~w^<a6)09)<!@oqo^eOEN#|4JL`!loGNV~4{l?QxFr&G2$Z
zh`n?~mb??<CORS0*$I0*oUmBmiG6JjNZeu$>0VPz_MFHY9Lz@~nc>eQb4)L?M2M0V
zCT+7q<weFdZ#eJHIOm6wC8EVG@%)hmv>D$-Sa8?xUU@7JlY`$dS&YBU*r7%QU$n;H
zRhKAMT@}H|!7*6)N*M}P%9uM;7NNssAeSYBE&Vchs3M2KAUVAKEQ2+Sced!t;r>)P
zxNKz1^GFV>^W<?|M-k6mlra3Y5`O5j=f;73#LwjMC0`MF4<u0=D}pMY(GWf-j9dL9
zvFFrC<}^oO_t+6=8a@KPw}m)kDumIugpmDX1T5l2(DrRK-ldF2ew!$k%w$3#NF2s1
zrE#r69t%`7koS+TKQq{yvqAyl3OdYrYh%?1ZG4%agCT2mFy)#Kj!)1<biXc)bM?@5
zQXeK)jBs+bG4G@q;-jk(W*soXvR6h>D&}kRYb|sv*TH+fR=boNVNQk#goMnHG29ZS
zdu_0M$0S_b>j=BulVQXiikr83!|%p)B-r`GpnnzwJ?BHHej$W{mO^woU*kB(I@fL`
z##FAwx&9rvbB6tGm-Zqij`d=5=AD;jz)UI!?rV83cuWa`9+gA(b~Q%6tB0Lb3-58Y
zA)Nb0uRrU+8rus{D7}Ig>epf4-i;}1?qiw9W7vB<N5#4~%>8}D4X=LO88!%qkAK-$
zH;m357o;t+!n8j_geq*sY4$HkdVErr0yinrXqoYJDMgj+UTKhzoDMD4*QXFYBhpnd
zrEo>gi)&fY6(7)p+)1=W(UA%+xsZ6>6q?cPMfWhBUcaA7nr~-QgvC5sa%&M8wXLA}
zkJr#F{mryt&vu&jDVoj-#8LUb-E_StfxaI|B3ag{s+;!HA*FPRTAo35d70$imPH4e
zb0{)7kG?w>(1u%uq+(r6gX>DDC$5xMt}0_MZ8`0~TuyTwDoE;J1!Y{Tpl|Og$hDX8
z3u73xSJ3-C6_n&vL8pZ(=t*li4K6IF+hXN3q_vE?=ax~@$TA9QETy^gN~wnNi`SJB
z(u^;mL9Y^8s!&4my~XsqvzVSX7c;L@Oyx(4sp~;8HH<5vozc9VB~(hgic9G~#yN+z
z%V|A-9<gr~w5Y$5Rt>497Ya25hgv$dx{ewv>S@Kh2HI-WMCUg(b8fzcoY|A}WARbC
za-4g5WZQYG^f(3HIYGg;r|4(-X-ZW(L-P-urFg0HG$iu^<?$}M%8|>Y?%hd-@2*nq
zo-PtHx<Qe5ZjxtIH|6Qwq5XI7l3C1s+F|g3*4%zbpEf_HyYf$ISkr&x;QE|`Z@!=j
zzOVTI`kHDj-jdjXcjWN$J&o7=Nc(1fqNMe`Bp>~mJR-i3S4bZP8~4+-cl{)k{FS1V
zzR}fv-zfRZH`;17Kp~3<NP61<#UxljVuJ-fvtRA^F>|~PGsB`>V?6QELA8VqK86@z
zCG#Kc3phJ@&J4f#eAhH!2IV*0gKlCC)w6(W)g&l+*<*9M1E%D&KF=D}Zc}G;{&t4c
zEf?I~G8r?zxFg%u3+fwKKl;m?R<rz|&p1XpHvl860?@oP0IIFL2X}EMZ1w%IWZg^*
zoS2F4%%|p^@x}XOUpxhOp^Eup)sq>h-{1?iX5NwGF2RU^>G=9~DtE|w<3g$@_62!h
zMAZ~b(|5$|HxsecN*^Ca=wR)7P4tFnpm31}B4_Zm#*5FqQyJe(F+!dy?*-qnMy8fM
z?hCpiJHs1ck7r_j6?Z3(T!Nq@E0KC;Em99|#O6&~5E2;)^*fxYT(ceXzeZtLV>B}N
z#$wo9KKJ}&9%oTJ99#Cl@J9kBnC-=ll}V7SOh(GP6sQ~SgV=`sP-IWek+JEJSirfV
z)(mKm$>covVR$!ZAwwt!H`t3S$9}$)7x@TMFGPB95&V*iQQBI9{r5_t{HY91KguCE
zP=UpqC0}*13VMgD(HvX@Q?Xif*VN)M>QK^LhXacB&<d+ZUwJ(WZ`5Pxi+X%|T92g{
z>yeOIk0k$koc~vc;{A1Kk*q^rR4v9mV4shA4JHOvBR94RlXEIzP*H)9@^XC4D#P=g
zrI@;qy@tfEX><`Xy9+QTBOl*p=D|rK7f&wpS#L)c!YvQu=Z8b+FUY``z=K#Kk&e2~
zG|nCGN3iQYoFBoR#FvxtcW)AF0eg`xpNOMR_aL||9--m8ao%bd%7*g$dL<US(_<jX
z8L3eA5gUm_q36t2yx0_hyw1&V_S%F|+twlG*&4X44To$?7#tbf{EAwFq*047B0H3~
z8s_l^^c>FN2eF@W7JDZBaXQEkvEz9Ar_TqHkG!Gx)Dv@ta6g*s6pX*-ic?p(Lw40<
zyzO_vYddG?GL|3TVT(JqHb{-+b3v9VZ>k!h<BC44r|H4;pDvme^-z$YhYW3f+>7Mh
zgpvB_X1~abr&`=2s{*x)%CHkwz~hOs==&sv<Os$!YGY9(ITmAvi6iNT7*<NKE)qBv
z{Yp}}Y$Jp85?MYwD8klX5yn2s@NQB;?;TY%-e(-rrHY@0YN%FIN6P{=r1hvlBakyz
zvo$bowk8TcY2prNhc?A%qpXy5r!F0wx7C3Lb4eA~*q=N^2e%mqEjX=@QH+K3`5bWE
zR0C}zG~jbq9p>RCP?%tkOSSemF2~%Bqa)*ECtNgkLNMc;*E=0Bj5~jZO|-ydAyb^o
zGett8Da5(|cb~i&q7HMW`llkauPI=imONxbWf8Jb1}FB6MB!LLjF~BbfT;pFUMGMO
zX+a!r6vP!fA+$de!qwCfaGNg-`9-6UTRIv=^&*hCI|lu(Vz}$9gb2oojw#Y8y(*33
z2pRl&EDPN>8I(Mh!EnY$YTsqCe4HG6)Z{QDRSw!F@<=|ZfRB#SAPZ^y4wA;1@iN#s
zQyzn6QfPa}yv?}LP;C*$hDjq4c4;I6az`Smdjz($3qfb65H5HLVcP^DJXt0L$;%^f
zVdZGt-8dS%-AAKf!f4dXis6jCI37Nd!kJC7ykDY<3Edj_%Y4tyfBgJq3aDJ84bj!w
zn3t}N*dESUsq5fngbwQEb#beTy*k_UIESSV&$|W)^fW~60QV*97@|_g2!%D;=;!{P
z)<xR5l*D<l1<c1~86(7CBIfB^;A*Kg3bJhBlkWhvI2R;Lo`Ms)yp8NW4K=I$;QKcK
zIuB-JedJtJScRhF?;>c2Ekk#67+&_QMS<ZC-n`m{)7pEn_HP1)@i}d<_aM?(W8P_y
zk0t-{S*D?kxxXsd)zqQ4vkBI}So5|xj`xwLQFQ$rig^<zsQMZX+TFtE-n-b;@esLf
z|H0<LE1cB)fb85aXmtCIGa7&JBzq{~q5vgJjij)c(eztfjFvS@5Q1b#;<r4-$Fa}&
z>jY|bRHu1snZrrYrIo1$B$jGS?FS~3ex(KFJ+P+vMz*9`ZBN%`InzyBHwv2VN#3nK
zG$GKB7C6kJJ8S0B-}ejXddV`{cQBl4sy5L3u1G3bw}bYqj->|fGk!RJ5B<54NPlK0
z)3O_>6k(7?JHylI=AjH~D$S(p2eWAN;v5o_%A;wA^C{<70g0OxQH*mj?J+2!fX^jV
znp8@eqnLMDR7SDIWi<S98A;qNqr$E-%4sR1+eu~geO4JoE0@u+d!<yw&mZPcO7p*z
zkVbw9rA#ZK8^cRT^k^{!uPLT8#_VhURzzz~7150SMO3`9i1TAbG|IV%6r74^yk8Ob
z?G(|Fb4AppT})Nw#q`FxgiOAdQ0DPc5=!USu)UmCZLA=pZR|tdS4HdUt7)L8hCYk1
zAIPPi)^T@W{)I-W5N+o7tA+k6IYL4|TB*mgjm{NwPtNe;oHIW`hp(O_;52oz$tzU*
zEPXFMM{^7=kZ<cn+TwPZ0v}zW#T%}Ypkx<CH(aN2zBg&q_gl0$^ESEJ-ldku_o!xT
z52?sMq=DK;ye093CSQ3<tGu4k?DNm*p&9o99(+abKfWe?#yKliy`!9r_jIc613hl}
zNOBdQC?~#`ntVS~?WiyGtKtjEnD<d?SszIV_LGl$Kgn$BCx@hd+7)Yoid7bD#<f64
zt~plEH-oybF+9X|ctb!Nac#N~c%+Xk10y7>vG0-nY#wJOBJi0xQuypT$bREx&uq~-
z!XA@nJ3x4eBUTf84f~z&;G#2@?{+~y@BHd`x<iM%H~;%O6}jBmuE-tZxBmFS;onTi
zF7?MtZ-49&@JGv%nb_Ipi_kaIq3u2s8Uwsv7{S@bv)*V+oDNCGG5-azH%4v-w)@V&
zgN7MM&z_1vZ*T199Qyt+Pc*^<wTGraFp}>}C6nOOrH4^-v|(1j8vO-z1in^>FZX5)
zJ*@?uGrDl8HiYUD?rncz#XV}1(3<6fU)`RNPWQ!&ydVhMF2I(<%TQjl3V+kq@<z=D
zJnr9wx{F(I{d6QOrf<ihDLb%dL<}NNa(C|ZIP{#_h4a$!@Ce$2(!&W@aW4^@rIWDN
zHyJtUDH!!I6>aMKVIQ7`&&LmNUi}~z24!GI<so$c$i#5xEDTG`#vtoe;$nFSbK^~{
zhyrXcDa67{Mfm=#7!G|U@Ej<GD&w3R_sX%fxdH+)m3&>T!tIY$*brZh&+Pxri?3mC
zQw=syEm}9$B0Purm^$vNsjkI|1GTVTS_@6>THL=`gX2qUFzaJAJm*xy{W$m7jIKh4
zTP4ajRKPL49COOb@Ta<z`#nl<`(QC-wiU5<Q3xl~0;mtmM`UL%Z1&~A**6<oBzf!W
zdL~*D4`Gs926~4dg#W1n%rB+k8|_EKpH!HiO2LNMWXyC=Lc!R*Q0z%SWZ@pfgvDdf
zk~hGI?Lz+5ow$?{3y%dcn8`kyKIY6~_HD<l@mqN_X$$X2ug8GF8Yqof4ZY6Q_!u1y
z%P}jlU3WQm1uen4zJ<6F7m5M(dFVYq2d2w|a71DjUY+#E+x31>v*K-7;c3wM$Qi#E
zp3v-dhnn~lT=I5>-31p|U3Er6A9t%sF}_*iguQ)sxF<-gTd{8>(F{GSjFD$y0FQ26
zc=G+pKSc*?+I8S$#+t)-UEB%RL(OkpL@dz3pF~Z}QC30s6eWz<D32pMWgs<Hig$t}
zcw1bYc?~f%*Nb8KVKFSaEQThzvDm&=5(1l~QSzS*X5Cc4co_xOIF#@=R|VVFt0LQ5
z6{q*AqK*B=nkH(v<D`a)6g33*sbO@AI_y(f8<EpQP8#E!EG;Y+W~|S>fJcfoISZfy
z*%(y>j?~6Y_J?>V>cUw|mwWcqU@oEoEkAW!6;X%X=JAlpmP7ANNjzp;v7kj1Niw2P
zZX1IeXGbI8;TRNtl!Rxf6mOD9@fNi-7P=c^h5_fl*bg-EnG)PbD8bS~5zY_f_+BRm
z#Wgapc)_^jz7Uco3n17{fX842pg%?sr-}vf&0GkldxTIL$9Q1-NO-OgMkeEkgWE)4
z-982;??u^%tO%)93TXHuh5kd*_&iw#-796F&s{X%FUcV6w+z0uvu5>07Oy_YBAhW!
zX|WtO?2y3+V`-#LkjAh{{93D}q0-78@3)d@spXG<brilRj6@Y<ny?KcP-Hg(kADhb
zrmGM}3JbyioFH}|5=6|O;kXerl0EE<y?%^>=%~@$TQv%s50Az*c`=l%mcsD|vN*<>
zw+!ZRo_h$PYWfJIY#WJ1Yn0IOQ47t2+6Z;f#==e7*nL48M$S5TTl0Tk3)va^7#GGF
z^84(a`eVR73Hn&Njkz3mZ7e;-JaoJkrWER8p11+*lZ|l1&Xl!Qb6jKYx{3R;=N_|%
zNTM?qNxCsF>46f~sKk!VKr#COre@7T)ucICBru;jyoK1mcqu+ST!G!io1iMhz2AR!
zV};!woGnQ}-|D@17<mwZ{#iKrB^PBih1k5K6k=hOs9sr%%IHQ2lpMjuJIAn3>J;xv
zoW-OQ7xBaRDol>wz}UriFw67-rl>wej`d4)Zh41EZ+cN3`;9rRUvRG)LK8cOa~I4A
z3R*IXT%L|0yJcf3;HMPz#>mk=_8A`u9#3zJRLS9%2EG5FO;>;D(Z7F&)GcC41v=(*
zWUdvx<ZY@3b35w#&yiM~noK@-+^Jh@DjjK^L1hU6v?)KBhW(gNa`8*({FIgCsk@Hc
zWVevW@9p&bSu`m&?xc;&c2mfR1nP_5OM`vMG|p)sg>OkCy^M6aoS8w0$|UW{S>*CA
zn+`9_r6cF^=<-kAs}e4xFMWlSeWZxo0*mRx{bIU~656n%gmzb#(Aka>>N~@@rLly@
z94H~PWhGQ*UP6ZknTuimC53U!Gu>kHe^x}6nMKsX_~nOK5lLMyB#8rs^k!}$Z89mO
zd@gmE`=NmT-YuZ;t^$g`T|k1L3aD49kkS|O_W7kkn(9_Wub&iA?cQQCm|8-sRT<+5
zmC^AZWwd&*oL&o6Qpb4Ks~oGTC#;6-^J*#aULAEvvtH%Tn^xSDqw%ttl1%>B+0#na
zcaD<2(J>O<+fEvvkJH(JlQj0iDN;ZO-E2BTqGso4PwRPl>u`~dc3q;?^RCdwflg9M
zx<>U{*J*CY4VoT&i;BK?(}I*cbWrminH{}PhOQ52W!FQx?emzNE<d6DHvdt6-ZOIl
z^_*5tc}e~8ujulz*Yx4h8=BYqmL~PSqq2wZ=}yZB8oTu)Z=QZ4{VSj7*R)<b($Px-
z%AYAa<TH8g{Y+m9K2uPX1!7*vus$IN`7{MgchSHSKIfV4QAfcfEo7|J#eHXeydG+Z
zW4HL+P-=o<TX>@-i~Br0t>E?B1|<!)80fOYIRgh&SUVzL#0j@qqq=y=8Hq03`_A`>
zo6VE)pRqe@J5!PBFasMG`$GJbAIi_ngp`dxd+hxov3@3Q&*a^zU|+a#??~DL9}KDV
z#bYVnELZe~;wdl0&76udscEpfISuRIO@mmv5AWMfMdpAPzJK+^`IjEJp~~21s2f7>
zGQQd62yqz)Ok>WbldtuA{;Ki*i8>Y!)qw9XEzBFGi}`Okzp$J02e&Qpp~)68GR`<Y
z;11bI)1g!mh-q~p>>XSL^<&FnaeNgD`_{lQc0H;aH{w~}X7*-B;#<>p)I{$<uvQGF
zH^jnxLLB~V+=UaTce77p557AjAYywWz90YJc_;}B98z$NdEa^0_F=DV8Zze{KwD)x
zB)=WR6*>gr=uG@La~KzgG6!OlgVSNTn3S0Z$CJDt@t^?HdJD0Kv)5$<#W?+@1n;kw
z!mhXs&%(;_PqPA@T@^@Tf6T#;ylXkH3KKf2kSA7+WmBqQx~du%qN*Xbts0{iSK|h%
zk@>p{4dqpcajHV%<w`8HV-2dH0^Z-4tFbJ{hUJW*QcCfean0qXVgxi7A+53yTQUkT
zWNSVYXXSCXb1piE=3wHrEOezG#-<?F+{Pb*)U$)wQkc%2Z3nPZGY#Y4ve%|66%uPw
z&}x;8dqa|#Q%S^_0}1E|*~7*G&fYTzw);XHex~hY%pHsSMx0&yvjakxqfn-|9dC|B
zVzcQM2)S;=Sgd6}EF3$WSHUu76|T-+iDLd*ec!(X=k6{<#N1Fg51j{_syWD;6@)dT
zxkvkyKb~!2ZPSkNvG_FX_~MNt&pn~_+8rZ*x?!t=E53NRpsa-RfT@mnz19iq_dD~h
zt0PKUCc(^%J@&O0kf}1oT1i9bGL8wU)W-J!EvTqz;}~;;iJjW;?AAfre)inx=;B#4
z_pd~1AR%`GmbEaK@k0g%0g^~|lElRMW8v{r46gG;v0OzI9V<k!(_ajCa>UW|SOO}y
zq&VXy1NR-WxYjNQhe#zn->iZ&-&GJdP8Iv?RPnY)6*q;{&?lwFy;5p$-mZp$V`|vy
z!(NroD$L!eAhuf-Be<8vEnX2K`xVg<tAt4c%19id3au7xOym3IT|FH{j923vkO~r>
zj>F=U^7wX48ZIBCxx-l!3SDBjDZ;oWVl+OT9fj1)kq}=d4yk(1OaEcc>5(K{)uk}`
zvJ^~gO|VT#9|I4Skf5#vQT9~3GQNpmd}F})M&!9P#$Oe}19btoTM4jtMgTMa31G-R
zK`3Yo;mi#o{y7@K*}0KWa}tKH#wfh~GYZ#J$DpKK6pN(95psVlUe1+5b*MBBD#);>
zO9t|(G7xQMY!f021%A(ZrpbbmW#Qc;i)v9hn1;weL0%ebhe_kRs5G2(q%m*46h0i1
z<c=#5Tv<5^Pj8Mu^P&+Lu}KK-KLnA`B8W|w1fjokIIMpUL+qJh>=hc0FvdZj6h~px
zp;4^02xEAtFa-Gb8EZvgGn_e`+wwSbX9BkF9m*RO!?0080JcqnXp$cR<7o;g7m$a1
zg&e%kGQTFQjV)`nQN#R9-&JkA%+kRtMFS|WHGs<`19-RV;o>arJ6biKy?qm~>aiNc
zJG3Fx%74De5X+pnKhW9?o_j2D`v&(nKCy#9ffM_rCqq8O9oxAddz}1qTs!Z_KH@;$
zk_$rRwGi%;UV!LjOEC5$_pNT)gt+dlIMKHoOGa>J5ebmkyAR`B((!dL6SWOFXxUhR
zDvuH@GOa+i9(w@I8{p#If>-O>5LJEx`#yHyn$rce)?R^{`E_*P?S}o~`v{MIjFBnN
zvHbiSe9-y?$s_$ZyI~ME%l^UP@-WK$B}j&D!X!~8La)b*)3HO6Bq=6KDjO9@uTz=+
z8?8c$mg*EaO^c-G>C%HR19}u?Opb*UDe{p8^;p=@;bvRj({^B;!i8SBOd+QPFKQS)
zowP2^q@!nMQ`e6Wa+tq}9)DZT8qR8>kc}iDwUx{Qc95G?EUC7|(G}NtI@g{+PVz|<
z5u8HS@%yOia2l=JpH5{fGiZulCQZF?nEu&k)7XPKbmd|$iQUPg=_m6kX(xB(=oXS<
zZ6VE)DWWO!i)hJ#B64piqNHO*bhWXF_GR<$BZ|m&Y7y;JETS>b3MsLO9}5a;y;>oq
zJS!ln!UED*R6vjP3uy9>e42HMyOi_uX=+qHSue>ay;=Dr8jw$07UWZ5bUwXk&gYza
z0q-^xkR|{7>J5byHlc{-J}RQqmBpmJr-bHhEal958GD7<my=jQiDi|voIOG@oI`JB
z|Fdjp9S!Ex)BDE_bX2v8f|fPY&*m1oA=pX=vyRg2#x^P#+fG*NkJF^bC#c=?6y=>d
zO((3+(9I)fX@|{unsxdD^>|$3ywhb0UD`>azphg0-YyE7aD%)WZc?&iHyK{P&9DD1
zZF_Q$6z2BO+q)0ww%a2*UH6#wi$0}^^Zujw(r2`rb5@$8U(!;|S2V}uHL2;nq305B
z>EY|QyhZVjPS1Q#+j`%V?Ai~c_TLBcu>MGH5g+MI$wv}y|48<-GJMZqZ{ABjo4t}p
zfz3F~{4@d61=SI=NE3$?bfL{ZQ!m#U;9r;#L}#<^%NuplbIozv&5Ctx8%T36*eTwq
z`n}5*XRRlp>?7|8@|H^33TGS^alr!i8UN(&@K*+|I9kQK$r^4@na&ue(-)Ed{7_ND
zd*v^A%fZhN8P{e&ZD0oAHv_X}eK5z`6aEf-?%3#s3)Wtc%lF1d0Uz8i_d!ga57h2@
z<9d-dd=Gk|u+|egr5*@Yn}Ywx(RD!e{J!n>-dj>Cl0wOTuG<VDyKIFNvS)r~3T>74
z-iwIrWDD6Vgp`p=QIbL_@BM#|^F16oqVasL=bA5m@P5v<iQM5g4*%4-_cPlR`+6JW
zu$u|iC-p>DlR0i{S>Ur}U;LKn4-fx=Xg|vN_&J;j+ddBSWhTS2at6|F&&4_JTYl4H
zIVvPp!Bu4~b}ZWfB}?8uPuz-Q>f3o=a|gU1?uIF6lb1*N;IRIF+}(cw6%_|@#L^$L
z*9IVmJvokFg3x6Vf+rq_Fd-)tJL<wPsNZ39ZRDPubCDR@5rtl(V*nrCu(}q93xDGg
zI5-jWS0-UzR5H$8OhL-)R1Eo^hNRAPG)rV+=ATTIf6hYgt!#XX&B6IaxtOPs2kEnU
zke_@6m+v1z7#&4>!ciQmIf}2w`FL!f5BWLy*gZQRHpF=>E&jc-qr5+M6z_T-#nAjC
zcxiqFOM~-p=p%DAR=H?%%R$YdY`o#j_tndpI8c&-?$^`dbtVn_Q&VBHKLyK|a)0}9
z&Vnf=qTzKs_qoR5z=l{v4~s!@R}?~SMItaH0zG&~D#j=rRyCpcdGZk2c89=tN-!oF
z1Yu%r0QX<}<Aa|cH0K?}*ggj^;g>JI-}b@3=zVx{XD<p}cO(1D4mi)*jvE`c;*sJe
z?ps}twbtwKDqtNPz1KqWfG1=*Cl%LkIV$fgh4D^Te70MRkKYzRGJhVrt(^lk-W9Ny
zn+27c>F9Vd6~A9j;a!eNaF=pGndt;*ao+jIy0HlSH5y(&z=pFS_%mw=D$E7;tQd~T
zN<+}|yDenht+6%I67KAS+4rCq-d-?;?p+InIxt6a#S$IOePPA^ug4$zut%*Ij+pht
z!Z(KS*I>M|K?_%7)FE?M6=e-7c>GBfb8f0)k*XT#t{UT5b(~tP#h*(Z7@W~X(kngQ
z#?VKcl@ZRCc8A$vQ@l|!MIhssQ>V>%cb2gXdveydnju`p9Dl3K;BdqgUrm`mnAj6(
zE+$xGW(-w{?%cOv$UV;8nPXzTOTh?kdfhRNbu2Zj-v8g*+jzMf_V3ikh6TFt8mSF~
zDlG(kP{X@@DzLe#1kVsfSiWLkN`L}P-8As&hz71QzPXyHjdx$QvHPPo_C;$UPL=)3
zHhmD?lk-Y8-7(Xt8#>nL!-DlGY1VeWmPkX|R~l!oNn^343?3hof$U&eROZO?tg9TB
zEtF?|PXViYD`L(MMMzyxg4-<4E^Sal2IrdQ4AsP=KiV*vpv$`xy4X*8ynCmI1L=Ag
z;;N5@PWrItUw^mO#~^QgIK=6ruaZ8x4(gz=QU@V#bYOd52i9YCuqZ?uUd4<p!xV7;
ziX048<e@cH4!=IjVs(-%rl(3{jiNLnJfyH$UkYhcrC@WhQ~bEYcxRvt_wq`?x@VWz
z?9nOo7D=Emyj^siki(3`F5xyrg1?p|QNLUoLx;%1O<5i`Q`nollyhFJpKbo5i{zO)
zC@<A!zi%I8=W-9>9SgkZZ2>Fe{%AkQxJQL|#w7Znx_ftAX*7m(K67`<eK4E59ZlIw
z@P3pn;<z8VZtze9c5>!kV>J8l#-YWR_cuRJ!utHFX!e<j>!X~x2WK9hom>RLTR0Pg
z*_Y<E615}N@J>5>*pBYTVZNJPyBmn}pM&rq>@f0IMnf__9xFE`qiJM1Mrvil?q?pB
zeLjZ0pHHHu<qTXkioorfX!g2}QMYa(sqa1b<UYc}anCWQ^A#q%tw2HL2b3w+qQUhm
z#QSE>-u=SrO`WK}AxXb$Waxmo0zKTSOsmV(sB?rC-A&S^MXd&Oag-62dYO=2iW!++
z=uP!^Eh+GgHMKSmpg9J1G-J*XDmXEm7W60D@q9F;<=E5HqYkwGlOriQ&7|M$&UEk1
z0*YyJr8$G#DLToMa*Q{S|Iw|q$95-4Wq4E34qw`!c96b>_|y2Gfz*F|2zTX$(zB#+
zvd@a3{fDFIi(3q(>%~#c(Ri{kPNW4JlITWqGEF^_Ldzmjscu0U4U$Z!=D>6^XiKM8
z4jFX7Cxb$=GicSh3_8TG<#`!Ip&8_`I)j2oWl*(B26rK+)7q4D3S_*JZI#aduQW<8
zPNN$EY2>~zje-WIk&Sv9E%=#A{XV9W&YM)Kf0s&1-%~j(mqyd3q>*=Q8vX4^qrC0u
z<e|v#_e2JbS(Qn$c3JeFayG4L$)=*Z91``pbgnIrzN#FhWb1rNo^^~q?Kw_|PZW?u
z%?Y~7IrfMpr^qwoG|m5Vh7JurN8=8hr?cf3$jI&@Wri2iGT!H$=W>NM+-L9csOvQ1
z+zpy;SHd{%78MM+L$$@FG<*C#8ujQtHOzlVA)g-6q;;IHYJE!o?R-vhf6J)mzn7%-
z{S|pHeND!9-%tf_JLK(sM_cYz(m;jx^k&2d+T`+)BsYJe46kbD);`nesWt40s-<18
zYw3z_9hK{Up&4;s=(2P@_b1mg_f=1iIB&lGbv;cr)#8o~J#<uaPsHPH2>I0mS6B6f
z)kbq{eAgS5>~)-~VujH+d6!tl2G;rmp;uyuQ`3iF%gN!GH&kFFcW^PChMbin(V@xu
zPRS^YI5rxddSl=nJO-=SlQVt81bn+P5$4Nzqxh)<)-|%fA!R0K1D!ydoUygU8IktR
z*ze(lhbNuz-oXijvu9&Oup@3XPsXcRQ(%)Y748kw@W5yW7N*XCbJ`3H9z7kJVN+2R
z>WKRncn`?S5yQi{qvV7Gw%?q9t=Gn)*UvFHlcA4&X-25=HpbN!6SVI!L;0^>$ThQs
z%P(szTxE-6E4c6cz2N>V-X7#<T;<Mbi1^8PtH(l==I~a;26xQcy9(zT)<C6VJ>Nw(
zflRh?M#u~ED|g_2>u&Uv-N$!qA1F3(C*b%4Fg|<`k6!p8TQ>lDPJwt77KDli!FZ~4
z2(KrH;%Q(Q7TpWSH>C)Co)U@Bz$o;;9SsSISU3!g<GreQ^o~xzz)Oips7OM1Ycl6S
zQ}IkS4I7kshrBBti#}(d<7y@*9?HVzso5z0n~hb;Ie6GV7gLYtVs7_5tlgf6Uf22c
zo4v_KN1)vQ2nwu@Kv(w&7JSV^{>eNzyXE1mWFE$a<sv|tan1G|lswBuUz2R~VDFFD
zp-gBNX5e{gI?`XH!J|AC3eQro;bt=S9Z%vNpG0)ukbt?9;*s1V4riNUu=r{;><>jj
zWnm;7EF!S!M>v|Ugh3-D6!HrX;iqK?Dt-syG@ql35(ALWTMJvpa7Ie|AQEc!BfQ8L
zeM2~}=DH94UH4+g-(7eYxr03^+cC#~D|VmTgfRu{@y>cJ<j=20$%8eR*yN7b_?58w
z=Y|bimU2&rD=L~7!M11tZtmkVeexW9)^ozd2F}agn-0$lQ!(KD6xiS4^Z)ZiG%NB3
zQ(t?84jqdI%hCAONN5@%_!&P0t%HVOwKZ=!t+v6xcFriX?$nz-5cBjcaK73cbMBg=
zU`bEZbs6LKd1LIAGRKLCzL+(<FK+qu!MgRmFsZ}@nbqBp*rbE!BQ%j7u7=n)72L8=
z!B>86>+q{7Ru!g}YPiE1$Id1VMDotjyQf-MY|ENXh8~oPbYWzoi@INiIMiVRn$r`1
zUzj5Po*8t~&CoZ8@lBo?*2$TnM8gz4R`$f%I%8<c_Q1l!Jy1B;5FJwuVSCIF)7lNu
z#ds%ovmqv|>W2Jh><Rj*i;QJDa8uDn_FoOS?9{^CbWJSu)_~n&6|7&N1l#`<aq*%8
zCVDDxKaB#)E~=yLv?j1#8;fpf<JSPryqjsk(nb>#r!h_l;%E9Y8BC0khLeUgs!F90
z^N?{tH)%|8k;cRfX_RZqAS;o5IOejr${Y@5%R$#l9%Vxnus}%>->)jdY>yK9EK|k;
z6E%!*RENTL4Ln_-gFi2IknO6AO}+GBH(if$ksc<2&o<_rrj6CZq9=Ngu+eAiq>m?`
zbnue#%n(Z*ROINOZ-5T>ZR=p&Ds5DaP(WLoJenBK6upwg5Is30FmH2AO$zPjCE@-{
z0=;fa;>9dU40HM`BA>kyn=0~z(yct<^6RLGOL!~v_E!m&t<_?v-%nwx)h?d$+x^}2
zPdJT`MBgGQ-ma0xlq7kiT+-rmRvn2KHLzu{E@b%q=U&yqvmy)J&9Z>9i3Ml6d*h#x
z1=@8ik=U&lB=_iYE|fX7ecVGh+Z1_Wy&04C#Z$+A@ELA{ELA)1yBG|~2h6Wa0WMq&
z-SOTYO$!}xz{nBy-={-`cfm4!=0bAWLUea=#e;Qjyp6sBYFVqHRks0qfA8YE)jkyY
z2cqthKW-I<qun72LuST7dtee`rP9#nWhS&w^BI$rk2?`3cw6)|=PY@<<z6x8F0Nwt
zxDsTglwy(2LtM{!imUTqa_9S7T<`rJV<&uOKC2#kCB9=uS}UHf??A>m3CemXO{@b_
zsZgR)e^q+$MuT?t)1d=i`jm3UkgBVD&_{VwTG+D}b=dWxh6z?QXeno@0&U6f&LFyD
zI*gVk3+gv|6nQF-qsy`rsmH)6loB?bjLe*<xON^bd&dQCoy!<ktfcDbwe+iX6IIXL
zPPgKB)4z}VsAI)`+WGPzjj#=%=q*8HkQPFH&W6&f!f>jMilFC<qv(%(4Bd%}rT3EY
zG=6#lE!mMshWnG~<(gz#IXH!+D^p0YcJ-byNGIcx>zmSOdr}&0K9fdrm(pm@*)&qk
zOrr;WX;kQzMghaqs6#Q0g3D7WB|DY8H>Z;Om{i)PkxKVJrO@``6xth?LZ5e~(B%~=
zG-P2470pW_wPh*nzfK{Av=rL)CWYerrqYVgRQjxtMu<xz4KT0MlupymW{_M^CVg4U
zTUv{=sd#=4tysd|W%fg_2{=Ojc}Gd<aX$6#I7UPJ70|<FC+JAxNjmYmkZQW0ruA-T
zXw&ht)T8qpW1$PAnO{T)6^rTD+DkO;#buhy`Km{!uhHVZHz+#eCjBwKMeWJA`R7ua
zl75#4_PWoU$^-T~KccmHk7--)r*tdj8M$kh(H!3w6j{$1E5~xG$aqb!zP};K!4))b
z%{zJ<T}kn$swnH)dosNGfxT59Nh;_QwK-SQ0@cqHed;rrjjy4g>ow$QRZDC4*3!(&
zwNzbGOY*#1o^ehAr@3Ez%29Q^8>@k9t(rJ(rHQuLT9~z@H(GZ0fh+fF1j+Qnx}pKN
zUv0}h+jg+g9|GS@&QkRhI1>vf#&gg6)R9oB8;PZfqp)(#Xw=t@#-XKSa87a@wuO(!
zsqGU{IKY83nIn)B(jgktyF}L?3CvBFLY=h?cWcVy#tHU0ggfDoa6<DD?gD&11=HV6
zVjsmQNV4@j{5iPuP2iQvP_(E5hgXl}zkd|8hmV3^q9bN2JEFm5GF-Mfz;w_AbSeyk
z^BOJA0qCJ_vOW$Oa(00;7o}}IG5n%AWdB;Q$HfZs2H4=U(qMSZ9f4Sh(Wq!0kFR>{
zQ?{Fh)w1jla$byRsbw%%_dwgBRhW~q1|!$3$2p5l+-tf8hrW2hxM>HbHt)ueci!-t
z<AaXlzA)+7k3nM&!eE~tq%Zp;<aYpy2L-`$LoihGL*QI{2xra1P`o4@Wt<^*;S71M
zauhm8M<aJ*4EuOvp<WV)nV;iP|1SZ$>}ioV=KGUL3Oe;u;VR3TRc#txUt*kdFay$K
zcnAG!CTat+@IW^kwh`HUkITlAdF<6m&B5QY9N7HML9s$Eg5+|M)|>;KdpXbx&%syw
z9JGASMuJy1R<>oK#Ul%!Ze`-TUM4CQWME5JI)W~wq4Py5j?|@KT{HI(Hz(mnT_RLo
zC!qR9JdAVWc$Y91^~+))Eus;m9tF!!5!i9|Fy#EhF?oI%GWv2KaR>JyJ`83bD)$j@
z<&DhA0hn&=kM@>>Xuo>^Mrr$D&Arjhjy~vTx)0h7d!TrD7gpNsK;C^Xe4DcsH6@#{
zN@)YyhpolP^PcE`Z6(yLcn5spa#$BGL*2V2xOu__FGCk$q2~hp$GMTE+H>%+b~dJ8
zn~7Cv(=jAuDsK!<fl~A&JUcTHlRu7!x0XHU&o~QcI~q^E5T5A>eAzM>_U~<RT9<pE
zq=sPQ$o^30E|X^VtV9y`=f;@>YfW*O^&GL-5Nl>}KU}mCeue3v{ht<&S!yD!r#eR7
zQRSXx6-<=Z!;a_t{C}v9-g8xPrBfLp4$9bdT8X!fR3NRZ3jM9B2pgz|o6#EZSgDEn
zaa!0Kri=U*_L(rJviF$|R=Dcm=psW54l~BJlpe5+HA3V>Q+#4f^Ui~{D1QE39Ak`6
zS9>u3(1ZOcMwnl2h$k`KP{(=crn}u>p=5}E28Ph#XJ*ky&KE7vM`fHYvJ`co@K_Vl
zk2UZiSOc$3HE}9YgMCQseeS1(y>k>{6rzCCBx#rj%3!*M9P0bY<M~i|lwOvD?KswV
z@-*T6S_ASo<?#7GS#&zfK<AJ&R1~C<c0~$}7o_mJMG96!q+z^E8vWX&@hM0KiUzW1
zV?1QR+E7Cy>r^@NFso6319L{NCosOzQpV9{_PRV%L+)vH=s4?OcBu|dv1housUAK#
zaF%K||2#yGdy{n$&HkD{K6-F%(u3bOJ?>7{g<A`2ZSC5;>8%5mIXVdL(uUPo6^xQr
zKo#p!E(>LGB})dc8)Q%(B!$OcB=LNmB-XBwKu>E)OtNYhOS9WW*okAJ+$u+GkIE6D
z*+)gix&q-UUnxA+eiZ3_e~L}l{)q5H9iq#rOSp$i;Dv$&g0ky`)oT@8o~*{cZ*_DA
za{qCl4)z(=iIR?b5inO1P8WG=Y(4wDhxf+P^xlwIXo3aPG!c1D2h!YID64LSf$zC1
zU|26am$HP(Snj$QIskDUwrHC<2pBk=bEVwZxM3tt7>+{`-}gL&Cgar7Y1qk|BSw02
z5UIQXawaZFoz5G6aUM9Wv6>I_4QLp*3wL$)aSm)h8bkciZ$=pA8${wacX4iPNx-Gs
zDInISV>jgB#LT1kZ)5>1hMhv?sB`Ex_adI|y@GAUH}FvH4leJy4^!DE$U0tz!tJlI
z&7%?x|9wLJ={gKEXvE&!7WSL}L89Y7sGsC6<43YIrb~e&g$ljjrA|4Qwdn63UFz4r
z8->j=qWCQ)bT-h8B$Im6mlKwhci);Ww+^87gY9Vlp&?`<H-bu;9`Rf`hNjOOPfFV+
z(WDzwY4_+^w5MSX8Qfe*Mc0<l<O+A%uDqI7j@v*bTep&Y><-$Sx|cTJ^q~R04$zj3
zel$NffclgMQO{=~G_@p@ypqFd_~HmM{1-`Q4n)(Fh8PMM6i15|#na&B3ADj6ksj(K
z(S~bDbbCfJ`IRJ7bk7u$VocH!lR`<Hy|OJyp$dLoDoUZJSt-=z&v?iqg?vWx+h{R<
zsYxcwi^=2}n#{TIWRe<{Ob5Cr6S|UUU~LlZd6h(So+MGv=SdX&fnS|TG^TGdonMj6
zeRj!o+$e?m@cUY5o=O95rqa@_X_P-Coj%BBP{XGTI{PS-Y}ki9?shhnJ;|YGpK@tL
zS00&~9Hl!G@+o}NF;dCkUY?f))K}vqWlk%kZDFVA(u>oS(DN+y-*}FKO3u^y9z}F?
z*F~C8SxoaLT&Bd6SLlMtH5wmrol4|y(lPH6TKeY}jo5mJe40whYxO;vQhlE+7C)rE
zPao0haZl(^!BeW#eNH{S%INB&7gTKgigZ27>2ca?YJ9?b<Uik1xY9cstyM{<m3XhS
zrHYF0y{C8KAE<KtN9tGek?gmAq7#juNMlYl>Eu?^PTqjfVC+-W^D`YuR=_MrB|NH9
z!m%65XjM=}^K1>ADA2@>VcOh9$zB}xytNVQyS@V;^?>=$wRZTUJ{TXv2V?hb1w_O)
z2vw(g5x=obq|Y3|_k$6*VaXjilYsHQgu4c#;l0%ks*PVotx~fXy!MyqoYWyy!@I<^
zpAw9Xr0~f`1}_ZcaOs6SJeMk9$8QB_MJVF4gA&&NRm7_O8oY~4FdPYZaF6om0!ifl
z{wmHlHi$cRU&Z{%-$ZFeqcAdR5q&EDh)W$3aCeYJxU3Q`gsCB7pB6^-*29ev-MF*e
z2t&Pi+t<wuGm3hnUua*9spya2@pf2paTxdcjl{zD?72A28@zjXb4cD9!_5|8>=_qm
z7caxg2_A@8w+h4Ot--Ut>(MN|32ooEpciAD;))$Ob9pzW9`r_aHy`A!@MWLMespU%
zfI)ryuyC0_LK9fmejW%b73Oiqhro2tA)G%G3XO&^B$ymV;j9Q8-xrCh{3t{}i$=Gf
zF*vOrhpb+_e{Pom|6z&vHYf>E7RflRmV$i^DHwVw6*~LU;KIAyu9fN7!20<A^ed;G
ze8>8fiAy32P5ZL2kbN`PUuB`Qmb-;n*D89N1!vZr{%p=dy=4~mJ<NpVGR}BcXW-Jj
z47@2$ho@#bXK~W-EHV{zgY({>lM&UKgaIl^c&Eue8}$T~$;RV4dxuot$H1Grh+pMI
z;qie;=q-)Fv=N7)tjYMTHWWrz4#7Su1d*$QF>Y)SHW~!N`nx}8zx>cQoqNT09>7kg
z{kYW6m%I0U7`JgB@!CCDZnz8AO8&zuPcM+%R%k?R!g<Cvz1-JgMWH7?^jQguW9~5R
z;LZ6r%g{V}DFRX!!?|H0oPTgW@55Z!+;+xC<^+cvoQZL(relWFRFqAb0#m0+aM{4v
zJ9RvQD#zhLkFmHlVKl}K9f_730&hcx!M}jFpssNDtWh73X)nkd^nyNfGZwwfQ0Bv2
z-~dxdNSpBPyAdik8$gxs;&W@eVf_HsKKPDqQozq;TSXjwsDK~-3i!iWn{^Hv7(R}*
zkPpfTW>2b_ixOIul(1V*2_Zo$IB--2nJ%hmT&V_w4;nanLld<pw73^j8%guD@m^@7
zPLrR@CE9pA)|mZkJ#crA5pMl4MD|bvgsj$qCTm66A>9x)v<IHA;x67@-SI=$5S#sU
z5t46!$zKfc+`Jp!5aXQg`p{3-gYIEn_Ezhnka?KR*EBKEM+1G<YM^YZ2He+aK-x$H
z?G1{Ie-&6Cl|sm3X}G<S#vW5yjAz}&#a<S7Ib-zll^hy1<dN7Qht(V9xDQJfJ%-9)
z@pfqh|CEAbt`rnwrJ(wV?>8#a(48cWtM8-{;331i7&7dOmE}CW9IVgE!6r=}eU2$$
z{9Hx$@G7D6o)T-)oK<pCfpV4#>zFDSz`mImgH_?eyq7NLv+s@5z`#XXu#nMVFSIUx
z9M^<vohHgBYa#uI7AR92haYHT;X`dG^XF+M>uEO_GgW_<hh?W6)LARjS}cP_Y0Nj>
zlY|o_Ayq7asGa}Bo^zd|hxu>u%CAgZkjN3XJF-QSN{+~L%N2T!$ArwEx59ex2XXDh
zCt)4@OYC~qCThZe3kSOnv8a2s&|F?CQtqg5r<EH1vA?>a;GxL8@Jw8!H)7%QN|CXp
zQCL;>!s()3(2DN`%`N&UaZp3=91Tp&(?+<SKKvsMVK&1UF`Z@@_R9k0_Ezkq<3558
z8+;#N2WOXI2%9njzeW+x&l>|-i}7%4b%5?uNBlW616A2hn0s;_qBuu&z-lREGu_d=
z$P;sRu7`UibE75R=xwqevIYTo|LPESRUL->t7x3P5D$ysWb|3U{IOLwXS|Q#ENfIH
zWhZg*=^1!cUSO}>CG@bp4&9x%aOT}z?&EocUVon<srVH-k}IH+{Q>_ytKp8quNaZj
zjMC}95M<hk*icC_Pn98$tMa7LphO);YV>i2CJp!2p(nZeG@{gyUcc`_k6L<Cu}m-W
zH0VQagRN+*Yk!)aWJ}aCi2kh{Mz^~J^(`JnrfK7-=EOw0+c1S1oMzCM8Yc=)pHI8@
zy3j-~HwyP&Nf*M_(xpW9;lz89llLwPHs440xB1eIGY4p9l^=!w44{mbAUgOqgjDlG
zsc~sIJ?%J5n|DUirMFS^!8nFC*vHb`$#GO~8&6g3@iZncf!swReK?g!-r7lI?V3bi
zqnO_*N+OBJNwn)_5>0-VMDt3LXnR2tEsjW{#7#+bd|DDU^+}>U$s|&%NF;{~iL^d8
zk)n1cQmI=a>CZ}}l!=KHF+P!IO--cDOA_g_cOvPYNTe}MiIg=qi8OPQD90q3z8+1c
ze-7*oYE7Ymr&FoBH|tmI*O|w9SgH+skgc<5Dd)1DjLe~e*|}u4K94qr9U-}MM`<hX
zq;KSXtLGC6C}i&m$}B!feLD-u&i*u&aaT^acV|h0y~jHPFOYjp5gl?UrsKyh(K=n;
zA@{jT-M?QWxn(yf{P|7NakxbvF5V^#D5V8y?90)(Pf>dw(9fEOl<xSLjvakMrqa(y
z+4(tbN-Cp8Z(nd${uO;4UQVayyr%IUZ|LTVx3qXc1v$_=y02DA;z=d7dROsg*L&)f
z_nz`hK2XB`50wA>10|||Bn<mVYo~l911~;rr4%_^s02z@#^Wg}Xk4rg)$W?e=CgF_
zMhl$mZi!)o44@R(B|^Xa680Lc;_}=d;^ey~k$LHxaNp1%Zr}ML9CE*ib?0k^*_{#4
zel`M6yD>L9oe=QWmUXFWG392R_&Kaem@jS<<36;D8@sziukRAj<1;P?GO&^5>}7;J
z6y_*kQi}qn98^RW>+P@hEAqWj9uJBnVD+X|eARCf6F)VIm(dO4+Sz*1Cet9o27DDi
zHaCi@;vZsf)*n&Z#2m7nETnT4@$ii*#6C?N`O4@0JOf<c-yO9Q#whsN6H8os@t%++
zZxQsvFg@-<R~rhR3&b+ksLsuvh@dh@Y%-pOCPm&(-nWo5o36-LcjFyv5A1eW1$*l?
z$X8j9K5ZMJS-l01nOiW)+yNi=-AGpV#>waV@K@g#*ZCfF{qO<IxyOE4Nq-#UY|wPh
z2rbD7LgC9`G%6lKHbOCdO&FZwc$?tHVf?6#;JZ{5#`liKBE;bClvq5U9mn~sc-)zs
zz<c0{ur*AA>JR30t|Vjm9@dI%QnB}WDw<ZMp>I<fGCk5!bw3^JO*7Ey!5g0O8Mt>b
z14Z{T@Ub)llg?&9Dm()`@P=WU8JK(_o%d|gvFrtBwx*^b_Dm}L)Kc+nVG05glF@^`
z!&}=DnR8A=hGhZ_2gl>^usFOJ#GawPF{tYnjghiZSp78uCQq2NVP5-USQzfF4aLPt
z?6omxJ*tcMNnZ!z?3n;ahI96NjUVzH4&tZz0c?`kkB$l-tSH)tE-P>LChvw<>`pjO
z`48LvY(rhf7KGYwLgvf$SUz_xE?@J6h1p6BKI{(7ie>1sbH&sCi?HIPD{Qz6`{}^>
zP+T~dpHt4fm%)7C@R@jSI-NI<rs9S46xRAC;fvu!+~=LxQ|v{Le=`P~`j3X3+eqA=
z&T*bA!_m2v`|=;Mez}i3WD+fKD#o1moOu6ft|@9Z^u#&Sp4io3j0fk9P{Lg&N#pcj
zVy+D-_6vJ-sv({6%5L_Y2J+YRws-|ZFHqpzy#hkq<nWrGZN}r2Fz=!wd_5JhaHb-L
z?NfvmWBRpMl<}XR3Th427z=8kknirpg%-RXY2o4@Eo}Rug*!*IuxgVw%*?yv(2^dQ
z=3#`g2Zkt+H-L?X4xY5A!>U{x<|hr2=G`5K?-*i8-)^Y))4@4M9eA(Q0aA1@>Z1<q
zHFdErP6z5fjF%p2qEtfz`kZlgga&?1)POVh_?9ow!1nGMFe*~Sj|EaFI3tB^+S0gl
zS{j+gvY5%=bCzRev1foBRGE8uTO`M4iySshltWL(J<5_Ycs5rWQm>@Y<ST_z*T3ST
zb*Gs2Q3~bE=`<JdYrYKjSIJ<<Zua1q$l+D6910}maXnWa!|W9x-Y8%ke>~58WG_yQ
zGJaaCppv_1q+C@{Y^;jYMrts0VP0&K2I|K0ImUkH7g3yzTBnJ;LQR~})WXm^TKM9w
zjf5O+{5YnK>R@e*Pgci;4@?TP9@dL9TyvRc8qgty`V~^xpUb?Bz9gQ6N+4XLORU}d
zOI-NSD%#H+72ArlgnLAmxV0%;41Sy=W-rMX9$v4-bN4qwPx6gunf6-vDwd1&!(NM7
zChtVz#gF3RjvDdbolC-C-BqzQqeO@&_r$!tPsO<USHionLg>qW6SkK%VBSv)67|}+
z$~n79*{T@K8sIj6-tuI>_YSpgxW}D@-@ck+oT>$W&gqMNQ~I%vWrKn_1Nq)*hpMkb
z(Cwr^SN<qO`i;Y9mx)juGX>pjrsIU|Y&@Jk7cPev;&hiQhNmw_Y4|F9ZC{6FcI+)Y
z!@UchzVO=P&-*kXFv$x?KzbCu?u*0XnMv5zEe)&q?j+9|Ra!9T5jLDa?6T8Xv-~`Q
zwim-A?J9P?zlq7B6a!8^z|!$g@$vTy_&$Dv_jjwX{!2A?HrF#p-GnuEt;lO?N1&$!
zWvrE^MSJDwTY@6hT~?tdAJyrmtTrk3(jyP*Ms_odXojl^HLo?JdEUKgZGt6v+_dH$
z$N@CL$&QxY973^9BdEXpNV@la49)*Io?fd=Caa~>=;oVQv~I^->M?8)HK;751wY*B
z<vUN>duu&4rf#7&_y0&*Z4c?)-$$)~eQ93zgEXkOKZSJ*B)4Bd6mU6&?ye4{ubp9Z
zciUlF`80wwG@{65U^I;#5<@CRvHZM`CHrXh=k$%IyV>y+tCT=J7xGqTOaf_MPoT9G
z3FK6tK&QSY(9+NRXD<^d_F4kX%1NNB0SP2L6G(n?0xjs9z*`3ir2jph3d-Zj`gS~}
z7R6KVGx5}XE}rJ!jHmK9@udAPp3V+Tpy^u^$e=WV9`;Y9=<G!5W|Kr|cavz;)@1tG
zKZQyd=bX96dRKlLX~(5g6#IXiLo>-ECX1@Gvq|<6_v&y~{qP^|LS`;U&oQ5l?&QtR
zg5%U)Q$R)LC#iwA$m5TmqT4@D(`x&(v?=Bso&9#6BqtY<THZzSQ@TVF8!wYw`4!R{
zca6f2U#Ab<Zqj_e5*qpA7Cl^iheqDwjFr_r@(8(4u3sL|r7@4Vhxsx6dh>)@yFa5;
zr{^T^Q%3fgFKAuSOUk?Pij1$6lT`j|iVS!|Mswd%n^pz+U#TGb`R}Np>K!d|uB67Z
zm1H2py@8{u$kC&U<o~Op@=*%7Y|Qr})|#hYR>1pmWhDBj;Yf%ElE-Rc;!YjBOEkdF
z|CmEA=n!|m{t|IgtzyN5AHw=hlUN@3P3X^U5P^rj2>-{mLPfqt4DzWF#f5#*{6oOb
z_k)<9^j_R({V49H)r$4=8^yI9ts*k7UHsY9C4PQm%n~DoJ)>kW;jb(rcF1GtbOq>r
zSAgLjMc^J|96LpLZ<j-flmt?z{uIl4e-lHFH;75kz6hW3Uqt4KFQWZzy_j3|O(<)9
z7aIG13B@!CxSo(f3-ja;n3H-Cq=7*-+W6ba`G^cdv}`fL#Em96(`W|AQVW!}TcOg&
z2Irgy^XA+L_+^gbUN3vT_fE!Jhv_g=U_X%iJl+dmgyMQv?vQuG1M)yezf~yEUW2uK
zCJcDL5jt14;9iIqBqr_PobfKqIj|QS5AK7>O&^r}^+oI;?x$OI5UNRj2z}%aO^HBk
zvkk(gCBaAu;VpvmhY(#AiqYH$Sl#O|K8}e%#JosMUKNG8o1-DWEe6@_+1<V@4mT&q
z<58ak&aEb3^o>M}+>?ZU*2!3PKN)je*zfy31uER3Q+}Si0+rJ+WqKN>F~>7FJq>S8
zrorY!8ippNLH55il#WZomA|PNmXpeT6{+kaNkOZ93Y1SKW2jm(dM`;rP-Y^`DiUx~
zk#kY~;&E$Y9Nx|2u0GbH)R#x&v}+VjIY**&LIiT`4&zVva4eGw!}gj(C@BrW%_G71
z5Ez8*p6o-K6oBrQ{_v6agLdr!Jiomk=W=|J<L3jP75gwcWG_;t>_(>iPMo>19ql`}
zafiVc?u_1u&HLA*$!INR=Xin#U*Q|(j<sFO;5~5(Rv0ga!O8_V&RUej+qqa7JO?{U
zov`NhESz~c0}l76v3BB!u+x+Avd{qou1vt%N_*VU9|xzkW8nL86wZ$u$<I{Yu--cY
zSKbeW<CZ~~plJhJTPyf8->1&r7Kh(Gk)L9Mj96pLXz78fawAwrvwvl(KD4EDu)SUb
zzaOf?k@>@-b26A_s)&Jw3K+}W<Uk7r^y!etlyq5)h?YT6tOCyOXS}mU0YjPF>3FYz
z0LJ!Bxk`wQQ-<|tRlHAA$M#AMRK3td^m;8+<!E7W3VY{V8Ry*5gpRKU#tC)o8m9`?
zaVp#$t-|{T+^3tW2K_uWGz?P1{Kvd&-H$s3k85MqZfy+bPQk?e+R(kMjUD_9b-t<t
z(<wT*ZL1Bx&6@02RmbeNYPig=@I##KYE(yHZw<WYqX6Hll4#SBg4-(21HF;L1I9R>
z)1*-#Cj*j|MQnvEs*L1t;MzYC?$sp%BihBP^M8cdm<~}Qq!D{m3R2D;!t7gzxKaIA
zl%1D?_Ep9by3*WZDTBJlGWfAtmc8C`?A>E*R4B*0>hkbVQNYa<1w?jN<TFYU)7liV
zdAt&Q_bFirf2_8IsbHbAD$aYTqiL=NX7tj8VKHME_S)#^YvRdvO%&eNL_wMsY({Cr
zg+EV9xf+-@NCT@P)v;7k9XD>uV#0G7#IirA=>+39lET9S?3-zjfX|sOF--QCSQ7nR
z{7$$nhAzq!KKnAo-_4n#+vO}#IX_2~rXCSnbzTYmVK0R0xz{4c^0jbEd?}VTz7i)T
ziiD<Bu`mcYFBB>+iXk#rMe(<r;$ZPzvA+4SINS0<+<yN?cy~02h(GEuU`#g4SetoT
zMcheI!Sw<)T$!zj%*8tBbwD3e6AalyXad<XQ&^Yw!li&d{CT!Qa24-zEEovcdxLRI
zbp(DWjKq@KF&I+9`u{QRl+T_Dg>y48wZ<7LqZi<7DR1S3yJ5nKmGGLj4v&s+;a%EY
zxLWIj`>B4MwF>4==rGi6jD*e9SX|RfM8xY9>`CO#ie))iU~?3D3I$l*ScvNCvyf>l
z!f2H%=ootg`w!oSWam9>@q3J1n=+ui9Lcrscz^37l5FbmFZdh&Q)|JlbH6d&w+qh0
zq-Zo`DPpn$?Q&HnGcPq7c1V*}rRz|3p+4DPF{Jv^9<;fvC#6)I)6-51I@`;NzPt3N
z`s22gZ7`UQqz|JeN6?FIqp4lRo{TLV=!?4}oxd}K+#H>$XTyAoNOPg~<!)5iZzbJr
zUPDDkH&V-tZ6x0AptBBp>HQ?mQ!Uz0O$!dv9eaPep&v-s%YrC%M+l9O4JBW%Fp7T=
zPOs%6sJ2%m6<I`4qDnMbzKW)a{xQ_kIF<~OW9f=)9Nk_JN1X|AH1d8Nd47$fB@*$}
zq7YBt<l}kMJdTQ*;wbNJ9BWr`lzTLeriaIovsWBdy2O#}q&RvrB#zcu#?hmmab(D^
zy?x@yes~-`bBd!QyW(g{VI1kU#Zgbkc+xG5Cv(dLQaqVJ11BZYQ|5IPa+0XgGnwK<
z3itJ;QcPDW>3&b6Gxg~d{w;%E{mP^}3fXkYB!{dhmvUV4=+piqlwWX^KEBJRoy_HI
znp{A451t_3+b1br`4s(gX1yxo42}DFmba<S)4hxfr1GzbOk9g8^y(#gWp;%;{jXBq
z*K5=;>jv#Rcax5EyG7~TdmLYWhl=g)Qon$Eq*Q*NwwOMoJ<A`_*yP7l{_qJc{P~n>
z)t^(UNf{kCdqHCiUQ*P*m-LN00+Z9q>EHa<<Sp@rVq)IVeZ#kOCWJ9h!&{m#yn;M7
zRZ!p93fg*vI|QrbVBsu@+6l7o-k^ZBlUbwUEQhkP2GpgsaIRPu+owvxtni;G*x4ca
zwEPk`8-9xKeSV0onN7lF?l<Awr$ID~_`>&=TJh)bXEFT2Ct=b5quB2GK}h9Piolsw
zV#SyD;(}(4*m3)t`1t0hpz21^;@l)s{xk`{w(ml{p;_!MWA5cli=cyyVaB$K!(Up(
zIX!9YeJzD=iL&@A{ZB|gY8Dfu>xEWcop^JrR$R`i6?Mbv#Mo(H#Ivbig`#bf=zP~I
zWG;7#t52lxS3?fJdn@46NoBa*P{&YxKD#yb&?m$Yb6)XtX`LySP3euGjeViOXF^w~
z9g?>WL)Tuyf3ag>m@$#_B97eEIuoxJIiq&me7q}L#JT4sI5Nx)Znho>GhBryU8`~T
z-8xJ@y%8gJZoxtu=5e0=2R-*)sH@$>UXFd3vC0R-l6~>$@qVOB97LB5=bv5tu{S6H
zj%NcQUl{~@nGo#ha|qTBp(t_-L&%PB$OIq8?wANTCPv~`TogoTG;jFEa7Rrn_A__V
zz}iq*X*{C56X4gEy)d`=?BPwT;Ibs_oREyKImy`bJDIl<Q*eHD3hIJVFexDgrEw{|
zrImuo%To|&lLF6%WDJW<#_U1ttGSj0`%y{gU66?N8i}~(k%0RJ@fi0t4j+u;u)-l0
zhgZbFd`~p0f}`LY8Hu}55m*;~7(Ll%6SgA^OIL+rfb$^)j|{=zUcvaK6oi_l0GO2d
zqyD@fb|)UhpS=g*<+dNcCh$G6uMeuYv(NJV9?o9vg7@kji0Ho^I~%uh&*EnESg{dZ
z-PYsetu>gx!V^A!R>0lc9pQhM!C~_fgx_0?*)0oj&~qLpdGgMVt23nj%tqO$S?r~l
z0kLcvcDXvD$AZatJl6qzT_#`_cTpNAkHx0y(a4`V3QwLAd|ZL9TO*ij8-^oRLooc@
zKvb^pkCOYAXyA^4<@`1C!_5R$FMD9p5hJv}?T%#0?pT-0o(T4>`k&K;<ry_BPE>}K
zs{+PMlYp%|>o@%U<!PqCJv;Jv<R*=C3bOcRDuvV@@)+$Xk7>*0@w8ALBZLA{lI8Iv
zNfCq7ln@r7is^Fdc(|K2qWzl8<7+`<sTS6_Xkmbn7GixhvDs4tlLn}xIzttCy;;{9
ztO5rgRqXksg1%Q&;gG3{><OxvSI^xp+>JM2q#Eo()X>_jhP<ijNOf1oj+2~?TCE1T
zQEKohR>Q&@YJ9Iz!x=lyg7;TJ-US5=vX)2xG)c@zm4wOzN$g~d<69_&%wJM487_?>
zf2HBvE`!;8UvhpRi?SL2gxify@pfywXz~6dJW76xP|aUre)o3a_O@NLE$$H8^8Si(
z1zqClN-5p}mx9?v)~a^Oz)x3}JE3Kv#P=R6Kh}%v<e^+HkBci6ko$r=g-0slK_X+E
z&x(*YSHj0uMVx2dZ0u7-q{b;?&0^lWWK7dPMH_-MYA!bz%h+pT(rQh(S!hB<g}I&Q
zybB)2d=%rh2Yb{p(^(dWt)<{o^-stMDL8NCe3YIPZU;(W`Kf<mph}Ba-q9pxSKSm7
zlM_YgpLC(@lp*BhGsFgsETO_f>{lf6L=VHGV*c$iA%EeynBw(Z{7x(sSr%u+%7=x*
zZ}k~bU3N}f>TyvN(G?M@ep7fwl!~S69*MhlWg^kzwfI(DFH+-wi1JJg<hn`0<u!Na
z1S`Y!pehzBX>gaK7Q#B&i*?!nvlbcSBkz&9e>KPa(-u%V)CW85c;lhc27PxA!V})@
zddaq$*^@?Nge7MJbseB$;0T8yyw~96gms^I6UTosyf!Z5&YTsn-?bLCEt|32c_$9c
z_TkO9gZRXEu%Aw$h~&FVf?*8eD&z4YJ{fmrr!zLmM%R-(yiPubzW<%X6Sp(ix%2|k
zwp`*~plk5_R05Oncd_HfL;kvb#(3=&c6Yr+s^kZRTh`#wmIf^P^c~y1+Td*c7c2is
zQ13QrYHO3Dz%E6KP*f!zg{BY_ZA$E;$9vS>$aRnrabk|M<YtuU+ME1*EvfefYx0t|
zA@fyswEW8ustFuHC5|I$vF%tI#98rDx5?yDFpYle&!(<~xfC{V5yh1)p;5ctX;n{8
zax7X$b)z?vbAcB<|F@G)*m_fEiVtN(>?f<u2Wda~Q`GkWayS%3!I~koX7?d7d>Ben
z(&6;H$6-3yBZ8JnMN;d7Nb0^fit6;D=}25OmC3}Av`Y+q$%vuJ<uO#>6+_Vmv82=|
zmi}7Dl5*czx@{6mS(>pVFA+=hHHLg&#gN?17^*uFLt|27$Rr|$A_HUSxL*t%2#TSx
zF^r$`V`$sG7)ou8q4xf<bYp8QHQkG)?gQg!S8g0t^^d2=%<&|zN+6%^iFD*`B8j{t
zTC^{jY?r6d+Zn0kJApIp_UUwRQU-ZAXOi=ZEPAvjo4fefms6ZeFFxeaG|pIAAfKGq
zAERg4$0_bZ0VSHAq&psk)Oq|AX}6!I$CJ-eUD7#HY&lQ&9gFC4&P8&SzC>orF4Ov}
zS17H=HB#Jlotj_Xpjg`y8ti|I+Fsu#CCgHZ-guXaPTix}Z}(|>&xh1G`4Qdnd`u7a
zJ|PNtO7;hzk>R%IBsH&$CRn|o5si#@GG5ZX39rbh{1vTOP)>=Z<&@C-H5qPvO;x$C
z>B@uG6kqw8v0jHbmGwy+(XA0_+^f0tff6e1RpGvwGnu_);3O^08;X+9w)`h77IX;1
zcfW-B-JfE5SF?D(wMj%7e-n@E>&3WFbz*vBjaV-ISrqsDBu+bh5Xl-<;{C^W!ua_+
zF??mESR`2~Yz{Yy(#%F-dB0JNm1+`&1DnL+U*E-_InBb{;D^{|*dk<feu`(ee~J&9
ztwQcatEfKQCR(-IgpyvH2*3JCOl|or3X(nx`-snC%E!-QZf=b*KUph!PpA`xo!`Xn
zyRG6>ai>t>``{dFSrkv@=LvJ$tD2Qi6Qc&%4lO(iFhISn5$;r(!1anbwv_ilzD+-L
zUa^Jl?jblgPoQ|@Xee+Ojx*<K&SM(d-DhFPs5uCGFdxZ=i?MY05_k=D!*(+d^pRPG
zf$vshYW_OtuGt6}Zo$k8+YmVMKWN|C2~XO?`v%@{koUpnVZPYqu^%T{ulnBx_S<Sd
z1S<!i!6p#Vvx1<#F&H0%Lg1Bt2sq7q1y{rH^Hw-=Zn2N$3VU@6A~7K*3ME^kF==uP
zEVW}%{4^FW2jZ~NDjth(GS-=%fXMp^csC*u3h{|3tV=|YWfEr0PvSdl5*q!uBl%zw
zUTjIig;_~hZNgaQT_T1AC!&=x(zBujOa%8SGo~@tjmMewaTt3k7F)YwATcxsiOknX
z1Vo`OGZJbSB4EY;wwC+hFnAb-Nq0kGb^Q<)oejYj?gs0M3c@ZQ*5WuPuR7Ns6{9%&
z-RB@asIeZ^=8N)oKHQDH58LK@W4y&4_Al<lsjL6tc#szwoVMbb;bufvZh%qBI@Ge~
zNm0oYi%zY8!wh%WeprU0)k|=!VKHtjScnA|=Hd73IoRUj1c^VhQ0_GoKQA%=Ry7sD
z-=@I&$0Y7%oQSo*$D=@&cazM<V$s~uC^<S3j(rI3_XVtdMnFP)IGlzIMPc$Fyf(8%
zL!~vQIQ79hTXV>DnqbNK9@y8}9aD1+dEdJmPW;ymPi6IS*hvfaj1?4pRWNCXBDkd-
znt%U^2{IkRr9%=m>m~8;gCsV|OQZO|Kf+p80>P^#&?qIz8yAw8JCrfPZy6;2lEcsf
z1*mdf$W=`hdh68iWuOM`OwdGg4R_`A*1|ALE%dL~MA<w|=*`!FgA(V=gH;f}T^8%F
zO5-u}9mVsNp&-XRlD;Z7US`ZQNCn;5_mMu7u}+jKN>tQPu~iM_%=64zs)ollY8WJ~
zhCp*QR59l>)0n+2S<2XCpa6Y+_QY_0Xm*4oMs1VCw*8WjyWA{JE&L*^JHH5n$S&cx
zN&;#>WSH}k<t}kqq;&r$hG=$*UM+uwK5N@EWqygX$J#_~@-K1C{kK>U+b&iF{1uCu
zB@m$~h3dyrFwK|7%5^gM%=qu?Zdt^YbB`Wpv{ir0F{dSuYr{Az9<Bge=6Lr`U>z?|
z5lR;n5n7{&8P65*<rMc0A5_Hp=Zr1?Dd1kE5)|81k=#=qD_M76`cDHt3|X)4(!e~%
zJX<&uCTXYv%~MKfOP7Fu&p$$T;6IVc9=-+HlDOj{iBEa#6|DFv^jQD?ck;TZZ-^EC
z&G90AOQHyFPZEWL(!^Vx452bVOGq8h5t+>yqRKi$xJhM-$i>;hC+&#vQ#mfO9Zm{4
zxznQG)3ZWhM3I>3bXnx7-4JW$-w`{%JrHAqo{4+Y%Z2{@FG3-^Srlph6{d?gr!`ZW
z^BGDoAFYCr<BUT?HSxwr2P>`gk+-fJ7OZ7YA@hkL2Ig4K7-uB!kgZ!b0Com;C~p~p
z;>WyYnl}=`fn%|8*96!doQ!M7r@_5*7B=mfi`wCf5I$-tZ((|%%Wn;D`EEw)+#Sd%
z;QPkc15oN4h#A8VA;RP^*hm81Jnnv2n1ldD-UGgviTsVZP#&0%`&|X-@$M8}JvxUS
zFE3(f+Z9OK-sH394reRwV@%K!{J;yGRC|NP(p88vtVW^p7i=qRgp}=1_&oZ9KY{<y
z^Mn*>X3A1?vI5m6DN|sw8kwbNQcJQ9IV9`Tres4ZNbA8{0X^y2C3E^+VL?OHtf;?p
zf3i7cOMR>dlh5^Gq_Z6)Gi@~G*xS>*g$}eJ%#lu2&7htR&SZ6MJ~@wfq5mE%qk*$l
zkjwqmw0rOdYW3Sf{jO}Ml`nTu)&p;{dhA27S^H_f$3d#l@TZvU0N!v6A}9Y~I>)}_
zxiX>D&^?Scb_-{ZN;oCmIZV+zB1lgslIF!l(gpb_y0kosu|O2{s*9p(?PzMUi>Bs@
z(G)x@nr1pj)AgCr<Txps{Uy;PF(8`qjiYI+dNl2jjHbt}+@Jg{igLe1k$poHt#64U
z2WfsgqiA|HDw;IcMN>m=H0}EyO(z{<NU0!(O3Y*FYF;eG@sFa)INBN-Px6ilbVNIm
z!fF$#{#FumUCC^xOrhhksno=|@(RwTzt7L0|E^@xsFzvv;YT)E=;TuMh&<keIzlTV
zIAe7ypQ<=x^=xPX-P?A8vM-#Z!T$=0CZDFQF=xoH<}9g=I8TX(F3^cjMRZLRlZd!P
zyT4qfiqTgoH|iRh*IuXKp*Kn2m%EQkZ_#|EJG6X4DUI22mvpl4k!<OG8uj4;)iph&
zp)HTdwf-?BmvOGV@F@j&KcmV~&ndU@IXQ)vkw=dg<eB(_I^<u{ZMT=?oBxu2et1bO
zvaiTZ{}oYAl{h}<qd0n`S|lE;6Y`7rdz?MPdYfcnttJDHA}I{yvp3{cml)iuLtH-f
zOZ<rYDGW=Sg)cGx^8TxslU*;I0_(*3T{U9!mTFPG@uP_IdM_rsRSM-l72+2vM4iuD
zQ9r0cJP2zP87YlIrKC~JZf_KO%$r0L^EYFin#KBxA0kM(MNAhh;(qB*aZtBaB;9Wn
ze&%f=K&DNEJ!lm*-tWcIA&i~=kEHXCtNH)mf2E~FNyvQ5-a_bcJ&3ZC%FfQp43U|%
z_nz8AG8#rwQc+QqjA)Tj5h5BYB7L8q-|vst%{iyj$?Zn3*W-0PuIqkz)zH&rwUjIW
zf$C*HQueGmYFhV+T8xDUDE2GedHaJx=d_W@<Sxq5=!S=hviRLD&Vl0o*5i!|{$A*d
zHO>P>mU#$1kJ1qLyMk9@IS$k7CL#IEH0+u*3+L)s^i$8nS;6kn-M<w3h1b|-@e0^K
z6J13^OwdnfBfjcv!Ga0f&|h&U65pCZDSj_TY&n4K@`vCSV}a1|mN30^9G&W?aMsQW
zH)^e69c6>7g|<j*u|vuL2ZZQ23bu+9R@ynkF3JTPZ@a?msT&4YyQ8JS18ZA7alOq8
zyPCXF`oRaAgs){?iXSvQ{h_ug0G=9x30x(zO71~;Juw(d^MYZZe-363&SAJ_2>N@8
z?A7ZK7z_x7r10%jnhB=xu~4{MhGOH6P|Vg7neN_V&-FY6L+wIvXn2UwEY9Hvor9X_
z$I>1odO%MGp{6JhS9=FS+AskAwqmYH@WYtLf@Sm32lBtXF<IIhw-mgf*TYkA$~@2@
z<&NpUUGegp3zocfhU60`Y!UNZ+GW8a7Cj3q>_k55fGy^l*uc-l8mgPE@MY>Le3U(b
z2bGqXl5hm6rw$`;(Ln^M9Dru|UMvh2o+qPS@RJa;%|%mq&ffylR~zwg_j=(&SPR#~
zt1)2oO8gMHH?1&3JnS++#k^&rFKsEVonDMxP79&tH6Pp0>0)uhTzn~>gT38_pIofE
z{_D`j^qg7PePt$eEk%#@{pl#-sTiC#3A9Nt67okPHf9()Hw}SI^FZ8H8Gu5c{)n^b
zCp^}wNPMn>*l@8&Qc;5MS_O>VCMWs|rBNXNkJ2?d$m?_~O{{97LysD%{Z=#m%5A58
zcmI;c=zp}wwu|KRC9rOZ6e6ZbBTY}R|7%3wk?`5qbP5mJAVo}2ZKf^az96HfL0nr^
z;M}DGiHE{-EHsHR3zShFsD#@((paV|fzy`Vu&74NRzrlY(M=IKmI^R2Re-aeA|8tz
zL7<p(#s~Ef-LOh<zodlPCM8T&>?!z?N|>3Ygko1E(W#^e)y3Tr+)ZS=FG-`Jn>0+X
zOX9YpBph@lk(lt4-2Q$e>l+o+()N~SUHw2iyFSskF(N}P_D_!MB_Su>MZ+xq(h}R>
zq_wM^it>I^<bXCR)oQ054efN(`!_Z86uabd3G`2r#MgK!9EcIw>D|)kCM|<m<}xt6
zCNvy5S=fkOv!b;eqE+P)vRED}7vz!DERUa)h1Yp^cO+-a<7swx)CY7Iem0R!$dwm-
z8+lw6@0s^T?EU7-iyg8o?miYBR%$AE-mZ)m!6ei<q>NK{Wnr!_fv&t3ih9yQbI<*y
zd%hBam-LUyZ2ysJ(E}=Sccxh{Txn!KKRP}rh$gH#M|!%U)S(zov;T{v85xl@xo0%>
znixawS7RvAD~5_r#8Te1IJ!3Z5*@Nlpzy27WEy*&t|#53@0#hfP(PDeo9@w<Uiq}v
zyokb{Jfrvd<y4zkN4_16lzR9N^*kwoX|__(=`Dk}43W2BB^=Zc$=}|pm{FmI2PgU?
z#daW`-WwwHP7UNIj>HkovDou(0_I(r0-L?lu|jJWBIOtXpLDRjN)OiG7NKa!GT59p
z!jvAXu|o8)xp!<99_<|nI<gl7B@d&1n&{j6cUEK$9pF381?xw6An>0z6mtAwawHfP
z1H;g9Hxie3#Nv3*cwEc5Eb=5tf){oja@wi5J^eO5EY8BR<2g8)^FU-y3b8Kv2^Ow=
z0rUP9u>V(uKQc9fBk&1BPJY9Y+9r(L-;O=Xoxm<hel0r1{aB8_PgdZ{kxE=MNSV$0
zsPc^-eMK*%I**VU#I4<jvTc729y@&`uiiR_4Wh?$-}XstWHyz@v}p0naBV)geKvP2
zp2vfiEnu~MOIYf>0oVR8=D1~R*f3>1n+)E<5)Rw>#(Oi~IdMNf-f@Tz+8yNxyA!-{
zn-%ZsWy1}xY&lx!I7hS`c~`CzyUlW8#b8$!<&qq$;K6@~c=DnFUYz&Wi+|*Mv&%^z
zwo~(EiEv*w?CHmq2mDwg(~s|V`0>*5{=D7LpJR86ZmSdiTxjFZ33mP*b=IHT#q+j<
z{`_>OKVRD5&tWV4`HsFn_gU=E*TfOJ+@Gu0`*WnZKexE~bI<GkJgCN>=L`*C%RK??
znG?W@V*+_pL?9Rb7sQut1hLMFU~c^v%<Adq_@`9}=Pwgop`*fBUm=`NwS{x6;4rJ!
zMe_L1QM^IytVaKf;jt>`S$S+6M=rR)@w?(#&f_9qPQAp(YcKOAwFF+EpUBFNLdy{y
z=?UGgvC?AE?;tW(?v*M0ePk+6I(kcJM`^6z=QgYC%wYS}JN#EVi=VE@=J}C#IjZg+
z4;r7#O?&g$_`-cYQ2Brl%NDTv_=o&uk<fG27xLuoMQpeAG5;_w=G!x#@Z_FP`9k?q
zw(@(%L+3naxmV9cUqlJteP6=&mcHN}=U?!y`WNgac3LTdA(+1MC998pOLN;^lZjLn
zNn3xS(ft+hQBn9Dm&w3ioX;*gNTElC1a_bLN1H!%ko@s>Qr+;A%pIC&#GmgJAO4Ll
znAB7GgwNz8yf=O|wUnA&O`l`lk*m)ea{2a}4oAPDHO&?DY4A&$zqFX5y&uwRt^1<>
z;Ryw*3xBfMaj7nEq8{&?NkPmv`C2VBviK)$Th&TM`NGTmxs@h$w9;7Dr-aB-I`OB9
zo}PP0s&ds7Fsz0iY^Wv6f)BKNT^)^a{!A4szS5SkA7mBVN<(!!sdI=Fc6-ZU-gn_k
z^y`8BQG&DD-Uo-O`=jj2AQ(FRhfb#vSaM?we#uP4zEhfr={5shz65(!9n@LsVdv#V
zSaDt-+qM`A7Rd_y9=8T%`X<me*obpN#~IN_=r}bygh#~;yk;+=fA7a7&x6oXwZJp?
zqgeL)7*1_EiJ!MmLq^#e%f!A)V~s6hPud|r)E>KUJ77tvBUHbOo~3RsxZB4SH--rx
z?pSxw1P`Q)@r26|FAP)iMoo)1CX@(o%SB&QTlhhFx<6Lb`$Hu#KxEGYkuUN~$xDOK
zkrO2Pse_>)^4fh$M7Bv}urgT8E$h$0OLXOV>^_Hvwde3a?6^q&9Bdy2qvmult_~I5
z-l;)ihY^J0G~ufe{$>m30O5V{$LBHr__)Up3u1h6^QjNswR*#%k2eNP@WR>Io>;WR
z1FA;ug3IBCfyS;lvCIW(3!JeJPN<vghz}YLNb4i^b237=X|)l(j%RV;@)^iGpGL{{
zlj8b(9KHiAG4!VedObdb-!bMewiFqu1$(hlWjDG6k9f?v?T9xrMYY5hthu}qX8P+P
zC0J#yHmf1RP&gwxTuoY+LtV!Jncn)MvtfzI0xv>{;zDQ*oR8wsy2zX{7xe~n@aQPR
zGnL>yL>mUVGtnZL#443q824kk@TiFHG9^vSvYd#MpT;0uL$F8cgx5jjzU_Jsz?nY%
zF~O`a%DSoHewhk(-Bbp>5#E=F3Ro~pTu)785ZzxeQ%3zI$;GWy9o9(Q!y0J6#5a2M
z=_}oz(m)<H!Y4nok^U(*(<kj#%0AISMamM$aFN2mfpTbBC65<PzsdDkJ4K(Xrm~>-
zv@7sERi`%4+=D9klcs`xtOD7?y)bcPH#jx7(fotL^YT?_mm7t5Wn*_-yr+O+Z^h20
zx;v(g5L{lt9G0D~h(jKVs5a<<D|J24wM+@8?<yfs?6TVX_rUR51zi0oeEweYXv~!r
zJuA|1l@)&C1W6d2k`!}k8$F-%lV1KUqy90Oq#}2N4xPS46W3m(jdX!jqR-Q;5$8!x
z{Q2beE%K@o_Zs(Fs9owOJ#%cO`892{MdXUsnf#&)zq-gTSpqZGN{avQQc!U1hO&@u
z*k31&9ouB^#z6*O6J+qoOBQ|sZ4~$G7d0O1qEl<-;eKB1s><b&9V)!rG4e3$Ci3dV
zaxhJi7x`#;j1pd*m(}vf&lLV9b$Qff$YQ6hG`xzWa9TqO3m(Y9EkF*-`$?fH@h9!c
z`A(i0kI6?moKALgqnR--G&94Il2<!Y;Y3IJY3xL^4!P2lTo2MR_Mw{L0rXe68e9&B
z)62F9+N>Kz@5e^dF^^~pxgAYw#GIp<bDm}w$J6k?m+8oeB$`6kXr4t1N%Tw;=fFEO
zF#9h3&bUtxe>|f1Do-i+VkKR9R7?7CO%&PIOs;A|OG=f*1ruo;eJzg(nTnVz_MhVu
zRWMla2Q&`!gRAENY=1QvGjxVwcl-#j${0+H8;=b$C&TdFRD^}h5L`h*%)B`mHA@%z
z1`AQ(yc8KfmZQdNCB7aKKK!^%FxK9V*Lr)<qx2vywj2}x{?-^s_82(T84i8jF{H)|
zU(N{)d2SH8-h{yTSOm0siXGLJI6O4G1V^bvIObnPOJoX+ywczobO#2v@1nRp57SmY
z#EHj`aeT*f!DKFjLCY)rsegwrQXd6ZwjKj~f52I~70dlPU^h^L_1mR*#VZ*;l`GHf
zNs8<h){}3$sqo#?efY*f(Yv{O0QcQJm~R~ZkEgo~=cMFO+~@OH)|fewXGDrD)zIl2
za%Uzh9|d1tti%4Z^tgxaB3`jgpQ8c|naWpk{IGRgZN8CjrEleazjyG~aeLTn(E*W7
zI?S?KmK^r)1n-HrVg)rDP7}HD8Jq3dyVilLS2*#`1ZO_;&4p9d-1y}vcdi`j!4i_5
ze6P@xAD{7JIicy;pZDgI{e4*Hv=1*R_F;KtUshh=%U=%ovbu*aYsC2S$z)%4O7-RQ
zX}&z*mM@RI;ma+FzI@@LFRwW7%iYC$UWg+t-j_44`Eva|@p^?XEBx~1$-~6#w91cr
z3og%tcY?7z%b#_l{5eN8fQJSJ@PJXm3zQeg{SE|iqed`~{vh^Y*M#@kF+^n2Lpe!g
zu(UXwe@>0y6;mQPe_9k*a5Rt9kKv7*W0?g5InXbT=cQcWVHNQ_Nb(Y^PZFKxo3C)R
ze*(WPN@QEPWL~!LDz`Xa<6)1*oYOmnt=EgL&hx4KsNoj>nwicSwzt{5FoX9B_VK^f
zS$sG+o25(da;s7fTg}VmXXbf4GVnepr##@ONBKOttbl_m9<u7QNBsFtA)AI3@sk6O
z*<)%kOMEHj5T7UfV$4&Xas4T;8T*W1`aI*!pPuoJ3C}rrtMKmFJ!eOMq4(^sqNBr0
zX=QK;MW#KW&qoVM=Xn7gT~$EsTOUy6L@B(zEAA_G|Iy*F4jMCFc<YqK?4j92->-is
zzm?x8sCzwi7vADH|BvKtTuVxd)pX+VTUvdoiZ&I$qTzjnzpk*1rdpLyQQc!|UiOeo
zv+~GT?Jm8#l|ffu+@iBPQ%ExU8cF0PQ<`itiN0$3ZhMUao~2N&PdZIsb(co0e?*&J
zme4q{pB`ILMN6N(rJ>V=uP45m28f@VG4msRiTOlE>Gl6__qFHRPs*_VL(2zAqWG9J
z!ph_@=YS&48TG>EZK_b7*AGtJ2MXWj5Da^+AviLlU@>$&#vGZ9%JONb(3u5~Fvhai
za}m@uA8MtGk!iaOnZFE$=CTq#o7dp#W)pl^v=K=|x8T8NQ{;<`rJj))o;K`(yVZUy
z=`_dY!-A3X>IlARAHzZ46X;)a3b_-`;P2tH=vHZi18ue#sA7-OF%B@8?T8V}ozN({
z?X<VMz;2f-7VdDv?+t>NV<0q<8J>_)_d;f~7f$7QBi_vimlyfs`)^;=Mf*XT{gGGb
zkBiHMPbVh;Dq{li)Fu!^g!Xb<HVC_>1>vtz5KOlO!Fh8KjxG-ZPZq2iN%4AKAoiaK
z6gikcNTmhf=ez(k-}Q&Jrawjmi*s6&FZ#~&MYxj>l5)NAqty%Dhj}4pz9%-BdZ7Ba
zJBnP~kRITQ+2Jnu9OaC8(N1^};V90D4p8;8N35$IYR}q2-O>h4CDt&^u!3FmDO_+o
zfozfe&Rlj>WN$3+u;(GHXgGk3NBdxXeh<1%3g6>0!5&iIF7nh{u_AV}$V6?#>R#({
zENd-B>|Twxy;owuBO~;T5j|3Z(>l0jDc%Z}b%fxW$qRn#hzLCdU7Lr#IXZB8F&pO^
zfP%gR)%COB`EUl5#9q?FZ#v#6Ovh;ZX^7Z475Dp1!O!iYf8J^|s<w+P_@BX;8aDvF
z2lR)5d|yOpitN$G-mn*07g=k;(lt|nV}Qu`jFv^C@YpO~EP<lf-*omu3&|h*K~sXh
z(VnV$lHFEM>(#zei|#jaKioh|bbio7$7ULG>=%8A{73DQQs}u#T+da0QOLtu`qW-S
zFMqu!t1s`UR_;Ch`thFb4zHol(jUmew~pS*)YI;TKWI+NPpV)1kCHv5AS2gBrlTa_
zZ7zWegJhAuP7dSMx?|R}?#R?uzz(75e4H+i@FCrCs=hm{)8vHrP9E_C6mV<50+tA$
z%g;x0@bVKms7e_~OGshPB}qse7VH=8HoDTQiKLaw$Rz3xW$d|5X9F&g?(&Od>2ZM!
zE}o|@+4FSjLM)x-3$$>_C0f;5PXB&2Q{TToX<tek&DL!vv!(6yrGFa@JR^ax@{-U|
z6<H;hZYVm}4L{yWgT##Zb)$^<J#3-Vvs&o9UK=Sr`9-m&f9cBpF3J+~-!VIRe6^Iv
z8<DN5elG`)bio4@?7TNh@{qhKhe=^__@gF=fx^QRFjE%xa?&^wCnY*yrEt;aBL(Qa
zC9nP!H0VMJ)o(5)?Y%K%w#=Ox+gxd5UuW_SbEIFD4s_Vnf&AAv($0I%bm_Aj$=>#&
zl<$6IJ0qAH)WgVPRs^ZJMbf5uQKWn<ieBG|BJ1R6S|1flS-URKyc3rwH6npRijv82
z_zj8{JE~<3x5@ZeHuc<)M=Op#qz4CzY5ukns*$cHSN|qzxZ6zmul|bc_#aAqB8l!I
zx1KXY0izZwiGDt1;rmsE^~%1G_f^OGAA^u=^dIV?M0eQdQFyg<90uK+hzmxV_}VcY
zu9>sY9>hYwoC}GI^D*S}VjMFuz&q&`7}&N3l_b3RHKs7>FvDF>b8HH;#DrerJhjCR
zF|(cEpyY;F(I=jK#20qG0x=>+WH~2<qv2K*vgZqy`s;YnL30J>TB5H(;s!?7rb6r8
zZFFnPLd=+4G#}4L)`vpTclZ<w)nB5ix&kjBy}{#WHCWpA3EJ8Xs19z%%U-|G5z&ba
zQIb61xX6mHlH+Sy3j9M|iBlw$d0u^Q_AXcBnT7pDzIq^=KODl-D~Ivk-y`_W_%R%2
zF`maho5bo1r*c%O7Oy(1&BOF&^XUKP@ufZsSYpT$j+$k_!?qd=&0#h7$}wU7v57bL
z*~U}H@8XV0d)aHCId5$^%%(Ax{B`<Cj<{gOMoni$ucs}4mA7YK(ZxCIz9Z{PIJ4<U
z7hW*Um4A(O<CO01{Hes9wVed_c!Ve4P4?u0qrAAn-HVN?g|;)=o5z}Z^Kmb4HoWf5
z8b#i`{*5=U_~Okbjoy5t#hZP9dh=j$91)M>#gW_U%|#vF{6P48uB-U4-$)-mG}njM
zZ1-U!Zy(<Nz=xe=ec5-VFHcGmvyiGE-*)!nDtUia2=?a>ngRTyB7h^@19`=gATeVE
zvw+?6pzL$JEhdC_dxWyTwaA+v59guBBG~;@Bu{mS;s<`wEEyle*Rw>|L*;pX{yUDl
z2FG*1g%`Qw@Flj5xy)x@Ug2zoMDEfPU7cr>d3p9#j{kR!{pQ}_F#8nNDY(f$6>f=M
z@id<9lg?JpZnJUEJKVK6lb@c=VyPS1T>kbhyUFD6nvuDDbaoyOHMlRjLmzN&qkMid
zuYl(WkB+0%BUTWz&P1O=UZ-0m7|TT*cj_@e?Rw1pEsJ^ht749w^n{Hqp77teCww~p
z3EwPx!u@JrQE1<1v{v~!RXi%DkwXe8&#{0uZG1p3ws}<3mPZ#J<&urtS5lwUK@Od5
zRF>949}634^7ikPB>Ro_-ufcW(VytQVIRpkr-r_*eox1n-%y|HuW59AB`t_9rzJf~
ziHAI=X8p&cR9rxd#P9RU57{*Qd<IR=xkU+vDKyOP8XXKzrtwutv|@KMtzUJG-o>O)
z%kp%(m6J^q0t(1-@G}ZORz__bUeSrvD$3J&OJ#NMNcUManRbeMo`^d7UHye5M|`JB
z`xfeU|2Mr;m%tOjM;&%k7XM}`2<EF2f+|$-Ax;gUJJs=W%3zrH8HTHBBN07qETVS{
z4JSbp@{L+(8bi3acn%!&bg@xoAvPv25%<*wDBEC!F~?WpwCFiMwcZ5(CT>Jd(`Mwy
zn_{HF4wQV@g@NXK@bHW9i<p>We%2w>^*aLXy_PUcJ&xAjCvjtz6$%bpBV>gQw(PML
zTv9t|d)gy8!~uT=3+CnpCv1vy#>ogV=lHmy(aH^4+uV_(;{ho(PwcAp#QHdqDc|Ic
z8;U;YO!mR0g`%^v#24cPL#Kb9@B)tZ$4YB|v2*mtY{>w$Ob~sZO9C)(T>$p14Zu6W
zB=sH}fXKh%>q2)<I_?kt&mRl#_+i=-KO8Icg_n*mBrf}4y_^rGZt%u`7rc=B(G%%|
zJrN^jt|7<V@hsR4t5RGs|B(x{UO6MZ&IxOOI3n<;1M1uCkuQ8UK0j>{-Dm@+va_&!
zbOy&VPNVd);695^_G1o~cz^H+I!zAan(jdqj6HyVJ@<)jgWb?C*@Z1vcfj3c8<aO~
z#U#NaHfh}e+jP+lc5E%CO<9di|LbEtZ-i5Om!s5R88X%`#ZvplFwR&A<!|$$sH%sN
zQ|7^Wq3GjTKU?@<fP-$@c%L#8XWF!Ia_4k}XiNi}PDKx~lk%H36+_)Lal~x0;Kz@L
zzxpV_wHN&c!b{xWL$DGQ`=X;!Tw6L+U?jS}M`kI4b-JT;p4jc47G9B2(&%d?iB;;I
zRMpT%%i@JkMddq9T>q7No&G{j(VuCp@T~oe|3a%g>#2FdH#)ETL(B<nG{~isi~}Wc
zWJ4?c_vSraonA|cXKHBaqW7fP`IgMKy`vY_?<wJDHNCE{rM5$L6qNdfhDm&<rj5<C
zYyBTme)gA44*esQv@Y7XQ3C0sWe_TMw9z8FwY6Rzo?`E`L1;T~MHX65Fd{l<$sx~F
z4wwGOiJ4j+<ze!WZI#2j;c^(hUK-Dy$RPZ)Bt{+-KADFd)GxV}_<lKAG-pwCW(Fxa
zUZZ^lm&sN3BHhT0BX{@nG_WC-mamDWkp^*OWqOIME6V81^(Ns}X`#*bKS^y+8-+NC
zV{0pIk!mC3->ua6xP{&fZ=~fKZ>eKO6}ehpC(|M^gY^GFMHj?QPw*j@Kl(*Wv;NYF
z%70`lCyxulk27SAJnHl1&~{r60a0?e>@J6O-$l;bN)89L<S<rvcTQQ#B4xbjd+_ZB
zjhZIf`|1;w=v9;MqAD7-zntO^lu+g8Fw%YMMp570M2^Xol)9V<yB$eQ-+^W)+mo}L
z16loWB;D^Wv_r|0-lqD}BAp;{j~Ggs5#iMOJ%TKcMAEU1k)#_QNjFbNk>1J}+S($t
ztfqKsRJ%fxS0|Bg-ZdI&bCb43q|@U;S@f_mhxYu*CmZF*RHOBr%$>_=YM(bW=IlpG
znAk)i*}tjox!9L2l!0BEJpNWH!tYB@DEw1F#AG#$_Ue!FZi67RYbfNyG~g+6e)GSN
zft}t2^vRlx^0m`osXP;vb%b+o=0N4Q&~S8w$6hes0}dDo9_wm+$X<_SmRq3{vJ2lQ
z9l)B&N3qJ`G#VXj(X8)?t@5sjxaom0#y&Xn-XDi|2IF9DC|Wm1!sJa1=I*!v#ow1;
z6rBjSP1i7L+D+WmNXLPRndn@34{E{pF~9vGWGstuN4W$sg=NTzeT_Hg-eXnbNA#+y
z$L>*$7~<NBosxeby-<Qndv{~6k1|}9DbE$biabVSsd}zc;k|SE@a1v+_{NX{oHJ-J
z>y7@8Q|1om<NHSOy~|_SvU38f?ViH%_0u@qaV9^SC4N1->G0B8T{bOS$YU#)viqmy
zT=I7X*Qu`MGov@~wy9hA^5pIOa)23s`mvYyCYZCyQVUitwB+eyPI9c+Rc$+dme)+R
z<s$)hysp}wPxp7^liE%^e7-ZE6g#VwzOJnO&Xq6wx$$F7cMi;QXZ4vL>=xz0Q+|2y
z4qZ>-A@SsH*FAZ5g(nBMdGdZGFYYzOi|>yWho%?Lp6SKL@Zx?tUL2_>vgk{^SbC)w
z*KF}(hr?c6=OVPBi(dTcv3UN+i=R*P=2Vg8YQOKz(}()-8%N;-`t8GmPx`XAj2};q
z^b<Xr{;b*T&-RI8);aV)-%b#RsReWR&tRS_a$8LgLwHP9D4V5)u})ez8)u5XiTp@z
zcq#bHpQ72QD~7WMoaf;b$D1}?;KdH{d^`Cf+f`lSJxW)2(!2zAI-bY_uO+cZeKH>!
zb&XeSzs@(JZ?MAK6fPK$%0_E$v5Zd|&nZafWq)q->G5}1ZEYrRI+MkL5!u{%>n@i+
zyvN&0M5eqXmqQ-qacSy(t_^y?>ihHg<@5qR(OAI8A|LXCIgfac=<z&gT*xhXg&a1j
zhzFc3;z@UkIR0A^uUC4^nu8y+b9Fx|DJ`UNXNo9#Xfe%odqkh43h3<M2V`xSM<W7q
z=*`6}vi47-fU4{CEaVnVh%6_i(~V@W{G9^QzEbAaFXYzu6M0|%K$6pIXnE8-YEgMZ
z{>iWCzEuTTE-WLvfEVQQ{VBzT6;Y~w0ZsaoOOG#Q(>JRODiWXhc!1DrXI-PDImy(^
zB8h&tC(*LntK_ViLdxlB<a#xWjNd&V|FU8_+wCRI9#=s<^Ip+^^Q-8W`dgZ$_nw9%
z)zJHsA1V6PXPVfvf!>X4rsGn-$Zy;~3ZEwhUl$q7{w<HS=XxMWusnJx_JQ}^ez33}
z2&olA5j<x&w(5<>!42aPZ9N%BW2a&4otaR{0{TYkplgjD@L?f7E?9~gXANK(XoOUu
z?Rp<tgQX&4>DIXczKNSry37>8Z?|K?wq1y?*o`N;`_LJ40LR)7V*7jx+;u&Q#Sf3+
zjKoQVO+Ae(o6q35!&$t1Zi6b}we41Chc933G3tv07S}qWwA=~H^PQoY?1DBgSE%e3
zdd@6&43!jpop;4?)Dr<iy|C=A7a~@AL#fIeW7qk>slW$$V}0Rb?F%>I|8@E2i{)ee
zkS;P1i7Wl^pEw507c9Y1e(31%#nc>M_@DH}Zjnn@$n`<h3LhAhd!tOSOMhp2VYc`g
zQ;&OM-vbZWD0rZ|$c&yp?IwKFt~mVE1r<M>aZSM)ZvQ!<S<_K;4?E!a5_>os*+F5o
zElk(iApfd0(vz&v@A4@ujXi;wkYiZoeH4!N7MOSJ5bVv&F?H>JXo!7OKh52^H(;0G
zQ}4i;kEU>Zum#g%HsR0l4RBpzg5PRu(ev#p=$~JKvAc~hLUTEe%8K6imL<sVT!c*;
z3-Q%>J}#WqMf(N8wGq7`@^2W|1h+u0yEc+W%oKTUEyTD@6S*Bt%)d4T1C*zruh|q#
zlF~%NuPLHSYcg&tkB7^h5rP>z1YeJ-!_c5FDnF~@PKgSB7xuzmE0IrHsDJ@NJJ4Sz
zi|MDO5inf}Qx|m6(3`($U5}qMcli%G;q;A6+P=_Taa}!T@s<vcc|&)UswheAHN|gy
zMJnsQlBMi-8j{;VmA5OYX~}Eq;wmyPcuhV<uW7^4D&j|PXi}J9<NkX`p^4R$lUhr8
z6YA*dq%Sn|_BZl=_Jgtq{-GCJ{?deJos>Pdi;A=*uwbq<RJ&wwYqK0A`^kfMh>X@2
zc?^6igC&NtVmB)^nxCR?N$53$#U7_neEod742GQ(d2gX}R2}J{=tT{bAm#z>eZ{2N
z^pFPD<dH08QD)(7dXsjQc6_=*J2WqnPFS4a*PW-C>9KUXA%=pBVyVyPOq$^NjiN6$
zQj_r6wOjwB*#)iiRJM(d4{4<4lE)NjQ%G+50i@jQM>a+N<Ty2mxFdw}-(=7Xo$s{q
zTNAY>v{I*32mLz#my%EaqoRAl<MdY!5wFBNb4C2!gGI*5Q4W?T<lwYJ4l=Xl1d~w?
z-hzSbW-53?sokI|cHl`_4K&B^1I@bpj{YvMq5&y;srB<AQctv^VIN#+=v_CO|Hzdb
zQk*GovLk&xZ%?1%?8tSu9i@G>C-)2|I<?51oa}vQ<E}u8vJ9ajgp<^%2pW(dK^c=H
zsBm)x<@bvu?e|gCXdX+?wp}1U*Gp7inn1a$u2T0wDHKF$^y|hQdUg69t+jnXyORrv
z20f#YMP=j?R7K+-f1ucs@8oc>o%)Hr<D<fEcs*YZ{SGUjpNo>{?o&ph;1Ps4^@Hx9
zfjH+N`e#CiVS?DVA9Np$y1U~rW%wivzN(4r&007&MH?EUW}}<lJXClrfQjN#*gsqj
z(`PH8JjDc_Et|1c<R#AC*$0^v3rubloQumg*tyXGJ7irD9_x<Ln%+1n{C{fWgAfrK
z0t?v)ku!^ibG=}R9=wPy*#vw~O@^lR4g6<%3(i|I@a1GS60hZAtaJeqjuzpd)H8S|
zy+pfZC01{Ei$5D{QF8J#mRxIq{=a4zoBqPtvQ8XMlH^yG(!54bj)$o!aASQBmVVHS
zN5}Q%SXVV3eY`)Xnh)et7DM=|!!SO1aRfiB8qM2Bjb|IjNi6e6lNUK_ao1RF)~;b*
znmm{D+~>2^@x^@pz%qWd$B50$R<W7sI+j?ukppLL<vN8OJnWGfe>UI83xAt)#7+wi
zy>7`GuTSuj_osQyU28VIW5ZEWcKl(wJ-;+|U_Fr`pQ`V~rK6qM_lGn8j&<Qjx~}}`
zsVkf4yRptKH!e|e=e1kh`M+p)b|`h{*AgClYPbg{&+*`_l^$HX-Gf&g^599QJUGzK
zgM-~XSjXRky(2u>;fe?MyCa@I^Wf%s4<6aWlb=ra<aawgnWH^9<AW#nneN4JyuA2q
zs~2nT_U6tyZ_YdH!?WdmxyKb>zO%-UA1e8C{1bo9I~TyN`vQ6U!XUmoCYV)wiwwHx
znuux-;nwC*wrL6D#$Vwq*)39J!=rfIh-m&yG3>EAmK%?p=X%i>YIx@YH`K<nYp+XO
zJ^M0WHWz)JF$p}OERl!wNaiNpt32k^HJ*L_I>&vw!3BeEa(AOt&bPV6tCQ1MUUXSa
z=)BE}gYWRS8JXN~X%;_Qoz0s~?y|zFd;H&m9R4mmI)#FT+_NH&2Zi6~%|h2%-1&gl
zgyeIt(FH6MU%;hm55?U3kW=y>@=Uo$Y(G=<m#=-qpLRXsBc`gfN++B=L++Es*85}^
zmQUZz@~L!T9=);8AqlH=vR6!_>SZ_S{g10;Vv|VTe-h|Y^-U_SZlLI$U#Xw`7t+nC
zqjL*CkU@gr-F$jWgU43U>zGP1TUkyj(xvpspoF;k361+)NS6!qNh&;-p6|@2%*7dW
zZObh>{O|_p<y<A5i%EonL^9AzqOOTo>9f-fihFa5v_@vqyIJ>X<<}yib3dmly)r83
zUP+x(Uy<?bDms1TEsbAVO;#j)(pNrFpQ5jHIH{51rnU*~`7b&ACy55ZhF^P44l|}H
z;{MZ~Xgu5-XGOoL(nEFRxD7_xwqe+7I1=j1#^Q#t=o4Kd&IzJVv2dm~)=JEV)c>xL
z8uPL1#3GRkUW&!H4X`E22=m=mB5maw%<N%;38@>PZ@3wXueaj&y6vcZBz(t{cEj6c
zub6fAqig6voY`_%=p9FJ=#eEtejdjP;XOV&(+XFXTVwWQ8;qW2i;Ssu!YgPGz2Ocp
zQFX+oPSIoa-U$=3oYCUzg0Ri5h}3X{<r_DIdbz`0%L4_^Jy5#M6LUU$BFNMWmQTFI
zuE!gjYz1F0-y5#N8)!Dh2lM9pKtjweq2fMFP0t4oqkJIySG-5;yLwrBLw&qAbc(!C
zz1a(U>O7$@K2!RI2l{FM|BmmYpWN_P<VD{dc7<Ys3p%QtF-Kl_Y$iHE*}xIE%^dLc
zj6KHq*r7hsR&@2*;Go%A^tC*LTI<u|cKD>=`HS4Qn<cVbk6^U(VZ63GD6%pKu+~!O
zUHkXo-BvTitlWw2^S2{Y(-bQPZ4p_%O;}X79)<S>k2rD-79CxM^$S-B-mej~KN;Xz
zraqb?mf(cnBHWE!fNgj6(Eeo}ZuXgrJv0aJOc(=?5_H^WVSKE}MrBUN!S_?~rmrS0
zTTDj(p_8EJF%j_&li<)e8BNP4<J+7;X!<k&Jsu5(nZ+OsN$!i6UsbUzT?L-kdtvBB
z!Ta8%2y@~6&|e{o`S#Ly=`96)V+owO_=jv)w^6^YMtb7@jc#_-(}qo-X_oO@dbOvD
z4Bx$?GkYpYO|60=FPG8m0i~qf@{+C$uAm#<m6WaVio*4UFW>4lRXM+=g`(HULTC*y
zGv3gZL+?mKWE9Ky)zY)#k95`TGo7>hO3oDxB$wPlH$MHLs|!0xy6-Q#C^VetHG=mY
zA%(({ZV(VGOl|9iDUU=Z$58Y^mC9hyL|JSOkVRdJET##c%h&358e`H%x3hmzURpCH
z7L}7ibt#?7E1{LTg>=OK9_^z$w9)G}xtQFflWJEf{>l|v+!;@|ub-#-QL%JKCx*Op
zqv*w-XtK)^K9d{YXzGYYiV19@o#xHdeZ)^vo>EN4Ltaq2UIwi;3nv99Z(1wkMP1kZ
z==^(slAI&vhV@}IVqF7W+|nej{UU=UA$)+|f2o`4Ke}{Au!diVneDzDdbr79%TYOO
z*dyi}G28SP%&eiZaw4-RhhX6=-Yn+0eV0Y{e0MW-Y;2(6cdN-F>>XVy-$}D;_E34%
zL8{zviu$ZxNlN{UskqUQ)~gs%zvbI$#B>|_<YY@157^M`BwNZG<v_7lTq&xL2T8d5
zQZJ<-`Y0JrX44|5_tFUZQyNBA3gIMkCY<yZMbew{Xfi1}Pg~lA569pN^{PmsQ5Ub1
zOMWUPX=l)yKiM=_;yzuP_lTZUKA}kpf`gO#n#yn1QbN%;Dwy3$zs7aa`Gcab>%G_&
z_7J}B;XSZ)jbL6p7QDjKeQ|#F0F<Q+Mn~*_m|-v+n?!e&fzYG<Z6_eMYcf3Ir@`mQ
zOiVKqRU!^L*zrsc(JK}MY6gfLu>#|~)<SR7CP)X1ev=t{VLbLQBvzk*Oz2tMFt<m4
zIcJc+8?02laNNZgOPd2keq4C@lEQGidlXd6W3jXPf?zXT#+mL(7<vC17KGk}fk!&b
zLo-oy{~nwaAK<R}Be;Jr7MzL_T-Gbc?*3J<kgrBc?>Z!FeZ`4`jVQ=!MeiYh@J?tr
znljy3@{tT*@|EX$!NZZ&?8(>VRk+V5;WaMq%WJaLxhZ21+dUY{E$=i~rpG9ru0NJ9
zL{8wIa#Q${!!#byZzk`}B5t#s!~5sx@^jUN?AN@6qst69_>M7u3t!Cxj+t<=?k4{C
zZ!61Q+`%z3ck{QTeY{5QAp6g=VBhtYeAMU!dk;I!dmo?Sy3sbgd9^JsF}GtY3w!P`
zb>LQZ<ic)FES=`WYt}mRxsT3#Z<`BOKXu_@lUzB?+Lb@#x^m+mR~|9mjSGz3c;I0-
zUhn3{>5*={>8cx>Wx4UUM{fM9)J^n=yYY=iH(n*_F4(~C{B5E;-_v*Jii7SvILw_b
zpSW|nf(IuVd2qKF51#+mgI{d(<ocJM9Js=Z@4WV6jYHmitd|cjyY0i1h3_Y9oF8{@
z^5cfPg1;<0I^T{4a)(I}?^zJc1vAcZvdER^3btU%IHBcC3FomC!E+Z!^6K?btaB)u
zUx?n&y7*XLQgEL0>*IL1ity#oMc%dZ5?}GV%pN&cIJPN)tu>N3Ze=q2J6z>^H?MJM
z?R9omPT_jG$pg2g@=?cIydW-(-)5z=$Mf49{x*ZxeYnFtYBE{AJc}c8vw29|UH)|H
z9<N=P!=*C9r<0z`cQ@zp@b)}*w!P11JMVMC-Upl}a#$Ps<@35V`Fz$fpA+Nq`F=`1
z*RSeLbA3apmvSDx?Z~5sE%#~mn){?|nM)e7IrQgbI@vg-(xPdpRQ}`|Nlr{839|&6
zqLM&MMFv1c{yAyB_)PPU){#?nEe#!CO)Dn7rRiF)DZsTt^skhWbLk7(KK?nm{U|0g
zDipb=d<vVBOU;9`>7qgg?eBYwdTh8s{|&xM2VNvnLuCTRFHWRIn#p95be&X7Q_10Q
z2G#D$re)H3)b93>yse6<srPfLtSYA*qgN!`{SC=iyrY7e8VVIY<AcKAV6Xgx>WqHU
zw2t3Y+}uSE)CK>2i!9V1b;l5Wk(sR!Ox6Q^FuG@daV;E(;r2tZe%)}0b{jk$F&<t@
zld<&MRE$fXffIWOE*-NmWA{A#yFMS$m7*h5_&bg~GeE*cBW&2W5|xA3Ane{coL#j6
zBVKL7_!V2RK5HAI2JJ-GAv4(C*@FR{`*3cCx!8dnLZrX&9;Y3}?9yY%`+5Rue@-D!
z>I`(G&f>6@4SrkL!f=Nj<c#c*pyhxEeH>Bt%@L+qPN;Qs#+wB$c>LD|Q?9te!@vzY
z>qP(dQFmmwxeJcIhv4IT;Et9ja-BSJsn8QsB)u?UoEJ_n@Pe#hS>&(q5_@wmNR9Qv
z^)65RdguvdM^ETX_e6P_2eiazWVef~y4XMU|0?F0Rc^R*(-qSdT~WTy1yduPQTEyi
z#=V`Or7M_)`^7z;kKi<46`Vei5A&|J!G|wraYDfw7e`uQ44*>ua-rF5I)=DCqGRER
z1*V@m1cS5Yu(K1)4g0+~Yrh+_ZFh-X-3~+^+lI0OTd{H5W|XhoC|I=XF@3^1obJ5_
zX`=UZ{tIKsUo}Las{wZH)Q921CAg-!2t08C?m<uFcjlpooeuIv_FMHCVA!aQ0>xRP
z|9b}FN%Rk{oQe^Lr$9G+61vq+faJdMsJS@~AEPF~_~Ll{xvvhTbai<8s6$u3KQheJ
z1fNh9<`ybQI@=3he3T&Pt%&Az-I4!67Lk{w(KAi>dQ&7Y_~&0L8SsmAe>amuaRb#V
zf1!0TwbUH>j-0HkC~nUy3iPd@x~wvKt6oZmze{M()2B4v<N>XUe?qrSD=BedCEc4|
zPPW%x(CLtu^!!{ooqF<$YKy9<^Oe|B2@g(-YAtnb{7CARpJ<6$J;`kOMq~edqg%n>
zNuQg9r@x)F?)@e^hrhH_=pCQ3B~Z6a0>_34PEC&R_*6*Y;5r#`-Iu}n?lMrymBi}C
zPTID&ozj9^Y5JEIGQa<d<nES`Vd8W0e^E?-+8&dvLm~C+b%%Be{o&8&6f$#7rU{F#
z&;ieQDpom9{fc8~t!fmF634<LQKZr{k_J?UlG}=Nq_IDQhQ5!c7sC>$`@vgeJ}s17
z*7}jz9A7F~=tcU2JjhzxgMzR7kgr7`E!!GO0n>8HSh9&8jA^G?`W<v<qVV77|Dyzv
zt#1%JDT5nAx6$o}wPj*vvysLqk-Pe`S>(q)%HnOC$d%j4;@f25&y5n@SKWWmbNOnz
zZX<G-rCaH)+)g^WXE!CDG^g)djA+(IV=A9(NDUp!=}@U5)%Y4y_piICxZwn4N?FlY
z3u{_C&6Zq?9LPgxI5S;bX?2_%ZL@S0*>M-r9xi-4(?e-cN+^v94W&TWaEhE1MM-00
zNo~Ugvd_3gE0!nHYmIBPNB1U~Tu&z@^DI(4noHdt6wtjB#iV-s1;zWkqPT4}r2Ms>
zZjTgPu<sp2<0NrHbbX}=t?K=4k)>)>La3YAk?vE&Uy&0(@ox}5rVoXcq6W4LpRs?~
zXnZakhkc7DLF1Dqw61DlYqU06GiM{VyDkzv7h?5FeOxp%LS@BjINjI)?O&#djNXmM
z0SBR&dJGe5&WMgIJ4{e;!XPJCw6%L6YKzEm-tiY5zk(UDDimH35y<>5IH883=Otfg
zIBOG7C4Cj@Pi`PR<re%>GhqKB8^@LMV7k2kGfRt5vF#aTdX!?xvr5z_zr~yQTEW8n
zEIMu)@Vsve(zgA=@a#@l`Ac$+$Wj?9$#KW??)*Ef2e<C)#jU!%d9{WbUsdeS_ap|g
zVYeZyqBcxuFe7;E{?S~RGLDscP2yhGn!Hy+i~kFp#UaET`CW%MCF}9oV~d#d^!cQ+
zAxpek!RNx(@I>SF+}O5>x4WA17MY!5$F-X!ukPcTa&u03f0)k~9A#Da;~X~j6f4JC
zv2%+xU-z=%OWwAeXl2LWR@?KvfexJT#DUiwbYzczjy&Ux6PtW<VnqXIwv2Y>ho78z
z_6Qg5wbq4C+Pd)dI2W$VabdkzF8t}c3)f4y@_H3l)*0r?!J6W+o-3bO>&oBET{+Or
zl{Y7dUh@)HuI+N=by{w`L3Dt|XSnejMR&G0bLX~VcQ)7Y;7+0Uc<Xs`OR3O#EWG%q
zx;Ni1^yUf&A5K~1%g=iIaq>q$E)sLjmFNI2cMjy*qd~mSESURlImf}GQ|r>kQ0^&u
zA^i7+v;0Xh=eR|(QdAVrOB0<|B{AHmDOPkt#qnpY3!J+lo=@3d<m-u-L>KyH?vza6
z@e>opJ~4?0+a+_+m8)$3{2E_wz0Q$+Q#fndP2RLDl}onV;zI}0*y(6G?>>5)cka*N
zk_~rQRyUK^56R+5jamHpRyH>szRP$2yT^}n?{WCr9L|25!-Z>ed3$ay&m5A+|3nt+
z?4>+*Da+&bpLu+<+kFm`RiSIa=jfnO1Wk#{BN^HIG^{j_0%UXPwAnrS;E+!DtW#-+
zRw}&=xki1iCeVMA6X?V5E7TNsh4vd>q3RonG*577PQ};KlC<}fb@UB6X}%(jZRIqo
z^QF+POQ?M0GditSOvOQuNV)U@b$-Yp!xveUoqwBN=cdx3vg<T(X);~NPN1fXS7^Iz
z0#()~(t%soXw>zaq}G~Fu4$QcKr@F<JkO`JzeUu)<SFeKTuM_nRnU*^uPJ=}TM9W=
zP5tXdzl+#W?ONMFJ2wa|{Bb+A26fV|a!D)~OpQ6~<#Fb|BIeBQg|GK|L*2M9EE?5C
z#r$C0*gOo8V@6`|uQ8CxpMWE7Q-rr}ItnUg;^i8k^5z^2Zk~tNy%r!*S+G>vm*U|=
z1I%|4-jAs(@$SiL?A^EyrLWiHxxpq3OWA^pJ+{Gk!wx7%@4~&;yYXMIeaNQ+P~UP8
z##V=M((ec?Vl2@ra^zu`PU3ykX^iwbgF6;yVIFUTH=eejgLde>)E++uJ77({1B#Qy
z-fFKCS_g}rP=V;QGIJ5LlM8B{U12Tf1}CvM3a@n&J0f>n7W$6Lb9d~Q^?>Ij53Cj$
zr>ko`;3p1g@wmTutzO0hm!7%9S-khL_?`zJ+)#MN4LXW$@bq-WSQ%HiTZoya%o*pW
zIAe#M6S|5V(W&H!uZtb9Lg+Pj66}!h(iU;QY%rp~4d$C!LnGJ<x9*9#?ZXND>vkM+
zLoKm?#u3CVJq)#t2cf+00G=P;2V1*6Aa^rJ`R&B>pzZJsF@<vI7K}Q#2|oTCAnjp-
z8MbTj$znB{x2_afUSkwZHAHJau_yey6wx({@vL|uR^`rz@qJx{md?e$mN{rtpAFY}
zz@1&%NOYcwfpJ>M%oI$U5>4oQo{ZyN6A?3Lg6OLsCpg+;aCYZt#B>bDG~u(U$W+JH
zICXfRRYyTtKfKwficgaSC#AI~W(j}i{UeG{vlSd&;ae#jD1+1{!J+&v0ja!BI%D#i
zmWBVM<ftE%<6KWc8$MFr)HkGOTu$YKUy#{jkp&!AOm~L9q<%9>L>B5PwVW;@8lOi8
zf-)#KFqQJh<WtDBhg8s9Lc0`SQifj{MVnXBvhvqdF#j#xaCuKXQfuf_^#?N1_(aXS
zzKFB;SNgT-8=3uUpgd|MZ@U%>nfRN`EJe<Ecqf@&{zs)h|IwoCe^e1Gi6rNCvNdid
zt93u<)ZuzM^!qandiR-*y{e^q+pFnddlgM9t)$~gg3EIBIR#Y|Q@_uTX`5dm4es%X
zs-NAWeF`@zO)7!1N-xrcRTn5kEtV7wqv)egB=ypdq<hmMX-B^ZI&U3HA3Or-V{8zO
zS9Kw?1}6%RbD`_YTxl)b=;IwX(s=7ezE19BR^?90nx1qv#h3d338X&$A(W~aNXza0
zNzTfjl>hrn+k{Wa`e7HnFqOc>S_yQ{kb==r(GyY84T0&>f>$p3JVRyCFIyI_4${~#
zRvL|s4YY4YEouCCOpXm}NLP6SNr_;Q{{3APRI-|$^fDrgSVJ;=y_`O#FDJM4hO|f8
zm>y@YB8x4@XqoOYk~BUkd^xA7Uc#OnwVi3{65-SN?MQP+I?<=SE@EDCrF~<4$i+T@
zJQoULY?7GkEF<XanP}2V6RhLG7m3oZ&@`W9>UZ@9jT?|gZ%gh_dF4HNp_NY?#knnT
zSqTMeS5iy;dvY1{g}R0}QBCeI!q+Z3F{c~;3Lely;h|c0OmG)Pe}>7*K1jIPAJ>l#
z64^J=wR-zMln)#ZpSF>Z)EJA)E)!5abP9UbPQ&u&GqLHL=+2on7jrY`<G9@t{5rQB
z&b?NlXNw6e7jA{a-(C37egIj$Enzgl3eq9Au<YfCkEdL);jKG*YkMQz$q%<hR&)KI
zb9gm592@tGUYOglVpkrIsWF!^er}TZoNLG{y@^dl>8Pv9gsMUgZmf8Kpj(gNs`Ugj
zpO!$wz8qdgRX8!D8XnW?Fn-}z9N*uFJ#nr0`?Uj!|4FdTb19Z^kzv{S@?0dN$RCQ8
zSSLc6RV`Kd$(p|0si)2}<_zM_g+n=Iiw2)@9m$#x#_%H537mgqGHcgO<@0-Hu;pKE
z_6wiQMvLaL`|tUD=jvis*|m&A`Wx}qyp`;?ZLQFC)^qnmo4K)Ebc~PQ$-6f0X2YZV
zSpR@I2Q4_v(SMF`FRx>q_vZw^*E=n;+-G>c=~*5gX2Zc2wro7pj*Gw8@fA0FzBs^v
z&xbm&pQ0mMn>(_t@FZvVbmFbcojB0JiQR8F@%~CD-qPX34t<@ud6F|{2rtivb<TWf
zzca71ab}-DXO>QK=BNT^o?qw8W!+s^bD9fJ+U~-kp)RZ~=AujuSH5%9l~bO%a@Pzu
zF1zT)%Z9u2%FFJYr|rQeFFiQtgeM0Huk*e)UR)UN&DuMBc+NCm-rC)d$JF`p=tm-R
zem#IEMGKaWe-MYe2lE`4b3E2HghzXYvW)1#@;e{SF)0x&Q4q;iZ=$%WEt;!)$MRy$
z^L*bpjwOy>;F#ceo|}1*Ed(1dSM~~DAD_V4hKc-t6rFcikN+FRHSH}UtBmZG*|~3}
zv?LTED-A@2_R`)%Mhj^VX;Y+9iZaW{Zcqr}Yh{c4-oHPd>&mAqSJLNs?)SORd7Ulx
zdHg3bpJh%LaC7Y`e*XM4+kGqKHpw#_*-Okhy^Hym%sDpwd5%q<m#|@VDIbX|<7mtC
zTr{PeEn3TYba(}CopOOsmtEi!s+IisOeMb`RmHK+RYEta;-i17IBe=gR@c4A8S5_c
zHue4@zo3NG>*JB<R1Jqm)%cPn{KURh`1qt!<hV-_{G$-DpM@u`FbDH=axnH~HomlF
zL-}Aff?KjsJx=J-Hn}LV6WYtE794!jjI)Cp@OI)uOt^6m@mcjyR=JJq26Z^`?;5)A
zxQrQ*7co7!0;_#X5oCQ9<x5T@LMso;%(C%yW)?I(v*0;12c_o=U{O$nm<4C?Jg*E(
z?JJS+v=XmnE~8Mh7WO`OVeIe_Q!h23$HQjC{%S$0#S6GEc#Z7=?;vOR39<);e{p&z
zzDEeYdbZGTUdhr7q0-Du?n~nb4W#jhhmhHr;iOnLlAa1K?P$62bgoutyFODzulY<m
zeO2r)R?MfRrHiOaFeH{PQWIUb8gz517KM!0rC7m@b537DqxB5vN6TvJZn>6*wXUaU
zBDXvxNBEArOejRhjEsXU=+F7>r1-{?R`;`_57TUDn#LX)VrWYd#`d(<*pd1gIMX6E
zSJIg1PW}=eq<+tnijR3ycd_F+>m%+l!pm}JxIY;-_>-^T1GX;-q=BykDa1dBn&rgo
zkq}Id@*(sgJcLR=hmimBP?9_rN<SJxX-2;=a-SbYm8*njW^)*Q5Z5E|@fq=2s`y-8
zV<_#26S?#iq4fJ}2*n%{y?P2Eq<JEkW+(=WzSAJG`4vd#*9VeMUI3|e`jeNsKh^k%
z4$)Fya{N!|H6wiJlIWzkQRhKd`s|}<EjN;Ib)mX+CrYhzAjMC1bhp1PO`pC;e9nf9
zH(ODu?M`~`w}akCZll-9=5!<1lmd%Q=vT#N>b$g()?PNI+coQG_C+IFQMQ`oPZ<bb
z;!3KD)F&BVJ?g$!m)cFVY5y8c+G?Odw#G~7osBBV?q>=+4Z8Mt5vg=7pf_XZ)3n8N
zX_(kMDq9P-tLF?lvws>n#Z95(Cnizff{A1lJ)Zt#E0UYTNQ!?qoIckMqfc9hQHS9E
z29*c~jmS~=3+Y4e1_^(s@aMMd>rRpOGE}86MILh`$Z&NRBxVW!?d@+EF!eK(m;Z-6
z_1A(!`3%aVn;;+f00FbboRA~5%=0&}I7E2eqH2(=Qil}^f}7w_h1we>IB@k0bf;d%
z>}OYTuDl-I7T&`;&j;8v<uL|}z1FdFPvEW83aNF^QDokR-Md~v#iJeb)kTgf<sDMB
z-=jdVrD{t$@JQ<?BKH3lb5JKrCj5b|`Ct5!|BEBmKXGWnH#mvT+w@?;)KPqkJD=NO
z_^2I6Hr~Yr-8;}!xdmPM2EiP^hp1z>acAx=>@lpvD>2{vzFvhVKE>#Lwg_*XvN2sF
z6T=^-1GULm`aBWi{v}|%O9HmaCt%sYWALs!f{?63xY<7pC!Po6xpNRchKE2}=pCVd
zLtzpYhW%AxSdkxw+grlnm9Zax)efNGXe91F5xc5MQE)De#J;nUxLOy9-H3+S&={;)
z*^PP^cB6*_L{DcoDY{rGMeiFVsL!}c418XSd#B2Tm#G{FJ<HLS6NRFRWzg-p8i)6;
z!vv*`u&~^MTVu9k;awvPytxXhJy+u2WqtThTY;7ND^clbfHF;EY`we-tIpa&QP~m3
zo1LNiS}Z&3o$<mV5Sn#?c-uV)ww=Ma@+Jfq3=hCh{xDka#UQKlC`?ZsN7uDPT<evJ
zGcFm}G&Bo;KIURU-_w}scovT*mmy_VC0rwG5VHLSk}lmr@%cxv+|Yv9^{;V#!ADf*
zeMdslUp(rPpc!AJ>Gmy=hkMeCDjNC`?Ho+EiieU`%LtO}Glq<pkEKcOinLc_0(~)?
zL?yz9v%+8|o$seCy36JZ4HV?LS(Q|uYLHQ?4t+hRPiy9^7Q7^5IwJty^2@hV|Ft&s
z#MXf<-Q6kvwD1NF4Im}EU^2N9Mt$WD&^M#Q^w2k&+)|HFN>v=4eU(7YV^ipb=<2=R
zmPsv6Ikb0T0qyK46gqD)$vi6~XNgLx=W4nYd6mXV)sa=i9a5O{fI{9k(8-HWNi+8a
zHRrTb)rI#o?nMVJ9QKpUH~gjPMiQJMA<ci!$nvf|J=uO{AAbL>AAh_o^5RKDxH4cE
zuXmQ`R;ST?C_sUCq>bmH_msG6@KheKV+P;6Hk;?uJiaFK;$FJM_n)fr!+jb&f0Q;`
zlrQ64;}yL1y#YUUHsXrU>-px!O+3E9gysJ;W5fR2Ic4Zh9@EX52i@4sKkRMU|Cc?d
zZgyh7<1QR|){PHj?_+m&PtKa?&0E&_@bVtMJU`2q@2L3khcZ8Qnc>en1wW_mtv{D7
z4B%Dn0i0J9z~9>fcv=5IuAUpn?rQ?sTWCHFAwu^_4&+1U0=e;4ARE69<PM1-!EY9O
zFZCdvXBos_B7?a9<se?s6~sH#f>|;&m<yYN`M-rBESng@3;PM}CnS_d^$p|n<6-=5
zZa7~Oo}be``+1A-?HGTG;5lawu#@*eK5cl2U8Wx9&0@~!d>_fb9z}`yC7Lgn#jtcy
zEPEFm;jV(Cyh8L|QF$CkUp>x~ML%fO>jWPECy@sXPG-&NDQu^c%9`8K*vv1Tx20t8
zi>ecB)tbq^|4wqo=xjDX4u3Vu<-az0{LnX_pG6k1chV^yka?Ovo-AaS^dgoyc80$N
zp5=6_Vs_FNoe#rHxbL$P&Py!i(S~K*^rMV-MV{w13gtW_rJRopso;g~6@00_g1^XL
z;3(nWaq_*uE{PX7Aol{VSLzGBRsGO+R8Mr9DkC&+DRk7yqMMs7+=6YOW_1!n7oWr!
zyDaFNXJh<_Y^3^UBV}+lrv1!<!RaidW{ImK3tLQcacx8%UN{I%#`_*-eZP+(A$Kt&
z_YV4Oy9Kr6TJ&?ihUXHOaYgSU9xbZCEVWYf7;zRzKZQrVa^aSL68~&6akDTJ86&e`
z7o7`sD!|l=BA8t+fvf0m@tSu*JQJ#+S$iG&3-4g(iTl_-?J=;j32l3xLhAZ6XzhCm
znPYD-Va^BqRQ-ZEg+C!F^!YSx!IBr+^(DcEE$r5ZUYPbLN6}Xpv}~xrBaWbX+ecGP
zH$~AUJ%N_<WU-^3MjzMCq9a$8NmFqFS(vKOUw@|X{c3b<p9UGO(V{*>b?M_}J*wWe
zf}VX@MFx(m>6z$tS*yREtTRNnOUEWUG2evD_L$P61ar!)*+vF0cF?rnJIT7IHOUKh
z-PnP9Nv)?HDg3ghm8}BUaKV}MBVFm%CU=sQ_aMbOPg*PT<Q4%wbY0h%Y-IeXvDA-}
zw)>NQ55WOW382*3f#h~JkfKEILu6tQDM|)Y$A)04%m^m^m%+4sxX^c0L+GT~Ev0P_
zp;~cG+7Lq5)I#XSh!9%)I+zkq2GjE`!DJ<N?ZG0y6}u>iepLn1NWmbqE(##iF#+V@
z<4@xo{pkHPKQeOmrN^Z{)bZY%_HOo~F;yPaF+${`M7P4pA{SEq=0v8m9qH<Jdm0&Q
zOW!W;q2@O>q$RRZQL}c@(3O_d&uTkm`&*FgaWk58YAemC-a;SlZK93OMLz1odOG=I
zEzRv(L$wmCX`Q42HUC>dM}9A-KEg*l@vRQ2KhvV+jZ3NIkvc7DR--9z7Ylzd(}QuK
z2fB-ByX^w9j}`l<Qf2D@XtwZ7&ZLRbGbn7(G}4uyLXpEK3H@p!-M%r7&Ppp#n2bDC
z?UEzi-~H)!a!<ib?m>4Od(h$jz3AtIUeqBncq8BSpoJ^CQ?Qrllnj-kW&0#(Lt_`B
zZ~Vru6C&@tx&!YoiSClFcHHgWhTi9&;$Y2V^me}or%|^tac?b}W($vQdktdBF5>X*
z3e>DBg>vi}RGcct)W3yLe^QN^XKv!?!8<4u8qXV}`{>#B5HU|0V66TG5_ejV_4gT$
zw!J_`?n}%SY=rrn-(cH=x2X604;M{7Af)jFB;J3<MbZ7#`QkUeZtujyL4VNi)>n+~
z`5D@}B4c>uBPxVn?X|{pxN0?^%ke%QG}WW)#BKBm7Q8B*TJWE1X!pN@B`2$e>*gZ1
zrBuK?RGfWw6=Ud}Q<z*SI<za&@M}_v$PFY3-6jENwkP0QR6IUbi2U@%qgW%mAFhAG
z(eqX)V)KHL-57}M!2ysHxuo=@0L=6XLaSK_s+&X6y&?=3#_q?Va}l^HaR?tOBB56t
zi7wwrd<~C;d}SmqcSebvX0+Ii$6}rQTWE{t{y#;rA5@j3#RD@j`DP}19a@8o9i!mB
zK?x5P=HOSMDpt>4hKp|u;Cg;7uJ~`jXTQyO@Xi7euh&3yPNJ8oKF;^ihk2?#;-9R*
z=3WLE|7;^FrFS4b#u~SE_Tq`19X`ypL-Hd>xMz7|$VgwDjq=9<DSuoY7lKLm_oLGP
zAR-S%;o{9$Y<h7FYJCzgZ(TAvuckqDS0<V@a-gxL0G5?SP}^GqX~zoGH(bJsGuQA^
z{x;l3KSa)%Conkn5}jM#V@%;!T(IrL@f(8Qm?$L}HnOx?^is*k_N9&fg7H{5fGTYU
z(s1n&q<(BP4XzMpHLr1^XLcgZct4qXxK0<o|Jmd`Vjii_T10)ESuk#vP+_4Ky{Xot
ze{4WAl-JQJ!IaXBu%PexR&?R69eujxN=t+X$5g?ew08zkXhA5w{33D<!nbF*B#H(L
z*4C?4$H-+%JT-YI(Wa_YvKe=RE@xy>m2MuT^c1|m*JtQ!O9_>JDW|837fI6UGF`8|
zE;{UP(XPh3^f&Mk9oB88KMK#NweKrB+xHzk9rKAUt9_%cJ-<bs^&d@lm*lz$GJN$x
zcRmu)i~s2N<qty!@WEGuSiV}0Z|4l>mXwja;lvmoUZ%)JPbTo5fs=Xq+G%WlawZSy
zrOY>6=W|TQBBA>*d;V49vXG_hFj$A@#_6%|h?V>;!jM;eHR7OU##|S+iKEY&@PVsl
zoOWRwe?Mx;O-5GS{K<w(t@pBZjU7+^?Z7+553>EY3+G>U<L<lnap-3c9%$snr7_<8
zev=PNJoVwPhQ5Na<16$eKQ1u$<8CMY*r3gi&kpxzO>KYfzsH}I4*T<gQ~s=X)1TM2
z`*WIP0FN6Tz-5cXJhU!=E1U!PsnCJ~ssq^ja{xzA3}heEK>nH<$OqmB^5Z2ztdS7J
z3R1zGWgpC6-UajDJt2IuGlVZ44(0B1!gzc`7?*{H^A3&uTq!B$9Kq_jeDnY(>^{iu
zx`()H@?pL$c!Y7kB3bTD6#r?C=0Oi)cvO8Xzq%v3BkmvNca6umqAiX+J|E`~-4eKd
zP$IV~CGjo6%~@uY!ryFC*)}MRS0|_Ql(G!=zITEjyvyWF$t<=Wmd%k<ayWERE|-W*
zd7xfCuhuVMU)@t|uX>t;W)$)mxgwG4Dq_ppGu(dkEJtiE=4*22cyP@*e!ruHpZqOh
z`;b!Bk}cyO&SgBOv5d24oo5A?^StHMd7j^No<n||XEm8}UVg0?%x?6;x`W;Exl9^O
zSyBjoFNK?N(olXP4d<QG*sxR<1^Z9p@suna+L48Gqq8AnlZBLpSr{sG7Nvt((0!1F
z*7>=p>X(PI@O+%SU5ZOecQ9VF9+zrv<H?DeSRqx56~C|IeO3(~4ZVo*FUkcbq!bGu
zp25(;r||7?Ha;CZ0SmJXoGlbg^BX4+C^Vz8x%t?lSco&T3UPMqIV7l;L;L4NxTs%)
z$tH2Wl)4A0%!jbJ(ttL>#c><miin&S7!>;&X}{kgq3IJYPWq0~b)9H!?na+JOVQsk
zf(>gTGB0_3X?D+nG{t_1=y@DQ|F(~$%c2W<l;b#R|22_Lx=f+Y*VAdV_H5dCU@k4Y
zuz;T3SD_7enI4@}qmiO>dYFb5ofOa69e#Rrq}K{s9I%R}e=?+rtBmMH$~v0(#+Y&@
zZlamSTWG!SRw_#|quNpnY8Kw(QFkpV_U<m4TWd|_=XX<c+Frq4v7_M|9B9NuC-Qsc
zObPL>l&$SftuOY`ZU;}=_tlGr&-I~(7d|v-zb~y9bB5YQKQdq8Pn$)z)g5i&gAudH
z72(&(wF;zrX97v$OCZTi45Bi<Ad=b^M7M2&XoOu5y%Zm7>Ic#4Ny0<(J&+Vi18J3Q
zAjysjq}X!NcfLA+(x3PXeZ-$49{Q1tmf#U)`;tx%Uy?TQp)aCSLA%|Pep`8vjNtmL
zUFk*-Pq~nFFK2qY$&n_Uu&2jwY^iM0UfN^2Td>ru>2S?1`u@q1Mv82d;gW6i*utDz
z157D2&4gZ8Zl<n=jdbm^G0p3~o+in!CI6{wX!0UM>eN_8U-eedWP|1OWVPVhtkDrH
zR4p32W+^oqsZ+>CHCnr4F@10o-Iy^dlvE;mcUtC?VUKxaJWH8=tQL7FyP2dBK3#C%
z#6Bu(G9_dw(ZypEXs*pT8u)iCJ!}|78=noMybA-Vy1Wlf-zA<Of4fuHE0K+|>Otk%
zJ?OulJ?L`1@N<Z)Zj4~Hbfinr>^1*Ty|NRhw142(<S*FS`T@VUzlC?%OBjia_pY)g
z{1Q3B9ZTx*K`@|>jk=DQ?8~^WaS6TdU4X>vGRzGrhJvJ+YkK8jqvI*D)4GJfEjRJX
z>o(d))??bRdw3A`052{*7Vo2GEDvkJkfLYkw_9YW+S(9y_Z4gd+oAXG4fdwJ!|`{b
zt7P*B40L^m^(HUTQX}$(;`>{5`aAkNd_m`0v6E7IhXpOK@l~-6j*%_+HLU?jdiSC2
zUXLr?Zo{&;4o8x2Kt1sqI>lD?%i(IVC%TBtz2yj8cUI)m^YQHKNhG<XW89Q9Y&1@W
zYhfY=q{m}$SRCxLj^N#}!)P=NgG@p=CWVE=H#HPrErao4Tp*m|{872WAES2#;6ZgD
zhK~<Hz_3u{{|m*o1K~(q9)aF+2Ql>3Vf@sI#Gg@z(dWovL~V`4_q-^)u8kI%_*ks}
zcN9P6ghmv91b5yYLvqM(mC;o)_|bm=6zfM~*kUD|%%2VYg2hO1)J2(%0lq7*g^ctD
zICgA8|H&5kGj0tUKN+GhKo71;%Te%aIadDG$Mf4p@Ydagzu(O;uw@6PrdlCOb~jce
z?uNv18!R1ehhj@dtoynTd8U2>s1$&UJ3}z>SOk{)9LA7m(O5X>D8?^4j(%~8K>t*1
zJDq{8p;<^y%EN<B;WN%C7F~Vk(b?-FmbG8OoV7PG-26WByEP+3_!_4k7ypO%I*_yV
z7tU+`!x%$JT5v;}J`e3qsYN}>c6Wc$a}Z4983W0G=5Xo=A4xthM^kK;0>xe!Pxl8*
zqDQe)$wz-C?VK$%9E}B}5(vusqe^Z?OKD)aE;Y_xNpofxQPcU2<Xde<MLl*=6x&i?
z(E+Fb*ptRi_oIzl18Kyu5E}I`oD{nb&}E58vJpPx(kDk~{+&3oYfK>X?kRL;Lpr73
z&Lq<vIkabF0d@XYNLLz)=|^iB{S+FGk<f5b4_+mk&RQDpeTN3fKOlqJ2D)|lDJ`>k
zK_(m8DSrKX!SCpx5q>{tZ0;Wl{oIW$qonvT$@21dJ=i+6H_Ms#<K@!^^2C3Gx#H<i
zR;V4pDwjub&CRjA_~kgR@2$i}>Qi`r;B<C*IE(Y=&SlBt3wU-f@Nuujy#Ct~*0R>*
zNl$dRSZ6txXRc(W--g_W*K(zmF<<xJ#Qz1D@HiVYKE7}pm%iK~W}{u~&}Pjo%6m9^
zt1S<(vgaYI9XWlNGxs|0!uQ#Y?<BZ$@0NXBC*jFq-MzT|vlka!^5*A@efaP}AKvxI
zhn<G{3SW^g8+-Y(hR}KfMGu5RhcEXV<i{Gb{Wx!hAJ4P)<ML2HR?PNe-1Ot7FMfPy
zj6XkHDYT#ff7ZI_&)=j2*ufxxHPQpPu3I2)+Y!jB4+7b7MG*J69mK60g89z-V2%$9
z;UBX?IpRqucZ(7p=2hYRVeo!Vdb*zz3L-ct=m7sS7YrUv!PS{|m=6q%<SDX(pYtn<
z13pLdz5inP_&dQwet(2(I*xMj?_;bXBea}B@vNwrz^!u<d5C5bo3Br1E9(@l@lIul
z=rqpHOy^Z%&M~h&!RC#bT>a`KfBPu<9=>I>_xBvm?8s%!H+fv%kk5|Q1)}3hWUhQp
zv%=~^ej{JRy&8+S^S~KisB)GyAD`u1>tdGpQp`_<XQ#bM^j&cYs|mJ_^urQ<)wh(-
zFDzxVHKlyTw3K&k?t$RH-QoXI2BSVnL1Cg4{uoPPSGyF97E5EiiZt>{rJ)fg3scjR
zm|c^J147fWG0Xzwv&4NX3%-4_(BYbet8a2dt~(b;cIBh*y%HSkFY@5S??C_0EzFU+
zi5#~Zco%+E+>dHdFTBff&gFRiv;<%C&!8h;WTom(V%pO*@pWnN7W|B_f$4~I&xTWT
zF8(&>pk>7=jQ(*3<@3v6p<0Es3s*3tt`1>`>qR#60o0c~#*;-&cxmwzJ8nFKkI-=D
z<i5c}!Ra}){|lD&|Ak_4#?G^qq>^wM3N7tHRo{D)t9pO(KRSp)f6I}p;NfUL97R&g
z6lg%{c=A@9M82L=>E5*&G_>0s>aIACqNXmSvQZ$TpG?&k)F{JFgJK71QMWW58azRd
zuEy%qr_Pn6XJkm>$!o~!*;+a_%$TZ|Zlr)6n`xVm3Dq4orC!I)NjH8QMaJx)yFokY
z-d-!xHL#)UllD;Kx4rbP)Q(IX94J81iB48Klk5gpYG`#M!;SkWsm6mokMSbMo!&IV
zz=z&F_o0C{zNG%nmlRC>$oPgI#ZU34G~tVxb;X~$O9fEZ+yMGu96*(J0c7G8K-a`2
z;SfM0HU-f1h2pjD0n}RSPwxW6&Pmyy&O8vFpFM)f^T(IM#BOYFr;qSs`_Q^PZ@RVD
zi@L@M-{IALq`lRh_^m6Q*ySR+xSeS4GDix}v8Tf_cBE^(m#ouwQ}$bHN}Xm!uPk@c
zkE9)>b$=Tv%UDqVxn?xdWGjUQZ6UARP1NVk23qoYJq;VMj)JEfk*CN<wQVt=O#78o
z5};4=(R%bTS(gHHv}xNJO)@LjptPDLlyys$W;8Rs`2ZT-Lxp4~FQiDZQ#|fGmx7Y#
zQ2Vu6bnpEPy480&9iB9m(o`nX6?G;0NfT&;!Z@1vRP>Oo98GVA$dkR`BB}WepfJJl
z4eb(L-dnm;@dH`faaWd(=gEqkzsOl%m8QjdV%K?0?4-i}VRWI$N>%^BpDSXvDf|da
zF;i?F^cpSR!XH{Cyzh07QCoNqNo}|AI_3r%ORiw;_-a@uRN}96IX(_4!PXl^A|F+N
z&tcgx*_eSnav2yVSAoIyXJJ})8vVW(U}xU~_<zh7JD_HKS>J+v9?uXk>IFuc-hp&O
zJ)T{-iz}~$)>B!Jn7cYKwbMr9!de7qT|-ApCCo%WKvYg55*B}emf!+TuzQ2=F0Wv6
z^Eos{?osVq1N=JfA-88e#;(4F!A-T|xqe-oF|I*oQVm)RD>1yN6z?XSL;s~`5T=)d
zK^L<i(~^e0Z7J~Hm4q#i5^%EhI7%Ojd87AXl&=Ycf$-k8P6~pSMG#g53Vw`R0M^(9
zU{9+boD2QXb-^D2`hh5m4}zjz2+T%@BDzmFUi=ORCmp~%pMxk^c}VQ;4#9itVc~U&
z#Da}cIFTBSI-OW_w>Scem{@FSc&1X5^;U(K{8Y&vEQ80#`XTXxJhYXR&~x)_s4P|$
z46mi=Kg9q~9~;3e*BBGeY!XaUb2#i-4LJubbT;b3URe*LbeCh@P<`kvG(x|j8?n1{
z3)Y&OV~OexeE7B#t%hRGDYt?P+hEQdThvc<#{3(eXzSxIJjTHommiMm{ReUHPb3Zt
zM$RvvWBBth9&c78L-%8vU~^?6`+g1rr<}sN+B5i6Qwr^gm3Y+cGBRyzF-mk%T~up?
z^_=HO`uqkj<UXTr&kxA<6FaKD5`y6^MP<WeC?ZyLwrcmFIN=R@n=km&@8tv+eFRPV
zIg%b+9z%9bij+TfA|212Od2NBDMWR)$T`j<v#>=}{)?$5U7b>swCQv&eQNE!nm!d9
z)5Z5&MW3xDwG{24vq!~ydW;9Tz?YsG1yHqbFxi%eQEqbtRbM$oS4Hoys#Pp)Qawhi
zX2#QEog|tamP%S*GKjsh==P*Ms(OElB5$3c<=0DS@so0T)$=0VU3po&kFS&M*jw~G
z>n=&@J)(NiyE*A<D^;hwq>T}8DKF$BrAB@wlk8u#<#rcc>L$t2$<iFE)1C3XCtKz9
z;ck2S^Or?~nEK0c_2*%NOC!%mpN{5NZxnd5)C8`aI*Hx4PUXYNGdSbF+5Aa;9<NMX
z$U6VPqs&$LX|+20C~EP0kr^Ldvz#YLt>VSXtNFpowH&d|m|15NzZ+x14Na!(<YB?)
z|F*M>#ZIouwBn#z8`itEhwBS$d5o>S$SFDU)&owQ@ZOnc&U9rRLpRnpcIR5nee5#a
zgU3AZ;0zB>{x`sjr$vjahc`>^_2z;b-W=P{hv&`n;fd>gc)hC+pFZltLrZ<wqS1%<
z{`KMB<9%6H-<K!2`La@uFV{Tx<!l8%cHi#Dmrnb!gseZuoBQ*NtNz?;K>*Ll6S~ig
zK;B#&$g^~VL>4QEDKMCy%?jb^&qKKCcqprG4r3kR)A9Tyyvf!3Su#F?Gu;mGgv|$8
zdFdgJn0c6gir&yg1EY9ok7(W`8N>DeVpv`DfZ9nN<%vCyvHZX|?k)N|bCu#bdu{^f
zswc9MK@!u}Wd6Q8g$H`1vRr5y`$wm9LqY~0Oh3URPG)j@=1I0n%i`W~**q^ihYg)`
z`NqaP?zJeNz4{h#Ok)9GOgzPI8&31WUWM#+s*vw5FXC%WMf_>&8GiHV3~yR`mcON(
zW!o=j`QXB09<!&Izs42|A5k&ORu;2*mkbIH$slr$G=kbC;j~8*Gm0b;ts#Z8r=_6N
zA%&4H(s*krgNe%B5p>uUnJl_ijN-6DJsm;L8ECkgfy=EYu`VMA?yb2P)bBL>bxLr2
z#4Xe*-xgWSTgY**Lr<w2f?a+U62odB{jUm>%g-UwD;H6QIheXN8@sxa&=Q%9B)??T
z=%t{3tH_w{Ohe&}6X-Q08?jUJQFOKtTO~?x!TN%@*NXG6S1nxD-NCs}_b}%CLqr{D
zz>uV7#5J|xx6TXXEq#r^hwrem^%G8Ne#fp)op|%M8?}9vqU3?y>Cxg|Wa-jZoYe=A
zUY{X!$6^@mtdb|2F=HslO_9#toIol=CzI)_X*A4vCaH#q%y_^&va()CjSQObn<*(>
zjj|SN2rjuM=`Yct*EwQuBJLFxn^ux(iUE~1ttPwPYXzTWJz1>SK-s37XtVVey0CjI
z{kAltVH+(-N^?8ypJ*vMV|LP_TUHbhZ9_E%d&o@EmWof>(U7GMbm@*GNf<g)pK2Ex
zspLl6-P}o`d>;kB_n?P;1PgeCH;w7(L$3RLXq(V>Vwd<*ZKSU_H~EsGoFAzR?Zsi2
z=r8y6BaH)oR4cCC{(b}-KU%iTk1opl(d|}W$~o>!G0T1Fkzny%3iqM>QDXn3<4prK
zy(ny^CoSyZK|kx=>GXa#TBhPknU9^R$XfUizc|oVTYEbH!j=}S+)MGNc2j<D8!>NL
z(eeD9RPkd64dCsfztn=L#Ee=$ZKd@}Ce&gq_|G95>GfG-%4l6j-+Qd3=9z0~&1yq3
zbXX;N8dp$u-g5eRWf@&~qC-j@TI46KNyc)*Ycp{P-J7pUw{@8AZ5G*Y=S3tDvw&um
z&ZDMh%Cw-b;54etqHz|&M;tnh+OnsR(d9|h?a@T)_GCPn-&LgE`D1C3$!L<Vl&2R5
zhEd4=!4x;PKh2)on^x%cpqs(6)R8Sif$GAqdQ&h4{z#EZo#?q1eF7mC|8VYJCq~Ns
zgm8}G!SYYAUHTtdh6^vu)fc$7q6I&41;hK*1K9PiNBQ15yq<YYu=;A?cB~2#o#i;B
zTZ(w2vv3}I8tun(vG!>uQma$3NiiN8&Cw_m&&Su52M{+a9C|Xr*x2HWeuJE#)M|yR
zL$;yomkC<j*TS>kYLq#zL~*+wVpi#(He@N<wyPs@#}fG2s$p+0HQarthPrb`(26yJ
zX{>m!PI!h1S6k7);u%K$_XPi<1ZO<<9&E%;Z0^yU2zY)2S7u+w&{>ypD)b^YCs&{}
zxCnjL6yaqLk#97}!{EX!j2MxI1+q!V2~9-TfCMy*J&p(Cjv@NUL3qf7BHb_q<6j42
z-g|$vRruqDb^x|C`op-<7yO?u{<-->a#0|jZ3w~((_rC84#DE`P}~szW)rjh7(4kO
ziaHKrn$01+9eWrGcZDW2BnsVJqQNcE@cS2oNl}d|O<JugZ!F%Z)c5_aQdTPkw(bja
z#ojQq9EINx#)$iYB3^nZL$PQMs$VK&%H4Sg?7axRdoM<!k~$8$YT~M`4mRm5L*G<A
zw3sc&iF4WrTW5gI!VR#JF+sy1bF7}S9oH6G;(qB)DEF{N>4?2>zvl?6*ZWX$*B4rs
z1M#<A<f-x^@Kxb3UfM+C@|hz@?0X!(0f{glp9=nxA#xH~h|<l6TWcXq?wo_t<O(SM
zy@c)7*P#)02g@BEqa>^q`**aXo6kqYrhLPxj^F6Y`v;G=5>((QML&g~I`c|*ikIm{
z7w+{T-|s`{v&%3t_%(v;?vEm?_hU(<;Yq(#iIgm-Qo~ZgKUk?uazP8|z&nssuo|`R
z)}(V+m(juNtLSO$I_llqgw*P`)1W20>CaF{>auYscYPmPukBCkEraMwW++))5&gLb
z1%E<2k{*7Gra_satIj=+dUz+$&Z1<&6HF)VLzz@JGlzD*$tSxDg;bwgOt*^4sPXOv
za_?C!SXWoc;e0Jc&A&sJYwpvioedN=<|(~@^_)gly{4Zh|D*lMpK17s?=+>flOEV~
z<L^>ZY>+C$*1Dp%;d?Kx$nVR?YzMH#;=%k@ZYVeZ9?lp38_CV@#<1#7MNS$xku4TZ
z=IEW%xG`-e2fm)e;dAEmQSU`8cbj<0Xf;-`)Zh`>S}gfOm#6pF=Wo+i@v8-^`P_`P
zg12kTw;yg~ufQ$bG~AS<51MmW`!;?s&XTJa@8ae~*8FntZhkK2oIzH${OzS3Yc6%*
zPol$W)CngZmG8_}@h+TZ@5;W)Ze0JwjUAlb*`Ujv*SYLt=NJ1pUekjM<30H3dk<Dp
z@#Lw#p1kv_Cm-wU#nx-R*eBVG?LT?(O>J);l;O=AWqmkI&4({|`EX>74@V62<;gpI
z`N}0<E}SIvo)|x#)mzLuq5eE$U;zJ@9KfFz1@iC5fqc;?h`-MY=B2``Y>*bh)jL92
zeRdc-{R`t>_rtj$Z$FRRAHgzv5AgJLf~~AB7&@~Li%!u<_8uO^y9P${qP{U)*gKYw
z^*O@1{f~0`kYl`HWE}S&f1Dpri|4uX6S!P8k$dVUah5?cPc}~B-lFfJ-Ykv3o2T;^
z(+plL@>ZILne3)<lJ!OB!;~S}g1eK=m9;r+9G}Y<w&k(qgnYi$D0H3R0&bdeiVs~p
z#S>SaX64(bd4pylYl!~x<eo)5V?z-a953RpcZ&G#Px0r$XZZ8jGpun&3SR4_Fve99
zuZktGI8p-I|0K{OT@v-PrJxrtg#k09k)|kvRLSnJozMr(BgG>zJX~<64q-`AEDCqU
z<AzrX3U;1A&x9N_+nmCmLp2!J>n6g}ZwYRW$cgl=Lu=x7e7}1IXYW?yMOzhYJd5EO
zkc-HLIY`vV#(tSZR942rt3Ck>ekUT=AQ{@tDbN!-j&Hw{V%MI7!3L-BA@VGadY22n
zgy?J$eG-?AZ{yxA!B(I108Nh`LGf-Qdc1ssmLIJmf7^y6<u`acPIP(Ne8J?-pK!d`
z1^<nbbajV}$N}}BVL82NP)k4ZA2o<(naPo6_Hf$rbtGldSQ_gwj-H>MNRADYN%qq;
ziu*E?2EUj?lP}DpMS%-xlnUrwBU8P#8om2Zofd7?r08mGs+M0Sn7GSnv+x}ce7uU3
zg?D6)f)V|kzmB3cjLA=XBS~m%riH?nGhM-yZc3Wdf6W%8nZKP}oh@m^{9Tms#)|ew
z+0Y#2J@me2FZEk)M=jO%lseOqjtX{+#3N@KIoOpZtGdyi_3kv^WFOTVdQjXPPx|)9
zlhzb@(JUix8rI@Xcc+M1#n^`+a_8fseaI$R?6y*TXjH5Z9r5;|ge^X_dbST$fAl8F
zUEb7E=S8nZdQs6jPx|cXL2D!TQEIfvcMEpSxE-$4ZJ`Uv{c<ApJV!D&auBRyJNoNo
zOEaYRQi=C&+TCGI{kB-q`YNH<jIyLChwXH!#)A3|G^fwTrqmQ~LQ|h?rkV1aXo}GW
zG6-K!wUukB`12YXr?{GiuQZ@zt}7`yS)V3f(W4n}bm?qQ9eOlLi!Lo)N;kHslfAPV
z*&SF+=S~orR;f@*^FliKXFfG3%p;Fw%9P?Xo66H>QcL}G>en@uY?Y>neymA!aoa>X
z>^z=~_bJlR-D9bK>1dLCFHc)U27J_9IVuu)+rO!OY3aybRJOT0RRoHDM=Qa{E0UsJ
z5>n(XIs|H;bR!+11r{WC;@`xdSnBx|`_Fztz=QwLv#A{oRc*-dZ^hj4&DfCg2;apH
zb*9-Z%x%Ap<HdqgdE+AP^tphAo69g(@GVPI3vqpSK1#&%u2Cxk8sn28As2^9ZBe-W
z-~eo+B2Xiqm-BK0kr3vC;NgzAGu;ZCj4Uv0n=w@88KLl)0bY0LBYo>KjA+)v6<-a=
zE?I)uSE{)8YBAJ87emiR6?r=}@OR=G)KqN|EQJ=-6}O<hZwsE?XckO`hu|yM;Hp@S
z(<205{P}e(S$qvXNi|4NzKE%T6=J7WB<7_8q|4;v{VKuYjL5?6eW?gGOT^0d1Q?md
zBS&PqPj5R4Rp|pb8yAeSzQIsG9*A)b{(?*B59f9M*x2ZUDIPw!G*-+wQ~Y2X?hl?2
zh#kH`XzUC^xB6iG+7N~&p@W<%j6hS&L45po5dZym2tD5)6mwDpRy#-G)tzV<YTi?c
zX?v(rb?2!{^^aF7<8F7T=ypmXewhU79}I`g#!+y*KL(%l6tOyZ0(!Vkg3;TlVj!M_
zflUiA`Vr%t)e<aQvlIt(v=L;h3z-pmxYeYG%Zl3Y7e1W5=Z$csY6ET#GZEQsGwh>n
zxO89#R=DrN(9^q7zt{oAZEn~a>w~461n+oJ2*$RAV~74h+>eXI;+HY-5<DN5qIgVN
znGBC%Vy8D;Xx1BY(bjefgRY&0`rz}>Z?3{#jjK55atq1v57F)8Q$$L<MyUJ;82$Nz
z*<XHQ`li3oitk3xhDws{3rYIcBt`oV^dw*DJ|x@Fk8IY;(b%`5_w|uH*?t>MZR(05
z6Fz}PTTZ53OQw^Z&TJa%CVV*e7SYH}i^*L{gTlAzP}tTLv}>Ial@H%ciW=J}_?<On
z<=fMzi-Lo5!;5rg`Oze8;m3;$p@EmesbJ#)+SYN1c8k8<;XPx?zWgYi6Fi)k_Y<k_
zq*U^V&mjBRS=6gFmprph(aG>LbUm<yRvjs)ipnaA?y4dEHP>lL#Z4-tyHs=YA^F)i
zQQPEJYWdhkE?3^r?VJy^B<Tz7O#UhI?SJX#b_q`Clw!$avV2jsCm(#@o2Q)U#|FCw
za?!#e{AS=VHu*V%_1=vV{Nk}Z?)x|%+gFLRW(uzHhUxqwd=^XBh&<K5`E0-{Z0o|j
zGGC2nJk{W}(%L*;^f$;&(dS#^R`KYbtNCAp5hq2h=MD2V^0~6jY(H@;yEvHft27H9
zRJ@(Tvn+Y2*DgLg*P2h<uwm1sdpI(2FDpE;Wrc3`e6NQCr~Gu_!S#-OJIaY8G@Mz!
z#hLZ2UHC`43rlQtWz8yAJ~G;k>m1xTzs8Nb2D-ET7I!{((p_YX+<BqaK8`uIkNbVy
z$8zh0zvrR{8!hl;g>zyC()8l%EnXZI>dl)JefW8x4@Y(Q@P12Qe$eL2=2m_@?57`J
zkML);X#u?OP5|%l5P7ayK|JkE5D(7^=1;<>V~$WB**lEyw1)A5^WoexZa??%jNoq;
z2RLTsK^D^m_Z51MqnLB@Mntiv=(tK39p}&cif*gEN7zryIogAcal2d`_mn@*x?|#b
z@wfzbpO`3eXi5BgQZnl)rSLDsRQ4AvWV8P1+_PH-e|(q0L3dAxj;l;Q9C4E0nr3nJ
z^lU!-DVytaa#+G7m(ykQcxgr++o<I8S<&fPwxoc^6cup9h*MndaEf<|xo4f^X<onJ
zG;3}-&HtQFvy=a6_V$p#M->T}mUKhb_HI~qtQ%f0k-(>3l2~_65+7BiAXP6s$NAFO
z;U|lq3wpyt)G^BJ4nx+7LwFb$i?WIFC{Rs-%$yULwp`@LD+}NlS19s~H!#rh7MjX$
zqN}tPcKYI886+6w5!HAcQw2Z6v&hgr1-Jdtu#ggEmXXo8sB%nXB97x_vgoUW=;M`3
z7Jh+L{P>lDFX>sZsS@13x<Y)kFT<y<OYkbbE}kE^Apf!+OOM>c8gYNV-K_!pJ~!c&
z-&65-J%^pp-wTZ1K{fajj61)<BfS$z<=sf8REnHy1hc-mCq@11Lz||F-pwt8>3x*w
zOuR9IJ_(jY>~sZEHyBUXcPP=3-BakM`E**THH%7yD3e<KJdwj*NOnU&f8&`Z3{|5#
zA9bpJvXrW4Xp@tjF6|Q>K%0m9)bw*D84NTeZG|<Yq_kFW`_|LrVH-rh@+NxxTFg1s
zCiFYTl%8%ir~L}sDDKgAiU_czW~E*9xx$L3FSDVSS9g=OieOSCh@E(c9hEM0pmn<)
zX=bDo?ay$gniDQGBgU1c+Pcw%h3*vf#+`l!?W33dJm|NN2fc3gAZxKx+GOZS$9H+s
zIBQS(v(}Szrh3x6b`O$0<UxbSdC;ZAeY9V8AN|_kPKn3dXzUeNn%?R{-tErR)Z#=v
zuR2m=qR2w+w5Q_ncC`2IUYg*vhqC*KtoJc%+A-RSW**&1uX|aNw)1xS)@VV|D&|xi
zX)1izCUj`Q7Mkw2iR5l<pkw`v>F=s_<aEG@rr%giOQa0xs>&+5xqAf}CoLzdTgyng
zQ*h)bYE!$uCT+IYpt6`HwEmnb<vn7u`~g}&Oobd5Eu`{I^Qo8LTx!ahLn9kzk!;VI
zq>kw{%z7%FIX0PA)F{!wHxo!wc03&&E*P()$I^g)qe<$yJUtB+45SZoH0SFe`VrHg
zj0X3i;mtj0l!h$r+bT^D<fN!;yd+7fOHlG6(FOVKFS-xw#O${}Anp4V3*UXh*Lm+T
zf7=@rxxPf`-e-d4@&vN~9%IJw`<OajXpGsl$X<OFb+fDSL9-Hhp6BtV>Krb7DZ<A;
z1=w{b2M4V(v9&4{UyBkjWav?7SVdx^O9T!Yg}_{RRSr3NVQZKZy!Y5(xc)Z$H*E`)
zOxI&;_cf4ET7^$m%W?Id4y3keVw$=-a{p6Bknv*3t1(>a2;)2$JI1JC;1&Z&L~lY+
z;uCb$K8323;0+#c!p-^z*y>V`?k_GO^H2rW%@f?s<5w~7<V8fbS7FKe3ek~LggVg;
zm$_Q>s7MHZoL3qYEt3#q9Ebcdhmp2SXf9s@vFk|yYE6PLcwP`r<^`gkK_LFS;g5bv
z{!mu+M}INfggbk}aE%Xoe(*)UgFmJ{3IJCIV#|j>*iH<_nm?hi_!5Sl84-w>dl2;p
zBSbHQKin<^BB)afsft%rVwPM}kw1S^B}n;!%I~+$D#~7MDq{z=s_eG-qtYJJP4vVJ
z$MrR%#J+9}8aFAza`^-tl$eB!h^ZJhP8sKt7GTUyASg`@ic>U@zDx_o`8rUjT_&_=
zb*znEjP3We;P+}30^`@>i_S(kE0`d_&<xMDwqZz)B_d>P5F<Rt>D4YMJn4m)cYb&l
z8HB?l!*Iqk0yftV;r5_t^tU<!Zi+)M+e9eO5&adDPT=sSY!p7uhfHn}yk3>yM#2Ti
zcGTd-)H*!Aa34x0Pw+@E92GkLL$`nq)b#y<Fh$Wvb+rpxV$MGINrF}l{lAXWgN&B-
zCW-F-$fIcx9lIq*d+Uc&ia4uvs*R=ebK^+PM2UFbR9doVChgy*O!kEfD03VM-A0X8
zJkg{cHF{L?!H}L0*hsVPn$wjUEAriCM|0&|C~kl!1uFYe<rp7YEBK?Hr^4u`QUnc2
zK1fRaBWaa+G(~j83QyrNQvDE5-xefOdukdrOwXi?_1S`ZnNOd#7m7?qvFL^=qcXb-
zRF`szM!dd4vg)<;EdMsO%(_ows~=O5<rB&o^_(U>dnNYJ@90*{C+Z*gjpY4))9i?U
zWVc3=CEBI=UO;#DpW2H{8v5|?sQ%3B2XV5J94AN&=MC-he4%kPJ3dz6$F1YJvqOoG
z_npeqX3gMxYiIL5@45UiYXR@QtHPz<nMV#;!cvo$^5<FF>^)@}TMg3Zxb~GiD%+4}
ztu<n5U&oKlHt>zAP25x3gwIbk<;C;NSwUeNPyeu;EfOucYxXYQeZq>@|Fh;O-py0C
z?%`!S_p<&5Th5$o$N&Ab<4f80+^XlmJ(?U?a*HF!-*x2V#ZLSv$%$n;o%of3Gfz)*
z=72BGf?wmpFQQ$z-y0V`r0&WsNv^z2#*JOY9JKA78*kX-&W?ZGxj@W5Z<c!S!#^I}
z<FY5W#d~pPfH%)o_2HhEe7HjRd?r8h<za%+vqJcHg2WEX)g^!*i2arR+dw{;E3#OQ
z!3^~f?lU-)_q_?_r8QwZF)f_i0tGkFCW8B}JHWC_5At{AL!2@3Fh7(RUFd_NxT<$F
zcgV(YlvJ$X5FX*ye@D2n>nOkYdyHfL#IgTx!3q2k&#OBUc<H-Dj(L{ESMDdXQB4Y$
z6r}R5*fef+N$1El8T@_n2@d#if{QO?a(Azj{C!>)%f86sir{R{8kfVHigMUmC70dK
z<npu$c|0p5kGVCE2TadrC+mDhqPS}E+4xC5`@Iot!SpVb<TYI?H=h1gY3ul_@_f)g
zl|>%iu+UWk2el+|rbQBmEv0aJkTjNA$sprLPgvSJ;K~a}43l<1bEPe=b?rlVi8m&>
z_#vQ1^tF5o#)+2!Fbvy=;U$ik=;(ku-O@04K?>p;6LG8LIIg}FnF_%N^cp1Qn|aZ=
zUng{I!$`~=7>o2BM=?t?4$|`R=sPJ9b?1}u>|Pp_mS&=)yWk3X6yW265)`ewgpyU)
zQ6Ms`XAJ6bqOKmKd>?wV9-(q*BThbSLZMCzW@Nm8jkDmp#=M8~X94H;{)waoUD(+}
zl153((C9urX#1Gn^m$P~^4c(v96X0mL&`96s+Om3Pe;>)&k8j5$9M|-phQs*rqHaM
z>D0Do7F|+MrX?5Wk(}W|8dk4DV>Oumq^pw6?<F)+XbidDT2y>Whps(bM%Eq6X^ga(
zarzk0HnHQ*kT4>>k86c~yq*+JZy>F}O;lpAg;WM_C8s)5>fvEdo{HPZuzWjxT`u}M
zZ|@{E9V<#aZB45Ec9ZR9!NfbhmoDG1rI0svr1-<0-gG#SYO^DS6*`fslQTUW??Md~
zF7#BxRlMd(2l}|tX#+R9?Bzy9v2NsW)Q$QFxQSk3H`+GBjSf`1lC!ZZc|CTaQJOAP
zf83d7wK~zgfljo2jw8jXIneaQ_M)@MPH4-vbfaMpCB^TiYpZQ&$X~%bjo(GfM5n^V
z@*U)=ww)%GS<t>2=HjepN-3i2JHc!V9j@3!--c|YxjT*N#JP2pD7BVM_1BPMq#^Zt
zu!?l#R?@50`qUYrM~ANH3jd=v8PC?Dc(bLn|ByN@t5Bm}?TcyrKo)x^&>GW4R2jU0
z#+;f*8jZ^INLDb+7R;hWqC;$I{4{E~GlkAePbP&0N~CQuf!2qOBM8r>X2lpPzBGz{
z70Hut$Z+zUJCusj29wJ50d(_ZUutvgMLJ&HsaZ;fvPMc#-6;uL7uk)h2K>Wb(e0F}
z{u`C%KQKk@E4<r2L3gXjN|nCBj~_45W7KnKO?Zm0{Trd)`T+lq)+1HpCO*EuhS0(q
zcqUh2K}k8n{}X%G$!Fod?le5j^6+GF7UsT8$6VphxMO}CC*))Cz4j0$2o2!+?;w$x
z_Cu142fl7~z{tB+h?cUz>T#QK+iV>?s|>L*cLjzwFT=GZ+Q@#YfrXFM@X28@PMjgQ
zxdKBPRq%QO@VT7OuumVu9zBMq=!f`|Ao?G6HzD{*BlNF4#FK6Huzh+HiRu-wUp^P}
z(kH{{XDaHRl_OE746g(WYxjdZ#7qzz&aIj7j!40l{_)UW9*eg}51~|P3;bf&z+2G}
zP6Gn*=Xn4uW(7dAdmye{^2g*O{^+~I6X!}D;N0Pcaecfo`kXH|nEB&JN&qs2?|AE{
z09X$fT2DbRJm-d>ZE-M0uL{P`hraMT>kpUN*Hje4F00_wWtF<Dt19I!wJPx*_f*=}
zJy3Zr-=NZWMze~feW%LO@_#BRI?{++KMFf;kAccsMMx=5z+@98$jMH{+=yBD;IaU#
z#}eAxRiWEY9s35%h3@#dkb5`<ol|DPZ3b{nb}90Gm%}8}5Nndx;qRS|ka09Y*DEuG
z<!p!FB(W=W-HVbSCrlaU0UK9ey#5j(em)^sQxuM!KM&ya;z*o46oYwxj>1ha|L!kI
zM&{IXj5at4h4NgOd7Kv3<6=A#`H3kmmtY=q9iA`lqUdfDw3^#6V8nmut@;^%`5W#E
z9op3G4~E(QL*Y-+pSnVltWBi_&!Rgy{^>=Xdj^m~!C;zsc_@wjZv-XGA5A;N@BRDb
z6G(ULWHM2lPFc%l)B5mv;yu2I0*n{a)R`JIp<RdO8?U0cSL;bS!i>CHchOeCuTVYV
zM0aL7)4V)4Qj_qaZa;h|%_EfheG8{Gs{|`F?GWA1kED?!VyMTvBQ*109L-&lNa<-Q
z^goKuJDlsci{nVqAXFNZXehHG+~*u2B_$(!WbeIGDwI-5W->A&+PkzxN+|IwB}tk}
zsc4c>&;9)6@{jN3y7+wV`<(at^_n@m4A0&?ig{rtu^!K2qU=R9oNxs*<!;~~y(;Wq
ze+LC_)nUMl2N;*~5JMzd(Bn`WZZ~|1WB$CwmDjrP+U_qn+2=c2S@hv?;~{)-On^d`
zjG!f)Ykb&Clx)V1rN`IA=wPG-DX<B2P*j@K-%g=Zbux70$_%=6UY<^u&!&(np2c~n
zMB_iG(!zfW=)z<TTBfE&QF=hF#yT`$qDPyT8_+~mL&_VuibNZXX<Mu*S;?%WzJrz|
zE^b2sZgw;-&w&&!uOsm*&g7NvN<Tc@>9V*dz0dceZgC%S_wuD_#eOvIwm;c71kkFR
zfux@oM3rvAWG5X$=dOlOj7cc@H-{3MgwgREVH7$moR%eo(}Ne`bWJ0Ij_i)0`5h6|
zusD(=lOrj#E0Wf(jv}cGQ6xBfJ=qtoCxcnh)OaD9KAUf#aL(fye{mzN-4sK4Yd4Xy
z)@G7X*+Om$V(GC#9Pv&HCEt%H@qn%5BDsy&&23~ImOwsAyr=bXJJG2f6tF&#d^mrn
zTbg^(zwDx&b-PJ{vvf>(=c*|<nL=#$($8fnG;3ih-I7nE)QRb2%zNdh|7B2cUnWuS
zKDzxei;lhBPpL1mN$P11)ivdkL2VvMRUF{{^L)Bj%y(C*g(MYIM8>WM>8*Y-S<Wb-
z$-heISj{0yOf4lPi!zcHJ4_9ghpE}`2$hdHN+U{+(pvRngy)ZuhthEh&pu95{~RX?
zvlBEW_XG{KoS<aUlT^l;f=Y%bX^h25TH-dqKH&h<Sv0_E-VHF-h9Q;^_>Yb35rBU$
z1Yyp0Au#9{f_VE8;QDF=)V~zw{m=+FQ5ylrmxh5`a47e9hr;I-p&<4l1b(~^fXP+?
zuxV`oyczO`l|TJKr#}GHVuE1t+70j)H-p!mIC#vRE^*^`fZ*yKFz@YdIOnk!oZ3_2
zAIk*k`&kg&l?~!Oxtzb654TS8y;T994?QY{FAI-C-L(_2<mhR*V{jg-cb$iKVf@~c
zbrCZ2@4~=vJp>=;Y>_YjLBRP|IAhihS1Mk??%kabnez!`I(k6D;um;)8Gw=20vMn_
z0!!H_EXC32wqhI>TaCwe-csnfaT3nqcY4`O8C<t-2I?lsV`lhlJZ-Fqn<pw^bb|_J
zanHEom_>LqO#{V5wRk6q;f?!<+omqX+YWjtowf``uPsM~XNLItixFP$UyV}+)}Z-!
zQ#|q793AQ`F#50+j^1p8!$x+fEargrb&jYN<Ah!cE*M$mid~lOsKxhI|JixrlzMLr
zRq@4zoP~F>-XG0`0?}SM2<I;kM%gtXn6WGrzs(K9qy1s1b~zkJx<~NOjKEJZk^JEu
zi5_}UsGJam{%505q%I1FxF5p$QWUE0jl%P$Q8;HX5+7$o;!M78->@SB9bbo|rBXQd
zyN2P{9ijLtHw3pA1Y=fu5H_z5#8IXJ_+zp^dOYz(xt%`vRn;5Q?|EXKlLxNpa>GV{
zSM2X}#sCK=tl+;_%t8m0&$7e0KW*@bsWrYXvqb$rYtd$vIoj>#?x<I5coxwZ69QLZ
z>FJfY?CT2jRbGy>yqDqDLOslXycDxVb?_u0DhDv!nx}<OzXS*R7Gs3sB3x^?0A=^6
zp;(m){`}4RQR?$C*GG}hn&zNG^DOiop@7P&GjXfw47}heixPp;(9C5Dwrfh`TJDqD
z;v<PFC&jU<ax5m@6UEh{qmXf)e?g@n&eRdWfj`4g&7BKkx||1G@EbC&|A2zRUKnrA
z*)h*P!mDNPp(gDOoVwEi8#~&-rlSQ!?mmKs+(vlg##uKiJX1WT8YDz-!jC!Epws>m
zIAorOZ<VJ(u<1CsmLG<D0VPoNu>fAp&V`3F_woHjDqLQf1hyGF;P-^Bpp~@|koS^(
z#`}R>iW^Mhz2z%P*6<(a?rxsD1{@5GK<ChMIAWv^zt${;ozD>t^fPeXrwysswO~<-
z79@&lL+3mr*uJ_I?l!l9tjKdPdEEy7oHwrjrUm{~JcJ`3AHWas2B1(?*qk>PTCYh!
z*%2|gdHE_>N?wApji*4e;y9$%6u?P7zx{Y27c|D~2QQTX5IZFWqH<DD+H4B>l_nr?
zJOP$`-v;T;o8ePmC<xRBf|^S(SW-BQ%~}r`pErQC!)A#390v)pTfyIG8?-;&3b#yR
zVQo(gtd5L<$3F2eKF$_W8yw+DaxF{Ta+@U?++iKh@38nociF5n^=w>r6C0>(X6el>
zY~oNWyDsvS#eR9hf(#p(RAnQZbo@Uyc)x|oooHusw~vS6)lzUrTOC%YX+lL8K(6Id
zIH#%yQw$eDnb#sv9i9iKBh=t&uNFAB^O;x4O7Qz&0?9Kh;ZMFTs2+3#%?&OvbC(CG
z%K1WpV-WmW5CN~FH-fcLJZRf-Pu0;xsO#PhHi~<p*DDQ@&u2n~bT)Kyu5r6zAskaE
zfn{2UVQ=bjh-94KZg>%%*Ixzu$2<$pXWi9vA3~UR8x$J7gE7au;k9}%tcm&owXc7J
z`-%bRSvCw8`~E>)j3Cx0jldQI?iYW^nUhPz(JhN}jmsxuPy1w4oF#)SV+IB+P(WM3
zxu`fo2}dZY;d9qT7?!Pxr7fJLV5N%-d1ik;&+&VU*zgXOGj0g-#>SPr+j%hr2O2q_
zt1KGd7Hq=!vN+V$;ylkINw`m7FWy;|ig(|qVVH6zX7y#^1(96*u`C~76&7KH^dTHR
ze3<)Aj-#E}X-s;34#(UtN8kFZ$l5D#)QD<arE?d3lkVZwkBzw1p&9=^e~KYN?Ko5X
zHAYu-qTr@axY@7=U&`^UvB)3v86L*9TLr0itT2fdjiSjonsQ%_r9Y|Sw0D&x?G&3x
zJ<p{n@8VRNnJr80+vKQoy#kF7pF^L+=TT#{GChe^qn;fL=~I#h#qQA}`yD_^>via?
zqaIyfY(P3gD`@!zBNDYWrp`_ia(6Z-wMq+`Bf>K|OKhpj$ew3Y9BKXxC(?fDOk1|Q
zlB&2n9pf3~UH^Gf@)&O_QTCw@4PS~@^rLEFe^P1mC$=MiyCnlDy*iNO*91{ZeGskF
z2_~nb!4y9(guHq8{P^7v>W~Yi<PD*;<WVS@s)dn3N*G1{2&08A;dJv+IIS^@px;j-
z=(bNJ9T*u!#%H4_&UZcCP>Uwt5gX{%`wjH2bt64^7DM(QH_;lAEo8AcmPQ7}(I5MG
z?vdC^7cOk2Air&Nc6I`3bSBV+<J;+dB<JrKB$6lR4KCols|&UKtW&m|1`?CVIA9Nj
zSR|7Y?WJY&QfR$&D(x7RM)Ug9XmWQt3BJsrtmaJGP`8iFDzfPEx&5@NESvJOa|m|j
z(xvrzl+Ag_ZpQhPx1fNkB@4;!Yau<aDWZGX2dUJfn7Zbb(3B4)bgAGF8LufNm+z&t
zHMNXZ%sR~THiv17>Ji?ZJ3{Zr9VMN>qoiDQl)^-hQTy^^q#JdN*5~qW*12Q!_r@_=
zG50sqnE9Ipx&3CzIeo0nb%0rR53}u&0+6y@5Ed=sxtwGn(8v|yJ#is$L;;vLLjV@*
zMnKJ@K(KBPfk)dyVf)z-m@XOu&z}T<sAT}`GY^36=KfHa=?`x_1HjQV2ySnQhJCG@
zz%DxuBsuS*xMw?bF5LktJV!Y8*KQbGz8755(qPewOeo^_9RC~HP$ifL2a@uk+o1>!
z+m*mt+cFreI|6G?o`mq(=fQl{dDsFM;9$f>*#GV}7>%vt83)em@MwbRA0LCW)-#Z*
zc>%)wz0;HS0e;-#zVqqd;lF}D_^dO``zL(1>m`i-wjw;wI|dh+iQy`935;4h0ej7)
z@vG5PtYWg*CntwVf8_DQo!R&=UJ(scl`y?t1!Zj1(V}@F*6VAaJa-!X{i%&Z{QY}A
zQU_0z=weWfK6gYJAat)l#h#T|@NpG>Z8Jvu+a@^fkQvIwti^02OLP^tMw|OKs2694
z59d4J*?W!{&gUIPU!3t$uq!IPb3;E9?z1}PiS7NI@ymR0VxTXMO!Y&d5`PRT4Zt&5
zfw(v-h%>!{@#Ro3jw=npU0VFib14*m%7>xxhA_OvJ6US)!*KI(7?%AD!}8Bzd<Gqc
zqjrViIn6MXei(|WwxRs}9)fX7A^0mg7-KI4;iVUWXwVmcj|BoyZqN^(zxT!ZJ3e@f
zcfZiV3-u>>;-*{fSmx!18w6bONQN`E%km6$kt2SPallXMcKkeMiy1-IIQOw7hH6^i
zy?y4mqsNpxkxfwcfHAs#UWKh-gawI)Sl6^1gQN{G&p{vE@^v{MOa~`U<!nVO!2G@1
z7+9l;Q-?Hg=Ay;iH@XlL3e|B$iz@bstKjxkN*K0-&upsaV#eRu7{_K|aF{&aD3!z9
zhdhh$O9sD-PebjgQ}FG0Y0f2_z;}C+SfV)|Uvl^A!Pe1e&3m__<AkwMO9+pcai77)
zVW_w}0FMp&;mMU>P(ShqoR#Z^M3FC`efuN4uzwGQ58uF6wU=NT{u~Y!JO#&doUwnV
z2|ncChwAuRxa55semnB_mfv;oO}z|icP~Jwz*(54eiAH}9|b|pQji%g0@Y03J^Ha9
zzA0wF3cI}^adH>*&Dahr<KkeGY&5(W42GX7J|NTU0uG_}aJt44I^LSX#kAEhsn8Iv
z_b-FT2X*1UBLo)|0h(uP!%&tch;3d1-yds0znuo`c3A{9g;!yeMjIr^wSu{88ysHF
z8B+IJ0M<VQ`tSfwLjz1wod+4c6Tx@#cql$Nn!5;uIjcnwT%HUwNw-0^^vobDupVN%
z{QEaM1wcDd5Pp;kgPhh#xGW$J>fz%+xnvdGG28~Ve6F#AKQE381;PpAASiPWg$cVN
zU_@awT$!*LCcle=E1p{+zi}&gtMae5o8juMjqqZjHH>S!%`%tYVSldPW;rp{>|}5?
zs};J<&eh*$8jdyW<fd9?XjaD*zddAjmmjm$ku7YiOe0GdZDflsK41$I8`;V9|5&Ei
zV`gsijOD6#v8%4%Sdzv>xT2y7ZKnXniY<kR`?}y|w-6@F$b)^8BAmId47M|tfL$fQ
z3CZO!(_uB7s5FD~@mBEYu^sIByAJGEyTQZXUNAH>01|6L!TJ7r*e10FYAd$F<e%H2
zX3j1+W4#CB_oP5?Q#z<CXTkN-9Jp_g4==|bge1vQh_O5hr#Y`(?J?*5X<r5|{RS)5
z+aPnh0j5rS2#Qzv8Rz_S7&7?;DUsj6wBaXc&+Uhavj$<_w_zB)M*uCegivnrNE}=*
ziXXMcq4SRM7=K6#YilQAhv-xs7cPtHV`p;K=PbPSLJ?g>RPegy0?g)Js`sn3QNo(=
zZ=W8vJmLKaVGDd9Vu~s}4|p%gA1!7F<A+lb_`)s<cfv+2iQIxF0$Z_eUjh~>C8DPJ
zE*$@sd#UE6b6!;@Dv4&}ZrwcW*jIp}!o~PBsT76h9mQL1CosO`3_gy%fOa95@kGRR
zblp>l1Le1|=X)(4)NerV;{PyR@(D(yKf}P89r&Z-4VL(QK%H5i@y4fKG^zT97NrB|
zko^zEO@yetMVQ9<iO?3|G30h^91Yoyr!WaAy8dV)cTi2H+Y!?!Xzg^mr8$%G6lPKG
zgt;_VlxK1NDpST!HS+6PNdG<6pv)>Q;?)+4iqatm9X(p|dl{(}uAt<_Ms)q+YFazf
zghof1(fxC4>BtjHdhyblo;BGLE3>D17f0IpZylBLF8Ssj7b?(mqsk5LoGs%)<;9+K
zX`dH~M|hLWQXewy^C6{!zC_?hGPQo>Zsbp=EBz@?Er2Tb2hh1c0rbx<kQQ7Dq+3!!
zq!=1RO-(^Gbupj)90;Z-qeJL!TnJhG4WT`Wp%gSVj9M>;QKox1=}+T4o|h5ie?F2D
zdDiE}w)GSeA5C(}8|Z4;M*7|mLji)D>FTmA<dqgnUp~f>&yjf2b=gWFyNxDwY@@xU
z3FH>Coh+8_pxKg%bl^iGZMnt0p`067#Qj-;Zb?*SxQEuMap%>Py%atog*<yxD6J!v
zjyy;s+lq8reJX?I6lBtl<b5Q%DT|i-?xzOJY+9g`LxZ#UIY%UqK7Pof$#)J=&4GN1
z3@)G}SV*shim3E<5k+q~NH5ik$+xYT4)XrhWa&fXb@&j?RxPEzW2LlRwv0Ay=UuGV
zoUgO^Fp0+;rqSg*ul(*X6^b08p&3WWMCl06W_@F7{omLfx9?0~@CR#e>|+VLhS=5d
z0`N*l5Z?C)g8Kp?Skom4!Jh=+-JHKH76+OC?m@O7bv$R;IKy-;M;OWT0C$B0;n#aV
zxU$v<X8iVni+B8Bew8mw9_0)1)xHoU9S*xqVxW6gJY3rp54vaLA$?yEbZ_4b{znp_
z^H>sik4l04&1sOoG!rVfXMtX8HpDpO!HV$(;LuS7ZmA`(DYg`BA02|7LnSbW9R-DX
zCt<3`1^D}d^Zy=Hz?hC|Q2kp6|EWKOcRY{NQq=~K{1Wuy-$8l7N4W9&D`*=1gi{TF
zpw9mvEJ+o@#;}p7?#R8$e1Cm{iQ^b0NvxYZ5zPfAqw?#iIPHooYA49y7<~oY-8&m)
z(-kpAK?w&-Rq)ajb<~Vqh@)B-<Hebp*yyZ{9ch4`mkD1yT8cN{>S4sEW$6B4IlgH#
z#HBSxc<%6O^o(1Bfo7)IAZ?BjPu3#tvBVxNYczargC8U9FjCk7BX>Ju&<H1d7RGbP
zk6o~Nz8h+WyW_Q!9(cFa6L0+T!Vx2UP*}hhRX_XUx4V9LKh+;~4Eb5)V*t*J3B>2W
z1Gy_Y2-PnJp~KH0{392Pf4Bq7iUs2&-bJ?+2u8!2Ae3DngtI3E;pO~5)S1B9%G{eU
zg3pP}NBiRmZ9lwV>x<)ke7Gyq8x!rlaO_e~ygbPR4PU!)zl1CPv~j_0LeAJ!x(;_P
zb;Py2>)qgBhiM%)sI<-+<@mYgBkz0nT{Op2GtAIyy9oxpH^$N2#~#i*-_QCD(aUrN
zY84xx?>Bvv(AC3bNlQ_xnJ{}Qa*rj$RR^`O_0<x5FQ<WzoEG7@0}FV5TMb<&sbZj|
zGH%b9kCz`QV%>zfIMaGIKFLzROOI!whlCuiUOAofjAYQSXeth0pNy05P2%UI37CFV
z5}i0d<gmCH1|*L`@$Vw|#C{~|y%)kh?vh$xI1G921F$El4{pBx3Gwh9dZK&4m;12W
zH++O(o%bOA>J3DCy#&FB&mnJiD_FQahAq4PgOvORXe+J*eV)%{g;ntS=neSDIh6&!
zE<)YHb8su<6wE(#3?i$_;MUDzsLkd~Ji}ZF<6M^}^)winz6YAda28(4HmJF?nY&aX
zA*VbL?A5&BxvDdyx7b28vw+?;CeYDk#51$Y;f$_6jQ^kmw%q_Nj@mE<HNkSL2K3HZ
z3@EV(%-R=#zpokut(ga6FBBnEq5_iqS|P)bcf725zx-<xJmB-2{8dxo>fT9^utW+p
zT*W{sTog`>6o!vs0^lYz#O$*FF!_OgHpBA|+k9kz*-jp2JNRzfBti)8b_l@+em6SY
zEdYgwN5X{RAvSB#Aj|or1!bH3LF!Ea6vPI@T8%K+WDv$X+YwNl5C_^*w?ZeM;ha|5
z1Tq(6U~1uN@IH5!O>(-;q?@Xlq;fTTYIBP%)4#>$-MGaJQmWak-rFp%q=u>8y~`eL
zs$*z=kF7m-pY>TZu-EwwtmtC{doaF{-M9FU?fu)#w%%@KB7HAeeq<Lb{L;r98Nbgx
z5dxnVBG55goaeVE!)ON?m^V{_Gx?O@mgXXuagKr1Wj#>sHw2ORHDIxEEu4|GfhDmH
zP{I3avX?xduF)4Zt_+4?sYvLY8Uxe$bMBB{0w}sBLie`au;O$w{OL-CXM9E-bYwrw
z*Up3Y-wWX2%VO9gdKj`d9fy0yXP_<RB53hF;Pm&Ea4qjX%&%#JJno`ecH}uY{p^B0
zB0ccz#&=jW?Kf~S1UR-0Lb}L5xRxM@%0j|uEhvKLt;XPsEu7DuCV@hyCg8{S(s;>e
z8XkK-9nFu*<JYru@c7&LyhEynW}_G3%Iiz;2Qd_^A*{X6XQh?K=;dyLFMpWgnN)L}
zsT+usc>d4TJsfX5T#r8Ln=tfR9FBJ5UaHUA`F&w0+N31o!%^HV{y81p1hX(&GY9*2
z9Ke}h3Nh5b1c!#oP^t77?r}MVHEQQDL9!eJMXusZi3(IzzJ(L*YH;QLdQ|CrfOD5W
z!kqIh_*4BkR$hOJ`PT2yez*%ipZ|he)_+GgqduHCe+cuE1?bwi5ws#}B<-6kN<CF$
zD9KZdHc3j*`C2LJNt{G?OsCN988YPjYdXz;G?Q{K%p%2{xg;0M9h@F2bjwhkyyq<9
zygv>4(V#_F(}7x6>5$J4T`EmpMqLwD(Ak|Uso?!8idJ7k3*1cUO^i7`+GIfy9#-VN
z(1td2+S1`TdzvobNSZ$DXnna8C3icM*%(*4Cgnz*Bi%`~)15-kc~E?yCmo;cMS&N+
zXw7nOdRpgAxduL@aLI>;<bCPPE?=tX_NC%Ae)Qm^ADt5Or-PyXG~D7(V#@+Z>_Px3
z@%c~ui9njNB#7SC1<_5PVDgX(p~(6W3fmh>H=M(0zfL&GD@Bmo>`2m6iQ@U4^)${i
znp`tCkYdwD`XRlEjQuy$xw<V>wlIz?EaGXu1m9gX^O;WGHcIeHAa9NB)IMql9c$k~
z+2!0Nk-n39Lbxx>d^b(dNFp|M4@nColV?{lwLjcT>NiqI{75P(WTcVgmUN2s$)Hhd
zGfAKAqguHvnkT%Ues=Ds%XhMAs3?b4Md#Ao)p;~+$^km|>Hs-#|HDNW?zx&&NV{(r
z(schKzH>TAfrSUjR=Joyoi3*Pe7<uir-T&#a=woJA^Lmt5WW0zh$0kADaE>!BsP@N
z%al^;&Ml=I7rwAVm0wuu{IATXtA||_|HUSH4zlcwe{7eyAT(7ALdh3FIBzcqA_@PP
z@wP$Mcjgb1&KzJZ_2a>g^g!#d18n)Z4jj6ip{3dbMo#tyaqe@Oaoh*C>v_YU7Ef3{
zXvIBsHjsPD0N(NpWaVrtxL)oI-bVr;PBI1t@ABN?{9O>3vj<*8rohG@X<)ZH6JpA;
zz-?v@7~af-<G!4U`k@H!TrCE}X(e!KTM4KwDFe-%V{qm&cbi}1E}Yiuu#_{hf`8nF
z*|&MFx#tnoU3v<c@!V<X^BQEr-$QIsH@yGc3wv#T!?xBz2)!)8UA80evgaslUN#yf
zxR+OM#CR-xC5cK`C*r20$rx+FdkEvF<EKhF>~>ecr+u^0KUNWs|DKNzT~&DgUkwAs
zEyVv!7vtycOHk*G7P3Z$mTwUIx^*zATNnLa>tp8w1Jpma0x#}ai5^a?&}FVMo_xOs
zH49C#eT_N#d|!+EQY>-2yftn(W`iDzc4(Vtk52_1(ZOjQTAg%4jjzt=qu`2mMs66!
z^T={O9{9-F6aTL8LK#_aJpRfXH8Op0)nZ@V$@@#0;D@t{{jl$cA9gJA$BRz>=o9IW
z-`D%2i<dvv>-wYPD1S7$>W5u+ez>*M7dN~3qH!(nt55eq7k6*Gx!()ru6yF#|2#0J
z#T{!K-LUeiD@y0MV64A0o?PgJmR}wD=Qv>fYJ2?DZHr1fY*1mcHF}p=VyvPCem-oD
zit=Vyu-^nvi>$%&$kjOf!U!7-R-)tK71%zKyKvk&6Z(c8?wFv9s=hktat(3u7{EKW
z+PLVDCLaB)fleBWF(+mro~~3!Az?KXGE%{*$x2xI-#l!aq==P{b5OEm7V34$V<_K8
zC56sFy|c0?-8~H*<fif-+GMQfKEj=`6S!MS5@Y<w<2?|=%r9fGa=R$M<BsCb*%A0i
zPY`vM{RQJ~1CXoT2O%qdLPp#-$h!3vn!k0!q7fhA;I~eYx%mbf175<Szt1^KuN9U(
zf6Q~L4`G$Z129Xj2fL#+5M9B&Y23HQzF&o=8JED*{ygX1orWo|kHZF;Bapc45EPph
z!Fo6V!XvU_=dlb}uAKt4cXmU!#|}6;V=Jf+Y=j=+aPV~Yhn`6upkllZK25QKFMG^k
zRK78EtFDCFxdxydtqX@`2qsNqP!XdEw{$ch(P0r>xwQZaP1WIPfhsI{ssy1@bK%Hn
zdG5NDg|Y5aAX8y79P6G04$mfnnXDAZ{U-%}&!k{a?|5)IG8W9$L}2XP5fIq;kBzY9
z*_MC(OxLoXZNA*k2HyW+GrI<v4A1k)Ci0%bc>x&t_aFNnHOvm0{9$j?2AQn*0GnK}
z1nQD~A%*#aNo63k?$n1}_l@8}+keb;UIUxxcAuSExCUz4ETPfb0^(xcv87Hm?BC{U
zcHZ+AdvvslrH!d#4*iwvfOi#ZTXKu3MpUz#pKddW#Wn2d?Yqp-tBzf~QqSTt?lGms
z_nGsL`z+hz0UP(Ai5a;yvrCg&nPE-`+oAb^S?GLYy}bj>eZCOnjuU~XW#eG>S1EY;
zaw>#G%>?t8irkB#&d)EJ5T2j|ZC4E-)o~T~m72nkjU{{-u!Y$Cb&%-c1{3_f;n>vx
z_-+yg%cn(yjyPxGByWZO-0hHbdMC_iOoIC1y&z_g&iRY`V9#vs{cAV?t4oSt@0mlK
z)65wgr+Am@!8uqJ#W|*i6_6@(8|rEsVb05DaC+AcrrSCpx%e~exY-MPX8nY)f4|}H
z*+0<NGz3rW1yJ#d5PHXq#Dkkeu|8%jCdY{5!DLCau9%3z(o=9ku?+t8m&1#jW?^%c
zBAWIo<K0K<D6F{{qZBnUX{|Qy_@;v$V2l#C*WhY0<+EWkG>q^^o25Q@OO?-b!y<9l
z>kT+;zXi>dw<1*~V9c0AH2Sm~tGJi9v^5noerBN0y#3r=l8Zu*^6}q_gDBE?2(SAc
z!KNuE@L$_$6ghhy^Rh3Yd-^pT$iInSuU2Cv&l<m(b06y?n()Ja%~(OL_~KGK{#^PR
zA3yHI$PJ(H*ZdwV`uGDoF83p34dWbrL29lXLEjBW(X|#)S{pK!swKF);hF^P51PRD
zDAGJbJB3=$%TP=F3=*`Ir+138>E-ZTQg5D5Kg(2zBGpNp7E#ck2Kk)TqN(;k<v)ma
z#p=><pFR~iFQ*yj49WDX5h+bFrYEya=zy#l{Twu>ZPgYe9co1<g>5J(-j+_iwWCjJ
z4%Fc8NYmrjQNvCr?!b1YEC&~wG|!c8ymO`QbT^V*=uSs&yVHDQ50b9;px&jPwB)oW
zxlHk*&R8#6^45#47<yCVac{aW;X}h4eW>_@4|O~E(%;A2ZEoX77hn64{(67%niN1S
zl>xM1eIR|;4x$**U>e&QO!bW+biF2&9@U4@yS8w;$a`GB6{2|Fc|DnNw?s<U2HK|=
zL!Bj?Xyo`Uv|)QJxe3NoIq#eg?b%90Hk_|BCxMpsC6GkJcDizC2MumXq=ODSdCqqi
z$xho%kwe_~$vap7?(Ct$<H>YAZ7&H%r_dCqR64XQjSLjiDRWE)DSytOqWhVYaC9Ge
zZsQysJDx{Y&ZZ52vU#ta=X4TtNpwXXNet&v;OPT&!Y-d&2J=Z~e*s0$D<sMDg_N|Y
zh;|<<qU+KJ>EXtMwDZwH%AZ+G&-h+TEWen<>x-%4Q!xn&mQelJ6571sBO4d}k(rf$
zWSeGov+ok$S@xv?_QdKRn>HW-v(E^^vqnMK|4#rmxD2yO|9<wXw2w8`^)r=5Md)z1
z0>fYSa6NAwl&3pGe3S>IW_!ZQa4$%E>Ba9oUeGY)0mnqGVEzw1F#NU*&b?R-@+G{V
zD&`FDW&GjCz<RI>+6wC)?f@;r-JB(w4E1MH;H!8#I0j~d*~=_Q^v?nN0q&`DDS#~o
zMIhpK5PqL8hL>rjoC9(cdSgz*(7}t){pu=YNmYUZ&v9P4cn|VBn!x46V^9cr23kfP
z(5(9w=Gb<DM)DWl{rnCN%lbj}#xOi65yYS%VH`ms7&~qZo^KzAKF7zScZ3vXESZFY
zJ(F>4A@30DPse{x<<Q(s0fk=7Mp2%}NjkyLIDeHnUq%fL;uhfQQ;V>siSMgEXkzUT
zZIt~Dxalh)zF3N7)q2=hxC}kQm*W^6Lmd3O61A?Z!jrzns4Zc_yO^eU!@wNno7Q5i
zxh4A5T4DHN8=SGv7Q4Ib(3)phngSiMA#WYtyyk?H8=TRo(FJRHR%iSnH!O&B$3Ge#
zc;>AK+QfTe${0T1+2)0T?>S#b%Nv(Od7~lc>zyk1Mz1U0xbcWLj!f`Ibz^T_AmEK_
zO1!Xjp%*Sb<B9jCdE%rf56rpgjxYM$Fh<@D$C4{9G;+axhR*mx%L%W`uH!612kg3S
zkC{8|aJ#-Ox_`IEraUY3K})Q9ycYeqoA?922VRabL9-!aoVsx}Mt$RXBrpEEx2!-l
zgXQ?_<T4DLppTN#y7=X(4$5f~R`W_-?<Z}%t*?dl`<GyI*J5;Hi%@0P0!(<KhG*ne
zadeO}_FkTk6+-jyn#o)g&z+4I-YKBl0(nf`B!}TO(@|JV7RRlf#_zII@Kl{N*7r|D
znlS;bH6^i&J61O?6vOvoV|kBU6usj|VcaZXJl*hr*8e~FFk%oa%=$p!?+>W#>jiPy
zub{D}8!m-)L3vy!ybXQBxk@jgYUnu}IM52=v!1|k)<Y=&`2d{t??HO(T~Ik&4GGV0
z!U&P;(7|^PgOS{~&An>#MNh(NtD~?vyA-PH55k#ud=L9J2Ts)N1LI_#E7nhjM?bmG
z=yC$2E5^a;nbF|p5dxck`9OT9D_0*ozz(#6FO{a?^=K96-mQR(7W#0jPzPq|0tjeo
z!GlB%&{JOorgIiRrnMTZ%vOP!nM&}jToF|AXF=~WIdBP(ft?+bK`eF>SoTeThxhpV
z{jntIl}W<V+Y%7?SqwUVjRu9sBjJ*q5TNy6#+D7RMU(p3(T#mfwzH3|`rFT@W(=_K
zbwjMVV3--Z4YT_X2HB;>e^}PGU+nMIUgkQspIHkGvNI|g@SW#@k}v8*+qECeMD8&=
zEZ@j}EV#!sWa?N~VlDgHeU}vt-DQ_X*R#C0x7h@pTg>50C6k|3$?g>2WVQQmvX%dC
zvaH%lw)j&OTcKOcVpDIkHKH}_;qJTaLTfFX_@a)jyi?DT+wQT;I~$mjSQD#DeZ;KB
zKV@@k+Sxsw_iR;mFN@wjzy}vXAksbxPFxtv=V_AgKxPVTDwzSUN9TaqGZkoSTnsuv
z0P}P7V4kTVEInupT4m<&H`*G$YdOFnVHa4{$DeaE{GgK0rKU`Xgtmtpq3BE;_&nMM
zlUjGchflj;vv@K{uS^Ba+zi+|dOs{J$c1o+0*GH#%sFFau&bXl;lH1THTyZowC@^x
z;vQms<$Lhs{3FQBeh$Co-tlwRCz!`)(6uYS!_I-9Jb%~+<H83(op&ax90XDH!wCF%
zXcXRz9*v%E<Iu%>JkC#+!duTKVT{96o?)4e&rlu@ADn|fzs|?93#xddW&tLuFUEJf
zG!R-BVb_BtC@ZlBBX^o$K$|IEkTU1l98b&&4@3zWf0W4y<j>1cJR2E>FSc#MuQxeY
z)N&_=6enSO<zCc!n}$=Q_o2LFHV&N6!;0yJXnCL*4=9&mXyZ{llyDM97@tM+*%wiF
z{1x0V_6BB7sX|%JJDB59$C;emP4)RdY&3j=mS>*fsW}~Z?${gThF!eY){T>Q^<urj
zFEkJzK;f>xSlcB?$Gn8;=_l@|+Ax}&#m7<FNpZSlAxU-u6Y1eaX<8LBl`gBvQs=iB
z6nI&ls<zA~4Fg5a#Zw})J1T^m)#<G=KjS>qpm70OB+|#IB7*4H%cbPa^yyrj0r{R=
zLFo@xl5O)U600^QwOpRdu`{EM|I8^Q#e)1tTG16xYid4eL-rlEG<JY{9t0ey;D-a{
zHapU{;&t@N#fe(Qok{krGs&%RA+shInqlrrg7vQCgKl)|s2gpRa;I+_-D%e=cgk4l
zL5^oVXtt~;DI|MRoS+w3Zt<d{0^Ssr>`e#ceW<g_hlB%sDQ%`73BU29iZlN7cXt4N
z2@Rwy&mijd45s^GA@naPl=RMqk@o9w+BH3rYCNN8>!tO)*S&!bMsK7^?_+3z(`MS#
zzJ=mF;^+hC3~n^oO1eCcEbwL<Z7)xt%st!byT=alUYbZ}r|qQF{+(3ybQcX?-c7%9
zlE^h?4}Ek@CZ@BOeos%KTf-@&-JVK)SJG%-W;&GyWbl4>CV5Hiqb0BQ(WPTqv_5!0
z38`h%`w!W4{y+}>G|Hu=J$$y4l1Izt93Zue2Pg^h>CfeSQdB9RDt_kK)K@?gO$%vF
zULk#bUP$#)MRa*-5#_iPk!4g7J&P+M)1{qk=Il-;HolV`+SAFLOWv|govm!ki&i$X
zxs???ZDspEw=$2n?JQ!|Pxid*CtKpx%}!1H!nQr03;SfPplqK#FeiSO40nc!${sM<
z&lBe8dcpBRFR=dZ4%$W@aK6O?3jYJ>xuOpVw^u<>vjr$jZ~|X1U-0G6g5~#Op{FbX
zO2+Smb)R>`#fW4$+MEJ!)c6@EHxn`^?T15|Iq<tD4;=gQVQx_&?By)Lw8MvB;QbM}
zKJyfGUq26ypDshDFn8nFSA$ebEeLfqfN=3ceqU~ZcM8v8gUBn88_l!Eb3eh^bv=-H
z^d}^Y8UXo-f8Y@%gikd_;-6n4SaoR(N=J!dkD3HpzmdZDy_4__=br1-Ov3`R>FDu5
z4mFo5;9&7=tm~hPEz9O(RJ<|@oL9y6r|O(pz7Qu0X>iAYCMpVPqrne`PudXWuj-&%
ziY`|2T#3UB1JrxA97FaPVvB|mTJXL1#Xw`c!86D+;!JVtA2UqwTZ@+)E%4|(D{NbD
zjSsHa;LsOad^yP;zp6XnL3G4m?REHgjuV=Xa7N$%obhln?<#4z^7{|Z?bx~Dqz7*3
zuI`RcW8HDpb$85p=Z+T!+%bFD9ZkQu<M(=ZJf7)}Cs(^8_PF7Zt!`){<c9mhT=7?v
z3x1sLf`1&HITywW&m3QeyKXq*t11U{xMGh5hwRWIfp@^{Z1A*#H5R|M#M}cGxY&3t
zif|X#;Y?G^Q#C=aTgJG}Y&E*K8e#Cdl_=c40yo(%#~Dq_aG`-d_FmA%z4A*@BbjjA
zPe4t3hE3PC@TZI>-rA^v)=w5;v)V%RNmR$>?W%ZWfeN-JD4}A@JbbIDh;|$2VCB78
zs3EI>r9m^X{^ksv!RKwAPBN%;d@9y`pNt_qN1PWi5tENfAv8<is&C>bB_M`fLu2r2
zrzqN98--S}!gv9M@aEt@kUhuy-{<-OM*jps&bApS{sNwtKEa>X51{({9o!Lo1NFZ;
zAojoK5TDTsUOG?U&7+49Z`BADx9@?!LM@0#+y<?ym9Xp2b+BYtV8@n=FsuG7sK}lI
z--u(dsJ@JM{Yqe?bs>z6&x5g<`(fYS3`h=70a~2|FXa=VWz;sX<-I9676}i}2SQVf
z7vvN=!;=Mea8cg^&b6)q?{6#N@)iSl7^Dl1mk^XJwBd^05|G@s2wq4ofU0+@kkqLR
zA)n`i+<8U#@?<tlStt*OT&9Cp&r~QMkcQZviSTXj1h^l<XEs5SpqMWKpC^omsdD2W
z!dMiRdkRDRPXW%S8DiN@{p{<D-|TqsZzlKsH=FpkkBwXKhxMxtu=$&LSE1++>weJ3
znw5Sr;Va+R=to~zQRgRi(Y23N3-mLCoxfR@Lpxhu(#EzoK4Q{*zG9wI&wl9DGFiF1
z?8@327A#)FYE5gH$E0eu_*x}9>U@*+rd6=x@)b--u7Z8vQNa$ayvZzVE1CBGDmMD<
zEp~hLZ5Dau4r>s(%TiQpS@W7Y7FJr%R&2Y^hGiSstCWYV^}!Q1=|mg*==+wPGWo)m
zxAn6LM!Yu~I0{U^j)D10C1AImG?-tM1$n<&a9E%B$om(7caRngB<sL)&d9lX&j_T3
zP2g0&1?TVBf>+@>SR3O8w>Ee~Vp#w@8W9Hhsq0~q{${8a-3qh*CBQb}ouJ11;s%C$
zL1%XwJpGUf%HG*9fAj%3^t2F4+DkxO^$0lpIswNs&cTb_SGbR=3bL-(gId5NQ0i)f
zrx)MAqvkG<Q|By#)Lsx1`T<SSzd-nFAKcnE2wE@xLC|U;oL47|`Jo~>(qIg}Wn#GA
zQUaf3PQXW_CS&LAX*lM~4BWDP7Csc5hbu2Dqd}w^F85x5bJi~8U4jMpCrlmJW-Z3b
zTYQJEY>F9RhNX$FXei9@Ng}>@NzxyuHU;5~SK$~uY9s$&jz{Bf+p#`;7x&%m!RE#k
zG!@9esd`yxpO%B2-wvRJR}udFSb}Nshw<UO<Cyj46z6uH!$-Tzv1<KQoE%w!VOwvZ
zb72hz+^fg=BO7sn=_6DuYr%%m&#`aQODr6Gi%)lT;geZk@Lj_<^xXa%Z>}ChH3<P~
z*d|1d!@|@RCqh1B#?Yz!apb2tp5&iO(xTXjRI4<Z-aMa5HfgfdU?@io!}5H8F`K%*
z6)AU;5)D?Ukdd=GUHiF^^5S@}YFLxz1Tt#4hxBIhQc}0jqrwf#NGEAI=_VS|-vA>T
zfYsFg-I$^?O{j*?b3Ub*)51^O@1Sl;63$lC7i~?sTWqL1#Fhrl?CA3hd+K{(Pwpw4
z7r4-oj?_7luJt-P-nx!Vt(+*m-icb6Gi^HTOwS}-$RNgrgx<N3y_qY;Rk%{2x*N3~
zcH=M-cd9$;PQgn&Xw-cVviA3+Q4_uBWStjP?((JrD<9ga;!9-`el$wZpNfb5ITJ9D
z{!R-bGYqC7?+`j!5K6-xVYGwquv+3GDXt@m#0{fK;?f41pcX@Ar#6wT#umDIE0#_x
zai>IIJUzL)m9FP(qgR0mBx=C@pwo7c|IZyH&pi;$$9IzU&Rrzvxtm_-B~kSBJrw<C
z55;qL`Qp=iNrn3#MsY5(t$G?={+C9V8q%pND}!D*WYTOIo=0xkM_>2wo;mKP|K9B<
z*qKfK3OV$rJcoAb^UThjTvFBL`Q)>CbcFA-CPp8i{QCz;P&%I^ZSuMAJfAjS%BReh
ze5&uxC-L9;)HeDx8~yMVle2lnjxKx2B6qxE=iat5^XIK>Sz{~9ebma1erRQ`0qrdH
zNe9#R>tq@GJJ`expIF%N7gm_Hkl$hK!AWl&2>9{7<N-HOs`UVgR8M$YU<&WXTflfV
zOK=HX0HG$@u>6-URKyxVbp2W|f8z-L2RQ58Aq-w9Zvy8-Tj2!nR5n-bgyE%0Ae);E
z#$QvQ!7LrdRAqwv;{DK<mIH4p^I&6BJ~(R^K{(%030E8jfeR;~s^=`M=_?2Nbd59R
zE8(a_4a9z`ht)Zapr+Rh=KW7$cYQlFmA{5x*WW|rGtR^r`wb*ae?xlVARO5&fDWo7
zaOcaBxIA4HOZ3KKdAAsz<U8yasuQr~)+F@i47k$zY1o6)v8z}PeFx=nz0qvM-E;AG
z#XRhPuY_6L-8gTuI*yiGi0QJ6QA%P7KL4YMi`%r(_8gz(#PJ>X@}+1csE6w1`e@~9
zfER>T;QV|;+@)fKnHN@}&_ZL3FI|JD#7uEkpc#f%nxmYE1(q+f#0z0oxIfbxpC7S7
z!PB;wSZ0Tx_S)llZwK^Kcf`UEj`%TU9e$kcgoekQ@S=<}_HJ^<+x5;UJd)2|R9(=^
zzy-q$UGT1^3vQa|f|p-9<LMk{EYfwxw#QC*+0F?M++T-%s_W1o-VudvJ7C3cd+w#O
z$H>KYcu>a{mC**vRjl!)6!&_6wZNjvwfw$kj=qLwIAPcXMNY24hnB|J_j48gNi{-~
zIV;h<d<7QhEk~!?W%zrQK3=HP#is^KvFJJ>$HwEUVuq(BwDJCCO|*Wmfq^EAQT6;n
z&Kp{Qi=x!<ZnFxOE>Pxf@cBF^t%z>AbMamFZ2aA$fYmGI(V>WQK7LO}I}=&#FPVmq
zeox{1tH~IhH3|D(Ou#8KrEr;p1QsQU<J>dj@I=iRRBaH&&|9NW?65G_ZV|!~a{-)c
zJp_*y_QO__pZssD7vg#UyL>tC-<f`Z7vAqcD)u$F?&^TuanE6_TPqw>c>)G+AA(F`
zBkYjA54~BnFn81)Nb#!z_uDt%@Z_s-*S{RDRGouoGN)ns=Hu|G<1n1mJp^qzMIhFB
z0KUlOa6i;OIANa#o{q_2XSoZk)@<jTm3TPtlo!00^WM}Co^Si=&Uu*YK;^GBEL1ax
zqSn>0ZeRrr?b3&N8+1VNCgV98O*nvy!7O+I?E9b!{NfBhBj&?l6Gh1Boy~pO3ZU0G
z0~CB^z}{&JeEu*AH1sAy(v%4hFChg+e<WbXrSY)-v=}sR9s_NMMnO%x5J<27$M)F|
zu#2PmSg+$RHcImsThRH7-TnQWMLg|ePiOYC=UIL1(74~s=hqKrwfY;A8}pS(rhj4y
zwI7)DiB7ik##^=|_7$6dxQ(^CKV@TWA2La+2IhOPj$Kf`%O;+<!&VmD=0CHVIXYFd
z(y>+S&d8hWMfnYu^5r_qO1{npGOjbX!Ru`Ql^e|cdIi(yzR7+rsABK;++qvAR5O3w
zJ1i%zhCRD?mql9Dv3R?COl8Caw#WNFRuuS{31+l1TeCOJ*z+^9)9zuZ%m1>OcLkxg
zdNj-}7l-dj6Jg-OH0V;02hnqi@K{w1<YG0TB?I8GuO1i)8A7m&F(}8I!}|bhuwCK+
zr+z!b=X##Q;aS6$?jX=GjDQOb8{qYZSXeh}8$6!21O9Qp);qmDFoN&kr&Ohb{JbnE
zJCg(Pj`<*=aS$AhOX1{yM<K}gG@N@{4*lmV!NiVp!;~MvCc8E$FMJJaw|syRyi0Xv
zWDgvl{tYTKzC*YEFNl5I5BGb9py7xB%DxlAV7HN2JzW%Ej2esUCy1lpa!J%In}{PV
zr=a6bSw07r=Wn{XxVw8kj^{hxQ4drxmH&^4ELKB$r-}}VoN?T?7#*WcFmID7)^Jbo
zcNs^F?eN01MLt;D>yHzrhGMVodK}uZ1%;ZnVTw8T+wI+r7p^4Ztq-Z#EuV=2+}&`l
zI+vf53h+zmL9|vc#pCym@cY^cytDWWE*89iA382!O2aj5tiOrNTB`Yu>@F^nxsTlz
zP3V@>jP~76@uYD(uD$vSZ8SRZ{f&>PXZaP+^?k?m<9)a&a0r761;{~s1n0Gkq~F4#
z<h*AL)lLzk@{{9fhoKbtcTOb#U6aX6Z5mBzk)?65a&&LD0?(h!rpdmFv{F!sQZo1)
zXPP>l+P{#@M`_SuA5B_)RhxeP0opiEhaMQ{(r9CS^3gFM!)Yr>^|c}0O*f+V%Bv~y
zj4{>9now_qDFv6CQR8cKs{XT<r2kq_d5<L>Y_g)>LThSrwxK8qTT(2yB@I(M>S?#5
zbME%^_?11KwR7O_YzO)ReAaWsk^7<6(OkacYWlK{rZ_v%tVd2X(%6~q);p84nG0QL
zccGkcSGq6hM&GV+M^>;qk*WtR>hqwA22T<>?L`U)ylKonA1cl9rJAFDq;bcej(iHB
z`!YdfU>nTOfgx1b9ZIJR!YQDXXOqRE2xHe%=<jGc5Veua24iUE&dpRfC6=7d#L@b)
zcnXNxN>(ek(T`~fG`cr|EbeZnq=Fsf6O~AXCOhfHoL$sByo=^N*-cwdB+<6`JydU*
zOh0Dtr2{{>*Rzt(b#|tb|C%)F8kf%f?&)+gE`wwiX40wlO!~8RA7#(XB84kiq;0gH
zV*cAt$>!NKtuC7$YUYr4eh#Jg=a7_fE}hBD<sGwJiXN3m<_q$u*es8%JoCsgIFCNR
zYG>n;+FAa&=j>eJGqz~tBer~8Jv&l+hef}yX1;6qbLLJByPkNTalaosyq_~dTiTfJ
z#&%X?(ZS~4d%-SAy=3KM-?FzKyV%_wy=>9nKP<OR2$GhLfs^4<aKTj;998GT63ZoU
za_mxwp0pC;vd!T@x&!aUdV<S;!4Uc`8kQu*L0<1R?h@Yt0+PGH#w7{VPb7o>h*W6u
zONZi?OmN5jP?wqm!hDW%<Xiz%s`G4Ka2aH49fwZ!GoUBRvw8QffM$9HG}u+co_}{C
zI_*AO;e55%b&nw+@foB#b%2n`TX3}Kf-C-CU_|D35U=Zl^ijjGZ;l`;)sMiXfur!>
zxY1a4Y%Ip=iQ{mS1m3lofLEF);g1zlu;t`591@w1OKjzEQ=U9_Kb(c-!*ft}#ys?2
ztb|{6RIp!14O<s4z-Mxca2cQFSUz8ZD^6&kVK~DW{x0w9<ma4>rTAWx=V%}4<8@C1
zob`1%>IEBO!@HHZe(frJdto)68M_8U?M(1Ot|^XeGQ*F3=GZyG0%H^`QFWFT#!s@w
zfqrY8^S}nv(rxkK3eHFVXoo8|+hgOfJ>Cy;K!FAagei__X5omlV;u2#sv~w}@I6<8
zBkpl`L_K~klKbR<UYQQ)r0jss=k2k6o;{9EwZna#w%Da&i@UsRP$<nB-=DO?dp9i6
zu!?7eFR#Uor93y3XofD%ruaw21RFT(C*Um4PkXOIy-7y6wZae&xvan`Lk4J?xeSla
z(Z{WPUZbkB6#J_QXRScw@d@0jtBsSdXyOyqCD?IrF<MD3LL0vC8`-UfTdh>F=eja(
zkXJ$z-i<2xFc&Wv&B2PJvv5#E0gJt7;(vE$V5-7&tcsIC+4iY8Uuz2Lr%7Yl`-wP(
zO~CR*Nt9_Ek1b;2Xt;bFo(UO)mMPppRyc~QQiO5EJ|SEZD}b^0h9LiNKX@Jf3E9tk
zLDuXG$o=~WNw40+_cw1r?$;}b743jiQqLhwtd-Aj9z)B$hrDOl2(5G<>YmraxPUug
z^SKH--7284@hS`~xdbN<oQDrW|6}Mp<FR0)IE-W@l89twh9oohImaH^WF`%oDlI9!
z4YINl32hD98b<aknq)+!kcPJQ)_Cs^{d~%oem&2*&-wqa>jI>&JPYX$Pr_|i?pG^6
z0ujoG;B(YI$iI>Ud;jhNY2BTm?Zn^tbGF0k$T*leZ7WPZydDbdR>5|yrEqIz2=sLZ
zfz>5nuo>YAHI2@Y{>~nr$6A9>1VU`tbSQQ)g-1F@AnGv{#un?tCRbhfz?pVaD>Y%r
zQ5_b<s=}_-oasAn0tj6g4+@2{5V(==ZVt)7n$oclpC%1z-=*MI!e|f~76Yjt!tl#N
z5TZhVv!*BCS=!q#Y#{wJ>+Jr_-Zgz;%YS}lCI`PUOZvuao_t}AS)Z6b`@j|@4Y7Y=
zuUV$S3-&YY8H?0^!XAzpV8`YAn8V@w?8J>OHY=}<$&b9n{zhJ7sxO+@TlFg}w5yRF
zyxYL^Z`ZSXZ~4B)poVS#T*ZdWtJtNdmCWEvB|9Ec#k7^G*}(W37C^NuZ&MwszE#h}
zwHn!D-mz+FyTX(#o7soASJ^qK8?2=H7PD7rV-x;%vhI&}nfbW~Y_rB=HbIEbH}`+D
z4SIq+Gb9QnB9id>ybL@(t^kMbs(_HD7EBA*hoJpNP+l+{JUIv2*U=u9v^v9!v7V4D
z=?h)0fsnR(4z!O8h5EaTxwCjVEcmh-3}&u}Cq0`W!eJZSG>GT_w%fsNdJ2fm-T`8}
zGeG587Fc=Z!q>a|pg-mi)c7BTr7_2$?I6z@pDBhK&idP=%I6#LPWbn^2fAAaVELx!
zV6^cy?0fJMiZx%ro3i&175)hVe(@RS>rZGF{R<C2380MA2$cUY5=~mg@N|m=%Dt6B
zZ!;N8xGsk$e^0>BS*rN@q9z_)Fc}5x^>D{reZI@pM~!Fts4hDNr>r-^rqL#7Ug3^f
z4|&%3cpz>qorO($!MK&Pt$uVY<(;OLxWjfG)=%AprMI|m7ZPw>R1)X*rJ{IiI{p#Z
zh5y-RV{qmk%zd{X&&|oh#}AI8`?BNsLG(0MlojHZ^%rrZLn%tBSD@3F8uXWFzyY}?
zOrLrk#r#`vR7yJr-|WI3g&vHG>_gkW$2iOPIS$=@jW%KLF-P<Z>R;k)<F$WqP+5@H
zq>dofzayz?qZmE^If}auB<ZoFG}#}Pp%7C!n$#js?u!*E<*yPg%vI$X5Or#*)+9M6
z9XfSgm!8;8A>)fvNnXZ?w3e9A$g^g&<<WFnAq3o&P1H8VicbBoriK<<I+|urzf2wJ
z$aN=5b8@DK=Uu2%(2ed{yHmt64|0w1B+WQ4>W}cIwLUZH#Uvk^`^1L?b9`wOw~SVH
z`%%CWe=>OIPdioxP(5b|j++-q9lV!?*0U(%+$<WQ6hv3kf@t_p5LK_6O|tK1lklow
z+WaM$0^{b;VuiUhr+hB;&7Vi>Wad*AzjtNr38C`kp`_=sfCNkylEvgjR5N)o&kHXh
zLARyk8XiWU_bj6mt;<PRGMvu(uOzdRt7y958ggE^mVUQH&`ZlmI(BJ2)tW?6Tlq#>
zVY`X6+BegVg<EOba5Q<iZKKnQJd6A$mUpJ&NIjEt0~aO`o1RERqqdW0KlghUB~epc
zGHLmyQ0t^rn)5D|R#l|Yp_m<%V4qHvynA(r&pOMaGH9G$CM~&_Nx!0Z(K59x@~qCH
z8D6{T>%-kN9Fk2V`m?FrCx>p-<<R>{xiq*rmsYgolIyrV<i%$nulPNrUATwJZtkJi
z{rt}Lat}G)xX)~7-)FV9_gF-I7i)QYn>D*!XH{XGD>3gf8xe4s{nTz^H?Lh|mRhZ>
zX6GGNop_He>FQyJ8XmH_hkDuI>3(L@_=E-Kzht5Kfn7WGlX>R}!Ig{Ru(NF}7;l{j
zc2YVJyv7)OqzOJ}JA$yM7igOV!92-@kh3lvK0J$rX;NDtL^u{ia^m6iOYWlKy~;g{
zQ=z+L2YAY4@XpmP2zs>}*39F#s^PsbN#r2pl^+I&kp<vdati#T&hht5F?5_NgBN<$
z(0#HV1{In>Hu*Z#jA(_r><(CLbr;SEJcOq`k0AfrGtj;E3X;3uLGAm`P$=^gCgERr
zQ7nk3X9%O#0}(6=6UWMr5_o-`6z(08M)`T;aAlJm{!&%IfHjKfa!CoVyi>t0rAc_i
zS_9_=YN5v>9V}U*i}&a0qo9WYMj9F78lF9#)^Ch2PMBi-5_7z(Y=O2d48<Z4HOE+D
zLcSG>8QNe%ku5gp+2eov9WX?|303(mC-9sz9{J4OFjL*Ic&<C<Zu7v)xt`c?*bB!W
z@J6rXnfP^y5AHMf#nx}WIOU`t`p@vk4|o0XoPPk$xDtTdRJn_Lc_7N=2cq%KKwR}G
z5U)K6L>JB%lqe14OdX!@u?obf*8!-+bHP)`a0g9_KPrCq!&VPJJh0CfMeq4wj))K5
znasUVR^E7Qh8NCp^h9|J&U9CE$CrOyQLoDd8+jLs@obHXx)XlB=YZ3b>~X;~JA8J>
z1}8^Y<4y@H-k0KYjS=FzT86SSEU>-B9A|l%A<sGBHaBBj&GP^^$q<W+4A5C~3ZBo_
zL$MLM7_(XjZ60XhR(nlcaaJ8|WG3OXjjFh?PZ>YkDq;U|MI0$U0aaJXW7RD=w9u2q
zyNNP*x=$L9O&f#Ryi;O$MFPbHM&T)AF)RrZ!6}i#_&HVxS0)Hx#NXc_^YI7F`1l2i
zq(4I8+#$$)^a>nvo`c+mr_d2O0Os+1P?7xrz8<~@Px89JVkhVKh2Mt1>=tx?y9T2Q
znxM<B5t_SdVJNr?1|F2b5x)|!Z@B==9k_qC?iAF(anLD03J<5{!Sn*|GLYN@yO!^U
zt(6%dIh+O##>vpKC;@!-bM8yoe{f<>1pF{v0T)~rK~=$ANcRqey1<!m>#`eM&T)ci
zZG3mL%o6SdGkAK~3{LaRO}2<3B&be-a}l~w_*t8?e>I_diaI!+RfW@^mEgOyA`IBd
zL--?ED0UhL*AC0Ter*}py=W{<HIjz23#7nxw*>F*h{9aXF*kbhmv!m?WT}(BGQq@8
zY}Sg8Y{9HgEG6eNvz_pjJ>313-P`_!{R{cTRK~w&H+Q~e3m?2<i*G+?c|}jzoc9CF
zEx(^B?d)YUMm}Iq!tSz<+dA3BE3GUr?gnd@zsjmNUtw~IjcoOBJ&P=>W1m;?-Ar>e
z^JY~na6%<p9$vxPbSl_&`wG@oR>AhKt7LY|tJtgLYIeAxhW!?=V-<7jSzB2H%b0kX
zH66RccCBt^>Zh(T;bk|OCFdL$C3i5-jBXZxtcOjz+RqC5p0k$JPt5qnAGY%6NKiou
zINBr)c{|2KOM((i-m4Dz7bZh=?Nqo?Ys&jv0H#?sFwfZunu^@ubd@(4hx<eSiy){C
zoCnU27r^?wrF<_J4j!#*;iyRzc=7JrY|B_k)JuSyx=GN1sjwh49kdT*LOB0S({tYg
zi>~a4kePW<G4dGLc$@&8|1LmsT{UcbTnAy&m*J-d_Xp4J1KGeQ&~|4K8awa6x3z<C
z#lIg0X1s?rQlDYYw{OsP`xms_{{w=Gg7~ds1nQ-W;F}n6l-MyE|CEoxrE24_&_tf|
zX_WEzCJoFs*TKH|lQBwK7uBZf;(|<FtTf|JVIzIC&@e_xFLPY)>wz0geepl_0F(?4
zLftL%aUf<9`im~d3yN#dwRQupzOa=ur{d6RQ6gSUP3Hc&G(6e26R*o<p?q)-R-fOC
zbH*OTfcV3BUpODX<eb2x=4Y_!`8hN^QG}->FX8K%m3YFk7SpCTqO)}~?)JOEdz`J<
zd!hrAo_3?M!2>)V-;WkAp5ULr7pQsn4L%9|fQyH}V9T+es50v>9!(dd?_Wny_aYJM
z>k*@DKM4xzlBBR%(scKc4DDSeM{!@|$t6aS(!`YM&Q4W2B&|;Kc4*QL0UgR)p-XZt
z`V_A=l~R{ZqsNDhY0h<1dj80qQU)z()qSA;aw5lgE1Cv2<kM?Q(aY@V!ea*#babNP
z-7`q*HlKCAxKM<c8=Vw)C(obm<aghLD!HFLBg~7wPV^@EE8e6xe<m&eIFsag7c1hA
z57qwXOYqT`v={sFth68L&G4sV<^JTP8$d>f0w_{8kj7;OlJMABWVm-0g=+`VwwfSP
z3Z6~<Lc!!#8cf}*=TMi~TuK$;-t#B(sPy`LvaSlDOJ$+-sdfQfZCglR-Yg;$nI&Xo
zx0Lp74kNL;Wuzv~xjb{isknS4%~M=W;}h0U#Fw=szG5A%eZf0e%eWu?1MgpL{*O9G
zZKk0^TgcornoL?_$f$H19g2&kJu~Czu39{e<~yBNbqO>vJ&{~zZzpB#Bw`<u$f_cl
zTDPW9JilKJbKg}v_dpz6w}XCar&COKI^B%kN$b=zNTGq>u>v!x{RL<1tldSzpLdb;
z$}IZTpGE7PcN3i6&F^m6bkHlCu4ZSGZ+kW=2<6Zx-5hG3kwaU8b4YY?4jsJK#VlO9
z*oCooSQ_U;-2QiqMSi-<d~>fb_PCLaec8y4?YY7(KDf$g^DP!!+|CxI^ZhQ|W9ch<
zn5of2_N=FmnW;Tt!{RTQ{@(XY`ppkkFo}0VKZyYiNW-wHA{_sz34-qp;fO47H?afk
z8}<MV-ffK!4Fv%w?$ObTfK0y2y%iJ<e!F5}Wm7!ddXorB%E`b*x4`$EpsAJtHo3bv
z8!a0=Kj*@8wjVSn<w0c1F_6(e2|CSZVU5K__?%n{D|rWZ{?8gX&7C<`Tbtp~qZ=>~
z%srnUJ7E$3UV5;g7gE?`xTf|TTIJqAliUZWQU1z%slQ+w`v(b}#XtI#Fy5Lhip~Y%
zs6Amc8f}-trLU#&tHU_7<@b*M2lBX6P7#@-5<Xp~f+ynD@Isb4n(%wcs$JT6ICe6Q
z4Aa9thbb7UFqO{}hKL15_;ij5J{WC=#---?-p2wZUoni@fH-ED@XQ7)?hvrXkv_JF
zMRqt(!~tbzI^uAq6ThX*!18y_cu~R?HI>}(h^jjl$$8))zXjdC=ZSxhdEuD(-k2vb
zlXpO8V!Npic3$+saZ0`zw80l6>wVGgyDy$r=JSfNAIh8ip)Sw$yb$xl$oszdFV`2Z
zIQZhoCq5{@+6QO8or#LGXJT)KH%f_nqn?`=mhfEAw38mFf7u;PJKXSLmn#;vx}a-~
zGnyUcxtgs`Xy@&Sn<hG-(IY#2aljVkeQeN7$Qp|ZE%C!lLZ457hC3PF*RsIeN^>l9
zF+)Gzt-mtg7<Hdd!;8xdG3K=aR)$T%*k^k9W3DdRbn9TWt2REU)x>B+?pHWG34^56
z_}xtfQ(h{equ)d{tDk^=x(XPbH6Fu%$zoB+IIL+Ji+8o8Q9E7=%N~x#byFqKWRp17
z)QaML?g8JTBaGtiLRhv~0CnH~21UW2AS?G3zMFjnh!}!h-LGKclINhK@f2Q<9DwQ(
zeNZp`0H&(ngTsbh;A_$WgSxk&XzVSpetivEOPWAxbt6bC*MVbw6(k3igZj%7C|q|D
z>VFr)_Qca*IpzdB+{f>#nuo#n>;V{<vKM-fXM=(4F8H<~9VFUPz*BWQG>6B*q%B*a
z*mMINmtGAwCWgU=kWiTMayHzq@dLdVp0H_^3z!Bt!16=ZkZg^h!!yNwn@m7PY#J2a
zp8~zlbzzg94(#8f2~o=Gd?%#}0TY#>ZIvQ?JTDK0(sG>JD+A9jjOFgPvEWiS7M%6Q
zLh9!+Q1DL@3<gHQ2U8K=4HAT1pMJ51g0JlG#82!)_j_j0`;Ogy`hlg-9b&cckr}g(
zthW6v6Sn@q{w;aK4t;pZf~=pj&C{Q<`5FUEXI>w38S{|Ys`s$=DfgHgb}`MccDAea
z7Te!_jeX2&;(w=+)pyo20l7M6nOnn7X;rh4dn(!FX%$RLx17oEEMr@O%2;Du8GHY?
zj0rT9Gbujjw6|9>`w`X5*{Oyl<<zpu4|VJ{my($cH?s7;D{O%d?^S)c&Qxt$Sch0U
zTcO*<j-I*CIz9T?4|>in9sj^q<qfmRe@1|6h&aUEmIAruvT)Tv5eg*KK=P{=xIWc~
zX+1{ppm#cq<NbT>PJ0Mh<N~!DJz+@J7eW^V!m-?7&fc33<x3WUsqr%SXuS%864ycY
zkB#szD;iQK$3cZcA`DDOhNt>z;OMy%N@I4xw%TmCB(xVIoen@iR34ZuJqjb%<U>_J
zHEijsgBpR$(B5|y9`@aWhxu)=Y+@&*3g3ZQ*abHdd*IBQUMPO?9xR7Hg8<(}Uy1z*
zrbT~1f1@DA@weK>siJ78JqmvrN@78<G+z8W4#hb0Pvn|9mSkw-s5EWd`&1i$#A>6X
zp$;ksOvW28bup=V3Ysez;eV@bFf!hYGc|p2v$_vHci?<3nNWP+z63Aa3&*v%4ukG;
zFO1$cRPu|*z3aB)jjR+j<aerF{7xmWu^UH*=c3A$eb_zu5QZN*f`XF@a9_zu)bKxx
zTSi{M$fjbvy{inJR#stlKpn2}xQw|ouW~2OO&r{I8|%+@qWjakxKrmL=5Bn1v3H+h
zgzZbrtbB_<tv>SJ`d92&`wOL1|KYB;f^@}8nCdG;NL)vp1ounO3*M=EkSb00MaEHm
zyd3!nD3If3MKbuTL<g6sl492+G5}4Qzek%KUr(m3M*38}!ho9h7}Ca4Bf8aKLS5Bn
zbh%(U??W*vwnQp_%k#%cRy@yaLv7n^Y5XHQTBGYg79o!GD&C2da%OOEwllqsb0L#Z
zSMr_aM&2LY$ojZD*?4%+=GPuHiSv<{h<nktbT2w7;Z4$s-t_*LH$|+PNlgPYNzB)W
zW?l86tLRI!OMPjKksoyz`_T^YC*d3ZG|cZ^s(%7Vv>=eo{brG(R1isY29fI#zUTQb
zn7rrAAye16bkuGhZQ^<6ET0f6Sr$rrQWwzY(uEZHauL;QE}{RHEhU?>Fe)0goWho@
zAb}g<G|qSx!O_(uIesl^WJZw8m`DoGT2EIdL{UKjcV-!HB8@AXso8%k{dyHm2e|Vy
za#So)cPyPa97pq(#gmM20xkQQK%?ptX<yuSs<2NYQK4kgtWBoENbWG7oJ#BNrP7<&
zG_ujzK{D5NkU(fUo&AzdWwATyx@ZRH{AAFXKN-{$o=LavWD*+hqE|_~Xn)r(Qc%bu
z6Q3-;C(5Ezr?cqt)hznklSN)nvq<VuJM)~~&ffC9?O=NgTQ%hdOSEZbiz6?yDyc?h
zrQ68ltN9)LbTj+N8KGm;ZZo6y4%Si9#cr71W0HG%*tLpY=4<$vy;Xa`R1d#nuWo*4
zTfzk)^Pec(IXebajwwJ&uR6#aFaT>c3$Ts013c#rtw{keyL&#Ax`u)CvDGlEaXl<-
z-^4#tVgQcC!pRHq@Tol!o{vffnK`NOvT+COv&sOmC%fSI9nN<3+{>LWoO3R86v_&Y
zL%ZS`_!e{??u8e_;m|VZcc}ttmpbTLc^Oj5u7bMCEeJW%2E(iiCcoq)xl?`66!irD
z_`U!~cnkM+KEmS(-(a2OFzo&J4+=L3;fq(oIB}LJ3N?u1X0_3{bE6bSUX#YG(c^H-
z3^{zqJ-yctO+b;#iCB4qyNueo)2L+<j;qr^hZ9=(FhK_;XX~Q0wm!<dnu5fAj~D!S
z|L2bpUOi-jTg}WcugM%0+$^w#qdPJ!5vQFdT*Y@Bf{Uzif3Xcd`DV-SS@!sEo&$=<
zIikTnC;Xc~1IHY5#t&I8Xug5>k=)$SNyZ&#+;m5Ubso5Iv?t!p@kHG*UZ}Lu3pcj&
z*+;<}8@abKGSVBBQn=TAhc^bsc;l#0Z~SWHjZUAu(2UPWBkjCUxYH9;eLV3y=Ma`?
zd*HOq?l{OZ!k)vf_+G~qJ)B+e(;R19wqyp{EOtWS*^X%A=z!aF?9pA&4t4piX5S%e
zoDgb-sw$SK+>N-Et2>SxGyK#u9jkc0#$%!xe!9f_^zO#Eeqb72Uc-B;e+_VP@)VSm
z)<?A*U2K$}j52$*(R{oX&dk!ll+o%any7|fhgERfT4ih>oQR`?6w%|B0v_QM(;tO$
zXrd&G;;Awy_e~nx=8wUhS0yn`cQigv;yarG-kUWRMa!)tQM`TxF8VHrzqJG~HTO4M
z?f3zIMtp^1<{x3%`XR7sc?Hc5&q4Ot6KFmC2y!!fp(&#WHXgbQYcJjbgGTOmzR6kH
z*Kfkv%4;z9P!nuh*9iNl4l;gLLE+hQ*ymRY!QU=|;qG(bWN-%7Ts;9L^Yb~j<S;DX
zbr1%%_rcWi9FVce0-=VTu+2OT(hHJcaC8Fa|8Ij;_5Z={!U)K(UIE9xFM^2W^I*F2
zEV!WH1Ko?<p;mMTEEl!|kJ*-RdnAKAK{FWTYy?N|7{G|bdhjb>2fQC>!WDCMh&!VS
zPc)Q4B})<1-^s%>CpnmNLk3>)U4?$v7zlYT4Q$<5=$7LBq5ESXx?2(iVn#u5pD@hn
z{>zrdd}CeLKeCvfcT9ZfEsJS*!*030VbS;Au(9)on8K_fCN1`cZ9VXc1uTETPAwc{
z*DgI_D*2CCcw#TRc&Ue-UV4u`yVT8mlDn9=MJEf}beqjCxxutVuCn#-FEefT2Bs%f
z$7ZY7Fz0nutoUgK-v^a5L)kL+P2&=?K3K|c<E89bQ7Pk73G)ytW6yt;F=yEdcGa$u
zd2Fj<Q*KqWo7%N3IIoU<a%^Cg(=RitolR`V+-t1-+)Z|9<878+(#igAxyQm6^s&r|
zgY3}7cWjXwzpHcBS4WH}&-aanHftGZ?BFb%t;%qQX~20=U1<9_70!y7!9iPuNh@t3
z&Tt0kdAdVOi#Hs3><7Xfv*7pLIp8}f6#D8G!{3w3A>{UI2-jK<$pxFhgEMhf2**Rk
zmqc**oD6Y))8N3k3<x#I0&}|@xNEf+#(mij+NTb}mLvN>SMw0OIavqR&l|yc*HtKA
zbPJ|wbDwZV2gI{Z$XIX#?6-8m_4pnTdEF1yHqW41<rO3*et}?tpYUw%Ur7BUh;Mp?
zQH0O*Q;&^8<pfD|Um=ab3NpARMg}jumO*nRCHy#`iE8^aFuqa~7cS7oP5cfVQ=*T`
z8q;ut!E`($Y=d$sGtf2Ci#s?2aZJHnG~BTetvOG9<@wckbHoNb;j;z5ImTkboCK_2
zmxQgEsVG*Sj>#`G@t}4#%17+Mm|Oc%(<Bd#Pank!1MY5UIE8`>3Ndc%MHFl+LHqsX
zD8{*eV}j~Y+5HOscDshLvu@$ewQV?d&m9!MaSykN_VUlC0kl5z49lfn;fQTRsPXqB
z3Z;DGyX0Z~edQnSVM25`SD5zw5TRG|#HsVT1Z_5xqBr@{Br$FrU5J&VtatLf|D;HD
zwMt~ItV%KKCy_$42Av<JO`jYlQ+|XV<?u{S3HOfwDVRpPbBy_SvMC+0HK)Ph=_HfS
zsMj26BH!m+)3Bl+Tde8jWgB|;)s|w$+mpVQ1LdhZ(*7|{^!=?99jTr{nOmHxV;X1a
z^tjN`C|8OY?MAZu+^AT^oi^sV)9wi#q`%997K?b&f=!-;?>uS5LN99U@uCgB-Xz!I
zO^LoU>Bz&GG<CTTz53@vQHOjf+}@9zKl#yyGydcr9zdfl0=fHg7QGspMcVg+XjjW@
zlDZX4QFrE0?%-T96r4{#v_dG~H<bEP7Le=pg_J3^m^|k!A<d$tbaC`DN?*lYT6b5_
zY5SElv1}FD8?52ZPrlQcypFD&iKNZbHqg0>DC)QSk5X@MBHiFEH09-1`WG2P3EQ?&
zmt!m`kBTGl);O};8BgIJ31l!Tk#w38>G!7X<Tf>lO8b)NZgMhrFQia&dkS@ir;_4`
zG_u>BMjw@S(Cx!JX#2!;x{;AiF2B>MA!H}Ltl3HN(is#KoI$3CGU$1C2Dy&pbB}%|
z9i5R$i)UxjmW7%0SGtuYq_?otkXvkgzzz0rDeqSLUSW?Vx#{v#J*!r4V5J8dnYV2d
zJ9YaS^DVo@=3*Pmx9((`!rg4ny?e~r{~;6q_K3y59b{YmhS+P*Z%izK_sQ>xz}x9k
z@ay(?us@*&!+!d3&Db2)4%<M*DmU(N@P{?La}~OM3GC%;^P{gL;BQzIe0JOn&3~gI
zdvz>~*cK1!7ZYLEa1w0hd7QpSJ78-@1{~!5%I`M0aItVdZ2a#q_~~$O&*oF`F}@ID
z0=UykpcG1T%0X#XHGf0b!=9uo;Lmdr1+Fb{vb`N7mUqKMS<cXJ=?90*r?6n|OHei%
zf*WF=;2dY43sernKSu%lc~%HDBuAoam?+M#5yvR8(b(=Lh24qLIH61ieS2kb-jDHU
zFlqwc7&{SNd1vRYkSgwdtA=$g>bN<d^N!bOW0cusl=`TPFY@(qlBWUQem@oe@SgLz
z3C4K$s0l8aWQJCI&CyK60w=8C8QZIX7ZeEFLM*W=&kDzPT4MtDS{;$MLt$Nebe`&f
zLOPCkUB(GbKRRJwBhT(8IO9SK7fkMR!A<M<9P-l@kF0S+uX}EI)4&~{t#juNUw4eT
z?T-CV-0=YC?$r0VV`z;#9^1*Y%--%;CFG9c$KCMxbT=%hcEzL9T+wci3x0X;jFZtB
zXa6??N1o-oq;^N_eC>e$e%fR8UpvhBX^VnyY;f-#YwRhpLWdoe7%-PGSqIVc4MX|U
z7U&W>9bIJ2am{5@&R#a*pJ&D>R5A^_ybaOxjR6M6O~LT-`gr7|E-IQ&#($047~!Uc
zdpb2RGf*A3Jy65+U{&mUpp1KF@ts+hB0l%xeCZqVm|!;^p-L9Jrj0{=&LaLhK^otr
z@mX|uH2z&Gfp^-)ak7~hs%4Ad-M7NH#A*bJCJW-MtbY(W;x`;#^aHdizd)zRM|kEx
z#MzUt;I`^>NG#;Oh1rilMXwk3j_-ld6YheA?j7!(Y6nxlR#>p`CWI`z29}GOK*qNb
zip}a^!{}<5biW+B_LagY-y+z~J=@bu&cLs^C&5p&05mQhft9lk!Q6lQV10frys+C1
zLflPqYUK_n{GJR`HYY;uuUKgRvIVmLu7_de)gU!548(7Qf^}FhxH<TP!D8;x>~MjH
z8GILd)Eat45U$mj!$6%ej1n{C*$#b}GkY?e3e<x8aq1A!p$c2fmEmAM-$#sAfOV_n
zz~!C{++QILv8$vwi%^p9Go>N3^(Tw?Z-}weZ<*)sH*C_%H!Mo*A5(kxi%oF)$krbi
zVk?)vVfq8F*oOF*Oj>D>tr&jAep?N(gBC+9^5aYP<@ICMp8J$t|Mr-bSwCWZ(!Ffb
z{rk+~OE(*N><(M6*U5s_I@!C;?M(V=3tJL&jmaoAF%9!Z=9$g+F06)))U0BwTq~I3
zkup}Gc8OU$FJV_lm9WWsidjNZF?-xm%(BBu*h(yA>3q&<^(|xn#gwzV<rS>zPbE7M
zSk0<Bc^)~gjva|^U?2M~GlgQ##~E>hrQB&@e>FPT;Do!Zd}1#vJn)qL&UwoUBEPZd
zC;`}<Hxhbpi~_MS((q=D9KahzP~5D>*+SYNEi?rVOg4snc@~haWDQ3Yc_-h_75?t^
zgr`e6XHg^on(Km~{nlI<ZMgu#gqMPmR5)C4T??x(Y=DvGTYwF1gZnM<+(EV-?iZ#&
z)`=Y;pO*onwsSv~eGZ)4vj@K4*b6nTdthE<4tV$SyXE<8F#4GblNaxYcYS$qU{L{l
zjHrgT=xdNC-wxlL?!rm2UeNb_4BF?PgTT3W&_DGHxHbF))dl}RB|!)``f|3htQf|g
zABE=UM&mw`#)W<|__Rd^1NrXrQ@#>@GE>8q+~F`aNE21LZ?7&?52uS8;+UOg7!_cN
zeRU3~<m-;Bu5o9a<7^!35Q06EmZ0K<aC~nZfvf#D;_tQ5=;|1UyL}V!<kDmeOi075
zvpcb%e-|!O$-#9i_M*zw19)`mVU$094C8cppQ`#aJ`X;Jx#C55zNr+y?WjPr1vO~u
z(11<GO;~Mk9siiNV3cz^YKC>8NcMfyYw5*2>BqP-bPx~Jy+W1A@37<0CyZ46jz<cI
zae=x3?OZ2B0XKwcl7=W*Y!j!ek0hwqMvB%Plctan<0xu@9IdUBr@wNFbbYB3Z7f#d
z{V6rFHPE1z*;=HxMTa0$mp<j_)BGI<bSlD-Mmuw-gM<my)|!&qTywhlYC2^vWVGQX
zP?8GKpZS)gmtjTGh1L{bVMDzow&ZZwj*>UqQ>u#t;V3@u)H%|HrB0L~GJ|{%&L9On
zXA&%Orm4m*beU&(dUaeWH_w&CWZlRv#f`lFxsgGXJ9T|@C!;kUr2g51Hf{E#VZQeX
zI_yQ}<V|Wk&z!w;CjEBsp|iri^t9ENI*#~J=2m~wTNFUKeu31$^UQHRL3D2ZY!clN
zOzt^z$fR~IEgzakd3@Gczc`e37cQXvzZTLIuf_D@^b!gi6-MrBmXX23<y7DrPBZFO
z(kZjm6m)S7JvWG;d*{~CdBgSews-@HPv1zJs{f;N)|<I|YYYA2Zu7fMG4$uiHtL%j
zOP3VlNUtM~o@|e&Z7hLKy-c9ryAw$nx6}E)?KE?95}h5FOd%(dX}n1aZLCZoBbG|<
zic)ETMjF|tr;*$HG%|PDK{vVI{PxowG=^uELxR)kU`je2<vlI`o9T4CH=W`K)5&8l
z&)p4NXYDQ5n78;<24}CZ4V*RcU}Zgf$r%&QeRXV?T>~>Myv&w8Y-T${Z!l<RVMp(^
zvD+P;?AW(%wyCm*E!f=8ZmxaC0&3r|`&+-T5j}rd_?D3noje*iTMEWHslWpM{&D8H
z#}O&kVCm)xnofS4r!^M>{w;!5?sVAWyarO#B4N&=D0q8u6Rhxw1_9kznCcx5n{yMP
z{d*Fu-k1uaQ_^9!Z6+j)&W7Qxz3}SWA&{&&25(=T0IiBMFf`*le9tcei~FTu99#jv
zr`N#inGG;=cN6UUa2<{>Y=tW?JAk&{g>xzo;pWXpU={NWj<Qz}^?e9T%07YRy6^B;
z|2KSJC4k;ngm7r=NW2v!iW2+9@oR?!eiM+wquSDV)=mar`pKfx-0|2PtbiN66>&52
z&j}?J6#lA;{Z}U8nG_AoaMHrkU)pGKVltX=e?a9^ee7IsfLs4f#gc8)aNbWN3|(e|
zIqjx+W2!k`j-QU!Z5H@d8ZgHh@fwe`8|}5kbr-B~O0_k9tFqy5CR^?cx5JKA_V@!F
z(C()L2Ap=pK36A9>T}|InHi|rJp=bM{tI?GV}B=~e?(odOy31ttzGbhy$gCy=S~Dg
z7hJ;cUXzNQ(PoJ=3P?EPnIkhWLuUqxA93RBE+=$d>WDAP9PkF8XO{3?5(eAh%}utr
zJA=DK_gmw${Z_ax%Mx8;2?v)VHah_>P+@5FemWlGH>gR`W+-cEiap<q@#a|`yqPnN
z&x(fFe39>K{HNe<?u&}e=C^;t$ry4?8;{S?LjU&~m=mLpNs^P$bDt_oYN+6gGfF6E
zG!bJiO~CCIoXlG_9!~+^qg9SW2@4q<DwW1nJZIy5MiR}{M`PihQK&IW9KUT8#nn$o
z;!S5^v^gb&;{*lK?f73XVZ$I$_Z_s%zCiEM53o#p2>wO9f{`x<p=sq4=#hK`Pj5bi
z{(}3kY;QMI9PWfCC2b&cy9GjC-T<q=S7Gz$CQu&R2+9(5F!@&%MD&-#{)$pKm{J5y
zKIh?p+*xS7a}qx6EC6rkqdfnW2a*;0;lav1@QOR8?zLq?=zr;ORVfuFR&9q&pLo#i
zh=II+8{r^}fXOK<!1c!>*pxXBB4^J6+j%}PE#Dn>@ZG^rQ9HOd)e^)CEC3dp!d~8&
z+EzUUhMaU^sf0G{64Ky2Wi`0BR2e=uC<4<|;LbKVC>fH0!uiq=@=Fq=UP^$&;Zd-C
z*)V&}cQZ#LhM3aPw`}XTH|%cU8+P6B1xq^phV8I@!!{VaVs_mxSpU`MtasNSi|~BP
z?93mt*3td!P;?&?j(Wtrga?@6(g9{&+0WR(L*~DY&o+wPY>a#-J5t)l>^*L?=fiF6
z;H2v;hTpxzTbo$42G9Dfsb`J1YgqcMDmF>6f}K?;W0(C(nd+Tl_VIWTJJENMeT}}z
z_9k3pZci_=TWLkiKfIU?MwBqw%u?1_d5QJEEn_>TR4|9QN;ct770bWKXPn|X<|yCD
z4!pa<{@7k){^M`4>+9Rt_xLVWn)-lU?0C#@=4+;S`ZIf-{D-O4i~xmtF(|n&2@w+G
zAlzR8R+KBl5NSZ=2wjK(Looem27-eGCk*XjLWnac$GC&kcP|jL@P+D}0O(yX8?K(2
z2Nr${VVOf1_=c^7w<QtKthy0Co!APMd_NtsIROl2C&5#e3VCDGA+>|w--ENb(=QvE
zVsc>AvuxPkzZ;Ho-f^Z$Htdbhh3(wQ*BF)u=YQqHlaxY`e%S=?99!U-+#MKAJuvxL
zKeWsC!-<ivAV1(eoKyV@!PkGmuc-pK;H?l=>>P<|??q87NgVSk#L#I#0yCxgTdh?d
z$J`r__pdAAa^69dA$4rl*TTNLlW|<80ZN?XH}%Va4_?~hZ!;Hk4EM&fe**Ba;9MN$
z%!#@!VR$fP748a(#7DC>;dukz7dMW_hqN8@y;3kZd<WWQXQ1izEW9n2i|YgS;j*HG
zczePT&PUG25#vr`Y5p0^vptWwZ;LRz;1UjnRpJX)i-z)zm?zMT6TV%?wC^qWMz{kl
zRJw7UeGlgIyz%v-$JqO85U+Z@#-NMu&`s$xPTl<-m&^XfgI)sUwNHpDp9_=LbWysR
zB2Hy}67<DLidG~_)0ew4G)qg40$0n^*^&t)`Eepm*H)pSnQHVaLY+U*H0fZbHYuk~
zrn?*T=&R2ZYEqd>$9oJ(G2V#k6-{X75mP!f&YWr@r&InF3)(3F)T)nk(~(Hr%aXL6
zt*FDunmR|@(7zrVve|D-DV}!p;5!#G@3E&U0|#1J<3M2!j&${^BOP>bB7u4*5}7`O
zD$mUzbxmioJK#)XWnCyH!-ZB!xl+Y0SJF{*qfMvWxVy@o!h77wbgKs$=z3CZpC_>c
zUNjKuO^=OcQlNwndA;+Y%MW}huicNXwDG&v{Q#QsDv*8)1(AW)Y>M;>rs&u?^u1~>
zB@NG`PP-7=ni)!ypDmzumW$~3fyEU1ZwbNDFjBj-j81Zn@`6j@)UUUS3XZKNQd&z|
zdm`weOeA&hTu+Ikq9`D7BNhGqkG`+pOi8b|P*rd=4PK9-b9=Y(ds!@<{}D@#`EfMX
zEuQS&#nbSP1RBszq`lRNbklV^sda59$G{|dc_)dgT#|V%IhkryQz$ehg_Q55P>FUb
zy<eG1d4;Kz`y`c$#-!10lQin~OrzqEGzwalM&*~A*?#Y4HZHh{?R<Edjr!EUzKGVd
zjzhI<t#B>t38`g^%j;OrhXxjS?+VM3yT-D8Z?c&6t?c8TcDCdD9hUI+9xLO#tZ`*e
znAxLOEN9*)*6#b8?fp0c(vFM*<>fMb-#QVd9n%J*5M$6aw1O`lF0eh!7xKO4KvwNS
zxRk+ruYD_F&zCivf4&aJ`fPwLCI7*<?5&(>vkhiViHCWqiEvmn86wJ3Ve{2=$Sl|e
z@^5qDWYR&9TYVJ%TYemVem?~o_Jxq|eF5}F6+@N#C3rQu67F-K%NVCdh{$RNv(Gmm
zbJ1-W<Zj+StM2hkMKAn3%sVl*gYawc6-cJN1NW()!L9W>6fXV^xtj#Ar$z|R{1C?J
z2BO#+B#w`^O5ne}lK7Z20R2kFqHXawJbF?Nx9yh4PwOY(c&~{l$vIlT2USq4Kn?wZ
z)KNf06LrsM;WI}awC<UVc8m1z?Q?yc7HWWhI;Wxu@qNxg{@wq@7?0bTVquaQieE9u
zm0zY~@<fIL7J!>v5chj=PmGHtI$3acfuc2M4_ae+nGN==vBfdU-04|uhf%KfIR2JB
z7LWtJ$#cM}kNl^pBZh`KqUa7s>^kg-H}W0v{T@f0vxPJEc*ZAlJm)I6IN-f?4ro2v
z0U_5ORb=eZZ5#IrbN2FhGh58qVuK<T*7)g#6}FGILM0u(r<qRphu_sQ5HM{T!xcO`
zeOhQbR&$TS*a}mekYa-5XN(Ug7@_|ILln%Niqm-RdWygl&Ux0uBm91s$@zS?C$({q
zccM<b(7+4v>ZqYS33W?V@rknvu6d}0-&Rb-++P!T2VDU#%E;r2gK~ITQx<QX<qSrn
zv8Yfs23?pG3RaIsA(X%$72>$wR1AL=is1W6Be5uJ1m^q|#P^E?a7OAMSZDbQK0o=!
z_a)pVCHet!(%!-`iC5fPIS6KoPoS@?9}X{j2x(^bp+d16BF1;ZXN@*kVc7z|=G=hg
zTdzXa{wuJmr~yP9YQeUd-=rGKVP0`5bnP#KLC%~k@jVOK+Na>hmja-Mqp&3<59ZB0
z0Keq-LSJt-??vr`j=9`RB%cOjTa#eR)&y9k7Ykv9o550SJ*2Ez1>WzLLP=^Utnv(o
z0$YFX;Pis*J1%fM+5wc;T7x=wCzbKN=+HkSSUYwqNG#NYPw%ziYOw~m9#;e524#@`
zr3mpZ3Q$}i2S0?z!Kvla5IG<TS^r6Zmc2McN{GTeuVGew=^e8V8Djps-m<C>Z`eAm
zH_X)SIjdjuf;k^}#+>q>v2&)+SW59z=F<O!1^Mv)(ZNSdA-#`f3H7m<$9=3yrk_=x
z>|?{l51H(K-d*_6#U49%Fe~nZ`f-TQ7F{=3IDb#=*?E<{Io!++^ZU`TK?8F;!9Q;V
zs@W!<A<hymWB9Oy4gW4;ykEoId@eB8gXdY6?|F7=&3Sg{`FZA7aDj>BUStBNikRfh
zVz&8H3CmQy#4H2L*ue2}R;*OXn%k<_fjc#fr=Qr$ahF-%QhsN*y1{Oo;5p;EPIk8X
zJ{!LIh~+(f!M<Gnz)C0mVrTCO!r%oFNIoP1IY*_zpjr-$hZNzF6Q2{`Yr%_YQ{a?~
z2?VwRjM-w#nM6(?pXdr6nI2HP&>Lbo*I4dJ0K6R;45xGFL;spZ5EZ=)Jd3$gR5%g@
zR{sb4hoWJYWgNJ1r+UHdBuGq2g<sm~{5_b#{h7O9+0HCb&&Yxk7Fpn9mj$95cf;EI
zIq-qIJ1*27f=%=CA^Y$d_)qdOZ2Wc|-sCpHmcpBGb8IKraSwdXy{GUf>oqJ$_y8k~
zzd>dNXE~(`;7HjKcwy#9H0KLHK_yX)9~4CuAqm|5a14q*mPK8@#~GGY#$ldqTkfcV
z*A#T{K)XKP+HZvK4J~lwC~Fi~a>7T(9=N^37e#Idp_bo#)a_Y>E~(4;yJQVo^1k>Z
zy)8H}CKk6!C!pAbBox$7#jF|Wn7cX?&*kq%`)9d$%xFJ;Njrq@pO4__r3Lus<w?|w
zI*X%aF5rKa#pt-QjC=m7(Di36PVZ>Mccsl3TyO(dA8tiWo;Ti5(T$;Zd$9jsA3ijF
zg68X=V@1Pjl#+jsO;MjYwB|cr^#6_7VFFZlT8K^z2~)C(D8+3Or^AgB)GaDSquixw
ze7X$pTg%cD0eLbpnn3YE6Y1ziWeQ7I<s7+5l)qbp&oWx%v`B|S%yp^khaUG(PN816
zsnq$<kbGwwk;OG*+N5quQ^L(Cy1<;OZ%rqcCl*xjl2Pa*ApL7d^+$-_gj>?iDOR*=
zz=~d{Skqoz8#>%zL*m}H^z^<h;T$_sd1y!b1MI1+&7S_7;Xng*4m4=)NKqFZDSC<%
z-8<_<zJ@cXyqwQJj?OgV0cR?&ccBV-SJJrbO0AJ@BtFfZ#(!}q%|;LM-s4HfH+Yf8
zd~d4oo=MJbe81!AOXp_$k?d-JI-C|jBQ6Hgmxr@x(wNy~;u1_<DRby_`&<fCo=-8W
zL#Xj;DAlMhq|xz<$mPXilJMe8!OAdNqO+VP?pZ;pVk>F-=2dj$-D<L4u$Gd#Bk0wP
zNNTHCPkA~~RK=YTF~9$#8KIkLS>qP6P>LpzwK247{x*6q6ifEUV#(Apj>7N6(e6d@
z<jeEPvo<8q%x?)av^tR<KS`vgp4+LgWIGK@a~IZ<Br+>a;;ybFx^JFLTUR90mOaT-
zUX@I8-N_X9JefYdOD1#9{mks)85`k7_Vsc-vtxB^Wn~T9biA51*;TVurPb_+KrK^~
zt!L4f8`<)kP3-sEYwYN_TioYzn>~Km!OA~(vw{Z?SWVpkds+X2Y5#c7PHp?igps?)
z<Hg`t+Za$SQh=Wdny_)*G+6k9xLd&)CN}$k#_?eIdvgKkj1Pn03oGE()0MDr$7;B9
zcP;!Jjs)4O8^QAT78tJC2HNuRa49_z<PDQyp;j84__z~}e9nf2mIokq<`FpiHy^|u
zPQn(KGw@UT9HgmSfS+2$5Z84Hl-ety{BtctOuGz6Vy?oazMC-9xecOf?|?P8I+oV=
z!Y${=5ZW~eb|J4p|MffQUHutU-hYQqKH~^16TqRpLin;#7@eMq;MU(_IEgcFpUFyM
z*0?b|e>)aEzRIBAeOa7VJRXPR6tK-z5pRhr;n)UctX-*!Z+X|J?1DNPIB4SBb}e-C
z)4?xolhK`bGp*0-;{|C0{BQYG46imszY#|G$<`Q0M46!FF;n!qVumkzcouhXIyyeL
zKz<U$*e1XgM-eO55=xm_qCEFNY};*xzYMGqtN7ewVS`d9ZE&inEt<@-#UJ}@aY2hM
z&V6r-kA>{;y@(xV4%=eSfGxf%v&9!%ZE>WzEk-}G!R$@^kEjiHZnwshKdf;5Tq`WN
zWQj)vE%5>WZ`jQ9&<~CP?rLDT_^t(ZJeiJO&&@IDu^BGyGR3#`COD_S7`5Vz@M6$3
zT&Qn|hrb(OZQT?c7t4F7R(g2s$7BpUuY>U++Bjtlze81N;8*Uxh!C5E)6c5nGB*`8
zc%_6V5+>qTWks~Sq=1K9<#Bqy9O|u+<=*0PxHDrc>MKiQ<Y_5fVk*h|SQ7ZsaTHGA
zOdDfwQS9g#$vM!%yt^udkL?7pm;1(Y9Dl<H*`L7fe+8S2Pp}W(!{Ea=FlNI`SSmRP
zD^5NJaP5c6;Rle}z&)7<y1;Hn2aMl!8(tj01-F~8L(Q9J&ZxZt$`%dqBd``~*HnR1
zY&l4#l!8=p5uA!X4_Uki-sFD@!YqzMHD~y}{+I`^Zyo@zV|xKM=D<AnEKpF%fYWc&
zAhJ9e9HSE<%{C5XM@B>6_9%#vS_|{@mqWhCBEAEf2mT{w!E6;D$X@9V>ffC}rNI_h
z3&CP#3#cqHf&J-*AbUz5T>o(|<W5bHoH+>sY*Zk^e<D~UE5JaT9Jp$Z1E+1$aQc}f
zteY(X;_t;^^F9$MxH<yjvwyG!hj`y<`)gLX{ViMf@(nu?_L!-UzRL!k?z8!q`dNqO
z01M9@U|Qz}Sm?|F);IeR>sRb&el@);0eV^1`d+q}|MakU$WA`H&sMMQX4xSfENp%&
zGyZdfy;*dPO~q!Wd#Hh3k!)uBe45yaclB(9XC1rzxSDM=tYmkh%9zO061FL$h+WFQ
zz|ISwXSJ^inXpqKtCB5bwQhyXvbT_ZFFwZ(Ri0<64=%75LPczrUNQ6aE@63FOIcsz
zCDu8uoJBvXV5N_%*kb2eR-)Fx0wS+48~1B0mEYM*THD#V^1JL;W*^h_9%KPcLoB}d
z8xtJ($95kS2I+a?aK=Om!j;BB1?L%S+bP3vojNG~o(%1A)8M}W3-Fh*0Sh-rsJ!S5
zb4}bpGt~n&{P2PfdtZoh4}=-Fg5mh?5Ll4D7=D~t4o1&bgPqBG_?5K@T>ivBUuhf^
zFHVGhkz`0rN`(!#cEE*#oiOQc286k0!i#O0@SStq7sqFT=C5oB3FrBo-h<G$@EAN*
zI?Y+|Mc`*v1t((~;qvWkutvBI$|l@{2JU&?`(Y5iet84ek9`EuiQhqYF7F0%HjZAm
z5YFBxj2BjpM4=uL)W0K+DzZ`-eL@CDj^XdKy^8p-TNxkOOu~;-wK&&87ypJB;^|#x
z_(hdaNx&YH$GW0tvp3ck@owguVB9n%6!)H6f=zS6aabdQzXhW(>;6_e^duIK4J2Ux
zizF=lk&0(!ccLEc!egtmap8qMczAd}uJX&n`;|xW4rd#?o;ii8W`#JV<pR2`E<rz)
za#ZiB!mPt}n7Z~de)7DEHPdh61O3}*uiuGp&F*58+XE~L@5iLQPf)b&Io==h2BT-c
z$JysUWB#ZgxM<aHv<w%Z>-j=7u1}a|$cxg3nc{RTRf1|7CF%6nG32NvLrc75>2bt(
z%Gjwuo<|i)_>2;rJgGu#pBi;+Ri{0En$)7CP1m}0=sR~;J(t%fc4P`&l%Go4(T0?D
zXBvqp7*mm(3H7WprT$nm;t~*w`j0;V=UWg?XGFYXF53unEE4ICGSR_Wq91;i^!%A6
ztypD6sh_N9Muat;d}mE(!feRxu?@ur*%Ecy(h_$&>bq`7QO@=x&|*(!ehzevceZA)
zccfUJ@2S7&L`gw2hzIaU|B5r^Cb&?crz@pSawDrhZgk_ZJ1Mt$kXo}Rox9>iIXAuO
zXwOW_`rtzvWBf?MjAxZY1L$Z@Af0NPMOA#(+2S8eyN}GFqaWsyBhMzQ=7rGpU!l}9
zdm+83T|@~QOUNp1DINL}MmqDC)7+aYsERwW%8#s~=22_NcimdL*%LvTylXC+x1Q2}
zZJ;ea8_9sXKEHk3L<SaHNGob9=l@01ncf)sz%x6!QnB<oH<l_U@L4A}jw~eOY0lPo
zzJuee9hU@>EJ&b~?+N5!pGXo(iFEODB8~c%NMF?WeT#d~#pm)~`l{_@zhOHCw%4&&
zQFZKJV=Zf~s9`P9)$F@}75kc6$((0ZGTHu0mZMS4j!vj$`wrH#QqGmo+1||7p1aOE
zc-QJnK^tRt?yxEK_nGjneipEJkQL1wVpGGuv14-uV5OM|Tz)+olFZ~_%Hm1zbjws&
z`vqayw;5o+$On#l1;dU1IXdsCp8xj`w?tGL+I#PzeUGyDii~U$5sE^YC8O*;N{UD+
zO_fnmLP$lF6iT*`(MKWqJ-@#{I>$Nw(J8O@{kreR<GP&v@NnBAynA~Ie*O}O+t)0`
zGmgtq*=z;AYzf7=>fxw#U;}0*MB=wTTX4<sZ5TK=2A#@9kM`q1Jh(m%`%9d}GkZ?p
zR>8tKH7gbWcBNs|pLD$6Bv@*Db5VM00cOTu#m;-z5ah~m-Q;pCN~^$ey&vGFm5*`H
zi+W6)@Ei+oyu>H2Z}I85kC>|3hED5#;D-7ROttO8ttdf?zEU(kNJi{qW$C}DUUVk1
zHyv0lPl@vtsc5JYz0+4A>#wSG<gywK4A&4{1TDJuT$^rf(WMY0eR^6T9GydrNb#;Q
zy&Gmq2d|ouq~Pxw?y{suZ>^{uY{b3VmdbD0QSWYhdSUBGU4kj9vD``QjGXD;78f#!
zaHVfSZlvPlPBu0k^zNqzwdDh~1t7)B6D6G?DifHh!W5>7{XEGh#FLJcdQ!Wj7fo^Y
zqLe9K)Uw!%WS4u9zMmI)jr5|e241B0-jj@zJn7XqPl{`0;WuFF*G&`=O62eiDO&80
z+G2p}Dm^Hzr*Me6xzmT4ZZt5=m1gV^`=*1=lzGUBYIgUhotqqKi0DBDj<ctQHg<GW
z+LmV3S(D;fE0S1lNs;|6=*c%T8j)!#oSi1LNY|L6stoDiMguCd(kIVHx@5jihm!QQ
zskTIun&xTH%Wl!jkLyQcT!q`cQki5HD$&W`igbUk0v$7wr(e0fscEnrnLX@DM`p{?
z+jlZl6f8{_+9heg8VP#y?l10a_>F7JI`IAJcJy8S4J$mq;LkrFaedA^)SUksGZkNA
z@9WQTMOXti^AjAa`WV}~9^mg^m3X071ztD1g9_triACBC<nw|dSyzNj|E^$9lS|m?
zm4~__Z!=?VCjMNIjxqC7@r`dXDvn7)4M-Fmm3T~+kHzR8NAO|&L7Y;w56_<5jl5xp
z=oN2AOUEtPM>Y~?=C4JV7m5Q^mZNFSA{>`84=2UWz}T$GsP}EW$oh=JV~RuZZo3zn
zYYW|Xq$@TG*3C~vJ2Y{&!kYbNc+tcdOW*2?+^`P98%<oKqK@@Gs<<^t3Dw0-DsW~W
zk$dfhujk0(DoGh^k(b1Z<05~u@I7d+XaN6(55Pz5K7^mCg6(gs;CIGj@Z0|s;;R~A
zbI<26v-mk&e)SwStauJNNzdSPaU)zD-3S*<8lhl&1KjKL6gG{ogOz?YaJaG(d_}g;
z|I#g(tW*Y5J4+zn=Nc?nPz?Liis7@&H8^unJbSFJLfxE8;;x+sKl5^+#3>7;H8UW5
zV><W+roqdn=iz7RdAOpM1|c`oVBO_(m{NZMZuZUuW6pv&zicRv$bpQ+Tqvr&2=_hm
zVQ5nUoP2f}0tXgCn7YXJ1eC&K!Nl2;d>2;cRYCl=$B<#&2>Nj?5SsZB?z;bggT1<7
z=r+N((C&$5H~L`AS|zj?qK5gROQjuXfPUM|aQ+@UbnowuseK0E)B!`WKz0ObJ|Bsg
zIT|I!{As;;0>0SbgGS${p?StE+<9a^1|%%P57(FCRI#HT@3sb)N36%d1(BHgFA7J9
z-E`oOXxwSM6IZqG!Vg3C;0fiuxbgB{w94FvcC!zl;-^FSc>Xb5bL#}gTAju{=TcBE
zC=1V+UBbg~4TEmo#JIZ^IBD`DjHzrupOr7sa`ii$ulE@@M}Nn!c^!CTUl(Gi1eyJh
z(VZ$yRqK0D*@9lAIY6FLLzL*po__SpT9Y;$(jjFN1KMP0LOF#N<S1)LTU$g|+#CdR
zU9iTbM$n1oarEel4;_#2rHqi-WOT=m*7z=_A+kX<=VmDV-n*7|E{dRS=b~tO(l)x0
z5<~kg?V`Cgd&T^7KyY-9(Cfv=$@=06N|H;Y(aVxZsWF*eOgT@7o?Re?;B2x|$fL4?
zOH{P9kUDLy)4Yxw^s@3c1!mo&&nK#A>Y*CicchL&lN#v3l@~Oxv4w8PzN0W|rDx%7
zWOeNaEfq7))>&P2WV!^0ZkJ+(JQ?2fQkEYp$g!M9AMQI>fvY$7W&1>BmM>6c?^|m8
zrb2_S-O=KPt2#XDv>scpH{k3MMr<f+!j1W+Jb9`)>%6nz`T#2)TW`%i9=7~8+K$!A
z?Zs!U1NT<z&nrxwINQ{jz13a#(=Qixdf>`Uv2I*4&7CuPdvMG}5BBr{F8T=Ey$<=0
z3~|;DV)fq4R?*BaB|X_`y(jl-_vG=*z4-MzFFxZxfPG&N;JX0>IkRmb=WH0nA$<q)
zrqsbaRm?q0+J>-3!cfkdJd9VU4(FJL;jErEf~}%P@=d={+&E@5`+AOHP4}_<)ngoA
z7(AYpCksB$GH<TfJ(2fboW%9bK0K}86fPJyl@oSP<AMjid|GP;O9jm2@XKP>QJBN(
zi{^4r={)u{@MGnO1svAw&mDsovC6ps9woPg)s_WvYvoe*vR}^Yb_cQC`(U0pa3x<n
z9K!ovhw>1S*Rfo+hU0R>`0Iyo-m0~Zvj(o`ia8tDHFP77*|>>UY=~ePk4P>_k7Q%5
z&HOH6GgmZi=KUT~+_fW$>&33;huRk5w%EeXQCnE?!WKSQxrNV(IS7R>YtX+fye6~|
z>@5pHW&c%}dHD*slwO8~&7z0ta|vdSz65P8mmtOT3h2rdif3Li6o|a>a_ushHsCf4
zI(Zjf`c%S7!92eaRtLZOKZAL-Ef5^?5u#Opz|Cu2aP68jj=wI4?eCPZWS>4d1=*sx
z7vaWBBk=L)Nw{s!Of>1~hmTAbp|$g3JS#e|cMJk>%DBZi*m5ae{Sbs~5{hjJVfb{&
z23)>15`#OUuwK0D4ik)!o8x!luEB?~*T2Ikw<lK6rA}ZmC*VM>(|BxlDlQ63M@7oQ
zo&7J0&ch|#=vRod_FqSfyJfhb&mGK~avu}VJ;1YFk5OsrQ`E?Lj*t4Z;1vJ2m~#0e
zHpsT2>bM_xXh#Q@T>XO%P5<!GZ%I1XElq*HdyvAHo;3KW9K9|S{-PraWEaqv9yuse
zW19;3ru3tEQ`M>GUkzG#LW_2b+{oVRx)e4{pT?INP(LuD8|RE^th_0mTw+GC*UU+~
zk0o`Cx1v8gtpA^9clfp7^-9~*6X7Ut5<P}xrv1rA+llVTIMbI`&NT9p3vChEmNDQ)
zKEii3HpZR4>3GoJ(;nok4K#bJaBepPRSAS`#WEy|lSnn!kjfq+b=4xpR3NR*Mf$W8
zspk}=PLcQdc@1d5Tp;c59#j|RLASoT(~#NjlwIsbD^*0NeWoiNiF2XHf>opQ%ZXlS
zInloU{pr*|(F+^lK&iv+X@i&Wc{|wBb1fTk?Y5#v&n!u~z=B5ZG^g_EqE}{WN)@ff
z^fc3mh6EYX0BZx1dZkB2$8{-oj1HCj(W2?+H7RDY1_k|9Bipn6#GOi&vRaiXXSWhb
z+xDfRy9)Gup*-bCh+AQ-9M$$0`QbaVR6MH(t!tAe)kwkXQxqO4r+>m}*oApxexd5v
zpEz&G5B%-;6%Y0MEN(v^@b9&^=oa$|W5zaPaNp->RQtaiLmif{dxSIm9$>|cN}S|>
z54UU-`Sde4(eY6!`uDnyIRlDNX3Z7+DtJqIbr;dHI~%j~Gf~GQ9pgu*V#d^Dl$o7`
z5pxo;&#ZXdF*Ozo#veubVTZ)EWItXJIkE!1ow!Io8c%g@!JRFU7<*(r&Rn<}*&`Td
zt1iLkU49t;XBHZ$Ps39_6LIO)F*w(II3{Ti!T>$SteNhp`@{*K?6b#=t=1@x(&(;f
zf;+Or=kO|Bbe^w;XM@$T@VF{Yu2n)k14T3r>4PuDY!k*}Z*x}~XGTb3MDt%b<J%7H
zHt*n{=s8VZR}DKI?m>0>T`<hL4`cp&06n@NgIe+v_`dHcbiHqYsfQb3@UceFc5H-4
zf<-)abOWS(cnao!>tUGw6A1e82$rm=hG#DKK;HNkyck#t+9$8U*cV06e6kSYiVDGW
zUJ*2g6+!PFg%H2xGAurr4^``P;pEyZaJg{-(xcK~!n0J+s!4%sQ&Zr~j1=&$N`Xa*
zsgRt19*+D-6A6tAa3m}PO2z%EGBXSI6la6%%N%&)d=XYwiCL+v0L~a+fvfF>Fn;iL
zxTabL7X?dDWzBu46<O!-ll8FYW)qB!dIx1$ZSc*a6O;laQ0cJ@o*pNM-j5V8*Ixzi
z^wz*zpL8(sf)Uz!Sz}I%tH?%qp>*aDJRCC&HMS1N(#8?ED|RG4JvauNj!r-s`N?Q<
z%@+?Gn~gSm{7~+A06rFuje+$mv0gF^bImv4`}j!wsk#Lza2xvU61S?^J20VQCvI%s
zh3(C|(M0eb+xqOse>n%y$oeRDZ;ZpZtBJUwH5r3tGEmz+4=*VdV$tbRG!DIkyAM?3
z`r~yNt@Iqn$-csMXWrwmf1goL_lIB@bYSJEE_CcGLE9ckl9F&$t-dKsozr{MtRadt
z*FuFf%LS)>loly})up5#hNQU8jF#-Prlp3C^iSQ5v}2i6rG`?x!x*yr<4r0(r;thV
zbV`0QNAxrOY4@8YbZt*CwGCb^c<JluR$wG~&fG$Eled$v?+#kHXg7sM?4!mr2dUxF
z5jxyAmUO1Y)5^3&+AVj6rms3j#UIa8?6M51{FP03Fpo-I1^e%HA?2LBPM_zO(HqBd
z+S<E<a=%p3#OF2SR#zu(WDV5#<^{d|(?WN(-_eUftu%W@8%<08LAgyGWUt>v*$xui
z=q<(9R?6`51G4-rvlkD)+nfEG<k|VFBB%dW;;Vl}Ca1F>Z~36kBOhq;^>l4kUN5r8
zp87oGy8+i9H{u*e6P|a)ln*+Yvs}CdC;w;3GY42Rt+nBxGq&t{&5qyQwdZfQ9k@2%
zku8t)=lx5axXjs^w|#XMnH3lA8tuwKA6>a+vm1|4apyhp?wn%e!O=+`Y+wjH=LE2$
znt1&YsurVO#DAiQpZ+7>8o_*3%9HJPdh%9vFP@evI6MOf@SYa~`018`9O^iTQ(h0^
zyi<euj{gvr>OYiM^%%xREyGx{TzDlej^N*^Bl&*vC?1kFnym`Ne&_C39`I@$zm}fB
zE!N)re%eG%-Z_bH-tgfs|0eU@K~wox^fWf8@#RK?8C<?X<aKV&;zQbVI5uQ1>)oHn
zY36=>bMpfJ{LG(|+!yhz0|C7M!(z@H5XdHnmhy~O%lMLmxM76_^XjWB_-MyU_OlOV
zw>hhL<@VJad~OXNC=O$fif}HkTFc4h>)11YJ$D`5z`n~i@?F8&dC<0rf5b;{vtA^>
zTNuf6&PB3jQzU=syP2<hZ04ZpoB8$1&8!%;nP2VL%-_u}gX_CX5D|C@Y_AnS$iI9j
zn~@KfB=aFeFCV6r=fm2bm*8scWpU3ggvRT|u;^R~B;GEAuU6%7{=+>OY*`IKy&l7o
zb5G%pS~DD-{|=mXw}JWIP6!rE(75@sxc7S>>~U8WrL>H&yHX#Ex0qpQhB?k}nSsx{
z=VR=Rg*aF~Kx7sc;=W0K*jIA_UXfUYeho{o)+9*Wu|m*yXc#8`STFuQ5>;cPMCWBI
z{@NOYs#YR<S8*7v`W{0X7>l*yvqWiWA{OhM#VtQmuxC>`&S}WPMXxU6o{menOsxpz
zxCA2uZ(>#A9lX<UAJw!UqRXsWj7)rrUM<ft(O7t1=DbC@BOfuk><gy6|1Q|G9r#)4
z5AIbK-o?$5v?5e|hR*ImN4<K|0}VM!f8U#Qa^%Tyr6S$1R3gQCW$G89N^kYl=xClg
zxeU=1{BbP`8mU8?V$P|s6CLNn24pRsA-SWBsq2^ty?bs-!*$H*;{<W{UT;Zy3Bp5g
z$(o{X+fY-5Exo^OM?RPAsXE?)`h+=B{;>YEsh1OJTz4X8kx$+w>q7NMT}0MJ>{Rx-
zl3|A{{q=UE`|%>j(;$4Ga_)5B+MPl@-Kk`d=xMvVQ<#A}DgSb#4AD~$k8-1PPHwcS
z)|D)lx>B!p7h1Z)g_4_{soyYX(n@e5txx@Fql36REp{aR;|}D0&7J}q?a1V-E!A|{
z&?`wBD*Y#%qdzTa-8;e3s4*wCOM*3g)RY#lG@*qfL>5WQh&H}8ApLCNiCUva1s=Mj
z(Wy<Ld0O;ixu(d~XpoHX0Y2Z^k5-^6J%6W6gAXYQpKxCae62wKJLLs4wGZukC`T_s
zdWk$vPx_nPgD#Gfp`7<pbVvMru4oEg-;BTbdgE{Gb)o~ca@ukF^>1i+>kD?2wqj2H
zdpvmR4Tf%M!Pj$}aFE9{OzZm;AAP9B)O$7fFuxjYGw<X4{JVIe;x@Llm0__#DGv0#
zhSo<5(YNk0_EIlE37?C2Y*#j3xR!wlt!e0>l8T$0lCgEH;4=9oVs=P89*&5`%{z{w
z!GS~g^5}k)II#yep4^EZ3DMZ+<W~GIb~A3Y-iX5t!mzyGO3YDQilxd6@sP<}96NeC
zj@<2ociP5b;kuC+<1_@{>3Crp0`5EDf=b4YIQx?g`n6f$f0m{webf-Mob+(PZ!N5O
zr;dNxRdJS%GESMSh=K!xi{JM`&jGTy^Q<(kv5~}oy}H41-*2$0`3|!r-ol(Y^&ma6
z64Kw?f!$hnV1wd4@EKbPp7{^J`{W~-=v)UjZuM|~;ZyLgc?x(_{J-D!6jDb$h3b3t
zP<#3bNUf}ejE0Axlye_W9=Z)y*}}1~y9mlfw+VADLu$fh@ann(UXF#}MB*JQ{R-^<
zRsbFMFT(J!Y<TC80lA`;eZ4vb-1N@D5&dMCQ+O7-o}Y!@tCK-*`8gPvo&vw*&x1>F
z8f?3i4tWh1VDF0z*!m<B2EEFH;YK+C7jglQ<w3)j0$5Xc1rnu;!LO<WW|`fB<!bjJ
zDZLs_f35|anrHBG;A^<3^$B))wu9}_E?B%%3hU}*@t1WUOo-}>+0Rw6{}@fo5<Rrn
zf>)|~-UeNV*kY)h9j@r*g`GNp6O4zU?3kf=Yt;zcAe`<ewY~9X`D7d=9DVP@=7>)D
z0vs8<80W2Bj)vPp@WqZW9I}1`)|`n%oz^IHkluz04ck$5VGQoPxC2xE2v5!S-DrPg
zFFqHZn#4PY(WZASe%DRFAW6YBJ(h;XGjni`c&_zYaSad2-@=ZX3fvf3gLQZ6Q6=>S
z-gSK~?r9%T<KP#;7yE&y1Rtlc;SbhU|HBe%DUy)vLH!o?B40Imvajw-I!grKt5coK
z(zR(vhCbcVH6dx?sVW*{M=QTLQLF`!j=~_)yg8EUPK>8V!jV^gbqZZ-m_fq^%%d}<
z3n_Q<Qd-@<f=W)Up-(e5&;pgsBrCC%j5@Z{#LgY0BD05zwfECuVH>)><S0!}ilv!v
z<Eh&96rI|ChA#g+M_mDFG~smy9a$<mRe$pYd*?E>4=SQBKdzH+b{T1fm6OWI3NkPe
zz42a;=-f}i#`)4fv)W$JuFe(;lY2)~j9Q5Yw2_A256X<`pro8X6x{faJfx&JL`{a*
zn9K55mtOqSvo{A1mgkG2TUFw&#K+B5c({B&ZhEK2+b?Ty&_*q;chzCtm%5z1NuR$d
z7;@ZRBOWVZ!c~5z{4d{(Yk!&ZTzgBt>0`x%0<GC9#D?E2wdHTq?6}d*o(Ic0a9y<n
zd+&1OwcMXgUiWA7El%twGRj41&aCI^!a-MEIF?*F=87vDxw`SFEH`elbmzh(ch1ri
zvrn7{%V+~Xj|VO>K^~ihtUi!fD7rW#l=-c?C(kPNWTQ|oUTHgk*MA<sa|;HtL-Zj2
zI(IOi7&wFt1fOS))-ZNf9?mP3M(`{3kvz_H6tD0Y&9XjY_;%=6UJy5q_ud}Q|8-2@
z!S)kbf5{|vNcCaO*2%oK|5X0AZW?Qs`|@tJ8T@qNOg?yN77yty_{p>9a%%QG(M|JX
zttkt5RI)$&|5(V+y#m-EdNCiZSi<M~Eahv%moaWw&JpQB{OwUNe{EkO7}Ftap&!bY
zW~+F-aGXo2t>KGZYq+jHj9XH}dCIc2oNK&}w^Xd-8H?6)_4oC>ZH2gRHEm!&uZ`Tb
zN4T;oH}bN-8~KOG_juTD;z4inz&tk(?8oOpZu$Rqy5fF)I~U%?=7MK!E_~X15frQQ
zVCen=SbyU(%=9jVL<`}x8&nF$$v0tGz#VWub06wfJ%q;Zb)aGU9PG?rLw;i`w7R#$
z5c$90*dc}cGljpVT)2omwJ=2V5C(5G!^RwQ42ZVGpH(h6CUGIIUAsu+85W>-+AQ3>
zX*T+0%)`S4{%ESb1n(y=$NPgqL<V^c_L{#Q<)R~Sfca)DGL1xMjKG-7QCPfs2i|$L
z2S@oFL>-CasQ)h!?N^+|4H2oB5P1RBw`Ak(oq1Su{4xe*6ruN>5_~H92L_6FvD>Xu
z^e7*q?A}_`z48<bUOq=-*%rY+e~Y*1BUX?4g8t(Km~PllbaniVa}~Pr*JlYzsg$CK
zWEq;eT9#Bid(qi0IqH?yhXyTBAm847#STJ=CJj`fiPfs)KUIwm*Q(RUv6>W6q(yS3
zI&??etqwN|Uy_Xh{SGmtDY-^8=z}rs)evrSPcvFI-JA|B6Zfk%mQ=UeiUOBf)BdS8
z^v6xyd*tmXpw5me58Kl)u{YWJ*@1>fJJQI0{b~H+{=!$~L_3x_(fpfE6shP;cSbvt
z$y#UH66;Ja)1ApH+nJKjIg{l+XKE3%QHPr|9T5Gd*QriqJziv--}a}3kp994?nonq
zdoQ`lff_}QX7na|nkMF~CBn1y(ZiO8`rA;;c58AxZAC{f3TDPN3z~StTsX7L$g9AV
zv{OuImf#-F4mYCZ(+#Q7)quVz=+mc{y3{XMhfZ(Orh6l_s8m6d@*k*E&3<v)8{dyi
z<y0xDRGFTIDbY8ZzVz?80-5ZUr!ekA{XWQ%;^AJj+OsG5yy-z{yJRTTQJPLXk|e3s
z64YhcjTaXD!JV;!X><2KoYnpv|7v_iPnXZ4r}q&rPJV|`zOQl4WYMP`^#W&lG@`X>
zJ<jV}iv#{XM2BxxIQdHjns?kmHHF(a)3FSD&MQI3xMCdr>?&?Cxr|!N^TqFHF24UO
zIJ1K?a86_z-nx>4VV}=pgiaFL4o$@FfOtF}9g7*F2mZ0>5War2A7wx6!2?}8@wj3P
zY8q@4ezqud4cml+_lILmYKZVvEyHE67NM`gd>lV~Cbk`#f{I-eaM{+;Vs09So_z*l
zmkJ?_6Rd@z{`hac9V&ZSVfF+wd~w(a3w!HhQie7<uhl^R#r<&GYGqs%C!9(T`rrsv
zIkcE9i`^HbF;-R*m;dR4_!phfkk<i2$9#fJn`Uq=uYuX)??HobcxQHtTaIHnEQ!Ah
zDlzwA$(d@X_+A4k7iz&|_7liAt%uq3>mk#x9(<<M!_NCppg6n^CX4r-5akE(MCTs7
zcz**bTMA+Au1nBiln<vz<w4w}e3&%+GDPB4xHsV{$nLrfZ7cIZQ#lt@M1FXFT^f`h
zPXXy)XCa=>K-9P-*eja^hGUYzQSJ=**_{QIxMY}Xo&xt%QsG+nd8pD)he)Lh@cHuv
zc-oi&@Bd}OX)!C6Ov#1Pvw4s%-r+PZh&^?Xa9LdzU5X>+u)R+uoY8*-^KLzbp{>oZ
zdi8s-AN3VvCU(MLF9|$0M+T*%dSUNV^7uGI3Ev*?hYKAvarYY?Y~F2y4+J0j>Qh@x
z^S2khHCL1$4tUFW0D|KX%s4m#Hye({$O>;v+C2rw&z*rc2FyiQYk#y7&$NqjL74tE
z1T)LRP%d->{t&#Qz0pzVyKpNW?7ba(#zbS1+zwoGSM(|V?#AG|`>=lCAzT!647;bs
zqhtGN>|>mYPogui;7uOJx)q|{$x`gnyn|&MsxacnV;nlY0S7*A!Vcv(7_0UX7dL#t
zHZkL9C3N7!mOnUlq6FPClBV4GvQ+U#j@}5qcesM+OPuOQ@1|<fib=ZkCDl-HL(ND&
zz*@{Z4s>v$D@7deqC;B-P~@f|w0HD4I+ig}@b#wB0_$0n5;vblXa!KhnPv18LdfTN
z7?teXNQ=*JroV@_62(N*$?ZESV&5JzKeL~TOApb!)}y3l7)S5to}|*WQ*=<`EDfEX
zLdEyfsCjTE9W2YCOXKs&?e%5)v89L}+m%q^%QA{iEEoT-3R3f^ra3B)Xy1QzgfAPY
z?(qvMu5KaAhi}QT{v+Lf^@R$*f2Ts}4w|6$hg!`3QA@0(=wJ(%!>Jw|bFwEl9+6|c
zm_9r$RDq9A?aS?s%Dh`rm3Ni*W4kDIE_Br7!!=rL8KA>|KkIVlB7JUsU?4KTMm&72
zG2<l@4tQ(I-(}5NP0NA}^(=Xfsul0;vf^*zhUI?VhGm0nxu2yS8`RtJ*G=|3UC)7?
z@*H@|P)9yn<;eeh`tzvT{=C4)iEXQ$IAWADyOue#fu{@iy6nRFF0TCYqARnT8wVG;
zvDyfCF0OOu)}<b-+7tL>D)38hk!O}5E+`<*Ud4>)$pJmR*z=heU%otm%i{;Kcg!Gh
z`x?vxR}bM)YlgD=rePegdpLKU8o_%CNAiQlQM^`oCOmA%@|Bt6IAz~>PAi|li=`)W
z)Zj@RyVZwB-JQ%fN>ll<?=-GY^5vVKr?Uyn<PBSA@x<!cY@{`p7tf!^@hS7!rPYrU
zZ2Wo5;)Q(f)FQ!B3}8j6CERKi$Qomo@-P2o{CmxE;i(GZKas&)6SjiG{a5nb;UWA_
zH<WANg^JGjDz@@n&DXkC3r6%BKCK(Z4#&b+PCA_5%nj$rjBrkF4Hvz#wfx3oE#Dcj
zmV0I8z?SVfaLgkIe%55e)n(aWWtI(k`q^-ARW=lH4t(603$CttP(QT*dOW%ebK(o3
zZ{{@!RlEUNmABxy<UJT)QU%dEk0DyG0h(_&gH50JV0Etz>aKM_AK~Py7%PQ)21#O-
zi;TD_Y2(GUW;ik19KUB;;^cGIn0UeltCMUoYyB*&EuMn4>!+co^GsY8H5YZ|{4wav
zVzeKy9Odt?#1dO^yH;O^;l&#;sCfgrcW*?wf17cKS2R|Y>_V%F`!Tij2)290<Alyr
zC}o_Co6XN-sZj=&>*ipmc0MNRT|ou2V*Kk^isb`uVYts-d>v4U`5PYMs{OTsDe@E#
zWIo4x`7g0J{|$Cr5WSkX&-i!!H;kD0A70h^g&!Yv;gtar;<H+c7C)AzUWa;6@u;2@
z(b<c}B?@n&mpqx*D$o%BzO=7ZiTW*7p@tS!dNWCl+zQp{u8t-hS|xhU!b!eIQI~d2
z(4)CA`qX^cfL=Bml7oaX9Z)x+3KLU0VIlS(Mq;m{Y(Whj7Bu*gCDos?qKQkaX@Q9i
zl|Qs0w^g<j+}n<v4%^W%WqS(UY)?C1+mjzU(32<!swj3K>sAMP*V~bP>pBYNjw88h
zI?_5xM;iCsfojqn$Z5IA0UJ4x_g#C6_7`72>?mQ49qE3tCEppgbU`?v)~MK!=3Hy~
zPs}r`8ZGIMxT$?}wxDZ1=5%MN8Li$RdScrJ-*AU9nQb>B%Z-L~Hb~4gGxTYSV8WZ3
z>57}14r#p5qQX2)x)-fM>!ztww}~2U`K(GVdE&!6T$!$d5<Tfuq?~*OTCzf(w5|G3
zOOqVci@8Q^WKT-_)q_&b%TPaGY5FcBMUX8)Sr@x;fpizrgkM;5^e0|;`U3}Qe8WHE
zzu>6IR!mNNk4cqp#Ec=j-(5|(UG6!4?%RMHl%JrB(qlZT_z<`Dt->E_6&P=N2NOJR
zA^P3G3kR>GZB-E(s$RvJb1&i2>^y9d&c&V6vM~Dm1zgy99(9MLV2>SV@$uc$7~LZg
zg(**Po?_AV<WZb-=Mdibx*w%9_TpOFh0Er|;J*lQV?7mxIh!Kz!_~ED`ZE-zoPw}@
zh2UKl_@SEYY}8pe4Y$`%#H@v5u~~Tp>bDHSt_H@HqD+@D#TkD-v&S9Bta1M~bF?~V
zjGH?2@x>Y)d}OGJ@U0&XdZmopzbT@dx;#!ABixa@Wbx4hX;jpb#9ifGkQv?qWxu~e
zfZ9h;NNI)xflZM2@BzGSz76-nZ$ijAaf{Xv{q5RYVBm8H_QY2}uLso-H|-G|*<1_L
z4%ERWxhEo&@dT9DK7n`bb#P!?E$sPF1Gh7(p!{Yz^cUB-1KSJ0#XcAQ{K*pey)0-x
zkpl^D3*cGbOYrS{0T>78L6}t@jGdbebpaPZYH2Ebtv?Ix|DA?0PN$%JRw9gcPXz7L
ziBPiR6kKdN4Ns%az@I_MaP|E;$ns8w_NC`x<?u9cl}rcOd+88fbpe7LGU0M}7IY8E
zg&9)$(1$O>B5|jR7cNeRj++oX?jFo{egGp*)xqwl=P;t|HT*aFlkj1+gH!V#D0w4^
ziVZz*eTf_n+Nyxl1hAoVvML^HRYJ45HdxhegF0EZ_~NlSnyqod;9-s!Un4k+&%E$$
z<PdB%9El1QV=-m(L{u9(71t}xMAg@GQKHZvn?z^SV`dQ6s)VBH`7pFr*nn#4k$B<H
zX6(MS1>eowh6R11G2~|qo*B6dJ&pI`c)@Rfk$D712gPB^J#nK-5?s@QbX0zlgZn!#
z37*C^j9zdPUq|1=lOYdqqGlbA+SG^|7n)Hq`3){v{1NRszM!+;57d3pfuUuBYiTP<
zqbABwdTviD6+6*&If}GwkO~F-r$$w8wCJ?4J}t{IrusD&<T_W}oXz`Ftg8pnc~6?1
zF^F21jHJI?$5H(CN#w5VOEN2GQ(B`R#dt5K;0Md8c~&Ux?+6#!zD*?U5Jm3l+sLwK
zj9{AYqNJXC>9ocHn&5Pp3a1^TchPZl|Het0+3Pe#`kp0Y!8f+jOsDYZO!`kEm$q%q
zr`{@8#H>_A8%LLr@qcAxb-tW@7gdnIZ8f$3uA##Z>!>KJfo>gsK_4SpX!we^ROA1V
z7R~xXSEhWY`4fIp{fOTb>Df&aH6;0}yfmwJ%W&{V;d6M<i&tg$=Fu_o963!<F#nXe
zs9u?2hbs56SL4IC)p_hBO)h$%#ajpI@U*kKEcr{1{RSBD$}mI0`ZeOFB4dubXTo{+
zOnGmy8ONS6=V==)c;*;O9;Rp|x=2<Wz1o_m>)7zW%QifEf-TE8*@_*B9Z&ya$47(h
zIs3D{a6UNj<5vzGHQSMWo;b4nr2ZUO)t`;WIdOB96Az#4%padSi`$J0D@usoR;(*4
zfEz0dkHq6$?!1412e1C^!KaIWt+$Ik&otr%4$RVhJ^96dp8V*Qm-yZWaKPh%e7bHB
z+cyp7s?S5%OKKRq>kjATG=dM$9m&&UMsZ;NXkPht46Esk<KL6UbD3Z*D^z$3R`Dd>
zH^zs5?3v8{A5P&~;g4A2>&u}hrt|9OGx(F<ES@rJHk%)x!?h3Qa&3?KeAvy8xA=)J
z)((F@d43@$l`Z0}bpf3Faxou!y@b!c2;@=KOSwz*&z%k~=Oqh*#Lg+08(V_eeE$kQ
z=(v&#%2sm!Q6ZdK62cW$q3pdilsl_Kd5`=m7M44HHf<HpTeOOeH)n#DUnZQf$b=BF
z+ii%;fLY>}J!5|cBz9+j_oGa(QO}0@TRG7E^de}C5F9P(EAXXH5!_sO9h&sYVEwRi
zc=5RcCYU~iFJJ25(WvK8XZ#wRC$_?XsPE8M_7{8|^B0nDbb;%`PDqpf4|He$2c_}`
z__xjye;u{PguOPXbk!C^Bs{TX(KxiWn22c!Q*g@A8Mv!`4n91!0EY|^&;1+A@Pxuj
zjA>ql`d7p7(a5zp_V#)Vsf@&vuni}~?Z6-6Ip8<+5DtHR9NUkb#M*nO@xRJs3@AB|
zmYEqiI6ep8?as$9kymhGcrmtwlwxA=Ej+gTE*@W6iDMT(#8G~=sO9?<hmCxW^ZLKU
zJ}Pfe>ce}~&HsdxR)57CmhG5X)q(35{lV-uvCDChq{xq=mvCH$)_KX&^Jk)a87fDi
zqI;RUL!L&<Dbm2reJSpn63v*QLJ@_kB%>xA4~x{PD_es`|Inm#J8k+mSBDIC>eBOc
zJ?ec+pMq))sOp&^eQq$K9o5D(M|6zLPnc5NN;3-aG^fSg=H#4bK?mkolILGb;pw)b
z7<Fs9xZj$Fbz76lTpRk9EAB<VY-o|QE!`K-r#G8~&uPCcSsbyYnY+avXpJom_OT@g
z6I&W9`s(HJHdH^{hKgQWQ^6W*Qu=E}8sS!C@Lud@##z$AR0}HjWlm!UnNxm*8BNYN
zrF|`;OWM<f{LGB0ghZxhj3JGjY(O^C^o2V?k3Npkr64aI(zDg3!D?D$^jm|v8q`I9
zT#b$&=ts8$RH+hGsIj*)Y1Q|o>?B2Tw^tC{W_fD+(VK?m%hBOgqU+tiC++>*g91`y
z=*V1YGFO)(qY9Dx{@abG=5^tdl3%z<uLDP~X~(8}-vraT4Noupgpboc;KvVdakcSl
zG!{FX`<q_i{-j3XqOZrKC$(tZT7wh1s&SG+B^DXpLvtv{5570?#b&{^$-jn8-wSaI
zUcro=1-SX;MQq|6JQJ6Rk2{68e#&|5n|%&%$)3dl(@&#qN&-fFJ%I~cW6^2NQS`fb
z2-kn!k1}R^F>J;zblWR#4<*|$`CAlb7)9c_)9dg-_bU7_GZ+)EF2Tuq{%9FJ2S5Mu
z6|4jwk*67t0e44=T>TJ~*y)9o1bFwg3m%>9h!=m@;NUU~TvBF&zCR4mccQL%ebvP5
z-D+6tr-FsPeR0SVd92?nx<F-uw<)s3@<SytO71UIZ0vy6#&6K9@CnwczXPQO%`j?0
z6C|rtK|x#@lwB!>KglH!TwVfuT1(+aM;Vm+mBZ49_hFFedq+Hb1df7hV|J}p^bhL5
zu&fT!N7aEx#bfXhtefGFD`7(ME%=jI2+^Dky;od-rQ6cLYH%8~uFZgg*gTlIAsy7z
zQeflq)8PCi3$os2g5sPsc>FOL_H~|yXIB#7(4TntFghMSjg5zQFXF-bzmuS`B@qTM
zIt_t-XW)}-GVIPd2l=;B;7C*|R4bo{%nRqiV^<oe*QA5xfef&?p9TA(1S{vFxVg{2
z0$qEG;f?1Fcph2~<LxV<%>NNQ?%4pRq+f!~koWL5w++e$cEI3<Zb)&FMz43W=(VIb
zE?lpG;r@Na?|~v7>XgTAzZFsQrz#dK(Zz!@X4v1s4xM|t;+`X-f)_jh#~2Jnr#mBY
z$*OVKZaxWXnx^8IlQYq6);xTyu@EEfF2OO2gRuI2h?u3r&~N&BG;oPP@4(G?R`@t$
zq_*MyzR_5>b_Yhy-i_1l@53e4hcF=EI6Ch*iN`h0pycaR+<Ye!gD&Oa$=s_rFt-FJ
zC)`Gj+4nK})kExXdV=Kr46k{;#8VP)G4sSnJZRE}dbfVywbD*39NUd<u2Ph=PTY52
z^df5?ar+g0@zF7=6dWiz5_@#$LYwF_Mw(KUniah<6nooS_M|f2nL-7-^Hb6g`r$K*
zMw}lnW+ES&Hf%Z>#?PS{e-==Z{}Q_VG>B%*T1A#$MBaFR1o<XK(d}*9=xay}=?3ni
z4NLcubJzj$-gB55GmcSpeH<AoCQ#A1)3o>4S=!i|LcycbX;e-oeKF3Z!3p_fY;uK`
z<rL9>BTHz<hcfzju$=A>t)S}eD%xC9LsNFtQL=9XW!b%;13g=4%$qk<bn649q<p4~
z9p7kr;D4k&@)wOU`b&E2B>2!=DK;G_!=JTfxldb9PAZh+h6v$wKn3pcNs%KDD)B`-
z6_zbl<<0?WtX`nb(gvCwwoZ$^?r1ai(&hQ6$JgiR^Xk<Gd})&*SFbZ-jX-1W8f(J8
zO-wnV&6Fdv&3I&hIVbnC;E7i(SYfIqpKi6}r|YbEtBf_T-eJufdy5QDj16na*z%}N
zwtTn4mS?TB<I+|;&IqvQhi~lp(IN+)(CWZXLLK>=M1R&kBv?GwPCWRI6aQN3%vK`P
z+_&6?pKNmFBu_WqChg8K58OHJtOqwm04w{6SnC+#FUUN+zbALvd2yxV0NxG*`PGO)
zoH%nZuUs{R`yL$1mU+W?xyUKU$c$uD!5@6*H=0)+9V6V9V>wV}Jj;7c;5Xskyen@a
z?-E_C5ZlRoFJKC*B~9fg4bym!@^rp9YzCJKo}m5NSsYY3n}2+p!w$;x_<;3%-r?!T
zF{2i6_c(u^GkPIwc`f1;%K(m)Tg(Q{i&^Hv622(%%K1i1`MBt3eO<JSr~O>ULE+2!
z`KRSP)hCD#h&$JW|AKh7V=(WZ70l&pgZXoGFegq=hwj1Ya6~H|zP(R_mK$l%U7QAk
zzKdtfs&uHBd;#KPGhp11EN~y21I8CFg43)5@Skx79MX%R#<v9SthfoIB<{j>r7Fle
z`v^AXJ%uu>W=OAl3!Mu-L#g!-xc~bnyb9}pHz)stIbA=*-ugfBeEJ0+TqJR&m~VWx
z+o1F~SA4q30nJtq#n}a;(OJqH!(^u5-R9|7d1?*@3>BQ58v(+*vJBlCg0Z|i1m|yE
zjo~ff7!$V%gFUxk&a!BftJ{U0=lA1|;v+b5LL8cGNWgjFN!UF19IhCUhVpuuxZ_U_
znm@}&ynY4uoh`;4yGn6X=q)UoaTn7DR^n}whgcvb7{|}*@l@tB{1e`czu+}Setm~W
zPqgAZ(J3~&{~gD9|HRq1f8igyzqn$E1g&^4_$u3^>8(x=@=22=50QC2o-IdfEk*wL
zxI9%#D$?KueaW*}i3<9vP|^(11v=S}c8aXY6a@`(1WoFgr9~UUwdvq49g;q(OGl6C
zk@9|ha)~gYVt+$gDteop%En~))R;~kF(IcBrgZePDeaCiqdE<9;uGc+tztnj8!brX
znFal`vn2aaONvgnq@RM<75twieUi1Jm3^#8R@#cLeYT{W3c(AGx1{R%mNZ`5k}lr1
zAddhG((f{-^lj#3BV$f&VP-V?r70aAXiAnx#VpfeOlQPAGjg60<sCGnVZ{bCyIG%R
zbm>u?vL1yQ>e3i19a<)48e3~EYBbiQo9g0wlvO9cZ~Z8)UX|2}R7fsfnJ$DYk^Ll*
z8MaiU$S!$Gs_a9<;(OCB;Ws_%)Qh5i%hJ-49<+3;48;wXranETsQQ)!z3%Z3J2rG-
z-Pd0jI=@5Aob5Q(<2za=wBf^lpYY49kEovi4(ofq!TXb2#Lun?e>Xl8JhTSEgL#79
zVZzmM>LE_JRfVDND=@atU6ks78^6sc!&SRV(EO&@@%1P~hl!UFlk@SX#6`U4myJz#
zGH_CVk?V>}#r0C>FmUl1^r|?8)2$P5(Y6!l{p>ipI2^^<n+~Dlz5Qscx)(F1@4|rz
zF$gcW;Tz*E7%93lQN!0`?akF#Ja`5Amj>d-!3zbCZ7#M?nSsx~Oh&_F6VS|OGzMD?
z!?l_NG0K{7@q9PjT+$!IN86#3q!nIzV~XaThIq|g<U*3P(A`}fPkmECzpB1ya#tR^
zp2^{1nV#r@GT0d=i5kJeBl)TW{5`(Iq9LE)#iBP5W+84&nfCx<1-txpspt#{PeRo-
zs5@K?hldvfJt>Chnb*PY$_@A@cMsOZRzpa54XE9D1g6f_u*16=bO+VJ?CM$=J?1gY
zKkyK2r&L1!6<5G}MFw<!Oa=S-=Ro=PS%@4Yz9yarHJuCKf9@PKy-SAN^{2qlJ^=<<
zif+`dRJc9%4AgceK%>J6&^s6lmo;LAvNaY~Nyov9Q752D?Ib)Ing}&JPs6~&XJDIo
zGAwdD2MOipz}_$goT5_TiE}FC%{~v``lLe@XTaRnEHLPI5tIuGz_;cqNH4t(zanqK
zcIA5z+3x{3ZxHu&_vhf@^cp&6w1R5lclhw}7w8_6z?bSWXy?}x+g*F(J~??@>MD<>
zR`OV|QV~D8siOR1ZHzKD#<&YsIK9*nvzNNzF@M6F`U5fP=1^=~I11x>jmKe`lMtp(
z!(QDpG4=F3;g4E~k!4G;$}tF~)`p<np*1*h!#dpKwFynTBJs?ZD6CZ9hIbD|qt&LJ
z*!%4sT=D1t_8xu|A32>s$<3#*RsI}WUrxu`&>Xa(OPDEBjK_;_;HpV?aOjmvbo^F>
zL;gO&U(L^O@u`<+G4w4?ZT^VI;@WW1*>(&z`Hk6%67-U#>H03=oBbw79IQYsO3HMs
zr5~MW)uhLkdNd}+h%Ub{qx;vbXn3M6#fd$R{6crS!vm<&VHl|$9!=%-6NL9+3U$qx
zK|{{YrDH$+smmvj23-rL9`38@!u55OZ5v5DrM6Jn%WZV_UJT7E-bK1s_tM9b162F)
zFg1KSMqiarP|raLq_FNZeJnalTjf%zWNtcbzMe^`mbs*QBA?>auTb9;MZ}gR)KpwX
zsWZ#TRkDIqGOH+Oehmfb)KPT9Q!+|;E*vs1>Aw9N+W+f4dEWd)OZI#feC2ksRO_VZ
z2Y*Ov>pyBgD#?QviCvCE4?gr&mj631b~!V8^HfQBZab;K2@ZX^E?@9+996jeq$>CO
z*N>fj)LAc1gYzG0^7nsQ+@!C=CmeNox~m?~w$tZsO#?pLX~1{x8}gSUMw~X;nB(P4
zcv-#)SNfWAc)KY(>@eee4RfwJWzG}LEcj2d1;?9Ma#w;Sztps1g@ab?)yJA`wpsHw
z2^;=zqYdkI+wh^yw*01-9UnPt$2w;A{OYPbkDltl$sG>tk?6<+#`Wjdl1@Cj*on7A
zIP<BIBHN?s$^mVz+;QKHZF1bX<%9=+-3@HM6?xodV%x3ETXuW0Qk)m>5&Y%svVm;Z
zG>G%N2lG#(p?qTGFdiE+oU;-}a6siqR_Pwa<AqZzcY$Cl9~;Lb?vLl=65c%7Wg<%k
zOcHw=A0ARZnV<Za!gKYe@suIH?6`0`2Sm-_6>&59+J#ws`|@o5TRexK6wc*)IrDfz
z;(Tt};>R6x7O<wHKbQUV=cwF;-0ZiAgL?+>?Zg1yY_ph$r!D4Uoh2L_v4oG-F5%UB
z!i7F9kTs(M`CvjI&$<xEcimE9w?!)al~09@-%{Y@>lDcUl>*koQo-VBD*U=}9?Im>
z;aKJcc$JX}za(=&@7zUjj4gn6(H)P9D28Fe_p9*x7RbE52j>S1=3k#Wh&Ou%9`jqE
zbct|oW`2e$pKsuG_B#v^ZsAnF?{H^RI}C9B1!~tMaEwGx45z+=^Qny!A6jAgEXD<a
zL-9n&Xgp9c0ljM`qt?0UxWRi4&ZzRk>wN=6Za5HQ7X{&)qbqUj_SJaVW<3t^*o+A-
z+b}(J2f9e^#T&T?v6r@BoV%UCgXW2NN8$|jeRvL2Pp09GC7JlzEf-JsD8Ns(S8zf~
zG4@_tiq@lV;X$ptsPtCkN;4neAkiHUwiA4umyOuAy9xW*ze4AmZ*krDkLY&)Ggk9A
zkx_2PN9`RLOMkF6v>R)yBxv3&DLT|5P0Qx@pilL(q&}(_wdTmtGSxnGe7QUYmMBm|
zpT5*EL5XJVR;Cf9Dirlal|HMgkzapxx;$KiN_;eFq^}m~OxC8q!*%GQqb_wT=~2Qv
zJ(`rKPfyku&}Vx?;fOG#dz*}?pSm%1o;D^gV-v~~^M}np6Iwgdl-8w-4tASx%b1%{
zoVVEf1ewuUv3GIWYDQ)o%xHgr8SM~!Wk)SDN^deHvp7?#8EHzbt^ecfi8)5egvN^e
zOre~}1&0~Y3gJocb1@{xZ3g68sZUk%`s6!QkNU6DrM!3@^17x?)=#yl>a!+o>C&LM
z9-==cr%sN&)M&U=Kl<FELJL1AQ(K)9-74uz`YDQJyHkM-1LetLL?2pW)SDE3^&*cd
z!HYgEOJ`U1Amf2DbV^>D+@46%??efbs_MoZ-!9Z?{)MCcJ8;08cKqb`9bKNc;YXj(
zSabU$o_2kY5+Xl*sK+b(wXhi*OP=F5wMOi_upXT<Yw@sn?q76xfD4yY;?u-?c<gbx
z$co;=z=1b#{rc+|TvUWvGFMS`$|bDM$ir~OT;UMU!lAD&;PT07__{m=H8>ehXC>hr
zqf;2P|0H5p94_-ej&JWA!6K_eXdAsBcfQ<%17H^xh`r6!mTkDfWecv}7Kwe**W;T}
zYf$_B3XIvl6a&o{iLAvu{Iqf=#v4z?rWf8Ao;F7KV~6A4h(YLdknwMsJGN>%qx@le
ztZ}!-jvnS1_|FJOnCfF)s5Tz`q>iI^tD=mz5_-5OV5v)Q{64lPCWXl$W=Z12#=oHM
z+X*^l-$6>^GiW_{3xAh1!T7b0p=$SCuo3gi$o%VIm{|-XZxzB(!HPJ0;tG5?dIijv
zUWLs5#gLm)4q*vZ0B`TZ+3}Td?_fFvZ%Kt$j%VQA)kL_{I{~^B65z&&Q!r<561=K8
z52vn*KleWcM_wdA#jummZ_#mhY?T5nyUxMcd&%&@KM`QWNzvtwgSBy|VBEnYP-_wo
zi|vkqUi%TSymbUtS{#K0g=27H>v0%*Bo5m3PlDR@M5vTYg75ifV0_M5sMkz}BHLuJ
z9+?czx00c)`W#d*PlcExY4BCdIAdmKL&T4Z(A|9r+#`!1ZA~dOzqk#OA_tw{TmwFH
zp2AVhW_a-TEj<788IE?hL*A!9@ZMWe+zn-Lg<((JCo6|-U*trWvNv`FE8yeb%6P?F
z6Ya(r;9O7PX}@iYL;akv)88Fyj2Iu64n&J-!!YH;C>*+BJX-hl!K72uaF*RH+>$*H
zADS%0R}o9lt9Utv*R4dm`>SzI+FEqlvk|vEh{Q^TEqG5bF{-!6pv&7`*ipO>N#QWw
z5Dtf{FHWM0;~9K>HWh!^Wujx*MI1Kc3O;*v4fDNkV%M>|c<YjI+vf;h+Rl26ANCwa
ze|d>9$KGPayjC<?@fGK^{3keIfADXo1f{yjQ0&g0)cL134cV_q8r~}OXn;D|&(tQp
z1NwBi$e8|#_or+<YdUbjo<<5@g+PYVJmW#6x^*~xEgD1DKYG)C&8f6%%uHIjeI8X*
zETn~cOX=B$72;06nvw(8)Ao&#R5E@GjkVfNvlVub_McsJ{QLiLbe>@${%sg;@4feu
zSu!%Zhmk}=C=|-bCVQ{`NfK!nX_rD#Rx*;1T_}V?BveL`XrSKT@qTa|KKX#2=YH<{
zy3RA4-v5lE&aPN`sU1&i1}9SP>=fF0Fr7;7X3|oX6XZJkBux@~92d=Vq_h2kVDDTa
zkHuGMe0?z~tt+A7Qg>-~!UOu?@Q4P#si3{Vl_WQ$ii$tHrnJNF=)Z{{NWZ(DRL*{(
zuk)H|sM1fGnfIIWT|4Q_n}77|oCF7~m12$FGJK;_mY?sC=M)1)UVl=Fuh^^bw==37
zZKckmLo_(OR+Bf{X>;&u9WF}OWwm>H98{~%++@H@n+$pUdn2}cV9fs=6Iq=trc5Kv
z_*|zM=jNL8BToyy(P6<3v6lR`x436ruwto^J-FdU4^|&z%}{L3eX%DepYO@jY<mg!
zRWF`u)|=bodvly_AK}dI!^Zl3*(|v)Pww51O$+*Q>jWDfRA<9hfwrvHyFc4L5k84M
zg0)NoczEXkUS2$qIdl*o_ZiGbhS_tQ?hsD@BQnkPLwRu3Fuo}EJ9i&CaL;n!s29k0
z>WKYYnO7-}WEZPZTs(3#@0#n#19p$$?Z=#0_R&~g)8@=xR^xe&n+vP&n81oBT=~_r
ziL50#nZeeLt7c5$O}nOYRK_$ub<3T{)_HK{zv=u^dnVtv_GGhxUOa8+EUvPj&4X;_
z@KYmiCMh5O^iHq^3+A#$;5<I!Bv{Iw^Lf&-1)Mm}mt|i0a<7F8`FPDj9x8S|d-4|X
zB&o%sAHSGa2Q230Ig7d9oyDB^bTOOiXMu-W795qxf=O+eP}eQ`2{z){a3Bj*`X7UN
z=EtFB=?O?y$bpsJCn3p2?2l{Dz=V6}VUy%V_!4~uB7?5MO5u0;o^}_gtPI|{3vYwb
zOSsee2AX2(;8{Qe1oij|LoA!1hi)^Bp8FkcdH;lyCG8M@QUU|Z<#6*mHJs~fh<$(b
zLer?BSi5l~zL#{y+;mr*`cQnso}Z4N7R^G>k3MK`>x;hqmf-j|%ds_R7500&4r^Cz
z!Mt()IR1Me9+ljOF;heF{KW|L{uzUB+YjUOmx;LNsK`VtJc?ds$MN=y9N{U?!>bd|
zp|*ShN|#*5&;CXD6>gy7&s*3t`yT4gDZ|t9<#;3aDfXH40-x4aW67!-)M>88&P5;b
z-lGQmZ1WWxcQoUoYb~h#^A}z->_ErCe=(4|5ltjXK1Pa;Ny(7GN?GzLlcV*$6llga
zMe4q#M9UOaX!kf(!kucgGe@2JSBM?PcTJituPts&Ius^)R}Ln6G+nTIv%B@FtIk05
zqztJb(uk71jOn1B@Mb?Sp#vLC$z02f?q`|NG&^$&6l^>jQwy>XK688{*vU4Q<gv<<
z%#T{qG2#Av+h|F*J1uE;w<Tq_iH>@mC0#8MJD9_kR58zz*u;{0KeV8GBJb>@Y9Tu1
z=CsPnoVF&L(P1?+G7){JInPWer>_Yu+-yuWSBz*=n;`|-8q)mP2DD_iK7CEsBbTeX
zR8}td!f&-Hx>1W(HEB{tlLl@6tWJI3t5K)OCRyB3p{M7RX{^}QWCSXb;SvRUGESaa
ztmOogU6$Iaq{-;4l*sW&lKm_R%Aeka3x58=nD7o<(yI-due2gfYC)IkW>jAM4VQoY
zjIdVpVFlaF)vpf!)xASi&l>D|`xVODRpGiQkw<8LiizVZ@OZ)_{4RDjj`sKQ;`Tdu
z>3Rt^$lXBYDcA60LLth1y@WYV1>%l%9=|G_#YLO*(E9U9?DEaV(6`5M#@tM7d6SNo
zKB@S;ItfS3Ou&ba4r8COacEZ%jbE)JF)A_)mv$V$2ZGmRcuV9ddhNmXeF5lOz5{1Q
zY{nIn*P*AfAF4lCg4Kdiyh8B2n+!cs>)SMJFP(%ISH|PG!ZGOlbOhGR0^XT9M6eA8
z;7gJHFfi?n+k}7NxrRA9xflz!pFX~}(?S1g4IF+#6_3X%;hRKxTyR;;HVu+k<=h3Q
z?zF?Ku|Hwx$#2kBT@QaEYr$whH58YYLyy?Ia98gJJn}ArPo-j($-D&bTMFRdqYKb5
z-~v4FDYC))3t*h!e4f3L3-#K0fb&j)Q*sX63^)$?(HWrEHx*iUBt!m+Wa#P?HzbV=
z@Lv)Q2NxfKQzi*eAbkX`dK`jNw-16@$UgYH`xrQMrhxLD6xcc@3En6sfPdg&NK=gl
z<7rXgawHmp`$U7hS2P@NiiSRaV;~|u4zB(^4Ch-DKut#Md{WXNeQ5@ixgCW$+mFJl
z;YXplGXwNRF1fvDCak!S1%E1!L*SMiP!V_cBSvRoi0%bY`|mPj%oOae!zGYK_hHVk
z#~`)65^nu|1&xXC;eux)sOx=)yG6et&gLKF$4X-GAQ{wmkVENgd0gtKi2f=n_;jHL
zem2&_#hXpgwblv?1^=&blO39@w@0IX4#JDUSnc8{I(g1G$ki2n?zy4)SPy)A#S_o=
z@y3Xd`B?pI5w1~Kj=2`A@RG$k4Ak9(9}TvNz2PqW{!93(4(!1@+k(;S@qSE>3d8Ni
zQCK`hbj1fIV#1^}JRc?)k{@&Mf$JGueeMFDQ7^=yGq2-#|J%Y@FT_Zz9%G}!Gu++u
z61T^{!Amahalc_bKJN1sHNt-2nn`UK>+=tDUP@A}nJj4tw%0^?74ptkr?^mU8WEyT
z|0NnzRJL&IMf4!IMZM`#iw#Mw8ARHF!|9;S2(sTXh9rgeSF30e^{t#n`afn0{+2h5
zb6h~5wk@HD#VcvL&N@oox|yQC?V$F{0c0J#hxW_~CdU1=%<3R%=|s>-^=PWsjw4B{
zc-jGp<S1qw3QDI*7c<H8=W!y3lT^MpkG{M*OSA9-)ude_d-ba%wYQk+B}+uj{tk7i
zJs`=HhjeLB1-adSMwJU+lC#8X+Lip4#$X+dtNcX0R(+;55>1pA_Jam#{}Qg5b{Z!6
zmm-&RliEB<9xN@*5~(s=Hb{;;uE=vMC^F=UeU82gf83}lvN&qoLq&rZIcqYm*W%=8
zZ9aTVhn@0t`Ej-$Z$F~XF}n<S`%FVlvNjT)ZX-UBVa#hiO!&qh6Sht=<uOCe`0-sc
z4w!Dv>aWdt_&f{Fud!gKIhLGKY03U>Rvdpv?0m%RGfd={l?Pe#<x|%DykAfLak3|e
z4eZ4|FZbep<9f5i%ib)xMtIMa1!Fn4FW1fJ$Ae^S_`1mW9NS>a%|rTg^56cP{lJc;
zQU`G4wt+lUu$N0m59VI|?ODfs2rtkZ%1^Y0abKO`Y;EMg{?@?X2P5})CT^I`T(MyU
z#|XDpRqiNuc{G}XzKh(BjuY28jOCGYow;()I3ZLYFJ>PXF8VQno3$r$k0Fz|X4+)V
z@^j;kJyY20@Kkm>HjRgzapxT2L$^FXo%K%6VA-UZyk@T_KUw0%XGaMisNf5BR?g<l
z(Q`O%tT%uD;?1fdKAd1Mm+h11a-i-!9=dxTOT3%MM+eR4+D-F$|Ec-BzH&bMf1l5X
z|IO!#9qC~BI~{_4rbD6leY{QvtaHzRi8nK#c->JLw=xsl#Pea8`*Cn`%!UC0CqYRy
z51!PYfqvTg&=qqLdao^ng{O+)&xBh*WA8&m*dw^F{0vlMsvz-g4XB72CuN-AyFYD&
zCzro~u1ph@-TndL1#NJ1gakevD~I3I)bL83KIYu-fvcqlie9S&J}4Z824T+VTIq^E
z?h7V;;B*`+KMPZ5`ry)43ozYtF<N;nNA<VD>#$-2KJVCueNzK)>z*L|kPw1bZXZOC
z7m>K<W-RJP#N!y3Bs}mt6{n^i#T(AYF+p^BPly@E)IoS0YA)cfpv$P-uL!jt3f}&z
zTlh@j9{Qw~;(-B=ar~tx813*Jx1X*;E0Z^Pf7d&_^HeZXOzZK;?9Ui~;2UD@cXYn_
z6Qk}6KWB-^f#eD9@xg8^?k`E(!==csL7HBoENuuDyom?$B15N0kt3C8?<!^L5vM{^
zMUM8{Lp6F?qfQO=g0<D4Ne^nZ2%cz@>J=SQJ)%q9tMw?G^u?W7pN3pApv0wyRH`C+
zoX3qQWw<dtyJ}4B!%V0q%Y<ernu-pBDf#D`(wA?hv`ORyRRk0G*ithZwbhKOcbd`e
zO=dLK*NhyT%xI^984Z4KN_*2x>8OV(iJw6-K4L=6HYT(?-<S;T#l7r=$U<uyk-}!-
z6Mbwz)|LkUztKyb(j)s%x-{ERmr9*==+#PXstMI1YvH>5dqsoJK2WEFm1^|#l`7qT
zBfL<rl<88X67?(--pgVIYRQwQvx#yPwO5ud`^nJMY0|XCUW%q@OHxy9H|iYuht8A!
zVBVj0+;;poKA6&qS6hGJ!b8niDmqcEk3VC@(gvLL_alxDtHWJJwRrbv4Zi8~8n<Ux
z;jEr7a8%MWlvES{j@uuj-1{=j8v6jhXWYe;|3uDa#!aj^Rg7KIMcD3r1t(v=i1S1a
zeavpL<7+yNxxTq*@+JpKFqNd=XW@tyM=_-}4L|&sf?t%9usHSzCYv0_f9bKf&suPa
zGb8Z4MHmJo9Ka$?@qaN`_|rN>H)dr3PQADTtq*O+ha1*m`&2(%W49FTRedq0$s0#J
z^F+7n?l}3vWV9`C!G3p~@Or~Y+-`|jwP7gQ)(ynX>um*xxDOU|_CT99b6jm?f?<mc
zu<n_T;LmBIy_Fh9bSmMWW_i5SB_r-8QfQjn1<E!Z5OeM)R19f?Rmlw?Q}!NK39q;E
zf=W=-e+cD*Cp@I#8gx9m0%k7-6Jpf`xVP~<*f*So19@j4^4MwU|2Pl4m*v74adWDl
zmJ8$EPlB0}VDG&@2K!Ue!DvhhSf?g|ta$*WW$u8Sg?nJ(zA$i7jDa09V`2327-;d0
zgeP<NLsF03F!uOHki51UjNAOcWXN_n=H(9`CHKOZ&~SKW6almR!^AE*97<k<!^+eM
zxc)E_C@~sp$Hu|W+wt&gUJ|UDlM1@d>9AAGQtm&}L267oe2h(lrx9ro<ed(yYBS)+
zzf9PgcpS`fb6}or9t_tz2fvpUz*4J1m_713<XpH7`u?Tx=s*RGdiVlP4XpvG^B>^q
zpf8Yf<OfW9&;~7W|DbH1B&LQ)W0topmMoD+X>TPwx=#&j?`fk|sv&yEidoa6CmN>r
z!>-T)7~n7jAKi4oY#M<YryOyP!8nu(bj9!nH`H<Uz~G~vsQ6<JM%c~A2Cqe!EV|>-
zTl`Soe=Q~^Y{dAct@t2q7v`<pjYhMA@XXPDcv5`3y_*|>Bk#mu->vbe<C~1m>oZU`
z@i@M%I3+sbqSH6(BE~Hc?8y~3(0$e&tm<Eijb9#P{*h-G=U9a|e!da8#rJ4XQIA7<
zeZ!K{7L2^wj^~@Y@O-WmotBiNbDI@ukA^Dgmut}TGdgtjm;q&{n2^bSvA^AGO&yAT
z$)D`#(Q|tey)la0Hj;uvo#;Y@aM(pnrkQc>G&9YULh`-o?b8L++P#Df(T|qxSx5I?
zZ>E-UJBexosP_0CirErO%ct(A4TBF-zG(#Y(~71C>TwjL7f<7QB~sf+vB#O8P6Y=t
zDX{1`U2e}ITNbyfeP=~BF`v#0K2FBrLYg5_Ode})lE#PI^v&nK;F6Y+!rXF-eE*bg
zFMC0e&DB)0sfLU?YsqoPM|#@UK=W6Br8RGwDbVRBnWp}xlBN!l8T5~6umsnZNOJik
zX<k<*!wzHQ_;{{7yQwMiqlHSsE2+%3UsX8OOpOhk)j4#Y21~Ej<Rfdfc*jz0e($cs
zCic3#M?sHwKh@*VSbgp>$$%?64cH;wkQ+xCarg@(u2^QwzOBambGr${KND`;Vak`=
z#C>a>83#9-aov1#ws>jI@1|Jr?Ryry#L1FpUbkeE(W0|eV#U2&d+?-6vG-YQ&38Mj
z`Fc!GX8T^e_GK>)+0&bk4e7%*BI7eEyD#7O>&IJ%+3*NCTb6ol%h?zEv-%OiBn%k9
zetrX4a_%6GnlAFoQ|vi!$`HQnF_f>&8OF<&4(F|#9XK%r*gqNhaX#^oN6gb3NAMHb
zQ9QZFX!dn<<eYh9xM-IX_fH(l2QP|F){}8O;mdd~lAORUx~{yY=S03ZXcCJJEuUgH
z_QokZaPU;l=`oEbsJV0W4|m}j_h5VB1x;8ngYVnSWVgDRTomcar|iA>La7&T^_s;A
zRkQfE+iX60aW+fp%;9|t=dkp#IXw2&9JY}5<{cu}JhGoR_kW!NkKd%gtq&<MR@}QY
ztx{pi+En1CRB$<wCc2mD@Vq4hdLGP#PaBRw^NABMtlvpcY!}}_I%na;o_ts`;Sv-r
zy9&;=*FiS-HnbK!fTqsJ(4Q+|ic2-zcwPf)neXA+v5$~6u>njM2&c`hW_WA%8*X)V
z!Rv*xIKD~+Q#*BW`vD8w+s_7f6xpLi7UJuu(fIPKGj@M*MfIF1XlFkiWe$4bzRSW}
z^>{v}$}YyYmzHDk@HN<cU=t45vlBi4>_&Hyap-!vA8T%gp=@*%{&R^#mF9T7ejo`i
zSft^L^J1qfW}Mw(k7F)wRY9F+5kfCuq|s%}KYkT247!fd7fW!A!(F^~?g7g5dL;Hr
z71;CVGn_c_B|eRRg@&(cFk16H-ldP|HLC&Jmw!RORZSSY@CWvq)QUrTw_#dq2hPj?
zi%XVtqo=GS9iA#hUt*<c(i0iVQ;;Jm2YDJkUx9}17L33I@xIShrso%hqx_00eZHVZ
zzfY*sEx~=U*sMv<r)W{Gg*L5f5ImS;I#e=8mp+Oti}DFQ`ZrRaf=l&j!(;=ByJbMv
z1{%_xC_~y`Z%7S;jOgxKBht<^qQX)mx>#>S((Oi+@mI`5zl=z(&WOI>GNP@CMs$6N
z5&h_8L=LYF>0YoQRrfNalUEIBmy76nJ<um`(x*$pKQZ4>k9Kd-rPU90XuGivwJ#9g
zc<EY{SfxqJlr`z+FbxX#R;OUGo9c5|m5NTO(7_^QdU;=o>IAdqbEN|HelEB{Pvpp`
zOqPb;lp(hZ(iEO0MdmS*WbH3O0S~*dSHfT1;M<95z1p$n(=VKT^e3|S4?L#XjOR<f
zig*2I;n!}!(RV)Lh?RBtMx_?Ja%yn1^J~<4U4=)4U#wiP8+Qdi5w65?j0$~-&*e&S
z-mZJNrTI3d`IKO^-~@~vbPeCd7h;jbWprFofR6;PFLC@?OfJepKRAWM3bV0z%yC@v
zI1_#6XJANcDi$3`#`l(q_~TMMDvEFEEp;(?Nw|Zy$wpvI^g)yr?7pv=`*3%^y{M2E
zh))Lupx^u*s4#CcT6(X;xn6#_ZTeDlo#u<Lrud-$G%tMZ>4AS2x#9R76Y#&}v3T;)
zD3sD7e7Is5j(a@_cP#IZ19ki2&01^Jt+2qYA51XG#1O-_>EcOoCpn*@hBp=}qxX2>
zT^J{ekv>wWq1g=ydplsES}UwhZ32aUjj%AG4(>jA14A-jz_aplxGDI;9=&eDVE3zF
zseB1+OY>pA<vA#>Jq<s8<O-knDd;xOfsNmff$E!hcxssovjlIZWJwTs*l&dI9c!T=
zdj~9=xI_5rcEC~pO|ZVlc9`&ZH<)#Vz^Maa@ZwiE#M*{Kl3OS=U)>9J%l+Z|)(uc}
zV-@uIu@ch0uL6mTb)ek51#I;LVB(2AVE#G;>=qvYy~&|a#0SA@cNoBm2-vhB8cszY
zf=|^6pyieVm9x^owL1;A2Bm@Sq*Ul5?qj8e$?$ni3Y`6(3hPuez&AS+W?wiC)qPLG
z2<g)>e$;tb@aiIr6@IEA7B`{n(p{Jn^$@ltJq5qgD(F&u2PdX|f^YwQ1*4rU;A`0q
zo5uZv%FhxQE+>s&d}PsZjRH>Fq=F}BY2u=rdZ_AdBIZ*|jC|V*v(DO}^W=f}>FyAe
z9|-6WJ_04`9g%n(Hix<5lWI52u<*cGZ%=Fuor8mN=i$~H3-QIHrMU6^O03jei>li<
z;xgWbsoVT9rB@&h7cK|$e<3*R+d=fVi^AOeICNi|fcEaGf<>5#!@RQ5dRZPG*l-?C
zth*#w?nS6K_$J<Mxq})RrD!s)93M73!{hOyJF%<=b;Zxj`_m10P9#mO)mu^8vJ<}y
zlOUNL(xm=Yj+QM}qCskEG__KblFsVV`$LAbJ;0QFLM=(UyC-F|^rII;2hx1Mq2%O`
z6t-E|ZhXhmh)EM@l7ky1^z)#co?hf<<3p=Q_|hNmrPMRjkIL??qhoqo$Z69~+Sd|5
zw&(ZI$(_NZEnJ){`W>V^wFr9gCyJi`h^6@Nhe@tIfn*d?XoE#M4Rp*TjTOh~Q$h~y
zc$!NQ+Gpwc?0o7tagiebT_J0)Yt--34U#gwP3?Q{(dd>^>X`qS`c^!ln~u*Z?Q9i=
zn!lkD2j0=#?{ySDwVoDbf2Q>v-)Q2f9~8W;m98CYqomuN^y1k+N?a}>xVe(NS<E>-
zPRod{wH((^m*<&>6?oetMef<D#G{Q=_@<pIpC7KqwnNp~v$qEOscCY-XHDLJRf{JD
zX$!uN4zF(0;fkZW+}lNu<v;4N)((CCtYp9tXTUi|hWsVbkWU#JaYC#SFH$w;3qi&_
z?2j>rt~2534JK?e-;^_}OnKrAGtPQy#&$Ezc~-SKhcC2XyB`)DzQ>Z=jjY6b-->Hy
z_25EDYhHNTn!ot=WMAQyc<`zho2U2Y>3)59#)!TwtJ9C2zW3wC3LCz6$(B!N_UGks
zc6{}~0A3L^kOv12;tIj&=?<~y4N*h*ZPHL~%^Ak?3y1SVam%W!1s>mqoS@3Q%X$Qd
z(n$7kAH@fRTOw<hBae<6!wnfu-0$32j=JW|nSw9)rEENZ71<q|2NSse4OiZBZX$;z
zPvX|ylX>VoH-0{33j0b;<t2Bg^4dMqSZA0!f2wik-Mc(^kLGj^PngaV3}*2B12Z`I
z`wZUbI+Gg@&*W3W85*YI$qVg0S!1#%|2NB%?_5h1?9N0u`7{x>wj@Gl-y~=YN`kX$
z$#AeD89r8~z$(i$Fu#}%6^D<4T}c)^bvOae9Xa43mj|W3;@eL10x0TShQwt>5TI}q
z?ti=ks*R<9&E+6*zY<zURYQE=8kibZ3&!_8K>e-;nCtrut}C~~?w@}lc9;xWPgBOf
zceT;h*%Z%Q>xJqc?QlZTP`r1RFjm(QU5v)z{#UN(y=)3Tzb~E%N?!Qc(HnKs=HtH=
zi?R6C3cT=lE#{rsg6@_6=w`hKt($}K{P_boa7j2C%SGe;!*Qr+aRiTMB%!rc8k(dX
zMQ_vNXq%9OYxVN*dem7AP`-f8ftN&wvJfXPEk@5LH*xy#J2*A&J}&%HhWQTVxMSy2
zlrDIVc^|5<RbKR~%->;DzdEG8pU}d%5!EHW;)fSaxHsbmzM0pGZ`9jx>$wih82=Zi
zJnq7anG*E7T9RfdNmKb?8JaL%mU36i(TCmg^e9Y0WL*@=Awr3k2PsqjIu-gnQ)G4Q
z#WO}ho%ElnQ~Mzex;9ml21sg=-U%&YXEF1LtnS7II%M@uNBnuZR8%B(DdM*EXN?~1
zxui!`zx1fFw?1XL>65hRL`ZEH+k1a~`msfyIv4BHQx|<YVxdod-}GqxY2gtTy{>X?
zJ&GvMrD;=iX?l&=$t={N{_nI!E>D|gUDKjc3-RLJp-I0VYS0uz4La<jPUeYfH0+7s
zI7_S2R67;=>Y+@o8<pr>gd#0Fra-xu<mu)eIeK3%OU^H3$g*0Rva5s*^SLB#dMrWa
zV&5gb@gLUA{DZASIxtYJ4NK};@mqci4hZ~?1rwVvTk|XSe%6SSV(PK)l#lp8z7Fk*
z-r<xrHNxlgT5v+Du>ZOj*h{w(`(Aj0($mT@q~W3Pc$A{LlE@Fo3NNPNEmTRoflUU*
z*bpPWb5*Y3$DoU7`!65cHlM@)ex4Tll3aY#o`c$ZPT+B^W4QGAQG7Z&9edWK;2QrV
zJZ+JHpGpp6@1=2=r5J;SIg!{jAsm}TMj~qWetc@U56eymVYX8sdY=3b9m=+2%;!z0
zBEJq@to*QzmtwW2FIH^u!KNrL3_9<DgT*`Vo4hOj8|jRlyGLWeQ^EnmhvTZ#gK_03
zJM3)iE8d$uF)`i}W716V<^w~#tFI><vszecsE*xb%D6LC0UN_*aYeio!n$r4_o4$}
zb}LwXZ-R|sjnKpT13XNxfu^r7;qQCFvRPOLjUl&TSI0FNy81Fih}%s_##!jTE;xCk
zPKi5nHY~|F4zurN!dkgB_>dG0?uGkcjmd6!dTu+6p1c7v>ej#;vo&zNY&BTFUIUtM
z*TXlzEpU6;e_+%Z1PVDJaOK~AnB%@*@U?<rjcXuOTWyDWS_kWt{lL#@1#CRL5>o3{
z!-er1#Vu<G^!O40J8uL*Zb~pHU*88^_d{Tx?g1Dsdr&a=B0$+Y7B>HP1dR42L+1HZ
z7~eAu9&;)z_DTk$8Hu2LApv%ICqlscBzP~A3aT0Da9PYaRyrqO%bSznA}zY;f%(wl
zc^TYxUW3W?B``MpKGZCF49k5gp>WwNh}%>PN7vNDf|1`~!|fLM__+<vpZW{Svm~%=
zpfv7sm&0enm5?gbaCeP1uFx|?nGt6A*R}^{)%M0w%S0!%d?1eOH57f<3I1Qf2sCLE
z-moF#u-?xV(^B0q<bgZte4U9>GIOv{XCBJ;T!?2K#oX(^5@)oo5e~GCXxY6DpWXcr
zvvv32`S@Uz_B$Y$>ftC|AB|JC9L7oglhFHL8iu^d!o#<6uqgX9uG*WALGHpWtac4G
z3UA`F*>|zBr4(br%5l%|N?a>ZEnI*#c+KVmdZabt$r;Vq*X0*VZTf@GO%n7<WZur6
zm!}Qml&RpC8oj!rMd!lx$Yi|{T@}0EyVX`CnIKpd@wQ|v{F~4A4Wq{kiFOSeO)YZH
zq**<I2A_AMl2{MnvGF30tv;l?%a@uWmQvwaKk8SvjviQVA=xcEN$*<#q1fYeFAt_w
zL-$jl;z1hrDV%ndM$y>Iv9wXRs($1qP*Q#}O)E*Gk5xyhPx~=?Z<<3-TyyE9{~6M~
ze4b8p6wudESI8^0h^D-{PUZt{kyP+qGJgAjw!<Sbh^?U4-=0yT^Gh;Fctt9oYsh-Y
zdy?Dok!GB2pmVif=$2eF4Kr_{;=aGgx@S9)#vl6H_>YpjBslS$B!^2$vx$og{|%Jo
z-{<7`?F)HM`=P*g5=tB^Ejm|!lzCdC3d@zLvRkGa?^v(SO~W);t3`w7Woq(FH!TkQ
zuEiTewfUC04r}G<aMv(hwky=-38csE7xnnVAbrly)@Kz91CA22&pRbUjtwy6F>Qvt
zb&U~UYc%56CC0q|lQG|2Zo-}4O?b&pQ=X?HW}b92K|wWV5OdEn(c_ZpYsoKPTJoZB
zG51X9!P!dI>|SBbG08pIV@)q!CvIE!t@^N?Okdvcxi9-x_T$^PY<QlScUn&OXFh4i
z?b!qP&B=kBab^&ATpY}iH|$yWp>R%A59RQNVf?UdI7bQo&Q^V7v);s}!-PA1><HdA
zbtM0rJ&LUtitNrZM~+!BhJP$`VvB`iImgSH^(KttulD1)z`%w7{dVC=k0!8PiYpIZ
zGLaj4PU5=Pg1>WMGD{C|<GJ_Uc*U$K{IEvsd}d7*U9qWr(MIfj#0+%t(KIeqb!U&!
z?%ZdIJKO%}&J#l2d1P8VL}bT9a&bI#)yG3*zavl(c?3pTC%}}Z1X%hz5nekcL)p_5
z2+B<p-(wjt6tdvkm*enN^sPn?&V#e}&qC(u3$Um5GT3@ugOPG2u;uw($jT{$W8M{z
zcS}4g?pHy~&Nnc-zj$8E{|KM6KSS|>AFyaeCv3VSg{Sr^qWl6)T-$Ak1LLg(ORpbp
zmKcPuz79j1sf^Au9Pzu}I4sX{MOC#ac-6-P`{a6JUQciQ@^?Pw`Yge9?tWN!Z#}-u
z+=gAEn{rex2+dFI!`2a@*m64@r;Lflg#~eF-SY@uj!eSf&QzQv@;K8=kKuvAIT#b4
zi*<hm18{CWt|_{Rk_LssWq1uo7u`T}x!X8%{5{kQD8;nXk8oUN1+M-6439{>L^s)2
zIH2ndHhq4FlS=DQKlKy#UebtjExuyb!zOH8^#hG0Te0uH-zX>1foUuM;M@EEaHgsR
z{j`&${0UMN<ReYvSIAJ^8nMd|vyRbXd0IJLfx1YMvW%5zu;@syzo|_2qH8{SyecjE
zt4ga+sFC$Vb(--(ogQuwIpq!wdb&}Q<UeWB<4Ia1m#syiomw=;QJb`OX;WjaHeGqD
zO*=%6*``bMu|?K*#b0gG_@XUb3))n1T$_4t)}~I-rp-Ua{F0$Xwk}%qq)wAMH)+zx
zF0rTf7aX78;(jKWK1rp*$=6Sfs&}hWcZCpN8LQA6A7vUO_E16Piex3JNFKHdG~ZpG
z!Zyf}X{0PoJt0F2uS!#7nG|VMNs{h+2|6k`s`bzQ;pAt3u-BswEWX($9NfRqIr%4^
z+w%ioFK)&Tr*G(B_618?8?gB9C!BovgV>Y3#~rrsaCplbEXsd{G8?O~YrqSP6yK*w
z$xj6HrW}L*K17vcrMPC|eT?~f2UqUDg>qImF|42%r;RVd(q~sNW8Nh^*>FL4p3kFu
z#~JLiCl9x%pTZ%j*=S*Z9P>*v(P+8g^(m*}wzJ9TIVTa@<d5LS0>SKC9E(A^(HL?o
z0)5sCZl75wzPJ~Hvg?9TOFsy;3U}k|EB@&4XB#@hCe+`w7AFgj?7@a57-H#*3Uhq$
zN1PYVdFp|2I#bYojw>2wJBxRZBOYJK=vz4)mwDM^aI@gYCHKQ`iv(xI#R~gOFvEk(
zjc{3x9!^rxMu<?y<iRR9<Cg-StCYq6Dx^@ctQ)K?I-%omD-5@3hSlYbaBk-Z_-6GM
z3eQ!6i~chxcX$MM6z{?BQ#WATkwQSX0>IX@aBOZK>~cE^ZBZv+d3%=dWgdmxl2kBo
zNdRe`NZ?n&;J+&XR8Mb(*!Fc`K72J)UiE_ull>stVHIpEUjr_R8^I!TI}Dw=8#G1-
z!MDYG;l`67IGY{_`#W}lYWODT;kX)}pI#0rol9Xr-7=VRbtO!lxfZq?Z-x`PJHbhJ
zH@s~Rgp6T9Q0KZAf<uGB$TtK&oEQ6?-qFxGJ04O^lOfnQ6(qw_;cBlG7_=%80zKlv
zVa;Ix^LS`1I08LBCPGMX3d{&ihgn}ULHpPVuq`+R9foIN_rnYD?ZFk8t8^W718>9K
zUZwExi^wC1zPobrYY1EV9;T}`z@nYs!2D<n6o$6JyLEqI>n91UR+qt5FGPM@{N0~y
z(ZE?Xx)^cG2*2(%$8|>5*cQ?UH@+5(W3xe+KSRto!GNdFjlkKj9MN229GY6V;skU<
zEjM>`^qz^mmd(bWTj%1J1Ydmlb_ougyAt2qtrhpsP59h;2SyeIV3T!_;Op$eEiXgy
zgnA_Q+!Bi$q>rFtPBMlr$iS7B$MH_>Nt8)CgP9XA;Dx5k_<Zj*G|?%+eyMkH*MKtI
zUS5s~5tTT3do>!0PGMZz2Mmw=jJ)VODzE#EQ!f5R|0Lm~6;h-T+Z3o-Q-zk?P^ag6
zM3&B5d=L2;)5@dfBtO)ec6IiluG0SWB38KP+=tVCB_^vIqbYB@GufdleUzL+2~Rz!
z=!_Trm*7M1qI~J};iZ&t%1`k3*3m85EmY*XlX{j1(6~K&=)uTfIxDrGS|5edzN6vf
z84yKoi(_f3`(YY6K7oEZC6kj&8g+RbrO!)_(d~e2a!fx(a`#Tt$<}jprB?yv_*|yi
z_^WjGg<u^UmXL(c9lD-=pN@Pgqn-WAX@cKVs>pdxRUfLz*zgTa8uyOOR@RYY@F!Y%
zsF7wQe5E^)%@n@1g`%haqTBlIWdE>}y4L(7-|lXjGeVLtiX3vp5os>GD8p8zvV8Ta
z97k2i^QBt~JR?_;Q$m&a**s-->8HXozo_t{qpDo*uEt)SYCJMQoqO7AaPn;p9__Bl
zIaQjx&0CAzUTN{enc95lu{KW{rz2Q<I@~rumwROD@-YKFRu9+X1{pE??9%7vc70ae
zBJw@$1{}Z3kZ;Ht@i?*fS=Gy!Hx!AT&ny$J=``VR;m<lf*^H-loAKrXb2i^#!J~#+
z@}0kyoOIuc(-V8}`SsR(!mTGyAJB{Ui0-;XS8t9HH?4oK`*L?hKlXfJ!$r4kxv``_
zN8YmIg?9(AX4ya<_hb;Csv67<wf3C)c?fU#Ig};-4ilaS2fnBzdgvO+d$oyA=!kho
zdjvbEjpSS6o&_DFxJSci))3y%0|jH)JkE)|SB>RIBb`}ZaU35h72d4i@vQ6M!h>pE
zSa$aW9whwevyZxRFLC2?Oq$5v3X{11@=3h<+9ZB2KbfaEPv+pwlX-sPWWIKOG9N6S
z%&-271KWdfV00o5LSM##uk|5#7;^{~4>%0ieHc`fjzIPF1i1e`5q4fkh7%9PGsGqX
z)|U%b`qks`K=jC8t<Qrid=7YW0Sr#Q0{RxkAYW1f;ePiZqN_}>xhlXz<n})JRSE9J
z8?aS-4*^H(LGnZs>{#ChTTLW!_#S!O_*4y-U(v(NR5Sct-xIeix5Y<}gK=-taQv6V
z_<N4X=zVrZEl*eMd)5uFYIvZ@5y1i6=#8o+3vgM)QuKbf3U4mih*hI^;AXSk_$DO?
zmm7qjOh_m$ZwkljlcRCe3Guz%9gkKsl5jzGDrWu7z#6w>SSa3SSDQ{D(HYzxbY5hE
z3a~}-3T_))gp0Rc$NJ0?6cSK8@!`J6@s#1@uE)6g&lB|jR*7cMUSfCND}1-B2E#|!
zV$JV5Jd^ebKaXg{O^?3dDEB5DdjC6)v;T>6BY)wdcWpSbM<<qv4p4nS7oM9eL5kBQ
zX^w{!g-n#DH7r97J!Pp>UQTd+<;Xl=p6vY<sC|^^ZgwbAX|@t=oTg0D-<8Q_uL@~u
zs?wks!HZE=qmpfE^zn@v6%AJ>n|<n}Q>IP}<uph}+>AcY)*y{d8niA*gEaSRP-mb9
zq44T`n5jWpgEXj4LW6eRQ74%|bvk9QPDfv<QIG9vv`AfzjwPy6A9Gb2n4m&3s$w^@
zMVSV_QX=!=N+fwu%rn&r^jL^erY;ou;1oIf`%IP!WMyg5AhDwo48G&rrG#Hfl5U=p
zpifcVsCe-odQ|>F>u(+SysHh1m44#@P0?4?YQfuT-*LTM6Q1k%f>#?FvE*qz{=V`N
z{Zs1D-@jI5zu)5gA#ZS{(kpa&^AbOwdXC=Pp5gv6PjH?>IbMJ85X1JD;?c49@!g+0
z*!%P?Ts-$C#w%XOCufRq%=AKZXuX6sQ3Y6FlaFb)&SIJOX&m<J6xN34;N5;Fu;hLg
z{^xfTS7@i>$!jTSwK@scStg*&%fr|+Dh`z<#bCXB6n2$_W7dBMaVj3b*q{5b_sPBZ
zZvGxD(AkZCVY^UXWGFHxZp0hM*5H@!6?n~a3BJu+fKr0r*W%-a2^T$3L(Dd*8(r~P
zwKHxPS=6**#;7q4c<ZS>KHM|_$Jy9m;qP7;TWy8EUz%azPa~XetB*JLXbT35IwnS{
zh?|@u${5IDp17UNRhK}ed7UuoUMsBiXolPEjo@?o1E{Tf3*UQI!yCb<33Pi5(JA+#
z=yC}REh&P0P2o)FJ_oL%CmAXjd75R%p-}h`(xftAZBq)oH%f$s5pmEuF$^Xx+Y9Dz
z{Xuc{W>}!S7K#l0Ah`bu2s*eNE;_D&7n}Uxu+3VqkJ$veg*)K$fB-PI+6{^s0WkXD
zE_icj3!J#V7G$rkfbVygz>$HA!R5(fP#C!k{QCKUoX$Fk9=91>r|bZ)5B`vr9{>kR
zcEj{ffnd8M2zqbb2b&i}!qMe%FxM##7B!3K*V|-hUX~0ajvRp{)v@sRbqq*dj0L+j
zhamTKJoNWUgq3qrAn$xS;M^=Qnw1To(sQA0$T?_Lxd>;iu7dlj8xZvV4xABgDn-*L
zu;T6u7`x;R^l5w#LmeBSeCaok_G^KPz&4no@DGCXByq!RS<DYm#PlvzJaAkKS9<8<
zp?YK7H`@ZA<XPj*FMTl2pg)eGLFnQ+6o0H1x2ioO&@t8#FQq%<_>&XR=Imtrmp={9
zUzvd?@6W=pZ{l7#%ok^0T7p#}D^am%E#6qN8N;J@qV>_;IC%74lr`CpMgzjIb4?T`
zKa0cvJQA?8E(LA<kD`?E3Czhqg#*Qn^<DV|oZ)^2w>-LrA+9CpBs}B!3(LgLx&o_y
zRHA%$HM%;zL+9%sapR^h*xUC9-U@ER_z(YZ!bmB4cuba_^;M+hx5Ur<Mh)6HN{2EA
z8IY}y3ANp}ptG}kQb~7TlFYOtMJIc@UnATv+nM?pInt40XR=u3N(-&0Q1K@Z^0?qd
zT2Vf9Yr8LLY*<QLxBJnOsCCr)lE~vU?Vw)?0TeNA54m>kC9{GM>fRPgZ=Ay^%`l4W
ze#Ov|mxriF$q|AJNwhgHmD2MvXw!u(I(G8}RaKs(<4t*#pmL6E2V9`rGcM8P{|d?N
z)HOO=E$&rHw`l_0qdkImd?e-(ExlSn*WW#(mEAAMR{s^<=v70FwzagZ&j*q>sHgbO
zM*8#UE1AbOlk|iZD*D<=+x**T=f4i>vHUMJUhks8ffAf4ILPOZNwHV9G#^Qq;m`<K
zcHbZ;IDGQ#V5-2ibqc&WU6E(aRN@XPWiC4|_B+lhtnpffWBpV`2T7G3cB%2cE;Vl7
zq0YU2t8?Be4YsM%U<Xf4UhzPaJx6G9*cmN8Y^}|mN3?mUrVbY$&|w96UEUU~%j*<{
zD=SiuCmQQ>agIK}8)?8bFAVs@CPUT`_pa?XjCjBrV;*W{!lBhBqPK0z?X%6e!OEOp
zHw$;@4GSI-Z^^SZSn(709=v&&HLIBQ<k^b7*s@*x`rMoK-uB^3&--$5c|U$rX2T^9
zY}xWbe@+*9<*%g!IQ!v1-tl-4OIL{ePK7<^Rt#Y|;S>!JUG(#JhV$TS4jd`Qgn37i
z2SyS{Z)Mi?9>EiEBwy1S#d04<@wqdj*>jVac?OT+_w{4AGuDaMz*v6!Y%Dh~aptkj
z&b)usI9^#hj#FI6^UD+Cx$@6=_H=UL-v7C<-f0)^eBvVRf-XEv?1LnHqT%JrXec-s
z4IfLQVU>OiY!8ot*nzQPryB=5H4lN}oWqdwAs*n2=s#Rbf(ngPm>_aFdV4aVf7Wqm
zRy`@cyYir1xL(%1EP(6wh2WA`48I58f_*vnAXV}q<l9$3wpS(8o~nYl12yn^<_B1?
z`U^}jYz42~U2xb#R`@hkFk_Q8?!IG$6S6Jwc}s7M%j=KL9fQ%}$8hYG!>B&Y5%Z2Z
z<5_7}O!src?7!~Fm14%x@WEukd=0w33@4kcL8tSZ@X^Yh*nMa>mbV7stBE1_Dlrrl
zK853Vn`mrW9fuvK<1w>25&PJs;>ML3IP!QFs@I;tDF&x-q7ZwG*?A6E92dFcvP-D<
zwGb6$icw4FChjr5jYWp{aFALl&ToH+H809>;wj-`-1Hpp4X(oZA6{Ym{u*4-qZSj+
z)ZyAepYVBB171@7f)dNWiT7DE`gQ!kS$3^BX6A3~UfYg~{X5ag|1Zv8+l8b0Nsyhc
zq+s_-l2)Y@UCfrIBb#L?ltc%sU6$URlA}5=c^dXtp0W=qkWwE-vJ;G*Pj*VAn5jgo
z)s^YTT4h>$PnjMHr_47m6)K8Sp?=p@sP6|AI`>b7ZmFo!0x=WCD67)4KPvR|oeE99
zszNbgD%3b#MKFrR_u3t0`YisO;abWRoTEgVs6>W06sbncSeXS1bhf9srG?2;(Kk7o
zJYJ5R6J)9BlMF@nmk}LwX{yeaBK3EYG)+^I>PASAd|4O1*8YcAJpQ2C!44c#)P|Sq
ze_@4kE7taH!I5LW<FFY`sIl-XYWjV~ZK4OYd+jG&;P*jvz~AF?uXp&#r3OC^eT{p}
ztFcP<CAK$I;+K0*@j_Mwdi?hougrLeFZz^XZ2LWYe(MhY7j_FbyWhl2?d#Z8UW6eB
z3USux%b3w3oS5nPDB*SvjoVM-m9$(u?Q#;|d_RG93CD1^b0#MIOh>IOvE!SajH|R0
z(Wg8fA4DF)q0?jWpJ6nvtBt@{*<l#5UU(33zhJTL!^V3-_#`Y4jiv?Q@}4_!PSzG2
zZ?yqKlU8Gr&I*(YU5p+Q3vlozZ=Ctr6A#Syz`~bq*d-V@N1i&P@+?PG{lpl%(*f_8
z4Z)U21JFOhMzH#N3;u8qv=trX`HPG(C0;Co8?~{Arv~16uY%?W74ew3m3R)7#`A+E
za9>m>jQHFNZ<ja27sb!wd*cHLz5%$+t%d`(m2l<DW7r$<0N$$Ih8wGj!Mp4-7_ZES
z$^)li{W!tG3Of!eoC#q)(?Ll;1>)`|K>4jhFl9v)Y+88$W<~FT=Q=y#^5YFqadZ_F
z#xIBZ1xsP!fh90Waw(jBvJ3*-Rzio@T7Vv#VB-C)u+eop{9d{Z7CCPQ!};sr)BTm8
z{CEk(ty~EGI~G963ST(AX%QTHxD-BiuY{HzYeD|TM(|SJ3TwBDJJ`CN@N<?wRL=|m
zwMBt&%Pbr=V6@20L_^xWXc*mn2qIo5!OLOsurN&Qn0`mVlwMJwtri2ROX48aARfGo
z6Jhqs6ws2-0NZa_kZqg;<0JCG-swCv4ZH+vrxt-{{7o?Ix(mHGKZN$5Phk3rmoU2Q
z4O~#KgZ=gm;5gwMT=Q>%4C8hfw5$uB-Il^%MWO>LrHnJzs-sN3HeQ`#fPrZyXjx@}
zhF#Y9N~f=IDfGwgK7;Uu%}~5x1E|$^1nOEl;y)AdJkyzg%T*`i0p)49MQ;Y$!YrH~
z=!5&EeevGoCAgqtC00bNLkZy-mhS71cMk`m+>Bs+;(P!leZxftIT}+s4q^17L|j*$
zif^Z9;`v7>aO2oq%*q!tj(Gvrgotmo)@zuzyaea`xQhXCWr7J-fh)b9qfO8&bouxW
zO?Q66beFF<ZDI==?`_B6-(BeTSc)7+$cej<BH1obr79y$^6b>1e@ce5+|87}U9_ZE
z_PuCien0wPKY+T=*;9eN1MRrPw9Uzp0-igIS;m!C8BL+AhaQv^CT5&DK9n`gm!4TH
zC0Qds3bI;9!MK?Q`tG3nlLE-=dmt4j?xhyD5PGi|N?Xdp=+L1^GG7)$Uq>IJCl*I2
zSw4xhf2Gj(hIIP$K9dsPAE)g49Kle{r7zuQNLDMKoO)lRuOqL}Fs~vSxBfcW9W0@r
znRjS^;eC4fsEmr<JSO$}CzRM&NdrH;q=d>>B5znj!{ckoV(AB()~B9sS2xnWU0(&S
zteHY0e~?S}Px7AgTkw6_DZZ+cwEq01Acb!F9w5Q-8zkAyTZ)I1GzVyitj=p09-1o4
zBWKBRt&%*4<jZsBOa(qI?pZd$ikzyW#D|iVINeg2-P4tMw5bY99#Y|cDyr-rpvo#g
zRr$*zHQx1HjVHONbIlcXK5wVN&Y2plXd-5xLz?_hTZ<!NwOGbbo42KFvz?s|&o9>D
z5i@kTxmhrmBlLLe5Pepz(dU+f2K;-pn0Y!4dEg}@cHCjieVt7BuDU7zdS@zjDQ4^y
zY0kE*Ex2NeC94g!;^U@0*hkTtzqMJjc4JQtdEJX8p7iDc_xtee8-00gVLyI-!G_PA
zvE`VP{khLEJJw7az+dAB^32FV{AI7;4sNsO*5yO^u;)-Nb{xhM!l~7y=)k7+4lLFo
z92$lE(?{?IdoXL)G5<*z!DA+l<lvtpd2!4r4(d0WLoSVGqY;kWaM6*|`i$ZB@G*Si
z^BDFcCsx?)#Lk6I{G-u{KdFpmS?jUvFkmeA9v%UvE)k%=Dgs_-N5DntNca&D2?P5?
z!K3adn4}s55;J0<vnmdTrW}Tk8Asq<Ln3@xEAo4k4!W+Ha7pwM#tb+K+LEW?pwf9*
z;dl{Z4j01bKgICc`xeyX-UFrYWl%;HaPM{{tf{Jong{P7w7wonS9}MnjhztKC4~Wt
z713d!20BaY<F*h}Tz$9)77y%;Q<jR{!Bl(vESU6T_6hZQ%V<oS?u?c<Cg2oz!OZI6
zfwx?|#5cPS%4ROa&^ya<lKEO}jo*x^=DRTT$!?Lc+>0w+L(t;CP_#P}j<GdSsG}T*
z7Q^DPeqJI@4-wzzr_-_aX(mbwCugnBNxVBC4}UqI#gjAhvG)SO&02H?!{-;F<@D?L
zXk-a0Tiih(;U}7Xs}wVXAK|wV71+}F6#Wl8$A#uqxc$T{Jl3-YOAo!nia+l$WBNzz
zm0pkIYXwVJ=PUjoL64mI9gi+-!9NRIQOEN)7H~Tr)9J+Ynm@Q8;U69u-;Mv)OVDAI
zq&31@?sZOztbL^^Q&I5L#jQtSl&rY*$ddjvIa(|F=NpCV^WQOfs+Uy|&kzOj6`7af
zdIi!pS0uMdinL^nB25obB;CV`l$WSTy$>l;c(5Y<@>8UZ6BMbhi6Ra8s6baU6@(X0
zfqu#>kn}0xcp5KHrqyzkyhx5T>xG9xaDCoClcC$<Mt?F(nqEswQ}P@s8j>qX4T7Wb
zZ=?jR=@wm#&EgyIgXlfEbfR}=JKk;kjSHN9VeG!2c<0s+kp*wYESn~jnfVp(Z2yeI
z6B;lj{}YaU@BwGMdXFXz@9^ID8a(jpHHQ7J#*Uva@yEC481>;9et+=<E$@}1-GxVJ
zl3Iq1q6Ztg{2u<Aa0j(}3wNr-O~I`%#yi<XShKYdJ;q+f0V)?UtX%ACqR!#vDQD10
zIS(J)I*Gq`W@E^Z<LLHP+@i8F@WQ+_d}NY>O|KJiZQ2psy!0?O4T{6)?r8k^I8tP*
z!ZBy_LA*Kf01mPY!Ii)F;=c!iH=D8>6Knl(;-2j&Z?zfU7Og{N&sF&L+cGQ-T!d5g
z=4021Ik=rX#rL#3miW5igpLVl6yuCr2RY)CO2#L<9dOF<A?V&d0HyBQU{Y3Z^f}Z6
zHz%6o_X1-~|E!OO4muc_rh!$~s_1iD5w!y4aLa6IjGHWh^7)<c<WDQet!{>CvY#Qj
z@B?Tccney~tHF9|B}fh{{~tx?9oF;vM)7tM4WUpZl4S4u+>c!-$w;=cH`#oRjA%(|
zS%pYSh)Sg*Qj$?96{%3sK!p@pzvuTy*Za!#$yHr_-tYT4_c^aaFLfT%xx)LDaOf_L
zk{P1O594X`<?Gb%$7PDoh@e|)7wBAQD4kq$n(T+2qJH;|(f8~A^eW^CP1^5Hfjjn7
z^qn2_IeP<{<gKP}x0h4KjHUEu;u1RUvV=-%m(rDkD`-rMHS{ZEJ>{I<NHa4xQCRav
znl;mz9!^?Ao;Q}0lmB9}zve_)OC4!uydy1Jw~$iDFCnec<#Z-_HN`AjPxcEo(#6x8
z>6+tK%73wqUK#D60|Wk}esaz!9Oh5MT>R-tj6b!`@u!?Xd0y-8P4@i{ll{HJba~4W
z`hDD&uD9~1yAK1&@5ymG)i#)JpF2YvH=ifragppxqG;9nYcex+gNiJYXvf4<Qaz9^
z`G#54&^3pqZOWrD@h@pxMF~w-lD=%?D*ChhGqrF1jUHM5q>V<>Z~RFaoBygKXjm((
z-=hQnTWw+b(EtP0jnG}MD_*rRg<?x{?D^Fjw$=S{`Nd$&N*#{Z7X=!8Mx)gZOB61h
zh`)1e@xXDa?5dxE1L?CRi)jHWRTiV(cm)*Buf;LHP3YdRLwe%(V!%OHSj}+9=E;Yl
zveFl}Apz)6ejFu3Lh!~X6o;!X;OW>)SaR~J{QHYR_SgiR4oSvI#XAW4Pi7}tXF>1&
z6R1Z%M@e!4v=oZrv*R7IyH~<V^|RD~zCm;OPuzb04;^hK`|q|I51!bPb?RF4(Ij1N
zh;GLRo_63{x}AC03KPDT(SzT0x8O>*e*FI9AU@+ToS)wptUYoJUza*g(Igx8EVttz
zcYD4*d^T@tn$I&c7IDhi<^0)mjpSKu;2npzvU=D~e)e!LSL{6?xfF+Z^AmS2-Q&eU
zLyz$GMqgfV+n)^&2k}0q6KscI9^2^*Ybl-MIkjPY`^`np%)i86o?hXr+1GehPAv2D
zcs`bYll_ZQ*!E2-KP^k+jujc)wM;UPUq9kMPqX>x?Pt6&ERV<UD_~>$SNyzVF;6Hc
z<sAp#@tW=xJS6@DD-5k>jf)@of=Vrqm{u=6K41A*Yy(GUHu8=KKe+5hGym}U&E*b%
z`HPBzh{{(GGfpdtp>p0i*G5^?-&GbKD^x^*nyRpqGtYXNJ5RW+Cazkli_%PWQDE6Z
znBHk2<_*;l8kaT19ix__m2XS2`(I0O!dX*%cq6?%Q(FoB^j6~I(AMI`mDYkx<?Iux
zC4P3+7JoyvMPUyek#Su|{28e$VxH>?ja6+#H&s2+{-U0!o77e$G`1D1FSZk^^V*9Y
zZS}>qTz%m${XGf`I*7ua9YyVrj-o?$C($$9P*m<S5?vN{7RSbR5n4Ts#RQ$M;$L%D
zQD12yo)>fzvmSI8Lz7H};gufZ=2<flAJ9`IyPFH=UA@G%^%i2&LYdK;)<<NF>MLgS
z?k6U<?=L3)=`Tv&4iFda4itkz28nK)2aA?gL&RR)p`z2vp<>60VPe4a;i61Mc1m0)
zaoHY1y$<4oWDnM93X$V2qhTr|#Iubf#Jv0w!l&;@VY+9ecz<`KIR0~_XlXV|w6h*1
zPC1PduB%6h^48vT!q}U>PxK~RFK@D{_NJ9<j!<BGAKFy!LyCWWX~H-^5>nf#-Wy0^
zJCD)48z(5YPcU7qJWV~yLg}-i%#a_8pcosuuOD=c^sM8^!bQ$Fk`K1L&0XRJ8B`^8
z97nxu^3Qlq-~SX)xM~UM7*>$S#!tjXjnv9Z0SkTAU{&1;+o$Lu-lYQ~rguTw{qC?U
zH%D$tUz~6n1fP;&7-lKp8#WqIT2@F2wua^+I~;VDew-ZX(frR5>y|Ht{-M<}U$X%L
zy4xWZ%U<Tf{do8DAZp~Us&?>UoUrr3*5!V9y*m(9e#dbz{1nzCp2naDp_ug|3@E*b
zffbizkKPsdy}O3Pg|Tuq6)#!FH?hel1y`1(;?bZq%=nRxcd?R%zT_byT0DVWSPlvX
zK8JQ<9&-B?;BUxFZ2MazXQC3kKm7(#uis%-YuV>J@B@BKs=}4oHMl?f6FS(|V*9{)
zL}`A-_QG#C6Z9Rgt(s6#`2%ZqHN#lpH=?)xkv{6b_+y~J%C{6abEYC2R4ek+O-g*{
zn-Zt3QRZmr*IguMj+-GWe5_7|ONXoS&K;`k5Ut9Ki&Q!Aw<>>VtH!gssd0^&8b_F@
zafqH87d6Y*y_8+(SENT~yWES(uN(M9g=0fic-I6KzV}|4Q)Q+iN<*3N`6{uMmJ)A0
zq{so^CHK@(fo)v>VXXLz!PUPJ?)3|XGKbOW>JN15*Myje@3I4`0ah+wF`z{5od(xo
z!@<uulKm0pIyDHktpd4z!1Nmx==Zh^`s(jsWBvxJ6H2gu;cL`yeT4(=g;*SvkHoWi
zICt?mMn*lwg{W*)L_CJ(`G+t%`2Zb{+{dpy>4;yGhG8@BVAF`(nASA~4-}K|s5Aj{
zZr;FtpI9tjbR7c*M1%icMpf1&r1(W(+MICAHV(s)%1}gKlm6lVLa^TI6neBhf$H}`
zc$pl4_cDi<zs?sECLNKz%aYfp>4}q{-B9xMAXH){cPU^Wo>c9Yp06GFzH|!)%Ne_}
zU=7B4uYmukB^dq15f9GIMf|jxh*F%6xv`Vsx6~TLI#^-e^U)aWBVaadI39KyjMsJj
zVD;1jt!|mY>Si~nWp=?#x$|h>*#OOJ+G0bl4uYn)MxW0b2s*EZJDZfze}MuN?lnt3
zX9KlxtfN~s)%5CA1>IQrhJH*gqD$lQX^7-+299}5g_KFhyQfi4g<GVUpFqEpW5~Sl
z3O%unq`AE>P|W&Jx+{ATY-3JJ|M4-hEAl7Tr#|GZ>P7zk((}8(g)*AA(8P7?$RT_M
zIruH1C#s9c;gJ(%-f^Pz+=cX|Y%z_xyo|;KtfZBht106A8gi;yO(j=W(v#`S$iib0
z?bLUqffwe{%KdXmq|BxBedd#Mup=2<TSU%E%gD0tN~+IYO)Xx_{P^7U^l_Fm#T|2|
zzA5YIPQn5z9^g-k&i<4V>Q9x4{<Q0;Kh61fnC_<^q+3hf>0q=cg|_h~&2S$&v(ArB
zZwRF8*N>Bf<RX9kbcPOBou^Nd@ioc!GHoAlo$CJGptj$VXsOa2y4gL0f~P*DkGpfI
z%lSOIdFv$&dtO4{>&xi7s3Na-pUJmH1KrAMrde8wX#T1Km)|Wgs$FYb>a7bWYKJ+)
zJHWM<5w^AID!E$SaW$tWhDG#7!=C=gw;zlo^WnIoAdpft3W>?%a3FjF2ArRS70FXD
zqS+o1&a*Myd;tazSq#0iD=^e~9b7MMM%&3dVG*?t@8%wq-a!w^&Gwe+$5G683WU+o
z69~u(!Bv%WxG^RiM_ex9S;|$skj(ara>l6+N`~j}RCIQcov|ia*i-ui-zuL&OLnmJ
zSW^tergxYoncvA5KjYxzZx}51=+kB@uuGgWzcp89ms>5_V}lk4*tFrUHtqS-ijHh@
zvNK1$Fk$a@W;|xD1@{Z<$A`WQ;`8H&bAFiM$t}llyIoe?w#k}x_u6rSo;@FppCx<C
z=X0pZBIyZS#;>wgb8(C_AG@@LTV4N;2j6$$-3|LW_{c#n@8izH^F8@LS8v`w+?P!n
z{J2*_ATMw^&O2;Rv4!zz?(*}jWQLySo=M?sb2gG6cwFXxTcX)=NemZGzrmlzCUW?|
zWY+I`n<urt%cC0a@rR=O+%Gwc6$2j2p1>R~?){t-Kj!hhvjyBZ;T6v+f6d=Fl=7Fl
zx4he_oJ&*R^M=+{e9@*xGBQ3f`_^)EP(2%YedWM44ZH@8e7~`Y%R_&1C;G*&@BHSd
zA%D4R@IT(MLqUA%q$nJ2DvIuQN+PF3NjzVnETTUti;rtngjThRIJZbu3@B6;6K&MQ
zrW7?X)Iwbtol+N<TeJ`hx3>^CKeQ0NXK9E&4>d%O5iP}@xR%1Nx28yp)D$Cowi3>l
zTM3VWt;L<&t%c=eEivMqmh39f7C+nTi1M2{!plil)VFLSv{KrLm(F?u=53{h(pD_F
z)K1hmw-?@$SlU^`Kx}$#ARJ;khybsS;+ym_W4xg-?`b6VO8;{6*Un=4>n>vCePglp
zdRK8V#6-Ayb`we4WVh%NQ&BjzhtM8jCcmGaLamLt&}cRn@89$ilx`tH!+MK^eSJjx
zxqZdFe*J{8LVxl1aev|GKR|4?A1Eeh4H6Oe28lr%1`A)w;&F-}BCKtPil!pjCAw&s
zm{vMW=Cg*2&$ot)n?@vdZjpSRyCmKyL1+VF%5sS9?hwmmhFv`pqTsCu{b=x@_Qsyn
zX_+UPWO-6+J1^4t?IoG)he^M}n;d)i(8o|;dg$Osi^c~~%;q54R(hQJx}T!&>rPXD
z|4?fGHH-#(MbO_Fm+731%w8^vBP*Xo+Mbm{Hm&Z`*M%80xg?ABUdkrJU%3?7>J{zP
zdrMJ6tLTQ?7y9@87rl>EhGwV++%Ib5rdvCB7<Yp48e?o(Yl_(Zy|D9sUrZT32vx_1
z!M2`p+hH^o7FpunQEQaC*`fHJJ$8(pgQve8@vLGgPJUdCHp&}eX}le?g!Dl~?T2;T
zL3B-)p5s(67|HB-Vfs<@%Lssa`Y{;ZJqh)c5ZJ_>g?;3C>^m8bo^Fu{U3VFn5)Hdv
zF~HwAtay|FYo8=ITHnH;U#VCemInVZau1T9iHD94F|*_`K2FWS!lY;LZ=Z)HEAx>M
zU5HK9udv&o7%5x|UHi9iax8=Ef(p!<`T^I6R$+mB*7q*?i1Lun7&k@o>Kf|t&Fd>-
z^cv83<U2P1Y($>J52T;}iBE5u(cJ1cPMH6}<{^J&U-3UYZKc4McPOy$YXzP&L~>)~
z{4pq7k##kdc>5?Np0!$u!w)O*{4gcnC7F3@2}-OguLsX7@i8wYp1MkjbL76gpM3qX
z97XQ!smPsJk^L)V$M7KquI{YBs~7&mL&<1fx#JHq*8j%5S-)^?U^B)l{=|v&CJfx!
zh(OctlFjiA(euAzadka%*VbW6<!2<%`GhtZHK^=SjgSMC2q=7yo;@mXZG9Q`#Jojm
zZ7DjMm0;W4*BIdW3T|<QvL7=a`M>heuVXIW4}Jzi+Z;4FJ%Q(jNBAl;-lIL_4$3zJ
z@d5Wx5qKA>0jX%?a|?eCB_ny~P0-o|Y?^xmPFAr{?Q<Plb))h0>t#%OehCAwMxe$m
z93LIS@L<R}7`Hl$(6SJGOFD(y{wLtM?ie201WGTCKhC#4iYecY;PmUm7@p~g8?o*<
z7J3MdzOL|f*$?N}F8FwJCmI8{p|kTQ7*AM_Qp43)TE7gXX^Z5!TV|x3=3qe|2dr(L
zilhg&_#QM7)ypigXZUC&Y6)aj4}<#SLHHim4<j#H;7XVox<_}zi3eTK_(i@KWN&r;
zzP7kfrGs}XTEp+329{k{L)=bfta4O<Swb^ie$zl@L+j}9P3a|GSV0NB-jIWC5hZKo
zlb_);>OA-{x!YyZf6LM+V9PCPyE~CCAB?5x=daT7FOl>`av*;03nkmE5Yn1*k~TF3
zQD4cM>8|NZyY&uJ_)#}H5xSptyZ%Q7MH^^t!YVTEzLf5FSV%p$%Xve0J}rMVk5t^|
zQyU|x$4ppAmkutbzyV8XWA;*VHeO1#SWIV}ov4q^e44ao4jIgzMJ?yeq^XBy($(9u
z$o|<JQra?~a;G^_knUpouzV>k8n~QRuUkRMzgLi0wu0uoTtX*)%%CL&Hl)5}0#$9A
zNIkmSl4a3MTEBTY+3C8_mDaBGr0ft)Kj=ZN1|6nZYCfc*;YWp|1L?DzagNRnrmZ8+
zQsJyH+I=yCp2`foYW8)Sej=Vs_a@W!|L)MBy&3e$;~^RP<;XK!9`!r(l8SDYkh;`y
zsza*i>7`mS#&=4!{!P2zD#AZi6|1gl;O&{#80V#nrZw$QFs=i3>KQ?)MBdBJcgL<}
zJ+acPHyo<_Veyqg_`7i!CJDw-tx<^hJXU)1#^ZjS4YGPnLAjqjT91@GjA3&B?y(p?
z!&f59W<8FiZb4h$UC^1kAN!>?{if0btqZ-;@6}OMRRtnM@gyR<oW>=Y_lsH*jxnB>
za4haB96rRryGH`%?M%kHs#J{Fo{sswv+zwb8;J(Fu$)nd;5)@|n^%U&o>iD+R*QkN
z8jzLPjO{Y_Iap1FeO%Rfps^<ZDAnR`GDo=~xjlD#*paW4cV<`BZv3pb8E>9r!QH(3
zarON{T&O;rzfTw3H*z%3YGK8uTdaBbr^(!Q<8<z+Jd2;4pU2~;F61oDWxP3m70<c4
zo?QYq^Go*~Ji|-&j!)jte#Nfrxyg+OYJ2jM*u(r~o)1TAO5br>0H5A{jIBqU<YuK1
zo|JutvxCpE;`$3be_RAxc8X%H?^jsq*)^VXIhIGb#dD6+O*R~s!u>T<xrNMFoeodu
zidC7M)8nDcTs>w-w;cZ7<2f7L%Hsmd0=BvPlF#;e&D!22Tvhajk8~*GtrIGEz|s#=
z53k}ED{I((>L<=Ot>vZFb=)oZ3)}HG9{#9-Z6-F#`)m{c)%(e>7B}<suwUH2;5Xa+
z{=;tV|8aF21*xAXh`V}<vZGQ_#HuKX{X3LIWwnx6HAh)?mn(~BgH=R=^z|fbsfro9
zWoD~JRcOys6KdINV#FAAp?y<bJR8tLWJk9UHvKfj!8i>O!!5<%2XgkAttmV{Yl<Ik
ztweqI)?&wl*1~VSmKf4eTg-W+EoxkJMBXr6(bS+Th9<NTYj)|0b(U>~wr)FN{H~pF
zzR_N!d+CdcB?jWys1D+UQAZK=r=xIx(@9j_GZcp+jD+ry&SK=|F5>na*)KY}t60>-
zM7-AQCZ^PP6N~e@i<>u0#kYVS;>HFuamq$^jCM5_(;Lji!t`FEm5=P3u<tFB+V+wC
zn0>_bLw&{S!Tm%<aX+!lrN0O=7$7=dmLBFI14XUOZ5<mhNX$AvNK|SJ7XMZZ77K0-
z7O84O#E0=i#Od8bL|nuW`Ew5u^Gb$@gi85w?GXLRI7GHz57AJ#(fm+1(l&ReFE8E6
z`-TVox#LMGf4ykWW^b}K_8}j&qhv303%0J(pVLwLk84lLStx|MjteD?yJ7TlT?D<e
zlAU&=uhF3yaa17hpAI)uC`|h<4Lp%SZMHt7u)G|}+j&8IU%V!xd*!5%^oiz2HPYl8
z3ixKP26I)Z@rLLk`cHd!bT*XF&aOBeZHoK>z3}B&U!1KOhztzFn_~<Y?a>%<!xB&B
zj1wGXhyDsPu*+)>u8(xWJe_55|F#-OOE*CI@isWd%d@rJek6`Mh;#ki(W<i-^fl#K
zywMkBZ~Za!VGtHvJAq<<>B-rA22oSaA>Hf(vYRi$OrG-x_+G)3DcA5#Jr)5{3rJX)
zh?;*%$PT(CGdXu)eL?Cuh8d{!%tYm<EVQ+HjOqT_*#F`w4ru0rhrPhjSp`_O_9ael
zmAQq@#fV>8ib*za5!0;<N%b=49P=JUODkngy&CT>)F7e%CnQCG#@Wtw=(w*QMY&(#
z)BYRY+ce;a^LGqB*oea(O|aVi0|rh~OYYx{FZIpXarPHF3aO=J{zl!%KPb5H2P1#{
zfgb&Z?TWwnaNsYh5C6q7dEK=9F9r?&ix*9Q;2!=5USs~i_{nb^xB88-(Z8@;?H4ZE
zHKT|7PprEB18*NS;a@=`M&*A;hs*|OMoFgO{;#l@`~?oJ>*0H^7Bkj;MrW;0SRO8W
z5(ZUcbW$bahJL`3D;2oeu^f5_-XX8-4W^7Jm5ifeJbqh*<-K0v`}#up^X8*o?)hFA
z=OWnt8MYqC!O^Qv&^BN8UjEENWrqjYFzh~FPD{tYRcY|pa|h)<w=wQ?3L+wtpb?XZ
zg$Yt?yBUXji7~hncMZp`T*Z|0m!S}J36Bm&pnPLEj?E9lD64bQr+XH&+ens8(<ykD
zoWNH(>qcG<MBgBPRP8&8j?O+PbMgl4z0fq#15Zc0VHpl$(vSnt9k36cb9Y10eg_I|
zwqVbg4e%Vg7Rlx-(b#b*#<yIE+@^VWQ$7oibL^p(GzA^PCL#F91dRA^95R-Vl74wc
zuSvtu-*ON(j_ZecHWqN5Ylf3syJ5-cE{M(Rgy|g(pt!j$7QE5H%Y{-O_@RNc2sNay
zSH@4dlM450rqsv=I@eH3O>3$tNwb26JS(Lx@vrDi{0o}*<SB)Idqgg#nbd238b$iw
zBEwAC-Q5&R8~R66|9zKezGUG2ObI2Q`ymwH?j#ij2gzKtKkX~_p$}6IQ`RPTDjeV{
zIemL5d(#%W=e(AV#xA43vlf!~g876ib4a1vY&sY=i}bW-lQZVft~YZ@*?&I8K3_l?
z!yM^l-~!tBeIC6#J%^V6ok@YmXV8>_=~Ve^8lCMlo!)k{r?u5HsG!d*(u$ZPvt9FP
zLpw)O4R@j?p$o~VU?Ek+JJCU@`FuY-jWTarQ%Q{#)kKY_rni$wIl+O(4PHVmcW<Fv
z*)pTm<pAk@J4oM>-04@47mYi9gvMkZrNd@{H1YOvdgvZZ4-cKCz#C!IODB?62V9}4
z!(-&0E}mLFNhY1EcgQLzL-Ls)(gEijN?MUezqh}n{fQ;Ccx*Y{8&FN^0d-WjxRGA@
z{Gok&l+eRU4O7~*gl0)=>^i55RL6FRY}Wy44-9dAn=u-?bw}|tGn{j=fPde9m|r#!
z3qpqCuM^O2;7C|>7>f|2@mOhXgQN&Mw2-W0WAoW)xo`ox)-A@`S1U2Cvom6owqost
z-3YANk1KN4)^zs7`l&~tG};drC<q2Pi7}&2!*1d^Ja-63`l?IV?tT^9;$qO}TRdEB
zlhOTtDvTGW!@*4Qp!BmbVQ?-QTnaJ&M={LKl_C1SDty{siyKJ|aOnCA)+GuYZmGi4
z9;)-NwVJ%QuQpH7ki9aE?b-fIN8Z-ZnVtTcu%DJ0uQIjZ^W*w)$JK-Q`pIEDD35uu
z{%EdPZpoAGP2{${CUeA@X*|n#Ca<_Qm(^^Y*r9O=FOOcyyk#9{PuRq2=G*y!{%-!}
zxsQ)&x$>2hhgiqLg9Gn*vD(5TY@vLVQ!o3=c`As%X`bNM_fB!@9;xMsP~P_EJoiZp
z=d*5++;Q?{F4l`?gQDx)a3+qu7bS4LaS~e;r|@Y1RPHx6jlb8Vv)|!NR_*qXXI+2H
zCkE#5o5*K8y-gmQuFvP-ltOOV{EA=qEN1Tsr5x+<R?cl@+;LO||LE|6UCS#uBe<H2
z$9`m$SD)D1v6f%v)$vmL!nb_C^8Uhayj1r)n~rQ`f5#?v-~5BC_WYEbj%JSD_KQ_k
z{N@W&|FF;Cznr1>k0;jsW6h-sBIv1txD7?|Gh9)m>MDtyE=uC~CncG;QWh>b%HrQx
z6)`zkMZ^r2JFi$(ad(iKD7vX8s^#pX`&3;xFKi(?HntF5j%o;xzAc5r^Okak(iBSF
zTZyc^R^pa>Yq>+x68C;-3BMF=p}AK_sEpSYQ*_#hn<Z^Tt7tu;x4*3jnAuKD?kjsm
zTj-0;<@%B_XdqUe=^z&D=_o$V?<DRCLlJ0bB(#1TiK3#;;=;`?qHU1u8FlU|c1<)9
zDV@8CL0`LxL-%Ap-Pcs`>>i@b&`eYnn~A#PJw@<jbK(2PTx^QyC2DLe#JNffF=kJ1
zu|u<uX!PqN*6Q>XK5o+6(;)lRXZI7`ZuAp9)%uIQcKyXi@BU(NT7O~ovA;OgQueHO
z86ftX4-gY~9-z%f573r72gpyum6SHQ(y8CBwCCzUO7%EIhW>8!_=!6$Vo!SU#*2=n
zc+(ZBQ)e3b5r+lP$Hm7;Yu-t!^b8@7PiN`8OBkI-1Q~Rc3_$&Bbh%F)jdV(+&haT^
zKJG64GP_R(UXN(1|1+vsT1bf<-_X#MN}4|G3!Pg2i+og+QM<AQF5PO4v6XG`x<nu2
zq)*R(a97lanWE;0IW!~(<mZ8b@XZ|xb9wGj&Xx=`UrT&&vqoN=9eU`@z<2*S$Q|GW
z&$6XBAF>*Tb2ngt{x*c>?!vkI`*1we6{B{#A$YncLd?C<_MZ=K<@#aBg+R32d>n}*
zPvNTCY0P+Z7HvGw!)9zamVb{x-nl5KO}L6Z71wa@zgVnPi-)$Ce4fglBAu<bQ1Uue
zp0U!T?vjqTx%bguCkts~A7R_-Cm7(KgZvZEFyU-2R-AePed*H~xc(&q#uq_HuNb@X
zN^su&4ea{8!-Pj=u$^52qgOK1U|R`|*edvG)Zp&)kFY-S3B@-*qfdS<KE12M)6#kz
ze*6VR5nr)m^Ea4E-*RMO10KuyXJMn%Qr0xWDyI>1x-?<6V-waNZo-m_O_+L3&PDP%
zSAHHN|9!J@6LOz5LSut`?ceWcwOeXN)eWc^-+<J>Z&;G|6|DRfH@bg;7+x>46?L#1
zR*U=HKEq7q6F$AHkzEGW@LE)b>FwnXKkYqCmsFtL?=qbBeutjzq<`^PDc)$8Am_kq
zSk}G535S<3k}RRly7~C9JP*l<&rzf-`5cpS@X6;1Y;qnUTjL=@$2`ENt@rWbVmh|J
zNJIVKJ8<ZcikmjK(0#4UczfN%kc$Z@Nxgw%FJdvL@;bUUMdO3=Ra|U+8Oz&VLO+8D
zbm$O{iL!I~YnyXeAoZFG<<t1`^AwcoPT)-WG0b=wh}dj@h>W9Hamxoj3ErrS^+H69
z2V!E~P#u2|x+w=xkhTw31NLA+_<#7CxD|1aHsW9LI?SkDg(-iRA)(b`tnc86eck6m
zr|(QG8af@1M%W?A(gv}U#>3ivEPBl!34;|t{g$D)>L&R%r~6`gaxW-IHr$O?-C$_d
z1#dk&A*NU#BSy7_Nt_PeP-{GTp#jB1YFIK!83%hRz?Pb6{geg@_OGQWiq$mdWI279
zRZ5=piYDL%ML9gB0=GwWIqN?CXqQGMt8USMcN0m|AdbRh_OYn&5-oBFCy(Eu<SqTV
z-oH;!^6VfoFZZMNZa&m#dYFb>aHoX#uGIhDUMg9#okkCJriA+|sATmba-TS#_AQ<*
z*`E${NO1<mPqC*7o9(IV=oyr<#eo9iXHrw}EV9d=MXH!Z$$1XcvdEr-BBoJR*c56u
zv!lyvCzHdW$&|Isj)vAup%W$3=!U`!s%bNm9>1PNrw7fE{pEA$?B_XDePlK{q|Km>
zCnuBI#R(LWV@YctSka9OHq>sWJzdauBA2MObnVYp`uK7;Wh~iGXT?F<*x8-#8A@$>
z>=7~zJW6%#0?8)pIIW8crmo4dSK?b3Ew_&(i@Ynec3lj0Gfbdkuac?dxjR(qoI$D+
z9@3-UIn=Lj9vydlN&YWN=xJEF>=CY}z0yaxYibjX-1L{qUn^m(n;LWmw8Y|<tx>X5
z7xxU>!7$DMBPSVRUs)HJ@9c(ydS=kL-U~;k$unE?0K`TQf!8L&Lfa8&8aD=O7F!`T
z#u_HV4n~@?v#Q5z*aR*>`phK=Shq^P|DECJz74;3?ZN9A2e4Mz4ZkuxF(l{+mTvGv
zgnbaYjyQ?Vy-&lU>p97W3deiXO9&NL(RoD-f}-MK*eV&rkIGyX(h(>9IKEQHSw18e
z7T$&M(J8^Ndu0d?tAb~2EvD8tz~Au~OcfM)pJa@W`>xJ|Pik_=3_0VN=<)17?b)=f
zBd6wc=Joev@9ym${P<2UKL4mMdz2330l$ax`0mV3vqs6@;Bg%Ob^=EXv*nq=Q{`;p
zz)_xa_*7d*esN_n&!4`6%~jWOkNX>Vu>V#zUn@O1)Aw?N$pH>He~|Tiy7QDcPyRQ~
zTe5I`_{J7LZr?VLqY{s?s?$lnsu99QiDx)t`8ifFxWJ~5F7ki7FY(a6SJ<mGnooGg
z@Sb5ecv5Ks>+Vb9mz{3$<?E^JKO&8T)6%(+GkMYFES{(Rm>ZU7^QxGqtXlh=%?w}g
z2rl3!HZOUNO%Xq$V!mQf%AY>GVV4W<c=fb$e)hRS_ECJ`pWiEa<>G3Nx>LiJTA%ot
z-Dm!}zm~T~*0E-KJs-*a!Uct2`ElMip7lWPw&K3??-Pxz_+JxmobiLl_5R6g6`Q%_
zMKkY@_{Af4{^oC!|M0eMf7$=nUv5$Ok0;Di5G%74#Ne?~|4C95?*}Uht2iYwZiupI
zouVu<ELDVCu8LT>R8{=ftSasWstL<M>f%?vx+vb=LKK;4h&Kfq;<*g<2aeVhYLeZf
zkkm@d+1XmWAE_m_sA-GbXWHV?X&upRovt`PvW?JaqbCe1^@R8BwxVlLJK?jgz3>{Z
zFCNK0b%mb>qWF0S(JHE=7_q;T=rGGrOtLT%|NW6n!tBms#i=f0(^6xhW6@Pqe(owx
zT`>_)mv<Ar4ZDk(Pr8fm+e}4VryfFv{Kc^8W@2`^nV7t;r+8S`Q|w=DE-XsS#oh^W
z*LA6v&{MJyPo`N2(-RglUnTQg>b*rs|K4Kcl-|N^NpCT4-T!xCF0?G#g=)0+(zJ_v
z$!x?vdY!e88vol*J!TxB7js?d$I*kN`16qL5_YFDxx+GFC%t=@kI-e~qokhgPa)TW
z=t<@Y`l%g48{E&*v7TYHsrDk>d=W+ey^1C$nK{*%D)W~mDWsi#mx_!tsZWO|^zdOW
z%^F`s69USp^xQ|u+xSj~Oa76{GZlQO(7>A_Eewv(L%WFvIG$#R_MN&SWq~Qm1I+O=
zs}HO-24d=>p;%B1yg4umE#_EaxuZ39_}Jn7J9|`3nS)199bqtcDJE2}lJkf&+%&c#
z`1nrb=<dVyf&*B5>JZZBd*H0mVO%Xf0-uvdk!K%(KP`{JHT47v*9Bv#!5N&m9g1>!
z-oN=h90Bf;_@R3lTLZ7+tk!j@tHi>u>IRZ0Bx1(tB=o3G!QWn~h+TXa9gp5aySNM-
zc$^8#S6LWY{0P%u$nMUx983y(hR^@y!r1BsZmSd^@Ma;lFMNgmzl-2@xELd}N}+%7
z4W52_i>|US=6+B)Hs@EMRP6&&%`1^PLi%^cR-@<e8Z;Pv#FK9yu|DY&Lf3zWxn3;_
zFV<p@c^$sXo?v6?<9%;kkHPNsP>QRUonH0mUQiDoc@3AJdwJAjsjd9@SBHD&>Tr8d
z9ZtvB;+oW_$9sLo?UGN}Z1PFw**{{y{u;Cnsm9YwRj7`V9X68p)8O$QPnTC9kIJ!K
zsSLf--eUPS$<;6^Mfc2N>|692I^SPm=>C_Op;n0d2lFLsC=dVE=VHz4XV^IQDdt>|
z_g2LxD4g>MAFgKM(Vt8tTHi;JUpgEM(y&JVuFUkOBFy&|T0BU`@?SUcw09z=&Wpzv
zw>TtTi$PfaHQe}d6(3{<Zt&13oS7O4la&|oVD|-t_?$=5=}?S|I)e>2LU7>LDLlU?
znbr@ENj`lbW<B-C-`u01LLVF`_QvRPFJyi6!1P8pG%FpFJqxaIG1(8z)h=l7zY|+-
zZG&^wCOEWnhRfJBnB%+x>w}kIS*8=z>gPc&UZ6T^hU`w53LQ^dlwFyKwa+Zkw_!Ar
z3`QV+B;oDyA<*?7fRSl^u<g4!YWnp+^+pqv-0BR=Hl1*7vp!<V^zeG24zAX<g4a<E
zT$Oz$U0W+-Vg5hLNcl;X+282AW-ayEQbmSp<+Sup2`xMQlIkUgcYbmX9sTl<dW^hJ
zp22r%h^pL4ZBL}0wXw8wRWvRAc!{p>45u?D=jcIw2<6nv&I7X`D)jfGTc$qbkmE&f
zwz`w8%0b$`d>_?GUvJBm8%ZU973qIiOjeciDd6`k((gNimT#O!!(ygTr#w6Qkzhx+
z2TY*@>Qm`l%~ZPWK26Rz(<C2o8tu%PLcP1%NsgZ_T@JD#w;9%S`Nl-Lku{M<dRWt)
zd>a~h(3YYX*wN@-Qz<iX8mX5~CxvbHWNm3rn_f>Ni(__lV39RV`esS18^=<WsU?lQ
zJdw^XoJwZCbLrUqWi&2qJ!zfUN-N`b(Y)LH$myIbWu9`AeKVePp~;)3EjvnA{{@g!
z=5e~17)*Z8WXEUg3)H+alJ1vYp$)5I2r3D*^>Q+8TX2W+J7v(dPg(Te>ug%~BbP2X
zyrg8MQi@TpplOS1C{gVTC6qPMhO)n8(X50lIfDwFmUw$y=Ee<mQGTW^esnRwkI+u&
z)~ySkMVP?1e-F&LV~!^?`(Sl*f4qnujQanEqkgeK*G;3b|GFi5m|4T6bh6YIB_l`b
zI0_jHu=?r}tSeiEiwicOPJcUEYq(%Q-~pr!b%R5dC(ec+L94ZX_#t(i$66<$`7s1e
z&qCpnb^#jqWJa{$3ce`CV7F~N)Dx5N#VQp~e%-^?q6e@pe*$a$T=d^wh&{?B*mbK6
zThCXaeL^iB{AfV-+FuymQjurwP~mm|)Y&;ilPx5N%T4y?&ChAi4^DUFx-FeK&fbI_
z(StXd_u_+H`|_Y}16jA<P?mTn-m+pO7YB~z5jo@eo!%r?Tt0;#X4v!W{<AqHay~cp
zS;WeBm+>RV)!e70GppR&%=&IS*nH+5ZZz1>hrC_cN7aoBT|L-H{V=!pJ;Lic9p!sc
z%egWth;I}e=eZkBaqCv6dB~-++;hTtnVY!4rw&B0Z|5i;ee(+E+Fj#C?_yZZ`34s>
zCUC{}B)0mJ!a9pndC9}Oys$?)58iQ~4N@QQ^rnYAxZ4x%J|;&p=$~=>X}LUl><iB5
zR=}&j7qU*`D~?$CnuoV6;j-XTerfcUmmGb^Z@!gr%ESuZ>HD6ev*gUvP|2F@syWQO
zhCK#-WS>Ew*tgebUf-dX+bGuYva&jMNUi5*L0|aZs;{z%=Nl)sY~X{h8hC!>cW%3*
zk#|`)@eHFMJg4CYi>E*NOms8r2mIoJdw%npb$@u^(!X52=pSz!q#$CG6vT^hisIvQ
zMWMb_N%U)05-S3gMbtnQQTIYc^xUN?#+s@LQ_1eRaZp|4Pzy2bYYU+lts!dGwiJ<l
zHATm7nnEkNm9TYfErRW|#NAHXVpol}@J`kdF^6@<g#~SdQGY%0LaD9TliyZ6j*xq;
z|Jn<E8-1~{gMsK(X&^qub`XnpcN7E1brS08hQi~4p>RBCB>IhzI?oq5_nea*6t>1<
zmt-)T`gIkLdzlEMR1=XcyC-~~brS|tx(nmX?!sw^sqham6`OyVihI-K9CWsaI9J|7
z#CJ9mBc_@O)g5M{B*;vJUNRFy<K#X}c?XGsJIEz)2Q~Nkk1~{Z(vMENXxN_J)T6Hp
z?d!OY&QID;=~oWWjZp{bm8u){SMZ>D{k-UGfHytq?Mo)jeq`PpNXL4eAX|@M3erAH
zmu{V>NN>q<-y22Sjz`nRH?eelMIsHIaEpvjrO~t9lK-ckLmD>uG-!MYwXJ<mn@s9x
zPy7$kAVv9nR)c3jOMKX&jm=GZNS<MUQ^ySPHACvVRozk8${f4757PGzz>@MI2wn$t
zH5nx{GnUX)v&MZ3JA|yZhknXzM6`DV?O1{-#Ve6EdOhY{*o+-&{~>v~3s#ry$B?53
z;VRsrQRyieI^NRD=Zo)!elU~VtZ{9Q<5uKJv_J?#?w`R|n{)7a7>35t7m<2B5_=6U
zqjdXKc;#HfaKl(cExdv7-~=?wO!=;cWXaaQjkTtCaH4M-R`f~7t*-a6UGo8U)nwsf
z>LV1mKS7064kCX%Mbp{mILmoh|M&&^PANcGW+D6syh4(15enYE#)a-BD4kmhc71~_
zr{7}irFU3xu?+hH%du*61@gzfml@>`u#WwJvT2p*_O24|oT|{AT?IRfYCPCpjTv#(
zs4b~RV3XwkD%7CIk804HYJ|mC!+TdXx($%tn1U({SzCocUn}uuWhK0yeSnW-GmqW!
z9xY-jVEMKj3zW)nSHBFej3tYv^IOztzd?Fq3H0)cVIBTj&V@zj+2<9a%H=HMSAcht
zm%6wh50UF~q4W0{zPUd|gk}zQdp<#$;$z&|@({^o5700v6H(VQ&|WtkYu2Tq<i7mf
zbxcL(np-%MlngDEB&?j2h#S7~Xz@G_TeV}MG3h#<xJKiC(iJp*ih`5TB`8daK;hPK
zRGbb&uk>?>dw&+bs%KDW5`wr9r{FO21pci(2BiamxGnckry`G{Rf-S1v%TT-)(iU@
zJaA9L9ihgDU`MW~wBL`sw=URev<piYY=>^>W_+%2#!QQ~xUy*_<|Z#ik>VnxO<aIf
zpE)?4=OCXE)A4YQ9YzM*p#1fCv@jfp_VY(co)e=&`s53Y2czBc{@4`V8%51MG2PY_
z3KzP<Sh+Ja)^tSolJ=;ctB0rWwPCln6|#D2z@uChvDcNb)#M)~YyKpQ&fjS8+|M%0
zR7u5?%V?Ey32i7Xq`k$tbg?;`+=o1*rJfmN`SlLztxci#UlYi|HJ0=`MN>l3C5l`W
zPQ?c2$fqhq&J8DMi+mriw)Uej_m7bAQZKsS-kr`qawU~@`^e_qe>C~>CVI7K4Rw%N
zDe-avrS_doV|Uuq()&~BhKAH%hD@S$18gXv#F`%I+R)|WHbi|Uk&4eG+U7cm#(%Y;
z+Beqpdc#B-^me@LxU-^Pi!7;a(KzbyaUAXRwWQc+D~dIlKz(mdq+641Xj{Z2ni^(H
zB^@TyvV2=IJT!^6=S`#||18NXa||WyA5CR%$B=j1@pP@;mcpJm(5OXD^jP|yU5>0L
z>#JMn+^(HuY`l-wv~i_6V>fy|-IFY?c+;uAM`>VD0EJvVPEHqtsdL&{%KsTgrqXA9
zE$#|=^p2sDxOhsRm`npcr&8U;bW+%tMIR4k)2au#lr{AwP3T!l%C;3WGq#3S+kT;V
zjUP1Z^<TPMq=X9-)vzO81DVFHQRAh9#0JS?bkaxo<Bpico$=*bSM0Me#ntOQ(J;0*
zo_*+t+;f8vzH=Cgb}$ZK7=;(Amgs+BBJvMSmKo3KfXQs+l`epF*%H*4t%mcZ4KjPN
z9k<82;7a%bj3hU7dF2Up*CTKp;s?F@Kn#vMfi?R=U_Lh#vEweFcyuHlImqmU+jZ%I
zzahEXN$6a58#B(`lV_C&XnE=h7D$HaE-u8QC&jqoS%%6@RXFKci!DzZ&}8)s=Nc7w
z+aeV%d#lb{HfnN`wl+81Y{PTcv}c7L9r<jP5zmh6%3f}!e08xoTTSf4dBX<q^8rKn
z+EC)`@gsQe;xVl0WyKM<t$Ej<$sBGyod=ws$tLQudwhozw~+g)hkIADSG#qrpSV$a
zUbgXJ!(IIB&0fx)d4O--J1C#y?tCHAla*vYXYaE<oYKXQM}!6NK=Wg)8*_p!WyY#!
z#%b2E3zf|D^Q__&&Ic+Y_~Y^@UitnCcW}DKo%3QiX!H#ZxtPEXDoI>9JB9a!-DZdK
zI~=Khk26PQaI1NlT(L5XAFq7Gc5|Pw&WIeouk(zD<v(XjuRJd8m(SXn1^mtSC8uP(
z;w#d#W9(7Pdta3B1id$`Z1tAyR=?vs*D|*9FXv^)D){fw_x$d_2d-RO$)BfGaY65D
zKB!W|#f3F|FXAKXZ~eq6qd&8|W-U)EsO9(P>R7|Mo_CM@!u2{|S*!9ZZ%zBg@u3au
zaOgYdI5)D_!X~bn^@B&u{K*gJHFNsPUp!OJK4(w<;pp3c*{k{=AM;TVV|y!#<ZMMz
zyGco8c2X8*S<2$;b`_ywp(=*GQx%I(sfpP$)kSsN7NR`Ah0r~zA%tT~VQ#7^PD@YE
zoWxe5(}C9Fob)kY?4T{m-fN52u{xr7kFHR*ZX*Wh>4}cTdSZ20Tk&*dJMqJ!z3}<e
zUW}4_o*PRI#K}$_#L1iv!b$G8PMCEPU!HUl2I~yPZFM8z7h)tPTFC5HTxT(GNEcCX
zt&8Ap#=^+gSnxMvF->OIjl;T%<+WYK`~D_k!D<r`c*;cBr<sVAr6%I)7ZaiS%S34G
z*+^DbBuh(s6ZMMPM924Smj1deWZ$%v&WCNM+Vk?B*1U@}rN;C8?LK-Fe1Lj*9;ES?
z+-T2l4+=eSnEt_sY`goBzX+s}`;SZ3)+ri&@C+G@JWtz<E>d8RC`woyP33Q6WtVRv
zH8$O%sdw+uLX(GN+Upq&dRa*Gy1b<c6;-65@Rdrfeo?WN5&~bUVUO%>7<5$|oj>cL
zq@97>y%<7!q%orGy2EExPYgZL8>+H%aKzLh=vhuE3?7L!+s8>3?L_?8KN*uROox8%
zEF?&l+S8GXWgp!NtcY2QnCgvik-lR`hux@8--kQ%T#@tT5bhrGfQ9~HoRa6Cfc{5i
z2bDkCAPByR$KlrR6t<lULABg#tloGIM{~mPv)e`3Zj1yaL}BOGE66duhCfqcP#{^j
zy*=XbDku>*PbMKUAO&8ow=rtf9bB?Zll%B|bSloko8y^yHZcqLYaha2^5|N2$VPKW
z4*IrxhV}cOW6wLO?+kf?cDwV@N$zoK%L-wm^a>yKir}sPT7ECZvMa9`xsOZmCa@II
zc5k5m^9_;$-Xf^;J8VAr4g*!na6@{T%OlIA7Ep%vt;*5Hv>aFZl;f+sHn%QE`TH{D
zT`j}+Wzy@bCB20a?_>wpJAA(K7H2xVMaLcT|I8}IE6q~898&^AnX_o=@fxj97U9U5
zSNL(_C6YY~F?xN0%xCAr(onLU-{)ek%sSONK0~7RQ|KpUWA(fz&}@2yDc+Ki*eMGw
z!!yyz{63T-)1e}{!V8bz#ncveptbcjo|L2@d2F)G{NF^rqSR^T$4if79MYx#>D%n<
zponNh%HO#yUq=7^moP6Q0?$=0B5>>lT-b9SN(u5_{_zZwI-iEglwc_AIf-?V$8q#|
z5Q2XMAlt|vo5vi*`z1cIW7!*bFM2^qvWF9@-7vb<AvE=KMOTOYC~k5=?ZjQ!eR@0Q
ze&38?Q#atk)wM8ey$WsCFT?kzi!f@SBX$PP#n{G~u$yI%4T-W3w!JOVwoH_Nf6^b;
zcMOIe9)aje!gb4`$T&X`o=W|2W2psn9+_d^pzcsQZ;Z?Bjd0n&1FZDhBRjke&X3ZD
z*#}L;pK5`Vj;c@{qy(*je@U_14;q256u#pV1($xHA8X!GOXFhl|5-pejn8RUyKJhM
zo<;SS(}|4kka<`#nIeHQOJiub&sDlP{t^vu5l%Y!p|mtMgfb#d(8b$<RNj1)Ml3!;
zb@iT<AbSy-9bGBIWFH+$lzw8x%~afTEiIh7j0_VUsqWlt8a=_DG#}WJzMNsQwoarJ
z|MB!|ofWOEv7}CAmeR9pMXh&QQM*f4q<qMV44N#-LD`a8KNw5xH_4fVN7FcmQKVZo
zlBRtbN!|92qGj7h)03nzbf)t-y1UDgH00O5*gu|>-;Jk*C&yFPW-FREc^v)cGMW|-
zA4#!ZBj};dD7w}>mYj^Nsl0qD=~&L8BcB$M>-iPbDPbM8blFVXhy6$6H1<-~p#$W7
z@euX;D0{ISylLGBU)mHBK(Sko(`4shdM5oiF7aV<K8&D_)2~oc(sk-FAfCLFlgM&j
zDk&PKlWvPFT4Rw-#lE>T1}~{m&N%urDro(+8tUEW3+=kwL_?hZlI4CSIOeHBm})?I
zdn?R%sEsa0dhp!b9$&LNK)J6GUIZCKuSIwC@-V}SjuvQ{(HA2R55&i9L*e8jJNRCX
zM5_hku+exTYI{t^x=qt@Q%QP`^B3UWhb5@)zZ!v8H=xO3I~I0x!E^TmNN(wdr6)Y$
zDfxf5B+p>me}T~NdjbV@!8n_C7UdViFf%Ly=aVm^SN$~<jF&8^w42x~`4>q&?%{6R
z2UuqD1Y_1bN6w1^99k*++|0_bLAMHdJ!@gOz5&1AHpB0Z0_&Nm$bMjTE^4RA&(CSG
z<FGdDli!XNc6Z>!UPfI1&X`-C@6Man^yCvGdvjH%{+y{im^~YZ@xBJ;nm?nsxs4^~
z51Pow7uxcqlT*2&WCnZoox_b@3)tw}BF<aBoZr7+&FwZgvw`XsK7L^b_qW@_Hx>5t
zM@Ls)nRbXv%{=&vpBIaN-aN(Gm!seLalS(!4|sfx&BmN$?ZjXn-unz!o(<)>x?!BT
zH=O6zMDW<DQ5<>w3dgp*#-0mecy44I&!~y#Xp@^9IW?KRH{arGUa7n_=q}p@+~ZW&
z44%C_lhcsJF$#~kA@Q-yU1YO*a}MVoe8wLXa(VOmJWk1e!Ma@v`1RsK_B;8KZ>GQE
zig!hPv*9(r{awuMewA>C`cl48{Du?Lq?hOHJD#wkO!8gJ*}r22kN!}>Yp%WLnr$E0
zdqgE0t5oso$5m_|Sj~D)HJso3Be(hckr(BC;we`@^8)W$R@hj_q4Vnbr0o}m<yX!a
z`;FI4Y+xsc@0_rzkzEfpadzYn_R0IniK@T2->Bc5yY~-ofAE(ZS}KTb@d~m#K~X%>
zQxf(Ml!VtFWif83im>~lBDO`V3XgSa!ltjf?3a*z>TxpDwL?R=%N$pC*-O!or75oa
zv=Z7gT8m5rEpfP5OX!}}7A+R)h^}39MeAZ+vGZ6P(cVr^_)4uOFs`l0UD!@EOEzIr
zbbFCzr!Ug0^~I*W2EtWa_Hdo-ATCNjvqf-6(Wu)=WXgTl?3zwu-6TVialue@X*3ig
zM;VECdyT}SL?e;>LGHn1zALY1XK{u*i(8hRMcJp-w90S|b>AmhTpicaullvrL47?X
zE_SBL8XHMZVKd#Dw3Y1Yw^REsJL$qy7h2b1KW*vYN|EypQCW^Vd2I5cT~<fPVEj?C
z*%Uwp4~|iu<tcKmI!(VNE40;>aN6<e5<RevrYNmANh7>T&L3{m3Zo2KUGs=aZF9+U
zcoDe<m66-nkMy`h18r;iO=o53E3Th9tXFE{o2ND=p3sv_8-274kZe$&E+`7<CfT26
z&@Hw=ghhWeh787F*<)u=J_6ff#zO1Z1k5>Xi_zZG;PXF*&O5BfJ__R{QV69aBqD^6
z3g<qLWQ43YWR_iK_TJi5Bu&xMq@_Ky)1)Yc5;8L)WM;kJKm6gkx^RW(cb;?Z`}6T$
z0<};}1SG7%<y;#WRc^uM264ZBwHGg158?8+<M{R4PH@#6k^IvcZ}*%=;afN4tUQZV
zw*&)yycc`}d@%8+AN&>sz%=X<8s7)u?^uyf6^z^S!4cS55rxE-E6|jWg+uRnq-kD*
zN53So+ekrpdn$q-rXwp!_+U<4$AhUk2oT+?_AB|Qn^%Yq%|%!y@+HwfN^oPR$of4h
zhr97jlzWNpW@9C8X^8n?w(vu4tHw`<8uU1G7w1mZLfT&^I#u_eG3-8S-rPq?-~$v+
zt%vV}dZ?{#K;L@}aG3NE6MP<G;G>7a&De;wlNyn@xDmRm8gYL~BkoLY#8{0+Xg+_4
zg20EEZ}t#>UN)fH-Ucx{)nkKAJ?4}@K(W>XjNE)54<qiu=w2Nzw%1}$uUcpix{IDe
zY7o`G8uNt1()ME&w%n>joM6%fY!SX2y$bwzT?U=NQtX{w0^?7@%`R?EE~5$&R*;Wf
z3-U0&F$d$;Wh4A~7K+!1K6iaOn&*q2T7C*NM<?Tye<H5^OF-)ScqHA3MZUIR)*KUT
zk9$#29~FsyC&F>y_GRd6hGOofVBESEh@@{9A%_5zANEI&OyN@a;*Gtdy-;W4iAfhd
z@Ve>@v}E03G|m;y>rP?sd1tIIbVTZBdz1{eLq*JSgsU9Eo}&k_;msa6F5ijhTedhm
zbu((SY|v43c0;eN!YDmUjJUcK`ok>nI?NnO1Lp}=jTw3hpU4Z3sZds&j9Ye$we1ry
zdXFKZTE=4GMm>}X?m_0dp?Lj58~R(d1WT<SKD+dR_dqoS7AWIC+wS;0s4HrocZPm)
zN0@rbV!_|v6#wfRP3ixM%!H@b_vTBovuvX0Js*<lyE^*t`VLho+@io)rSv$WkgjXy
zl2%+MRhy?$NT)=)SQbm0{Gus-XE+5|gpif#C9<69PbF)-=|BH-B(=Gd-Ja7F*4vr#
z>+Gr9yc5);{UANKw43a8ZRu*H4SgIZI^Z(PsaDK3uZ-qXX|ov}-#mksW=^AJH>Xgh
zn+ctMGMN;@CzEabWSU<!nWoiFrk)oki=M$`Dl;^uF4~i5z+a}Oub}ngK<#&kMn551
zvjX&7pK0N2iO$+hq7O;Nl$te}HjA03veblXT}<e)_hd4AJBe;QV`>v$|8pov#3X3U
zp2-yYV>+ca&Y|Rai%56<3gL-aLmetN&=-4KdaS!!u#gYX(DGxnuFZ}<XgksFeW%Fn
zof~;NdQgM07yVNABUz=3RM0n=bd19&Xm1o<%8nu3-q*-pxP(tDW>RrN4qZJ_NI&gM
z=~Tfj(wSUC%iHf$c*`StBXZs6=L+Uf*hf-T_)gJ#{s^bJtmuJvg4}uq!~}MQv$(s*
z^;N-<)oKX3+6QC*_Cvju7Fx@-amH*Y!kcu_?5&3gTYVhdZiE*hga$=pOo^Y0bDlGC
zGIcImG!}{*=28>{h&)dGdQ58DjN$8c;_TP`_z-mrb6}5eF;1u-b{hV1?ueRn4m0Y#
zu;1Pf5BMVN6oPT%-DP}w9SKdj7_>_XDDh8*S)UBN$<9WWO98wN3GRj0O_BGygRztE
z;bqN3q?~(-U5A=+_RM?iy48kOgLd>^Bg04&JWq|z{P<KC&XeuVZVt*^t)#{WL;7$e
z_vf%DT3q5jn0HJY#((8Taz)i>)(IcS7LMcj{0?yU4aTBpK2`WOXRy}U*=(OQpVvKG
z$X7L&@+50ZZcAOwzAD!I?$9QVd}qshHtyn&E&Dk3$YIv)b%M9n+i_PtNABzB%;O)6
zIR|cBci{}LZuek0tMfb}*P9In`te>n!9IR^kxPw(*y=(E>wmn=ho(mGCXXl{bMFfO
zQ;X#u^W)j?*fo9-mc)JYQ#heIjlbT{;KemrJS#t&mxShuy-GeCav^VND`Ka>V%|ul
z+^MvTt!Gqlh?sX?_O0aIJFD0~<2Gls-QlU)HJm>EE=O3_a@58;Ubp!kyR5y>x6L2$
zuL<>Bqu#(;pBgyv#zS7=)hOKOk64KAxR>x_O^bWNzB`(@O8+Te{qmHbW;|o<6VF*i
z+`Y_IU$X6sm;5o~6)!p8%sqF#X8Xl&Sl;9<H;;YCHp2y%aOiuM9rc0NPi*D)b3Ss!
zhEII@)Mq}G*v9Tpzwo|3-?-0`?}E?zgNt7L<b_7RIM(eqJ2d}cr&<44Z>fy5pqs38
zLb%lJk9LrTOpudCwaH0qu62|gws(@2>d8wNTjZs#v7IHGtqRh*5sFggYei{rWEaU|
zoyai@X5o{%Zc?DTlGJL_U2^{1UFsFuLz+3ar&QY}JfwcgQo(o?$+=oZ+AX~6^rM&L
z<De={R#cNdpHh=-JM@-p_w|;Np7xexW~xgX5$aOzFLfz#b{{F#vyXK8P9N#LyoPky
zP(#|fTtiY3%;wE58q$B~G^9R@ti=w-n)sD9Ii0kjQHwWF_@<3iledYIR&No{H(Mh2
z?euTpE;`(OFAZ9HfC_#eCKs6#bU4wD-Y7egQm8Y1UwN9$XS&ltOAo5>KTpbkeQ3Pf
z1yY+DNLz=6lIyex^1FP6zAuWWTbq(8`cnpF|H>uRv&A$e>K1jBYH9a`M(T9z1qHOU
zkp0H5;>Q1ncB;ssY*=R;8_^XDHG4p@Q!m_T?hU6~eUX$k0LL=5LAQq@P-Y}%nva1~
zu5k63O~l>z5`IQbfx*rh=wvzvfrHHPx5Hw5eYG6lZ?3}UYwOY1XA^E6*oIxpc4Pba
z1L&`O6i@e@#LY%~G);9vT!f3r#kj&(bZ7EY&*Fx>C-fG2;f(Nf9=YcSpY8!TIa%y;
zHU+`KH3Z$lFJpRYIKJmZ;&V<k6jQIFcUY|W%!<dc_1EBQl!O7IU;ZsS6@PZ6Lsc^q
z>bY5XZjp@zPjfJFeIB+p<m38`Li~s-!uvlrux?ff2D+5uZDtvqn#z&XUI7Ea*{M*d
z#0241GJaNt(5%}S=5z<XQ>t;jwHoU^YOrSDU7Wrw^15oZs6AAR+^SlvR~7u2*>y<X
zRfnag>k#8vhiun61n(8Dz5nX4uwNZ^JgCK2=UQkEtHq$qyD*(}7q^pZprl!Y{|;3P
zR`4C5+a35#yNz`_s_@IB5;r1lK|Ap#4hhftsH^1&^)CaCmtwhP3D%4)M(m#=TqqLU
zXZr%=A|D|wxj5*b10UmTOn#Du+{2l$P|m>Qury?kO+|;SWXzbHgv|VFa4}90S?V}6
zj)_G~*i~#6IpMMPQRw?V0=zgJ5m}dEBW9XY$AY2yFc2!^g;(rs0QS7_N7Dp9)SU8x
z#eFaA8GIh5+s@%t%2^y~7dJ>_H_SPC8XXH<5GNyY=Au(K{J1@mvQJ|0_hT^Je;B&o
z_v7H<J=iU?6Q-wa(XP4~yTfd-$8a5@N>?G;$`bNzOEKTo0;<{z@H2BB;ug%t=hrh3
zeqtI@)J!luM&u|=2sO1vu-_~)@qb5SqxVR-jU0}1H-=!)nnC#aR}(Fveev8(9S!0Z
z`y#d{25wOj--|BzAln)59?PMrOa>3_ev!esuXG`|l{~t?CAND(R_ae^T6H~D#MhEV
z!fm=;T|t?uB{XDz0hP36Q#ZE^8cr!R@mB(Esfr=PYf%&&6h`6x!NSdOkv1mz(c}AG
z!iON3p0@6^^~EVtJ?=#9<L&9T%5nN#e}JxB+eJanTWP0-4b=}?O{$8^2?mQOxpE$L
z`8SJ-nx@mq<5Q^mzA;5VlgR80lUg3gYZ<7`02De5l++c}y@2Ta1R~{K6KVa)2~_7e
zo_<_5qRYPxY5P?}(#$ZV=D|kP`oxGn*N&%<mlH_k^F&H&Ai7{D;!G8wbbnAk@$;yF
ziR3wY0u}OjGTAhq6tpK&ya!YFb0#!9YX*JUGLQcK6kROs71ZnFYI1kAq0N1^()9G5
z^x(*Tdb{T+b+oXf?PncH=B*1Y+u%mp9Xx2+_49Px!<U+m1d#mkAgcDdOrx_SX~*}g
zR5de!k`j~YB4^O<_t~U>yMP8amyq$qTVzyFO+iQRQ@3@GD0|~`k{sXA<%Cum@!~7-
zfZx=8v@8necZC10&Y17l1+Do?sBZ0v5FJ(QwpPcB_`VpUFaXAT2O;tM5Zv(_4$Bpz
zU~W1V&Z`WuAZ8+Zj-7-@ty5syG6M!WbMei80d`DZigklmqI&FlY}&XP4+?f*GVh1&
z?W4GmV~05doM3nQ6r%sR;klg$<W#({A;}jtHUXG6M(j3~FQZe>NJNjiirmfd=un=7
z8gc(MZMzPG{Ct=u7ek@40)Lh6AVs`ORdjxc3X!e57cLxo*IJPN^fShf|0$R}|B%>E
zj+KNvZ%2h9zaOK-U4nY@MO9T!^it>J{ra(4>;UokugwGB>Tso>E+_t{$4V;ttp3E1
zf2Ije&IO6{Tuiv=<aBOwoW(u7=Q72c^OU<5{IA1uHkq`NOYGM2oI5sraQGJfaC$pW
z{<53jZ#cl)o*rRO{ga%q&z=*r9eJd@3&$)t%_%`{oZWIp%p&Lb#Od?=^PV^V(DvgJ
z+Y5X(`69pi7RYHvA-rVAWtIsG=a07|+2(6Bhp5Ex+o5sntDnHX^b`5yuw?d8PUVc&
zG(K36!MmKZ*xWdqYhUMZrBfbjsuggSZy{gpa)UeVD&}wHC7h*JCbHM%{L8t5r^Mal
zx23{;b-$9&G*)ru{o5Q<eusM{SMycx8rIu!m)B0K<!&l<{Ow^KzX`p^@z(dbo6Z9+
zfBt|YLhE_e>IS}|`H;syddTrXjeKq0BfhKqn8Ut2=J28?JlC&@yX<_*#`B-?q;b!A
zVgDDL-R&jE%Dv(_vdx^*={2|ae8bv<-|`4abiAxu_@e!LzLW5Qk3MhZ9h#qb&&tm{
zH%#2SK73)D$=|q}&v#zj`h#E1Yv<M1ez9fmKfL$EUlxF94vv+PoUCP~UYZ>w#d{s3
z;U035p74&=$#fF^txnSS9rDt1ozBwYhR#yQGYV4pBt_}nJ4MOePYB9HmO1`qSIOU{
zo0OoVBt_*ZNwm1Tr1`45)K&OM^M3S@z8&f*?fBYLdbdSc8rh&MeVM8vb%_vnFPUD_
z7>iy~Sx7G_`(-bwQd?EZT&gM=JE}_kh3KkJk*XvsazM$CRHdkdmGt@XN_szK6@715
zMVsPQ)0~_&q}F2{SwyX;?c%wxDSIQ8n`{wToNd%RXD4Oe+d~7g575QmM@Vt*Nvbcg
zr{CM0XsyX9x+A%f&bqUdmms+4Lw#sXsXuM-zeLR;A(ZznoXnlCkp2nLM}3h@viX_x
zQ9Yl2zbO&i&?=gEvyNO79?{~Qmo%~AJ-L7UN~b^lp~r7KK&?t1C&Ifxb$54YN-D@u
z=#7<+HL$Qq6KXkw1jAehxqo!AbmnLbj2VZCLq)zJh4FBq2^RgHj?y?Y*jmrWzyXV3
zUcVIe-d1?{-&zDKZh&3E7ChR&0}8r(G3)+8G@LvpcwBZ^u+{<Yu}(-8PN8MvU4_@$
z9cHm-k^A@@WV(7Geyk66|L2Ea8!kZRz(p9J2*jZi!8mdt6umZvL4AG%R*s57q3jh5
zsk(}_p0Ti+6OXoE2~Y@2#1WHZ46jYW`;}=3f0T~Fi!w1q%sbTsvJtvB2PgA#LH`76
z)Sv*CR)tt7n4^<jZ{X$GVko+lVEgV;JTxzZnszx}zAQ)YOBL8N{U+`_zKOr0uT}s0
z7A7qdeo1i`8ZELe=jT^pq(c?jqO0&GuL`}(s^B7W&i}<$iH>;{RxGK)E43<EmR929
zmP$0rRAQ~j#}sxHdBZ(7F}<z=okmumNn~ZlB$r{%t5SG%FNO7}5@?weLwoiO+!218
zQ&S6ZMPD$1`iNQPdoJ{D=fEpC8^bnV$K+9281W_p<3j`wdO;fMg|GW}c(QPpCPC-r
zHG~~cKqu9B9FK^Fu|W)W=3IgC^l02Fjl`r`5r{7i1DjmN@AMGJ>jh(Ta3BnOT*BU?
z0T|otFXl8qaH<b{`*`Ejq4Usqa1IuR9+-7j<Sk#hA)8%s%;OY}K6S=teJ2dCbAZ`x
zI~4XmiJZGf@qF7MTvFH%yYSsmpSlATPq(7kVG}M4x4|26hbla@3LkVVp;NmQT^%hj
z*=T|2YR$v13$sx<cP3o^O~cnj6SS=tJ47|Wxnw-1954{?(PPEEbrf!%*2QZcip{Os
zcpR>Uw^sczt&av$?~8uTCE+gG&;t<@x}l+`B8IkgLaf+fY9#!l!t|fC<zXAO_xnIv
zM_!Xj`!li)cti`W9*|^mmpaX;qW;^;$sy$i)%MM&*+JK-Qc9<eUy^7;K|HOxe3f>3
zN781W%M=+FL{_;0^zNlE_3iIP%GMs#THr<>=#==(cBCuzcJv_pD0%GJPjc2f#k<B9
z(%x-N@h4W%JpX0nQnQf8na`ux@v~@i<TNTvolIAENMv10wCCXj>bP$_Rk|9Hvy2fn
zr5I8dM?;!8*N`T)8PKw=2IQVPj-E8>Q;57i1x_DJ$!TL~->xy#etZlmJ|08e9mdkk
z75Zd9dmQzcXh0J?8&Y|oAtgUIq-Qq_sbZob`CAy!r1j%S`=vhpuo_4GbqvXS)dc$c
zo@s=bwOTx8()x&b^ykDPdeVP6?eJbjm*lMJz5Qm28?u9ZzV4-lKZi*z^duQRb0Dp$
zE>xB0O5V%Ql9keVY7s8JQCSzrE+LTgGD1aHB7#giU!_v<e)b?QiPRRRlag{aRmm1m
z)X);LJ${qUcdVwaN%zS6cq4UP_l%A#eoYaJK9J_}FQjGNPW>T+U(@BVTre;UHYno3
z@os1d>VcDGD$x9^h9^@sa4eud<|+=vNbkW|V=@d&dX0pI${6&YYJf|T6A&;}!U^>$
zn5r`aJGRWh{1@gJ>ahf#8?128W*w%S+JwX9+flBw532)?AYVMkmaTWhsyr7w8s>(>
zL1(c<=R7{;`rz-j3otMWgpPVBhW3ttf#DT=-5rO*yNQTfnuc@#vY>t|4|j5Ifa}Up
zt8yD^2WqjPLj!iCJi&SQm$-N89hO9Xg45IQ$Q=I{ix+fY8_!N0*Py`Bqq^}x;lJzk
zR)xbB_hy9(4R)HT$!n?x^5a!Q*yQIh&I=sL?sLa*RVM=uD<97T{K2z#8MDsfsT?tF
z2B%B2nN8+%xA_bCtIZP5c3Ht1$*VZ4c^w<;Y-HZLmB(l9;K=TK`SX^8ysPpkCkuy_
zT(vz<RCMC{sV=<s*lC`T;l`gopW(7m=UCh3JX>7!=C~?f-uu^|Kj~iNP3D2zZ&xsz
zyM^++pfFx_C4y^XqF5o~3R`-|Fb>6W|3wMxJSdU7zDVLZ0V(`rS{m0jrSr9&nY_C#
zi@R;kW|!(5{y8d--Hzn*ft&*F)K<u6``_S0C}y5l!b=vH@{&bmTsyN|<c2D^uJ=t&
z{(O^j#q5*fTgm2Ys(7)^ZMJ`PoBhM@@X|Hae57v;=iaU1TKBtbJENAz%GPmyP93Km
zy~p9k_qj*M2fV-X0nfWw&%d@caFgjno;SFWUF9EfUF#!W`Q$Orse8g1)lHmH`;^Te
zKjV`ho-=iN$(yxb@n_R!P7oZQ-NAy-T=$l#XA9?>zvpGXA2_9{mF>rTVq5#qtXAK~
zng(Ba;JI&{)cTz_FZjvTg4ujP<2OG&{)fF@{N>I%qR(|-M$$bkE6JmSbX{<KWc=kM
z57Fb&Y85>$zfO|&BzZ}$NnR>E(OK%#UqQN_q9AQBQ<Q4z6{RejE>dGl7wPNvu2RXn
zu9EWlZc;#PH))uOl5{LWN!rk{yA*9DW}x`)lEshik}3C)LiYEN243wU+12%sPKddv
zL8Yf;HKeEXXjD%rsKZj)p}&-tdM%}6BbL#ppUY@Nrxo<x+>%n7tf(ww6}@O&L+0k|
zsYY!Bg^k%vJA-X$`G%eJ-f=Gt{dJHwl^vs^eRi~4-;vgcJYxM|S4v!dhRQR}QSdZx
zYWV9%S3X^&>pCHn7#U7q1Z$&HemuF`rqE8;ELxymKvlwFe0RZZ(rdU!Q`{cYjQOuf
z>iI!%NWW6V+CMa8ZwIW^lgFJWiehJ=1cPbH==V<*qsseWbYy?zTo{NC5ksJRb2!Wt
z^zd?pK8o{<u*?7$kYtSS6Q^NT@l14DJQo*VnWORGVl;GFfjbvhq5DXY(@WndSf92?
zE!_z{vwbixK7=`Z9D_yvBJq?1o~1is&NCO-b#}$TVeTj~ISUPob0}GJ9>+I$<EOPR
zepvcr@|*w!jJt$-<sjU79gL;vp_p?l46ca4=MRyv4v0pfbQSOFVvuJedYUcqc)sZx
z-aJf17t>_)4^Khck5sIgmJUOQ3?L&DCeK8!RyG^9syP_aU);3R^N`j#A6oD7@u#2w
z8t#QKHY*ahvm(@Ayn#<*XR;!r7%loGI2Bj|<zFS(IlB}-ZsI<aTZ+!lOELdvDHh6>
zVZg6al!`fJ!;Mn(^Dc$bl2UByQi@fH!bvHzzW?qQLw8v*-rc+b4}%-9_9(*jCxv)B
zq!2S#7ht4EK0c@9;oF^D_&m?Si1*p(_2D|MzRJR!`<eJJKLgz^r=#I$8oDe<#fJeY
zxcf@5_O2x2rp+~M(M-UQJ8_tJJQh!e#b8n86_jp|hHbYf?23$l%%pIb+_{V;YeKQ^
zeK70~i0|v)OR#ach((=5hu_g3cCvnWc*F<AZC;qU`8+l?p2OIA9*DSp26x5Z)eTQq
zxc)o^taicYLMPM-|FW`^Jsvzii5mOk5N!aow;jOfn7x>}ZWkVDY{#*hEeLksi0E0?
zcp|?R+e=qswd)Fm{<jpJ)h)1~$sGDIBHwjjHs(#Ai3a8Ah<sv#-Eou9{Rm(=YXZ#G
z4Ds#hSZt2hL*rr51D-hyjjBU1^!Y%%Nzz2TV_)=Iu8z{Ns&Mb7j0x|R5PP=^p5=E&
zLuy9^C(FWN{~rpz_?<!@e4>M6TIfUQEAr5PN>NW9l75`X+5}WnLG&$peXo?VhZNC5
z&s<vDFN@3yQfZJ~A{8x*qc^5kX!?u@y1X=$`X3AwJ;@6c`P7H(Mx3V|_GhV2iz_v3
zaUqosjx<JiEW_oFP;$*)YO35$d){s$4bAmrvT`N;i(X166&8}7$6V4`K9g2EO{Ga=
zjA`5!&}fqh^d-@dre=(zj;8uFR?IDqm&Q=~*fI3<>1dkhHJZ+h9!-kTdel`$j}A>8
zMWqfS$+&R@-C8$-e1?ypvExTjj_U|I(|;sYH;kk=38UzdgC2PpkEZqIqba247=rv5
z(L*0i*~+7--A<2mR_oE*o})?Q=@<%qH;zv9oj@yeCF+<ug>tQCQrMfhRBOJF9tSU_
zn0Hq6ZRR>Ua$}?Dqi-XlS$o8d{Sc{KIU$bP4m9wfGc|rbO{vajs7BwD>_2;xvv@z-
za_th`iwmLZf^d4=8cnyQIMR<#Bpb6dvRAoIi@N4h>9}I*>{LP8zi*R@XC2*_9@2uJ
zO;lL$ioTz3q0w7EQ-j$L+B)VhxpnV=)m=Jae>Vke>)jQmBf8_jY-O|`QiVmfI?k!}
z!x+~Akkit}&j&iVo2`q6Dm{3qjKfja@nSE@$nR@{NBX9CaCkOsKmUg;@5S(5X(_UN
zYjJMjM!1Q+v}x)dNL>!2BcDXtQ3s5;>5PV9u1NJhgWtV81!vnEzKi^k)8`Vheg<Rv
zk1*Kwj)ttr9ZRX#&^93zcOPY9Yg8@_e2U;7B{+bOE3sX4Aj+~IKx5rwgy;$uyWo=g
zs(wVC{x@v2`HeuqRWEVq$X7Evv*M2~Tw~mw`<+#0rB|x_ct#(txYm!m5782gUu`}z
zY$$ik(B+VYdc5`fSl%CL$mOdi^7_F7{PAfrPb{CtQ(|Vap5Gk4>HZ%#xh>+M-b;B-
zgt%GduIA*X^{m=u6MMpzeRl2S!|{9B?c+g~nQ)Bl@=tQjD|`N?<is{cE_}}FG&kA1
zaeCAl-d-U*4{t?2N8XzUYx%PGIDg)262Qu{E^+3pATF2^!uI1X^UA*Aob@Aul`5mS
z&i4ujERNx?3UTayEuMSLy(V~oiLACNnayfbxM@-vKMhFd_3tv+ZA=!Q+jyNby|Q^(
zN)FeQ=dw?&*z?ro^QW5yd^xj_%Y%wo;ph!kTv*J92bFNQcO|?(p_F&-DC4A&<^1(U
zIe)rX!OIui<fcxycwE*k{;|E1?+&Tr)nW%U+V?i!TzH4GRH}LRy=q<;RKp*)+~s{z
zEl=xJ$APVNoL7C1A7qLdDEa{(zf{lT{Tg_(|3mf<Ze*jlM=V$Hn41N&FsHqVEeAeh
z50U-RKKg=d6JPTBmRIaJ@-@HN{)Q)|y=7h57B>IyJ)4Jo;G(uxo@(}qSB8A%@E>g)
zzThiAzxs_|%l%+OtDihGxt)U)esi+rAAyPg%in+e<1V2xlC|h^O@A&cS)c469q%hA
z%}o>xLoxe&z1vZmwOVw!UUibfHpxqRujHlhRh^|u;bY%ESwUJDr65gFRFo8K6r~+m
ziqcfYF4D{eU8E`JyGWmJc9B&7c9B*M7c9fsU8T9!U8NoSx=KaIx=Ov~n$!5h<`mpu
zF0yqCs0S~kc~cfqv6J9#bz4GF50=vO56j8W+=_b1uBN{o*3sFeHuUu0Cc(+DrHFaE
zsPOYX>iq67MT@Q#Ufa`?M^04vOS}tCbEmj84=PyUMIFcck=lZb)IBYjj_e4h&ti{b
zE0;iN%_%fN|2l2DUqC}|l~J!hx2bCJeF`XiOoiiKkzv4l>iYQ$&7Jt0&Zo=b@4QZ!
z`CI{=k9NaQ&7RnFvlqU*s^ib9ey}jp5{{|CShjW;)ZItoZRr@KDjQ<`hKbNBl3=GZ
z1?SyNQPyUL4_5P`RI(6DMlXf3k0s9hT#de#){xEI1ZS0Pc)np5=B4h#m*0obY0_~l
zgB?5<IN;WLCq(RYLBWC3s5t0`?R(B(-9`_rweZ9!c;R#(AIxa=g+Z=Abesb4b=oBy
zZx2LBbTB5)4~4F{c`p=M5uYxRsBw?N>rPjYy6-A_H43-m#5lCL#iN(#I1g1##E)4?
zSh6P>J-t%U8kLIt#5ByimX1|n8OU|bgwcj9co<!W-;e9)n~;sy%W`n{YYq~fa<N=w
zbgv5To&C!^R7}ptX7_w7xS5XwzeV0hs{q@K3NU?A0X9u2K+E6)yy;MYqqX_C<)05Z
zbMdw9d3fuWhc3E#XpPIoBf*;K=beMEU$c=hHydgG*D<vr3rWgZSUEWp7uIK>;CMRp
zJkwxwF%|lODLCYpjCHO_Fxj678%x2XK>~_Y;^8fv7mt!+@ZzwT^Ck&y;_oPYPL0H-
ztq~~gACBo2m!Y&b6!ZIrU|dcR9$N)s@6U_q?G=C~-3xe^<0m+azR-H?jcr@JFzbsa
zt{*vvY8elBiOz^(M|Vi}t}ys}3he^c9o*`KaW;-9e_#*$*>;fGcLMbbj^f+MLnxKo
zkE;6J@QUAoNPFS;S+W@)Mr;tA$o05&Z#6U%tg!L)a@?_50vo|u)6g+Tap!p$*J6f#
zH8U_SbDH3wnIQi3B-rf+Hd#)<#Oa2(H%cF@HX1wsjDXk6VYpj41Uu6PAvbscv|Rf^
zaep7kSc{yEg$kT!^}qp>ZkRYp5fPK+5jjf^ddp?7<3&3;>56-lLo0cIc|)ehUyz0N
z6ViWHPs(|<RG)d92Hvfp!CgzJaBTsdyq8T*n=&Z0cM2_kl0Yr_F|;)^ib_htsP$Da
zeO13ihRgivWTZE(>mc}5wr6PL%hS|<r!)2I;XpdBC#Y)JAyWFWo3wt|(%nHDDZ_Rx
z8QifXvl&Y$smh#I?3zQ}CeNTH|4pH;*%Fywnn>Tpo~GrqJ`LG3hSnd^Bi(-^$-QL+
z_0}6fDp9(0W3n!}J|0f{cMhj1?}kyq{9&{#d??xf)}hW@btu1s4!x}&LZwfJP|`>p
za*5NS-%EzlX|-W=vUV7CKQWwIWp!!VJY7mys!M@Gbm>FoaQf#qoXoBbr<^mobYaa%
z+GaAEKJ^<%71`qjYlo>wV+xh5m_dh6&!OfRb2@!zG2NE8q!vrzj;ObxXD(YweakM2
zvp*ny|6{axq&-==I8lgrb_F}T)9<0@sPv&16@~kWuenHD4g{0Ba~R!AilPM_V(G`h
zYl8EYN@~rS^!0Hr?fY9qZYJf#!BuovwU%B5)DuLg$nNF~lHc`~#tryL%kO?AKhIxe
z6(<8fe>vfql^5M4MZEUv27A$?sIFH*Q2*XgKCA)nZ~YPKJ5b!GhCplKaQN;Wg;8Z=
z(S5!V%;bTJR%4u0n2vMv%y9Ple3Z|(fT!FF92Wm>FWs=gjhkB${&p9d2Oh-C9mi3)
z)gHs+o!}*N8a-^>ar==6RJRFsyP6*&?*||`EeP?+m+|0sB&b^q!ZwTludkD#E!a}y
z<_p(}Q6cgsm%?$|EzHQRMqmB=SbL)p4f{klVd`s~6uD#bAzz?l*pBfdWw@uQ9534_
z&tE+i`Gateua)V^N2d4U?Vi1PTeAjVouSDV2?P23pdsuKIE+mOkL3EK(QGky9A9fS
z;-*W))r%(a2GuEi;IYUbC(dHivvb*Dk2z0XYrzASF5?UfD-Kw`hD~g&x$l8ZeCn(%
zH(cGx^Kb5D)z*V-rgn^v1)XHiBzs<7;>c6#ojLjKDOPTG<qY{VEZftAr>S{zwVD^p
zDEshGc|Q*L;?J261NeE?CARkr5`C)>eyM$#qZ-4w(Jg|t#zt}9@@RHha+OCth~cJr
zaeOL0p8Km^<KPX6JUlju8(${#hhC{1IzEj}=cn^T%M5<8GL!R{WbvFC*I8LFo5Q>2
zaPYGnzMPcHi;w2<I+J|f|1Y1XW)<*(U4?u@r-<wC7qRuJ8>}?3n2$CWv$<ah>&z<U
zbM2)(Afb#GZ7AosffXG3xPtvdZgSg(TikVAC6DYNdhGYB_*Bwu9_e+5BMw!w%7z-Y
zTXvU!o7eKm|LVA-#Xa7?>ON<0f50h@^=uT<z$yg~xmR-||5bj>o>QK%^S&lN7x$Fc
zzIn!>!mIvs%S-NZ?G^9&-prLIZ#dEIEhpZ8$5un$^ORj5xH+el?<;;{FN@C{9N5N(
zUVPz@;osO}^LH+c{J}|2esV*vUz{}cH>Yj=!)wm|<xyAv@!Zvd<NQ)MyEe*7%U;S#
zt5<ap+#bO(oFXS##mGsaJvvGT+dE26OFBviH9AQpHl3vNF`cB#g5{j1Eibt&l9$Sk
z%1fhy<RzPQc}c%aUb4C)FO76JqwSey^j&>6UA{7#x{2qP)vmdu5I>KCM*K(T1-nAy
z(?aSoWwF@fETs!2D=6dNN?LMyEvZ@C(BN~M$y8}OwKVLe<v$M4YwKe)>6{%Y_&O4b
zUC4f*o9JPlC29M4TD9Gm7NiD{^4wrSWeBHzV#c|wpFlR0N`HN>(;pPl#y;ibBRKLa
z#@{D7!6=K{^OEYFT4=Ug8y#|Lrw!RMcxTj6_`y5l+C<^&t>}*XTUGE=t2frY)_{Gn
zCW6K1h+>uwHdN}O!v{TRYmCE$x#OYW1Y{N%qg-Yh<fhET7P~nZQScue{w+cwFN1+_
zIuD6igNKdQxS+Th0b{qJ_X6Q}*s>2Rj~v1UhhxwbT(@c0?a`dzh?F2_taLeri<@1c
zYwC`h8fRhF?15cz!rf`>1%<&r7*^wpd58S5v`+wrCS1hDNrCV!3c^p55JaSg;?STl
zSU845`LXcCjE+L^zG!GAU%|6iS8=^(ERKwg!)McYd@xS{FSv&28HxBeHVNH&Cgalc
zWE_b}!NHBGSg)Q2rR!;!yD%MJ8`H)9Dg!@nW<YJM=zY3mLbf^+@k&|fKPd~QtFth2
zZx(8gWMSU^EJUr(!mnvr*r=X`ZjUq3+dmW6Of%7=IRjgcX5eq940Lf!N8LAZM_ZMK
z8`o1|*F6=ZmZm`6HyJ)RlkoF<A};hxM8)`Pcxsk_af{>8b9o%9#Jgbj9MRF>tFRt&
z1t~%bV*N4_895PXI~R_{tHVTp>M~Az2th_dFn-wv;lR*9SPHLYwaC$QFus6ZZ~dU-
z<txZ=Vn@>`X2@gbk<rf+8`C`yX?_-t&F<)a)D8c2cZGS-DfAuXf>g0@)0^T5_cD>C
znrDXtmrr1G#8KP{KZMqh{m=;5gSc}$(c5VoDi3VIg3TK-bcHo)X0JspuR@r}3`Y%I
z21B*QIMr3K`Q-kC{NFhk*FFnlewgCc_o=A-F&Q7)B{csb?30}U%T9){R@BEfrO{Za
zJW^!2hoiNR4l?^|BWs`*#_IIPwh<br8rvJi6MLa%YEL--rv&@eU2tGeXAE`kh*ei*
z@h$KV6@UFs9aeoN*9R>$VPiAh?fQ(A${J~V;C-5ZM)Z>XD=9g@j4pJ#K}j3)sOeP}
zCEBOaDkM>Lw|I*CafK{@M^NA1m#N3}AhK}`5Z!8D3N`W~&2SGoD4cWMV@^>QQzx4C
z(T+Z^JVp`n2k88xon-!e3;C(p&|cwCSpIf7-8*4HA4dw7O{*F8shLjGpG~Ic227!i
z<3+Y*9AzCHLx){Q(XzfHsAcGI(vBD^b{RU9oHT?ksSly^mj;vVsKN9)O`8-Av}t_E
zAS&)Oh}w4!q_>~6sJp)w8ST)beaE!uzd|i~J$@h!xIK{6jtn9tU2Pg(s!c;p2h&#H
z!89m!Fdc{*OvjxE(>eRW^!U+WvU)g#EMtdKjkT_DmXD&nq)-0uji~;<sMy_}Oy_P)
zr_<?X)G28`733`vdE#Z%!*~^W=d7op+c(q01v^M@^FEp<I621Zb~Mq}k*w~!kk3*#
zGH5wVLwwHD*hRk7dw2l(^$w!R+Lx()zR2W+T&4LM2^5i)OkNHdG-p>1If&eG|0|{R
z==Ck?f*QK(f1i$bctkt)JtKur&E&D~Jt=l<qme;B$b5$AkpGiK{;Q7I*3cP89(932
zn-T&BC_~X&6>Cb>ku<j-UVI;by|=W%l|wQ8_Xvco7z5ka21pN|h_o}4@H%8FW<Q*X
zi!=}Z84E#+m!YNGD%^Oq9v4eC3#Z-=P}4rl>U9)b^zG1Lvm<5*zsUF=Zg^sM7L&T2
z7g-t~Oy42y>~jKfW_l>zSw-ND_Z8tci-YloBsj^WV`27n+_{jCR{vtm%&b7m&)ZnP
zybjTI4OnU01f_nju<gk^c&2|6F3|7D3KYEbI$5@R-jRp2cV-ohuDsQxJ1^g<%w4Xi
za`anuwlwO;GfxlT)2{~c_<1^Pl{1|GjUUAWvc_=i3<LIU9M3sNz<&lC^U%5}{O_zO
zPhVih(f#Ie^?P%+Dz)HyQOmgStQC76S;JSiS@T-!P5gV6Ehk#-<S47X+_v%{UtE2Z
z=Z-zeZ^ziP`e;W140q<t!Ke6(hAW$Ob7$4|Go1X)gYAnw`B|tJk3ZnU&C~tZy2}L~
zS{lGHhcEH*Awg_(BbawC59P<tFEg(X=LPp8_{ZcZRu7Ekh!0md$}ooYcf|6Bpg2BQ
z7|)Rp6IgnCjlDl4a<4Z@Jfb0)7Zs+k%H>pUJCw$*v(tH}Y6hD=$lw^?Og0mJt>RAC
zxgz~K%WTZ%?y5N)ke9<Vx8!n)Y91fT%i~d7^Vzab0iP)?;Aw{nxlOM~?8b`tXXp)9
zSy#-lLrb{V#}alH+#czC8L!(~&i!Uru>I(p9Np&@dn;73>EB9D`B}wP?YH@;Y&8ce
z)o?ehyL{28R<M2Qc*d@K{Lbe-7v?@-`*-#HxZgu|`md3-oFB18-eWFqf5JT{KIQ8D
z&p0UUId}X1f@3DU;;sjqIW_$?8-9DkN`miGvaN-mM!e@I4Ig-Fw~u^Eu$=YRedcVZ
zHog+}g%4i;%CWb;aoIh=^r`>B_eHO(skoj0#Q)+V;VLb+{=;!&|8lp_qSqe%k6-A@
zNU4WpB$px?X=rCL3r&}m9vzmIY_G~n$u(jwYLk`vcIzPd_ZLpnAsr-@qNy~qbt?JH
znMOJF)5!4BbZYT2C9R4X)JK{{UBol()5|%uZ}5C-x?)bMM;4L0{}Ngzw}J*`ucR?W
zYiUoP4HSHR3oVG<L5tq*CE5Ll$!7RTdS>iE!|a^Nxb3va1D+9gaZg$u;X|+01IW5D
zh-QBeqsGnRIV@(Jf*Gl#9DJQDM-|e{CuQ^?`Zo1GdXJ3$dqjhThx1U=TN)DhiINuo
zpoAX(=-cZKc)36xlS&jZ+)N2-k9y*WgDT3UKG>`vI5{u0u=37en3oO1>WYy-?HJTO
zH^7eX6X2yJL0QKHZ>8y&ZZQk1w#>!J<L3D3VFC4^WzdPX5)PR)7#3%ZpvX<=dBGOh
z4m<J6W-lHXAH)--qgYgP0ym58;Bv_UeYZJb+5{Jrwx7a|bXRoU=8kRs&SGn!2b$J;
zBKqffgr4?BoU*UT^!nkrqTuNq3V?3YMHCnZBGEetwNHbgFjU;Z*ItH#R~Q_whr_lZ
z0)DNLsBMqJw_nk?`S}V4J-Ld~{1~|ViG0fXIK&PX9q9-0kh4#~d97<0nJ8wU@rigT
zSa(H&$20U;629I^f>xhocr8rE?32kb4o$|c)MU|<NJescGM+>v!^tHX1y;#O)e&8#
zXGz%Lm4uTMlhCy$5hiOAvHJNnM2Q{Ds`3Qf9F+iaj>ng}IGFVp?4RYakUtxP?pasy
z<<S)=wnrnVdo<+wN5Mrq5(l&+pe6P)Wj(@>^(Pc~6$1D2V6l@5La1FJ8kdNB-mr_f
z_2mMZbNsQw!4C^eeeqS!2ex@$czEDE*6E3TdXoqA&YeZXq%*km+6_a+-N|CYX&63q
z!QG?Ikor5}aHa!%EbN8%@+A5zo`9UeQFL5>2-A=3N9%<>Fizcx-Id!=|7;8HeBFpF
zc^fG9UWai*R^yt!75*77hxgniXk4)f6StV->tS(&Ix`#Ap)+BYJ{@1GreMx%V+@g#
zFh`4EFOA2U6$a2dG!~=%^{_5|1fuT^L(5mu_f^xz4nr;6Tht$Jdo*y%t2bVz^uoCN
zJ;hy62|A)1Rc)q#+<l$k7104bQbf*XSUWw6`a;UWA=R(;4P`mJpsO>UkoSNF>ZVjj
z$9vr&U4xsXwylH|3Jb_?Tn?q>W{~TK6!O-+Mguy>lJ4IqIv_Z8eg+}*ar-4|$nmFZ
z1AOSChbLL~JWCqEuJn*y=-Vp?+92{Lo9hqLcAvep;P7^`cHKlOCF{s;<VvbcTS}kS
zEhHW}kBpRNkzJQ*G;xG6bv;CMx3dw|z8*`lqx9&`6I~jmIE*r{459pNZQ46=kmyMc
zqzXMPQnnvJtG;WJ?Ga5HC!<M4$NH0gTR+m>(2wRm?MpGc`qH<-eZ_vJFJ0}~mx})D
zOZzj$k7@mAN>e|IJlUVNs%g^qbDCuNMw3kY4Is181E_!o(7z1>Xj#huS}Ze=f^H6?
znX8A;fy807E@UL#{WXT7y$ortDbXMqW4ci=mF|g5PTTIeRJ3OSd0$vU@7`F_-c4dZ
zrM{60zS)wW{2p3idWgm)o}k`C9B96yGZp<fO-^=a$V$tTUfuPkGok)u<#vfeyhCVh
zemIr&yh2gVae`NvL?0fd(e%vgR2DCMdzr;_>0Sk`lD$LO6NK0FV1sb5KcN?DFKOo9
zxAg6CD@|DTmD--Qll2tgd><f(Pl|$hte}XPe%)X$c6!O4DscOyCNgjuur$_056wZs
zqo9M)t8|epc#s1YjKk9|6Hwm57%P~O69&w{?~QYCx5^y*&6mLFoh4L))}nFKMqutX
zgq!Vw#ri|gy>tS3zwB{Da>mf}r{VM0T|7I_VV$-&!W;cyoqSQ;qJq&SD-6ruMWM44
zi~X_J@LEd6!S|WyoSO@+_#!;XE)%XRajzPC7t1^!;IQIjl>0u1!vw+Bdiz1_ioW2O
zeLIdD$Z(mS91kBW&%&O<7w2{3FPnPsJ9iZh%~azFt$nyyuRpKer^P)>wE2<NP%f|+
z`Qs0x*nj<4wtQ;9H5(^z^*8W<v&MW_Zz}uNn(_b#GftQ|j~{+9=b`BqtZ-r(x6HBP
zQJQO5_49guUcQlw!?yCn<2!i#3gNRdJiuv6M>zZWaelGFj+aky;D{kktk=<nmpnPe
z4ymr(*V&yN%+Io}>N!4o$CKL}y|}Bs56`Lf<!Rgf`Rv~dT<v&~^JD^9eRmK?JP77#
zhN1k>@iMQk2;(OmBY3@GBr7kE;@*3rInn+K4?lI4YaC*@$NpG0TOG&WCdadPp9GG2
zlfXI2*Lc%`M4oJv#82KQ@vG2eZd;JT(SK7oGCGydny2v}!5;Lul+KM)GI)P;2DhBa
z<YJ>Nc6^e>m))+jmq-;Bv}Ci#<s4pOmCH7L^H}~-9`6g!=hr(6*mY_lAJ-`2#;--3
zcK-(V$}8qCF(qssRLXsP%h=GXoSl6t`0b^ef`xdC$7fY?+U+Vn{N^@K>0Hgl!)rKa
z-d(QWSIcD~b!<>|k8}Ur=k8<bId5$P@9=rZDdmkk<nJS1H0}xCwrS#_{!jU3*)#U~
z{+zE57L1>{qT{~1nN7T3vqRz=t}lPfrH${n_Du_)`uLu0KYn1PH?7=W|B;IeKe2q+
zXI^@^jTeh<dqlUd+*0zDYxjTS&i%jh7~%8!H2sIjYyaRF<DZ-q{gXF$YG)mbcGmN5
z=NmWLS$f&d%3s@A<`dJeeiEhcljy}ii5_H6BFhA0+TAdj^ruWA&sS4vN#%6<@qGqe
z+GIuwgXU6%_J4GI`$979w1k#4FQ=QTtH|@rIy%2#BYj_EOGmSJ(Q}&vv}?^VN_k*M
zqsKbY<?E+N+0LDmUCxnatv4layFhD}22rDD813wPMYxXRN%3b24WsLn7+65A8f7#o
zw2FR=tfM|D4@q~zGYU?9O-skM(vQTiv;x2B{U%vh<#hxPR=`!?uK3xv2hLqnLHm;4
z=%~~epYCWPIb;w99oE5cOI<vhqK6o9HyF^@2t5@E4nHIWHk%-zPB=OXX2B+AF7A4o
zW5r$z%v!h%%_FRE;?HU<D_@W7GaK>4Y%47PY{#Ui-S{(OKjaz@A=l;@YImJP*a&;{
zdgOrCQ%>kS+67~7of7khD|)uM;pCAs*zx-;a`v6WoMun_HupkOk~ije^F`uH!4Hh|
z$K2)%Fwwe*-SaOYdv74Vdx{-TL@@lXg&;C96yKsR!_qel;fDm<Wnlz#2SuW@*n=nq
zMWN^1XuS9kjWy1K0i$viN$0O(TIU#i+!2GITQQiR8H+1xW6^v;{3wn^q__|1{EWpG
znK;<}5<8-|vDjW2i@o8oNZ1i8dKIy-d>#WG_ZZaa#vnBNDvGCH#qq)`=rR5ZvV+B}
z^)Cv}OQJ9_JQAbcMBtiE1bVIzj9}+5gvMOP*78uyY6^kF$6#z|4+8xS6z`swQ1Rs=
zdb|$6!h07`o#&6Ta6fEw64|A-K7yd=4VNBXc-`cQxAEtYa>xT)r<}z&`7_WhbHi2_
zSEx-t4e757exy2M#a1Wm8Q_RTx9!nUI4-?MoJ8NN$MC8BFbZcLMA*4~h_BlXS@oSL
zwAhAz9$R2qvJuDIt<h`LIxMqXjrdbmIG(Z`Mo*T&RM7$h4Hm#~<$OGMnFE*PS=iiU
zik$A#5N#|zKXy*S2BDVMdN2XT-HqU5Iu0L?je&alDAaw_Mf-?hSY<N=FM<c*Si=D5
zsP)I2|1|K@wKt}h^um-5$`~@aJ7ym13gsLHeE-`CC#K0ka+ATuxL;H<?;F|v_(;7I
z-qDIZuc&dxQ!*aeNCUL*Q&+tjIy|qE@*T^lQ*9ARQXXmLW|8)aG&;~Ti8?)vqpdfu
zP>0(Q^!#loz10k&vJC;${e~}1GW4QHsU8#{`i(1JpQ3GkPUIrk<ZmY*r@mhg(5Rcc
zXkhVH%6lp9&ZE~*L&ypmhsD(V{Xa4k{>qAKQ(F4QggR<TwEfTox*}&ls<%eduxBGk
z#$p(mj~YT6y9Uvu!CK^RtVzeS`q6sVzLX!NLDOFKp-unwp_LWt)HYk423GW@v>Clg
zp;(R1&r_q;x2kkJN|oljsnUf2RZ^-`rMpIIG$BQea;NvEV-3B@b(1=sd#+AJ)B4ac
zmp)XI)`#-NuNw^wI`l+?`gZ6?Yj5<YVMbcCex5c(eAc1Nzq(X5do-=d6qy`MpqO%|
zh|MN+c%X1qeKDioE%RxYk_CAzUry59Riu5ynm(Ivq05VQQl8y@%CA34W8&=SpsW+U
z*nEo6>_#6Cd(g09UUcMxFQwcKAdC7S>Mt8cSLQ^Klo2ETyw^zaLkcaqmPu(Yx#Y03
zh|)Kfk;>Le8hfaQPI}%a!<0rE`|K&nXf{*jh8D6Z{6uW<o%ZDX7QK5}$ZB-N&t9D|
zU|1JCSgeH8VLh?2mn!B*tHay6ADYDt-f;O~Ts%Jv`@W9E#*?C7rEi4e;$Bt$+gRlO
zrepCmGe{ouq4H%B{_b3khn-dnudp>j4sXGml{=BQdOt=x97U0h9r^@0VtR`U8WxB?
z;hnQ+*mNGj{e9usasg)bfpC8wisG)3_-~=;N2JG#=|p4%MHgaFU^a{o7hv3;60CN<
zi8qON@T{#)@E;yRF{KI1$GpP+>+evs^b=moeHU5C-$*o-<=qAyd5UglmWFj<M?EE;
zDD~vsi+Zuyp59z_QG@^I=sd%*{NF!Lk!)E}NGK^u(|x|ruTi41Dutq`G>|lCAtQvy
z%wCaHk~CD5k~ZzVcWCdu{-59fL67P<ibMBxUhnl%IIqW2a)w+p!I-<`^yfF7W<1$p
zFw0k4a;NqPK9FL?R^6HPQ*Btg=NN9;=D_>yCUC?97v2y%iG9qbvg36xE|2%&x>0_-
z<K-M~D4Wk;rv`C|d<dHyUdpP$%ej}%Y94qrl2wJK)A>7=Q%=V5+~fp)G%k@BOC^hb
z`g+csnaaJS)404io%ak8x=uqDU%^H;-;={tee$?VTt2Ja6FsbfMZ)7L=Kr>p@Q|CO
zyzN&RpVTVnz5Ob9$Dqyp*|d`H_u9h8WvW=Qy^2kaZso)bar=00=Ye8NdClEwMz+ZH
zIMlG?=NdjKv>zqMTHg1vmc2IY;#7+|Zog2+-$Hh?lj<H0tlPua-1l<#Z+khsWFO}_
z*7K>4^*pa)Keu`u;L*wlS?%mWUY6Xz1#XSJTjvnp?l{EehYxc@P7^1E9pQp$N4b4;
zGy7Q{<9Eg_yjt%#-_<?AuX>;42%}T%HTX1-`|k`ZPCUzV=C!g+bQ@O{o#UV4_Sqq_
zKXaun@!5WtIojn4n}l6ukK${*<HU77^7;lFY2M-h+-4)u_j(X}m(42far)8wTyXmV
zUwikE&qzP!e=1M7ThFH)p!<w>Ydz;s#dfy++0NCEI|MVelTYQp;Jpi8@}a@6IQiu(
zK2!3V2abEg{_StLZ^T>Hlzqn!;@+{ww|6|k_dQn~e9wDjKXAs_4;&EoftPRm!1|RR
zc)0dRYPB6nn{!9f9KBIA;p!;rdc=x?9!iAP4XV7%bbhNf^*?G$HM(O+v0*F~m5!rX
zZBEp~&W-y1nM`e6JZZ>eZ;I&jCBw7+G~japr7sMjJ;TDN(k-0WKZZ75j-#C`6N!9M
zs3A6;8sBA;+_rrBSY1L_zEseowcBW|*vFBz-AB8<8c8;(neLoCMdReplbY*QYDv3I
zt1djGP?dHnbbn2qWgqGGukUny;Xj)CvkSVpcEiA3@?wicSul{)uxywn9542S?aJN=
z9nc5TkBu;>tUm%~4}^bz3w(b)6hronghd2l=xC9hkR6T1r^mu|-FPe=?E>@f69s=@
z3U+vSLG_n6Tnl{BIAjiXHO)h+*gvUn3Pz#D671R-iWToxV1-RM#>B2cqfZPx-^aow
zDh^??3D};n4rj#kc<+*AjK8}c!8WOoD%^n1H))VX1}G{M^BS`d^fntNy>hVBIv3{C
z@-QSQAMT+A2o5d8l%OKq^AZ{dm!OY!DPr11#<r#mj{`O#xmP)gkCo&7%nJCutH6on
zoAK-YX0*?$#Qc+$m}{^FZ6eo_ymJdOpKifD<toe*doe?7t6=L`CAdyi@E=-*5-pK`
zdAkKq4sC&D)E2BBxdksDS7KjWC6szsV!*b|a2zD=E75_E)2u-Mh;m%LwFy@SZ^FXx
zGISm(MfR@})LWLI(6<;V3Bq66Qi%Ga1z2-6AB!L7;lhht+<Tpa+b=hw%d>1`-O57C
znM|D7mw{Ww=`a#n`SsZwpl_QhJjE0!yidlj<4MrWNko|cI!J~jVAc0HI3Hb$({Zu*
z>ktEJr6@EcM!@FZDr{Z00*X&U5#_oRTJ<65)Coq9)q$wHJ|7(xbMZWWHtydPdGY}>
zaCG@}j6UXxcM4N6aEj<a6}qAQkuyka;U8G!h^Bq^DE@6H7^5~2vbErbOK@|*D5R|%
zj$NlMp|5L!x`l%vIXVFAH2Nbbz!-;`4B??6@->0Ip>$LSy|uK(4W@yo$JFphM+NJb
zC}P0b?oc-FhSI36aK0x6lhwaT%I+KasC=ZD=dbDFsSaw``-G0}yiWo9Z;|yy@w5Nq
z0tH&NQbzJgTKb`x`Yt(4vE2^RfAxE*IdT_a<__BExs_G~Y^HHpWi;Yy5seUA;ySq-
zDO)R(+_r9@pA*(o(C>A$y(x~gH^k7VYvI(dCX5D@hmhX>01ACGhn7y5N!e{))F;fH
zK96>#i^Im#YkPa@5o$x{j}qmZkEHCQmK3_yoDyRP(3i6&)Uu#2Y0otv;}gAUTaqqG
z*Y>0sb8S-op-JCmHA!Qv2KA~`CpR5+Qp)W?Dm{D9=K?kQ)mM$4)T<Kvs#2`6DtRfX
z(ilBelA5AQ2{oz|BYxjzn;Kmn*@LDY?LpqGPWv~j)5&k@G|^Urx^2*)&c7NI+o(x+
zu0`cVJ;hz8E4Fs@Xwpza%J^bT(Z5VdeV{pwUuH>x7exQumdN*vHH}?7hUz3Ddpy8}
zevO_id~#1Z@zsZ#vu0D(^aV84X%Q9qhf-4IYVzq8LoRM{bhCOLP1ave`Qm%|!a0+Q
zlya!$RX&}6SxgoR<uqc#7V1&8os<p4j_J-lG)VNT>f{d7iMD3)D?Uj!t6J%f??swD
z=^E9z-lk#G9+1qkr*yHTlO8;HLwfXyw0D1}kIsK-r(74*NXeq7W_Ph|p@><zDri<v
zhhwD{R?pSN2p4^n`1FNoRzGZgYlg*(EwEgDIHIqMO#A--aAl9^ke_wL%<s-38#ft4
zOFa>+H3P0yv$0ELk3Z-w!Y#Q_1Zl6v3Fl};U0;jn-s`}tlkxOjDxw8b_T8Y3c>OUC
z3O9<-bY1i$K5j-g^KH1ewgypO>yW>$9utQ(V$Js>*m?6fMqM}qzdPp<|5I=s2H(P{
zrT4M><P&ro*oj*uZ}8jj6GHZW$9RvwnBKJuzZafW{5d(cY*Ju_{mN{5K#g;nHQ48h
zHb=hJ<xeX5yoU^VORzET+1g)ZX3RKp<Y2y&Xel;ZhO^5UD;CIO-rUoMSEh{Si!u&u
z9ygx5c5~rIkwf00J%x=nd$R9PZ_a3*$@eGD=9v%Xvi`yVHvF=XXGJXL3%{2M#>5JK
z@_IFUPG7?=zA=1OVJ+wHj^p9(3EcDbIzAhf#Mk84b8Bh}JIZa~SF6%E=y5tn+GTQk
zP8P3zoXuLsIlS3BmwP4UaqNzK?l@7v3$GROs5?a*bEBA3&X#b+-cr`jDdU+ToA{iq
z(0k;?j`H~mKAF6kEA1+I;KxdSR=S1fjH}|2udDdOhOOLx@HP%;-Np{{x3i0MHFqnn
z=AWZ>@U|;EI8At9hAKPxa_vr*bFbxo-)lLcd>5~Ft7GSXb^PDH-JG~+5Bm(-%TM0y
zWv<`HQL*)W*j?yB1H=s^b5P7n9OSyQ4cxG|k=Irn;ukrG`SgY+{*ZixN31`}K55Nt
zmV1n6mA7!?uH*c%`2;syJIO6CPjN$+Gdxc3ENfV`@-g=|{t$AGN2i=;*R2<LXUj$Y
ze(w?=|8kitl&-RA-)sD2*md?FeS^v6reKoZ68*T_?CpMs`#arbz5nj<CDZ$SSLp%Y
ze))jsAA86HG9K}9&&S-Z`GoCSpK#)eryMG7p`&}AvB{X{JpJ@@?mniSTX(m!v33V9
z3+do*4IO+%<brybcJgJ%PWJKa<S)Z4=%2R*Rn%Hg{E#8^;NcLmZ?UA{+e1ltz;J4+
z8$owO-!o~ogvy#gyX>t=O4*Lmdf5w&TzFWrPW17KE4}IFPK%a$l990wx#-U#jXCq^
z%I6>oxU+<Obyw2$hBdTt&RQxlT1UJ4uBR;jG+O*Ji=6i7QRazal2fZ7#mucV+-4_@
zQr;_AcVc7W#SuztKS8V7Tj^x`C35PxL9H+D(b^ABsJ6=sk`+9MkyF1=V9769A|(wE
zI~g>m$zlF01q6DiAhN9o6kUYAdtC>u^8~y1S09w;7{hy*DUM$nh)(gj@Ihr5^mmTL
z{s{!tXa5PU#SXJo$3mfO9Il!<LvxoKOf1|52f+iqdrn77^bFKL^uydy{umQCUwBl3
zi2oCe-)2j2+9?!q;@#-xg4GD|kHqiE(J0Z4#nQO7nAsi&yD<s)m%9#;&k|8IFd4QY
zx8t-a1*I2Lq5o|IDz($0VIg+7tTUlIHVgB{WTUHOBic-Iu(7+yEDQc))V@4Sh|EVY
z3efnr015?#7(qqoaj^&y-o==7w-}2imtg<l5~%5yV$QNs6dW$amS3e<J){h()5_q#
zybLQ6%P=vm4BwK=5F_rB-+pCCfXF~Al)>s^DYhq<;vkhG>tP9kB1*7axdg9silLxY
z4CRy}SbZ<V*%^iC-cSIEaskSw=Odvs50>|G(W#z`4`XtW9lQ}58QIvoGYk7qW@4u3
zYPUa6NBY||WPcQz%*RwDze&OKXX~-#Rx+lG?clb3iTF?=y4!IH_%uHr+T-F7VX_u>
z|6=g3HX4=QYfvN|0p-e7=rcxmPIp65v~($~yDrA-j9?t;6@)XJ7Qo1O9-frXfr<Vs
zbjz8Eo$@}yqnd_)FFc^@IR!V5PC|mI8&+*_hT_NZDD`j@On7@Z_Z|(SC>wE)F(MrZ
zN!6p!qBa7~tA@hzu?3bno1@DCGn_UyMM<U!W=I*~_ToNx^H2{<CiTMDV?E(9ObdCN
z)y4Np4M)}~<J@-z%v{(VTOY{ct-D}iiY=Se=-)JY**DS+`bhc#Z)oYFPI8TSN`JB*
z&_lue_IZAd9vfYvM~mC2<@_mnV%<V<$C{{ZdIL4e)sxY=Ix62<LnVdVX!Mp!T6KC8
zeUd7n{*w!6@_`&m7?wq~$I__GKZW}DOd_4T@f27aOY02R2%l&rh5cGW`_;v+=eT(k
zUFb)xD&Arba4LPAHIc^Ii5#B21LZ8RqYpcnnpLc5M8+`sC_XF7f1A<SFa7B603%ve
zZa^yod(-749s1Q(n{w}J&|D>TDo$6UB|fTj#b1TOij+y}ixNf5RifNSid5&XNDi+R
zXiua9x%N_^e{J%#FJGRXCCF1mt~@!l%F|bU1@cc(prs0m^gdIO8g-QDYpxPi$SG5H
zkTP95rA%?=DuRu!LStf7DSf9JrFy8-$aqaUKdL8%&g?}$9_mwJWnYrY?ML2KW;Fid
zV0x`OjEaLsk^EDT;u>3e&tu8LXaY?g>P8wtQ)te)X;km!OO}dr>GktK`mbX#y;oU5
zSu-PPLQpIXJ`_({#!2*ca|-RXO((tYS>%2pm$tPOQslK#qAr^$VZv5Q+q#3ib?Zp4
zcpoht+(0R9P2{k=g^UNDCXbhGv~~X_T9a{|uB^U8mdhW~!GvdIu<Hfw`0|#_eLvIF
zH$P~8^*=gzt_!9-5*q>P^5`>P2?h65MSoBOl~z45P`@`yP5a=lhcT{JnZiTI9I4wZ
zvDRZGX6pl+{@OtQyFDcG6Yy-P8|s!#LG!t3I6YeI8(j5AV?rQaPhX5#Q<mdpSOiKR
z#bBX+Jbo@s#Mp=Hp*KGbBjmE6b36wl3kz@~wFG^O%At933sxvp<9tvpyq@ntx77!b
zY;qX$+na?Zc@ipHTcNc1B0lZ9hL`7VWBL!l(;e{)PSG!rcIzz;*nP(H13zGA^AEoQ
zyKv!uvfQI*cV7BKflobE=AJjy*#D9SSBX34$~|4a_EwKGx*PISGh;4y@6Y+dw@PU+
z=MmqAumcZg|AbL&ew}#8Kx_UKXUFdz+jExvIH9FDv8}2bKV3PQ{U1-|*G|*Kjp4&3
zz5Mu9)Ev%!IFGYAklQnZIri~lo@N%xvjbLeM#*Y^aypW|+N1eTL@dAmvz9lmis#r@
z2|V31kxv{<;^pe=d7ys^S68QUTjvH|sGrU&#%J*Jz)U_Ao5czl+5CFrMo!7d;a+jM
zd^0#t^xN}U!L)!az8CQQV}-m{cw8BF#T@gsm@ikAu&jG2Z~k7&!-~q-2AkOZ#wPx;
zxSWs3Rj^rU1-A^}%t~iBbE0o0@BCcJ-$c&&s!0{s9<Ab+Gq!U3@2&i;Y#T3f+|C=`
zZ|B_7YSHl)|9RzV*sP_7Yh!ovLx);Eud<7y@9pAmwRJpr-EP*NvxiH@?ByM%``DmI
zJzxD>&zs&0Mn(GpKJxG&SKMphQ}-IV<KZE0d48Cc-!`$hX}LwFnUmFzaa7+H9yjzj
zCyzeC?UPUP-`S@)Wa(+KiF}4nq@Lxwd97S4a?jVQ&hg{zf>XNv0(aSRk?o2vaYNc=
zmWjN=4gOa-%;6eeF}%*9U#|0?mK&^_e3L`R-QuHvMD}^dZC0CnhqK<?;eRoAxtHQS
zE>F3~<GbAFx{&+qd+9#!>i>XmE_uKYc0S<U_a3n0Z;=sFe8_p@OzFiUQ<6JkN@ZdR
z?DdlYwDPbS-9IysY8A}MJ$^9#u^K{KOox(`$8btGHj?ItNT^^2)BRK%I{t4o9j$jD
z+g%gr>T6f>UhGbG5-)0*FoT9}olS#17Et5#MPzn3ln%v*)1Uh>^lD!`*|myo^gbKt
z)4oh9Ps|k^$|4$cX%oF0RYi1U2kFn;O{Izl$n*GNx+yYyCvDD9%fItvdg3aT2`%UN
zoQG6A_BjogyrTZ2KalI3Z?sC}GPnJZ#;&2V&|J|S`%WstypJlr#;9Zc3oR`5(iQvY
zdLpxGh(-I1v2Bnk9#ss4mv|TVP8$ka>5-VTRD!+_7`Gg4(NR4{JQp32G-m?t?sI{H
z)FedNPr;TbPkgGIj;FV0;QMz!G%EX}spotYY6s$zd@z=L2*I&)OHou5hK89daa%nc
zO^uNl<PnW8?_)6AaxD%A#NkarJgyy2fLFU<;z=i=PE%++hU@XuI0c*gq{2{R1DyVD
zz^_MXP&$+jkJJq86g{n18d>N(n}zTd*@)2Hh?BzW8Y=FIlG8cp7$a_ry}2mum51vq
z@(_I{56P<ec<!7pHt_NhSecK+=6pQ8oR77)gsyTe9}Oq-v8*N^`;zie;GGY9BeCQD
zFb}J<^Wbfrhi&(BMRy|?-d%DrI57tkL@ws_ijDAml8w_2*@!LA!l^HrxHBdbQpp*J
zJd=(r>2$0ek%kPR$NWl4Mb{lEXg|Fkj~*st;HM<~5npe*QX;f8*5Qa&0uF1&18Q;b
z@4gnZ{=}f>K{PHOUV|k$5jYdD8a=F5B1&pGY)&piRKgO>9UTJo9}BU)CJ+uY7T}-U
zJiOaA2ZN^0!r9L=@hH;=NrR^2p3rE1O`Qt)=i)tf*+gvq?SjI1CwMB1N1rqYII7vB
zPo^EZ_OO9l24k8k2|nv6{81hObMZ4_r#b|4GtF^8b0EeSm_nydKj>8%!N$T6eh2h1
zb98U0pVNVlmo~JYYGB~v9*Fy;0^KAfywg;GdW9THhREVTQy1j8Na4-FUo_?JS5lY$
zNM8oNp{LV3>2}&vdeQoTtW`zEX6AJ&ufHUo{pTn!^)$IE9~WCTN2t=Sk&?dDQ@10#
zNx5hz`EJ-wX*pZyr`QK?6+7W?hZmAwPA=V0%%+gMbdtA9rMeeMG^I9y9>uMtoGVe}
za%dF=H!q`KkAkUEZvi!h&L-OqA1YbpNwH>=Ddd+kEqLchwX$PqpS3l)<x0p>Wdv=m
z973+%gXr~8Q`%{1Oq#xZX!P;kbTLqeqP?}KYq2^F9;QYS`YNR4qeS<gDUjnHdD?lf
zJNdtsqhq7x$hEc`-7)JXHV9>@ueB__xh6yH;WAV-LPqc@WoY~NuJlD-hQe8fR<D<#
z7awFuaiJ_(y_Kc&q20*pOE<C)krR9gIVyMQP9Z0|({^8Z`fs=b`T8nSg}9;qlTjs|
z<Q~*xl_uS8>Ph1rdW*ic0bRasM7!Gh3l8}}8Y(x00w)b8zeX!+8uuUN{<I^<YYwzh
zFg-{8ok)G%JSgXcH+9aMC4Bh#bW%Q;9&0ZnPv=!MaeEY9EfCD%uL<P2D4EXuO(or$
z42oO6k(PPq)7t69RJ~*qtt_de$@jO>@xeQ3Z|ZJ}|6Wf+mNe4%Pe<ratYACnpP}0)
z&(Vme%al6)2Hi5fOL>A%(>DA$y$^Uv9mn6%jR{}q-@l)9a=8??Cw0Z8#%}nftbous
zWqj`4Lue+Nu->nOll%3=hJztyh#b4o3Nz%s6S<tYVVE_<3bQ^j@-K*=?PCrwZF7R=
z+lhjg=7IKj!L<LM@18pkK2sKA&nU5pFlHsp!`EQOyI2gQ1UMBXVYYTEzE!7V$@FZf
z8Rg-JOc4ybl)+hdGi02$A}4zXWWR{6=c0Y6_<2z5Z#7}?q!#SaI*rY*+A#d`C0sap
z9qMQA;Pjn`xc&1PT+CjI-sd~)+x;2u6@Nl``9I{X=)w}na;R)~cDtp(r}rrH?Tu=@
zB1(f(7HjjL1)^)UK#v1L`iSSD5s%I2$L-YvxVUW)_xNbR&-90JwcAL36DQ$8P0Y7{
z*su?cVbur+9(`mykLu#W{T(N=O|m<ew|a1p{B+(ib_Oqw@Z+O`RbFy)9xMI|U`6d<
z-fX&<Z(A(mBLkQ7f?lim>+f(bxV(l>I>)g5)>t0@Q|#(E#B*U{0@JZ|{N_U<=jbN0
zp=3QLxTf%#>8bo<#s(hlmBvNm)7gJ$2A}Ma$#SBLzN#^c7sX}sIERhAP%4MV*XQuW
zxw-sPIgeZR<gwY5e9;@r=i5mIJg840o9!)RFNY$wx>3Xv<`=VqxP?9^l<*quQr6s7
z%Kf>FbuN|hgaw-fo3orxm6Y=?>k97ixI$!pHuHetm2CT@lCyHQu#IaK`^bwM=<HVR
zmA;KnP20|2`d4$&k7};Ev_o{bYj|DCPVTk1ma{x}am(mBUS+wPqmA}(QO~`cp}LRt
z6zcg$xBa54et_e;9TYj`25wPo<O8aQSYPun7wR<eWj&GqF+9p&jGDQx$uW)>-dCgH
zarWyi_+1(&*+=daKmT@$9iE+L?Q>`NQ|(!fOl{>!bK6*B)Hx24KF`vp&hxss3tUSV
z*|PH@Ur4*ePJ=J=;^UWDUT{sHU%$fVCST>K6IZ!nuwa~~USr$G*TnGEb>1}pI&;o-
zzPRr?dkep8*;GS%w91hBpERUy+?U$!^`+l6Mr2%TOtnu;=&nnD+WXd&q+89%=g}b2
z8E!#7$1O=MYdC!^8%5jS5bchzrXr`&r0e5ALux0`Tn9IjGM+*<PSYs-=uDC%`IF*~
zK(ZRLgbejo(wbFKv`^&s8pbEmv#lxg*fxXwh2HBcC*Fh1%IJ7dCAlBpPCd+bkxkh?
zI;GP{3v!QAQO}b!tE81)m|Y^9X7T>xf0z2`Jf?Av+UdlR*EFR5BN?6ePXE37OT`<z
zK>3I)8s2w@{U9Zb4_3vpJ?dEdTMJ$jbn!k<4?!J$@O6|i?j-cbf~#iuW-u797g(Zd
z=WwLF8-;)7qVGG;8iVugaNvYJLSH)KN_QuW>*ETy!IKa$Yzho4Jn^^BbZnBFf%0d*
z*t&l<eAmpyO34CLcL*PAeK5TGF2>XyOA$684AJ>3VEt<q!Y4)GZ{8YYU5-Zd-x!e>
zS_{&Q!#|C9SScmo$nOL!d%g~qrxI~lbb^L0PKLSFdK7$J5BHrZ80wP>RjCa)Q?vnA
z76MXpAPu#n(h*i8K96)VkPx1M62ZdWY?z7oxtSPNnu*HunYi*X6Aj8)m|>8GW|J(u
zHO#_S^(?&jnTZFtGV!h^6aTHuM5k3IblznM=0yf}3r5o_(It&ql8&E0(_p+N4I_o#
z6co4tQrA*pWt9ro>=dl+T#ugO-{>5j48LPZ$o!Ltz9SOhGH)GRGZQfTU_6fAiNmJv
zYavk=H=b!Md?^OGX;GNBITDe(!(q{|3a1-aAY*?R+G>`ep=1fNlS7cPBp6TJgK)qy
zKxj<!vHqbyF4fHzo5LbQGj1j#m3^@N;xss?d1Cd1shA+`j;r+(F=Bx$lC+%hrFlFm
z108W(XmjiL3Ko=yEfl}}2ajyP-%^64b0g6+U^wP>9g3z>3oL>;_TDhVsZdkFn(v1f
zbw=nn*$@dI^&y?p8wnC!e7~nHdgz+49H5T+3#zDHt_&G{MQlCM9ht%1a7t4KQ=6o*
z{oG%=*XJi~3Hw6PZSU!y$!oe4+d=hDp3wU759q&rx9M{K>olwU5~=k&N7{Q&({7jJ
zWF(%4)ms}WVc~vyWV?s1^sA-m2G!&~u!@#CRgi6TDQTQ7B#i-iG*s}D2KCFJw3bvl
zB=`z%t=5sFN}Sk)i6*nVtI7FAD9zRkq24|LRCag{l}cuk??o@FiFK#LVyo|s^>~VL
zvM2YIHnjBw(ZFFNNn3bPn}f`$aKr%mp>IOrW`=a%M~{vj(jk?}TGYivU33ssNvTYU
z3??bi&WYWre^ED*43;Ikzg?+ArYpT2)rAb#OVh+JQsl5yif(`WC$ZJ}C;2?}uOwpE
zAITBRKa%&4e@miTe@jx{{+1}Y{gG5W`y<h>{wrCL{ZFFeBt?5JN|AwwH09rrrj1j&
z(5g#a$a-2=dhxp}Rb7{%u%EJYBT9}67s-?MX+`S0R)y@-d(gjonpE$jLqGoYCd1=>
zD6+zsdY78gh697>)B7PbPizk6wuyb@#n$9bV`$@maWu->g-))ROsB4S(nrr3H2cqN
z+J9yNjcQy(*RF<=s!ljP7#l-3bK=NYDv{#i*Ha(kG+J~wlSbC$(C+L4YECbqzNO_<
zb9f6Syx2~*#<k*Z*+XdDPp&-=(X*JNH2V8-GLJq(?|PmmkCT^a>-rnCb>Ur#3w}gL
zGM|%<=u^y{A^PNsUuo`#Uo=Wy8rLK;*q<ba^dAb?kfH*A8+Ew%(1uOdUO206fKVGF
z)W`S7fM)}Re>DXBayT;5B`9;Y#tVZnn6EGnajGs*vYd=wA)aE#-3JOTvr+$kJ{%7R
zqrGk!)?HeKM|#mX(6$!!hU?I|AsKc5d(YY<_7?&-LIUyrV_b|_lT8>ewyE|7Z^NG5
zHRz$Z8yPF>@%u{y#E3j*m>fsN6~PfsKZiu`%Sai113|s+!btTIntMFQOrw|RIr<%x
z1jou}`w#T~^cS`Ly0H3f8E(v$;}Q=A-esW7_us0r(kXTBRjI{~l5}`*SZ_Wv&w!WA
z?92DPO}KHUDc8&!$j-|KbN@I?Zpj<YdAmljdMj~W`+poEW5-Iy_M9@>k%#zA;QkRV
zoVRfzmsPv-w8I{J?aVapchQG^F8FfX$=SSn?_B<sw}2y-2C;|jB0eCsgm3R##?n5^
zIqu_1J`o<y_!P;De4=^H?HFzq*`27Kahz-x&(DS>aOUuJylGG(oApj&_b$mi?r}1Y
zKDeIOCZ_O|iK*OOZ3F)}yMdF}q_M%EbnbsOoqMdvU{B3VUR0aOSH@<s^sOvj8Ia9;
zzh?8$xQ)C>C5Ip7=WrL}Tz1$gd@xBKzdD}BeJ1C#_M?0rw6uWt$Q1I=!a{Z$QN&L!
z74h#S#oVS@!mAFK@PL3){@$~U|Fo5{WBew*V^_|Hr7QUPsS0+<5Si$im7FwW3lEg7
z;>rhAeEHB;PAJ~S*)iMs)Pid6oV<gZY-(7}Vkal`spa?TyI87w9WVS_$4|cOX3e*I
zcvt&gp7C%WpSUS9(HHl#@2LZP<Iq9Q-_^j5<&C^1;}AcHJj~tZH*t>h5gs`BD9=}D
z<`++z*=zqX_KX)hh%U#uM&$(iwVvRa(I?q&u;{{H6PrMxq6@EhhTj&R;iE=pd49!N
zKBU{qiAk*-^|qCVJGXI{Qt_+3jk_70<1LfKowVc}>&Kkq&gr@|Hd2>-PUw=xh+cH<
zd@r)t*qh3d^=L(%KD8?Jp_PS(WH{T1M$R&!&-wjnp^_P`IW&kmt1QUv;!v6}awJ9G
zme4-2hoSPwj!YaJ=)boUX!!#-da65>vbId8xru&s^yEB}TM$eymxNM&M>s{!kEK89
z38eQnnLb9Qk)Bny=(Fb2WAzeR)xUz$oVL=})iv~`Za2+&vtO`&57U(8$7uTTQ#7*o
zIl3KvnZ|s+LH#4{(GKk=BstST(=*;sQ^+TBne&631OL(K<}UEKCJW`a-Elxk31&l7
z;UzXloFlX_aI@G3Z0n7(j|RA*BN#x|{h;YL058@If=+=2RO*I8(lP>FFIwU8b-?k4
zwRncwVcju%`0j8-W9kI-pXUO}h>1w~F&QoUrXpdU7vdGY5n43^%A@?C);Jq>!{%ak
z>3rDy4M6GSg>cJXgkDz`<J{k+urvsRiscFn8L<l91H<929*Ntp)}U^GG@Kh^P>~)B
z_Faq5CUK~G8iyf;@dz3(_FUd3U~bAf-0hQyl{*rlZIL884@r2`GZ|N+lHqwT8B;9R
z<3RX&EIqUyVIS9He;;ud*{9&ij1;s5r{Ke~6!<Mpf#IAKjCW4KtAQynmQI24h4p9^
z-J~cNaZCM8#*(ec=r%rCaKn?}yCezgzbE3)DzTOKVI4M$T#VC&1egp<KuCH#<_RCp
z!72_eYu94m=~x_+i$%eh7^r`ULj069_)-&ri+@&QrQ<40N?d`4<}fIJT80vXrD(Qa
zjFvfz@Frp*64C?FR<Z!Us^-CUhd;hn&xT2rA7o2r;%K@LRz^+7#6U0jPx8RU;Zsnf
zF&RDHyFu%$3oev8!F}m?G}<|$uku)IxiK1nCAP?!Z4EDD#*-%!EZsCp=sqK`S$i06
zTpS|UhJ*3eb`V0o4Zx?`{%DzR0?j@~Fu2tRvRV3gCVIMU-F0!~q&DhBPp^E8I)b~X
z;dhHNM#d}Rr-M93c9+AYRv8@K&;@@cNkPx-H+9|pjUJ0^#lPck>Bz*Flzsm>DTX|z
zR_XhcRB($lM_nWLTNi0mY#W^!dYTMgwUBr15!w>jNQ1rhlYzKvN=DYwajvFp_bU1k
zUO_GOrL?hY5xtw6NB!<((=`7Kij>+wUb~YicKJFAbcrLskuhZ9Cwfs~VbroMgkpaN
z(1-wk+WUScarQKdn>dA9`nid%oC)IjJeK@iY$+sL_%=UAQp}p6w8d~RDF`ih;xQ8{
zIbukI?&{H2U0s?VrA1>^)v5PY6_UBENCB$yWS-rPH2h>}R%jPGeO`);uKtyDiTN!#
zlKE3I?dx|*XVEu_`?9Z+OTk|xz2ZJg?(Y613Htd_a&GoVN#A!LBrmpqkT~UkkeqM&
zASv$kQR1`Xqa-6x{27<elCu-ONTzswl~gSJCaH`5F8QAELsC`wQ$lUOBr!&RB+BCN
z;&>^l`Xo(BEnR8RM_EdW>`wEiD$>%0DrC8@hhYC|5!dU8_aQy<9%4wvIwqvlcYt{R
zGN<$$OZwh*Bz4Uty5?*{4*l$f?>(MQi2RXHygU7R;YAZdXVQwk{$%qtfUbNBp-p<r
zNg*_X*8Ps5w_fpdqb-q+h%WiYpJ{Y{PZp&`<<di+Lb^Y`l-lhpXuoq6Mf+9Lhp1Yz
zsoq1nPxjM((?fJD@+i%JdYl}5&rsO=bF`x5G7Sp6K`mZ)sc6w7GT!x^stsQePQ9nu
zo4=CMq2IJ8Q5v@{$zTw5M{u(uzRy#|?E!*4EY}nD(!C+CB|LaLV}!+-iY<jfVi&^_
zS1*qc99Y8LzBUMdI|jMe$3gqH3$%WT&n#;%OxipH4+qb|;)e@xxnU8?j)cNiF!_S}
z$G|)|4x`Sl!%6G)cyo3G$`)l}(|{br{mjRXC&d^iIs|?nD{;T)c5I%w69s}3cjQAo
zMtX~V&&x-!!t*#hUz|pb*g>Cebs1AXUPnda9e8Ct6pYblaG(1Et#jT&X7MMeB!9=N
zx<7CYm1Y?&8BTBN#&?3{`J9mwkA0}ZdKEp`c%dfuLr;FH(u=iT=<%X6eK@nmi2LO9
z<LcM}+_ZEMU!QBi3%rN&{mCOZ&&i5A?ZLgRtvPh09h;fkbFiT!2lSZ0WBxj`Zo3;R
zo|()eH&5lgE4+BiXm2*}I+K?*`f=6#IV>$Vk53gYV8^~e{IDRH|8!l<hJH)=dUYr_
zw=d@$^;K+O5zZ2d<Qv1IL~b^QJATD*<D*zEXj;pAvf@}WE1m=UCveL11b$Ppj-O3T
z<hEalEK`xhU&kb~^2204uwp$2D5P+9VG7rprt;7|siGIPft`+R;8w>pPHIczFB3)X
z=UO@^PtV|(hZ%fgekRwx%jDefEIu!n%|&_H>^N{ED>QFpgXuZE=SvR1%gN<$l*jFl
z^Z3$+d_G}az-K=c@RRL@+<kr#5AR>h?k|gZ@}3gbh$`g>*D|5MY~rn7H?j7ma*nB~
z;5{juIc!lSCr{qOd#$Q?cb~0nEx(P`K5yfP54W@5xoWm;*ufT6HGC^;C%=fQWrd(!
z{CG+oYyG#IorMNu+HEhpb?jyTllxeHQ$4q@*w3XC4)Cp>2if}ZL0+|`fk(}4WHbFk
zEO+S;uNSvasFBD*A8X>==|_0!%Om_X^eD@GJ<2gln%VhjvzWC$#)i9(3ATC*Uz^dw
zT9qyQ^I;1=S2@l)LyohS?QxzrO<m|I>LhnqoecYFkl%g{3RtE|dwjHLX_z);9_c9<
zFS-={rWa{m*Q1J024v*amqKnElTAf`;w@&v&o(FD8J5&beFQC(mC$Fa|AcREN4|3%
zNN%1Jt=lt^4BS0v@dO`HqnRzX)dPf%yO^>+tsvcdQKXa|N4C!s>AGJkDgVi!>is!l
z-=dIS%r7JRDV1b2emkwUucfT9d#Tg;AieT#q9dU#^eF!{Jw7Mich#?uU(ikZeB&NX
znf63z+8wkr?+p!`@`-L6{GgTJ|I*WtE*Lvs7H+e<qpP1HB79Zw+*f!Bewx_q*Ao@K
zz3|LOAJ;q$@zBi}%kBF^euNoR4bAbV+YmT)4#m>rBk&>53a;}YHjk}wtkV|XOU7W5
zlLO|z8HcCqoDiz*Dm0ae!mDz})^$^{>y;-~I!wpuY##*On2ABkvvAXD4*v6;i~a%g
z;lC(A>=*~3>*Qekun56|e~Y2fBA)dtLSdo590v6(;3hih5eHYp&o}}Hq9S3_x(2&s
zqv5X;gH2mvuy|lB>Z)Sl*lR7c#4c~wj<xt^8;1w!aj3iz2MF&cdwe{$hQ=c#Cmt_$
z#-nFbJpKxn<hG;n=({%_t4rc>GAbShUhydFA1}V1IOOe!6I+RKxTF#X^L?UoJ$Wt0
zb;KfQc`Qc$j={r-81Rc|c$|+yMCBS7Esn(D!4YVBv>N$It8ld6N^Cy3T<AYxn09O#
zq$EqxqiV6xr9<#1Fc>+_LC{bWcg(Z}P%D}T%bWfp=OZ#RHnR{NDE2$jXTW^7H{P6|
z21&ao_^0SMDNKQq_GFkCOcdR7SGXHHqs(vu5_^q9-yRNFD`Srq(UtxB*aoA{{fE-M
zfOEbCzTu<rclroSvl@mHbxU-8Js78t4}y828I~_L#Y_8sxYpAct6v#H|EK}}X6eCW
zK`*$8y#W<<ZS3mQfL(JB!5~xxFH=Ugiy{u0%A=&Koan8K=Z)~6b`*-6=7ZRw-1(jM
z27RXF0q<#O$18eJ)j>A|lPGN9Lo$1Jhr;*Ypsul3XxYRIlxoyU{eGRK{&$X1VsjHY
z?`oj9t@Y$sy_;$d?xYJ>MYmAw*o2R$px+Ut)azOyWm)HuR6{o15<J_0j#Ro-m`rlB
z*U?Riqo*b@G|4`kRQ$pyt0;t~d<&rCOZ;g;H(&a&e;RqmPN7e8-6(wa1nRj`u)M2m
zN&Y>^#A6ht+!#vF!Uj{Ur5X9j_apazhBRJXpZ1Q?rMH<{l+i_<Hsq_2`gBFoaq3Qo
zmdaAZ^)8gST8b7n|CY?x_#wI1@kO$H<R^*q`}dL#`FE0tWp5-Cd%c#7{Pa?C|HBK(
z36&R;LdQ;tLScvGxl)JZZe_bfE}&hK>eDWXTH7ueeZO6@Ah1Ky&8Sl{S?Yyk{_htO
zX_Z%!euH02uDHLEgonSCysLaCxqk1x1o$Wkdi+U}{_BgxqvpHhnEY=^oRSpv3g|-d
z%CfZby&RFOA}t)FLXG7;XnP+m`qZdHheP$K!^x2HMw`&BX#*%M)102ZvLx4lkyP9b
zRC>yW`fjtQ-aE$Axr?rJM|%qS#!jPlHDB^M?N8UL0?A|dVv6frPTsDO^kZ=>)d{Y}
z<FQFJ>_Q632Bix=X*SKfl1q1r3aN8xDczb<L4l*I=mM%~pIt3kPu)XOEBBM!jz&8B
z@d%ksK29NLPm`tJIdap!Ocy?0rz79*kh=LJT9xsfhM2vgC!$-oLHZl{kNrc>A4#L`
zKUs9I?+)DwO1Sw=75dE@IJT)L>hgO-W@{hlwFwViW&j$dnB&+%OXQ9iiNG_2hUGSJ
zw6PaH_;`_-cf}(&cPQn1VMMo?=w37j-KPa$txX70CWb*OH5~TgF*tT24r%=okz2eT
z<z{IZe=ZY0;&Sk1VgbB|l)$oYIhL7hfstf8ru*!~_zk-;{(L=d_G(1#@*{A(--6?l
zPvh^EHvF7*2`#^_VZr9x_~-Qi#kx=7*4Y8W6R)AM^8<`Ge}!$;FPQpBan6e_JSj?+
zNw+%>YE<Cwvy^#}vKqS{Rp+TIM7MmH4wrxI#lsHhb9}TR_jfktV%`2c=j8zIJUoc=
zGAy`t?ohrd8Nu>OR$O_X_-W05{AZOd8;u^r(IS^@(=v{eRy%Q`nJdq~Jdr(?xO0-M
z2XlrO?@{$;>$n*_;-xSDv!BhH8UFm^{5&4>X8{-J2XVbcFweCN;RK^4tSGyTf8Pw{
zu%hK$I&mc*{w!{vl>hNRBiVh!8rJ(6#VUhi`1_3*HWAsLH@a(i|K7FyWPBXoyB)`>
z3*!0Z`*^;$GJ!pQC-AALbzJsu9Z!r-6t`F+KaEV{qF+gTE<BlU{!C`w$o1UucRgpv
zrEri!D(}foWs&*d%-Ri{YnR5I*VDK*B%S3IGgxCs20wDk<fyNr^Io3C_LD{WNM<AZ
z91{1?iX72t&t=!IxxD>Q9(%>-v*n}$mKPpb{rf`ha<YgUi;8(`SP7?0Ead^_Wo)gm
zi4$LJ;-<6ZTu@!X=fw@AB5oi>w=EoIQ6)B`#SQd&D}O$-jZbXe&R&t#?CHLP1Nzml
z<Vy`NI<}MdCf9Pk>n`rAT*q5a*YVHD-F(?xbmVXB;a}l<`G(Fuwy)dA+N0}va9ceu
zblcBOt@~Ns?f}oIJ;3o=2gRnqLC!mWkTdlgc*@)c4%*bfUoSTB<M$2hF5SqpCn?g8
zMT)ecMv;<Ll<0)W@yU!=7F+M){cMOTl?JGh($yYRx>$oEY_({ceNU2Kqf0@rds9r7
z0exQFm%<WE=-5?LI^#2lju{N0i3Y=|e1;Vz-(yNGx1-y;9LPjT>;@j1L=G*U^u}Z+
z^=bE~bKMqFS;8_p=o?PA9AfEfYyuTLNGA47qghY0NG3Lqz78xV-3Oa!P5u`0n_f-w
zeRt93_j~E%*@L7hIKEWTLi4tqrjd>3=+NWKG*|B?eGR-vWoI9gJUS?<;Wa5aex$AU
zztNbeKlCw68s332h#oD6f?f*v@JR^|&#B@>rO1N{O~+`W4x$ZvBk!#~o*y*C<0xY&
z+V#i7Ujy)D_aNlYw!nIYp@Ks-919#rA?lU{=5zS}ndPs%?cjNK3{))~a5-rlX5OBF
zus$w$<n4yQ6v2e3bBA2pR5aZ7MDx9AcyrYoDo1CCJgqP0ESUw1QFCDM&L2Mu=OKv}
zAhk6BZLUF(Ild4-%@$$d+7M`;TZ}{9mZI0NW%%G4imu*a(C}Q2=c89(to};;=vaws
z<*TsWX*DK2S&c0L;n;9L9PjNSa5yUh9akc-NIDYk29YQ;i^P26NE}s+#Mvhi*s(1_
zY`jNczsT>p=7gh2Cmii5s}b>T6>98P;Zo8{*dJX1|BmI@D7_s2sf6K}awyWJmO;H^
z3F-yguRTNLVmub1N+nqMgF)EiABaEy7QiceK8y_Jp?Q-(GJDU#hqPI+`sIs;Su=6w
zs1Nq&c;nc@Y2XG=%#rrM^08B}bM0i<AD)Q74`PGH$OU>6o$z<%c=RlBMDmfbSaxp=
z-hQ=%j*2b18Cm1fFor-^!^PPOgQkpx?X=;j5S)%BVh`<{*I?XsAB1DhX4p5z6zx|1
zkUq#5uf?W_wQ3)n`lp98?|Na~LmgOO)Q0I%O|0Cdj_I4!@HSHgZt+S;TB(2vkrU3J
z)eU~0GU#;a0>iOVc)Id8r3ZYcKHi^cmCJidwtY=5!#nAc=`$Lt|A;ne+#^55TXb6H
z8jX><L=*qEQOv*76e)9pekwQ9alON|XXrtyb=^m`OY3N7X$|eZzKxoDZlMSB%1QoM
z2@M}uNL-OiZ_Kjk+p%<d8I(%H`X!UrhXks>u$I&!qp4}lYHBSHC6|sxWb7J1vRCF%
zcGyf>Fnk(WDo>#kQf_3WJb~T~9ZNMKwv>7t)Me->I(%R#B~BBp15Gm;_{xOd-#4Vp
z*LtMYQ<t{Q6dms79yDFt2(v}5rleYqstRPpF1a*)Q~oQ-GWa3!sQN4!y8XSx%k+&z
zMgOJb<Jt~MyxVh0_t2-3!Y7X<0qKt<UxFV>MlF3H@y@v~F}`t6vVZtJ$@{vyl4<UD
zC9n1FO8#ixl{i`7m8{IYEBR({PvUy@o+PdIzT{2y14-YOhmvJ)9!a+Ldm`EB|5S3l
z;+bSc$8*Vd`wq$F+nthgmtRW8PkJMn8~k3<bMR-0dCGUm?d5+Y-Gim6z)gm<yyYk;
zNrB#9SEi44JxJ@UCLIsap{8MabXv=hVuhaL!UM=X%AD3ewxpYLM^fMJpk-|~q_xYQ
zcJ3Wd)^}XV)MN^23+?OX0ADib@E4ioK<fKsG41TLf_f%K()5&A>d_ug&hAN+DDIpo
z)6*&VLl)f@xu`!5g|u0^l>VJ5r-tG!^mxs7`o3r<b%yLFsknOjRMSAM@0#eoTMIRb
zJh}doHrh{@$k^gKHBY)jag`70!Jy|<(fN|z-hEGI-M*2+f<JWSuQbNamBp7Q-SKp_
z5)=ohVa;m|EIri|e)YZ4wV@BfE*nGdpXgMKF-LTXC2G}1;$;cpqoWOqmF;ot(>M(I
z<$~gV?kEWJLduI7s7{!J9}@!LXcYo8w=n!m3m3Lb3}&8=!@B;7P%d1LpuTDNa3~Yk
z0&+0Sv;cj-7Ng<XCYZKVik|5<>^)V3gZJt%``<n+wP?Vxz$RGlK89y9r*LR)D=uHU
zfZ-Fa;?wz?=<Rn8`+h&d(W>VdKJz8I7`}tryH7ytcew5T19^98Hn`lC8$G)5&WGK(
zDMXRCbx~o{A~o)?(%_6+T6`o@hviIq^XK#WEFW&jO{T^ieWM@mPZ+>QM-1Y3PX_bG
zbW8q9!`b7p$m_&Q_=hp`&L(RfJ=u;s?~md8a~#<F;W)vIaN@6rT)3v!MD86jnXBrj
z@R9c({KQ}ym)Lr9iu(-q@$%&}uCsXa$T=dnJD2J9Jbt%v0n3dE<myL3ti3dt^FA%&
z)PTi2=h70+8MKUdhljGukuaA3ww&|(t>jv#RjjyRH7}0{=L1O*T%I1u-7?p3<Ax~C
zSQpI}HsT&S5W`ACWBJndSgtf$%N<2)`HNN@n`FlEN#%GRkQ&cfN(nqJJ%JN6*Kus|
zIu0>S<dyY_-0Yadfp?R5eP}XUsjlZWwd>h`atgotmLfI=Q~AQg4ctR?;9Kj{_|c+t
zwld4$x{eGs-JU5ln=D>9BAf63$>zMX8+m?Sj<~^cdAjh&WEJ!I%7c84+FQWKlM30#
zS9oH>i+P5837bDCVZ%M8+!0^KLp?WfO}}!M`&7<vn<`j!-DcKtuH^knTez@Q+(Pl<
z79zpEc)69^a)l>my`5h_*v_Tl)f}d_gU4*z!B5O;n0D6i$bmcg+@_uUN3E7y!)w{%
zK`m?YE=J}q?(t|B7xb@V8~-}Kk}ZDK*YTTEb?iGvhK9|Qp=r4?RQ+CtrZ1MIX<FS#
z_jxzk_(+a?<>l$@ECu@Bp-8XGlqoe?m9}l^LBBt1kku+}>T0P=YWiY(!mbb56dF;w
zd4FpBZbmnx1aD^CFfzC_il&w_&Dw7#c8MKH?Yc8rKb=gz6Q)s*?tb)Ee?EC9E}~?c
z<<!b+sLP?Xbl}4}GO|vgbNPb%C$*9GEz2jT*TvKxEIjScTPQxhnmY7%QAx{QN(ehh
z(?&K?M7I{Y*LjNW-fJU=Czoi>_v_Tv;4W>M{D_i;pMCxAOWJ1fo`xrVp~LMz$-_hn
zV{UZ8iX>S~M0Z$;&Q(dFGQN#dL-R*<gcWLGqm2&AAM}DjxIVJg43SiBg!ShAV7A*7
zKP?6#e~URbX%4}Gh@nWlGaN@PM?oV>0-F}#&+q@BZea@tkJ0d5W{-gh4p_NC=)x%z
zaAS=#lIOXi%b1B!6TV9PEq4sgnu<HNp74C^h2Ry_(ec+C8^ULx-;0@8?&Jr%(phkN
zHX9y#{^;j87i(tEgJH;gTwk~VbG!qPL4g=39|WnhL3p-mA!>AjF??GvrkO5+RIb=z
z{=5juB9l8QDg*)5A-H-n1VxuZaJf~yr|lD6>J1@K^a_EYS_q~Kjrz6wBJ{i;4E<@r
z7<XbJp7dP^+oeIM-xG+E_5gS)2f*5N0rroW50g>z@OIE#tk(92`B%YeYMX_sTtCn>
zU!<tbgmkkHe$4mAHQDKSTkM5R=AIZ`KNbC~rl5GwWPCE6gs?m}%$0V9+<a#go}7T`
zhT{>j!Vx3SjKzK}dzkx;#?eY!<h5H1e$;;`@&jb^B*I@D1<T(fU^s9%v^|E3J;NcG
zS1}l`P7K2NXJ+WvWq|0B^+(M}6BxJ{A$67^OhOHy9j}MH>|S_WE<DOz+GskYDHxvW
z$hoS9sSj0P{YnWj-xOdkBadx8<Zz<T|8aEQ;aIreA2+izvSsg4HupKl%t*9}(xg76
zrD$qc6{1vjh{$N5qD6(0($J)Vv^11TMT5ro{{8-VuDZHPdg6KR`<(at^%BMMvBEgR
zMG*53`~|tjA?SPc9pogwL6GHVC<y!j&CxtlQ~nAT+<OMi!=2D?^qBXZ+dx0|F636-
zg0tN%@LaALc6(id@mnwQ`AjV=;n`D1=W2MHSOwFc9EI67<<Osg2!#I?!p^n%uy!y9
zs^T+2RWl9Znp64Siua*Bk|0TK2UrWl!_`k)Vg26?u-GyZ#>K6Gh;NHQcSi_JwVwlR
z+$|T`Hw`}YPXq;7AL#P(fT#=?m^}nAVS_ycN?F76GZwIElL_1jG=Rj3I$%0yG{`5Y
z!nzxZu+3B+G7rf>BuGMf-w2r6APjw{1mJV=FdMe{&b0KuuyvdJn9jI&EMij++i38D
zbvSi1>5HB0X5JI_y!8=N9RHBH54W*}&mOSk_xD+n@_mL2?y=BocbUfIyX?=8J52xj
z9fr+!*riW*SgH42W`64~GuwKPZ3(>3W(7Q8D<a$2#AEF&{KrF<Jn=DmQ}~2k{Mf->
zjq75Y8=kUpsn3}3!xt?6cn?$1f5)bq_p#pEezxqwAX}+3%nX+cfXz~2?m!;_VV5M}
zj;aiF?ULh}B1NdWr2>)rH6VSvHaMi}!`w5*F#OvbBtv=j=(QcFCjm(Mx<IIf2h6qd
zfmK11K>yHm=u?~teU)>eIea1iwWVOVWEF(xuLCx?84?}ipnThQIR7*e$~{xKS0)t(
z?bD#7Arqbi><8=b`Ea7-Ak>GHfRW{4ke96hL6H-j6?h7w**RDqeF46|ZiIx$W(YIB
z0msDdfQWJ%R0TYNfaY!xTG|6TZtr1(Z$FGn{0<(2zhV7#L3H6ccGY9zXzL`6uU^Pv
zb>1jkzElN`{4~(lT^n~#)W=UNjrpwC9KC*8;`l|}UDL;KVwy7^4H}1XF5~fwH_zX%
znTBJU12NZW4i<NW;N8L{7`1C9o;|n*7jouTvUV&=F5-Ku^9i`hAQ^ub>_%C$efaQ7
z2F8Tv-~^+5O!!cUWfu>jO@28(NH~VRTaV+am>L|j`z&rPJCC6^8?a}%31wLeeu}w;
zyB^%dKAU!wEqH<-6uU7Z`z0nAzD2Qy4|sK9KPHVDMDg3da3Jk3em@~d?wTT;^*4fi
zge7S87Af-hB}1W6auoJkfnLWcQK*C}%}-M&|IwOMQKn7ztn^5)&VUBpjp@KOQ(Ek6
zLHn<dr6^Bp@~yL_NoEdoC7sd0Ajhdqb*6(wu5`Ya_s)$y$!)ebZQ|YYUxmKpc48v+
z)J&$Nqf_Z=`gAH;6+kDQ1L@75Ac7OKs48GK&HKQ!%WLP6djEVnJTHVI8bhf|X%Y8-
z|3^^=7Sn~hOQ_*j7#XQ8qXU-9>9O+)+UL2F5<SBy#$^>b+peaa`VnL<7fBy}MN-O(
zHFW3NT3TDVj%?D`)A^_kR2Q<59{5F35!*zkO*T{Z=q=Q&xRq+;qp4+744ssTrKK6M
zWT_QLy2s)ukmBjZ?RaYaZyV_g@T|~*?Ih#6gR)-jpoo+NN_9x2_MSxUv`V5)zB?&P
zFq!6^Po^UqQpnd~7s>zFMYn2q)A;CA^6}h5S$rlm@PNBY5A37oi_@srDxHRg(&_q@
z43baHq#c1-)TNot-RIf#x17&|mhC55vs{|;HkUuQd31JWKKV!=AaR}#s#;P&-$xcw
z-pN8*Fy$bLzdJ}Fn~R8ehko_GVtP615YO)%A|_TsN*hZ^?^Ov!`jnE&;Zjl*DkJy6
zGBPSD<GuAVnr~80QNiUT5M9oDY~}Q1?hu={dx({P8)6r-huJH>H(z!75A(!-Y}-x&
za48am@Yg~h5hMcdzKcRzvlt{?k${SCQt;7#Bz$@;2Mdpm0_9?55WKAhb1XIC!ZjVx
z+iw68WhSudiv?VYwuW=l96&dmAmNc4T;J;r=kq4QnV;NsRW^$=I2M3-&k~RdTLl`1
z8(@b*4BX{AD^`^VD(xxoNOUjnPi4TUm>g)inh!b3MR0aiDU|Zr{Se<<SzWG%s3~XR
zP~UkF%x#3lLCs)oa06!iyA7M(K7f&(kKw<^PvOSXm$0w*E$9vP!5^8gP_O?3G>~((
z$^~%HS{O6yMA6S(9QQX%qR@C5lx~(qH&+ETJ*|i{j8t%X8uvW^)Ig<KS}0SagH9v#
zar|^ce3WI3c|4yyUfcrH?8YL^v_hdZHdwLU4n=l5V0#Kfj?+c06;8+|xuBeZ8%FhW
zMn$y;`Y!dtnM&i)<BSh>`1xXAyC3=lOv1XW+>gaG%u<QdxEE?V8Y=~0yUz@qxGE6+
zlY;Pg`b;$6Hw$;ia_3{nY_zkRgNi@-qms{<{paD0-g)R4F(0ja=kv_Z0$g@<0VaN1
zfbT6rFg-8?<5zPR_NEZt6%WCv1tF;86oOmDdDgIb0sdOE0Jn@<fE!EZqlNK&?8=*m
zcZKF*;2fTPIXnmFzMGB78ndx*TrmEcI|~O^&g5AhzQ<n~h>>$=;8~9VJ_GYd^MPro
z%6r%)aZ}LDX)=m@o`^}gekkeci%+|JaLU^8SSjj_F8e&uN5unWa@_IOC^u|Kc0rv#
zPN=t>@JAcqO`dAJk?(*S!*=*`hAp~PT4R8q6~gqfD0R>RpS?H3+cu{7VS_PFtTV(B
zL;5(zk@vPEbWr!G7V148jXy_e@UE8{)~r*(H;0rk^)AmW49jz7v>bX*8Hsl`NMm=Q
zBp$gYj{kigfu1U&7~v|6@|+<Vr}q!WtRDvVn?ImMcK|L$^@H|<KDaf8cY1fehJrUQ
zV25Wn_#EnhdqR&uW#I!*zj6oO@~m({QVT5Yy8?;RE`j&Ci!egB4%ja4Q5dQQsTIfJ
z`HN$)XU<`G)Kvo878b#>F9%?8VlD(5WrOLRbf}Ks3-{c2gTG8N{Cb@L5w|#3_fjlq
zByR?l6YD@~a1~h2S_YfC7Qw;j`Jm<+3}%`$I3s*2_>J`g$NAondc+;Pq@5u=fx%=I
zJ6KY032~dv;mBlTuweSIooR!|R1Fx4Q32zYQDAB+$Nw$TaK}LcutyY@R0zR--G9v4
z_!lc(^^J)LePWZ|ykm<Edf4QqXUw^!lL<IHX35g+ti$I%(|&!2Sv_fGV>E8DSqE=0
z!Fku&O5YZCc<wc(o_3Wr^fWVt8O<!a=L$2;y2310UST&^Twx2cuP~iYSC~>nGYi(f
z%GUk3%8GwnV>z1F+1meZu-I!i*+=_UMg_Oov#)m;Kb^D4%=;|PwvF9zeaN)hAG3qM
zJK4jN&)A!{uNbAiW7-)Xnd*nHOgiZ&i(L4Zg*yrH{~u9UQ7R7U!o2IXRTlCTM}bs}
zGWg}F!=fFWhrCA*49*zAk6&h>Kid*qJ8VI3BZD{g&Tw0C99Suj2M?MEL))i8+qXct
zlsE@u+(IEtgU`%V!{NN`S{S!t6I9QSg*=`^)_RlxE&9o@Ve@WS+`SjfCuP8emTXu$
zJrACLC;<KJVn_@ugTod_VZrDsIAu}|O#x@%$e}t|rrH2^PdCA|=oa|7rj`5K@AJFf
zBbc%JDfk+^f;h2W_^A3BY=Q>icH=N;m<gglk_g8C5<{nGDHK(f#UG9Qz80g5dehbM
zBWU3-{=JbqiMt-Qm|<-5SnT9?gU(7v{5Hi2Hz~N|q3>R}`HwHY<oAx`^)v8WZ!n)7
zbIyX>Vl32Kf%i-!aWG&b8Xt<r;+^sM_udZdH{FS6_wK?h(Y@G}n2vqQ+2~ZAiy^ZL
zP*l1Y-(TlEt+XTfVR0qC_Bx5}mZ$NOK`mArT*MQWm++ETGxjdN&OI!xxbnt5JTCnZ
z@6YPsJ*#dsR(yqnv2U^K#|PZCu^*2K4dT7bUzlY57t5Ck(u77~3Q`+Eo7aic^fpOa
zZ!be@b7ZOgw>&9@DpJ)oWxf|uqr!a}bZJ17?$6L6(^Gn+CTmFF{xkl6zh|C^1&R8N
zrSxPg%D!wv%fH&uGBrmUb^zMR73Qh_&OCqON)n##bi~Yq<V3yb`+aZH&G4cBe0=Hf
zXFqzja}vp^Ora^+Q>jUEIvtGgC-YkYbXq--3g-q<VCGB;ygZBaUI&xk-`TWm<XqBG
zo=3{c^J!@00*Vm~p?RM|Xv)J-8d<lH0&^Gfe*1s4!hbQ9k6A+F#lq<4^DqiMzm$x!
zm(kO;%Smm<3R=ik5^0B%isUL<^lKGu`?#7~-$YRA>qwG#zlPQitR+Fw_0+1dfxfbh
zv~^AtMJ8;bTW2=Y=8sz_%q*HBR>qLy>{!YYilg%takOJrJlz%FMjPt3(UkD*G=_7Y
zpLOma`^*Fy;`yQPQc2X`l0;PrJE`9@nKlZikmzNep^n`}+Ah0EZ+JJUoK2<E5qs#g
z*<QN!VlUNX?IZ08Y1HyFjTV=tQ}tv%6B@`Mm#j=0vdyB|tyyFon$1~2*<`adhg604
zQ{A@xB=~PXy;#d<M6Yvc%hWsyIg>|LD*2?eCZDWs<kObX2k5}k15|Y40QZ_7pm+KO
zlr*7$I_4D60>0ONG^39_uj*sNZXekx#ZT;o)n_&<x1WUud}V$iJP&yJJImbki*@h(
z&BS^C!MsBdM0$n6)Mx~_risJA7%5O1l7XBba&XdE5#FSzfQFF<6#di!`w{x^VVW^0
zJu?T9lU8u$hCN&!OR%Qf4SqlIhA-NafGYh#b89e!bEnmx8(|=itKrA(4X`IZ213iX
z!K&;;=-QLQAA4YKb~?{BXG2s&9!!5(2)E=)z;xnah{&vffR86YJ>WFtUatkON%dga
zeVKDfuR*QeEm-~FE}Tng=XbFV_#piZe%yZr-MQ~z*}{*o!Tc+H`|=$gpB@Gk5<tQI
zLg+9cg6<Q=IR97zwZ2H9lJiKk+AfFUR~67nNC{Uk&JGMy!@2u3aD9y?I^WjDWiNEG
zwoe~Fd^W`9_r@sql=n5Snxjei7|e>c<SrR&-gmG?-&=OLdaDEG8ZhLjW}Fg0xV_5>
zM+Up_Ub!o-uye;}d&gn#3l9u3_QG!qyfGzpJoX;<;di77sM+d^$t`~P_RK^S%9(`a
z%O~Su>nZrCmrtTHrlKP6M?AVX4XvD~qi_}X(a8B@{X&0CIP8y!-Tv4h8i3YX0r=K9
z06*#lV6bcedVloio??GAj`qhD*8bS`a5`p&PsiS$(@<gkG@SNkD)vv9ic~ZOWBMke
zoW*2xUoZ*7c22}+<$n0*yf129oq+jQeemAJ@u*Va%`?hg=(fre?|XV+o5DEEf5Ojo
z`(5$kbQi1^b;gov!j4&pIm6t4n&XI-mJYa;=ee&<<+HF3Ys_9?h0!m^;(-Oc^V(^S
zhSSZkz1ajOIT+)SL;M^w(f~KF;~C%%9nO%~MxQ)Q{Qg4&jri;Mha<eR{ZAS9Ojp9f
zL!<EMS9!ecDu<T5cQV)}jlR6+G-r_n&O0iGai2wT$X0~!zJ##xya2xZ`v*2B3_<?L
zACQ#&4L0cZ!@i0>=x4o9eDyVy&3Fl4Uw6aW4V|DU`54X}XajGXdvN$#E40qN0hI$+
z!EaX+oHl3#iTVo=F|!to2TsHKebsQz>No`5KL)qfABG~WQV73M1dlfrz&@)y2<pp*
z{;CWxiP{JE{ZgU9I0foul3@Se?I1fG2ZzL>;jhg`@QR9r;La8Bmb2(W218)Dat>5&
z3j$a_ozHG2LH%(ba2@o3GXbt()QYfot^<gO+Q5m+W8g!kDR@S6hb4DAUD&M&I;Yh@
z{i_lfdnv%xs*$k5QVKR*69caeB4Ft%2x+dr*&o~QY|FfUX4>?g&0q4G#ZP<A;*&ZV
zje5jR{<zO(;~lo|=}l($qJ`}qca^RG-NY`6U1ld|H?rrS>e=(N7nw%Y1-A9Zc{U(+
zo|P`FV~@LPS>D=OW;dpmZIGyC7SgpW!?BiG?yhBxGIcDiwvM&#J<m?3UtnIBFS0%A
z4a_97k)`NdX8R5{u}4E!*v7F}S=PL3Y)?}QbKP@;)pp-vs^{;pX6*+oUf>aXIjNJa
z{P~O(eC=VK*6&$a)o13rXpq6wVODca03N&&h9Xljn360BVk1UEPrf{So~#7cqt&2R
zL=#Sn>%tZjL)g!~Ru3+Y0bP3=D7@(a5%B~EgWX^tzzf<Ue8K1>X9g+E0NIRScx$%+
zK7aiWjBhQ6lKKcZ*Rlad_ilylvhiSI&v%LcB|&Xb3TV7f1!cE10Pd?<sIec;m*m4S
z5AG-*Is^l^%Hi{cW02E&9F7Z~f`-}WU_|!?n4f$JY(uZY^2Ik{dG=k<9Bha0(VZ~T
z?geluB3NjAgk8Shz~JC7h>{UNlexn9yM6?YvX#VZwK90hUmi~iD&e&&s%TL#8g2LJ
zpvQg#yn4|D_5NDm(b?9R|HvK{*6`V=r7L<$d*WKD2`DjUGWtdMqu=A1*b_1j&&n)9
z^UkHX^7<-V|7aa5i)_JvMsfIl7SHnKCZf>WWc=frikT(*FjO%U75C)eYQ=n<TvCXc
zE{D+ZRT<iF#(JV(6*|dQW6;yn*mb&=yL~UBLBb{M*xHOwW3Hp(u2!61d=Jwuw_{Y_
z6Yd4>#^5<G@l??pjQ#u`r+9tFlHzapeZ)_!jQEZ65dw7Qm=IOH6Cp2SG16TkLD$Qr
z=;AvWT4yRpi<T<TK)E9A=~1Q(Ej3y=LxX~nH0kCUZBluzOIgAObV$dD4mg-__mmmE
z<1-$4pRx3qtmvYl4NVlaBh?Ojo+oi6UaO;6MIxp1PLwsng?eAPlJH`8`p`L!PI-A!
z+#xU8Ht0>i$cJn;OrV5vUz*eEN5;JqDRg)eWr$4SA8#t@{GCe2KTo6lj_H)z;7?YC
z0c5>t2DwfPBniVHa{C@csm(K~Fnt!Og#=T<*x7XR_iS>wHHTIom`ef?^C)cmeA3cg
zK<R%Mkj3*5^12#IRwoydW6>hY%KDGC?ORMHX-lXqCyYjwE~Tw!m(iO$%gOK43Ob+=
zPFI~)(cu-VNuxM|zI8>?(CD=^7_yFvj;*J&zc$c2J|}unw26F%H&a8<R(f?NnvA76
z3-Mws>8_5W0lqU6dl*leJGapf+D<b*ZKuIPJ{y{vK;L+N=td=<4b4iTyTUtZ!?B%|
zFg2NKza~@bz7&cdvx{<?cTwA{-SoS6H+8N{CD%Wx^e<))egCtEBsc6O(+_)TO7K3~
zab+L9Hcg}4<TQHzCXN2Mq?2PhKPSCPCs(Ts65g0WO0^lJ+n+({s+qhyo=KMno-z}q
zZss(<o1J^w%}(b&WBU(0XNn(Qu<tuwvA?Tcv&f@wS-8V{CaCn0xs2~;bo(1qI`)Ho
z>K<k>_5v{Fun<)Fh{7HfaTu*E1ub49VeVmh_`~x$#<FT~P*oFV&DRC?#t^!Wo52)*
ze>|&T4+@nCe|TWxKm})RNKS&PgZ{w6f?-u~2y9s$1|v#VLH6hM@NRN6Bs_};EKGov
zi<2SVJ{5dq(%{CYOy29*4_hA`fa1qRVEv&K7E2ukO{jv0YpS7<_c5<YpNFGi4Itgx
z1o5l|%8%RvP3?QIKf4`X%6CBfu5Oqw_6oMdz6Hl0AE0DeKRkOf0Fs`+VDjNV&?h8_
zAIAvel>kwc+aQLkGbAv%LJCD1WH9-rEZ)2)k6Z4ILjM~|IP!uDt}Nqx(VZG-u}Bkj
z9JTSepe{OH(L=Fl15~v%La#^0_#fv~OqDjrXE_%5USlkd&$PrlBGwqV%m%BQZSl0S
zJ$D4~JS>0De|??dxzB*!l7w^BobcvoXPl?#f<6CSaLhATl&E&Y2^-z<lJz(o-#rdb
zZuP(}-hnR5@Z>pOFDzQ;g@^8WVUM;q&gAE$A8Fp$aMl~EZ+m0MBX9J0;Ee*8ywRe_
z8(*#QMj!CTgfCurJ>LsG$O|o6J+WcBCpzEqz&GH5BKhO+*;jXLa&gChF>cs%#ucx;
za=~vST=0sHGw!x_!Xh9}_vbr0E5Hvr3?n6Y|LnazCN|sQ@%^?~y2J*LTUw*yS4&hn
zK9+O*#^8H33v_5PLzzfZJfvcRrnN?RC(sZZ`t?ySk)L7Ib+P@3HZF0{!fy?uF~eH}
zLvN_zdOubE94TXyml7H@jKa|l3V89T9FEqM<=!Y6Ts$m=@k=DJ?w&YWpcwu>D2fwD
zh@jUpAyjA;z=pYh;ez%sY<m6!f)5Tr-TW``P3sfP?B>4bg12yKUJp3vynx!5Pa&kN
z1Duz0zom5>?D}~ZdK+8$d*cl_62Nor+E?K9hejypGaB85ItU6l3$yf3!MJZHfUZ=+
zmaL=TyR003xt733l_JRg$~%I0b3wB@8+39qz-Ze(c)lhTM3$$(H0}dPTD61kDdJ(x
zm>39-jDq*K*1%GiaQJ&IjDL+Tgt5l+fc>2bWBNIV;m>4PV(1Gi!@QuU*$wQSouI7M
z5klA7LV|-O&)%3r@-HKJJfsIZ<g~#QH6UYy3bgQ9&CnP*?njV@Mtcdcej*AkX+peX
z{+G?%{ga*F@s&pb``8<iw`}U(7fjH<i?vRB#O9{mXU*!j+4uhIOk4dblg+)%8X_Cm
z-$NJJ9P>KnCvuKOn4Do2drvV7+ZtvnQO$Nso@67fPcY?;$Jszv6}vmHiiP~GWa}<g
zvWS98)_SOtHQcFWvIbSms<?_J`5kBNx+mBg<CE;mylN(Ny@qX>f11sbIm=cYJ;#b9
z>)0Co^DIO20-xPnWOJMA*^JgkwrzeBlU{$7troe#?hm)Jyh-<&)W?VH_nl5A_xU*s
z^LoRK?tWk$yT34j=pX#7^oKni5`+#HQSL$)he@OOOy`g+C{G>*778j5(XS56d$r)^
zA3az())*#iF^5_2EMdzUJD9Bn;MMEQJz?X(w|hL#6HWr>iPPa{MG&|d&xK{Rq3|yx
z3|?!8!@93)VBeD{sBDjcPv7F<wRi%&G}{S5vv<MYygktVDGf#iW`R!Ien?z?0A6Yp
z!S((UIREM}hz(W1S^JX^opTyW^*P)9aXtLwGo76mTfp*DD~uWc0OGDa2FdVlI0*df
z)$~2L81{pE#CQ1K@*7Hv1aWP@2yU1rj**-R7Boo~UEeF<`~%APmuKkz^V7mkA3fY3
zVuTG@X6WBL7BScsZQeWL{ryguxYQl@%<|?uK0hqWor>(k4Ah@J8xQyLu0-ZyJTPws
zwvLPBKHH6$G&35Xdc<Sr;vLwUk%ZG)Q&3ZG5B{8!hP5@BI9_o-W<=-X{;!1?xcU&@
z{!xaaJCEWY?$JBbauUrqoJN1kTAtItfZ-P!(QMxpG+xz$wt=@$!TT<5b!kIG*T;Bl
zd>0-GdX9&~d(e2-JIp@bhjSkH;|<Y4G<E!iCzt<0pL+td`lAp%kq{*jBQaX(E<rnI
zOVQ0WGIU_4EG^HMr`%&ayK_c~sxPR}qKj%IdrpH!S7_3&9BopK)}?R3`Xps(Nbi0b
zk=A(=@`*GfS#1k4xip5R&9tQ89xEFCpAE%!*wR){ds02*K#RURQmhM59`}NNt|V%1
zccRtboN0=fE49kI(cDq)v{iZ>o%}P765e{yo*SODsN9Q+qrK_v)bZ4=<wNqlKID92
z0{vO*OZ|?1{G8!OeRUJbc*`Uja-K|I1g21W^AwuDYbv$Qm`0QJrjyUt=`{MPKWP^P
z@chaQdOAChRK^9-mN7HQL2DL;C<c?F9Ph%*&mmNuOBDw5sEj)-<R>qn`^!Q&M<$dS
zY8KL-XN%~s>|(m@xr8ok4Wq4fOR0Hi8M(Twpp5M+DXcY|W~r_w;r}A&Pjw`939cop
zY3nHA;Cfp6c>{$wN72B}O;p~tnV*%m((2G?N-Bw=(1=)?HZqRx9gd??w|GizkEf|i
zw$aJKZL~3VJ4J}?pykOs2qhBe+_nU|`YVAZFHa=z$BA^pEr|}4B$2JiPO6;0ljfY=
zNu^TB^nDKBp&d)6mLL2_w-lPUJ%#ieQfTIf6lxUOMdhdOu<ToRSib08mKc7Q%^!7-
zsr<gj<i|W<8xFUz1*;yi9r2G@>dOx1u=gpOzUw(_dGv~<uX)QF7kprA4}NC9Z3kG_
z=wIyJoWE@Bups;%6oEt?ak#cc3jT5Kdc^|;So}a4PLI@p$BEjI;AQ}2U<%4>#zNc&
zTkt&ykiFX#Vye8r|BoO1Su-7$NzZ~Wjq{;u*J8M`d?m!pUJLVrH^V<$?u5C#9b9K6
zf%UHx@I15!LT09ew^}x=dy@+~jRjy-Rt(BHWnh<i6n5oT!M}=XnAUs-sHYCri8nyK
zRTI3KeGP6T-GrHE?r^uw1BjM=3<O>9e*SYf9p3}32j9UGo=vXjyb^^+15noS6Pnt8
zL)St9><bdYOWq<VX*GiHam4X{zXV>pCWWJOWH2j4mVd9x<LGA!cqL5{n_ZOg+e;NR
zj8Q`yz9ad@?@b<hTKGF(8-I+{Mfr_-sP|YOMJx?*?-nCG&}fW8Jj<M6Zic2j51Srt
zfu*ry&^Bo-O7E~l(G6DEH`f}oxFM)b!WK8(w#9jgb~uA~BUrmVzFO*lwqG4kGTagU
zo;qT;E5qZN46C{s-c<*b@daG80`O5BV0bd1Ljs`NI>3@3!00i6pT9F4eTt#we+*|y
zGQ4@n5$mlTarr5pEwXUH!5n)G;(MZ4e>*%^WQ%@pY%oFBhTjeUFZ*JJrUjNLeReEr
zws6<iJqr|WGsjhT%=j746i-&0pmHAH|8M2}>tI9NWn+MoMD@}2fi51%*TK9HZ4Bqn
z&G=`d(LYB6>nEw>iGQj%rAh@GW+`*Fm=f+O=L|hxo&|r;_ezPfIH)@k&FZ9ac#0H8
zy^_E)YsImdcbaPViej0D2o{wH;bn6{^a%Y2bxVJ9U-U1ywTi#)bB4gz<=pkO<RhG2
z@E*p_dJAn+d*JW*7x2%e8}uAH;mDZBaM!RMnzimjh3Xy18+8kcWUoV`^i}vK*#u)G
z8)2U0MOY(M2mE~(1bKcUWYkGmt5OASqmMzG?qTpXDTN!B#gN1b;mWvtXq~(t1m|W!
z%gS`fjoS-DIlE!~sbo0ZmIyXO+u@>NJPaI<23!3o$S>pXiNu|kufia>a3SQco(C{<
z7N`aV@NDW7Sa!e{?!NYd;PLJteAx;1&tPyt&<@@-Tf(X$bI48S%+5T0s6M9+%I`Hm
z-%=I)6BS|NS2=jPN(LkZB;iN#2<Yc5nQ3|gAo%YWyTtoQ<sUz?Ej-6!v;8Ggk$=j>
zZard$&faIsAGNY`me*NI^A+Y$(8ylYUtoe3we0b`(@gS5HJcN7f)$EXu^WGmF>j}%
zY<}%wc6DVrvz=bXzRxdZ>-Uwgy)O^3D^m|Kqt0Sx9$(Bn#}~7$7R5}-u9)X?irH{=
zF$)`eh<#~0#GW20VaHFFvbSH#SX=mEre%1P`SFiExv`QRH#yFZX`Nu#+fJ~HhflJ(
zFRR)5qEoEw{TUW@x|S^%xWLkK8kuCq6;{^K!m2}C*|5QVmNxnkTj<@z&K-QgE}Ok&
zu04HhZ0lE6{`x1oGUgw<eP0O5`CNKungo<9%Yc849EdGdghfmh%=Jctov{uC`WV2J
z1QR&&!UA}H9n`+q!_E=}iM6f}I?oe~LMOnzt&`!?aets6GvUhgdGP1bLU`!86q-82
zp*e9ag!^uSeuY>F<%}XjwFDSxzY`1>?Scb`_rP_LbQrNY3r@)7f>gr+u-aDyKX;VE
zrvlE>dCHlA?$vPg))~-^J`d+-G{W8K&9Hs#4R~662fB3IA@^7Z%n5uBM|I!80+~Lj
zQ~L@t$NvP^?7y&VhY-HICyHt|5~x@%jY2kZDBCs)|4UWjoCpn^FjX7xaKF`#7-Lks
zY>rBLR(x-3heru;=b$qNwvNNgm&c=T`$TjVnU2>d2jS-$p3|`j#g^+!&}SoOUpTKp
zUzI2vF(QWN^5fBU4=1A^O2P@3Qc(9@Dyph-U-`65oS4p8quqJ<#HJ9v_7-Ey_fnj@
z_z3>$uHZSj6Zr3a4cbMYMG^J$=vH5kCzf4CdF`vH+Ibzr3tI6+*gcGdc0Qkeg0X#F
zc)ImDLQN0O%YTQ<llm}seLt!$9zc=6pSZ^BH=b1!pv|&E6eBD`PJ^Ok*ds<2cO+=s
zSt$z2m7&=iWjXU#o__0%qQp;%^t4);wyjbnAp>>lyQ4u}k(#tlTAP;S>(DYi&KWGw
zr>lI&b$h)Lc|R~FLqk(qy~2#x5p&Axw4f&8vGi8glIK^f2;Hq|q=yZeI@!_>3p@It
zWKXd_><RBXP*Is9xvyi~6$!Lf7|G@e(v~=)_vA#115T8B)R{icccDv)u9S7%l?I~u
z8Oh$ArhIfKxAJjxaG?i1((@#**Pb-)m>21+^ClzCa86h9p@KdiTGujxoR0XC{2o6N
z+cc5PmrbI$`IAXFcnX!zno7a*rjcOSblS4spNe+|(D<Sm<Zv;N9G(V|xZo^WY7$Ix
z{<Ennb`E_zF_)5F&7)DO3&>$|2<_VuN*6CKq=4^>=)>5>q_|`WEhq@1d2LH+zxZ-G
zfGcQV=}I!r3a9$|Rpj|<HH{OFBu}k1yidQDl0DYZz{K@b$@^yhej90zYZS>_Y@$rL
z&GfEsGreuxLW#S!Qs<OtQkIILocb8rY#&RpNwM_&aV$9)$C2jRIO=QQ>_zU@m^m$;
zKJ1Mr^~>>eqKnU!A}_J*t(RCw@g?^1+a)$C<}$lIv5CD{aD`=GXlCvUuCYPC>+D_3
zO&0yTmHA!1%cSl<U>O#VSl9gy_Px2AX^OpK&Dn3+@u)ubv+4`8!yjzD)*n`RTo5)1
zh(hN!aZq=XhGph*Q0K1*>(8jd^Z-pz8^iCDuEt=q(*i0+*g$%XBXlM^gKnY+N1acA
z^-p-O+$a#X?3e>8gP{<;W+^xetcJ6B>*0&{R;Uz=2k)gjK&&$fGJ<wN@UuM-7nu%y
za@m~6l?&%q7C??!F+BQQ3PBf+z_-*&h*^9R9=V@}m)f<E@$UkBde#U+jaT4fK?^+E
zd<(wJxC?77+MwX?BRF-llk@VQL5u$@SS0fnOdH-qWY{M-DE<{R3I}0|`7fxc_zfZQ
z0vK~u5I3z8Ml~%_G;17zn`evTv(FM}x>X8ogk|tv+(>-+ljnKD<nh`a1%5|S#HIu#
z{MN3F4Wm_2WVRaS?^VapiyCPCVl+M*)<ka^Z6qZf%o?SO7UFt%@2eipeV~uUM+|Vo
zIzy~+FhYxuMp&0)j0{XL_O=N|&NRhOoJ;p`f*C5DG{Y1=CmK7)96uGBW5XkJ%oDc2
zSbYnObhN<rt`@kTEU?eq0=LOo;HF;w`_7o7P?R}#Tbg5Krx_OT9et028J;aL#VI<b
z*qd#F&7vk~xXc*+nvHO)iV+HjaCU030UCDdV}P_i{;=1>pqaXOVT}%kCuw7KmKJBy
zXyS;X(U^Nc18-%h<C6q6{If<CBj)g#mzy#g=_ujt;Zd0JKmp6k<<V-R9M1BRMQ_=W
zsQyqIf9{vU-@%gTsV0HFZDJ^#Isy-Sa^JyEVa}Nr!m?mNbQTxDDBdYqQ8xq|C4R!C
zfI*1J{|YtFKf^iQk5DuJJv`6low^4-a983bXuCXvy479qsptuey#ElE3bw&yt9u|m
z|2A~)z6qlIete;?8B8^sVAbSCI23yk9@Nx<;JdS6uE{&x{?*`_d>m9RS3uR!5%^<U
z4#O);p!{$Vi1zSvwoX3i&)p9?d08-tb833k_W{jKg<QU$Si(J*@-_(|7qboC-HC-s
zzFXkZ-}Uex-)Tf8uK>?QOJL-rP*C%p3;9!LLj6jA*jq3eBzh*mu!|QcpKyb2M<=L!
z=m^Jm*}}_NmZ0up4#M0}Yv8NT8B^L&QK126f2%;y0!4WJP!4X)mw|ml67Z&M1Qae3
zhFV(zIBGJ)7Rn4TSLILai^n@=m;I7mlY7cGo_xeSHr;1YD_U9a&K4%!-o%urHLy>D
z=h?pxXV{lfH4LMVvy)>hn5M}Q_H=m}tNe9{<=!b`quv)XOV0u(`8c1s6y~u->A5Vc
zYCn7SA%`W7&tXmt*=*$WZ00PG&3@j<VgjeK*piD`Y))?$GsA4Qsydsc%*bKPWIwa8
z&Se;t$F@iuU~gX)Fu5-Wnbf;twq<S!%Qq@zLJ6hp!JkrA@VboY&pyl~=N@G(-4#so
z-f>oEP{XWmonh_g>)7n~^{i!T6SH}LjV-OY#kL>4$6{_jWN&#_D>Lr}OZI%r9;kg}
z+cmzi#tFaJ<&uAFqp~opi5LM*uedW}r3}=`$-}i<itxBt6;jeiLqMSpgkCj(ba7Ky
z{ofcc@3MxP<qmK|fq1Wi^Mr~$p(tqryp5X-w|D!)q>7pF<mp^ksJ{^AM~6Xk|4I<w
zx(1YtqQLWWG#H1+!~TssKxuyxT)LM6jRt!_D<cgq>1M%<JNv=C-~eRqDFW$|QV{Gq
z0vjEx!0=2pYzaCG4eA#_?O!8Q315YC#y25n*<JW>vmJ8$I^jw0bMP#G11;P8;OMrm
zu%YxP{NoJN0~JD8^g|RiW=f#?9ce6_D~AOlium~o&l(nLU|5DWR+Z>u%w1!YQMJHD
zo2_uskR4vm02KFi!E7}TtPt|SB_fmXf#!4^m=uJ!(&ylcPa%9>vjmTKufV>@NVHJf
zh~oFQqIzo_y47vRp5vUKRhf)CPwmEproA}Wk&bOcS=_0Ti;w-e2WHnnoP7HbzEdg3
z151yhZ&M}D>YT*eyHBCr-?Mmh-FbZRtsbK{UPd+Ht2i$GI&Lv(Mfb~hacF)UN(es2
z#*$84>iY~`zr4iAoHuCV`W~ZuK4L=R7xXnA#9NntqV%laDAfB8@1GN-!HvSS*iMuV
zz8^t`dE&I*OOi(SO3|SmGE|^0ONYz(e#=6Ewj3Tsr5Z|9xI>w2-l&khvl=BPtCQkw
z4QdnPyDmp9lANba$9RA1;ci{3&(@=ZIr`MJ*MOp84M{W1h!lK`>9URq@3otdQN1bM
z*=9ysp60awpE+GWV?nTN46RciOA;+(DRzw|{ikU~?pLiSC)}DoD%;S-1{;cBYD<-h
zc2sqVpOseIQ?s4}Wj%7B<9i$_*pE@CB+&BPK(1LxF`-0LY@O)3xHAd9;*ZNNbi34*
z7N)w<na%EWdgVBpu*id^&-bLNd0u1{>P<(Nj;A+kedri>wAW_(QtMGaD!k0I)m@Wl
z;;+fHO=T+CI8LK~LDOk>ls|3B3!t!zGw5S?ARYP_M7r9uXxJ&3bCPG1Qsf-km^7ER
z<?+7R(fRcF^aAp|5JJ@#L+M!ULb`W+5nU_%k8F1?rsm~Kh>Z`U$r?+k`|VO{tyo4H
z%a+q??G+S$dj*-VUrDVh;WXt`IQ^TriZ;JmMJZ9M>6lyu2_1=`Rj!fr^hP8t#Wm!z
zaSb_J)UYe2HSDZo4O0oPVI#Y1m<FFKOD3LXqD^Pma*uPYRIHAzlD)tl2G{e~heoDQ
z-^9!xU1h-zH&}jGD;s_D9s}WrY=1%r!=Ps@c0&(yec#I(sz0-1ZG&vk|2LbSD+pfB
zqOeI(0(5m{pn0J@=yWN8;7)a*Fm0$^ZvdGmI3ro!61+0(V6Pp*)sL<qe8CI!^Znpe
z$~5TR83d;@=fbw*3qk+!Qn)0y8mfcWgZo+j-jE&#N)NX~y<QR&N2S1j9jUO^GYw?V
zX2KNf{oq)g4;R@%c+kXs<<rYy->akGw!R87C92_M{%M{Gss(|D3vg^=BS<}Jg5`6s
zf%dZ-uyyurcy{j|l(@Hp_0h-BC)ovF3!gz{HTRkSe9hUey<osSz&Ce&2A#uSq2cTx
zc%T0XbE=1-BL6RZy&!;34TSK^4q=qxoD><`5uCp#hR`IA1_F{8VK0UMLZz`XNd{+@
zjl}2YWl{UO9DZq)N9!93+}St^7avwc>0~8L2vx=kYZVOop@JbbyxaAk8V<;)qkV-s
zy18lKz;zAYx8!G|i~M1%iRy`(xTRAQ2Mx4v!$K|m$<G;UIosuJmln?Wtc6#;YoY5m
z&Nl4P!u*?Bc(hy#HKMdo){X!Cw<g~m@joA=iPGPBJ~w$Zu2UY3=lNVkR6zq%<JB?q
zy&BH)Rm0K6s;K={1xrx{zid`U?=wpHwpS6ARTNRlWfU6BSHNwXc&ciz9GVo#;_C{2
z)~k`h8E2$1@{|;YS4m=2i3B!giDSeLF+8(+1fH5His{ZGyk9Sj8+iw|zC!>nh5du?
z%74I@cb7ga`U!%f-=XH<H!x&hVAIV{a6h;Yro8C|x$w8ZzV^WT$d};P{|t64dkR-x
zbN~iE291_>IOO;MWJ~YDA-UU7y6Gm&d({HUepf-Ix(SY`UV@P^^^o}bJorwmg~XaO
zuteh&h;QfnfscIe8B_sEoUQF>Sq={QB_JwU%-@p=;nmxG=m^S%4Lmnm2bpmASQ<>z
z+6%4wc7yYX6u7%N2`+ry0ppi%gZkI8Fmz=L+-ceX9W9X{-M$i*_J)Cw*diDhJ0JcF
z4Tk9A8L+&6DomX;5q#^$!+*>Ja#~#YEED0do&)rJw1zLOV>p-B6!KaPA>^YjOw`i?
z`4#HmeM1?-oJPUROMK4~EDisONC0aT1)1GK@U`tPTYTatGphK?%v$=`VZM8k2<5$}
z=bh|A+(S0c>mJ)+af`J(USsRSF0(@o7nz1@Ei>&s#m+XKVD8r|SmlT#Ok!&(v$HB@
zQTm0<aaKOtcYi-~+mp?<@6KdSH`1A^dm8KS+sprsJ<PN<mDPXW%@W*pGmG+FY@f+4
z*3^{3dN!u8cRnd>wtWgSaZ6!amZz{MwJ9uM>@KF&x{G<`?`GvCsjTkn9(FEs9}CM#
zXBqD@*+kPEw*Az8mgteo{_^urhf^-w)|$%(e&sRQvj^C*uZ7Gaub2%UEM+C%4>QGm
z6>Q0p6YO>5DOO)~j-64t$XbqFVmFpvWd_r3vb2!9OdzkFt^U)&29ln$D(5$>L!*y%
z>3?P81AnqNHGf%mfe=J-PI4G$B#-|o&F5lra63p5ep+z%milN|uB!u?t_HmGW&(B%
z7O+~&8jSbZLzptcwn`UJ4fFt0IUg|Wnh5=M)8Kh!AY48(8y-Gd09QvXhJuyLq5j=!
zxSq5g_D<UZoAu*hnZtHiwJ;G}Dv}{sAr-{3_kj~-Ldm}z$m+@G8J>giL$(x7%{T%l
z>MFs`rJC>Y&%n@u^B^772uoKtgTVG1Absi%s0g;h@bV7m8hpm{Ij>>$iVu)7=L-aK
z2U`8{KQKc|2siQ_yxu9!(=nGu={i|V2^ocwdMfy2SRIY|>&4$;J#^JG#<CD|Y^mkD
zWivZ`ev-4LraI#(k#YF1)thrq{qRTLR1_|pfo(Oxc>l?K+%ESY?hIOnxmBxhhtfJU
z<xCxvh{Z8&@z__k1J}nT;f;kU*fS{=FS_o-iI9PDn2kOjxwwO$ca)cNkMGVyxbt`!
z3UnR8?;|Tw$L9o^CD!1nn`baksSXPkUc`}S8gaDr6}12F8rC)5Kq0N$7_{Xc4tMaK
z7Cc7btWK=!@5V5nmuQs#8bv?6Lq)qj3|aq~ce1|Xi<g6FE%ghBO@3oA{ljxQf@FU{
zh#D0{=y$9rWxO3h;4MyN`4apbBuU>br732G3~3!4NtbWPQfQwX35zOFjUx9-XeiPN
zH6@xVt4zhemFfKp6^dw3rQ}RCN?W8(?=3WF=#vK792-p{p_<e?N{eP+(xS)9wCS0g
z4t=fBp?v|mMBj8Na<3kpw9uzDE&61&$be=G7}BBxhIE=68cuc^QQcN!GS)L8-|Hr{
zdW|V{>6p>2duFsD!JH&rEvRzXg5oZWp@O)vq~vEw{i;@E|G|p7FIm&8d>gXcVoP=N
z?P%6`d-`JIK(qB6X_*qEk5WJ<M3FWM5s3>qk*0_<MToo5`H`+vrR+vS+U_*bY#dFt
z_n>k&Px|fWMg4)^WDqi*9GCe}bHoH%vB8(FZStcC&W#p{nna`5Os0UPQ)t5MsoY;N
zjry#n)ALdOr2oaA7PSP>?EN$7`ocikY#c-rdxFR!cP8mgm_>nKW|2yIFn#3tqvGb-
zG<W74p0k=mMH}bRVaa*qTr`if3+Izf_k1#nT0mw)3+Myq72(?w_TqjCyLF|6IXx|5
z3$03-$JJ6MbFhr9yim?6)sC<&XO1$}v<l`{QN`p$PqM!J8n$BX8NREkWjkJ9VCr#=
zESS%G?rv{k0sXg_dg(oOw)7#>8SG^FSufa;-EUdz(~oRs@&HrI8fM$$1YtN#6b$1e
z;B|`(c)2LRwr*vxD${^D`*h$zo*^u}WCj|4Eg@mDJv=;)@K@Un>JE58f{7o*H%^7g
ziv!`A`W%?{ECkM-S^`6PE1^1*--U{!puaH&?){91bUybTHEt*Dh}Z>U<$K`Ri!_MQ
z$%4J1`{74fKHTdo1Yg@jkQr45-|COR5ur+u@H_!a;%h+X;u+2#tOWzZi{L!1k@snv
zpf>9&2p_)=MGdz=zU2-)xpp76)we_8vB%Jq+6lXsbaU413&{KV3LH+n0nItRAolwM
zRPFf$x0Szu$-Zx}`R^d?n)eg5P7c8!?{&6Y{DVtJ1+e?9AeM4Q+_&?>c(hss%Zo(u
zdeR8|xI_&5UBt0UMgm3JB(Qa#B);~R;tnS%+>j-W$E;+q<dO_}Oc;qvuZ_eZlEtnv
z-kA`SL;Lx1xV=IS+ds>phN(PGo+*z@x60$P9C`FVB9G=L<Z-w{9v>9SV^fkmzFZ=Y
zzRvQfA|a2<Z_42^o~L%QlSAoFSrm(s#rY%oJSA@=e$^g{`}1XRt-K7@M@!?$9w|)m
zkiw)q{<-x@;B!X_Jhep}51tpp=K~|K(qsgB1c_ottOy=06~^9XA-wrQ5aWLdpexTo
zhT>n4@c#`hi-w?o)lay!mJ<=y3_#P0uMo1JA1+M(1k$Vzc8q=xfxkJs@zHCju6hM2
zaW7!*)MwzM@f23R<hk+E$D9}Q5SHk*L22hbIJ@r-T=r~*4+A$~PH_vkPr3@X2AjYp
z{}NQVHNb{f7vRBmK94ax2cB0>Lw9Hm$O@hWrNSyOajt+(o%|fLsT{VemBQxQVz3TA
z2q%XRfIvYWXu9(~)9WlK*_i=DW7A-$Z4cz~e#|4i6gbY^YUXPaAZRpap<IoFi7%tU
z5H~?$#ya@?Wi@=9z5<NSE#W(%g+OWZpk6!}t|iQX$uiR*=)gqy;OGPD-5&5|uPbbx
zMo_Ec2>JhPV8rLK(D>O5HVPR-y|F$BEYOC;V;ZnYQWZM4C_;jeJUDTmWr4ROfQT3*
zKM)2{-sg7m9cD!n2idgn&n&H~mvu|`uun-mZ>RT!E$(>00#4j!!zI_*$8%TMjlKpp
z&iy>oI(vrAm{QG*q^g*G-w}3jpp=Oj7qjgN1#Fvg9^2iK&3?3JuxzO`7QS^4TV=7E
zO;=4}Nv=ED^uk2;!8U>2|GAwR_HJXt|KiyV7SC$-#IZ)fI3}<=mZe(7vgeOuSWkQm
zdleGH)<ngy-qslAYZJ??i(}aiiepit@$8-CHYPe}J6j@~z?gFq6MmV@LQ;3L3g<n{
zKXwmW@+g(f2}x!1vQwEc|2zbK?PX8Cq_KVenM`<O4jZAE$G!y@usIz?tUaZa)h;{2
z+SXJu#nO`u{+wp}HrKHx?FM$Vw~4tvX<>UlwX*Z3512s8V^*io%_1+oVlT3L+54=|
zETdtN{ZRPL<R%EhPyye;ONqmSR4KS<A`3l_72s%rGPp&l!<i*oAiGo#=4>&7+Xv0y
zPWxDp*RX}c)sFCzE0ilHyMf~qPcYp$0aC_HhU>%Aq4aeSn17lB#j2sODQpQ?zgPiI
z36bz2n0v26qak%lJlxv311>*L0?8S>;OCn?ke!tdwF|RhXi^^B4=9Av4TqrZbUCz+
ztN@+(6Ob%>8U)YQLjA^iaGlWvev?|@ZAdH3%D4}TZyrJQ%%|Mt`4U3P-$75pCy-4Y
zfK~NFaB7zTs^1mHkEUXnnJbCGdLwc1C3(EOK?$ew4r05hCVn>1MNJ1oyfViW$L5T|
z+_%=~<m-UG4TwVCu6W|92cF&GgQ?Dwu!--n{&jGV4d-0#tyzGXr~kvB&C9T=XBD25
zUx$AuZo)lTF*tTZJdQEjfyn*C-lfUtx^g#4IPJwm`E*SCn2CSx<Y3ylJX9_(K+n7)
zJ_jzr;Pi6bm30(<9jL^yM^0en*&6)Zat1@YYH>Nw@GKH*#6ry`Tx5F{2gYB=9l^KI
zZ`mD;SbrZiquX&^>|@N`(uoxj-FRo-3pDZQ!DpIp(cn`rTGaF*Tk#pwRKMb$vjd!w
z`UB-}{=yWO-<VPS7ng1ppy&gF6n<8S%&rU5(fcCw^?@jjY#l)=^<rdEE>2NN5_}#a
zNxSW&sQsT51zeG)+3_;ujU#DI-$?S=FH2+G<!En@9KF~kPifi;q;^h$-cK4un_r9~
z_ci>SB&<Yy*+kRylquvS_l~-$kkmaDS{bTJ0fVZflB`A}bk!-mPMtCWHR$x921(_N
zrmN1Hr0`agMrCNxYfo(|`lU@Xs&z<ul`ged>QT~w9*w)GPw#db(Bfc2DjsV@(IbrM
z(F<eR-e^LzicQHd(Tr-s&FSV$3$pYYLq)b@sY~CI@|3K&f6AIR|F$N<FE;e>wJnu(
z*wN@a_VoO+1LdA}r2aBS30Xjb+mS?85!nPek(HA(6>GZC*kKpWVs<6_BW@%S<xYn^
z#?dWt53+0FXQen#8n*Ky`IlbY$>B{KOvjVxmGSg?rVmxV^C6eD|KsSqqp|+~H%^Em
z38h3SQ5ur^ct5Y_o3y2&(9oc%6zxsJs+5tHL`1YyqEIAKQi{rI2<=^@t@(ZZet$Uk
zaX7d8+~@XwKVR4Lx*nIk{NlI0VCRgJ-P?_qU&oD?k628QvnnRYPqs{y$MhHZ{bC3C
zW4B52r6Vy^>m5Vx<6=m2cnmeKi=p*RF|^?1KH7FXmhL~>PjhC(Q=M7@6*MPOqfRoN
zSdl`@RE|(x%`v+9<OEUQ)8v?Oh62{*&^B?s7+WlK9;YkxtoL;qHMX2Qaw=)=y8EOe
zuBjJ3KBpd;@91vnC$Y!+Nu3oGVXUEovY~CDzPBBgs%yhqTqi2*yCSQt5$@IZ!i7s_
z2+A}UtgXSQJx_Rk%LXT#N8o3Vu{b|>B5tHR;gg07PW!l`|F8MrU=NJ#w;XF~*C1ry
zCiI)PP3$H7aYtMq_G}Kqh}FU9vNQ~K3nI|gH5w<T?ZbviakxGv0e|Eqc$=r-nBfup
zP&<y=rW3eYk&eOXnMeuD!d~GU4L8h#(W87+#};DvxXV!bT#VevYfv-0fw=54^q|{t
z$f&?W!)k2zy^FT@@58*;L*y*1hsXX$(75~rUJcJMx$y=3n_uC^t2b!A^9}_kKj6D}
z6XwdzDERORl@VXy+WQ-f(!ZnDqy^)`eq!LOUl?m4&Ra|VqGRMgm`)Y5(R~Hho1(~B
zVn>!>(TZ~;#GLe|mDpJ+@%n8_oSd)3=^vFiO88MG3QgqURAshxSLQ9A!k^-y%-OS*
zxy(+PzndxZWF=*8s#4;w@k;DFONn!|mH1*&D=zd9dwbPZyy&1Jdkj?Mp63<#8Wgyb
z=)f53_!p7=|6u;FUoaKdM~8JSu<!8$+b@5^vKe2o;^k+2Uh)ZZpMQkGv?lB<{D4+H
z-s6?`TkI)+jY(ZzVT0QXOiOr%A$Ol(dYi{klp3&VP8}Zk*TO&fKEf{Cg+)a*VxCsw
zRnr~Vw+LPBZy9bV+(e&$*P-{b1X<0+=<@P17T&vrjH?AO%*@A){dst}?L5{l$ibyi
zS$J!F2F8jRI9PWIdvi`;=$_-KS#%UJHiwa~lLEgtNiZl(#HHv1*uOLmw}*+|M7<c4
zH%8%7Ndy)j+=FWyL-EKd7$*kof^i#hzJ3viuSNbCcF-4AKQ}|`+(vX=w-&31ti;f-
zp15#sF@|hj0M(Ima71Gkq8?0xdzuq2ZF3NwvT<lFk3n$H5xCu9C@!>N*s5D$lAa|V
znVX~DsV@cwnj-T`PxRI94#$;Uak)Vce_VCp-l&D4;q7pIWLu=Eslxb?5^75nps?*1
z&5HO+kFPY*{*G@cV*7J?qWy^S^B>R_|7tqD;5KQ`y+N@XuF|5kOLYCm1?s!t9JPCU
zhLpq7DSJ{HJsx<3Rt-EvF_RLhCL)fy{)nOCz$gmka9U#!O4|qQrqq=?Nq!wjfA@;n
z#oL$q_TNH5<(nv@dOdZN*V5A;t0+!oC7pFyF7#Y4T6fly^iD6OqYswQ+CfWbd5#CY
zTH-;I9Xv>Bu?LMg??EwkOUT_|DJ>iBN%u;;=)~3KG}?U?eSf-^x<+rJ8>hF?DU$%2
z+I~Crd>=?RR|Zhp0)JXz>rds+{Aq}BAQ{weCxc%*Dcfr|bu<Yjef@B{Ju-?+55&;v
zZgJ#zJ%N75B$NBD!&H`doUY$JMP`G}(34X+^nJnw3f8<tcRv-A`Oi|)Hn~l`)>f1G
z>j%`y{}J6B@qz~Tdq?f8o9W@Y?^JyM4~@2Lg<&BoxY42p3vUf1YiMCih7KGT>Ored
zSG4}s4PWbe;$68ZCfw?a2My-PP#6q{{{#bdtBkM)TRfOP3ML=MVqWq@9Ngf9zpJMs
zI%YPeyqJfUd5iI0a~Z4}SHnYb6XNW*;p{Vithd>LX3@z}F)S1*Z^H5HL^RZQ#v;-C
z0EX>ILhf;)FTOa2S#$~mQZr$0mV>d^^N`?Qh>bIfF~*@3yXN1*&3#pfYPye^o(=HR
zcn<sOH@G1>K0iMEiuH!Sv3!&wOM8^rph=Zg7OHdJ2Tgt+tj!&Sw=hR^Uk$D^<l2%R
z+;Z83>u#HIjbMY1>SD<UrdaXZMCN|qhKg*~NKU$CC)n^4cx>HdE)lu&iQQ)M-AXr}
z9le0>&-dVwLzi*5>1saFbpvlQ+|2Lx`tmsA0M5_Z&fO*lv0c+{ZXF%UCPTw{;fF}x
zlDwCVX2$ZjPVt<0FM%`SlQ?gF3J>ppgbTkP<Kx9A*dsEXUFT-<(*ap*|1*c{i*xx|
zNIu6o7V=Q-%Yvy>%mV|j@!P>Sctu?qw`{)6S9L47Q%W^k^}fd&k{<97%{s1H(ZJV>
zA9JL}GrlzD1vjpK#bx1dxJ%MIPD%Q}6C;~geO)s@8~d5xxBJRY#oyS$>j#^+{>cuJ
zzgWM+A8sG;m$P60W5;9#`Qlhb`O8~H*)gb<{9g|xSs_PB*0)oZ-&QNj59X-IPOnwu
z!sV^y-Je^_h8tAnh89(!v#7~_f7N863&@+5+sX|=ZRLZS>as?Ry1cTxhOBi$LpDP@
z`D<xA`KYU=?D<7gUJ}+`F6yTxhm~o`FIIMtYdUMo?#0@2{2Cp3jj68uqE1(S9@9~F
zo6<?P=_owb5A@`$q|UOMXBU}BU)EFWDzC5YD)%^TAV1k|C~tA=CZD73@+4g&S-IIr
z-gB#moF@7)+6EiT?n`>f%j`_#%3h}OXvN<0n!4U{+nhdfPN<pOYe8RG!K$D9Ri(dt
zwz9wcGJb$;JI`ER)pMY1`DmbQv(G}FJZ_LY_{Sje_buhqBL~YGPX^1^Hx7}%wf#>n
z-}j%K(cMb675?oN#*%zIS!CF{lRP<@<Tx|PPcA}Uv4G`o2C}^TnJgE?S<81dY~=D4
zL3GwQh{j&uNma2s$t8U!-BAysfRjP=Zuc%KKD3*LHV0Go+EAL>e-Ay>i=equ6paer
zOC5D%DZDa{;;to--IpYKHY1h3wH&2McTbRV^J!W+A&YvvJ5N_{UnJ!}mud34>w?~W
zi<UZ6Q>)Bcx;5_!eR6wElMgphqQ!R#|Mi#J`zoP1UKLii)bT(^3j>zw;@_>#I5A9Q
zSqplgevB#VYx`oe_dvl283O4M;m%<jT=gD-ZEkija-0AkCr4<{pN4MRW?}EKxsdDK
zk=fM~Tj#HYamG4ySMtH1{=P_g;E$Re+wpF2km%^#4X5BxY#I^{i-(b-D`qdKdn~qJ
zh=Z$h0>-=)S@boBu;rJ?zid8=%<sps$nzvl+!H%+>rBB{K8yM9vjyAhJW@90Vbz&@
zWIic?scI2odR{^4;H%gzCW6COrFbRw<J~pOq5tL<j-S7S*v*y58B&egk86<bdk@`p
zAK+bFEgtIDA;+&CBc3$CZt!CSZFqw6Q%_M_^9&oBo@2?+7ufXeCDuF>OrN6Hg4^*1
z4HHGj(eJmI82=7$%>+mH*n7<CB<6{=A26io178360K>kGxI4NL-5eXCGNBQNtQw)y
zp%Fe0KR`L=1L}u;fKA<d%<*`Sz!&e(>@2>{zD1?VTfDJ<gWbNb(IM>>?w7p8FrlR!
zd-xn{s-Hpm@>5()eFEhzk0Fm2jJx&?P`+6Qmn{!r)VmhB*Y9J?{Cl|ir3NpyRl`EL
z3Sa#y;MZ~+24Y#<?p_(9ZEm7Du@rNEU4x#p$XOpL#(=Mv@y@mgr*{-0x%8sQp5!B8
zR36@LK9AR#ImmjJ1rP1B=r%MHO^Z(BTgWL)$T)#^6~{5P=_tmiAAy2ND$=YEVX0G+
z$lxW4PT2$awkr-#_Q#^;SPa^ni$+FqB>q)|!{kvICch8C@fPu!P~8QOjys{)Ydb6k
z2jI^LKRi6L70RXFSoVHB)H|#ZoEX8db@fE-_QeQIUx2Xtb8t*)HogpWL7eLp+zOeD
znu`;l+&mUVrla9EXE<)g+h9l?Ft>*ll)NnQAm1FPbo*h_+TMt+Glq+U5mw(Yz&Xb*
z`1-sf8UwX4)v`Tq3emXkoElofRp7I!6|A5AC8f_lXj0G5)VxyoM;czy=A}=mw{ATZ
z)ZU||7b|J%g>uTNx=s&OifOR$&`i(HBTdU38h0&|%+{Tvck*#+>Ux-tc1))IW(nlx
zvY*lp?WI|5A}KgDj1E`?)A%ntDWf)!2EFv7+m>7D{fCWow3V1S)~%pEV?C+!mc=wr
zXCcL_&7=J;Zj|w778y3rprC#(6nS_WxwuTF&o)!&x}!5)2zL_TSV!s;>_`Xg9O;&Y
zBi*%kq`DYKT59M-(hDbwZgHlYo>R$u`g96@J%i?Xxza56`4sNGgp$-&Q*7jVI?3zk
z<>M7}XoVMT@b{!4Grh>&YdKw6x{CIgtfO~6n`p$;tz>)Kj~1r}(wfwr!Z5#^N;-y-
zzJCP0*W63%F2vI4od?KxbrSvbNu`R^V|3;HNm81UNl$CDslqRh=Gqt1aLZz$AC;1h
z=PlZsUq!B#59n@L0~v%or>`sCQt5_|bmGuAa&P)g<I)u||BEul+pA%v;KF+twuge~
zSRdb~6OLZ)f+9~t#P{lf-Hp94<YpfP<@ZN^u?2>{9D?J<K+ZNB^!_s(=Mu-D)#CA3
z?J^l=>!u2gc_yr+x%klFj>9>g@E2Y!`)?bt+I1@qZt}<Chue|pybBFaLhxXFIOf|%
zW0PhqoIc0Hx$z(j6jHIp>=@RI>sJ438F*rojdyo)aoDc_o}wRO_wehGoy&1xXC)3l
zx+hqV^(g!E6pB}0qwSeSd@BBekUu|h;gkYvG%2yu2vy-xZp+06ntbz62i`>;d0u^I
z&I>nS6(=LUYSfE^e+w4s^L||VV4z?=4Cc&N68kAwv!Tf_UO#md?}!-7V;)T4)BPPe
zFL)}?_%?%myxh2T<9vSLy;$^ydhwy`m29$L9k=f4%}sS%*{0f$4-5@t=Tke_&}bJQ
ziwWi=tuQ_k8qQAbqS$KBUVf+@%UAZrvAbac4@o)5@qG^Q?~|#VJ@6=Z&p6JZ15Wa!
zqv?FcAd_2#p5@^0*&I9TJa0Uo$6MN6<i9ftd3r(-zrTBh&03f6j=tA9aO4fnm|VuG
zj<;A%WVxaSRI<8SHIJ;RVdK5`IKl3L;Dgk1a9|yyeFL|IJmPkLAM?%mPdW0!Gp<m3
z!OG)ba;E<)UXt;euUEg}l^@@-hVpx!r1gP6bZq40j!kT)^^u3SZsrp|L|4V@PwY|i
znFkepVXKp0xzE0DeAfRv`>goE-Q8OFzezvYe&jEH!oPWg)gSh;`pZ+rZtZkW1$kYX
zg52#tMY;5{q8#enO5XCSm3(BYlH8$-vV18=S<alUBG>;C-4`jX<*bRSviwU`K6^q<
zR-DsDR_fSRjwx>|M+T|OWurCZjw<bB>ymb|<}OXyX<~c1LR(9Ac%UT*Cv}j=dT7fd
zEOcb)hmQQXP*?u8qoX`%Vkg;7TTgDP(Ua@;b(XJA?;^W()R$-7(U;c*b(LLh4CIPN
z19|rWLwU@YZt}Hv-Q=2(?y|L+kz9GnNZ#b!L+<vZhkSWSPkHNyo^ptnu{`CKvFzj4
zOIEJxC9fD`BLB=Yk(&ma%4#{La=Be^c~o<6c}lMEG$jb#X=`6OQ@fwMGjc8EB(9+x
z@73g9vWl8AS5cnTYN{H(hUzY_rMPr)7W=!A>`wWRVcu4H*1?aufAyy}s-i)8>`wZ6
zeitpC8A3}2>>=a9k%Gg%mp0YKQc%DFvRIfz>AtDt`0yCbU4M!iozKvTb>}F(E}zEj
zx=fF}uhW&&w}?$@DCpip>U8-joqqI&ZuAvrql6z+-9-UcKP%yGttz6+)uDc~JuX%1
zAoaN(ekmAWbRQ!a%<2WhxIVb~qCb3w4#L>f|IkK{F*#-^R`eJNzjJn2J$(YS6da+G
zF%@QuW<s&Y987w&02K#4FnQiG9PYUqIZxMPU6K!04Df|dkUw5M4#ZHaozM*0g*SpR
zxS+e>RV)lgd`hI~NQ}n84*Rg!c0ZyQ#pB)%k>@;k5MG(du(*&S*cFGdH0LOm96gTg
zJtv^Q;uIE*K8>Z?nfP9F22b{6;iqj5^j@BWhJP+j>RiB$xO|Xaf!MznVnWj;L<n8T
zBkl?k8jA6^QwgSzyoU7)uES(QDJ-|%z><wOvCq8>A*0LDrSmOl*5AU2=-UXfxr3|s
z@1V@R0{+iMkIj@y#O73@qD>V%9jY)lunMP6RAG8?6&Bs9!kbbtW8_pJWp5R97Kz{K
zQ-$ifN^A|N#6g2ftT<MI0lh2GW&a(qTf2<|({JN?>McxuS&l%1a^%{TVU)*BY!R>N
z!tPQ`5511#JFcP4h7#O%y9%eF#dz8A3ZA?u!rt^tLVqa4!hQwNf1Ho}unP#6@=#lQ
z9uGI3!<`N}_<A@Ce*c}t@k^OVu**Q$jdUEbKLyRJC-B}n4GS`lA-d;L90)&*xL+wa
zvfz;Luq45u-$8T<PJq+vc(flGhb8f`*w`F{>7)0ehhQdseHno-Vs;(j7lzhlA!23;
zMwII=<fQJz-?!TlXAy|pb^h3y=ZkrNx8VFvZ_Ml<Sd;N<u*`HN_>33Y*e=0`GIuo1
znh)*wZkV)X782UK;M1Wg7%n*?uxui7=8r>d%NYEN7zyq-45zPJBXl01{ZC|U5-l-t
z*g#x+-Va_OeQ-rG!R%K(P!rz`_ojA*PN&WoUayPpCp*9^KohSQsAHmo8k|O|AZcVP
zoH+QG6f1tve+HkagZ~GzRC-N{Nl)pVOFg*^xKH_pRTO4+i>6F26<MWXvJ<-|&AAsy
z=Y0+x-E)QpjZ3E?UDIe<t0T1FUouVBP9#e?PWXspsPcBC$UN<#j+cU|Y*rAp*V;}e
z>;1$zW(&=W+(1{WR*`|T7wyqqOz{Tusc?lW_0yg~GgPNihNC0de40ppugB3s%dzB8
zHk#&Vj-<N}htr(V!|3>{p;VS{L#GO@Y5aScKG@3CJD2J9bfyOeOj28>8$FqVmNGqh
z$+YXBOur9W)3X*E+E{8!lZTEVuhXMxLD@Ju^v;R&_qkD$#X{P!VjcyoaG^W<#mr>k
zM8`6nsKk8=#ebhh^Q4&+uHr`e<LA@tZ;R+$iznS4w~9VJSx+gsn@MA_A9;lZ(tygH
zf{hbQI2A^NMn%$<Z+l7Y>V6t{I)S!jCzIR#!(`t%jUKE^r`+Z<v^V4&Y1`#fR*xd8
zG$<iIx=Fj&-l3&;YG}&%hcx{8V^WNJNwYlPlgac?<UaccE%p9OYh7AF$yWtZmKsid
zRfqlP_9C>XjW0T#&@ZM7D*GEky;k%XC-uTwzdl&)+aFsFTA=>%5M+!2sw!>pV8;mj
zU0{dqo)eINz!ATGOhfv%*;r&TAAxN>FtOcoDAQW}-X%V}e!lo}CjjMxcS1L9H+l~V
zLwIEbRCnx!*Cf$#J}?1Y`y}HS4r9OjaTuOD1(yzI;1-nwvmqA{^`#I~YQ(kYZYjPB
zt#r-MYWSr-z>j{9;9LD1(Ft!6u;U}vhke7wtluc~QDo~A%53^tm46OWXAaWj4R1T}
zWS5S-x}r13Pc`8E&%5)et;XytJUntmUtY4?oE01_`FB?<p7I5(c+ZBvUm4E5E{x{Y
z`SyI}+C=_X=g2pHO=bOFGsSbbu|oKK{#dz)Rl0d{`N|c1<Hj0Z-ftuKh}_JE+kDyg
zoIgKk4&;jic5?pGUF>>HbOyf+<sjp5R+<~ht>UA(xhjUK?S6($JiB-$@Ryi_yevPN
zjq6g_=j&mvRTbKgb{e~MJjt%@)7eTPgO|L@<n6_0*>qnve{ws=ab0t{zATS}R_F6t
zjRFpiE94@BA~uY=%$a|#aKh}XtesK96MkId=>JN2xW^673ctxSPnU68aXFh-+~R39
zw*_D14jWvn;8)p|T$WhHtNg2Z^Sm04wZ6;CbnkKMn|u5`|2|iQJz&kbwJZ&K$j@7e
zY<x`}Yo*n5l5Ydo%y=Xi@{jp<$0uCa@`SHHdCE&}J>$29&v{qQ3-&(ql3$;F#b5JY
z^Z9FU*tYg9M|^t6$HYwKJ*1I8xi<0VkdN$A)Xd+0e&RTM;kH}8a_g(#*sjA5_M6wj
z^Rs^n?dmsYFZ#nvul!|`4hr&yA_aN%YDKx;SY*;_TFG5^D#@!!S+;wlEW7Mi5!}Pp
za=&k_<t2wy<%bj0WZSQTzZl;}t{U1_^n(ij_7-*7Lr+8Qm#!h34{s+AyVFh{JWEr4
z@kCRW7q*xCJZ&#$%+it@?`X+JBRa^ZPIr)P^tI(0e{EUiowjT?RY$&GtRn~5>dI8E
zEBD{jQSLmslUz4VPyQxyZ||ipvW~Zl;80AX{ti<~+j<H;dE-nD;&G_%R7##TjV>y?
z&=%F1G=0u&dfR3WS+t%{zntCa)MpQB_tJ}`{;Meb+Bz!F^`_0Oe5s(upLFkUCq4CD
z6uvTqJlcj+d~FozKHEo;CI@J9N)nxQJxnX7q*3>sr)lh$EZVa_j|$dYqE}l>C@!ar
z3_Dd(PV58WxOhy(Utd!5^$*mT@r5)`{-SfU6$Nu%8B4aSL7|TZvZ~udIYb9D9rW<K
zYghdFBA^rxdcyI#DNYLyPk6C8RPR_q<Czr#e={a_wZ-tEBQb2D9V|k|LpggA&euC*
znu-fP_7(f@iE|O<wGfRvJn%B!3pr^k5r1YaYR_y$LE2{AoZt&rTYp^cA1Kb-J5cp)
zCmJhv;ry{+eAy5R_hEZbsThIQ`H{G_TzE6O#K869K9srahr#<e^xAL$1uY3ET5}L;
zFOuLj@er<_Oc9)&!^oX<1WHjyaj5JVmi|5ti=HQNL-;`2Oge?jQ`2$H@ig9z$bh0*
zChC;VU~<(N4BmGZvnFLB=~ou&;<NE#U=A|R<)H20b9j_|4xN?H<Is%rBFk}JFcr_^
zSi4-bvk+&t(Ye^<ATng*a?yr!(P)s1xu4Ht>qYUqKIbuGfOxDrhlXY1d4A{Mgl`VE
zwPfSK;%tn%nI-zkvM^!oSq#lPgZgioSkpHX4pTF5VZ&*tMW(|y^%Pd0K8ZeOPvFDp
zG^igrj?%ryuzcH5*e*DN&BG64v3@FS8V>;%lM%f$2>}xiLT;akcV!1~Wk<Z2XW|gi
z6pMR_`|xyJ45~gyqjY~1o>)iXzxr@Ac<;gIj$sH+4?&*b;K$tGg+WV$kleBZzl1L5
ztR0BU@&1@AyqaN7+t6>;W>nAJgt7(eF>B!(lnXs?<(y^MJ98;QrYy#>i3>4g%sgxv
z>V^`lS#S{gi+!J|c+tZNqYWk@KyLz0>)4}8YYaYWj)aotFeq!<;H@SjT}y(U&JbjG
z8ie2a<{~@Q5B8>g5IDdD3RXSw#kM<A#*6;y8C|e`Nhi2%*1^t5Ei6AN{4zJ%;`j$u
z^zEb~c4)0&{N*nNEo-4)-#$}OL?d;weM3eH&*^hb1Nj#{pp+}sq*8yI%r$S4&-4=N
zmRm#ygD;ZP)$`=DLNI3pXXa&_lcX!O)GmKhXh!EG!L2$#qj$%W@{?#9Fg}8!u7^_J
z8M~?VpB*$UD}Y?L`O>?5Z)%^qmW;BO)3|02YMHTszJGM3jptox;8|xH^W1?ZPZ~#t
zKSq<!)8W+EVJOYn#Z)#<qW05=(AiUisA<ALIy-0pJ)Yi=_T-t7=afFw*{C-~bT%a)
z;RIi@trz`$WlSxrjLF>4n3lcmNnQ_o(#=mjNgiTMuTqU^%!FR#`=1HjT4hQ_27O4|
zy)Vfr1875%C7s+QQ}mJvG<}sb1%7d$zHNq44@0KKfkb(?iChIHVaEb%vNX4)u#F=q
zb-)<%wHZf`;vLA-)|m$Na3Kq8S9%vRpR_x8(12^pXm<J<>Q%Cd%v618)yM!+jNL(s
zD!VCTcPK3~i=c4vo@JbkrD;hAsN=CDT5~g%9=APC2N#{9OAVQ%y(WiD1dC^UV<8=`
zDyAWorS$RjEqdCinrPMoT9Dm9rY6rRE$t2M8q-8~TfR__62Xg~CVC@gD523)<k`a8
zz^GIMo_bmc-=c%Ve>x$5zdmj{cf%2pGkx916pds03Z{iQY?>^Q>n-?3{jAYRaX2>r
z8VxP|@xbiKa4DLKr(<Wq^V3{hytoKY(!9_wZ#9O#-w5>)+pwg;A8iKiz?G<72x%9J
z1qtDB92JdOO0n4XFdoOt4<f5JMSOn@wqsAC?sx`*`($HFaW2kpDS*M0E0}F}9j&LA
zW1D{^v`X*crCB`;Q=f|enb-LI_yaB_f5w!JEofZu51%qyarFfio_|}7SG-r}d)n<e
zV1zc$6J37`FLvhbY6g69c6V-_Co-2trhH_N89!GYz`uelxI%jfk2);6<c7&SMKHDk
z!$$D>;bR1M+@9xNoXGpPJF>x)sqEQj1}nC5<#CTiFU;kI>~YM4AB2m38lRP%x@av=
za@r{N@|$^@jxV2U?a%f<M2FU^?fjx@CpYBp;%CXh?CBrMZ(aBBy}=QDLMe)$mqoMk
zo*0g@kL7N^_p|kpcn-Ep;4_aBdExRTp7lAI=Pgg+F^^LD-RL7MA3DnYzZ~OYIgNET
zoe)~fNuFABijRCr=Ot}3_;sgD_SQecn>(Im7qu+j@;Qs=-^=EmSvkBS_#CIqKF=9_
za@q7_F2|hB<LEUPILjiRcfZT$BdHfTc6I?b=@hc1=(Tw1e~ClnB2N5T#CtL?^Q{$E
zxX7ZIuYD<IDgP?Z+gZXP)2<1>!F9f;Sjw{-N;$3Y20uD{leHqs*xs+4!`9#8hLyK@
zz{)$kbZrIuY^mfC!BxC3p_(1CYWTm~cUiab9-k7~`vR+4{yXO(_YA4y`+4<T_O5|b
zyFF$V=O-Kz^px)xJY&DF&$(Owmpo$rEB4#>nwQ*s!$E)Eve$t3oHX+TX9PBKpA$_S
zRsNB;zH4T0<<H!?^A~P3`N}Z^geGPFog4f7;6GhkI7#&<XT1B#de?vPivz#eap@nf
zw)o4`{Fkqt{KrL>3i8N33bIq9f_%<dQTDv5D8HG|O0N3RN}gG-BoES1kuR)lEuY`1
zDmR|BrdRGv6BLPdgjkVxD=XU9$BOn;SkdDr64l#-E-J`$TG56q#@SNW2gAuUeH85}
zwWFW?#*@=s2P!Ldq6XFJG$C>p_3@lb52D?vtF|XKzgS7JIvZ$C;ub0xCA@;~wo~QR
zU1V7mN&|Ewg<UX)jLhSyN-2p_S|6qngVX5i_S1CzXBJ%v%_CF!67BtWmB!w`NnNf~
z(B6W3^rf_(lwLfewpwo~W&B5qj{Qc*zyGF^EJaNFu8hc$YM8ZO9dmzZqSst)DBkG^
z3p;T>DmFlioe@snHpb_fy>b6_UwCdbN3ohERQLY}w>}WeENhqyA13<uMnZq0oj8w=
z!?7_Acp~x+Vd6fgy+`CbKg`0UQFHJ%Vm^%Ty2D3n2{sP%!V&ish~2gtuS3^iNccu{
z-{FHPE4HHWurJKoiFs&Q02W;fM6>Wbk8s}!>39%Mzu1Lc`oTD47lK-^P-yNBgTC;0
z)SU>&z>EkqiVV(|#3*zX9#(njURaEXL50Y^|ESpq--uX@7`Y$%@Al(GP#pXW;!%D$
z9w&Moz`t10Z__FP$L1xVGCKj?{w2WACJ`T3CSuY4L_Et$ghy#2{@h8#sq#eOhe^c2
zBZ*KCNW_~diEuGUgwKNnc!wll+kXkjtv-MmOAjFGTRaYKibwY6ICOB2!^iUd2shu4
z!D8p%`Tjom_1K5Z1u=M&D9(sg(ePG`##FN?SdWjy%Eb}rzcn1E1cTcAKp0-7ghK0R
z2<i?8V_VX0nC;sIwcSCm-?S5*+;?E_#O-)*83>z>0XXr+4^9=nNS?M0zA9V9+~kcT
zQ#NAE*LCoTSc9cztHg(NIn*Y4;g87GvG6NAd$ADR77Lx_#T>kw<BFPlGod`u1*Zz9
zqK$=^XA>u*s;vWx*Nw;WM`O`xtk7IUm*R50;iBhYsNl-Tu$%~d&9Z{6;SdbiF$i~s
zzjy4+{s_HlhQot;V^3l)jMC_d0`Knl_Qn9OXXqp1h917l9bufVjXpiJU?e<cgW9Si
z$X5++KUDC2m0;$6P=MFH-=f3dJ1u_liGF|lKyTIFkkt1%eV_V>hK1IWMMVug?pZ;L
z0?MfT^EJ|5cZCeq3aQ{!9;GeHq4k5$P>gOmxwK8Am%2yDPVi&gJQ77`WgH!8w~utT
zM$tvZaOxHtLfON1k@BzYbfCna#*f)X7mPNMjlmkaK+9<A`o%P@W*!ZhHk-D$Orw8=
zj<h*#0x9O%(VeCdw07c9x?2qz8T215TRMoGw+^7**UU(JtSS9bGA0MH1V8$pA?2LX
zCmr|BWHYl99p0@=b)U59Mpy?@n5jjrC%32LUYa!eWIOt*-j4EiY0$P_8l+#RPL!!m
zGY+d$Wu7{<d{L)J7Yzz~uR+ci+R?LWO$v6>q7iP|R6V*Qoh<4?Zj*b`(plDY`PnFH
zTQHQShMUm!huvwPZ#SBf+l>Z2>P{7zJ*Z5h7frlpN?-n(QQ^7)^x9&OKq&l2d#8eS
zXW7uN;UnnIuQ8PIcsyNinoNiLOrs9Lv#5>UJW4NHL}O2S(a4<DH2B>{sv5tIw!QbK
zJN9B0EZ9YU<3cH|F`U#+L{sWEG52^Kpo=S$$UY>Mh8G;83YAl2G(VH(Rb~qmUY^j=
z3TW)!|I623Co7|JidCwlh~IbVnTF^YwR%EPn_tqjC!!<WqnTWFztfgCzbR;)BJ9JI
zar%-fMrf*I-bPIfR?tTBk&bZm>LPme3{g0(2X^?F;B2889`+P0u~bW#jFK=#(HglA
zhrz6BG)})7hnOyt;ju*Qx5WQ|-J-cT-f<BkUU?#}eib&hY`|OVt*DvekF2EasBN_i
z3Y$X&7d0H&*-^0Cun$p=@%T3CAQm~Mpk)0~RA-%lg?a{DHf15>?|JNqyNE6mE+a|%
z8k|1d#IlEXz;!j~*i;KooyRC1`vMoj-a_F;6Q+;<ioHcYQD&^bM;lwQQ<2C~#H+Eg
z=#u!oSd)*==)f&by4>Kb$HQjpv!a(F|MfRwjbvltNjBx3&1SsbcmQvjWx*_1p)=~N
zcv&yuX<lW+aYe)UM8{E_;$_EAua4tOMh?7dvm^W6pTds^yYT%`;ctH7CODt-xn#FH
zufOZTlMKArebx%T7P*>(iq>)A%Z<E4^bh~m+Qz1B{n&ZCKOb%h;3B~*bS~S$T782!
zd)qEnE#J-eI)w1wiJ?5lFO1)w*uyd9;k@`|1bcpq<orKT?DIRC?|s_Kt)Im3*lYXv
z<>6Q!v}HdBOpIgJix-*Dc-|CyfEy+!@I2*2_CA})scr|^wRIBrJdwn!CMI*~hh#ny
za)<-^rtpJnDQw`L%7v{C^T6c8Tr=zl>pnfg<F+5=TYZmlbLBBs*m|68`ls>B`ZTtP
zI>Gi6PV%far#RroDX!m>&KqZ(=J~xd_&`ero8HOfZ)s=v_pY;SvMh_8r)0B-Z4O(T
zpW`D&=Xp)1Twc{aj~A(5VBNO)EURDSXw3q?FXpT~{Y(6|XA$f5zszs{yTTg7in-Ug
zt6b$=!uhkW@wSE6Id5qxtFO4h?N{C8c`M8KpjSCBbic)uXWnMT@po8Xs^DSWgwI={
zis#i;@s%^xtQ=6oA1B=9#aj1xrs!R*4Y<!nLmqH^!vod`tmWe-4|&GbhumdR9j9p2
z^S(><>=)L+0l|-W(Zk2AbN(sc_I$zj|GegU)AyXXyaz>;8Pdi8eJVNKnF?C;$nv)y
z&G6|=JEwOcHD7)D@UAPZj4>qLE#2vCVh?KmX-uCZObH9kXvgaQ^x)z^^0FID=d>hR
z-&Np^me^9{pOKXDU@WC6IFQvkXZna4)X#B_@Ho0t;TA7i5WR+q{&~|QH$Mt+-%fiY
zcG29&p>)?Fk}f=mp^7c>q)bT^ts-)$&ySN)Wjb}gb(VJAJ5N0tFOsqL6;d2^ou&qs
zQ$bZFUADYW=aTE`tkE;tdipg*Pi!R9pP#8$Rtt^w{znmqTOm491@i@qr{4y3^jfBg
z?F&1I?jK$3nxcnG<Mk0b+z{jcGlH9`F$U_IqVL~6xcsaimI)2$@WDZNy><xBj}%!J
zb%s%yHEIHFMUUYK^sgTU)y;MY>M#y75+<N`_enS!<A`vjDM)aihR}Q$(MLQBhGShZ
zX7d~@Oq~bYOAC;A+a2R7Meo-2B{-hpiMPQ*OPC>g`nrhw^-ZxeTfP>@mDZzb*9Lr6
z*o0CKZ|D~Jh)nwyXpGnjUb;>6A^T!cu`k?g{ctJM4=cO+W9DYDi?8;_cH;n~ED1pN
zkpP^o4Zv=dKn(94h-%wFjF}XOxT%4ta|(o>T_9Y=EK#T%i0O?1uqg~c#LfV$vI`JB
zkN)sV_s89-;_<5=W{3Kr#?TLKj{2gnnJ;R`ZbN&^t#H%af*}n)P&nib^O>7quDTJP
z$JfJo_&Q9wy+$yER>S1B$iEI>fleov!Bf)<#!HrBZ@vdQwONes4(`|ywg7Ff&%@1M
zbI`W08&sTU<J<a~(2a3HM8-5EmQ2B-`%ch#IT<&b9B}6A1ibk^4&mR%qNI5Yj=vd&
zsSP6pPi7c=3x>ib%^Jg^7$-Lo(nLqWJli4o+szVZeh<Xy`+`?;wjbO=%&=`?Z`_bg
zuuj_;xvz~-TF?!5qYa?$sgJPXow2oZCwyztf!g&BC`@XPUf%6MQ`FI7-UbWXs$$Jc
zW%Rhx3QLm|knsDr*nj+>9~-{V1?eMoX?aJJ^Iy>^zh|_^{t+z}J=y1<-X-1aO6n1E
zi=sSkkoou$DjIT`Oic@Dgi#*-HqN18mS?DCd^)MGPNS)ZkI>PVhv>4^L2`<Trw=Nz
zq`4!SItnk`$lOrM7e3nUgLYDoY9Q@x^rezAACl_V(V2EDX|C&1^1Qi_RHo0N&HrYQ
zb@3FcO`Sv?lgClr`O)IuJB;?q*3|2$L~8~PCjA<7s)+3?dNWN)Z);CdIn<5rz1F9V
zQ}n3ED;=^+(V`u}?P$fJw&eO&m4>>j(0pAbn)E|~+<N?#g3kYxK4pHF6xF^;T3Mf@
zSLq+6q0Nob(6t|=PlMh|2`2BPrtxnjgOoSYgT8O1(a&E?Z!2C)rQcsm`Lo_g{ae44
z)+oJ`v=+UWcwD2j@Zm>kvFjJ<k!_1~bg%*$hqtD^{8j3_^M^FN_@@-ArA9h0lu7eH
zWhz^*Ov^n~$o^$(8h%%e=JZmh;YIDJb+UNPc{+5hwH}3w>zQ`1ZgjA*2Q6$cA&XCD
zG}g?VMg&+=(?2U3BX(m8=MAHPF{A09!#HyEnMAr}Q>bP5OuAA(hlV7%lj3eqIv2l+
zPS$Lo57JiBpXX1Xa<`M3=`OmO8bW^mh12ePqO5VZ;QUOCr_}BTsa`RK9)3PTTHn%W
zbeq!@V}6#r7oVfN><hHA>m~XbS4^L+N~!0ka`L=gNqH6b=+mz{s+jzgo>#siufPv9
zdGTk`-qJ!PCI4twPbGBT+!_iW+dyrr=nFOJfG)3eF}<iWY_bfnz0e39p7p}}Zf2Oe
zW&jEwi}`%96*jezkyB!e%Sofq9BVJGI}Vur#u*z%&4B9_H_TqV5QhdW1skt~vdwy&
z-nJPJ*ZCpiQXpm;2`*1`FftAI;C@~t$~<F)790mv)kGM6OvdrIhtc}maYS}V$9d~B
zXk4Fzi-mb8>{N(@K38DScnyln%h31N9c<oPgJ;&YaBObC#WT;K@An$BXMez|QJ*jZ
z-y!b0n2@T#3q&vA)?yX*aaCjU-)(tuQak=&uf@JUw7K>~M^0PNnfG_^%A+3%Uf3Zc
z?(1R9Yb{Lq(9b^HyQCj4i!|rEZiBc`WZ5@;wPL3Na73UrZ?v=JF|9|i)umDFFM9A6
z_8iB0r4#tx;z|7Sk0bBiJ%u~8o6ffaXK>HvS-j5Cjja#Q<zMgS^KcV)PM^G(KP_9r
z<(oaZ{g!2XVEGDO?6`{8o2(J6_O)D*yq<l>ZDe|}iKngh;ZdJ9v$y+J_AT4SU!#1v
zpS7P@#`^KPE&e=QFF@p21GtTCAm1qu<lGtCdBC&n9JO=@Py4uo>(=h%&>uVbrcV(6
z{2Rm@{C9C;>)l)&EHY-=!8|iAnCJEi;i!xd9&8)R8}Ed2^ujRC`V}U+mG|(z!QmWJ
z9nM;tBDiCZNKP+{<OZK8PBD+>@6V(8X5wDfnjOQZ4EAx*<9+OZG?q84+Rrva<M>3o
zcn*IQFWA!uxM^1c|D2b|2jqhs(<zCMeoEq?+sXXp<RMPpox-C$QaOC|VYcddgkSwR
z!jAWka?i|TtQK^fGiIl8iTMd``E!E5m7e4kk*8R9dOFAHo#y<i)BIm>2H&vBWV?5n
z91wMe7g(L;M^$H8)hmlT{LA8bq1k*`FNeP*=CGQ{IZi%#j{EX?{#bopbOhz{xn+5r
zkamIVf-iD;_$4mwEjm6vl(O5m*5uhqk=Ctkkv#ELsz2~qy0!ANWZnCV6tLy1)Zp}8
z@`-7YLT!Ib!S?^8J~I?4!di)RC#g_}2vzE-)RuIPXi$T9dphf{O`i)o5(agl$qx-^
zNTCt^Y3N0_t<5N_b^vY8v!tkciSCTDp-CS{kmIAVWU4iZ`uk3$hJLf@o8|&K+-oUS
zE?z}$RU5_g_|i<ZKpOU8C;2r5lgYz9bl_<eUHGz(k~<urBS<2fWg_2x@)&*lagqug
zGs*5;Ho2SSQTovW+B)b8DU@8JfeXv%ZQBZJIbTD1%WCPD$s?No^ciU+zou(*Kai)<
zC#q}sM(Xju=%<+iCI+^GN}V!JS*W5~WZ2bjtK+4fCjQRSLQ|YJT&i_Zr>ZCVXoM~_
zt1A+H41wtGm~prV^3si=pJ4)<<Gt}}zZstT^~2A(1MpjL^3N*`!oLzr!HFA!jTTmL
z6l^xd)eu^+jKRCDF|+khByX@qZT&D*(+Ic)jl}4(QE*WngZUPAu$(*=Q48$hv~(P-
z7mojbFLk-aMCd9x;QBQO6l|S@+r1~F`toF?PjZCyEk`UK>IA(cC)j*;Lb>FO!e!3T
z-s_ATr<~Dr&KVJBolz`$F;ccUWBg=iJkxf@-fK<>U+x4|H7EQREi%hGj!54=8NTl(
zq2uUD$cuEq#p;Pzr9KgTqzNFG@q!0D4i|jvQNML8vNzfxe#scTa2$=dgGS+Bn~~U1
zI~<2ohT*=uE#~wXitTr;5xHH)EpvuXC81ViXZyCbLTcg=_?ZvJ*W5uEG~5F2Vz$g5
zIRI{j{jf&r3&WFru*I-9&IFs_!8c=Q%;_oSJ0tuv?JjnThM4fYtLRPAN5;O+c>SRh
zJ`eAR4^cY!@~i{;T52I{t0r>99$==9I&REr1IrXu6unVF%0OkPtZN0`JOzCE^M|r_
z{-mr9-^u*IXPVmkBRQUZPs4`2A*)+2$kz2K&H2zkQCmfqqt<=uo?1<dse;nV%jxBU
z8`Sex2`!HjjLiO*=+L!%db}W)Y87*+F8K@%A0>SEZ%)voh-0KP^e`Q0OeWpbMCv&w
zp4J-0QsL`px_CZ<%pwFcW_2(Pb_*hD%63{k)t|l>Z>0<0H&XXeYv^U#GCJPhgDOhr
zQ;feWwVLiipT;;-7bgd5v(cWU^P@;fdl+2}l&QMCL}6zvX_1#Xy|(U41;(b7W7Lx_
z5AH^vX6uvpflibn^p?sw?J3~7I@$ZF(E}?L(rn#|8Xx_Ua^HQIc3FOsUY>g|Wp8^W
zX@otMKEJG&^uiuU-aBd}qsj`YcGWGZv-?fy#qsM>i*<?A+@)BGmM=?u;F8qOs!(d9
zTp;yOyeOSB%a{84T#(eh<VhW)^CYX;dD13_JZZpk@pUOr3L1YwI;5X3sn}eUG;S0~
zF3&GX^YgDr`Ma-4SX?G)MpsDo8|tL)Z68UR#?Pcl8*WRkQkAsAyGk-Lt(N+y*GK{^
zC}nqjC@EZRkn%n}mCpFTlG?6&FD=S#mPXormpnTEk^YNTq@H=o^!K+a^_{Iw6Ygme
zFW07bMxAIwQx|GmYe-Fxd(aMLQyMqAFI`PBr%0p0WS1||$_>`E$!RzpnK*`idyJ=1
zsgp_Z-&FG6IE#wg&!d_vizqJ2i*&cHrWO7hsaw)k+UV&|5mDPI=XwzBR1cxIv-gl%
zK_pGGh@qv~`^kDz0^R$UM5{_uX-fJrs?In`RFy%Qdf7D6H<!+77Ena-Wl~AGMjz73
zsG_cdp3yyOTUbY*mOmx^v9IZseIw<s`9kfk{-nG_1<Y+!LZG86T&}kTi!S6dpE_XG
z(T<q7x(jRsUv;8G55&zg!Pq@!@DcO=_@S04Dzd_@1v2bRhM~UoXmo8e4h#BE!uyp|
zaPjU8WVp`3DiwEZD_@F1*(;HMVLj;aX4HiHVMk#gPW;-560s|Gixa$_-w~L<WG@V!
z#^SU40c3tThyh_Ku(LdhH;>bx5|@rJ*E8_$mxCfD(P{GXA_i6!!Rp3UINvPAqPyi7
z`>p~v+T2CPpjw=t-+;&iPw}GRCG5@K!NI2qtE)ew?~os;-TxcQ{1uq<1Z(WGGWQuM
zx^z6-@Z@9aJmEz<Zs?)KTU@m{Zl5kE-|oaemAkOI)RoEIke`HgXOFWzcu=J=AAD=V
z`oa^f*}5;kSMAUJ70r1<^FXe-H;7%b2J^}t|MAs{61Np^>EAr&6?3gw`f0<qLAJcP
z?Qs6$JA!lHjpX;^Mst4R7+(0wj?GQ%dAZ9t-tIS^8<Hk)bIwGTiye4O=_Ec_GMVf1
z9XbA_6MM%vb61}!Y&vTykC3KuC#~sh|8hF}=eh90&>0*yXC@~Mn#FyTXLHfL+1z~6
zm2Csv*u-rP>&tU_SJ!!b_uo98`FcJtsaYU2qlIi$;LaX-i?~hxV&*FztbJz*A9}i!
z8-IH8=g!Oci1l**<i3J0M6cwA>#Nwj)f)C5ww8ke)^V4c>-j~ejeKa)CjOi2&B<Du
z*<$Gy9#^uJM``;Co}w?mT<ym{jQx3NjX(SB2;enRAisPS$kF?@v-OxAobY)EuRFMt
z{q2L;uQ`YZ?c2qowUrZVcXPvrV6GAx*yv+I2jft_QyR(<Q^L6C!!TYhc7-b*?qRE`
z;oN*fFd(fX_}+;K-Y(eJ-hq+a<x?b&bdTbo=TW@cCz`j|?q$r1VX03nKU@&c?~%;o
zi;nT=_=l3-_iNI(zxh%?%y~(z)j3I_Fh{zKbJE|z=cPO0xzdy27o_{P7p0Fo3Z*$6
zFH62}i>3LkuS*e2Zc2^1x23c;Rg$g!J!x&tLuq!#W68Ddh17fYJ1O0=S!%!FyR`1(
zAE|kHD{7_Jn$|V9p_hN#(L!5oay`?DLZ@}5-)2VCuCED2IrpW1hXzuR?tdg?IQr!<
zOzg<V(5F}9$+yChMwLw$ybD)K__TmJbX!W+Ggnf{(e*S-bqih3_oF%g0%_}b!N^Ds
zro?|?bZ=24wY$5Qd?)NDm(l}tYJ3ufRHjfXSK&)|n?^4-rjxJQ8B&kUrZc9wly)|s
z%!ghgzpGcMaqKlJzIKCV48KLH*%dUPXAM~d-KR}2AJPVE(b>1_DJ>~~LDN;=&_}EH
zR5+`NT-JS}DFI(;()J(J<o%24L|^ffg9_-U+6t+QMIULAGGcUE<A$3m4jod%pN2Nb
z6ZhGiVH&vO-VQJPHBq&%J=Uda;h5;}vJ?Kp{`+(=$WIr8=68fLcS4|=9(I@O;Zsm&
z?By<Sc-jTOx9MZEPFJKK=!&&n4RCgk0UTQlAh{S~%1J}){c4DhgS%m-;BwsA(+v)%
zx}mnP8#Z3+hX0DY!6>I2CM1ZTZR&;_<GYD@rW@k#2>m<25buTWadMdf4CfnQM`Kr9
zS=AL9@AYxORUerpT@ci}3+DTFhHr%)dUh4B&9xJL#dbtaxh{VH(t&0#9rPQcji~t@
z5U@!LQgC~`*{3PI-R;mjMFZUrspI{DwkU{dBRn^1=(AoGdh=T&XPgT9Stuh;TM0cs
zDdOoZ1vEMSqZWlfq;ujYEt~v<ESkU4&d|>kW713kxs9~i{ypuec|*@!Us3wg=cKXl
zDVaWfM7>?>$)={3;>X>m^%rYM)x3&=58R;<ZEn%>wKvJRLF^6eO33B(75dYuh~{lA
zpn?|{XzavX8jzhs5r$b5wIh?peN3mTGf$F6$#Lpsd6d)-rc%DfA+q#2NFHww(9D@}
zq;_*3ZM5D?+3Ar~*(03Z?G2?~>cK+42@=}qc8XsUK#@(p^yA@XTHmmNHoRI*&%Q6C
zGg?dNt>r@6KW7fr?3+nBbyLZCpd;mnPoQ6lW2trENV=?NOMfmhE!kj2v7;>MkBK=w
z@7kBVO--p}bWb|9sT;Kv>C^OXdNe0YhY~tyQPfclDxTDa&bMhzsx_@BXZ}Cwm350W
zojyxVQ$9$gk*}lypPopeYwDz}diSK<h6<_sjWVgI_L}se!xf2E6-ZG(bEVN4+0w`8
zOliZ3Q_|CK$ECjBM<gXmk&-Qwq-~xFlI62F$zgx2<Q5Vmea(%Qs=Gx=M&}}=Q(MEO
zVXO8?k9UVj9j}K<olHZe#zP@euuX_`;cKuope$Hgc{y0pco-}dcL|X?28KwTdWK5X
z|3am{1NTUdrQuTF_mPrb(q0J@Vx_)?F;Ydu73tcntI|Px2g&0?wDj-i0qKA3o#|Un
z{ny5u4b2fHq%x;cwAWhSMp7y=lqAh*CX`T!<`79KMWHA}#+x*eG>K-7+-^!j$xKoF
z_VaH%hwIJtb|34--fK9|^V7QZ6a+0e4gKzCz-2}jC~iIn3UkiG<|Vn1b3G42J+H%R
zyIXJ}tN<Q7D*<DlGI%_#3Z9ERhOlvsptigfqOLrJWcHGOU%dvuhJJY2`589J48h%i
zKm79uVT_X)dNhr}C(e>MQ&<*b?~TXbNebu^HW^n$O~b{fr(=A>Z1kU~ffKiBV||w{
zj_|jC9OiXO*(P|_%^VZ7t#HBwYdkDthc{InF?WR<eopa1r{5b<aDM>Kft_d~9g1BC
z!f`C``6_6O#lV?yc!2MBl_w?Qr2EOdZZQoLH)UXYKqjiEWTQwY?{U|^!uLxzaJ5rD
z8cZt2h_9viyRQN(1RkO4f(FctY(eQSofs7H0%aDw##LhdSUK_u6_p0@x65xF=Pg7h
zZ;J9>OMV}eAVJ3bZ8f%L9GM=PNabEiq`qF6!oAe!V%#k9c&JX+%JX@j;zG)oC3<(+
zfSx!lrt4};=`PRLG3;AKd;4tZ*~ogTR&k|=Ydt9>-H&2FZ=qZ3w$qckUF0@5j23*0
zp!O<0^D=8c?TU&c-_3ka<a(6M?T?Yd`s3u{euAI>GN^q&?{&{PO978^sDj@Q1<uZ+
zxvtkJHG|h4zPL@|vkOTlsD##(m-2IHIjz}TMM70IWU5j}oAxwPLE97Z(rKr2aZl+~
z*K^v=b&+{Y4{0>MAvU3(BG-K2y$PR5uHYL*{X0l;KYo#=*dGcK5o9a(2{D^5!fckS
zC=0F-W4}~Kv(3R{*zt<7JQGlYtud2iIUzhF=A1NZuaIH4p7R{v0Xb&yeH>H$HlF46
z$uqgm2`sU6BLAINU=lkO+0~_!Sc$9>%V<$zcakQv@a0pOpTJbse|0JgcTi^8!qZsG
zg=s8dnF_P+S7E7fs_fYuHI`hZ#+WC6j1!*0M6zbE--~9l!`(BPWb7=~Hg7g#Ewfo<
z*c|p%b1tiS!837E)Y;CJ^Vm5F4c1kw!ImXyvX72h?AhG;Z2ZW4meQ=v?62tXoIalM
zvvna0u+wFpi#ax)=kcsjL-tjH7$1AY!X@?C4{?3gC1Jouqz&1h@kT5|$(Y5cEn>d&
z7PC*fCaltE35&NdW&PG>EYe{q`|NJcI(#hHlTDT^a`Q5_&wn}d@wH+dZYx;e+Lf%?
zY!y4LvzqNuv}W?ZteH}q4YSBw!(2mcna<L+ENiSC%dN6wo_p6ZB`tflw$q*shplHZ
z(;e8#DhH<O@5sz2JF%GmoY?9bXJ#Vg%F<iin6jWJYxUg7y03<?vPB12a$z30XU4&}
z)V&bewF6e#1cAbnZSd452%4{MhfAS5A+RI_;yw02hW|bg-53r5D<WZnTQpdmjfG3{
z2jS<PLqPFI;8;o`Y-vu04Q6RTpVHxV%Nf}7{T#^HUWCjqc`(p?1FV1KL$Y%TRF8Q8
z(+BxkZx-JlrL@33<7Y5RwHqGkzJswlKS9}xK~Mw%JRm88J)cHlWtTYW{3ngcug77@
z;6%JHKN&w5OvCE|(=p`wY|I{+hXHG~vA0MUpJ_1cy<~`<x+WOK>x4|bm*YXn)i|Zd
zmX~R)#~HI+P_BCenq~Un5m$e#R1U=Ye|O-CwB0B>Ck%BXBe1kH3iY*Ok-`q3S#2B+
z@t)E?`vlBSO~m2GW9TG$9K&_eaGd=~ydRu_#z#-1X;vm4zH}B_E@$J*?DM!I<sv%m
zy@YN~dAMr9Rs8wq8frbbfn|IC!I*is(W)sQ`~7)_@}ENd8C{IhLU(b=wtMK;QHrbi
zdqs8R13Xz(jwg5x(oyv)v|3h;C!8Phc_1}7#pMwWEPIRx)NAqHNG-aT)uBa5JuX*m
zz$GOOC}-J-$EzE$0Gn{bi6)GG--NO{%_!{CjCMzwQSL%B8r*2c-n?d<ezF;*cQs>$
zWixJ;Y)0*}CN%bK!djsw9CxG<S4cJDy|4z%dtHw%#`UNgUx%g*wWu(v7C-7dMwK;>
zP;g@n_U?R$>bt9P`t~YR_u|>kD=M&9vm8H(J-`EP_p#+nDbCz*4{N60MTsZHxG=H^
z56&yZRSkErYI8nnj=qg0>HpvrK2uxw&UK73;`u^lS5Rwl9=<HOgu+}d7Uo{SnUl|>
z*THN|{dN{xowAS~oWZ9vPUDm44AkpBiL(q(;Ej}2{NA5}5{r^iHi^$Z={|~n+6gGM
zHy+17JcQM9acJrAfB6To=<zlh?X>tTm7oZGR}h9pBcYfYy9=+$?nJ}%ZMc2bRt&wl
z34@G%aZ8;i?sjs=lkc6eZm$DM$*#k-=hk2buErO&RyfVW65V-S$DJe-)YC9VscL=n
z^dJo7b!}N$I;dr;g@@j$qh``<Of#B+&L34!K4&U=xhdgqRRv7w8;`9;vUu>A6ngFu
z$0_ck(QTb5w%QBf?7H7zE;s}~tiHg|Kkwl+_Xegv>4Jen|H2Blc2Hg23@VOw@I3q>
zG~O=<7o}3@I$Q*|6!Jl@=mxwAx&q5gb3uAu4hYT30(t#YVDEVXM$V;x)vrXz^Em>?
z-^an(gZp8>aWu@7i2$W<p&%_344h>kNHlH&j}zWtkn0XnBF<2G*&g0z+rraNtHJr`
z3V0E@47OF7L-#r}c(Y&$_}eUokJpXixrGr7$s0nhtN|=0eVBbv4?YO#fenw6Hr~Xb
z-i5)Ia0Ujg3`#upKu}K~wtE?XynqpC&t3%kM@(R^<5IYjXaR+<j3LPV7dO`NHP?Pj
z3f!g|!>9O_kTYZrb8Xgu$9G$3wOR+qY#pHWn={bF4G@&$4PM7K@$**zymH+ME9Qqn
z!s&2OsEC4_qxM6gR~*#5jE9@SiLgmM1;+eLgZx(+P}-LXQPMfEdsQxIUA+Pe=G}xz
z)%oBTSqyQ`_u-CxB@6~V0-0+K5IeRN&PP6l!;@YDSN9syGy0)2_A|VX8-#-we*<k5
z!tJS|*j76l|Hw#Sle;wTDwo3zFaZ~qC}N%66f_*8!u$C7_gTSgY$=|H5-;YX`3zl*
zjwg(pVu&5hi}Cc0rC3<F9B=fm#zJ#DEUR<GkiG8cZO7}u`JAB>5kXkqybC3A!Z80^
zB)(r4i?8Yr;?Vkdy!rDe?^jL6;azFC$|(aM*=1t%#%#<?yNCwedAQYr-!InQ##390
z&|bR~RmW9elza`oz<S&m`UK~7ci?Z||M0ME53cWchj&Xq;-|{*DA_lRk|~08s9c1?
z1V)p;g*ZJuB}J~hUg)I1Jn6kwpb_86lsl+Q3W;hohnqzeujbOz3{7%z)1jqW932?N
zNb#Kky=`4YJ+-Fvt<Hi5JFKW`U^VGTuBG1#*VAiHXUaJ3PSr2HNMPPZY7X5(_glA-
z0T)b{Pwt@wkzv$P7(qT?qbNu#mf|-aAVc2wA=Y!4JX8|sv11}#OF2fNbt!aGIE}XI
zo+Q@|8T9zjX`0A;U(eN_rS{%znlyBt+J$l{T;wt(jO0<o`>XV#`8pX~yGaWoZjqlw
zJ{=b=pq;l1snemDQoof@dgMLo6unROd&+2TFVAx^t03DGm895RMZzi%$$L!=oezCP
z8X1pi>&;pkD6J#S%6ck&&_Gn!NDFhDXw|W1(hhk-yX{-3O}CZzEwqtiPa8ECwA1H9
z9c1UyNoTa5lDo(=(rSN3iFyB$bTrS(bNP>q^k2}E$uG%7u!~;5?IP*6ZW4R=ibj|A
zkY({}8Z7B0ZQkcE)AW`yU%jLL;Xdl0^q!6z4p6)I2R>`*BW2foqAuYtq_FraNkxC7
z$j9Glfx;j)cn^_y(NBt#9j32-Bh*{@oA${GFz+$}_AOA5eNq!*(^`a>S&T570U|7{
zSA@wQ7iH&6#aRDmF?Q+HD0a$xG>iH;nkl7>VJbA1t?L-e!gq)>R~ZR*H%EebktCa4
zEy@11=CdR_q}X)_X;#o7&5~?n*w%+K?3$h|`;sfm^c3WndbAw#{mS!U{l>AvfpIMP
z$ap5~E6+p<Ca`By6q$U?R90!N$!;p!vyw}(uypQtxV%~bF8Kc7>|(!i4+p+*Z$iFu
z*~Q<u?fZXlx>ZBmBmZIU#Ew5)@D@RMWi1RxJw>7A{3tM+G#0LuNWhpRX^1%~2eIw)
zkh4q?E`6U2aZjg#&G2-X=r9MGhBUyaQwMI30Q~eY07V%SSpM4_UTCZY{d8OCGj)V}
zGdI9!<_ERA0zl?fFvMXP)JsLdm@l!=)Ex&}FOI<B-XzE$Iu2zM(qYK(3{(c5gOnQ=
zpz8N!NV2;QVkP{XsaXhpm+yit_W&GfszAW!G0c)`gp)-rFnLQSc+C2bpEtX~BJ&OG
zaD5L)l|Mt+zu!SS?H4r67Qi<VLg>>ef*qQp@X@|8Soc^Qcgjnli@gl0r1E>G#_@Pd
zWFj{5cQ}W2N+=XO1$QMVV{oPl7UZho@TD0jl+AlalIP&IP<0G+)IdiaEfo4QA9c!f
z@c!<F7^cplNh7b{-bARvOT)XO^zn;;A)muxgr@DrIBUUTOpP!>@5&{-zs(eF<}Sst
zmgZ>fWPyVlEb)=kGCXd%95v@!q06up{wZC7T{~Cei)pL)@8c@WG+&KsRjbhitg$cE
z8YTLzF?GHT9`Lfk=Z9_ZM2-zsU9-W!%Qkp}_t%~dwn2Lf8x)hU!MajweCchCL4wv;
zdU!Po^1WfmzE$Yovl3?+t;ElVR^a4%D;yNH!glTDC~LC}hkPyZ%?=A(xXT>5Aik&c
zG{XWbQ|#4Pf@)$W_@H$WPCR3bTitn$`cy;oXx7K12tE9$&ajyGz0hXhIUpSBr7uM7
zc?+;4UmInN=A(U?CYCSKz>pGkJgqwy?Q>^iwbCpUKR5%8zNulDlPbP0n}!ExC}VW=
z6ddSM!siB)aAS%BUhkiPHOBJToH!2qx@2*IwhY#XN?}j61ZK#I^Sq)ln3**S&EATl
zp_T~V4ie(|00NjfG6K=DKOsx@2k7$m+O4xb!k(M`u-*7A1lIMyUZ*ZN_5L|D?tKb!
zvh84h?g^a1M(D1sh3OtO;QOl*?k7Efr5dF$t-2UOJqjR0=r%mcx&iK{SE1(JB?w8n
z0AGx<q4HxU#OIuXk8UR+O)V7+`j0_!@ljZx91lx(#(~U+{or9A4R6**fOo@QI3*kc
zuB)~|V7@>2kPqawx<l|0CurHQ4$j)xK&tZ!u!yn%&GIF1d5SSiOVER!NdV3zIv}`H
z6C9S#1trayAgQhjs)kcxqxU43czyx|jL5+=KWS+FAP!#-jfR26qL3pe1P(t(xD%rW
zxrwVjbCzxWoJeLbXK=lX>lA;^dHmDC-Og*_3WplG`Dt}rWZWaJx~`gQbgAUlvvSVf
zzKpZYFXhTt+~c-Py319{mvG7k#a#E{BJQkM5jXXGA!oR?kW=yC|6>Zdil>F#noULA
z9Ft<McWViERPr9T0r`W;m~uX;p^Ce{wVJzAUc^}xL~vsh{kdDm61a{DC7hFc9p`Yr
zi3?x%gqu97mD}aq&WU<F<z$5Z<I+{SInS%ToL+iAx9!;{&dmD<XUvUoZdroRP$dEm
zV@Ja}Uvb#*N(z?kmIG1E36L|a2oAkdVAFsK=*!InBb&J(a6=R1v=)HaBLJ5eeTa5j
z1UDSbpnvBw7`(9xoFvzRcZ>rpoZ<%M4PGF7W)t7X1;V2EVCcBC7d!$Y;9^=dDAw->
zJGnz(=5YjMD-%J5Qb74$8uZy`z)|r`Fls#qtoQ;vzIPe0`x=~`eG4QK3xHF(3uoHO
zVE64R$h!L&PJL{I?B#86uIU-9jOYR_hd2E541j6OSMX~42~W=pV9XC;yuW@F-ftX>
zU5-+iH!O<{*X40qs3IPAnS!e8RdCRM27iN{gFQVOIDfegPHf<Cf1n=PEHuVB6HM`#
zf+c>{TZ#L^)}a1tdld6`;rHvFX!^_#lOJrw>koI}t@nG-_INl}RYdVVl>L}s8HZm_
z$Kw{kM9kWpjC#GP_`@$9&kUVL@uah;2<P!;56=lZbp<Wm`P_o}w=rOJAqw=BU@bof
zRn=8sbNxeX>a4|#4~-}&)ryt6op|2sIok4h2-~|}W7fPr^x69X&D*}<Ep8CUr4OT1
zqyYUY6rz26_E4DCD89ep@6Q*+Y2cM48S?w6R7W}5oHCx~)lDEhVMP+uRiZBUDP(j=
znRZ`Op`ki8n%X;qtOsY2kkDK@B|MML3~NwqzZP{iX_E$@FRK);OUulG0z`<G-(qxW
zy*_z=H6Y&zBf226i015GOd~xeWM*ngI%#G!{BkJ`C|i)DjU{F8Sw@h)oT6`7kxVI{
z?OCyse)C#_m;zp_dC{7RlWgd6@EQ`cx1|8xwRB9%jwbNA))@ussOFG88M>^eWNinY
zx8g`f?T&Oc&xs(~nQUELC{*8-l%}{*qM$n&zH_I|Z5zn%p$7$&dQyI|7x~=vrgIfO
z6xrlU@;!d!GP04JmHcUx(PlF9*+L^JTlxHz05TETM%xw#(T3>l)bw}<?NSIPX`c{U
zRJ4ok$nGJ<jiLN*Y%j&~Y*@dtF!~Z0PH)vBNUkM<bYdb&9HPjzH;O)_L{qS73>AEi
zp}<qI<YK;`3_tFt`s4#tj|WMh;~=T+h@%ndLu8x7=W*j<%C9<1zSi;dsXd;U!x382
zafH-t5@_th1hUsVO8?{@r5Or|v^F}C1ivNHKO2+C|5Fl)CLg0CLCF;IB!z7J(@3)V
mG)?<)hn(Z)FoD0nzZLjffxi{_TY<k7_*;R$75INvf&T${=?-iF

literal 0
HcmV?d00001


From 2f62b81f411975f7622d0b137e18470c96b31178 Mon Sep 17 00:00:00 2001
From: Kyle I S Harrington <kephale@users.noreply.github.com>
Date: Fri, 28 Jun 2024 20:14:14 -0400
Subject: [PATCH 078/183] Add alpha parameter to DiceFocalLoss (#7841)

Fixes https://github.com/Project-MONAI/MONAI/issues/7682.

### Description

This PR introduces the `alpha` parameter from `FocalLoss` into
`DiceFocalLoss`.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] New tests added to cover the changes.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Kyle Harrington <czi@kyleharrington.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/losses/dice.py          | 13 ++++++++++---
 tests/test_dice_focal_loss.py | 29 +++++++++++++++++++++++++++++
 2 files changed, 39 insertions(+), 3 deletions(-)

diff --git a/monai/losses/dice.py b/monai/losses/dice.py
index f1c357d31f..07a38d9572 100644
--- a/monai/losses/dice.py
+++ b/monai/losses/dice.py
@@ -811,7 +811,7 @@ class DiceFocalLoss(_Loss):
     The details of Focal Loss is shown in ``monai.losses.FocalLoss``.
 
     ``gamma`` and ``lambda_focal`` are only used for the focal loss.
-    ``include_background``, ``weight`` and ``reduction`` are used for both losses
+    ``include_background``, ``weight``, ``reduction``, and ``alpha`` are used for both losses,
     and other parameters are only used for dice loss.
 
     """
@@ -837,6 +837,7 @@ def __init__(
         weight: Sequence[float] | float | int | torch.Tensor | None = None,
         lambda_dice: float = 1.0,
         lambda_focal: float = 1.0,
+        alpha: float | None = None,
     ) -> None:
         """
         Args:
@@ -871,7 +872,8 @@ def __init__(
                 Defaults to 1.0.
             lambda_focal: the trade-off weight value for focal loss. The value should be no less than 0.0.
                 Defaults to 1.0.
-
+            alpha: value of the alpha in the definition of the alpha-balanced Focal loss. The value should be in
+                [0, 1]. Defaults to None.
         """
         super().__init__()
         weight = focal_weight if focal_weight is not None else weight
@@ -890,7 +892,12 @@ def __init__(
             weight=weight,
         )
         self.focal = FocalLoss(
-            include_background=include_background, to_onehot_y=False, gamma=gamma, weight=weight, reduction=reduction
+            include_background=include_background,
+            to_onehot_y=False,
+            gamma=gamma,
+            weight=weight,
+            alpha=alpha,
+            reduction=reduction,
         )
         if lambda_dice < 0.0:
             raise ValueError("lambda_dice should be no less than 0.0.")
diff --git a/tests/test_dice_focal_loss.py b/tests/test_dice_focal_loss.py
index 814a174762..f769aac69f 100644
--- a/tests/test_dice_focal_loss.py
+++ b/tests/test_dice_focal_loss.py
@@ -91,6 +91,35 @@ def test_script(self):
         test_input = torch.ones(2, 1, 8, 8)
         test_script_save(loss, test_input, test_input)
 
+    @parameterized.expand(
+        [
+            ("sum_None_0.5_0.25", "sum", None, 0.5, 0.25),
+            ("sum_weight_0.5_0.25", "sum", torch.tensor([1.0, 1.0, 2.0]), 0.5, 0.25),
+            ("sum_weight_tuple_0.5_0.25", "sum", (3, 2.0, 1), 0.5, 0.25),
+            ("mean_None_0.5_0.25", "mean", None, 0.5, 0.25),
+            ("mean_weight_0.5_0.25", "mean", torch.tensor([1.0, 1.0, 2.0]), 0.5, 0.25),
+            ("mean_weight_tuple_0.5_0.25", "mean", (3, 2.0, 1), 0.5, 0.25),
+            ("none_None_0.5_0.25", "none", None, 0.5, 0.25),
+            ("none_weight_0.5_0.25", "none", torch.tensor([1.0, 1.0, 2.0]), 0.5, 0.25),
+            ("none_weight_tuple_0.5_0.25", "none", (3, 2.0, 1), 0.5, 0.25),
+        ]
+    )
+    def test_with_alpha(self, name, reduction, weight, lambda_focal, alpha):
+        size = [3, 3, 5, 5]
+        label = torch.randint(low=0, high=2, size=size)
+        pred = torch.randn(size)
+
+        common_params = {"include_background": True, "to_onehot_y": False, "reduction": reduction, "weight": weight}
+
+        dice_focal = DiceFocalLoss(gamma=1.0, lambda_focal=lambda_focal, alpha=alpha, **common_params)
+        dice = DiceLoss(**common_params)
+        focal = FocalLoss(gamma=1.0, alpha=alpha, **common_params)
+
+        result = dice_focal(pred, label)
+        expected_val = dice(pred, label) + lambda_focal * focal(pred, label)
+
+        np.testing.assert_allclose(result, expected_val, err_msg=f"Failed on case: {name}")
+
 
 if __name__ == "__main__":
     unittest.main()

From 55386e0bb9415867bded3707c93bc9bc76f527cb Mon Sep 17 00:00:00 2001
From: Pengfei Guo <32000655+guopengf@users.noreply.github.com>
Date: Mon, 1 Jul 2024 09:39:40 -0400
Subject: [PATCH 079/183] Adding Tailored ControlNet Implementations into
 Generative Model Application (#7875)

Fixes #7874.

### Description

Integrating a tailored ControlNet model into the generative model
application to enable the training using high-dimensional 3D images (up
to 512 x 512 x 768).

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Pengfei Guo <pengfeig@nvidia.com>
Signed-off-by: alkamid <adam.klimont@cydar.co.uk>
Signed-off-by: Pengfei Guo <32000655+guopengf@users.noreply.github.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Adam Klimont <adamkli@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .pre-commit-config.yaml                       |   2 +-
 monai/apps/generation/__init__.py             |  10 +
 monai/apps/generation/maisi/__init__.py       |  10 +
 .../generation/maisi/networks/__init__.py     |  10 +
 .../maisi/networks/controlnet_maisi.py        | 178 ++++++++++++++++++
 monai/data/torchscript_utils.py               |   2 +-
 requirements-dev.txt                          |   1 +
 tests/test_controlnet_maisi.py                | 169 +++++++++++++++++
 8 files changed, 380 insertions(+), 2 deletions(-)
 create mode 100644 monai/apps/generation/__init__.py
 create mode 100644 monai/apps/generation/maisi/__init__.py
 create mode 100644 monai/apps/generation/maisi/networks/__init__.py
 create mode 100644 monai/apps/generation/maisi/networks/controlnet_maisi.py
 create mode 100644 tests/test_controlnet_maisi.py

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index b9debaf08f..3fff6ed631 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -9,7 +9,7 @@ ci:
 
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
+    rev: v4.6.0
     hooks:
       - id: end-of-file-fixer
       - id: trailing-whitespace
diff --git a/monai/apps/generation/__init__.py b/monai/apps/generation/__init__.py
new file mode 100644
index 0000000000..1e97f89407
--- /dev/null
+++ b/monai/apps/generation/__init__.py
@@ -0,0 +1,10 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
diff --git a/monai/apps/generation/maisi/__init__.py b/monai/apps/generation/maisi/__init__.py
new file mode 100644
index 0000000000..1e97f89407
--- /dev/null
+++ b/monai/apps/generation/maisi/__init__.py
@@ -0,0 +1,10 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
diff --git a/monai/apps/generation/maisi/networks/__init__.py b/monai/apps/generation/maisi/networks/__init__.py
new file mode 100644
index 0000000000..1e97f89407
--- /dev/null
+++ b/monai/apps/generation/maisi/networks/__init__.py
@@ -0,0 +1,10 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
diff --git a/monai/apps/generation/maisi/networks/controlnet_maisi.py b/monai/apps/generation/maisi/networks/controlnet_maisi.py
new file mode 100644
index 0000000000..3641124b7d
--- /dev/null
+++ b/monai/apps/generation/maisi/networks/controlnet_maisi.py
@@ -0,0 +1,178 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Sequence, cast
+
+import torch
+
+from monai.utils import optional_import
+
+ControlNet, has_controlnet = optional_import("generative.networks.nets.controlnet", name="ControlNet")
+get_timestep_embedding, has_get_timestep_embedding = optional_import(
+    "generative.networks.nets.diffusion_model_unet", name="get_timestep_embedding"
+)
+
+if TYPE_CHECKING:
+    from generative.networks.nets.controlnet import ControlNet as ControlNetType
+else:
+    ControlNetType = cast(type, ControlNet)
+
+
+class ControlNetMaisi(ControlNetType):
+    """
+    Control network for diffusion models based on Zhang and Agrawala "Adding Conditional Control to Text-to-Image
+    Diffusion Models" (https://arxiv.org/abs/2302.05543)
+
+    Args:
+        spatial_dims: number of spatial dimensions.
+        in_channels: number of input channels.
+        num_res_blocks: number of residual blocks (see ResnetBlock) per level.
+        num_channels: tuple of block output channels.
+        attention_levels: list of levels to add attention.
+        norm_num_groups: number of groups for the normalization.
+        norm_eps: epsilon for the normalization.
+        resblock_updown: if True use residual blocks for up/downsampling.
+        num_head_channels: number of channels in each attention head.
+        with_conditioning: if True add spatial transformers to perform conditioning.
+        transformer_num_layers: number of layers of Transformer blocks to use.
+        cross_attention_dim: number of context dimensions to use.
+        num_class_embeds: if specified (as an int), then this model will be class-conditional with `num_class_embeds`
+            classes.
+        upcast_attention: if True, upcast attention operations to full precision.
+        use_flash_attention: if True, use flash attention for a memory efficient attention mechanism.
+        conditioning_embedding_in_channels: number of input channels for the conditioning embedding.
+        conditioning_embedding_num_channels: number of channels for the blocks in the conditioning embedding.
+        use_checkpointing: if True, use activation checkpointing to save memory.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        num_res_blocks: Sequence[int] | int = (2, 2, 2, 2),
+        num_channels: Sequence[int] = (32, 64, 64, 64),
+        attention_levels: Sequence[bool] = (False, False, True, True),
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        resblock_updown: bool = False,
+        num_head_channels: int | Sequence[int] = 8,
+        with_conditioning: bool = False,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        num_class_embeds: int | None = None,
+        upcast_attention: bool = False,
+        use_flash_attention: bool = False,
+        conditioning_embedding_in_channels: int = 1,
+        conditioning_embedding_num_channels: Sequence[int] | None = (16, 32, 96, 256),
+        use_checkpointing: bool = True,
+    ) -> None:
+        super().__init__(
+            spatial_dims,
+            in_channels,
+            num_res_blocks,
+            num_channels,
+            attention_levels,
+            norm_num_groups,
+            norm_eps,
+            resblock_updown,
+            num_head_channels,
+            with_conditioning,
+            transformer_num_layers,
+            cross_attention_dim,
+            num_class_embeds,
+            upcast_attention,
+            use_flash_attention,
+            conditioning_embedding_in_channels,
+            conditioning_embedding_num_channels,
+        )
+        self.use_checkpointing = use_checkpointing
+
+    def forward(
+        self,
+        x: torch.Tensor,
+        timesteps: torch.Tensor,
+        controlnet_cond: torch.Tensor,
+        conditioning_scale: float = 1.0,
+        context: torch.Tensor | None = None,
+        class_labels: torch.Tensor | None = None,
+    ) -> tuple[Sequence[torch.Tensor], torch.Tensor]:
+        emb = self._prepare_time_and_class_embedding(x, timesteps, class_labels)
+        h = self._apply_initial_convolution(x)
+        if self.use_checkpointing:
+            controlnet_cond = torch.utils.checkpoint.checkpoint(
+                self.controlnet_cond_embedding, controlnet_cond, use_reentrant=False
+            )
+        else:
+            controlnet_cond = self.controlnet_cond_embedding(controlnet_cond)
+        h += controlnet_cond
+        down_block_res_samples, h = self._apply_down_blocks(emb, context, h)
+        h = self._apply_mid_block(emb, context, h)
+        down_block_res_samples, mid_block_res_sample = self._apply_controlnet_blocks(h, down_block_res_samples)
+        # scaling
+        down_block_res_samples = [h * conditioning_scale for h in down_block_res_samples]
+        mid_block_res_sample *= conditioning_scale
+
+        return down_block_res_samples, mid_block_res_sample
+
+    def _prepare_time_and_class_embedding(self, x, timesteps, class_labels):
+        # 1. time
+        t_emb = get_timestep_embedding(timesteps, self.block_out_channels[0])
+
+        # timesteps does not contain any weights and will always return f32 tensors
+        # but time_embedding might actually be running in fp16. so we need to cast here.
+        # there might be better ways to encapsulate this.
+        t_emb = t_emb.to(dtype=x.dtype)
+        emb = self.time_embed(t_emb)
+
+        # 2. class
+        if self.num_class_embeds is not None:
+            if class_labels is None:
+                raise ValueError("class_labels should be provided when num_class_embeds > 0")
+            class_emb = self.class_embedding(class_labels)
+            class_emb = class_emb.to(dtype=x.dtype)
+            emb = emb + class_emb
+
+        return emb
+
+    def _apply_initial_convolution(self, x):
+        # 3. initial convolution
+        h = self.conv_in(x)
+        return h
+
+    def _apply_down_blocks(self, emb, context, h):
+        # 4. down
+        if context is not None and self.with_conditioning is False:
+            raise ValueError("model should have with_conditioning = True if context is provided")
+        down_block_res_samples: list[torch.Tensor] = [h]
+        for downsample_block in self.down_blocks:
+            h, res_samples = downsample_block(hidden_states=h, temb=emb, context=context)
+            for residual in res_samples:
+                down_block_res_samples.append(residual)
+
+        return down_block_res_samples, h
+
+    def _apply_mid_block(self, emb, context, h):
+        # 5. mid
+        h = self.middle_block(hidden_states=h, temb=emb, context=context)
+        return h
+
+    def _apply_controlnet_blocks(self, h, down_block_res_samples):
+        # 6. Control net blocks
+        controlnet_down_block_res_samples = []
+        for down_block_res_sample, controlnet_block in zip(down_block_res_samples, self.controlnet_down_blocks):
+            down_block_res_sample = controlnet_block(down_block_res_sample)
+            controlnet_down_block_res_samples.append(down_block_res_sample)
+
+        mid_block_res_sample = self.controlnet_mid_block(h)
+
+        return controlnet_down_block_res_samples, mid_block_res_sample
diff --git a/monai/data/torchscript_utils.py b/monai/data/torchscript_utils.py
index cabf06ce89..507cf411d6 100644
--- a/monai/data/torchscript_utils.py
+++ b/monai/data/torchscript_utils.py
@@ -116,7 +116,7 @@ def load_net_with_metadata(
     Returns:
         Triple containing loaded object, metadata dict, and extra files dict containing other file data if present
     """
-    extra_files = {f: "" for f in more_extra_files}
+    extra_files = dict.fromkeys(more_extra_files, "")
     extra_files[METADATA_FILENAME] = ""
 
     jit_obj = torch.jit.load(filename_prefix_or_stream, map_location, extra_files)
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 517c842d1e..b598f301f6 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -58,3 +58,4 @@ lpips==0.1.4
 nvidia-ml-py
 huggingface_hub
 pyamg>=5.0.0
+git+https://github.com/KumoLiu/GenerativeModels.git@cuda#egg=monai-generative
diff --git a/tests/test_controlnet_maisi.py b/tests/test_controlnet_maisi.py
new file mode 100644
index 0000000000..b522b750c8
--- /dev/null
+++ b/tests/test_controlnet_maisi.py
@@ -0,0 +1,169 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.utils import optional_import
+from tests.utils import SkipIfBeforePyTorchVersion
+
+_, has_generative = optional_import("generative")
+
+if has_generative:
+    from monai.apps.generation.maisi.networks.controlnet_maisi import ControlNetMaisi
+
+TEST_CASES = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "conditioning_embedding_in_channels": 1,
+            "conditioning_embedding_num_channels": (8, 8),
+            "use_checkpointing": False,
+        },
+        6,
+        (1, 8, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "conditioning_embedding_in_channels": 1,
+            "conditioning_embedding_num_channels": (8, 8),
+            "use_checkpointing": True,
+        },
+        6,
+        (1, 8, 4, 4, 4),
+    ],
+]
+
+TEST_CASES_CONDITIONAL = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "conditioning_embedding_in_channels": 1,
+            "conditioning_embedding_num_channels": (8, 8),
+            "use_checkpointing": False,
+            "with_conditioning": True,
+            "cross_attention_dim": 2,
+        },
+        6,
+        (1, 8, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "conditioning_embedding_in_channels": 1,
+            "conditioning_embedding_num_channels": (8, 8),
+            "use_checkpointing": True,
+            "with_conditioning": True,
+            "cross_attention_dim": 2,
+        },
+        6,
+        (1, 8, 4, 4, 4),
+    ],
+]
+
+TEST_CASES_ERROR = [
+    [
+        {"spatial_dims": 2, "in_channels": 1, "with_conditioning": True, "cross_attention_dim": None},
+        "ControlNet expects dimension of the cross-attention conditioning "
+        "(cross_attention_dim) when using with_conditioning.",
+    ],
+    [
+        {"spatial_dims": 2, "in_channels": 1, "with_conditioning": False, "cross_attention_dim": 2},
+        "ControlNet expects with_conditioning=True when specifying the cross_attention_dim.",
+    ],
+    [
+        {"spatial_dims": 2, "in_channels": 1, "num_channels": (8, 16), "norm_num_groups": 16},
+        "ControlNet expects all num_channels being multiple of norm_num_groups",
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_channels": (8, 16),
+            "attention_levels": (True,),
+            "norm_num_groups": 8,
+        },
+        "ControlNet expects num_channels being same size of attention_levels",
+    ],
+]
+
+
+@SkipIfBeforePyTorchVersion((2, 0))
+@skipUnless(has_generative, "monai-generative required")
+class TestControlNet(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_shape_unconditioned_models(self, input_param, expected_num_down_blocks_residuals, expected_shape):
+        net = ControlNetMaisi(**input_param)
+        with eval_mode(net):
+            x = torch.rand((1, 1, 16, 16)) if input_param["spatial_dims"] == 2 else torch.rand((1, 1, 16, 16, 16))
+            timesteps = torch.randint(0, 1000, (1,)).long()
+            controlnet_cond = (
+                torch.rand((1, 1, 32, 32)) if input_param["spatial_dims"] == 2 else torch.rand((1, 1, 32, 32, 32))
+            )
+            result = net.forward(x, timesteps, controlnet_cond)
+            self.assertEqual(len(result[0]), expected_num_down_blocks_residuals)
+            self.assertEqual(result[1].shape, expected_shape)
+
+    @parameterized.expand(TEST_CASES_CONDITIONAL)
+    def test_shape_conditioned_models(self, input_param, expected_num_down_blocks_residuals, expected_shape):
+        net = ControlNetMaisi(**input_param)
+        with eval_mode(net):
+            x = torch.rand((1, 1, 16, 16)) if input_param["spatial_dims"] == 2 else torch.rand((1, 1, 16, 16, 16))
+            timesteps = torch.randint(0, 1000, (1,)).long()
+            controlnet_cond = (
+                torch.rand((1, 1, 32, 32)) if input_param["spatial_dims"] == 2 else torch.rand((1, 1, 32, 32, 32))
+            )
+            context = torch.randn((1, 1, input_param["cross_attention_dim"]))
+            result = net.forward(x, timesteps, controlnet_cond, context=context)
+            self.assertEqual(len(result[0]), expected_num_down_blocks_residuals)
+            self.assertEqual(result[1].shape, expected_shape)
+
+    @parameterized.expand(TEST_CASES_ERROR)
+    def test_error_input(self, input_param, expected_error):
+        with self.assertRaises(ValueError) as context:  # output shape too small
+            _ = ControlNetMaisi(**input_param)
+        runtime_error = context.exception
+        self.assertEqual(str(runtime_error), expected_error)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 2c7a26b74245a2665d37f29a72019a252ded678b Mon Sep 17 00:00:00 2001
From: Dong Yang <don.yang.mech@gmail.com>
Date: Mon, 1 Jul 2024 09:03:16 -0600
Subject: [PATCH 080/183] Integrating a Tailored Auto-Encoder Model into
 Generative Model Application (#7861)

Fixes #7858.

### Description

Integrating a tailored auto-encoder model into the generative model
application to improve the production of high-dimensional 3D images,
specifically sized at 512 x 512 x 512.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: dongyang0122 <don.yang.mech@gmail.com>
Signed-off-by: Dong Yang <yangd05@Dongs-MacBook-Pro.local>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Dong Yang <yangd05@Dongs-MacBook-Pro.local>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .../maisi/networks/autoencoderkl_maisi.py     | 975 ++++++++++++++++++
 tests/test_autoencoderkl_maisi.py             | 229 ++++
 2 files changed, 1204 insertions(+)
 create mode 100644 monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
 create mode 100644 tests/test_autoencoderkl_maisi.py

diff --git a/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py b/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
new file mode 100644
index 0000000000..533da32fa0
--- /dev/null
+++ b/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
@@ -0,0 +1,975 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import gc
+import logging
+from typing import TYPE_CHECKING, Sequence, cast
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+from monai.networks.blocks import Convolution
+from monai.utils import optional_import
+from monai.utils.type_conversion import convert_to_tensor
+
+AttentionBlock, has_attentionblock = optional_import("generative.networks.nets.autoencoderkl", name="AttentionBlock")
+AutoencoderKL, has_autoencoderkl = optional_import("generative.networks.nets.autoencoderkl", name="AutoencoderKL")
+ResBlock, has_resblock = optional_import("generative.networks.nets.autoencoderkl", name="ResBlock")
+
+
+if TYPE_CHECKING:
+    from generative.networks.nets.autoencoderkl import AutoencoderKL as AutoencoderKLType
+else:
+    AutoencoderKLType = cast(type, AutoencoderKL)
+
+
+# Set up logging configuration
+logger = logging.getLogger(__name__)
+
+
+def _empty_cuda_cache(save_mem: bool) -> None:
+    if torch.cuda.is_available() and save_mem:
+        torch.cuda.empty_cache()
+    return
+
+
+class MaisiGroupNorm3D(nn.GroupNorm):
+    """
+    Custom 3D Group Normalization with optional print_info output.
+
+    Args:
+        num_groups: Number of groups for the group norm.
+        num_channels: Number of channels for the group norm.
+        eps: Epsilon value for numerical stability.
+        affine: Whether to use learnable affine parameters, default to `True`.
+        norm_float16: If True, convert output of MaisiGroupNorm3D to float16 format, default to `False`.
+        print_info: Whether to print information, default to `False`.
+        save_mem: Whether to clean CUDA cache in order to save GPU memory, default to `True`.
+    """
+
+    def __init__(
+        self,
+        num_groups: int,
+        num_channels: int,
+        eps: float = 1e-5,
+        affine: bool = True,
+        norm_float16: bool = False,
+        print_info: bool = False,
+        save_mem: bool = True,
+    ):
+        super().__init__(num_groups, num_channels, eps, affine)
+        self.norm_float16 = norm_float16
+        self.print_info = print_info
+        self.save_mem = save_mem
+
+    def forward(self, input: torch.Tensor) -> torch.Tensor:
+        if self.print_info:
+            logger.info(f"MaisiGroupNorm3D with input size: {input.size()}")
+
+        if len(input.shape) != 5:
+            raise ValueError("Expected a 5D tensor")
+
+        param_n, param_c, param_d, param_h, param_w = input.shape
+        input = input.view(param_n, self.num_groups, param_c // self.num_groups, param_d, param_h, param_w)
+
+        inputs = []
+        for i in range(input.size(1)):
+            array = input[:, i : i + 1, ...].to(dtype=torch.float32)
+            mean = array.mean([2, 3, 4, 5], keepdim=True)
+            std = array.var([2, 3, 4, 5], unbiased=False, keepdim=True).add_(self.eps).sqrt_()
+            if self.norm_float16:
+                inputs.append(((array - mean) / std).to(dtype=torch.float16))
+            else:
+                inputs.append((array - mean) / std)
+
+        del input
+        _empty_cuda_cache(self.save_mem)
+
+        input = torch.cat(inputs, dim=1) if max(inputs[0].size()) < 500 else self._cat_inputs(inputs)
+
+        input = input.view(param_n, param_c, param_d, param_h, param_w)
+        if self.affine:
+            input.mul_(self.weight.view(1, param_c, 1, 1, 1)).add_(self.bias.view(1, param_c, 1, 1, 1))
+
+        if self.print_info:
+            logger.info(f"MaisiGroupNorm3D with output size: {input.size()}")
+
+        return input
+
+    def _cat_inputs(self, inputs):
+        input_type = inputs[0].device.type
+        input = inputs[0].clone().to("cpu", non_blocking=True) if input_type == "cuda" else inputs[0].clone()
+        inputs[0] = 0
+        _empty_cuda_cache(self.save_mem)
+
+        for k in range(len(inputs) - 1):
+            input = torch.cat((input, inputs[k + 1].cpu()), dim=1)
+            inputs[k + 1] = 0
+            _empty_cuda_cache(self.save_mem)
+            gc.collect()
+
+            if self.print_info:
+                logger.info(f"MaisiGroupNorm3D concat progress: {k + 1}/{len(inputs) - 1}.")
+
+        return input.to("cuda", non_blocking=True) if input_type == "cuda" else input
+
+
+class MaisiConvolution(nn.Module):
+    """
+    Convolutional layer with optional print_info output and custom splitting mechanism.
+
+    Args:
+        spatial_dims: Number of spatial dimensions (1D, 2D, 3D).
+        in_channels: Number of input channels.
+        out_channels: Number of output channels.
+        num_splits: Number of splits for the input tensor.
+        dim_split: Dimension of splitting for the input tensor.
+        print_info: Whether to print information.
+        save_mem: Whether to clean CUDA cache in order to save GPU memory, default to `True`.
+        Additional arguments for the convolution operation.
+        https://docs.monai.io/en/stable/networks.html#convolution
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        num_splits: int,
+        dim_split: int,
+        print_info: bool,
+        save_mem: bool = True,
+        strides: Sequence[int] | int = 1,
+        kernel_size: Sequence[int] | int = 3,
+        adn_ordering: str = "NDA",
+        act: tuple | str | None = "PRELU",
+        norm: tuple | str | None = "INSTANCE",
+        dropout: tuple | str | float | None = None,
+        dropout_dim: int = 1,
+        dilation: Sequence[int] | int = 1,
+        groups: int = 1,
+        bias: bool = True,
+        conv_only: bool = False,
+        is_transposed: bool = False,
+        padding: Sequence[int] | int | None = None,
+        output_padding: Sequence[int] | int | None = None,
+    ) -> None:
+        super().__init__()
+        self.conv = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            strides=strides,
+            kernel_size=kernel_size,
+            adn_ordering=adn_ordering,
+            act=act,
+            norm=norm,
+            dropout=dropout,
+            dropout_dim=dropout_dim,
+            dilation=dilation,
+            groups=groups,
+            bias=bias,
+            conv_only=conv_only,
+            is_transposed=is_transposed,
+            padding=padding,
+            output_padding=output_padding,
+        )
+
+        self.dim_split = dim_split
+        self.stride = strides[self.dim_split] if isinstance(strides, list) else strides
+        self.num_splits = num_splits
+        self.print_info = print_info
+        self.save_mem = save_mem
+
+    def _split_tensor(self, x: torch.Tensor, split_size: int, padding: int) -> list[torch.Tensor]:
+        overlaps = [0] + [padding] * (self.num_splits - 1)
+        last_padding = x.size(self.dim_split + 2) % split_size
+
+        slices = [slice(None)] * 5
+        splits: list[torch.Tensor] = []
+        for i in range(self.num_splits):
+            slices[self.dim_split + 2] = slice(
+                i * split_size - overlaps[i],
+                (i + 1) * split_size + (padding if i != self.num_splits - 1 else last_padding),
+            )
+            splits.append(x[tuple(slices)])
+
+        if self.print_info:
+            for j in range(len(splits)):
+                logger.info(f"Split {j + 1}/{len(splits)} size: {splits[j].size()}")
+
+        return splits
+
+    def _concatenate_tensors(self, outputs: list[torch.Tensor], split_size: int, padding: int) -> torch.Tensor:
+        slices = [slice(None)] * 5
+        for i in range(self.num_splits):
+            slices[self.dim_split + 2] = slice(None, split_size) if i == 0 else slice(padding, padding + split_size)
+            outputs[i] = outputs[i][tuple(slices)]
+
+        if self.print_info:
+            for i in range(self.num_splits):
+                logger.info(f"Output {i + 1}/{len(outputs)} size after: {outputs[i].size()}")
+
+        if max(outputs[0].size()) < 500:
+            x = torch.cat(outputs, dim=self.dim_split + 2)
+        else:
+            x = outputs[0].clone().to("cpu", non_blocking=True)
+            outputs[0] = torch.Tensor(0)
+            _empty_cuda_cache(self.save_mem)
+            for k in range(len(outputs) - 1):
+                x = torch.cat((x, outputs[k + 1].cpu()), dim=self.dim_split + 2)
+                outputs[k + 1] = torch.Tensor(0)
+                _empty_cuda_cache(self.save_mem)
+                gc.collect()
+                if self.print_info:
+                    logger.info(f"MaisiConvolution concat progress: {k + 1}/{len(outputs) - 1}.")
+
+            x = x.to("cuda", non_blocking=True)
+        return x
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if self.print_info:
+            logger.info(f"Number of splits: {self.num_splits}")
+
+        # compute size of splits
+        l = x.size(self.dim_split + 2)
+        split_size = l // self.num_splits
+
+        # update padding length if necessary
+        padding = 3
+        if padding % self.stride > 0:
+            padding = (padding // self.stride + 1) * self.stride
+        if self.print_info:
+            logger.info(f"Padding size: {padding}")
+
+        # split tensor into a list of tensors
+        splits = self._split_tensor(x, split_size, padding)
+
+        del x
+        _empty_cuda_cache(self.save_mem)
+
+        # convolution
+        outputs = [self.conv(split) for split in splits]
+        if self.print_info:
+            for j in range(len(outputs)):
+                logger.info(f"Output {j + 1}/{len(outputs)} size before: {outputs[j].size()}")
+
+        # update size of splits and padding length for output
+        split_size_out = split_size
+        padding_s = padding
+        non_dim_split = self.dim_split + 1 if self.dim_split < 2 else 0
+        if outputs[0].size(non_dim_split + 2) // splits[0].size(non_dim_split + 2) == 2:
+            split_size_out *= 2
+            padding_s *= 2
+        elif splits[0].size(non_dim_split + 2) // outputs[0].size(non_dim_split + 2) == 2:
+            split_size_out //= 2
+            padding_s //= 2
+
+        # concatenate list of tensors
+        x = self._concatenate_tensors(outputs, split_size_out, padding_s)
+
+        del outputs
+        _empty_cuda_cache(self.save_mem)
+
+        return x
+
+
+class MaisiUpsample(nn.Module):
+    """
+    Convolution-based upsampling layer.
+
+    Args:
+        spatial_dims: Number of spatial dimensions (1D, 2D, 3D).
+        in_channels: Number of input channels to the layer.
+        use_convtranspose: If True, use ConvTranspose to upsample feature maps in decoder.
+        num_splits: Number of splits for the input tensor.
+        dim_split: Dimension of splitting for the input tensor.
+        print_info: Whether to print information.
+        save_mem: Whether to clean CUDA cache in order to save GPU memory, default to `True`.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        use_convtranspose: bool,
+        num_splits: int,
+        dim_split: int,
+        print_info: bool,
+        save_mem: bool = True,
+    ) -> None:
+        super().__init__()
+        self.conv = MaisiConvolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=in_channels,
+            strides=2 if use_convtranspose else 1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+            is_transposed=use_convtranspose,
+            num_splits=num_splits,
+            dim_split=dim_split,
+            print_info=print_info,
+            save_mem=save_mem,
+        )
+        self.use_convtranspose = use_convtranspose
+        self.save_mem = save_mem
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        if self.use_convtranspose:
+            x = self.conv(x)
+            x_tensor: torch.Tensor = convert_to_tensor(x)
+            return x_tensor
+
+        x = F.interpolate(x, scale_factor=2.0, mode="trilinear")
+        _empty_cuda_cache(self.save_mem)
+        x = self.conv(x)
+        _empty_cuda_cache(self.save_mem)
+
+        out_tensor: torch.Tensor = convert_to_tensor(x)
+        return out_tensor
+
+
+class MaisiDownsample(nn.Module):
+    """
+    Convolution-based downsampling layer.
+
+    Args:
+        spatial_dims: Number of spatial dimensions (1D, 2D, 3D).
+        in_channels: Number of input channels.
+        num_splits: Number of splits for the input tensor.
+        dim_split: Dimension of splitting for the input tensor.
+        print_info: Whether to print information.
+        save_mem: Whether to clean CUDA cache in order to save GPU memory, default to `True`.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        num_splits: int,
+        dim_split: int,
+        print_info: bool,
+        save_mem: bool = True,
+    ) -> None:
+        super().__init__()
+        self.pad = (0, 1) * spatial_dims
+        self.conv = MaisiConvolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=in_channels,
+            strides=2,
+            kernel_size=3,
+            padding=0,
+            conv_only=True,
+            num_splits=num_splits,
+            dim_split=dim_split,
+            print_info=print_info,
+            save_mem=save_mem,
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x = F.pad(x, self.pad, mode="constant", value=0.0)
+        x = self.conv(x)
+        return x
+
+
+class MaisiResBlock(nn.Module):
+    """
+    Residual block consisting of a cascade of 2 convolutions + activation + normalisation block, and a
+    residual connection between input and output.
+
+    Args:
+        spatial_dims: Number of spatial dimensions (1D, 2D, 3D).
+        in_channels: Input channels to the layer.
+        norm_num_groups: Number of groups for the group norm layer.
+        norm_eps: Epsilon for the normalization.
+        out_channels: Number of output channels.
+        num_splits: Number of splits for the input tensor.
+        dim_split: Dimension of splitting for the input tensor.
+        norm_float16: If True, convert output of MaisiGroupNorm3D to float16 format, default to `False`.
+        print_info: Whether to print information, default to `False`.
+        save_mem: Whether to clean CUDA cache in order to save GPU memory, default to `True`.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        norm_num_groups: int,
+        norm_eps: float,
+        out_channels: int,
+        num_splits: int,
+        dim_split: int,
+        norm_float16: bool = False,
+        print_info: bool = False,
+        save_mem: bool = True,
+    ) -> None:
+        super().__init__()
+        self.in_channels = in_channels
+        self.out_channels = in_channels if out_channels is None else out_channels
+        self.save_mem = save_mem
+
+        self.norm1 = MaisiGroupNorm3D(
+            num_groups=norm_num_groups,
+            num_channels=in_channels,
+            eps=norm_eps,
+            affine=True,
+            norm_float16=norm_float16,
+            print_info=print_info,
+            save_mem=save_mem,
+        )
+        self.conv1 = MaisiConvolution(
+            spatial_dims=spatial_dims,
+            in_channels=self.in_channels,
+            out_channels=self.out_channels,
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+            num_splits=num_splits,
+            dim_split=dim_split,
+            print_info=print_info,
+            save_mem=save_mem,
+        )
+        self.norm2 = MaisiGroupNorm3D(
+            num_groups=norm_num_groups,
+            num_channels=out_channels,
+            eps=norm_eps,
+            affine=True,
+            norm_float16=norm_float16,
+            print_info=print_info,
+            save_mem=save_mem,
+        )
+        self.conv2 = MaisiConvolution(
+            spatial_dims=spatial_dims,
+            in_channels=self.out_channels,
+            out_channels=self.out_channels,
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+            num_splits=num_splits,
+            dim_split=dim_split,
+            print_info=print_info,
+            save_mem=save_mem,
+        )
+
+        self.nin_shortcut = (
+            MaisiConvolution(
+                spatial_dims=spatial_dims,
+                in_channels=self.in_channels,
+                out_channels=self.out_channels,
+                strides=1,
+                kernel_size=1,
+                padding=0,
+                conv_only=True,
+                num_splits=num_splits,
+                dim_split=dim_split,
+                print_info=print_info,
+                save_mem=save_mem,
+            )
+            if self.in_channels != self.out_channels
+            else nn.Identity()
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        h = self.norm1(x)
+        _empty_cuda_cache(self.save_mem)
+
+        h = F.silu(h)
+        _empty_cuda_cache(self.save_mem)
+        h = self.conv1(h)
+        _empty_cuda_cache(self.save_mem)
+
+        h = self.norm2(h)
+        _empty_cuda_cache(self.save_mem)
+
+        h = F.silu(h)
+        _empty_cuda_cache(self.save_mem)
+        h = self.conv2(h)
+        _empty_cuda_cache(self.save_mem)
+
+        if self.in_channels != self.out_channels:
+            x = self.nin_shortcut(x)
+            _empty_cuda_cache(self.save_mem)
+
+        out = x + h
+        out_tensor: torch.Tensor = convert_to_tensor(out)
+        return out_tensor
+
+
+class MaisiEncoder(nn.Module):
+    """
+    Convolutional cascade that downsamples the image into a spatial latent space.
+
+    Args:
+        spatial_dims: Number of spatial dimensions (1D, 2D, 3D).
+        in_channels: Number of input channels.
+        num_channels: Sequence of block output channels.
+        out_channels: Number of channels in the bottom layer (latent space) of the autoencoder.
+        num_res_blocks: Number of residual blocks (see ResBlock) per level.
+        norm_num_groups: Number of groups for the group norm layers.
+        norm_eps: Epsilon for the normalization.
+        attention_levels: Indicate which level from num_channels contain an attention block.
+        with_nonlocal_attn: If True, use non-local attention block.
+        use_flash_attention: If True, use flash attention for a memory efficient attention mechanism.
+        num_splits: Number of splits for the input tensor.
+        dim_split: Dimension of splitting for the input tensor.
+        norm_float16: If True, convert output of MaisiGroupNorm3D to float16 format, default to `False`.
+        print_info: Whether to print information, default to `False`.
+        save_mem: Whether to clean CUDA cache in order to save GPU memory, default to `True`.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        num_channels: Sequence[int],
+        out_channels: int,
+        num_res_blocks: Sequence[int],
+        norm_num_groups: int,
+        norm_eps: float,
+        attention_levels: Sequence[bool],
+        num_splits: int,
+        dim_split: int,
+        norm_float16: bool = False,
+        print_info: bool = False,
+        save_mem: bool = True,
+        with_nonlocal_attn: bool = True,
+        use_flash_attention: bool = False,
+    ) -> None:
+        super().__init__()
+
+        # Check if attention_levels and num_channels have the same size
+        if len(attention_levels) != len(num_channels):
+            raise ValueError("attention_levels and num_channels must have the same size")
+
+        # Check if num_res_blocks and num_channels have the same size
+        if len(num_res_blocks) != len(num_channels):
+            raise ValueError("num_res_blocks and num_channels must have the same size")
+
+        self.save_mem = save_mem
+
+        blocks: list[nn.Module] = []
+
+        blocks.append(
+            MaisiConvolution(
+                spatial_dims=spatial_dims,
+                in_channels=in_channels,
+                out_channels=num_channels[0],
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+                num_splits=num_splits,
+                dim_split=dim_split,
+                print_info=print_info,
+                save_mem=save_mem,
+            )
+        )
+
+        output_channel = num_channels[0]
+        for i in range(len(num_channels)):
+            input_channel = output_channel
+            output_channel = num_channels[i]
+            is_final_block = i == len(num_channels) - 1
+
+            for _ in range(num_res_blocks[i]):
+                blocks.append(
+                    MaisiResBlock(
+                        spatial_dims=spatial_dims,
+                        in_channels=input_channel,
+                        norm_num_groups=norm_num_groups,
+                        norm_eps=norm_eps,
+                        out_channels=output_channel,
+                        num_splits=num_splits,
+                        dim_split=dim_split,
+                        norm_float16=norm_float16,
+                        print_info=print_info,
+                        save_mem=save_mem,
+                    )
+                )
+                input_channel = output_channel
+                if attention_levels[i]:
+                    blocks.append(
+                        AttentionBlock(
+                            spatial_dims=spatial_dims,
+                            num_channels=input_channel,
+                            norm_num_groups=norm_num_groups,
+                            norm_eps=norm_eps,
+                            use_flash_attention=use_flash_attention,
+                        )
+                    )
+
+            if not is_final_block:
+                blocks.append(
+                    MaisiDownsample(
+                        spatial_dims=spatial_dims,
+                        in_channels=input_channel,
+                        num_splits=num_splits,
+                        dim_split=dim_split,
+                        print_info=print_info,
+                        save_mem=save_mem,
+                    )
+                )
+
+        if with_nonlocal_attn:
+            blocks.append(
+                ResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=num_channels[-1],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=num_channels[-1],
+                )
+            )
+
+            blocks.append(
+                AttentionBlock(
+                    spatial_dims=spatial_dims,
+                    num_channels=num_channels[-1],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    use_flash_attention=use_flash_attention,
+                )
+            )
+            blocks.append(
+                ResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=num_channels[-1],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=num_channels[-1],
+                )
+            )
+
+        blocks.append(
+            MaisiGroupNorm3D(
+                num_groups=norm_num_groups,
+                num_channels=num_channels[-1],
+                eps=norm_eps,
+                affine=True,
+                norm_float16=norm_float16,
+                print_info=print_info,
+                save_mem=save_mem,
+            )
+        )
+        blocks.append(
+            MaisiConvolution(
+                spatial_dims=spatial_dims,
+                in_channels=num_channels[-1],
+                out_channels=out_channels,
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+                num_splits=num_splits,
+                dim_split=dim_split,
+                print_info=print_info,
+                save_mem=save_mem,
+            )
+        )
+
+        self.blocks = nn.ModuleList(blocks)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        for block in self.blocks:
+            x = block(x)
+            _empty_cuda_cache(self.save_mem)
+        return x
+
+
+class MaisiDecoder(nn.Module):
+    """
+    Convolutional cascade upsampling from a spatial latent space into an image space.
+
+    Args:
+        spatial_dims: Number of spatial dimensions (1D, 2D, 3D).
+        num_channels: Sequence of block output channels.
+        in_channels: Number of channels in the bottom layer (latent space) of the autoencoder.
+        out_channels: Number of output channels.
+        num_res_blocks: Number of residual blocks (see ResBlock) per level.
+        norm_num_groups: Number of groups for the group norm layers.
+        norm_eps: Epsilon for the normalization.
+        attention_levels: Indicate which level from num_channels contain an attention block.
+        with_nonlocal_attn: If True, use non-local attention block.
+        use_flash_attention: If True, use flash attention for a memory efficient attention mechanism.
+        use_convtranspose: If True, use ConvTranspose to upsample feature maps in decoder.
+        num_splits: Number of splits for the input tensor.
+        dim_split: Dimension of splitting for the input tensor.
+        norm_float16: If True, convert output of MaisiGroupNorm3D to float16 format, default to `False`.
+        print_info: Whether to print information, default to `False`.
+        save_mem: Whether to clean CUDA cache in order to save GPU memory, default to `True`.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        num_channels: Sequence[int],
+        in_channels: int,
+        out_channels: int,
+        num_res_blocks: Sequence[int],
+        norm_num_groups: int,
+        norm_eps: float,
+        attention_levels: Sequence[bool],
+        num_splits: int,
+        dim_split: int,
+        norm_float16: bool = False,
+        print_info: bool = False,
+        save_mem: bool = True,
+        with_nonlocal_attn: bool = True,
+        use_flash_attention: bool = False,
+        use_convtranspose: bool = False,
+    ) -> None:
+        super().__init__()
+        self.print_info = print_info
+        self.save_mem = save_mem
+
+        reversed_block_out_channels = list(reversed(num_channels))
+
+        blocks: list[nn.Module] = []
+
+        blocks.append(
+            MaisiConvolution(
+                spatial_dims=spatial_dims,
+                in_channels=in_channels,
+                out_channels=reversed_block_out_channels[0],
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+                num_splits=num_splits,
+                dim_split=dim_split,
+                print_info=print_info,
+                save_mem=save_mem,
+            )
+        )
+
+        if with_nonlocal_attn:
+            blocks.append(
+                ResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=reversed_block_out_channels[0],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=reversed_block_out_channels[0],
+                )
+            )
+            blocks.append(
+                AttentionBlock(
+                    spatial_dims=spatial_dims,
+                    num_channels=reversed_block_out_channels[0],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    use_flash_attention=use_flash_attention,
+                )
+            )
+            blocks.append(
+                ResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=reversed_block_out_channels[0],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=reversed_block_out_channels[0],
+                )
+            )
+
+        reversed_attention_levels = list(reversed(attention_levels))
+        reversed_num_res_blocks = list(reversed(num_res_blocks))
+        block_out_ch = reversed_block_out_channels[0]
+        for i in range(len(reversed_block_out_channels)):
+            block_in_ch = block_out_ch
+            block_out_ch = reversed_block_out_channels[i]
+            is_final_block = i == len(num_channels) - 1
+
+            for _ in range(reversed_num_res_blocks[i]):
+                blocks.append(
+                    MaisiResBlock(
+                        spatial_dims=spatial_dims,
+                        in_channels=block_in_ch,
+                        norm_num_groups=norm_num_groups,
+                        norm_eps=norm_eps,
+                        out_channels=block_out_ch,
+                        num_splits=num_splits,
+                        dim_split=dim_split,
+                        norm_float16=norm_float16,
+                        print_info=print_info,
+                        save_mem=save_mem,
+                    )
+                )
+                block_in_ch = block_out_ch
+
+                if reversed_attention_levels[i]:
+                    blocks.append(
+                        AttentionBlock(
+                            spatial_dims=spatial_dims,
+                            num_channels=block_in_ch,
+                            norm_num_groups=norm_num_groups,
+                            norm_eps=norm_eps,
+                            use_flash_attention=use_flash_attention,
+                        )
+                    )
+
+            if not is_final_block:
+                blocks.append(
+                    MaisiUpsample(
+                        spatial_dims=spatial_dims,
+                        in_channels=block_in_ch,
+                        use_convtranspose=use_convtranspose,
+                        num_splits=num_splits,
+                        dim_split=dim_split,
+                        print_info=print_info,
+                        save_mem=save_mem,
+                    )
+                )
+
+        blocks.append(
+            MaisiGroupNorm3D(
+                num_groups=norm_num_groups,
+                num_channels=block_in_ch,
+                eps=norm_eps,
+                affine=True,
+                norm_float16=norm_float16,
+                print_info=print_info,
+                save_mem=save_mem,
+            )
+        )
+        blocks.append(
+            MaisiConvolution(
+                spatial_dims=spatial_dims,
+                in_channels=block_in_ch,
+                out_channels=out_channels,
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+                num_splits=num_splits,
+                dim_split=dim_split,
+                print_info=print_info,
+                save_mem=save_mem,
+            )
+        )
+
+        self.blocks = nn.ModuleList(blocks)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        for block in self.blocks:
+            x = block(x)
+            _empty_cuda_cache(self.save_mem)
+        return x
+
+
+class AutoencoderKlMaisi(AutoencoderKLType):
+    """
+    AutoencoderKL with custom MaisiEncoder and MaisiDecoder.
+
+    Args:
+        spatial_dims: Number of spatial dimensions (1D, 2D, 3D).
+        in_channels: Number of input channels.
+        out_channels: Number of output channels.
+        num_res_blocks: Number of residual blocks per level.
+        num_channels: Sequence of block output channels.
+        attention_levels: Indicate which level from num_channels contain an attention block.
+        latent_channels: Number of channels in the latent space.
+        norm_num_groups: Number of groups for the group norm layers.
+        norm_eps: Epsilon for the normalization.
+        with_encoder_nonlocal_attn: If True, use non-local attention block in the encoder.
+        with_decoder_nonlocal_attn: If True, use non-local attention block in the decoder.
+        use_flash_attention: If True, use flash attention for a memory efficient attention mechanism.
+        use_checkpointing: If True, use activation checkpointing.
+        use_convtranspose: If True, use ConvTranspose to upsample feature maps in decoder.
+        num_splits: Number of splits for the input tensor.
+        dim_split: Dimension of splitting for the input tensor.
+        norm_float16: If True, convert output of MaisiGroupNorm3D to float16 format, default to `False`.
+        print_info: Whether to print information, default to `False`.
+        save_mem: Whether to clean CUDA cache in order to save GPU memory, default to `True`.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        num_res_blocks: Sequence[int],
+        num_channels: Sequence[int],
+        attention_levels: Sequence[bool],
+        latent_channels: int = 3,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        with_encoder_nonlocal_attn: bool = False,
+        with_decoder_nonlocal_attn: bool = False,
+        use_flash_attention: bool = False,
+        use_checkpointing: bool = False,
+        use_convtranspose: bool = False,
+        num_splits: int = 16,
+        dim_split: int = 0,
+        norm_float16: bool = False,
+        print_info: bool = False,
+        save_mem: bool = True,
+    ) -> None:
+        super().__init__(
+            spatial_dims,
+            in_channels,
+            out_channels,
+            num_res_blocks,
+            num_channels,
+            attention_levels,
+            latent_channels,
+            norm_num_groups,
+            norm_eps,
+            with_encoder_nonlocal_attn,
+            with_decoder_nonlocal_attn,
+            use_flash_attention,
+            use_checkpointing,
+            use_convtranspose,
+        )
+
+        self.encoder = MaisiEncoder(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            num_channels=num_channels,
+            out_channels=latent_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            attention_levels=attention_levels,
+            with_nonlocal_attn=with_encoder_nonlocal_attn,
+            use_flash_attention=use_flash_attention,
+            num_splits=num_splits,
+            dim_split=dim_split,
+            norm_float16=norm_float16,
+            print_info=print_info,
+            save_mem=save_mem,
+        )
+
+        self.decoder = MaisiDecoder(
+            spatial_dims=spatial_dims,
+            num_channels=num_channels,
+            in_channels=latent_channels,
+            out_channels=out_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            attention_levels=attention_levels,
+            with_nonlocal_attn=with_decoder_nonlocal_attn,
+            use_flash_attention=use_flash_attention,
+            use_convtranspose=use_convtranspose,
+            num_splits=num_splits,
+            dim_split=dim_split,
+            norm_float16=norm_float16,
+            print_info=print_info,
+            save_mem=save_mem,
+        )
diff --git a/tests/test_autoencoderkl_maisi.py b/tests/test_autoencoderkl_maisi.py
new file mode 100644
index 0000000000..e88dc469c9
--- /dev/null
+++ b/tests/test_autoencoderkl_maisi.py
@@ -0,0 +1,229 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.utils import optional_import
+from tests.utils import SkipIfBeforePyTorchVersion
+
+tqdm, has_tqdm = optional_import("tqdm", name="tqdm")
+_, has_einops = optional_import("einops")
+_, has_generative = optional_import("generative")
+
+if has_generative:
+    from monai.apps.generation.maisi.networks.autoencoderkl_maisi import AutoencoderKlMaisi
+
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+
+
+CASES_NO_ATTENTION = [
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": (1, 1, 1),
+            "norm_num_groups": 4,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "num_splits": 2,
+            "print_info": False,
+        },
+        (1, 1, 32, 32, 32),
+        (1, 1, 32, 32, 32),
+        (1, 4, 8, 8, 8),
+    ]
+]
+
+CASES_ATTENTION = [
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, True),
+            "num_res_blocks": (1, 1, 1),
+            "norm_num_groups": 4,
+            "with_encoder_nonlocal_attn": True,
+            "with_decoder_nonlocal_attn": True,
+            "num_splits": 2,
+            "print_info": False,
+        },
+        (1, 1, 32, 32, 32),
+        (1, 1, 32, 32, 32),
+        (1, 4, 8, 8, 8),
+    ]
+]
+
+if has_einops:
+    CASES = CASES_NO_ATTENTION + CASES_ATTENTION
+else:
+    CASES = CASES_NO_ATTENTION
+
+
+@unittest.skipUnless(has_generative, "monai-generative required")
+class TestAutoencoderKlMaisi(unittest.TestCase):
+    @parameterized.expand(CASES)
+    def test_shape(self, input_param, input_shape, expected_shape, expected_latent_shape):
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.forward(torch.randn(input_shape).to(device))
+            self.assertEqual(result[0].shape, expected_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+            self.assertEqual(result[2].shape, expected_latent_shape)
+
+    @parameterized.expand(CASES)
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_with_convtranspose_and_checkpointing(
+        self, input_param, input_shape, expected_shape, expected_latent_shape
+    ):
+        input_param = input_param.copy()
+        input_param.update({"use_checkpointing": True, "use_convtranspose": True})
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.forward(torch.randn(input_shape).to(device))
+            self.assertEqual(result[0].shape, expected_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+            self.assertEqual(result[2].shape, expected_latent_shape)
+
+    def test_model_channels_not_multiple_of_norm_num_group(self):
+        with self.assertRaises(ValueError):
+            AutoencoderKlMaisi(
+                spatial_dims=3,
+                in_channels=1,
+                out_channels=1,
+                num_channels=(24, 24, 24),
+                attention_levels=(False, False, False),
+                latent_channels=8,
+                num_res_blocks=(1, 1, 1),
+                norm_num_groups=16,
+                num_splits=2,
+                print_info=False,
+            )
+
+    def test_model_num_channels_not_same_size_of_attention_levels(self):
+        with self.assertRaises(ValueError):
+            AutoencoderKlMaisi(
+                spatial_dims=3,
+                in_channels=1,
+                out_channels=1,
+                num_channels=(24, 24, 24),
+                attention_levels=(False, False),
+                latent_channels=8,
+                num_res_blocks=(1, 1, 1),
+                norm_num_groups=16,
+                num_splits=2,
+                print_info=False,
+            )
+
+    def test_model_num_channels_not_same_size_of_num_res_blocks(self):
+        with self.assertRaises(ValueError):
+            AutoencoderKlMaisi(
+                spatial_dims=3,
+                in_channels=1,
+                out_channels=1,
+                num_channels=(24, 24),
+                attention_levels=(False, False, False),
+                latent_channels=8,
+                num_res_blocks=(8, 8, 8),
+                norm_num_groups=16,
+                num_splits=2,
+                print_info=False,
+            )
+
+    def test_shape_reconstruction(self):
+        input_param, input_shape, expected_shape, _ = CASES[0]
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.reconstruct(torch.randn(input_shape).to(device))
+            self.assertEqual(result.shape, expected_shape)
+
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_reconstruction_with_convtranspose_and_checkpointing(self):
+        input_param, input_shape, expected_shape, _ = CASES[0]
+        input_param = input_param.copy()
+        input_param.update({"use_checkpointing": True, "use_convtranspose": True})
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.reconstruct(torch.randn(input_shape).to(device))
+            self.assertEqual(result.shape, expected_shape)
+
+    def test_shape_encode(self):
+        input_param, input_shape, _, expected_latent_shape = CASES[0]
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.encode(torch.randn(input_shape).to(device))
+            self.assertEqual(result[0].shape, expected_latent_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_encode_with_convtranspose_and_checkpointing(self):
+        input_param, input_shape, _, expected_latent_shape = CASES[0]
+        input_param = input_param.copy()
+        input_param.update({"use_checkpointing": True, "use_convtranspose": True})
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.encode(torch.randn(input_shape).to(device))
+            self.assertEqual(result[0].shape, expected_latent_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+
+    def test_shape_sampling(self):
+        input_param, _, _, expected_latent_shape = CASES[0]
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.sampling(
+                torch.randn(expected_latent_shape).to(device), torch.randn(expected_latent_shape).to(device)
+            )
+            self.assertEqual(result.shape, expected_latent_shape)
+
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_sampling_convtranspose_and_checkpointing(self):
+        input_param, _, _, expected_latent_shape = CASES[0]
+        input_param = input_param.copy()
+        input_param.update({"use_checkpointing": True, "use_convtranspose": True})
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.sampling(
+                torch.randn(expected_latent_shape).to(device), torch.randn(expected_latent_shape).to(device)
+            )
+            self.assertEqual(result.shape, expected_latent_shape)
+
+    def test_shape_decode(self):
+        input_param, expected_input_shape, _, latent_shape = CASES[0]
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.decode(torch.randn(latent_shape).to(device))
+            self.assertEqual(result.shape, expected_input_shape)
+
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_decode_convtranspose_and_checkpointing(self):
+        input_param, expected_input_shape, _, latent_shape = CASES[0]
+        input_param = input_param.copy()
+        input_param.update({"use_checkpointing": True, "use_convtranspose": True})
+        net = AutoencoderKlMaisi(**input_param).to(device)
+        with eval_mode(net):
+            result = net.decode(torch.randn(latent_shape).to(device))
+            self.assertEqual(result.shape, expected_input_shape)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 15d0771e8949a1658f0f0e79e71b5e0822668bde Mon Sep 17 00:00:00 2001
From: Dong Yang <don.yang.mech@gmail.com>
Date: Mon, 1 Jul 2024 20:44:35 -0600
Subject: [PATCH 081/183] Integrating a Tailored Diffusion U-Net Model into
 Generative Model Application (#7867)

Fixes #7858.

Integrating a tailored diffusion U-Net model into the generative model
application to improve the production of high-dimensional 3D images,
specifically sized at 512 x 512 x 512.

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: dongyang0122 <don.yang.mech@gmail.com>
Signed-off-by: Dong Yang <yangd05@Dongs-MacBook-Pro.local>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Dong Yang <yangd05@Dongs-MacBook-Pro.local>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .../networks/diffusion_model_unet_maisi.py    | 410 ++++++++++++
 requirements-dev.txt                          |   2 +-
 tests/test_diffusion_model_unet_maisi.py      | 592 ++++++++++++++++++
 3 files changed, 1003 insertions(+), 1 deletion(-)
 create mode 100644 monai/apps/generation/maisi/networks/diffusion_model_unet_maisi.py
 create mode 100644 tests/test_diffusion_model_unet_maisi.py

diff --git a/monai/apps/generation/maisi/networks/diffusion_model_unet_maisi.py b/monai/apps/generation/maisi/networks/diffusion_model_unet_maisi.py
new file mode 100644
index 0000000000..d5f5f6136b
--- /dev/null
+++ b/monai/apps/generation/maisi/networks/diffusion_model_unet_maisi.py
@@ -0,0 +1,410 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# =========================================================================
+# Adapted from https://github.com/huggingface/diffusers
+# which has the following license:
+# https://github.com/huggingface/diffusers/blob/main/LICENSE
+#
+# Copyright 2022 UC Berkeley Team and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =========================================================================
+
+from __future__ import annotations
+
+from collections.abc import Sequence
+
+import torch
+from torch import nn
+
+from monai.networks.blocks import Convolution
+from monai.utils import ensure_tuple_rep, optional_import
+from monai.utils.type_conversion import convert_to_tensor
+
+get_down_block, has_get_down_block = optional_import(
+    "generative.networks.nets.diffusion_model_unet", name="get_down_block"
+)
+get_mid_block, has_get_mid_block = optional_import(
+    "generative.networks.nets.diffusion_model_unet", name="get_mid_block"
+)
+get_timestep_embedding, has_get_timestep_embedding = optional_import(
+    "generative.networks.nets.diffusion_model_unet", name="get_timestep_embedding"
+)
+get_up_block, has_get_up_block = optional_import("generative.networks.nets.diffusion_model_unet", name="get_up_block")
+xformers, has_xformers = optional_import("xformers")
+zero_module, has_zero_module = optional_import("generative.networks.nets.diffusion_model_unet", name="zero_module")
+
+__all__ = ["DiffusionModelUNetMaisi"]
+
+
+class DiffusionModelUNetMaisi(nn.Module):
+    """
+    U-Net network with timestep embedding and attention mechanisms for conditioning based on
+    Rombach et al. "High-Resolution Image Synthesis with Latent Diffusion Models" https://arxiv.org/abs/2112.10752
+    and Pinaya et al. "Brain Imaging Generation with Latent Diffusion Models" https://arxiv.org/abs/2209.07162
+
+    Args:
+        spatial_dims: Number of spatial dimensions.
+        in_channels: Number of input channels.
+        out_channels: Number of output channels.
+        num_res_blocks: Number of residual blocks (see ResnetBlock) per level. Can be a single integer or a sequence of integers.
+        num_channels: Tuple of block output channels.
+        attention_levels: List of levels to add attention.
+        norm_num_groups: Number of groups for the normalization.
+        norm_eps: Epsilon for the normalization.
+        resblock_updown: If True, use residual blocks for up/downsampling.
+        num_head_channels: Number of channels in each attention head. Can be a single integer or a sequence of integers.
+        with_conditioning: If True, add spatial transformers to perform conditioning.
+        transformer_num_layers: Number of layers of Transformer blocks to use.
+        cross_attention_dim: Number of context dimensions to use.
+        num_class_embeds: If specified (as an int), then this model will be class-conditional with `num_class_embeds` classes.
+        upcast_attention: If True, upcast attention operations to full precision.
+        use_flash_attention: If True, use flash attention for a memory efficient attention mechanism.
+        dropout_cattn: If different from zero, this will be the dropout value for the cross-attention layers.
+        include_top_region_index_input: If True, use top region index input.
+        include_bottom_region_index_input: If True, use bottom region index input.
+        include_spacing_input: If True, use spacing input.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        num_res_blocks: Sequence[int] | int = (2, 2, 2, 2),
+        num_channels: Sequence[int] = (32, 64, 64, 64),
+        attention_levels: Sequence[bool] = (False, False, True, True),
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        resblock_updown: bool = False,
+        num_head_channels: int | Sequence[int] = 8,
+        with_conditioning: bool = False,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        num_class_embeds: int | None = None,
+        upcast_attention: bool = False,
+        use_flash_attention: bool = False,
+        dropout_cattn: float = 0.0,
+        include_top_region_index_input: bool = False,
+        include_bottom_region_index_input: bool = False,
+        include_spacing_input: bool = False,
+    ) -> None:
+        super().__init__()
+        if with_conditioning is True and cross_attention_dim is None:
+            raise ValueError(
+                "DiffusionModelUNetMaisi expects dimension of the cross-attention conditioning (cross_attention_dim) "
+                "when using with_conditioning."
+            )
+        if cross_attention_dim is not None and with_conditioning is False:
+            raise ValueError(
+                "DiffusionModelUNetMaisi expects with_conditioning=True when specifying the cross_attention_dim."
+            )
+        if dropout_cattn > 1.0 or dropout_cattn < 0.0:
+            raise ValueError("Dropout cannot be negative or >1.0!")
+
+        # All number of channels should be multiple of num_groups
+        if any((out_channel % norm_num_groups) != 0 for out_channel in num_channels):
+            raise ValueError(
+                f"DiffusionModelUNetMaisi expects all num_channels being multiple of norm_num_groups, "
+                f"but get num_channels: {num_channels} and norm_num_groups: {norm_num_groups}"
+            )
+
+        if len(num_channels) != len(attention_levels):
+            raise ValueError(
+                f"DiffusionModelUNetMaisi expects num_channels being same size of attention_levels, "
+                f"but get num_channels: {len(num_channels)} and attention_levels: {len(attention_levels)}"
+            )
+
+        if isinstance(num_head_channels, int):
+            num_head_channels = ensure_tuple_rep(num_head_channels, len(attention_levels))
+
+        if len(num_head_channels) != len(attention_levels):
+            raise ValueError(
+                "num_head_channels should have the same length as attention_levels. For the i levels without attention,"
+                " i.e. `attention_level[i]=False`, the num_head_channels[i] will be ignored."
+            )
+
+        if isinstance(num_res_blocks, int):
+            num_res_blocks = ensure_tuple_rep(num_res_blocks, len(num_channels))
+
+        if len(num_res_blocks) != len(num_channels):
+            raise ValueError(
+                "`num_res_blocks` should be a single integer or a tuple of integers with the same length as "
+                "`num_channels`."
+            )
+
+        if use_flash_attention and not has_xformers:
+            raise ValueError("use_flash_attention is True but xformers is not installed.")
+
+        if use_flash_attention is True and not torch.cuda.is_available():
+            raise ValueError(
+                "torch.cuda.is_available() should be True but is False. Flash attention is only available for GPU."
+            )
+
+        self.in_channels = in_channels
+        self.block_out_channels = num_channels
+        self.out_channels = out_channels
+        self.num_res_blocks = num_res_blocks
+        self.attention_levels = attention_levels
+        self.num_head_channels = num_head_channels
+        self.with_conditioning = with_conditioning
+
+        # input
+        self.conv_in = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=num_channels[0],
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+
+        # time
+        time_embed_dim = num_channels[0] * 4
+        self.time_embed = self._create_embedding_module(num_channels[0], time_embed_dim)
+
+        # class embedding
+        self.num_class_embeds = num_class_embeds
+        if num_class_embeds is not None:
+            self.class_embedding = nn.Embedding(num_class_embeds, time_embed_dim)
+
+        self.include_top_region_index_input = include_top_region_index_input
+        self.include_bottom_region_index_input = include_bottom_region_index_input
+        self.include_spacing_input = include_spacing_input
+
+        new_time_embed_dim = time_embed_dim
+        if self.include_top_region_index_input:
+            self.top_region_index_layer = self._create_embedding_module(4, time_embed_dim)
+            new_time_embed_dim += time_embed_dim
+        if self.include_bottom_region_index_input:
+            self.bottom_region_index_layer = self._create_embedding_module(4, time_embed_dim)
+            new_time_embed_dim += time_embed_dim
+        if self.include_spacing_input:
+            self.spacing_layer = self._create_embedding_module(3, time_embed_dim)
+            new_time_embed_dim += time_embed_dim
+
+        # down
+        self.down_blocks = nn.ModuleList([])
+        output_channel = num_channels[0]
+        for i in range(len(num_channels)):
+            input_channel = output_channel
+            output_channel = num_channels[i]
+            is_final_block = i == len(num_channels) - 1
+
+            down_block = get_down_block(
+                spatial_dims=spatial_dims,
+                in_channels=input_channel,
+                out_channels=output_channel,
+                temb_channels=new_time_embed_dim,
+                num_res_blocks=num_res_blocks[i],
+                norm_num_groups=norm_num_groups,
+                norm_eps=norm_eps,
+                add_downsample=not is_final_block,
+                resblock_updown=resblock_updown,
+                with_attn=(attention_levels[i] and not with_conditioning),
+                with_cross_attn=(attention_levels[i] and with_conditioning),
+                num_head_channels=num_head_channels[i],
+                transformer_num_layers=transformer_num_layers,
+                cross_attention_dim=cross_attention_dim,
+                upcast_attention=upcast_attention,
+                use_flash_attention=use_flash_attention,
+                dropout_cattn=dropout_cattn,
+            )
+
+            self.down_blocks.append(down_block)
+
+        # mid
+        self.middle_block = get_mid_block(
+            spatial_dims=spatial_dims,
+            in_channels=num_channels[-1],
+            temb_channels=new_time_embed_dim,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            with_conditioning=with_conditioning,
+            num_head_channels=num_head_channels[-1],
+            transformer_num_layers=transformer_num_layers,
+            cross_attention_dim=cross_attention_dim,
+            upcast_attention=upcast_attention,
+            use_flash_attention=use_flash_attention,
+            dropout_cattn=dropout_cattn,
+        )
+
+        # up
+        self.up_blocks = nn.ModuleList([])
+        reversed_block_out_channels = list(reversed(num_channels))
+        reversed_num_res_blocks = list(reversed(num_res_blocks))
+        reversed_attention_levels = list(reversed(attention_levels))
+        reversed_num_head_channels = list(reversed(num_head_channels))
+        output_channel = reversed_block_out_channels[0]
+        for i in range(len(reversed_block_out_channels)):
+            prev_output_channel = output_channel
+            output_channel = reversed_block_out_channels[i]
+            input_channel = reversed_block_out_channels[min(i + 1, len(num_channels) - 1)]
+
+            is_final_block = i == len(num_channels) - 1
+
+            up_block = get_up_block(
+                spatial_dims=spatial_dims,
+                in_channels=input_channel,
+                prev_output_channel=prev_output_channel,
+                out_channels=output_channel,
+                temb_channels=new_time_embed_dim,
+                num_res_blocks=reversed_num_res_blocks[i] + 1,
+                norm_num_groups=norm_num_groups,
+                norm_eps=norm_eps,
+                add_upsample=not is_final_block,
+                resblock_updown=resblock_updown,
+                with_attn=(reversed_attention_levels[i] and not with_conditioning),
+                with_cross_attn=(reversed_attention_levels[i] and with_conditioning),
+                num_head_channels=reversed_num_head_channels[i],
+                transformer_num_layers=transformer_num_layers,
+                cross_attention_dim=cross_attention_dim,
+                upcast_attention=upcast_attention,
+                use_flash_attention=use_flash_attention,
+                dropout_cattn=dropout_cattn,
+            )
+
+            self.up_blocks.append(up_block)
+
+        # out
+        self.out = nn.Sequential(
+            nn.GroupNorm(num_groups=norm_num_groups, num_channels=num_channels[0], eps=norm_eps, affine=True),
+            nn.SiLU(),
+            zero_module(
+                Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=num_channels[0],
+                    out_channels=out_channels,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+            ),
+        )
+
+    def _create_embedding_module(self, input_dim, embed_dim):
+        model = nn.Sequential(nn.Linear(input_dim, embed_dim), nn.SiLU(), nn.Linear(embed_dim, embed_dim))
+        return model
+
+    def _get_time_and_class_embedding(self, x, timesteps, class_labels):
+        t_emb = get_timestep_embedding(timesteps, self.block_out_channels[0])
+
+        # timesteps does not contain any weights and will always return f32 tensors
+        # but time_embedding might actually be running in fp16. so we need to cast here.
+        # there might be better ways to encapsulate this.
+        t_emb = t_emb.to(dtype=x.dtype)
+        emb = self.time_embed(t_emb)
+
+        if self.num_class_embeds is not None:
+            if class_labels is None:
+                raise ValueError("class_labels should be provided when num_class_embeds > 0")
+            class_emb = self.class_embedding(class_labels)
+            class_emb = class_emb.to(dtype=x.dtype)
+            emb += class_emb
+        return emb
+
+    def _get_input_embeddings(self, emb, top_index, bottom_index, spacing):
+        if self.include_top_region_index_input:
+            _emb = self.top_region_index_layer(top_index)
+            emb = torch.cat((emb, _emb), dim=1)
+        if self.include_bottom_region_index_input:
+            _emb = self.bottom_region_index_layer(bottom_index)
+            emb = torch.cat((emb, _emb), dim=1)
+        if self.include_spacing_input:
+            _emb = self.spacing_layer(spacing)
+            emb = torch.cat((emb, _emb), dim=1)
+        return emb
+
+    def _apply_down_blocks(self, h, emb, context, down_block_additional_residuals):
+        if context is not None and self.with_conditioning is False:
+            raise ValueError("model should have with_conditioning = True if context is provided")
+        down_block_res_samples: list[torch.Tensor] = [h]
+        for downsample_block in self.down_blocks:
+            h, res_samples = downsample_block(hidden_states=h, temb=emb, context=context)
+            down_block_res_samples.extend(res_samples)
+
+        # Additional residual conections for Controlnets
+        if down_block_additional_residuals is not None:
+            new_down_block_res_samples: list[torch.Tensor] = []
+            for down_block_res_sample, down_block_additional_residual in zip(
+                down_block_res_samples, down_block_additional_residuals
+            ):
+                down_block_res_sample += down_block_additional_residual
+                new_down_block_res_samples.append(down_block_res_sample)
+
+            down_block_res_samples = new_down_block_res_samples
+        return h, down_block_res_samples
+
+    def _apply_up_blocks(self, h, emb, context, down_block_res_samples):
+        for upsample_block in self.up_blocks:
+            res_samples = down_block_res_samples[-len(upsample_block.resnets) :]
+            down_block_res_samples = down_block_res_samples[: -len(upsample_block.resnets)]
+            h = upsample_block(hidden_states=h, res_hidden_states_list=res_samples, temb=emb, context=context)
+
+        return h
+
+    def forward(
+        self,
+        x: torch.Tensor,
+        timesteps: torch.Tensor,
+        context: torch.Tensor | None = None,
+        class_labels: torch.Tensor | None = None,
+        down_block_additional_residuals: tuple[torch.Tensor] | None = None,
+        mid_block_additional_residual: torch.Tensor | None = None,
+        top_region_index_tensor: torch.Tensor | None = None,
+        bottom_region_index_tensor: torch.Tensor | None = None,
+        spacing_tensor: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        """
+        Forward pass through the UNet model.
+
+        Args:
+            x: Input tensor of shape (N, C, SpatialDims).
+            timesteps: Timestep tensor of shape (N,).
+            context: Context tensor of shape (N, 1, ContextDim).
+            class_labels: Class labels tensor of shape (N,).
+            down_block_additional_residuals: Additional residual tensors for down blocks of shape (N, C, FeatureMapsDims).
+            mid_block_additional_residual: Additional residual tensor for mid block of shape (N, C, FeatureMapsDims).
+            top_region_index_tensor: Tensor representing top region index of shape (N, 4).
+            bottom_region_index_tensor: Tensor representing bottom region index of shape (N, 4).
+            spacing_tensor: Tensor representing spacing of shape (N, 3).
+
+        Returns:
+            A tensor representing the output of the UNet model.
+        """
+
+        emb = self._get_time_and_class_embedding(x, timesteps, class_labels)
+        emb = self._get_input_embeddings(emb, top_region_index_tensor, bottom_region_index_tensor, spacing_tensor)
+        h = self.conv_in(x)
+        h, _updated_down_block_res_samples = self._apply_down_blocks(h, emb, context, down_block_additional_residuals)
+        h = self.middle_block(h, emb, context)
+
+        # Additional residual conections for Controlnets
+        if mid_block_additional_residual is not None:
+            h += mid_block_additional_residual
+
+        h = self._apply_up_blocks(h, emb, context, _updated_down_block_res_samples)
+        h = self.out(h)
+        h_tensor: torch.Tensor = convert_to_tensor(h)
+        return h_tensor
diff --git a/requirements-dev.txt b/requirements-dev.txt
index b598f301f6..1bba930273 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -58,4 +58,4 @@ lpips==0.1.4
 nvidia-ml-py
 huggingface_hub
 pyamg>=5.0.0
-git+https://github.com/KumoLiu/GenerativeModels.git@cuda#egg=monai-generative
+git+https://github.com/Project-MONAI/GenerativeModels.git@7428fce193771e9564f29b91d29e523dd1b6b4cd
diff --git a/tests/test_diffusion_model_unet_maisi.py b/tests/test_diffusion_model_unet_maisi.py
new file mode 100644
index 0000000000..b5c14192d9
--- /dev/null
+++ b/tests/test_diffusion_model_unet_maisi.py
@@ -0,0 +1,592 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.utils import optional_import
+
+_, has_einops = optional_import("einops")
+_, has_generative = optional_import("generative")
+
+if has_generative:
+    from monai.apps.generation.maisi.networks.diffusion_model_unet_maisi import DiffusionModelUNetMaisi
+
+
+UNCOND_CASES_2D = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": (1, 1, 2),
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, True, True),
+            "num_head_channels": (0, 2, 4),
+            "norm_num_groups": 8,
+        }
+    ],
+]
+
+UNCOND_CASES_3D = [
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": (0, 0, 4),
+            "norm_num_groups": 8,
+        }
+    ],
+]
+
+COND_CASES_2D = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "upcast_attention": True,
+        }
+    ],
+]
+
+DROPOUT_OK = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "dropout_cattn": 0.25,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+        }
+    ],
+]
+
+DROPOUT_WRONG = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "num_channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "dropout_cattn": 3.0,
+        }
+    ]
+]
+
+
+@skipUnless(has_generative, "monai-generative required")
+class TestDiffusionModelUNetMaisi2D(unittest.TestCase):
+    @parameterized.expand(UNCOND_CASES_2D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_unconditioned_models(self, input_param):
+        net = DiffusionModelUNetMaisi(**input_param)
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, 1, 16, 16)), torch.randint(0, 1000, (1,)).long())
+            self.assertEqual(result.shape, (1, 1, 16, 16))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_timestep_with_wrong_shape(self):
+        net = DiffusionModelUNetMaisi(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            num_channels=(8, 8, 8),
+            attention_levels=(False, False, False),
+            norm_num_groups=8,
+        )
+        with self.assertRaises(ValueError):
+            with eval_mode(net):
+                net.forward(torch.rand((1, 1, 16, 16)), torch.randint(0, 1000, (1, 1)).long())
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_with_different_in_channel_out_channel(self):
+        in_channels = 6
+        out_channels = 3
+        net = DiffusionModelUNetMaisi(
+            spatial_dims=2,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            num_res_blocks=1,
+            num_channels=(8, 8, 8),
+            attention_levels=(False, False, False),
+            norm_num_groups=8,
+        )
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, in_channels, 16, 16)), torch.randint(0, 1000, (1,)).long())
+            self.assertEqual(result.shape, (1, out_channels, 16, 16))
+
+    def test_model_channels_not_multiple_of_norm_num_group(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNetMaisi(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                num_channels=(8, 8, 12),
+                attention_levels=(False, False, False),
+                norm_num_groups=8,
+            )
+
+    def test_attention_levels_with_different_length_num_head_channels(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNetMaisi(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                num_channels=(8, 8, 8),
+                attention_levels=(False, False, False),
+                num_head_channels=(0, 2),
+                norm_num_groups=8,
+            )
+
+    def test_num_res_blocks_with_different_length_channels(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNetMaisi(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=(1, 1),
+                num_channels=(8, 8, 8),
+                attention_levels=(False, False, False),
+                norm_num_groups=8,
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models(self):
+        net = DiffusionModelUNetMaisi(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            num_channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            with_conditioning=True,
+            transformer_num_layers=1,
+            cross_attention_dim=3,
+            norm_num_groups=8,
+            num_head_channels=8,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 32)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                context=torch.rand((1, 1, 3)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 32))
+
+    def test_with_conditioning_cross_attention_dim_none(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNetMaisi(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                num_channels=(8, 8, 8),
+                attention_levels=(False, False, True),
+                with_conditioning=True,
+                transformer_num_layers=1,
+                cross_attention_dim=None,
+                norm_num_groups=8,
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_context_with_conditioning_none(self):
+        net = DiffusionModelUNetMaisi(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            num_channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            with_conditioning=False,
+            transformer_num_layers=1,
+            norm_num_groups=8,
+        )
+
+        with self.assertRaises(ValueError):
+            with eval_mode(net):
+                net.forward(
+                    x=torch.rand((1, 1, 16, 32)),
+                    timesteps=torch.randint(0, 1000, (1,)).long(),
+                    context=torch.rand((1, 1, 3)),
+                )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models_class_conditioning(self):
+        net = DiffusionModelUNetMaisi(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            num_channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            norm_num_groups=8,
+            num_head_channels=8,
+            num_class_embeds=2,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 32)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                class_labels=torch.randint(0, 2, (1,)).long(),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 32))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_conditioned_models_no_class_labels(self):
+        net = DiffusionModelUNetMaisi(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            num_channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            norm_num_groups=8,
+            num_head_channels=8,
+            num_class_embeds=2,
+        )
+
+        with self.assertRaises(ValueError):
+            net.forward(x=torch.rand((1, 1, 16, 32)), timesteps=torch.randint(0, 1000, (1,)).long())
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_model_channels_not_same_size_of_attention_levels(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNetMaisi(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                num_channels=(8, 8, 8),
+                attention_levels=(False, False),
+                norm_num_groups=8,
+                num_head_channels=8,
+                num_class_embeds=2,
+            )
+
+    @parameterized.expand(COND_CASES_2D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_conditioned_2d_models_shape(self, input_param):
+        net = DiffusionModelUNetMaisi(**input_param)
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, 1, 16, 16)), torch.randint(0, 1000, (1,)).long(), torch.rand((1, 1, 3)))
+            self.assertEqual(result.shape, (1, 1, 16, 16))
+
+    @parameterized.expand(UNCOND_CASES_2D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_with_additional_inputs(self, input_param):
+        input_param["include_top_region_index_input"] = True
+        input_param["include_bottom_region_index_input"] = True
+        input_param["include_spacing_input"] = True
+        net = DiffusionModelUNetMaisi(**input_param)
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 16)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                top_region_index_tensor=torch.rand((1, 4)),
+                bottom_region_index_tensor=torch.rand((1, 4)),
+                spacing_tensor=torch.rand((1, 3)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 16))
+
+
+@skipUnless(has_generative, "monai-generative required")
+class TestDiffusionModelUNetMaisi3D(unittest.TestCase):
+    @parameterized.expand(UNCOND_CASES_3D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_unconditioned_models(self, input_param):
+        net = DiffusionModelUNetMaisi(**input_param)
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, 1, 16, 16, 16)), torch.randint(0, 1000, (1,)).long())
+            self.assertEqual(result.shape, (1, 1, 16, 16, 16))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_with_different_in_channel_out_channel(self):
+        in_channels = 6
+        out_channels = 3
+        net = DiffusionModelUNetMaisi(
+            spatial_dims=3,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            num_res_blocks=1,
+            num_channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            norm_num_groups=4,
+        )
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, in_channels, 16, 16, 16)), torch.randint(0, 1000, (1,)).long())
+            self.assertEqual(result.shape, (1, out_channels, 16, 16, 16))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models(self):
+        net = DiffusionModelUNetMaisi(
+            spatial_dims=3,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            num_channels=(16, 16, 16),
+            attention_levels=(False, False, True),
+            norm_num_groups=16,
+            with_conditioning=True,
+            transformer_num_layers=1,
+            cross_attention_dim=3,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 16, 16)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                context=torch.rand((1, 1, 3)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 16, 16))
+
+    # Test dropout specification for cross-attention blocks
+    @parameterized.expand(DROPOUT_WRONG)
+    def test_wrong_dropout(self, input_param):
+        with self.assertRaises(ValueError):
+            _ = DiffusionModelUNetMaisi(**input_param)
+
+    @parameterized.expand(DROPOUT_OK)
+    @skipUnless(has_einops, "Requires einops")
+    def test_right_dropout(self, input_param):
+        _ = DiffusionModelUNetMaisi(**input_param)
+
+    @parameterized.expand(UNCOND_CASES_3D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_with_additional_inputs(self, input_param):
+        input_param["include_top_region_index_input"] = True
+        input_param["include_bottom_region_index_input"] = True
+        input_param["include_spacing_input"] = True
+        net = DiffusionModelUNetMaisi(**input_param)
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 16, 16)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                top_region_index_tensor=torch.rand((1, 4)),
+                bottom_region_index_tensor=torch.rand((1, 4)),
+                spacing_tensor=torch.rand((1, 3)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 16, 16))
+
+
+if __name__ == "__main__":
+    unittest.main()

From 410109a8dab9b244e79697ee89e13c8b044da6dd Mon Sep 17 00:00:00 2001
From: Can Zhao <69829124+Can-Zhao@users.noreply.github.com>
Date: Tue, 2 Jul 2024 06:50:41 -0700
Subject: [PATCH 082/183] Maisi morphological funcs (#7893)

Fixes # .

### Description

Maisi morphological funcs

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Can-Zhao <canz@nvidia.com>
Signed-off-by: Can Zhao <69829124+Can-Zhao@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 docs/source/apps.rst                          |   8 +
 monai/apps/generation/maisi/utils/__init__.py |  10 ++
 .../maisi/utils/morphological_ops.py          | 170 ++++++++++++++++++
 tests/test_morphological_ops.py               | 102 +++++++++++
 4 files changed, 290 insertions(+)
 create mode 100644 monai/apps/generation/maisi/utils/__init__.py
 create mode 100644 monai/apps/generation/maisi/utils/morphological_ops.py
 create mode 100644 tests/test_morphological_ops.py

diff --git a/docs/source/apps.rst b/docs/source/apps.rst
index 7fa7b9e9ff..c6ba8c0b9a 100644
--- a/docs/source/apps.rst
+++ b/docs/source/apps.rst
@@ -261,3 +261,11 @@ FastMRIReader
 
 .. autoclass:: monai.apps.nnunet.nnUNetV2Runner
   :members:
+
+`Generative AI`
+---------------
+
+`MAISI Utilities`
+~~~~~~~~~~~~~~~~~
+.. automodule:: monai.apps.generation.maisi.utils.morphological_ops
+    :members:
diff --git a/monai/apps/generation/maisi/utils/__init__.py b/monai/apps/generation/maisi/utils/__init__.py
new file mode 100644
index 0000000000..1e97f89407
--- /dev/null
+++ b/monai/apps/generation/maisi/utils/__init__.py
@@ -0,0 +1,10 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
diff --git a/monai/apps/generation/maisi/utils/morphological_ops.py b/monai/apps/generation/maisi/utils/morphological_ops.py
new file mode 100644
index 0000000000..14786d60a2
--- /dev/null
+++ b/monai/apps/generation/maisi/utils/morphological_ops.py
@@ -0,0 +1,170 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import Sequence
+
+import torch
+import torch.nn.functional as F
+from torch import Tensor
+
+from monai.config import NdarrayOrTensor
+from monai.utils import convert_data_type, convert_to_dst_type, ensure_tuple_rep
+
+
+def erode(mask: NdarrayOrTensor, filter_size: int | Sequence[int] = 3, pad_value: float = 1.0) -> NdarrayOrTensor:
+    """
+    Erode 2D/3D binary mask.
+
+    Args:
+        mask: input 2D/3D binary mask, [N,C,M,N] or [N,C,M,N,P] torch tensor or ndarray.
+        filter_size: erosion filter size, has to be odd numbers, default to be 3.
+        pad_value: the filled value for padding. We need to pad the input before filtering
+                   to keep the output with the same size as input. Usually use default value
+                   and not changed.
+
+    Return:
+        eroded mask, same shape and data type as input.
+
+    Example:
+
+        .. code-block:: python
+
+            # define a naive mask
+            mask = torch.zeros(3,2,3,3,3)
+            mask[:,:,1,1,1] = 1.0
+            filter_size = 3
+            erode_result = erode(mask, filter_size)  # expect torch.zeros(3,2,3,3,3)
+            dilate_result = dilate(mask, filter_size)  # expect torch.ones(3,2,3,3,3)
+    """
+    mask_t, *_ = convert_data_type(mask, torch.Tensor)
+    res_mask_t = erode_t(mask_t, filter_size=filter_size, pad_value=pad_value)
+    res_mask: NdarrayOrTensor
+    res_mask, *_ = convert_to_dst_type(src=res_mask_t, dst=mask)
+    return res_mask
+
+
+def dilate(mask: NdarrayOrTensor, filter_size: int | Sequence[int] = 3, pad_value: float = 0.0) -> NdarrayOrTensor:
+    """
+    Dilate 2D/3D binary mask.
+
+    Args:
+        mask: input 2D/3D binary mask, [N,C,M,N] or [N,C,M,N,P] torch tensor or ndarray.
+        filter_size: dilation filter size, has to be odd numbers, default to be 3.
+        pad_value: the filled value for padding. We need to pad the input before filtering
+                   to keep the output with the same size as input. Usually use default value
+                   and not changed.
+
+    Return:
+        dilated mask, same shape and data type as input.
+
+    Example:
+
+        .. code-block:: python
+
+            # define a naive mask
+            mask = torch.zeros(3,2,3,3,3)
+            mask[:,:,1,1,1] = 1.0
+            filter_size = 3
+            erode_result = erode(mask,filter_size) # expect torch.zeros(3,2,3,3,3)
+            dilate_result = dilate(mask,filter_size) # expect torch.ones(3,2,3,3,3)
+    """
+    mask_t, *_ = convert_data_type(mask, torch.Tensor)
+    res_mask_t = dilate_t(mask_t, filter_size=filter_size, pad_value=pad_value)
+    res_mask: NdarrayOrTensor
+    res_mask, *_ = convert_to_dst_type(src=res_mask_t, dst=mask)
+    return res_mask
+
+
+def get_morphological_filter_result_t(mask_t: Tensor, filter_size: int | Sequence[int], pad_value: float) -> Tensor:
+    """
+    Apply a morphological filter to a 2D/3D binary mask tensor.
+
+    Args:
+        mask_t: input 2D/3D binary mask, [N,C,M,N] or [N,C,M,N,P] torch tensor.
+        filter_size: morphological filter size, has to be odd numbers.
+        pad_value: the filled value for padding. We need to pad the input before filtering
+                   to keep the output with the same size as input.
+
+    Return:
+        Tensor: Morphological filter result mask, same shape as input.
+    """
+    spatial_dims = len(mask_t.shape) - 2
+    if spatial_dims not in [2, 3]:
+        raise ValueError(
+            f"spatial_dims must be either 2 or 3, "
+            f"got spatial_dims={spatial_dims} for mask tensor with shape of {mask_t.shape}."
+        )
+
+    # Define the structuring element
+    filter_size = ensure_tuple_rep(filter_size, spatial_dims)
+    if any(size % 2 == 0 for size in filter_size):
+        raise ValueError(f"All dimensions in filter_size must be odd numbers, got {filter_size}.")
+
+    structuring_element = torch.ones((mask_t.shape[1], mask_t.shape[1]) + filter_size).to(mask_t.device)
+
+    # Pad the input tensor to handle border pixels
+    # Calculate padding size
+    pad_size = [size // 2 for size in filter_size for _ in range(2)]
+
+    input_padded = F.pad(mask_t.float(), pad_size, mode="constant", value=pad_value)
+
+    # Apply filter operation
+    conv_fn = F.conv2d if spatial_dims == 2 else F.conv3d
+    output = conv_fn(input_padded, structuring_element, padding=0) / torch.sum(structuring_element[0, ...])
+
+    return output
+
+
+def erode_t(mask_t: Tensor, filter_size: int | Sequence[int] = 3, pad_value: float = 1.0) -> Tensor:
+    """
+    Erode 2D/3D binary mask with data type as torch tensor.
+
+    Args:
+        mask_t: input 2D/3D binary mask, [N,C,M,N] or [N,C,M,N,P] torch tensor.
+        filter_size: erosion filter size, has to be odd numbers, default to be 3.
+        pad_value: the filled value for padding. We need to pad the input before filtering
+                   to keep the output with the same size as input. Usually use default value
+                   and not changed.
+
+    Return:
+        Tensor: eroded mask, same shape as input.
+    """
+
+    output = get_morphological_filter_result_t(mask_t, filter_size, pad_value)
+
+    # Set output values based on the minimum value within the structuring element
+    output = torch.where(torch.abs(output - 1.0) < 1e-7, 1.0, 0.0)
+
+    return output
+
+
+def dilate_t(mask_t: Tensor, filter_size: int | Sequence[int] = 3, pad_value: float = 0.0) -> Tensor:
+    """
+    Dilate 2D/3D binary mask with data type as torch tensor.
+
+    Args:
+        mask_t: input 2D/3D binary mask, [N,C,M,N] or [N,C,M,N,P] torch tensor.
+        filter_size: dilation filter size, has to be odd numbers, default to be 3.
+        pad_value: the filled value for padding. We need to pad the input before filtering
+                   to keep the output with the same size as input. Usually use default value
+                   and not changed.
+
+    Return:
+        Tensor: dilated mask, same shape as input.
+    """
+    output = get_morphological_filter_result_t(mask_t, filter_size, pad_value)
+
+    # Set output values based on the minimum value within the structuring element
+    output = torch.where(output > 0, 1.0, 0.0)
+
+    return output
diff --git a/tests/test_morphological_ops.py b/tests/test_morphological_ops.py
new file mode 100644
index 0000000000..6f29415759
--- /dev/null
+++ b/tests/test_morphological_ops.py
@@ -0,0 +1,102 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.apps.generation.maisi.utils.morphological_ops import dilate, erode, get_morphological_filter_result_t
+from tests.utils import TEST_NDARRAYS, assert_allclose
+
+TESTS_SHAPE = []
+for p in TEST_NDARRAYS:
+    mask = torch.zeros(1, 1, 5, 5, 5)
+    filter_size = 3
+    TESTS_SHAPE.append([{"mask": p(mask), "filter_size": filter_size}, [1, 1, 5, 5, 5]])
+    mask = torch.zeros(3, 2, 5, 5, 5)
+    filter_size = 5
+    TESTS_SHAPE.append([{"mask": p(mask), "filter_size": filter_size}, [3, 2, 5, 5, 5]])
+    mask = torch.zeros(1, 1, 1, 1, 1)
+    filter_size = 5
+    TESTS_SHAPE.append([{"mask": p(mask), "filter_size": filter_size}, [1, 1, 1, 1, 1]])
+    mask = torch.zeros(1, 1, 1, 1)
+    filter_size = 5
+    TESTS_SHAPE.append([{"mask": p(mask), "filter_size": filter_size}, [1, 1, 1, 1]])
+
+TESTS_VALUE_T = []
+filter_size = 3
+mask = torch.ones(3, 2, 3, 3, 3)
+TESTS_VALUE_T.append([{"mask": mask, "filter_size": filter_size, "pad_value": 1.0}, torch.ones(3, 2, 3, 3, 3)])
+mask = torch.zeros(3, 2, 3, 3, 3)
+TESTS_VALUE_T.append([{"mask": mask, "filter_size": filter_size, "pad_value": 0.0}, torch.zeros(3, 2, 3, 3, 3)])
+mask = torch.ones(3, 2, 3, 3)
+TESTS_VALUE_T.append([{"mask": mask, "filter_size": filter_size, "pad_value": 1.0}, torch.ones(3, 2, 3, 3)])
+mask = torch.zeros(3, 2, 3, 3)
+TESTS_VALUE_T.append([{"mask": mask, "filter_size": filter_size, "pad_value": 0.0}, torch.zeros(3, 2, 3, 3)])
+
+TESTS_VALUE = []
+for p in TEST_NDARRAYS:
+    mask = torch.zeros(3, 2, 5, 5, 5)
+    filter_size = 3
+    TESTS_VALUE.append(
+        [{"mask": p(mask), "filter_size": filter_size}, p(torch.zeros(3, 2, 5, 5, 5)), p(torch.zeros(3, 2, 5, 5, 5))]
+    )
+    mask = torch.ones(1, 1, 3, 3, 3)
+    filter_size = 3
+    TESTS_VALUE.append(
+        [{"mask": p(mask), "filter_size": filter_size}, p(torch.ones(1, 1, 3, 3, 3)), p(torch.ones(1, 1, 3, 3, 3))]
+    )
+    mask = torch.ones(1, 2, 3, 3, 3)
+    filter_size = 3
+    TESTS_VALUE.append(
+        [{"mask": p(mask), "filter_size": filter_size}, p(torch.ones(1, 2, 3, 3, 3)), p(torch.ones(1, 2, 3, 3, 3))]
+    )
+    mask = torch.zeros(3, 2, 3, 3, 3)
+    mask[:, :, 1, 1, 1] = 1.0
+    filter_size = 3
+    TESTS_VALUE.append(
+        [{"mask": p(mask), "filter_size": filter_size}, p(torch.zeros(3, 2, 3, 3, 3)), p(torch.ones(3, 2, 3, 3, 3))]
+    )
+    mask = torch.zeros(3, 2, 3, 3)
+    mask[:, :, 1, 1] = 1.0
+    filter_size = 3
+    TESTS_VALUE.append(
+        [{"mask": p(mask), "filter_size": filter_size}, p(torch.zeros(3, 2, 3, 3)), p(torch.ones(3, 2, 3, 3))]
+    )
+
+
+class TestMorph(unittest.TestCase):
+
+    @parameterized.expand(TESTS_SHAPE)
+    def test_shape(self, input_data, expected_result):
+        result1 = erode(input_data["mask"], input_data["filter_size"])
+        assert_allclose(result1.shape, expected_result, type_test=False, device_test=False, atol=0.0)
+
+    @parameterized.expand(TESTS_VALUE_T)
+    def test_value_t(self, input_data, expected_result):
+        result1 = get_morphological_filter_result_t(
+            input_data["mask"], input_data["filter_size"], input_data["pad_value"]
+        )
+        assert_allclose(result1, expected_result, type_test=False, device_test=False, atol=0.0)
+
+    @parameterized.expand(TESTS_VALUE)
+    def test_value(self, input_data, expected_erode_result, expected_dilate_result):
+        result1 = erode(input_data["mask"], input_data["filter_size"])
+        assert_allclose(result1, expected_erode_result, type_test=True, device_test=True, atol=0.0)
+        result2 = dilate(input_data["mask"], input_data["filter_size"])
+        assert_allclose(result2, expected_dilate_result, type_test=True, device_test=True, atol=0.0)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 58106a6255f5936d26e479d11d3a34e3d536c112 Mon Sep 17 00:00:00 2001
From: Hans Johnson <hans-johnson@uiowa.edu>
Date: Wed, 3 Jul 2024 08:01:52 -0500
Subject: [PATCH 083/183] Remove use of deprecated python 3.12 strtobool
 (#7900)

distutils is not available in python 3.12, so a substitute is needed for
the strtobool code.

Fixes #7899

### Description

Replace distutils strtobool with local _strtobool version.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.

---------

Signed-off-by: Hans Johnson <hans-johnson@uiowa.edu>
---
 monai/utils/misc.py  | 22 ++++++++++++++++++++--
 requirements-min.txt |  3 ++-
 2 files changed, 22 insertions(+), 3 deletions(-)

diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index ab9fe259aa..96a59e1b35 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -24,7 +24,6 @@
 import warnings
 from ast import literal_eval
 from collections.abc import Callable, Iterable, Sequence
-from distutils.util import strtobool
 from math import log10
 from pathlib import Path
 from typing import TYPE_CHECKING, Any, TypeVar, cast, overload
@@ -78,6 +77,25 @@
     "run_cmd",
 ]
 
+
+def _strtobool(val: str) -> bool:
+    """
+    Replaces deprecated (pre python 3.12)
+    distutils strtobool function.
+
+    True values are y, yes, t, true, on and 1;
+    False values are n, no, f, false, off and 0.
+    Raises ValueError if val is anything else.
+    """
+    val = val.lower()
+    if val in ("y", "yes", "t", "true", "on", "1"):
+        return True
+    elif val in ("n", "no", "f", "false", "off", "0"):
+        return False
+    else:
+        raise ValueError(f"invalid truth value {val}")
+
+
 _seed = None
 _flag_deterministic = torch.backends.cudnn.deterministic
 _flag_cudnn_benchmark = torch.backends.cudnn.benchmark
@@ -400,7 +418,7 @@ def _parse_var(s):
                 d[key] = literal_eval(value)
             except ValueError:
                 try:
-                    d[key] = bool(strtobool(str(value)))
+                    d[key] = bool(_strtobool(str(value)))
                 except ValueError:
                     d[key] = value
     return d
diff --git a/requirements-min.txt b/requirements-min.txt
index ad0bb1ef20..a091ef0568 100644
--- a/requirements-min.txt
+++ b/requirements-min.txt
@@ -1,5 +1,6 @@
 # Requirements for minimal tests
 -r requirements.txt
-setuptools>=50.3.0,<66.0.0,!=60.6.0
+setuptools>=50.3.0,<66.0.0,!=60.6.0 ; python_version < "3.12"
+setuptools>=70.2.0; python_version >= "3.12"
 coverage>=5.5
 parameterized

From 3b9683c312a992217bca1a1325d5086163ac23b1 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 4 Jul 2024 20:02:04 +0800
Subject: [PATCH 084/183] Clean disk space in conda test pipeline (#7902)

Fixes #7901


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/conda.yml | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/.github/workflows/conda.yml b/.github/workflows/conda.yml
index 367a24cbde..8e2807111e 100644
--- a/.github/workflows/conda.yml
+++ b/.github/workflows/conda.yml
@@ -32,6 +32,10 @@ jobs:
         maximum-size: 16GB
         disk-root: "D:"
     - uses: actions/checkout@v4
+    - name: Clean up disk space
+      run: |
+        find /opt/hostedtoolcache/* -maxdepth 0 ! -name 'Python' -exec rm -rf {} \;
+        rm -rf /usr/share/dotnet/
     - uses: conda-incubator/setup-miniconda@v3
       with:
         auto-update-conda: true

From 64ea76d83a92b7cf7f13c8f93498d50037c3324c Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Sat, 6 Jul 2024 11:11:57 +0800
Subject: [PATCH 085/183] Fix deprecated argument in 'scipy.sparse.linalg.cg'
 (#7897)

Fixes #7896

### Description

- 'scipy.sparse.linalg.cg' keyword argument `tol` is deprecated in favor
of `rtol` and will be removed in SciPy v1.14.0. Until then, if set, it
will override `rtol`. So update to use `rtol` in `cg`.

- Drop python 3.8 test in packaging and premerge-gpu-test.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 .github/workflows/pythonapp.yml                   | 13 +++++++------
 docs/requirements.txt                             |  2 +-
 monai/data/ultrasound_confidence_map.py           | 10 +++++-----
 monai/optimizers/lr_finder.py                     |  2 +-
 requirements-dev.txt                              |  2 +-
 requirements.txt                                  |  2 +-
 setup.cfg                                         |  4 ++--
 tests/test_ultrasound_confidence_map_transform.py |  4 ++++
 tests/utils.py                                    |  2 +-
 9 files changed, 23 insertions(+), 18 deletions(-)

diff --git a/.github/workflows/pythonapp.yml b/.github/workflows/pythonapp.yml
index d1e77bb567..cd6b6ccede 100644
--- a/.github/workflows/pythonapp.yml
+++ b/.github/workflows/pythonapp.yml
@@ -29,10 +29,10 @@ jobs:
         opt: ["codeformat", "pytype", "mypy"]
     steps:
     - uses: actions/checkout@v4
-    - name: Set up Python 3.8
+    - name: Set up Python 3.9
       uses: actions/setup-python@v5
       with:
-        python-version: '3.8'
+        python-version: '3.9'
     - name: cache weekly timestamp
       id: pip-cache
       run: |
@@ -45,6 +45,7 @@ jobs:
         key: ${{ runner.os }}-pip-${{ steps.pip-cache.outputs.datew }}
     - name: Install dependencies
       run: |
+        find  /opt/hostedtoolcache/* -maxdepth 0 ! -name 'Python' -exec rm -rf {} \;
         python -m pip install --upgrade pip wheel
         python -m pip install -r requirements-dev.txt
     - name: Lint and type check
@@ -130,10 +131,10 @@ jobs:
     - uses: actions/checkout@v4
       with:
         fetch-depth: 0
-    - name: Set up Python 3.8
+    - name: Set up Python 3.9
       uses: actions/setup-python@v5
       with:
-        python-version: '3.8'
+        python-version: '3.9'
     - name: cache weekly timestamp
       id: pip-cache
       run: |
@@ -211,10 +212,10 @@ jobs:
     runs-on: ubuntu-latest
     steps:
     - uses: actions/checkout@v4
-    - name: Set up Python 3.8
+    - name: Set up Python 3.9
       uses: actions/setup-python@v5
       with:
-        python-version: '3.8'
+        python-version: '3.9'
     - name: cache weekly timestamp
       id: pip-cache
       run: |
diff --git a/docs/requirements.txt b/docs/requirements.txt
index 6caddce666..fe415a07b5 100644
--- a/docs/requirements.txt
+++ b/docs/requirements.txt
@@ -6,7 +6,7 @@ itk>=5.2
 nibabel
 parameterized
 scikit-image>=0.19.0
-scipy>=1.7.1
+scipy>=1.12.0; python_version >= '3.9'
 tensorboard
 commonmark==0.9.1
 recommonmark==0.6.0
diff --git a/monai/data/ultrasound_confidence_map.py b/monai/data/ultrasound_confidence_map.py
index 5c716b62be..865e4a0a0f 100644
--- a/monai/data/ultrasound_confidence_map.py
+++ b/monai/data/ultrasound_confidence_map.py
@@ -19,10 +19,10 @@
 __all__ = ["UltrasoundConfidenceMap"]
 
 cv2, _ = optional_import("cv2")
-csc_matrix, _ = optional_import("scipy.sparse", "1.7.1", min_version, "csc_matrix")
-spsolve, _ = optional_import("scipy.sparse.linalg", "1.7.1", min_version, "spsolve")
-cg, _ = optional_import("scipy.sparse.linalg", "1.7.1", min_version, "cg")
-hilbert, _ = optional_import("scipy.signal", "1.7.1", min_version, "hilbert")
+csc_matrix, _ = optional_import("scipy.sparse", "1.12.0", min_version, "csc_matrix")
+spsolve, _ = optional_import("scipy.sparse.linalg", "1.12.0", min_version, "spsolve")
+cg, _ = optional_import("scipy.sparse.linalg", "1.12.0", min_version, "cg")
+hilbert, _ = optional_import("scipy.signal", "1.12.0", min_version, "hilbert")
 ruge_stuben_solver, _ = optional_import("pyamg", "5.0.0", min_version, "ruge_stuben_solver")
 
 
@@ -285,7 +285,7 @@ def _solve_linear_system(self, lap, rhs):
             lap_sparse = lap.tocsr()
             ml = ruge_stuben_solver(lap_sparse, coarse_solver="pinv")
             m = ml.aspreconditioner(cycle="V")
-            x, _ = cg(lap, rhs, tol=self.cg_tol, maxiter=self.cg_maxiter, M=m)
+            x, _ = cg(lap, rhs, rtol=self.cg_tol, maxiter=self.cg_maxiter, M=m)
         else:
             x = spsolve(lap, rhs)
 
diff --git a/monai/optimizers/lr_finder.py b/monai/optimizers/lr_finder.py
index 045135628d..aa2e4567b3 100644
--- a/monai/optimizers/lr_finder.py
+++ b/monai/optimizers/lr_finder.py
@@ -524,7 +524,7 @@ def plot(
         # Plot the LR with steepest gradient
         if steepest_lr:
             lr_at_steepest_grad, loss_at_steepest_grad = self.get_steepest_gradient(skip_start, skip_end)
-            if lr_at_steepest_grad is not None:
+            if lr_at_steepest_grad is not None and loss_at_steepest_grad is not None:
                 ax.scatter(
                     lr_at_steepest_grad,
                     loss_at_steepest_grad,
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 1bba930273..ced783443e 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -2,7 +2,7 @@
 -r requirements-min.txt
 pytorch-ignite==0.4.11
 gdown>=4.7.3
-scipy>=1.7.1
+scipy>=1.12.0; python_version >= '3.9'
 itk>=5.2
 nibabel
 pillow!=8.3.0  # https://github.com/python-pillow/Pillow/issues/5571
diff --git a/requirements.txt b/requirements.txt
index 1d6ae13eec..aae455f58c 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,2 +1,2 @@
 torch>=1.9
-numpy>=1.20,<2.0
+numpy>=1.20,<=1.26.0
diff --git a/setup.cfg b/setup.cfg
index 05bf181c70..b2b32066ab 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -49,7 +49,7 @@ all =
     nibabel
     ninja
     scikit-image>=0.14.2
-    scipy>=1.7.1
+    scipy>=1.12.0; python_version >= '3.9'
     pillow
     tensorboard
     gdown>=4.7.3
@@ -92,7 +92,7 @@ ninja =
 skimage =
     scikit-image>=0.14.2
 scipy =
-    scipy>=1.7.1
+    scipy>=1.12.0; python_version >= '3.9'
 pillow =
     pillow!=8.3.0
 tensorboard =
diff --git a/tests/test_ultrasound_confidence_map_transform.py b/tests/test_ultrasound_confidence_map_transform.py
index 87c08b3ac3..1c6b8f7635 100644
--- a/tests/test_ultrasound_confidence_map_transform.py
+++ b/tests/test_ultrasound_confidence_map_transform.py
@@ -20,8 +20,11 @@
 from PIL import Image
 
 from monai.transforms import UltrasoundConfidenceMapTransform
+from monai.utils import optional_import
 from tests.utils import assert_allclose
 
+_, has_scipy = optional_import("scipy")
+
 TEST_INPUT = np.array(
     [
         [1, 2, 3, 23, 13, 22, 5, 1, 2, 3],
@@ -482,6 +485,7 @@
 )
 
 
+@unittest.skipUnless(has_scipy, "Requires scipy")
 class TestUltrasoundConfidenceMapTransform(unittest.TestCase):
 
     def setUp(self):
diff --git a/tests/utils.py b/tests/utils.py
index d1939e590b..77b53cebb8 100644
--- a/tests/utils.py
+++ b/tests/utils.py
@@ -475,7 +475,7 @@ def run_process(self, func, local_rank, args, kwargs, results):
             if self.verbose:
                 os.environ["NCCL_DEBUG"] = "INFO"
                 os.environ["NCCL_DEBUG_SUBSYS"] = "ALL"
-            os.environ["NCCL_BLOCKING_WAIT"] = str(1)
+            os.environ["TORCH_NCCL_BLOCKING_WAIT"] = str(1)
             os.environ["OMP_NUM_THREADS"] = str(1)
             os.environ["WORLD_SIZE"] = str(self.nproc_per_node * self.nnodes)
             os.environ["RANK"] = str(self.nproc_per_node * self.node_rank + local_rank)

From 3a0c2d5afcc91379819fd8428fa83b00fb18c9fd Mon Sep 17 00:00:00 2001
From: monai-bot <64792179+monai-bot@users.noreply.github.com>
Date: Mon, 8 Jul 2024 12:00:09 +0100
Subject: [PATCH 086/183] auto updates (#7903)

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>

Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
---
 monai/apps/generation/maisi/networks/autoencoderkl_maisi.py | 2 --
 tests/test_autoencoderkl_maisi.py                           | 2 +-
 tests/test_controlnet_maisi.py                              | 1 +
 tests/test_diffusion_model_unet_maisi.py                    | 3 ++-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py b/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
index 533da32fa0..f27f73ec60 100644
--- a/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
+++ b/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
@@ -27,13 +27,11 @@
 AutoencoderKL, has_autoencoderkl = optional_import("generative.networks.nets.autoencoderkl", name="AutoencoderKL")
 ResBlock, has_resblock = optional_import("generative.networks.nets.autoencoderkl", name="ResBlock")
 
-
 if TYPE_CHECKING:
     from generative.networks.nets.autoencoderkl import AutoencoderKL as AutoencoderKLType
 else:
     AutoencoderKLType = cast(type, AutoencoderKL)
 
-
 # Set up logging configuration
 logger = logging.getLogger(__name__)
 
diff --git a/tests/test_autoencoderkl_maisi.py b/tests/test_autoencoderkl_maisi.py
index e88dc469c9..392a3d7db2 100644
--- a/tests/test_autoencoderkl_maisi.py
+++ b/tests/test_autoencoderkl_maisi.py
@@ -29,7 +29,6 @@
 
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
-
 CASES_NO_ATTENTION = [
     [
         {
@@ -82,6 +81,7 @@
 
 @unittest.skipUnless(has_generative, "monai-generative required")
 class TestAutoencoderKlMaisi(unittest.TestCase):
+
     @parameterized.expand(CASES)
     def test_shape(self, input_param, input_shape, expected_shape, expected_latent_shape):
         net = AutoencoderKlMaisi(**input_param).to(device)
diff --git a/tests/test_controlnet_maisi.py b/tests/test_controlnet_maisi.py
index b522b750c8..7b0e69f2c8 100644
--- a/tests/test_controlnet_maisi.py
+++ b/tests/test_controlnet_maisi.py
@@ -130,6 +130,7 @@
 @SkipIfBeforePyTorchVersion((2, 0))
 @skipUnless(has_generative, "monai-generative required")
 class TestControlNet(unittest.TestCase):
+
     @parameterized.expand(TEST_CASES)
     def test_shape_unconditioned_models(self, input_param, expected_num_down_blocks_residuals, expected_shape):
         net = ControlNetMaisi(**input_param)
diff --git a/tests/test_diffusion_model_unet_maisi.py b/tests/test_diffusion_model_unet_maisi.py
index b5c14192d9..059a4a4ba8 100644
--- a/tests/test_diffusion_model_unet_maisi.py
+++ b/tests/test_diffusion_model_unet_maisi.py
@@ -26,7 +26,6 @@
 if has_generative:
     from monai.apps.generation.maisi.networks.diffusion_model_unet_maisi import DiffusionModelUNetMaisi
 
-
 UNCOND_CASES_2D = [
     [
         {
@@ -294,6 +293,7 @@
 
 @skipUnless(has_generative, "monai-generative required")
 class TestDiffusionModelUNetMaisi2D(unittest.TestCase):
+
     @parameterized.expand(UNCOND_CASES_2D)
     @skipUnless(has_einops, "Requires einops")
     def test_shape_unconditioned_models(self, input_param):
@@ -512,6 +512,7 @@ def test_shape_with_additional_inputs(self, input_param):
 
 @skipUnless(has_generative, "monai-generative required")
 class TestDiffusionModelUNetMaisi3D(unittest.TestCase):
+
     @parameterized.expand(UNCOND_CASES_3D)
     @skipUnless(has_einops, "Requires einops")
     def test_shape_unconditioned_models(self, input_param):

From 8cfbcbabd1529ef4090fb6f7ffbeef47d6b70cc2 Mon Sep 17 00:00:00 2001
From: Hans Johnson <hans-johnson@uiowa.edu>
Date: Tue, 9 Jul 2024 01:29:29 -0500
Subject: [PATCH 087/183] Replace deprecated pkgutil.find_loader use (#7906)

### Description

Replace deprecated function with modern version

<string>:1: DeprecationWarning: 'pkgutil.find_loader' is deprecated and
slated for removal in Python 3.14; use importlib.util.find_spec()
instead


### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).

Signed-off-by: Hans Johnson <hans-johnson@uiowa.edu>
---
 runtests.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/runtests.sh b/runtests.sh
index 0b3e20ce49..65e3a2bb6b 100755
--- a/runtests.sh
+++ b/runtests.sh
@@ -167,7 +167,7 @@ function clang_format {
 }
 
 function is_pip_installed() {
-	return $("${PY_EXE}" -c "import sys, pkgutil; sys.exit(0 if pkgutil.find_loader(sys.argv[1]) else 1)" $1)
+	return $("${PY_EXE}" -c "import sys, importlib.util; sys.exit(0 if importlib.util.find_spec(sys.argv[1]) else 1)" $1)
 }
 
 function clean_py {

From 9554f471a416d2f111dce08cf2d729254a502e48 Mon Sep 17 00:00:00 2001
From: Mathijs de Boer <8137653+MathijsdeBoer@users.noreply.github.com>
Date: Thu, 11 Jul 2024 19:00:38 +0200
Subject: [PATCH 088/183] Fix docstring indentation in `SaveImage` (#7913)

### Description

The docstring for `monai.transforms.io.array.SaveImage` had some faulty
indentation, causing weirdly formatted documentation on the website.
This pull request fixes the indentation.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Mathijs de Boer <m.deboer-41@umcutrecht.nl>
Co-authored-by: Mathijs de Boer <m.deboer-41@umcutrecht.nl>
---
 monai/transforms/io/array.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/monai/transforms/io/array.py b/monai/transforms/io/array.py
index 7222a26fc3..e0ecc127f2 100644
--- a/monai/transforms/io/array.py
+++ b/monai/transforms/io/array.py
@@ -307,11 +307,11 @@ class SaveImage(Transform):
 
     Args:
         output_dir: output image directory.
-        Handled by ``folder_layout`` instead, if ``folder_layout`` is not ``None``.
+            Handled by ``folder_layout`` instead, if ``folder_layout`` is not ``None``.
         output_postfix: a string appended to all output file names, default to `trans`.
-        Handled by ``folder_layout`` instead, if ``folder_layout`` is not ``None``.
+            Handled by ``folder_layout`` instead, if ``folder_layout`` is not ``None``.
         output_ext: output file extension name.
-        Handled by ``folder_layout`` instead, if ``folder_layout`` is not ``None``.
+            Handled by ``folder_layout`` instead, if ``folder_layout`` is not ``None``.
         output_dtype: data type (if not None) for saving data. Defaults to ``np.float32``.
         resample: whether to resample image (if needed) before saving the data array,
             based on the ``"spatial_shape"`` (and ``"original_affine"``) from metadata.

From 14b086b553693f5d344ff054f37d12ce6839da06 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 12 Jul 2024 20:16:54 +0800
Subject: [PATCH 089/183] 7908 drop python 3.8 (#7909)

Fix #7908

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/cron-ngc-bundle.yml |  4 ++--
 .github/workflows/pythonapp-gpu.yml   |  2 +-
 .github/workflows/pythonapp-min.yml   | 12 ++++++------
 .github/workflows/setupapp.yml        |  6 +++---
 CONTRIBUTING.md                       |  2 +-
 docs/images/python.svg                |  2 +-
 docs/source/installation.md           |  2 +-
 monai/__init__.py                     |  2 +-
 monai/apps/auto3dseg/auto_runner.py   |  2 +-
 monai/auto3dseg/utils.py              |  2 +-
 setup.cfg                             |  4 ++--
 11 files changed, 20 insertions(+), 20 deletions(-)

diff --git a/.github/workflows/cron-ngc-bundle.yml b/.github/workflows/cron-ngc-bundle.yml
index bd45bc8d1e..d4b45e1d55 100644
--- a/.github/workflows/cron-ngc-bundle.yml
+++ b/.github/workflows/cron-ngc-bundle.yml
@@ -18,10 +18,10 @@ jobs:
     runs-on: ubuntu-latest
     steps:
     - uses: actions/checkout@v4
-    - name: Set up Python 3.8
+    - name: Set up Python 3.9
       uses: actions/setup-python@v5
       with:
-        python-version: '3.8'
+        python-version: '3.9'
     - name: cache weekly timestamp
       id: pip-cache
       run: echo "datew=$(date '+%Y-%V')" >> $GITHUB_OUTPUT
diff --git a/.github/workflows/pythonapp-gpu.yml b/.github/workflows/pythonapp-gpu.yml
index f83d52f8e3..ead622b39c 100644
--- a/.github/workflows/pythonapp-gpu.yml
+++ b/.github/workflows/pythonapp-gpu.yml
@@ -62,7 +62,7 @@ jobs:
         if [ ${{ matrix.environment }} = "PT110+CUDA111" ] || \
           [ ${{ matrix.environment }} = "PT113+CUDA116" ]
         then
-        PYVER=3.8 PYSFX=3 DISTUTILS=python3-distutils && \
+        PYVER=3.9 PYSFX=3 DISTUTILS=python3-distutils && \
         apt-get update && apt-get install -y --no-install-recommends \
           curl \
           pkg-config \
diff --git a/.github/workflows/pythonapp-min.yml b/.github/workflows/pythonapp-min.yml
index dffae10558..02d8f5058e 100644
--- a/.github/workflows/pythonapp-min.yml
+++ b/.github/workflows/pythonapp-min.yml
@@ -31,10 +31,10 @@ jobs:
     timeout-minutes: 40
     steps:
     - uses: actions/checkout@v4
-    - name: Set up Python 3.8
+    - name: Set up Python 3.9
       uses: actions/setup-python@v5
       with:
-        python-version: '3.8'
+        python-version: '3.9'
     - name: Prepare pip wheel
       run: |
         which python
@@ -73,7 +73,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ['3.8', '3.9', '3.10', '3.11']
+        python-version: ['3.9', '3.10', '3.11', '3.12']
     timeout-minutes: 40
     steps:
     - uses: actions/checkout@v4
@@ -118,14 +118,14 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        pytorch-version: ['1.9.1', '1.10.2', '1.11.0', '1.12.1', '1.13', 'latest']
+        pytorch-version: ['1.10.2', '1.11.0', '1.12.1', '1.13', '2.0.1', 'latest']
     timeout-minutes: 40
     steps:
     - uses: actions/checkout@v4
-    - name: Set up Python 3.8
+    - name: Set up Python 3.9
       uses: actions/setup-python@v5
       with:
-        python-version: '3.8'
+        python-version: '3.9'
     - name: Prepare pip wheel
       run: |
         which python
diff --git a/.github/workflows/setupapp.yml b/.github/workflows/setupapp.yml
index c6ad243b81..a76635e224 100644
--- a/.github/workflows/setupapp.yml
+++ b/.github/workflows/setupapp.yml
@@ -77,7 +77,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: ['3.8', '3.9', '3.10']
+        python-version: ['3.9', '3.10', '3.11']
     steps:
     - uses: actions/checkout@v4
       with:
@@ -119,10 +119,10 @@ jobs:
   install:  # pip install from github url, the default branch is dev
     runs-on: ubuntu-latest
     steps:
-    - name: Set up Python 3.8
+    - name: Set up Python 3.9
       uses: actions/setup-python@v5
       with:
-        python-version: '3.8'
+        python-version: '3.9'
     - name: cache weekly timestamp
       id: pip-cache
       run: |
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 5c886cff30..8db796637f 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -123,7 +123,7 @@ or (for new features that would not break existing functionality):
 ```
 
 It is recommended that the new test `test_[module_name].py` is constructed by using only
-python 3.8+ build-in functions, `torch`, `numpy`, `coverage` (for reporting code coverages) and `parameterized` (for organising test cases) packages.
+python 3.9+ build-in functions, `torch`, `numpy`, `coverage` (for reporting code coverages) and `parameterized` (for organising test cases) packages.
 If it requires any other external packages, please make sure:
 - the packages are listed in [`requirements-dev.txt`](requirements-dev.txt)
 - the new test `test_[module_name].py` is added to the `exclude_cases` in [`./tests/min_tests.py`](./tests/min_tests.py) so that
diff --git a/docs/images/python.svg b/docs/images/python.svg
index b7aa7c59bd..8ef6b61c03 100644
--- a/docs/images/python.svg
+++ b/docs/images/python.svg
@@ -1 +1 @@
-<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="102.2" height="20"><linearGradient id="smooth" x2="0" y2="100%"><stop offset="0" stop-color="#bbb" stop-opacity=".1"/><stop offset="1" stop-opacity=".1"/></linearGradient><clipPath id="round"><rect width="102.2" height="20" rx="3" fill="#fff"/></clipPath><g clip-path="url(#round)"><rect width="65.5" height="20" fill="#555"/><rect x="65.5" width="36.7" height="20" fill="#007ec6"/><rect width="102.2" height="20" fill="url(#smooth)"/></g><g fill="#fff" text-anchor="middle" font-family="DejaVu Sans,Verdana,Geneva,sans-serif" font-size="110"><image x="5" y="3" width="14" height="14" xlink:href="data:image/svg+xml;base64,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"/><text x="422.5" y="150" fill="#010101" fill-opacity=".3" transform="scale(0.1)" textLength="385.0" lengthAdjust="spacing">python</text><text x="422.5" y="140" transform="scale(0.1)" textLength="385.0" lengthAdjust="spacing">python</text><text x="828.5" y="150" fill="#010101" fill-opacity=".3" transform="scale(0.1)" textLength="267.0" lengthAdjust="spacing">3.8+</text><text x="828.5" y="140" transform="scale(0.1)" textLength="267.0" lengthAdjust="spacing">3.8+</text></g></svg>
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="102.2" height="20"><linearGradient id="smooth" x2="0" y2="100%"><stop offset="0" stop-color="#bbb" stop-opacity=".1"/><stop offset="1" stop-opacity=".1"/></linearGradient><clipPath id="round"><rect width="102.2" height="20" rx="3" fill="#fff"/></clipPath><g clip-path="url(#round)"><rect width="65.5" height="20" fill="#555"/><rect x="65.5" width="36.7" height="20" fill="#007ec6"/><rect width="102.2" height="20" fill="url(#smooth)"/></g><g fill="#fff" text-anchor="middle" font-family="DejaVu Sans,Verdana,Geneva,sans-serif" font-size="110"><image x="5" y="3" width="14" height="14" xlink:href="data:image/svg+xml;base64,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"/><text x="422.5" y="150" fill="#010101" fill-opacity=".3" transform="scale(0.1)" textLength="385.0" lengthAdjust="spacing">python</text><text x="422.5" y="140" transform="scale(0.1)" textLength="385.0" lengthAdjust="spacing">python</text><text x="828.5" y="150" fill="#010101" fill-opacity=".3" transform="scale(0.1)" textLength="267.0" lengthAdjust="spacing">3.9+</text><text x="828.5" y="140" transform="scale(0.1)" textLength="267.0" lengthAdjust="spacing">3.9+</text></g></svg>
diff --git a/docs/source/installation.md b/docs/source/installation.md
index 644dd623c1..4308a07647 100644
--- a/docs/source/installation.md
+++ b/docs/source/installation.md
@@ -19,7 +19,7 @@
 
 ---
 
-MONAI's core functionality is written in Python 3 (>= 3.8) and only requires [Numpy](https://numpy.org/) and [Pytorch](https://pytorch.org/).
+MONAI's core functionality is written in Python 3 (>= 3.9) and only requires [Numpy](https://numpy.org/) and [Pytorch](https://pytorch.org/).
 
 The package is currently distributed via Github as the primary source code repository,
 and the Python package index (PyPI). The pre-built Docker images are made available on DockerHub.
diff --git a/monai/__init__.py b/monai/__init__.py
index eb05ac993d..cb0ccd36f8 100644
--- a/monai/__init__.py
+++ b/monai/__init__.py
@@ -17,7 +17,7 @@
 from ._version import get_versions
 
 PY_REQUIRED_MAJOR = 3
-PY_REQUIRED_MINOR = 8
+PY_REQUIRED_MINOR = 9
 
 version_dict = get_versions()
 __version__: str = version_dict.get("version", "0+unknown")
diff --git a/monai/apps/auto3dseg/auto_runner.py b/monai/apps/auto3dseg/auto_runner.py
index 5b6b501555..99bf92c481 100644
--- a/monai/apps/auto3dseg/auto_runner.py
+++ b/monai/apps/auto3dseg/auto_runner.py
@@ -551,7 +551,7 @@ def set_device_info(
             cmd_prefix: command line prefix for subprocess running in BundleAlgo and EnsembleRunner.
                 Default using env "CMD_PREFIX" or None, examples are:
 
-                    - single GPU/CPU or multinode bcprun: "python " or "/opt/conda/bin/python3.8 ",
+                    - single GPU/CPU or multinode bcprun: "python " or "/opt/conda/bin/python3.9 ",
                     - single node multi-GPU running "torchrun --nnodes=1 --nproc_per_node=2 "
 
                 If user define this prefix, please make sure --nproc_per_node matches cuda_visible_device or
diff --git a/monai/auto3dseg/utils.py b/monai/auto3dseg/utils.py
index 58b900d410..211f23c415 100644
--- a/monai/auto3dseg/utils.py
+++ b/monai/auto3dseg/utils.py
@@ -407,7 +407,7 @@ def _prepare_cmd_default(cmd: str, cmd_prefix: str | None = None, **kwargs: Any)
 
     Args:
         cmd: the command or script to run in the distributed job.
-        cmd_prefix: the command prefix to run the script, e.g., "python", "python -m", "python3", "/opt/conda/bin/python3.8 ".
+        cmd_prefix: the command prefix to run the script, e.g., "python", "python -m", "python3", "/opt/conda/bin/python3.9 ".
         kwargs: the keyword arguments to be passed to the script.
 
     Returns:
diff --git a/setup.cfg b/setup.cfg
index b2b32066ab..202e7b0e24 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -21,9 +21,9 @@ classifiers =
     Intended Audience :: Healthcare Industry
     Programming Language :: C++
     Programming Language :: Python :: 3
-    Programming Language :: Python :: 3.8
     Programming Language :: Python :: 3.9
     Programming Language :: Python :: 3.10
+    Programming Language :: Python :: 3.11
     Topic :: Scientific/Engineering
     Topic :: Scientific/Engineering :: Artificial Intelligence
     Topic :: Scientific/Engineering :: Medical Science Apps.
@@ -33,7 +33,7 @@ classifiers =
     Typing :: Typed
 
 [options]
-python_requires = >= 3.8
+python_requires = >= 3.9
 # for compiling and develop setup only
 # no need to specify the versions so that we could
 # compile for multiple targeted versions.

From 848005d2d325f20e78c82efff6df987c9e63eb9f Mon Sep 17 00:00:00 2001
From: Hans Johnson <hans-johnson@uiowa.edu>
Date: Sun, 14 Jul 2024 09:30:34 -0500
Subject: [PATCH 090/183] Fix failing unit-test test_wsireader (#7905)

Conversion of units when the unit if '' caused the test to fail.

```bash
pytest -k tiff
```

FAILED
tests/test_wsireader.py::TestTiffFile::test_resolution_mpp_0__home_johnsonhj_src_MONAI_tests_testing_data_temp_wsi_generic_tiff_tiff
- ValueError: Currently, it only supports length conversion but `` is
given.

FAILED
tests/test_wsireader.py::TestTiffFile::test_resolution_mpp_0__home_johnsonhj_src_MONAI_tests_testing_data_temp_wsi_generic_tiff_tiff
- AttributeError: 'ConvertUnits' object has no attribute
'conversion_factor'

### Description

Fixes a failing test.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.

Signed-off-by: Hans Johnson <hans-johnson@uiowa.edu>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/data/wsi_reader.py | 2 +-
 monai/utils/misc.py      | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/monai/data/wsi_reader.py b/monai/data/wsi_reader.py
index b31d4d9c3a..96d84d8cf1 100644
--- a/monai/data/wsi_reader.py
+++ b/monai/data/wsi_reader.py
@@ -1098,7 +1098,7 @@ def get_mpp(self, wsi, level: int) -> tuple[float, float]:
             unit = wsi.pages[level].tags.get("ResolutionUnit")
             if unit is not None:
                 unit = str(unit.value)[8:]
-            else:
+            if unit is None or len(unit) == 0:
                 warnings.warn("The resolution unit is missing. `micrometer` will be used as default.")
                 unit = "micrometer"
 
diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index 96a59e1b35..e8a46ecc61 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -814,7 +814,7 @@ def __init__(self, input_unit: str, target_unit: str) -> None:
                 "Both input and target units should be from the same quantity. "
                 f"Input quantity is {input_base} while target quantity is {target_base}"
             )
-        self._calculate_conversion_factor()
+        self.conversion_factor = self._calculate_conversion_factor()
 
     def _get_valid_unit_and_base(self, unit):
         unit = str(unit).lower()
@@ -841,7 +841,7 @@ def _calculate_conversion_factor(self):
             return 1.0
         input_power = self._get_unit_power(self.input_unit)
         target_power = self._get_unit_power(self.target_unit)
-        self.conversion_factor = 10 ** (input_power - target_power)
+        return 10 ** (input_power - target_power)
 
     def __call__(self, value: int | float) -> Any:
         return float(value) * self.conversion_factor

From 4fbe8004b69cc9a80576a3b7419b22a86d3fe933 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 16 Jul 2024 19:28:44 +0800
Subject: [PATCH 091/183] Fix wsireader get mpp issue (#7921)

Fixes #7918

### Description

The main issue is that enum is expressed differently between different
pythons
Related PR: https://github.com/Project-MONAI/MONAI/pull/7905

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/data/wsi_reader.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/data/wsi_reader.py b/monai/data/wsi_reader.py
index 96d84d8cf1..2a4fe9f7a8 100644
--- a/monai/data/wsi_reader.py
+++ b/monai/data/wsi_reader.py
@@ -1097,7 +1097,7 @@ def get_mpp(self, wsi, level: int) -> tuple[float, float]:
         ):
             unit = wsi.pages[level].tags.get("ResolutionUnit")
             if unit is not None:
-                unit = str(unit.value)[8:]
+                unit = str(unit.value.name)
             if unit is None or len(unit) == 0:
                 warnings.warn("The resolution unit is missing. `micrometer` will be used as default.")
                 unit = "micrometer"

From 50d518051f6bb7e186513154ab5cfcd6a8a6149d Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 17 Jul 2024 20:17:22 +0800
Subject: [PATCH 092/183] Support download bundles from ngc private registry
 (#7907)

### Description

Support download from ngc private registry, this download option
requires ngc api key.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 .github/workflows/blossom-ci.yml    |   3 +
 .github/workflows/conda.yml         |   4 +
 .github/workflows/cron.yml          |  12 +++
 .github/workflows/docker.yml        |   3 +
 .github/workflows/integration.yml   |   3 +
 .github/workflows/pythonapp-min.yml |   9 +++
 .github/workflows/setupapp.yml      |   8 ++
 monai/bundle/scripts.py             | 114 ++++++++++++++++++++++++++--
 tests/test_bundle_download.py       |  19 ++++-
 9 files changed, 166 insertions(+), 9 deletions(-)

diff --git a/.github/workflows/blossom-ci.yml b/.github/workflows/blossom-ci.yml
index bf507bab3b..736d4bfdf6 100644
--- a/.github/workflows/blossom-ci.yml
+++ b/.github/workflows/blossom-ci.yml
@@ -93,6 +93,9 @@ jobs:
         run: blossom-ci
         env:
           OPERATION: 'START-CI-JOB'
+          NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+          NGC_ORG: ${{ secrets.NGC_ORG }}
+          NGC_TEAM: ${{ secrets.NGC_TEAM }}
           CI_SERVER: ${{ secrets.CI_SERVER }}
           REPO_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 
diff --git a/.github/workflows/conda.yml b/.github/workflows/conda.yml
index 8e2807111e..394685acd3 100644
--- a/.github/workflows/conda.yml
+++ b/.github/workflows/conda.yml
@@ -60,6 +60,10 @@ jobs:
         conda deactivate
     - name: Test env (CPU ${{ runner.os }})
       shell: bash -el {0}
+      env:
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
       run: |
         conda activate monai
         $(pwd)/runtests.sh --build --unittests
diff --git a/.github/workflows/cron.yml b/.github/workflows/cron.yml
index 0f9e6cd480..cc113b0446 100644
--- a/.github/workflows/cron.yml
+++ b/.github/workflows/cron.yml
@@ -50,6 +50,10 @@ jobs:
         python -m pip install -r requirements-dev.txt
         python -m pip list
     - name: Run tests report coverage
+      env:
+          NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+          NGC_ORG: ${{ secrets.NGC_ORG }}
+          NGC_TEAM: ${{ secrets.NGC_TEAM }}
       run: |
         export LAUNCH_DELAY=$[ $RANDOM % 16 * 60 ]
         echo "Sleep $LAUNCH_DELAY"
@@ -94,6 +98,10 @@ jobs:
         python -m pip install -r requirements-dev.txt
         python -m pip list
     - name: Run tests report coverage
+      env:
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
       run: |
         export LAUNCH_DELAY=$[ $RANDOM % 16 * 60 ]
         echo "Sleep $LAUNCH_DELAY"
@@ -196,6 +204,10 @@ jobs:
     - name: Run tests report coverage
       # The docker image process has done the compilation.
       # BUILD_MONAI=1 is necessary for triggering the USE_COMPILED flag.
+      env:
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
       run: |
         cd /opt/monai
         nvidia-smi
diff --git a/.github/workflows/docker.yml b/.github/workflows/docker.yml
index 65716f86f9..17ffe4cf90 100644
--- a/.github/workflows/docker.yml
+++ b/.github/workflows/docker.yml
@@ -100,3 +100,6 @@ jobs:
       shell: bash
       env:
         QUICKTEST: True
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
diff --git a/.github/workflows/integration.yml b/.github/workflows/integration.yml
index c82530a551..5be2ebb86c 100644
--- a/.github/workflows/integration.yml
+++ b/.github/workflows/integration.yml
@@ -68,6 +68,9 @@ jobs:
       shell: bash
       env:
         BUILD_MONAI: 1
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
       run: ./runtests.sh --build --net
 
     - name: Add reaction
diff --git a/.github/workflows/pythonapp-min.yml b/.github/workflows/pythonapp-min.yml
index 02d8f5058e..b0d37937e9 100644
--- a/.github/workflows/pythonapp-min.yml
+++ b/.github/workflows/pythonapp-min.yml
@@ -67,6 +67,9 @@ jobs:
       shell: bash
       env:
         QUICKTEST: True
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
 
   min-dep-py3:  # min dependencies installed tests for different python
     runs-on: ubuntu-latest
@@ -112,6 +115,9 @@ jobs:
         ./runtests.sh --min
       env:
         QUICKTEST: True
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
 
   min-dep-pytorch:  # min dependencies installed tests for different pytorch
     runs-on: ubuntu-latest
@@ -161,3 +167,6 @@ jobs:
         ./runtests.sh --min
       env:
         QUICKTEST: True
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
diff --git a/.github/workflows/setupapp.yml b/.github/workflows/setupapp.yml
index a76635e224..7e01f55cd9 100644
--- a/.github/workflows/setupapp.yml
+++ b/.github/workflows/setupapp.yml
@@ -49,6 +49,10 @@ jobs:
         python -m pip install --upgrade torch torchvision
         python -m pip install -r requirements-dev.txt
     - name: Run unit tests report coverage
+      env:
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
       run: |
         python -m pip list
         git config --global --add safe.directory /__w/MONAI/MONAI
@@ -104,6 +108,10 @@ jobs:
         python -m pip install --upgrade pip wheel
         python -m pip install -r requirements-dev.txt
     - name: Run quick tests CPU ubuntu
+      env:
+        NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
+        NGC_ORG: ${{ secrets.NGC_ORG }}
+        NGC_TEAM: ${{ secrets.NGC_TEAM }}
       run: |
         python -m pip list
         python -c 'import torch; print(torch.__version__); print(torch.rand(5,3))'
diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index 598d938cbd..975b6cbdbd 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -16,6 +16,7 @@
 import os
 import re
 import warnings
+import zipfile
 from collections.abc import Mapping, Sequence
 from pathlib import Path
 from pydoc import locate
@@ -171,6 +172,10 @@ def _get_ngc_bundle_url(model_name: str, version: str) -> str:
     return f"https://api.ngc.nvidia.com/v2/models/nvidia/monaitoolkit/{model_name.lower()}/versions/{version}/zip"
 
 
+def _get_ngc_private_bundle_url(model_name: str, version: str, repo: str) -> str:
+    return f"https://api.ngc.nvidia.com/v2/{repo}/models/{model_name.lower()}/versions/{version}/zip"
+
+
 def _get_monaihosting_bundle_url(model_name: str, version: str) -> str:
     monaihosting_root_path = "https://api.ngc.nvidia.com/v2/models/nvidia/monaihosting"
     return f"{monaihosting_root_path}/{model_name.lower()}/versions/{version}/files/{model_name}_v{version}.zip"
@@ -219,6 +224,48 @@ def _download_from_ngc(
     extractall(filepath=filepath, output_dir=extract_path, has_base=True)
 
 
+def _download_from_ngc_private(
+    download_path: Path, filename: str, version: str, remove_prefix: str | None, repo: str, headers: dict | None = None
+) -> None:
+    # ensure prefix is contained
+    filename = _add_ngc_prefix(filename)
+    request_url = _get_ngc_private_bundle_url(model_name=filename, version=version, repo=repo)
+    if has_requests:
+        headers = {} if headers is None else headers
+        response = requests_get(request_url, headers=headers)
+        response.raise_for_status()
+    else:
+        raise ValueError("NGC API requires requests package. Please install it.")
+
+    zip_path = download_path / f"{filename}_v{version}.zip"
+    with open(zip_path, "wb") as f:
+        f.write(response.content)
+    logger.info(f"Downloading: {zip_path}.")
+    if remove_prefix:
+        filename = _remove_ngc_prefix(filename, prefix=remove_prefix)
+    extract_path = download_path / f"{filename}"
+    with zipfile.ZipFile(zip_path, "r") as z:
+        z.extractall(extract_path)
+        logger.info(f"Writing into directory: {extract_path}.")
+
+
+def _get_ngc_token(api_key, retry=0):
+    """Try to connect to NGC."""
+    url = "https://authn.nvidia.com/token?service=ngc"
+    headers = {"Accept": "application/json", "Authorization": "ApiKey " + api_key}
+    if has_requests:
+        response = requests_get(url, headers=headers)
+        if not response.ok:
+            # retry 3 times, if failed, raise an error.
+            if retry < 3:
+                logger.info(f"Retrying {retry} time(s) to GET {url}.")
+                return _get_ngc_token(url, retry + 1)
+            raise RuntimeError("NGC API response is not ok. Failed to get token.")
+        else:
+            token = response.json()["token"]
+        return token
+
+
 def _get_latest_bundle_version_monaihosting(name):
     url = "https://api.ngc.nvidia.com/v2/models/nvidia/monaihosting"
     full_url = f"{url}/{name.lower()}"
@@ -227,12 +274,28 @@ def _get_latest_bundle_version_monaihosting(name):
         resp = requests_get(full_url)
         resp.raise_for_status()
     else:
-        raise ValueError("NGC API requires requests package.  Please install it.")
+        raise ValueError("NGC API requires requests package. Please install it.")
     model_info = json.loads(resp.text)
     return model_info["model"]["latestVersionIdStr"]
 
 
-def _get_latest_bundle_version(source: str, name: str, repo: str) -> dict[str, list[str] | str] | Any | None:
+def _get_latest_bundle_version_private_registry(name, repo, headers=None):
+    url = f"https://api.ngc.nvidia.com/v2/{repo}/models"
+    full_url = f"{url}/{name.lower()}"
+    requests_get, has_requests = optional_import("requests", name="get")
+    if has_requests:
+        headers = {} if headers is None else headers
+        resp = requests_get(full_url, headers=headers)
+        resp.raise_for_status()
+    else:
+        raise ValueError("NGC API requires requests package. Please install it.")
+    model_info = json.loads(resp.text)
+    return model_info["model"]["latestVersionIdStr"]
+
+
+def _get_latest_bundle_version(
+    source: str, name: str, repo: str, **kwargs: Any
+) -> dict[str, list[str] | str] | Any | None:
     if source == "ngc":
         name = _add_ngc_prefix(name)
         model_dict = _get_all_ngc_models(name)
@@ -242,6 +305,10 @@ def _get_latest_bundle_version(source: str, name: str, repo: str) -> dict[str, l
         return None
     elif source == "monaihosting":
         return _get_latest_bundle_version_monaihosting(name)
+    elif source == "ngc_private":
+        headers = kwargs.pop("headers", {})
+        name = _add_ngc_prefix(name)
+        return _get_latest_bundle_version_private_registry(name, repo, headers)
     elif source == "github":
         repo_owner, repo_name, tag_name = repo.split("/")
         return get_bundle_versions(name, repo=f"{repo_owner}/{repo_name}", tag=tag_name)["latest_version"]
@@ -308,6 +375,9 @@ def download(
         # Execute this module as a CLI entry, and download bundle via URL:
         python -m monai.bundle download --name <bundle_name> --url <url>
 
+        # Execute this module as a CLI entry, and download bundle from ngc_private with latest version:
+        python -m monai.bundle download --name <bundle_name> --source "ngc_private" --bundle_dir "./" --repo "org/org_name"
+
         # Set default args of `run` in a JSON / YAML file, help to record and simplify the command line.
         # Other args still can override the default args at runtime.
         # The content of the JSON / YAML file is a dictionary. For example:
@@ -328,10 +398,13 @@ def download(
             Default is `bundle` subfolder under `torch.hub.get_dir()`.
         source: storage location name. This argument is used when `url` is `None`.
             In default, the value is achieved from the environment variable BUNDLE_DOWNLOAD_SRC, and
-            it should be "ngc", "monaihosting", "github", or "huggingface_hub".
+            it should be "ngc", "monaihosting", "github", "ngc_private", or "huggingface_hub".
+            If source is "ngc_private", you need specify the NGC_API_KEY in the environment variable.
         repo: repo name. This argument is used when `url` is `None` and `source` is "github" or "huggingface_hub".
             If `source` is "github", it should be in the form of "repo_owner/repo_name/release_tag".
             If `source` is "huggingface_hub", it should be in the form of "repo_owner/repo_name".
+            If `source` is "ngc_private", it should be in the form of "org/org_name" or "org/org_name/team/team_name",
+            or you can specify the environment variable NGC_ORG and NGC_TEAM.
         url: url to download the data. If not `None`, data will be downloaded directly
             and `source` will not be checked.
             If `name` is `None`, filename is determined by `monai.apps.utils._basename(url)`.
@@ -363,11 +436,18 @@ def download(
 
     bundle_dir_ = _process_bundle_dir(bundle_dir_)
     if repo_ is None:
-        repo_ = "Project-MONAI/model-zoo/hosting_storage_v1"
-    if len(repo_.split("/")) != 3 and source_ != "huggingface_hub":
-        raise ValueError("repo should be in the form of `repo_owner/repo_name/release_tag`.")
+        org_ = os.getenv("NGC_ORG", None)
+        team_ = os.getenv("NGC_TEAM", None)
+        if org_ is not None:
+            repo_ = f"org/{org_}/team/{team_}" if team_ is not None else f"org/{org_}"
+        else:
+            repo_ = "Project-MONAI/model-zoo/hosting_storage_v1"
+    if len(repo_.split("/")) not in (2, 4) and source_ == "ngc_private":
+        raise ValueError(f"repo should be in the form of `org/org_name/team/team_name` or `org/org_name`, got {repo_}.")
+    if len(repo_.split("/")) != 3 and source_ == "github":
+        raise ValueError(f"repo should be in the form of `repo_owner/repo_name/release_tag`, got {repo_}.")
     elif len(repo_.split("/")) != 2 and source_ == "huggingface_hub":
-        raise ValueError("Hugging Face Hub repo should be in the form of `repo_owner/repo_name`")
+        raise ValueError(f"Hugging Face Hub repo should be in the form of `repo_owner/repo_name`, got {repo_}.")
     if url_ is not None:
         if name_ is not None:
             filepath = bundle_dir_ / f"{name_}.zip"
@@ -376,10 +456,19 @@ def download(
         download_url(url=url_, filepath=filepath, hash_val=None, progress=progress_)
         extractall(filepath=filepath, output_dir=bundle_dir_, has_base=True)
     else:
+        headers = {}
         if name_ is None:
             raise ValueError(f"To download from source: {source_}, `name` must be provided.")
+        if source == "ngc_private":
+            api_key = os.getenv("NGC_API_KEY", None)
+            if api_key is None:
+                raise ValueError("API key is required for ngc_private source.")
+            else:
+                token = _get_ngc_token(api_key)
+                headers = {"Authorization": f"Bearer {token}"}
+
         if version_ is None:
-            version_ = _get_latest_bundle_version(source=source_, name=name_, repo=repo_)
+            version_ = _get_latest_bundle_version(source=source_, name=name_, repo=repo_, headers=headers)
         if source_ == "github":
             if version_ is not None:
                 name_ = "_v".join([name_, version_])
@@ -394,6 +483,15 @@ def download(
                 remove_prefix=remove_prefix_,
                 progress=progress_,
             )
+        elif source_ == "ngc_private":
+            _download_from_ngc_private(
+                download_path=bundle_dir_,
+                filename=name_,
+                version=version_,
+                remove_prefix=remove_prefix_,
+                repo=repo_,
+                headers=headers,
+            )
         elif source_ == "huggingface_hub":
             extract_path = os.path.join(bundle_dir_, name_)
             huggingface_hub.snapshot_download(repo_id=repo_, revision=version_, local_dir=extract_path)
diff --git a/tests/test_bundle_download.py b/tests/test_bundle_download.py
index 89fbe5e8b2..fe7caf5c17 100644
--- a/tests/test_bundle_download.py
+++ b/tests/test_bundle_download.py
@@ -56,7 +56,7 @@
 TEST_CASE_5 = [
     ["models/model.pt", "models/model.ts", "configs/train.json"],
     "brats_mri_segmentation",
-    "https://api.ngc.nvidia.com/v2/models/nvidia/monaihosting/brats_mri_segmentation/versions/0.3.9/files/brats_mri_segmentation_v0.3.9.zip",
+    "https://api.ngc.nvidia.com/v2/models/nvidia/monaihosting/brats_mri_segmentation/versions/0.4.0/files/brats_mri_segmentation_v0.4.0.zip",
 ]
 
 TEST_CASE_6 = [["models/model.pt", "configs/train.json"], "renalStructures_CECT_segmentation", "0.1.0"]
@@ -173,6 +173,23 @@ def test_monaihosting_url_download_bundle(self, bundle_files, bundle_name, url):
                     file_path = os.path.join(tempdir, bundle_name, file)
                     self.assertTrue(os.path.exists(file_path))
 
+    @parameterized.expand([TEST_CASE_5])
+    @skip_if_quick
+    def test_ngc_private_source_download_bundle(self, bundle_files, bundle_name, _url):
+        with skip_if_downloading_fails():
+            # download a single file from url, also use `args_file`
+            with tempfile.TemporaryDirectory() as tempdir:
+                def_args = {"name": bundle_name, "bundle_dir": tempdir}
+                def_args_file = os.path.join(tempdir, "def_args.json")
+                parser = ConfigParser()
+                parser.export_config_file(config=def_args, filepath=def_args_file)
+                cmd = ["coverage", "run", "-m", "monai.bundle", "download", "--args_file", def_args_file]
+                cmd += ["--progress", "False", "--source", "ngc_private"]
+                command_line_tests(cmd)
+                for file in bundle_files:
+                    file_path = os.path.join(tempdir, bundle_name, file)
+                    self.assertTrue(os.path.exists(file_path))
+
     @parameterized.expand([TEST_CASE_6])
     @skip_if_quick
     def test_monaihosting_source_download_bundle(self, bundle_files, bundle_name, version):

From bdbfa3e4c52766b53f6419eacc2c3490a2093759 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 18 Jul 2024 13:37:45 +0800
Subject: [PATCH 093/183] Revert change in blossom-ci yml (#7927)

Revert change in blossom-ci.yml

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/blossom-ci.yml | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/.github/workflows/blossom-ci.yml b/.github/workflows/blossom-ci.yml
index 736d4bfdf6..bf507bab3b 100644
--- a/.github/workflows/blossom-ci.yml
+++ b/.github/workflows/blossom-ci.yml
@@ -93,9 +93,6 @@ jobs:
         run: blossom-ci
         env:
           OPERATION: 'START-CI-JOB'
-          NGC_API_KEY: ${{ secrets.NGC_API_KEY }}
-          NGC_ORG: ${{ secrets.NGC_ORG }}
-          NGC_TEAM: ${{ secrets.NGC_TEAM }}
           CI_SERVER: ${{ secrets.CI_SERVER }}
           REPO_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 

From 85ab9f4542198ab00cabac92b6b33ca0c3ff89fa Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 18 Jul 2024 14:52:00 +0800
Subject: [PATCH 094/183] Try to fix experiment already exist issue in
 `MLFlowHandler` (#7916)

Try to fixes https://github.com/NVIDIA/NVFlare/issues/2698.


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/handlers/mlflow_handler.py | 25 +++++++++++++++++++++----
 1 file changed, 21 insertions(+), 4 deletions(-)

diff --git a/monai/handlers/mlflow_handler.py b/monai/handlers/mlflow_handler.py
index df209c1c8b..6d19579d9e 100644
--- a/monai/handlers/mlflow_handler.py
+++ b/monai/handlers/mlflow_handler.py
@@ -21,6 +21,7 @@
 import torch
 from torch.utils.data import Dataset
 
+from monai.apps.utils import get_logger
 from monai.config import IgniteInfo
 from monai.utils import CommonKeys, ensure_tuple, min_version, optional_import
 
@@ -29,6 +30,9 @@
 mlflow.entities, _ = optional_import(
     "mlflow.entities", descriptor="Please install mlflow.entities before using MLFlowHandler."
 )
+MlflowException, _ = optional_import(
+    "mlflow.exceptions", name="MlflowException", descriptor="Please install mlflow before using MLFlowHandler."
+)
 pandas, _ = optional_import("pandas", descriptor="Please install pandas for recording the dataset.")
 tqdm, _ = optional_import("tqdm", "4.47.0", min_version, "tqdm")
 
@@ -41,6 +45,8 @@
 
 DEFAULT_TAG = "Loss"
 
+logger = get_logger(module_name=__name__)
+
 
 class MLFlowHandler:
     """
@@ -236,10 +242,21 @@ def start(self, engine: Engine) -> None:
     def _set_experiment(self):
         experiment = self.experiment
         if not experiment:
-            experiment = self.client.get_experiment_by_name(self.experiment_name)
-            if not experiment:
-                experiment_id = self.client.create_experiment(self.experiment_name)
-                experiment = self.client.get_experiment(experiment_id)
+            for _retry_time in range(3):
+                try:
+                    experiment = self.client.get_experiment_by_name(self.experiment_name)
+                    if not experiment:
+                        experiment_id = self.client.create_experiment(self.experiment_name)
+                        experiment = self.client.get_experiment(experiment_id)
+                    break
+                except MlflowException as e:
+                    if "RESOURCE_ALREADY_EXISTS" in str(e):
+                        logger.warning("Experiment already exists; delaying before retrying.")
+                        time.sleep(1)
+                        if _retry_time == 2:
+                            raise e
+                    else:
+                        raise e
 
         if experiment.lifecycle_stage != mlflow.entities.LifecycleStage.ACTIVE:
             raise ValueError(f"Cannot set a deleted experiment '{self.experiment_name}' as the active experiment")

From 7e4f14180925f81a77c9f4d5d1858a7fe1e6aaec Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 18 Jul 2024 16:17:14 +0800
Subject: [PATCH 095/183] Fix incorrect repo name during bundle download
 (#7929)

Fixes #7928

### Description

Only use env var when source is `ngc_private`.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/bundle/scripts.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index 975b6cbdbd..56146546e8 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -438,7 +438,7 @@ def download(
     if repo_ is None:
         org_ = os.getenv("NGC_ORG", None)
         team_ = os.getenv("NGC_TEAM", None)
-        if org_ is not None:
+        if org_ is not None and source_ == "ngc_private":
             repo_ = f"org/{org_}/team/{team_}" if team_ is not None else f"org/{org_}"
         else:
             repo_ = "Project-MONAI/model-zoo/hosting_storage_v1"

From 46e2b0eb9a7c8249177df775047fa098af26f6a6 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 18 Jul 2024 19:49:20 +0800
Subject: [PATCH 096/183] Fix load pretrain weight issue in ResNet (#7924)

Fixes #7923


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/networks/nets/resnet.py | 7 +++----
 tests/test_resnet.py          | 8 +++-----
 2 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/monai/networks/nets/resnet.py b/monai/networks/nets/resnet.py
index 6e61db07ca..d62722478e 100644
--- a/monai/networks/nets/resnet.py
+++ b/monai/networks/nets/resnet.py
@@ -510,7 +510,7 @@ def _resnet(
                     # Check model bias_downsample and shortcut_type
                     bias_downsample, shortcut_type = get_medicalnet_pretrained_resnet_args(resnet_depth)
                     if shortcut_type == kwargs.get("shortcut_type", "B") and (
-                        bool(bias_downsample) == kwargs.get("bias_downsample", False) if bias_downsample != -1 else True
+                        bias_downsample == kwargs.get("bias_downsample", True)
                     ):
                         # Download the MedicalNet pretrained model
                         model_state_dict = get_pretrained_resnet_medicalnet(
@@ -518,8 +518,7 @@ def _resnet(
                         )
                     else:
                         raise NotImplementedError(
-                            f"Please set shortcut_type to {shortcut_type} and bias_downsample to"
-                            f"{bool(bias_downsample) if bias_downsample!=-1 else 'True or False'}"
+                            f"Please set shortcut_type to {shortcut_type} and bias_downsample to {bias_downsample} "
                             f"when using pretrained MedicalNet resnet{resnet_depth}"
                         )
                 else:
@@ -681,7 +680,7 @@ def get_medicalnet_pretrained_resnet_args(resnet_depth: int):
     # After testing
     # False: 10, 50, 101, 152, 200
     # Any: 18, 34
-    bias_downsample = -1 if resnet_depth in [18, 34] else 0  # 18, 10, 34
+    bias_downsample = resnet_depth in (18, 34)
     shortcut_type = "A" if resnet_depth in [18, 34] else "B"
     return bias_downsample, shortcut_type
 
diff --git a/tests/test_resnet.py b/tests/test_resnet.py
index e873f1238a..a55d18f5de 100644
--- a/tests/test_resnet.py
+++ b/tests/test_resnet.py
@@ -266,7 +266,7 @@ def test_resnet_shape(self, model, input_param, input_shape, expected_shape):
     @parameterized.expand(PRETRAINED_TEST_CASES)
     @skip_if_quick
     @skip_if_no_cuda
-    def test_resnet_pretrained(self, model, input_param, input_shape, expected_shape):
+    def test_resnet_pretrained(self, model, input_param, _input_shape, _expected_shape):
         net = model(**input_param).to(device)
         # Save ckpt
         torch.save(net.state_dict(), self.tmp_ckpt_filename)
@@ -290,9 +290,7 @@ def test_resnet_pretrained(self, model, input_param, input_shape, expected_shape
                 and input_param.get("n_input_channels", 3) == 1
                 and input_param.get("feed_forward", True) is False
                 and input_param.get("shortcut_type", "B") == shortcut_type
-                and (
-                    input_param.get("bias_downsample", True) == bool(bias_downsample) if bias_downsample != -1 else True
-                )
+                and (input_param.get("bias_downsample", True) == bias_downsample)
             ):
                 model(**cp_input_param)
             else:
@@ -303,7 +301,7 @@ def test_resnet_pretrained(self, model, input_param, input_shape, expected_shape
             cp_input_param["n_input_channels"] = 1
             cp_input_param["feed_forward"] = False
             cp_input_param["shortcut_type"] = shortcut_type
-            cp_input_param["bias_downsample"] = bool(bias_downsample) if bias_downsample != -1 else True
+            cp_input_param["bias_downsample"] = bias_downsample
             if cp_input_param.get("spatial_dims", 3) == 3:
                 with skip_if_downloading_fails():
                     pretrained_net = model(**cp_input_param).to(device)

From d020facccbd3afe979fce68c24703dcda47234f6 Mon Sep 17 00:00:00 2001
From: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Date: Fri, 19 Jul 2024 17:10:36 +0100
Subject: [PATCH 097/183] Merge genaidev Into Dev (#7886)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes #6676 .

### Description

This merges the Generative Models code into dev. Everything has been
checked by the generative team, tests all pass, and the changes that
have been done recently are integrated. This is ready to merge.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Mark Graham <markgraham539@gmail.com>
Signed-off-by: KumoLiu <yunl@nvidia.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Signed-off-by: Mark Graham <mark@Marks-MacBook-Pro.local>
Signed-off-by: vgrau98 <victor.grau93@gmail.com>
Signed-off-by: vgrau98 <35843843+vgrau98@users.noreply.github.com>
Signed-off-by: Wenqi Li <wenqil@nvidia.com>
Signed-off-by: dongy <dongy@nvidia.com>
Signed-off-by: myron <amyronenko@nvidia.com>
Signed-off-by: kaibo <ktang@unc.edu>
Signed-off-by: monai-bot <monai.miccai2019@gmail.com>
Signed-off-by: elitap <elias.tappeiner@gmx.at>
Signed-off-by: Felix Schnabel <f.schnabel@tum.de>
Signed-off-by: YanxuanLiu <yanxuanl@nvidia.com>
Signed-off-by: ytl0623 <david89062388@gmail.com>
Signed-off-by: Dženan Zukić <dzenan.zukic@kitware.com>
Signed-off-by: Ishan Dutta <ishandutta0098@gmail.com>
Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: heyufan1995 <heyufan1995@gmail.com>
Signed-off-by: binliu <binliu@nvidia.com>
Signed-off-by: axel.vlaminck <axel.vlaminck@gmail.com>
Signed-off-by: Ibrahim Hadzic <ibrahimhadzic45@gmail.com>
Signed-off-by: Behrooz <3968947+drbeh@users.noreply.github.com>
Signed-off-by: Timothy Baker <bakertim@umich.edu>
Signed-off-by: Mathijs de Boer <m.deboer-41@umcutrecht.nl>
Signed-off-by: Fabian Klopfer <fabian.klopfer@ieee.org>
Signed-off-by: Lucas Robinet <robinet.lucas@iuct-oncopole.fr>
Signed-off-by: Lucas Robinet <67736918+Lucas-rbnt@users.noreply.github.com>
Signed-off-by: chaoliu <chaoliu@nvidia.com>
Signed-off-by: cxlcl <chaoliucxl@gmail.com>
Signed-off-by: chaoliu <chaoliucxl@gmail.com>
Signed-off-by: Suraj Pai <b.pai@maastrichtuniversity.nl>
Signed-off-by: Juan Pablo de la Cruz Gutiérrez <juampatronics@gmail.com>
Signed-off-by: John Zielke <j.l.zielke@gmail.com>
Signed-off-by: Mingxin Zheng <mingxinz@nvidia.com>
Signed-off-by: Vladimir Chernyi <57420464+scalyvladimir@users.noreply.github.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: Szabolcs Botond Lorincz Molnar <szabolcs.lorincz@mvision.ai>
Signed-off-by: Lucas Robinet <luca.robinet@gmail.com>
Signed-off-by: Mingxin <mingxinz@nvidia.com>
Signed-off-by: Han Wang <freddie.wanah@gmail.com>
Signed-off-by: Konstantin Sukharev <k.sukharev@gmail.com>
Signed-off-by: Ben Murray <ben.murray@gmail.com>
Signed-off-by: Matthew Vine <32849887+MattTheCuber@users.noreply.github.com>
Signed-off-by: Peter Kaplinsky <peterkaplinsky@gmail.com>
Signed-off-by: Simon Jensen <61684806+simojens@users.noreply.github.com>
Signed-off-by: NabJa <nabil.jabareen@gmail.com>
Signed-off-by: virginiafdez <virginia.fernandez@kcl.ac.uk>
Signed-off-by: Eric Kerfoot <eric.kerfoot@kcl.ac.uk>
Signed-off-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: Mark Graham <markgraham539@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: KumoLiu <yunl@nvidia.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: vgrau98 <35843843+vgrau98@users.noreply.github.com>
Co-authored-by: Wenqi Li <831580+wyli@users.noreply.github.com>
Co-authored-by: Dong Yang <don.yang.mech@gmail.com>
Co-authored-by: myron <myron@users.noreply.github.com>
Co-authored-by: Kaibo Tang <99367900+kvttt@users.noreply.github.com>
Co-authored-by: monai-bot <64792179+monai-bot@users.noreply.github.com>
Co-authored-by: elitap <elitap@users.noreply.github.com>
Co-authored-by: Felix Schnabel <f.schnabel@tum.de>
Co-authored-by: YanxuanLiu <104543031+YanxuanLiu@users.noreply.github.com>
Co-authored-by: ytl0623 <david89062388@gmail.com>
Co-authored-by: Dženan Zukić <dzenan.zukic@kitware.com>
Co-authored-by: Ishan Dutta <ishandutta0098@gmail.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Kaibo Tang <ktang@unc.edu>
Co-authored-by: Yufan He <59374597+heyufan1995@users.noreply.github.com>
Co-authored-by: binliunls <107988372+binliunls@users.noreply.github.com>
Co-authored-by: Ben Murray <ben.murray@gmail.com>
Co-authored-by: axel.vlaminck <axel.vlaminck@gmail.com>
Co-authored-by: Mingxin Zheng <18563433+mingxin-zheng@users.noreply.github.com>
Co-authored-by: Ibrahim Hadzic <ibrahimhadzic45@gmail.com>
Co-authored-by: Dr. Behrooz Hashemian <3968947+drbeh@users.noreply.github.com>
Co-authored-by: Timothy J. Baker <62781117+tim-the-baker@users.noreply.github.com>
Co-authored-by: Mathijs de Boer <8137653+MathijsdeBoer@users.noreply.github.com>
Co-authored-by: Mathijs de Boer <m.deboer-41@umcutrecht.nl>
Co-authored-by: Fabian Klopfer <fabian.klopfer@ieee.org>
Co-authored-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Co-authored-by: Lucas Robinet <67736918+Lucas-rbnt@users.noreply.github.com>
Co-authored-by: Lucas Robinet <robinet.lucas@iuct-oncopole.fr>
Co-authored-by: cxlcl <chaoliucxl@gmail.com>
Co-authored-by: Suraj Pai <b.pai@maastrichtuniversity.nl>
Co-authored-by: Juampa <1523654+juampatronics@users.noreply.github.com>
Co-authored-by: johnzielke <j.l.zielke@gmail.com>
Co-authored-by: Vladimir Chernyi <57420464+scalyvladimir@users.noreply.github.com>
Co-authored-by: Lőrincz-Molnár Szabolcs-Botond <lorincz_szabolcs@yahoo.com>
Co-authored-by: Nic Ma <nma@nvidia.com>
Co-authored-by: Lucas Robinet <luca.robinet@gmail.com>
Co-authored-by: Han Wang <freddie.wanah@gmail.com>
Co-authored-by: Konstantin Sukharev <50718389+k-sukharev@users.noreply.github.com>
Co-authored-by: Matthew Vine <32849887+MattTheCuber@users.noreply.github.com>
Co-authored-by: Pkaps25 <43655728+Pkaps25@users.noreply.github.com>
Co-authored-by: Peter Kaplinsky <peterkaplinsky@gmail.com>
Co-authored-by: Simon Jensen <61684806+simojens@users.noreply.github.com>
Co-authored-by: NabJa <32510324+NabJa@users.noreply.github.com>
Co-authored-by: Virginia Fernandez <61539159+virginiafdez@users.noreply.github.com>
Co-authored-by: virginiafdez <virginia.fernandez@kcl.ac.uk>
Co-authored-by: Yu <146002968+Yu0610@users.noreply.github.com>
---
 docs/source/engines.rst                       |    5 +
 docs/source/inferers.rst                      |   23 +
 docs/source/utils.rst                         |    5 +
 monai/apps/detection/utils/anchor_utils.py    |    4 +-
 monai/apps/pathology/transforms/post/array.py |    1 +
 monai/bundle/utils.py                         |    1 +
 monai/data/dataset_summary.py                 |    1 +
 monai/data/utils.py                           |   15 +-
 monai/engines/__init__.py                     |    4 +-
 monai/engines/trainer.py                      |  283 ++-
 monai/engines/utils.py                        |   77 +-
 monai/inferers/__init__.py                    |    5 +
 monai/inferers/inferer.py                     | 1280 ++++++++++-
 monai/networks/blocks/__init__.py             |    3 +
 monai/networks/blocks/attention_utils.py      |  128 ++
 monai/networks/blocks/crossattention.py       |  166 ++
 monai/networks/blocks/rel_pos_embedding.py    |   56 +
 monai/networks/blocks/selfattention.py        |   77 +-
 monai/networks/blocks/spade_norm.py           |   95 +
 monai/networks/blocks/spatialattention.py     |   82 +
 monai/networks/blocks/transformerblock.py     |   28 +-
 monai/networks/blocks/upsample.py             |   32 +-
 monai/networks/layers/__init__.py             |    3 +-
 monai/networks/layers/factories.py            |   13 +-
 monai/networks/layers/utils.py                |   15 +-
 monai/networks/layers/vector_quantizer.py     |  233 ++
 monai/networks/nets/__init__.py               |    9 +
 monai/networks/nets/autoencoderkl.py          |  702 ++++++
 monai/networks/nets/controlnet.py             |  465 ++++
 monai/networks/nets/diffusion_model_unet.py   | 1913 +++++++++++++++++
 monai/networks/nets/patchgan_discriminator.py |  230 ++
 monai/networks/nets/quicknat.py               |    2 +
 monai/networks/nets/spade_autoencoderkl.py    |  480 +++++
 .../nets/spade_diffusion_model_unet.py        |  934 ++++++++
 monai/networks/nets/spade_network.py          |  435 ++++
 monai/networks/nets/swin_unetr.py             |    7 +-
 monai/networks/nets/transformer.py            |  157 ++
 monai/networks/nets/vqvae.py                  |  472 ++++
 monai/networks/schedulers/__init__.py         |   17 +
 monai/networks/schedulers/ddim.py             |  294 +++
 monai/networks/schedulers/ddpm.py             |  250 +++
 monai/networks/schedulers/pndm.py             |  316 +++
 monai/networks/schedulers/scheduler.py        |  205 ++
 monai/networks/utils.py                       |   22 +
 monai/transforms/regularization/array.py      |    4 +
 .../transforms/utils_create_transform_ims.py  |    6 +-
 monai/utils/__init__.py                       |    1 +
 monai/utils/misc.py                           |    5 +-
 monai/utils/ordering.py                       |  207 ++
 tests/hvd_evenly_divisible_all_gather.py      |    8 +-
 tests/min_tests.py                            |    1 +
 tests/test_autoencoderkl.py                   |  337 +++
 tests/test_controlnet.py                      |  215 ++
 tests/test_controlnet_inferers.py             | 1310 +++++++++++
 tests/test_crossattention.py                  |  131 ++
 tests/test_diffusion_inferer.py               |  236 ++
 tests/test_diffusion_model_unet.py            |  585 +++++
 tests/test_ensure_channel_first.py            |    7 +-
 tests/test_ensure_channel_firstd.py           |    7 +-
 .../test_evenly_divisible_all_gather_dist.py  |    8 +-
 tests/test_handler_metrics_saver_dist.py      |    4 +-
 tests/test_hilbert_transform.py               |  123 +-
 tests/test_integration_unet_2d.py             |    1 +
 .../test_integration_workflows_adversarial.py |  173 ++
 tests/test_latent_diffusion_inferer.py        |  824 +++++++
 tests/test_ordering.py                        |  289 +++
 tests/test_patch_gan_dicriminator.py          |  179 ++
 tests/test_prepare_batch_diffusion.py         |  104 +
 tests/test_reg_loss_integration.py            |    3 +
 tests/test_scheduler_ddim.py                  |   83 +
 tests/test_scheduler_ddpm.py                  |  104 +
 tests/test_scheduler_pndm.py                  |  108 +
 tests/test_selfattention.py                   |   42 +-
 tests/test_spade_autoencoderkl.py             |  295 +++
 tests/test_spade_diffusion_model_unet.py      |  574 +++++
 tests/test_spade_vaegan.py                    |  140 ++
 tests/test_spatialattention.py                |   55 +
 tests/test_synthetic.py                       |    2 +-
 tests/test_transformer.py                     |  109 +
 tests/test_transformerblock.py                |   29 +-
 tests/test_vector_quantizer.py                |   89 +
 tests/test_vis_cam.py                         |    3 +
 tests/test_vis_gradcam.py                     |    2 +
 tests/test_vqvae.py                           |  274 +++
 tests/test_vqvaetransformer_inferer.py        |  295 +++
 tests/testing_data/data_config.json           |   20 +
 86 files changed, 16359 insertions(+), 178 deletions(-)
 create mode 100644 monai/networks/blocks/attention_utils.py
 create mode 100644 monai/networks/blocks/crossattention.py
 create mode 100644 monai/networks/blocks/rel_pos_embedding.py
 create mode 100644 monai/networks/blocks/spade_norm.py
 create mode 100644 monai/networks/blocks/spatialattention.py
 create mode 100644 monai/networks/layers/vector_quantizer.py
 create mode 100644 monai/networks/nets/autoencoderkl.py
 create mode 100644 monai/networks/nets/controlnet.py
 create mode 100644 monai/networks/nets/diffusion_model_unet.py
 create mode 100644 monai/networks/nets/patchgan_discriminator.py
 create mode 100644 monai/networks/nets/spade_autoencoderkl.py
 create mode 100644 monai/networks/nets/spade_diffusion_model_unet.py
 create mode 100644 monai/networks/nets/spade_network.py
 create mode 100644 monai/networks/nets/transformer.py
 create mode 100644 monai/networks/nets/vqvae.py
 create mode 100644 monai/networks/schedulers/__init__.py
 create mode 100644 monai/networks/schedulers/ddim.py
 create mode 100644 monai/networks/schedulers/ddpm.py
 create mode 100644 monai/networks/schedulers/pndm.py
 create mode 100644 monai/networks/schedulers/scheduler.py
 create mode 100644 monai/utils/ordering.py
 create mode 100644 tests/test_autoencoderkl.py
 create mode 100644 tests/test_controlnet.py
 create mode 100644 tests/test_controlnet_inferers.py
 create mode 100644 tests/test_crossattention.py
 create mode 100644 tests/test_diffusion_inferer.py
 create mode 100644 tests/test_diffusion_model_unet.py
 create mode 100644 tests/test_integration_workflows_adversarial.py
 create mode 100644 tests/test_latent_diffusion_inferer.py
 create mode 100644 tests/test_ordering.py
 create mode 100644 tests/test_patch_gan_dicriminator.py
 create mode 100644 tests/test_prepare_batch_diffusion.py
 create mode 100644 tests/test_scheduler_ddim.py
 create mode 100644 tests/test_scheduler_ddpm.py
 create mode 100644 tests/test_scheduler_pndm.py
 create mode 100644 tests/test_spade_autoencoderkl.py
 create mode 100644 tests/test_spade_diffusion_model_unet.py
 create mode 100644 tests/test_spade_vaegan.py
 create mode 100644 tests/test_spatialattention.py
 create mode 100644 tests/test_transformer.py
 create mode 100644 tests/test_vector_quantizer.py
 create mode 100644 tests/test_vqvae.py
 create mode 100644 tests/test_vqvaetransformer_inferer.py

diff --git a/docs/source/engines.rst b/docs/source/engines.rst
index afb2682822..a015c7b2a3 100644
--- a/docs/source/engines.rst
+++ b/docs/source/engines.rst
@@ -30,6 +30,11 @@ Workflows
 .. autoclass:: GanTrainer
     :members:
 
+`AdversarialTrainer`
+~~~~~~~~~~~~~~~~~~~~
+.. autoclass:: AdversarialTrainer
+    :members:
+
 `Evaluator`
 ~~~~~~~~~~~
 .. autoclass:: Evaluator
diff --git a/docs/source/inferers.rst b/docs/source/inferers.rst
index 33f9e14d83..326f56e96c 100644
--- a/docs/source/inferers.rst
+++ b/docs/source/inferers.rst
@@ -49,6 +49,29 @@ Inferers
     :members:
     :special-members: __call__
 
+`DiffusionInferer`
+~~~~~~~~~~~~~~~~~~
+.. autoclass:: DiffusionInferer
+    :members:
+    :special-members: __call__
+
+`LatentDiffusionInferer`
+~~~~~~~~~~~~~~~~~~~~~~~~
+.. autoclass:: LatentDiffusionInferer
+    :members:
+    :special-members: __call__
+
+`ControlNetDiffusionInferer`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autoclass:: ControlNetDiffusionInferer
+    :members:
+    :special-members: __call__
+
+`ControlNetLatentDiffusionInferer`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+.. autoclass:: ControlNetLatentDiffusionInferer
+    :members:
+    :special-members: __call__
 
 Splitters
 ---------
diff --git a/docs/source/utils.rst b/docs/source/utils.rst
index 527247799f..fef671e1f8 100644
--- a/docs/source/utils.rst
+++ b/docs/source/utils.rst
@@ -81,3 +81,8 @@ Component store
 ---------------
 .. autoclass:: monai.utils.component_store.ComponentStore
   :members:
+
+Ordering
+--------
+.. automodule:: monai.utils.ordering
+  :members:
diff --git a/monai/apps/detection/utils/anchor_utils.py b/monai/apps/detection/utils/anchor_utils.py
index 283169b653..cbde3ebae9 100644
--- a/monai/apps/detection/utils/anchor_utils.py
+++ b/monai/apps/detection/utils/anchor_utils.py
@@ -189,7 +189,7 @@ def generate_anchors(
             w_ratios = 1 / area_scale
             h_ratios = area_scale
         # if 3d, w:h:d = 1:aspect_ratios[:,0]:aspect_ratios[:,1]
-        elif self.spatial_dims == 3:
+        else:
             area_scale = torch.pow(aspect_ratios_t[:, 0] * aspect_ratios_t[:, 1], 1 / 3.0)
             w_ratios = 1 / area_scale
             h_ratios = aspect_ratios_t[:, 0] / area_scale
@@ -199,7 +199,7 @@ def generate_anchors(
         hs = (h_ratios[:, None] * scales_t[None, :]).view(-1)
         if self.spatial_dims == 2:
             base_anchors = torch.stack([-ws, -hs, ws, hs], dim=1) / 2.0
-        elif self.spatial_dims == 3:
+        else:  # elif self.spatial_dims == 3:
             ds = (d_ratios[:, None] * scales_t[None, :]).view(-1)
             base_anchors = torch.stack([-ws, -hs, -ds, ws, hs, ds], dim=1) / 2.0
 
diff --git a/monai/apps/pathology/transforms/post/array.py b/monai/apps/pathology/transforms/post/array.py
index 42ca385fa0..0aa8e14655 100644
--- a/monai/apps/pathology/transforms/post/array.py
+++ b/monai/apps/pathology/transforms/post/array.py
@@ -379,6 +379,7 @@ def _generate_contour_coord(self, current: np.ndarray, previous: np.ndarray) ->
         """
 
         p_delta = (current[0] - previous[0], current[1] - previous[1])
+        row, col = -1, -1
 
         if p_delta in ((0.0, 1.0), (0.5, 0.5), (1.0, 0.0)):
             row = int(current[0] + 0.5)
diff --git a/monai/bundle/utils.py b/monai/bundle/utils.py
index a0f39d236f..0f17422ba5 100644
--- a/monai/bundle/utils.py
+++ b/monai/bundle/utils.py
@@ -221,6 +221,7 @@ def load_bundle_config(bundle_path: str, *config_names: str, **load_kw_args: Any
                 raise ValueError(f"Cannot find config file '{full_cname}'")
 
             ardata = archive.read(full_cname)
+            cdata = {}
 
             if full_cname.lower().endswith("json"):
                 cdata = json.loads(ardata, **load_kw_args)
diff --git a/monai/data/dataset_summary.py b/monai/data/dataset_summary.py
index 769ae33b46..5b9e32afca 100644
--- a/monai/data/dataset_summary.py
+++ b/monai/data/dataset_summary.py
@@ -84,6 +84,7 @@ def collect_meta_data(self):
         """
 
         for data in self.data_loader:
+            meta_dict = {}
             if isinstance(data[self.image_key], MetaTensor):
                 meta_dict = data[self.image_key].meta
             elif self.meta_key in data:
diff --git a/monai/data/utils.py b/monai/data/utils.py
index 585f02ec9e..7a08300abb 100644
--- a/monai/data/utils.py
+++ b/monai/data/utils.py
@@ -53,10 +53,6 @@
     pytorch_after,
 )
 
-if pytorch_after(1, 13):
-    # import private code for reuse purposes, comment in case things break in the future
-    from torch.utils.data._utils.collate import collate_tensor_fn, default_collate_fn_map
-
 pd, _ = optional_import("pandas")
 DataFrame, _ = optional_import("pandas", name="DataFrame")
 nib, _ = optional_import("nibabel")
@@ -454,8 +450,13 @@ def collate_meta_tensor_fn(batch, *, collate_fn_map=None):
     Collate a sequence of meta tensor into a single batched metatensor. This is called by `collage_meta_tensor`
     and so should not be used as a collate function directly in dataloaders.
     """
-    collate_fn = collate_tensor_fn if pytorch_after(1, 13) else default_collate
-    collated = collate_fn(batch)  # type: ignore
+    if pytorch_after(1, 13):
+        from torch.utils.data._utils.collate import collate_tensor_fn  # imported here for pylint/mypy issues
+
+        collated = collate_tensor_fn(batch)
+    else:
+        collated = default_collate(batch)
+
     meta_dicts = [i.meta or TraceKeys.NONE for i in batch]
     common_ = set.intersection(*[set(d.keys()) for d in meta_dicts if isinstance(d, dict)])
     if common_:
@@ -496,6 +497,8 @@ def list_data_collate(batch: Sequence):
 
     if pytorch_after(1, 13):
         # needs to go here to avoid circular import
+        from torch.utils.data._utils.collate import default_collate_fn_map
+
         from monai.data.meta_tensor import MetaTensor
 
         default_collate_fn_map.update({MetaTensor: collate_meta_tensor_fn})
diff --git a/monai/engines/__init__.py b/monai/engines/__init__.py
index d8dc51f620..93cc40e292 100644
--- a/monai/engines/__init__.py
+++ b/monai/engines/__init__.py
@@ -12,12 +12,14 @@
 from __future__ import annotations
 
 from .evaluator import EnsembleEvaluator, Evaluator, SupervisedEvaluator
-from .trainer import GanTrainer, SupervisedTrainer, Trainer
+from .trainer import AdversarialTrainer, GanTrainer, SupervisedTrainer, Trainer
 from .utils import (
+    DiffusionPrepareBatch,
     IterationEvents,
     PrepareBatch,
     PrepareBatchDefault,
     PrepareBatchExtraInput,
+    VPredictionPrepareBatch,
     default_make_latent,
     default_metric_cmp_fn,
     default_prepare_batch,
diff --git a/monai/engines/trainer.py b/monai/engines/trainer.py
index f1513ea73b..c1364fe015 100644
--- a/monai/engines/trainer.py
+++ b/monai/engines/trainer.py
@@ -24,7 +24,7 @@
 from monai.engines.workflow import Workflow
 from monai.inferers import Inferer, SimpleInferer
 from monai.transforms import Transform
-from monai.utils import GanKeys, min_version, optional_import
+from monai.utils import AdversarialIterationEvents, AdversarialKeys, GanKeys, min_version, optional_import
 from monai.utils.enums import CommonKeys as Keys
 from monai.utils.enums import EngineStatsKeys as ESKeys
 from monai.utils.module import pytorch_after
@@ -37,7 +37,7 @@
     Metric, _ = optional_import("ignite.metrics", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Metric")
     EventEnum, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "EventEnum")
 
-__all__ = ["Trainer", "SupervisedTrainer", "GanTrainer"]
+__all__ = ["Trainer", "SupervisedTrainer", "GanTrainer", "AdversarialTrainer"]
 
 
 class Trainer(Workflow):
@@ -471,3 +471,282 @@ def _iteration(
             GanKeys.GLOSS: g_loss.item(),
             GanKeys.DLOSS: d_total_loss.item(),
         }
+
+
+class AdversarialTrainer(Trainer):
+    """
+    Standard supervised training workflow for adversarial loss enabled neural networks.
+
+    Args:
+        device: an object representing the device on which to run.
+        max_epochs: the total epoch number for engine to run.
+        train_data_loader: Core ignite engines uses `DataLoader` for training loop batchdata.
+        g_network: ''generator'' (G) network architecture.
+        g_optimizer: G optimizer function.
+        g_loss_function: G loss function for adversarial training.
+        recon_loss_function: G loss function for reconstructions.
+        d_network: discriminator (D) network architecture.
+        d_optimizer: D optimizer function.
+        d_loss_function: D loss function for adversarial training..
+        epoch_length: number of iterations for one epoch, default to `len(train_data_loader)`.
+        non_blocking: if True and this copy is between CPU and GPU, the copy may occur asynchronously with respect to
+            the host. For other cases, this argument has no effect.
+        prepare_batch: function to parse image and label for current iteration.
+        iteration_update: the callable function for every iteration, expect to accept `engine` and `batchdata` as input
+            parameters. if not provided, use `self._iteration()` instead.
+        g_inferer: inference method to execute G model forward. Defaults to ``SimpleInferer()``.
+        d_inferer: inference method to execute D model forward. Defaults to ``SimpleInferer()``.
+        postprocessing: execute additional transformation for the model output data. Typically, several Tensor based
+            transforms composed by `Compose`. Defaults to None
+        key_train_metric: compute metric when every iteration completed, and save average value to engine.state.metrics
+            when epoch completed. key_train_metric is the main metric to compare and save the checkpoint into files.
+        additional_metrics: more Ignite metrics that also attach to Ignite Engine.
+        metric_cmp_fn: function to compare current key metric with previous best key metric value, it must accept 2 args
+            (current_metric, previous_best) and return a bool result: if `True`, will update 'best_metric` and
+            `best_metric_epoch` with current metric and epoch, default to `greater than`.
+        train_handlers: every handler is a set of Ignite Event-Handlers, must have `attach` function, like:
+            CheckpointHandler, StatsHandler, etc.
+        amp: whether to enable auto-mixed-precision training, default is False.
+        event_names: additional custom ignite events that will register to the engine.
+            new events can be a list of str or `ignite.engine.events.EventEnum`.
+        event_to_attr: a dictionary to map an event to a state attribute, then add to `engine.state`.
+            for more details, check: https://pytorch.org/ignite/generated/ignite.engine.engine.Engine.html
+            #ignite.engine.engine.Engine.register_events.
+        decollate: whether to decollate the batch-first data to a list of data after model computation, recommend
+            `decollate=True` when `postprocessing` uses components from `monai.transforms`. default to `True`.
+        optim_set_to_none: when calling `optimizer.zero_grad()`, instead of setting to zero, set the grads to None.
+            more details: https://pytorch.org/docs/stable/generated/torch.optim.Optimizer.zero_grad.html.
+        to_kwargs: dict of other args for `prepare_batch` API when converting the input data, except for
+            `device`, `non_blocking`.
+        amp_kwargs: dict of the args for `torch.cuda.amp.autocast()` API, for more details:
+            https://pytorch.org/docs/stable/amp.html#torch.cuda.amp.autocast.
+    """
+
+    def __init__(
+        self,
+        device: torch.device | str,
+        max_epochs: int,
+        train_data_loader: Iterable | DataLoader,
+        g_network: torch.nn.Module,
+        g_optimizer: Optimizer,
+        g_loss_function: Callable,
+        recon_loss_function: Callable,
+        d_network: torch.nn.Module,
+        d_optimizer: Optimizer,
+        d_loss_function: Callable,
+        epoch_length: int | None = None,
+        non_blocking: bool = False,
+        prepare_batch: Callable = default_prepare_batch,
+        iteration_update: Callable | None = None,
+        g_inferer: Inferer | None = None,
+        d_inferer: Inferer | None = None,
+        postprocessing: Transform | None = None,
+        key_train_metric: dict[str, Metric] | None = None,
+        additional_metrics: dict[str, Metric] | None = None,
+        metric_cmp_fn: Callable = default_metric_cmp_fn,
+        train_handlers: Sequence | None = None,
+        amp: bool = False,
+        event_names: list[str | EventEnum | type[EventEnum]] | None = None,
+        event_to_attr: dict | None = None,
+        decollate: bool = True,
+        optim_set_to_none: bool = False,
+        to_kwargs: dict | None = None,
+        amp_kwargs: dict | None = None,
+    ):
+        super().__init__(
+            device=device,
+            max_epochs=max_epochs,
+            data_loader=train_data_loader,
+            epoch_length=epoch_length,
+            non_blocking=non_blocking,
+            prepare_batch=prepare_batch,
+            iteration_update=iteration_update,
+            postprocessing=postprocessing,
+            key_metric=key_train_metric,
+            additional_metrics=additional_metrics,
+            metric_cmp_fn=metric_cmp_fn,
+            handlers=train_handlers,
+            amp=amp,
+            event_names=event_names,
+            event_to_attr=event_to_attr,
+            decollate=decollate,
+            to_kwargs=to_kwargs,
+            amp_kwargs=amp_kwargs,
+        )
+
+        self.register_events(*AdversarialIterationEvents)
+
+        self.state.g_network = g_network
+        self.state.g_optimizer = g_optimizer
+        self.state.g_loss_function = g_loss_function
+        self.state.recon_loss_function = recon_loss_function
+
+        self.state.d_network = d_network
+        self.state.d_optimizer = d_optimizer
+        self.state.d_loss_function = d_loss_function
+
+        self.g_inferer = SimpleInferer() if g_inferer is None else g_inferer
+        self.d_inferer = SimpleInferer() if d_inferer is None else d_inferer
+
+        self.state.g_scaler = torch.cuda.amp.GradScaler() if self.amp else None
+        self.state.d_scaler = torch.cuda.amp.GradScaler() if self.amp else None
+
+        self.optim_set_to_none = optim_set_to_none
+        self._complete_state_dict_user_keys()
+
+    def _complete_state_dict_user_keys(self) -> None:
+        """
+        This method appends to the _state_dict_user_keys AdversarialTrainer's elements that are required for
+        checkpoint saving.
+
+        Follows the example found at:
+            https://pytorch.org/ignite/generated/ignite.engine.engine.Engine.html#ignite.engine.engine.Engine.state_dict
+        """
+        self._state_dict_user_keys.extend(
+            ["g_network", "g_optimizer", "d_network", "d_optimizer", "g_scaler", "d_scaler"]
+        )
+
+        g_loss_state_dict = getattr(self.state.g_loss_function, "state_dict", None)
+        if callable(g_loss_state_dict):
+            self._state_dict_user_keys.append("g_loss_function")
+
+        d_loss_state_dict = getattr(self.state.d_loss_function, "state_dict", None)
+        if callable(d_loss_state_dict):
+            self._state_dict_user_keys.append("d_loss_function")
+
+        recon_loss_state_dict = getattr(self.state.recon_loss_function, "state_dict", None)
+        if callable(recon_loss_state_dict):
+            self._state_dict_user_keys.append("recon_loss_function")
+
+    def _iteration(
+        self, engine: AdversarialTrainer, batchdata: dict[str, torch.Tensor]
+    ) -> dict[str, torch.Tensor | int | float | bool]:
+        """
+        Callback function for the Adversarial Training processing logic of 1 iteration in Ignite Engine.
+        Return below items in a dictionary:
+            - IMAGE: image Tensor data for model input, already moved to device.
+            - LABEL: label Tensor data corresponding to the image, already moved to device. In case of Unsupervised
+                Learning this is equal to IMAGE.
+            - PRED: prediction result of model.
+            - LOSS: loss value computed by loss functions of the generator (reconstruction and adversarial summed up).
+            - AdversarialKeys.REALS: real images from the batch. Are the same as IMAGE.
+            - AdversarialKeys.FAKES: fake images generated by the generator. Are the same as PRED.
+            - AdversarialKeys.REAL_LOGITS: logits of the discriminator for the real images.
+            - AdversarialKeys.FAKE_LOGITS: logits of the discriminator for the fake images.
+            - AdversarialKeys.RECONSTRUCTION_LOSS: loss value computed by the reconstruction loss function.
+            - AdversarialKeys.GENERATOR_LOSS: loss value computed by the generator loss function. It is the
+                discriminator loss for the fake images. That is backpropagated through the generator only.
+            - AdversarialKeys.DISCRIMINATOR_LOSS: loss value computed by the discriminator loss function. It is the
+                discriminator loss for the real images and the fake images. That is backpropagated through the
+                discriminator only.
+
+        Args:
+            engine: `AdversarialTrainer` to execute operation for an iteration.
+            batchdata: input data for this iteration, usually can be dictionary or tuple of Tensor data.
+
+        Raises:
+            ValueError: must provide batch data for current iteration.
+
+        """
+
+        if batchdata is None:
+            raise ValueError("Must provide batch data for current iteration.")
+        batch = engine.prepare_batch(batchdata, engine.state.device, engine.non_blocking, **engine.to_kwargs)
+
+        if len(batch) == 2:
+            inputs, targets = batch
+            args: tuple = ()
+            kwargs: dict = {}
+        else:
+            inputs, targets, args, kwargs = batch
+
+        engine.state.output = {Keys.IMAGE: inputs, Keys.LABEL: targets, AdversarialKeys.REALS: inputs}
+
+        def _compute_generator_loss() -> None:
+            engine.state.output[AdversarialKeys.FAKES] = engine.g_inferer(
+                inputs, engine.state.g_network, *args, **kwargs
+            )
+            engine.state.output[Keys.PRED] = engine.state.output[AdversarialKeys.FAKES]
+            engine.fire_event(AdversarialIterationEvents.GENERATOR_FORWARD_COMPLETED)
+
+            engine.state.output[AdversarialKeys.FAKE_LOGITS] = engine.d_inferer(
+                engine.state.output[AdversarialKeys.FAKES].float().contiguous(), engine.state.d_network, *args, **kwargs
+            )
+            engine.fire_event(AdversarialIterationEvents.GENERATOR_DISCRIMINATOR_FORWARD_COMPLETED)
+
+            engine.state.output[AdversarialKeys.RECONSTRUCTION_LOSS] = engine.state.recon_loss_function(
+                engine.state.output[AdversarialKeys.FAKES], targets
+            ).mean()
+            engine.fire_event(AdversarialIterationEvents.RECONSTRUCTION_LOSS_COMPLETED)
+
+            engine.state.output[AdversarialKeys.GENERATOR_LOSS] = engine.state.g_loss_function(
+                engine.state.output[AdversarialKeys.FAKE_LOGITS]
+            ).mean()
+            engine.fire_event(AdversarialIterationEvents.GENERATOR_LOSS_COMPLETED)
+
+        # Train Generator
+        engine.state.g_network.train()
+        engine.state.g_optimizer.zero_grad(set_to_none=engine.optim_set_to_none)
+
+        if engine.amp and engine.state.g_scaler is not None:
+            with torch.cuda.amp.autocast(**engine.amp_kwargs):
+                _compute_generator_loss()
+
+            engine.state.output[Keys.LOSS] = (
+                engine.state.output[AdversarialKeys.RECONSTRUCTION_LOSS]
+                + engine.state.output[AdversarialKeys.GENERATOR_LOSS]
+            )
+            engine.state.g_scaler.scale(engine.state.output[Keys.LOSS]).backward()
+            engine.fire_event(AdversarialIterationEvents.GENERATOR_BACKWARD_COMPLETED)
+            engine.state.g_scaler.step(engine.state.g_optimizer)
+            engine.state.g_scaler.update()
+        else:
+            _compute_generator_loss()
+            (
+                engine.state.output[AdversarialKeys.RECONSTRUCTION_LOSS]
+                + engine.state.output[AdversarialKeys.GENERATOR_LOSS]
+            ).backward()
+            engine.fire_event(AdversarialIterationEvents.GENERATOR_BACKWARD_COMPLETED)
+            engine.state.g_optimizer.step()
+        engine.fire_event(AdversarialIterationEvents.GENERATOR_MODEL_COMPLETED)
+
+        def _compute_discriminator_loss() -> None:
+            engine.state.output[AdversarialKeys.REAL_LOGITS] = engine.d_inferer(
+                engine.state.output[AdversarialKeys.REALS].contiguous().detach(),
+                engine.state.d_network,
+                *args,
+                **kwargs,
+            )
+            engine.fire_event(AdversarialIterationEvents.DISCRIMINATOR_REALS_FORWARD_COMPLETED)
+
+            engine.state.output[AdversarialKeys.FAKE_LOGITS] = engine.d_inferer(
+                engine.state.output[AdversarialKeys.FAKES].contiguous().detach(),
+                engine.state.d_network,
+                *args,
+                **kwargs,
+            )
+            engine.fire_event(AdversarialIterationEvents.DISCRIMINATOR_FAKES_FORWARD_COMPLETED)
+
+            engine.state.output[AdversarialKeys.DISCRIMINATOR_LOSS] = engine.state.d_loss_function(
+                engine.state.output[AdversarialKeys.REAL_LOGITS], engine.state.output[AdversarialKeys.FAKE_LOGITS]
+            ).mean()
+            engine.fire_event(AdversarialIterationEvents.DISCRIMINATOR_LOSS_COMPLETED)
+
+        # Train Discriminator
+        engine.state.d_network.train()
+        engine.state.d_network.zero_grad(set_to_none=engine.optim_set_to_none)
+
+        if engine.amp and engine.state.d_scaler is not None:
+            with torch.cuda.amp.autocast(**engine.amp_kwargs):
+                _compute_discriminator_loss()
+
+            engine.state.d_scaler.scale(engine.state.output[AdversarialKeys.DISCRIMINATOR_LOSS]).backward()
+            engine.fire_event(AdversarialIterationEvents.DISCRIMINATOR_BACKWARD_COMPLETED)
+            engine.state.d_scaler.step(engine.state.d_optimizer)
+            engine.state.d_scaler.update()
+        else:
+            _compute_discriminator_loss()
+            engine.state.output[AdversarialKeys.DISCRIMINATOR_LOSS].backward()
+            engine.state.d_optimizer.step()
+
+        return engine.state.output
diff --git a/monai/engines/utils.py b/monai/engines/utils.py
index 02c718cd14..5339d6965a 100644
--- a/monai/engines/utils.py
+++ b/monai/engines/utils.py
@@ -13,9 +13,10 @@
 
 from abc import ABC, abstractmethod
 from collections.abc import Callable, Sequence
-from typing import TYPE_CHECKING, Any, cast
+from typing import TYPE_CHECKING, Any, Mapping, cast
 
 import torch
+import torch.nn as nn
 
 from monai.config import IgniteInfo
 from monai.transforms import apply_transform
@@ -36,6 +37,8 @@
     "PrepareBatch",
     "PrepareBatchDefault",
     "PrepareBatchExtraInput",
+    "DiffusionPrepareBatch",
+    "VPredictionPrepareBatch",
     "default_make_latent",
     "engine_apply_transform",
     "default_metric_cmp_fn",
@@ -238,6 +241,78 @@ def _get_data(key: str) -> torch.Tensor:
         return cast(torch.Tensor, image), cast(torch.Tensor, label), tuple(args_), kwargs_
 
 
+class DiffusionPrepareBatch(PrepareBatch):
+    """
+    This class is used as a callable for the `prepare_batch` parameter of engine classes for diffusion training.
+
+    Assuming a supervised training process, it will generate a noise field using `get_noise` for an input image, and
+    return the image and noise field as the image/target pair plus the noise field the kwargs under the key "noise".
+    This assumes the inferer being used in conjunction with this class expects a "noise" parameter to be provided.
+
+    If the `condition_name` is provided, this must refer to a key in the input dictionary containing the condition
+    field to be passed to the inferer. This will appear in the keyword arguments under the key "condition".
+
+    """
+
+    def __init__(self, num_train_timesteps: int, condition_name: str | None = None) -> None:
+        self.condition_name = condition_name
+        self.num_train_timesteps = num_train_timesteps
+
+    def get_noise(self, images: torch.Tensor) -> torch.Tensor:
+        """Returns the noise tensor for input tensor `images`, override this for different noise distributions."""
+        return torch.randn_like(images)
+
+    def get_timesteps(self, images: torch.Tensor) -> torch.Tensor:
+        """Get a timestep, by default this is a random integer between 0 and `self.num_train_timesteps`."""
+        return torch.randint(0, self.num_train_timesteps, (images.shape[0],), device=images.device).long()
+
+    def get_target(self, images: torch.Tensor, noise: torch.Tensor, timesteps: torch.Tensor) -> torch.Tensor:
+        """Return the target for the loss function, this is the `noise` value by default."""
+        return noise
+
+    def __call__(
+        self,
+        batchdata: dict[str, torch.Tensor],
+        device: str | torch.device | None = None,
+        non_blocking: bool = False,
+        **kwargs: Any,
+    ) -> tuple[torch.Tensor, torch.Tensor, tuple, dict]:
+        images, _ = default_prepare_batch(batchdata, device, non_blocking, **kwargs)
+        noise = self.get_noise(images).to(device, non_blocking=non_blocking, **kwargs)
+        timesteps = self.get_timesteps(images).to(device, non_blocking=non_blocking, **kwargs)
+
+        target = self.get_target(images, noise, timesteps).to(device, non_blocking=non_blocking, **kwargs)
+        infer_kwargs = {"noise": noise, "timesteps": timesteps}
+
+        if self.condition_name is not None and isinstance(batchdata, Mapping):
+            infer_kwargs["condition"] = batchdata[self.condition_name].to(device, non_blocking=non_blocking, **kwargs)
+
+        # return input, target, arguments, and keyword arguments where noise is the target and also a keyword value
+        return images, target, (), infer_kwargs
+
+
+class VPredictionPrepareBatch(DiffusionPrepareBatch):
+    """
+    This class is used as a callable for the `prepare_batch` parameter of engine classes for diffusion training.
+
+    Assuming a supervised training process, it will generate a noise field using `get_noise` for an input image, and
+    from this compute the velocity using the provided scheduler. This value is used as the target in place of the
+    noise field itself although the noise is field is in the kwargs under the key "noise". This assumes the inferer
+    being used in conjunction with this class expects a "noise" parameter to be provided.
+
+    If the `condition_name` is provided, this must refer to a key in the input dictionary containing the condition
+    field to be passed to the inferer. This will appear in the keyword arguments under the key "condition".
+
+    """
+
+    def __init__(self, scheduler: nn.Module, num_train_timesteps: int, condition_name: str | None = None) -> None:
+        super().__init__(num_train_timesteps=num_train_timesteps, condition_name=condition_name)
+        self.scheduler = scheduler
+
+    def get_target(self, images, noise, timesteps):
+        return self.scheduler.get_velocity(images, noise, timesteps)
+
+
 def default_make_latent(
     num_latents: int,
     latent_size: int,
diff --git a/monai/inferers/__init__.py b/monai/inferers/__init__.py
index 960380bfb8..fc78b9f7c4 100644
--- a/monai/inferers/__init__.py
+++ b/monai/inferers/__init__.py
@@ -12,13 +12,18 @@
 from __future__ import annotations
 
 from .inferer import (
+    ControlNetDiffusionInferer,
+    ControlNetLatentDiffusionInferer,
+    DiffusionInferer,
     Inferer,
+    LatentDiffusionInferer,
     PatchInferer,
     SaliencyInferer,
     SimpleInferer,
     SliceInferer,
     SlidingWindowInferer,
     SlidingWindowInfererAdapt,
+    VQVAETransformerInferer,
 )
 from .merger import AvgMerger, Merger, ZarrAvgMerger
 from .splitter import SlidingWindowSplitter, Splitter, WSISlidingWindowSplitter
diff --git a/monai/inferers/inferer.py b/monai/inferers/inferer.py
index 0b4199938d..769b6cc0e7 100644
--- a/monai/inferers/inferer.py
+++ b/monai/inferers/inferer.py
@@ -11,24 +11,41 @@
 
 from __future__ import annotations
 
+import math
 import warnings
 from abc import ABC, abstractmethod
 from collections.abc import Callable, Iterable, Iterator, Mapping, Sequence
+from functools import partial
 from pydoc import locate
 from typing import Any
 
 import torch
 import torch.nn as nn
+import torch.nn.functional as F
 
 from monai.apps.utils import get_logger
+from monai.data import decollate_batch
 from monai.data.meta_tensor import MetaTensor
 from monai.data.thread_buffer import ThreadBuffer
 from monai.inferers.merger import AvgMerger, Merger
 from monai.inferers.splitter import Splitter
 from monai.inferers.utils import compute_importance_map, sliding_window_inference
-from monai.utils import BlendMode, PatchKeys, PytorchPadMode, ensure_tuple, optional_import
+from monai.networks.nets import (
+    VQVAE,
+    AutoencoderKL,
+    ControlNet,
+    DecoderOnlyTransformer,
+    DiffusionModelUNet,
+    SPADEAutoencoderKL,
+    SPADEDiffusionModelUNet,
+)
+from monai.networks.schedulers import Scheduler
+from monai.transforms import CenterSpatialCrop, SpatialPad
+from monai.utils import BlendMode, Ordering, PatchKeys, PytorchPadMode, ensure_tuple, optional_import
 from monai.visualize import CAM, GradCAM, GradCAMpp
 
+tqdm, has_tqdm = optional_import("tqdm", name="tqdm")
+
 logger = get_logger(__name__)
 
 __all__ = [
@@ -752,3 +769,1264 @@ def network_wrapper(
             return out
 
         return tuple(out_i.unsqueeze(dim=self.spatial_dim + 2) for out_i in out)
+
+
+class DiffusionInferer(Inferer):
+    """
+    DiffusionInferer takes a trained diffusion model and a scheduler and can be used to perform a signal forward pass
+    for a training iteration, and sample from the model.
+
+    Args:
+        scheduler: diffusion scheduler.
+    """
+
+    def __init__(self, scheduler: Scheduler) -> None:  # type: ignore[override]
+        super().__init__()
+
+        self.scheduler = scheduler
+
+    def __call__(  # type: ignore[override]
+        self,
+        inputs: torch.Tensor,
+        diffusion_model: DiffusionModelUNet,
+        noise: torch.Tensor,
+        timesteps: torch.Tensor,
+        condition: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        """
+        Implements the forward pass for a supervised training iteration.
+
+        Args:
+            inputs: Input image to which noise is added.
+            diffusion_model: diffusion model.
+            noise: random noise, of the same shape as the input.
+            timesteps: random timesteps.
+            condition: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            seg: if model is instance of SPADEDiffusionModelUnet, segmentation must be
+            provided on the forward (for SPADE-like AE or SPADE-like DM)
+        """
+        if mode not in ["crossattn", "concat"]:
+            raise NotImplementedError(f"{mode} condition is not supported")
+
+        noisy_image: torch.Tensor = self.scheduler.add_noise(original_samples=inputs, noise=noise, timesteps=timesteps)
+        if mode == "concat":
+            if condition is None:
+                raise ValueError("Conditioning is required for concat condition")
+            else:
+                noisy_image = torch.cat([noisy_image, condition], dim=1)
+                condition = None
+        diffusion_model = (
+            partial(diffusion_model, seg=seg)
+            if isinstance(diffusion_model, SPADEDiffusionModelUNet)
+            else diffusion_model
+        )
+        prediction: torch.Tensor = diffusion_model(x=noisy_image, timesteps=timesteps, context=condition)
+
+        return prediction
+
+    @torch.no_grad()
+    def sample(
+        self,
+        input_noise: torch.Tensor,
+        diffusion_model: DiffusionModelUNet,
+        scheduler: Scheduler | None = None,
+        save_intermediates: bool | None = False,
+        intermediate_steps: int | None = 100,
+        conditioning: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        verbose: bool = True,
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor | tuple[torch.Tensor, list[torch.Tensor]]:
+        """
+        Args:
+            input_noise: random noise, of the same shape as the desired sample.
+            diffusion_model: model to sample from.
+            scheduler: diffusion scheduler. If none provided will use the class attribute scheduler
+            save_intermediates: whether to return intermediates along the sampling change
+            intermediate_steps: if save_intermediates is True, saves every n steps
+            conditioning: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            verbose: if true, prints the progression bar of the sampling process.
+            seg: if diffusion model is instance of SPADEDiffusionModel, segmentation must be provided.
+        """
+        if mode not in ["crossattn", "concat"]:
+            raise NotImplementedError(f"{mode} condition is not supported")
+        if mode == "concat" and conditioning is None:
+            raise ValueError("Conditioning must be supplied for if condition mode is concat.")
+        if not scheduler:
+            scheduler = self.scheduler
+        image = input_noise
+        if verbose and has_tqdm:
+            progress_bar = tqdm(scheduler.timesteps)
+        else:
+            progress_bar = iter(scheduler.timesteps)
+        intermediates = []
+        for t in progress_bar:
+            # 1. predict noise model_output
+            diffusion_model = (
+                partial(diffusion_model, seg=seg)
+                if isinstance(diffusion_model, SPADEDiffusionModelUNet)
+                else diffusion_model
+            )
+            if mode == "concat" and conditioning is not None:
+                model_input = torch.cat([image, conditioning], dim=1)
+                model_output = diffusion_model(
+                    model_input, timesteps=torch.Tensor((t,)).to(input_noise.device), context=None
+                )
+            else:
+                model_output = diffusion_model(
+                    image, timesteps=torch.Tensor((t,)).to(input_noise.device), context=conditioning
+                )
+
+            # 2. compute previous image: x_t -> x_t-1
+            image, _ = scheduler.step(model_output, t, image)
+            if save_intermediates and t % intermediate_steps == 0:
+                intermediates.append(image)
+        if save_intermediates:
+            return image, intermediates
+        else:
+            return image
+
+    @torch.no_grad()
+    def get_likelihood(
+        self,
+        inputs: torch.Tensor,
+        diffusion_model: DiffusionModelUNet,
+        scheduler: Scheduler | None = None,
+        save_intermediates: bool | None = False,
+        conditioning: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        original_input_range: tuple = (0, 255),
+        scaled_input_range: tuple = (0, 1),
+        verbose: bool = True,
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor | tuple[torch.Tensor, list[torch.Tensor]]:
+        """
+        Computes the log-likelihoods for an input.
+
+        Args:
+            inputs: input images, NxCxHxW[xD]
+            diffusion_model: model to compute likelihood from
+            scheduler: diffusion scheduler. If none provided will use the class attribute scheduler.
+            save_intermediates: save the intermediate spatial KL maps
+            conditioning: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            original_input_range: the [min,max] intensity range of the input data before any scaling was applied.
+            scaled_input_range: the [min,max] intensity range of the input data after scaling.
+            verbose: if true, prints the progression bar of the sampling process.
+            seg: if diffusion model is instance of SPADEDiffusionModel, segmentation must be provided.
+        """
+
+        if not scheduler:
+            scheduler = self.scheduler
+        if scheduler._get_name() != "DDPMScheduler":
+            raise NotImplementedError(
+                f"Likelihood computation is only compatible with DDPMScheduler,"
+                f" you are using {scheduler._get_name()}"
+            )
+        if mode not in ["crossattn", "concat"]:
+            raise NotImplementedError(f"{mode} condition is not supported")
+        if mode == "concat" and conditioning is None:
+            raise ValueError("Conditioning must be supplied for if condition mode is concat.")
+        if verbose and has_tqdm:
+            progress_bar = tqdm(scheduler.timesteps)
+        else:
+            progress_bar = iter(scheduler.timesteps)
+        intermediates = []
+        noise = torch.randn_like(inputs).to(inputs.device)
+        total_kl = torch.zeros(inputs.shape[0]).to(inputs.device)
+        for t in progress_bar:
+            timesteps = torch.full(inputs.shape[:1], t, device=inputs.device).long()
+            noisy_image = self.scheduler.add_noise(original_samples=inputs, noise=noise, timesteps=timesteps)
+            diffusion_model = (
+                partial(diffusion_model, seg=seg)
+                if isinstance(diffusion_model, SPADEDiffusionModelUNet)
+                else diffusion_model
+            )
+            if mode == "concat" and conditioning is not None:
+                noisy_image = torch.cat([noisy_image, conditioning], dim=1)
+                model_output = diffusion_model(noisy_image, timesteps=timesteps, context=None)
+            else:
+                model_output = diffusion_model(x=noisy_image, timesteps=timesteps, context=conditioning)
+
+            # get the model's predicted mean,  and variance if it is predicted
+            if model_output.shape[1] == inputs.shape[1] * 2 and scheduler.variance_type in ["learned", "learned_range"]:
+                model_output, predicted_variance = torch.split(model_output, inputs.shape[1], dim=1)
+            else:
+                predicted_variance = None
+
+            # 1. compute alphas, betas
+            alpha_prod_t = scheduler.alphas_cumprod[t]
+            alpha_prod_t_prev = scheduler.alphas_cumprod[t - 1] if t > 0 else scheduler.one
+            beta_prod_t = 1 - alpha_prod_t
+            beta_prod_t_prev = 1 - alpha_prod_t_prev
+
+            # 2. compute predicted original sample from predicted noise also called
+            # "predicted x_0" of formula (15) from https://arxiv.org/pdf/2006.11239.pdf
+            if scheduler.prediction_type == "epsilon":
+                pred_original_sample = (noisy_image - beta_prod_t ** (0.5) * model_output) / alpha_prod_t ** (0.5)
+            elif scheduler.prediction_type == "sample":
+                pred_original_sample = model_output
+            elif scheduler.prediction_type == "v_prediction":
+                pred_original_sample = (alpha_prod_t**0.5) * noisy_image - (beta_prod_t**0.5) * model_output
+            # 3. Clip "predicted x_0"
+            if scheduler.clip_sample:
+                pred_original_sample = torch.clamp(pred_original_sample, -1, 1)
+
+            # 4. Compute coefficients for pred_original_sample x_0 and current sample x_t
+            # See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
+            pred_original_sample_coeff = (alpha_prod_t_prev ** (0.5) * scheduler.betas[t]) / beta_prod_t
+            current_sample_coeff = scheduler.alphas[t] ** (0.5) * beta_prod_t_prev / beta_prod_t
+
+            # 5. Compute predicted previous sample µ_t
+            # See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
+            predicted_mean = pred_original_sample_coeff * pred_original_sample + current_sample_coeff * noisy_image
+
+            # get the posterior mean and variance
+            posterior_mean = scheduler._get_mean(timestep=t, x_0=inputs, x_t=noisy_image)
+            posterior_variance = scheduler._get_variance(timestep=t, predicted_variance=predicted_variance)
+
+            log_posterior_variance = torch.log(posterior_variance)
+            log_predicted_variance = torch.log(predicted_variance) if predicted_variance else log_posterior_variance
+
+            if t == 0:
+                # compute -log p(x_0|x_1)
+                kl = -self._get_decoder_log_likelihood(
+                    inputs=inputs,
+                    means=predicted_mean,
+                    log_scales=0.5 * log_predicted_variance,
+                    original_input_range=original_input_range,
+                    scaled_input_range=scaled_input_range,
+                )
+            else:
+                # compute kl between two normals
+                kl = 0.5 * (
+                    -1.0
+                    + log_predicted_variance
+                    - log_posterior_variance
+                    + torch.exp(log_posterior_variance - log_predicted_variance)
+                    + ((posterior_mean - predicted_mean) ** 2) * torch.exp(-log_predicted_variance)
+                )
+            total_kl += kl.view(kl.shape[0], -1).mean(dim=1)
+            if save_intermediates:
+                intermediates.append(kl.cpu())
+
+        if save_intermediates:
+            return total_kl, intermediates
+        else:
+            return total_kl
+
+    def _approx_standard_normal_cdf(self, x):
+        """
+        A fast approximation of the cumulative distribution function of the
+        standard normal. Code adapted from https://github.com/openai/improved-diffusion.
+        """
+
+        return 0.5 * (
+            1.0 + torch.tanh(torch.sqrt(torch.Tensor([2.0 / math.pi]).to(x.device)) * (x + 0.044715 * torch.pow(x, 3)))
+        )
+
+    def _get_decoder_log_likelihood(
+        self,
+        inputs: torch.Tensor,
+        means: torch.Tensor,
+        log_scales: torch.Tensor,
+        original_input_range: tuple = (0, 255),
+        scaled_input_range: tuple = (0, 1),
+    ) -> torch.Tensor:
+        """
+        Compute the log-likelihood of a Gaussian distribution discretizing to a
+        given image. Code adapted from https://github.com/openai/improved-diffusion.
+
+        Args:
+            input: the target images. It is assumed that this was uint8 values,
+                      rescaled to the range [-1, 1].
+            means: the Gaussian mean Tensor.
+            log_scales: the Gaussian log stddev Tensor.
+            original_input_range: the [min,max] intensity range of the input data before any scaling was applied.
+            scaled_input_range: the [min,max] intensity range of the input data after scaling.
+        """
+        if inputs.shape != means.shape:
+            raise ValueError(f"Inputs and means must have the same shape, got {inputs.shape} and {means.shape}")
+        bin_width = (scaled_input_range[1] - scaled_input_range[0]) / (
+            original_input_range[1] - original_input_range[0]
+        )
+        centered_x = inputs - means
+        inv_stdv = torch.exp(-log_scales)
+        plus_in = inv_stdv * (centered_x + bin_width / 2)
+        cdf_plus = self._approx_standard_normal_cdf(plus_in)
+        min_in = inv_stdv * (centered_x - bin_width / 2)
+        cdf_min = self._approx_standard_normal_cdf(min_in)
+        log_cdf_plus = torch.log(cdf_plus.clamp(min=1e-12))
+        log_one_minus_cdf_min = torch.log((1.0 - cdf_min).clamp(min=1e-12))
+        cdf_delta = cdf_plus - cdf_min
+        log_probs = torch.where(
+            inputs < -0.999,
+            log_cdf_plus,
+            torch.where(inputs > 0.999, log_one_minus_cdf_min, torch.log(cdf_delta.clamp(min=1e-12))),
+        )
+        return log_probs
+
+
+class LatentDiffusionInferer(DiffusionInferer):
+    """
+    LatentDiffusionInferer takes a stage 1 model (VQVAE or AutoencoderKL), diffusion model, and a scheduler, and can
+    be used to perform a signal forward pass for a training iteration, and sample from the model.
+
+    Args:
+        scheduler: a scheduler to be used in combination with `unet` to denoise the encoded image latents.
+        scale_factor: scale factor to multiply the values of the latent representation before processing it by the
+            second stage.
+        ldm_latent_shape: desired spatial latent space shape. Used if there is a difference in the autoencoder model's latent shape.
+        autoencoder_latent_shape:  autoencoder_latent_shape: autoencoder spatial latent space shape. Used if there is a
+             difference between the autoencoder's latent shape and the DM shape.
+    """
+
+    def __init__(
+        self,
+        scheduler: Scheduler,
+        scale_factor: float = 1.0,
+        ldm_latent_shape: list | None = None,
+        autoencoder_latent_shape: list | None = None,
+    ) -> None:
+        super().__init__(scheduler=scheduler)
+        self.scale_factor = scale_factor
+        if (ldm_latent_shape is None) ^ (autoencoder_latent_shape is None):
+            raise ValueError("If ldm_latent_shape is None, autoencoder_latent_shape must be None, and vice versa.")
+        self.ldm_latent_shape = ldm_latent_shape
+        self.autoencoder_latent_shape = autoencoder_latent_shape
+        if self.ldm_latent_shape is not None and self.autoencoder_latent_shape is not None:
+            self.ldm_resizer = SpatialPad(spatial_size=self.ldm_latent_shape)
+            self.autoencoder_resizer = CenterSpatialCrop(roi_size=self.autoencoder_latent_shape)
+
+    def __call__(  # type: ignore[override]
+        self,
+        inputs: torch.Tensor,
+        autoencoder_model: AutoencoderKL | VQVAE,
+        diffusion_model: DiffusionModelUNet,
+        noise: torch.Tensor,
+        timesteps: torch.Tensor,
+        condition: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        """
+        Implements the forward pass for a supervised training iteration.
+
+        Args:
+            inputs: input image to which the latent representation will be extracted and noise is added.
+            autoencoder_model: first stage model.
+            diffusion_model: diffusion model.
+            noise: random noise, of the same shape as the latent representation.
+            timesteps: random timesteps.
+            condition: conditioning for network input.
+            mode: Conditioning mode for the network.
+            seg: if diffusion model is instance of SPADEDiffusionModel, segmentation must be provided.
+        """
+        with torch.no_grad():
+            latent = autoencoder_model.encode_stage_2_inputs(inputs) * self.scale_factor
+
+        if self.ldm_latent_shape is not None:
+            latent = torch.stack([self.ldm_resizer(i) for i in decollate_batch(latent)], 0)
+
+        prediction: torch.Tensor = super().__call__(
+            inputs=latent,
+            diffusion_model=diffusion_model,
+            noise=noise,
+            timesteps=timesteps,
+            condition=condition,
+            mode=mode,
+            seg=seg,
+        )
+        return prediction
+
+    @torch.no_grad()
+    def sample(  # type: ignore[override]
+        self,
+        input_noise: torch.Tensor,
+        autoencoder_model: AutoencoderKL | VQVAE,
+        diffusion_model: DiffusionModelUNet,
+        scheduler: Scheduler | None = None,
+        save_intermediates: bool | None = False,
+        intermediate_steps: int | None = 100,
+        conditioning: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        verbose: bool = True,
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor | tuple[torch.Tensor, list[torch.Tensor]]:
+        """
+        Args:
+            input_noise: random noise, of the same shape as the desired latent representation.
+            autoencoder_model: first stage model.
+            diffusion_model: model to sample from.
+            scheduler: diffusion scheduler. If none provided will use the class attribute scheduler.
+            save_intermediates: whether to return intermediates along the sampling change
+            intermediate_steps: if save_intermediates is True, saves every n steps
+            conditioning: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            verbose: if true, prints the progression bar of the sampling process.
+            seg: if diffusion model is instance of SPADEDiffusionModel, or autoencoder_model
+             is instance of SPADEAutoencoderKL, segmentation must be provided.
+        """
+
+        if (
+            isinstance(autoencoder_model, SPADEAutoencoderKL)
+            and isinstance(diffusion_model, SPADEDiffusionModelUNet)
+            and autoencoder_model.decoder.label_nc != diffusion_model.label_nc
+        ):
+            raise ValueError(
+                f"If both autoencoder_model and diffusion_model implement SPADE, the number of semantic"
+                f"labels for each must be compatible, but got {autoencoder_model.decoder.label_nc} and"
+                f"{diffusion_model.label_nc}"
+            )
+
+        outputs = super().sample(
+            input_noise=input_noise,
+            diffusion_model=diffusion_model,
+            scheduler=scheduler,
+            save_intermediates=save_intermediates,
+            intermediate_steps=intermediate_steps,
+            conditioning=conditioning,
+            mode=mode,
+            verbose=verbose,
+            seg=seg,
+        )
+
+        if save_intermediates:
+            latent, latent_intermediates = outputs
+        else:
+            latent = outputs
+
+        if self.autoencoder_latent_shape is not None:
+            latent = torch.stack([self.autoencoder_resizer(i) for i in decollate_batch(latent)], 0)
+            latent_intermediates = [
+                torch.stack([self.autoencoder_resizer(i) for i in decollate_batch(l)], 0) for l in latent_intermediates
+            ]
+
+        decode = autoencoder_model.decode_stage_2_outputs
+        if isinstance(autoencoder_model, SPADEAutoencoderKL):
+            decode = partial(autoencoder_model.decode_stage_2_outputs, seg=seg)
+        image = decode(latent / self.scale_factor)
+
+        if save_intermediates:
+            intermediates = []
+            for latent_intermediate in latent_intermediates:
+                decode = autoencoder_model.decode_stage_2_outputs
+                if isinstance(autoencoder_model, SPADEAutoencoderKL):
+                    decode = partial(autoencoder_model.decode_stage_2_outputs, seg=seg)
+                intermediates.append(decode(latent_intermediate / self.scale_factor))
+            return image, intermediates
+
+        else:
+            return image
+
+    @torch.no_grad()
+    def get_likelihood(  # type: ignore[override]
+        self,
+        inputs: torch.Tensor,
+        autoencoder_model: AutoencoderKL | VQVAE,
+        diffusion_model: DiffusionModelUNet,
+        scheduler: Scheduler | None = None,
+        save_intermediates: bool | None = False,
+        conditioning: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        original_input_range: tuple | None = (0, 255),
+        scaled_input_range: tuple | None = (0, 1),
+        verbose: bool = True,
+        resample_latent_likelihoods: bool = False,
+        resample_interpolation_mode: str = "nearest",
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor | tuple[torch.Tensor, list[torch.Tensor]]:
+        """
+        Computes the log-likelihoods of the latent representations of the input.
+
+        Args:
+            inputs: input images, NxCxHxW[xD]
+            autoencoder_model: first stage model.
+            diffusion_model: model to compute likelihood from
+            scheduler: diffusion scheduler. If none provided will use the class attribute scheduler
+            save_intermediates: save the intermediate spatial KL maps
+            conditioning: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            original_input_range: the [min,max] intensity range of the input data before any scaling was applied.
+            scaled_input_range: the [min,max] intensity range of the input data after scaling.
+            verbose: if true, prints the progression bar of the sampling process.
+            resample_latent_likelihoods: if true, resamples the intermediate likelihood maps to have the same spatial
+                dimension as the input images.
+            resample_interpolation_mode: if use resample_latent_likelihoods, select interpolation 'nearest', 'bilinear',
+                or 'trilinear;
+            seg: if diffusion model is instance of SPADEDiffusionModel, or autoencoder_model
+             is instance of SPADEAutoencoderKL, segmentation must be provided.
+        """
+        if resample_latent_likelihoods and resample_interpolation_mode not in ("nearest", "bilinear", "trilinear"):
+            raise ValueError(
+                f"resample_interpolation mode should be either nearest, bilinear, or trilinear, got {resample_interpolation_mode}"
+            )
+        latents = autoencoder_model.encode_stage_2_inputs(inputs) * self.scale_factor
+
+        if self.ldm_latent_shape is not None:
+            latents = torch.stack([self.ldm_resizer(i) for i in decollate_batch(latents)], 0)
+
+        outputs = super().get_likelihood(
+            inputs=latents,
+            diffusion_model=diffusion_model,
+            scheduler=scheduler,
+            save_intermediates=save_intermediates,
+            conditioning=conditioning,
+            mode=mode,
+            verbose=verbose,
+            seg=seg,
+        )
+
+        if save_intermediates and resample_latent_likelihoods:
+            intermediates = outputs[1]
+            resizer = nn.Upsample(size=inputs.shape[2:], mode=resample_interpolation_mode)
+            intermediates = [resizer(x) for x in intermediates]
+            outputs = (outputs[0], intermediates)
+        return outputs
+
+
+class ControlNetDiffusionInferer(DiffusionInferer):
+    """
+    ControlNetDiffusionInferer takes a trained diffusion model and a scheduler and can be used to perform a signal
+    forward pass for a training iteration, and sample from the model, supporting ControlNet-based conditioning.
+
+    Args:
+        scheduler: diffusion scheduler.
+    """
+
+    def __init__(self, scheduler: Scheduler) -> None:
+        Inferer.__init__(self)
+        self.scheduler = scheduler
+
+    def __call__(  # type: ignore[override]
+        self,
+        inputs: torch.Tensor,
+        diffusion_model: DiffusionModelUNet,
+        controlnet: ControlNet,
+        noise: torch.Tensor,
+        timesteps: torch.Tensor,
+        cn_cond: torch.Tensor,
+        condition: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        """
+        Implements the forward pass for a supervised training iteration.
+
+        Args:
+            inputs: Input image to which noise is added.
+            diffusion_model: diffusion model.
+            controlnet: controlnet sub-network.
+            noise: random noise, of the same shape as the input.
+            timesteps: random timesteps.
+            cn_cond: conditioning image for the ControlNet.
+            condition: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            seg: if model is instance of SPADEDiffusionModelUnet, segmentation must be
+            provided on the forward (for SPADE-like AE or SPADE-like DM)
+        """
+        if mode not in ["crossattn", "concat"]:
+            raise NotImplementedError(f"{mode} condition is not supported")
+
+        noisy_image = self.scheduler.add_noise(original_samples=inputs, noise=noise, timesteps=timesteps)
+        down_block_res_samples, mid_block_res_sample = controlnet(
+            x=noisy_image, timesteps=timesteps, controlnet_cond=cn_cond
+        )
+        if mode == "concat" and condition is not None:
+            noisy_image = torch.cat([noisy_image, condition], dim=1)
+            condition = None
+
+        diffuse = diffusion_model
+        if isinstance(diffusion_model, SPADEDiffusionModelUNet):
+            diffuse = partial(diffusion_model, seg=seg)
+
+        prediction: torch.Tensor = diffuse(
+            x=noisy_image,
+            timesteps=timesteps,
+            context=condition,
+            down_block_additional_residuals=down_block_res_samples,
+            mid_block_additional_residual=mid_block_res_sample,
+        )
+
+        return prediction
+
+    @torch.no_grad()
+    def sample(  # type: ignore[override]
+        self,
+        input_noise: torch.Tensor,
+        diffusion_model: DiffusionModelUNet,
+        controlnet: ControlNet,
+        cn_cond: torch.Tensor,
+        scheduler: Scheduler | None = None,
+        save_intermediates: bool | None = False,
+        intermediate_steps: int | None = 100,
+        conditioning: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        verbose: bool = True,
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor | tuple[torch.Tensor, list[torch.Tensor]]:
+        """
+        Args:
+            input_noise: random noise, of the same shape as the desired sample.
+            diffusion_model: model to sample from.
+            controlnet: controlnet sub-network.
+            cn_cond: conditioning image for the ControlNet.
+            scheduler: diffusion scheduler. If none provided will use the class attribute scheduler
+            save_intermediates: whether to return intermediates along the sampling change
+            intermediate_steps: if save_intermediates is True, saves every n steps
+            conditioning: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            verbose: if true, prints the progression bar of the sampling process.
+            seg: if diffusion model is instance of SPADEDiffusionModel, segmentation must be provided.
+        """
+        if mode not in ["crossattn", "concat"]:
+            raise NotImplementedError(f"{mode} condition is not supported")
+
+        if not scheduler:
+            scheduler = self.scheduler
+        image = input_noise
+        if verbose and has_tqdm:
+            progress_bar = tqdm(scheduler.timesteps)
+        else:
+            progress_bar = iter(scheduler.timesteps)
+        intermediates = []
+        for t in progress_bar:
+            # 1. ControlNet forward
+            down_block_res_samples, mid_block_res_sample = controlnet(
+                x=image, timesteps=torch.Tensor((t,)).to(input_noise.device), controlnet_cond=cn_cond
+            )
+            # 2. predict noise model_output
+            diffuse = diffusion_model
+            if isinstance(diffusion_model, SPADEDiffusionModelUNet):
+                diffuse = partial(diffusion_model, seg=seg)
+
+            if mode == "concat" and conditioning is not None:
+                model_input = torch.cat([image, conditioning], dim=1)
+                model_output = diffuse(
+                    model_input,
+                    timesteps=torch.Tensor((t,)).to(input_noise.device),
+                    context=None,
+                    down_block_additional_residuals=down_block_res_samples,
+                    mid_block_additional_residual=mid_block_res_sample,
+                )
+            else:
+                model_output = diffuse(
+                    image,
+                    timesteps=torch.Tensor((t,)).to(input_noise.device),
+                    context=conditioning,
+                    down_block_additional_residuals=down_block_res_samples,
+                    mid_block_additional_residual=mid_block_res_sample,
+                )
+
+            # 3. compute previous image: x_t -> x_t-1
+            image, _ = scheduler.step(model_output, t, image)
+            if save_intermediates and t % intermediate_steps == 0:
+                intermediates.append(image)
+        if save_intermediates:
+            return image, intermediates
+        else:
+            return image
+
+    @torch.no_grad()
+    def get_likelihood(  # type: ignore[override]
+        self,
+        inputs: torch.Tensor,
+        diffusion_model: DiffusionModelUNet,
+        controlnet: ControlNet,
+        cn_cond: torch.Tensor,
+        scheduler: Scheduler | None = None,
+        save_intermediates: bool | None = False,
+        conditioning: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        original_input_range: tuple = (0, 255),
+        scaled_input_range: tuple = (0, 1),
+        verbose: bool = True,
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor | tuple[torch.Tensor, list[torch.Tensor]]:
+        """
+        Computes the log-likelihoods for an input.
+
+        Args:
+            inputs: input images, NxCxHxW[xD]
+            diffusion_model: model to compute likelihood from
+            controlnet: controlnet sub-network.
+            cn_cond: conditioning image for the ControlNet.
+            scheduler: diffusion scheduler. If none provided will use the class attribute scheduler.
+            save_intermediates: save the intermediate spatial KL maps
+            conditioning: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            original_input_range: the [min,max] intensity range of the input data before any scaling was applied.
+            scaled_input_range: the [min,max] intensity range of the input data after scaling.
+            verbose: if true, prints the progression bar of the sampling process.
+            seg: if diffusion model is instance of SPADEDiffusionModel, segmentation must be provided.
+        """
+
+        if not scheduler:
+            scheduler = self.scheduler
+        if scheduler._get_name() != "DDPMScheduler":
+            raise NotImplementedError(
+                f"Likelihood computation is only compatible with DDPMScheduler,"
+                f" you are using {scheduler._get_name()}"
+            )
+        if mode not in ["crossattn", "concat"]:
+            raise NotImplementedError(f"{mode} condition is not supported")
+        if verbose and has_tqdm:
+            progress_bar = tqdm(scheduler.timesteps)
+        else:
+            progress_bar = iter(scheduler.timesteps)
+        intermediates = []
+        noise = torch.randn_like(inputs).to(inputs.device)
+        total_kl = torch.zeros(inputs.shape[0]).to(inputs.device)
+        for t in progress_bar:
+            timesteps = torch.full(inputs.shape[:1], t, device=inputs.device).long()
+            noisy_image = self.scheduler.add_noise(original_samples=inputs, noise=noise, timesteps=timesteps)
+            down_block_res_samples, mid_block_res_sample = controlnet(
+                x=noisy_image, timesteps=torch.Tensor((t,)).to(inputs.device), controlnet_cond=cn_cond
+            )
+
+            diffuse = diffusion_model
+            if isinstance(diffusion_model, SPADEDiffusionModelUNet):
+                diffuse = partial(diffusion_model, seg=seg)
+
+            if mode == "concat" and conditioning is not None:
+                noisy_image = torch.cat([noisy_image, conditioning], dim=1)
+                model_output = diffuse(
+                    noisy_image,
+                    timesteps=timesteps,
+                    context=None,
+                    down_block_additional_residuals=down_block_res_samples,
+                    mid_block_additional_residual=mid_block_res_sample,
+                )
+            else:
+                model_output = diffuse(
+                    x=noisy_image,
+                    timesteps=timesteps,
+                    context=conditioning,
+                    down_block_additional_residuals=down_block_res_samples,
+                    mid_block_additional_residual=mid_block_res_sample,
+                )
+            # get the model's predicted mean,  and variance if it is predicted
+            if model_output.shape[1] == inputs.shape[1] * 2 and scheduler.variance_type in ["learned", "learned_range"]:
+                model_output, predicted_variance = torch.split(model_output, inputs.shape[1], dim=1)
+            else:
+                predicted_variance = None
+
+            # 1. compute alphas, betas
+            alpha_prod_t = scheduler.alphas_cumprod[t]
+            alpha_prod_t_prev = scheduler.alphas_cumprod[t - 1] if t > 0 else scheduler.one
+            beta_prod_t = 1 - alpha_prod_t
+            beta_prod_t_prev = 1 - alpha_prod_t_prev
+
+            # 2. compute predicted original sample from predicted noise also called
+            # "predicted x_0" of formula (15) from https://arxiv.org/pdf/2006.11239.pdf
+            if scheduler.prediction_type == "epsilon":
+                pred_original_sample = (noisy_image - beta_prod_t ** (0.5) * model_output) / alpha_prod_t ** (0.5)
+            elif scheduler.prediction_type == "sample":
+                pred_original_sample = model_output
+            elif scheduler.prediction_type == "v_prediction":
+                pred_original_sample = (alpha_prod_t**0.5) * noisy_image - (beta_prod_t**0.5) * model_output
+            # 3. Clip "predicted x_0"
+            if scheduler.clip_sample:
+                pred_original_sample = torch.clamp(pred_original_sample, -1, 1)
+
+            # 4. Compute coefficients for pred_original_sample x_0 and current sample x_t
+            # See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
+            pred_original_sample_coeff = (alpha_prod_t_prev ** (0.5) * scheduler.betas[t]) / beta_prod_t
+            current_sample_coeff = scheduler.alphas[t] ** (0.5) * beta_prod_t_prev / beta_prod_t
+
+            # 5. Compute predicted previous sample µ_t
+            # See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
+            predicted_mean = pred_original_sample_coeff * pred_original_sample + current_sample_coeff * noisy_image
+
+            # get the posterior mean and variance
+            posterior_mean = scheduler._get_mean(timestep=t, x_0=inputs, x_t=noisy_image)
+            posterior_variance = scheduler._get_variance(timestep=t, predicted_variance=predicted_variance)
+
+            log_posterior_variance = torch.log(posterior_variance)
+            log_predicted_variance = torch.log(predicted_variance) if predicted_variance else log_posterior_variance
+
+            if t == 0:
+                # compute -log p(x_0|x_1)
+                kl = -super()._get_decoder_log_likelihood(
+                    inputs=inputs,
+                    means=predicted_mean,
+                    log_scales=0.5 * log_predicted_variance,
+                    original_input_range=original_input_range,
+                    scaled_input_range=scaled_input_range,
+                )
+            else:
+                # compute kl between two normals
+                kl = 0.5 * (
+                    -1.0
+                    + log_predicted_variance
+                    - log_posterior_variance
+                    + torch.exp(log_posterior_variance - log_predicted_variance)
+                    + ((posterior_mean - predicted_mean) ** 2) * torch.exp(-log_predicted_variance)
+                )
+            total_kl += kl.view(kl.shape[0], -1).mean(dim=1)
+            if save_intermediates:
+                intermediates.append(kl.cpu())
+
+        if save_intermediates:
+            return total_kl, intermediates
+        else:
+            return total_kl
+
+
+class ControlNetLatentDiffusionInferer(ControlNetDiffusionInferer):
+    """
+    ControlNetLatentDiffusionInferer takes a stage 1 model (VQVAE or AutoencoderKL), diffusion model, controlnet,
+    and a scheduler, and can be used to perform a signal forward pass for a training iteration, and sample from
+    the model.
+
+    Args:
+        scheduler: a scheduler to be used in combination with `unet` to denoise the encoded image latents.
+        scale_factor: scale factor to multiply the values of the latent representation before processing it by the
+            second stage.
+        ldm_latent_shape: desired spatial latent space shape. Used if there is a difference in the autoencoder model's latent shape.
+        autoencoder_latent_shape:  autoencoder_latent_shape: autoencoder spatial latent space shape. Used if there is a
+             difference between the autoencoder's latent shape and the DM shape.
+    """
+
+    def __init__(
+        self,
+        scheduler: Scheduler,
+        scale_factor: float = 1.0,
+        ldm_latent_shape: list | None = None,
+        autoencoder_latent_shape: list | None = None,
+    ) -> None:
+        super().__init__(scheduler=scheduler)
+        self.scale_factor = scale_factor
+        if (ldm_latent_shape is None) ^ (autoencoder_latent_shape is None):
+            raise ValueError("If ldm_latent_shape is None, autoencoder_latent_shape must be None" "and vice versa.")
+        self.ldm_latent_shape = ldm_latent_shape
+        self.autoencoder_latent_shape = autoencoder_latent_shape
+        if self.ldm_latent_shape is not None and self.autoencoder_latent_shape is not None:
+            self.ldm_resizer = SpatialPad(spatial_size=self.ldm_latent_shape)
+            self.autoencoder_resizer = CenterSpatialCrop(roi_size=self.autoencoder_latent_shape)
+
+    def __call__(  # type: ignore[override]
+        self,
+        inputs: torch.Tensor,
+        autoencoder_model: AutoencoderKL | VQVAE,
+        diffusion_model: DiffusionModelUNet,
+        controlnet: ControlNet,
+        noise: torch.Tensor,
+        timesteps: torch.Tensor,
+        cn_cond: torch.Tensor,
+        condition: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        """
+        Implements the forward pass for a supervised training iteration.
+
+        Args:
+            inputs: input image to which the latent representation will be extracted and noise is added.
+            autoencoder_model: first stage model.
+            diffusion_model: diffusion model.
+            controlnet: instance of ControlNet model
+            noise: random noise, of the same shape as the latent representation.
+            timesteps: random timesteps.
+            cn_cond: conditioning tensor for the ControlNet network
+            condition: conditioning for network input.
+            mode: Conditioning mode for the network.
+            seg: if diffusion model is instance of SPADEDiffusionModel, segmentation must be provided.
+        """
+        with torch.no_grad():
+            latent = autoencoder_model.encode_stage_2_inputs(inputs) * self.scale_factor
+
+        if self.ldm_latent_shape is not None:
+            latent = torch.stack([self.ldm_resizer(i) for i in decollate_batch(latent)], 0)
+
+        if cn_cond.shape[2:] != latent.shape[2:]:
+            cn_cond = F.interpolate(cn_cond, latent.shape[2:])
+
+        prediction = super().__call__(
+            inputs=latent,
+            diffusion_model=diffusion_model,
+            controlnet=controlnet,
+            noise=noise,
+            timesteps=timesteps,
+            cn_cond=cn_cond,
+            condition=condition,
+            mode=mode,
+            seg=seg,
+        )
+
+        return prediction
+
+    @torch.no_grad()
+    def sample(  # type: ignore[override]
+        self,
+        input_noise: torch.Tensor,
+        autoencoder_model: AutoencoderKL | VQVAE,
+        diffusion_model: DiffusionModelUNet,
+        controlnet: ControlNet,
+        cn_cond: torch.Tensor,
+        scheduler: Scheduler | None = None,
+        save_intermediates: bool | None = False,
+        intermediate_steps: int | None = 100,
+        conditioning: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        verbose: bool = True,
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor | tuple[torch.Tensor, list[torch.Tensor]]:
+        """
+        Args:
+            input_noise: random noise, of the same shape as the desired latent representation.
+            autoencoder_model: first stage model.
+            diffusion_model: model to sample from.
+            controlnet: instance of ControlNet model.
+            cn_cond: conditioning tensor for the ControlNet network.
+            scheduler: diffusion scheduler. If none provided will use the class attribute scheduler.
+            save_intermediates: whether to return intermediates along the sampling change
+            intermediate_steps: if save_intermediates is True, saves every n steps
+            conditioning: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            verbose: if true, prints the progression bar of the sampling process.
+            seg: if diffusion model is instance of SPADEDiffusionModel, or autoencoder_model
+             is instance of SPADEAutoencoderKL, segmentation must be provided.
+        """
+
+        if (
+            isinstance(autoencoder_model, SPADEAutoencoderKL)
+            and isinstance(diffusion_model, SPADEDiffusionModelUNet)
+            and autoencoder_model.decoder.label_nc != diffusion_model.label_nc
+        ):
+            raise ValueError(
+                "If both autoencoder_model and diffusion_model implement SPADE, the number of semantic"
+                "labels for each must be compatible. Got {autoencoder_model.decoder.label_nc} and {diffusion_model.label_nc}"
+            )
+
+        if cn_cond.shape[2:] != input_noise.shape[2:]:
+            cn_cond = F.interpolate(cn_cond, input_noise.shape[2:])
+
+        outputs = super().sample(
+            input_noise=input_noise,
+            diffusion_model=diffusion_model,
+            controlnet=controlnet,
+            cn_cond=cn_cond,
+            scheduler=scheduler,
+            save_intermediates=save_intermediates,
+            intermediate_steps=intermediate_steps,
+            conditioning=conditioning,
+            mode=mode,
+            verbose=verbose,
+            seg=seg,
+        )
+
+        if save_intermediates:
+            latent, latent_intermediates = outputs
+        else:
+            latent = outputs
+
+        if self.autoencoder_latent_shape is not None:
+            latent = torch.stack([self.autoencoder_resizer(i) for i in decollate_batch(latent)], 0)
+            latent_intermediates = [
+                torch.stack([self.autoencoder_resizer(i) for i in decollate_batch(l)], 0) for l in latent_intermediates
+            ]
+
+        decode = autoencoder_model.decode_stage_2_outputs
+        if isinstance(autoencoder_model, SPADEAutoencoderKL):
+            decode = partial(autoencoder_model.decode_stage_2_outputs, seg=seg)
+
+        image = decode(latent / self.scale_factor)
+
+        if save_intermediates:
+            intermediates = []
+            for latent_intermediate in latent_intermediates:
+                decode = autoencoder_model.decode_stage_2_outputs
+                if isinstance(autoencoder_model, SPADEAutoencoderKL):
+                    decode = partial(autoencoder_model.decode_stage_2_outputs, seg=seg)
+                intermediates.append(decode(latent_intermediate / self.scale_factor))
+            return image, intermediates
+
+        else:
+            return image
+
+    @torch.no_grad()
+    def get_likelihood(  # type: ignore[override]
+        self,
+        inputs: torch.Tensor,
+        autoencoder_model: AutoencoderKL | VQVAE,
+        diffusion_model: DiffusionModelUNet,
+        controlnet: ControlNet,
+        cn_cond: torch.Tensor,
+        scheduler: Scheduler | None = None,
+        save_intermediates: bool | None = False,
+        conditioning: torch.Tensor | None = None,
+        mode: str = "crossattn",
+        original_input_range: tuple | None = (0, 255),
+        scaled_input_range: tuple | None = (0, 1),
+        verbose: bool = True,
+        resample_latent_likelihoods: bool = False,
+        resample_interpolation_mode: str = "nearest",
+        seg: torch.Tensor | None = None,
+    ) -> torch.Tensor | tuple[torch.Tensor, list[torch.Tensor]]:
+        """
+        Computes the log-likelihoods of the latent representations of the input.
+
+        Args:
+            inputs: input images, NxCxHxW[xD]
+            autoencoder_model: first stage model.
+            diffusion_model: model to compute likelihood from
+            controlnet: instance of ControlNet model.
+            cn_cond: conditioning tensor for the ControlNet network.
+            scheduler: diffusion scheduler. If none provided will use the class attribute scheduler
+            save_intermediates: save the intermediate spatial KL maps
+            conditioning: Conditioning for network input.
+            mode: Conditioning mode for the network.
+            original_input_range: the [min,max] intensity range of the input data before any scaling was applied.
+            scaled_input_range: the [min,max] intensity range of the input data after scaling.
+            verbose: if true, prints the progression bar of the sampling process.
+            resample_latent_likelihoods: if true, resamples the intermediate likelihood maps to have the same spatial
+                dimension as the input images.
+            resample_interpolation_mode: if use resample_latent_likelihoods, select interpolation 'nearest', 'bilinear',
+                or 'trilinear;
+            seg: if diffusion model is instance of SPADEDiffusionModel, or autoencoder_model
+             is instance of SPADEAutoencoderKL, segmentation must be provided.
+        """
+        if resample_latent_likelihoods and resample_interpolation_mode not in ("nearest", "bilinear", "trilinear"):
+            raise ValueError(
+                f"resample_interpolation mode should be either nearest, bilinear, or trilinear, got {resample_interpolation_mode}"
+            )
+
+        latents = autoencoder_model.encode_stage_2_inputs(inputs) * self.scale_factor
+
+        if cn_cond.shape[2:] != latents.shape[2:]:
+            cn_cond = F.interpolate(cn_cond, latents.shape[2:])
+
+        if self.ldm_latent_shape is not None:
+            latents = torch.stack([self.ldm_resizer(i) for i in decollate_batch(latents)], 0)
+
+        outputs = super().get_likelihood(
+            inputs=latents,
+            diffusion_model=diffusion_model,
+            controlnet=controlnet,
+            cn_cond=cn_cond,
+            scheduler=scheduler,
+            save_intermediates=save_intermediates,
+            conditioning=conditioning,
+            mode=mode,
+            verbose=verbose,
+            seg=seg,
+        )
+
+        if save_intermediates and resample_latent_likelihoods:
+            intermediates = outputs[1]
+            resizer = nn.Upsample(size=inputs.shape[2:], mode=resample_interpolation_mode)
+            intermediates = [resizer(x) for x in intermediates]
+            outputs = (outputs[0], intermediates)
+        return outputs
+
+
+class VQVAETransformerInferer(nn.Module):
+    """
+    Class to perform inference with a VQVAE + Transformer model.
+    """
+
+    def __init__(self) -> None:
+        Inferer.__init__(self)
+
+    def __call__(
+        self,
+        inputs: torch.Tensor,
+        vqvae_model: VQVAE,
+        transformer_model: DecoderOnlyTransformer,
+        ordering: Ordering,
+        condition: torch.Tensor | None = None,
+        return_latent: bool = False,
+    ) -> torch.Tensor | tuple[torch.Tensor, torch.Tensor, tuple]:
+        """
+        Implements the forward pass for a supervised training iteration.
+
+        Args:
+            inputs: input image to which the latent representation will be extracted.
+            vqvae_model: first stage model.
+            transformer_model: autoregressive transformer model.
+            ordering: ordering of the quantised latent representation.
+            return_latent: also return latent sequence and spatial dim of the latent.
+            condition: conditioning for network input.
+        """
+        with torch.no_grad():
+            latent = vqvae_model.index_quantize(inputs)
+
+        latent_spatial_dim = tuple(latent.shape[1:])
+        latent = latent.reshape(latent.shape[0], -1)
+        latent = latent[:, ordering.get_sequence_ordering()]
+
+        # get the targets for the loss
+        target = latent.clone()
+        # Use the value from vqvae_model's num_embeddings as the starting token, the "Begin Of Sentence" (BOS) token.
+        # Note the transformer_model must have vqvae_model.num_embeddings + 1 defined as num_tokens.
+        latent = F.pad(latent, (1, 0), "constant", vqvae_model.num_embeddings)
+        # crop the last token as we do not need the probability of the token that follows it
+        latent = latent[:, :-1]
+        latent = latent.long()
+
+        # train on a part of the sequence if it is longer than max_seq_length
+        seq_len = latent.shape[1]
+        max_seq_len = transformer_model.max_seq_len
+        if max_seq_len < seq_len:
+            start = int(torch.randint(low=0, high=seq_len + 1 - max_seq_len, size=(1,)).item())
+        else:
+            start = 0
+        prediction: torch.Tensor = transformer_model(x=latent[:, start : start + max_seq_len], context=condition)
+        if return_latent:
+            return prediction, target[:, start : start + max_seq_len], latent_spatial_dim
+        else:
+            return prediction
+
+    @torch.no_grad()
+    def sample(
+        self,
+        latent_spatial_dim: tuple[int, int, int] | tuple[int, int],
+        starting_tokens: torch.Tensor,
+        vqvae_model: VQVAE,
+        transformer_model: DecoderOnlyTransformer,
+        ordering: Ordering,
+        conditioning: torch.Tensor | None = None,
+        temperature: float = 1.0,
+        top_k: int | None = None,
+        verbose: bool = True,
+    ) -> torch.Tensor:
+        """
+        Sampling function for the VQVAE + Transformer model.
+
+        Args:
+            latent_spatial_dim: shape of the sampled image.
+            starting_tokens: starting tokens for the sampling. It must be vqvae_model.num_embeddings value.
+            vqvae_model: first stage model.
+            transformer_model: model to sample from.
+            conditioning: Conditioning for network input.
+            temperature: temperature for sampling.
+            top_k: top k sampling.
+            verbose: if true, prints the progression bar of the sampling process.
+        """
+        seq_len = math.prod(latent_spatial_dim)
+
+        if verbose and has_tqdm:
+            progress_bar = tqdm(range(seq_len))
+        else:
+            progress_bar = iter(range(seq_len))
+
+        latent_seq = starting_tokens.long()
+        for _ in progress_bar:
+            # if the sequence context is growing too long we must crop it at block_size
+            if latent_seq.size(1) <= transformer_model.max_seq_len:
+                idx_cond = latent_seq
+            else:
+                idx_cond = latent_seq[:, -transformer_model.max_seq_len :]
+
+            # forward the model to get the logits for the index in the sequence
+            logits = transformer_model(x=idx_cond, context=conditioning)
+            # pluck the logits at the final step and scale by desired temperature
+            logits = logits[:, -1, :] / temperature
+            # optionally crop the logits to only the top k options
+            if top_k is not None:
+                v, _ = torch.topk(logits, min(top_k, logits.size(-1)))
+                logits[logits < v[:, [-1]]] = -float("Inf")
+            # apply softmax to convert logits to (normalized) probabilities
+            probs = F.softmax(logits, dim=-1)
+            # remove the chance to be sampled the BOS token
+            probs[:, vqvae_model.num_embeddings] = 0
+            # sample from the distribution
+            idx_next = torch.multinomial(probs, num_samples=1)
+            # append sampled index to the running sequence and continue
+            latent_seq = torch.cat((latent_seq, idx_next), dim=1)
+
+        latent_seq = latent_seq[:, 1:]
+        latent_seq = latent_seq[:, ordering.get_revert_sequence_ordering()]
+        latent = latent_seq.reshape((starting_tokens.shape[0],) + latent_spatial_dim)
+
+        return vqvae_model.decode_samples(latent)
+
+    @torch.no_grad()
+    def get_likelihood(
+        self,
+        inputs: torch.Tensor,
+        vqvae_model: VQVAE,
+        transformer_model: DecoderOnlyTransformer,
+        ordering: Ordering,
+        condition: torch.Tensor | None = None,
+        resample_latent_likelihoods: bool = False,
+        resample_interpolation_mode: str = "nearest",
+        verbose: bool = False,
+    ) -> torch.Tensor:
+        """
+        Computes the log-likelihoods of the latent representations of the input.
+
+        Args:
+            inputs: input images, NxCxHxW[xD]
+            vqvae_model: first stage model.
+            transformer_model: autoregressive transformer model.
+            ordering: ordering of the quantised latent representation.
+            condition: conditioning for network input.
+            resample_latent_likelihoods: if true, resamples the intermediate likelihood maps to have the same spatial
+                dimension as the input images.
+            resample_interpolation_mode: if use resample_latent_likelihoods, select interpolation 'nearest', 'bilinear',
+                or 'trilinear;
+            verbose: if true, prints the progression bar of the sampling process.
+
+        """
+        if resample_latent_likelihoods and resample_interpolation_mode not in ("nearest", "bilinear", "trilinear"):
+            raise ValueError(
+                f"resample_interpolation mode should be either nearest, bilinear, or trilinear, got {resample_interpolation_mode}"
+            )
+
+        with torch.no_grad():
+            latent = vqvae_model.index_quantize(inputs)
+
+        latent_spatial_dim = tuple(latent.shape[1:])
+        latent = latent.reshape(latent.shape[0], -1)
+        latent = latent[:, ordering.get_sequence_ordering()]
+        seq_len = math.prod(latent_spatial_dim)
+
+        # Use the value from vqvae_model's num_embeddings as the starting token, the "Begin Of Sentence" (BOS) token.
+        # Note the transformer_model must have vqvae_model.num_embeddings + 1 defined as num_tokens.
+        latent = F.pad(latent, (1, 0), "constant", vqvae_model.num_embeddings)
+        latent = latent.long()
+
+        # get the first batch, up to max_seq_length, efficiently
+        logits = transformer_model(x=latent[:, : transformer_model.max_seq_len], context=condition)
+        probs = F.softmax(logits, dim=-1)
+        # target token for each set of logits is the next token along
+        target = latent[:, 1:]
+        probs = torch.gather(probs, 2, target[:, : transformer_model.max_seq_len].unsqueeze(2)).squeeze(2)
+
+        # if we have not covered the full sequence we continue with inefficient looping
+        if probs.shape[1] < target.shape[1]:
+            if verbose and has_tqdm:
+                progress_bar = tqdm(range(transformer_model.max_seq_len, seq_len))
+            else:
+                progress_bar = iter(range(transformer_model.max_seq_len, seq_len))
+
+            for i in progress_bar:
+                idx_cond = latent[:, i + 1 - transformer_model.max_seq_len : i + 1]
+                # forward the model to get the logits for the index in the sequence
+                logits = transformer_model(x=idx_cond, context=condition)
+                # pluck the logits at the final step
+                logits = logits[:, -1, :]
+                # apply softmax to convert logits to (normalized) probabilities
+                p = F.softmax(logits, dim=-1)
+                # select correct values and append
+                p = torch.gather(p, 1, target[:, i].unsqueeze(1))
+
+                probs = torch.cat((probs, p), dim=1)
+
+        # convert to log-likelihood
+        probs = torch.log(probs)
+
+        # reshape
+        probs = probs[:, ordering.get_revert_sequence_ordering()]
+        probs_reshaped = probs.reshape((inputs.shape[0],) + latent_spatial_dim)
+        if resample_latent_likelihoods:
+            resizer = nn.Upsample(size=inputs.shape[2:], mode=resample_interpolation_mode)
+            probs_reshaped = resizer(probs_reshaped[:, None, ...])
+
+        return probs_reshaped
diff --git a/monai/networks/blocks/__init__.py b/monai/networks/blocks/__init__.py
index e67cb3376f..47abc4a1c4 100644
--- a/monai/networks/blocks/__init__.py
+++ b/monai/networks/blocks/__init__.py
@@ -17,6 +17,7 @@
 from .backbone_fpn_utils import BackboneWithFPN
 from .convolutions import Convolution, ResidualUnit
 from .crf import CRF
+from .crossattention import CrossAttentionBlock
 from .denseblock import ConvDenseBlock, DenseBlock
 from .dints_block import ActiConvNormBlock, FactorizedIncreaseBlock, FactorizedReduceBlock, P3DActiConvNormBlock
 from .downsample import MaxAvgPool
@@ -30,6 +31,8 @@
 from .regunet_block import RegistrationDownSampleBlock, RegistrationExtractionBlock, RegistrationResidualConvBlock
 from .segresnet_block import ResBlock
 from .selfattention import SABlock
+from .spade_norm import SPADE
+from .spatialattention import SpatialAttentionBlock
 from .squeeze_and_excitation import (
     ChannelSELayer,
     ResidualSELayer,
diff --git a/monai/networks/blocks/attention_utils.py b/monai/networks/blocks/attention_utils.py
new file mode 100644
index 0000000000..8c9002a16e
--- /dev/null
+++ b/monai/networks/blocks/attention_utils.py
@@ -0,0 +1,128 @@
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import Tuple
+
+import torch
+import torch.nn.functional as F
+from torch import nn
+
+
+def get_rel_pos(q_size: int, k_size: int, rel_pos: torch.Tensor) -> torch.Tensor:
+    """
+    Get relative positional embeddings according to the relative positions of
+    query and key sizes.
+
+    Args:
+        q_size (int): size of query q.
+        k_size (int): size of key k.
+        rel_pos (Tensor): relative position embeddings (L, C).
+
+    Returns:
+        Extracted positional embeddings according to relative positions.
+    """
+    rel_pos_resized: torch.Tensor = torch.Tensor()
+    max_rel_dist = int(2 * max(q_size, k_size) - 1)
+    # Interpolate rel pos if needed.
+    if rel_pos.shape[0] != max_rel_dist:
+        # Interpolate rel pos.
+        rel_pos_resized = F.interpolate(
+            rel_pos.reshape(1, rel_pos.shape[0], -1).permute(0, 2, 1), size=max_rel_dist, mode="linear"
+        )
+        rel_pos_resized = rel_pos_resized.reshape(-1, max_rel_dist).permute(1, 0)
+    else:
+        rel_pos_resized = rel_pos
+
+    # Scale the coords with short length if shapes for q and k are different.
+    q_coords = torch.arange(q_size)[:, None] * max(k_size / q_size, 1.0)
+    k_coords = torch.arange(k_size)[None, :] * max(q_size / k_size, 1.0)
+    relative_coords = (q_coords - k_coords) + (k_size - 1) * max(q_size / k_size, 1.0)
+
+    return rel_pos_resized[relative_coords.long()]
+
+
+def add_decomposed_rel_pos(
+    attn: torch.Tensor, q: torch.Tensor, rel_pos_lst: nn.ParameterList, q_size: Tuple, k_size: Tuple
+) -> torch.Tensor:
+    r"""
+    Calculate decomposed Relative Positional Embeddings from mvitv2 implementation:
+    https://github.com/facebookresearch/mvit/blob/19786631e330df9f3622e5402b4a419a263a2c80/mvit/models/attention.py
+
+    Only 2D and 3D are supported.
+
+    Encoding the relative position of tokens in the attention matrix: tokens spaced a distance
+    `d` apart will have the same embedding value (unlike absolute positional embedding).
+
+    .. math::
+        Attn_{logits}(Q, K) = (QK^{T} + E_{rel})*scale
+
+    where
+
+    .. math::
+        E_{ij}^{(rel)} = Q_{i}.R_{p(i), p(j)}
+
+    with :math:`R_{p(i), p(j)} \in R^{dim}` and :math:`p(i), p(j)`,
+    respectively spatial positions of element :math:`i` and :math:`j`
+
+    When using "decomposed" relative positional embedding, positional embedding is defined ("decomposed") as follow:
+
+    .. math::
+        R_{p(i), p(j)} = R^{d1}_{d1(i), d1(j)} + ... + R^{dn}_{dn(i), dn(j)}
+
+    with :math:`n = 1...dim`
+
+    Decomposed relative positional embedding reduces the complexity from :math:`\mathcal{O}(d1*...*dn)` to
+    :math:`\mathcal{O}(d1+...+dn)` compared with classical relative positional embedding.
+
+    Args:
+        attn (Tensor): attention map.
+        q (Tensor): query q in the attention layer with shape (B, s_dim_1 * ... * s_dim_n, C).
+        rel_pos_lst (ParameterList): relative position embeddings for each axis: rel_pos_lst[n] for nth axis.
+        q_size (Tuple): spatial sequence size of query q with (q_dim_1, ..., q_dim_n).
+        k_size (Tuple): spatial sequence size of key k with (k_dim_1, ...,  k_dim_n).
+
+    Returns:
+        attn (Tensor): attention logits with added relative positional embeddings.
+    """
+    rh = get_rel_pos(q_size[0], k_size[0], rel_pos_lst[0])
+    rw = get_rel_pos(q_size[1], k_size[1], rel_pos_lst[1])
+
+    batch, _, dim = q.shape
+
+    if len(rel_pos_lst) == 2:
+        q_h, q_w = q_size[:2]
+        k_h, k_w = k_size[:2]
+        r_q = q.reshape(batch, q_h, q_w, dim)
+        rel_h = torch.einsum("bhwc,hkc->bhwk", r_q, rh)
+        rel_w = torch.einsum("bhwc,wkc->bhwk", r_q, rw)
+
+        attn = (attn.view(batch, q_h, q_w, k_h, k_w) + rel_h[:, :, :, :, None] + rel_w[:, :, :, None, :]).view(
+            batch, q_h * q_w, k_h * k_w
+        )
+    elif len(rel_pos_lst) == 3:
+        q_h, q_w, q_d = q_size[:3]
+        k_h, k_w, k_d = k_size[:3]
+
+        rd = get_rel_pos(q_d, k_d, rel_pos_lst[2])
+
+        r_q = q.reshape(batch, q_h, q_w, q_d, dim)
+        rel_h = torch.einsum("bhwdc,hkc->bhwdk", r_q, rh)
+        rel_w = torch.einsum("bhwdc,wkc->bhwdk", r_q, rw)
+        rel_d = torch.einsum("bhwdc,wkc->bhwdk", r_q, rd)
+
+        attn = (
+            attn.view(batch, q_h, q_w, q_d, k_h, k_w, k_d)
+            + rel_h[:, :, :, :, None, None]
+            + rel_w[:, :, :, None, :, None]
+            + rel_d[:, :, :, None, None, :]
+        ).view(batch, q_h * q_w * q_d, k_h * k_w * k_d)
+
+    return attn
diff --git a/monai/networks/blocks/crossattention.py b/monai/networks/blocks/crossattention.py
new file mode 100644
index 0000000000..dc1d5d388e
--- /dev/null
+++ b/monai/networks/blocks/crossattention.py
@@ -0,0 +1,166 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import Optional, Tuple
+
+import torch
+import torch.nn as nn
+
+from monai.networks.layers.utils import get_rel_pos_embedding_layer
+from monai.utils import optional_import
+
+Rearrange, _ = optional_import("einops.layers.torch", name="Rearrange")
+
+
+class CrossAttentionBlock(nn.Module):
+    """
+    A cross-attention block, based on: "Dosovitskiy et al.,
+    An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale <https://arxiv.org/abs/2010.11929>"
+    One can setup relative positional embedding as described in <https://arxiv.org/abs/2112.01526>
+    """
+
+    def __init__(
+        self,
+        hidden_size: int,
+        num_heads: int,
+        dropout_rate: float = 0.0,
+        hidden_input_size: int | None = None,
+        context_input_size: int | None = None,
+        dim_head: int | None = None,
+        qkv_bias: bool = False,
+        save_attn: bool = False,
+        causal: bool = False,
+        sequence_length: int | None = None,
+        rel_pos_embedding: Optional[str] = None,
+        input_size: Optional[Tuple] = None,
+        attention_dtype: Optional[torch.dtype] = None,
+    ) -> None:
+        """
+        Args:
+            hidden_size (int): dimension of hidden layer.
+            num_heads (int): number of attention heads.
+            dropout_rate (float, optional): fraction of the input units to drop. Defaults to 0.0.
+            hidden_input_size (int, optional): dimension of the input tensor. Defaults to hidden_size.
+            context_input_size (int, optional): dimension of the context tensor. Defaults to hidden_size.
+            dim_head (int, optional): dimension of each head. Defaults to hidden_size // num_heads.
+            qkv_bias (bool, optional): bias term for the qkv linear layer. Defaults to False.
+            save_attn (bool, optional): to make accessible the attention matrix. Defaults to False.
+            causal: whether to use causal attention.
+            sequence_length: if causal is True, it is necessary to specify the sequence length.
+            rel_pos_embedding (str, optional): Add relative positional embeddings to the attention map.
+                For now only "decomposed" is supported (see https://arxiv.org/abs/2112.01526). 2D and 3D are supported.
+            input_size (tuple(spatial_dim), optional): Input resolution for calculating the relative
+                positional parameter size.
+            attention_dtype: cast attention operations to this dtype.
+        """
+
+        super().__init__()
+
+        if not (0 <= dropout_rate <= 1):
+            raise ValueError("dropout_rate should be between 0 and 1.")
+
+        if dim_head:
+            inner_size = num_heads * dim_head
+            self.head_dim = dim_head
+        else:
+            if hidden_size % num_heads != 0:
+                raise ValueError("hidden size should be divisible by num_heads.")
+            inner_size = hidden_size
+            self.head_dim = hidden_size // num_heads
+
+        if causal and sequence_length is None:
+            raise ValueError("sequence_length is necessary for causal attention.")
+
+        self.num_heads = num_heads
+        self.hidden_input_size = hidden_input_size if hidden_input_size else hidden_size
+        self.context_input_size = context_input_size if context_input_size else hidden_size
+        self.out_proj = nn.Linear(inner_size, self.hidden_input_size)
+        # key, query, value projections
+        self.to_q = nn.Linear(self.hidden_input_size, inner_size, bias=qkv_bias)
+        self.to_k = nn.Linear(self.context_input_size, inner_size, bias=qkv_bias)
+        self.to_v = nn.Linear(self.context_input_size, inner_size, bias=qkv_bias)
+        self.input_rearrange = Rearrange("b h (l d) -> b l h d", l=num_heads)
+
+        self.out_rearrange = Rearrange("b h l d -> b l (h d)")
+        self.drop_output = nn.Dropout(dropout_rate)
+        self.drop_weights = nn.Dropout(dropout_rate)
+
+        self.scale = self.head_dim**-0.5
+        self.save_attn = save_attn
+        self.attention_dtype = attention_dtype
+
+        self.causal = causal
+        self.sequence_length = sequence_length
+
+        if causal and sequence_length is not None:
+            # causal mask to ensure that attention is only applied to the left in the input sequence
+            self.register_buffer(
+                "causal_mask",
+                torch.tril(torch.ones(sequence_length, sequence_length)).view(1, 1, sequence_length, sequence_length),
+            )
+            self.causal_mask: torch.Tensor
+
+        self.att_mat = torch.Tensor()
+        self.rel_positional_embedding = (
+            get_rel_pos_embedding_layer(rel_pos_embedding, input_size, self.head_dim, self.num_heads)
+            if rel_pos_embedding is not None
+            else None
+        )
+        self.input_size = input_size
+
+    def forward(self, x: torch.Tensor, context: torch.Tensor | None = None):
+        """
+        Args:
+            x (torch.Tensor): input tensor. B x (s_dim_1 * ... * s_dim_n) x C
+            context (torch.Tensor, optional): context tensor. B x (s_dim_1 * ... * s_dim_n) x C
+
+        Return:
+            torch.Tensor: B x (s_dim_1 * ... * s_dim_n) x C
+        """
+        # calculate query, key, values for all heads in batch and move head forward to be the batch dim
+        b, t, c = x.size()  # batch size, sequence length, embedding dimensionality (hidden_size)
+
+        q = self.to_q(x)
+        kv = context if context is not None else x
+        _, kv_t, _ = kv.size()
+        k = self.to_k(kv)
+        v = self.to_v(kv)
+
+        if self.attention_dtype is not None:
+            q = q.to(self.attention_dtype)
+            k = k.to(self.attention_dtype)
+
+        q = q.view(b, t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, t,  hs)
+        k = k.view(b, kv_t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, kv_t, hs)
+        v = v.view(b, kv_t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, kv_t, hs)
+        att_mat = torch.einsum("blxd,blyd->blxy", q, k) * self.scale
+
+        # apply relative positional embedding if defined
+        att_mat = self.rel_positional_embedding(x, att_mat, q) if self.rel_positional_embedding is not None else att_mat
+
+        if self.causal:
+            att_mat = att_mat.masked_fill(self.causal_mask[:, :, :t, :kv_t] == 0, float("-inf"))
+
+        att_mat = att_mat.softmax(dim=-1)
+
+        if self.save_attn:
+            # no gradients and new tensor;
+            # https://pytorch.org/docs/stable/generated/torch.Tensor.detach.html
+            self.att_mat = att_mat.detach()
+
+        att_mat = self.drop_weights(att_mat)
+        x = torch.einsum("bhxy,bhyd->bhxd", att_mat, v)
+        x = self.out_rearrange(x)
+        x = self.out_proj(x)
+        x = self.drop_output(x)
+        return x
diff --git a/monai/networks/blocks/rel_pos_embedding.py b/monai/networks/blocks/rel_pos_embedding.py
new file mode 100644
index 0000000000..e53e5841b0
--- /dev/null
+++ b/monai/networks/blocks/rel_pos_embedding.py
@@ -0,0 +1,56 @@
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import Iterable, Tuple
+
+import torch
+from torch import nn
+
+from monai.networks.blocks.attention_utils import add_decomposed_rel_pos
+from monai.utils.misc import ensure_tuple_size
+
+
+class DecomposedRelativePosEmbedding(nn.Module):
+    def __init__(self, s_input_dims: Tuple[int, int] | Tuple[int, int, int], c_dim: int, num_heads: int) -> None:
+        """
+        Args:
+            s_input_dims (Tuple): input spatial dimension. (H, W) or (H, W, D)
+            c_dim (int): channel dimension
+            num_heads(int): number of attention heads
+        """
+        super().__init__()
+
+        # validate inputs
+        if not isinstance(s_input_dims, Iterable) or len(s_input_dims) not in [2, 3]:
+            raise ValueError("s_input_dims must be set as follows: (H, W) or (H, W, D)")
+
+        self.s_input_dims = s_input_dims
+        self.c_dim = c_dim
+        self.num_heads = num_heads
+        self.rel_pos_arr = nn.ParameterList(
+            [nn.Parameter(torch.zeros(2 * dim_input_size - 1, c_dim)) for dim_input_size in s_input_dims]
+        )
+
+    def forward(self, x: torch.Tensor, att_mat: torch.Tensor, q: torch.Tensor) -> torch.Tensor:
+        """"""
+        batch = x.shape[0]
+        h, w, d = ensure_tuple_size(self.s_input_dims, 3, 1)
+
+        att_mat = add_decomposed_rel_pos(
+            att_mat.contiguous().view(batch * self.num_heads, h * w * d, h * w * d),
+            q.contiguous().view(batch * self.num_heads, h * w * d, -1),
+            self.rel_pos_arr,
+            (h, w) if d == 1 else (h, w, d),
+            (h, w) if d == 1 else (h, w, d),
+        )
+
+        att_mat = att_mat.reshape(batch, self.num_heads, h * w * d, h * w * d)
+        return att_mat
diff --git a/monai/networks/blocks/selfattention.py b/monai/networks/blocks/selfattention.py
index 7b410b1a7c..9905e7d036 100644
--- a/monai/networks/blocks/selfattention.py
+++ b/monai/networks/blocks/selfattention.py
@@ -11,9 +11,12 @@
 
 from __future__ import annotations
 
+from typing import Optional, Tuple
+
 import torch
 import torch.nn as nn
 
+from monai.networks.layers.utils import get_rel_pos_embedding_layer
 from monai.utils import optional_import
 
 Rearrange, _ = optional_import("einops.layers.torch", name="Rearrange")
@@ -33,6 +36,12 @@ def __init__(
         qkv_bias: bool = False,
         save_attn: bool = False,
         dim_head: int | None = None,
+        hidden_input_size: int | None = None,
+        causal: bool = False,
+        sequence_length: int | None = None,
+        rel_pos_embedding: Optional[str] = None,
+        input_size: Optional[Tuple] = None,
+        attention_dtype: Optional[torch.dtype] = None,
     ) -> None:
         """
         Args:
@@ -42,6 +51,14 @@ def __init__(
             qkv_bias (bool, optional): bias term for the qkv linear layer. Defaults to False.
             save_attn (bool, optional): to make accessible the attention matrix. Defaults to False.
             dim_head (int, optional): dimension of each head. Defaults to hidden_size // num_heads.
+            hidden_input_size (int, optional): dimension of the input tensor. Defaults to hidden_size.
+            causal: whether to use causal attention (see https://arxiv.org/abs/1706.03762).
+            sequence_length: if causal is True, it is necessary to specify the sequence length.
+            rel_pos_embedding (str, optional): Add relative positional embeddings to the attention map.
+                For now only "decomposed" is supported (see https://arxiv.org/abs/2112.01526). 2D and 3D are supported.
+            input_size (tuple(spatial_dim), optional): Input resolution for calculating the relative
+                positional parameter size.
+            attention_dtype: cast attention operations to this dtype.
 
         """
 
@@ -53,12 +70,23 @@ def __init__(
         if hidden_size % num_heads != 0:
             raise ValueError("hidden size should be divisible by num_heads.")
 
+        if dim_head:
+            self.inner_dim = num_heads * dim_head
+            self.dim_head = dim_head
+        else:
+            if hidden_size % num_heads != 0:
+                raise ValueError("hidden size should be divisible by num_heads.")
+            self.inner_dim = hidden_size
+            self.dim_head = hidden_size // num_heads
+
+        if causal and sequence_length is None:
+            raise ValueError("sequence_length is necessary for causal attention.")
+
         self.num_heads = num_heads
-        self.dim_head = hidden_size // num_heads if dim_head is None else dim_head
-        self.inner_dim = self.dim_head * num_heads
+        self.hidden_input_size = hidden_input_size if hidden_input_size else hidden_size
+        self.out_proj = nn.Linear(self.inner_dim, self.hidden_input_size)
 
-        self.out_proj = nn.Linear(self.inner_dim, hidden_size)
-        self.qkv = nn.Linear(hidden_size, self.inner_dim * 3, bias=qkv_bias)
+        self.qkv = nn.Linear(self.hidden_input_size, self.inner_dim * 3, bias=qkv_bias)
         self.input_rearrange = Rearrange("b h (qkv l d) -> qkv b l h d", qkv=3, l=num_heads)
         self.out_rearrange = Rearrange("b h l d -> b l (h d)")
         self.drop_output = nn.Dropout(dropout_rate)
@@ -66,11 +94,50 @@ def __init__(
         self.scale = self.dim_head**-0.5
         self.save_attn = save_attn
         self.att_mat = torch.Tensor()
+        self.attention_dtype = attention_dtype
+        self.causal = causal
+        self.sequence_length = sequence_length
+
+        if causal and sequence_length is not None:
+            # causal mask to ensure that attention is only applied to the left in the input sequence
+            self.register_buffer(
+                "causal_mask",
+                torch.tril(torch.ones(sequence_length, sequence_length)).view(1, 1, sequence_length, sequence_length),
+            )
+            self.causal_mask: torch.Tensor
+
+        self.rel_positional_embedding = (
+            get_rel_pos_embedding_layer(rel_pos_embedding, input_size, self.dim_head, self.num_heads)
+            if rel_pos_embedding is not None
+            else None
+        )
+        self.input_size = input_size
 
     def forward(self, x):
+        """
+        Args:
+            x (torch.Tensor): input tensor. B x (s_dim_1 * ... * s_dim_n) x C
+
+        Return:
+            torch.Tensor: B x (s_dim_1 * ... * s_dim_n) x C
+        """
         output = self.input_rearrange(self.qkv(x))
         q, k, v = output[0], output[1], output[2]
-        att_mat = (torch.einsum("blxd,blyd->blxy", q, k) * self.scale).softmax(dim=-1)
+
+        if self.attention_dtype is not None:
+            q = q.to(self.attention_dtype)
+            k = k.to(self.attention_dtype)
+
+        att_mat = torch.einsum("blxd,blyd->blxy", q, k) * self.scale
+
+        # apply relative positional embedding if defined
+        att_mat = self.rel_positional_embedding(x, att_mat, q) if self.rel_positional_embedding is not None else att_mat
+
+        if self.causal:
+            att_mat = att_mat.masked_fill(self.causal_mask[:, :, : x.shape[1], : x.shape[1]] == 0, float("-inf"))
+
+        att_mat = att_mat.softmax(dim=-1)
+
         if self.save_attn:
             # no gradients and new tensor;
             # https://pytorch.org/docs/stable/generated/torch.Tensor.detach.html
diff --git a/monai/networks/blocks/spade_norm.py b/monai/networks/blocks/spade_norm.py
new file mode 100644
index 0000000000..343dfa9ec0
--- /dev/null
+++ b/monai/networks/blocks/spade_norm.py
@@ -0,0 +1,95 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+from monai.networks.blocks import Convolution
+from monai.networks.layers.utils import get_norm_layer
+
+
+class SPADE(nn.Module):
+    """
+    Spatially Adaptive Normalization (SPADE) block, allowing for normalization of activations conditioned on a
+    semantic map. This block is used in SPADE-based image-to-image translation models, as described in
+    Semantic Image Synthesis with Spatially-Adaptive Normalization (https://arxiv.org/abs/1903.07291).
+
+    Args:
+        label_nc: number of semantic labels
+        norm_nc: number of output channels
+        kernel_size: kernel size
+        spatial_dims: number of spatial dimensions
+        hidden_channels: number of channels in the intermediate gamma and beta layers
+        norm: type of base normalisation used before applying the SPADE normalisation
+        norm_params: parameters for the base normalisation
+    """
+
+    def __init__(
+        self,
+        label_nc: int,
+        norm_nc: int,
+        kernel_size: int = 3,
+        spatial_dims: int = 2,
+        hidden_channels: int = 64,
+        norm: str | tuple = "INSTANCE",
+        norm_params: dict | None = None,
+    ) -> None:
+        super().__init__()
+
+        if norm_params is None:
+            norm_params = {}
+        if len(norm_params) != 0:
+            norm = (norm, norm_params)
+        self.param_free_norm = get_norm_layer(norm, spatial_dims=spatial_dims, channels=norm_nc)
+        self.mlp_shared = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=label_nc,
+            out_channels=hidden_channels,
+            kernel_size=kernel_size,
+            norm=None,
+            act="LEAKYRELU",
+        )
+        self.mlp_gamma = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=hidden_channels,
+            out_channels=norm_nc,
+            kernel_size=kernel_size,
+            act=None,
+        )
+        self.mlp_beta = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=hidden_channels,
+            out_channels=norm_nc,
+            kernel_size=kernel_size,
+            act=None,
+        )
+
+    def forward(self, x: torch.Tensor, segmap: torch.Tensor) -> torch.Tensor:
+        """
+        Args:
+            x: input tensor with shape (B, C, [spatial-dimensions]) where C is the number of semantic channels.
+            segmap: input segmentation map (B, C, [spatial-dimensions]) where C is the number of semantic channels.
+            The map will be interpolated to the dimension of x internally.
+        """
+
+        # Part 1. generate parameter-free normalized activations
+        normalized = self.param_free_norm(x.contiguous())
+
+        # Part 2. produce scaling and bias conditioned on semantic map
+        segmap = F.interpolate(segmap, size=x.size()[2:], mode="nearest")
+        actv = self.mlp_shared(segmap)
+        gamma = self.mlp_gamma(actv)
+        beta = self.mlp_beta(actv)
+        out: torch.Tensor = normalized * (1 + gamma) + beta
+        return out
diff --git a/monai/networks/blocks/spatialattention.py b/monai/networks/blocks/spatialattention.py
new file mode 100644
index 0000000000..75319853d9
--- /dev/null
+++ b/monai/networks/blocks/spatialattention.py
@@ -0,0 +1,82 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import Optional
+
+import torch
+import torch.nn as nn
+
+from monai.networks.blocks import SABlock
+from monai.utils import optional_import
+
+Rearrange, _ = optional_import("einops.layers.torch", name="Rearrange")
+
+
+class SpatialAttentionBlock(nn.Module):
+    """Perform spatial self-attention on the input tensor.
+
+    The input tensor is reshaped to B x (x_dim * y_dim [ * z_dim]) x C, where C is the number of channels, and then
+    self-attention is performed on the reshaped tensor. The output tensor is reshaped back to the original shape.
+
+    Args:
+        spatial_dims: number of spatial dimensions, could be 1, 2, or 3.
+        num_channels: number of input channels. Must be divisible by num_head_channels.
+        num_head_channels: number of channels per head.
+        attention_dtype: cast attention operations to this dtype.
+
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        num_channels: int,
+        num_head_channels: int | None = None,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        attention_dtype: Optional[torch.dtype] = None,
+    ) -> None:
+        super().__init__()
+
+        self.spatial_dims = spatial_dims
+        self.norm = nn.GroupNorm(num_groups=norm_num_groups, num_channels=num_channels, eps=norm_eps, affine=True)
+        # check num_head_channels is divisible by num_channels
+        if num_head_channels is not None and num_channels % num_head_channels != 0:
+            raise ValueError("num_channels must be divisible by num_head_channels")
+        num_heads = num_channels // num_head_channels if num_head_channels is not None else 1
+        self.attn = SABlock(
+            hidden_size=num_channels, num_heads=num_heads, qkv_bias=True, attention_dtype=attention_dtype
+        )
+
+    def forward(self, x: torch.Tensor):
+        residual = x
+
+        if self.spatial_dims == 1:
+            h = x.shape[2]
+            rearrange_input = Rearrange("b c h -> b h c")
+            rearrange_output = Rearrange("b h c -> b c h", h=h)
+        if self.spatial_dims == 2:
+            h, w = x.shape[2], x.shape[3]
+            rearrange_input = Rearrange("b c h w -> b (h w) c")
+            rearrange_output = Rearrange("b (h w) c -> b c h w", h=h, w=w)
+        else:
+            h, w, d = x.shape[2], x.shape[3], x.shape[4]
+            rearrange_input = Rearrange("b c h w d -> b (h w d) c")
+            rearrange_output = Rearrange("b (h w d) c -> b c h w d", h=h, w=w, d=d)
+
+        x = self.norm(x)
+        x = rearrange_input(x)  # B x (x_dim * y_dim [ * z_dim]) x C
+
+        x = self.attn(x)
+        x = rearrange_output(x)  # B x  x C x x_dim * y_dim * [z_dim]
+        x = x + residual
+        return x
diff --git a/monai/networks/blocks/transformerblock.py b/monai/networks/blocks/transformerblock.py
index ddf959dad2..2458902cba 100644
--- a/monai/networks/blocks/transformerblock.py
+++ b/monai/networks/blocks/transformerblock.py
@@ -11,10 +11,10 @@
 
 from __future__ import annotations
 
+import torch
 import torch.nn as nn
 
-from monai.networks.blocks.mlp import MLPBlock
-from monai.networks.blocks.selfattention import SABlock
+from monai.networks.blocks import CrossAttentionBlock, MLPBlock, SABlock
 
 
 class TransformerBlock(nn.Module):
@@ -31,6 +31,9 @@ def __init__(
         dropout_rate: float = 0.0,
         qkv_bias: bool = False,
         save_attn: bool = False,
+        causal: bool = False,
+        sequence_length: int | None = None,
+        with_cross_attention: bool = False,
     ) -> None:
         """
         Args:
@@ -53,10 +56,27 @@ def __init__(
 
         self.mlp = MLPBlock(hidden_size, mlp_dim, dropout_rate)
         self.norm1 = nn.LayerNorm(hidden_size)
-        self.attn = SABlock(hidden_size, num_heads, dropout_rate, qkv_bias, save_attn)
+        self.attn = SABlock(
+            hidden_size,
+            num_heads,
+            dropout_rate,
+            qkv_bias=qkv_bias,
+            save_attn=save_attn,
+            causal=causal,
+            sequence_length=sequence_length,
+        )
         self.norm2 = nn.LayerNorm(hidden_size)
+        self.with_cross_attention = with_cross_attention
 
-    def forward(self, x):
+        if self.with_cross_attention:
+            self.norm_cross_attn = nn.LayerNorm(hidden_size)
+            self.cross_attn = CrossAttentionBlock(
+                hidden_size=hidden_size, num_heads=num_heads, dropout_rate=dropout_rate, qkv_bias=qkv_bias, causal=False
+            )
+
+    def forward(self, x: torch.Tensor, context: torch.Tensor | None = None) -> torch.Tensor:
         x = x + self.attn(self.norm1(x))
+        if self.with_cross_attention:
+            x = x + self.cross_attn(self.norm_cross_attn(x), context=context)
         x = x + self.mlp(self.norm2(x))
         return x
diff --git a/monai/networks/blocks/upsample.py b/monai/networks/blocks/upsample.py
index dee9966919..50fd39a70b 100644
--- a/monai/networks/blocks/upsample.py
+++ b/monai/networks/blocks/upsample.py
@@ -17,8 +17,8 @@
 import torch.nn as nn
 
 from monai.networks.layers.factories import Conv, Pad, Pool
-from monai.networks.utils import icnr_init, pixelshuffle
-from monai.utils import InterpolateMode, UpsampleMode, ensure_tuple_rep, look_up_option
+from monai.networks.utils import CastTempType, icnr_init, pixelshuffle
+from monai.utils import InterpolateMode, UpsampleMode, ensure_tuple_rep, look_up_option, pytorch_after
 
 __all__ = ["Upsample", "UpSample", "SubpixelUpsample", "Subpixelupsample", "SubpixelUpSample"]
 
@@ -50,6 +50,7 @@ def __init__(
         size: tuple[int] | int | None = None,
         mode: UpsampleMode | str = UpsampleMode.DECONV,
         pre_conv: nn.Module | str | None = "default",
+        post_conv: nn.Module | None = None,
         interp_mode: str = InterpolateMode.LINEAR,
         align_corners: bool | None = True,
         bias: bool = True,
@@ -71,6 +72,7 @@ def __init__(
             pre_conv: a conv block applied before upsampling. Defaults to "default".
                 When ``conv_block`` is ``"default"``, one reserved conv layer will be utilized when
                 Only used in the "nontrainable" or "pixelshuffle" mode.
+            post_conv: a conv block applied after upsampling. Defaults to None. Only used in the "nontrainable"  mode.
             interp_mode: {``"nearest"``, ``"linear"``, ``"bilinear"``, ``"bicubic"``, ``"trilinear"``}
                 Only used in the "nontrainable" mode.
                 If ends with ``"linear"`` will use ``spatial dims`` to determine the correct interpolation.
@@ -154,15 +156,25 @@ def __init__(
             linear_mode = [InterpolateMode.LINEAR, InterpolateMode.BILINEAR, InterpolateMode.TRILINEAR]
             if interp_mode in linear_mode:  # choose mode based on dimensions
                 interp_mode = linear_mode[spatial_dims - 1]
-            self.add_module(
-                "upsample_non_trainable",
-                nn.Upsample(
-                    size=size,
-                    scale_factor=None if size else scale_factor_,
-                    mode=interp_mode.value,
-                    align_corners=align_corners,
-                ),
+
+            upsample = nn.Upsample(
+                size=size,
+                scale_factor=None if size else scale_factor_,
+                mode=interp_mode.value,
+                align_corners=align_corners,
             )
+
+            # Cast to float32 as 'upsample_nearest2d_out_frame' op does not support bfloat16
+            # https://github.com/pytorch/pytorch/issues/86679. This issue is solved in PyTorch 2.1
+            if pytorch_after(major=2, minor=1):
+                self.add_module("upsample_non_trainable", upsample)
+            else:
+                self.add_module(
+                    "upsample_non_trainable",
+                    CastTempType(initial_type=torch.bfloat16, temporary_type=torch.float32, submodule=upsample),
+                )
+            if post_conv:
+                self.add_module("postconv", post_conv)
         elif up_mode == UpsampleMode.PIXELSHUFFLE:
             self.add_module(
                 "pixelshuffle",
diff --git a/monai/networks/layers/__init__.py b/monai/networks/layers/__init__.py
index 3a6e4aa554..48c10270b1 100644
--- a/monai/networks/layers/__init__.py
+++ b/monai/networks/layers/__init__.py
@@ -14,7 +14,7 @@
 from .conjugate_gradient import ConjugateGradient
 from .convutils import calculate_out_shape, gaussian_1d, polyval, same_padding, stride_minus_kernel_padding
 from .drop_path import DropPath
-from .factories import Act, Conv, Dropout, LayerFactory, Norm, Pad, Pool, split_args
+from .factories import Act, Conv, Dropout, LayerFactory, Norm, Pad, Pool, RelPosEmbedding, split_args
 from .filtering import BilateralFilter, PHLFilter, TrainableBilateralFilter, TrainableJointBilateralFilter
 from .gmm import GaussianMixtureModel
 from .simplelayers import (
@@ -38,4 +38,5 @@
 )
 from .spatial_transforms import AffineTransform, grid_count, grid_grad, grid_pull, grid_push
 from .utils import get_act_layer, get_dropout_layer, get_norm_layer, get_pool_layer
+from .vector_quantizer import EMAQuantizer, VectorQuantizer
 from .weight_init import _no_grad_trunc_normal_, trunc_normal_
diff --git a/monai/networks/layers/factories.py b/monai/networks/layers/factories.py
index 4fc2c16f73..29b72a4f37 100644
--- a/monai/networks/layers/factories.py
+++ b/monai/networks/layers/factories.py
@@ -70,7 +70,7 @@ def use_factory(fact_args):
 from monai.networks.utils import has_nvfuser_instance_norm
 from monai.utils import ComponentStore, look_up_option, optional_import
 
-__all__ = ["LayerFactory", "Dropout", "Norm", "Act", "Conv", "Pool", "Pad", "split_args"]
+__all__ = ["LayerFactory", "Dropout", "Norm", "Act", "Conv", "Pool", "Pad", "RelPosEmbedding", "split_args"]
 
 
 class LayerFactory(ComponentStore):
@@ -201,6 +201,10 @@ def split_args(args):
 Conv = LayerFactory(name="Convolution layers", description="Factory for creating convolution layers.")
 Pool = LayerFactory(name="Pooling layers", description="Factory for creating pooling layers.")
 Pad = LayerFactory(name="Padding layers", description="Factory for creating padding layers.")
+RelPosEmbedding = LayerFactory(
+    name="Relative positional embedding layers",
+    description="Factory for creating relative positional embedding factory",
+)
 
 
 @Dropout.factory_function("dropout")
@@ -468,3 +472,10 @@ def constant_pad_factory(dim: int) -> type[nn.ConstantPad1d | nn.ConstantPad2d |
     """
     types = (nn.ConstantPad1d, nn.ConstantPad2d, nn.ConstantPad3d)
     return types[dim - 1]
+
+
+@RelPosEmbedding.factory_function("decomposed")
+def decomposed_rel_pos_embedding() -> type[nn.Module]:
+    from monai.networks.blocks.rel_pos_embedding import DecomposedRelativePosEmbedding
+
+    return DecomposedRelativePosEmbedding
diff --git a/monai/networks/layers/utils.py b/monai/networks/layers/utils.py
index ace1af27b6..8676f74638 100644
--- a/monai/networks/layers/utils.py
+++ b/monai/networks/layers/utils.py
@@ -11,9 +11,11 @@
 
 from __future__ import annotations
 
+from typing import Optional
+
 import torch.nn
 
-from monai.networks.layers.factories import Act, Dropout, Norm, Pool, split_args
+from monai.networks.layers.factories import Act, Dropout, Norm, Pool, RelPosEmbedding, split_args
 from monai.utils import has_option
 
 __all__ = ["get_norm_layer", "get_act_layer", "get_dropout_layer", "get_pool_layer"]
@@ -124,3 +126,14 @@ def get_pool_layer(name: tuple | str, spatial_dims: int | None = 1):
     pool_name, pool_args = split_args(name)
     pool_type = Pool[pool_name, spatial_dims]
     return pool_type(**pool_args)
+
+
+def get_rel_pos_embedding_layer(name: tuple | str, s_input_dims: Optional[tuple], c_dim: int, num_heads: int):
+    embedding_name, embedding_args = split_args(name)
+    embedding_type = RelPosEmbedding[embedding_name]
+    # create a dictionary with the default values which can be overridden by embedding_args
+    kw_args = {"s_input_dims": s_input_dims, "c_dim": c_dim, "num_heads": num_heads, **embedding_args}
+    # filter out unused argument names
+    kw_args = {k: v for k, v in kw_args.items() if has_option(embedding_type, k)}
+
+    return embedding_type(**kw_args)
diff --git a/monai/networks/layers/vector_quantizer.py b/monai/networks/layers/vector_quantizer.py
new file mode 100644
index 0000000000..9c354e1009
--- /dev/null
+++ b/monai/networks/layers/vector_quantizer.py
@@ -0,0 +1,233 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import Sequence, Tuple
+
+import torch
+from torch import nn
+
+__all__ = ["VectorQuantizer", "EMAQuantizer"]
+
+
+class EMAQuantizer(nn.Module):
+    """
+    Vector Quantization module using Exponential Moving Average (EMA) to learn the codebook parameters based on  Neural
+    Discrete Representation Learning by Oord et al. (https://arxiv.org/abs/1711.00937) and the official implementation
+    that can be found at https://github.com/deepmind/sonnet/blob/v2/sonnet/src/nets/vqvae.py#L148 and commit
+    58d9a2746493717a7c9252938da7efa6006f3739.
+
+    This module is not compatible with TorchScript while working in a Distributed Data Parallelism Module. This is due
+    to lack of TorchScript support for torch.distributed module as per https://github.com/pytorch/pytorch/issues/41353
+    on 22/10/2022. If you want to TorchScript your model, please turn set `ddp_sync` to False.
+
+    Args:
+        spatial_dims: number of spatial dimensions of the input.
+        num_embeddings: number of atomic elements in the codebook.
+        embedding_dim: number of channels of the input and atomic elements.
+        commitment_cost: scaling factor of the MSE loss between input and its quantized version. Defaults to 0.25.
+        decay: EMA decay. Defaults to 0.99.
+        epsilon: epsilon value. Defaults to 1e-5.
+        embedding_init: initialization method for the codebook. Defaults to "normal".
+        ddp_sync: whether to synchronize the codebook across processes. Defaults to True.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        num_embeddings: int,
+        embedding_dim: int,
+        commitment_cost: float = 0.25,
+        decay: float = 0.99,
+        epsilon: float = 1e-5,
+        embedding_init: str = "normal",
+        ddp_sync: bool = True,
+    ):
+        super().__init__()
+        self.spatial_dims: int = spatial_dims
+        self.embedding_dim: int = embedding_dim
+        self.num_embeddings: int = num_embeddings
+
+        assert self.spatial_dims in [2, 3], ValueError(
+            f"EMAQuantizer only supports 4D and 5D tensor inputs but received spatial dims {spatial_dims}."
+        )
+
+        self.embedding: torch.nn.Embedding = torch.nn.Embedding(self.num_embeddings, self.embedding_dim)
+        if embedding_init == "normal":
+            # Initialization is passed since the default one is normal inside the nn.Embedding
+            pass
+        elif embedding_init == "kaiming_uniform":
+            torch.nn.init.kaiming_uniform_(self.embedding.weight.data, mode="fan_in", nonlinearity="linear")
+        self.embedding.weight.requires_grad = False
+
+        self.commitment_cost: float = commitment_cost
+
+        self.register_buffer("ema_cluster_size", torch.zeros(self.num_embeddings))
+        self.register_buffer("ema_w", self.embedding.weight.data.clone())
+        # declare types for mypy
+        self.ema_cluster_size: torch.Tensor
+        self.ema_w: torch.Tensor
+        self.decay: float = decay
+        self.epsilon: float = epsilon
+
+        self.ddp_sync: bool = ddp_sync
+
+        # Precalculating required permutation shapes
+        self.flatten_permutation = [0] + list(range(2, self.spatial_dims + 2)) + [1]
+        self.quantization_permutation: Sequence[int] = [0, self.spatial_dims + 1] + list(
+            range(1, self.spatial_dims + 1)
+        )
+
+    def quantize(self, inputs: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
+        """
+        Given an input it projects it to the quantized space and returns additional tensors needed for EMA loss.
+
+        Args:
+            inputs: Encoding space tensors of shape [B, C, H, W, D].
+
+        Returns:
+            torch.Tensor: Flatten version of the input of shape [B*H*W*D, C].
+            torch.Tensor: One-hot representation of the quantization indices of shape [B*H*W*D, self.num_embeddings].
+            torch.Tensor: Quantization indices of shape [B,H,W,D,1]
+
+        """
+        with torch.cuda.amp.autocast(enabled=False):
+            encoding_indices_view = list(inputs.shape)
+            del encoding_indices_view[1]
+
+            inputs = inputs.float()
+
+            # Converting to channel last format
+            flat_input = inputs.permute(self.flatten_permutation).contiguous().view(-1, self.embedding_dim)
+
+            # Calculate Euclidean distances
+            distances = (
+                (flat_input**2).sum(dim=1, keepdim=True)
+                + (self.embedding.weight.t() ** 2).sum(dim=0, keepdim=True)
+                - 2 * torch.mm(flat_input, self.embedding.weight.t())
+            )
+
+            # Mapping distances to indexes
+            encoding_indices = torch.max(-distances, dim=1)[1]
+            encodings = torch.nn.functional.one_hot(encoding_indices, self.num_embeddings).float()
+
+            # Quantize and reshape
+            encoding_indices = encoding_indices.view(encoding_indices_view)
+
+            return flat_input, encodings, encoding_indices
+
+    def embed(self, embedding_indices: torch.Tensor) -> torch.Tensor:
+        """
+        Given encoding indices of shape [B,D,H,W,1] embeds them in the quantized space
+        [B, D, H, W, self.embedding_dim] and reshapes them to [B, self.embedding_dim, D, H, W] to be fed to the
+        decoder.
+
+        Args:
+            embedding_indices: Tensor in channel last format which holds indices referencing atomic
+                elements from self.embedding
+
+        Returns:
+            torch.Tensor: Quantize space representation of encoding_indices in channel first format.
+        """
+        with torch.cuda.amp.autocast(enabled=False):
+            embedding: torch.Tensor = (
+                self.embedding(embedding_indices).permute(self.quantization_permutation).contiguous()
+            )
+            return embedding
+
+    def distributed_synchronization(self, encodings_sum: torch.Tensor, dw: torch.Tensor) -> None:
+        """
+        TorchScript does not support torch.distributed.all_reduce. This function is a bypassing trick based on the
+        example: https://pytorch.org/docs/stable/generated/torch.jit.unused.html#torch.jit.unused
+
+        Args:
+            encodings_sum: The summation of one hot representation of what encoding was used for each
+                position.
+            dw: The multiplication of the one hot representation of what encoding was used for each
+                position with the flattened input.
+
+        Returns:
+            None
+        """
+        if self.ddp_sync and torch.distributed.is_initialized():
+            torch.distributed.all_reduce(tensor=encodings_sum, op=torch.distributed.ReduceOp.SUM)
+            torch.distributed.all_reduce(tensor=dw, op=torch.distributed.ReduceOp.SUM)
+        else:
+            pass
+
+    def forward(self, inputs: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
+        flat_input, encodings, encoding_indices = self.quantize(inputs)
+        quantized = self.embed(encoding_indices)
+
+        # Use EMA to update the embedding vectors
+        if self.training:
+            with torch.no_grad():
+                encodings_sum = encodings.sum(0)
+                dw = torch.mm(encodings.t(), flat_input)
+
+                if self.ddp_sync:
+                    self.distributed_synchronization(encodings_sum, dw)
+
+                self.ema_cluster_size.data.mul_(self.decay).add_(torch.mul(encodings_sum, 1 - self.decay))
+
+                # Laplace smoothing of the cluster size
+                n = self.ema_cluster_size.sum()
+                weights = (self.ema_cluster_size + self.epsilon) / (n + self.num_embeddings * self.epsilon) * n
+                self.ema_w.data.mul_(self.decay).add_(torch.mul(dw, 1 - self.decay))
+                self.embedding.weight.data.copy_(self.ema_w / weights.unsqueeze(1))
+
+        # Encoding Loss
+        loss = self.commitment_cost * torch.nn.functional.mse_loss(quantized.detach(), inputs)
+
+        # Straight Through Estimator
+        quantized = inputs + (quantized - inputs).detach()
+
+        return quantized, loss, encoding_indices
+
+
+class VectorQuantizer(torch.nn.Module):
+    """
+    Vector Quantization wrapper that is needed as a workaround for the AMP to isolate the non fp16 compatible parts of
+    the quantization in their own class.
+
+    Args:
+        quantizer (torch.nn.Module):  Quantizer module that needs to return its quantized representation, loss and index
+            based quantized representation.
+    """
+
+    def __init__(self, quantizer: EMAQuantizer):
+        super().__init__()
+
+        self.quantizer: EMAQuantizer = quantizer
+
+        self.perplexity: torch.Tensor = torch.rand(1)
+
+    def forward(self, inputs: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]:
+        quantized, loss, encoding_indices = self.quantizer(inputs)
+        # Perplexity calculations
+        avg_probs = (
+            torch.histc(encoding_indices.float(), bins=self.quantizer.num_embeddings, max=self.quantizer.num_embeddings)
+            .float()
+            .div(encoding_indices.numel())
+        )
+
+        self.perplexity = torch.exp(-torch.sum(avg_probs * torch.log(avg_probs + 1e-10)))
+
+        return loss, quantized
+
+    def embed(self, embedding_indices: torch.Tensor) -> torch.Tensor:
+        return self.quantizer.embed(embedding_indices=embedding_indices)
+
+    def quantize(self, encodings: torch.Tensor) -> torch.Tensor:
+        output = self.quantizer(encodings)
+        encoding_indices: torch.Tensor = output[2]
+        return encoding_indices
diff --git a/monai/networks/nets/__init__.py b/monai/networks/nets/__init__.py
index de5d1adc7e..c777fe6442 100644
--- a/monai/networks/nets/__init__.py
+++ b/monai/networks/nets/__init__.py
@@ -14,9 +14,11 @@
 from .ahnet import AHnet, Ahnet, AHNet
 from .attentionunet import AttentionUnet
 from .autoencoder import AutoEncoder
+from .autoencoderkl import AutoencoderKL
 from .basic_unet import BasicUNet, BasicUnet, Basicunet, basicunet
 from .basic_unetplusplus import BasicUNetPlusPlus, BasicUnetPlusPlus, BasicunetPlusPlus, basicunetplusplus
 from .classifier import Classifier, Critic, Discriminator
+from .controlnet import ControlNet
 from .daf3d import DAF3D
 from .densenet import (
     DenseNet,
@@ -34,6 +36,7 @@
     densenet201,
     densenet264,
 )
+from .diffusion_model_unet import DiffusionModelUNet
 from .dints import DiNTS, TopologyConstruction, TopologyInstance, TopologySearch
 from .dynunet import DynUNet, DynUnet, Dynunet
 from .efficientnet import (
@@ -52,6 +55,7 @@
 from .hovernet import Hovernet, HoVernet, HoVerNet, HoverNet
 from .milmodel import MILModel
 from .netadapter import NetAdapter
+from .patchgan_discriminator import MultiScalePatchDiscriminator, PatchDiscriminator
 from .quicknat import Quicknat
 from .regressor import Regressor
 from .regunet import GlobalNet, LocalNet, RegUNet
@@ -104,9 +108,13 @@
     seresnext50,
     seresnext101,
 )
+from .spade_autoencoderkl import SPADEAutoencoderKL
+from .spade_diffusion_model_unet import SPADEDiffusionModelUNet
+from .spade_network import SPADENet
 from .swin_unetr import PatchMerging, PatchMergingV2, SwinUNETR
 from .torchvision_fc import TorchVisionFCModel
 from .transchex import BertAttention, BertMixedLayer, BertOutput, BertPreTrainedModel, MultiModal, Pooler, Transchex
+from .transformer import DecoderOnlyTransformer
 from .unet import UNet, Unet
 from .unetr import UNETR
 from .varautoencoder import VarAutoEncoder
@@ -114,3 +122,4 @@
 from .vitautoenc import ViTAutoEnc
 from .vnet import VNet
 from .voxelmorph import VoxelMorph, VoxelMorphUNet
+from .vqvae import VQVAE
diff --git a/monai/networks/nets/autoencoderkl.py b/monai/networks/nets/autoencoderkl.py
new file mode 100644
index 0000000000..35d80e0565
--- /dev/null
+++ b/monai/networks/nets/autoencoderkl.py
@@ -0,0 +1,702 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from collections.abc import Sequence
+from typing import List
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+from monai.networks.blocks import Convolution, SpatialAttentionBlock, Upsample
+from monai.utils import ensure_tuple_rep, optional_import
+
+Rearrange, _ = optional_import("einops.layers.torch", name="Rearrange")
+
+__all__ = ["AutoencoderKL"]
+
+
+class AsymmetricPad(nn.Module):
+    """
+    Pad the input tensor asymmetrically along every spatial dimension.
+
+    Args:
+        spatial_dims: number of spatial dimensions, could be 1, 2, or 3.
+    """
+
+    def __init__(self, spatial_dims: int) -> None:
+        super().__init__()
+        self.pad = (0, 1) * spatial_dims
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x = nn.functional.pad(x, self.pad, mode="constant", value=0.0)
+        return x
+
+
+class AEKLDownsample(nn.Module):
+    """
+    Convolution-based downsampling layer.
+
+    Args:
+        spatial_dims: number of spatial dimensions (1D, 2D, 3D).
+        in_channels: number of input channels.
+    """
+
+    def __init__(self, spatial_dims: int, in_channels: int) -> None:
+        super().__init__()
+        self.pad = AsymmetricPad(spatial_dims=spatial_dims)
+
+        self.conv = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=in_channels,
+            strides=2,
+            kernel_size=3,
+            padding=0,
+            conv_only=True,
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x = self.pad(x)
+        x = self.conv(x)
+        return x
+
+
+class AEKLResBlock(nn.Module):
+    """
+    Residual block consisting of a cascade of 2 convolutions + activation + normalisation block, and a
+    residual connection between input and output.
+
+    Args:
+        spatial_dims: number of spatial dimensions, could be 1, 2, or 3.
+        in_channels: input channels to the layer.
+        norm_num_groups: number of groups involved for the group normalisation layer. Ensure that your number of
+            channels is divisible by this number.
+        norm_eps: epsilon for the normalisation.
+        out_channels: number of output channels.
+    """
+
+    def __init__(
+        self, spatial_dims: int, in_channels: int, norm_num_groups: int, norm_eps: float, out_channels: int
+    ) -> None:
+        super().__init__()
+        self.in_channels = in_channels
+        self.out_channels = in_channels if out_channels is None else out_channels
+
+        self.norm1 = nn.GroupNorm(num_groups=norm_num_groups, num_channels=in_channels, eps=norm_eps, affine=True)
+        self.conv1 = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=self.in_channels,
+            out_channels=self.out_channels,
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+        self.norm2 = nn.GroupNorm(num_groups=norm_num_groups, num_channels=out_channels, eps=norm_eps, affine=True)
+        self.conv2 = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=self.out_channels,
+            out_channels=self.out_channels,
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+
+        self.nin_shortcut: nn.Module
+        if self.in_channels != self.out_channels:
+            self.nin_shortcut = Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=self.in_channels,
+                out_channels=self.out_channels,
+                strides=1,
+                kernel_size=1,
+                padding=0,
+                conv_only=True,
+            )
+        else:
+            self.nin_shortcut = nn.Identity()
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        h = x
+        h = self.norm1(h)
+        h = F.silu(h)
+        h = self.conv1(h)
+
+        h = self.norm2(h)
+        h = F.silu(h)
+        h = self.conv2(h)
+
+        x = self.nin_shortcut(x)
+
+        return x + h
+
+
+class Encoder(nn.Module):
+    """
+    Convolutional cascade that downsamples the image into a spatial latent space.
+
+    Args:
+        spatial_dims: number of spatial dimensions, could be 1, 2, or 3.
+        in_channels: number of input channels.
+        channels: sequence of block output channels.
+        out_channels: number of channels in the bottom layer (latent space) of the autoencoder.
+        num_res_blocks: number of residual blocks (see _ResBlock) per level.
+        norm_num_groups: number of groups for the GroupNorm layers, num_channels must be divisible by this number.
+        norm_eps: epsilon for the normalization.
+        attention_levels: indicate which level from num_channels contain an attention block.
+        with_nonlocal_attn: if True use non-local attention block.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        channels: Sequence[int],
+        out_channels: int,
+        num_res_blocks: Sequence[int],
+        norm_num_groups: int,
+        norm_eps: float,
+        attention_levels: Sequence[bool],
+        with_nonlocal_attn: bool = True,
+    ) -> None:
+        super().__init__()
+        self.spatial_dims = spatial_dims
+        self.in_channels = in_channels
+        self.channels = channels
+        self.out_channels = out_channels
+        self.num_res_blocks = num_res_blocks
+        self.norm_num_groups = norm_num_groups
+        self.norm_eps = norm_eps
+        self.attention_levels = attention_levels
+
+        blocks: List[nn.Module] = []
+        # Initial convolution
+        blocks.append(
+            Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=in_channels,
+                out_channels=channels[0],
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+
+        # Residual and downsampling blocks
+        output_channel = channels[0]
+        for i in range(len(channels)):
+            input_channel = output_channel
+            output_channel = channels[i]
+            is_final_block = i == len(channels) - 1
+
+            for _ in range(self.num_res_blocks[i]):
+                blocks.append(
+                    AEKLResBlock(
+                        spatial_dims=spatial_dims,
+                        in_channels=input_channel,
+                        norm_num_groups=norm_num_groups,
+                        norm_eps=norm_eps,
+                        out_channels=output_channel,
+                    )
+                )
+                input_channel = output_channel
+                if attention_levels[i]:
+                    blocks.append(
+                        SpatialAttentionBlock(
+                            spatial_dims=spatial_dims,
+                            num_channels=input_channel,
+                            norm_num_groups=norm_num_groups,
+                            norm_eps=norm_eps,
+                        )
+                    )
+
+            if not is_final_block:
+                blocks.append(AEKLDownsample(spatial_dims=spatial_dims, in_channels=input_channel))
+        # Non-local attention block
+        if with_nonlocal_attn is True:
+            blocks.append(
+                AEKLResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=channels[-1],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=channels[-1],
+                )
+            )
+
+            blocks.append(
+                SpatialAttentionBlock(
+                    spatial_dims=spatial_dims,
+                    num_channels=channels[-1],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+            blocks.append(
+                AEKLResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=channels[-1],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=channels[-1],
+                )
+            )
+        # Normalise and convert to latent size
+        blocks.append(nn.GroupNorm(num_groups=norm_num_groups, num_channels=channels[-1], eps=norm_eps, affine=True))
+        blocks.append(
+            Convolution(
+                spatial_dims=self.spatial_dims,
+                in_channels=channels[-1],
+                out_channels=out_channels,
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+
+        self.blocks = nn.ModuleList(blocks)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        for block in self.blocks:
+            x = block(x)
+        return x
+
+
+class Decoder(nn.Module):
+    """
+    Convolutional cascade upsampling from a spatial latent space into an image space.
+
+    Args:
+        spatial_dims: number of spatial dimensions, could be 1, 2, or 3.
+        channels: sequence of block output channels.
+        in_channels: number of channels in the bottom layer (latent space) of the autoencoder.
+        out_channels: number of output channels.
+        num_res_blocks: number of residual blocks (see _ResBlock) per level.
+        norm_num_groups: number of groups for the GroupNorm layers, num_channels must be divisible by this number.
+        norm_eps: epsilon for the normalization.
+        attention_levels: indicate which level from num_channels contain an attention block.
+        with_nonlocal_attn: if True use non-local attention block.
+        use_convtranspose: if True, use ConvTranspose to upsample feature maps in decoder.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        channels: Sequence[int],
+        in_channels: int,
+        out_channels: int,
+        num_res_blocks: Sequence[int],
+        norm_num_groups: int,
+        norm_eps: float,
+        attention_levels: Sequence[bool],
+        with_nonlocal_attn: bool = True,
+        use_convtranspose: bool = False,
+    ) -> None:
+        super().__init__()
+        self.spatial_dims = spatial_dims
+        self.channels = channels
+        self.in_channels = in_channels
+        self.out_channels = out_channels
+        self.num_res_blocks = num_res_blocks
+        self.norm_num_groups = norm_num_groups
+        self.norm_eps = norm_eps
+        self.attention_levels = attention_levels
+
+        reversed_block_out_channels = list(reversed(channels))
+
+        blocks: List[nn.Module] = []
+
+        # Initial convolution
+        blocks.append(
+            Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=in_channels,
+                out_channels=reversed_block_out_channels[0],
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+
+        # Non-local attention block
+        if with_nonlocal_attn is True:
+            blocks.append(
+                AEKLResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=reversed_block_out_channels[0],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=reversed_block_out_channels[0],
+                )
+            )
+            blocks.append(
+                SpatialAttentionBlock(
+                    spatial_dims=spatial_dims,
+                    num_channels=reversed_block_out_channels[0],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+            blocks.append(
+                AEKLResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=reversed_block_out_channels[0],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=reversed_block_out_channels[0],
+                )
+            )
+
+        reversed_attention_levels = list(reversed(attention_levels))
+        reversed_num_res_blocks = list(reversed(num_res_blocks))
+        block_out_ch = reversed_block_out_channels[0]
+        for i in range(len(reversed_block_out_channels)):
+            block_in_ch = block_out_ch
+            block_out_ch = reversed_block_out_channels[i]
+            is_final_block = i == len(channels) - 1
+
+            for _ in range(reversed_num_res_blocks[i]):
+                blocks.append(
+                    AEKLResBlock(
+                        spatial_dims=spatial_dims,
+                        in_channels=block_in_ch,
+                        norm_num_groups=norm_num_groups,
+                        norm_eps=norm_eps,
+                        out_channels=block_out_ch,
+                    )
+                )
+                block_in_ch = block_out_ch
+
+                if reversed_attention_levels[i]:
+                    blocks.append(
+                        SpatialAttentionBlock(
+                            spatial_dims=spatial_dims,
+                            num_channels=block_in_ch,
+                            norm_num_groups=norm_num_groups,
+                            norm_eps=norm_eps,
+                        )
+                    )
+
+            if not is_final_block:
+                if use_convtranspose:
+                    blocks.append(
+                        Upsample(
+                            spatial_dims=spatial_dims, mode="deconv", in_channels=block_in_ch, out_channels=block_in_ch
+                        )
+                    )
+                else:
+                    post_conv = Convolution(
+                        spatial_dims=spatial_dims,
+                        in_channels=block_in_ch,
+                        out_channels=block_in_ch,
+                        strides=1,
+                        kernel_size=3,
+                        padding=1,
+                        conv_only=True,
+                    )
+                    blocks.append(
+                        Upsample(
+                            spatial_dims=spatial_dims,
+                            mode="nontrainable",
+                            in_channels=block_in_ch,
+                            out_channels=block_in_ch,
+                            interp_mode="nearest",
+                            scale_factor=2.0,
+                            post_conv=post_conv,
+                            align_corners=None,
+                        )
+                    )
+
+        blocks.append(nn.GroupNorm(num_groups=norm_num_groups, num_channels=block_in_ch, eps=norm_eps, affine=True))
+        blocks.append(
+            Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=block_in_ch,
+                out_channels=out_channels,
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+
+        self.blocks = nn.ModuleList(blocks)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        for block in self.blocks:
+            x = block(x)
+        return x
+
+
+class AutoencoderKL(nn.Module):
+    """
+    Autoencoder model with KL-regularized latent space based on
+    Rombach et al. "High-Resolution Image Synthesis with Latent Diffusion Models" https://arxiv.org/abs/2112.10752
+    and Pinaya et al. "Brain Imaging Generation with Latent Diffusion Models" https://arxiv.org/abs/2209.07162
+
+    Args:
+        spatial_dims: number of spatial dimensions, could be 1, 2, or 3.
+        in_channels: number of input channels.
+        out_channels: number of output channels.
+        num_res_blocks: number of residual blocks (see _ResBlock) per level.
+        channels: number of output channels for each block.
+        attention_levels: sequence of levels to add attention.
+        latent_channels: latent embedding dimension.
+        norm_num_groups: number of groups for the GroupNorm layers, num_channels must be divisible by this number.
+        norm_eps: epsilon for the normalization.
+        with_encoder_nonlocal_attn: if True use non-local attention block in the encoder.
+        with_decoder_nonlocal_attn: if True use non-local attention block in the decoder.
+        use_checkpoint: if True, use activation checkpoint to save memory.
+        use_convtranspose: if True, use ConvTranspose to upsample feature maps in decoder.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int = 1,
+        out_channels: int = 1,
+        num_res_blocks: Sequence[int] | int = (2, 2, 2, 2),
+        channels: Sequence[int] = (32, 64, 64, 64),
+        attention_levels: Sequence[bool] = (False, False, True, True),
+        latent_channels: int = 3,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        with_encoder_nonlocal_attn: bool = True,
+        with_decoder_nonlocal_attn: bool = True,
+        use_checkpoint: bool = False,
+        use_convtranspose: bool = False,
+    ) -> None:
+        super().__init__()
+
+        # All number of channels should be multiple of num_groups
+        if any((out_channel % norm_num_groups) != 0 for out_channel in channels):
+            raise ValueError("AutoencoderKL expects all num_channels being multiple of norm_num_groups")
+
+        if len(channels) != len(attention_levels):
+            raise ValueError("AutoencoderKL expects num_channels being same size of attention_levels")
+
+        if isinstance(num_res_blocks, int):
+            num_res_blocks = ensure_tuple_rep(num_res_blocks, len(channels))
+
+        if len(num_res_blocks) != len(channels):
+            raise ValueError(
+                "`num_res_blocks` should be a single integer or a tuple of integers with the same length as "
+                "`num_channels`."
+            )
+
+        self.encoder = Encoder(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            channels=channels,
+            out_channels=latent_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            attention_levels=attention_levels,
+            with_nonlocal_attn=with_encoder_nonlocal_attn,
+        )
+        self.decoder = Decoder(
+            spatial_dims=spatial_dims,
+            channels=channels,
+            in_channels=latent_channels,
+            out_channels=out_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            attention_levels=attention_levels,
+            with_nonlocal_attn=with_decoder_nonlocal_attn,
+            use_convtranspose=use_convtranspose,
+        )
+        self.quant_conv_mu = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=latent_channels,
+            out_channels=latent_channels,
+            strides=1,
+            kernel_size=1,
+            padding=0,
+            conv_only=True,
+        )
+        self.quant_conv_log_sigma = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=latent_channels,
+            out_channels=latent_channels,
+            strides=1,
+            kernel_size=1,
+            padding=0,
+            conv_only=True,
+        )
+        self.post_quant_conv = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=latent_channels,
+            out_channels=latent_channels,
+            strides=1,
+            kernel_size=1,
+            padding=0,
+            conv_only=True,
+        )
+        self.latent_channels = latent_channels
+        self.use_checkpoint = use_checkpoint
+
+    def encode(self, x: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor]:
+        """
+        Forwards an image through the spatial encoder, obtaining the latent mean and sigma representations.
+
+        Args:
+            x: BxCx[SPATIAL DIMS] tensor
+
+        """
+        if self.use_checkpoint:
+            h = torch.utils.checkpoint.checkpoint(self.encoder, x, use_reentrant=False)
+        else:
+            h = self.encoder(x)
+
+        z_mu = self.quant_conv_mu(h)
+        z_log_var = self.quant_conv_log_sigma(h)
+        z_log_var = torch.clamp(z_log_var, -30.0, 20.0)
+        z_sigma = torch.exp(z_log_var / 2)
+
+        return z_mu, z_sigma
+
+    def sampling(self, z_mu: torch.Tensor, z_sigma: torch.Tensor) -> torch.Tensor:
+        """
+        From the mean and sigma representations resulting of encoding an image through the latent space,
+        obtains a noise sample resulting from sampling gaussian noise, multiplying by the variance (sigma) and
+        adding the mean.
+
+        Args:
+            z_mu: Bx[Z_CHANNELS]x[LATENT SPACE SIZE] mean vector obtained by the encoder when you encode an image
+            z_sigma: Bx[Z_CHANNELS]x[LATENT SPACE SIZE] variance vector obtained by the encoder when you encode an image
+
+        Returns:
+            sample of shape Bx[Z_CHANNELS]x[LATENT SPACE SIZE]
+        """
+        eps = torch.randn_like(z_sigma)
+        z_vae = z_mu + eps * z_sigma
+        return z_vae
+
+    def reconstruct(self, x: torch.Tensor) -> torch.Tensor:
+        """
+        Encodes and decodes an input image.
+
+        Args:
+            x: BxCx[SPATIAL DIMENSIONS] tensor.
+
+        Returns:
+            reconstructed image, of the same shape as input
+        """
+        z_mu, _ = self.encode(x)
+        reconstruction = self.decode(z_mu)
+        return reconstruction
+
+    def decode(self, z: torch.Tensor) -> torch.Tensor:
+        """
+        Based on a latent space sample, forwards it through the Decoder.
+
+        Args:
+            z: Bx[Z_CHANNELS]x[LATENT SPACE SHAPE]
+
+        Returns:
+            decoded image tensor
+        """
+        z = self.post_quant_conv(z)
+        dec: torch.Tensor
+        if self.use_checkpoint:
+            dec = torch.utils.checkpoint.checkpoint(self.decoder, z, use_reentrant=False)
+        else:
+            dec = self.decoder(z)
+        return dec
+
+    def forward(self, x: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
+        z_mu, z_sigma = self.encode(x)
+        z = self.sampling(z_mu, z_sigma)
+        reconstruction = self.decode(z)
+        return reconstruction, z_mu, z_sigma
+
+    def encode_stage_2_inputs(self, x: torch.Tensor) -> torch.Tensor:
+        z_mu, z_sigma = self.encode(x)
+        z = self.sampling(z_mu, z_sigma)
+        return z
+
+    def decode_stage_2_outputs(self, z: torch.Tensor) -> torch.Tensor:
+        image = self.decode(z)
+        return image
+
+    def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
+        """
+        Load a state dict from an AutoencoderKL trained with [MONAI Generative](https://github.com/Project-MONAI/GenerativeModels).
+
+        Args:
+            old_state_dict: state dict from the old AutoencoderKL model.
+        """
+
+        new_state_dict = self.state_dict()
+        # if all keys match, just load the state dict
+        if all(k in new_state_dict for k in old_state_dict):
+            print("All keys match, loading state dict.")
+            self.load_state_dict(old_state_dict)
+            return
+
+        if verbose:
+            # print all new_state_dict keys that are not in old_state_dict
+            for k in new_state_dict:
+                if k not in old_state_dict:
+                    print(f"key {k} not found in old state dict")
+            # and vice versa
+            print("----------------------------------------------")
+            for k in old_state_dict:
+                if k not in new_state_dict:
+                    print(f"key {k} not found in new state dict")
+
+        # copy over all matching keys
+        for k in new_state_dict:
+            if k in old_state_dict:
+                new_state_dict[k] = old_state_dict[k]
+
+        # fix the attention blocks
+        attention_blocks = [k.replace(".attn.qkv.weight", "") for k in new_state_dict if "attn.qkv.weight" in k]
+        for block in attention_blocks:
+            new_state_dict[f"{block}.attn.qkv.weight"] = torch.cat(
+                [
+                    old_state_dict[f"{block}.to_q.weight"],
+                    old_state_dict[f"{block}.to_k.weight"],
+                    old_state_dict[f"{block}.to_v.weight"],
+                ],
+                dim=0,
+            )
+            new_state_dict[f"{block}.attn.qkv.bias"] = torch.cat(
+                [
+                    old_state_dict[f"{block}.to_q.bias"],
+                    old_state_dict[f"{block}.to_k.bias"],
+                    old_state_dict[f"{block}.to_v.bias"],
+                ],
+                dim=0,
+            )
+            # old version did not have a projection so set these to the identity
+            new_state_dict[f"{block}.attn.out_proj.weight"] = torch.eye(
+                new_state_dict[f"{block}.attn.out_proj.weight"].shape[0]
+            )
+            new_state_dict[f"{block}.attn.out_proj.bias"] = torch.zeros(
+                new_state_dict[f"{block}.attn.out_proj.bias"].shape
+            )
+
+        # fix the upsample conv blocks which were renamed postconv
+        for k in new_state_dict:
+            if "postconv" in k:
+                old_name = k.replace("postconv", "conv")
+                new_state_dict[k] = old_state_dict[old_name]
+        self.load_state_dict(new_state_dict)
diff --git a/monai/networks/nets/controlnet.py b/monai/networks/nets/controlnet.py
new file mode 100644
index 0000000000..ed3654733d
--- /dev/null
+++ b/monai/networks/nets/controlnet.py
@@ -0,0 +1,465 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# =========================================================================
+# Adapted from https://github.com/huggingface/diffusers
+# which has the following license:
+# https://github.com/huggingface/diffusers/blob/main/LICENSE
+#
+# Copyright 2022 UC Berkeley Team and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =========================================================================
+
+from __future__ import annotations
+
+from collections.abc import Sequence
+
+import torch
+from torch import nn
+
+from monai.networks.blocks import Convolution
+from monai.networks.nets.diffusion_model_unet import get_down_block, get_mid_block, get_timestep_embedding
+from monai.utils import ensure_tuple_rep
+
+
+class ControlNetConditioningEmbedding(nn.Module):
+    """
+    Network to encode the conditioning into a latent space.
+    """
+
+    def __init__(self, spatial_dims: int, in_channels: int, out_channels: int, channels: Sequence[int]):
+        super().__init__()
+
+        self.conv_in = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=channels[0],
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            adn_ordering="A",
+            act="SWISH",
+        )
+
+        self.blocks = nn.ModuleList([])
+
+        for i in range(len(channels) - 1):
+            channel_in = channels[i]
+            channel_out = channels[i + 1]
+            self.blocks.append(
+                Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=channel_in,
+                    out_channels=channel_in,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    adn_ordering="A",
+                    act="SWISH",
+                )
+            )
+
+            self.blocks.append(
+                Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=channel_in,
+                    out_channels=channel_out,
+                    strides=2,
+                    kernel_size=3,
+                    padding=1,
+                    adn_ordering="A",
+                    act="SWISH",
+                )
+            )
+
+        self.conv_out = zero_module(
+            Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=channels[-1],
+                out_channels=out_channels,
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+
+    def forward(self, conditioning):
+        embedding = self.conv_in(conditioning)
+
+        for block in self.blocks:
+            embedding = block(embedding)
+
+        embedding = self.conv_out(embedding)
+
+        return embedding
+
+
+def zero_module(module):
+    for p in module.parameters():
+        nn.init.zeros_(p)
+    return module
+
+
+class ControlNet(nn.Module):
+    """
+    Control network for diffusion models based on Zhang and Agrawala "Adding Conditional Control to Text-to-Image
+    Diffusion Models" (https://arxiv.org/abs/2302.05543)
+
+    Args:
+        spatial_dims: number of spatial dimensions.
+        in_channels: number of input channels.
+        num_res_blocks: number of residual blocks (see ResnetBlock) per level.
+        channels: tuple of block output channels.
+        attention_levels: list of levels to add attention.
+        norm_num_groups: number of groups for the normalization.
+        norm_eps: epsilon for the normalization.
+        resblock_updown: if True use residual blocks for up/downsampling.
+        num_head_channels: number of channels in each attention head.
+        with_conditioning: if True add spatial transformers to perform conditioning.
+        transformer_num_layers: number of layers of Transformer blocks to use.
+        cross_attention_dim: number of context dimensions to use.
+        num_class_embeds: if specified (as an int), then this model will be class-conditional with `num_class_embeds`
+            classes.
+        upcast_attention: if True, upcast attention operations to full precision.
+        conditioning_embedding_in_channels: number of input channels for the conditioning embedding.
+        conditioning_embedding_num_channels: number of channels for the blocks in the conditioning embedding.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        num_res_blocks: Sequence[int] | int = (2, 2, 2, 2),
+        channels: Sequence[int] = (32, 64, 64, 64),
+        attention_levels: Sequence[bool] = (False, False, True, True),
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        resblock_updown: bool = False,
+        num_head_channels: int | Sequence[int] = 8,
+        with_conditioning: bool = False,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        num_class_embeds: int | None = None,
+        upcast_attention: bool = False,
+        conditioning_embedding_in_channels: int = 1,
+        conditioning_embedding_num_channels: Sequence[int] = (16, 32, 96, 256),
+    ) -> None:
+        super().__init__()
+        if with_conditioning is True and cross_attention_dim is None:
+            raise ValueError(
+                "DiffusionModelUNet expects dimension of the cross-attention conditioning (cross_attention_dim) "
+                "to be specified when with_conditioning=True."
+            )
+        if cross_attention_dim is not None and with_conditioning is False:
+            raise ValueError(
+                "DiffusionModelUNet expects with_conditioning=True when specifying the cross_attention_dim."
+            )
+
+        # All number of channels should be multiple of num_groups
+        if any((out_channel % norm_num_groups) != 0 for out_channel in channels):
+            raise ValueError(
+                f"DiffusionModelUNet expects all channels to be a multiple of norm_num_groups, but got"
+                f" channels={channels} and norm_num_groups={norm_num_groups}"
+            )
+
+        if len(channels) != len(attention_levels):
+            raise ValueError(
+                f"DiffusionModelUNet expects channels to have the same length as attention_levels, but got "
+                f"channels={channels} and attention_levels={attention_levels}"
+            )
+
+        if isinstance(num_head_channels, int):
+            num_head_channels = ensure_tuple_rep(num_head_channels, len(attention_levels))
+
+        if len(num_head_channels) != len(attention_levels):
+            raise ValueError(
+                f"num_head_channels should have the same length as attention_levels, but got channels={channels} and "
+                f"attention_levels={attention_levels} . For the i levels without attention,"
+                " i.e. `attention_level[i]=False`, the num_head_channels[i] will be ignored."
+            )
+
+        if isinstance(num_res_blocks, int):
+            num_res_blocks = ensure_tuple_rep(num_res_blocks, len(channels))
+
+        if len(num_res_blocks) != len(channels):
+            raise ValueError(
+                f"`num_res_blocks` should be a single integer or a tuple of integers with the same length as "
+                f"`num_channels`, but got num_res_blocks={num_res_blocks} and channels={channels}."
+            )
+
+        self.in_channels = in_channels
+        self.block_out_channels = channels
+        self.num_res_blocks = num_res_blocks
+        self.attention_levels = attention_levels
+        self.num_head_channels = num_head_channels
+        self.with_conditioning = with_conditioning
+
+        # input
+        self.conv_in = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=channels[0],
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+
+        # time
+        time_embed_dim = channels[0] * 4
+        self.time_embed = nn.Sequential(
+            nn.Linear(channels[0], time_embed_dim), nn.SiLU(), nn.Linear(time_embed_dim, time_embed_dim)
+        )
+
+        # class embedding
+        self.num_class_embeds = num_class_embeds
+        if num_class_embeds is not None:
+            self.class_embedding = nn.Embedding(num_class_embeds, time_embed_dim)
+
+        # control net conditioning embedding
+        self.controlnet_cond_embedding = ControlNetConditioningEmbedding(
+            spatial_dims=spatial_dims,
+            in_channels=conditioning_embedding_in_channels,
+            channels=conditioning_embedding_num_channels,
+            out_channels=channels[0],
+        )
+
+        # down
+        self.down_blocks = nn.ModuleList([])
+        self.controlnet_down_blocks = nn.ModuleList([])
+        output_channel = channels[0]
+
+        controlnet_block = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=output_channel,
+            out_channels=output_channel,
+            strides=1,
+            kernel_size=1,
+            padding=0,
+            conv_only=True,
+        )
+        controlnet_block = zero_module(controlnet_block.conv)
+        self.controlnet_down_blocks.append(controlnet_block)
+
+        for i in range(len(channels)):
+            input_channel = output_channel
+            output_channel = channels[i]
+            is_final_block = i == len(channels) - 1
+
+            down_block = get_down_block(
+                spatial_dims=spatial_dims,
+                in_channels=input_channel,
+                out_channels=output_channel,
+                temb_channels=time_embed_dim,
+                num_res_blocks=num_res_blocks[i],
+                norm_num_groups=norm_num_groups,
+                norm_eps=norm_eps,
+                add_downsample=not is_final_block,
+                resblock_updown=resblock_updown,
+                with_attn=(attention_levels[i] and not with_conditioning),
+                with_cross_attn=(attention_levels[i] and with_conditioning),
+                num_head_channels=num_head_channels[i],
+                transformer_num_layers=transformer_num_layers,
+                cross_attention_dim=cross_attention_dim,
+                upcast_attention=upcast_attention,
+            )
+
+            self.down_blocks.append(down_block)
+
+            for _ in range(num_res_blocks[i]):
+                controlnet_block = Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=output_channel,
+                    out_channels=output_channel,
+                    strides=1,
+                    kernel_size=1,
+                    padding=0,
+                    conv_only=True,
+                )
+                controlnet_block = zero_module(controlnet_block)
+                self.controlnet_down_blocks.append(controlnet_block)
+            #
+            if not is_final_block:
+                controlnet_block = Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=output_channel,
+                    out_channels=output_channel,
+                    strides=1,
+                    kernel_size=1,
+                    padding=0,
+                    conv_only=True,
+                )
+                controlnet_block = zero_module(controlnet_block)
+                self.controlnet_down_blocks.append(controlnet_block)
+
+        # mid
+        mid_block_channel = channels[-1]
+
+        self.middle_block = get_mid_block(
+            spatial_dims=spatial_dims,
+            in_channels=mid_block_channel,
+            temb_channels=time_embed_dim,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            with_conditioning=with_conditioning,
+            num_head_channels=num_head_channels[-1],
+            transformer_num_layers=transformer_num_layers,
+            cross_attention_dim=cross_attention_dim,
+            upcast_attention=upcast_attention,
+        )
+
+        controlnet_block = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=output_channel,
+            out_channels=output_channel,
+            strides=1,
+            kernel_size=1,
+            padding=0,
+            conv_only=True,
+        )
+        controlnet_block = zero_module(controlnet_block)
+        self.controlnet_mid_block = controlnet_block
+
+    def forward(
+        self,
+        x: torch.Tensor,
+        timesteps: torch.Tensor,
+        controlnet_cond: torch.Tensor,
+        conditioning_scale: float = 1.0,
+        context: torch.Tensor | None = None,
+        class_labels: torch.Tensor | None = None,
+    ) -> tuple[list[torch.Tensor], torch.Tensor]:
+        """
+        Args:
+            x: input tensor (N, C, H, W, [D]).
+            timesteps: timestep tensor (N,).
+            controlnet_cond: controlnet conditioning tensor (N, C, H, W, [D])
+            conditioning_scale: conditioning scale.
+            context: context tensor (N, 1, cross_attention_dim), where cross_attention_dim is specified in the model init.
+            class_labels: context tensor (N, ).
+        """
+        # 1. time
+        t_emb = get_timestep_embedding(timesteps, self.block_out_channels[0])
+
+        # timesteps does not contain any weights and will always return f32 tensors
+        # but time_embedding might actually be running in fp16. so we need to cast here.
+        # there might be better ways to encapsulate this.
+        t_emb = t_emb.to(dtype=x.dtype)
+        emb = self.time_embed(t_emb)
+
+        # 2. class
+        if self.num_class_embeds is not None:
+            if class_labels is None:
+                raise ValueError("class_labels should be provided when num_class_embeds > 0")
+            class_emb = self.class_embedding(class_labels)
+            class_emb = class_emb.to(dtype=x.dtype)
+            emb = emb + class_emb
+
+        # 3. initial convolution
+        h = self.conv_in(x)
+
+        controlnet_cond = self.controlnet_cond_embedding(controlnet_cond)
+
+        h += controlnet_cond
+
+        # 4. down
+        if context is not None and self.with_conditioning is False:
+            raise ValueError("model should have with_conditioning = True if context is provided")
+        down_block_res_samples: list[torch.Tensor] = [h]
+        for downsample_block in self.down_blocks:
+            h, res_samples = downsample_block(hidden_states=h, temb=emb, context=context)
+            for residual in res_samples:
+                down_block_res_samples.append(residual)
+
+        # 5. mid
+        h = self.middle_block(hidden_states=h, temb=emb, context=context)
+
+        # 6. Control net blocks
+        controlnet_down_block_res_samples = []
+
+        for down_block_res_sample, controlnet_block in zip(down_block_res_samples, self.controlnet_down_blocks):
+            down_block_res_sample = controlnet_block(down_block_res_sample)
+            controlnet_down_block_res_samples.append(down_block_res_sample)
+
+        down_block_res_samples = controlnet_down_block_res_samples
+
+        mid_block_res_sample: torch.Tensor = self.controlnet_mid_block(h)
+
+        # 6. scaling
+        down_block_res_samples = [h * conditioning_scale for h in down_block_res_samples]
+        mid_block_res_sample *= conditioning_scale
+
+        return down_block_res_samples, mid_block_res_sample
+
+    def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
+        """
+        Load a state dict from a ControlNet trained with
+        [MONAI Generative](https://github.com/Project-MONAI/GenerativeModels).
+
+        Args:
+            old_state_dict: state dict from the old ControlNet model.
+        """
+
+        new_state_dict = self.state_dict()
+        # if all keys match, just load the state dict
+        if all(k in new_state_dict for k in old_state_dict):
+            print("All keys match, loading state dict.")
+            self.load_state_dict(old_state_dict)
+            return
+
+        if verbose:
+            # print all new_state_dict keys that are not in old_state_dict
+            for k in new_state_dict:
+                if k not in old_state_dict:
+                    print(f"key {k} not found in old state dict")
+            # and vice versa
+            print("----------------------------------------------")
+            for k in old_state_dict:
+                if k not in new_state_dict:
+                    print(f"key {k} not found in new state dict")
+
+        # copy over all matching keys
+        for k in new_state_dict:
+            if k in old_state_dict:
+                new_state_dict[k] = old_state_dict[k]
+
+        # fix the attention blocks
+        attention_blocks = [k.replace(".attn1.qkv.weight", "") for k in new_state_dict if "attn1.qkv.weight" in k]
+        for block in attention_blocks:
+            new_state_dict[f"{block}.attn1.qkv.weight"] = torch.cat(
+                [
+                    old_state_dict[f"{block}.attn1.to_q.weight"],
+                    old_state_dict[f"{block}.attn1.to_k.weight"],
+                    old_state_dict[f"{block}.attn1.to_v.weight"],
+                ],
+                dim=0,
+            )
+
+            # projection
+            new_state_dict[f"{block}.attn1.out_proj.weight"] = old_state_dict[f"{block}.attn1.to_out.0.weight"]
+            new_state_dict[f"{block}.attn1.out_proj.bias"] = old_state_dict[f"{block}.attn1.to_out.0.bias"]
+
+            new_state_dict[f"{block}.attn2.out_proj.weight"] = old_state_dict[f"{block}.attn2.to_out.0.weight"]
+            new_state_dict[f"{block}.attn2.out_proj.bias"] = old_state_dict[f"{block}.attn2.to_out.0.bias"]
+
+        self.load_state_dict(new_state_dict)
diff --git a/monai/networks/nets/diffusion_model_unet.py b/monai/networks/nets/diffusion_model_unet.py
new file mode 100644
index 0000000000..8a9ac859a3
--- /dev/null
+++ b/monai/networks/nets/diffusion_model_unet.py
@@ -0,0 +1,1913 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# =========================================================================
+# Adapted from https://github.com/huggingface/diffusers
+# which has the following license:
+# https://github.com/huggingface/diffusers/blob/main/LICENSE
+#
+# Copyright 2022 UC Berkeley Team and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =========================================================================
+
+from __future__ import annotations
+
+import math
+from collections.abc import Sequence
+
+import torch
+from torch import nn
+
+from monai.networks.blocks import Convolution, CrossAttentionBlock, MLPBlock, SABlock, SpatialAttentionBlock, Upsample
+from monai.networks.layers.factories import Pool
+from monai.utils import ensure_tuple_rep, optional_import
+
+Rearrange, _ = optional_import("einops.layers.torch", name="Rearrange")
+
+__all__ = ["DiffusionModelUNet"]
+
+
+def zero_module(module: nn.Module) -> nn.Module:
+    """
+    Zero out the parameters of a module and return it.
+    """
+    for p in module.parameters():
+        p.detach().zero_()
+    return module
+
+
+class DiffusionUNetTransformerBlock(nn.Module):
+    """
+    A Transformer block that allows for the input dimension to differ from the hidden dimension.
+
+    Args:
+        num_channels: number of channels in the input and output.
+        num_attention_heads: number of heads to use for multi-head attention.
+        num_head_channels: number of channels in each attention head.
+        dropout: dropout probability to use.
+        cross_attention_dim: size of the context vector for cross attention.
+        upcast_attention: if True, upcast attention operations to full precision.
+
+    """
+
+    def __init__(
+        self,
+        num_channels: int,
+        num_attention_heads: int,
+        num_head_channels: int,
+        dropout: float = 0.0,
+        cross_attention_dim: int | None = None,
+        upcast_attention: bool = False,
+    ) -> None:
+        super().__init__()
+        self.attn1 = SABlock(
+            hidden_size=num_attention_heads * num_head_channels,
+            hidden_input_size=num_channels,
+            num_heads=num_attention_heads,
+            dim_head=num_head_channels,
+            dropout_rate=dropout,
+            attention_dtype=torch.float if upcast_attention else None,
+        )
+        self.ff = MLPBlock(hidden_size=num_channels, mlp_dim=num_channels * 4, act="GEGLU", dropout_rate=dropout)
+        self.attn2 = CrossAttentionBlock(
+            hidden_size=num_attention_heads * num_head_channels,
+            num_heads=num_attention_heads,
+            hidden_input_size=num_channels,
+            context_input_size=cross_attention_dim,
+            dim_head=num_head_channels,
+            dropout_rate=dropout,
+            attention_dtype=torch.float if upcast_attention else None,
+        )
+        self.norm1 = nn.LayerNorm(num_channels)
+        self.norm2 = nn.LayerNorm(num_channels)
+        self.norm3 = nn.LayerNorm(num_channels)
+
+    def forward(self, x: torch.Tensor, context: torch.Tensor | None = None) -> torch.Tensor:
+        # 1. Self-Attention
+        x = self.attn1(self.norm1(x)) + x
+
+        # 2. Cross-Attention
+        x = self.attn2(self.norm2(x), context=context) + x
+
+        # 3. Feed-forward
+        x = self.ff(self.norm3(x)) + x
+        return x
+
+
+class SpatialTransformer(nn.Module):
+    """
+    Transformer block for image-like data. First, project the input (aka embedding) and reshape to b, t, d. Then apply
+    standard transformer action. Finally, reshape to image.
+
+    Args:
+        spatial_dims: number of spatial dimensions.
+        in_channels: number of channels in the input and output.
+        num_attention_heads: number of heads to use for multi-head attention.
+        num_head_channels: number of channels in each attention head.
+        num_layers: number of layers of Transformer blocks to use.
+        dropout: dropout probability to use.
+        norm_num_groups: number of groups for the normalization.
+        norm_eps: epsilon for the normalization.
+        cross_attention_dim: number of context dimensions to use.
+        upcast_attention: if True, upcast attention operations to full precision.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        num_attention_heads: int,
+        num_head_channels: int,
+        num_layers: int = 1,
+        dropout: float = 0.0,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        cross_attention_dim: int | None = None,
+        upcast_attention: bool = False,
+    ) -> None:
+        super().__init__()
+        self.spatial_dims = spatial_dims
+        self.in_channels = in_channels
+        inner_dim = num_attention_heads * num_head_channels
+
+        self.norm = nn.GroupNorm(num_groups=norm_num_groups, num_channels=in_channels, eps=norm_eps, affine=True)
+
+        self.proj_in = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=inner_dim,
+            strides=1,
+            kernel_size=1,
+            padding=0,
+            conv_only=True,
+        )
+
+        self.transformer_blocks = nn.ModuleList(
+            [
+                DiffusionUNetTransformerBlock(
+                    num_channels=inner_dim,
+                    num_attention_heads=num_attention_heads,
+                    num_head_channels=num_head_channels,
+                    dropout=dropout,
+                    cross_attention_dim=cross_attention_dim,
+                    upcast_attention=upcast_attention,
+                )
+                for _ in range(num_layers)
+            ]
+        )
+
+        self.proj_out = zero_module(
+            Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=inner_dim,
+                out_channels=in_channels,
+                strides=1,
+                kernel_size=1,
+                padding=0,
+                conv_only=True,
+            )
+        )
+
+    def forward(self, x: torch.Tensor, context: torch.Tensor | None = None) -> torch.Tensor:
+        # note: if no context is given, cross-attention defaults to self-attention
+        batch = channel = height = width = depth = -1
+        if self.spatial_dims == 2:
+            batch, channel, height, width = x.shape
+        if self.spatial_dims == 3:
+            batch, channel, height, width, depth = x.shape
+
+        residual = x
+        x = self.norm(x)
+        x = self.proj_in(x)
+
+        inner_dim = x.shape[1]
+
+        if self.spatial_dims == 2:
+            x = x.permute(0, 2, 3, 1).reshape(batch, height * width, inner_dim)
+        if self.spatial_dims == 3:
+            x = x.permute(0, 2, 3, 4, 1).reshape(batch, height * width * depth, inner_dim)
+
+        for block in self.transformer_blocks:
+            x = block(x, context=context)
+
+        if self.spatial_dims == 2:
+            x = x.reshape(batch, height, width, inner_dim).permute(0, 3, 1, 2).contiguous()
+        if self.spatial_dims == 3:
+            x = x.reshape(batch, height, width, depth, inner_dim).permute(0, 4, 1, 2, 3).contiguous()
+
+        x = self.proj_out(x)
+        return x + residual
+
+
+def get_timestep_embedding(timesteps: torch.Tensor, embedding_dim: int, max_period: int = 10000) -> torch.Tensor:
+    """
+    Create sinusoidal timestep embeddings following the implementation in Ho et al. "Denoising Diffusion Probabilistic
+    Models" https://arxiv.org/abs/2006.11239.
+
+    Args:
+        timesteps: a 1-D Tensor of N indices, one per batch element.
+        embedding_dim: the dimension of the output.
+        max_period: controls the minimum frequency of the embeddings.
+    """
+    if timesteps.ndim != 1:
+        raise ValueError("Timesteps should be a 1d-array")
+
+    half_dim = embedding_dim // 2
+    exponent = -math.log(max_period) * torch.arange(start=0, end=half_dim, dtype=torch.float32, device=timesteps.device)
+    freqs = torch.exp(exponent / half_dim)
+
+    args = timesteps[:, None].float() * freqs[None, :]
+    embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1)
+
+    # zero pad
+    if embedding_dim % 2 == 1:
+        embedding = torch.nn.functional.pad(embedding, (0, 1, 0, 0))
+
+    return embedding
+
+
+class DiffusionUnetDownsample(nn.Module):
+    """
+    Downsampling layer.
+
+    Args:
+        spatial_dims: number of spatial dimensions.
+        num_channels: number of input channels.
+        use_conv: if True uses Convolution instead of Pool average to perform downsampling. In case that use_conv is
+            False, the number of output channels must be the same as the number of input channels.
+        out_channels: number of output channels.
+        padding: controls the amount of implicit zero-paddings on both sides for padding number of points
+            for each dimension.
+    """
+
+    def __init__(
+        self, spatial_dims: int, num_channels: int, use_conv: bool, out_channels: int | None = None, padding: int = 1
+    ) -> None:
+        super().__init__()
+        self.num_channels = num_channels
+        self.out_channels = out_channels or num_channels
+        self.use_conv = use_conv
+        if use_conv:
+            self.op = Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=self.num_channels,
+                out_channels=self.out_channels,
+                strides=2,
+                kernel_size=3,
+                padding=padding,
+                conv_only=True,
+            )
+        else:
+            if self.num_channels != self.out_channels:
+                raise ValueError("num_channels and out_channels must be equal when use_conv=False")
+            self.op = Pool[Pool.AVG, spatial_dims](kernel_size=2, stride=2)
+
+    def forward(self, x: torch.Tensor, emb: torch.Tensor | None = None) -> torch.Tensor:
+        del emb
+        if x.shape[1] != self.num_channels:
+            raise ValueError(
+                f"Input number of channels ({x.shape[1]}) is not equal to expected number of channels "
+                f"({self.num_channels})"
+            )
+        output: torch.Tensor = self.op(x)
+        return output
+
+
+class WrappedUpsample(Upsample):
+    """
+    Wraps MONAI upsample block to allow for calling with timestep embeddings.
+    """
+
+    def forward(self, x: torch.Tensor, emb: torch.Tensor | None = None) -> torch.Tensor:
+        del emb
+        upsampled: torch.Tensor = super().forward(x)
+        return upsampled
+
+
+class DiffusionUNetResnetBlock(nn.Module):
+    """
+    Residual block with timestep conditioning.
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        temb_channels: number of timestep embedding  channels.
+        out_channels: number of output channels.
+        up: if True, performs upsampling.
+        down: if True, performs downsampling.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        temb_channels: int,
+        out_channels: int | None = None,
+        up: bool = False,
+        down: bool = False,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+    ) -> None:
+        super().__init__()
+        self.spatial_dims = spatial_dims
+        self.channels = in_channels
+        self.emb_channels = temb_channels
+        self.out_channels = out_channels or in_channels
+        self.up = up
+        self.down = down
+
+        self.norm1 = nn.GroupNorm(num_groups=norm_num_groups, num_channels=in_channels, eps=norm_eps, affine=True)
+        self.nonlinearity = nn.SiLU()
+        self.conv1 = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=self.out_channels,
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+
+        self.upsample = self.downsample = None
+        if self.up:
+            self.upsample = WrappedUpsample(
+                spatial_dims=spatial_dims,
+                mode="nontrainable",
+                in_channels=in_channels,
+                out_channels=in_channels,
+                interp_mode="nearest",
+                scale_factor=2.0,
+                align_corners=None,
+            )
+        elif down:
+            self.downsample = DiffusionUnetDownsample(spatial_dims, in_channels, use_conv=False)
+
+        self.time_emb_proj = nn.Linear(temb_channels, self.out_channels)
+
+        self.norm2 = nn.GroupNorm(num_groups=norm_num_groups, num_channels=self.out_channels, eps=norm_eps, affine=True)
+        self.conv2 = zero_module(
+            Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=self.out_channels,
+                out_channels=self.out_channels,
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+        self.skip_connection: nn.Module
+        if self.out_channels == in_channels:
+            self.skip_connection = nn.Identity()
+        else:
+            self.skip_connection = Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=in_channels,
+                out_channels=self.out_channels,
+                strides=1,
+                kernel_size=1,
+                padding=0,
+                conv_only=True,
+            )
+
+    def forward(self, x: torch.Tensor, emb: torch.Tensor) -> torch.Tensor:
+        h = x
+        h = self.norm1(h)
+        h = self.nonlinearity(h)
+
+        if self.upsample is not None:
+            x = self.upsample(x)
+            h = self.upsample(h)
+        elif self.downsample is not None:
+            x = self.downsample(x)
+            h = self.downsample(h)
+
+        h = self.conv1(h)
+
+        if self.spatial_dims == 2:
+            temb = self.time_emb_proj(self.nonlinearity(emb))[:, :, None, None]
+        else:
+            temb = self.time_emb_proj(self.nonlinearity(emb))[:, :, None, None, None]
+        h = h + temb
+
+        h = self.norm2(h)
+        h = self.nonlinearity(h)
+        h = self.conv2(h)
+        output: torch.Tensor = self.skip_connection(x) + h
+        return output
+
+
+class DownBlock(nn.Module):
+    """
+    Unet's down block containing resnet and downsamplers blocks.
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        out_channels: number of output channels.
+        temb_channels: number of timestep embedding channels.
+        num_res_blocks: number of residual blocks.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        add_downsample: if True add downsample block.
+        resblock_updown: if True use residual blocks for downsampling.
+        downsample_padding: padding used in the downsampling block.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        temb_channels: int,
+        num_res_blocks: int = 1,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        add_downsample: bool = True,
+        resblock_updown: bool = False,
+        downsample_padding: int = 1,
+    ) -> None:
+        super().__init__()
+        self.resblock_updown = resblock_updown
+
+        resnets = []
+
+        for i in range(num_res_blocks):
+            in_channels = in_channels if i == 0 else out_channels
+            resnets.append(
+                DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=in_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+
+        self.resnets = nn.ModuleList(resnets)
+
+        if add_downsample:
+            self.downsampler: nn.Module | None
+            if resblock_updown:
+                self.downsampler = DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    down=True,
+                )
+            else:
+                self.downsampler = DiffusionUnetDownsample(
+                    spatial_dims=spatial_dims,
+                    num_channels=out_channels,
+                    use_conv=True,
+                    out_channels=out_channels,
+                    padding=downsample_padding,
+                )
+        else:
+            self.downsampler = None
+
+    def forward(
+        self, hidden_states: torch.Tensor, temb: torch.Tensor, context: torch.Tensor | None = None
+    ) -> tuple[torch.Tensor, list[torch.Tensor]]:
+        del context
+        output_states = []
+
+        for resnet in self.resnets:
+            hidden_states = resnet(hidden_states, temb)
+            output_states.append(hidden_states)
+
+        if self.downsampler is not None:
+            hidden_states = self.downsampler(hidden_states, temb)
+            output_states.append(hidden_states)
+
+        return hidden_states, output_states
+
+
+class AttnDownBlock(nn.Module):
+    """
+    Unet's down block containing resnet, downsamplers and self-attention blocks.
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        out_channels: number of output channels.
+        temb_channels: number of timestep embedding  channels.
+        num_res_blocks: number of residual blocks.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        add_downsample: if True add downsample block.
+        resblock_updown: if True use residual blocks for downsampling.
+        downsample_padding: padding used in the downsampling block.
+        num_head_channels: number of channels in each attention head.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        temb_channels: int,
+        num_res_blocks: int = 1,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        add_downsample: bool = True,
+        resblock_updown: bool = False,
+        downsample_padding: int = 1,
+        num_head_channels: int = 1,
+    ) -> None:
+        super().__init__()
+        self.resblock_updown = resblock_updown
+
+        resnets = []
+        attentions = []
+
+        for i in range(num_res_blocks):
+            in_channels = in_channels if i == 0 else out_channels
+            resnets.append(
+                DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=in_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+            attentions.append(
+                SpatialAttentionBlock(
+                    spatial_dims=spatial_dims,
+                    num_channels=out_channels,
+                    num_head_channels=num_head_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+
+        self.attentions = nn.ModuleList(attentions)
+        self.resnets = nn.ModuleList(resnets)
+
+        self.downsampler: nn.Module | None
+        if add_downsample:
+            if resblock_updown:
+                self.downsampler = DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    down=True,
+                )
+            else:
+                self.downsampler = DiffusionUnetDownsample(
+                    spatial_dims=spatial_dims,
+                    num_channels=out_channels,
+                    use_conv=True,
+                    out_channels=out_channels,
+                    padding=downsample_padding,
+                )
+        else:
+            self.downsampler = None
+
+    def forward(
+        self, hidden_states: torch.Tensor, temb: torch.Tensor, context: torch.Tensor | None = None
+    ) -> tuple[torch.Tensor, list[torch.Tensor]]:
+        del context
+        output_states = []
+
+        for resnet, attn in zip(self.resnets, self.attentions):
+            hidden_states = resnet(hidden_states, temb)
+            hidden_states = attn(hidden_states).contiguous()
+            output_states.append(hidden_states)
+
+        if self.downsampler is not None:
+            hidden_states = self.downsampler(hidden_states, temb)
+            output_states.append(hidden_states)
+
+        return hidden_states, output_states
+
+
+class CrossAttnDownBlock(nn.Module):
+    """
+    Unet's down block containing resnet, downsamplers and cross-attention blocks.
+
+    Args:
+        spatial_dims: number of spatial dimensions.
+        in_channels: number of input channels.
+        out_channels: number of output channels.
+        temb_channels: number of timestep embedding channels.
+        num_res_blocks: number of residual blocks.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        add_downsample: if True add downsample block.
+        resblock_updown: if True use residual blocks for downsampling.
+        downsample_padding: padding used in the downsampling block.
+        num_head_channels: number of channels in each attention head.
+        transformer_num_layers: number of layers of Transformer blocks to use.
+        cross_attention_dim: number of context dimensions to use.
+        upcast_attention: if True, upcast attention operations to full precision.
+        dropout_cattn: if different from zero, this will be the dropout value for the cross-attention layers
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        temb_channels: int,
+        num_res_blocks: int = 1,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        add_downsample: bool = True,
+        resblock_updown: bool = False,
+        downsample_padding: int = 1,
+        num_head_channels: int = 1,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        upcast_attention: bool = False,
+        dropout_cattn: float = 0.0,
+    ) -> None:
+        super().__init__()
+        self.resblock_updown = resblock_updown
+
+        resnets = []
+        attentions = []
+
+        for i in range(num_res_blocks):
+            in_channels = in_channels if i == 0 else out_channels
+            resnets.append(
+                DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=in_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+
+            attentions.append(
+                SpatialTransformer(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    num_attention_heads=out_channels // num_head_channels,
+                    num_head_channels=num_head_channels,
+                    num_layers=transformer_num_layers,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    cross_attention_dim=cross_attention_dim,
+                    upcast_attention=upcast_attention,
+                    dropout=dropout_cattn,
+                )
+            )
+
+        self.attentions = nn.ModuleList(attentions)
+        self.resnets = nn.ModuleList(resnets)
+
+        self.downsampler: nn.Module | None
+        if add_downsample:
+            if resblock_updown:
+                self.downsampler = DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    down=True,
+                )
+            else:
+                self.downsampler = DiffusionUnetDownsample(
+                    spatial_dims=spatial_dims,
+                    num_channels=out_channels,
+                    use_conv=True,
+                    out_channels=out_channels,
+                    padding=downsample_padding,
+                )
+        else:
+            self.downsampler = None
+
+    def forward(
+        self, hidden_states: torch.Tensor, temb: torch.Tensor, context: torch.Tensor | None = None
+    ) -> tuple[torch.Tensor, list[torch.Tensor]]:
+        output_states = []
+
+        for resnet, attn in zip(self.resnets, self.attentions):
+            hidden_states = resnet(hidden_states, temb)
+            hidden_states = attn(hidden_states, context=context).contiguous()
+            output_states.append(hidden_states)
+
+        if self.downsampler is not None:
+            hidden_states = self.downsampler(hidden_states, temb)
+            output_states.append(hidden_states)
+
+        return hidden_states, output_states
+
+
+class AttnMidBlock(nn.Module):
+    """
+    Unet's mid block containing resnet and self-attention blocks.
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        temb_channels: number of timestep embedding channels.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        num_head_channels: number of channels in each attention head.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        temb_channels: int,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        num_head_channels: int = 1,
+    ) -> None:
+        super().__init__()
+
+        self.resnet_1 = DiffusionUNetResnetBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=in_channels,
+            temb_channels=temb_channels,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+        )
+        self.attention = SpatialAttentionBlock(
+            spatial_dims=spatial_dims,
+            num_channels=in_channels,
+            num_head_channels=num_head_channels,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+        )
+
+        self.resnet_2 = DiffusionUNetResnetBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=in_channels,
+            temb_channels=temb_channels,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+        )
+
+    def forward(
+        self, hidden_states: torch.Tensor, temb: torch.Tensor, context: torch.Tensor | None = None
+    ) -> torch.Tensor:
+        del context
+        hidden_states = self.resnet_1(hidden_states, temb)
+        hidden_states = self.attention(hidden_states).contiguous()
+        hidden_states = self.resnet_2(hidden_states, temb)
+
+        return hidden_states
+
+
+class CrossAttnMidBlock(nn.Module):
+    """
+    Unet's mid block containing resnet and cross-attention blocks.
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        temb_channels: number of timestep embedding channels
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        num_head_channels: number of channels in each attention head.
+        transformer_num_layers: number of layers of Transformer blocks to use.
+        cross_attention_dim: number of context dimensions to use.
+        upcast_attention: if True, upcast attention operations to full precision.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        temb_channels: int,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        num_head_channels: int = 1,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        upcast_attention: bool = False,
+        dropout_cattn: float = 0.0,
+    ) -> None:
+        super().__init__()
+
+        self.resnet_1 = DiffusionUNetResnetBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=in_channels,
+            temb_channels=temb_channels,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+        )
+        self.attention = SpatialTransformer(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            num_attention_heads=in_channels // num_head_channels,
+            num_head_channels=num_head_channels,
+            num_layers=transformer_num_layers,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            cross_attention_dim=cross_attention_dim,
+            upcast_attention=upcast_attention,
+            dropout=dropout_cattn,
+        )
+        self.resnet_2 = DiffusionUNetResnetBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=in_channels,
+            temb_channels=temb_channels,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+        )
+
+    def forward(
+        self, hidden_states: torch.Tensor, temb: torch.Tensor, context: torch.Tensor | None = None
+    ) -> torch.Tensor:
+        hidden_states = self.resnet_1(hidden_states, temb)
+        hidden_states = self.attention(hidden_states, context=context)
+        hidden_states = self.resnet_2(hidden_states, temb)
+
+        return hidden_states
+
+
+class UpBlock(nn.Module):
+    """
+    Unet's up block containing resnet and upsamplers blocks.
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        prev_output_channel: number of channels from residual connection.
+        out_channels: number of output channels.
+        temb_channels: number of timestep embedding channels.
+        num_res_blocks: number of residual blocks.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        add_upsample: if True add downsample block.
+        resblock_updown: if True use residual blocks for upsampling.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        prev_output_channel: int,
+        out_channels: int,
+        temb_channels: int,
+        num_res_blocks: int = 1,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        add_upsample: bool = True,
+        resblock_updown: bool = False,
+    ) -> None:
+        super().__init__()
+        self.resblock_updown = resblock_updown
+        resnets = []
+
+        for i in range(num_res_blocks):
+            res_skip_channels = in_channels if (i == num_res_blocks - 1) else out_channels
+            resnet_in_channels = prev_output_channel if i == 0 else out_channels
+
+            resnets.append(
+                DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=resnet_in_channels + res_skip_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+
+        self.resnets = nn.ModuleList(resnets)
+
+        self.upsampler: nn.Module | None
+        if add_upsample:
+            if resblock_updown:
+                self.upsampler = DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    up=True,
+                )
+            else:
+                post_conv = Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+                self.upsampler = WrappedUpsample(
+                    spatial_dims=spatial_dims,
+                    mode="nontrainable",
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    interp_mode="nearest",
+                    scale_factor=2.0,
+                    post_conv=post_conv,
+                    align_corners=None,
+                )
+
+        else:
+            self.upsampler = None
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        res_hidden_states_list: list[torch.Tensor],
+        temb: torch.Tensor,
+        context: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        del context
+        for resnet in self.resnets:
+            # pop res hidden states
+            res_hidden_states = res_hidden_states_list[-1]
+            res_hidden_states_list = res_hidden_states_list[:-1]
+            hidden_states = torch.cat([hidden_states, res_hidden_states], dim=1)
+
+            hidden_states = resnet(hidden_states, temb)
+
+        if self.upsampler is not None:
+            hidden_states = self.upsampler(hidden_states, temb)
+
+        return hidden_states
+
+
+class AttnUpBlock(nn.Module):
+    """
+    Unet's up block containing resnet, upsamplers, and self-attention blocks.
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        prev_output_channel: number of channels from residual connection.
+        out_channels: number of output channels.
+        temb_channels: number of timestep embedding channels.
+        num_res_blocks: number of residual blocks.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        add_upsample: if True add downsample block.
+        resblock_updown: if True use residual blocks for upsampling.
+        num_head_channels: number of channels in each attention head.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        prev_output_channel: int,
+        out_channels: int,
+        temb_channels: int,
+        num_res_blocks: int = 1,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        add_upsample: bool = True,
+        resblock_updown: bool = False,
+        num_head_channels: int = 1,
+    ) -> None:
+        super().__init__()
+        self.resblock_updown = resblock_updown
+
+        resnets = []
+        attentions = []
+
+        for i in range(num_res_blocks):
+            res_skip_channels = in_channels if (i == num_res_blocks - 1) else out_channels
+            resnet_in_channels = prev_output_channel if i == 0 else out_channels
+
+            resnets.append(
+                DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=resnet_in_channels + res_skip_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+            attentions.append(
+                SpatialAttentionBlock(
+                    spatial_dims=spatial_dims,
+                    num_channels=out_channels,
+                    num_head_channels=num_head_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+
+        self.resnets = nn.ModuleList(resnets)
+        self.attentions = nn.ModuleList(attentions)
+
+        self.upsampler: nn.Module | None
+        if add_upsample:
+            if resblock_updown:
+                self.upsampler = DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    up=True,
+                )
+            else:
+
+                post_conv = Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+                self.upsampler = WrappedUpsample(
+                    spatial_dims=spatial_dims,
+                    mode="nontrainable",
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    interp_mode="nearest",
+                    scale_factor=2.0,
+                    post_conv=post_conv,
+                    align_corners=None,
+                )
+        else:
+            self.upsampler = None
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        res_hidden_states_list: list[torch.Tensor],
+        temb: torch.Tensor,
+        context: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        del context
+        for resnet, attn in zip(self.resnets, self.attentions):
+            # pop res hidden states
+            res_hidden_states = res_hidden_states_list[-1]
+            res_hidden_states_list = res_hidden_states_list[:-1]
+            hidden_states = torch.cat([hidden_states, res_hidden_states], dim=1)
+
+            hidden_states = resnet(hidden_states, temb)
+            hidden_states = attn(hidden_states).contiguous()
+
+        if self.upsampler is not None:
+            hidden_states = self.upsampler(hidden_states, temb)
+
+        return hidden_states
+
+
+class CrossAttnUpBlock(nn.Module):
+    """
+    Unet's up block containing resnet, upsamplers, and self-attention blocks.
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        prev_output_channel: number of channels from residual connection.
+        out_channels: number of output channels.
+        temb_channels: number of timestep embedding channels.
+        num_res_blocks: number of residual blocks.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        add_upsample: if True add downsample block.
+        resblock_updown: if True use residual blocks for upsampling.
+        num_head_channels: number of channels in each attention head.
+        transformer_num_layers: number of layers of Transformer blocks to use.
+        cross_attention_dim: number of context dimensions to use.
+        upcast_attention: if True, upcast attention operations to full precision.
+        dropout_cattn: if different from zero, this will be the dropout value for the cross-attention layers
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        prev_output_channel: int,
+        out_channels: int,
+        temb_channels: int,
+        num_res_blocks: int = 1,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        add_upsample: bool = True,
+        resblock_updown: bool = False,
+        num_head_channels: int = 1,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        upcast_attention: bool = False,
+        dropout_cattn: float = 0.0,
+    ) -> None:
+        super().__init__()
+        self.resblock_updown = resblock_updown
+
+        resnets = []
+        attentions = []
+
+        for i in range(num_res_blocks):
+            res_skip_channels = in_channels if (i == num_res_blocks - 1) else out_channels
+            resnet_in_channels = prev_output_channel if i == 0 else out_channels
+
+            resnets.append(
+                DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=resnet_in_channels + res_skip_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+            attentions.append(
+                SpatialTransformer(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    num_attention_heads=out_channels // num_head_channels,
+                    num_head_channels=num_head_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    num_layers=transformer_num_layers,
+                    cross_attention_dim=cross_attention_dim,
+                    upcast_attention=upcast_attention,
+                    dropout=dropout_cattn,
+                )
+            )
+
+        self.attentions = nn.ModuleList(attentions)
+        self.resnets = nn.ModuleList(resnets)
+
+        self.upsampler: nn.Module | None
+        if add_upsample:
+            if resblock_updown:
+                self.upsampler = DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    up=True,
+                )
+            else:
+
+                post_conv = Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+                self.upsampler = WrappedUpsample(
+                    spatial_dims=spatial_dims,
+                    mode="nontrainable",
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    interp_mode="nearest",
+                    scale_factor=2.0,
+                    post_conv=post_conv,
+                    align_corners=None,
+                )
+        else:
+            self.upsampler = None
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        res_hidden_states_list: list[torch.Tensor],
+        temb: torch.Tensor,
+        context: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        for resnet, attn in zip(self.resnets, self.attentions):
+            # pop res hidden states
+            res_hidden_states = res_hidden_states_list[-1]
+            res_hidden_states_list = res_hidden_states_list[:-1]
+            hidden_states = torch.cat([hidden_states, res_hidden_states], dim=1)
+
+            hidden_states = resnet(hidden_states, temb)
+            hidden_states = attn(hidden_states, context=context)
+
+        if self.upsampler is not None:
+            hidden_states = self.upsampler(hidden_states, temb)
+
+        return hidden_states
+
+
+def get_down_block(
+    spatial_dims: int,
+    in_channels: int,
+    out_channels: int,
+    temb_channels: int,
+    num_res_blocks: int,
+    norm_num_groups: int,
+    norm_eps: float,
+    add_downsample: bool,
+    resblock_updown: bool,
+    with_attn: bool,
+    with_cross_attn: bool,
+    num_head_channels: int,
+    transformer_num_layers: int,
+    cross_attention_dim: int | None,
+    upcast_attention: bool = False,
+    dropout_cattn: float = 0.0,
+) -> nn.Module:
+    if with_attn:
+        return AttnDownBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            temb_channels=temb_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            add_downsample=add_downsample,
+            resblock_updown=resblock_updown,
+            num_head_channels=num_head_channels,
+        )
+    elif with_cross_attn:
+        return CrossAttnDownBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            temb_channels=temb_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            add_downsample=add_downsample,
+            resblock_updown=resblock_updown,
+            num_head_channels=num_head_channels,
+            transformer_num_layers=transformer_num_layers,
+            cross_attention_dim=cross_attention_dim,
+            upcast_attention=upcast_attention,
+            dropout_cattn=dropout_cattn,
+        )
+    else:
+        return DownBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            temb_channels=temb_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            add_downsample=add_downsample,
+            resblock_updown=resblock_updown,
+        )
+
+
+def get_mid_block(
+    spatial_dims: int,
+    in_channels: int,
+    temb_channels: int,
+    norm_num_groups: int,
+    norm_eps: float,
+    with_conditioning: bool,
+    num_head_channels: int,
+    transformer_num_layers: int,
+    cross_attention_dim: int | None,
+    upcast_attention: bool = False,
+    dropout_cattn: float = 0.0,
+) -> nn.Module:
+    if with_conditioning:
+        return CrossAttnMidBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            temb_channels=temb_channels,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            num_head_channels=num_head_channels,
+            transformer_num_layers=transformer_num_layers,
+            cross_attention_dim=cross_attention_dim,
+            upcast_attention=upcast_attention,
+            dropout_cattn=dropout_cattn,
+        )
+    else:
+        return AttnMidBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            temb_channels=temb_channels,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            num_head_channels=num_head_channels,
+        )
+
+
+def get_up_block(
+    spatial_dims: int,
+    in_channels: int,
+    prev_output_channel: int,
+    out_channels: int,
+    temb_channels: int,
+    num_res_blocks: int,
+    norm_num_groups: int,
+    norm_eps: float,
+    add_upsample: bool,
+    resblock_updown: bool,
+    with_attn: bool,
+    with_cross_attn: bool,
+    num_head_channels: int,
+    transformer_num_layers: int,
+    cross_attention_dim: int | None,
+    upcast_attention: bool = False,
+    dropout_cattn: float = 0.0,
+) -> nn.Module:
+    if with_attn:
+        return AttnUpBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            prev_output_channel=prev_output_channel,
+            out_channels=out_channels,
+            temb_channels=temb_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            add_upsample=add_upsample,
+            resblock_updown=resblock_updown,
+            num_head_channels=num_head_channels,
+        )
+    elif with_cross_attn:
+        return CrossAttnUpBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            prev_output_channel=prev_output_channel,
+            out_channels=out_channels,
+            temb_channels=temb_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            add_upsample=add_upsample,
+            resblock_updown=resblock_updown,
+            num_head_channels=num_head_channels,
+            transformer_num_layers=transformer_num_layers,
+            cross_attention_dim=cross_attention_dim,
+            upcast_attention=upcast_attention,
+            dropout_cattn=dropout_cattn,
+        )
+    else:
+        return UpBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            prev_output_channel=prev_output_channel,
+            out_channels=out_channels,
+            temb_channels=temb_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            add_upsample=add_upsample,
+            resblock_updown=resblock_updown,
+        )
+
+
+class DiffusionModelUNet(nn.Module):
+    """
+    Unet network with timestep embedding and attention mechanisms for conditioning based on
+    Rombach et al. "High-Resolution Image Synthesis with Latent Diffusion Models" https://arxiv.org/abs/2112.10752
+    and Pinaya et al. "Brain Imaging Generation with Latent Diffusion Models" https://arxiv.org/abs/2209.07162
+
+    Args:
+        spatial_dims: number of spatial dimensions.
+        in_channels: number of input channels.
+        out_channels: number of output channels.
+        num_res_blocks: number of residual blocks (see _ResnetBlock) per level.
+        channels: tuple of block output channels.
+        attention_levels: list of levels to add attention.
+        norm_num_groups: number of groups for the normalization.
+        norm_eps: epsilon for the normalization.
+        resblock_updown: if True use residual blocks for up/downsampling.
+        num_head_channels: number of channels in each attention head.
+        with_conditioning: if True add spatial transformers to perform conditioning.
+        transformer_num_layers: number of layers of Transformer blocks to use.
+        cross_attention_dim: number of context dimensions to use.
+        num_class_embeds: if specified (as an int), then this model will be class-conditional with `num_class_embeds`
+        classes.
+        upcast_attention: if True, upcast attention operations to full precision.
+        dropout_cattn: if different from zero, this will be the dropout value for the cross-attention layers
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        num_res_blocks: Sequence[int] | int = (2, 2, 2, 2),
+        channels: Sequence[int] = (32, 64, 64, 64),
+        attention_levels: Sequence[bool] = (False, False, True, True),
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        resblock_updown: bool = False,
+        num_head_channels: int | Sequence[int] = 8,
+        with_conditioning: bool = False,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        num_class_embeds: int | None = None,
+        upcast_attention: bool = False,
+        dropout_cattn: float = 0.0,
+    ) -> None:
+        super().__init__()
+        if with_conditioning is True and cross_attention_dim is None:
+            raise ValueError(
+                "DiffusionModelUNet expects dimension of the cross-attention conditioning (cross_attention_dim) "
+                "when using with_conditioning."
+            )
+        if cross_attention_dim is not None and with_conditioning is False:
+            raise ValueError(
+                "DiffusionModelUNet expects with_conditioning=True when specifying the cross_attention_dim."
+            )
+        if dropout_cattn > 1.0 or dropout_cattn < 0.0:
+            raise ValueError("Dropout cannot be negative or >1.0!")
+
+        # All number of channels should be multiple of num_groups
+        if any((out_channel % norm_num_groups) != 0 for out_channel in channels):
+            raise ValueError("DiffusionModelUNet expects all num_channels being multiple of norm_num_groups")
+
+        if len(channels) != len(attention_levels):
+            raise ValueError("DiffusionModelUNet expects num_channels being same size of attention_levels")
+
+        if isinstance(num_head_channels, int):
+            num_head_channels = ensure_tuple_rep(num_head_channels, len(attention_levels))
+
+        if len(num_head_channels) != len(attention_levels):
+            raise ValueError(
+                "num_head_channels should have the same length as attention_levels. For the i levels without attention,"
+                " i.e. `attention_level[i]=False`, the num_head_channels[i] will be ignored."
+            )
+
+        if isinstance(num_res_blocks, int):
+            num_res_blocks = ensure_tuple_rep(num_res_blocks, len(channels))
+
+        if len(num_res_blocks) != len(channels):
+            raise ValueError(
+                "`num_res_blocks` should be a single integer or a tuple of integers with the same length as "
+                "`num_channels`."
+            )
+
+        self.in_channels = in_channels
+        self.block_out_channels = channels
+        self.out_channels = out_channels
+        self.num_res_blocks = num_res_blocks
+        self.attention_levels = attention_levels
+        self.num_head_channels = num_head_channels
+        self.with_conditioning = with_conditioning
+
+        # input
+        self.conv_in = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=channels[0],
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+
+        # time
+        time_embed_dim = channels[0] * 4
+        self.time_embed = nn.Sequential(
+            nn.Linear(channels[0], time_embed_dim), nn.SiLU(), nn.Linear(time_embed_dim, time_embed_dim)
+        )
+
+        # class embedding
+        self.num_class_embeds = num_class_embeds
+        if num_class_embeds is not None:
+            self.class_embedding = nn.Embedding(num_class_embeds, time_embed_dim)
+
+        # down
+        self.down_blocks = nn.ModuleList([])
+        output_channel = channels[0]
+        for i in range(len(channels)):
+            input_channel = output_channel
+            output_channel = channels[i]
+            is_final_block = i == len(channels) - 1
+
+            down_block = get_down_block(
+                spatial_dims=spatial_dims,
+                in_channels=input_channel,
+                out_channels=output_channel,
+                temb_channels=time_embed_dim,
+                num_res_blocks=num_res_blocks[i],
+                norm_num_groups=norm_num_groups,
+                norm_eps=norm_eps,
+                add_downsample=not is_final_block,
+                resblock_updown=resblock_updown,
+                with_attn=(attention_levels[i] and not with_conditioning),
+                with_cross_attn=(attention_levels[i] and with_conditioning),
+                num_head_channels=num_head_channels[i],
+                transformer_num_layers=transformer_num_layers,
+                cross_attention_dim=cross_attention_dim,
+                upcast_attention=upcast_attention,
+                dropout_cattn=dropout_cattn,
+            )
+
+            self.down_blocks.append(down_block)
+
+        # mid
+        self.middle_block = get_mid_block(
+            spatial_dims=spatial_dims,
+            in_channels=channels[-1],
+            temb_channels=time_embed_dim,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            with_conditioning=with_conditioning,
+            num_head_channels=num_head_channels[-1],
+            transformer_num_layers=transformer_num_layers,
+            cross_attention_dim=cross_attention_dim,
+            upcast_attention=upcast_attention,
+            dropout_cattn=dropout_cattn,
+        )
+
+        # up
+        self.up_blocks = nn.ModuleList([])
+        reversed_block_out_channels = list(reversed(channels))
+        reversed_num_res_blocks = list(reversed(num_res_blocks))
+        reversed_attention_levels = list(reversed(attention_levels))
+        reversed_num_head_channels = list(reversed(num_head_channels))
+        output_channel = reversed_block_out_channels[0]
+        for i in range(len(reversed_block_out_channels)):
+            prev_output_channel = output_channel
+            output_channel = reversed_block_out_channels[i]
+            input_channel = reversed_block_out_channels[min(i + 1, len(channels) - 1)]
+
+            is_final_block = i == len(channels) - 1
+
+            up_block = get_up_block(
+                spatial_dims=spatial_dims,
+                in_channels=input_channel,
+                prev_output_channel=prev_output_channel,
+                out_channels=output_channel,
+                temb_channels=time_embed_dim,
+                num_res_blocks=reversed_num_res_blocks[i] + 1,
+                norm_num_groups=norm_num_groups,
+                norm_eps=norm_eps,
+                add_upsample=not is_final_block,
+                resblock_updown=resblock_updown,
+                with_attn=(reversed_attention_levels[i] and not with_conditioning),
+                with_cross_attn=(reversed_attention_levels[i] and with_conditioning),
+                num_head_channels=reversed_num_head_channels[i],
+                transformer_num_layers=transformer_num_layers,
+                cross_attention_dim=cross_attention_dim,
+                upcast_attention=upcast_attention,
+                dropout_cattn=dropout_cattn,
+            )
+
+            self.up_blocks.append(up_block)
+
+        # out
+        self.out = nn.Sequential(
+            nn.GroupNorm(num_groups=norm_num_groups, num_channels=channels[0], eps=norm_eps, affine=True),
+            nn.SiLU(),
+            zero_module(
+                Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=channels[0],
+                    out_channels=out_channels,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+            ),
+        )
+
+    def forward(
+        self,
+        x: torch.Tensor,
+        timesteps: torch.Tensor,
+        context: torch.Tensor | None = None,
+        class_labels: torch.Tensor | None = None,
+        down_block_additional_residuals: tuple[torch.Tensor] | None = None,
+        mid_block_additional_residual: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        """
+        Args:
+            x: input tensor (N, C, SpatialDims).
+            timesteps: timestep tensor (N,).
+            context: context tensor (N, 1, ContextDim).
+            class_labels: context tensor (N, ).
+            down_block_additional_residuals: additional residual tensors for down blocks (N, C, FeatureMapsDims).
+            mid_block_additional_residual: additional residual tensor for mid block (N, C, FeatureMapsDims).
+        """
+        # 1. time
+        t_emb = get_timestep_embedding(timesteps, self.block_out_channels[0])
+
+        # timesteps does not contain any weights and will always return f32 tensors
+        # but time_embedding might actually be running in fp16. so we need to cast here.
+        # there might be better ways to encapsulate this.
+        t_emb = t_emb.to(dtype=x.dtype)
+        emb = self.time_embed(t_emb)
+
+        # 2. class
+        if self.num_class_embeds is not None:
+            if class_labels is None:
+                raise ValueError("class_labels should be provided when num_class_embeds > 0")
+            class_emb = self.class_embedding(class_labels)
+            class_emb = class_emb.to(dtype=x.dtype)
+            emb = emb + class_emb
+
+        # 3. initial convolution
+        h = self.conv_in(x)
+
+        # 4. down
+        if context is not None and self.with_conditioning is False:
+            raise ValueError("model should have with_conditioning = True if context is provided")
+        down_block_res_samples: list[torch.Tensor] = [h]
+        for downsample_block in self.down_blocks:
+            h, res_samples = downsample_block(hidden_states=h, temb=emb, context=context)
+            for residual in res_samples:
+                down_block_res_samples.append(residual)
+
+        # Additional residual conections for Controlnets
+        if down_block_additional_residuals is not None:
+            new_down_block_res_samples: list[torch.Tensor] = []
+            for down_block_res_sample, down_block_additional_residual in zip(
+                down_block_res_samples, down_block_additional_residuals
+            ):
+                down_block_res_sample = down_block_res_sample + down_block_additional_residual
+                new_down_block_res_samples += [down_block_res_sample]
+
+            down_block_res_samples = new_down_block_res_samples
+
+        # 5. mid
+        h = self.middle_block(hidden_states=h, temb=emb, context=context)
+
+        # Additional residual conections for Controlnets
+        if mid_block_additional_residual is not None:
+            h = h + mid_block_additional_residual
+
+        # 6. up
+        for upsample_block in self.up_blocks:
+            res_samples = down_block_res_samples[-len(upsample_block.resnets) :]
+            down_block_res_samples = down_block_res_samples[: -len(upsample_block.resnets)]
+            h = upsample_block(hidden_states=h, res_hidden_states_list=res_samples, temb=emb, context=context)
+
+        # 7. output block
+        output: torch.Tensor = self.out(h)
+
+        return output
+
+    def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
+        """
+        Load a state dict from a DiffusionModelUNet trained with
+        [MONAI Generative](https://github.com/Project-MONAI/GenerativeModels).
+
+        Args:
+            old_state_dict: state dict from the old DecoderOnlyTransformer  model.
+        """
+
+        new_state_dict = self.state_dict()
+        # if all keys match, just load the state dict
+        if all(k in new_state_dict for k in old_state_dict):
+            print("All keys match, loading state dict.")
+            self.load_state_dict(old_state_dict)
+            return
+
+        if verbose:
+            # print all new_state_dict keys that are not in old_state_dict
+            for k in new_state_dict:
+                if k not in old_state_dict:
+                    print(f"key {k} not found in old state dict")
+            # and vice versa
+            print("----------------------------------------------")
+            for k in old_state_dict:
+                if k not in new_state_dict:
+                    print(f"key {k} not found in new state dict")
+
+        # copy over all matching keys
+        for k in new_state_dict:
+            if k in old_state_dict:
+                new_state_dict[k] = old_state_dict[k]
+
+        # fix the attention blocks
+        attention_blocks = [k.replace(".attn1.qkv.weight", "") for k in new_state_dict if "attn1.qkv.weight" in k]
+        for block in attention_blocks:
+            new_state_dict[f"{block}.attn1.qkv.weight"] = torch.cat(
+                [
+                    old_state_dict[f"{block}.attn1.to_q.weight"],
+                    old_state_dict[f"{block}.attn1.to_k.weight"],
+                    old_state_dict[f"{block}.attn1.to_v.weight"],
+                ],
+                dim=0,
+            )
+
+            # projection
+            new_state_dict[f"{block}.attn1.out_proj.weight"] = old_state_dict[f"{block}.attn1.to_out.0.weight"]
+            new_state_dict[f"{block}.attn1.out_proj.bias"] = old_state_dict[f"{block}.attn1.to_out.0.bias"]
+
+            new_state_dict[f"{block}.attn2.out_proj.weight"] = old_state_dict[f"{block}.attn2.to_out.0.weight"]
+            new_state_dict[f"{block}.attn2.out_proj.bias"] = old_state_dict[f"{block}.attn2.to_out.0.bias"]
+        # fix the upsample conv blocks which were renamed postconv
+        for k in new_state_dict:
+            if "postconv" in k:
+                old_name = k.replace("postconv", "conv")
+                new_state_dict[k] = old_state_dict[old_name]
+        self.load_state_dict(new_state_dict)
+
+
+class DiffusionModelEncoder(nn.Module):
+    """
+    Classification Network based on the Encoder of the Diffusion Model, followed by fully connected layers. This network is based on
+    Wolleb et al. "Diffusion Models for Medical Anomaly Detection" (https://arxiv.org/abs/2203.04306).
+
+    Args:
+        spatial_dims: number of spatial dimensions.
+        in_channels: number of input channels.
+        out_channels: number of output channels.
+        num_res_blocks: number of residual blocks (see _ResnetBlock) per level.
+        channels: tuple of block output channels.
+        attention_levels: list of levels to add attention.
+        norm_num_groups: number of groups for the normalization.
+        norm_eps: epsilon for the normalization.
+        resblock_updown: if True use residual blocks for downsampling.
+        num_head_channels: number of channels in each attention head.
+        with_conditioning: if True add spatial transformers to perform conditioning.
+        transformer_num_layers: number of layers of Transformer blocks to use.
+        cross_attention_dim: number of context dimensions to use.
+        num_class_embeds: if specified (as an int), then this model will be class-conditional with `num_class_embeds` classes.
+        upcast_attention: if True, upcast attention operations to full precision.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        num_res_blocks: Sequence[int] | int = (2, 2, 2, 2),
+        channels: Sequence[int] = (32, 64, 64, 64),
+        attention_levels: Sequence[bool] = (False, False, True, True),
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        resblock_updown: bool = False,
+        num_head_channels: int | Sequence[int] = 8,
+        with_conditioning: bool = False,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        num_class_embeds: int | None = None,
+        upcast_attention: bool = False,
+    ) -> None:
+        super().__init__()
+        if with_conditioning is True and cross_attention_dim is None:
+            raise ValueError(
+                "DiffusionModelEncoder expects dimension of the cross-attention conditioning (cross_attention_dim) "
+                "when using with_conditioning."
+            )
+        if cross_attention_dim is not None and with_conditioning is False:
+            raise ValueError(
+                "DiffusionModelEncoder expects with_conditioning=True when specifying the cross_attention_dim."
+            )
+
+        # All number of channels should be multiple of num_groups
+        if any((out_channel % norm_num_groups) != 0 for out_channel in channels):
+            raise ValueError("DiffusionModelEncoder expects all num_channels being multiple of norm_num_groups")
+        if len(channels) != len(attention_levels):
+            raise ValueError("DiffusionModelEncoder expects num_channels being same size of attention_levels")
+
+        if isinstance(num_head_channels, int):
+            num_head_channels = ensure_tuple_rep(num_head_channels, len(attention_levels))
+
+        if isinstance(num_res_blocks, int):
+            num_res_blocks = ensure_tuple_rep(num_res_blocks, len(channels))
+
+        if len(num_head_channels) != len(attention_levels):
+            raise ValueError(
+                "num_head_channels should have the same length as attention_levels. For the i levels without attention,"
+                " i.e. `attention_level[i]=False`, the num_head_channels[i] will be ignored."
+            )
+
+        self.in_channels = in_channels
+        self.block_out_channels = channels
+        self.out_channels = out_channels
+        self.num_res_blocks = num_res_blocks
+        self.attention_levels = attention_levels
+        self.num_head_channels = num_head_channels
+        self.with_conditioning = with_conditioning
+
+        # input
+        self.conv_in = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=channels[0],
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+
+        # time
+        time_embed_dim = channels[0] * 4
+        self.time_embed = nn.Sequential(
+            nn.Linear(channels[0], time_embed_dim), nn.SiLU(), nn.Linear(time_embed_dim, time_embed_dim)
+        )
+
+        # class embedding
+        self.num_class_embeds = num_class_embeds
+        if num_class_embeds is not None:
+            self.class_embedding = nn.Embedding(num_class_embeds, time_embed_dim)
+
+        # down
+        self.down_blocks = nn.ModuleList([])
+        output_channel = channels[0]
+        for i in range(len(channels)):
+            input_channel = output_channel
+            output_channel = channels[i]
+            is_final_block = i == len(channels)  # - 1
+
+            down_block = get_down_block(
+                spatial_dims=spatial_dims,
+                in_channels=input_channel,
+                out_channels=output_channel,
+                temb_channels=time_embed_dim,
+                num_res_blocks=num_res_blocks[i],
+                norm_num_groups=norm_num_groups,
+                norm_eps=norm_eps,
+                add_downsample=not is_final_block,
+                resblock_updown=resblock_updown,
+                with_attn=(attention_levels[i] and not with_conditioning),
+                with_cross_attn=(attention_levels[i] and with_conditioning),
+                num_head_channels=num_head_channels[i],
+                transformer_num_layers=transformer_num_layers,
+                cross_attention_dim=cross_attention_dim,
+                upcast_attention=upcast_attention,
+            )
+
+            self.down_blocks.append(down_block)
+
+        self.out = nn.Sequential(nn.Linear(4096, 512), nn.ReLU(), nn.Dropout(0.1), nn.Linear(512, self.out_channels))
+
+    def forward(
+        self,
+        x: torch.Tensor,
+        timesteps: torch.Tensor,
+        context: torch.Tensor | None = None,
+        class_labels: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        """
+        Args:
+            x: input tensor (N, C, SpatialDims).
+            timesteps: timestep tensor (N,).
+            context: context tensor (N, 1, ContextDim).
+            class_labels: context tensor (N, ).
+        """
+        # 1. time
+        t_emb = get_timestep_embedding(timesteps, self.block_out_channels[0])
+
+        # timesteps does not contain any weights and will always return f32 tensors
+        # but time_embedding might actually be running in fp16. so we need to cast here.
+        # there might be better ways to encapsulate this.
+        t_emb = t_emb.to(dtype=x.dtype)
+        emb = self.time_embed(t_emb)
+
+        # 2. class
+        if self.num_class_embeds is not None:
+            if class_labels is None:
+                raise ValueError("class_labels should be provided when num_class_embeds > 0")
+            class_emb = self.class_embedding(class_labels)
+            class_emb = class_emb.to(dtype=x.dtype)
+            emb = emb + class_emb
+
+        # 3. initial convolution
+        h = self.conv_in(x)
+
+        # 4. down
+        if context is not None and self.with_conditioning is False:
+            raise ValueError("model should have with_conditioning = True if context is provided")
+        for downsample_block in self.down_blocks:
+            h, _ = downsample_block(hidden_states=h, temb=emb, context=context)
+
+        h = h.reshape(h.shape[0], -1)
+        output: torch.Tensor = self.out(h)
+
+        return output
diff --git a/monai/networks/nets/patchgan_discriminator.py b/monai/networks/nets/patchgan_discriminator.py
new file mode 100644
index 0000000000..74da917694
--- /dev/null
+++ b/monai/networks/nets/patchgan_discriminator.py
@@ -0,0 +1,230 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from collections.abc import Sequence
+
+import torch
+import torch.nn as nn
+
+from monai.networks.blocks import Convolution
+from monai.networks.layers import Act
+from monai.networks.utils import normal_init
+
+
+class MultiScalePatchDiscriminator(nn.Sequential):
+    """
+    Multi-scale Patch-GAN discriminator based on Pix2PixHD:
+    High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs (https://arxiv.org/abs/1711.11585)
+
+    The Multi-scale discriminator made up of several PatchGAN discriminators, that process the images
+    at different spatial scales.
+
+    Args:
+        num_d: number of discriminators
+        num_layers_d: number of Convolution layers (Conv + activation + normalisation + [dropout]) in the first
+            discriminator. Each subsequent discriminator has one additional layer, meaning the output size is halved.
+        spatial_dims: number of spatial dimensions (1D, 2D etc.)
+        channels: number of filters in the first convolutional layer (doubled for each subsequent layer)
+        in_channels: number of input channels
+        out_channels: number of output channels in each discriminator
+        kernel_size: kernel size of the convolution layers
+        activation: activation layer type
+        norm: normalisation type
+        bias: introduction of layer bias
+        dropout: probability of dropout applied, defaults to 0.
+        minimum_size_im: minimum spatial size of the input image. Introduced to make sure the architecture
+            requested isn't going to downsample the input image beyond value of 1.
+        last_conv_kernel_size: kernel size of the last convolutional layer.
+    """
+
+    def __init__(
+        self,
+        num_d: int,
+        num_layers_d: int,
+        spatial_dims: int,
+        channels: int,
+        in_channels: int,
+        out_channels: int = 1,
+        kernel_size: int = 4,
+        activation: str | tuple = (Act.LEAKYRELU, {"negative_slope": 0.2}),
+        norm: str | tuple = "BATCH",
+        bias: bool = False,
+        dropout: float | tuple = 0.0,
+        minimum_size_im: int = 256,
+        last_conv_kernel_size: int = 1,
+    ) -> None:
+        super().__init__()
+        self.num_d = num_d
+        self.num_layers_d = num_layers_d
+        self.num_channels = channels
+        self.padding = tuple([int((kernel_size - 1) / 2)] * spatial_dims)
+        for i_ in range(self.num_d):
+            num_layers_d_i = self.num_layers_d * (i_ + 1)
+            output_size = float(minimum_size_im) / (2**num_layers_d_i)
+            if output_size < 1:
+                raise AssertionError(
+                    f"Your image size is too small to take in up to {i_} discriminators with num_layers = {num_layers_d_i}."
+                    "Please reduce num_layers, reduce num_D or enter bigger images."
+                )
+            subnet_d = PatchDiscriminator(
+                spatial_dims=spatial_dims,
+                channels=self.num_channels,
+                in_channels=in_channels,
+                out_channels=out_channels,
+                num_layers_d=num_layers_d_i,
+                kernel_size=kernel_size,
+                activation=activation,
+                norm=norm,
+                bias=bias,
+                padding=self.padding,
+                dropout=dropout,
+                last_conv_kernel_size=last_conv_kernel_size,
+            )
+
+            self.add_module("discriminator_%d" % i_, subnet_d)
+
+    def forward(self, i: torch.Tensor) -> tuple[list[torch.Tensor], list[list[torch.Tensor]]]:
+        """
+        Args:
+            i: Input tensor
+
+        Returns:
+            list of outputs and another list of lists with the intermediate features
+            of each discriminator.
+        """
+
+        out: list[torch.Tensor] = []
+        intermediate_features: list[list[torch.Tensor]] = []
+        for disc in self.children():
+            out_d: list[torch.Tensor] = disc(i)
+            out.append(out_d[-1])
+            intermediate_features.append(out_d[:-1])
+
+        return out, intermediate_features
+
+
+class PatchDiscriminator(nn.Sequential):
+    """
+    Patch-GAN discriminator based on Pix2PixHD:
+    High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs (https://arxiv.org/abs/1711.11585)
+
+
+    Args:
+        spatial_dims: number of spatial dimensions (1D, 2D etc.)
+        channels: number of filters in the first convolutional layer (doubled for each subsequent layer)
+        in_channels: number of input channels
+        out_channels: number of output channels
+        num_layers_d: number of Convolution layers (Conv + activation + normalisation + [dropout]) in the discriminator.
+        kernel_size: kernel size of the convolution layers
+        act: activation type and arguments. Defaults to LeakyReLU.
+        norm: feature normalization type and arguments. Defaults to batch norm.
+        bias: whether to have a bias term in convolution blocks. Defaults to False.
+        padding: padding to be applied to the convolutional layers
+        dropout: proportion of dropout applied, defaults to 0.
+        last_conv_kernel_size: kernel size of the last convolutional layer.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        channels: int,
+        in_channels: int,
+        out_channels: int = 1,
+        num_layers_d: int = 3,
+        kernel_size: int = 4,
+        activation: str | tuple = (Act.LEAKYRELU, {"negative_slope": 0.2}),
+        norm: str | tuple = "BATCH",
+        bias: bool = False,
+        padding: int | Sequence[int] = 1,
+        dropout: float | tuple = 0.0,
+        last_conv_kernel_size: int | None = None,
+    ) -> None:
+        super().__init__()
+        self.num_layers_d = num_layers_d
+        self.num_channels = channels
+        if last_conv_kernel_size is None:
+            last_conv_kernel_size = kernel_size
+
+        self.add_module(
+            "initial_conv",
+            Convolution(
+                spatial_dims=spatial_dims,
+                kernel_size=kernel_size,
+                in_channels=in_channels,
+                out_channels=channels,
+                act=activation,
+                bias=True,
+                norm=None,
+                dropout=dropout,
+                padding=padding,
+                strides=2,
+            ),
+        )
+
+        input_channels = channels
+        output_channels = channels * 2
+
+        # Initial Layer
+        for l_ in range(self.num_layers_d):
+            if l_ == self.num_layers_d - 1:
+                stride = 1
+            else:
+                stride = 2
+            layer = Convolution(
+                spatial_dims=spatial_dims,
+                kernel_size=kernel_size,
+                in_channels=input_channels,
+                out_channels=output_channels,
+                act=activation,
+                bias=bias,
+                norm=norm,
+                dropout=dropout,
+                padding=padding,
+                strides=stride,
+            )
+            self.add_module("%d" % l_, layer)
+            input_channels = output_channels
+            output_channels = output_channels * 2
+
+        # Final layer
+        self.add_module(
+            "final_conv",
+            Convolution(
+                spatial_dims=spatial_dims,
+                kernel_size=last_conv_kernel_size,
+                in_channels=input_channels,
+                out_channels=out_channels,
+                bias=True,
+                conv_only=True,
+                padding=int((last_conv_kernel_size - 1) / 2),
+                dropout=0.0,
+                strides=1,
+            ),
+        )
+
+        self.apply(normal_init)
+
+    def forward(self, x: torch.Tensor) -> list[torch.Tensor]:
+        """
+        Args:
+            x: input tensor
+
+        Returns:
+            list of intermediate features, with the last element being the output.
+        """
+        out = [x]
+        for submodel in self.children():
+            intermediate_output = submodel(out[-1])
+            out.append(intermediate_output)
+
+        return out[1:]
diff --git a/monai/networks/nets/quicknat.py b/monai/networks/nets/quicknat.py
index cbcccf24d7..bbc4e7e490 100644
--- a/monai/networks/nets/quicknat.py
+++ b/monai/networks/nets/quicknat.py
@@ -168,6 +168,8 @@ def _get_layer(self, in_channels, out_channels, dilation):
     def forward(self, input, _):
         i = 0
         result = input
+        result1 = input  # this will not stay this value, needed here for pylint/mypy
+
         for l in self.children():
             # ignoring the max (un-)pool and droupout already added in the initial initialization step
             if isinstance(l, (nn.MaxPool2d, nn.MaxUnpool2d, nn.Dropout2d)):
diff --git a/monai/networks/nets/spade_autoencoderkl.py b/monai/networks/nets/spade_autoencoderkl.py
new file mode 100644
index 0000000000..294b121c94
--- /dev/null
+++ b/monai/networks/nets/spade_autoencoderkl.py
@@ -0,0 +1,480 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from collections.abc import Sequence
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+from monai.networks.blocks import Convolution, SpatialAttentionBlock, Upsample
+from monai.networks.blocks.spade_norm import SPADE
+from monai.networks.nets.autoencoderkl import Encoder
+from monai.utils import ensure_tuple_rep
+
+__all__ = ["SPADEAutoencoderKL"]
+
+
+class SPADEResBlock(nn.Module):
+    """
+    Residual block consisting of a cascade of 2 convolutions + activation + normalisation block, and a
+    residual connection between input and output.
+    Enables SPADE normalisation for semantic conditioning (Park et. al (2019): https://github.com/NVlabs/SPADE)
+
+    Args:
+        spatial_dims: number of spatial dimensions (1D, 2D, 3D).
+        in_channels: input channels to the layer.
+        norm_num_groups: number of groups involved for the group normalisation layer. Ensure that your number of
+            channels is divisible by this number.
+        norm_eps: epsilon for the normalisation.
+        out_channels: number of output channels.
+        label_nc: number of semantic channels for SPADE normalisation
+        spade_intermediate_channels: number of intermediate channels for SPADE block layer
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        norm_num_groups: int,
+        norm_eps: float,
+        out_channels: int,
+        label_nc: int,
+        spade_intermediate_channels: int,
+    ) -> None:
+        super().__init__()
+        self.in_channels = in_channels
+        self.out_channels = in_channels if out_channels is None else out_channels
+        self.norm1 = SPADE(
+            label_nc=label_nc,
+            norm_nc=in_channels,
+            norm="GROUP",
+            norm_params={"num_groups": norm_num_groups, "affine": False},
+            hidden_channels=spade_intermediate_channels,
+            kernel_size=3,
+            spatial_dims=spatial_dims,
+        )
+        self.conv1 = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=self.in_channels,
+            out_channels=self.out_channels,
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+        self.norm2 = SPADE(
+            label_nc=label_nc,
+            norm_nc=out_channels,
+            norm="GROUP",
+            norm_params={"num_groups": norm_num_groups, "affine": False},
+            hidden_channels=spade_intermediate_channels,
+            kernel_size=3,
+            spatial_dims=spatial_dims,
+        )
+        self.conv2 = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=self.out_channels,
+            out_channels=self.out_channels,
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+
+        self.nin_shortcut: nn.Module
+        if self.in_channels != self.out_channels:
+            self.nin_shortcut = Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=self.in_channels,
+                out_channels=self.out_channels,
+                strides=1,
+                kernel_size=1,
+                padding=0,
+                conv_only=True,
+            )
+        else:
+            self.nin_shortcut = nn.Identity()
+
+    def forward(self, x: torch.Tensor, seg: torch.Tensor) -> torch.Tensor:
+        h = x
+        h = self.norm1(h, seg)
+        h = F.silu(h)
+        h = self.conv1(h)
+        h = self.norm2(h, seg)
+        h = F.silu(h)
+        h = self.conv2(h)
+
+        x = self.nin_shortcut(x)
+
+        return x + h
+
+
+class SPADEDecoder(nn.Module):
+    """
+    Convolutional cascade upsampling from a spatial latent space into an image space.
+    Enables SPADE normalisation for semantic conditioning (Park et. al (2019): https://github.com/NVlabs/SPADE)
+
+    Args:
+        spatial_dims: number of spatial dimensions (1D, 2D, 3D).
+        channels: sequence of block output channels.
+        in_channels: number of channels in the bottom layer (latent space) of the autoencoder.
+        out_channels: number of output channels.
+        num_res_blocks: number of residual blocks (see ResBlock) per level.
+        norm_num_groups: number of groups for the GroupNorm layers, channels must be divisible by this number.
+        norm_eps: epsilon for the normalization.
+        attention_levels: indicate which level from channels contain an attention block.
+        label_nc: number of semantic channels for SPADE normalisation.
+        with_nonlocal_attn: if True use non-local attention block.
+        spade_intermediate_channels: number of intermediate channels for SPADE block layer.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        channels: Sequence[int],
+        in_channels: int,
+        out_channels: int,
+        num_res_blocks: Sequence[int],
+        norm_num_groups: int,
+        norm_eps: float,
+        attention_levels: Sequence[bool],
+        label_nc: int,
+        with_nonlocal_attn: bool = True,
+        spade_intermediate_channels: int = 128,
+    ) -> None:
+        super().__init__()
+        self.spatial_dims = spatial_dims
+        self.channels = channels
+        self.in_channels = in_channels
+        self.out_channels = out_channels
+        self.num_res_blocks = num_res_blocks
+        self.norm_num_groups = norm_num_groups
+        self.norm_eps = norm_eps
+        self.attention_levels = attention_levels
+        self.label_nc = label_nc
+
+        reversed_block_out_channels = list(reversed(channels))
+
+        blocks: list[nn.Module] = []
+
+        # Initial convolution
+        blocks.append(
+            Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=in_channels,
+                out_channels=reversed_block_out_channels[0],
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+
+        # Non-local attention block
+        if with_nonlocal_attn is True:
+            blocks.append(
+                SPADEResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=reversed_block_out_channels[0],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=reversed_block_out_channels[0],
+                    label_nc=label_nc,
+                    spade_intermediate_channels=spade_intermediate_channels,
+                )
+            )
+            blocks.append(
+                SpatialAttentionBlock(
+                    spatial_dims=spatial_dims,
+                    num_channels=reversed_block_out_channels[0],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+            blocks.append(
+                SPADEResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=reversed_block_out_channels[0],
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    out_channels=reversed_block_out_channels[0],
+                    label_nc=label_nc,
+                    spade_intermediate_channels=spade_intermediate_channels,
+                )
+            )
+
+        reversed_attention_levels = list(reversed(attention_levels))
+        reversed_num_res_blocks = list(reversed(num_res_blocks))
+        block_out_ch = reversed_block_out_channels[0]
+        for i in range(len(reversed_block_out_channels)):
+            block_in_ch = block_out_ch
+            block_out_ch = reversed_block_out_channels[i]
+            is_final_block = i == len(channels) - 1
+
+            for _ in range(reversed_num_res_blocks[i]):
+                blocks.append(
+                    SPADEResBlock(
+                        spatial_dims=spatial_dims,
+                        in_channels=block_in_ch,
+                        norm_num_groups=norm_num_groups,
+                        norm_eps=norm_eps,
+                        out_channels=block_out_ch,
+                        label_nc=label_nc,
+                        spade_intermediate_channels=spade_intermediate_channels,
+                    )
+                )
+                block_in_ch = block_out_ch
+
+                if reversed_attention_levels[i]:
+                    blocks.append(
+                        SpatialAttentionBlock(
+                            spatial_dims=spatial_dims,
+                            num_channels=block_in_ch,
+                            norm_num_groups=norm_num_groups,
+                            norm_eps=norm_eps,
+                        )
+                    )
+
+            if not is_final_block:
+                post_conv = Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=block_in_ch,
+                    out_channels=block_in_ch,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+                blocks.append(
+                    Upsample(
+                        spatial_dims=spatial_dims,
+                        mode="nontrainable",
+                        in_channels=block_in_ch,
+                        out_channels=block_in_ch,
+                        interp_mode="nearest",
+                        scale_factor=2.0,
+                        post_conv=post_conv,
+                        align_corners=None,
+                    )
+                )
+
+        blocks.append(nn.GroupNorm(num_groups=norm_num_groups, num_channels=block_in_ch, eps=norm_eps, affine=True))
+        blocks.append(
+            Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=block_in_ch,
+                out_channels=out_channels,
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+
+        self.blocks = nn.ModuleList(blocks)
+
+    def forward(self, x: torch.Tensor, seg: torch.Tensor) -> torch.Tensor:
+        for block in self.blocks:
+            if isinstance(block, SPADEResBlock):
+                x = block(x, seg)
+            else:
+                x = block(x)
+        return x
+
+
+class SPADEAutoencoderKL(nn.Module):
+    """
+    Autoencoder model with KL-regularized latent space based on
+    Rombach et al. "High-Resolution Image Synthesis with Latent Diffusion Models" https://arxiv.org/abs/2112.10752
+    and Pinaya et al. "Brain Imaging Generation with Latent Diffusion Models" https://arxiv.org/abs/2209.07162
+    Enables SPADE normalisation for semantic conditioning (Park et. al (2019): https://github.com/NVlabs/SPADE)
+
+    Args:
+        spatial_dims: number of spatial dimensions (1D, 2D, 3D).
+        label_nc: number of semantic channels for SPADE normalisation.
+        in_channels: number of input channels.
+        out_channels: number of output channels.
+        num_res_blocks: number of residual blocks (see ResBlock) per level.
+        channels: sequence of block output channels.
+        attention_levels: sequence of levels to add attention.
+        latent_channels: latent embedding dimension.
+        norm_num_groups: number of groups for the GroupNorm layers, channels must be divisible by this number.
+        norm_eps: epsilon for the normalization.
+        with_encoder_nonlocal_attn: if True use non-local attention block in the encoder.
+        with_decoder_nonlocal_attn: if True use non-local attention block in the decoder.
+        spade_intermediate_channels: number of intermediate channels for SPADE block layer.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        label_nc: int,
+        in_channels: int = 1,
+        out_channels: int = 1,
+        num_res_blocks: Sequence[int] | int = (2, 2, 2, 2),
+        channels: Sequence[int] = (32, 64, 64, 64),
+        attention_levels: Sequence[bool] = (False, False, True, True),
+        latent_channels: int = 3,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        with_encoder_nonlocal_attn: bool = True,
+        with_decoder_nonlocal_attn: bool = True,
+        spade_intermediate_channels: int = 128,
+    ) -> None:
+        super().__init__()
+
+        # All number of channels should be multiple of num_groups
+        if any((out_channel % norm_num_groups) != 0 for out_channel in channels):
+            raise ValueError("SPADEAutoencoderKL expects all channels being multiple of norm_num_groups")
+
+        if len(channels) != len(attention_levels):
+            raise ValueError("SPADEAutoencoderKL expects channels being same size of attention_levels")
+
+        if isinstance(num_res_blocks, int):
+            num_res_blocks = ensure_tuple_rep(num_res_blocks, len(channels))
+
+        if len(num_res_blocks) != len(channels):
+            raise ValueError(
+                "`num_res_blocks` should be a single integer or a tuple of integers with the same length as "
+                "`channels`."
+            )
+
+        self.encoder = Encoder(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            channels=channels,
+            out_channels=latent_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            attention_levels=attention_levels,
+            with_nonlocal_attn=with_encoder_nonlocal_attn,
+        )
+        self.decoder = SPADEDecoder(
+            spatial_dims=spatial_dims,
+            channels=channels,
+            in_channels=latent_channels,
+            out_channels=out_channels,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            attention_levels=attention_levels,
+            label_nc=label_nc,
+            with_nonlocal_attn=with_decoder_nonlocal_attn,
+            spade_intermediate_channels=spade_intermediate_channels,
+        )
+        self.quant_conv_mu = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=latent_channels,
+            out_channels=latent_channels,
+            strides=1,
+            kernel_size=1,
+            padding=0,
+            conv_only=True,
+        )
+        self.quant_conv_log_sigma = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=latent_channels,
+            out_channels=latent_channels,
+            strides=1,
+            kernel_size=1,
+            padding=0,
+            conv_only=True,
+        )
+        self.post_quant_conv = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=latent_channels,
+            out_channels=latent_channels,
+            strides=1,
+            kernel_size=1,
+            padding=0,
+            conv_only=True,
+        )
+        self.latent_channels = latent_channels
+
+    def encode(self, x: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor]:
+        """
+        Forwards an image through the spatial encoder, obtaining the latent mean and sigma representations.
+
+        Args:
+            x: BxCx[SPATIAL DIMS] tensor
+
+        """
+        h = self.encoder(x)
+        z_mu = self.quant_conv_mu(h)
+        z_log_var = self.quant_conv_log_sigma(h)
+        z_log_var = torch.clamp(z_log_var, -30.0, 20.0)
+        z_sigma = torch.exp(z_log_var / 2)
+
+        return z_mu, z_sigma
+
+    def sampling(self, z_mu: torch.Tensor, z_sigma: torch.Tensor) -> torch.Tensor:
+        """
+        From the mean and sigma representations resulting of encoding an image through the latent space,
+        obtains a noise sample resulting from sampling gaussian noise, multiplying by the variance (sigma) and
+        adding the mean.
+
+        Args:
+            z_mu: Bx[Z_CHANNELS]x[LATENT SPACE SIZE] mean vector obtained by the encoder when you encode an image
+            z_sigma: Bx[Z_CHANNELS]x[LATENT SPACE SIZE] variance vector obtained by the encoder when you encode an image
+
+        Returns:
+            sample of shape Bx[Z_CHANNELS]x[LATENT SPACE SIZE]
+        """
+        eps = torch.randn_like(z_sigma)
+        z_vae = z_mu + eps * z_sigma
+        return z_vae
+
+    def reconstruct(self, x: torch.Tensor, seg: torch.Tensor) -> torch.Tensor:
+        """
+        Encodes and decodes an input image.
+
+        Args:
+            x: BxCx[SPATIAL DIMENSIONS] tensor.
+            seg: Bx[LABEL_NC]x[SPATIAL DIMENSIONS] tensor of segmentations for SPADE norm.
+        Returns:
+            reconstructed image, of the same shape as input
+        """
+        z_mu, _ = self.encode(x)
+        reconstruction = self.decode(z_mu, seg)
+        return reconstruction
+
+    def decode(self, z: torch.Tensor, seg: torch.Tensor) -> torch.Tensor:
+        """
+        Based on a latent space sample, forwards it through the Decoder.
+
+        Args:
+            z: Bx[Z_CHANNELS]x[LATENT SPACE SHAPE]
+            seg: Bx[LABEL_NC]x[SPATIAL DIMENSIONS] tensor of segmentations for SPADE norm.
+        Returns:
+            decoded image tensor
+        """
+        z = self.post_quant_conv(z)
+        dec: torch.Tensor = self.decoder(z, seg)
+        return dec
+
+    def forward(self, x: torch.Tensor, seg: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
+        z_mu, z_sigma = self.encode(x)
+        z = self.sampling(z_mu, z_sigma)
+        reconstruction = self.decode(z, seg)
+        return reconstruction, z_mu, z_sigma
+
+    def encode_stage_2_inputs(self, x: torch.Tensor) -> torch.Tensor:
+        z_mu, z_sigma = self.encode(x)
+        z = self.sampling(z_mu, z_sigma)
+        return z
+
+    def decode_stage_2_outputs(self, z: torch.Tensor, seg: torch.Tensor) -> torch.Tensor:
+        image = self.decode(z, seg)
+        return image
diff --git a/monai/networks/nets/spade_diffusion_model_unet.py b/monai/networks/nets/spade_diffusion_model_unet.py
new file mode 100644
index 0000000000..75d1687df3
--- /dev/null
+++ b/monai/networks/nets/spade_diffusion_model_unet.py
@@ -0,0 +1,934 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# =========================================================================
+# Adapted from https://github.com/huggingface/diffusers
+# which has the following license:
+# https://github.com/huggingface/diffusers/blob/main/LICENSE
+#
+# Copyright 2022 UC Berkeley Team and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =========================================================================
+
+from __future__ import annotations
+
+from collections.abc import Sequence
+
+import torch
+from torch import nn
+
+from monai.networks.blocks import Convolution, SpatialAttentionBlock
+from monai.networks.blocks.spade_norm import SPADE
+from monai.networks.nets.diffusion_model_unet import (
+    DiffusionUnetDownsample,
+    DiffusionUNetResnetBlock,
+    SpatialTransformer,
+    WrappedUpsample,
+    get_down_block,
+    get_mid_block,
+    get_timestep_embedding,
+    zero_module,
+)
+from monai.utils import ensure_tuple_rep
+
+__all__ = ["SPADEDiffusionModelUNet"]
+
+
+class SPADEDiffResBlock(nn.Module):
+    """
+    Residual block with timestep conditioning and SPADE norm.
+    Enables SPADE normalisation for semantic conditioning (Park et. al (2019): https://github.com/NVlabs/SPADE)
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        temb_channels: number of timestep embedding  channels.
+        label_nc: number of semantic channels for SPADE normalisation.
+        out_channels: number of output channels.
+        up: if True, performs upsampling.
+        down: if True, performs downsampling.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        spade_intermediate_channels: number of intermediate channels for SPADE block layer
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        temb_channels: int,
+        label_nc: int,
+        out_channels: int | None = None,
+        up: bool = False,
+        down: bool = False,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        spade_intermediate_channels: int = 128,
+    ) -> None:
+        super().__init__()
+        self.spatial_dims = spatial_dims
+        self.channels = in_channels
+        self.emb_channels = temb_channels
+        self.out_channels = out_channels or in_channels
+        self.up = up
+        self.down = down
+
+        self.norm1 = SPADE(
+            label_nc=label_nc,
+            norm_nc=in_channels,
+            norm="GROUP",
+            norm_params={"num_groups": norm_num_groups, "eps": norm_eps, "affine": True},
+            hidden_channels=spade_intermediate_channels,
+            kernel_size=3,
+            spatial_dims=spatial_dims,
+        )
+
+        self.nonlinearity = nn.SiLU()
+        self.conv1 = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=self.out_channels,
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+
+        self.upsample = self.downsample = None
+        if self.up:
+            self.upsample = WrappedUpsample(
+                spatial_dims=spatial_dims,
+                mode="nontrainable",
+                in_channels=in_channels,
+                out_channels=in_channels,
+                interp_mode="nearest",
+                scale_factor=2.0,
+                align_corners=None,
+            )
+        elif down:
+            self.downsample = DiffusionUnetDownsample(spatial_dims, in_channels, use_conv=False)
+
+        self.time_emb_proj = nn.Linear(temb_channels, self.out_channels)
+
+        self.norm2 = SPADE(
+            label_nc=label_nc,
+            norm_nc=self.out_channels,
+            norm="GROUP",
+            norm_params={"num_groups": norm_num_groups, "eps": norm_eps, "affine": True},
+            hidden_channels=spade_intermediate_channels,
+            kernel_size=3,
+            spatial_dims=spatial_dims,
+        )
+        self.conv2 = zero_module(
+            Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=self.out_channels,
+                out_channels=self.out_channels,
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+        self.skip_connection: nn.Module
+
+        if self.out_channels == in_channels:
+            self.skip_connection = nn.Identity()
+        else:
+            self.skip_connection = Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=in_channels,
+                out_channels=self.out_channels,
+                strides=1,
+                kernel_size=1,
+                padding=0,
+                conv_only=True,
+            )
+
+    def forward(self, x: torch.Tensor, emb: torch.Tensor, seg: torch.Tensor) -> torch.Tensor:
+        h = x
+        h = self.norm1(h, seg)
+        h = self.nonlinearity(h)
+
+        if self.upsample is not None:
+            x = self.upsample(x)
+            h = self.upsample(h)
+        elif self.downsample is not None:
+            x = self.downsample(x)
+            h = self.downsample(h)
+
+        h = self.conv1(h)
+
+        if self.spatial_dims == 2:
+            temb = self.time_emb_proj(self.nonlinearity(emb))[:, :, None, None]
+        else:
+            temb = self.time_emb_proj(self.nonlinearity(emb))[:, :, None, None, None]
+        h = h + temb
+
+        h = self.norm2(h, seg)
+        h = self.nonlinearity(h)
+        h = self.conv2(h)
+        output: torch.Tensor = self.skip_connection(x) + h
+        return output
+
+
+class SPADEUpBlock(nn.Module):
+    """
+    Unet's up block containing resnet and upsamplers blocks.
+    Enables SPADE normalisation for semantic conditioning (Park et. al (2019): https://github.com/NVlabs/SPADE)
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        prev_output_channel: number of channels from residual connection.
+        out_channels: number of output channels.
+        temb_channels: number of timestep embedding channels.
+        label_nc: number of semantic channels for SPADE normalisation.
+        num_res_blocks: number of residual blocks.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        add_upsample: if True add downsample block.
+        resblock_updown: if True use residual blocks for upsampling.
+        spade_intermediate_channels: number of intermediate channels for SPADE block layer.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        prev_output_channel: int,
+        out_channels: int,
+        temb_channels: int,
+        label_nc: int,
+        num_res_blocks: int = 1,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        add_upsample: bool = True,
+        resblock_updown: bool = False,
+        spade_intermediate_channels: int = 128,
+    ) -> None:
+        super().__init__()
+        self.resblock_updown = resblock_updown
+        resnets = []
+
+        for i in range(num_res_blocks):
+            res_skip_channels = in_channels if (i == num_res_blocks - 1) else out_channels
+            resnet_in_channels = prev_output_channel if i == 0 else out_channels
+
+            resnets.append(
+                SPADEDiffResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=resnet_in_channels + res_skip_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    label_nc=label_nc,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    spade_intermediate_channels=spade_intermediate_channels,
+                )
+            )
+
+        self.resnets = nn.ModuleList(resnets)
+
+        self.upsampler: nn.Module | None
+        if add_upsample:
+            if resblock_updown:
+                self.upsampler = DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    up=True,
+                )
+            else:
+                post_conv = Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+                self.upsampler = WrappedUpsample(
+                    spatial_dims=spatial_dims,
+                    mode="nontrainable",
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    interp_mode="nearest",
+                    scale_factor=2.0,
+                    post_conv=post_conv,
+                    align_corners=None,
+                )
+        else:
+            self.upsampler = None
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        res_hidden_states_list: list[torch.Tensor],
+        temb: torch.Tensor,
+        seg: torch.Tensor,
+        context: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        del context
+        for resnet in self.resnets:
+            # pop res hidden states
+            res_hidden_states = res_hidden_states_list[-1]
+            res_hidden_states_list = res_hidden_states_list[:-1]
+            hidden_states = torch.cat([hidden_states, res_hidden_states], dim=1)
+            hidden_states = resnet(hidden_states, temb, seg)
+
+        if self.upsampler is not None:
+            hidden_states = self.upsampler(hidden_states, temb)
+
+        return hidden_states
+
+
+class SPADEAttnUpBlock(nn.Module):
+    """
+    Unet's up block containing resnet, upsamplers, and self-attention blocks.
+    Enables SPADE normalisation for semantic conditioning (Park et. al (2019): https://github.com/NVlabs/SPADE)
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        prev_output_channel: number of channels from residual connection.
+        out_channels: number of output channels.
+        temb_channels: number of timestep embedding channels.
+        label_nc: number of semantic channels for SPADE normalisation
+        num_res_blocks: number of residual blocks.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        add_upsample: if True add downsample block.
+        resblock_updown: if True use residual blocks for upsampling.
+        num_head_channels: number of channels in each attention head.
+        spade_intermediate_channels: number of intermediate channels for SPADE block layer
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        prev_output_channel: int,
+        out_channels: int,
+        temb_channels: int,
+        label_nc: int,
+        num_res_blocks: int = 1,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        add_upsample: bool = True,
+        resblock_updown: bool = False,
+        num_head_channels: int = 1,
+        spade_intermediate_channels: int = 128,
+    ) -> None:
+        super().__init__()
+        self.resblock_updown = resblock_updown
+        resnets = []
+        attentions = []
+
+        for i in range(num_res_blocks):
+            res_skip_channels = in_channels if (i == num_res_blocks - 1) else out_channels
+            resnet_in_channels = prev_output_channel if i == 0 else out_channels
+
+            resnets.append(
+                SPADEDiffResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=resnet_in_channels + res_skip_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    label_nc=label_nc,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    spade_intermediate_channels=spade_intermediate_channels,
+                )
+            )
+            attentions.append(
+                SpatialAttentionBlock(
+                    spatial_dims=spatial_dims,
+                    num_channels=out_channels,
+                    num_head_channels=num_head_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                )
+            )
+
+        self.resnets = nn.ModuleList(resnets)
+        self.attentions = nn.ModuleList(attentions)
+
+        self.upsampler: nn.Module | None
+        if add_upsample:
+            if resblock_updown:
+                self.upsampler = DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    up=True,
+                )
+            else:
+                post_conv = Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+                self.upsampler = WrappedUpsample(
+                    spatial_dims=spatial_dims,
+                    mode="nontrainable",
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    interp_mode="nearest",
+                    scale_factor=2.0,
+                    post_conv=post_conv,
+                    align_corners=None,
+                )
+        else:
+            self.upsampler = None
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        res_hidden_states_list: list[torch.Tensor],
+        temb: torch.Tensor,
+        seg: torch.Tensor,
+        context: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        del context
+        for resnet, attn in zip(self.resnets, self.attentions):
+            # pop res hidden states
+            res_hidden_states = res_hidden_states_list[-1]
+            res_hidden_states_list = res_hidden_states_list[:-1]
+            hidden_states = torch.cat([hidden_states, res_hidden_states], dim=1)
+            hidden_states = resnet(hidden_states, temb, seg)
+            hidden_states = attn(hidden_states).contiguous()
+
+        if self.upsampler is not None:
+            hidden_states = self.upsampler(hidden_states, temb)
+
+        return hidden_states
+
+
+class SPADECrossAttnUpBlock(nn.Module):
+    """
+    Unet's up block containing resnet, upsamplers, and self-attention blocks.
+    Enables SPADE normalisation for semantic conditioning (Park et. al (2019): https://github.com/NVlabs/SPADE)
+
+    Args:
+        spatial_dims: The number of spatial dimensions.
+        in_channels: number of input channels.
+        prev_output_channel: number of channels from residual connection.
+        out_channels: number of output channels.
+        temb_channels: number of timestep embedding channels.
+        label_nc: number of semantic channels for SPADE normalisation.
+        num_res_blocks: number of residual blocks.
+        norm_num_groups: number of groups for the group normalization.
+        norm_eps: epsilon for the group normalization.
+        add_upsample: if True add downsample block.
+        resblock_updown: if True use residual blocks for upsampling.
+        num_head_channels: number of channels in each attention head.
+        transformer_num_layers: number of layers of Transformer blocks to use.
+        cross_attention_dim: number of context dimensions to use.
+        upcast_attention: if True, upcast attention operations to full precision.
+        spade_intermediate_channels: number of intermediate channels for SPADE block layer.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        prev_output_channel: int,
+        out_channels: int,
+        temb_channels: int,
+        label_nc: int,
+        num_res_blocks: int = 1,
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        add_upsample: bool = True,
+        resblock_updown: bool = False,
+        num_head_channels: int = 1,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        upcast_attention: bool = False,
+        spade_intermediate_channels: int = 128,
+    ) -> None:
+        super().__init__()
+        self.resblock_updown = resblock_updown
+        resnets = []
+        attentions = []
+
+        for i in range(num_res_blocks):
+            res_skip_channels = in_channels if (i == num_res_blocks - 1) else out_channels
+            resnet_in_channels = prev_output_channel if i == 0 else out_channels
+
+            resnets.append(
+                SPADEDiffResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=resnet_in_channels + res_skip_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    label_nc=label_nc,
+                    spade_intermediate_channels=spade_intermediate_channels,
+                )
+            )
+            attentions.append(
+                SpatialTransformer(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    num_attention_heads=out_channels // num_head_channels,
+                    num_head_channels=num_head_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    num_layers=transformer_num_layers,
+                    cross_attention_dim=cross_attention_dim,
+                    upcast_attention=upcast_attention,
+                )
+            )
+
+        self.attentions = nn.ModuleList(attentions)
+        self.resnets = nn.ModuleList(resnets)
+
+        self.upsampler: nn.Module | None
+        if add_upsample:
+            if resblock_updown:
+                self.upsampler = DiffusionUNetResnetBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    temb_channels=temb_channels,
+                    norm_num_groups=norm_num_groups,
+                    norm_eps=norm_eps,
+                    up=True,
+                )
+            else:
+                post_conv = Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+                self.upsampler = WrappedUpsample(
+                    spatial_dims=spatial_dims,
+                    mode="nontrainable",
+                    in_channels=out_channels,
+                    out_channels=out_channels,
+                    interp_mode="nearest",
+                    scale_factor=2.0,
+                    post_conv=post_conv,
+                    align_corners=None,
+                )
+        else:
+            self.upsampler = None
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        res_hidden_states_list: list[torch.Tensor],
+        temb: torch.Tensor,
+        seg: torch.Tensor | None = None,
+        context: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        for resnet, attn in zip(self.resnets, self.attentions):
+            # pop res hidden states
+            res_hidden_states = res_hidden_states_list[-1]
+            res_hidden_states_list = res_hidden_states_list[:-1]
+            hidden_states = torch.cat([hidden_states, res_hidden_states], dim=1)
+            hidden_states = resnet(hidden_states, temb, seg)
+            hidden_states = attn(hidden_states, context=context).contiguous()
+
+        if self.upsampler is not None:
+            hidden_states = self.upsampler(hidden_states, temb)
+
+        return hidden_states
+
+
+def get_spade_up_block(
+    spatial_dims: int,
+    in_channels: int,
+    prev_output_channel: int,
+    out_channels: int,
+    temb_channels: int,
+    num_res_blocks: int,
+    norm_num_groups: int,
+    norm_eps: float,
+    add_upsample: bool,
+    resblock_updown: bool,
+    with_attn: bool,
+    with_cross_attn: bool,
+    num_head_channels: int,
+    transformer_num_layers: int,
+    label_nc: int,
+    cross_attention_dim: int | None,
+    upcast_attention: bool = False,
+    spade_intermediate_channels: int = 128,
+) -> nn.Module:
+    if with_attn:
+        return SPADEAttnUpBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            prev_output_channel=prev_output_channel,
+            out_channels=out_channels,
+            temb_channels=temb_channels,
+            label_nc=label_nc,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            add_upsample=add_upsample,
+            resblock_updown=resblock_updown,
+            num_head_channels=num_head_channels,
+            spade_intermediate_channels=spade_intermediate_channels,
+        )
+    elif with_cross_attn:
+        return SPADECrossAttnUpBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            prev_output_channel=prev_output_channel,
+            out_channels=out_channels,
+            temb_channels=temb_channels,
+            label_nc=label_nc,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            add_upsample=add_upsample,
+            resblock_updown=resblock_updown,
+            num_head_channels=num_head_channels,
+            transformer_num_layers=transformer_num_layers,
+            cross_attention_dim=cross_attention_dim,
+            upcast_attention=upcast_attention,
+            spade_intermediate_channels=spade_intermediate_channels,
+        )
+    else:
+        return SPADEUpBlock(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            prev_output_channel=prev_output_channel,
+            out_channels=out_channels,
+            temb_channels=temb_channels,
+            label_nc=label_nc,
+            num_res_blocks=num_res_blocks,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            add_upsample=add_upsample,
+            resblock_updown=resblock_updown,
+            spade_intermediate_channels=spade_intermediate_channels,
+        )
+
+
+class SPADEDiffusionModelUNet(nn.Module):
+    """
+    UNet network with timestep embedding and attention mechanisms for conditioning, with added SPADE normalization for
+    semantic conditioning (Park et.al (2019): https://github.com/NVlabs/SPADE). An example tutorial can be found at
+    https://github.com/Project-MONAI/GenerativeModels/tree/main/tutorials/generative/2d_spade_ldm
+
+    Args:
+        spatial_dims: number of spatial dimensions.
+        in_channels: number of input channels.
+        out_channels: number of output channels.
+        label_nc: number of semantic channels for SPADE normalisation.
+        num_res_blocks: number of residual blocks (see ResnetBlock) per level.
+        channels: tuple of block output channels.
+        attention_levels: list of levels to add attention.
+        norm_num_groups: number of groups for the normalization.
+        norm_eps: epsilon for the normalization.
+        resblock_updown: if True use residual blocks for up/downsampling.
+        num_head_channels: number of channels in each attention head.
+        with_conditioning: if True add spatial transformers to perform conditioning.
+        transformer_num_layers: number of layers of Transformer blocks to use.
+        cross_attention_dim: number of context dimensions to use.
+        num_class_embeds: if specified (as an int), then this model will be class-conditional with `num_class_embeds`
+        classes.
+        upcast_attention: if True, upcast attention operations to full precision.
+        spade_intermediate_channels: number of intermediate channels for SPADE block layer
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        label_nc: int,
+        num_res_blocks: Sequence[int] | int = (2, 2, 2, 2),
+        channels: Sequence[int] = (32, 64, 64, 64),
+        attention_levels: Sequence[bool] = (False, False, True, True),
+        norm_num_groups: int = 32,
+        norm_eps: float = 1e-6,
+        resblock_updown: bool = False,
+        num_head_channels: int | Sequence[int] = 8,
+        with_conditioning: bool = False,
+        transformer_num_layers: int = 1,
+        cross_attention_dim: int | None = None,
+        num_class_embeds: int | None = None,
+        upcast_attention: bool = False,
+        spade_intermediate_channels: int = 128,
+    ) -> None:
+        super().__init__()
+        if with_conditioning is True and cross_attention_dim is None:
+            raise ValueError(
+                "SPADEDiffusionModelUNet expects dimension of the cross-attention conditioning (cross_attention_dim) "
+                "when using with_conditioning."
+            )
+        if cross_attention_dim is not None and with_conditioning is False:
+            raise ValueError(
+                "SPADEDiffusionModelUNet expects with_conditioning=True when specifying the cross_attention_dim."
+            )
+
+        # All number of channels should be multiple of num_groups
+        if any((out_channel % norm_num_groups) != 0 for out_channel in channels):
+            raise ValueError("SPADEDiffusionModelUNet expects all num_channels being multiple of norm_num_groups")
+
+        if len(channels) != len(attention_levels):
+            raise ValueError("SPADEDiffusionModelUNet expects num_channels being same size of attention_levels")
+
+        if isinstance(num_head_channels, int):
+            num_head_channels = ensure_tuple_rep(num_head_channels, len(attention_levels))
+
+        if len(num_head_channels) != len(attention_levels):
+            raise ValueError(
+                "num_head_channels should have the same length as attention_levels. For the i levels without attention,"
+                " i.e. `attention_level[i]=False`, the num_head_channels[i] will be ignored."
+            )
+
+        if isinstance(num_res_blocks, int):
+            num_res_blocks = ensure_tuple_rep(num_res_blocks, len(channels))
+
+        if len(num_res_blocks) != len(channels):
+            raise ValueError(
+                "`num_res_blocks` should be a single integer or a tuple of integers with the same length as "
+                "`num_channels`."
+            )
+
+        self.in_channels = in_channels
+        self.block_out_channels = channels
+        self.out_channels = out_channels
+        self.num_res_blocks = num_res_blocks
+        self.attention_levels = attention_levels
+        self.num_head_channels = num_head_channels
+        self.with_conditioning = with_conditioning
+        self.label_nc = label_nc
+
+        # input
+        self.conv_in = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=channels[0],
+            strides=1,
+            kernel_size=3,
+            padding=1,
+            conv_only=True,
+        )
+
+        # time
+        time_embed_dim = channels[0] * 4
+        self.time_embed = nn.Sequential(
+            nn.Linear(channels[0], time_embed_dim), nn.SiLU(), nn.Linear(time_embed_dim, time_embed_dim)
+        )
+
+        # class embedding
+        self.num_class_embeds = num_class_embeds
+        if num_class_embeds is not None:
+            self.class_embedding = nn.Embedding(num_class_embeds, time_embed_dim)
+
+        # down
+        self.down_blocks = nn.ModuleList([])
+        output_channel = channels[0]
+        for i in range(len(channels)):
+            input_channel = output_channel
+            output_channel = channels[i]
+            is_final_block = i == len(channels) - 1
+
+            down_block = get_down_block(
+                spatial_dims=spatial_dims,
+                in_channels=input_channel,
+                out_channels=output_channel,
+                temb_channels=time_embed_dim,
+                num_res_blocks=num_res_blocks[i],
+                norm_num_groups=norm_num_groups,
+                norm_eps=norm_eps,
+                add_downsample=not is_final_block,
+                resblock_updown=resblock_updown,
+                with_attn=(attention_levels[i] and not with_conditioning),
+                with_cross_attn=(attention_levels[i] and with_conditioning),
+                num_head_channels=num_head_channels[i],
+                transformer_num_layers=transformer_num_layers,
+                cross_attention_dim=cross_attention_dim,
+                upcast_attention=upcast_attention,
+            )
+
+            self.down_blocks.append(down_block)
+
+        # mid
+        self.middle_block = get_mid_block(
+            spatial_dims=spatial_dims,
+            in_channels=channels[-1],
+            temb_channels=time_embed_dim,
+            norm_num_groups=norm_num_groups,
+            norm_eps=norm_eps,
+            with_conditioning=with_conditioning,
+            num_head_channels=num_head_channels[-1],
+            transformer_num_layers=transformer_num_layers,
+            cross_attention_dim=cross_attention_dim,
+            upcast_attention=upcast_attention,
+        )
+
+        # up
+        self.up_blocks = nn.ModuleList([])
+        reversed_block_out_channels = list(reversed(channels))
+        reversed_num_res_blocks = list(reversed(num_res_blocks))
+        reversed_attention_levels = list(reversed(attention_levels))
+        reversed_num_head_channels = list(reversed(num_head_channels))
+        output_channel = reversed_block_out_channels[0]
+        for i in range(len(reversed_block_out_channels)):
+            prev_output_channel = output_channel
+            output_channel = reversed_block_out_channels[i]
+            input_channel = reversed_block_out_channels[min(i + 1, len(channels) - 1)]
+
+            is_final_block = i == len(channels) - 1
+
+            up_block = get_spade_up_block(
+                spatial_dims=spatial_dims,
+                in_channels=input_channel,
+                prev_output_channel=prev_output_channel,
+                out_channels=output_channel,
+                temb_channels=time_embed_dim,
+                num_res_blocks=reversed_num_res_blocks[i] + 1,
+                norm_num_groups=norm_num_groups,
+                norm_eps=norm_eps,
+                add_upsample=not is_final_block,
+                resblock_updown=resblock_updown,
+                with_attn=(reversed_attention_levels[i] and not with_conditioning),
+                with_cross_attn=(reversed_attention_levels[i] and with_conditioning),
+                num_head_channels=reversed_num_head_channels[i],
+                transformer_num_layers=transformer_num_layers,
+                cross_attention_dim=cross_attention_dim,
+                upcast_attention=upcast_attention,
+                label_nc=label_nc,
+                spade_intermediate_channels=spade_intermediate_channels,
+            )
+
+            self.up_blocks.append(up_block)
+
+        # out
+        self.out = nn.Sequential(
+            nn.GroupNorm(num_groups=norm_num_groups, num_channels=channels[0], eps=norm_eps, affine=True),
+            nn.SiLU(),
+            zero_module(
+                Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=channels[0],
+                    out_channels=out_channels,
+                    strides=1,
+                    kernel_size=3,
+                    padding=1,
+                    conv_only=True,
+                )
+            ),
+        )
+
+    def forward(
+        self,
+        x: torch.Tensor,
+        timesteps: torch.Tensor,
+        seg: torch.Tensor,
+        context: torch.Tensor | None = None,
+        class_labels: torch.Tensor | None = None,
+        down_block_additional_residuals: tuple[torch.Tensor] | None = None,
+        mid_block_additional_residual: torch.Tensor | None = None,
+    ) -> torch.Tensor:
+        """
+        Args:
+            x: input tensor (N, C, SpatialDims).
+            timesteps: timestep tensor (N,).
+            seg: Bx[LABEL_NC]x[SPATIAL DIMENSIONS] tensor of segmentations for SPADE norm.
+            context: context tensor (N, 1, ContextDim).
+            class_labels: context tensor (N, ).
+            down_block_additional_residuals: additional residual tensors for down blocks (N, C, FeatureMapsDims).
+            mid_block_additional_residual: additional residual tensor for mid block (N, C, FeatureMapsDims).
+        """
+        # 1. time
+        t_emb = get_timestep_embedding(timesteps, self.block_out_channels[0])
+
+        # timesteps does not contain any weights and will always return f32 tensors
+        # but time_embedding might actually be running in fp16. so we need to cast here.
+        # there might be better ways to encapsulate this.
+        t_emb = t_emb.to(dtype=x.dtype)
+        emb = self.time_embed(t_emb)
+
+        # 2. class
+        if self.num_class_embeds is not None:
+            if class_labels is None:
+                raise ValueError("class_labels should be provided when num_class_embeds > 0")
+            class_emb = self.class_embedding(class_labels)
+            class_emb = class_emb.to(dtype=x.dtype)
+            emb = emb + class_emb
+
+        # 3. initial convolution
+        h = self.conv_in(x)
+
+        # 4. down
+        if context is not None and self.with_conditioning is False:
+            raise ValueError("model should have with_conditioning = True if context is provided")
+        down_block_res_samples: list[torch.Tensor] = [h]
+        for downsample_block in self.down_blocks:
+            h, res_samples = downsample_block(hidden_states=h, temb=emb, context=context)
+            for residual in res_samples:
+                down_block_res_samples.append(residual)
+
+        # Additional residual conections for Controlnets
+        if down_block_additional_residuals is not None:
+            new_down_block_res_samples: list[torch.Tensor] = [h]
+            for down_block_res_sample, down_block_additional_residual in zip(
+                down_block_res_samples, down_block_additional_residuals
+            ):
+                down_block_res_sample = down_block_res_sample + down_block_additional_residual
+                new_down_block_res_samples.append(down_block_res_sample)
+
+            down_block_res_samples = new_down_block_res_samples
+
+        # 5. mid
+        h = self.middle_block(hidden_states=h, temb=emb, context=context)
+
+        # Additional residual conections for Controlnets
+        if mid_block_additional_residual is not None:
+            h = h + mid_block_additional_residual
+
+        # 6. up
+        for upsample_block in self.up_blocks:
+            res_samples = down_block_res_samples[-len(upsample_block.resnets) :]
+            down_block_res_samples = down_block_res_samples[: -len(upsample_block.resnets)]
+            h = upsample_block(hidden_states=h, res_hidden_states_list=res_samples, seg=seg, temb=emb, context=context)
+
+        # 7. output block
+        output: torch.Tensor = self.out(h)
+
+        return output
diff --git a/monai/networks/nets/spade_network.py b/monai/networks/nets/spade_network.py
new file mode 100644
index 0000000000..9164541f27
--- /dev/null
+++ b/monai/networks/nets/spade_network.py
@@ -0,0 +1,435 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import Sequence
+
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+from monai.networks.blocks import Convolution
+from monai.networks.blocks.spade_norm import SPADE
+from monai.networks.layers import Act
+from monai.networks.layers.utils import get_act_layer
+from monai.utils.enums import StrEnum
+
+__all__ = ["SPADENet"]
+
+
+class UpsamplingModes(StrEnum):
+    bicubic = "bicubic"
+    nearest = "nearest"
+    bilinear = "bilinear"
+
+
+class SPADENetResBlock(nn.Module):
+    """
+    Creates a Residual Block with SPADE normalisation.
+
+    Args:
+        spatial_dims: number of spatial dimensions
+        in_channels: number of input channels
+        out_channels: number of output channels
+        label_nc: number of semantic channels that will be taken into account in SPADE normalisation blocks
+        spade_intermediate_channels: number of intermediate channels in the middle conv. layers in SPADE normalisation blocks
+        norm: base normalisation type used on top of SPADE
+        kernel_size: convolutional kernel size
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        label_nc: int,
+        spade_intermediate_channels: int = 128,
+        norm: str | tuple = "INSTANCE",
+        act: str | tuple = (Act.LEAKYRELU, {"negative_slope": 0.2}),
+        kernel_size: int = 3,
+    ):
+        super().__init__()
+        self.in_channels = in_channels
+        self.out_channels = out_channels
+        self.int_channels = min(self.in_channels, self.out_channels)
+        self.learned_shortcut = self.in_channels != self.out_channels
+        self.conv_0 = Convolution(
+            spatial_dims=spatial_dims, in_channels=self.in_channels, out_channels=self.int_channels, act=None, norm=None
+        )
+        self.conv_1 = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=self.int_channels,
+            out_channels=self.out_channels,
+            act=None,
+            norm=None,
+        )
+        self.activation = get_act_layer(act)
+        self.norm_0 = SPADE(
+            label_nc=label_nc,
+            norm_nc=self.in_channels,
+            kernel_size=kernel_size,
+            spatial_dims=spatial_dims,
+            hidden_channels=spade_intermediate_channels,
+            norm=norm,
+        )
+        self.norm_1 = SPADE(
+            label_nc=label_nc,
+            norm_nc=self.int_channels,
+            kernel_size=kernel_size,
+            spatial_dims=spatial_dims,
+            hidden_channels=spade_intermediate_channels,
+            norm=norm,
+        )
+
+        if self.learned_shortcut:
+            self.conv_s = Convolution(
+                spatial_dims=spatial_dims,
+                in_channels=self.in_channels,
+                out_channels=self.out_channels,
+                act=None,
+                norm=None,
+                kernel_size=1,
+            )
+            self.norm_s = SPADE(
+                label_nc=label_nc,
+                norm_nc=self.in_channels,
+                kernel_size=kernel_size,
+                spatial_dims=spatial_dims,
+                hidden_channels=spade_intermediate_channels,
+                norm=norm,
+            )
+
+    def forward(self, x, seg):
+        x_s = self.shortcut(x, seg)
+        dx = self.conv_0(self.activation(self.norm_0(x, seg)))
+        dx = self.conv_1(self.activation(self.norm_1(dx, seg)))
+        out = x_s + dx
+        return out
+
+    def shortcut(self, x, seg):
+        if self.learned_shortcut:
+            x_s = self.conv_s(self.norm_s(x, seg))
+        else:
+            x_s = x
+        return x_s
+
+
+class SPADEEncoder(nn.Module):
+    """
+    Encoding branch of a VAE compatible with a SPADE-like generator
+
+    Args:
+        spatial_dims: number of spatial dimensions
+        in_channels: number of input channels
+        z_dim: latent space dimension of the VAE containing the image sytle information
+        channels: number of output after each downsampling block
+        input_shape: spatial input shape of the tensor, necessary to do the reshaping after the linear layers
+        of the autoencoder (HxWx[D])
+        kernel_size: convolutional kernel size
+        norm: normalisation layer type
+        act: activation type
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        z_dim: int,
+        channels: Sequence[int],
+        input_shape: Sequence[int],
+        kernel_size: int = 3,
+        norm: str | tuple = "INSTANCE",
+        act: str | tuple = (Act.LEAKYRELU, {"negative_slope": 0.2}),
+    ):
+        super().__init__()
+        self.in_channels = in_channels
+        self.z_dim = z_dim
+        self.channels = channels
+        if len(input_shape) != spatial_dims:
+            raise ValueError("Length of parameter input shape must match spatial_dims; got %s" % (input_shape))
+        for s_ind, s_ in enumerate(input_shape):
+            if s_ / (2 ** len(channels)) != s_ // (2 ** len(channels)):
+                raise ValueError(
+                    "Each dimension of your input must be divisible by 2 ** (autoencoder depth)."
+                    "The shape in position %d, %d is not divisible by %d. " % (s_ind, s_, len(channels))
+                )
+        self.input_shape = input_shape
+        self.latent_spatial_shape = [s_ // (2 ** len(self.channels)) for s_ in self.input_shape]
+        blocks = []
+        ch_init = self.in_channels
+        for _, ch_value in enumerate(channels):
+            blocks.append(
+                Convolution(
+                    spatial_dims=spatial_dims,
+                    in_channels=ch_init,
+                    out_channels=ch_value,
+                    strides=2,
+                    kernel_size=kernel_size,
+                    norm=norm,
+                    act=act,
+                )
+            )
+            ch_init = ch_value
+
+        self.blocks = nn.ModuleList(blocks)
+        self.fc_mu = nn.Linear(
+            in_features=np.prod(self.latent_spatial_shape) * self.channels[-1], out_features=self.z_dim
+        )
+        self.fc_var = nn.Linear(
+            in_features=np.prod(self.latent_spatial_shape) * self.channels[-1], out_features=self.z_dim
+        )
+
+    def forward(self, x):
+        for block in self.blocks:
+            x = block(x)
+        x = x.view(x.size(0), -1)
+        mu = self.fc_mu(x)
+        logvar = self.fc_var(x)
+        return mu, logvar
+
+    def encode(self, x):
+        for block in self.blocks:
+            x = block(x)
+        x = x.view(x.size(0), -1)
+        mu = self.fc_mu(x)
+        logvar = self.fc_var(x)
+        return self.reparameterize(mu, logvar)
+
+    def reparameterize(self, mu, logvar):
+        std = torch.exp(0.5 * logvar)
+        eps = torch.randn_like(std)
+        return eps.mul(std) + mu
+
+
+class SPADEDecoder(nn.Module):
+    """
+    Decoder branch of a SPADE-like generator. It can be used independently, without an encoding branch,
+    behaving like a GAN, or coupled to a SPADE encoder.
+
+    Args:
+        label_nc: number of semantic labels
+        spatial_dims: number of spatial dimensions
+        out_channels: number of output channels
+        label_nc: number of semantic channels used for the SPADE normalisation blocks
+        input_shape: spatial input shape of the tensor, necessary to do the reshaping after the linear layers
+        channels: number of output after each downsampling block
+        z_dim: latent space dimension of the VAE containing the image sytle information (None if encoder is not used)
+        is_vae: whether the decoder is going to be coupled to an autoencoder or not (true: yes, false: no)
+        spade_intermediate_channels: number of channels in the intermediate layers of the SPADE normalisation blocks
+        norm: base normalisation type
+        act:  activation layer type
+        last_act: activation layer type for the last layer of the network (can differ from previous)
+        kernel_size: convolutional kernel size
+        upsampling_mode: upsampling mode (nearest, bilinear etc.)
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        out_channels: int,
+        label_nc: int,
+        input_shape: Sequence[int],
+        channels: list[int],
+        z_dim: int | None = None,
+        is_vae: bool = True,
+        spade_intermediate_channels: int = 128,
+        norm: str | tuple = "INSTANCE",
+        act: str | tuple = (Act.LEAKYRELU, {"negative_slope": 0.2}),
+        last_act: str | tuple | None = (Act.LEAKYRELU, {"negative_slope": 0.2}),
+        kernel_size: int = 3,
+        upsampling_mode: str = UpsamplingModes.nearest.value,
+    ):
+        super().__init__()
+        self.is_vae = is_vae
+        self.out_channels = out_channels
+        self.label_nc = label_nc
+        self.num_channels = channels
+        if len(input_shape) != spatial_dims:
+            raise ValueError("Length of parameter input shape must match spatial_dims; got %s" % (input_shape))
+        for s_ind, s_ in enumerate(input_shape):
+            if s_ / (2 ** len(channels)) != s_ // (2 ** len(channels)):
+                raise ValueError(
+                    "Each dimension of your input must be divisible by 2 ** (autoencoder depth)."
+                    "The shape in position %d, %d is not divisible by %d. " % (s_ind, s_, len(channels))
+                )
+        self.latent_spatial_shape = [s_ // (2 ** len(self.num_channels)) for s_ in input_shape]
+
+        if not self.is_vae:
+            self.conv_init = Convolution(
+                spatial_dims=spatial_dims, in_channels=label_nc, out_channels=channels[0], kernel_size=kernel_size
+            )
+        elif self.is_vae and z_dim is None:
+            raise ValueError(
+                "If the network is used in VAE-GAN mode, parameter z_dim "
+                "(number of latent channels in the VAE) must be populated."
+            )
+        else:
+            self.fc = nn.Linear(z_dim, np.prod(self.latent_spatial_shape) * channels[0])
+
+        self.z_dim = z_dim
+        blocks = []
+        channels.append(self.out_channels)
+        self.upsampling = torch.nn.Upsample(scale_factor=2, mode=upsampling_mode)
+        for ch_ind, ch_value in enumerate(channels[:-1]):
+            blocks.append(
+                SPADENetResBlock(
+                    spatial_dims=spatial_dims,
+                    in_channels=ch_value,
+                    out_channels=channels[ch_ind + 1],
+                    label_nc=label_nc,
+                    spade_intermediate_channels=spade_intermediate_channels,
+                    norm=norm,
+                    kernel_size=kernel_size,
+                    act=act,
+                )
+            )
+
+        self.blocks = torch.nn.ModuleList(blocks)
+        self.last_conv = Convolution(
+            spatial_dims=spatial_dims,
+            in_channels=channels[-1],
+            out_channels=out_channels,
+            padding=(kernel_size - 1) // 2,
+            kernel_size=kernel_size,
+            norm=None,
+            act=last_act,
+        )
+
+    def forward(self, seg, z: torch.Tensor | None = None):
+        """
+        Args:
+            seg: input BxCxHxW[xD] semantic map on which the output is conditioned on
+            z: latent vector output by the encoder if self.is_vae is True. When is_vae is
+            False, z is a random noise vector.
+
+        Returns:
+
+        """
+        if not self.is_vae:
+            x = F.interpolate(seg, size=tuple(self.latent_spatial_shape))
+            x = self.conv_init(x)
+        else:
+            if (
+                z is None and self.z_dim is not None
+            ):  # Even though this network is a VAE (self.is_vae), you should be able to sample from noise as well.
+                z = torch.randn(seg.size(0), self.z_dim, dtype=torch.float32, device=seg.get_device())
+            x = self.fc(z)
+            x = x.view(*[-1, self.num_channels[0]] + self.latent_spatial_shape)
+
+        for res_block in self.blocks:
+            x = res_block(x, seg)
+            x = self.upsampling(x)
+
+        x = self.last_conv(x)
+        return x
+
+
+class SPADENet(nn.Module):
+    """
+    SPADE Network, implemented based on the code by Park, T et al. in
+    "Semantic Image Synthesis with Spatially-Adaptive Normalization"
+    (https://github.com/NVlabs/SPADE)
+
+    Args:
+        spatial_dims: number of spatial dimensions
+        in_channels: number of input channels
+        out_channels: number of output channels
+        label_nc: number of semantic channels used for the SPADE normalisation blocks
+        input_shape:  spatial input shape of the tensor, necessary to do the reshaping after the linear layers
+        channels: number of output after each downsampling block
+        z_dim: latent space dimension of the VAE containing the image sytle information (None if encoder is not used)
+        is_vae: whether the decoder is going to be coupled to an autoencoder (true) or not (false)
+        spade_intermediate_channels: number of channels in the intermediate layers of the SPADE normalisation blocks
+        norm: base normalisation type
+        act: activation layer type
+        last_act: activation layer type for the last layer of the network (can differ from previous)
+        kernel_size: convolutional kernel size
+        upsampling_mode: upsampling mode (nearest, bilinear etc.)
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        label_nc: int,
+        input_shape: Sequence[int],
+        channels: list[int],
+        z_dim: int | None = None,
+        is_vae: bool = True,
+        spade_intermediate_channels: int = 128,
+        norm: str | tuple = "INSTANCE",
+        act: str | tuple = (Act.LEAKYRELU, {"negative_slope": 0.2}),
+        last_act: str | tuple | None = (Act.LEAKYRELU, {"negative_slope": 0.2}),
+        kernel_size: int = 3,
+        upsampling_mode: str = UpsamplingModes.nearest.value,
+    ):
+        super().__init__()
+        self.is_vae = is_vae
+        self.in_channels = in_channels
+        self.out_channels = out_channels
+        self.channels = channels
+        self.label_nc = label_nc
+        self.input_shape = input_shape
+
+        if self.is_vae:
+            if z_dim is None:
+                ValueError("The latent space dimension mapped by parameter z_dim cannot be None is is_vae is True.")
+            else:
+                self.encoder = SPADEEncoder(
+                    spatial_dims=spatial_dims,
+                    in_channels=in_channels,
+                    z_dim=z_dim,
+                    channels=channels,
+                    input_shape=input_shape,
+                    kernel_size=kernel_size,
+                    norm=norm,
+                    act=act,
+                )
+
+        decoder_channels = channels
+        decoder_channels.reverse()
+
+        self.decoder = SPADEDecoder(
+            spatial_dims=spatial_dims,
+            out_channels=out_channels,
+            label_nc=label_nc,
+            input_shape=input_shape,
+            channels=decoder_channels,
+            z_dim=z_dim,
+            is_vae=is_vae,
+            spade_intermediate_channels=spade_intermediate_channels,
+            norm=norm,
+            act=act,
+            last_act=last_act,
+            kernel_size=kernel_size,
+            upsampling_mode=upsampling_mode,
+        )
+
+    def forward(self, seg: torch.Tensor, x: torch.Tensor | None = None):
+        z = None
+        if self.is_vae:
+            z_mu, z_logvar = self.encoder(x)
+            z = self.encoder.reparameterize(z_mu, z_logvar)
+            return self.decoder(seg, z), z_mu, z_logvar
+        else:
+            return (self.decoder(seg, z),)
+
+    def encode(self, x: torch.Tensor):
+        if self.is_vae:
+            return self.encoder.encode(x)
+        else:
+            return None
+
+    def decode(self, seg: torch.Tensor, z: torch.Tensor | None = None):
+        return self.decoder(seg, z)
diff --git a/monai/networks/nets/swin_unetr.py b/monai/networks/nets/swin_unetr.py
index 6f96dfd291..3900c866b3 100644
--- a/monai/networks/nets/swin_unetr.py
+++ b/monai/networks/nets/swin_unetr.py
@@ -347,7 +347,7 @@ def window_partition(x, window_size):
         x: input tensor.
         window_size: local window size.
     """
-    x_shape = x.size()
+    x_shape = x.size()  # length 4 or 5 only
     if len(x_shape) == 5:
         b, d, h, w, c = x_shape
         x = x.view(
@@ -363,10 +363,11 @@ def window_partition(x, window_size):
         windows = (
             x.permute(0, 1, 3, 5, 2, 4, 6, 7).contiguous().view(-1, window_size[0] * window_size[1] * window_size[2], c)
         )
-    elif len(x_shape) == 4:
+    else:  # if len(x_shape) == 4:
         b, h, w, c = x.shape
         x = x.view(b, h // window_size[0], window_size[0], w // window_size[1], window_size[1], c)
         windows = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, window_size[0] * window_size[1], c)
+
     return windows
 
 
@@ -613,7 +614,7 @@ def forward_part1(self, x, mask_matrix):
             _, dp, hp, wp, _ = x.shape
             dims = [b, dp, hp, wp]
 
-        elif len(x_shape) == 4:
+        else:  # elif len(x_shape) == 4
             b, h, w, c = x.shape
             window_size, shift_size = get_window_size((h, w), self.window_size, self.shift_size)
             pad_l = pad_t = 0
diff --git a/monai/networks/nets/transformer.py b/monai/networks/nets/transformer.py
new file mode 100644
index 0000000000..1af725abda
--- /dev/null
+++ b/monai/networks/nets/transformer.py
@@ -0,0 +1,157 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import torch
+import torch.nn as nn
+
+from monai.networks.blocks import TransformerBlock
+
+__all__ = ["DecoderOnlyTransformer"]
+
+
+class AbsolutePositionalEmbedding(nn.Module):
+    """Absolute positional embedding.
+
+    Args:
+        max_seq_len: Maximum sequence length.
+        embedding_dim: Dimensionality of the embedding.
+    """
+
+    def __init__(self, max_seq_len: int, embedding_dim: int) -> None:
+        super().__init__()
+        self.max_seq_len = max_seq_len
+        self.embedding_dim = embedding_dim
+        self.embedding = nn.Embedding(max_seq_len, embedding_dim)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        batch_size, seq_len = x.size()
+        positions = torch.arange(seq_len, device=x.device).repeat(batch_size, 1)
+        embedding: torch.Tensor = self.embedding(positions)
+        return embedding
+
+
+class DecoderOnlyTransformer(nn.Module):
+    """Decoder-only (Autoregressive) Transformer model.
+
+    Args:
+        num_tokens: Number of tokens in the vocabulary.
+        max_seq_len: Maximum sequence length.
+        attn_layers_dim: Dimensionality of the attention layers.
+        attn_layers_depth: Number of attention layers.
+        attn_layers_heads: Number of attention heads.
+        with_cross_attention: Whether to use cross attention for conditioning.
+        embedding_dropout_rate: Dropout rate for the embedding.
+    """
+
+    def __init__(
+        self,
+        num_tokens: int,
+        max_seq_len: int,
+        attn_layers_dim: int,
+        attn_layers_depth: int,
+        attn_layers_heads: int,
+        with_cross_attention: bool = False,
+        embedding_dropout_rate: float = 0.0,
+    ) -> None:
+        super().__init__()
+        self.num_tokens = num_tokens
+        self.max_seq_len = max_seq_len
+        self.attn_layers_dim = attn_layers_dim
+        self.attn_layers_depth = attn_layers_depth
+        self.attn_layers_heads = attn_layers_heads
+        self.with_cross_attention = with_cross_attention
+
+        self.token_embeddings = nn.Embedding(num_tokens, attn_layers_dim)
+        self.position_embeddings = AbsolutePositionalEmbedding(max_seq_len=max_seq_len, embedding_dim=attn_layers_dim)
+        self.embedding_dropout = nn.Dropout(embedding_dropout_rate)
+
+        self.blocks = nn.ModuleList(
+            [
+                TransformerBlock(
+                    hidden_size=attn_layers_dim,
+                    mlp_dim=attn_layers_dim * 4,
+                    num_heads=attn_layers_heads,
+                    dropout_rate=0.0,
+                    qkv_bias=False,
+                    causal=True,
+                    sequence_length=max_seq_len,
+                    with_cross_attention=with_cross_attention,
+                )
+                for _ in range(attn_layers_depth)
+            ]
+        )
+
+        self.to_logits = nn.Linear(attn_layers_dim, num_tokens)
+
+    def forward(self, x: torch.Tensor, context: torch.Tensor | None = None) -> torch.Tensor:
+        tok_emb = self.token_embeddings(x)
+        pos_emb = self.position_embeddings(x)
+        x = self.embedding_dropout(tok_emb + pos_emb)
+
+        for block in self.blocks:
+            x = block(x, context=context)
+        logits: torch.Tensor = self.to_logits(x)
+        return logits
+
+    def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
+        """
+        Load a state dict from a DecoderOnlyTransformer trained with
+        [MONAI Generative](https://github.com/Project-MONAI/GenerativeModels).
+
+        Args:
+            old_state_dict: state dict from the old DecoderOnlyTransformer  model.
+        """
+
+        new_state_dict = self.state_dict()
+        # if all keys match, just load the state dict
+        if all(k in new_state_dict for k in old_state_dict):
+            print("All keys match, loading state dict.")
+            self.load_state_dict(old_state_dict)
+            return
+
+        if verbose:
+            # print all new_state_dict keys that are not in old_state_dict
+            for k in new_state_dict:
+                if k not in old_state_dict:
+                    print(f"key {k} not found in old state dict")
+            # and vice versa
+            print("----------------------------------------------")
+            for k in old_state_dict:
+                if k not in new_state_dict:
+                    print(f"key {k} not found in new state dict")
+
+        # copy over all matching keys
+        for k in new_state_dict:
+            if k in old_state_dict:
+                new_state_dict[k] = old_state_dict[k]
+
+        # fix the attention blocks
+        attention_blocks = [k.replace(".attn.qkv.weight", "") for k in new_state_dict if "attn.qkv.weight" in k]
+        for block in attention_blocks:
+            new_state_dict[f"{block}.attn.qkv.weight"] = torch.cat(
+                [
+                    old_state_dict[f"{block}.attn.to_q.weight"],
+                    old_state_dict[f"{block}.attn.to_k.weight"],
+                    old_state_dict[f"{block}.attn.to_v.weight"],
+                ],
+                dim=0,
+            )
+
+        # fix the renamed norm blocks first  norm2 -> norm_cross_attention , norm3 -> norm2
+        for k in old_state_dict:
+            if "norm2" in k:
+                new_state_dict[k.replace("norm2", "norm_cross_attn")] = old_state_dict[k]
+            if "norm3" in k:
+                new_state_dict[k.replace("norm3", "norm2")] = old_state_dict[k]
+
+        self.load_state_dict(new_state_dict)
diff --git a/monai/networks/nets/vqvae.py b/monai/networks/nets/vqvae.py
new file mode 100644
index 0000000000..f198bfbb2b
--- /dev/null
+++ b/monai/networks/nets/vqvae.py
@@ -0,0 +1,472 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from collections.abc import Sequence
+from typing import Tuple
+
+import torch
+import torch.nn as nn
+
+from monai.networks.blocks import Convolution
+from monai.networks.layers import Act
+from monai.networks.layers.vector_quantizer import EMAQuantizer, VectorQuantizer
+from monai.utils import ensure_tuple_rep
+
+__all__ = ["VQVAE"]
+
+
+class VQVAEResidualUnit(nn.Module):
+    """
+    Implementation of the ResidualLayer used in the VQVAE network as originally used in Morphology-preserving
+    Autoregressive 3D Generative Modelling of the Brain by Tudosiu et al. (https://arxiv.org/pdf/2209.03177.pdf).
+
+    The original implementation that can be found at
+    https://github.com/AmigoLab/SynthAnatomy/blob/main/src/networks/vqvae/baseline.py#L150.
+
+    Args:
+        spatial_dims: number of spatial spatial_dims of the input data.
+        in_channels: number of input channels.
+        num_res_channels: number of channels in the residual layers.
+        act: activation type and arguments. Defaults to RELU.
+        dropout: dropout ratio. Defaults to no dropout.
+        bias: whether to have a bias term. Defaults to True.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        num_res_channels: int,
+        act: tuple | str | None = Act.RELU,
+        dropout: float = 0.0,
+        bias: bool = True,
+    ) -> None:
+        super().__init__()
+
+        self.spatial_dims = spatial_dims
+        self.in_channels = in_channels
+        self.num_res_channels = num_res_channels
+        self.act = act
+        self.dropout = dropout
+        self.bias = bias
+
+        self.conv1 = Convolution(
+            spatial_dims=self.spatial_dims,
+            in_channels=self.in_channels,
+            out_channels=self.num_res_channels,
+            adn_ordering="DA",
+            act=self.act,
+            dropout=self.dropout,
+            bias=self.bias,
+        )
+
+        self.conv2 = Convolution(
+            spatial_dims=self.spatial_dims,
+            in_channels=self.num_res_channels,
+            out_channels=self.in_channels,
+            bias=self.bias,
+            conv_only=True,
+        )
+
+    def forward(self, x):
+        return torch.nn.functional.relu(x + self.conv2(self.conv1(x)), True)
+
+
+class Encoder(nn.Module):
+    """
+    Encoder module for VQ-VAE.
+
+    Args:
+        spatial_dims: number of spatial spatial_dims.
+        in_channels: number of input channels.
+        out_channels: number of channels in the latent space (embedding_dim).
+        channels: sequence containing the number of channels at each level of the encoder.
+        num_res_layers: number of sequential residual layers at each level.
+        num_res_channels: number of channels in the residual layers at each level.
+        downsample_parameters: A Tuple of Tuples for defining the downsampling convolutions. Each Tuple should hold the
+            following information stride (int), kernel_size (int), dilation (int) and padding (int).
+        dropout: dropout ratio.
+        act: activation type and arguments.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        channels: Sequence[int],
+        num_res_layers: int,
+        num_res_channels: Sequence[int],
+        downsample_parameters: Sequence[Tuple[int, int, int, int]],
+        dropout: float,
+        act: tuple | str | None,
+    ) -> None:
+        super().__init__()
+        self.spatial_dims = spatial_dims
+        self.in_channels = in_channels
+        self.out_channels = out_channels
+        self.channels = channels
+        self.num_res_layers = num_res_layers
+        self.num_res_channels = num_res_channels
+        self.downsample_parameters = downsample_parameters
+        self.dropout = dropout
+        self.act = act
+
+        blocks: list[nn.Module] = []
+
+        for i in range(len(self.channels)):
+            blocks.append(
+                Convolution(
+                    spatial_dims=self.spatial_dims,
+                    in_channels=self.in_channels if i == 0 else self.channels[i - 1],
+                    out_channels=self.channels[i],
+                    strides=self.downsample_parameters[i][0],
+                    kernel_size=self.downsample_parameters[i][1],
+                    adn_ordering="DA",
+                    act=self.act,
+                    dropout=None if i == 0 else self.dropout,
+                    dropout_dim=1,
+                    dilation=self.downsample_parameters[i][2],
+                    padding=self.downsample_parameters[i][3],
+                )
+            )
+
+            for _ in range(self.num_res_layers):
+                blocks.append(
+                    VQVAEResidualUnit(
+                        spatial_dims=self.spatial_dims,
+                        in_channels=self.channels[i],
+                        num_res_channels=self.num_res_channels[i],
+                        act=self.act,
+                        dropout=self.dropout,
+                    )
+                )
+
+        blocks.append(
+            Convolution(
+                spatial_dims=self.spatial_dims,
+                in_channels=self.channels[len(self.channels) - 1],
+                out_channels=self.out_channels,
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+
+        self.blocks = nn.ModuleList(blocks)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        for block in self.blocks:
+            x = block(x)
+        return x
+
+
+class Decoder(nn.Module):
+    """
+    Decoder module for VQ-VAE.
+
+    Args:
+        spatial_dims: number of spatial spatial_dims.
+        in_channels: number of channels in the latent space (embedding_dim).
+        out_channels: number of output channels.
+        channels: sequence containing the number of channels at each level of the decoder.
+        num_res_layers: number of sequential residual layers at each level.
+        num_res_channels: number of channels in the residual layers at each level.
+        upsample_parameters: A Tuple of Tuples for defining the upsampling convolutions. Each Tuple should hold the
+            following information stride (int), kernel_size (int), dilation (int), padding (int), output_padding (int).
+        dropout: dropout ratio.
+        act: activation type and arguments.
+        output_act: activation type and arguments for the output.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        channels: Sequence[int],
+        num_res_layers: int,
+        num_res_channels: Sequence[int],
+        upsample_parameters: Sequence[Tuple[int, int, int, int, int]],
+        dropout: float,
+        act: tuple | str | None,
+        output_act: tuple | str | None,
+    ) -> None:
+        super().__init__()
+        self.spatial_dims = spatial_dims
+        self.in_channels = in_channels
+        self.out_channels = out_channels
+        self.channels = channels
+        self.num_res_layers = num_res_layers
+        self.num_res_channels = num_res_channels
+        self.upsample_parameters = upsample_parameters
+        self.dropout = dropout
+        self.act = act
+        self.output_act = output_act
+
+        reversed_num_channels = list(reversed(self.channels))
+
+        blocks: list[nn.Module] = []
+        blocks.append(
+            Convolution(
+                spatial_dims=self.spatial_dims,
+                in_channels=self.in_channels,
+                out_channels=reversed_num_channels[0],
+                strides=1,
+                kernel_size=3,
+                padding=1,
+                conv_only=True,
+            )
+        )
+
+        reversed_num_res_channels = list(reversed(self.num_res_channels))
+        for i in range(len(self.channels)):
+            for _ in range(self.num_res_layers):
+                blocks.append(
+                    VQVAEResidualUnit(
+                        spatial_dims=self.spatial_dims,
+                        in_channels=reversed_num_channels[i],
+                        num_res_channels=reversed_num_res_channels[i],
+                        act=self.act,
+                        dropout=self.dropout,
+                    )
+                )
+
+            blocks.append(
+                Convolution(
+                    spatial_dims=self.spatial_dims,
+                    in_channels=reversed_num_channels[i],
+                    out_channels=self.out_channels if i == len(self.channels) - 1 else reversed_num_channels[i + 1],
+                    strides=self.upsample_parameters[i][0],
+                    kernel_size=self.upsample_parameters[i][1],
+                    adn_ordering="DA",
+                    act=self.act,
+                    dropout=self.dropout if i != len(self.channels) - 1 else None,
+                    norm=None,
+                    dilation=self.upsample_parameters[i][2],
+                    conv_only=i == len(self.channels) - 1,
+                    is_transposed=True,
+                    padding=self.upsample_parameters[i][3],
+                    output_padding=self.upsample_parameters[i][4],
+                )
+            )
+
+        if self.output_act:
+            blocks.append(Act[self.output_act]())
+
+        self.blocks = nn.ModuleList(blocks)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        for block in self.blocks:
+            x = block(x)
+        return x
+
+
+class VQVAE(nn.Module):
+    """
+    Vector-Quantised Variational Autoencoder (VQ-VAE) used in Morphology-preserving Autoregressive 3D Generative
+    Modelling of the Brain by Tudosiu et al. (https://arxiv.org/pdf/2209.03177.pdf)
+
+    The original implementation can be found at
+    https://github.com/AmigoLab/SynthAnatomy/blob/main/src/networks/vqvae/baseline.py#L163/
+
+    Args:
+        spatial_dims: number of spatial spatial_dims.
+        in_channels: number of input channels.
+        out_channels: number of output channels.
+        downsample_parameters: A Tuple of Tuples for defining the downsampling convolutions. Each Tuple should hold the
+            following information stride (int), kernel_size (int), dilation (int) and padding (int).
+        upsample_parameters: A Tuple of Tuples for defining the upsampling convolutions. Each Tuple should hold the
+            following information stride (int), kernel_size (int), dilation (int), padding (int), output_padding (int).
+        num_res_layers: number of sequential residual layers at each level.
+        channels: number of channels at each level.
+        num_res_channels: number of channels in the residual layers at each level.
+        num_embeddings: VectorQuantization number of atomic elements in the codebook.
+        embedding_dim: VectorQuantization number of channels of the input and atomic elements.
+        commitment_cost: VectorQuantization commitment_cost.
+        decay: VectorQuantization decay.
+        epsilon: VectorQuantization epsilon.
+        act: activation type and arguments.
+        dropout: dropout ratio.
+        output_act: activation type and arguments for the output.
+        ddp_sync: whether to synchronize the codebook across processes.
+        use_checkpointing if True, use activation checkpointing to save memory.
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int,
+        in_channels: int,
+        out_channels: int,
+        channels: Sequence[int] = (96, 96, 192),
+        num_res_layers: int = 3,
+        num_res_channels: Sequence[int] | int = (96, 96, 192),
+        downsample_parameters: Sequence[Tuple[int, int, int, int]] | Tuple[int, int, int, int] = (
+            (2, 4, 1, 1),
+            (2, 4, 1, 1),
+            (2, 4, 1, 1),
+        ),
+        upsample_parameters: Sequence[Tuple[int, int, int, int, int]] | Tuple[int, int, int, int, int] = (
+            (2, 4, 1, 1, 0),
+            (2, 4, 1, 1, 0),
+            (2, 4, 1, 1, 0),
+        ),
+        num_embeddings: int = 32,
+        embedding_dim: int = 64,
+        embedding_init: str = "normal",
+        commitment_cost: float = 0.25,
+        decay: float = 0.5,
+        epsilon: float = 1e-5,
+        dropout: float = 0.0,
+        act: tuple | str | None = Act.RELU,
+        output_act: tuple | str | None = None,
+        ddp_sync: bool = True,
+        use_checkpointing: bool = False,
+    ):
+        super().__init__()
+
+        self.in_channels = in_channels
+        self.out_channels = out_channels
+        self.spatial_dims = spatial_dims
+        self.channels = channels
+        self.num_embeddings = num_embeddings
+        self.embedding_dim = embedding_dim
+        self.use_checkpointing = use_checkpointing
+
+        if isinstance(num_res_channels, int):
+            num_res_channels = ensure_tuple_rep(num_res_channels, len(channels))
+
+        if len(num_res_channels) != len(channels):
+            raise ValueError(
+                "`num_res_channels` should be a single integer or a tuple of integers with the same length as "
+                "`num_channls`."
+            )
+        if all(isinstance(values, int) for values in upsample_parameters):
+            upsample_parameters_tuple: Sequence = (upsample_parameters,) * len(channels)
+        else:
+            upsample_parameters_tuple = upsample_parameters
+
+        if all(isinstance(values, int) for values in downsample_parameters):
+            downsample_parameters_tuple: Sequence = (downsample_parameters,) * len(channels)
+        else:
+            downsample_parameters_tuple = downsample_parameters
+
+        if not all(all(isinstance(value, int) for value in sub_item) for sub_item in downsample_parameters_tuple):
+            raise ValueError("`downsample_parameters` should be a single tuple of integer or a tuple of tuples.")
+
+        # check if downsample_parameters is a tuple of ints or a tuple of tuples of ints
+        if not all(all(isinstance(value, int) for value in sub_item) for sub_item in upsample_parameters_tuple):
+            raise ValueError("`upsample_parameters` should be a single tuple of integer or a tuple of tuples.")
+
+        for parameter in downsample_parameters_tuple:
+            if len(parameter) != 4:
+                raise ValueError("`downsample_parameters` should be a tuple of tuples with 4 integers.")
+
+        for parameter in upsample_parameters_tuple:
+            if len(parameter) != 5:
+                raise ValueError("`upsample_parameters` should be a tuple of tuples with 5 integers.")
+
+        if len(downsample_parameters_tuple) != len(channels):
+            raise ValueError(
+                "`downsample_parameters` should be a tuple of tuples with the same length as `num_channels`."
+            )
+
+        if len(upsample_parameters_tuple) != len(channels):
+            raise ValueError(
+                "`upsample_parameters` should be a tuple of tuples with the same length as `num_channels`."
+            )
+
+        self.num_res_layers = num_res_layers
+        self.num_res_channels = num_res_channels
+
+        self.encoder = Encoder(
+            spatial_dims=spatial_dims,
+            in_channels=in_channels,
+            out_channels=embedding_dim,
+            channels=channels,
+            num_res_layers=num_res_layers,
+            num_res_channels=num_res_channels,
+            downsample_parameters=downsample_parameters_tuple,
+            dropout=dropout,
+            act=act,
+        )
+
+        self.decoder = Decoder(
+            spatial_dims=spatial_dims,
+            in_channels=embedding_dim,
+            out_channels=out_channels,
+            channels=channels,
+            num_res_layers=num_res_layers,
+            num_res_channels=num_res_channels,
+            upsample_parameters=upsample_parameters_tuple,
+            dropout=dropout,
+            act=act,
+            output_act=output_act,
+        )
+
+        self.quantizer = VectorQuantizer(
+            quantizer=EMAQuantizer(
+                spatial_dims=spatial_dims,
+                num_embeddings=num_embeddings,
+                embedding_dim=embedding_dim,
+                commitment_cost=commitment_cost,
+                decay=decay,
+                epsilon=epsilon,
+                embedding_init=embedding_init,
+                ddp_sync=ddp_sync,
+            )
+        )
+
+    def encode(self, images: torch.Tensor) -> torch.Tensor:
+        output: torch.Tensor
+        if self.use_checkpointing:
+            output = torch.utils.checkpoint.checkpoint(self.encoder, images, use_reentrant=False)
+        else:
+            output = self.encoder(images)
+        return output
+
+    def quantize(self, encodings: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor]:
+        x_loss, x = self.quantizer(encodings)
+        return x, x_loss
+
+    def decode(self, quantizations: torch.Tensor) -> torch.Tensor:
+        output: torch.Tensor
+
+        if self.use_checkpointing:
+            output = torch.utils.checkpoint.checkpoint(self.decoder, quantizations, use_reentrant=False)
+        else:
+            output = self.decoder(quantizations)
+        return output
+
+    def index_quantize(self, images: torch.Tensor) -> torch.Tensor:
+        return self.quantizer.quantize(self.encode(images=images))
+
+    def decode_samples(self, embedding_indices: torch.Tensor) -> torch.Tensor:
+        return self.decode(self.quantizer.embed(embedding_indices))
+
+    def forward(self, images: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor]:
+        quantizations, quantization_losses = self.quantize(self.encode(images))
+        reconstruction = self.decode(quantizations)
+
+        return reconstruction, quantization_losses
+
+    def encode_stage_2_inputs(self, x: torch.Tensor) -> torch.Tensor:
+        z = self.encode(x)
+        e, _ = self.quantize(z)
+        return e
+
+    def decode_stage_2_outputs(self, z: torch.Tensor) -> torch.Tensor:
+        e, _ = self.quantize(z)
+        image = self.decode(e)
+        return image
diff --git a/monai/networks/schedulers/__init__.py b/monai/networks/schedulers/__init__.py
new file mode 100644
index 0000000000..29e9020d65
--- /dev/null
+++ b/monai/networks/schedulers/__init__.py
@@ -0,0 +1,17 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from .ddim import DDIMScheduler
+from .ddpm import DDPMScheduler
+from .pndm import PNDMScheduler
+from .scheduler import NoiseSchedules, Scheduler
diff --git a/monai/networks/schedulers/ddim.py b/monai/networks/schedulers/ddim.py
new file mode 100644
index 0000000000..2a0121d063
--- /dev/null
+++ b/monai/networks/schedulers/ddim.py
@@ -0,0 +1,294 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# =========================================================================
+# Adapted from https://github.com/huggingface/diffusers
+# which has the following license:
+# https://github.com/huggingface/diffusers/blob/main/LICENSE
+#
+# Copyright 2022 UC Berkeley Team and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =========================================================================
+
+from __future__ import annotations
+
+import numpy as np
+import torch
+
+from .ddpm import DDPMPredictionType
+from .scheduler import Scheduler
+
+DDIMPredictionType = DDPMPredictionType
+
+
+class DDIMScheduler(Scheduler):
+    """
+    Denoising diffusion implicit models is a scheduler that extends the denoising procedure introduced in denoising
+    diffusion probabilistic models (DDPMs) with non-Markovian guidance. Based on: Song et al. "Denoising Diffusion
+    Implicit Models" https://arxiv.org/abs/2010.02502
+
+    Args:
+        num_train_timesteps: number of diffusion steps used to train the model.
+        schedule: member of NoiseSchedules, name of noise schedule function in component store
+        clip_sample: option to clip predicted sample between -1 and 1 for numerical stability.
+        set_alpha_to_one: each diffusion step uses the value of alphas product at that step and at the previous one.
+            For the final step there is no previous alpha. When this option is `True` the previous alpha product is
+            fixed to `1`, otherwise it uses the value of alpha at step 0.
+        steps_offset: an offset added to the inference steps. You can use a combination of `steps_offset=1` and
+            `set_alpha_to_one=False`, to make the last step use step 0 for the previous alpha product, as done in
+            stable diffusion.
+        prediction_type: member of DDPMPredictionType
+        clip_sample_min: minimum clipping value when clip_sample equals True
+        clip_sample_max: maximum clipping value when clip_sample equals True
+        schedule_args: arguments to pass to the schedule function
+
+    """
+
+    def __init__(
+        self,
+        num_train_timesteps: int = 1000,
+        schedule: str = "linear_beta",
+        clip_sample: bool = True,
+        set_alpha_to_one: bool = True,
+        steps_offset: int = 0,
+        prediction_type: str = DDIMPredictionType.EPSILON,
+        clip_sample_min: float = -1.0,
+        clip_sample_max: float = 1.0,
+        **schedule_args,
+    ) -> None:
+        super().__init__(num_train_timesteps, schedule, **schedule_args)
+
+        if prediction_type not in DDIMPredictionType.__members__.values():
+            raise ValueError("Argument `prediction_type` must be a member of DDIMPredictionType")
+
+        self.prediction_type = prediction_type
+
+        # At every step in ddim, we are looking into the previous alphas_cumprod
+        # For the final step, there is no previous alphas_cumprod because we are already at 0
+        # `set_alpha_to_one` decides whether we set this parameter simply to one or
+        # whether we use the final alpha of the "non-previous" one.
+        self.final_alpha_cumprod = torch.tensor(1.0) if set_alpha_to_one else self.alphas_cumprod[0]
+
+        # standard deviation of the initial noise distribution
+        self.init_noise_sigma = 1.0
+
+        self.timesteps = torch.from_numpy(np.arange(0, self.num_train_timesteps)[::-1].astype(np.int64))
+
+        self.clip_sample = clip_sample
+        self.clip_sample_values = [clip_sample_min, clip_sample_max]
+        self.steps_offset = steps_offset
+
+        # default the number of inference timesteps to the number of train steps
+        self.num_inference_steps: int
+        self.set_timesteps(self.num_train_timesteps)
+
+    def set_timesteps(self, num_inference_steps: int, device: str | torch.device | None = None) -> None:
+        """
+        Sets the discrete timesteps used for the diffusion chain. Supporting function to be run before inference.
+
+        Args:
+            num_inference_steps: number of diffusion steps used when generating samples with a pre-trained model.
+            device: target device to put the data.
+        """
+        if num_inference_steps > self.num_train_timesteps:
+            raise ValueError(
+                f"`num_inference_steps`: {num_inference_steps} cannot be larger than `self.num_train_timesteps`:"
+                f" {self.num_train_timesteps} as the unet model trained with this scheduler can only handle"
+                f" maximal {self.num_train_timesteps} timesteps."
+            )
+
+        self.num_inference_steps = num_inference_steps
+        step_ratio = self.num_train_timesteps // self.num_inference_steps
+        if self.steps_offset >= step_ratio:
+            raise ValueError(
+                f"`steps_offset`: {self.steps_offset} cannot be greater than or equal to "
+                f"`num_train_timesteps // num_inference_steps : {step_ratio}` as this will cause timesteps to exceed"
+                f" the max train timestep."
+            )
+
+        # creates integer timesteps by multiplying by ratio
+        # casting to int to avoid issues when num_inference_step is power of 3
+        timesteps = (np.arange(0, num_inference_steps) * step_ratio).round()[::-1].copy().astype(np.int64)
+        self.timesteps = torch.from_numpy(timesteps).to(device)
+        self.timesteps += self.steps_offset
+
+    def _get_variance(self, timestep: int, prev_timestep: int) -> torch.Tensor:
+        alpha_prod_t = self.alphas_cumprod[timestep]
+        alpha_prod_t_prev = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
+        beta_prod_t = 1 - alpha_prod_t
+        beta_prod_t_prev = 1 - alpha_prod_t_prev
+
+        variance: torch.Tensor = (beta_prod_t_prev / beta_prod_t) * (1 - alpha_prod_t / alpha_prod_t_prev)
+
+        return variance
+
+    def step(
+        self,
+        model_output: torch.Tensor,
+        timestep: int,
+        sample: torch.Tensor,
+        eta: float = 0.0,
+        generator: torch.Generator | None = None,
+    ) -> tuple[torch.Tensor, torch.Tensor]:
+        """
+        Predict the sample at the previous timestep by reversing the SDE. Core function to propagate the diffusion
+        process from the learned model outputs (most often the predicted noise).
+
+        Args:
+            model_output: direct output from learned diffusion model.
+            timestep: current discrete timestep in the diffusion chain.
+            sample: current instance of sample being created by diffusion process.
+            eta: weight of noise for added noise in diffusion step.
+            generator: random number generator.
+
+        Returns:
+            pred_prev_sample: Predicted previous sample
+            pred_original_sample: Predicted original sample
+        """
+        # See formulas (12) and (16) of DDIM paper https://arxiv.org/pdf/2010.02502.pdf
+        # Ideally, read DDIM paper in-detail understanding
+
+        # Notation (<variable name> -> <name in paper>
+        # - model_output -> e_theta(x_t, t)
+        # - pred_original_sample -> f_theta(x_t, t) or x_0
+        # - std_dev_t -> sigma_t
+        # - eta -> η
+        # - pred_sample_direction -> "direction pointing to x_t"
+        # - pred_prev_sample -> "x_t-1"
+
+        # 1. get previous step value (=t-1)
+        prev_timestep = timestep - self.num_train_timesteps // self.num_inference_steps
+
+        # 2. compute alphas, betas
+        alpha_prod_t = self.alphas_cumprod[timestep]
+        alpha_prod_t_prev = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
+
+        beta_prod_t = 1 - alpha_prod_t
+
+        # predefinitions satisfy pylint/mypy, these values won't be ultimately used
+        pred_original_sample = sample
+        pred_epsilon = model_output
+
+        # 3. compute predicted original sample from predicted noise also called
+        # "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
+        if self.prediction_type == DDIMPredictionType.EPSILON:
+            pred_original_sample = (sample - (beta_prod_t**0.5) * model_output) / (alpha_prod_t**0.5)
+            pred_epsilon = model_output
+        elif self.prediction_type == DDIMPredictionType.SAMPLE:
+            pred_original_sample = model_output
+            pred_epsilon = (sample - (alpha_prod_t**0.5) * pred_original_sample) / (beta_prod_t**0.5)
+        elif self.prediction_type == DDIMPredictionType.V_PREDICTION:
+            pred_original_sample = (alpha_prod_t**0.5) * sample - (beta_prod_t**0.5) * model_output
+            pred_epsilon = (alpha_prod_t**0.5) * model_output + (beta_prod_t**0.5) * sample
+
+        # 4. Clip "predicted x_0"
+        if self.clip_sample:
+            pred_original_sample = torch.clamp(
+                pred_original_sample, self.clip_sample_values[0], self.clip_sample_values[1]
+            )
+
+        # 5. compute variance: "sigma_t(η)" -> see formula (16)
+        # σ_t = sqrt((1 − α_t−1)/(1 − α_t)) * sqrt(1 − α_t/α_t−1)
+        variance = self._get_variance(timestep, prev_timestep)
+        std_dev_t = eta * variance**0.5
+
+        # 6. compute "direction pointing to x_t" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
+        pred_sample_direction = (1 - alpha_prod_t_prev - std_dev_t**2) ** 0.5 * pred_epsilon
+
+        # 7. compute x_t-1 without "random noise" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
+        pred_prev_sample = alpha_prod_t_prev**0.5 * pred_original_sample + pred_sample_direction
+
+        if eta > 0:
+            # randn_like does not support generator https://github.com/pytorch/pytorch/issues/27072
+            device: torch.device = torch.device(model_output.device if torch.is_tensor(model_output) else "cpu")
+            noise = torch.randn(model_output.shape, dtype=model_output.dtype, generator=generator).to(device)
+            variance = self._get_variance(timestep, prev_timestep) ** 0.5 * eta * noise
+
+            pred_prev_sample = pred_prev_sample + variance
+
+        return pred_prev_sample, pred_original_sample
+
+    def reversed_step(
+        self, model_output: torch.Tensor, timestep: int, sample: torch.Tensor
+    ) -> tuple[torch.Tensor, torch.Tensor]:
+        """
+        Predict the sample at the next timestep by reversing the SDE. Core function to propagate the diffusion
+        process from the learned model outputs (most often the predicted noise).
+
+        Args:
+            model_output: direct output from learned diffusion model.
+            timestep: current discrete timestep in the diffusion chain.
+            sample: current instance of sample being created by diffusion process.
+
+        Returns:
+            pred_prev_sample: Predicted previous sample
+            pred_original_sample: Predicted original sample
+        """
+        # See Appendix F at https://arxiv.org/pdf/2105.05233.pdf, or Equation (6) in https://arxiv.org/pdf/2203.04306.pdf
+
+        # Notation (<variable name> -> <name in paper>
+        # - model_output -> e_theta(x_t, t)
+        # - pred_original_sample -> f_theta(x_t, t) or x_0
+        # - std_dev_t -> sigma_t
+        # - eta -> η
+        # - pred_sample_direction -> "direction pointing to x_t"
+        # - pred_post_sample -> "x_t+1"
+
+        # 1. get previous step value (=t+1)
+        prev_timestep = timestep + self.num_train_timesteps // self.num_inference_steps
+
+        # 2. compute alphas, betas at timestep t+1
+        alpha_prod_t = self.alphas_cumprod[timestep]
+        alpha_prod_t_prev = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
+
+        beta_prod_t = 1 - alpha_prod_t
+
+        # predefinitions satisfy pylint/mypy, these values won't be ultimately used
+        pred_original_sample = sample
+        pred_epsilon = model_output
+
+        # 3. compute predicted original sample from predicted noise also called
+        # "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
+
+        if self.prediction_type == DDIMPredictionType.EPSILON:
+            pred_original_sample = (sample - beta_prod_t ** (0.5) * model_output) / alpha_prod_t ** (0.5)
+            pred_epsilon = model_output
+        elif self.prediction_type == DDIMPredictionType.SAMPLE:
+            pred_original_sample = model_output
+            pred_epsilon = (sample - alpha_prod_t ** (0.5) * pred_original_sample) / beta_prod_t ** (0.5)
+        elif self.prediction_type == DDIMPredictionType.V_PREDICTION:
+            pred_original_sample = (alpha_prod_t**0.5) * sample - (beta_prod_t**0.5) * model_output
+            pred_epsilon = (alpha_prod_t**0.5) * model_output + (beta_prod_t**0.5) * sample
+
+        # 4. Clip "predicted x_0"
+        if self.clip_sample:
+            pred_original_sample = torch.clamp(
+                pred_original_sample, self.clip_sample_values[0], self.clip_sample_values[1]
+            )
+
+        # 5. compute "direction pointing to x_t" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
+        pred_sample_direction = (1 - alpha_prod_t_prev) ** (0.5) * pred_epsilon
+
+        # 6. compute x_t+1 without "random noise" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
+        pred_post_sample = alpha_prod_t_prev ** (0.5) * pred_original_sample + pred_sample_direction
+
+        return pred_post_sample, pred_original_sample
diff --git a/monai/networks/schedulers/ddpm.py b/monai/networks/schedulers/ddpm.py
new file mode 100644
index 0000000000..93ad833031
--- /dev/null
+++ b/monai/networks/schedulers/ddpm.py
@@ -0,0 +1,250 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# =========================================================================
+# Adapted from https://github.com/huggingface/diffusers
+# which has the following license:
+# https://github.com/huggingface/diffusers/blob/main/LICENSE
+#
+# Copyright 2022 UC Berkeley Team and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =========================================================================
+
+from __future__ import annotations
+
+import numpy as np
+import torch
+
+from monai.utils import StrEnum
+
+from .scheduler import Scheduler
+
+
+class DDPMVarianceType(StrEnum):
+    """
+    Valid names for DDPM Scheduler's `variance_type` argument. Options to clip the variance used when adding noise
+    to the denoised sample.
+    """
+
+    FIXED_SMALL = "fixed_small"
+    FIXED_LARGE = "fixed_large"
+    LEARNED = "learned"
+    LEARNED_RANGE = "learned_range"
+
+
+class DDPMPredictionType(StrEnum):
+    """
+    Set of valid prediction type names for the DDPM scheduler's `prediction_type` argument.
+
+    epsilon: predicting the noise of the diffusion process
+    sample: directly predicting the noisy sample
+    v_prediction: velocity prediction, see section 2.4 https://imagen.research.google/video/paper.pdf
+    """
+
+    EPSILON = "epsilon"
+    SAMPLE = "sample"
+    V_PREDICTION = "v_prediction"
+
+
+class DDPMScheduler(Scheduler):
+    """
+    Denoising diffusion probabilistic models (DDPMs) explores the connections between denoising score matching and
+    Langevin dynamics sampling. Based on: Ho et al., "Denoising Diffusion Probabilistic Models"
+    https://arxiv.org/abs/2006.11239
+
+    Args:
+        num_train_timesteps: number of diffusion steps used to train the model.
+        schedule: member of NoiseSchedules, name of noise schedule function in component store
+        variance_type: member of DDPMVarianceType
+        clip_sample: option to clip predicted sample between -1 and 1 for numerical stability.
+        prediction_type: member of DDPMPredictionType
+        clip_sample_min: minimum clipping value when clip_sample equals True
+        clip_sample_max: maximum clipping value when clip_sample equals True
+        schedule_args: arguments to pass to the schedule function
+    """
+
+    def __init__(
+        self,
+        num_train_timesteps: int = 1000,
+        schedule: str = "linear_beta",
+        variance_type: str = DDPMVarianceType.FIXED_SMALL,
+        clip_sample: bool = True,
+        prediction_type: str = DDPMPredictionType.EPSILON,
+        clip_sample_min: float = -1.0,
+        clip_sample_max: float = 1.0,
+        **schedule_args,
+    ) -> None:
+        super().__init__(num_train_timesteps, schedule, **schedule_args)
+
+        if variance_type not in DDPMVarianceType.__members__.values():
+            raise ValueError("Argument `variance_type` must be a member of `DDPMVarianceType`")
+
+        if prediction_type not in DDPMPredictionType.__members__.values():
+            raise ValueError("Argument `prediction_type` must be a member of `DDPMPredictionType`")
+
+        self.clip_sample = clip_sample
+        self.clip_sample_values = [clip_sample_min, clip_sample_max]
+        self.variance_type = variance_type
+        self.prediction_type = prediction_type
+
+    def set_timesteps(self, num_inference_steps: int, device: str | torch.device | None = None) -> None:
+        """
+        Sets the discrete timesteps used for the diffusion chain. Supporting function to be run before inference.
+
+        Args:
+            num_inference_steps: number of diffusion steps used when generating samples with a pre-trained model.
+            device: target device to put the data.
+        """
+        if num_inference_steps > self.num_train_timesteps:
+            raise ValueError(
+                f"`num_inference_steps`: {num_inference_steps} cannot be larger than `self.num_train_timesteps`:"
+                f" {self.num_train_timesteps} as the unet model trained with this scheduler can only handle"
+                f" maximal {self.num_train_timesteps} timesteps."
+            )
+
+        self.num_inference_steps = num_inference_steps
+        step_ratio = self.num_train_timesteps // self.num_inference_steps
+        # creates integer timesteps by multiplying by ratio
+        # casting to int to avoid issues when num_inference_step is power of 3
+        timesteps = (np.arange(0, num_inference_steps) * step_ratio).round()[::-1].astype(np.int64)
+        self.timesteps = torch.from_numpy(timesteps).to(device)
+
+    def _get_mean(self, timestep: int, x_0: torch.Tensor, x_t: torch.Tensor) -> torch.Tensor:
+        """
+        Compute the mean of the posterior at timestep t.
+
+        Args:
+            timestep: current timestep.
+            x0: the noise-free input.
+            x_t: the input noised to timestep t.
+
+        Returns:
+            Returns the mean
+        """
+        # these attributes are used for calculating the posterior, q(x_{t-1}|x_t,x_0),
+        # (see formula (5-7) from https://arxiv.org/pdf/2006.11239.pdf)
+        alpha_t = self.alphas[timestep]
+        alpha_prod_t = self.alphas_cumprod[timestep]
+        alpha_prod_t_prev = self.alphas_cumprod[timestep - 1] if timestep > 0 else self.one
+
+        x_0_coefficient = alpha_prod_t_prev.sqrt() * self.betas[timestep] / (1 - alpha_prod_t)
+        x_t_coefficient = alpha_t.sqrt() * (1 - alpha_prod_t_prev) / (1 - alpha_prod_t)
+
+        mean: torch.Tensor = x_0_coefficient * x_0 + x_t_coefficient * x_t
+
+        return mean
+
+    def _get_variance(self, timestep: int, predicted_variance: torch.Tensor | None = None) -> torch.Tensor:
+        """
+        Compute the variance of the posterior at timestep t.
+
+        Args:
+            timestep: current timestep.
+            predicted_variance: variance predicted by the model.
+
+        Returns:
+            Returns the variance
+        """
+        alpha_prod_t = self.alphas_cumprod[timestep]
+        alpha_prod_t_prev = self.alphas_cumprod[timestep - 1] if timestep > 0 else self.one
+
+        # For t > 0, compute predicted variance βt (see formula (6) and (7) from https://arxiv.org/pdf/2006.11239.pdf)
+        # and sample from it to get previous sample
+        # x_{t-1} ~ N(pred_prev_sample, variance) == add variance to pred_sample
+        variance: torch.Tensor = (1 - alpha_prod_t_prev) / (1 - alpha_prod_t) * self.betas[timestep]
+        # hacks - were probably added for training stability
+        if self.variance_type == DDPMVarianceType.FIXED_SMALL:
+            variance = torch.clamp(variance, min=1e-20)
+        elif self.variance_type == DDPMVarianceType.FIXED_LARGE:
+            variance = self.betas[timestep]
+        elif self.variance_type == DDPMVarianceType.LEARNED and predicted_variance is not None:
+            return predicted_variance
+        elif self.variance_type == DDPMVarianceType.LEARNED_RANGE and predicted_variance is not None:
+            min_log = variance
+            max_log = self.betas[timestep]
+            frac = (predicted_variance + 1) / 2
+            variance = frac * max_log + (1 - frac) * min_log
+
+        return variance
+
+    def step(
+        self, model_output: torch.Tensor, timestep: int, sample: torch.Tensor, generator: torch.Generator | None = None
+    ) -> tuple[torch.Tensor, torch.Tensor]:
+        """
+        Predict the sample at the previous timestep by reversing the SDE. Core function to propagate the diffusion
+        process from the learned model outputs (most often the predicted noise).
+
+        Args:
+            model_output: direct output from learned diffusion model.
+            timestep: current discrete timestep in the diffusion chain.
+            sample: current instance of sample being created by diffusion process.
+            generator: random number generator.
+
+        Returns:
+            pred_prev_sample: Predicted previous sample
+        """
+        if model_output.shape[1] == sample.shape[1] * 2 and self.variance_type in ["learned", "learned_range"]:
+            model_output, predicted_variance = torch.split(model_output, sample.shape[1], dim=1)
+        else:
+            predicted_variance = None
+
+        # 1. compute alphas, betas
+        alpha_prod_t = self.alphas_cumprod[timestep]
+        alpha_prod_t_prev = self.alphas_cumprod[timestep - 1] if timestep > 0 else self.one
+        beta_prod_t = 1 - alpha_prod_t
+        beta_prod_t_prev = 1 - alpha_prod_t_prev
+
+        # 2. compute predicted original sample from predicted noise also called
+        # "predicted x_0" of formula (15) from https://arxiv.org/pdf/2006.11239.pdf
+        if self.prediction_type == DDPMPredictionType.EPSILON:
+            pred_original_sample = (sample - beta_prod_t ** (0.5) * model_output) / alpha_prod_t ** (0.5)
+        elif self.prediction_type == DDPMPredictionType.SAMPLE:
+            pred_original_sample = model_output
+        elif self.prediction_type == DDPMPredictionType.V_PREDICTION:
+            pred_original_sample = (alpha_prod_t**0.5) * sample - (beta_prod_t**0.5) * model_output
+
+        # 3. Clip "predicted x_0"
+        if self.clip_sample:
+            pred_original_sample = torch.clamp(
+                pred_original_sample, self.clip_sample_values[0], self.clip_sample_values[1]
+            )
+
+        # 4. Compute coefficients for pred_original_sample x_0 and current sample x_t
+        # See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
+        pred_original_sample_coeff = (alpha_prod_t_prev ** (0.5) * self.betas[timestep]) / beta_prod_t
+        current_sample_coeff = self.alphas[timestep] ** (0.5) * beta_prod_t_prev / beta_prod_t
+
+        # 5. Compute predicted previous sample µ_t
+        # See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
+        pred_prev_sample = pred_original_sample_coeff * pred_original_sample + current_sample_coeff * sample
+
+        # 6. Add noise
+        variance = 0
+        if timestep > 0:
+            noise = torch.randn(
+                model_output.size(), dtype=model_output.dtype, layout=model_output.layout, generator=generator
+            ).to(model_output.device)
+            variance = (self._get_variance(timestep, predicted_variance=predicted_variance) ** 0.5) * noise
+
+        pred_prev_sample = pred_prev_sample + variance
+
+        return pred_prev_sample, pred_original_sample
diff --git a/monai/networks/schedulers/pndm.py b/monai/networks/schedulers/pndm.py
new file mode 100644
index 0000000000..c0728bbdff
--- /dev/null
+++ b/monai/networks/schedulers/pndm.py
@@ -0,0 +1,316 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# =========================================================================
+# Adapted from https://github.com/huggingface/diffusers
+# which has the following license:
+# https://github.com/huggingface/diffusers/blob/main/LICENSE
+#
+# Copyright 2022 UC Berkeley Team and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =========================================================================
+
+from __future__ import annotations
+
+from typing import Any
+
+import numpy as np
+import torch
+
+from monai.utils import StrEnum
+
+from .scheduler import Scheduler
+
+
+class PNDMPredictionType(StrEnum):
+    """
+    Set of valid prediction type names for the PNDM scheduler's `prediction_type` argument.
+
+    epsilon: predicting the noise of the diffusion process
+    v_prediction: velocity prediction, see section 2.4 https://imagen.research.google/video/paper.pdf
+    """
+
+    EPSILON = "epsilon"
+    V_PREDICTION = "v_prediction"
+
+
+class PNDMScheduler(Scheduler):
+    """
+    Pseudo numerical methods for diffusion models (PNDM) proposes using more advanced ODE integration techniques,
+    namely Runge-Kutta method and a linear multi-step method. Based on: Liu et al.,
+    "Pseudo Numerical Methods for Diffusion Models on Manifolds"  https://arxiv.org/abs/2202.09778
+
+    Args:
+        num_train_timesteps: number of diffusion steps used to train the model.
+        schedule: member of NoiseSchedules, name of noise schedule function in component store
+        skip_prk_steps:
+            allows the scheduler to skip the Runge-Kutta steps that are defined in the original paper as being required
+            before plms step.
+        set_alpha_to_one:
+            each diffusion step uses the value of alphas product at that step and at the previous one. For the final
+            step there is no previous alpha. When this option is `True` the previous alpha product is fixed to `1`,
+            otherwise it uses the value of alpha at step 0.
+        prediction_type: member of DDPMPredictionType
+        steps_offset:
+            an offset added to the inference steps. You can use a combination of `offset=1` and
+            `set_alpha_to_one=False`, to make the last step use step 0 for the previous alpha product, as done in
+            stable diffusion.
+        schedule_args: arguments to pass to the schedule function
+    """
+
+    def __init__(
+        self,
+        num_train_timesteps: int = 1000,
+        schedule: str = "linear_beta",
+        skip_prk_steps: bool = False,
+        set_alpha_to_one: bool = False,
+        prediction_type: str = PNDMPredictionType.EPSILON,
+        steps_offset: int = 0,
+        **schedule_args,
+    ) -> None:
+        super().__init__(num_train_timesteps, schedule, **schedule_args)
+
+        if prediction_type not in PNDMPredictionType.__members__.values():
+            raise ValueError("Argument `prediction_type` must be a member of PNDMPredictionType")
+
+        self.prediction_type = prediction_type
+
+        self.final_alpha_cumprod = torch.tensor(1.0) if set_alpha_to_one else self.alphas_cumprod[0]
+
+        # standard deviation of the initial noise distribution
+        self.init_noise_sigma = 1.0
+
+        # For now we only support F-PNDM, i.e. the runge-kutta method
+        # For more information on the algorithm please take a look at the paper: https://arxiv.org/pdf/2202.09778.pdf
+        # mainly at formula (9), (12), (13) and the Algorithm 2.
+        self.pndm_order = 4
+
+        self.skip_prk_steps = skip_prk_steps
+        self.steps_offset = steps_offset
+
+        # running values
+        self.cur_model_output = torch.Tensor()
+        self.counter = 0
+        self.cur_sample = torch.Tensor()
+        self.ets: list = []
+
+        # default the number of inference timesteps to the number of train steps
+        self.set_timesteps(num_train_timesteps)
+
+    def set_timesteps(self, num_inference_steps: int, device: str | torch.device | None = None) -> None:
+        """
+        Sets the discrete timesteps used for the diffusion chain. Supporting function to be run before inference.
+
+        Args:
+            num_inference_steps: number of diffusion steps used when generating samples with a pre-trained model.
+            device: target device to put the data.
+        """
+        if num_inference_steps > self.num_train_timesteps:
+            raise ValueError(
+                f"`num_inference_steps`: {num_inference_steps} cannot be larger than `self.num_train_timesteps`:"
+                f" {self.num_train_timesteps} as the unet model trained with this scheduler can only handle"
+                f" maximal {self.num_train_timesteps} timesteps."
+            )
+
+        self.num_inference_steps = num_inference_steps
+        step_ratio = self.num_train_timesteps // self.num_inference_steps
+        # creates integer timesteps by multiplying by ratio
+        # casting to int to avoid issues when num_inference_step is power of 3
+        self._timesteps = (np.arange(0, num_inference_steps) * step_ratio).round().astype(np.int64)
+        self._timesteps += self.steps_offset
+
+        if self.skip_prk_steps:
+            # for some models like stable diffusion the prk steps can/should be skipped to
+            # produce better results. When using PNDM with `self.skip_prk_steps` the implementation
+            # is based on crowsonkb's PLMS sampler implementation: https://github.com/CompVis/latent-diffusion/pull/51
+            self.prk_timesteps = np.array([])
+            self.plms_timesteps = self._timesteps[::-1]
+
+        else:
+            prk_timesteps = np.array(self._timesteps[-self.pndm_order :]).repeat(2) + np.tile(
+                np.array([0, self.num_train_timesteps // num_inference_steps // 2]), self.pndm_order
+            )
+            self.prk_timesteps = (prk_timesteps[:-1].repeat(2)[1:-1])[::-1].copy()
+            self.plms_timesteps = self._timesteps[:-3][
+                ::-1
+            ].copy()  # we copy to avoid having negative strides which are not supported by torch.from_numpy
+
+        timesteps = np.concatenate([self.prk_timesteps, self.plms_timesteps]).astype(np.int64)
+        self.timesteps = torch.from_numpy(timesteps).to(device)
+        # update num_inference_steps - necessary if we use prk steps
+        self.num_inference_steps = len(self.timesteps)
+
+        self.ets = []
+        self.counter = 0
+
+    def step(self, model_output: torch.Tensor, timestep: int, sample: torch.Tensor) -> tuple[torch.Tensor, Any]:
+        """
+        Predict the sample at the previous timestep by reversing the SDE. Core function to propagate the diffusion
+        process from the learned model outputs (most often the predicted noise).
+        This function calls `step_prk()` or `step_plms()` depending on the internal variable `counter`.
+
+        Args:
+            model_output: direct output from learned diffusion model.
+            timestep: current discrete timestep in the diffusion chain.
+            sample: current instance of sample being created by diffusion process.
+        Returns:
+            pred_prev_sample: Predicted previous sample
+        """
+        # return a tuple for consistency with samplers that return (previous pred, original sample pred)
+
+        if self.counter < len(self.prk_timesteps) and not self.skip_prk_steps:
+            return self.step_prk(model_output=model_output, timestep=timestep, sample=sample), None
+        else:
+            return self.step_plms(model_output=model_output, timestep=timestep, sample=sample), None
+
+    def step_prk(self, model_output: torch.Tensor, timestep: int, sample: torch.Tensor) -> torch.Tensor:
+        """
+        Step function propagating the sample with the Runge-Kutta method. RK takes 4 forward passes to approximate the
+        solution to the differential equation.
+
+        Args:
+            model_output: direct output from learned diffusion model.
+            timestep: current discrete timestep in the diffusion chain.
+            sample: current instance of sample being created by diffusion process.
+
+        Returns:
+            pred_prev_sample: Predicted previous sample
+        """
+        if self.num_inference_steps is None:
+            raise ValueError(
+                "Number of inference steps is 'None', you need to run 'set_timesteps' after creating the scheduler"
+            )
+
+        diff_to_prev = 0 if self.counter % 2 else self.num_train_timesteps // self.num_inference_steps // 2
+        prev_timestep = timestep - diff_to_prev
+        timestep = self.prk_timesteps[self.counter // 4 * 4]
+
+        if self.counter % 4 == 0:
+            self.cur_model_output = 1 / 6 * model_output
+            self.ets.append(model_output)
+            self.cur_sample = sample
+        elif (self.counter - 1) % 4 == 0:
+            self.cur_model_output += 1 / 3 * model_output
+        elif (self.counter - 2) % 4 == 0:
+            self.cur_model_output += 1 / 3 * model_output
+        elif (self.counter - 3) % 4 == 0:
+            model_output = self.cur_model_output + 1 / 6 * model_output
+            self.cur_model_output = torch.Tensor()
+
+        # cur_sample should not be an empty torch.Tensor()
+        cur_sample = self.cur_sample if self.cur_sample.numel() != 0 else sample
+
+        prev_sample: torch.Tensor = self._get_prev_sample(cur_sample, timestep, prev_timestep, model_output)
+        self.counter += 1
+
+        return prev_sample
+
+    def step_plms(self, model_output: torch.Tensor, timestep: int, sample: torch.Tensor) -> Any:
+        """
+        Step function propagating the sample with the linear multi-step method. This has one forward pass with multiple
+        times to approximate the solution.
+
+        Args:
+            model_output: direct output from learned diffusion model.
+            timestep: current discrete timestep in the diffusion chain.
+            sample: current instance of sample being created by diffusion process.
+
+        Returns:
+            pred_prev_sample: Predicted previous sample
+        """
+        if self.num_inference_steps is None:
+            raise ValueError(
+                "Number of inference steps is 'None', you need to run 'set_timesteps' after creating the scheduler"
+            )
+
+        if not self.skip_prk_steps and len(self.ets) < 3:
+            raise ValueError(
+                f"{self.__class__} can only be run AFTER scheduler has been run "
+                "in 'prk' mode for at least 12 iterations "
+            )
+
+        prev_timestep = timestep - self.num_train_timesteps // self.num_inference_steps
+
+        if self.counter != 1:
+            self.ets = self.ets[-3:]
+            self.ets.append(model_output)
+        else:
+            prev_timestep = timestep
+            timestep = timestep + self.num_train_timesteps // self.num_inference_steps
+
+        if len(self.ets) == 1 and self.counter == 0:
+            model_output = model_output
+            self.cur_sample = sample
+        elif len(self.ets) == 1 and self.counter == 1:
+            model_output = (model_output + self.ets[-1]) / 2
+            sample = self.cur_sample
+            self.cur_sample = torch.Tensor()
+        elif len(self.ets) == 2:
+            model_output = (3 * self.ets[-1] - self.ets[-2]) / 2
+        elif len(self.ets) == 3:
+            model_output = (23 * self.ets[-1] - 16 * self.ets[-2] + 5 * self.ets[-3]) / 12
+        else:
+            model_output = (1 / 24) * (55 * self.ets[-1] - 59 * self.ets[-2] + 37 * self.ets[-3] - 9 * self.ets[-4])
+
+        prev_sample = self._get_prev_sample(sample, timestep, prev_timestep, model_output)
+        self.counter += 1
+
+        return prev_sample
+
+    def _get_prev_sample(self, sample: torch.Tensor, timestep: int, prev_timestep: int, model_output: torch.Tensor):
+        # See formula (9) of PNDM paper https://arxiv.org/pdf/2202.09778.pdf
+        # this function computes x_(t−δ) using the formula of (9)
+        # Note that x_t needs to be added to both sides of the equation
+
+        # Notation (<variable name> -> <name in paper>
+        # alpha_prod_t -> α_t
+        # alpha_prod_t_prev -> α_(t−δ)
+        # beta_prod_t -> (1 - α_t)
+        # beta_prod_t_prev -> (1 - α_(t−δ))
+        # sample -> x_t
+        # model_output -> e_θ(x_t, t)
+        # prev_sample -> x_(t−δ)
+        alpha_prod_t = self.alphas_cumprod[timestep]
+        alpha_prod_t_prev = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
+        beta_prod_t = 1 - alpha_prod_t
+        beta_prod_t_prev = 1 - alpha_prod_t_prev
+
+        if self.prediction_type == PNDMPredictionType.V_PREDICTION:
+            model_output = (alpha_prod_t**0.5) * model_output + (beta_prod_t**0.5) * sample
+
+        # corresponds to (α_(t−δ) - α_t) divided by
+        # denominator of x_t in formula (9) and plus 1
+        # Note: (α_(t−δ) - α_t) / (sqrt(α_t) * (sqrt(α_(t−δ)) + sqr(α_t))) =
+        # sqrt(α_(t−δ)) / sqrt(α_t))
+        sample_coeff = (alpha_prod_t_prev / alpha_prod_t) ** (0.5)
+
+        # corresponds to denominator of e_θ(x_t, t) in formula (9)
+        model_output_denom_coeff = alpha_prod_t * beta_prod_t_prev ** (0.5) + (
+            alpha_prod_t * beta_prod_t * alpha_prod_t_prev
+        ) ** (0.5)
+
+        # full formula (9)
+        prev_sample = (
+            sample_coeff * sample - (alpha_prod_t_prev - alpha_prod_t) * model_output / model_output_denom_coeff
+        )
+
+        return prev_sample
diff --git a/monai/networks/schedulers/scheduler.py b/monai/networks/schedulers/scheduler.py
new file mode 100644
index 0000000000..acdccc60de
--- /dev/null
+++ b/monai/networks/schedulers/scheduler.py
@@ -0,0 +1,205 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# =========================================================================
+# Adapted from https://github.com/huggingface/diffusers
+# which has the following license:
+# https://github.com/huggingface/diffusers/blob/main/LICENSE
+#
+# Copyright 2022 UC Berkeley Team and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =========================================================================
+
+
+from __future__ import annotations
+
+import torch
+import torch.nn as nn
+
+from monai.utils import ComponentStore, unsqueeze_right
+
+NoiseSchedules = ComponentStore("NoiseSchedules", "Functions to generate noise schedules")
+
+
+@NoiseSchedules.add_def("linear_beta", "Linear beta schedule")
+def _linear_beta(num_train_timesteps: int, beta_start: float = 1e-4, beta_end: float = 2e-2):
+    """
+    Linear beta noise schedule function.
+
+    Args:
+        num_train_timesteps: number of timesteps
+        beta_start: start of beta range, default 1e-4
+        beta_end: end of beta range, default 2e-2
+
+    Returns:
+        betas: beta schedule tensor
+    """
+    return torch.linspace(beta_start, beta_end, num_train_timesteps, dtype=torch.float32)
+
+
+@NoiseSchedules.add_def("scaled_linear_beta", "Scaled linear beta schedule")
+def _scaled_linear_beta(num_train_timesteps: int, beta_start: float = 1e-4, beta_end: float = 2e-2):
+    """
+    Scaled linear beta noise schedule function.
+
+    Args:
+        num_train_timesteps: number of timesteps
+        beta_start: start of beta range, default 1e-4
+        beta_end: end of beta range, default 2e-2
+
+    Returns:
+        betas: beta schedule tensor
+    """
+    return torch.linspace(beta_start**0.5, beta_end**0.5, num_train_timesteps, dtype=torch.float32) ** 2
+
+
+@NoiseSchedules.add_def("sigmoid_beta", "Sigmoid beta schedule")
+def _sigmoid_beta(num_train_timesteps: int, beta_start: float = 1e-4, beta_end: float = 2e-2, sig_range: float = 6):
+    """
+    Sigmoid beta noise schedule function.
+
+    Args:
+        num_train_timesteps: number of timesteps
+        beta_start: start of beta range, default 1e-4
+        beta_end: end of beta range, default 2e-2
+        sig_range: pos/neg range of sigmoid input, default 6
+
+    Returns:
+        betas: beta schedule tensor
+    """
+    betas = torch.linspace(-sig_range, sig_range, num_train_timesteps)
+    return torch.sigmoid(betas) * (beta_end - beta_start) + beta_start
+
+
+@NoiseSchedules.add_def("cosine", "Cosine schedule")
+def _cosine_beta(num_train_timesteps: int, s: float = 8e-3):
+    """
+    Cosine noise schedule, see https://arxiv.org/abs/2102.09672
+
+    Args:
+        num_train_timesteps: number of timesteps
+        s: smoothing factor, default 8e-3 (see referenced paper)
+
+    Returns:
+        (betas, alphas, alpha_cumprod) values
+    """
+    x = torch.linspace(0, num_train_timesteps, num_train_timesteps + 1)
+    alphas_cumprod = torch.cos(((x / num_train_timesteps) + s) / (1 + s) * torch.pi * 0.5) ** 2
+    alphas_cumprod /= alphas_cumprod[0].item()
+    alphas = torch.clip(alphas_cumprod[1:] / alphas_cumprod[:-1], 0.0001, 0.9999)
+    betas = 1.0 - alphas
+    return betas, alphas, alphas_cumprod[:-1]
+
+
+class Scheduler(nn.Module):
+    """
+    Base class for other schedulers based on a noise schedule function.
+
+    This class is meant as the base for other schedulers which implement their own way of sampling or stepping. Here
+    the class defines beta, alpha, and alpha_cumprod values from a noise schedule function named with `schedule`,
+    which is the name of a component in NoiseSchedules. These components must all be callables which return either
+    the beta schedule alone or a triple containing (betas, alphas, alphas_cumprod) values. New schedule functions
+    can be provided by using the NoiseSchedules.add_def, for example:
+
+    .. code-block:: python
+
+        from monai.networks.schedulers import NoiseSchedules, DDPMScheduler
+
+        @NoiseSchedules.add_def("my_beta_schedule", "Some description of your function")
+        def _beta_function(num_train_timesteps, beta_start=1e-4, beta_end=2e-2):
+            return torch.linspace(beta_start, beta_end, num_train_timesteps, dtype=torch.float32)
+
+        scheduler = DDPMScheduler(num_train_timesteps=1000, schedule="my_beta_schedule")
+
+    All such functions should have an initial positional integer argument `num_train_timesteps` stating the number of
+    timesteps the schedule is for, otherwise any other arguments can be given which will be passed by keyword through
+    the constructor's `schedule_args` value. To see what noise functions are available, print the object NoiseSchedules
+    to get a listing of stored objects with their docstring descriptions.
+
+    Note: in previous versions of the schedulers the argument `schedule_beta` was used to state the beta schedule
+    type, this now replaced with `schedule` and most names used with the previous argument now have "_beta" appended
+    to them, eg. 'schedule_beta="linear"' -> 'schedule="linear_beta"'. The `beta_start` and `beta_end` arguments are
+    still used for some schedules but these are provided as keyword arguments now.
+
+    Args:
+        num_train_timesteps: number of diffusion steps used to train the model.
+        schedule: member of NoiseSchedules,
+            a named function returning the beta tensor or (betas, alphas, alphas_cumprod) triple
+        schedule_args: arguments to pass to the schedule function
+    """
+
+    def __init__(self, num_train_timesteps: int = 1000, schedule: str = "linear_beta", **schedule_args) -> None:
+        super().__init__()
+        schedule_args["num_train_timesteps"] = num_train_timesteps
+        noise_sched = NoiseSchedules[schedule](**schedule_args)
+
+        # set betas, alphas, alphas_cumprod based off return value from noise function
+        if isinstance(noise_sched, tuple):
+            self.betas, self.alphas, self.alphas_cumprod = noise_sched
+        else:
+            self.betas = noise_sched
+            self.alphas = 1.0 - self.betas
+            self.alphas_cumprod = torch.cumprod(self.alphas, dim=0)
+
+        self.num_train_timesteps = num_train_timesteps
+        self.one = torch.tensor(1.0)
+
+        # settable values
+        self.num_inference_steps: int | None = None
+        self.timesteps = torch.arange(num_train_timesteps - 1, -1, -1)
+
+    def add_noise(self, original_samples: torch.Tensor, noise: torch.Tensor, timesteps: torch.Tensor) -> torch.Tensor:
+        """
+        Add noise to the original samples.
+
+        Args:
+            original_samples: original samples
+            noise: noise to add to samples
+            timesteps: timesteps tensor indicating the timestep to be computed for each sample.
+
+        Returns:
+            noisy_samples: sample with added noise
+        """
+        # Make sure alphas_cumprod and timestep have same device and dtype as original_samples
+        self.alphas_cumprod = self.alphas_cumprod.to(device=original_samples.device, dtype=original_samples.dtype)
+        timesteps = timesteps.to(original_samples.device)
+
+        sqrt_alpha_cumprod: torch.Tensor = unsqueeze_right(self.alphas_cumprod[timesteps] ** 0.5, original_samples.ndim)
+        sqrt_one_minus_alpha_prod: torch.Tensor = unsqueeze_right(
+            (1 - self.alphas_cumprod[timesteps]) ** 0.5, original_samples.ndim
+        )
+
+        noisy_samples = sqrt_alpha_cumprod * original_samples + sqrt_one_minus_alpha_prod * noise
+        return noisy_samples
+
+    def get_velocity(self, sample: torch.Tensor, noise: torch.Tensor, timesteps: torch.Tensor) -> torch.Tensor:
+        # Make sure alphas_cumprod and timestep have same device and dtype as sample
+        self.alphas_cumprod = self.alphas_cumprod.to(device=sample.device, dtype=sample.dtype)
+        timesteps = timesteps.to(sample.device)
+
+        sqrt_alpha_prod: torch.Tensor = unsqueeze_right(self.alphas_cumprod[timesteps] ** 0.5, sample.ndim)
+        sqrt_one_minus_alpha_prod: torch.Tensor = unsqueeze_right(
+            (1 - self.alphas_cumprod[timesteps]) ** 0.5, sample.ndim
+        )
+
+        velocity = sqrt_alpha_prod * noise - sqrt_one_minus_alpha_prod * sample
+        return velocity
diff --git a/monai/networks/utils.py b/monai/networks/utils.py
index 152911f443..6a97434215 100644
--- a/monai/networks/utils.py
+++ b/monai/networks/utils.py
@@ -42,6 +42,7 @@
     "predict_segmentation",
     "normalize_transform",
     "to_norm_affine",
+    "CastTempType",
     "normal_init",
     "icnr_init",
     "pixelshuffle",
@@ -1167,3 +1168,24 @@ def freeze_layers(model: nn.Module, freeze_vars=None, exclude_vars=None):
                 warnings.warn(f"The exclude_vars includes {param}, but requires_grad is False, change it to True.")
 
     logger.info(f"{len(frozen_keys)} of {len(src_dict)} variables frozen.")
+
+
+class CastTempType(nn.Module):
+    """
+    Cast the input tensor to a temporary type before applying the submodule, and then cast it back to the initial type.
+    """
+
+    def __init__(self, initial_type, temporary_type, submodule):
+        super().__init__()
+        self.initial_type = initial_type
+        self.temporary_type = temporary_type
+        self.submodule = submodule
+
+    def forward(self, x):
+        dtype = x.dtype
+        if dtype == self.initial_type:
+            x = x.to(self.temporary_type)
+        x = self.submodule(x)
+        if dtype == self.initial_type:
+            x = x.to(self.initial_type)
+        return x
diff --git a/monai/transforms/regularization/array.py b/monai/transforms/regularization/array.py
index a7436bda84..4bf6cff649 100644
--- a/monai/transforms/regularization/array.py
+++ b/monai/transforms/regularization/array.py
@@ -87,12 +87,14 @@ def apply(self, data: torch.Tensor):
 
     def __call__(self, data: torch.Tensor, labels: torch.Tensor | None = None, randomize=True):
         data_t = convert_to_tensor(data, track_meta=get_track_meta())
+        labels_t = data_t  # will not stay this value, needed to satisfy pylint/mypy
         if labels is not None:
             labels_t = convert_to_tensor(labels, track_meta=get_track_meta())
         if randomize:
             self.randomize()
         if labels is None:
             return convert_to_dst_type(self.apply(data_t), dst=data)[0]
+
         return (
             convert_to_dst_type(self.apply(data_t), dst=data)[0],
             convert_to_dst_type(self.apply(labels_t), dst=labels)[0],
@@ -149,11 +151,13 @@ def apply_on_labels(self, labels: torch.Tensor):
 
     def __call__(self, data: torch.Tensor, labels: torch.Tensor | None = None, randomize=True):
         data_t = convert_to_tensor(data, track_meta=get_track_meta())
+        augmented_label = None
         if labels is not None:
             labels_t = convert_to_tensor(labels, track_meta=get_track_meta())
         if randomize:
             self.randomize(data)
         augmented = convert_to_dst_type(self.apply(data_t), dst=data)[0]
+
         if labels is not None:
             augmented_label = convert_to_dst_type(self.apply(labels_t), dst=labels)[0]
         return (augmented, augmented_label) if labels is not None else augmented
diff --git a/monai/transforms/utils_create_transform_ims.py b/monai/transforms/utils_create_transform_ims.py
index 4b5990abd3..a29fd4dbf9 100644
--- a/monai/transforms/utils_create_transform_ims.py
+++ b/monai/transforms/utils_create_transform_ims.py
@@ -269,11 +269,9 @@ def update_docstring(code_path, transform_name):
 
 
 def pre_process_data(data, ndim, is_map, is_post):
-    """If transform requires 2D data, then convert to 2D"""
+    """If transform requires 2D data, then convert to 2D by selecting the middle of the last dimension."""
     if ndim == 2:
-        for k in keys:
-            data[k] = data[k][..., data[k].shape[-1] // 2]
-
+        data = {k: v[..., v.shape[-1] // 2] for k, v in data.items()}
     if is_map:
         return data
     return data[CommonKeys.LABEL] if is_post else data[CommonKeys.IMAGE]
diff --git a/monai/utils/__init__.py b/monai/utils/__init__.py
index 2c32eb2cf4..03fa1ceed1 100644
--- a/monai/utils/__init__.py
+++ b/monai/utils/__init__.py
@@ -126,6 +126,7 @@
     version_leq,
 )
 from .nvtx import Range
+from .ordering import Ordering
 from .profiling import (
     PerfContext,
     ProfileHandler,
diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index e8a46ecc61..40370ca2c6 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -118,6 +118,7 @@ def star_zip_with(op, *vals):
 
 
 T = TypeVar("T")
+NT = TypeVar("NT", np.ndarray, torch.Tensor)
 
 
 @overload
@@ -907,11 +908,11 @@ def is_sqrt(num: Sequence[int] | int) -> bool:
     return ensure_tuple(ret) == num
 
 
-def unsqueeze_right(arr: NdarrayOrTensor, ndim: int) -> NdarrayOrTensor:
+def unsqueeze_right(arr: NT, ndim: int) -> NT:
     """Append 1-sized dimensions to `arr` to create a result with `ndim` dimensions."""
     return arr[(...,) + (None,) * (ndim - arr.ndim)]
 
 
-def unsqueeze_left(arr: NdarrayOrTensor, ndim: int) -> NdarrayOrTensor:
+def unsqueeze_left(arr: NT, ndim: int) -> NT:
     """Prepend 1-sized dimensions to `arr` to create a result with `ndim` dimensions."""
     return arr[(None,) * (ndim - arr.ndim)]
diff --git a/monai/utils/ordering.py b/monai/utils/ordering.py
new file mode 100644
index 0000000000..1be61f98ab
--- /dev/null
+++ b/monai/utils/ordering.py
@@ -0,0 +1,207 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import numpy as np
+
+from monai.utils.enums import OrderingTransformations, OrderingType
+
+
+class Ordering:
+    """
+    Ordering class that projects a 2D or 3D image into a 1D sequence. It also allows the image to be transformed with
+    one of the following transformations:
+    Reflection (see np.flip for more details).
+    Transposition (see np.transpose for more details).
+    90-degree rotation (see np.rot90 for more details).
+
+    The transformations are applied in the order specified by the transformation_order parameter.
+
+    Args:
+        ordering_type: The ordering type. One of the following:
+            - 'raster_scan': The image is projected into a 1D sequence by scanning the image from left to right and from
+            top to bottom. Also called a row major ordering.
+            - 's_curve': The image is projected into a 1D sequence by scanning the image in a circular snake like
+            pattern from top left towards right gowing in a spiral towards the center.
+            - random': The image is projected into a 1D sequence by randomly shuffling the image.
+        spatial_dims: The number of spatial dimensions of the image.
+        dimensions: The dimensions of the image.
+        reflected_spatial_dims: A tuple of booleans indicating whether to reflect the image along each spatial dimension.
+        transpositions_axes: A tuple of tuples indicating the axes to transpose the image along.
+        rot90_axes: A tuple of tuples indicating the axes to rotate the image along.
+        transformation_order: The order in which to apply the transformations.
+    """
+
+    def __init__(
+        self,
+        ordering_type: str,
+        spatial_dims: int,
+        dimensions: tuple[int, int, int] | tuple[int, int, int, int],
+        reflected_spatial_dims: tuple[bool, bool] | None = None,
+        transpositions_axes: tuple[tuple[int, int], ...] | tuple[tuple[int, int, int], ...] | None = None,
+        rot90_axes: tuple[tuple[int, int], ...] | None = None,
+        transformation_order: tuple[str, ...] = (
+            OrderingTransformations.TRANSPOSE.value,
+            OrderingTransformations.ROTATE_90.value,
+            OrderingTransformations.REFLECT.value,
+        ),
+    ) -> None:
+        super().__init__()
+        self.ordering_type = ordering_type
+
+        if self.ordering_type not in list(OrderingType):
+            raise ValueError(
+                f"ordering_type must be one of the following {list(OrderingType)}, but got {self.ordering_type}."
+            )
+
+        self.spatial_dims = spatial_dims
+        self.dimensions = dimensions
+
+        if len(dimensions) != self.spatial_dims + 1:
+            raise ValueError(f"dimensions must be of length {self.spatial_dims + 1}, but got {len(dimensions)}.")
+
+        self.reflected_spatial_dims = reflected_spatial_dims
+        self.transpositions_axes = transpositions_axes
+        self.rot90_axes = rot90_axes
+        if len(set(transformation_order)) != len(transformation_order):
+            raise ValueError(f"No duplicates are allowed. Received {transformation_order}.")
+
+        for transformation in transformation_order:
+            if transformation not in list(OrderingTransformations):
+                raise ValueError(
+                    f"Valid transformations are {list(OrderingTransformations)} but received {transformation}."
+                )
+        self.transformation_order = transformation_order
+
+        self.template = self._create_template()
+        self._sequence_ordering = self._create_ordering()
+        self._revert_sequence_ordering = np.argsort(self._sequence_ordering)
+
+    def __call__(self, x: np.ndarray) -> np.ndarray:
+        x = x[self._sequence_ordering]
+
+        return x
+
+    def get_sequence_ordering(self) -> np.ndarray:
+        return self._sequence_ordering
+
+    def get_revert_sequence_ordering(self) -> np.ndarray:
+        return self._revert_sequence_ordering
+
+    def _create_ordering(self) -> np.ndarray:
+        self.template = self._transform_template()
+        order = self._order_template(template=self.template)
+
+        return order
+
+    def _create_template(self) -> np.ndarray:
+        spatial_dimensions = self.dimensions[1:]
+        template = np.arange(np.prod(spatial_dimensions)).reshape(*spatial_dimensions)
+
+        return template
+
+    def _transform_template(self) -> np.ndarray:
+        for transformation in self.transformation_order:
+            if transformation == OrderingTransformations.TRANSPOSE.value:
+                self.template = self._transpose_template(template=self.template)
+            elif transformation == OrderingTransformations.ROTATE_90.value:
+                self.template = self._rot90_template(template=self.template)
+            elif transformation == OrderingTransformations.REFLECT.value:
+                self.template = self._flip_template(template=self.template)
+
+        return self.template
+
+    def _transpose_template(self, template: np.ndarray) -> np.ndarray:
+        if self.transpositions_axes is not None:
+            for axes in self.transpositions_axes:
+                template = np.transpose(template, axes=axes)
+
+        return template
+
+    def _flip_template(self, template: np.ndarray) -> np.ndarray:
+        if self.reflected_spatial_dims is not None:
+            for axis, to_reflect in enumerate(self.reflected_spatial_dims):
+                template = np.flip(template, axis=axis) if to_reflect else template
+
+        return template
+
+    def _rot90_template(self, template: np.ndarray) -> np.ndarray:
+        if self.rot90_axes is not None:
+            for axes in self.rot90_axes:
+                template = np.rot90(template, axes=axes)
+
+        return template
+
+    def _order_template(self, template: np.ndarray) -> np.ndarray:
+        depths = None
+        if self.spatial_dims == 2:
+            rows, columns = template.shape[0], template.shape[1]
+        else:
+            rows, columns, depths = (template.shape[0], template.shape[1], template.shape[2])
+
+        sequence = eval(f"self.{self.ordering_type}_idx")(rows, columns, depths)
+
+        ordering = np.array([template[tuple(e)] for e in sequence])
+
+        return ordering
+
+    @staticmethod
+    def raster_scan_idx(rows: int, cols: int, depths: int | None = None) -> np.ndarray:
+        idx: list[tuple] = []
+
+        for r in range(rows):
+            for c in range(cols):
+                if depths is not None:
+                    for d in range(depths):
+                        idx.append((r, c, d))
+                else:
+                    idx.append((r, c))
+
+        idx_np = np.array(idx)
+
+        return idx_np
+
+    @staticmethod
+    def s_curve_idx(rows: int, cols: int, depths: int | None = None) -> np.ndarray:
+        idx: list[tuple] = []
+
+        for r in range(rows):
+            col_idx = range(cols) if r % 2 == 0 else range(cols - 1, -1, -1)
+            for c in col_idx:
+                if depths:
+                    depth_idx = range(depths) if c % 2 == 0 else range(depths - 1, -1, -1)
+
+                    for d in depth_idx:
+                        idx.append((r, c, d))
+                else:
+                    idx.append((r, c))
+
+        idx_np = np.array(idx)
+
+        return idx_np
+
+    @staticmethod
+    def random_idx(rows: int, cols: int, depths: int | None = None) -> np.ndarray:
+        idx: list[tuple] = []
+
+        for r in range(rows):
+            for c in range(cols):
+                if depths:
+                    for d in range(depths):
+                        idx.append((r, c, d))
+                else:
+                    idx.append((r, c))
+
+        idx_np = np.array(idx)
+        np.random.shuffle(idx_np)
+
+        return idx_np
diff --git a/tests/hvd_evenly_divisible_all_gather.py b/tests/hvd_evenly_divisible_all_gather.py
index 78c6ca06bc..732ad13b83 100644
--- a/tests/hvd_evenly_divisible_all_gather.py
+++ b/tests/hvd_evenly_divisible_all_gather.py
@@ -30,10 +30,10 @@ def test_data(self):
         self._run()
 
     def _run(self):
-        if hvd.rank() == 0:
-            data1 = torch.tensor([[1, 2], [3, 4]])
-            data2 = torch.tensor([[1.0, 2.0]])
-            data3 = torch.tensor(7)
+        # if hvd.rank() == 0:
+        data1 = torch.tensor([[1, 2], [3, 4]])
+        data2 = torch.tensor([[1.0, 2.0]])
+        data3 = torch.tensor(7)
 
         if hvd.rank() == 1:
             data1 = torch.tensor([[5, 6]])
diff --git a/tests/min_tests.py b/tests/min_tests.py
index 8128bb7b84..3a143df84b 100644
--- a/tests/min_tests.py
+++ b/tests/min_tests.py
@@ -154,6 +154,7 @@ def run_testsuit():
         "test_plot_2d_or_3d_image",
         "test_png_rw",
         "test_prepare_batch_default",
+        "test_prepare_batch_diffusion",
         "test_prepare_batch_extra_input",
         "test_prepare_batch_hovernet",
         "test_rand_grid_patch",
diff --git a/tests/test_autoencoderkl.py b/tests/test_autoencoderkl.py
new file mode 100644
index 0000000000..d15cb79084
--- /dev/null
+++ b/tests/test_autoencoderkl.py
@@ -0,0 +1,337 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import os
+import tempfile
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.apps import download_url
+from monai.networks import eval_mode
+from monai.networks.nets import AutoencoderKL
+from monai.utils import optional_import
+from tests.utils import SkipIfBeforePyTorchVersion, skip_if_downloading_fails, testing_data_config
+
+tqdm, has_tqdm = optional_import("tqdm", name="tqdm")
+_, has_einops = optional_import("einops")
+
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+
+
+CASES_NO_ATTENTION = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+        },
+        (1, 1, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+        },
+        (1, 1, 16, 16, 16),
+        (1, 1, 16, 16, 16),
+        (1, 4, 4, 4, 4),
+    ],
+]
+
+CASES_ATTENTION = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": (1, 1, 2),
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "with_encoder_nonlocal_attn": False,
+        },
+        (1, 1, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, True),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, True),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16, 16),
+        (1, 1, 16, 16, 16),
+        (1, 4, 4, 4, 4),
+    ],
+]
+
+if has_einops:
+    CASES = CASES_NO_ATTENTION + CASES_ATTENTION
+else:
+    CASES = CASES_NO_ATTENTION
+
+
+class TestAutoEncoderKL(unittest.TestCase):
+    @parameterized.expand(CASES)
+    def test_shape(self, input_param, input_shape, expected_shape, expected_latent_shape):
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.forward(torch.randn(input_shape).to(device))
+            self.assertEqual(result[0].shape, expected_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+            self.assertEqual(result[2].shape, expected_latent_shape)
+
+    @parameterized.expand(CASES)
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_with_convtranspose_and_checkpointing(
+        self, input_param, input_shape, expected_shape, expected_latent_shape
+    ):
+        input_param = input_param.copy()
+        input_param.update({"use_checkpoint": True, "use_convtranspose": True})
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.forward(torch.randn(input_shape).to(device))
+            self.assertEqual(result[0].shape, expected_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+            self.assertEqual(result[2].shape, expected_latent_shape)
+
+    def test_model_channels_not_multiple_of_norm_num_group(self):
+        with self.assertRaises(ValueError):
+            AutoencoderKL(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(24, 24, 24),
+                attention_levels=(False, False, False),
+                latent_channels=8,
+                num_res_blocks=1,
+                norm_num_groups=16,
+            )
+
+    def test_model_num_channels_not_same_size_of_attention_levels(self):
+        with self.assertRaises(ValueError):
+            AutoencoderKL(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(24, 24, 24),
+                attention_levels=(False, False),
+                latent_channels=8,
+                num_res_blocks=1,
+                norm_num_groups=16,
+            )
+
+    def test_model_num_channels_not_same_size_of_num_res_blocks(self):
+        with self.assertRaises(ValueError):
+            AutoencoderKL(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(24, 24, 24),
+                attention_levels=(False, False, False),
+                latent_channels=8,
+                num_res_blocks=(8, 8),
+                norm_num_groups=16,
+            )
+
+    def test_shape_reconstruction(self):
+        input_param, input_shape, expected_shape, _ = CASES[0]
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.reconstruct(torch.randn(input_shape).to(device))
+            self.assertEqual(result.shape, expected_shape)
+
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_reconstruction_with_convtranspose_and_checkpointing(self):
+        input_param, input_shape, expected_shape, _ = CASES[0]
+        input_param = input_param.copy()
+        input_param.update({"use_checkpoint": True, "use_convtranspose": True})
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.reconstruct(torch.randn(input_shape).to(device))
+            self.assertEqual(result.shape, expected_shape)
+
+    def test_shape_encode(self):
+        input_param, input_shape, _, expected_latent_shape = CASES[0]
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.encode(torch.randn(input_shape).to(device))
+            self.assertEqual(result[0].shape, expected_latent_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_encode_with_convtranspose_and_checkpointing(self):
+        input_param, input_shape, _, expected_latent_shape = CASES[0]
+        input_param = input_param.copy()
+        input_param.update({"use_checkpoint": True, "use_convtranspose": True})
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.encode(torch.randn(input_shape).to(device))
+            self.assertEqual(result[0].shape, expected_latent_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+
+    def test_shape_sampling(self):
+        input_param, _, _, expected_latent_shape = CASES[0]
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.sampling(
+                torch.randn(expected_latent_shape).to(device), torch.randn(expected_latent_shape).to(device)
+            )
+            self.assertEqual(result.shape, expected_latent_shape)
+
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_sampling_convtranspose_and_checkpointing(self):
+        input_param, _, _, expected_latent_shape = CASES[0]
+        input_param = input_param.copy()
+        input_param.update({"use_checkpoint": True, "use_convtranspose": True})
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.sampling(
+                torch.randn(expected_latent_shape).to(device), torch.randn(expected_latent_shape).to(device)
+            )
+            self.assertEqual(result.shape, expected_latent_shape)
+
+    def test_shape_decode(self):
+        input_param, expected_input_shape, _, latent_shape = CASES[0]
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.decode(torch.randn(latent_shape).to(device))
+            self.assertEqual(result.shape, expected_input_shape)
+
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_decode_convtranspose_and_checkpointing(self):
+        input_param, expected_input_shape, _, latent_shape = CASES[0]
+        input_param = input_param.copy()
+        input_param.update({"use_checkpoint": True, "use_convtranspose": True})
+        net = AutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.decode(torch.randn(latent_shape).to(device))
+            self.assertEqual(result.shape, expected_input_shape)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_compatibility_with_monai_generative(self):
+        # test loading weights from a model saved in MONAI Generative, version 0.2.3
+        with skip_if_downloading_fails():
+            net = AutoencoderKL(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(4, 4, 4),
+                latent_channels=4,
+                attention_levels=(False, False, True),
+                num_res_blocks=1,
+                norm_num_groups=4,
+            ).to(device)
+
+            tmpdir = tempfile.mkdtemp()
+            key = "autoencoderkl_monai_generative_weights"
+            url = testing_data_config("models", key, "url")
+            hash_type = testing_data_config("models", key, "hash_type")
+            hash_val = testing_data_config("models", key, "hash_val")
+            filename = "autoencoderkl_monai_generative_weights.pt"
+
+            weight_path = os.path.join(tmpdir, filename)
+            download_url(url=url, filepath=weight_path, hash_val=hash_val, hash_type=hash_type)
+
+            net.load_old_state_dict(torch.load(weight_path), verbose=False)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_controlnet.py b/tests/test_controlnet.py
new file mode 100644
index 0000000000..4746c7ce22
--- /dev/null
+++ b/tests/test_controlnet.py
@@ -0,0 +1,215 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import os
+import tempfile
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.apps import download_url
+from monai.networks import eval_mode
+from monai.networks.nets.controlnet import ControlNet
+from monai.utils import optional_import
+from tests.utils import skip_if_downloading_fails, testing_data_config
+
+_, has_einops = optional_import("einops")
+UNCOND_CASES_2D = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+        },
+        (1, 8, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        },
+        (1, 8, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (4, 4, 4),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 4,
+        },
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        },
+        (1, 8, 4, 4),
+    ],
+]
+
+UNCOND_CASES_3D = [
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+        },
+        (1, 8, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (4, 4, 4),
+            "num_head_channels": 4,
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 4,
+            "resblock_updown": True,
+        },
+        (1, 4, 4, 4, 4),
+    ],
+]
+
+COND_CASES_2D = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+        },
+        (1, 8, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "resblock_updown": True,
+        },
+        (1, 8, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "upcast_attention": True,
+        },
+        (1, 8, 4, 4),
+    ],
+]
+
+
+class TestControlNet(unittest.TestCase):
+    @parameterized.expand(UNCOND_CASES_2D + UNCOND_CASES_3D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_unconditioned_models(self, input_param, expected_output_shape):
+        input_param["conditioning_embedding_in_channels"] = input_param["in_channels"]
+        input_param["conditioning_embedding_num_channels"] = (input_param["channels"][0],)
+        net = ControlNet(**input_param)
+        with eval_mode(net):
+            x = torch.rand((1, 1) + (16,) * input_param["spatial_dims"])
+            timesteps = torch.randint(0, 1000, (1,)).long()
+            controlnet_cond = torch.rand((1, 1) + (16,) * input_param["spatial_dims"])
+            result = net.forward(x, timesteps=timesteps, controlnet_cond=controlnet_cond)
+            self.assertEqual(len(result[0]), 2 * len(input_param["channels"]))
+            self.assertEqual(result[1].shape, expected_output_shape)
+
+    @parameterized.expand(COND_CASES_2D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models(self, input_param, expected_output_shape):
+        input_param["conditioning_embedding_in_channels"] = input_param["in_channels"]
+        input_param["conditioning_embedding_num_channels"] = (input_param["channels"][0],)
+        net = ControlNet(**input_param)
+        with eval_mode(net):
+            x = torch.rand((1, 1) + (16,) * input_param["spatial_dims"])
+            timesteps = torch.randint(0, 1000, (1,)).long()
+            controlnet_cond = torch.rand((1, 1) + (16,) * input_param["spatial_dims"])
+            result = net.forward(x, timesteps=timesteps, controlnet_cond=controlnet_cond, context=torch.rand((1, 1, 3)))
+            self.assertEqual(len(result[0]), 2 * len(input_param["channels"]))
+            self.assertEqual(result[1].shape, expected_output_shape)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_compatibility_with_monai_generative(self):
+        # test loading weights from a model saved in MONAI Generative, version 0.2.3
+        with skip_if_downloading_fails():
+            net = ControlNet(
+                spatial_dims=2,
+                in_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 8),
+                attention_levels=(False, False, True),
+                norm_num_groups=8,
+                with_conditioning=True,
+                transformer_num_layers=1,
+                cross_attention_dim=3,
+                resblock_updown=True,
+            )
+
+            tmpdir = tempfile.mkdtemp()
+            key = "controlnet_monai_generative_weights"
+            url = testing_data_config("models", key, "url")
+            hash_type = testing_data_config("models", key, "hash_type")
+            hash_val = testing_data_config("models", key, "hash_val")
+            filename = "controlnet_monai_generative_weights.pt"
+
+            weight_path = os.path.join(tmpdir, filename)
+            download_url(url=url, filepath=weight_path, hash_val=hash_val, hash_type=hash_type)
+
+            net.load_old_state_dict(torch.load(weight_path), verbose=False)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_controlnet_inferers.py b/tests/test_controlnet_inferers.py
new file mode 100644
index 0000000000..e3b0aeb5a2
--- /dev/null
+++ b/tests/test_controlnet_inferers.py
@@ -0,0 +1,1310 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.inferers import ControlNetDiffusionInferer, ControlNetLatentDiffusionInferer
+from monai.networks.nets import (
+    VQVAE,
+    AutoencoderKL,
+    ControlNet,
+    DiffusionModelUNet,
+    SPADEAutoencoderKL,
+    SPADEDiffusionModelUNet,
+)
+from monai.networks.schedulers import DDIMScheduler, DDPMScheduler
+from monai.utils import optional_import
+
+_, has_scipy = optional_import("scipy")
+_, has_einops = optional_import("einops")
+
+
+CNDM_TEST_CASES = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [8],
+            "norm_num_groups": 8,
+            "attention_levels": [True],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "channels": [8],
+            "attention_levels": [True],
+            "norm_num_groups": 8,
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (2, 1, 8, 8),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [8],
+            "norm_num_groups": 8,
+            "attention_levels": [True],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "channels": [8],
+            "attention_levels": [True],
+            "num_res_blocks": 1,
+            "norm_num_groups": 8,
+            "num_head_channels": 8,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (2, 1, 8, 8, 8),
+    ],
+]
+LATENT_CNDM_TEST_CASES = [
+    [
+        "AutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "channels": [4, 4],
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "num_head_channels": 4,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (1, 1, 8, 8),
+        (1, 3, 4, 4),
+    ],
+    [
+        "VQVAE",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [4, 4],
+            "num_res_layers": 1,
+            "num_res_channels": [4, 4],
+            "downsample_parameters": ((2, 4, 1, 1), (2, 4, 1, 1)),
+            "upsample_parameters": ((2, 4, 1, 1, 0), (2, 4, 1, 1, 0)),
+            "num_embeddings": 16,
+            "embedding_dim": 3,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [8, 8],
+            "norm_num_groups": 8,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "channels": [8, 8],
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "norm_num_groups": 8,
+            "num_head_channels": 8,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (1, 1, 16, 16),
+        (1, 3, 4, 4),
+    ],
+    [
+        "VQVAE",
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [4, 4],
+            "num_res_layers": 1,
+            "num_res_channels": [4, 4],
+            "downsample_parameters": ((2, 4, 1, 1), (2, 4, 1, 1)),
+            "upsample_parameters": ((2, 4, 1, 1, 0), (2, 4, 1, 1, 0)),
+            "num_embeddings": 16,
+            "embedding_dim": 3,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 3,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [8, 8],
+            "norm_num_groups": 8,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        {
+            "spatial_dims": 3,
+            "in_channels": 3,
+            "channels": [8, 8],
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "norm_num_groups": 8,
+            "num_head_channels": 8,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (1, 1, 16, 16, 16),
+        (1, 3, 4, 4, 4),
+    ],
+]
+LATENT_CNDM_TEST_CASES_DIFF_SHAPES = [
+    [
+        "AutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "channels": [4, 4],
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "num_head_channels": 4,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (1, 1, 12, 12),
+        (1, 3, 8, 8),
+    ],
+    [
+        "VQVAE",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [4, 4],
+            "num_res_layers": 1,
+            "num_res_channels": [4, 4],
+            "downsample_parameters": ((2, 4, 1, 1), (2, 4, 1, 1)),
+            "upsample_parameters": ((2, 4, 1, 1, 0), (2, 4, 1, 1, 0)),
+            "num_embeddings": 16,
+            "embedding_dim": 3,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [8, 8],
+            "norm_num_groups": 8,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "channels": [8, 8],
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "norm_num_groups": 8,
+            "num_head_channels": 8,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (1, 1, 12, 12),
+        (1, 3, 8, 8),
+    ],
+    [
+        "VQVAE",
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [4, 4],
+            "num_res_layers": 1,
+            "num_res_channels": [4, 4],
+            "downsample_parameters": ((2, 4, 1, 1), (2, 4, 1, 1)),
+            "upsample_parameters": ((2, 4, 1, 1, 0), (2, 4, 1, 1, 0)),
+            "num_embeddings": 16,
+            "embedding_dim": 3,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 3,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [8, 8],
+            "norm_num_groups": 8,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        {
+            "spatial_dims": 3,
+            "in_channels": 3,
+            "channels": [8, 8],
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "norm_num_groups": 8,
+            "num_head_channels": 8,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (1, 1, 12, 12, 12),
+        (1, 3, 8, 8, 8),
+    ],
+    [
+        "SPADEAutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "channels": [4, 4],
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "num_head_channels": 4,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (1, 1, 8, 8),
+        (1, 3, 4, 4),
+    ],
+    [
+        "AutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "SPADEDiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "channels": [4, 4],
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "num_head_channels": 4,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (1, 1, 8, 8),
+        (1, 3, 4, 4),
+    ],
+    [
+        "SPADEAutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "SPADEDiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "channels": [4, 4],
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "num_head_channels": 4,
+            "conditioning_embedding_num_channels": [16],
+            "conditioning_embedding_in_channels": 1,
+        },
+        (1, 1, 8, 8),
+        (1, 3, 4, 4),
+    ],
+]
+
+
+class ControlNetTestDiffusionSamplingInferer(unittest.TestCase):
+    @parameterized.expand(CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_call(self, model_params, controlnet_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        controlnet = ControlNet(**controlnet_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        controlnet.to(device)
+        controlnet.eval()
+        input = torch.randn(input_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        noise = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetDiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],), device=input.device).long()
+        sample = inferer(
+            inputs=input, noise=noise, diffusion_model=model, controlnet=controlnet, timesteps=timesteps, cn_cond=mask
+        )
+        self.assertEqual(sample.shape, input_shape)
+
+    @parameterized.expand(CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_intermediates(self, model_params, controlnet_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        controlnet = ControlNet(**controlnet_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        controlnet.to(device)
+        controlnet.eval()
+        noise = torch.randn(input_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetDiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        sample, intermediates = inferer.sample(
+            input_noise=noise,
+            diffusion_model=model,
+            scheduler=scheduler,
+            controlnet=controlnet,
+            cn_cond=mask,
+            save_intermediates=True,
+            intermediate_steps=1,
+        )
+        self.assertEqual(len(intermediates), 10)
+
+    @parameterized.expand(CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_ddpm_sampler(self, model_params, controlnet_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        controlnet = ControlNet(**controlnet_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        controlnet.to(device)
+        controlnet.eval()
+        mask = torch.randn(input_shape).to(device)
+        noise = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=1000)
+        inferer = ControlNetDiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        sample, intermediates = inferer.sample(
+            input_noise=noise,
+            diffusion_model=model,
+            scheduler=scheduler,
+            controlnet=controlnet,
+            cn_cond=mask,
+            save_intermediates=True,
+            intermediate_steps=1,
+        )
+        self.assertEqual(len(intermediates), 10)
+
+    @parameterized.expand(CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_ddim_sampler(self, model_params, controlnet_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        controlnet = ControlNet(**controlnet_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        controlnet.to(device)
+        controlnet.eval()
+        mask = torch.randn(input_shape).to(device)
+        noise = torch.randn(input_shape).to(device)
+        scheduler = DDIMScheduler(num_train_timesteps=1000)
+        inferer = ControlNetDiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        sample, intermediates = inferer.sample(
+            input_noise=noise,
+            diffusion_model=model,
+            scheduler=scheduler,
+            controlnet=controlnet,
+            cn_cond=mask,
+            save_intermediates=True,
+            intermediate_steps=1,
+        )
+        self.assertEqual(len(intermediates), 10)
+
+    @parameterized.expand(CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sampler_conditioned(self, model_params, controlnet_params, input_shape):
+        model_params["with_conditioning"] = True
+        model_params["cross_attention_dim"] = 3
+        model = DiffusionModelUNet(**model_params)
+        controlnet = ControlNet(**controlnet_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        controlnet.to(device)
+        controlnet.eval()
+        mask = torch.randn(input_shape).to(device)
+        noise = torch.randn(input_shape).to(device)
+        scheduler = DDIMScheduler(num_train_timesteps=1000)
+        inferer = ControlNetDiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        conditioning = torch.randn([input_shape[0], 1, 3]).to(device)
+        sample, intermediates = inferer.sample(
+            input_noise=noise,
+            diffusion_model=model,
+            controlnet=controlnet,
+            cn_cond=mask,
+            scheduler=scheduler,
+            save_intermediates=True,
+            intermediate_steps=1,
+            conditioning=conditioning,
+        )
+        self.assertEqual(len(intermediates), 10)
+
+    @parameterized.expand(CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_get_likelihood(self, model_params, controlnet_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        controlnet = ControlNet(**controlnet_params)
+        controlnet.to(device)
+        controlnet.eval()
+        input = torch.randn(input_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetDiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        likelihood, intermediates = inferer.get_likelihood(
+            inputs=input,
+            diffusion_model=model,
+            scheduler=scheduler,
+            controlnet=controlnet,
+            cn_cond=mask,
+            save_intermediates=True,
+        )
+        self.assertEqual(intermediates[0].shape, input.shape)
+        self.assertEqual(likelihood.shape[0], input.shape[0])
+
+    @unittest.skipUnless(has_scipy, "Requires scipy library.")
+    def test_normal_cdf(self):
+        from scipy.stats import norm
+
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetDiffusionInferer(scheduler=scheduler)
+        x = torch.linspace(-10, 10, 20)
+        cdf_approx = inferer._approx_standard_normal_cdf(x)
+        cdf_true = norm.cdf(x)
+        torch.testing.assert_allclose(cdf_approx, cdf_true, atol=1e-3, rtol=1e-5)
+
+    @parameterized.expand(CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sampler_conditioned_concat(self, model_params, controlnet_params, input_shape):
+        # copy the model_params dict to prevent from modifying test cases
+        model_params = model_params.copy()
+        n_concat_channel = 2
+        model_params["in_channels"] = model_params["in_channels"] + n_concat_channel
+        model_params["cross_attention_dim"] = None
+        model_params["with_conditioning"] = False
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        controlnet = ControlNet(**controlnet_params)
+        controlnet.to(device)
+        controlnet.eval()
+        noise = torch.randn(input_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        conditioning_shape = list(input_shape)
+        conditioning_shape[1] = n_concat_channel
+        conditioning = torch.randn(conditioning_shape).to(device)
+        scheduler = DDIMScheduler(num_train_timesteps=1000)
+        inferer = ControlNetDiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        sample, intermediates = inferer.sample(
+            input_noise=noise,
+            diffusion_model=model,
+            controlnet=controlnet,
+            cn_cond=mask,
+            scheduler=scheduler,
+            save_intermediates=True,
+            intermediate_steps=1,
+            conditioning=conditioning,
+            mode="concat",
+        )
+        self.assertEqual(len(intermediates), 10)
+
+
+class LatentControlNetTestDiffusionSamplingInferer(unittest.TestCase):
+    @parameterized.expand(LATENT_CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_prediction_shape(
+        self,
+        ae_model_type,
+        autoencoder_params,
+        dm_model_type,
+        stage_2_params,
+        controlnet_params,
+        input_shape,
+        latent_shape,
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+        controlnet = ControlNet(**controlnet_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        controlnet.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        controlnet.eval()
+
+        input = torch.randn(input_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        noise = torch.randn(latent_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetLatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],), device=input.device).long()
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                controlnet=controlnet,
+                cn_cond=mask,
+                seg=input_seg,
+                noise=noise,
+                timesteps=timesteps,
+            )
+        else:
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                noise=noise,
+                timesteps=timesteps,
+                controlnet=controlnet,
+                cn_cond=mask,
+            )
+        self.assertEqual(prediction.shape, latent_shape)
+
+    @parameterized.expand(LATENT_CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_shape(
+        self,
+        ae_model_type,
+        autoencoder_params,
+        dm_model_type,
+        stage_2_params,
+        controlnet_params,
+        input_shape,
+        latent_shape,
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+        controlnet = ControlNet(**controlnet_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        controlnet.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        controlnet.eval()
+
+        noise = torch.randn(latent_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetLatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if ae_model_type == "SPADEAutoencoderKL" or dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                controlnet=controlnet,
+                cn_cond=mask,
+                scheduler=scheduler,
+                seg=input_seg,
+            )
+        else:
+            sample = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                controlnet=controlnet,
+                cn_cond=mask,
+            )
+        self.assertEqual(sample.shape, input_shape)
+
+    @parameterized.expand(LATENT_CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_intermediates(
+        self,
+        ae_model_type,
+        autoencoder_params,
+        dm_model_type,
+        stage_2_params,
+        controlnet_params,
+        input_shape,
+        latent_shape,
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+        controlnet = ControlNet(**controlnet_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        controlnet.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        controlnet.eval()
+
+        noise = torch.randn(latent_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetLatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if ae_model_type == "SPADEAutoencoderKL" or dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                seg=input_seg,
+                controlnet=controlnet,
+                cn_cond=mask,
+            )
+
+            # TODO: this isn't correct, should the above produce intermediates as well?
+            # This test has always passed so is this branch not being used?
+            intermediates = None
+        else:
+            sample, intermediates = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                save_intermediates=True,
+                intermediate_steps=1,
+                controlnet=controlnet,
+                cn_cond=mask,
+            )
+
+        self.assertEqual(len(intermediates), 10)
+        self.assertEqual(intermediates[0].shape, input_shape)
+
+    @parameterized.expand(LATENT_CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_get_likelihoods(
+        self,
+        ae_model_type,
+        autoencoder_params,
+        dm_model_type,
+        stage_2_params,
+        controlnet_params,
+        input_shape,
+        latent_shape,
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+        controlnet = ControlNet(**controlnet_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        controlnet.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        controlnet.eval()
+
+        input = torch.randn(input_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetLatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample, intermediates = inferer.get_likelihood(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                controlnet=controlnet,
+                cn_cond=mask,
+                scheduler=scheduler,
+                save_intermediates=True,
+                seg=input_seg,
+            )
+        else:
+            sample, intermediates = inferer.get_likelihood(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                controlnet=controlnet,
+                cn_cond=mask,
+                save_intermediates=True,
+            )
+        self.assertEqual(len(intermediates), 10)
+        self.assertEqual(intermediates[0].shape, latent_shape)
+
+    @parameterized.expand(LATENT_CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_resample_likelihoods(
+        self,
+        ae_model_type,
+        autoencoder_params,
+        dm_model_type,
+        stage_2_params,
+        controlnet_params,
+        input_shape,
+        latent_shape,
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+        controlnet = ControlNet(**controlnet_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        controlnet.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        controlnet.eval()
+
+        input = torch.randn(input_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetLatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample, intermediates = inferer.get_likelihood(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                controlnet=controlnet,
+                cn_cond=mask,
+                save_intermediates=True,
+                resample_latent_likelihoods=True,
+                seg=input_seg,
+            )
+        else:
+            sample, intermediates = inferer.get_likelihood(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                controlnet=controlnet,
+                cn_cond=mask,
+                save_intermediates=True,
+                resample_latent_likelihoods=True,
+            )
+        self.assertEqual(len(intermediates), 10)
+        self.assertEqual(intermediates[0].shape[2:], input_shape[2:])
+
+    @parameterized.expand(LATENT_CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_prediction_shape_conditioned_concat(
+        self,
+        ae_model_type,
+        autoencoder_params,
+        dm_model_type,
+        stage_2_params,
+        controlnet_params,
+        input_shape,
+        latent_shape,
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        stage_2_params = stage_2_params.copy()
+        n_concat_channel = 3
+        stage_2_params["in_channels"] = stage_2_params["in_channels"] + n_concat_channel
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+        controlnet = ControlNet(**controlnet_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        controlnet.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        controlnet.eval()
+
+        input = torch.randn(input_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        noise = torch.randn(latent_shape).to(device)
+        conditioning_shape = list(latent_shape)
+        conditioning_shape[1] = n_concat_channel
+        conditioning = torch.randn(conditioning_shape).to(device)
+
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetLatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],), device=input.device).long()
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                noise=noise,
+                controlnet=controlnet,
+                cn_cond=mask,
+                timesteps=timesteps,
+                condition=conditioning,
+                mode="concat",
+                seg=input_seg,
+            )
+        else:
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                noise=noise,
+                controlnet=controlnet,
+                cn_cond=mask,
+                timesteps=timesteps,
+                condition=conditioning,
+                mode="concat",
+            )
+        self.assertEqual(prediction.shape, latent_shape)
+
+    @parameterized.expand(LATENT_CNDM_TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_shape_conditioned_concat(
+        self,
+        ae_model_type,
+        autoencoder_params,
+        dm_model_type,
+        stage_2_params,
+        controlnet_params,
+        input_shape,
+        latent_shape,
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        stage_2_params = stage_2_params.copy()
+        n_concat_channel = 3
+        stage_2_params["in_channels"] = stage_2_params["in_channels"] + n_concat_channel
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+        controlnet = ControlNet(**controlnet_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        controlnet.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        controlnet.eval()
+
+        noise = torch.randn(latent_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        conditioning_shape = list(latent_shape)
+        conditioning_shape[1] = n_concat_channel
+        conditioning = torch.randn(conditioning_shape).to(device)
+
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetLatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                controlnet=controlnet,
+                cn_cond=mask,
+                scheduler=scheduler,
+                conditioning=conditioning,
+                mode="concat",
+                seg=input_seg,
+            )
+        else:
+            sample = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                controlnet=controlnet,
+                cn_cond=mask,
+                scheduler=scheduler,
+                conditioning=conditioning,
+                mode="concat",
+            )
+        self.assertEqual(sample.shape, input_shape)
+
+    @parameterized.expand(LATENT_CNDM_TEST_CASES_DIFF_SHAPES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_shape_different_latents(
+        self,
+        ae_model_type,
+        autoencoder_params,
+        dm_model_type,
+        stage_2_params,
+        controlnet_params,
+        input_shape,
+        latent_shape,
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+        controlnet = ControlNet(**controlnet_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        controlnet.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        controlnet.eval()
+
+        input = torch.randn(input_shape).to(device)
+        noise = torch.randn(latent_shape).to(device)
+        mask = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        # We infer the VAE shape
+        autoencoder_latent_shape = [i // (2 ** (len(autoencoder_params["channels"]) - 1)) for i in input_shape[2:]]
+        inferer = ControlNetLatentDiffusionInferer(
+            scheduler=scheduler,
+            scale_factor=1.0,
+            ldm_latent_shape=list(latent_shape[2:]),
+            autoencoder_latent_shape=autoencoder_latent_shape,
+        )
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],), device=input.device).long()
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                controlnet=controlnet,
+                cn_cond=mask,
+                noise=noise,
+                timesteps=timesteps,
+                seg=input_seg,
+            )
+        else:
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                noise=noise,
+                controlnet=controlnet,
+                cn_cond=mask,
+                timesteps=timesteps,
+            )
+        self.assertEqual(prediction.shape, latent_shape)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_incompatible_spade_setup(self):
+        stage_1 = SPADEAutoencoderKL(
+            spatial_dims=2,
+            label_nc=6,
+            in_channels=1,
+            out_channels=1,
+            channels=(4, 4),
+            latent_channels=3,
+            attention_levels=[False, False],
+            num_res_blocks=1,
+            with_encoder_nonlocal_attn=False,
+            with_decoder_nonlocal_attn=False,
+            norm_num_groups=4,
+        )
+        stage_2 = SPADEDiffusionModelUNet(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=3,
+            out_channels=3,
+            channels=[4, 4],
+            norm_num_groups=4,
+            attention_levels=[False, False],
+            num_res_blocks=1,
+            num_head_channels=4,
+        )
+        controlnet = ControlNet(
+            spatial_dims=2,
+            in_channels=1,
+            channels=[4, 4],
+            norm_num_groups=4,
+            attention_levels=[False, False],
+            num_res_blocks=1,
+            num_head_channels=4,
+            conditioning_embedding_num_channels=[16],
+        )
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        controlnet.to(device)
+        controlnet.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        controlnet.eval()
+        noise = torch.randn((1, 3, 4, 4)).to(device)
+        mask = torch.randn((1, 1, 4, 4)).to(device)
+        input_seg = torch.randn((1, 3, 8, 8)).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = ControlNetLatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        with self.assertRaises(ValueError):
+            _ = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                controlnet=controlnet,
+                cn_cond=mask,
+                seg=input_seg,
+            )
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_crossattention.py b/tests/test_crossattention.py
new file mode 100644
index 0000000000..4ab0ab1823
--- /dev/null
+++ b/tests/test_crossattention.py
@@ -0,0 +1,131 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import numpy as np
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.blocks.crossattention import CrossAttentionBlock
+from monai.networks.layers.factories import RelPosEmbedding
+from monai.utils import optional_import
+
+einops, has_einops = optional_import("einops")
+
+TEST_CASE_CABLOCK = []
+for dropout_rate in np.linspace(0, 1, 4):
+    for hidden_size in [360, 480, 600, 768]:
+        for num_heads in [4, 6, 8, 12]:
+            for rel_pos_embedding in [None, RelPosEmbedding.DECOMPOSED]:
+                for input_size in [(16, 32), (8, 8, 8)]:
+                    test_case = [
+                        {
+                            "hidden_size": hidden_size,
+                            "num_heads": num_heads,
+                            "dropout_rate": dropout_rate,
+                            "rel_pos_embedding": rel_pos_embedding,
+                            "input_size": input_size,
+                        },
+                        (2, 512, hidden_size),
+                        (2, 512, hidden_size),
+                    ]
+                    TEST_CASE_CABLOCK.append(test_case)
+
+
+class TestResBlock(unittest.TestCase):
+
+    @parameterized.expand(TEST_CASE_CABLOCK)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape(self, input_param, input_shape, expected_shape):
+        net = CrossAttentionBlock(**input_param)
+        with eval_mode(net):
+            result = net(torch.randn(input_shape), context=torch.randn(2, 512, input_param["hidden_size"]))
+            self.assertEqual(result.shape, expected_shape)
+
+    def test_ill_arg(self):
+        with self.assertRaises(ValueError):
+            CrossAttentionBlock(hidden_size=128, num_heads=12, dropout_rate=6.0)
+
+        with self.assertRaises(ValueError):
+            CrossAttentionBlock(hidden_size=620, num_heads=8, dropout_rate=0.4)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_attention_dim_not_multiple_of_heads(self):
+        with self.assertRaises(ValueError):
+            CrossAttentionBlock(hidden_size=128, num_heads=3, dropout_rate=0.1)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_inner_dim_different(self):
+        CrossAttentionBlock(hidden_size=128, num_heads=4, dropout_rate=0.1, dim_head=30)
+
+    def test_causal_no_sequence_length(self):
+        with self.assertRaises(ValueError):
+            CrossAttentionBlock(hidden_size=128, num_heads=4, dropout_rate=0.1, causal=True)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_causal(self):
+        block = CrossAttentionBlock(
+            hidden_size=128, num_heads=1, dropout_rate=0.1, causal=True, sequence_length=16, save_attn=True
+        )
+        input_shape = (1, 16, 128)
+        block(torch.randn(input_shape))
+        # check upper triangular part of the attention matrix is zero
+        assert torch.triu(block.att_mat, diagonal=1).sum() == 0
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_context_input(self):
+        block = CrossAttentionBlock(
+            hidden_size=128, num_heads=1, dropout_rate=0.1, causal=True, sequence_length=16, context_input_size=12
+        )
+        input_shape = (1, 16, 128)
+        block(torch.randn(input_shape), context=torch.randn(1, 3, 12))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_context_wrong_input_size(self):
+        block = CrossAttentionBlock(
+            hidden_size=128, num_heads=1, dropout_rate=0.1, causal=True, sequence_length=16, context_input_size=12
+        )
+        input_shape = (1, 16, 128)
+        with self.assertRaises(RuntimeError):
+            block(torch.randn(input_shape), context=torch.randn(1, 3, 24))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_access_attn_matrix(self):
+        # input format
+        hidden_size = 128
+        num_heads = 2
+        dropout_rate = 0
+        input_shape = (2, 256, hidden_size)
+
+        # be  not able to access the matrix
+        no_matrix_acess_blk = CrossAttentionBlock(
+            hidden_size=hidden_size, num_heads=num_heads, dropout_rate=dropout_rate
+        )
+        no_matrix_acess_blk(torch.randn(input_shape))
+        assert isinstance(no_matrix_acess_blk.att_mat, torch.Tensor)
+        # no of elements is zero
+        assert no_matrix_acess_blk.att_mat.nelement() == 0
+
+        # be able to acess the attention matrix
+        matrix_acess_blk = CrossAttentionBlock(
+            hidden_size=hidden_size, num_heads=num_heads, dropout_rate=dropout_rate, save_attn=True
+        )
+        matrix_acess_blk(torch.randn(input_shape))
+        assert matrix_acess_blk.att_mat.shape == (input_shape[0], input_shape[0], input_shape[1], input_shape[1])
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_diffusion_inferer.py b/tests/test_diffusion_inferer.py
new file mode 100644
index 0000000000..7f37025d3c
--- /dev/null
+++ b/tests/test_diffusion_inferer.py
@@ -0,0 +1,236 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.inferers import DiffusionInferer
+from monai.networks.nets import DiffusionModelUNet
+from monai.networks.schedulers import DDIMScheduler, DDPMScheduler
+from monai.utils import optional_import
+
+_, has_scipy = optional_import("scipy")
+_, has_einops = optional_import("einops")
+
+TEST_CASES = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [8],
+            "norm_num_groups": 8,
+            "attention_levels": [True],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        (2, 1, 8, 8),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [8],
+            "norm_num_groups": 8,
+            "attention_levels": [True],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        (2, 1, 8, 8, 8),
+    ],
+]
+
+
+class TestDiffusionSamplingInferer(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_call(self, model_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        input = torch.randn(input_shape).to(device)
+        noise = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],), device=input.device).long()
+        sample = inferer(inputs=input, noise=noise, diffusion_model=model, timesteps=timesteps)
+        self.assertEqual(sample.shape, input_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_intermediates(self, model_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        noise = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        sample, intermediates = inferer.sample(
+            input_noise=noise, diffusion_model=model, scheduler=scheduler, save_intermediates=True, intermediate_steps=1
+        )
+        self.assertEqual(len(intermediates), 10)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_ddpm_sampler(self, model_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        noise = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=1000)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        sample, intermediates = inferer.sample(
+            input_noise=noise, diffusion_model=model, scheduler=scheduler, save_intermediates=True, intermediate_steps=1
+        )
+        self.assertEqual(len(intermediates), 10)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_ddim_sampler(self, model_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        noise = torch.randn(input_shape).to(device)
+        scheduler = DDIMScheduler(num_train_timesteps=1000)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        sample, intermediates = inferer.sample(
+            input_noise=noise, diffusion_model=model, scheduler=scheduler, save_intermediates=True, intermediate_steps=1
+        )
+        self.assertEqual(len(intermediates), 10)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sampler_conditioned(self, model_params, input_shape):
+        model_params["with_conditioning"] = True
+        model_params["cross_attention_dim"] = 3
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        noise = torch.randn(input_shape).to(device)
+        scheduler = DDIMScheduler(num_train_timesteps=1000)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        conditioning = torch.randn([input_shape[0], 1, 3]).to(device)
+        sample, intermediates = inferer.sample(
+            input_noise=noise,
+            diffusion_model=model,
+            scheduler=scheduler,
+            save_intermediates=True,
+            intermediate_steps=1,
+            conditioning=conditioning,
+        )
+        self.assertEqual(len(intermediates), 10)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_get_likelihood(self, model_params, input_shape):
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        input = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        likelihood, intermediates = inferer.get_likelihood(
+            inputs=input, diffusion_model=model, scheduler=scheduler, save_intermediates=True
+        )
+        self.assertEqual(intermediates[0].shape, input.shape)
+        self.assertEqual(likelihood.shape[0], input.shape[0])
+
+    @unittest.skipUnless(has_scipy, "Requires scipy library.")
+    def test_normal_cdf(self):
+        from scipy.stats import norm
+
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = DiffusionInferer(scheduler=scheduler)
+
+        x = torch.linspace(-10, 10, 20)
+        cdf_approx = inferer._approx_standard_normal_cdf(x)
+        cdf_true = norm.cdf(x)
+        torch.testing.assert_allclose(cdf_approx, cdf_true, atol=1e-3, rtol=1e-5)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sampler_conditioned_concat(self, model_params, input_shape):
+        # copy the model_params dict to prevent from modifying test cases
+        model_params = model_params.copy()
+        n_concat_channel = 2
+        model_params["in_channels"] = model_params["in_channels"] + n_concat_channel
+        model_params["cross_attention_dim"] = None
+        model_params["with_conditioning"] = False
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        noise = torch.randn(input_shape).to(device)
+        conditioning_shape = list(input_shape)
+        conditioning_shape[1] = n_concat_channel
+        conditioning = torch.randn(conditioning_shape).to(device)
+        scheduler = DDIMScheduler(num_train_timesteps=1000)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        sample, intermediates = inferer.sample(
+            input_noise=noise,
+            diffusion_model=model,
+            scheduler=scheduler,
+            save_intermediates=True,
+            intermediate_steps=1,
+            conditioning=conditioning,
+            mode="concat",
+        )
+        self.assertEqual(len(intermediates), 10)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_call_conditioned_concat(self, model_params, input_shape):
+        # copy the model_params dict to prevent from modifying test cases
+        model_params = model_params.copy()
+        n_concat_channel = 2
+        model_params["in_channels"] = model_params["in_channels"] + n_concat_channel
+        model_params["cross_attention_dim"] = None
+        model_params["with_conditioning"] = False
+        model = DiffusionModelUNet(**model_params)
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        model.to(device)
+        model.eval()
+        input = torch.randn(input_shape).to(device)
+        noise = torch.randn(input_shape).to(device)
+        conditioning_shape = list(input_shape)
+        conditioning_shape[1] = n_concat_channel
+        conditioning = torch.randn(conditioning_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        scheduler.set_timesteps(num_inference_steps=10)
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],), device=input.device).long()
+        sample = inferer(
+            inputs=input, noise=noise, diffusion_model=model, timesteps=timesteps, condition=conditioning, mode="concat"
+        )
+        self.assertEqual(sample.shape, input_shape)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_diffusion_model_unet.py b/tests/test_diffusion_model_unet.py
new file mode 100644
index 0000000000..7f764d85de
--- /dev/null
+++ b/tests/test_diffusion_model_unet.py
@@ -0,0 +1,585 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import os
+import tempfile
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.apps import download_url
+from monai.networks import eval_mode
+from monai.networks.nets import DiffusionModelUNet
+from monai.utils import optional_import
+from tests.utils import skip_if_downloading_fails, testing_data_config
+
+_, has_einops = optional_import("einops")
+
+UNCOND_CASES_2D = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": (1, 1, 2),
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, True, True),
+            "num_head_channels": (0, 2, 4),
+            "norm_num_groups": 8,
+        }
+    ],
+]
+
+UNCOND_CASES_3D = [
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": (0, 0, 4),
+            "norm_num_groups": 8,
+        }
+    ],
+]
+
+COND_CASES_2D = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "resblock_updown": True,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "upcast_attention": True,
+        }
+    ],
+]
+
+DROPOUT_OK = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "dropout_cattn": 0.25,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+        }
+    ],
+]
+
+DROPOUT_WRONG = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "dropout_cattn": 3.0,
+        }
+    ]
+]
+
+
+class TestDiffusionModelUNet2D(unittest.TestCase):
+    @parameterized.expand(UNCOND_CASES_2D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_unconditioned_models(self, input_param):
+        net = DiffusionModelUNet(**input_param)
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, 1, 16, 16)), torch.randint(0, 1000, (1,)).long())
+            self.assertEqual(result.shape, (1, 1, 16, 16))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_timestep_with_wrong_shape(self):
+        net = DiffusionModelUNet(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, False),
+            norm_num_groups=8,
+        )
+        with self.assertRaises(ValueError):
+            with eval_mode(net):
+                net.forward(torch.rand((1, 1, 16, 16)), torch.randint(0, 1000, (1, 1)).long())
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_with_different_in_channel_out_channel(self):
+        in_channels = 6
+        out_channels = 3
+        net = DiffusionModelUNet(
+            spatial_dims=2,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, False),
+            norm_num_groups=8,
+        )
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, in_channels, 16, 16)), torch.randint(0, 1000, (1,)).long())
+            self.assertEqual(result.shape, (1, out_channels, 16, 16))
+
+    def test_model_channels_not_multiple_of_norm_num_group(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNet(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 12),
+                attention_levels=(False, False, False),
+                norm_num_groups=8,
+            )
+
+    def test_attention_levels_with_different_length_num_head_channels(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNet(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 8),
+                attention_levels=(False, False, False),
+                num_head_channels=(0, 2),
+                norm_num_groups=8,
+            )
+
+    def test_num_res_blocks_with_different_length_channels(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNet(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=(1, 1),
+                channels=(8, 8, 8),
+                attention_levels=(False, False, False),
+                norm_num_groups=8,
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models(self):
+        net = DiffusionModelUNet(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            with_conditioning=True,
+            transformer_num_layers=1,
+            cross_attention_dim=3,
+            norm_num_groups=8,
+            num_head_channels=8,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 32)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                context=torch.rand((1, 1, 3)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 32))
+
+    def test_with_conditioning_cross_attention_dim_none(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNet(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 8),
+                attention_levels=(False, False, True),
+                with_conditioning=True,
+                transformer_num_layers=1,
+                cross_attention_dim=None,
+                norm_num_groups=8,
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_context_with_conditioning_none(self):
+        net = DiffusionModelUNet(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            with_conditioning=False,
+            transformer_num_layers=1,
+            norm_num_groups=8,
+        )
+
+        with self.assertRaises(ValueError):
+            with eval_mode(net):
+                net.forward(
+                    x=torch.rand((1, 1, 16, 32)),
+                    timesteps=torch.randint(0, 1000, (1,)).long(),
+                    context=torch.rand((1, 1, 3)),
+                )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models_class_conditioning(self):
+        net = DiffusionModelUNet(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            norm_num_groups=8,
+            num_head_channels=8,
+            num_class_embeds=2,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 32)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                class_labels=torch.randint(0, 2, (1,)).long(),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 32))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_conditioned_models_no_class_labels(self):
+        net = DiffusionModelUNet(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            norm_num_groups=8,
+            num_head_channels=8,
+            num_class_embeds=2,
+        )
+
+        with self.assertRaises(ValueError):
+            net.forward(x=torch.rand((1, 1, 16, 32)), timesteps=torch.randint(0, 1000, (1,)).long())
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_model_channels_not_same_size_of_attention_levels(self):
+        with self.assertRaises(ValueError):
+            DiffusionModelUNet(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 8),
+                attention_levels=(False, False),
+                norm_num_groups=8,
+                num_head_channels=8,
+                num_class_embeds=2,
+            )
+
+    @parameterized.expand(COND_CASES_2D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_conditioned_2d_models_shape(self, input_param):
+        net = DiffusionModelUNet(**input_param)
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, 1, 16, 16)), torch.randint(0, 1000, (1,)).long(), torch.rand((1, 1, 3)))
+            self.assertEqual(result.shape, (1, 1, 16, 16))
+
+
+class TestDiffusionModelUNet3D(unittest.TestCase):
+    @parameterized.expand(UNCOND_CASES_3D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_unconditioned_models(self, input_param):
+        net = DiffusionModelUNet(**input_param)
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, 1, 16, 16, 16)), torch.randint(0, 1000, (1,)).long())
+            self.assertEqual(result.shape, (1, 1, 16, 16, 16))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_with_different_in_channel_out_channel(self):
+        in_channels = 6
+        out_channels = 3
+        net = DiffusionModelUNet(
+            spatial_dims=3,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            norm_num_groups=4,
+        )
+        with eval_mode(net):
+            result = net.forward(torch.rand((1, in_channels, 16, 16, 16)), torch.randint(0, 1000, (1,)).long())
+            self.assertEqual(result.shape, (1, out_channels, 16, 16, 16))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models(self):
+        net = DiffusionModelUNet(
+            spatial_dims=3,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(16, 16, 16),
+            attention_levels=(False, False, True),
+            norm_num_groups=16,
+            with_conditioning=True,
+            transformer_num_layers=1,
+            cross_attention_dim=3,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 16, 16)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                context=torch.rand((1, 1, 3)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 16, 16))
+
+    # Test dropout specification for cross-attention blocks
+    @parameterized.expand(DROPOUT_WRONG)
+    def test_wrong_dropout(self, input_param):
+        with self.assertRaises(ValueError):
+            _ = DiffusionModelUNet(**input_param)
+
+    @parameterized.expand(DROPOUT_OK)
+    @skipUnless(has_einops, "Requires einops")
+    def test_right_dropout(self, input_param):
+        _ = DiffusionModelUNet(**input_param)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_compatibility_with_monai_generative(self):
+        # test loading weights from a model saved in MONAI Generative, version 0.2.3
+        with skip_if_downloading_fails():
+            net = DiffusionModelUNet(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 8),
+                attention_levels=(False, False, True),
+                with_conditioning=True,
+                cross_attention_dim=3,
+                transformer_num_layers=1,
+                norm_num_groups=8,
+            )
+
+            tmpdir = tempfile.mkdtemp()
+            key = "diffusion_model_unet_monai_generative_weights"
+            url = testing_data_config("models", key, "url")
+            hash_type = testing_data_config("models", key, "hash_type")
+            hash_val = testing_data_config("models", key, "hash_val")
+            filename = "diffusion_model_unet_monai_generative_weights.pt"
+
+            weight_path = os.path.join(tmpdir, filename)
+            download_url(url=url, filepath=weight_path, hash_val=hash_val, hash_type=hash_type)
+
+            net.load_old_state_dict(torch.load(weight_path), verbose=False)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_ensure_channel_first.py b/tests/test_ensure_channel_first.py
index 0c9ad5869e..fe046a4cdf 100644
--- a/tests/test_ensure_channel_first.py
+++ b/tests/test_ensure_channel_first.py
@@ -50,9 +50,10 @@ class TestEnsureChannelFirst(unittest.TestCase):
     @parameterized.expand([TEST_CASE_1, TEST_CASE_2, TEST_CASE_3, TEST_CASE_4, TEST_CASE_5, TEST_CASE_6])
     @unittest.skipUnless(has_itk, "itk not installed")
     def test_load_nifti(self, input_param, filenames, original_channel_dim):
-        if original_channel_dim is None:
-            test_image = np.random.rand(8, 8, 8)
-        elif original_channel_dim == -1:
+        # if original_channel_dim is None
+        test_image = np.random.rand(8, 8, 8)
+
+        if original_channel_dim == -1:
             test_image = np.random.rand(8, 8, 8, 1)
 
         with tempfile.TemporaryDirectory() as tempdir:
diff --git a/tests/test_ensure_channel_firstd.py b/tests/test_ensure_channel_firstd.py
index 63a437894b..e9effad951 100644
--- a/tests/test_ensure_channel_firstd.py
+++ b/tests/test_ensure_channel_firstd.py
@@ -35,9 +35,10 @@ class TestEnsureChannelFirstd(unittest.TestCase):
 
     @parameterized.expand([TEST_CASE_1, TEST_CASE_2, TEST_CASE_3])
     def test_load_nifti(self, input_param, filenames, original_channel_dim):
-        if original_channel_dim is None:
-            test_image = np.random.rand(8, 8, 8)
-        elif original_channel_dim == -1:
+        # if original_channel_dim is None:
+        test_image = np.random.rand(8, 8, 8)
+
+        if original_channel_dim == -1:
             test_image = np.random.rand(8, 8, 8, 1)
 
         with tempfile.TemporaryDirectory() as tempdir:
diff --git a/tests/test_evenly_divisible_all_gather_dist.py b/tests/test_evenly_divisible_all_gather_dist.py
index d6d26c7e23..f1d45ba48f 100644
--- a/tests/test_evenly_divisible_all_gather_dist.py
+++ b/tests/test_evenly_divisible_all_gather_dist.py
@@ -27,10 +27,10 @@ def test_data(self):
         self._run()
 
     def _run(self):
-        if dist.get_rank() == 0:
-            data1 = torch.tensor([[1, 2], [3, 4]])
-            data2 = torch.tensor([[1.0, 2.0]])
-            data3 = torch.tensor(7)
+        # if dist.get_rank() == 0
+        data1 = torch.tensor([[1, 2], [3, 4]])
+        data2 = torch.tensor([[1.0, 2.0]])
+        data3 = torch.tensor(7)
 
         if dist.get_rank() == 1:
             data1 = torch.tensor([[5, 6]])
diff --git a/tests/test_handler_metrics_saver_dist.py b/tests/test_handler_metrics_saver_dist.py
index 46c9ad27d7..2e12b08aa9 100644
--- a/tests/test_handler_metrics_saver_dist.py
+++ b/tests/test_handler_metrics_saver_dist.py
@@ -51,8 +51,10 @@ def _val_func(engine, batch):
 
         engine = Engine(_val_func)
 
+        # define here to ensure symbol always exists regardless of the following if conditions
+        data = [{PostFix.meta("image"): {"filename_or_obj": [fnames[0]]}}]
+
         if my_rank == 0:
-            data = [{PostFix.meta("image"): {"filename_or_obj": [fnames[0]]}}]
 
             @engine.on(Events.EPOCH_COMPLETED)
             def _save_metrics0(engine):
diff --git a/tests/test_hilbert_transform.py b/tests/test_hilbert_transform.py
index 879a74969d..b91ba3f6b7 100644
--- a/tests/test_hilbert_transform.py
+++ b/tests/test_hilbert_transform.py
@@ -19,11 +19,11 @@
 
 from monai.networks.layers import HilbertTransform
 from monai.utils import OptionalImportError
-from tests.utils import SkipIfModule, SkipIfNoModule, skip_if_no_cuda
+from tests.utils import SkipIfModule, SkipIfNoModule
 
 
 def create_expected_numpy_output(input_datum, **kwargs):
-    x = np.fft.fft(input_datum.cpu().numpy() if input_datum.device.type == "cuda" else input_datum.numpy(), **kwargs)
+    x = np.fft.fft(input_datum.cpu().numpy(), **kwargs)
     f = np.fft.fftfreq(x.shape[kwargs["axis"]])
     u = np.heaviside(f, 0.5)
     new_dims_before = kwargs["axis"]
@@ -44,19 +44,15 @@ def create_expected_numpy_output(input_datum, **kwargs):
 # CPU TEST DATA
 
 cpu_input_data = {}
-cpu_input_data["1D"] = torch.as_tensor(hann_windowed_sine, device=cpu).unsqueeze(0).unsqueeze(0)
-cpu_input_data["2D"] = (
-    torch.as_tensor(np.stack([hann_windowed_sine] * 10, axis=1), device=cpu).unsqueeze(0).unsqueeze(0)
-)
-cpu_input_data["3D"] = (
-    torch.as_tensor(np.stack([np.stack([hann_windowed_sine] * 10, axis=1)] * 10, axis=2), device=cpu)
-    .unsqueeze(0)
-    .unsqueeze(0)
-)
-cpu_input_data["1D 2CH"] = torch.as_tensor(np.stack([hann_windowed_sine] * 10, axis=1), device=cpu).unsqueeze(0)
+cpu_input_data["1D"] = torch.as_tensor(hann_windowed_sine, device=cpu)[None, None]
+cpu_input_data["2D"] = torch.as_tensor(np.stack([hann_windowed_sine] * 10, axis=1), device=cpu)[None, None]
+cpu_input_data["3D"] = torch.as_tensor(
+    np.stack([np.stack([hann_windowed_sine] * 10, axis=1)] * 10, axis=2), device=cpu
+)[None, None]
+cpu_input_data["1D 2CH"] = torch.as_tensor(np.stack([hann_windowed_sine] * 10, axis=1), device=cpu)[None]
 cpu_input_data["2D 2CH"] = torch.as_tensor(
     np.stack([np.stack([hann_windowed_sine] * 10, axis=1)] * 10, axis=2), device=cpu
-).unsqueeze(0)
+)[None]
 
 # SINGLE-CHANNEL CPU VALUE TESTS
 
@@ -97,64 +93,21 @@ def create_expected_numpy_output(input_datum, **kwargs):
     1e-5,  # absolute tolerance
 ]
 
+TEST_CASES_CPU = [
+    TEST_CASE_1D_SINE_CPU,
+    TEST_CASE_2D_SINE_CPU,
+    TEST_CASE_3D_SINE_CPU,
+    TEST_CASE_1D_2CH_SINE_CPU,
+    TEST_CASE_2D_2CH_SINE_CPU,
+]
+
 # GPU TEST DATA
 
 if torch.cuda.is_available():
     gpu = torch.device("cuda")
-
-    gpu_input_data = {}
-    gpu_input_data["1D"] = torch.as_tensor(hann_windowed_sine, device=gpu).unsqueeze(0).unsqueeze(0)
-    gpu_input_data["2D"] = (
-        torch.as_tensor(np.stack([hann_windowed_sine] * 10, axis=1), device=gpu).unsqueeze(0).unsqueeze(0)
-    )
-    gpu_input_data["3D"] = (
-        torch.as_tensor(np.stack([np.stack([hann_windowed_sine] * 10, axis=1)] * 10, axis=2), device=gpu)
-        .unsqueeze(0)
-        .unsqueeze(0)
-    )
-    gpu_input_data["1D 2CH"] = torch.as_tensor(np.stack([hann_windowed_sine] * 10, axis=1), device=gpu).unsqueeze(0)
-    gpu_input_data["2D 2CH"] = torch.as_tensor(
-        np.stack([np.stack([hann_windowed_sine] * 10, axis=1)] * 10, axis=2), device=gpu
-    ).unsqueeze(0)
-
-    # SINGLE CHANNEL GPU VALUE TESTS
-
-    TEST_CASE_1D_SINE_GPU = [
-        {},  # args (empty, so use default)
-        gpu_input_data["1D"],  # Input data: Random 1D signal
-        create_expected_numpy_output(gpu_input_data["1D"], axis=2),  # Expected output: FFT of signal
-        1e-5,  # absolute tolerance
-    ]
-
-    TEST_CASE_2D_SINE_GPU = [
-        {},  # args (empty, so use default)
-        gpu_input_data["2D"],  # Input data: Random 1D signal
-        create_expected_numpy_output(gpu_input_data["2D"], axis=2),  # Expected output: FFT of signal
-        1e-5,  # absolute tolerance
-    ]
-
-    TEST_CASE_3D_SINE_GPU = [
-        {},  # args (empty, so use default)
-        gpu_input_data["3D"],  # Input data: Random 1D signal
-        create_expected_numpy_output(gpu_input_data["3D"], axis=2),  # Expected output: FFT of signal
-        1e-5,  # absolute tolerance
-    ]
-
-    # MULTICHANNEL GPU VALUE TESTS, PROCESS ALONG FIRST SPATIAL AXIS
-
-    TEST_CASE_1D_2CH_SINE_GPU = [
-        {},  # args (empty, so use default)
-        gpu_input_data["1D 2CH"],  # Input data: Random 1D signal
-        create_expected_numpy_output(gpu_input_data["1D 2CH"], axis=2),
-        1e-5,  # absolute tolerance
-    ]
-
-    TEST_CASE_2D_2CH_SINE_GPU = [
-        {},  # args (empty, so use default)
-        gpu_input_data["2D 2CH"],  # Input data: Random 1D signal
-        create_expected_numpy_output(gpu_input_data["2D 2CH"], axis=2),
-        1e-5,  # absolute tolerance
-    ]
+    TEST_CASES_GPU = [[args, image.to(gpu), exp_data, atol] for args, image, exp_data, atol in TEST_CASES_CPU]
+else:
+    TEST_CASES_GPU = []
 
 # TESTS CHECKING PADDING, AXIS SELECTION ETC ARE COVERED BY test_detect_envelope.py
 
@@ -162,42 +115,10 @@ def create_expected_numpy_output(input_datum, **kwargs):
 @SkipIfNoModule("torch.fft")
 class TestHilbertTransformCPU(unittest.TestCase):
 
-    @parameterized.expand(
-        [
-            TEST_CASE_1D_SINE_CPU,
-            TEST_CASE_2D_SINE_CPU,
-            TEST_CASE_3D_SINE_CPU,
-            TEST_CASE_1D_2CH_SINE_CPU,
-            TEST_CASE_2D_2CH_SINE_CPU,
-        ]
-    )
-    def test_value(self, arguments, image, expected_data, atol):
-        result = HilbertTransform(**arguments)(image)
-        result = result.squeeze(0).squeeze(0).numpy()
-        np.testing.assert_allclose(result, expected_data.squeeze(), atol=atol)
-
-
-@skip_if_no_cuda
-@SkipIfNoModule("torch.fft")
-class TestHilbertTransformGPU(unittest.TestCase):
-
-    @parameterized.expand(
-        (
-            []
-            if not torch.cuda.is_available()
-            else [
-                TEST_CASE_1D_SINE_GPU,
-                TEST_CASE_2D_SINE_GPU,
-                TEST_CASE_3D_SINE_GPU,
-                TEST_CASE_1D_2CH_SINE_GPU,
-                TEST_CASE_2D_2CH_SINE_GPU,
-            ]
-        ),
-        skip_on_empty=True,
-    )
+    @parameterized.expand(TEST_CASES_CPU + TEST_CASES_GPU)
     def test_value(self, arguments, image, expected_data, atol):
         result = HilbertTransform(**arguments)(image)
-        result = result.squeeze(0).squeeze(0).cpu().numpy()
+        result = np.squeeze(result.cpu().numpy())
         np.testing.assert_allclose(result, expected_data.squeeze(), atol=atol)
 
 
diff --git a/tests/test_integration_unet_2d.py b/tests/test_integration_unet_2d.py
index 918190775c..3b40682de0 100644
--- a/tests/test_integration_unet_2d.py
+++ b/tests/test_integration_unet_2d.py
@@ -35,6 +35,7 @@ def __getitem__(self, _unused_id):
         def __len__(self):
             return train_steps
 
+    net = None
     if net_name == "basicunet":
         net = BasicUNet(spatial_dims=2, in_channels=1, out_channels=1, features=(4, 8, 8, 16, 16, 32))
     elif net_name == "unet":
diff --git a/tests/test_integration_workflows_adversarial.py b/tests/test_integration_workflows_adversarial.py
new file mode 100644
index 0000000000..f323fc9917
--- /dev/null
+++ b/tests/test_integration_workflows_adversarial.py
@@ -0,0 +1,173 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import os
+import shutil
+import tempfile
+import unittest
+from glob import glob
+
+import numpy as np
+import torch
+
+import monai
+from monai.data import create_test_image_2d
+from monai.engines import AdversarialTrainer
+from monai.handlers import CheckpointSaver, StatsHandler, TensorBoardStatsHandler
+from monai.networks.nets import AutoEncoder, Discriminator
+from monai.transforms import Compose, EnsureChannelFirstd, LoadImaged, RandFlipd, ScaleIntensityd
+from monai.utils import AdversarialKeys as Keys
+from monai.utils import CommonKeys, optional_import, set_determinism
+from tests.utils import DistTestCase, TimedCall, skip_if_quick
+
+nib, has_nibabel = optional_import("nibabel")
+
+
+def run_training_test(root_dir, device="cuda:0"):
+    learning_rate = 2e-4
+    real_label = 1
+    fake_label = 0
+
+    real_images = sorted(glob(os.path.join(root_dir, "img*.nii.gz")))
+    train_files = [{CommonKeys.IMAGE: img, CommonKeys.LABEL: img} for img in zip(real_images)]
+
+    # prepare real data
+    train_transforms = Compose(
+        [
+            LoadImaged(keys=[CommonKeys.IMAGE, CommonKeys.LABEL]),
+            EnsureChannelFirstd(keys=[CommonKeys.IMAGE, CommonKeys.LABEL], channel_dim=2),
+            ScaleIntensityd(keys=[CommonKeys.IMAGE]),
+            RandFlipd(keys=[CommonKeys.IMAGE, CommonKeys.LABEL], prob=0.5),
+        ]
+    )
+    train_ds = monai.data.CacheDataset(data=train_files, transform=train_transforms, cache_rate=0.5)
+    train_loader = monai.data.DataLoader(train_ds, batch_size=2, shuffle=True, num_workers=4)
+
+    # Create Discriminator
+    discriminator_net = Discriminator(
+        in_shape=(1, 64, 64), channels=(8, 16, 32, 64, 1), strides=(2, 2, 2, 2, 1), num_res_units=1, kernel_size=5
+    ).to(device)
+    discriminator_opt = torch.optim.Adam(discriminator_net.parameters(), learning_rate)
+    discriminator_loss_criterion = torch.nn.BCELoss()
+
+    def discriminator_loss(real_logits, fake_logits):
+        real_target = real_logits.new_full((real_logits.shape[0], 1), real_label)
+        fake_target = fake_logits.new_full((fake_logits.shape[0], 1), fake_label)
+        real_loss = discriminator_loss_criterion(real_logits, real_target)
+        fake_loss = discriminator_loss_criterion(fake_logits.detach(), fake_target)
+        return torch.div(torch.add(real_loss, fake_loss), 2)
+
+    # Create Generator
+    generator_network = AutoEncoder(
+        spatial_dims=2,
+        in_channels=1,
+        out_channels=1,
+        channels=(8, 16, 32, 64),
+        strides=(2, 2, 2, 2),
+        num_res_units=1,
+        num_inter_units=1,
+    )
+    generator_network = generator_network.to(device)
+    generator_optimiser = torch.optim.Adam(generator_network.parameters(), learning_rate)
+    generator_loss_criterion = torch.nn.MSELoss()
+
+    def reconstruction_loss(recon_images, real_images):
+        return generator_loss_criterion(recon_images, real_images)
+
+    def generator_loss(fake_logits):
+        fake_target = fake_logits.new_full((fake_logits.shape[0], 1), real_label)
+        recon_loss = discriminator_loss_criterion(fake_logits.detach(), fake_target)
+        return recon_loss
+
+    key_train_metric = None
+
+    train_handlers = [
+        StatsHandler(
+            name="training_loss",
+            output_transform=lambda x: {
+                Keys.RECONSTRUCTION_LOSS: x[Keys.RECONSTRUCTION_LOSS],
+                Keys.DISCRIMINATOR_LOSS: x[Keys.DISCRIMINATOR_LOSS],
+                Keys.GENERATOR_LOSS: x[Keys.GENERATOR_LOSS],
+            },
+        ),
+        TensorBoardStatsHandler(
+            log_dir=root_dir,
+            tag_name="training_loss",
+            output_transform=lambda x: {
+                Keys.RECONSTRUCTION_LOSS: x[Keys.RECONSTRUCTION_LOSS],
+                Keys.DISCRIMINATOR_LOSS: x[Keys.DISCRIMINATOR_LOSS],
+                Keys.GENERATOR_LOSS: x[Keys.GENERATOR_LOSS],
+            },
+        ),
+        CheckpointSaver(
+            save_dir=root_dir,
+            save_dict={"g_net": generator_network, "d_net": discriminator_net},
+            save_interval=2,
+            epoch_level=True,
+        ),
+    ]
+
+    num_epochs = 5
+
+    trainer = AdversarialTrainer(
+        device=device,
+        max_epochs=num_epochs,
+        train_data_loader=train_loader,
+        g_network=generator_network,
+        g_optimizer=generator_optimiser,
+        g_loss_function=generator_loss,
+        recon_loss_function=reconstruction_loss,
+        d_network=discriminator_net,
+        d_optimizer=discriminator_opt,
+        d_loss_function=discriminator_loss,
+        non_blocking=True,
+        key_train_metric=key_train_metric,
+        train_handlers=train_handlers,
+    )
+    trainer.run()
+
+    return trainer.state
+
+
+@skip_if_quick
+@unittest.skipUnless(has_nibabel, "Requires nibabel library.")
+class IntegrationWorkflowsAdversarialTrainer(DistTestCase):
+    def setUp(self):
+        set_determinism(seed=0)
+
+        self.data_dir = tempfile.mkdtemp()
+        for i in range(40):
+            im, _ = create_test_image_2d(64, 64, num_objs=3, rad_max=14, num_seg_classes=1, channel_dim=-1)
+            n = nib.Nifti1Image(im, np.eye(4))
+            nib.save(n, os.path.join(self.data_dir, f"img{i:d}.nii.gz"))
+
+        self.device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu:0")
+        monai.config.print_config()
+
+    def tearDown(self):
+        set_determinism(seed=None)
+        shutil.rmtree(self.data_dir)
+
+    @TimedCall(seconds=200, daemon=False)
+    def test_training(self):
+        torch.manual_seed(0)
+
+        finish_state = run_training_test(self.data_dir, device=self.device)
+
+        # Assert AdversarialTrainer training finished
+        self.assertEqual(finish_state.iteration, 100)
+        self.assertEqual(finish_state.epoch, 5)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_latent_diffusion_inferer.py b/tests/test_latent_diffusion_inferer.py
new file mode 100644
index 0000000000..2e04ad6c5c
--- /dev/null
+++ b/tests/test_latent_diffusion_inferer.py
@@ -0,0 +1,824 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.inferers import LatentDiffusionInferer
+from monai.networks.nets import VQVAE, AutoencoderKL, DiffusionModelUNet, SPADEAutoencoderKL, SPADEDiffusionModelUNet
+from monai.networks.schedulers import DDPMScheduler
+from monai.utils import optional_import
+
+_, has_einops = optional_import("einops")
+TEST_CASES = [
+    [
+        "AutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        (1, 1, 8, 8),
+        (1, 3, 4, 4),
+    ],
+    [
+        "VQVAE",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [4, 4],
+            "num_res_layers": 1,
+            "num_res_channels": [4, 4],
+            "downsample_parameters": ((2, 4, 1, 1), (2, 4, 1, 1)),
+            "upsample_parameters": ((2, 4, 1, 1, 0), (2, 4, 1, 1, 0)),
+            "num_embeddings": 16,
+            "embedding_dim": 3,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [8, 8],
+            "norm_num_groups": 8,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        (1, 1, 16, 16),
+        (1, 3, 4, 4),
+    ],
+    [
+        "VQVAE",
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [4, 4],
+            "num_res_layers": 1,
+            "num_res_channels": [4, 4],
+            "downsample_parameters": ((2, 4, 1, 1), (2, 4, 1, 1)),
+            "upsample_parameters": ((2, 4, 1, 1, 0), (2, 4, 1, 1, 0)),
+            "num_embeddings": 16,
+            "embedding_dim": 3,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 3,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [8, 8],
+            "norm_num_groups": 8,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        (1, 1, 16, 16, 16),
+        (1, 3, 4, 4, 4),
+    ],
+    [
+        "AutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "SPADEDiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        (1, 1, 8, 8),
+        (1, 3, 4, 4),
+    ],
+]
+TEST_CASES_DIFF_SHAPES = [
+    [
+        "AutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        (1, 1, 12, 12),
+        (1, 3, 8, 8),
+    ],
+    [
+        "VQVAE",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [4, 4],
+            "num_res_layers": 1,
+            "num_res_channels": [4, 4],
+            "downsample_parameters": ((2, 4, 1, 1), (2, 4, 1, 1)),
+            "upsample_parameters": ((2, 4, 1, 1, 0), (2, 4, 1, 1, 0)),
+            "num_embeddings": 16,
+            "embedding_dim": 3,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [8, 8],
+            "norm_num_groups": 8,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        (1, 1, 12, 12),
+        (1, 3, 8, 8),
+    ],
+    [
+        "VQVAE",
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [4, 4],
+            "num_res_layers": 1,
+            "num_res_channels": [4, 4],
+            "downsample_parameters": ((2, 4, 1, 1), (2, 4, 1, 1)),
+            "upsample_parameters": ((2, 4, 1, 1, 0), (2, 4, 1, 1, 0)),
+            "num_embeddings": 16,
+            "embedding_dim": 3,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 3,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [8, 8],
+            "norm_num_groups": 8,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        (1, 1, 12, 12, 12),
+        (1, 3, 8, 8, 8),
+    ],
+    [
+        "SPADEAutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "DiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        (1, 1, 8, 8),
+        (1, 3, 4, 4),
+    ],
+    [
+        "AutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "SPADEDiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        (1, 1, 8, 8),
+        (1, 3, 4, 4),
+    ],
+    [
+        "SPADEAutoencoderKL",
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "latent_channels": 3,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+            "norm_num_groups": 4,
+        },
+        "SPADEDiffusionModelUNet",
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 3,
+            "out_channels": 3,
+            "channels": [4, 4],
+            "norm_num_groups": 4,
+            "attention_levels": [False, False],
+            "num_res_blocks": 1,
+            "num_head_channels": 4,
+        },
+        (1, 1, 8, 8),
+        (1, 3, 4, 4),
+    ],
+]
+
+
+class TestDiffusionSamplingInferer(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_prediction_shape(
+        self, ae_model_type, autoencoder_params, dm_model_type, stage_2_params, input_shape, latent_shape
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+        noise = torch.randn(latent_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = LatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],), device=input.device).long()
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                seg=input_seg,
+                noise=noise,
+                timesteps=timesteps,
+            )
+        else:
+            prediction = inferer(
+                inputs=input, autoencoder_model=stage_1, diffusion_model=stage_2, noise=noise, timesteps=timesteps
+            )
+        self.assertEqual(prediction.shape, latent_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_shape(
+        self, ae_model_type, autoencoder_params, dm_model_type, stage_2_params, input_shape, latent_shape
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        noise = torch.randn(latent_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = LatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if ae_model_type == "SPADEAutoencoderKL" or dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                seg=input_seg,
+            )
+        else:
+            sample = inferer.sample(
+                input_noise=noise, autoencoder_model=stage_1, diffusion_model=stage_2, scheduler=scheduler
+            )
+        self.assertEqual(sample.shape, input_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_intermediates(
+        self, ae_model_type, autoencoder_params, dm_model_type, stage_2_params, input_shape, latent_shape
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        noise = torch.randn(latent_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = LatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if ae_model_type == "SPADEAutoencoderKL" or dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample, intermediates = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                seg=input_seg,
+                save_intermediates=True,
+                intermediate_steps=1,
+            )
+        else:
+            sample, intermediates = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                save_intermediates=True,
+                intermediate_steps=1,
+            )
+        self.assertEqual(len(intermediates), 10)
+        self.assertEqual(intermediates[0].shape, input_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_get_likelihoods(
+        self, ae_model_type, autoencoder_params, dm_model_type, stage_2_params, input_shape, latent_shape
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = LatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample, intermediates = inferer.get_likelihood(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                save_intermediates=True,
+                seg=input_seg,
+            )
+        else:
+            sample, intermediates = inferer.get_likelihood(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                save_intermediates=True,
+            )
+        self.assertEqual(len(intermediates), 10)
+        self.assertEqual(intermediates[0].shape, latent_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_resample_likelihoods(
+        self, ae_model_type, autoencoder_params, dm_model_type, stage_2_params, input_shape, latent_shape
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = LatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample, intermediates = inferer.get_likelihood(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                save_intermediates=True,
+                resample_latent_likelihoods=True,
+                seg=input_seg,
+            )
+        else:
+            sample, intermediates = inferer.get_likelihood(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                save_intermediates=True,
+                resample_latent_likelihoods=True,
+            )
+        self.assertEqual(len(intermediates), 10)
+        self.assertEqual(intermediates[0].shape[2:], input_shape[2:])
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_prediction_shape_conditioned_concat(
+        self, ae_model_type, autoencoder_params, dm_model_type, stage_2_params, input_shape, latent_shape
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        stage_2_params = stage_2_params.copy()
+        n_concat_channel = 3
+        stage_2_params["in_channels"] = stage_2_params["in_channels"] + n_concat_channel
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+        noise = torch.randn(latent_shape).to(device)
+        conditioning_shape = list(latent_shape)
+        conditioning_shape[1] = n_concat_channel
+        conditioning = torch.randn(conditioning_shape).to(device)
+
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = LatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],), device=input.device).long()
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                noise=noise,
+                timesteps=timesteps,
+                condition=conditioning,
+                mode="concat",
+                seg=input_seg,
+            )
+        else:
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                noise=noise,
+                timesteps=timesteps,
+                condition=conditioning,
+                mode="concat",
+            )
+        self.assertEqual(prediction.shape, latent_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_shape_conditioned_concat(
+        self, ae_model_type, autoencoder_params, dm_model_type, stage_2_params, input_shape, latent_shape
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        stage_2_params = stage_2_params.copy()
+        n_concat_channel = 3
+        stage_2_params["in_channels"] = stage_2_params["in_channels"] + n_concat_channel
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        noise = torch.randn(latent_shape).to(device)
+        conditioning_shape = list(latent_shape)
+        conditioning_shape[1] = n_concat_channel
+        conditioning = torch.randn(conditioning_shape).to(device)
+
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = LatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            sample = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                conditioning=conditioning,
+                mode="concat",
+                seg=input_seg,
+            )
+        else:
+            sample = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                conditioning=conditioning,
+                mode="concat",
+            )
+        self.assertEqual(sample.shape, input_shape)
+
+    @parameterized.expand(TEST_CASES_DIFF_SHAPES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_shape_different_latents(
+        self, ae_model_type, autoencoder_params, dm_model_type, stage_2_params, input_shape, latent_shape
+    ):
+        stage_1 = None
+
+        if ae_model_type == "AutoencoderKL":
+            stage_1 = AutoencoderKL(**autoencoder_params)
+        if ae_model_type == "VQVAE":
+            stage_1 = VQVAE(**autoencoder_params)
+        if ae_model_type == "SPADEAutoencoderKL":
+            stage_1 = SPADEAutoencoderKL(**autoencoder_params)
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            stage_2 = SPADEDiffusionModelUNet(**stage_2_params)
+        else:
+            stage_2 = DiffusionModelUNet(**stage_2_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+        noise = torch.randn(latent_shape).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        # We infer the VAE shape
+        autoencoder_latent_shape = [i // (2 ** (len(autoencoder_params["channels"]) - 1)) for i in input_shape[2:]]
+        inferer = LatentDiffusionInferer(
+            scheduler=scheduler,
+            scale_factor=1.0,
+            ldm_latent_shape=list(latent_shape[2:]),
+            autoencoder_latent_shape=autoencoder_latent_shape,
+        )
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],), device=input.device).long()
+
+        if dm_model_type == "SPADEDiffusionModelUNet":
+            input_shape_seg = list(input_shape)
+            if "label_nc" in stage_2_params.keys():
+                input_shape_seg[1] = stage_2_params["label_nc"]
+            else:
+                input_shape_seg[1] = autoencoder_params["label_nc"]
+            input_seg = torch.randn(input_shape_seg).to(device)
+            prediction = inferer(
+                inputs=input,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                noise=noise,
+                timesteps=timesteps,
+                seg=input_seg,
+            )
+        else:
+            prediction = inferer(
+                inputs=input, autoencoder_model=stage_1, diffusion_model=stage_2, noise=noise, timesteps=timesteps
+            )
+        self.assertEqual(prediction.shape, latent_shape)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_incompatible_spade_setup(self):
+        stage_1 = SPADEAutoencoderKL(
+            spatial_dims=2,
+            label_nc=6,
+            in_channels=1,
+            out_channels=1,
+            channels=(4, 4),
+            latent_channels=3,
+            attention_levels=[False, False],
+            num_res_blocks=1,
+            with_encoder_nonlocal_attn=False,
+            with_decoder_nonlocal_attn=False,
+            norm_num_groups=4,
+        )
+        stage_2 = SPADEDiffusionModelUNet(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=3,
+            out_channels=3,
+            channels=[4, 4],
+            norm_num_groups=4,
+            attention_levels=[False, False],
+            num_res_blocks=1,
+            num_head_channels=4,
+        )
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+        noise = torch.randn((1, 3, 4, 4)).to(device)
+        input_seg = torch.randn((1, 3, 8, 8)).to(device)
+        scheduler = DDPMScheduler(num_train_timesteps=10)
+        inferer = LatentDiffusionInferer(scheduler=scheduler, scale_factor=1.0)
+        scheduler.set_timesteps(num_inference_steps=10)
+
+        with self.assertRaises(ValueError):
+            _ = inferer.sample(
+                input_noise=noise,
+                autoencoder_model=stage_1,
+                diffusion_model=stage_2,
+                scheduler=scheduler,
+                seg=input_seg,
+            )
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_ordering.py b/tests/test_ordering.py
new file mode 100644
index 0000000000..e6b235e179
--- /dev/null
+++ b/tests/test_ordering.py
@@ -0,0 +1,289 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import numpy as np
+from parameterized import parameterized
+
+from monai.utils.enums import OrderingTransformations, OrderingType
+from monai.utils.ordering import Ordering
+
+TEST_2D_NON_RANDOM = [
+    [
+        {
+            "ordering_type": OrderingType.RASTER_SCAN,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (),
+            "transpositions_axes": (),
+            "rot90_axes": (),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [0, 1, 2, 3],
+    ],
+    [
+        {
+            "ordering_type": OrderingType.S_CURVE,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (),
+            "transpositions_axes": (),
+            "rot90_axes": (),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [0, 1, 3, 2],
+    ],
+    [
+        {
+            "ordering_type": OrderingType.RASTER_SCAN,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (True, False),
+            "transpositions_axes": (),
+            "rot90_axes": (),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [2, 3, 0, 1],
+    ],
+    [
+        {
+            "ordering_type": OrderingType.S_CURVE,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (True, False),
+            "transpositions_axes": (),
+            "rot90_axes": (),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [2, 3, 1, 0],
+    ],
+    [
+        {
+            "ordering_type": OrderingType.RASTER_SCAN,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (),
+            "transpositions_axes": ((1, 0),),
+            "rot90_axes": (),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [0, 2, 1, 3],
+    ],
+    [
+        {
+            "ordering_type": OrderingType.S_CURVE,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (),
+            "transpositions_axes": ((1, 0),),
+            "rot90_axes": (),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [0, 2, 3, 1],
+    ],
+    [
+        {
+            "ordering_type": OrderingType.RASTER_SCAN,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (),
+            "transpositions_axes": (),
+            "rot90_axes": ((0, 1),),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [1, 3, 0, 2],
+    ],
+    [
+        {
+            "ordering_type": OrderingType.S_CURVE,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (),
+            "transpositions_axes": (),
+            "rot90_axes": ((0, 1),),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [1, 3, 2, 0],
+    ],
+    [
+        {
+            "ordering_type": OrderingType.RASTER_SCAN,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (True, False),
+            "transpositions_axes": ((1, 0),),
+            "rot90_axes": ((0, 1),),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [0, 1, 2, 3],
+    ],
+    [
+        {
+            "ordering_type": OrderingType.S_CURVE,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (True, False),
+            "transpositions_axes": ((1, 0),),
+            "rot90_axes": ((0, 1),),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [0, 1, 3, 2],
+    ],
+]
+
+
+TEST_3D = [
+    [
+        {
+            "ordering_type": OrderingType.RASTER_SCAN,
+            "spatial_dims": 3,
+            "dimensions": (1, 2, 2, 2),
+            "reflected_spatial_dims": (),
+            "transpositions_axes": (),
+            "rot90_axes": (),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        },
+        [0, 1, 2, 3, 4, 5, 6, 7],
+    ]
+]
+
+TEST_ORDERING_TYPE_FAILURE = [
+    [
+        {
+            "ordering_type": "hilbert",
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (True, False),
+            "transpositions_axes": ((1, 0),),
+            "rot90_axes": ((0, 1),),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        }
+    ]
+]
+
+TEST_ORDERING_TRANSFORMATION_FAILURE = [
+    [
+        {
+            "ordering_type": OrderingType.S_CURVE,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (True, False),
+            "transpositions_axes": ((1, 0),),
+            "rot90_axes": ((0, 1),),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                "flip",
+            ),
+        }
+    ]
+]
+
+TEST_REVERT = [
+    [
+        {
+            "ordering_type": OrderingType.S_CURVE,
+            "spatial_dims": 2,
+            "dimensions": (1, 2, 2),
+            "reflected_spatial_dims": (True, False),
+            "transpositions_axes": (),
+            "rot90_axes": (),
+            "transformation_order": (
+                OrderingTransformations.TRANSPOSE.value,
+                OrderingTransformations.ROTATE_90.value,
+                OrderingTransformations.REFLECT.value,
+            ),
+        }
+    ]
+]
+
+
+class TestOrdering(unittest.TestCase):
+    @parameterized.expand(TEST_2D_NON_RANDOM + TEST_3D)
+    def test_ordering(self, input_param, expected_sequence_ordering):
+        ordering = Ordering(**input_param)
+        self.assertTrue(np.array_equal(ordering.get_sequence_ordering(), expected_sequence_ordering, equal_nan=True))
+
+    @parameterized.expand(TEST_ORDERING_TYPE_FAILURE)
+    def test_ordering_type_failure(self, input_param):
+        with self.assertRaises(ValueError):
+            Ordering(**input_param)
+
+    @parameterized.expand(TEST_ORDERING_TRANSFORMATION_FAILURE)
+    def test_ordering_transformation_failure(self, input_param):
+        with self.assertRaises(ValueError):
+            Ordering(**input_param)
+
+    @parameterized.expand(TEST_REVERT)
+    def test_revert(self, input_param):
+        sequence = np.random.randint(0, 100, size=input_param["dimensions"]).flatten()
+
+        ordering = Ordering(**input_param)
+
+        reverted_sequence = sequence[ordering.get_sequence_ordering()]
+        reverted_sequence = reverted_sequence[ordering.get_revert_sequence_ordering()]
+
+        self.assertTrue(np.array_equal(sequence, reverted_sequence, equal_nan=True))
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_patch_gan_dicriminator.py b/tests/test_patch_gan_dicriminator.py
new file mode 100644
index 0000000000..c19898e70d
--- /dev/null
+++ b/tests/test_patch_gan_dicriminator.py
@@ -0,0 +1,179 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.nets import MultiScalePatchDiscriminator, PatchDiscriminator
+from tests.utils import test_script_save
+
+TEST_PATCHGAN = [
+    [
+        {
+            "num_layers_d": 3,
+            "spatial_dims": 2,
+            "channels": 8,
+            "in_channels": 3,
+            "out_channels": 1,
+            "kernel_size": 3,
+            "activation": "LEAKYRELU",
+            "norm": "instance",
+            "bias": False,
+            "dropout": 0.1,
+        },
+        torch.rand([1, 3, 256, 512]),
+        (1, 8, 128, 256),
+        (1, 1, 32, 64),
+    ],
+    [
+        {
+            "num_layers_d": 3,
+            "spatial_dims": 3,
+            "channels": 8,
+            "in_channels": 3,
+            "out_channels": 1,
+            "kernel_size": 3,
+            "activation": "LEAKYRELU",
+            "norm": "instance",
+            "bias": False,
+            "dropout": 0.1,
+        },
+        torch.rand([1, 3, 256, 512, 256]),
+        (1, 8, 128, 256, 128),
+        (1, 1, 32, 64, 32),
+    ],
+]
+
+TEST_MULTISCALE_PATCHGAN = [
+    [
+        {
+            "num_d": 2,
+            "num_layers_d": 3,
+            "spatial_dims": 2,
+            "channels": 8,
+            "in_channels": 3,
+            "out_channels": 1,
+            "kernel_size": 3,
+            "activation": "LEAKYRELU",
+            "norm": "instance",
+            "bias": False,
+            "dropout": 0.1,
+            "minimum_size_im": 256,
+        },
+        torch.rand([1, 3, 256, 512]),
+        [(1, 1, 32, 64), (1, 1, 4, 8)],
+        [4, 7],
+    ],
+    [
+        {
+            "num_d": 2,
+            "num_layers_d": 3,
+            "spatial_dims": 3,
+            "channels": 8,
+            "in_channels": 3,
+            "out_channels": 1,
+            "kernel_size": 3,
+            "activation": "LEAKYRELU",
+            "norm": "instance",
+            "bias": False,
+            "dropout": 0.1,
+            "minimum_size_im": 256,
+        },
+        torch.rand([1, 3, 256, 512, 256]),
+        [(1, 1, 32, 64, 32), (1, 1, 4, 8, 4)],
+        [4, 7],
+    ],
+]
+TEST_TOO_SMALL_SIZE = [
+    {
+        "num_d": 2,
+        "num_layers_d": 6,
+        "spatial_dims": 2,
+        "channels": 8,
+        "in_channels": 3,
+        "out_channels": 1,
+        "kernel_size": 3,
+        "activation": "LEAKYRELU",
+        "norm": "instance",
+        "bias": False,
+        "dropout": 0.1,
+        "minimum_size_im": 256,
+    }
+]
+
+
+class TestPatchGAN(unittest.TestCase):
+    @parameterized.expand(TEST_PATCHGAN)
+    def test_shape(self, input_param, input_data, expected_shape_feature, expected_shape_output):
+        net = PatchDiscriminator(**input_param)
+        with eval_mode(net):
+            result = net.forward(input_data)
+            self.assertEqual(tuple(result[0].shape), expected_shape_feature)
+            self.assertEqual(tuple(result[-1].shape), expected_shape_output)
+
+    def test_script(self):
+        net = PatchDiscriminator(
+            num_layers_d=3,
+            spatial_dims=2,
+            channels=8,
+            in_channels=3,
+            out_channels=1,
+            kernel_size=3,
+            activation="LEAKYRELU",
+            norm="instance",
+            bias=False,
+            dropout=0.1,
+        )
+        i = torch.rand([1, 3, 256, 512])
+        test_script_save(net, i)
+
+
+class TestMultiscalePatchGAN(unittest.TestCase):
+    @parameterized.expand(TEST_MULTISCALE_PATCHGAN)
+    def test_shape(self, input_param, input_data, expected_shape, features_lengths=None):
+        net = MultiScalePatchDiscriminator(**input_param)
+        with eval_mode(net):
+            result, features = net.forward(input_data)
+            for r_ind, r in enumerate(result):
+                self.assertEqual(tuple(r.shape), expected_shape[r_ind])
+            for o_d_ind, o_d in enumerate(features):
+                self.assertEqual(len(o_d), features_lengths[o_d_ind])
+
+    def test_too_small_shape(self):
+        with self.assertRaises(AssertionError):
+            MultiScalePatchDiscriminator(**TEST_TOO_SMALL_SIZE[0])
+
+    def test_script(self):
+        net = MultiScalePatchDiscriminator(
+            num_d=2,
+            num_layers_d=3,
+            spatial_dims=2,
+            channels=8,
+            in_channels=3,
+            out_channels=1,
+            kernel_size=3,
+            activation="LEAKYRELU",
+            norm="instance",
+            bias=False,
+            dropout=0.1,
+            minimum_size_im=256,
+        )
+        i = torch.rand([1, 3, 256, 512])
+        test_script_save(net, i)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_prepare_batch_diffusion.py b/tests/test_prepare_batch_diffusion.py
new file mode 100644
index 0000000000..d969c06368
--- /dev/null
+++ b/tests/test_prepare_batch_diffusion.py
@@ -0,0 +1,104 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.engines import SupervisedEvaluator
+from monai.engines.utils import DiffusionPrepareBatch
+from monai.inferers import DiffusionInferer
+from monai.networks.nets import DiffusionModelUNet
+from monai.networks.schedulers import DDPMScheduler
+
+TEST_CASES = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [8],
+            "norm_num_groups": 8,
+            "attention_levels": [True],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        (2, 1, 8, 8),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": [8],
+            "norm_num_groups": 8,
+            "attention_levels": [True],
+            "num_res_blocks": 1,
+            "num_head_channels": 8,
+        },
+        (2, 1, 8, 8, 8),
+    ],
+]
+
+
+class TestPrepareBatchDiffusion(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_output_sizes(self, input_args, image_size):
+        device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+        dataloader = [{"image": torch.randn(image_size).to(device)}]
+        scheduler = DDPMScheduler(num_train_timesteps=20)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        network = DiffusionModelUNet(**input_args).to(device)
+        evaluator = SupervisedEvaluator(
+            device=device,
+            val_data_loader=dataloader,
+            epoch_length=1,
+            network=network,
+            inferer=inferer,
+            non_blocking=True,
+            prepare_batch=DiffusionPrepareBatch(num_train_timesteps=20),
+            decollate=False,
+        )
+        evaluator.run()
+        output = evaluator.state.output
+        # check shapes are the same
+        self.assertEqual(output["pred"].shape, image_size)
+        self.assertEqual(output["label"].shape, output["image"].shape)
+
+    @parameterized.expand(TEST_CASES)
+    def test_conditioning(self, input_args, image_size):
+        device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+        dataloader = [{"image": torch.randn(image_size).to(device), "context": torch.randn((2, 4, 3)).to(device)}]
+        scheduler = DDPMScheduler(num_train_timesteps=20)
+        inferer = DiffusionInferer(scheduler=scheduler)
+        network = DiffusionModelUNet(**input_args, with_conditioning=True, cross_attention_dim=3).to(device)
+        evaluator = SupervisedEvaluator(
+            device=device,
+            val_data_loader=dataloader,
+            epoch_length=1,
+            network=network,
+            inferer=inferer,
+            non_blocking=True,
+            prepare_batch=DiffusionPrepareBatch(num_train_timesteps=20, condition_name="context"),
+            decollate=False,
+        )
+        evaluator.run()
+        output = evaluator.state.output
+        # check shapes are the same
+        self.assertEqual(output["pred"].shape, image_size)
+        self.assertEqual(output["label"].shape, output["image"].shape)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_reg_loss_integration.py b/tests/test_reg_loss_integration.py
index 1fb81689e6..8afc2da6ad 100644
--- a/tests/test_reg_loss_integration.py
+++ b/tests/test_reg_loss_integration.py
@@ -83,6 +83,9 @@ def forward(self, x):
         # initialize a SGD optimizer
         optimizer = optim.Adam(net.parameters(), lr=learning_rate)
 
+        # declare first for pylint
+        init_loss = None
+
         # train the network
         for it in range(max_iter):
             # set the gradient to zero
diff --git a/tests/test_scheduler_ddim.py b/tests/test_scheduler_ddim.py
new file mode 100644
index 0000000000..1a8f8cab67
--- /dev/null
+++ b/tests/test_scheduler_ddim.py
@@ -0,0 +1,83 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.networks.schedulers import DDIMScheduler
+from tests.utils import assert_allclose
+
+TEST_2D_CASE = []
+for beta_schedule in ["linear_beta", "scaled_linear_beta"]:
+    TEST_2D_CASE.append([{"schedule": beta_schedule}, (2, 6, 16, 16), (2, 6, 16, 16)])
+
+TEST_3D_CASE = []
+for beta_schedule in ["linear_beta", "scaled_linear_beta"]:
+    TEST_3D_CASE.append([{"schedule": beta_schedule}, (2, 6, 16, 16, 16), (2, 6, 16, 16, 16)])
+
+TEST_CASES = TEST_2D_CASE + TEST_3D_CASE
+
+TEST_FULl_LOOP = [
+    [{"schedule": "linear_beta"}, (1, 1, 2, 2), torch.Tensor([[[[-0.9579, -0.6457], [0.4684, -0.9694]]]])]
+]
+
+
+class TestDDPMScheduler(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_add_noise(self, input_param, input_shape, expected_shape):
+        scheduler = DDIMScheduler(**input_param)
+        scheduler.set_timesteps(num_inference_steps=100)
+        original_sample = torch.zeros(input_shape)
+        noise = torch.randn_like(original_sample)
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (original_sample.shape[0],)).long()
+
+        noisy = scheduler.add_noise(original_samples=original_sample, noise=noise, timesteps=timesteps)
+        self.assertEqual(noisy.shape, expected_shape)
+
+    @parameterized.expand(TEST_CASES)
+    def test_step_shape(self, input_param, input_shape, expected_shape):
+        scheduler = DDIMScheduler(**input_param)
+        scheduler.set_timesteps(num_inference_steps=100)
+        model_output = torch.randn(input_shape)
+        sample = torch.randn(input_shape)
+        output_step = scheduler.step(model_output=model_output, timestep=500, sample=sample)
+        self.assertEqual(output_step[0].shape, expected_shape)
+        self.assertEqual(output_step[1].shape, expected_shape)
+
+    @parameterized.expand(TEST_FULl_LOOP)
+    def test_full_timestep_loop(self, input_param, input_shape, expected_output):
+        scheduler = DDIMScheduler(**input_param)
+        scheduler.set_timesteps(50)
+        torch.manual_seed(42)
+        model_output = torch.randn(input_shape)
+        sample = torch.randn(input_shape)
+        for t in range(50):
+            sample, _ = scheduler.step(model_output=model_output, timestep=t, sample=sample)
+        assert_allclose(sample, expected_output, rtol=1e-3, atol=1e-3)
+
+    def test_set_timesteps(self):
+        scheduler = DDIMScheduler(num_train_timesteps=1000)
+        scheduler.set_timesteps(num_inference_steps=100)
+        self.assertEqual(scheduler.num_inference_steps, 100)
+        self.assertEqual(len(scheduler.timesteps), 100)
+
+    def test_set_timesteps_with_num_inference_steps_bigger_than_num_train_timesteps(self):
+        scheduler = DDIMScheduler(num_train_timesteps=1000)
+        with self.assertRaises(ValueError):
+            scheduler.set_timesteps(num_inference_steps=2000)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_scheduler_ddpm.py b/tests/test_scheduler_ddpm.py
new file mode 100644
index 0000000000..f0447aded2
--- /dev/null
+++ b/tests/test_scheduler_ddpm.py
@@ -0,0 +1,104 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.networks.schedulers import DDPMScheduler
+from tests.utils import assert_allclose
+
+TEST_2D_CASE = []
+for beta_schedule in ["linear_beta", "scaled_linear_beta"]:
+    for variance_type in ["fixed_small", "fixed_large"]:
+        TEST_2D_CASE.append(
+            [{"schedule": beta_schedule, "variance_type": variance_type}, (2, 6, 16, 16), (2, 6, 16, 16)]
+        )
+
+TEST_3D_CASE = []
+for beta_schedule in ["linear_beta", "scaled_linear_beta"]:
+    for variance_type in ["fixed_small", "fixed_large"]:
+        TEST_3D_CASE.append(
+            [{"schedule": beta_schedule, "variance_type": variance_type}, (2, 6, 16, 16, 16), (2, 6, 16, 16, 16)]
+        )
+
+TEST_CASES = TEST_2D_CASE + TEST_3D_CASE
+
+TEST_FULl_LOOP = [
+    [{"schedule": "linear_beta"}, (1, 1, 2, 2), torch.Tensor([[[[-1.0153, -0.3218], [0.8454, -0.7870]]]])]
+]
+
+
+class TestDDPMScheduler(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_add_noise(self, input_param, input_shape, expected_shape):
+        scheduler = DDPMScheduler(**input_param)
+        original_sample = torch.zeros(input_shape)
+        noise = torch.randn_like(original_sample)
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (original_sample.shape[0],)).long()
+
+        noisy = scheduler.add_noise(original_samples=original_sample, noise=noise, timesteps=timesteps)
+        self.assertEqual(noisy.shape, expected_shape)
+
+    @parameterized.expand(TEST_CASES)
+    def test_step_shape(self, input_param, input_shape, expected_shape):
+        scheduler = DDPMScheduler(**input_param)
+        model_output = torch.randn(input_shape)
+        sample = torch.randn(input_shape)
+        output_step = scheduler.step(model_output=model_output, timestep=500, sample=sample)
+        self.assertEqual(output_step[0].shape, expected_shape)
+        self.assertEqual(output_step[1].shape, expected_shape)
+
+    @parameterized.expand(TEST_FULl_LOOP)
+    def test_full_timestep_loop(self, input_param, input_shape, expected_output):
+        scheduler = DDPMScheduler(**input_param)
+        scheduler.set_timesteps(50)
+        torch.manual_seed(42)
+        model_output = torch.randn(input_shape)
+        sample = torch.randn(input_shape)
+        for t in range(50):
+            sample, _ = scheduler.step(model_output=model_output, timestep=t, sample=sample)
+        assert_allclose(sample, expected_output, rtol=1e-3, atol=1e-3)
+
+    @parameterized.expand(TEST_CASES)
+    def test_get_velocity_shape(self, input_param, input_shape, expected_shape):
+        scheduler = DDPMScheduler(**input_param)
+        sample = torch.randn(input_shape)
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (input_shape[0],)).long()
+        velocity = scheduler.get_velocity(sample=sample, noise=sample, timesteps=timesteps)
+        self.assertEqual(velocity.shape, expected_shape)
+
+    def test_step_learned(self):
+        for variance_type in ["learned", "learned_range"]:
+            scheduler = DDPMScheduler(variance_type=variance_type)
+        model_output = torch.randn(2, 6, 16, 16)
+        sample = torch.randn(2, 3, 16, 16)
+        output_step = scheduler.step(model_output=model_output, timestep=500, sample=sample)
+        self.assertEqual(output_step[0].shape, sample.shape)
+        self.assertEqual(output_step[1].shape, sample.shape)
+
+    def test_set_timesteps(self):
+        scheduler = DDPMScheduler(num_train_timesteps=1000)
+        scheduler.set_timesteps(num_inference_steps=100)
+        self.assertEqual(scheduler.num_inference_steps, 100)
+        self.assertEqual(len(scheduler.timesteps), 100)
+
+    def test_set_timesteps_with_num_inference_steps_bigger_than_num_train_timesteps(self):
+        scheduler = DDPMScheduler(num_train_timesteps=1000)
+        with self.assertRaises(ValueError):
+            scheduler.set_timesteps(num_inference_steps=2000)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_scheduler_pndm.py b/tests/test_scheduler_pndm.py
new file mode 100644
index 0000000000..69e5e403f5
--- /dev/null
+++ b/tests/test_scheduler_pndm.py
@@ -0,0 +1,108 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.networks.schedulers import PNDMScheduler
+from tests.utils import assert_allclose
+
+TEST_2D_CASE = []
+for beta_schedule in ["linear_beta", "scaled_linear_beta"]:
+    TEST_2D_CASE.append([{"schedule": beta_schedule}, (2, 6, 16, 16), (2, 6, 16, 16)])
+
+TEST_3D_CASE = []
+for beta_schedule in ["linear_beta", "scaled_linear_beta"]:
+    TEST_3D_CASE.append([{"schedule": beta_schedule}, (2, 6, 16, 16, 16), (2, 6, 16, 16, 16)])
+
+TEST_CASES = TEST_2D_CASE + TEST_3D_CASE
+
+TEST_FULl_LOOP = [
+    [
+        {"schedule": "linear_beta"},
+        (1, 1, 2, 2),
+        torch.Tensor([[[[-2123055.2500, -459014.2812], [2863438.0000, -1263401.7500]]]]),
+    ]
+]
+
+
+class TestDDPMScheduler(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_add_noise(self, input_param, input_shape, expected_shape):
+        scheduler = PNDMScheduler(**input_param)
+        original_sample = torch.zeros(input_shape)
+        noise = torch.randn_like(original_sample)
+        timesteps = torch.randint(0, scheduler.num_train_timesteps, (original_sample.shape[0],)).long()
+        noisy = scheduler.add_noise(original_samples=original_sample, noise=noise, timesteps=timesteps)
+        self.assertEqual(noisy.shape, expected_shape)
+
+    @parameterized.expand(TEST_CASES)
+    def test_step_shape(self, input_param, input_shape, expected_shape):
+        scheduler = PNDMScheduler(**input_param)
+        scheduler.set_timesteps(600)
+        model_output = torch.randn(input_shape)
+        sample = torch.randn(input_shape)
+        output_step = scheduler.step(model_output=model_output, timestep=500, sample=sample)
+        self.assertEqual(output_step[0].shape, expected_shape)
+        self.assertEqual(output_step[1], None)
+
+    @parameterized.expand(TEST_FULl_LOOP)
+    def test_full_timestep_loop(self, input_param, input_shape, expected_output):
+        scheduler = PNDMScheduler(**input_param)
+        scheduler.set_timesteps(50)
+        torch.manual_seed(42)
+        model_output = torch.randn(input_shape)
+        sample = torch.randn(input_shape)
+        for t in range(50):
+            sample, _ = scheduler.step(model_output=model_output, timestep=t, sample=sample)
+        assert_allclose(sample, expected_output, rtol=1e-3, atol=1e-3)
+
+    @parameterized.expand(TEST_FULl_LOOP)
+    def test_timestep_two_loops(self, input_param, input_shape, expected_output):
+        scheduler = PNDMScheduler(**input_param)
+        scheduler.set_timesteps(50)
+        torch.manual_seed(42)
+        model_output = torch.randn(input_shape)
+        sample = torch.randn(input_shape)
+        for t in range(50):
+            sample, _ = scheduler.step(model_output=model_output, timestep=t, sample=sample)
+        torch.manual_seed(42)
+        model_output2 = torch.randn(input_shape)
+        sample2 = torch.randn(input_shape)
+        scheduler.set_timesteps(50)
+        for t in range(50):
+            sample2, _ = scheduler.step(model_output=model_output2, timestep=t, sample=sample2)
+        assert_allclose(sample, sample2, rtol=1e-3, atol=1e-3)
+
+    def test_set_timesteps(self):
+        scheduler = PNDMScheduler(num_train_timesteps=1000, skip_prk_steps=True)
+        scheduler.set_timesteps(num_inference_steps=100)
+        self.assertEqual(scheduler.num_inference_steps, 100)
+        self.assertEqual(len(scheduler.timesteps), 100)
+
+    def test_set_timesteps_prk(self):
+        scheduler = PNDMScheduler(num_train_timesteps=1000, skip_prk_steps=False)
+        scheduler.set_timesteps(num_inference_steps=100)
+        self.assertEqual(scheduler.num_inference_steps, 109)
+        self.assertEqual(len(scheduler.timesteps), 109)
+
+    def test_set_timesteps_with_num_inference_steps_bigger_than_num_train_timesteps(self):
+        scheduler = PNDMScheduler(num_train_timesteps=1000)
+        with self.assertRaises(ValueError):
+            scheduler.set_timesteps(num_inference_steps=2000)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_selfattention.py b/tests/test_selfattention.py
index 0ebed84159..d069d6aa30 100644
--- a/tests/test_selfattention.py
+++ b/tests/test_selfattention.py
@@ -20,6 +20,7 @@
 
 from monai.networks import eval_mode
 from monai.networks.blocks.selfattention import SABlock
+from monai.networks.layers.factories import RelPosEmbedding
 from monai.utils import optional_import
 
 einops, has_einops = optional_import("einops")
@@ -28,12 +29,20 @@
 for dropout_rate in np.linspace(0, 1, 4):
     for hidden_size in [360, 480, 600, 768]:
         for num_heads in [4, 6, 8, 12]:
-            test_case = [
-                {"hidden_size": hidden_size, "num_heads": num_heads, "dropout_rate": dropout_rate},
-                (2, 512, hidden_size),
-                (2, 512, hidden_size),
-            ]
-            TEST_CASE_SABLOCK.append(test_case)
+            for rel_pos_embedding in [None, RelPosEmbedding.DECOMPOSED]:
+                for input_size in [(16, 32), (8, 8, 8)]:
+                    test_case = [
+                        {
+                            "hidden_size": hidden_size,
+                            "num_heads": num_heads,
+                            "dropout_rate": dropout_rate,
+                            "rel_pos_embedding": rel_pos_embedding,
+                            "input_size": input_size,
+                        },
+                        (2, 512, hidden_size),
+                        (2, 512, hidden_size),
+                    ]
+                    TEST_CASE_SABLOCK.append(test_case)
 
 
 class TestResBlock(unittest.TestCase):
@@ -53,6 +62,27 @@ def test_ill_arg(self):
         with self.assertRaises(ValueError):
             SABlock(hidden_size=620, num_heads=8, dropout_rate=0.4)
 
+    def test_attention_dim_not_multiple_of_heads(self):
+        with self.assertRaises(ValueError):
+            SABlock(hidden_size=128, num_heads=3, dropout_rate=0.1)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_inner_dim_different(self):
+        SABlock(hidden_size=128, num_heads=4, dropout_rate=0.1, dim_head=30)
+
+    def test_causal_no_sequence_length(self):
+        with self.assertRaises(ValueError):
+            SABlock(hidden_size=128, num_heads=4, dropout_rate=0.1, causal=True)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_causal(self):
+        block = SABlock(hidden_size=128, num_heads=1, dropout_rate=0.1, causal=True, sequence_length=16, save_attn=True)
+        input_shape = (1, 16, 128)
+        block(torch.randn(input_shape))
+        # check upper triangular part of the attention matrix is zero
+        assert torch.triu(block.att_mat, diagonal=1).sum() == 0
+
+    @skipUnless(has_einops, "Requires einops")
     def test_access_attn_matrix(self):
         # input format
         hidden_size = 128
diff --git a/tests/test_spade_autoencoderkl.py b/tests/test_spade_autoencoderkl.py
new file mode 100644
index 0000000000..9353ceedc2
--- /dev/null
+++ b/tests/test_spade_autoencoderkl.py
@@ -0,0 +1,295 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.nets import SPADEAutoencoderKL
+from monai.utils import optional_import
+
+einops, has_einops = optional_import("einops")
+
+CASES_NO_ATTENTION = [
+    [
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+        },
+        (1, 1, 16, 16),
+        (1, 3, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "with_encoder_nonlocal_attn": False,
+            "with_decoder_nonlocal_attn": False,
+        },
+        (1, 1, 16, 16, 16),
+        (1, 3, 16, 16, 16),
+        (1, 1, 16, 16, 16),
+        (1, 4, 4, 4, 4),
+    ],
+]
+
+CASES_ATTENTION = [
+    [
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16),
+        (1, 3, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": (1, 1, 2),
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16),
+        (1, 3, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16),
+        (1, 3, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, True),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16),
+        (1, 3, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, False),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "with_encoder_nonlocal_attn": False,
+        },
+        (1, 1, 16, 16),
+        (1, 3, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, True),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+        },
+        (1, 1, 16, 16, 16),
+        (1, 3, 16, 16, 16),
+        (1, 1, 16, 16, 16),
+        (1, 4, 4, 4, 4),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "label_nc": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4, 4),
+            "latent_channels": 4,
+            "attention_levels": (False, False, True),
+            "num_res_blocks": 1,
+            "norm_num_groups": 4,
+            "spade_intermediate_channels": 32,
+        },
+        (1, 1, 16, 16),
+        (1, 3, 16, 16),
+        (1, 1, 16, 16),
+        (1, 4, 4, 4),
+    ],
+]
+
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+
+if has_einops:
+    CASES = CASES_ATTENTION + CASES_NO_ATTENTION
+else:
+    CASES = CASES_NO_ATTENTION
+
+
+class TestSPADEAutoEncoderKL(unittest.TestCase):
+    @parameterized.expand(CASES)
+    def test_shape(self, input_param, input_shape, input_seg, expected_shape, expected_latent_shape):
+        net = SPADEAutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.forward(torch.randn(input_shape).to(device), torch.randn(input_seg).to(device))
+            self.assertEqual(result[0].shape, expected_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_model_channels_not_multiple_of_norm_num_group(self):
+        with self.assertRaises(ValueError):
+            SPADEAutoencoderKL(
+                spatial_dims=2,
+                label_nc=3,
+                in_channels=1,
+                out_channels=1,
+                channels=(24, 24, 24),
+                attention_levels=(False, False, False),
+                latent_channels=8,
+                num_res_blocks=1,
+                norm_num_groups=16,
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_model_channels_not_same_size_of_attention_levels(self):
+        with self.assertRaises(ValueError):
+            SPADEAutoencoderKL(
+                spatial_dims=2,
+                label_nc=3,
+                in_channels=1,
+                out_channels=1,
+                channels=(24, 24, 24),
+                attention_levels=(False, False),
+                latent_channels=8,
+                num_res_blocks=1,
+                norm_num_groups=16,
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_model_channels_not_same_size_of_num_res_blocks(self):
+        with self.assertRaises(ValueError):
+            SPADEAutoencoderKL(
+                spatial_dims=2,
+                label_nc=3,
+                in_channels=1,
+                out_channels=1,
+                channels=(24, 24, 24),
+                attention_levels=(False, False, False),
+                latent_channels=8,
+                num_res_blocks=(8, 8),
+                norm_num_groups=16,
+            )
+
+    def test_shape_encode(self):
+        input_param, input_shape, _, _, expected_latent_shape = CASES[0]
+        net = SPADEAutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.encode(torch.randn(input_shape).to(device))
+            self.assertEqual(result[0].shape, expected_latent_shape)
+            self.assertEqual(result[1].shape, expected_latent_shape)
+
+    def test_shape_sampling(self):
+        input_param, _, _, _, expected_latent_shape = CASES[0]
+        net = SPADEAutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.sampling(
+                torch.randn(expected_latent_shape).to(device), torch.randn(expected_latent_shape).to(device)
+            )
+            self.assertEqual(result.shape, expected_latent_shape)
+
+    def test_shape_decode(self):
+        input_param, _, input_seg_shape, expected_input_shape, latent_shape = CASES[0]
+        net = SPADEAutoencoderKL(**input_param).to(device)
+        with eval_mode(net):
+            result = net.decode(torch.randn(latent_shape).to(device), torch.randn(input_seg_shape).to(device))
+            self.assertEqual(result.shape, expected_input_shape)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_wrong_shape_decode(self):
+        net = SPADEAutoencoderKL(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=1,
+            out_channels=1,
+            channels=(4, 4, 4),
+            latent_channels=4,
+            attention_levels=(False, False, False),
+            num_res_blocks=1,
+            norm_num_groups=4,
+        )
+        with self.assertRaises(RuntimeError):
+            _ = net.decode(torch.randn((1, 1, 16, 16)).to(device), torch.randn((1, 6, 16, 16)).to(device))
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_spade_diffusion_model_unet.py b/tests/test_spade_diffusion_model_unet.py
new file mode 100644
index 0000000000..481705f56f
--- /dev/null
+++ b/tests/test_spade_diffusion_model_unet.py
@@ -0,0 +1,574 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.nets import SPADEDiffusionModelUNet
+from monai.utils import optional_import
+
+einops, has_einops = optional_import("einops")
+UNCOND_CASES_2D = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": (1, 1, 2),
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, True, True),
+            "num_head_channels": (0, 2, 4),
+            "norm_num_groups": 8,
+            "label_nc": 3,
+        }
+    ],
+]
+
+UNCOND_CASES_3D = [
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "label_nc": 3,
+            "spade_intermediate_channels": 256,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, False),
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 8,
+            "norm_num_groups": 8,
+            "resblock_updown": True,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": (0, 0, 4),
+            "norm_num_groups": 8,
+            "label_nc": 3,
+        }
+    ],
+]
+
+COND_CASES_2D = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "resblock_updown": True,
+            "label_nc": 3,
+        }
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "num_res_blocks": 1,
+            "channels": (8, 8, 8),
+            "attention_levels": (False, False, True),
+            "num_head_channels": 4,
+            "norm_num_groups": 8,
+            "with_conditioning": True,
+            "transformer_num_layers": 1,
+            "cross_attention_dim": 3,
+            "upcast_attention": True,
+            "label_nc": 3,
+        }
+    ],
+]
+
+
+class TestSPADEDiffusionModelUNet2D(unittest.TestCase):
+    @parameterized.expand(UNCOND_CASES_2D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_unconditioned_models(self, input_param):
+        net = SPADEDiffusionModelUNet(**input_param)
+        with eval_mode(net):
+            result = net.forward(
+                torch.rand((1, 1, 16, 16)),
+                torch.randint(0, 1000, (1,)).long(),
+                torch.rand((1, input_param["label_nc"], 16, 16)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 16))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_timestep_with_wrong_shape(self):
+        net = SPADEDiffusionModelUNet(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, False),
+            norm_num_groups=8,
+        )
+        with self.assertRaises(ValueError):
+            with eval_mode(net):
+                net.forward(
+                    torch.rand((1, 1, 16, 16)), torch.randint(0, 1000, (1, 1)).long(), torch.rand((1, 3, 16, 16))
+                )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_label_with_wrong_shape(self):
+        net = SPADEDiffusionModelUNet(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, False),
+            norm_num_groups=8,
+        )
+        with self.assertRaises(RuntimeError):
+            with eval_mode(net):
+                net.forward(torch.rand((1, 1, 16, 16)), torch.randint(0, 1000, (1,)).long(), torch.rand((1, 6, 16, 16)))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_with_different_in_channel_out_channel(self):
+        in_channels = 6
+        out_channels = 3
+        net = SPADEDiffusionModelUNet(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, False),
+            norm_num_groups=8,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                torch.rand((1, in_channels, 16, 16)), torch.randint(0, 1000, (1,)).long(), torch.rand((1, 3, 16, 16))
+            )
+            self.assertEqual(result.shape, (1, out_channels, 16, 16))
+
+    def test_model_channels_not_multiple_of_norm_num_group(self):
+        with self.assertRaises(ValueError):
+            SPADEDiffusionModelUNet(
+                spatial_dims=2,
+                label_nc=3,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 12),
+                attention_levels=(False, False, False),
+                norm_num_groups=8,
+            )
+
+    def test_attention_levels_with_different_length_num_head_channels(self):
+        with self.assertRaises(ValueError):
+            SPADEDiffusionModelUNet(
+                spatial_dims=2,
+                label_nc=3,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 8),
+                attention_levels=(False, False, False),
+                num_head_channels=(0, 2),
+                norm_num_groups=8,
+            )
+
+    def test_num_res_blocks_with_different_length_channels(self):
+        with self.assertRaises(ValueError):
+            SPADEDiffusionModelUNet(
+                spatial_dims=2,
+                label_nc=3,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=(1, 1),
+                channels=(8, 8, 8),
+                attention_levels=(False, False, False),
+                norm_num_groups=8,
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models(self):
+        net = SPADEDiffusionModelUNet(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            with_conditioning=True,
+            transformer_num_layers=1,
+            cross_attention_dim=3,
+            norm_num_groups=8,
+            num_head_channels=8,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 32)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                seg=torch.rand((1, 3, 16, 32)),
+                context=torch.rand((1, 1, 3)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 32))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_with_conditioning_cross_attention_dim_none(self):
+        with self.assertRaises(ValueError):
+            SPADEDiffusionModelUNet(
+                spatial_dims=2,
+                label_nc=3,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 8),
+                attention_levels=(False, False, True),
+                with_conditioning=True,
+                transformer_num_layers=1,
+                cross_attention_dim=None,
+                norm_num_groups=8,
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_context_with_conditioning_none(self):
+        net = SPADEDiffusionModelUNet(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            with_conditioning=False,
+            transformer_num_layers=1,
+            norm_num_groups=8,
+        )
+
+        with self.assertRaises(ValueError):
+            with eval_mode(net):
+                net.forward(
+                    x=torch.rand((1, 1, 16, 32)),
+                    timesteps=torch.randint(0, 1000, (1,)).long(),
+                    seg=torch.rand((1, 3, 16, 32)),
+                    context=torch.rand((1, 1, 3)),
+                )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models_class_conditioning(self):
+        net = SPADEDiffusionModelUNet(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            norm_num_groups=8,
+            num_head_channels=8,
+            num_class_embeds=2,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 32)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                seg=torch.rand((1, 3, 16, 32)),
+                class_labels=torch.randint(0, 2, (1,)).long(),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 32))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_conditioned_models_no_class_labels(self):
+        net = SPADEDiffusionModelUNet(
+            spatial_dims=2,
+            label_nc=3,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            norm_num_groups=8,
+            num_head_channels=8,
+            num_class_embeds=2,
+        )
+
+        with self.assertRaises(ValueError):
+            net.forward(
+                x=torch.rand((1, 1, 16, 32)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                seg=torch.rand((1, 3, 16, 32)),
+            )
+
+    def test_model_channels_not_same_size_of_attention_levels(self):
+        with self.assertRaises(ValueError):
+            SPADEDiffusionModelUNet(
+                spatial_dims=2,
+                label_nc=3,
+                in_channels=1,
+                out_channels=1,
+                num_res_blocks=1,
+                channels=(8, 8, 8),
+                attention_levels=(False, False),
+                norm_num_groups=8,
+                num_head_channels=8,
+                num_class_embeds=2,
+            )
+
+    @parameterized.expand(COND_CASES_2D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_conditioned_2d_models_shape(self, input_param):
+        net = SPADEDiffusionModelUNet(**input_param)
+        with eval_mode(net):
+            result = net.forward(
+                torch.rand((1, 1, 16, 16)),
+                torch.randint(0, 1000, (1,)).long(),
+                torch.rand((1, input_param["label_nc"], 16, 16)),
+                torch.rand((1, 1, 3)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 16))
+
+
+class TestDiffusionModelUNet3D(unittest.TestCase):
+    @parameterized.expand(UNCOND_CASES_3D)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_unconditioned_models(self, input_param):
+        net = SPADEDiffusionModelUNet(**input_param)
+        with eval_mode(net):
+            result = net.forward(
+                torch.rand((1, 1, 16, 16, 16)),
+                torch.randint(0, 1000, (1,)).long(),
+                torch.rand((1, input_param["label_nc"], 16, 16, 16)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 16, 16))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_with_different_in_channel_out_channel(self):
+        in_channels = 6
+        out_channels = 3
+        net = SPADEDiffusionModelUNet(
+            spatial_dims=3,
+            label_nc=3,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            num_res_blocks=1,
+            channels=(8, 8, 8),
+            attention_levels=(False, False, True),
+            norm_num_groups=4,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                torch.rand((1, in_channels, 16, 16, 16)),
+                torch.randint(0, 1000, (1,)).long(),
+                torch.rand((1, 3, 16, 16, 16)),
+            )
+            self.assertEqual(result.shape, (1, out_channels, 16, 16, 16))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape_conditioned_models(self):
+        net = SPADEDiffusionModelUNet(
+            spatial_dims=3,
+            label_nc=3,
+            in_channels=1,
+            out_channels=1,
+            num_res_blocks=1,
+            channels=(16, 16, 16),
+            attention_levels=(False, False, True),
+            norm_num_groups=16,
+            with_conditioning=True,
+            transformer_num_layers=1,
+            cross_attention_dim=3,
+        )
+        with eval_mode(net):
+            result = net.forward(
+                x=torch.rand((1, 1, 16, 16, 16)),
+                timesteps=torch.randint(0, 1000, (1,)).long(),
+                seg=torch.rand((1, 3, 16, 16, 16)),
+                context=torch.rand((1, 1, 3)),
+            )
+            self.assertEqual(result.shape, (1, 1, 16, 16, 16))
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_spade_vaegan.py b/tests/test_spade_vaegan.py
new file mode 100644
index 0000000000..3fdb9b74cb
--- /dev/null
+++ b/tests/test_spade_vaegan.py
@@ -0,0 +1,140 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import numpy as np
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.nets import SPADENet
+
+CASE_2D = [
+    [[2, 1, 1, 3, [64, 64], [16, 32, 64, 128], 16, True]],
+    [[2, 1, 1, 3, [64, 64], [16, 32, 64, 128], None, False]],
+]
+CASE_3D = [
+    [[3, 1, 1, 3, [64, 64, 64], [16, 32, 64, 128], 16, True]],
+    [[3, 1, 1, 3, [64, 64, 64], [16, 32, 64, 128], None, False]],
+]
+
+
+def create_semantic_data(shape: list, semantic_regions: int):
+    """
+    To create semantic and image mock inputs for the network.
+    Args:
+        shape: input shape
+        semantic_regions: number of semantic region
+    Returns:
+    """
+    out_label = torch.zeros(shape)
+    out_image = torch.zeros(shape) + torch.randn(shape) * 0.01
+    for i in range(1, semantic_regions):
+        shape_square = [i // np.random.choice(list(range(2, i // 2))) for i in shape]
+        start_point = [np.random.choice(list(range(shape[ind] - shape_square[ind]))) for ind, i in enumerate(shape)]
+        if len(shape) == 2:
+            out_label[
+                start_point[0] : (start_point[0] + shape_square[0]), start_point[1] : (start_point[1] + shape_square[1])
+            ] = i
+            base_intensity = torch.ones(shape_square) * np.random.randn()
+            out_image[
+                start_point[0] : (start_point[0] + shape_square[0]), start_point[1] : (start_point[1] + shape_square[1])
+            ] = (base_intensity + torch.randn(shape_square) * 0.1)
+        elif len(shape) == 3:
+            out_label[
+                start_point[0] : (start_point[0] + shape_square[0]),
+                start_point[1] : (start_point[1] + shape_square[1]),
+                start_point[2] : (start_point[2] + shape_square[2]),
+            ] = i
+            base_intensity = torch.ones(shape_square) * np.random.randn()
+            out_image[
+                start_point[0] : (start_point[0] + shape_square[0]),
+                start_point[1] : (start_point[1] + shape_square[1]),
+                start_point[2] : (start_point[2] + shape_square[2]),
+            ] = (base_intensity + torch.randn(shape_square) * 0.1)
+        else:
+            ValueError("Supports only 2D and 3D tensors")
+
+    # One hot encode label
+    out_label_ = torch.zeros([semantic_regions] + list(out_label.shape))
+    for ch in range(semantic_regions):
+        out_label_[ch, ...] = out_label == ch
+
+    return out_label_.unsqueeze(0), out_image.unsqueeze(0).unsqueeze(0)
+
+
+class TestSpadeNet(unittest.TestCase):
+    @parameterized.expand(CASE_2D)
+    def test_forward_2d(self, input_param):
+        """
+        Check that forward method is called correctly and output shape matches.
+        """
+        net = SPADENet(*input_param)
+        in_label, in_image = create_semantic_data(input_param[4], input_param[3])
+        with eval_mode(net):
+            if not net.is_vae:
+                out = net(in_label, in_image)
+                out = out[0]
+            else:
+                out, z_mu, z_logvar = net(in_label, in_image)
+                self.assertTrue(torch.all(torch.isfinite(z_mu)))
+                self.assertTrue(torch.all(torch.isfinite(z_logvar)))
+
+            self.assertTrue(torch.all(torch.isfinite(out)))
+            self.assertEqual(list(out.shape), [1, 1, 64, 64])
+
+    @parameterized.expand(CASE_2D)
+    def test_encoder_decoder(self, input_param):
+        """
+        Check that forward method is called correctly and output shape matches.
+        """
+        net = SPADENet(*input_param)
+        in_label, in_image = create_semantic_data(input_param[4], input_param[3])
+        with eval_mode(net):
+            out_z = net.encode(in_image)
+            if net.is_vae:
+                self.assertEqual(list(out_z.shape), [1, 16])
+            else:
+                self.assertEqual(out_z, None)
+            out_i = net.decode(in_label, out_z)
+            self.assertEqual(list(out_i.shape), [1, 1, 64, 64])
+
+    @parameterized.expand(CASE_3D)
+    def test_forward_3d(self, input_param):
+        """
+        Check that forward method is called correctly and output shape matches.
+        """
+        net = SPADENet(*input_param)
+        in_label, in_image = create_semantic_data(input_param[4], input_param[3])
+        with eval_mode(net):
+            if net.is_vae:
+                out, z_mu, z_logvar = net(in_label, in_image)
+                self.assertTrue(torch.all(torch.isfinite(z_mu)))
+                self.assertTrue(torch.all(torch.isfinite(z_logvar)))
+            else:
+                out = net(in_label, in_image)
+                out = out[0]
+            self.assertTrue(torch.all(torch.isfinite(out)))
+            self.assertEqual(list(out.shape), [1, 1, 64, 64, 64])
+
+    def test_shape_wrong(self):
+        """
+        We input an input shape that isn't divisible by 2**(n downstream steps)
+        """
+        with self.assertRaises(ValueError):
+            _ = SPADENet(1, 1, 8, [16, 16], [16, 32, 64, 128], 16, True)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_spatialattention.py b/tests/test_spatialattention.py
new file mode 100644
index 0000000000..70b78263c5
--- /dev/null
+++ b/tests/test_spatialattention.py
@@ -0,0 +1,55 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.blocks.spatialattention import SpatialAttentionBlock
+from monai.utils import optional_import
+
+einops, has_einops = optional_import("einops")
+
+TEST_CASES = [
+    [
+        {"spatial_dims": 2, "num_channels": 128, "num_head_channels": 32, "norm_num_groups": 32, "norm_eps": 1e-6},
+        (1, 128, 32, 32),
+        (1, 128, 32, 32),
+    ],
+    [
+        {"spatial_dims": 3, "num_channels": 16, "num_head_channels": 8, "norm_num_groups": 8, "norm_eps": 1e-6},
+        (1, 16, 8, 8, 8),
+        (1, 16, 8, 8, 8),
+    ],
+]
+
+
+class TestBlock(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_shape(self, input_param, input_shape, expected_shape):
+        net = SpatialAttentionBlock(**input_param)
+        with eval_mode(net):
+            result = net(torch.randn(input_shape))
+            self.assertEqual(result.shape, expected_shape)
+
+    def test_attention_dim_not_multiple_of_heads(self):
+        with self.assertRaises(ValueError):
+            SpatialAttentionBlock(spatial_dims=2, num_channels=128, num_head_channels=33)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_synthetic.py b/tests/test_synthetic.py
index 7db3c3e77a..4ab2144568 100644
--- a/tests/test_synthetic.py
+++ b/tests/test_synthetic.py
@@ -47,7 +47,7 @@ def test_create_test_image(self, dim, input_param, expected_img, expected_seg, e
         set_determinism(seed=0)
         if dim == 2:
             img, seg = create_test_image_2d(**input_param)
-        elif dim == 3:
+        else:  # dim == 3
             img, seg = create_test_image_3d(**input_param)
         self.assertEqual(img.shape, expected_shape)
         self.assertEqual(seg.max(), expected_max_cls)
diff --git a/tests/test_transformer.py b/tests/test_transformer.py
new file mode 100644
index 0000000000..b371809d47
--- /dev/null
+++ b/tests/test_transformer.py
@@ -0,0 +1,109 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import os
+import tempfile
+import unittest
+from unittest import skipUnless
+
+import numpy as np
+import torch
+from parameterized import parameterized
+
+from monai.apps import download_url
+from monai.networks import eval_mode
+from monai.networks.nets import DecoderOnlyTransformer
+from monai.utils import optional_import
+from tests.utils import skip_if_downloading_fails, testing_data_config
+
+_, has_einops = optional_import("einops")
+TEST_CASES = []
+for dropout_rate in np.linspace(0, 1, 2):
+    for attention_layer_dim in [360, 480, 600, 768]:
+        for num_heads in [4, 6, 8, 12]:
+            TEST_CASES.append(
+                [
+                    {
+                        "num_tokens": 10,
+                        "max_seq_len": 16,
+                        "attn_layers_dim": attention_layer_dim,
+                        "attn_layers_depth": 2,
+                        "attn_layers_heads": num_heads,
+                        "embedding_dropout_rate": dropout_rate,
+                    }
+                ]
+            )
+
+
+class TestDecoderOnlyTransformer(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_unconditioned_models(self, input_param):
+        net = DecoderOnlyTransformer(**input_param)
+        with eval_mode(net):
+            net.forward(torch.randint(0, 10, (1, 16)))
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_conditioned_models(self, input_param):
+        net = DecoderOnlyTransformer(**input_param, with_cross_attention=True)
+        with eval_mode(net):
+            net.forward(torch.randint(0, 10, (1, 16)), context=torch.randn(1, 3, input_param["attn_layers_dim"]))
+
+    def test_attention_dim_not_multiple_of_heads(self):
+        with self.assertRaises(ValueError):
+            DecoderOnlyTransformer(
+                num_tokens=10, max_seq_len=16, attn_layers_dim=8, attn_layers_depth=2, attn_layers_heads=3
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_dropout_rate_negative(self):
+
+        with self.assertRaises(ValueError):
+            DecoderOnlyTransformer(
+                num_tokens=10,
+                max_seq_len=16,
+                attn_layers_dim=8,
+                attn_layers_depth=2,
+                attn_layers_heads=2,
+                embedding_dropout_rate=-1,
+            )
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_compatibility_with_monai_generative(self):
+        # test loading weights from a model saved in MONAI Generative, version 0.2.3
+        with skip_if_downloading_fails():
+            net = DecoderOnlyTransformer(
+                num_tokens=10,
+                max_seq_len=16,
+                attn_layers_dim=8,
+                attn_layers_depth=2,
+                attn_layers_heads=2,
+                with_cross_attention=True,
+                embedding_dropout_rate=0,
+            )
+
+            tmpdir = tempfile.mkdtemp()
+            key = "decoder_only_transformer_monai_generative_weights"
+            url = testing_data_config("models", key, "url")
+            hash_type = testing_data_config("models", key, "hash_type")
+            hash_val = testing_data_config("models", key, "hash_val")
+            filename = "decoder_only_transformer_monai_generative_weights.pt"
+            weight_path = os.path.join(tmpdir, filename)
+            download_url(url=url, filepath=weight_path, hash_val=hash_val, hash_type=hash_type)
+
+            net.load_old_state_dict(torch.load(weight_path), verbose=False)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_transformerblock.py b/tests/test_transformerblock.py
index 5a8dbba83c..a850cc6f74 100644
--- a/tests/test_transformerblock.py
+++ b/tests/test_transformerblock.py
@@ -12,6 +12,7 @@
 from __future__ import annotations
 
 import unittest
+from unittest import skipUnless
 
 import numpy as np
 import torch
@@ -19,28 +20,33 @@
 
 from monai.networks import eval_mode
 from monai.networks.blocks.transformerblock import TransformerBlock
+from monai.utils import optional_import
 
+einops, has_einops = optional_import("einops")
 TEST_CASE_TRANSFORMERBLOCK = []
 for dropout_rate in np.linspace(0, 1, 4):
     for hidden_size in [360, 480, 600, 768]:
         for num_heads in [4, 8, 12]:
             for mlp_dim in [1024, 3072]:
-                test_case = [
-                    {
-                        "hidden_size": hidden_size,
-                        "num_heads": num_heads,
-                        "mlp_dim": mlp_dim,
-                        "dropout_rate": dropout_rate,
-                    },
-                    (2, 512, hidden_size),
-                    (2, 512, hidden_size),
-                ]
-                TEST_CASE_TRANSFORMERBLOCK.append(test_case)
+                for cross_attention in [False, True]:
+                    test_case = [
+                        {
+                            "hidden_size": hidden_size,
+                            "num_heads": num_heads,
+                            "mlp_dim": mlp_dim,
+                            "dropout_rate": dropout_rate,
+                            "with_cross_attention": cross_attention,
+                        },
+                        (2, 512, hidden_size),
+                        (2, 512, hidden_size),
+                    ]
+                    TEST_CASE_TRANSFORMERBLOCK.append(test_case)
 
 
 class TestTransformerBlock(unittest.TestCase):
 
     @parameterized.expand(TEST_CASE_TRANSFORMERBLOCK)
+    @skipUnless(has_einops, "Requires einops")
     def test_shape(self, input_param, input_shape, expected_shape):
         net = TransformerBlock(**input_param)
         with eval_mode(net):
@@ -54,6 +60,7 @@ def test_ill_arg(self):
         with self.assertRaises(ValueError):
             TransformerBlock(hidden_size=622, num_heads=8, mlp_dim=3072, dropout_rate=0.4)
 
+    @skipUnless(has_einops, "Requires einops")
     def test_access_attn_matrix(self):
         # input format
         hidden_size = 128
diff --git a/tests/test_vector_quantizer.py b/tests/test_vector_quantizer.py
new file mode 100644
index 0000000000..43533d0377
--- /dev/null
+++ b/tests/test_vector_quantizer.py
@@ -0,0 +1,89 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from math import prod
+
+import torch
+from parameterized import parameterized
+
+from monai.networks.layers import EMAQuantizer, VectorQuantizer
+
+TEST_CASES = [
+    [{"spatial_dims": 2, "num_embeddings": 16, "embedding_dim": 8}, (1, 8, 4, 4), (1, 4, 4)],
+    [{"spatial_dims": 3, "num_embeddings": 16, "embedding_dim": 8}, (1, 8, 4, 4, 4), (1, 4, 4, 4)],
+]
+
+
+class TestEMA(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_ema_shape(self, input_param, input_shape, output_shape):
+        layer = EMAQuantizer(**input_param)
+        x = torch.randn(input_shape)
+        layer = layer.train()
+        outputs = layer(x)
+        self.assertEqual(outputs[0].shape, input_shape)
+        self.assertEqual(outputs[2].shape, output_shape)
+
+        layer = layer.eval()
+        outputs = layer(x)
+        self.assertEqual(outputs[0].shape, input_shape)
+        self.assertEqual(outputs[2].shape, output_shape)
+
+    @parameterized.expand(TEST_CASES)
+    def test_ema_quantize(self, input_param, input_shape, output_shape):
+        layer = EMAQuantizer(**input_param)
+        x = torch.randn(input_shape)
+        outputs = layer.quantize(x)
+        self.assertEqual(outputs[0].shape, (prod(input_shape[2:]), input_shape[1]))  # (HxW[xD], C)
+        self.assertEqual(outputs[1].shape, (prod(input_shape[2:]), input_param["num_embeddings"]))  # (HxW[xD], E)
+        self.assertEqual(outputs[2].shape, (input_shape[0],) + input_shape[2:])  # (1, H, W, [D])
+
+    def test_ema(self):
+        layer = EMAQuantizer(spatial_dims=2, num_embeddings=2, embedding_dim=2, epsilon=0, decay=0)
+        original_weight_0 = layer.embedding.weight[0].clone()
+        original_weight_1 = layer.embedding.weight[1].clone()
+        x_0 = original_weight_0
+        x_0 = x_0.unsqueeze(0).unsqueeze(-1).unsqueeze(-1)
+        x_0 = x_0.repeat(1, 1, 1, 2) + 0.001
+
+        x_1 = original_weight_1
+        x_1 = x_1.unsqueeze(0).unsqueeze(-1).unsqueeze(-1)
+        x_1 = x_1.repeat(1, 1, 1, 2)
+
+        x = torch.cat([x_0, x_1], dim=0)
+        layer = layer.train()
+        _ = layer(x)
+
+        self.assertTrue(all(layer.embedding.weight[0] != original_weight_0))
+        self.assertTrue(all(layer.embedding.weight[1] == original_weight_1))
+
+
+class TestVectorQuantizer(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_vector_quantizer_shape(self, input_param, input_shape, output_shape):
+        layer = VectorQuantizer(EMAQuantizer(**input_param))
+        x = torch.randn(input_shape)
+        outputs = layer(x)
+        self.assertEqual(outputs[1].shape, input_shape)
+
+    @parameterized.expand(TEST_CASES)
+    def test_vector_quantizer_quantize(self, input_param, input_shape, output_shape):
+        layer = VectorQuantizer(EMAQuantizer(**input_param))
+        x = torch.randn(input_shape)
+        outputs = layer.quantize(x)
+        self.assertEqual(outputs.shape, output_shape)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_vis_cam.py b/tests/test_vis_cam.py
index b641599af2..68b12de2f8 100644
--- a/tests/test_vis_cam.py
+++ b/tests/test_vis_cam.py
@@ -70,6 +70,8 @@ class TestClassActivationMap(unittest.TestCase):
 
     @parameterized.expand([TEST_CASE_0, TEST_CASE_1, TEST_CASE_2, TEST_CASE_3])
     def test_shape(self, input_data, expected_shape):
+        model = None
+
         if input_data["model"] == "densenet2d":
             model = DenseNet121(spatial_dims=2, in_channels=1, out_channels=3)
         if input_data["model"] == "densenet3d":
@@ -80,6 +82,7 @@ def test_shape(self, input_data, expected_shape):
             model = SEResNet50(spatial_dims=2, in_channels=3, num_classes=4)
         if input_data["model"] == "senet3d":
             model = SEResNet50(spatial_dims=3, in_channels=3, num_classes=4)
+
         device = "cuda:0" if torch.cuda.is_available() else "cpu"
         model.to(device)
         model.eval()
diff --git a/tests/test_vis_gradcam.py b/tests/test_vis_gradcam.py
index 325b74b3ce..f77d916a5b 100644
--- a/tests/test_vis_gradcam.py
+++ b/tests/test_vis_gradcam.py
@@ -153,6 +153,8 @@ class TestGradientClassActivationMap(unittest.TestCase):
 
     @parameterized.expand(TESTS)
     def test_shape(self, cam_class, input_data, expected_shape):
+        model = None
+
         if input_data["model"] == "densenet2d":
             model = DenseNet121(spatial_dims=2, in_channels=1, out_channels=3)
         elif input_data["model"] == "densenet2d_bin":
diff --git a/tests/test_vqvae.py b/tests/test_vqvae.py
new file mode 100644
index 0000000000..4916dc2faa
--- /dev/null
+++ b/tests/test_vqvae.py
@@ -0,0 +1,274 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.nets.vqvae import VQVAE
+from tests.utils import SkipIfBeforePyTorchVersion
+
+TEST_CASES = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "num_res_layers": 1,
+            "num_res_channels": (4, 4),
+            "downsample_parameters": ((2, 4, 1, 1),) * 2,
+            "upsample_parameters": ((2, 4, 1, 1, 0),) * 2,
+            "num_embeddings": 8,
+            "embedding_dim": 8,
+        },
+        (1, 1, 8, 8),
+        (1, 1, 8, 8),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "num_res_layers": 1,
+            "num_res_channels": 4,
+            "downsample_parameters": ((2, 4, 1, 1),) * 2,
+            "upsample_parameters": ((2, 4, 1, 1, 0),) * 2,
+            "num_embeddings": 8,
+            "embedding_dim": 8,
+        },
+        (1, 1, 8, 8, 8),
+        (1, 1, 8, 8, 8),
+    ],
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "num_res_layers": 1,
+            "num_res_channels": (4, 4),
+            "downsample_parameters": (2, 4, 1, 1),
+            "upsample_parameters": ((2, 4, 1, 1, 0),) * 2,
+            "num_embeddings": 8,
+            "embedding_dim": 8,
+        },
+        (1, 1, 8, 8),
+        (1, 1, 8, 8),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (4, 4),
+            "num_res_layers": 1,
+            "num_res_channels": (4, 4),
+            "downsample_parameters": ((2, 4, 1, 1),) * 2,
+            "upsample_parameters": (2, 4, 1, 1, 0),
+            "num_embeddings": 8,
+            "embedding_dim": 8,
+        },
+        (1, 1, 8, 8, 8),
+        (1, 1, 8, 8, 8),
+    ],
+]
+
+TEST_LATENT_SHAPE = {
+    "spatial_dims": 2,
+    "in_channels": 1,
+    "out_channels": 1,
+    "downsample_parameters": ((2, 4, 1, 1),) * 2,
+    "upsample_parameters": ((2, 4, 1, 1, 0),) * 2,
+    "num_res_layers": 1,
+    "channels": (8, 8),
+    "num_res_channels": (8, 8),
+    "num_embeddings": 16,
+    "embedding_dim": 8,
+}
+
+
+class TestVQVAE(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_shape(self, input_param, input_shape, expected_shape):
+        device = "cuda" if torch.cuda.is_available() else "cpu"
+
+        net = VQVAE(**input_param).to(device)
+
+        with eval_mode(net):
+            result, _ = net(torch.randn(input_shape).to(device))
+
+        self.assertEqual(result.shape, expected_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @SkipIfBeforePyTorchVersion((1, 11))
+    def test_shape_with_checkpoint(self, input_param, input_shape, expected_shape):
+        device = "cuda" if torch.cuda.is_available() else "cpu"
+        input_param = input_param.copy()
+        input_param.update({"use_checkpointing": True})
+
+        net = VQVAE(**input_param).to(device)
+
+        with eval_mode(net):
+            result, _ = net(torch.randn(input_shape).to(device))
+
+        self.assertEqual(result.shape, expected_shape)
+
+    # Removed this test case since TorchScript currently does not support activation checkpoint.
+    # def test_script(self):
+    #     net = VQVAE(
+    #         spatial_dims=2,
+    #         in_channels=1,
+    #         out_channels=1,
+    #         downsample_parameters=((2, 4, 1, 1),) * 2,
+    #         upsample_parameters=((2, 4, 1, 1, 0),) * 2,
+    #         num_res_layers=1,
+    #         channels=(8, 8),
+    #         num_res_channels=(8, 8),
+    #         num_embeddings=16,
+    #         embedding_dim=8,
+    #         ddp_sync=False,
+    #     )
+    #     test_data = torch.randn(1, 1, 16, 16)
+    #     test_script_save(net, test_data)
+
+    def test_channels_not_same_size_of_num_res_channels(self):
+        with self.assertRaises(ValueError):
+            VQVAE(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(16, 16),
+                num_res_channels=(16, 16, 16),
+                downsample_parameters=((2, 4, 1, 1),) * 2,
+                upsample_parameters=((2, 4, 1, 1, 0),) * 2,
+            )
+
+    def test_channels_not_same_size_of_downsample_parameters(self):
+        with self.assertRaises(ValueError):
+            VQVAE(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(16, 16),
+                num_res_channels=(16, 16),
+                downsample_parameters=((2, 4, 1, 1),) * 3,
+                upsample_parameters=((2, 4, 1, 1, 0),) * 2,
+            )
+
+    def test_channels_not_same_size_of_upsample_parameters(self):
+        with self.assertRaises(ValueError):
+            VQVAE(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(16, 16),
+                num_res_channels=(16, 16),
+                downsample_parameters=((2, 4, 1, 1),) * 2,
+                upsample_parameters=((2, 4, 1, 1, 0),) * 3,
+            )
+
+    def test_downsample_parameters_not_sequence_or_int(self):
+        with self.assertRaises(ValueError):
+            VQVAE(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(16, 16),
+                num_res_channels=(16, 16),
+                downsample_parameters=(("test", 4, 1, 1),) * 2,
+                upsample_parameters=((2, 4, 1, 1, 0),) * 2,
+            )
+
+    def test_upsample_parameters_not_sequence_or_int(self):
+        with self.assertRaises(ValueError):
+            VQVAE(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(16, 16),
+                num_res_channels=(16, 16),
+                downsample_parameters=((2, 4, 1, 1),) * 2,
+                upsample_parameters=(("test", 4, 1, 1, 0),) * 2,
+            )
+
+    def test_downsample_parameter_length_different_4(self):
+        with self.assertRaises(ValueError):
+            VQVAE(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(16, 16),
+                num_res_channels=(16, 16),
+                downsample_parameters=((2, 4, 1),) * 3,
+                upsample_parameters=((2, 4, 1, 1, 0),) * 2,
+            )
+
+    def test_upsample_parameter_length_different_5(self):
+        with self.assertRaises(ValueError):
+            VQVAE(
+                spatial_dims=2,
+                in_channels=1,
+                out_channels=1,
+                channels=(16, 16),
+                num_res_channels=(16, 16, 16),
+                downsample_parameters=((2, 4, 1, 1),) * 2,
+                upsample_parameters=((2, 4, 1, 1, 0, 1),) * 3,
+            )
+
+    def test_encode_shape(self):
+        device = "cuda" if torch.cuda.is_available() else "cpu"
+
+        net = VQVAE(**TEST_LATENT_SHAPE).to(device)
+
+        with eval_mode(net):
+            latent = net.encode(torch.randn(1, 1, 32, 32).to(device))
+
+        self.assertEqual(latent.shape, (1, 8, 8, 8))
+
+    def test_index_quantize_shape(self):
+        device = "cuda" if torch.cuda.is_available() else "cpu"
+
+        net = VQVAE(**TEST_LATENT_SHAPE).to(device)
+
+        with eval_mode(net):
+            latent = net.index_quantize(torch.randn(1, 1, 32, 32).to(device))
+
+        self.assertEqual(latent.shape, (1, 8, 8))
+
+    def test_decode_shape(self):
+        device = "cuda" if torch.cuda.is_available() else "cpu"
+
+        net = VQVAE(**TEST_LATENT_SHAPE).to(device)
+
+        with eval_mode(net):
+            latent = net.decode(torch.randn(1, 8, 8, 8).to(device))
+
+        self.assertEqual(latent.shape, (1, 1, 32, 32))
+
+    def test_decode_samples_shape(self):
+        device = "cuda" if torch.cuda.is_available() else "cpu"
+
+        net = VQVAE(**TEST_LATENT_SHAPE).to(device)
+
+        with eval_mode(net):
+            latent = net.decode_samples(torch.randint(low=0, high=16, size=(1, 8, 8)).to(device))
+
+        self.assertEqual(latent.shape, (1, 1, 32, 32))
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_vqvaetransformer_inferer.py b/tests/test_vqvaetransformer_inferer.py
new file mode 100644
index 0000000000..36b715f588
--- /dev/null
+++ b/tests/test_vqvaetransformer_inferer.py
@@ -0,0 +1,295 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.inferers import VQVAETransformerInferer
+from monai.networks.nets import VQVAE, DecoderOnlyTransformer
+from monai.utils import optional_import
+from monai.utils.ordering import Ordering, OrderingType
+
+einops, has_einops = optional_import("einops")
+TEST_CASES = [
+    [
+        {
+            "spatial_dims": 2,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (8, 8),
+            "num_res_channels": (8, 8),
+            "downsample_parameters": ((2, 4, 1, 1),) * 2,
+            "upsample_parameters": ((2, 4, 1, 1, 0),) * 2,
+            "num_res_layers": 1,
+            "num_embeddings": 16,
+            "embedding_dim": 8,
+        },
+        {
+            "num_tokens": 16 + 1,
+            "max_seq_len": 4,
+            "attn_layers_dim": 4,
+            "attn_layers_depth": 2,
+            "attn_layers_heads": 1,
+            "with_cross_attention": False,
+        },
+        {"ordering_type": OrderingType.RASTER_SCAN.value, "spatial_dims": 2, "dimensions": (2, 2, 2)},
+        (2, 1, 8, 8),
+        (2, 4, 17),
+        (2, 2, 2),
+    ],
+    [
+        {
+            "spatial_dims": 3,
+            "in_channels": 1,
+            "out_channels": 1,
+            "channels": (8, 8),
+            "num_res_channels": (8, 8),
+            "downsample_parameters": ((2, 4, 1, 1),) * 2,
+            "upsample_parameters": ((2, 4, 1, 1, 0),) * 2,
+            "num_res_layers": 1,
+            "num_embeddings": 16,
+            "embedding_dim": 8,
+        },
+        {
+            "num_tokens": 16 + 1,
+            "max_seq_len": 8,
+            "attn_layers_dim": 4,
+            "attn_layers_depth": 2,
+            "attn_layers_heads": 1,
+            "with_cross_attention": False,
+        },
+        {"ordering_type": OrderingType.RASTER_SCAN.value, "spatial_dims": 3, "dimensions": (2, 2, 2, 2)},
+        (2, 1, 8, 8, 8),
+        (2, 8, 17),
+        (2, 2, 2, 2),
+    ],
+]
+
+
+class TestVQVAETransformerInferer(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_prediction_shape(
+        self, stage_1_params, stage_2_params, ordering_params, input_shape, logits_shape, latent_shape
+    ):
+        stage_1 = VQVAE(**stage_1_params)
+        stage_2 = DecoderOnlyTransformer(**stage_2_params)
+        ordering = Ordering(**ordering_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+
+        inferer = VQVAETransformerInferer()
+        prediction = inferer(inputs=input, vqvae_model=stage_1, transformer_model=stage_2, ordering=ordering)
+        self.assertEqual(prediction.shape, logits_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_prediction_shape_shorter_sequence(
+        self, stage_1_params, stage_2_params, ordering_params, input_shape, logits_shape, latent_shape
+    ):
+        stage_1 = VQVAE(**stage_1_params)
+        max_seq_len = 3
+        stage_2_params_shorter = dict(stage_2_params)
+        stage_2_params_shorter["max_seq_len"] = max_seq_len
+        stage_2 = DecoderOnlyTransformer(**stage_2_params_shorter)
+        ordering = Ordering(**ordering_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+
+        inferer = VQVAETransformerInferer()
+        prediction = inferer(inputs=input, vqvae_model=stage_1, transformer_model=stage_2, ordering=ordering)
+        cropped_logits_shape = (logits_shape[0], max_seq_len, logits_shape[2])
+        self.assertEqual(prediction.shape, cropped_logits_shape)
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample(self):
+
+        stage_1 = VQVAE(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            channels=(8, 8),
+            num_res_channels=(8, 8),
+            downsample_parameters=((2, 4, 1, 1),) * 2,
+            upsample_parameters=((2, 4, 1, 1, 0),) * 2,
+            num_res_layers=1,
+            num_embeddings=16,
+            embedding_dim=8,
+        )
+        stage_2 = DecoderOnlyTransformer(
+            num_tokens=16 + 1,
+            max_seq_len=4,
+            attn_layers_dim=4,
+            attn_layers_depth=2,
+            attn_layers_heads=1,
+            with_cross_attention=False,
+        )
+        ordering = Ordering(ordering_type=OrderingType.RASTER_SCAN.value, spatial_dims=2, dimensions=(2, 2, 2))
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        inferer = VQVAETransformerInferer()
+
+        starting_token = 16  # from stage_1 num_embeddings
+
+        sample = inferer.sample(
+            latent_spatial_dim=(2, 2),
+            starting_tokens=starting_token * torch.ones((2, 1), device=device),
+            vqvae_model=stage_1,
+            transformer_model=stage_2,
+            ordering=ordering,
+        )
+        self.assertEqual(sample.shape, (2, 1, 8, 8))
+
+    @skipUnless(has_einops, "Requires einops")
+    def test_sample_shorter_sequence(self):
+        stage_1 = VQVAE(
+            spatial_dims=2,
+            in_channels=1,
+            out_channels=1,
+            channels=(8, 8),
+            num_res_channels=(8, 8),
+            downsample_parameters=((2, 4, 1, 1),) * 2,
+            upsample_parameters=((2, 4, 1, 1, 0),) * 2,
+            num_res_layers=1,
+            num_embeddings=16,
+            embedding_dim=8,
+        )
+        stage_2 = DecoderOnlyTransformer(
+            num_tokens=16 + 1,
+            max_seq_len=2,
+            attn_layers_dim=4,
+            attn_layers_depth=2,
+            attn_layers_heads=1,
+            with_cross_attention=False,
+        )
+        ordering = Ordering(ordering_type=OrderingType.RASTER_SCAN.value, spatial_dims=2, dimensions=(2, 2, 2))
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        inferer = VQVAETransformerInferer()
+
+        starting_token = 16  # from stage_1 num_embeddings
+
+        sample = inferer.sample(
+            latent_spatial_dim=(2, 2),
+            starting_tokens=starting_token * torch.ones((2, 1), device=device),
+            vqvae_model=stage_1,
+            transformer_model=stage_2,
+            ordering=ordering,
+        )
+        self.assertEqual(sample.shape, (2, 1, 8, 8))
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_get_likelihood(
+        self, stage_1_params, stage_2_params, ordering_params, input_shape, logits_shape, latent_shape
+    ):
+        stage_1 = VQVAE(**stage_1_params)
+        stage_2 = DecoderOnlyTransformer(**stage_2_params)
+        ordering = Ordering(**ordering_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+
+        inferer = VQVAETransformerInferer()
+        likelihood = inferer.get_likelihood(
+            inputs=input, vqvae_model=stage_1, transformer_model=stage_2, ordering=ordering
+        )
+        self.assertEqual(likelihood.shape, latent_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_get_likelihood_shorter_sequence(
+        self, stage_1_params, stage_2_params, ordering_params, input_shape, logits_shape, latent_shape
+    ):
+        stage_1 = VQVAE(**stage_1_params)
+        max_seq_len = 3
+        stage_2_params_shorter = dict(stage_2_params)
+        stage_2_params_shorter["max_seq_len"] = max_seq_len
+        stage_2 = DecoderOnlyTransformer(**stage_2_params_shorter)
+        ordering = Ordering(**ordering_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+
+        inferer = VQVAETransformerInferer()
+        likelihood = inferer.get_likelihood(
+            inputs=input, vqvae_model=stage_1, transformer_model=stage_2, ordering=ordering
+        )
+        self.assertEqual(likelihood.shape, latent_shape)
+
+    @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
+    def test_get_likelihood_resampling(
+        self, stage_1_params, stage_2_params, ordering_params, input_shape, logits_shape, latent_shape
+    ):
+        stage_1 = VQVAE(**stage_1_params)
+        stage_2 = DecoderOnlyTransformer(**stage_2_params)
+        ordering = Ordering(**ordering_params)
+
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        stage_1.to(device)
+        stage_2.to(device)
+        stage_1.eval()
+        stage_2.eval()
+
+        input = torch.randn(input_shape).to(device)
+
+        inferer = VQVAETransformerInferer()
+        likelihood = inferer.get_likelihood(
+            inputs=input,
+            vqvae_model=stage_1,
+            transformer_model=stage_2,
+            ordering=ordering,
+            resample_latent_likelihoods=True,
+            resample_interpolation_mode="nearest",
+        )
+        self.assertEqual(likelihood.shape, input_shape)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/testing_data/data_config.json b/tests/testing_data/data_config.json
index a570c787ba..8b1d2868b7 100644
--- a/tests/testing_data/data_config.json
+++ b/tests/testing_data/data_config.json
@@ -138,6 +138,26 @@
             "url": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/ssl_pretrained_weights.pth",
             "hash_type": "sha256",
             "hash_val": "c3564f40a6a051d3753a6d8fae5cc8eaf21ce8d82a9a3baf80748d15664055e8"
+        },
+        "decoder_only_transformer_monai_generative_weights": {
+            "url": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/decoder_only_transformer.pth",
+            "hash_type": "sha256",
+            "hash_val": "f93de37d64d77cf91f3bde95cdf93d161aee800074c89a92aff9d5699120ec0d"
+        },
+        "diffusion_model_unet_monai_generative_weights": {
+            "url": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/diffusion_model_unet.pth",
+            "hash_type": "sha256",
+            "hash_val": "0d2171b386902f5b4fd3e967b4024f63e353694ca45091b114970019d045beee"
+        },
+        "autoencoderkl_monai_generative_weights": {
+            "url": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/autoencoderkl.pth",
+            "hash_type": "sha256",
+            "hash_val": "6e02c9540c51b16b9ba98b5c0c75d6b84b430afe9a3237df1d67a520f8d34184"
+        },
+        "controlnet_monai_generative_weights": {
+            "url": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/controlnet.pth",
+            "hash_type": "sha256",
+            "hash_val": "cd100d0c69f47569ae5b4b7df653a1cb19f5e02eff1630db3210e2646fb1ab2e"
         }
     },
     "configs": {

From 7a8680e84457cb374859639ab6a078313da85926 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 23 Jul 2024 00:07:41 +0800
Subject: [PATCH 098/183] Fix mypy issue introduced in 1.11.0 (#7941)

Fixes #7940 .

### Description
Refer: https://mypy.readthedocs.io/en/stable/common_issues.html

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/apps/auto3dseg/hpo_gen.py               |  2 +-
 monai/apps/pathology/transforms/post/array.py | 10 ++++++----
 monai/data/meta_tensor.py                     |  4 ++--
 monai/networks/layers/simplelayers.py         |  2 +-
 monai/networks/nets/quicknat.py               | 12 ++++++------
 monai/transforms/croppad/array.py             | 16 ++++++++--------
 monai/transforms/croppad/dictionary.py        |  8 ++++----
 monai/transforms/spatial/array.py             |  2 +-
 monai/visualize/class_activation_maps.py      | 10 +++++-----
 tests/test_subpixel_upsample.py               |  6 +++---
 10 files changed, 37 insertions(+), 35 deletions(-)

diff --git a/monai/apps/auto3dseg/hpo_gen.py b/monai/apps/auto3dseg/hpo_gen.py
index b755b99feb..ed6d903897 100644
--- a/monai/apps/auto3dseg/hpo_gen.py
+++ b/monai/apps/auto3dseg/hpo_gen.py
@@ -53,7 +53,7 @@ def update_params(self, *args, **kwargs):
         raise NotImplementedError
 
     @abstractmethod
-    def set_score(self):
+    def set_score(self, *args, **kwargs):
         """Report the evaluated results to HPO."""
         raise NotImplementedError
 
diff --git a/monai/apps/pathology/transforms/post/array.py b/monai/apps/pathology/transforms/post/array.py
index 0aa8e14655..035bce2c69 100644
--- a/monai/apps/pathology/transforms/post/array.py
+++ b/monai/apps/pathology/transforms/post/array.py
@@ -28,7 +28,7 @@
     SobelGradients,
 )
 from monai.transforms.transform import Transform
-from monai.transforms.utils_pytorch_numpy_unification import max, maximum, min, sum, unique
+from monai.transforms.utils_pytorch_numpy_unification import max, maximum, min, sum, unique, where
 from monai.utils import TransformBackends, convert_to_numpy, optional_import
 from monai.utils.misc import ensure_tuple_rep
 from monai.utils.type_conversion import convert_to_dst_type, convert_to_tensor
@@ -162,7 +162,8 @@ def __call__(self, prob_map: NdarrayOrTensor) -> NdarrayOrTensor:
         pred = label(pred)[0]
         if self.remove_small_objects is not None:
             pred = self.remove_small_objects(pred)
-        pred[pred > 0] = 1
+        pred_indices = np.where(pred > 0)
+        pred[pred_indices] = 1
 
         return convert_to_dst_type(pred, prob_map, dtype=self.dtype)[0]
 
@@ -338,7 +339,8 @@ def __call__(self, mask: NdarrayOrTensor, instance_border: NdarrayOrTensor) -> N
         instance_border = instance_border >= self.threshold  # uncertain area
 
         marker = mask - convert_to_dst_type(instance_border, mask)[0]  # certain foreground
-        marker[marker < 0] = 0
+        marker_indices = where(marker < 0)
+        marker[marker_indices] = 0  # type: ignore[index]
         marker = self.postprocess_fn(marker)
         marker = convert_to_numpy(marker)
 
@@ -635,7 +637,7 @@ def __call__(  # type: ignore
 
         seg_map_crop = convert_to_dst_type(seg_map_crop == instance_id, type_map_crop, dtype=bool)[0]
 
-        inst_type = type_map_crop[seg_map_crop]
+        inst_type = type_map_crop[seg_map_crop]  # type: ignore[index]
         type_list, type_pixels = unique(inst_type, return_counts=True)
         type_list = list(zip(type_list, type_pixels))
         type_list = sorted(type_list, key=lambda x: x[1], reverse=True)
diff --git a/monai/data/meta_tensor.py b/monai/data/meta_tensor.py
index cad0851a8e..2df4da4a35 100644
--- a/monai/data/meta_tensor.py
+++ b/monai/data/meta_tensor.py
@@ -505,7 +505,7 @@ def peek_pending_rank(self):
         a = self.pending_operations[-1].get(LazyAttr.AFFINE, None) if self.pending_operations else self.affine
         return 1 if a is None else int(max(1, len(a) - 1))
 
-    def new_empty(self, size, dtype=None, device=None, requires_grad=False):
+    def new_empty(self, size, dtype=None, device=None, requires_grad=False):  # type: ignore[override]
         """
         must be defined for deepcopy to work
 
@@ -580,7 +580,7 @@ def ensure_torch_and_prune_meta(
             img.affine = MetaTensor.get_default_affine()
         return img
 
-    def __repr__(self):
+    def __repr__(self):  # type: ignore[override]
         """
         Prints a representation of the tensor.
         Prepends "meta" to ``torch.Tensor.__repr__``.
diff --git a/monai/networks/layers/simplelayers.py b/monai/networks/layers/simplelayers.py
index 4ac621967f..4acd4a3622 100644
--- a/monai/networks/layers/simplelayers.py
+++ b/monai/networks/layers/simplelayers.py
@@ -452,7 +452,7 @@ def get_binary_kernel(window_size: Sequence[int], dtype=torch.float, device=None
 
 def median_filter(
     in_tensor: torch.Tensor,
-    kernel_size: Sequence[int] = (3, 3, 3),
+    kernel_size: Sequence[int] | int = (3, 3, 3),
     spatial_dims: int = 3,
     kernel: torch.Tensor | None = None,
     **kwargs,
diff --git a/monai/networks/nets/quicknat.py b/monai/networks/nets/quicknat.py
index bbc4e7e490..7e0f9c6b38 100644
--- a/monai/networks/nets/quicknat.py
+++ b/monai/networks/nets/quicknat.py
@@ -42,7 +42,7 @@ class SkipConnectionWithIdx(SkipConnection):
     Inherits from SkipConnection but provides the indizes with each forward pass.
     """
 
-    def forward(self, input, indices):
+    def forward(self, input, indices):  # type: ignore[override]
         return super().forward(input), indices
 
 
@@ -57,7 +57,7 @@ class SequentialWithIdx(nn.Sequential):
     def __init__(self, *args):
         super().__init__(*args)
 
-    def forward(self, input, indices):
+    def forward(self, input, indices):  # type: ignore[override]
         for module in self:
             input, indices = module(input, indices)
         return input, indices
@@ -165,7 +165,7 @@ def _get_layer(self, in_channels, out_channels, dilation):
         )
         return nn.Sequential(conv.get_submodule("adn"), conv.get_submodule("conv"))
 
-    def forward(self, input, _):
+    def forward(self, input, _):  # type: ignore[override]
         i = 0
         result = input
         result1 = input  # this will not stay this value, needed here for pylint/mypy
@@ -215,7 +215,7 @@ def __init__(self, in_channels: int, max_pool, se_layer, dropout, kernel_size, n
         super().__init__(in_channels, se_layer, dropout, kernel_size, num_filters)
         self.max_pool = max_pool
 
-    def forward(self, input, indices=None):
+    def forward(self, input, indices=None):  # type: ignore[override]
         input, indices = self.max_pool(input)
 
         out_block, _ = super().forward(input, None)
@@ -243,7 +243,7 @@ def __init__(self, in_channels: int, un_pool, se_layer, dropout, kernel_size, nu
         super().__init__(in_channels, se_layer, dropout, kernel_size, num_filters)
         self.un_pool = un_pool
 
-    def forward(self, input, indices):
+    def forward(self, input, indices):  # type: ignore[override]
         out_block, _ = super().forward(input, None)
         out_block = self.un_pool(out_block, indices)
         return out_block, None
@@ -270,7 +270,7 @@ def __init__(self, in_channels: int, se_layer, dropout, max_pool, un_pool, kerne
         self.max_pool = max_pool
         self.un_pool = un_pool
 
-    def forward(self, input, indices):
+    def forward(self, input, indices):  # type: ignore[override]
         out_block, indices = self.max_pool(input)
         out_block, _ = super().forward(out_block, None)
         out_block = self.un_pool(out_block, indices)
diff --git a/monai/transforms/croppad/array.py b/monai/transforms/croppad/array.py
index ce3701b263..813f8c1d44 100644
--- a/monai/transforms/croppad/array.py
+++ b/monai/transforms/croppad/array.py
@@ -362,10 +362,10 @@ def __init__(self, lazy: bool = False):
 
     @staticmethod
     def compute_slices(
-        roi_center: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_size: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_start: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_end: Sequence[int] | NdarrayOrTensor | None = None,
+        roi_center: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_size: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_start: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_end: Sequence[int] | int | NdarrayOrTensor | None = None,
         roi_slices: Sequence[slice] | None = None,
     ) -> tuple[slice]:
         """
@@ -459,10 +459,10 @@ class SpatialCrop(Crop):
 
     def __init__(
         self,
-        roi_center: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_size: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_start: Sequence[int] | NdarrayOrTensor | None = None,
-        roi_end: Sequence[int] | NdarrayOrTensor | None = None,
+        roi_center: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_size: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_start: Sequence[int] | int | NdarrayOrTensor | None = None,
+        roi_end: Sequence[int] | int | NdarrayOrTensor | None = None,
         roi_slices: Sequence[slice] | None = None,
         lazy: bool = False,
     ) -> None:
diff --git a/monai/transforms/croppad/dictionary.py b/monai/transforms/croppad/dictionary.py
index be9441dc4a..cea11d9676 100644
--- a/monai/transforms/croppad/dictionary.py
+++ b/monai/transforms/croppad/dictionary.py
@@ -438,10 +438,10 @@ class SpatialCropd(Cropd):
     def __init__(
         self,
         keys: KeysCollection,
-        roi_center: Sequence[int] | None = None,
-        roi_size: Sequence[int] | None = None,
-        roi_start: Sequence[int] | None = None,
-        roi_end: Sequence[int] | None = None,
+        roi_center: Sequence[int] | int | None = None,
+        roi_size: Sequence[int] | int | None = None,
+        roi_start: Sequence[int] | int | None = None,
+        roi_end: Sequence[int] | int | None = None,
         roi_slices: Sequence[slice] | None = None,
         allow_missing_keys: bool = False,
         lazy: bool = False,
diff --git a/monai/transforms/spatial/array.py b/monai/transforms/spatial/array.py
index 094afdd3c4..3739a83e71 100644
--- a/monai/transforms/spatial/array.py
+++ b/monai/transforms/spatial/array.py
@@ -3441,7 +3441,7 @@ def filter_count(self, image_np: NdarrayOrTensor, locations: np.ndarray) -> tupl
             idx = self.R.permutation(image_np.shape[0])
             idx = idx[: self.num_patches]
             idx_np = convert_data_type(idx, np.ndarray)[0]
-            image_np = image_np[idx]
+            image_np = image_np[idx]  # type: ignore[index]
             locations = locations[idx_np]
             return image_np, locations
         elif self.sort_fn not in (None, GridPatchSort.MIN, GridPatchSort.MAX):
diff --git a/monai/visualize/class_activation_maps.py b/monai/visualize/class_activation_maps.py
index 6d1e8dfd03..489a563818 100644
--- a/monai/visualize/class_activation_maps.py
+++ b/monai/visualize/class_activation_maps.py
@@ -290,7 +290,7 @@ def __init__(
         )
         self.fc_layers = fc_layers
 
-    def compute_map(self, x, class_idx=None, layer_idx=-1, **kwargs):
+    def compute_map(self, x, class_idx=None, layer_idx=-1, **kwargs):  # type: ignore[override]
         logits, acti, _ = self.nn_module(x, **kwargs)
         acti = acti[layer_idx]
         if class_idx is None:
@@ -302,7 +302,7 @@ def compute_map(self, x, class_idx=None, layer_idx=-1, **kwargs):
         output = torch.stack([output[i, b : b + 1] for i, b in enumerate(class_idx)], dim=0)
         return output.reshape(b, 1, *spatial)  # resume the spatial dims on the selected class
 
-    def __call__(self, x, class_idx=None, layer_idx=-1, **kwargs):
+    def __call__(self, x, class_idx=None, layer_idx=-1, **kwargs):  # type: ignore[override]
         """
         Compute the activation map with upsampling and postprocessing.
 
@@ -361,7 +361,7 @@ class GradCAM(CAMBase):
 
     """
 
-    def compute_map(self, x, class_idx=None, retain_graph=False, layer_idx=-1, **kwargs):
+    def compute_map(self, x, class_idx=None, retain_graph=False, layer_idx=-1, **kwargs):  # type: ignore[override]
         _, acti, grad = self.nn_module(x, class_idx=class_idx, retain_graph=retain_graph, **kwargs)
         acti, grad = acti[layer_idx], grad[layer_idx]
         b, c, *spatial = grad.shape
@@ -369,7 +369,7 @@ def compute_map(self, x, class_idx=None, retain_graph=False, layer_idx=-1, **kwa
         acti_map = (weights * acti).sum(1, keepdim=True)
         return F.relu(acti_map)
 
-    def __call__(self, x, class_idx=None, layer_idx=-1, retain_graph=False, **kwargs):
+    def __call__(self, x, class_idx=None, layer_idx=-1, retain_graph=False, **kwargs):  # type: ignore[override]
         """
         Compute the activation map with upsampling and postprocessing.
 
@@ -401,7 +401,7 @@ class GradCAMpp(GradCAM):
 
     """
 
-    def compute_map(self, x, class_idx=None, retain_graph=False, layer_idx=-1, **kwargs):
+    def compute_map(self, x, class_idx=None, retain_graph=False, layer_idx=-1, **kwargs):  # type: ignore[override]
         _, acti, grad = self.nn_module(x, class_idx=class_idx, retain_graph=retain_graph, **kwargs)
         acti, grad = acti[layer_idx], grad[layer_idx]
         b, c, *spatial = grad.shape
diff --git a/tests/test_subpixel_upsample.py b/tests/test_subpixel_upsample.py
index 5abbe57e11..fe9fb1c328 100644
--- a/tests/test_subpixel_upsample.py
+++ b/tests/test_subpixel_upsample.py
@@ -55,9 +55,9 @@
     (2, 1, 32, 16, 8),
 ]
 
-TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_2D_EXTRA)
-TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_3D_EXTRA)
-TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_CONV_BLOCK_EXTRA)
+TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_2D_EXTRA)  # type: ignore
+TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_3D_EXTRA)  # type: ignore
+TEST_CASE_SUBPIXEL.append(TEST_CASE_SUBPIXEL_CONV_BLOCK_EXTRA)  # type: ignore
 
 # add every test back with the pad/pool sequential component omitted
 for tests in list(TEST_CASE_SUBPIXEL):

From 37917e009f49f89149bed5b2c2451e2693132bc1 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 24 Jul 2024 01:58:56 +0800
Subject: [PATCH 099/183] Make ViT and Unetr to be torchscript comaptible
 (#7937)

Fixes #7936

### Description
- Pre-define `self.causal_mask = torch.Tensor()` before register buffer
- Move norm_cross_attn and cross_attn out of if block

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/networks/blocks/crossattention.py   |  4 +++-
 monai/networks/blocks/selfattention.py    |  2 ++
 monai/networks/blocks/transformerblock.py | 13 +++++++------
 tests/test_vit.py                         | 10 +++++-----
 4 files changed, 17 insertions(+), 12 deletions(-)

diff --git a/monai/networks/blocks/crossattention.py b/monai/networks/blocks/crossattention.py
index dc1d5d388e..b888ea3942 100644
--- a/monai/networks/blocks/crossattention.py
+++ b/monai/networks/blocks/crossattention.py
@@ -109,6 +109,8 @@ def __init__(
                 torch.tril(torch.ones(sequence_length, sequence_length)).view(1, 1, sequence_length, sequence_length),
             )
             self.causal_mask: torch.Tensor
+        else:
+            self.causal_mask = torch.Tensor()
 
         self.att_mat = torch.Tensor()
         self.rel_positional_embedding = (
@@ -118,7 +120,7 @@ def __init__(
         )
         self.input_size = input_size
 
-    def forward(self, x: torch.Tensor, context: torch.Tensor | None = None):
+    def forward(self, x: torch.Tensor, context: Optional[torch.Tensor] = None):
         """
         Args:
             x (torch.Tensor): input tensor. B x (s_dim_1 * ... * s_dim_n) x C
diff --git a/monai/networks/blocks/selfattention.py b/monai/networks/blocks/selfattention.py
index 9905e7d036..3ab1e1fd10 100644
--- a/monai/networks/blocks/selfattention.py
+++ b/monai/networks/blocks/selfattention.py
@@ -105,6 +105,8 @@ def __init__(
                 torch.tril(torch.ones(sequence_length, sequence_length)).view(1, 1, sequence_length, sequence_length),
             )
             self.causal_mask: torch.Tensor
+        else:
+            self.causal_mask = torch.Tensor()
 
         self.rel_positional_embedding = (
             get_rel_pos_embedding_layer(rel_pos_embedding, input_size, self.dim_head, self.num_heads)
diff --git a/monai/networks/blocks/transformerblock.py b/monai/networks/blocks/transformerblock.py
index 2458902cba..0aa1697479 100644
--- a/monai/networks/blocks/transformerblock.py
+++ b/monai/networks/blocks/transformerblock.py
@@ -11,6 +11,8 @@
 
 from __future__ import annotations
 
+from typing import Optional
+
 import torch
 import torch.nn as nn
 
@@ -68,13 +70,12 @@ def __init__(
         self.norm2 = nn.LayerNorm(hidden_size)
         self.with_cross_attention = with_cross_attention
 
-        if self.with_cross_attention:
-            self.norm_cross_attn = nn.LayerNorm(hidden_size)
-            self.cross_attn = CrossAttentionBlock(
-                hidden_size=hidden_size, num_heads=num_heads, dropout_rate=dropout_rate, qkv_bias=qkv_bias, causal=False
-            )
+        self.norm_cross_attn = nn.LayerNorm(hidden_size)
+        self.cross_attn = CrossAttentionBlock(
+            hidden_size=hidden_size, num_heads=num_heads, dropout_rate=dropout_rate, qkv_bias=qkv_bias, causal=False
+        )
 
-    def forward(self, x: torch.Tensor, context: torch.Tensor | None = None) -> torch.Tensor:
+    def forward(self, x: torch.Tensor, context: Optional[torch.Tensor] = None) -> torch.Tensor:
         x = x + self.attn(self.norm1(x))
         if self.with_cross_attention:
             x = x + self.cross_attn(self.norm_cross_attn(x), context=context)
diff --git a/tests/test_vit.py b/tests/test_vit.py
index d27c10f95e..d638c0116a 100644
--- a/tests/test_vit.py
+++ b/tests/test_vit.py
@@ -30,7 +30,7 @@
                         for mlp_dim in [3072]:
                             for num_layers in [4]:
                                 for num_classes in [8]:
-                                    for pos_embed in ["conv", "perceptron"]:
+                                    for proj_type in ["conv", "perceptron"]:
                                         for classification in [False, True]:
                                             for nd in (2, 3):
                                                 test_case = [
@@ -42,7 +42,7 @@
                                                         "mlp_dim": mlp_dim,
                                                         "num_layers": num_layers,
                                                         "num_heads": num_heads,
-                                                        "pos_embed": pos_embed,
+                                                        "proj_type": proj_type,
                                                         "classification": classification,
                                                         "num_classes": num_classes,
                                                         "dropout_rate": dropout_rate,
@@ -87,7 +87,7 @@ def test_ill_arg(
         mlp_dim,
         num_layers,
         num_heads,
-        pos_embed,
+        proj_type,
         classification,
         dropout_rate,
     ):
@@ -100,12 +100,12 @@ def test_ill_arg(
                 mlp_dim=mlp_dim,
                 num_layers=num_layers,
                 num_heads=num_heads,
-                pos_embed=pos_embed,
+                proj_type=proj_type,
                 classification=classification,
                 dropout_rate=dropout_rate,
             )
 
-    @parameterized.expand(TEST_CASE_Vit)
+    @parameterized.expand(TEST_CASE_Vit[:1])
     @SkipIfBeforePyTorchVersion((1, 9))
     def test_script(self, input_param, input_shape, _):
         net = ViT(**(input_param))

From 316934aba4093ff4657140819c8797177b2b0259 Mon Sep 17 00:00:00 2001
From: Mingxin Zheng <18563433+mingxin-zheng@users.noreply.github.com>
Date: Wed, 24 Jul 2024 21:58:06 +0800
Subject: [PATCH 100/183] Add checks for monai bundles after download and warn
 if incompatible (#7938)

Fixes #7930 .

### Description

Check the monai version in metadata JSON and warn if the version is
newer than the package being used.


### Demonstration

Warning when the version is hardcoded to 1.2 from monaihosting
```
root@MS-7D31:/workspace/MONAI# python -m monai.bundle download spleen_ct_segmentation
2024-07-23 11:00:31,286 - INFO - --- input summary of monai.bundle.scripts.download ---
2024-07-23 11:00:31,286 - INFO - > name: 'spleen_ct_segmentation'
2024-07-23 11:00:31,286 - INFO - > source: 'monaihosting'
2024-07-23 11:00:31,286 - INFO - > remove_prefix: 'monai_'
2024-07-23 11:00:31,286 - INFO - > progress: True
2024-07-23 11:00:31,286 - INFO - ---


2024-07-23 11:00:31,985 - INFO - Expected md5 is None, skip md5 check for file /root/.cache/torch/hub/bundle/spleen_ct_segmentation_v0.5.8.zip.
2024-07-23 11:00:31,986 - INFO - File exists: /root/.cache/torch/hub/bundle/spleen_ct_segmentation_v0.5.8.zip, skipped downloading.
2024-07-23 11:00:31,986 - INFO - Writing into directory: /root/.cache/torch/hub/bundle.
2024-07-23 11:00:32,176 - WARNING - Your MONAI version is 1.2, but the bundle is built on MONAI version 1.3.2.
```

Auto select version if the download src is from NGC
```
root@MS-7D31:/workspace/MONAI# BUNDLE_DOWNLOAD_SRC=ngc python -m monai.bundle download spleen_ct_segmentation
2024-07-23 11:02:12,277 - INFO - --- input summary of monai.bundle.scripts.download ---
2024-07-23 11:02:12,277 - INFO - > name: 'spleen_ct_segmentation'
2024-07-23 11:02:12,277 - INFO - > source: 'ngc'
2024-07-23 11:02:12,277 - INFO - > remove_prefix: 'monai_'
2024-07-23 11:02:12,277 - INFO - > progress: True
2024-07-23 11:02:12,277 - INFO - ---


monai_spleen_ct_segmentation_v0.3.7.zip: 34.0MB [00:01, 24.1MB/s]
2024-07-23 11:02:17,953 - INFO - Downloaded: /root/.cache/torch/hub/bundle/monai_spleen_ct_segmentation_v0.3.7.zip
2024-07-23 11:02:17,954 - INFO - Expected md5 is None, skip md5 check for file /root/.cache/torch/hub/bundle/monai_spleen_ct_segmentation_v0.3.7.zip.
2024-07-23 11:02:17,954 - INFO - Writing into directory: /root/.cache/torch/hub/bundle/spleen_ct_segmentation.
```

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Mingxin Zheng <mingxinz@nvidia.com>
Signed-off-by: Mingxin Zheng <18563433+mingxin-zheng@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/bundle/scripts.py       | 138 +++++++++++++++++++++++++++-------
 tests/test_bundle_download.py |  51 +++++++++++++
 2 files changed, 162 insertions(+), 27 deletions(-)

diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index 56146546e8..4967b6cf50 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -27,7 +27,7 @@
 import torch
 from torch.cuda import is_available
 
-from monai.apps.mmars.mmars import _get_all_ngc_models
+from monai._version import get_versions
 from monai.apps.utils import _basename, download_url, extractall, get_logger
 from monai.bundle.config_item import ConfigComponent
 from monai.bundle.config_parser import ConfigParser
@@ -67,6 +67,9 @@
 DEFAULT_DOWNLOAD_SOURCE = os.environ.get("BUNDLE_DOWNLOAD_SRC", "monaihosting")
 PPRINT_CONFIG_N = 5
 
+MONAI_HOSTING_BASE_URL = "https://api.ngc.nvidia.com/v2/models/nvidia/monaihosting"
+NGC_BASE_URL = "https://api.ngc.nvidia.com/v2/models/nvidia/monaitoolkit"
+
 
 def update_kwargs(args: str | dict | None = None, ignore_none: bool = True, **kwargs: Any) -> dict:
     """
@@ -169,16 +172,19 @@ def _get_git_release_url(repo_owner: str, repo_name: str, tag_name: str, filenam
 
 
 def _get_ngc_bundle_url(model_name: str, version: str) -> str:
-    return f"https://api.ngc.nvidia.com/v2/models/nvidia/monaitoolkit/{model_name.lower()}/versions/{version}/zip"
+    return f"{NGC_BASE_URL}/{model_name.lower()}/versions/{version}/zip"
+
+
+def _get_ngc_private_base_url(repo: str) -> str:
+    return f"https://api.ngc.nvidia.com/v2/{repo}/models"
 
 
 def _get_ngc_private_bundle_url(model_name: str, version: str, repo: str) -> str:
-    return f"https://api.ngc.nvidia.com/v2/{repo}/models/{model_name.lower()}/versions/{version}/zip"
+    return f"{_get_ngc_private_base_url(repo)}/{model_name.lower()}/versions/{version}/zip"
 
 
 def _get_monaihosting_bundle_url(model_name: str, version: str) -> str:
-    monaihosting_root_path = "https://api.ngc.nvidia.com/v2/models/nvidia/monaihosting"
-    return f"{monaihosting_root_path}/{model_name.lower()}/versions/{version}/files/{model_name}_v{version}.zip"
+    return f"{MONAI_HOSTING_BASE_URL}/{model_name.lower()}/versions/{version}/files/{model_name}_v{version}.zip"
 
 
 def _download_from_github(repo: str, download_path: Path, filename: str, progress: bool = True) -> None:
@@ -267,8 +273,7 @@ def _get_ngc_token(api_key, retry=0):
 
 
 def _get_latest_bundle_version_monaihosting(name):
-    url = "https://api.ngc.nvidia.com/v2/models/nvidia/monaihosting"
-    full_url = f"{url}/{name.lower()}"
+    full_url = f"{MONAI_HOSTING_BASE_URL}/{name.lower()}"
     requests_get, has_requests = optional_import("requests", name="get")
     if has_requests:
         resp = requests_get(full_url)
@@ -279,18 +284,100 @@ def _get_latest_bundle_version_monaihosting(name):
     return model_info["model"]["latestVersionIdStr"]
 
 
-def _get_latest_bundle_version_private_registry(name, repo, headers=None):
-    url = f"https://api.ngc.nvidia.com/v2/{repo}/models"
-    full_url = f"{url}/{name.lower()}"
-    requests_get, has_requests = optional_import("requests", name="get")
-    if has_requests:
-        headers = {} if headers is None else headers
-        resp = requests_get(full_url, headers=headers)
-        resp.raise_for_status()
-    else:
-        raise ValueError("NGC API requires requests package. Please install it.")
+def _examine_monai_version(monai_version: str) -> tuple[bool, str]:
+    """Examine if the package version is compatible with the MONAI version in the metadata."""
+    version_dict = get_versions()
+    package_version = version_dict.get("version", "0+unknown")
+    if package_version == "0+unknown":
+        return False, "Package version is not available. Skipping version check."
+    if monai_version == "0+unknown":
+        return False, "MONAI version is not specified in the bundle. Skipping version check."
+    # treat rc versions as the same as the release version
+    package_version = re.sub(r"rc\d.*", "", package_version)
+    monai_version = re.sub(r"rc\d.*", "", monai_version)
+    if package_version < monai_version:
+        return (
+            False,
+            f"Your MONAI version is {package_version}, but the bundle is built on MONAI version {monai_version}.",
+        )
+    return True, ""
+
+
+def _check_monai_version(bundle_dir: PathLike, name: str) -> None:
+    """Get the `monai_version` from the metadata.json and compare if it is smaller than the installed `monai` package version"""
+    metadata_file = Path(bundle_dir) / name / "configs" / "metadata.json"
+    if not metadata_file.exists():
+        logger.warning(f"metadata file not found in {metadata_file}.")
+        return
+    with open(metadata_file) as f:
+        metadata = json.load(f)
+    is_compatible, msg = _examine_monai_version(metadata.get("monai_version", "0+unknown"))
+    if not is_compatible:
+        logger.warning(msg)
+
+
+def _list_latest_versions(data: dict, max_versions: int = 3) -> list[str]:
+    """
+    Extract the latest versions from the data dictionary.
+
+    Args:
+        data: the data dictionary.
+        max_versions: the maximum number of versions to return.
+
+    Returns:
+        versions of the latest models in the reverse order of creation date, e.g. ['1.0.0', '0.9.0', '0.8.0'].
+    """
+    # Check if the data is a dictionary and it has the key 'modelVersions'
+    if not isinstance(data, dict) or "modelVersions" not in data:
+        raise ValueError("The data is not a dictionary or it does not have the key 'modelVersions'.")
+
+    # Extract the list of model versions
+    model_versions = data["modelVersions"]
+
+    if (
+        not isinstance(model_versions, list)
+        or len(model_versions) == 0
+        or "createdDate" not in model_versions[0]
+        or "versionId" not in model_versions[0]
+    ):
+        raise ValueError(
+            "The model versions are not a list or it is empty or it does not have the keys 'createdDate' and 'versionId'."
+        )
+
+    # Sort the versions by the 'createdDate' in descending order
+    sorted_versions = sorted(model_versions, key=lambda x: x["createdDate"], reverse=True)
+    return [v["versionId"] for v in sorted_versions[:max_versions]]
+
+
+def _get_latest_bundle_version_ngc(name: str, repo: str | None = None, headers: dict | None = None) -> str:
+    base_url = _get_ngc_private_base_url(repo) if repo else NGC_BASE_URL
+    version_endpoint = base_url + f"/{name.lower()}/versions/"
+
+    if not has_requests:
+        raise ValueError("requests package is required, please install it.")
+
+    version_header = {"Accept-Encoding": "gzip, deflate"}  # Excluding 'zstd' to fit NGC requirements
+    if headers:
+        version_header.update(headers)
+    resp = requests_get(version_endpoint, headers=version_header)
+    resp.raise_for_status()
     model_info = json.loads(resp.text)
-    return model_info["model"]["latestVersionIdStr"]
+    latest_versions = _list_latest_versions(model_info)
+
+    for version in latest_versions:
+        file_endpoint = base_url + f"/{name.lower()}/versions/{version}/files/configs/metadata.json"
+        resp = requests_get(file_endpoint, headers=headers)
+        metadata = json.loads(resp.text)
+        resp.raise_for_status()
+        # if the package version is not available or the model is compatible with the package version
+        is_compatible, _ = _examine_monai_version(metadata["monai_version"])
+        if is_compatible:
+            if version != latest_versions[0]:
+                logger.info(f"Latest version is {latest_versions[0]}, but the compatible version is {version}.")
+            return version
+
+    # if no compatible version is found, return the latest version
+    return latest_versions[0]
 
 
 def _get_latest_bundle_version(
@@ -298,17 +385,13 @@ def _get_latest_bundle_version(
 ) -> dict[str, list[str] | str] | Any | None:
     if source == "ngc":
         name = _add_ngc_prefix(name)
-        model_dict = _get_all_ngc_models(name)
-        for v in model_dict.values():
-            if v["name"] == name:
-                return v["latest"]
-        return None
+        return _get_latest_bundle_version_ngc(name)
     elif source == "monaihosting":
         return _get_latest_bundle_version_monaihosting(name)
     elif source == "ngc_private":
         headers = kwargs.pop("headers", {})
         name = _add_ngc_prefix(name)
-        return _get_latest_bundle_version_private_registry(name, repo, headers)
+        return _get_latest_bundle_version_ngc(name, repo=repo, headers=headers)
     elif source == "github":
         repo_owner, repo_name, tag_name = repo.split("/")
         return get_bundle_versions(name, repo=f"{repo_owner}/{repo_name}", tag=tag_name)["latest_version"]
@@ -470,9 +553,8 @@ def download(
         if version_ is None:
             version_ = _get_latest_bundle_version(source=source_, name=name_, repo=repo_, headers=headers)
         if source_ == "github":
-            if version_ is not None:
-                name_ = "_v".join([name_, version_])
-            _download_from_github(repo=repo_, download_path=bundle_dir_, filename=name_, progress=progress_)
+            name_ver = "_v".join([name_, version_]) if version_ is not None else name_
+            _download_from_github(repo=repo_, download_path=bundle_dir_, filename=name_ver, progress=progress_)
         elif source_ == "monaihosting":
             _download_from_monaihosting(download_path=bundle_dir_, filename=name_, version=version_, progress=progress_)
         elif source_ == "ngc":
@@ -501,6 +583,8 @@ def download(
                 f"got source: {source_}."
             )
 
+    _check_monai_version(bundle_dir_, name_)
+
 
 @deprecated_arg("net_name", since="1.2", removed="1.5", msg_suffix="please use ``model`` instead.")
 @deprecated_arg("net_kwargs", since="1.2", removed="1.5", msg_suffix="please use ``model`` instead.")
diff --git a/tests/test_bundle_download.py b/tests/test_bundle_download.py
index fe7caf5c17..331d228f1e 100644
--- a/tests/test_bundle_download.py
+++ b/tests/test_bundle_download.py
@@ -16,6 +16,7 @@
 import tempfile
 import unittest
 from unittest.case import skipUnless
+from unittest.mock import patch
 
 import numpy as np
 import torch
@@ -24,6 +25,7 @@
 import monai.networks.nets as nets
 from monai.apps import check_hash
 from monai.bundle import ConfigParser, create_workflow, load
+from monai.bundle.scripts import _examine_monai_version, _list_latest_versions, download
 from monai.utils import optional_import
 from tests.utils import (
     SkipIfBeforePyTorchVersion,
@@ -207,6 +209,55 @@ def test_monaihosting_source_download_bundle(self, bundle_files, bundle_name, ve
                     file_path = os.path.join(tempdir, bundle_name, file)
                     self.assertTrue(os.path.exists(file_path))
 
+    @patch("monai.bundle.scripts.get_versions", return_value={"version": "1.2"})
+    def test_examine_monai_version(self, mock_get_versions):
+        self.assertTrue(_examine_monai_version("1.1")[0])  # Should return True, compatible
+        self.assertTrue(_examine_monai_version("1.2rc1")[0])  # Should return True, compatible
+        self.assertFalse(_examine_monai_version("1.3")[0])  # Should return False, not compatible
+
+    @patch("monai.bundle.scripts.get_versions", return_value={"version": "1.2rc1"})
+    def test_examine_monai_version_rc(self, mock_get_versions):
+        self.assertTrue(_examine_monai_version("1.2")[0])  # Should return True, compatible
+        self.assertFalse(_examine_monai_version("1.3")[0])  # Should return False, not compatible
+
+    def test_list_latest_versions(self):
+        """Test listing of the latest versions."""
+        data = {
+            "modelVersions": [
+                {"createdDate": "2021-01-01", "versionId": "1.0"},
+                {"createdDate": "2021-01-02", "versionId": "1.1"},
+                {"createdDate": "2021-01-03", "versionId": "1.2"},
+            ]
+        }
+        self.assertEqual(_list_latest_versions(data), ["1.2", "1.1", "1.0"])
+        self.assertEqual(_list_latest_versions(data, max_versions=2), ["1.2", "1.1"])
+        data = {
+            "modelVersions": [
+                {"createdDate": "2021-01-01", "versionId": "1.0"},
+                {"createdDate": "2021-01-02", "versionId": "1.1"},
+            ]
+        }
+        self.assertEqual(_list_latest_versions(data), ["1.1", "1.0"])
+
+    @skip_if_quick
+    @patch("monai.bundle.scripts.get_versions", return_value={"version": "1.2"})
+    def test_download_monaihosting(self, mock_get_versions):
+        """Test checking MONAI version from a metadata file."""
+        with patch("monai.bundle.scripts.logger") as mock_logger:
+            with tempfile.TemporaryDirectory() as tempdir:
+                download(name="spleen_ct_segmentation", bundle_dir=tempdir, source="monaihosting")
+                # Should have a warning message because the latest version is using monai > 1.2
+                mock_logger.warning.assert_called_once()
+
+    @skip_if_quick
+    @patch("monai.bundle.scripts.get_versions", return_value={"version": "1.2"})
+    def test_download_ngc(self, mock_get_versions):
+        """Test checking MONAI version from a metadata file."""
+        with patch("monai.bundle.scripts.logger") as mock_logger:
+            with tempfile.TemporaryDirectory() as tempdir:
+                download(name="spleen_ct_segmentation", bundle_dir=tempdir, source="ngc")
+                mock_logger.warning.assert_not_called()
+
 
 @skip_if_no_cuda
 class TestLoad(unittest.TestCase):

From 12d00ce1369e37cb06f483735ef83674a208b031 Mon Sep 17 00:00:00 2001
From: Hans Johnson <hans-johnson@uiowa.edu>
Date: Wed, 24 Jul 2024 10:35:04 -0500
Subject: [PATCH 101/183] Cleanup warnings when collecting tests (#7914)

Minimize warning information so that important warnings and errors are
more easily accessible

warnings identified with
```bash
python3  -W ignore::DeprecationWarning -m pytest -k dummy
```

Fixing warnings like:
cannot collect test class 'TestTimeAugmentation' because it has a
__init__ constructor

Use `__test__ = False` to suppress attempted collection in those non
test-classes

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.

Signed-off-by: Hans Johnson <hans-johnson@uiowa.edu>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/data/test_time_augmentation.py     | 2 ++
 tests/test_arraydataset.py               | 1 +
 tests/test_auto3dseg.py                  | 6 ++++++
 tests/test_convert_data_type.py          | 1 +
 tests/test_handler_prob_map_producer.py  | 2 ++
 tests/test_handler_validation.py         | 1 +
 tests/test_image_filter.py               | 1 +
 tests/test_prepare_batch_default.py      | 1 +
 tests/test_prepare_batch_default_dist.py | 1 +
 tests/test_prepare_batch_extra_input.py  | 1 +
 tests/test_prepare_batch_hovernet.py     | 1 +
 tests/test_torchscript_utils.py          | 1 +
 12 files changed, 19 insertions(+)

diff --git a/monai/data/test_time_augmentation.py b/monai/data/test_time_augmentation.py
index 23572dcef4..bcd5ea91a9 100644
--- a/monai/data/test_time_augmentation.py
+++ b/monai/data/test_time_augmentation.py
@@ -106,6 +106,8 @@ class TestTimeAugmentation:
             mode, mean, std, vvc = tt_aug(test_data)
     """
 
+    __test__ = False  # indicate to pytest that this class is not intended for collection
+
     def __init__(
         self,
         transform: InvertibleTransform,
diff --git a/tests/test_arraydataset.py b/tests/test_arraydataset.py
index b61b3c139c..03239a9764 100644
--- a/tests/test_arraydataset.py
+++ b/tests/test_arraydataset.py
@@ -40,6 +40,7 @@
 
 
 class TestCompose(Compose):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def __call__(self, input_, lazy=False):
         img = self.transforms[0](input_)
diff --git a/tests/test_auto3dseg.py b/tests/test_auto3dseg.py
index 6be33bf6ca..5273f0663a 100644
--- a/tests/test_auto3dseg.py
+++ b/tests/test_auto3dseg.py
@@ -123,6 +123,8 @@ class TestOperations(Operations):
     Test example for user operation
     """
 
+    __test__ = False  # indicate to pytest that this class is not intended for collection
+
     def __init__(self) -> None:
         self.data = {"max": np.max, "mean": np.mean, "min": np.min}
 
@@ -132,6 +134,8 @@ class TestAnalyzer(Analyzer):
     Test example for a simple Analyzer
     """
 
+    __test__ = False  # indicate to pytest that this class is not intended for collection
+
     def __init__(self, key, report_format, stats_name="test"):
         self.key = key
         super().__init__(stats_name, report_format)
@@ -149,6 +153,8 @@ class TestImageAnalyzer(Analyzer):
     Test example for a simple Analyzer
     """
 
+    __test__ = False  # indicate to pytest that this class is not intended for collection
+
     def __init__(self, image_key="image", stats_name="test_image"):
         self.image_key = image_key
         report_format = {"test_stats": None}
diff --git a/tests/test_convert_data_type.py b/tests/test_convert_data_type.py
index b95539f4b7..a27a05cf28 100644
--- a/tests/test_convert_data_type.py
+++ b/tests/test_convert_data_type.py
@@ -73,6 +73,7 @@
 
 
 class TestTensor(torch.Tensor):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
     pass
 
 
diff --git a/tests/test_handler_prob_map_producer.py b/tests/test_handler_prob_map_producer.py
index 347f8cb92c..406fe77c8f 100644
--- a/tests/test_handler_prob_map_producer.py
+++ b/tests/test_handler_prob_map_producer.py
@@ -30,6 +30,7 @@
 
 
 class TestDataset(Dataset):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def __init__(self, name, size):
         super().__init__(
@@ -64,6 +65,7 @@ def __getitem__(self, index):
 
 
 class TestEvaluator(Evaluator):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def _iteration(self, engine, batchdata):
         return batchdata
diff --git a/tests/test_handler_validation.py b/tests/test_handler_validation.py
index 752b1d3df7..92f8578f11 100644
--- a/tests/test_handler_validation.py
+++ b/tests/test_handler_validation.py
@@ -22,6 +22,7 @@
 
 
 class TestEvaluator(Evaluator):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def _iteration(self, engine, batchdata):
         engine.state.output = "called"
diff --git a/tests/test_image_filter.py b/tests/test_image_filter.py
index adc9dade9c..76e38d94f4 100644
--- a/tests/test_image_filter.py
+++ b/tests/test_image_filter.py
@@ -38,6 +38,7 @@
 
 
 class TestModule(torch.nn.Module):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def __init__(self):
         super().__init__()
diff --git a/tests/test_prepare_batch_default.py b/tests/test_prepare_batch_default.py
index 9aa498866f..093468ce27 100644
--- a/tests/test_prepare_batch_default.py
+++ b/tests/test_prepare_batch_default.py
@@ -21,6 +21,7 @@
 
 
 class TestNet(torch.nn.Module):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def forward(self, x: torch.Tensor):
         return x
diff --git a/tests/test_prepare_batch_default_dist.py b/tests/test_prepare_batch_default_dist.py
index 0c53a74834..53a79575e6 100644
--- a/tests/test_prepare_batch_default_dist.py
+++ b/tests/test_prepare_batch_default_dist.py
@@ -43,6 +43,7 @@
 
 
 class TestNet(torch.nn.Module):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def forward(self, x: torch.Tensor):
         return x
diff --git a/tests/test_prepare_batch_extra_input.py b/tests/test_prepare_batch_extra_input.py
index f20c6e7352..3c53cc6481 100644
--- a/tests/test_prepare_batch_extra_input.py
+++ b/tests/test_prepare_batch_extra_input.py
@@ -36,6 +36,7 @@
 
 
 class TestNet(torch.nn.Module):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def forward(self, x: torch.Tensor, t1=None, t2=None, t3=None):
         return {"x": x, "t1": t1, "t2": t2, "t3": t3}
diff --git a/tests/test_prepare_batch_hovernet.py b/tests/test_prepare_batch_hovernet.py
index 773fcb53bf..ae9554a3e8 100644
--- a/tests/test_prepare_batch_hovernet.py
+++ b/tests/test_prepare_batch_hovernet.py
@@ -28,6 +28,7 @@
 
 
 class TestNet(torch.nn.Module):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def forward(self, x: torch.Tensor):
         return {HoVerNetBranch.NP: torch.tensor([1, 2]), HoVerNetBranch.NC: torch.tensor([4, 4]), HoVerNetBranch.HV: 16}
diff --git a/tests/test_torchscript_utils.py b/tests/test_torchscript_utils.py
index 6f8f231829..5a5fb47864 100644
--- a/tests/test_torchscript_utils.py
+++ b/tests/test_torchscript_utils.py
@@ -23,6 +23,7 @@
 
 
 class TestModule(torch.nn.Module):
+    __test__ = False  # indicate to pytest that this class is not intended for collection
 
     def forward(self, x):
         return x + 10

From 3f0c768c491168202b558721667806a2172f4b8a Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 25 Jul 2024 23:57:59 +0800
Subject: [PATCH 102/183] Fix Incompatible types in assignment issue (#7950)

Fixes #7947

### Description
Add type ignore in several redefinitions cases

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/auto3dseg/analyzer.py            | 2 +-
 monai/metrics/cumulative_average.py    | 2 ++
 monai/metrics/panoptic_quality.py      | 2 +-
 monai/metrics/rocauc.py                | 4 ++--
 monai/transforms/croppad/functional.py | 2 +-
 monai/visualize/img2tensorboard.py     | 4 +++-
 6 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/monai/auto3dseg/analyzer.py b/monai/auto3dseg/analyzer.py
index 37f3faea21..e60327b551 100644
--- a/monai/auto3dseg/analyzer.py
+++ b/monai/auto3dseg/analyzer.py
@@ -470,7 +470,7 @@ def __call__(self, data: Mapping[Hashable, MetaTensor]) -> dict[Hashable, MetaTe
 
         unique_label = unique(ndas_label)
         if isinstance(ndas_label, (MetaTensor, torch.Tensor)):
-            unique_label = unique_label.data.cpu().numpy()
+            unique_label = unique_label.data.cpu().numpy()  # type: ignore[assignment]
 
         unique_label = unique_label.astype(np.int16).tolist()
 
diff --git a/monai/metrics/cumulative_average.py b/monai/metrics/cumulative_average.py
index e55e7b8576..dccf7b094b 100644
--- a/monai/metrics/cumulative_average.py
+++ b/monai/metrics/cumulative_average.py
@@ -65,6 +65,7 @@ def get_current(self, to_numpy: bool = True) -> NdarrayOrTensor:
         if self.val is None:
             return 0
 
+        val: NdarrayOrTensor
         val = self.val.clone()
         val[~torch.isfinite(val)] = 0
 
@@ -96,6 +97,7 @@ def aggregate(self, to_numpy: bool = True) -> NdarrayOrTensor:
             dist.all_reduce(sum)
             dist.all_reduce(count)
 
+        val: NdarrayOrTensor
         val = torch.where(count > 0, sum / count, sum)
 
         if to_numpy:
diff --git a/monai/metrics/panoptic_quality.py b/monai/metrics/panoptic_quality.py
index 05175ba0fb..7c9d59c264 100644
--- a/monai/metrics/panoptic_quality.py
+++ b/monai/metrics/panoptic_quality.py
@@ -274,7 +274,7 @@ def _get_paired_iou(
 
         return paired_iou, paired_true, paired_pred
 
-    pairwise_iou = pairwise_iou.cpu().numpy()
+    pairwise_iou = pairwise_iou.cpu().numpy()  # type: ignore[assignment]
     paired_true, paired_pred = linear_sum_assignment(-pairwise_iou)
     paired_iou = pairwise_iou[paired_true, paired_pred]
     paired_true = torch.as_tensor(list(paired_true[paired_iou > match_iou_threshold] + 1), device=device)
diff --git a/monai/metrics/rocauc.py b/monai/metrics/rocauc.py
index 56d9faa9dd..57a8a072b4 100644
--- a/monai/metrics/rocauc.py
+++ b/monai/metrics/rocauc.py
@@ -88,8 +88,8 @@ def _calculate(y_pred: torch.Tensor, y: torch.Tensor) -> float:
 
     n = len(y)
     indices = y_pred.argsort()
-    y = y[indices].cpu().numpy()
-    y_pred = y_pred[indices].cpu().numpy()
+    y = y[indices].cpu().numpy()  # type: ignore[assignment]
+    y_pred = y_pred[indices].cpu().numpy()  # type: ignore[assignment]
     nneg = auc = tmp_pos = tmp_neg = 0.0
 
     for i in range(n):
diff --git a/monai/transforms/croppad/functional.py b/monai/transforms/croppad/functional.py
index a8286fb90c..361ec48dcd 100644
--- a/monai/transforms/croppad/functional.py
+++ b/monai/transforms/croppad/functional.py
@@ -48,7 +48,7 @@ def _np_pad(img: NdarrayTensor, pad_width: list[tuple[int, int]], mode: str, **k
             warnings.warn(f"Padding: moving img {img.shape} from cuda to cpu for dtype={img.dtype} mode={mode}.")
         img_np = img.detach().cpu().numpy()
     else:
-        img_np = img
+        img_np = np.asarray(img)
     mode = convert_pad_mode(dst=img_np, mode=mode).value
     if mode == "constant" and "value" in kwargs:
         kwargs["constant_values"] = kwargs.pop("value")
diff --git a/monai/visualize/img2tensorboard.py b/monai/visualize/img2tensorboard.py
index e7884e9b1f..677640bd04 100644
--- a/monai/visualize/img2tensorboard.py
+++ b/monai/visualize/img2tensorboard.py
@@ -176,7 +176,9 @@ def plot_2d_or_3d_image(
     # as the `d` data has no batch dim, reduce the spatial dim index if positive
     frame_dim = frame_dim - 1 if frame_dim > 0 else frame_dim
 
-    d: np.ndarray = data_index.detach().cpu().numpy() if isinstance(data_index, torch.Tensor) else data_index
+    d: np.ndarray = (
+        data_index.detach().cpu().numpy() if isinstance(data_index, torch.Tensor) else np.asarray(data_index)
+    )
 
     if d.ndim == 2:
         d = rescale_array(d, 0, 1)  # type: ignore

From 2e53df78e580131046dc8db7f7638063db1f5045 Mon Sep 17 00:00:00 2001
From: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Date: Thu, 25 Jul 2024 18:56:25 +0100
Subject: [PATCH 103/183] Adding metadata schema to the code base itself
 (#7409)

Fixes #7303 #6959.

### Description

This adds the schema file into the code base (but this maybe should be
elsewhere). The changes implement a number of new things:

* Moved definitions into a `$defs` section per the JSON schema standard
* Permits multiple input arguments and return results from networks with
arbitrary names using the `patternProperties` mechanism
* Allows the types of inputs and outputs to be, additional to just
tensors, numbers, booleans, or strings
* Outputs after post processing can be specified with the
`post_processed_outputs` section if they are significantly changed with
the post-process transforms defined in scripts
* Multiple network IO formats can be specified in addition to
`network_data_format`, these must follow the pattern
`<name>_data_format`
* `required_packages_version` added in addition to
`optional_packages_version`

#7253 depends on this schema change.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Eric Kerfoot <eric.kerfoot@kcl.ac.uk>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
---
 docs/source/mb_specification.rst | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/docs/source/mb_specification.rst b/docs/source/mb_specification.rst
index cedafa0d23..56d660e35c 100644
--- a/docs/source/mb_specification.rst
+++ b/docs/source/mb_specification.rst
@@ -63,12 +63,12 @@ This file contains the metadata information relating to the model, including wha
 * **monai_version**: version of MONAI the bundle was generated on, later versions expected to work.
 * **pytorch_version**: version of Pytorch the bundle was generated on, later versions expected to work.
 * **numpy_version**: version of Numpy the bundle was generated on, later versions expected to work.
-* **optional_packages_version**: dictionary relating optional package names to their versions, these packages are not needed but are recommended to be installed with this stated minimum version.
+* **required_packages_version**: dictionary relating required package names to their versions. These are packages in addition to the base requirements of MONAI which this bundle absolutely needs. For example, if the bundle must load Nifti files the Nibabel package will be required.
 * **task**: plain-language description of what the model is meant to do.
 * **description**: longer form plain-language description of what the model is, what it does, etc.
 * **authors**: state author(s) of the model.
 * **copyright**: state model copyright.
-* **network_data_format**: defines the format, shape, and meaning of inputs and outputs to the model, contains keys "inputs" and "outputs" relating named inputs/outputs to their format specifiers (defined below).
+* **network_data_format**: defines the format, shape, and meaning of inputs and outputs to the (primary) model, contains keys "inputs" and "outputs" relating named inputs/outputs to their format specifiers (defined below). There is also an optional "post_processed_outputs" key stating the format of "outputs" after postprocessing transforms are applied, this is used to describe the final output from the bundle if it varies from the raw network output. These keys can also relate to primitive values (number, string, boolean), instead of the tensor format specified below.
 
 Tensor format specifiers are used to define input and output tensors and their meanings, and must be a dictionary containing at least these keys:
 
@@ -89,6 +89,8 @@ Optional keys:
 * **data_source**: description of where training/validation can be sourced.
 * **data_type**: type of source data used for training/validation.
 * **references**: list of published referenced relating to the model.
+* **supported_apps**: list of supported applications which use bundles, eg. 'monai-label' would be present if the bundle is compatible with MONAI Label applications.
+* **\*_data_format**: defines the format, shape, and meaning of inputs and outputs to additional models which are secondary to the main model. This contains the same sort of information as **network_data_format** which describes networks providing secondary functionality, eg. a localisation network used to identify ROI in an image for cropping before data is sent to the primary network of this bundle.
 
 The format for tensors used as inputs and outputs can be used to specify semantic meaning of these values, and later is used by software handling bundles to determine how to process and interpret this data. There are various types of image data that MONAI is uses, and other data types such as point clouds, dictionary sequences, time signals, and others. The following list is provided as a set of supported definitions of what a tensor "format" is but is not exhaustive and users can provide their own which would be left up to the model users to interpret:
 
@@ -124,7 +126,7 @@ An example JSON metadata file:
       "monai_version": "0.9.0",
       "pytorch_version": "1.10.0",
       "numpy_version": "1.21.2",
-      "optional_packages_version": {"nibabel": "3.2.1"},
+      "required_packages_version": {"nibabel": "3.2.1"},
       "task": "Decathlon spleen segmentation",
       "description": "A pre-trained model for volumetric (3D) segmentation of the spleen from CT image",
       "authors": "MONAI team",

From 54019e4137ecccc9b80883299590a349dcaf6f6e Mon Sep 17 00:00:00 2001
From: Virginia Fernandez <61539159+virginiafdez@users.noreply.github.com>
Date: Tue, 30 Jul 2024 08:15:07 +0100
Subject: [PATCH 104/183] Addition of norm_eps (#7962)

Fixes # .

### Description

Addition of norm_eps to spade_autoencoderkl.py as per
https://github.com/Project-MONAI/MONAI/issues/7946.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Virginia Fernandez <vf19@lihe044-pc.isd.kcl.ac.uk>
Signed-off-by: Virginia Fernandez <virginia.fernandez@kcl.ac.uk>
Co-authored-by: Virginia Fernandez <vf19@lihe044-pc.isd.kcl.ac.uk>
Co-authored-by: Virginia Fernandez <virginia.fernandez@kcl.ac.uk>
---
 monai/networks/nets/spade_autoencoderkl.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/monai/networks/nets/spade_autoencoderkl.py b/monai/networks/nets/spade_autoencoderkl.py
index 294b121c94..d5794a9227 100644
--- a/monai/networks/nets/spade_autoencoderkl.py
+++ b/monai/networks/nets/spade_autoencoderkl.py
@@ -59,7 +59,7 @@ def __init__(
             label_nc=label_nc,
             norm_nc=in_channels,
             norm="GROUP",
-            norm_params={"num_groups": norm_num_groups, "affine": False},
+            norm_params={"num_groups": norm_num_groups, "affine": False, "eps": norm_eps},
             hidden_channels=spade_intermediate_channels,
             kernel_size=3,
             spatial_dims=spatial_dims,
@@ -77,7 +77,7 @@ def __init__(
             label_nc=label_nc,
             norm_nc=out_channels,
             norm="GROUP",
-            norm_params={"num_groups": norm_num_groups, "affine": False},
+            norm_params={"num_groups": norm_num_groups, "affine": False, "eps": norm_eps},
             hidden_channels=spade_intermediate_channels,
             kernel_size=3,
             spatial_dims=spatial_dims,

From f1ef3e88f4bc594779c0c4188883f78bf6d1efff Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 30 Jul 2024 18:57:57 +0800
Subject: [PATCH 105/183] Rename `optional_packages_version` to
 `required_packages_version` (#7253)

Fixes #6959.

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: KumoLiu <yunl@nvidia.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: Mingxin Zheng <18563433+mingxin-zheng@users.noreply.github.com>
---
 monai/bundle/config_parser.py       |  4 ++--
 monai/bundle/reference_resolver.py  | 19 ++++++++++++++++++-
 monai/bundle/utils.py               |  4 +++-
 tests/testing_data/data_config.json |  4 ++--
 tests/testing_data/metadata.json    |  4 ++--
 5 files changed, 27 insertions(+), 8 deletions(-)

diff --git a/monai/bundle/config_parser.py b/monai/bundle/config_parser.py
index 829036af6f..a2ffeedc92 100644
--- a/monai/bundle/config_parser.py
+++ b/monai/bundle/config_parser.py
@@ -118,7 +118,7 @@ def __init__(
         self.ref_resolver = ReferenceResolver()
         if config is None:
             config = {self.meta_key: {}}
-        self.set(config=config)
+        self.set(config=self.ref_resolver.normalize_meta_id(config))
 
     def __repr__(self):
         return f"{self.config}"
@@ -221,7 +221,7 @@ def set(self, config: Any, id: str = "", recursive: bool = True) -> None:
                 if isinstance(conf_, dict) and k not in conf_:
                     conf_[k] = {}
                 conf_ = conf_[k if isinstance(conf_, dict) else int(k)]
-        self[ReferenceResolver.normalize_id(id)] = config
+        self[ReferenceResolver.normalize_id(id)] = self.ref_resolver.normalize_meta_id(config)
 
     def update(self, pairs: dict[str, Any]) -> None:
         """
diff --git a/monai/bundle/reference_resolver.py b/monai/bundle/reference_resolver.py
index b36f2cc4a5..050cd75fa7 100644
--- a/monai/bundle/reference_resolver.py
+++ b/monai/bundle/reference_resolver.py
@@ -17,7 +17,7 @@
 from typing import Any, Iterator
 
 from monai.bundle.config_item import ConfigComponent, ConfigExpression, ConfigItem
-from monai.bundle.utils import ID_REF_KEY, ID_SEP_KEY
+from monai.bundle.utils import DEPRECATED_ID_MAPPING, ID_REF_KEY, ID_SEP_KEY
 from monai.utils import allow_missing_reference, look_up_option
 
 __all__ = ["ReferenceResolver"]
@@ -202,6 +202,23 @@ def normalize_id(cls, id: str | int) -> str:
         """
         return str(id).replace("#", cls.sep)  # backward compatibility `#` is the old separator
 
+    def normalize_meta_id(self, config: Any) -> Any:
+        """
+        Update deprecated identifiers in `config` using `DEPRECATED_ID_MAPPING`.
+        This will replace names that are marked as deprecated with their replacement.
+
+        Args:
+            config: input config to be updated.
+        """
+        if isinstance(config, dict):
+            for _id, _new_id in DEPRECATED_ID_MAPPING.items():
+                if _id in config.keys():
+                    warnings.warn(
+                        f"Detected deprecated name '{_id}' in configuration file, replacing with '{_new_id}'."
+                    )
+                    config[_new_id] = config.pop(_id)
+        return config
+
     @classmethod
     def split_id(cls, id: str | int, last: bool = False) -> list[str]:
         """
diff --git a/monai/bundle/utils.py b/monai/bundle/utils.py
index 0f17422ba5..50d2608f4c 100644
--- a/monai/bundle/utils.py
+++ b/monai/bundle/utils.py
@@ -36,7 +36,7 @@
     "monai_version": _conf_values["MONAI"],
     "pytorch_version": str(_conf_values["Pytorch"]).split("+")[0].split("a")[0],  # 1.9.0a0+df837d0 or 1.13.0+cu117
     "numpy_version": _conf_values["Numpy"],
-    "optional_packages_version": {},
+    "required_packages_version": {},
     "task": "Describe what the network predicts",
     "description": "A longer description of what the network does, use context, inputs, outputs, etc.",
     "authors": "Your Name Here",
@@ -157,6 +157,8 @@
 
 DEFAULT_EXP_MGMT_SETTINGS = {"mlflow": DEFAULT_MLFLOW_SETTINGS}  # default experiment management settings
 
+DEPRECATED_ID_MAPPING = {"optional_packages_version": "required_packages_version"}
+
 
 def load_bundle_config(bundle_path: str, *config_names: str, **load_kw_args: Any) -> Any:
     """
diff --git a/tests/testing_data/data_config.json b/tests/testing_data/data_config.json
index 8b1d2868b7..79033dd0d6 100644
--- a/tests/testing_data/data_config.json
+++ b/tests/testing_data/data_config.json
@@ -162,9 +162,9 @@
     },
     "configs": {
         "test_meta_file": {
-            "url": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
+            "url": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
             "hash_type": "md5",
-            "hash_val": "662135097106b71067cd1fc657f8720f"
+            "hash_val": "06954cad2cc5d3784e72077ac76f0fc8"
         }
     }
 }
diff --git a/tests/testing_data/metadata.json b/tests/testing_data/metadata.json
index 98a17b73c5..29737e3a9d 100644
--- a/tests/testing_data/metadata.json
+++ b/tests/testing_data/metadata.json
@@ -1,5 +1,5 @@
 {
-    "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
+    "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
     "version": "0.1.0",
     "changelog": {
         "0.1.0": "complete the model package",
@@ -8,7 +8,7 @@
     "monai_version": "0.9.0",
     "pytorch_version": "1.10.0",
     "numpy_version": "1.21.2",
-    "optional_packages_version": {
+    "required_packages_version": {
         "nibabel": "3.2.1"
     },
     "task": "Decathlon spleen segmentation",

From 139b62c4aa161969ad1126c4feeec88c9833d4ef Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 1 Aug 2024 19:30:05 +0800
Subject: [PATCH 106/183] Fix outdated link in the docs (#7971)

Fixes #7968

### Description

Fix outdated link in the docs

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/index.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/source/index.rst b/docs/source/index.rst
index b6c8c22f98..85adee7e44 100644
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -37,7 +37,7 @@ Features
 Getting started
 ---------------
 
-`MedNIST demo <https://colab.research.google.com/drive/1wy8XUSnNWlhDNazFdvGBHLfdkGvOHBKe>`_ and `MONAI for PyTorch Users <https://colab.research.google.com/drive/1boqy7ENpKrqaJoxFlbHIBnIODAs1Ih1T>`_ are available on Colab.
+`MedNIST demo <https://colab.research.google.com/github/Project-MONAI/tutorials/blob/main/2d_classification/mednist_tutorial.ipynb>`_ and `MONAI for PyTorch Users <https://colab.research.google.com/github/Project-MONAI/tutorials/blob/main/modules/developer_guide.ipynb>`_ are available on Colab.
 
 Examples and notebook tutorials are located at `Project-MONAI/tutorials <https://github.com/Project-MONAI/tutorials>`_.
 

From 1ece8a5c0b187a737f02ec44e7436239e3343e43 Mon Sep 17 00:00:00 2001
From: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Date: Fri, 2 Aug 2024 13:19:13 +0800
Subject: [PATCH 107/183] 7982-fix-ci-issue (#7983)

Fixes #7982

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/pythonapp.yml | 1 +
 requirements-dev.txt            | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/.github/workflows/pythonapp.yml b/.github/workflows/pythonapp.yml
index cd6b6ccede..fe04f96a80 100644
--- a/.github/workflows/pythonapp.yml
+++ b/.github/workflows/pythonapp.yml
@@ -149,6 +149,7 @@ jobs:
         key: ${{ runner.os }}-pip-${{ steps.pip-cache.outputs.datew }}
     - name: Install dependencies
       run: |
+        find  /opt/hostedtoolcache/* -maxdepth 0 ! -name 'Python' -exec rm -rf {} \;
         python -m pip install --user --upgrade pip setuptools wheel twine
         # install the latest pytorch for testing
         # however, "pip install monai*.tar.gz" will build cpp/cuda with an isolated
diff --git a/requirements-dev.txt b/requirements-dev.txt
index ced783443e..72ba210093 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -21,7 +21,7 @@ black>=22.12
 isort>=5.1
 ruff
 pytype>=2020.6.1; platform_system != "Windows"
-types-pkg_resources
+types-setuptools
 mypy>=1.5.0
 ninja
 torchvision

From ae5a04d685ade10d886db1918d68e292a6096a17 Mon Sep 17 00:00:00 2001
From: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Date: Sat, 3 Aug 2024 22:37:36 +0800
Subject: [PATCH 108/183] 7973-add-ngc-prefix (#7974)

Fixes #7973 .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/bundle/scripts.py | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index 4967b6cf50..6dd83c1f81 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -217,10 +217,15 @@ def _remove_ngc_prefix(name: str, prefix: str = "monai_") -> str:
 
 
 def _download_from_ngc(
-    download_path: Path, filename: str, version: str, remove_prefix: str | None, progress: bool
+    download_path: Path,
+    filename: str,
+    version: str,
+    prefix: str = "monai_",
+    remove_prefix: str | None = "monai_",
+    progress: bool = True,
 ) -> None:
     # ensure prefix is contained
-    filename = _add_ngc_prefix(filename)
+    filename = _add_ngc_prefix(filename, prefix=prefix)
     url = _get_ngc_bundle_url(model_name=filename, version=version)
     filepath = download_path / f"{filename}_v{version}.zip"
     if remove_prefix:
@@ -231,10 +236,16 @@ def _download_from_ngc(
 
 
 def _download_from_ngc_private(
-    download_path: Path, filename: str, version: str, remove_prefix: str | None, repo: str, headers: dict | None = None
+    download_path: Path,
+    filename: str,
+    version: str,
+    repo: str,
+    prefix: str = "monai_",
+    remove_prefix: str | None = "monai_",
+    headers: dict | None = None,
 ) -> None:
     # ensure prefix is contained
-    filename = _add_ngc_prefix(filename)
+    filename = _add_ngc_prefix(filename, prefix=prefix)
     request_url = _get_ngc_private_bundle_url(model_name=filename, version=version, repo=repo)
     if has_requests:
         headers = {} if headers is None else headers
@@ -491,7 +502,7 @@ def download(
         url: url to download the data. If not `None`, data will be downloaded directly
             and `source` will not be checked.
             If `name` is `None`, filename is determined by `monai.apps.utils._basename(url)`.
-        remove_prefix: This argument is used when `source` is "ngc". Currently, all ngc bundles
+        remove_prefix: This argument is used when `source` is "ngc" or "ngc_private". Currently, all ngc bundles
             have the ``monai_`` prefix, which is not existing in their model zoo contrasts. In order to
             maintain the consistency between these two sources, remove prefix is necessary.
             Therefore, if specified, downloaded folder name will remove the prefix.

From 56ee32e36c5c0c7a5cb10afa4ec5589c81171e6b Mon Sep 17 00:00:00 2001
From: David Carreto Fidalgo <davidc.fidalgo@gmail.com>
Date: Sat, 3 Aug 2024 22:29:45 +0200
Subject: [PATCH 109/183] Fix: Small logic mistake in the `AsDiscrete.__call__`
 method (#7984)

Hi MONAI Team!
Thank you very much for this super nice framework, really appreciate it!
Just found a small logic mistake in one of the transform classes. To
reproduce:
```python
import torch
from monai.transforms.post.array import AsDiscrete

transform = AsDiscrete(argmax=True)
prediction = torch.rand(2, 3, 3)

transform(prediction, argmax=False)
# will still apply argmax
```

### Description

Proposed fix: `argmax` is explicitly checked for `None` in the `__cal__`
method.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: David Carreto Fidalgo <davidc.fidalgo@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/transforms/post/array.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/monai/transforms/post/array.py b/monai/transforms/post/array.py
index da9b23ce57..2e733c4f6c 100644
--- a/monai/transforms/post/array.py
+++ b/monai/transforms/post/array.py
@@ -211,7 +211,8 @@ def __call__(
             raise ValueError("`to_onehot=True/False` is deprecated, please use `to_onehot=num_classes` instead.")
         img = convert_to_tensor(img, track_meta=get_track_meta())
         img_t, *_ = convert_data_type(img, torch.Tensor)
-        if argmax or self.argmax:
+        argmax = self.argmax if argmax is None else argmax
+        if argmax:
             img_t = torch.argmax(img_t, dim=self.kwargs.get("dim", 0), keepdim=self.kwargs.get("keepdim", True))
 
         to_onehot = self.to_onehot if to_onehot is None else to_onehot

From 6c23fd06fc11667beedd0ba730d4104076a8db2d Mon Sep 17 00:00:00 2001
From: Virginia Fernandez <61539159+virginiafdez@users.noreply.github.com>
Date: Tue, 6 Aug 2024 17:18:22 +0100
Subject: [PATCH 110/183] Flash attention (#7977)

Fixes #7944.

### Description

In response to Issue https://github.com/Project-MONAI/MONAI/issues/7944,
I added the new functionality scaled_dot_product_attention from PyTorch
to re-enable flash attention, present in the original MONAI Generative
Models repository. This is allowed for torch >= 2.0 and when argument
save_attn = False. Errors are raised otherwise. I ran quick tests and
added some checks on test_selfattention and test_crossattention scripts
to make sure the outputs are the same as not using flash attention.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Virginia Fernandez <virginia.fernandez@kcl.ac.uk>
Co-authored-by: Virginia Fernandez <virginia.fernandez@kcl.ac.uk>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
---
 monai/networks/blocks/crossattention.py     | 73 +++++++++++++++------
 monai/networks/blocks/selfattention.py      | 58 ++++++++++++----
 monai/networks/blocks/spatialattention.py   |  8 ++-
 monai/networks/blocks/transformerblock.py   | 13 +++-
 monai/networks/nets/diffusion_model_unet.py |  5 ++
 tests/test_crossattention.py                | 66 +++++++++++++++----
 tests/test_selfattention.py                 | 61 +++++++++++++----
 7 files changed, 223 insertions(+), 61 deletions(-)

diff --git a/monai/networks/blocks/crossattention.py b/monai/networks/blocks/crossattention.py
index b888ea3942..daa5abdd56 100644
--- a/monai/networks/blocks/crossattention.py
+++ b/monai/networks/blocks/crossattention.py
@@ -17,7 +17,7 @@
 import torch.nn as nn
 
 from monai.networks.layers.utils import get_rel_pos_embedding_layer
-from monai.utils import optional_import
+from monai.utils import optional_import, pytorch_after
 
 Rearrange, _ = optional_import("einops.layers.torch", name="Rearrange")
 
@@ -44,6 +44,7 @@ def __init__(
         rel_pos_embedding: Optional[str] = None,
         input_size: Optional[Tuple] = None,
         attention_dtype: Optional[torch.dtype] = None,
+        use_flash_attention: bool = False,
     ) -> None:
         """
         Args:
@@ -55,13 +56,16 @@ def __init__(
             dim_head (int, optional): dimension of each head. Defaults to hidden_size // num_heads.
             qkv_bias (bool, optional): bias term for the qkv linear layer. Defaults to False.
             save_attn (bool, optional): to make accessible the attention matrix. Defaults to False.
-            causal: whether to use causal attention.
-            sequence_length: if causal is True, it is necessary to specify the sequence length.
-            rel_pos_embedding (str, optional): Add relative positional embeddings to the attention map.
-                For now only "decomposed" is supported (see https://arxiv.org/abs/2112.01526). 2D and 3D are supported.
-            input_size (tuple(spatial_dim), optional): Input resolution for calculating the relative
-                positional parameter size.
+            causal (bool, optional): whether to use causal attention.
+            sequence_length (int, optional): if causal is True, it is necessary to specify the sequence length.
+            rel_pos_embedding (str, optional): Add relative positional embeddings to the attention map. For now only
+            "decomposed" is supported (see https://arxiv.org/abs/2112.01526). 2D and 3D are supported.
+            input_size (tuple(spatial_dim), optional): Input resolution for calculating the relative positional
+            parameter size.
             attention_dtype: cast attention operations to this dtype.
+            use_flash_attention: if True, use Pytorch's inbuilt
+            flash attention for a memory efficient attention mechanism (see
+            https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
         """
 
         super().__init__()
@@ -81,6 +85,20 @@ def __init__(
         if causal and sequence_length is None:
             raise ValueError("sequence_length is necessary for causal attention.")
 
+        if use_flash_attention and not pytorch_after(minor=13, major=1, patch=0):
+            raise ValueError(
+                "use_flash_attention is only supported for PyTorch versions >= 2.0."
+                "Upgrade your PyTorch or set the flag to False."
+            )
+        if use_flash_attention and save_attn:
+            raise ValueError(
+                "save_attn has been set to True, but use_flash_attention is also set"
+                "to True. save_attn can only be used if use_flash_attention is False"
+            )
+
+        if use_flash_attention and rel_pos_embedding is not None:
+            raise ValueError("rel_pos_embedding must be None if you are using flash_attention.")
+
         self.num_heads = num_heads
         self.hidden_input_size = hidden_input_size if hidden_input_size else hidden_size
         self.context_input_size = context_input_size if context_input_size else hidden_size
@@ -94,6 +112,7 @@ def __init__(
         self.out_rearrange = Rearrange("b h l d -> b l (h d)")
         self.drop_output = nn.Dropout(dropout_rate)
         self.drop_weights = nn.Dropout(dropout_rate)
+        self.dropout_rate = dropout_rate
 
         self.scale = self.head_dim**-0.5
         self.save_attn = save_attn
@@ -101,6 +120,7 @@ def __init__(
 
         self.causal = causal
         self.sequence_length = sequence_length
+        self.use_flash_attention = use_flash_attention
 
         if causal and sequence_length is not None:
             # causal mask to ensure that attention is only applied to the left in the input sequence
@@ -142,26 +162,39 @@ def forward(self, x: torch.Tensor, context: Optional[torch.Tensor] = None):
             q = q.to(self.attention_dtype)
             k = k.to(self.attention_dtype)
 
-        q = q.view(b, t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, t,  hs)
+        q = q.view(b, t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, t,  hs) #
         k = k.view(b, kv_t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, kv_t, hs)
         v = v.view(b, kv_t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, kv_t, hs)
-        att_mat = torch.einsum("blxd,blyd->blxy", q, k) * self.scale
 
-        # apply relative positional embedding if defined
-        att_mat = self.rel_positional_embedding(x, att_mat, q) if self.rel_positional_embedding is not None else att_mat
+        if self.use_flash_attention:
+            x = torch.nn.functional.scaled_dot_product_attention(
+                query=q.transpose(1, 2),
+                key=k.transpose(1, 2),
+                value=v.transpose(1, 2),
+                scale=self.scale,
+                dropout_p=self.dropout_rate,
+                is_causal=self.causal,
+            ).transpose(
+                1, 2
+            )  # Back to (b, nh, t, hs)
+        else:
+            att_mat = torch.einsum("blxd,blyd->blxy", q, k) * self.scale
+            # apply relative positional embedding if defined
+            if self.rel_positional_embedding is not None:
+                att_mat = self.rel_positional_embedding(x, att_mat, q)
 
-        if self.causal:
-            att_mat = att_mat.masked_fill(self.causal_mask[:, :, :t, :kv_t] == 0, float("-inf"))
+            if self.causal:
+                att_mat = att_mat.masked_fill(self.causal_mask[:, :, :t, :kv_t] == 0, float("-inf"))
 
-        att_mat = att_mat.softmax(dim=-1)
+            att_mat = att_mat.softmax(dim=-1)
 
-        if self.save_attn:
-            # no gradients and new tensor;
-            # https://pytorch.org/docs/stable/generated/torch.Tensor.detach.html
-            self.att_mat = att_mat.detach()
+            if self.save_attn:
+                # no gradients and new tensor;
+                # https://pytorch.org/docs/stable/generated/torch.Tensor.detach.html
+                self.att_mat = att_mat.detach()
 
-        att_mat = self.drop_weights(att_mat)
-        x = torch.einsum("bhxy,bhyd->bhxd", att_mat, v)
+            att_mat = self.drop_weights(att_mat)
+            x = torch.einsum("bhxy,bhyd->bhxd", att_mat, v)
         x = self.out_rearrange(x)
         x = self.out_proj(x)
         x = self.drop_output(x)
diff --git a/monai/networks/blocks/selfattention.py b/monai/networks/blocks/selfattention.py
index 3ab1e1fd10..124c00acc6 100644
--- a/monai/networks/blocks/selfattention.py
+++ b/monai/networks/blocks/selfattention.py
@@ -15,9 +15,10 @@
 
 import torch
 import torch.nn as nn
+import torch.nn.functional as F
 
 from monai.networks.layers.utils import get_rel_pos_embedding_layer
-from monai.utils import optional_import
+from monai.utils import optional_import, pytorch_after
 
 Rearrange, _ = optional_import("einops.layers.torch", name="Rearrange")
 
@@ -42,6 +43,7 @@ def __init__(
         rel_pos_embedding: Optional[str] = None,
         input_size: Optional[Tuple] = None,
         attention_dtype: Optional[torch.dtype] = None,
+        use_flash_attention: bool = False,
     ) -> None:
         """
         Args:
@@ -59,6 +61,9 @@ def __init__(
             input_size (tuple(spatial_dim), optional): Input resolution for calculating the relative
                 positional parameter size.
             attention_dtype: cast attention operations to this dtype.
+            use_flash_attention: if True, use Pytorch's inbuilt
+            flash attention for a memory efficient attention mechanism (see
+            https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
 
         """
 
@@ -82,6 +87,20 @@ def __init__(
         if causal and sequence_length is None:
             raise ValueError("sequence_length is necessary for causal attention.")
 
+        if use_flash_attention and not pytorch_after(minor=13, major=1, patch=0):
+            raise ValueError(
+                "use_flash_attention is only supported for PyTorch versions >= 2.0."
+                "Upgrade your PyTorch or set the flag to False."
+            )
+        if use_flash_attention and save_attn:
+            raise ValueError(
+                "save_attn has been set to True, but use_flash_attention is also set"
+                "to True. save_attn can only be used if use_flash_attention is False."
+            )
+
+        if use_flash_attention and rel_pos_embedding is not None:
+            raise ValueError("rel_pos_embedding must be None if you are using flash_attention.")
+
         self.num_heads = num_heads
         self.hidden_input_size = hidden_input_size if hidden_input_size else hidden_size
         self.out_proj = nn.Linear(self.inner_dim, self.hidden_input_size)
@@ -91,12 +110,14 @@ def __init__(
         self.out_rearrange = Rearrange("b h l d -> b l (h d)")
         self.drop_output = nn.Dropout(dropout_rate)
         self.drop_weights = nn.Dropout(dropout_rate)
+        self.dropout_rate = dropout_rate
         self.scale = self.dim_head**-0.5
         self.save_attn = save_attn
         self.att_mat = torch.Tensor()
         self.attention_dtype = attention_dtype
         self.causal = causal
         self.sequence_length = sequence_length
+        self.use_flash_attention = use_flash_attention
 
         if causal and sequence_length is not None:
             # causal mask to ensure that attention is only applied to the left in the input sequence
@@ -130,23 +151,34 @@ def forward(self, x):
             q = q.to(self.attention_dtype)
             k = k.to(self.attention_dtype)
 
-        att_mat = torch.einsum("blxd,blyd->blxy", q, k) * self.scale
+        if self.use_flash_attention:
+            x = F.scaled_dot_product_attention(
+                query=q.transpose(1, 2),
+                key=k.transpose(1, 2),
+                value=v.transpose(1, 2),
+                scale=self.scale,
+                dropout_p=self.dropout_rate,
+                is_causal=self.causal,
+            ).transpose(1, 2)
+        else:
+            att_mat = torch.einsum("blxd,blyd->blxy", q, k) * self.scale
 
-        # apply relative positional embedding if defined
-        att_mat = self.rel_positional_embedding(x, att_mat, q) if self.rel_positional_embedding is not None else att_mat
+            # apply relative positional embedding if defined
+            if self.rel_positional_embedding is not None:
+                att_mat = self.rel_positional_embedding(x, att_mat, q)
 
-        if self.causal:
-            att_mat = att_mat.masked_fill(self.causal_mask[:, :, : x.shape[1], : x.shape[1]] == 0, float("-inf"))
+            if self.causal:
+                att_mat = att_mat.masked_fill(self.causal_mask[:, :, : x.shape[-2], : x.shape[-2]] == 0, float("-inf"))
 
-        att_mat = att_mat.softmax(dim=-1)
+            att_mat = att_mat.softmax(dim=-1)
 
-        if self.save_attn:
-            # no gradients and new tensor;
-            # https://pytorch.org/docs/stable/generated/torch.Tensor.detach.html
-            self.att_mat = att_mat.detach()
+            if self.save_attn:
+                # no gradients and new tensor;
+                # https://pytorch.org/docs/stable/generated/torch.Tensor.detach.html
+                self.att_mat = att_mat.detach()
 
-        att_mat = self.drop_weights(att_mat)
-        x = torch.einsum("bhxy,bhyd->bhxd", att_mat, v)
+            att_mat = self.drop_weights(att_mat)
+            x = torch.einsum("bhxy,bhyd->bhxd", att_mat, v)
         x = self.out_rearrange(x)
         x = self.out_proj(x)
         x = self.drop_output(x)
diff --git a/monai/networks/blocks/spatialattention.py b/monai/networks/blocks/spatialattention.py
index 75319853d9..1cfafb1585 100644
--- a/monai/networks/blocks/spatialattention.py
+++ b/monai/networks/blocks/spatialattention.py
@@ -33,6 +33,7 @@ class SpatialAttentionBlock(nn.Module):
         num_channels: number of input channels. Must be divisible by num_head_channels.
         num_head_channels: number of channels per head.
         attention_dtype: cast attention operations to this dtype.
+        use_flash_attention: if True, use flash attention for a memory efficient attention mechanism.
 
     """
 
@@ -44,6 +45,7 @@ def __init__(
         norm_num_groups: int = 32,
         norm_eps: float = 1e-6,
         attention_dtype: Optional[torch.dtype] = None,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
 
@@ -54,7 +56,11 @@ def __init__(
             raise ValueError("num_channels must be divisible by num_head_channels")
         num_heads = num_channels // num_head_channels if num_head_channels is not None else 1
         self.attn = SABlock(
-            hidden_size=num_channels, num_heads=num_heads, qkv_bias=True, attention_dtype=attention_dtype
+            hidden_size=num_channels,
+            num_heads=num_heads,
+            qkv_bias=True,
+            attention_dtype=attention_dtype,
+            use_flash_attention=use_flash_attention,
         )
 
     def forward(self, x: torch.Tensor):
diff --git a/monai/networks/blocks/transformerblock.py b/monai/networks/blocks/transformerblock.py
index 0aa1697479..28d9c563ac 100644
--- a/monai/networks/blocks/transformerblock.py
+++ b/monai/networks/blocks/transformerblock.py
@@ -36,6 +36,7 @@ def __init__(
         causal: bool = False,
         sequence_length: int | None = None,
         with_cross_attention: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         """
         Args:
@@ -43,8 +44,10 @@ def __init__(
             mlp_dim (int): dimension of feedforward layer.
             num_heads (int): number of attention heads.
             dropout_rate (float, optional): fraction of the input units to drop. Defaults to 0.0.
-            qkv_bias (bool, optional): apply bias term for the qkv linear layer. Defaults to False.
+            qkv_bias(bool, optional): apply bias term for the qkv linear layer. Defaults to False.
             save_attn (bool, optional): to make accessible the attention matrix. Defaults to False.
+            use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+             (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
 
         """
 
@@ -66,13 +69,19 @@ def __init__(
             save_attn=save_attn,
             causal=causal,
             sequence_length=sequence_length,
+            use_flash_attention=use_flash_attention,
         )
         self.norm2 = nn.LayerNorm(hidden_size)
         self.with_cross_attention = with_cross_attention
 
         self.norm_cross_attn = nn.LayerNorm(hidden_size)
         self.cross_attn = CrossAttentionBlock(
-            hidden_size=hidden_size, num_heads=num_heads, dropout_rate=dropout_rate, qkv_bias=qkv_bias, causal=False
+            hidden_size=hidden_size,
+            num_heads=num_heads,
+            dropout_rate=dropout_rate,
+            qkv_bias=qkv_bias,
+            causal=False,
+            use_flash_attention=use_flash_attention,
         )
 
     def forward(self, x: torch.Tensor, context: Optional[torch.Tensor] = None) -> torch.Tensor:
diff --git a/monai/networks/nets/diffusion_model_unet.py b/monai/networks/nets/diffusion_model_unet.py
index 8a9ac859a3..a885339d0d 100644
--- a/monai/networks/nets/diffusion_model_unet.py
+++ b/monai/networks/nets/diffusion_model_unet.py
@@ -66,6 +66,8 @@ class DiffusionUNetTransformerBlock(nn.Module):
         dropout: dropout probability to use.
         cross_attention_dim: size of the context vector for cross attention.
         upcast_attention: if True, upcast attention operations to full precision.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+        (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
 
     """
 
@@ -77,6 +79,7 @@ def __init__(
         dropout: float = 0.0,
         cross_attention_dim: int | None = None,
         upcast_attention: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.attn1 = SABlock(
@@ -86,6 +89,7 @@ def __init__(
             dim_head=num_head_channels,
             dropout_rate=dropout,
             attention_dtype=torch.float if upcast_attention else None,
+            use_flash_attention=use_flash_attention,
         )
         self.ff = MLPBlock(hidden_size=num_channels, mlp_dim=num_channels * 4, act="GEGLU", dropout_rate=dropout)
         self.attn2 = CrossAttentionBlock(
@@ -96,6 +100,7 @@ def __init__(
             dim_head=num_head_channels,
             dropout_rate=dropout,
             attention_dtype=torch.float if upcast_attention else None,
+            use_flash_attention=use_flash_attention,
         )
         self.norm1 = nn.LayerNorm(num_channels)
         self.norm2 = nn.LayerNorm(num_channels)
diff --git a/tests/test_crossattention.py b/tests/test_crossattention.py
index 4ab0ab1823..44458147d6 100644
--- a/tests/test_crossattention.py
+++ b/tests/test_crossattention.py
@@ -22,6 +22,7 @@
 from monai.networks.blocks.crossattention import CrossAttentionBlock
 from monai.networks.layers.factories import RelPosEmbedding
 from monai.utils import optional_import
+from tests.utils import SkipIfBeforePyTorchVersion
 
 einops, has_einops = optional_import("einops")
 
@@ -31,25 +32,29 @@
         for num_heads in [4, 6, 8, 12]:
             for rel_pos_embedding in [None, RelPosEmbedding.DECOMPOSED]:
                 for input_size in [(16, 32), (8, 8, 8)]:
-                    test_case = [
-                        {
-                            "hidden_size": hidden_size,
-                            "num_heads": num_heads,
-                            "dropout_rate": dropout_rate,
-                            "rel_pos_embedding": rel_pos_embedding,
-                            "input_size": input_size,
-                        },
-                        (2, 512, hidden_size),
-                        (2, 512, hidden_size),
-                    ]
-                    TEST_CASE_CABLOCK.append(test_case)
+                    for flash_attn in [True, False]:
+                        test_case = [
+                            {
+                                "hidden_size": hidden_size,
+                                "num_heads": num_heads,
+                                "dropout_rate": dropout_rate,
+                                "rel_pos_embedding": rel_pos_embedding if not flash_attn else None,
+                                "input_size": input_size,
+                                "use_flash_attention": flash_attn,
+                            },
+                            (2, 512, hidden_size),
+                            (2, 512, hidden_size),
+                        ]
+                        TEST_CASE_CABLOCK.append(test_case)
 
 
 class TestResBlock(unittest.TestCase):
 
     @parameterized.expand(TEST_CASE_CABLOCK)
     @skipUnless(has_einops, "Requires einops")
+    @SkipIfBeforePyTorchVersion((2, 0))
     def test_shape(self, input_param, input_shape, expected_shape):
+        # Without flash attention
         net = CrossAttentionBlock(**input_param)
         with eval_mode(net):
             result = net(torch.randn(input_shape), context=torch.randn(2, 512, input_param["hidden_size"]))
@@ -62,6 +67,25 @@ def test_ill_arg(self):
         with self.assertRaises(ValueError):
             CrossAttentionBlock(hidden_size=620, num_heads=8, dropout_rate=0.4)
 
+    @SkipIfBeforePyTorchVersion((2, 0))
+    def test_save_attn_with_flash_attention(self):
+        with self.assertRaises(ValueError):
+            CrossAttentionBlock(
+                hidden_size=128, num_heads=3, dropout_rate=0.1, use_flash_attention=True, save_attn=True
+            )
+
+    @SkipIfBeforePyTorchVersion((2, 0))
+    def test_rel_pos_embedding_with_flash_attention(self):
+        with self.assertRaises(ValueError):
+            CrossAttentionBlock(
+                hidden_size=128,
+                num_heads=3,
+                dropout_rate=0.1,
+                use_flash_attention=True,
+                save_attn=False,
+                rel_pos_embedding=RelPosEmbedding.DECOMPOSED,
+            )
+
     @skipUnless(has_einops, "Requires einops")
     def test_attention_dim_not_multiple_of_heads(self):
         with self.assertRaises(ValueError):
@@ -75,6 +99,22 @@ def test_causal_no_sequence_length(self):
         with self.assertRaises(ValueError):
             CrossAttentionBlock(hidden_size=128, num_heads=4, dropout_rate=0.1, causal=True)
 
+    @skipUnless(has_einops, "Requires einops")
+    @SkipIfBeforePyTorchVersion((2, 0))
+    def test_causal_flash_attention(self):
+        block = CrossAttentionBlock(
+            hidden_size=128,
+            num_heads=1,
+            dropout_rate=0.1,
+            causal=True,
+            sequence_length=16,
+            save_attn=False,
+            use_flash_attention=True,
+        )
+        input_shape = (1, 16, 128)
+        # Check it runs correctly
+        block(torch.randn(input_shape))
+
     @skipUnless(has_einops, "Requires einops")
     def test_causal(self):
         block = CrossAttentionBlock(
@@ -119,7 +159,7 @@ def test_access_attn_matrix(self):
         # no of elements is zero
         assert no_matrix_acess_blk.att_mat.nelement() == 0
 
-        # be able to acess the attention matrix
+        # be able to acess the attention matrix.
         matrix_acess_blk = CrossAttentionBlock(
             hidden_size=hidden_size, num_heads=num_heads, dropout_rate=dropout_rate, save_attn=True
         )
diff --git a/tests/test_selfattention.py b/tests/test_selfattention.py
index d069d6aa30..3e98f4c5c4 100644
--- a/tests/test_selfattention.py
+++ b/tests/test_selfattention.py
@@ -22,6 +22,7 @@
 from monai.networks.blocks.selfattention import SABlock
 from monai.networks.layers.factories import RelPosEmbedding
 from monai.utils import optional_import
+from tests.utils import SkipIfBeforePyTorchVersion
 
 einops, has_einops = optional_import("einops")
 
@@ -31,24 +32,27 @@
         for num_heads in [4, 6, 8, 12]:
             for rel_pos_embedding in [None, RelPosEmbedding.DECOMPOSED]:
                 for input_size in [(16, 32), (8, 8, 8)]:
-                    test_case = [
-                        {
-                            "hidden_size": hidden_size,
-                            "num_heads": num_heads,
-                            "dropout_rate": dropout_rate,
-                            "rel_pos_embedding": rel_pos_embedding,
-                            "input_size": input_size,
-                        },
-                        (2, 512, hidden_size),
-                        (2, 512, hidden_size),
-                    ]
-                    TEST_CASE_SABLOCK.append(test_case)
+                    for flash_attn in [True, False]:
+                        test_case = [
+                            {
+                                "hidden_size": hidden_size,
+                                "num_heads": num_heads,
+                                "dropout_rate": dropout_rate,
+                                "rel_pos_embedding": rel_pos_embedding if not flash_attn else None,
+                                "input_size": input_size,
+                                "use_flash_attention": flash_attn,
+                            },
+                            (2, 512, hidden_size),
+                            (2, 512, hidden_size),
+                        ]
+                        TEST_CASE_SABLOCK.append(test_case)
 
 
 class TestResBlock(unittest.TestCase):
 
     @parameterized.expand(TEST_CASE_SABLOCK)
     @skipUnless(has_einops, "Requires einops")
+    @SkipIfBeforePyTorchVersion((2, 0))
     def test_shape(self, input_param, input_shape, expected_shape):
         net = SABlock(**input_param)
         with eval_mode(net):
@@ -62,6 +66,23 @@ def test_ill_arg(self):
         with self.assertRaises(ValueError):
             SABlock(hidden_size=620, num_heads=8, dropout_rate=0.4)
 
+    @SkipIfBeforePyTorchVersion((2, 0))
+    def test_rel_pos_embedding_with_flash_attention(self):
+        with self.assertRaises(ValueError):
+            SABlock(
+                hidden_size=128,
+                num_heads=3,
+                dropout_rate=0.1,
+                use_flash_attention=True,
+                save_attn=False,
+                rel_pos_embedding=RelPosEmbedding.DECOMPOSED,
+            )
+
+    @SkipIfBeforePyTorchVersion((1, 13))
+    def test_save_attn_with_flash_attention(self):
+        with self.assertRaises(ValueError):
+            SABlock(hidden_size=128, num_heads=3, dropout_rate=0.1, use_flash_attention=True, save_attn=True)
+
     def test_attention_dim_not_multiple_of_heads(self):
         with self.assertRaises(ValueError):
             SABlock(hidden_size=128, num_heads=3, dropout_rate=0.1)
@@ -74,6 +95,22 @@ def test_causal_no_sequence_length(self):
         with self.assertRaises(ValueError):
             SABlock(hidden_size=128, num_heads=4, dropout_rate=0.1, causal=True)
 
+    @skipUnless(has_einops, "Requires einops")
+    @SkipIfBeforePyTorchVersion((2, 0))
+    def test_causal_flash_attention(self):
+        block = SABlock(
+            hidden_size=128,
+            num_heads=1,
+            dropout_rate=0.1,
+            causal=True,
+            sequence_length=16,
+            save_attn=False,
+            use_flash_attention=True,
+        )
+        input_shape = (1, 16, 128)
+        # Check it runs correctly
+        block(torch.randn(input_shape))
+
     @skipUnless(has_einops, "Requires einops")
     def test_causal(self):
         block = SABlock(hidden_size=128, num_heads=1, dropout_rate=0.1, causal=True, sequence_length=16, save_attn=True)

From 49a1e343129909f7e01c67882a71c12beed3ef30 Mon Sep 17 00:00:00 2001
From: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Date: Wed, 7 Aug 2024 16:28:31 +0800
Subject: [PATCH 111/183] 7994-enhance-mlpblock (#7995)

Fixes #7994  .

### Description
The current implementation does not support tuple input of "GEGLU" since
it only change the out features of the first linear layer when the input
is a string of "GEGLU".

This PR enhances it, and also enable "vista3d" mode to support #7987
Tests are added to cover the changes.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/pythonapp.yml |  1 +
 monai/networks/blocks/mlp.py    | 20 ++++++++++++++++----
 tests/test_mlp.py               | 28 ++++++++++++++++++++++++++++
 3 files changed, 45 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/pythonapp.yml b/.github/workflows/pythonapp.yml
index fe04f96a80..65f9a4dcf2 100644
--- a/.github/workflows/pythonapp.yml
+++ b/.github/workflows/pythonapp.yml
@@ -99,6 +99,7 @@ jobs:
       name: Install itk pre-release (Linux only)
       run: |
         python -m pip install --pre -U itk
+        find  /opt/hostedtoolcache/* -maxdepth 0 ! -name 'Python' -exec rm -rf {} \;
     - name: Install the dependencies
       run: |
         python -m pip install --user --upgrade pip wheel
diff --git a/monai/networks/blocks/mlp.py b/monai/networks/blocks/mlp.py
index d3510b64d3..8771711d25 100644
--- a/monai/networks/blocks/mlp.py
+++ b/monai/networks/blocks/mlp.py
@@ -11,12 +11,15 @@
 
 from __future__ import annotations
 
+from typing import Union
+
 import torch.nn as nn
 
 from monai.networks.layers import get_act_layer
+from monai.networks.layers.factories import split_args
 from monai.utils import look_up_option
 
-SUPPORTED_DROPOUT_MODE = {"vit", "swin"}
+SUPPORTED_DROPOUT_MODE = {"vit", "swin", "vista3d"}
 
 
 class MLPBlock(nn.Module):
@@ -39,7 +42,7 @@ def __init__(
                 https://github.com/google-research/vision_transformer/blob/main/vit_jax/models.py#L87
                 "swin" corresponds to one instance as implemented in
                 https://github.com/microsoft/Swin-Transformer/blob/main/models/swin_mlp.py#L23
-
+                "vista3d" mode does not use dropout.
 
         """
 
@@ -48,15 +51,24 @@ def __init__(
         if not (0 <= dropout_rate <= 1):
             raise ValueError("dropout_rate should be between 0 and 1.")
         mlp_dim = mlp_dim or hidden_size
-        self.linear1 = nn.Linear(hidden_size, mlp_dim) if act != "GEGLU" else nn.Linear(hidden_size, mlp_dim * 2)
+        act_name, _ = split_args(act)
+        self.linear1 = nn.Linear(hidden_size, mlp_dim) if act_name != "GEGLU" else nn.Linear(hidden_size, mlp_dim * 2)
         self.linear2 = nn.Linear(mlp_dim, hidden_size)
         self.fn = get_act_layer(act)
-        self.drop1 = nn.Dropout(dropout_rate)
+        # Use Union[nn.Dropout, nn.Identity] for type annotations
+        self.drop1: Union[nn.Dropout, nn.Identity]
+        self.drop2: Union[nn.Dropout, nn.Identity]
+
         dropout_opt = look_up_option(dropout_mode, SUPPORTED_DROPOUT_MODE)
         if dropout_opt == "vit":
+            self.drop1 = nn.Dropout(dropout_rate)
             self.drop2 = nn.Dropout(dropout_rate)
         elif dropout_opt == "swin":
+            self.drop1 = nn.Dropout(dropout_rate)
             self.drop2 = self.drop1
+        elif dropout_opt == "vista3d":
+            self.drop1 = nn.Identity()
+            self.drop2 = nn.Identity()
         else:
             raise ValueError(f"dropout_mode should be one of {SUPPORTED_DROPOUT_MODE}")
 
diff --git a/tests/test_mlp.py b/tests/test_mlp.py
index 54f70d3318..2598d8877d 100644
--- a/tests/test_mlp.py
+++ b/tests/test_mlp.py
@@ -15,10 +15,12 @@
 
 import numpy as np
 import torch
+import torch.nn as nn
 from parameterized import parameterized
 
 from monai.networks import eval_mode
 from monai.networks.blocks.mlp import MLPBlock
+from monai.networks.layers.factories import split_args
 
 TEST_CASE_MLP = []
 for dropout_rate in np.linspace(0, 1, 4):
@@ -31,6 +33,14 @@
             ]
             TEST_CASE_MLP.append(test_case)
 
+# test different activation layers
+TEST_CASE_ACT = []
+for act in ["GELU", "GEGLU", ("GEGLU", {})]:  # type: ignore
+    TEST_CASE_ACT.append([{"hidden_size": 128, "mlp_dim": 0, "act": act}, (2, 512, 128), (2, 512, 128)])
+
+# test different dropout modes
+TEST_CASE_DROP = [["vit", nn.Dropout], ["swin", nn.Dropout], ["vista3d", nn.Identity]]
+
 
 class TestMLPBlock(unittest.TestCase):
 
@@ -45,6 +55,24 @@ def test_ill_arg(self):
         with self.assertRaises(ValueError):
             MLPBlock(hidden_size=128, mlp_dim=512, dropout_rate=5.0)
 
+    @parameterized.expand(TEST_CASE_ACT)
+    def test_act(self, input_param, input_shape, expected_shape):
+        net = MLPBlock(**input_param)
+        with eval_mode(net):
+            result = net(torch.randn(input_shape))
+            self.assertEqual(result.shape, expected_shape)
+        act_name, _ = split_args(input_param["act"])
+        if act_name == "GEGLU":
+            self.assertEqual(net.linear1.in_features, net.linear1.out_features // 2)
+        else:
+            self.assertEqual(net.linear1.in_features, net.linear1.out_features)
+
+    @parameterized.expand(TEST_CASE_DROP)
+    def test_dropout_mode(self, dropout_mode, dropout_layer):
+        net = MLPBlock(hidden_size=128, mlp_dim=512, dropout_rate=0.1, dropout_mode=dropout_mode)
+        self.assertTrue(isinstance(net.drop1, dropout_layer))
+        self.assertTrue(isinstance(net.drop2, dropout_layer))
+
 
 if __name__ == "__main__":
     unittest.main()

From 660891f37665874a38b289fdb8cd339800aff9cc Mon Sep 17 00:00:00 2001
From: Balamurali <balamuralim.1993@gmail.com>
Date: Thu, 8 Aug 2024 00:58:12 -0700
Subject: [PATCH 112/183] Initial commit -- Adding calibration loss specific to
 segmentation (#7819)

### Description

Model calibration has helped in developing reliable deep learning
models. In this pull request, I have added a new loss function NACL
(https://arxiv.org/abs/2303.06268, https://arxiv.org/abs/2401.14487)
which has shown promising results for both discriminative and
calibration in segmentation.

**Future Plans:** Currently, MONAI has some of the alternative loss
functions (Label Smoothing, and Focal Loss), but it doesn't have the
calibration specific loss functions (https://arxiv.org/abs/2111.15430,
https://arxiv.org/abs/2209.09641). Besides, these methods are better
evaluated with calibration metrics, Expected Calibration Error
(https://lightning.ai/docs/torchmetrics/stable/classification/calibration_error.html).

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Balamurali <balamuralim.1993@gmail.com>
Signed-off-by: bala93 <balamuralim.1993@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/losses.rst    |   5 ++
 monai/losses/__init__.py  |   1 +
 monai/losses/nacl_loss.py | 139 +++++++++++++++++++++++++++++++
 tests/test_nacl_loss.py   | 166 ++++++++++++++++++++++++++++++++++++++
 4 files changed, 311 insertions(+)
 create mode 100644 monai/losses/nacl_loss.py
 create mode 100644 tests/test_nacl_loss.py

diff --git a/docs/source/losses.rst b/docs/source/losses.rst
index ba794af3eb..528ccd1173 100644
--- a/docs/source/losses.rst
+++ b/docs/source/losses.rst
@@ -93,6 +93,11 @@ Segmentation Losses
 .. autoclass:: SoftDiceclDiceLoss
     :members:
 
+`NACLLoss`
+~~~~~~~~~~
+.. autoclass:: NACLLoss
+    :members:
+
 Registration Losses
 -------------------
 
diff --git a/monai/losses/__init__.py b/monai/losses/__init__.py
index e937b53fa4..41935be204 100644
--- a/monai/losses/__init__.py
+++ b/monai/losses/__init__.py
@@ -37,6 +37,7 @@
 from .hausdorff_loss import HausdorffDTLoss, LogHausdorffDTLoss
 from .image_dissimilarity import GlobalMutualInformationLoss, LocalNormalizedCrossCorrelationLoss
 from .multi_scale import MultiScaleLoss
+from .nacl_loss import NACLLoss
 from .perceptual import PerceptualLoss
 from .spatial_mask import MaskedLoss
 from .spectral_loss import JukeboxLoss
diff --git a/monai/losses/nacl_loss.py b/monai/losses/nacl_loss.py
new file mode 100644
index 0000000000..3303e89bce
--- /dev/null
+++ b/monai/losses/nacl_loss.py
@@ -0,0 +1,139 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import Any
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from torch.nn.modules.loss import _Loss
+
+from monai.networks.layers import GaussianFilter, MeanFilter
+
+
+class NACLLoss(_Loss):
+    """
+    Neighbor-Aware Calibration Loss (NACL) is primarily developed for developing calibrated models in image segmentation.
+    NACL computes standard cross-entropy loss with a linear penalty that enforces the logit distributions
+    to match a soft class proportion of surrounding pixel.
+
+    Murugesan, Balamurali, et al.
+    "Trust your neighbours: Penalty-based constraints for model calibration."
+    International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2023.
+    https://arxiv.org/abs/2303.06268
+
+    Murugesan, Balamurali, et al.
+    "Neighbor-Aware Calibration of Segmentation Networks with Penalty-Based Constraints."
+    https://arxiv.org/abs/2401.14487
+    """
+
+    def __init__(
+        self,
+        classes: int,
+        dim: int,
+        kernel_size: int = 3,
+        kernel_ops: str = "mean",
+        distance_type: str = "l1",
+        alpha: float = 0.1,
+        sigma: float = 1.0,
+    ) -> None:
+        """
+        Args:
+            classes: number of classes
+            dim: dimension of data (supports 2d and 3d)
+            kernel_size: size of the spatial kernel
+            distance_type: l1/l2 distance between spatial kernel and predicted logits
+            alpha: weightage between cross entropy and logit constraint
+            sigma: sigma of gaussian
+        """
+
+        super().__init__()
+
+        if kernel_ops not in ["mean", "gaussian"]:
+            raise ValueError("Kernel ops must be either mean or gaussian")
+
+        if dim not in [2, 3]:
+            raise ValueError(f"Support 2d and 3d, got dim={dim}.")
+
+        if distance_type not in ["l1", "l2"]:
+            raise ValueError(f"Distance type must be either L1 or L2, got {distance_type}")
+
+        self.nc = classes
+        self.dim = dim
+        self.cross_entropy = nn.CrossEntropyLoss()
+        self.distance_type = distance_type
+        self.alpha = alpha
+        self.ks = kernel_size
+        self.svls_layer: Any
+
+        if kernel_ops == "mean":
+            self.svls_layer = MeanFilter(spatial_dims=dim, size=kernel_size)
+            self.svls_layer.filter = self.svls_layer.filter / (kernel_size**dim)
+        if kernel_ops == "gaussian":
+            self.svls_layer = GaussianFilter(spatial_dims=dim, sigma=sigma)
+
+    def get_constr_target(self, mask: torch.Tensor) -> torch.Tensor:
+        """
+        Converts the mask to one hot represenation and is smoothened with the selected spatial filter.
+
+        Args:
+            mask: the shape should be BH[WD].
+
+        Returns:
+            torch.Tensor: the shape would be BNH[WD], N being number of classes.
+        """
+        rmask: torch.Tensor
+
+        if self.dim == 2:
+            oh_labels = F.one_hot(mask.to(torch.int64), num_classes=self.nc).contiguous().permute(0, 3, 1, 2).float()
+            rmask = self.svls_layer(oh_labels)
+
+        if self.dim == 3:
+            oh_labels = F.one_hot(mask.to(torch.int64), num_classes=self.nc).contiguous().permute(0, 4, 1, 2, 3).float()
+            rmask = self.svls_layer(oh_labels)
+
+        return rmask
+
+    def forward(self, inputs: torch.Tensor, targets: torch.Tensor) -> torch.Tensor:
+        """
+        Computes standard cross-entropy loss and constraints it neighbor aware logit penalty.
+
+        Args:
+            inputs: the shape should be BNH[WD], where N is the number of classes.
+            targets: the shape should be BH[WD].
+
+        Returns:
+            torch.Tensor: value of the loss.
+
+        Example:
+            >>> import torch
+            >>> from monai.losses import NACLLoss
+            >>> B, N, H, W = 8, 3, 64, 64
+            >>> input = torch.rand(B, N, H, W)
+            >>> target = torch.randint(0, N, (B, H, W))
+            >>> criterion = NACLLoss(classes = N, dim = 2)
+            >>> loss = criterion(input, target)
+        """
+
+        loss_ce = self.cross_entropy(inputs, targets)
+
+        utargets = self.get_constr_target(targets)
+
+        if self.distance_type == "l1":
+            loss_conf = utargets.sub(inputs).abs_().mean()
+        elif self.distance_type == "l2":
+            loss_conf = utargets.sub(inputs).pow_(2).abs_().mean()
+
+        loss: torch.Tensor = loss_ce + self.alpha * loss_conf
+
+        return loss
diff --git a/tests/test_nacl_loss.py b/tests/test_nacl_loss.py
new file mode 100644
index 0000000000..51ec275cf4
--- /dev/null
+++ b/tests/test_nacl_loss.py
@@ -0,0 +1,166 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import numpy as np
+import torch
+from parameterized import parameterized
+
+from monai.losses import NACLLoss
+
+inputs = torch.tensor(
+    [
+        [
+            [
+                [0.1498, 0.1158, 0.3996, 0.3730],
+                [0.2155, 0.1585, 0.8541, 0.8579],
+                [0.6640, 0.2424, 0.0774, 0.0324],
+                [0.0580, 0.2180, 0.3447, 0.8722],
+            ],
+            [
+                [0.3908, 0.9366, 0.1779, 0.1003],
+                [0.9630, 0.6118, 0.4405, 0.7916],
+                [0.5782, 0.9515, 0.4088, 0.3946],
+                [0.7860, 0.3910, 0.0324, 0.9568],
+            ],
+            [
+                [0.0759, 0.0238, 0.5570, 0.1691],
+                [0.2703, 0.7722, 0.1611, 0.6431],
+                [0.8051, 0.6596, 0.4121, 0.1125],
+                [0.5283, 0.6746, 0.5528, 0.7913],
+            ],
+        ]
+    ]
+)
+targets = torch.tensor([[[1, 1, 1, 1], [1, 1, 1, 0], [0, 0, 1, 0], [0, 1, 0, 0]]])
+
+TEST_CASES = [
+    [{"classes": 3, "dim": 2}, {"inputs": inputs, "targets": targets}, 1.1442],
+    [{"classes": 3, "dim": 2, "kernel_ops": "gaussian"}, {"inputs": inputs, "targets": targets}, 1.1433],
+    [{"classes": 3, "dim": 2, "kernel_ops": "gaussian", "sigma": 0.5}, {"inputs": inputs, "targets": targets}, 1.1469],
+    [{"classes": 3, "dim": 2, "distance_type": "l2"}, {"inputs": inputs, "targets": targets}, 1.1269],
+    [{"classes": 3, "dim": 2, "alpha": 0.2}, {"inputs": inputs, "targets": targets}, 1.1790],
+    [
+        {"classes": 3, "dim": 3, "kernel_ops": "gaussian"},
+        {
+            "inputs": torch.tensor(
+                [
+                    [
+                        [
+                            [
+                                [0.5977, 0.2767, 0.0591, 0.1675],
+                                [0.4835, 0.3778, 0.8406, 0.3065],
+                                [0.6047, 0.2860, 0.9742, 0.2013],
+                                [0.9128, 0.8368, 0.6711, 0.4384],
+                            ],
+                            [
+                                [0.9797, 0.1863, 0.5584, 0.6652],
+                                [0.2272, 0.2004, 0.7914, 0.4224],
+                                [0.5097, 0.8818, 0.2581, 0.3495],
+                                [0.1054, 0.5483, 0.3732, 0.3587],
+                            ],
+                            [
+                                [0.3060, 0.7066, 0.7922, 0.4689],
+                                [0.1733, 0.8902, 0.6704, 0.2037],
+                                [0.8656, 0.5561, 0.2701, 0.0092],
+                                [0.1866, 0.7714, 0.6424, 0.9791],
+                            ],
+                            [
+                                [0.5067, 0.3829, 0.6156, 0.8985],
+                                [0.5192, 0.8347, 0.2098, 0.2260],
+                                [0.8887, 0.3944, 0.6400, 0.5345],
+                                [0.1207, 0.3763, 0.5282, 0.7741],
+                            ],
+                        ],
+                        [
+                            [
+                                [0.8499, 0.4759, 0.1964, 0.5701],
+                                [0.3190, 0.1238, 0.2368, 0.9517],
+                                [0.0797, 0.6185, 0.0135, 0.8672],
+                                [0.4116, 0.1683, 0.1355, 0.0545],
+                            ],
+                            [
+                                [0.7533, 0.2658, 0.5955, 0.4498],
+                                [0.9500, 0.2317, 0.2825, 0.9763],
+                                [0.1493, 0.1558, 0.3743, 0.8723],
+                                [0.1723, 0.7980, 0.8816, 0.0133],
+                            ],
+                            [
+                                [0.8426, 0.2666, 0.2077, 0.3161],
+                                [0.1725, 0.8414, 0.1515, 0.2825],
+                                [0.4882, 0.5159, 0.4120, 0.1585],
+                                [0.2551, 0.9073, 0.7691, 0.9898],
+                            ],
+                            [
+                                [0.4633, 0.8717, 0.8537, 0.2899],
+                                [0.3693, 0.7953, 0.1183, 0.4596],
+                                [0.0087, 0.7925, 0.0989, 0.8385],
+                                [0.8261, 0.6920, 0.7069, 0.4464],
+                            ],
+                        ],
+                        [
+                            [
+                                [0.0110, 0.1608, 0.4814, 0.6317],
+                                [0.0194, 0.9669, 0.3259, 0.0028],
+                                [0.5674, 0.8286, 0.0306, 0.5309],
+                                [0.3973, 0.8183, 0.0238, 0.1934],
+                            ],
+                            [
+                                [0.8947, 0.6629, 0.9439, 0.8905],
+                                [0.0072, 0.1697, 0.4634, 0.0201],
+                                [0.7184, 0.2424, 0.0820, 0.7504],
+                                [0.3937, 0.1424, 0.4463, 0.5779],
+                            ],
+                            [
+                                [0.4123, 0.6227, 0.0523, 0.8826],
+                                [0.0051, 0.0353, 0.3662, 0.7697],
+                                [0.4867, 0.8986, 0.2510, 0.5316],
+                                [0.1856, 0.2634, 0.9140, 0.9725],
+                            ],
+                            [
+                                [0.2041, 0.4248, 0.2371, 0.7256],
+                                [0.2168, 0.5380, 0.4538, 0.7007],
+                                [0.9013, 0.2623, 0.0739, 0.2998],
+                                [0.1366, 0.5590, 0.2952, 0.4592],
+                            ],
+                        ],
+                    ]
+                ]
+            ),
+            "targets": torch.tensor(
+                [
+                    [
+                        [[0, 1, 0, 1], [1, 2, 1, 0], [2, 1, 1, 1], [1, 1, 0, 1]],
+                        [[2, 1, 0, 2], [1, 2, 0, 2], [1, 0, 1, 1], [1, 1, 0, 0]],
+                        [[1, 0, 2, 1], [0, 2, 2, 1], [1, 0, 1, 1], [0, 0, 2, 1]],
+                        [[2, 1, 1, 0], [1, 0, 0, 2], [1, 0, 2, 1], [2, 1, 0, 1]],
+                    ]
+                ]
+            ),
+        },
+        1.15035,
+    ],
+]
+
+
+class TestNACLLoss(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_result(self, input_param, input_data, expected_val):
+        loss = NACLLoss(**input_param)
+        result = loss(**input_data)
+        np.testing.assert_allclose(result.detach().cpu().numpy(), expected_val, atol=1e-4, rtol=1e-4)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 4a3117fe6bbfbd8e6e33d6bc5d36f8ae70135ddd Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 9 Aug 2024 13:27:38 +0800
Subject: [PATCH 113/183] Ensure location as tuple in wsireader (#8007)

Fixes #8006

### Description

Ensure location as tuple in wsireader

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/data/wsi_datasets.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/monai/data/wsi_datasets.py b/monai/data/wsi_datasets.py
index 3488029a7a..2ee8c9d363 100644
--- a/monai/data/wsi_datasets.py
+++ b/monai/data/wsi_datasets.py
@@ -23,7 +23,7 @@
 from monai.data.utils import iter_patch_position
 from monai.data.wsi_reader import BaseWSIReader, WSIReader
 from monai.transforms import ForegroundMask, Randomizable, apply_transform
-from monai.utils import convert_to_dst_type, ensure_tuple_rep
+from monai.utils import convert_to_dst_type, ensure_tuple, ensure_tuple_rep
 from monai.utils.enums import CommonKeys, ProbMapKeys, WSIPatchKeys
 
 __all__ = ["PatchWSIDataset", "SlidingPatchWSIDataset", "MaskedPatchWSIDataset"]
@@ -123,9 +123,9 @@ def _get_label(self, sample: dict):
     def _get_location(self, sample: dict):
         if self.center_location:
             size = self._get_size(sample)
-            return [sample[WSIPatchKeys.LOCATION][i] - size[i] // 2 for i in range(len(size))]
+            return ensure_tuple(sample[WSIPatchKeys.LOCATION][i] - size[i] // 2 for i in range(len(size)))
         else:
-            return sample[WSIPatchKeys.LOCATION]
+            return ensure_tuple(sample[WSIPatchKeys.LOCATION])
 
     def _get_level(self, sample: dict):
         if self.patch_level is None:

From 0bb05d7bca54db8c3cf670b1d27883f4116c21dc Mon Sep 17 00:00:00 2001
From: ytl0623 <david89062388@gmail.com>
Date: Fri, 9 Aug 2024 14:35:57 +0800
Subject: [PATCH 114/183] Add label smoothing param in DiceCELoss (#8000)

Fixes #7957

### Description
In this modified version I made the following changes:
1. Added `label_smoothing: float = 0.0` parameter in `__init__` method,
default value is 0.0.
2. When creating the `self.cross_entropy` instance, pass the
`label_smoothing` parameter to `nn.CrossEntropyLoss`.
3. Added `self.label_smoothing = label_smoothing` in the `__init__`
method to save this parameter for access when needed.

For example:
```
from monai.losses import DiceCELoss

# Before
criterion = DiceCELoss()
criterion.cross_entropy.label_smoothing = 0.1

# Now
criterion = DiceCELoss(label_smoothing=0.1)
```

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: ytl0623 <david89062388@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/losses/dice.py | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/monai/losses/dice.py b/monai/losses/dice.py
index 07a38d9572..44cde41e5d 100644
--- a/monai/losses/dice.py
+++ b/monai/losses/dice.py
@@ -666,6 +666,7 @@ def __init__(
         weight: torch.Tensor | None = None,
         lambda_dice: float = 1.0,
         lambda_ce: float = 1.0,
+        label_smoothing: float = 0.0,
     ) -> None:
         """
         Args:
@@ -704,6 +705,9 @@ def __init__(
                 Defaults to 1.0.
             lambda_ce: the trade-off weight value for cross entropy loss. The value should be no less than 0.0.
                 Defaults to 1.0.
+            label_smoothing: a value in [0, 1] range. If > 0, the labels are smoothed
+                by the given factor to reduce overfitting.
+                Defaults to 0.0.
 
         """
         super().__init__()
@@ -728,7 +732,12 @@ def __init__(
             batch=batch,
             weight=dice_weight,
         )
-        self.cross_entropy = nn.CrossEntropyLoss(weight=weight, reduction=reduction)
+        if pytorch_after(1, 10):
+            self.cross_entropy = nn.CrossEntropyLoss(
+                weight=weight, reduction=reduction, label_smoothing=label_smoothing
+            )
+        else:
+            self.cross_entropy = nn.CrossEntropyLoss(weight=weight, reduction=reduction)
         self.binary_cross_entropy = nn.BCEWithLogitsLoss(pos_weight=weight, reduction=reduction)
         if lambda_dice < 0.0:
             raise ValueError("lambda_dice should be no less than 0.0.")

From 069519dfc89e984dd10d997497e5c7c1aa963b5c Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 9 Aug 2024 16:00:04 +0800
Subject: [PATCH 115/183] Add include_fc and `use_combined_linear` argument in
 the `SABlock` (#7996)

Fixes #7991
Fixes #7992

### Description
Add `include_fc` and `use_combined_linear` argument in the `SABlock`.


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/networks/blocks/crossattention.py       |  33 ++--
 monai/networks/blocks/selfattention.py        |  60 ++++---
 monai/networks/blocks/spatialattention.py     |  11 +-
 monai/networks/blocks/transformerblock.py     |   8 +-
 monai/networks/nets/autoencoderkl.py          |  73 +++++---
 monai/networks/nets/controlnet.py             |  36 ++--
 monai/networks/nets/diffusion_model_unet.py   | 160 +++++++++++++++---
 monai/networks/nets/spade_autoencoderkl.py    |  22 +++
 .../nets/spade_diffusion_model_unet.py        |  41 ++++-
 monai/networks/nets/transformer.py            |  34 ++--
 tests/test_crossattention.py                  |  17 +-
 tests/test_selfattention.py                   |  66 ++++++--
 tests/test_unetr.py                           |   2 +-
 13 files changed, 426 insertions(+), 137 deletions(-)

diff --git a/monai/networks/blocks/crossattention.py b/monai/networks/blocks/crossattention.py
index daa5abdd56..bdecf63168 100644
--- a/monai/networks/blocks/crossattention.py
+++ b/monai/networks/blocks/crossattention.py
@@ -59,13 +59,12 @@ def __init__(
             causal (bool, optional): whether to use causal attention.
             sequence_length (int, optional): if causal is True, it is necessary to specify the sequence length.
             rel_pos_embedding (str, optional): Add relative positional embeddings to the attention map. For now only
-            "decomposed" is supported (see https://arxiv.org/abs/2112.01526). 2D and 3D are supported.
+                "decomposed" is supported (see https://arxiv.org/abs/2112.01526). 2D and 3D are supported.
             input_size (tuple(spatial_dim), optional): Input resolution for calculating the relative positional
-            parameter size.
+                parameter size.
             attention_dtype: cast attention operations to this dtype.
-            use_flash_attention: if True, use Pytorch's inbuilt
-            flash attention for a memory efficient attention mechanism (see
-            https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
+            use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+                (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
         """
 
         super().__init__()
@@ -109,7 +108,7 @@ def __init__(
         self.to_v = nn.Linear(self.context_input_size, inner_size, bias=qkv_bias)
         self.input_rearrange = Rearrange("b h (l d) -> b l h d", l=num_heads)
 
-        self.out_rearrange = Rearrange("b h l d -> b l (h d)")
+        self.out_rearrange = Rearrange("b l h d -> b h (l d)")
         self.drop_output = nn.Dropout(dropout_rate)
         self.drop_weights = nn.Dropout(dropout_rate)
         self.dropout_rate = dropout_rate
@@ -152,31 +151,20 @@ def forward(self, x: torch.Tensor, context: Optional[torch.Tensor] = None):
         # calculate query, key, values for all heads in batch and move head forward to be the batch dim
         b, t, c = x.size()  # batch size, sequence length, embedding dimensionality (hidden_size)
 
-        q = self.to_q(x)
+        q = self.input_rearrange(self.to_q(x))
         kv = context if context is not None else x
         _, kv_t, _ = kv.size()
-        k = self.to_k(kv)
-        v = self.to_v(kv)
+        k = self.input_rearrange(self.to_k(kv))
+        v = self.input_rearrange(self.to_v(kv))
 
         if self.attention_dtype is not None:
             q = q.to(self.attention_dtype)
             k = k.to(self.attention_dtype)
 
-        q = q.view(b, t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, t,  hs) #
-        k = k.view(b, kv_t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, kv_t, hs)
-        v = v.view(b, kv_t, self.num_heads, c // self.num_heads).transpose(1, 2)  # (b, nh, kv_t, hs)
-
         if self.use_flash_attention:
             x = torch.nn.functional.scaled_dot_product_attention(
-                query=q.transpose(1, 2),
-                key=k.transpose(1, 2),
-                value=v.transpose(1, 2),
-                scale=self.scale,
-                dropout_p=self.dropout_rate,
-                is_causal=self.causal,
-            ).transpose(
-                1, 2
-            )  # Back to (b, nh, t, hs)
+                query=q, key=k, value=v, scale=self.scale, dropout_p=self.dropout_rate, is_causal=self.causal
+            )
         else:
             att_mat = torch.einsum("blxd,blyd->blxy", q, k) * self.scale
             # apply relative positional embedding if defined
@@ -195,6 +183,7 @@ def forward(self, x: torch.Tensor, context: Optional[torch.Tensor] = None):
 
             att_mat = self.drop_weights(att_mat)
             x = torch.einsum("bhxy,bhyd->bhxd", att_mat, v)
+
         x = self.out_rearrange(x)
         x = self.out_proj(x)
         x = self.drop_output(x)
diff --git a/monai/networks/blocks/selfattention.py b/monai/networks/blocks/selfattention.py
index 124c00acc6..ac96b077bd 100644
--- a/monai/networks/blocks/selfattention.py
+++ b/monai/networks/blocks/selfattention.py
@@ -11,7 +11,7 @@
 
 from __future__ import annotations
 
-from typing import Optional, Tuple
+from typing import Tuple, Union
 
 import torch
 import torch.nn as nn
@@ -40,9 +40,11 @@ def __init__(
         hidden_input_size: int | None = None,
         causal: bool = False,
         sequence_length: int | None = None,
-        rel_pos_embedding: Optional[str] = None,
-        input_size: Optional[Tuple] = None,
-        attention_dtype: Optional[torch.dtype] = None,
+        rel_pos_embedding: str | None = None,
+        input_size: Tuple | None = None,
+        attention_dtype: torch.dtype | None = None,
+        include_fc: bool = True,
+        use_combined_linear: bool = True,
         use_flash_attention: bool = False,
     ) -> None:
         """
@@ -61,9 +63,10 @@ def __init__(
             input_size (tuple(spatial_dim), optional): Input resolution for calculating the relative
                 positional parameter size.
             attention_dtype: cast attention operations to this dtype.
-            use_flash_attention: if True, use Pytorch's inbuilt
-            flash attention for a memory efficient attention mechanism (see
-            https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
+            include_fc: whether to include the final linear layer. Default to True.
+            use_combined_linear: whether to use a single linear layer for qkv projection, default to True.
+            use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+                (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
 
         """
 
@@ -105,9 +108,22 @@ def __init__(
         self.hidden_input_size = hidden_input_size if hidden_input_size else hidden_size
         self.out_proj = nn.Linear(self.inner_dim, self.hidden_input_size)
 
-        self.qkv = nn.Linear(self.hidden_input_size, self.inner_dim * 3, bias=qkv_bias)
-        self.input_rearrange = Rearrange("b h (qkv l d) -> qkv b l h d", qkv=3, l=num_heads)
-        self.out_rearrange = Rearrange("b h l d -> b l (h d)")
+        self.qkv: Union[nn.Linear, nn.Identity]
+        self.to_q: Union[nn.Linear, nn.Identity]
+        self.to_k: Union[nn.Linear, nn.Identity]
+        self.to_v: Union[nn.Linear, nn.Identity]
+
+        if use_combined_linear:
+            self.qkv = nn.Linear(self.hidden_input_size, self.inner_dim * 3, bias=qkv_bias)
+            self.to_q = self.to_k = self.to_v = nn.Identity()  # add to enable torchscript
+            self.input_rearrange = Rearrange("b h (qkv l d) -> qkv b l h d", qkv=3, l=num_heads)
+        else:
+            self.to_q = nn.Linear(self.hidden_input_size, self.inner_dim, bias=qkv_bias)
+            self.to_k = nn.Linear(self.hidden_input_size, self.inner_dim, bias=qkv_bias)
+            self.to_v = nn.Linear(self.hidden_input_size, self.inner_dim, bias=qkv_bias)
+            self.qkv = nn.Identity()  # add to enable torchscript
+            self.input_rearrange = Rearrange("b h (l d) -> b l h d", l=num_heads)
+        self.out_rearrange = Rearrange("b l h d -> b h (l d)")
         self.drop_output = nn.Dropout(dropout_rate)
         self.drop_weights = nn.Dropout(dropout_rate)
         self.dropout_rate = dropout_rate
@@ -117,6 +133,8 @@ def __init__(
         self.attention_dtype = attention_dtype
         self.causal = causal
         self.sequence_length = sequence_length
+        self.include_fc = include_fc
+        self.use_combined_linear = use_combined_linear
         self.use_flash_attention = use_flash_attention
 
         if causal and sequence_length is not None:
@@ -144,8 +162,13 @@ def forward(self, x):
         Return:
             torch.Tensor: B x (s_dim_1 * ... * s_dim_n) x C
         """
-        output = self.input_rearrange(self.qkv(x))
-        q, k, v = output[0], output[1], output[2]
+        if self.use_combined_linear:
+            output = self.input_rearrange(self.qkv(x))
+            q, k, v = output[0], output[1], output[2]
+        else:
+            q = self.input_rearrange(self.to_q(x))
+            k = self.input_rearrange(self.to_k(x))
+            v = self.input_rearrange(self.to_v(x))
 
         if self.attention_dtype is not None:
             q = q.to(self.attention_dtype)
@@ -153,13 +176,8 @@ def forward(self, x):
 
         if self.use_flash_attention:
             x = F.scaled_dot_product_attention(
-                query=q.transpose(1, 2),
-                key=k.transpose(1, 2),
-                value=v.transpose(1, 2),
-                scale=self.scale,
-                dropout_p=self.dropout_rate,
-                is_causal=self.causal,
-            ).transpose(1, 2)
+                query=q, key=k, value=v, scale=self.scale, dropout_p=self.dropout_rate, is_causal=self.causal
+            )
         else:
             att_mat = torch.einsum("blxd,blyd->blxy", q, k) * self.scale
 
@@ -179,7 +197,9 @@ def forward(self, x):
 
             att_mat = self.drop_weights(att_mat)
             x = torch.einsum("bhxy,bhyd->bhxd", att_mat, v)
+
         x = self.out_rearrange(x)
-        x = self.out_proj(x)
+        if self.include_fc:
+            x = self.out_proj(x)
         x = self.drop_output(x)
         return x
diff --git a/monai/networks/blocks/spatialattention.py b/monai/networks/blocks/spatialattention.py
index 1cfafb1585..665442b55e 100644
--- a/monai/networks/blocks/spatialattention.py
+++ b/monai/networks/blocks/spatialattention.py
@@ -32,8 +32,13 @@ class SpatialAttentionBlock(nn.Module):
         spatial_dims: number of spatial dimensions, could be 1, 2, or 3.
         num_channels: number of input channels. Must be divisible by num_head_channels.
         num_head_channels: number of channels per head.
+        norm_num_groups: Number of groups for the group norm layer.
+        norm_eps: Epsilon for the normalization.
         attention_dtype: cast attention operations to this dtype.
-        use_flash_attention: if True, use flash attention for a memory efficient attention mechanism.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
 
     """
 
@@ -45,6 +50,8 @@ def __init__(
         norm_num_groups: int = 32,
         norm_eps: float = 1e-6,
         attention_dtype: Optional[torch.dtype] = None,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
         use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
@@ -60,6 +67,8 @@ def __init__(
             num_heads=num_heads,
             qkv_bias=True,
             attention_dtype=attention_dtype,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
             use_flash_attention=use_flash_attention,
         )
 
diff --git a/monai/networks/blocks/transformerblock.py b/monai/networks/blocks/transformerblock.py
index 28d9c563ac..05eb3b07ab 100644
--- a/monai/networks/blocks/transformerblock.py
+++ b/monai/networks/blocks/transformerblock.py
@@ -37,6 +37,8 @@ def __init__(
         sequence_length: int | None = None,
         with_cross_attention: bool = False,
         use_flash_attention: bool = False,
+        include_fc: bool = True,
+        use_combined_linear: bool = True,
     ) -> None:
         """
         Args:
@@ -47,7 +49,9 @@ def __init__(
             qkv_bias(bool, optional): apply bias term for the qkv linear layer. Defaults to False.
             save_attn (bool, optional): to make accessible the attention matrix. Defaults to False.
             use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
-             (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
+                (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
+            include_fc: whether to include the final linear layer. Default to True.
+            use_combined_linear: whether to use a single linear layer for qkv projection, default to True.
 
         """
 
@@ -69,6 +73,8 @@ def __init__(
             save_attn=save_attn,
             causal=causal,
             sequence_length=sequence_length,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
             use_flash_attention=use_flash_attention,
         )
         self.norm2 = nn.LayerNorm(hidden_size)
diff --git a/monai/networks/nets/autoencoderkl.py b/monai/networks/nets/autoencoderkl.py
index 35d80e0565..836027796f 100644
--- a/monai/networks/nets/autoencoderkl.py
+++ b/monai/networks/nets/autoencoderkl.py
@@ -157,6 +157,10 @@ class Encoder(nn.Module):
         norm_eps: epsilon for the normalization.
         attention_levels: indicate which level from num_channels contain an attention block.
         with_nonlocal_attn: if True use non-local attention block.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -170,6 +174,9 @@ def __init__(
         norm_eps: float,
         attention_levels: Sequence[bool],
         with_nonlocal_attn: bool = True,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.spatial_dims = spatial_dims
@@ -220,6 +227,9 @@ def __init__(
                             num_channels=input_channel,
                             norm_num_groups=norm_num_groups,
                             norm_eps=norm_eps,
+                            include_fc=include_fc,
+                            use_combined_linear=use_combined_linear,
+                            use_flash_attention=use_flash_attention,
                         )
                     )
 
@@ -243,6 +253,9 @@ def __init__(
                     num_channels=channels[-1],
                     norm_num_groups=norm_num_groups,
                     norm_eps=norm_eps,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
             )
             blocks.append(
@@ -291,6 +304,10 @@ class Decoder(nn.Module):
         attention_levels: indicate which level from num_channels contain an attention block.
         with_nonlocal_attn: if True use non-local attention block.
         use_convtranspose: if True, use ConvTranspose to upsample feature maps in decoder.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -305,6 +322,9 @@ def __init__(
         attention_levels: Sequence[bool],
         with_nonlocal_attn: bool = True,
         use_convtranspose: bool = False,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.spatial_dims = spatial_dims
@@ -350,6 +370,9 @@ def __init__(
                     num_channels=reversed_block_out_channels[0],
                     norm_num_groups=norm_num_groups,
                     norm_eps=norm_eps,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
             )
             blocks.append(
@@ -389,6 +412,9 @@ def __init__(
                             num_channels=block_in_ch,
                             norm_num_groups=norm_num_groups,
                             norm_eps=norm_eps,
+                            include_fc=include_fc,
+                            use_combined_linear=use_combined_linear,
+                            use_flash_attention=use_flash_attention,
                         )
                     )
 
@@ -463,6 +489,10 @@ class AutoencoderKL(nn.Module):
         with_decoder_nonlocal_attn: if True use non-local attention block in the decoder.
         use_checkpoint: if True, use activation checkpoint to save memory.
         use_convtranspose: if True, use ConvTranspose to upsample feature maps in decoder.
+        include_fc: whether to include the final linear layer in the attention block. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection in the attention block, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -480,6 +510,9 @@ def __init__(
         with_decoder_nonlocal_attn: bool = True,
         use_checkpoint: bool = False,
         use_convtranspose: bool = False,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
 
@@ -509,6 +542,9 @@ def __init__(
             norm_eps=norm_eps,
             attention_levels=attention_levels,
             with_nonlocal_attn=with_encoder_nonlocal_attn,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
         self.decoder = Decoder(
             spatial_dims=spatial_dims,
@@ -521,6 +557,9 @@ def __init__(
             attention_levels=attention_levels,
             with_nonlocal_attn=with_decoder_nonlocal_attn,
             use_convtranspose=use_convtranspose,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
         self.quant_conv_mu = Convolution(
             spatial_dims=spatial_dims,
@@ -665,27 +704,18 @@ def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
         # copy over all matching keys
         for k in new_state_dict:
             if k in old_state_dict:
-                new_state_dict[k] = old_state_dict[k]
+                new_state_dict[k] = old_state_dict.pop(k)
 
         # fix the attention blocks
-        attention_blocks = [k.replace(".attn.qkv.weight", "") for k in new_state_dict if "attn.qkv.weight" in k]
+        attention_blocks = [k.replace(".attn.to_q.weight", "") for k in new_state_dict if "attn.to_q.weight" in k]
         for block in attention_blocks:
-            new_state_dict[f"{block}.attn.qkv.weight"] = torch.cat(
-                [
-                    old_state_dict[f"{block}.to_q.weight"],
-                    old_state_dict[f"{block}.to_k.weight"],
-                    old_state_dict[f"{block}.to_v.weight"],
-                ],
-                dim=0,
-            )
-            new_state_dict[f"{block}.attn.qkv.bias"] = torch.cat(
-                [
-                    old_state_dict[f"{block}.to_q.bias"],
-                    old_state_dict[f"{block}.to_k.bias"],
-                    old_state_dict[f"{block}.to_v.bias"],
-                ],
-                dim=0,
-            )
+            new_state_dict[f"{block}.attn.to_q.weight"] = old_state_dict.pop(f"{block}.to_q.weight")
+            new_state_dict[f"{block}.attn.to_k.weight"] = old_state_dict.pop(f"{block}.to_k.weight")
+            new_state_dict[f"{block}.attn.to_v.weight"] = old_state_dict.pop(f"{block}.to_v.weight")
+            new_state_dict[f"{block}.attn.to_q.bias"] = old_state_dict.pop(f"{block}.to_q.bias")
+            new_state_dict[f"{block}.attn.to_k.bias"] = old_state_dict.pop(f"{block}.to_k.bias")
+            new_state_dict[f"{block}.attn.to_v.bias"] = old_state_dict.pop(f"{block}.to_v.bias")
+
             # old version did not have a projection so set these to the identity
             new_state_dict[f"{block}.attn.out_proj.weight"] = torch.eye(
                 new_state_dict[f"{block}.attn.out_proj.weight"].shape[0]
@@ -698,5 +728,8 @@ def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
         for k in new_state_dict:
             if "postconv" in k:
                 old_name = k.replace("postconv", "conv")
-                new_state_dict[k] = old_state_dict[old_name]
-        self.load_state_dict(new_state_dict)
+                new_state_dict[k] = old_state_dict.pop(old_name)
+        if verbose:
+            # print all remaining keys in old_state_dict
+            print("remaining keys in old_state_dict:", old_state_dict.keys())
+        self.load_state_dict(new_state_dict, strict=True)
diff --git a/monai/networks/nets/controlnet.py b/monai/networks/nets/controlnet.py
index ed3654733d..8b08eaae10 100644
--- a/monai/networks/nets/controlnet.py
+++ b/monai/networks/nets/controlnet.py
@@ -143,6 +143,10 @@ class ControlNet(nn.Module):
         upcast_attention: if True, upcast attention operations to full precision.
         conditioning_embedding_in_channels: number of input channels for the conditioning embedding.
         conditioning_embedding_num_channels: number of channels for the blocks in the conditioning embedding.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to True.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -163,6 +167,9 @@ def __init__(
         upcast_attention: bool = False,
         conditioning_embedding_in_channels: int = 1,
         conditioning_embedding_num_channels: Sequence[int] = (16, 32, 96, 256),
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         if with_conditioning is True and cross_attention_dim is None:
@@ -282,6 +289,9 @@ def __init__(
                 transformer_num_layers=transformer_num_layers,
                 cross_attention_dim=cross_attention_dim,
                 upcast_attention=upcast_attention,
+                include_fc=include_fc,
+                use_combined_linear=use_combined_linear,
+                use_flash_attention=use_flash_attention,
             )
 
             self.down_blocks.append(down_block)
@@ -326,6 +336,9 @@ def __init__(
             transformer_num_layers=transformer_num_layers,
             cross_attention_dim=cross_attention_dim,
             upcast_attention=upcast_attention,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
 
         controlnet_block = Convolution(
@@ -441,25 +454,16 @@ def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
         # copy over all matching keys
         for k in new_state_dict:
             if k in old_state_dict:
-                new_state_dict[k] = old_state_dict[k]
+                new_state_dict[k] = old_state_dict.pop(k)
 
         # fix the attention blocks
-        attention_blocks = [k.replace(".attn1.qkv.weight", "") for k in new_state_dict if "attn1.qkv.weight" in k]
+        attention_blocks = [k.replace(".out_proj.weight", "") for k in new_state_dict if "out_proj.weight" in k]
         for block in attention_blocks:
-            new_state_dict[f"{block}.attn1.qkv.weight"] = torch.cat(
-                [
-                    old_state_dict[f"{block}.attn1.to_q.weight"],
-                    old_state_dict[f"{block}.attn1.to_k.weight"],
-                    old_state_dict[f"{block}.attn1.to_v.weight"],
-                ],
-                dim=0,
-            )
-
             # projection
-            new_state_dict[f"{block}.attn1.out_proj.weight"] = old_state_dict[f"{block}.attn1.to_out.0.weight"]
-            new_state_dict[f"{block}.attn1.out_proj.bias"] = old_state_dict[f"{block}.attn1.to_out.0.bias"]
-
-            new_state_dict[f"{block}.attn2.out_proj.weight"] = old_state_dict[f"{block}.attn2.to_out.0.weight"]
-            new_state_dict[f"{block}.attn2.out_proj.bias"] = old_state_dict[f"{block}.attn2.to_out.0.bias"]
+            new_state_dict[f"{block}.out_proj.weight"] = old_state_dict.pop(f"{block}.to_out.0.weight")
+            new_state_dict[f"{block}.out_proj.bias"] = old_state_dict.pop(f"{block}.to_out.0.bias")
 
+        if verbose:
+            # print all remaining keys in old_state_dict
+            print("remaining keys in old_state_dict:", old_state_dict.keys())
         self.load_state_dict(new_state_dict)
diff --git a/monai/networks/nets/diffusion_model_unet.py b/monai/networks/nets/diffusion_model_unet.py
index a885339d0d..f57fe251d2 100644
--- a/monai/networks/nets/diffusion_model_unet.py
+++ b/monai/networks/nets/diffusion_model_unet.py
@@ -67,7 +67,9 @@ class DiffusionUNetTransformerBlock(nn.Module):
         cross_attention_dim: size of the context vector for cross attention.
         upcast_attention: if True, upcast attention operations to full precision.
         use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
-        (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
 
     """
 
@@ -80,6 +82,8 @@ def __init__(
         cross_attention_dim: int | None = None,
         upcast_attention: bool = False,
         use_flash_attention: bool = False,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
     ) -> None:
         super().__init__()
         self.attn1 = SABlock(
@@ -89,6 +93,8 @@ def __init__(
             dim_head=num_head_channels,
             dropout_rate=dropout,
             attention_dtype=torch.float if upcast_attention else None,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
             use_flash_attention=use_flash_attention,
         )
         self.ff = MLPBlock(hidden_size=num_channels, mlp_dim=num_channels * 4, act="GEGLU", dropout_rate=dropout)
@@ -134,6 +140,11 @@ class SpatialTransformer(nn.Module):
         norm_eps: epsilon for the normalization.
         cross_attention_dim: number of context dimensions to use.
         upcast_attention: if True, upcast attention operations to full precision.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
+
     """
 
     def __init__(
@@ -148,6 +159,9 @@ def __init__(
         norm_eps: float = 1e-6,
         cross_attention_dim: int | None = None,
         upcast_attention: bool = False,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.spatial_dims = spatial_dims
@@ -175,6 +189,9 @@ def __init__(
                     dropout=dropout,
                     cross_attention_dim=cross_attention_dim,
                     upcast_attention=upcast_attention,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
                 for _ in range(num_layers)
             ]
@@ -529,6 +546,10 @@ class AttnDownBlock(nn.Module):
         resblock_updown: if True use residual blocks for downsampling.
         downsample_padding: padding used in the downsampling block.
         num_head_channels: number of channels in each attention head.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -544,6 +565,9 @@ def __init__(
         resblock_updown: bool = False,
         downsample_padding: int = 1,
         num_head_channels: int = 1,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.resblock_updown = resblock_updown
@@ -570,6 +594,9 @@ def __init__(
                     num_head_channels=num_head_channels,
                     norm_num_groups=norm_num_groups,
                     norm_eps=norm_eps,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
             )
 
@@ -636,7 +663,11 @@ class CrossAttnDownBlock(nn.Module):
         transformer_num_layers: number of layers of Transformer blocks to use.
         cross_attention_dim: number of context dimensions to use.
         upcast_attention: if True, upcast attention operations to full precision.
-        dropout_cattn: if different from zero, this will be the dropout value for the cross-attention layers
+        dropout_cattn: if different from zero, this will be the dropout value for the cross-attention layers.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -656,6 +687,9 @@ def __init__(
         cross_attention_dim: int | None = None,
         upcast_attention: bool = False,
         dropout_cattn: float = 0.0,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.resblock_updown = resblock_updown
@@ -688,6 +722,9 @@ def __init__(
                     cross_attention_dim=cross_attention_dim,
                     upcast_attention=upcast_attention,
                     dropout=dropout_cattn,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
             )
 
@@ -745,6 +782,10 @@ class AttnMidBlock(nn.Module):
         norm_num_groups: number of groups for the group normalization.
         norm_eps: epsilon for the group normalization.
         num_head_channels: number of channels in each attention head.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -755,6 +796,9 @@ def __init__(
         norm_num_groups: int = 32,
         norm_eps: float = 1e-6,
         num_head_channels: int = 1,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
 
@@ -772,6 +816,9 @@ def __init__(
             num_head_channels=num_head_channels,
             norm_num_groups=norm_num_groups,
             norm_eps=norm_eps,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
 
         self.resnet_2 = DiffusionUNetResnetBlock(
@@ -808,6 +855,10 @@ class CrossAttnMidBlock(nn.Module):
         transformer_num_layers: number of layers of Transformer blocks to use.
         cross_attention_dim: number of context dimensions to use.
         upcast_attention: if True, upcast attention operations to full precision.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -822,6 +873,9 @@ def __init__(
         cross_attention_dim: int | None = None,
         upcast_attention: bool = False,
         dropout_cattn: float = 0.0,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
 
@@ -844,6 +898,9 @@ def __init__(
             cross_attention_dim=cross_attention_dim,
             upcast_attention=upcast_attention,
             dropout=dropout_cattn,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
         self.resnet_2 = DiffusionUNetResnetBlock(
             spatial_dims=spatial_dims,
@@ -989,6 +1046,10 @@ class AttnUpBlock(nn.Module):
         add_upsample: if True add downsample block.
         resblock_updown: if True use residual blocks for upsampling.
         num_head_channels: number of channels in each attention head.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -1004,6 +1065,9 @@ def __init__(
         add_upsample: bool = True,
         resblock_updown: bool = False,
         num_head_channels: int = 1,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.resblock_updown = resblock_updown
@@ -1032,6 +1096,9 @@ def __init__(
                     num_head_channels=num_head_channels,
                     norm_num_groups=norm_num_groups,
                     norm_eps=norm_eps,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
             )
 
@@ -1116,7 +1183,11 @@ class CrossAttnUpBlock(nn.Module):
         transformer_num_layers: number of layers of Transformer blocks to use.
         cross_attention_dim: number of context dimensions to use.
         upcast_attention: if True, upcast attention operations to full precision.
-        dropout_cattn: if different from zero, this will be the dropout value for the cross-attention layers
+        dropout_cattn: if different from zero, this will be the dropout value for the cross-attention layers.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -1136,6 +1207,9 @@ def __init__(
         cross_attention_dim: int | None = None,
         upcast_attention: bool = False,
         dropout_cattn: float = 0.0,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.resblock_updown = resblock_updown
@@ -1169,6 +1243,9 @@ def __init__(
                     cross_attention_dim=cross_attention_dim,
                     upcast_attention=upcast_attention,
                     dropout=dropout_cattn,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
             )
 
@@ -1250,6 +1327,9 @@ def get_down_block(
     cross_attention_dim: int | None,
     upcast_attention: bool = False,
     dropout_cattn: float = 0.0,
+    include_fc: bool = True,
+    use_combined_linear: bool = False,
+    use_flash_attention: bool = False,
 ) -> nn.Module:
     if with_attn:
         return AttnDownBlock(
@@ -1263,6 +1343,9 @@ def get_down_block(
             add_downsample=add_downsample,
             resblock_updown=resblock_updown,
             num_head_channels=num_head_channels,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
     elif with_cross_attn:
         return CrossAttnDownBlock(
@@ -1280,6 +1363,9 @@ def get_down_block(
             cross_attention_dim=cross_attention_dim,
             upcast_attention=upcast_attention,
             dropout_cattn=dropout_cattn,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
     else:
         return DownBlock(
@@ -1307,6 +1393,9 @@ def get_mid_block(
     cross_attention_dim: int | None,
     upcast_attention: bool = False,
     dropout_cattn: float = 0.0,
+    include_fc: bool = True,
+    use_combined_linear: bool = False,
+    use_flash_attention: bool = False,
 ) -> nn.Module:
     if with_conditioning:
         return CrossAttnMidBlock(
@@ -1320,6 +1409,9 @@ def get_mid_block(
             cross_attention_dim=cross_attention_dim,
             upcast_attention=upcast_attention,
             dropout_cattn=dropout_cattn,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
     else:
         return AttnMidBlock(
@@ -1329,6 +1421,9 @@ def get_mid_block(
             norm_num_groups=norm_num_groups,
             norm_eps=norm_eps,
             num_head_channels=num_head_channels,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
 
 
@@ -1350,6 +1445,9 @@ def get_up_block(
     cross_attention_dim: int | None,
     upcast_attention: bool = False,
     dropout_cattn: float = 0.0,
+    include_fc: bool = True,
+    use_combined_linear: bool = False,
+    use_flash_attention: bool = False,
 ) -> nn.Module:
     if with_attn:
         return AttnUpBlock(
@@ -1364,6 +1462,9 @@ def get_up_block(
             add_upsample=add_upsample,
             resblock_updown=resblock_updown,
             num_head_channels=num_head_channels,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
     elif with_cross_attn:
         return CrossAttnUpBlock(
@@ -1382,6 +1483,9 @@ def get_up_block(
             cross_attention_dim=cross_attention_dim,
             upcast_attention=upcast_attention,
             dropout_cattn=dropout_cattn,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
     else:
         return UpBlock(
@@ -1419,9 +1523,13 @@ class DiffusionModelUNet(nn.Module):
         transformer_num_layers: number of layers of Transformer blocks to use.
         cross_attention_dim: number of context dimensions to use.
         num_class_embeds: if specified (as an int), then this model will be class-conditional with `num_class_embeds`
-        classes.
+            classes.
         upcast_attention: if True, upcast attention operations to full precision.
-        dropout_cattn: if different from zero, this will be the dropout value for the cross-attention layers
+        dropout_cattn: if different from zero, this will be the dropout value for the cross-attention layers.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to True.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -1442,6 +1550,9 @@ def __init__(
         num_class_embeds: int | None = None,
         upcast_attention: bool = False,
         dropout_cattn: float = 0.0,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         if with_conditioning is True and cross_attention_dim is None:
@@ -1536,6 +1647,9 @@ def __init__(
                 cross_attention_dim=cross_attention_dim,
                 upcast_attention=upcast_attention,
                 dropout_cattn=dropout_cattn,
+                include_fc=include_fc,
+                use_combined_linear=use_combined_linear,
+                use_flash_attention=use_flash_attention,
             )
 
             self.down_blocks.append(down_block)
@@ -1553,6 +1667,9 @@ def __init__(
             cross_attention_dim=cross_attention_dim,
             upcast_attention=upcast_attention,
             dropout_cattn=dropout_cattn,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
 
         # up
@@ -1587,6 +1704,9 @@ def __init__(
                 cross_attention_dim=cross_attention_dim,
                 upcast_attention=upcast_attention,
                 dropout_cattn=dropout_cattn,
+                include_fc=include_fc,
+                use_combined_linear=use_combined_linear,
+                use_flash_attention=use_flash_attention,
             )
 
             self.up_blocks.append(up_block)
@@ -1714,31 +1834,23 @@ def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
         # copy over all matching keys
         for k in new_state_dict:
             if k in old_state_dict:
-                new_state_dict[k] = old_state_dict[k]
+                new_state_dict[k] = old_state_dict.pop(k)
 
         # fix the attention blocks
-        attention_blocks = [k.replace(".attn1.qkv.weight", "") for k in new_state_dict if "attn1.qkv.weight" in k]
+        attention_blocks = [k.replace(".out_proj.weight", "") for k in new_state_dict if "out_proj.weight" in k]
         for block in attention_blocks:
-            new_state_dict[f"{block}.attn1.qkv.weight"] = torch.cat(
-                [
-                    old_state_dict[f"{block}.attn1.to_q.weight"],
-                    old_state_dict[f"{block}.attn1.to_k.weight"],
-                    old_state_dict[f"{block}.attn1.to_v.weight"],
-                ],
-                dim=0,
-            )
-
             # projection
-            new_state_dict[f"{block}.attn1.out_proj.weight"] = old_state_dict[f"{block}.attn1.to_out.0.weight"]
-            new_state_dict[f"{block}.attn1.out_proj.bias"] = old_state_dict[f"{block}.attn1.to_out.0.bias"]
+            new_state_dict[f"{block}.out_proj.weight"] = old_state_dict.pop(f"{block}.to_out.0.weight")
+            new_state_dict[f"{block}.out_proj.bias"] = old_state_dict.pop(f"{block}.to_out.0.bias")
 
-            new_state_dict[f"{block}.attn2.out_proj.weight"] = old_state_dict[f"{block}.attn2.to_out.0.weight"]
-            new_state_dict[f"{block}.attn2.out_proj.bias"] = old_state_dict[f"{block}.attn2.to_out.0.bias"]
         # fix the upsample conv blocks which were renamed postconv
         for k in new_state_dict:
             if "postconv" in k:
                 old_name = k.replace("postconv", "conv")
-                new_state_dict[k] = old_state_dict[old_name]
+                new_state_dict[k] = old_state_dict.pop(old_name)
+        if verbose:
+            # print all remaining keys in old_state_dict
+            print("remaining keys in old_state_dict:", old_state_dict.keys())
         self.load_state_dict(new_state_dict)
 
 
@@ -1782,6 +1894,9 @@ def __init__(
         cross_attention_dim: int | None = None,
         num_class_embeds: int | None = None,
         upcast_attention: bool = False,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         if with_conditioning is True and cross_attention_dim is None:
@@ -1866,6 +1981,9 @@ def __init__(
                 transformer_num_layers=transformer_num_layers,
                 cross_attention_dim=cross_attention_dim,
                 upcast_attention=upcast_attention,
+                include_fc=include_fc,
+                use_combined_linear=use_combined_linear,
+                use_flash_attention=use_flash_attention,
             )
 
             self.down_blocks.append(down_block)
diff --git a/monai/networks/nets/spade_autoencoderkl.py b/monai/networks/nets/spade_autoencoderkl.py
index d5794a9227..cc8909194a 100644
--- a/monai/networks/nets/spade_autoencoderkl.py
+++ b/monai/networks/nets/spade_autoencoderkl.py
@@ -137,6 +137,10 @@ class SPADEDecoder(nn.Module):
         label_nc: number of semantic channels for SPADE normalisation.
         with_nonlocal_attn: if True use non-local attention block.
         spade_intermediate_channels: number of intermediate channels for SPADE block layer.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -152,6 +156,9 @@ def __init__(
         label_nc: int,
         with_nonlocal_attn: bool = True,
         spade_intermediate_channels: int = 128,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.spatial_dims = spatial_dims
@@ -200,6 +207,9 @@ def __init__(
                     num_channels=reversed_block_out_channels[0],
                     norm_num_groups=norm_num_groups,
                     norm_eps=norm_eps,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
             )
             blocks.append(
@@ -243,6 +253,9 @@ def __init__(
                             num_channels=block_in_ch,
                             norm_num_groups=norm_num_groups,
                             norm_eps=norm_eps,
+                            include_fc=include_fc,
+                            use_combined_linear=use_combined_linear,
+                            use_flash_attention=use_flash_attention,
                         )
                     )
 
@@ -331,6 +344,9 @@ def __init__(
         with_encoder_nonlocal_attn: bool = True,
         with_decoder_nonlocal_attn: bool = True,
         spade_intermediate_channels: int = 128,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
 
@@ -360,6 +376,9 @@ def __init__(
             norm_eps=norm_eps,
             attention_levels=attention_levels,
             with_nonlocal_attn=with_encoder_nonlocal_attn,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
         self.decoder = SPADEDecoder(
             spatial_dims=spatial_dims,
@@ -373,6 +392,9 @@ def __init__(
             label_nc=label_nc,
             with_nonlocal_attn=with_decoder_nonlocal_attn,
             spade_intermediate_channels=spade_intermediate_channels,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
         self.quant_conv_mu = Convolution(
             spatial_dims=spatial_dims,
diff --git a/monai/networks/nets/spade_diffusion_model_unet.py b/monai/networks/nets/spade_diffusion_model_unet.py
index 75d1687df3..a9609b1d39 100644
--- a/monai/networks/nets/spade_diffusion_model_unet.py
+++ b/monai/networks/nets/spade_diffusion_model_unet.py
@@ -325,6 +325,10 @@ class SPADEAttnUpBlock(nn.Module):
         resblock_updown: if True use residual blocks for upsampling.
         num_head_channels: number of channels in each attention head.
         spade_intermediate_channels: number of intermediate channels for SPADE block layer
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -342,6 +346,9 @@ def __init__(
         resblock_updown: bool = False,
         num_head_channels: int = 1,
         spade_intermediate_channels: int = 128,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.resblock_updown = resblock_updown
@@ -371,6 +378,9 @@ def __init__(
                     num_head_channels=num_head_channels,
                     norm_num_groups=norm_num_groups,
                     norm_eps=norm_eps,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
             )
 
@@ -457,6 +467,8 @@ class SPADECrossAttnUpBlock(nn.Module):
         cross_attention_dim: number of context dimensions to use.
         upcast_attention: if True, upcast attention operations to full precision.
         spade_intermediate_channels: number of intermediate channels for SPADE block layer.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism.
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -477,6 +489,9 @@ def __init__(
         cross_attention_dim: int | None = None,
         upcast_attention: bool = False,
         spade_intermediate_channels: int = 128,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.resblock_updown = resblock_updown
@@ -510,6 +525,9 @@ def __init__(
                     num_layers=transformer_num_layers,
                     cross_attention_dim=cross_attention_dim,
                     upcast_attention=upcast_attention,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
             )
 
@@ -592,6 +610,9 @@ def get_spade_up_block(
     cross_attention_dim: int | None,
     upcast_attention: bool = False,
     spade_intermediate_channels: int = 128,
+    include_fc: bool = True,
+    use_combined_linear: bool = False,
+    use_flash_attention: bool = False,
 ) -> nn.Module:
     if with_attn:
         return SPADEAttnUpBlock(
@@ -608,6 +629,9 @@ def get_spade_up_block(
             resblock_updown=resblock_updown,
             num_head_channels=num_head_channels,
             spade_intermediate_channels=spade_intermediate_channels,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
     elif with_cross_attn:
         return SPADECrossAttnUpBlock(
@@ -627,6 +651,7 @@ def get_spade_up_block(
             cross_attention_dim=cross_attention_dim,
             upcast_attention=upcast_attention,
             spade_intermediate_channels=spade_intermediate_channels,
+            use_flash_attention=use_flash_attention,
         )
     else:
         return SPADEUpBlock(
@@ -667,9 +692,11 @@ class SPADEDiffusionModelUNet(nn.Module):
         transformer_num_layers: number of layers of Transformer blocks to use.
         cross_attention_dim: number of context dimensions to use.
         num_class_embeds: if specified (as an int), then this model will be class-conditional with `num_class_embeds`
-        classes.
+            classes.
         upcast_attention: if True, upcast attention operations to full precision.
-        spade_intermediate_channels: number of intermediate channels for SPADE block layer
+        spade_intermediate_channels: number of intermediate channels for SPADE block layer.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -691,6 +718,9 @@ def __init__(
         num_class_embeds: int | None = None,
         upcast_attention: bool = False,
         spade_intermediate_channels: int = 128,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         if with_conditioning is True and cross_attention_dim is None:
@@ -783,6 +813,9 @@ def __init__(
                 transformer_num_layers=transformer_num_layers,
                 cross_attention_dim=cross_attention_dim,
                 upcast_attention=upcast_attention,
+                include_fc=include_fc,
+                use_combined_linear=use_combined_linear,
+                use_flash_attention=use_flash_attention,
             )
 
             self.down_blocks.append(down_block)
@@ -799,6 +832,9 @@ def __init__(
             transformer_num_layers=transformer_num_layers,
             cross_attention_dim=cross_attention_dim,
             upcast_attention=upcast_attention,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
+            use_flash_attention=use_flash_attention,
         )
 
         # up
@@ -834,6 +870,7 @@ def __init__(
                 upcast_attention=upcast_attention,
                 label_nc=label_nc,
                 spade_intermediate_channels=spade_intermediate_channels,
+                use_flash_attention=use_flash_attention,
             )
 
             self.up_blocks.append(up_block)
diff --git a/monai/networks/nets/transformer.py b/monai/networks/nets/transformer.py
index 1af725abda..3a278c112a 100644
--- a/monai/networks/nets/transformer.py
+++ b/monai/networks/nets/transformer.py
@@ -51,6 +51,10 @@ class DecoderOnlyTransformer(nn.Module):
         attn_layers_heads: Number of attention heads.
         with_cross_attention: Whether to use cross attention for conditioning.
         embedding_dropout_rate: Dropout rate for the embedding.
+        include_fc: whether to include the final linear layer. Default to True.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to True.
+        use_flash_attention: if True, use Pytorch's inbuilt flash attention for a memory efficient attention mechanism
+            (see https://pytorch.org/docs/2.2/generated/torch.nn.functional.scaled_dot_product_attention.html).
     """
 
     def __init__(
@@ -62,6 +66,9 @@ def __init__(
         attn_layers_heads: int,
         with_cross_attention: bool = False,
         embedding_dropout_rate: float = 0.0,
+        include_fc: bool = True,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
         self.num_tokens = num_tokens
@@ -86,6 +93,9 @@ def __init__(
                     causal=True,
                     sequence_length=max_seq_len,
                     with_cross_attention=with_cross_attention,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
+                    use_flash_attention=use_flash_attention,
                 )
                 for _ in range(attn_layers_depth)
             ]
@@ -133,25 +143,15 @@ def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
         # copy over all matching keys
         for k in new_state_dict:
             if k in old_state_dict:
-                new_state_dict[k] = old_state_dict[k]
-
-        # fix the attention blocks
-        attention_blocks = [k.replace(".attn.qkv.weight", "") for k in new_state_dict if "attn.qkv.weight" in k]
-        for block in attention_blocks:
-            new_state_dict[f"{block}.attn.qkv.weight"] = torch.cat(
-                [
-                    old_state_dict[f"{block}.attn.to_q.weight"],
-                    old_state_dict[f"{block}.attn.to_k.weight"],
-                    old_state_dict[f"{block}.attn.to_v.weight"],
-                ],
-                dim=0,
-            )
+                new_state_dict[k] = old_state_dict.pop(k)
 
         # fix the renamed norm blocks first  norm2 -> norm_cross_attention , norm3 -> norm2
-        for k in old_state_dict:
+        for k in list(old_state_dict.keys()):
             if "norm2" in k:
-                new_state_dict[k.replace("norm2", "norm_cross_attn")] = old_state_dict[k]
+                new_state_dict[k.replace("norm2", "norm_cross_attn")] = old_state_dict.pop(k)
             if "norm3" in k:
-                new_state_dict[k.replace("norm3", "norm2")] = old_state_dict[k]
-
+                new_state_dict[k.replace("norm3", "norm2")] = old_state_dict.pop(k)
+        if verbose:
+            # print all remaining keys in old_state_dict
+            print("remaining keys in old_state_dict:", old_state_dict.keys())
         self.load_state_dict(new_state_dict)
diff --git a/tests/test_crossattention.py b/tests/test_crossattention.py
index 44458147d6..e034e42290 100644
--- a/tests/test_crossattention.py
+++ b/tests/test_crossattention.py
@@ -22,7 +22,7 @@
 from monai.networks.blocks.crossattention import CrossAttentionBlock
 from monai.networks.layers.factories import RelPosEmbedding
 from monai.utils import optional_import
-from tests.utils import SkipIfBeforePyTorchVersion
+from tests.utils import SkipIfBeforePyTorchVersion, assert_allclose
 
 einops, has_einops = optional_import("einops")
 
@@ -166,6 +166,21 @@ def test_access_attn_matrix(self):
         matrix_acess_blk(torch.randn(input_shape))
         assert matrix_acess_blk.att_mat.shape == (input_shape[0], input_shape[0], input_shape[1], input_shape[1])
 
+    @parameterized.expand([[True], [False]])
+    @skipUnless(has_einops, "Requires einops")
+    @SkipIfBeforePyTorchVersion((2, 0))
+    def test_flash_attention(self, causal):
+        input_param = {"hidden_size": 128, "num_heads": 1, "causal": causal, "sequence_length": 16 if causal else None}
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        block_w_flash_attention = CrossAttentionBlock(**input_param, use_flash_attention=True).to(device)
+        block_wo_flash_attention = CrossAttentionBlock(**input_param, use_flash_attention=False).to(device)
+        block_wo_flash_attention.load_state_dict(block_w_flash_attention.state_dict())
+        test_data = torch.randn(1, 16, 128).to(device)
+
+        out_1 = block_w_flash_attention(test_data)
+        out_2 = block_wo_flash_attention(test_data)
+        assert_allclose(out_1, out_2, atol=1e-4)
+
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/tests/test_selfattention.py b/tests/test_selfattention.py
index 3e98f4c5c4..88919fd8b1 100644
--- a/tests/test_selfattention.py
+++ b/tests/test_selfattention.py
@@ -22,7 +22,7 @@
 from monai.networks.blocks.selfattention import SABlock
 from monai.networks.layers.factories import RelPosEmbedding
 from monai.utils import optional_import
-from tests.utils import SkipIfBeforePyTorchVersion
+from tests.utils import SkipIfBeforePyTorchVersion, assert_allclose, test_script_save
 
 einops, has_einops = optional_import("einops")
 
@@ -32,20 +32,23 @@
         for num_heads in [4, 6, 8, 12]:
             for rel_pos_embedding in [None, RelPosEmbedding.DECOMPOSED]:
                 for input_size in [(16, 32), (8, 8, 8)]:
-                    for flash_attn in [True, False]:
-                        test_case = [
-                            {
-                                "hidden_size": hidden_size,
-                                "num_heads": num_heads,
-                                "dropout_rate": dropout_rate,
-                                "rel_pos_embedding": rel_pos_embedding if not flash_attn else None,
-                                "input_size": input_size,
-                                "use_flash_attention": flash_attn,
-                            },
-                            (2, 512, hidden_size),
-                            (2, 512, hidden_size),
-                        ]
-                        TEST_CASE_SABLOCK.append(test_case)
+                    for include_fc in [True, False]:
+                        for use_combined_linear in [True, False]:
+                            test_case = [
+                                {
+                                    "hidden_size": hidden_size,
+                                    "num_heads": num_heads,
+                                    "dropout_rate": dropout_rate,
+                                    "rel_pos_embedding": rel_pos_embedding,
+                                    "input_size": input_size,
+                                    "include_fc": include_fc,
+                                    "use_combined_linear": use_combined_linear,
+                                    "use_flash_attention": True if rel_pos_embedding is None else False,
+                                },
+                                (2, 512, hidden_size),
+                                (2, 512, hidden_size),
+                            ]
+                            TEST_CASE_SABLOCK.append(test_case)
 
 
 class TestResBlock(unittest.TestCase):
@@ -175,6 +178,39 @@ def count_sablock_params(*args, **kwargs):
         nparams_default_more_heads = count_sablock_params(hidden_size=hidden_size, num_heads=num_heads * 2)
         self.assertEqual(nparams_default, nparams_default_more_heads)
 
+    @parameterized.expand([[True, False], [True, True], [False, True], [False, False]])
+    @skipUnless(has_einops, "Requires einops")
+    @SkipIfBeforePyTorchVersion((2, 0))
+    def test_script(self, include_fc, use_combined_linear):
+        input_param = {
+            "hidden_size": 360,
+            "num_heads": 4,
+            "dropout_rate": 0.0,
+            "rel_pos_embedding": None,
+            "input_size": (16, 32),
+            "include_fc": include_fc,
+            "use_combined_linear": use_combined_linear,
+        }
+        net = SABlock(**input_param)
+        input_shape = (2, 512, 360)
+        test_data = torch.randn(input_shape)
+        test_script_save(net, test_data)
+
+    @skipUnless(has_einops, "Requires einops")
+    @SkipIfBeforePyTorchVersion((2, 0))
+    def test_flash_attention(self):
+        for causal in [True, False]:
+            input_param = {"hidden_size": 360, "num_heads": 4, "input_size": (16, 32), "causal": causal}
+        device = "cuda:0" if torch.cuda.is_available() else "cpu"
+        block_w_flash_attention = SABlock(**input_param, use_flash_attention=True).to(device)
+        block_wo_flash_attention = SABlock(**input_param, use_flash_attention=False).to(device)
+        block_wo_flash_attention.load_state_dict(block_w_flash_attention.state_dict())
+        test_data = torch.randn(2, 512, 360).to(device)
+
+        out_1 = block_w_flash_attention(test_data)
+        out_2 = block_wo_flash_attention(test_data)
+        assert_allclose(out_1, out_2, atol=1e-4)
+
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/tests/test_unetr.py b/tests/test_unetr.py
index 46018d2bc0..1217c9d85f 100644
--- a/tests/test_unetr.py
+++ b/tests/test_unetr.py
@@ -123,7 +123,7 @@ def test_ill_arg(self):
             )
 
     @parameterized.expand(TEST_CASE_UNETR)
-    @SkipIfBeforePyTorchVersion((1, 9))
+    @SkipIfBeforePyTorchVersion((2, 0))
     def test_script(self, input_param, input_shape, _):
         net = UNETR(**(input_param))
         net.eval()

From 6be7b13a901918071be1cf10aee8701d6e751484 Mon Sep 17 00:00:00 2001
From: "Kelvin R." <138339140+K-Rilla@users.noreply.github.com>
Date: Fri, 9 Aug 2024 02:29:47 -0700
Subject: [PATCH 116/183] Replaced package "pkg_resources" with "packaging"
 (#7953)

Fixes #7559 .

### Description
Replaced "pkg_resources" references with "packaging" in
MONAI/monai/utils/module.py & setup.py
Changes were made in functions "pytorch_after", "version_leq",
"version_geq".

### Types of changes
- Non-breaking change (fix or new feature that would not break existing
functionality).

---------

Signed-off-by: dedeepyasai <dedeepyasai.sai@gmail.com>
Signed-off-by: saelra <rasaelee@gmail.com>
Signed-off-by: Kelvin R <kelvinrbNsc@gmail.com>
Signed-off-by: ken-ni <kennett.vera@gmail.com>
Signed-off-by: Dureti <98233210+DuretiShemsi@users.noreply.github.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: saelra <rasaelee@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: dedeepyasai <dedeepyasai.sai@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Ratanachat Saelee <146144408+Saelra@users.noreply.github.com>
Co-authored-by: ken-ni <kennett.vera@gmail.com>
Co-authored-by: Dureti <98233210+DuretiShemsi@users.noreply.github.com>
---
 .github/workflows/pythonapp.yml | 2 +-
 docs/requirements.txt           | 1 +
 monai/utils/module.py           | 7 ++++---
 requirements-min.txt            | 1 +
 setup.cfg                       | 2 ++
 setup.py                        | 4 ++--
 6 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/.github/workflows/pythonapp.yml b/.github/workflows/pythonapp.yml
index 65f9a4dcf2..3c39166c1e 100644
--- a/.github/workflows/pythonapp.yml
+++ b/.github/workflows/pythonapp.yml
@@ -151,7 +151,7 @@ jobs:
     - name: Install dependencies
       run: |
         find  /opt/hostedtoolcache/* -maxdepth 0 ! -name 'Python' -exec rm -rf {} \;
-        python -m pip install --user --upgrade pip setuptools wheel twine
+        python -m pip install --user --upgrade pip setuptools wheel twine packaging
         # install the latest pytorch for testing
         # however, "pip install monai*.tar.gz" will build cpp/cuda with an isolated
         # fresh torch installation according to pyproject.toml
diff --git a/docs/requirements.txt b/docs/requirements.txt
index fe415a07b5..ff94f7b6de 100644
--- a/docs/requirements.txt
+++ b/docs/requirements.txt
@@ -41,3 +41,4 @@ onnxruntime; python_version <= '3.10'
 zarr
 huggingface_hub
 pyamg>=5.0.0
+packaging
diff --git a/monai/utils/module.py b/monai/utils/module.py
index 4d28f8d986..251232d62f 100644
--- a/monai/utils/module.py
+++ b/monai/utils/module.py
@@ -564,7 +564,7 @@ def version_leq(lhs: str, rhs: str) -> bool:
     """
 
     lhs, rhs = str(lhs), str(rhs)
-    pkging, has_ver = optional_import("pkg_resources", name="packaging")
+    pkging, has_ver = optional_import("packaging.Version")
     if has_ver:
         try:
             return cast(bool, pkging.version.Version(lhs) <= pkging.version.Version(rhs))
@@ -591,7 +591,8 @@ def version_geq(lhs: str, rhs: str) -> bool:
 
     """
     lhs, rhs = str(lhs), str(rhs)
-    pkging, has_ver = optional_import("pkg_resources", name="packaging")
+    pkging, has_ver = optional_import("packaging.Version")
+
     if has_ver:
         try:
             return cast(bool, pkging.version.Version(lhs) >= pkging.version.Version(rhs))
@@ -629,7 +630,7 @@ def pytorch_after(major: int, minor: int, patch: int = 0, current_ver_string: st
         if current_ver_string is None:
             _env_var = os.environ.get("PYTORCH_VER", "")
             current_ver_string = _env_var if _env_var else torch.__version__
-        ver, has_ver = optional_import("pkg_resources", name="parse_version")
+        ver, has_ver = optional_import("packaging.version", name="parse")
         if has_ver:
             return ver(".".join((f"{major}", f"{minor}", f"{patch}"))) <= ver(f"{current_ver_string}")  # type: ignore
         parts = f"{current_ver_string}".split("+", 1)[0].split(".", 3)
diff --git a/requirements-min.txt b/requirements-min.txt
index a091ef0568..21cf9d5e5c 100644
--- a/requirements-min.txt
+++ b/requirements-min.txt
@@ -4,3 +4,4 @@ setuptools>=50.3.0,<66.0.0,!=60.6.0 ; python_version < "3.12"
 setuptools>=70.2.0; python_version >= "3.12"
 coverage>=5.5
 parameterized
+packaging
diff --git a/setup.cfg b/setup.cfg
index 202e7b0e24..dfa94fcfa1 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -137,6 +137,8 @@ pyyaml =
     pyyaml
 fire =
     fire
+packaging =
+    packaging
 jsonschema =
     jsonschema
 pynrrd =
diff --git a/setup.py b/setup.py
index b90d9d0976..576743c1f7 100644
--- a/setup.py
+++ b/setup.py
@@ -17,7 +17,7 @@
 import sys
 import warnings
 
-import pkg_resources
+from packaging import version
 from setuptools import find_packages, setup
 
 import versioneer
@@ -40,7 +40,7 @@
 
     BUILD_CUDA = FORCE_CUDA or (torch.cuda.is_available() and (CUDA_HOME is not None))
 
-    _pt_version = pkg_resources.parse_version(torch.__version__).release
+    _pt_version = version.parse(torch.__version__).release
     if _pt_version is None or len(_pt_version) < 3:
         raise AssertionError("unknown torch version")
     TORCH_VERSION = int(_pt_version[0]) * 10000 + int(_pt_version[1]) * 100 + int(_pt_version[2])

From f8480027888e365d25a1b85429b200dca58b9f19 Mon Sep 17 00:00:00 2001
From: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Date: Fri, 9 Aug 2024 22:32:38 +0800
Subject: [PATCH 117/183] Add utils for vista3d (#7999)

This PR is a part of #7987

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [x] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Signed-off-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 docs/source/apps.rst                          |   8 -
 docs/source/transforms.rst                    |   3 +
 monai/transforms/__init__.py                  |   1 +
 monai/transforms/utils.py                     | 183 ++++++++++++++++++
 .../utils_morphological_ops.py}               |   2 +
 tests/min_tests.py                            |   1 +
 tests/test_morphological_ops.py               |   2 +-
 tests/test_vista3d_utils.py                   | 133 +++++++++++++
 8 files changed, 324 insertions(+), 9 deletions(-)
 rename monai/{apps/generation/maisi/utils/morphological_ops.py => transforms/utils_morphological_ops.py} (99%)
 create mode 100644 tests/test_vista3d_utils.py

diff --git a/docs/source/apps.rst b/docs/source/apps.rst
index c6ba8c0b9a..7fa7b9e9ff 100644
--- a/docs/source/apps.rst
+++ b/docs/source/apps.rst
@@ -261,11 +261,3 @@ FastMRIReader
 
 .. autoclass:: monai.apps.nnunet.nnUNetV2Runner
   :members:
-
-`Generative AI`
----------------
-
-`MAISI Utilities`
-~~~~~~~~~~~~~~~~~
-.. automodule:: monai.apps.generation.maisi.utils.morphological_ops
-    :members:
diff --git a/docs/source/transforms.rst b/docs/source/transforms.rst
index a359821679..637f0873f1 100644
--- a/docs/source/transforms.rst
+++ b/docs/source/transforms.rst
@@ -2310,6 +2310,9 @@ Utilities
 .. automodule:: monai.transforms.utils_pytorch_numpy_unification
     :members:
 
+.. automodule:: monai.transforms.utils_morphological_ops
+    :members:
+
 By Categories
 -------------
 .. toctree::
diff --git a/monai/transforms/__init__.py b/monai/transforms/__init__.py
index ef1da2d855..9548443768 100644
--- a/monai/transforms/__init__.py
+++ b/monai/transforms/__init__.py
@@ -688,6 +688,7 @@
     weighted_patch_samples,
     zero_margins,
 )
+from .utils_morphological_ops import dilate, erode
 from .utils_pytorch_numpy_unification import (
     allclose,
     any_np_pt,
diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index d8461d927b..e32bf6fc48 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -22,6 +22,7 @@
 
 import numpy as np
 import torch
+from torch import Tensor
 
 import monai
 from monai.config import DtypeLike, IndexSelection
@@ -30,6 +31,7 @@
 from monai.networks.utils import meshgrid_ij
 from monai.transforms.compose import Compose
 from monai.transforms.transform import MapTransform, Transform, apply_transform
+from monai.transforms.utils_morphological_ops import erode
 from monai.transforms.utils_pytorch_numpy_unification import (
     any_np_pt,
     ascontiguousarray,
@@ -65,6 +67,8 @@
     min_version,
     optional_import,
     pytorch_after,
+    unsqueeze_left,
+    unsqueeze_right,
 )
 from monai.utils.enums import TransformBackends
 from monai.utils.type_conversion import (
@@ -103,6 +107,8 @@
     "generate_spatial_bounding_box",
     "get_extreme_points",
     "get_largest_connected_component_mask",
+    "get_largest_connected_component_mask_point",
+    "convert_points_to_disc",
     "remove_small_objects",
     "img_bounds",
     "in_bounds",
@@ -1172,6 +1178,183 @@ def get_largest_connected_component_mask(
     return convert_to_dst_type(out, dst=img, dtype=out.dtype)[0]
 
 
+def get_largest_connected_component_mask_point(
+    img_pos: NdarrayTensor,
+    img_neg: NdarrayTensor,
+    point_coords: NdarrayTensor,
+    point_labels: NdarrayTensor,
+    pos_val: Sequence[int] = (1, 3),
+    neg_val: Sequence[int] = (0, 2),
+    margins: int = 3,
+) -> NdarrayTensor:
+    """
+    Gets the connected component of img_pos and img_neg that include the positive points and
+    negative points separately. The function is used for combining automatic results with interactive
+    results in VISTA3D.
+
+    Args:
+        img_pos: bool type tensor, shape [B, 1, H, W, D], where B means the foreground masks from a single 3D image.
+        img_neg: same format as img_pos but corresponds to negative points.
+        pos_val: positive point label values.
+        neg_val: negative point label values.
+        point_coords: the coordinates of each point, shape [B, N, 3], where N means the number of points.
+        point_labels: the label of each point, shape [B, N].
+    """
+
+    cucim_skimage, has_cucim = optional_import("cucim.skimage")
+
+    use_cp = has_cp and has_cucim and isinstance(img_pos, torch.Tensor) and img_pos.device != torch.device("cpu")
+    if use_cp:
+        img_pos_ = convert_to_cupy(img_pos.short())  # type: ignore
+        img_neg_ = convert_to_cupy(img_neg.short())  # type: ignore
+        label = cucim_skimage.measure.label
+        lib = cp
+    else:
+        if not has_measure:
+            raise RuntimeError("skimage.measure required.")
+        img_pos_, *_ = convert_data_type(img_pos, np.ndarray)
+        img_neg_, *_ = convert_data_type(img_neg, np.ndarray)
+        # for skimage.measure.label, the input must be bool type
+        if img_pos_.dtype != bool or img_neg_.dtype != bool:
+            raise ValueError("img_pos and img_neg must be bool type.")
+        label = measure.label
+        lib = np
+
+    features_pos, _ = label(img_pos_, connectivity=3, return_num=True)
+    features_neg, _ = label(img_neg_, connectivity=3, return_num=True)
+
+    outs = np.zeros_like(img_pos_)
+    for bs in range(point_coords.shape[0]):
+        for i, p in enumerate(point_coords[bs]):
+            if point_labels[bs, i] in pos_val:
+                features = features_pos
+            elif point_labels[bs, i] in neg_val:
+                features = features_neg
+            else:
+                # if -1 padding point, skip
+                continue
+            for margin in range(margins):
+                if isinstance(p, np.ndarray):
+                    x, y, z = np.round(p).astype(int).tolist()
+                else:
+                    x, y, z = p.float().round().int().tolist()
+                l, r = max(x - margin, 0), min(x + margin + 1, features.shape[-3])
+                t, d = max(y - margin, 0), min(y + margin + 1, features.shape[-2])
+                f, b = max(z - margin, 0), min(z + margin + 1, features.shape[-1])
+                if (features[bs, 0, l:r, t:d, f:b] > 0).any():
+                    index = features[bs, 0, l:r, t:d, f:b].max()
+                    outs[[bs]] += lib.isin(features[[bs]], index)
+                    break
+    outs[outs > 1] = 1
+    return convert_to_dst_type(outs, dst=img_pos, dtype=outs.dtype)[0]
+
+
+def convert_points_to_disc(
+    image_size: Sequence[int], point: Tensor, point_label: Tensor, radius: int = 2, disc: bool = False
+):
+    """
+    Convert a 3D point coordinates into image mask. The returned mask has the same spatial
+    size as `image_size` while the batch dimension is the same as 'point' batch dimension.
+    The point is converted to a mask ball with radius defined by `radius`. The output
+    contains two channels each for negative (first channel) and positive points.
+
+    Args:
+        image_size: The output size of the converted mask. It should be a 3D tuple.
+        point: [B, N, 3], 3D point coordinates.
+        point_label: [B, N], 0 or 2 means negative points, 1 or 3 means postive points.
+        radius: disc ball radius size.
+        disc: If true, use regular disc, other use gaussian.
+    """
+    masks = torch.zeros([point.shape[0], 2, image_size[0], image_size[1], image_size[2]], device=point.device)
+    _array = [
+        torch.arange(start=0, end=image_size[i], step=1, dtype=torch.float32, device=point.device) for i in range(3)
+    ]
+    coord_rows, coord_cols, coord_z = torch.meshgrid(_array[2], _array[1], _array[0])
+    # [1, 3, h, w, d] -> [b, 2, 3, h, w, d]
+    coords = unsqueeze_left(torch.stack((coord_rows, coord_cols, coord_z), dim=0), 6)
+    coords = coords.repeat(point.shape[0], 2, 1, 1, 1, 1)
+    for b, n in np.ndindex(*point.shape[:2]):
+        point_bn = unsqueeze_right(point[b, n], 6)
+        if point_label[b, n] > -1:
+            channel = 0 if (point_label[b, n] == 0 or point_label[b, n] == 2) else 1
+            pow_diff = torch.pow(coords[b, channel] - point_bn[b, n], 2)
+            if disc:
+                masks[b, channel] += pow_diff.sum(0) < radius**2
+            else:
+                masks[b, channel] += torch.exp(-pow_diff.sum(0) / (2 * radius**2))
+    return masks
+
+
+def sample_points_from_label(
+    labels: Tensor,
+    label_set: Sequence[int],
+    max_ppoint: int = 1,
+    max_npoint: int = 0,
+    device: torch.device | str | None = "cpu",
+    use_center: bool = False,
+):
+    """Sample points from labels.
+
+    Args:
+        labels: [1, 1, H, W, D]
+        label_set: local index, must match values in labels.
+        max_ppoint: maximum positive point samples.
+        max_npoint: maximum negative point samples.
+        device: returned tensor device.
+        use_center: whether to sample points from center.
+
+    Returns:
+        point: point coordinates of [B, N, 3]. B equals to the length of label_set.
+        point_label: [B, N], always 0 for negative, 1 for positive.
+    """
+    if not labels.shape[0] == 1:
+        raise ValueError("labels must have batch size 1.")
+
+    if device is None:
+        device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+    labels = labels[0, 0]
+    unique_labels = labels.unique().cpu().numpy().tolist()
+    _point = []
+    _point_label = []
+    for id in label_set:
+        if id in unique_labels:
+            plabels = labels == int(id)
+            nlabels = ~plabels
+            _plabels = get_largest_connected_component_mask(erode(plabels.unsqueeze(0).unsqueeze(0))[0, 0])
+            plabelpoints = torch.nonzero(_plabels).to(device)
+            if len(plabelpoints) == 0:
+                plabelpoints = torch.nonzero(plabels).to(device)
+            nlabelpoints = torch.nonzero(nlabels).to(device)
+            num_p = min(len(plabelpoints), max_ppoint)
+            num_n = min(len(nlabelpoints), max_npoint)
+            pad = max_ppoint + max_npoint - num_p - num_n
+            if use_center:
+                pmean = plabelpoints.float().mean(0)
+                pdis = ((plabelpoints - pmean) ** 2).sum(-1)
+                _, sorted_indices_tensor = torch.sort(pdis)
+                sorted_indices = sorted_indices_tensor.cpu().tolist()
+            else:
+                sorted_indices = list(range(len(plabelpoints)))
+                random.shuffle(sorted_indices)
+            _point.append(
+                torch.stack(
+                    [plabelpoints[sorted_indices[i]] for i in range(num_p)]
+                    + random.choices(nlabelpoints, k=num_n)
+                    + [torch.tensor([0, 0, 0], device=device)] * pad
+                )
+            )
+            _point_label.append(torch.tensor([1] * num_p + [0] * num_n + [-1] * pad).to(device))
+        else:
+            # pad the background labels
+            _point.append(torch.zeros(max_ppoint + max_npoint, 3).to(device))
+            _point_label.append(torch.zeros(max_ppoint + max_npoint).to(device) - 1)
+    point = torch.stack(_point)
+    point_label = torch.stack(_point_label)
+
+    return point, point_label
+
+
 def remove_small_objects(
     img: NdarrayTensor,
     min_size: int = 64,
diff --git a/monai/apps/generation/maisi/utils/morphological_ops.py b/monai/transforms/utils_morphological_ops.py
similarity index 99%
rename from monai/apps/generation/maisi/utils/morphological_ops.py
rename to monai/transforms/utils_morphological_ops.py
index 14786d60a2..b3134c1865 100644
--- a/monai/apps/generation/maisi/utils/morphological_ops.py
+++ b/monai/transforms/utils_morphological_ops.py
@@ -20,6 +20,8 @@
 from monai.config import NdarrayOrTensor
 from monai.utils import convert_data_type, convert_to_dst_type, ensure_tuple_rep
 
+__all__ = ["erode", "dilate"]
+
 
 def erode(mask: NdarrayOrTensor, filter_size: int | Sequence[int] = 3, pad_value: float = 1.0) -> NdarrayOrTensor:
     """
diff --git a/tests/min_tests.py b/tests/min_tests.py
index 3a143df84b..479c4c8dc2 100644
--- a/tests/min_tests.py
+++ b/tests/min_tests.py
@@ -209,6 +209,7 @@ def run_testsuit():
         "test_zarr_avg_merger",
         "test_perceptual_loss",
         "test_ultrasound_confidence_map_transform",
+        "test_vista3d_utils",
     ]
     assert sorted(exclude_cases) == sorted(set(exclude_cases)), f"Duplicated items in {exclude_cases}"
 
diff --git a/tests/test_morphological_ops.py b/tests/test_morphological_ops.py
index 6f29415759..422e8c4b9d 100644
--- a/tests/test_morphological_ops.py
+++ b/tests/test_morphological_ops.py
@@ -16,7 +16,7 @@
 import torch
 from parameterized import parameterized
 
-from monai.apps.generation.maisi.utils.morphological_ops import dilate, erode, get_morphological_filter_result_t
+from monai.transforms.utils_morphological_ops import dilate, erode, get_morphological_filter_result_t
 from tests.utils import TEST_NDARRAYS, assert_allclose
 
 TESTS_SHAPE = []
diff --git a/tests/test_vista3d_utils.py b/tests/test_vista3d_utils.py
new file mode 100644
index 0000000000..a940854d88
--- /dev/null
+++ b/tests/test_vista3d_utils.py
@@ -0,0 +1,133 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest.case import skipUnless
+
+import numpy as np
+import torch
+from parameterized import parameterized
+
+from monai.transforms.utils import (
+    convert_points_to_disc,
+    get_largest_connected_component_mask_point,
+    sample_points_from_label,
+)
+from monai.utils import min_version
+from monai.utils.module import optional_import
+from tests.utils import skip_if_no_cuda, skip_if_quick
+
+cp, has_cp = optional_import("cupy")
+cucim_skimage, has_cucim = optional_import("cucim.skimage")
+measure, has_measure = optional_import("skimage.measure", "0.14.2", min_version)
+
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+
+
+TESTS_SAMPLE_POINTS_FROM_LABEL = []
+for use_center in [True, False]:
+    labels = torch.zeros(1, 1, 32, 32, 32)
+    labels[0, 0, 5:10, 5:10, 5:10] = 1
+    labels[0, 0, 10:15, 10:15, 10:15] = 3
+    labels[0, 0, 20:25, 20:25, 20:25] = 5
+    TESTS_SAMPLE_POINTS_FROM_LABEL.append(
+        [{"labels": labels, "label_set": (1, 3, 5), "use_center": use_center}, (3, 1, 3), (3, 1)]
+    )
+
+TEST_CONVERT_POINTS_TO_DISC = []
+for radius in [1, 2]:
+    for disc in [True, False]:
+        image_size = (32, 32, 32)
+        point = torch.randn(3, 1, 3)
+        point_label = torch.randint(0, 4, (3, 1))
+        expected_shape = (point.shape[0], 2, *image_size)
+        TEST_CONVERT_POINTS_TO_DISC.append(
+            [
+                {"image_size": image_size, "point": point, "point_label": point_label, "radius": radius, "disc": disc},
+                expected_shape,
+            ]
+        )
+
+TEST_LCC_MASK_POINT_TORCH = []
+for bs in [1, 2]:
+    for num_points in [1, 3]:
+        shape = (bs, 1, 128, 32, 32)
+        TEST_LCC_MASK_POINT_TORCH.append(
+            [
+                {
+                    "img_pos": torch.randint(0, 2, shape, dtype=torch.bool),
+                    "img_neg": torch.randint(0, 2, shape, dtype=torch.bool),
+                    "point_coords": torch.randint(0, 10, (bs, num_points, 3)),
+                    "point_labels": torch.randint(0, 4, (bs, num_points)),
+                },
+                shape,
+            ]
+        )
+
+TEST_LCC_MASK_POINT_NP = []
+for bs in [1, 2]:
+    for num_points in [1, 3]:
+        shape = (bs, 1, 32, 32, 64)
+        TEST_LCC_MASK_POINT_NP.append(
+            [
+                {
+                    "img_pos": np.random.randint(0, 2, shape, dtype=bool),
+                    "img_neg": np.random.randint(0, 2, shape, dtype=bool),
+                    "point_coords": np.random.randint(0, 5, (bs, num_points, 3)),
+                    "point_labels": np.random.randint(0, 4, (bs, num_points)),
+                },
+                shape,
+            ]
+        )
+
+
+@skipUnless(has_measure or cucim_skimage, "skimage or cucim.skimage required")
+class TestSamplePointsFromLabel(unittest.TestCase):
+
+    @parameterized.expand(TESTS_SAMPLE_POINTS_FROM_LABEL)
+    def test_shape(self, input_data, expected_point_shape, expected_point_label_shape):
+        point, point_label = sample_points_from_label(**input_data)
+        self.assertEqual(point.shape, expected_point_shape)
+        self.assertEqual(point_label.shape, expected_point_label_shape)
+
+
+class TestConvertPointsToDisc(unittest.TestCase):
+
+    @parameterized.expand(TEST_CONVERT_POINTS_TO_DISC)
+    def test_shape(self, input_data, expected_shape):
+        result = convert_points_to_disc(**input_data)
+        self.assertEqual(result.shape, expected_shape)
+
+
+@skipUnless(has_measure or cucim_skimage, "skimage or cucim.skimage required")
+class TestGetLargestConnectedComponentMaskPoint(unittest.TestCase):
+
+    @skip_if_quick
+    @skip_if_no_cuda
+    @skipUnless(has_cp and cucim_skimage, "cupy and cucim.skimage required")
+    @parameterized.expand(TEST_LCC_MASK_POINT_TORCH)
+    def test_cp_shape(self, input_data, shape):
+        for key in input_data:
+            input_data[key] = input_data[key].to(device)
+        mask = get_largest_connected_component_mask_point(**input_data)
+        self.assertEqual(mask.shape, shape)
+
+    @skipUnless(has_measure, "skimage required")
+    @parameterized.expand(TEST_LCC_MASK_POINT_NP)
+    def test_np_shape(self, input_data, shape):
+        mask = get_largest_connected_component_mask_point(**input_data)
+        self.assertEqual(mask.shape, shape)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 62430315cd176478d06dd197b7a6dfdd6cd90c44 Mon Sep 17 00:00:00 2001
From: myron <myron@users.noreply.github.com>
Date: Sat, 10 Aug 2024 06:34:27 -0700
Subject: [PATCH 118/183] Adding a network CellSamWrapper (#7981)

Adding a network CellSamWrapper, a thin wrapper around SAM, which can be
used for 2D segmentation tasks.



### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: am <am>
Signed-off-by: myron <amyronenko@nvidia.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: am <am>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/installation.md             |  4 +-
 monai/networks/nets/cell_sam_wrapper.py | 92 +++++++++++++++++++++++++
 requirements-dev.txt                    |  1 +
 setup.cfg                               |  5 +-
 tests/test_cell_sam_wrapper.py          | 58 ++++++++++++++++
 5 files changed, 157 insertions(+), 3 deletions(-)
 create mode 100644 monai/networks/nets/cell_sam_wrapper.py
 create mode 100644 tests/test_cell_sam_wrapper.py

diff --git a/docs/source/installation.md b/docs/source/installation.md
index 4308a07647..70a8b6f1d4 100644
--- a/docs/source/installation.md
+++ b/docs/source/installation.md
@@ -254,10 +254,10 @@ Since MONAI v0.2.0, the extras syntax such as `pip install 'monai[nibabel]'` is
 - The options are
 
 ```
-[nibabel, skimage, scipy, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, cucim, openslide, pandas, einops, transformers, mlflow, clearml, matplotlib, tensorboardX, tifffile, imagecodecs, pyyaml, fire, jsonschema, ninja, pynrrd, pydicom, h5py, nni, optuna, onnx, onnxruntime, zarr, lpips, pynvml, huggingface_hub]
+[nibabel, skimage, scipy, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, cucim, openslide, pandas, einops, transformers, mlflow, clearml, matplotlib, tensorboardX, tifffile, imagecodecs, pyyaml, fire, jsonschema, ninja, pynrrd, pydicom, h5py, nni, optuna, onnx, onnxruntime, zarr, lpips, pynvml, huggingface_hub, segment-anything]
 ```
 
 which correspond to `nibabel`, `scikit-image`,`scipy`, `pillow`, `tensorboard`,
-`gdown`, `pytorch-ignite`, `torchvision`, `itk`, `tqdm`, `lmdb`, `psutil`, `cucim`, `openslide-python`, `pandas`, `einops`, `transformers`, `mlflow`, `clearml`, `matplotlib`, `tensorboardX`, `tifffile`, `imagecodecs`, `pyyaml`, `fire`, `jsonschema`, `ninja`, `pynrrd`, `pydicom`, `h5py`, `nni`, `optuna`, `onnx`, `onnxruntime`, `zarr`, `lpips`, `nvidia-ml-py`, `huggingface_hub` and `pyamg` respectively.
+`gdown`, `pytorch-ignite`, `torchvision`, `itk`, `tqdm`, `lmdb`, `psutil`, `cucim`, `openslide-python`, `pandas`, `einops`, `transformers`, `mlflow`, `clearml`, `matplotlib`, `tensorboardX`, `tifffile`, `imagecodecs`, `pyyaml`, `fire`, `jsonschema`, `ninja`, `pynrrd`, `pydicom`, `h5py`, `nni`, `optuna`, `onnx`, `onnxruntime`, `zarr`, `lpips`, `nvidia-ml-py`, `huggingface_hub`, `pyamg` and `segment-anything` respectively.
 
 - `pip install 'monai[all]'` installs all the optional dependencies.
diff --git a/monai/networks/nets/cell_sam_wrapper.py b/monai/networks/nets/cell_sam_wrapper.py
new file mode 100644
index 0000000000..308c3a6bcb
--- /dev/null
+++ b/monai/networks/nets/cell_sam_wrapper.py
@@ -0,0 +1,92 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import torch
+from torch import nn
+from torch.nn import functional as F
+
+from monai.utils import optional_import
+
+build_sam_vit_b, has_sam = optional_import("segment_anything.build_sam", name="build_sam_vit_b")
+
+_all__ = ["CellSamWrapper"]
+
+
+class CellSamWrapper(torch.nn.Module):
+    """
+    CellSamWrapper is thin wrapper around SAM model https://github.com/facebookresearch/segment-anything
+    with an image only decoder, that can be used for segmentation tasks.
+
+
+    Args:
+        auto_resize_inputs: whether to resize inputs before passing to the network.
+            (usually they need be resized, unless they are already at the expected size)
+        network_resize_roi: expected input size for the network.
+            (currently SAM expects 1024x1024)
+        checkpoint: checkpoint file to load the SAM weights from.
+            (this can be downloaded from SAM repo https://dl.fbaipublicfiles.com/segment_anything/sam_vit_b_01ec64.pth)
+        return_features: whether to return features from SAM encoder
+            (without using decoder/upsampling to the original input size)
+
+    """
+
+    def __init__(
+        self,
+        auto_resize_inputs=True,
+        network_resize_roi=(1024, 1024),
+        checkpoint="sam_vit_b_01ec64.pth",
+        return_features=False,
+        *args,
+        **kwargs,
+    ) -> None:
+        super().__init__(*args, **kwargs)
+
+        self.network_resize_roi = network_resize_roi
+        self.auto_resize_inputs = auto_resize_inputs
+        self.return_features = return_features
+
+        if not has_sam:
+            raise ValueError(
+                "SAM is not installed, please run: pip install git+https://github.com/facebookresearch/segment-anything.git"
+            )
+
+        model = build_sam_vit_b(checkpoint=checkpoint)
+
+        model.prompt_encoder = None
+        model.mask_decoder = None
+
+        model.mask_decoder = nn.Sequential(
+            nn.BatchNorm2d(num_features=256),
+            nn.ReLU(inplace=True),
+            nn.ConvTranspose2d(256, 128, kernel_size=3, stride=2, padding=1, output_padding=1, bias=False),
+            nn.BatchNorm2d(num_features=128),
+            nn.ReLU(inplace=True),
+            nn.ConvTranspose2d(128, 3, kernel_size=3, stride=2, padding=1, output_padding=1, bias=True),
+        )
+
+        self.model = model
+
+    def forward(self, x):
+        sh = x.shape[2:]
+
+        if self.auto_resize_inputs:
+            x = F.interpolate(x, size=self.network_resize_roi, mode="bilinear")
+
+        x = self.model.image_encoder(x)
+
+        if not self.return_features:
+            x = self.model.mask_decoder(x)
+            if self.auto_resize_inputs:
+                x = F.interpolate(x, size=sh, mode="bilinear")
+
+        return x
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 72ba210093..76f1952345 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -59,3 +59,4 @@ nvidia-ml-py
 huggingface_hub
 pyamg>=5.0.0
 git+https://github.com/Project-MONAI/GenerativeModels.git@7428fce193771e9564f29b91d29e523dd1b6b4cd
+git+https://github.com/facebookresearch/segment-anything.git@6fdee8f2727f4506cfbbe553e23b895e27956588
diff --git a/setup.cfg b/setup.cfg
index dfa94fcfa1..e240445e36 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -85,6 +85,7 @@ all =
     nvidia-ml-py
     huggingface_hub
     pyamg>=5.0.0
+    segment-anything
 nibabel =
     nibabel
 ninja =
@@ -162,11 +163,13 @@ pynvml =
     nvidia-ml-py
 # # workaround https://github.com/Project-MONAI/MONAI/issues/5882
 # MetricsReloaded =
-#     MetricsReloaded @ git+https://github.com/Project-MONAI/MetricsReloaded@monai-support#egg=MetricsReloaded
+    # MetricsReloaded @ git+https://github.com/Project-MONAI/MetricsReloaded@monai-support#egg=MetricsReloaded
 huggingface_hub =
     huggingface_hub
 pyamg =
     pyamg>=5.0.0
+segment-anything =
+    segment-anything @ git+https://github.com/facebookresearch/segment-anything@6fdee8f2727f4506cfbbe553e23b895e27956588#egg=segment-anything
 
 [flake8]
 select = B,C,E,F,N,P,T4,W,B9
diff --git a/tests/test_cell_sam_wrapper.py b/tests/test_cell_sam_wrapper.py
new file mode 100644
index 0000000000..2f1ee2b901
--- /dev/null
+++ b/tests/test_cell_sam_wrapper.py
@@ -0,0 +1,58 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.nets.cell_sam_wrapper import CellSamWrapper
+from monai.utils import optional_import
+
+build_sam_vit_b, has_sam = optional_import("segment_anything.build_sam", name="build_sam_vit_b")
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+TEST_CASE_CELLSEGWRAPPER = []
+for dims in [128, 256, 512, 1024]:
+    test_case = [
+        {"auto_resize_inputs": True, "network_resize_roi": [1024, 1024], "checkpoint": None},
+        (1, 3, *([dims] * 2)),
+        (1, 3, *([dims] * 2)),
+    ]
+    TEST_CASE_CELLSEGWRAPPER.append(test_case)
+
+
+@unittest.skipUnless(has_sam, "Requires SAM installation")
+class TestResNetDS(unittest.TestCase):
+
+    @parameterized.expand(TEST_CASE_CELLSEGWRAPPER)
+    def test_shape(self, input_param, input_shape, expected_shape):
+        net = CellSamWrapper(**input_param).to(device)
+        with eval_mode(net):
+            result = net(torch.randn(input_shape).to(device))
+            self.assertEqual(result.shape, expected_shape, msg=str(input_param))
+
+    def test_ill_arg0(self):
+        with self.assertRaises(RuntimeError):
+            net = CellSamWrapper(auto_resize_inputs=False, checkpoint=None).to(device)
+            net(torch.randn([1, 3, 256, 256]).to(device))
+
+    def test_ill_arg1(self):
+        with self.assertRaises(RuntimeError):
+            net = CellSamWrapper(network_resize_roi=[256, 256], checkpoint=None).to(device)
+            net(torch.randn([1, 3, 1024, 1024]).to(device))
+
+
+if __name__ == "__main__":
+    unittest.main()

From 250c18d71b39f414d4ef91e353d69ca8c2ce3f92 Mon Sep 17 00:00:00 2001
From: Pengfei Guo <32000655+guopengf@users.noreply.github.com>
Date: Sun, 11 Aug 2024 23:46:48 -0400
Subject: [PATCH 119/183] Refactor DiffusionModelUNetMaisi (#7989)

Fixes #7988 .

### Description

Refactor DiffusionModelUNetMaisi to use monai core components.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Pengfei Guo <pengfeig@nvidia.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 .../networks/diffusion_model_unet_maisi.py    | 36 +++++++++----------
 tests/test_diffusion_model_unet_maisi.py      |  7 +---
 2 files changed, 19 insertions(+), 24 deletions(-)

diff --git a/monai/apps/generation/maisi/networks/diffusion_model_unet_maisi.py b/monai/apps/generation/maisi/networks/diffusion_model_unet_maisi.py
index d5f5f6136b..e990b5fc98 100644
--- a/monai/apps/generation/maisi/networks/diffusion_model_unet_maisi.py
+++ b/monai/apps/generation/maisi/networks/diffusion_model_unet_maisi.py
@@ -37,21 +37,15 @@
 from torch import nn
 
 from monai.networks.blocks import Convolution
-from monai.utils import ensure_tuple_rep, optional_import
-from monai.utils.type_conversion import convert_to_tensor
-
-get_down_block, has_get_down_block = optional_import(
-    "generative.networks.nets.diffusion_model_unet", name="get_down_block"
-)
-get_mid_block, has_get_mid_block = optional_import(
-    "generative.networks.nets.diffusion_model_unet", name="get_mid_block"
-)
-get_timestep_embedding, has_get_timestep_embedding = optional_import(
-    "generative.networks.nets.diffusion_model_unet", name="get_timestep_embedding"
+from monai.networks.nets.diffusion_model_unet import (
+    get_down_block,
+    get_mid_block,
+    get_timestep_embedding,
+    get_up_block,
+    zero_module,
 )
-get_up_block, has_get_up_block = optional_import("generative.networks.nets.diffusion_model_unet", name="get_up_block")
-xformers, has_xformers = optional_import("xformers")
-zero_module, has_zero_module = optional_import("generative.networks.nets.diffusion_model_unet", name="zero_module")
+from monai.utils import ensure_tuple_rep
+from monai.utils.type_conversion import convert_to_tensor
 
 __all__ = ["DiffusionModelUNetMaisi"]
 
@@ -78,6 +72,8 @@ class DiffusionModelUNetMaisi(nn.Module):
         cross_attention_dim: Number of context dimensions to use.
         num_class_embeds: If specified (as an int), then this model will be class-conditional with `num_class_embeds` classes.
         upcast_attention: If True, upcast attention operations to full precision.
+        include_fc: whether to include the final linear layer. Default to False.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
         use_flash_attention: If True, use flash attention for a memory efficient attention mechanism.
         dropout_cattn: If different from zero, this will be the dropout value for the cross-attention layers.
         include_top_region_index_input: If True, use top region index input.
@@ -102,6 +98,8 @@ def __init__(
         cross_attention_dim: int | None = None,
         num_class_embeds: int | None = None,
         upcast_attention: bool = False,
+        include_fc: bool = False,
+        use_combined_linear: bool = False,
         use_flash_attention: bool = False,
         dropout_cattn: float = 0.0,
         include_top_region_index_input: bool = False,
@@ -152,9 +150,6 @@ def __init__(
                 "`num_channels`."
             )
 
-        if use_flash_attention and not has_xformers:
-            raise ValueError("use_flash_attention is True but xformers is not installed.")
-
         if use_flash_attention is True and not torch.cuda.is_available():
             raise ValueError(
                 "torch.cuda.is_available() should be True but is False. Flash attention is only available for GPU."
@@ -210,7 +205,6 @@ def __init__(
             input_channel = output_channel
             output_channel = num_channels[i]
             is_final_block = i == len(num_channels) - 1
-
             down_block = get_down_block(
                 spatial_dims=spatial_dims,
                 in_channels=input_channel,
@@ -227,6 +221,8 @@ def __init__(
                 transformer_num_layers=transformer_num_layers,
                 cross_attention_dim=cross_attention_dim,
                 upcast_attention=upcast_attention,
+                include_fc=include_fc,
+                use_combined_linear=use_combined_linear,
                 use_flash_attention=use_flash_attention,
                 dropout_cattn=dropout_cattn,
             )
@@ -245,6 +241,8 @@ def __init__(
             transformer_num_layers=transformer_num_layers,
             cross_attention_dim=cross_attention_dim,
             upcast_attention=upcast_attention,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
             use_flash_attention=use_flash_attention,
             dropout_cattn=dropout_cattn,
         )
@@ -280,6 +278,8 @@ def __init__(
                 transformer_num_layers=transformer_num_layers,
                 cross_attention_dim=cross_attention_dim,
                 upcast_attention=upcast_attention,
+                include_fc=include_fc,
+                use_combined_linear=use_combined_linear,
                 use_flash_attention=use_flash_attention,
                 dropout_cattn=dropout_cattn,
             )
diff --git a/tests/test_diffusion_model_unet_maisi.py b/tests/test_diffusion_model_unet_maisi.py
index 059a4a4ba8..f9384e6d82 100644
--- a/tests/test_diffusion_model_unet_maisi.py
+++ b/tests/test_diffusion_model_unet_maisi.py
@@ -17,14 +17,11 @@
 import torch
 from parameterized import parameterized
 
+from monai.apps.generation.maisi.networks.diffusion_model_unet_maisi import DiffusionModelUNetMaisi
 from monai.networks import eval_mode
 from monai.utils import optional_import
 
 _, has_einops = optional_import("einops")
-_, has_generative = optional_import("generative")
-
-if has_generative:
-    from monai.apps.generation.maisi.networks.diffusion_model_unet_maisi import DiffusionModelUNetMaisi
 
 UNCOND_CASES_2D = [
     [
@@ -291,7 +288,6 @@
 ]
 
 
-@skipUnless(has_generative, "monai-generative required")
 class TestDiffusionModelUNetMaisi2D(unittest.TestCase):
 
     @parameterized.expand(UNCOND_CASES_2D)
@@ -510,7 +506,6 @@ def test_shape_with_additional_inputs(self, input_param):
             self.assertEqual(result.shape, (1, 1, 16, 16))
 
 
-@skipUnless(has_generative, "monai-generative required")
 class TestDiffusionModelUNetMaisi3D(unittest.TestCase):
 
     @parameterized.expand(UNCOND_CASES_3D)

From 7a6f680642e4fba4ac6465237292f43f5755e869 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 12 Aug 2024 20:46:23 +0800
Subject: [PATCH 120/183] Remove segment-anything in setup.cfg (#8010)

Fixes #8009

### Description

Remove segment-anything in setup.cfg

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 .github/workflows/release.yml       |  4 ++--
 setup.cfg                           |  5 ++---
 tests/ngc_bundle_download.py        |  2 +-
 tests/test_handler_ignite_metric.py |  4 ++--
 tests/test_patchembedding.py        | 12 ++++++------
 tests/test_unetr.py                 | 12 ++++++------
 tests/test_vitautoenc.py            | 10 +++++-----
 7 files changed, 24 insertions(+), 25 deletions(-)

diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
index 60b610565e..a014a4ed1d 100644
--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -24,7 +24,7 @@ jobs:
         python-version: ${{ matrix.python-version }}
     - name: Install setuptools
       run: |
-        python -m pip install --user --upgrade setuptools wheel
+        python -m pip install --user --upgrade setuptools wheel packaging
     - name: Build and test source archive and wheel file
       run: |
         find  /opt/hostedtoolcache/* -maxdepth 0 ! -name 'Python' -exec rm -rf {} \;
@@ -104,7 +104,7 @@ jobs:
         run: |
           find  /opt/hostedtoolcache/* -maxdepth 0 ! -name 'Python' -exec rm -rf {} \;
           git describe
-          python -m pip install --user --upgrade setuptools wheel
+          python -m pip install --user --upgrade setuptools wheel packaging
           python setup.py build
           cat build/lib/monai/_version.py
       - name: Upload version
diff --git a/setup.cfg b/setup.cfg
index e240445e36..2115c30a7f 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -85,7 +85,6 @@ all =
     nvidia-ml-py
     huggingface_hub
     pyamg>=5.0.0
-    segment-anything
 nibabel =
     nibabel
 ninja =
@@ -168,8 +167,8 @@ huggingface_hub =
     huggingface_hub
 pyamg =
     pyamg>=5.0.0
-segment-anything =
-    segment-anything @ git+https://github.com/facebookresearch/segment-anything@6fdee8f2727f4506cfbbe553e23b895e27956588#egg=segment-anything
+# segment-anything =
+#     segment-anything @ git+https://github.com/facebookresearch/segment-anything@6fdee8f2727f4506cfbbe553e23b895e27956588#egg=segment-anything
 
 [flake8]
 select = B,C,E,F,N,P,T4,W,B9
diff --git a/tests/ngc_bundle_download.py b/tests/ngc_bundle_download.py
index 01dc044870..107114861c 100644
--- a/tests/ngc_bundle_download.py
+++ b/tests/ngc_bundle_download.py
@@ -127,7 +127,7 @@ def test_loading_mmar(self, item):
                 in_channels=1,
                 img_size=(96, 96, 96),
                 patch_size=(16, 16, 16),
-                pos_embed="conv",
+                proj_type="conv",
                 hidden_size=768,
                 mlp_dim=3072,
             )
diff --git a/tests/test_handler_ignite_metric.py b/tests/test_handler_ignite_metric.py
index 28e0b69621..3e42bda35d 100644
--- a/tests/test_handler_ignite_metric.py
+++ b/tests/test_handler_ignite_metric.py
@@ -16,7 +16,7 @@
 import torch
 from parameterized import parameterized
 
-from monai.handlers import IgniteMetric, IgniteMetricHandler, from_engine
+from monai.handlers import IgniteMetricHandler, from_engine
 from monai.losses import DiceLoss
 from monai.metrics import LossMetric
 from tests.utils import SkipIfNoModule, assert_allclose, optional_import
@@ -172,7 +172,7 @@ def _val_func(engine, batch):
     @parameterized.expand(TEST_CASES[0:2])
     def test_old_ignite_metric(self, input_param, input_data, expected_val):
         loss_fn = DiceLoss(**input_param)
-        ignite_metric = IgniteMetric(loss_fn=loss_fn, output_transform=from_engine(["pred", "label"]))
+        ignite_metric = IgniteMetricHandler(loss_fn=loss_fn, output_transform=from_engine(["pred", "label"]))
 
         def _val_func(engine, batch):
             pass
diff --git a/tests/test_patchembedding.py b/tests/test_patchembedding.py
index d059145033..71ac767966 100644
--- a/tests/test_patchembedding.py
+++ b/tests/test_patchembedding.py
@@ -43,7 +43,7 @@
                                             "patch_size": (patch_size,) * nd,
                                             "hidden_size": hidden_size,
                                             "num_heads": num_heads,
-                                            "pos_embed": proj_type,
+                                            "proj_type": proj_type,
                                             "pos_embed_type": pos_embed_type,
                                             "dropout_rate": dropout_rate,
                                         },
@@ -127,7 +127,7 @@ def test_ill_arg(self):
                 patch_size=(16, 16, 16),
                 hidden_size=128,
                 num_heads=12,
-                pos_embed="conv",
+                proj_type="conv",
                 pos_embed_type="sincos",
                 dropout_rate=5.0,
             )
@@ -139,7 +139,7 @@ def test_ill_arg(self):
                 patch_size=(64, 64, 64),
                 hidden_size=512,
                 num_heads=8,
-                pos_embed="perceptron",
+                proj_type="perceptron",
                 pos_embed_type="sincos",
                 dropout_rate=0.3,
             )
@@ -151,7 +151,7 @@ def test_ill_arg(self):
                 patch_size=(8, 8, 8),
                 hidden_size=512,
                 num_heads=14,
-                pos_embed="conv",
+                proj_type="conv",
                 dropout_rate=0.3,
             )
 
@@ -162,7 +162,7 @@ def test_ill_arg(self):
                 patch_size=(4, 4, 4),
                 hidden_size=768,
                 num_heads=8,
-                pos_embed="perceptron",
+                proj_type="perceptron",
                 dropout_rate=0.3,
             )
         with self.assertRaises(ValueError):
@@ -183,7 +183,7 @@ def test_ill_arg(self):
                 patch_size=(16, 16, 16),
                 hidden_size=768,
                 num_heads=12,
-                pos_embed="perc",
+                proj_type="perc",
                 dropout_rate=0.3,
             )
 
diff --git a/tests/test_unetr.py b/tests/test_unetr.py
index 1217c9d85f..8c5ecb32e1 100644
--- a/tests/test_unetr.py
+++ b/tests/test_unetr.py
@@ -30,7 +30,7 @@
                         for num_heads in [8]:
                             for mlp_dim in [3072]:
                                 for norm_name in ["instance"]:
-                                    for pos_embed in ["perceptron"]:
+                                    for proj_type in ["perceptron"]:
                                         for nd in (2, 3):
                                             test_case = [
                                                 {
@@ -42,7 +42,7 @@
                                                     "norm_name": norm_name,
                                                     "mlp_dim": mlp_dim,
                                                     "num_heads": num_heads,
-                                                    "pos_embed": pos_embed,
+                                                    "proj_type": proj_type,
                                                     "dropout_rate": dropout_rate,
                                                     "conv_block": True,
                                                     "res_block": False,
@@ -75,7 +75,7 @@ def test_ill_arg(self):
                 hidden_size=128,
                 mlp_dim=3072,
                 num_heads=12,
-                pos_embed="conv",
+                proj_type="conv",
                 norm_name="instance",
                 dropout_rate=5.0,
             )
@@ -89,7 +89,7 @@ def test_ill_arg(self):
                 hidden_size=512,
                 mlp_dim=3072,
                 num_heads=12,
-                pos_embed="conv",
+                proj_type="conv",
                 norm_name="instance",
                 dropout_rate=0.5,
             )
@@ -103,7 +103,7 @@ def test_ill_arg(self):
                 hidden_size=512,
                 mlp_dim=3072,
                 num_heads=14,
-                pos_embed="conv",
+                proj_type="conv",
                 norm_name="batch",
                 dropout_rate=0.4,
             )
@@ -117,7 +117,7 @@ def test_ill_arg(self):
                 hidden_size=768,
                 mlp_dim=3072,
                 num_heads=12,
-                pos_embed="perc",
+                proj_type="perc",
                 norm_name="instance",
                 dropout_rate=0.2,
             )
diff --git a/tests/test_vitautoenc.py b/tests/test_vitautoenc.py
index c68c583a0e..9a503948d0 100644
--- a/tests/test_vitautoenc.py
+++ b/tests/test_vitautoenc.py
@@ -23,7 +23,7 @@
 for in_channels in [1, 4]:
     for img_size in [64, 96, 128]:
         for patch_size in [16]:
-            for pos_embed in ["conv", "perceptron"]:
+            for proj_type in ["conv", "perceptron"]:
                 for nd in [2, 3]:
                     test_case = [
                         {
@@ -34,7 +34,7 @@
                             "mlp_dim": 3072,
                             "num_layers": 4,
                             "num_heads": 12,
-                            "pos_embed": pos_embed,
+                            "proj_type": proj_type,
                             "dropout_rate": 0.6,
                             "spatial_dims": nd,
                         },
@@ -54,7 +54,7 @@
             "mlp_dim": 3072,
             "num_layers": 4,
             "num_heads": 12,
-            "pos_embed": "conv",
+            "proj_type": "conv",
             "dropout_rate": 0.6,
             "spatial_dims": 3,
         },
@@ -93,7 +93,7 @@ def test_shape(self, input_param, input_shape, expected_shape):
         ]
     )
     def test_ill_arg(
-        self, in_channels, img_size, patch_size, hidden_size, mlp_dim, num_layers, num_heads, pos_embed, dropout_rate
+        self, in_channels, img_size, patch_size, hidden_size, mlp_dim, num_layers, num_heads, proj_type, dropout_rate
     ):
         with self.assertRaises(ValueError):
             ViTAutoEnc(
@@ -104,7 +104,7 @@ def test_ill_arg(
                 mlp_dim=mlp_dim,
                 num_layers=num_layers,
                 num_heads=num_heads,
-                pos_embed=pos_embed,
+                proj_type=proj_type,
                 dropout_rate=dropout_rate,
             )
 

From 68581146502b3f0c9c876b12902df3197b6aa98a Mon Sep 17 00:00:00 2001
From: Pengfei Guo <32000655+guopengf@users.noreply.github.com>
Date: Mon, 12 Aug 2024 23:54:48 -0400
Subject: [PATCH 121/183] Refactor AutoencoderKlMaisi (#7993)

Fixes #7988 .

### Description

Refactor AutoencoderKlMaisi to use monai core components.


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Pengfei Guo <pengfeig@nvidia.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Signed-off-by: Pengfei Guo <32000655+guopengf@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .../maisi/networks/autoencoderkl_maisi.py     | 68 ++++++++++++-------
 monai/networks/nets/autoencoderkl.py          |  4 +-
 tests/test_autoencoderkl_maisi.py             |  6 +-
 3 files changed, 46 insertions(+), 32 deletions(-)

diff --git a/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py b/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
index f27f73ec60..a52274b24a 100644
--- a/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
+++ b/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
@@ -13,25 +13,17 @@
 
 import gc
 import logging
-from typing import TYPE_CHECKING, Sequence, cast
+from typing import Sequence
 
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
 from monai.networks.blocks import Convolution
-from monai.utils import optional_import
+from monai.networks.blocks.spatialattention import SpatialAttentionBlock
+from monai.networks.nets.autoencoderkl import AEKLResBlock, AutoencoderKL
 from monai.utils.type_conversion import convert_to_tensor
 
-AttentionBlock, has_attentionblock = optional_import("generative.networks.nets.autoencoderkl", name="AttentionBlock")
-AutoencoderKL, has_autoencoderkl = optional_import("generative.networks.nets.autoencoderkl", name="AutoencoderKL")
-ResBlock, has_resblock = optional_import("generative.networks.nets.autoencoderkl", name="ResBlock")
-
-if TYPE_CHECKING:
-    from generative.networks.nets.autoencoderkl import AutoencoderKL as AutoencoderKLType
-else:
-    AutoencoderKLType = cast(type, AutoencoderKL)
-
 # Set up logging configuration
 logger = logging.getLogger(__name__)
 
@@ -518,11 +510,13 @@ class MaisiEncoder(nn.Module):
         in_channels: Number of input channels.
         num_channels: Sequence of block output channels.
         out_channels: Number of channels in the bottom layer (latent space) of the autoencoder.
-        num_res_blocks: Number of residual blocks (see ResBlock) per level.
+        num_res_blocks: Number of residual blocks (see AEKLResBlock) per level.
         norm_num_groups: Number of groups for the group norm layers.
         norm_eps: Epsilon for the normalization.
         attention_levels: Indicate which level from num_channels contain an attention block.
         with_nonlocal_attn: If True, use non-local attention block.
+        include_fc: whether to include the final linear layer in the attention block. Default to False.
+        use_combined_linear: whether to use a single linear layer for qkv projection in the attention block, default to False.
         use_flash_attention: If True, use flash attention for a memory efficient attention mechanism.
         num_splits: Number of splits for the input tensor.
         dim_split: Dimension of splitting for the input tensor.
@@ -547,6 +541,8 @@ def __init__(
         print_info: bool = False,
         save_mem: bool = True,
         with_nonlocal_attn: bool = True,
+        include_fc: bool = False,
+        use_combined_linear: bool = False,
         use_flash_attention: bool = False,
     ) -> None:
         super().__init__()
@@ -603,11 +599,13 @@ def __init__(
                 input_channel = output_channel
                 if attention_levels[i]:
                     blocks.append(
-                        AttentionBlock(
+                        SpatialAttentionBlock(
                             spatial_dims=spatial_dims,
                             num_channels=input_channel,
                             norm_num_groups=norm_num_groups,
                             norm_eps=norm_eps,
+                            include_fc=include_fc,
+                            use_combined_linear=use_combined_linear,
                             use_flash_attention=use_flash_attention,
                         )
                     )
@@ -626,7 +624,7 @@ def __init__(
 
         if with_nonlocal_attn:
             blocks.append(
-                ResBlock(
+                AEKLResBlock(
                     spatial_dims=spatial_dims,
                     in_channels=num_channels[-1],
                     norm_num_groups=norm_num_groups,
@@ -636,16 +634,18 @@ def __init__(
             )
 
             blocks.append(
-                AttentionBlock(
+                SpatialAttentionBlock(
                     spatial_dims=spatial_dims,
                     num_channels=num_channels[-1],
                     norm_num_groups=norm_num_groups,
                     norm_eps=norm_eps,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
                     use_flash_attention=use_flash_attention,
                 )
             )
             blocks.append(
-                ResBlock(
+                AEKLResBlock(
                     spatial_dims=spatial_dims,
                     in_channels=num_channels[-1],
                     norm_num_groups=norm_num_groups,
@@ -699,11 +699,13 @@ class MaisiDecoder(nn.Module):
         num_channels: Sequence of block output channels.
         in_channels: Number of channels in the bottom layer (latent space) of the autoencoder.
         out_channels: Number of output channels.
-        num_res_blocks: Number of residual blocks (see ResBlock) per level.
+        num_res_blocks: Number of residual blocks (see AEKLResBlock) per level.
         norm_num_groups: Number of groups for the group norm layers.
         norm_eps: Epsilon for the normalization.
         attention_levels: Indicate which level from num_channels contain an attention block.
         with_nonlocal_attn: If True, use non-local attention block.
+        include_fc: whether to include the final linear layer in the attention block. Default to False.
+        use_combined_linear: whether to use a single linear layer for qkv projection in the attention block, default to False.
         use_flash_attention: If True, use flash attention for a memory efficient attention mechanism.
         use_convtranspose: If True, use ConvTranspose to upsample feature maps in decoder.
         num_splits: Number of splits for the input tensor.
@@ -729,6 +731,8 @@ def __init__(
         print_info: bool = False,
         save_mem: bool = True,
         with_nonlocal_attn: bool = True,
+        include_fc: bool = False,
+        use_combined_linear: bool = False,
         use_flash_attention: bool = False,
         use_convtranspose: bool = False,
     ) -> None:
@@ -758,7 +762,7 @@ def __init__(
 
         if with_nonlocal_attn:
             blocks.append(
-                ResBlock(
+                AEKLResBlock(
                     spatial_dims=spatial_dims,
                     in_channels=reversed_block_out_channels[0],
                     norm_num_groups=norm_num_groups,
@@ -767,16 +771,18 @@ def __init__(
                 )
             )
             blocks.append(
-                AttentionBlock(
+                SpatialAttentionBlock(
                     spatial_dims=spatial_dims,
                     num_channels=reversed_block_out_channels[0],
                     norm_num_groups=norm_num_groups,
                     norm_eps=norm_eps,
+                    include_fc=include_fc,
+                    use_combined_linear=use_combined_linear,
                     use_flash_attention=use_flash_attention,
                 )
             )
             blocks.append(
-                ResBlock(
+                AEKLResBlock(
                     spatial_dims=spatial_dims,
                     in_channels=reversed_block_out_channels[0],
                     norm_num_groups=norm_num_groups,
@@ -812,11 +818,13 @@ def __init__(
 
                 if reversed_attention_levels[i]:
                     blocks.append(
-                        AttentionBlock(
+                        SpatialAttentionBlock(
                             spatial_dims=spatial_dims,
                             num_channels=block_in_ch,
                             norm_num_groups=norm_num_groups,
                             norm_eps=norm_eps,
+                            include_fc=include_fc,
+                            use_combined_linear=use_combined_linear,
                             use_flash_attention=use_flash_attention,
                         )
                     )
@@ -870,7 +878,7 @@ def forward(self, x: torch.Tensor) -> torch.Tensor:
         return x
 
 
-class AutoencoderKlMaisi(AutoencoderKLType):
+class AutoencoderKlMaisi(AutoencoderKL):
     """
     AutoencoderKL with custom MaisiEncoder and MaisiDecoder.
 
@@ -886,6 +894,8 @@ class AutoencoderKlMaisi(AutoencoderKLType):
         norm_eps: Epsilon for the normalization.
         with_encoder_nonlocal_attn: If True, use non-local attention block in the encoder.
         with_decoder_nonlocal_attn: If True, use non-local attention block in the decoder.
+        include_fc: whether to include the final linear layer. Default to False.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
         use_flash_attention: If True, use flash attention for a memory efficient attention mechanism.
         use_checkpointing: If True, use activation checkpointing.
         use_convtranspose: If True, use ConvTranspose to upsample feature maps in decoder.
@@ -909,6 +919,8 @@ def __init__(
         norm_eps: float = 1e-6,
         with_encoder_nonlocal_attn: bool = False,
         with_decoder_nonlocal_attn: bool = False,
+        include_fc: bool = False,
+        use_combined_linear: bool = False,
         use_flash_attention: bool = False,
         use_checkpointing: bool = False,
         use_convtranspose: bool = False,
@@ -930,12 +942,14 @@ def __init__(
             norm_eps,
             with_encoder_nonlocal_attn,
             with_decoder_nonlocal_attn,
-            use_flash_attention,
             use_checkpointing,
             use_convtranspose,
+            include_fc,
+            use_combined_linear,
+            use_flash_attention,
         )
 
-        self.encoder = MaisiEncoder(
+        self.encoder: nn.Module = MaisiEncoder(
             spatial_dims=spatial_dims,
             in_channels=in_channels,
             num_channels=num_channels,
@@ -945,6 +959,8 @@ def __init__(
             norm_eps=norm_eps,
             attention_levels=attention_levels,
             with_nonlocal_attn=with_encoder_nonlocal_attn,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
             use_flash_attention=use_flash_attention,
             num_splits=num_splits,
             dim_split=dim_split,
@@ -953,7 +969,7 @@ def __init__(
             save_mem=save_mem,
         )
 
-        self.decoder = MaisiDecoder(
+        self.decoder: nn.Module = MaisiDecoder(
             spatial_dims=spatial_dims,
             num_channels=num_channels,
             in_channels=latent_channels,
@@ -963,6 +979,8 @@ def __init__(
             norm_eps=norm_eps,
             attention_levels=attention_levels,
             with_nonlocal_attn=with_decoder_nonlocal_attn,
+            include_fc=include_fc,
+            use_combined_linear=use_combined_linear,
             use_flash_attention=use_flash_attention,
             use_convtranspose=use_convtranspose,
             num_splits=num_splits,
diff --git a/monai/networks/nets/autoencoderkl.py b/monai/networks/nets/autoencoderkl.py
index 836027796f..af191e748b 100644
--- a/monai/networks/nets/autoencoderkl.py
+++ b/monai/networks/nets/autoencoderkl.py
@@ -532,7 +532,7 @@ def __init__(
                 "`num_channels`."
             )
 
-        self.encoder = Encoder(
+        self.encoder: nn.Module = Encoder(
             spatial_dims=spatial_dims,
             in_channels=in_channels,
             channels=channels,
@@ -546,7 +546,7 @@ def __init__(
             use_combined_linear=use_combined_linear,
             use_flash_attention=use_flash_attention,
         )
-        self.decoder = Decoder(
+        self.decoder: nn.Module = Decoder(
             spatial_dims=spatial_dims,
             channels=channels,
             in_channels=latent_channels,
diff --git a/tests/test_autoencoderkl_maisi.py b/tests/test_autoencoderkl_maisi.py
index 392a3d7db2..0e9f427fb6 100644
--- a/tests/test_autoencoderkl_maisi.py
+++ b/tests/test_autoencoderkl_maisi.py
@@ -16,16 +16,13 @@
 import torch
 from parameterized import parameterized
 
+from monai.apps.generation.maisi.networks.autoencoderkl_maisi import AutoencoderKlMaisi
 from monai.networks import eval_mode
 from monai.utils import optional_import
 from tests.utils import SkipIfBeforePyTorchVersion
 
 tqdm, has_tqdm = optional_import("tqdm", name="tqdm")
 _, has_einops = optional_import("einops")
-_, has_generative = optional_import("generative")
-
-if has_generative:
-    from monai.apps.generation.maisi.networks.autoencoderkl_maisi import AutoencoderKlMaisi
 
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
@@ -79,7 +76,6 @@
     CASES = CASES_NO_ATTENTION
 
 
-@unittest.skipUnless(has_generative, "monai-generative required")
 class TestAutoencoderKlMaisi(unittest.TestCase):
 
     @parameterized.expand(CASES)

From 9dbfe160635312069dced0f6babad6f89d8dc8e7 Mon Sep 17 00:00:00 2001
From: Pengfei Guo <32000655+guopengf@users.noreply.github.com>
Date: Tue, 13 Aug 2024 01:30:40 -0400
Subject: [PATCH 122/183] Refactor ControlNetMaisi (#8005)

Fixes #7988 .

### Description

Refactor ControlNetMaisi to use monai core components.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Pengfei Guo <pengfeig@nvidia.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .../maisi/networks/controlnet_maisi.py        | 33 +++++++++----------
 monai/networks/nets/controlnet.py             | 10 +++---
 tests/test_controlnet_maisi.py                | 19 ++++++-----
 3 files changed, 29 insertions(+), 33 deletions(-)

diff --git a/monai/apps/generation/maisi/networks/controlnet_maisi.py b/monai/apps/generation/maisi/networks/controlnet_maisi.py
index 3641124b7d..269086d971 100644
--- a/monai/apps/generation/maisi/networks/controlnet_maisi.py
+++ b/monai/apps/generation/maisi/networks/controlnet_maisi.py
@@ -11,24 +11,15 @@
 
 from __future__ import annotations
 
-from typing import TYPE_CHECKING, Sequence, cast
+from typing import Sequence
 
 import torch
 
-from monai.utils import optional_import
+from monai.networks.nets.controlnet import ControlNet
+from monai.networks.nets.diffusion_model_unet import get_timestep_embedding
 
-ControlNet, has_controlnet = optional_import("generative.networks.nets.controlnet", name="ControlNet")
-get_timestep_embedding, has_get_timestep_embedding = optional_import(
-    "generative.networks.nets.diffusion_model_unet", name="get_timestep_embedding"
-)
 
-if TYPE_CHECKING:
-    from generative.networks.nets.controlnet import ControlNet as ControlNetType
-else:
-    ControlNetType = cast(type, ControlNet)
-
-
-class ControlNetMaisi(ControlNetType):
+class ControlNetMaisi(ControlNet):
     """
     Control network for diffusion models based on Zhang and Agrawala "Adding Conditional Control to Text-to-Image
     Diffusion Models" (https://arxiv.org/abs/2302.05543)
@@ -49,10 +40,12 @@ class ControlNetMaisi(ControlNetType):
         num_class_embeds: if specified (as an int), then this model will be class-conditional with `num_class_embeds`
             classes.
         upcast_attention: if True, upcast attention operations to full precision.
-        use_flash_attention: if True, use flash attention for a memory efficient attention mechanism.
         conditioning_embedding_in_channels: number of input channels for the conditioning embedding.
         conditioning_embedding_num_channels: number of channels for the blocks in the conditioning embedding.
         use_checkpointing: if True, use activation checkpointing to save memory.
+        include_fc: whether to include the final linear layer. Default to False.
+        use_combined_linear: whether to use a single linear layer for qkv projection, default to False.
+        use_flash_attention: if True, use flash attention for a memory efficient attention mechanism.
     """
 
     def __init__(
@@ -71,10 +64,12 @@ def __init__(
         cross_attention_dim: int | None = None,
         num_class_embeds: int | None = None,
         upcast_attention: bool = False,
-        use_flash_attention: bool = False,
         conditioning_embedding_in_channels: int = 1,
-        conditioning_embedding_num_channels: Sequence[int] | None = (16, 32, 96, 256),
+        conditioning_embedding_num_channels: Sequence[int] = (16, 32, 96, 256),
         use_checkpointing: bool = True,
+        include_fc: bool = False,
+        use_combined_linear: bool = False,
+        use_flash_attention: bool = False,
     ) -> None:
         super().__init__(
             spatial_dims,
@@ -91,9 +86,11 @@ def __init__(
             cross_attention_dim,
             num_class_embeds,
             upcast_attention,
-            use_flash_attention,
             conditioning_embedding_in_channels,
             conditioning_embedding_num_channels,
+            include_fc,
+            use_combined_linear,
+            use_flash_attention,
         )
         self.use_checkpointing = use_checkpointing
 
@@ -105,7 +102,7 @@ def forward(
         conditioning_scale: float = 1.0,
         context: torch.Tensor | None = None,
         class_labels: torch.Tensor | None = None,
-    ) -> tuple[Sequence[torch.Tensor], torch.Tensor]:
+    ) -> tuple[list[torch.Tensor], torch.Tensor]:
         emb = self._prepare_time_and_class_embedding(x, timesteps, class_labels)
         h = self._apply_initial_convolution(x)
         if self.use_checkpointing:
diff --git a/monai/networks/nets/controlnet.py b/monai/networks/nets/controlnet.py
index 8b08eaae10..8b8813597f 100644
--- a/monai/networks/nets/controlnet.py
+++ b/monai/networks/nets/controlnet.py
@@ -174,24 +174,22 @@ def __init__(
         super().__init__()
         if with_conditioning is True and cross_attention_dim is None:
             raise ValueError(
-                "DiffusionModelUNet expects dimension of the cross-attention conditioning (cross_attention_dim) "
+                "ControlNet expects dimension of the cross-attention conditioning (cross_attention_dim) "
                 "to be specified when with_conditioning=True."
             )
         if cross_attention_dim is not None and with_conditioning is False:
-            raise ValueError(
-                "DiffusionModelUNet expects with_conditioning=True when specifying the cross_attention_dim."
-            )
+            raise ValueError("ControlNet expects with_conditioning=True when specifying the cross_attention_dim.")
 
         # All number of channels should be multiple of num_groups
         if any((out_channel % norm_num_groups) != 0 for out_channel in channels):
             raise ValueError(
-                f"DiffusionModelUNet expects all channels to be a multiple of norm_num_groups, but got"
+                f"ControlNet expects all channels to be a multiple of norm_num_groups, but got"
                 f" channels={channels} and norm_num_groups={norm_num_groups}"
             )
 
         if len(channels) != len(attention_levels):
             raise ValueError(
-                f"DiffusionModelUNet expects channels to have the same length as attention_levels, but got "
+                f"ControlNet expects channels to have the same length as attention_levels, but got "
                 f"channels={channels} and attention_levels={attention_levels}"
             )
 
diff --git a/tests/test_controlnet_maisi.py b/tests/test_controlnet_maisi.py
index 7b0e69f2c8..bfdf25ec6e 100644
--- a/tests/test_controlnet_maisi.py
+++ b/tests/test_controlnet_maisi.py
@@ -17,14 +17,12 @@
 import torch
 from parameterized import parameterized
 
+from monai.apps.generation.maisi.networks.controlnet_maisi import ControlNetMaisi
 from monai.networks import eval_mode
 from monai.utils import optional_import
 from tests.utils import SkipIfBeforePyTorchVersion
 
-_, has_generative = optional_import("generative")
-
-if has_generative:
-    from monai.apps.generation.maisi.networks.controlnet_maisi import ControlNetMaisi
+_, has_einops = optional_import("einops")
 
 TEST_CASES = [
     [
@@ -103,8 +101,8 @@
 TEST_CASES_ERROR = [
     [
         {"spatial_dims": 2, "in_channels": 1, "with_conditioning": True, "cross_attention_dim": None},
-        "ControlNet expects dimension of the cross-attention conditioning "
-        "(cross_attention_dim) when using with_conditioning.",
+        "ControlNet expects dimension of the cross-attention conditioning (cross_attention_dim) "
+        "to be specified when with_conditioning=True.",
     ],
     [
         {"spatial_dims": 2, "in_channels": 1, "with_conditioning": False, "cross_attention_dim": 2},
@@ -112,7 +110,8 @@
     ],
     [
         {"spatial_dims": 2, "in_channels": 1, "num_channels": (8, 16), "norm_num_groups": 16},
-        "ControlNet expects all num_channels being multiple of norm_num_groups",
+        f"ControlNet expects all channels to be a multiple of norm_num_groups, but got"
+        f" channels={(8, 16)} and norm_num_groups={16}",
     ],
     [
         {
@@ -122,16 +121,17 @@
             "attention_levels": (True,),
             "norm_num_groups": 8,
         },
-        "ControlNet expects num_channels being same size of attention_levels",
+        f"ControlNet expects channels to have the same length as attention_levels, but got "
+        f"channels={(8, 16)} and attention_levels={(True,)}",
     ],
 ]
 
 
 @SkipIfBeforePyTorchVersion((2, 0))
-@skipUnless(has_generative, "monai-generative required")
 class TestControlNet(unittest.TestCase):
 
     @parameterized.expand(TEST_CASES)
+    @skipUnless(has_einops, "Requires einops")
     def test_shape_unconditioned_models(self, input_param, expected_num_down_blocks_residuals, expected_shape):
         net = ControlNetMaisi(**input_param)
         with eval_mode(net):
@@ -145,6 +145,7 @@ def test_shape_unconditioned_models(self, input_param, expected_num_down_blocks_
             self.assertEqual(result[1].shape, expected_shape)
 
     @parameterized.expand(TEST_CASES_CONDITIONAL)
+    @skipUnless(has_einops, "Requires einops")
     def test_shape_conditioned_models(self, input_param, expected_num_down_blocks_residuals, expected_shape):
         net = ControlNetMaisi(**input_param)
         with eval_mode(net):

From 34ce94db424445b38eb56a6c842e55a2122d4a9d Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 13 Aug 2024 14:44:53 +0800
Subject: [PATCH 123/183] Fix ci issue in test_vit (#8013)

Fixes #8012


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 requirements-dev.txt | 1 -
 tests/test_vit.py    | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/requirements-dev.txt b/requirements-dev.txt
index 76f1952345..9aad0804e6 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -58,5 +58,4 @@ lpips==0.1.4
 nvidia-ml-py
 huggingface_hub
 pyamg>=5.0.0
-git+https://github.com/Project-MONAI/GenerativeModels.git@7428fce193771e9564f29b91d29e523dd1b6b4cd
 git+https://github.com/facebookresearch/segment-anything.git@6fdee8f2727f4506cfbbe553e23b895e27956588
diff --git a/tests/test_vit.py b/tests/test_vit.py
index d638c0116a..a3ffd0b2ef 100644
--- a/tests/test_vit.py
+++ b/tests/test_vit.py
@@ -106,7 +106,7 @@ def test_ill_arg(
             )
 
     @parameterized.expand(TEST_CASE_Vit[:1])
-    @SkipIfBeforePyTorchVersion((1, 9))
+    @SkipIfBeforePyTorchVersion((2, 0))
     def test_script(self, input_param, input_shape, _):
         net = ViT(**(input_param))
         net.eval()

From 4877767cf92649a38ffda0fc590f2b92ba59f019 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 14 Aug 2024 09:55:53 +0800
Subject: [PATCH 124/183] Fix module can not import correctly issue (#8015)

Fixes #8014


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/utils/module.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/monai/utils/module.py b/monai/utils/module.py
index 251232d62f..1ac8140b39 100644
--- a/monai/utils/module.py
+++ b/monai/utils/module.py
@@ -13,7 +13,6 @@
 
 import enum
 import functools
-import importlib.util
 import os
 import pdb
 import re
@@ -209,11 +208,13 @@ def load_submodules(
     ):
         if (is_pkg or load_all) and name not in sys.modules and match(exclude_pattern, name) is None:
             try:
+                mod = import_module(name)
                 mod_spec = importer.find_spec(name)  # type: ignore
                 if mod_spec and mod_spec.loader:
-                    mod = importlib.util.module_from_spec(mod_spec)
-                    mod_spec.loader.exec_module(mod)
+                    loader = mod_spec.loader
+                    loader.exec_module(mod)
                     submodules.append(mod)
+
             except OptionalImportError:
                 pass  # could not import the optional deps., they are ignored
             except ImportError as e:

From e85580af2267404ff0f022e5372a44e8effe6316 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 15 Aug 2024 12:40:21 +0800
Subject: [PATCH 125/183] Fix 'torch.device' object has no attribute 'gpu_id'
 issue in trt export (#8019)

Part of #8017


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/utils.py | 6 +++---
 monai/utils/module.py   | 1 -
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/monai/networks/utils.py b/monai/networks/utils.py
index 6a97434215..f301c2dd5c 100644
--- a/monai/networks/utils.py
+++ b/monai/networks/utils.py
@@ -822,7 +822,7 @@ def _onnx_trt_compile(
     output_names = [] if not output_names else output_names
 
     # set up the TensorRT builder
-    torch_tensorrt.set_device(device)
+    torch.cuda.set_device(device)
     logger = trt.Logger(trt.Logger.WARNING)
     builder = trt.Builder(logger)
     network = builder.create_network(1 << int(trt.NetworkDefinitionCreationFlag.EXPLICIT_BATCH))
@@ -931,7 +931,7 @@ def convert_to_trt(
         warnings.warn(f"The dynamic batch range sequence should have 3 elements, but got {dynamic_batchsize} elements.")
 
     device = device if device else 0
-    target_device = torch.device(f"cuda:{device}") if device else torch.device("cuda:0")
+    target_device = torch.device(f"cuda:{device}")
     convert_precision = torch.float32 if precision == "fp32" else torch.half
     inputs = [torch.rand(ensure_tuple(input_shape)).to(target_device)]
 
@@ -986,7 +986,7 @@ def scale_batch_size(input_shape: Sequence[int], scale_num: int):
                     ir_model,
                     inputs=input_placeholder,
                     enabled_precisions=convert_precision,
-                    device=target_device,
+                    device=torch_tensorrt.Device(f"cuda:{device}"),
                     ir="torchscript",
                     **kwargs,
                 )
diff --git a/monai/utils/module.py b/monai/utils/module.py
index 1ac8140b39..78087aef84 100644
--- a/monai/utils/module.py
+++ b/monai/utils/module.py
@@ -214,7 +214,6 @@ def load_submodules(
                     loader = mod_spec.loader
                     loader.exec_module(mod)
                     submodules.append(mod)
-
             except OptionalImportError:
                 pass  # could not import the optional deps., they are ignored
             except ImportError as e:

From 77304dd114227b8b9b1059665aedba295db0ffc7 Mon Sep 17 00:00:00 2001
From: Yufan He <59374597+heyufan1995@users.noreply.github.com>
Date: Thu, 15 Aug 2024 01:50:33 -0500
Subject: [PATCH 126/183] Add vista network (#7987)

Fixes # .

### Description

Add VISTA3D model architecture to MONAI core

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: heyufan1995 <heyufan1995@gmail.com>
Signed-off-by: Yufan He <heyufan1995@gmail.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Co-authored-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/networks.rst            |  10 +
 monai/networks/nets/__init__.py     |   3 +-
 monai/networks/nets/segresnet_ds.py | 128 +++-
 monai/networks/nets/vista3d.py      | 908 ++++++++++++++++++++++++++++
 monai/transforms/utils.py           |   4 +-
 tests/test_segresnet_ds.py          |  86 ++-
 tests/test_vista3d.py               |  85 +++
 7 files changed, 1189 insertions(+), 35 deletions(-)
 create mode 100644 monai/networks/nets/vista3d.py
 create mode 100644 tests/test_vista3d.py

diff --git a/docs/source/networks.rst b/docs/source/networks.rst
index 249375dfc1..1810fec49b 100644
--- a/docs/source/networks.rst
+++ b/docs/source/networks.rst
@@ -481,6 +481,11 @@ Nets
 .. autoclass:: SegResNetDS
   :members:
 
+`SegResNetDS2`
+~~~~~~~~~~~~~~
+.. autoclass:: SegResNetDS2
+  :members:
+
 `SegResNetVAE`
 ~~~~~~~~~~~~~~
 .. autoclass:: SegResNetVAE
@@ -556,6 +561,11 @@ Nets
 .. autoclass:: UNETR
     :members:
 
+`VISTA3D`
+~~~~~~~~~
+.. autoclass:: VISTA3D
+    :members:
+
 `SwinUNETR`
 ~~~~~~~~~~~
 .. autoclass:: SwinUNETR
diff --git a/monai/networks/nets/__init__.py b/monai/networks/nets/__init__.py
index c777fe6442..0570c9fcc1 100644
--- a/monai/networks/nets/__init__.py
+++ b/monai/networks/nets/__init__.py
@@ -76,7 +76,7 @@
     resnet200,
 )
 from .segresnet import SegResNet, SegResNetVAE
-from .segresnet_ds import SegResNetDS
+from .segresnet_ds import SegResNetDS, SegResNetDS2
 from .senet import (
     SENet,
     SEnet,
@@ -118,6 +118,7 @@
 from .unet import UNet, Unet
 from .unetr import UNETR
 from .varautoencoder import VarAutoEncoder
+from .vista3d import VISTA3D, vista3d132
 from .vit import ViT
 from .vitautoenc import ViTAutoEnc
 from .vnet import VNet
diff --git a/monai/networks/nets/segresnet_ds.py b/monai/networks/nets/segresnet_ds.py
index 6430f5fdc9..1ac5a79ee3 100644
--- a/monai/networks/nets/segresnet_ds.py
+++ b/monai/networks/nets/segresnet_ds.py
@@ -11,6 +11,7 @@
 
 from __future__ import annotations
 
+import copy
 from collections.abc import Callable
 from typing import Union
 
@@ -23,7 +24,7 @@
 from monai.networks.layers.utils import get_act_layer, get_norm_layer
 from monai.utils import UpsampleMode, has_option
 
-__all__ = ["SegResNetDS"]
+__all__ = ["SegResNetDS", "SegResNetDS2"]
 
 
 def scales_for_resolution(resolution: tuple | list, n_stages: int | None = None):
@@ -425,3 +426,128 @@ def _forward(self, x: torch.Tensor) -> Union[None, torch.Tensor, list[torch.Tens
 
     def forward(self, x: torch.Tensor) -> Union[None, torch.Tensor, list[torch.Tensor]]:
         return self._forward(x)
+
+
+class SegResNetDS2(SegResNetDS):
+    """
+    SegResNetDS2 adds an additional decorder branch to SegResNetDS and is the image encoder of VISTA3D
+     <https://arxiv.org/abs/2406.05285>`_.
+
+    Args:
+        spatial_dims: spatial dimension of the input data. Defaults to 3.
+        init_filters: number of output channels for initial convolution layer. Defaults to 32.
+        in_channels: number of input channels for the network. Defaults to 1.
+        out_channels: number of output channels for the network. Defaults to 2.
+        act: activation type and arguments. Defaults to ``RELU``.
+        norm: feature normalization type and arguments. Defaults to ``BATCH``.
+        blocks_down: number of downsample blocks in each layer. Defaults to ``[1,2,2,4]``.
+        blocks_up: number of upsample blocks (optional).
+        dsdepth: number of levels for deep supervision. This will be the length of the list of outputs at each scale level.
+                 At dsdepth==1,only a single output is returned.
+        preprocess: optional callable function to apply before the model's forward pass
+        resolution: optional input image resolution. When provided, the network will first use non-isotropic kernels to bring
+                    image spacing into an approximately isotropic space.
+                    Otherwise, by default, the kernel size and downsampling is always isotropic.
+
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int = 3,
+        init_filters: int = 32,
+        in_channels: int = 1,
+        out_channels: int = 2,
+        act: tuple | str = "relu",
+        norm: tuple | str = "batch",
+        blocks_down: tuple = (1, 2, 2, 4),
+        blocks_up: tuple | None = None,
+        dsdepth: int = 1,
+        preprocess: nn.Module | Callable | None = None,
+        upsample_mode: UpsampleMode | str = "deconv",
+        resolution: tuple | None = None,
+    ):
+        super().__init__(
+            spatial_dims=spatial_dims,
+            init_filters=init_filters,
+            in_channels=in_channels,
+            out_channels=out_channels,
+            act=act,
+            norm=norm,
+            blocks_down=blocks_down,
+            blocks_up=blocks_up,
+            dsdepth=dsdepth,
+            preprocess=preprocess,
+            upsample_mode=upsample_mode,
+            resolution=resolution,
+        )
+
+        self.up_layers_auto = nn.ModuleList([copy.deepcopy(layer) for layer in self.up_layers])
+
+    def forward(  # type: ignore
+        self, x: torch.Tensor, with_point: bool = True, with_label: bool = True
+    ) -> tuple[Union[None, torch.Tensor, list[torch.Tensor]], Union[None, torch.Tensor, list[torch.Tensor]]]:
+        """
+        Args:
+            x: input tensor.
+            with_point: if true, return the point branch output.
+            with_label: if true, return the label branch output.
+        """
+        if self.preprocess is not None:
+            x = self.preprocess(x)
+
+        if not self.is_valid_shape(x):
+            raise ValueError(f"Input spatial dims {x.shape} must be divisible by {self.shape_factor()}")
+
+        x_down = self.encoder(x)
+
+        x_down.reverse()
+        x = x_down.pop(0)
+
+        if len(x_down) == 0:
+            x_down = [torch.zeros(1, device=x.device, dtype=x.dtype)]
+
+        outputs: list[torch.Tensor] = []
+        outputs_auto: list[torch.Tensor] = []
+        x_ = x.clone()
+        if with_point:
+            i = 0
+            for level in self.up_layers:
+                x = level["upsample"](x)
+                x = x + x_down[i]
+                x = level["blocks"](x)
+
+                if len(self.up_layers) - i <= self.dsdepth:
+                    outputs.append(level["head"](x))
+                i = i + 1
+
+            outputs.reverse()
+        x = x_
+        if with_label:
+            i = 0
+            for level in self.up_layers_auto:
+                x = level["upsample"](x)
+                x = x + x_down[i]
+                x = level["blocks"](x)
+
+                if len(self.up_layers) - i <= self.dsdepth:
+                    outputs_auto.append(level["head"](x))
+                i = i + 1
+
+            outputs_auto.reverse()
+
+        return outputs[0] if len(outputs) == 1 else outputs, outputs_auto[0] if len(outputs_auto) == 1 else outputs_auto
+
+    def set_auto_grad(self, auto_freeze=False, point_freeze=False):
+        """
+        Args:
+            auto_freeze: if true, freeze the image encoder and the auto-branch.
+            point_freeze: if true, freeze the image encoder and the point-branch.
+        """
+        for param in self.encoder.parameters():
+            param.requires_grad = (not auto_freeze) and (not point_freeze)
+
+        for param in self.up_layers_auto.parameters():
+            param.requires_grad = not auto_freeze
+
+        for param in self.up_layers.parameters():
+            param.requires_grad = not point_freeze
diff --git a/monai/networks/nets/vista3d.py b/monai/networks/nets/vista3d.py
new file mode 100644
index 0000000000..fe7f93d493
--- /dev/null
+++ b/monai/networks/nets/vista3d.py
@@ -0,0 +1,908 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import math
+from typing import Any, Callable, Optional, Sequence, Tuple
+
+import numpy as np
+import torch
+import torch.nn.functional as F
+from torch import nn
+
+import monai
+from monai.networks.blocks import MLPBlock, UnetrBasicBlock
+from monai.networks.nets import SegResNetDS2
+from monai.transforms.utils import convert_points_to_disc
+from monai.transforms.utils import get_largest_connected_component_mask_point as lcc
+from monai.transforms.utils import sample_points_from_label
+from monai.utils import optional_import, unsqueeze_left, unsqueeze_right
+
+rearrange, _ = optional_import("einops", name="rearrange")
+
+__all__ = ["VISTA3D", "vista3d132"]
+
+
+def vista3d132(encoder_embed_dim: int = 48, in_channels: int = 1):
+    """
+    Exact VISTA3D network configuration used in https://arxiv.org/abs/2406.05285>`_.
+    The model treats class index larger than 132 as zero-shot.
+
+    Args:
+        encoder_embed_dim: hidden dimension for encoder.
+        in_channels: input channel number.
+    """
+    segresnet = SegResNetDS2(
+        in_channels=in_channels,
+        blocks_down=(1, 2, 2, 4, 4),
+        norm="instance",
+        out_channels=encoder_embed_dim,
+        init_filters=encoder_embed_dim,
+        dsdepth=1,
+    )
+    point_head = PointMappingSAM(feature_size=encoder_embed_dim, n_classes=512, last_supported=132)
+    class_head = ClassMappingClassify(n_classes=512, feature_size=encoder_embed_dim, use_mlp=True)
+    vista = VISTA3D(image_encoder=segresnet, class_head=class_head, point_head=point_head)
+    return vista
+
+
+class VISTA3D(nn.Module):
+    """
+    VISTA3D based on:
+        `VISTA3D: Versatile Imaging SegmenTation and Annotation model for 3D Computed Tomography
+        <https://arxiv.org/abs/2406.05285>`_.
+
+    Args:
+        image_encoder: image encoder backbone for feature extraction.
+        class_head: class head used for class index based segmentation
+        point_head: point head used for interactive segmetnation
+    """
+
+    def __init__(self, image_encoder: nn.Module, class_head: nn.Module, point_head: nn.Module):
+        super().__init__()
+        self.image_encoder = image_encoder
+        self.class_head = class_head
+        self.point_head = point_head
+        self.image_embeddings = None
+        self.auto_freeze = False
+        self.point_freeze = False
+        self.NINF_VALUE = -9999
+        self.PINF_VALUE = 9999
+
+    def get_foreground_class_count(self, class_vector: torch.Tensor | None, point_coords: torch.Tensor | None) -> int:
+        """Get number of foreground classes based on class and point prompt."""
+        if class_vector is None:
+            if point_coords is None:
+                raise ValueError("class_vector and point_coords cannot be both None.")
+            return point_coords.shape[0]
+        else:
+            return class_vector.shape[0]
+
+    def convert_point_label(
+        self,
+        point_label: torch.Tensor,
+        label_set: Sequence[int] | None = None,
+        special_index: Sequence[int] = (23, 24, 25, 26, 27, 57, 128),
+    ):
+        """
+        Convert point label based on its class prompt. For special classes defined in special index,
+        the positive/negative point label will be converted from 1/0 to 3/2. The purpose is to separate those
+        classes with ambiguous classes.
+
+        Args:
+            point_label: the point label tensor, [B, N].
+            label_set: the label index matching the indexes in labels. If labels are mapped to global index using RelabelID,
+                this label_set should be global mapped index. If labels are not mapped to global index, e.g. in zero-shot
+                evaluation, this label_set should be the original index.
+            special_index: the special class index that needs to be converted.
+        """
+        if label_set is None:
+            return point_label
+        if not point_label.shape[0] == len(label_set):
+            raise ValueError("point_label and label_set must have the same length.")
+
+        for i in range(len(label_set)):
+            if label_set[i] in special_index:
+                for j in range(len(point_label[i])):
+                    point_label[i, j] = point_label[i, j] + 2 if point_label[i, j] > -1 else point_label[i, j]
+        return point_label
+
+    def sample_points_patch_val(
+        self,
+        labels: torch.Tensor,
+        patch_coords: Sequence[slice],
+        label_set: Sequence[int],
+        use_center: bool = True,
+        mapped_label_set: Sequence[int] | None = None,
+        max_ppoint: int = 1,
+        max_npoint: int = 0,
+    ):
+        """
+        Sample points for patch during sliding window validation. Only used for point only validation.
+
+        Args:
+            labels: shape [1, 1, H, W, D].
+            patch_coords: a sequence of sliding window slice objects.
+            label_set: local index, must match values in labels.
+            use_center: sample points from the center.
+            mapped_label_set: global index, it is used to identify special classes and is the global index
+                for the sampled points.
+            max_ppoint/max_npoint: positive points and negative points to sample.
+        """
+        point_coords, point_labels = sample_points_from_label(
+            labels[patch_coords],
+            label_set,
+            max_ppoint=max_ppoint,
+            max_npoint=max_npoint,
+            device=labels.device,
+            use_center=use_center,
+        )
+        point_labels = self.convert_point_label(point_labels, mapped_label_set)
+        return (point_coords, point_labels, torch.tensor(label_set).to(point_coords.device).unsqueeze(-1))
+
+    def update_point_to_patch(
+        self, patch_coords: Sequence[slice], point_coords: torch.Tensor, point_labels: torch.Tensor
+    ):
+        """
+        Update point_coords with respect to patch coords.
+        If point is outside of the patch, remove the coordinates and set label to -1.
+
+        Args:
+            patch_coords: a sequence of the python slice objects representing the patch coordinates during sliding window inference.
+                This value is passed from sliding_window_inferer.
+            point_coords: point coordinates, [B, N, 3].
+            point_labels: point labels, [B, N].
+        """
+        patch_ends = [patch_coords[-3].stop, patch_coords[-2].stop, patch_coords[-1].stop]
+        patch_starts = [patch_coords[-3].start, patch_coords[-2].start, patch_coords[-1].start]
+        # update point coords
+        patch_starts_tensor = unsqueeze_left(torch.tensor(patch_starts, device=point_coords.device), 2)
+        patch_ends_tensor = unsqueeze_left(torch.tensor(patch_ends, device=point_coords.device), 2)
+        # [1 N 1]
+        indices = torch.logical_and(
+            ((point_coords - patch_starts_tensor) > 0).all(2), ((patch_ends_tensor - point_coords) > 0).all(2)
+        )
+        # check if it's within patch coords
+        point_coords = point_coords.clone() - patch_starts_tensor
+        point_labels = point_labels.clone()
+        if indices.any():
+            point_labels[~indices] = -1
+            point_coords[~indices] = 0
+            # also remove padded points, mainly used for inference.
+            not_pad_indices = (point_labels != -1).any(0)
+            point_coords = point_coords[:, not_pad_indices]
+            point_labels = point_labels[:, not_pad_indices]
+            return point_coords, point_labels
+        return None, None
+
+    def connected_components_combine(
+        self,
+        logits: torch.Tensor,
+        point_logits: torch.Tensor,
+        point_coords: torch.Tensor,
+        point_labels: torch.Tensor,
+        mapping_index: torch.Tensor,
+        thred: float = 0.5,
+    ):
+        """
+        Combine auto results with point click response. The auto results have shape [B, 1, H, W, D] which means B foreground masks
+        from a single image patch.
+        Out of those B foreground masks, user may add points to a subset of B1 foreground masks for editing.
+        mapping_index represents the correspondence between B and B1.
+        For mapping_index with point clicks, NaN values in logits will be replaced with point_logits. Meanwhile, the added/removed
+        region in point clicks must be updated by the lcc function.
+        Notice, if a positive point is within logits/prev_mask, the components containing the positive point will be added.
+
+        Args:
+            logits: automatic branch results, [B, 1, H, W, D].
+            point_logits: point branch results, [B1, 1, H, W, D].
+            point_coords: point coordinates, [B1, N, 3].
+            point_labels: point labels, [B1, N].
+            mapping_index: [B].
+            thred: the threshold to convert logits to binary.
+        """
+        logits = logits.as_tensor() if isinstance(logits, monai.data.MetaTensor) else logits
+        _logits = logits[mapping_index]
+        inside = []
+        for i in range(_logits.shape[0]):
+            inside.append(
+                np.any(
+                    [
+                        _logits[i, 0, p[0], p[1], p[2]].item() > 0
+                        for p in point_coords[i].cpu().numpy().round().astype(int)
+                    ]
+                )
+            )
+        inside_tensor = torch.tensor(inside).to(logits.device)
+        nan_mask = torch.isnan(_logits)
+        # _logits are converted to binary [B1, 1, H, W, D]
+        _logits = torch.nan_to_num(_logits, nan=self.NINF_VALUE).sigmoid()
+        pos_region = point_logits.sigmoid() > thred
+        diff_pos = torch.logical_and(torch.logical_or(_logits <= thred, unsqueeze_right(inside_tensor, 5)), pos_region)
+        diff_neg = torch.logical_and((_logits > thred), ~pos_region)
+        cc = lcc(diff_pos, diff_neg, point_coords=point_coords, point_labels=point_labels)
+        # cc is the region that can be updated by point_logits.
+        cc = cc.to(logits.device)
+        # Need to replace NaN with point_logits. diff_neg will never lie in nan_mask,
+        # only remove unconnected positive region.
+        uc_pos_region = torch.logical_and(pos_region, ~cc)
+        fill_mask = torch.logical_and(nan_mask, uc_pos_region)
+        if fill_mask.any():
+            # fill in the mean negative value
+            point_logits[fill_mask] = -1
+        # replace logits nan value and cc with point_logits
+        cc = torch.logical_or(nan_mask, cc).to(logits.dtype)
+        logits[mapping_index] *= 1 - cc
+        logits[mapping_index] += cc * point_logits
+        return logits
+
+    def gaussian_combine(
+        self,
+        logits: torch.Tensor,
+        point_logits: torch.Tensor,
+        point_coords: torch.Tensor,
+        point_labels: torch.Tensor,
+        mapping_index: torch.Tensor,
+        radius: int | None = None,
+    ):
+        """
+        Combine point results with auto results using gaussian.
+
+        Args:
+            logits: automatic branch results, [B, 1, H, W, D].
+            point_logits: point branch results, [B1, 1, H, W, D].
+            point_coords: point coordinates, [B1, N, 3].
+            point_labels: point labels, [B1, N].
+            mapping_index: [B].
+            radius: gaussian ball radius.
+        """
+        if radius is None:
+            radius = min(point_logits.shape[-3:]) // 5  # empirical value 5
+        weight = 1 - convert_points_to_disc(point_logits.shape[-3:], point_coords, point_labels, radius=radius).sum(
+            1, keepdims=True
+        )
+        weight[weight < 0] = 0
+        logits = logits.as_tensor() if isinstance(logits, monai.data.MetaTensor) else logits
+        logits[mapping_index] *= weight
+        logits[mapping_index] += (1 - weight) * point_logits
+        return logits
+
+    def set_auto_grad(self, auto_freeze: bool = False, point_freeze: bool = False):
+        """
+        Freeze auto-branch or point-branch.
+
+        Args:
+            auto_freeze: whether to freeze the auto branch.
+            point_freeze: whether to freeze the point branch.
+        """
+        if auto_freeze != self.auto_freeze:
+            if hasattr(self.image_encoder, "set_auto_grad"):
+                self.image_encoder.set_auto_grad(auto_freeze=auto_freeze, point_freeze=point_freeze)
+            else:
+                for param in self.image_encoder.parameters():
+                    param.requires_grad = (not auto_freeze) and (not point_freeze)
+            for param in self.class_head.parameters():
+                param.requires_grad = not auto_freeze
+            self.auto_freeze = auto_freeze
+
+        if point_freeze != self.point_freeze:
+            if hasattr(self.image_encoder, "set_auto_grad"):
+                self.image_encoder.set_auto_grad(auto_freeze=auto_freeze, point_freeze=point_freeze)
+            else:
+                for param in self.image_encoder.parameters():
+                    param.requires_grad = (not auto_freeze) and (not point_freeze)
+            for param in self.point_head.parameters():
+                param.requires_grad = not point_freeze
+            self.point_freeze = point_freeze
+
+    def forward(
+        self,
+        input_images: torch.Tensor,
+        point_coords: torch.Tensor | None = None,
+        point_labels: torch.Tensor | None = None,
+        class_vector: torch.Tensor | None = None,
+        prompt_class: torch.Tensor | None = None,
+        patch_coords: Sequence[slice] | None = None,
+        labels: torch.Tensor | None = None,
+        label_set: Sequence[int] | None = None,
+        prev_mask: torch.Tensor | None = None,
+        radius: int | None = None,
+        val_point_sampler: Callable | None = None,
+        **kwargs,
+    ):
+        """
+        The forward function for VISTA3D. We only support single patch in training and inference.
+        One exception is allowing sliding window batch size > 1 for automatic segmentation only case.
+        B represents number of objects, N represents number of points for each objects.
+
+        Args:
+            input_images: [1, 1, H, W, D]
+            point_coords: [B, N, 3]
+            point_labels: [B, N], -1 represents padding. 0/1 means negative/positive points for regular class.
+                2/3 means negative/postive ponits for special supported class like tumor.
+            class_vector: [B, 1], the global class index
+            prompt_class: [B, 1], the global class index. This value is associated with point_coords to identify if
+                the points are for zero-shot or supported class. When class_vector and point_coords are both
+                provided, prompt_class is the same as class_vector. For prompt_class[b] > 512, point_coords[b]
+                will be considered novel class.
+            patch_coords: a sequence of the python slice objects representing the patch coordinates during sliding window inference.
+                This value is passed from sliding_window_inferer. This is an indicator for training phase or validation phase.
+            labels: [1, 1, H, W, D], the groundtruth label tensor, only used for point-only evaluation
+            label_set: the label index matching the indexes in labels. If labels are mapped to global index using RelabelID,
+                this label_set should be global mapped index. If labels are not mapped to global index, e.g. in zero-shot
+                evaluation, this label_set should be the original index.
+            prev_mask: [B, N, H_fullsize, W_fullsize, D_fullsize].
+                This is the transposed raw output from sliding_window_inferer before any postprocessing.
+                When user click points to perform auto-results correction, this can be the auto-results.
+            radius: single float value controling the gaussian blur when combining point and auto results.
+                The gaussian combine is not used in VISTA3D training but might be useful for finetuning purposes.
+            val_point_sampler: function used to sample points from labels. This is only used for point-only evaluation.
+
+        """
+        image_size = input_images.shape[-3:]
+        device = input_images.device
+        if point_coords is None and class_vector is None:
+            return self.NINF_VALUE + torch.zeros([1, 1, *image_size], device=device)
+
+        bs = self.get_foreground_class_count(class_vector, point_coords)
+        if patch_coords is not None:
+            # if during validation and perform enable based point-validation.
+            if labels is not None and label_set is not None:
+                # if labels is not None, sample from labels for each patch.
+                if val_point_sampler is None:
+                    # TODO: think about how to refactor this part.
+                    val_point_sampler = self.sample_points_patch_val
+                point_coords, point_labels, prompt_class = val_point_sampler(labels, patch_coords, label_set)
+                if prompt_class[0].item() == 0:  # type: ignore
+                    point_labels[0] = -1  # type: ignore
+                labels, prev_mask = None, None
+            elif point_coords is not None:
+                # If not performing patch-based point only validation, use user provided click points for inference.
+                # the point clicks is in original image space, convert it to current patch-coordinate space.
+                point_coords, point_labels = self.update_point_to_patch(patch_coords, point_coords, point_labels)  # type: ignore
+
+        if point_coords is not None and point_labels is not None:
+            # remove points that used for padding purposes (point_label = -1)
+            mapping_index = ((point_labels != -1).sum(1) > 0).to(torch.bool)
+            if mapping_index.any():
+                point_coords = point_coords[mapping_index]
+                point_labels = point_labels[mapping_index]
+                if prompt_class is not None:
+                    prompt_class = prompt_class[mapping_index]
+            else:
+                if self.auto_freeze or (class_vector is None and patch_coords is None):
+                    # if auto_freeze, point prompt must exist to allow loss backward
+                    # in training, class_vector and point cannot both be None due to loss.backward()
+                    mapping_index.fill_(True)
+                else:
+                    point_coords, point_labels = None, None
+
+        if point_coords is None and class_vector is None:
+            return self.NINF_VALUE + torch.zeros([bs, 1, *image_size], device=device)
+
+        if self.image_embeddings is not None and kwargs.get("keep_cache", False) and class_vector is None:
+            out, out_auto = self.image_embeddings, None
+        else:
+            out, out_auto = self.image_encoder(
+                input_images, with_point=point_coords is not None, with_label=class_vector is not None
+            )
+        # release memory
+        input_images = None  # type: ignore
+
+        # force releasing memories that set to None
+        torch.cuda.empty_cache()
+        if class_vector is not None:
+            logits, _ = self.class_head(out_auto, class_vector)
+            if point_coords is not None:
+                point_logits = self.point_head(out, point_coords, point_labels, class_vector=prompt_class)
+                if patch_coords is None:
+                    logits = self.gaussian_combine(
+                        logits, point_logits, point_coords, point_labels, mapping_index, radius  # type: ignore
+                    )
+                else:
+                    # during validation use largest component
+                    logits = self.connected_components_combine(
+                        logits, point_logits, point_coords, point_labels, mapping_index  # type: ignore
+                    )
+        else:
+            logits = self.NINF_VALUE + torch.zeros([bs, 1, *image_size], device=device, dtype=out.dtype)
+            logits[mapping_index] = self.point_head(out, point_coords, point_labels, class_vector=prompt_class)
+            if prev_mask is not None and patch_coords is not None:
+                logits = self.connected_components_combine(
+                    prev_mask[patch_coords].transpose(1, 0).to(logits.device),
+                    logits[mapping_index],
+                    point_coords,  # type: ignore
+                    point_labels,  # type: ignore
+                    mapping_index,
+                )
+
+        if kwargs.get("keep_cache", False) and class_vector is None:
+            self.image_embeddings = out.detach()
+        return logits
+
+
+class PointMappingSAM(nn.Module):
+    def __init__(self, feature_size: int, max_prompt: int = 32, n_classes: int = 512, last_supported: int = 132):
+        """Interactive point head used for VISTA3D.
+        Adapted from segment anything:
+        `https://github.com/facebookresearch/segment-anything/blob/main/segment_anything/modeling/mask_decoder.py`.
+
+        Args:
+            feature_size: feature channel from encoder.
+            max_prompt: max prompt number in each forward iteration.
+            n_classes: number of classes the model can potentially support. This is the maximum number of class embeddings.
+            last_supported: number of classes the model support, this value should match the trained model weights.
+        """
+        super().__init__()
+        transformer_dim = feature_size
+        self.max_prompt = max_prompt
+        self.feat_downsample = nn.Sequential(
+            nn.Conv3d(in_channels=feature_size, out_channels=feature_size, kernel_size=3, stride=2, padding=1),
+            nn.InstanceNorm3d(feature_size),
+            nn.GELU(),
+            nn.Conv3d(in_channels=feature_size, out_channels=transformer_dim, kernel_size=3, stride=1, padding=1),
+            nn.InstanceNorm3d(feature_size),
+        )
+
+        self.mask_downsample = nn.Conv3d(in_channels=2, out_channels=2, kernel_size=3, stride=2, padding=1)
+
+        self.transformer = TwoWayTransformer(depth=2, embedding_dim=transformer_dim, mlp_dim=512, num_heads=4)
+        self.pe_layer = PositionEmbeddingRandom(transformer_dim // 2)
+        self.point_embeddings = nn.ModuleList([nn.Embedding(1, transformer_dim), nn.Embedding(1, transformer_dim)])
+        self.not_a_point_embed = nn.Embedding(1, transformer_dim)
+        self.special_class_embed = nn.Embedding(1, transformer_dim)
+        self.mask_tokens = nn.Embedding(1, transformer_dim)
+
+        self.output_upscaling = nn.Sequential(
+            nn.ConvTranspose3d(transformer_dim, transformer_dim, kernel_size=3, stride=2, padding=1, output_padding=1),
+            nn.InstanceNorm3d(transformer_dim),
+            nn.GELU(),
+            nn.Conv3d(transformer_dim, transformer_dim, kernel_size=3, stride=1, padding=1),
+        )
+
+        self.output_hypernetworks_mlps = MLP(transformer_dim, transformer_dim, transformer_dim, 3)
+        # class embedding
+        self.n_classes = n_classes
+        self.last_supported = last_supported
+        self.class_embeddings = nn.Embedding(n_classes, feature_size)
+        self.zeroshot_embed = nn.Embedding(1, transformer_dim)
+        self.supported_embed = nn.Embedding(1, transformer_dim)
+
+    def forward(
+        self,
+        out: torch.Tensor,
+        point_coords: torch.Tensor,
+        point_labels: torch.Tensor,
+        class_vector: torch.Tensor | None = None,
+    ):
+        """Args:
+        out: feature from encoder, [1, C, H, W, C]
+        point_coords: point coordinates, [B, N, 3]
+        point_labels: point labels, [B, N]
+        class_vector: class prompts, [B]
+        """
+        # downsample out
+        out_low = self.feat_downsample(out)
+        out_shape = tuple(out.shape[-3:])
+        # release memory
+        out = None  # type: ignore
+        torch.cuda.empty_cache()
+        # embed points
+        points = point_coords + 0.5  # Shift to center of pixel
+        point_embedding = self.pe_layer.forward_with_coords(points, out_shape)  # type: ignore
+        point_embedding[point_labels == -1] = 0.0
+        point_embedding[point_labels == -1] += self.not_a_point_embed.weight
+        point_embedding[point_labels == 0] += self.point_embeddings[0].weight
+        point_embedding[point_labels == 1] += self.point_embeddings[1].weight
+        point_embedding[point_labels == 2] += self.point_embeddings[0].weight + self.special_class_embed.weight
+        point_embedding[point_labels == 3] += self.point_embeddings[1].weight + self.special_class_embed.weight
+        output_tokens = self.mask_tokens.weight
+
+        output_tokens = output_tokens.unsqueeze(0).expand(point_embedding.size(0), -1, -1)
+        if class_vector is None:
+            tokens_all = torch.cat(
+                (
+                    output_tokens,
+                    point_embedding,
+                    self.supported_embed.weight.unsqueeze(0).expand(point_embedding.size(0), -1, -1),
+                ),
+                dim=1,
+            )
+            # tokens_all = torch.cat((output_tokens, point_embedding), dim=1)
+        else:
+            class_embeddings = []
+            for i in class_vector:
+                if i > self.last_supported:
+                    class_embeddings.append(self.zeroshot_embed.weight)
+                else:
+                    class_embeddings.append(self.supported_embed.weight)
+            tokens_all = torch.cat((output_tokens, point_embedding, torch.stack(class_embeddings)), dim=1)
+        # cross attention
+        masks = []
+        max_prompt = self.max_prompt
+        for i in range(int(np.ceil(tokens_all.shape[0] / max_prompt))):
+            # remove variables in previous for loops to save peak memory for self.transformer
+            src, upscaled_embedding, hyper_in = None, None, None
+            torch.cuda.empty_cache()
+            idx = (i * max_prompt, min((i + 1) * max_prompt, tokens_all.shape[0]))
+            tokens = tokens_all[idx[0] : idx[1]]
+            src = torch.repeat_interleave(out_low, tokens.shape[0], dim=0)
+            pos_src = torch.repeat_interleave(self.pe_layer(out_low.shape[-3:]).unsqueeze(0), tokens.shape[0], dim=0)
+            b, c, h, w, d = src.shape
+            hs, src = self.transformer(src, pos_src, tokens)
+            mask_tokens_out = hs[:, :1, :]
+            hyper_in = self.output_hypernetworks_mlps(mask_tokens_out)
+            src = src.transpose(1, 2).view(b, c, h, w, d)  # type: ignore
+            upscaled_embedding = self.output_upscaling(src)
+            b, c, h, w, d = upscaled_embedding.shape
+            mask = hyper_in @ upscaled_embedding.view(b, c, h * w * d)
+            masks.append(mask.view(-1, 1, h, w, d))
+
+        return torch.vstack(masks)
+
+
+class ClassMappingClassify(nn.Module):
+    """Class head that performs automatic segmentation based on class vector."""
+
+    def __init__(self, n_classes: int, feature_size: int, use_mlp: bool = True):
+        """Args:
+        n_classes: maximum number of class embedding.
+        feature_size: class embedding size.
+        use_mlp: use mlp to further map class embedding.
+        """
+        super().__init__()
+        self.use_mlp = use_mlp
+        if use_mlp:
+            self.mlp = nn.Sequential(
+                nn.Linear(feature_size, feature_size),
+                nn.InstanceNorm1d(1),
+                nn.GELU(),
+                nn.Linear(feature_size, feature_size),
+            )
+        self.class_embeddings = nn.Embedding(n_classes, feature_size)
+        self.image_post_mapping = nn.Sequential(
+            UnetrBasicBlock(
+                spatial_dims=3,
+                in_channels=feature_size,
+                out_channels=feature_size,
+                kernel_size=3,
+                stride=1,
+                norm_name="instance",
+                res_block=True,
+            ),
+            UnetrBasicBlock(
+                spatial_dims=3,
+                in_channels=feature_size,
+                out_channels=feature_size,
+                kernel_size=3,
+                stride=1,
+                norm_name="instance",
+                res_block=True,
+            ),
+        )
+
+    def forward(self, src: torch.Tensor, class_vector: torch.Tensor):
+        b, c, h, w, d = src.shape
+        src = self.image_post_mapping(src)
+        class_embedding = self.class_embeddings(class_vector)
+        if self.use_mlp:
+            class_embedding = self.mlp(class_embedding)
+        # [b,1,feat] @ [1,feat,dim], batch dimension become class_embedding batch dimension.
+        masks = []
+        for i in range(b):
+            mask = class_embedding @ src[[i]].view(1, c, h * w * d)
+            masks.append(mask.view(-1, 1, h, w, d))
+
+        return torch.cat(masks, 1), class_embedding
+
+
+class TwoWayTransformer(nn.Module):
+    def __init__(
+        self,
+        depth: int,
+        embedding_dim: int,
+        num_heads: int,
+        mlp_dim: int,
+        activation: tuple | str = "relu",
+        attention_downsample_rate: int = 2,
+    ) -> None:
+        """
+        A transformer decoder that attends to an input image using
+        queries whose positional embedding is supplied.
+        Adapted from `https://github.com/facebookresearch/segment-anything/blob/main/segment_anything/modeling/transformer.py`.
+
+        Args:
+            depth: number of layers in the transformer.
+            embedding_dim: the channel dimension for the input embeddings.
+            num_heads: the number of heads for multihead attention. Must divide embedding_dim.
+            mlp_dim: the channel dimension internal to the MLP block.
+            activation: the activation to use in the MLP block.
+            attention_downsample_rate: the rate at which to downsample the image before projecting.
+        """
+        super().__init__()
+        self.depth = depth
+        self.embedding_dim = embedding_dim
+        self.num_heads = num_heads
+        self.mlp_dim = mlp_dim
+        self.layers = nn.ModuleList()
+
+        for i in range(depth):
+            self.layers.append(
+                TwoWayAttentionBlock(
+                    embedding_dim=embedding_dim,
+                    num_heads=num_heads,
+                    mlp_dim=mlp_dim,
+                    activation=activation,
+                    attention_downsample_rate=attention_downsample_rate,
+                    skip_first_layer_pe=(i == 0),
+                )
+            )
+
+        self.final_attn_token_to_image = Attention(embedding_dim, num_heads, downsample_rate=attention_downsample_rate)
+        self.norm_final_attn = nn.LayerNorm(embedding_dim)
+
+    def forward(
+        self, image_embedding: torch.Tensor, image_pe: torch.Tensor, point_embedding: torch.Tensor
+    ) -> Tuple[torch.Tensor, torch.Tensor]:
+        """
+        Args:
+            image_embedding: image to attend to. Should be shape
+                B x embedding_dim x h x w for any h and w.
+            image_pe: the positional encoding to add to the image. Must
+                have the same shape as image_embedding.
+            point_embedding: the embedding to add to the query points.
+                Must have shape B x N_points x embedding_dim for any N_points.
+
+        Returns:
+            torch.Tensor: the processed point_embedding.
+            torch.Tensor: the processed image_embedding.
+        """
+        # BxCxHxW -> BxHWxC == B x N_image_tokens x C
+        image_embedding = image_embedding.flatten(2).permute(0, 2, 1)
+        image_pe = image_pe.flatten(2).permute(0, 2, 1)
+
+        # Prepare queries
+        queries = point_embedding
+        keys = image_embedding
+
+        # Apply transformer blocks and final layernorm
+        for layer in self.layers:
+            queries, keys = layer(queries=queries, keys=keys, query_pe=point_embedding, key_pe=image_pe)
+
+        # Apply the final attention layer from the points to the image
+        q = queries + point_embedding
+        k = keys + image_pe
+        attn_out = self.final_attn_token_to_image(q=q, k=k, v=keys)
+        queries = queries + attn_out
+        queries = self.norm_final_attn(queries)
+
+        return queries, keys
+
+
+class TwoWayAttentionBlock(nn.Module):
+    def __init__(
+        self,
+        embedding_dim: int,
+        num_heads: int,
+        mlp_dim: int = 2048,
+        activation: tuple | str = "relu",
+        attention_downsample_rate: int = 2,
+        skip_first_layer_pe: bool = False,
+    ) -> None:
+        """
+        A transformer block with four layers: (1) self-attention of sparse
+        inputs, (2) cross attention of sparse inputs to dense inputs, (3) mlp
+        block on sparse inputs, and (4) cross attention of dense inputs to sparse
+        inputs.
+        Adapted from `https://github.com/facebookresearch/segment-anything/blob/main/segment_anything/modeling/transformer.py`.
+
+        Args:
+            embedding_dim: the channel dimension of the embeddings.
+            num_heads: the number of heads in the attention layers.
+            mlp_dim: the hidden dimension of the mlp block.
+            activation: the activation of the mlp block.
+            skip_first_layer_pe: skip the PE on the first layer.
+        """
+        super().__init__()
+        self.self_attn = Attention(embedding_dim, num_heads)
+        self.norm1 = nn.LayerNorm(embedding_dim)
+
+        self.cross_attn_token_to_image = Attention(embedding_dim, num_heads, downsample_rate=attention_downsample_rate)
+        self.norm2 = nn.LayerNorm(embedding_dim)
+
+        self.mlp = MLPBlock(hidden_size=embedding_dim, mlp_dim=mlp_dim, act=activation, dropout_mode="vista3d")
+        self.norm3 = nn.LayerNorm(embedding_dim)
+
+        self.norm4 = nn.LayerNorm(embedding_dim)
+        self.cross_attn_image_to_token = Attention(embedding_dim, num_heads, downsample_rate=attention_downsample_rate)
+
+        self.skip_first_layer_pe = skip_first_layer_pe
+
+    def forward(
+        self, queries: torch.Tensor, keys: torch.Tensor, query_pe: torch.Tensor, key_pe: torch.Tensor
+    ) -> Tuple[torch.Tensor, torch.Tensor]:
+        # Self attention block
+        if self.skip_first_layer_pe:
+            queries = self.self_attn(q=queries, k=queries, v=queries)
+        else:
+            q = queries + query_pe
+            attn_out = self.self_attn(q=q, k=q, v=queries)
+            queries = queries + attn_out
+        queries = self.norm1(queries)
+
+        # Cross attention block, tokens attending to image embedding
+        q = queries + query_pe
+        k = keys + key_pe
+        attn_out = self.cross_attn_token_to_image(q=q, k=k, v=keys)
+        queries = queries + attn_out
+        queries = self.norm2(queries)
+
+        # MLP block
+        mlp_out = self.mlp(queries)
+        queries = queries + mlp_out
+        queries = self.norm3(queries)
+
+        # Cross attention block, image embedding attending to tokens
+        q = queries + query_pe
+        k = keys + key_pe
+        attn_out = self.cross_attn_image_to_token(q=k, k=q, v=queries)
+        keys = keys + attn_out
+        keys = self.norm4(keys)
+
+        return queries, keys
+
+
+class Attention(nn.Module):
+    """
+    An attention layer that allows for downscaling the size of the embedding
+    after projection to queries, keys, and values.
+    Adapted from `https://github.com/facebookresearch/segment-anything/blob/main/segment_anything/modeling/transformer.py`.
+
+    Args:
+        embedding_dim: the channel dimension of the embeddings.
+        num_heads: the number of heads in the attention layers.
+        downsample_rate: the rate at which to downsample the image before projecting.
+    """
+
+    def __init__(self, embedding_dim: int, num_heads: int, downsample_rate: int = 1) -> None:
+        super().__init__()
+        self.embedding_dim = embedding_dim
+        self.internal_dim = embedding_dim // downsample_rate
+        self.num_heads = num_heads
+        if not self.internal_dim % num_heads == 0:
+            raise ValueError("num_heads must divide embedding_dim.")
+
+        self.q_proj = nn.Linear(embedding_dim, self.internal_dim)
+        self.k_proj = nn.Linear(embedding_dim, self.internal_dim)
+        self.v_proj = nn.Linear(embedding_dim, self.internal_dim)
+        self.out_proj = nn.Linear(self.internal_dim, embedding_dim)
+
+    def _separate_heads(self, x: torch.Tensor, num_heads: int) -> torch.Tensor:
+        b, n, c = x.shape
+        x = x.reshape(b, n, num_heads, c // num_heads)
+        # B x N_heads x N_tokens x C_per_head
+        return x.transpose(1, 2)
+
+    def _recombine_heads(self, x: torch.Tensor) -> torch.Tensor:
+        b, n_heads, n_tokens, c_per_head = x.shape
+        x = x.transpose(1, 2)
+        # B x N_tokens x C
+        return x.reshape(b, n_tokens, n_heads * c_per_head)
+
+    def forward(self, q: torch.Tensor, k: torch.Tensor, v: torch.Tensor) -> torch.Tensor:
+        # Input projections
+        q = self.q_proj(q)
+        k = self.k_proj(k)
+        v = self.v_proj(v)
+
+        # Separate into heads
+        q = self._separate_heads(q, self.num_heads)
+        k = self._separate_heads(k, self.num_heads)
+        v = self._separate_heads(v, self.num_heads)
+
+        # Attention
+        _, _, _, c_per_head = q.shape
+        attn = q @ k.permute(0, 1, 3, 2)  # B x N_heads x N_tokens x N_tokens
+        attn = attn / math.sqrt(c_per_head)
+        attn = torch.softmax(attn, dim=-1)
+
+        # Get output
+        out = attn @ v
+        out = self._recombine_heads(out)
+        out = self.out_proj(out)
+
+        return out
+
+
+class PositionEmbeddingRandom(nn.Module):
+    """
+    Positional encoding using random spatial frequencies.
+    Adapted from `https://github.com/facebookresearch/segment-anything/blob/main/segment_anything/modeling/prompt_encoder.py`.
+
+    Args:
+        num_pos_feats: the number of positional encoding features.
+        scale: the scale of the positional encoding.
+    """
+
+    def __init__(self, num_pos_feats: int = 64, scale: Optional[float] = None) -> None:
+        super().__init__()
+        if scale is None or scale <= 0.0:
+            scale = 1.0
+        self.register_buffer("positional_encoding_gaussian_matrix", scale * torch.randn((3, num_pos_feats)))
+
+    def _pe_encoding(self, coords: torch.torch.Tensor) -> torch.torch.Tensor:
+        """Positionally encode points that are normalized to [0,1]."""
+        # assuming coords are in [0, 1]^2 square and have d_1 x ... x d_n x 2 shape
+        coords = 2 * coords - 1
+        # [bs=1,N=2,2] @ [2,128]
+        # [bs=1, N=2, 128]
+        coords = coords @ self.positional_encoding_gaussian_matrix
+        coords = 2 * np.pi * coords
+        # outputs d_1 x ... x d_n x C shape
+        # [bs=1, N=2, 128+128=256]
+        return torch.cat([torch.sin(coords), torch.cos(coords)], dim=-1)
+
+    def forward(self, size: Tuple[int, int, int]) -> torch.torch.Tensor:
+        """Generate positional encoding for a grid of the specified size."""
+        h, w, d = size
+        device: Any = self.positional_encoding_gaussian_matrix.device
+        grid = torch.ones((h, w, d), device=device, dtype=torch.float32)
+        x_embed = grid.cumsum(dim=0) - 0.5
+        y_embed = grid.cumsum(dim=1) - 0.5
+        z_embed = grid.cumsum(dim=2) - 0.5
+        x_embed = x_embed / h
+        y_embed = y_embed / w
+        z_embed = z_embed / d
+        pe = self._pe_encoding(torch.stack([x_embed, y_embed, z_embed], dim=-1))
+        # C x H x W
+        return pe.permute(3, 0, 1, 2)
+
+    def forward_with_coords(
+        self, coords_input: torch.torch.Tensor, image_size: Tuple[int, int, int]
+    ) -> torch.torch.Tensor:
+        """Positionally encode points that are not normalized to [0,1]."""
+        coords = coords_input.clone()
+        coords[:, :, 0] = coords[:, :, 0] / image_size[0]
+        coords[:, :, 1] = coords[:, :, 1] / image_size[1]
+        coords[:, :, 2] = coords[:, :, 2] / image_size[2]
+        # B x N x C
+        return self._pe_encoding(coords.to(torch.float))
+
+
+class MLP(nn.Module):
+    """
+    Multi-layer perceptron. This class is only used for `PointMappingSAM`.
+    Adapted from `https://github.com/facebookresearch/segment-anything/blob/main/segment_anything/modeling/mask_decoder.py`.
+
+    Args:
+        input_dim: the input dimension.
+        hidden_dim: the hidden dimension.
+        output_dim: the output dimension.
+        num_layers: the number of layers.
+        sigmoid_output: whether to apply a sigmoid activation to the output.
+    """
+
+    def __init__(
+        self, input_dim: int, hidden_dim: int, output_dim: int, num_layers: int, sigmoid_output: bool = False
+    ) -> None:
+        super().__init__()
+        self.num_layers = num_layers
+        h = [hidden_dim] * (num_layers - 1)
+        self.layers = nn.ModuleList(nn.Linear(n, k) for n, k in zip([input_dim] + h, h + [output_dim]))
+        self.sigmoid_output = sigmoid_output
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        for i, layer in enumerate(self.layers):
+            x = F.relu(layer(x)) if i < self.num_layers - 1 else layer(x)
+        if self.sigmoid_output:
+            x = F.sigmoid(x)
+        return x
diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index e32bf6fc48..363fce91be 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -1274,10 +1274,10 @@ def convert_points_to_disc(
     coords = unsqueeze_left(torch.stack((coord_rows, coord_cols, coord_z), dim=0), 6)
     coords = coords.repeat(point.shape[0], 2, 1, 1, 1, 1)
     for b, n in np.ndindex(*point.shape[:2]):
-        point_bn = unsqueeze_right(point[b, n], 6)
+        point_bn = unsqueeze_right(point[b, n], 4)
         if point_label[b, n] > -1:
             channel = 0 if (point_label[b, n] == 0 or point_label[b, n] == 2) else 1
-            pow_diff = torch.pow(coords[b, channel] - point_bn[b, n], 2)
+            pow_diff = torch.pow(coords[b, channel] - point_bn, 2)
             if disc:
                 masks[b, channel] += pow_diff.sum(0) < radius**2
             else:
diff --git a/tests/test_segresnet_ds.py b/tests/test_segresnet_ds.py
index 5372fcc8ae..eab7bac9a0 100644
--- a/tests/test_segresnet_ds.py
+++ b/tests/test_segresnet_ds.py
@@ -17,7 +17,7 @@
 from parameterized import parameterized
 
 from monai.networks import eval_mode
-from monai.networks.nets import SegResNetDS
+from monai.networks.nets import SegResNetDS, SegResNetDS2
 from tests.utils import SkipIfBeforePyTorchVersion, test_script_save
 
 device = "cuda" if torch.cuda.is_available() else "cpu"
@@ -71,7 +71,7 @@
 ]
 
 
-class TestResNetDS(unittest.TestCase):
+class TestSegResNetDS(unittest.TestCase):
 
     @parameterized.expand(TEST_CASE_SEGRESNET_DS)
     def test_shape(self, input_param, input_shape, expected_shape):
@@ -80,47 +80,71 @@ def test_shape(self, input_param, input_shape, expected_shape):
             result = net(torch.randn(input_shape).to(device))
             self.assertEqual(result.shape, expected_shape, msg=str(input_param))
 
+    @parameterized.expand(TEST_CASE_SEGRESNET_DS)
+    def test_shape_ds2(self, input_param, input_shape, expected_shape):
+        net = SegResNetDS2(**input_param).to(device)
+        with eval_mode(net):
+            result = net(torch.randn(input_shape).to(device), with_label=False)
+            self.assertEqual(result[0].shape, expected_shape, msg=str(input_param))
+            self.assertTrue(result[1] == [])
+
+            result = net(torch.randn(input_shape).to(device), with_point=False)
+            self.assertEqual(result[1].shape, expected_shape, msg=str(input_param))
+            self.assertTrue(result[0] == [])
+
     @parameterized.expand(TEST_CASE_SEGRESNET_DS2)
     def test_shape2(self, input_param, input_shape, expected_shape):
         dsdepth = input_param.get("dsdepth", 1)
-        net = SegResNetDS(**input_param).to(device)
-
-        net.train()
-        result = net(torch.randn(input_shape).to(device))
-        if dsdepth > 1:
-            assert isinstance(result, list)
-            self.assertEqual(dsdepth, len(result))
-            for i in range(dsdepth):
-                self.assertEqual(
-                    result[i].shape,
-                    expected_shape[:2] + tuple(e // (2**i) for e in expected_shape[2:]),
-                    msg=str(input_param),
-                )
-        else:
-            assert isinstance(result, torch.Tensor)
-            self.assertEqual(result.shape, expected_shape, msg=str(input_param))
-
-        net.eval()
-        result = net(torch.randn(input_shape).to(device))
-        assert isinstance(result, torch.Tensor)
-        self.assertEqual(result.shape, expected_shape, msg=str(input_param))
+        for net in [SegResNetDS, SegResNetDS2]:
+            net = net(**input_param).to(device)
+            net.train()
+            if isinstance(net, SegResNetDS2):
+                result = net(torch.randn(input_shape).to(device), with_label=False)[0]
+            else:
+                result = net(torch.randn(input_shape).to(device))
+            if dsdepth > 1:
+                assert isinstance(result, list)
+                self.assertEqual(dsdepth, len(result))
+                for i in range(dsdepth):
+                    self.assertEqual(
+                        result[i].shape,
+                        expected_shape[:2] + tuple(e // (2**i) for e in expected_shape[2:]),
+                        msg=str(input_param),
+                    )
+            else:
+                assert isinstance(result, torch.Tensor)
+                self.assertEqual(result.shape, expected_shape, msg=str(input_param))
+
+            if not isinstance(net, SegResNetDS2):
+                # eval mode of SegResNetDS2 has same output as training mode
+                # so only test eval mode for SegResNetDS
+                net.eval()
+                result = net(torch.randn(input_shape).to(device))
+                assert isinstance(result, torch.Tensor)
+                self.assertEqual(result.shape, expected_shape, msg=str(input_param))
 
     @parameterized.expand(TEST_CASE_SEGRESNET_DS3)
     def test_shape3(self, input_param, input_shape, expected_shapes):
         dsdepth = input_param.get("dsdepth", 1)
-        net = SegResNetDS(**input_param).to(device)
-
-        net.train()
-        result = net(torch.randn(input_shape).to(device))
-        assert isinstance(result, list)
-        self.assertEqual(dsdepth, len(result))
-        for i in range(dsdepth):
-            self.assertEqual(result[i].shape, expected_shapes[i], msg=str(input_param))
+        for net in [SegResNetDS, SegResNetDS2]:
+            net = net(**input_param).to(device)
+            net.train()
+            if isinstance(net, SegResNetDS2):
+                result = net(torch.randn(input_shape).to(device), with_point=False)[1]
+            else:
+                result = net(torch.randn(input_shape).to(device))
+            assert isinstance(result, list)
+            self.assertEqual(dsdepth, len(result))
+            for i in range(dsdepth):
+                self.assertEqual(result[i].shape, expected_shapes[i], msg=str(input_param))
 
     def test_ill_arg(self):
         with self.assertRaises(ValueError):
             SegResNetDS(spatial_dims=4)
 
+        with self.assertRaises(ValueError):
+            SegResNetDS2(spatial_dims=4)
+
     @SkipIfBeforePyTorchVersion((1, 10))
     def test_script(self):
         input_param, input_shape, _ = TEST_CASE_SEGRESNET_DS[0]
diff --git a/tests/test_vista3d.py b/tests/test_vista3d.py
new file mode 100644
index 0000000000..d3b4e0c10e
--- /dev/null
+++ b/tests/test_vista3d.py
@@ -0,0 +1,85 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.nets import VISTA3D, SegResNetDS2
+from monai.networks.nets.vista3d import ClassMappingClassify, PointMappingSAM
+from tests.utils import SkipIfBeforePyTorchVersion, skip_if_quick
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+TEST_CASES = [
+    [{"encoder_embed_dim": 48, "in_channels": 1}, {}, (1, 1, 64, 64, 64), (1, 1, 64, 64, 64)],
+    [{"encoder_embed_dim": 48, "in_channels": 2}, {}, (1, 2, 64, 64, 64), (1, 1, 64, 64, 64)],
+    [
+        {"encoder_embed_dim": 48, "in_channels": 1},
+        {"class_vector": torch.tensor([1, 2, 3], device=device)},
+        (1, 1, 64, 64, 64),
+        (3, 1, 64, 64, 64),
+    ],
+    [
+        {"encoder_embed_dim": 48, "in_channels": 1},
+        {
+            "point_coords": torch.tensor([[[1, 2, 3], [1, 2, 3]]], device=device),
+            "point_labels": torch.tensor([[1, 0]], device=device),
+        },
+        (1, 1, 64, 64, 64),
+        (1, 1, 64, 64, 64),
+    ],
+    [
+        {"encoder_embed_dim": 48, "in_channels": 1},
+        {
+            "class_vector": torch.tensor([1, 2], device=device),
+            "point_coords": torch.tensor([[[1, 2, 3], [1, 2, 3]], [[1, 2, 3], [1, 2, 3]]], device=device),
+            "point_labels": torch.tensor([[1, 0], [1, 0]], device=device),
+        },
+        (1, 1, 64, 64, 64),
+        (2, 1, 64, 64, 64),
+    ],
+]
+
+
+@SkipIfBeforePyTorchVersion((1, 11))
+@skip_if_quick
+class TestVista3d(unittest.TestCase):
+
+    @parameterized.expand(TEST_CASES)
+    def test_vista3d_shape(self, args, input_params, input_shape, expected_shape):
+        segresnet = SegResNetDS2(
+            in_channels=args["in_channels"],
+            blocks_down=(1, 2, 2, 4, 4),
+            norm="instance",
+            out_channels=args["encoder_embed_dim"],
+            init_filters=args["encoder_embed_dim"],
+            dsdepth=1,
+        )
+        point_head = PointMappingSAM(feature_size=args["encoder_embed_dim"], n_classes=512, last_supported=132)
+        class_head = ClassMappingClassify(n_classes=512, feature_size=args["encoder_embed_dim"], use_mlp=True)
+        net = VISTA3D(image_encoder=segresnet, class_head=class_head, point_head=point_head).to(device)
+        with eval_mode(net):
+            result = net.forward(
+                torch.randn(input_shape).to(device),
+                point_coords=input_params.get("point_coords", None),
+                point_labels=input_params.get("point_labels", None),
+                class_vector=input_params.get("class_vector", None),
+            )
+            self.assertEqual(result.shape, expected_shape)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 7b9a523c97ff9e7f1fad7fb2a761ce5322947500 Mon Sep 17 00:00:00 2001
From: Balamurali <balamuralim.1993@gmail.com>
Date: Thu, 15 Aug 2024 01:01:03 -0700
Subject: [PATCH 127/183] NACLLoss memory management  (#8020)

Fixes # .

### Description

Calling contiguous after applying the permute option to work with view
operation in apply_filter
(https://github.com/Project-MONAI/MONAI/blob/59a7211070538586369afd4a01eca0a7fe2e742e/monai/networks/layers/simplelayers.py#L293).

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Balamurali <balamuralim.1993@gmail.com>
Signed-off-by: bala93 <balamuralim.1993@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/losses/nacl_loss.py | 4 ++--
 tests/test_nacl_loss.py   | 1 +
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/monai/losses/nacl_loss.py b/monai/losses/nacl_loss.py
index 3303e89bce..27a712d308 100644
--- a/monai/losses/nacl_loss.py
+++ b/monai/losses/nacl_loss.py
@@ -95,11 +95,11 @@ def get_constr_target(self, mask: torch.Tensor) -> torch.Tensor:
         rmask: torch.Tensor
 
         if self.dim == 2:
-            oh_labels = F.one_hot(mask.to(torch.int64), num_classes=self.nc).contiguous().permute(0, 3, 1, 2).float()
+            oh_labels = F.one_hot(mask.to(torch.int64), num_classes=self.nc).permute(0, 3, 1, 2).contiguous().float()
             rmask = self.svls_layer(oh_labels)
 
         if self.dim == 3:
-            oh_labels = F.one_hot(mask.to(torch.int64), num_classes=self.nc).contiguous().permute(0, 4, 1, 2, 3).float()
+            oh_labels = F.one_hot(mask.to(torch.int64), num_classes=self.nc).permute(0, 4, 1, 2, 3).contiguous().float()
             rmask = self.svls_layer(oh_labels)
 
         return rmask
diff --git a/tests/test_nacl_loss.py b/tests/test_nacl_loss.py
index 51ec275cf4..704bbdb9b1 100644
--- a/tests/test_nacl_loss.py
+++ b/tests/test_nacl_loss.py
@@ -47,6 +47,7 @@
 
 TEST_CASES = [
     [{"classes": 3, "dim": 2}, {"inputs": inputs, "targets": targets}, 1.1442],
+    [{"classes": 3, "dim": 2}, {"inputs": inputs.repeat(4, 1, 1, 1), "targets": targets.repeat(4, 1, 1)}, 1.1442],
     [{"classes": 3, "dim": 2, "kernel_ops": "gaussian"}, {"inputs": inputs, "targets": targets}, 1.1433],
     [{"classes": 3, "dim": 2, "kernel_ops": "gaussian", "sigma": 0.5}, {"inputs": inputs, "targets": targets}, 1.1469],
     [{"classes": 3, "dim": 2, "distance_type": "l2"}, {"inputs": inputs, "targets": targets}, 1.1269],

From 9f56a3a02eef613546a1a19e98e36627c961c650 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 19 Aug 2024 23:09:10 +0800
Subject: [PATCH 128/183] Move PyType test to weekly test (#8025)

Fixes #8022

### Description

- Add format test to weekly test
- Set pytype test as not required in each PR
- Add packaging in weekly-preview pipeline

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/weekly-preview.yml | 35 +++++++++++++++++++++++++++-
 1 file changed, 34 insertions(+), 1 deletion(-)

diff --git a/.github/workflows/weekly-preview.yml b/.github/workflows/weekly-preview.yml
index e94e1dac5a..8d8cccffad 100644
--- a/.github/workflows/weekly-preview.yml
+++ b/.github/workflows/weekly-preview.yml
@@ -5,6 +5,39 @@ on:
   - cron: "0 2 * * 0"  # 02:00 of every Sunday
 
 jobs:
+  flake8-py3:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        opt: ["codeformat", "pytype", "mypy"]
+    steps:
+    - uses: actions/checkout@v4
+    - name: Set up Python 3.9
+      uses: actions/setup-python@v5
+      with:
+        python-version: '3.9'
+    - name: cache weekly timestamp
+      id: pip-cache
+      run: |
+        echo "datew=$(date '+%Y-%V')" >> $GITHUB_OUTPUT
+    - name: cache for pip
+      uses: actions/cache@v4
+      id: cache
+      with:
+        path: ~/.cache/pip
+        key: ${{ runner.os }}-pip-${{ steps.pip-cache.outputs.datew }}
+    - name: Install dependencies
+      run: |
+        find  /opt/hostedtoolcache/* -maxdepth 0 ! -name 'Python' -exec rm -rf {} \;
+        python -m pip install --upgrade pip wheel
+        python -m pip install -r requirements-dev.txt
+    - name: Lint and type check
+      run: |
+        # clean up temporary files
+        $(pwd)/runtests.sh --build --clean
+        # Github actions have 2 cores, so parallelize pytype
+        $(pwd)/runtests.sh --build --${{ matrix.opt }} -j 2
+
   packaging:
     if: github.repository == 'Project-MONAI/MONAI'
     runs-on: ubuntu-latest
@@ -19,7 +52,7 @@ jobs:
         python-version: '3.9'
     - name: Install setuptools
       run: |
-        python -m pip install --user --upgrade setuptools wheel
+        python -m pip install --user --upgrade setuptools wheel packaging
     - name: Build distribution
       run: |
         export HEAD_COMMIT_ID=$(git rev-parse HEAD)

From 3a6f6200861722df4176274f45277923d877d1ef Mon Sep 17 00:00:00 2001
From: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Date: Mon, 19 Aug 2024 17:26:44 +0100
Subject: [PATCH 129/183] Updating to match Numpy 2.0 requirements (#7857)

Fixes #7856.

### Description

This introduces changes to meet Numpy 2.0 requirements. MONAI itself is
compatible with Numpy 2.0 however some dependencies are not such as
older versions of Pytorch. This PR adjusts the MAX_SEED value to be
compatible with Numpy 2.0 behaviour changes, uses the `ptp` function,
and some other minor tweaks. The versions for dependencies are also
fixed to exclude Numpy 2.0.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Eric Kerfoot <eric.kerfoot@kcl.ac.uk>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 environment-dev.yml                    | 4 ++--
 monai/data/utils.py                    | 2 +-
 monai/transforms/io/array.py           | 2 +-
 monai/transforms/spatial/functional.py | 2 +-
 monai/transforms/transform.py          | 4 ++--
 requirements.txt                       | 2 +-
 setup.cfg                              | 2 +-
 tests/test_meta_tensor.py              | 2 +-
 tests/test_nifti_endianness.py         | 2 +-
 tests/test_signal_fillempty.py         | 4 ++--
 tests/test_signal_fillemptyd.py        | 4 ++--
 11 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/environment-dev.yml b/environment-dev.yml
index d23958baba..a4651ec7e4 100644
--- a/environment-dev.yml
+++ b/environment-dev.yml
@@ -5,10 +5,10 @@ channels:
   - nvidia
   - conda-forge
 dependencies:
-  - numpy>=1.20
+  - numpy>=1.24,<2.0
   - pytorch>=1.9
   - torchvision
-  - pytorch-cuda=11.6
+  - pytorch-cuda>=11.6
   - pip
   - pip:
     - -r requirements-dev.txt
diff --git a/monai/data/utils.py b/monai/data/utils.py
index 7a08300abb..f35c5124d8 100644
--- a/monai/data/utils.py
+++ b/monai/data/utils.py
@@ -927,7 +927,7 @@ def compute_shape_offset(
     corners = in_affine_ @ corners
     all_dist = corners_out[:-1].copy()
     corners_out = corners_out[:-1] / corners_out[-1]
-    out_shape = np.round(corners_out.ptp(axis=1)) if scale_extent else np.round(corners_out.ptp(axis=1) + 1.0)
+    out_shape = np.round(np.ptp(corners_out, axis=1)) if scale_extent else np.round(np.ptp(corners_out, axis=1) + 1.0)
     offset = None
     for i in range(corners.shape[1]):
         min_corner = np.min(all_dist - all_dist[:, i : i + 1], 1)
diff --git a/monai/transforms/io/array.py b/monai/transforms/io/array.py
index e0ecc127f2..7c0e8f7123 100644
--- a/monai/transforms/io/array.py
+++ b/monai/transforms/io/array.py
@@ -86,7 +86,7 @@ def switch_endianness(data, new="<"):
         if new not in ("<", ">"):
             raise NotImplementedError(f"Not implemented option new={new}.")
         if current_ != new:
-            data = data.byteswap().newbyteorder(new)
+            data = data.byteswap().view(data.dtype.newbyteorder(new))
     elif isinstance(data, tuple):
         data = tuple(switch_endianness(x, new) for x in data)
     elif isinstance(data, list):
diff --git a/monai/transforms/spatial/functional.py b/monai/transforms/spatial/functional.py
index add4e7f5ea..22726f06a5 100644
--- a/monai/transforms/spatial/functional.py
+++ b/monai/transforms/spatial/functional.py
@@ -373,7 +373,7 @@ def rotate(img, angle, output_shape, mode, padding_mode, align_corners, dtype, l
     if output_shape is None:
         corners = np.asarray(np.meshgrid(*[(0, dim) for dim in im_shape], indexing="ij")).reshape((len(im_shape), -1))
         corners = transform[:-1, :-1] @ corners  # type: ignore
-        output_shape = np.asarray(corners.ptp(axis=1) + 0.5, dtype=int)
+        output_shape = np.asarray(np.ptp(corners, axis=1) + 0.5, dtype=int)
     else:
         output_shape = np.asarray(output_shape, dtype=int)
     shift = create_translate(input_ndim, ((np.array(im_shape) - 1) / 2).tolist())
diff --git a/monai/transforms/transform.py b/monai/transforms/transform.py
index 3d09cea545..15c2499a73 100644
--- a/monai/transforms/transform.py
+++ b/monai/transforms/transform.py
@@ -203,8 +203,8 @@ def set_random_state(self, seed: int | None = None, state: np.random.RandomState
 
         """
         if seed is not None:
-            _seed = id(seed) if not isinstance(seed, (int, np.integer)) else seed
-            _seed = _seed % MAX_SEED
+            _seed = np.int64(id(seed) if not isinstance(seed, (int, np.integer)) else seed)
+            _seed = _seed % MAX_SEED  # need to account for Numpy2.0 which doesn't silently convert to int64
             self.R = np.random.RandomState(_seed)
             return self
 
diff --git a/requirements.txt b/requirements.txt
index aae455f58c..e184322c13 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,2 +1,2 @@
 torch>=1.9
-numpy>=1.20,<=1.26.0
+numpy>=1.24,<2.0
diff --git a/setup.cfg b/setup.cfg
index 2115c30a7f..1ce4a3f34c 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -42,7 +42,7 @@ setup_requires =
     ninja
 install_requires =
     torch>=1.9
-    numpy>=1.20
+    numpy>=1.24,<2.0
 
 [options.extras_require]
 all =
diff --git a/tests/test_meta_tensor.py b/tests/test_meta_tensor.py
index f31a07eba4..60b6019703 100644
--- a/tests/test_meta_tensor.py
+++ b/tests/test_meta_tensor.py
@@ -448,7 +448,7 @@ def test_shape(self):
 
     def test_astype(self):
         t = MetaTensor([1.0], affine=torch.tensor(1), meta={"fname": "filename"})
-        for np_types in ("float32", "np.float32", "numpy.float32", np.float32, float, "int", np.compat.long, np.uint16):
+        for np_types in ("float32", "np.float32", "numpy.float32", np.float32, float, "int", np.uint16):
             self.assertIsInstance(t.astype(np_types), np.ndarray)
         for pt_types in ("torch.float", torch.float, "torch.float64"):
             self.assertIsInstance(t.astype(pt_types), torch.Tensor)
diff --git a/tests/test_nifti_endianness.py b/tests/test_nifti_endianness.py
index 4475d8aaab..f8531dc08f 100644
--- a/tests/test_nifti_endianness.py
+++ b/tests/test_nifti_endianness.py
@@ -82,7 +82,7 @@ def test_switch(self):  # verify data types
         after = switch_endianness(before)
         np.testing.assert_allclose(after.astype(float), expected_float)
 
-        before = np.array(["1.12", "-9.2", "42"], dtype=np.string_)
+        before = np.array(["1.12", "-9.2", "42"], dtype=np.bytes_)
         after = switch_endianness(before)
         np.testing.assert_array_equal(before, after)
 
diff --git a/tests/test_signal_fillempty.py b/tests/test_signal_fillempty.py
index a3ee623cc5..2be4bd8600 100644
--- a/tests/test_signal_fillempty.py
+++ b/tests/test_signal_fillempty.py
@@ -30,7 +30,7 @@ class TestSignalFillEmptyNumpy(unittest.TestCase):
     def test_correct_parameters_multi_channels(self):
         self.assertIsInstance(SignalFillEmpty(replacement=0.0), SignalFillEmpty)
         sig = np.load(TEST_SIGNAL)
-        sig[:, 123] = np.NAN
+        sig[:, 123] = np.nan
         fillempty = SignalFillEmpty(replacement=0.0)
         fillemptysignal = fillempty(sig)
         self.assertTrue(not np.isnan(fillemptysignal).any())
@@ -42,7 +42,7 @@ class TestSignalFillEmptyTorch(unittest.TestCase):
     def test_correct_parameters_multi_channels(self):
         self.assertIsInstance(SignalFillEmpty(replacement=0.0), SignalFillEmpty)
         sig = convert_to_tensor(np.load(TEST_SIGNAL))
-        sig[:, 123] = convert_to_tensor(np.NAN)
+        sig[:, 123] = convert_to_tensor(np.nan)
         fillempty = SignalFillEmpty(replacement=0.0)
         fillemptysignal = fillempty(sig)
         self.assertTrue(not torch.isnan(fillemptysignal).any())
diff --git a/tests/test_signal_fillemptyd.py b/tests/test_signal_fillemptyd.py
index ee8c571ef8..7710279495 100644
--- a/tests/test_signal_fillemptyd.py
+++ b/tests/test_signal_fillemptyd.py
@@ -30,7 +30,7 @@ class TestSignalFillEmptyNumpy(unittest.TestCase):
     def test_correct_parameters_multi_channels(self):
         self.assertIsInstance(SignalFillEmptyd(replacement=0.0), SignalFillEmptyd)
         sig = np.load(TEST_SIGNAL)
-        sig[:, 123] = np.NAN
+        sig[:, 123] = np.nan
         data = {}
         data["signal"] = sig
         fillempty = SignalFillEmptyd(keys=("signal",), replacement=0.0)
@@ -46,7 +46,7 @@ class TestSignalFillEmptyTorch(unittest.TestCase):
     def test_correct_parameters_multi_channels(self):
         self.assertIsInstance(SignalFillEmptyd(replacement=0.0), SignalFillEmptyd)
         sig = convert_to_tensor(np.load(TEST_SIGNAL))
-        sig[:, 123] = convert_to_tensor(np.NAN)
+        sig[:, 123] = convert_to_tensor(np.nan)
         data = {}
         data["signal"] = sig
         fillempty = SignalFillEmptyd(keys=("signal",), replacement=0.0)

From cea80a686b907b22d90c8024cea5a17ef9d9f58a Mon Sep 17 00:00:00 2001
From: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Date: Tue, 20 Aug 2024 19:55:22 +0800
Subject: [PATCH 130/183] 8029 update load old weights function for
 diffusion_model_unet.py  (#8031)

Fixes #8029  .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/nets/diffusion_model_unet.py | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/monai/networks/nets/diffusion_model_unet.py b/monai/networks/nets/diffusion_model_unet.py
index f57fe251d2..65d6053acc 100644
--- a/monai/networks/nets/diffusion_model_unet.py
+++ b/monai/networks/nets/diffusion_model_unet.py
@@ -1837,9 +1837,26 @@ def load_old_state_dict(self, old_state_dict: dict, verbose=False) -> None:
                 new_state_dict[k] = old_state_dict.pop(k)
 
         # fix the attention blocks
-        attention_blocks = [k.replace(".out_proj.weight", "") for k in new_state_dict if "out_proj.weight" in k]
+        attention_blocks = [k.replace(".attn.to_k.weight", "") for k in new_state_dict if "attn.to_k.weight" in k]
         for block in attention_blocks:
+            new_state_dict[f"{block}.attn.to_q.weight"] = old_state_dict.pop(f"{block}.to_q.weight")
+            new_state_dict[f"{block}.attn.to_k.weight"] = old_state_dict.pop(f"{block}.to_k.weight")
+            new_state_dict[f"{block}.attn.to_v.weight"] = old_state_dict.pop(f"{block}.to_v.weight")
+            new_state_dict[f"{block}.attn.to_q.bias"] = old_state_dict.pop(f"{block}.to_q.bias")
+            new_state_dict[f"{block}.attn.to_k.bias"] = old_state_dict.pop(f"{block}.to_k.bias")
+            new_state_dict[f"{block}.attn.to_v.bias"] = old_state_dict.pop(f"{block}.to_v.bias")
+
             # projection
+            new_state_dict[f"{block}.attn.out_proj.weight"] = old_state_dict.pop(f"{block}.proj_attn.weight")
+            new_state_dict[f"{block}.attn.out_proj.bias"] = old_state_dict.pop(f"{block}.proj_attn.bias")
+
+        # fix the cross attention blocks
+        cross_attention_blocks = [
+            k.replace(".out_proj.weight", "")
+            for k in new_state_dict
+            if "out_proj.weight" in k and "transformer_blocks" in k
+        ]
+        for block in cross_attention_blocks:
             new_state_dict[f"{block}.out_proj.weight"] = old_state_dict.pop(f"{block}.to_out.0.weight")
             new_state_dict[f"{block}.out_proj.bias"] = old_state_dict.pop(f"{block}.to_out.0.bias")
 

From de2a819e82e9c0575a959170d8e534fefe002d08 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 23 Aug 2024 11:16:29 +0800
Subject: [PATCH 131/183] Fix AttributeError when using torch.min and max
 (#8041)

Fixes #8040.

### Description

Only return values if got a namedtuple when using torch.min and max

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .../transforms/utils_pytorch_numpy_unification.py  |  4 ++--
 tests/test_utils_pytorch_numpy_unification.py      | 14 +++++++++++++-
 2 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/monai/transforms/utils_pytorch_numpy_unification.py b/monai/transforms/utils_pytorch_numpy_unification.py
index 020d99af16..98b75cff76 100644
--- a/monai/transforms/utils_pytorch_numpy_unification.py
+++ b/monai/transforms/utils_pytorch_numpy_unification.py
@@ -480,7 +480,7 @@ def max(x: NdarrayTensor, dim: int | tuple | None = None, **kwargs) -> NdarrayTe
         else:
             ret = torch.max(x, int(dim), **kwargs)  # type: ignore
 
-    return ret
+    return ret[0] if isinstance(ret, tuple) else ret
 
 
 def mean(x: NdarrayTensor, dim: int | tuple | None = None, **kwargs) -> NdarrayTensor:
@@ -546,7 +546,7 @@ def min(x: NdarrayTensor, dim: int | tuple | None = None, **kwargs) -> NdarrayTe
         else:
             ret = torch.min(x, int(dim), **kwargs)  # type: ignore
 
-    return ret
+    return ret[0] if isinstance(ret, tuple) else ret
 
 
 def std(x: NdarrayTensor, dim: int | tuple | None = None, unbiased: bool = False) -> NdarrayTensor:
diff --git a/tests/test_utils_pytorch_numpy_unification.py b/tests/test_utils_pytorch_numpy_unification.py
index 6e655289e4..90c0401e46 100644
--- a/tests/test_utils_pytorch_numpy_unification.py
+++ b/tests/test_utils_pytorch_numpy_unification.py
@@ -17,7 +17,7 @@
 import torch
 from parameterized import parameterized
 
-from monai.transforms.utils_pytorch_numpy_unification import mode, percentile
+from monai.transforms.utils_pytorch_numpy_unification import max, min, mode, percentile
 from monai.utils import set_determinism
 from tests.utils import TEST_NDARRAYS, assert_allclose, skip_if_quick
 
@@ -27,6 +27,13 @@
     TEST_MODE.append([p(np.array([3.1, 4.1, 4.1, 5.1])), p(4.1), False])
     TEST_MODE.append([p(np.array([3.1, 4.1, 4.1, 5.1])), p(4), True])
 
+TEST_MIN_MAX = []
+for p in TEST_NDARRAYS:
+    TEST_MIN_MAX.append([p(np.array([1, 2, 3, 4, 4, 5])), {}, min, p(1)])
+    TEST_MIN_MAX.append([p(np.array([[3.1, 4.1, 4.1, 5.1], [3, 5, 4.1, 5]])), {"dim": 1}, min, p([3.1, 3])])
+    TEST_MIN_MAX.append([p(np.array([1, 2, 3, 4, 4, 5])), {}, max, p(5)])
+    TEST_MIN_MAX.append([p(np.array([[3.1, 4.1, 4.1, 5.1], [3, 5, 4.1, 5]])), {"dim": 1}, max, p([5.1, 5])])
+
 
 class TestPytorchNumpyUnification(unittest.TestCase):
 
@@ -74,6 +81,11 @@ def test_mode(self, array, expected, to_long):
         res = mode(array, to_long=to_long)
         assert_allclose(res, expected)
 
+    @parameterized.expand(TEST_MIN_MAX)
+    def test_min_max(self, array, input_params, func, expected):
+        res = func(array, **input_params)
+        assert_allclose(res, expected, type_test=False)
+
 
 if __name__ == "__main__":
     unittest.main()

From a5fbe716378948630783deef8ee435e7e3bdc918 Mon Sep 17 00:00:00 2001
From: Han123su <107395380+Han123su@users.noreply.github.com>
Date: Fri, 23 Aug 2024 12:09:23 +0800
Subject: [PATCH 132/183] Refactor Export for Model Conversion and Saving
 (#7934)

Fixes #6375 .

### Description
Changes to be made based on the [previous discussion
#7835](https://github.com/Project-MONAI/MONAI/pull/7835).

Modify the `_export` function to call the `saver` parameter for saving
different models. Rewrite the `onnx_export` function using the updated
`_export` to achieve consistency in model format conversion and saving.

* Rewrite `onnx_export` to call `_export` with `convert_to_onnx` and
appropriate `kwargs`.
* Add a `saver: Callable` parameter to `_export`, replacing
`save_net_with_metadata`.
* Pass `save_net_with_metadata` function wrapped with `partial` to set
parameters like `include_config_vals` and `append_timestamp`.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Han123su <popsmall212@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/bundle/scripts.py | 46 ++++++++++++++++++++++++++---------------
 1 file changed, 29 insertions(+), 17 deletions(-)

diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index 6dd83c1f81..142a366669 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -18,6 +18,7 @@
 import warnings
 import zipfile
 from collections.abc import Mapping, Sequence
+from functools import partial
 from pathlib import Path
 from pydoc import locate
 from shutil import copyfile
@@ -1254,6 +1255,7 @@ def verify_net_in_out(
 
 def _export(
     converter: Callable,
+    saver: Callable,
     parser: ConfigParser,
     net_id: str,
     filepath: str,
@@ -1268,6 +1270,8 @@ def _export(
     Args:
         converter: a callable object that takes a torch.nn.module and kwargs as input and
             converts the module to another type.
+        saver: a callable object that accepts the converted model to save, a filepath to save to, meta values
+            (extracted from the parser), and a dictionary of extra JSON files (name -> contents) as input.
         parser: a ConfigParser of the bundle to be converted.
         net_id: ID name of the network component in the parser, it must be `torch.nn.Module`.
         filepath: filepath to export, if filename has no extension, it becomes `.ts`.
@@ -1307,14 +1311,9 @@ def _export(
     # add .json extension to all extra files which are always encoded as JSON
     extra_files = {k + ".json": v for k, v in extra_files.items()}
 
-    save_net_with_metadata(
-        jit_obj=net,
-        filename_prefix_or_stream=filepath,
-        include_config_vals=False,
-        append_timestamp=False,
-        meta_values=parser.get().pop("_meta_", None),
-        more_extra_files=extra_files,
-    )
+    meta_values = parser.get().pop("_meta_", None)
+    saver(net, filepath, meta_values=meta_values, more_extra_files=extra_files)
+
     logger.info(f"exported to file: {filepath}.")
 
 
@@ -1413,17 +1412,23 @@ def onnx_export(
         input_shape_ = _get_fake_input_shape(parser=parser)
 
     inputs_ = [torch.rand(input_shape_)]
-    net = parser.get_parsed_content(net_id_)
-    if has_ignite:
-        # here we use ignite Checkpoint to support nested weights and be compatible with MONAI CheckpointSaver
-        Checkpoint.load_objects(to_load={key_in_ckpt_: net}, checkpoint=ckpt_file_)
-    else:
-        ckpt = torch.load(ckpt_file_)
-        copy_model_state(dst=net, src=ckpt if key_in_ckpt_ == "" else ckpt[key_in_ckpt_])
 
     converter_kwargs_.update({"inputs": inputs_, "use_trace": use_trace_})
-    onnx_model = convert_to_onnx(model=net, **converter_kwargs_)
-    onnx.save(onnx_model, filepath_)
+
+    def save_onnx(onnx_obj: Any, filename_prefix_or_stream: str, **kwargs: Any) -> None:
+        onnx.save(onnx_obj, filename_prefix_or_stream)
+
+    _export(
+        convert_to_onnx,
+        save_onnx,
+        parser,
+        net_id=net_id_,
+        filepath=filepath_,
+        ckpt_file=ckpt_file_,
+        config_file=config_file_,
+        key_in_ckpt=key_in_ckpt_,
+        **converter_kwargs_,
+    )
 
 
 def ckpt_export(
@@ -1544,8 +1549,12 @@ def ckpt_export(
 
     converter_kwargs_.update({"inputs": inputs_, "use_trace": use_trace_})
     # Use the given converter to convert a model and save with metadata, config content
+
+    save_ts = partial(save_net_with_metadata, include_config_vals=False, append_timestamp=False)
+
     _export(
         convert_to_torchscript,
+        save_ts,
         parser,
         net_id=net_id_,
         filepath=filepath_,
@@ -1715,8 +1724,11 @@ def trt_export(
     }
     converter_kwargs_.update(trt_api_parameters)
 
+    save_ts = partial(save_net_with_metadata, include_config_vals=False, append_timestamp=False)
+
     _export(
         convert_to_trt,
+        save_ts,
         parser,
         net_id=net_id_,
         filepath=filepath_,

From 872585df178b71df88399a395083577165aeb5ac Mon Sep 17 00:00:00 2001
From: Yufan He <59374597+heyufan1995@users.noreply.github.com>
Date: Mon, 26 Aug 2024 00:12:05 -0500
Subject: [PATCH 133/183] Add vista3d inferers (#8021)

Fixes # .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: heyufan1995 <heyufan1995@gmail.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/apps.rst                          |  16 ++
 .../maisi/utils => vista3d}/__init__.py       |   0
 monai/apps/vista3d/inferer.py                 | 177 ++++++++++++++
 monai/apps/vista3d/sampler.py                 | 172 ++++++++++++++
 monai/apps/vista3d/transforms.py              | 224 ++++++++++++++++++
 monai/inferers/utils.py                       |   1 +
 monai/networks/nets/vista3d.py                |  43 +++-
 monai/transforms/utils.py                     |  55 ++++-
 tests/min_tests.py                            |   1 +
 tests/test_point_based_window_inferer.py      |  77 ++++++
 tests/test_vista3d_sampler.py                 | 100 ++++++++
 tests/test_vista3d_transforms.py              |  94 ++++++++
 tests/test_vista3d_utils.py                   |  45 +++-
 13 files changed, 988 insertions(+), 17 deletions(-)
 rename monai/apps/{generation/maisi/utils => vista3d}/__init__.py (100%)
 create mode 100644 monai/apps/vista3d/inferer.py
 create mode 100644 monai/apps/vista3d/sampler.py
 create mode 100644 monai/apps/vista3d/transforms.py
 create mode 100644 tests/test_point_based_window_inferer.py
 create mode 100644 tests/test_vista3d_sampler.py
 create mode 100644 tests/test_vista3d_transforms.py

diff --git a/docs/source/apps.rst b/docs/source/apps.rst
index 7fa7b9e9ff..cc4cea8c1e 100644
--- a/docs/source/apps.rst
+++ b/docs/source/apps.rst
@@ -248,6 +248,22 @@ FastMRIReader
 ~~~~~~~~~~~~~
 .. autofunction:: monai.apps.reconstruction.complex_utils.complex_conj
 
+`Vista3d`
+---------
+.. automodule:: monai.apps.vista3d.inferer
+.. autofunction:: point_based_window_inferer
+
+.. automodule:: monai.apps.vista3d.transforms
+.. autoclass:: VistaPreTransformd
+    :members:
+.. autoclass:: VistaPostTransformd
+    :members:
+.. autoclass:: Relabeld
+    :members:
+
+.. automodule:: monai.apps.vista3d.sampler
+.. autofunction:: sample_prompt_pairs
+
 `Auto3DSeg`
 -----------
 .. automodule:: monai.apps.auto3dseg
diff --git a/monai/apps/generation/maisi/utils/__init__.py b/monai/apps/vista3d/__init__.py
similarity index 100%
rename from monai/apps/generation/maisi/utils/__init__.py
rename to monai/apps/vista3d/__init__.py
diff --git a/monai/apps/vista3d/inferer.py b/monai/apps/vista3d/inferer.py
new file mode 100644
index 0000000000..709f81f624
--- /dev/null
+++ b/monai/apps/vista3d/inferer.py
@@ -0,0 +1,177 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import copy
+from collections.abc import Sequence
+from typing import Any
+
+import torch
+
+from monai.data.meta_tensor import MetaTensor
+from monai.utils import optional_import
+
+tqdm, _ = optional_import("tqdm", name="tqdm")
+
+__all__ = ["point_based_window_inferer"]
+
+
+def point_based_window_inferer(
+    inputs: torch.Tensor | MetaTensor,
+    roi_size: Sequence[int],
+    predictor: torch.nn.Module,
+    point_coords: torch.Tensor,
+    point_labels: torch.Tensor,
+    class_vector: torch.Tensor | None = None,
+    prompt_class: torch.Tensor | None = None,
+    prev_mask: torch.Tensor | MetaTensor | None = None,
+    point_start: int = 0,
+    center_only: bool = True,
+    margin: int = 5,
+    **kwargs: Any,
+) -> torch.Tensor:
+    """
+    Point-based window inferer that takes an input image, a set of points, and a model, and returns a segmented image.
+    The inferer algorithm crops the input image into patches that centered at the point sets, which is followed by
+    patch inference and average output stitching, and finally returns the segmented mask.
+
+    Args:
+        inputs: [1CHWD], input image to be processed.
+        roi_size: the spatial window size for inferences.
+            When its components have None or non-positives, the corresponding inputs dimension will be used.
+            if the components of the `roi_size` are non-positive values, the transform will use the
+            corresponding components of img size. For example, `roi_size=(32, -1)` will be adapted
+            to `(32, 64)` if the second spatial dimension size of img is `64`.
+        sw_batch_size: the batch size to run window slices.
+        predictor: the model. For vista3D, the output is [B, 1, H, W, D] which needs to be transposed to [1, B, H, W, D].
+            Add transpose=True in kwargs for vista3d.
+        point_coords: [B, N, 3]. Point coordinates for B foreground objects, each has N points.
+        point_labels: [B, N]. Point labels. 0/1 means negative/positive points for regular supported or zero-shot classes.
+            2/3 means negative/positive points for special supported classes (e.g. tumor, vessel).
+        class_vector: [B]. Used for class-head automatic segmentation. Can be None value.
+        prompt_class: [B]. The same as class_vector representing the point class and inform point head about
+            supported class or zeroshot, not used for automatic segmentation. If None, point head is default
+            to supported class segmentation.
+        prev_mask: [1, B, H, W, D]. The value is before sigmoid. An optional tensor of previously segmented masks.
+        point_start: only use points starting from this number. All points before this number is used to generate
+            prev_mask. This is used to avoid re-calculating the points in previous iterations if given prev_mask.
+        center_only: for each point, only crop the patch centered at this point. If false, crop 3 patches for each point.
+        margin: if center_only is false, this value is the distance between point to the patch boundary.
+    Returns:
+        stitched_output: [1, B, H, W, D]. The value is before sigmoid.
+    Notice: The function only supports SINGLE OBJECT INFERENCE with B=1.
+    """
+    if not point_coords.shape[0] == 1:
+        raise ValueError("Only supports single object point click.")
+    if not len(inputs.shape) == 5:
+        raise ValueError("Input image should be 5D.")
+    image, pad = _pad_previous_mask(copy.deepcopy(inputs), roi_size)
+    point_coords = point_coords + torch.tensor([pad[-2], pad[-4], pad[-6]]).to(point_coords.device)
+    prev_mask = _pad_previous_mask(copy.deepcopy(prev_mask), roi_size)[0] if prev_mask is not None else None
+    stitched_output = None
+    for p in point_coords[0][point_start:]:
+        lx_, rx_ = _get_window_idx(p[0], roi_size[0], image.shape[-3], center_only=center_only, margin=margin)
+        ly_, ry_ = _get_window_idx(p[1], roi_size[1], image.shape[-2], center_only=center_only, margin=margin)
+        lz_, rz_ = _get_window_idx(p[2], roi_size[2], image.shape[-1], center_only=center_only, margin=margin)
+        for i in range(len(lx_)):
+            for j in range(len(ly_)):
+                for k in range(len(lz_)):
+                    lx, rx, ly, ry, lz, rz = (lx_[i], rx_[i], ly_[j], ry_[j], lz_[k], rz_[k])
+                    unravel_slice = [
+                        slice(None),
+                        slice(None),
+                        slice(int(lx), int(rx)),
+                        slice(int(ly), int(ry)),
+                        slice(int(lz), int(rz)),
+                    ]
+                    batch_image = image[unravel_slice]
+                    output = predictor(
+                        batch_image,
+                        point_coords=point_coords,
+                        point_labels=point_labels,
+                        class_vector=class_vector,
+                        prompt_class=prompt_class,
+                        patch_coords=unravel_slice,
+                        prev_mask=prev_mask,
+                        **kwargs,
+                    )
+                    if stitched_output is None:
+                        stitched_output = torch.zeros(
+                            [1, output.shape[1], image.shape[-3], image.shape[-2], image.shape[-1]], device="cpu"
+                        )
+                        stitched_mask = torch.zeros(
+                            [1, output.shape[1], image.shape[-3], image.shape[-2], image.shape[-1]], device="cpu"
+                        )
+                    stitched_output[unravel_slice] += output.to("cpu")
+                    stitched_mask[unravel_slice] = 1
+    # if stitched_mask is 0, then NaN value
+    stitched_output = stitched_output / stitched_mask
+    # revert padding
+    stitched_output = stitched_output[
+        :, :, pad[4] : image.shape[-3] - pad[5], pad[2] : image.shape[-2] - pad[3], pad[0] : image.shape[-1] - pad[1]
+    ]
+    stitched_mask = stitched_mask[
+        :, :, pad[4] : image.shape[-3] - pad[5], pad[2] : image.shape[-2] - pad[3], pad[0] : image.shape[-1] - pad[1]
+    ]
+    if prev_mask is not None:
+        prev_mask = prev_mask[
+            :,
+            :,
+            pad[4] : image.shape[-3] - pad[5],
+            pad[2] : image.shape[-2] - pad[3],
+            pad[0] : image.shape[-1] - pad[1],
+        ]
+        prev_mask = prev_mask.to("cpu")  # type: ignore
+        # for un-calculated place, use previous mask
+        stitched_output[stitched_mask < 1] = prev_mask[stitched_mask < 1]
+    if isinstance(inputs, torch.Tensor):
+        inputs = MetaTensor(inputs)
+    if not hasattr(stitched_output, "meta"):
+        stitched_output = MetaTensor(stitched_output, affine=inputs.meta["affine"], meta=inputs.meta)
+    return stitched_output
+
+
+def _get_window_idx_c(p: int, roi: int, s: int) -> tuple[int, int]:
+    """Helper function to get the window index."""
+    if p - roi // 2 < 0:
+        left, right = 0, roi
+    elif p + roi // 2 > s:
+        left, right = s - roi, s
+    else:
+        left, right = int(p) - roi // 2, int(p) + roi // 2
+    return left, right
+
+
+def _get_window_idx(p: int, roi: int, s: int, center_only: bool = True, margin: int = 5) -> tuple[list[int], list[int]]:
+    """Get the window index."""
+    left, right = _get_window_idx_c(p, roi, s)
+    if center_only:
+        return [left], [right]
+    left_most = max(0, p - roi + margin)
+    right_most = min(s, p + roi - margin)
+    left_list = [left_most, right_most - roi, left]
+    right_list = [left_most + roi, right_most, right]
+    return left_list, right_list
+
+
+def _pad_previous_mask(
+    inputs: torch.Tensor | MetaTensor, roi_size: Sequence[int], padvalue: int = 0
+) -> tuple[torch.Tensor | MetaTensor, list[int]]:
+    """Helper function to pad inputs."""
+    pad_size = []
+    for k in range(len(inputs.shape) - 1, 1, -1):
+        diff = max(roi_size[k - 2] - inputs.shape[k], 0)
+        half = diff // 2
+        pad_size.extend([half, diff - half])
+    if any(pad_size):
+        inputs = torch.nn.functional.pad(inputs, pad=pad_size, mode="constant", value=padvalue)  # type: ignore
+    return inputs, pad_size
diff --git a/monai/apps/vista3d/sampler.py b/monai/apps/vista3d/sampler.py
new file mode 100644
index 0000000000..b7aeb89a2e
--- /dev/null
+++ b/monai/apps/vista3d/sampler.py
@@ -0,0 +1,172 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import copy
+import random
+from collections.abc import Callable, Sequence
+from typing import Any
+
+import numpy as np
+import torch
+from torch import Tensor
+
+__all__ = ["sample_prompt_pairs"]
+
+ENABLE_SPECIAL = True
+SPECIAL_INDEX = (23, 24, 25, 26, 27, 57, 128)
+MERGE_LIST = {
+    1: [25, 26],  # hepatic tumor and vessel merge into liver
+    4: [24],  # pancreatic tumor merge into pancreas
+    132: [57],  # overlap with trachea merge into airway
+}
+
+
+def _get_point_label(id: int) -> tuple[int, int]:
+    if id in SPECIAL_INDEX and ENABLE_SPECIAL:
+        return 2, 3
+    else:
+        return 0, 1
+
+
+def sample_prompt_pairs(
+    labels: Tensor,
+    label_set: Sequence[int],
+    max_prompt: int | None = None,
+    max_foreprompt: int | None = None,
+    max_backprompt: int = 1,
+    max_point: int = 20,
+    include_background: bool = False,
+    drop_label_prob: float = 0.2,
+    drop_point_prob: float = 0.2,
+    point_sampler: Callable | None = None,
+    **point_sampler_kwargs: Any,
+) -> tuple[Tensor | None, Tensor | None, Tensor | None, Tensor | None]:
+    """
+    Sample training pairs for VISTA3D training.
+
+    Args:
+        labels: [1, 1, H, W, D], ground truth labels.
+        label_set: the label list for the specific dataset. Note if 0 is included in label_set,
+            it will be added into automatic branch training. Recommend removing 0 from label_set
+            for multi-partially-labeled-dataset training, and adding 0 for finetuning specific dataset.
+            The reason is region with 0 in one partially labeled dataset may contain foregrounds in
+            another dataset.
+        max_prompt: int, max number of total prompt, including foreground and background.
+        max_foreprompt: int, max number of prompt from foreground.
+        max_backprompt: int, max number of prompt from background.
+        max_point: maximum number of points for each object.
+        include_background: if include 0 into training prompt. If included, background 0 is treated
+            the same as foreground. Always be False for multi-partial-dataset training. If needed,
+            can be true for finetuning specific dataset, .
+        drop_label_prob: probability to drop label prompt.
+        drop_point_prob: probability to drop point prompt.
+        point_sampler: sampler to augment masks with supervoxel.
+        point_sampler_kwargs: arguments for point_sampler.
+
+    Returns:
+        label_prompt: [B, 1]. The classes used for training automatic segmentation.
+        point: [B, N, 3]. The corresponding points for each class.
+        Note that background label prompt requires matching point as well ([0,0,0] is used).
+        point_label: [B, N]. The corresponding point labels for each point (negative or positive).
+        -1 is used for padding the background label prompt and will be ignored.
+        prompt_class: [B, 1], exactly the same with label_prompt for label indexing for training loss.
+        label_prompt can be None, and prompt_class is used to identify point classes.
+    """
+    # class label number
+    if not labels.shape[0] == 1:
+        raise ValueError("only support batch size 1")
+    labels = labels[0, 0]
+    device = labels.device
+    unique_labels = labels.unique().cpu().numpy().tolist()
+    if include_background:
+        unique_labels = list(set(unique_labels) - (set(unique_labels) - set(label_set)))
+    else:
+        unique_labels = list(set(unique_labels) - (set(unique_labels) - set(label_set)) - {0})
+    background_labels = list(set(label_set) - set(unique_labels))
+    # during training, balance background and foreground prompts
+    if max_backprompt is not None:
+        if len(background_labels) > max_backprompt:
+            random.shuffle(background_labels)
+            background_labels = background_labels[:max_backprompt]
+
+    if max_foreprompt is not None:
+        if len(unique_labels) > max_foreprompt:
+            random.shuffle(unique_labels)
+            unique_labels = unique_labels[:max_foreprompt]
+
+    if max_prompt is not None:
+        if len(unique_labels) + len(background_labels) > max_prompt:
+            if len(unique_labels) > max_prompt:
+                unique_labels = random.sample(unique_labels, max_prompt)
+                background_labels = []
+            else:
+                background_labels = random.sample(background_labels, max_prompt - len(unique_labels))
+    _point = []
+    _point_label = []
+    # if use regular sampling
+    if point_sampler is None:
+        num_p = min(max_point, int(np.abs(random.gauss(mu=0, sigma=max_point // 2))) + 1)
+        num_n = min(max_point, int(np.abs(random.gauss(mu=0, sigma=max_point // 2))))
+        for id in unique_labels:
+            neg_id, pos_id = _get_point_label(id)
+            plabels = labels == int(id)
+            nlabels = ~plabels
+            plabelpoints = torch.nonzero(plabels)
+            nlabelpoints = torch.nonzero(nlabels)
+            # final sampled positive points
+            num_pa = min(len(plabelpoints), num_p)
+            # final sampled negative points
+            num_na = min(len(nlabelpoints), num_n)
+            _point.append(
+                torch.stack(
+                    random.choices(plabelpoints, k=num_pa)
+                    + random.choices(nlabelpoints, k=num_na)
+                    + [torch.tensor([0, 0, 0], device=device)] * (num_p + num_n - num_pa - num_na)
+                )
+            )
+            _point_label.append(
+                torch.tensor([pos_id] * num_pa + [neg_id] * num_na + [-1] * (num_p + num_n - num_pa - num_na)).to(
+                    device
+                )
+            )
+        for _ in background_labels:
+            # pad the background labels
+            _point.append(torch.zeros(num_p + num_n, 3).to(device))  # all 0
+            _point_label.append(torch.zeros(num_p + num_n).to(device) - 1)  # -1 not a point
+    else:
+        _point, _point_label = point_sampler(unique_labels, **point_sampler_kwargs)
+        for _ in background_labels:
+            # pad the background labels
+            _point.append(torch.zeros(len(_point_label[0]), 3).to(device))  # all 0
+            _point_label.append(torch.zeros(len(_point_label[0])).to(device) - 1)  # -1 not a point
+    if len(unique_labels) == 0 and len(background_labels) == 0:
+        # if max_backprompt is 0 and len(unique_labels), there is no effective prompt and the iteration must
+        # be skipped. Handle this in trainer.
+        label_prompt, point, point_label, prompt_class = None, None, None, None
+    else:
+        label_prompt = torch.tensor(unique_labels + background_labels).unsqueeze(-1).to(device).long()
+        point = torch.stack(_point)
+        point_label = torch.stack(_point_label)
+        prompt_class = copy.deepcopy(label_prompt)
+        if random.uniform(0, 1) < drop_label_prob and len(unique_labels) > 0:
+            label_prompt = None
+            # If label prompt is dropped, there is no need to pad with points with label -1.
+            pad = len(background_labels)
+            point = point[: len(point) - pad]  # type: ignore
+            point_label = point_label[: len(point_label) - pad]
+            prompt_class = prompt_class[: len(prompt_class) - pad]
+        else:
+            if random.uniform(0, 1) < drop_point_prob:
+                point = None
+                point_label = None
+    return label_prompt, point, point_label, prompt_class
diff --git a/monai/apps/vista3d/transforms.py b/monai/apps/vista3d/transforms.py
new file mode 100644
index 0000000000..3e8145cd80
--- /dev/null
+++ b/monai/apps/vista3d/transforms.py
@@ -0,0 +1,224 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import warnings
+from typing import Sequence
+
+import numpy as np
+import torch
+
+from monai.config import DtypeLike, KeysCollection
+from monai.transforms import MapLabelValue
+from monai.transforms.transform import MapTransform
+from monai.transforms.utils import keep_components_with_positive_points
+from monai.utils import look_up_option
+
+__all__ = ["VistaPreTransformd", "VistaPostTransformd", "Relabeld"]
+
+
+def _get_name_to_index_mapping(labels_dict: dict | None) -> dict:
+    """get the label name to index mapping"""
+    name_to_index_mapping = {}
+    if labels_dict is not None:
+        name_to_index_mapping = {v.lower(): int(k) for k, v in labels_dict.items()}
+    return name_to_index_mapping
+
+
+def _convert_name_to_index(name_to_index_mapping: dict, label_prompt: list | None) -> list | None:
+    """convert the label name to index"""
+    if label_prompt is not None and isinstance(label_prompt, list):
+        converted_label_prompt = []
+        # for new class, add to the mapping
+        for l in label_prompt:
+            if isinstance(l, str) and not l.isdigit():
+                if l.lower() not in name_to_index_mapping:
+                    name_to_index_mapping[l.lower()] = len(name_to_index_mapping)
+        for l in label_prompt:
+            if isinstance(l, (int, str)):
+                converted_label_prompt.append(
+                    name_to_index_mapping.get(l.lower(), int(l) if l.isdigit() else 0) if isinstance(l, str) else int(l)
+                )
+            else:
+                converted_label_prompt.append(l)
+        return converted_label_prompt
+    return label_prompt
+
+
+class VistaPreTransformd(MapTransform):
+    def __init__(
+        self,
+        keys: KeysCollection,
+        allow_missing_keys: bool = False,
+        special_index: Sequence[int] = (25, 26, 27, 28, 29, 117),
+        labels_dict: dict | None = None,
+        subclass: dict | None = None,
+    ) -> None:
+        """
+        Pre-transform for Vista3d.
+
+        It performs two functionalities:
+
+        1. If label prompt shows the points belong to special class (defined by special index, e.g. tumors, vessels),
+           convert point labels from 0 (negative), 1 (positive) to special 2 (negative), 3 (positive).
+
+        2. If label prompt is within the keys in subclass, convert the label prompt to its subclasses defined by subclass[key].
+           e.g. "lung" label is converted to ["left lung", "right lung"].
+
+        The `label_prompt` is a list of int values of length [B] and `point_labels` is a list of length B,
+        where each element is an int value of length [B, N].
+
+        Args:
+            keys: keys of the corresponding items to be transformed.
+            special_index: the index that defines the special class.
+            subclass: a dictionary that maps a label prompt to its subclasses.
+            allow_missing_keys: don't raise exception if key is missing.
+        """
+        super().__init__(keys, allow_missing_keys)
+        self.special_index = special_index
+        self.subclass = subclass
+        self.name_to_index_mapping = _get_name_to_index_mapping(labels_dict)
+
+    def __call__(self, data):
+        label_prompt = data.get("label_prompt", None)
+        point_labels = data.get("point_labels", None)
+        # convert the label name to index if needed
+        label_prompt = _convert_name_to_index(self.name_to_index_mapping, label_prompt)
+        try:
+            # The evaluator will check prompt. The invalid prompt will be skipped here and captured by evaluator.
+            if self.subclass is not None and label_prompt is not None:
+                _label_prompt = []
+                subclass_keys = list(map(int, self.subclass.keys()))
+                for i in range(len(label_prompt)):
+                    if label_prompt[i] in subclass_keys:
+                        _label_prompt.extend(self.subclass[str(label_prompt[i])])
+                    else:
+                        _label_prompt.append(label_prompt[i])
+                data["label_prompt"] = _label_prompt
+            if label_prompt is not None and point_labels is not None:
+                if label_prompt[0] in self.special_index:
+                    point_labels = np.array(point_labels)
+                    point_labels[point_labels == 0] = 2
+                    point_labels[point_labels == 1] = 3
+                    point_labels = point_labels.tolist()
+                data["point_labels"] = point_labels
+        except Exception:
+            # There is specific requirements for `label_prompt` and `point_labels`.
+            # If B > 1 or `label_prompt` is in subclass_keys, `point_labels` must be None.
+            # Those formatting errors should be captured later.
+            warnings.warn("VistaPreTransformd failed to transform label prompt or point labels.")
+
+        return data
+
+
+class VistaPostTransformd(MapTransform):
+    def __init__(self, keys: KeysCollection, allow_missing_keys: bool = False) -> None:
+        """
+        Post-transform for Vista3d. It converts the model output logits into final segmentation masks.
+        If `label_prompt` is None, the output will be thresholded to be sequential indexes [0,1,2,...],
+        else the indexes will be [0, label_prompt[0], label_prompt[1], ...].
+        If `label_prompt` is None while `points` are provided, the model will perform postprocess to remove
+        regions that does not contain positive points.
+
+        Args:
+            keys: keys of the corresponding items to be transformed.
+            dataset_transforms: a dictionary specifies the transform for corresponding dataset:
+                key: dataset name, value: list of data transforms.
+            dataset_key: key to get the dataset name from the data dictionary, default to "dataset_name".
+            allow_missing_keys: don't raise exception if key is missing.
+
+        """
+        super().__init__(keys, allow_missing_keys)
+
+    def __call__(self, data):
+        """data["label_prompt"] should not contain 0"""
+        for keys in self.keys:
+            if keys in data:
+                pred = data[keys]
+                object_num = pred.shape[0]
+                device = pred.device
+                if data.get("label_prompt", None) is None and data.get("points", None) is not None:
+                    pred = keep_components_with_positive_points(
+                        pred.unsqueeze(0),
+                        point_coords=data.get("points").to(device),
+                        point_labels=data.get("point_labels").to(device),
+                    )[0]
+                pred[pred < 0] = 0.0
+                # if it's multichannel, perform argmax
+                if object_num > 1:
+                    # concate background channel. Make sure user did not provide 0 as prompt.
+                    is_bk = torch.all(pred <= 0, dim=0, keepdim=True)
+                    pred = pred.argmax(0).unsqueeze(0).float() + 1.0
+                    pred[is_bk] = 0.0
+                else:
+                    # AsDiscrete will remove NaN
+                    # pred = monai.transforms.AsDiscrete(threshold=0.5)(pred)
+                    pred[pred > 0] = 1.0
+                if "label_prompt" in data and data["label_prompt"] is not None:
+                    pred += 0.5  # inplace mapping to avoid cloning pred
+                    label_prompt = data["label_prompt"].to(device)  # Ensure label_prompt is on the same device
+                    for i in range(1, object_num + 1):
+                        frac = i + 0.5
+                        pred[pred == frac] = label_prompt[i - 1].to(pred.dtype)
+                    pred[pred == 0.5] = 0.0
+                data[keys] = pred
+        return data
+
+
+class Relabeld(MapTransform):
+    def __init__(
+        self,
+        keys: KeysCollection,
+        label_mappings: dict[str, list[tuple[int, int]]],
+        dtype: DtypeLike = np.int16,
+        dataset_key: str = "dataset_name",
+        allow_missing_keys: bool = False,
+    ) -> None:
+        """
+        Remap the voxel labels in the input data dictionary based on the specified mapping.
+
+        This list of local -> global label mappings will be applied to each input `data[keys]`.
+        if `data[dataset_key]` is not in `label_mappings`, label_mappings['default']` will be used.
+        if `label_mappings[data[dataset_key]]` is None, no relabeling will be performed.
+
+        Args:
+            keys: keys of the corresponding items to be transformed.
+            label_mappings: a dictionary specifies how local dataset class indices are mapped to the
+                global class indices. The dictionary keys are dataset names and the values are lists of
+                list of (local label, global label) pairs. This list of local -> global label mappings
+                will be applied to each input `data[keys]`. If `data[dataset_key]` is not in `label_mappings`,
+                label_mappings['default']` will be used. if `label_mappings[data[dataset_key]]` is None,
+                no relabeling will be performed. Please set `label_mappings={}` to completely skip this transform.
+            dtype: convert the output data to dtype, default to float32.
+            dataset_key: key to get the dataset name from the data dictionary, default to "dataset_name".
+            allow_missing_keys: don't raise exception if key is missing.
+
+        """
+        super().__init__(keys, allow_missing_keys)
+        self.mappers = {}
+        self.dataset_key = dataset_key
+        for name, mapping in label_mappings.items():
+            self.mappers[name] = MapLabelValue(
+                orig_labels=[int(pair[0]) for pair in mapping],
+                target_labels=[int(pair[1]) for pair in mapping],
+                dtype=dtype,
+            )
+
+    def __call__(self, data):
+        d = dict(data)
+        dataset_name = d.get(self.dataset_key, "default")
+        _m = look_up_option(dataset_name, self.mappers, default=None)
+        if _m is None:
+            return d
+        for key in self.key_iterator(d):
+            d[key] = _m(d[key])
+        return d
diff --git a/monai/inferers/utils.py b/monai/inferers/utils.py
index a080284e7c..bd99765348 100644
--- a/monai/inferers/utils.py
+++ b/monai/inferers/utils.py
@@ -300,6 +300,7 @@ def sliding_window_inference(
 
     # remove padding if image_size smaller than roi_size
     if any(pad_size):
+        kwargs.update({"pad_size": pad_size})
         for ss, output_i in enumerate(output_image_list):
             zoom_scale = [_shape_d / _roi_size_d for _shape_d, _roi_size_d in zip(output_i.shape[2:], roi_size)]
             final_slicing: list[slice] = []
diff --git a/monai/networks/nets/vista3d.py b/monai/networks/nets/vista3d.py
index fe7f93d493..9148e36542 100644
--- a/monai/networks/nets/vista3d.py
+++ b/monai/networks/nets/vista3d.py
@@ -23,7 +23,7 @@
 from monai.networks.blocks import MLPBlock, UnetrBasicBlock
 from monai.networks.nets import SegResNetDS2
 from monai.transforms.utils import convert_points_to_disc
-from monai.transforms.utils import get_largest_connected_component_mask_point as lcc
+from monai.transforms.utils import keep_merge_components_with_points as lcc
 from monai.transforms.utils import sample_points_from_label
 from monai.utils import optional_import, unsqueeze_left, unsqueeze_right
 
@@ -78,6 +78,35 @@ def __init__(self, image_encoder: nn.Module, class_head: nn.Module, point_head:
         self.NINF_VALUE = -9999
         self.PINF_VALUE = 9999
 
+    def update_slidingwindow_padding(
+        self,
+        pad_size: list | None,
+        labels: torch.Tensor | None,
+        prev_mask: torch.Tensor | None,
+        point_coords: torch.Tensor | None,
+    ):
+        """
+        Image has been padded by sliding window inferer.
+        The related padding need to be performed outside of slidingwindow inferer.
+
+        Args:
+            pad_size: padding size passed from sliding window inferer.
+            labels: image label ground truth.
+            prev_mask: previous segmentation mask.
+            point_coords: point click coordinates.
+        """
+        if pad_size is None:
+            return labels, prev_mask, point_coords
+        if labels is not None:
+            labels = F.pad(labels, pad=pad_size, mode="constant", value=0)
+        if prev_mask is not None:
+            prev_mask = F.pad(prev_mask, pad=pad_size, mode="constant", value=0)
+        if point_coords is not None:
+            point_coords = point_coords + torch.tensor(
+                [pad_size[-2], pad_size[-4], pad_size[-6]], device=point_coords.device
+            )
+        return labels, prev_mask, point_coords
+
     def get_foreground_class_count(self, class_vector: torch.Tensor | None, point_coords: torch.Tensor | None) -> int:
         """Get number of foreground classes based on class and point prompt."""
         if class_vector is None:
@@ -317,6 +346,7 @@ def forward(
         prev_mask: torch.Tensor | None = None,
         radius: int | None = None,
         val_point_sampler: Callable | None = None,
+        transpose: bool = False,
         **kwargs,
     ):
         """
@@ -329,7 +359,7 @@ def forward(
             point_coords: [B, N, 3]
             point_labels: [B, N], -1 represents padding. 0/1 means negative/positive points for regular class.
                 2/3 means negative/postive ponits for special supported class like tumor.
-            class_vector: [B, 1], the global class index
+            class_vector: [B, 1], the global class index.
             prompt_class: [B, 1], the global class index. This value is associated with point_coords to identify if
                 the points are for zero-shot or supported class. When class_vector and point_coords are both
                 provided, prompt_class is the same as class_vector. For prompt_class[b] > 512, point_coords[b]
@@ -346,8 +376,12 @@ def forward(
             radius: single float value controling the gaussian blur when combining point and auto results.
                 The gaussian combine is not used in VISTA3D training but might be useful for finetuning purposes.
             val_point_sampler: function used to sample points from labels. This is only used for point-only evaluation.
-
+            transpose: bool. If true, the output will be transposed to be [1, B, H, W, D]. Required to be true if calling from
+                sliding window inferer/point inferer.
         """
+        labels, prev_mask, point_coords = self.update_slidingwindow_padding(
+            kwargs.get("pad_size", None), labels, prev_mask, point_coords
+        )
         image_size = input_images.shape[-3:]
         device = input_images.device
         if point_coords is None and class_vector is None:
@@ -424,9 +458,10 @@ def forward(
                     point_labels,  # type: ignore
                     mapping_index,
                 )
-
         if kwargs.get("keep_cache", False) and class_vector is None:
             self.image_embeddings = out.detach()
+        if transpose:
+            logits = logits.transpose(1, 0)
         return logits
 
 
diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index 363fce91be..7027c07d67 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -107,7 +107,8 @@
     "generate_spatial_bounding_box",
     "get_extreme_points",
     "get_largest_connected_component_mask",
-    "get_largest_connected_component_mask_point",
+    "keep_merge_components_with_points",
+    "keep_components_with_positive_points",
     "convert_points_to_disc",
     "remove_small_objects",
     "img_bounds",
@@ -1178,7 +1179,7 @@ def get_largest_connected_component_mask(
     return convert_to_dst_type(out, dst=img, dtype=out.dtype)[0]
 
 
-def get_largest_connected_component_mask_point(
+def keep_merge_components_with_points(
     img_pos: NdarrayTensor,
     img_neg: NdarrayTensor,
     point_coords: NdarrayTensor,
@@ -1188,8 +1189,8 @@ def get_largest_connected_component_mask_point(
     margins: int = 3,
 ) -> NdarrayTensor:
     """
-    Gets the connected component of img_pos and img_neg that include the positive points and
-    negative points separately. The function is used for combining automatic results with interactive
+    Keep connected regions of img_pos and img_neg that include the positive points and
+    negative points separately. The function is used for merging automatic results with interactive
     results in VISTA3D.
 
     Args:
@@ -1199,6 +1200,7 @@ def get_largest_connected_component_mask_point(
         neg_val: negative point label values.
         point_coords: the coordinates of each point, shape [B, N, 3], where N means the number of points.
         point_labels: the label of each point, shape [B, N].
+        margins: include points outside of the region but within the margin.
     """
 
     cucim_skimage, has_cucim = optional_import("cucim.skimage")
@@ -1249,6 +1251,49 @@ def get_largest_connected_component_mask_point(
     return convert_to_dst_type(outs, dst=img_pos, dtype=outs.dtype)[0]
 
 
+def keep_components_with_positive_points(
+    img: torch.Tensor, point_coords: torch.Tensor, point_labels: torch.Tensor
+) -> torch.Tensor:
+    """
+    Keep connected regions that include the positive points. Used for point-only inference postprocessing to remove
+    regions without positive points.
+    Args:
+        img: [1, B, H, W, D]. Output prediction from VISTA3D. Value is before sigmoid and contain NaN value.
+        point_coords: [B, N, 3]. Point click coordinates
+        point_labels: [B, N]. Point click labels.
+    """
+    if not has_measure:
+        raise RuntimeError("skimage.measure required.")
+    outs = torch.zeros_like(img)
+    for c in range(len(point_coords)):
+        if not ((point_labels[c] == 3).any() or (point_labels[c] == 1).any()):
+            # skip if no positive points.
+            continue
+        coords = point_coords[c, point_labels[c] == 3].tolist() + point_coords[c, point_labels[c] == 1].tolist()
+        not_nan_mask = ~torch.isnan(img[0, c])
+        img_ = torch.nan_to_num(img[0, c] > 0, 0)
+        img_, *_ = convert_data_type(img_, np.ndarray)  # type: ignore
+        label = measure.label
+        features = label(img_, connectivity=3)
+        pos_mask = torch.from_numpy(img_).to(img.device) > 0
+        # if num features less than max desired, nothing to do.
+        features = torch.from_numpy(features).to(img.device)
+        # generate a map with all pos points
+        idx = []
+        for p in coords:
+            idx.append(features[round(p[0]), round(p[1]), round(p[2])].item())
+        idx = list(set(idx))
+        for i in idx:
+            if i == 0:
+                continue
+            outs[0, c] += features == i
+        outs = outs > 0
+        # find negative mean value
+        fill_in = img[0, c][torch.logical_and(~outs[0, c], not_nan_mask)].mean()
+        img[0, c][torch.logical_and(pos_mask, ~outs[0, c])] = fill_in
+    return img
+
+
 def convert_points_to_disc(
     image_size: Sequence[int], point: Tensor, point_label: Tensor, radius: int = 2, disc: bool = False
 ):
@@ -1269,7 +1314,7 @@ def convert_points_to_disc(
     _array = [
         torch.arange(start=0, end=image_size[i], step=1, dtype=torch.float32, device=point.device) for i in range(3)
     ]
-    coord_rows, coord_cols, coord_z = torch.meshgrid(_array[2], _array[1], _array[0])
+    coord_rows, coord_cols, coord_z = torch.meshgrid(_array[0], _array[1], _array[2])
     # [1, 3, h, w, d] -> [b, 2, 3, h, w, d]
     coords = unsqueeze_left(torch.stack((coord_rows, coord_cols, coord_z), dim=0), 6)
     coords = coords.repeat(point.shape[0], 2, 1, 1, 1, 1)
diff --git a/tests/min_tests.py b/tests/min_tests.py
index 479c4c8dc2..f80d06f5d3 100644
--- a/tests/min_tests.py
+++ b/tests/min_tests.py
@@ -210,6 +210,7 @@ def run_testsuit():
         "test_perceptual_loss",
         "test_ultrasound_confidence_map_transform",
         "test_vista3d_utils",
+        "test_vista3d_transforms",
     ]
     assert sorted(exclude_cases) == sorted(set(exclude_cases)), f"Duplicated items in {exclude_cases}"
 
diff --git a/tests/test_point_based_window_inferer.py b/tests/test_point_based_window_inferer.py
new file mode 100644
index 0000000000..1b293288c4
--- /dev/null
+++ b/tests/test_point_based_window_inferer.py
@@ -0,0 +1,77 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.apps.vista3d.inferer import point_based_window_inferer
+from monai.networks import eval_mode
+from monai.networks.nets.vista3d import vista3d132
+from monai.utils import optional_import
+from tests.utils import SkipIfBeforePyTorchVersion, skip_if_quick
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+_, has_tqdm = optional_import("tqdm")
+
+TEST_CASES = [
+    [
+        {"encoder_embed_dim": 48, "in_channels": 1},
+        (1, 1, 64, 64, 64),
+        {
+            "roi_size": [32, 32, 32],
+            "point_coords": torch.tensor([[[1, 2, 3], [1, 2, 3]]], device=device),
+            "point_labels": torch.tensor([[1, 0]], device=device),
+        },
+    ],
+    [
+        {"encoder_embed_dim": 48, "in_channels": 1},
+        (1, 1, 64, 64, 64),
+        {
+            "roi_size": [32, 32, 32],
+            "point_coords": torch.tensor([[[1, 2, 3], [1, 2, 3]]], device=device),
+            "point_labels": torch.tensor([[1, 0]], device=device),
+            "class_vector": torch.tensor([1], device=device),
+        },
+    ],
+    [
+        {"encoder_embed_dim": 48, "in_channels": 1},
+        (1, 1, 64, 64, 64),
+        {
+            "roi_size": [32, 32, 32],
+            "point_coords": torch.tensor([[[1, 2, 3], [1, 2, 3]]], device=device),
+            "point_labels": torch.tensor([[1, 0]], device=device),
+            "class_vector": torch.tensor([1], device=device),
+            "point_start": 1,
+        },
+    ],
+]
+
+
+@SkipIfBeforePyTorchVersion((1, 11))
+@skip_if_quick
+class TestPointBasedWindowInferer(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_vista3d(self, vista3d_params, inputs_shape, inferer_params):
+        vista3d = vista3d132(**vista3d_params).to(device)
+        with eval_mode(vista3d):
+            inferer_params["predictor"] = vista3d
+            inferer_params["inputs"] = torch.randn(*inputs_shape).to(device)
+            stitched_output = point_based_window_inferer(**inferer_params)
+            self.assertEqual(stitched_output.shape, inputs_shape)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_vista3d_sampler.py b/tests/test_vista3d_sampler.py
new file mode 100644
index 0000000000..6945d250d2
--- /dev/null
+++ b/tests/test_vista3d_sampler.py
@@ -0,0 +1,100 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.apps.vista3d.sampler import sample_prompt_pairs
+
+label = torch.zeros([1, 1, 64, 64, 64])
+label[:, :, :10, :10, :10] = 1
+label[:, :, 20:30, 20:30, 20:30] = 2
+label[:, :, 30:40, 30:40, 30:40] = 3
+label1 = torch.zeros([1, 1, 64, 64, 64])
+
+TEST_VISTA_SAMPLE_PROMPT = [
+    [
+        {
+            "labels": label,
+            "label_set": [0, 1, 2, 3, 4],
+            "max_prompt": 5,
+            "max_foreprompt": 4,
+            "max_backprompt": 1,
+            "drop_label_prob": 0,
+            "drop_point_prob": 0,
+        },
+        [4, 4, 4, 4],
+    ],
+    [
+        {
+            "labels": label,
+            "label_set": [0, 1],
+            "max_prompt": 5,
+            "max_foreprompt": 4,
+            "max_backprompt": 1,
+            "drop_label_prob": 0,
+            "drop_point_prob": 1,
+        },
+        [2, None, None, 2],
+    ],
+    [
+        {
+            "labels": label,
+            "label_set": [0, 1, 2, 3, 4],
+            "max_prompt": 5,
+            "max_foreprompt": 4,
+            "max_backprompt": 1,
+            "drop_label_prob": 1,
+            "drop_point_prob": 0,
+        },
+        [None, 3, 3, 3],
+    ],
+    [
+        {
+            "labels": label1,
+            "label_set": [0, 1],
+            "max_prompt": 5,
+            "max_foreprompt": 4,
+            "max_backprompt": 1,
+            "drop_label_prob": 0,
+            "drop_point_prob": 1,
+        },
+        [1, None, None, 1],
+    ],
+    [
+        {
+            "labels": label1,
+            "label_set": [0, 1],
+            "max_prompt": 5,
+            "max_foreprompt": 4,
+            "max_backprompt": 0,
+            "drop_label_prob": 0,
+            "drop_point_prob": 1,
+        },
+        [None, None, None, None],
+    ],
+]
+
+
+class TestGeneratePrompt(unittest.TestCase):
+    @parameterized.expand(TEST_VISTA_SAMPLE_PROMPT)
+    def test_result(self, input_data, expected):
+        output = sample_prompt_pairs(**input_data)
+        result = [i.shape[0] if i is not None else None for i in output]
+        self.assertEqual(result, expected)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_vista3d_transforms.py b/tests/test_vista3d_transforms.py
new file mode 100644
index 0000000000..9d61fe2fc2
--- /dev/null
+++ b/tests/test_vista3d_transforms.py
@@ -0,0 +1,94 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+from unittest.case import skipUnless
+
+import torch
+from parameterized import parameterized
+
+from monai.apps.vista3d.transforms import VistaPostTransformd, VistaPreTransformd
+from monai.utils import min_version
+from monai.utils.module import optional_import
+
+measure, has_measure = optional_import("skimage.measure", "0.14.2", min_version)
+
+device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
+
+
+TEST_VISTA_PRETRANSFORM = [
+    [
+        {"label_prompt": [1], "points": [[0, 0, 0]], "point_labels": [1]},
+        {"label_prompt": [1], "points": [[0, 0, 0]], "point_labels": [3]},
+    ],
+    [
+        {"label_prompt": [2], "points": [[0, 0, 0]], "point_labels": [0]},
+        {"label_prompt": [2], "points": [[0, 0, 0]], "point_labels": [2]},
+    ],
+    [
+        {"label_prompt": [3], "points": [[0, 0, 0]], "point_labels": [0]},
+        {"label_prompt": [4, 5], "points": [[0, 0, 0]], "point_labels": [0]},
+    ],
+    [
+        {"label_prompt": [6], "points": [[0, 0, 0]], "point_labels": [0]},
+        {"label_prompt": [7, 8], "points": [[0, 0, 0]], "point_labels": [0]},
+    ],
+]
+
+
+pred1 = torch.zeros([2, 64, 64, 64])
+pred1[0, :10, :10, :10] = 1
+pred1[1, 20:30, 20:30, 20:30] = 1
+output1 = torch.zeros([1, 64, 64, 64])
+output1[:, :10, :10, :10] = 2
+output1[:, 20:30, 20:30, 20:30] = 3
+
+# -1 is needed since pred should be before sigmoid.
+pred2 = torch.zeros([1, 64, 64, 64]) - 1
+pred2[:, :10, :10, :10] = 1
+pred2[:, 20:30, 20:30, 20:30] = 1
+output2 = torch.zeros([1, 64, 64, 64])
+output2[:, 20:30, 20:30, 20:30] = 1
+
+TEST_VISTA_POSTTRANSFORM = [
+    [{"pred": pred1.to(device), "label_prompt": torch.tensor([2, 3]).to(device)}, output1.to(device)],
+    [
+        {
+            "pred": pred2.to(device),
+            "points": torch.tensor([[25, 25, 25]]).to(device),
+            "point_labels": torch.tensor([1]).to(device),
+        },
+        output2.to(device),
+    ],
+]
+
+
+class TestVistaPreTransformd(unittest.TestCase):
+    @parameterized.expand(TEST_VISTA_PRETRANSFORM)
+    def test_result(self, input_data, expected):
+        transform = VistaPreTransformd(keys="image", subclass={"3": [4, 5], "6": [7, 8]}, special_index=[1, 2])
+        result = transform(input_data)
+        self.assertEqual(result, expected)
+
+
+@skipUnless(has_measure, "skimage.measure required")
+class TestVistaPostTransformd(unittest.TestCase):
+    @parameterized.expand(TEST_VISTA_POSTTRANSFORM)
+    def test_result(self, input_data, expected):
+        transform = VistaPostTransformd(keys="pred")
+        result = transform(input_data)
+        self.assertEqual((result["pred"] == expected).all(), True)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_vista3d_utils.py b/tests/test_vista3d_utils.py
index a940854d88..5a0caedd61 100644
--- a/tests/test_vista3d_utils.py
+++ b/tests/test_vista3d_utils.py
@@ -18,11 +18,7 @@
 import torch
 from parameterized import parameterized
 
-from monai.transforms.utils import (
-    convert_points_to_disc,
-    get_largest_connected_component_mask_point,
-    sample_points_from_label,
-)
+from monai.transforms.utils import convert_points_to_disc, keep_merge_components_with_points, sample_points_from_label
 from monai.utils import min_version
 from monai.utils.module import optional_import
 from tests.utils import skip_if_no_cuda, skip_if_quick
@@ -57,6 +53,31 @@
                 expected_shape,
             ]
         )
+        image_size = (16, 32, 64)
+        point = torch.tensor([[[8, 16, 42], [2, 8, 21]]])
+        point_label = torch.tensor([[1, 0]])
+        expected_shape = (point.shape[0], 2, *image_size)
+        TEST_CONVERT_POINTS_TO_DISC.append(
+            [
+                {"image_size": image_size, "point": point, "point_label": point_label, "radius": radius, "disc": disc},
+                expected_shape,
+            ]
+        )
+
+TEST_CONVERT_POINTS_TO_DISC_VALUE = []
+image_size = (16, 32, 64)
+point = torch.tensor([[[8, 16, 42], [2, 8, 21]]])
+point_label = torch.tensor([[1, 0]])
+expected_shape = (point.shape[0], 2, *image_size)
+for radius in [5, 10]:
+    for disc in [True, False]:
+        TEST_CONVERT_POINTS_TO_DISC_VALUE.append(
+            [
+                {"image_size": image_size, "point": point, "point_label": point_label, "radius": radius, "disc": disc},
+                [point, point_label],
+            ]
+        )
+
 
 TEST_LCC_MASK_POINT_TORCH = []
 for bs in [1, 2]:
@@ -108,9 +129,17 @@ def test_shape(self, input_data, expected_shape):
         result = convert_points_to_disc(**input_data)
         self.assertEqual(result.shape, expected_shape)
 
+    @parameterized.expand(TEST_CONVERT_POINTS_TO_DISC_VALUE)
+    def test_value(self, input_data, points):
+        result = convert_points_to_disc(**input_data)
+        point, point_label = points
+        for i in range(point.shape[0]):
+            for j in range(point.shape[1]):
+                self.assertEqual(result[i, point_label[i, j], point[i, j][0], point[i, j][1], point[i, j][2]], True)
+
 
 @skipUnless(has_measure or cucim_skimage, "skimage or cucim.skimage required")
-class TestGetLargestConnectedComponentMaskPoint(unittest.TestCase):
+class TestKeepMergeComponentsWithPoints(unittest.TestCase):
 
     @skip_if_quick
     @skip_if_no_cuda
@@ -119,13 +148,13 @@ class TestGetLargestConnectedComponentMaskPoint(unittest.TestCase):
     def test_cp_shape(self, input_data, shape):
         for key in input_data:
             input_data[key] = input_data[key].to(device)
-        mask = get_largest_connected_component_mask_point(**input_data)
+        mask = keep_merge_components_with_points(**input_data)
         self.assertEqual(mask.shape, shape)
 
     @skipUnless(has_measure, "skimage required")
     @parameterized.expand(TEST_LCC_MASK_POINT_NP)
     def test_np_shape(self, input_data, shape):
-        mask = get_largest_connected_component_mask_point(**input_data)
+        mask = keep_merge_components_with_points(**input_data)
         self.assertEqual(mask.shape, shape)
 
 

From 1a8afd18d5fb132efa26d6f73c3bf1fdbccd985d Mon Sep 17 00:00:00 2001
From: mylapallilavanyaa <149993494+mylapallilavanyaa@users.noreply.github.com>
Date: Mon, 26 Aug 2024 11:32:45 +0530
Subject: [PATCH 134/183] Monthly downloads badge added (#7891)

Fixes # .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: mylapallilavanyaa <149993494+mylapallilavanyaa@users.noreply.github.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Mingxin Zheng <18563433+mingxin-zheng@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 README.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/README.md b/README.md
index 5345cdb926..498d3c6149 100644
--- a/README.md
+++ b/README.md
@@ -14,6 +14,7 @@
 [![postmerge](https://img.shields.io/github/checks-status/project-monai/monai/dev?label=postmerge)](https://github.com/Project-MONAI/MONAI/actions?query=branch%3Adev)
 [![Documentation Status](https://readthedocs.org/projects/monai/badge/?version=latest)](https://docs.monai.io/en/latest/)
 [![codecov](https://codecov.io/gh/Project-MONAI/MONAI/branch/dev/graph/badge.svg?token=6FTC7U1JJ4)](https://codecov.io/gh/Project-MONAI/MONAI)
+[![monai Downloads Last Month](https://assets.piptrends.com/get-last-month-downloads-badge/monai.svg 'monai Downloads Last Month by pip Trends')](https://piptrends.com/package/monai)
 
 MONAI is a [PyTorch](https://pytorch.org/)-based, [open-source](https://github.com/Project-MONAI/MONAI/blob/dev/LICENSE) framework for deep learning in healthcare imaging, part of [PyTorch Ecosystem](https://pytorch.org/ecosystem/).
 Its ambitions are:

From b62d1e118711b2665435bdac9e3cefb5d496a84a Mon Sep 17 00:00:00 2001
From: Yufan He <59374597+heyufan1995@users.noreply.github.com>
Date: Wed, 28 Aug 2024 03:14:44 -0500
Subject: [PATCH 135/183] Fix transpose and patch coords bug (#8047)

Fixes # .

### Description

Fix the bug that causes wrong results in model zoo finetuning. Patch
coords was not passed from sliding window to vista3d.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: heyufan1995 <heyufan1995@gmail.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/requirements.txt          |  1 +
 monai/apps/vista3d/sampler.py  | 29 ++++++++++++++++++-----------
 monai/networks/nets/vista3d.py |  7 +++++--
 3 files changed, 24 insertions(+), 13 deletions(-)

diff --git a/docs/requirements.txt b/docs/requirements.txt
index ff94f7b6de..7307d8e5f9 100644
--- a/docs/requirements.txt
+++ b/docs/requirements.txt
@@ -42,3 +42,4 @@ zarr
 huggingface_hub
 pyamg>=5.0.0
 packaging
+polygraphy
diff --git a/monai/apps/vista3d/sampler.py b/monai/apps/vista3d/sampler.py
index b7aeb89a2e..17b2d34911 100644
--- a/monai/apps/vista3d/sampler.py
+++ b/monai/apps/vista3d/sampler.py
@@ -20,8 +20,6 @@
 import torch
 from torch import Tensor
 
-__all__ = ["sample_prompt_pairs"]
-
 ENABLE_SPECIAL = True
 SPECIAL_INDEX = (23, 24, 25, 26, 27, 57, 128)
 MERGE_LIST = {
@@ -30,6 +28,8 @@
     132: [57],  # overlap with trachea merge into airway
 }
 
+__all__ = ["sample_prompt_pairs"]
+
 
 def _get_point_label(id: int) -> tuple[int, int]:
     if id in SPECIAL_INDEX and ENABLE_SPECIAL:
@@ -66,22 +66,29 @@ def sample_prompt_pairs(
         max_backprompt: int, max number of prompt from background.
         max_point: maximum number of points for each object.
         include_background: if include 0 into training prompt. If included, background 0 is treated
-            the same as foreground. Always be False for multi-partial-dataset training. If needed,
-            can be true for finetuning specific dataset, .
+            the same as foreground and points will be sampled. Can be true only if user want to segment
+            background 0 with point clicks, otherwise always be false.
         drop_label_prob: probability to drop label prompt.
         drop_point_prob: probability to drop point prompt.
         point_sampler: sampler to augment masks with supervoxel.
         point_sampler_kwargs: arguments for point_sampler.
 
     Returns:
-        label_prompt: [B, 1]. The classes used for training automatic segmentation.
-        point: [B, N, 3]. The corresponding points for each class.
-        Note that background label prompt requires matching point as well ([0,0,0] is used).
-        point_label: [B, N]. The corresponding point labels for each point (negative or positive).
-        -1 is used for padding the background label prompt and will be ignored.
-        prompt_class: [B, 1], exactly the same with label_prompt for label indexing for training loss.
-        label_prompt can be None, and prompt_class is used to identify point classes.
+        tuple:
+            - label_prompt (Tensor | None): Tensor of shape [B, 1] containing the classes used for
+              training automatic segmentation.
+            - point (Tensor | None): Tensor of shape [B, N, 3] representing the corresponding points
+              for each class. Note that background label prompts require matching points as well
+              (e.g., [0, 0, 0] is used).
+            - point_label (Tensor | None): Tensor of shape [B, N] representing the corresponding point
+              labels for each point (negative or positive). -1 is used for padding the background
+              label prompt and will be ignored.
+            - prompt_class (Tensor | None): Tensor of shape [B, 1], exactly the same as label_prompt
+              for label indexing during training. If label_prompt is None, prompt_class is used to
+              identify point classes.
+
     """
+
     # class label number
     if not labels.shape[0] == 1:
         raise ValueError("only support batch size 1")
diff --git a/monai/networks/nets/vista3d.py b/monai/networks/nets/vista3d.py
index 9148e36542..979a090df0 100644
--- a/monai/networks/nets/vista3d.py
+++ b/monai/networks/nets/vista3d.py
@@ -336,11 +336,11 @@ def set_auto_grad(self, auto_freeze: bool = False, point_freeze: bool = False):
     def forward(
         self,
         input_images: torch.Tensor,
+        patch_coords: Sequence[slice] | None = None,
         point_coords: torch.Tensor | None = None,
         point_labels: torch.Tensor | None = None,
         class_vector: torch.Tensor | None = None,
         prompt_class: torch.Tensor | None = None,
-        patch_coords: Sequence[slice] | None = None,
         labels: torch.Tensor | None = None,
         label_set: Sequence[int] | None = None,
         prev_mask: torch.Tensor | None = None,
@@ -421,7 +421,10 @@ def forward(
                     point_coords, point_labels = None, None
 
         if point_coords is None and class_vector is None:
-            return self.NINF_VALUE + torch.zeros([bs, 1, *image_size], device=device)
+            logits = self.NINF_VALUE + torch.zeros([bs, 1, *image_size], device=device)
+            if transpose:
+                logits = logits.transpose(1, 0)
+            return logits
 
         if self.image_embeddings is not None and kwargs.get("keep_cache", False) and class_vector is None:
             out, out_auto = self.image_embeddings, None

From b6d6d77745ccacfc2e5b478bfc20975104f6ff12 Mon Sep 17 00:00:00 2001
From: stayd <77039165+staydelight@users.noreply.github.com>
Date: Wed, 28 Aug 2024 18:33:00 +0800
Subject: [PATCH 136/183] Add a mapping function in image_reader.py and
 image_writer.py (#7769)

Add a function to create a JSON file that maps input and output paths.

Fixes #7557  .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: staydelight <kevin295643815697236@gmail.com>
Co-authored-by: staydelight <kevin295643815697236@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/transforms.rst        |  12 +++
 monai/transforms/__init__.py      |  14 +++-
 monai/transforms/io/array.py      |  60 ++++++++++++++-
 monai/transforms/io/dictionary.py |  31 +++++++-
 monai/utils/enums.py              |   1 +
 tests/test_mapping_file.py        | 117 ++++++++++++++++++++++++++++
 tests/test_mapping_filed.py       | 122 ++++++++++++++++++++++++++++++
 7 files changed, 351 insertions(+), 6 deletions(-)
 create mode 100644 tests/test_mapping_file.py
 create mode 100644 tests/test_mapping_filed.py

diff --git a/docs/source/transforms.rst b/docs/source/transforms.rst
index 637f0873f1..3e45d899ec 100644
--- a/docs/source/transforms.rst
+++ b/docs/source/transforms.rst
@@ -554,6 +554,12 @@ IO
     :members:
     :special-members: __call__
 
+`WriteFileMapping`
+""""""""""""""""""
+.. autoclass:: WriteFileMapping
+    :members:
+    :special-members: __call__
+
 
 NVIDIA Tool Extension (NVTX)
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
@@ -1642,6 +1648,12 @@ IO (Dict)
     :members:
     :special-members: __call__
 
+`WriteFileMappingd`
+"""""""""""""""""""
+.. autoclass:: WriteFileMappingd
+    :members:
+    :special-members: __call__
+
 Post-processing (Dict)
 ^^^^^^^^^^^^^^^^^^^^^^
 
diff --git a/monai/transforms/__init__.py b/monai/transforms/__init__.py
index 9548443768..f37016e63f 100644
--- a/monai/transforms/__init__.py
+++ b/monai/transforms/__init__.py
@@ -238,8 +238,18 @@
 )
 from .inverse import InvertibleTransform, TraceableTransform
 from .inverse_batch_transform import BatchInverseTransform, Decollated, DecollateD, DecollateDict
-from .io.array import SUPPORTED_READERS, LoadImage, SaveImage
-from .io.dictionary import LoadImaged, LoadImageD, LoadImageDict, SaveImaged, SaveImageD, SaveImageDict
+from .io.array import SUPPORTED_READERS, LoadImage, SaveImage, WriteFileMapping
+from .io.dictionary import (
+    LoadImaged,
+    LoadImageD,
+    LoadImageDict,
+    SaveImaged,
+    SaveImageD,
+    SaveImageDict,
+    WriteFileMappingd,
+    WriteFileMappingD,
+    WriteFileMappingDict,
+)
 from .lazy.array import ApplyPending
 from .lazy.dictionary import ApplyPendingd, ApplyPendingD, ApplyPendingDict
 from .lazy.functional import apply_pending
diff --git a/monai/transforms/io/array.py b/monai/transforms/io/array.py
index 7c0e8f7123..4e71870fc9 100644
--- a/monai/transforms/io/array.py
+++ b/monai/transforms/io/array.py
@@ -15,6 +15,7 @@
 from __future__ import annotations
 
 import inspect
+import json
 import logging
 import sys
 import traceback
@@ -45,11 +46,19 @@
 from monai.transforms.utility.array import EnsureChannelFirst
 from monai.utils import GridSamplePadMode
 from monai.utils import ImageMetaKey as Key
-from monai.utils import OptionalImportError, convert_to_dst_type, ensure_tuple, look_up_option, optional_import
+from monai.utils import (
+    MetaKeys,
+    OptionalImportError,
+    convert_to_dst_type,
+    ensure_tuple,
+    look_up_option,
+    optional_import,
+)
 
 nib, _ = optional_import("nibabel")
 Image, _ = optional_import("PIL.Image")
 nrrd, _ = optional_import("nrrd")
+FileLock, has_filelock = optional_import("filelock", name="FileLock")
 
 __all__ = ["LoadImage", "SaveImage", "SUPPORTED_READERS"]
 
@@ -505,7 +514,7 @@ def __call__(
             else:
                 self._data_index += 1
                 if self.savepath_in_metadict and meta_data is not None:
-                    meta_data["saved_to"] = filename
+                    meta_data[MetaKeys.SAVED_TO] = filename
                 return img
         msg = "\n".join([f"{e}" for e in err])
         raise RuntimeError(
@@ -514,3 +523,50 @@ def __call__(
             "    https://docs.monai.io/en/latest/installation.html#installing-the-recommended-dependencies.\n"
             f"   The current registered writers for {self.output_ext}: {self.writers}.\n{msg}"
         )
+
+
+class WriteFileMapping(Transform):
+    """
+    Writes a JSON file that logs the mapping between input image paths and their corresponding output paths.
+    This class uses FileLock to ensure safe writing to the JSON file in a multiprocess environment.
+
+    Args:
+        mapping_file_path (Path or str): Path to the JSON file where the mappings will be saved.
+    """
+
+    def __init__(self, mapping_file_path: Path | str = "mapping.json"):
+        self.mapping_file_path = Path(mapping_file_path)
+
+    def __call__(self, img: NdarrayOrTensor):
+        """
+        Args:
+            img: The input image with metadata.
+        """
+        if isinstance(img, MetaTensor):
+            meta_data = img.meta
+
+        if MetaKeys.SAVED_TO not in meta_data:
+            raise KeyError(
+                "Missing 'saved_to' key in metadata. Check SaveImage argument 'savepath_in_metadict' is True."
+            )
+
+        input_path = meta_data[Key.FILENAME_OR_OBJ]
+        output_path = meta_data[MetaKeys.SAVED_TO]
+        log_data = {"input": input_path, "output": output_path}
+
+        if has_filelock:
+            with FileLock(str(self.mapping_file_path) + ".lock"):
+                self._write_to_file(log_data)
+        else:
+            self._write_to_file(log_data)
+        return img
+
+    def _write_to_file(self, log_data):
+        try:
+            with self.mapping_file_path.open("r") as f:
+                existing_log_data = json.load(f)
+        except (FileNotFoundError, json.JSONDecodeError):
+            existing_log_data = []
+        existing_log_data.append(log_data)
+        with self.mapping_file_path.open("w") as f:
+            json.dump(existing_log_data, f, indent=4)
diff --git a/monai/transforms/io/dictionary.py b/monai/transforms/io/dictionary.py
index 4da1d422ca..be1e78db8a 100644
--- a/monai/transforms/io/dictionary.py
+++ b/monai/transforms/io/dictionary.py
@@ -17,16 +17,17 @@
 
 from __future__ import annotations
 
+from collections.abc import Hashable, Mapping
 from pathlib import Path
 from typing import Callable
 
 import numpy as np
 
 import monai
-from monai.config import DtypeLike, KeysCollection
+from monai.config import DtypeLike, KeysCollection, NdarrayOrTensor
 from monai.data import image_writer
 from monai.data.image_reader import ImageReader
-from monai.transforms.io.array import LoadImage, SaveImage
+from monai.transforms.io.array import LoadImage, SaveImage, WriteFileMapping
 from monai.transforms.transform import MapTransform, Transform
 from monai.utils import GridSamplePadMode, ensure_tuple, ensure_tuple_rep
 from monai.utils.enums import PostFix
@@ -320,5 +321,31 @@ def __call__(self, data):
         return d
 
 
+class WriteFileMappingd(MapTransform):
+    """
+    Dictionary-based wrapper of :py:class:`monai.transforms.WriteFileMapping`.
+
+    Args:
+          keys: keys of the corresponding items to be transformed.
+              See also: :py:class:`monai.transforms.compose.MapTransform`
+          mapping_file_path: Path to the JSON file where the mappings will be saved.
+              Defaults to "mapping.json".
+          allow_missing_keys: don't raise exception if key is missing.
+    """
+
+    def __init__(
+        self, keys: KeysCollection, mapping_file_path: Path | str = "mapping.json", allow_missing_keys: bool = False
+    ) -> None:
+        super().__init__(keys, allow_missing_keys)
+        self.mapping = WriteFileMapping(mapping_file_path)
+
+    def __call__(self, data: Mapping[Hashable, NdarrayOrTensor]) -> dict[Hashable, NdarrayOrTensor]:
+        d = dict(data)
+        for key in self.key_iterator(d):
+            d[key] = self.mapping(d[key])
+        return d
+
+
 LoadImageD = LoadImageDict = LoadImaged
 SaveImageD = SaveImageDict = SaveImaged
+WriteFileMappingD = WriteFileMappingDict = WriteFileMappingd
diff --git a/monai/utils/enums.py b/monai/utils/enums.py
index b786e92151..eba1be18ed 100644
--- a/monai/utils/enums.py
+++ b/monai/utils/enums.py
@@ -543,6 +543,7 @@ class MetaKeys(StrEnum):
     SPATIAL_SHAPE = "spatial_shape"  # optional key for the length in each spatial dimension
     SPACE = "space"  # possible values of space type are defined in `SpaceKeys`
     ORIGINAL_CHANNEL_DIM = "original_channel_dim"  # an integer or float("nan")
+    SAVED_TO = "saved_to"
 
 
 class ColorOrder(StrEnum):
diff --git a/tests/test_mapping_file.py b/tests/test_mapping_file.py
new file mode 100644
index 0000000000..97fa4312ed
--- /dev/null
+++ b/tests/test_mapping_file.py
@@ -0,0 +1,117 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import json
+import os
+import shutil
+import tempfile
+import unittest
+
+import numpy as np
+from parameterized import parameterized
+
+from monai.data import DataLoader, Dataset
+from monai.transforms import Compose, LoadImage, SaveImage, WriteFileMapping
+from monai.utils import optional_import
+
+nib, has_nib = optional_import("nibabel")
+
+
+def create_input_file(temp_dir, name):
+    test_image = np.random.rand(128, 128, 128)
+    output_ext = ".nii.gz"
+    input_file = os.path.join(temp_dir, name + output_ext)
+    nib.save(nib.Nifti1Image(test_image, np.eye(4)), input_file)
+    return input_file
+
+
+def create_transform(temp_dir, mapping_file_path, savepath_in_metadict=True):
+    return Compose(
+        [
+            LoadImage(image_only=True),
+            SaveImage(output_dir=temp_dir, output_ext=".nii.gz", savepath_in_metadict=savepath_in_metadict),
+            WriteFileMapping(mapping_file_path=mapping_file_path),
+        ]
+    )
+
+
+@unittest.skipUnless(has_nib, "nibabel required")
+class TestWriteFileMapping(unittest.TestCase):
+    def setUp(self):
+        self.temp_dir = tempfile.mkdtemp()
+
+    def tearDown(self):
+        shutil.rmtree(self.temp_dir)
+
+    @parameterized.expand([(True,), (False,)])
+    def test_mapping_file(self, savepath_in_metadict):
+        mapping_file_path = os.path.join(self.temp_dir, "mapping.json")
+        name = "test_image"
+        input_file = create_input_file(self.temp_dir, name)
+        output_file = os.path.join(self.temp_dir, name, name + "_trans.nii.gz")
+
+        transform = create_transform(self.temp_dir, mapping_file_path, savepath_in_metadict)
+
+        if savepath_in_metadict:
+            transform(input_file)
+            self.assertTrue(os.path.exists(mapping_file_path))
+            with open(mapping_file_path) as f:
+                mapping_data = json.load(f)
+            self.assertEqual(len(mapping_data), 1)
+            self.assertEqual(mapping_data[0]["input"], input_file)
+            self.assertEqual(mapping_data[0]["output"], output_file)
+        else:
+            with self.assertRaises(RuntimeError) as cm:
+                transform(input_file)
+            cause_exception = cm.exception.__cause__
+            self.assertIsInstance(cause_exception, KeyError)
+            self.assertIn(
+                "Missing 'saved_to' key in metadata. Check SaveImage argument 'savepath_in_metadict' is True.",
+                str(cause_exception),
+            )
+
+    def test_multiprocess_mapping_file(self):
+        num_images = 50
+
+        single_mapping_file = os.path.join(self.temp_dir, "single_mapping.json")
+        multi_mapping_file = os.path.join(self.temp_dir, "multi_mapping.json")
+
+        data = [create_input_file(self.temp_dir, f"test_image_{i}") for i in range(num_images)]
+
+        # single process
+        single_transform = create_transform(self.temp_dir, single_mapping_file)
+        single_dataset = Dataset(data=data, transform=single_transform)
+        single_loader = DataLoader(single_dataset, batch_size=1, num_workers=0, shuffle=True)
+        for _ in single_loader:
+            pass
+
+        # multiple processes
+        multi_transform = create_transform(self.temp_dir, multi_mapping_file)
+        multi_dataset = Dataset(data=data, transform=multi_transform)
+        multi_loader = DataLoader(multi_dataset, batch_size=4, num_workers=3, shuffle=True)
+        for _ in multi_loader:
+            pass
+
+        with open(single_mapping_file) as f:
+            single_mapping_data = json.load(f)
+        with open(multi_mapping_file) as f:
+            multi_mapping_data = json.load(f)
+
+        single_set = {(entry["input"], entry["output"]) for entry in single_mapping_data}
+        multi_set = {(entry["input"], entry["output"]) for entry in multi_mapping_data}
+
+        self.assertEqual(single_set, multi_set)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_mapping_filed.py b/tests/test_mapping_filed.py
new file mode 100644
index 0000000000..d0f8bcf938
--- /dev/null
+++ b/tests/test_mapping_filed.py
@@ -0,0 +1,122 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import json
+import os
+import shutil
+import tempfile
+import unittest
+
+import numpy as np
+import torch
+from parameterized import parameterized
+
+from monai.data import DataLoader, Dataset, decollate_batch
+from monai.inferers import sliding_window_inference
+from monai.networks.nets import UNet
+from monai.transforms import Compose, EnsureChannelFirstd, LoadImaged, SaveImaged, WriteFileMappingd
+from monai.utils import optional_import
+
+nib, has_nib = optional_import("nibabel")
+
+
+def create_input_file(temp_dir, name):
+    test_image = np.random.rand(128, 128, 128)
+    input_file = os.path.join(temp_dir, name + ".nii.gz")
+    nib.save(nib.Nifti1Image(test_image, np.eye(4)), input_file)
+    return input_file
+
+
+# Test cases that should succeed
+SUCCESS_CASES = [(["seg"], ["seg"]), (["image", "seg"], ["seg"])]
+
+# Test cases that should fail
+FAILURE_CASES = [(["seg"], ["image"]), (["image"], ["seg"]), (["seg"], ["image", "seg"])]
+
+
+@unittest.skipUnless(has_nib, "nibabel required")
+class TestWriteFileMappingd(unittest.TestCase):
+    def setUp(self):
+        self.temp_dir = tempfile.mkdtemp()
+        self.output_dir = os.path.join(self.temp_dir, "output")
+        os.makedirs(self.output_dir)
+        self.mapping_file_path = os.path.join(self.temp_dir, "mapping.json")
+
+    def tearDown(self):
+        shutil.rmtree(self.temp_dir)
+        if os.path.exists(self.mapping_file_path):
+            os.remove(self.mapping_file_path)
+
+    def run_test(self, save_keys, write_keys):
+        name = "test_image"
+        input_file = create_input_file(self.temp_dir, name)
+        output_file = os.path.join(self.output_dir, name, name + "_seg.nii.gz")
+        data = [{"image": input_file}]
+
+        test_transforms = Compose([LoadImaged(keys=["image"]), EnsureChannelFirstd(keys=["image"])])
+
+        post_transforms = Compose(
+            [
+                SaveImaged(
+                    keys=save_keys,
+                    meta_keys="image_meta_dict",
+                    output_dir=self.output_dir,
+                    output_postfix="seg",
+                    savepath_in_metadict=True,
+                ),
+                WriteFileMappingd(keys=write_keys, mapping_file_path=self.mapping_file_path),
+            ]
+        )
+
+        dataset = Dataset(data=data, transform=test_transforms)
+        dataloader = DataLoader(dataset, batch_size=1)
+        device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+        model = UNet(spatial_dims=3, in_channels=1, out_channels=2, channels=(16, 32), strides=(2,)).to(device)
+        model.eval()
+
+        with torch.no_grad():
+            for batch_data in dataloader:
+                test_inputs = batch_data["image"].to(device)
+                roi_size = (64, 64, 64)
+                sw_batch_size = 2
+                batch_data["seg"] = sliding_window_inference(test_inputs, roi_size, sw_batch_size, model)
+                batch_data = [post_transforms(i) for i in decollate_batch(batch_data)]
+
+        return input_file, output_file
+
+    @parameterized.expand(SUCCESS_CASES)
+    def test_successful_mapping_filed(self, save_keys, write_keys):
+        input_file, output_file = self.run_test(save_keys, write_keys)
+        self.assertTrue(os.path.exists(self.mapping_file_path))
+        with open(self.mapping_file_path) as f:
+            mapping_data = json.load(f)
+        self.assertEqual(len(mapping_data), len(write_keys))
+        for entry in mapping_data:
+            self.assertEqual(entry["input"], input_file)
+            self.assertEqual(entry["output"], output_file)
+
+    @parameterized.expand(FAILURE_CASES)
+    def test_failure_mapping_filed(self, save_keys, write_keys):
+        with self.assertRaises(RuntimeError) as cm:
+            self.run_test(save_keys, write_keys)
+
+        cause_exception = cm.exception.__cause__
+        self.assertIsInstance(cause_exception, KeyError)
+        self.assertIn(
+            "Missing 'saved_to' key in metadata. Check SaveImage argument 'savepath_in_metadict' is True.",
+            str(cause_exception),
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()

From 29ce1a743bc067c259ac6646ec67c111a84ee80a Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 29 Aug 2024 11:52:51 +0800
Subject: [PATCH 137/183] Use torch_tensorrt.Device instead of torch.device in
 trt compile (#8051)

Fixes #8050

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/monai/networks/utils.py b/monai/networks/utils.py
index f301c2dd5c..bd65ffa33e 100644
--- a/monai/networks/utils.py
+++ b/monai/networks/utils.py
@@ -851,7 +851,7 @@ def _onnx_trt_compile(
     # wrap the serialized TensorRT engine back to a TorchScript module.
     trt_model = torch_tensorrt.ts.embed_engine_in_new_module(
         f.getvalue(),
-        device=torch.device(f"cuda:{device}"),
+        device=torch_tensorrt.Device(f"cuda:{device}"),
         input_binding_names=input_names,
         output_binding_names=output_names,
     )

From b209347c0b804d966a83141d602777da5e27f1b7 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Sat, 31 Aug 2024 00:45:29 +0800
Subject: [PATCH 138/183] Ensure synchronization by adding cuda.synchronize()
 (#8058)

Fixes #8054

### Description
Add cuda sync after invoke cuda

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/layers/filtering.py | 8 ++++++--
 monai/transforms/utils.py          | 3 ++-
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/monai/networks/layers/filtering.py b/monai/networks/layers/filtering.py
index 0ff1187dcc..c48c77cf98 100644
--- a/monai/networks/layers/filtering.py
+++ b/monai/networks/layers/filtering.py
@@ -51,6 +51,8 @@ def forward(ctx, input, spatial_sigma=5, color_sigma=0.5, fast_approx=True):
         ctx.cs = color_sigma
         ctx.fa = fast_approx
         output_data = _C.bilateral_filter(input, spatial_sigma, color_sigma, fast_approx)
+        if torch.cuda.is_available():
+            torch.cuda.synchronize()
         return output_data
 
     @staticmethod
@@ -139,7 +141,8 @@ def forward(ctx, input_img, sigma_x, sigma_y, sigma_z, color_sigma):
             do_dsig_y,
             do_dsig_z,
         )
-
+        if torch.cuda.is_available():
+            torch.cuda.synchronize()
         return output_tensor
 
     @staticmethod
@@ -301,7 +304,8 @@ def forward(ctx, input_img, guidance_img, sigma_x, sigma_y, sigma_z, color_sigma
             do_dsig_z,
             guidance_img,
         )
-
+        if torch.cuda.is_available():
+            torch.cuda.synchronize()
         return output_tensor
 
     @staticmethod
diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index 7027c07d67..1d1f070568 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -2512,6 +2512,7 @@ def distance_transform_edt(
                 block_params=block_params,
                 float64_distances=float64_distances,
             )
+        torch.cuda.synchronize()
     else:
         if not has_ndimage:
             raise RuntimeError("scipy.ndimage required if cupy is not available")
@@ -2545,7 +2546,7 @@ def distance_transform_edt(
 
     r_vals = []
     if return_distances and distances_original is None:
-        r_vals.append(distances)
+        r_vals.append(distances_ if use_cp else distances)
     if return_indices and indices_original is None:
         r_vals.append(indices)
     if not r_vals:

From d0ba8a60950e4a7ffbffa3bab7349117123f1ddb Mon Sep 17 00:00:00 2001
From: Kennett Vera <kennett.vera@gmail.com>
Date: Fri, 30 Aug 2024 17:07:06 -0700
Subject: [PATCH 139/183] Added docstring to address 'Scaling of
 RandImageFilterd transform #6857' (#8055)

Fixes #6857

### Description

Added docstring to RandImageFilterd method which informs the user that
they need to manually scale the result image when using this method.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] In-line docstrings updated.

---------

Signed-off-by: dedeepyasai <dedeepyasai.sai@gmail.com>
Signed-off-by: saelra <rasaelee@gmail.com>
Signed-off-by: Kelvin R <kelvinrbNsc@gmail.com>
Signed-off-by: ken-ni <kennett.vera@gmail.com>
Signed-off-by: Dureti <98233210+DuretiShemsi@users.noreply.github.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: saelra <rasaelee@gmail.com>
Co-authored-by: Kelvin R. <138339140+K-Rilla@users.noreply.github.com>
Co-authored-by: Kelvin R <kelvinrbNsc@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: dedeepyasai <dedeepyasai.sai@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Ratanachat Saelee <146144408+Saelra@users.noreply.github.com>
Co-authored-by: Dureti <98233210+DuretiShemsi@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/transforms/utility/dictionary.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/monai/transforms/utility/dictionary.py b/monai/transforms/utility/dictionary.py
index 7e3a7b0454..2475060f4e 100644
--- a/monai/transforms/utility/dictionary.py
+++ b/monai/transforms/utility/dictionary.py
@@ -1714,6 +1714,10 @@ class RandImageFilterd(MapTransform, RandomizableTransform):
             Probability the transform is applied to the data
         allow_missing_keys:
             Don't raise exception if key is missing.
+
+    Note:
+        - This transform does not scale output image values automatically to match the range of the input.
+          The output should be scaled by later transforms to match the input if this is desired.
     """
 
     backend = ImageFilter.backend

From fa1ef8be157d5eb96de17aa78642384f68d99397 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Sat, 31 Aug 2024 22:25:42 +0800
Subject: [PATCH 140/183] Update base image to 2408 (#8049)

Fixes #8048

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/cron.yml          | 24 ++++++++++++------------
 .github/workflows/pythonapp-gpu.yml |  4 ++--
 Dockerfile                          |  2 +-
 3 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/.github/workflows/cron.yml b/.github/workflows/cron.yml
index cc113b0446..6732ab7256 100644
--- a/.github/workflows/cron.yml
+++ b/.github/workflows/cron.yml
@@ -13,24 +13,24 @@ jobs:
     strategy:
       matrix:
         environment:
-          - "PT191+CUDA113"
           - "PT110+CUDA113"
-          - "PT113+CUDA113"
-          - "PTLATEST+CUDA121"
+          - "PT113+CUDA118"
+          - "PT210+CUDA121"
+          - "PTLATEST+CUDA124"
         include:
           # https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes
           - environment: PT110+CUDA113
             pytorch: "torch==1.10.2 torchvision==0.11.3 --extra-index-url https://download.pytorch.org/whl/cu113"
             base: "nvcr.io/nvidia/pytorch:21.06-py3"  # CUDA 11.3
-          - environment: PT113+CUDA113
-            pytorch: "torch==1.13.1 torchvision==0.14.1 --extra-index-url https://download.pytorch.org/whl/cu113"
-            base: "nvcr.io/nvidia/pytorch:21.06-py3"  # CUDA 11.3
-          - environment: PT113+CUDA122
+          - environment: PT113+CUDA118
             pytorch: "torch==1.13.1 torchvision==0.14.1 --extra-index-url https://download.pytorch.org/whl/cu121"
-            base: "nvcr.io/nvidia/pytorch:23.08-py3"  # CUDA 12.2
+            base: "nvcr.io/nvidia/pytorch:22.10-py3"  # CUDA 11.8
+          - environment: PT210+CUDA121
+            pytorch: "pytorch==2.1.0 torchvision==0.16.0 --extra-index-url https://download.pytorch.org/whl/cu121"
+            base: "nvcr.io/nvidia/pytorch:23.08-py3"  # CUDA 12.1
           - environment: PTLATEST+CUDA124
             pytorch: "-U torch torchvision --extra-index-url https://download.pytorch.org/whl/cu121"
-            base: "nvcr.io/nvidia/pytorch:24.03-py3"  # CUDA 12.4
+            base: "nvcr.io/nvidia/pytorch:24.08-py3"  # CUDA 12.4
     container:
       image: ${{ matrix.base }}
       options: "--gpus all"
@@ -80,7 +80,7 @@ jobs:
     if: github.repository == 'Project-MONAI/MONAI'
     strategy:
       matrix:
-        container: ["pytorch:23.08", "pytorch:24.03"]
+        container: ["pytorch:23.08", "pytorch:24.08"]
     container:
       image: nvcr.io/nvidia/${{ matrix.container }}-py3  # testing with the latest pytorch base image
       options: "--gpus all"
@@ -129,7 +129,7 @@ jobs:
     if: github.repository == 'Project-MONAI/MONAI'
     strategy:
       matrix:
-        container: ["pytorch:24.03"]
+        container: ["pytorch:24.08"]
     container:
       image: nvcr.io/nvidia/${{ matrix.container }}-py3  # testing with the latest pytorch base image
       options: "--gpus all"
@@ -233,7 +233,7 @@ jobs:
     if: github.repository == 'Project-MONAI/MONAI'
     needs: cron-gpu  # so that monai itself is verified first
     container:
-      image: nvcr.io/nvidia/pytorch:24.03-py3  # testing with the latest pytorch base image
+      image: nvcr.io/nvidia/pytorch:24.08-py3  # testing with the latest pytorch base image
       options: "--gpus all --ipc=host"
     runs-on: [self-hosted, linux, x64, integration]
     steps:
diff --git a/.github/workflows/pythonapp-gpu.yml b/.github/workflows/pythonapp-gpu.yml
index ead622b39c..70c3153076 100644
--- a/.github/workflows/pythonapp-gpu.yml
+++ b/.github/workflows/pythonapp-gpu.yml
@@ -44,9 +44,9 @@ jobs:
             pytorch: "-h"  # we explicitly set pytorch to -h to avoid pip install error
             base: "nvcr.io/nvidia/pytorch:23.08-py3"
           - environment: PT210+CUDA121DOCKER
-            # 24.03: 2.3.0a0+40ec155e58.nv24.3
+            # 24.08: 2.3.0a0+40ec155e58.nv24.3
             pytorch: "-h"  # we explicitly set pytorch to -h to avoid pip install error
-            base: "nvcr.io/nvidia/pytorch:24.03-py3"
+            base: "nvcr.io/nvidia/pytorch:24.08-py3"
     container:
       image: ${{ matrix.base }}
       options: --gpus all --env NVIDIA_DISABLE_REQUIRE=true  # workaround for unsatisfied condition: cuda>=11.6
diff --git a/Dockerfile b/Dockerfile
index 8e255597d1..e97836e3ce 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -11,7 +11,7 @@
 
 # To build with a different base image
 # please run `docker build` using the `--build-arg PYTORCH_IMAGE=...` flag.
-ARG PYTORCH_IMAGE=nvcr.io/nvidia/pytorch:24.03-py3
+ARG PYTORCH_IMAGE=nvcr.io/nvidia/pytorch:24.08-py3
 FROM ${PYTORCH_IMAGE}
 
 LABEL maintainer="monai.contact@gmail.com"

From c9f8d328fc0196ef166007a81dffc20a321f30af Mon Sep 17 00:00:00 2001
From: Boris Fomitchev <borisfom@users.noreply.github.com>
Date: Sun, 1 Sep 2024 07:04:41 -0700
Subject: [PATCH 141/183] Added TRTWrapper (#7990)

### Description

Added alternative class to ONNX->TRT export and wrap TRT engines for
inference.
It encapsulates filesystem persistence and does not rely on
torch-tensortrt for execution.
Also can be used to run ONNX with onnxruntime.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Co-authored-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: binliunls <107988372+binliunls@users.noreply.github.com>
---
 Dockerfile                        |   1 +
 docs/source/config_syntax.md      |  42 +++
 monai/bundle/config_parser.py     |   8 +-
 monai/bundle/scripts.py           |   4 +-
 monai/bundle/utils.py             |  36 +-
 monai/handlers/__init__.py        |   1 +
 monai/handlers/trt_handler.py     |  61 ++++
 monai/networks/__init__.py        |   2 +
 monai/networks/nets/swin_unetr.py |   8 +-
 monai/networks/trt_compiler.py    | 565 ++++++++++++++++++++++++++++++
 monai/networks/utils.py           | 264 +++++++++++---
 requirements-dev.txt              |   2 +
 setup.cfg                         |   3 +
 tests/min_tests.py                |   1 +
 tests/test_config_parser.py       |  32 ++
 tests/test_sure_loss.py           |   2 +-
 tests/test_trt_compile.py         | 140 ++++++++
 17 files changed, 1121 insertions(+), 51 deletions(-)
 create mode 100644 monai/handlers/trt_handler.py
 create mode 100644 monai/networks/trt_compiler.py
 create mode 100644 tests/test_trt_compile.py

diff --git a/Dockerfile b/Dockerfile
index e97836e3ce..e45932c6bb 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -56,4 +56,5 @@ RUN apt-get update \
   && rm -rf /var/lib/apt/lists/*
 # append /opt/tools to runtime path for NGC CLI to be accessible from all file system locations
 ENV PATH=${PATH}:/opt/tools
+ENV POLYGRAPHY_AUTOINSTALL_DEPS=1
 WORKDIR /opt/monai
diff --git a/docs/source/config_syntax.md b/docs/source/config_syntax.md
index c932879b5a..742841acca 100644
--- a/docs/source/config_syntax.md
+++ b/docs/source/config_syntax.md
@@ -16,6 +16,7 @@ Content:
   - [`$` to evaluate as Python expressions](#to-evaluate-as-python-expressions)
   - [`%` to textually replace configuration elements](#to-textually-replace-configuration-elements)
   - [`_target_` (`_disabled_`, `_desc_`, `_requires_`, `_mode_`) to instantiate a Python object](#instantiate-a-python-object)
+  - [`+` to alter semantics of merging config keys from multiple configuration files](#multiple-config-files)
 - [The command line interface](#the-command-line-interface)
 - [Recommendations](#recommendations)
 
@@ -175,6 +176,47 @@ _Description:_ `_requires_`, `_disabled_`, `_desc_`, and `_mode_` are optional k
   - `"debug"` -- execute with debug prompt and return the return value of ``pdb.runcall(_target_, **kwargs)``,
     see also [`pdb.runcall`](https://docs.python.org/3/library/pdb.html#pdb.runcall).
 
+## Multiple config files
+
+_Description:_ Multiple config files may be specified on the command line.
+The content of those config files is being merged. When same keys are specifiled in more than one config file,
+the value associated with the key is being overridden, in the order config files are specified.
+If the desired behaviour is to merge values from both files, the key in second config file should be prefixed with `+`.
+The value types for the merged contents must match and be both of `dict` or both of `list` type.
+`dict` values will be merged via update(), `list` values - concatenated via extend().
+Here's an example. In this case, "amp" value will be overridden by extra_config.json.
+`imports` and `preprocessing#transforms` lists will be merged. An error would be thrown if the value type in `"+imports"` is not `list`:
+
+config.json:
+```json
+{
+    "amp": "$True"
+    "imports": [
+	"$import torch"
+    ],
+    "preprocessing": {
+        "_target_": "Compose",
+        "transforms": [
+	  "$@t1",
+	  "$@t2"
+        ]
+    },
+}
+```
+
+extra_config.json:
+```json
+{
+    "amp": "$False"
+    "+imports": [
+	"$from monai.networks import trt_compile"
+    ],
+    "+preprocessing#transforms": [
+        "$@t3"
+    ]
+}
+```
+
 ## The command line interface
 
 In addition to the Pythonic APIs, a few command line interfaces (CLI) are provided to interact with the bundle.
diff --git a/monai/bundle/config_parser.py b/monai/bundle/config_parser.py
index a2ffeedc92..1d9920a230 100644
--- a/monai/bundle/config_parser.py
+++ b/monai/bundle/config_parser.py
@@ -20,7 +20,7 @@
 
 from monai.bundle.config_item import ComponentLocator, ConfigComponent, ConfigExpression, ConfigItem
 from monai.bundle.reference_resolver import ReferenceResolver
-from monai.bundle.utils import ID_REF_KEY, ID_SEP_KEY, MACRO_KEY
+from monai.bundle.utils import ID_REF_KEY, ID_SEP_KEY, MACRO_KEY, merge_kv
 from monai.config import PathLike
 from monai.utils import ensure_tuple, look_up_option, optional_import
 from monai.utils.misc import CheckKeyDuplicatesYamlLoader, check_key_duplicates
@@ -423,8 +423,10 @@ def load_config_files(cls, files: PathLike | Sequence[PathLike] | dict, **kwargs
         if isinstance(files, str) and not Path(files).is_file() and "," in files:
             files = files.split(",")
         for i in ensure_tuple(files):
-            for k, v in (cls.load_config_file(i, **kwargs)).items():
-                parser[k] = v
+            config_dict = cls.load_config_file(i, **kwargs)
+            for k, v in config_dict.items():
+                merge_kv(parser, k, v)
+
         return parser.get()  # type: ignore
 
     @classmethod
diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index 142a366669..f1d1286e4b 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -32,7 +32,7 @@
 from monai.apps.utils import _basename, download_url, extractall, get_logger
 from monai.bundle.config_item import ConfigComponent
 from monai.bundle.config_parser import ConfigParser
-from monai.bundle.utils import DEFAULT_INFERENCE, DEFAULT_METADATA
+from monai.bundle.utils import DEFAULT_INFERENCE, DEFAULT_METADATA, merge_kv
 from monai.bundle.workflows import BundleWorkflow, ConfigWorkflow
 from monai.config import IgniteInfo, PathLike
 from monai.data import load_net_with_metadata, save_net_with_metadata
@@ -105,7 +105,7 @@ def update_kwargs(args: str | dict | None = None, ignore_none: bool = True, **kw
         if isinstance(v, dict) and isinstance(args_.get(k), dict):
             args_[k] = update_kwargs(args_[k], ignore_none, **v)
         else:
-            args_[k] = v
+            merge_kv(args_, k, v)
     return args_
 
 
diff --git a/monai/bundle/utils.py b/monai/bundle/utils.py
index 50d2608f4c..53d619f234 100644
--- a/monai/bundle/utils.py
+++ b/monai/bundle/utils.py
@@ -13,6 +13,7 @@
 
 import json
 import os
+import warnings
 import zipfile
 from typing import Any
 
@@ -21,12 +22,21 @@
 
 yaml, _ = optional_import("yaml")
 
-__all__ = ["ID_REF_KEY", "ID_SEP_KEY", "EXPR_KEY", "MACRO_KEY", "DEFAULT_MLFLOW_SETTINGS", "DEFAULT_EXP_MGMT_SETTINGS"]
+__all__ = [
+    "ID_REF_KEY",
+    "ID_SEP_KEY",
+    "EXPR_KEY",
+    "MACRO_KEY",
+    "MERGE_KEY",
+    "DEFAULT_MLFLOW_SETTINGS",
+    "DEFAULT_EXP_MGMT_SETTINGS",
+]
 
 ID_REF_KEY = "@"  # start of a reference to a ConfigItem
 ID_SEP_KEY = "::"  # separator for the ID of a ConfigItem
 EXPR_KEY = "$"  # start of a ConfigExpression
 MACRO_KEY = "%"  # start of a macro of a config
+MERGE_KEY = "+"  # prefix indicating merge instead of override in case of multiple configs.
 
 _conf_values = get_config_values()
 
@@ -233,3 +243,27 @@ def load_bundle_config(bundle_path: str, *config_names: str, **load_kw_args: Any
             parser.read_config(f=cdata)
 
     return parser
+
+
+def merge_kv(args: dict | Any, k: str, v: Any) -> None:
+    """
+    Update the `args` dict-like object with the key/value pair `k` and `v`.
+    """
+    if k.startswith(MERGE_KEY):
+        """
+        Both values associated with `+`-prefixed key pair must be of `dict` or `list` type.
+        `dict` values will be merged, `list` values - concatenated.
+        """
+        id = k[1:]
+        if id in args:
+            if isinstance(v, dict) and isinstance(args[id], dict):
+                args[id].update(v)
+            elif isinstance(v, list) and isinstance(args[id], list):
+                args[id].extend(v)
+            else:
+                raise ValueError(ValueError(f"config must be dict or list for key `{k}`, but got {type(v)}: {v}."))
+        else:
+            warnings.warn(f"Can't merge entry ['{k}'], '{id}' is not in target dict - copying instead.")
+            args[id] = v
+    else:
+        args[k] = v
diff --git a/monai/handlers/__init__.py b/monai/handlers/__init__.py
index 641f9aae7d..fa6e158be8 100644
--- a/monai/handlers/__init__.py
+++ b/monai/handlers/__init__.py
@@ -40,5 +40,6 @@
 from .stats_handler import StatsHandler
 from .surface_distance import SurfaceDistance
 from .tensorboard_handlers import TensorBoardHandler, TensorBoardImageHandler, TensorBoardStatsHandler
+from .trt_handler import TrtHandler
 from .utils import from_engine, ignore_data, stopping_fn_from_loss, stopping_fn_from_metric, write_metrics_reports
 from .validation_handler import ValidationHandler
diff --git a/monai/handlers/trt_handler.py b/monai/handlers/trt_handler.py
new file mode 100644
index 0000000000..0e36b59d8c
--- /dev/null
+++ b/monai/handlers/trt_handler.py
@@ -0,0 +1,61 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+from monai.config import IgniteInfo
+from monai.networks import trt_compile
+from monai.utils import min_version, optional_import
+
+Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
+if TYPE_CHECKING:
+    from ignite.engine import Engine
+else:
+    Engine, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Engine")
+
+
+class TrtHandler:
+    """
+    TrtHandler acts as an Ignite handler to apply TRT acceleration to the model.
+    Usage example::
+        handler = TrtHandler(model=model, base_path="/test/checkpoint.pt", args={"precision": "fp16"})
+        handler.attach(engine)
+        engine.run()
+    """
+
+    def __init__(self, model, base_path, args=None, submodule=None):
+        """
+        Args:
+            base_path: TRT path basename. TRT plan(s) saved to "base_path[.submodule].plan"
+            args: passed to trt_compile(). See trt_compile() for details.
+            submodule : Hierarchical ids of submodules to convert, e.g. 'image_decoder.decoder'
+        """
+        self.model = model
+        self.base_path = base_path
+        self.args = args
+        self.submodule = submodule
+
+    def attach(self, engine: Engine) -> None:
+        """
+        Args:
+            engine: Ignite Engine, it can be a trainer, validator or evaluator.
+        """
+        self.logger = engine.logger
+        engine.add_event_handler(Events.STARTED, self)
+
+    def __call__(self, engine: Engine) -> None:
+        """
+        Args:
+            engine: Ignite Engine, it can be a trainer, validator or evaluator.
+        """
+        trt_compile(self.model, self.base_path, args=self.args, submodule=self.submodule, logger=self.logger)
diff --git a/monai/networks/__init__.py b/monai/networks/__init__.py
index 4c429ae813..5a240021d6 100644
--- a/monai/networks/__init__.py
+++ b/monai/networks/__init__.py
@@ -11,7 +11,9 @@
 
 from __future__ import annotations
 
+from .trt_compiler import trt_compile
 from .utils import (
+    add_casts_around_norms,
     convert_to_onnx,
     convert_to_torchscript,
     convert_to_trt,
diff --git a/monai/networks/nets/swin_unetr.py b/monai/networks/nets/swin_unetr.py
index 3900c866b3..714d986f4b 100644
--- a/monai/networks/nets/swin_unetr.py
+++ b/monai/networks/nets/swin_unetr.py
@@ -320,7 +320,7 @@ def _check_input_size(self, spatial_shape):
             )
 
     def forward(self, x_in):
-        if not torch.jit.is_scripting():
+        if not torch.jit.is_scripting() and not torch.jit.is_tracing():
             self._check_input_size(x_in.shape[2:])
         hidden_states_out = self.swinViT(x_in, self.normalize)
         enc0 = self.encoder1(x_in)
@@ -1046,14 +1046,14 @@ def __init__(
 
     def proj_out(self, x, normalize=False):
         if normalize:
-            x_shape = x.size()
+            x_shape = x.shape
+            # Force trace() to generate a constant by casting to int
+            ch = int(x_shape[1])
             if len(x_shape) == 5:
-                n, ch, d, h, w = x_shape
                 x = rearrange(x, "n c d h w -> n d h w c")
                 x = F.layer_norm(x, [ch])
                 x = rearrange(x, "n d h w c -> n c d h w")
             elif len(x_shape) == 4:
-                n, ch, h, w = x_shape
                 x = rearrange(x, "n c h w -> n h w c")
                 x = F.layer_norm(x, [ch])
                 x = rearrange(x, "n h w c -> n c h w")
diff --git a/monai/networks/trt_compiler.py b/monai/networks/trt_compiler.py
new file mode 100644
index 0000000000..a9dd0d9e9b
--- /dev/null
+++ b/monai/networks/trt_compiler.py
@@ -0,0 +1,565 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import inspect
+import os
+import tempfile
+import threading
+from collections import OrderedDict
+from pathlib import Path
+from types import MethodType
+from typing import Any, Dict, List, Union
+
+import torch
+
+from monai.apps.utils import get_logger
+from monai.networks.utils import add_casts_around_norms, convert_to_onnx, convert_to_torchscript, get_profile_shapes
+from monai.utils.module import optional_import
+
+polygraphy, polygraphy_imported = optional_import("polygraphy")
+if polygraphy_imported:
+    from polygraphy.backend.common import bytes_from_path
+    from polygraphy.backend.trt import (
+        CreateConfig,
+        Profile,
+        engine_bytes_from_network,
+        engine_from_bytes,
+        network_from_onnx_path,
+    )
+
+trt, trt_imported = optional_import("tensorrt")
+torch_tensorrt, _ = optional_import("torch_tensorrt", "1.4.0")
+cudart, _ = optional_import("cuda.cudart")
+
+
+lock_sm = threading.Lock()
+
+
+# Map of TRT dtype -> Torch dtype
+def trt_to_torch_dtype_dict():
+    return {
+        trt.int32: torch.int32,
+        trt.float32: torch.float32,
+        trt.float16: torch.float16,
+        trt.bfloat16: torch.float16,
+        trt.int64: torch.int64,
+        trt.int8: torch.int8,
+        trt.bool: torch.bool,
+    }
+
+
+def get_dynamic_axes(profiles):
+    """
+    This method calculates dynamic_axes to use in onnx.export().
+    Args:
+       profiles: [[min,opt,max],...] list of profile dimensions
+    """
+    dynamic_axes: dict[str, list[int]] = {}
+    if not profiles:
+        return dynamic_axes
+    for profile in profiles:
+        for key in profile:
+            axes = []
+            vals = profile[key]
+            for i in range(len(vals[0])):
+                if vals[0][i] != vals[2][i]:
+                    axes.append(i)
+            if len(axes) > 0:
+                dynamic_axes[key] = axes
+    return dynamic_axes
+
+
+def cuassert(cuda_ret):
+    """
+    Error reporting method for CUDA calls.
+    Args:
+     cuda_ret: CUDA return code.
+    """
+    err = cuda_ret[0]
+    if err != 0:
+        raise RuntimeError(f"CUDA ERROR: {err}")
+    if len(cuda_ret) > 1:
+        return cuda_ret[1]
+    return None
+
+
+class ShapeError(Exception):
+    """
+    Exception class to report errors from setting TRT plan input shapes
+    """
+
+    pass
+
+
+class TRTEngine:
+    """
+    An auxiliary class to implement running of TRT optimized engines
+
+    """
+
+    def __init__(self, plan_path, logger=None):
+        """
+        Loads serialized engine, creates execution context and activates it
+        Args:
+          plan_path: path to serialized TRT engine.
+          logger: optional logger object
+        """
+        self.plan_path = plan_path
+        self.logger = logger or get_logger("trt_compile")
+        self.logger.info(f"Loading TensorRT engine: {self.plan_path}")
+        self.engine = engine_from_bytes(bytes_from_path(self.plan_path))
+        self.tensors = OrderedDict()
+        self.cuda_graph_instance = None  # cuda graph
+        self.context = self.engine.create_execution_context()
+        self.input_names = []
+        self.output_names = []
+        self.dtypes = []
+        self.cur_profile = 0
+        dtype_dict = trt_to_torch_dtype_dict()
+        for idx in range(self.engine.num_io_tensors):
+            binding = self.engine[idx]
+            if self.engine.get_tensor_mode(binding) == trt.TensorIOMode.INPUT:
+                self.input_names.append(binding)
+            elif self.engine.get_tensor_mode(binding) == trt.TensorIOMode.OUTPUT:
+                self.output_names.append(binding)
+                dtype = dtype_dict[self.engine.get_tensor_dtype(binding)]
+                self.dtypes.append(dtype)
+
+    def allocate_buffers(self, device):
+        """
+        Allocates outputs to run TRT engine
+        Args:
+            device: GPU device to allocate memory on
+        """
+        ctx = self.context
+
+        for i, binding in enumerate(self.output_names):
+            shape = list(ctx.get_tensor_shape(binding))
+            if binding not in self.tensors or list(self.tensors[binding].shape) != shape:
+                t = torch.empty(shape, dtype=self.dtypes[i], device=device).contiguous()
+                self.tensors[binding] = t
+                ctx.set_tensor_address(binding, t.data_ptr())
+
+    def set_inputs(self, feed_dict, stream):
+        """
+        Sets input bindings for TRT engine according to feed_dict
+        Args:
+           feed_dict: a dictionary [str->Tensor]
+           stream: CUDA stream to use
+        """
+        e = self.engine
+        ctx = self.context
+
+        last_profile = self.cur_profile
+
+        def try_set_inputs():
+            for binding, t in feed_dict.items():
+                if t is not None:
+                    t = t.contiguous()
+                    shape = t.shape
+                    ctx.set_input_shape(binding, shape)
+                    ctx.set_tensor_address(binding, t.data_ptr())
+
+        while True:
+            try:
+                try_set_inputs()
+                break
+            except ShapeError:
+                next_profile = (self.cur_profile + 1) % e.num_optimization_profiles
+                if next_profile == last_profile:
+                    raise
+                self.cur_profile = next_profile
+                ctx.set_optimization_profile_async(self.cur_profile, stream)
+
+        left = ctx.infer_shapes()
+        assert len(left) == 0
+
+    def infer(self, stream, use_cuda_graph=False):
+        """
+        Runs TRT engine.
+        Args:
+            stream: CUDA stream to run on
+            use_cuda_graph: use CUDA graph. Note: requires all inputs to be the same GPU memory between calls.
+        """
+        if use_cuda_graph:
+            if self.cuda_graph_instance is not None:
+                cuassert(cudart.cudaGraphLaunch(self.cuda_graph_instance, stream))
+                cuassert(cudart.cudaStreamSynchronize(stream))
+            else:
+                # do inference before CUDA graph capture
+                noerror = self.context.execute_async_v3(stream)
+                if not noerror:
+                    raise ValueError("ERROR: inference failed.")
+                # capture cuda graph
+                cuassert(
+                    cudart.cudaStreamBeginCapture(stream, cudart.cudaStreamCaptureMode.cudaStreamCaptureModeThreadLocal)
+                )
+                self.context.execute_async_v3(stream)
+                graph = cuassert(cudart.cudaStreamEndCapture(stream))
+                self.cuda_graph_instance = cuassert(cudart.cudaGraphInstantiate(graph, 0))
+                self.logger.info("CUDA Graph captured!")
+        else:
+            noerror = self.context.execute_async_v3(stream)
+            cuassert(cudart.cudaStreamSynchronize(stream))
+            if not noerror:
+                raise ValueError("ERROR: inference failed.")
+
+        return self.tensors
+
+
+class TrtCompiler:
+    """
+    This class implements:
+      - TRT lazy persistent export
+      - Running TRT with optional fallback to Torch
+        (for TRT engines with limited profiles)
+    """
+
+    def __init__(
+        self,
+        model,
+        plan_path,
+        precision="fp16",
+        method="onnx",
+        input_names=None,
+        output_names=None,
+        export_args=None,
+        build_args=None,
+        input_profiles=None,
+        dynamic_batchsize=None,
+        use_cuda_graph=False,
+        timestamp=None,
+        fallback=False,
+        logger=None,
+    ):
+        """
+        Initialization method:
+         Tries to load persistent serialized TRT engine
+         Saves its arguments for lazy TRT build on first forward() call
+        Args:
+            model: Model to "wrap".
+            plan_path : Path where to save persistent serialized TRT engine.
+            precision: TRT builder precision o engine model. Should be 'fp32'|'tf32'|'fp16'|'bf16'.
+            method: One of 'onnx'|'torch_trt'.
+                    Default is 'onnx' (torch.onnx.export()->TRT). This is the most stable and efficient option.
+                    'torch_trt' may not work for some nets. Also AMP must be turned off for it to work.
+            input_names: Optional list of input names. If None, will be read from the function signature.
+            output_names: Optional list of output names. Note: If not None, patched forward() will return a dictionary.
+            export_args: Optional args to pass to export method. See onnx.export() and Torch-TensorRT docs for details.
+            build_args: Optional args to pass to TRT builder. See polygraphy.Config for details.
+            input_profiles: Optional list of profiles for TRT builder and ONNX export.
+                            Each profile is a map of the form : {"input id" : [min_shape, opt_shape, max_shape], ...}.
+            dynamic_batchsize: A sequence with three elements to define the batch size range of the input for the model to be
+                               converted. Should be a sequence like [MIN_BATCH, OPT_BATCH, MAX_BATCH].
+            [note]: If neither input_profiles nor dynamic_batchsize specified, static shapes will be used to build TRT engine.
+            use_cuda_graph: Use CUDA Graph for inference. Note: all inputs have to be the same GPU memory between calls!
+            timestamp: Optional timestamp to rebuild TRT engine (e.g. if config file changes).
+            fallback: Allow to fall back to Pytorch when TRT inference fails (e.g, shapes exceed max profile).
+        """
+
+        method_vals = ["onnx", "torch_trt"]
+        if method not in method_vals:
+            raise ValueError(f"trt_compile(): 'method' should be one of {method_vals}, got: {method}.")
+        precision_vals = ["fp32", "tf32", "fp16", "bf16"]
+        if precision not in precision_vals:
+            raise ValueError(f"trt_compile(): 'precision' should be one of {precision_vals}, got: {precision}.")
+
+        self.plan_path = plan_path
+        self.precision = precision
+        self.method = method
+        self.return_dict = output_names is not None
+        self.output_names = output_names or []
+        self.profiles = input_profiles or []
+        self.dynamic_batchsize = dynamic_batchsize
+        self.export_args = export_args or {}
+        self.build_args = build_args or {}
+        self.engine: TRTEngine | None = None
+        self.use_cuda_graph = use_cuda_graph
+        self.fallback = fallback
+        self.disabled = False
+
+        self.logger = logger or get_logger("trt_compile")
+
+        # Normally we read input_names from forward() but can be overridden
+        if input_names is None:
+            argspec = inspect.getfullargspec(model.forward)
+            input_names = argspec.args[1:]
+        self.input_names = input_names
+        self.old_forward = model.forward
+
+        # Force engine rebuild if older than the timestamp
+        if timestamp is not None and os.path.exists(self.plan_path) and os.path.getmtime(self.plan_path) < timestamp:
+            os.remove(self.plan_path)
+
+    def _inputs_to_dict(self, input_example):
+        trt_inputs = {}
+        for i, inp in enumerate(input_example):
+            input_name = self.input_names[i]
+            trt_inputs[input_name] = inp
+        return trt_inputs
+
+    def _load_engine(self):
+        """
+        Loads TRT plan from disk and activates its execution context.
+        """
+        try:
+            self.engine = TRTEngine(self.plan_path, self.logger)
+            self.input_names = self.engine.input_names
+        except Exception as e:
+            self.logger.debug(f"Exception while loading the engine:\n{e}")
+
+    def forward(self, model, argv, kwargs):
+        """
+        Main forward method:
+         Builds TRT engine if not available yet.
+         Tries to run TRT engine
+         If exception thrown and self.callback==True: falls back to original Pytorch
+
+        Args: Passing through whatever args wrapped module's forward() has
+        Returns: Passing through wrapped module's forward() return value(s)
+
+        """
+        if self.engine is None and not self.disabled:
+            # Restore original forward for export
+            new_forward = model.forward
+            model.forward = self.old_forward
+            try:
+                self._load_engine()
+                if self.engine is None:
+                    build_args = kwargs.copy()
+                    if len(argv) > 0:
+                        build_args.update(self._inputs_to_dict(argv))
+                    self._build_and_save(model, build_args)
+                    # This will reassign input_names from the engine
+                    self._load_engine()
+            except Exception as e:
+                if self.fallback:
+                    self.logger.info(f"Failed to build engine: {e}")
+                    self.disabled = True
+                else:
+                    raise e
+            if not self.disabled and not self.fallback:
+                # Delete all parameters
+                for param in model.parameters():
+                    del param
+                # Call empty_cache to release GPU memory
+                torch.cuda.empty_cache()
+            model.forward = new_forward
+        # Run the engine
+        try:
+            if len(argv) > 0:
+                kwargs.update(self._inputs_to_dict(argv))
+                argv = ()
+
+            if self.engine is not None:
+                # forward_trt is not thread safe as we do not use per-thread execution contexts
+                with lock_sm:
+                    device = torch.cuda.current_device()
+                    stream = torch.cuda.Stream(device=device)
+                    self.engine.set_inputs(kwargs, stream.cuda_stream)
+                    self.engine.allocate_buffers(device=device)
+                    # Need this to synchronize with Torch stream
+                    stream.wait_stream(torch.cuda.current_stream())
+                    ret = self.engine.infer(stream.cuda_stream, use_cuda_graph=self.use_cuda_graph)
+                    # if output_names is not None, return dictionary
+                    if not self.return_dict:
+                        ret = list(ret.values())
+                        if len(ret) == 1:
+                            ret = ret[0]
+                    return ret
+        except Exception as e:
+            if model is not None:
+                self.logger.info(f"Exception: {e}\nFalling back to Pytorch ...")
+            else:
+                raise e
+        return self.old_forward(*argv, **kwargs)
+
+    def _onnx_to_trt(self, onnx_path):
+        """
+        Builds TRT engine from ONNX file at onnx_path and saves to self.plan_path
+        """
+
+        profiles = []
+        if self.profiles:
+            for input_profile in self.profiles:
+                if isinstance(input_profile, Profile):
+                    profiles.append(input_profile)
+                else:
+                    p = Profile()
+                    for name, dims in input_profile.items():
+                        assert len(dims) == 3
+                        p.add(name, min=dims[0], opt=dims[1], max=dims[2])
+                    profiles.append(p)
+
+        build_args = self.build_args.copy()
+        build_args["tf32"] = self.precision != "fp32"
+        build_args["fp16"] = self.precision == "fp16"
+        build_args["bf16"] = self.precision == "bf16"
+
+        self.logger.info(f"Building TensorRT engine for {onnx_path}: {self.plan_path}")
+        network = network_from_onnx_path(onnx_path, flags=[trt.OnnxParserFlag.NATIVE_INSTANCENORM])
+        return engine_bytes_from_network(network, config=CreateConfig(profiles=profiles, **build_args))
+
+    def _build_and_save(self, model, input_example):
+        """
+        If TRT engine is not ready, exports model to ONNX,
+        builds TRT engine and saves serialized TRT engine to the disk.
+        Args:
+             input_example: passed to onnx.export()
+        """
+
+        if self.engine is not None:
+            return
+
+        export_args = self.export_args
+
+        add_casts_around_norms(model)
+
+        if self.method == "torch_trt":
+            enabled_precisions = [torch.float32]
+            if self.precision == "fp16":
+                enabled_precisions.append(torch.float16)
+            elif self.precision == "bf16":
+                enabled_precisions.append(torch.bfloat16)
+            inputs = list(input_example.values())
+            ir_model = convert_to_torchscript(model, inputs=inputs, use_trace=True)
+
+            def get_torch_trt_input(input_shape, dynamic_batchsize):
+                min_input_shape, opt_input_shape, max_input_shape = get_profile_shapes(input_shape, dynamic_batchsize)
+                return torch_tensorrt.Input(
+                    min_shape=min_input_shape, opt_shape=opt_input_shape, max_shape=max_input_shape
+                )
+
+            tt_inputs = [get_torch_trt_input(i.shape, self.dynamic_batchsize) for i in inputs]
+            engine_bytes = torch_tensorrt.convert_method_to_trt_engine(
+                ir_model,
+                "forward",
+                inputs=tt_inputs,
+                ir="torchscript",
+                enabled_precisions=enabled_precisions,
+                **export_args,
+            )
+        else:
+            dbs = self.dynamic_batchsize
+            if dbs:
+                if len(self.profiles) > 0:
+                    raise ValueError("ERROR: Both dynamic_batchsize and input_profiles set for TrtCompiler!")
+                if len(dbs) != 3:
+                    raise ValueError("dynamic_batchsize has to have len ==3 ")
+                profiles = {}
+                for id, val in input_example.items():
+                    sh = val.shape[1:]
+                    profiles[id] = [[dbs[0], *sh], [dbs[1], *sh], [dbs[2], *sh]]
+                self.profiles = [profiles]
+
+            if len(self.profiles) > 0:
+                export_args.update({"dynamic_axes": get_dynamic_axes(self.profiles)})
+
+            # Use temporary directory for easy cleanup in case of external weights
+            with tempfile.TemporaryDirectory() as tmpdir:
+                onnx_path = Path(tmpdir) / "model.onnx"
+                self.logger.info(
+                    f"Exporting to {onnx_path}:\n\toutput_names={self.output_names}\n\texport args: {export_args}"
+                )
+                convert_to_onnx(
+                    model,
+                    input_example,
+                    filename=str(onnx_path),
+                    input_names=self.input_names,
+                    output_names=self.output_names,
+                    **export_args,
+                )
+                self.logger.info("Export to ONNX successful.")
+                engine_bytes = self._onnx_to_trt(str(onnx_path))
+
+        open(self.plan_path, "wb").write(engine_bytes)
+
+
+def trt_forward(self, *argv, **kwargs):
+    """
+    Patch function to replace original model's forward() with.
+    Redirects to TrtCompiler.forward()
+    """
+    return self._trt_compiler.forward(self, argv, kwargs)
+
+
+def trt_compile(
+    model: torch.nn.Module,
+    base_path: str,
+    args: Dict[str, Any] | None = None,
+    submodule: Union[str, List[str]] | None = None,
+    logger: Any | None = None,
+) -> torch.nn.Module:
+    """
+    Instruments model or submodule(s) with TrtCompiler and replaces its forward() with TRT hook.
+    Args:
+      model: module to patch with TrtCompiler object.
+      base_path: TRT plan(s) saved to f"{base_path}[.{submodule}].plan" path.
+                 dirname(base_path) must exist, base_path does not have to.
+                 If base_path does point to existing file (e.g. associated checkpoint),
+                 that file becomes a dependency - its mtime is added to args["timestamp"].
+      args: Optional dict : unpacked and passed to TrtCompiler() - see TrtCompiler above for details.
+      submodule: Optional hierarchical id(s) of submodule to patch, e.g. ['image_decoder.decoder']
+                  If None, TrtCompiler patch is applied to the whole model.
+                  Otherwise, submodule (or list of) is being patched.
+      logger: Optional logger for diagnostics.
+    Returns:
+      Always returns same model passed in as argument. This is for ease of use in configs.
+    """
+
+    default_args: Dict[str, Any] = {
+        "method": "onnx",
+        "precision": "fp16",
+        "build_args": {"builder_optimization_level": 5, "precision_constraints": "obey"},
+    }
+
+    default_args.update(args or {})
+    args = default_args
+
+    if trt_imported and polygraphy_imported and torch.cuda.is_available():
+        # if "path" filename point to existing file (e.g. checkpoint)
+        # it's also treated as dependency
+        if os.path.exists(base_path):
+            timestamp = int(os.path.getmtime(base_path))
+            if "timestamp" in args:
+                timestamp = max(int(args["timestamp"]), timestamp)
+            args["timestamp"] = timestamp
+
+        def wrap(model, path):
+            wrapper = TrtCompiler(model, path + ".plan", logger=logger, **args)
+            model._trt_compiler = wrapper
+            model.forward = MethodType(trt_forward, model)
+
+        def find_sub(parent, submodule):
+            idx = submodule.find(".")
+            # if there is "." in name, call recursively
+            if idx != -1:
+                parent_name = submodule[:idx]
+                parent = getattr(parent, parent_name)
+                submodule = submodule[idx + 1 :]
+                return find_sub(parent, submodule)
+            return parent, submodule
+
+        if submodule is not None:
+            if isinstance(submodule, str):
+                submodule = [submodule]
+            for s in submodule:
+                parent, sub = find_sub(model, s)
+                wrap(getattr(parent, sub), base_path + "." + s)
+        else:
+            wrap(model, base_path)
+    else:
+        logger = logger or get_logger("trt_compile")
+        logger.warning("TensorRT and/or polygraphy packages are not available! trt_compile() has no effect.")
+
+    return model
diff --git a/monai/networks/utils.py b/monai/networks/utils.py
index bd65ffa33e..d0150b4e5b 100644
--- a/monai/networks/utils.py
+++ b/monai/networks/utils.py
@@ -36,6 +36,8 @@
 onnx, _ = optional_import("onnx")
 onnxreference, _ = optional_import("onnx.reference")
 onnxruntime, _ = optional_import("onnxruntime")
+polygraphy, polygraphy_imported = optional_import("polygraphy")
+torch_tensorrt, _ = optional_import("torch_tensorrt", "1.4.0")
 
 __all__ = [
     "one_hot",
@@ -61,6 +63,7 @@
     "look_up_named_module",
     "set_named_module",
     "has_nvfuser_instance_norm",
+    "get_profile_shapes",
 ]
 
 logger = get_logger(module_name=__name__)
@@ -68,6 +71,26 @@
 _has_nvfuser = None
 
 
+def get_profile_shapes(input_shape: Sequence[int], dynamic_batchsize: Sequence[int] | None):
+    """
+    Given a sample input shape, calculate min/opt/max shapes according to dynamic_batchsize.
+    """
+
+    def scale_batch_size(input_shape: Sequence[int], scale_num: int):
+        scale_shape = [*input_shape]
+        scale_shape[0] = scale_num
+        return scale_shape
+
+    # Use the dynamic batchsize range to generate the min, opt and max model input shape
+    if dynamic_batchsize:
+        min_input_shape = scale_batch_size(input_shape, dynamic_batchsize[0])
+        opt_input_shape = scale_batch_size(input_shape, dynamic_batchsize[1])
+        max_input_shape = scale_batch_size(input_shape, dynamic_batchsize[2])
+    else:
+        min_input_shape = opt_input_shape = max_input_shape = input_shape
+    return min_input_shape, opt_input_shape, max_input_shape
+
+
 def has_nvfuser_instance_norm():
     """whether the current environment has InstanceNorm3dNVFuser
     https://github.com/NVIDIA/apex/blob/23.05-devel/apex/normalization/instance_norm.py#L15-L16
@@ -606,6 +629,9 @@ def convert_to_onnx(
     rtol: float = 1e-4,
     atol: float = 0.0,
     use_trace: bool = True,
+    do_constant_folding: bool = True,
+    constant_size_threshold: int = 16 * 1024 * 1024 * 1024,
+    dynamo=False,
     **kwargs,
 ):
     """
@@ -632,7 +658,10 @@ def convert_to_onnx(
         rtol: the relative tolerance when comparing the outputs of PyTorch model and TorchScript model.
         atol: the absolute tolerance when comparing the outputs of PyTorch model and TorchScript model.
         use_trace: whether to use `torch.jit.trace` to export the torchscript model.
-        kwargs: other arguments except `obj` for `torch.jit.script()` to convert model, for more details:
+        do_constant_folding: passed to onnx.export(). If True, extra polygraphy folding pass is done.
+        constant_size_threshold: passed to polygrapy conatant forling, default = 16M
+        kwargs: if use_trace=True: additional arguments to pass to torch.onnx.export()
+            else: other arguments except `obj` for `torch.jit.script()` to convert model, for more details:
             https://pytorch.org/docs/master/generated/torch.jit.script.html.
 
     """
@@ -642,6 +671,7 @@ def convert_to_onnx(
         if use_trace:
             # let torch.onnx.export to trace the model.
             mode_to_export = model
+            torch_versioned_kwargs = kwargs
         else:
             if not pytorch_after(1, 10):
                 if "example_outputs" not in kwargs:
@@ -654,32 +684,37 @@ def convert_to_onnx(
                 del kwargs["example_outputs"]
             mode_to_export = torch.jit.script(model, **kwargs)
 
+        if torch.is_tensor(inputs) or isinstance(inputs, dict):
+            onnx_inputs = (inputs,)
+        else:
+            onnx_inputs = tuple(inputs)
+
         if filename is None:
             f = io.BytesIO()
-            torch.onnx.export(
-                mode_to_export,
-                tuple(inputs),
-                f=f,
-                input_names=input_names,
-                output_names=output_names,
-                dynamic_axes=dynamic_axes,
-                opset_version=opset_version,
-                **torch_versioned_kwargs,
-            )
+        else:
+            f = filename
+
+        torch.onnx.export(
+            mode_to_export,
+            onnx_inputs,
+            f=f,
+            input_names=input_names,
+            output_names=output_names,
+            dynamic_axes=dynamic_axes,
+            opset_version=opset_version,
+            do_constant_folding=do_constant_folding,
+            **torch_versioned_kwargs,
+        )
+        if filename is None:
             onnx_model = onnx.load_model_from_string(f.getvalue())
         else:
-            torch.onnx.export(
-                mode_to_export,
-                tuple(inputs),
-                f=filename,
-                input_names=input_names,
-                output_names=output_names,
-                dynamic_axes=dynamic_axes,
-                opset_version=opset_version,
-                **torch_versioned_kwargs,
-            )
             onnx_model = onnx.load(filename)
 
+    if do_constant_folding and polygraphy_imported:
+        from polygraphy.backend.onnx.loader import fold_constants
+
+        fold_constants(onnx_model, size_threshold=constant_size_threshold)
+
     if verify:
         if device is None:
             device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
@@ -814,7 +849,6 @@ def _onnx_trt_compile(
 
     """
     trt, _ = optional_import("tensorrt", "8.5.3")
-    torch_tensorrt, _ = optional_import("torch_tensorrt", "1.4.0")
 
     input_shapes = (min_shape, opt_shape, max_shape)
     # default to an empty list to fit the `torch_tensorrt.ts.embed_engine_in_new_module` function.
@@ -916,8 +950,6 @@ def convert_to_trt(
             to compile model, for more details: https://pytorch.org/TensorRT/py_api/torch_tensorrt.html#torch-tensorrt-py.
     """
 
-    torch_tensorrt, _ = optional_import("torch_tensorrt", version="1.4.0")
-
     if not torch.cuda.is_available():
         raise Exception("Cannot find any GPU devices.")
 
@@ -935,23 +967,9 @@ def convert_to_trt(
     convert_precision = torch.float32 if precision == "fp32" else torch.half
     inputs = [torch.rand(ensure_tuple(input_shape)).to(target_device)]
 
-    def scale_batch_size(input_shape: Sequence[int], scale_num: int):
-        scale_shape = [*input_shape]
-        scale_shape[0] *= scale_num
-        return scale_shape
-
-    # Use the dynamic batchsize range to generate the min, opt and max model input shape
-    if dynamic_batchsize:
-        min_input_shape = scale_batch_size(input_shape, dynamic_batchsize[0])
-        opt_input_shape = scale_batch_size(input_shape, dynamic_batchsize[1])
-        max_input_shape = scale_batch_size(input_shape, dynamic_batchsize[2])
-    else:
-        min_input_shape = opt_input_shape = max_input_shape = input_shape
-
     # convert the torch model to a TorchScript model on target device
     model = model.eval().to(target_device)
-    ir_model = convert_to_torchscript(model, device=target_device, inputs=inputs, use_trace=use_trace)
-    ir_model.eval()
+    min_input_shape, opt_input_shape, max_input_shape = get_profile_shapes(input_shape, dynamic_batchsize)
 
     if use_onnx:
         # set the batch dim as dynamic
@@ -960,7 +978,6 @@ def scale_batch_size(input_shape: Sequence[int], scale_num: int):
         ir_model = convert_to_onnx(
             model, inputs, onnx_input_names, onnx_output_names, use_trace=use_trace, dynamic_axes=dynamic_axes
         )
-
         # convert the model through the ONNX-TensorRT way
         trt_model = _onnx_trt_compile(
             ir_model,
@@ -973,6 +990,8 @@ def scale_batch_size(input_shape: Sequence[int], scale_num: int):
             output_names=onnx_output_names,
         )
     else:
+        ir_model = convert_to_torchscript(model, device=target_device, inputs=inputs, use_trace=use_trace)
+        ir_model.eval()
         # convert the model through the Torch-TensorRT way
         ir_model.to(target_device)
         with torch.no_grad():
@@ -1189,3 +1208,168 @@ def forward(self, x):
         if dtype == self.initial_type:
             x = x.to(self.initial_type)
         return x
+
+
+def cast_tensor(x, from_dtype=torch.float16, to_dtype=torch.float32):
+    """
+    Utility function to cast a single tensor from from_dtype to to_dtype
+    """
+    return x.to(dtype=to_dtype) if x.dtype == from_dtype else x
+
+
+def cast_all(x, from_dtype=torch.float16, to_dtype=torch.float32):
+    """
+    Utility function to cast all tensors in a tuple from from_dtype to to_dtype
+    """
+    if isinstance(x, torch.Tensor):
+        return cast_tensor(x, from_dtype=from_dtype, to_dtype=to_dtype)
+    else:
+        if isinstance(x, dict):
+            new_dict = {}
+            for k in x.keys():
+                new_dict[k] = cast_all(x[k], from_dtype=from_dtype, to_dtype=to_dtype)
+            return new_dict
+        elif isinstance(x, tuple):
+            return tuple(cast_all(y, from_dtype=from_dtype, to_dtype=to_dtype) for y in x)
+
+
+class CastToFloat(torch.nn.Module):
+    """
+    Class used to add autocast protection for ONNX export
+    for forward methods with single return vaue
+    """
+
+    def __init__(self, mod):
+        super().__init__()
+        self.mod = mod
+
+    def forward(self, x):
+        dtype = x.dtype
+        with torch.amp.autocast("cuda", enabled=False):
+            ret = self.mod.forward(x.to(torch.float32)).to(dtype)
+        return ret
+
+
+class CastToFloatAll(torch.nn.Module):
+    """
+    Class used to add autocast protection for ONNX export
+    for forward methods with multiple return values
+    """
+
+    def __init__(self, mod):
+        super().__init__()
+        self.mod = mod
+
+    def forward(self, *args):
+        from_dtype = args[0].dtype
+        with torch.amp.autocast("cuda", enabled=False):
+            ret = self.mod.forward(*cast_all(args, from_dtype=from_dtype, to_dtype=torch.float32))
+        return cast_all(ret, from_dtype=torch.float32, to_dtype=from_dtype)
+
+
+def wrap_module(base_t: type[nn.Module], dest_t: type[nn.Module]) -> Callable[[nn.Module], nn.Module | None]:
+    """
+    Generic function generator to replace base_t module with dest_t wrapper.
+    Args:
+        base_t : module type to replace
+        dest_t : destination module type
+    Returns:
+        swap function to replace base_t module with dest_t
+    """
+
+    def expansion_fn(mod: nn.Module) -> nn.Module | None:
+        out = dest_t(mod)
+        return out
+
+    return expansion_fn
+
+
+def simple_replace(base_t: type[nn.Module], dest_t: type[nn.Module]) -> Callable[[nn.Module], nn.Module | None]:
+    """
+    Generic function generator to replace base_t module with dest_t.
+    base_t and dest_t should have same atrributes. No weights are copied.
+    Args:
+        base_t : module type to replace
+        dest_t : destination module type
+    Returns:
+        swap function to replace base_t module with dest_t
+    """
+
+    def expansion_fn(mod: nn.Module) -> nn.Module | None:
+        if not isinstance(mod, base_t):
+            return None
+        args = [getattr(mod, name, None) for name in mod.__constants__]
+        out = dest_t(*args)
+        return out
+
+    return expansion_fn
+
+
+def _swap_modules(model: nn.Module, mapping: dict[str, nn.Module]) -> nn.Module:
+    """
+    This function swaps nested modules as specified by "dot paths" in mod with a desired replacement. This allows
+    for swapping nested modules through arbitrary levels if children
+
+    NOTE: This occurs in place, if you want to preserve model then make sure to copy it first.
+
+    """
+    for path, new_mod in mapping.items():
+        expanded_path = path.split(".")
+        parent_mod = model
+        for sub_path in expanded_path[:-1]:
+            submod = parent_mod._modules[sub_path]
+            if submod is None:
+                break
+            else:
+                parent_mod = submod
+        parent_mod._modules[expanded_path[-1]] = new_mod
+
+    return model
+
+
+def replace_modules_by_type(
+    model: nn.Module, expansions: dict[str, Callable[[nn.Module], nn.Module | None]]
+) -> nn.Module:
+    """
+    Top-level function to replace modules in model, specified by class name with a desired replacement.
+    NOTE: This occurs in place, if you want to preserve model then make sure to copy it first.
+    Args:
+        model : top level module
+        expansions : replacement dictionary: module class name -> replacement function generator
+    Returns:
+        model, possibly modified in-place
+    """
+    mapping: dict[str, nn.Module] = {}
+    for name, m in model.named_modules():
+        m_type = type(m).__name__
+        if m_type in expansions:
+            # print (f"Found {m_type} in expansions ...")
+            swapped = expansions[m_type](m)
+            if swapped:
+                mapping[name] = swapped
+
+    print(f"Swapped {len(mapping)} modules")
+    _swap_modules(model, mapping)
+    return model
+
+
+def add_casts_around_norms(model: nn.Module) -> nn.Module:
+    """
+    Top-level function to add cast wrappers around modules known to cause issues for FP16/autocast ONNX export
+    NOTE: This occurs in place, if you want to preserve model then make sure to copy it first.
+    Args:
+        model : top level module
+    Returns:
+        model, possibly modified in-place
+    """
+    print("Adding casts around norms...")
+    cast_replacements = {
+        "BatchNorm1d": wrap_module(nn.BatchNorm1d, CastToFloat),
+        "BatchNorm2d": wrap_module(nn.BatchNorm2d, CastToFloat),
+        "BatchNorm3d": wrap_module(nn.BatchNorm2d, CastToFloat),
+        "LayerNorm": wrap_module(nn.LayerNorm, CastToFloat),
+        "InstanceNorm1d": wrap_module(nn.InstanceNorm1d, CastToFloat),
+        "InstanceNorm3d": wrap_module(nn.InstanceNorm3d, CastToFloat),
+    }
+    replace_modules_by_type(model, cast_replacements)
+    return model
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 9aad0804e6..6d0ccd378a 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -59,3 +59,5 @@ nvidia-ml-py
 huggingface_hub
 pyamg>=5.0.0
 git+https://github.com/facebookresearch/segment-anything.git@6fdee8f2727f4506cfbbe553e23b895e27956588
+onnx_graphsurgeon
+polygraphy
diff --git a/setup.cfg b/setup.cfg
index 1ce4a3f34c..c97118d43a 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -160,6 +160,9 @@ lpips =
     lpips==0.1.4
 pynvml =
     nvidia-ml-py
+polygraphy =
+    polygraphy
+
 # # workaround https://github.com/Project-MONAI/MONAI/issues/5882
 # MetricsReloaded =
     # MetricsReloaded @ git+https://github.com/Project-MONAI/MetricsReloaded@monai-support#egg=MetricsReloaded
diff --git a/tests/min_tests.py b/tests/min_tests.py
index f80d06f5d3..632355b5c6 100644
--- a/tests/min_tests.py
+++ b/tests/min_tests.py
@@ -186,6 +186,7 @@ def run_testsuit():
         "test_torchvisiond",
         "test_transchex",
         "test_transformerblock",
+        "test_trt_compile",
         "test_unetr",
         "test_unetr_block",
         "test_vit",
diff --git a/tests/test_config_parser.py b/tests/test_config_parser.py
index cf1edc8f08..2b00c9f9d1 100644
--- a/tests/test_config_parser.py
+++ b/tests/test_config_parser.py
@@ -125,6 +125,22 @@ def __call__(self, a, b):
     [0, 4],
 ]
 
+TEST_CASE_MERGE_JSON = ["""{"key1": [0], "key2": [0] }""", """{"key1": [1], "+key2": [4] }""", "json", [1], [0, 4]]
+
+TEST_CASE_MERGE_YAML = [
+    """
+    key1: 0
+    key2: [0]
+    """,
+    """
+    key1: 1
+    +key2: [4]
+    """,
+    "yaml",
+    1,
+    [0, 4],
+]
+
 
 class TestConfigParser(unittest.TestCase):
 
@@ -357,6 +373,22 @@ def test_parse_json_warn(self, config_string, extension, expected_unique_val, ex
             self.assertEqual(parser.get_parsed_content("key#unique"), expected_unique_val)
             self.assertIn(parser.get_parsed_content("key#duplicate"), expected_duplicate_vals)
 
+    @parameterized.expand([TEST_CASE_MERGE_JSON, TEST_CASE_MERGE_YAML])
+    @skipUnless(has_yaml, "Requires pyyaml")
+    def test_load_configs(
+        self, config_string, config_string2, extension, expected_overridden_val, expected_merged_vals
+    ):
+        with tempfile.TemporaryDirectory() as tempdir:
+            config_path1 = Path(tempdir) / f"config1.{extension}"
+            config_path2 = Path(tempdir) / f"config2.{extension}"
+            config_path1.write_text(config_string)
+            config_path2.write_text(config_string2)
+
+            parser = ConfigParser.load_config_files([config_path1, config_path2])
+
+            self.assertEqual(parser["key1"], expected_overridden_val)
+            self.assertEqual(parser["key2"], expected_merged_vals)
+
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/tests/test_sure_loss.py b/tests/test_sure_loss.py
index 903f9bd2ca..fb8f5dda72 100644
--- a/tests/test_sure_loss.py
+++ b/tests/test_sure_loss.py
@@ -65,7 +65,7 @@ def operator(x):
 
         loss_real = sure_loss_real(operator, x_real, y_pseudo_gt_real, complex_input=False)
         loss_complex = sure_loss_complex(operator, x_complex, y_pseudo_gt_complex, complex_input=True)
-        self.assertAlmostEqual(loss_real.item(), loss_complex.abs().item(), places=6)
+        self.assertAlmostEqual(loss_real.item(), loss_complex.abs().item(), places=5)
 
 
 if __name__ == "__main__":
diff --git a/tests/test_trt_compile.py b/tests/test_trt_compile.py
new file mode 100644
index 0000000000..21125d203f
--- /dev/null
+++ b/tests/test_trt_compile.py
@@ -0,0 +1,140 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import tempfile
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.handlers import TrtHandler
+from monai.networks import trt_compile
+from monai.networks.nets import UNet, cell_sam_wrapper, vista3d132
+from monai.utils import optional_import
+from tests.utils import skip_if_no_cuda, skip_if_quick, skip_if_windows
+
+trt, trt_imported = optional_import("tensorrt")
+polygraphy, polygraphy_imported = optional_import("polygraphy")
+build_sam_vit_b, has_sam = optional_import("segment_anything.build_sam", name="build_sam_vit_b")
+
+TEST_CASE_1 = ["fp32"]
+TEST_CASE_2 = ["fp16"]
+
+
+@skip_if_windows
+@skip_if_no_cuda
+@skip_if_quick
+@unittest.skipUnless(trt_imported, "tensorrt is required")
+@unittest.skipUnless(polygraphy_imported, "polygraphy is required")
+class TestTRTCompile(unittest.TestCase):
+
+    def setUp(self):
+        self.gpu_device = torch.cuda.current_device()
+
+    def tearDown(self):
+        current_device = torch.cuda.current_device()
+        if current_device != self.gpu_device:
+            torch.cuda.set_device(self.gpu_device)
+
+    def test_handler(self):
+        from ignite.engine import Engine
+
+        net1 = torch.nn.Sequential(*[torch.nn.PReLU(), torch.nn.PReLU()])
+        data1 = net1.state_dict()
+        data1["0.weight"] = torch.tensor([0.1])
+        data1["1.weight"] = torch.tensor([0.2])
+        net1.load_state_dict(data1)
+        net1.cuda()
+
+        with tempfile.TemporaryDirectory() as tempdir:
+            engine = Engine(lambda e, b: None)
+            args = {"method": "torch_trt"}
+            TrtHandler(net1, tempdir + "/trt_handler", args=args).attach(engine)
+            engine.run([0] * 8, max_epochs=1)
+            self.assertIsNotNone(net1._trt_compiler)
+            self.assertIsNone(net1._trt_compiler.engine)
+            net1.forward(torch.tensor([[0.0, 1.0], [1.0, 2.0]], device="cuda"))
+            self.assertIsNotNone(net1._trt_compiler.engine)
+
+    @parameterized.expand([TEST_CASE_1, TEST_CASE_2])
+    def test_unet_value(self, precision):
+        model = UNet(
+            spatial_dims=3,
+            in_channels=1,
+            out_channels=2,
+            channels=(2, 2, 4, 8, 4),
+            strides=(2, 2, 2, 2),
+            num_res_units=2,
+            norm="batch",
+        ).cuda()
+        with torch.no_grad(), tempfile.TemporaryDirectory() as tmpdir:
+            model.eval()
+            input_example = torch.randn(2, 1, 96, 96, 96).cuda()
+            output_example = model(input_example)
+            args: dict = {"builder_optimization_level": 1}
+            trt_compile(
+                model,
+                f"{tmpdir}/test_unet_trt_compile",
+                args={"precision": precision, "build_args": args, "dynamic_batchsize": [1, 4, 8]},
+            )
+            self.assertIsNone(model._trt_compiler.engine)
+            trt_output = model(input_example)
+            # Check that lazy TRT build succeeded
+            self.assertIsNotNone(model._trt_compiler.engine)
+            torch.testing.assert_close(trt_output, output_example, rtol=0.01, atol=0.01)
+
+    @parameterized.expand([TEST_CASE_1, TEST_CASE_2])
+    @unittest.skipUnless(has_sam, "Requires SAM installation")
+    def test_cell_sam_wrapper_value(self, precision):
+        model = cell_sam_wrapper.CellSamWrapper(checkpoint=None).to("cuda")
+        with torch.no_grad(), tempfile.TemporaryDirectory() as tmpdir:
+            model.eval()
+            input_example = torch.randn(1, 3, 128, 128).to("cuda")
+            output_example = model(input_example)
+            trt_compile(
+                model,
+                f"{tmpdir}/test_cell_sam_wrapper_trt_compile",
+                args={"precision": precision, "dynamic_batchsize": [1, 1, 1]},
+            )
+            self.assertIsNone(model._trt_compiler.engine)
+            trt_output = model(input_example)
+            # Check that lazy TRT build succeeded
+            self.assertIsNotNone(model._trt_compiler.engine)
+            torch.testing.assert_close(trt_output, output_example, rtol=0.01, atol=0.01)
+
+    @parameterized.expand([TEST_CASE_1, TEST_CASE_2])
+    def test_vista3d(self, precision):
+        model = vista3d132(in_channels=1).to("cuda")
+        with torch.no_grad(), tempfile.TemporaryDirectory() as tmpdir:
+            model.eval()
+            input_example = torch.randn(1, 1, 64, 64, 64).to("cuda")
+            output_example = model(input_example)
+            model = trt_compile(
+                model,
+                f"{tmpdir}/test_vista3d_trt_compile",
+                args={"precision": precision, "dynamic_batchsize": [1, 1, 1]},
+                submodule=["image_encoder.encoder", "class_head"],
+            )
+            self.assertIsNotNone(model.image_encoder.encoder._trt_compiler)
+            self.assertIsNotNone(model.class_head._trt_compiler)
+            trt_output = model.forward(input_example)
+            # Check that lazy TRT build succeeded
+            # TODO: set up input_example in such a way that image_encoder.encoder and class_head are called
+            # and uncomment the asserts below
+            # self.assertIsNotNone(model.image_encoder.encoder._trt_compiler.engine)
+            # self.assertIsNotNone(model.class_head._trt_compiler.engine)
+            torch.testing.assert_close(trt_output, output_example, rtol=0.01, atol=0.01)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 7219ee7db771930179d9f219c59463c2c6d227ef Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 2 Sep 2024 13:12:15 +0800
Subject: [PATCH 142/183] Add box and points convert transform (#8053)

Add box and points convert transform
Cherrypick ApplyTransformToPoints

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: Mingxin Zheng <18563433+mingxin-zheng@users.noreply.github.com>
---
 docs/source/transforms.rst               |  36 ++++++
 monai/transforms/__init__.py             |  12 ++
 monai/transforms/spatial/array.py        |  44 +++++++
 monai/transforms/spatial/dictionary.py   |  60 ++++++++++
 monai/transforms/spatial/functional.py   |  71 +++++++++++-
 monai/transforms/utility/array.py        | 140 ++++++++++++++++++++++-
 monai/transforms/utility/dictionary.py   |  74 ++++++++++++
 monai/transforms/utils.py                |  23 ++++
 monai/utils/__init__.py                  |   1 +
 monai/utils/type_conversion.py           |   8 ++
 tests/test_apply_transform_to_points.py  |  81 +++++++++++++
 tests/test_apply_transform_to_pointsd.py | 133 +++++++++++++++++++++
 tests/test_convert_box_points.py         | 121 ++++++++++++++++++++
 13 files changed, 799 insertions(+), 5 deletions(-)
 create mode 100644 tests/test_apply_transform_to_points.py
 create mode 100644 tests/test_apply_transform_to_pointsd.py
 create mode 100644 tests/test_convert_box_points.py

diff --git a/docs/source/transforms.rst b/docs/source/transforms.rst
index 3e45d899ec..41bb4ae79a 100644
--- a/docs/source/transforms.rst
+++ b/docs/source/transforms.rst
@@ -976,6 +976,18 @@ Spatial
     :members:
     :special-members: __call__
 
+`ConvertBoxToPoints`
+""""""""""""""""""""
+.. autoclass:: ConvertBoxToPoints
+    :members:
+    :special-members: __call__
+
+`ConvertPointsToBoxes`
+""""""""""""""""""""""
+.. autoclass:: ConvertPointsToBoxes
+    :members:
+    :special-members: __call__
+
 
 Smooth Field
 ^^^^^^^^^^^^
@@ -1222,6 +1234,12 @@ Utility
     :members:
     :special-members: __call__
 
+`ApplyTransformToPoints`
+""""""""""""""""""""""""
+.. autoclass:: ApplyTransformToPoints
+    :members:
+    :special-members: __call__
+
 Dictionary Transforms
 ---------------------
 
@@ -1973,6 +1991,18 @@ Spatial (Dict)
     :members:
     :special-members: __call__
 
+`ConvertBoxToPointsd`
+"""""""""""""""""""""
+.. autoclass:: ConvertBoxToPointsd
+    :members:
+    :special-members: __call__
+
+`ConvertPointsToBoxesd`
+"""""""""""""""""""""""
+.. autoclass:: ConvertPointsToBoxesd
+    :members:
+    :special-members: __call__
+
 
 Smooth Field (Dict)
 ^^^^^^^^^^^^^^^^^^^
@@ -2277,6 +2307,12 @@ Utility (Dict)
     :members:
     :special-members: __call__
 
+`ApplyTransformToPointsd`
+"""""""""""""""""""""""""
+.. autoclass:: ApplyTransformToPointsd
+    :members:
+    :special-members: __call__
+
 
 MetaTensor
 ^^^^^^^^^^
diff --git a/monai/transforms/__init__.py b/monai/transforms/__init__.py
index f37016e63f..2cdd965c91 100644
--- a/monai/transforms/__init__.py
+++ b/monai/transforms/__init__.py
@@ -396,6 +396,8 @@
 from .spatial.array import (
     Affine,
     AffineGrid,
+    ConvertBoxToPoints,
+    ConvertPointsToBoxes,
     Flip,
     GridDistortion,
     GridPatch,
@@ -427,6 +429,12 @@
     Affined,
     AffineD,
     AffineDict,
+    ConvertBoxToPointsd,
+    ConvertBoxToPointsD,
+    ConvertBoxToPointsDict,
+    ConvertPointsToBoxesd,
+    ConvertPointsToBoxesD,
+    ConvertPointsToBoxesDict,
     Flipd,
     FlipD,
     FlipDict,
@@ -503,6 +511,7 @@
 from .utility.array import (
     AddCoordinateChannels,
     AddExtremePointsChannel,
+    ApplyTransformToPoints,
     AsChannelLast,
     CastToType,
     ClassesToIndices,
@@ -542,6 +551,9 @@
     AddExtremePointsChanneld,
     AddExtremePointsChannelD,
     AddExtremePointsChannelDict,
+    ApplyTransformToPointsd,
+    ApplyTransformToPointsD,
+    ApplyTransformToPointsDict,
     AsChannelLastd,
     AsChannelLastD,
     AsChannelLastDict,
diff --git a/monai/transforms/spatial/array.py b/monai/transforms/spatial/array.py
index 3739a83e71..6e39fb2e19 100644
--- a/monai/transforms/spatial/array.py
+++ b/monai/transforms/spatial/array.py
@@ -25,6 +25,7 @@
 
 from monai.config import USE_COMPILED, DtypeLike
 from monai.config.type_definitions import NdarrayOrTensor
+from monai.data.box_utils import BoxMode, StandardMode
 from monai.data.meta_obj import get_track_meta, set_track_meta
 from monai.data.meta_tensor import MetaTensor
 from monai.data.utils import AFFINE_TOL, affine_to_spacing, compute_shape_offset, iter_patch, to_affine_nd, zoom_affine
@@ -34,6 +35,8 @@
 from monai.transforms.inverse import InvertibleTransform
 from monai.transforms.spatial.functional import (
     affine_func,
+    convert_box_to_points,
+    convert_points_to_box,
     flip,
     orientation,
     resize,
@@ -3544,3 +3547,44 @@ def __call__(self, img: torch.Tensor, randomize: bool = True) -> torch.Tensor:
 
         else:
             return img
+
+
+class ConvertBoxToPoints(Transform):
+    """
+    Converts an axis-aligned bounding box to points. It can automatically convert the boxes to the points based on the box mode.
+    Bounding boxes of the shape (N, C) for N boxes. C is [x1, y1, x2, y2] for 2D or [x1, y1, z1, x2, y2, z2] for 3D for each box.
+    Return shape will be (N, 4, 2) for 2D or (N, 8, 3) for 3D.
+    """
+
+    backend = [TransformBackends.TORCH, TransformBackends.NUMPY]
+
+    def __init__(self, mode: str | BoxMode | type[BoxMode] | None = None) -> None:
+        """
+        Args:
+            mode: the mode of the box, can be a string, a BoxMode instance or a BoxMode class. Defaults to StandardMode.
+        """
+        super().__init__()
+        self.mode = StandardMode if mode is None else mode
+
+    def __call__(self, data: Any):
+        data = convert_to_tensor(data, track_meta=get_track_meta())
+        points = convert_box_to_points(data, mode=self.mode)
+        return convert_to_dst_type(points, data)[0]
+
+
+class ConvertPointsToBoxes(Transform):
+    """
+    Converts points to an axis-aligned bounding box.
+    Points representing the corners of the bounding box. Shape (N, 8, 3) for the 8 corners of a 3D cuboid or
+    (N, 4, 2) for the 4 corners of a 2D rectangle.
+    """
+
+    backend = [TransformBackends.TORCH, TransformBackends.NUMPY]
+
+    def __init__(self) -> None:
+        super().__init__()
+
+    def __call__(self, data: Any):
+        data = convert_to_tensor(data, track_meta=get_track_meta())
+        box = convert_points_to_box(data)
+        return convert_to_dst_type(box, data)[0]
diff --git a/monai/transforms/spatial/dictionary.py b/monai/transforms/spatial/dictionary.py
index 01fadcfb69..82dee15c7c 100644
--- a/monai/transforms/spatial/dictionary.py
+++ b/monai/transforms/spatial/dictionary.py
@@ -26,6 +26,7 @@
 
 from monai.config import DtypeLike, KeysCollection, SequenceStr
 from monai.config.type_definitions import NdarrayOrTensor
+from monai.data.box_utils import BoxMode, StandardMode
 from monai.data.meta_obj import get_track_meta
 from monai.data.meta_tensor import MetaTensor
 from monai.networks.layers.simplelayers import GaussianFilter
@@ -33,6 +34,8 @@
 from monai.transforms.inverse import InvertibleTransform
 from monai.transforms.spatial.array import (
     Affine,
+    ConvertBoxToPoints,
+    ConvertPointsToBoxes,
     Flip,
     GridDistortion,
     GridPatch,
@@ -2611,6 +2614,61 @@ def __call__(self, data: Mapping[Hashable, NdarrayOrTensor]) -> dict[Hashable, N
         return d
 
 
+class ConvertBoxToPointsd(MapTransform):
+    """
+    Dictionary-based wrapper of :py:class:`monai.transforms.ConvertBoxToPoints`.
+    """
+
+    backend = ConvertBoxToPoints.backend
+
+    def __init__(
+        self,
+        keys: KeysCollection,
+        point_key="points",
+        mode: str | BoxMode | type[BoxMode] | None = StandardMode,
+        allow_missing_keys: bool = False,
+    ):
+        """
+        Args:
+            keys: keys of the corresponding items to be transformed.
+            point_key: key to store the point data.
+            mode: the mode of the input boxes. Defaults to StandardMode.
+            allow_missing_keys: don't raise exception if key is missing.
+        """
+        super().__init__(keys, allow_missing_keys)
+        self.point_key = point_key
+        self.converter = ConvertBoxToPoints(mode=mode)
+
+    def __call__(self, data):
+        d = dict(data)
+        for key in self.key_iterator(d):
+            data[self.point_key] = self.converter(d[key])
+        return data
+
+
+class ConvertPointsToBoxesd(MapTransform):
+    """
+    Dictionary-based wrapper of :py:class:`monai.transforms.ConvertPointsToBoxes`.
+    """
+
+    def __init__(self, keys: KeysCollection, box_key="box", allow_missing_keys: bool = False):
+        """
+        Args:
+            keys: keys of the corresponding items to be transformed.
+            box_key: key to store the box data.
+            allow_missing_keys: don't raise exception if key is missing.
+        """
+        super().__init__(keys, allow_missing_keys)
+        self.box_key = box_key
+        self.converter = ConvertPointsToBoxes()
+
+    def __call__(self, data):
+        d = dict(data)
+        for key in self.key_iterator(d):
+            data[self.box_key] = self.converter(d[key])
+        return data
+
+
 SpatialResampleD = SpatialResampleDict = SpatialResampled
 ResampleToMatchD = ResampleToMatchDict = ResampleToMatchd
 SpacingD = SpacingDict = Spacingd
@@ -2635,3 +2693,5 @@ def __call__(self, data: Mapping[Hashable, NdarrayOrTensor]) -> dict[Hashable, N
 GridPatchD = GridPatchDict = GridPatchd
 RandGridPatchD = RandGridPatchDict = RandGridPatchd
 RandSimulateLowResolutionD = RandSimulateLowResolutionDict = RandSimulateLowResolutiond
+ConvertBoxToPointsD = ConvertBoxToPointsDict = ConvertBoxToPointsd
+ConvertPointsToBoxesD = ConvertPointsToBoxesDict = ConvertPointsToBoxesd
diff --git a/monai/transforms/spatial/functional.py b/monai/transforms/spatial/functional.py
index 22726f06a5..b693e7d023 100644
--- a/monai/transforms/spatial/functional.py
+++ b/monai/transforms/spatial/functional.py
@@ -24,6 +24,7 @@
 import monai
 from monai.config import USE_COMPILED
 from monai.config.type_definitions import NdarrayOrTensor
+from monai.data.box_utils import get_boxmode
 from monai.data.meta_obj import get_track_meta
 from monai.data.meta_tensor import MetaTensor
 from monai.data.utils import AFFINE_TOL, compute_shape_offset, to_affine_nd
@@ -32,7 +33,7 @@
 from monai.transforms.intensity.array import GaussianSmooth
 from monai.transforms.inverse import TraceableTransform
 from monai.transforms.utils import create_rotate, create_translate, resolves_modes, scale_affine
-from monai.transforms.utils_pytorch_numpy_unification import allclose
+from monai.transforms.utils_pytorch_numpy_unification import allclose, concatenate, stack
 from monai.utils import (
     LazyAttr,
     TraceKeys,
@@ -610,3 +611,71 @@ def affine_func(
         out = _maybe_new_metatensor(img, dtype=torch.float32, device=resampler.device)
     out = out.copy_meta_from(meta_info) if isinstance(out, MetaTensor) else out
     return out if image_only else (out, affine)
+
+
+def convert_box_to_points(bbox, mode):
+    """
+    Converts an axis-aligned bounding box to points.
+
+    Args:
+        mode: The mode specifying how to interpret the bounding box.
+        bbox: Bounding boxes of the shape (N, C) for N boxes. C is [x1, y1, x2, y2] for 2D or [x1, y1, z1, x2, y2, z2]
+            for 3D for each box. Return shape will be (N, 4, 2) for 2D or (N, 8, 3) for 3D.
+
+    Returns:
+        sequence of points representing the corners of the bounding box.
+    """
+
+    mode = get_boxmode(mode)
+
+    points_list = []
+    for _num in range(bbox.shape[0]):
+        corners = mode.boxes_to_corners(bbox[_num : _num + 1])
+        if len(corners) == 4:
+            points_list.append(
+                concatenate(
+                    [
+                        concatenate([corners[0], corners[1]], axis=1),
+                        concatenate([corners[2], corners[1]], axis=1),
+                        concatenate([corners[2], corners[3]], axis=1),
+                        concatenate([corners[0], corners[3]], axis=1),
+                    ],
+                    axis=0,
+                )
+            )
+        else:
+            points_list.append(
+                concatenate(
+                    [
+                        concatenate([corners[0], corners[1], corners[2]], axis=1),
+                        concatenate([corners[3], corners[1], corners[2]], axis=1),
+                        concatenate([corners[3], corners[4], corners[2]], axis=1),
+                        concatenate([corners[0], corners[4], corners[2]], axis=1),
+                        concatenate([corners[0], corners[1], corners[5]], axis=1),
+                        concatenate([corners[3], corners[1], corners[5]], axis=1),
+                        concatenate([corners[3], corners[4], corners[5]], axis=1),
+                        concatenate([corners[0], corners[4], corners[5]], axis=1),
+                    ],
+                    axis=0,
+                )
+            )
+
+    return stack(points_list, dim=0)
+
+
+def convert_points_to_box(points):
+    """
+    Converts points to an axis-aligned bounding box.
+
+    Args:
+        points: Points representing the corners of the bounding box. Shape (N, 8, 3) for the 8 corners of
+            a 3D cuboid or (N, 4, 2) for the 4 corners of a 2D rectangle.
+    """
+    from monai.transforms.utils_pytorch_numpy_unification import max, min
+
+    mins = min(points, dim=1)
+    maxs = max(points, dim=1)
+    # Concatenate the min and max values to get the bounding boxes
+    bboxes = concatenate([mins, maxs], axis=1)
+
+    return bboxes
diff --git a/monai/transforms/utility/array.py b/monai/transforms/utility/array.py
index 5dfbcb0e91..fee546bea3 100644
--- a/monai/transforms/utility/array.py
+++ b/monai/transforms/utility/array.py
@@ -31,7 +31,7 @@
 from monai.config.type_definitions import NdarrayOrTensor
 from monai.data.meta_obj import get_track_meta
 from monai.data.meta_tensor import MetaTensor
-from monai.data.utils import is_no_channel, no_collation
+from monai.data.utils import is_no_channel, no_collation, orientation_ras_lps
 from monai.networks.layers.simplelayers import (
     ApplyFilter,
     EllipticalFilter,
@@ -42,16 +42,17 @@
     SharpenFilter,
     median_filter,
 )
-from monai.transforms.inverse import InvertibleTransform
+from monai.transforms.inverse import InvertibleTransform, TraceableTransform
 from monai.transforms.traits import MultiSampleTrait
 from monai.transforms.transform import Randomizable, RandomizableTrait, RandomizableTransform, Transform
 from monai.transforms.utils import (
+    apply_affine_to_points,
     extreme_points_to_image,
     get_extreme_points,
     map_binary_to_indices,
     map_classes_to_indices,
 )
-from monai.transforms.utils_pytorch_numpy_unification import concatenate, in1d, moveaxis, unravel_indices
+from monai.transforms.utils_pytorch_numpy_unification import concatenate, in1d, linalg_inv, moveaxis, unravel_indices
 from monai.utils import (
     MetaKeys,
     TraceKeys,
@@ -66,7 +67,7 @@
 )
 from monai.utils.enums import TransformBackends
 from monai.utils.misc import is_module_ver_at_least
-from monai.utils.type_conversion import convert_to_dst_type, get_equivalent_dtype
+from monai.utils.type_conversion import convert_to_dst_type, get_dtype_string, get_equivalent_dtype
 
 PILImageImage, has_pil = optional_import("PIL.Image", name="Image")
 pil_image_fromarray, _ = optional_import("PIL.Image", name="fromarray")
@@ -106,6 +107,7 @@
     "ToCupy",
     "ImageFilter",
     "RandImageFilter",
+    "ApplyTransformToPoints",
 ]
 
 
@@ -1715,3 +1717,133 @@ def __call__(self, img: NdarrayOrTensor, meta_dict: Mapping | None = None) -> Nd
         if self._do_transform:
             img = self.filter(img)
         return img
+
+
+class ApplyTransformToPoints(InvertibleTransform, Transform):
+    """
+    Transform points between image coordinates and world coordinates.
+    The input coordinates are assumed to be in the shape (C, N, 2 or 3), where C represents the number of channels
+    and N denotes the number of points. It will return a tensor with the same shape as the input.
+
+    Args:
+        dtype: The desired data type for the output.
+        affine: A 3x3 or 4x4 affine transformation matrix applied to points. This matrix typically originates
+            from the image. For 2D points, a 3x3 matrix can be provided, avoiding the need to add an unnecessary
+            Z dimension. While a 4x4 matrix is required for 3D transformations, it's important to note that when
+            applying a 4x4 matrix to 2D points, the additional dimensions are handled accordingly.
+            The matrix is always converted to float64 for computation, which can be computationally
+            expensive when applied to a large number of points.
+            If None, will try to use the affine matrix from the input data.
+        invert_affine: Whether to invert the affine transformation matrix applied to the points. Defaults to ``True``.
+            Typically, the affine matrix is derived from an image and represents its location in world space,
+            while the points are in world coordinates. A value of ``True`` represents transforming these
+            world space coordinates to the image's coordinate space, and ``False`` the inverse of this operation.
+        affine_lps_to_ras: Defaults to ``False``. Set to `True` if your point data is in the RAS coordinate system
+            or you're using `ITKReader` with `affine_lps_to_ras=True`.
+            This ensures the correct application of the affine transformation between LPS (left-posterior-superior)
+            and RAS (right-anterior-superior) coordinate systems. This argument ensures the points and the affine
+            matrix are in the same coordinate system.
+
+    Use Cases:
+        - Transforming points between world space and image space, and vice versa.
+        - Automatically handling inverse transformations between image space and world space.
+        - If points have an existing affine transformation, the class computes and
+          applies the required delta affine transformation.
+
+    """
+
+    def __init__(
+        self,
+        dtype: DtypeLike | torch.dtype | None = None,
+        affine: torch.Tensor | None = None,
+        invert_affine: bool = True,
+        affine_lps_to_ras: bool = False,
+    ) -> None:
+        self.dtype = dtype
+        self.affine = affine
+        self.invert_affine = invert_affine
+        self.affine_lps_to_ras = affine_lps_to_ras
+
+    def transform_coordinates(
+        self, data: torch.Tensor, affine: torch.Tensor | None = None
+    ) -> tuple[torch.Tensor, dict]:
+        """
+        Transform coordinates using an affine transformation matrix.
+
+        Args:
+            data: The input coordinates are assumed to be in the shape (C, N, 2 or 3),
+                where C represents the number of channels and N denotes the number of points.
+            affine: 3x3 or 4x4 affine transformation matrix. The matrix is always converted to float64 for computation,
+                which can be computationally expensive when applied to a large number of points.
+
+        Returns:
+            Transformed coordinates.
+        """
+        data = convert_to_tensor(data, track_meta=get_track_meta())
+        # applied_affine is the affine transformation matrix that has already been applied to the point data
+        applied_affine = getattr(data, "affine", None)
+
+        if affine is None and self.invert_affine:
+            raise ValueError("affine must be provided when invert_affine is True.")
+
+        affine = applied_affine if affine is None else affine
+        affine = convert_data_type(affine, dtype=torch.float64)[0]  # always convert to float64 for affine
+        original_affine: torch.Tensor = affine
+        if self.affine_lps_to_ras:
+            affine = orientation_ras_lps(affine)
+
+        # the final affine transformation matrix that will be applied to the point data
+        _affine: torch.Tensor = affine
+        if self.invert_affine:
+            _affine = linalg_inv(affine)
+            if applied_affine is not None:
+                # consider the affine transformation already applied to the data in the world space
+                # and compute delta affine
+                _affine = _affine @ linalg_inv(applied_affine)
+        out = apply_affine_to_points(data, _affine, dtype=self.dtype)
+
+        extra_info = {
+            "invert_affine": self.invert_affine,
+            "dtype": get_dtype_string(self.dtype),
+            "image_affine": original_affine,  # record for inverse operation
+            "affine_lps_to_ras": self.affine_lps_to_ras,
+        }
+        xform: torch.Tensor = original_affine if self.invert_affine else linalg_inv(original_affine)
+        meta_info = TraceableTransform.track_transform_meta(
+            data, affine=xform, extra_info=extra_info, transform_info=self.get_transform_info()
+        )
+
+        return out, meta_info
+
+    def __call__(self, data: torch.Tensor, affine: torch.Tensor | None = None):
+        """
+        Args:
+            data: The input coordinates are assumed to be in the shape (C, N, 2 or 3),
+                where C represents the number of channels and N denotes the number of points.
+            affine: A 3x3 or 4x4 affine transformation matrix, this argument will take precedence over ``self.affine``.
+        """
+        if data.ndim != 3 or data.shape[-1] not in (2, 3):
+            raise ValueError(f"data should be in shape (C, N, 2 or 3), got {data.shape}.")
+        affine = self.affine if affine is None else affine
+        if affine is not None and affine.shape not in ((3, 3), (4, 4)):
+            raise ValueError(f"affine should be in shape (3, 3) or (4, 4), got {affine.shape}.")
+
+        out, meta_info = self.transform_coordinates(data, affine)
+
+        return out.copy_meta_from(meta_info) if isinstance(out, MetaTensor) else out
+
+    def inverse(self, data: torch.Tensor) -> torch.Tensor:
+        transform = self.pop_transform(data)
+        # Create inverse transform
+        dtype = transform[TraceKeys.EXTRA_INFO]["dtype"]
+        invert_affine = not transform[TraceKeys.EXTRA_INFO]["invert_affine"]
+        affine = transform[TraceKeys.EXTRA_INFO]["image_affine"]
+        affine_lps_to_ras = transform[TraceKeys.EXTRA_INFO]["affine_lps_to_ras"]
+        inverse_transform = ApplyTransformToPoints(
+            dtype=dtype, invert_affine=invert_affine, affine_lps_to_ras=affine_lps_to_ras
+        )
+        # Apply inverse
+        with inverse_transform.trace_transform(False):
+            data = inverse_transform(data, affine)
+
+        return data
diff --git a/monai/transforms/utility/dictionary.py b/monai/transforms/utility/dictionary.py
index 2475060f4e..1279ca93ab 100644
--- a/monai/transforms/utility/dictionary.py
+++ b/monai/transforms/utility/dictionary.py
@@ -35,6 +35,7 @@
 from monai.transforms.utility.array import (
     AddCoordinateChannels,
     AddExtremePointsChannel,
+    ApplyTransformToPoints,
     AsChannelLast,
     CastToType,
     ClassesToIndices,
@@ -180,6 +181,9 @@
     "ClassesToIndicesd",
     "ClassesToIndicesD",
     "ClassesToIndicesDict",
+    "ApplyTransformToPointsd",
+    "ApplyTransformToPointsD",
+    "ApplyTransformToPointsDict",
 ]
 
 DEFAULT_POST_FIX = PostFix.meta()
@@ -1744,6 +1748,75 @@ def __call__(self, data: Mapping[Hashable, NdarrayOrTensor]) -> dict[Hashable, N
         return d
 
 
+class ApplyTransformToPointsd(MapTransform, InvertibleTransform):
+    """
+    Dictionary-based wrapper of :py:class:`monai.transforms.ApplyTransformToPoints`.
+    The input coordinates are assumed to be in the shape (C, N, 2 or 3),
+    where C represents the number of channels and N denotes the number of points.
+    The output has the same shape as the input.
+
+    Args:
+        keys: keys of the corresponding items to be transformed.
+            See also: monai.transforms.MapTransform
+        refer_key: The key of the reference item used for transformation.
+            It can directly refer to an affine or an image from which the affine can be derived.
+        dtype: The desired data type for the output.
+        affine: A 3x3 or 4x4 affine transformation matrix applied to points. This matrix typically originates
+            from the image. For 2D points, a 3x3 matrix can be provided, avoiding the need to add an unnecessary
+            Z dimension. While a 4x4 matrix is required for 3D transformations, it's important to note that when
+            applying a 4x4 matrix to 2D points, the additional dimensions are handled accordingly.
+            The matrix is always converted to float64 for computation, which can be computationally
+            expensive when applied to a large number of points.
+            If None, will try to use the affine matrix from the refer data.
+        invert_affine: Whether to invert the affine transformation matrix applied to the points. Defaults to ``True``.
+            Typically, the affine matrix is derived from the image, while the points are in world coordinates.
+            If you want to align the points with the image, set this to ``True``. Otherwise, set it to ``False``.
+        affine_lps_to_ras: Defaults to ``False``. Set to `True` if your point data is in the RAS coordinate system
+            or you're using `ITKReader` with `affine_lps_to_ras=True`.
+            This ensures the correct application of the affine transformation between LPS (left-posterior-superior)
+            and RAS (right-anterior-superior) coordinate systems. This argument ensures the points and the affine
+            matrix are in the same coordinate system.
+        allow_missing_keys: Don't raise exception if key is missing.
+    """
+
+    def __init__(
+        self,
+        keys: KeysCollection,
+        refer_key: str | None = None,
+        dtype: DtypeLike | torch.dtype = torch.float64,
+        affine: torch.Tensor | None = None,
+        invert_affine: bool = True,
+        affine_lps_to_ras: bool = False,
+        allow_missing_keys: bool = False,
+    ):
+        MapTransform.__init__(self, keys, allow_missing_keys)
+        self.refer_key = refer_key
+        self.converter = ApplyTransformToPoints(
+            dtype=dtype, affine=affine, invert_affine=invert_affine, affine_lps_to_ras=affine_lps_to_ras
+        )
+
+    def __call__(self, data: Mapping[Hashable, torch.Tensor]):
+        d = dict(data)
+        if self.refer_key is not None:
+            if self.refer_key in d:
+                refer_data = d[self.refer_key]
+            else:
+                raise KeyError(f"The refer_key '{self.refer_key}' is not found in the data.")
+        else:
+            refer_data = None
+        affine = getattr(refer_data, "affine", refer_data)
+        for key in self.key_iterator(d):
+            coords = d[key]
+            d[key] = self.converter(coords, affine)
+        return d
+
+    def inverse(self, data: Mapping[Hashable, torch.Tensor]) -> dict[Hashable, torch.Tensor]:
+        d = dict(data)
+        for key in self.key_iterator(d):
+            d[key] = self.converter.inverse(d[key])
+        return d
+
+
 RandImageFilterD = RandImageFilterDict = RandImageFilterd
 ImageFilterD = ImageFilterDict = ImageFilterd
 IdentityD = IdentityDict = Identityd
@@ -1784,3 +1857,4 @@ def __call__(self, data: Mapping[Hashable, NdarrayOrTensor]) -> dict[Hashable, N
 RandCuCIMD = RandCuCIMDict = RandCuCIMd
 AddCoordinateChannelsD = AddCoordinateChannelsDict = AddCoordinateChannelsd
 FlattenSubKeysD = FlattenSubKeysDict = FlattenSubKeysd
+ApplyTransformToPointsD = ApplyTransformToPointsDict = ApplyTransformToPointsd
diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index 1d1f070568..b1f1bbd0f6 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -27,6 +27,7 @@
 import monai
 from monai.config import DtypeLike, IndexSelection
 from monai.config.type_definitions import NdarrayOrTensor, NdarrayTensor
+from monai.data.utils import to_affine_nd
 from monai.networks.layers import GaussianFilter
 from monai.networks.utils import meshgrid_ij
 from monai.transforms.compose import Compose
@@ -35,6 +36,7 @@
 from monai.transforms.utils_pytorch_numpy_unification import (
     any_np_pt,
     ascontiguousarray,
+    concatenate,
     cumsum,
     isfinite,
     nonzero,
@@ -2555,5 +2557,26 @@ def distance_transform_edt(
     return convert_data_type(r_vals[0] if len(r_vals) == 1 else r_vals, output_type=type(img), device=device)[0]
 
 
+def apply_affine_to_points(data: torch.Tensor, affine: torch.Tensor, dtype: DtypeLike | torch.dtype | None = None):
+    """
+    apply affine transformation to a set of points.
+
+    Args:
+        data: input data to apply affine transformation, should be a tensor of shape (C, N, 2 or 3),
+            where C represents the number of channels and N denotes the number of points.
+        affine: affine matrix to be applied, should be a tensor of shape (3, 3) or (4, 4).
+        dtype: output data dtype.
+    """
+    data_: torch.Tensor = convert_to_tensor(data, track_meta=False, dtype=torch.float64)
+    affine = to_affine_nd(data_.shape[-1], affine)
+
+    homogeneous: torch.Tensor = concatenate((data_, torch.ones((data_.shape[0], data_.shape[1], 1))), axis=2)  # type: ignore
+    transformed_homogeneous = torch.matmul(homogeneous, affine.T)
+    transformed_coordinates = transformed_homogeneous[:, :, :-1]
+    out, *_ = convert_to_dst_type(transformed_coordinates, data, dtype=dtype)
+
+    return out
+
+
 if __name__ == "__main__":
     print_transform_backends()
diff --git a/monai/utils/__init__.py b/monai/utils/__init__.py
index 03fa1ceed1..4e36e3cd47 100644
--- a/monai/utils/__init__.py
+++ b/monai/utils/__init__.py
@@ -148,6 +148,7 @@
     dtype_numpy_to_torch,
     dtype_torch_to_numpy,
     get_dtype,
+    get_dtype_string,
     get_equivalent_dtype,
     get_numpy_dtype_from_string,
     get_torch_dtype_from_string,
diff --git a/monai/utils/type_conversion.py b/monai/utils/type_conversion.py
index e4f97fc4a6..420e935b33 100644
--- a/monai/utils/type_conversion.py
+++ b/monai/utils/type_conversion.py
@@ -33,6 +33,7 @@
     "get_equivalent_dtype",
     "convert_data_type",
     "get_dtype",
+    "get_dtype_string",
     "convert_to_cupy",
     "convert_to_numpy",
     "convert_to_tensor",
@@ -102,6 +103,13 @@ def get_dtype(data: Any) -> DtypeLike | torch.dtype:
     return type(data)
 
 
+def get_dtype_string(dtype: DtypeLike | torch.dtype) -> str:
+    """Get a string representation of the dtype."""
+    if isinstance(dtype, torch.dtype):
+        return str(dtype)[6:]
+    return str(dtype)[3:]
+
+
 def convert_to_tensor(
     data: Any,
     dtype: DtypeLike | torch.dtype = None,
diff --git a/tests/test_apply_transform_to_points.py b/tests/test_apply_transform_to_points.py
new file mode 100644
index 0000000000..0c16603996
--- /dev/null
+++ b/tests/test_apply_transform_to_points.py
@@ -0,0 +1,81 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.data import MetaTensor
+from monai.transforms.utility.array import ApplyTransformToPoints
+from monai.utils import set_determinism
+
+set_determinism(seed=0)
+
+DATA_2D = torch.rand(1, 64, 64)
+DATA_3D = torch.rand(1, 64, 64, 64)
+POINT_2D_WORLD = torch.tensor([[[2, 2], [2, 4], [4, 6]]])
+POINT_2D_IMAGE = torch.tensor([[[1, 1], [1, 2], [2, 3]]])
+POINT_2D_IMAGE_RAS = torch.tensor([[[-1, -1], [-1, -2], [-2, -3]]])
+POINT_3D_WORLD = torch.tensor([[[2, 4, 6], [8, 10, 12]], [[14, 16, 18], [20, 22, 24]]])
+POINT_3D_IMAGE = torch.tensor([[[-8, 8, 6], [-2, 14, 12]], [[4, 20, 18], [10, 26, 24]]])
+POINT_3D_IMAGE_RAS = torch.tensor([[[-12, 0, 6], [-18, -6, 12]], [[-24, -12, 18], [-30, -18, 24]]])
+AFFINE_1 = torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])
+AFFINE_2 = torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])
+
+TEST_CASES = [
+    [MetaTensor(DATA_2D, affine=AFFINE_1), POINT_2D_WORLD, None, True, False, POINT_2D_IMAGE],
+    [None, MetaTensor(POINT_2D_IMAGE, affine=AFFINE_1), None, False, False, POINT_2D_WORLD],
+    [None, MetaTensor(POINT_2D_IMAGE, affine=AFFINE_1), AFFINE_1, False, False, POINT_2D_WORLD],
+    [MetaTensor(DATA_2D, affine=AFFINE_1), POINT_2D_WORLD, None, True, True, POINT_2D_IMAGE_RAS],
+    [MetaTensor(DATA_3D, affine=AFFINE_2), POINT_3D_WORLD, None, True, False, POINT_3D_IMAGE],
+    [
+        MetaTensor(DATA_3D, affine=AFFINE_2),
+        MetaTensor(POINT_3D_IMAGE, affine=AFFINE_2),
+        None,
+        False,
+        False,
+        POINT_3D_WORLD,
+    ],
+    [MetaTensor(DATA_3D, affine=AFFINE_2), POINT_3D_WORLD, None, True, True, POINT_3D_IMAGE_RAS],
+]
+
+TEST_CASES_WRONG = [
+    [POINT_2D_WORLD, True, None],
+    [POINT_2D_WORLD.unsqueeze(0), False, None],
+    [POINT_3D_WORLD[..., 0:1], False, None],
+    [POINT_3D_WORLD, False, torch.tensor([[[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]]])],
+]
+
+
+class TestCoordinateTransform(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_transform_coordinates(self, image, points, affine, invert_affine, affine_lps_to_ras, expected_output):
+        transform = ApplyTransformToPoints(
+            dtype=torch.int64, affine=affine, invert_affine=invert_affine, affine_lps_to_ras=affine_lps_to_ras
+        )
+        affine = image.affine if image is not None else None
+        output = transform(points, affine)
+        self.assertTrue(torch.allclose(output, expected_output))
+        invert_out = transform.inverse(output)
+        self.assertTrue(torch.allclose(invert_out, points))
+
+    @parameterized.expand(TEST_CASES_WRONG)
+    def test_wrong_input(self, input, invert_affine, affine):
+        transform = ApplyTransformToPoints(dtype=torch.int64, invert_affine=invert_affine)
+        with self.assertRaises(ValueError):
+            transform(input, affine)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_apply_transform_to_pointsd.py b/tests/test_apply_transform_to_pointsd.py
new file mode 100644
index 0000000000..4cedfa9d66
--- /dev/null
+++ b/tests/test_apply_transform_to_pointsd.py
@@ -0,0 +1,133 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.data import MetaTensor
+from monai.transforms.utility.dictionary import ApplyTransformToPointsd
+from monai.utils import set_determinism
+
+set_determinism(seed=0)
+
+DATA_2D = torch.rand(1, 64, 64)
+DATA_3D = torch.rand(1, 64, 64, 64)
+POINT_2D_WORLD = torch.tensor([[[2, 2], [2, 4], [4, 6]]])
+POINT_2D_IMAGE = torch.tensor([[[1, 1], [1, 2], [2, 3]]])
+POINT_2D_IMAGE_RAS = torch.tensor([[[-1, -1], [-1, -2], [-2, -3]]])
+POINT_3D_WORLD = torch.tensor([[[2, 4, 6], [8, 10, 12]], [[14, 16, 18], [20, 22, 24]]])
+POINT_3D_IMAGE = torch.tensor([[[-8, 8, 6], [-2, 14, 12]], [[4, 20, 18], [10, 26, 24]]])
+POINT_3D_IMAGE_RAS = torch.tensor([[[-12, 0, 6], [-18, -6, 12]], [[-24, -12, 18], [-30, -18, 24]]])
+
+TEST_CASES = [
+    [
+        MetaTensor(DATA_2D, affine=torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        POINT_2D_WORLD,
+        None,
+        True,
+        False,
+        POINT_2D_IMAGE,
+    ],
+    [
+        None,
+        MetaTensor(POINT_2D_IMAGE, affine=torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        None,
+        False,
+        False,
+        POINT_2D_WORLD,
+    ],
+    [
+        None,
+        MetaTensor(POINT_2D_IMAGE, affine=torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]]),
+        False,
+        False,
+        POINT_2D_WORLD,
+    ],
+    [
+        MetaTensor(DATA_2D, affine=torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        POINT_2D_WORLD,
+        None,
+        True,
+        True,
+        POINT_2D_IMAGE_RAS,
+    ],
+    [
+        MetaTensor(DATA_3D, affine=torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        POINT_3D_WORLD,
+        None,
+        True,
+        False,
+        POINT_3D_IMAGE,
+    ],
+    ["affine", POINT_3D_WORLD, None, True, False, POINT_3D_IMAGE],
+    [
+        MetaTensor(DATA_3D, affine=torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        MetaTensor(POINT_3D_IMAGE, affine=torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        None,
+        False,
+        False,
+        POINT_3D_WORLD,
+    ],
+    [
+        MetaTensor(DATA_3D, affine=torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        POINT_3D_WORLD,
+        None,
+        True,
+        True,
+        POINT_3D_IMAGE_RAS,
+    ],
+]
+
+TEST_CASES_WRONG = [
+    [POINT_2D_WORLD, True, None],
+    [POINT_2D_WORLD.unsqueeze(0), False, None],
+    [POINT_3D_WORLD[..., 0:1], False, None],
+    [POINT_3D_WORLD, False, torch.tensor([[[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]]])],
+]
+
+
+class TestCoordinateTransform(unittest.TestCase):
+    @parameterized.expand(TEST_CASES)
+    def test_transform_coordinates(self, image, points, affine, invert_affine, affine_lps_to_ras, expected_output):
+        data = {
+            "image": image,
+            "point": points,
+            "affine": torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]]),
+        }
+        refer_key = "image" if (image is not None and image != "affine") else image
+        transform = ApplyTransformToPointsd(
+            keys="point",
+            refer_key=refer_key,
+            dtype=torch.int64,
+            affine=affine,
+            invert_affine=invert_affine,
+            affine_lps_to_ras=affine_lps_to_ras,
+        )
+        output = transform(data)
+
+        self.assertTrue(torch.allclose(output["point"], expected_output))
+        invert_out = transform.inverse(output)
+        self.assertTrue(torch.allclose(invert_out["point"], points))
+
+    @parameterized.expand(TEST_CASES_WRONG)
+    def test_wrong_input(self, input, invert_affine, affine):
+        transform = ApplyTransformToPointsd(keys="point", dtype=torch.int64, invert_affine=invert_affine, affine=affine)
+        with self.assertRaises(ValueError):
+            transform({"point": input})
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_convert_box_points.py b/tests/test_convert_box_points.py
new file mode 100644
index 0000000000..5e3d7ee645
--- /dev/null
+++ b/tests/test_convert_box_points.py
@@ -0,0 +1,121 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.data.box_utils import convert_box_to_standard_mode
+from monai.transforms.spatial.array import ConvertBoxToPoints, ConvertPointsToBoxes
+from tests.utils import assert_allclose
+
+TEST_CASE_POINTS_2D = [
+    [
+        torch.tensor([[10, 20, 30, 40], [50, 60, 70, 80]]),
+        "xyxy",
+        torch.tensor([[[10, 20], [30, 20], [30, 40], [10, 40]], [[50, 60], [70, 60], [70, 80], [50, 80]]]),
+    ],
+    [torch.tensor([[10, 20, 20, 20]]), "ccwh", torch.tensor([[[0, 10], [20, 10], [20, 30], [0, 30]]])],
+]
+TEST_CASE_POINTS_3D = [
+    [
+        torch.tensor([[10, 20, 30, 40, 50, 60], [70, 80, 90, 100, 110, 120]]),
+        "xyzxyz",
+        torch.tensor(
+            [
+                [
+                    [10, 20, 30],
+                    [40, 20, 30],
+                    [40, 50, 30],
+                    [10, 50, 30],
+                    [10, 20, 60],
+                    [40, 20, 60],
+                    [40, 50, 60],
+                    [10, 50, 60],
+                ],
+                [
+                    [70, 80, 90],
+                    [100, 80, 90],
+                    [100, 110, 90],
+                    [70, 110, 90],
+                    [70, 80, 120],
+                    [100, 80, 120],
+                    [100, 110, 120],
+                    [70, 110, 120],
+                ],
+            ]
+        ),
+    ],
+    [
+        torch.tensor([[10, 20, 30, 10, 10, 10]]),
+        "cccwhd",
+        torch.tensor(
+            [
+                [
+                    [5, 15, 25],
+                    [15, 15, 25],
+                    [15, 25, 25],
+                    [5, 25, 25],
+                    [5, 15, 35],
+                    [15, 15, 35],
+                    [15, 25, 35],
+                    [5, 25, 35],
+                ]
+            ]
+        ),
+    ],
+    [
+        torch.tensor([[10, 20, 30, 40, 50, 60]]),
+        "xxyyzz",
+        torch.tensor(
+            [
+                [
+                    [10, 30, 50],
+                    [20, 30, 50],
+                    [20, 40, 50],
+                    [10, 40, 50],
+                    [10, 30, 60],
+                    [20, 30, 60],
+                    [20, 40, 60],
+                    [10, 40, 60],
+                ]
+            ]
+        ),
+    ],
+]
+
+TEST_CASES = TEST_CASE_POINTS_2D + TEST_CASE_POINTS_3D
+
+
+class TestConvertBoxToPoints(unittest.TestCase):
+
+    @parameterized.expand(TEST_CASES)
+    def test_convert_box_to_points(self, boxes, mode, expected_points):
+        transform = ConvertBoxToPoints(mode=mode)
+        converted_points = transform(boxes)
+        assert_allclose(converted_points, expected_points, type_test=False)
+
+
+class TestConvertPointsToBoxes(unittest.TestCase):
+
+    @parameterized.expand(TEST_CASES)
+    def test_convert_box_to_points(self, boxes, mode, points):
+        transform = ConvertPointsToBoxes()
+        converted_boxes = transform(points)
+        expected_boxes = convert_box_to_standard_mode(boxes, mode)
+        assert_allclose(converted_boxes, expected_boxes, type_test=False)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 6a0e1b043ba2890e1463fa49df76f66e56a68b08 Mon Sep 17 00:00:00 2001
From: Yufan He <59374597+heyufan1995@users.noreply.github.com>
Date: Mon, 2 Sep 2024 01:10:10 -0500
Subject: [PATCH 143/183] Fix vista3d transpose bug (#8059)

Fixes # .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: heyufan1995 <heyufan1995@gmail.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
---
 monai/apps/vista3d/inferer.py  |  2 +-
 monai/networks/nets/vista3d.py | 16 ++++++++++------
 2 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/monai/apps/vista3d/inferer.py b/monai/apps/vista3d/inferer.py
index 709f81f624..8f622ef6cd 100644
--- a/monai/apps/vista3d/inferer.py
+++ b/monai/apps/vista3d/inferer.py
@@ -100,7 +100,7 @@ def point_based_window_inferer(
                         point_labels=point_labels,
                         class_vector=class_vector,
                         prompt_class=prompt_class,
-                        patch_coords=unravel_slice,
+                        patch_coords=[unravel_slice],
                         prev_mask=prev_mask,
                         **kwargs,
                     )
diff --git a/monai/networks/nets/vista3d.py b/monai/networks/nets/vista3d.py
index 979a090df0..4215a9a594 100644
--- a/monai/networks/nets/vista3d.py
+++ b/monai/networks/nets/vista3d.py
@@ -336,7 +336,7 @@ def set_auto_grad(self, auto_freeze: bool = False, point_freeze: bool = False):
     def forward(
         self,
         input_images: torch.Tensor,
-        patch_coords: Sequence[slice] | None = None,
+        patch_coords: list[Sequence[slice]] | None = None,
         point_coords: torch.Tensor | None = None,
         point_labels: torch.Tensor | None = None,
         class_vector: torch.Tensor | None = None,
@@ -364,8 +364,12 @@ def forward(
                 the points are for zero-shot or supported class. When class_vector and point_coords are both
                 provided, prompt_class is the same as class_vector. For prompt_class[b] > 512, point_coords[b]
                 will be considered novel class.
-            patch_coords: a sequence of the python slice objects representing the patch coordinates during sliding window inference.
-                This value is passed from sliding_window_inferer. This is an indicator for training phase or validation phase.
+            patch_coords: a list of sequence of the python slice objects representing the patch coordinates during sliding window
+                inference. This value is passed from sliding_window_inferer.
+                This is an indicator for training phase or validation phase.
+                Notice for sliding window batch size > 1 (only supported by automatic segmentation), patch_coords will inlcude
+                coordinates of multiple patches. If point prompts are included, the batch size can only be one and all the
+                functions using patch_coords will by default use patch_coords[0].
             labels: [1, 1, H, W, D], the groundtruth label tensor, only used for point-only evaluation
             label_set: the label index matching the indexes in labels. If labels are mapped to global index using RelabelID,
                 this label_set should be global mapped index. If labels are not mapped to global index, e.g. in zero-shot
@@ -395,14 +399,14 @@ def forward(
                 if val_point_sampler is None:
                     # TODO: think about how to refactor this part.
                     val_point_sampler = self.sample_points_patch_val
-                point_coords, point_labels, prompt_class = val_point_sampler(labels, patch_coords, label_set)
+                point_coords, point_labels, prompt_class = val_point_sampler(labels, patch_coords[0], label_set)
                 if prompt_class[0].item() == 0:  # type: ignore
                     point_labels[0] = -1  # type: ignore
                 labels, prev_mask = None, None
             elif point_coords is not None:
                 # If not performing patch-based point only validation, use user provided click points for inference.
                 # the point clicks is in original image space, convert it to current patch-coordinate space.
-                point_coords, point_labels = self.update_point_to_patch(patch_coords, point_coords, point_labels)  # type: ignore
+                point_coords, point_labels = self.update_point_to_patch(patch_coords[0], point_coords, point_labels)  # type: ignore
 
         if point_coords is not None and point_labels is not None:
             # remove points that used for padding purposes (point_label = -1)
@@ -455,7 +459,7 @@ def forward(
             logits[mapping_index] = self.point_head(out, point_coords, point_labels, class_vector=prompt_class)
             if prev_mask is not None and patch_coords is not None:
                 logits = self.connected_components_combine(
-                    prev_mask[patch_coords].transpose(1, 0).to(logits.device),
+                    prev_mask[patch_coords[0]].transpose(1, 0).to(logits.device),
                     logits[mapping_index],
                     point_coords,  # type: ignore
                     point_labels,  # type: ignore

From c9b8bdb93b8a5e5fd4e9a72c34dba760f9c66a04 Mon Sep 17 00:00:00 2001
From: BenjaminLi <76795078+25benjaminli@users.noreply.github.com>
Date: Mon, 2 Sep 2024 11:06:34 -0400
Subject: [PATCH 144/183] Add deterministic support for
 RandSimulateLowResolutiond (#8057)

Fixes #7911, which describes how the RandSimulateLowResolutiond
dictionary transform produces non-deterministic outputs, yet the typical
array transform RandSimulateLowResolution produces deterministic ones.

### Description

Inside of `RandSimulateLowResolutiond`, added the line
`self.sim_lowres_tfm.set_random_state(seed, state)` in
`set_random_state` to ensure the helper function `sim_lowres_tfm` is
seeded and the transform can be performed deterministically.

Note: I also sifted through the other dictionary transforms with helper
functions and did not find anything that looked problematic similar to
this.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [ ] Non-breaking change (fix or new feature that would not break
existing functionality).
- [x] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [x] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: BenjaminLi <25benjaminli@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/transforms/spatial/dictionary.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/monai/transforms/spatial/dictionary.py b/monai/transforms/spatial/dictionary.py
index 82dee15c7c..2b80034a07 100644
--- a/monai/transforms/spatial/dictionary.py
+++ b/monai/transforms/spatial/dictionary.py
@@ -2588,6 +2588,7 @@ def set_random_state(
         self, seed: int | None = None, state: np.random.RandomState | None = None
     ) -> RandSimulateLowResolutiond:
         super().set_random_state(seed, state)
+        self.sim_lowres_tfm.set_random_state(seed, state)
         return self
 
     def __call__(self, data: Mapping[Hashable, NdarrayOrTensor]) -> dict[Hashable, NdarrayOrTensor]:

From dbfe418c03073baf07a0e14cd7606571f3d0de18 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 3 Sep 2024 00:35:26 +0800
Subject: [PATCH 145/183] Ignore warning from nptyping as workaround (#8062)

workaround for #8061

### Description

Ignore warning from nptyping as workaround

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/data/image_reader.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/monai/data/image_reader.py b/monai/data/image_reader.py
index f5e199e2a3..aab1e03898 100644
--- a/monai/data/image_reader.py
+++ b/monai/data/image_reader.py
@@ -34,6 +34,9 @@
 )
 from monai.utils import MetaKeys, SpaceKeys, TraceKeys, ensure_tuple, optional_import, require_pkg
 
+# workaround for https://github.com/Project-MONAI/MONAI/issues/8061
+warnings.filterwarnings("ignore", category=DeprecationWarning, module="nptyping")
+
 if TYPE_CHECKING:
     import itk
     import nibabel as nib

From befb5f6af1521e53ee3e757ea66df30b40d24801 Mon Sep 17 00:00:00 2001
From: Bastian Wittmann <73648286+bwittmann@users.noreply.github.com>
Date: Tue, 3 Sep 2024 14:55:00 +0200
Subject: [PATCH 146/183] Forced `Fourier` class to output contiguous tensors.
 (#7969)

Forced `Fourier` class to output contiguous tensors, which potentially
fixes a performance bottleneck.

### Description

Some transforms, such as `RandKSpaceSpikeNoise`, rely on the `Fourier`
class.
In its current state, the `Fourier` class returns non-contiguous
tensors, which potentially limits performance.
For example, when followed by `RandHistogramShift`, the following
warning occurs:

```
<path_to_monai>/monai/transforms/intensity/array.py:1852: UserWarning: torch.searchsorted(): input value tensor is non-contiguous, this will lower the performance due to extra data copy when converting non-contiguous tensor to contiguous, please use contiguous input value tensor if possible. This message will only appear once per program. (Triggered internally at /opt/conda/conda-bld/pytorch_1716905975447/work/aten/src/ATen/native/BucketizationUtils.h:32.)
  indices = ns.searchsorted(xp.reshape(-1), x.reshape(-1)) - 1
```

A straightforward fix is to force the `Fourier` class to output
contiguous tensors (see commit).
To reproduce, please run:

```
from monai.transforms import RandKSpaceSpikeNoise
from monai.transforms.utils import Fourier
import numpy as np

### TEST WITH TRANSFORMS ###
t = RandKSpaceSpikeNoise(prob=1)

# for torch tensors
a_torch = torch.rand(1, 128, 128, 128)
print(a_torch.is_contiguous())

a_torch_mod = t(a_torch)
print(a_torch_mod.is_contiguous())

# for np arrays
a_np = np.random.rand(1, 128, 128, 128)
print(a_np.flags['C_CONTIGUOUS'])

a_np_mod = t(a_np)  # automatically transformed to torch.tensor
print(a_np_mod.is_contiguous())

### TEST DIRECTLY WITH FOURIER ###
f = Fourier()

# inv_shift_fourier
# for torch tensors
real_torch = torch.randn(1, 128, 128, 128)
im_torch = torch.randn(1, 128, 128, 128)
k_torch = torch.complex(real_torch, im_torch)
print(k_torch.is_contiguous())

out_torch = f.inv_shift_fourier(k_torch, spatial_dims=3)
print(out_torch.is_contiguous())

# for np arrays
real_np = np.random.randn(1, 100, 100, 100)
im_np = np.random.randn(1, 100, 100, 100)
k_np = real_np + 1j * im_np
print(k_np.flags['C_CONTIGUOUS'])

out_np = f.inv_shift_fourier(k_np, spatial_dims=3)
print(out_np.flags['C_CONTIGUOUS'])

# shift_fourier
# for torch tensors
a_torch = torch.rand(1, 128, 128, 128)
print(a_torch.is_contiguous())

out_torch = f.shift_fourier(a_torch, spatial_dims=3)
print(out_torch.is_contiguous())

# for np arrays
a_np = np.random.rand(1, 128, 128, 128)
print(a_np.flags['C_CONTIGUOUS'])

out_np = f.shift_fourier(a_np, spatial_dims=3)
print(out_np.flags['C_CONTIGUOUS'])
```

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Bastian Wittmann <bastian.wittmann@uzh.ch>.
Signed-off-by: Bastian Wittmann <bastian.wittmann@uzh.ch>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/transforms/utils.py | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index b1f1bbd0f6..32fffc25f0 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -1863,7 +1863,7 @@ class Fourier:
     """
 
     @staticmethod
-    def shift_fourier(x: NdarrayOrTensor, spatial_dims: int) -> NdarrayOrTensor:
+    def shift_fourier(x: NdarrayOrTensor, spatial_dims: int, as_contiguous: bool = False) -> NdarrayOrTensor:
         """
         Applies fourier transform and shifts the zero-frequency component to the
         center of the spectrum. Only the spatial dimensions get transformed.
@@ -1871,6 +1871,7 @@ def shift_fourier(x: NdarrayOrTensor, spatial_dims: int) -> NdarrayOrTensor:
         Args:
             x: Image to transform.
             spatial_dims: Number of spatial dimensions.
+            as_contiguous: Whether to convert the cached NumPy array or PyTorch tensor to be contiguous.
 
         Returns
             k: K-space data.
@@ -1885,10 +1886,12 @@ def shift_fourier(x: NdarrayOrTensor, spatial_dims: int) -> NdarrayOrTensor:
                 k = np.fft.fftshift(np.fft.fftn(x.cpu().numpy(), axes=dims), axes=dims)
         else:
             k = np.fft.fftshift(np.fft.fftn(x, axes=dims), axes=dims)
-        return k
+        return ascontiguousarray(k) if as_contiguous else k
 
     @staticmethod
-    def inv_shift_fourier(k: NdarrayOrTensor, spatial_dims: int, n_dims: int | None = None) -> NdarrayOrTensor:
+    def inv_shift_fourier(
+        k: NdarrayOrTensor, spatial_dims: int, n_dims: int | None = None, as_contiguous: bool = False
+    ) -> NdarrayOrTensor:
         """
         Applies inverse shift and fourier transform. Only the spatial
         dimensions are transformed.
@@ -1896,6 +1899,7 @@ def inv_shift_fourier(k: NdarrayOrTensor, spatial_dims: int, n_dims: int | None
         Args:
             k: K-space data.
             spatial_dims: Number of spatial dimensions.
+            as_contiguous: Whether to convert the cached NumPy array or PyTorch tensor to be contiguous.
 
         Returns:
             x: Tensor in image space.
@@ -1910,7 +1914,7 @@ def inv_shift_fourier(k: NdarrayOrTensor, spatial_dims: int, n_dims: int | None
                 out = np.fft.ifftn(np.fft.ifftshift(k.cpu().numpy(), axes=dims), axes=dims).real
         else:
             out = np.fft.ifftn(np.fft.ifftshift(k, axes=dims), axes=dims).real
-        return out
+        return ascontiguousarray(out) if as_contiguous else out
 
 
 def get_number_image_type_conversions(transform: Compose, test_data: Any, key: Hashable | None = None) -> int:

From aea46ff26b39c0c88e3d00cb88cb03442df61dd5 Mon Sep 17 00:00:00 2001
From: Boris Fomitchev <borisfom@users.noreply.github.com>
Date: Wed, 4 Sep 2024 01:31:28 -0700
Subject: [PATCH 147/183] Trt compiler fixes (#8064)

Fixes https://github.com/Project-MONAI/MONAI/issues/8061.

### Description

Post-merge fixes for trt_compile()

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Co-authored-by: Yiheng Wang <vennw@nvidia.com>
Co-authored-by: binliunls <107988372+binliunls@users.noreply.github.com>
---
 monai/networks/trt_compiler.py | 8 ++++++--
 tests/test_trt_compile.py      | 4 ++--
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/monai/networks/trt_compiler.py b/monai/networks/trt_compiler.py
index a9dd0d9e9b..00d2eb61af 100644
--- a/monai/networks/trt_compiler.py
+++ b/monai/networks/trt_compiler.py
@@ -342,6 +342,7 @@ def forward(self, model, argv, kwargs):
                     self._build_and_save(model, build_args)
                     # This will reassign input_names from the engine
                     self._load_engine()
+                    assert self.engine is not None
             except Exception as e:
                 if self.fallback:
                     self.logger.info(f"Failed to build engine: {e}")
@@ -403,8 +404,10 @@ def _onnx_to_trt(self, onnx_path):
 
         build_args = self.build_args.copy()
         build_args["tf32"] = self.precision != "fp32"
-        build_args["fp16"] = self.precision == "fp16"
-        build_args["bf16"] = self.precision == "bf16"
+        if self.precision == "fp16":
+            build_args["fp16"] = True
+        elif self.precision == "bf16":
+            build_args["bf16"] = True
 
         self.logger.info(f"Building TensorRT engine for {onnx_path}: {self.plan_path}")
         network = network_from_onnx_path(onnx_path, flags=[trt.OnnxParserFlag.NATIVE_INSTANCENORM])
@@ -502,6 +505,7 @@ def trt_compile(
 ) -> torch.nn.Module:
     """
     Instruments model or submodule(s) with TrtCompiler and replaces its forward() with TRT hook.
+    Note: TRT 10.3 is recommended for best performance. Some nets may even fail to work with TRT 8.x
     Args:
       model: module to patch with TrtCompiler object.
       base_path: TRT plan(s) saved to f"{base_path}[.{submodule}].plan" path.
diff --git a/tests/test_trt_compile.py b/tests/test_trt_compile.py
index 21125d203f..2f9db8f0c2 100644
--- a/tests/test_trt_compile.py
+++ b/tests/test_trt_compile.py
@@ -20,10 +20,10 @@
 from monai.handlers import TrtHandler
 from monai.networks import trt_compile
 from monai.networks.nets import UNet, cell_sam_wrapper, vista3d132
-from monai.utils import optional_import
+from monai.utils import min_version, optional_import
 from tests.utils import skip_if_no_cuda, skip_if_quick, skip_if_windows
 
-trt, trt_imported = optional_import("tensorrt")
+trt, trt_imported = optional_import("tensorrt", "10.1.0", min_version)
 polygraphy, polygraphy_imported = optional_import("polygraphy")
 build_sam_vit_b, has_sam = optional_import("segment_anything.build_sam", name="build_sam_vit_b")
 

From 4e70bf694c5178637f4749a84e7a59a8d07332e7 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 4 Sep 2024 17:52:52 +0800
Subject: [PATCH 148/183] Allow ApplyTransformToPointsd receive a sequence of
 refer keys (#8063)

Enhance `ApplyTransformToPointsd` to receive a sequence of refer keys.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/transforms/utility/array.py        |  64 ++++++-----
 monai/transforms/utility/dictionary.py   |  28 ++---
 tests/test_apply_transform_to_pointsd.py | 136 ++++++++++++++++-------
 3 files changed, 146 insertions(+), 82 deletions(-)

diff --git a/monai/transforms/utility/array.py b/monai/transforms/utility/array.py
index fee546bea3..bfd2f506c2 100644
--- a/monai/transforms/utility/array.py
+++ b/monai/transforms/utility/array.py
@@ -1764,6 +1764,30 @@ def __init__(
         self.invert_affine = invert_affine
         self.affine_lps_to_ras = affine_lps_to_ras
 
+    def _compute_final_affine(self, affine: torch.Tensor, applied_affine: torch.Tensor | None = None) -> torch.Tensor:
+        """
+        Compute the final affine transformation matrix to apply to the point data.
+
+        Args:
+            data: Input coordinates assumed to be in the shape (C, N, 2 or 3).
+            affine: 3x3 or 4x4 affine transformation matrix.
+
+        Returns:
+            Final affine transformation matrix.
+        """
+
+        affine = convert_data_type(affine, dtype=torch.float64)[0]
+
+        if self.affine_lps_to_ras:
+            affine = orientation_ras_lps(affine)
+
+        if self.invert_affine:
+            affine = linalg_inv(affine)
+            if applied_affine is not None:
+                affine = affine @ applied_affine
+
+        return affine
+
     def transform_coordinates(
         self, data: torch.Tensor, affine: torch.Tensor | None = None
     ) -> tuple[torch.Tensor, dict]:
@@ -1780,35 +1804,25 @@ def transform_coordinates(
             Transformed coordinates.
         """
         data = convert_to_tensor(data, track_meta=get_track_meta())
-        # applied_affine is the affine transformation matrix that has already been applied to the point data
-        applied_affine = getattr(data, "affine", None)
-
         if affine is None and self.invert_affine:
             raise ValueError("affine must be provided when invert_affine is True.")
-
+        # applied_affine is the affine transformation matrix that has already been applied to the point data
+        applied_affine: torch.Tensor | None = getattr(data, "affine", None)
         affine = applied_affine if affine is None else affine
-        affine = convert_data_type(affine, dtype=torch.float64)[0]  # always convert to float64 for affine
-        original_affine: torch.Tensor = affine
-        if self.affine_lps_to_ras:
-            affine = orientation_ras_lps(affine)
+        if affine is None:
+            raise ValueError("affine must be provided if data does not have an affine matrix.")
 
-        # the final affine transformation matrix that will be applied to the point data
-        _affine: torch.Tensor = affine
-        if self.invert_affine:
-            _affine = linalg_inv(affine)
-            if applied_affine is not None:
-                # consider the affine transformation already applied to the data in the world space
-                # and compute delta affine
-                _affine = _affine @ linalg_inv(applied_affine)
-        out = apply_affine_to_points(data, _affine, dtype=self.dtype)
+        final_affine = self._compute_final_affine(affine, applied_affine)
+        out = apply_affine_to_points(data, final_affine, dtype=self.dtype)
 
         extra_info = {
             "invert_affine": self.invert_affine,
             "dtype": get_dtype_string(self.dtype),
-            "image_affine": original_affine,  # record for inverse operation
+            "image_affine": affine,
             "affine_lps_to_ras": self.affine_lps_to_ras,
         }
-        xform: torch.Tensor = original_affine if self.invert_affine else linalg_inv(original_affine)
+
+        xform = orientation_ras_lps(linalg_inv(final_affine)) if self.affine_lps_to_ras else linalg_inv(final_affine)
         meta_info = TraceableTransform.track_transform_meta(
             data, affine=xform, extra_info=extra_info, transform_info=self.get_transform_info()
         )
@@ -1834,16 +1848,12 @@ def __call__(self, data: torch.Tensor, affine: torch.Tensor | None = None):
 
     def inverse(self, data: torch.Tensor) -> torch.Tensor:
         transform = self.pop_transform(data)
-        # Create inverse transform
-        dtype = transform[TraceKeys.EXTRA_INFO]["dtype"]
-        invert_affine = not transform[TraceKeys.EXTRA_INFO]["invert_affine"]
-        affine = transform[TraceKeys.EXTRA_INFO]["image_affine"]
-        affine_lps_to_ras = transform[TraceKeys.EXTRA_INFO]["affine_lps_to_ras"]
         inverse_transform = ApplyTransformToPoints(
-            dtype=dtype, invert_affine=invert_affine, affine_lps_to_ras=affine_lps_to_ras
+            dtype=transform[TraceKeys.EXTRA_INFO]["dtype"],
+            invert_affine=not transform[TraceKeys.EXTRA_INFO]["invert_affine"],
+            affine_lps_to_ras=transform[TraceKeys.EXTRA_INFO]["affine_lps_to_ras"],
         )
-        # Apply inverse
         with inverse_transform.trace_transform(False):
-            data = inverse_transform(data, affine)
+            data = inverse_transform(data, transform[TraceKeys.EXTRA_INFO]["image_affine"])
 
         return data
diff --git a/monai/transforms/utility/dictionary.py b/monai/transforms/utility/dictionary.py
index 1279ca93ab..db5f19c0de 100644
--- a/monai/transforms/utility/dictionary.py
+++ b/monai/transforms/utility/dictionary.py
@@ -1758,8 +1758,9 @@ class ApplyTransformToPointsd(MapTransform, InvertibleTransform):
     Args:
         keys: keys of the corresponding items to be transformed.
             See also: monai.transforms.MapTransform
-        refer_key: The key of the reference item used for transformation.
-            It can directly refer to an affine or an image from which the affine can be derived.
+        refer_keys: The key of the reference item used for transformation.
+            It can directly refer to an affine or an image from which the affine can be derived. It can also be a
+            sequence of keys, in which case each refers to the affine applied to the matching points in `keys`.
         dtype: The desired data type for the output.
         affine: A 3x3 or 4x4 affine transformation matrix applied to points. This matrix typically originates
             from the image. For 2D points, a 3x3 matrix can be provided, avoiding the need to add an unnecessary
@@ -1782,7 +1783,7 @@ class ApplyTransformToPointsd(MapTransform, InvertibleTransform):
     def __init__(
         self,
         keys: KeysCollection,
-        refer_key: str | None = None,
+        refer_keys: KeysCollection | None = None,
         dtype: DtypeLike | torch.dtype = torch.float64,
         affine: torch.Tensor | None = None,
         invert_affine: bool = True,
@@ -1790,23 +1791,24 @@ def __init__(
         allow_missing_keys: bool = False,
     ):
         MapTransform.__init__(self, keys, allow_missing_keys)
-        self.refer_key = refer_key
+        self.refer_keys = ensure_tuple_rep(refer_keys, len(self.keys))
         self.converter = ApplyTransformToPoints(
             dtype=dtype, affine=affine, invert_affine=invert_affine, affine_lps_to_ras=affine_lps_to_ras
         )
 
     def __call__(self, data: Mapping[Hashable, torch.Tensor]):
         d = dict(data)
-        if self.refer_key is not None:
-            if self.refer_key in d:
-                refer_data = d[self.refer_key]
-            else:
-                raise KeyError(f"The refer_key '{self.refer_key}' is not found in the data.")
-        else:
-            refer_data = None
-        affine = getattr(refer_data, "affine", refer_data)
-        for key in self.key_iterator(d):
+        for key, refer_key in self.key_iterator(d, self.refer_keys):
             coords = d[key]
+            affine = None  # represents using affine given in constructor
+            if refer_key is not None:
+                if refer_key in d:
+                    refer_data = d[refer_key]
+                else:
+                    raise KeyError(f"The refer_key '{refer_key}' is not found in the data.")
+
+                # use the "affine" member of refer_data, or refer_data itself, as the affine matrix
+                affine = getattr(refer_data, "affine", refer_data)
             d[key] = self.converter(coords, affine)
         return d
 
diff --git a/tests/test_apply_transform_to_pointsd.py b/tests/test_apply_transform_to_pointsd.py
index 4cedfa9d66..978113931c 100644
--- a/tests/test_apply_transform_to_pointsd.py
+++ b/tests/test_apply_transform_to_pointsd.py
@@ -30,72 +30,90 @@
 POINT_3D_WORLD = torch.tensor([[[2, 4, 6], [8, 10, 12]], [[14, 16, 18], [20, 22, 24]]])
 POINT_3D_IMAGE = torch.tensor([[[-8, 8, 6], [-2, 14, 12]], [[4, 20, 18], [10, 26, 24]]])
 POINT_3D_IMAGE_RAS = torch.tensor([[[-12, 0, 6], [-18, -6, 12]], [[-24, -12, 18], [-30, -18, 24]]])
+AFFINE_1 = torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])
+AFFINE_2 = torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])
 
 TEST_CASES = [
+    [MetaTensor(DATA_2D, affine=AFFINE_1), POINT_2D_WORLD, None, True, False, POINT_2D_IMAGE],  # use image affine
+    [None, MetaTensor(POINT_2D_IMAGE, affine=AFFINE_1), None, False, False, POINT_2D_WORLD],  # use point affine
+    [None, MetaTensor(POINT_2D_IMAGE, affine=AFFINE_1), AFFINE_1, False, False, POINT_2D_WORLD],  # use input affine
+    [None, POINT_2D_WORLD, AFFINE_1, True, False, POINT_2D_IMAGE],  # use input affine
     [
-        MetaTensor(DATA_2D, affine=torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        MetaTensor(DATA_2D, affine=AFFINE_1),
         POINT_2D_WORLD,
         None,
         True,
-        False,
-        POINT_2D_IMAGE,
-    ],
+        True,
+        POINT_2D_IMAGE_RAS,
+    ],  # test affine_lps_to_ras
+    [MetaTensor(DATA_3D, affine=AFFINE_2), POINT_3D_WORLD, None, True, False, POINT_3D_IMAGE],
+    ["affine", POINT_3D_WORLD, None, True, False, POINT_3D_IMAGE],  # use refer_data itself
     [
-        None,
-        MetaTensor(POINT_2D_IMAGE, affine=torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        MetaTensor(DATA_3D, affine=AFFINE_2),
+        MetaTensor(POINT_3D_IMAGE, affine=AFFINE_2),
         None,
         False,
         False,
-        POINT_2D_WORLD,
+        POINT_3D_WORLD,
     ],
+    [MetaTensor(DATA_3D, affine=AFFINE_2), POINT_3D_WORLD, None, True, True, POINT_3D_IMAGE_RAS],
+    [MetaTensor(DATA_3D, affine=AFFINE_2), POINT_3D_WORLD, None, True, True, POINT_3D_IMAGE_RAS],
+]
+TEST_CASES_SEQUENCE = [
     [
+        (MetaTensor(DATA_2D, affine=AFFINE_1), MetaTensor(DATA_3D, affine=AFFINE_2)),
+        [POINT_2D_WORLD, POINT_3D_WORLD],
         None,
-        MetaTensor(POINT_2D_IMAGE, affine=torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])),
-        torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]]),
-        False,
+        True,
         False,
-        POINT_2D_WORLD,
-    ],
+        ["image_1", "image_2"],
+        [POINT_2D_IMAGE, POINT_3D_IMAGE],
+    ],  # use image affine
     [
-        MetaTensor(DATA_2D, affine=torch.tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]])),
-        POINT_2D_WORLD,
+        (MetaTensor(DATA_2D, affine=AFFINE_1), MetaTensor(DATA_3D, affine=AFFINE_2)),
+        [POINT_2D_WORLD, POINT_3D_WORLD],
         None,
         True,
         True,
-        POINT_2D_IMAGE_RAS,
-    ],
+        ["image_1", "image_2"],
+        [POINT_2D_IMAGE_RAS, POINT_3D_IMAGE_RAS],
+    ],  # test affine_lps_to_ras
     [
-        MetaTensor(DATA_3D, affine=torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])),
-        POINT_3D_WORLD,
+        (None, None),
+        [MetaTensor(POINT_2D_IMAGE, affine=AFFINE_1), MetaTensor(POINT_3D_IMAGE, affine=AFFINE_2)],
         None,
+        False,
+        False,
+        None,
+        [POINT_2D_WORLD, POINT_3D_WORLD],
+    ],  # use point affine
+    [
+        (None, None),
+        [POINT_2D_WORLD, POINT_2D_WORLD],
+        AFFINE_1,
         True,
         False,
-        POINT_3D_IMAGE,
-    ],
-    ["affine", POINT_3D_WORLD, None, True, False, POINT_3D_IMAGE],
+        None,
+        [POINT_2D_IMAGE, POINT_2D_IMAGE],
+    ],  # use input affine
     [
-        MetaTensor(DATA_3D, affine=torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])),
-        MetaTensor(POINT_3D_IMAGE, affine=torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])),
+        (MetaTensor(DATA_2D, affine=AFFINE_1), MetaTensor(DATA_3D, affine=AFFINE_2)),
+        [MetaTensor(POINT_2D_IMAGE, affine=AFFINE_1), MetaTensor(POINT_3D_IMAGE, affine=AFFINE_2)],
         None,
         False,
         False,
-        POINT_3D_WORLD,
-    ],
-    [
-        MetaTensor(DATA_3D, affine=torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]])),
-        POINT_3D_WORLD,
-        None,
-        True,
-        True,
-        POINT_3D_IMAGE_RAS,
+        ["image_1", "image_2"],
+        [POINT_2D_WORLD, POINT_3D_WORLD],
     ],
 ]
 
 TEST_CASES_WRONG = [
-    [POINT_2D_WORLD, True, None],
-    [POINT_2D_WORLD.unsqueeze(0), False, None],
-    [POINT_3D_WORLD[..., 0:1], False, None],
-    [POINT_3D_WORLD, False, torch.tensor([[[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]]])],
+    [POINT_2D_WORLD, True, None, None],
+    [POINT_2D_WORLD.unsqueeze(0), False, None, None],
+    [POINT_3D_WORLD[..., 0:1], False, None, None],
+    [POINT_3D_WORLD, False, torch.tensor([[[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]]]), None],
+    [POINT_3D_WORLD, False, None, "image"],
+    [POINT_3D_WORLD, False, None, []],
 ]
 
 
@@ -107,10 +125,10 @@ def test_transform_coordinates(self, image, points, affine, invert_affine, affin
             "point": points,
             "affine": torch.tensor([[1, 0, 0, 10], [0, 1, 0, -4], [0, 0, 1, 0], [0, 0, 0, 1]]),
         }
-        refer_key = "image" if (image is not None and image != "affine") else image
+        refer_keys = "image" if (image is not None and image != "affine") else image
         transform = ApplyTransformToPointsd(
             keys="point",
-            refer_key=refer_key,
+            refer_keys=refer_keys,
             dtype=torch.int64,
             affine=affine,
             invert_affine=invert_affine,
@@ -122,11 +140,45 @@ def test_transform_coordinates(self, image, points, affine, invert_affine, affin
         invert_out = transform.inverse(output)
         self.assertTrue(torch.allclose(invert_out["point"], points))
 
+    @parameterized.expand(TEST_CASES_SEQUENCE)
+    def test_transform_coordinates_sequences(
+        self, image, points, affine, invert_affine, affine_lps_to_ras, refer_keys, expected_output
+    ):
+        data = {"image_1": image[0], "image_2": image[1], "point_1": points[0], "point_2": points[1]}
+        keys = ["point_1", "point_2"]
+        transform = ApplyTransformToPointsd(
+            keys=keys,
+            refer_keys=refer_keys,
+            dtype=torch.int64,
+            affine=affine,
+            invert_affine=invert_affine,
+            affine_lps_to_ras=affine_lps_to_ras,
+        )
+        output = transform(data)
+
+        self.assertTrue(torch.allclose(output["point_1"], expected_output[0]))
+        self.assertTrue(torch.allclose(output["point_2"], expected_output[1]))
+        invert_out = transform.inverse(output)
+        self.assertTrue(torch.allclose(invert_out["point_1"], points[0]))
+
     @parameterized.expand(TEST_CASES_WRONG)
-    def test_wrong_input(self, input, invert_affine, affine):
-        transform = ApplyTransformToPointsd(keys="point", dtype=torch.int64, invert_affine=invert_affine, affine=affine)
-        with self.assertRaises(ValueError):
-            transform({"point": input})
+    def test_wrong_input(self, input, invert_affine, affine, refer_keys):
+        if refer_keys == []:
+            with self.assertRaises(ValueError):
+                ApplyTransformToPointsd(
+                    keys="point", dtype=torch.int64, invert_affine=invert_affine, affine=affine, refer_keys=refer_keys
+                )
+        else:
+            transform = ApplyTransformToPointsd(
+                keys="point", dtype=torch.int64, invert_affine=invert_affine, affine=affine, refer_keys=refer_keys
+            )
+            data = {"point": input}
+            if refer_keys == "image":
+                with self.assertRaises(KeyError):
+                    transform(data)
+            else:
+                with self.assertRaises(ValueError):
+                    transform(data)
 
 
 if __name__ == "__main__":

From 19cc6f01766120132f964beecb06d1d561f83801 Mon Sep 17 00:00:00 2001
From: "Wei_Chuan, Chiang" <45346252+slicepaste@users.noreply.github.com>
Date: Wed, 4 Sep 2024 18:42:49 +0800
Subject: [PATCH 149/183] Make MetaTensor optional printed in DataStats and
 DataStatsd #5905 (#7814)

Fixes #5905

### Description

We simply add one argument for DataStats and DataStatsd to make
MetaTensor optional printed.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Wei_Chuan, Chiang <jamesis1356@gmail.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Signed-off-by: Suraj Pai <b.pai@maastrichtuniversity.nl>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: Suraj Pai <b.pai@maastrichtuniversity.nl>
Co-authored-by: Ben Murray <ben.murray@gmail.com>
---
 monai/transforms/utility/array.py      |  7 ++++
 monai/transforms/utility/dictionary.py | 28 +++++++++++--
 tests/test_data_stats.py               | 41 ++++++++++++++++++-
 tests/test_data_statsd.py              | 54 +++++++++++++++++++++++++-
 4 files changed, 123 insertions(+), 7 deletions(-)

diff --git a/monai/transforms/utility/array.py b/monai/transforms/utility/array.py
index bfd2f506c2..72dd189009 100644
--- a/monai/transforms/utility/array.py
+++ b/monai/transforms/utility/array.py
@@ -656,6 +656,7 @@ def __init__(
         data_shape: bool = True,
         value_range: bool = True,
         data_value: bool = False,
+        meta_info: bool = False,
         additional_info: Callable | None = None,
         name: str = "DataStats",
     ) -> None:
@@ -667,6 +668,7 @@ def __init__(
             value_range: whether to show the value range of input data.
             data_value: whether to show the raw value of input data.
                 a typical example is to print some properties of Nifti image: affine, pixdim, etc.
+            meta_info: whether to show the data of MetaTensor.
             additional_info: user can define callable function to extract additional info from input data.
             name: identifier of `logging.logger` to use, defaulting to "DataStats".
 
@@ -681,6 +683,7 @@ def __init__(
         self.data_shape = data_shape
         self.value_range = value_range
         self.data_value = data_value
+        self.meta_info = meta_info
         if additional_info is not None and not callable(additional_info):
             raise TypeError(f"additional_info must be None or callable but is {type(additional_info).__name__}.")
         self.additional_info = additional_info
@@ -707,6 +710,7 @@ def __call__(
         data_shape: bool | None = None,
         value_range: bool | None = None,
         data_value: bool | None = None,
+        meta_info: bool | None = None,
         additional_info: Callable | None = None,
     ) -> NdarrayOrTensor:
         """
@@ -727,6 +731,9 @@ def __call__(
                 lines.append(f"Value range: (not a PyTorch or Numpy array, type: {type(img)})")
         if self.data_value if data_value is None else data_value:
             lines.append(f"Value: {img}")
+        if self.meta_info if meta_info is None else meta_info:
+            metadata = getattr(img, "meta", "(input is not a MetaTensor)")
+            lines.append(f"Meta info: {repr(metadata)}")
         additional_info = self.additional_info if additional_info is None else additional_info
         if additional_info is not None:
             lines.append(f"Additional info: {additional_info(img)}")
diff --git a/monai/transforms/utility/dictionary.py b/monai/transforms/utility/dictionary.py
index db5f19c0de..79d0be522d 100644
--- a/monai/transforms/utility/dictionary.py
+++ b/monai/transforms/utility/dictionary.py
@@ -793,6 +793,7 @@ def __init__(
         data_shape: Sequence[bool] | bool = True,
         value_range: Sequence[bool] | bool = True,
         data_value: Sequence[bool] | bool = False,
+        meta_info: Sequence[bool] | bool = False,
         additional_info: Sequence[Callable] | Callable | None = None,
         name: str = "DataStats",
         allow_missing_keys: bool = False,
@@ -812,6 +813,8 @@ def __init__(
             data_value: whether to show the raw value of input data.
                 it also can be a sequence of bool, each element corresponds to a key in ``keys``.
                 a typical example is to print some properties of Nifti image: affine, pixdim, etc.
+            meta_info: whether to show the data of MetaTensor.
+                it also can be a sequence of bool, each element corresponds to a key in ``keys``.
             additional_info: user can define callable function to extract
                 additional info from input data. it also can be a sequence of string, each element
                 corresponds to a key in ``keys``.
@@ -825,15 +828,34 @@ def __init__(
         self.data_shape = ensure_tuple_rep(data_shape, len(self.keys))
         self.value_range = ensure_tuple_rep(value_range, len(self.keys))
         self.data_value = ensure_tuple_rep(data_value, len(self.keys))
+        self.meta_info = ensure_tuple_rep(meta_info, len(self.keys))
         self.additional_info = ensure_tuple_rep(additional_info, len(self.keys))
         self.printer = DataStats(name=name)
 
     def __call__(self, data: Mapping[Hashable, NdarrayOrTensor]) -> dict[Hashable, NdarrayOrTensor]:
         d = dict(data)
-        for key, prefix, data_type, data_shape, value_range, data_value, additional_info in self.key_iterator(
-            d, self.prefix, self.data_type, self.data_shape, self.value_range, self.data_value, self.additional_info
+        for (
+            key,
+            prefix,
+            data_type,
+            data_shape,
+            value_range,
+            data_value,
+            meta_info,
+            additional_info,
+        ) in self.key_iterator(
+            d,
+            self.prefix,
+            self.data_type,
+            self.data_shape,
+            self.value_range,
+            self.data_value,
+            self.meta_info,
+            self.additional_info,
         ):
-            d[key] = self.printer(d[key], prefix, data_type, data_shape, value_range, data_value, additional_info)
+            d[key] = self.printer(
+                d[key], prefix, data_type, data_shape, value_range, data_value, meta_info, additional_info
+            )
         return d
 
 
diff --git a/tests/test_data_stats.py b/tests/test_data_stats.py
index 05453b0694..f9b424f8e1 100644
--- a/tests/test_data_stats.py
+++ b/tests/test_data_stats.py
@@ -23,6 +23,7 @@
 import torch
 from parameterized import parameterized
 
+from monai.data.meta_tensor import MetaTensor
 from monai.transforms import DataStats
 
 TEST_CASE_1 = [
@@ -130,20 +131,55 @@
 ]
 
 TEST_CASE_8 = [
+    {
+        "prefix": "test data",
+        "data_type": True,
+        "data_shape": True,
+        "value_range": True,
+        "data_value": True,
+        "additional_info": np.mean,
+        "name": "DataStats",
+    },
     np.array([[0, 1], [1, 2]]),
     "test data statistics:\nType: <class 'numpy.ndarray'> int64\nShape: (2, 2)\nValue range: (0, 2)\n"
     "Value: [[0 1]\n [1 2]]\nAdditional info: 1.0\n",
 ]
 
+TEST_CASE_9 = [
+    np.array([[0, 1], [1, 2]]),
+    "test data statistics:\nType: <class 'numpy.ndarray'> int64\nShape: (2, 2)\nValue range: (0, 2)\n"
+    "Value: [[0 1]\n [1 2]]\n"
+    "Meta info: '(input is not a MetaTensor)'\n"
+    "Additional info: 1.0\n",
+]
+
+TEST_CASE_10 = [
+    MetaTensor(
+        torch.tensor([[0, 1], [1, 2]]),
+        affine=torch.as_tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 2, 0], [0, 0, 0, 1]], dtype=torch.float64),
+        meta={"some": "info"},
+    ),
+    "test data statistics:\nType: <class 'monai.data.meta_tensor.MetaTensor'> torch.int64\n"
+    "Shape: torch.Size([2, 2])\nValue range: (0, 2)\n"
+    "Value: tensor([[0, 1],\n        [1, 2]])\n"
+    "Meta info: {'some': 'info', affine: tensor([[2., 0., 0., 0.],\n"
+    "        [0., 2., 0., 0.],\n"
+    "        [0., 0., 2., 0.],\n"
+    "        [0., 0., 0., 1.]], dtype=torch.float64), space: RAS}\n"
+    "Additional info: 1.0\n",
+]
+
 
 class TestDataStats(unittest.TestCase):
 
-    @parameterized.expand([TEST_CASE_1, TEST_CASE_2, TEST_CASE_3, TEST_CASE_4, TEST_CASE_5, TEST_CASE_6, TEST_CASE_7])
+    @parameterized.expand(
+        [TEST_CASE_1, TEST_CASE_2, TEST_CASE_3, TEST_CASE_4, TEST_CASE_5, TEST_CASE_6, TEST_CASE_7, TEST_CASE_8]
+    )
     def test_value(self, input_param, input_data, expected_print):
         transform = DataStats(**input_param)
         _ = transform(input_data)
 
-    @parameterized.expand([TEST_CASE_8])
+    @parameterized.expand([TEST_CASE_9, TEST_CASE_10])
     def test_file(self, input_data, expected_print):
         with tempfile.TemporaryDirectory() as tempdir:
             filename = os.path.join(tempdir, "test_data_stats.log")
@@ -158,6 +194,7 @@ def test_file(self, input_data, expected_print):
                 "data_shape": True,
                 "value_range": True,
                 "data_value": True,
+                "meta_info": True,
                 "additional_info": np.mean,
                 "name": name,
             }
diff --git a/tests/test_data_statsd.py b/tests/test_data_statsd.py
index ef88300c10..a28a938c40 100644
--- a/tests/test_data_statsd.py
+++ b/tests/test_data_statsd.py
@@ -21,6 +21,7 @@
 import torch
 from parameterized import parameterized
 
+from monai.data.meta_tensor import MetaTensor
 from monai.transforms import DataStatsd
 
 TEST_CASE_1 = [
@@ -150,22 +151,70 @@
 ]
 
 TEST_CASE_9 = [
+    {
+        "keys": "img",
+        "prefix": "test data",
+        "data_shape": True,
+        "value_range": True,
+        "data_value": True,
+        "meta_info": False,
+        "additional_info": np.mean,
+        "name": "DataStats",
+    },
     {"img": np.array([[0, 1], [1, 2]])},
     "test data statistics:\nType: <class 'numpy.ndarray'> int64\nShape: (2, 2)\nValue range: (0, 2)\n"
     "Value: [[0 1]\n [1 2]]\nAdditional info: 1.0\n",
 ]
 
+TEST_CASE_10 = [
+    {"img": np.array([[0, 1], [1, 2]])},
+    "test data statistics:\nType: <class 'numpy.ndarray'> int64\nShape: (2, 2)\nValue range: (0, 2)\n"
+    "Value: [[0 1]\n [1 2]]\n"
+    "Meta info: '(input is not a MetaTensor)'\n"
+    "Additional info: 1.0\n",
+]
+
+TEST_CASE_11 = [
+    {
+        "img": (
+            MetaTensor(
+                torch.tensor([[0, 1], [1, 2]]),
+                affine=torch.as_tensor([[2, 0, 0, 0], [0, 2, 0, 0], [0, 0, 2, 0], [0, 0, 0, 1]], dtype=torch.float64),
+                meta={"some": "info"},
+            )
+        )
+    },
+    "test data statistics:\nType: <class 'monai.data.meta_tensor.MetaTensor'> torch.int64\n"
+    "Shape: torch.Size([2, 2])\nValue range: (0, 2)\n"
+    "Value: tensor([[0, 1],\n        [1, 2]])\n"
+    "Meta info: {'some': 'info', affine: tensor([[2., 0., 0., 0.],\n"
+    "        [0., 2., 0., 0.],\n"
+    "        [0., 0., 2., 0.],\n"
+    "        [0., 0., 0., 1.]], dtype=torch.float64), space: RAS}\n"
+    "Additional info: 1.0\n",
+]
+
 
 class TestDataStatsd(unittest.TestCase):
 
     @parameterized.expand(
-        [TEST_CASE_1, TEST_CASE_2, TEST_CASE_3, TEST_CASE_4, TEST_CASE_5, TEST_CASE_6, TEST_CASE_7, TEST_CASE_8]
+        [
+            TEST_CASE_1,
+            TEST_CASE_2,
+            TEST_CASE_3,
+            TEST_CASE_4,
+            TEST_CASE_5,
+            TEST_CASE_6,
+            TEST_CASE_7,
+            TEST_CASE_8,
+            TEST_CASE_9,
+        ]
     )
     def test_value(self, input_param, input_data, expected_print):
         transform = DataStatsd(**input_param)
         _ = transform(input_data)
 
-    @parameterized.expand([TEST_CASE_9])
+    @parameterized.expand([TEST_CASE_10, TEST_CASE_11])
     def test_file(self, input_data, expected_print):
         with tempfile.TemporaryDirectory() as tempdir:
             filename = os.path.join(tempdir, "test_stats.log")
@@ -180,6 +229,7 @@ def test_file(self, input_data, expected_print):
                 "data_shape": True,
                 "value_range": True,
                 "data_value": True,
+                "meta_info": True,
                 "additional_info": np.mean,
                 "name": name,
             }

From b539cbb99b263e4b69a8fb2cbc21ef020ff669b2 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Fri, 6 Sep 2024 20:06:11 +0800
Subject: [PATCH 150/183] Suppress deprecated warning when import monai (#8067)

- add packaging in setup.cfg
- fix test_gdsdataset.py issue
- add test_matshow3d to the skip list for min test
- suppress deprecated warning when import monai (workaround for #8060)

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 monai/__init__.py          | 44 +++++++++++++++++++++++++++++++++++++-
 monai/data/image_reader.py |  3 ---
 pyproject.toml             |  1 +
 setup.cfg                  |  1 +
 tests/min_tests.py         |  1 +
 tests/test_gdsdataset.py   |  7 +++---
 6 files changed, 50 insertions(+), 7 deletions(-)

diff --git a/monai/__init__.py b/monai/__init__.py
index cb0ccd36f8..3f6b12c407 100644
--- a/monai/__init__.py
+++ b/monai/__init__.py
@@ -13,9 +13,51 @@
 
 import os
 import sys
-
+import logging
+import warnings
 from ._version import get_versions
 
+
+old_showwarning = warnings.showwarning
+
+
+def custom_warning_handler(message, category, filename, lineno, file=None, line=None):
+    ignore_files = ["ignite/handlers/checkpoint", "modelopt/torch/quantization/tensor_quant"]
+    if any(ignore in filename for ignore in ignore_files):
+        return
+    old_showwarning(message, category, filename, lineno, file, line)
+
+
+class DeprecatedTypesWarningFilter(logging.Filter):
+    def filter(self, record):
+        message_bodies_to_ignore = [
+            "np.bool8",
+            "np.object0",
+            "np.int0",
+            "np.uint0",
+            "np.void0",
+            "np.str0",
+            "np.bytes0",
+            "@validator",
+            "@root_validator",
+            "class-based `config`",
+            "pkg_resources",
+            "Implicitly cleaning up",
+        ]
+        for message in message_bodies_to_ignore:
+            if message in record.getMessage():
+                return False
+        return True
+
+
+# workaround for https://github.com/Project-MONAI/MONAI/issues/8060
+# TODO: remove this workaround after upstream fixed the warning
+# Set the custom warning handler to filter warning
+warnings.showwarning = custom_warning_handler
+# Get the logger for warnings and add the filter to the logger
+logging.getLogger("py.warnings").addFilter(DeprecatedTypesWarningFilter())
+
+
 PY_REQUIRED_MAJOR = 3
 PY_REQUIRED_MINOR = 9
 
diff --git a/monai/data/image_reader.py b/monai/data/image_reader.py
index aab1e03898..f5e199e2a3 100644
--- a/monai/data/image_reader.py
+++ b/monai/data/image_reader.py
@@ -34,9 +34,6 @@
 )
 from monai.utils import MetaKeys, SpaceKeys, TraceKeys, ensure_tuple, optional_import, require_pkg
 
-# workaround for https://github.com/Project-MONAI/MONAI/issues/8061
-warnings.filterwarnings("ignore", category=DeprecationWarning, module="nptyping")
-
 if TYPE_CHECKING:
     import itk
     import nibabel as nib
diff --git a/pyproject.toml b/pyproject.toml
index 53ca608d20..c2ab92a43d 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,6 +4,7 @@ requires = [
   "setuptools",
   "torch>=1.9",
   "ninja",
+  "packaging"
 ]
 
 [tool.black]
diff --git a/setup.cfg b/setup.cfg
index c97118d43a..694dc969d9 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -40,6 +40,7 @@ python_requires = >= 3.9
 setup_requires =
     torch
     ninja
+    packaging
 install_requires =
     torch>=1.9
     numpy>=1.24,<2.0
diff --git a/tests/min_tests.py b/tests/min_tests.py
index 632355b5c6..f39d3f9843 100644
--- a/tests/min_tests.py
+++ b/tests/min_tests.py
@@ -212,6 +212,7 @@ def run_testsuit():
         "test_ultrasound_confidence_map_transform",
         "test_vista3d_utils",
         "test_vista3d_transforms",
+        "test_matshow3d",
     ]
     assert sorted(exclude_cases) == sorted(set(exclude_cases)), f"Duplicated items in {exclude_cases}"
 
diff --git a/tests/test_gdsdataset.py b/tests/test_gdsdataset.py
index f0a419dcf5..5d2e2aa013 100644
--- a/tests/test_gdsdataset.py
+++ b/tests/test_gdsdataset.py
@@ -23,7 +23,7 @@
 from monai.data import GDSDataset, json_hashing
 from monai.transforms import Compose, Flip, Identity, LoadImaged, SimulateDelayd, Transform
 from monai.utils import optional_import
-from tests.utils import TEST_NDARRAYS, assert_allclose
+from tests.utils import TEST_NDARRAYS, assert_allclose, skip_if_no_cuda
 
 _, has_cp = optional_import("cupy")
 nib, has_nib = optional_import("nibabel")
@@ -70,9 +70,9 @@ def __call__(self, data):
         return data
 
 
+@skip_if_no_cuda
 @unittest.skipUnless(has_cp, "Requires CuPy library.")
-@unittest.skipUnless(has_nib, "Requires nibabel package.")
-@unittest.skipUnless(has_kvikio_numpy, "Requires scikit-image library.")
+@unittest.skipUnless(has_cp and has_kvikio_numpy, "Requires CuPy and kvikio library.")
 class TestDataset(unittest.TestCase):
 
     def test_cache(self):
@@ -131,6 +131,7 @@ def test_dtype(self):
                 self.assertEqual(ds[0].dtype, DTYPES[_dtype])
                 self.assertEqual(ds1[0].dtype, DTYPES[_dtype])
 
+    @unittest.skipUnless(has_nib, "Requires nibabel package.")
     @parameterized.expand([TEST_CASE_1, TEST_CASE_2, TEST_CASE_3])
     def test_shape(self, transform, expected_shape):
         test_image = nib.Nifti1Image(np.random.randint(0, 2, size=[128, 128, 128]).astype(float), np.eye(4))

From d02ba11d8069870d71316a616f047c499627c71c Mon Sep 17 00:00:00 2001
From: Suraj Pai <bspai@bwh.harvard.edu>
Date: Sat, 7 Sep 2024 05:56:46 -0400
Subject: [PATCH 151/183] Fix generalized dice computation (#7970)

Fixes #7966

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Suraj Pai <b.pai@maastrichtuniversity.nl>
Signed-off-by: Suraj Pai <bspai@bwh.harvard.edu>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/metrics/generalized_dice.py      | 125 +++++++++++-------
 tests/test_compute_generalized_dice.py | 170 ++++++++++++++++++-------
 2 files changed, 201 insertions(+), 94 deletions(-)

diff --git a/monai/metrics/generalized_dice.py b/monai/metrics/generalized_dice.py
index e56bd46592..516021949b 100644
--- a/monai/metrics/generalized_dice.py
+++ b/monai/metrics/generalized_dice.py
@@ -14,34 +14,47 @@
 import torch
 
 from monai.metrics.utils import do_metric_reduction, ignore_background
-from monai.utils import MetricReduction, Weight, look_up_option
+from monai.utils import MetricReduction, Weight, deprecated_arg, deprecated_arg_default, look_up_option
 
 from .metric import CumulativeIterationMetric
 
 
 class GeneralizedDiceScore(CumulativeIterationMetric):
-    """Compute the Generalized Dice Score metric between tensors, as the complement of the Generalized Dice Loss defined in:
+    """
+    Compute the Generalized Dice Score metric between tensors.
 
+    This metric is the complement of the Generalized Dice Loss defined in:
     Sudre, C. et. al. (2017) Generalised Dice overlap as a deep learning
-        loss function for highly unbalanced segmentations. DLMIA 2017.
+    loss function for highly unbalanced segmentations. DLMIA 2017.
 
-    The inputs `y_pred` and `y` are expected to be one-hot, binarized channel-first
-    or batch-first tensors, i.e., CHW[D] or BCHW[D].
+    The inputs `y_pred` and `y` are expected to be one-hot, binarized batch-first tensors, i.e., NCHW[D].
 
     Example of the typical execution steps of this metric class follows :py:class:`monai.metrics.metric.Cumulative`.
 
     Args:
-        include_background (bool, optional): whether to include the background class (assumed to be in channel 0), in the
+        include_background: Whether to include the background class (assumed to be in channel 0) in the
             score computation. Defaults to True.
-        reduction (str, optional): define mode of reduction to the metrics. Available reduction modes:
-            {``"none"``, ``"mean_batch"``, ``"sum_batch"``}. Default to ``"mean_batch"``. If "none", will not do reduction.
-        weight_type (Union[Weight, str], optional): {``"square"``, ``"simple"``, ``"uniform"``}. Type of function to transform
+        reduction: Define mode of reduction to the metrics. Available reduction modes:
+            {``"none"``, ``"mean"``, ``"sum"``, ``"mean_batch"``, ``"sum_batch"``,
+            ``"mean_channel"``, ``"sum_channel"``}, default to ``"mean"``. if "none", will not do reduction.
+        weight_type: {``"square"``, ``"simple"``, ``"uniform"``}. Type of function to transform
             ground truth volume into a weight factor. Defaults to ``"square"``.
 
     Raises:
-        ValueError: when the `weight_type` is not one of {``"none"``, ``"mean"``, ``"sum"``}.
+        ValueError: When the `reduction` is not one of MetricReduction enum.
     """
 
+    @deprecated_arg_default(
+        "reduction",
+        old_default=MetricReduction.MEAN_BATCH,
+        new_default=MetricReduction.MEAN,
+        since="1.4.0",
+        replaced="1.5.0",
+        msg_suffix=(
+            "Old versions computed `mean` when `mean_batch` was provided due to bug in reduction， "
+            "If you want to retain the old behavior (calculating the mean), please explicitly set the parameter to 'mean'."
+        ),
+    )
     def __init__(
         self,
         include_background: bool = True,
@@ -50,79 +63,90 @@ def __init__(
     ) -> None:
         super().__init__()
         self.include_background = include_background
-        reduction_options = [
-            "none",
-            "mean_batch",
-            "sum_batch",
-            MetricReduction.NONE,
-            MetricReduction.MEAN_BATCH,
-            MetricReduction.SUM_BATCH,
-        ]
-        self.reduction = reduction
-        if self.reduction not in reduction_options:
-            raise ValueError(f"reduction must be one of {reduction_options}")
+        self.reduction = look_up_option(reduction, MetricReduction)
         self.weight_type = look_up_option(weight_type, Weight)
+        self.sum_over_classes = self.reduction in {
+            MetricReduction.SUM,
+            MetricReduction.MEAN,
+            MetricReduction.MEAN_CHANNEL,
+            MetricReduction.SUM_CHANNEL,
+        }
 
     def _compute_tensor(self, y_pred: torch.Tensor, y: torch.Tensor) -> torch.Tensor:  # type: ignore[override]
-        """Computes the Generalized Dice Score and returns a tensor with its per image values.
+        """
+        Computes the Generalized Dice Score and returns a tensor with its per image values.
 
         Args:
-            y_pred (torch.Tensor): binarized segmentation model output. It must be in one-hot format and in the NCHW[D] format,
+            y_pred (torch.Tensor): Binarized segmentation model output. It must be in one-hot format and in the NCHW[D] format,
                 where N is the batch dimension, C is the channel dimension, and the remaining are the spatial dimensions.
-            y (torch.Tensor): binarized ground-truth. It must be in one-hot format and have the same shape as `y_pred`.
+            y (torch.Tensor): Binarized ground-truth. It must be in one-hot format and have the same shape as `y_pred`.
+
+        Returns:
+            torch.Tensor: Generalized Dice Score averaged across batch and class
 
         Raises:
-            ValueError: if `y_pred` and `y` have less than 3 dimensions, or `y_pred` and `y` don't have the same shape.
+            ValueError: If `y_pred` and `y` have less than 3 dimensions, or `y_pred` and `y` don't have the same shape.
         """
         return compute_generalized_dice(
-            y_pred=y_pred, y=y, include_background=self.include_background, weight_type=self.weight_type
+            y_pred=y_pred,
+            y=y,
+            include_background=self.include_background,
+            weight_type=self.weight_type,
+            sum_over_classes=self.sum_over_classes,
         )
 
+    @deprecated_arg(
+        "reduction",
+        since="1.3.3",
+        removed="1.7.0",
+        msg_suffix="Reduction will be ignored. Set reduction during init. as gen.dice needs it during compute",
+    )
     def aggregate(self, reduction: MetricReduction | str | None = None) -> torch.Tensor:
         """
         Execute reduction logic for the output of `compute_generalized_dice`.
 
-        Args:
-            reduction (Union[MetricReduction, str, None], optional): define mode of reduction to the metrics.
-                Available reduction modes: {``"none"``, ``"mean"``, ``"sum"``, ``"mean_batch"``, ``"sum_batch"``}.
-                Defaults to ``"mean"``. If "none", will not do reduction.
+        Returns:
+            torch.Tensor: Aggregated metric value.
+
+        Raises:
+            ValueError: If the data to aggregate is not a PyTorch Tensor.
         """
         data = self.get_buffer()
         if not isinstance(data, torch.Tensor):
             raise ValueError("The data to aggregate must be a PyTorch Tensor.")
 
-        # Validate reduction argument if specified
-        if reduction is not None:
-            reduction_options = ["none", "mean", "sum", "mean_batch", "sum_batch"]
-            if reduction not in reduction_options:
-                raise ValueError(f"reduction must be one of {reduction_options}")
-
         # Do metric reduction and return
-        f, _ = do_metric_reduction(data, reduction or self.reduction)
+        f, _ = do_metric_reduction(data, self.reduction)
 
         return f
 
 
 def compute_generalized_dice(
-    y_pred: torch.Tensor, y: torch.Tensor, include_background: bool = True, weight_type: Weight | str = Weight.SQUARE
+    y_pred: torch.Tensor,
+    y: torch.Tensor,
+    include_background: bool = True,
+    weight_type: Weight | str = Weight.SQUARE,
+    sum_over_classes: bool = False,
 ) -> torch.Tensor:
-    """Computes the Generalized Dice Score and returns a tensor with its per image values.
+    """
+    Computes the Generalized Dice Score and returns a tensor with its per image values.
 
     Args:
-        y_pred (torch.Tensor): binarized segmentation model output. It should be binarized, in one-hot format
+        y_pred (torch.Tensor): Binarized segmentation model output. It should be binarized, in one-hot format
             and in the NCHW[D] format, where N is the batch dimension, C is the channel dimension, and the
             remaining are the spatial dimensions.
-        y (torch.Tensor): binarized ground-truth. It should be binarized, in one-hot format and have the same shape as `y_pred`.
-        include_background (bool, optional): whether to include score computation on the first channel of the
+        y (torch.Tensor): Binarized ground-truth. It should be binarized, in one-hot format and have the same shape as `y_pred`.
+        include_background: Whether to include score computation on the first channel of the
             predicted output. Defaults to True.
         weight_type (Union[Weight, str], optional): {``"square"``, ``"simple"``, ``"uniform"``}. Type of function to
             transform ground truth volume into a weight factor. Defaults to ``"square"``.
+        sum_over_labels (bool): Whether to sum the numerator and denominator across all labels before the final computation.
 
     Returns:
-        torch.Tensor: per batch and per class Generalized Dice Score, i.e., with the shape [batch_size, num_classes].
+        torch.Tensor: Per batch and per class Generalized Dice Score, i.e., with the shape [batch_size, num_classes].
 
     Raises:
-        ValueError: if `y_pred` or `y` are not PyTorch tensors, if `y_pred` and `y` have less than three dimensions,
+        ValueError: If `y_pred` or `y` are not PyTorch tensors, if `y_pred` and `y` have less than three dimensions,
             or `y_pred` and `y` don't have the same shape.
     """
     # Ensure tensors have at least 3 dimensions and have the same shape
@@ -158,16 +182,21 @@ def compute_generalized_dice(
         b[infs] = 0
         b[infs] = torch.max(b)
 
-    # Compute the weighted numerator and denominator, summing along the class axis
-    numer = 2.0 * (intersection * w).sum(dim=1)
-    denom = (denominator * w).sum(dim=1)
+    # Compute the weighted numerator and denominator, summing along the class axis when sum_over_classes is True
+    if sum_over_classes:
+        numer = 2.0 * (intersection * w).sum(dim=1, keepdim=True)
+        denom = (denominator * w).sum(dim=1, keepdim=True)
+        y_pred_o = y_pred_o.sum(dim=-1, keepdim=True)
+    else:
+        numer = 2.0 * (intersection * w)
+        denom = denominator * w
+        y_pred_o = y_pred_o
 
     # Compute the score
     generalized_dice_score = numer / denom
 
     # Handle zero division. Where denom == 0 and the prediction volume is 0, score is 1.
     # Where denom == 0 but the prediction volume is not 0, score is 0
-    y_pred_o = y_pred_o.sum(dim=-1)
     denom_zeros = denom == 0
     generalized_dice_score[denom_zeros] = torch.where(
         (y_pred_o == 0)[denom_zeros],
diff --git a/tests/test_compute_generalized_dice.py b/tests/test_compute_generalized_dice.py
index e04444e988..985a01e993 100644
--- a/tests/test_compute_generalized_dice.py
+++ b/tests/test_compute_generalized_dice.py
@@ -22,17 +22,17 @@
 _device = "cuda:0" if torch.cuda.is_available() else "cpu"
 
 # keep background
-TEST_CASE_1 = [  # y (1, 1, 2, 2), y_pred (1, 1, 2, 2), expected out (1)
+TEST_CASE_1 = [  # y (1, 1, 2, 2), y_pred (1, 1, 2, 2), expected out (1, 1) with compute_generalized_dice
     {
         "y_pred": torch.tensor([[[[1.0, 0.0], [0.0, 1.0]]]], device=_device),
         "y": torch.tensor([[[[1.0, 0.0], [1.0, 1.0]]]], device=_device),
         "include_background": True,
     },
-    [0.8],
+    [[0.8]],
 ]
 
 # remove background
-TEST_CASE_2 = [  # y (2, 1, 2, 2), y_pred (2, 3, 2, 2), expected out (2) (no background)
+TEST_CASE_2 = [  # y (2, 3, 2, 2), y_pred (2, 3, 2, 2), expected out (2) (no background) with GeneralizedDiceScore
     {
         "y_pred": torch.tensor(
             [
@@ -47,32 +47,32 @@
             ]
         ),
         "include_background": False,
+        "reduction": "mean_batch",
     },
-    [0.1667, 0.6667],
+    [0.583333, 0.333333],
 ]
 
-# should return 0 for both cases
-TEST_CASE_3 = [
+TEST_CASE_3 = [  # y (2, 3, 2, 2), y_pred (2, 3, 2, 2), expected out (1) with GeneralizedDiceScore
     {
         "y_pred": torch.tensor(
             [
-                [[[0.0, 0.0], [0.0, 0.0]], [[0.0, 0.0], [0.0, 0.0]], [[1.0, 1.0], [1.0, 1.0]]],
-                [[[0.0, 0.0], [0.0, 0.0]], [[0.0, 0.0], [0.0, 0.0]], [[1.0, 1.0], [1.0, 1.0]]],
+                [[[1.0, 1.0], [1.0, 0.0]], [[0.0, 1.0], [0.0, 0.0]], [[0.0, 1.0], [1.0, 1.0]]],
+                [[[1.0, 0.0], [1.0, 1.0]], [[0.0, 1.0], [1.0, 1.0]], [[0.0, 1.0], [1.0, 0.0]]],
             ]
         ),
         "y": torch.tensor(
             [
                 [[[1.0, 1.0], [1.0, 1.0]], [[0.0, 0.0], [0.0, 0.0]], [[0.0, 0.0], [0.0, 0.0]]],
-                [[[0.0, 1.0], [1.0, 0.0]], [[1.0, 0.0], [0.0, 1.0]], [[0.0, 0.0], [0.0, 0.0]]],
+                [[[0.0, 0.0], [0.0, 1.0]], [[1.0, 1.0], [0.0, 0.0]], [[0.0, 0.0], [1.0, 0.0]]],
             ]
         ),
         "include_background": True,
+        "reduction": "mean",
     },
-    [0.0, 0.0],
+    [0.5454],
 ]
 
-TEST_CASE_4 = [
-    {"include_background": True, "reduction": "mean_batch"},
+TEST_CASE_4 = [  # y (2, 3, 2, 2), y_pred (2, 3, 2, 2), expected out (1) with GeneralizedDiceScore
     {
         "y_pred": torch.tensor(
             [
@@ -83,15 +83,36 @@
         "y": torch.tensor(
             [
                 [[[1.0, 1.0], [1.0, 1.0]], [[0.0, 0.0], [0.0, 0.0]], [[0.0, 0.0], [0.0, 0.0]]],
-                [[[0.0, 0.0], [0.0, 1.0]], [[1.0, 1.0], [0.0, 0.0]], [[0.0, 0.0], [1.0, 0.0]]],
+                [[[1.0, 1.0], [1.0, 1.0]], [[0.0, 0.0], [0.0, 0.0]], [[0.0, 0.0], [0.0, 0.0]]],
             ]
         ),
+        "include_background": True,
+        "reduction": "sum",
     },
-    [0.5455],
+    [1.045455],
+]
+
+TEST_CASE_5 = [  # y (2, 2, 3, 3) y_pred (2, 2, 3, 3) expected out (2, 2) with compute_generalized_dice
+    {"y": torch.ones((2, 2, 3, 3)), "y_pred": torch.ones((2, 2, 3, 3))},
+    [[1.0000, 1.0000], [1.0000, 1.0000]],
 ]
 
-TEST_CASE_5 = [
-    {"include_background": True, "reduction": "sum_batch"},
+TEST_CASE_6 = [  # y (2, 2, 3, 3) y_pred (2, 2, 3, 3) expected out (2, 2) with compute_generalized_dice
+    {"y": torch.zeros((2, 2, 3, 3)), "y_pred": torch.ones((2, 2, 3, 3))},
+    [[0.0000, 0.0000], [0.0000, 0.0000]],
+]
+
+TEST_CASE_7 = [  # y (2, 2, 3, 3) y_pred (2, 2, 3, 3) expected out (2, 2) with compute_generalized_dice
+    {"y": torch.ones((2, 2, 3, 3)), "y_pred": torch.zeros((2, 2, 3, 3))},
+    [[0.0000, 0.0000], [0.0000, 0.0000]],
+]
+
+TEST_CASE_8 = [  # y (2, 2, 3, 3) y_pred (2, 2, 3, 3) expected out (2, 2) with compute_generalized_dice
+    {"y": torch.zeros((2, 2, 3, 3)), "y_pred": torch.zeros((2, 2, 3, 3))},
+    [[1.0000, 1.0000], [1.0000, 1.0000]],
+]
+
+TEST_CASE_9 = [  # y (2, 3, 2, 2) y_pred (2, 3, 2, 2) expected out (2) with GeneralizedDiceScore
     {
         "y_pred": torch.tensor(
             [
@@ -102,61 +123,118 @@
         "y": torch.tensor(
             [
                 [[[1.0, 1.0], [1.0, 1.0]], [[0.0, 0.0], [0.0, 0.0]], [[0.0, 0.0], [0.0, 0.0]]],
-                [[[1.0, 1.0], [1.0, 1.0]], [[0.0, 0.0], [0.0, 0.0]], [[0.0, 0.0], [0.0, 0.0]]],
+                [[[0.0, 0.0], [0.0, 1.0]], [[1.0, 1.0], [0.0, 0.0]], [[0.0, 0.0], [1.0, 0.0]]],
             ]
         ),
+        "include_background": True,
+        "reduction": "mean_channel",
     },
-    1.0455,
+    [0.545455, 0.545455],
 ]
 
-TEST_CASE_6 = [{"y": torch.ones((2, 2, 3, 3)), "y_pred": torch.ones((2, 2, 3, 3))}, [1.0000, 1.0000]]
 
-TEST_CASE_7 = [{"y": torch.zeros((2, 2, 3, 3)), "y_pred": torch.ones((2, 2, 3, 3))}, [0.0000, 0.0000]]
-
-TEST_CASE_8 = [{"y": torch.ones((2, 2, 3, 3)), "y_pred": torch.zeros((2, 2, 3, 3))}, [0.0000, 0.0000]]
+TEST_CASE_10 = [  # y (2, 3, 2, 2) y_pred (2, 3, 2, 2) expected out (2, 3) with compute_generalized_dice
+    # and (3) with GeneralizedDiceScore "mean_batch"
+    {
+        "y_pred": torch.tensor(
+            [
+                [[[1.0, 1.0], [1.0, 0.0]], [[0.0, 1.0], [0.0, 0.0]], [[0.0, 1.0], [1.0, 1.0]]],
+                [[[1.0, 0.0], [1.0, 1.0]], [[0.0, 1.0], [1.0, 1.0]], [[0.0, 1.0], [1.0, 0.0]]],
+            ]
+        ),
+        "y": torch.tensor(
+            [
+                [[[1.0, 1.0], [1.0, 1.0]], [[0.0, 0.0], [0.0, 0.0]], [[0.0, 0.0], [0.0, 0.0]]],
+                [[[0.0, 0.0], [0.0, 1.0]], [[1.0, 1.0], [0.0, 0.0]], [[0.0, 0.0], [1.0, 0.0]]],
+            ]
+        ),
+        "include_background": True,
+    },
+    [[0.857143, 0.0, 0.0], [0.5, 0.4, 0.666667]],
+]
 
-TEST_CASE_9 = [{"y": torch.zeros((2, 2, 3, 3)), "y_pred": torch.zeros((2, 2, 3, 3))}, [1.0000, 1.0000]]
+TEST_CASE_11 = [  # y (2, 3, 2, 2) y_pred (2, 3, 2, 2) expected out (2, 1) with compute_generalized_dice (summed over classes)
+    # and (2) with GeneralizedDiceScore "mean_channel"
+    {
+        "y_pred": torch.tensor(
+            [
+                [[[1.0, 1.0], [1.0, 0.0]], [[0.0, 1.0], [0.0, 0.0]], [[0.0, 1.0], [1.0, 1.0]]],
+                [[[1.0, 0.0], [1.0, 1.0]], [[0.0, 1.0], [1.0, 1.0]], [[0.0, 1.0], [1.0, 0.0]]],
+            ]
+        ),
+        "y": torch.tensor(
+            [
+                [[[1.0, 1.0], [1.0, 1.0]], [[0.0, 0.0], [0.0, 0.0]], [[0.0, 0.0], [0.0, 0.0]]],
+                [[[0.0, 0.0], [0.0, 1.0]], [[1.0, 1.0], [0.0, 0.0]], [[0.0, 0.0], [1.0, 0.0]]],
+            ]
+        ),
+        "include_background": True,
+        "sum_over_classes": True,
+    },
+    [[0.545455], [0.545455]],
+]
 
 
 class TestComputeGeneralizedDiceScore(unittest.TestCase):
-
     @parameterized.expand([TEST_CASE_1])
     def test_device(self, input_data, _expected_value):
+        """
+        Test if the result tensor is on the same device as the input tensor.
+        """
         result = compute_generalized_dice(**input_data)
         np.testing.assert_equal(result.device, input_data["y_pred"].device)
 
-    # Functional part tests
-    @parameterized.expand([TEST_CASE_1, TEST_CASE_2, TEST_CASE_6, TEST_CASE_7, TEST_CASE_8, TEST_CASE_9])
+    @parameterized.expand([TEST_CASE_1, TEST_CASE_5, TEST_CASE_6, TEST_CASE_7, TEST_CASE_8])
     def test_value(self, input_data, expected_value):
+        """
+        Test if the computed generalized dice score matches the expected value.
+        """
         result = compute_generalized_dice(**input_data)
         np.testing.assert_allclose(result.cpu().numpy(), expected_value, atol=1e-4)
 
-    # Functional part tests
-    @parameterized.expand([TEST_CASE_3])
-    def test_nans(self, input_data, expected_value):
-        result = compute_generalized_dice(**input_data)
-        self.assertTrue(np.allclose(np.isnan(result.cpu().numpy()), expected_value))
-
-    # Samplewise tests
-    @parameterized.expand([TEST_CASE_1, TEST_CASE_2])
+    @parameterized.expand([TEST_CASE_2, TEST_CASE_3, TEST_CASE_4, TEST_CASE_9])
     def test_value_class(self, input_data, expected_value):
-        # same test as for compute_meandice
-        vals = {}
-        vals["y_pred"] = input_data.pop("y_pred")
-        vals["y"] = input_data.pop("y")
+        """
+        Test if the GeneralizedDiceScore class computes the correct values.
+        """
+        y_pred = input_data.pop("y_pred")
+        y = input_data.pop("y")
         generalized_dice_score = GeneralizedDiceScore(**input_data)
-        generalized_dice_score(**vals)
-        result = generalized_dice_score.aggregate(reduction="none")
+        generalized_dice_score(y_pred=y_pred, y=y)
+        result = generalized_dice_score.aggregate()
         np.testing.assert_allclose(result.cpu().numpy(), expected_value, atol=1e-4)
 
-    # Aggregation tests
-    @parameterized.expand([TEST_CASE_4, TEST_CASE_5])
-    def test_nans_class(self, params, input_data, expected_value):
-        generalized_dice_score = GeneralizedDiceScore(**params)
-        generalized_dice_score(**input_data)
-        result = generalized_dice_score.aggregate()
+    @parameterized.expand([TEST_CASE_10])
+    def test_values_compare(self, input_data, expected_value):
+        """
+        Compare the results of compute_generalized_dice function and GeneralizedDiceScore class.
+        """
+        result = compute_generalized_dice(**input_data)
         np.testing.assert_allclose(result.cpu().numpy(), expected_value, atol=1e-4)
 
+        y_pred = input_data.pop("y_pred")
+        y = input_data.pop("y")
+        generalized_dice_score = GeneralizedDiceScore(**input_data, reduction="mean_batch")
+        generalized_dice_score(y_pred=y_pred, y=y)
+        result_class_mean = generalized_dice_score.aggregate()
+        np.testing.assert_allclose(result_class_mean.cpu().numpy(), np.mean(expected_value, axis=0), atol=1e-4)
+
+    @parameterized.expand([TEST_CASE_11])
+    def test_values_compare_sum_over_classes(self, input_data, expected_value):
+        """
+        Compare the results when summing over classes between compute_generalized_dice function and GeneralizedDiceScore class.
+        """
+        result = compute_generalized_dice(**input_data)
+        np.testing.assert_allclose(result.cpu().numpy(), expected_value, atol=1e-4)
+
+        y_pred = input_data.pop("y_pred")
+        y = input_data.pop("y")
+        input_data.pop("sum_over_classes")
+        generalized_dice_score = GeneralizedDiceScore(**input_data, reduction="mean_channel")
+        generalized_dice_score(y_pred=y_pred, y=y)
+        result_class_mean = generalized_dice_score.aggregate()
+        np.testing.assert_allclose(result_class_mean.cpu().numpy(), np.mean(expected_value, axis=1), atol=1e-4)
+
 
 if __name__ == "__main__":
     unittest.main()

From 64eee8cb9cfad9ef5bd3eaf597fef0fbe85144b4 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 10 Sep 2024 17:49:57 +0800
Subject: [PATCH 152/183] Fix FileNotFoundError when download bundle (#8076)

Fixes #8075


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/bundle/scripts.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index f1d1286e4b..f31ad0e814 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -255,6 +255,7 @@ def _download_from_ngc_private(
     else:
         raise ValueError("NGC API requires requests package. Please install it.")
 
+    os.makedirs(download_path, exist_ok=True)
     zip_path = download_path / f"{filename}_v{version}.zip"
     with open(zip_path, "wb") as f:
         f.write(response.content)

From 3282e45cfb2770d564552c8f80954ff61ff717cf Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 16 Sep 2024 12:47:55 +0800
Subject: [PATCH 153/183] Remove deprecated args and class for v1.4 (#8079)

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/bundle/workflows.py               |  7 +++----
 monai/handlers/__init__.py              |  2 +-
 monai/handlers/ignite_metric.py         | 24 +-----------------------
 monai/losses/dice.py                    | 17 +----------------
 monai/networks/blocks/patchembedding.py |  8 +-------
 monai/networks/nets/unetr.py            |  9 +--------
 monai/networks/nets/vit.py              |  8 --------
 monai/networks/nets/vitautoenc.py       |  9 +--------
 monai/utils/enums.py                    | 13 -------------
 monai/utils/misc.py                     |  2 +-
 10 files changed, 10 insertions(+), 89 deletions(-)

diff --git a/monai/bundle/workflows.py b/monai/bundle/workflows.py
index 11c9bf0562..d728d7d930 100644
--- a/monai/bundle/workflows.py
+++ b/monai/bundle/workflows.py
@@ -26,7 +26,7 @@
 from monai.bundle.properties import InferProperties, MetaProperties, TrainProperties
 from monai.bundle.utils import DEFAULT_EXP_MGMT_SETTINGS, EXPR_KEY, ID_REF_KEY, ID_SEP_KEY
 from monai.config import PathLike
-from monai.utils import BundleProperty, BundlePropertyConfig, deprecated_arg, deprecated_arg_default, ensure_tuple
+from monai.utils import BundleProperty, BundlePropertyConfig, deprecated_arg, ensure_tuple
 
 __all__ = ["BundleWorkflow", "ConfigWorkflow"]
 
@@ -43,7 +43,7 @@ class BundleWorkflow(ABC):
         workflow_type: specifies the workflow type: "train" or "training" for a training workflow,
             or "infer", "inference", "eval", "evaluation" for a inference workflow,
             other unsupported string will raise a ValueError.
-            default to `None` for common workflow.
+            default to `train` for train workflow.
         workflow: specifies the workflow type: "train" or "training" for a training workflow,
             or "infer", "inference", "eval", "evaluation" for a inference workflow,
             other unsupported string will raise a ValueError.
@@ -274,7 +274,6 @@ class ConfigWorkflow(BundleWorkflow):
         new_name="workflow_type",
         msg_suffix="please use `workflow_type` instead.",
     )
-    @deprecated_arg_default("workflow_type", None, "train", since="1.2", replaced="1.4")
     def __init__(
         self,
         config_file: str | Sequence[str],
@@ -284,7 +283,7 @@ def __init__(
         run_id: str = "run",
         final_id: str = "finalize",
         tracking: str | dict | None = None,
-        workflow_type: str | None = None,
+        workflow_type: str | None = "train",
         workflow: str | None = None,
         properties_path: PathLike | None = None,
         **override: Any,
diff --git a/monai/handlers/__init__.py b/monai/handlers/__init__.py
index fa6e158be8..c1fa448f25 100644
--- a/monai/handlers/__init__.py
+++ b/monai/handlers/__init__.py
@@ -20,7 +20,7 @@
 from .earlystop_handler import EarlyStopHandler
 from .garbage_collector import GarbageCollector
 from .hausdorff_distance import HausdorffDistance
-from .ignite_metric import IgniteMetric, IgniteMetricHandler
+from .ignite_metric import IgniteMetricHandler
 from .logfile_handler import LogfileHandler
 from .lr_schedule_handler import LrScheduleHandler
 from .mean_dice import MeanDice
diff --git a/monai/handlers/ignite_metric.py b/monai/handlers/ignite_metric.py
index 021154d705..64ded4d5ea 100644
--- a/monai/handlers/ignite_metric.py
+++ b/monai/handlers/ignite_metric.py
@@ -20,7 +20,7 @@
 
 from monai.config import IgniteInfo
 from monai.metrics import CumulativeIterationMetric, LossMetric
-from monai.utils import MetricReduction, deprecated, min_version, optional_import
+from monai.utils import MetricReduction, min_version, optional_import
 
 idist, _ = optional_import("ignite", IgniteInfo.OPT_IMPORT_VERSION, min_version, "distributed")
 
@@ -153,25 +153,3 @@ def attach(self, engine: Engine, name: str) -> None:  # type: ignore[override]
         self._name = name
         if self.save_details and not hasattr(engine.state, "metric_details"):
             engine.state.metric_details = {}  # type: ignore
-
-
-@deprecated(since="1.2", removed="1.4", msg_suffix="Use IgniteMetricHandler instead of IgniteMetric.")
-class IgniteMetric(IgniteMetricHandler):
-
-    def __init__(
-        self,
-        metric_fn: CumulativeIterationMetric | None = None,
-        loss_fn: _Loss | None = None,
-        output_transform: Callable = lambda x: x,
-        save_details: bool = True,
-        reduction: MetricReduction | str = MetricReduction.MEAN,
-        get_not_nans: bool = False,
-    ) -> None:
-        super().__init__(
-            metric_fn=metric_fn,
-            loss_fn=loss_fn,
-            output_transform=output_transform,
-            save_details=save_details,
-            reduction=reduction,
-            get_not_nans=get_not_nans,
-        )
diff --git a/monai/losses/dice.py b/monai/losses/dice.py
index 44cde41e5d..3f02fae6b8 100644
--- a/monai/losses/dice.py
+++ b/monai/losses/dice.py
@@ -24,7 +24,7 @@
 from monai.losses.focal_loss import FocalLoss
 from monai.losses.spatial_mask import MaskedLoss
 from monai.networks import one_hot
-from monai.utils import DiceCEReduction, LossReduction, Weight, deprecated_arg, look_up_option, pytorch_after
+from monai.utils import DiceCEReduction, LossReduction, Weight, look_up_option, pytorch_after
 
 
 class DiceLoss(_Loss):
@@ -646,9 +646,6 @@ class DiceCELoss(_Loss):
 
     """
 
-    @deprecated_arg(
-        "ce_weight", since="1.2", removed="1.4", new_name="weight", msg_suffix="please use `weight` instead."
-    )
     def __init__(
         self,
         include_background: bool = True,
@@ -662,7 +659,6 @@ def __init__(
         smooth_nr: float = 1e-5,
         smooth_dr: float = 1e-5,
         batch: bool = False,
-        ce_weight: torch.Tensor | None = None,
         weight: torch.Tensor | None = None,
         lambda_dice: float = 1.0,
         lambda_ce: float = 1.0,
@@ -712,7 +708,6 @@ def __init__(
         """
         super().__init__()
         reduction = look_up_option(reduction, DiceCEReduction).value
-        weight = ce_weight if ce_weight is not None else weight
         dice_weight: torch.Tensor | None
         if weight is not None and not include_background:
             dice_weight = weight[1:]
@@ -825,9 +820,6 @@ class DiceFocalLoss(_Loss):
 
     """
 
-    @deprecated_arg(
-        "focal_weight", since="1.2", removed="1.4", new_name="weight", msg_suffix="please use `weight` instead."
-    )
     def __init__(
         self,
         include_background: bool = True,
@@ -842,7 +834,6 @@ def __init__(
         smooth_dr: float = 1e-5,
         batch: bool = False,
         gamma: float = 2.0,
-        focal_weight: Sequence[float] | float | int | torch.Tensor | None = None,
         weight: Sequence[float] | float | int | torch.Tensor | None = None,
         lambda_dice: float = 1.0,
         lambda_focal: float = 1.0,
@@ -885,7 +876,6 @@ def __init__(
                 [0, 1]. Defaults to None.
         """
         super().__init__()
-        weight = focal_weight if focal_weight is not None else weight
         self.dice = DiceLoss(
             include_background=include_background,
             to_onehot_y=False,
@@ -994,9 +984,6 @@ class GeneralizedDiceFocalLoss(_Loss):
         ValueError: if either `lambda_gdl` or `lambda_focal` is less than 0.
     """
 
-    @deprecated_arg(
-        "focal_weight", since="1.2", removed="1.4", new_name="weight", msg_suffix="please use `weight` instead."
-    )
     def __init__(
         self,
         include_background: bool = True,
@@ -1010,7 +997,6 @@ def __init__(
         smooth_dr: float = 1e-5,
         batch: bool = False,
         gamma: float = 2.0,
-        focal_weight: Sequence[float] | float | int | torch.Tensor | None = None,
         weight: Sequence[float] | float | int | torch.Tensor | None = None,
         lambda_gdl: float = 1.0,
         lambda_focal: float = 1.0,
@@ -1028,7 +1014,6 @@ def __init__(
             smooth_dr=smooth_dr,
             batch=batch,
         )
-        weight = focal_weight if focal_weight is not None else weight
         self.focal = FocalLoss(
             include_background=include_background,
             to_onehot_y=to_onehot_y,
diff --git a/monai/networks/blocks/patchembedding.py b/monai/networks/blocks/patchembedding.py
index 91bd73ebbb..fca566591a 100644
--- a/monai/networks/blocks/patchembedding.py
+++ b/monai/networks/blocks/patchembedding.py
@@ -21,7 +21,7 @@
 
 from monai.networks.blocks.pos_embed_utils import build_sincos_position_embedding
 from monai.networks.layers import Conv, trunc_normal_
-from monai.utils import deprecated_arg, ensure_tuple_rep, optional_import
+from monai.utils import ensure_tuple_rep, optional_import
 from monai.utils.module import look_up_option
 
 Rearrange, _ = optional_import("einops.layers.torch", name="Rearrange")
@@ -42,9 +42,6 @@ class PatchEmbeddingBlock(nn.Module):
 
     """
 
-    @deprecated_arg(
-        name="pos_embed", since="1.2", removed="1.4", new_name="proj_type", msg_suffix="please use `proj_type` instead."
-    )
     def __init__(
         self,
         in_channels: int,
@@ -52,7 +49,6 @@ def __init__(
         patch_size: Sequence[int] | int,
         hidden_size: int,
         num_heads: int,
-        pos_embed: str = "conv",
         proj_type: str = "conv",
         pos_embed_type: str = "learnable",
         dropout_rate: float = 0.0,
@@ -69,8 +65,6 @@ def __init__(
             pos_embed_type: position embedding layer type.
             dropout_rate: fraction of the input units to drop.
             spatial_dims: number of spatial dimensions.
-        .. deprecated:: 1.4
-            ``pos_embed`` is deprecated in favor of ``proj_type``.
         """
 
         super().__init__()
diff --git a/monai/networks/nets/unetr.py b/monai/networks/nets/unetr.py
index a88e5a92fd..79ea0e23f7 100644
--- a/monai/networks/nets/unetr.py
+++ b/monai/networks/nets/unetr.py
@@ -18,7 +18,7 @@
 from monai.networks.blocks.dynunet_block import UnetOutBlock
 from monai.networks.blocks.unetr_block import UnetrBasicBlock, UnetrPrUpBlock, UnetrUpBlock
 from monai.networks.nets.vit import ViT
-from monai.utils import deprecated_arg, ensure_tuple_rep
+from monai.utils import ensure_tuple_rep
 
 
 class UNETR(nn.Module):
@@ -27,9 +27,6 @@ class UNETR(nn.Module):
     UNETR: Transformers for 3D Medical Image Segmentation <https://arxiv.org/abs/2103.10504>"
     """
 
-    @deprecated_arg(
-        name="pos_embed", since="1.2", removed="1.4", new_name="proj_type", msg_suffix="please use `proj_type` instead."
-    )
     def __init__(
         self,
         in_channels: int,
@@ -39,7 +36,6 @@ def __init__(
         hidden_size: int = 768,
         mlp_dim: int = 3072,
         num_heads: int = 12,
-        pos_embed: str = "conv",
         proj_type: str = "conv",
         norm_name: tuple | str = "instance",
         conv_block: bool = True,
@@ -67,9 +63,6 @@ def __init__(
             qkv_bias: apply the bias term for the qkv linear layer in self attention block. Defaults to False.
             save_attn: to make accessible the attention in self attention block. Defaults to False.
 
-        .. deprecated:: 1.4
-            ``pos_embed`` is deprecated in favor of ``proj_type``.
-
         Examples::
 
             # for single channel input 4-channel output with image size of (96,96,96), feature size of 32 and batch norm
diff --git a/monai/networks/nets/vit.py b/monai/networks/nets/vit.py
index 4eada6aa76..07c5147cb2 100644
--- a/monai/networks/nets/vit.py
+++ b/monai/networks/nets/vit.py
@@ -18,7 +18,6 @@
 
 from monai.networks.blocks.patchembedding import PatchEmbeddingBlock
 from monai.networks.blocks.transformerblock import TransformerBlock
-from monai.utils import deprecated_arg
 
 __all__ = ["ViT"]
 
@@ -31,9 +30,6 @@ class ViT(nn.Module):
     ViT supports Torchscript but only works for Pytorch after 1.8.
     """
 
-    @deprecated_arg(
-        name="pos_embed", since="1.2", removed="1.4", new_name="proj_type", msg_suffix="please use `proj_type` instead."
-    )
     def __init__(
         self,
         in_channels: int,
@@ -43,7 +39,6 @@ def __init__(
         mlp_dim: int = 3072,
         num_layers: int = 12,
         num_heads: int = 12,
-        pos_embed: str = "conv",
         proj_type: str = "conv",
         pos_embed_type: str = "learnable",
         classification: bool = False,
@@ -75,9 +70,6 @@ def __init__(
             qkv_bias (bool, optional): apply bias to the qkv linear layer in self attention block. Defaults to False.
             save_attn (bool, optional): to make accessible the attention in self attention block. Defaults to False.
 
-        .. deprecated:: 1.4
-            ``pos_embed`` is deprecated in favor of ``proj_type``.
-
         Examples::
 
             # for single channel input with image size of (96,96,96), conv position embedding and segmentation backbone
diff --git a/monai/networks/nets/vitautoenc.py b/monai/networks/nets/vitautoenc.py
index d69f5df4be..3c20f9a784 100644
--- a/monai/networks/nets/vitautoenc.py
+++ b/monai/networks/nets/vitautoenc.py
@@ -20,7 +20,7 @@
 from monai.networks.blocks.patchembedding import PatchEmbeddingBlock
 from monai.networks.blocks.transformerblock import TransformerBlock
 from monai.networks.layers import Conv
-from monai.utils import deprecated_arg, ensure_tuple_rep, is_sqrt
+from monai.utils import ensure_tuple_rep, is_sqrt
 
 __all__ = ["ViTAutoEnc"]
 
@@ -33,9 +33,6 @@ class ViTAutoEnc(nn.Module):
     Modified to also give same dimension outputs as the input size of the image
     """
 
-    @deprecated_arg(
-        name="pos_embed", since="1.2", removed="1.4", new_name="proj_type", msg_suffix="please use `proj_type` instead."
-    )
     def __init__(
         self,
         in_channels: int,
@@ -47,7 +44,6 @@ def __init__(
         mlp_dim: int = 3072,
         num_layers: int = 12,
         num_heads: int = 12,
-        pos_embed: str = "conv",
         proj_type: str = "conv",
         dropout_rate: float = 0.0,
         spatial_dims: int = 3,
@@ -71,9 +67,6 @@ def __init__(
             qkv_bias: apply bias to the qkv linear layer in self attention block. Defaults to False.
             save_attn: to make accessible the attention in self attention block. Defaults to False. Defaults to False.
 
-        .. deprecated:: 1.4
-            ``pos_embed`` is deprecated in favor of ``proj_type``.
-
         Examples::
 
             # for single channel input with image size of (96,96,96), conv position embedding and segmentation backbone
diff --git a/monai/utils/enums.py b/monai/utils/enums.py
index eba1be18ed..7838a2e741 100644
--- a/monai/utils/enums.py
+++ b/monai/utils/enums.py
@@ -16,7 +16,6 @@
 from typing import TYPE_CHECKING
 
 from monai.config import IgniteInfo
-from monai.utils import deprecated
 from monai.utils.module import min_version, optional_import
 
 __all__ = [
@@ -56,7 +55,6 @@
     "DataStatsKeys",
     "ImageStatsKeys",
     "LabelStatsKeys",
-    "AlgoEnsembleKeys",
     "HoVerNetMode",
     "HoVerNetBranch",
     "LazyAttr",
@@ -615,17 +613,6 @@ class LabelStatsKeys(StrEnum):
     LABEL_NCOMP = "ncomponents"
 
 
-@deprecated(since="1.2", removed="1.4", msg_suffix="please use `AlgoKeys` instead.")
-class AlgoEnsembleKeys(StrEnum):
-    """
-    Default keys for Mixed Ensemble
-    """
-
-    ID = "identifier"
-    ALGO = "infer_algo"
-    SCORE = "best_metric"
-
-
 class HoVerNetMode(StrEnum):
     """
     Modes for HoVerNet model:
diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index 40370ca2c6..6386aae713 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -887,7 +887,7 @@ def run_cmd(cmd_list: list[str], **kwargs: Any) -> subprocess.CompletedProcess:
     if kwargs.pop("run_cmd_verbose", False):
         import monai
 
-        monai.apps.utils.get_logger("run_cmd").info(f"{cmd_list}")
+        monai.apps.utils.get_logger("run_cmd").info(f"{cmd_list}")  # type: ignore[attr-defined]
     try:
         return subprocess.run(cmd_list, **kwargs)
     except subprocess.CalledProcessError as e:

From 25589c377ac63d6be28ab9bb65dd8cb52f2bebdf Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 16 Sep 2024 14:43:52 +0800
Subject: [PATCH 154/183] move publish to test pypi (#8086)

Move publish to test pypi to blossom

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/release.yml | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
index a014a4ed1d..cb0e109bb7 100644
--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -78,12 +78,13 @@ jobs:
         rm dist/monai*.tar.gz
         ls -al dist/
 
-    - if: matrix.python-version == '3.9' && startsWith(github.ref, 'refs/tags/')
-      name: Publish to Test PyPI
-      uses: pypa/gh-action-pypi-publish@release/v1
-      with:
-        password: ${{ secrets.TEST_PYPI }}
-        repository-url: https://test.pypi.org/legacy/
+    # remove publishing to Test PyPI as it is moved to blossom
+    # - if: matrix.python-version == '3.9' && startsWith(github.ref, 'refs/tags/')
+    #   name: Publish to Test PyPI
+    #   uses: pypa/gh-action-pypi-publish@release/v1
+    #   with:
+    #     password: ${{ secrets.TEST_PYPI }}
+    #     repository-url: https://test.pypi.org/legacy/
 
   versioning:
     # compute versioning file from python setup.py

From 5bff4781ff436b61bda8ab40020dbd598e62c771 Mon Sep 17 00:00:00 2001
From: Christian Herz <christianherz1985@gmail.com>
Date: Wed, 18 Sep 2024 03:58:44 -0400
Subject: [PATCH 155/183] BUG: fixed wrong space directions interpretation in
 nrrd reader (#8091)

the vectors presented for space directions are column vectors

ref: https://teem.sourceforge.net/nrrd/format.html#spacedirections

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [ ] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Christian Herz <christianherz1985@gmail.com>
---
 monai/data/image_reader.py | 2 +-
 tests/test_nrrd_reader.py  | 8 ++++++--
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/monai/data/image_reader.py b/monai/data/image_reader.py
index f5e199e2a3..b4ae562911 100644
--- a/monai/data/image_reader.py
+++ b/monai/data/image_reader.py
@@ -1359,7 +1359,7 @@ def _get_affine(self, header: dict) -> np.ndarray:
         x, y = direction.shape
         affine_diam = min(x, y) + 1
         affine: np.ndarray = np.eye(affine_diam)
-        affine[:x, :y] = direction
+        affine[:x, :y] = direction.T
         affine[: (affine_diam - 1), -1] = origin  # len origin is always affine_diam - 1
         return affine
 
diff --git a/tests/test_nrrd_reader.py b/tests/test_nrrd_reader.py
index 649b9fa94d..5bf958e970 100644
--- a/tests/test_nrrd_reader.py
+++ b/tests/test_nrrd_reader.py
@@ -40,8 +40,8 @@
         "dimension": 4,
         "space": "left-posterior-superior",
         "sizes": [3, 4, 4, 1],
-        "space directions": [[0.0, 0.0, 0.0], [1.0, 0.0, 0.0], [0.0, 1.0, 0.0], [0.0, 0.0, 1.0]],
-        "space origin": [0.0, 0.0, 0.0],
+        "space directions": [[0.7, 0.0, 0.0], [0.0, 0.0, -0.8], [0.0, 0.9, 0.0]],
+        "space origin": [1.0, 5.0, 20.0],
     },
 ]
 
@@ -110,6 +110,10 @@ def test_read_with_header(self, data_shape, filename, expected_shape, dtype, ref
         np.testing.assert_allclose(image_array, test_image)
         self.assertIsInstance(image_header, dict)
         self.assertTupleEqual(tuple(image_header["spatial_shape"]), expected_shape)
+        np.testing.assert_allclose(
+            image_header["affine"],
+            np.array([[-0.7, 0.0, 0.0, -1.0], [0.0, 0.0, -0.9, -5.0], [0.0, -0.8, 0.0, 20.0], [0.0, 0.0, 0.0, 1.0]]),
+        )
 
     @parameterized.expand([TEST_CASE_8])
     def test_read_with_header_index_order_c(self, data_shape, filename, expected_shape, dtype, reference_header):

From d2d492ec045848b5872d8333e5174086c9cdfead Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 18 Sep 2024 16:52:37 +0800
Subject: [PATCH 156/183] Fix link in test bundle under MONAI-extra-test-data
 (#8092)

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 tests/test_bundle_download.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/test_bundle_download.py b/tests/test_bundle_download.py
index 331d228f1e..02a9f40846 100644
--- a/tests/test_bundle_download.py
+++ b/tests/test_bundle_download.py
@@ -89,7 +89,7 @@
 TEST_CASE_10 = [
     ["network.json", "test_output.pt", "test_input.pt", "large_files.yaml"],
     "test_bundle",
-    "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/test_bundle_v0.1.2.zip",
+    "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/test_bundle_v0.1.3.zip",
     {"model.pt": "27952767e2e154e3b0ee65defc5aed38", "model.ts": "97746870fe591f69ac09827175b00675"},
 ]
 

From fa1c1af79ef5387434f2a76744f75b5aaca09f0b Mon Sep 17 00:00:00 2001
From: Han123su <107395380+Han123su@users.noreply.github.com>
Date: Fri, 20 Sep 2024 17:16:29 +0800
Subject: [PATCH 157/183] Fix RandomWeightedCrop for Integer Weightmap Handling
 (#8097)

Fixes #7949 .

### Description
Regardless of the type of `weight map`, random numbers should be kept as
floating-point numbers for calculating the sampling location. However,
`searchsorted` requires matching data structures. I have modified
`convert_to_dst_type` to control converting only the data structure
while maintaining the original data type. Additionally, I have included
an example with integer weight maps in the test file.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Han123su <popsmall212@gmail.com>
Signed-off-by: Han123su <107395380+Han123su@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/transforms/utils.py        |  3 ++-
 tests/test_rand_weighted_crop.py | 30 ++++++++++++++++++++++++++++++
 2 files changed, 32 insertions(+), 1 deletion(-)

diff --git a/monai/transforms/utils.py b/monai/transforms/utils.py
index 32fffc25f0..e7e1616e13 100644
--- a/monai/transforms/utils.py
+++ b/monai/transforms/utils.py
@@ -582,7 +582,8 @@ def weighted_patch_samples(
     if not v[-1] or not isfinite(v[-1]) or v[-1] < 0:  # uniform sampling
         idx = r_state.randint(0, len(v), size=n_samples)
     else:
-        r, *_ = convert_to_dst_type(r_state.random(n_samples), v)
+        r_samples = r_state.random(n_samples)
+        r, *_ = convert_to_dst_type(r_samples, v, dtype=r_samples.dtype)
         idx = searchsorted(v, r * v[-1], right=True)  # type: ignore
     idx, *_ = convert_to_dst_type(idx, v, dtype=torch.int)  # type: ignore
     # compensate 'valid' mode
diff --git a/tests/test_rand_weighted_crop.py b/tests/test_rand_weighted_crop.py
index 47a8f3bfa2..f509065a56 100644
--- a/tests/test_rand_weighted_crop.py
+++ b/tests/test_rand_weighted_crop.py
@@ -90,6 +90,21 @@ def get_data(ndim):
                 [[63, 37], [31, 43], [66, 20]],
             ]
         )
+        im = SEG1_2D
+        weight_map = np.zeros_like(im, dtype=np.int32)
+        weight_map[0, 30, 20] = 3
+        weight_map[0, 45, 44] = 1
+        weight_map[0, 60, 50] = 2
+        TESTS.append(
+            [
+                "int w 2d",
+                dict(spatial_size=(10, 12), num_samples=3),
+                p(im),
+                q(weight_map),
+                (1, 10, 12),
+                [[60, 50], [30, 20], [45, 44]],
+            ]
+        )
         im = SEG1_3D
         weight = np.zeros_like(im)
         weight[0, 5, 30, 17] = 1.1
@@ -149,6 +164,21 @@ def get_data(ndim):
                 [[32, 24, 40], [32, 24, 40], [32, 24, 40]],
             ]
         )
+        im = SEG1_3D
+        weight_map = np.zeros_like(im, dtype=np.int32)
+        weight_map[0, 6, 22, 19] = 4
+        weight_map[0, 8, 40, 31] = 2
+        weight_map[0, 13, 20, 24] = 3
+        TESTS.append(
+            [
+                "int w 3d",
+                dict(spatial_size=(8, 10, 12), num_samples=3),
+                p(im),
+                q(weight_map),
+                (1, 8, 10, 12),
+                [[13, 20, 24], [6, 22, 19], [8, 40, 31]],
+            ]
+        )
 
 
 class TestRandWeightedCrop(CropTest):

From acfc508ebd35ad46bbccbd19f2eb57d34609f0e7 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 26 Sep 2024 12:38:36 +0800
Subject: [PATCH 158/183] Fix dtype bug in ScaleIntensityRangePercentile
 (#8109)

Fixes #8108


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/transforms/intensity/array.py             | 2 +-
 tests/test_scale_intensity_range_percentiles.py | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/monai/transforms/intensity/array.py b/monai/transforms/intensity/array.py
index 3b813809e4..20000c52c4 100644
--- a/monai/transforms/intensity/array.py
+++ b/monai/transforms/intensity/array.py
@@ -1411,7 +1411,7 @@ def __call__(self, img: NdarrayOrTensor) -> NdarrayOrTensor:
         else:
             img_t = self._normalize(img=img_t)
 
-        return convert_to_dst_type(img_t, dst=img)[0]
+        return convert_to_dst_type(img_t, dst=img, dtype=self.dtype)[0]
 
 
 class MaskIntensity(Transform):
diff --git a/tests/test_scale_intensity_range_percentiles.py b/tests/test_scale_intensity_range_percentiles.py
index 7c3a684a00..a7390efe72 100644
--- a/tests/test_scale_intensity_range_percentiles.py
+++ b/tests/test_scale_intensity_range_percentiles.py
@@ -14,6 +14,7 @@
 import unittest
 
 import numpy as np
+import torch
 
 from monai.transforms.intensity.array import ScaleIntensityRangePercentiles
 from tests.utils import TEST_NDARRAYS, NumpyImageTestCase2D, assert_allclose
@@ -34,6 +35,7 @@ def test_scaling(self):
         scaler = ScaleIntensityRangePercentiles(lower=lower, upper=upper, b_min=b_min, b_max=b_max, dtype=np.uint8)
         for p in TEST_NDARRAYS:
             result = scaler(p(img))
+            self.assertEqual(result.dtype, torch.uint8)
             assert_allclose(result, p(expected), type_test="tensor", rtol=1e-4)
 
     def test_relative_scaling(self):

From cac21f6936a2e8d6e4e57e4e958f8e32aae1585e Mon Sep 17 00:00:00 2001
From: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Date: Fri, 27 Sep 2024 04:23:47 +0100
Subject: [PATCH 159/183] Cleaning up some very old and now obsolete
 infrastructure (#8113)

### Description

This removes some functions which aren't needed any more. These related
to an early idea for doing importation to avoid defining `__all__`
components to modules, but this is being done instead as it is more
standard.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Eric Kerfoot <eric.kerfoot@kcl.ac.uk>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 docs/source/networks.rst                |   5 --
 docs/source/utils.rst                   |   6 --
 monai/__init__.py                       |  39 +++++----
 monai/bundle/scripts.py                 |   3 +-
 monai/config/__init__.py                |   1 -
 monai/config/deviceconfig.py            |  10 ---
 monai/engines/evaluator.py              |   4 +-
 monai/engines/trainer.py                |   3 +-
 monai/engines/utils.py                  |   3 +-
 monai/engines/workflow.py               |   3 +-
 monai/handlers/checkpoint_loader.py     |   3 +-
 monai/handlers/checkpoint_saver.py      |   3 +-
 monai/handlers/classification_saver.py  |   2 +-
 monai/handlers/decollate_batch.py       |   4 +-
 monai/handlers/earlystop_handler.py     |   3 +-
 monai/handlers/garbage_collector.py     |   3 +-
 monai/handlers/ignite_metric.py         |   3 +-
 monai/handlers/logfile_handler.py       |   3 +-
 monai/handlers/lr_schedule_handler.py   |   3 +-
 monai/handlers/metric_logger.py         |   3 +-
 monai/handlers/metrics_saver.py         |   2 +-
 monai/handlers/mlflow_handler.py        |   3 +-
 monai/handlers/nvtx_handlers.py         |   3 +-
 monai/handlers/parameter_scheduler.py   |   3 +-
 monai/handlers/postprocessing.py        |   3 +-
 monai/handlers/probability_maps.py      |   4 +-
 monai/handlers/smartcache_handler.py    |   3 +-
 monai/handlers/stats_handler.py         |   3 +-
 monai/handlers/tensorboard_handlers.py  |   3 +-
 monai/handlers/trt_handler.py           |   3 +-
 monai/handlers/utils.py                 |   4 +-
 monai/handlers/validation_handler.py    |   3 +-
 monai/networks/nets/hovernet.py         |   3 +-
 monai/networks/nets/unet.py             |   3 -
 monai/networks/nets/voxelmorph.py       |   5 --
 monai/transforms/adaptors.py            |   5 --
 monai/utils/__init__.py                 |   7 +-
 monai/utils/aliases.py                  | 103 ------------------------
 monai/utils/dist.py                     |   2 +-
 monai/utils/enums.py                    |  51 +++++++-----
 monai/utils/jupyter_utils.py            |   2 +-
 monai/utils/module.py                   |  60 +-------------
 tests/min_tests.py                      |   6 --
 tests/test_fastmri_reader.py            |   3 +-
 tests/test_handler_garbage_collector.py |   3 +-
 45 files changed, 90 insertions(+), 307 deletions(-)
 delete mode 100644 monai/utils/aliases.py

diff --git a/docs/source/networks.rst b/docs/source/networks.rst
index 1810fec49b..64a3a4c9d1 100644
--- a/docs/source/networks.rst
+++ b/docs/source/networks.rst
@@ -735,14 +735,9 @@ Nets
 .. autoclass:: VoxelMorphUNet
    :members:
 
-.. autoclass:: voxelmorphunet
-   :members:
-
 .. autoclass:: VoxelMorph
    :members:
 
-.. autoclass:: voxelmorph
-
 Utilities
 ---------
 .. automodule:: monai.networks.utils
diff --git a/docs/source/utils.rst b/docs/source/utils.rst
index fef671e1f8..ae3b476c3e 100644
--- a/docs/source/utils.rst
+++ b/docs/source/utils.rst
@@ -17,12 +17,6 @@ Module utils
   :members:
 
 
-Aliases
--------
-.. automodule:: monai.utils.aliases
-  :members:
-
-
 Misc
 ----
 .. automodule:: monai.utils.misc
diff --git a/monai/__init__.py b/monai/__init__.py
index 3f6b12c407..46f7510915 100644
--- a/monai/__init__.py
+++ b/monai/__init__.py
@@ -79,28 +79,25 @@ def filter(self, record):
         category=RuntimeWarning,
     )
 
-from .utils.module import load_submodules  # noqa: E402
-
-# handlers_* have some external decorators the users may not have installed
-# *.so files and folder "_C" may not exist when the cpp extensions are not compiled
-excludes = "|".join(
-    [
-        "(^(monai.handlers))",
-        "(^(monai.bundle))",
-        "(^(monai.fl))",
-        "((\\.so)$)",
-        "(^(monai._C))",
-        "(.*(__main__)$)",
-        "(.*(video_dataset)$)",
-        "(.*(nnunet).*$)",
-    ]
-)
-
-# load directory modules only, skip loading individual files
-load_submodules(sys.modules[__name__], False, exclude_pattern=excludes)
 
-# load all modules, this will trigger all export decorations
-load_submodules(sys.modules[__name__], True, exclude_pattern=excludes)
+from . import (  # noqa: E402
+    apps,
+    auto3dseg,
+    bundle,
+    config,
+    data,
+    engines,
+    fl,
+    handlers,
+    inferers,
+    losses,
+    metrics,
+    networks,
+    optimizers,
+    transforms,
+    utils,
+    visualize,
+)
 
 __all__ = [
     "apps",
diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index f31ad0e814..4251da0b6f 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -34,7 +34,7 @@
 from monai.bundle.config_parser import ConfigParser
 from monai.bundle.utils import DEFAULT_INFERENCE, DEFAULT_METADATA, merge_kv
 from monai.bundle.workflows import BundleWorkflow, ConfigWorkflow
-from monai.config import IgniteInfo, PathLike
+from monai.config import PathLike
 from monai.data import load_net_with_metadata, save_net_with_metadata
 from monai.networks import (
     convert_to_onnx,
@@ -45,6 +45,7 @@
     save_state,
 )
 from monai.utils import (
+    IgniteInfo,
     check_parent_dir,
     deprecated_arg,
     ensure_tuple,
diff --git a/monai/config/__init__.py b/monai/config/__init__.py
index c814e1f8eb..a83889aee0 100644
--- a/monai/config/__init__.py
+++ b/monai/config/__init__.py
@@ -14,7 +14,6 @@
 from .deviceconfig import (
     USE_COMPILED,
     USE_META_DICT,
-    IgniteInfo,
     get_config_values,
     get_gpu_info,
     get_optional_config_values,
diff --git a/monai/config/deviceconfig.py b/monai/config/deviceconfig.py
index a4580c741b..7ac1b29919 100644
--- a/monai/config/deviceconfig.py
+++ b/monai/config/deviceconfig.py
@@ -45,7 +45,6 @@
     "print_debug_info",
     "USE_COMPILED",
     "USE_META_DICT",
-    "IgniteInfo",
 ]
 
 
@@ -261,14 +260,5 @@ def print_debug_info(file: TextIO = sys.stdout) -> None:
     print_gpu_info(file)
 
 
-class IgniteInfo:
-    """
-    Config information of the PyTorch ignite package.
-
-    """
-
-    OPT_IMPORT_VERSION = "0.4.4"
-
-
 if __name__ == "__main__":
     print_debug_info()
diff --git a/monai/engines/evaluator.py b/monai/engines/evaluator.py
index 2c8dfe6b85..523c3dcbf6 100644
--- a/monai/engines/evaluator.py
+++ b/monai/engines/evaluator.py
@@ -17,14 +17,14 @@
 import torch
 from torch.utils.data import DataLoader
 
-from monai.config import IgniteInfo, KeysCollection
+from monai.config import KeysCollection
 from monai.data import MetaTensor
 from monai.engines.utils import IterationEvents, default_metric_cmp_fn, default_prepare_batch
 from monai.engines.workflow import Workflow
 from monai.inferers import Inferer, SimpleInferer
 from monai.networks.utils import eval_mode, train_mode
 from monai.transforms import Transform
-from monai.utils import ForwardMode, ensure_tuple, min_version, optional_import
+from monai.utils import ForwardMode, IgniteInfo, ensure_tuple, min_version, optional_import
 from monai.utils.enums import CommonKeys as Keys
 from monai.utils.enums import EngineStatsKeys as ESKeys
 from monai.utils.module import look_up_option, pytorch_after
diff --git a/monai/engines/trainer.py b/monai/engines/trainer.py
index c1364fe015..bbcc9c880b 100644
--- a/monai/engines/trainer.py
+++ b/monai/engines/trainer.py
@@ -18,13 +18,12 @@
 from torch.optim.optimizer import Optimizer
 from torch.utils.data import DataLoader
 
-from monai.config import IgniteInfo
 from monai.data import MetaTensor
 from monai.engines.utils import IterationEvents, default_make_latent, default_metric_cmp_fn, default_prepare_batch
 from monai.engines.workflow import Workflow
 from monai.inferers import Inferer, SimpleInferer
 from monai.transforms import Transform
-from monai.utils import AdversarialIterationEvents, AdversarialKeys, GanKeys, min_version, optional_import
+from monai.utils import AdversarialIterationEvents, AdversarialKeys, GanKeys, IgniteInfo, min_version, optional_import
 from monai.utils.enums import CommonKeys as Keys
 from monai.utils.enums import EngineStatsKeys as ESKeys
 from monai.utils.module import pytorch_after
diff --git a/monai/engines/utils.py b/monai/engines/utils.py
index 5339d6965a..11a0000989 100644
--- a/monai/engines/utils.py
+++ b/monai/engines/utils.py
@@ -18,9 +18,8 @@
 import torch
 import torch.nn as nn
 
-from monai.config import IgniteInfo
 from monai.transforms import apply_transform
-from monai.utils import ensure_tuple, min_version, optional_import
+from monai.utils import IgniteInfo, ensure_tuple, min_version, optional_import
 from monai.utils.enums import CommonKeys, GanKeys
 
 if TYPE_CHECKING:
diff --git a/monai/engines/workflow.py b/monai/engines/workflow.py
index 30622c2b93..3629659db1 100644
--- a/monai/engines/workflow.py
+++ b/monai/engines/workflow.py
@@ -20,10 +20,9 @@
 from torch.utils.data import DataLoader
 from torch.utils.data.distributed import DistributedSampler
 
-from monai.config import IgniteInfo
 from monai.engines.utils import IterationEvents, default_metric_cmp_fn, default_prepare_batch
 from monai.transforms import Decollated
-from monai.utils import ensure_tuple, is_scalar, min_version, optional_import
+from monai.utils import IgniteInfo, ensure_tuple, is_scalar, min_version, optional_import
 
 from .utils import engine_apply_transform
 
diff --git a/monai/handlers/checkpoint_loader.py b/monai/handlers/checkpoint_loader.py
index 9a867534a3..f48968ecfd 100644
--- a/monai/handlers/checkpoint_loader.py
+++ b/monai/handlers/checkpoint_loader.py
@@ -17,9 +17,8 @@
 
 import torch
 
-from monai.config import IgniteInfo
 from monai.networks.utils import copy_model_state
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 Checkpoint, _ = optional_import("ignite.handlers", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Checkpoint")
diff --git a/monai/handlers/checkpoint_saver.py b/monai/handlers/checkpoint_saver.py
index 0651c6ff33..2a3a467570 100644
--- a/monai/handlers/checkpoint_saver.py
+++ b/monai/handlers/checkpoint_saver.py
@@ -17,8 +17,7 @@
 from collections.abc import Mapping
 from typing import TYPE_CHECKING, Any
 
-from monai.config import IgniteInfo
-from monai.utils import is_scalar, min_version, optional_import
+from monai.utils import IgniteInfo, is_scalar, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 
diff --git a/monai/handlers/classification_saver.py b/monai/handlers/classification_saver.py
index 831808f4fb..ffcfe3c1fb 100644
--- a/monai/handlers/classification_saver.py
+++ b/monai/handlers/classification_saver.py
@@ -18,8 +18,8 @@
 
 import torch
 
-from monai.config import IgniteInfo
 from monai.data import CSVSaver, decollate_batch
+from monai.utils import IgniteInfo
 from monai.utils import ImageMetaKey as Key
 from monai.utils import evenly_divisible_all_gather, min_version, optional_import, string_list_all_gather
 
diff --git a/monai/handlers/decollate_batch.py b/monai/handlers/decollate_batch.py
index ac3aa94145..81415bd56e 100644
--- a/monai/handlers/decollate_batch.py
+++ b/monai/handlers/decollate_batch.py
@@ -13,10 +13,10 @@
 
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo, KeysCollection
+from monai.config import KeysCollection
 from monai.engines.utils import IterationEvents
 from monai.transforms import Decollated
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
diff --git a/monai/handlers/earlystop_handler.py b/monai/handlers/earlystop_handler.py
index 93334bf5c0..0562335192 100644
--- a/monai/handlers/earlystop_handler.py
+++ b/monai/handlers/earlystop_handler.py
@@ -14,8 +14,7 @@
 from collections.abc import Callable
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 EarlyStopping, _ = optional_import("ignite.handlers", IgniteInfo.OPT_IMPORT_VERSION, min_version, "EarlyStopping")
diff --git a/monai/handlers/garbage_collector.py b/monai/handlers/garbage_collector.py
index 3d7e948364..586fa10d33 100644
--- a/monai/handlers/garbage_collector.py
+++ b/monai/handlers/garbage_collector.py
@@ -14,8 +14,7 @@
 import gc
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 if TYPE_CHECKING:
     from ignite.engine import Engine, Events
diff --git a/monai/handlers/ignite_metric.py b/monai/handlers/ignite_metric.py
index 64ded4d5ea..44a5634c42 100644
--- a/monai/handlers/ignite_metric.py
+++ b/monai/handlers/ignite_metric.py
@@ -18,9 +18,8 @@
 import torch
 from torch.nn.modules.loss import _Loss
 
-from monai.config import IgniteInfo
 from monai.metrics import CumulativeIterationMetric, LossMetric
-from monai.utils import MetricReduction, min_version, optional_import
+from monai.utils import IgniteInfo, MetricReduction, min_version, optional_import
 
 idist, _ = optional_import("ignite", IgniteInfo.OPT_IMPORT_VERSION, min_version, "distributed")
 
diff --git a/monai/handlers/logfile_handler.py b/monai/handlers/logfile_handler.py
index df6ebd34a7..0c44ae47f4 100644
--- a/monai/handlers/logfile_handler.py
+++ b/monai/handlers/logfile_handler.py
@@ -15,8 +15,7 @@
 import os
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
diff --git a/monai/handlers/lr_schedule_handler.py b/monai/handlers/lr_schedule_handler.py
index a79722517d..8d90992a84 100644
--- a/monai/handlers/lr_schedule_handler.py
+++ b/monai/handlers/lr_schedule_handler.py
@@ -17,8 +17,7 @@
 
 from torch.optim.lr_scheduler import ReduceLROnPlateau, _LRScheduler
 
-from monai.config import IgniteInfo
-from monai.utils import ensure_tuple, min_version, optional_import
+from monai.utils import IgniteInfo, ensure_tuple, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
diff --git a/monai/handlers/metric_logger.py b/monai/handlers/metric_logger.py
index d59205a021..62cdee6509 100644
--- a/monai/handlers/metric_logger.py
+++ b/monai/handlers/metric_logger.py
@@ -17,8 +17,7 @@
 from threading import RLock
 from typing import TYPE_CHECKING, Any
 
-from monai.config import IgniteInfo
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 from monai.utils.enums import CommonKeys
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
diff --git a/monai/handlers/metrics_saver.py b/monai/handlers/metrics_saver.py
index 88a0926b91..6175b1242a 100644
--- a/monai/handlers/metrics_saver.py
+++ b/monai/handlers/metrics_saver.py
@@ -14,9 +14,9 @@
 from collections.abc import Callable, Sequence
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
 from monai.data import decollate_batch
 from monai.handlers.utils import write_metrics_reports
+from monai.utils import IgniteInfo
 from monai.utils import ImageMetaKey as Key
 from monai.utils import ensure_tuple, min_version, optional_import, string_list_all_gather
 
diff --git a/monai/handlers/mlflow_handler.py b/monai/handlers/mlflow_handler.py
index 6d19579d9e..c7e293ea7d 100644
--- a/monai/handlers/mlflow_handler.py
+++ b/monai/handlers/mlflow_handler.py
@@ -22,8 +22,7 @@
 from torch.utils.data import Dataset
 
 from monai.apps.utils import get_logger
-from monai.config import IgniteInfo
-from monai.utils import CommonKeys, ensure_tuple, min_version, optional_import
+from monai.utils import CommonKeys, IgniteInfo, ensure_tuple, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 mlflow, _ = optional_import("mlflow", descriptor="Please install mlflow before using MLFlowHandler.")
diff --git a/monai/handlers/nvtx_handlers.py b/monai/handlers/nvtx_handlers.py
index 38eef6f05b..bd22af0db8 100644
--- a/monai/handlers/nvtx_handlers.py
+++ b/monai/handlers/nvtx_handlers.py
@@ -16,8 +16,7 @@
 
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
-from monai.utils import ensure_tuple, min_version, optional_import
+from monai.utils import IgniteInfo, ensure_tuple, min_version, optional_import
 
 _nvtx, _ = optional_import("torch._C._nvtx", descriptor="NVTX is not installed. Are you sure you have a CUDA build?")
 if TYPE_CHECKING:
diff --git a/monai/handlers/parameter_scheduler.py b/monai/handlers/parameter_scheduler.py
index d12e6e072c..1ce6193b6d 100644
--- a/monai/handlers/parameter_scheduler.py
+++ b/monai/handlers/parameter_scheduler.py
@@ -16,8 +16,7 @@
 from collections.abc import Callable
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 if TYPE_CHECKING:
     from ignite.engine import Engine, Events
diff --git a/monai/handlers/postprocessing.py b/monai/handlers/postprocessing.py
index c698c84338..541b5924d1 100644
--- a/monai/handlers/postprocessing.py
+++ b/monai/handlers/postprocessing.py
@@ -14,9 +14,8 @@
 from collections.abc import Callable
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
 from monai.engines.utils import IterationEvents, engine_apply_transform
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
diff --git a/monai/handlers/probability_maps.py b/monai/handlers/probability_maps.py
index 8a60fcc983..e21bd199f8 100644
--- a/monai/handlers/probability_maps.py
+++ b/monai/handlers/probability_maps.py
@@ -17,10 +17,10 @@
 
 import numpy as np
 
-from monai.config import DtypeLike, IgniteInfo
+from monai.config import DtypeLike
 from monai.data.folder_layout import FolderLayout
 from monai.utils import ProbMapKeys, min_version, optional_import
-from monai.utils.enums import CommonKeys
+from monai.utils.enums import CommonKeys, IgniteInfo
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
diff --git a/monai/handlers/smartcache_handler.py b/monai/handlers/smartcache_handler.py
index ee043635db..e07e98e541 100644
--- a/monai/handlers/smartcache_handler.py
+++ b/monai/handlers/smartcache_handler.py
@@ -13,9 +13,8 @@
 
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
 from monai.data import SmartCacheDataset
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
diff --git a/monai/handlers/stats_handler.py b/monai/handlers/stats_handler.py
index c49fcda819..ab36d19bd1 100644
--- a/monai/handlers/stats_handler.py
+++ b/monai/handlers/stats_handler.py
@@ -19,8 +19,7 @@
 import torch
 
 from monai.apps import get_logger
-from monai.config import IgniteInfo
-from monai.utils import is_scalar, min_version, optional_import
+from monai.utils import IgniteInfo, is_scalar, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
diff --git a/monai/handlers/tensorboard_handlers.py b/monai/handlers/tensorboard_handlers.py
index 7b7e3968fb..44a03710de 100644
--- a/monai/handlers/tensorboard_handlers.py
+++ b/monai/handlers/tensorboard_handlers.py
@@ -18,8 +18,7 @@
 import numpy as np
 import torch
 
-from monai.config import IgniteInfo
-from monai.utils import is_scalar, min_version, optional_import
+from monai.utils import IgniteInfo, is_scalar, min_version, optional_import
 from monai.visualize import plot_2d_or_3d_image
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
diff --git a/monai/handlers/trt_handler.py b/monai/handlers/trt_handler.py
index 0e36b59d8c..45e2669f70 100644
--- a/monai/handlers/trt_handler.py
+++ b/monai/handlers/trt_handler.py
@@ -13,9 +13,8 @@
 
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
 from monai.networks import trt_compile
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
diff --git a/monai/handlers/utils.py b/monai/handlers/utils.py
index 0cd31b89c2..b6771f2dcc 100644
--- a/monai/handlers/utils.py
+++ b/monai/handlers/utils.py
@@ -19,8 +19,8 @@
 import numpy as np
 import torch
 
-from monai.config import IgniteInfo, KeysCollection, PathLike
-from monai.utils import ensure_tuple, look_up_option, min_version, optional_import
+from monai.config import KeysCollection, PathLike
+from monai.utils import IgniteInfo, ensure_tuple, look_up_option, min_version, optional_import
 
 idist, _ = optional_import("ignite", IgniteInfo.OPT_IMPORT_VERSION, min_version, "distributed")
 if TYPE_CHECKING:
diff --git a/monai/handlers/validation_handler.py b/monai/handlers/validation_handler.py
index 89c7715f42..38dd511aa4 100644
--- a/monai/handlers/validation_handler.py
+++ b/monai/handlers/validation_handler.py
@@ -13,9 +13,8 @@
 
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
 from monai.engines.evaluator import Evaluator
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
diff --git a/monai/networks/nets/hovernet.py b/monai/networks/nets/hovernet.py
index 5f340c9be6..3745b66bb5 100644
--- a/monai/networks/nets/hovernet.py
+++ b/monai/networks/nets/hovernet.py
@@ -43,7 +43,7 @@
 from monai.networks.layers.factories import Conv, Dropout
 from monai.networks.layers.utils import get_act_layer, get_norm_layer
 from monai.utils.enums import HoVerNetBranch, HoVerNetMode, InterpolateMode, UpsampleMode
-from monai.utils.module import export, look_up_option
+from monai.utils.module import look_up_option
 
 __all__ = ["HoVerNet", "Hovernet", "HoVernet", "HoVerNet"]
 
@@ -409,7 +409,6 @@ def forward(self, xin: torch.Tensor, short_cuts: list[torch.Tensor]) -> torch.Te
         return x
 
 
-@export("monai.networks.nets")
 class HoVerNet(nn.Module):
     """HoVerNet model
 
diff --git a/monai/networks/nets/unet.py b/monai/networks/nets/unet.py
index 7b16b6c923..eac0ddab39 100644
--- a/monai/networks/nets/unet.py
+++ b/monai/networks/nets/unet.py
@@ -20,13 +20,10 @@
 from monai.networks.blocks.convolutions import Convolution, ResidualUnit
 from monai.networks.layers.factories import Act, Norm
 from monai.networks.layers.simplelayers import SkipConnection
-from monai.utils import alias, export
 
 __all__ = ["UNet", "Unet"]
 
 
-@export("monai.networks.nets")
-@alias("Unet")
 class UNet(nn.Module):
     """
     Enhanced version of UNet which has residual units implemented with the ResidualUnit class.
diff --git a/monai/networks/nets/voxelmorph.py b/monai/networks/nets/voxelmorph.py
index 0496cfc8f8..4923b6ad60 100644
--- a/monai/networks/nets/voxelmorph.py
+++ b/monai/networks/nets/voxelmorph.py
@@ -21,13 +21,10 @@
 from monai.networks.blocks.upsample import UpSample
 from monai.networks.blocks.warp import DVF2DDF, Warp
 from monai.networks.layers.simplelayers import SkipConnection
-from monai.utils import alias, export
 
 __all__ = ["VoxelMorphUNet", "voxelmorphunet", "VoxelMorph", "voxelmorph"]
 
 
-@export("monai.networks.nets")
-@alias("voxelmorphunet")
 class VoxelMorphUNet(nn.Module):
     """
     The backbone network used in VoxelMorph. See :py:class:`monai.networks.nets.VoxelMorph` for more details.
@@ -340,8 +337,6 @@ def forward(self, concatenated_pairs: torch.Tensor) -> torch.Tensor:
 voxelmorphunet = VoxelMorphUNet
 
 
-@export("monai.networks.nets")
-@alias("voxelmorph")
 class VoxelMorph(nn.Module):
     """
     A re-implementation of VoxelMorph framework for medical image registration as described in
diff --git a/monai/transforms/adaptors.py b/monai/transforms/adaptors.py
index f5f1a4fc18..5a0c24c7f6 100644
--- a/monai/transforms/adaptors.py
+++ b/monai/transforms/adaptors.py
@@ -125,12 +125,9 @@ def __call__(self, img, seg):
 
 from typing import Callable
 
-from monai.utils import export as _monai_export
-
 __all__ = ["adaptor", "apply_alias", "to_kwargs", "FunctionSignature"]
 
 
-@_monai_export("monai.transforms")
 def adaptor(function, outputs, inputs=None):
 
     def must_be_types_or_none(variable_name, variable, types):
@@ -215,7 +212,6 @@ def _inner(ditems):
     return _inner
 
 
-@_monai_export("monai.transforms")
 def apply_alias(fn, name_map):
 
     def _inner(data):
@@ -236,7 +232,6 @@ def _inner(data):
     return _inner
 
 
-@_monai_export("monai.transforms")
 def to_kwargs(fn):
 
     def _inner(data):
diff --git a/monai/utils/__init__.py b/monai/utils/__init__.py
index 4e36e3cd47..b5b2d7a525 100644
--- a/monai/utils/__init__.py
+++ b/monai/utils/__init__.py
@@ -11,8 +11,6 @@
 
 from __future__ import annotations
 
-# have to explicitly bring these in here to resolve circular import issues
-from .aliases import alias, resolve_name
 from .component_store import ComponentStore
 from .decorators import MethodReplacer, RestartGenerator
 from .deprecate_utils import DeprecatedError, deprecated, deprecated_arg, deprecated_arg_default
@@ -40,6 +38,7 @@
     GridSamplePadMode,
     HoVerNetBranch,
     HoVerNetMode,
+    IgniteInfo,
     InterpolateMode,
     JITMetadataKeys,
     LazyAttr,
@@ -109,12 +108,10 @@
     allow_missing_reference,
     damerau_levenshtein_distance,
     exact_version,
-    export,
     get_full_type_name,
     get_package_version,
     get_torch_version_tuple,
     instantiate,
-    load_submodules,
     look_up_option,
     min_version,
     optional_import,
@@ -153,3 +150,5 @@
     get_numpy_dtype_from_string,
     get_torch_dtype_from_string,
 )
+
+# have to explicitly bring these in here to resolve circular import issues
diff --git a/monai/utils/aliases.py b/monai/utils/aliases.py
deleted file mode 100644
index 2974eec2eb..0000000000
--- a/monai/utils/aliases.py
+++ /dev/null
@@ -1,103 +0,0 @@
-# Copyright (c) MONAI Consortium
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#     http://www.apache.org/licenses/LICENSE-2.0
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-"""
-This module is written for configurable workflow, not currently in use.
-"""
-
-from __future__ import annotations
-
-import importlib
-import inspect
-import sys
-import threading
-
-alias_lock = threading.RLock()
-GlobalAliases = {}
-
-__all__ = ["alias", "resolve_name"]
-
-
-def alias(*names):
-    """
-    Stores the decorated function or class in the global aliases table under the given names and as the `__aliases__`
-    member of the decorated object. This new member will contain all alias names declared for that object.
-    """
-
-    def _outer(obj):
-        for n in names:
-            with alias_lock:
-                GlobalAliases[n] = obj
-
-        # set the member list __aliases__ to contain the alias names defined by the decorator for `obj`
-        obj.__aliases__ = getattr(obj, "__aliases__", ()) + tuple(names)
-
-        return obj
-
-    return _outer
-
-
-def resolve_name(name):
-    """
-    Search for the declaration (function or class) with the given name. This will first search the list of aliases to
-    see if it was declared with this aliased name, then search treating `name` as a fully qualified name, then search
-    the loaded modules for one having a declaration with the given name. If no declaration is found, raise ValueError.
-
-    Raises:
-        ValueError: When the module is not found.
-        ValueError: When the module does not have the specified member.
-        ValueError: When multiple modules with the declaration name are found.
-        ValueError: When no module with the specified member is found.
-
-    """
-    # attempt to resolve an alias
-    with alias_lock:
-        obj = GlobalAliases.get(name)
-
-    if name in GlobalAliases and obj is None:
-        raise AssertionError
-
-    # attempt to resolve a qualified name
-    if obj is None and "." in name:
-        modname, declname = name.rsplit(".", 1)
-
-        try:
-            mod = importlib.import_module(modname)
-            obj = getattr(mod, declname, None)
-        except ModuleNotFoundError as not_found_err:
-            raise ValueError(f"Module {modname!r} not found.") from not_found_err
-
-        if obj is None:
-            raise ValueError(f"Module {modname!r} does not have member {declname!r}.")
-
-    # attempt to resolve a simple name
-    if obj is None:
-        # Get all modules having the declaration/import, need to check here that getattr returns something which doesn't
-        # equate to False since in places __getattr__ returns 0 incorrectly:
-        # https://github.com/tensorflow/tensorboard/blob/a22566561d2b4fea408755a951ac9eaf3a156f8e/
-        # tensorboard/compat/tensorflow_stub/pywrap_tensorflow.py#L35
-        mods = [m for m in list(sys.modules.values()) if getattr(m, name, None)]
-
-        if len(mods) > 0:  # found modules with this declaration or import
-            if len(mods) > 1:  # found multiple modules, need to determine if ambiguous or just multiple imports
-                foundmods = set(filter(None, {inspect.getmodule(getattr(m, name)) for m in mods}))  # resolve imports
-
-                if len(foundmods) > 1:  # found multiple declarations with the same name
-                    modnames = [m.__name__ for m in foundmods]
-                    msg = f"Multiple modules ({modnames!r}) with declaration name {name!r} found, resolution is ambiguous."
-                    raise ValueError(msg)
-                mods = list(foundmods)
-
-            obj = getattr(mods[0], name)
-
-        if obj is None:
-            raise ValueError(f"No module with member {name!r} found.")
-
-    return obj
diff --git a/monai/utils/dist.py b/monai/utils/dist.py
index 2418b43591..c7ff988027 100644
--- a/monai/utils/dist.py
+++ b/monai/utils/dist.py
@@ -24,7 +24,7 @@
 import torch
 import torch.distributed as dist
 
-from monai.config import IgniteInfo
+from monai.utils.enums import IgniteInfo
 from monai.utils.module import min_version, optional_import
 
 idist, has_ignite = optional_import("ignite", IgniteInfo.OPT_IMPORT_VERSION, min_version, "distributed")
diff --git a/monai/utils/enums.py b/monai/utils/enums.py
index 7838a2e741..900133f2f2 100644
--- a/monai/utils/enums.py
+++ b/monai/utils/enums.py
@@ -15,7 +15,6 @@
 from enum import Enum
 from typing import TYPE_CHECKING
 
-from monai.config import IgniteInfo
 from monai.utils.module import min_version, optional_import
 
 __all__ = [
@@ -61,6 +60,7 @@
     "BundleProperty",
     "BundlePropertyConfig",
     "AlgoKeys",
+    "IgniteInfo",
 ]
 
 
@@ -89,14 +89,6 @@ def __repr__(self):
         return self.value
 
 
-if TYPE_CHECKING:
-    from ignite.engine import EventEnum
-else:
-    EventEnum, _ = optional_import(
-        "ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "EventEnum", as_type="base"
-    )
-
-
 class NumpyPadMode(StrEnum):
     """
     See also: https://numpy.org/doc/1.18/reference/generated/numpy.pad.html
@@ -717,6 +709,35 @@ class AdversarialKeys(StrEnum):
     DISCRIMINATOR_LOSS = "discriminator_loss"
 
 
+class OrderingType(StrEnum):
+    RASTER_SCAN = "raster_scan"
+    S_CURVE = "s_curve"
+    RANDOM = "random"
+
+
+class OrderingTransformations(StrEnum):
+    ROTATE_90 = "rotate_90"
+    TRANSPOSE = "transpose"
+    REFLECT = "reflect"
+
+
+class IgniteInfo(StrEnum):
+    """
+    Config information of the PyTorch ignite package.
+
+    """
+
+    OPT_IMPORT_VERSION = "0.4.4"
+
+
+if TYPE_CHECKING:
+    from ignite.engine import EventEnum
+else:
+    EventEnum, _ = optional_import(
+        "ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "EventEnum", as_type="base"
+    )
+
+
 class AdversarialIterationEvents(EventEnum):
     """
     Keys used to define events as used in the AdversarialTrainer.
@@ -733,15 +754,3 @@ class AdversarialIterationEvents(EventEnum):
     DISCRIMINATOR_LOSS_COMPLETED = "discriminator_loss_completed"
     DISCRIMINATOR_BACKWARD_COMPLETED = "discriminator_backward_completed"
     DISCRIMINATOR_MODEL_COMPLETED = "discriminator_model_completed"
-
-
-class OrderingType(StrEnum):
-    RASTER_SCAN = "raster_scan"
-    S_CURVE = "s_curve"
-    RANDOM = "random"
-
-
-class OrderingTransformations(StrEnum):
-    ROTATE_90 = "rotate_90"
-    TRANSPOSE = "transpose"
-    REFLECT = "reflect"
diff --git a/monai/utils/jupyter_utils.py b/monai/utils/jupyter_utils.py
index 7dcd0e62cd..b1b43a6767 100644
--- a/monai/utils/jupyter_utils.py
+++ b/monai/utils/jupyter_utils.py
@@ -24,7 +24,7 @@
 import numpy as np
 import torch
 
-from monai.config import IgniteInfo
+from monai.utils import IgniteInfo
 from monai.utils.module import min_version, optional_import
 
 try:
diff --git a/monai/utils/module.py b/monai/utils/module.py
index 78087aef84..3f07dd9e91 100644
--- a/monai/utils/module.py
+++ b/monai/utils/module.py
@@ -16,15 +16,13 @@
 import os
 import pdb
 import re
-import sys
 import warnings
 from collections.abc import Callable, Collection, Hashable, Mapping
 from functools import partial, wraps
 from importlib import import_module
-from pkgutil import walk_packages
 from pydoc import locate
 from re import match
-from types import FunctionType, ModuleType
+from types import FunctionType
 from typing import Any, Iterable, cast
 
 import torch
@@ -43,13 +41,11 @@
     "InvalidPyTorchVersionError",
     "OptionalImportError",
     "exact_version",
-    "export",
     "damerau_levenshtein_distance",
     "look_up_option",
     "min_version",
     "optional_import",
     "require_pkg",
-    "load_submodules",
     "instantiate",
     "get_full_type_name",
     "get_package_version",
@@ -172,60 +168,6 @@ def damerau_levenshtein_distance(s1: str, s2: str) -> int:
     return d[string_1_length - 1, string_2_length - 1]
 
 
-def export(modname):
-    """
-    Make the decorated object a member of the named module. This will also add the object under its aliases if it has
-    a `__aliases__` member, thus this decorator should be before the `alias` decorator to pick up those names. Alias
-    names which conflict with package names or existing members will be ignored.
-    """
-
-    def _inner(obj):
-        mod = import_module(modname)
-        if not hasattr(mod, obj.__name__):
-            setattr(mod, obj.__name__, obj)
-
-            # add the aliases for `obj` to the target module
-            for alias in getattr(obj, "__aliases__", ()):
-                if not hasattr(mod, alias):
-                    setattr(mod, alias, obj)
-
-        return obj
-
-    return _inner
-
-
-def load_submodules(
-    basemod: ModuleType, load_all: bool = True, exclude_pattern: str = "(.*[tT]est.*)|(_.*)"
-) -> tuple[list[ModuleType], list[str]]:
-    """
-    Traverse the source of the module structure starting with module `basemod`, loading all packages plus all files if
-    `load_all` is True, excluding anything whose name matches `exclude_pattern`.
-    """
-    submodules = []
-    err_mod: list[str] = []
-    for importer, name, is_pkg in walk_packages(
-        basemod.__path__, prefix=basemod.__name__ + ".", onerror=err_mod.append
-    ):
-        if (is_pkg or load_all) and name not in sys.modules and match(exclude_pattern, name) is None:
-            try:
-                mod = import_module(name)
-                mod_spec = importer.find_spec(name)  # type: ignore
-                if mod_spec and mod_spec.loader:
-                    loader = mod_spec.loader
-                    loader.exec_module(mod)
-                    submodules.append(mod)
-            except OptionalImportError:
-                pass  # could not import the optional deps., they are ignored
-            except ImportError as e:
-                msg = (
-                    "\nMultiple versions of MONAI may have been installed?\n"
-                    "Please see the installation guide: https://docs.monai.io/en/stable/installation.html\n"
-                )  # issue project-monai/monai#5193
-                raise type(e)(f"{e}\n{msg}").with_traceback(e.__traceback__) from e  # raise with modified message
-
-    return submodules, err_mod
-
-
 def instantiate(__path: str, __mode: str, **kwargs: Any) -> Any:
     """
     Create an object instance or call a callable object from a class or function represented by ``_path``.
diff --git a/tests/min_tests.py b/tests/min_tests.py
index f39d3f9843..f5b715e979 100644
--- a/tests/min_tests.py
+++ b/tests/min_tests.py
@@ -232,12 +232,6 @@ def run_testsuit():
 
 
 if __name__ == "__main__":
-    # testing import submodules
-    from monai.utils.module import load_submodules
-
-    _, err_mod = load_submodules(sys.modules["monai"], True)
-    assert not err_mod, f"err_mod={err_mod} not empty"
-
     # testing all modules
     test_runner = unittest.TextTestRunner(stream=sys.stdout, verbosity=2)
     result = test_runner.run(run_testsuit())
diff --git a/tests/test_fastmri_reader.py b/tests/test_fastmri_reader.py
index af2eed7db5..06c3954eae 100644
--- a/tests/test_fastmri_reader.py
+++ b/tests/test_fastmri_reader.py
@@ -17,7 +17,7 @@
 from parameterized import parameterized
 
 from monai.apps.reconstruction.fastmri_reader import FastMRIReader
-from tests.utils import assert_allclose
+from tests.utils import SkipIfNoModule, assert_allclose
 
 TEST_CASE1 = [
     {
@@ -64,6 +64,7 @@
 ]
 
 
+@SkipIfNoModule("h5py")
 class TestMRIUtils(unittest.TestCase):
 
     @parameterized.expand([TEST_CASE1, TEST_CASE2])
diff --git a/tests/test_handler_garbage_collector.py b/tests/test_handler_garbage_collector.py
index 317eba1b11..4254a73a6b 100644
--- a/tests/test_handler_garbage_collector.py
+++ b/tests/test_handler_garbage_collector.py
@@ -19,10 +19,9 @@
 from ignite.engine import Engine
 from parameterized import parameterized
 
-from monai.config import IgniteInfo
 from monai.data import Dataset
 from monai.handlers import GarbageCollector
-from monai.utils import min_version, optional_import
+from monai.utils import IgniteInfo, min_version, optional_import
 
 Events, has_ignite = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 

From 8546be098daaf6841c39a2748412bbda83929c92 Mon Sep 17 00:00:00 2001
From: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Date: Fri, 27 Sep 2024 14:39:20 +0100
Subject: [PATCH 160/183] Re-adding load_submodules  (#8118)

### Description

This restores load_modules for now to resolve importation issues. Doing
a two-pass loading process seems to allow some references which normally
Python wouldn't permit. The fact that these issues weren't caught in
testing is rather strange, and only showed up when symbols in
`monai.apps` were references in bundles. I'm not sure if this is all
related to circular imports, if parts of MONAI aren't being tested
properly, or the way the symbol resolution works within `monai.bundle`
should be improved.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Eric Kerfoot <eric.kerfoot@kcl.ac.uk>
---
 monai/__init__.py       | 38 +++++++++++++++++++++-----------------
 monai/utils/__init__.py |  1 +
 monai/utils/module.py   | 36 +++++++++++++++++++++++++++++++++++-
 tests/min_tests.py      |  6 ++++++
 4 files changed, 63 insertions(+), 18 deletions(-)

diff --git a/monai/__init__.py b/monai/__init__.py
index 46f7510915..f6fc8b0646 100644
--- a/monai/__init__.py
+++ b/monai/__init__.py
@@ -80,25 +80,29 @@ def filter(self, record):
     )
 
 
-from . import (  # noqa: E402
-    apps,
-    auto3dseg,
-    bundle,
-    config,
-    data,
-    engines,
-    fl,
-    handlers,
-    inferers,
-    losses,
-    metrics,
-    networks,
-    optimizers,
-    transforms,
-    utils,
-    visualize,
+from .utils.module import load_submodules  # noqa: E402
+
+# handlers_* have some external decorators the users may not have installed
+# *.so files and folder "_C" may not exist when the cpp extensions are not compiled
+excludes = "|".join(
+    [
+        "(^(monai.handlers))",
+        "(^(monai.bundle))",
+        "(^(monai.fl))",
+        "((\\.so)$)",
+        "(^(monai._C))",
+        "(.*(__main__)$)",
+        "(.*(video_dataset)$)",
+        "(.*(nnunet).*$)",
+    ]
 )
 
+# load directory modules only, skip loading individual files
+load_submodules(sys.modules[__name__], False, exclude_pattern=excludes)
+
+# load all modules, this will trigger all export decorations
+load_submodules(sys.modules[__name__], True, exclude_pattern=excludes)
+
 __all__ = [
     "apps",
     "auto3dseg",
diff --git a/monai/utils/__init__.py b/monai/utils/__init__.py
index b5b2d7a525..916c1a6c70 100644
--- a/monai/utils/__init__.py
+++ b/monai/utils/__init__.py
@@ -112,6 +112,7 @@
     get_package_version,
     get_torch_version_tuple,
     instantiate,
+    load_submodules,
     look_up_option,
     min_version,
     optional_import,
diff --git a/monai/utils/module.py b/monai/utils/module.py
index 3f07dd9e91..df5fe873ae 100644
--- a/monai/utils/module.py
+++ b/monai/utils/module.py
@@ -16,13 +16,15 @@
 import os
 import pdb
 import re
+import sys
 import warnings
 from collections.abc import Callable, Collection, Hashable, Mapping
 from functools import partial, wraps
 from importlib import import_module
+from pkgutil import walk_packages
 from pydoc import locate
 from re import match
-from types import FunctionType
+from types import FunctionType, ModuleType
 from typing import Any, Iterable, cast
 
 import torch
@@ -168,6 +170,38 @@ def damerau_levenshtein_distance(s1: str, s2: str) -> int:
     return d[string_1_length - 1, string_2_length - 1]
 
 
+def load_submodules(
+    basemod: ModuleType, load_all: bool = True, exclude_pattern: str = "(.*[tT]est.*)|(_.*)"
+) -> tuple[list[ModuleType], list[str]]:
+    """
+    Traverse the source of the module structure starting with module `basemod`, loading all packages plus all files if
+    `load_all` is True, excluding anything whose name matches `exclude_pattern`.
+    """
+    submodules = []
+    err_mod: list[str] = []
+    for importer, name, is_pkg in walk_packages(
+        basemod.__path__, prefix=basemod.__name__ + ".", onerror=err_mod.append
+    ):
+        if (is_pkg or load_all) and name not in sys.modules and match(exclude_pattern, name) is None:
+            try:
+                mod = import_module(name)
+                mod_spec = importer.find_spec(name)  # type: ignore
+                if mod_spec and mod_spec.loader:
+                    loader = mod_spec.loader
+                    loader.exec_module(mod)
+                    submodules.append(mod)
+            except OptionalImportError:
+                pass  # could not import the optional deps., they are ignored
+            except ImportError as e:
+                msg = (
+                    "\nMultiple versions of MONAI may have been installed?\n"
+                    "Please see the installation guide: https://docs.monai.io/en/stable/installation.html\n"
+                )  # issue project-monai/monai#5193
+                raise type(e)(f"{e}\n{msg}").with_traceback(e.__traceback__) from e  # raise with modified message
+
+    return submodules, err_mod
+
+
 def instantiate(__path: str, __mode: str, **kwargs: Any) -> Any:
     """
     Create an object instance or call a callable object from a class or function represented by ``_path``.
diff --git a/tests/min_tests.py b/tests/min_tests.py
index f5b715e979..f39d3f9843 100644
--- a/tests/min_tests.py
+++ b/tests/min_tests.py
@@ -232,6 +232,12 @@ def run_testsuit():
 
 
 if __name__ == "__main__":
+    # testing import submodules
+    from monai.utils.module import load_submodules
+
+    _, err_mod = load_submodules(sys.modules["monai"], True)
+    assert not err_mod, f"err_mod={err_mod} not empty"
+
     # testing all modules
     test_runner = unittest.TextTestRunner(stream=sys.stdout, verbosity=2)
     result = test_runner.run(run_testsuit())

From 76ef9f40c8da626928238c91eacddc789b0b4530 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 30 Sep 2024 15:41:14 +0800
Subject: [PATCH 161/183] Fix IgniteInfo can not import issue (#8121)

Fixes #8120 .

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/config/__init__.py     |  1 +
 monai/config/deviceconfig.py | 10 ++++++++++
 monai/utils/enums.py         |  2 +-
 3 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/monai/config/__init__.py b/monai/config/__init__.py
index a83889aee0..c814e1f8eb 100644
--- a/monai/config/__init__.py
+++ b/monai/config/__init__.py
@@ -14,6 +14,7 @@
 from .deviceconfig import (
     USE_COMPILED,
     USE_META_DICT,
+    IgniteInfo,
     get_config_values,
     get_gpu_info,
     get_optional_config_values,
diff --git a/monai/config/deviceconfig.py b/monai/config/deviceconfig.py
index 7ac1b29919..05842245ce 100644
--- a/monai/config/deviceconfig.py
+++ b/monai/config/deviceconfig.py
@@ -23,6 +23,8 @@
 import torch
 
 import monai
+from monai.utils.deprecate_utils import deprecated
+from monai.utils.enums import IgniteInfo as _IgniteInfo
 from monai.utils.module import OptionalImportError, get_package_version, optional_import
 
 try:
@@ -45,6 +47,7 @@
     "print_debug_info",
     "USE_COMPILED",
     "USE_META_DICT",
+    "IgniteInfo",
 ]
 
 
@@ -260,5 +263,12 @@ def print_debug_info(file: TextIO = sys.stdout) -> None:
     print_gpu_info(file)
 
 
+@deprecated(since="1.4.0", removed="1.6.0", msg_suffix="Please use `monai.utils.enums.IgniteInfo` instead.")
+class IgniteInfo:
+    """Deprecated Import of IgniteInfo enum, which was moved to `monai.utils.enums.IgniteInfo`."""
+
+    OPT_IMPORT_VERSION = _IgniteInfo.OPT_IMPORT_VERSION
+
+
 if __name__ == "__main__":
     print_debug_info()
diff --git a/monai/utils/enums.py b/monai/utils/enums.py
index 900133f2f2..1fbf3ffa05 100644
--- a/monai/utils/enums.py
+++ b/monai/utils/enums.py
@@ -727,7 +727,7 @@ class IgniteInfo(StrEnum):
 
     """
 
-    OPT_IMPORT_VERSION = "0.4.4"
+    OPT_IMPORT_VERSION = "0.4.11"
 
 
 if TYPE_CHECKING:

From 796271ccd4ee48065af3b2afd56bf46960a53112 Mon Sep 17 00:00:00 2001
From: Hsin-Tong Hsieh <75067139+ctongh@users.noreply.github.com>
Date: Sun, 13 Oct 2024 05:55:46 +0800
Subject: [PATCH 162/183] Add options to skip operations for RestoreLabeld
 Transform (#8125)

Fixes #6380

### Description

Four new bool parameters are added into `RestoreLabeld` to allow users
to selectively enable or disable each restoration operation as needed,
and a corresponding test case is added to verify that the function runs
correctly.

This design allows users to selectively enable or disable each
restoration operation as needed, providing greater flexibility.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Hsin Tong <zcc22592590@gmail.com>
Signed-off-by: Hsin-Tong Hsieh <75067139+ctongh@users.noreply.github.com>
Signed-off-by: kbbbbkb <139567836+Kbinn@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
Co-authored-by: kbbbbkb <139567836+Kbinn@users.noreply.github.com>
---
 monai/apps/deepgrow/transforms.py | 69 ++++++++++++++--------
 tests/test_deepgrow_transforms.py | 95 ++++++++++++++++++++++++++++++-
 2 files changed, 140 insertions(+), 24 deletions(-)

diff --git a/monai/apps/deepgrow/transforms.py b/monai/apps/deepgrow/transforms.py
index c2f97091fd..721c0db489 100644
--- a/monai/apps/deepgrow/transforms.py
+++ b/monai/apps/deepgrow/transforms.py
@@ -803,6 +803,14 @@ class RestoreLabeld(MapTransform):
         original_shape_key: key that records original shape for foreground.
         cropped_shape_key: key that records cropped shape for foreground.
         allow_missing_keys: don't raise exception if key is missing.
+        restore_resizing: used to enable or disable resizing restoration, default is True.
+            If True, the transform will resize the items back to its original shape.
+        restore_cropping: used to enable or disable cropping restoration, default is True.
+            If True, the transform will restore the items to its uncropped size.
+        restore_spacing: used to enable or disable spacing restoration, default is True.
+            If True, the transform will resample the items back to the spacing it had before being altered.
+        restore_slicing: used to enable or disable slicing restoration, default is True.
+            If True, the transform will reassemble the full volume by restoring the slices to their original positions.
     """
 
     def __init__(
@@ -819,6 +827,10 @@ def __init__(
         original_shape_key: str = "foreground_original_shape",
         cropped_shape_key: str = "foreground_cropped_shape",
         allow_missing_keys: bool = False,
+        restore_resizing: bool = True,
+        restore_cropping: bool = True,
+        restore_spacing: bool = True,
+        restore_slicing: bool = True,
     ) -> None:
         super().__init__(keys, allow_missing_keys)
         self.ref_image = ref_image
@@ -833,6 +845,10 @@ def __init__(
         self.end_coord_key = end_coord_key
         self.original_shape_key = original_shape_key
         self.cropped_shape_key = cropped_shape_key
+        self.restore_resizing = restore_resizing
+        self.restore_cropping = restore_cropping
+        self.restore_spacing = restore_spacing
+        self.restore_slicing = restore_slicing
 
     def __call__(self, data: Any) -> dict:
         d = dict(data)
@@ -842,38 +858,45 @@ def __call__(self, data: Any) -> dict:
             image = d[key]
 
             # Undo Resize
-            current_shape = image.shape
-            cropped_shape = meta_dict[self.cropped_shape_key]
-            if np.any(np.not_equal(current_shape, cropped_shape)):
-                resizer = Resize(spatial_size=cropped_shape[1:], mode=mode)
-                image = resizer(image, mode=mode, align_corners=align_corners)
+            if self.restore_resizing:
+                current_shape = image.shape
+                cropped_shape = meta_dict[self.cropped_shape_key]
+                if np.any(np.not_equal(current_shape, cropped_shape)):
+                    resizer = Resize(spatial_size=cropped_shape[1:], mode=mode)
+                    image = resizer(image, mode=mode, align_corners=align_corners)
 
             # Undo Crop
-            original_shape = meta_dict[self.original_shape_key]
-            result = np.zeros(original_shape, dtype=np.float32)
-            box_start = meta_dict[self.start_coord_key]
-            box_end = meta_dict[self.end_coord_key]
-
-            spatial_dims = min(len(box_start), len(image.shape[1:]))
-            slices = tuple(
-                [slice(None)] + [slice(s, e) for s, e in zip(box_start[:spatial_dims], box_end[:spatial_dims])]
-            )
-            result[slices] = image
+            if self.restore_cropping:
+                original_shape = meta_dict[self.original_shape_key]
+                result = np.zeros(original_shape, dtype=np.float32)
+                box_start = meta_dict[self.start_coord_key]
+                box_end = meta_dict[self.end_coord_key]
+
+                spatial_dims = min(len(box_start), len(image.shape[1:]))
+                slices = tuple(
+                    [slice(None)] + [slice(s, e) for s, e in zip(box_start[:spatial_dims], box_end[:spatial_dims])]
+                )
+                result[slices] = image
+            else:
+                result = image
 
             # Undo Spacing
-            current_size = result.shape[1:]
-            # change spatial_shape from HWD to DHW
-            spatial_shape = list(np.roll(meta_dict["spatial_shape"], 1))
-            spatial_size = spatial_shape[-len(current_size) :]
+            if self.restore_spacing:
+                current_size = result.shape[1:]
+                # change spatial_shape from HWD to DHW
+                spatial_shape = list(np.roll(meta_dict["spatial_shape"], 1))
+                spatial_size = spatial_shape[-len(current_size) :]
 
-            if np.any(np.not_equal(current_size, spatial_size)):
-                resizer = Resize(spatial_size=spatial_size, mode=mode)
-                result = resizer(result, mode=mode, align_corners=align_corners)  # type: ignore
+                if np.any(np.not_equal(current_size, spatial_size)):
+                    resizer = Resize(spatial_size=spatial_size, mode=mode)
+                    result = resizer(result, mode=mode, align_corners=align_corners)  # type: ignore
 
             # Undo Slicing
             slice_idx = meta_dict.get("slice_idx")
             final_result: NdarrayOrTensor
-            if slice_idx is None or self.slice_only:
+            if not self.restore_slicing:  # do nothing if restore slicing isn't requested
+                final_result = result
+            elif slice_idx is None or self.slice_only:
                 final_result = result if len(result.shape) <= 3 else result[0]
             else:
                 slice_idx = meta_dict["slice_idx"][0]
diff --git a/tests/test_deepgrow_transforms.py b/tests/test_deepgrow_transforms.py
index a491a8004b..091d00afcd 100644
--- a/tests/test_deepgrow_transforms.py
+++ b/tests/test_deepgrow_transforms.py
@@ -141,6 +141,21 @@
 
 DATA_12 = {"image": np.arange(27).reshape(3, 3, 3), PostFix.meta("image"): {}, "guidance": [[0, 0, 0], [0, 1, 1], 1]}
 
+DATA_13 = {
+    "image": np.arange(64).reshape((1, 4, 4, 4)),
+    PostFix.meta("image"): {
+        "spatial_shape": [8, 8, 4],
+        "foreground_start_coord": np.array([1, 1, 1]),
+        "foreground_end_coord": np.array([3, 3, 3]),
+        "foreground_original_shape": (1, 4, 4, 4),
+        "foreground_cropped_shape": (1, 2, 2, 2),
+        "original_affine": np.array(
+            [[[1.0, 0.0, 0.0, 0.0], [0.0, 1.0, 0.0, 0.0], [0.0, 0.0, 1.0, 0.0], [0.0, 0.0, 0.0, 1.0]]]
+        ),
+    },
+    "pred": np.array([[[[10, 20], [30, 40]], [[50, 60], [70, 80]]]]),
+}
+
 FIND_SLICE_TEST_CASE_1 = [{"label": "label", "sids": "sids"}, DATA_1, [0]]
 
 FIND_SLICE_TEST_CASE_2 = [{"label": "label", "sids": "sids"}, DATA_2, [0, 1]]
@@ -329,6 +344,74 @@
 
 RESTORE_LABEL_TEST_CASE_2 = [{"keys": ["pred"], "ref_image": "image", "mode": "nearest"}, DATA_11, RESULT]
 
+RESTORE_LABEL_TEST_CASE_3_RESULT = np.zeros((10, 20, 20))
+RESTORE_LABEL_TEST_CASE_3_RESULT[:5, 0:10, 0:10] = 1
+RESTORE_LABEL_TEST_CASE_3_RESULT[:5, 0:10, 10:20] = 2
+RESTORE_LABEL_TEST_CASE_3_RESULT[:5, 10:20, 0:10] = 3
+RESTORE_LABEL_TEST_CASE_3_RESULT[:5, 10:20, 10:20] = 4
+RESTORE_LABEL_TEST_CASE_3_RESULT[5:10, 0:10, 0:10] = 5
+RESTORE_LABEL_TEST_CASE_3_RESULT[5:10, 0:10, 10:20] = 6
+RESTORE_LABEL_TEST_CASE_3_RESULT[5:10, 10:20, 0:10] = 7
+RESTORE_LABEL_TEST_CASE_3_RESULT[5:10, 10:20, 10:20] = 8
+
+RESTORE_LABEL_TEST_CASE_3 = [
+    {"keys": ["pred"], "ref_image": "image", "mode": "nearest", "restore_cropping": False},
+    DATA_11,
+    RESTORE_LABEL_TEST_CASE_3_RESULT,
+]
+
+RESTORE_LABEL_TEST_CASE_4_RESULT = np.zeros((4, 8, 8))
+RESTORE_LABEL_TEST_CASE_4_RESULT[1, 2:6, 2:6] = np.array(
+    [[10.0, 10.0, 20.0, 20.0], [10.0, 10.0, 20.0, 20.0], [30.0, 30.0, 40.0, 40.0], [30.0, 30.0, 40.0, 40.0]]
+)
+RESTORE_LABEL_TEST_CASE_4_RESULT[2, 2:6, 2:6] = np.array(
+    [[50.0, 50.0, 60.0, 60.0], [50.0, 50.0, 60.0, 60.0], [70.0, 70.0, 80.0, 80.0], [70.0, 70.0, 80.0, 80.0]]
+)
+
+RESTORE_LABEL_TEST_CASE_4 = [
+    {"keys": ["pred"], "ref_image": "image", "mode": "nearest", "restore_resizing": False},
+    DATA_13,
+    RESTORE_LABEL_TEST_CASE_4_RESULT,
+]
+
+RESTORE_LABEL_TEST_CASE_5_RESULT = np.zeros((4, 4, 4))
+RESTORE_LABEL_TEST_CASE_5_RESULT[1, 1:3, 1:3] = np.array([[10.0, 20.0], [30.0, 40.0]])
+RESTORE_LABEL_TEST_CASE_5_RESULT[2, 1:3, 1:3] = np.array([[50.0, 60.0], [70.0, 80.0]])
+
+RESTORE_LABEL_TEST_CASE_5 = [
+    {"keys": ["pred"], "ref_image": "image", "mode": "nearest", "restore_spacing": False},
+    DATA_13,
+    RESTORE_LABEL_TEST_CASE_5_RESULT,
+]
+
+RESTORE_LABEL_TEST_CASE_6_RESULT = np.zeros((1, 4, 8, 8))
+RESTORE_LABEL_TEST_CASE_6_RESULT[-1, 1, 2:6, 2:6] = np.array(
+    [[10.0, 10.0, 20.0, 20.0], [10.0, 10.0, 20.0, 20.0], [30.0, 30.0, 40.0, 40.0], [30.0, 30.0, 40.0, 40.0]]
+)
+RESTORE_LABEL_TEST_CASE_6_RESULT[-1, 2, 2:6, 2:6] = np.array(
+    [[50.0, 50.0, 60.0, 60.0], [50.0, 50.0, 60.0, 60.0], [70.0, 70.0, 80.0, 80.0], [70.0, 70.0, 80.0, 80.0]]
+)
+
+RESTORE_LABEL_TEST_CASE_6 = [
+    {"keys": ["pred"], "ref_image": "image", "mode": "nearest", "restore_slicing": False},
+    DATA_13,
+    RESTORE_LABEL_TEST_CASE_6_RESULT,
+]
+
+RESTORE_LABEL_TEST_CASE_7 = [
+    {
+        "keys": ["pred"],
+        "ref_image": "image",
+        "mode": "nearest",
+        "restore_resizing": False,
+        "restore_cropping": False,
+        "restore_spacing": False,
+        "restore_slicing": False,
+    },
+    DATA_11,
+    np.array([[[[1, 2], [3, 4]], [[5, 6], [7, 8]]]]),
+]
+
 FETCH_2D_SLICE_TEST_CASE_1 = [
     {"keys": ["image"], "guidance": "guidance"},
     DATA_12,
@@ -445,7 +528,17 @@ def test_correct_results(self, arguments, input_data, expected_result):
 
 class TestRestoreLabeld(unittest.TestCase):
 
-    @parameterized.expand([RESTORE_LABEL_TEST_CASE_1, RESTORE_LABEL_TEST_CASE_2])
+    @parameterized.expand(
+        [
+            RESTORE_LABEL_TEST_CASE_1,
+            RESTORE_LABEL_TEST_CASE_2,
+            RESTORE_LABEL_TEST_CASE_3,
+            RESTORE_LABEL_TEST_CASE_4,
+            RESTORE_LABEL_TEST_CASE_5,
+            RESTORE_LABEL_TEST_CASE_6,
+            RESTORE_LABEL_TEST_CASE_7,
+        ]
+    )
     def test_correct_results(self, arguments, input_data, expected_result):
         result = RestoreLabeld(**arguments)(input_data)
         np.testing.assert_allclose(result["pred"], expected_result)

From 3fd73f199af518172ac48909caaeb30d14df63c8 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Sun, 13 Oct 2024 15:42:12 +0800
Subject: [PATCH 163/183] 8110 update changelog for v1.4 (#8135)

part of #8110

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 CHANGELOG.md                | 95 ++++++++++++++++++++++++++++++++++++-
 docs/source/installation.md |  4 +-
 2 files changed, 96 insertions(+), 3 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 804508c262..53f1049449 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,6 +5,98 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ## [Unreleased]
 
+## [1.4.0] - 2024-10-15
+## What's Changed
+### Added
+* Implemented Conjugate Gradient Solver to generate confidence maps. (#7876)
+* Added norm parameter to `ResNet` (#7752, #7805)
+* Introduced alpha parameter to `DiceFocalLoss` for improved flexibility (#7841)
+* Integrated Tailored ControlNet Implementations (#7875)
+* Integrated Tailored Auto-Encoder Model (#7861)
+* Integrated Tailored Diffusion U-Net Model (7867)
+* Added Maisi morphological functions (#7893)
+* Added support for downloading bundles from NGC private registry (#7907, #7929, #8076)
+* Integrated generative refactor into the core (#7886, #7962)
+* Made `ViT` and `UNETR` models compatible with TorchScript (#7937)
+* Implemented post-download checks for MONAI bundles and compatibility warnings (#7938)
+* Added NGC prefix argument when downloading bundles (#7974)
+* Added flash attention support in the attention block for improved performance (#7977)
+* Enhanced `MLPBlock` for compatibility with VISTA-3D (#7995)
+* Added support for Neighbor-Aware Calibration Loss (NACL) for calibrated models in segmentation tasks (#7819)
+* Added label_smoothing parameter to `DiceCELoss` for enhanced model calibration (#8000)
+* Add `include_fc` and `use_combined_linear` argument in the `SABlock` (#7996)
+* Added utilities, networks, and an inferer specific to VISTA-3D (#7999, #7987, #8047, #8059, #8021)
+* Integrated a new network, `CellSamWrapper`, for cell-based applications (#7981)
+* Introduced `WriteFileMapping` transform to map between input image paths and their corresponding output paths (#7769)
+* Added `TrtHandler` to accelerate models using TensorRT (#7990, #8064)
+* Added box and points conversion transforms for more flexible spatial manipulation (#8053)
+* Enhanced `RandSimulateLowResolutiond` transform with deterministic support (#8057)
+* Added a contiguous argument to the `Fourier` class to facilitate contiguous tensor outputs (#7969)
+* Allowed `ApplyTransformToPointsd` to receive a sequence of reference keys for more versatile point manipulation (#8063)
+* Made `MetaTensor` an optional print in `DataStats` and `DataStatsd` for more concise logging (#7814)
+#### misc.
+* Refactored Dataset to utilize Compose for handling transforms. (#7784)
+* Combined `map_classes_to_indices` and `generate_label_classes_crop_centers` into a unified function (#7712)
+* Introduced metadata schema directly into the codebase for improved structure and validation (#7409)
+* Renamed `optional_packages_version` to `required_packages_version` for clearer package dependency management. (#7253)
+* Replaced `pkg_resources` with the more modern packaging module for package handling (#7953)
+* Refactored MAISI-related networks to align with the new generative components (#7989, #7993, #8005)
+* Added a badge displaying monthly download statistics to enhance project visibility (#7891)
+### Fixed
+#### transforms
+* Ensured deterministic behavior in `MixUp`, `CutMix`, and `CutOut` transforms (#7813)
+* Applied a minor correction to `AsDiscrete` transform (#7984)
+* Fixed handling of integer weightmaps in `RandomWeightedCrop` (#8097)
+* Resolved data type bug in `ScaleIntensityRangePercentile` (#8109)
+#### data
+* Fixed negative strides issue in the `NrrdReader` (#7809)
+* Addressed wsireader issue with retrieving MPP (7921)
+* Ensured location is returned as a tuple in wsireader (#8007)
+* Corrected interpretation of space directions in nrrd reader (#8091)
+#### metrics and losses
+* Improved memory management for `NACLLoss` (#8020)
+* Fixed reduction logic in `GeneralizedDiceScore` (#7970)
+#### networks
+* Resolved issue with loading pre-trained weights in `ResNet` (#7924)
+* Fixed error where `torch.device` object had no attribute gpu_id during TensorRT export (#8019)
+* Corrected function for loading older weights in `DiffusionModelUNet` (#8031)
+* Switched to `torch_tensorrt.Device` instead of `torch.device` during TensorRT compilation (#8051)
+#### engines and handlers
+* Attempted to resolve the "experiment already exists" issue in `MLFlowHandler` (#7916)
+* Refactored the model export process for conversion and saving (#7934)
+#### misc.
+* Adjusted requirements to exclude Numpy version 2.0 (#7859)
+* Updated deprecated `scipy.ndimage` namespaces in optional imports (#7847, #7897)
+* Resolved `load_module()` deprecation in Python 3.12 (#7881)
+* Fixed Ruff type check issues (#7885)
+* Cleaned disk space in the conda test pipeline (#7902)
+* Replaced deprecated `pkgutil.find_loader` usage  (#7906)
+* Enhanced docstrings in various modules (#7913, #8055)
+* Test cases fixing (#7905, #7794, #7808)
+* Fix mypy issue introduced in 1.11.0 (#7941)
+* Cleaned up warnings during test collection (#7914)
+* Fix incompatible types in assignment issue (#7950)
+* Fix outdated link in the docs (#7971)
+* Addressed CI issues (#7983, #8013)
+* Fix module can not import correctly issue (#8015)
+* Fix AttributeError when using torch.min and max (#8041)
+* Ensure synchronization by adding `cuda.synchronize` (#8058)
+* Ignore warning from nptyping as workaround (#8062)
+* Suppress deprecated warning when importing monai (#8067)
+* Fix link in test bundle under MONAI-extra-test-data (#8092)
+### Changed
+* Base Docker image upgraded to `nvcr.io/nvidia/pytorch:24.08-py3` from `nvcr.io/nvidia/pytorch:23.08-py3`
+* Change blossom-ci to ACL security format (#7843)
+* Move PyType test to weekly test (#8025)
+* Adjusted to meet Numpy 2.0 requirements (#7857)
+### Deprecated
+* Dropped support for Python 3.8 (#7909)
+* Remove deprecated arguments and class for v1.4 (#8079)
+### Removed
+* Remove use of deprecated python 3.12 strtobool (#7900)
+* Removed the pipeline for publishing to testpypi (#8086)
+* Cleaning up some very old and now obsolete infrastructure (#8113, #8118, #8121)
+
 ## [1.3.2] - 2024-06-25
 ### Fixed
 #### misc.
@@ -1040,7 +1132,8 @@ the postprocessing steps should be used before calling the metrics methods
 
 [highlights]: https://github.com/Project-MONAI/MONAI/blob/master/docs/source/highlights.md
 
-[Unreleased]: https://github.com/Project-MONAI/MONAI/compare/1.3.2...HEAD
+[Unreleased]: https://github.com/Project-MONAI/MONAI/compare/1.4.0...HEAD
+[1.4.0]: https://github.com/Project-MONAI/MONAI/compare/1.3.2...1.4.0
 [1.3.2]: https://github.com/Project-MONAI/MONAI/compare/1.3.1...1.3.2
 [1.3.1]: https://github.com/Project-MONAI/MONAI/compare/1.3.0...1.3.1
 [1.3.0]: https://github.com/Project-MONAI/MONAI/compare/1.2.0...1.3.0
diff --git a/docs/source/installation.md b/docs/source/installation.md
index 70a8b6f1d4..4308a07647 100644
--- a/docs/source/installation.md
+++ b/docs/source/installation.md
@@ -254,10 +254,10 @@ Since MONAI v0.2.0, the extras syntax such as `pip install 'monai[nibabel]'` is
 - The options are
 
 ```
-[nibabel, skimage, scipy, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, cucim, openslide, pandas, einops, transformers, mlflow, clearml, matplotlib, tensorboardX, tifffile, imagecodecs, pyyaml, fire, jsonschema, ninja, pynrrd, pydicom, h5py, nni, optuna, onnx, onnxruntime, zarr, lpips, pynvml, huggingface_hub, segment-anything]
+[nibabel, skimage, scipy, pillow, tensorboard, gdown, ignite, torchvision, itk, tqdm, lmdb, psutil, cucim, openslide, pandas, einops, transformers, mlflow, clearml, matplotlib, tensorboardX, tifffile, imagecodecs, pyyaml, fire, jsonschema, ninja, pynrrd, pydicom, h5py, nni, optuna, onnx, onnxruntime, zarr, lpips, pynvml, huggingface_hub]
 ```
 
 which correspond to `nibabel`, `scikit-image`,`scipy`, `pillow`, `tensorboard`,
-`gdown`, `pytorch-ignite`, `torchvision`, `itk`, `tqdm`, `lmdb`, `psutil`, `cucim`, `openslide-python`, `pandas`, `einops`, `transformers`, `mlflow`, `clearml`, `matplotlib`, `tensorboardX`, `tifffile`, `imagecodecs`, `pyyaml`, `fire`, `jsonschema`, `ninja`, `pynrrd`, `pydicom`, `h5py`, `nni`, `optuna`, `onnx`, `onnxruntime`, `zarr`, `lpips`, `nvidia-ml-py`, `huggingface_hub`, `pyamg` and `segment-anything` respectively.
+`gdown`, `pytorch-ignite`, `torchvision`, `itk`, `tqdm`, `lmdb`, `psutil`, `cucim`, `openslide-python`, `pandas`, `einops`, `transformers`, `mlflow`, `clearml`, `matplotlib`, `tensorboardX`, `tifffile`, `imagecodecs`, `pyyaml`, `fire`, `jsonschema`, `ninja`, `pynrrd`, `pydicom`, `h5py`, `nni`, `optuna`, `onnx`, `onnxruntime`, `zarr`, `lpips`, `nvidia-ml-py`, `huggingface_hub` and `pyamg` respectively.
 
 - `pip install 'monai[all]'` installs all the optional dependencies.

From c56cfe7fc59cb92be910d1a0774bc37ec0995c45 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 14 Oct 2024 11:54:53 +0800
Subject: [PATCH 164/183] 8110 update highlights page for v1.4 (#8111)

Part of #8110

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Eric Kerfoot <17726042+ericspod@users.noreply.github.com>
---
 docs/images/maisi_train.png | Bin 0 -> 187025 bytes
 docs/images/vista2d.png     | Bin 0 -> 455243 bytes
 docs/images/vista3d.png     | Bin 0 -> 87074 bytes
 docs/source/whatsnew.rst    |   1 +
 docs/source/whatsnew_1_3.md |   2 +-
 docs/source/whatsnew_1_4.md |  68 ++++++++++++++++++++++++++++++++++++
 6 files changed, 70 insertions(+), 1 deletion(-)
 create mode 100644 docs/images/maisi_train.png
 create mode 100644 docs/images/vista2d.png
 create mode 100644 docs/images/vista3d.png
 create mode 100644 docs/source/whatsnew_1_4.md

diff --git a/docs/images/maisi_train.png b/docs/images/maisi_train.png
new file mode 100644
index 0000000000000000000000000000000000000000..8c4936456d897f20a6880cefb3d413b0ae89f9eb
GIT binary patch
literal 187025
zcmeFZXIPV47d99KM0)QfN)eT&RH;#sE<^+YsYelM0wOIyOcbPd5l~PfU8Dr0w?w)~
z?+|K0dP}H*gfMZwHs5>RAMboqu9@pP581z-guT{YYu|gdz0M}i<^gPmx(2!cDk>_#
zUz9(<*%Uw<Kz;7qpAY4sp?qjB(9+V-&@#}`oxi}uz{JGJz{tqV!okMO!p_3T$i~CQ
z&dJ5i&CSHh%g4jT$HB$T_2-{ZQB!_HLrYIfOV7p3$jtSB`Z@anV83u~fEGYaB?>sl
zPDRa5b=C#|0svIB6l(tv{9hlca}*lS(_NrvV5EGZiVbj%ikkWy4fP+YDc^oYc@Lmr
zr{%bG<Mw$@)5mn8-dyr;lCmy{-6?P8elUm=S9tO=gr0$imycgS;<DtGtJf5jlvQr3
zs%hWV(bdy8Ff=oNXklsf$lCF#(=%rmS2rJDKmPzoU{L7Wu<(e;sOWde?^9CKK735i
z&dJToFZf#c?R!OKRdr2mU42VyTYE?6&#qrX!y}_(;}eq@?84&jrR9~?wRQaN-u}Vi
z5#jjc4_{OiKL6YEUl{urzSt>zoui?lrlI@87uC4{%AjVap}lnDJjZQQy2sv}qVjJp
zaNSADDsQG2Q+R;me)4jVfk#{sD}n#R+W*ekf5uqI{}N~aow5JH7Y4vYO+^tNH9G(V
zI5jI3Y(og61PChDb%Uhac-EGB9;8Q2|L}o~CuSB>bH#5)u-56<_p+bA*@DM?9E1t1
zSVSmU42E&2tRYEcat)BchaiG3toO8{Ks(^K*Kh``|Avj{R?H_qjKaXfb}8i(c1fu`
z5%llL9UyLajt;b1Ob|Q+Fc^Wz>Ut!_qvrk-DXsHofR=atXMp%oAmN!;=ow(&P=oyp
z@OBAt1`y04^JmPT0Zus*Xy8~?=QBXZdSKyk%I5kpFB)5bVu24n!l{tCQsE^31cK{M
zC}Q6N%6tY0lc4-q_!|h<Ge>s@SbYx>I|I;ngUBj+Bn6e`Gr)<2$OX!@FDcW;4fcY2
zG(S6t9{r7=!7~7}CeDW(-TwxS%dI#AELQ3Ljh@vrK=?BP=nQZ-rx{2vd2|NA*Q*vD
zM^qI3wx_(xB&ePNbew<85gz<`g#rBw)nCs5tqM@!86ewqK7k}DeFh-R0RK+566pG&
zA2ep}|E^VBT`pJ_=|d(3lLdn3S5CPR)Nrf;WqCVY{teAJX8@*N9Go1b7Xre4`m+wc
z@V^l)G(u{f9Nm?Jw}PQU6m}}+y+~L8u(S0y(VN4uA5pAlfWa#`Ju)XncAg3UM)K*1
z55a0Eot?jlHRoSxc1!p-QvEy4K4yEY&Xzx${wCGRzte2`Z=?$PSE}RPV2}UN#Q%%?
z`5#UEKj6zhn)nCF|G~t6F!7($`JdDIzt~*=VB$ZR_>Z0VkH_+l$MQe?(0?%TA58oQ
z6aT@)e=zYMO#Gjtp#S*J|M<@T_|E??`OX1af_UAyMH!wcD;&ARv2=}By)jPPxXvh2
z0@E(|Jkqe5|B(qR4LhcB@o#y9jW=h25KnklDoOneP@6#h<oT^>Ql~a)r^8Uby5<zr
zbMge`K%nF;B=d--@M+{3Ab(A2o#7ZxNfTtIZRwl=UZTkvxw$C%AlO&0#R%kTf9{oR
z6!a7cyWx&zd=90}cbUC73}xkwt{Tj>BeYsnKib-lBz+WTF8`Sj{2r69UHMXfTh(NE
zRZXf{gOjx#4IqrQcsO@dqq$*xb5^#~b{p?p532((=H*vhewCDt*2RNAzbH?oVeuX-
z*xVvMgE%Z6MQc3zVq-JwdURr65#an6(4@&~dlMUfnU}*r{6eC>QQ7qdI>FnWq91He
z1fqAD<qoR}#wC#GVckl+TbJe%c6duppk+;wSoFCa`qj(&l9yO2?JWa($v3^vA;iN-
zLNKnijnl<m0?)8^GGf8*=J1z|QPgrOy@<Ishn^t6RV61hs_w@fed_dcD>;L<39X<Q
z5e`B-_GqjTSfW$BhVL62=8juPrB9>HPQA))uL6iI`l=Bx)dP1&M*_JU>2Pyx0dvfi
z9<mqqx6TtSoA4$@zIo4S8|5U^m`leL?lqeXO~ip`XNF$a?Yy`A;DY9nA;}R{R?+Ox
z$9{@s&DbJ42lKj`nmMrwo1CBB0swid<ktZHO{t2310r06AO;@WH!3VC@heeN?-ne)
zI3?|>cze-DFrM!`ojXeo%>(L}(S(8rHH0ZgqPYv2!Se9+n2j+SS4_U>af{%ryO*lJ
zeOL6lR&QLc&`PQMtj%FK4*1${iZy~PhHr3#sZ|kq5Z;2WL#N+%7p&eF&ziG8r}{2D
zNtY&qul>@l&ahHCXs?chjU(ta6iV0CEV32`-mTSuo{Y#{8s1(mD@WYC(GN*dWGmBP
z+m&UtwbWiX1F#E%q7f{<tE6X&kg6MB5d^*Qd%V=P^q$467;kdI@0Ff)jh?NoTiZd*
z+>0WTr{yp<h}O!(>QN1c?;C~<nZL-GDdWbv)G>V%30ICQk5aEg_VjoASor1_NB3m{
zOz;+Aj>RrV1M33zS)OUiY1O47s@Z-_@da5lY&sVeuOJyu-^0WSY3=o(xJIeP&w50&
z4l+Gncs3yzGN3~}GsUx4ynWC6T6RRZ4mE?q(x=$pneh3v&Ka{dAS+pxWRKS^24c5T
zi(pVyxAMC&A|IdRD}S`T=<NF2fJHci?|yCH)v*4&E66msqgTuvJJhrg648R-*<Q$K
zf9ZMxk5O#5xm$fTf42P$ATnio>+<YVNVl@pHoxST4qRUz-g!}OixuC{it*}THfpw}
z3q26l8V8q`4ODh#RR`Y|+YX#}23?nY2Rj;yVOSaWwN$|EwWc41K~;n!B1txWi+09O
z13avEf6Wf*K7g6;&fmQ!_my9BqaW=m7jr^M2@*)cSO*%+qZz_c2PSK%GIc~-sVZCC
zL#BQDTP}}Pj@{JlF7e<Y{>^>Mo>^%edW#i@yiG}(7U&UT5j@D}lOjjWN3+6}<wg@R
zRH7Io0ISez4$B@_s!pxb{w-BJ*n;4LGugY3`*|JGZ<{)G7q_p_k=KzEAlgoU<wuC)
z^bq&O*t?T7-uGmKX}K32zSp#Q1a~4>4w%(DRH9xjA{g9xGlj=JIcQ&lH8^0~-U6?`
zeQ&}2RE|Tdb{gVA69I>%_=Gmb4^RcX^#XsD16#(aQH%Ajb|J<>>nXc#4{tXMMRU$O
z1F1jwU1khc4Wy+IE`#<`IM9M*B=j*Jz8D`>v<`G1)=(*(oLsvfc%xtv8_bvIH~o!&
zce_*8V1>VvdugsRVoM*YO4No(VdK*lWz3o_8e~1IPW8aJHcVV=S^W9xnuu4GN$($D
zl@oU(zTkSVn%xX5pvdDrN&B==iw`P|7st>=l1yX{Vjx5AMK<b;MPn6l;Y-1xG$t1x
zps&kw^Q{ypp%+-=GHG$>5Z?qoH~Y)wF)f&?_NuDXazXx*yx?#~`q|fE5A?gMxr539
zocaOkK~hOLhjuR}=v~}%loaliRC7WD^Ez>uD2Rb1?8u9K39w}fzpu^YA*r;7-0UKP
zYdS%(bL`8t_|mNN5xH|SkJd!^;4Dogh>0wH`lifX8$Z7fm2t#~B!;hD+KJkX2wPKo
z1FQq40qYzG)hSp_*NcX3^cq@3fYV=Tt<Mgf{tz#z>M562_;j^1b?cLN`Y<7kc$Ywp
z4TuySs5<q0Tx#HTh~Df~Fs>gdeaDovV!&PbK+*d*d(#hoVOQFI{cKbq;fUxA(Z<UC
zm;o)wM#w{|$1Ndg20Tf&N=-u<AF@9*4z?|D3N(EF?jhY(2A#J20)ZdC#Q~$XHH2y7
zkuPO69F%?wNb?3goGDHNy9*k;VCd%faF|`*^G4fosZZy6y{16)^7w*W1Y$uEMJ*5x
zwG??@Kea}__vuk(FGKgT$7D}rca8tHLadK%S}5GLKOD}cw6p2OO32nUGG({zm3DP~
zQRG>@+xXzPZ!qj_L>x~1nw9L*bgkm+0|BWn4Jo1}MZU<$%2Q5#Jos~9q2a*V9=Kq}
z2IEjmlTuxmDiQ7zc5`>8B3{8huW;QWT{a`&s~hk2Nh|BSd(*blH}x0d%q!Gy%wGY#
z1K+xGu9y3_J^g7hnG1Rsuag$BFUKR(QShqPs-!(^wAgaWyqu^pe7C%RH)O8i*<xmj
z*Mi=bDsC975_vvi!GYry!iue=ty)=`*K&8NEyzHAs$KG30Uh(@0q^v4s|VMlBF3~-
zAiUZHcmZK~r)>;`i*)GrVy>>B<}dl}x@47ooG5f|ldJrPeE=I(C(x1jlq4{s#7Gp!
z_vSUVe>%4s)9oJPR2x~J-Bi&fQ1Vmy7u`DtiEstVxmamx%iDsC&!i)aixfkJOU?kU
zy}7><=P>=Qnt8oNYN=p({z38kKRAjlnL|TfxJM-%&JrXf@>3uv)rd`jce;27VXaM@
zJ3v9&wpokrC)!u1ZfJ+o2}Pe<sT;J1;}sTHUc*?7^V%1I?9JI)g|zLxX&<@1GroP>
zCVs)xil+w`25wyXalcbZuN_ob(As}~f&g1FA1k#m4r=$4(P|ktXsk!2j(n^6G?sK{
zSoh2IE?-*_mJ6b<HPxc$lNUQ%kG1&GZI>3BKxDR&V;c+EJ5tA)4yl%cZ!;g$FD?2n
z#yUqQU^_d})KIHYIYl&!JEu|6<c}IN*YtRVPk~*rw0|*5r^oxJqhuV`a3ywt?#rOP
zwlmQhDv81i2}o;6TVyT}iOPawW%YJyHYHQgw@o(~{gYovRq7jPI4U(|6W)}<ZbA?`
zS~rMpnAI^}5H9Kr;B*ZCYft<$Z1k`ZcGSE3>p1XOqxcNK4G5w@rVQ~3RM>$_n1(Pr
zVaJFr%V>4?K1CeHO+k^>$8hJlhNaGrpYUV#36Tg94LV$g`S)R{&B7I5n~kYQUNVo<
z`K|heyo5YC-4gpzE+e{H{xZ9T%|Ce6*QFu|mUv_{T%dA;*iYvU!bBDHAG-f^s!AVP
zz}$N(o!Ap4`t+sBm7i<rqZNb*EO7_uro|9`2GEim-h4b#DlI_IRZ>~HCKSDN=cbHc
zLP`{kPy7|?ob1LUN@YwXT`T3tBRI6fL<gV-wB#79O;M5Nv%RiVrOJ0R)V+zTjMobL
z4rswQCAs^pNlym7pxz=4bnz=`Ekh7NGKV>?g#dGuvI%Sm&z0CE4H?2VZ9JxfwtBB?
z#z!BUVb1_t<@r!o3JY30w~b18+hHPQW82S1E2ofBT~*VG)BMlgJeh6l=s0I2t1s1R
zX^cZ3b`X4uTr$rMR`m-stHd0{NC^h!6NOsF>~yW}SDs%IC93t<mhIMr2rgk*xvVVp
z1}Yyi!|lp(sL(=pwNGb&XT!c5u5z;Z6a7o=0%`u3FUesv^XE)%mRed7Cg;(tq$?GK
z6nOEm2z_I9MPsNrQF;p{RUIWW_3Oe@L5xc{!*9>y^3b;mb=i%?1>Vjb07+^LJ4kxC
zaHTTNB$#L9uPVspq?ItIc+RD_+D)Xc`xnM#U+O=g5{t*fxA?yhx(}Zd3KzY?VWKvl
zwOiA93Z;ECGuk!dHhJC|hS&s0wqB)?@<_U__k?1Tyv%My7!c;N2y*j!Z%9sfI*0J)
z0}nZ{tE`CVkiEUJXX%40PagDb2CyX+Y6vkg%n{u270n3Vr=yo}s1k|M?bM$&m{&_F
z?Qt{JI75D!ZLxSA3(Ulm+M72exk6Tabf5?v>)|ZU96>_{+C%m6Js@1_x;0d;a86k3
zwg=NTgqQO+@8nh#Q-u3_?3+%dojG>Yb;>!Qg(#RIAgp+_A7uP2uc@g?-Co``ZisHB
zxw30`XV?GgkKe<LpHsLOV5+BI6yY4?-ZKDKqbqKujb1AUJ40udu4+A+li6}8sFy-B
zw)f@fUw4$dIFn*SnB7;=0+n8LBsv^8mh`H${pm@5N<m`{Grg<q_VOFSr}x|PK2OH5
z3%P-9nHbKkd>ga{wK~MvGZAVrpJ2SBnBARwPmMMhiwrhPw~RvWcX_FaNAgjJm{W(w
zrz{{uKpqKHYj8%GXu%tjpQnan-dX^9ay)6$Kq9LtTJo<XTLbp$^jqenHyJlI1kSCD
z_*(G3K9KtC{(~$$A}79wl-=J5Tuyee>l>81@dLp04nobnOr|A_V*44OY`D&7r~_UU
z$gy@3c69g>QoU&4p{!oGR1e9vcw@=-ty)ZW!uImvMWG?;Nf#vV@EPDaA~YyvInO0i
zrMjlREX>nhaKk9WGF{nQps6-hNrmduw@?<+JHWbBF~I@f|Kkk6@7wf%;?P^Kud}WB
z1oGmiZH%2YbGD{w%KD;2&#BVBKS$Mk^NVc2d&>k!@L?Unc%cMtJfzsqeZ1$E=J0rU
z2WZj|-Ez1ne&I_aB&@Mm{~}G|JKgifUeG&)(|L!m`qApR1!jA{G0D&^_X5{?kFlH=
ziJnp`2Eu<Vf^8bF7hH+DkbGWEmT3GAGi)T2Weg=S4ZSlGHW#kJb7XWr?VH=Tx9jT!
z_AU>md%Mbqzlxpi7P$UmNg0MJ5wP><m;iwkWVl~I9eUVbuvhKy*FwwgjDtG=7DY90
zl~iJ*=jZ2#7(dT&{}xc$^1x-ZiCion+=N=c>xBoNBAkN8cum)i0#q+*GjH6LS$cV!
z>cJfYK7OT*eSoPB7t@vOXu$uMhSUn2aCwFF(W@0rj64IZb*!Tzk)O{1^BIU;Cp60$
zV4LF77YKe^Ex;L#!U`lD7W4+m5yRZOXMkqlv8Eg*hP(tnRWWe5Z`pkYpn?D1N4v=p
zQz%uDGZGI75^E>fz)n_&L4>|#G;E0Cs1-L-xnpY^AI*Pge27C^)TZ~h9qnSO*4GS{
zmg%2Xv7<k7>N7NWSA_O?+mGwK{9MqCU|b-0A}Z%z=be7^+{tLI$mr7Wa?fgN#+_In
z({(vNg8L?atfsXpn_x~n?=t|sI<nx7#3W<It-raTIiY$BcEM(YzQnk}Zi%K2glFoJ
zi+4!I$S$j3VEd~A9rftWW#iaP7Ci6f4i*uSp3sJl7D$L8_ZGPNKras`o7AAH8lo0x
zxou2Xx$f5ohWJ1CxD~(=ctY1X$4R<`OFP^~kCC4A=1%~_53gTtkjOFcdTNpfZFfq#
zeLVsLa14_GNJ`v38LO4@3PU>+b_${T<l&jaer60In5eKxnEX9s3tkCqsO!flZ%uXd
z{9d7&w$O+;G)2X90{OJmNxB1nWy+m{q&*l~7~H;G1L3@%is|pJZ-}xz=ji|R{*T9v
z*^^yLYrraJQu<DZ9Ffy6;q+2^lS#CWuSUcVl*SN@m;6hM6aTD!PcLVc%wE<|?+&qk
zMC0zxuWlN6xk<7GcDWp15LW5MwiYMmT3_{E<*-E2nDay4t<^_-=L+1JLgdv1Mza%v
zvXD^n15>OV@p28~<c0N`7Ms-+#rJze9<W3>$85nIA{L0c&CAb;oSz%&7d001f={{H
z9rq;L&|KhX&@+eVAiw_D{tyi#iP8p%xfTZ|i04#K72jvXR-DEI!-r{{1}{tae!o|L
z-9K~^K}tbSYLlA<kmPhMvS2kE>$pt8L2y+9<-D&H+VEiL^<Hf9dzPq6ZAjz&n(yd`
zIMd14+@?r7+Ho$7pHjpcCGOB$H0zSRSdODiu)rK}=tpQTf)uo*n%UZtY;gS~(ffI5
z8h8B9@`)ZB4dCU4TMC`qK_}bf0S6WbJpu^lb7+eT+xNGGi9w#1B6DjLii7;#x@*p~
zC579y7+734{~oc&@6-DVp~K6a5RH-`crB>>fX5;jwPYc~ohJ@4m627f1nq-|!=#&v
z$zG1ITLw?Py{|wPYva#%oL33x7jcm2`~iVe)K!ek3yegUmBtt5!=E7!Pstg}k`tCa
zH>*>q=eL$BV(}y6YcgE6s$hB?EFuSbi=dJ<NunoY1!9rR6Ei(xqm>5<)`2Kz*GsNS
zR0?e3iJw-)sBSadumyL*&k<U&`*XA~iT=QE(EB4Wq4k>nl78m5rgOGyM+1!t#==eY
z*A(uAYh1e`&6Zgwx&G@4Ban_f2p80##Sa@H$lP*mazJ+=m*=ML#~eaHb5LnsK|9Un
zPW8ee?L^Xrp0^F%Ar{wF+r0?HRxKq$Ii85^Y)xPx{1R3pV)Y`h<5dVYH|xMrM}&E%
zT!l?>l45ykayKt0olFv1Bb}%lEz6^Io8X*JsF??DMQMFp^y0Q6<jcjV8PrAG|2_n-
z5EID!6rB%}?0xU)ZrSzXN6q{5b7@LFFiJTzt;3?;Cinb=Z_&;sNMJR+y1WSDZVN^`
z!bM7g?jx+`-uqn)DTatM5L5q5qecm0!XZ<Zgs6r71c9AM-$xqi3-paT&MXCaU1tE<
zqZbqHOzWy3xsQS=E89tSooxxnS_z@RLup9p8Q@MOGRGww>%p1;;~`n#(l)0Scj7R5
z?ZWpUUN&Qmj((*S<j3h-S`M#7vp-Mm;fk`pUcw$F|4Z*))r%l-$c}HqBp4eP-;`yL
z>x$`Y@0BO_x1PMGvK}_6AJ_Q#*E_M6WtA@N0j&$`QakHcYOSOeKy(Bn8~n--^psR9
zN^sh*7u+c#T1Bzrm`cZ#Lx0hgd%r&eys?~me_tK=xjG2>h6KktL}&?wE+n+=@Dh^z
zjKx_ay!8rfW?YWg3<S69Obf)O+C9xAy02VAUsXlk#do&rSAsDP3l=FwkdC?eo^Fo<
zodP@86B^5)&glEMY<l7ivQwV1t5+H^nO0DLn*U@uIb!N<MOss{&nkpB-)Qt^;=zP1
zO^aM>XvkD8N%Cy0h3Z$_+ts;%DJ~SOkanbYWw`Oiz7XHP$Q!zYx(Mq-3rPEYJphKQ
zX(5X->mp(m(FPt{)(b60kaXjs6tJR)v@X?#!(&sOmd-9X6HL?v&5iaF2DfDjqz2DW
zp#-#@y$TIH#BbPD1{KHd(7#|xN>G%=DZdk>Stuv-5qQn;X(Uc`ry6hAU;2&Or9kd>
zzPB{Ng&Ah3STPlU--ij3)q?|;lkZJmlpjwye<J`p$=n1V>{qQTBojP5ir0MT79oHu
z$ggQE!J8%twUHw5QnS%wN#6E*NYQR``bom!1P<1;I5+9IAQFKsaz$3f#Bf12xAvCp
z^n?bxq#5H4-iapxgg%dM?NjF16NQa{n6)I`(JYN}^GFsO=$)~?5)ZMpcaxRISpA&1
zn%UmvJF|BaW3vl>_Vhfodg-1?faBF$<|@289O4cshN+(-t|My!%${GByN-ElB08$L
z=g#G=P4k*<hfzGz5qN$2I*a?ppM8sf<05A;<j$eb4%%r31C0iF`(cZrN@l~UZg&sa
z4f{7);rdq4toIeq#d3C4kscJ|cf;Te;NCy!x>Oym7r)nL7>f@vXTJK$QS<uG+~BXs
zyg%6c1IkJBf;@%b`pj;chl7P|oBvhuD*N>1E8G{@Ig~BI2ET(vrBMukttx`IW?na>
z+Z<Oii*Uw@4`1dDDj8Y37;nYRb{x4QT9cl;)hx-B%Q_@5kD?~Y{JH#q3vPkYkcY__
za`Sf^&=zaOu?UP_y|Jmc25H>6K~aP7DT3k7ZNeZ?!ev%mA9>Kr+nhea%lRw1&92=&
zuTwl`ocS{;zUP(udqNWvh!yRQ<RZj$%yDQK;A&d6L~-Pd^KlEB5*|FgnKSvElVGLx
z;tj#=M8ljbg2AsA<Io;|bglU$AcDl|O;jc9nK#uC(n=xP*7y<YB>cS=?hy+e2ru#F
z*iDaxNzVj<axF6j8h^~~(wsv$exEFQnw?cSL~_m}`3`Q86kfD;47%R9{VErBB9kgV
z(^}1c7ai<53T@+$;0HbZGu86@OzD9}Aa1w+HX&dkxa5(IMGctXj1+cdpS5ef)MVlE
z_*#*ohz+N_ey_N;i>lapR2t$QdJOFbj;r@i0Zsa;qa*5#n~IM+(-gW3uc^$p0b5Y1
z*)d@W{a?vakjruts;CRq2z97ACCF;msB{LnKsS-AmUdXE4a`{bAZEgMNON@w8GgYM
z=m=TGIZWuhOgQS&h*F6Z_V8P>?a!>A^}543?{vgpnAkq`3D>GRVCJ%whR)$0-XUWU
zmDT4}9zjgw)e+zAV#&8)ck1M(eA?<Get`HOdW*6Vj!H}W^`1HN4IfIH<)qWDSDyI8
zCv9r)ybB8KT!l6Lp|cMZiC{PYzNznv!x`O;x2&WjL3)9=YEiK~Q-O}xbiSFiIewIO
zdL}d7rdhPMp|0<H2AB$Av~fulglGv!n|6dUMAv5mM1C+F{Xt5C7nU_n(N1ALVT&J!
z**%qk`aSkrIDN3F{qd13$Kdb8+<Er!xVMAY_lW_q76qM*f*$IJv;+s{yhe+%cFDKa
zH?g;(-_BTb+!Cf2CqF|CP-u+TTRa~VxBz~OQXWzy7`KzNvgWL(+SD2oJlA6g?`WZ{
zc<RNu&_+9o-tP@)3A^&91)!9Md=V_JC)J77oW=FFsR0&u=4W_rmAxML9@6G1o#m5$
zdO7=TbI!`CTi@));V$KV%59+`8>rf-#GBu@IwaNNqIn*`)-c<3!XM(_j=y(1QCe<V
z=%1R>AKJ3tCKI*4*Hl-%b;?Tzv!pyUJzai6H!_a-gEF}I8hnL*GUf8o=6GQKdXR1s
zmN5F<4+L~ef7Bqt*HBZY-(!dj3l9{|1j^rfsEoSZ>j$cUmDN)s9AS%gcYwTZvuYKH
zW`STE_xk({ucxwN#8(riKSKCYt<1ws#jZ+g1A4NIU&ZL~G6FN^7-5<u6^cPW|G65?
zB3W7I#qXBM-7vgK5>J2Hvh%S1vp9SAxoqGK;olZ}=AUcDvG3nZk_>U-6}Zk0I8{}@
zFl4w*gc(Opnb*aQ<t;RwmkJQ@|H?D@u~fi4aOma=yZjDD%jL)HdF@v-e`aKYz*|oo
z;x(v=rf!)R#u{Dm{0IFP3~~#i4)Uvrmn&YT@=6K$yVF*RJWlAnFvNCyI?JHB7{ur9
z50fJ}5@jJr#(0yeJP5F*+#rt-VU1g=H}K1`tNxT&{9R9`vDxA(hgGg4uxo`g$>EEy
zS?x;%FG*@>9~_D1ZVWDhuy&#p3CUko-YgG{cckZ5J516mR`7;nltadV$}STM+f2V=
z#I7#Y?w&d?MznUr36i>bxbN<+!=34%w>I}ipkk*V#q1_#TizP^T;pzY`7U<PNm3|;
z{t4CT#Xp+nh=r4q!SwJ6QHsxfHxFu;Jvp`RP!9fKWK=6wUsX$p$!S;Qyy%-GX#e8Z
zFUZA^+jOPwJJDBJI(vbF(lYs{)#zt+ziXlmP^}9IvufwXJgN#G(68QpMhTI*oQxSC
zEOkbplwo56nJL)cc-cH)Xr<Kq!D`OaoNH5Vk01Y~>VbWDT~+K&pVT_$N^LRB1(MTw
zXn&nxgsUFDN64HNcXwcecx}O0adQze9*UniJr2G{<~@7c@98MnZSJaFPUdJRUqkR~
zDY^mA!Jh;9W>(8xzjv2F91M89H~fR%FZ#ZmiTwShl4z95Kz;q`d%fD%594nV;~UNF
z2)qluDhtlhP-omx=NO8&4sVF5MmiR+iamU@T=Q02sB9~fF80_h+PvcG)TeXC6ql`A
z02xu)4GaNMRiRl3N2a5V{8(15bu&Uvn>|N4qj%nnT})aaxmQE}!#n!YM}}SOe(|?n
zu=Z%G&u2iu9l%QNqsT_NMX&(L(F4@tGcipFKXGT>OnlcDUrqI;GNSimG%xUmxnM<z
zx5F(c4?&%^MOe;$PboNoZDaP*sMFX%ob`fu#Fol~YnQHaK+CgpUB2aL${UEM?;BcD
zwub=4_6IRRj9To06nrP5xu)RHCpxrfAS?PtO-9C-w-~?ae@mXqfL2EDrp4W1{eT0n
zX>#3HEao!R5e<pH^5xwBgM)?oFa)n#Fb{u?=78#D+n&Ze8YC$dy78!JS$8dgxr^WF
z5^cGk%ZG;32)(?3JWqjEIEJ?aNJA1_+~Qq)w;n_;|MF-<>^7b*Z1X%qvBE}tR@=Z!
z=t1_jM?<qXOGi`MAtkJwY)7DEBcFJdp8;|eEJI0*r!`va1cgI)f_u*Gh0A!An93dc
zwd#l#GO#lJ$TW3b{bte^t3eh)$W-d3+v{%i97meew@$Ot7HT%vunEF)kv3TWtbme7
zYeCUn`EdqWjW$cw>7CD5XiAx$*g|7O>pE`kJ4C`O<zAzBHU?ZWFOKCKyV=02;m-dx
zO7s-@S=@h6f3P#<anldm(0=aSHUyL3To|I<>+NQvsarif^1vZdK(`KFkh8<{($2li
z@Z!vLw)|Df9#WQ9Ek!PNlwiU1<bm}3a>+J3_jzy(La#C$#BOhM+dGR8%{G+89`AH*
zw<X|QR!;KsP|u*(mRiw`gt>Vs)^KmAHljVq4RXz8%&H1`lTVy!%3N0Unau7cwg2{w
z%xe*?K#U+I+}t7aVDBETx_;cc<kPuz54Dq)<(9R_GL?pxYU5R|HY`n3@&D*o+BI`<
ztC>3oqa_?!<DmYTe7OFE%cFipnsf%~d`?v}0YcU-lZivuc>O-7(UAAc_>f;vKvFQy
zvZFsPwP;?Gp}T`8$Up8z6CsoaVyw|v_`OBdTKebEUq649o{Mb0nA}pf{d@v8JjbE|
zt<OxtdI?Cm(?5QtS{r(=He(a$>?-=s;9(9X&t64JF!3p43RA8?h7n;6%SuJMN(d-{
zbaoyd;UMuHbDOZs^=0mS2YT+>-Yb>2{ik-~_zzJ7T06E3d*I2`zS;3Zsqv-o3)pwd
zaBd^@jyLy8m8ABW+FmVIvA+zyA`B?iGb*nh(c*)(l+D<*`>?%|y)E=pI6yL@!1uYx
z`mbWvCsbSaES;;Y$dd><B}TgkR-~s%#`!;!z>ob3?3$rH>pdY-k3>0*Q}ix&#9q%G
zi|g-XWq|4tI2VEZ1cgr1HS1>|9iftVwqfy1V-XioIk9eMQ93!c)(?+vKaSUWY|A_o
zRsnC;uk22(jw+>VDGHIe!Wg3Da_i!Oka|Ln0M_rTSPj9M=$ETsUg%-`)AUVUqdN7C
zlXnIr7vaf$kvjYaB`nJ13#Zpv1oAaA&(`Eu&7^F#=O!`Vy<2wt{o1$0*p3BhoujD4
zYazF(jj)!TAR6fN2=nA-%=qD_blXDagW|*F$!SbQ;nTfKpE-1xsID2#&oACYa@Cbr
zHxc%T*Ob~D&8pCE{U)<q*>1$OSAS5E4N~S0*Sje>mWMkXIq}*GBi^!m<hu3Eu5*5i
z$7NIOrh6f`mvSe<;)gA9h%xhxJJFsbh8n*#G>?Dz>sgb)bL)_`BlzTZ+ti6mHak+A
z?MK439DWORU^dv42nVizi(<Gll%6myy;023sQ+Ybw_jW{=Nd0jb0PJ&qLWbhC!jQY
zWwkLhY|$$+byoH?ach4<>k`d~qmAT<7C*$MRe5Yo;1X|QUC3`E&VJ9bpErH3@xD+F
z|H*y8+J@wJnPWHuc(F1NlH1d|vR2j5uRVPkg>{!~x-#yoec~u7`-?^Bi!@b-w@FLg
zI-Knc;06u{GHMtozgYlr#=LqmtC+q%*4MIrP`voK&VIg`Gm;D0!xW~gy4LG!ebnJa
zLva7$FQSab3#UU&PK;c(sv-^2XSKaWJL#GEyiz`nc!I9<82~sPZjxjOC0M!1n$9`F
z%7kdO54hQy5=h5}-SGC#d}1lGpr1>0Qs%=|j>^Z@VjNEI<3GC1=iuxP6NoIU%GTRq
zmy-(If$Yzf80A>kuhY64tF5%LcQJ9?h1hl+UkRdX7S`cmM7WV|5)fEq3=H_nrzI%H
zt=_Qjuw`d+_|7}|*9PDlyg`q{g<qCTq^}op-}wsoiv_?{B@xDDdW-hI4d@QV1f8Z}
zt_&gI4B!nsy*Xn!$i0Uqar%(&)nMmN+u-|)es)}B`XJ=U?;?`^325LytJnJGK+^;=
z*Ct>$P88M{u4LDl%x6bm0sQ}S?DkJ~lJ6Spkz4E2$nTpRuYIZ#GnpzN*pT`n2bXiy
zY6#X2ug(BJ57eaEP^ypz{&>mIgN@H_CkwQA-DK|kloTU(XkXSGz-fAZw{89gz_Nsp
zM%J0JBsWQ%0S1_z8T!mV*?M#n(!s2JEuR*p_5wX>=IZivAd;`4uEj2BMidt$pml8Q
z<%acADBsG6N=7gwD^@Z7*INbP$6J7Su=VN!;qyYAFzIpjI!33oVBwXZv7JSi@B5@n
zBbWaA0f>E+5}NgZN_DIEF_WLFj&gbvis)mrt>)%5>UdRGMqTAM=D`v=k3_jM4z|i}
z^ueaJmw2_Gd<i!~vuu<J8eC2n7VmNoaJu%|LiN(e?(5G}rU8I>cM!I}1IbRPu^-y<
z{Eb^pUwDvKe=e#-NmKJhr#Gk8Wy~!a(a=5}_RhgH8RU)!0gf;tGfAY5CunYMub%;c
z4W88`(`{c<duQAZ+3=`6;e<4d-VayXC^UO_KMp17)r=m%78qf4bCI|r@>N(tzNv#I
z)VTt8u+wqMuVIA!tle%Q)><L|Xs!Bzo&LL5`5O8UwC=pAb<Ve9Ip|qWmzYh)r7gUz
zuGh-L?$Ax!x^CsH1$}**Dk>3c(xNPp^r#vO&YHK(FcZP5cD`Fz*(i-ZS<7Cl+Zg(_
z_<(Dv-&xAnV$8N)%e}g$Dm7~Tm8A}J1@G3`sxk+xCS<gvr$uj#w<<47y;5$msfj|3
zf_m;k-^9zi<$uNz+c7n*aCWQJ6A$v|ECZxYU_K$>LA|?8)N`&-CDg9b-S6y)eE9x*
zyR&N0-M5c}9iHtJ_<6{!&n4!hwobOZEbI>Ud}n<K#Z)c7+tW(RM!^X_yl;ISq69RQ
zt$C_&Q`V{U*=jN)PR|c+*AQ$K#N-`ch|v(oT^GT}vIwR#z(&6nMZ07+AW!NF_o_x^
zfc|MuPd0~6P24`n8lrktJS}A`pU&wHH`N%EBl1n{uOPS@{fY=6GkjlifeQi`Yb_X+
zHC%eP<763g-YJ0N{6wt3>cfd3mcT-2AiltKNGXn8k@#8i_CS@UtmYkI=Qf^-MvkAG
z-se{*EtDCnqMpET3lZ_zIXycbTCSU?=f7jNq0!@`n*Q#-q*FP9!sJ&ZKR$qB?Xj(D
z3E*DsHRrl4S{R__ifrlS_KC{3l$i&v|4bGylzC0n=WRFF3gVn2)U=8cP1T-O(EpB;
ztqe|5s_OZR^u0hl|28r9<ptFe*M-87!p+4d<S}S`|9W9)k#1UbN=dNSWYya6!Q^K<
z?qzso8Y#((-5WtTIs*)SD3ZixNVX$5?FpSWP+OOy2NP>#;rBkd-#;lPUj=T=@%QVP
zy#||dw(X8v`Ich9OC{tD=+1g~LFjU;hIyV#{@Bd(lB2mL&(fcV*Q9#LwD6yx4#c7I
z`rOHVm(v{d@5FV)!Epg4Mdr)xK#_+tB<6WWO%~`OC3t3K?}4#A*wvJe-t(~Ou$LOw
zpty|gEvBx@g*i6OG6#p{Lf?xFuFD8{=Xw@w7fM8J-z;@ECoGjt5d^008_SNhn$?Wu
z4f<t#Tu5K;2SaK+GoC~I<H|m$^38P}d~|fZbJUb_;y-vgH`d-NS7MfyI_^JoQdCrQ
zf*dZ=t1-1#I0G;Sw<VF3#;P`gJ$T#9Ca2rEQCV3Hz3$Q`W>6$Ug<-g6ZykL<zmSg&
znjUq@4W`_Hj$fu^Z%&qCZ-{k%U6mc&da3W?6r$(K{u&Shxp*?j<aD*~x=j-tw7X9n
zD0p3B^68u5SWL|lIam`~>6|;L11$_(ba%wFI7A~bhs(3Wws!Q-V;2Js@1N(te8I$f
z1@v*a+I6AawDsi-rL{=NI{9{Jb2oqQuPvu3|3e3b^?IpZN{Ixdz=9AJjP;_|&bw1@
z_sP;a>=4QIU3SeV&|9>lyUSd>92CL5JjV$2C?;slo~$1Z^{ESMcb<fWmr*MSK9aa5
z<}!EjIR`2C`!1hv@k;#LXF1Luerx#&+v~1i*+MnX0L$<G8>K}%HqeCJucaz2nKw{)
zy?6~vCoYS{TZ2!df2oIA>^*CkL_Kb}iFuRcO{vXv@2hm+*TbDS#KA75&Xl&tBLz$9
zh4_V<8M0TZAHD8%)vy(PU$1<yZ?c1vF9<1Ax$-JEdu#4vZd&R}!=Sw-rF%<39O+Uq
z!4>M@k!_ssp5|WQtY?sFBNd_R(#q@4arK&c8Dl(am5)DKu+hkSVWUK;Gosk@(n$KC
zhU(Yj>YBYD5*Ug2K%u+8k$qfOf6fanWLvG94jB|wmF$u0)AeLa)N2(-k-=+IC#cMz
zoY^Z#W_(%)f}N76AFG7q;>gj}PPAdAsN}ntvMYQaN>z#tQYFscZWbD^e$;cs^TJ5l
z{_Z>7;z3X5vZQzNH*X$rS4rQMGBk~tR4sb!_IQ!M-_;Eb+Sr~>kCtL=v)90lzx7fi
zY6i5a*~j>O^&$$Xg>Amj?CWi4dpyBA)ua%+_~M|n>G0;1k&l1$Qw2S51|zN*X%iKx
zELy9ekgNOQ6wem$pNzj*Ee>}r`_2G3<Njm$xd`|XrPa=G+^DY^L1G<k_6~^`TRj74
zuSNfBojLNW)DlJKK)XQ>RlKLaC@pk6-xWeR_D@Y&|6A>O^pg7js;kW!1$Q;p?i8rm
zig=7gGLi0%L%CgnZ^TCL2Ki<>HS{UbXHVl)C)$btCJMV>`)JiV7wEQR=BF<-8qCi{
zT67<wt1HKcw{79yJ^FsY8XqRaB6i^x=)-&CX8;qVT^L*oU&Ft`e?H*gxv<XWvFNew
zlhN6esSLbUgJc)(!=_l_LoH9|Zyqo~*m<3Vt@1?&*2ho}oQgxV5^)^H)fDlN3DhVm
z$ParvZiAhhBF>jq`Bn!-M9FGMe$A|gi47N2_?^_5@tpxsMVAZL(>25E+&roT1wHOe
zAB3-SC$Y}|h~6uM$vzQtifit$f03Mkm4){TKZH1tp%(DNUGVObRvZr5#>xa)pX<lX
zMeZZmv}2Fc?e;CDj&lU&kRet5lRnt^x)-u~VIKQM?dmyogTI3%<C-jgIq{ROj)54V
zf%DTqTBrv;HL-|XI&S?=(|!^(X|*iG`h0wPo&?5{j8mM|z{J6N;==DUz^++mQQ|W$
z6kp+qCR`BaJnL)OUJc>{xq|8DZFT11krpu{DH+C%@HS31Qc-sKY5jpFE8`+EWcK=i
zck8svGnpqdnv(-r)HtOhE3HdidyKCvc}<M>mnq4QdHD(wt4Jv?r#7K7513ptTZ-n~
zV32BtN)mlZta#s+8I3@voLTCst*X`BVT{76wbMh=hw;i)nAi{${7g>K<3YDYrKYU3
zCx=TviQa&F-$cf3u#3OLfw3+}hs!+mKhgXD>^sH*R-+$3T>IMf>?R;+fch))I=)~_
z6PJPA`9KNew#l(hf`PNuM!z$?52EG!Pb^a$D0LclG}ym7X>ej41faKkTEMw3Na32%
z+z(l%JC!3N?_F9k58ZBT&!o-nUow%0mIu5_MQ&^5W~0<lrv)et0#d9M!PuxiziP8>
zODM8(ukI|2p1x|fX;2z;(z#d@M>%`wfgOR0HqVG-S{y@9s-_Uz)5=dt8;kf)POTGy
z-zI*lU{ny~#B}21+Jo-$(KA4Vt^<hTrx=zWo-Q4rGTbG4XWA^P@v)7=H}9IR8_ZiK
z5{3^u2~zVM8{bHJ1G7)DUxo`*hC){#0Dc3XNG&U$rb2~_>?G!*ks8{I4WsH|ol#XE
zw{5g)nZA%k8|Ijy$~Z*moio5=h?n8eBluEY-Y4(6lQMo#6f(R6f_!gU-*W%6-inJH
zdKP|sb@KtX(F43-c!Nk$G#x^Q<y_5jT9K0mE3Us&<htXLYzCNz$M_83k*5kg>h&3)
ztK{9{Uf7HiYeq5SXd4bZ+{kt&OLRmnlM%`F8+LZwxY~Xm1;ZLEL$>_J9Z5nZde>|s
zua%3Z>f0~-RHFlU)Lz7v@HeO@Zlwjp1}UOc^%WWUJ-fKOsR42pNnton?&POk5Y+{!
zY9zYvqtF_ZGOV3rWi%gr3PDc2?U0pnvIB6o+B~{a7~Lb;uVi6r()H+ynXC0xnJF!Y
zo6Uni7GI^%I(6|j1kKt_ZBjG~mKaE-!*!p^{`MD=D8A-xR%7J=`I7ad>)tFFQf7ob
zNJ;Fz^iqZYxLGTt2^QUwcfrE-*O6U@L$aQ+kg*|s2T#T9nr4%ksslgaGML}W<+{<=
zK#sNalFYBWkt%^5bS!3q;WHB(eVWX^0o8?YXU)p6w?4Xq)w8u?bimJG9qR(F>35?a
z)hfJAy>x1Nc8x*v9*1dIG?TRr<ZF8g<cWc$;V3WIR-!SIJu6j|kjSR`*Dr|h4QmU~
z1<!O4r@b3Wvat>?Tj!Y?VKtmX<g~u~deiFRqiAf@6C(~EOO1z=qJxMxeqH-sl$a(i
zA)+es+{emOOS`J-+F1GU2ZN1o!p%eERcQn$S^n@6f+R)IF+V^Avxz@V&Q{_XV*L{q
zh&HIhA>!wkJvTU-mI$&(TU7_u<C5hBqi?g2@7?xz$*A#%jb&AtV`Ii3V;g$C^eq0&
zTbovIJ~_J3tZ-Y8&(Br_d?k30Zc}cgY)t`Kr#t;z(Z*Ig!b$pI<=+iF>ne|LKYtlK
zjK&th=qc^Qp5R!?T!)}Pt>E0|Lg4#WjntHEsAdAC6<p4`Z?G1z$$!qy-zMTP48Ck*
z<FZ;cWvmh`h-Ta(ftgWgl_<}GTAMfvSKrm`oH&W_@z>GK^upSI`E}55yovb=mBZVi
z;~MplkbsV{6KU(|L<6N<%(WSAmoNI^Lcf~)SH5iA`!{O=`d8u#v5&9lax_%eG=Py0
zkq>y}y$i1;8u&;lJsquF9kyeq6dRu>k9PzGL{%AkC~4lA(@i$W=&oRs?%B&lxlr{@
z(=gU#*7dug@nX#gTv;RUR|0$s@fXoKjhwU;n4zqXJbJ0{xprf=PVVo^Ko+Q(N1XD=
z0nPweup_14o7?GIRS6S}(W_G?a)lX2x8xumBO|?2-Dv|1A)%uzMP!K2pfjSHQkyVK
z<75rgGt$vqT<zAU!+gr+hU@s{1o!9ZDIH0#oC3K|KOfXGJ<inFk^nPZZ{ybVu|<-U
zyqZBL()Lu83dYP%N{_!kE%j`0U~&xaYdEfe63SkzoZ-7!tCTZs#BOS$#5tx8jj{G>
z>Gj-e7tpO*nB#@F-1Q9zESB-89Qht6;dIRow`m(DNqx~`E=GgVdx4b?V%eq~^kulz
zSLLI%WHqNun?m3db7FT{5?`xN?7mVjyw!rlN$E$|J`W^a-Gu*XM>mS5p)>_7htN?%
zZoxpZ`u(zz5nGQ2a6oX8jm=!PVyb*`eu1BF`lN(&@M=m`!|KE2dCx+x>~$^vVGSo)
z#C-Rfgj_x&ZE$3EN{kVcPnw`<jEcGVd5y8d8r%w{d@J6I8S-T{RkTA>b3db7;g$1J
z{+;i-PwnC=w<IMmGL1R>2kTPBNT++P3)DmwSZz!VYkU=`UMIO5U$eeHy4ElP%fJPg
z;Y#0Nr&yT?s1<_@gLueXUo|IYCUA(Kuc_AezBXrN=x%7dyS!KUtOO1mzPV3vY21dC
zk%f>XB~9H!M*XZB_A-%|AEMO7Fricd96wkWE?53dS^vJ1a&hJ*F8P(!ackd)t0SvE
zSTy|}9@}1XZx}afvfK1Sb7=5WW@2m(N{Ap&9<o*k!q-GBKUO@AJ_7`ZsZPZ=ZGg-P
z9u1viFquI=;HLX-HnRy(tS7?stuISf8iyRNt51zh=kb3aHXAUjNp4+k)iACJt?Xmg
z{;)tqzuP$2uj&jSab)MFT|l~mFNi=^<qtH%-3zSTtUvb^96qX~{hZ^PD{C3}!I_`>
zrm&i&&0;@w_hLU_8WOZ<raZ17&E$7so{pnkR3F&gcisBN?n>5%ZRdX?G4__>SP7cX
znW+u%QiPa+qr0K^SIlxuy?=Co-DcJCbWI?x(=MWob%B|)-Fr=K=HL@geMMJ3{nn0%
zM9Z&oeyX@|8=wM{kWcLJb28iPsy%SHWM7VZb`8_pw2z4!%Mrh0bwIz&y!>%~`rD7+
ze?9fUTu@yu`VW@MX#IWZ3}W`wcU*5d`s5&3!tY2nx=T~Zz$JOW4m96%8kgDdGE=&?
z0xv>|3B0<TmxqxlzBcfh{T@}MpJzb0BU)qB9T{Fvj-nVRdeMTm9%k+1=?23kPlptL
zmiR;4Q>qnXg^#$8KZXN}{=UKV{<eA}r-muT$rgLCpdD00%4wGUQL@P)D#pn}c}gdH
zVEwtT1DhR5A=D5(nmX6^eJ#dIE5Dp>cbh5L`qr1PPxz9X0Z9nA<;SO=pp3k4EL%HU
zF=o@RA~qu|l-W<eqI&Z^2r~^JmD$BSg}))5%fA8XAuDUM8VxTe55Iw5H-C-nY%X;N
zyiy!LnPkU61;-U_w$vXElAcssV_Xy0g5rAWt$Vq5br7;(>7UB_R2>2ATV}3s>L(SH
z+$54=cEe{~(!Qw{?Quu{juF=ox6@0iR{46mpHs;bX7GPK4gS8tipLnDiJB*<hJ%ff
z(Ph4f!v@Wi{ZV)KwSf&SA%YZ{4anG@ZxDpZ+c1_HUH*dKbT9n<a_5rmCl2tvOhBpU
z-h2rvFK{u^OG!85Hvax2$uc(DRQi_eC=QRA^FQvSY{a{v|0a*W<8oC|>x1`VbizDy
zeyE^;8}ag$&ex*t_iLt_$kwG{8NjfhnZxodJ4jJ`XPZTkING(MG68I3xA*qiyB_RR
z@RcXJ=VCt*E&Qu+U@JU#vP*qh?C%i`#c_li&ry0%FEq4w4_a04{1ZU>vMDad-*N5l
zTWqu!c=(>6vmj&LZ#0CuN{})e*^#i4DJ~{yJ4+A$>|}_Hxg_0|2%wf!N^UlFZ7)D(
zjE~z{CVlIX>M%6O^Qer^jK*jNSSLbiuv$$Ea$CeJ(5GFLW4c!hQRvzoG~juw_+nD7
zh&4VjPyfM7_3<yO_2z0k36A+wY3{){#cE=IzHB$W6Pvkmu5{4mPMF$DpXkx!n&YkA
z8-<aD77U*<W4h#IV;nwMQbKh#mf^lgJ`&_KbN`iFhp^GTdJ$1h>puB}>am${SnQY6
zTgXbO-*QvDWE1e=%QHY-@)<yXzT}!gM~d#}^jP$Ms-wxZhlRR^Ykk0z2O6_;hmN?N
z2q)J@ek7#m2PbDkK!j3VmMyV}hbGwawxc)kHbf3M0q5Cl1bd=icDL8dtDGKa+)NyY
z>p-0}hIv_Q;C66@eTI+9=L<*;6yKeBd?_o;k!S^V!=ZTFK^GL!94+V#IjBU*1^jlI
z31iK$(vT|}cfUV;Y%>YcYJm4@os41tNBXMX4Q|Q=Du~UU`;zPXh&^dUzO31TpC-sM
zi)r0qcQ+jthuov2tTb7V{T?FC5lha5%<g$KCHb1xW*md<7a_nOemt?!`Kl{Pdh$dF
zCfdY+62?;sdm}RCru`h~TseQv1-X=@@`Fh%nk6S=v<aKLD2`T$VCSq7?k2d-J%2;q
zz)ds#s>*qhwsEM!b7`_v=Mub|lpvpP5LbZP?Ot&{(gbntpN}SlbhbNGCWPQm5qRoW
zEwwjl+6%AvFj{k9`H2v4U1@ad;ZH&awoxt?c45Z81s(L~Vnt;@FnXrug%C`}{_PFE
zAT;e=vOj4v(C>QfH&9!i#4ehg?6ZN9S}x%wGqB@A)*)Msw*yNEJjVncl;q$_&GQqU
zTdN=*qzlfgy`P$-{ZNY!4~syi@=q4HvEN&q_*Q<A$ZlsMZ+JMc_)v5vd1S*SJ392%
zf6w?dI=0RUYB(+ORk@eYVW$Pc;t_DoP<_maj(7rlbBm&8voP@hd5hq+I0cHbXRjQ8
z{w00znVeAdI5mg%wHub&8zHQk9csR%hF0Gmrr!_VsIc6xAK6BRAR}T@?pFHyyH27G
zJ?9h(vZJMnu?$l0@pSR%vc^|fIp$Xp{p4ArPNHQ-%}=6+3t_5qdj>76dt<h~KC(~#
zr4c3O2i|oVGGC(QH*l5kTk}UNwxB4E=KtKcplCJvc#ce8cLum@LrGUGfKI&YOY7k@
zO_u8~9+Jgh0ast$aF8QW0!_(7r(e<AlxW_tZ^{8J6isPi1U5R%rfdd^TWvh_P%gwC
z@$C}n#lb26J|!P#Tj0HrjtJR;U4{+9q=%|XY}y9a(!pVh>P1GHylt4&WDX<mz@p+I
zc?_7%->Lew)7~4t4%L222!g45e&>XNKs*Dms6@h#)pBrYq|dx<U?L}!QoEOj*$SF@
zUah)Rw?yf+jH2Ai_<~G8;}#(i6dEZVRKou^-rh5+sr~C3#e<3piY?NK3IbA<UK5o=
zmm(+~BGQR~^iZQ9y$J|NjZ&l&>7CFK5Re*r3B3doY9Jxreg5}8@3`-Ko)6EsAMS??
z$G}b4<+s;wt-0o$%bMVIcEJy5U4^oBKq(pKXmdSc(Nbb}yAZssnVA4o6+80qrP$jF
zEyD+PrNhG_{Ij%!^2gUMMMy9N$r;8kT;R!1D+3IJ{W$*6JcWUgi(N!Uz{=spzt5};
z0;UcdhlbgZLcfvnT6IY(U2~Uz#OgK<U(89a#!BvWbG36z@kM@`(|b7czF-?3L;&M0
zh@su{SKR?}h1d2mXYOLO)`Yoos&>>ON@q{Ywa@uM<<sfdS}<?5<~dIv3`I;N2(vD*
zv`dkmyZ47?dGp^VW}ra`#SMlANKqiejh2hP;XK107v8G8%QQXO69C1t=<!;GwG^=L
zNpoM<wtixk`QKa!;Nc9TL;g1Co+r69D*mDA8QVP2VY2;<Y!@ZTQ3IPH05~3v15>pC
zPw(W``2!H~n9A-9w3_!ke41Qftj*lWHo#~AsPzDga9boC*#`ZtT>vc6J-<IRLiBI<
zC97waSf*Uo?dU5q{{4&qXYc>xD%ai7Z17V69tm8oAmn8%;+>!S;uQ_3L(#|z?!6_1
z#h!X-+ON4K*9GBUaX(9^-Z#kq`_%ruK8nV+XmNBAn8gc#*jZlf2v&o%+^CU_o%9<N
z8Q9CRmvm~K&Zr%@CieR`lP&Pj|IelP@B1_kuqCh*A;#bU0CS!|RzMf;4;3FG%e<SN
zC#ge!X{_u{HB|3*;j^U&=+rmY=_`8w{Q&iO+hNzb7htgfv^;<+gkw{=%~^kYFqc&I
z`yA^(Hr4bf>K4eJ?HAx#seTXArQ2!ka)r+n{j3CD4(BlN5f1|Xheo4nD&_4<F9pd6
zzGAKS|0_a8@X-QK<^O<GT>xnsbeUR$+)hDLGt3*VP?d=)RCOlmrGV#WTp^el7%Atr
zL1Mtr+#@1mJNmClQq8k^yAp9T-SIV}YQvkb%8kx8tsvL<oQsTgs$3S|1BSpis-Scf
zsVd0L`e;9GoB&x9tcZHkn`J)}uy23<kzqULZp=maej~&H2HNJ3EYa5z@Mo<$sT|@S
zsc#Pb{j#GC1(ljKIK(*D&*c@rfI$L=FgJjnt#_r!HN?deYa%E!!vm+FZ{(Prm{(r)
zL_RB6RZl+ZHi#xLk9plIK$c?II0Cidu*A^?c`%ln_I#g)Z5(Ins#RGdet6a=^1|Z+
z1-@shgy;V#6^3Gx1wDKuPxO*fd5cV{6FkL|MAQAV?8y<kJI$FH+BR8@X>AzBfV;(0
z%DMhTr2eR*qN0n!$5fsSR;?6<p*x0(d=`nlfVNHsZG;v8g@$mN{wQKaEv}CIq4_Hc
zN&7uR<%L;i_xOCajaJj50%NDniR^fE-X_tKO{7HE;^dLU8q<6^y1hUIGMl?$V9q=m
zI@|PH<a<S2zy}i`$!&&X?MfT(xTIKPb_~OTFJ(xDxq3cA2e!XKMr(;4s%2VRF3d*;
zFmzMyL)j_MEVPoE`Kv<>^ue>Ds<h{JWJ*K2myYo)i|kho+Bg}AXXWmWoTp}@E8S#s
z&-K<*1_L7{|5e70=KF1vFAG?U$V6DWw>2sAfjN82OVX{-Y-B<`<w`ZQW;`QjQQnQv
z-pzU4;&><fhuO0p1*HFt9?G32##U6dL8~^WC{HpGtQ29h@U-C9TW((RH`_~UN1%2C
z8Ju<~1VoIlIob_5C^1SS?Fuf%bq|I8get<^f`~45WfRc!2Y+a)d!T20#7#nzQVA;n
z`QJR#J7X-}_Mmsv8WVoIW6@Ca$(rXwd$#|jg|P^d$`Opf6G|97njI5|F#xScs|_g_
zV<y!^`g>fK{Xp(b=c!-sJoDLIu%p=7F>;q#Om6Dy4a<C<>=$M*WJVe+oOt+fAlhiC
z0Ny#Ri2-(UfuXzqXb?}jk8ICUsMiq*M=m72B8T}aezyu8YN)q{OG|gresZ?r$F~&H
zpYN&SS!_MusJ&6+j)xB;7ZaK<A&1yVNPt=11YlLH*+*KyHe^JS0@!{w(E!`;5w;Nf
z^ryE0MbxX6@stY*1QkMZ8GxO<M7Hz#0TDDy3fPi>$_xNAB+QGHX#R(X>LNBMaDQCU
z7d%A`zQB0d*ra=9b6CmI<-uWlywQ`-ajf|?uIG5aapm3cCc1S-lEq2r4jdJXFA?6;
zFcET?oBZ3L>_OZ889}Eji-}*{yIDYMFT0<m#eX&rZ%*CYb#&fvPuTGpEuTuvrn}s^
zFI`$M<LCwV$(c5LF$wGj|2lV-e_i3L*nE2^NSyHs<IRm``PdHERtFK7^5{V%B7X7N
zu9eZfE4>pxIrBIKM*Pk#3Bbb4>4zvPR?(p5Y^RD#>C0U{qPJ#3&f88^=Z3%H?cu9!
zfSEWnYo;Imh>23jvB_nd+B}5}Y(TX_LBRWwK~SWI&A0!dp*#S9qA@U`XZ}2NJ(>aq
z(nNCry^7ZFN?7WbQ~XUaD278GQb$HAXWziRZK#8G0WDJZOR7O?$ri8Wf3fcz<-a-m
z+RUk$uVD*37KKXJZnj?TR2i-irKd@cdwXwcL2J7e3eFrNYUGEV8(^y<1vu_`Zua<1
zpePU#lcKW7Erg!kzyRq8N#h?xwGjHRRZ^PQiUi4v5U*?=VV{mcupTy`yqfsSuj6hX
zIL7nGH+s-S^fUqrUp5Y9Uzx@y-@c-@M@)r^cO7rnb#<lRi`GJ{)aD(gUg39g7%UkQ
zpbK|%ACqwt8p{|;`Jdbh0~oL+gmZEb1jj#ZN#QT_Pgq_kX=<zyQ)SwH6!XBQWyd9=
zn=sfFP2W+k)lkYmm`AcymqS*g13Ry_UP#i6&>~jmGh3feV!l0hKgxD{CXpN<dvp>Z
zZM>I<e+d@xbuOvgp~S`gkIqGHQ9H8v7O^zneU|s7y#%AcfC)U!s3Zr`CA$7VX-<Z{
z6kb;Ha5VT;zm^(W=y{=TfwF1U8ysaUmLEtYv`9cw^tUs>6{X$;&|`;KlGSb?_&}Ko
z20&5~03_`y8nY{5_-~Ib$5_uw4l0q1yuWXZe=@WmhX@jtN=h<{-$*+y#=<bFOVPJ!
z@(aH`dk>7hX+b-nInUCO%cw2rkqoKg4~;Izvn&%<MYmC)B=|Y=PYe+VA?9^MLy^}S
zTXD{8>#~9c<8!+@;Wnq2j-8<I(GP)aCfrQZmds9hnlxvB_cJ{WrzHP`Z14{*uF;!N
z);%->br5{jkL$-qCjGGf-{kU{)}s`<p_Q5TeRKpaMxgCoiL6+M-sf3!INzj*SJJWL
z22sEWl9&CyZ_DNt?-k7lY-SCI%{@L|Dmj}df(|-3TA+2&1bNHvX`QY?A7MOPJlpI0
z#^08l3vs5i$9NYavf~7HnZg|O_lIVn%8c~*3?pyO-D^b7zp7(+Jl#>I#M;Rk#ebLP
z-1>!g7>?4$7kPdUh{E3%`UcAo@ol4XQ#GFwF3V?e;3hVo&PDWXB{AQ}+%LeD%CVR;
z`(0zr5VlENT@f&w+L^&9fWIB|g?!Vx=CRqUKR%=u;6?e}hu1rbAbv+yL-^L`7(Y<Y
z@;)BCN+8@ay}5M=8k`I0m(`A7hDKA2mKecxGIob}C!0_#JuB#nLLoMqDs2R`yDLht
z9VOz$kuEtZUU}Wo)1^GG3@y3p_Bmcg0qXM(k=uyfHsobL#>HtfM=%Z*?j$5$HO5^P
zp2QMyQsZ^;dczeJRy7KA7M)VkF=LccnHPDauJ^9|SP+MnoUsfr<Vnw%#{NA(TJhHB
z+?{lrgu#6ME1j_Z;(+rT1BHnI4yxt=x5+AeE*kczR8FQ+9|@geZAU(|pan_JL=hcg
zUduh4MhF=ihIc1_k+?8%E`&NG@Ag-1%Ng!gl;wHPQu>AFNTLcZv<-6=QM$V%Twv{1
zWGvJ--TrkW_<B6zd~4<yF_FOBs90k1s3T`%Fs`nyrw(3~Eu?@gD%&0@QI7xk=-?o}
zIb2A5-)u0)?)$S6cD_<7yzu?N6Sf!`>D#9#dLsZ^@%=v^dF>;BK#%|@hMoa6--Z?F
zlm(WFtoEX$!J?q}K_HF1i~ik9Ehpqr{{}&~0O0u`LS<E6=?@JTnCuEYec2Ylhn^Tr
z`>*f&&v#BQpeUaiV^BvjfCPF|A={uBo8Wir4^8|B;s~`{4tb_h>2R1j`_{G36X?G^
zzHr-sjyTdWl^hgs6Om>-v=h)>Fgfl0NqePe+i_E@>H}S%PegBtl5%>XhE*ClI)6CP
zr!1Fad6;}4*DJ6Fy90;a2;lG&FDFhTNbpw6s!Yv0WcBKB;lFdwMRPtdnB6DFsvl`H
z`>7Ik-%kxIz;(#t$5tnIezyvB3@;TJ;`HBK^p@^))9ZR>I<pnuqu^N4RPE?4$G^a@
zm^@hpvB5U@FTvM`{SD$qUYG)PB*lMm_cqDX-|P9<Ym{2e{`q5}FC7;ynyRw@6|BPj
zh~3T5oP&g4a0YEjp6aKR7T45pM|vbp+3C7obYZLK$AD*fs!uyXecotaNB!oE0gvDg
z2wX2<=Tt$#mje%KC!v>Y11YzP)p9pae+TJ=zgi7svoB@fZ@=5Kl70TX+rDLtkN)>u
zB3rwUN{EoWl3P)WxLjWNz2;c=xVpH&JbuRW=~e_e9BQ0zP9wv&VYBjZTci+M39Il-
zno{tW%Xk3lAOYIoAbD#v($b!FYSuw8TF^zJvF!3^54*|Vg72>d=pt6Xjt9sh!0o41
za~b}nmr4bfug}UF4{}THAVbs=8SK2NJOJTf?3y{!uB2)w@-&kdHA#Nx<_wpcuA3V&
zS6f0)2|Oi0JA~-lYZ&&X&6LDzurH=&G?ve?%0moQe}0DSc-lJ2bh@Wvy3iNxj4SND
zrhbR&ay}*VRj1_Ke{C!Z(uu#Ms|NOpp7@z>_)E?;R`%E|v&j*;wKh@nzLT%Zynjl4
zOQi~b=2N<kUoUo~>payCM|=2cO77yNSTzSHcJ=z)_i|h0Tw5+K#29eR6&U-e6Z&@p
zO*)4*hj>G8%Si1DbYxqFNG`TGH|MIbr(4Q^FjaZ*ec>iw=VOm;H(%D88m{%~%_H)u
zm#l+82w}Ht_X4RDw>qym{=7z#pfJ)ZkwZ`&;_Cy9N^)C(^pmxfq(^?bj=FhRVymGo
zI<+&clgTDQ7HToVXTP!#s8GHmDdQX$W)YJ<4mNGySLV0hcPw4g9(`<Dk??&u=2Su0
zrwP<zL_rXEz7;*qYeg4n04;9o5KBkVX$@hQGTw(PWrOUG`|S{YIP6CXZNA9_+^2pT
zrj_Yn+<w01=JtTp=V-&iHBuODkilUpd#T@MU&#<^lqlcJIbh%`AeZ3}S^@lG?wR{N
z8Ef?7PEB?2d6kJ?-iTKY13CUu?>(nO?=fJ)Q{wl`nbjN-FXu01QbRBZ7TTteQP(v<
zje8pJh8+!GJ&;zSJzr+7@%p9UOKUa#k0#nx4bx2}Yi>rb^RDa!tMOH$OHOmh=BL|n
z8{+>VDOJlGfm`OoQfrQirKUndcOO)aG%7hevUdAK{FEQ!ymFKN5x2FrQ)1P%NTNBI
zZW0lzPTYfIOKg%c6ZJ{aPI0Ezv|m1c3by9`3f0PhXIL%?xIWX?*l+|*?Uv4pEW0<B
zzQa@&FmLstB};n%y?R;+e$8?^ztnaS9pl*i0Kue86w1*&Z~PHr5ghZ}>9T?8ak~J+
zai{y(dYelDNLnthP`>c<in?o_giQvj5C~<(zXW=r_UIFyDTA{Zb0C6^hH9Y=(0pg#
zG0SQ8up=3S_CNc+?1z#;6`*oR2P{45k2-;PnsAnm*|7qsm(l`TQ~&;%ET*-9%7WYw
z9F8S!AY%ZGSYkaZg0gUC^-tRHU!Rbjfq9)8gz(pxRXS$7?@u=8br2G30bFNkfJh)W
zDmnA&h{czTIgda#mD#iH4g*w)&nr-s>=<9eNP>Ax7wFvqyy#V6{JvV|D@y{b1H#x!
zQ0c_5k*p1Nj0W@loBJP+T<?8+idC{C3LGU7oc`L%q$h>G#rTr-+caf2y2(7WD?x|M
zr2`QM-_qVJHr6%iB4dC=arxy>f*8rNZ+(g)Nns#^r+qt{!EJz&$17au1p<zyH)}5v
zKvw;DYt7lQ+5}#7#wQRkxV4B9d7&H!Y&C?x{@R-%4@XH~laPsMF4w+_U}bqJ2b{nl
zka8saI?F`vya7-}4_`yU9$aCeSfFNJp>vkAZNThc<}cUaz^tv2{3%Vl;mzS$6O84K
z17B8CLMz0aB({j*aG<gf)0^)QgG+;%DBP@px#?571IDK4{?+6Yw@;A8F0p?7G~-ui
zsq$RbUzMlcz~FKA#`QC^_1F6aGm#mx!V9H~t}RvqxaLm;bdWppg7NIM-8j-RaA#9v
zGP&Pd>N#m{@{LRSH9F2Qb#O=kbCBY=xJAyhQ+>3@gB5&E@vw^E@O7j2Z#?(DzOSW;
zdN0rtW2P%UYANJx>{HW}nLG`bGa*1SjjA4wxqWtDjY<kvxTSS8yXE0~y6G(yL}AP)
z(c-WnF3yXwS4=jq&niL^vp1RaB)=BjS!k0AekuA$*W~UeZS%yn)1z28jxmJFPs~jY
zu^LU%Z)m6{hEx=nRN+82SAHuOPQ3D%Do7M+S-Stv#k9e$AO!SeV-j+i(G<)BCnsT>
zx;QQ>j>Rm3(Yn*7rUgvXFn5P=`emo1*%339w<mT#(bKSITlp?3<q22*-BE3kB7VP!
z*#HAIkqW0+tD1S=4Q+{R*5+X?O>+ITujK3*`Ou^XUp>C!A<*MY<@2MjtNqmH0!b-Q
z<EqiOv_HqY-|uVfdV1Ury^2U3RH+_ypN9zDEhrO^k85N;r_5m}`G`h_S9(oi)Sx6Z
zS5`MISsz7JPhLkD7LINyr-q~#ped=kFx0~Tj0NfQ`9FdVVj%z_Kw=LdU(m_aO8}C&
z5ikln765XcdiB%5f4TDVC%1}dk6ouq`%t4!pGEMM>ecQ8($k(HzrX*`lu5P2PR&ZO
zCD?d`230E($9opu%nH2ac+V!M8WZ;68G#65a<n{DD(oJ~*##$1LfdKnV0ibzh;a{#
z@OhVfHJd>F=&YOFs{$9^W`^3+U;#IdPG}(Y3TZ6Y*q$o(+hW|JA`O2h!8`t`3!L|O
zF_XvWz7ri^yY%fI2#f$pa3P?8QmuPFG?}%kJ9-sEunOQw?o37RO1wzSoE0oxF;#AT
z<<)hA>zq-uuA0Oq>_^jKbut-55o&CzjTj!Df1+kno6)Ai`d3HA;{_VI)B{GKF0Ghv
z-wmF+Sd~q(ZFjLBPn{~?bgk+bnKcj0Mf$pMRifYjx@fw@%Z!!>fQrBX3f&0aQ4rhe
zC1vbUa8XE-+`*bLmb&znEMFtU!k&30?g=xJpk2G%L}2XLz4^4RQ3e+qc6wCvRjsnH
zC$C3T&?&;_q1`eUyPHuNg_(fv(w?SYGDAnb%^k0dG^1~a5qXmq$9uh8kl{Sxsy9W?
z-;=Tu`9iS@gcOa4_bD)GRgD~MsR{}Ty)kseJ6amzLL``9D(U={56t##`K!DVb8g*O
zB&uo`WMp_EOKrGNQl0I!raI^6Ip_STfgWVm?B1AverHXZnTDjd2;N>=7z~5b28bIY
zGKDl?5PG6i*a4WKC*Q&AX2zYV{+Vh2&rEHicX*M4@4m(t7K%4>k%E92f1~*hwVTSC
zC<rt#m(1?7HMxSXJ0#z^<5M-`^IWg<!YDzBD~_wO{H;JZ>)pH{m=)lnAr&nDAe)3z
zA9hEV8zPc*8pk{g1BL|7zOQbt;+|K;rwtOm(I78@6pcxY3nDitqC-@$LjZeCGa)AS
zLRE8yg?PM}u~6e1X;zsU@5uXnEuSQ*7%86j5ID(TQTvV`Jt>d)*PwPK0U9inOOT%e
z40Qb-mY4$#neUws4V`saN-Sc|zcc5#e4KI}FU{@k)I?BOq<#bRQEQ`A!I`-kZ6ci!
zA!su-Hi~<Un{0;P9zsb@8HiN<Zm8r`V_v6WIESvIviQ*xR9a7}1Y+s!`t6YgZ^=&C
z-rg?$-tTV7fp&zs<Mm(lv?aClw3M>$AF+|Zep9ummUjx(`8;H$FieAqlD*<(PBs^j
zMA}dN`yGYtyytc3xc<;cY#jmJq%tvp4nOSA-?+B0+r}7*0z+;f=5a{ihRd#X(HFzx
zHf5Kw<++0)+10g4;vthkZy0&FRoD#TC+}6jUIXAx82|<z2YaBm=Se9f^WG)xVcvaU
zV^I#Y4!729_(A@7gQ50~BtcmV0b>sM3BBa!i;En*4iC5*vi71m(WcqZS=bSe&S7ph
zo$kEfF0)|uGdkTZm1YkdiJfC2<pVlp^-g&;qRmxpwPcg(Vdq^Y?eg*M;W>~7-e^wT
zX7z86=r<=B)9wfwgBygdsogdL6+6dB?HG{<=WGNpRW<~*)pG(UT1cAs5fe`D7W13E
zs6b|;>`OG#TqXW4M6Lzi4-mnt45Wl-`=#Qd^OIhw{wrN54=a{u^_>cpaX%w^Y53Xb
zXt_RU5ky*?X(-~gSRMj41b<L7j{jOQ8rZ{CGWzahnzpFyLWRp?o31rsjXTm`8IY3_
zK>>Z>i$7(+gb=&-*i&LilW`q57{IhBLIm4(-uAg`BPOate#dntVjkifY?7k4`aSv5
z8Sa0|{sF=Q{79v2AbgJaJ^AM${ojGG6gyZy*fDs?PwuRs5$HKwhD97`Gx%4uW^$0F
zjs~)?=Ap{Trt=lQEE=3v24tOBHu<f~TxO2>1^O|I-yzqLL$3h-*lm3z#kE+4tVrQc
zeMF=@C-{Z=h;EI41XIEd(=vmU-hm0fVf2{w9FpQ68oFu>#kEAFYd7@yx8&|Q=DT~B
z#vL)FdK$c4>Ei$iKLr~a#C|(>CJk#R<_`^%K5`LG;-!Xi(Eg#h_z6jLKR5Qmv=Mo`
zkwyFi7tu3_D>I`0GdGIj(>T4vi%%icENIm%nt(y@J`QvkD^9vf;A)MIaizJzDUn6^
zTjNmR?by$2|M-)C@&nSlgj0Z@RBnSg_3+si7-;my^)q#Yt9O!15Wf25Sn)`72;Xm*
z{V~0~C?noreN$Cpy^0czi=})Pp<Jh=oN8Z1)?$dNzlc(y#f@f*{d$U{S^Je|>f<JI
zVSi|v*dtzrE^GRwk>#4pi1awA;6^dD5)b6s#H`_*>f~j}*I|XKiq~fgLT`lGc#|pc
zZDWCjcQTbfq>U#$=<6uL0Cmg0OH;V<cY>Kj#fIf<N${(}XSsMs%AuxzKkxA|#Ds+W
zhM2~!bue-Qb<tpypuAk}yg3K0mc}fD1H>aEh0iKFiilT9`7Yhs16rHoiQjDnt{2vY
z46ZMkGTRKeoGERQ!u2@3*oh~?*ZZ%Vp0TM)({<G!X?a1%LZIzoocFUzYyO8L9}9XP
zn--w0utKZZ-oA}P*y6f+!{sE?bXGmTel@8gjO3|#p9wTEHmR9;6IGFc3X_*S1xF;o
zZdU&2w5LVHzWAchz6g@f`;vXLKk7ydeAAc@jrvvA2Y&IgaTb&5x4V2|X}#=au};2J
zyRoN#X81x0w9N7JA6hT)p#u|~soI-c)Sfit#+U&93S~C|*hVQ2W2cQgVoA^ZESAuf
z6o@41HNiaLEZgIb$K5%2cavwAq@z$~adwlh0YxhB%mm#H*rGwn{zx|hU`R0`LJ;Xk
zMX$1276CaiO78_s5P-pbkYn`}pvTz;1hkqVfT*$Yhem&kxDR;D#8T&BFdOe<mC;dy
zJ}6-hb?Afv1J1wzANFtoT;@dqF?OQvG^$uEpjb)tIski1_%=NAW~C^|bm>=J&3c4v
zWkd_(&I=p%pX8wbuC0(5&@;aWhXOO0x1SFI3PaXI<C)d}05S<Ux(?tBCP<Ml2{j%5
zoO6SUhKsc$yNA+O`%9=VcU$&;#;8tnv2X{)#Uw5px6LBkg(xo&x|<^Okni2?MDx&V
z)#iTEQ{dGbJF}IXK3ME+g9H;=*&Wl7Sv4Vv<A#RKWF;%#FMeIj))%fgtK2!mi!k}-
zI<jr<oKBi3wmAbP0q9E_3lhVP3f{aC@=6kcQzje6W{vSb_r4sx?{)$`-Beqb?guo#
zunV((^1yC}VVnm^ReMgLEN1Q~zuEqJOjE{kR5}&0o*SC(;A>popx)a%<*`ZTN~-2h
zFYsl3<f6GR*HC?^ku_v?u<6%|vo?azQEc>edgp{|FR1?H!b{E=I1-1uCjBf30gs)p
z$B5hzv`f1<2IE^eH{n-wa^)yhfH9w3DnGq<faewITGKvCSkqmZ?AfIZy}%D+<x7>W
zQ%A%!+2vLt*T_ZhVN9X4yr<YfA8b_eD-v7j1nMTz%7GEoH@M)>s~oLAavb%G$Non8
zx#Os`<V-s0!RgyC4f<A$Bv_^>v*4*3n+6|9*O#}$;_Rjcb2Ci_lNdzP?r^`m?XUUk
zjrP!XO!(Zjv2E&4VIBf{GKV$FfbkY@YLp}!A~(*EGpBU;5=+08gXd+ObD(V1YHU%o
z57AFQmD=Q-HQlKDq~xwWSIol%WbZdaCdG`83kwl|Q`Sw2*qCtRJE{Ny8-{%{CM+1=
zh8Hfruq~z>V|2-E{nZD{p&y8)@!fU+;)*9J>SC*+H4OZ{VB^cdXc5RjH4z^!sT|Il
zP)Vb$t2W3}J9ni@z5HV{9Wrrz%WO6&M&WRGu7W{9_E9=SdT*n&+N<LDZM^1Wf#*m~
zJi=+KJwSkzw=gNL&4S45PWBXXd1n)EcxPTq@y@uIW3}S`np1zv8VfpmJ*z2Qg6^sM
z4!(hU*~9T*2f^H^-ch?(#8>2m(m&;{KLFlq1EUtu(tJfMCE?7L+6}#f+*Hgr_<@ZL
z7=Z3+AEE$A-RKWZ!L@HMgre`&ZHm88vT_1ADBZCTUWDGF$M+O396q4Cz1k;n&%V0G
z_+u8I`C{CxrEtKj2X9>IuS<8iKDo>J0C3O)=nDyABw&+#w|gc%{Yn;y_ZRM@2t(h@
zgZpk3$JKBADh<8&VboFe?lDLoli3KPa>BNNG{yyN-S&s3v%^Pm|BJ3@sBSUsL*~0N
zSA{^7xq$b6dJD7H_M9VFRazbj<+3t;^wV&|g#>tVJbX52d4qR1w$c<2aIe30I~<5y
zw{PYnwUX5w49r194s@tv0C-Y3J=rT2cJ$HJe75wYO(cz(p-fO+OH1m;T=VCw=7E}<
zTDjWJSJbrQx=qyF`sZ?{T<q(Lx(2O|J>|bxZ$4gj9X`fqkzS2#)a_w|N)A!Mnc}3P
z1=x+vqtjo__lMSq)`|G4<Oy87zH$iuZ<G0}iQ!>aXVTpO1@j&9k!KZ2d-0*XChmhZ
zP8ELxeHU_n)Z#K9_foG~pokMqr~i_YBPUOjOxL?~YlsT3mP5wO{iX6Cef#gjn`o@6
zw^w7TeXj3Al{8OOK+wGu;d^S9j>P&9eU4U^izp=!p}YLQ<UR-#9w3MCV3!!lZa`A_
zQ{X`KztMY~Subi+!uXL`Ak=Q$j{Sv*q$)#?mXQYq0PP4?hLZfRuLBl9R})|z7jTDY
z6TO&W+aVi}h=%j+<{8dCzwg@=ta$18wb);MdqJ5#I&^YCVxmshtxaaf|DieL94Z=z
z_cUJy5}+p};O|&wV5r$s5<dM-)iTx%K#cto*w}uQ20+|bmqbqY!KbX)ydlH!JbZzA
z2L4wnc_2!`xoA8X4ol2CmO}}XrWdb;FJOXBp|Oo}_VNMuelJvvy{~OZ-F*1|lJ!Ly
zv%~=1ZvJ!3OQ2jL>*hE($`~3cp?%AN%0#FxS)5obAFmvX)_o{HaxX+5k#mK^^pU(3
zV4DsUtosj=TJ1IW&|Vt$9{%odcl@3Tntje`pwKnosAS9DL{o3FsN~q8yBEvJ2D{_O
zOR5w3Xhoz8gI4BN3lumwf+?B|atz<k$MBK(a>asbKSUhx1&EWRx>!R$io>q06A>sK
zsZR+uM=RIDBJ$kv&7wE5*=ZiB{XB0_F=W7Sdt<T<UNO?&Br)REMBU@Jc~qQQpxGda
zZF0nAm!?t#oXndTX$a0Seo#VaC-_FOysgVwT!91Y!Bk=;_Id^VmKw+}?iP(#V!Lqb
zPpDpuLL=1cJGI}(N;V^CavFvxoZFn%s(;SxXx*%@I?4DvP-|WpU9?`1WWK^gry#^#
zNwxL&z!HRIvZ6kZN;H~aPhKTfjv%32JptVLuo%q6>i3U@i!ahWJ>u$K7wBoUD781|
zOz1AkC1tL)vu1S8PR}6(%*#z0{EhX={Wsq;3(n<tmud-Y$?_MyHZ$PKg*R*;9KZAt
zKlO398)Qu3=f9#;A5pyp^X@BzmsB?=Yl5MqrH(suH;B%y;2WgCSwETFDf{Cg^NY%R
z#nwsV*FUBIrNZAUb~d%?cfJ~n;%w%IDaZA8YYlbi;n_YqCB&cGJU5jSLsrQJfZ*bH
zF%mK-mGw!p-rl~HBhgG5_2dejiWQfpQ@4G*^D7Jmoax6#r2iZ6VCTX)rQwWLe~e0S
zyoODf2vx=_ENkDy?*$gjxjz0dPb}3_8QIA>OZKDnaX;C8TZz2Ip`uqFA-XeOHa~L`
zWH*4%DY%RDY%a?JIzbi7yG;MNoi>^ETX0VmgQJ``quJVXz#R=RtljvYjPbJUtwcr!
zFp+|ZOo|nR@ie;dxjD9QCyAA?{Nc~}=Rm3QJD(QS0~;CA4@$*aQ$GvNv-n*~e$&^*
z+KeuowLfwVF^HFO{G`HPEs6grzgFdcL&;fCNUiE&zxi%%rNg)}RR~kN8Y7t~zND2_
zm53JD%-A`oYK({ZPm{*yoUb{+t{k8kW=QBR13=!F>}W&1R85xGQ<9ba>nodS7T49*
zo^?7Iv@Vi)9dm^WA<dCHN02rqTS-2?_QoH7S151X6xF!&`WjmRn{z-@ux}@5P+KHX
zPP*UBcnT6;;=j8KDe&fa<z}eA8dcD2WKP{{E)`Mfb1tvZtZI-KIkIl98a<2+KTeV?
zH<~g<1m{MueA_b=_&eujXySld(4ah4iFga=71{{JvBh3RkW~Wb+pwMm3e5$`8+8>8
zpCHCe+Jc3mpNge!Wirx%HeVXNd^D2xR5$7n`#lD~0Y5pGF0_0&)=|G7RH0#SRK%U2
zHN7#0OxaV|0WToCyxr|~z7g2tmEAAxC<ae|+>d(Na{vAx8h7`tUGvbKfZ@z*Bl5I4
zlXlo)L_&wl2BGz0?DuI6@p3J$SC`(^(t7EkOl_LRBi8ah6e-OC<kFh<)w$G-rugx0
zv**3hh`sTg`6gr}xd6V?Fo$*4s2-qQ5c#kRVW->&e9jK9W}S{Rd5>v5laH4#uG<YH
zQ3Nnq5eFN3(X1}H99i+?Sk@-{H~a&Y#|e(7O#qTs^88t2_E9PjA{}#)6Dy0*GgbG^
zLmwgl{+Mwxk?v0QjO<pu$U~&HOQBv5FNn>~ocJ6Do#=F3n&+}~DD{3r=X23#7PPlZ
zk!nt@@Ybc=AWhQJ2CX9apVn-iPdR%Cq~RBUxh^Ty+QAs6d$haanb*?j58G{iRLBa)
z7W|v8T}pQp1;ANZPZywc8^(Eo@*tej=*6PjUM>jC)6A}MxE_1%(1Y2%zS?`{O>BXp
z0wK+3rHu+?Q@=N7sKX4VzsD*Cnq(TEf(}c+U49hK*~=AXOZ)b#qs)4HRQ}Ha54|G^
z@Ub;O`Kd4#^7YXm1GJvuxC0MhPfA0T{4aN!|8lAM=U;CMX3rZ{F?jY1cIt$Ff`%qj
zwZ)FS0ZMHoK&c(uwg&IU9VAoZ2kC&8&^uHy@}L&w22?snwk<#pHJ*gi)b4^YSI(dB
zE^%PJ6MHS3r_7BfQKqK0&h1gK-Y@n-^qh3Jxv%AwG5^ncRnwj@u{=hV>7JrV?&$#4
z!$N^}aeqq#0xGQeK1t%RNzt%k<lt$c{c+7mif~D!@<P|}+3NB|U2o6`g`R*vfYAT;
zQ!GRz7-BuH>Zw-6|J__CblxIcpN)%pbm>BmU^jQ>JDCk!Q}O6+-}&Bwd9lK3`ia)j
z&j-U(3Qna>;xh-md|3^HkN$siSqJV^uNSBNqR5G?NEqsfhXf!N088?)e(&E1@A#Hw
zstJ%kJpWXf+WhS-A!lyhKNWYp{R6o3HWjhIQq2lYtl*UAF7M^y`DfiwH-*2K0g92x
zH2a*I+EDfS_=j0<N~7+*VNqSauKe`&Ikz?l@NRx=Xq!^5IZMf3$4uQRn?>3>98r~4
zst>QO^mA(UFiBhYw4T>*@-wa<jFJb+z=?LPD&lz4H#1V6hVr<FhDpMM&dpfH#yytj
zDL3Ur%dRa()ZJuEQIb<5x#2XRQ$-zMIxOJboYn%k$yF1LVjS&PMs+>wQ-`u*PFhp>
zVp{LLeC>5EFq^*xZ1)0o4?)|VI{zLC(3)b$ebzeCivA`@g-dYUQX|HOaXBh)4oplN
z_tC0D4u9nzD;dZLKV#rb?6#xcC5p78!Z8<lhjU2n9C=|?69@ipj&XOsR%}e=xK5e&
z>=`%p^*BC41m&x~Ij4AWqw>A~?2C4JuPa>-?>s5i`TY7_%R+3v1OeKK2|ML1SwhB4
zsdPXisC)$H2>0;R!8T6L-x5WU7m~&*NtrD70{?<ciYLgYCAKHFMe#`&vEtVv<-yt1
zLDZ#6@TILxFe@g6he#Lh4n8+GwVqr)$E{C&dt*?1=+aAVarE72N>|&$lFL`dCO&jR
z0HYewO(g+`4&@FqCLE#8LTWNWX*jwtK@Z<E$XWc$`$l(bQ&p<x?!B{?Ij<_}X3E$$
z50C&`0I|2F^^%Yq&g(;?1ro^-QaC5l?C91Wz!Qh_|81<n9(}dActPGopYwL<_#h1!
z0Sq+wXQ%;<!64^dlHLL;e7;Hwh&PLm37#KtgFi@O?HFt7%<w`J8A8T8Ba>H)9B0eK
zqHOI3pTZQeadrTg+I{{ybo)60OHGDXiB8n+=p-B|k~jo|rx)oyv_fYIXN9ppRU#|{
zfLS>lM%r!Q3pC2oze_$)+6>S`Mr@OuA^;`DgaD?7gFhk(dQ7C3dejcgIj2W#iz2}n
zp=+6`;gt7i$E?U7<?>=wX6FT_vzyKg$c5{LdjYMOitQLf3Lk;QOa4(F`%}Vq_O@AY
z+=2+1mOAMAP>+!JlqSkAe}QUI^ph%vY<(DT+fRMGk>5||EsD|I_hPNTu>z~4Z^jd;
z<dyr4#eJN!<s8!{yY4%XsjSE{BXH)O`M>;_$npUbNdX{&aKc$aF0!qbNz5s6B?;$N
zkZ5n^$(?lXwkH7T)4XYFuKgEf8F`<S4QK=D0qh#e6Oh;r+YzBWw<^zh+!!z0&HC4`
z?RU4Y<)GBt2!kCWS4hLYVe$$%N3dDy{F%FtHO1g6So$RJaW|xD6;sWOfu8d~m0=R?
zgx?P2@_x3Swi31!i~sPs{=&ctgJ@dFlb%x6dBs3*B$(crn-sJd{p{@fmyPP?TZEs1
z8bz=Q%$L~vvp?_5J6%)%#A3jX?SAd*BVA12`Lb{nC9C!j8pv-vCE|PNvSNPJAo8Y~
z-+jwhx@${gv%bLtx#l11H`dIDka`-8(jz(r=iQ%aN|Tt&Wq6|XW&Va>`<iJK`bAu)
z)68a)r?Kz+j9Kf22;Nkzu4>WE?_#xO(zJ7cWs(7~x3nA2)^wSv=>9I2VB3nlwK+a)
zYW}1g^BJW2#E_#X$@0U8gw*{s?}fA@$Fub5ju>z&Y|2xuBrWKRq@Zi03k0$Ech(~>
zzeZtS11nc{1tYTTs|Sc4WdjY9HofCl$H%6+zSY$e-CI@n>vBPbj((}eXUBVxN&+UP
zQEDm7Capd5-f$J^+g&xDWM5G4HwD_;mNB8|Tt%r4=gZTGXg{b_gt%SR>RsWa(~tMp
z*S_BuYB)g41?i9iAi@fHLGFD<6n8&`ck@%SGTP-P(<hPTav3qVE7Stt7qJVtW}~AJ
zp5PD^p;PvA>acxi2krBE>X7db)D&`>8^d67oQXu$IC(GnzSQaA8~4M%e39@`cSTDB
zl{V+InJk>jD{wG7N=yjJ#1A_ik7(F$p)+F961z9-%|RnpqaGz%gDN9SExrxpkAV*c
zpErD>TQf&~qnJauN$YdBhj+xAwA-&P8dgnsuu1KktEKL!Ub-pp`C%2^)AY5>u@#?H
zwr*4ffs}ri45o1Hvn0Rlp2So)hKa52oJcM8Pd1P$VjN&}04xZw#IAZTiIASeqyqTy
zM-KV<?K6xm(>&Y>0!x<OtFeq)lV7dLPSD8>DUg|SbzgZP;2SX}Bek2my9W0VOA>Z1
zxLs-5V}fpXn>WF!e9R{DQn*TR(ICI4wIW9^M(;5hDbOW0cp|Tql_KWbGnwP9#y!qN
z)DW-9FC+cE*&o~!s`%yj>rq;>S$7l>c7x=rSThQ>u1Pi_eOUBo9e?0b(9RqDYig`C
z>w(-p?=)y?tAAPYcX!}LX|#7*n=C8>W8cq!2x;SRG6B<eyMT|PYwv0wHGS8+@sqL7
zC$XosuydJN{(%~=zNTV^oKjJ(>-KV1RdY{YvWl0BZuhIA+oQTp<Jcy$<Uvg$CDt!S
zDOJM#^?JK&d3qq@v#YXkFYgEIfS$C_yn6^=h<&?6y$aOaXJ*JrG3PVcT>oz{CdxmU
z%qfuam)p#rdZ?41Vd`q3RJCm;64UBn*FNL%<{uiI1ppCY)9;z0+F>85*GO`U69*y}
z5N_Ipxuh>VY|PJ6m>Hbns0Y#1k$C^DVt@y?^*=;Z*Jb_!>}Rjke=o4kGSu|XpEH>-
zde)FzSD$DZ^LRw?*Av<;a|>Yd;UTt@?W%@B(`cgtD55g@T8iaEsiZi;m=8N@t``^5
zLT)o$cZHAc3gQpx?S*y#ipSj3&<Lw*Q3{vjSx;E!HQ>mJuDM;U>X4a!|GQ(tmNg7`
zyZX-`j$em(&<HE5^j`YOf(Ek~=dzZ44gj&J&*Gi;6}B;qDmRpMlHxxv-UgUh$mGL<
zL!O|e*lUPF@7F`XGB8NJk;3((F~;+0x>c}cjOy0b^Kn2t7JCi$AgKhWz^l>t?#lGs
zyb?on<`@0fnh3MzNVp?WU~#3FG_N=+lIjiVbY4U;cGLGeuYez0xz{@W!0)_#uL6qm
z$e(l=qG*r4T_hyMI-F*VWo!^xd%W{ae@5I|d9rRv2X1(pkU19li(gr`^C1kQ0n7G0
z!9cmPWXv9F8f$j-WgFL-6l&s!obyruz~wqs6ydg5H6H9-2%9RNDp}U=jrL$=e)A51
zc!XI(?q7Ih@?EtwO`E6{Su|tPYFRX4KIr0jG+|j&r^c!(<+%Tnw%6&d6T8)t#J>JF
zM++B%!2(Zqp^_t?rmaQ@Zio01U$gg}jbl2?eIebwbal)4H${f5ZkMd>ZJB3xY<?|i
zs@M53S4~_e4$|S=aQswWFE1v4rC-3c8I}wH^7Zkdf+D}_c1|G%6GtHIL09eQYcu8F
zgWrSU6CGRiqoTWde1ku)P})63WeUeRMSn}ES;h33jyQA2%sR~q?JT+ecyz(~F6Z^Y
zOHLnIiBN_ueaL;H0iHJBKKTXAupe$$Hq$nponRf54RZSEq|d_6?qbOKkdRv%M80R|
zCCh_yv^?sEYM>13BWC2wpHTIF4Y;y&sXp*x{CuN@{bkwyiU-&2VZDIHa(ff3Zyb9$
zO5R>eQ$|{-+BG5plZ$uC8)9XQk(a?0tOO}x51P#Fl&xag_GBh6y2OKGWVp~JVmid*
z5b+Wt_lEf!8<IY0*YEV=PPxA%{5~(#fXe-!FO5M&YFpQcHYp{4!;8vBycjNVwp>3D
ziR05O&1Psnd-GKKK|{c<>jJxD@yeH1QEV;-+VxY|NvYH3YKa(wXC628Jnh}#1_DmQ
z`0?IT)6;>7PJ-sB!5qVHr`I8H1kVV5IfgAn=>EfKRRi84ns?eIjs^`ZZb}+;`);w^
z=`cXj!5?;o1laGFt55{j{fuzVBr`BmBk$2J3kgN#rZN$=Co0MQsVfJ_`p{m-hH57#
zEu$~T=$fK#j4k+t_jK?qCIIN|jE3suCv|<A*+Y|FV@_+)u_%(sADW{G=wcwia>X@6
zkc&W{V@hDtK|rp;glh%$zaK`7oyw+OM&kett-#Y^zk5U&Fo@LI#3X#DgnA#5hfj4(
ztPOmJz5n>SCEuGc$q?y8QY&{r$dJ;6YxaH~4!fNV-7W#L9V%dVuDvwt)phe%z0(%F
zN78;gl&m}&=Xg%g@>nirrxvs`=0?2&2u^rxiS*RP`BRtbLJ!LR5*=ah4~;P}$%+iQ
zGhE6w3{G#05E#VRaTSTQ>~dr<_OqgoB(QPIw!a3G+$z|2x?EyN=(Q*za4|OT+0m-K
z*+E0qU{e9q_OceXyl#+U2aLzST=bJC-YHlVa&A9e3v&<0-%r*i8=EqJtohVJwQ~8%
zp?Z_uD#j)K0If=tifV)|q=`2W?=*a0myDO}K<PEaOtLi5%Fv)3PHrtFQ}|z<eWpO<
zCEY<l>o(lrNT*_tI%<Nye$r4)nVej8o~M`vyak##HwRxfofUl)Lj*k|oFufrxOz?d
zx;@l6#1zol*sHq9KYh>>RW0>cvx#PH7y5&c@QpFzvU-J+`mn>5cs1_a+-YA3%-7eH
z7*O|0b9?6-Tmf{_FkQY?nlYGDdG@AVe}rrFt#b)l{N(!2WY2xUIMJA5cJ>>sYYIgI
z?U)<#50~isA{FR5P9<x+oKdp#T9^o&>W_C`{ExXZWzg{+c!+|?0<@E_a{gvFPJ|KI
z^<O!7RR2E6wy<@kOL0D0#|QQcFAMCeRW3Lsa3z>eL%tCrrg3!bwZM!9tx<U;v0vi$
zexjCOtI+MWUrEjnN4?n|)xK^?G6e#ETG-F^-yd^Xw0s7}qf7cO4bJISkd5(H>4`^D
zA(C$P;z`(sIedvw!2I{n98wUDHyE%rGnv@R<Q+PIZ^a%;P6bnR)P%Jzj(MKn-!MKt
z%}q+o>qq)}UC4-fxT@yi!k?pxnX}=g@l8F!EDAy{#UC#LdI2*K<?qY~&5k--f%ms8
zg74Js%k0T#H2AvuVo~<BP#WsRSWq~0@h^ygh09rPMo7E=;E~sql*<aTfK0KpfqS%9
z<_asxm<vB_uot#dM>b$#AwB-3c1<F0AQ<k_;W5doseTdPU2Ij#0U+P$A_KKKFtu77
z%-|p^i~-~vH&qJL8n$FQdsM0-n3Q5v=k0nv?VRU)9A)5G0uZ~CLk`2-Ghm&p!pVgL
z=OVg}bwXH{>At<RVQ||PX!V#UJwq=3fZPV21;0BQ58lSZ$V%=dY0pkBFDFm#?fjuJ
zqK?xHvcba%nXxW~K=B2_c9;XDTBK|0T#g4BdL7|zHhU>D;Z_Rkj_43`gg4>|4UMo%
z>XtX>wJKQ&wnS;N_>}#=6o1wIv=_RQ+-Pg07d++6wb)vEgZJ0zM$Vhls`=*BUCV7D
zlMWxPpa+#cnFY3aYHq9E^j`1cSUm5=9ffdtj_O;+d{C3WlB_x|DF5{P^ro!YVRdfa
z3~$QS+4LeoUG#C{=qHt0iG41F?Q^0KYP&nfVP6f`#IIZj@hq{kfd3rPNWs}#*f@RD
zNqplfyT;S}t)il41aDbg6(|Qsbd84Gz7(NA!+s7X^ku<r!9weC!snG=N9E<|i&&Fp
zlgX+1<lN8_*#XU>2hN?m8>2+zv=&tA$eNhyWHlpnaYdTzO!S&cldlTp5$w#=0O*Jd
z{h=W%AqM~5smY~0{6iCWLHqO!dUikVWbMY*7ZHLA;B1z%H~7E*Hki<gJamBp&F~ta
z84fy>0FgN+_w`kPF8TlaKW-!2FZ`hy>-77lzdjF4Ix`p@ec^fM|Knkx{dqjl;X7Xk
zjJEzm!;X-heo8{&FgMoUwZHQavc(D4dZ-DC4&gNn$3Wq)nd90rf&}HLi7)<%$7|?`
z*u}S~i{ii{ZDvg9s7FV#sT0i_>0XX=&&yB#uFUJ%dM~y8s%I)JG9e0;``@;X0L}W`
zqS9@>>;FLs6!QHP38e|O^N2K+3VJzXR3)sFlgHF4Hi^8%B`|7bplD9J0HY-#O@^r=
zFb7tJjKk%rqXMUqnZLPRSo8S{P2`>kcSpIpnHok^J*@3YHSn76;S9MG%wP6-v&P!7
zakf6Q2O8Zd2QQl5(1V|v8s~P+^K@za9QXF}pUpRM8Bl&C)31$2WGfR&3TL*At72l`
z_E*_p>Qt|zo|t|}J@FK|v|dDJ-Z-t_@GL62R8p!$3iJP}Dm<j<jq;UKo@<icF!Sml
z%k>RTiD|^F84tS)c>ads3|xvg+V<$u1|hC@?R5le`CP^J-YdWdslSVU<}8$s8e>n;
zP57!twA&c9FB{MPayDx<Gi}cV4)Z<oRBO`q9r1{~3fI=aLyqQ7d!MMAPOFzIw|Lz9
z{0ZLQ%<cDX<Ur_7?^0WRb9kva1Z{8O#^5F3vZ5lGW*Em7ShA5RHi7-L>3KDma*s52
zEeLl3OOtuAj{0*Xpxc(-0y28{F9U|>Pj<U#xPFg^cI}e%v)du(fy&H;d>bI5y`Aun
zB#YUmIv$ter@g0-#W2ls;}T_u*2<>Uk?Q1=s@%#AQjWXDUhiWxC%i|-#CfN{7h6Hr
z->9(tEh$_5O)92JK~`gy<zT|`UXp<71Iv8nZ9_5jl$NNamAkG_3%GzJi)S2DwnXHN
zKW)aNCb#@=puF;u|Eb{iGVH4^kZOoJ8xOh)<?o%Ve7xqB5gGlZ+kc`I-j1EWRQZSI
zeKS+C&dBzm35vg8Gc!}ip&Z6!)~a5w%6)T(ca@!XI$|Bsqf}cY;HChxg^O%y{aQBi
z<N2I1djiTz+KUDPU#`E0%5TVBV$z32iLkBI)wJx($gzxPJJM2TQgNxP5Me)+<W|%b
zG{xYnLQD=(qUO`yEy0_D$u~N*UW6drqBy_$3k*<RAf5C5-VvWvgtp5De9N`3{ynp4
zE|~aTKYFzGx?<g5Uk3-s_2~w5s{#Q(n`QQvLp1FCR9D69*}DJgl3-?goE_i8Df%qo
zlk9$KS2M)3=+cJWz~BIsfc)-mQQqs7Ei*g!awoH2n@X426nD5hXB$t|pO><q#K^hD
zh~Z7K!G15f2{F+~^cE}q?EIrJGkXK0^4i3d>SBg9K~)YJ?jW<P8TjUGI7fdD(ahwh
zXu`9_q0oG;Aee}Ky;*4;S%v}{+5Kh`DNqv7#q#aV6L;Hm(R5F)h=a${;V<IfVP~ED
z5a`i80>Tg*spol|RgHZmbqfC5i1Lh3{q4f`c=Lmy9g8xS`gk@{hGUmY$|dJyl&OA5
zC{9SA;|S_FbSS4_Jm{ibX*k0@v;&y+bKAysHz7hJ8|fB1&vN}PJJsLGOd7G5iWaIj
zx!-v&Du~63mf>Z*^3!h4z;9l|esqNt|2Xk(>d%ng0ekT?`4hRC$y)zlm7m(WEiOH4
zpn<YPOJ4lVaz}+z6`;xD8LHQ}K5%(Bp&=rEx_DHm`52q$EH+uW8^*~rE}{6{5RstT
zH2UH0<n(+ecS}t6>$k*%0*_r*gl}89B4-uaK2<*94^5#4`R9TtvYsOZdF2GD)0v2B
zlfLPE(YXDrE>|HpCir%|d5+CXeIW=1bnnM9`d4=;JPRN|yAPhHcoB;njFpqoX|s?{
zE;GFsmE~l}jSAMYf>C+LPaWOqoR@fhP>18ASL4S7w#KB&v3Z;BNUNs<za7_SE49@g
zG>nu`^&1p0ih#$<;B`Bu)MeHP6T9k)FKjx^dXDthZ|~}V;<J+LVt(h{DpB&4l%_(|
zEHYiMuZlrC<-s*ATob-RJ{E9E>|bnHoGzhn#he%KEAds+<YY_BtsUbIb6|bzC0IP(
z+y3j-ySljBNi71IW8)+#jUU4Se29`YqFHH$9Kt^BMGavtM(F;ln||?PIfE%}yH_5^
zv&rBJGa(h}meQrkV?w6RbEl3KlL;n*-<Z|iS5K#v4|G??d$FF)Tka(#gSKVB8!n+~
zR9Pf)V?)X(E1tENem(QGijI!t?r=@f@>NNC@FkLEODxA=sGs%Xi^9q36p6qz4p$bA
z=4a1tvKK&Q(;v6eO>doUx4y{@*Um613V??8EBM7Qw9_$g#a1}x`BN~U_3_MqAdj5?
z$I9{rSrgeKIgHtMC7`G|vVbXvw+de34-NOUI>6rSYeSxX+8Dj+<f8y^BGz7XQuIi2
z3xFot2a=XUn*KYqgY{J~#V1|gKydCw#*HD3d@i34Tz(~q3aw~hZ8_jM+X0SHCEfN&
zp1r5!sT1mLCRYr-hIA5N)(hKfpchZBD7;6P|GVQ>3?MTX^GN^ciK#=obE(qE4OJ{4
zoY@aL{FL?IOiuVeJ7DXeHK2y+Sq&J39zRBI)Hg+CpY1{7fySEA!WwGl)?q9)`4*rE
z$=?yaUlpkKJ<cLTLU>QL_0>y;c(rvJbOnSaz7EXlCo$4!pKm{vVD7*QDa#8ranx4-
zq9mJrw#;25@UUB%Zo1(AV(&eJnripGVH6dWCcOy}K~ZVad!iyu1XQG#s7Q@~^Z<b<
zNJkJ5P!NzVQX<l8gixhPmrz1Nlb%pQfROuH?)N?OJZHOS&i-)DoG<qW2Av5jt6b}U
z{mZWu04c^ik@bjG2<SGoRQ-mM+C1|V9n4Fp?-qFR{k!LBjnm-dP_lVs?X;fYRkPM^
z!+z=EO=TPlRcuIIrC{}oWPkQ2+to<B5e5DIwi;vRuv1n}_c%FTa2{K>0fkozAkR{~
z?hLvL=}CG|>Xs1vBc;0ePdsR&JJT9F%bIUg<T9qJQ9pq6cei8IEtg1@Epf4l<GLy8
zsYl1MSUMbb&3pP|sr!hDxXEYK{Tk%3Q426qv^tN60v*N{CCGeFFrZd51*BVsN+vs6
z|H=bH3ld^mysL8Ue}W$|V20*u<N)m6vq{D9feuxMSEj&e%?NQv*o<EI$|Gd6QM2eD
z43j!)7*_6ADhr5SgWyYo0qKONzoiqxwBiAo&(g6@`5Fz=XrSJPfdCG-1sDX3Pwv|R
z{0AUE4=?}C^wvrON&^}fH=q;xUhMNIbD&N{n<aN{-c)6jdywm;l2W!~t@X*5^iDW3
z#D>}eAR=ZvKo=lJ3|PL@YZCtukQ{>mx&U7h5a5C5bONfDKTcqc*7oqz+&Dbvb=#F#
zG?Qy{PwswV$K6%(sq(%K5rO!(3(W>uimXN{)gly~+)mP0Wc2aH!^2>ARvmmn^)|;E
z+Ce|HmW9V*Qx5UUG59(|PbQ3k#@`9(0B{>c06G8<0ig;K_RmZHczRcQBDJL+7fKVQ
z*ctO2geS~R+w3+yP;opvXwADmc$X<U|G0p8hF5L)iTT)LF|fl+zqVmB=f-w_r?BH4
z{GDMwS9T?3w{1oqz3HsJdQpZm?!>KFQKr6{UORq=yT@7fs0#6b@PtGbP?r)v{T806
zL}JXT=K$e}ZeWT4sA$FjDw;PyGElh4p#UMG?QoNC0s;Hs^V+JdV#axy_-zF_!BZV5
z5;yh?>lSG^nB!mIi3SH;;oaQ?F6|fqkc-CA(?H!My9eXm@lgRMYLrxwk>0Cavt#l*
zi8|(EdTf?r38mEh<U3#Gq%S0-Z30jApPzm1pLdH6g*TdrX$@eGe{DA=wTl#30n1hL
z2R9VMQ<fd-8mdnJ`obzHqMOSog16IJs2D#5JkvwbUum{QcxN9dKYX~o-YBdrw`ovb
zGVp%5mD<(02Vc6kZY6#n4BuxO0$*o#{}=ZG5Ig+myLD`((k%IZ;G<RZmVi2yQXA_b
zhdL0c0a+_xk^HgB>*B*9=K!S$aq=5V%x8*DTbfREWi3?<-(%J^=uT)1yKL3xnkGxC
zxbnR0RhcJ!(Z)Xz_hewyUatDPbNwGbtT8lMWcQKfUvv|4&?D6*U#?CnFp>CAi1vMf
zcF_J6C`EW*ZYPc(05u{BZze=6a-aIJ#43LM(sg0?gegNrupq^(WiOQ{fHmAZh<#-V
zj5zRlWx6S4Xt=0#vTnC5Jer!L;BrffElF<!g)Z7<k#SY5(h`tlH5|5U)HeW+jY?k*
zj7bjXx~Y!ZoBjrpQ#_?NKbICgiBZ+a3`rAgPO>`{o!~EC(D^LpnFwt*mboFQJbE}M
zwW^3ZfbM&CzI^Qo9NGdFBx+Sw)6V~pZ#q8XfD^T}`QY2NrpnHC<MKx!g?P1r&~Cnv
z-g+QkzxBrKVGx+ZTV|L&c^aTUYB}$wQMZcv_wySX8g@6N9T~4oSny?DAoj--))gw(
zz4Xe*D#?%fV4ti+N$jfZai<6H_nc^FiF*-lIdwyof^v6jqem=s6f!<|s9J?f37>q(
z#{0y|{cfU18n#TS4acX>IJ-V)L2^s5yS5v)F_`}s9r{tqkvKwL=3N#UA^R3Zw)3Ni
zs3zv>?<HH>y7iEx2Z(d6X0{ih=M3;c+^+t*;-B>jTgPw2y}D!Ow+cM8pJWgghebcB
zp$%_=d&sz}h)pryP3d4I*XjPLIrZ^mENxg-I_g8`Y!)M>p>xa8$(8>0d&0n>A^wKY
zF52$#;cfUZ=qLJYW!4!h^i(6st+F)BH=OP%G#gdhO4Q+8<OYPh-kOwdRtZRrB^l1>
z30=CoDUloa@Bwn#fOgf2q;cxKME40-(YW%CT#4T2bi0?olpvp?V}q|?;HWvtvQYY<
zL(mzg9w4Mn-Svw<T&y>crkz<OHtVswbKAmm30$oilT^JjgNqwv^TzUcyR<`f(5rXL
z6LTK-A}Ge`RDT}Az%>-7Ce3=#eKddH3-WCn_acq_p1?Xed@ogh#8u%+HpPwjot#wi
zIg$B+?EUDZaLm<~r4s?;lcVia2nA@ug;}My3b770s>yZEcY8d3n$UOgsUYftz|W`j
zAx1ZC@{yVg{s`F;=JO|2RIYj4Zx{k}4G7{U$pICm3XWpWfz)I~?jH#He>ooNhn}Fi
z&c%l#Pa#g?nT=-SUvKz;SbuTz8}yF(9nfovezDTwsI;$AcoO#J&7IKLr!2`>Wa}Eq
z?C2ai5=0TeZ_+Hs_H-n(?8cl>mw|n6f`B>@MxKKZl`uGJRW6OkF&h0Vj0!MrzTsGr
zTVpXFE7k(~?ho&1sOb)$O1CS-PB?$(Wz>h8Du)v+ns3N-KQTI=7zFEs;C<o6v`A<h
zE0vkJ2S}~~JLW$fJ%3|nA+Xy*a3f;YotfgWVpR%I{cV%Hu3XeBd6pZ76BE1ikVDU7
z7Q$=;V=&9Y)ZfJxKX=b7k1ABRJ9=z0t+5*V$1(pr;|j9(dXK%Gk+bE7b^qqbjtW&Q
zd)mo4#~{~v>(}6?+rxQPuS-XDZit9D(J$YF!dWq@1*wh&ws1eahT&%&k|j&!QTWiK
zHt{g($>&I7gSrVI?;=fLpq-)w{zaEk`Omj=_<6yeF4vv%iRL4}jQ|ByxOX5|y_M_b
zj%l|!b2o1Br}yPu8caa+i}&(AF}zSNHa&37&&=X_#r*~InDf@@b-P#zJzjxZC+Uyf
zm=l!w&75PbXov3HqFq4Fx<{b*VLDW)0WRbqBjqJvE#55j7u`bXgZozJ7F66OW_|^q
zBI3f`^6Lkwko;vO*$~mRj_=AgTb`aKHx-2PiHEJ&u>u82O^l@mEk`D%A^mh-YFw7u
zRY>V`+<$#(BY~>E3t!8F$7+-ya_GQ{e{;=L)yw=|G+gul##an_a|gv+{HbeG^2cp%
zrgIAD2qkVj9rupf5!60}j{DdZOzSgM+~k@;`T3k}&=gJB+$xM07JMNZ`{AM9FXs~*
zzpD+4^xw!g5U|0uXankny)<Fu6p*nRDvpNcP?GdWm(W>xE-99IyaVa>&y%DV^P&&p
zt$%nAa5^jT+g21>x?t_U9fzs?;&!s93!xV|&T^U|VdN}vlM_I!oIn^ncbg+x+k(*Z
z`M8J;hc8Y8pG%%`0J$XUC1HcTu-*HUw#oF>2a7cgy;(|92dD+Ak+Ed&XZvI^<rY0X
zscJ0EsBU1;A{F-q7wLVz=4G=I1*OmU-3p$_>k}DqDDwOk!&3ATs+l`@Xok64A94b5
zs|s2#v0*>#4q6cvx<_RW7LFwrs^t2G6&Zfs^+4{+>H~8aaD6JrWBYvbI}mYH3yvrO
zXz+g<WF%%X61Nz)O8sl@ya>uGjxZMM-hQ)EBKP4O>wU}SOT_TdIeo6Lz`(2!R_!v`
z-kk3&*!H-|w^l^q8xCPhLWXXsyAxp@Ga%rVu<$FiEGu#bH5i=ffSgVGU7amqSlRXG
zF@!?_@i)Ge`~|CGrnKelTiZLRxX$M|jz>>FKY6|zgsVh}qSt>-OQzaJ6zj<Kz@#jF
zL{K_~vugoD>DkLzsZ^o#!ew%`s<+F1iS=xAyotsQExKdt(S5`v%qAmTb!Bn&25~Gt
zdukeqN;mitJ-x65Q33Y-rL^(kU2y};2{(zd?8X}R2Fb>kKaNf-7b|SBlsb9c>@MCY
z+uhyV)(GRh(|4($Q-i=75zI+)$aW7ZgWKthxdXGVk*O7agmE`F3jlr-^;)*9Pl<nA
z>RjQB4SU^0&XQeA8CtxQC<fj5di_HDlWPlHQcb2gAF>*%8v=YmqCUPjuZ%(3d5RGR
z_2|GXT;5e?XZvaeOapx7bxj&wY7aKof%sW~xBrw^v;r;y%t6vFaxa2`c?PS3v~iK{
z(v=j4G<BrUgdXT+S#&hI)NF~IIk@fD(A=<MR1sF^ZxZcishAbyZ|CP>=j!42sF$47
zbG9;Nt?Pd5Gi{*zo)`O+UDf3Kl`=mu@ylfSuR2k*3EY-z@O6M0204-lL2HU{5B#aV
zGxmE*(G^m76r4p%(4}@pqUV8<#|^0D?z{sKjE8*NK%xH+THCST)VB0Lg>{GqJeGjq
zFDNpL(+11~&uX{yi&Quavg~{?C@mirmcy@|;qtx5D~KtYU}XWHh;_a2VO;BvKxcKi
z#H(`S-wYhX^_K_ASLMx;{y9(p6u5s1pjtmH$Qlks*k9P7SU-=+pOaRde&T(e3Aa99
zPM{su(l0t}7+r4c`D5>kr`TiywQ3FziT*D-dW6PbbUl`~t5f;C)fzlBHdmg*Gu}pP
zF|2F$_i_ynWrIG@O}=28b34YjJf7y;fG-tk=gt|o7%Hg>w0CV4SoMB?Y!~}2cKpMC
zeZs~)jF#Wqy+(}QD+LYc4|{7UX%A~>{eCGF&z|P4{vqK(7w|foZfV(q8^=TiIh?nm
z7;C;zFIMvp@D{JvQZZAuH5a<M!tC*HY+pmq3)a0WuRbgms1idJEe|d$USBbGk(sb6
z$d-q3*1*`Ko@$AHu9y>L`fwcHUGfvbfKLx{T9pB|%NR2W^OZAp=W4rG5Pc05lyQbX
zZvDU5yOek}hAtWAxjP&p;{KvDCmKZ{v;vtIi$6LCWXy}OS=(B4$|vmYTf9x6m?RT7
z606NpVEGd<YTZJxbPu;o>y0I;{#uNjke+(x?jlf8{+Ao{U=L%By)QVR`p!nz&XKhK
z*Fg#4^5K&8HHmDSax2TW@=o5>FcZ3CJD<zpR$3vg9UU4MTh+wRaKrTNt8WT6#w4Qn
z=S~Bh@kg;~+o*rBCjLl$oVPGAjy|;--+*UW#Vn7{Rx1yFPU?S}%++C@Yb|g}Zr&=M
zyC5_x<YXXV6t_smOSq;GDA%U%eEAgkqPXgI1mB}$7aWxtxE7)R#??86Emaq~KL*Te
z0?erXaHv+r(U$dJbh*pK!p(aNqWIvgOidq#e1BR65#5%OsY+t0Dfmf&wvAyinio+~
zcSB&ZQ8hiUQ&f({t2_c*b87^oL;mZp9ap%kaV<)+v@C1<{kT4l?b{Eiz+yFA_rZ=f
z3M2iBNHJ<T$V@236s%~Pj5v<CJ$qaYdi<(OXXDB1?Zc|GG700C8<-hd52V9fXK=TN
zB;*7AT>H4fzTPMa(!Fs^Vd>wv^0BwShVl}lCpE8)uA08hS3Fwso7_-KheR}U5AWK2
z#!_H9)uT#ohtCw*L?@g4!ltu2>d$TgP>erb>OUSJOt<6Q&{dJEck&G!CN&(Q|GmF*
z*@#^`6spkjSNbz!1O0hRZB7DNIP4F#pWl7J|D3RoUnPmP3&~OnbAMq^qFY&cej!N@
zeyP5~#uW$U1f|U|8dwf(IC!w+ibm``2Ok$<e=gJSR4<riHx20ed|UqSQwYj_cP-J7
zZa&Ex64k5>@Tr1<RvSjQIvKpf=>7E$WIc@?m+ySm>(lz_YHI9flR_=8hl%=qKR2$R
zM9PnHC8*}U6oD37S<>PL`<%|8JvYuY#UNX!j_b!yN+3Co^bs_HWTrx9-+?x);l-BQ
zlR^dJk1UMJQFr)kr2<=Ls-c;M3C615hOr17-hd`<)V8sr7W3{I-oG&ULf9!wJ>F|C
zHw^55(mK&x2>o!9Uq|)u1$Xpgyrht_3~jyb6`H1IqnH3pV0S6{=Le@K$?7-fR@;8t
zRqYlnsR8Wk`$j)P#D9bX9qWaK+f5wEjCk3?bEthUJ+x`+La)DB6vPP^p2;CmLz5fA
zn#{I4UafxvGAKWFVa%}=fRNczpBxo~St&6X8V5-gkT$`XE7dR15+s{aH`gpgCPi5E
z>^oS%#MQ~Q%~J?IA`j*ZU<V49yY<5+z~FIVJ7?iktxelul1*~J&u1r@B#!c=7RqQY
zlUBRxcWoV`I)=Rq?Ri|2nqO;7Zmn@e%%S$#)iwX3d$byV#5r1L(JVS_Kc=qs`RA?g
z6*iVV$K_vzF?BuUcmeM&tB<OcZ#d7H_FA`Q8l;9`BAJDMq}byF#Rnp=V}*Y6r*Zcp
zFb3+f>`RN=)%X_!P?sI|^YFtqOHj1futXU`8t=+KhE+1qZySD7V>t2scFe`cW^~+Q
zsGBy@t<nrLa0YyHDE{b{`Q@fBq2E*V9UNja+Nk+2l$UC`I$fZQR3m$mNL#S*&=cZJ
zF-zr_V~UN53hfao>1R8*be}>3-HbB}o8VvaZR7Yqm>$@E?{gk69-D5S`&9^KhVqyi
zxd)8<8O>Nhwen|bcADa>Iv{KU1f_35Ae)A`nK2&{^9P<&GX_NNXyR4xXIrnBzh&7T
z!D@zsrEFo`GZ#q@xxd^b9_m!TCp5knSXcd!5MqJf4=J$_xOpK(L&iKM>x9B2GVD6#
z2{7|@y{UeqzSyWPMJdw8zwMD2iy7rc2fSaU)CwA^q$)Yq^l`buss_^+JodD9Xeqg^
z-n+kquQYr}QPHb1p)xjJ^IV*yo<S)AMU!+zif_70rAp;}kzx;l>D`&GvwFS0<Q`57
zdiFd8T_reO!f#vp$z_JEvAjO*Qr)2UguQB5iRX3e2#E*U1{06C0?rHV?&QPCS+nTq
zGar}1eCf7>&nk;AhhJ4FroTRNxh94eDpA(vaQ2w%QW2pkyy45WdLwK#dD_ah3A*pA
zzSfSPjQ^tNUeTPPD(<s!81}Ea@_#y0+fM(ZQ4jX~-|}<(qf@wmtm27(7A#1cvh>Xk
zax8+Dc4BJRg=>aNW>)+!xf(j0Wz)$~5{tVQ2;!q!k_sSrHVPj>nB!;?#bu4jyB5F6
zJAWyXe13d8UXWPkm9bP-M$fXqqALaa)VZ0Ydnl1A-!l6pj%&rzdO%;1HvfLP(oD>b
z2=KK?!7RlD+({~=CO9U~LceJ0Bj7OGH^~RljQ&-Ed}ISL#ZZ<2J6?6`&P}WqqCP0;
zV39W*c^L&_z=L6?13b2z<H_?t%t@u)^&KSyFy4{Rm$tfefd1A=5^HlmOHo!LZ$|Ev
zUP`4%ON+c77}V<q`VN2zju7LhROWCi`Cu;&^>+66ZkD;cy`1>Dkf^XT5Q?-j3dy9u
za^Q%`;I>Fj02i{MlOQwGDxR#hs7K|)>)s9K>m}>imDRD2g+s1K=Kf?k88pV)8e8x>
zQ>t+VB7QSyW}9ks5yfujf@G12$B7*T>$GvJrglR3?PzT1kl|(SzCHIwtIzF@HN!r&
zy;;fLijWHF&H?Vnjano%OfXYe9Sq6%YdwHM+kF50VbLBRs{9kGzJ=({`sms6qgkB4
zE%Lw$wxJVmCfjL7yX1^$!W~`i?D2&Vjp9$PC?f^n4IP<a2cVL4{|u96E+U&}pMDtV
zhI&ARuGJ9Z@f#gKnguDrLr%2y-fDdh<pX`Gr6btLC)Afh-oc29ytj4};?i|a{&8$c
ziz%pv(WA9z9wknnqS-hfU5lmDIH?+aE#`M4?Z4$`5n)5@Yz3XWC<EA2W0fK~W07tB
zt>Pizu#q`pSz5Fo?6G`i>^=qnaD)U=`A6J@HgyJfrBAsq1B%R<{G*Ol9y<?uLy49P
z?r1wur0w}0xb$2~v4sBKvqr0e_D`E+=Z-PZy{g?luJ)Lg$1*EFzsr0=H*rY=ig}(n
zo7#rp*QhQ^%3%w#vM^MRj6Ma|?<51Pa7lHIVujfDpgU8sZe&mxV1@{CqlzSx1~5kF
z;R<G(GuF<M9uD9ntvASH(4avmYb^0FoRo?*BfB?>j?Zp_zU?kM@`VY0o<{nM(Tqr7
z0ph?qGjXzgQv3a!!KHck2CiO!e)i^w-KGS2ZHLz(jsVoZ7Iu%5y)T4l+(fwm;<D%n
zBpU)la6jS1KY|(~Y#K!s97HK@F#Ljb`WR8fP_3hhdep041Z2dxgr9#V|JY)z+wSDm
zH+DatH1qJ`g}BY&64tp3%0<C49>84H=t(Na6A`M5-hm>T0ofZJ{jK>Z5quqPCy9r|
zjvdT1?n$Uy<%%P5W^95>wEZ&QLxZCn1W(p_R#-hGW)_O}!%I3F3#~GLW5L~oGRG1d
zAeiYN^PkZT;*fd-x@heI!5XM{BlB5JkVlu7t{yrfyQ?S|+7uu(`}OP*6YBdWHfv6J
zEci4f6_eH(f_OZ8<gmKddS!QA7$5fCx^2DsezJ&5jk+V(K`ykC;+H(<8JkuB0_hEB
zRV7ud)LktuTUlOd0EuKg`D7<EGxX|hs;Ftk=5?U8yEE7E-?AM4-(~%$)+3<*2qm{s
z>_`F~5RPVmN`~JHU#Dt`MVu-pS2Z@rjqx@QGO=l!mrn(#w1xy5PUK&yYkh82+a0tV
zC--p0_(S!BnBZu=@)0wI$A`+aAI*|2W7I%A=(tr4wSq`{e>qncl`WTSGcy<cYYj4+
zH3J&GPE(?U%^?M34#^cR$fp(V)|%%|02<EQ=ek`W1OsV(^cFOV%%vQjNaIpGdX@@|
z<>}v(l-i{kC`#}#s!`r@T+@@<#@A1&9BT2i;*k;LkHqQ*H7kmy9!V*_?BV)Wqf*X*
z$98~yNPI_gArK;>h>pM@HpSaLg8ys}Xgc#?a=k*jHqmp9wHY?%x}II}@<L4%kQK5k
zk6in-L+4#avFCN7y7p{9m-aL5?{4!!UTQ-Ek%jXJHK5tSxZUxN_Kl*!6H#ilfrHnW
z$DKzuxeGnz8Z;074h%XG#^J!(X7IFrsX6E|I`gYzHsROhTH(~yQ0S9E+gI0XU63ag
zw#T>Y79VyzSw*r}Hd`W>6~Q*cfSqk-7Axv3I(GLjI$p%Bq+*>in5P6EAh0StQ5pDj
zyc8}?cBfbnw|m@K;q;gT@8vx=(^eHR5P7r(Io5PBdTlYeBr}4c;`M%7!{*41<1gsG
z<^LxKpnpO!pa21j+4QU}do;^mbQ>an(G?1OKDhWVe0OrPTBJ=5J|rWfKspDYyJo=L
z<OFhl2kjz#m<+rN)0cv7{^`4;Kbu;(5{W#ofz%EX2H+o8r~a?j#lPn${U4uKatW8L
zbh~i;=lshLtRpn<mH#^_{Evy@|Mc%u^(E2Yzw?<tDrO4NxH<yxq5ofQ^#9hs|GgZ|
zhVvg{lk>$?-2=#vgD1_Lp=2+LKMCLm3?1d5@fyTE@*dfUDuR9jFu7~1w;HF__Oe!<
z8cb{@Z8{3TI}C!xp;A;XU;qo)Dk1-WTcts*D&L0SPxEb}BgUHtvLK2dWDPjvZD61y
zZV5mnMDGKZNJ~eVkYAZVjRXV!Hy?d)Cwel|rr+I>rE9;2COB}YjvNGfW#7kg0pJXP
zparxY9ODnf(EFAn$o-4Wi8*opeyo3g`0u;%_uBaTP5C<%{2dto-X{LuH2=;D{?3v9
z&KUpBpZ_ip{)fsXu56&Z2YOsdk`%w+VxdRM(`G+Rwl}F9UbMS4c-&DNa%axx{sY5%
zqr1?eQr=G^gWkyOopwY6(d@U)F7W67pUtk>h{@Jw3v^#Y8XG}dKprXSSqvQl#=8nA
zauCoI;In;9>X~z<+;qkFXsBUQZKX)iC=?*?=k3Zq244?rYgJ6>P14)yh>mP4?_PcZ
z8(zgAorCD{xHe|vZ$hU@DW%2mz_zsJTm=AGp26)b>E<f^@l}Pq-`TU;3+G>;Fqw3s
zOjfd1NpR5IUQ_|)eF2=6AclEQWh_}$teAG@NwNIwQXll|g}4FeLwUzw*XNe%uXHz*
z-Ci6>zod!&g0SV2Bm%gGr_d2TQI(Rn)g?j6e(c&Cq0a?guow2qvILNzYGtJ)&}>$R
z5gk>P6cHy0zDAMmTn{SfP|!cUt`Nw<RdL6>QdvJTI;7-IkwB-q*Alt{*_y>jl?4Fp
z>k4y_Q<PgB(!3P2Hk$D8&Y26b?=OanGx`fx8wPJ|!2v!?!GZ+{J_t=yb!;8rn4}HZ
za#QjPEfGq@=K3*g@K-daJJqJBx=?*f{cgt^BWT9Iq~r37iVHIjS>Kxd5K^Id;E~bD
z8tuYt_?Pq}m1Y6+o}=(_15Tx%BdE9U?H(|Fe%ZYIJ!CfFO`clB)m<NWXWM3c8ZM&w
zvL#iW2zr^-nJKCIXv{q~-NZn#zjZLrDIqT-@s`cBmCH4bmzgI$vcAS;F}$Q*NBB_!
z@yqt^$KYP~hvtVQEqzlY!LS&J?8?x^HauIhfRnQERqNx-0nm<txD?14{IGnwz!`+m
zT%Ug$Gkq)h?uosx$9fvPqfe}CD}DbcR7jU`(p8(vLD__n>{<_wHj45LX8YVRs&BD|
z=)4uV+!)5lWSeqTyl8B0@tIS{PBjTFZfFlJ)DO?a(TvX|&d6MO_-_6CYsv-s&c4ub
z_3=oA+)>7wJSa3+2Jm<{azwd+mFjE95Rb2YCtAFBFk$92sIw4`Vq>$qKuxa|i?=5h
zMC8MPuLi&V*dkaRQ8<Bde#|6gBU1DF`0dYai8pS1WH)q<n#|tT8ecadg%1$c<1x<Q
zQ%;z+@wD4%Q@ZNDOb`dX+jsSkBEqM1gpzlbA~vcR*BK1zMoUD7b}dSC6E`{p;wvvz
zw%>%0*%U9HeD=vT*H`Z42eZEB3xCnQO#+NOx~rCGsih{w@s>>{#GRur%`z*&m#VT%
zXqZ&Yo+7ubbys!y6yvQPwcB|TiC1q{Hm^6X(<Qq2T+ItIv!+#*!p6W-#Ko*aiu<9F
zX7iyDxuBf<i$Lq0;B`WJV!UTy3*ZNeb7IldEz(lN4w|1zPe~?u-5`p<p17~u7#UGA
z2%DUYf={?#FR<x7K7U`=E=u%9f@br@*(=J^X$OvN{9zrXq++sFbHm#KKc9k?Z37WH
z#s=pEZuNjf=P8wEL07d9za^l^K1AvENl4s5YHVe9qoycWY5ZbEO?|`Ln@?|LZwtRR
zzjjBjJLx3Ngpcl+?ZhDuCG<W?XRU~cHXcP?9#P<{)ICvS#$u~?HsoW4kU3skZ+=(a
z+Ti?a^tuXwdB|b|1Fk0D03C6F_RMYxS2E_eZ{NkmjJ0o^Z~J<@Tp{<^8;_p3nI?wW
z)Xr5%OilBJA3n_V_lZzIXGGMw1ny|+gmpR*Q?Fu|sCZT0P%grF^5YyBps)ZXhx-wL
zv{>^17$kb^ImECLTL2{&5(CFr{%~X8)BXV#nsnlC8|#$0oOcj9aQ-ePWc9SlMqRsb
zW!zzB7D!<r9CD%r@t9zAT+f~`xk`~QMxN36+%C0Rw9y#JRyK<GQgXKLnc&BS5Urae
zhPlq@MvhMiWnvThrNk~YGFX-;HVs1FOR{rFDKAN784sUIq0@bL{Cv4vDAyJ!8q7zr
z>VTFjeOuSJ8kF=Ary!ut1)b_E`jS;0m8<HSS@kP)kUKkBuR&jNtw@T?=EPrg0A~9Y
zhaRnCWM?{9pa&wc@0DL#S~HOx9-vcKedN?2n6P(}Ya973>H!41O3v2yF*}4th++*{
zaBFV`n%|#r{5c*^;m~vEr*&qr@4Qy(Fa@*Kj)5<%jFAKK7kb)VJvTouNT!)bK2LhN
z;2>8UrQHAAru};m9zYwt+${~X9bBp<@kBuIJZSx%NRh9H-hP=+&xlLNZQh{UJ}17m
z((uKfYYY7?gQW*BXyVKHr5cYZ0-sy0?}&iEai;N&+BFe38o$Wj-IiU~i6cnPF%5>8
z{N#2MkE$C}V1!J>>kXFf3S<cAkd!*~`xvZ8o-I0wh8ksA)o*)Tdt=99A|i46+nep!
z!RHHLT2AwAk{JfX7~Gdt$<U%jfdUl3)4em-M$2}Z9B+K7O<Frhepi`jTARrIgt2v*
zZ<~&Jdp42fY9{?r7L|=u2v{>5rzsE3lP<jlUkl;&ZfGbv`Qfg2!^NS<<QK9ZUN(Jw
zc!jY>vbNT97u<|GPK2zx7Ep@GS){C%^dlC)4N#l$=>6xW@%KU{Jxs<%{HT;>tBxe?
zNlmbs$K3h$!{e(w2TN-ivr)U^ldrWGc}AY45LPrr!>=d^yt{ZZUhJ7i4lp$KZS-j7
zSb=bYF9Z+CQ5>As5J7k}?{eBwxO<+OiFD4e@I|o;@zZzn3C|4`yAxSX^894CEC@t)
z#A6!_C_1Ecd<`bj+d^R)+CEOiPW1`4glk;mvq`)Pop<eBxP*Og9cexPqzT*1Osc`y
zATexIeo}ngIw6#x^BR18Sv94C>48M?V}9&Re(5XEtHOz=Us#9S^iBBcmA1neA9292
z&U|QvFd^=>H3*}AWpWJd*aJZ!V$<FOwlLKxk;;>K@6lICTC?fanR_puKK(77FlMp@
z5ddFDwq{fjAL1rJSOCJ82uB$ZJ8+fzAjd=NeD%NRYA4BdYRUt9VF)CFRRH#TlHn+g
zRl{WRk%aOx`H3|#rInImHcIaD2L`WuQ(*p_UNTZz=q%YH2@6@3eHfA)bCgLFpz_vH
zIQ(XdjVL-BWwgtbfUA$fX*xFd4Zhh)eA+-v@VL~Z?}aZeF?p7quU`oou=OGh69Asx
zRV1?$U>avc?j)x)6*>hQVG;4t>&7LvZoZ1X8~syEb9{oZMOGdPq%BJj?GhhjX_!5W
zW1tW^TL+hYv9`%`8g5%dYU2LuCkhs4uDCjAJ71CwRn$bKoU=PmphV-$hLH9+CZ~ge
zNcs44f*`_J#$%TQaIOZ`)O7i?u)b1lrt*wwsH6}9?ZcC)?8gH1;>G)sPC(2d)&-xU
zjJJHcHSLo2EXr9J7=iieMM>gl^w+d;VF>*>WRzv$I&%TVoj^;v)MG&OYdrw+FUd#W
zBStUR>5P<41n&E$7#9gPl7_<9E(v6PoKYP0$zW}@Y|h57Gt$_DJh2lC6@#>&DOu0n
zvZhF@w3zP+DP8|o(q~2ReT^UBgiB*odzvC55f<cE!zIc5Ce9aL45;#Px(eJAGCbDF
z1&j;Y)QXLwp^N}f!aaJdUuvg&N3T&V+s0A2WUcglp*LhYun0hIqBTh)lp`YKTd?|3
zKE%P7#1OWoU`IGzrEg=re%c62s2Q@<W!s+@p<-r|JHOXw8UFf<juTzgtO!rX?~xso
zk79DoU2s^4(?ACKy&Ff*7Ma=*JdiON=eNVV)QC#l1fPNH6CBr(b3q(!rm{GSpCQro
zRaAD&dy$Ity}gx8Zx@}kb62q?cS71`-{g5izN~-!Wy#ZyX0%i<{z!2oTrSFUB75E$
zl1xr(cfEKcEJ~OwAHYtf`Els@%*DUP7Q~P$Fi@sO1-uhgs|4V8r!8iC`4q+QdLOBX
z>GIz{EMgksd4A%NRN&V**cfciCU6#lIfBv7syS-xg~=bjcOXhMi})q5SUre;Sfjg9
z{})|}$t>v3Hk65~L{jSPK}HKnjH?6M%jaB(@FcTgoE1qdT7gkg|AxN5#l`TG(fv1{
zm20!Te7!X>HI;bP8u96f^Ak!cx4bfQ(XRT+{XEZ8&p$q;g*3~M96qAYH4D-Pf>}r!
zedFy&c9>!!uBXsRHGaYydG1nGtyx`j=0hE)V+kN9qZyUMSmhGbtH0=2<@+=Vpw5AC
z)xYRM<uw>v8Ey{wTt#r?udEj>SMRB9nBV!>&Xc%4|7}s>9{Az253xYPNmjxPfSk~r
zjU2g@3QP^FU-rrm#&o^;s2s=G#ADaKu?=T)4Vo>!g)k$;XR|1mlu6vLbFj_<HIb4C
zh*t^jR&8K)jM<CgE-$JuE^|U*XKCHSkAW)5I?>9i3Q?hYf0YOfU%YSU$#`t)I3~+u
zRST19yv#4YV8o0NB6hZ+0Cz8;LI7&M3tdwIj`-N_Qf~}Q@LQ!sMdh7a-51{q=x<E+
z|2T8G(F(AJqZ*L5aD<2$g#iW%sExl-NOW{(Z;Yx(czvtq!4LKsQ{C*BPb%J6N!)++
z$PQ8Mrz9FJb-wUdJ{L(Rwo!~A9fM<{9lPm6K%S=F8F#JRN4Ymj7r7X>6vErNd4Ejh
z?pu9m(|zL}nJt@WwlSVf>-2ICFf1m?wfd&9RL#2T=#049HxdnY*UVD*Az5yG%&!I$
znb=>+y6*c{&vFIa_wf7`bk#hQpVS%A=qObz^s;pMV0hWgz%J-0%QJ})7DiWT_%NgF
z7@u+1CiF7(k^7ldP?cZQ8dwyrW=ToUwuHkBlC*{^gD%gRN9blO)NRLB6N~-j^s8kn
zu7j@GJzusqi)xgcbEl*AB9DQEY@P2pBZ+S)MV0HuhNC`L0ScZrtggg+wISU*Vch%4
zj1rc&JpQq=gS<7O@D~@?mUIgw{9v;cgjF*t*9RR#fCqPj&MucV3gB(r$eZhZoC)tX
z*=MYDb0a$}5<ali>UW6(N+YbZtfA_9hm4f%&a|rJv3x?46s|ojCFzq<prZHtQq(R(
zB=*yzFSISl^>sc6hi(x%9v0)BThYLK*^y5`M`JtJaCq@_L9w=a$4G>@VwHbBCG}zL
zmoH=PbJf+Z74p}=+h6&rcO|lHOhWJQquEU&s-=DjqoFxx7Vh?x822(45^hS9J^{Bg
zZy4dxf1i`V@53?KCam{XdSg7c{mm%@;QQ8G#Anxe@S7oEep(y&^jH_^3JDTdDb2jh
zj&rI3DAfuhq8X2_|J0HwV?Q0x8HVN7__-fvfM%pH2cM^nLYXQVp%Ip>7(tP?G{bUO
zk+Ui&(qMUX*6QJNw{>ddH>as{mzWl6u3wG;$}}L2`uFNn8w~{Dc168PL}R2`X~VX1
zME6$`Pe}1$2qS_Z<VND@?!+rkzSJZNJYCw@6pJSphYW+c;h8vc0I{ZBQgs-Q4%;28
z9^Or+<RS^{230Cc2?PhFN9MHcJ_i~3lkZP$eMID>Uf;ieU}Nh=9RIx~N=SpK`hd@w
z5gADpPRnS{Afqbxvz~3c=4J&?hxT;7ui<+(ry?lc?lW-w7X*ZS7B2v|BPakgR%u4K
zmwx>Zb%@V%!YCHMsHBwn;0x1A_xZrM=Q8ULg4?k|>G=3)Xr*>SF0QZ0IRz|MSrY<_
zsqp?)Y<F%eHZO6;RvrB=<=Az-3&YJh)e%KhJ<7yqxD}!E6<7!kYb~dw+m1`RDX2x%
zkF>wyR=n5fw3V_oFRS8AXFjU!(M_MWn?}JA#J(XTXre%l9|FG<WYn4#w1$lcldg)X
zNj@WId(+G1P5;!<yzQ}_QgkSifhso##cIw$qkSeVb`3Odk&-LZtmUlH6D5oM_DTNB
z=Y;PTE(rW&KTao;((TJ_3NFoQK1aDp^lE_$pdF^Y2%T+a@0&$z^h&02@b{aG;hUIk
zO%GMkd)Gb4J}>R=UVD+p^hVR9DoAGz1ymq*Dviz_NI_bywAe^NRYs{}9q4Bb(}XwX
zT5s|8dWWetUF;&TLiR3kkuE14D#KZgj=rjE60x|uwKWu<pxGEGyEVm7U^&SkjXE>n
zVh`8EqkYld)cocrXU^&r(K;+8z^uc;l5l;a#K!<^h6PSOG#ZjvOwO|s(1bZEhsd_Q
ziQd+GnJK5$Hn^k3wGH@tJ_K&n3;2rSqpy0i%#wq8E%rP$7v8A7(B<1hU9GAQ-^TN-
zPe?Wn4zM+ibE-e%^OQGl;S+krw##X)VpO}@k5D2ZY3!_V!EzL@qa-qtq9fl2xlbE9
zG~BEV`t)7VKk!svvzbk&T0N83)<8RJNHY(hD%?5S*(P~hpo@x7*TUCcCP}vM49!fn
ztC#2f=xYD+#NHQrh3>_plR!0DF&lUoBu&OraDF~l!e?-;X`DlKG)`z$ih96VDK)1c
zF>9f6lK!FCoMU(}2<hODp0re;l^*4iWBnlxDbwm`Y4^T{b-w9ZJ4ek-2pBB-%fph^
zqgvCBV}<kU?w;`~HAg%kIYte!vUl}8=Rfn{N{$b9;&e6b$?<mrp0E|xxEa<T)S$yh
zh$}xDwY)Aarv~$k)Bo)Bb?x|Q3cGx$mMolc`s8=|?bAfhhL^yWG2(63!7up!D63h9
z>P^gQZ7>r84bwnc1pS~>XDs!D)GR>G=7eZPF~H&Mbr&x~oT}Ly=PX8bBlklMV7aGc
zI1_QL2dh@j2j8jp(Inpj?1$|xlAOR738epG1#jB%-cbL0;QKn)J+r*w?!Vt!QP6p+
zBXQ|aYz@e=<DAoibC2@L891lI`){9Zk(tvGWestD-n>hL#f$D|rl)=abTPp)iI0ai
zYqo4Z=0k9_fcluP#gq@M9nH4Hcnp*ajw3r(Q%WzzV~*t7#!X+B=#z{ek{oim%k90t
zvJdI9w<n+Ql)|``*&XxUQpxw&$|^Kj0y9!t0H;1=M^+S=j@ET3Miu_zY>ad|Dnw3#
zCpm}gdVZ=mWtxJw-KMF^ORZEJe1_WG(z;DHMU0>WINRA`@Kn{^%#aG*6#g)3jqCZn
zOUIycJr8euo?F;fdc<KI{5_a!kjimrMjHwc3iXrhPqt?bPvmuIfF(Jn2X<U%cw&By
z{Z=tY<zpSXvwmEa;I~a<Rln4|@|S&S=MRk{q$HLyH=&2#&Sbs8H(Cr6T{E;tj%qEf
z`)~-{?4~XA9B5wy8IB~2!Egu0h2!BA2+;R1YiA7y2`QjCCTsxpS#&Jb>T-s}cw*Sl
zeB8!tUdpEuZl&pMo|Y%i&s?KB+1fenZMIaFOB~RF{@lbDc66lPuceeE0&O*SZtE2e
z{eVpmB#fy%MKb<M-bS;n;@o+TzJR4pSC9?`b*R;XkEdvsHD7@`fd#D^;#pv-AAD^E
zzdA2^aJ7BRZLE<Hv8V6xdB``L1}wAD(I}v3q6D`3VYO1KN`E?=rL}ie43}<7v8za5
zRz8}#V`x)%o@o~qT4?|j?MMJ0gyrw(SaX8@6jDspuNbb(@F*Mm{-gTC@bZ|S7Tf#u
z{@QSv%5z(<=#^1=6iwnJ#xr=xmX8F&7=>ZcFVrOqoy^EsDz+m`Dp_!}Ezzp#b0_Bt
zzuv=6ary7pZ_?Gx-Umu@Ob%KO5NN@*XYn;Nz~c$p6?ok7!l*LGQ8zcLPZ_-8LFJe{
z-)yFRYvkHHtI{u8jO^gnPiOk)YzIhIEvlOxrzlDulH;MYYcAEn+_a(5M*7byo>=o$
z+b34xBLH{P*Y9MK@ZCYK<M&(3bMW7TlC==Khx{|n&*5GGWp#LGJrLu^FCA$gve}BA
z9`V!5(IshJsHsd-d_?~FXnEk|Pr6elgBb6oegxF<+WrHj_P_O;>z_OG(3@rhXrjTu
zzAP>RxyJ_BRlgp+N@X1l&NrrA-1#><_4oDH1A;#dp(l2a_;&0)aqIwD)0nUqK6#TA
z7Xd>#YiqrKRA>Bd-g-SjQ1_`vQuu?@-HY{U;+h=2VE$kL3C!Hg1{+$<<RjuHcI$)>
zWkxnP%X}E~V^n1FYGieP<-JUPaID>V8F`$_mQ6Jr0bf`?keh&`?mejSdu}@=YIfq&
zgH1QlEerAsrMJ_amYY}d)G9hF0Ywm2Ja;OAG01kj<wyIarX}(#Zjfu8s<)EL)Mb4R
zv58TPC$IKt^*Gpj()b$Aj$}qo8&$taz_BpvwCc1}tNLW-3gT-r1%7S13u#3W5*#D<
zjSDQEpZ1jX=i1TWG*@~59iaet??OUpg08GBSW*7VY1>u3MI3UC+los15^?%nOY{<J
zdf%3!=fD?(yOE5|*Wk-kdxFL(@?2Y=Q;OKVig6!Cu`SMdok%gQ4#NOXe}&bNdpwTG
zzi<_-nAfZ=;7i0<eudqEh(?`r5yllGJ`IOB6}k&_+lHGU<Gr(Fjj#Yvc%xb&9D=Dg
z4b8=t#wFxqwd@>SUL}X~z7cggBl?Ex#(Yw{)8Gy3g$*z-#buRXQW(E5><6zNHV6_D
z)Es#0TIqk{H7K>AN98#Sp`pQ%%Uy@481<1-*d`?M<?EJ01`-JLSTiuJCTm_ePWo-R
z2{teOEz@3iDfSB6oLq1xgkgi_>sPpjK|zVs$+1iYfpYtL@26?H>#v$0%F0#*Z^@gT
zm_O%VMl%Wz7DzHB78GFJ+#AK2zBp~|Zvj-;yz<j|ywefiukbx@vlC%RTmJRY&ruBz
zOcp?ZQgoAhzB2%9jj%!2j_h)6=OMw1&rhII>}sChZ=Gd&eo}81a{Lz(V65g*tT3jK
ziF%e~x|)j0?Wd{VV;vcrwH-795)It%--u@;DXl>PyRMll<La74aKC%1Wpzqeg{zY4
zT1D9W-t=1wInfKoAL#`=Y~zWW`1mj|JH<mD#NfmY7929^&2lJ@yx^zYQ@67AT((zl
z)~nrn=bl={03iEwlfpz{y-jo^^H7Z~C8?GfVX2d1M}56s?+5qx(<j*ze>z;)5jk(V
zQAOOu0L-Dn@J6doz(qNc7$Vp=G^Dos6N39|u0$#SEaBRdZ~{N_h?7aO5#&5MAW4Zh
zyiS#+6cM_p1_XD;d~$6yVQ>TASX-rZE;NC2l_|19>`?X;?_9p~B=tBY(0x_|)}oKM
z;vp7P_=B?p^S<-z3i;=6+l-oLWHn;0+&yu@mGPy`5$DN?B?v3f0n0jukQvV#mlqV8
zJ9IAX`#9ob@L=;nZS?V*GCG%fzsft^2{Ad%Kp*`hW)>A9LcCl$TSHtg9S^I5rKdMc
z%MzQ~jP=#m)k_c^cD*ImkMM?ybM%5$z~L8htPn>?1e(!jSz&<nmsHyDE!WihBR-6!
zw+L4Z<v2a9dqlJ&=;v=S2#OaImtJ~9&aBK_9N8%by3g<sk`J>XR8ePp$vp8K#X%V9
z$1cTrY-@>6bZi$mc0b^)>4)D81ElzfongSMzShJtxR}TYRn@DOuTUQn+>(>&O;q!Q
z_n)FW9?uMnr+_v#@M_z*>kiYpiztaJYJH|iRh`4z-Ry0ztKNNn%%y$qScgX@EH~%n
z`K4HZBN9kBktvh}9j49DK%<Yxc7~VqKnecr^_|jP7@*@QV0F!_OPT8>No<z&C0NoW
zC9N>ZTPMr3@@47cku~WJ@5A<2AKo<{I*YvN&=Y&?S|>_qIvdU4>&Y(-7@PnnzcX`c
zn5s}{MbUP88sd847C&F&-Tll*Dl=B16^eRdqT?rXXzhT`92Me`XKg`&DDi6@0TQ|f
zUY~8J^c>|`gxEOr<Tztxi?-fIPP%LuzqFmuI_B#?-Ugl6{08Cj>9KKdvlQqqoxI^{
zH!8n+m;v7G7ER(Xa20hMoKGr`vNbwCG7hC{7Mej^1q4uBXk6t+C!Rgzzb%}jg^8jM
ze{dd5Lc?7r>-vEgH~RQmHvWiZ0K?Ccp6k3U>dPNg>6@}FKOk4)@#!|RNQ$2_y+m92
zik|xTrCIb#@Ck~4E(LYxciT35cxRJyX37QdG0u=E_K{z$e&AP*&4>SdcfX&myKF0(
zZu>}V7WASL6#5k*K1>CtG#^hKg7FNA)TJ{s_4ZwR@3;T{$(p|VbkbM>C5Z@Wf%3=;
z%IT0yV|^Gkq)al~vl}wU0v%1T{3_Yr35G*AF`ld{Wh=lbFIF0b(*%i|(Na{c!k_Z0
zqhlGcd?W91_AXoORM+EMe39SZKiw~{-I)Jy)-M#$e+tA?Gli3G4AZ{RE{r;i?TRev
ziDmg;%juTx?{qVj7ce_3^NmkoaPUT)2e3l?X(C#46@^Rf9?J$fm$R4j+SBh=$lKbk
zf9Q&ytlpU8>*Q*O50gzORwVkifbfP=IJ9@f=aR{Y<-ooAikSOn>5i{lvDe+WP6ui-
zor~U$z9KlWL{p<M%t}Y`n?>v#d0|NJ)B!=a3;yz}QI-2U_%D}#Uao=9N8R~cWb)|r
zz0<P~XXq$EBQLDJX`=`qbV@_)UQ<={_Z9C<Gv~QDJ<$h;QyW3scRDn;%iZ6{ol^UJ
zg|VWiQ5=XUGDK(u_5Md^A?`Rq`nK+cm4LhBejF3^4w}`<q6vw|-dhtcgo=XjyaU9n
zwFqKri;|_*OadNqZCke~n%1!~cfX-7bH+S3QMt8SqGZOhWQbK9485@4bs~_N)8gbr
zST052_78+HftBs2g+G5knv7f*&r+Bi*gDtA^WFUO`U9Qsrvm54&&(N3nBm>qSlL$i
zH>&VRx%(r@X7^zlw^sTkZ(pHf{FI$6Qu~cAc-JaUeC*0cseX<Q9KAXtiBW4%z6pl;
zQB@Z4GOsH3>(9PIx|eh{X{Ypr1#Vc29DBjw(m=!k{y<>%pHNGR)g9UZ&-PKl76BIJ
zR_2zE<v~>RZi>F5qm$_58oz(TCQBG$M>(bx1y+L>w%bZkWOzE7IPO)H*ZfEuzgJP6
zt{Zpdr@@JYsh+c6{4EyP#nm3S&Yt9&KrkMqkKypFQB+|P&&|qIjOADjIJXcECXlL5
zSL(|x?3?zeEP2+;X9>5;;f?qpT7!2mi-AIcMBMMN!QoPqh6>-Y5FXu_wX&$vxJ6sK
z*Wkpt@OXRUY2r=Z3xY30=2?_W89|}V=Z2g4R*<ZvJMq%3{BM<h*f`^y-|a`ix0<@P
z8TUg2jjnb*GV|++W%7M!%r((MLCqSyRQH(85NHj+MG%??<xRr8yLyN6#%v!vd!3zO
ztuGM9J5wewlhgt^j$kHkYL9#OF=%psS^f6fln_|7;fDGe1_|;KuSA7CIb}U3##=@P
z-Y9~Cdr0u<@n-G`gz^Xia&j|N0+2j#t?ZspXngn4#N4;bs+Z{{4{L+3wpS)htQA=S
z#h^}5V(yIZlpfa_udI-%9+g%QZ)iZxbS~8TJ0xSZCf<Br2$^4}OR`M|;@9&4=r{BU
zCnQG|Youy$EZJ6kgtK&dil^obSIafdhRG=Pl%S8y3rZeib)^u%MM^855`r_UB6`c{
zx%(ewpNyDvGt6M__dAo=)=l<#J{8Uu!YYzR<_8vjf3p~%x5-MozJm2FCX}9-als;7
znX9v%gFNg!6Ow8>>}wJPlhoIqn2D15oPfkmayBcp50DJv!yh793rZ!7XS7^47rIDo
z<-(_XO~6`o>-1ux^NFykoEHr$J*bFGP!s7W;&s9B9woE2)kYS3*#G8Pb@J%E!OQCN
zU$1sQ&{h^nOT}cK{6Q5fhQ}BWl?3+$5g6M?6jTrspn;4hx|;ZRVMkNCwf7i$K0Hmb
z8o4LZ)u?xP32~X^9*Np5HExyZ^ZGOuR7RT-Uw!usRj_%fJ?GAsyPXZXu||k-|Ibtn
zBCV<n4mEEWzQ4=Lq$h#BozO)pE!L->Y-ZCnJtbz3IeyCX`VEC2gEuz%J%EVk4;Im%
zD=(aD_jK0J0Ru6tFRgc2;&%+wT?1m+5)i|r!x|a0Dc6>uwSy(Wor+CGVAQ9YbkWS9
zVV&~^V!fSi4|qF9Td8%xHWv_m!XMqvIx&o2>TtEYJKAUkw1sL`>YuqXR~?nEx4&KG
zIA5u)E2<6e?*7?^5{om?2kg(#tOzwch#L;oC-yaG0Dr-C`-jUxP|4R$PEyrZ?AK4M
zhaVqnL8sT3vl{hKnmI40KV(HUa{{w$A^yQeNbxcfHY6XIuHPeoth|%?aIB=1-p!P0
z@=g(kokiu;$5Ai~1!4{$dJs432QhH8VO1cn&y|f!v6W0^1uAxCXJlWB%<7g_R%^2=
zOaN&zk$e<IW2{d7<m@_I<$9TPNwc835bo;w>eYTY7Nllh<l1}Yg>Z{Vc3CGu=Nt4G
z!g^FfL7JNqpz(y(inY?$kEl&ks?ift^5WC6LDlniy?H!-{(AQSQj43hB8p-~G;2Yg
zTyrl)2xfQY#q8f6Pi2;`;o&hnc6Mk$Ts`sJnT|!}?sm^KBoKFF^hqb1QLx!xPZDRg
z4{tHDR!?Q93Vhh$xD=JNwt}x?5;-BHqp&6qlnjEO5Flb)5lKy-5&=$#=UuKYwRiul
z=_P;rTT4saZJjgM;$^zn#a$&6LPM>CfQ$(^y7Bl%6)c|N)|QPe^R2~pcQ-S`+B)Wj
z{=_a_$>}{$i(Ju610bHCocM-#NMW2cdM)ehNkQMD4K@t*q>Hb@>ek9i_g)A-fD4MA
z{l-+bcHR?6C76RxP&j63MWri3EW|Z_u9+&q%7D-oi4v(3$8WM->%HMV(`E63X`%Z`
zXWAwZ(#hA|)fux_gcOI>Z9ME2;SmNEwZrSu>Jtp9C;9~Lq|-lBOXKa=6?L7B{V?`}
z#z%o+0)(1!m$B{>JL|q%+xJEuZ!frm18W>>gL>P_$iB)C=mx%@1d?+GHt80lQwBt|
zA;tGV5g}NZpXXcejx9LCVtdY4&-LOK1yP)iDJqkpZ*=8>Y4`w`j*?A)LD}G;nu#{P
z*vU2hPd`_9OgXyU8~hT)t$O%UB-^qi_oRV^8=wH(mhH|C=tNFG=w-7gF#-c(SswE}
z=r4EvyfOoVg@-Vka>9IStMvq)SIgdXi_AYBb`89tZ}tHWB74<WQOaj+exp!M<x#j`
zxhu?mWiGCzR=NlIg^4TA!kS99N}@NqCi5pN9*cLQDn8{u-j?e^=pBkue+PpqW|%QC
z)^pm$4CFE*iEpkC>8Cwg>aDAaGYRn(OMI#=)(p#~gb|fmK%D5W>20cF>0Wu^Hehn=
zMj&Ic)+2gR<H<(Pi*JgPnT~zyD%*(m<(lY0s8XPK<g0`8fWWByyPaZ{4WSL>jGlZA
zM>D;?#8)}>xEpj8fKywhcu`LD|6%XV!=e7${$Zt3NmO<*6-Cx0WSus92r>5gmaLO4
z*~d%~${IouQ-rLOJ?oGygsc-{EZGe+*1^p5ynXNc{yq0q-}|_}zvFn0=lEUEbN%5s
zqM7%6KIeSi=lfh<uh)6^|E8SgY$s$mrp@8aL*9C_L<YapS83}15_tJ-<W)OwyHlc%
z4hfMM-Y~Pn62NdblU<pow1zPV_531^LIjL-%)XxtvmYox<>#=!6kSPsvRFksY=uB^
z>Iy`!nIaiMY^st7q|bnqYVjC`#tC5n370+qUbHxTG)v7;^L3@l+Sfz3#Z~tJjg!<}
zoyKnP5SoFx*Ws*>(4?%QU2RxP@h-!(>hj;$0NSK^6k(vi3kK^@j0Vx<AxTdAYy)eH
z-`7C@di7k)GJrV_21oq51av%C_0Ug-?mHw_I{OBk>J=XR3e3g*TLw`3^krRCtR4O9
z64Ad|ByHTzFP>ux_nQrY+Uu_tar-~js-;5|`IcH9O8C3l>~Z~Lj;g}F$Bh`R9q`>&
zpNbd`T_YDW)LndSfBDYB6|>|L12yU2*pzN)2+}y)higo~5q;dDH{8lbgEX4)eo*go
zfnchbnNVcdhtnV0jvhQUnY0Y->eRI*ONfAivMXd;P4;wtJC5`YX#^gCE}+J^#|Nk}
z#@|MNiXrRMxS&AXaz7BaG=}AOTYm4V7;YD6iT(Be48Vh{(F9$yqyuS1E+DOFY+bH)
zvg_l6mck#Hf4;N{=4q|MocXz$Y)2Dw$Wvt<)x`3hYgWfU>vOynt3nidwmbjG`4Np}
z2TCx=ha`IF80M%?+7Q72e`!}6=DN)t&R2&^dUQ`c$T1#~{&N4Fy+U1qMnqag4XvDt
zQ#m@7h)0%Xk)2mzaKC=|+qmDZR)cyT0v7M~*j0jjWWB2dX>3_cp`{m;+GH+NvOcXo
zeReI4*&@=fQO)&IXxgD;BiA9<)co%C_+F?^HrvQ{et)rr@r^)fzVZ#RbK_#)n#(vd
z4#aDk@Az)Zw1T~=Ny(yzV}OF^-dzPx(f43O3r=9RC3p2fnSnlN4pxKISp`b@azXal
zpddk*gF6@!Z2HDlT+aN6P`o!rPOUkZuRS0$t*tft#a2I7@1l8nEA(%yYmtVLS<)Bx
z7ts1!ckMTFQ6c^1$0#)qzWZ+P`12pR1spv{ot+M7i1$*)(AlHI(6bePikM0P5!0M#
zilORx;LK)rh1?jbcgd+a>y;wl<q0a$pe6XrU8euGK#fb70>*7B+GYk4i8=_dQqUl4
z6>>;2xwsi`9v!;IOfw&19%kA9Ce~EqM?B}1q`QUJKJPi)=B9U08D9<Cvjc>cTl1**
zfD*op8PGW3$;IXWF?=B>wlr{eO!eMcaLNa|pj#8DK$-GmH2t+p?yTt6CjAfqgP*e?
zEz>8U+wPSa(8t@@z8rL;b@ZC*OKc+QFj#dC)d$@a14HU6wYS4+ttjB1!WlcafD$_Z
zU1!dTo)5t4_w&MbRB`F55kP|AI$PQ=gz<ls-?0+3T<8z{uA1Sulsxntd;h%g6t2{s
znzY;u**Q@fO*@km_YL83F-)Yus9Geg)_^QbJPqdV;!1&HHSvMN3`HmFC&RtDO~aQl
z7MdSxo+a^xxK%~y3h6~zeW<;AEb0m4{JdCewb^7oLYMXqB|xzr*s*(@>oAltGV(r5
zZuQ=i`YE}Usu20|!&ek$l#RQ%P(1{eYUT;%{1ay6i&x(d5_j$n4_R8a*vG_dJUbH^
zE1*D|lsdWghE+H1qZFS{(TGKzyqCbmaP`6g*_VGjslO{_GhjTHpjHOsId<T-<XOmS
z7tdA>E&}JE&p=_atnOY<d7!z@%$?x>E&tL^a-l7s<dlC%h`i}lc=AP#$<!BnW=mj#
z6RzRb$4KblJY|jSg|fNyc~#TGkL4^6Ty+Y5unApFGHm~R@Q`4V_al_34}9Vwj5yw*
zPmFb35r-udTeLfS-2ZYhP1Ejn%*8#;Q#|!8*RSx~a*RPCs=|mNXp=r8<%BLdAi5!!
z-*0GsUjMkYd6QDBkS=e2-Wk^sCg-j81B(o;^$RHOLF!FVMhY*L5N`6T!A4L;2i!*-
zYtLEuI7>ch`t0v`=PA)O@o16g`}w2#S6B_ws^8Q5FwC$<_%{M6;hGDoN4&O2!d}kW
z*SQuxM8yeuO-8sbG~M%=oD9Y_L`MTD2!Seq6)+)vsEO*UpVpF{tCo%yKf`m+h8ucj
z=8@jzbB|Pt!hRrG$cRtS<I3{BSTMUNWL3K;LDOaWmB2{;y$IWZd-2BUq6dX8pFFvi
z2R%&hCa@|y7MRZQ>ObYx>A3FwO}8VsqeTAOK1^|kIO62@y=bqFx()_EV~&|oMG9>U
z60%J-7r<fj%dkS%JnUT&Dt$3G^!CL*HK4u&qu@58{l-vi^sgp8Kn*&462e0XQ=qE6
zqu_4=9Y}8HhBPx;=4Iz&HxBAVv^`Qiq*SD*@P-wg_5rCinR4QjVn~o}t~D)o!Lo)I
zg@L?Sp6N8e*cofW<GH6kTwbu`YJl=D_<?u|8bOEjxmMHo8p?UK9m}RgdKVXD1KXv(
zu*=xCLZvUyUby^*)grCBL64TH$~!+aRS=Juu=RB*sBLh5IN*KA)?JlYdmHA=9jN~O
z!rqG3G2>+cBns$}skqIK3gPkk{k82J6Q=ojs~zHo%1=XNBe_2}y-l3D`Qe}zgU1zt
zpj|!CTl5ax$_1o7aB9sEnzSQ_tOxW^?I|C`coNkGkAxL`^YPaVYQC^s-6w1s2mPj~
z$9^1Tr}E-n>Zby}SGOtpX|n|y<IpDfQR;mXXUZh>&`4{*I8kjz$Kus#hlsTEKZ28O
z80t8m6+S;IT~|<9zB>msb60}&fBjfn?3kci3#fO%H7Efe+jBm*FZb8LLHNVK3<yWS
zB=%q+=^)@e5qV&zkykwv$3oRer}$AuIxAlc47lshSp6X;9+f5C_L1i<Pbkkx*fUVX
zorf@~GHGOfD@$fcj4IhVFOdY7zhBs@SV4IFQO347dID=VfZ8{g6jG5r@8|jysWBef
zFSLp!uIZ4E$X~SCKWN5>IM$cFuXla<5uJZP`NTMZ+1%2JmlFUlTi%<tXWq=cc=Iu-
zDFrV4Frsl%sRFyVjI+i}TU)L>f6Xs5e^?FcP4Jk%K=~re-7Kj)pE`gTiS<S!V3`}R
zVH=r^91A%L%LJx$%lDZ2AFGO6qfYmS=>q{`J<75hH~_i8mM`s~2hZ%N|756{zRy>@
z{2I?>Ph!$^wrerOdoCaIsnt1jW20<)eWiWh#iLPzH`K(z$Q3>8wMfONlH3mSIWgDL
z5yGgiX;?{mXo$e;io9W$fswPen+R}{Ze?~^lR0xqOylAoe4-7C`qy=xItK!bIvvm1
zdcEY|iC$)6oVp+M_&MUb38}X54buK=^;!E#EIaz{!nS8q)3*xTYL-cG5bHU0o^Y=#
zO44<gn;)o1HjGV&XHhghrSeD3)AFk2{aI-5BG33n3)zM}JNkuVtJ7`I+~<5}ojP+=
z_GAsDDYoN5WmKZGY@W2BNN|Fsi@4m6)XVjdNZ*D12F=Pl1$x+widtHLzM)wpcPA3V
zyb*0%g%cjhdOMwZqF;*2;!Z}WSCOs0SGqht!X?L>StV%I>NsN|6T66V<gnb!l|8E<
z+hD$s)Ep$=K+|hmZCTZQkKhf+-$=*aJdomha_RV$GU7R#&c#f&KaT4AZAM-n*9*<H
z^!cRI?d}N)yA#4p;jdFOO_#<dK05sR<d3+|+)W<csIE#2El|jr&_BKa{Z{U^nZW$s
zx_N>5+IYOz)4EecFf_;J)P)Y+o9E87T^O_;A}-6}n0?p2Rhlk}K6tU|Rhzu?3`$KR
z67<UMPb&LP5G7DjKG8a4Or)plVMi}7xgtllIRK#*AMTGRxQYBc{%8-M^1}la5^|Gp
zQda{mc$ByI(uj7$E%Hr#VrAN#hXxoRICV-CIcRoG>x}RHj^~Z%TV%T_<qfM1WE+6Y
zt8$)9=5}c+`ex-LWVt1O;hias94h>UC1&-HKaRM!&VlCs<Sp{dBNIZi5yA`4>DP$&
zV_E+)p<><h{p7VC%{{d7{VRORYzN0+km6iWhbD}c`+ErI`df)<=wgg;L){HuvxG@Q
z@>Wux2cDU4#8sEp;MJu*Nly?3Cyu6PI(bz#Wrf=Js`Xv)2{$k6lT7<)b}c3NtgmvJ
z$^0M#*4o{=+Y%mx;+&(i;-zHE`_^^QmX(j4;RXYeoMt9$!%sAi85?_hJgQD2ip^el
zk)fU_UR5&!L``B=zRiEdTBLFg-cPF9Dx<sZ%J7%0c8SvR&_4!H7t!Bd<$(z6NIwDp
zm!X{Fy28RkkY$%CeENSYlEVgG>KY6k#}H5NJV5nANrgyR5SMdp$WsNOfBa<7uPj|b
z5-)6F=FbAbm-L;%=27W5OwGa+Zg7>ZAGc8q*nB!b7bezN^@)bgH|Y2`bTO260#3y`
zp{a2iR3+98;Mdc@CXXlT|73_o%xjAjElN6itAcDh^bQ?B0oHWUD>WbsSkZ-$7NLU}
z`ZFvYRtm;qWABhPnuhbaT3{&lZsh(6S6(MCbLWC=uWj+l`vw+G?h=i+Lgc-lbJ`)a
zC|AB6zNoHd3}W#7NyzqTYgI6K0HBV+<blLWSZ$ZcKmRn?U%WS(Vm>MjuAgIanacX-
zGy+He^pgMQR&&_A)~lRO=MUYH0r?oPeA@<yEL$(38$JNFH1zg4+?I9UPll~?HE=p-
zc?5bk*k=bQUQt2j(QXbMNBFeRpTeSXB<P<*7e<(Gu0&J3Ad8?ueFG1ac7`zEp#-Io
zqYboIpyZMS5<P#A_x?y)NB?$XyJHr9HC^;W37NKk{+;%Nl<${I`Tc5DzW&>8zwCv7
z+wGUF`tRrMUuK`)AnUgfz3WwZ`l)IKNKW!JCe4x4!-)$Y>HX$2P|Q0Lt7$B&Za0+E
za`@eAw?h*(fk{*ThgS2uKt=R#91`aTy@Zpj{^pCpiE^@M30eLZj_w!d1vfNMOZ;Ox
z(PIktxBTRGY?ycz)1@j5sQCWsl7IKK_`Vkgw%-&ozHhny4VPZ~E-hTLzC&`R9sQ5z
z-tXsu?D|=DGq0n6@6^l|No)V$p;NI!`0exsKK|D;H$Y2`(<<oSrv9&^_ur=eHGcU2
z_ftJVmq*t_Yxhv4DG;($lQj<svf)ybK`}D@SX@Q_hD&UFZu&*0NNZ5vsD2-881!cT
zv5**kI$LSU4pa@kcm*S;#v#meC^=d>z^pTx7lW}CT74y5tQMBTt=*Fdi#HmDYl6?O
z2*$wGyRSe_P^I<RDf%0gAiC?~;92J+6+Fw;c(+jO+4pb|i78H0nTl@rA<*i4JN03t
z#EH-+wfp9*RY}-Jh0eqYM`f)XCsu@q)3Rg@`$}S%h38=;ja+oDzjeRc6>P4<8`dZ=
zdN07fezzHC$1e=%4lPEZ&4r%_JWfnoXq>Ih5V}z2yZ?+&Ur?`$C?JE*gir;^MlI!D
zaJEv`C)VdG9dt*cDr`AM<0ejh5PkRcMtemod_+EWeC4B9D$aWM1}HiFMnD64kkDMY
zsEGLl-1FRa0I~KxQr;n)7JP}+^5gT}1fC~uA0nSGSvCtXy0v08X&8LN&}MTqyH9At
z=y@_MdNNwQythEvAiD;;@$kzzU(V4J`>U^+3GRR9X1eDbL3{gJ7z2qAovKRe44TkY
zx<W}A7X!ub5cBx(HstjHC%l_EA@9*ayQ_S@SDL-W-$lQ)i8x`cPfFD6ajg~h$<m{6
zy?3%1FU1EJq*b`e+4o4Sd;cNPk*oVTAfx?Y=U3^}I)DIYKzfWpBS#ODvH)2O7lKD|
z2%IGoQ-W(_+(`o)9Zxo4VQmHXV9%bbTk_m|$HDk|y2fFs2}+E2Ro)=Weo@_f7P2Se
z9$~z!kca1_SBJkX`yO)FQ|0q{6ubtlU=4Cquomm<yLY&rYxE^(h}3oP?o@`rAM~dj
z%?0AhmPnPJH6S%iU{glqtK~axYK_ZsvbDd{Vm@$jB)CvL<g(lK2N@yPW@I;N0Av!}
zJxb8Wb<Dek5box)GCP}CD}*cVyA*kCVr0;h&%h~0T>4`6_wLHd#go+Uj7clW{EDvZ
zsAix~8}c~KT0Ol5pGbBTyKnO`WRzcfz&oN*bwS9W%&6=;Y_#tdQ|s>7RauJ0BvW+l
zL9H>Oy6%G>LK9ib-G6^c?mqDiZ>2N#whWqAY>M9>e!#?2c(_ZrMQTX@(}fYVEg0!j
zsrzCSvicZa65JEPBJTY4=~SIExhcRvjR6%f*gEd)4W^Y*Spz1){U(kI@2EtjNHyrN
z7Y443PRmDd7$X?SO(!JDGIDZnl%6Tmr0wnZ;mV>4ACS~H*(iIPY+gO*x-t|wJ;r+L
z=mB^9-kPE?1~9#XL2O!|6ZL77jaJ}^4fh^O&Ci|{5bmWKag;`-BTgsm*I(2<U)Ji<
z0=N@X+6rTwrx;X|-=pmaTimK3HTyQnN`F}r_Hq8pyj*<uI1cQYF%ZbGBBe-OTD{7C
z<&|Aw`J-c2a7W$|Vogb#i;#nMGygS=NHh1I&7cQ<sWWu`T#`Kfu{oIe%AQD|dV$lw
zY)V5)=0C!gehMGZUC2j+rj^bzT?ypPG8_KO<1AM&MicxRKN+HTo$$SrU4I_xQX8Oj
z`xXjXGoM01D{^uT-2fWi_So$c>#S4}PY4~l8|FO!&k*2$_tiei@<FaC{<%J(2K54=
zn_%%@Qc-Z_QO5s6lq1`K1gbQFekhQEvZD4QwaN0K)IBsz{+udTzpHowuFUh0a<*Zo
zVSee-;j)9Xb<s{cUe)Crpehn)MQQlSaM!EwC&N1<gL|^V0bLHDdh(LEi#ibe_%W<=
z)NCf^Zpc~56@@xuMxnUb{dat%T81Ev5~>!oy-PX^mkI4Mr6nJ7A-6AzE<=ct8Z|al
zm)_-dNHFT0F4Q<qD^`j5*}d4y$Bo1hCla29?qg_(Y={}E_G@WY6{0wjjK(`&^AZQL
z&=Z<@X~nTBr-kEmOF46vV$|5ae0jrA^;lRaHGXPBIx5)3Ov+QNJj?%VCAD^P@Y9%J
z*xL>s|0NS0t9&m)<#HB=P`K95F@ZUPbf0LgMpeqaQC2H(kT-D^>6gVGReiY;_jV`v
zi9=MuyUB<^(9k)7v?HHO9H+Tj!=JLX=bGt@X}p<JJYU@#C=_~3%j=A=`_*>eusyOD
zzhqy-lsK*vduQxd?tir9m^a=qh^@tR;9ZM7%gVJ-OnuIthxMLcdS86`tc^J11%6Hb
zdGoFE7DbJx%4$2;ea_A3ldYP=wxgmw+#RO|Ejl(hj<wXvj^e6EA85I#`_y6eNzkx$
zD;vN&tl?c*=<I#y`*5l-v|VlGMp>Gc>K}XVmxHzd@v(Vt^Z7nEcf{zFzHY#%3k|2Q
zWMd)qa85KP3AHR^KDYc}7@-0mqhe<t!n&lX#$*d5BN_V?Z^TmIxZSsrkjS#{iV(u{
zq*I2MGgRzY?4cPIg#H9qc(pC5<x2OWtMWCVX@FGA&O78`llE-?lhe%FrygJIiFw3@
z(rOic92QMHjI5>>4CrT>^jItnN$rSYp~cg%Y4@_4<Icn)eSDEhXBb&OMD;TH-YC=x
zND7?6U*gM^f8msb4~5+H*ZNBP^y!iY%+S60s)3~*hKy*wHBwiJ4+Gf=gnS2QI{ay9
zIyRy5VK<L^kyy<uVf)kJD?Y=9tD<05XZXNEmX7b>az^s~_fxVN=dWrx#4QZaBs>vL
zLt36*4%;DNB3M&0=rX)v{Xe<%2RxX4*OsdOP#1kliYBK8H-xpLi)W}9VInBnJxHV{
z3*`hH8TbOSAdwuQ)o3PAn(9X0*zJ)tz_%?5@CEo7A1GZfd_tJ^Y?K$YO-8(Gyecv(
zhFOCFX2Z}F5TF?T;m1<H2@;qJgeS5FO(_cvoC3lnMK3y>S++S%Uq^6q3%+)T3xoNJ
zGN?UN<<POE$K*rGXS1Xtojt9rxn&D^c5XOG+CLIen43C|xqb!WVE`8a;$0h|q__=B
ziVfCBXS`-&c6+IY7HU5p05wA7KIKK^J~vmevuyi9E<8NN?)y?Qry~Cu&I!x!fII$%
zf-azPRhN?Our}x3j#C~2c!kA?`KSh8a57Qm#nt;ZNAi5)lpa}qxpE4k(`|j0)VuLQ
zk)BLGkMtc+q@eSu%8f(Btl<Ruh=lpK?-`A-!_R7NUVX-Qb-x=TvXoXN8_pti1<01w
z3S1utsaH~eN1pV2`-G;XEHn4gJx$_tRnf6%31%&UvDnX_`C{tX_pH%4#s#El>9Q2M
zGy={-Wg7E`b0UTWQe3m0phK10j(cL=rZ1Hl9m!X^b96FT?d<i-=c=)jc%m%%`VP~W
zNzE8^zvY~qwFp@^KAxCbfhyyQ?NsZjO}wPfxaD3LiXOWClVR_;=KZfsgAZX(d>S^w
z0U^b6nT2w(rJhs%C&RIImXip-ggDO$p_1OU+yJ_XHR)5wX-CGKeNPfrzCC|qEQTH$
zm7;($B8Re(|F~TtkjHtM;cU~uWE5p=$2l<}$lvGG*4@sx7h8NPWd8#Z##wE#85X0B
zJhFT+yuo}qOqG3f7JAfkctPi2_0@A&_QZtf2lc07CqAE&Zv4`AXqEZBdj+GdMzv3s
zLEok5BlF%vw>jk%gKMhVkC><Q4qF^ghgQc6#>BYCYN{(ge8b`g&h~GxCIUOvlZ4R&
zwE8?C6ta9m$G2d3UeKx{bxAgLJc-Y<B=(!o1tELpGOtfLRGjqi+6cp6k@9?DL>mB>
zC??0z8#}-@(6lR{e`FnMq3haDg2z0mi~hXBiiXh7?y>*`KaRrAzgpyf{SDS`%?b|0
zI3x#IHos&j<ZYghfy?D*rTLOq`JTJE3#s1q+;=ave@XQSrGJd|X);p?((%ngEL8Kp
z=fQU`{J9%kn7)N&Uo#jv#v7CPsk0|cBwKtpocY^7dHN?PlMShdN_X_1meQoImNzfU
z8&yY*xGyB$u|7W$J-j>?9WsNkKy#KqlWG|$A<<soBYrYy7=tL1JpaDgaPZjqX2sTT
z+q_{DC}yg8kE$Y4wny3Z&G4xCoZ{QpS2^E=#qD8Vxu&z<j!}X~Fs>e=P0aE7`fjNp
z5ct14|J7x+!~p<6^fTNMW%$@ewc-Oxs77^<?pS!o1z(b~wcVG<Vec6pp5}{U@6#SB
zL+I?wO}K3lIx~oI1eSp|%^J}8p@0I98l)agxej_QE2A|NVNtgoiD$o)&=eyaC#Y|y
zQ>1`6lRu7b0czfdy(b7XcJzMW&{@4fx!i9vOl0cnzxc~37Z)bE!kwWAP1Su`<P-Ha
zu*91{wGBf(jkh+f7W`sy5i?XR9lDuZ?s?ih&U*7N%&^7R2rXm{DZq`TnV7(=Wyp`J
z`!6X}KadUg&%IY2Zm2`bd4F^N4egJgh{BI#9*5DSX->Ccw=FaK^SZzR5UfOYrH4TS
z0S`R13iK0dAfzl5sE&1Okl=K-O*FL}>ywrnf9zk}(0}nv;cZE<aa2C3XSm65A!!}+
zRMwSkfzc#20D9B^;F+eB{Iupm<yHBS5)&cZ*)9fHx`D2m#FHuEbP&M4fAlLj>hFS(
zhQLmrQu+k8(h(f{ERGaQv4{02FBhX7%sL!r?vG4uJ%xE)l!=((;b?Zk9T@Ld<);`m
zq?;u48j;$-kk@>RPgRpi2R|KBE)(MT<LIr|9{31+dbjEAw|coQPMaoM{Fe%49o}VE
znGt5nr+RI=l$A(bu;?Z@={caqT=WTBQIabv3mocQlq2PytXhQK?eudIex%2A^8tj^
zSvf&fsUV4?RAZ%>x!;8hjjEjI^fWme*edeeH;ntoBewEu?lIO=Y98&3*R(rSYplyl
zmzKSYzRyS5JPWG6R~hhf(0rq)G`?VC$rl@;dYrJJm=#=_!rW+c1^<xGOuT1a=a}co
zZpEAr(utOk#8W}v*Wa&qJ*gy6h0R-@DN9YBSC$*<yOo)7f9Rq2UG-;PLNN(<3IpA{
zL;cp~MW=$&&yA!O?3Cu1)JE%G@*-p(Yq@(SR@dxYsn+bs;!-KzIH1yW#RC1v(GiqQ
z1P(`A#ZUIj@<<JABLbGW9eZv?$VblnsJ~y}@%m`|08*3U%i1U**?c-cGA?_rJccm0
z9bYL{Q{DQSKfmGpzQC*?3d~APHaMO8qsZL7%0;YeS6^aOqqb9eLUwtt_dWWKgq-qb
z1Nz9i4-y8ttgK;3L;9KOu#UX^H)X{;EY+edk~<vVieKBfEAMqvv`MO_Fy+su^64LT
zQdI=ayH8?@3Z)-=4X{bF&ou?*)oAMS_eY}E7uRm<?7wC(<+wh-Ub~66s!TlDaE!-8
zEq89iCZnvVG-28J50FWHT-|lZ{|-q1dq4_%jB3C!Q*}w`jQ~K+oFK~V2wH^S?kf<$
zcZJ>8koZxgKAdi6ZSZ1Jfv3J68<6%$RfB}B!<?MM6@vydhUNP#%DQwM!#9lJEyWIv
zLG4O(s<L@?>o;mSZ!6tO*s8pD(hnX9MmdL?((-D#eL8#fSCO(nval(hww%JdC8Os|
z&G=^F@3KOOH$0#qc&4T-jA4s#i^7~k83tU?lE{!^M#owK%Hs@EAiMD{ws~>*zEB|S
zAa~IV?~|8#T6j1nQM|Miss<!PSz=spM+Pa|jWn{6O#ukt!xJr<l^<fcwSDoouALyV
zT*}FR;`3EZRmUCf1tytG(7!^M*IoIL*ZYLX!u9-86NC`+A;==XzGiJ%Z)xP2D_Nad
z_NfjSget!^SKFGXzHH|LoKaPLnBIv#aH{%Qweuuyug|C9#!W@bp&XtiXB+=F34u#@
zIi{k-p5nheVC0Tg4uac&k2*~?r9g6hiYI15=#ul${cGr_&*uerqt)=rECpIw`o5Vr
z<6fnuj%uyCET(DMxCz}-?>UqQJ&9E6M{@0)qIV}ny#+Jqam+;7NCD#7GK-H(=YWDU
zF}+lj<@><=aWk{<GWHKw>JBL5hdG{2!Vk=XNLE=e&&Q@w4n4#xfJK~*Ct{Pxz8@;%
zZAYJ6{2`EYb@L5tLbd03R|L8Q(m-HAUKpr_nEH4(^jgh%^KORQ42&oBO!o}=6dWG9
z0_|b^eD~%dj%)WKI|gb)C;<5A-wfH0^nF9I)=VRFX1^UBbR^0w8Yn-nJZC}ib*ZVu
zBM=CV$UHex!RI?_kz7LKSL%3F`z08BVT)6ll;o=v#CQj@83{{0J>>N)?VF?5kaK^Z
zJXh*{>HDun@<+y3I{OAg#8G$aFLKtH{58^+u?>fSY{_CGFdOF((CzE8-s^R!0q6<>
z$^%`zIe=aZanjXqPE{`LD;O`Pcx)_aPCpD;*m-b}`s2v8+MWu_h?4^92rR3*ak2qo
zx`#S+xlmS6S6GZ1hmN&>V_*$X8;1yu`rBs#0>RA!O}>Qw<V`+7e}<-8auTSz-T``8
zQ0*CW1Biks)_fzP6!if72S7MZ<47q~kS3<Ve=;QQ2T|Jz4XQBu7<4uf1Pt}gK-aH(
zU9kof_#T6bPwc<n`M+bKAGK%5n~j&L(CjLtqO*$0fT2sTb-4}GXO!FCAMp#1=S<;y
z=;tXjO&uJy*=cfF8PP8mOxc{bXQrO-NU|0nTo>m@n~kx28au6YJp$H73TwDgm|^ct
z!*8d99Ca|0dkIjvA1D~yiwypJ&+Q;|?&3NJt^mnFl7AX5yBu<fMr@58g`RF*mk{Vp
zYM_B?^^Z^rf<B5~%QRmXrJZ2a<G8Ta*S`jjp$v%TYnxUUyiVH8C@wc~_tSVxy*~B@
za)5f1oEUNpHI?S%?BM_<M!aGo%MLq21aJ;mCL&{_H9Pq`b05Oo#_VRpx8dg%n`e^M
zJo?&5@!r8TM=KRZRipYPtJjqx1vj?Ia?zzody>6rl<W>3#f=L$_^xu}h9K)(r)8x4
zPlo;bC}IVA36-Hw3#!kf2rF|F5{d{DcrH;odS|{`G?(DGQTgHkbGw7fTNRts3-f43
zYhyh~y3}|FwsBL0Y<&jWl!?j2Ob7IiL!Z`ikbmrTW${VWB=0jFsk&JkHnwNEFiPw#
zww(wYgFcpZd@|E8upuw%Ewt?+n=9xSE&jf1+`LjRhRjs&j~-{E)K3u4!<H<Z5}9)>
zgg->xd8Crtmi>G}kaVf*K{WfgaE9hbIy7e6ObvPB9RP%v$UoW0AM%lJ@yhMb1WYa>
zJ(x8CDoQMgrik3wcIhW$Ke4jUjwDT9kCiaz0MZiiB3Gy^zQz}+#!_L)Ds2j72XOVq
zD2;FIu*aZ}v_yeS4BlX#Ot#`&O&>;lZU3ViqwZ*Iv|?l3|AvyTPBz92t?8P#n3n!z
zNZsH%uZSjSYGPHIS|0ipzOW@#CEYeqJF^=?><tT>Ut#9+m6WuL8a>!-ZDuL@eqD$p
z_L*uAG_1HuN7wu5f(~}~cD_1pLY!Yfu03b^Bj9Ofod2O#iLN_owKw}g?vct}`8}0|
z#1&BuMqSoFDBIiFjU(pe$NXeK!cOGU@fd!jLB&JpJ~03Lf;@RazGq(M{W@Ii?##Tj
zRSD^G5@8?D^uQOA4|>8_2Yf%bJtO^Zl8DLHPL+|FuN-iG{;dPtE&VSa!&A0v3(0ZK
z$+&SLN_*b$l@Vv8v@_>RHorq`u27?n2MZ*=PcT^Erdh~N<4~x7%>%kcHY?ua54!Pu
z@8?1zKv-M&Oj>YPJ&YmsCbm<%N|1MU+>xxLdyyPnjVww=m7Sb~8?5hTFjxLCv7~8Y
z<pHW^?Co>H6xcH9O{1q>$1ujn98H#&3Z^<$bHv&%p^wZ72t?Z@W|%o<M>v<1YlhMl
z_dCLJ^+V^d6S}zTpYogHZ8LZs{pKB<2&G)a{S*1^S#z~4B#B%!d#X#i7v_2l&rgP0
z#Iyc<)HzD8H}x!r2WjjVxk+48*F6Uk;d^<4F;=o1^Sy0u59v)b{i9_8FHL|ePO#R%
zIR0eVgJiLJfH!go&l#0&|2`jE(_68wdPi`rU|6pwuf#_=oLR|V1U-{dt;0*dVq7WK
z8!Y;ee(;@7zgT~MzLxy_l7A@m1crVSKfX+Bgsx3`z_nl<IFcTEO9mc_+hp676zT{#
z3y;yCuwO2r@1F(LkgL05NdH(1{QdVmD}*qpJJ`18NY_(Q-K3@MQz~+!Asqo)a0$x7
zCWK^w30_i=p$Oi7y0ib41!&c4pI_U}uAcvzUHwnL=L}`Ovd^C@_w!13+)u}NGY@|s
zvQjTnyX2B##m~*wR`(oG@MBCscXXHHcYBv5c*7sC1aF$5$vM#R_&+Ir0)XOof)h@4
zs?iQJ;U5HSytdsY>TI+R$d3<&OVMRDs28&UKT&XW4+z@tw*O>^gk>A31ug=Pqpd#R
zIBHZg0&$Wia2)cc(|L>l$MM+mF2_+Ca2%U+``)M3djXE)uWK0mWH^!yIF3&@0mt#>
zF2~V${<gs%Rm$bRt?~Z%t6vut`Cyf5@!J}=(G8-1GBi~HL-U+Cj?(WJ&u%K&^*0S5
z&SekE$iFTD`OPHO8M|C)z;XP=hS9%S1Z$WUzJGA4eMbz*p?(C%MBSHn^r`?)(dVBG
zpN;1){W)Cm=2bWVc-G59f3&0c0YypUCxf26@frtm=eOY$hL}1UY~C0;A@v`_DRKM%
zoJ|S&ui2DpzYi~Duf=E0M~cxfaNr?-D~V&X#r^{Q8iKF`P#SjQ<8G;v6yfn`Udk)#
zb>ucJs_^V<x#Qnwt<Uw9#9cb3j6JSdmUv8z=s11qgOG1zmn?V)j?9d^ZnGe-06Rub
zK<IjT0oUv0a3|0XSk5`~f&2-FhaJM7D*5{}Qo|>lb+n$H{aDpM;K<`F{8Z`U@Rf^O
zbrCivXEMN7ZbR~>?N<%}rDm>1C^f}QL2S{LyEUIjhtAXWiLl1I%l8Wf^&e`A22}<V
zjGL9+C9)lre~R8xEC8P_O;Mr;rUJ`0gzsJlcs8T!32W@+P5MbV*w!nak`5bUu~uzx
zQYQ-dnKyVq@w+^F8iWAz(DBz)@W@=n0s^Ib^>xrG==c8Z?GVxz6p-M4lLa8F6O>TU
z^m{)Y{l|~Q2Z_I3PBg_H%8dSs01VBOu77^E)$cb5?gA2on>`QExtDAIybOi(ZwhC>
z16%xRQ*bZ8m;i<ze@O)o&sBX0dc27Y%Cy3}JFI1FdHe7#%J!(#u34fSHjZj}pGP++
zP%P*{a6n+3^!f9`!PR(>%4r8`Poc+~|9rPGfno<`LVq=CAyC|ryC&#E|7Ox8m7sqc
z@ZSdfa})mi3H;L${Y!4)FOfzYaFSj{rqx_!%5vnogyITUc|~zG<uSHnA6YTp=ZAWY
z=xm#&=+%A4S^PSXKN)teB3ZvyGm*F^^q-Zec(){&kA{!u>lhSPxIX1wOA~wQd@ojB
zWHT>6#8}Cc%{}noMD3&8aD5wvrWP*{L76@OpCYK%w^7qPr;}t8&a<aCk8#)v?ql-{
zGo945F`ZP4lI1hL{Re|01DHzugJ}xB*a8><&<6O%>Q9D@i}-9<0?_JSxlWb+$>0jz
zHj?m*N0kTvWbn?K#R-ybdhhhZ$vUf`i32nE2PMl}{U!V|*5Pe^YYV#SlF5>9%rt(~
z;kn#NT=ls7T$NrJ)yMCpzE^b@OqIr{Ci5|_01cNypo^_~Icr7n{DhbR{p8S^Jv4m;
zX+bg}P}Jx$l+ZT4gi)a|3A2Orr8)I6)z@LfhtWgrlQY2^1=w-)L7(XnKe-eS$>aw=
z8J4+5xS?0gZo^&~iPZP)$gQd5DA4a{5DEQUC4^zZ1h$Fg!0>GntihGlK~7%j32cat
zJv&i0AkwIDT)pAQJC8X_I(9NTiZ1hc040vTV<A6UOsY-fs8x+CyGKKG0Yn^h>ogs7
zw97HH^9eQD6{HHi6NA?tL=J1sd)X8(qRUc5^jhUgtX|C|V)i|Rvsp@Xg_Y{F`3%hI
zjmtmvbS%&daDNWfo-2XwrFb+bp^F|$Kz9`^5)w8(V<{hapeZih)*Z?!yzqU;w;=K`
zuBTY4R{1IjW-E{0`Xzd0{0<#^4NhM>94mSWoN`%wN!!}Bpo|je0nV7DT{6hWP(bxY
z%!yLH&?lg8&r_Mdg0?}>m48^M8Zwp@sO(IZ8N)xSzLSl-PcAJ;pJ3VBU{Vp-)^OES
zFu~tv$}>Xs<rg-Y=ZrEZ_r&VU4ubggA9WE-kh8;snrr4_ow$Ea$4^gTDvANb;*@``
zprE11YAe|7fK>!7rhn85_?<-`^-!QqBdy!o6STMfQHSBLlnqYKs=SC+ORHY^jg9|T
zbbf`e{^td*l4Ulbq+4BqOE^DQW`u!Uz=38+et?#>L^i6`!-=lgO||tU2ergek~PYc
zdRlJDRUULWlPGXEnAf_uit)W{&#yOZICK5s;NP=^ehTydR}7p><*LP>Fr&Nz(5A*j
zx)?}$JQ`!>2K=J3_vgVR80Qo{2VtCF8+aNM4ac4^aWh@+kQG`xcl*2X;9LoCc`yGb
zm-l!7{=3v|!2Z9;9ByZkpvK$r`jqf)dT1}lFJtN82C(*DvIaG&(z)$)ZGfM_ui1CU
zuMifuK=|0YV+L{&5PW8~ip^_pc{f9fpwTmwk;$&7se32{PtrQoW`nhS?kmM<hzcpP
z5I=@mafA|FI8gMXZY)MJR{HOA(iS*bbG0%g){C8D>dN)+orNphv9mQhZ0~mM^O)QP
z?U--)F9cEtp+t(t@AVcKR5j%%Lldlxz>m8N9kOV~lq`qJvItM8%sgEdzoZGB7`L=&
zd7GK4+U<sF%XQJwY)IU?3h>MyE;v*z%Cc_*`Y?!m<I*V40Qk;31h-ahg|)&0cVmw}
zLP@Uw$-q->li`DF!cGRD2=ypli%7@C^b#3J*}_NKkel?pmT)#UUF?vH=k@WcYV5ZY
zz)8gbJoGVY0D?5rpqtZ$kfrwMb7W*hmMOO<KjRivuNXV-pIS!~OD1hL1vpl_W>Vly
zC^pw4qw^nhRWX+o!>pU+U+jtD&M6pR2>KfBM^x4V#LZ8~=2o0G1KX5#J}*j)#Jn>1
zl_}J^<u&8DmCQ(pv6lCL(?Ob?+zb+qgVmDF$a_c4&NwZd{n#6C^hIjVtsJ(s`vvab
z@$q5VhAEnDCwj=bju|7R3;AoM`L$W)LFwB0=hfXd^UtML4#M?|GYXdd7xNHvXkKOe
ziAascLLlDxoi{%rG|zH&*=>9K%1SFUgQj|DM2J!3w;7&^Do6<C#3EgAXkI#;FQ3n>
zCu>J8<K5`v84FEySWJT$gXSIlfsdfNT26Pu(4v#D^qw3XOMxJniO5eDH0znRNIAo&
zq@rr^RL{mg-ZFllTho<O+x8q?*Gg>{v(B%2lqD6ddYfp?xs)pwa{I9k_Xf*C?BaA4
zyu*sC5EgR?fe|$yeADugK?-gX(jXyYAd}1^;pvqma?YW;|ImFene$AJmp|{b&$@}1
z9^Y{%mAAmTFJTT1)51PATnREN>V2LVDS6}K$@h2NKR%B6ioZ>c%`$Fk7}3Mm;#lPd
zwp`qM+b2!JUfVBUUNs-Gz{L4eH&i3?!c%mN!y7kDF3gUiqQ#GEspf9HgP*yg{*9U@
zFE^(3{W3Q;Y27unmbV*i&*%Eo3+5plk*FJF1V1kEJ!C?sOtk5cL?Cl)^1+Q}HNQ)V
z-dsN!1XKkqN1A?E>5hk2oQvlEHuXkadQmGcLC&z~#eITUH$h9ui(V-MOZ=qV^Q0nn
z#yl6n5g;K|Mb;?lhVwfKa0&2{nr6ae`4yC}X7Nq>b2Ub@ItlM+bL?o0`n+o<W^LNu
zCf3cRB6^h7CaU0%(Z{Qea@wO8p(o~8^UMo+3rm?FHZ)6*-VhmFa=P=%vBU6WuFgV{
zckv+H7?GR*(6;M5J|SZ3-LkFoQ0c(Zkka70wSEcKfCAQf!f`38<0LEF+=k!rrZ=<Y
zv5rOaJ)_YVb>10^hadPt<ZJn2L3$66KCA!~|76=ME~jN?be^r%Gsm^lHfd5y#5I$M
zBk_`<2PsCnV7OwfECr}b-tZbd)m`dr*q~n#>h6#53_B4#lV+QY(a!mn)O|Y(aV|M_
zA+4tDC&L*xisA{+>{X9TeRqkOMWyt!+jC#HYs0oghwBuxLLRan^2uqII!JM{akszd
z6R2``!y<WEh9~OQAB~x}G*7JwN5G6HA_es*!DMLj#isG=lW#M`IUmTC=P)brD>VA2
zoZ4en<WceZef_L^>&xlm?gN~Dj^6$9)bhFE>Dxsg&pR5DwMGVAl1wNSem@yxsb#ow
zpZ4Y?_@g!V9H*65j<=~wZr#n6%Q1oP>o(7nBPKD&95zs*v^XkXj#IZ}2HGYkSqYJy
zA#?HCQ232oP`<~S-5ycb!qy|v4SxE8wHK*eK=e<uWJGdQ!u?KpL`sge$oI!_eT6n#
zJz3G&@?Rz@%rgfnca}yfQK^246}Fc(Vsuxtj8!hthH8`6X~@!zd9h}y(xnIVerxfA
z^Ha(S4*4ZBj;mVrYs%4`)thzJT;-(1#o=t9P0jLDG`Hj{)ae>utDzyNoZ6YQo&ATo
z(2TK%6!KFQMuvDL{TF&9Brf$=5h*V<Buro_$v?_c?2_?T$IG_2Aw~x9vlc!~w@ICk
z$E7$-o5mq5H(kW1_1QG?tlK>TrfsC#j&Q$dJG4*W2|4ULdFHV_nKioFHcv7<i>m!$
zOJa1y^Mvbrhlm&k!?Smag_k4P8rM~U2XW9(pm$j*)+D(Av9j%l`Zb&K^tP-nS43vz
ze_DP$mgtOK>)>dn^7ZF?b^&&NO;K%4U;e#f1O9<`3?JPp;G;Hy4cx-Yld`L+HSbI*
z#r3t9$-&rQ3(F8KtZh(J(sYr8aFE^9fgxFwp_s~X_V|%UEDueZcSPM@+n{VwRpzu4
zKN+NCl35dXoQks)9;{T|FLxQS^jDEW3(oWVgReYfJ(5|cHsMCqK$gG1tZDlJPpih`
zitY;*5%srY-I#=uw!B+XnY)#3^QcGqJZn@Xw0A7SE_?FK2$i;lUm25!jtY^#<8eMY
z{K5_BqwBYUI$*&iUHO|aNzD_|Lg%>Cgl)v^#e!Qd&E9po9S;+FVS_->`qq2DyjDM&
zBGP@!xeZpE@@_J-7Cz8QKTIBYHMyXxy`XhbB|nln<ol{@@zvxhnP%gtA3`LbPF8-g
zG%O!28Ob;f>OWKkQbxO@`EAwn6}_kti5`m^<sk|$apDJPZxpd_>iBa9#gG!(@ea+N
z-VW2vy%`$E6dzi+>*ga%c7ylMT~(yN=*u>tdc8Tz(8aDz&zAF{CE{tMguK<ELpQCF
zcRLMRs~=Q_sE_w5<l#xMMuSMq@w={8Lq|Yq&aU9Q%H=-&Bq!B!+KJ2VvS$zEGfC;7
z@&%&Rd|RkzAjkSIsGb_CLYJng%WTP>@qtYku9G`cL5AcoN8_!eWA*814S;3@6*@gB
zSLoKd6cyAPV^z+Ct-4aw>B0vlo6QY)mNTNs839ZaAt*Te>^pR6)&+V=fH`D!1G*&-
z{EAeb4n3fq^pioRM+7TRZ>GG<Zh5-K<hqe}5-oVm5zTS?$3iL}+O9Zpn^!d-vb6(g
zp@y!58jtD}1^sG{9hMvrG|>eLlOX<k1oF;EF$<~%%gP8vIohW{V?Ar#_$LEYa1-Du
zxCS89ox~AP!(9s#>nNjIt8LvQS%%U|s~~N`uHg%@?PCetvza7bICx7|%udUC^peh!
z%DJ#rICNr}y5plnr&}48ZZ^dWq3wK(YnmECQC)u?yEVBPIsjYwkC(a<p<%NM-AI|!
z>pB%0tL*6eNMmdL&}!w$om~G@(D&){vGm=xwkF&R-Y!*<8mH#V6^2wJ%MluQtBWcA
znx&7$ryYeuZ`lYrvBkEx+){vOH&K|tn3QL{6ITVWSpikoT6k06+>~)%Hwv_+*5+HF
z?Yq9Ul`G$`#;>nj++I}4b9~SaKZjJ?kp>AID+y1l2C5J@OLt|ciXqhJU~8j?yXY_4
ztux^1<LHsOZK=;x7(aa-Mm3~<1EAAB_unM(b`@zX{`q>AAK{yRsPGjfcfztUx=E@7
zNZD9vu>D0my=MvXdrJpGZ?HZ~whrA9r2u7cItz$D6Lm}rQe>rwn)&ioR$o@K62?9Z
z3M%{3C6Yt;!K(fdF@Q1_=hry5451!{PNUY3+-5f{!I5~-F>|+JyYQy}4laQF3L!B1
zZ;%18h1)wipplihB@nOX`x?ildipva=novZ@RS*t^hqG$5p+y;z=yF3LwZet>f%xD
z8&YM0&_QaKj2$<2w605EyEBtFL&T$PU|36yuDc<bdWPfdXKcW_F3OP~7}qjO6WC-=
z9jcW8;1madcIwS;+*%0ly7>xrPzhjx#|QumEKh&lV4Z6j`i9*Aev%qMF$y3vZv^+P
zhr^Fj;FDgM42QCWaumZrX`HQhnkG>uqra%AL@Cpf+OywvgyH0#I;L-ZusWE)xgxTD
z9p(}JNYMb**k&x!zvTf1(a2gbdcQB-)O;=^UZPPqHd0*Ypl3kGNhjt92tqX@2{X=B
zhl?68mk(TW0psKNZLX!7kWepI_~n*H&&SEfwiG5P7`!}FyC%+`$P&e%!ZZeN(x+HK
zS@M<5Xo<AQjd0?%u0^!N>IN-g+uv$<qrlsjj~pd$Bg~_+j`DDT7cU1=wRR5r?6eHk
zM1r|$9bm57P(sXm{eb_RtM+6_<Ui)B#e=zO-!<#_;Z6BEO&HEn>|pLEE)zwKXC=5}
zK8DUtt2$)4T=zS7^5HGcPQ&&Nr(d+$m<heO(Ow#U95u_nXVcTS97Mu#be^3tOxW&p
z<9ldV^qsmmw0%I<;wWm)P}w&b;vzq@lp4Rv6>`5Cv7<VRD-q&V^3k{wpj6*c;F!up
zD!_VYvz%G~1H&Z}H+-W|=q=lp*7KXnd*ZA;chrF9FpDyP2k%?1A#cQZTp(L#y>%>4
zR22y@H2H|Pd-e7rbAtC7x2ci93rd=XA4M<h*@l@y*pz{zIVklA4Dr)1qE+ocbSwH$
z?bE&}@Awm1#a>6h6Pw-KQ$wmnx1PXB2E<bp2hUQBy<(Oo{aLKPKUdLHXvy%U%3l~k
zUZYiw9x}YgYmMP?_;%q)Ekv}v#jP$-iA{|mz=(utfs;y9IVkeY@INI{o`kCx_tTrE
zw{Z45Hx5Z9;+&R-aeQSFlb6BV1neVa+W_o5q1b-WW2{DwnXiVqNsiYlGYgU9<T1Aa
z>)?z!8`$YP7NxaU>&(m_*Q=Y^$Bt_lke2J|5)>e5*2FgFhkium*^du&^}I}n_~=_y
z*Ko<-`mCJ%BhW(A=80|6IngV3+n;NGX4np&gf&L*g8dIt;gjgaWuhxZW36>v>cbXY
zA2apVIn{1p__d9ao(juDRn7t6ZO@WAV{*5kcJ!~sA<L~JRqd;6a^J`7d79nCR&mx8
zL2!)R+O1s8qd8q%U4$t?d6cW`x*E;LoniwDHZ~l(lAXm8hi0vYScM`UG8|yQ6EMu^
zAFq5g=ye<D?Xx0`@6&J!B~pW%^(Z0Q6Xk$OBG5WAB-<srP7rl8`-E$g6>kpUw}U=8
zvtKYp`$Ar7M-<Dr2>rpTLB0_UWEOG7&_HExUw4l&kfi!to3ZM<Z$3T|ZQUnWS}JyA
zSX1)-xmY6sw?=B+qYHaBkYDDQOB|;!8yd?*wwye&kq6y_;9Y*I49S=WV_KG}XZ(nb
z7E5dJvKIdhjaEM8**e%H(<yK4YdRanTbClUW9t(%UjFz=nKfet>4xI^K<LS3iGsxU
z%)N(wwHu!zXTznhMr>IWcKE`dRCD=~`8HT-Db*^O3rVsy)E@Ki!XchJt<LwUH)Di4
z&IVPRAvtK%7us$D@$#>XV3g}P<iJ8A<iHvsI>|d!nKhew0q8J{CE$tGCmhj7-@aYa
zTskhcu=f$$m$}gBd;SrQD{*m_s)?k~##9JJV@&^<pIUT-K9f`4wq7RYxGfo#Etmej
zr005t%i}_bgZ!Q$nG=P#?oIA-U|dVu1%6a}jB8L|V54vwmxs~wD(SF7d=ydy1QhXP
zJiUd_m8MbR^wGuEXk0}OXeS0e7XkXalc9}L&r%t82U`Stl9K71(L><0D+fgjE$4Ru
zp{|KjKN+w_)2<`JBRjQlsE;8Jkj({>2;G>Hdr*m%#kUW!cs9zWUX1(tAr0y!T+27#
zkF<DMTch63<S5Mfu8+)3V$ty!LjyQyKQ`P(+J#Piq$N}vV*7j**I%SV3<yJUbUxMz
z792GI^#yb6xhKS)b2Y&4LfWi?*}4SB3Z5!|F3bKW!)eyRmPyO&16>Vb9@GqWbi=3W
z(~h0_B*H{NV0x~THQOtOZKB-<b++=TNq~+Tvc?r46~t*6<P?}i9<UVT*0Pu1=o;<`
z9i_!cmYqj2(6VZeQw=LadOs14DJ$i{zqs>9m_I>0__2MhQ2_ur97mYrH&hgX7Tu|T
zDAfKn*pFfNHCPOKEE%(6yF18uNpQCmoU?uScd4js{V`PTx+sRsl}=;L6q_Oh;#U0N
zrL45JpA6eBv9)QxYkr2o5xwi`@$@e2Y*#uc9)E4zrI`FB=PTF=KPpV(YC#{T#I}qt
z;U*;@W~9;RtS0xjR*r=)?Mo~Bh2rm7RDI-&uYMBv_VHE)cRrS4HF_Q(0(o{D&WuP<
zvTiq@iS<^TJlj^P;u5w34yND!A3E;slSO}O51Py@e`+r|-QVHr5SYpDw$}OyrpN;Y
z-s`GBf!7r%@FFy*lFRIXhHzxJ$Njig03Z$XIRk={RJt_CQX!YH+zxnNz0x~JGqUug
z&sHzQU;Y8P%X&?Lql*F^BLr15<;Z2E&3<CJn!Xa;-_(e|m?MyrFIH%<C~evD7b`<Z
zRn@>L-jD(9=>rshP}eVg@qzTBU+b!ZWEp;0K1%f6sX6s@1OMDvx*F~JhAYKCd)_>F
zz4@8K$3Dq>)s=%eR%tq7WixlfMXdGj+mw3Ic}N8@V<Q6mV%K}~*E&?iUzC^Y=En75
z4E^I;vEXqD9svrvr!k-bE&r$36P?Q&h|`Abq6X>+yE<2yshsiWJ&ZErR<yW`c+5*~
zi}pXzNq0`~UapqOoHyS{%(t}gO-g+kr{M^F+0brw?b(5^ZV!(61x2`rg>`*W)C~=j
zWj%J)acY70iFCb($F#!gapgrmts&_+a1h`@Qt@dCNF=j$fet?(kVrP~{_RUJ_ZJzZ
zR*vp?mSPv2RgTnD=eKud<l}TkM-vA|*5#b(43xciqo{4@(MAD<+HU@ek1vPbz3QAQ
zmzrWs^)X+NR2It^p(M079v@onUmS}aV2+k<{`@^~p@cu{tcH+%*y>+BN54HMh|A;I
z_2w6J3Yz1Zjo&XTMy~bOq|Oc{j<G(<>=PiC?5F(z5}W?Ul_}2VH<!LUH*9I7qKF;@
zj_PB1fU1fMd5FLsRpWZH7OAeHH))<UEr*DVd3_}&sOGxNoo5WlH({9HtRMXs4{)=O
zs8LX0WSrB6j>SVie)d(~zUdr=)k{e*=b>#Y1aAC?--wMvEo(PbF{#qjOk6A1S_kb{
zKs=S?xHq@(2pA7=p!;!)_A22o*8csb^Lwjw^P@Om@s~2vSLH9EAJO-H^+f2Zw_i2Q
zDoG}Cc9(wXx#}vsIT`%AIdtUq-8=+lKayuc0ySxl>o&2vw%}0Vuw)mdl#<Ro-5hoN
zChKx`f$QJ8&xZLIzP<rTWjlgd3zR4qw9|6wR_aK(k`%>6fE$G40&u33vwRH?xFqiO
zDDXJme77X=5k*SiRz7LN&7pA%Ox+l<__XaEioFz4(X%#Zj!=tT^0*}V#WX${{i@Z5
zY8~{U1tg+>W5*jc7$!s|4Glje&i&ACHVLoH*{oeg(QExvLZ8q%eJtv>AJ=|ed%7q&
z^v$fgT|B~CS`a|aZ7Ry8m7+{ZGLQB<`Jg`~=?7OskA`NN<?-Iuo-ga{IMq;Sx35Ad
z(5y0beFpv8)-OT-vN8JApPV0#J!H{daiXy=+gqU=wOmH<9IQyEGLg;&<(fOmnICRa
zc}H+4i!88<W%h;QPKr-oQk_rduxZ5Xr}9s9+?~AL_b@?zQ#9LMHvh*4!Qs}0eSyNa
zsP3zbEgT(xV<(!?45Ouli%p345bIP8d1$Hk5Vdn-r|VYF!(<hvo29)a3l58m&Nj!c
zyB;RqxOIVnqfwW{R3|es`W+hDM=Sgh>#)CXV(-#Hd$!P({)(NpH$k6I4aomJSLo5u
z`#$#d2^~bbqEAjq^27N{qZ?kh>~A0z`oG9~@2IA_bzc|@A|OUVdI<^$N>jR00xBRP
zA|ky+MWhKRHPk359U`D0AVfr@gY-_MiwH;y5^5;YA)$l-A@4WYd!KvGmi^uLJ?EY?
z?l;EwM~04}Ypprwnrl9DKF_Z>zoh8nYvtAY)(fA0va5stoMdU|t}paBXTlm7z#g$Y
zQ|Dk;1IuW`Po}mx3H2lnEb2ImN(Ti$YKYJhoJO==k_F)M-`C##FW~$Iol;)QBsZZ6
zVZjf-7Lzv@#`=~i0^_`lU&=m>sTVms6p40<^Ae72SG{s}i!<p#fJgolsVB5<tU$p1
z&u_1hn89bP@ii3dE;w_^qAo9`T-{6fuZ@kNU&t?%D|WU(@GmUs;VShVd{bkkC`w{i
zi|!7KEPD5JV)Zt`IWk3oSl-kXsG>~pPatTwkZ`CF8OGhSIEVJ#1NWj#mOh4ZU8zdC
zpxQ})TC3#=<i=(qmOhOkPth1e^}a8Aq5WU|G}WUyd9R8dW4-@MGhO`D?f*W&z+PEs
z@8{+G=LiIsf22H?T9LKj_;m^+2d;GI0p+EC7wrAV-{~rl^&mD*b@+pTKMWvOx8!Bl
z-}nxJbKVT7YRT?}=BJQdh2eo71M(vEEb@vznSGD8d*r2~K2@z8xdFDN>n!G)U!Yrp
zeB6J~<VzuE2lB@NN&1!J*zQy0?%{!x1Ujk~UC<vi3lZpT*@Y_Dax7>R@*EE+b9n*y
z;#Sih2DReWm%Ys2v)xz>Q2GRkk)H`^$hl*(^QCLS^)Md95FTBRK7fGflZ7UIk~JH*
zX@g>2Ws|ATerSoYbvG?>I_emwJ``krKgoG0-DCeGEN!>>xx6=X7CNY>{X2mPSn9-4
z_az^7^p1xDZ2{?1eAdWZyzWuKkdJX#W8AgFF(=z-Vw)fRzBcU-+Un?g;JzNt&oB@P
z74}xbFWsYasPVT0^l2p~p|@6d>O1jgSK<DiPJ%N^eNaZE4Ti>-y|z!XM(BN#xW+D<
zX!4cS?Fs7g^}cDVwPnE=)o1S(naY(WijcqW-)#3y`43&6T{SCvjVas&|0F`IKH1AT
zZOb3hms(d|-Rs|bagCWc9W(3ol2Tur@p|flX!Z1$L#)0g1okhMmk281cwX1WZn&-Y
z`G%^8kG$^qK+h>v;J9Bt6>~ae<z3<I-MH_8Y5M}&-@dUKO~T|jI-rilWd1Y4>Z+M^
z&UK+ar~G`On##I+>36%7mRGf(UFZHe&JtBRBTcPGACxHR(A0_la#8CO{0P$n+?$ob
zrPZO(GF0c#V=dvgS9qk4zX$|l{aa-Ce;a+PiSk8|a)d_q+OhLp{DCYfkcsp7TvuLG
zQB+wzO7o;x?aX!7XPg#46*yVtp8>ZWM!dI!9x%U@t5UlB{X#wRu-MY6Ic6DK25DA0
zhq4J83FA1iTMl&ldi$@oDPlWxGCpySk&mE3T*yXIRB$2r4yP}-Y+DVNMgRA;T<M1=
ze$>QBH<ll=jeIRkm?){^q|g%N8@QLQ>@;$T4%vGL2pXBb+1Thz^%Aq&w326+(mGq8
z!tT*MXS{zR?^~?9ZGK)#8*ZxE4WiPG$>j?Zcp#41^mW}2B$u(>rneh3gNC)IRi~tV
ziRVAJ)w3=oS|&JtnDg-Vd+o22{PCRgF`w|q*KTJ^X$jLD*Xr={{7dU_i3#br?dD|%
zXrMpvGv@#{YvSuj$@=Zr^4LvpXWS!|2bM}j7LW^=xAG;c?}BJOeJf}ll!s9dks+;o
zN4sl;^2q+8BM|lL`ZrNsahDcrGAUr@y#O<hWg2*_@#@_Fs~=uE*@ghGN7Lc^SQ@;~
z%=Ow!iIoo&n`^s!dVIU^$ssg(I`$C1lhrdsZgs(6E53QNVV500-5PX1sloJgWJ$YR
z#oP9WbICF4PcfZZ?5_74gE*DeOve5^iv4}u0aO)O2Epj3`h^LjQH0U?=WN>`MN<ck
zDa!`Zg@P%+v;?g!I$nxDk9}@l7rMho;U}zwh?Pdd>I<)mN6wI5CWpf<RM2*=??O|H
z>~v~wK~Be9P@8%HBz^wk<p0kE!<oGO8Hd7sTSJ37MiYl32RMt8T0X)aG$gReQV?~!
zsJB#W$-45y+1LzIHdjZP?ua%A2^yCt2l#DRpWgcAF8I#PCIbYA%5VP`uKw*$b$18a
zb#bPqi^ErjWn40?VYeMXzMp)~rXF)(_79pO=O`tM421lZUSIuBb!efKQtJ$L+9@&I
z%c6@WdRJZY8w<R54C~B5;wuH&7D)lXkqR5?yQTBig?i*@AnvGlcQ4(%$B#Gw@=~-)
zCnudl(MM`%Cz-S-Z6fpBp(kwFxJtjX8FDey^Ar!Yiss&5Z@KsU-80U&<HusW^1Jq7
zn?CIiMD5>ql7|Zy!A@nr{R?!TZH)mAI1fC}@V(w`)3^TB>3w<7ytal}iGky+svzwn
zj+rQXNC^B?jUu5{+o9HJ>&}(*4@JQcS9y3v0w)jOEt+L@Hy?)Og3o%sLuOemL}P=i
z3%9iNOILbYLEhpoz_z@df^!-tm{@fC#j)4Q>&l~k$-bt+TU@146||u%lx=nx<B@?v
z=De5F#LH5j+NI5(OV<^u5xLs=^Q%s$)Tvk83Z#xxv)CNrqi4q`-2PTMp^di#vhHZ9
z=TKrWQ&wP7{4=mkZ(pk{%PsF9ZC4;(I=(Wf6Y{}^NPcpNG4eeG{b+Zm{P9SOM{nAd
zvNO~+dFKV<qL_v#(~I)*n8=+ir(JfVqB9`&{27p~_QonzRQa+G_U!dDp`TPNk53;`
z@+O+3&=cL!C?(MhhizZiMjJoQ7@0}~6Pjb1-wz06F6vmft7xi(4$bTe_<LJ+-|0=r
z8EP|wH}lZ_wfDN<X6V~hS@56m$T{Vzg8x@ONCwD&P27Y7j715SM~(p*JtQFt5u<#$
zORt~5G;I34pYW~f^Ak7@D}|VO^Apb|X=XSJtO|I@V5o&eCku9?<i@s9w*60?ez~7{
z?WXx@pzC*I@_&ieGAN<a&13Xa{u|O;u#>)5h`D1dr7$|Obc6SiqHRl*RgR^lL)~e#
z4o7w3s>o#IgTjgt)aS2SLjK$1e0F0vG}tE2s*-_L+shG9v^<wZ-4CG36*`;H&!PTt
za1z*P|9$~*D<O~lL34!;Y)?S2!ffeS)|;dV0xskD2=ju;z=jzwM6=}f09R1DOB{)i
zbz04f=cQf@op9Ev$^^?!Wt7yu8)%X;fO_Bd2^A<v!F&1vF;n|F;7>Y)FK80i-;r+^
zJ3$ehu+Vf`DV*INDINZ~F~}W@YN$@U^wvj#q=ekjpG58+g@S$%s_Q`JOcuJY(K$aw
z<t;%I@ealk?s<l-03b1;c>NQm;ySa|$9siM=^?a(Aq;6Tw`h63E5GyMv)aufVyG`5
zpbSj57(=cmqLy?c!Csr+7q%A|cK)Dw-M@HB?|b85YvErVB=x`F8unLi&6=D`j@2Tp
z{6v<PYKu!QZR-8R`4!<$@2EG<JiNFp;2RnEl2e%OVALZ%FrpX!%7~H}M1I|0ocu}r
zgJxe}`jGa%ARnB%2O$8Pt})7|f_oqSesAdi{d;!|)aay<O|gv04hDri;xiRdJ?9FR
znGeU<afm%V_uzoXgM*n<vClQ7AvS!iZ~vhAy)S9|CWsI1Dc1|6r;DIazJKb$N>xC*
z=ErkT>)8v<D^z`>&w78qIqHAn<{9&R0zGb#+B&@}rP`vRyE>zyjAr34qRh(GB^H=j
z;?KV?WH=7otZcjg$EMpqX7acSjQoa(#Fy_?442ICGXd5;B}{TN2GvpREt@FCtA}+&
zL;6cz-kH8V&UX8jl%^>uY#jE_B>uA*;}|{*dm~p9ipCDp$5_Qihuz@ADL2@J^q)R1
zJ90!<n6>}P`5b#!y$#H4LR*MajQK0)p0OOwzccoa3IoRedFYn&5R!No*+35n`(eRU
zGtwBvz{Afno<c>9z^K<INE@)3KFlhddVM`xv;&DhyaxbPzxoG__>MF6Jc_KQat7HA
zDs}DtLx0eOrOmHnX4d`Z01Ex|(mb*&3ps_M{9D=nKRyCV2-p*d`;ZD!;=7>8!>ZXo
zhwPf&?FEPQt3!Q%(4g-C$OCdqpeM6H7NiDQfOH%ygi^kQAp?HpZ;i<50sp^yCj$`w
zg3Gf2o+KTVnm`jY)BmPT@W1IP)Cl}nSlXuUA2iVrXML(4Knn<gCZKM7s2QN1&<#Pa
z--WG?0o2??=5MV~DbNO(nPoHpzVMA^P`2Z#;M%r5oIwGF9}PqdGLh35M-c2l6#J(Z
z{QKZ4*XsT|(#9@kR5;EEP%*?hXAY=N#rQDXb&qE44ir5Wr2k4Y&r33da)@|b7h)ea
zcGB#&Vd-mNqRnsk?lFv?e~N7`x=Z#Yfg4|dsyR2+mF+-xvqxo<B=a`;tv@21deC)V
z)H$n%B$c%^>E3JMzmCQ`%UEw@*I{LL!;2R_p<R&)MChL<T>Mcl|3Ud6n(wE47^;*{
z@-jBwxRsat)2@PZ$Q?K*uOFTo5Qens9=G4rvBS6fS+%f#d*bK|OYd;oT0C;<&=!_e
zv?ra|CYjK$&1i`sydV3Z%pCT$M@f{}*z)!=q<*JEVx-Cvk#;+*)dqVk;cu+=p97Qd
zP`UDuI-p(Ka8h0z671Gm=j-kC?h~Y<!^>Sdf)`ebyoyU}9DM1BZ=VN|?mNSbY*z@C
zjnIkD-ZPPPUpN(i?iX^5W_ee1R0c&B)_wj7N4-FVM<_YXtPG9W2aWb^Kj%4poo;u2
z15Hc)hB@jF^>R$5DE<)tg<3W@H_7|HupVd8wLS1gKg8+}8VUSn$`T(-EVdM02|YT8
z;S{47oDQ!a<tA!I33iT=U(NR3e6~|ab6~|HB^sy$J|o8Urc<N~3ANR`LtUrrXJKqX
zJYQkwcbUQj17xFQj3@O<^)zVsp6LVTf69XyLP;&e1;upFj-u7x59aC?29F-JAw2^O
zQblp*-bNKpVkg?BCWEB%8M}rbc{eqD<{W?F?BxeVvPj+qD&w${4Pl_glHFiu(`C-_
zsqMu97R^^khF$p&#cA&zWTta}iY^lf{;UvssifCWA0{OOi{zst@^xEJ_$gJpMNPm@
zcYQXIL-$mRJkk!qUQd1)8D#Kk3YuC_qGU(V#BVm~%ruQM{BW;?=WZtY>WXYKvkpTl
zWGA*~C@jTz-aeFkuu&5qZKc!AI$rD}6xyOF;glZ@>w^9SbZbj@S&3m-7#j(%<&#x+
zufi-|BtowYQ<=#azKjz)aVjD`teBOo2}RWIn)$5)0-E%a;53OOY-$pH={NIzFXm+P
zc!y*#EUg13d*|mj@*W#XmxnK*(h1Gl-G9&|hszip$_5^4Cmidf3Bln?%tk@+Cf(`A
z{yArYjwvsgT_LC8q0CEM&LQhV3EUsywg$0O@&oSm0>hN^FW~`plrH9cMwYAO*SIZS
z^;5}Sr%HSN7p#g8gz^0u#5L!Ay!Xjsa>6dX?zrW10Dk9UE!EvR0^n-{>>&K<JcZIA
zr(PkPsTFO1(4<*b`~7b+j^GekrAq<cZ%^BIf5GtaGc0=y9=zhgzv(UPX)gRX{PrI}
z_h<^vR0b~q5tJv2Qt{|**OIKb@0pbAx@3fWkOXXsmry+MtKbhBOH0TbOcKhu6Ge6s
zH(jHu{XtV4+>)iNGJ^kD3VNHayfAMn(kD?36-PFjQSN{m|0~Z$%w(49rvHeKgTqqi
z#M7?14k)1;UShq;g<Q;X0p;3u?G@fRg5##>=h*dvEZYxV6&4;U;F$cfaK{jAky|le
zoEhg#=140XnsO?2rbb{Oi;zu_A$DVO>x7@AMv%CEgQ4ltbQxIg>8)yT2C;dKxrJsH
z$UMgvtOg;UISyyCZfEI4$%J}*$d=G*h&X;rhiwfiA6@^KfL*P2ldvK*B{ol++USYi
zW*)F025Npz!jFY-oD*{#(+gIs3sWo$y_p$AFO^DsbxFyl_MRQAcqAhjpTZ>EQYX=~
z^eJJPYZOa5rL*YL1QDi8R4-{aURdCRR1=m)`unh(hLHQRy_XY4Or^OmJ+|^X{BDwS
z30Iq|<9(;a@6o&EcLl|!Q0ah%34wUO8>Na0Eo~z*BbY<43yJ2IbY!|u4n@xzT2;&6
zoP2(ItW7n%HHQ`0l^Hdy&})x(4I*z9RnzU<a#(O^|8>t(wy9{vcPGFD)QcU4WSxV&
zfpC=tq_JGzLN4KBoBaZZd&ie&OretP0Uw`Tr+FCFFW=-Gj6*h!9jgn@8n!SqEy<Ek
zcr;&h<Z3kFixrgG1-INuGi09Y`o7CHmP9%BG{MpL7T-s){45tu%Yq5YhxqSvLZQNu
zlhJI~hgl!qfAwYKuqP`>n=ogT6JO>~&JnC%AC0=BY&g*)WM*-84B=8?F8TJSg4#;r
zLJ9h%^ql7he2x3W0)Dy;8tduDqPdoKcJb)*Z_C4%#xs@HbD8K!u0@2}+&G+WWDPu{
zQ7}>_oOY41w#(3J%0)LTm(T?Xj!nw)$i+qtEbm@MYOO>?Nwe$Rx$DSjp*@O{mq?Tt
z9+K}Cm$fUng3S~7OmN)IHCavl93wOA@<f0id7glTEZ{z&fh(Oybh>3nx=z9TiMg2h
z3(u>LCQONcPKmlQ?yvG@UHwx^*|xcyRGZMMCPTxj#fg;*i~;ax^vdL!vlXI}&o9$v
z1fC_-#3$yBz9GG=EMaF?ZM!$GrGF?pph7UkwXdh)fk<y)q=nxUsqULLp~}Mtb)kMj
z^lrqg-SQ1Z+`v;0Ornm3V`bN{={%IdUl<|PGSuXaE&oV<zAP^Bcq6>n?du%hdYqh-
z5naIL2Ng(GI3h{2g2<PeKc8~G2<8}Lq1ePF)Lq#l{vNVQ@7NvngQoGwF=PRdgu+Xp
zQ^_&B-YbS!CG{M$h973;4(y_mgWW+K{C98D^J*PNo3;TJy0k97bUWvKhi#|G%x`Pv
zvn0=uvV8ODFSLy3%b##CETu;_w%B!y44z%8%bQy*>X%u{nx2sQZhb>X-_X3HMQIFD
z6^D@-tLlwh_jwCto&k-^MK~?v9IcE9-M!b!4M&;l*vcJ6Ph8t^;hhlua$AFb;paL~
zs@?d6RXXQ;ywoQ*+(N^aW#`V>C{q8Cp3mk^oObL=<xe!zZ5_tw*E`_1dkYrbne-du
zNo;|-VKbT)#y^J+GBnlqzj=K!Iy$kJp!wVlAV%jlfnw5#MTh5;0Z*g&9<SKp8$W52
zKP%EG&^$GFFFYmu@@`tykghytE5|FXr*WKkm$+P0$L_P>ABG)VE_>56BAvEBgR$s7
ziDrOJrTe<?d02uCv)&R{O^A~plkSUk+&=r=g7-NyW8||llmzES%uWb^{<h4Q|4pUB
zRiAoh9hth`>|0GRqUvgsz&5%P4(!!>jQR|IW2Nb7FM0cC_Fx7#b~)FrPwAO*04T32
z-&2r3+`23of+h(g9R8rucE@etRwe1c7WOygW|?xu`-JxeNwD!W%<4l>(E<go7Rc`T
zpt{%s0iacPaskQ*vq5noAgYyx#aP)KTYP!ao#>tslSffkYomFeHSmRbHEY>?{ey;M
zxj;h9pT(Yfp3H}#^Oby$2-fbGq@Ll-m6$GylGn+kjV*~5?9}7ZI;u`5$P(cvIf;tk
zg8`a}qdq>PAqmdZ=<|Ng1NW&spPVY3f@5*%6kqxfM5ke69XVY1gr*e~;fRdmnAR>7
z6k>!gh?oI<EN5D{Ukb0~PUEK0z;*<-`+Ahs%SQnslQyegMz!y;Jh#Av)iIGL>oFYe
zkPtM(Bf{dEgSSq8bS@d$)T`_4EIFvXn>ZmFVio<SU-04iuWB4qTr^+cLkD;2kqrJ4
z^C?P%jS@atvei@Gl=M%wo5!%Y@XqMc*RP}_XJfa%%6)!h>A&F>yrhgI7W%RfHmPiK
z$>W1ELn}OFzai?$ph~ThHuc4uh3zxa;tlA8o1w+W*r!g1!gZW~Qh~|~c*X_0I(Z^f
zE>)hKtmK&(l%D=&!7f9~DLOCWi74;#D=)p|?*>We>gI>OSICk(S16BuPz2Hu<tAig
z#*nSb>7EUrz_g@tKLf`MO%E$URkL@Z7a|WGY*vfD%lm~detaw{{xe0BjKRZs^3k7=
z4P0lixlT(~h?vQ1iEQo(o7GN;(Q}g?7CCRAA|1}u4yp=Kz8TPLIEeA#GCRt`cL&}(
zYKU^5a=-2<=;3gCC^^*iK<Hb7{KYS<YsjUpaWeI^?|t%pJc_~LGmc!{95mTrd8V*5
zR-Oqzu)3T-s*@pd^1+tpF=z>odxu|+jm>ei9c^BTHdD>fLUpcV=g&vbC_g)OV9lrj
z1+oInN@gF)1+9E%dnw`{d}Row%eBs4G6y~ToUIB|+)S)2rq37{zm{8V{&|TZyj(>E
zP0(vkmTcHIiYkS(QADc%f2cfh?aQ{sdRM_u+|Bt)7S5Zly3BPmAy)#YpKw3Z1!qZm
ze_CxkCNSkZK{%YRhdKc7+ZfDeYtTH8jo;M&$|9D1b<33o!Ot@u-SLDEO1`?II*Wm<
zLxc1X?Esz<RPv^*5YBD4<R&{P#&12uQRa5xV+YsK?{%)nm^~gnyWDk(a+@4Hfr`Wd
zAYxZAnh;?JB}C52nWm&?AyRM?i^HY8H_rqey`Oj{`@<3ObC2rp$$6^ijl;wFP7l;E
zQO*-lVMDiwaRn0vA=quVvh7av;W(UF^zbmQ+HTbNPBq_P;<<;C{KvJKbzGrbN{6z1
z?RWEQy!fK?$Trt22-wi{xuVvIq*{$mJ7$A8d}L+sig=2d)vxTrw-sxweKd5+$Ob={
zA$fx^(2ywcCLLkcNE&0L3Jf*$BK%D>qBBj6^G&sRa^$p<><k5+HNU_3L6?dMoY-u_
z@KGKRgP*IgnvzXh#_FR7^9%D4MWRU|zp5(3JGC7U?-A4~GtOjm0N#U`I3ufz>un+-
zi#?lAZ1TY}JJ67nLs-~ncV*smpYU@M*9ejcEKZRTF6x)7oX)E$T*A5bF2Lzk1juxg
z!VyaDggWFgd$|;;u>R6WNHfc@X|8-6(d(S+;UAikOdP7jqGPYWK8-aWnORJl(8Aat
zLsHn2;Jxm;Au5-#L741$7r{8J=o2Rp!i5fuJctruW#^C4xF!ANNK`5@c4D&;6NWyX
z>+AWEEKfZN=bn$TC7RkgyKQb8)CXvNSUdYFWSM>QX0D+E;g*KauCS#R5rdZ~B#O2`
z>D?`)ezh<n=@H59In&jiMlmNHWy~HhL@+MDxy6zBMTwMILpKTGb3m6EH6@?f5St-8
zf5aJ|jd$gJ_o{<;;j9zC7xdtrBRsEZue5V?E;T$80v4Ca4)s9jGdNI*J=63jijG{Z
zRm0`Jnil@C-!bG3-ecr*3u&q-^8Iq?)}=3Nx+nYx5=SF#h?-`_KXToZxwV0A+w+>!
z4lGL2^Hs%*QIj37izQzdx0jr{Il+0!>dQI?XN8xmBwBruW5jcJz-|(|f;W70d>S`#
zpTHho-ShCf-0B0);uCgO_YQpZo)p@RvD6*3X#gaYdjoOp&54JxiPM;nnzYNwPUon7
zn_{_Y?lB(m{I5K|`h?LKQzd<k$a=&>U$!-yePDF+m+2x(1wIVU*D-77c=WE|gyT3q
zX$K=^AD*lVzVqVbw>4vb{@qL^PjJ9upGVV=HEvsXwQ%QIDo`%p!MXS3uX|}%^b7Ib
zzxPV>RF>F-n+bYXUI}uHAz#2w`HO>W<Cm)PW<@mpoHt#dEOe<_N3<c<!n`akQ%sjf
z3OMRQUq^@X28g(5yZp>|y}SF`*J?roTPHZr_3I5?>zn?8J3nMO&L-+e-z*9AePZV&
zl8;@zvKkr^!R9qciV0UeNqM;Y`6B_{qGN_1O`3!Sr_fapc3<YTq~{>!rtiByz07`H
zJa01b%)$3YUR+mEvNZZ_1xn<|9aZ%mb^ne!4s1y8t+X_XVRwo06$`ehVPjn-dBW#J
z+?4~iAET}(yGY58Pelw=I>70vLI~Idshr4-?<Jf5T4+vme*;EYae}-0ts`~daYVHF
z&~P(L`S5*C!w{#V8Rrq4mI}McR6)er_{}Pn;z<jbF*&U6Y&p?7ck`ka-y26U`ExRc
zA~aHr?sswA@oi$7z`8rNMk~8L72$ce?|osCeo80SEM(D<_-k0|^nETH^4kcAD_n$f
zY&S>Ak7&8jaQRJM7pO^`GO^#(tItDdH?HY8=acXA@U%S<e!JDc#{KZ1$ToA*OFtDX
z7pZ*wkhP!B@xY1j4(Ya(wd-dQx%s1FGOsxAvfq6bQ{DGa<SXZyI=epDqpncQNhKY^
z>P!oP0xPc}#1+_NwNEA=Ge<hvYdfi(n=TRgO#k_CU`O<mTkPpHZfm}V<W{glfCD_{
zxDSNR4r_U$rWaq1t+~qj{e#-m`?QVMel(ts_v_JGTaGmFV9M7%Df|Ls2jM^zj|+$U
zUbPu-S1btSW5eeA%4g1AoIfafsy5)E*4eeGdgH?`VwA`s1m{=b5W)zi*%)aDDl`(C
z_~zD^&H|V7BbhqiOrTYcf?=>SWlgy>ZuEHk&BdY<Pd-YGFW2T!Tk6jA!E7va8)C>P
zt$yF9xrhqHm}%lRR=r2oq0yyS`D6D%b2HW`g`IaA8PEnA85M5-UfgbLtMf66;sC{f
zFw|<(#O4{xh@u%T>iF2@%%wd&+_cHUrgakiF!0e0p3JOupy$e^BBe7i;*58SAFqZV
zo%PQ9g|dQ%zU+%noWkbb(Gox;3zSEu8>)*53Ui+JHpqjF2RAwDPr@6KJmerCnuydU
z)R;zA&+>7uy%S0Kc`d5A=3>HeRc?ls%(TcU;+59%_!jhw_DSe*iUgLzLp?z}p5>E0
zI7|*)(i@wVvMBVrb1KwC;^f0eZpXT#-!YHa9?B!yph}@+W8i0>UpZcTw$Yd<)W15a
zmi5uIJW;UiW0zuHUXx>?nXkXu<tZs<g)W{AOtW(&=A_CQiXNfsc?rm<*p*qH#0q>4
zG$t}z+z^J?IKy7w3sP-vywTKNdFq_%$<B$(b}9!KJnHz#bWc*w#KME|>+zA<^WM|n
zi{SHTEu<XuL<Hq#q;7DQUuc*)JbtL#SBPwtjexYmSvT^oOmsTH*#NWIuvt8La5D6<
zW8SxK49~cIC7z@Q9i6#EbLg841~+L*)Lon*U|KF4g5)5gu{DKXm!aO#Pcey6v|jf6
zo$u4vJtO?0?qnW)Bo_YamV++7D@;Xr9$I=lV|;~|oTGzqlRQ4;U`X<3?QJfNdLJ(}
ztg7>^bM0n~29YUM2&Z~D>Z#>vpt7=DNWhTplP?nBX8kfEvK`tjYv0^OvmS9tQhdkk
zPqxa;-%sWULOH+guD<d7>Y{jc`R==4ef|s-zd(HRZhQbm4LMOOF=<deP;<^J(|imU
zekYu3taPK}dmf8d^8*!InxF$dz{cb59<1)a)3n0h;5SVmzMKNP#A7N>>6tjYZuc?w
z%S|x3Gu2(7xKl6Vs02$jwY$91433?4*!nUP+Xu>0k_q7#U<;hj#eAY_bFfOnJ4b<g
z^1JpYN=y!UmNt^E1v=t8mo}{5g!ya3ZFoq@=k+p<T6~P&8YVx;^Iprj^Fga@XSpkK
zSw$Hky%f-{+jn-L^vF(GqHJ#YEpQqN58N8i{AlKDM*Os=9dry4{>>>~MKDHOh=xiU
zymfus#bF($N~nv(YhSbjwxpMpGv|(=AHNdbU46#To6nFyI*UAbeM5v$$ABhs$9IlJ
z6XK$GgnfO9x3K*bT|DHFSvg^j)H82%j!N(=j9uSTrzPCl<yvzfcTE^Y_<|ZntFxm^
zqF>tfFOfcUTh4;F=(j9X6PEm|%O%Pv5Ud`@V|Gxi1m{IKLHA2;jNRDr5}(k%<+TlC
z-0mjQM>@WK+o;fP+woJh9d8~`i8d^Z5R@Z^bo+-Lp*8+cVdm`Gqo+3rh=|wU@9{o<
z`yJrCnq_$4b;Jp8t~EH9|91dg@IUSBh{SIOuY8JKG0zLT3Q&<{g)foVT8S_ghx;-q
zGv9EnAbW}Li9;S;7!@Z7Y+Nr%wwNmC`yNXt=;OrN1oG*HHZ79Jr)#8%u&cytS6`PA
z<zlcOAy>s{!~L$al6F}KbvZ-Q2s&w9dGYv*(56r5uax=z8VGT$KH>5}gKNH-q}dqb
ze6+dE+@S}$Jrz4$kvQStWy7K7%a1VQIc191z@g<)+2d^+J7}K&hdv{ZC-!VTW+Q66
z9W{Xt%~Et4d@uNzyi3-;-&GO4PFQ%-c_gDAu?<^;T`Ga`B0QUhkVWntu-lKQj6Sbz
zD0g{K-^xYjp<Mp0W3&|ChmLT3XVBT$IJ9oyqEz9A<N*!g;l$MuIF)_?pJa{+^!1yR
zJPIS2G2(q}>(EuvF36Rt4Ljm7DDm_rAFmTjA~w=&I}Mz;_sc4+IS;mu&yf!i92%(y
zFPD>MNLrQ6K+c1LlF@b*Yn&P$dLq<1sw{083uhPhl0$XBPYk;W^0zVHa2b?9K-FIe
z2bh7g6{u&h)$2;2HfjzI+}#Bhb)cxlKZMx^NwIDEe1UNYB?bbZqXV3<b%y~r<cKg;
zgnCC|nz{xhw6BHk>8L(Q28rg|bg^?7u$~6E=gIZz^**j~%(5GJD%4!M6#q<i?tMjQ
zpOh9)=tG0<*8!Ybhfk<An*g!G2q0E?z#oVeR_=)v@?(P+llN?H2+qLfh8x)2+%GW5
zy)@Xr_yI9Mk+A{Vkc@j&9i}cIf_MZ)w(SC9jO>8z!i_!wDsW6b;-_3clk}7vs9e%O
zS(0{-brUoLlqCZJN9(t%YvaN${qzw@0rfs6PM~yokfI6>NUlI(QY~%Y<(BR0Jd4V&
zb-DI886^FGzrMiwLzD;KH({U<p@A@?KEnWd%rF^X4qn;rzkZ;oTT>BY&C-Iwrzwd8
zzdbl$vN-JtHGY-ZcfG@)ZCY}05zvw+bf>TZ%}Pl?Ly}iH4T^|duzlCdE|%zZHV$4R
z7VH}-kn#TgdXlNg%>Dll5A=ZEI4qP4+LWWLF?r30xSQljVr=)**RJZ#cbr*Gdn-`y
z%Htwol_-2T|L7An+0p%wyv(-p2f{k)7JNzSrwl(GPzScCJTSYJAY?I;BuNp2GXo5W
z;%@QGZs6|0`Gw9^UyStbHQjQ)$4DQ}k(MV<g6mqd{!ZxYf;~cRpM!6y^?|$=7mDmv
zu(&+jJgGg{OSXhT<^|<@FWjkuid94ZK>+-wLN9+}j2ty_tGgMWGtoNtHVcNIgfg84
zCW<f!Vn?rQWJmI))2$WrhsV)9;qRyYP+7JGaTxsgW!6?oU>(HwAy!$XaeR<l<?RUn
z!mbFIM?Z-YsV&hva=w-cs1a&U_hcd8p!n%l=X7?pZ(Ugqz)Skv4CU=uQ^8IUa;~!{
z3CvTW9n}RMTVxUTL-xp7Yz^G%5&LlJ9o-*B=ZG*oeCgbd5&6u{mr?m50;Cas*uF>A
z9a1=KZf}}-JtNBc<SRpHE!T|T?U5fn#X4x)ZgtjcAQY(X{rH#&x~+V205odz#S+p-
zrALg;=n<&T?|&TbutC0;Vb-QCd8(*krf(=>LVq|@p+$AhBI92`V!g##(P!jHMM?{Q
z2_w92Dl(<E^w4YIU)&114{KUFTcY<h5cl;j(zUv6)1v80qi5jzZnO_9Jn*4g^9;m6
z0I1LXMh*lY{BIPw0`(73*Z8oSKz;7dh-;SN>O3LJxv0zgZjwDln_D_4TM(MUKcFOw
z#H&q6GHYg<*x^2siiSg2w0iRxFO?<QwmdA2r@5p+^GZ5weg0f#dX`k?RBpq5W@gUp
z`F;7|?7!uHb^gl=#Q?FI8nOo+o#6v3`|na-^2CgKz>s8?1<s&sWR^yibA6{@m^FLm
z*^)M6KHd*MY!uD$vi;&0g?5(FN&wdSU7$Mmoy4*;SEwjow>kpz7c{Bw)+UMK@>@&v
zEG2;;!$0d)@7J2<>j*BjK-qq%f&Uc^Yvm+}eKlXK4FDE3Y$L-9YK$E4wP%K!*=Gwa
z7M}dP7GHieJ3B`Z?^(f+O(Qri!kgp&pSkg0?V*nxU%!W90XO9|LAa?cZ<-h!01obz
zjsuS+^X+b1%3O_oDV285#oK#8{o^c&>qwGwhMkAuwiDW^`gCbY=z}n!^}Dxe_`VoL
zgE;t``1XF~Y|Y~POGZ{WkpfUu{!>YE)T74#*}?x$jgj()Fd}l_jM^`Go=JMFoW1Cy
zzE>UGryRatOkHpBr;owUjqpAAuVs4Hyqu|H@~70SDs80a;iEAlRs4}$Mu#K}Snsk4
zGrT`eYik5{U;p>V*?=e<B;(96yEb>38f%qr=Nh4Hm1QQ>#B8<mbv%or{s<wA4-WL_
zs#j7uL5K^+8{6==^M!tDd{lJu$LB~RWuH9??*BwLTG0C)4{Ius*3JQ$_3@yb^z6;^
zuL6y7CeSU0T8;dR)6(xM1HwddqyX#b*`NOkP7>vpQVGEgDk1|)%2|l;+g$m<@`D>9
zOG=7aBSIxfSMR5X4aTO@DHdM$aJCxC%!lK<nj!QQbHcDcKfxwM%vUq*YYVK@_#2OC
z1as8oG#WMB7d}lMtN`@~95PA1t|F{8p)f(X)H!lT+bi}d3x`~;x>7@n>Rl1@*oZ^X
zswOGH>R#ocKs*aWLzZu}<Qb5SG+gZ*Xu4(9lZPx^vD{vIHvFy3fga6&?3EP<?^DZI
zRnS4OZE>a25lMkW0ZsH>!V*9_M!sN==kkr&l2Q*HfAb;X!hFWG+x>U<Ci$Z~-2UNi
zV6GeFFGH4Wfu8i$8&?)Ek6LK1<1|Ja^n90UA&7->a!6;I9ATJBy;uF3hxxoNkM9G5
zXCv|?LSK!@SI(D|i^yJE<dRJY(&AgbUw&Th9PY?;yWA-!p0(@V$!|{S-BeR|O_3vq
z;#Y$8W{VKPx2at^HU?$gVvtzDI``>D#~{h*TO%DUzZ#;Zj5JkgSvwI66T-n%uDLSP
z;80n};)3?jj-M30Qi9N!t5Sf1{?S~GU;GaVimyb|_JK?o{^|5m(?0GT?xqaG>jvj7
zmnN&=CKTQ&>04+-VQO5z3ZJPp=zgf7ao|nRQRQYAI3d%Q6+oQJ-hk<oLuDP(NEf<s
zH+#%5C2khI_3n*bA1ONP7Vd26TKN_il8_8QGC@qDKpEh{KM;8fjmY&2c5pH`SoBy;
z(Eoh9Q^<x?g3&Gm&WL)jfUIA}Qd2ig1y|ZF0p+Pk82MHWIAUEr0-pW65J@!XK)@#B
zUn<@A6MA9PZk72l@<e&{Y2TjX(%d!<hkn|q^aJn+LvTHOT@*Z!0fRR$4+P3vD-2|(
zZS9?($TBCKMl>^?CV>Bzs_{0D+g5S!`@Hebe4W0_MstM`qw*<A*>Ve&g*(E?wp<E0
zN)<|L04rP$hkl1|n1>|s9dTbC!720++a|ypw7f8oJ<x@|#;R%Fk0($=Gl3b5z|ha3
zHrF#v2V64KL%<H=`EXdR`9M?G*4aJ9)XzF&)KG8i>$~$Bl8?HB&NwA=b}D`x8+Q=}
zZn!Y3&wbY26g9_GOOAf`%G@i1slGIv<ueDOdQ=4j0yJfdCn4cD=fE0Wy5NsJ8<9qa
z6(#JihUPpzJc{;w!|&U8*ikU{<)J!6!P!2FS}tN?-2V+oxgK$F$c7u6;210OkUy9&
zPNNmOI(y_Y?_G}AG0HHVEsw+So~~Y%Nbj8_d@kK_%lw|lU)70?<&e<qh^k!}&3W0_
z<Jc)zF86K-heax4%xry=k9kO+cx~4?x9T$-R~#=dRlFFpURN%*0A+u?KJvite3$A#
z>Cld`RpKhIMEAoT;kVV%f@1UXZ)WVLdJ9yf5l+4Jaw@_mPEL-J>1GZN+6YDq?I&7r
ze$8FN)q@Zo_c=>=4*6_vBbvF!%zZ+V36W&cTWS|WYS>Zr`EcrWWAhocGvfg?w86I~
zJ|GX<Z_5WNaZWHw9_@PMmh82mje_9X#XGv9h1+XFc@xsV7@mg^J${uKD{<fMUvPBF
zU17tG)g_6#U-XKz;xy>C>qt*}dP7T?kWLtdv-}#db8~Z{=u{9TTadxqkL9XXf}e-$
z`HV>FM+|MXPi2J;8iXkvD+r7YfZUj`QJ&O+T>VjFmuo)noo2WFsL(CmWxfY|lI-^`
zeqmP#7d&iPDU7W&anI9kg>(5H$X3HEyWZ@PA-kW7*|}3>du#Khs-b130*82Ha|qN8
z`zUXm$+Zc7%wK4(3@d+fo}Evm0#j0+d{f(#tDO;58~L^VIOK}zOqffW#*KHv%c2_F
zd6JjlaNX+6tcj-=+_QUjCu@e?>`Cd2CSG+zwdIO=?2{e|Sqqmflf6<Mwh#1kTSs>0
zJuJNXWXO}&@HkNNJ4pKg!96LL1*FVV1b*cE%I00QI$YL0<Xjx7^UBuqQl-lMFIL|s
zmH7J8t%z5fz|t?H%>=E-mBq8%gc?>2ET$ZN|4j0Gp&Y*MN)wHSCS72P^t!9hdtV3Q
zcX?Bd&yOiuBxvD$oMA4BxaBcw7XSJwCsVI%K-!Iqndv&`Tt*}ufdA|+Zd>B}eB9ZZ
zX^q>t!+z;WQ$=?5)hY{#-?mLk$~9iLwl1-8I7v=+m+mAV9MwNnwa%uNJK*4G)44l+
zfKKdGl&DSf7@JH1!#tNqGHVJD!@fghai@(7r%Abb6abhr#Ijj_%%t>>SHYeyQRK%%
zDVMT+VML9>iKM4OxsHym-ZDcS-|Km*ZU|h%X^cM|!&dL)C~M<UO@7>P=i~OK`LXgr
zsom4z>Y*!KEw=A27?{UhjBe2qz9cxHev^$^x>OMphD+~oFw5DMEb5y|P58x}(P5JA
zZ&vbgSYl-P-JM>?k#bLsf(973&Lk{!?BqM`%+~u(RrZY;^W4#S(Dv6i>P!@_n0U3C
zAL?T5C(ld-C5-dxYxb@@-rydpHP#^+m+U;1R#Yz*E$5%w3gqAzk5Og)<}2u^uTKW4
zPjk!<Stf=7Jq!88oXl%_>372ngq5WuKV}N%k3(e3CzhmEXC~<y7k)Yl$8}=I?wTw!
z%Ux`;RJcNS??QEmLYFz$WZtdt7MC`kZ2Cc!?BpfnU2g^Zije7HLVVr4tVc>uZa?6)
zRxMYbSqbz%Z|8rZ0xNFXb^C{N$r~+6v5m|)dV$E-MQx22hMswK@h+fIq>vw6qvBU|
zpiOL4zBn~5M1TWRnXq_?+|p1$>~Nm71wvkbK;^uzZitz#okI@UpPSoZ>w)#lkG2WN
zp&UY5Gn{8Cm$VHA+|dC9U#0Lb+f#gJ95~!E>>hvn*`jaHS=YTOhxa_@8cW6!8t$hH
zS-WjXu9&@%^`Po5&$}3Wzt}h4>+zHn=bN`y@vYm?%)(>r)AtRpQpR&C;wFQPOU7{I
zgEqIa!ta>Dt`iBjtA+I?jPI=vB{n$Y7_XRl^udZ`d-44hI~R#G#qdI#@2l-I{<}#P
zBgLE(@z)6>q*!A3HLGK5KL^Qbxu$@o<9sfi{RAbw%IUPWW9JhlFNyTjc@GbEGyz4n
zYEI!IMhtkxY8Stp`H>OZv9k!jcuKSrQNI)>a{tLjV#LV0g=4iR#$e%5yg~LcOGWnJ
zJ8xv2y-iKZ7SkS9sZ?3Wch7ct5Otdwp-c;fvt`?sITK9g^`tDZbp)GHkl47=%LD@#
z*XQrvxy_lT93y^0Pv)J~UZAZAZcAQr5_djxQP)`tns|x5_!#QD5sExiANq-6+*dkw
z0)aFeC~>rG8M)(pyv@xLJ8`-+I)<J#?82pEYBQfM<@b*J2goD~n^dD3V-b%=w|A$@
z%WG2XoIrfb5o<7ViF)F1PU0)=FoOgKvEE|;RWzV&v0Pr4lC0D@EFsOQ&wJj0`F72l
z(kg8{lBpDwiY;UT=pUR9pto7~L#=9~eix=PXmmXfio_*)>TW!H3DX1rC%=Co82&$F
zHS})>`9Jf?`XBjsqwmlnI3*5LL!GIu9LV)((YfudPh%u2ax5^vL!Y2%jU&UikwHsJ
z)>)CaCR$b+7=200XN<p?m@cV(;hes>bmyDJ4L(RQsvbko1LqtZQ`O*X-D$F);{=l>
z_tTK1J34Ahldm(!6|Yt+4;>zepgidJbteXsbjk7)?SIhlj+nqL5Opo1bNs!Jzja;e
zkGO5l)4W=8`T#?PA@xH42BseV%mYpDT%%TC&qYm})a53JT^}<dp3OaJonUoX^O)Xy
z#$zJ%4U#S{x*StUB;BqrC|X~z6?~CFRj>6PFsJrXJ3$%{b#^PCQXtx~C0+e!sg<J>
za>LNt!h97H2|eb|4~|$y{#_&}p<=^%Ni<m=uI*^F1JPL%=Mpt@zJ!GdTX7au2Oea1
zZYw`qJA?;hn-gR-DU7^^kJP=rb1N7hp2wzqcAza%<$Lj5n@poirp?G{@wK>Cwdc)l
zDKcajfFU7U&_Sax>cEaL<@9I!&p>P4F9|uMGIZ+V%=b~H+7v_o1*Z{;!b}tqwm3Ht
zLX||swptz_$4fpRYBnSlI$)V+KhKLk6{_*StN*2QVd|?u!)ks8%^QVNi=v;R7}dxf
zL@UePw*V^;+t^(mrsSMXF4pfA69{X#Boy|I_v*LZk*dQiT0Gp}vM#x#38Cn&lHAD6
z*$B);>@1i7top50x+L>hdoIn;^@rU`S3g;Ze!^<I1wf^a7~X42#noR}dS|4M#|#?O
zND^Ev2;!)yjU2_Cu@l4Op!O!Yd+J^eV)1a>Yp<Bw6*zwGJYzmMc77%jN>4dL<n}f0
zuHy%il5C@5SQ&AZnOi-nJtY^@)tHAlC<Cv(dS&vj-GhKt>ALy5o<2|06!WnX&#1@L
z!-FG3mZoEfu_pA%6{VADV{p@q=tmj)6+xE-qm-XF84~1<s69v%jB)MNnUD+x9++aj
ziaowYlg(rl9?lK#NK=ZhQZIcunuOr@O0|uw%C6Cwt}3dEGY_m-&uiI@UEk>n97s7@
zLuBn4g|L{D3vLeNietGW<u+EPYF<tnO5|GE48}GdBUDacpYN!UQFqCa_%05L7IB-5
zTWC~qolW!$im0+kuIXu=u5HQ`dfhJ3P=9OMVEFoNue94$3QJrI6PwK|p<^5}I>Hl=
z{pmpxE2wlf*KI3LM$xRjzN!Id#VIY(B_b0sc)qjw>+o|Lm6tn$lsniuY5y+&Q$**G
zW(;Y10Jl`Y*Dt3j7`N$a#(5AF8drop6rGyg;K%nYxf%{kbcJ}2k<b8yikHAz_1B<^
zw~!Bq>}WCZ)lbLJ0v?N!qb&+;IXg-eC`fiP5>T21O%jH*Lh{`#_N!hTvsqK=_LS{Z
zs$Ge!^BAjfEmyKRFt2Uj+L~Ap<PL{WyvZxX*anXnK;HRcDN>TS+%RmtVZo&%HMWF(
z`*j$9SBK9vIOLvnc6sFKN^<EWCJ@8BVQd`$Jweg!ktv`zHJ2DD{b3cf5qIY(t>ye}
zn|sI?SEQc8ANk6X)xbe)vlV~=DLT7(yPy<DpOWF6xf{PY3jayTu^k_*_F9vkUhZV)
zY=M}nNd2t`mF|RX=OyG~0r?um$Zo<9^JuNEe@b1-3+m_Lv{sDy)iPG)+E8fT`pzNc
zJUNgUJ~<PmL_5C1i4OK=4-i{l^UK7|d29~E55zw$`4&!k`S5xSf8`b<n{#c8D5`EZ
zOUZd08UR9&05+{>V=QEpcEFKx$<Bv|iT2aS%}AQ!${2p6hic`KYlf}Mokzd6=8${G
zRjP%Xmb5`nAWBYkM~TM^t&X5`uX=ry$guYR%)TmQTk|u)T1`OY9vS!KF4hsV7>eI1
z2vm_L$W`arn@$6bWQj3-EJf?=*Zk=^u}Gfo?8w59M3I`3Lkt4EUV?oRD309^HLOI%
z21$Yz#e~QQ<(nITvD7G%-ez>a{~UuwmIQnnd4K4lAHh(y<6AQ2glC<Q{|`Qqfx63x
zu?8-=+DQn5t^Y~3IE*a;`8=I?xO^a1RoGP~X!K#2jETVQXQGA2JP(A;yLSzmwNjO;
z2x6o1<ZijTl#@ih<O19fj_e^Sl22_vQ)%|Bn0fWHIsJj7)o-_MK8>nwM$%B!AkDlE
zt(fB!S7<;f>V%oukV9NKPRicHvxW9yfM3evTQS*kBlX{2Wi!0f%e3HQBmyP9aFsLu
z->LL9#C#Mo-2n?@af`h4t4}7Oq^$2j_#HK-iJ_Z^T7)9fSe3b-@Lo`vDmXX*v!=z?
z9YG(`nshWpo<u%wolu(?nvL_qlwhaK=_5Glr<b{}@jqjy;fWY1NB9riF8tv(<Y-D&
z);9>kDJ-v~mhW6P|9*-t`oZw1@uPOyVUHP3g$^0*@+-UlL6hP4c>Awpfe9!E%7Fpj
zlXyR9V<NPrfbXYmmHNBOev$7>emp!e)D)o5>pEwhWS<>TC^!k6bwK5pA*}_rZs7~q
z+Rym-0PEVobyd0mca3y8E#K}0Iq+JaOm$GS_)6YXS@Z{f1*1D31JpR{mzlhbDCbce
zpnmKQeEDb)K7Y`>0X)x4B5Ic%4n>aB0X0Me1CDRvvXxS=h1oeDcNqZ1Ghf&TyKRf=
z>m-sKaY&=*m3g+vZ7rH(;~obKjG}(OT66ycZb-#CzOX*3uAwOKbGCp7tQ>li&~1r2
z6@bWt0QJFa6xo39A|QS)053k8J+;R3*<PHMJ;H_EmH>*dud7f(wvrVfoDKl5yaPZ?
z&_M5N<6hTATVeRSh*oNN0k8->1cVN=)2OVP5y;sp^3ESLG)l@~&zyaU!{>TRt2$z4
ze>?YeMF3*Q@>g4&Idy;Oj1Mo^VeV_)zvu!_gz`sSs}D7<`y05g57EITp^I*RX{22)
zkfN2N!iZh_1~ha3s7Gn=fg#>sfBn4yd0j!EC^>+~k%LTpujU}q^an~$XT#^czm}SO
zd~G^}%Vcysd@(yD?*Rw>y8tc0{uXg5VOgPv9x{6B_O5%xf`6+&{+Isr1@G?gFsMz5
zlRbeT7j#!)#OQ~?+6slmm3p%rKepS1T%UBmfjqPU&${~uO-L<t*6-qeU(OBIeY}Xf
zcXlk%*)jzrKct2qCnz=to&G=B>i_@s+W+bir=bi$h~pbExI?6#-47VNGGUXnNz$bV
zpg}rTr=o1NW<+rlJ^xT=a5Z+th8xkM<JS+D+_RgF?b*#lE*k-_V7E5`u@Mh@fuE9T
zoAw*9proAvE~PGeTlIBa<Z3CH_N(ep!ZLgiBtloHhNV#I^@5MH`AA3oE%d0(B3Mb!
z@1T-k_+d;jW|w7*l4lR2L~BwM<4It$eb`$c^E3f*T$H6A%1u@DeN2c-pIiyv)_r!G
zkn0&Ax9uO8TRN4P4#ErZLX7QCO=p3(Z3Y)BIL+Vj3|LQ~hPWGG$9<&<qnoX_(MJ|i
zE>xD4B{goBPMxr@J7^H=*WB>+t9p$y%jtNIcHu(`5L`diW)t<ZaZ3-B_l{5R0+`J#
zsN76D=mG3V*l*3W?8v295F~?rr8CGVLt;j-8wO&3KF>PG(*E;RI~NY?n=ZvKo@ff#
zU1CiWj;fJ`yclYhLImH?J(jECflVZxekedk`+ca)v5T|woR2-`JC#7XO*Po$JNXx$
zJ!C~xtjA@9J1L3UOmrMyX(H`J;KaCIOZu=~UVb4*jM_SdW})2o1Upqc5WUvC|9+lN
z1k?L?Dg}fPyqA-+^`WUyR)eHpj}NLUa4IOI4H<y02Nwme7E};nWKYS(6`+v|Vu7eB
z3fK;U{_h)(msBa=M`VXdNbp79>pi}lt{)IXMyPArFedJFOi#XGkk;+$8{gxP)|nVt
z8<RF>6bd{GLG>e9yOk?GFwedE5=H;hkfxXC$B*y34ogNxwg2Qn?aX?I`C|*#U^U7S
zus43K!N|r+qz!Ug7_(*Chuk`Y+Sf88-5ni7@dTz3bA%k&qH?&JU_YC}_alG9YzTJJ
zZordWy5?ltIpb08kbbqhG){Ey-c4lsf&&>5927j?Q`E>=`D8k6!l&9aV|%vn^dWQO
zHb>c?A<aBR!{(QH60Y5&VN?|1eWL4N1ZFH>3mS|M*c<-Y9Jg~0qaM|P+{GrfTs}C9
zGiW!l=L+rT-%)-yf57%rtpCs9BX<l!1fEPL#;cK^<G~3U;k05z2o9Bjoxo}2XZhXt
z<L$`)%C6vb<^c^4OXpMV|M-5O`$pkky`K%tVe+;d*=u}Pzj;)Ekk#*=c|X!wMjOAI
zfOP9FNqU^3F>?NmrZuOSf!3mfZUdA>2}(R-Hd`Y;Q447*GRDMvB&wt_4VF74Jv>>6
zLn!gM{nIS@Ypqy^;UJSPl#_&%WLKNC_6hhYhpN|?B?40$`6COQtvmZ|__v;B8rryA
zNcC&>Di5y{@b5;_`^pPP`0C)h4w5md#f)+psbt-y?s*SF%pn=(8zvH*2L$uKwlO@#
zqDylqT`LeP$n>?be_m$NsJ!hvieRg?x9z~D0TLbSvUf%k<Ty}?eT9fO#V*S5G8m`5
zh0kx&+cD=nGjhGIdPi;69(f!A+%Ycxh&THoJ0fr1{q~}TrcO=N#PHJ2&uDr}#kIe`
ziSR-Fy-$_#Uwjw8FB2GOOIL8Ly{XE(6lFr=JlA(El?2<W=E}Hjc}~se^1syY|H$Mg
z<;%CCOUoN|gT~lPFu`@Fi5jT%c_YJE3kwTN`?4QbGfo)~3RI5qU#!VFUvyCA9zBmf
zsT7&8JNdUYgC~C9nklk>mil;<f>ndcU0YGyUt^Lw=jqYq%tP9&in!3m(rIj#BYNTd
zLlOLeogp*M`b;G)qVtzgrU7vo=M!K^aBPL1sN%WpxE$Krra=X60ncN@+YFYtzkRE(
z_rBGLKmS%;iJKFnd4#eB{6lZ$+Q=S6PD_~zPmM#9=b)pK{P8Zf0rTs}L<2mJv8<+C
z`~_pdJXY4lOD<Vkk1Km#F9Arfp4^nRri*Ez>qG5X>0xxq;Y)0b-H(Ou;W|H-tHS@+
zWPm@-`STZJ0qS(&rkFsLOB61G^_$YyRq%*qcUyP*{FC2_jAm>VsJ7D$SrYH%rXNSW
zRT{9E=nA)_Be130WgwDT{6SPp^^kV9%dvlD+J2!^DpVj~nE2(NJ`m(z`b?KqWc~a2
z4w5062#xlps*wnG6#+o`h?2vUhvvPHeh71XeSKcN=q;%vke^L&wDthS6<<LS00(S+
za-cSxard1i(y?^$LHk#MGjNzk#|xh2PUTrrKbRZ%#Z*epvJ!OPrvMc4|7lsRT`(pV
z1lK6<#<*Mf-0X#wE?X7%`VV^Xz#UU`H=bUt41T*bXe3BtMNc|&;{G4@{yV6ty=@=G
zQBb5e0jU8&L5c`cL`p!UiHNB55~V`~qz4H^dItdk1qBt5PDGmY-b?5ul#l?@Awj?h
zA^R-%`~GI0XZy@=?=$nBGjqQ4{^8)PnU%HHr`-4TxyyB}RdZKo+n23z@QdqTTz&AR
zYI2v#qe$!P@$J5I9&QQul?oR5YdfV}|FoEWpuVQ1PQ>4O`1geD?I(+vuxljwF}N2F
z!dTXfd%YdrQRkLVYc^Z*vYuzy{^hj_$LNVmioIgj-%e~I(M$&AL==C#SBVn`nIffC
z7x#)lUvsCtJfNKQ+E*EWEe<VuMnz%f=^OdmE`S8maXaAIP3Anoe`_grn>Yx*;Y;z{
ze316%lkad0mu+tQ9a)Y0T~nl{l8CzV7NlW)im-WD64{c?)d0eewJ%8EFGm+)N;|u1
z3SeLMWuHHFN|L>H0l*~sH(s4z=#!6oF_ZHry!bC4>-UfLt9GoaBn;GIp8Dw>&Yg>h
zXs-*O9&G$nTBHt@Xur`XoIvN*eY*M}k;U1Xf8^W5fz>_K-O&#QG8UWtVKGz&yVuKU
zscU5$t03PHumDj6fVHy!Vx9g2)|%c&#${1VHC!)%CF&DU5eFb9=XQm<sxqXzWZvh4
zWFt?Y9^>@IugFjYV6Y%@7$!%8Yhv{NyNe5h4(Wzp#;f$~v+u<yF0e?pu(GhKwrmWj
ziERXo{b{cLYe_2pGvD#!#G=#uYUtT))%+4u=iv<nX)`ljKhGq|!H#?uO)eXS`Yd*n
zOZAjy3j)MUpxT9I&T@7$OJaE6dE1yXOH#uy;%u~ogQ*_%<3~zD;iv_%7RR&c)xSIj
z=a>9nJ45^LcSa5%iqpWfD0ye>9zh|F1jf$a+HzPJGX`k#(w@AvXVpGayA*7`AV6@4
z#7Z(+Ia*G7pYUi$YxEYla2mF{#XM&ddig#`_D_6AIR0}EPW^6~|B8{H<Fn)8DogX1
zd<^I}(T-7-T{a;<bQ=;jSlbIQJS^E=#jr7)LKvB6M#CS6Iq$T<QSpZKwM?$`7SD}k
zw|b*od#Ynvm4BIc#=n22CEmR&o0>@Gd%meL5k8svQMJF+3Y#o`wrhKky}cm8&MJ;;
z$-yw8Drd1dhM$4>vPKds#j$2BYGFgQ{c+>|!nx=*o_>c};>dQ<pSx{F1K4e_f8?2-
zlgkMeSWKi5j}c)vLbfxY*ZV*rqq9N1?4xUPi^7c)+%>V@AHf#{xYw#sY))jZk;{`l
z@#fvSqk`(fb>Rnn8t&X#aih|-)I!Mt|Mq<N$=`bd>uq51@WY@uj@LdQI#{eO;c6E0
z^ZeS6a;pxW<N76=>)RQHA4m`JkZ9z^su2?8{l?UTX@raRnS^|*p~HZXlU%X#%aR>T
z*LRkKGyZkh&;R|fJK(cBcTdB9JgOUmUta>$rxFdTs8Ei%8=tN8&rwr9*IF)P;*G&J
zagZl5460Y*Y6Paxeb9NQAKYbce5<Qt`HNkp4!`bE?Ow{r4bQu6)f&Zn-u=#hgXljU
z^QC_<0ze-MX9Eo=4vb|(a1(vt55I~W^0<7a+wOk9r?A`Gv&8-ZXpSX1Vi(ft(qkGl
zbBT`k99d!qZH~ls^!Q_&v?x_)^8Rh!4gQIR^99qX%3H&46W0kQZHCq9Uf5>(nKeIi
ztCL{IfTRGbJsoTP6+ZQ7-&DF*#rfH|lU#s=ZaRpQq>G2NZ&P<>of-BSv@iQKAqT!-
zP3fAWPOp2>e?*vY&A*{VXl{cCYWwfDU+M2Re7zB1OHSw>q!^6<!bm>|bLviqHRf^B
zcPfUiQ)!w7GYK5W#??F>=NBf{EV`fFOv_M+Qg*~{)0x(H&+MP*8HFt1p1x&Bs*i9B
zJoB$!u<P$_?;0PZqfR@rGkQYKLQ)noB<ZR;^DHUwqfH#{#t)XpzMSfjlrY<2?$v!%
z91x+dD!pzzUb)NnizxTrD34o?9~2-IWl!<y7%IE%82Te>UHeaS$7Ij60`XY-dx*T7
z#ZHI_T`J|51RUQ2;P}KI7%Ody?I^wGO*qFYD5e95i3}2;CW(h-l|&oKlIG72_(u!E
z(D=kdWM$sIF_zMOc4h13Th}4ApXHt4|Aqj+4o%p<hXB(Dn7E)*_~xh&cyh&3BEvfO
z{_JgYtHV-vhOyO8({I{&4_nFta3-C_&8OiOL@2iFCY+lyD#bYEK|^C-Rh{{)L{HCg
zTS2c<EM?2~Wa7U75Y*p4eO{T5Kn=+KNZ2>w^P{~HY%>}nsh5Tl5$tnLe^AUz@-Q=%
z7fBHUT9SzEWEKSgFPU^Ig&>|(7gIFNJ!cXMBE3zJ{<mXV_CLdapNLh(-Nz>8aA}T7
z$n<rXm^-Un7jOan;7I&gaLxAKq0T+|RRQ=ku5J*D`w^@xhjUU^EGAC1*O^4Q$^X2m
zzf(-5;OTZ;Z{x`5HA|WCuX`U@CI1ff{=*5niI~$_2MUq(uR^duZw3YD-BL?FepHu$
zrb6kDD$-uxf%Yq3ClDH*9P2d)@9Q;kn&NGV7eQ<m8)$e|K)({E|EZC^F1;)NboVK*
zda(Sz1rt{Oe&nY~RyYQT9w9%(`vm*|Kc5$o@8iLKP=-{>J3(Kn?<s!#0=xX6{_V9J
z@ur9Qu+Syy7Y!nOOPg8p_-}D(ez&?-su;`{PMvAh^q@=ksJ@j@TKPSK`ohKka`pec
z_91@@`nI|AJCT%Y8J|)<ZK%rNPa<-P#Z=!p`xL43P5}<>SzFv<504O!`Vds3b3qZ0
zX+M!VHt6IR7$qru+Mzidv_4ca_2UnP$DIwV_}6U*_?EwY_AWq6=McRl$-!r<1^@^z
zo!4ya<V^2$<{NyzJz<3;hG*v4+v@o0t|M_e1Mc-8!HK*&3&A|1=aV99iLkf35wihZ
z#UW;MXDmH6`Tylimi-r?{~u<*0v7vgQ3)bv*UW#!!<9Zl{RA$^&T9i^?A!h4?D2Eg
z9jA@+V@S?Gl;YI7WYMqcJh;*)@iWkQ^F1X~b(K^87p@|>|JCcA`FpR2z@uU+A>2-Q
z7c&2(gxtIs{HF=pzRyzq^384UO9qS32miHwrR)s=5CNb0dlv@L9nr2X{1pxO-&_87
z#Qt}|{1&hM3;Ouq{qldTxv0QFrWFn$c!x-z5@KkZ?@f+Gm-BNj-04xfrQ6suAt8Gz
zJx2Z4C`AfSe`O**0Dp#@P?IPDqYW7DpYlzqCpm~^^%2CD7e1SudowVA&zU<_xJl$k
zI=3ul2I7GS18nuvQshuFaFT5Y>MKZ@+P__azzvNCAO6HFm`tqyndJ!vG7&oDR>VFC
z&qU_s1QIG}@Lw++<6;zx5K71G8~Z@&C$PH(aGkk`$IVAfL~6hT0MXVxS3vIIFINHy
zn!FR>cPH3`=LjI!w21ialM;Ws(nS22;F3K=#ve6;0ciYBKS^88U=;xiwEl%?i6B6^
zVGH;Wbs)P9d|WGnte^AyC1jImPDcLs<_1;dpRNq(^P|;11U}_GP&|OZFE#_M+#~<t
z+7#xz;hzHtGWiv&Pz44cT<92>p<zS|nD`6=<lz8Z`qxX)-5Aktg{0kGrs6RmKn((B
zwJqrcn4>lDB8YSY=<DmPEGIFXgy#C(o`O^Dbe}T8(TAn1fybb`&-ak%Za^P-Nd(YG
zMv-3*{_EH7?fxS)dvq_Rd-f5KN_rC@_G>b}p?<r~X!kD!dQIWTtmeOJV=i2OwfOVh
z0i*zXq3h#{@d!C`jD7D%EBL316grZ;f;dsoh1{R&$4~(aX}{S8bPTkN?9X2U+5E~W
z7dQOMDf8WZlX1f;iN35zM@>y4P{w)Rgc(G?RtIW9#ANVS5U#h86@b=cN6ZDe!H$e&
z`I|@1H5`6PJsW>exR?7~OTLpPP_V#L259kgWQjKs^N*FteOb!HL}%N}<z56j1AV<m
z4dr^?83Q~-T!z{rRe~LJZ)n(M!xu7V0zcrGLJ*83Gh|t9DUq(7d#E~>eY}c+dm-SZ
z_AOy=$?nM-k|tJ#t@DT%E#C;V5-cQMqq~}2fX?fz4<43vqkXB;Y1LOtKy@&v?i!v{
zN(z({Ou_(N3BAQ)5A2nyZ5kh?E5nw$hn1D5H>h-L!>Z>8Mz+>+vcQm0AOLl!cvAR&
z6#m2>w+X7A<Fe2k29mw9#d^(ROqJSBP4L!N)swCz6uxdU`hl&*BoRC~N?FPl`|e?D
z6KK3a(WWRZL*-jKRkiO%4y{o5%a1Q8x+s8>D&lv%0htXo)7^<-XyAh8!U7Le<hDu#
z_sE0xr-b>9g_TS~e>^%7{c8CWXt@f>#40b)5!IUM^6;dlYbl*avRT}&=RY(&6M6V-
z^wr63`tlq2{0Ry*7-&I&;;bTDF_Gq&xC4FggYo@A{!3r$YDS-!xlvh6SgV!v(S9uV
zT~#uSqFUtIZw@Cj43iA~aIOk%H#xF#N)0DDrd4B4V)mXMU}Oq>QMF}nM(m7@@AtKz
zF|}h^zk^e~L>%$J|Ii>-0i?JFZ5UQ_Xf|_$9xhV}^clU;j81al_Cawb@?PDMOtrcH
z&Ggn8w1T5gpkD4KglUynQMO*^oxw$nX9$xjl3>x21ATG)VGRv#crS5{re*G?olznk
zAg%)nK8Ss1(Vj(v-;F%5D2d*!Y`je8ywcmduoj$|{5%6)gD;D9#)&ruF<flp^W8gN
zT{|IEtyz~LES&2u%Jb$?O1KE6vf(oz5nVyn_gi8*iLUD=C+S1LB+c?jpt>UDF<hnM
zt1&bno~`v-PtjIbv9<Gy7k%o~)CB@D$0Ej~Pph`Ce!n}uwSB-Rf5=%~S5Y5r9|3!J
zr@<}moV~p;yATyC!~M)wYyLX?U`^!ZVeekJipxvwv&n}ZRk?bU;%bE~Rxv_O!|y3q
zM00IVhqe67ISd|Y;9JD-!&&iX!xxdX({ly^h=R54lJbfX3E`TW+m29J3+MIJSLcQo
zL+IQtoUO&okPmf%mc8u6K6G=0k3XpWxviBYOtsBIiTa)SPe!9}uXM#;u)MS2;(5kp
zYBJe$4G)S45?r^vOE8J_&SrD|(D4yvS{}2*yp;1aNvG$alN<bkbRGLsI8eO3?q0^8
zx%T62N89t7r8yF|VX20%$Lp~-8McFaKOUVy#pK|J(w9IpCWq1^RmGwQd}7}_xF!7L
zQ|!8B5Ox^W1aS1dYme|+>Qt=a<E!|gMBsQfyV=_N!P(ec!Y`ZYK)5#1phkzoUT@YW
zID~?6^xpL`xqV*iGx#}fx9I$C?6#N^1LN(?J5?5<Vdn6txG4PPx!ZjS8?+Q^Hq=xT
zg>Z#j0~}qTufABHzURXHw%~ZiJj1}V`m?Y!OSf}(3dI;{C=Nejy;c#k;6Z#)72XdQ
z*jYIh#Dh18CY^Zo#`U;L><GZPw_kZ0U$z&H1JrO9r)&v&ht))X{4RDDmU)lBn1Va*
z&ovsUwBj_32`IJp6SR#ub8Da@jW_%TZ^og&jqVa`Su<JTjM)XdYbxq|ogNDyqB#D0
zoB8+63=lAuIZ%USbXs=KY^&$Y<{K7D-V-OvR#gez{0&G|ers>grJ0!-qPaW$22;M-
zMp^QVCX(|-<<lW5SAiEj%r9vKtG^qmf-XlgZJQC^sq*J#z$9R*sN-kOf|@U<6i}5{
zQ7Hq*@`K0zLkeCHLt;+bP0j>afA4ITwK_zUY@v>f=|)=lT+6ceC0=%L>h<<4dZ11r
zwDQu3ECZeX@V-4g=e@C~OEon-O@~5ioLMBpXSIMq!R<ygoyY$mvARuJk2R>|n{Rfv
zKesMEDXHFYBab`OsQSIt!neOr#HtyFQ<zkf+(t>_w#QH5{7nS*Pl@iff$yP(l0${O
zdVC!{lD^{C<CKc_sfC$bj^tN?vu*gUGQaXA51=U>L0fX=sjQ-e4U~oTr7>8bZ18<>
zc-Pr83SHTZq$>m=gP3lhF?57vXiJ~<)E7MwW|EY%#1qm}a;n?o-cl`Al@ZSwscdxY
z3V%G|^NzLKxOvnY%n(b1d}0<fV8=N-Lun@tlGIw1IjbhXmc_Qn*W9+QEJ-LOOKM}M
z2<F6MQ&ll{mJ%(W>hC;TL>V~XqK*~PSGIpXyN5nzoN7O^N+?GZTR2yz9hDvd{2crD
z#ZaC}ZhL9klwM&A$Lb<(hg~vIx{hobIKz^p=a?CRI#gMuY?I^lvFe9H=lhnH8lg#F
zv~TsJKIC~=5)t|h69BGLI8<vU+)Y83=66`hXm}OMk1<5}oT>0P#pm=uJyN;{#HV&i
zobO4GPRfJCV2;*RiI0ZmkP$~#RUL{w$qMy#?c$7K&-&V5stF}Q?h;el>O|ta&GrKk
z+MlOnqgz<lk5Y>)EqbWcM!t$6SoA;R6T<Uc28(8#>U^?3O_oo~Z8zpROIOFG@Lp<}
zUnvw08#;UM3@2SSCDEiKO9YSE4#!h3?#817w*6WaGO%MlGpB@1M^;81I8rJQDEqyI
zE7U)7P>PqUl!ZRHWOR2Lpt<=1x>Zl%yFMxFJ)1gIy^jClqQJWL?ZV*OEGR~U9KXDH
zX68kkMs`Ni(zr6{{l>z@(W>KlwtEdz-bokNkMcTm-F+N{t-hSI7xI2G6@k{$KKliz
z_{j!PDYLt)l=8zx3==VJXN;_}BNup;u266lETj{LRDJa`(hJ96+fL>yKgk5!ZT80-
zDnI;)hO3Re=f5~vne-T<f0=X($(tycK=Db~YDj}3o}^|nXDx`22qsWt%L=hR7LNFZ
zs!#g)EcP{h$I9ieRIM@T7u=-Htp$K#!wsIbVzV6aK5^7cy``J)!cmTL+qg-|6Ov2M
zIN_){;=RkTE}vnX@&uZ<p5`IJF=>3i<XS6toY+NC+9GO-Q^C6BmL&ApK887HY~i#4
z-SxW*lEh0`o+!TZ2!4?ee5Z?FHg0eK3r4T$d6ISUB9e19Xn28%ctWUKV6C&73I$`i
zM(DHs1!D~Yip$3m_cL?2e$*Ty%lF;Stp)KmO-Oma<1`rEl`|uc6<C^~yXn)Dm-y`l
z-gEKrx;@PZ)dtLc0~i~vtdK|a(*=S_7yA-(&89oqb0osvJZdXqj+B^<P@MCc%)Yh*
z4<u;l4d>v&TVBOb-we;0vcg>Rd58;3mk>*Js<h6s-vu<}V@gQi=CF?eB}}giQ#};n
zUKx1Zd`uQN(1?3|(iGN}EO9(mE7Ilmj6BzIrJDli%xc~Q7Djgz>1r1DM1x}PMi#Hm
zQnYMr|AkMLkJL_!hQE4GOVQGO;X<Kz_8DS;dTqtmF<HLr_r8AltS>fcp?3oZlNb(3
z>oHO{ef8qD?aGA<=gY*isE@_g)3QLs!DAigoAxC)gVL#EwlBvFO-l1hH)MD-2#GPC
za#&s}EXrnAmdbGf)tIz#)Qw8GnXk5IY^A&@O&TbP7#5P(9(eSOW*yy>Fb231r@DxI
zME<(5u>v}`Smk7}s!m&SF6IOKiakaBnN#XdUsCXi*5VhE{L5ybe%wZ{jt1i%s6z{S
z-Qqo{JZ%SbmW4cvT`gazDQP_ij3~Z2%v<uIDY5WaLjwt0ZBe~2-W~16P44ar;?UQ^
zS@E`WRiq3CtRY4k-cr<GyP)Is-H`7v*!Ea~JQ&34SVauxFDz3WmTTtHmQT#<65y1v
z==E@K(zqBE(@ih2I?yJs-WiyU+YKXA6GH=>{c+i60m|kY-)zFJNAdbID_thVpWL@E
z`R|v4ScxL-O|ruY(KrU`yHnn_H8*|p9QitHXMZ4kH%t=dJPcm@sp=CIxB&n=6;~ps
zuie@t_#Mu~YeFCz+)YOJe1WX6Vy&H%%_i&=AFI>$o#QomO%&H30Ww4(=ES3R{Te{H
z1}j=UcCa#DTAC~!!1}4!yX9`q>Z$vuDZMF7FhL+<0M?ym&{aMXTME)jLyv7S>*h~u
zQfj(3Ba-vZyCxzAKYL_(WH=<@0e)oAl$<M658o8~X_NbNS5AT1I=<86xyt3FN3z-Y
zi%tOKwKz;57B22`!XSq}pNAU^@}LS+U}j~hK)$AI@08v{pLa%iV!5gJhho!O`cIAH
z>XkqgM6WJj9VAE{%(M2J){c@uAFEOzw=P@w24v!gie~N;?^D_ea8SJHR_*>sVj=oq
zzs|^JQxG9~TJ^E_#qDz}ZZQnp<-dY20K?)a#Sb3HoGZ%!xHWu_<|ivSGMzb)KJ>m*
z+#zZqI41xs#V=EOiKTm<e|vf`;Zy1{?P^;V%W##xEiQkXfw6H3_+i_J0Pcn!gSd!h
zay-%AUi1z_26DR}{-Ds>dv<!O<%u9_@Bx}=(tfjMYwB2WJrd_!GwxdDT&669V`78(
zr0X}qCRckyzD??y8LAHX-Wt5RRBpy-!G35G+r&l$;TX2_R%75ifWKv(M<co$yO4RU
zKEleY%g)THtD{wLbu1rUX9jP^)tZm%M?>)!!vdA=S0x1a8V+BWV;2a#@||&Y2=Qg7
zv}P&0GoHt2xIw{rNwn0vzSP1DZ?ufegA_~jGj93;tzrS;w6nnu5_v&3=ebwFPKhx$
z6<RT~Y&ug_jSBh%dQSc`@=0bJ8C4PKha$%THJMcD-!<vuMSofu))WNOT(Cz?@-qR3
zs)>I+@*chhz@-e%TgQW+5^dRYHJcm1$Wf{ikvzmCJf!O(+_9Ok8tMH91vPPW_Wr1A
zW!Ff9tMG9AjdtM#>q)zaXT?&l74p+$Y#-Lgoj!_u_zkG{BYOVOIEjOUFEt^tU*T5@
zNjiz9lV{(ELAm7<bLTc#?^N%u^hC?D=o~KC6pJN&^<F-Lgp*`55ZhMxzN1|nBHRbX
zteReU>Jw4C?KEmT!#^0Yw1<^)prP$uuC#v@vRHJ2(=H7{0a1ImCLSORfG8j)s4#6e
z$&ROycW(_#zLCw8Gi7Drby{?Hq&3&2Ejx6tleUgsM1?oe5uZ)DV!`LuR5_s)b&G3t
zcJdK6H#zqv{DNdHuL70au~?JPDr7jKQmecW=BPJbWdSVvCazs6*EHw5&OXmrC~sVP
zWcRW-I8CJRas$dN(BwW*{Fp;Gpv>7^DuR20b=Qb+HTxtdUjuRSHpk#Q?)q&e-3b~*
z>LD+$$r?jqxVP2SE^;4A=W}Jn_~ndmdkb`SIy?5S(=v3=(<><oSMa6@ugX!vK<)0>
z+xU7ILeHH+`}FMPxTiVC+-+y6Z20O=zqny(6B~}oAX5>2B{>f(2#=5iY7K&Xy+XJk
zd1_2q1dG_3We%Dk$Eysq?bs^_!dK3z+44BvFw7RmGsZ81!-442U2!Q^YnVL7(A7($
zx_TmGVR6O)?&fA2d2(_8^!ho8%!FqJnQASko0jqJ?SNt$Nofpn51${gpGfb#L$)<v
zo0K-cJ$QmkDxf;Xm6#I2>SJ)3ES`!H*;xa|+_N*XQIbM>no$RFx6ZaSWT#*FmfE)3
zY2H&i3U+3iL3g<tVEa(+g_hdxpI|)qGOlcGL-cETtQQCRhUHt>?>bq25DI@{mE96A
zr2(Aqoxo^~dt<J$R^ZMgmqX-7bT6;;z2M5%eU&<DkHdC^3tD*^H`kC%a6UX|m`z3O
zExdN5tm#&n{&tb`k$%Hzx{!FTH>U;AD}3qCR_V2#>08$VZ2fwe3+AxurS_Z3WU-34
zPS-NsCu|V4aI-tPD}DuHTVJVjSms6mfL@OUGoUxJ%WtQ^5^?z<4Jx*@I(35$O>_$U
z7r#t&bPo(Pqv?bee$Ewd(cc?A&nUiT<^M}jkf9W7MEoEZKlqa9-HcD@$B8D3`!yXK
z9tTjpOtnY>CF@YVeCsQ>A3xOwI&Dp?RW|ZiL^7gqVS?!v>34#Wj}pS}K7@5omJ$20
zyWv4X&V*@LMf*)~jD!tBM#;*dP5f)p9~2E7nva<pmBLZ45q1Vd4KTJC?3_(h%V;46
zkgsg#-07a7A{*^yUF_~pR0#5g*T?z|y1{(`BGRHDG^C<tRyLNWx!RPNy{#Jf1^n*I
z<d_2~G1Cn=94ecPa5II#m9WtM`bJK*(f!!sukS;?1<>7C^v?;u)0WWp{!3R$t2$O)
zz*f%YN4vRJUa50k=R&HzPw{%J*z^XQMb4Z#jT&oIu8=1+?T6~5ucAX9caCCKQ4mJ>
zDQq)K&@_hUVBX?(`S^jWwCsf~H>9AM`M65bRh#*jad)noJ-au2J=jvDMF|l;3bR<U
zAekq&BXLy(<InI%3qq!Go37s!X7F}D$*4njss7o$#dlSQd|APtoR0wJCG>iA8agd!
z<)nJ=0{i9Ky{V@`w!!J>+~BWoDT7U|g^7BL?p)Bc3{;dd7k1mvw6c5VfI;t02oA@~
z8=riW!#a=w0O@Q%ZJHOc|M1~S(!M*d>0p6{HSF6tH0oimbBXF;QKsfA$8Trn)Kit2
zbp%2*pF?7rSYe`Mj4E|3nZ-)LLVq#yxKPq~kcvrG!BFBs<zrVz%?z~*AzC{(=>s1Y
z5(Dl`X5S0czFW~8cyq;FB4cA-SiYj@i$u8_wQacH{r=F1Ct3RP5QcR&VhF}6<<6HB
z%bJ=19iPiJ8eY6tTeB{u^MtS#-3bj*Qz`hm12fo~eZNjVPh3CNTIW`Id($ly6jj;(
zdaLxlHm$CF3RJkH;KrGE(KmP}_@MWzM!}08vp@lxQ6;0t*wl%4IhHrnm4ROC>g^Rl
zu>WD)HZQIzvK&uN&)YCh`{4KqZ4TLROU78KpU$muixSdGZRt<dG&&RARzMkS<byg3
zLv)}cQV}Ii=5D2dxv0Z;s^5-=6!R+5$<J>M(8M>RExE4>ec!a6l0~rxp%C0~&D>d%
z5<VdqXfH%tn-P=6<1HY~wG6WQ4bdN={xhlOqAe-?^ttpljnv(KS!`c{n5Jz&Vw4F-
z;>W?fIoVMtQ>On_9(|yr1BbaXkQW4~6j!glHJKO#>;OY|lvLF^oC#a&(@$LBY1H4(
z^z!oHDKyc&G*dhA>ZQ6c^E2qJ<6r0FDkWF>h41a=n)6MWl<4VBm9TqhpX<&o+z7V3
zC_|BFLQz1Or}|{>XKefG?Uchg0v%lBrr{%^_a+iy_nO=$MV`zzVD(3cWY-KG&;3M%
z!wLU)4aR<YD;a?8?TvDk0L71&E`BHr^-b=-TymoE74G0{xDg#POqHsU;0Y^QYPIIe
z?mVY)wdQInrF9|Sx!0cviKV=)UacRRtw4tdcv~pvg5?hktwTi&?x$wG)jY$UnK!3X
z$e&-Bz88+cGo2!O20|BiJF5uU9R?II-xax1D4F#H++oekiGJnjzqkEzV2V)rCaUYn
zMd{%3hf1CZ4Y_|bzx_a^3FtJeqC@`Zuof_8n>V+sF+vZ=WeY@VWBKI!*33L+&&I~A
z_L{$XE`eZfwij)>=#p=Rw;`JEZrx0eE>-PkpDdBwZisgKko4w(L*#&HS!Zb4s*gMA
zk}41pkBow$$jp~V(I_(eLGxkMNHtI;4EP6mygWR!p;_D)gA)PEZyX^zgBXUD{d1a9
za8^wE1ddl*d?h|mot^g%d^EbJ9%vc36azCKsmg|bx2*6S6u$4uy+bGXsZ%vOzmqBn
z9SIQD+ugsdK9l+Qr>Ab0zIejwE4U|+m`FIuY~E5RMXaao)FMK%=>c?zNdQ$^=a<2}
z^e2(o9mCGwth8p_q}@-_9lGQg&)wJKU1XTHTy^RmHCX_S9_Z`ok4wqU>eZh}FIeyt
zE}Eo`%+tG#D01EE8OSZ<G8Vu)jRBH{Yd@aHPef!o+xisVz3T2z9V`@3cuS1t(&JOU
z(=qj#BDB<MqF)nHOs7cV_&5yyPB!~SGSSZ<+nC#-7JL{tERn8iU%K7lJNT{ILOQQO
z?5i>rg-Hd;c&bzmh33H~gmR<<5gDJvu>46^lExDB4Syf{ig1D=>%zE9^p0tFZRYb=
znnU-tOvZrzZ6=bGGxkb@ICTWjvneg3$D1MHD=nE)QZgu3Dd@|*wKZ^Vxm-k1jwv?i
z0#T`Zvf8&v1fLa&Q-TWBP4ZdjN309n=rXi>aDLG@-(!xL(jvpB0q~8cyMv;<J?J7|
zZe%f!t=5c%A#=LF^}AEQ5y!rU50uUJNaQ<HSQD==#WO9LbU=cukA4ZrP|d+EUlFMz
zqX@<+0Y2XGk0-y4PVy)&zv?oZnC@dcE!Zk|Akv8pKIB&-nk>1~!8}UgT!pZ-LCJ3u
z-kX~7b6wHAHddK;qlDsZ1hNkhqAi{Nj8(@9L=5h36UUgc1~xH5m1b1d%DUeu@L}6U
z|0){QL9w3l*~pqO3-{T|>Yvw2v{m%J@=Gp4nC3UlnYv{OjvY*qWWI*Pj*NO(jscL)
zNF)DzA|oB5aUF8oob6O%Z`&CPirOv%=OP&N&ZIKav9u;rM_3iWZSu(g9DV@ecA}r;
z9!EGtn_9N)k$2ohyL#Z5YxXg#k`QxZCKeP7cP=&G<QK$4BCEEor%o^YLGf_({nspB
zPkEg+GX$l%wHV$cvcV_kok&Lp;~N~`m8i?zMimL)PHStFk0U{MEL*)=5O#dlG!3qV
z6y5B0^yNgH1)uehzQb{w-xq_A8`pbXq$OCs^56boc)NVh?{S*&x2D%6iL?N$!(V-w
zVG^;gqR@^um$oY(EfuN`*6q@{eM)`Ncia3xLS6Ix1Lpr0ivO)C`MY~4*l9O<K^Y$u
zj&r?^MTRYn`DC7x<FTtJl#bpl*XQmDp78#3s`_&ey_#aH3e=X&Dbm?|B=V~Pr46?W
zBum*l$%B|ZfUlYkv1J4Z=Iken?Ype4Xw4N20=B!C;7*S22GW4|uk}Db8_<(is|wI4
zkKDjR$ef%AQbDQMfctr;5n__EGP*HlbJz87Vl=*w?o-X2$smiPhki@hmaPz1PAat+
z=WmY&INg?sH9))AQY(UJX$2%{KV}y|j~r40^Jg$F@@ak<Vet&_8ap<UE$a#$Ovm5W
z($WZeKBFn#_k6#i<xaY45R2Hqrwn9L!#goRt{CWC`R#>uOQ3b`k);n-jm9Ot`yJD6
za6o(0gq`_9KA!G4J<uV_rz=3)s4d3$$N?CTzrIZ((J~v@plfeoK>m|1wwe6MIQ9HL
ze>T{1Y9B?vp5k(|E5|u^9Us^47kIOsmC4vwfT?3biWaun9;956w%e{MING2IWqX<&
zc==E~S8E@r8D!usP&9=8lOd}tO6{rwqU~wV?S347b}=awPOJ93TP(ZDkC9HI=|8iO
z8O7_ywIRBq-WS!3^@Ljh4U!&<&7`YWfuy~m_HUyoekVrpzGsfYKYMmWHgY|q>hOc3
zZ2{Wa+IdexA?@&caqrWu52<$dJ;MsA&bGFmO?bpx;P~5v{?l|h8=X5ik#Ti#<yO%e
zIe-aTyzxGGr{CcrE%rs$f9`y$fb<1n&>i6YK4kxOCFcD0!t8mPzeX_=n7b4l=7cQ~
zNP=TMfHQs{(31JrfiQn$*cY10)%@3=1$d2+oeU;8dt{8wlq?UxKfS;|c{atvJ%7Uh
zK@~QGo-c_OZ>0;It$IL9&#s(gxvhCD2hfWnmcS<o_W_}|Se?RG56szwlU1r?L`&|r
z{RdS~CaK>lCt$1Fz}fpEV=RElvcvwMI1n7$oZf^zh4KM3m8EpFcn}Yu(!`jUON{M!
zo*h>SG+WCR8(dDu_DYtzsZP)%Hu`%^YyEFNIoJB6+vUf`PgFPJyef#;-j2>oOxTwj
zRfgKsa7?0~R`2P<zMO{VSG}{Pifp?k{(@#TIAz6wHO_fc)ugn_10vsiCNMzBd3)vl
ztD<#-cUBP1%jwC1AQM-m;>}WmP<NJ8#u!$Z+u;+L4aftOfBLY3?o_&kqE)>ArPvnA
zs^a3#z6oT%g$}{Y(4fnSU0Ey&(iSeds_)oK2Qze9sd`49n4T42Zf{areztI90!b48
z7=${COt@-3bU5^VU|ghbXpO-AT_mv8BkL#X`C5=9Tzynk6`+q~2gn|C(;;6c%3M3Q
zNz6KuC6Y_2!v6fX-a9T+oO-)DMe$PxiGW`9ExKxL<QK4)?`APGznkwh7s0>Kr8;<1
zzdy^eXmZuLDZt9i6=}Pmtp3bZmhX;-|H)h1>Ll{aihq+9bM|$hOJ%dBN!l_szhj=V
zjm&~C1<o`N`9>y6I;znnlksjfUn2))p3+72O9?VE>KLYUx)XnpZ$nrR)AFkT-+>&g
zE8dlTG-zO5SyzURQo!u|HTpW^6}Ny0iqPp7(CLoJcQ>Oo{9Jb@bzKqHTUxQ}lezPH
zGYVD6+)I*M7z2%_(8DNVfCtp)CM;UL+9yHJukzhhg=-S>ZTEF%Xzv;6;OY#RNSb&|
zjIxwM)c(9VO{FPS1hi(JIr4FTB;%=D)WX-;TiYE^lPA+3MA?3~Xie&YIL+5@m(`-m
zhN*{51SOa2!0l2AUT7!$<^!pitBKQ>qjTfiwV{n~IMZn96e-KbEHM^Hy1F!HFG;pz
zu^;m%nCPhkUX^>Mqu=!RRLl`<Rn-v0;`97SRRx$-XGUY$n@^hzmsX~2T`Fc!r{^BC
z!t;~ZAukO2^!X)-^NW}>5bGJ2=eZ#}ePH=-;ueZK$}dcwedm_7zoji8%?E9lQr0}%
zh)WE|p}N^%eG3>!wXuckWge*bttaU&k&hTl)C9Cl@wuO=XiLjrWFX8cLoCG3>b;X6
zLYa(%1~VU3G-NoL>v@~_$Peqpy<Rbz*V$2&>-&BuDSTGX0+0?3N`?liN+r$mX1yx9
z#4eJz9T+6;6A}P*gS^@GAlZyOC5hM~H%I1(o(#7h^1UlX`%hg@DCp%?>UXo0=e%*Y
zngm+frtY4cy1r%}vOs61T)5<v#r4EShO%p;q?S)cadosQkl>p2E^iQ@Vp&bF&S46c
z-R9Q6qO6@$l;SQB&texi(!;DHBns;hQ@>a*6(YTMl->b3zsY^M6iLjQLtCH+ksA-%
zT+1w3TQAbk2lSoxoDp#A_OO!msIoVg&4W5`oXpOfZ6n0}-@cA(3VXJP<iVIs_Ak_k
z*DYWcnp3M(0Aqvn{`np$c2a^^o(z&fOus1sOcMRXM%4<WDv^|d>M}THlgmC4%y^+f
z8zWk#@m72l`A^DhHuCX$6G6G%#MpmYVFQ^?&)<mo?+P1;0t_hhz>O%g5G*>uOj|)0
z;P(ha6<FjKfYQ|fNUbw{`ryuq7BIv0(}Z3%iu(D=E@w%=p(zQV?RRYcsR}`lkyoTn
z<aI5hoWR(IKPWcln^X~479c0T25>BW|AV5~mk6YO!?b~#96)6_M{W2gl+1%7m(DJ&
zl373}O^HNz@KKF>Vm9$M;_&DwEk>4s36KQ%ANVPu?k1UXqSqpXg;?=OHy20%$*(@D
z)3av~*QC-H52Jcv;lg`bTcy(jgRlNYh;myTAVdKHgec}?=39|_$~pgk-tz!KwD^;A
zg&p8rVfw|nBKfPBs~jL^5wunXh*^M<L+njvi*G6U^8?M@41fVL-X;L!aW0XJ(IzQm
zpIFp%qevIQ$AQnANs@n(%FqI&GStKXfc+BS-k%0YWtfowsSE-jm4RxDZ*lyOyitP5
zwB)k?zrO$ozkoP!!+^l}s6QyS4}kKizYYjc-$rK(Fy`E92n1|BnIC|lrM$?`7Cn3a
zK7D^#0-F<nqkjtPCP$e7lrms|Ql=`0%mFkSuZ6>o0mfmH3Lrymt_XQtleF^hFR_nk
z_y2=pq8<(h;GobS6g5-Dz<>TSCch7xXH%=YP~TtP;y)}zlm9SRiNDY2N?^!>_N%|X
z3NRS|KAqOc|1dnD-^Q@-_)J5h=x@8?-<Me<<o5vr|8)#g59$7QRsBD-sx%Y3-l@Y)
zumFVuh@NDDWg_ZG36BB##>Rn{p}t@B%y#0QO1gha@?x*-;CtQoLZ>IsUGQ%=!iM=3
zC&~XIyY@e5BL9#7DW0}w<_5{7Vg$H7NO%;k&Un39uPJ63u@nalimy@kM*YvYn*Za+
z@qa(?{~ZXES9MaaI^jbP!lNMkg5UWb4y3nQ7<zi7sUj%+<POf9cj5+#{)a%(_w1Q^
zvOWgueiBm$At!5E$6H;-LbyqR#K1Dx*kMzkL7PHEf7b#Rp1rL6V}&A%fsBiS*x_DW
zTyA+q7h-{TN$$|dKNmuOpaTFurR*rJ0D~<v_$7`Ifd!1y_sKooe)icKhbc;>inoLW
zC(0f<RE)WM7l<=xNL6TTDZosaz(C@xgU{(d@l9isnA{6F#TcK|!*xOBT-ky<z@FW7
zk+{@}HfRq((3`IFvo~1D>>3ipEbqFzD_14!;7qOHRcdQ_ugiO}P|gm_m-%C$w26b5
ziY>h}*{!BmE5{I^^=2MEi>l~7ZQ1gHcGvzTb+6?X!*{l>?{HaSIDWN@EHx3tH;66O
z&><M%Hqy-zQF&i$pyDV=+ho(M3zz|ub8&_tFSQv4DIRZv_$<0WR8OlmF?8#WW)%dV
zbU+G+(>Tsr-5Hkg%6RV!C|#oVyO3MAb8Dw1v7GST0uZe{*TCdhL^No<FQ3VG&{V#3
z`^|^(Wj~*57O#>oJ*YRi^tRga36nX5=rI4#fT-w-jSaM!dO%=D#gsNYEOAM{Gpz09
zaY)=LqZNw~-}_+1-!5#Iv95ZNDAj$G9wSGypNmJvDT^1(YO)dCJF&Q!p+^hcrnw1R
zev$O`NmE7mcH#Tfa!W|Gwh1u+pV*es9pzmr)NXy1aQ3}yW$dG7&`jEZau>AbjlYxf
z*Y<)7w%;z|J#T~00uvg#1d7pH4AkhbW+He`J@?KC&njJ<O0v;l6HVsA`U{J8+^~PV
z%v{<ccQ*C->Lw7M&{#YX(gSE%mF=-sZdM5WuN-@PH${?l^=6`exO03{%uucWN_~>C
z&va%D!sf%o`>Aos>Y>({Y%!FhDzv?cHwzl&@aB!~*9>RdxMT{k0yaty;S7pzzl6xh
zf`stZ7Y*tkVZQy{9AwV%PmPWEv=7f-b{7=4jTB#bd0`+)dhP?UcXRrCGaYcAUa}?<
z?0o>}{HYcw{x%zPth#ym!{an*m&qmQ^|NsocHoWgLwKH15W<&6gQNhKc)uvQM0#Ro
z%YtB}F5zsbaQm^+Bbn+qg&JaKoOW1c&e^xvJ)?P;#RyyVHUNqQn3Zh_wrx#-w1*$I
z{~YNGALh_F1)HLb(=D^vvrBgVcKZg++l%j)ms8^IpOuOP0v}euA-5jIL!{Qi=Qf?^
z<kZbuzgcX_pkW-t#jS#Sym|htmKK`Jl2K+_k%kcafS>_~5ihBR=9nikTR9!A$=8+k
zm_CiY6znCgT^}^0H9oWy$4v9&Y$_gt!?a`0Zh)EKEJ&iuSnKzGZ`l(U{M@I)Ty+m2
zMV%$J2PSbn$Gmm6D@$=8jXL7PLaC0#@T2jY>9gLuH0#_6nLhG4O1%Nz-}8#P1#K7)
z!H%9=1B|Bpp3hDV!9YunQTy)@xO>5bi;@?thzk0l=7D<cgnqt;!(?soYTMMA)Un<t
zKcW=3t+kqKG`q5~+e%&BiP1rPmd*;S8j@3C-p9Qb<`vbc<92%A&Zjv|)F|6r0*O=1
zdF?>)#({dc?v6%x)8xq|U||kR>_fhE3I*nzt0<XmwR+F~`f-FvnOd2NCefBqtcN#F
zTs%j5h~4C)*5^13?M%~JI=0%viidG*MO;cvy2YhD$(vFmd#VwKie6BdYP#yee~Rc2
zecygHGJkf1yQ>t(o?-fOXh^UoLDcL*t?5iGM+;vfLKqLo1+o&E^iK*%5dTl6=EQvM
z`h@KgQdic<r`UUmRLq7M^5_dpsh?YTd8bN<71->AN~SpO$iyIiOIVOsne*gPP<b9K
z{9A;)6W4$OOE0bB9Ea9}JdXtOYio{^w<LWZ>~6fdW8A^^&7CD2oxOWyfz_F>P<5Jt
zyEo3Y(<(^Y-#1;(px96qg<DU;+3?-1K|<x#<!Q&}6S4u~Q9D9j+<9|RU(TOur8=wR
zsl#i>-H9i~g0}OP9DUElE;h=h*}kAgmnq5(ZeY%;$`F@Y7tZ5XqvZ_Fn$Hlu9*pjc
z>{ob?y3j$D?JC2^SU{4dTlZ%?n{O5<?knv^s9vl%`KX<+Us#!$&9spr{9L}=K4?_3
z!{&-`vZ4Sw52ETYX##{7srKJ9(PSQOd~Lu38HQhxOIEm2<eeA&D$Sw0Wh+)L-_Y{z
zKV5`pc!!Ep0hL!X^`}#%u;L}ocpqd`69;~(y`F(_W442zGo^zCv$^0IYs98jEF|HY
zZdGC1Pr*8n|GdDZdsvm3WKHy2v`CD!W}YG`P=_zcwB`|G2xW*W*?u!&hneLw(u2ZJ
z4s+fx!;9_^tYh8rV-Kr|Gaq2J$dC2Eia=>Con)%o%j+}zYO7!As9EK*hkx;8)LG(v
z!E~$QSgqoPRi&%X3i9o62y>IF&HH4vA&-=YfG{unUQLI=Z1U`(=D4cFtoGudMaT25
zeB9UE$~vb3ZcE2wraM+&EFXVUukm}8qogBTrj9IviZ4PUy^$OSgl^r*{aJm8UL65$
zh_VknJxE`C5}y04rzJ%7<3|gZyVc4M4-r7AVhV^Ierd!9a)G2mSU}>G((9m^%``-F
zv;D5q-we6i$WP`tuEluj&Xp$XKL?SNdD_AIcbb5hKb8+jW^h8(NJ>O6EGle)`3t@#
zY`<sU_4MGkfPf2c-dvO!_<qBx<0W0sGX0_>CB-SLOSbv8f1V;17r21zfe`_Sp`Tkv
z5bF$xu(JuIc06QBvm-D!wl(3=$@NQt`kLHlBm2R4OdaHeg(*Ips96IPt^w!GA9}#u
zf`kg-Zh%B`sp>%Pyz{iIsu^%LJpocrYf<p0z-e2unYa!5%gxBbf}DG;ANEirg`IRD
zWTLU;NM>;bBzc_?*run;ILx9#w^H@aXP9U*HS#*`uW-&5H%RhtkI;|@BU-^92-%6v
zP<Nu)FwvkLg8!s?6=029xQ_28-BUKix|cg29?SJ}fddpWX8oR`8+a>3cOmXm2|**x
zdjf0Chl)glKPVpXsWv85M|34k0~)Yj+G50Gh|L5s6xJkIPQ}yTRe#P<VS3x`vMG5?
zUiH!adluv2s}z$yVp?&0y7j_Ptyi1l@nSA}Zav#hWo#|k%om&R^r`{`P!vh-qV&o3
zVfu}PzKd$AW`0RcPu}t-?Xyof1qqG94IJF|10SG_s~9R#w$D;72P#|<l9S!M^-*7C
zg4ADlD^KLorH{VNdKt3DXPP@ebXHLC3^OcUC)0SoVv7aSF}B0k@Oid^XrdHI=x;N3
z@9<hYF12n$uRk5^pVtqhy`3C=8{we%H2R*O>t^sn2~AP)KD_{VUj#>^*7)O{)Wh|;
zeQvRL303(UDOT<Ffv)~?Z=3W7p4}}Q){WbG#?&{x<zG~Y+osjO%xC0Ns#4jTw`bF6
zQ7|B*WBFbazAf3n-lEvkZ9bKyG7$Lo5Lg(Ia>WCJL=&7l(}wvh(V{EKncKG%aWzk}
z^xMlC_w%DK?XHNte#_QFUmCLw#@@iM9_n`iAsU*q=_ha81tXxC>h9%kfiA;r-{E$l
z5U)Y=VUl((F>7Fc?I<s+)P=`6s|$Pi?vdLAEhtS_yl=|${fY__*VsKD{Kh?T?VLLO
zpex01fCfA|M?9uY{y9V3m$~j!{=#1~dA=ph)z`MP^h+;#U*IdwE8&ysZsM9e&>oZB
zWpP$GigEOLSGxb;uJn{>i2>-&LaB3r+!=0wj77S1+um@f>StEnPk>a{{PmgNg}aeP
zzeQ-rPpttW-=4f!hzg=r0leHF#P<iq91zXCr(m=%2|xIqT-OL3VqJW28K^Qz+*f^#
zOmCtAdN$XC#=$?tfsk+Aouh~*Yof>>6e4-YKw4@Jh#h@e7bABxf*%=KEo~lMU)%Tz
zJp_V)DJv(X;GG8mORHV#fh4@@7zmd2?vW?6e_TPVf4Q@4r4^{BdHzupu+Axd{o6zR
zd()Zqp!M}*V?=Km=?S2VrC<u^VqGboDRv3q?tc6+M*IP=HwJ=~Nf%R=<U@0SLX4Bc
zbZo4vy1UPy)IDw1P2ZkGa+sYidlFaQb=Tg<nXn-z%CQ1u4A|}~Ng8@=2x~)OyW)%B
ziuk+yoRQmEG*chb+Hgzv6$E05PcJJiMcbT==~ijf@89TM20uZ~fUuxW%nYFX1{6CR
zy#t7b-F)JWAf_>9uDfUZ*HdZx*@~~+r=VgPeF0W4KxP2xwq~}?YVN5{MEK2}T7O(3
z(WGwK__<qoL2dGel`8I(G7C!Zo37BASG8SgkUQ%^LP5jmrbt2>28#%V3pwHt=il-9
z^!`CnbucmNIys*QOSQ2*kCe@Pxh&>-U(1v&hT`x@yxoThos79n6mnjAXVHqDOewS4
z(0kvC6-Qkaul13C+wxMP<hFzSV@v&aO%T-fUA)3JU$Hg({zwDO67t+M(F7<pal7gb
zEfwlu**L6mJzYFVc`ZgH!AJ`e5^gH~BtevN-U*pZvOj(T!x@C5(>nku29@&I7O*?F
zPk!`4yitZXq?%5v(N+FNs50%!YteUh_7qsJZZ8}n(OMes-r>rIjZY6@N+j8X!*eF{
z$Dg`$-YdT@(RBBdx&Y$h#lAZ)M^&v%pzt9Im7SQMk!`>SM|ObtkkR{zOmOD<T0%{@
zly$%L=&rKnnZ`J&O66O0D2X@d+mBoP?@%0fde*MlogihQ$l#ze#L&Dog&K5XyxYaG
zN_R$mS>Nv}oBo-&G?SmUI;1zTcVM({T7^%a+O0H+fy<tJr~-uc?f|iU%B~>J3V?IR
z&3kpidvf%kW1<B4K7Qh|(JO5>;V8SfS593Tc5I>6-p$pK@p4Szb<G|6LW`vnKf?C8
z<Urd2y(A5WTIQL#LEcew!KkXPY3zdExfYwAk`<cwOeLboBvo+GFodlV6P<k;)t<;`
z;gZEcuq*XouAO`^uyMSV``Y48sk>mX;Vm6usvRLWYV!kk4!9j*J~5`UnHFY?_1Yv0
zrsyOZkMBrTAwE0HpNWQ4l&&TTW?1!M#JYbx>vobowN+`-jp1!}ap!RXeE}yb+b(9)
z>tc7=iN1$FG$%)Q4dYZJJ)X|mWw6?K*K=K|e4-)orl(81XsW?Au*j*I%AG7iQmm>W
z0J-h>(x;U@rOLOzm%b`};&l;{u9L&cDwIS|UmpbDxpDr~ej__p^~%XdWqp5wc!xDD
zASk>^W}Drmm;5^Aj+FblU#w^5kT{UApy0GD)Bia_K-6gN)MHr5iN#^Z7~q*5RUC1t
z`QcSp%Jx-k_~?xAjU6FMvwC9)hPT74u>y`7*Pfl{TXI_RRihprHwkvmu1t>s;9U%v
z31FoobD|cvuXo8<s(*CJo_%KBzp8oaZKn^FbCl`FYI$AJo>rAjjMJa0_sC1c**_?R
zezY_n^X`BT_$Q&0#l1oDKvMk#FGcQFN<7i7%6Kpnz4ct>yrCta+4y&2KkR&s;|fVM
zk5ezlB~x~_MyR(g2MklMK(=O&!9>n(%xSoI$p%`AbT51R_J(mf>ntj@?#-heEiOSd
z-qm31#(a_-0TcQR&PgclGQWwOtD3x|Z;b}4l@HwDpD8hzUM1_dw}B3S{y~wKGF55P
zVS8P|ir(I-E$7Rt%IBpRZEv_x>4UcziHz|TU_9b5mFc1Qs{l1MvnyvI?Aw?-i15G$
zl<E!C=+76feTDFnI0-?K3H(d9q@m1GwRWOvBnCZAD`j#6BKllE9ubipJVD>r;OD0V
zrXkORsg%ubR2%TktWDLT57GM%*0;I#C1r|&?n2SBQmoGWq7|GO^2PSWX#R8#1sY{X
zGF5KOPYp|;r5G0le`l*xKJl?j9%3Qf4$?Z?uyK57MfAV4J6ky@w$yI^^0+30VSmz7
zWLVr4l=|HJ`<!fZG>!Hsij+Wfz@XUnxnqb@26$vA^2`#DR!dO5xR<hnCVpJtvob~a
z!C%NEdJKA3gDOR~*|59}CT&3sgp)7(H}$vb?E4cB8|Fts_b(vtWjzTk#!o<0z$#$2
zI<7Q+i+Ygb2;|Ci5GKPnHH^^R&cp-Ljl|SN^!m`r9t@1gBQ7B|@$TDi@DID=qCdg4
zMr__CP;tNg)TWP_r@L{>V6^dEXRnur(F$_ePpe9iB%@t1VzebkXZ2uGsaX8AyJRkx
zZs{^({Jv1Av6wzkQSqA&=IOt=#`Fr|u=`U15#D?#wvIT`Wq&RQ1r*irhSL>r5S~q1
zQJ84c4|rR_2Z;77BsKu|ssKC+mnE8s6Rz6~{&7X(CS)#y#q1N7T?zn`OWXfJ(NzH8
z;D^6l^z8tEso|~}f>5OL-izyzF_I*c-3mSuS~vy3W1D_prvs@y`h1|LKI#WR1@Rh4
znl3#GQ2QoRPzoX1gF_Je+6|O|f$RjJGB+T-``70j1`pU9nm6P75J1i6S|W@Z08p_~
zWH~@~_pi^{rhMJ^u?m5GMF!x+2tPo<!Nvh0xg`Gn7&NtXs%DTYc^Y$}i487(QlTu8
z1M|cBh1laGLxuO?y0%d4tRR1m@FLB~EA7Z@T_z4eTbU=~`}Nu<8DvUWD6kM^F+iyc
zle_HZ8tXUoH!1q*S7{>mdp6)@>JQExO(>eeC&fno(R)8SR{#Ihz!?S&BKAN;6aYV7
zfsbrP#i75=T5%QL@daEOzj`O-E<QU_Hk)q!%Uw5qlxBI%D95W8d3{<%bj(H9<FML!
zck_H~EX(3IL_byt{QqFuzy93=s?Zh4xA{29-C$r^H#G9606~f0&oea<UFE;dGsB;e
zS568_N_U#h5WncYk8u{L!m#xO$GJ9YJ<5utu6kZ&G$C6r>DC72(;8YJ_u0=%R*mZk
zdoq5tf2<BNU+s8!IPR!d2i?8-xvB|i{my;G`DEbd6s{~;;s6<TP1`G%=%=1^^};;F
zc~j-1^=l#NH?|8grZ2FjSv0B>&^BjBEQp7s-UKY^WF}@n*EDFPa0cdghr{?dB|ymC
z>}}xtGW@mcA&f92d|XyV!DQ^-yTE+(!us+V`E+K9B<7n8VYZa(hE748BMsLb4l`6!
z{xA04JRZuw?H`_26+)t{QxRpYY+0w0Eg@uym<riqitNLLki7^YrYJ&8vSpnzmL&T!
zj4U(ueFlxe%=G*8y_e^@>U%%e_jf<n^Lwu6^}7Ek!^}Bz&gVQo$MQbj$NR8*MX3DN
zX*nc%qB-Z%8wfs~k=ciELZfZeM{F>N*=#(s*D&IJLV$Hn-MYe18NGDH@w3KJr$P-j
z+AV8s^6gaT<9)GhVd2k@f9MktBgrM+6-1655K2AW=b0dhFKgxCuw?f3C%BXo9BOMj
z!bBYc8e&bX(dD^ED%i!oZINm(j1nsz-szt?Pjutux_|ZCp0yc^b_rr{-JvO*#Ji3Q
zFTV2xT>nY)y<P5Rr`_}2yxfhf@IC+XEri*({XNab*GyY;cLfw|)nzXJfT*qw5Q^Lk
zv)>y6?kc7h@nzVyI#Zck(O4xOO82Drt9zQ4bRW0zTnw_AXyrbf#v?Jtov>fkVuzzE
z@5`+tzJ??BrFSxq=nn6{n&f7FPUqHHnAN7I3E3zDh@}dFT;N>ya@pCt$7(A};`9=Q
z6e~@%)nh?2zb7Y>k;0qKh|dMlftD;<2t8=pqNkLq^ss3M;&@5S#B`zc$Eyi<J<LnH
zK`}kef73E7wL2!;<TUbICjwE-Tl&4WgN;Tcwz_dNq1oO_p|!~s;Sce_4plwL)+d(e
zuU2Y4S7c4Rx$!V_e_&o6Pxsoec+)6nSaTGyFO+sa!xp;V=?BC<1coiY{HmRQ{3a42
z7a}=Y<|d4@Ely}PEiUi9GqFWozb92((Pzo&BE$k3e8Lv3@LJt(??>NuYZ73!58nIz
z=sCUuKh*NDma9@wRX}-JqW-4S7v<b7l~>F63wz)JY3t6|S(wP=u&^Y2AAv4!k&d&K
zB`;G8)0~ykOah}V!_pr*bL~9<;aENtqPtOvDU?L_FrmKIx_ySZH;IM9ttO(0eu$d~
zJ?+1%)=#4g@AGtp*0-{yAp(L}tMSW2m2J=LL}H#^H|^fqs9q@DDh&Fhb<iTq@2IWW
z7RaE$kYW3uLxurT(#==!bI7nObVC6hgQkkP!ZtM(ApSfc5{fu$(E%qLbZnd%jX<q)
zfebFTFOc<5639NIv6Os<UOeScHiIKaa&kEYwc-yi<6wOG0oim<sC)T)_J}Q(HR|KB
z@ZDs_Y^u~*F*1RC&IZcWXzM9<ZSZ)YK3p2KoU9<RWx0$m@Xb33m4}ScCS6)Q$Npnl
zxLOGnP>&LAXt*DcTkgd_Ag@d<N-ScSedrZe>|xY`K|gFQ%7yAe$ow~()B6r1O9QN4
zRv1869y;34T*|9HbuSBjkKZ@k%JZ(@%~tHnlW!_YNmA}+_$2dBZ1)=d!wzm+MFi8K
z6b_Q(qR1<{2_+)bj)wSYU8=g?-p?yj4Vd5Ba2Jgo(X;wi>U>}i(`=;e!*mR(W{3*-
zzhTQ;>U3!IN8FguRLPt3Z$`g0zG)#9%8Ya+2c#aqFpxR{Vd1OiUWB`KofDScl0)+0
z>-c7wQj|lnh__|XM;Z!!hQ)<-X|JrTIim#?-|=osJDYuFhmvS=kv=ts=2<`+YMN14
z4&~Qrg6$C4$$PAu)Zby%joDH0>|(`%2DMzcRH$IomoFZ<AsWe94>V+ZQ4Vln-)ic$
zt{P?1o`o_j_l&TN@X0Q^63Kcs&pI$pHBzlmVs1(cbv7{Vnuv@h<WgLyLN`)|2KREH
zxL%-WRV-|cVMD-N3pbg`UfDX_>yCWe%ap@iZuTW?HzPH5uMLKtRrQ7#1YXmE1l^z=
zs;^UqhbGc0jLAWTBy32upC4&?!CIEzYQP!q9qo0Ta49Wd(60sh<$bhcb(79lxg^Z{
z8iAZq+B`M9lPTq#d?tRUD|G*j5CYZvVq$ypqPvM=wD)(!$zXAT^W(~-w|ZNANZ}_F
znnXPdg9P)`HQn#g6W!NRqZRM$Bn>&=G8~l^?pr*C6DmE8V{<)d{N`dTWMBSn2;@HG
z3QH87o#sh`ve0kQW{X@lQG@c>1M~epRacq!5!7cUO@}ke3wST9oWecuZ9m=S`x?KH
zO)7*_Z_@O0;4}-Jex%KqjYpsrYOimIuGstD#fi)b{X5H%$Jn|q7sY7qa^Pz=rZ$3G
zNEWcD!YcaVhq5Tgb3bDCAbH8dnMtC%ju!VoEo-Zvq%5^MRNe}{qNvSizchL7U<WHw
zfD)d*wU0JWoNS7pq~T0@kQNB%97YcJl=@^JJvsX;$7Jlhn$GbDHlvDr&Te`~K{hN%
zEcNkF=eoU&=Vi4lAb%z&xB{)RwBa%sd-EIXAp?o>wEp*xrk*_1dNLr$i(?zxoTgu<
zIgW!I)A<8MJW7M-8}bU-J@>`6q1h3EM0k1l(79?HBTvR{GU?XyN-xIRtqVJS_i0Pc
z=bzXy4&BRa4LTi9)uauR`wdvN$weh318cL&uF^$C2NnNDiG~}NpOtl6E`;7SKir}<
zYtvpyTz&*USjrKi<e&0DMos(;&8~@Bv6++MnLqgL*;`W264CXEs0ok5nafd}NCtKM
z?4~*RUilpCNNpMiC4!_~K}2n@Xh7_vkyb`6bW0iL{a;oMYOTFO?4Mq*l5Aj_oU*^w
znxH6L_IWs{txheTYDZ~G^>lPbQFb)dDVI``U*<Tk>rn$Pcv<t=9-d6Rag(R)P0Lv+
zv*HBpAY9f1HU04v9Ga7c8$a~1)>F5KiGNLVwXLb`b6m|fD!D+FyBJ&cOm1>mn-fez
ziE|xt%6jmJIig?*$3~{MtY|t8xx=X2E_hq%B&jkZDShh6-O>4~u&1L^>DhWBxVvh@
z!&fcdTO`l{)cj;?49<x2fo{gTDrD&9eQIR71xcES^MJwIEA%KV8N)mdB4l8!TRlMb
z8U<#U{eU!JHzlDP?$%%;{hYi-%6FB(Nx)+RQC<RQ1kA_PV96&@sgtAs_*mn=Ih5mn
z4*dIt{g+)Fj5N)N96@qVCABVIbWQL@83j}dn`fSCTAx-%l}tMg&)wDbZ41=Vx~W$6
zbegYHf0((3H3vSAPk)D@NleiibCm&~2=Zb!gWD?n9yxCJZp10r;)*}jpMv^()ZqpF
z5-m6fDNR<ykTT$Y#XVW8{Aa^$&82um;MCOAs{v`Wuk|vPn_W?T?eI-g4Gm+DH>8R?
zvojcOn}G2K0+hNkBh3EjFtfAnaDsJqKn34Q+h+~S6hXVzq0H|y8tk;#Twe?kNPBOE
zx$Ul>8d~HQdvkGjZ-`m{g@GNe+t2u(?eHJZ;5%<SyVf;o{N-sD2K%zF%aX}QNe&*@
z*ORGHFLJJ0eCt*`)gv?hK?`x=v+nLUF7ShD0@?ZLKOnp_=0}U{OH4iUG=&|t2^6+>
z8+{)ndxiJB-4GA5c5!T#BoH#xd8X0ab*jGlMD^(iH?KC6?vvjOW(tk@mkxjBI_pWz
z+wrbJ!gSZ>_Ky~tFkpAA1{?j9BG;VUj+Zby<46V<j>j(gNNf<JZ!}kSBDN}T$bHVI
znn}HG%^tuws*1a{M^*lSaM9Jt*chJhO5(7rxpdQ3tM0_jwOeOkJM=O1b7X9oo107-
z_JA`GE*{uD;&y>~;=m2<^V;6qK8U>b_yG~D6SqOC(lp<I0YqE-v*piRd&G}yMI1Bm
zV$-VA=G(9r<@6taLn?aevf4a2thkW*X^)|ao>%7==zE}Kb-y~dn%vk_Jr*#ftYpjI
zH(iF@46b|{w7o)p3m#|7+6<<w@yjR#ILz9S+GHf#nCZEP6IU?jvZ`mN1_-DX3f2at
znug=HLgU14&H_l5r5U|%%r}~d4yQZpBjbe+_YqnsjAfZIMWc6(0C1uo6#3*Ibf~G0
zi<0{w71e*;#b4FHMZMorpI3OxrIu&5r(;f*)ML~=Psfk>Xv$%PzH7viIpRw>qJ$6h
z(2a<Jm(<}Gu>0LGBHCJscBO7O2#zgAS*YLzCLK2R9J_KG#erHn`U8?HkhW=9*Zs=x
zHmM3~vDZ^e?>jKXB~pKzmi~7hZ#YsDKv?;i@f0<`-hkS4zqFZFca8Ci2_F)m8r#(~
zsv9)b-``*Rh{?;({EV<aXso6Czk~Pe=NNZ<=!FsR<&r|GB_EjQO>2)Y7{cX!tF5gW
zz13TY;a5r$d6MWq^gevgfYg3=6A-Lo#~m5m*q^YDAE>8%aal`~c$zlwS}Yzwn7=&<
zBg)bS0pzm`ARk*>&;yrND;8Dct#RA0f({&t0Bk1Xr-mtO>nPI{pgv%bbjPN6&`%&}
zW&H=_Y2U28aXT(G6F8_}pXx=`!zm@q$WZ_h)zkv&w1U6ELC<$nj_JQW0VSu<0l=H(
z128ICS=5GsZK9Uof^zEA=&w)c{B}|nuex+p_y3EY?sZI~D=ZdtKJn6;KgEurw>i6#
zr1@8Mm2gcJB?73@a}>W!IQ7Ev5LVSk=3Y!(+0y-=%Lxi~*B<k%;{SeFCHw;ZxBSY>
zly-a$Yhp*&n)w0IoQ>KD7(>%DQx-LUKs>O_^jDZSuD_;Nm<9cw?^<Rf8rW=&xL^Aj
z;3oK1c*!UoExYIp-XevWE?T73>(~p%+W+T$%OS-mfKA|oPW^xkq`yVY9_(A_8;6xS
z($kikpj+Ht@r+|>wQ1$K#c#dA9xMB8k5OPeL49e5B7%<kqCXN)4{L6cMtOXLRIe@F
z2k4t~al7rw@dHv(ixS6zAMbhcI58ubS}d4LS85K%AlaCwx@siHF`P8-cEv!Kb;Hg&
z8@87kE;btp;vI)gY;LXYE8062;*Sl}fcD8R6HLd8@_ceetY*{VsfPwIV!_=47@4+y
zqm!R;i?_BB4EKJ*DAHhK@tm{R_YqBtyzA$B7roRq%_yFO&g;A;qIf{knBo?6@R&bx
zv%=}<`_s=mSxS)?887Ik$y?{iMK9jX*@{(C?4zL(YZwj!BWu}Ink)Tu!I}IwF$kH3
z{Omj)_(xc&V+7JwJbF=qUCEMcSWG-CQqzAQ!)@KnM(uBpo4w0_QZ#m4QV@Mc)y|nk
z(BhFV1?fe^?qGKCIvP@vQbMFlFNQW6jI}9D_b<Bhe%VzK`Gl+J*duJ3=&2&|)g7gt
zSLl1mh)43vF>pSwblKg$m1~8j^;SrKc@$@vKzgczve|M8TjF*3aGqPO&-A(Zb>Dt@
z{pA6N2^q5(M#LO|`x+D>PmXhdA#;s*E-RpYRn?wx&_C$4Xuj>P3yEq^HuBG8-f^xL
z;dzonw|h_Xp(Iix`jF0~Yj0~rhy!aUkH)PKM+D$!+FGT!?1>YBSP8aju1^Xi+uh`)
z5+ZPFNR=|08PO<8g9Xdq_mTn)Me4YE!oo_o42CdlayGjoFU5|STD2s2(`zF%DXLUI
zN<=)&i6gR3p;KDa0FFyb4lSJF<)7eSL}T>WA4ZDG>=|f(MHIXK?9~y<^9LLlBs(cm
z&6TJwcRfY352=Xn+8dr)*evMtIBzTKYV5GtqfbNCB4>{X2G?H4fMiFgq5-{QHuFfz
z`F^$2o?g&R?I)OPdahosCy3(?UIGu4aYbg~dqnx?-uzJ#%8EMipFjT}m4|$NLdR*5
zls)tKek><|R0<2d0tyZnqKc)b>$OK^zuxudR9>66VWESfc_eDh9JcZeP*q3;9pN%&
zB5nK5fK3?z!uPY;kDlV?om((lKHxWX?I<8~Yyp@|08W34nyduYxDgoZec(tqgD0L|
z0XQxB*tb`IK+tD?YGSnTECR$#<25)ISbUu~UA8vAz8Xe^(*eSk1(1^D$%-Ek9-V#(
zR7Ywc82~S#Lq^O%@FegL6Y<7bHPwHgYk<+vSXv6`JhTJtL9Ga;MEu`;GD|s{zL~-V
zN4E%PYx(*7@<l-nA`C^5WkxkKt&>OTW>CvKaH0<D0JTZraD#1pm!*Su2H!ZVtSV%4
zM0qhMj<k38M6wv$TlMpT{Hw5Em%$<x$wGrspt#MD(T=w*#KM!K$mBD@lGoYRAJX&C
zrs2X#4~vtsO*pR}cSKG0xVJH-{rL2`%?wXd4Lf?)5R_sfw77$w9`O*i2-=wR#S36%
zKH=YC2aw!-O;2tUVyWnRu6cz~>`lZQ8KU=hR1)5vTRO*MO;rYA!Ci6$qYgD||9z2%
zLL<Z_+nHz0D#PfhXp#2W{%ITNoL?`jR0)??!v);u4T|^Es&f(TxA3~hO=C80t1TVy
z9*mieu}G$0rs(_5>(h>6d~w<c4w{Z}zd=)B&YW!BS(_J$3&+NiSWD~bH6hz>a9pBp
z0AC~HcGuO2$iD0Q<DsS(1beY@A53b6ci^h^W^%goTgx^Flq>u9=kxEa1D-WJ`CJ$C
zic_b*GTS`q<jD)N%R~Ck8>62;^yV6jocsZS_fLnG`X2=5{UQF-$y*7i8>6UtUHNw*
zM|aQEwM;E+_-v$8(CNZ`Ddgpk!UBD2J1B`JwVewH#8QnG>Fi`>Rj`9c^u1GNyMhUz
z_Hx5Mu{Y6ni-_JmBg|TwfX5Mp(_dPB9vTrk)2({{_IHT~+g;j<LUSw$!$a22{mE2(
zLaV~p)93ON^^S_^+}+Lh9I%=V4P8bg4J}7=YVW(vuQPY4X(z{S2C<h+ELx4!Pmd`r
zwB@-JTVHxI?=R}3<P-Hu3hMC8&Br}5BYWrgS&UvkU7Faz!|bu;j+!>;`EDSC|AbF5
zzgy~hyy{&|nVA>7>aO5;$hJ>{4_xs}Y6358eZACR8WA~dFJ8GGnDwyI?$3{2zj{%-
zljUa2zMyoX4Yy8OyEQx1s5rSDzo6LR=JKSJcf)+;$^*sbB-7G+NrLyxq^$f4a+iaS
z>vLunbWqe(af)?<u+o&qJ#;(aG%udA9&E8)pSgN;&K@xFtR#ZTrId;l`nVXiQs}ub
zPPL}h6Hb7nUIunpN9Ivi%tvYS&9>L4#^XhG#g4?C5)ib&GsVbC;dEEQlO!Oc=f85n
z=VDdHQ=emUb1!?x$py<<vn$Tx`mU_Ed)RFC-aB2bw`)?o)1)|Fe%eg%R&>nUwaK2`
z>LEP6S@2S{;b<!jm8zaG!5>!Yqgc}Hi8%5t@dB|c;*ChCshNSe?z_(WLxHf%?<;yr
zx4Z_~3t#CUG?vW*V%ya1h4^QSf%{n|K%9F-5&L@t4x+pe53u)BfA+wAh+sZ<cb!#X
zgj7(D&vkU3MD5r3$?Iw8{qKb0e|rp|61@QX?=2$<yC4T}-`PEwJl9DKA(UxYLIoM_
zyLQ0!dUdSsb$-B65Nw$o`C9}3&wTtR3h|q~b^c%#+)*vbbXBkj@n8`O^3k7U%{CLE
zBw3wJ=w*xF^x*?Ff1hEE`l%T&QAGL9;9K+hjVQI5{2NT}A(O4~Ix9-zBDgw`pa1+E
z7XhR%R!xauHVgx6t($|nUBpew4Q_5&4i}asd)f*qBQ>)}oW;mmQ_s|B!m@I=wHK>b
zV`9enc-lbWgN^_$A{h7tP>a!DW?t_vPOwhXdgagHnSc($1b{nw@lz_T{=n7`NQyk_
zx7YWr9;<vZzh?3H_ouoTi4-g|iUnAH`cYZzhNsS%{q}$Pjk~^vrV3-V{8*?)O`xc7
z0?GjZ&WE6aH`(?lC+2x%N(a+rd;YhN{QYRb&&vOE!apzd|GO*veF7_hnh2sD0u||u
zW@a{_P3zZ-1ELw?b^1k6{ZQj8py8S_5BJ_*W}EYvE57EG$y@m@%`dZQ?(6Pk8|}NC
z(&Obj{f=8aX%jzwQ(02DSnHtqZlG*)<v+MPEGa()Cs%e*j2JN>zSul~e}n&%Kk>3K
zMua284$-(kVpDDS?iWk3W3}K{w@(5w-rq(K7$;0yiR?N7FeXx_Iy&-^9zSu2nsK?$
z8oQWXRgO)C`VCfkJ@%btg~C%>TY@EiSyO}dPGJphWggXCPao{>hP1SK*&d)h0TF{y
zgORYW^_J=amK@#caEL%>fS#8vtq<v{im~aQ6$OqMA<1c@h`MckUeHj2PaN{7@UkGy
z+O}W3^np;eNx4#qAJu41C}=r5IL1D$n7mGUNjb-Oh4dTyo}eEo-^x04V$f66H0P;D
zOzohk$?JfmW3^S!E_o#B=sA_HAzK35f@<SUe0`Oa48ZR``V=_EZa(T0T)<g+JWuV2
zIAU>Nt*<YQL%N)i@!BcWy-BWCx5={+nPf*^w(1TNrkOSo0x=;-<sfULIyk=`Npe<J
zmMvcR;`jo#ZZ-7a=@L#QN_TZ$9Vboc3ZoQZJB=$hRiYnz=bGvOqtLzfvJg}lbcxkP
zOEu}UtDcvrMp?tw9UZ-`bL7}%nfPV%RgF|Y`D{JA{omO%>=>H3EC|1Fqg#(qvrSWV
ziGKzYVQ&=D_54fL_Q7%SS>WmvD}ew04Y~hs{?xEy6ksRDe?T0tTYSJ|uEo{82AQ|Z
z^2XU8kU<2q$@&L`A%e+s$^3ltZ@I01>(c%ikN*#eEBsQ6{o}j-y=p@An~prEtf(iu
z4jP~PXa5tn*#a)-Ge;wgdIk<WkiKoU?eBlOZV6rY^%r~wAG;_SN%3il-%aDTV7B5F
z;@^$>xymL{uf5tZyzB8u$8MzBWn~e{G*oJUQMXlq8cR+Ax5=x6W~yStWfsd**fRy~
z_l&QU4V70h-@u4DOny{n3rJB8!RYSsG-qr$dhHBqA`F-p0gQ20ZX&Q~FcC092g|&i
zd0l6<8R*=i=ubcq2JjeRGoZ}g!ir`p>CojzaYmbbu$2qbfNzC=)1@OO%M8-_<JHeR
zNU%&8jrC{x1v)Ci(`g_r)D&mxUjG51RQ+-ewc1h2r%|gb|Mq%Z3KyLMbQ-|Ii2CuC
zZC&x*J*tCEF~FP88|#X?I*GooWEFj>rzPHcc-K3XxLRi{NFtpLSV#)J{0@yZ1j7nh
zR`-vs2O}6MJK|}aP*dC7Lgy><1a)Srho$w%`QtU=a^Zxzt7DsE{T87?P#Wr^BFv$)
z^z)soF>3ir`r=#dL3XsX=E3p6Qiny6R{St(3M@#X+(cu*)eXF?4QveCv8=($h(k?$
zQv<*)0=;=nJwVGByWj~T6(W&i3rHh$x6BhHNuK{{M`QCStv_SN)tImFJv1d9vRK^x
z+Wj31iei36NQ}Y+W1`6Z$@_`SYmrTN<(hUGlXG!UHecvf%I@r!&RouTE5^sF-ZUkT
zk}pcgp}i87-v$Z0^+fT;TJX}$s%0!X@A?YfW?gMmpOFzHG_%=Ql<iFXKuYMJUQF)K
zne#5GNN4)%+__#K!*L{kJIAFVFccQe>f6(xVqRU~T&P(5ddzC%yq7{v)Jxogoj}pr
zw`~K1aF=;kB`b<pBC#3;v*?p5{h+duQn7a*^NyDDqN|}^o+87hm6{}gm8*QM<){66
z%NVI#TElCuGDHhCMpQb!B<TAtGiS9VP%7A5S_dg5=A9aptlF^JH0nLw@V+s-Iez(;
zi5}F@z`N%QrOfMU+V`H-;wvxBb8)=>T0CvDe@uUJ>Ns>`PU9Ha;(@gFi=M^Ty-Pj#
zV+O^EK%oiTQ}5LH%c0Fs9=aPTnW<{6`m{$HI)Cs6x}~5u_pv#uF%2GonxtiPwz{@@
z*|9d*s**<^UMa7@%vPP-akjQP?pRRpBXM#OK#N^uciN69q(m4NkE@^e-4c(G&s%zq
zwkthJXAe{!8yzSZiw_>Oe@F9qWz)X@@@AFdi%D#4YU>!|v~6?FthbePnw2xvo2r<%
zexy9oPHR)*!7l%+D{4}($dmd@@ac_=&My{2Pl)JusE6M%JD6WIc6A{&U)#)7)xN8h
z8yLONWx%^NA7S$uCnE91Z$T?y@t{T5m+RdgOMyG>T6u&4+vV!P|A>nJ;ZKVYx(H3@
z72S}$R6&Ar&Ic)=O;Z|eMaslFn)&;lp5&=33qY_rvmJ8gm(V)svi;3gU8MqW`M5K_
zs?3l&4e%(I1T0FSA0IL|tIHPjc(S#OEpP6lfiQW$(j(M{N<8hh5euplaI<OmYBYgS
z=nYWRFzs@}sDS-$PxfK#LIwgTNT2~m&(NXEiHBlmB!MK$*%7d1Cg#-B_y0@078o~I
zOv*TB;Ug{_)E=5{X3`W3p8$;csS6qaOE@==Rl}Ozh!_RdT1C@)cK5u-hyAZf+l%ZX
z-L5?@An0({RiDY1_-S;F2S~&SA^>2v;-ztJQk^$y74zEyp?+Q7*9!w4nbBH)_}|tC
zMod)`0t3+CP!zF=0#XBU<A3Sbex)6{Mc{Z;C++L6u4CRS-*%|c*lK{?(X6sJCq%sK
zesy?^fv0E!&#Gh6JIisuCYFQG?-c3B##@5u=gftFKqdZ<;OM{n`6ub>jMc_Gb1$e?
zidmDFnIzcey{e*o)XrLqxyz_Wwa}Ts)4!*y-`SD1fHG{PPks3VQVAT8LAMzr(m&`;
zeswc`&9SR*1;4NkE=B8>73v#+CUaUO=d6D33Hn?W1KR6O<Z&9?8W+uDIgxsl=<DS8
zhhBil=a19ryx(Sit*mhSxM?y;zdH89v3$GQ@;z^)uC5$TJUU^QK1%+h%P%Vul$wq#
zqc%J(z6@S}yOt)|h9z?Ttb}QCgG$)>5pb(b#nUcoD5AP90@g9750+k;`?>i`hhA3A
z5c6kS;hM494scToE&qVX021Is&=bfwX!Nt}#u5Nn7-`CfJxaQrFh7fKzdi-~?W5wf
z1^?nnWIu_JR2Pd{Mf`Gd*}tFE#LfRX-9Ikm|E#Wz)4$&gN~FT+CQR5UEB)p;7zX~J
z>?(OF)~_I)hita6d&2tT%Em3DMeo7t+W>#%Yv;JME|J1pal8No>3@J^J6f@hK(})n
zEtlB@Fnq>d*Iw5@`BU@`<}!z7A}f7Qn4fn%*wg|f=oVxMMB_a_qw#Fu(3)rw%#UDm
zY|3{<((+KzYd7KkC=!EjPTp9{I|HOa|5tK8ZSh<1*HpI~5I#%M;Ti%nBL^!w)kKK8
zcOF*O;=k{%OnsQBmTI{eeL^JYRCC2yjAaj?d}4%squIxe;`pl?M*TbL{qI|OcQ`y-
zvimLxmyT?4y>;5mrLs}0rL1rb@nstLEk}Hw*y1e%O_S1YHv(yMPT=gJw-pC=8q*S^
zRx<*q&uEUsjhaZcYmR{)%`WF(hu8QZmf}yG>KL~j2FT+lSP<K+&q<#Gr?iL|uU!EA
z>Nwz%IDC4uXi}3_ct1!5P@C&`HOF*o_+HMAaO=k(9I$Ga*S}i2nuU<NkC+UuG~9ju
zv-(W3egy7QZ-J7ew_cEsQnKajr}kH1IOQUYgpQ#*1dpLZZRPQ1*uG&G^VZjdmZkS!
z@5R?w>2o}Um7-%#IaH91aP#I{^>ZYfOB9We)05B!{Ln#*5+mtPCNyX9c^@D~Mpi%T
z@KmHOx5XR!uCB7^X3afHNu&Qewc3h^TM;^sen8%PM~DVp!~;#|fgx%T85<oQ$K;N%
zrN9Q7UMyl-?*(;VfVPf|IG+w&It9@fz`UAo#ZFCQmyCcI9$_}mjM!@?n&2kQgOwCI
zkZTwrD{5XSQbS%3lQ-Hk$Vj}<wZ`DZGku+@dD~H=Cm@AZCyGKpOfn@<EvG0bM({Fh
zO|uIS)bv0BW`P6TdszqKqtGjt)qn%J5zxWlTtiBNZ0jaS40ZtPJ@c>#x)BYy-YC;V
z<}KLbXzv`z3l>%C|2Lf=hYBMVHjy?5+N*s5RBR%oU3;qrO$F&$f@?mgdry)&7`k5k
z-=pTN1qwqd(fsr)V33Ao;cJRd^nKXAW{;s8fvzUuUoYX_n2N}ZuN9@$>C)PW175O(
zEH~?V;WOmA49c;N`o}}xd=bqA)9cRbX#v+SwMf@|>KebX)km)#Us|$FwbRIpu<db|
zcjzPay4iPIK})4K4Rdzz%GKy)RNrg&j4>AvVM_XnlQf6LK^^^t@9V|=GKaK!$7)WY
zbyfz;nF{2fio%(+iUn7qd<3r37f$R$H=ejtV|UXo2w!E~(_@mjXmVasBK&<oymaw{
z{MQd=;g^QyMt6b|qEq}fOJ7Iro@IJ~M#>w7qL)O)nDaMp@UVriw>-Jpw7?}Y62Hpm
ztSN)QX+~RDocQq%p;u_ZB%sQzZ<^j<RUU8W=7kbUA=s5+(+JYh@~<<#G*y>B=M{(C
z$JpWN`mcOVwLbJK)S9>Ibi3h^_|35T^E>r>q(vUT@K6@3lcc$lvp-d#7ZNEQ*0toO
zl)$%eIIW=r<>OLfh%Bh?;<UJNX=l^9>yU<b8<FDADfG0uJTjVk`~wzWT6hTOIa^+c
zdJE<bGP`=ldxvx$4U{!#J8AMFN_X835<TJ92nocG{eVDy9Z6!zUb6dW@p=@+=a|X#
zHsMXl51OEV?2dL|iLLS0g`i%3791iB%cTI_LrbqGjK<Z-=IZ(0=aU`(kgRxN8~1s^
zf+IWBR<KqSApu-kc_hVi{9v6BEh3ACCb5mOFR2a7eyZbDad=h@{Lzg}l0)@#ffvZe
zs~>O0J|}7;aU^3JC;{%r_M{sj*ox{#_{tOuIg%J}yxgLxu|u$csZ=$J_#MCWy33s`
zGnpyl(lT@7x;-*p1{zS#Io=&t;`O&KNH;uzyqwUD95KGu%lirPc<)pmExmcu8~xh1
zUR~0(2T^mLIIN73Dee1k?Rc3qcMA)ezkI@Cputff<hiiqG~$47=w&i$F^TMIjQ0|-
z;jiB!GGBvB7<dMEuwuKb9^0j_=Z4?$Hovx`D^tW$#yjHS<-oK>$iBo&Ig*)Sh+Tc5
zeq9E~uECo~iq|7>1>&vfRNXmt2ndHhL8A@E(Sy^Uuv|zBjr1&yyq80BRS&-aLt9cI
zd)V>F=jmJ9QIYZeRhV5!U2;l8!AT5I-_JPIsQCDup1Zuw1%c09ad-K$Z$rrbn-(AW
zM+z8aTfEFK+C%WJ1-gi9-mGHXkzNG><2SQZQ|aut+o@bqHySE-tAAN~Y1w3WcN-NN
z&J+lQQa`-^0E)MABq(nyQ593>Y|901nZVNiaLbxh#BJ-r2a;lHHAN($k7vamsCSSe
zn&Vl0#kDAIN%2uN%Cd3tvHZ&>p*TE$jP`8W31O{acR}5&NqkLbmn0p>;$ty&$B~D~
z9MNTG3&(oP0G(xav#{4Zs&${fdZ|a%i`%yj%01t%*L23nie^phpr0J2`Jf}=P0}x9
z<JvMYjG9e!5~GBHooZ`EwIX2cD^$DZcba*qMhh7r)-E9<m<??LpiXo11^c2YARota
zS_$cpSi;`!nruk+Zkh*su02zk&EsPFb9b-1#!o#Cb85Th_S4V5ra9`O&H5E2)g|>S
zjLx{u&vqXcfiKGy?_;S{pa9v4Fk$vyP|!ig`42CL%U3Vn+5{beVgXy#W@(<2HY1Nh
zs<&SeJF|Uk_wk6Ydh%uZ(dzmiN8j}_Rg?LcLLIr*dT;geIr|a$LOlDQyR*`B&-e}b
zp!neakS)_sSTIbCB5>JFryhvWsTt)KJR>Y?*F`NI!R4pKK64yW2qv#LTX54A5L~q8
z24SuzH$PM#;;)Fn=yfNYSQ3r(lq$r%xboh9Ot(nYB5dX^X#J$pdO&9dFIo^r4!7Yr
zTole}1yqXbK27go8o@NYztOkk=n7Mtx1)&|uAq7*XHSg*2}RTFMnFe1RLwI#+g{TN
zsyl)W@T!S?7wi7@!_A|66DpN~MJOJ_K4Nmy673Y0A0myjXgTeN29GyIztQyLS6aXJ
z^Y-&qFqDlE-hpnCC7z(W5eL=GV~k6W_Z7C0Z}uKmTJ9jirF~gV@TFjuSEfRjj67=3
zF!IU=8cK~Se1a_0@NV#%j^L*oaVu@ndvaHWF6r954}Nguo-Owa+U;gj#-laK3?ouU
z$QrxGVYirmYS3;aVbS9Li+MGYu@^6`XN?KgHQt^qe{dH9GjVz&g&Zxm-uP;ZkoGA#
zVq(2R^nvemnbEmjzU%d<jgxR8)Pe_~kMss;V{T%Ef9ixB2iz$}ocZWxyi*_dK<XWT
zFVMI%bySO|)pwj!5fXHm5zXX=9-a~A-tq>n{|Cg~<?gk>?Ayakcv@?<`&zqPEKXhP
zluDZ9S0W>~ZvS;LTAdV)->h9pJW}d2<9Golvukv3k>y+1K{Y<fZ3)~BybqZP#<y9r
zjIXP~gfs-M=Q#x(p_!Yp2qapSi)cJocd@2lBFBzg6qe``6TzgssUM)HJy5wM0pv35
z02`$F^<piaaW4tJ!C#^KU#6)4m3RNwOkKIAMgWwNm_>^Ofhi&!yH;U?xsDMbN6<44
zH2^_mqyp&Vu3Ge;yx(l+_0Cv}NAuSL#94F{j1MVEb~SomOEK$X6xL4U95x7l;`iqC
zjWZtQUAkXR4?o~(Ldr6V>$vH+@oNl$PXuaq!3eyxz<nB5^^|wPPTOa<pIiU&BO-xQ
zf-(Dbc@`B`n(5s7OauK(0Qu_&_)PBJnH@9mh*8e_98sB^?<hK9AlEIE2l1BN!n*C{
z$IF8v+5kt6dV6XH_gKa8^?ftSTey(M3763lj!$SVXN@BzaDG~L)2SJ}#b#KH0YXaF
z%SU+dnAY_Zmf0OC#c|(QO!UndcBVAq*fK!?H_8T*XP3cwD@4rwfHrM)yDQ|yQXeZ-
zLJRa+c?)mvW0fy2{D!Dw3DyFZJLan-CowE^7Ye#QLzeomn8GZtQ$vJ%KXK_-!471-
z|4<iV&0A^d5XZw6ZgC;_DO+e_)XL%^Y6Pv?F`EW&d^|!wM@cU5yXUA_Fg8jvOYyEU
zE>G>eT%Hm#Xm5HzRi#C$2s631QG5;>aMjdC(`|C2`<1QTqn?0mmGN49bH20i<IGO@
zVKwKm(HKHpdXaDQbwIbD5;`;DCelDq0;(0#vNGwY%!(ehGtpj}lMh~qO-eOsVU!s!
z5}QZ4YdE%KX;JeeL$qsPp6|FpiJb4Hk^Xf9r7+o=TEo4^V^h;xulq?|zEQpX)f@S3
z>kOV1mPSwt0O}YOjW#=o3YaLmnZRrvfAlrHngtOJJh7XLPl=u!?!ca770Ik#8du}K
z-a4{mcHQ%lw1x)VpairztGf6{%_molhTh5jKwgh7>)6+lt(SMv4cYsca9#Ug%faBT
zcj*Fqykl0jQ5ASrU=2X3iEd9jI+npg>Mx^gZ@u2dyLCApPJZYw#dQtm>ZNv9)%x2u
zYN!O?<M7)O4cx`&2^tBsRR1bR9QMGd-dkHIu4}Wa!+LqMBxtF%sFO{G+RQ}!k&|1W
zafAm7YZ(ETDcfy8bV+d0bM`!*fwMAv|Hv_BF+rSQh&RQZoDj~!;9Ez@z;cCcu7@(W
z`6|Kt=*}c|6F8TpMuk__+}_)6$t0>#X)68GoeJrr9PJy=9ERenE9UtBP>1=oYPNzQ
zG&2nx$Hf_uX&ubRZN@Hj+bC^Qd4BhtDKf6Cdbj<Kmhr3YKYF1d-(6ffI5-i{BfDt#
z)Av#mf$s(Rhwg36g`&$^C;IfjHW9;5bhTMMcM*M!N_1UO14K9=bF?W@0i}t8N)bPd
zXiYy4QkH^yfbgM@qJRR>hq_#S(08MZN^Se@5;)4-CA<1p4vWw0UY4crhfhJxbBBLG
z#!SE-d`z<&I~gQA<Krzi7yI@)7-<tI-E;1!)QX+Ym%T#TDE|4GtSaZ8*b^r-?Gbyl
zpI;SuzwehyR;_kTwGJ>vu@nx*0}%4BITzhAYSgF0vz6zUjld5;Ej&@w^Y-tS5%UG}
z08l#Xlx*qBU1bc(ao)FLA69(^=gYh1?kTopKt^S3v^no9W1PETu9$e?Lxd`PC|zVc
zz>6s@wqm15Y~HeUpXlhkRXc0{tdHkx)9n1qRZ7RZ{qE26JxUjrCjtj;Z74%b=V?Y9
zzYnt)!~e%9GoiA2Uoa^zrbZNa1&?I?KqoA<U$KRL-~%ECen^9dewVU4qJSpS?0*%H
zH%#_P#wREwHn$(ReJz9&A}VM1fVU2=_%aeloZQM8Nkg_jH<ZY@teUhtxBTl?T@;S2
zk(e`Md!W0Nag3x|t3lWmr~2sm*%Tr8k+B31qsqcBv#|@6NvZSNi(>2hR+cv^m$wpu
zhCH=ayUq#`Eqnw{&<D}%LwD}Cfr(5miW@a~u%ctNuB+PJdvT*U?m0d+zHpyL2l{P_
zW+<PzJOizCO=QmML{_b4r!wmQHlxiZ<<b*5K;3Xacoi`9)uH<U#{bM$llhT~+B_;5
z&AbKV%SU@Ben6hZZ~1&ziTX2c0-&#|cmREqn&~EVsZafKL^{4f14d@1s|3#9#>P(n
zfM74_qrcPyfL5Q77^K{{E}H2jibcF=JI|VduvUb|vQkR$a;SBEm{nY7^lb%6(^nW}
z(Lf_Dq-Ftg2!2RSs}d<oF3=-SI4;@HvMuDMzu_R}@SO4MhQ$=t@#$D4b~x#khEO@G
z){z(}cD9(guyXtS1|Fj$+HAhX`@NIm>Npt4J_qkHT{DT=04uwfqF6jGI|%!@*<PD(
z>uOcjg_Rq1;Vxahq{XO@ZEr{ngRo^nc9aEfvlY#56&G%<<@40i{{Y<OuD>lQ3wv#y
zLA5CKfl!Fz$b6*|>)+^mDCGWfG?s0RVXf6oKR%Y>V~}XuuVC#u-*L$H+GUFEoP|;p
zX5XyGLx)4#_KV)QG_3}@Mris>00yf9F!)dkmY!1-XYY33f6`M}kp>^@h<rSR8|T<F
zV=Ec{Eje5rVGvHaC7x$2QO?o*0rSg|*~|mLj-%HM=>R_tn*=`fWZvr0RgvyRr0Q$S
z8Rx7a=Vb3q6|D&wG*6ZEhxo$-B0Jo_c{fRld_u7z`M^kV6!{TN=RCQBFcjE9l}Tks
zKgU&XxKzjYf586Z&w4X;{%icjQgTq}&8!ip{wlF_w^z2Mi(a`nXa1q~9y$5O@g1&w
zti^ZYpMkMo@`E)z#Ci0-W@2Qi&>1+pj(f6mw)*qiZ8x;`J=t?<VNX7UH3kCNUbFuD
zw*ukYZ~Z39k39kU<A@t&MJoRsIf@Q1MX}EQ969P#&SiD+E9%iJmltTHxjcG%5Pw_X
zTU#uyV_)r5iGXI_H^s<1_I27v-hUmEJ0i*!{`o9V`#)|a`AHk|v6Z3SBg#JFSK~JV
zLrSt}iI;*y8|?~a$;0ObwmmQtHF*)-bgn=J8ma~(M&~WvcVji^c9%ICPH=vEXFO~o
zinIRkz?+@<EP?sHHj6R#_{_ifxhh6D)Sej?#oSlUnY>;17san;EFl)$<Y#V$EIIBI
z*0VF)xA5sI?+v|ua^2};=<py8nyagN#Y44|FKOl@-t1B;Vj1jdx;hu?6|NZde~ABU
zS2wO`b*ux&&}>KvI$A}R4KMQA`kcv{VeWQpcqum)@0$I^)Kl`L#QC#F#_$HOW))TT
zCpqRgC!242r#tMkwi+Z{EY{qq`_e>S3nG#KV0-WbaCxi$`sMAC22<$=oGQ{3iXwA{
zoDUVM+BlO5+U2XIU)xUmek3$qSg;nElG1B8P(<wV^`EXUrrmO5xtxMC9(tYT)Zxfi
zV##&k)rAAq+W{MR90Gx0)c@p0`n_=}&R+2IGExAr5dpz}|NW+(8gXK+2#!BM`AMO?
z>z=}6QvSudecFK(qv(%y0--oLUk@0EvoX!z?qrId^@*_Tjks>TD(UL|>$h)zo_v<r
zf`R?z&t0tfsdqZcE<IgAj(E5R+xO|NrU7qc4(|4QuIId5`JpkHb!FG-P_h6R3vr-l
z480%Cvoy-6W+Wo0iz9VGZ5!)t9<YDh?PUCu*I23L2mP%N%J@$@uzvCyi!9=i;sCZx
zqTrd$k_d)QsY}Z5*D~Meq3|J!M)VusEh~ycPGYW$tvEi>OkYUgO7f<FZ=ykQ$-dhO
zdst5^-T{IwAGdvMX>t+mx!6)591Xa1V&|_{#AzI)iIaQ=u*B13?_g5LyX7re>Qu_E
z<-zsE(aDW`pj~-+oR1*aX<@2$N-Y^N2ptQ=Eg>EX?`Ao?p{DK>y22HOQCjdM%Mi~Y
z9@?@&JXf5OXX#HZijB1fmO8l7sUoz!cZbr6vSH_&Gz`=^gEV<kKwPyGR3S3m-$9AC
zQ8jk!K`y8n8Oa%m6V<FGQ~Dh=G4#S<AhU#+R26{B9|iIAi*|`zx+;9E5TjyH5n0N+
zKtMzyuEgie3XHBQ071~D`h|^~*c+|}s#D-bYKF$tL2s_xRZ#5oI7sn)9n-cb3IR8s
zyMQ29nA4JSk$2Ig{lPC=T-=dLq}u%~0S}X>PMne%zj@@y7Y-ua6LeFNO_w)LV<O?}
zAiL?wcrE{(-~SJ3ik)HqPJR-$3#8;;LGaKLHNi*XV7xRelWXhq+Sa1xCA41nu)A<S
z>r4LVf*JceTAq8wZ=Y${eisxRf1A3JOFRBkwyfD`OK=fP>=N!O=r8c%u`}bbYsUUM
zO5^uw8{chYul0Oy^+p@imG>U8a3WY=OE%lDk|&}2c}c>0e%|d&yK8mpqeAJMCM!m&
z4=)<`mIy;OyGy0p0(UtP7#!fUe><b!XN$Pvr9p|IB8oI}HFNJ3J;CC44qhThg9SNt
z{9gtFxX-{krKWE|+Pts(Bq8AGdF~2{!o-!t0pU`GJq6_)Ns{G$DZh_D`E_cHEzUBc
z2}pq1ZHdlrTK4&Dpl)v22L&yqd%F*#@0g^B28rK577u57g1*LI$GiMCzXXqIIi@L~
zCnBN!qC`#*!eU}@KG(%Ot?{Wv@}Vmugv&}g_Yg-Cl!VS47KGLpBrCn!unQ0d`O05-
z=XYI3PdJi<fpz;!FV;D>|Ell$Z(qNxhpt3f#5Q2|iB%$U;m*+6lr-dK(Cf*-tPxWG
zob|PF&w~lt_oNO$j0ZTkv7{{(G{YQaH#og8TY~g!KOlMgUmBlyReH==Px$4Zh_$T!
zncAm$DRZwsruWPLF}2TC#2!>j7xE&Y(<<TMo7P46B8v_sG!HPlSn3N(iEl@f3d=)t
z05p#X^G4Y89eIC9*k-9>$}h-OhEd&fwq2>nKWYI*d>ga})n!jZftyt;9<+fKJzH=u
zS2vGZ!O&cxA<(sJ8c={i7=w(j8EmoeVjAGnXeAPVK$twdFuL~}$NVl^GcX9pj;zUy
z7RLc@x`Y(!r(SSBY$t5y{U2oatmYq(xK5V;k6(#T4r+t#z!Ec<2MILUEieI?;}~^i
zb#c^wOL19)I5kufFz(K%9B!8j?oR(}-$?EyvlTWly07lAuN?CLN|nipK7b=ZD^QM$
zTaGDeD;uh6J(JTNO^a>oo^f|o0n9E09B9NiHl$7z<;p-axoTyv^q@*K+Nu|!E~he>
z?vxTlQjc@{Qnq0CpR9KL@7H9zwNbDNGV?U8IDc;4wO|&;Stu=qkWZ-?DAr^z82TKx
zE=1Duq)Wy<|Bg`s@97}Rha+}y&_qmW&2L<w?4^2(R_roD!m4g2q>;g-7iF1@;~V3<
z<=M;+dhgg(sW9{VmF)U`aUjMN)TdFw7Rrum6j*SbxKencb;zDFcBpkXtX!%u>q4~j
zdp^>s)26Fz?~5TxZ5}D;;@a1kkE1(%xp5;>mE8*QEb}d-K&6I5yr;i$EOJS$sJ7hU
z<7{f0x;Yv0^88_CzM^R+fAd%bgV!kH2ZVMNWMvcr?6=$Z1sfF1+71%yqZ<$rLNh?q
z->$RngtI@4Dx6)!&MyA0Z}D=7d7P`<35V@hY!ol=TfUjxb{8(F!953M$J(O^EXH`c
zb-=h0H!VV|wzgM6@{v~Q-tyjNK}G7^*Q1eK+PkMj6Z8+b_JdvV>)y%!{R|<i(Xc_E
zMvb#zMt3d^M*<G_5fkdnJ?Qe-*WcjS;i9F??eEVE4bR;98uiIzr<1jll-^WrGYN~^
zLPY@<9MWQ}y1J!w3&H7}!f&5qPCIk)ebZ;l9iN((A~2UGf8W1<b`E_sb@H?|&4)Oo
zDiqKtNzzZ(XSF%Q<#3aMb~t$I=vDiQRndkRj`qgSs#i~(J8c{M|C1g4&$m3=ei!%Z
zESQE7>;S(3Mv@)GlFE}_BhAU$x%qHkBlEJ-;u4C`^OA3usUG<{{8C}5gO9eI6HS1;
zsj(o-QN#P_#&6+H*wSU{?pQBBd8(BW{RqET$G{6mX+dGvx>?1OthJSHm1=2U0@)@4
zMYNEhK*qMdbs2D~38*Qh9vBUbLj$ai4u()J(=JdBm{D`@)gyrJwE^@9m#L{gAlaa-
z@)ePTqJU{<nShy$-c&WBH*Gos<%_J58$cQQFTAHQ;*`c~j11V<6XQrtP>&TrG5B2p
zc=PK!7TI=;EjOR6J2uBJja}rRB6!c=6l&}Ms6BCSF{*S(?xr;J3;dQFBK<3i_{$<e
zll44TDpuRWVgyaYY6qtOm6NgXs%bBovjSK)%@Ip*WXARUfSj%a-?aGzu!sSS9MsGr
zY*k<uP4|fQ3;r(6vjXUhql;i|bO1s3*(g}NaWJY)o!J9h^G0vPu5~b#wQ4?YCKosp
zGBS&Ar+6Q_YfGrRV1zjKIX2`xZbxXy9=hZd?AK3|oJ0>yQ(=r;qzFaUjKW8006A2R
zf<~@9-gxw5iDK~<E)`X>I@ydxr+#@KzTt7={gSHR(;txAu#KZke!#duGfpPgUaF~}
z0_T|^M8s_Q0it%QYi|LSF1u9=)7pS`ST3*UV@WAf^~DMuJ3*KEAhW+qo}m>_`DQIr
zG1g#jKFrf~B2QAmGvUlrA<{yJS{V)cVnzuJo(R;mA=``#HI3s4rTuMac#Y1iO?W)u
z^VG2t<c*>|T+Lg%TO0Z9Ev&I4&$o$BxEe;S=1+c585L645H1uP6^fuqU^B6)U9-^g
zlqNX)BK!bdu2ao|pS&*2-)GfU3`9xpHpag@R$7W4vLjv$(lFxn#nv-r$TIKCQMUqO
zO?92!&b9kQJgl^fPw^gkZo#AF>l=6YlX+QDAtuns1AXvTQj*h2>xn?OF)SgJDBB<_
zeINQL2wIb7RPA<r?!593NVy3Qs=HR%#@@`~Xi~W5j>kHMK`yRXHl#Sot=W{kl<p<s
zl~mYg{{vF@k$tmYJB1uBy9b*(*1<z7_{6sVqQ6*){5?m;t6LxAT~=JkjZX8IoYCiX
zqVKMBmnqgNXe=>at+B%`X^w%O*>D>|hkfDk#@f=i(Vp&CBaM3nPqD6g4sea6Z^_T7
z)FqPpzsz65R#&BMnCrU+q^9;024s$1G`af%)7E?|y9^oonr_T^Q6mCPv7>P7Q3Br_
z<_(hQ>fNr+Z+U8whIh*f`UXFMR1&;P%N}Et=eix<=T4U_kU6-&6(O+%Tkk{Gy7M;~
zHwHR4)=E3%e^*yg&asAbEsmSF_m}jnY7nyO;GF=-&H~j!QC*~NrU+RGCsYT2RO)Q(
z<L@5*_3-*xj3XJ|Y!rdwQsiLw9X;RwIw+Z1_0ZtdW^?sjnDO=1Y?Jg-QrFtMLc^>Y
z(bR8UmckOWU}F6Cp$(=eML6+=X3R#~>jm?%_=0yBZi*89AnO=9V#?K8mb-%#P3i8z
z^>f>l1PY#r`2iU<JDsacml#jd)x4$()#`%t&w*Qd{gGE->lijvWwKsrxQe2%VC4k&
zJUn95$#q>Dvvn@0%U1T=?n<KUwxNn)=I5TwO`=Za^QYo7r)>)b(E`@Q9DGs46Kj=^
zS7`a5-RH^Xj_W=#%rm~+L;{vB$~@CqddN7DhxziL3G_x-1^2?PAmkjAouU(a2I)Z2
zX|fg7-l9goN!JHOw(`?A_C;vYj)ST6oc)=H-#N1!(|^P4-1z=AEY5W>8+SI+)x=qt
z-Ce34cj<};o#V~-W!j1OPg{FmRdl+SbqLzL{8T|6c_QDe5&UXTE?Mk@4V=0}3odd{
z6Cryo+(J|bl48w{Hs{=k;RU9rASz@;(lYskoo==*kI0o5fsi~&LN?CQ33b2_9z51h
z`g#=v(7T|`^haLtG4N8Im<rYoVJ8i?VCI@IU-!;t;Sl|huN)tWB+Y5ou$cEq=W)a#
z-%h2Giav$NppG@m&Ij287Cu@5@*Z!6py6rsj@{*6@feC3N_xN11)A~LGW(*6Pk8Bo
zsBJ=7Rm%||T2^lTy{Pc2#Dm?Cv2FTV{`Vf<DCDvGoT1n2lkWC&uz~sQfD}4+3iEOY
zahcHoTO0R;>*3qbWJCCd94rF6&iPZ0vMo>?9%jxIt-Z!Lq))rIcK#=J@$zq|hTHyG
zkKxj&Pe#~@rn=*(Z@^X?djB!Y;0I)v2B^01(~^Eb?shPg31&L?6)G1GVM|%+8A?nI
zrl_wro>}c$o=2Z33R<Ps(S+=n6{PhLYvzafeh|*i$hc&rEL<ZC;}{n8Mtdq4-rZR#
zp!a8Y6)S8IP23Fvd13TEBsVDKTt}+j#(q(eXWxHjl}crxrBMAy^G*TghtDhEeI#7b
zH2~-bIYKfFMY2C5{NVhn(N^@<)DGKeM{@2b6$IViSM?ll&_ycO_tKLq7S^hzVNCTq
zi;LL#m`mu-#xI<ZAkX2Dpg{o$bA+MOU{#SjP@{w#1lB7{h=xIXEz=}l)$!0a@@L@W
z5HKlc=Z64?o+LNGW_T&h7If>G{pXZ3%LMGvqY+<yG2KzbMUA~Q;SN$zHiZYN<RBZ1
z8IR*r$m{u<A#!m~0pFcRO<kd-+XE6CRf}+)lfdynxn*O@5HTtQ=c?q#DI?Nl&n)Mt
zO@t*uT7dS3L(KDn0E<d|0{y^G?Y!|^MjRkC{G>h1n^g_;-SPbaS*QiY>n&qY?CzUp
zK9tlMZ+@_VG#IP(xr(jAKDj?EHs^GIt@5PHZOe=Vi}UNLMwg%8>n>K?H5MO<-haQK
zR?g69?=Yb?K!MQW<>djLzB3jUEfnUj6D4n>*)YoC6?n9*)T-#^>o%9hYc*kMx|f~{
zx(^)w-gJ*pbLk)>hjs4zk)`ZDFSV*G)Ad_?jl(jpW%s!BZ*tTdaGO-SC71h)9y(BQ
zw_tlQW0x-?_I`RaVz9}^@rpqe=8>1LSHSZauBk8kWIK--s(RX1lZ+y&iYmz!7s(HO
z6g$K8Rc#Dy7Q-uIWYdmq_}wNz+RuuMUd8fjKu3X!80(C`+#umw<XAGSOiuK?-Hm%I
z1zaD;qn;b=OLF?6R(+whtqe1aXiT(yI%n!Kq}|&01CoU0u6=U($@_v*hv02^KKQ-I
zI#ZG#H6(m@)Q_;yTv<ZbqIY5`hIuN@Ijs^RaJ|z{RSmm$?cBxJdiQjrzebW$LwU^(
zlQMFa)ty9XS5J>_#K6#5r;kmH%ITbYZFiYGWR`~I{qri}6&{%f`pf307*pAPKKE_U
zZfS9H5;^_%*UnM?7klp=*W}i2i=wEgfY?B(QL0i!njjEe3W$hEml_qR5u(xp1R@~4
zgMfm7Qk52w-V!=0O==|6gx(XR1PEDovhLpJe4l>zo^|fNXaDZGzx@YNOx`l{&N-jw
z8RHql@ezOe$66<hV_fNCH8!osHQ0wBD#~V-H)M^U-sUUKZ5HP{UK8q}JQ&;DKS}?2
z_=^ZdZc_CirMb(G&Ol{#S-6EMO)<K)w!rQ2Hs8nU3qVn0I-42dq7bGV0i|j8y4A`u
zrn1^M=oVdrnin_|**<;hynOevir7~doIh4Dh}aqRUI(33pNDj!U+v)9J{Z%z!Eq(!
z{(c3fa{*meonm+HIP9n?Y0@Noq|x4S<!`c(eKUI5*fv<`uB6j_<WRE@52Y~dJ+gt7
z@4j4-sL7Ze-9c2FYgxMLi{_-ux%EO*L`14k&4Ap3_Z6fYbHU~HrF8OidXL(NQT{Ro
zfy5&TJTKUTYF&*|k5F_%P>$(vtYosL-`LHQDk*`nJzoUh;O%2Z`gIu7eW?RXVPN{T
zHf>veOt-`ZuD4x%Bgm1|@*&o{s#9L@!@ZAwUsDf1T)t{7p;Pc&^)B3E2;1h9XGWOY
zT2q|&wu)c1#P?#`)@w0n8Do-2vxQcpCI9!5*O1%fHPeU;^RgV0P;?z8TG_Yg6%#|O
z)}^z%o?}5i=lW|j-+g_SPcR)CQF9h#mJ7#Pml7^=+XPRhM5Ml)@xSkLj*JN9kJGZ?
zU(gyT`60VjRM}H=E&qIcU%2@pL+6^v!{@>cF)w`>s;KDn`Fj?%>Dp+t(B_Y|sfTcf
zvlmV@on?$lRAI7=$7?M>Oi=hK8b{d5mdU2l#T^7<#d3^1h92*nuW{aZ4rV@+>0SqR
zbJ}+FJo}>SL|SnoEm#S^8|fnscu_(3Ft@#Udw#4R{jHq^^B(@26M5lZA`P!K{-2tT
z|NXvL9z=wyEJ~H?_@Oa}THWfyNiL;zdr(XlOq*-pepc7KH~(2Fw$g}`=P~5=433zm
zdIaQ94*@iJdILstD*)K=-7pg1IAB3w-ZSD>`@HmCRb_g}1eDyL1`6i0pY}A}j(SPa
zuWHcbY=O4FxavLto20gZlnY<1t1}q&ijqck=mJvy_fd>suU7%P46xARDh8GT91sSO
z0qIzSKmk7ZA!9J;;@>^4M)5KI7WA)|8}-|XFgEE?Has~XE^9PGodN`kM$oNuaq%x>
zu3vA!vO|rpT~UV<WJ~!}5m2oQ0{uJsgP=?g{q;n(NuesQpQnQTnAHSyB89O5WxAZt
zAb~1D|9T!Txzk>krhr++uXnsx75wEq8x9vX9)3f1{Wg*y!~Eq$rHg&Ki;e#^w*M=~
zwgPmUM)`eKN=L!Sr>c1l!cLi-)OokN<Qd_y@;jC1!z+(lv3nf5{|3h#{2zC$;5kG{
zS#+uHxAB`f;9F`25KFfo_{Y9c9|NCu^y&?o9Oz(%ZtF{U#(N+ZLA)dgZG>(71oXbk
z_>acRS74i-K4GiSH0iMwLv*uL2MA;>G>+@4oo4vQuR#ILjK?wT-GN6xeSro%Bh1ue
z4j|*6VaY9TJI6e4JzC}2HC4%3FYF8r`9sDiwlY)nZgPzv{{1+t7}Jn;cmc-ZqV2Bt
z-BFI$Xtv!tm2#)PtczdjThoVz4?M^EmlL=gvlAKoz723+(YUoVaTpiPK0Q_pp+$qo
z7d_NPf7NE0WnOVbUG^NQAe|Pu*pN9BlA9&m6GsDG6@|SWzmgO8ephLu9|4Pruq}C$
zPJ(Pds|0N!(_O!*FEr$#Pg>^F!?I{swt)tZ*r|Kz|6OT~v0Z34qlTwEabRO(RI*>S
z+5Oo=xt6%`2#4*VE<fYI$0p+?$g#&URVVY+4Z<m!;BvW%u<iY{Yhc{~Lf!Q1dFWyh
zTAYP3y;X_>J%8f^KHrEIjF0C7x%jUy1#6o^18(gg)fY_g8PHbiZ`rEG6RfB^kw)&u
zvIF1{*lW@hQ~@3^33y!@dtE2jV8IZo(_RVbi`=4%?ic_6I`aQ6qcM*<ND=sh8gKhy
z*bd<XEq0>UdmwV{-A!FPYUl>e2au9hk!tPEH^<}aj~y1S&ZVrp1RbGzSf#z*ACO&r
z*oe@-kXa~lI%cj<SM9OQKhAw1-suUPr?gK?_sNAZGgN-9thz;s#37?@T;qxd<-5~6
z^gD2p@h))0oUym8F7WBW9j3|WyXu7U9V-&+7sfjp(tT0wack_W#!GEG`l!p)^&M4e
z9ud}vI8h!+^T23Y@dn1d`!Q_6dB=TVwely2%8LULQKL(Sr<e7HcQ!zGtl+;h_I^hB
z!B(Lt7Z3$?j{V7?yTn!NP=jlRGN_)RPMB8=)Mx^uLDh2ulKVqSS@r)Yt*l6N2(ai;
z2E5s8oXQML4qw51_{9}yQ0K;7tHBSLy*>Ar|APMJ)MKam_g~7>TC8gcw&!2Hg?{re
z;`^=`Ws_cG`<qM6hu<#=co^|@sDbVIH*=xiPXtViRFXnL==1knjJ)KKGm9V<<v#`z
zWvjF>==%A-i)p|48d-pgLP?}krEZjVqf|2Q+bOAhc!<cEOZe)DI5sUlByf%Wd3)3N
zb4`_bYstrVvYhq)9&f$iBkklm?BlEdK+APb-PH;Sv=fZb)|qt09o--lCG9su7mD@7
z_q!6&Cvgu!kpOyD+3^Flx|x13$b`a?F}0h!Rmxluo@#D?5>?cmdnwreg_0+ppGzfY
z$F!A>miUul1r*OBJTxMJt7*U(tB?2(lrqIQ<lT3<+8cMn^h&C*4JNmv#ETBd%s5=R
zRMfb1R?JVcj))@?(8WNySPV_$0*@+Xdldpf*Ms6R6-Sy*%bim=iBzcLAh_t7+bJ2u
zb~t;u3B{-0kX@vkp0+b)IO&#F-B-)>e%;p3HgaL(E!9`{viPU3<2aToP3}WR`d)?6
z*LknAFS7r{EdnjD@<!+(9|a`7!RL4{`g4gG<Ph@74a9U!j_{6!V1k>trfN1exFDk_
zch7+6{T@|=2iN2kb@~P|JF0#odKm70s(HWm^KSyB5wDq2;)6%opSwkBwlZaT-~&)5
zBnt>>I))fhWA8Yzt3>Z4d9{Xio_*YFs{BO!;RHO^;LyxD{5zT^1=I;##JnwBydo^W
zI*~$m7Ml`F%bx^Q9A*!DIUy=tFQRN5X-6mza0LU3JylILh|a%hp%6UXvy{p#|DtvE
z$cuZ~f6y?TU8WRi*VFLE1fTVZEJDvNM_1?O=S6RhLvg8ms6M<z){2RW-h#x0qoO0c
zd|*1tH<KDbk<q2}H{+U2)~<SsG;I>J3)1a2;?kQ>UA&WUtu?6CoH9_&_L8oGmy<e<
zjH<^<a1@X%4m67liT7n15A?8PzDEZHsq<b|2SVV-X^%kvV}};WEsZ4M$$EKPhugvZ
zh|}!i66fQ4b)s)4^Vk}yR{YEGmeSPZ3AW&loDusZ^^Akql&n@S&PcQ142W_wn`owQ
zxr#bGk=anQXTy~zb7<PUW3F(1nLrqFyZWre+{%p5tLvu~zxrWec{WKaX?}C8a@O(^
zHI9g37&f_qnU9(KY?+Ax4l6_LCl*Kd844)IFz{+VVX#*|MG<R6pU9{I#;!0&`6nNp
zx$6Z_-n@UZI#$hylVN<~E!QN?3&GyniU~-4qA8UK&sRd+bNduIcEA(jIVD+A-C;YS
z(4u|y?p3K3Whuvl)sLlCOfRw@^T~9!=$AHpZ<rYur67^NLUu;@Y(McRA-jm)(=N+u
z>+s4$pNfOS9Nn+xp$W{y^??GulrtgcWi%BAd#nzm4^DQ<hkWz6ITOiMx-Rjcks_ns
zizW)}TuN;qAr>KkMP)RgJL5s!lDFk{?p$#otH4@Qx%xr*YXK27Am6Jw!PNVp_e>+&
z`k5;`;gyk%-j#jz-ZTFAkC>aW1JkV&t&MH)ugVJ*C~0?|9^0#LW%osTcfwOsFT0-9
z&LKM#7P#knF5`&Z{rXZlYGrYe@v`wLTUM?@*<Ezb6d9BUnSZ`jM``<rTGt>8IBW7^
z`je%>jfhku=ho4}<eL<?IFf|c{ob`5jMC<<w`QXa-Ab~O#?PZ(v-l~pTxKgkz0ZYL
zIRy~8m##KejkLZIBA!G^P}qYjt<G0`4ZZwMY5&-jbMZ5U@b9Wu7EGfWAZHcAO7b+1
zP>fy?#EV*VzJ9zQ-1CRJ`co615Ne-(d+zh9c!@S{%aIh?m>I&hwnz5kD?R53QDeRM
zEyRe)Li+4zM;@WaGqwLcb>SvOGRPy}J~4YmyCl{wtK$A+Mwgo*@_z3s4)7Tqf7Sy3
z>T{fC1M=Gt3S;w0b7Z<*N!!(rq@wP~R}cLci*g|)U!PI}jju+1Y0tD;#($*AlTk1>
z8q<5KC!bHfL(!K;>LgS)f?8=s#fQ%(@nw;80%3e5wdcw=;_)G__lbsAE}vpiyuREB
z!o~fC;WsJFKfbQm5RCh?ybxQ%(+#~1Z9~nDF0WEb?iyjoRKz`ff+$^ebpHMV*&)`a
zn$F1St46T{$oCZwQw8F`3#1;CFY0nAC1Jw(PLgf%U263WANLuZmmv@cIxdtJPnqDD
zGB&qn)KAu#K$H#<Py+49$m$lbzhnC!=Mf*nk(Mp@zom51Ls{%kK)eKe>kBhXDfkU^
za}<x+2r|S6GQ3h2Im#^|B#9^uxR2GmaVt`A-q7%jFaB-PbfmRz!&0j&Mmw1fhv`h4
zEsYhLXbb6mz5P|esY(_@wj}(?AU5PP(%LLVU{fQD_?390wPWO7uiNV{+Q`Od#7fM6
zR$bRS{o&ba*%4)yXn0`{_5u=an%GD&g?;H+Sa?owc}~zibq5suEuv?Sf;bDoxs<fk
zUtk-zA=6`XKilPVKVf>}f<mT!Zd+|A=G|lfkMj>>XUZwkA%AeQtqq9aBR<09Cla|s
zTKQ_+CDJu9nfBF;=gvg5JfjH*Tjh=s?rMr~IgM1NWUhE85_38X+U$k~^~0X)<#CU>
zpB)t)7};I1oF8pQkamR^nvE^FLs#5N;ZG9dOt+lmn_3;p79AA@-YIV6Ls(G`WLUUV
zM`}!=(BuN+u<E&ZjthOnu;OzD%C;=m1)`-pJU}9ncIrO0aU+_-SCGw_)$63J=(yn(
za^7yh)nr5~d;Ew+ec-dN`UWUAVs(e}hH6c(?3DqDyYv>CXp?+Nr|m2DiI3LY9_#Gg
zJl3X>Ym0d7u7T-o!a8<F{PVVKd~R9V^1z%)hx|&8mSfJxIVDIdl#v={RMPuIu3o5P
z;`NfLmI*147>JBCz7~nRLTFXVg`J|=k=^~*;VCi({RFWE0~1LrzaFos;yf0AG4*hP
z1I|BBw95TlQ%L2cK!a#P8TF)<vs8c69m7~mKuiJAi_=Qf9lCu%I`gLYVkFa{a++(d
z-6$W`cjDqwjhP2|!K~uqMwrUoZpnPP_!A7V8l2{Dqh2jK4lbqE;0SWHX>t=)_<jXM
zC>|he2`N`Ml(U0MCripaUAwc;fgAWt%UVY-Z-Jb74nk=_?b0jCh{>?}=SJl_oPz3{
zEQjY0%QY>)_>@_RyLuZ9o906*uTl+ro9&gq58*Z+#~)8WF)YJk80A;o<x0+Llrf*B
z*qy3PDz^XBVs-2J-Tfa0KL{`}Wy5E*FkEsoz9?sFsbHOeb+Zsxh(f*T-3abGgU9l1
z>~&uU!QM<ByvA97w%L{RtS9UCsR)dax3}`J*Y+0AedQ|l1%sTcrDq%=$;LgGmk6V9
zOfl{D?t5885LSXb5V=(qr2A;o{lrk$Iluh@u{SS-3kpVtYBDCR1>%>WpU#+so12@+
zRSm!M8!!=e5k31zNN+M)qpe)Ef1K}9-#PM15U5uN={h^6&!t1kEZK$5jXmC<P<Q2k
zpV*MH2}uCb=xt@$Jy`f{P6lRDaxPA4S*yjsGgW#$uaP8r+tqxsr1&-=la%Idys&v9
zw4vwa1<_~j?%HGOeMw6hrC6hN6sA5cJctxfu=vv+UuG|~{%vm2rTR6=>(-EiOs7Hf
z^kaDg=m9IvL$#_k85@ZHk~FFzovYjBjp^=_k_zjpngHkh=9-DIyt|VuW@6!y3J;>i
z`WJEzUhHB!a@W|A$K>UC9ri2wQKC9G0ZLQ{f~aNXagf|I{^}if53Ih1KqhlrlnY1S
z*JunRqyEWT&I9Z^WNpYG;mH%S+T__k8B%@t`#RBA{moy8UMg_%nj0+Ll65x9I+`$z
z8^up{?X`;i!|^m|qL?$FIC>9kndD<!KiNS*`-FBd5+W+aI!~6ovP#qZ5P`csLrP`?
zZVTh_#T-F*DW~tKAJI7l^WWAoN9(<x0q~<1trMNw^K)lxEA>67oaO;y)0%-Cn#LFy
zMK!dP0wv1#c>vAkk(-JDl;kn0Y7d8rviaLdQEK`o<6L4+7x?QU^`Z1Z*EHT6k=@k7
z#hIU08hszY3bEJAb!odm3JTz;H}ep3ZVe1nF>io@sJjBHjsJKvmX<iOj#N-lgOz_X
z`h=tJj5M-r){5K-y?qNko=QJj?oC1NkX|BspChs3kSTEdtzVB1{iox@wd!!pg*6vR
z$-8J>iV!{+vAksTy|(=#r=iiX>dMx1jWupO&GM}HE$T<gQ!lo&pr+HLKfUYg^5*Pm
zYA+@GAS1UGCG+*Jm7H#UclfLL<?6CO8RV}JKdOC%=Om*)uCueRj_>*{JqpjlwPGUO
z-Y5z~xG&M4ow$kq<Qvk6Svp0Tjp=odxWV?N-Xh6e$wf6I9cu#eCC+~6>;!1IA>PH&
zY~GF)L4y;S*_MPTzE+G3ngvZu)U*pW5my6)`d%5gp~=m3$pD=}zz0HversB_?K}U<
zhchbl`VJ3rN%iq=HddI4C!*KvQ&+`c2bVkJyy-XcSePt~)4m0Wfc58>CJV*9hlg<v
zJT^H^{2njDlH->X?TPOIPs|teDd_AB%^L_dZs2IzsE!>_0IFB@Q3RSBYBLdXQNj4J
zW8&F0vG$I#gL>1O20E9bZNrz#@6|1DHN8Bwxi7t~VzK)5ujMpLvmI%mfpd(E-vu!2
z+?~>S!yTV9yQ698w5XYU_nm$iMe8Tfz%kZs`sd2Rzv(vrUQ0}%|59&!PDGE)ZA*Pw
zYDL!?{>g9}Cl8L~$^#(p|0b;S?`Ss@sDaljuXkM4-i`8s42{AF1oI0xtJ6P4FR`MZ
ze|vh@<G_hUHE&L^!y1Kv5es9Z9o3;=5`Y$bpehFivzRh20W<HGaFw`ye}&)cjjwnc
z?;=GV)Jl%>U)a#K+yOs8Krpg<v<$Kd>IqzQ?*eh-()bMCF=M*M8oNU9q>+}>3N128
z=d7LdxN3Qh$PWHl)Pgo(kK!B0*7$m%<jYXHXyq*Z5b}`nL_^ZEKN&og0eE*8%#ZWq
zQrqmZT|28{+@u|i74)P%O2ck<g%&YDX$=z4>%h!8sAw-*f!H-0-GC;dw>f2_Xve~U
zoO-@1Rb*WDfyT?Vnw`4CyAC>3ebs}1GQ8Ki1?C=DwvI}LEkVGnjE0>cK)nFBUY`!Y
z2<AT-zJJX`kGsHVhHBtfu?FaxoE?rT9Rv=U3n6Etz$^YphyjmG2%TMoCJJE$@5o38
zGm!;N)YyU9(V>%&CEg5b0*+`6pqjt}aMVMWK*np>I`=0-NtC~&vcC8Gk+H0o(b`it
zJeT<f#@(#rkr1*to3n+&KOXP^jHdm(3bxoi-UdBAy5Mf4g|bO*_p6T<%BomQmu^JY
zRcsV`;$Su2X_DJ3MFBd6mdbCjUP;i;v=D_xAU-hiID+<&RA#InT}So5cg}zKgk`Cv
z){az5G)=lMBZA{fZwsH8q1xG_Ou=#t47@Mc7YtVlN>u_r8JxjuKD1S7_YHsA(sRi&
zb6;H4x19WyljtwixoCVoZhK#rh+WymtHNE%K^r08v%3pQ-=`yA+UbeDk%N8mM<8PN
zjn@CHzan&l;kisR0$EKOh#%jDY4-|#?khFxxII!RA<#2^>WyEtJf>aod~xdQ1F>RX
zl%b>o9fRGpCp01J@g<v|`q&<NGeU{5^4U82!5y2<{yN7W%7Ksw&x4qNm9?}`z61IP
zHszBLLD|JgTUJ)bqdl(er`CzspX1X;o-Q(+zyAE&=G4HD{5s5i*yA*XJy-2{^;n{r
zqx;9|T-5#I-u1(>3*KyI{iQ~wT5Vy>%#S^ubhj92SN0o53D^me?K>#`jYU*p*r_&&
zCi!P~94!ZK9j-EaF@*Amz+_Z#JQGPjSI#2US*1v2wfXR*L$w7!J+!uX$lUIv1xI;;
z;3`4!B$M8ChD(CSWtJV@rn2@Csjl#kK$N7iceJtj8GJ6iwyD@+<l0NgnV-DD@j+Le
zTe@33%R`&h^-b+NvQpLtm-^w@+C!AH0T(@B8FR%$GuT>I$O+WFD!i%yN~L$a7*CV!
z8T;ZSUV5PeM|dRt_9S~~m7mh7dVxrI8F?jS&5QL$tGly9Qcg^^rOkS&=8fx)!Tu+l
zm_%QzuqhUbG?^}EO{@fuu{q_}h+qcot6XWH%`F||r4BW8$AERR#(TmymZ5_y=!I+4
zDLdTRwF*a@(li6CagoROmPGDH;Su@BbMZr|Z~OizFjVa1qZblP=`&liHysT0US+Jn
zv(tvbL@wx|E6jNYEGM$J1YzdnJu|-|tUo$%@L#nbX|dYp&%wQ8t?qoEz}BK-tV^xT
zqdrF*n{U((cyO!8!FSw7P_`#cCOGPz`rWIB5eg)Y;B_@JJth;Wu%~+S!>TX4zhpDC
zbATOED_!w`hw_r)r*4x}jTB|m7+cZ3ezoAL`R9`I^4ia(_U}Xb;t_H|+n<ysH>OKn
zbK7EaBZt4OBn;)`jBO<;#U%>p_snmbW$bD@-k7jSw3W}?AV>m{*QzH4tfP53gZT;F
zm)S<2o68)W5i;-%?V(6DYjZb#nx)?{PMRA1l<FD~8ON3)OL1g*%n-f*g@#}xww3Q<
z4pJE3z@=W44!2~bsF&t9tJD`)_6pa^yMI~i;IumZUSQ|PgKOjNiiNq_Z2G{26vjw>
zfpnT*(JA{rrdi#)_SoyII8V&E3rEBk-exbRGW8S(N}VPRFP34WMh3dl=cacSZ$4vR
z9InA&lNNeuJB#*6rd9=2aq1YRsUX+nM`K3A-M9~j!iPGhABIE;-8kW2VS6qTUQQ`*
z@|LM{H&w&FKe)K+5Ua^vHCR7g<f2_K?_ZH{#js8}=It?f5Sb+@wUW}<#6_d?^{qKx
zp3n&7_K;(C=8Voy*%5Qy((Jr+SH?mrqGYHV1GqCz?@E*4aI@s;ili%Tg69i(4XXDA
zd_BjqF2dfeo@E&HJgnMzbcCnC%adC#UQMlJeR`?;+mPoOWe7#7wV5!o5$-agf3v^f
zyjfsze?ik-igaC!Tk-zNiGqikJJ;;ASjs7LO6#pbr0ACIHTkRgf!y|vL*?vDR6kOL
zR%3ZpS}W^q82>b{(o6&7;hBX~9&j=9N&<H_Imc~yhP}mJDZBC=jW@ZDZM?$)eZ1fm
z!BkfM<&Uqp=OHMdTv^UqF~d||Ypbk=^eP|>69}6Q14ch-o~K0nGcvqjW(c?mco|<%
z&OD(kbySt?7jeVj@)%*{E1dOh+XiR1-sO(V>sCQ}FTDo9_R;Fi)ikeP7*{ZAx9!+y
zI<g>Nmb($7RED4NnLer`WDuHzAZp)Gqu4f>8Tr9Klz8*6cFr$f8O+c)ZuV~YN8PY)
zQEG+U%&1@wLSmA_LO=uoF^I7<ZO&qXo`QPj_NM*`LPCj8r5zEZPU*P?xgz$tk5}^G
zLpw!2$!k5koN{nOKSwb*U>IWXJu@Z9IFd~Yac6;<*-PSvWgp_*j)A=AO);@8MaGUt
z?YVhz;c*=nWJIL6GQ!?fBGXvjUEZ17E024&CNIo`f$3PK%Jl=sZ>iS8ztZf-7lM^N
zGN}GdN_?0Cdpwt*(HTwRmhDNC5dkMb!DIQ4t|t3Nv&jvhIJ>QilAAmO$^)HfidJ6F
zP7Z-cZdgZXs;A|pM{Is!v!dXI@d=E>Y*sKvTthHFel%%xpnGY1%L#tB)ZS>y%WYtz
zY{v%|GFPIj=EaoFJ~4I#z;szCSAcqa0|(nfJpyaIh)~)qOeBzWGegg-WEki?vM$;~
zZ=7ETc-XeGy}*&HaYbkniMmq<9ZW^jg12Wj5CmG6)GqqV(1r(E>J9>Jx~_E57hpJ-
zY<JCdTDxcxe=_W-iO??brPA6cT@$EfpS4(IcbEcx({Qhuu+?pkkA7^EoIz1&S_xA)
zu^9<Fd)r|5DPSB_`f!2(2a>cI<;t_WQi{x-U0&-vJhux}PxrbPr_!iy=(?*>TZgIc
z_XkvbMo3pTx(eOeK~#0SuNOW37F`?}`|SZ0_F(E+RLPh{S1-Nl&LAf9L;9j**ST&q
z2zY$y5YZ>}lhi9pyZAJ9KR3SdFZs?ZmT%rsVd~_9oIw|D;Z}{(g~=agf?&K*@Aj9>
zW&#xU;y;1U{|H3?`tyI1-#0*+&+UQA%0ms1RTtO^=%O|mPfu7VU5K9;<)?Z~*kMKH
zQ>DqaVY003rP_Rrn$^X4)K%3?XtzjBf}#KtQ?N0!)Oq+kmK3cayRw)T0kXGye|35t
z0g-)M(&%MPu8(4W-!Eh<dh7q^CLmAw_^H<3q%Mfvc4r=mQPnqNxL`fipZE1ONkjSk
zDEUQfkI4&ybP~voKl><v{QBBnPu)wQ%{?((`d2EzZ`2q;$WK3>I7X|?&s9$`c(pNx
z(dp{7a=0II&&gmUS=(`}9AWtJZ@Ryj{uJQ*FZ-8&Gt&A0h+ho{?w?CFp6tk+CbmHI
zSydq<TT|+Zejk1$Kd%(5SJEY~Qk=$yjLVm4>Q;<MHnu*--g!iCU(5A5OT0-(PSHvG
zMUb7^T>Eu_`G?>4ZR!4%Gqe3e&OC+tg{A_l__wT=j`|;^z3f^B10}mBAy;*ArJus+
zo)&kCFw;$XjuQ8Kp1WzJ{6^nlTErF1^`IbRdTQbZ27QKbAR+Ja8-0b7pT5lqobzAf
zul1??6rrqg)eROi6s7%Werzjk`mdDx{J%=ME7A3e=*f0E-=7S_HvmCg0f-^(ai<SV
z0fM*;6=a*w*mkcB<9l4Gh6s{Ye_FQ-E(o&c-1Nu+{PUN6)82Ov{36Jx1@R!oFKYXX
z<|wzoPD(=86~G#oJ~V?V80I5S-gx8j(Q@AZPH|bcb>I=F<*cgL0y;UG!oA`>$2otz
zc=mRm5snC2B<W$8C>)uwag3H*u+@#ZtP=PRTdK9AxRfa_kR%@wy3P>J<KsGsBY&iU
z{Y4GmI!4_`T%jB4f@C?Z1-fRjr%L~V58Tw0Ay2m_6JX9Z_;8kz5rJz<C%<WCl@z_a
z!99H3{KIk+gUs6g(go0&{?DfczU9zx^la6h6h4Ep3h=MZ)qVQX!0Pj9^GC0YzHD%b
z{hhZIx)_6N52dV74d|fIlD+`6@fpZqaS{atfdHKe$hPF341XOP8+tW2(uL>m9#`?Z
zv%n8K47k5v4-)nJnPz~_%ER;;uix(o`s+=nfRXy&?Q(yOr~cJ5;6#(BKEWNsH(hQc
z@tQSrrPhAjvnbXdP|NKQOdlBSzh-84KEgl^e!T9^<Vj88bUp1;7lk8RLNDZp19I8B
z{r{1#zel3@59`+V>8d^3i<tIJY&R!{(7Y%^O{v^PRDTRrMx(n%A!A<p`g_y4&zw0`
zRecljneb}OE@s=q`)>RGv{jt2rJs%p#m!cs0B5aEYVT=GCbDfQNd>2E2vb2+uH{Nt
z|27+HB(HgY6-L!l6$KCT90e4V7;65S8&KjcY92idEW5P*e?8wo^C6gur#Nd+*~rx}
z$~4Z=+aqJPNDzN~t|(*pa!J_PABD4-Q>{ikHo0w|79*X1yMU;Fahj>Mu_39OQKsbe
zs8B2fxkFsf#<{ZwrsN|NZwEZFu6#whkaS<<m3pTq;wT%2336T6`iFRw5SS(cB+RyC
z+HokEd2dn4`0Z@94i$hm!%>3v_vA~=T`<DqjkY5(*6B@UCo-x!TrY*nKj?gS&gNjS
zpzn9SS0--E0?B{tsG;K32V=lOm5Bl?qgpSC324WGMKvCnlg3btchBem>fXV<rOUWL
zS1s|uXXV7-F700p#Cq;@jm1PTO++{{Bx8R2=R8*d_SHf<AuR5Z^?|Ukm<o@l7Z{3v
zMzqy;a9H?#A>n@geD>C#e;yNzX?mDs{bLU)+C>OEy4z)sf%wAmuO$@n+vOWB82nER
z?f<etAo+;XOuOl<v!0fD1!gb{vw5Z3gs_u8)FT(<`Wh|_`YXNgea&$$mg$=3-sMIj
z{^sCb_`JxW5et~;|NI&4EiD?&ur&+SbLjCiLAWhL%Tl$y@%_7V;zN4dClfxc-wvbk
z`<T1)ou?$y1n7d~ftcN({CU`ra|hGV+K!I=Kx13*SM2<qsEgcA{Z|wpF?r5@`1{wZ
zTSI9Td{z_sSu-WHecuz7x5sFIeUcY{dm#SRB>C%;{F&H-J3t*Hrqu&6*G6sX-~_xt
zqjCEny}<F{QlQ^=O>vIZ+hOOlrSHFg%n)FGMdSb<d^E2M{m)Sva)x%5X#<cR(!u%?
zzXb5ISh!;5&HR6Bo<3H;`vy0g1a@#QeUL;aL5V-0#R|KpU^I*anigj=s~*%>Vhm($
z79oSl^bic)xD-m)%%Jo7@5rFL<-mr$kprL&SvxL>UN>liYy5<^L2L`-wk-RAae#0X
z)s1bj1IXCn0at*K211wacXog8)at(P|IhyCUoRh^WI+$;?PF`&=+)YffTyX=b?AS^
zMOtc;p&K8l);l){m1I~*0Y>iZ^&R!7=cuq*>G~1o=hx_~%P>#iLHz42;$Qo;<H(w0
zOemp&CpEeg-Zm4>mO>@3%}oh&t6%M{?8c{VS@WCRnus;vo3;Dy$*1^#5$dk|H$pb7
z{NE#8&wq$?wPSAv%6o9?o_<`+vyY0`ZBh!Ikn0LhJxm@YqDkJ0ygqAB+l8O48w{gw
z+r;e2SEP<n|Df|wTbqM{-TU&;ZKNYxwoM#^4}YcTBk~$LL>6o7&2vcX1Wn127D)(|
zjq*(^F%Fl1O}hvhZR^HbG4t8W_pt9BeK|Lv5pD#HMK;nd9FA<8kk2eCA1wwl_w12f
z_e37G`((je3ev_uE=eM8jAS$_23S0P<qg~7;hkE-rZH-}D!YBhm<T@&g`V0|48+$Z
ziKiN11xl;lrsS~(E@a!R?Y}gP@2B$u{MM6prh8f|MgL}>@<UXLWQ|&`_Vd<x-*`4s
z=~C(mpo$5e2TCwahG@avIDZIAGI_{e3DpZdLIYRO4EZ^qHtN-fypC?0$2|w7=byc|
z@HFQM?TmV0VoG~hAmH7Ww{Shj6zloOB=~IMkj5_9aqOZvmf)|IdB?BX?6i5f3cq(r
zyMBDNiv(Y(5^E`FI!j2_O8mb0fjkP+Fy^zJ?uslyQ$;%HA4_ZCl&B^`wg()|R(~uW
zvXdDx8TBOz2!zq=VT?41%fmQ75*|}kI)HPQ_zB(lDsXzQt!7qKtR*kJ$;kbqJl(cZ
zgDL1;!R;iNKr_l@hl6&U^43cV%QfqkbCxZ$&SiryNPbFj8~+9lA>Xatk9VfZkfg$E
zgaM}KC_x${h-CA}T&Im<%hS_Sg-b<{+L$-4m-^sh27YJ(q6BP1gSt-i?#MuvH>h4I
z2mSyOLcLBr=+6SfFLv*Wh4in141&%)0<b@5eT2`gxPq?0>@Y(k$mkKyVJ3Xiu0VQi
zsN#lF!Yb|~aez2nRZgoe&kmZZf%4Vx@~-@F)f!4;b6{LCg&RF5e8E|?`EKEDGpRwh
z@uZ)!9k;u5qc}(Gm2N}jgmKs9##rb0E)>gqb|zZq5%|hB2j*mLp^KmRYs{L*K4I$q
zWH@`f_v`po3H}vDvxuXbvKu$%IXDbxefUz1P!McL+3&Xco4vHhjrYU*VYH}}3z(bz
z0Cr1_$Z<j(L|ITQ13fdjdYW_Vu=V%PGg1JN1X2uHVkW9|SU(#d<0W9!OoV!u$H)(-
z*mmY?laz1xeWWNr^cxBqsPmo;JzKBr3SsyAumVW2rxf@4E{?emIG>qY0di(8t!}ZW
zUZ^Yx+@Vs&EpCqw$>WZtx#Yly=5K8j27TXp!9tf?>}h?Nhd!2$mZ>jDm;i8JQzjmW
zPQ;9L8|J%;Y3uD%iSJX0PK~{O&J`LgXH#AN<7X0F9prfX|KDh)0Q#T*-NODq^XLCG
z2V@{*e(kS}xcnt_au&;v{*g??>`LI_c(E_E^T2~fjdF_KG{%aabE>wiw2=NW@&p)W
z(N4S5@1&8^c9CHEXWaw&b`uIt)ySYBsTLS28>|vr+`axGmOG?m@Ps70ky`W<TaA0q
z{c}@g3}hyy2+R$zQ*=I>7`ePD4MN@2rpAmz4~)(cmR4ShyFeFBB;_@^n&`+Rgj~Aq
zF#*TmW)GEvo^GEBxTO<2caY`vGRVdz><O~%OBdeD+hQ=5`OBaOA--cJiS82D8p84{
z^SayqLnO_v+e3*PBHeORz4<gNiU3I@il$;EtB(mnUC%QE9twB5FWJ#%mzGGEb3cD1
zEE=DC%(X-pq<1Mno7m2pQn77KeCNr~Fk&&i)7!aW^I4-xbKx%ONW%T8NHrEM)-ShX
zpQ<sShDcITmWosna?nQN_jC$-SI9adaI=gfi7iC#LF}-g(#w0YXglwU_@(eglONAd
zI~_^~8t?}kAk4H=WMDimglub8wT(nU`92~@-nYhXxJ_?}Yn}Y`CxiNfiSv)n%NOs@
znG#zo4?^3Z_s>FDK(n>|lfu#hsb*H?s`cptm)_C&xI6D3C)rnAC7(Iy7C?~q4IQ7^
z4OEuSGo|{BSyEP-*iO|GN7TFYy3+^bUQcPADR*^I(>;9b@V$UcHsPLn3wP;rXE$+Y
zm(d`=34?Hpqpb2q=lOWzZmO!1XEp+QJBm&yCl8&~iYpOsV;0nUaj=!aINzt;hnJ!e
zh&@QX4dHgc6S~5d3(zI5W)Xi7r(+}~&$g5Onuz>Am9uo~PJ|z&zzEJn%5ElBl_pI#
z_c=~EVCw#IJ!;9qtND!l$?Mr#kIqLTJ~hj*CaSvR?XN0yJ8O1@A6xy7Eu7BWL9$RF
z1~s9>lg1*IjY&ZKV~ko<f@5v_0NwCxbaZw*sVLBCi-<dH9vyl@*VsuX$*wA`Wzgpo
z5>8SN$V6%NDO)?UCX`BI@{wa2FjfcYZ&tDQBjPN#o(|PKY2*9SqxS8jbJ9*6vAiLL
zn}`lT(4aXg9rQG>4`cKClIePXbh$qhT+R8Gd(<iZ#Q|HVeIf-O7MMk~@mbj&iT$us
z)NdfUi*I)~qBB<a8)=_sd45D7-2Y5!x7dpY;nz8gQ+3S@Phs4u0!W{+jidpZ97T)d
zR7zfGF!JFo!jc!2?v~5P@UAz-4omdKYBP#`<==WH+R)^8<UE^x8%pLg*um+Yxc!K8
zp@18iG#26`Xp-`1fz944f9R{TOOdwcWiE}>5s8t_U6KdmyC2lb@v^9mSCrOZ6wHcZ
ztl1;wvMZ0j(T8P6+r3_SAF6OUtovb=j-qg^8tl?RV3!f=Jd9$aYLgHUD#xIYfF(6c
zTo%H^cATn)x-P46f9Ld3)y~9@(vY^2R`?_Btjl#|hKq%X1wP%l65D35Q(@&PIswX3
z<MO?-yU^R@!UEw$i6%Tp`m|YI2!B)LvGto4LziAo2-j@Fs96(vI)geB17T9Y!4P+X
z%G4~-xa}S$fEHbaqf1O`w8fhAFX!I1dS2~*a3(=+4IKrwSrTaBQ>vh(#*ombz#Gma
z#Qd1&iV8sRznnj=QQc-Tk#y0@M!TKk+WS79L(LkXMWuNBHTLj4nz;a4l3${tH?dTX
z|LQtDf5u(dB~>fLCLV^0oK=uvY-K!qmSHuD4{6&3+wUWT;^-Hr3sT#fo>TLh5bqbt
zOs8{Z?ptUGW%eK<`!0lz#FO-}dWB3({MhnB5z?6k(6L}tIi6yyUy1dbXfUl9$T#Zl
z*3*1>S*`5em!|B&`^4LCeY+f{P)uYDtQ$l`m=IMcu5X3ETdrD*nVFYM`+DK|k^K*s
z^^_03xO5?1x+1}ZC+#i6z5t3<6IWLwgkVNQ1jCGQsz)h$+T*(#+Fn8OZksD_79}~v
zmrQtL?=!zRa_mK%Zxq`AT^c!Y58yh8V6gNc@*T1KsB676PNG@j>n{6@Is2#Lru4I~
zUw2QrR(k)Kak)9ez*gWRi#pfsql8lKQ-%}{!+s2oXtGeo>}3v1NeG7;CR}_Uo%h7`
z@c|yb^)%NVRiwy0>UnE&>Wa2k8k5P~#y09wNtTV;3#s@EF;8#ie!<i{Smu1k#!hkb
zE%0f>oy4BBv`d9jf<j$@kVz7h1<oZECg<LNAl0>|N5hp_^XDJkHPB0UF0w|XSHQTC
zGtEh7miD>sTqRJK=_*W3g#NN@thl4Pd!LJ$eUDd&i(19k$Eo^GcUa6XH;P%s-`*Ag
z+fVVaC?c>4w&xSW*O-yUf=ye3V<p4s;<!-cS}*^idkBB1&}H^5)K^CA)jk;cE0#5x
z7!{(->+X?1)<$AATpt=cX>Q)STd<q{_@tUUPuto1lP@#je0@jrxC(n@qS=6LI0?sv
z(%8*b_870`<tA|y6EO$owjW=*Pvvgrs1kUy{N6rkdfLS4U6PZ0Iz#)5*7FZE3vF9X
zqoNA;R#dhhs2hRHDmt3f^3419wQ?Oe+{;&gFwgQax^2WN`*?=1IDWC33KW?Bx{Yob
zD|SjGt+1OS3tOA9qGWW~L~lmVa_P#k(d#UPfU1+<%1?Y{iT-j!<`P1jefa6Mx?E$}
zXQ;2FD(eVM(WrmCsH!ipT)rf+xzxCiHkG{8LiBCl>EnrSIV2r<fw8{^$P+2c<fRBt
zkG2`?@QgjKtmX{7SjHPUQ59IPT8D3h9Ftq&Kr%H$SkF_IaTJ+)KKYvdrTkJ0ttR)5
z0;+LSfw1n<Qn;S1e1zEX@3a`lFHAmSOMPV5lrCT6hNSjQK7mhh+Vr;CWCLpt%!IVZ
zbJ=EggA#wC(A@B%I$p6aYtSD%joap<oaH!5T7<F&5yLW4@0CHdGU<Dt(9m-WIx)7~
zP<)&J-073w#vOT|UE{Hg8uxaeP*`H6tY`pqgxac3#zuUclfG|n$=5&i=~7LSc79>o
zW!~B35LI()gW~E-kJJ?{m?xM*#Bnx|5Kqf$3g3>_08L^d8FS!1wb^9?{b8&n@m?^{
z>Vb`pDBXTAe0s^%Mv<U7SCTe30K9-p5Uo+izZM|z#I?<;ggEwua^g^&jU2aikE3Cq
zM^TNw=<UMlh6MFdMwU;Q+sPZ6WM<Ua5-J?Q_t7fLWULV5m^(Cv+jwgl+HP~Ef%y3j
z{h>sKpJI!&?=?ON@(PT?<O{(1u0T4Zmqg9ULBg97xZZlcgr1bU+4rb6mK!-83V$-(
zCV2AVFz@Y_%q#nld3%Ca7<!+QMr4g^A?+R|@nzw8%CWr6SpqspF6dKncfpdioq)`>
z*CtcnJR)s|k6qiR!M-rcyPN3ZC_s`5M7e*Y7B{DgQ!HO3>Y6!lO)XwVgd%IoA~GUR
zM;NetTJ1k#l_=#Kj$x;_s&Z0}m6Eh-Wty?FWVaNOSBPZk5?iC^=b;}C&bO*<V!AWj
zA~JatLoy`ZzD>1b6fmW@<zrQM6P5~4E?Cjx{=(9;r4m+Z(KoCLw>%}<LXxgz+6~?b
z7G$oBzrSDU=!~)mdDDVA1`NNl5{fAbE!)tb67<PN`y6{8-hs!T<flr8G&x_k?bN+N
z1;5I5vb8yO-*HpX=Md7uRVOgf=u%ha5KTn}{b+7OJkRN*mZJE^El#b=`OkSC`VQ|C
zTs@WsJbC7P7%68W<O)<tt8fg5gMIFpP+NxJ^MV>9;-h`nnLnSt@~BntXE?(?N`F)<
z^bE|WaqMu7CJB1hfy)}3Me6?!8q)IAPgrVXYoOiX(;-aK)g|&@dPLd}L^h_v3wM*&
zAV;Xl8>tkX;2ME^2j~eJB$Mj-jaQ~QX^OU8^a<zTq}EZ8vbfm(rTgq)zsU=mdVzo(
zl>Kf>FA)dUtpHG8n2HK>?b=U8H|slW7tnprqfZQq2QIrN_0^rINS$mM)q8W`?zIC`
zxifX%3e7-cidXil@_8?c9wIO-%L0ukZx*iR*ZSPO8*L&nd*<lGX4V%z%kR+O?@ai=
zVtV7>JNGdvyZPwJ0iVn0A8+yjiI$~*)-e8Vpfg|!6v_Onp7ekJ8x(@Gh4EOnlFM&&
z%jz6L<~1k1<wM5EY71*l{L%ZRt?KS+mP<3GUI{T|%!WTDp*IJ0rH3ZMxZ8#~?Jz$U
z63SAwdl>V5$}qyv@Q>W^gNzJpw@tV0ZK+tbyD90hMXPN=mn1WD`|fI&Z7!p<GkqGN
zr{tC}q^_{C?VB1aKd7)<kj&MTM*3Fa41oUpTxi1OF~StVR6)uVFa)Hj<$qhK-u+Qi
z6lImP?H9W0FhKz|>=51J!Hh&A$Qv876Tc9F%a{le--r>X!t1t=MTojol=hNLaGdhX
z(2&CUP%+=Kbz4u(Ct&c-|GUog!GF`6{-^H`ka=Lw381+mCXsQCo;tzjTgIdeib5x?
zX6NBPq3-Yp!&_ZVz=359lwFowt%+}dw#|qnP342OUEA8V7J#my@cZ)Ub*lW`n^1pM
znT4nPt7TY`qA)$yS(XedsXj)Sb+^T<qKYJF;|F{@slc!Ck!+qeG-hizj}B<@C9szY
zC*C7<S`!vF(M9XRMHLU}-z)a*+vXy<qPdXC^&Ws%&_+au*T{!$E`v={dnA3EzO5X(
ztB?*-^y?8h6ger7qBoj&TUebrGuf34xH03HML3yy!Vg;V9J(w3q`>q<b6G?KI7NIT
zHNQerBv6x6T_Tb*ed@|Zx_ij;^?yZ+{&#xxf7<uItG;Y(2=%kfTNnNhs4rhguXwT7
zB2L-0+-agSQYj?Z$@?@FJk2^|L6xg}jan!xRv<6KvQciD8Ji#G?#}qUJUZIo5>!Ee
zy%><5%Rx2%4?3vEBS1ACZW*O!mpu^hQ>GJ2bOU%)Fbv>Ptb2Hr1gOmh3;_)2JQ`p?
z3IGEt#b~5Qy79FCjG|g#_u#BEdvF#b3cy)yroX)%-aVU<5+tKMK|g~4I9b4+wnN<*
zz{v*w)!WxRFU&k_@4E8aIWvBHdo5q6;t%Klx?qX_8sUF`qW<R}39?Q^syqeP)a9?L
zLiTw>)@s=xL<P$-?+O#(ZM*v(M4BvRhH(8+m3--1=aKW8El+`;a@B>8&!3-i2b)o5
z^#5%g4hw*Czi@jFV3!9YFKG2S_WabrQiI+PnGb!r{JCjWsS5Jw2qeOa{(Y<%@=|pl
zO^m|U^e2P=Mg)1HMfFsfQjMOy!YPRq+Ed!9nS~i*bxPw}V@&k3&V7gSuR<Pm!=Uq)
zs@*6JipF?gA@&4UR7lMYq9e?pM{2S!^TquSe%~}+wCvOPiCVK*`sU%mFo_4Nb@6C1
zg=@0~2Ep<}UJ@h=Cc>b{VfN4F@s^_d;p`z2pxAywKGWkwhI0C~LmT^UA3Mvz40(6j
zxDNL@tqI=s0?@?{B~uOo{iu%U8a&3OB-uF1{PGCZPpcXyRJd&NUV{65^7$hu{})Es
z5^6lfpUw_yK7py6^sn^qU?b?#ujs}0Z=^{r`WTkPtqgtIdb7GQAr?L(lEuNy7ompw
zS?JTp$6rkui$X!i(Lv9sWl+Y?t$E22K3ACs%^Tt#_4G9@T_B7fatkr>J*s2cL(z#;
zRv<+|e7ZaXW?rs{;3%veOD%EFKIllKw16VTpz~e2W5wB+Z(=uf>2r_NZ)H6wT7G5R
zwV76u6Nb_xAvjD>vQ!HZQ1L$ss&`!3dCv+Pv$qn9F4(^-jmU5o)2O&8Ze=mN|Eaci
zp%0!*7Rrd?rCh!CT@G^G9?L%<?h??0KLNWgxCyn1QLd4n%ZOIm^4Rz0Sp{Q~ctdNg
zt|ccCM&NFmnK%?M4&g&xK^z!H#(rDg{G%qwEKK(d#O82s^X2plp=u_l$vqcm&*A|@
za*&hUrz%Tj%|QB)>>Uf%5oc!0PHG=%Z23GQ$^m=I&7V^G38JS9<0jC1u#)A}8w!nz
z!hjWN=jCmPI{u-F_h*T9#=xzaO^EfO*;wgH6VbHAunAV?jmNuDv}gJ7MB~INsh2d}
z0yEC1f>qTQy(VwId&=@zQH}S|%;JS9W{86>3~Ea6NtGmb1=A$p6e;3k?4fRDoeyf6
zbA(cC^|8$JHLp&wC_Kxzdnl>?{o7-U9|ywyjce3)Zcr0;Na$c?PVxe=JykKTasNJy
zgMw1K>0^%7cE6kDFRP#_QVBOstcp(VuS+sw=9^JN!GPPRT@Rg$#!AIc>{{LJJO6^F
zjTa2lu<J)yl1R&ffq45TNAARm^1QwHRfHjF9nIp?V>$^+f?a{iUgOn}qmWZI3ESEm
z@uVxIb`jsyt#_1(kHVW4gC~#k9x-?NdU}yVWjb`Lbpzm=(TPh;Kv>?`2uQ2JMZZ&;
z8>u@Rbz~|3W%gusa-=Ubd?wt_Dq+JUBL0ojpxT*86|xuO?Ru-R510-|0Xk<no$JPi
zAK0Qm+;zAMbd?Dm*-hK6u13dgJRT|LEjRA@i83cA)?tp&#Jg3GBRQ1~iCtV4mTDP;
zRL{8w*V+&goohc2^@Pv5%Uph`*VQ?E=0VX~JUfx^!ftwv=6kp$581oiH{S^>Fwdns
z;L6KC(`4_MGiNkvA1s@g&2gFI={Mi=f{RbaS^7+2ORnidpoq5_Ow-LUx8v&aEwm~g
zaN+j)fHCj5+~TNK^EDve_GgBH2%`PgxkvG@?{9t+tCNKgY%Pi?Il#l4kIL89kl=H%
zN}kK+eKs|g*Fb16Vlr}cisM52-39I!_r4^*;I8a&9^=3{u!W8tMX6=lKS!PEakZCQ
zGM`gAEk7y;f#VbfwI;&^rgB>pD#eYY`>U)61EM}$z#OBLFR|wh(liHZ*vHX_2a$Pt
zWZH4<{NZ^5%~P$kxIOUJyNGbTlFzRQla7Y_@RBm2BvlF2&2s9@j?e;7!(U@IanWI?
z9-E#pi5wvIXqI&7-;PNl?f-JS;<2Wc&|N;oS~By&sN4(rm?PBCv1U3)cj>~VHv3$Y
z8{UVV-V0t^NJCvJFpc@AkojMQ(f?V0L#+dfAZVD0lc2f-iXde7%82pz(w}(HyGp-P
zP8{29La$A@S{VFg&+!_#^m#(RZ$H9X?kVGmWbNDoY(vbS4B2yH^IThA<7u?i#3hh|
zJjc<KGxK&%uI=g?x<_GVyD~nchm!uYAtS~a43OTQ3B#VJsgj=~r5aaAj>c+~z{EnW
z91Zt_#mXGZ%T$k)Q$YH0y#sZxoI13lg|u8CE~XvRCu#(&=bOQk=tYI5FpJ&bnCWKt
z!Qxw3t3qR!aE1ev?@_rQax0dRt-;tc4)*}KBb_^&^nmX)xiF+)u3YAbm9FEAP^g)3
zLwz#;kwXT@<$s2UPty)ktw^wgKWNTTCcQo_bRh~hFW>$~R|vh=UaquCP2^jvji~}R
zj79njC;N)5UWNG<i%%z35x4sSB}85lm&iw<-w(~<;n}0+??aX)Bzq>l*jw4?R-C_A
z7kz~DN{9V|l1}s3o_irHHQ9t}3-OcHIl5qcMqyx6X{umpRH!IbMrt|PHla}WnqFti
zo%m~+X}NYePb*j1T2@}F@>6ocYV?Jx$xt226fIVojEhiK$TPD)vvK_OD-GU0jI48D
zr1qq;<pa@cKNOjYi>TTY)Kb`Bg10A{swoF$N-idO+b)TsJ%nRaD&iz>v{l&}4(}7W
z8q9cQTQXGgjOqijPXI+{!>wN#osL8h3AF3q<wkrqo0KFo;030G-Z6zgG=*vjcfS$!
z{-DO5KQF3oz|~c>n?y4sLI9bSh<KTC96l=0i}J{#rdcEn7=#`GnQYFG<S6t~q28r-
zhMo2^zHy=3Sa33BeUf`4?l?7UBb9odGy>Jx(Fmd`z)6SKwYrq#cCMNa4P`P~U9uhS
zuGqTXqS&H#b)%5#PGSqOWFlkJh_%oi_2sr^g8WBiAD3rO%KVlD->fdaTe^3bN#v7o
zZFZ<JV>IHW4_h257Ny~MbeK8_IsxMQS4SU|jp1yDmpV1KimG{Rd5k+v5sS^BA^ilX
z`<|#__2S{<qo`&~uxd0Zi6`m{eox?vt}d}>dGbS3{=x$vYZf2#aAW){UGSl7Mz}yi
zaV6G3Rsgss1ci}*Uh7_8knl3=D=bYm8!MH|*a|_|M9WCCo%`eQEA>D8SL2Kvm9J71
z#?dX9U_xR5ZGYNvWs!-nA#clz1?yp6oGxw6zUSVEoQk%TU+r#GKO3pu3%jFxcO0dg
zN&iMyrQ|H)DH;uQ(OwsCnHEo6)Jly-&QOP_g?*y-Mn$2(u|pG+T1CB2{GS@NNg@l=
zh?@-->I)=<e<>?;;~5#+z|AXcBI)Y=us}91x$?Do8;=pIL)P~*=ib2D9=waFlv@ut
z?R>E7&~7S?8H`9E^pvX9TvFK!?ndzJ$#sRUjh$%pVAeQa5jiU3@ZIUBVedDqiozZx
z9*W`mSsj1i9_=WiTcH-&+LC%C?sj=HlZ4kp7yLn)`S0e{=pn`x*tG$pN}`28Q}LYs
zyxZ|n?W~c4-SAkpq)}HRU4FiIOKz~IO?<3kz7x1&TqE=(^$aPe)xyfW*QYTDnb=e`
z&vj=MrRSZ!&nXz@Abs)Y{!8N!hhb{oxP^058W;8moZuQokWD-0kd)H-@!50>&Nk9Q
zl5}_oe#y2ymY;ZCK=BF)2C6PpKqOHp>OR$vlof4iGNh50ZQ@Xj>PuN(okq`G6Yem#
zNVf#;YJE55^k;$}<}EJE$Gqj5S;1MSIFmH$V4<pf2u7p~(XcBR;fRoBW1BPLcrQW!
z(%p8rPq17^(n>sd=*6)QvB#N>f4Jy_MFLU}J?wgQ1ws{_Zrp&AAq~j2yl$`jg(dsU
z5v_A-jz#v<BX>gEr_4H{ZyH>z6C6MUPExzFY0eZxlg=sH(E(+hTp4UU5%-xIbt@9D
z)e~PFoS+pw>e8<8W%B<RbHgeFarQT>G=LT{1IOU+{L(+3FIjQ#{e!y3kF%XW_LzmM
zUHg%_c16ba4F{i`=*l-+eOSR`_S4B)2RAGc46{jHoE^_mpSZtc+pPI*f2<$+3;bx}
zer?UqS7CO|x<WVoiu9XpWv7xtgwEL9X!c+@&3uD7KKhz}hmHE;AFF(IzgBbPpPODa
zt9Sk!PJufL36mNx7rnDkUsimryY<w4{y!!);t%Y3f7G0OHdoze+Xkch>QkM;_I<bR
zFX|fdwXEd_{Uhhu|5*LqvOhEbhu5F*g+G$t&i~N!clPh}z_!chZ!S0cJ1?2{cKN!(
z%LV5H<CDXAANe=h^VX!Ujd?An@T2l!Y<c+3+s|J~P3z1pwAme^^lsvX&55BQDIXYQ
zS3S-)`8DfG_tdo)p4&fMX2%cQbQAW${hvr#{hy-#ADR8Y?Di#ek07vNDGOXyC$pCS
zN1A-|{4bt+c!B+<yFekS#s4JRfz#<i`ONhX&C~xg99va8*WTDJYMxvDACu$2f!tqO
z_c(y#i+#Y3ULSDHC9mO!`hQ*9o9sU<29EhWp1Ke?8+rIc@xM;<#{C~=gM>or4;%cE
z0S@B$KFI&0_6B%U(N(!~^Jf%Y`D6?%{i)-~72it!9G<p(BG3nD4g%hL;!`6Jw0-S^
z{y!QZ+n;7&I!OM*8GCm77n3Wbfoqs{+y66l-@N>p%vPHopn8l@10DkkT0ZtyG@k{y
zS05Ng$F-f&Lv01HBuxBa30%WC`@sDlzQB6!<0{*^_6AQ^J^^aLaFYCo%ku2NGnZFa
zC<E7@THF7#jKXx%A7lQ5=YOf(1NMv$<pG1gH}K8NpY6K-%Jvuz(g0cyyq*WRsyX_g
z{EtA8^{0ScV+<$RKb%p=SAS9MgZV#UU>eElPxn7#nN=wXJk<i-LGGB*AJ&fPB+Tfq
zwyH}!9ee`lU<@Z=q|q%H4#G^JH!pwY8%>UgOgx$#p{3Dia>P-DkCu+8#r0_EI9gdF
zH*rQQYj|62w6aEO%#OAlN8660U769+akO+C?ODUR`lD^f(YE7g+YvbSG&;}&92_0(
zS&#OtM~A^-b>pBL2G<78cQobIpPcr``ai>t{(p1wR$oRRVctET7j)EJUU%-|AN8Ia
xFKp?~z2`Q6*=^N5f_X`Ax0x`8>zp~xJjwHY$j)T3#a9YjrtDn$H>dvpO#snn=_mjI

literal 0
HcmV?d00001

diff --git a/docs/images/vista2d.png b/docs/images/vista2d.png
new file mode 100644
index 0000000000000000000000000000000000000000..5d09c1a27578a6b72b1a2700a5fb29999c99157d
GIT binary patch
literal 455243
zcmce-Ra6^a*f)w>up+^sw8ewFQ=}9rP~6?!-L(`e#l6r{T!Xt5f`<adodU&OLXea9
z;#=p>`<?4EE4i7K%$k|K_w)R0$EvBw<Ka-^AR!^)DZH0eM?%6FMnXbm!@@wE0gweE
zP9ULqNGoV!A&x*S%TGv1FOd{v-)j2i9<JE=8v6L%oSkF^{eI88=%ducAbifa<R)Mf
z^^2a3-VT+)4qG8Kdb_ZY-OrZg$0R@>C$+c%6ei}hFiM1?@Cn<LD0f^h_t$yXP!mXL
zwkv@zds~dtIZ0js{|zC85f)T^AUiuI{B$8H?|0LF4EHvi^Lp{$!;kbV_aEM;{~jI7
zBwu6yb7H6>QJUx3fBt-Y&-m&;M~w*1u<U=1zgqJ2S^v4hr;dH*H^l$BM6wP|l;wXe
z@anw$`xoW^TtU{@>;I3NSc_c^{m*@Ngm1f_R+#j51x+sFUR_zTIgn@JCl5=xaDp5$
z3)UvQFk>=t1}SK7eov#aX@~C0X51exWBaV(6%vw=zI}tT;kseNXI}s@CyYyWM~zSF
zzZFteS>)!of{ZC|3|SrySs(GYa`<*OSqkAw6a=L!_D&Ll^wD;6Q4=qwB#T6ahQfj+
z%S^e)$DtnTj21cmQNG(^{~d~s4#S(~kb?2PZ2r^L8)*kN!$9uVO`#P%PN;KqVljtD
z4m)`tIyGjA(7Va4!m$%|67(dGf9=F}2imvukZ+Z^`}^POiAF|`ZqD6HOGEeVOUuHf
zLQ|<sO_A`aLnAN?yXi16&3xy>!FAk#iLVNP9(9)YB}ZP-{jRHiTs|Bm5F|{-=xMS0
z+J5Dc(S?d0Vf;d~riz@23B}UH-^-U`OhH>zu1Aec!#YUM2G~7NS*5S4S^sW&85aO_
ztu4(OUwUY0Y~|yl-pZCj;#|Md+nB``6vtDQuzSLfsWPuD9V&+ueT0L^(Yn{BBp|(8
zy1iOlw(?y(pfpy^q%f^WKQV7TRx3%UFy`a6iy<qU&5;>s7`-CU5+ymK>oX`VoT=aq
z?-sqDrotR!j9i>yc|}sW?o!JO+8~nZ)*Xh`p4jJ8E|IRAZf88gYBbYUZ!o=@`KY9Y
zPhdg>;<d40lf_XTnVp^TntF7pu3w)WNcvVQO)F_&YgV6|hy~4(7P_H1>}D7!f!W<n
zm>MPR{dWnpv&7~v78O`kj^hj){f@#Dx@o?cPE)cQutOo_8u~jij^g0@{8tJR{z+g-
z+?wP;N64eizkLyv93k#!Gk@spETuq~Xy$jXCM2uUuRIqF4IBCVeZs@&wvX)e9Wkzn
z2l4ld9StpyBFtpu*RNgs@n!|xVD3WOB+M*zGi)XbKwKb7y&-<Ro>*|jn+v!wb?P?T
z@RXfsy<U6NnoGv`SdW&Do29R>*wLbu$EKp9E3`Y*XvyO8QdencrQJzY{@+n!LU;JH
z6`Y)5?7|s%8sWNV_hX{-i_=1DZUAj+2F;dX5~g!I-qUu5hB0NtIs<MSU}1&W8s<(0
z9Zk;I-mRc1$M8VjEs<<|x%XA+pZG~aDtlRfL00{BoX5^Q4AyI&lA`UDKD-XlXUKwL
zhV>-NO9+iNTYxyC8I&9w7SeQDZWDFfx7tp(=u*lG`g^A|O7k*%yo?wf^yeG9KOHSO
z1?%&`#OAthI#08p)FtD2<j-o;-{Kp=5SgDA469VDich&3W$XmBqAyW%;G|eRCim$1
zB?Zw_Xt5qBxbGJA+GinqHZ74d)ZMk93^0=I0*|eWO^?QKVjg};FwDUhDiO2FwmNak
zQu6Wl4r&qr4=>$$+uao^uI(luVEsU%q<9J>-MaC^$212M)XZ^=LSMF@RdE1-x*r$a
zc=^#KAGvYyH!t92>~UMObG&~h2Uk~LihB1wK5+3_!u|h#oNp)4-+}6Hg*?r51)@JM
zx}2TwBaa={KK$`hDkXSI^pJ$>-A?#`Sa4#@;ND_&<QLHRb(ieXhnoQ1#*yNpdm(hI
zYQW7ZAR>YgYh8dr_eu84FN(R}V$e~wX?I`)?9?Bby0nVj5N;EK;j@e^>a+M57w7gl
z?hGD;hBEOY&4M<mQIDFgx!`G6oP&mj^9*bW{(HK9Mr;2h#AS*+H(ND1V;MwK@^dIv
z+s&0()<^L2Qp%(t$}%P88b1H;<AI8*0he#-)Nz{Bv)<5gDRGKi&PIuiX42Ez*r*-b
z=s-3zqpPN^8Cg>L>CWn{^`iZJ=Mv<SMD{z!_BO<<o0vGAghoiJ!z#RK7U@7bM8$Ua
z<p+Rtw|H**U!vP%AG0Na_itPkm9SG~&)}Ugs5~`u=f#e|ck)gR3oIqnF}w5iQK&N?
zJrt>-VW_8jss>rQAZzUT84^n#%X<9RuV>Bcn#u0qPP7a2nR)3a5`ObB5@?dQVb7Xi
z@>+Ro)+9juX56J|=HK1HMh~)sfvzT69r<I?tDQ#pKep;3hnkqIR2d_h^`1v(TNj_Q
za)0dWg9|;;-VL++hfa`4apduX#POQi@RAN~9yv{MHGXvq%Hv*vqqg1$yvMKV;crhi
zfN``*+e%sS+nT-W+~K8?nmT05ZkT(;H8z!Z)SCqlp!>VIYmRBV3eSqlNX^g6vu2(%
zeRW-3((eVh&CqQpeiyD6$u;dstuw4%{;z_k0;(b7Xvgh>eei__fErm?!hESW-2!9L
ztUN5@(|~n01Y&1D8Znl;>4M#p`oqyrn{tvAbMpi6C9b&{)AQwy`N=*S5;M@SFcP!S
z2YSLGtHyw@ff7ow_?K$QrRy`Kq>hV+wGPewxqDT(vXp+NMw#FVPdjG``-HDV2H#6C
zvvba7e_&xq-v-Jz8K40Q-CMU#X>U$q+K-FOrh&QgU4SG>^H{^KaNNVZ#H193U2@9s
zAo$_G129grCuMa9(V~7*U9eId>of3StZvL%X^gj<XXEs6{#Nf5GCw~WWzMO8963!C
zjp(=bb}o4b2)@vXp=}<`A)@;xW62fVVcp*@0Z$t^2aIaTiNO9+iAnv$s=A_mf$wM6
zlDcbh&_BE&Dweb4spXs+Srx~BcsMOE@oYIB=TJ%?8;NISHSI;I*rB9}+GDw|Grtes
zbyZ836r&a5_&6Po?m9@}OXrIu;InvN=Q!&uB;ouvYYuR9G&VL|=fml^U~WK5QCDY?
zg?qSwT5z8#6C}PIIIU!K?0v5G3}!LKNFh15ARqrvj|PAnD5=hrRk?XBR>*x0Yxhew
z%-Fu}A4U0ejcyP>nUfly2?-^l7Qk>1SDonRXRB@n9eYf4Y?#ncfV5UTUzGK+|IDPa
zFRl?VFrb-pm)qz3qqzjT6!dpF=PdAZA^8OU*;qMow{yxPO)XaUmgd1OBgrQsOM>(s
zM7{r^<_N@(xG7e#<)K)5&ukesf7=pt+;$C{&y4bkh?t!0kvfvAG@`QV$AGh@)QRDl
zZoLa|vZ)EDS~aiPYo4%#zb<};_EczTlh8%Cwr1NVx5CO<bsIl#n9X-AB=)`q3{Ba!
zJGv7&2j}#Uq`58>b7UBzJnD|=3$1#Q1wLM+S6|&2oC0x3N7}?_bDGms4EXgVezS5y
zH1rh&4Ep|+Mp?+nXgkhMqdwi$;5RRC);lf~60&4~f9y@x)@4zQ7JpT+E%hmj{jhmK
z`g_G_sKJ=XplROY`m!e}*fNgJVSS<SCD%<w%)GAfNCZJTkAU@(_lk6Wy0X630vOu+
z^ne7?L!qG|#P98Z)#UalKdWuIVmsqt*&Q26@Eb?_k?;}^=yq~doNoK<B4}bv7noXb
z#4l_YHj^)o0G+~)i?|_oOdJ<GTKyjc^`x#Bp!)x_hUxWQnjT?31jB>;7l9YBZ!P5I
z6>d|!mOM?@y%FBHczV%-Tu5|JHZcStLsn<zAX`yQj&y(=1%=l?aU@bwjs!do*%F1Z
zOP)4&iH$ESI|Q=}*v+drPzi@UZ5}&LhJqeHvFnq24<PA=UROMfMh;}+obf?rk!>Qz
z3g*)mAb1Z65qd+UrjG+4>znJcrr2Z)ySZL+>3)!Pk^T>2N2_(+|1EF!;IHH`W#<sq
z;zUJf=)(>5O#7lpx`?{N_N`t|)XRTSCFGF%_DUos;MfgIPJFD}TR5L)$rSGG5SKZ=
z_-e0#1~$Vk2^rmukVR>(4tpRTa9qYRXx7hx^uBfYC?st6b|HNnKS}(!w$?fIbXmUx
zK=vgu<3hQlfLGhPOmXbXa(#x0m-cxl{ME8&ij325a);x5>qYF3)ZmU|xr6<11Lv?9
zx+c8s(ftnNrI44Vy+(ZY?2HcfrhDow_dK}_aSA^llvNF)(<S-I8#=hCn<V8FVz(WM
ziJQi9RxbmNuipGn{4W_2hW$;+aV6K+Uq5+FtZ1YB_waT6<*1DwxE*I_30)*H%1Twt
z>}j>-cQ1U;o#TEiged#g!*?ASpUH_fDN50}#wfLBaKyT>2*n4s7ay<N)MOE(Rrf(z
zmo}}Y<?c$v*ZH;MamS}WWUVusT)bPZ!%@8P&ih+sGy|1uvtxrQ1S>1Z<;R)kqdSCD
zKGEb)1P(9sjL<Voj@|dmI5%m6Hi=IsW#lDdpf;oy!QGwRYC}_1Q|hG^5!c?fwl4{<
zd0g_N0_Eg@AQe@zMJer68819TkA%26zyhvp5g~4uruZvWvFA|!{0v@oSJp*4AxfaH
zFVxt#abE=t#>3SuO&Qmh_<KS6x#ZOgxDTVn3MCWcdX=EaSdPv(fx}X%Td;GizQhZ~
z5~pO%5tRBfw~Db5>*0JiiKUyydS@C>pCx~oKoabWqj{h7PgG1Xv5Q<Q^)RHs6FBl$
zVz2nb0DLG<PF7gby~C~R9|J~IDnb2$*^xAS*LXd1Xh|715%v;zYhE<R2p<?bf8U6B
z%IfIBcbS3&%|%2V#)?W+?N(g2T5d3#veHm07tgX9FhkPm$nel}xUFD0$KiV{4De%R
zGkXlQb&h@?Nkpb^;3T2TY`LJ;^<n)U2IONPsy)}lib5uvqImzo;FzBW2bFhD4ZQQT
zFaR@7FtsxfaClNFVaKDQvw@I>3^or=SRBm0u7FtUR0$^r<#S4pCLu310>o$uT9=bY
z+p9@~&?B1Ga$KS{bb72ZIzNTtuGc&*s#l7hG*H0<(<cYaV{E=Bzt=_AlZimL+TL!7
zRd@GwrDmI|D68wK4hqgb`%TRrhk6@yoj*Ei@>z0!SmAjgqOR9#IgtPiD=K~~A7LG4
zWqnEhCbe<$XWOE5-mpT8=P|H6eeSobB5uK2an&CVuf&gMd@JHTcTN>U4={t%z-^_d
zUu;k}94A1;+tRAy{agW+qX?&^N_`MrYHPKXx}2PMs8;N@e6AHZD%l>kB$Q}G94lpz
z6J>sN<$6U0n)CnNVeAWEV#k6PJk|Z>0)o$LqcKF8C<mMjq>^fq%LiX}g<>+1Fn)Do
z=JYA6K!2K5sjmWj5}J-DC|+n49ZF}PoCBN$ti|CTKYn^nMN;}@2znus&e)m~con6g
zUy0l$B#mp(m6Ng19)u^DgTFR6iYfdnQqo%^+sf$nWM=b<g(Mly(omY0a`vxu$Df(3
z;nnVVoV;?KBdD}_t;@A@Mq=$J_saW3tAe@Jl!ENCceOKp1rHy%_{P7}^%2HZ!#(af
z3tj()?lV0dm26{Sp*@hUs3#@h!vQkEK=3&wjg*yZ0-}UIgH;qZ?VfCYF5%%WQk{*w
zIb|oS?BS{v+>EjBEsUQ_ux@vges|ybn>#f&%&GS7jJIrXZ6)^q4C1~%ZiG23HR-u6
zh6KN;?moW3JyUUi&Sd^ht`WR`t!7|=H8e8v64-x*o@nbvPENHx)ng=!!r#&gqD`_g
ze*GFeJ8J3Le(i2r9todKbKh;g2P&rxnoUqEbp%QP22qeJezCIF2TbH(mgI&rY0Jvt
z134PLIY4RK5i1d;&cMZMuax!&w94GE&&(Qwha$s}H6q7jq1o({4V-npQ+yP?{dm1T
zk%aV$Ub-8nuck9<guk2a^JnsEJEmoIRpwun?n1lGP!tbp-G-q>4T*CsnPh+SUh+4F
zuS$T#?j62Cx(Z9;H)^zOy$b9Ro8-<~{HVSD0+l!QR&h~b;^Now8XC1uIT6rZ_vwTv
zpC`q_=is43c~H7uwoe~4JFEXM<-OF$7kUurSiF{9E3e(+7Uri92{P3<Oj&PeN7eQB
z83=_MR;!IHJ@8i+^n)K|GuuP#g{N&K4>k`(kPhYylm^tW`ZlDc(%;X2sTTIAT(J;(
z&u3w<f6|=(+0=;y0@-L<qAo8lhmCc&&dj!?rdv`_yEAi6tOwb$XEmDajPxp~Pjj3(
z(yhY5wE;e6u6kVq-yx`SawPl{1MS~PU(wq-TDBEN^``Td>KVS9z1qBygu*?jmrfWm
zTooshAOxOX+z$@{6d_yVnu*RaE<pfkdNI}Nm>H5v!(mxJVpYjqs}+$7+JghIp5^e5
z2L(g2<E|yMQQ+;gq?_=Ar|aY7esfCXGqNIlhEE%r6ifo-j7A51tDcaq-!y0b%$;|^
zE8z_8G7l1Mb-FP4R*e30-_$a5v5BOlfR>)nHB+Uv#lr47X7}P&LxR>1O~ARq%A%Jq
z5Wb}S7Ko$Q-go~Z$sh-kO$o@|bfXRlHLd-UhiF`K3p1%&U&XI8y=*#c8)aT0yR<ON
zG4E{sG?11xxe*Dh380aa`uHfU!E$?9!k*y&H@WX(*PTl|Ap`>S_3zYbit3NFw=Xql
zOCA2>@~wFfcZVbK>sd^JBnW^pyc<bEx>3Qw6nNvaUKqfRu2fRB>}-jqsI-?<v%V|Y
zTtzKr3$^*<+Qd*mkncC>-f1~yW*1}ixd!d|+W8eHyH0Zw%Af=)#lc{2jD-vXNj=uT
zX?<ij{_AdLK(eZX2EUOppo&i-9*(aI^vnGHG3EhIGnj{dxMIDb;?q2i0*80|oWTtd
z;>ZjeVXZ3&2<ED;91CwPXI+v3VFc5KNo^d|aaUN6(zWKITeo{J-0+4mhrqluY%l%t
zdC4OQ0(+)PkB|Fys~v`C1iO&eU@B%C>;|U0Dt$@SCEf#BEo>QFfg{3^8BIhLSTwEI
zr(BM!jz?6y6t}a@r>d%3A^R+Vfx$u08`Oh}OvSUmy%)D@CxhhF^)>FoqUQ$g)Jyx5
zUplkzeg5x<hiqIgCEYuk_AH`C_oa&B*JTEj=9!BddQ;p(pLC)9dv>`qe92#_zxz^9
zH_Uzt3Q7lj_|TiP`p<nf3kmr)=_%B}CCcjn2U*Xqtk{s0RDMi)G{%yF;ra|Q_WH%s
z=`vwXy<1^Et0*V!e;A1Tkf8>emc>rsR*u{diH7DB=Uhn5G3|Dr{3EtHOYUQE-`bPN
zGI3FdAFT$BjkH5vv2L{k)e;yp-?vX9C#%lX(R#3a&f2;X7wUCnEJ24$fKX3l`e%*Z
zuIJ>HiVuYgPT-R7o{#WfO=J9dP+fif_pbIkd*53#KnQ-C;P&Yk6@w@p9y>{}9^2|U
zrnQlQCR&A#0W9M{A4OZ+n$)tfadOgBW@u#84fQRG9(>Wwk$tVulOd-ObzS@<tUE9U
z4sU+7+)9j~&Fwx;>|N*=ddu2xPlmGM6Z>}L<;$K1o@)ON*AUGw$$8G<2RK9O4m^Yd
z(gQ!I)hLe=08y3BQIG>3e=@c_`m2b7vn#Iqz9;pqf3tFe?T+Yt-R8uL3bvRFnRyOM
z=n7otQRpDZHDq(UuT#9<7voeB+E|1B#SX07NP4s!@?`9n%gmX$p<iD1blw21lmGnb
zy$9A`?C|c?^tc5^8W9u{YU;>%1!n)WXJ*0PxX}BOs;u9;)V{VY@swM;#+XvKVO;=I
zF4P=7f$S})S~<i^sIo-7=F-y};naRW%~`DZ+3}_yc7}JF;Ib}6NINfIT}iYvF+hyB
zKuJtIt9?GqbVLhAt#1?g_b(K7!NPIncDDyiJX*GMBE9(I4#VXjto>d;mF_3;0If3n
zpvpUEye$2w^&?>RY{g}@f2;l7w;!?79G!^{&@zN#rp3ws40E1CAa0<#GdMj-A6|4;
zc73Rg|KcL*Pj#RrI!bzscpWVzRqvS;Bh9P5{DJ#h71jzRSHu8PQDx;$Ph5xLEK`r=
z9T4G;?Q;r=NxL2ZwLwM-LWIAz&bS3IjEuT1YO-)Wy~809RH|H?H7}EU)Jy5^k&fG;
zUz=RX@u_Qt?X0YX`1yranTLNQP*Kr(&GWZ?Qp2I8JbCGLhd1-5cl@`@#-6USbFZ|b
zqV<Z;fp*KWh*H=3Y1>$qh0G!mcDPy0oxiP9qsFTj<TQ;pnZxr3RaH@6&4ms?WCc*+
zo!8frr`Qswjd*W%O(6Sm2Bk>kCeRGYYxG%~Hu2X~ExVHKLjNAfoAXIS>O#&ka-S7<
z*w&U1kd>57=HrW2)6Bnxtfgfv6{p$R_1{2wO#-ndt^M?b%p@*&czJ^;AuMzvAp}(k
z8b2Q79O>S#1KM)yODDC@YRg$&slyjofWPFiI9SZAe}2l(mp16L@ZV>-os6k8cTyo%
zSukXt&^z~fdH|XWCo*Ve#b5DPmWB>1@Vv-Kei5^3Jl_nhZ@BlZ(;VTc31SZ!T!OmU
zaSHEuU0t1q5%iQuScMtBGCL69)78`E`@H7jz{fZ5FHzuRVcxfD@CHS2<!`<33%)2g
z)-7w;_mSZ#s@n|$fiLl#e-{O->O*nopLz)N1ksWQ=G>OXD@w`rAQ?|bRFzM^Dce-K
z(-}|bE<<|gp0n@!vol}NN?CREl)Abz;zsf5hAlVN5bB|PLgHd~iUJ^#hr%}<25R%f
ze`-D{AU6yaj%DQZ*tt)itf(h4%y6~LPM6L0%~OJ?pFTGA*sMwDcGJFkky}#0NQ<j)
z%g=8`HD-xZxb`dZp}m885R!E!WkQFLTSy)H`?XmzR1DObd}^#*iMQ~5mN#hq)@(^Q
zItUT+S$^=MG<efIAX*ls&n!?P#+TaO{xC08YxRTjJdlcNBfG-MArPLywTss+oLgCu
zGT|l(s0ks{Y20P|bq#(z<2Y?Pji%!KUW`)O5`Vy<I~v_Cx*&k_p_udB<&j9^nPwDX
z__JgK(HFm!v{E#8gvZ*^ZEaD%NJa4b%$npE<kSro?o&L>-E@kI>y=>Bu9~=I+3cAP
z4Q4htjh30K(Byec|BU$&R+63dAnQQaCBxFtugIu?L<^68!?~}eh7%G6nV|@)2paxY
z=)v<H&3w!6#4u#QEl*aT6^tP|nqURooRO7FFfxjTMjbFT*EzAsmY$BNc#DShzRS`)
z-7;=<QI@7=Yk%K>4r=ED7OXf51459J#2n|SM<?ID`{bcNU3n5C;@3u%364)J?}May
zHQw$Og`ho`#?Dc!z_lTg2w#f9Fd<<q@-xK|`(L$TOL`Ydn#TQ#LV9>*^+fr1O3L-@
z3Zsb=@?LmV!B@W}6ZMZY43rPzH;A!})yhBUgRf$)!4jkV7aTV(7S3Z01svo>oE#Do
zrJdNRGK>&FUII}LYR-G-9bp{_hWNp^GUQp|f`$X&GHW2a_i3&-Wu>9rO?hR**M@!j
zPXniS;v{4SIRgl@jTKs7GTOiYXL1T<wX=Q0!6fR~S?H)?tpnx=ZsTGU6`V|q`g|QV
z^N?S2>HELNkXoJP_VI@OOuuRDYjrs*<8)X*vR+r`=4}hj_wU}yD$-u0%V(q+aU9v5
zw$u{0&PXWTCtXd>QEb(<g!a~hiJCZqyPK&Q1`CUSEjQY;DB35s>ZaJc*Mi%s8O(Tj
zy&s{$@8<p@9?h09uc_uj&ek=)WB&}Yp;7Q^auTVPCEEn<@887#QJ-v(?34%#r6eyp
zVBbzCk$nf<Tq2{Wv`U<*G37US{}bf(taN=d8EwE=*Wq5>%VXQ(v|kHSb6qcNV^Lp=
zJ?b*qZGsGz-G7x~!u{V9Gtt6>Os<ulTgAD{*47|$_yFfO@U*qVpIK=uhF077GlZgA
zF@R7EUXJQ~O-NX<K`34=qYfr;q7(M4$dG&a`Q{|YkJx)V+R|wQY9HS|jVP5FYFhP9
zwRz2Qy15h422K)Hn@(Csrll>YLh8!Ffm(~j8Scgg?9lk$boOg%1dsExhqYS5Uh(ft
zL?z<;{-Yn-{*;jL!k{iZIZVTKpN*dO4Kd5lZl(_(Ous1m@eilQ#JUOjPPVbUBW7>#
zq~5O-;zN_X`&bCs8XBr&8>R%-b9oaubl;$NB^%4hJ&AO|H7QwDEVSU27w=|0<`y5+
z+D2}F)cl_;K=v&0vi*zaU4PiobyB+6s2$W0D+@BjiJ_-tNV}k`rX7tjhat*lq9YKY
zfpc`ud~P;YAO2CS#iqj(v|5Z?9ZZ~X#C-~T9~npQTAH^xk{0Ft+2j^Q3=aBbJ~<WM
zZppIrlhW2<%axQ;RUEysu>qR5({>Opc1?b`iWd|VHJhB0;tek=lL;uRjP&WI%J;(%
zaH@CZ$a>S|ip7{y*hcB%#J+uoRb3PLw;?9t0l%|z_Jf-vZgkB09-b>a#t3*R0Wi5Q
zI$|=}_68N=ubwi|hy_D1j$t(_@G*m4*qn)6+|8u0j05Z!X4hD>n<TqSSLrDX`^Ph4
zXKcLJP%IvBqhHZCH6=azHVKj_gN)F-YW=`kdgv&?#IU@r3+@!R%v{hI9Tg2b8iEdO
zi^c&FTELFFwl{TtaR;@;lV4^Qz2VnyB8Hwq2v1DH<+>Jp(+QQXG-@Z$3PB9W>p#6n
zsbeDDoiwz*p|T2zfQel5;O`kuM#xPrF*Ks4YZ4WYl$7;bdnE)3cA`&#w=*Rm&{7lL
z=g)_$F{?FO7aKLzjKqm4)Nc4Oaj#!D9fP5Z1q>PeBeR2p!51rZG6N2)P^OBxdg}*)
zcb}&HXYD6kV#!oK$;IBW{qyZN@PltNK~j01!=3XX>x~1G=BbiK2lhY&-WjQtS|9p`
zTRE_Xz}Gr{oxtncz-v7byhIrhBdp=*XwV@AP2+5&hDMX@8<ewOC#}1N6W4n}F`{@P
z6CZaqbx_x{a^KWCuq_CCP~_wC@ZdaP78Va6-p{{dMp9NYs9OB2hHBLID2rRix?8Zv
zxGYL_8%`2bet4d|&d+!U<Hlo$1v`J4Vt22E-DDXI`P>m5y&fa!DRiZofe@?~M1wAJ
z5DpkZR{V?ucukBwWa(-YAXt{#S#suffYItgSE-LaJ}#v*u^PZ-Yi1xfP^WZOZ#Q)t
z+gt!`Jlds$i{I&-S3|2KQ@(|RRG&##9hcrm^(2G4;BV;b4=2s)40DY=V_twX3yA_6
z+N8V!)dcDi!kJ&@*FhlbP7JV~dUgN5Qgktxg>MVt*KL=J01BF@5Exm@KCAl?K<~!$
zMPnmEW2f0#b_8NzxDwD>atS0&9MyW1h&vn6EMJp#y6)!rYMV@S#o}k@h~GZYCQb|{
zu1jjhX{(20egb_Wgq^9joQ_@CRwtdyelK9kubpf1V!PiZqt`xkU4nO_adwdh-)Z=6
zQ5^|0G9E@zQ`)9K5&xKu;<g?BVHF;7WjyzJ@bMnX-VqOBH?1fac_)o^x!fNk*?|G{
zoU99^7)|${RNr_SzfdU(#eH57&Q9Y@dTK(PSU!eyGiX^@0CV3Kz8XGYr-t=_LWET4
zr2rW0<fE2{YyA9Q6qC>A=j-!nMe%7HmXuZWX;a4~ppOXK|Mx+ZR47Yd<ZF3!sn*=6
zpmv|Fvht#~SF-klkii6w@$uQ9$*tb0;#h3=oMxVSS5na%^m#C35qNzaejxHj^49GY
zrq5Q=J3Y5|U-GN)_$C1Bw_=v3-26v`apF%0V;`Vj-0sngLl_D-%z$t1O7~q$S!;e)
zjw$ik*W8&r4jkXC-n9KfMXu<}$*i4mb)O1`S|^ph$;p?<&D*#!Fx`3l=bRgh4HzOq
zd|BC&j1+*UTQ-qLRI^i!X6!!5$%|2-gMZ&F>@Y_c4?g8uB2zq<1GhGIayIrWoUUgD
zyRiu*Y~L=de=Ft&2awUSkw?6iyn%uKDKN6bSUF>c6&Pt8gj*~y57%5;5u8v@jZp(@
zIPp_AVsZzpfzK<>-Isq_M~`Hv&1rl7Rc<JxFRg0q#3z_GH0jIKkoHN6ITAHhdHu_r
zqw6u4M~(q^LNuZ}pSj&<X{fRH?b3O*QEgX9<OLDj^ub?|kr7+4dGk_BN{?mQtLpwd
znOVmeCdS|r5utqbVKS|%Bq*7TDY?f$Q`WC3N2|of_WZE8wKZ$g-1puMk@Q&41#j>d
zS2GyH9#e*YFqmiIRv*w-zfxiZRI!NDw%ka>Bnome5aQ~;*EV)$XBZttyf<wxil^H&
zz<hUO%EUy;-tvsXZlcQU%F=LKKsR!s(1#-$PfyTR!!`|H8twOy!~cb;Ao6xBxj_Pz
zhS*EZ29y^QKKij*62l|bfIn6-)*&?Xbh&<=+Aq>-KA*!}b*hTyz^M@02zG<Rtw5pl
zQN5<Ln#^pKas7)P<lEDjqg4chN=TR)FL3#zJ}@8<GP(-}EG<@cux9-d6p$_IlXo>9
zEC2JAE5-YzP|Vqjn0#Y{yY;sYFd-guNWM6iP%?V7*&IjinOk_f&s#kTK0d-Datpa2
zHj4-B!Jqdtf<j|IXi$(%`!UFA83k)1VvXsjM4b%UeK2jE?5scXf^=z%?RSCp<@#9~
z8Cy(}S8m_PN+tMqogBx)>+9<J?J{$~J_s-FgncdF-F}4ie0fmV+esmI{8dGDQvn2L
z7^&mM{-TchT-Vn>NP92ZL`3(N8^R_&Ed2|#W;ar;+ODpF#qw3|%qAH@`iKpUb~+yr
z7MWz70pb4}dP~;fD}d@>A(;7RQtiCPknSihg!muPe?F}JIJA8@lo7R^l81`<&!ggY
zsBNy^yDIAA^z^J8cMf}`<(C=`k0EQbRk735G-)5fN)s2EvFPYZhQI6I*!e{4Z-%xj
z8!dz!B~g)cL+MQTLbtAXcr6?Va78Y!7StmmfM4+N7f(7n)wHk$nYvU5P6Z4+{M!hx
z6B0KsZ#2Zb2t(XcOG7Y{a=zOzzjqUgm|jUhP+>z&XqYj9s88R{`03^Jb!9c6llUXX
zG(&FdtAV0P{Hd^4U4b^Cu5vGE>hci!?6?~nN7G2{h@!Rm_T{DDLNmej5jY4RcYX_+
z2goqS>EaB;&(1zXI4bGDOy1pjcCw@$ycadjKfKQDD*eY941)*7ipnAm5B6*VUrm(N
zNT#%r)sCG#x7+++mAc4E8GZb7J$XL*Q40@<f{{8Z{^SK?1qX@_ntqYBi2vy}TEmki
zO$Pa%bmn$@@)XVQK>bxA1HgIt;t`%iXzvhF6HF>1S`}&UKGj!YITh{MT8-}ayYxYV
z7nPUSx1ORSAE`ArNKC?*d3Ge}aU?`Q0AqBq#12|)>N6tozW9(9MjR7U*i5FXLzdqr
zVr%=Yu!}ZVh)<B00#t0%oP>(`tzxwNPjt+N4uOc+wQm<WXV17GTu(1|i(MyaT9kS-
zg8<e-|Dpk&5U-0PAcm`<Vady}wUv&ygWC1`w+!#d&zC6}i%V7hj>m@eY$_46+8vy4
zUZPHI!56L#OB#(24}Vl}Dp2-N6&AH7-YT7TPthqp@0kc;pH1iPpZw7cqOB)T0Z@Ng
zBS0g<6wxjud_z?0ND(%dKmLc7O)=nFZ!|=`O!~K8*L(JyMkHH%9En>3{DLPRp_^qi
z9m7Ws{hC|AW^*U%ZZ)00`ro#w>%_27t>kh1VIGR19aowR$60y=Y54gQ=c4mm*dSO!
z-}9)G%dMs=&xO;E*;SN!A|c7(OWMh_kTCYeZc4%c1@+bRv;}lw#i?+1=#x>l$%NPc
zNb1$y&fF)r7jFcO1;@{~djjx}0(|H1nA0=*|1oD}QAb2*%cYxmIF99;jYCauoOJ{D
zivQkW4(0ia1Sq8wW#WH?UNjg`uR*Ur<cY1=_3G+C57jEuwAje%YlG+kWPC3OXiA=H
zjNd%NS?=bYM^3H>RlxA=(e9o;?j(qGW8hLh!b^Dc@|3cpaI55E&ub!Np}YJs4W!+c
zsSIRnbjZ*$KUzdTm=y-;s*Ra+9;GzMy+){Fx|x4(mL8<(b5wzG3q`OmW8+~O{~m>S
zElhSJ>KDxPKE5z?3kG&cXpnzN@8(KlHO#C5<>WR%JB^vy8Fbdh<}>@lP@|+Oy4QFb
z7mgS%Y7Pz7+->b*PZ#G<o=z%G*CV_)d!*<i4O_m<EQ3V7bziKkb~erJ*)6T!!_<f{
z(?Mp@gc=^Ar5yMh7yaD4_4)0$?Mq!sI>1x;(wz1$S~V{0k>-zE50aB7|BG_z^78{D
zNy*5<e;8J)y#?Y*zd>POP&0Ch7@(wbi~EfA(`ja5@{5%L*e0cTI_iZrF>%yO<~1TW
z*GbmN>yw(T-0gy`pgF&@z=v+P=ot1dak8LoR*oh@2kRZzS7~%<=_+LI)%V=uMp!!&
z!kF3*M_H30+?9G7%uW~nN11uGh{+TvQCdbnmu~Hgc70g4UwwQmt!4N~VYB-=FK69l
z?d0_7*i+AY$?0198yBXNE!N1$d`?Ryf&+f4GH|N8zX(m>g0ExJ9N#ZGVEeyJB^Wf!
zu=r`IS7{H@?CqP@`Tcu!q%R6!$11KcT_@AgycP^UV)7^cPXPIx)qBMYsMy%3C)yJ@
zpwYEr=EnVNOMtGTdV+O0J7;)SfQyG15sc+|Su@8|HKJHr6?Nx7IzxnQFnH3e^4T;@
zjnQJx%cnnj0*Qd2sm}BAE$h$rqc;Um4%`^0jX*6uH*OZDUqhcfT4Xl=m2Q8uu!th$
z$ZGg<I3+l^-dGdWe$qtjAq>bpe|UMp-~wLaR8f%@8dDCyOmb1JwN6NI8P!St@Ify<
zy(V*%Zm(h6G@}DwzedmTj~k4Hy<q=?@~NeZxJi&Z#nx7oGEaAv^5F5+g^3WKfV2ni
z>AZ!2Xm#hx>ZoHcd2A8V>N1vSfnVxXQuWsE&FVl<cW^p;z#T<axRgS;?-0EHf%8iG
zIi$)cIOyXH!9EG%Y%Q}}7JQ(HAnT0%yEz@}7_=x|B-OJMmX@kJm8RNFNrkW#)ni#P
zASj4r%KY}(yd5Gn@|D56Qj8z6EX<Ky>d_MLeT$kReY;}(>ngd93D1Wkuw*0G@D|J1
zo49xEFWFdn6Chdc{kV9-xVXB^{Ct-53;#;1(hA6S+7NXsmH*w59|F<7rZc3WNiSAU
zug1ee#I<q%aBi%H%nowSiQwV`amLS`fEVDoL<#RhL`X|j{kHqb$w2UlgN<z?!R437
zbb=?rJ93J;i=<$ty@n_^oi18^+9$^Qqc#1}!HSxuM(Ybwjk)^BY3<JRQI76N4}@FW
z#)tMAK7W3wA$D~ubAL|@829$R;muD}*Pz>#oy#g)Df(}YyyFhO5r(<WTM#y^aD878
z67h%Z_p;|dcs0{{_v5Hy_xXnAD`YDxZ98pa6^@T1zm&fqRI{If5~CrDi-r>-fIwJL
zam2%qAn}HAJ1Zlj)RO%0Jki1$C~MC-xKo{F2!`T1t5@5KilF7+Sd)c#VA)w=hKbZs
zVh({FuFA^uWz*9@5k!?LfK~Q;xkO(Jc13Ioc2cN1dC}XtuOQq_&g~Ivr?XY#pIsh;
z?bmtzzhnnqfFb9SNdsORu}tgP1P$(MjdOi--OeYq^+Z5WD1sZ7;k#WgG(FLXg4RbR
zjRhQF>*ZS7$%6#_vpMxY{x+|X|4&15VyIZsuD??aL!cx9C)56)Dk233mi;xy<_|J(
zg<O9&HdR-%tY-qYAIy^H^mkHU$dlk$j{I@WV|IQzAd^?LuH{<)e<H0_*cesgW?k8p
zmAI^|@8uNtj<4%IO11^bjL^2NIJ~^%S89q-o@Pj@;03R}{iy~9u!}wdYW)2V7Xw?i
zuTV+~@NELoC2oXo7oZ@y>`dlMwcPw@$3zg2H6h7Ec66Yw4w2_lSpw1qlZ%MbAl^QQ
z_?mps!vddhz)O<)mt1*G<H{fOut>3V<AxcOK6T%$7(VKvr>upo8ZW7Ig*$q4>l+8h
zFMTFGB{<26thrni1+jqz(yFS|^b}tHE$&(k9=Ar2<A6j2sTnc8pcJ~AkpXmX;GGVu
z0%`xry=Rgq&&suQcmJAo7}s^*qW(<4<t#v;YWQ(PIuKzP2{d+w?d#JuHi9Rz1ZyUv
zCwC*>|KQ4~Y#?c_%-zt8tTLdX^~YL$;K*Qz1010wtqV=gf0}bE4cOPoZN8Td>?-Pq
zr9y2a3r|zFM>q0LBCiero#YptBq<6*mhUaXrcP@~a}iJxx^U5VK43Qd!w3-&Op>51
zEW7u(m${<hD<HD0tfVNZPMnt~Xym<w9SY~<Y=i@_1hnfNF|H58cJskFEyD^++9F2(
zX0U)?&{4D<Z*D?%-L)T#ZwEL*rXbE{?X<>5!nh3rM5s()69K@Dn;{klxOly5_l={U
zNx`cYT({xU^`0SQc{w6S^HT~sA&-=j$&1=5?YBH4>E5-cZ^%`DJ{O84oSaUt5jj&~
z+~&NuMU|0@Z5C*TS@8^lz2BlN!R5pCbwWW#wBE;&6Dp%Ki4<S9X>dkzKjf5>FN_F}
zlC59Jxl!XoSw<~E1@6>`U6N0&45o!&7)(%6wG;S7aQ67ha)W|SziIhi-H`D#1cvgr
zM-)LU)(G%=kEPgFv%9x;!1VgSA}<B3dkxo7_Um{rFRl=YnLitNm@I5y6DB71Au8IY
zP-p)(SGu{zPEI`(x~EJfdCXdI8>opoN^vnTbzqASf9p&{_WSk64<8E0%|)Ex_oLmO
zzB^fZ4ED{-C?>Fxy0sHuCZC`O4%(CDcXN}KFWJ@Kj*eJ?3>M$YL!`46RWC{EU$gvo
zn@w_ts-N)vACoYM%<>tbPwpCS%lYuvIfvqpurDBB#3s?tJS<dHpc)>*!!fsBjFB&W
zegq~<_fvbN&;g~j`hX=*)C<#v!@@71yl9sY_~|`2H6+n>p0)00iGp>doS6=pr;mYa
znlg}e2(g!dFGJKFF#N_>j5hdEjm4oLigzNkCU7C@GfsPH>5z4dmeyN47L^{=k2Aql
zHO(3*$V-Y{aEsuRFr*F`)R8e~%k&953PjGv2L1-3jd_m-IiM%1OJZdMTnzmzsj_vx
z0N(yBp?=({A?RG5>%hnlJfFAwJLEkpzC0X!|7+KKvaTh<ie^+-s5^W}L5=%5@VRy}
zdZ1>~ZvhNx<>+{NNEd5t&}GmQa(t`D_n9hzM|j)Jd{gbuY#c?(_QhK(1{MG4^4@Wf
zqtV2_%mPGMt;ly7-?<Ij`;a-+rdU0S_Vd6%i%qpTGai*--)xn(r2xMc7UcY*Z(UHP
zzGKg>{rGE~hi{Wp6kAP$6dMryx4ePm^JmvU<M6js5N$m>_$8d4vJs{EzO+FZ!}!TR
zBgZ*A&BBT&3T6#ctgq);w#A)0BV?<0<S>y#aA^r4ZnL)wE0W>OtEMnYE35Q>eo4vx
z17z-h*h5sOH=||-jDM|dp*?pEn2bg65ZvwY;q<`7z*JMy(oXG-x=^3lPY5*vtn;=S
zlN1nCchj!5dCg?pOfMTMF{1wY<ND{oC_AUFzIJa<dXbOUf@S^}T6{usMfsMSzVB(T
zItB&~e2A>4bA3c${b<*j(e)wcFqoe|JsQ#fb%C+#{z1MS*Z&w8ni6!!q^mZ}Dy&6a
zd+L8VJbBI<Hdb>W0w2OfhLt~~9UrLMAX~DhYWiR4_kS?6epWYhJh*#k^5(&YIz=uy
zd)yAm8|F~>{Lz9`zlm$C-LI5+QaLWUXjN&uzUKX^oHD5T#`x{Zaump+MzoA|^!ZT7
z3i6@31cTGwU0v%N^nWD+j98;v3Y&yx>0NXMb0%XoG`>hc6K2@qDS5$K@lqSnmjkEx
zMUO_dLE`%d{Xv)9*DwP2Q`5bT#CX_fa8BUd;5{{8H(dw(rf;<qXsM;8H=r%S9QST`
zohh%$krN2^GxH*~6QRW1o6rOYcapYfX(9A8u=90YQtyh7jC@SzsIZ6Ws@PrG`UXVI
z(6vkHhmicdq{0stWh2k{9e2*OH10mLfSD|c+tnMhjaMyKHNkUAG-mxz!C9gq^<j^@
zWycp89F|uJC!1~CXKRh~qteg#UaoF6px{YN>?XJI0<nH1LStp=-vBV^NnXX>5CDeG
z5m9j@{_dmP!0hca-KD;wSGIP&7x_N!-fam?gH^ZeSq=8}fjI*fN8<;L-w<IIc|^jq
z!a@TKL8;rf2Y{_Ht<KP1YXo1PV0LrL2U$fMwWAmY7ss6M+39;=GUGugC~;C$mKKqg
zedxqaL#qURI2xLG{w(&&mWKUyBk-`FPA$eXorxnpB_3lu`xQ9Dt77GwyZ;rfu81&k
z3!yEX7IO-N40ha)h1Z3x8FaI&pB~VjuKPI<0x{ICxVSD{_6^bk1*1HCVXZNt`cdT=
z)=ZER`~+z=bNfqhkNUXrzs7&N4dD?mC?E<ext6xRnx<i3%V*r~f?NxF3KCXuBA}q=
zi+xl-&=R4=WZ7);3I5KF)~Wy<int;T4ptVirLbtWZp&x;bw;^aBqUtx?iOGo0<}aE
zg!KeDIaP}lO|$u(^<VeYa;uN<BO*?-_t|k_iVLma?9cWDe<?rhxL1;|DXiJyRR%EX
z&%+VUcaE(?1UkSqgWaNkh`K{VLc7v6tJw;#>t;m{YiDQ7%(@rnE-o8eP8`Itps>ww
zg%&Q?!Xb8sb`RgB(FXa%lG1G%G0^SC-1sDO7!Biv!11-A!rDmb)=&LD=UJRpR^vXi
zuD>@cN<}3?TFIWf;JLcf)9GAk>5GQdZ6W|yMDmlixz#cZ$+xpbC}xS$zcbyW)3%k`
zSwLy=DrHzo0JF7n+ccEH1opMYcBZSeJRG6ch>|f;77IDZ>G$_HC50g$kYp94rH|{Y
zAn-j=T3W5`i@;kMft#@dagx(<NzBJhP7qc+<S6_z&bzq%OzNwzA9j6@gehJ@FTvI(
z#5n>qi(7>_&1$d4tTHV9{ti=g!Pq5r9PiNT)UXf<wV#Z-Enm96YDfD&Xv!IO7@I<X
zx7(bB9m}g%xx%uTm=yM%UpBH~shqZj*E$Zc6hT8NzQ07c{tLCph>#kGeo(xryA9J)
zO0dZ4c_1=HXh9ozbz;fK>wE()x<}pmxs7<)>*N1gK8Rh1{Y{jVpawhJ&^v!oR@O?~
z(XB9L49NJ#nQof42%g~bDN1;~GREB<`K2aT8lKvI{M6b4<gO0gabbUQ+QMRzqkU$3
z*`si~%BAs~Wv)@N>D(P~WN!}|0-$4CA8jq_kKUg14l*+0WSNLKky;^asszD4(u{W5
z@A60EZtG4iZ1<}xc)^nW!$=;K;F%xST&oA?UpzPn#5TFOTxYHl63~n?3?_o2Z&vFv
zyW&=%Rg!*^fp^nfJbjwm)WI;@NK!T<eFFlci$TG@q+m2vNv9m;;+JVfW}YOAOB5>X
zJEq%u4YDbQflW&ABJ$~J4x3!J!;#L*A*_^D5&fKCAHYC*llJ|M%l8nh=k9uq{b)vi
zMmO%-W7C6#3+>DtuiCP*p2Z6le~h-aZ|zhjZY&Z7cxxlurdReErl-x7;@rqGGC$O7
zJtiTm5BMt8nVs#g#G1#=c2yiQgv4+OB>uT>C2$nnmMF1jj#@h6>%B6I)wL^&vX+oU
zcA=M~qHETmfr#*_s4PPF>~eIpjgHQL3U>AQd^91Dnnvu7qcjWT8zp`!==32-I<LNX
zG`Y!TN>Rp7u_SVTNl5gvAu`>Rr2*ImT%l_zmM|(aC^Mebs%WG=sBW~cXkU3KOZuZ9
z(%5z?#&Ljrq^chk1geTkm#&Lcw8f4g#1fW@&SlC{-tm<T3AEu!;SzqU&Kzt!=Vg+T
z)p~NWmk(RbI)Mj=NbHW@o=DpF>u;jlQZoDzzVCOULG!?waiD`?b)2F+mZE=O_%vO{
zbPa{ZAoNB=cEZ-~qq(V2C1vn6XaS(lJ>G@Ev%~X)lsBMBCHPUuNgx}UuB$M%o$NCP
zGGIpmAENx~Rn|+a@aLZlvKUk#M<i$P$1<tCJ<~>W{6+dvBud2w|1-IGKiH22UalOx
z<#Xpu9f@~Y!IIRHiTN3-CNm1(1>?Rm@u}p6UwaeVK(lR~v{3Y1YTU!w^-!8SM7=wz
zD>6UX*(u_Uj;JRIP-QA-Udp~1%%x`dB=>z^E#hsgB?c;51a3?W_3x&qTLN2WZ2|Cz
zB8Iirdb4G6MvS|!UKQGu(G1?0#asm?{HDK5gEicob?jZ<fXAIMko{W;X1%+iEkZ!>
z?~ECNiJ+qoi{@+uooEEXe=(&d;A>jI;e{nJ@#(;SnX0|~{J6TT!ogw*^*ptU>N7@P
zRDqa6p5MX<v^C>jhhClEb-j+nP5Zl>&^o;oeU>aw>-X1PP?`|!?eh<E&WAD;Z&6*)
z{GtA<4ikOK)YNZL{h1!TZbILFzRT+1SRZ;J<cZb!Bh^`nUOs2KRI=Rd{Kr)BMD9n7
z@+N`eVPWc1Rg<tMn0TICT+e%w{6O_s1?OgW<ALU7L)TU;TY)<`KL(~&L*IK@C$3x$
z-0vT;Z9I1Sj+&Vop|+7DioR!Z^KGPU?c8!Q^SJc5sK&LrSim&5qCdR_PnC$RUx?(`
zRPkxt)0D=IuuLY#-2ce}5cRIptuN@fFLW7g%^LLPIuB@PesJ!3yZwslgB!P}Pyg{s
zEIwL<;x8)l&{bpQAfx&35}t9S@fej9vj7VYk-W?9YbJbZ5!rq<V3b@`%i%C-^gr<n
zjF3B+>MeAxy^j5XF7Q5mYV??qnCbtXVkfvfy!5;&e5Q~0B(OS9n*D=~G@W&$lOoCQ
zUj{X-Rr0A5XWtu8^$QPm80!YIJFb*#X#eEnMv{0CM&#3e$YOir#}Mr~3e@UO^5I2b
zXjk|PcUf67=M9^h3f<^>3y_c{*q8KvJG5DQ6j*i#9InLC6ZKB6mHYz3E+{?;**a(x
z8GC@I!1=-YHpk1{gN#g=b110IO}@-st5xR>b{M!o)abgwX<N<KS-uak=4Z8WVWFRi
z_pYQ>2^JT?f5(N9<A85_okkc1i*k8pgB;mo9@l877alFTS|1>p)Ag=GW0}|+2S-+1
zBrdx=5@#D}%DJ76J1>%zpHHB-gpoefWLTEpHeWoRC8-&!w3L#lCV2_{*lhJsB){-o
zzxOLftujW>5jaRqk7q3@mcxFJFE9Nvu62nkr`J>}PR@caC(YZsnTa;x9Yf}dC$a7r
zVtclQ=PNeA>Vxy`kxOCS3qy`=LqHUEcJy@F##LWZNTr5Ex)hL#a6E0i%IAaIOUWx1
zM5K}_U#$@6MI%kmFBRN9qA@%D9fN};+`y$%?G5Tw`SH$%OIn@N0r=1#oY32A?2nn^
zUAz6!t)Eh&!;GC&Ps4Ewm;V2wn4AQnAp7nyY=^RPSWo841&F?G^4Lw4+PPw_+HGI^
zK>sC7eSK<5)48EV3h0q+#uH|ZWJHBuX3N7s`!=h0wraX8GecdKau_0s50$KuEltEv
zd&$lf>HLR+Qr+nauBG2Zv9%GS#em3EQ9@jGql~C2)wdSbY8KN{Qr>r@hc_&+wZhrh
zS_4Wmn!q%dcY%>yTSKfxw%S0uGHD6pcd~DY|M#Na$Wyf`aZm0%3Hp*D-#ix&`QA{{
zBDHhITqQrZF(JCK=%AkJu^x)EU<+o*wH)T)`Tj9VDc(-*cPWo9dIVyFBw(Z8=fLvA
zEW4Hw9st;5@>p~?IgKho7afoFyYxjjT{>sXKwOE#H}!9JYT7?Vh;hqz&wkklhe$I!
zxX|gmp&Cq=`@IZ$C!f_ZMfm?UVFZ4c;*e$L;c<`^UV=na9DK!CpXL|9if;~%w4`S@
z9X3v$jJs=dd*9C2$&}e$>@wwA3gjqC0f~o;Y-PVAemhC%=~dQYMkTlpDp!yBfSsbC
zHz*<+vtnuOL-_r7_|M)B_q?KJCATTBM!D~bpt$xCTnvWwJ&{g4Oxu55sfzWbErsEu
zG+rNNdHz>Hd`q~xAp3oC7Q@f)OG>}g&>-vr+wJgBkHON8GVtNR72=<#Z1dS&M06vH
zOjEk9vNG~LkZfqon!%O=H?p6V$IOJTp{cw+;}0{t-N^k?cW>id`JIIsnG*ozM{zNm
z(Xsj;Y9q`Kagsk6j|r7l>Gil!i?N6{hpw4mXJL~A&5(|G&Y8e4s|?}$qED9ptFeW@
z>;U^+>lJIJWqh}>{>^LpvyoI^e`a=}O!`&u;{}2u?3Z0c%6+f30(4vfQ`1?KL9b<6
zhu#SfdqyO?>e{^%|9{x}s;D^HHQnGY!JXjl?!ki-+}#^@C%8ihH16)uxD(vnEx3Dd
zKmDIId-klEebbjTRbBPUW1o=Y4z%q{mcJ7u{#UP7#M9-CDz#3R!&+(B;Zf&=Y}A9Y
za=vZWOumBA7>V5Gl*O3Ph8&(+Pf!$5WHl!IkCl95&uuTyD^V&$9Y$H&Xx;r@WAGhj
zgPE3pUzcQTg^oq;%N)BPhx$D7zaL6b&k_#aQd@mgI0zU1LqXW|40ZVadv~A)U|Z~Q
z9m*8mtPPc=8@c->3u91^d3antD}LNeZ0IX%Y5Do7>b@a{6oud`tM4vhc@Gnl)%oG@
zRE~lH2{a^uK}{S7@P!E*jBf>#epQH-wb0O!4HGRWlQf;GsJ->h3cNsfc$G9=AmErZ
ze1+7gR3;uBRyNV1gcQzEfq;WkA0DV=oaUsr|3EZs?PoSbi1<eRzbA@b0n%UyBGj;P
zd`=!freN8)nvl>a-hDqN|G<DL5n*hvrAURXrMbB*r3*jsn?*g9M4eX4z%XsXS+Xv3
zk0k79Yol|KjkHg@ji(Do{T~-7`{qH?5BPA@p7u?lk-t&VWHi#+8_2M%1R$-2G0Ul5
zzzf%Xx5aj{?iUK;$73ZR=H)m(wt>oAlH>9c97|{6x}A^^tIxgUU}6XRf2zsHMeT&2
za`-TH(?9R7CHRb*h)xX`jyaTUQ-bLZ>F`-sjhe8crTE17(?bzWNNod>y+d*LCTF#k
zbNcc8fqdr#{8%GS3}pHTEeDoIxa|O6qhG5DQ_}Nhc}teyUUPIgRI(;B#+nD-;!8wO
zGPvoG){p0SgM`d14_P@2W1JJI(Uq%f*7X+rgCP`D^=oyEjXt%taUsL%$Bv97%Hrv@
zac-w_5X_*!!s0Sg3F3XZnP#vdt;e@A9qRTA&*}oMe;mP;;raOspk8HczJ)3;=v;KQ
zS|OqY`25GqE$Afnf08rTZ99PNt+#Zlo9L|+%ZiD?FP}5|Q~~_vj#@>m@y8oUF-PMw
z<WgmzgRfs^n~!%AD=-^XQc{*OI$TG7QqqFGOgWkwldRi0!OdNa7Fk^#v!SL{Tf4>C
zc^WC&+8Q)Hi@0=4x}E6;Vb^85Dx}E0?uILP6GPCKNW+$GN5;bgSK)?X3eLrmj;AN9
z=q6par+QACpqLlc*UXqhs5zT~4|jV%xUR|H@Pr!Asd>&xgj!jVsCPN$+?x~r^tKY9
zWL_jy_FYD<>klgbvkyYCjob$UYDXJkj#JTpC<d><=?xS&H+019;ZZTArCQynvKk&%
zk7{<F*yE!%Eq(!YJPB#)k|`}-U#-?6dwrC(k6du`+^mr(E2qDOL7FB)-6yJFnNP;D
z_})4q4=!K7wts&@zK!1{kCsSsyc=au)`DMI!Lzma`xnUV!T2IInH}2clh>+k+E;Au
z`0;jiQ~(CguWw}#k!l8xIb$x>09HSyd||;{2%k&RFfg%Tl~{AY1T7(A_0#(;jX8nt
z#$woYRr5TmaLx&l(YL8DOcPS9rB1n%kQP}x?5y$+Pwn!0LV*-8|GO%;y!K<)e*NU_
zJ-xa{*Wmi=GV1S`p|z{)S6a^4yDVW8B|(UxxEWA)g2*}h^eSTWCaTEG3N$1>EzoKN
zql5I3R*3X9DQW@_nZ9S(Mx6|Imkb^S)|FF?tWk~?e$`V}&VvVx${_CJp`vx$veM?2
z7l_+kD@VP({eJmV;vB%m)srW_Cka+62d}Uz3Vqx=IKH+JMBlrsD}TwQkCY-Kkd%JI
zk1TD5g;h(mPhAS`F;iP?xv($#b9YAv8owMWlnfW=dcXFvvoR*Ear_TG+xQv0FGL~I
zN~<WL9b}^-*876xfC1hzeY<Ic&tTR?N4IA;@2(LijwH141I_p2g`bKlkyn@k>aFVq
z&vVK<lZRP_Jgt}xhLHys5iv;_2bBScaEFe{pnsg~F*NI1L{6m9yYO_=*Q=|x1RoAm
z#)B`Fj+V4m3GgTWheMhUSL?0!dl^0jY^8^OR^+yF3SJ6*ed+F-;HmKT_D+JtQ8L3p
zL=hOtpONVF7fFq&&Ampm3M6anUaEkud-q?+{f#cg6g5*Cg@>mcuZcwd3*mn^FUMmE
za>oPYj`uc<G){#wHeu4+dy^YCRX!RCYlIFriQQ1#R!&V7+QJints41aRkMocruX}S
z!&ZIH0$sK(ePsh@!$U{7Ha2R}xi+Kh`sjF}<sv!3t=G?lJgod|k_Kh;3<dnL4uh5_
z3qRY0aWfS=fP-ch7Xd;jN*{BBkurj&MLm0!o4XA?*3Qdj4Si|VK+{Wn(mv|d`lY%b
z-N=+?uTDTWMr((`f<d=otST?^Q6D*&)yH@YBFimMC0^RB32D8szgymn$ALHc8fk_T
zeKNWb@*mzzJ$^75a;zlo*Qw&S<Mj6GPqhrbRy6OjO10i;h8n9(VLybsED^Mvx61OG
z{Gg*}wCNqKuLdM~9vf8mR3bSSI|H{?%%k*7%6gTE6T@FQ2!^@6=-gZr()dLExl-gp
zgBWR*gVMk+PoWqNY%oGWO0lRnKUkSu(-~HN*|he1zqi2>m$-&#|7b>n8I*L?rqb>0
z1I!pdjioadbu(fl)ut^}lo8A~;lxlMx_hYDZEKsyS?kGVc*tBquD6Iy3^*r_ICkdw
z`hTYViXV=6am2x`E<M}Ih||i9F@}1yr>6sHr@6VmrKDhyVZLC%><+-Q_;(|oTQ&FH
zEn#-}^WD{C99_(^5wL(*44*Mp(xoPAAuZG<t9ML3$!jnmRrpm3t;tA*0nVH`QNYFH
z>89tL`efLM*68=8r50^8Kue0Ep|K|r>K{lXMz`rlb%D@_V9}7mk$~}>h38R)DZF}H
zwOY-j;=V6<Io~D^5BowSB|ad79Gc|5xKy%kECm$_F71i*>o%d8L`S4w<0ZnxISmw^
zhQDV1-yusLuB|$xt|c7<cjVQ-XcQSWXeyj)r^G6G@@}bbj0GTgH!m5%LQH%W%JNk;
z5o5o)J<dS-^Wmf@tssgY8{YI_V+6!8s{NY}E)2}>+1wmdyijpJ*lRC#zzO??aL%eN
z4gSA`@5VoBKn|o5Y_$8f`?AS~q&t|T2tO0pq=ec~N#dP@vjliD3u3#3)jcU8+Y~8(
zIVOT>{651uI{KIaBO*Ly)Vh0j?7_M+gf{*eabs_93`qk8zL9>J|2w?gqQ}w$k~V9&
z+lBIApWe1^@}6La$47skbI;5}uQdZ~Uu)wQ_(%8^Gk=J@A%H2sUa?dW($4azWM%<<
zedR`F{Cs8sJ3HC&u?h9_@u?wFHV)oYyto#Bybl39HU8ZI+gV!L`La~F!V=R7Cr49B
zn!#D^vOa+m{*Ko=jCSy&b#v>AP3q)`-fMijY(G;9_Th;}>z0X;6=jF^$smjcDXWM*
zre|2&hc*9Ef$g#4Ec^4m^mNRV;TU5-rOa|jfzE%WN@CNtrO(Q;a}j=NF;m3<&YZxo
zZA$)sCpBUPRgKb?3`XiG<3@TABa@vLcAXi_$S8Y42GaiiRc|yw^G8<Pil8X5pC7n~
z(NpJ5z*13x$w-I?+w@ZfSJ&Ua&^i^!wJ9NczK|85)SD+cgTr2Qn@VfL<x+yyVlTc~
zmQJxI_n_t|y$Nwbvvfi?|B1pFiIJ*ZaV$J$)MJ-I<qU>+-zJTi?+Rq&EnGKJP*Bz`
zn+&ZkkOuz7<^aY@ECIpwxgzZDLEM9<4Rrzr;Ca8|i%O!FKp{;;h+nv?hDmH1`#k_o
z;f?MO*C8A4h;g>ef9M(8vwXf<biCWODn7Q{z}ry^Rb4%Bc-ai3|Lok{@R%z3%AZcw
z>6+`A_&Ch5%PV8gee6aTN=TQ3(%`W%{MuG42hACjzPZo#PUPgDJ>R`NPno<cHiH4y
z_zHe}<4p0L0av-suKW9C>jsK6q8psOqBu+Fe!@W1%4QzW-JK{aA9T^oH!P@KN&fkY
zYU870roy}C0^;Xro6YH;-g8D@$^m_l;GT3o@sqLT*jo%DnqHD=-7}2G;H<dIppgwL
z>N0a;6eTBb*aHNWsx)Bko)Gw-iYfgJ?dwAT4Wo)MY3ju2tT~95M_Xt0>P3h32N#8`
zcU$1Nyn4S&cX+(Nt6FPYHFOzGj@o49?hsHVF$b;Q5*&9&Rpi-+>g}U`?I1T<`3*YT
z59j`j>_Pv|sw72KtrgF^YI+4hI<6EwTJf+}3onuY9#%U23>G%&3m8P`2Z2le!<LbX
zRv(9@(Hi$_v~Va1al4F;HCx=Q$=zNjASq+qmZVyw9N=`0RPa6R1p^LnqKA)cQ?Ju=
z-0!<)_j+DJ&!+-EB~(JrOFv87=1(O#CsOCY@b4bU>UVp&x00;-SaMY`f7THzAUCSo
zjc-+0?MHei6x;CPb0r;QRBP0Xqq?Q{4d}zQFo|`24snZfPrBK2{H-$kR()OA2f|d!
z0e|6sCqLfFC7H^t8+l+|3lvw!=KtNztEskPPWmXw+iyMxoMs_1AWD#mf_k=Si#Us`
zlAs#rQtqJ<XJ2xV>wGVj97mGh<9BAoAM-$tgEqcMusB}P!}rHCmU>lO^~E05jkw5d
zI+%wDnlW`^rG*1WC9OT^qyyuuEIxbiItNj-t?j|->eW~WF8vs>f0W~3OP;}4(H&(W
zl&eLxahtpg?$hKtN~N1+38U%yb?1+S9FBS~wh9K#t#ln1YD!Mmn7(r#z8Rk=_m3aX
z{x&${#jbkRE_9RHBY%5gcu@-Ci4J{ONy1F*GrQyJkVaJ>UjSuy^UkYlVw$;x0nn`c
zwYnQgW{icsoxf}>-A{rGhL&Ayd8$SZ$9YpS4r}sw)p;p?<mtF)mlrV{r*2;kHQe3w
zvS(HcXz6Zcw~(d>kJiO9Dj?0eT5!~$$b*PK5scHlV~u4M8075f?Vc|WPTr7+$rJnB
z6IbT!XRmr}6SFN9KI&PamXC6)WN|ODF%b!N1!Im>!%s}a0zG0@n&jI26%QLv>ZZcM
z`+F6U5F%JSp;870BwFxdPiAa2#L_mZ)yTR@!@^h#!$CSs%eJz<Ag!%fo~>1YOo;y3
zY4C}Sn;v*DadCFZYfy_cla$<=9OB8{exPx-o8X0oQd9fw5D?wugX_5p(jbCJk>mPf
zrIv>iJ7~P7p|%tz&NSFQx0JHn)(%uYc%b2w{0h`m{@rTNAsyd)w-o=`neE-=OD?pW
zg6eK!HYzn{La7vQ<*iV#=&^^0w@SgmpgiL4$l%T)9gQ@?Po6&!8^FQYBFB4M^8AB^
zpnt!w@`Pjo#4oyr7P5fe9+-6l6=7#%wv)W-ZG>#=vc8NH#gRa%^fm*FkK>6JJrBp1
zUaaV7ZzP)4CKp!fUkMy3jk}CSarQo;F1uIcdgcCE+>CoYPy2LZr<W@e+1yH%e@qsU
zm(~<2ECpU(hO8vm!)!;_NsZ1s1WJiqBze-l+)d~%FD!EGEWqv}CQp)5iJ=l77@fMe
zQLCO=5uqkeRrCl?dQvr)HM((h+bt{3;Z^Kabv6>&uJm$PGicRc2`{!VW3d%GywXSy
zw<lW0*_s3dlrOU}ZB-R?`3VRR$ZF}lo4@<JKSfABL_d>!VxF_>i|*a^BcZXFKOBeU
z!e(%crMj2LC#6yJPhnDMEbGqs3NR#gJ8S|KfWq#gNzK5SOD3;aU2r2-T;pXe%9Kbb
z7pBs0)CT7y;xWe}QmThzXMCObmXkUcm2D5bPvo=m`fKh4QjDeMrjv$!S$t&~yLYyx
z(twR&T(GjHh)zLTL^WyKy5!60=j2rtB7uC;mDCMF?R1>{PE&!G_L61PPmGT)Qqg9K
zIe^!<_LaFsPd4n`d3&=r#s|u7q|3rq>bt|826OC%8(9_7+3t+PjOJaLcVC+YtSJSu
zUkXEaE(A5|DFMP*(A#$A_qO149B_0X@Y$4=5(_(66hX)-fic;0tW=tnU_lo(eAb!d
zBXDpqmcO}Lag<>ZJS>y8>z}J>mG<Xo*Z)2b8)pn_1Ip3kg&bHEJ&c#8nzHvi-+|4O
z-5;G82w)t>r>=g%f}3GUmvPXTvRtCBC3yT)Aw>J^LL*}e4$K8^t|K;A%9cAgqSrJ1
zcJK!n&ybs=w*UAB4GxyNTLN1jhDTLoQwQ=T>LS2PWjYoqX%5;X^P2!-RJO#;eHq*2
zxY};J5yW_U=!`wJh@o{*aIh)7>!orvd8g8UjkzHeLBnkBV3XT~<)GFUq3F-*>MGm!
zo2BJ_lrG{b!oj(=G)+tPIoF2pVwa4^Ye3F!T3mMILp)LO{cGrpo0mbZb+1z<g6}-3
z&rj0%_)e=Vc`HX!fj^Jz8M;D=^nS&Io13kXq(q)3$bq(4+c>Bs0Y8W07Zki89<yQ&
zZ+V`M%Mu{dHZ;VXr1%EnY|lobsl^)wUs6b&9jz=@F;3P769fEX^i2`WDByR_{|tf?
zHES(ZS~<K+ABYHVI@{t3RIN{LKUbSen{G(9))&L10W6@EhJ*)g5{6U-RK5^Kk2$?%
zo;1p$hg4+Ty0(1yQz4_EP}!<nL+c6&MWzP5fnzm8a8SU%aHe5;F*&sTTccEfUZs*;
zFgSX(v|L3@&)C>xY%I8vG2dMvAtG$0GSh6$gBD~v2o8+OdB-}K()0s3gu5K(7KjTy
zcmvC^;7eZ(jPES<zMP&D^D#m}jf%Q2!Dd|s7?yDQfAr^kAeSh9PsKD^VvdcAV_Th@
z{sd-FG0q`DAr{wyrs-0=zu@n{t7^)d|Kl7;>{hxP1qNnRR73gGz<@06*m%Dc_l_%d
z*sBf3|CoGxW4`PCMF-prA|4j9Y9CT<cx{Sr=zfgGbroXUnj$O|dDO1!D}cC&d(Rv0
z&rAd6xKChByM4vo1bvRe_tvkzEG(uS$sm%u2?m&uj-^!M3Aipvq-xf_r#H56eMS3>
z=)QYgvm0^9l@1B&3ME`oeSj}czpV4m6gYe%7AT8nL0=oCi{qRnfDF6;tMf2KJEp+D
z=mqzn7eGO=<R|W_6EIwVJ>W1gF$?qX-IVk=nl%j(k79ZazAc?}S0sZF6X_{Ecqa(W
z@w5^<iO-q!bi?&w>HAt;?3+GYp!kEb$L;5Ag%`3@7ivcY`Ju;n3oD*(Zx_O<gnqgE
z7~PgeR}}S?BHmbZM9$giuEZ3b`|Ze2(AS&+@tYPOAwKJ6BBzN@B%cabh8z?0WF!Dy
zWye}O9jlsGc~m*3UQ#L_VzFmlr)=5B5lm0v_t(BI6mkqh@D=H=LU<{zQHh#tVvVom
z_H*#Y{kApR#VY`<U#g06T*N`=4esX_Q)$HWFQ%JiUzw&lTY;m5+%ftkut(Dittr%g
z5atCBNJP2SJiI2G2(8Qh#kr?v7VgQfKN}1Bz7+S1s_%FI9hpH2P0(>C`om3RGMm{L
z7(b!!+cQ%glQ31nzO;uW{M#}Uic%Uu0(=)<5$=a*a@hu{Cb%8gvwSQkxkma$M4(<u
zZ|saFU{piH+u{ZjUW(0NYjWm1tJ=yB#6{6n5|d%U4C~o4+pKe;b!4XIwX#1H6nZ3j
zBm&R1nnmCbXSMScsOAk=_jVZJ7wx;fLy>Syn)z#R#>e%ZBl<u@Pj30tXwal$?B<8T
z;oBuC&;wBXSP~G_eyK5<vXbdItzA_f3(k*4#hgiUsc6P+Yr}xlHq7xB5+N17!B?M$
z2QOLg{(39lEhd5jT6A|$(N5>*OY=S@?KIkDI488~*|_jZMY<9M2bZToLG`iBEtX%G
zP&K=(iMWGnUI76H<`>WkD<0RG`yHo0eGSgP*8=0g+7eAwpun|>RwmetmwiLI5ouqw
zJz0e2<l;RWl+l<;FU_YO%CPKqM3#LX#RQMqIqp*OM`@uqdPIZ^CI)r<Gcb5?6BJ3j
z!L`meLw=fSzRyb2>^%?pO0cNr&Fi`M+25^NlIci_aK+7xfs27VE`fAJ`fX*^m?x3L
zg91P<Vs_I?k{jMuAKuE*OwGnj&12q*E30p<Kl1ji<V$;4`7M$=6yD|l#_Etutjq;H
z_yquT4++7HY)=sJo>L|5*AyT0^LVi{cN{UD@pr`%I;{WHiOHC}CKdfzJf~YKcaqBS
zG3MLly)eugx=5fmNsx?Ve|Wr_GmQ>;w7pe!-m#R`Hlg}PqBn!gJ)2(WnS{LwE_zU`
zDK0lVFIF2bcy4WkDHE}4oIglaqLJyk;V|c1ZRX&^(}hkAE)2cl21cF|#z9Z{R@edn
zQUV-wF{t*kJ9{HLi>RPRjHo%GhI=?;E7I_e2NiA`k$}T4<0Un4p|1KP6N1$->`!gV
ztphI^l*?0h+V7scRGFbx%~w*NUOvMT=Chpmx~`)rX@3WOb2xuMP~Ryi2``M6y|>+#
z<m4qi$wh(x{(2Nk#CK^tk@>v}e|;!HOPH^|Ad|?@R915(#z3UxT*nPK0iGZWsVShc
z{Rv1sKl#7t7QOb{naq@`XpUPHHF<VgZKLP?WKSN5)f5+ZY(waJ{yBKJbc%&rkW@`2
zmsEe8v8i?vH&Lv{mVrVHMV-iA->L^rgm~Sh;XH{y!ou!Ed1Gc&djk`YSrHZAuEEm4
ze7N8cWd{$}-%mHwD`n5;H)YH~U{5gGNvbph59lv@)^~6cfTls<i%7+_p|0znyXs}(
zZC1he7Q`yN*z)`Sp;fT@I^?S|f^$}uY{aeif-y@sWV<^dZ^xV8^m6H{zH+Vv_iyyr
zpUK7KJcCNR3qc{eSpUTwZ6lI>^Bb!!6x3c(n+kB;gEV9xbk;C*AzhFW`O@Ar#sc*(
zEx_bk;PbbWjS`eoS^KC>m>Qqu0N=H}ULysV;EbXr;ofsT8KlUr^Nx8osqxg`1_q~b
z-6;x<?P2r}JI12Do2`k@jwj-lke98*2qg$x^xi+vu=P?uq_=Y`|B5bJP-h*!XPx0Q
z1kB3E2HdjNG|<2ZaMT#R<*S<>T(etN?q+*2&`#>~Ox%!cB}Ol@N}ZHfAPMn`eYp#V
zRk^gR+!nulC;spsQx){=qgdx?_V(82o#8hCIN*QZd>$x(uKRpWXF0+3;Ow#uyU6lR
zvZ?i^W)sLL<+Nx*g>yu5B2)mwN_uyQR^c=~nzozyU0WCd$y^d)NWIjDK|nAuijdE5
zSdZg8l^quqNmk+TLH{Si>>4PAn)tWa_eDJW@iDjm>iFHLoG9QgVKDrj_qpJ^Iogdk
zkbdqDs+}DeFwe?53j7ir9GNneqtsF<NK#s(WMylW?#wInT&m*bsvpTmjrm=zbYNV*
zNOK~+=+FDGG*_Dn)eZ{m_N_G@5hePxH9*y?jXBaPX*<MpV?!1J08mAk6(qTOL4cmm
zsj70kuUs8CRTqc8%I=1U_DX_x`%C$HByGgH^=&RqNH@~jLNzKAAS8q^aQ*6~E(54|
z7C<+0Wc~ip>QbKO=V$Xel^t61TADOWBZ2}}TDdLJ&D7Lm_HBJSM$L*rF3O1fH)+SZ
zaU@-KDWh}MShKgt*1Lpas^6n<Q3x}0Wqe}EPbCXaf4Cv3eYT`l`r#qC#*GPV(-yYm
zinOv3CKwzJoL_i%Cdd70M-}bz`%Fhc-S=ErZ|uF<j!o#B_|i<_wB<e{OV)7Kj)j+u
z!mo&(ZHYs?$;>o@00&|Pndq0vbuPy2uP=_tg8jjl-Vp+a)p?_7LiMPF3@|=?Fft4X
z*$0=P*0Q+jG@KIYIYT+|S^f?o2Vyx@sv4z1P?$i?#_&lHOv@m~+^5?w2jd}t!8uU;
z;@(4_+x%TWeKV8Aq{N*fp{Q7sdL=1Npf~UG)oC*qTIx90oXAr6M0A;Kw9xgUiU@g+
zF`653$$X9uLNq&Z;rtht(>s^HOZ1}X@1w)RKh!JcO?fND-ofOAdTA>@m{SOYy$uxQ
z7g(4}q`~Y}2<YZNn{lvqb&sRhT>_$GqXy5wQ)BY4)>;D=!AA>JHmDuDikCSTyl<mj
z&8teplN{;+LaMrFmYXjzo(bbc@es?pP?2eiY8ur;Wd+@ly}hr!F_ldD$)KQQ>*qKh
zuLi$+m~ry-^`SAav2tJr&wP~%Lj<*$P>mB3XYpdlMEoKqo}SI82UvZ|iO(kCTBK2=
zeXFilJzw3dY2tZ@JWp29YDP_F3O=m~zd2;fv?J>~qY==}nyrwbOmvL?7x4hbV6yUs
ztna4I^yFtkM7kKqM+_ob?F?=68p7`(&<f$Hc<m&hLWlsZ1)enVJk6hkPn7KPy0#$7
zi=S|l`Fz;02hWLO5YYbws=9s`qVGr#?>jsR!;RASswkcRv}<DHT+ioO4F?aNvTz+{
z2&jR*M=I00R@YQzxQn4`isW_4LOyKvc823@5F1cTA++6&Jw9QBB71-yfevcpo~%a4
zRSA|F4;Hs<wjGT@wiH$1Zr9&XZdDW>?6xvMZ0kin(|pz=fg1{{-ejnnnK}mzfDD@t
z$x)$+Bf(wK^L~cs;$b6*p(1QYYqD_q>))rKowH*BLv&~W3h+3?>Ie=NP)4}?&FKs)
ztnJ>JoLiI6rJQb&nhsu1LPQ!9RkyT!`J4I><|v4(CD_h#@hX(hLHeB<yfRo+d(;o_
zTa-*eJ%2yp_ikzBq&A>B?2@a<8znIdC?{vWxpQ$|Dqr@|ob#oH?|mFkg97LDzdlWY
zSuu=;urLgA6+=>J$dGE?k-%rf`*YQF71mN$5`9hPGM>$lj~nS0%iYqy;$%hhc(Fzx
z^tgXS9gwD8t=?xBVGct4QY`~Z2EnQ&x%G*;T(XulL;~PcYUhMECdQBx8_zrSg(gSd
zZTizwRX>Be5DI*T<wiMcL;vm*0Oheo0gIo^E?T$JGtHs2NYLh+Vr$#FJ){<VxH{HA
z9Xqy+6Et@Q!QnRv6pM~`WqA9xNS<_XihI0sY^WH;>5`keg|V@U7Z6B4Xky3I<ckJV
zO)i#qj3(mC_ix%c1ix=H;78ldE=!oX#k{fwoEQ}4=GV)H19_Xkaoe#TMRIA40HaST
zmZ@nZx9CQBmA({mLh(+lpODY@R#X9HrShO@Z#pacObsK}dqK`0hH$@dMc~|3*GL0)
zibp%Y+&70t1)1|so%bV!cpDT?c86tH3OYS~df!v#!r{SBY@M4=7cYj0c>9JH>V4H4
z8ePf3`Tl46s|ubdhvMn&{`hr+P|raVg8OJkio?3&P~l%WeFH|d$$*GE@%U?7OBGMq
z4QWk7CVjugGm(OO7k+>i|0&eEPTi+N&zwMICIQqXMdg~{{Uy77aj%n=;hTNDU0;~v
z6X7UJ<PzYB-e~4Hi0FrXxwG8z%qFBa#oakmWtJu+v^6j3)Cf>P^4&xpN|)NoCnoyn
z5JvLT@?ve5D;!D*0mA3PGHWC_OQHT*wva|^@q={yv{P{{+EM+mdM^QkAkHk2AI)l!
zG4F8_7Xq3Wi0fe+rdZBtH93i`8UrnD|BUa017JEI+rC|4;Lv9mrD!mu?}QC2=-E3+
z>4W9ZQDuHrRg5DqT?2XKMvk}aZZ!&jvYFkWm6~iqUr#PuFi$+*@W^*{=5A3M!L1Tj
zxxQnfzQ1qR@qH!2FyatmqWk%X`mznnfe_e3=xQ{s`k{tkF{ik|;zwPa0=4-GBDB;!
zcGAI8;@OGcHq5@~*tlA9OFsE?um+e~yT1dY?DW6Qf*kzhLfhuu)IE4B2cI9EmOIkk
zh?krW#%+UoBsdb;2vkEHUE8;dGq%?G>U`ocCJZBbjPj0)J|-@T<?uL6@hgNoqUx@R
z-W{O+2Rv-r(BvOKQMG%is@@1x!V&n276(sZ&aSup@H0cR)?JY+Jf6-nw8Wg_CHe!H
zF-I-vY_lL=d$i{5&Qo<w+I){aPboRY-fo{a+LrhR@kK}4u01^Ca6I=1JS({=cCM3C
z^a$qud)mb{srshG?z?6zD05*+WMC&Q>&(V4u<b&s12;q$dFSrJG{uATW$j`s`}DHN
zhF>!C^Wj(!pE^6j8<tO;Z>;`1)+Ge*<v!uqkWNIi@|i(kyEOv|v5tVvak^#@gLhKf
zg_f4U4Y+VJvP$VRrS%IPnlg$!aNR!!P7AwpE`nNY@|A;A!p6Bcc$P9;><dTRvfd!z
z?=A|2Ys2<6NbiWR*FEfU+Q-0AB{9N!@24f>X+d9NMw+`(z#&c2ag-*A1mIZ`7754b
zZBM%O+LKNCy*UaM51v+mrkd@z4G{&8*l#X+0~H@%qTZQzEn^TID}VgrdXSXY-yl!u
zu?}+pl2<Y}&pnkW;1BlS=SPBO&dV{ViNeOGspiIy9o<mZksa6=d-M!jy_+u!*+xqG
zV+kttHR}Tt^7PcZ9bThpnkr(3zquKIVNVb&gjr!3cuBqylUgRq#JZVN`Q8?Ev^PdH
zKdXZRf0T<5D8$_9I+Qg{WPw+OV!AOMZ~m~|p%HxBo?PIxMssu)95Y!$h8KSHKaWH-
zi?PnX#dIMIP^$bRgHR4u&J0s|${!I$D(zlDtcr6H6?>+Y_D@6teqRgYvQS=1%M93J
zxo)=<x(XJO;)J}&R79kzO3WY2HWK=bG>Px}YC;_O8PSzPIU6qrIV}6+pU-}{j#Fct
zE^R|}@b3ylej<lLre^U^C6;f170TMpv5?hp6!R1pHS!i4X>nczy9R?KPa8rp?;l2c
zgBTo$Jr1gN6*r_)9%xa7r1>^frNyV%+J0%Ay}hkHPEqU&y1tdqlRd`lGcxH2`g;m9
zW%0>CasKsvx`HdY^hHSo1<}7~2nPTDZGif6X>_**m><HKKC1?*;q?ft{AJG1F-=jQ
zNdJpH-_?jFGQH50rcWheHX*++o!Dw&ap>JAR_;=_5>O47A6|BP3tG?0e6fd-V-M_N
zi|amHnnI{nm(k0*$@>DoyW{wqMi@d#1m*l(mUVO(9Eq*fh^VR#v&5&c?uvm*tRe=^
z#KCa<CC(D}+eV!}ZY-3iAhmwNg4<t<Mu<2<YQ_yJz1TRdABF$=*217>)YARCL;+ZV
zR~7`;Ht16)+E{w??;ZA`@EBEBn<VImx`Jh{#oVm95<U6kS|G^3yY_#xNs89@_h?0H
z8j^B3QZ8<!o}Z0m?n=2mJFj}sdgWNCR<|5Wl_Nt#%V%Mbv@#ubz@a^NOX|YxIZB~^
zG13u&j-R?-&0}*u^k^td)rhGv8FfOgCwRzAa?gs<oIM64<M4lw<!Gm@NW%V)S`y6b
zvZ#EHCt|DbBxMU4*9aQ>R0V3Md;0@&NvE7U7Xo5gj0UcKZsv*6VU34H0JBP;kiw<b
zBSq*Mob9Xh`rf-sSxx`kWGv6<_c`08<DA09IrEo&zi;^U8R;zVVuMKDdE`-m=lb@j
z|A1n2O~LL1b}{a~M<C0}e%&2UnC@Llr%zch0o3Dzs5A>1!BFLT=E9}7bQ{}Eu-2Bk
z^|^m;6EaDuHB)V`g{Q8|?75f6z{v_#FO92(9iU~<^x2iP%Ch;YVt36t*3bV<e~pi~
zC2FBeqiX`m>*3XhE+YtYz@F5rOGh<}WkNxHrLz#F;*)6Gq=R|4$1hM)gR{-w^)vom
zh&L>nHt*D30rZMFQDD1;&14a{U3v0a+U<2RH43ibLU$&lD4v-eV~oYLjF?rYZMO>J
z)Xn<a)A$EvUhqe4st9DhOaj}TU#BJ-3`I(Yp_j_ds-K^f(L+EpZd70gqNpeYn3y?R
z!J9p6(PIth!mwq;DMCR(E<F?%m%YIk)xcTzqRHz)kIhoVQ9o|cD@~1TY(!3CqW%RG
zw@7y1294k$Tfw3v${cHvzF$e+GNAt;7T8*5yIZ$WXKqfNi(S{lrU&IiW-88_J{w4f
zufDx{0og0-AhTZ+;+XvV#-{vXV`FQL0{L@GBXYTdFYNarUsk4mF<YFUU)|O8WeXwG
zSjZ`bpV=;e{yOhVxbO}~l@{OLmv6MSC3PJ<R50O%5PzV2Y^q*`PI0jJn`Z)QU=#|h
zR&-Q~d@>mGrX|~cm!zpdApEz|vlyoN#o$7Z_q|Dw<9=+c<!4k(p`kHUJ0I`N#FaNv
z<00k+EHYBrcoZ-%O-Am00sCrNNKvcmLUfAKnMglE$bkajV)}AIa-Xv-mEy#YqZrKD
zc%i5>-pD%C$a)-Z_A{*Bhvw1*Vp^Biv&o5Im78Qm%_5mHTMA9pr`qyt0|MT;H*p;o
z&b+#H_bkU~bx_0wVnD{D&fV=yE+m-{=~5%x_xy7<UuXQH9KzI%q6FFwT09xIHEL9}
z@BqyiF6H^ASi1wkDXCxA<R!`EM$(sz(O&{u96ViiMrAHG(KiDZ*e#qtzLd1PyA%o}
z1Nzkz8c&^Rs#vy*p8%%kPytIg?yk@vr5D-waW2+Gm(JcZ9Z_}LpofQrx{?BW-Uog1
zk2O5`hdUQpdDkio=xrMsER~?K5IkgA;K-x~Uo|39&pu={ez>^hA|ox6dR+PI>-Leq
z7Ncy7*R@i?sBTxLl?;BpMHx8+wvFQ)b!DRuzRf8)MwIjY`cW9=wrl=MB`*h;tQ>HW
zlS+`}aYSh==R><pA14~oE;S(my)-IHZ3q$ZcXrqUqL&vdAlZBESPi!70EgFyyUaF2
z0YNt!hfaO$oZ6<2owE>HLg+m}2~gn-dV5}yt*zHlf8*nee3p|{E5bptwl1f;22GBs
zwp&=x@1`x36PYUsmw`;B$%*zFT#icC!N)b$dN}UrTN3<R3fv+#T-@g9WMZt~uvcWd
zCcS<wl3sm~Q%hsLy^-A>b>gXoU4`YLGaIxyEsS~>3o0eG+v8H24XV}<8!044Utz3y
z{4@YE)FivS^epqK8;XnVcH!u@y`eN<-5j0(Z}{)um>RPpVT{{va5Sd={4{9~KDq7Z
zb@ej%Mn)~qa?z`fg|^v9Sh#(wN<>89^YPPza4JhdXT7P)!t25e7f@pTsxVgSL=}6v
z<(_SKAJf_Wr?=Wv*ZQc&T&d60_hCU=O}Mxs9p}E|iL2}7M5^m`2k!9@Y7Ngaq&CCl
z7tSaXYZhPh_*m(P48P$!u`R%cD68m1j2c52+?{@vFB5OPR~y#TPzx@*XngmA=|5nz
z+!#22x(4=;pl17~RUg}@jB~Mpw=dGc+SHRrqP@)lxpR<13K)y1r?y{f0Hc7vL;hLf
ztG!|2wq56-alI7ecq`aBfIpJ5P>zqSq$u2V#Z;Z!f$qt`BKPJs;Qh_xZox_IO09ve
zsBzxx<|p$hepmQ01{P&UTb@CVl4EPMnT0ZjMj9`qaN}*U+g`ER+1T+6IIvW8`L7{C
zr)g-j0yRM>u3?CSKc-@i;@-Qgl4h(&YZ4?dCr7!D5l6l_By8WgETu}Ls5ut>cCjYB
z{W9#K6D^qussuBmo6s^Wj?KS7*?|S@)c?&fMN6n9m}Ktmp8p;Md6$}oV8MYCZDY`(
z?UXQntk%w(RZd8PCivhJJDntMM=RE8$H~R%Sol61x<=^Zbw41EEamq&*lORcuLtol
z```LP+Du6gCsh}wb`vCz_lumXJwjn_4ZB=+K9b}=eUZhae@_$LB9*^~IomTC0<pLm
zGEQDY1()<oa3<Z*+W$NwP%ZgI6~z@w&U{09+4Qz;QT*g`PoJ!a7r+o1BhccvsOZ0b
z0^!SfXAl%jJ};&}hLbJO7+<{&V*!V~|DX)KMVh=^QFyB*|EaLBJ_QmMxihKxs0_yh
ztxF=G?>An=f9F09-H!g^M+`u6|9jiI+-q$_+FkA!(*qD5n$RAJE}?(jsEM;J;%cpN
z;a27Y>hhPEzY;TMwa2CK@7i6@IPQ>T#|i%qA1#+Y;3FtEwS0S6p2<Mh&w4s#wh4U7
zl0>!MPHaFO>|%V=ecF$NS=asUXG~J+NFktWgP9M^OWLtkz&$d$UUzE|XdVpt@kN1#
zbZ6x1u*LEZqF#Zp2-yc4D)&cv7AAkJR31|<_0mK(4qAvG=xI>Q-2bkh?D-E4k(#D$
zZvG7or^JsQ{`BHj*fhYe)XCO&NJh0*Ky7n(3@PdRa(>~Q8-)w?%;0gSzhaP#K?SsE
z%QILTD>$Q1&rP`QQ{ybKaz0wO=SX`9EdJ{<&{%Y4i*3S2q@nvI_cmux%5xW`<^36f
zU~U!Bp{oqKMLFVW-_OdAZ)@(il^PH{eM3GZ#G_BLP|aFX3K4^Cdb>k;Wiq>oInyIk
zvArBq(*|z_hzFKpa!6-uC2{8`K|`dPP0WdO2_Iv}@~)IL(n1y%<*4c7cqZP)zDtj+
zj0VWjH3}zk)^xACBIQj4vC}J&<`4E!a`nUG<9pIdHm_rrVPO9o_9**>kqbX2V&+)U
z`{by~C1EU{slvpl3OFyEYTbl7kzZX~`{eHSHo9e~>;ryE_)fKn_$2l5peU0HYr@Rx
zK!!Yp>0g58Y9eGLEuZtw%<2sZRolSnwZ>4=?q>l#{2NDIV)Wy|T_LHZh;{;EHFf_q
zt^$8EzGxn2)v@pHPbMaVTZ->4n+EQ@xGJ{wUt2O3|CqW`g@0yaHd5GcW4kFU1pi)7
zQsqLM1@^s$!7|Y9pyLIo$o7u{D9>PFF^Y@#<}LbHqKBlmM1z~}O9mNQTMf(w7MY<|
zKJWDO3NNKdP`An+_pBZViflQvNsh{mR}dNxBBRw1T=#iBeu6j8!tT5{%C=KvD5UyO
z($>TT#}_WGB11w{suR&J3A-W$hp@tG21AJ`L-WfgBqxMjgG);7bneMp?Aq`!I=tt0
zJvpv%<8f}bdRiO@kS=p0aE}^ulqWoDE-xr(%2_)p&t{{nMVmb|HM$<gB>6Kr^=ekf
zYD|x09tUyFiNQrT!oTJ8bxp*?e-;<lTvy%0sF_Xl*|``k`gy)FWnNVGCEgOd9aI$!
za!Y{{tw}%OI7fAjzqzPiu(=$-UZklGp(zqaGxa`N*?7k|G@7YqYR8S)21dj)Z_4Pr
z6_^Owtf*lzX3sx#jCUeij@?T{2%DlTJmGF9wGR6$ep}^VOgY<XS75@t6fifd#vkFF
z;4(FGuPU6qb+&!|8mg0NcGPR#<sY||U!#ueymAPJD~gGOi~E62Y#H12IqEj!)%vp7
zQIYBdM%>4RMb^AovZ{YyfV?=!^q|2*)#v%s8i1Zhmauy^DBc$uOzv3O1i2i3p@5p3
zR_>(-W>+_*ek)1QWSw>x6e+xI{Q9+s%je1bvHd~Z-yNF7FDdAZMs2tk%)D1AaDm}_
zu6dL1`vWlLEmRB)3}ne*bK%{H)$bqAbM~5Tz>Vs`DO<&a4+g$w7fwVl17BB|^Y^d(
z`)T)<Aqi`F7avi1?c=9G?An&EDG7OJHtAs_sjQji=5#bhtfh*vJ-O~V9MbAfTfzK`
zj|xWHwHUjcC>ur+9L4+yooD_h;CNgs^l|IGeRETMekI-Jw2qY(Em$AkO}1~eRS}Y7
z$)FRzQ!Q7R{zL5lwugl<AlBcYUKrX~jBQWMKbc0Sx5YC%5bKwrvL@HMzKi>ow(Rvj
zfM1^OJ%|nm!JeMIx@VhXS!}6+h93*jq~;UzSnQ#0*nqJn)lEm+op$QoBjJt$+{;9^
zmNI&uA7~R5B(13;4biI?l(?!FOlmQQoU9AAoJ*qmjC4ueE;M)B=}FP?b75uacXHj4
zPZO7*dV%df*DrB1Kg-9!bo|cLH3){Gm{xhPCH1=d6ViYS$*%vokF9XpUL3rS)%pE&
z`VM?M_zKFpJl3au@_vR_k8sM~#Gr2$>xIook4Erf`b~a1>JdoG#Kb3j+-oGs_65*n
zFTdyIna($p7Wl`ApQRBlw^$7<GXrHBS)^FK&>69&e3D*bH_hipbULm<(YMp9`Waxu
zekw6ou(SmhEK$?_G~4*rXeZ!fvi)Oxg@JjJ&SmB;Gj*r+p|0cY^SzSCFnJn)*-Ymk
zosCt!o^wca<4LP;VytvBtpn3RrzU4a@2rs?CdCgd@iAV-N2GmFSI0&WrNT~!l}rA;
z-FI?n<+{vl0Xpe}t&c%&L_wuT8NJj@PufJw;O^UeXBy8Ik|LI(0HYs{OZu|cC#R~T
zs+{HPhG2>Bnh8X3VCjqmYbtsq`DeG~>7TQ~tQp(Uh9Phy%I9%wYx$r647;WfUToZv
zoqRPt*63(wjS6SyfvDJMu-gv<oY<_k(FC%mkCFm=hjf3-$$U>bvki^{K1M42277}~
zM#kmgT}L#y+V$^Ntop6g6e5>MmzJJi)VX-7LS8PHoM<+0z<6|dNfnB$?2$ox!WrB}
zj}dW^W7fvRBoCZS#)s(par9}~@Q**ZA3!AswwoRY|4<+|p6$w5Q_fqEc%JWQ)ExDr
zYp2k^9bvuL;e0N{q`IJ(=L5SAS9e?m;Z2z61Pfh0^`~Yh)fImFp=UOPEv*3|hS4(N
zH-G`ZrkrY2sIdB?j(t{rSe+bv+F|yalfBIgdpk23MB_@5LR>*6A74ca$PC+@?s;*c
zD<f?CN@GP0S50Y)vsPKX7Q^R7aUULGscv((pv@}mW>Bj_TlWUjX0`do_eSyLMTOCu
zU?XY2dS}C&)X4gvsL_&tkdzTCMW|o%2H@3bEl}fyx{DAyX>Qq*_PuX*2h9<19fTx~
zVzAG<fp(R9gJU4n8M_+S*8r=s_dLhg!fvCb$2n)Zmuovo;Y;D)HCe)Mcp_kspf3*e
z!RC6RQbl60EVTeJWC+;_{RoYEHP`%BIzTDURhgwKzA;0&w8Z(vMZu=nvRWooJoK+%
zjhM!&i)IXqY_SpedD~s!hy(z?zy}bA^i5xUkKc_l`?RDFcR0n+Q})l9oimW)T121E
z+;<(Z24j$^qQD(C(@dB&cc{N&#?`(2D!dI<x4E}cdlYB?p~739x^ZA3Ps_K*O0E5i
zx;AHd6n=Gahh8X?ppS;olF1@pv?LJQql5~{)TxL?NGLC4AcQX;?DjeZpoTxUSA(Y>
zmdD=1Av^*7aAo-5fS>Xu+30iN<KOC3!1zitg@H$he}#IZ_8co7#@*UE!W$#T_rAxc
zsnZszHOFAlQ~}MbJ)X^e^LH-aXJ$OAU!kjtb2uX=`{#hN2)-nyYrSmH2hFUkhVz4n
zz?I*fQ`b!edDp{s6uxVw5~{+Is(&STbq-+&?J-zg6*wY>jHZYc^!Ti;Ig0l48>Ybf
zq0-8jtnhVFV-h}%u*5$KHRdQsv~1D%(HRrYL}P{?^UfAQp1dF77kMPF334oQspQ{D
z^@GoE;lzztO`EZPCLXk$iNw&5bIu1@4omYQrV{P($~umI2=yTcl=MObr2{$TxT`sN
z?&+8>lyrT<1U(^zLMV*t%yt_rb~(p9UTdN}j`TPv<ah5>(V@w(!9b|J=Q-DIh$zv7
z;X7paW2h`^lhx<#m3>cG4sddVj+#&V>@1i?H1%m`XWKvQtMM;W)xcq(A<_VSemH}O
zmyT2>CYGHCtpT;LJTwMwT)>B2SLOaN3X66*R3oV8u-w{{0ZvvHK>1%mWl=4FytDl@
z_TlC@_rFC)n>i6`=>3#+cQ&^ax36~gY28p({>nL}V7ScI*)<OLFD-zcr7C77Z{CTZ
zuhy|1(*(b^Hr-%lD?a!Sih20{Je=%|rt!Dk1E!eHPIi9?yykoqK(~1^v!}={WMv5G
ziNJ*3hR*SND8^d*v`2kenS~sRD|2t)jHkdQ7AV#;KR&LtS%7ik7We7u8wyzMSq*q%
zCXH!*gEf3deRz|OXb-sMTBN+IIByO2yh8Kx-5z17J!7wNiAn_9=C`9LC5G~dUk^M4
zyxz<%hXp=jtg-ZcFaejZUn&j^>~)>t4X3;&rF}|7<|F<|w4gK{9X!vpxk#N)t+HQV
zyml#u=LqQuWaxY)PN5jIWypJ!ZIwAj+>?D#OtZ6(veQ)b2^sq>;98ybx%m2-4nF87
zV!aFvR)llz;;@^X>v&=-h4BbR5qg#&g0?o^&EF3e9}f~@X}`m_cYWO&9i3mg24>oS
z$xRuNxxWzpWaEZN+(Q2nJrV+r43S|^bQ@0Gr%V$FK^QY@1sfg{X?PUv;l0GD$q=|X
zcrlUtDUEQD2m=!!YJS;`G?~fS$#iOHF1+oQoY=4Yg&pKhyFAT|E7WjvQ|IFQd#n`D
zB{Ev@J^y>>G&KW*s$}h~7&z^*rb~*A5_|w@jjv>=X5P>8Z%^uU%DcF4U&m?`&DrUX
zMh}rxN#9`cv5*V}+?A0>s=Elf?zw<lRR&cgr0}J|e@~74gU3WX1IF{aue4}GEOmZr
zhGRvR64||ax*5E^(Fe3W^$|S0R}77(Nw<5FBp(|m`x~7u>Mjg;9~YTOPB_%U1c!c)
z_D`&IMQQl@_3g|nG-zW6EO@a0trb~{9c#j;`wjUM1(`M_1`NN<Eff`*_C_JEMpJm!
zOWwZKE{75(!bk}EPGw^LF3;w{pYY~ibtatpO)*dPgk`a?!zga`jef*R-3?*b8H@Gr
z$qtTdKsU<^%mgX0KjVpRY-;`ni7W(N4E~>w2o7qqgEFuVr+!n@c~V}zke^v07*G=m
z-o)=OBr9r4D$CQg82h(!N0U{`b|$7W!DlfXW(Rw8x|oR^bGo{g8fvIv1?MS$qY_Rp
z?j;lX93{2TYxquXR?1Q&7ptAz=J0lN^@mV8n}#5i#aC2_nIKe}<ogSg!tLwl@bCa)
z>trt=n{#&PZs~YsTYB%aQE?e6z!NXq6Vm=AfAHfeRNCz>VM#$e)YIGg8@D}?o_ksw
zPEkEpaL*3r5r52{&N~%|@s6QSIpr_}a?-TSp>R@b==nouVm58eD&~`$+dL?)sY%g3
zb5Y9^t-huNjj&e9Okuq_^?Jg=|0cuo`3v|aU{Phar3Ae1Tc|8B`JUz2A1+oLvb&sM
zolq~anXfV)7&IkSSnzeME>0fca+~t7Y`;3OUrPV~`}hApG7WU7gIl{Zw)$<d9}^8|
zHHdi}XE?F>71E)FgshSy4fJIexe@smzilI}X~5KO8v~#8{D7l@yq@U(?=INMPwKU^
z3q`M&Zk?&np<<y|qp=#Sthw09oP>DFa;Zgg=uC5^l_BLe@UXPp^T~|g>q*<&)UYet
zbT(v%7@rLR04ExS17%M^IvRP0oZ?Xn3QPikO>Q2y!fuOkW!;zKn8B-B0%JZ068W_J
zY@8qB;ounjb(f6ehEppa(ZfDc=GJTui&EFit)3mLk#B`9?yKJ_ChjYHbsUZx+KMBg
z&1O%)U&}Xx>mRd{KBEJv5AZq|h!q4`2w+y*o6DL<1WyTzu0<0r^FvED#S`0oj7)u)
zVu}yCCNn#jkK$*CGb*r?h45x9>uTS%yQ@fL7UjG%%AF3qGki|#%v=+m9ZB(DpG-Hb
zF+V<+i-e+F1nu}#$mso*5%v+KLHpC4ekb59bHYZTYrmHq&s@wl*HdmNJiMxqMR`Ws
zSg)h)KQ__OFYRtn7r%yc3;O#G6nxU*ky#_x*c|_rVZR;5NAY&Vwn!-_)Ey+C={EP8
zS5wE?`0Is4M|!A>GQ@;)Qh)Ssx3<)caXnB~U#aeIUSs9vssg#SDR4=Jry{gyY4h7@
zDd4BkX+ecn>xCr$1C0zndR1A$aZu5^r+CG4i0pAwe2vLICQF@p0AvO&XhdPW`&r7q
zqa%Sss8r>@ACw)tsU`%HGvnu5*k{zRr-g0+r+lE2?+zY7HxRKZnMu`pBH*3oY+a_I
z<JYRwu1t~Z?6FBABZ0ak2%KSa`#pA}U3&4{#_9))|L6DfO|?}Hb`^BxytLEuT=_XQ
zb9y2T0of9=oS_1|-&Bbw7q%Mb7Ke&0hlZSg?iAy7?Y|Ex*Lt#5>R0rZ4oY!x%s7Ca
z0@Wm8`ge&CV-LHt;T`+#NQx1F*x~x&wNeXT(Poflp@T5l6k;v216f{!2eUSVJi2v-
zs;8p}@zYA<8}CoyL`vf<J+&IgVBr>e?YABC=t}@3ZPmtjHl*1llK~fkkFSSuH|fNh
zF^jqm1uQq>8uqc^V~eCHF5jn37B^lfiQ|R84i&sN;kaFPB|KU75W3D%Vv0r7kDPvt
zM2XnbaJ)$gcl+YZnA#GQN6&nulq|RL$V}|ZrQ*Y=)Lai4ij25#7%$~;|4i$Qki_Ly
zVY?zAT+7+rm*;zne@%~bN~t^gbz?p6W3|~5c-KCv-D1_|EgV2vEIL&HI7$=DkxZGi
zJb=!a#O0a2_odlSVKVW?x_b6J@Tge+nL12$nJ!m_dY>%vH@dOL*z)0Tf8vIGhZ#Z@
z?qnhNrp(Lpq1HoGPX|?fMC>b@lAZsLwaZs7LJaJ2pCR^S=8n{3E}16b8=7pbwLmrx
zvU(F(qx!p)&ArJ@H=`x}QFTqdUdcm+BI@ppJ0;;k%$m^uN7YvawAF6gwuR!>;$EcC
z;_gnd;_mJah2T!H;-$D1C=S8hNpUAYaEAcF-MRVqK4<T9pF-ZkTHiNi%rVDE3%pZc
za1TOs6m8A+r3k>Z9xiNuSo}CD+Jse596Rvb9G<#zp)+F;dU^7{@f(F|_4|fty&`LY
zvb~8h^(d@L(pNA`RAz^3z%qlu0xfJ(z=#S@+uqlbqfBvk?dy^S<sIA1Tz-Zjp>YMl
zrO)%7*|%+Ei?{-{4_%g484QDPH;x0l^JMPBmEx%tOy3&`=PJ7-_-XR~Dd$xvNs~M`
zXk~IRlW4x$g3IF&6wPYYY>67>Mu3yO^hcuYv$}>L3q^zuh72|iO852<%~4a$xF`kv
z0j4_Tfw_S?ehbs#{#=>{GBdLsE)ufomh`gQcp)Qi6chqme$4y%6oVMz!7#CFgO1RH
z8k4E@nzgg!O|1?F+(NMO1HV2^vKrKD59e?V8)67mH5T;{cTkX7V$&I^7)f;<+u<yY
z4rhC1zbR$FUW~Nj%GSFxqa>I+2Xo$0Kgk(is6MRX`X_HP{oGrYdlNZfM;Fi@Tba(q
zt6SmSnK`~9Ms^xntVpimze6FyW4*mal-jrV&^xj3b@{q+*`3$Zz_|Fh%Z|nGX{2It
zQF>rfjUYAk(0t41TW&|Dv%xN;GlfI@g~K@TPiSN6d!~+qYJs!~m9eVvYs8V0_E6FT
znd2@G_Sm(%_l%2_raR~(&5pru-7f~8%J+*BcZ|D2Nl6~(=FL(@pZSg3tT7n_8{+zm
z^%X`JdKZgZYhcdnja5v=Q_9W2mf*Sjwp&wudVOrM6|d?g%I(hL`$*<n3{StjhH93*
z&2qhxwhNfh=^(3a-5yrTkk9eIbNxSY$k{*y+ut9tohhRz0M3INpa9$0oc3ddWAQo}
zP^ohAbbLAT7Dz{J_yZOg`h!-Uuc{f@x<}1DRQ<aGZmLq{#kAnynpa~0<g1#x&{)Ka
zErWh7J`<YNfM=?(as=cbG=z1!S1O>Ohm(jntk?{a_=|A_VY2=<6BcCs)iBP=C>uqT
z;eL^Z+E=6LZf_tan_n14!fhN!alqdUJPkia6C@jS)pFR_h>ciOD`3M;Es8tHuTP+p
zGjLI89ZNu+h#Fi0@YTzG)#MCfMtGs%FDVQprr1&0Q*F&1*W-#LltCr2KR%Za;fY0k
zOwM`+w+qLxKW;LVvEa?D4cq7#&0$;#6xVRtaJ5%|&16u2OV>y@$|uVHk$$oBV~n;=
zy-Xe{o+d(wNv}6j)ox!aF}Le@>lE@Ogy#C}_-#RaL5&=9bvmEdZNE9$N>(9LC2I%0
zG_2Q(NzLJCKL2=eMOq(QM=@G=neATL%;MT?Hy-(+;i|s)l|w_Cb!~hnNU7=U-uZMz
zP!zOzX4e*H4M}3#tT)oe7<{}i{AS6S(vf`S=szHg{GS31#TBE4%-z#`o8u3-Ecusg
zx4)>iI-c@y46(~G(;IsVQwfU%FN0VeCPwww=|w&HQWUuKDMGv5ui<Vrv(#>60HNm)
z{h>c)UokPnjC0GP0iNZ<^QUY;eFwFx%aW=9l@!?${lKopN7le#%BL;{XY+Bg+hr(<
z;fzAN$d?7xH?>J{Y4j3~!FH(~13<D2pTT*X4n{`CB-Zu2dV^6=Qelty-bpb%Z(OQT
z{;kMbv#SN;0H+Ek0`)t0b^2C&&SaV`M#4rPgBJ>=G5$lNzEsh3)niA&=c0~@iuG$j
zt!k$|JfV|J+JBb_x0Hr1N4SeZ(2}DH5EN-{MosWQ1TIcU5~Hdw;-hYF@o_fGcV5H}
zRwC%vl@j)LJ8O$l-%)bq9V|c>*Di`UfiIyOX>UQU3SsKLl<XZkNyCg{{kj-s$3u|q
z7*j`W+*N$V)+#H2)3e8Kcfz*FTM73}HTUa5p_?{Z3H8}xlF$ObPg9EsDf%8<1aja6
zhWej35m_KtoL7l?gg|KmE-I}st_msgqsV#Wgv%+Y_G83JK{Q7gGe6c^Gvd;X(<HJH
zlW9<;14Du2jTDec9EYW%470d`fPL*ni~_zY_N32qAnJE%+~|;C|A!e+TkHC|cyjsO
zTdpdLD4@lAE#OR=$k>)!E+Y+J>7mU|4X+G%azKyr3%J{LQcbRx3cH+(mbCP*w|eo}
zB8=qeiDv7Q3}V;)+mOOGyPPpvAv#4LMy2&H^MAsBt}kBF1)Xn+tTTV4O2?7Nk}}g|
z|DD+xiW*0`K2MAFF&L$0c*=F=U0(VG-`SfB)f?9k>M3?Djk^3V!JP&_X>kVU=f$|V
z=3J#BkzUiwF(8;sUFHck7dU;yes(1%j~A7F*a~`=cO3F@`802*Ua7XbM$YP0aVDV4
zpr^GU+n6p2)4F|{`9B%!tkH6*+7zKh+q3%HPT%3Gp$<iK?z6^Lbd$G2ZoqrE3|S~+
z)t9yUsEnc5nYsS`6K$$=OLCs~Y?)y%_pGy6ZdscG78zO?!(57maAUfRN<OYXow-8k
zmxYiT)6gDy*3x`kx&oyE(SW9Hgq>EPhue)X38B$8d)E$LLzR-WCcd6CCDjO(PKM9U
z#py8PAJNk$hx-DwL8s<tcXAD0`(wkNBxJ{G+)Du-gsyu5<G@=Y>>=8{W4WNIFPBU@
zzZ4|?sb-`9W@MCb3hb>uQfDivyRKB$%#*u2H1yb6?G8PfZ*kRNI~Tof=Eqlf`vy^k
znFoDv!JHfF@QuerM<+})?Hx%~b9$A5E2l?(Qe5hyt+W0Bm{j`$RN3J&;aIPwe}G-V
zojUP&%UGZDHqasLf4Vgks%HjWg3<I)k%k=4cF7%;(9i?eDVtEe+LrtaNU-G!n2}*(
z5@Z4)pi`<c3rv=y%||l{)5_TDjt|7}T&uLAPNCDw;cZ3Q9kJ5<=wR~7`coh1tm^Dx
zjT8f;Bx@*ePlv2H4dj0hCZU8=tOM_hllY3&;ciFs9hoK~avxmgAB%KifC#EazL?|L
zJEl2}+w@}f96=o&Nyh<K&p91kN#k20<N3#Bq`Kb-FuZ>zD4<Y{(_vQDD;f2L{!?0q
z6)rcf@7X6^J$ZA(aB*yHEuV>jALBpPRHcHJT&{aWkjNsn24+7=2R?3g8mV!S^W@Z|
z9*dA(>K*johT8U#r29(IxK%X=ms38&$KQ&i%1#QCKAOyc27<%=0$<U|Q?2a0|2(a~
zfo?lHUhq{c$t`P_KKwGn2IpK7!+9`*=>0qR1wIR#@MSXa3VGndu(1iU&MnM$5<T-G
zPmHva#lNnxCxj!;2T@sD73iD1BkqPrf_M^{T#4bOBwGXJX1Xk%7H$<>^KL6<s*VpH
zw>@39hbze3+&Hm8_(5>e!*_|YL;~heTXp4KiJVu45jrv7gopk{xdxo8-~O#E>08@$
zZ9G9u{0f(D`$i&J%U4PHf>cR~{BVVQd!KC|*>S_;#0Mp+%Zzzd<3$6QwI1-J|997~
zHZYw@P<Q>}{j)K5c_ehl*z)cWq<tOow;#FrMw|SPgyOQFI_#Z&U^eFa%*DMyF0g^k
zw*?2=lMi8iqi)jA$&+lk-m~Spb>8GK+)5^vF~qjHsX2tXm=><o&<=A<#sG-4!@lFP
zkmn^aL@CmI3Kc6z>LP4R$MLAbw>FeibWzCekD1+8=wmY`!D@9@%9-=*!hZpUM_v^z
zX}eX4XV?M*i{YND5$CqUprxkH%pE1`8Eas3P}hSmeCO(W*TXLv!xhv+&t62G>R07Y
zQvDO9dF?`9Uc})di}0dBt1@JzEgEiXp1;GYw~K7v3REi)do?^_dZ2N6)0G<yC2fZa
z!wpp!=HW*fXV>}oq~G{>v@u#f^Xdi!7iD#53X?b;q#@bBNh9#v5A-S)p+(TC4k`mW
z5g6FiRgSz7ny(G>yLfxJWUa=F%F?>W?6rz((r3-#DIRkB*3CA@!zykgHsN0fm?N>J
z=`<Oez8RK!RMC2IQE%9skg^avBTdCSBM&VCR4mOzklR`8L!HeRi1jTl=FRS-E1G$G
z&u<#1b)3LWt{-nI6~+AE+)HT*mVzkK)wN!*sNGiT+5`V`eOi%Tw1zcl{G$)|iVyjS
zNVCS(R9V4xqi6NK%vyiINt><h4_r=%_iX|?7b=1tCFv>Xn!^OT%~_<Bu=O69z(LQL
zon;qYTy{pOiK2!IzdbpDf@t0IL{5O^e?l(Rvp{nX&ckp~-I-;Qebq%IIS-z#5euju
zwvp3zUZG=p-u0Z?e-Z$zwTKEYDm8<oxs%f=hsTl<TVHcpo_{HtnAE99nQtRJZj~?8
zZZ&$;Bl&T+mRUrCN{l5b^;RJZ1)7Vks7O@Q&krpLVj|Nrr{-4lH*Q7BSo<JfOqCcE
z9K0zzr16y;SXBBg69|eS&eFlTmHrMQs<U#j%qJi9KIjw+swc>^X5|A49CPNkhGmK_
zeGAWdj7|F28xSA>K>Wf^xP^y?Bj&1|z$99ZOm_fUr3-xCy2B3nbW%}4yCL%^<0rY4
zW8a{MV*@F>ZAsz8do7W>g-rQHOPsdHy5$~(irX<gt>wBRm%u-rVLQgHg*p>)4d<m*
zbWzWK`1U7VshcSTHMSQmcPCD!rJ^ZQHrwG<FEn#cH#PS)_{&$wF9FfFS(eM!Q<>?$
zJahlI36nlgJ5<($^We}@;;Kzv6E?HFta(0od+0N_-E{93BilGL4ooqN-14%S`VHxs
zNBB<oviV$dNUkYCFNqOD#j4+QUXW|<*x56(&{`4oC5*CZh!J3vWv8Y}3Sp@iiJ8;P
z%&fQeEjxHM6U7<&v1@3TN8G*Tmb{Vu_|0Jr*W|(~7RbWqA|?!{BJhj}DOWms)HjSN
zC3jzb9?jh-)zkOpM+ioRt71tX=?in)u89D?;nLxidqmXScvg4%Gp-;f%^#;_e)1u&
z<Yg{8vk~97&G%l_fBFWmf4tPQ@S+e9XgPSa{-i8=eXrvs37yI90#%AS74H|kWC{YK
z#OcIuo)N#;l}1wvrhmNxa3%zcc{88QvyXEm|9&uev(yl&1$oaZ+TZB$qtm~GHtK!!
zKot2ArQv@&lg&RCb(6DilVC#B^!LFQOS@BRSm)8m?mQ=zyP>?^3Yl0ZElQSpGA?9l
zRUYhkxza)+KgYZ%gf1|c49m*Yi}-z3*aO?mX!8lamY0Q`ZE1>m9JhE88htVD<62zg
zG78YG<n-9XF=+X_MMJ6D_9=V4gx@>MQE-1q<Dxj#-fNdQmRbLIm2$)IBVg>@J8c0G
zTl8>h=5n{{#I@s*qrz|`{K9!7+Eb&}u24pE_VeP+@rvEcU3(dL=fY*p|G;N8P*kRK
zohLi+7HjS92J8KRE*;jZ^01`<4^n1pKH4O@e~Pc87?hf`cV+`?=ZAPdR3R>AGL%t}
z?D>1UL?S>UZacNscQ1dMvn}V(|Gm3P5^N}(nM$=0=DG$Kyt4*#wecq}BTB~)cN*8n
zVKOg!uB+Y*p~QV`ky*S)gEukok!FUbk&yyKNO=VsuM;DkSF5H8>W`0D)Fa}>f*bb~
zH}sO2!)LEm*SAYu%%}%&(Y%tX0buTUoKtiVdsORj*Xgmh8!8grdo){zHMplRgUN-y
z<->J3V0`sFVH99$gHq-3#uqunf+MRn$YFxw_B<8tRW!prTW^Na;&nD3zi&K((6#?z
zL+3Aa)Y?cgOO{V;-DV;W!oH7zZ1LfrD)c{?82m3Kdc}hFyzmMCBOGQ{%rn^9!Q)eL
zfr0DW#|dBM5AGv7>*IpaG<c!n$sUDjWoLL8Wku>zVPE6y>}DFdxm7f!L~u-9y7bK9
zzr+T!vpXNXxBx~&IlS6I%H#B1_Pc=}!_yN|RhN<U@s6z@TV1dev-U=_>6_QdeOK26
z5C&4R?~tbUhE!}g^YwQfE_73#Z-#)cfrcXRniKzh9HsJ&pmU*<j|%rizG}nXqE2$x
zk^X|2<jR$&%MBbGyP~a7d;jylZNfkQ+2$o0VhMb0A!DCHd+pT`c4vQamuPgEU99mk
z9J=uQH3PS>hmW%I#%e8Hdft}p%*?)JZU3#t&?T;=UZ7{O)nBTwiT^7(@q!79Jml#?
zlj>+XI9WlkI<lw7v)VF3PTr?@F;GZpe9D!qVP$K{+SSodIQF8g<$RbptXGe(f(HU<
z5Lr-7fe8RQ?-bK`FgINf%k(Az!z0R%8^_Lp6I(SmaK$jnhdm<5-H0*Lz(yFlXphuM
zI>w@D7X+&d;ZWgyBv|O?2sa`4yTw1RsAe#X*KOd4_dW4TBRS6hJ)MM#lDzA^UHlPH
z5fkvUvG3hRsH=w)x(cTe^q^b=wIX5UOM%N#!@RonwEs>7v8c7s2AkdEy0r*GCU1F=
z!!&K0kgk7asp?U&364TidqJX7Id>m${z=|S(GQacqC$){rKXNxgrcW^W@k5Oh3TM=
zKf}Pdu`Bw_tPN-%=j{~j7G(LpSjNGlp@;b9fT&KFO?ii?M>0Z#Elr`UXnC>0F3W=9
zZSbp<V-;;=xKPviQWpOK#GHieuQvc)5ZaL&@k0NlS7w2QEL`w-eKUzqp}Z)iWD1qO
zmSREo%RjX~PO+9Zxgs=cKlJ|`TxY?ep|>X)f63Lc&%pSg<%dwdn-%0houhZBgiUZ1
z?IeH@1WqGjU{!hs18)D)udC;(9o6X8nOr;6CXvYTCC9KT_g)_*$eJX_%Rrbb{4xZ5
zW^VgCMdoXzDh;*N5xBYetP++J&^kGYxpQ29Q(6TE(D}5C{oU}_=<`r%JbfcHL7S_c
zp5n^zGgSi(E~sYwC{c-?qEwDI-Z=Y7ZX~2fJZ3|ZYfjr`!4gUNQI014JPi<sMj%5>
zM%(2`FL+7!yk$Dys#yv080Ko27I0#QEeSMxdt59PDboF`Q>0sEM?$bKc!7So_ISUP
zLnR~p1wbC>!_W0%8X<h^=ra7@Yy9F^sCIyvJFfJ8%bNWcPf5Ap-Bf6XEt{<sv2iEJ
z($1W<%ZF)P^d$+jS&osy^cwVvfnhS3Myvmv(HdNwlpV<bRAURTQCi5@Y9wZRIEW=c
z{t7V=*PJtk%6_+l4-pX?>*BLK(b7GDvq={+ZVC7zWAXel#z_lSHB)_tU2mq*?qloM
zd#D0*k}9^tcg-S+DL2A8Hw>B!pI50zXWwy>VM>x4ONA}+5J<vcg++NO_xC6u5Zv^H
zWT~;#O7~L<ON3L<RBnN@7t(|i16(IkgT2z@5tSlot2>VT(r&BNuB00+rY3gw?>XzL
z!bul!T`&*(q6Vf{65SpJbIFjErGuZ8sZ*1T9N`FQOJOZBYYD>`7mJE678wQ$sT|(j
z$F;g2$g6`?b`ej#Zo(c83-M6fgC>gK-ZL{Ps)o4a{!T_ikh;fTVYeqtnPca-Tk%Ol
z7UP$B{}U|#W^y<!^X<Zyr}_Nqz1=raUY<`}p6^t&kbt1LY`TYQUjx4>QZIByoDXC{
z*a2V7haQ9*9m&k}tz#@|)3C{VE@nRUqT=r5dZQY8;BxlsB8KAW+gx2GOcX#bMu}GC
zR?uH$)3bvVTJ%OL+6`8wZ3lNI4X`KFcPb5(cC-t=6eN6oceFS<y05->hz6H`PIIo8
z-f2C(-8y|bDSsE89^<1np$>1>OA{PyWZ!BhqRh;_&zKkU)+JH|6OpK4`|l)Ay|xsg
zq)!Zj3%oXIA8l?#a03lC*}c!%<38s`(B}kQ!KSJVl!~>-FNqfuH>_U-C~hrDqk<Hn
zpqulZ2xjV8_Iw$1j7sn2^P_n78;!UkKLpx;CI;XylIF^gSJ*FhG_6JDf#hBk$+J|z
zHd~6kq1Tgf>aN$_Q$iTZg~~Y|Ki}4aZT~Rr`W&1JJ;umOg1Lky%x9~3*LQvdq6)8=
zPfZguHTxm;S&Y6EqAkaHMy-B2a=<1F^CF#xD>PvXopz^-uzN6y|1TE+<9`4mzK@${
ze0A0nBH#$uYuMp;xNhS7JkOVcFEkWaSJk#j{dN07ob%Gkeeijr%Qxpkov9B=1uDEP
zKf{h?qYrK1c5$6I!pr)iJvsB<zxtSlM+WA{xq7%z=Rt@(qQ%mT2RV8Xj3k;=x*Vle
zOs#go%w3ARtz>s1%ab@h0YQr`wscAuDw=>8?jEfQlfXl=VW8ocdWV>s{B!TyEy~$V
zvoLzHy!iH9xX#U-K3&0ad1+6YF!Dg@&bs*3x65mS3vP&J+b}fRoTg3}@Pf|T4ek0}
z<JU59j!jm0l{(hz9EjfLp~_kK+OwC|B<2stg3Q1!Rx7&w&xJqt53=MtY9^Q_lt-S$
ziT@-F*H#-7r~wwaLcV<w+K<yoOfQ?AuZz6G(faoh6M<<(c`$)ip!q%tm4BqexFm#|
zi$-Fcb5M{aj?+g|Vd7$vtFl^IG>ycb^yQ{-tqtfqx?a-Up-q%+Lxj}mxaCbzWQxNv
zjfg?23?3Aufd=;xc3v~nDVLa^nxOGo6QG&p98a$_q*8XA^9@=2-0Tw2_*=A*Pb%OI
z0nK!Sy3hK@QRF9%yvoYTn(uZqcfh6+S=5}?;}-of#Mbw0|5JKxVlFf~lIk})@Pi}H
zu3h#L>_qkj6tYU*kU{!BD8r#p^Md|h1iJFZ39}0;_@K*Vu|6q~L=nO^faY@l;K~&(
zEV&aC55c`Z(3UfB=Z;M>lX1?vGtdrGYy~<MmFnqPS{}5kedZeI2WWO%#M26pFu2!N
zBs#-gC3hlqcr-O}qR9)6u9w|n)*d;%EQ)+sURs2e3$<5P;2zHcECa+HMo~qjnWiN;
z&Ni~bO{TvaM(GpN`{h|}+s}XK363>Jx4EorLZ%OylS7Jg!&8l>*)Zchjf<tZ;>?KH
zyg4d#L$9IqrP%;1g-mR+>ktK<QPJg<H}uOG(Q3Zo`v0!evP1$3?6w2j6Z@Lnu_b~(
z;N#=-#f1W4MB%|e7~-4P3SipzQ7ZbR#Ufaafp^?pw`0-u1}bCBxsTLs51_-Prg#gE
zD95Gk3vUBR-$lFXntqBDOR6CCCr*q4oWI23;%r2JxfBf`<oLf@fYGKzIw@hc8DaW#
z!m>QvkwyB5H0<A6$Qm7Ss}=fFF*!f-MDAml?18-E2Ch5#K%o&P66C+}fe{mp7nkIF
zG|6AfPe&)&xpCDe*lQ|sSK5$Vg|ow3Q9?a1<pdRY;Qrla0Yo}J%Q;Z^`70`YKJ?Df
z=Hmr(E!0|~pZKmlz$3{fo?fHFLb?dz|J-UX$<CLjliYntuZKY^O=?b4iJI_jKAX5u
zOP}u`k_D#2wJPaXyGoXTPpP4?q;gu^r-zMb5<PdLS-iO_uZqZ7KOo@qwE)pFxhna_
z*~r2V?1)GtMxd@_$;NQ5iH(mxI$R1=iZJg@C49*G7M%@{D)y?Z%w%du#(P}j>zMd*
zqhi>%09mBIGHn6FiXifU_QJJdHM=j{!3Y-P=>wn#!c(|DJP6c>8vA_Y^6Z~cl*5t!
zeIh`~9O4wO9PoZP|08LHM71A(g+XcDQwwyx7hT>;C$8FTW_Cbny*JnJU(u7Z8$2`i
z(jEqF4HFx2cD*}dH-xaeBN#$HfCuW1kohb!oAusGhRW6Ic1#jnfGE2(cloe#pc~#4
zEe1dwqSs;2X?dR+_@*L}G?fJh1_Fr9g$!8{jDi%ZlfwPYfaJAsROU*Li7_EA{cN4{
zq$Gq{Xo4ntXnISUKOyctYTN+V`Mhq#ct?ay9}#ix-bNf^7)?4=MjkFk*pNjnQBkn1
z-mkX~y^A;<$l+5lA7XK-$%f-WbG>O+XUp;+5mBNM=!MMp_(n%UV4`gX(ZJ!se*vjT
zcKG8j7pmd|Pq<DBHL;OTVyD;Y#_1StVY-pQhwAMb85=_v#AM%JywuFK3H^@jSECwc
z4-VGJPgG!O<N|+=FG6dFzg|5Ur1%-_*wE~7Tpy|iQ~n+|1kXfD8)~<tAaC`P@9=|6
z0qk&!%!RnAZVF9Ejao#p;d28e8SO$#Ze^P=7CA#)I2nw}{4;sqZO4giTG4>Di2uCJ
z$U+_5eJ2#35iTwlt?csVPy<cBDf0X3o`j<_^S=N;cOu;c`L|j@yCKCJo10z~Kqx^7
zaGFg90^m}l!Kp?`!|HgruCBBXvY{QE2}h-MOnMyrk@swl0s4?TK`22jO3Ub|eJ7p{
z_2!FHx6!#%WCF3FQDsp6z?a|ze}QmQ`}_#?HQp?RHr4rSQ+oO}1|#?<cp4~-ka;_$
zx+~G7^17T;7*R|z0oWgvyS&yn@>MZT-=u_<z8+^22Nev<R-c#<C_-#wBKQ)@>7aUP
zMOr9?Pv(gDczQ+L9OKhs5O?<iFJbvikLQMW|1B7ugSJW*0`B~%XhDD}uBwEm+rAR@
zYl6{E|H;+3x=TH*Kh$ug=}ewdDBng8T|>j^ku!+5^L4z49$5zp!>95}Dfwc#;@`zG
zEAOKct0#EPk(dlnfYDoaf52=kz!Ymzu%>n?d^kBf5iq%<-R1ZEVD%g+^{H_jLvkfV
zoEzfWkI~N@h@2}+vdyijMoOaA)^M~L**9&quu&D7>w3I~4adu;*!s5aVc3EmhT3Bd
zPY>}=5_FB{+msZhNBx!V$I2!T-Wm8*tCORTQYTgmKk3Y@hpgv5{;M1*DdOq<7+stP
zL!xoTnzBjgcmf@C{=7~J<Wx*D`J^yk?|o1R;p0jx3Q*(bwB|DVvYGG|ua`McA%E@p
zJQ=9y;q_q~S^D`YBEsUC{nnAm5R77DdmiC&K1pe8>cR)wVaCL;7@53$<s^cf7ABnu
z^=_tNu5T~eh(*-Rz#?SLLITF^dfPZrU}T>GSg5Mv>%9|xXPSIUDt655uSbOoNs3=%
z#L-kCPmLwIT@nGFi){1vDYQNWLCbz*+z*$$`&neGrmik2ke%`m0QbKsG4>RUzr<3+
zK{HD`c7cYADG%xK0soZazIUEu%8CMD9CHv81-(UI-)hR*_7!s5?ZyuHhk0AoQFLI|
zl7$BMFghaFc4d1_%F}&)#6VN)2wxxMN@2dx;Dmb$*XojKo4eqUkj}PRm@ihp9v4&|
zvzP(QeiC#M`=}2ebx7Rx=<JhN1n^nuGLWH1%!(Yq@(SOQkffQ~GwJ4ks}uH}DWy#+
zYxd9y&_!XAN8l9o9NGCDXcA-Ib!T<^bfWpj^ViMv;Y#`4xG?r)A`vY0^L=AOjgLO5
zJ%kxrZrl-EG&$5^;QaRA7jJULaq}0+>fP$6wa48f4Gpal(Jw$5hGYqdJ6E=<BkRNk
zn<DPWj0>lKdus!wA;gxyD>&SZg*q|nw{4U}=;3KiNZeKoNz2~aNz=rm%OGVjib;Hh
z!~yTZ<Jw&?v6!s<=^*FP<srGn*~hDTBOg@ZA)=gADX>P{NaK;}78wGK+YiSjB;}(M
zVP3=0@=V@3t1H%K_Y6>-zM+xN7+u;)Pa`=NR#>ErFtClbcl?=y=SFJM9VJX?Qb-5_
zVvALU#p`MX1*PX1ti@B5QXgq1j0jk`wMfiETUm;1IlRv;?(gp{mh=D9)&3<-o`^>q
zc&t{q_)(fhIhr-u+1jdcx*Z|;>|^>I2X{63tl=+wXF?oT{w}H{D&k@{d%FfEppaem
z^CLu)yT(ANVfgyQy5#Qt^91lt=;~LvW4VJYc%4<bbkgBKW0I6q95PivbpX?x)M)ja
z{{)}oo3Fii1|r7`<x6lvm>c^aT}HWPN#M&K0#juS+*oko@!ip$Xie3`uwARGghf5g
zMnNwXOXt#GXd{ng7pLLrKsY`+0t|DT2-6qg>1S7hGu)OatGvHEX}#Qv3%n;dLr<+a
z9QNjd43sj&hh)?8&&b@&5$fIya98*~i9a9lj`^JwC}!<b#ry-}{I%VDZ9@Wt)mSU@
zZq+VGfQGQWPk#i8qchi<B%If>mHPX;-@vsQBlw-Qgj+S`eO%HWWg|5O-cK1vlyl89
zV2(ddGUiSc*y@=cn(g0;kEjLaKNh9irV)k5l|tud8^Q<$TZy&t&4xZg!MQ-%)EZYI
zLhNdpInfJJ2O!LB$s=!UHzM4NCEX!>#n90Gq&{lk&CCGPn_x7dvBK3^*+osEHpcL+
zy)><}dp_?5Rxp=Lr6d7b2A2lr#kAXmx?oKXDwuGZm#p8s`ZS)s%QCg;A`D0MKQ~w+
zi*`aiPmV<!BTZVGH3cZoGZ)##4j-A&i{wq`z#n-N8=l%mf^C^aafL#d#RglBlL?U>
zYE2`L*D=3N?KbU%sfpM}SvG+*v@qEsz`1v_@*VMSSQ*Ex?^^L0IM=J*Ab|gs`1-W`
z>bkGO+I@<^#M%>(C^EgM^$zwYOoCxFdcYKp`G}A^?<%z4BekD@9S$Xdvkymj97zrr
z->$4oE|s&CuQACNN_S}lbb8+yQc&Jcvd5iG$;>Lj;GDsyV?YJ0Hp=)6|M+gkKw2>i
zqc8H;K%}yr*|FtC>BHQS(CsLRfXK|>WzMXha{ubdqjJ$p1BprqeswA7*m~_Civ)|k
z0&}5;Nqm<S$J(jgzM$ZqH7!!G|JzTS>xpif(N#5l78&4M@k=zWrv9_my<nH^Zd2vs
zR5TElM@j-_xIS$IRGBUji+lt59d^@xBSw()>|ujf<pwMfrbJ<{l)zj+hb%VQe-QSm
z*w=ZJa2^!wS@HYte{kv-17n^8rRO-s!R!j;2-0)!-OwIb55Iz|nQ&JoOCoz032)R_
z(xF?JNm1bw^tmGmLsqDWZ<^unDYMXCTbpVWPG9-oXiPtJp|SaMs6=1x-fZ0@<KxIQ
zCF>{OLw0-}ILU)fn>N3<#zn{9SgR;Qft#CwmXsK3lVO)+g7XZ)=7Q5Xqkh|gkl?mI
zrWaNXG=aZN0o<eMpzqi$yIY|oLwN8P!>&BCGvnSbqh&vuzV*(h&KzOO^)@w7)tt6i
zK^01fQC5vLBZUz$+X8{WCpvtb)?j<35R)J;*#e<g8`5b!B63y|Km6VLYkZsf$FuDo
zH=m!%bm8I@i<SW<DMKYAa5#Y8r4laO^m>KpFvrO!s?Orf^!36<g>(u_&d~6Kz(Otd
zF2zST!T&8%o$c@P@KkI<?x<2Iz2K3}1;^(BdxH<_ssTTyig+U7T{o9|Nbm<Gqgc&)
zrt)g%0axjUrbm;}msRWmcSBL?Ps7lQu&uLDo(TZkGJ<=1Yc`~b(qfe#H%)~%Y21{Y
z)tRuDJMdw;Ch^((4wFX_eEMp_ZR)fn@;mXM_QuZenY4)69~e>`6o>69@j^BJCVt74
ztQSsu+%FoR=te_ii~W$iW7>*su~65RGz5kXA%V6XX0y-{-@r>Rl#53g_5VTjLqzkd
zFL~al>;pqBOS7OQxyl}7vr1o!H?EhWqix`HJz;#5lIlV1p|UcK?DG6K;Sp0=k0pg|
z;o-lFp(XrEQ*)K@b*R(x5jYSNCZGrR*%~AvF?seY0b2RVR(c}Ar4&<eNXaxJn$6yh
ziV!-r?&BHAkA&QRwn*uW=;n{z+Z_xWu`oF?0>mWI^~{tWGCEUJG7W&a%q&evC}Nm&
z!;5e^m>8DDi~e$1jYUqb*w#IWrNO~yq!w%C8dErJDD|ksK)XY_qvPO;;jY{Nv!FGC
zs@Z_T;4fLGa4A-g-B^PXb@_jBB@iA>igAd%jq(->Gt}-9*Y*)ZUn@9ALYj`u_>!z7
zgrxb1EI$>Sl1S%e?=($54usfgFEqw(;bm$E2##X+{BFh~#v2eyp(wW{K~x|GA8*ii
z`Mw(q^JK~xdbYrCW}#pi9~7`C#H^kEJ@(S+;t&)OQEZFBpX}v<2v!;6kmPkSwB9zT
zNHniSvoF0H(DlK3W^V)bclK}qjrc+%o@YiD8^uLbZv;~KqfOXk1+tMo5MdCIxc&F8
zZDO9bo0TLhmFrV|8o{X_MnMALr|f?;P2cK9z%H?xzP~Elt@rnJJ<QAXNw;EWik8;A
z`kbr127wgpxbemLEczL(bUr!87Rn0)Tcm}vG|=GR@T-F}1aDLS*;y~Ct^_LiU+G&k
zG}sxLn$vut{`-bPQISMa@*p8Z93YE;z`%RI@iM2=JXrudRy2|~9B8{xTDBd-VMTu7
zT*)cU8KmSUjy0o1;WW)}Et2uO7Q_otL0JJry)s?OYh;>W0@GX^C|$FZScA(6;AQos
z^F&TBIQ}TdcfF5r7Umu!U_wjM_pPzIZXo>Ld_v_Aw*OJ;m(3+&)tJ7lL~R#RW9(57
zzQBE;mxpx6>wpT!aYWy(iI%4(eVg`U%DZh#dK~uL(N0btB7(UZrG=WBqo(K9_JxbX
z+p!~|>e7_L+y{*4fv&g6pOo6BrIRVCUxy`B5Ye^Rknrt`Gy<+eioqJjO5rHpyz$Ux
zo19P}|6X{UrEY=r=X)wOkz7iAhPyT7Q1MsaXzU|jhxxyvq4^;E<~5<3aD|T_?m`9V
z=iL5xGq4;n(rb5vF;EeaMskj%t?X{o@(+kO@{E9EkyS8KI55waLTMvOj3mV}R!?TE
zORMnFH90P6pfCpB^8J!Q=KFQ#bzkVj(##e|Ucva?LX65l$d9)0KlRafzSqyqeHvdI
zIv#&JSi~8Z^wW@Nrfacjy#I-b4N?xs?<d>+#;_vLp0}+%8I$-!gF(|kWu(?*7_Z(T
z#X`-}5gDsw{Xp5CDtX$*_6#K=Sos;D!bmY)_;=2l>+g25n;##&QjNk@^$X%mPmE?Q
zNO_gGhv0*FlZ4@t;ZcV+zC?B$l?cH4E1@T@_G=$k)COJ!@FamT`*lh0XLac(uX@WK
zX=3ja1@r^NPy+EFc@pBT0Ac8g%WGWv@P*Tm*1x<f+p6m7=Q^SoZTocipYAcc%>C#6
z^f);y0rjGnua2O8F)cMUymf7JE^CyH5?5*&6CpRPA8CD;V7onmj~wMb8F%y8ojuc_
z7m(Y%&W&@^1qhlnV|amod~0)20}alxx>jM$Hq6$M<mXG4xqj?ezJKCzpc$X8@{f-n
zqYy;2H$J=ZgQl078@H+NjpY;z|N7m8{nMORw{Izd5-&N9_OdGlh`!rGmxB8A&{)$Y
zDqoTRlb_QJqeyW&)VJo_1ByWCo(BPVRu&54Gg>HVBr3rf#RfjE1*82H3zK@@>j}aa
z<Gt*)k<8|@Jlqw{T~e5MrmZC8aT)sPb~bvJmht+v%X#0{jdPJLK|j-d0l~%HV+r>p
zmooRTyWs!|(MQN%5SmU}co-0-qqQFCcADw-?2RWa<U{hA_wO$qX26;7p&*`>zH?4G
z2pKhu9w!!Y$O4{Mt{_hTUvdU$25mOM0lq6Y0n8!VU{{{xWF+|QUa0I{l86@tF3Rm+
zP7yMUKs6@|Y#E3+F&bX-sPJp03ivJvYRzTHNqI`xLrQ=7?B%|^34BnEb=~To4Gi%e
z=jm*LzL;tC8GWl8A24wLEC@AF28>)v^u#zW%_z0Dn#?iHvN19iEW4y}f0SYRqfUUS
zs#WyEIx!M$3_dTu=`!9jJo$bS04b<d!|xoAG{EAu-wH2CGcOd)2yZlbJ-0hnKb||Z
zJ;;WitjVO~qLbt>Msjsu-%uJhs$}tAy&~CCyW-E&8Mhxanss`aBX>8Pujc$yihYv8
z<#bC!_7yAO=QnDuKtEzd>mDe)9e%LirPXQLRKia%1dy)rjl8J^_wJjIO;G%~%E`19
zX#46;Z+Tp~1}$#N*)PB;R<rHiiZ^01Bpl@$OROIKfXB7J0Fk3ZPApL}W0TXo^~O3p
z*w`8KonPaX<;cvdU>dEk#kDEP{rj7fVxXe?FHJ=lyJD!jfVxiGh;zbztC3A$m;}{T
z9WdCc?2;juF5$%dYAe&LqdYeSoz7)!b7bf(Y>*#GT<AOwAx8jtq!tZ|fJb5xw}m$+
zTj%%?sZv(YTm%#F5K(5sp~jo3US^8hHBF7NV<K1HH)ee)N81%TpR*`ZUj2HKfFMo0
zBHW>n%bna#3L2+J-<aPx;`!fl4UUS32vy-kzQZU}4V)^u=THBl{IT>$SV@((?g$vh
zMM5f^uWk4iPK{TrRw$Y`_(q05cq^J3&Ux2v=U1p^r&<m7h0tKxzNz8@uL3iSfaSe8
z8>X}8Xh@p<MJ?*yfPGOmd)>IzeQg!qwx2^jk%>)%zlb?CEDtxMqczzL2Tp~Cj{>Og
z0`dtbta`sBg(KnLOBFDDTB<`1c)@A!l-gRf-;fO4ay@P-ewqzZ%p81di5+co`kRJR
zOkEV;L9u+e@I@%zjfJJa6e#n_eZJK-Ug(YJQ>hQ8LfvgfH~TJ!9#`NtqaVkWXv;~u
zlR%E1;()rjgVkS~=WfSk56#E1Y}jXNss-apM8}(zz8}T!XAHeb*KThOR(tbjbcgd@
zJ;+xwAuelWTF1{~Tj?SnsL&pR1}NX5^tlEX+WUA1ao38Z0!c00_~#rKL3v&7=fglI
zNf77{r)^fdJR6RQqzomP{$$%9k`-+lL%*T)rRj~0&Gt7B)1#W{h8}lAdK?ry5-T8V
zUBEkQ8OO9TI@x~F+k4~{bi8D@-(F7^$=d8)_q2fluQFXy&yclD6)(@X$AD-I4I-PR
z>v#4+cRkr_er{yQU^Wze@v>OM?=eaS%t~`k-2m=4!QCaNW6mJ{xB?Q}@YgvJ=34BX
zbm73Gf?7YJ1p~^I98i4-k@S6JV#5EaDrn)@WMz1SX7zT21qj3@6qBN)s%0eQSgGXS
z!C?P%C=$DK!dQI&@=P#S7Z{$*82x(TwA${6WnJI9n1~3=jG_<{P&fEeVgX~O=y}m1
zU7;q@iU=tXWpyydO?Y@D7zc5{M^F|rdnWq8v>Yf3iU#)iLfby&vc0CkGn4o@gNU1@
zj?PI(Gy_S+jV>)5g@+F?-F)p^ia>>}<2K=O(cCF>j5WKbr!R9A)SDmWvYHZrt!~5+
z9d?g8Merm7U1B#;7dld2UhvGVG_@(*kD-I5af}_pTYI4MxNyR8v&b_id_%2&eVIt4
z^oAg~Opl~^%o5(H*~aAx#}c3pJu?*AbiJ@VA1QbI1>Q2PHD7&tty`8vPpLPwh^fDS
zs2pPYE^}t|vE13>4nH>Leq!dh=qL}_d%LC59b4R84Yv~FpGHbw0bQl60<}JJE_*1t
zTK3Ms>~D7@<EK{n#m39b4#KqS9ud&xMBM4IogA(1j(9n5h+agieHV4^CTZ#&kBfN!
zS@q%o9Gf|Lop5B7D?fhy%i0xomoDgin)sjt7$RSGY1ge(qA%~P3%Sx_%)OiDkgjY<
z(;4`2Qd#6S_wD%hbb0fNcyTym4Y)lNtG`E4TQdwx^Bz)5>GEWc9Vk7{-t!*Lun6>j
zH0h_T+!GXPo^nUv%{4J0%4(OKoZ%}L>CZd(P(wnREi_fF#D+3dKVaYX-u;Mp8yUNO
z@UE_`#p{}D^oR7h9&w%u3PjV91;oatm?ZWg$Pzxq6~|DQ0lS~?o1X6*{_*uZjlf)S
zSXh}D?MDL|`+WAW;3<NeiU<qM+#BDuhi)~8ss0GSDTWE`yWdA9j9v(xN*hP3g`1k_
z(ruMFcyzVKt_gRhJNust_4brD?^)F8Du9`|AVIw-GJ$?(EY{|2NHrO*h?O1RX)F%~
zn{&Od#V6@Ey!#-(*^2%z&_E_ii;k$VHCU_By@_lI&wyx|=jZSNl8RwvKkjp`F>w}r
z<?`2<V7yzA(HF&*$~QHk$Q!}P8s-i!0avrwljq)79hRlr`#<H=Y91w4UuqeE06^?G
z#Gomy5>DLCT!WX*Hhw!J6qI);=DZbtmX<2U%3dHa*`ja4#*J|V0j12$cN5KDCWRwH
zo~&2+p)-AOS={PRjG=~Yzq=@TI#HSeB(|cBFK;i`D4{J%g$7C=6UE>pB8dx@!YWUS
z7B@Qe#MSt8`6=A2;XnJP$<)V-Qch|#QSL#v%|Ex=pM7+i6Y@|AaX!oUmTAH%U<tA%
zkb6FQyv(`YT3co&-)Q9<jf8~5Dy1KB#w$Y|@z?MFVIiKCrqm#C6enqVu8oHZGLf(v
z51$w5RVeV1D_5Fc#S89@E`xZ1XRtWbxu=u1+l>oUGbq#U%8A3@uMU<g%Fe}ZPBi)+
z?rH#WSIBi7a?8cBraK9s-T*)|#?2!P=a3%>HgPDQ{Hddl^RJhtGda=e%k~r<)FQF%
zu;L%FL!kAnf)0racN-q*-;Ro|NyAW)EO`J-54Rk1E1f@*Z>sXHC{w>3zPJ9lx{~dV
zlqttAwy&TeDDJ6Ezfxmy)|uy_^$6||TA(brJ1=C)#rZJRlG6~V({ogH^=;s2#8&7l
zkrxLv{Q&l-j!}x*(+tWy9uSbGZWELI-Tb*(Vx$%hb&QajO^6eYaFoV|wRjyHAYFa_
z6A@8;BxJh;x=>)qM@luwuo#F_%F(&unCO^SgdB&C3VX7Gt$9LqqfJSg<K$UVB_UV6
zZaz5?q0y$;iHC4$+&GB$PJ_{#4V=``%7Uv9ehq~*$(#=vyPMt#Lw|-AbZc9fTCx;H
zat8n_z1oW4YDYleK7T6D(AP}sX93%AQxz>O>CwIIsPWKka$HGX)G&}mwi)i&TCOgK
z$pXNTX{7f;=$kjGB&1t?shGn%%iu@5Qvr;mwXE}1@BeZEoCtckpGvr0|4+)-$SIcK
z0IBpu*_PoHRw#VbDxG<mSj~(WP+hrO&0X}-l&|!^pk_S}sjhiVTFsvJ5r-{ZjNoTX
z5@{OSHeG^exPnB^vWsN1b4u_w=0Y0rIfob>dy>~NEzTx}D@s})6Aw`+^GS63V6=ri
zbAwO$(Xwb<S=&d}v_7I{=!KA?{H$j2c&XEp(AECQNdy&{0;j?9+dm46b;#dK`njss
z!MOxk{Jvi8@L3^TS55Hbmj~mXRLWO5RJ?deebOv%Ix9!~BXvY*PWd~uo=*ST_vSo|
zlFH&Z%qRsbE}?i9RlbZi{TONN2wTq&r7cgHcxo$$EeBnk8;1fh8@{SqOGFvkRlC!=
zBxeT%2QbSY4p60*PXC8>{k7@a9oCI7=vtl@Wez<|0qr$B%Xl1H9DawMX*c~wr{uCp
z<9q9q@w<u8fs{N6&_?fa>`iiZd&-AS@JeOW9<hPuq$MJen_-Q10k70A(n~Q$J^Qoz
z&%>whxt6fN<)M|pJ?fTqI;XZXq7V$1IcDoSmWcWG%$6f`w;vtDEP`FhZJm#9b0Upd
zKzTy2)vLCYi&Oc}6;Y7$*d<`m?5pOOk#YYt=Rmj5y1(;o<%Vky5-r{N<#yNnka@?5
zDelh0+#e4EU2fc59+mxBM<^f3^|N_^oljW6Y@);VglMG~e&a6txAM>C%x<EhFvRE6
zL9S@-GZLqEzmH8XuAi}m*faH610L+Mo&2|7YmwEbU4#iXTZ+3h5VG^$;XA!-aij>j
z5wA7h+MM(*wJ7!ZCQfx!zbVyNEmwd5Fji>4z@39xa<<qyS3eMvO#D`t3GB&N`SihW
zGVBAR!h3u`mg_a%B7MxpP5g7Zwb9ZDWzPKH5RJIEgofW?11=6tI!LN?T@K$*-g%JM
zKEdv`aI2<PGR-CAv%=iyR91vuui5qDo036#0K#MwG)Y4yf;?do6m9D<pQ?xki$545
z(S#8dbULVY9=q@<)b{pHT)TbPc^BJXIEhQr3XDyN>fs-*%0793Qw6s8zW9mfshCXU
z2<rr3Dev7%{I(V7WR;F7Q2B(17ABGRKObJ9h;p^x+znF<73eI7q$c`qEMc|odIDL&
z!9NGj{cjf`T%EUXdx))V6tajS+{S~e0y{MN;g+9T6$$4RMbcku7xmoLiyQUWOhfv+
zgM<LiRm;LQ00Jzy^Rg$4!Oa3Lj>XR7*qWF#>&C;ec`fH(pfF{jZ4ci4(^oq?ObKX?
zyu*pn`2OG|kBCz1n&}p!x6uOG)-1%G$8n0!W!<rTwMym#bEuJPDc&J`rV>AL0|!l(
zYu_?3?u3;B`)9azbBk*1EYS%K%sA56T?|WPgKRO3SwTTz=$5vllT!zrZyr+4vISG8
zz+Aa7(XKSz0n}E9r}&qiBjtLxk*6lM0lw>i6U47SJDdk6ulBN;p1@kQI$Sf{2}0u$
zBN+gim&*~DE1CVD?{bZth41E$VR?`U9t0l<VERb>p*;)~;@}l?7TvtNyt90zP*y*+
zBmt#E{#mEvW#wJjL+d8e8QDmfItggEBr_Nq&c>P}j&8QwloiIybS1vf*E&7T&~*2y
zO9S)<dND1EexpIx4vfI>y2EFjXb-z{mI>u?W-QmbkuVb>%+?qAE%l(u>n86TK0j~%
z80M-rSL(&8np1}M-Vi=bg!$B9<SUri5(RA?e4!>NR?Zb0S4yBDqfH&7_M=pUxcUd5
zpdp1DR3u0(-GW=(j|C3OX_x|mnW?<)t{Znxe`#jDHa@}6#;0no?gq{oDXi122<>m{
zV!_e7^)5U3P}fgOVoLFOcLED}J^3bXWDyE<p|I9IWEDK6ogM5@#{t+AKJ7l7phAX)
zi`{}jL6ZN0dWLR`<F)qb)Rmg;-*b*Qv9rauHWDDNUYlo}OkVQc5Kh<-t|S=a`$--q
z#1G?3;eSr%zYgdo=6hM<wbat8ZXe=sJoZw&W|puJJoAo@+8<`dxlD;fe>&XBkK5(V
z&n=p_^h4y5;0EQ=Ipr<H6n;1d2MMu#VoevcjQ)E104&NBan1xartB0^V=t=u#@=4^
zx5>%QW^kEhp~sUK%&sTqt%rNDmk)Ec=<{CHahz7RlLFysL9PvNn&4ZA<C~io$n&*~
zqh;K5fdRvtk<Kx>Om_BpdE#^ROnBSAoP3`$i=+zFe)XAg#1h6!0=O|S-^mrXz}Ic(
z=OT4yCA4VwmMc}W@pMw!1F`}|xVYt7JnDV@awN-GrJPSVj_>aY2>hQ32y*W{mLo-5
zKFM7N^uWyea?R?E1DGo_0|tJqt&B}%9IvOmtx`8|nb4baI?9K-APu;>v||W9ydrbJ
zuKo=bW5p?Gu)IWGG#j4L`jl#cAbe>Rj9SSRlOhOE-zy0Q;tg1N@~%+mC=pJ;9evc;
zEoF(t8l#UT*<i7kyFUPMbL$D~qZLNo^YYoz&X!*}zF>>C{R4#lrf*+zdn^Rq#`(u<
z=<TzA+~7%|xEsnquJTJzzw4g17KyL2z$YiAz#mfrjraqRdA3+Zwm(O9x-Tpa6gb89
znsdUF7JIvdS~KOx70%Pkm7&MRo!uLqukDu1=5yjqiSWL8Bfc4vE8t&QerSWvPt8u$
z%O#JB2C^(Y9Jm++ms*SAdI3LOlerl--I7ICpKEA&FEZ{?68uVJO1121RRsU7Wr7nh
z$k5(J`;d(aexX*?V{k;urj@8e3EQrLlV-ef&RvBjKx`Q4v~I$eE@}4}pJ0vG>iXmZ
z1wTW!K75a7{d0&utjO>`(J^kk)^cPI2J;;l*H%{t<x8D}VG;a0aGv^qHWyMRVO?T*
zqQC;Bu_v&yanX)oAOAB;Q%l>q_*hK$-9ct<1=F})Zi`-&Kp|V;zyOZ;TZ=Ld7&gFt
zGbD4_g3DZ)xp*_l1Zev*9aD=m^%0EQ06MK5x<BRqv16Q0UB81;krd9cl?ZQP5S2*(
z-P|UNeh%2%XP@s_`khA=Ll{O1BsII{r}Ou{I25&*W{1&2Ff#?U=}tSJ0p7u(=szb$
zEOcq9n+m~dKP@@b{$OIth4Ona)yRc@r&pjePX6N-I^GI6h2*}EA&odh*46|p=U5E&
z`*0vP)gB@nn~uN(%?P~}Iq#%fnd;9&e@WpB=g!#7-C0Rz;q;{w(<^_^x<4B{1j@oc
zu!fy>HYiuerZGQ9&VuKr#e~pOf)D#N!?)(evQ#byP0gaV2Wc~sJ|f`6;!<|hy`!N(
zh@%<GYCpV5Hf)MPhLD@ngQ$}D`|Ss)c&yo@C-%2<pcL!c7Xr;^qfI44=fn=L7(KT2
zj3bi%Ny?#OB=8`WzlZAc?d|&^F2!D|Tn84JwQd}I(`*;M@U6S%xM;_vd602D>URyk
z_dSz<@2S=%A3*CSS$Ed4dDIS@UOZ7>e&x0QlIygkTE-inBzUehBZu+aOzKdYZnQ5P
zPH3!qV@cqmC`zJWE;Y(rlA|N)R_B<kidQxKzBH?DJP)uhDEwucA*D(XFI9pCsPn6e
zh2DLk5ojckfk=9FstZ|N0?-l7RgR|`gdNyrX;+8a>)zKl#L~b%R--R$1XLB$<}$RD
z{5uJ|y2A-@y_7Gv1D=;z4Gj%ne}4Yw5S|%{qu3T0G|`&Z>fQ}P^J?F-pGaU)eAhNk
z7H+%0-=^9L!T94j#kJqX<2;on0GLV5z(q|H)_=|#>ot$>=?VR7cVKHvCgWVC#GrXD
z^7J&>p&VYe`jZxc@5%kPmvEiYIQVGEvW%H592Ix6Kd!HlS?zQBPuNLBwxH$R0JHx4
z?-JGL={QA&Q|vJ2@vSSAu=O8|+h1B+E&d-_Zy6O=w{44lkl^m_F2UX19fCWByE_DT
z_rik(cXy{CcoKpH3-0dn);W9ccW>Kw{jd75+FWyt(MK2G5eMx)QSpU(HF{yU&Q!oU
zV5fQCESuJ|R#TpOyMv|YCiN!Y&z&<K$X)<v)W$je?fQM*vEPT|@H%W1V0ldG^tNz0
z#af_{PVL`s$~70X;$l}5B*Ob{gx4PIO)wSoz^VVfRLe}9A+z#CCTejQiR_rQF<D#4
z!rS>rC}34budBkrnq+`M=JON+NY!>@>nfow0flN&zd&+qt3wx~a_y-8iAD77xb5%x
zOQbx<JLgi9+n-sz%c-*OqNiQFUJnQU2bnZjro-kJ2_I!u&jfJJaY9KALktl?s^nR~
zMdDzK?ZzURgB&AA1FoYA2s!S=Muc%#&1B#h9%yq!9v*NP4rdVKJXE&+{i_veu>l^~
zZA)O>NI)4aHO5oG`tx2skDFNSpFz<kDZA|My3hS}+#w-BVe<^3`<VD?7Ei8jzKKa!
zvXcwx7qkWlCWfElBli1#1$9@bMjQkL1Iv$kzp6FD#6FE&aJ)^h4j+OK-?=FyKKV(m
zF2^|X?S{e}BTf%%oTHR`(~lOKXyd;0fJmi)LB9OoXDR@|-d?=8Y=cQcW1;?aZ>>xE
z6a;)}G3;VVlYf16IY6J$L-GNQViyE;{fJVEP6@5a3n0y8$*{z%PW(*jdgp*Ih^t*U
z;<HNwX0b1=;GZm(Y#REzH7p%98c*OnOo;NKt_zVgP-|?<&)?F$wM*bah}e20_3$vo
zw$pL7)LX!34s=)J{VjDq9N|Lry`txL1hmjg12(ABP(B?drvFM7{dCbYdOT){dv1Za
zr1kUPvMg>!CYUV1CQ4(3q?(oBAKcn&r=7$$RMLmTS-r-`KSn{92`%3=d!ic3S@tsf
zbEG+XGvWH)$@34dbh>>lC0TVzn4ccLeX|H6`_oG`UR7TS%Tf9y!)^Mp%gcjv$e)H#
zPG8J~wYLP*wzT_zkoyjU3&N358TN<Rx2-({9AWqlL9Gd!^1W!!=lhisoGmGE#TqtW
z>L;I3&oHkL2z_u`?tOE%>?E+{)pWbbuI<{iROz#%$ubJvsl}2X%kYWJW11Qi!Hz<*
z<c%>T`%qS3Rj`kv$O|~F9(Vn&3k#XuOdfY<bsoRw#%yF9O5syyina7Pc+7u!O40e<
z=pVhzssC_Z*c%!PUYx6?o_7hCn;5wtY|O{lMK)UcG+E`wjvL*4#SKfz#5Z#-<e|)(
zX~x3J$q-)j^=s%8WB#e&-KL#<Rcc0fNhY>eq+XBHE`)2`wZXC5Cuqsj)i8<C0&Vr)
zGDo;^z-(Tr4Slc`>7lP^bH%4><m<mWuTcc7n_%H(w115V=dRGtyImSxH#Zm0g(o|#
z!6~(kp$hTim>_at14^i;%M~92ST3H)5U-Pgwf@|4zrrU!Jrp*~6wPe!0`j}e=QCui
zI7pBQ`BJ6|UYbq_H4%PtXkCRYEs$DmcJIV6b4ND!yZiq4anR~qi-z9F#tW>-o9nWo
z(xOcHD+j@2`Gi1<+fcXna3V>A1>DM;Qx%h;o!yR0ux|Nfg1`Rn+<NIeu_W$lOA+3;
z`g6C%QAgdwTITL&rnR?RU|MbLM^Xbym(nXBrJDL&+u+Ra{q8{UNA~f8za;3`@5Z}^
z5Z7+TrQvIFZ$3V%fp$NK#=%YPruS@olW_`9c|6q9F97gPv2cDJ<`6KUlb4w?wE7v+
z>;yBLh-+ueJ9XY01xE%@Zbmd-rI%b!wc^f1g(t2)l$a!7#eFz*>XOZ)o?5XsHXBKg
zXZSAi3HDG$yN-mv<cB+&+mg}7*1qaJaUkaT0C4#@g@QhlCqfTm-Ow{-qW<7w9v&DV
z=T82bJexAR?`jA)?izaqX*=Bt&j_)!1)3?QqB`1mvx9I-+=OY*@F5B|hZYCCU?1~O
ze|mfKL@L2(hFnpS=0ZbC=!g-GO4ZpONb%a86N1|2%;d>B;sj%2>5qLUFkW1YjEvuu
zWn31!eOx6Yr7TgllBGSRv4t@s%qZ|m$3t)J!I74FRG^Z-d=E}g<5<~|vcGV5uA2f%
z(8W9DD)?A`rYZerl>Ar2u8dhauB1sDKV)Xx3C`ULMe5j#LIUK9#U+xk_1mtUah4EC
zWJyv|A$^0b#W<Ll5rn+wn&wjEm9_*s_vA5|WeH}&3=XGJO+was{wp?7=VqsT<nVlS
z0I(2q+F+r}hsaT7LcRdRtp(oOJB5uOi&Ue(izOGtEt@a4<6N$@xYh@?>r?&DBf8H*
zSM(qK)d)H18}Xx?){Ox*b(>Q}0#R{w+&A05&m8&fjzm2d&@a#Z_OjF2#5mO%kGyMx
z-p<ByxfzznRNj{vz|flE{X|F%M*~@38nJmFsWl9?ia+V1S@Yaj+43?)@la9@U4%55
z)D*d^@*3spG@q=mcG?@kWlc8?9Da58<n6VEvmdiG<iTySC*54V|75WXZmkJ2XYhXQ
zS9BFvk-eiXDR?Bymi&HfwVZTreyCZk*KI{Jb=IBVz0}55z3W--(9(F2b}cWE&)j)U
znpwN9aNp<D^{W<i_<m198eoSbc#7{V|3*~H+;v8{a!~Frc@(`AYJc7ROlozz$I``9
zZ~6G@-Bj8h(uRQo`)M!WfpCS2L3`bw4(-jBJD=!R(^ZQ9!;EA;UyPn%p!{m3LB+O_
zvYHj!13$WuENu&yhbB6r9=;#^2Ha{J1NN1)9yUBLf2PWa$8u{o-UXk9d1|ar9{32~
zv$fU6r{mC3sMC$xp4Wr3tvD7f(1p~>fn$g)Qz_lA9A<$b*>v}?!EF$adS$Xq^O3z|
zA3{%OsfP>vlzP~<AlUU2+_R^LrtJN-l1)O&EfKmU4Gt2}2)&d%e|_vrx(IA?O81%A
z7gM0By4~uAai}041<#33ScGDO|DUn3G?NnJ@!Pj#4y*6@diyRw2*NKbtM6YPug$JN
z1Y-)cfJ7IuSyxG+R!vT!sndO{=UQcR?NKPIRo@NI5S54*Q@9~*%r6N{cz94%Rl2Bf
z3l$jX1Lt>mh|$x5c=jxxiDnGd@BeU%r4A33pAX2frb<7Huo6`RXL4|-YsXz9`gE0!
zOjwvhJyQWPphrtpSQtcD;Hxe8Xxrqr?fNhP2kM-V1A$gp<uYMfDba|N@lJQMwtng$
zXl@Qb>Gm7nr5zuoHqCl&Oni)-0zfNT+{E??C*!f<YR&rK--6*)mKQ<e->kRroQt0m
zyLIn=%yx7i@;E)a5Rs7e$5h|{86KW}$vYSNV5%;9Oz8A<^i%1;O}E>YpXvOSF&{5D
zEV!#U%)Z|ZX(q7W{83Y}V;`a6NU-BzxZTasoWMl4*K>p+4M}8nX@P?DVVIG$&-;@u
z`-+v(Sb|Be>Qa(U^Dn+5CB^YVcP#JZo&A2}nbdBbCXi2ehJ(eMwe!}~l)=W6CqaMo
zalIkrG|3~$Z;;VA*Rndx?f$52)ZYCXPhanCSEY*>4z84#xI4N&kc713-JAJ+^ba!`
zDGiS+1@GJG?RQZX&LMA5eylh5E)8A?P*RsX8QcTVRy|#W3cnEy&7lN*1j{tGnEGBZ
zV>hwA25I;8;3&cq+B5kC1x7v{;p1G=Kl|}s{{B0AaiS;~jo<KKaUJusEe#Si7l0Nr
za7zqGi|)flG7PZIphCouO0~qu$3yvHR-wg#4K$!`nEC{iZZY)g=sEq}rma?Pq1LK6
zmzqq;z+NTC`NkoO%Nx6ZiZA81qpo4p#>p&9c=3Jb^SPBp<!)bYyNUOd&#j{zw;y4L
za1Lxb`pC0paTJCBGhzOF&|D`6(^{B`hXU80S5-08%BtBKUyBWph?vc%9$AkpDYAcE
z1~_9-DY9E}aMZSVFa+#SjoSw9b^#I-e|y$q+bf^1j*b^A_CJhyLRMC&FFQQM@7JiZ
zxw+xq9}7V<z43V7wl9}Sgw}=mJ4^kMkhDa1RXKRwW|^^pvmVV@j-`w=H{#A@c8el4
zQ?^eq3wNgroG8c>MVWuof8nnM>Gu(fcw`8@kdzcd0Sv0n$Sjl0)lh%GK31{<9ffQZ
zR0$E)H;Y0`NGhR#;O_GWjzjMs=eb1aNaGEO7}Ke{c&i5wqrxvQF)S)v@m@{9C@HMU
zw6y++zC`Mrr3)8#{sI#>)lMVOan}~S4F>MEMf7^xA`$jHgj`pWjrVuFG*4#pl&#m}
z{f38Qcij_yzP<?V9{al-<bSl8je{gs>U(0&jeW6H)(2ye&rdw(bD$_M%_SJUwrYhu
zwo)|fM6KC3_^jEN`<$K@QAY@Ez9l6n)RIKGJ8j{ikvPM}-Kho?Mv1tlXNB6Q9hF%R
z&}#h@3)Cp+e0_`Ru>En?cqhGf(<{Bk*CYBFrG-{@=BBpJeP+6UvpCnpdKv&i#G#QC
z$`Db82d;a}55*k$s?K~#BP0?#aKsl0aLu3KJ4<YvWTm)3DI#mrk#)mdKO!6A(VzXA
z^LR@lB>c$wUtf`g8r?YO8)nAwri+^nx!fM<q47-MiI*q}Xa8nqa51aKK-@DX31Jr)
ze*+N2#QrtaFHOc(AJ7IzFyAMg%8KQVGlZa58#+Crp>^^M`LxUF+R|2Bu8)19O)VC_
zRmuQ%JxrhJdn1Hquvl3ajmhrj;Y4P>-GS(ZI>P2L=YcZrnJ72;zeDH$Ddy|{fVE>o
zrAs)5J{vG8E7(`R@Pjbo!~=y}TIA5MZjjSVK4wAF9_;^IMKbCAOht$j@Gv~ickVtr
zIF^vTZ!&Q9XRgBrBjCO}e2=)aPDO!ZHtcY*$nP&wbh(wEO;NStFdKPFGR+4;-hry0
zS9yhxE#$<o1%3)+WdUSwS9k<atBsQS!W5u2M;A51-YUk98x7Qxs4U~q>0C?tskYsx
z&~^LW_JV>y+`bnyLeaN*!I!era$16o>Tk);U-ltL&Z|@rIdxNzibecBMMfIo;^WEI
z9k+4({$0DXj`O3xvt}JH+vza*9w@Z7!F1AF>_Dsd?p6cVUP;XK;$Su(gyUKrjg{4D
zpy7&BHQ}so$!E^5_u5JZNf<Qw*qno-^!jL}k9U_ET^w@G_&YLZBu(>w0n5Fh2Mg}n
zk;JqyfgD|kpMF|>|3>bg{ya?px~QZ`R@jwqxN-}ZevC*L19rnX@kNJ(Hk-bHI=~!-
z$bOiuv4$;ktD(0$x42~(Za==H`Iya$escF3Q|Jz`-o1K(MHUkAycJrZou)mNz6?(~
z6#%HpELlu<m`Dm3u<ou#-cqSd<Z^B>3BNwr!UEl2EOb>*mfofANASgRpCyt#&pCzl
zF(9qLOh6Q2tA-Wjz%@}cU4;>vC1I}zzzy=wuk-X0@Q4wiC+Aj2CuCJ-gE!2<x*28;
znuJT$BbNbOrQeY2I{%GP{Qt+L2f@QgAuBdkBVl+&-*&{#o`O`CQ<=H?#!WgxN%!Yx
zelBhaE;F|D`Gb|RM)wLOUFAa5>phbFp8Wg<Yc%CEV;;W@Y^i7$WfVWB=q~oT-|*Ij
zFk6oK(MH{e{TIVE{@i2hy-gpthta0Y91JEq&zvSKw-j|w(=m9I)36YzQ;!uYv5Rc4
z2DA{%l`RgBybAnTCNI3l<J?YcYE00&y#_blN#e;^S-0TBzYK~S3Wr-uv)A(G2oM!~
zVciK7cr|%LFb>3FcgJ1H#$|ZCJI|VLFM?#7sRcB%bcRnnz&dpy1U)L|q)&n9VQtoh
zaBnrO)!((Fd~?-Ol`fycKWQ8nfJPHP->k%SdaYxpmj1q4)5z?}9s>7dqbaeRlHPae
zJ&q`F7;yeB)R&TNj#&HyY1(*6WMPF2P0j8rcTq=d2TJ>Pj*2nEU{*Z{vPdp77wl^q
z;S9E5{_aDYc!fqNvxQ#dnkO)S4;daYryxu<dpnK@%k%a_sdqfxx({AXblTVJIWR2u
z{$3`E_rYZIiQjOECKt%FiB1pKCLaJG3$}O+4u))tWHuyYPUU7@x5j{)s~>s@A#+^D
z%YM;=6j`x8;pL2lFwNuWwSoyPg(K}G0kf<bc2bk=ZXwqAGL5o&E-FR`4}CFpNFKvO
zSR$|@b=6z8R~_@h)BpEU377<&{~7)QCd^S&px$wJ{G_EbwPC?NhoYnzne|njAp(``
z=3$_~#MNt-B~@o8>{@ldb=}I?PY(b9imI;o%<1zp<(rfu!MCcMmfG1%#-vjxNr9M)
z;b4U<Bs_dWUl=VfRF{VvB)!f$FrIO8U@HJXa7?;__?WZOH0+PK2Dkvl!?HWD`Sk6(
z@|W4YZ}l)zaoZ4?uwsv=?=nWTgkk91v^L5LW#BQiI_)~<e!CR`f=B$1zocvTKfpvZ
zvw|#k`5jj5E)GYXxBGq;OKv_3HFw5LhMTsTZRT?(_3FHQ`pfE=g=aQR=(7pM@Q+2K
zhI3~mY&JjEmqpbmB$i}nYwz|Gx(eIml}2!F)Tg70>93!lcDfG@+mlWZR2vHJ_#G?O
zaOmaBeFEdP>)x(D{k|6&Xp%kKx|0pXV?LVg$uw}*<)S3O&2^`0e?xhB5!JrhG-8Td
ztpy1pi3U?G6%t4mKlJCjiQ`6tuQ(#51EPwR64j_Bj66S@Nx2f&)8_&tk_?F-!fOt!
zf?i~x0E_6rUIHPL5>vsKV`p!6KCR8kIBT}QC)WgSX$Rn>^W96~div~ebm>kU>hVxt
zm}{2c%+R`)PY=C6h`x(b+G-~zLfKG%rA>%I6e#~OzowOtlPxw@Kh)|Ner`cXEAo{u
zNDZ1pw;%nhE};f|2%x|L%Lr{ZoF^+r<cM=tH3~z4)8hwQ;Swp&|Ld^S14AKUthf|2
zE>p%$oXqYqSQ}evxIdMU<6(Vp)Uf55)zJF^VFrB5$Enf{aDa%9j_nW6L==V?O>3je
zQUAjQ7)8`uWlItBzd|7N^RLY3A)zw)Fi)>-U$7Tx1$zolgt_1NWSNhr1Bs0h7G7xF
zz1B$05H<G4Vd?fj2272FZqNhH+^+(_+D(Sy3yj2>lovhMo6hf<Nsjep<7t>I`65kg
zU+bo(G<4vWjG0(<6q`-AM@nkyze0m^5e`{iUz@i7R^X9Y_AiyR<3&dPZmx(?4aB^d
zp(Jq%zh()A3<qYah?3l$guDgmk#Z7dta=)$@Dy4vQ1k5*AEJH0tFlGSPkM>>*EV48
zxfOBBgSn~O8!TilTNV5jE@L*mmHF&!Hp9Tu`Hebfd*oZrXsgLP`%<a@;h}O0jZB~>
z6QtycC9X};dAl3$kKB<dZ!bXuV_FLjt!ht2w#21Kg&}aYFrcf499#yH0bq~=o>7?Y
zkCAvS^2m>oX6E1~LksdkW)iokyT)$8X<1f+@$y$vECZAR{pP~3s7Ro#6m;J=lG0b)
z_9tvR2EK7Xacs?}L6ZIj8Pj}sn;zYFu<a`R_rbdH;N{>s!1qS~0xoCo(dP=VQmNgG
zhsbjTQStYt*5D0!UvGw4g0mx4>eL4AU$<`=&Ly?!v16~0#n|%hco!TFC*5WILu&)U
zFL#Nw=t1%w4C4QWHUC|2E@@+A%+3rMyp%LoXjd~u;W8Dmq}#C(Aw?QOsNGi-LB~de
zk$<ck*lsz_m_C1>drwPJUL7BZAfsHziA;vA@1e}iK%gKB5`Jl5_T=ND_7C&)wy5z_
zSE$a!%@!>zZm#lu%7ACmK5$rP9yS(tuDv>n8KlYyS7eo*Y;e(?5vA|M7lP);je~pu
zT%Pori<(k_A9Bj7Tn<3-_lk{l)*0RdDYS(z;hQJTyWD(?_Dc%6cu6G*M6#a!W8ZTc
z9CBo&BLHQ%$^Pra{uXb43Im67*pb~#4o}K@*iY*6LgmqsN-4_Xg|{x$!pHa*e$15P
z8-UdKi7=wq-!>cj?@11Vxq6(B;c%Py(kH?dfrANy7Hza7oXQFd88`@5252l?vw-pg
z4Fua6ok?BQe*0r}1R*XYQ<E%9i+(*J9f@j~O(y*lGNu(kLpzwun%wf|4}dCnGCdfJ
z=1Ax}w}@bt@Cylqy9;&V*#Ca0q)N8E5!|0B><^bM+?-Hd_d>4Mi*v=2XfV)-1mduD
z=>BpZhw<7A?TUS)#FXVX*e0f(Q85|X3bX2A#9CJ!@Mj3m=3|o@RdsvQIB8h<aR&_A
zBJ|et)e63VWT#;x8X%Si@O80FBji0S>ZyPw=?7BRK+(HRsD&J47*Uy-di*ots;y^Z
z*ZjI;+@-pIf!zP;K72k<r!Bl?mvc!~Vlx{tMx;81?HBu!W->^tb1|mL-QG5R-RKLv
zI4Z_%1L?!xi&wA91ltkl)IG6i+eD>ngoQQn;>#g$acLbDbF-2EHQT#D*mmV}`SM!K
z`S7zowMLJVgMZJr%tOb8Ne=!nidKo<Ages%t8>U7q^R8oNU4M0uXUD1z<nhnW9}F2
z2fcDVuony6Y_09w8a3bhi-qC%nbWd0Fy5CP<Z{5Iex?>mXyeh{Ksam}8<7ufv`D-T
z;zJd>!w0~qcoL+F{4wfbN<6`&(_SomUJ+N&MOhbN(32+r){1^M9|o;zd4(ItWpUVu
zr}%LJ-Z!vNf?7vNuYZ|=egB?zpSp<3Yd~GM=UYsm)4DWJUw~It@#Tt3J2?`OS~Hov
zSpv83OH~zHoAq4b%wN2P0~r`_dC76vUyFHqZf)N8&bNPRm0gt<cl))^tWST5Td5FD
zB?Jc|*t#<(eu{;#ueP}o!!6y$A~by=hrx9~lQJexg}$@|TdCJ1u#Dmh!5tH+H@Wh4
z&`7BzVj#l6J%%RM#hG)|sL^kBokf=#CI5poDfKigG3{vtjP8NiV=IzEU0z>?B||$o
zVa(i0^Wn!_uIhva@eKX{1wsFZtTIYOPHZJLJ>dGc6hpsD3$g6UuK`InlN@tCdfXDy
z_vCljPAnTa!a=~${B&d0m(dx5$5i3!yd_YzNTeXaylG?Nfm-#^mJk`aq2#TZ*+e9z
zxHv`PM3-3IVxkFD&rPpa?_>B&zNR+A%>jUnQs=m0WOkEkYmw<{cg!Ru-4r^BZ%75k
zAQ4~4UghfCPjqv%;$$eWC3ssfEM5;}3HS5q9wxZU7?Wk=0k_@Y{&eHEgtM*@PFn%M
zmtKcQ2{h(_MOLe;ml8lq8f+Nbb>br1XkS12^nK%U|4-WTlTAoHkJDEU;eiu9q?mCm
zS=NZ0vvPNTRxYLDlHJ}#H+gvu;Dzu-#Ggs>d{rt=-f%!Lpp7(h`&R$OxdLHyOugaS
z=HQDsA2ua1$AHMl-3NZ$nL4x4W&_`+-M|*EpKiBs`)oGKUz72E!@$j6$E7mU7wQ_H
zH3p!L?kuqb@LU-I$}m1fZm3&yS6$M9JC8su4mBjeWYagOaXi~GOF{6cON9Bl6dftm
zB7_=gMm@lWP@;WenRapPbCacWfesHEdz&=XInVYZp-ovZ1Z9|hM)f~m=)b<#OWZlG
zY<^9dD(#vtEBY<vw3O4_Bj&V?x)9E?#RVjSNaP|*j)ik%l)*mE8M^1u;4;7G#S7MV
z%JJ)0jlhwsYnay|09W3`WdUtlrgeJo_B@y$3Ke*lvLT`J)~XL3e&^is-v|RJb#Szp
zZYN+u8@p7F>P|hUMbEDo;*Pa4l3_PW1v-c&2o%HCIsxczzR^cO0t2^l`KEwi8AH63
z!Yay#VxotVRA>MDAuu;#B+IwT5Z~(^{#Yr>rmPj3tby12sc*)7$cbU|8?(!c<%5Zp
zH6&oAYy7Y=Sc;{ahrd@n$4kS7@5qN-D{5e2f796@ED<g|0l~_(9&lUb_1rDhv9y4F
zmdlc8ec2ZH{OUESy?2_dCM_xzNDh#2MWjF9`fS|BcEE}&M>PKPTctc^5<am|I&CgV
zDJ8)eN@qtw-0AP^Oz{!o&9bb!lHx62ktC6BIG4wPhzj-P-{R#(7OOKZ5Q*hD;ERh-
zR~)Wvxq5_wW~5i<BizU`XOaudxW8jOx5b@kq*5fzy#D~uGJ(_3z+6o~6pEb-uz7qx
zYSbH^;AmPBc2JJoX<>2yB;8-SY4Y?&71SvbIO4F8oe<>6vC{h$ja~R(sH>yt3YTuH
z0xekIpz(+ZIK*1onTDF=qv6S4nXyw>uY8xm#|Zb(BG;|grRh)m!I^@0w-8y=Vh<xy
z{5bh*9#2NVg3$An+t9a6$@KI$1&M)($&obf8X^ZW0ebhtwt7=#uhB2no@XcD0+2@H
z#vJ*gTN>TeM<Yt8N1>Q(MaNoc$uOZLbhjQkshQ$>w6{0h1hVy(d?xVHP=PSVC$G&f
zc}2&QBuEkFISNp}6P1bqVg#vGbAunZ&|_(&*?{mp{7kkU&$$|WU601w+8VOS6Xwm`
zBus@ygFT+Z+5|y`=jPs#Sv$)TUV2*m=w|&JA_qa<2w%Jk$~1?3f@s)Ix$*&#U}fP#
zVtIm@$J>mZ4LJhnNrUgpsQ>@eV#J7TY&GMt3*GwXKGm6@wzbvejm2Q2ZhyWx|1XQH
z4v6YVf95gZkx9N%PB+`Pshi{!iGD6>uPBV0DI8kk$D2E+W&pvtIuHT3uu)WihzWBJ
z{>>vJYlZH56mzmgNFL#z3TAvNdS5=U>Dv6?g!OUPJafiru4(g}u_8P@^1m`!^27|=
zTtnp%A(7|O2TTV$%4xP(UY9FggL;~r*lTS3MzYsQG-v<SpgYHm$#mT5Kh(d6o;ls@
z;WAd6|M-Dlx~OpnYF>bYkX~sp=9FXW<RHHszY^*g+o}VB<O&p)_zi9}FRT6XTP4Q}
z5;T|^rmd}mC(CS(gjUkokhEPs6qKe^Q$`w+^yAD_OTchs<BjO$XOvXsi|PNMunz|0
z6XhzS1w3K-omO_&3`chk)B*(Hqg;H#1y^EUIiq|J(?zCk`LXu7%#D$ovRTQUFD*7e
z?UkdaJ~fk$i0@nG8A=-h;3I{;0km=r4j-O#Asco-7w>L|saCeK{1i)cECl|$&7y!A
z5GY|P3~nCSVk*$?KAJabXw-G-8>h2dG)jj`Nf<Xjlg&K@i+G^!wJF;iFNrHQ=LPd?
z*6%${wO=$>WL^x|-Wv87-3n7orc-MIovfzPoFY@QnA*|)ZU^*|-S*dJgU8!JiZ<ts
z@tl1SLPVMkIO^C}H&f<3B#`)|J%^oPGGv{3OE-FoC0k`BCT)5zayL895l<$ah1f`>
zv4>ZWyk~M(PYt?J$Q9Rt_1wS;Y~`9)1Tt@UpD)#7qg&*PgJwZ@8@vRh3bWTP#8Gn8
zq36%igLu(`Qwv|nq4q;AApSVWe4pYgMMox=?p)vEg(9?2_*sGFomv?#&VOp`|5jWT
z*uUe$+)D#UtVJlO==E}Q`b1C|V{P1sFo)5GIs#Nitis`6fJCa`_tr6Gla@J_KHWrq
z|2AwL(U)Q+^UpepMBG-$-?Kyb_6^+<9MKWko$@?HHOmJwKXLTnq>C|Qo07xm7&CML
zclSV|*b)#OnV#l*!p^W&t~xI4NLx>0LrJ()=};YVfzPE`LE9Vm)WwsRWo(hzjkk1v
zIBjLSabv$72KVhd3A0{T(R{Yq#J-J|zT|K+d$$MS#<hHfrS#bA^X~Q5@IvwKHCrFb
z5tS|KxvNiSbp|ADL{7fVE<=cH!W|d!iul6^4j@HcUS3<Bf>lfzrKM$}lbg#_=3`2<
z;ahcmXs{+xDJq08Q?`0C{fy?bU{>5rKt#rUU<Ogp;#hi{U+N{_uX(3_0!KrySIZU>
zVabccV(J{=Qp|(qQXVFpaml(rL8R7d{P^`erJ^A5PT=d-Y0XB;o}V|J6NuDzTHHp2
zBE>v#32IwC?lQqlyBEB-Mv(A>N9AwBY09)^LzE^!m8jEQXHQnzErXTKB0=8!!10Mz
zI-Q%+2$y<dGR;W*aH0s=qG9H0-woPq-<Pyl?Hiv8JvJc)YS>lc?)5KE65we3g*`?}
zwfUz~0te+%h&u-QxUDRkgZ+Ek8~u>`=hXk&--70U!|NDAfg?rO#J0=!ZAF7A>hMO}
z;xb?^UsFy;#r*Q?gGqv3?GyTiSm2{lY%7uiq%ut$95a%D2}p>5-I?^*?UdNC!WA+a
zI>|WL(YRUb7lV*BYxe4=>(WO~3J3L2Kzx>c(4{n7w%A}6cB1)fez4l8Rq1>#dK^kA
zn+w6~MC^V1oL%Ic&Fh*+BI+*0kyAJ;Z5%FPbdo(@BY=fC^A_7kN1>WQ^MW`DFz@Wt
z)RsRvWQhEZE+Nn`gQhFgN!*3Z|5>c>A*~Q_!yzB%LbjzsiqJS6E5ZeFF&-<^rn#AH
zWL?#C=Ex1vk+QZbV3|KCLqLg!NMcoKmi>AP-g6vv^>H(ObrvQj_NdME{qb!}&&c_p
ze}CLkzBvO$@7l>bO^!s1T$h33Jghbfi6xD9mR){s$i(Zi9$#U=2tyRD{zUbTA`^KG
z>c1TLz7TM9cnmVbzQSHT64(93ox)vdGfGzvQ9_(#fi-8z)|q=}`mMcQ!HjMuu3B#@
zy!+ZKLpL)Sp~t@~JP+N7YnbFFuVak304y2Y06^bimE}&CC@>-7A62n_BghU<)X!ts
zN-p<@#IW*1#{aN@{@s4Lav((PxHH27e&O{sR#&Bq=%=j@eUayi>aaBGxIzFm_RUHv
zD#-yi_`6SwBJS4&0;;JGZ7poc_)|yLTq5_pROe^}`}xyaA2lm_AdhsIF5@JlCbLyU
zZ-bte9QMPo9f!2P`$!f&cfFL;wwZ`Czp?@49vXtp1)AC|ji1_Agr0IWUE|AZX0`k)
z<FkvCLrp}O%8Eb{wbFqh;x^?27$%Or(<`O`-aAPmjAgnmuEMIMCwy0i;#eJGGnO1~
zpCqt1=&FN+E_JyfrGT}?#U|@!GhPzK8jqP2Q6?n>6*th1onScC64!*4qMSGH3Y(w@
ziF&zF!5P?Uj6zBcJN}V<k9(!oWQmudn+RPBFiD=;z5J_wl&bD^0%^AOE_hsoyS$}d
zpW`w1*`g5=l$7)&QG9PnQ^ZvoQc|K$`;<JjZ&^9Il^VXgi!ha@vm4KnfdAdzgRIGP
zCPa=CIn0I65ZhAxidlE|@~sXnTpx(~6v?K*T4b9;`LN{9j*@0W9c!C2dKC*4wdc7!
z=ETBrno!eHEl%VQv)2>Y%tPF`>YPb+qfqP;L;tmjQK4+a{|<(ULc{XC{Y<MWij}Z)
zfmdBth(Nl<_G4+PbV*i}Ql_o{X+G(o?+TVs#1&(!Lyy@=w^+~cb*p`z`(#~NJjtBr
zr-cpa$sYqu`8p<hKEOKLVS~k<KXSO~#UfBRo@7exSj!wMcMk3eqQjj!Yfmjn@Vf=!
zS6xIHoTbqBdqq#WupK=S)${L3mlIU204`#}(jP$=UIWfn8Ajn9pcCC&=rp$8BM@k5
zA5NvYNP35#%7i?^Y_rhcemgD$>CtQ1Vea#BSpNDQ&F@NBtj68NlD~32YY%F+(DN`$
zg+z{Gm<X!2G+#{KwVyxrOXj)B!<OYP9?9&^V+^CA+d#K$pAXv**8A^<zs$PalXt$$
z_pJ@)DW?4tBYHDn;d_jZfuH2*+d53>fntrUpCTgKo!~F*nJqKb6}xzQc!4WqNxV9r
zTkYae_o~|-KS)S2dGqDZSQ1H=4XI&Jbq4qaVrnkL0ll$pZC2XOCl<qZl7?a-lP-z-
zCO#ocZ590PemIJ69WmuyAKk6JH}#c4&Atw+qh7>7@obdURa<H3S<>}YII)3LnK}&u
zG+(3zL-DGbfbZnBxYYd)t0LxWrlh<3w?b%t*w~Ttb$?J|v`0&0;e(UPANiJpYSduB
zy-6NN==}EoG;5Qmq8_7Zna3S%VHb#+2h1+bhbM{|n@lx{@ygz40xib>U9s~(cJ^j_
zN?(`7%WZn`Pyb?28X^n&lD4Qu#<50*8ECY$#M|ND%_nK_TcXM+JA`ksMMa5F0Inw8
z1`5>l1oXqnXd6-Eq$TL~p-3}>4T}u{o$30CR7tZ|Wj1y;1D%_gHS&Lp`U(bPnqy}2
zl%5Vzq$+-f=3AccFzK$^UTpP4+po$jyExw3J+uciP0KIt^tvyGJ7f+eOS0(L8R4#V
zVSe5DDTJ2E64BKmD^$sT9lr$#FSpLRx0<FvCLQFVxWqNKZU&vqNDH7kq>K#w=9wY)
z@rA__*8_A9&%s}{4UPt(WG`acc0w3=4!E+wd#k|;a9`I`>7@3r=B1f(_Q=XY5S({S
z(gq*@iNa^OB$yzogoKxGeUM467dnHddJEg!ssBuV@ec_Z3JFRG>qo*MHYlND3p+Pw
z6S>=BD1nvSbRQA73tkfuWq5vfBm8JAjAZ90sm;G7NGQtwWa<7ynL1Y%v_I>{Tv+sb
zUJH4HAKbN`YI+(kRc{M#udp2%xc8Fe>|s(N-bO;TiLReeboXG}PEH1=&?sK8Rr^|^
z_6Q4w{n0{5cfm~iYNB4WQEn`c0Z_BR#tsi4cOh_W+)t(IVwV~EE{}?N!C0cm&h`q0
z;M_38$&~!Bo@bO=kNg?9_6weMm;Tp6H_g^#>MNy_Iu>aW+nlub5=GM?lv^h>oIfWu
zBjpV;KsQtkrv0c5v#)$v8AE$Geau&<neOMNgfcug&&7uFL3s&%Xq-%?yVrO@`M>6L
z9suI5Q~cJDh7e|ONvlee)`*XvJXk&&2g7Oo#XBf`=wKD_`Iq~t4ks-B^t9Qw3ev2f
z&fR^TDiYNtPl=qO)wUz7c?hfo6_?A~f7U|)2IXw_(wUCr8~&^XKYbj(32I^$(-8@-
z`Ug(t#}5HlY4E0hVF45At;jtREdYqWxt|I$ZPcyg1sLpUHh6|&Q+e7Vmk2&r@`v@I
zt9HN)pNE1vDYg9P9dm0YIK2Nwn#}^1iwn8_HkJw0+*GDWfUy(?P$u>OjK6F#8B$az
z9HlJD>2s%-|N3m{DYuxu%bMgGr<%Gl87ul0Lbv2lc5I4Bcc5-H`i*R3a1E)>d|vvL
zdIP2mcgu(zvW^;#USrZ<kU+$%VPrHd4d}US3w|%a!omHIk!g%nXn8p;U~Q}NIDK7G
zOhLb6X#70ei@hL%&vyF)dsI#*#suk0(W~M3h)g;$-)#RSW}oXToaG<<x^usqlJ%85
z7<~AXbY&|=*pPu}A}y6#^P=iPDSH~3kI=<s2oom+rRGXB2UH)dsTSpTS<_!>$rp2n
z5%qNCB_n@LqW+kK!y<1fNunR;z+dB*+I<-L#vb|p%wxnj{?o@-RrH$nN2V2q(q0Ge
z3HPKIbKSPq6c5Em;bmh|y?1|(JM?Qh7E`CGoajvhbmwEp=JF|PgXxto)yIYa|MkG&
zNLfw+2(BD>z6^DDcJq<2m0O<iF9xbL3F^u6h?>gWkKbbHK?NqF1D4lx9FBX2uxs>Q
zEMezJh%iK}nW4<+mLdNzWx4GJ4o>dlMVA*LZy&Hyeza`@%4IqTNP){HMv^Mr_szDd
zRN9{Jo7;RA(Vn2hOi;@TFQ4-=K3bcyH$ZJZ-~6XjDN@TXxvvk(!-$GlKDnlf+L&NJ
z9sxEDT<ufF8%rm(+<z;Q_!n5FXFC1m_qE#=`JnPGo9A=rf3>&Rrr^011<><i%dCUB
zvnC0Vz*c5IUP}vdcl9J&ar<)*z%^wJW>8j@{18tzTUY&_$bLO$W+U?Qb`M0~xCONt
zDd;B<e3+hAGnvg?6#1TxrV{Fs?LsXdkt-|e@nw*mq{bgb<s4lr@Wi6Q90nQH96o9|
zkQOB&RE-4%kS#4Ox7cc=M+{$eb`b)rgHCq@o$(w_S>q@V{dKAzo`)^4u%N=GUVlq+
zmW<HVei#%K!o`Z=+a-j$E<2yz_X{+@`&%Ov#!E3;17%#SAS6BP!-u)$7#}y7_rA*#
z4=?bj5f=(7iNIa0vGpvVDq6=B(3JIOS3Ga~$-LSIrqeN=TI%!SK22%AuMOnvVWnOk
z82Cq3=<;LX^V+^{vlUA4(_&%A%heQy#3mnVv0Ai5u^1mQOF1#HOD0C;WMdncV(nLQ
zcjruz3$Lnb-zie1Bs5;PzLoTU{GmEKGL2C1w2UiqHLTYG|GhzlV&6PM*>GHv51Pq=
z68*PtmYSi(H^K2b>*s8)3Jk&hP2W(pp;VMt6|gQkHtOo8<k#NRn{w<Z6Je=Rme@$t
zy2xXH8?abASCb&;h(6OCG0OChD4gV^i;o2#%AvrOTb-msOLNF`N+d#L{r^b9HEBs%
z=zgXMhwVF+Puf9>PmZ~WP)dFP77fu`o=ECd9}ka3x++uFb4uI-h_}9+LmrEd>+x3n
zCHeX`WAY{Qc#ExckHiP@u!->$d82a#+RnFqpr^PX-K2!3M*<i|bgV*Gb&6{(Y$%mP
zs%4~;o&0A$C*3c%on+SLw;#`ZpNv<k#?`Q@#^!~>UD#XL^ypN-$5MtzRT|)TiI0&|
zI)p<y^^~S531h>~)4U?tP{JB8hVR^d*C~IGFFqtx^$>=<$nTqXOJWB^`Fz7nyH{E6
znHMGqD-62=c4=l{+OE349crIbN@)qKmij0P{9QkwB-|PbRfGLA9g52OQju7)Wg`hA
z7e4GD`d$pO{Prm@P$&cFPB4d!R~$(pa!7mX#%<`Jq<Yg0u_C~v<z}l?dm<HD0nl+}
zBSj5MBjCfg&mY4^N0$7cmY}Y!6>7U82WA+#a#M@_0FNPRCdO~Ek%My$UtRZ9YC&!#
zF{ITN?smv^4FmUco+!#`8Z2@{(p+$>UePjFpf!34td2wj!?^!~2}x*i!w5Lx&(|k^
zk12elczP;oy`?T#<T88m=t;~W?zvO^@kuD(HDw{Ii~pGU;a9uDvo(iL{{v1So}u#T
zjh}~62?>e^N|WB#_f}*s6n1x7VD*~19f@+$u7h%nw@O>usiiagqtdkC9pR<_4c}oo
z*5?f-epr|TqHV%Tqh~}CPY{ODY~z%<wB2$j5joAA@GaMBAf~Zd(c+I^!cSIPhu`E(
z;&=A|>zoY`vHF&!+X0@@y!{A(^-8+yRzo_l4_N4fvBZn=?>5s@`y3H;W>|MNcSw+q
zS*MnqWU-&trRR07A)twSYt3RlKiYhVZ=mW)iK-H*)RlEsTi<ZixLdQ<lV12*9!3_>
zmal0Pei=vs3FR=3LL=LY&#|6c+88fa&ed&qA_2ZLLl#+%96;WN%d9(4`2J#T-iZGg
zP>{oEU2t^R&$8S89I>zY2YdKBum!#J4z%1~T$tETIBenFJ0{KRaq*zquEx1;epQ3p
zCcF`b;SCJU5O@8{*Z1Ld|8a=tFYnc*z?C2Q`KLw()U}gUopNn74j`g<qz?3WSHNu?
zT7I350%bxml};>`*B9X<))fWh?>{ji|Iku@Ng3=V(!p6xt!b<NdAD9(=q4Jx=5A%Z
z)YO<@iQvW}@;5)9e_!ODLGZx6?(r>%h3-zt_2$i$jmF5}*mlF%eG*HK5@f2a8-85E
zj9igD`@J~I!>S->)Pc)oU+Adq7xd;#_xHOQjrKBBj}V>Y_;Z5c_(+JVmCeep?=ex4
z-)y2c%MI<?esTu;K#jLTkdoLZ|HB191I5(GP((=ig>K2pz|$dvNHL8vK>ODt;1W}R
zr5j}|RjI~t5=y48-$}ruRqU>dEUnf*OS)MWA|btLawvJ}rLO)YUAF?eVy_yBZtMt|
z?ldMn_y0`avOiTI26)KgZTNVldz268on{FA5|p-xfNa)GIiDVE=BkpGP&#SYM*80G
zSy?`*l+Y6C<L{izDCcc>kR}2+J<R#Im+;wHZ`MBTo%{S`{t1~+ffW6qhf`{^J`ZIJ
zI#abk=Wb*0Zq1!?eEC>TUd~>xf%8sZ2^vR!u9=5Z^1bsb`Nn!6kx}cZ60_1(@wG_!
zh-DqyZ0+oepgKlLYk34y7pth_^Sd3rs#8YAI6(cQF$<J0I~G`WNL&GB+fc{n7n-?H
zj^ga=T$Zh(a;Dw+?5{4F(=MgtPN+@bp<{9^bG2#eQjP~e?cJX+0>N*-bU3<Gi^hG1
zOz*Z}fiCwvh4w;X6m-29hK~n?@Z$8?(71}61t&iLWS5gu()m0cirkSvG+bcj8z7ON
zf95rFNX#N)O~Y^bp-Lzi<D;+_BvMwA@7wuA2*Kl~IfEmUQ3XOlYsJCz(mH|RDLpZN
z8Bt6M9FXAu>+uP*wGB~=ZfjKMaW=lZDtj?T6A;`odB?FM{8r~UfcVMZ#K5cGaA`h5
z_RmCf13KKW*g;pUM?%8jz&|Z*l!kaysd5Yxmg<ewHrNsrlo-H#%Rw;s76!dIC%3-@
zYypIcX2`^s$d|Dt9#AHf(vs^I|9q0CrM|sHelBS1e1$Ns2iJ}-6PZf3WVsU|>+1^w
z8Sy~i$(Gq?@TI{!MaP>_Y8^}Z&y`NBv%XxzS^jWsF7hk>jBozi2sD$7j9Mh4AdOqj
z<FiN-!46vLImrfRviL+sy<ZDg+mas#21<lhoHg(FlCF8lmL*zVdR<a&q*J^0emX4o
zC1m!$AUXM;u-v^L*|p_aDc&hkK<w=8J5QRX9@g*s#!pU|dDnfYuCK4PK7V%Iiom-%
zbq>xwuJ11(d)b7u0gnOK31DASA9P#YkDgjXd<g&c$a_8@{kMndS144ce8$HdAc|+t
z?Bs%2=dFLAy}>(k=CrlNv37^)JUBYCO<zd%;wBi`1tQzUZqHTF;~713`g62~I*)LW
zGJ8#MaAKZ<d8+7jTSbN}XTY6Y&~DL3?*kHRDWn2vqQ%s731MYpJ><eD9h=&qitQ_G
zpg-LukI!#T-zrvHi>*0mj#Kq)h+2PdjIA}|?z5ICv-S7#qczxYn(RO)R5KenqHvV6
z(6HK0DWb3822|fqa>}2E?3zmS_E2ZHC>l9=1ZZ*Wx4(;s%6d^N_6Waq5#z*+;B3Z>
za)cxvH&!;sWGK7d#ouwJG!snuy!`TYeA(IjcrM%@_tY?R9(3;;6yTCP*+uPq`faDJ
zSJDSs#hYMB-Wyy#DN{=KSh6ZImd-h4u+#%HonBL?H@bYA>zc<r{h1`-w$h-3(wG09
zv=0p>f%}vFVg8x5vL>2!Yr(W(P~v_$>Q!Z&sQfvI>@Wa)t?gXWYz6cFav++<$V^V0
z)SdVT5)0;;u@lJ+_IsG>@w_!?41)6O$+*9{l1?wTRm9X+Otq4rBCaJToR_zD`$0>!
z)1y!zDxhez0Fsu)TMm`sNgMJcL41UM4CqV8z8+`7a|3L8BC=SsUVPJM$H@S%2U9)m
z%J{EF^qaUEQoHEE*R9sN%b5Fl31QTX)zKqIE|pS?k~EJM$4{x4Rd090TWpufQzLT}
zr2byGtBaKix_R9qkghsd4HcB`{Zu(*z|v*G{U@8*X9W!`AQEH~$DbH*gt-yo`WT~D
z;9vz+tSB)LyC<v6{6{Q1E^rz8zaJGWK<Z())hp~Q)>^2G%;PArkF%!Q>gaPJ{yDU<
z*XWiqPD*P<VJQQ?nfd3Gy&o7{j+~#*jtKMP+0k)sA(Rwley&Jr@p`se>epHxqPs9H
zm>zqRSA^QuMoqFh)f&@2h>$L8v--VY_gJ*j!Ux{(7u~_{k15$j_BU)xM%uE>xKyN3
z!v268QGq`xkga$zh^8PxL5@jrT|R~1yCb80L#;*V;Dp+D0b@DP<}i4wpRlegTUCWq
zQI`Wsn2Fu8FbupU{8by4!;rLv5h7JoxHGUkEE%skJ}Q`n(6pOPV%XN4`~J4#ikdW(
z*|y|?Yr-4EK)t**9M#AimlSkU5}%N8i3T+C!2kRlF=ziD5j{Z97E^%y_<`+ze^%Dl
zM>_7Ra{t3bIC475xs#X+og!g5gbSQ5r@GBYD945jH;rxsoC<C+u?aQavN9(6QisGd
z5(y-EJDmves?u==@hYO(R)Z?}JFO~Td%`ZdIqD7>(rIjXA#_X~32A?t%0_I+qyqZp
z9?2fug6=IWxT2EOth`*I+g87lJF6V3ocT#4M2d0Vs%X3J%igy~pE6QyOkwkH3ofwm
zD5ZN8WpH5B0`h*f0g~)dPNSQH<LY8+Of)n!F+{5-9yC%SS$FrHD@?QysPJ$xIdNS)
zT?zA?(8HeLmmy;RHD+##59k$wax%4gQw5cOZ1O{A`<|UUy3dXM#{DZ>Qnd&X_D=Lj
z1GdcC+J0j@WPv(j`z}S47I(IB1PSW%Kt5L4WTb)7o(aMuo=6|Wf5dX|^FpY`Ne8El
z#MWJtJK&4io7iI0+A;X$b#9MndiQ2LUW`F+?i5lCIlEc#H`4li#A7XnT!u}Z$SKL=
zJ3sh1k@$JMc`+5+IGU4bOd4E?vmAH;{dk#Ue$7U<3FEwKRt8ejuJG_=ngtNR!KY&T
z*5!iGoyJ#bwvtOJ<Y}5s6^6vMM5n7xX0=%UIec1xIA?@wSU2-?mx^NVET|E=RCp9=
z{3%DCpPK2b!%_U=%G~It1P*?cT;uU_M`8l{rhEwPM?yldWhnX%5)4Rv6N-A6U`WZ$
zTxfjq&h9P=Jg~4(xx2tY{$I}~7l)vt2wfv#mCCZ?u}=8$kjZ<if&pM;oM^^ZGl>&9
z9G4^|@h2~FoSg~Nu%#@F8KZQcqOJ$idzWyp2{#A%`|R*d%avvywPMB33zHW(af_0j
z*r}iFd;<8p7a{1Xy{4|4VkE#eXvv6)e?M>g^ySXW#fY#6W4(=C;j`t3t%k(9;jQ6|
ztl+W*`_=Xj2?+<heYrTTF{5rVy}fM0(?f|<F~)baiF>QM&GSPp${z(rzJOBq^yQVg
zeM+#H(djt%O^*;VH=^5+c+jaHn<AAklxf)s#%Fp9=uTQMff`p;UEO*6%b;@)Y>!V&
zT-x1DWVEEnTlg9dXm0-3Crt_-@VeP?>(uc38{aE_H02Jp04@AA%;g_6;Tq0|A>x$h
zab5iB(BuQ$s*?hyhr@mni$6FC6Aje!*h4~yr~5ja-z8cohK7`2RX>pErCZituSF<)
zlm9Rm=t*<Qb?e7)q4~nVSbVoe7)eS{>h>9v>%sv|x$7VihHq0Y97;p3wEzn2luSCU
z#+O3V#dE~O5nhrakUlg<qlFb(1JomBvqABqN_MSj$rJu850lKtMr;%!BK!^E5YnFv
zd@$Sdj50||YOQTgw}M<>Xw3u+Ht+VM(B-%&vfGVr)PRAJy(uxOj0s_G5q)<m>87!f
zVQ@2kN|wc8iztb#Jm~X>sKodp#rl&k1+X7mKDR6gx`w_BJOt`jfjh4!lJrcL!xB+#
zO8*Gy2})98T5)izv?Ui;R>jc4k?!W(uS4$d-=homP{OC-Gqm##I|+goywToUpGAF-
zSK3ot5j%#gL^$J4SMb4uKke5C+wTLjjzFJ!-6gXtgjMe+b&Z|Wmu~UO#$T|r>W;I^
zL+Fs;k%utSvh_AE_vz@duN>Ap5O1j4^5H7+$~v1tJ-fh!d-*ym-0z*m8uuas3VNDy
z&7-lzf4%fHqO<Be2|gn@3*2Fg$sUBOKLR3B=m-kFbFhJTJNy3Jw0w4byQ}>^6{&GJ
z>W2JBj9hMZwt=mX!U62Cp(d}XNQ7ieB4P8pO%r<?up0*WP#r)ReD1-}z$NkL*tQO|
zt}j)(1?Y<>BIy>&is^?AFuED|by0sA`OnY2#s2FVAQ)IwO<VU5J*n?c0=*I$#yNZ8
z_B|VA@Z{!47<H!)hK>dq^-%-WieLIY?x;M54f$x*;ajbAH2k{WKHL6CHonI2^vuvs
zDW+v}bJfDu5HzT{%C5u={;V4<Pa?I^b?`wdyOaO(_C?@%hQpdJkB(7P2GgKtrT?}(
zi&M_9|Eho+Kh1g~SyW}^c2hBW@mSgY4&p4CmDQp_v~eTW(O-|^5QMF(h@<0qweHDS
z!mkpNEX;W=Yvx?I!yaghybhA9Q-$p@Tdo`<(WAzPG5nq{8gfltT>?%$mol)hu-~ey
z{|+Isg3QdI7--6E-44sLpc^d4QIC#}6v~DG^Kx3un2%W|LHqy+6dN16*h5DriA$0r
z%U9(-ZQ`KqC0!M+D{YkRF4_<pF!pzh)tMR?GBZ^jh8nco>h`Z>q&^1So_lYukmeG7
zveVzr`z;t_(OuJDL!j)2m<CFfvw%O;hd277Wm^HR{jBR6ml<bpQA)moE2*p2D-r`Q
ze>79411v;Pm<T9;k$&LPuQv>V7heSUz@avZ_!!*Z1iE8SY%-6okJ%WN0#^6!Otr!I
znesIhk^8=&s72s1lw(RPKg7?3^1;XWP$~Hn%qUe>>gSI6iQ1;GC}d45q%BUL&EnUw
zj{~{gL65gPr|-D*KeY&PV<7C=u>RnEKab6)0)6vx8Aq-C)H+573L7~a51lB$T$_m?
z=hH>;wf0xnr>RkCk2qDMJ8L1~2johJOjY9tdC;IFr-y`0_=+6&`MR4hjNo`-&m-1v
zmX=?2<mS3k-{Jece)<3Ad?C*AF#f)`hw}fC^_D?#ZQI&#LhwK!1b4UK?hu^D-QC^Y
zJwR~x;Lx}`1PJa9!6CT2)8E?X+<mLw`^~DZMHe*dPtP^S7>{MmI&n<K@J$Z7zcCZ2
z>bg+yPO%5eci$?jq$TV4FSJ`$+wA|jRcB4k*c4guBPismv}{u!xhhOyY#1?(Hr0$W
z#fCB2qm<g!;aOSvE}DPhXY}JJt-npR+@Vl!go39kxY)9&p@MPfg>C67$N0^uONZ9~
zA*p1ngXJ?t$S*hYElHR>2lhT^PrEhOq;j{pYhF&9&EdXTz9#>i=NvprRgWJZ8J8j{
zzevY~;;Ik2;*+X{{b{3IcT$AsUmX!D4vd4Ir1nE=16^^)1Ru9>H3e54dd&1S+3F1T
zC@U%=eaM$__}MLF-YG|oW^BR`QWX64JtjpyBg}_?%R{{k)<)b>mBX`1JP}#halE$9
zfd4Fr<H^!DRePe39|nQ;%dV<TDW?N&p}a@|NHdgbtlX*H|8la7Oay6BX&$h}6M%TO
zK>SQ=LKOC_48Ku~glHt0v}S=*(?-jfUm1Z0_@E8!7Z$RhhUgVi<`{iw35qQ%6O-Z)
zte!C`OLB4y4Wb=9Xv0+kgJLi+08W^SYPLx(8!{X6R_pH^ueguv3j-GpS;JE7cj1Dh
z9zV6~%*N|(b+DRS0R;{j2K`9X%h*9+71=w*bgowUgQL6^pQKL6t?XUZ<2&sHDx!h2
z?x^ZDM`$o(LCKc6?akaZaK440HZ(L$&CDoyRRRElOJ99!D-H>CbaeE)3f=ZQ(?Omc
z)BtG1Z!6sFR^4o<N<S@w!g8!a;y^-MwMNR(w7QtJWu=J9-A!nLztJeS3Ij!Fx*9RD
zT2;zB&f3q)CSyvAx6^;Vd4!NFVZA5*gQh$jhxlG;utr9Q1VL3BzO0NTq>Xq64GsnR
z=mU+OA@O^%ybn-oY|K2eDvWjcT1ilxXQ)<)#v#bb*x$$&%0e>V6{3TYijuexVm-Xk
z^Q46a!3*(FHz87dhIE9L>JC|dqx|Ac={H{9LubJ5Hw}EIRfiRV%hCT<_n_RF>AJZ9
z_vv%Qua#z(&&)0ha|4>{r?LCBOGCDgIws3=@@wo2(-GRt3}-&hwI@ZQi1OL~Btl*`
zb}ISmAZ#7%E$`S@EY9xd^Ef3X9FUXLV5lU?XEd-d0>hC<VgfZ<Q&W!iAG|<voHA(g
z<KqR=qlNA4nND6;I{D<sVV3{cw*tO-Lvv}^xJFhClaa-^_(n)b^3$E9a-``d<Wxgg
zSARcpQ$$4?AyFgzOib3fnKkw6ZSoO#yLK|93<2a*O5BnGj}<=)tiPOBR~vhH1&a8x
z^0^pf7#F~KEEav`GWSKf7vS&dlgA%$%r)xEuv49xlL**v6AH63jyXDlu2P$H0Ann=
z4#$z@i@j5gnQ5{G5Ix_|y&L-M_ET&GYe_0BVm#R;tst~0AgP)mvh3I*<H`G4S~lm>
za{Iq~A|QbE-Js)yq}1JQwSV>?k^c+w+Mc<<b-GXz4D>%M3xxE6nO(~0h1$@W;+2Yx
zhu3d{IqGh>taSUpBJ1^eYk6M<rMi9+*34|vT0c0Cgrcv(D2kM3US3uiZ4Aq_sdB&t
z|MsFoEFSaI*Lqdz&6Xe)5JGKwy4W`k_M#Db2D;KYb=sn=tR&j3h4oM4<ABY3jcI*h
zJD!%dz8R#geQVuYDhA(VG_0P>Z`Hs@2ccQtyLz(+?(fW8%Fa)!e6|WNf?$zh0-f7E
zh}k*6`{lCWs>tMm?M=gCe8qP|FOI`y#~7mJ(ZuJNM8|@gRTqr%$$)$V^A2+FGW$D5
zdBZH5&fW7@K2D+3$(`u-^dmf$z_^h)kSqeUJoS9R$0CCyf5q#W5QE*vXu7Vy_h)Dn
z!`e4pC3btlIL;)<kkABhyb`IxzND<QewMeRE)BHSvg19Qkd?GpENaw<6df%KD!gvZ
z(G^3;R<$DAgQ=VjylRsov4*2!Ij9|BzTD>WpwzrtjmZ;z@(Y^oQ~Ixd0ZG4BF`AqH
zSsfn<)c_qnbz)6zto^8SULpzM8I@LM$Q&Z1@2OC-WBW2pdDUGiAT9$^_>j*7scZ4i
zi7Rqm)tMgZUVSj_ZBxynDY6!Rttw+B`Sk@V`sx*3ItC|5-8W_e`sI4-BfWmiBPoc}
zbv|}1ATaOq-d?@U%2cnj#Gs>YY_|5wP!y$(ywOS%NTb!Rj3||{XGE~i=Ww^A_r)oA
z^ts^C6nAv=^V4ziq%J8=E{d{hZO%9WPbNpn0k6~d_FS$BbvvTJ1XvP}jn>U*(P9t7
zzLRLvvH8im;ZT9-gtvR}ugZkCs<1oh<cB`E@A?!-$G#r3s;z&moTD0do(hdIl*Bcv
zmx+$v_6l++40}uGI)IvVtmSW35ue#OzhX&x|L=-&ad&s#O;%^&=YI{eb)E;SR(mXy
zNyNNM>UwaYy@afBv9n`H*8TSzuPjUX0q#@H<B08#iA5%+ceb|Gj&zHmCAkCaeOXLx
zTpp6fCZZ8w@3YSp+Efv9^F7tCxITNVAP<Z$S_l2l3;jvhVYSplBmOBR!$!=dl`ty<
z^q{(QoU}BBm8=Y<?q`}h8#R6%tc8j|_s);&H<ZH^Smj%K(}Iq;c!qtjKltURJsq+5
z`Pq?B%54dFpf{d#HW&Rn#~ee1HME;M20lC~rQLL#@Az;zv+8@>e1AnbH-hw0Y0AjY
zIKL8Iz}c61XE#Isa1~Ib=bAa-n@iafLiSF-ZkY^!Uj%tZ)N&>`@$Z1-Tj5CJnnaI?
z+4SR*-wGDA?gq(Tn@3QBsSg{KLo`^bo$?x05FR;KM(^D%rMUF*rI1C_@oDH)y_A|H
z2DZDBJ9V|J9KF$GYE2jfi9m7OIlrCx3yYPE^VvRYW<*gxyDmegb`&F&vZa>aZaZFA
z56`fljrP{zB)=s5y<HyMWu@u<gf&?XI-yad`b<nfp%98@YRIpXQJxEhJ~~L3m=F39
zF;l&}c+QwYsQvr=RF$F32Pv<>lBYL)Zz|p}Wa_}~3*W^YRQ7nX&01S_!}r8t8GvP#
zQV8Dc-$nNwNwf|#OcK}_Nxg<iU>xC_{93g7f(IQ%fn{qpi(7?RP*$60Yp*5izNozF
zw37I8@ummZIJDkf*&cAIWH6o$(RA{uZb5a5@I)X+ReW{OMUho@28b}yd)w!^jIivo
zBBq%9V3AoFIT<Dam}B(;ZXO&lFT#wec)+Oe>$v{nEwLCT>VMY-Ys(S1T<#v-k1}5J
zZ<2JK4F{f2EA4;!>&@uR)_nsYE#AK5ZOB-Z(|&)rps`RXVZJ=M#H;_W6EL)8fc=1R
zyxhF;vsoRX+y?W^|D9b6VB7%H@w|31rwsT~gR}B;QD@mqYg}|Y_dc?&sNJI4YV*JB
zl6|HauG2heG4UfD7&Nqz;V&*uL9Fxfb8M{k5vzL*TJ%`0YX5Y((k<(>l6jz~D{EhE
zU?EvFf5HhS4p{D;fZV#<v|iB=`ud=R!%?Nl7QO#c4$g_GqW@<^oYThHvbZ|xATG@k
zAe{FL3m9Qlb-#HL0{T>-nlHwpAT_Xb8uRz}WHEa5V!VC6auiBVY1z;5{s-=r@3-rM
z!nD^SYH<!}w6`=k=a=AmzB)_5(o7QXPcZiDmC-~DV0k5uqBzA&lr6EEsNU<mQZGs(
z&STR5HSQ%Ta;o0I@gqrWSgA^d4}z;8c3*fCC%g%f$26A6mqE*>?aVjZx_tdmKv#G@
zwreZk(L+m%q=E}NCFw$=HH(NGrU9jMYM_6!Q$z>PWN0m8TSe*B5}R1+$G7yi^09J>
z`SJ-%sA6l5p?1|u!tTp6W#{(aKx4!s{-OO7RmQAIn|p8`MP!N@WTZK@?RgtHC_#q-
z?k;`ERV?Z<{*&LqjP!y0pe@i89+UFk`KNnWa;2kJc-`2<a>s3l_?0;8vQCT*$8*9i
zdf*He^2;PhFn+iDEQ-`$ZugLR@*1=wK^mVx?yQS7TE`??uj0p?Im*n<A3KGoQF~{#
z?rxjL=zBW&a_!GIkqa@zTkO8BL?ZCSlxK+KrJ;}$2&Y6uzv9S2+VZSRL;G-hPbYmf
z>AFTC#T5~*>oDySElMqBW|pu%4YglYX<)$d<0~?q^!VguN|iPP-2c|d|JKSA@1CBw
ze!ai-xo7=NYRFO96)>1v{p*^XpTF86Yd;rGcd)#|=+}YChF_FOxQbZ+p(RCGp~U6e
zml`+vmmR_05<z6aw^|m<oMsJBtxlR{MuJgO=Ta_RuKF;GZupUvhQ_a|q#v<;5nqm&
zIf|h*Im*>)ZP|6mp}JpJ1jC6xun+2dvNrqCG;e))P;+n^I;}XRKkgn;T%3kIrrcVE
zGwNllEahgq3$Cindem<WQhsJHUkn5u;7rWSz^JBh)-r~Gci`fJ=^Y#j;qPlcxAXOJ
z+vc@_mlr+&y}%dr%K~beEFFfy(`LY~9Yx^#ahr^+?B%$<U*cN3J6Vy5kx|gO&pt0c
zw+}Q8MffLzZNO*-v!I}0scF0?7;gSqB8G{DrI#$0uQyL5T<-WX4R5mUSul+2YLl3R
zH8M541aggi@8iiRH$ujmzCY2uF<8HG-cKV*2w7(?3<>Rl>XR^csAGurJ*AG_@Pe=2
ztmCtF5m%Jak>wZcTg{Pj;A8u6q8Yv}@ho-A7yy^v1p`Y&@JrKSSZ-f3b~o;q4}@<f
z*+je!iEmEc5FGH8CTcFbw@lh}|D2aa3~Z2CN22WDQQ?j+FG+ZUXKEoZYBI4Eg@V6X
zMV7dvwExQ_gGfMdY^%)k>?m{V{7mqdWLrSgsQ)R^)?WKPYlb6`*}dVj$EE*z_<5$v
zgiU$#8iin6B79&-YPjHa_T^y+9&%+tM0Usrc^jh^^swuYsp_niG+4~j^)bHZ9OC^H
zW;ny^p49y`Z0^|ram068UG!$u_t-QGtM9v!yQ}rKV0<UY|AY-QX?lJj@b*;Iy|X#?
zIm>8+<1dcS>l=~}$*wfbpktg;$_O3cjWY$UwnR<{ax~a@U#z6bQCYD${Gnj8SYuu>
z)_N{+bXR&8e)Kv1@wKTiNgBiexO^F67cKf){=Gk-Jr~)d@_1!S&6T<C&k!tTTy`1%
z8P}Yik@ffY1=rU_+pUpo{vWdA*E{WmrW^zms<j{T(#80}F=DH<Oe|XpIJ4A0m4`4u
z*R)vvZ?#v6rWsRHE(9W2b0K@*BM`!hZ6`GG%snVU97G}_r7TJPzVzoMjyy9x|JES?
zdu0qh2784B2Rl448BG#e4cyJD<r<q5%2;)xEA{pSJG^i`it>ja?j8DK!hNDuWEZG<
z3VfvalsdIuCAzE+SRN-P(ZwfO84fCvRwyVa(PgU2_G4liI4R+@kV(wlQBe*R^2ZP;
zcjs{Bb*!9Jpt<-3jj?TQXJ2$fUiSIyGGMzQ&)+YZt$yJAS;;GtdgQrIhHb{g+{D}X
z`IPR>6x$h)`t@R{AgE8mZ=yB3C+kwGZeT#1ZZyBH4vhqQFqViaU1nx=bqx6S0||@3
zW^sIb+;<+OZ@y@1?u_O=YYE1)YY8NJy9fuWE>>37KEMs0%k2bvrP+S}V7dSa1*P74
zk%6K8gypOqBCv<G$L(=J{Le4fz0Y?ezgBJLc%N;x9artd01)BRadipcHCS9+JnO}0
zgh6t&g2q-^wv$)gA*3sfW<O(cyXU6o;gdEj!tFJoIsAJGI&pB1`(e#+!rvV0@C(BZ
zYdwCKbJ&1tGP;uIC!jM$R&UO3Wj}(Jr1{P+8n1}LyGyFX;7mrN#hWP^JsBT`Lx&Gi
zZm!LV_>d@S-8+;-zJN+GpGu<inatHC<!N44<}^G1aa(+<pGKaz?8-Cf?ucu@vP=uS
zkN+<hV86;&q>vyM?iq7wT6L7nS96N4>upxR?>XW2geb#k@_<j!&oHub{BsV06l=|9
zt*<RF<mte%>Y&UVVv<tc<l^YzlCo0<HSCdO)fp5_Mss$YxhcCw&*$oc=xGxPS!EqV
zHfL8(X7hYFfL=ppR-31c-@eH{OTNtx*~Sz({Dv>>1nGHhHCiz}5xm1so`!bM^+>bk
z?8qkKzb5;G%Y!{~<Lowyc}C<Buu!b#=X4;q?sm58bDVyDF&qVn`%dVaE@0D*o%E4X
z|HtFxd%96lBb>Oi-53dZ)E+a{FDFlyDH&jD<lD-YULb*bhf7W>?$LL6<V(|T8Izz?
zQ^(j(#+i{>)i-?OT5Ld2(on@X@v{56Zs2z~1QZItAIResSW`P5b<Vs`G;{uV|Hxj+
z?U&7oR#%fz1}vGw*TRciOtdj1_DLVqtvt6!H*y#c-to~)UzSD}#f4hlWe?TVG57u!
z{^3FrS$s(Gc+di#EAjV9C6Aw6Lh1dtYyI~MF$@s~iATqP_ftY^=)9V=PhzW0Cv*Ps
zdx`m&LgQ$QVt8O7_bsBLKq4L{bFBCy7mH%w+lQcaz1`{&3e~)ytLk|*HqX=4Ivyl1
zryI0}Za1YYT2<{2MZSx=Y&OEB-`H#w)5gcQ<`?P(eQDa=H1P$ifC@%TTrkN@6J)Y>
zK`6!c<V?s(VsLF1m!NR|(PfURv^o~BoY~KLM(l$fjOk&W{K?4{@LqV4>DKP*F_AuN
zYslBV*8hxsLfzW%m|fe_GrGK6FY7Xajxg338+OY9TjP(i1boltCh^+ZT0^x3BSQVc
zf|^=|Mo((j4Fz!E0aHmASJzs*H6FXxUCcj!{v4c~n2cwzH#9dp&%)Vn->um1FVz}n
zaXCWu{gZ17d|FFO>p!Nl0I*IPFPT*(@;>16d<_@|BifKWK~yBfI6TMrywkJl=>e0r
zS^VK`1Cj(5#VBuL-;pAq38mF2LC|~Hf)Cf0``L<ln@z%@)4APnHaS`V63f7Vt-29X
z_N^pECV3(}qfo@7YHbP=R+)h)5ZziIsSXRYYW!m4)FWQYrR0ql+k@7d5Q#d>0(a@`
z9w;ebEodrYsQLaa+z^x#LB!S?`dlY7Q5LH}1NLxnNJyh-bulH-_p8X=7gstt$Lp!1
z%K&fFjS=qK+vRS%Z`4Ju`5M0*YA-6`F%Ka;K!~^}@&fz*h6K_9WVOHfF|@Es7b@on
z8QVO!%xi++5|?iM$R|m%7>7IqQi_V|^F7G(mVSL%WqA45YSOIv#~0vyQg7<~n0E-a
zDu$+UxkDD1+@Z?Zo?2oqKC#3^q#=#phFE7kDSLu0Mq#1A4W!aV(f>dd5k**Be^v%B
zWWosWte9L9(_8Vi)#lO4?MQMN0l6~BsDck?H&yl~S35-&r0}>x!}wJ@hr=0Y-)6>{
zkmyFWSh^U&FTT@k(LMsP$890PBrbgh9zGeO+VrV*+T)e&cT9Y%^1c<?rpu|sLW`23
z(CG+(i!eNiwC~i2vZ)`9TJz?n!K_Vm4HN62aM!OU*Li#*<l0TKc(lU(e2mz@s$+QB
zuLSP?{;s<BjLrXUpZ~qij%RKlVd+BOua}c5^n06!9oJo+csdar*C=D)a~U}<9HA#p
z%A+3YecN8LJ2?3*x4=wq#l>OZ^icD&+P;ifTO&Y5Jeda{J4DyL6ci<Wa27US4upHZ
z-N>tgZO!)nGINxjN$uP-xatm>r9YoPKctpi;qzhlHKk(qqX|EDSIIq$>4U2L2Fp(x
z@074%vs^J(olzQF-)G>eE2~p!)bC+b9D6$KXVW66Qwwn+U2+{asa&s0(zCKEsH%4L
zB6**6M4zK{476qd8|qn2>t?a#?f5qi{cuX^uLLDD#L%?tY}3u(?+#ruZjTq8UmhF*
z<{R)PBLg=qEbQIE;o(+-Gy_R8=yQZ9Ha7O=m)r;FcDGX_7ndYplu+ZiBW~CI>c+#%
z3vh0~@VcLw045Nkr145+%-toiNoXrgwR+Yh6S-LLB?E8xPO?5i?F}e*9ImZlXAH0D
zFOP<eYI#ET8H^xRhM)LcpbT5aP<FYV=BOj)aZl|*z&OH9)efnhZg9(fbL<=Zyaxzs
z<}$8L2+`pRMoqslvX*Kvots%CpF>z#9@$rrN3PEkd8v%<@sN+@7!9QLyFFpfro!xO
z35k@HIwN&kTyFOhx<4pF$jE{zcbqkC#L<Xn{4NIi<1&=*HTT_*$-26#gv&3p7Q$0%
z1bM|ABTY?KIeaW*zkcJVR$1p4kPOG(BiFSF^wb`*!4brk=GC_{k40-Q*7`eSgn6Wy
zhdpB|8!PFydzN?*To06(&D}xc@XE^^5knser)1i<q%-p0Q26m&kZbccIrYT=;Utik
zFMjv=knq|$HVN}*>`9rLVm}J|Bc5D+{(bQ?0SA=iJP`XHV=3|1LC6mpdzLe*Uox}5
z&~S3r-O&RLMvq{h21zm<%gVvz_|>GH2F5FQXpb;a+m(F2Fi8n|8O6yXpPq`aQd2c_
z);Am<t#_r^A<lIj_(lLlgPn8KBR}F}6Z_?9JBrgr;^5=dk>ib#Jd3`^`-8B+Vw!Jz
zUSN5^3gqv#r*Pg>g0JH1iB3kr7nx{zqqGEef<}Do9|oh;Fe_rS&>56*Rg5DeqY@w_
z5sG`A#cz`PSZzM$|AY7aDl@XOip;d{N<Tb2<h03D*Yl)M*S(FJ>LL9v+P;~D8rzqw
z-rI*#zGnzH;`vyzvWm9HeAC7=XXoe*_?b5WWe?kGMvv1OG~Mt`pXlbU_i~33(`5Jc
zmkr)iZ23<O>ING^;a?c)rJQFiAG%&lZ~r(cQify{pn{Y_`+xwg`S$AFFTA~UFYNdN
z6~bmfGqzpPhykb_hS}B2In1i@+WxU4ZS2)ifz#EkHpmVqR8qS7s;_vxcstVtScF@Y
za>o`(e6d;jxvtg>ZR}#VKC-Zv_#>6b46;tbGGD&j#`B1FeX>Wf?kU1uK!H<=*Ll2H
zJ;)~8*4DPv=K6KxVSsbUtUe3KcBs&5jO^+8)^%HHf7yfLhe|_ndV2a!$9>HeXt*}=
z>gp2QJCAieat7&}S9_mskC$tW;Z#*seO{0C_kpdLS5gA_A^XHt`<`0=LvPVY5p%BR
zY0u8i2A+<8R#rTZ%1VGkZqSmKP)S%;J!uefO&xu_qtdK$F?5f@%%8B=*3A_wJYu-G
zahBKhbz<V){->uazq`+wCPTa9het5d+CHcAzIiSc`l9D3Q3V_y#gB=pI#;v^jAm>v
zv!Fk>uBh<`W@$a%aNnD$O0`tSS95U0i3kgBXE=m!Nhrmg#|W|ttb8%lvcGwZY(Mb7
zvgIs(FDVPbascf{w#}bSYhq9qTUS{BP{d>Ab{9*S+<-u-XU`5F(I(34cgAO)^Y#wh
zfvPO!n6fAGKlbo0a9+)q#9Z&Tqg-k8s@K->j9?L+?Mpm=y6{eQQ`QH@vv4uShY1@S
ztqC5#|JJ*on;R=rq*hfKYiq+~ppw(2#vH-B4nuEjI0q5fCifS#B`L2ZN<qoi!oHoB
zG+)KYsw*0+ZmTt4cKhFSdyf7Y^>iNWUv1l0v-98lB=CACUt3$1+1cm)xc&y&%ZRow
zDR-i<2CH9}5&2}GoIJcE`3!X#rAu7hF{vi?l=L?(!Vov453@Jjz=BPTuW)u=tEWmL
zY^ZQ{j-9-=NQ^SKr|EQt%Wqkv+ophV6MxWC5l2igud#!?0Mg+TGwKMOeV|I6tH<;G
zYtN9Wv>aJw)!B)ZGoBnjzc^zFl$xs(Yb-3xO)rKf;~j309d}IllnR8c`!`v@F&Dfn
z8M)G&XmMIrS`VT+Y9+k0Oa6O#EXfrbkg(>H#kB$vko~MPDV~~8L-IYO$7z39Y{vX(
z(BR*FYH4a^1tuR#fn=To;S7X;WI>d8_`freH>v_lkeu&6@5K-qcrkUq%te{908O{A
z4=Cx7k&iw{8>!DR3D^NK_h8t&+p>2{OG{BO={t04@*UR<-|lfAhQfD;z1?-45M*R~
zY@5phZM-k}_2Y2Dz}eW*!oK`AtML1|^4cZCMXBIsxh#e9{cEf5(Rf?B=U&+`>GIC3
zxKmIK4R1df!%CDuF}~;CN7&)QRzYHiRKT}>@{@A271>}6rKM#k*eZZr+y542$*Yce
z(Cu<~ziW7EWb{FwXszd~ZG;;S*A9gyIocNgxhIM+;6QN&3PD|UnU(!z)V4Jv&A1D^
z=sOptrrrseno|7J&Pi8S>$D~azU>GyM6b=xipa~$AKqv4y4N}HPs~JbUr+F&&`@CM
z`!Fkk)*mO_yI|857|5Z8Q#wig+z#6*Qmd0Jpg3MZnYa=GaS2g^>-8qX_b)pzw5TLM
zPlG^q9R`_C)>wuis5;lp+A4?LyLJ^m#?_Z*m*t~{oiDJj?zg2Akzds|iRdFFI_Sfa
zXtPyZ?bucK)l%<TFUOGsN`%FsXuZ%>Y$WOI^3#TR8zoU|oL%XOTkL|t6<)7Tu<P@)
z0qAIfXxREKX8akb#4zU?E8oL}QWG}T7v4khz2^yv)nEG5fR$)~o9lH9YLtdrP?LB}
zwdMq>qrdL%E+l~(b3a?l-ycf7{WTI)ejdnt_kOPb__4P!S446?%1YG~Yo)sTC;G|r
zP-}PD8)c!F&vt(7d!=mJvdfsp#Vg>oojh_g%&K<Qg(gQ??87%H;zfKRS!9-$KCYxs
zK#7hx%gYpf!}ljcduW&6X_>CcHr31Rv2$odxyH}y#AHSqAis?FbU$v5Htr6y_{;de
z>^k2+e{Zb<%p1~jvkht2e?RV5Y%TX0uf9%^Fti&!B4g&=iGc96Z$EuK_f+d*RAeC^
zrx_@@!VZ(-QtPpxjX}rFfX!ftL~~HwmMRnQEbxjsI)GBF2euH`BYt{^oPaX?5k2gG
zkUnVW4x}(c)Cf_zM>kANOu0vn{}~ecy+1fmpuo!Y5RPE(F3j7Zo@E!Tu22`K2C*zG
z7`5{xCXVE0bLA2C_0j$S;O{URbj;_V$L2Oz*iC8up!-fN+OPUX0_r+>O}7&M4Rclq
zc}N=6(6wn`iKX>V)(b+q1o5Tf>Zxe1O0=mQ{FQsCb(_c!`}3pq)>wd+Z|Krd)VsaT
z<I5K?k&fmL+iYDa8@bn_nVC|u3gx_ZHxke|Rdl?h`tP`TqUHN6?;dxpo>#j(@0t$j
zVSq_KSrD5lUU<AilfRxSEFPiC7yZ5mQN<JsJa^PlS!PB~4tp0k6`8yjB9vHkCu?C!
zIrK#d0q#9~Ae9hxeeqq1P6|3FJ%1dom9)|!cDAW{yUq(qs34gv0>PYsHN1(X#N3!~
zUcmxJf9ItTn8l*kUz<T7UTm2>jut^l=Zg@cBED4R&yi|+IVJS|#3|~k;cbqFd^!!K
zYI$NVsa$f1&h{dskLh9{&q+Uh3WpvaW&ff}-WFmjSA$>^BLJEd>}q4|Dt}8NDCl>M
zL(>FlNlm?=ML9Xj*^@)DMJ{?j>g0I@dKGb72IY%FyI!T><FMtG9Y3T7T%l`vYO3p7
zg(6`S<eIqU{n9)>zQ9jS`j~C23a|v2$yEL6C!JGozFdDX-K~C1DDQ~Xc)Sv(mfF34
zg-Y=`qi5u|&M=eGdo+Fys_K9)V7w3)r#d{)=-+F6)|_3?csaR%7*S+UFW9D7$<RQz
z6{qeM?G+Fr*Gdf(*NKwh7SE3z2t%tM!1LYIb(wZ;CeDzUV6zSEOKH{LM4M?cA!%7b
zM@xlgRWdVG<$m(V{~s{re=iqGK1a&5XSebYXlLgUhSn+&_hXavB1w9PPfk!J{^Ugs
zC=MB!zIf<`(8SMc%s-%QV{QWlvJd$+C~$}795I+ziEL$sSzwV<(qyj*ic?hsqib51
zgIfEIuqjp4MU)dW!gEceOD5yat@p1zhONZkF0R>DYrgEX@;f}AUTg+mjIoZeiIZAE
zjG4FN;h#p3;Xm`QvHGuZ*(S(<01JIHH#fIo+4)5J=8pUEN!<H+x^AD&-AVbox34__
zf$RAAv+##QBXga)#7|0#-{;eQwo^;&kwhK=?vD{s4iv((pWzX?;TBUir3ydhKz{`r
zhqyH4KV0cJTr<pIz$=Szk}FZa?xgH>l@=93=UH0WQS3l>N%kr$5|Yg0#W2Q18Y5~x
zVQ9+m;Lwtmu>Cm2P}V#0r7=~Ud>EME6z@VvB3?@XtcpWM3iApJgCn(o*_ImR%u9B>
zQqllTCtc@uTY%ask^hx2djC%zu3dA@(*l(`aZm`uRLGyVhX;?tz1Ol|^5+^Up|&e6
zRZ;(Gb5WrN9aG?u7hM4eWDpQl@mh`y65a1snbJ@I!c|eyTG1aJb9Dl{!YdHp)({#5
z5>ZXk9~8zO(t8>UU!=jT)V_wpGi(c2NpsS#$j|wbpfd6wcO#H0efg2t)ZDHsX~c};
z2vbdqjM0Y!q}GnUj0r8XSsFyoo>sg+@M6XuIDx@*cQCpqi>eI2M(%yN-<<q3#TQ-m
z$@@pd(z~-Z+-*FG8n3KteX@XA)anacqH6T2$GDo=pcxZ?5r}{<B*~<XunO~AKFIQ^
zuG}T&vqpxTtx1XwLwH<*Bv+3EPnbHNTLuJ2B`06Rz>25bdS7ZB#n}SnU?-8LAvwjH
zX#k1flIt<%5&Y6X{(m(kT!~X`57&a~G?kNnYReSeg6TV@;l*V)3rF)ystLnVyRJ{q
z$<&!SYDObXp2OaP{ch$rxeBA%4Sz|0aqs`Kf5qf`BtKbePisI7PZkANsU=iY$wjm~
zVC!Q2-m!|Dp9d@bJvpJFt$on)7}Z8ynJDOL`b<{Qw@`UBy@Aqwi3f;EtsrO!jq=0f
z@lRf*cKfNjc>5*nQg>_TeyRV^#jCwKuQEHSj+3vIMj2Ag85iy0sMjPRnk!t(9PMpw
z9)IH{;3df54{Pryc6^>cz{MrQc3-v1&e=Oka?B?H<Uf}v4>~z;Tv1#)8K*k2-kQ)5
zQyTtIT(;YgIU+#HlTvktsm%N;)etgGhfD)A`-&^?Zw=qRQM|iKPJkssT}FoVO@IQ2
zJFABc`Do?a$9<ZXPiA!DVtiFsmQZ|U$q5?8RE`~dIbWwmH7EpJ-uL#n;=9^}2*FWg
zAO-)i(4p9K`t%IhdOm3SBpk=?n&a^Tv*6k|gNpK{Uf3f?x`9-DJy}L(B7TJH24!rj
zz6y4Y3BFm?2oHC;mK#jUoL|e7!9Z=|T<GLEpybT9u%H<KJ&<yk2S`YUB0lFqLXqF%
zhFW9#K0wDOAE!06HEt%!u`uPji+d6$?d1^@-qjfXG-9P@Q!!Y{j6;YoA+v0wHyg>C
zD=b1q_4BC0Y=$gR#kJxmxD$b93#i<PfV%Rx6vfPy^+_Y@r*KU`fH!TCw7#x~KS*=C
zhetqAR9hSW^yCFh#JSa~^~U4LBn}P^-WeDe1QbZbZlg~wmhT!9J!AXdnfr?~r0%4f
z=NQ`9{P@qfp0+4dX11`qn`os8hmg>0G=(0Z{{s^8va&LgY`Gmz!r9qbQPR(V7GTkp
z!+w42i#(qqhzBPu(01kbjw3+J#Ky%n^7G39gK5po%s|GbEI4ekQ%y!N#s$V)F(Q09
z#JI>`K7|kOUz@PtTs`vT>?<lNels>Ei`ik~<dk4YHTLq#2n&M+T&RG&Jrxxd1I-0%
z0SQ5rF4Svf6F){4h`Yoec;5c03-pY=+BW)ukDc6zF?}h?jV2wa%Q~Oh6@cI?#L!+e
zMOf+nWI?+}H+`^4zJE-h(grh$Xj!<XC(gjtoW`U1iBsT_M-5YA-qfGq?*s!hPwWK9
zG|+H>_z~|KAA7+;H=-EOo+b53O3|@M39t@ah^Dx%5!_6w5*N~?d;7O{=>Wb(A}uQ6
z16%@3G4%NjbK)LeZ13~Y4yEvU!;6m(!Y)}yH~uYw{-+WGYN4KxzKaNX>h_y{M%o>U
zqB>1K=V}Y|g=#8&Ms7d+7<|Qi8zW(Qy$WoN5mP$(#K&0wXU?AUml%`Rw$wCfzE9ov
z-eBmz`vN6&S;yc)RnsCCL$P>KzE3HkQ}YZeDy3qyv~$~GK$Z-zY%>)`1T<;9hvQ_;
zom<PJ8O6vaLLJ(7gM+Hzvmea9f<z^%3Cm5gJlO(8E7XkiO&!A*!)REHjc4WY1phHt
z_mZncH~h9qA7Cm)$0=8*Bq1r~XZPMx6W%aiqBn#d8E2FPDLQCRVfp9hN=(Vmd8WEl
zIW!2`JXJq}PL0f2@x5Kbj3N9%9A!Hb-Y!`yp%v*=xudL3Op{&JGD>NfHZ8^Jd)HsS
zKY1d_!%g-rgwO)>@(^^3oWSU)B;1_i%^{=|%S}(74A{0gm2DuKIrLqg@y|d7PcBR8
zzgwQ9TG)q&EhH8$y}$Qh8uOU6#5b2eef&@_>Y{<snC8xBl_1sYoUrCC?{GIEhv@3q
zf%DyjKR*mWp)o=v|3Wv$#IQ47IN^l6uT6__ffhzlbg2JreA}Xz7R-n=HHSOk>Wp5D
zI3~Z3&AlMMipr)lBb8qt`e*N(87!tnhO-7Lhq-=D=%BrC+(lAwc4L`MnQ6ZL503>{
z=owvf#3ox-K|x6V^?8Zyi}g?{8e=HHZzYs_^D4GXscN_3kY>Z}chwb_kYJWglqByR
z8$)t-b_TY@)6>%hU+mVFAucX%X=SCUp<!TGSJzDA!N-`?O!-!>%iY1HCG1O&th_uD
zU^Ma%KYA#hOh!&F-xAl@*m%`_!xtDgP4FDk%^b3&m@*3rP6DdCEr*sQxzYXWv9YnF
z!jv7ZxrK$9#YG7=T#CeDV2u1-yx>Pn41mt*SfQ=Y&4rkTrsZS+nZa>#C64&vE`JJZ
zYd-^u1|@}AHegc&V;DfKCkTKz*Z)9wF)^PexcBk%r!8o*-FaxRWNN7Q2nOGm?JR%%
zA<w9g3O}ntu>EFk7FQ)9q5M5dJJ5VG`FD)|Sxv{^E$O1+5$U0)BDB<M<@N9}{RPIs
z1Hq9*i_vbE&63uaNj+JM!Wr&?y##n;dSR;<{kwHGP+fi3tj~J;V9w|Cc<Eg=P48qd
zsHjo}Dg+vq4B?NB-Y>$*p}T<C72gRDoMvyESo1%p0&C6}4bQ=A!Gq>MYA@41u(GkZ
zk&#1d)PaR8s>K@jl#-v(p4}GR1y(!a70VJ(g^9zuCbwu)zqeejXWFlBcTB#%LJa?~
z{)qJQ6XLL(LBnAif7jbkDD)G4DpW+xl4+yHbCkZXgDV%xa8nxa^HOLYUn7Q}nhyvR
z&EB0yEUOSFNi-==>OoGR1~)%)DjeK3MG<Jq{J(1qg{T;T=*56~*hi2}KjzDVLd9a=
zpHLBziu>-UL83)2Z;pxyi<P|cj5PRum4eT9*>Vq8r7uPejanBg-SiLl5Iv{aJawE&
zS<W7!H<FhU2(-LFtEwGyCbFys4?$NRIPd$#mI=#?D-VHR^|2xL>GJ;O?`**u*2gu+
zS=oy+`A~N1$y}65<GIaKiBj^}%s*Q6S{|=@O{dR+HFH~T=wgQ{;1z~ZJ<INuMAYhC
zY`wdOJ%3jbvdsD+Go`syD9lNO)i!yMvLo-OO1Y0Dsw!lm$rB`@La^_vU}uz==wRqe
zh!ZFqBS(_$=a<uN9bW_T96yw`w9uy}6$u-Vv0>=KXUK*hCDTT;BI~`(K~X_$J{0M_
zo6OhaIB0<BG7ss@=kP8HrBWnC`oQaOaXH2Lk3+W5GF>uOK%8QD-*~9wduvsqS;=@w
z%NuQ2iIPZ}A13ff)ULQXn$_2*>4?|YvLPQW=Z(z$GlCG>ohU<v%r(F#Oz14!$mMar
z)bR_q^C&1X&hs1iLDsSv?B5hHS|~EhQ}=7}K3Q$XPvJ*;Myr5&?lB_M(VOv?dxm~W
znN1t+bnAqfWj3#J_KdqDD>PbL^b%*r;jgp*pl<#Qj$ErUD3YZuQCBRAnNrP<9si1e
zxp6<u$=X^i)+_2nmtmh#Oj1Oi49Yv^0Z;>gQ8Ha10i2Gj-shdG{*YIA5(|#ZLM0lf
zR{Y3(1={2T7oLjBN<&Rdz!5vc?B2I)*27EsM6BAfQ45~rVa!}0*mebY0MsqhrOSj}
zUOM`0{DasXv`zf^gZmA2%7V>77i*;n$b6SjRb5zIEK{}467bD%Y8~4hO>K4BMFbFt
zzf}HLRITS-&%jZhKYRESOxvmx<JB2syTX<!OE+p{XP1(kjKRUd!R7P76qz3fDk@c>
z9UC8?wrW&RO$MGz0POAo2sIq=OWHR~SR(d4D=I1+j^@fZ8)-a}rV^Cgl-(Ak=u}C4
zV!cMq?o9wc{SqBp18mIVvOH?5`+~*Ht2V_*K;Gvy5k1qu*;XvRT6$u3E<nnSh0wMx
zs%>xIKnzcvI@;+?(&h&3ml+?(hRhkU@BN^tDRKB&GPgppul5*emyV%=@{$SuuNs77
ziIy8@Jqwj7i>|h<Hs<^z>~n&SPOU{dk6{3ZPd1pgG;^PV^8c*8_P!=V-&eQ`@vzU4
zRPdwtZFbb7APO#jLP(<jg#F9)r;P`3Y?nwV^X4^J|3^W-LHifjj{8oB>+w7`JZ$)E
zHChbHP8a@;*r!qbFh-v(-ij}Pm!Q9(&*q0b*C-GJWukVbWe{eQ*K~u!LD`%d^y&A1
zxd6BM6>7mV1NX0EzV`TBs3~c5$-Q5+E5kT0CBAd2CUh#=$cKo%`g%$GTV@Pg?hHP*
z=00<u0pVaseID~-)2C5YHI1kQ2+epDvjMhI^M;e<y>I01Z`W%wYXSpfshl21v6vdZ
z6O;<FKSDxD*$U6MygB>~$e`f?q9iQX>f6sis!Q9<Wh~IkPIu>T$o1&t)q}r|!43t=
zSK}IE(r<2(WSpgWh$#H222TEhyk`f-FAMK^l40>Sbc$Sy`c4vM1t6LGkG&9=P;;WH
zohH`?p?uiPUU@5G^0vJEpSDfshz=o<=QK%JT66-lzx1HCBz*hHg{d{Pf+oB&WoRSH
z4qc7}<QjvwpmHWh!lEyHV4uGb>#+3?*4wJ>r=;VK8y@}STtO-YEL7^5>jz%?U1_al
zq1c6_9@zL|TZ?F|%PC&0K90v1@?lJW_Q;6}>ig}0tNzQTZ2USE%YvOB7Pg+QLj3|u
zmo|2cYyrAeCt=$l#rN6W;dTrP%y>EeNsBV@JcIlmdB>9QOa|w(O~mYaceiB*V`6=E
zihxQJ{fNYMMf1LLt%m3<8@iaGhX*zdg|f0TAolx*$>;|Z)+rGY2r{!KwzkML6n%Yt
zNu#C!94J{btLu2_rKp-582El<mqqZ-SWvuGXKyQ<s93Fh>serbGPgv7&gterF)TcM
zd1=YO);8LOCpsxXxkPUI&}FIJ-Jbeph9<I~6sVU0pBsQY1}F}J_KI+8u1xuz3lD+J
z>~FSlyT>6D7O76X!8$s*T!DWY03fCQ9T+@g)d=Y80q_L)@58%JSv;<hU)8H>%qMfI
zm}JP&WaQ=Ro0`PSR7i6tyDf__y%kk4WM|uGx+R}Vg-}~0+1#QFExmn{tO(ddNETBm
z;Z^Vl`>&h+&fO6jPP<yQqeoTAcV1Q8drtd#`D5K~tmyS}fWF<G38ES0?7Qwev!M_S
z8X6lP>qBb%^9-yc%^R&YVW9NAph=S&5r0xAWJ+YR;h`B8mASJ7w8;IwE&hX=p!=$-
z2BN^V?U{HV998}S8bs=)oRni60rUSmK95C)0>qb19UaSmY$3OK`<$;j!ui}3<i#l^
z)y(Oe<UTzsZoJC9&7b3)EPolHo<+({$q}41=$`ECS$ueaTkzQQ8?H63ai&5-wuR|V
z7@q2rLh!%lym<Loy}K!fEfq{3frY@L3nwP_{bJ={OUBH)-u9s$qC92iPc*TJH+<os
zVdm(|&*t6qN|sFStvpKKrRahs&Uu(@g318ovDSxc+jl0*P>L|57HCyE-JL8SV@4PF
zzF+G<TE9Qgm0Z^ga%?AV7l<GkiY9N}?ICA;-9VOu=px`W3~9WQqX*jKltxClbk?B6
zqg#Lj(cfnt4?ci}U=x1s)CjB?*%l)n(|0U?Qolj-HO6t<$fb0r4z+L7C=QMNS;O1I
z>*vgfH~C<p4}%eIvR>SkU*}~NJfanK=uey298>cg!$vBm0V@lFTJBMxQ~9L~EdaHF
z(J~%D^=1@&b;#j;&fv6NK=vJ9GLKmFSJ!Ls0Uk1sY;0<Zng)&M34z+7H1%H=gyP7A
zhCIa@Hgx0{AyVJmt?6$ec4!BPGp>Lf)#U>Xv^7Si*Zh`-A-dQ+voene?~_!k0GL*R
zJ~>Eo`bIawSIS3BEdG7lTAl$yN76uIYU&cN!0(H{Ov$0_uvR!`-2&rtr^e_5%47N(
zzbs+@hDWMkRmHpfS!(*`zd+t;9gxh}C3Iv9r+3F0u0VW4X-hhtM-5VE5K2QkrY~R0
zq>S7n2zebs423#%%e>y7WMyOf?R&pASPglqR<2ZQ^}3}cdK{9_(A3--rphf;Dywln
z*HcqdJ2*c#2ij5<+WBPk>8x=R7B<}IMI8Y)G+aQEc(TM?rQ0svzBW)xfHZ61;D9>l
z*v3Bt*jxRcFIT5Y29<(GfFCsV_03I9i?pFGe@A=)SfJ*D&zgRhSpM<AS!&gYtInXm
zmE(QhbIo^j%agM|ZNbKb^r4SQvPdb5-#aB34#ny3&}TJu_4h!qno7B943VH5ZL*QC
zZ#GqqM^JO;sVzW|bKd@qkT6XC=tDTmhJ%N<6AO8W^*s09kfu%clHK2W?zyuDk!R(L
zrYo?ASP}ce30h3sEFLs`pc4@}kb>Hfup23ntb6L(pbv&;Jom=@+^Jtp?oH9O!L;eQ
z?%LVyHOg=|(zd=~k7c<?@el$HRc9F-?FO4?LGCy$N+5s!A<y_bF7gUlI#X@1A9`Mm
z=4NC2qy;Nd^84~{uD+p4vZm*sl3U$~PT!%I{RAUs;1*v%FQ!Vo>UqHr{QiH}%i{OI
zzL2i51YYsh3dOSC#&U;6U9?fJ4;zweHfN^Td9RVsngS?rD81o=W9ykN>(_)2e-;fK
znUVMvX<2K0a_lYbcYbekww_ns$H|yVQ5t&j+z8AcsSzXfsu=@*>G<uB>_5K3_txTw
za|z5Mb*rl?Knv?Y9B3QobLo<U%yR-E_aB3D-)J89?kK(4V+cNuO256GhbuYM?;=-R
zMhiZEA=Gr!f5UC=4MI6vp1kaSSr=UO=zfxYSv`Nn=j<G=9QPw9*KBx)Prr9}Hyx<J
zwd!*{4o?EqEdaUL?QC^enz8ey4$Y|ZxkJ%Nc=6T_2q*BCy>HAUDHjTwu1?^$$@Be?
z5R5yV4J(`q4POX{h$f$GVc`)Nu7ciM#h)OA7*f3nhYMDaGVukn0Gpx0K2a{c?!2C~
zbB43Q1n%y@@CYj_sv@(O&rC0JUrGmCz|<fT!}<)^D=FqGS&VZ$Jau>WKJs4UKTUn2
zj5#;}$KGH-VL;`5uJMEmK#6t@{|IRa@sjG#y&pDbcYBF)L{|-h?p_-ZGW9eJR$Kda
z>LS^q8Ppa_^nfl4BbI8zrUR0uXGsAmeI`%fNhvdQ*KvW?BQ9gt`4cGI4UT7Kd$Gc!
z@ht?iE}7@iG6e1fV>!5b5>Qyifi$5(0mtoZN48Js-@uGLbswM+!8lA@>Bhr$SKF-^
zM|gzc=YPI28>6ACR`ie;rT#%8pJ_?PRkAwZty|*87+eRL-ZGR*Tn{lJ=`5a(l)<a2
zUix;guzt-~Hxkt^NZJ89W}v(F$dPE(F>`c`2g+gw{jMLymgI?2|7=sB`K~&4dVBHw
zM{ulGzINps?-KEp;M=?lgCRzeE(5gIz|;$<p_9L=1E;OY{n=W>iY-u)r^^&;XE3MB
z06EF-Tb_E&FGlste}4sjpkrW27&gHM=4Zg4>a;qA9v)f(v#zRRCEDbFIIY&!)^o?V
z$6hkB^76)-m}F#RK5y5#CBW2emaRQ^xTuI?U|=9l>FVviJ6E^eO_4Tvso9>W-zeUN
zXZ-PcpWE|_+}{6r#j<&&L5tq0b=BQfZk-xktOq#wgXqwtYq*J^0>OvSUE__WFk6-R
zmI7^GejfN@w*}f^AGCE*EZ7lulWq~M@zv-?+R7*Qq@W(tcQ3CL)Vk|;a#8@rwoMYL
zGCIuUbz8p2tbHhjCCr$*l12o1@m5`X%|sV-a&^}GTmiSe1#aK6iOnJ6Yh>SR51$|*
z(xi(eHdztsgdUY}Ia>^z2YkV$!QFZ2(-^{Hp?L=%KU7q`wC1<drJS=L5&vg-B6|>y
z93d7wKQCQ4S5zY8_`g?MgciJ<G-(#z#_O#$X5zw5d5(f!e&E9AhAIHuMtiGtnK+y0
zon<UJnQ}dTkafSgnN6TCD|7epyRRc^lpdwb&fF@PA$&%xzuxb=y-;s$=JQD$3Oe*^
z5d(sZJ`x!pZ<Uf7bA0sm-MXmO3)O!!RY32Vd}$DUeUC_{S-DLu#c{-=6js84c&Hl=
z2Dp*+m6K{JQ--&HD9oxmM=~20(z5m1FO9w8iNboL1?LhfnmK36Z=sKU{m#$xVK!ta
zgY&8B=;&&k_vP4Rk3qo)kDW^NS`KGxx%t$8owx-6FJu;G<{m1+N2$HNJ-vnIie^DM
zNg(mAZGqI%<Y42}VUY3fR2>w{`zrG(EzE{d0biKx-=#5A=un4S%Y;>xKg)3+3>FD+
zj|HCXN9!Si>p>?69-zckXl!3hEkv94@lYyXwvxVsX}VW3p?8f+iDPLJW<OEZ>Ljsr
z{a)NrnD1lD`EHv>ZL)NY-cI;Zb2b0^Lax8LZKR7T`alN>A)qP|>qcNg7s`0-`?n{S
z$Eq=-C!RRqlAuFx!o1(`^9!sc9*!9%*X$p}mB}wfOwJk!ZRLpok8hDF`yBU2wb+KP
z#30vmxhK~W1Sg@VUXNPi=h7u8_~N?eB1^<ox?|%H2cHttePE1BLqImnSSmMNZ5y25
zMNIUj((KSO<afY1l%BC`Twb!MbvJ@T;JHCTs{2z~H|>yyJL+^T8=R!Z<3WMdd18Df
zQ}ATi%}%45WP(^t#ykV-+bq9T#;`X3iU+uL0<ExGd8i@wQq$V&Yxy*9tc$L>iAjiC
z`<m<?a5^v7n|mJQg+HwB3k<HT-*He<QT_8Cm~#Q3Nvc<;T=_g{=D1Rs3eXS>tE%#3
zPkbHOJO`X5fl$jzH9$fRco0NIMZuFS)tjRL)Mf1}U{GeWV&>$OP+QBav$*>zc<EPc
zUW^bF0GJlAPV)6NS}%G$uWU>JiW4Apyn9Uk<+r>-`8?1*19z)mE}XXbdCa3UVmJJo
z7@o`fj_w;O(E7g{+I5<<Vh3Di#?o1VR$Qh}R8^I3wD;Z@xR<HvKMxt8vcY21BTtvP
zDwFq(lm~1w?K8+{*^sTT-EbFO+l^R{6!*w#(YPt8+?cwxg59F?gd<ybUD2x}<Jkb@
zO?Gz5l>n$-Slpg{oOPRdV@V$|bgR9pomZ-xryH5kq~aY`)>ls2SmHCC&3F|NE29ox
z#ZjpZtyI)?vdkiuVLsA4DF(x8_L<o*4QHhO+zDiWQW1msY$L_@zAv4}u0AC0eg7M@
z`5qh30W~}aWYM{D<7flZcf#oZt}f}IcRU0&wrq*x9Ksi;_I>>J`_aiJ0O#bTPY^J@
z2zRV7Vou^LD2N*+QTfJ!cZWB+pQjft+J$0fKJ0(}b?h91IxGM7rTtv7<#oxnr78yO
zv_>ha!EyChs>gX!+OGW;V_@d+7tyPtM}T7QIFm4Uv1O(dOVQVCQvVd0Fp<gqylz4K
zoe{*B!`xopv(Wf>&I_MTG(0?mj3xEOmV>juhgi@>GiKaF4S$1)kOn>0Y6YGw)SqmV
zwZ^U&avxR2GppT+gHL=Dl}#>07=~SRAyrt6&8hM~FrDDw-~c~m66i{`sO02ib#%DS
zUOF#dj?CRfzthDQ$lJ42k&=y#4FD7PY`5J@lcV8tIlMDwU3(ud7vDkvCx1P33X02?
z3>~J+Q_u@&oHgax)4hVp7C9d;_zNAKG1zEs!@%D$wrLSYc8=2vvi|-?*LIY*pT)Gh
zOEJwxpsHHa$+63xTLP&rhju1h*44c0_Kp)dA7*uIKxy2dRPe=DuVW@FqYq|YADmm`
z8D|{dZo<`Q#H?EuqLxC>3xJ-)sz}A((|0-;TS==Lc_4ssThIES`%V5Cnnmc_PvSKE
zaaL?>4N~I7aRWHQ81%H_Z$+y;w$Nkxdij$xOa8}9ILJvZ3GXta&7!Qzgt-&@r*VIr
z<@Cba+`z!CxJm*ykrh#ODWr_YJwW&{K9`P7Bh=Y{0>ZIoiCwdZn<QsOClE>X=;}Fn
zLF=Nt2odRAri_<njt3>4NE>hxoPdyxf01Vo&r~#*vEkn+eI%*>dyj-#s-3=Ax~O}O
zj~o=BRq2-Pu_oY8wv0;Q?6q&Gl2fZer&Os{rdF=|86EmPW7h-eKc_N488o-BD9xPU
z1~%cl;NV~tT8yj57uRi4TK9V|6U{sY17F|HuPNu<5I>-9uR6{@=4sT524G)QAO_V@
zH`T0uxPDS#bnFHx5DAch(lIh#tNY)o0}T40|BtM<jLIr(*F_Z(=`QI8X=&;14(aah
z?v^g;mImqW66tPfq#Nlz58pm(?6bb}{%{PD0m3_<Ij{SwyQJh3IIabXukSKi+cur;
z20#uRpqruvO!4r&M9>hJSXiyz{aW+J&d!NEp4aOw^db6;Y;1GK_E*b}<DXergxS);
zxEm*1)O&l&(O{h*TU0V@q3wQ~i{$-nV`Jm=-m6UkJ4Apaj$bm&JZ&Yv^z<Qj@50Ne
ztgHL!uXN?e*)lC&iun1+9n;}9lZl3l=Y|N(@8Tm<vY{Vj#8!!rQ9Gc(XkPmebc`6Z
z@7Jv|g)2DxY=kEb`tgZcdqMM{LG?Ebam$aRmVEjEJ*A_43tTza41SRzvcJs;I&I6F
z(RuGHv}ALJMqi2w{P+bt-h!2G-)$z$8LDgWsys<Xx1-=QyWnk=CLq%m?y(x#zszfv
zGGQ8j{;$^>xuT1qW)JK;Qo$f&c)4uECG~dw$mo#v!7LBzI+<WWsv+3-;svox=nDHz
zYC{u@>$`mR%$v4iC@;6HqBm=U%ukKAe}gF(p0Yk4Y4%sZ1&9E*J9K|%%Rg$vNfx)8
z){WfI21xbq6o)6Y+*si?iDb1u?YtJPIF3X~o#e;)kW9xO-;k--!!oq-+dbIRKXAEU
z*NrXK>fhU#!*Jqv--m3m`?T+}W_`e-=WoTw`E`Qh7(c1GHnex+KCou%Q%tU2=XU2I
zsdtrxTsCV#OGj4?vP_(rh~V8{TwF|s2|Ny<lhg)}qJaVE*Vy1WcV#-_0X}FeT6P$w
z10)9FaiIaa2+Y#)o90}Wmee;qU1;=*P*i(VU=;={TScPTtbUDkyljzrhuOs_pKgJ@
zL)6_CPRMsWpEaV=k{ldr4?$97&I(kDbaa(&WX!aWIT~+eNptSCQl+9tv$&?<35OK-
z7xt!pm6eY0RflUcsryvJ$|QF<1ynCFNw<;r-By!@RBU{pW;(Oi!0$Mphs$_66uFnE
zaU5r~pctmRDKKLjW|@0uB>zT6WD-(ItbbyM<`w~q4;uHy%{u0KpZtW$Ilh_2jIoAK
zsI*wV_3O5U6v?3qRQShhcvCfNXT|`(cNM+Sb=@iqeho4nc5igwWQDI?PU3wKq2l0-
zgx@hs(xfWoudLf6#O;wTCPToIar=HiFrV}%5~1*O{`h7{R4G)xa`4+ev3$sKlLR9k
zv9$%{@<pc9q?iV^zB1h6k5T+O41+c^>GY^D9jAGR9ybrG?)ctOLo~nusPR*-Xd2ve
zC-$8H!yXwK$;ibOQCfNgNSoBukAIHY#9|1xmzy26KQ&ZU`5CKoL_4nZbe=wF52AZl
z=)i$<$;m0nlxhS{qZQkdj7~XVysr*a05G`n$aLad!613Pr2yR?64+Vd;!wapczb)}
z@pyb+vTe=0x^f0{d*?rzPC;K^K2zE3jh;R@5RuC@#{u^oFt-#<ch^Z<Es07FX1cBp
zf_X)b?}>#{HlxcPudt{H;1{GM<LZWn->-Sn*O;od>xxK;@+&<cOd^msxmTD0euTU{
z+mmgsgW9?DqwrrP#_G@_1_?$tzc>nc$C;BMqK^+qlS|GX3pj5glO@QN$WX&77=y|W
z6^o|#!<WruZBEyqourtq3r9R27DVQ2R;O~>d47AW)vChYNUr0;)F00;@rT5GTG&E;
z%c{^N&y#kH>LK`dsuYqujDk!M;e7|Blae7xJ%5>kiT&s{T_a{wQ%2#j`d(>3XR7ic
zzH239<nKQZj{jQ|la?N_B#Srdh#?X5-E^;NgWuJG<(a|_`4N;3hH>fy7;7bpcVUnM
zVoup4<jYQI8BLmEcn@pqPp3t``?Lv-zA2#{*YvzK4P6_={?1-O!h-t^D2FoNt#|Zn
zA`81KpQ&do?6!x#X4e(Z{3!G<wW=7mz9N&N5b{qR4mPj(O0&#~f4k8CJ_aP;ZP#$!
z#ym>qH>&%g5to0BXry83qPATsay;&vtEu^!pgpf6LHeF!on?0tEw(1xbrC+OnVtVg
zcp&|SR8Dr7h_qHtwX|wmv|@9ymlxyEkM9bgDIk1LuTB1RG%aZg{$$6W0RjzP=-{A)
ziK(gk^`dE|=7Ma|blXMd>CyG?{lmk<k8xd+2RV05OhhcEupLMqsFy!vJ3cMgGe(${
zG?n`ezo_Z&pS#y&1r%s%J^Y698O^g+BV&r^^7TO`N6Sg4&f1;6=7NTY60KQ&l1fmH
zW>?U}nTrjJ59XhipFvEcpQxoU481SbvSxmXj4~_P6013C;p$#R+LAyZJ2Q3@xzI}f
z;}Ovpx~ovZHW`YcF^v0715;#Ho)ur_ifvrgt!{OPq#;Ri{XOpw<G-b-3<VQnemh)l
zkl4*EF0&>(8VNKA=>pS##p$EwroOZ<2;SnmxDy<+#&{G&<)(^{i|j!$*(u<CSWCR^
zAz}2UMN6Y7E;OoRh_LZ1l=LE&N2L%EDOf}`V$Ia@Nx5o>8b~P)3C~tWR*Xl-yq1jG
zQscA8mTV~Kb-*?8`XystXRZ31s=U74N_)oDyjA!xQ+4tt`mfgfse}SLP81Y4f^J!%
z$N@mJ0teD`EaNRvoVXo;h!npmf&V&Z(hLY@uPT?6loZh5z)`EVZl$3Do^x%98q4{L
zbMP3B8#Mv%%U`AJdvwCvt`9BQXxj6O?VjwTCJ7qlqD9lWI+eAaH#SyF6hLY(2PyS*
z!~SUf6n#P3OxfCwjxny)bHr!@o*~`mZC!`+O%cEogbDvhmu&IcX97^-HL0LX-G|kA
zD+2z*2?+@RLQL;FHQnu`gD*Aib6pE2Nx)k)dEGG;nlk}H6F`Br#~t?8ZEVM8ylscO
z5aYML@&E;JW<v4X3>iUJ>LwZI<`B)0wO-@vSv+#9s?r=~W9L*dbUT{PH9b@yC=PO`
zYTbuy8*JP&MWhwB(|(*g-mWhn*}Nt&URu;KVe?o!t_+)MZU1}W{qp<5j(Iz2Rs%ak
zv^<N%v`$_-IZi)unnp;R_^rgcR6v&)7J*P0Q`w<2+Yr4N65{ztB8#Hhw=N-O1Qz1h
z6|&naK_=?`yUd~7|K<RHwLvHbfE~HcW#?aHseEQs4|lIT&f4ty6^v*TVc)A$zfTID
zhbPZG4XxQ#^$}5|dSlepfP%$Pby!Hla>_dp@dx>#boyv`Ea6KUY37)ITJ91KwjbzD
zK_o2n<Zoc9IzF?`o6}Om%-}#_V~qHfX&mefCxfPpA%fs5T_A{(BpEDG<HUI+g((5<
z1JVxdr`a&PcfR5_9Uqku92U!|Lo3sXpmFPX9%z~3j99-6E=~tXCF_|`Ebr|0DlhrJ
z+t6S3jLBD9_VRAUrx*<HJJa*|WyA848*j9PWu$wm{lp%Ry6uqq+IGAmEKJ7UFrB{h
zO^2qy2;8>$UM~1{dkLQzUpF3ra`_JD6~WWzr-#&x3^4Qo?RD?)a4DV=6@r@+3<OFO
z!2)^tcKgbK`?kpv7QC+BXV@i~*60XF4S@yCmsjacWps>A(D0JG*2`~$hj-K9PsbVE
zV?t&GimAoJ4x3X#^v1*BS4Yjg#7JoUE#3pMN?J;Uie8kyH@dN4Fj>XH6G75+NnFNw
z{^@-@DQphW)0<C$qkYB_r8x%244d^{bCs-J6B8_f!qmaJk~3M3g_fK>6ZR}fr22ZT
z@m62u2aL~>Euo1~vZ8*vQZcBW1^U%dFY<i}VMJF9@6NK1Q7-JV_`NyEkiDUu5P10!
z$BiwG1)3IrF|LjqcTKs9<=u)+tR&?Et2$Zv-a6MpwD7Dm#M`fu`t$T~XlU<~XT!x>
z^bBw?RZZp=TOE;Ql+?7%C(aKq@mB@P9We4{bDcdtvVQqiU)1TDF}c>st};yn;dkiC
z*bdw7IOwd+lnNgu8VDGv3NLYr1n|>?50sYv+l*iUGl4k^&d<uqT0r6?VJ2f!Ap)CP
z|Im<RBSZ48dnFb3k1xKY3Lxtd!~G8$yqin5onXLv&87=N<Mrouc6Kg(u<nn>=e`(W
zUTBCi<O1S++9(CMp`b^+o@fw^HwE51kcChvQ3lQ*#gbnjh<jV0drwZ_bF2TnCHD*|
zGtFVY2??;GMZF}z^3dG{Qptdv0Hz-%fI6zx$l}B&Eb75?;{$=4-}d(QiG7jcC&d!<
zOxe)BK9R93o?*~XF>raw^2%3YUX6A!!ZW4!d3#Q4hW~Jy)-5efpgGw&S#te1@9P_~
z9mU243$<Ipnwr@$(y@d@)JJr%Q=kPcP&C~v3n}6#KM#n@B$20H4=H=muu|35&&2VP
z?Gn{G3O}h}6^AN_BKBm;R;Wgc51_al=umqM+Y^7mD2LCHG)x;6uu10x2}(hs@|{fv
z^~%@u4E@HTf>Pr8&|H-vXyAuRMCQiSb<;2uEjsTgqR{d;sJtfjzh;5YpV!C*kR_Ss
zMOLKkOQ`WUAqj8SaUXSbUpPU|Is}t44*LC~iM5D%N>fEs1y0Ts%|813q~y=Ce7$t8
zX<Do<7l~NcXV>YL@Ro;d$9@sUmw#IVTswpuZdP)j!&EJ*9+x3>Q@I8VW4h&D>bBmU
zo-Xe0nVG#OX3-0VsPTQnmnve$nT8r$?RCj3kB+d0wcxUs>Cp&)!srOR*QM!uE|;L=
zx#wFxahKC)#9gq7UB`e#fE^X<i8wz<^8L<Etmlo={C$%XardRuZG3aHd@O7FSuRqK
zDKA=i@>NFXbkRnww4T<<|9Ani5e|h8JTk~0d>r!#WY_2>f7semlW25#7obMB7j!o8
z5y1E}_?$;@Kc4l@SkxyE#yD|GJAx4s@Y8_wPCu>xh<yIuMqNz8Y=-)5WH&kGxm5*|
zA&Iz5Ka%Y3oKyr7RRy8x&j0N{lX$o_wF0IUgAE4^2|vD+9JXC~Z6s(fLS69*X>=Je
zNW_X=y%JcGL;YI_t||d&GWs-=0FtvOOn=NAgoDT}=X+Okv%)ZyST_<{u&j7v38F~f
zp!98{m)`dXU8~y185}iUZy{J-)RIGNi>7tj*jxX6Wx6<a*8Ct#mGoUX-R-fzXqP1r
z^(kU<Wo899tD+xfr?4B9uhrkf(}?t2=R*7q<`(b`h`>jZVgKw1l~zEEAyK>hi`wXn
zn{ZSmS67U1tm;xZtbuwkpG8p}Z4Jf_A4$8hSXINVLXIg;&)gk@Ik%>2GM`A_iA=zK
zb|+1jJVC|0`EBjEjAOs@TUY+rGlaH9*z%%xBa_@)y`6X0HO`aV2N+dx-@n~kx@UzJ
zny)Ej(E&!+c8wRDQYciMQ7G^`epTrH_Qx=TkwG6D9s$9b{0!`?0i@%rHV5WAn{Xyc
zOzZ5w?S?+}L5rX<Ci;8)s}%gTtPdy?nX(`H?ZI(wsu%{eV9Pc+&zjJzGGG&c->3X5
zz!tqVv|u^_ngMvy&&>hf@N1u5MvT`Msaj!Q>*^Ifz%I!Qx;tPz0TU!B=mur+t?Js^
z+rd~mM9IS&9EfmmaL~jp-fuoYL4nl%^Ro{?G9~ke;35H@gp}0Ojly)BO*&h5IM8ha
zZcB)OZ|d#bK!PBKY^g|)&VR!vUC>HT4NC=4@+Lgs^h4bfS>vi#ECY{EoYb_X*E+u8
z&lwi_G1}UfB<jVE&#|!I<by<Zc_gVCG|Fw#Xyb)jDs(BE0R^74840P_2yY;_-?zuv
z!sFbUX|@Ke0tx2rPlGMuVe*jgenSlx_<5~DmgG(Gv$Ig^@O2=?zkvutfBR$-tZ1Cr
z-A(HccX$jhOEmHJqu&C0=OhQIIaXNro|kZsoWkm!X8L~<iO)3rLyRQb9MC8^=CEM6
zv*f+_*xk>ES<u|^!HbZ<6W-^cv6nMiNl71`hPpmwQ11b!p`CmnnjqLvXxPpIN;V5V
zTeG~UFM=wm?^a=;%^53sw`L)eR;dI75kDUNexu7KW-5_$IFlQz^Y|i$0dH$Cp{n{V
z;nM)Z@1oM!QPfD8Dx(vU5;YbnYEn|7eBmoA#fw_(eeRusH~zK;OB?194Wo<%0?1$O
zXp?*W{`B+G-yY<}t7^C0KkjbCiRqoLR>5e|W7z;^V{MWJ=K3hFVmmcUDK#_Da=w9W
zJ~-y-1`vhfb;rN=<X+HCOnbW}KKeMX5Pnj)E|jV{7L=Z@v-z=oHB?jlVJr&srIBIo
zda9KJFVW*rW`3bTrfl}$dAarW0U%j>H)h}FrltYl7wx>#?8NAxTxadd46Rp8+?D&P
zc6$VI94@!2bXH68HvzL09a=_2M)F>43evSXrCFaEOH<3sCDYQ>64H?s_5o_qo-qo9
zULJGi1?FYYM~Pwyc+<b)WD~v7o}7@jD~O<H<?M*cg$s)3$Dpp^<7b8RFC|8Xh7q)=
z!@^IJosW&~{<Xyz{8Bi&{W*-m>0XR3O*Zp;9Dnh%3C!S`7aKL=6U+olrfhpy+1^i8
ze5l_a&I8%5E#S(x{R`tq^3O#0NlN~D(igPIx##uyuY7_>LR6P<S?x&3r&gf#KZA6s
z7b?{iQStWT<7cxvh#-I~$3%0*FZ8fU%U$^t1nE2ZTUEKjX2#6bNt89Fl);`0|DOef
zl%%wFd^H|RZ(g8L{}`T$WR<;nVn<yxlLbkj-|u>+)P`2~3{6KGXb30m(`$^w%+@&g
zNtSXh*Nz0D1Vtbp00Izb1g*9VApgJV>U-lVL6M-<>QYX_pOlmos~9g?fDKf8Z~*}p
z@GnP7db92{OQjT6N$Gvf=e8+R83lA;>jNs_<fH;{3;+v5eLX9SOx~m!Q2|-<@T5&+
zH7#7L;dC_@$UaX@AP*V=LZ}zCecXBR(v~ET8tg$)%4PmygtuDy%_74h^Y^6n)Wpp#
zr6}7c&9&_>K0G2Kpf7=E_{$fGvROeCSQ;9dnL}sbH3G)spA{7jJ0q`52pDW!T1Uq+
zx%QnnfuC5O7PEEgY$%bY4dk(pug~whNn!G3ikd~swu#!=s9EuJ$<v+A$eZ9NP;ew%
zlb)@*x>4M$D*EiYcgIMD+U$&@8fn`Ger>FL3kX2p;6@h&%OGyJg|{*CDwoi7+P#~9
zxX-4;9wv57Udc|^9V)6V6&zms=mI~Ht<E=sC7&k@p9q7_q11hfz`(y$qo~17;eFvb
z$5eaaf=kZG(C83TUo%aVNXwl$R9Av`@1RlE#R)MmAlz3v-%T!mXz8DOXjO~v```A1
zUu}p1M3I8c^ahSm;?shaS7*Js%J*4hiwD$?SGv8~PwHk);Ud^L0ng7>C91i8+534@
z^5$$puG4Esq0_aIKKH(%eU`au7i|*`<~UZX{Mt<`dInAbj#$cBx;TRLA<w)2$fKg_
zX_HOg?~ZMMai#zM_0;PF_zf=oHciW|TWU?V@S)V@OWY*oBb6v1->6A|7MZ=}xL>SA
zPkbLRscGhVSwZBOLHD76s;YzFp1a?_D9ZzZ_u#yLW6@Nom5m8?E`k<Q-;QxvRP+q(
z(-1B-4NX-IYF2z-U)8V8z!&_-u~Tu>Fx;aDDP7p?3|U8Q$P-{K)Q%Sse$odHvYV%q
zAJ0)9YhIK#WWO6IiEV`I?9wF*s%XoWG|fHZN&c9gK1&s#W<PFGwuXVGaTk#ol7bwS
z>rcAV;-+YIJ6Kg-nT2)T$aubX;eocpn-XC-x5a7v*qnlLIWXz?V)qMA-6)tvxNfKv
zH&>slG~dn=*=Y-FzAGWAdlMiZ>kp|S0;kL}N|yY2zX@EaA_6(k7_In>4;g{;h}c};
ziFo#`*;dhRVL4!DB&%-aXtw&Nm#;4ZBiF1PPO|@GJhh61#89R-(S#L%ty3W`2)$rS
z@IpOeUeeHvCZ)VHzHdo5yq5C#J923cD#0J7Ot;qH8LUu{&zHatNdQ48Lw)O}yd*E4
z7CSMK{#K_1#v4bYGg{{YnwpNOWBlKAExH4aY<b|W6d9~dnULgXd0Ixg#2;xnCr#2A
zAL2;(Z|Of2g;Zk%H(TNwO?)yzQ^twkwq$dcBW|z##vK1v8)n>k9PitGc>7e_SCvGU
z)~8ZqgvQ#sx?d$qMQ=E&Uf#&@-jP)tGO@Du0Z8x$fCFF{0uOd@^qQN$&}%hKV5kDv
z(A0sI`u!hk(IOGL2hfr_@6tO#K=ArxW%yN6=6yboaRf#}uQgL|Z=mW`(MEt92e1(B
zFW?6u!^6Xa+RNLA&KFPK>*|<lU~~d@;MB~_0>yH>CF@3oKUC*i334^QFFvKS)J1aH
zCLlcojDysPuz)pT!9f5)07m#cnwQdg>OKRMwVTxJ>|JmdSV2!TPeX@&56~x&M5QC;
z^irU}uyflXSE9TLr+W*5#CuCLX!MUOE0i4%#>O};5e{xdJ5vNozNB@_0p_bD9rv2C
z1662gX+{l3NT}m7OX}84DvO-9vfEmiU%?f;uKOdzsRF?dM!VR$EFlUs5YH+rCkhgt
zc8-q@+;v}Z*IQz%ICL4N@A;bcI;#xvijt#xq8Tj>M~He~{K#&UsudT%8f!eX)r0=q
zkc3mEa}qmRTE~O2<~pgIEld)qsdDDk*!mJQMN-hP@HuwJp@1KG;G_bb_{<wgq<Cd2
za&p0gT-%8@tuVLrAYH?Q`M-_BFO84_W*VIMMnv{2bTK~;&j_E@*4@u4UsMOXx-6HA
zzUPu3IISBJ2HJX&K6c1MlTQqYr=;NE2_qs_Z<x40M-t7e@q3s2>U29m5{E>Xk=hm1
zq%>9wsak{=88>-hRG^CGT!^VKP0d7*Lb9>qZ>Oh9Wcm8FhgbiY?eU<p$Mz{E-pgl6
z?dHgT?SV&}Y@nFpl)<K;I)fQYef-3rKM0dz(c#PY1j<i{Q4@H$d9$c1m+6JU9dxPi
zjrH~N@lUI#A~nY<b{P|!bacV}4eHOR(4^jdBprkVV)Zx#fA2kRuXy+nUryE}6iaZA
z*u$Gx)Acpz51KI-2MAGosL0g6d%ZSxrlh6yG+1|B*H?aKXTPa|?6pKOZXXZR!73nb
zxQYJr++t8Y(0;Qz3?5a<^xB)%V7c##O{aAD%H9)Rf3wXw`*#~u5!fL(1X<sb9(+EP
z^?~+x8M;^t3=*h@)DLTkLQ_8!N&@Bghj-1sE06IihG0ajoPLVM$0UP7|M@doA9?~`
zi!!H%c8}JLN;LF2T^eG;kQy}?*G=)qJ}lMS(jO=+)*2O2rH9x|{Zk37V=uc*I%sZ0
zdN)rWzbpGGLoz`#f?ex<oBOPDW3v+>@7dd3{Zji)#b?o2c6jc8JCI#NLi!0V)G{Q?
z5wm%*pj<=1{lQnB7ZRVS!IaEFoB1P-4zDfpC*CA4&_*%@3YDrN6rDcYWWu~LRl8Vk
z{+*3n%;rHnWp>&|l3@Zo0e+v9c#x(59GMc@`=e+<hY1q9-0DewYpKY{p)<QQeg%mG
zobXb`@2HFC@+|M?5!BKc^o^L6j}eC~(5w;gRQk^GM$-J7zUJ9kzMZOU=g!v0W=Jrt
z6p0lXPX`r6pjIe@+c}=ubS%dl2MiOy`-Co%yJ*G4#`bFXjT1iur1X>(TPqWZ<ZofV
zo9if#^7g6x?;~Sl|3IfhbyL%~A5`3DUGF#=dRN;$L2PMz+c?YnTKx^c1gX+cy)0E+
zY$eL!K=A~6M6=xn1o#VxGyMC<1mXk03=K>QzzjWpQ49nydU_^a-mzc5Acu#Cz3zr}
zMS(t<l0Wqc;Gl(ON6JETU{vOCK0yDEpbFg6bzqQ?Xs<OLj~_LWD4GVYr&n7u!($Hv
z@CCYlW@f$!cer?n68_gC74ewdPqUXMA$cQ?gX-D=fvS<oMq`N|KM2_y_>L^$i_7WK
zAeR&~x3fna{yPbqG^V?QTIc6Wr5FdR@!F%TG$W^R1s1x9(!&%c+7*jbii*QccO6xb
zlZ)DQ2mTQ{AzbnilEK;ZkfTJ&IP8cWRa<IW$#qc@I*+K+hxubA$*=|KEGX1`^vGQv
zou-AIew)#MGd02Re{SCry?@JFlYWmXeVw?+0b9Xn{{#uqa#~~#Av0TnE99gFev80u
zeKB*v4KP=tB>9O1CffggU3h#$)hNegik&&kT@Ra*N8-b~PxHN6s?^u*APi3(PTp|!
z`9QDb7aV+cbLX}2Vrc+DLD^gR_wRzY(1`A|`_ZIdEALa~oy481*|COMRsa!vu-Tua
zE%W*wI%qvJXyaY|p>0~zv97*|TuTd<o+4r0m+j6VzDM>-`G4~BVI!$fMjj`<h!zN(
z8ghEO>jG!qb^8%V9PvfmDTRwcT5OTV9EdCj@ifWg3|R^wwY;D0fv`(Q?XFHA3U@g<
zx?lJ)dek1;OQNV&WvE&i)_eY8wSV~i;t{RaHcc*Tq@r8-*CtyL7*|xO8g@it$h1Ox
zP^cxF!Qo$Mo>C}5owzxj{S5ZqJh2|{awVD^$#5?3H7lsMp;5}(flmaHH*nDb9}GaJ
zH#NJf;@_=xArTu#`waDV-2yrSG4FVz9-I35OsS8)Xyh<fY%Ew?BvQ8&{`kl_HD3K0
ztLkh4ezt(3@llY#(G;%0nkW%6#AeEz)m&A7|5LiZghZ95NPX80S4ub;G<yE!xibbG
zvRlV5Bs_G!9Bs!>*$p2D)4m0a{d1(N;X(*!*K~svAMa)45k*qdB4FRA<u`8;DMOXd
zM6k<*>T?R16NGDCTINJ5-ZMwrXn{O_LuZWIbbKbb1?k3$i|)S}K*S#hYrRx>%%!H0
z_I0{Q(UyrRp3)5Vs8Gy(G`BKV;Pd;CGwjvL(6nkhs}CyBAnpIoLNb<7+KTe1I{(L}
zlSaPihY8exE|_7#QdIQm9v7=K@|Mj!ccOHBQCNvTvR(Beq5c7b3(>qx-XD*yphk_e
zX9lhaKlU=7aaYoOP*yn(#{#Y)xGRN^lj1m5auTcG!09%O;rR^$1=t@y>I2hJyXTE!
z+3cBjwrJklps*ScQ&1@XTIu-|g_>CTJ4!MB4LH4EnV56m3cxSA-bZEJPb1ThJGRdh
zi4O1m02RSK4i}?IrIyqOr0>QR7u=gQU%scOCm^?g3mu8)68W`5rbL-q{L?3N*pcz^
zS}>u4J!Kz1e%yGy78<M_*Sv44J33_Z7WMM#x&bW%JktPxNN%H+mby&y1MmU7QXp2@
zvaQ53-V6NI^_;b}wSa~NTnZ3*+-7yG)@-3_?5zI17@m&yZgIjfudTUWD|tGrR9Z-s
zm)A=T^h8)yMIm6PL*g=r@I&>CjFJzIWD*-~P`EM;ay#xeC%6l$#Sg->8`Z4AzwI%o
z`j~(ndEOye^V56(YuQaf%XMk3_f1Tm5_SC&8<f=t2cPk2IoElIhWoL&d(ZKQ4mMn|
zd+h8|W_$hAuSx<`IXks#g<%!@@L0~G58gEr!MK$xUB9dnAF@)m)7mYu@%>Km(Yfu6
z!YE@#t4x>NS3QouqO*6mDe)WRm{SV^k7vmDA86Y99I})YGdEAe!gTY_T-9MH4728q
zHJv|OU4MEmPC5#Hqsqhx6aMBWIh{+nUo|7?Z;*2c0VDoVODhRh9DR`@{d*)Gy8`MV
zvcMi*1h3R<H$$FdmPj1XgyP0`2SFJZw|l${7T7WT9lyx*|EJs1{Bjl|wls%XL&|#`
z%R;zHTwIiwo2)xkQo3vH3>^2S_xd}Skjs&Gn2cvcN?K)(X0;Z*T#xHI>QPu`j;?Z6
zc=&#e-aCH9uxk63T~OcJ{fm<z5Q5_CHFDuf6=AZ`AWZwyIo$00LJ45|+KX{%?{ReA
z5k5C3??ZZ`=r9t}Kpw3kP|6vwV}44J-<Pl-Wldf-?S3SoB+%Z3o5PIeW>C|BJ?zuL
zoJUY;KS8MfDNj~5+l^#apQC~BzDDsAORBav$ODXkd((l^hCOV6l(tb87QuX9pBmYN
z0!a|anZGucM=udO1Q{`<$~rAaglXEDSBaoF(BPXg>~*-ut+C4d{f%q8HAGXn1w!_@
zYP2<AcM<UJcN`db+!(H)4J0$LXTUea;|Osn?Y&3-66icnxwp6|v<{sFx`tSzJcMEi
z<)O7EloLY^n0ej(HJtq|3}PpjJQiFE_~)%4J|aUBr0P`JOjIes7FzqCkSd%(1i#^r
zLxzMD?4NVO>lcZW*E~DZg!NvsmL({!@hX-0XOB_)K{ke!!?k$G`4pRsFQ@u=PG?zt
zZ!wF^tm%7ES+RTfeVM5EHkx=J(r#-3KWx}x>t32zEKDB0f=oW;q&Vpzq8o1y+HD6e
ztV&N&5oO?JS=}G!wxI9<igOERX%YRyDv1_Ke|hhqg6aorT%h$FEHIfU3ME4&I7;PD
z!=@_Zt|-FbVNJ88_L{m|#Kk@9q*+T)l#^<QRW_L=g`74~rR6lHh&1pBOOlJHEYt{4
zm*pN3q-~SNbDkJw>a~i^NI*G3R-bV6emhxcw%MH<s-D5uZFJC@J9H-SzEl8wXvgXF
z+FFcx+Fpw+N!&W{OGd`T{1v@kRGDyYzjz#SEJ&XnV^C4~h>uSW?Au^U!o<P^zyh!!
zr=Zy2^@{+$raza3hew7w5hS7QHhSLg?C!qab&`GdlL2R%3uJsinpL;8MIN?XUSB(a
zfTOIDZnfI0JsgkAC6d8=Un-?=(0N`P?C&EhF$!0v4nBPVq7C@U6D(062h^<fyTO`8
zC9km&>l1mR<a-sB#VY-Gh31<ai2sS9XlMuyF=Ad`9gv~%*KYjlS|E8j3^Ru4d(#)x
z)ic4$A|k--4Dr#<qPqL9=?<0ALEFPU_bG4e`peVT2(j$c?KB?qRc|Fi-G^1O3-VCU
z4#$jBZlAPrOBI_IhalEoOMC7J3(W#*W;4;ZeKU+BcHnaDwa<voqecDhCp$wjVSgYp
zKcARsxFTC~VrfSfEpn=Eok0VgkDigryvX%a!3FiUgAAHhORSGYP&GwZ0Q!9{S@8@v
zYKVZ)+?n_fGYBcHOoG9a=UTH&1!+?<cvxbPO2gxH`i6kZW_N$NulisZ^ItabyTM-j
z^}AjDVaz|aupCa%i1@Gw;#VHUE0K{w8uJl~mAz}vbdRmRSmCN!Sg?`}2+djxmGv&x
zRI$gaGL~!g85T&jI`f~MuvbXjdv%(gv?YptQBS-ro0ea$*VaAP;DWQv=#{_8!<5bH
zxu>2Zxc-a!;9`G&Szz>g4{kllwJV8^Z?7Ar%u*sR?;BCh=?GrCDmjCz!_|V-Qm&n4
zrQFgahX>0~ZcU7}P_R_ju)il8H){AjolKw;4%>GSanD0J(3FzeI6*=RHL1n2yP5@g
z&@)QkseQY=l@9dIraJTpBR<I=VKd-K3bx&XxB+m30g4`A8VtHJuA!5}U(BNPfyK!t
zAo&jvHlPwOIm<<Fgzf7x%(CeZx8q{>W<NWy`KyaTLe17}owZAi@G0usH~e0NBd__C
z-F}L*bFieO%uv_78P%JQhS*QQ-KV6>Gq%P2#lz=<nr&52p$J}FT2rsGF|J2Ug`b+!
zCkJJ~g+qasrH+47!rz9%tPTgdFK6=WqzB6Zs*_BW^9ynzirt!Nzx&Y%+CX>!*WYvI
zmw7gq<sTC^ZhPuc|Gdy>3JgC|)b?8#HIk!DL46XWEDCbTiL0XL<Vf<)FWiJfih09o
z9B@cP3vKaM>aQa@leuua|2#ymd{rcwWDfCrvrkdPr0)tt0{knXm4%FLN_(28;jDDo
z#nqKlUAZ%wjhS2+I<WMLamf1(O@2sco1^5v_m8dngV<7(gH0{@TQCk*R#Su(4@Zi3
z8_xKjrKfr;z1*<*XVsE`qB!X#Y=jRRZH)Av<HVgvzO;(?i7W4&7;4Z3*cXV;9kWFI
zcG|*8&(1z|k`QTkXCRM#o6KXIW+H|h_WM98Y`o$yWOHGYBv{a22A4f<(1dgMpcjcR
z(mV~cHE@sj-gpBF7I-{Bkt@iO%%}tA0;E8#mUb#yPr&$_l$;C%WH7LR@dsGWfzzbp
zwvQpc-7Vr5n2i1!{(Y^W1i}uuxo{!Q1P2F4i=)AW0;1<@+3Y??<PZGuOs@WEIlfVR
z*X2nR$C+e=OYbY}kCKkZHfSpY`uo;@$|*V+QnM;Ir7}axlpL)gU~F2)j<ixcZ%q=t
zxoWlA$-Ap-WwN*^2bWZA{-kg3Eiw&bL|rhG6MUg{-@NNhIjFtNr~4O7_RxUvnR_N8
zZBaDm3UDYPD6qSu5yt6bea|&!(oPXFlg>2{fBrmSz!*$^jv}$~ABL<i#zGP-x%+XG
z^m8_U@@#+qJQp##?rwEaYwsI7y4RYJBt{7|d~g2hwFDKa3NiX@&{2l%%3m%V!ivKw
zn??plqcMQzfD;lhC{?tFtPe+q|0Cb{T-Gfet9s+F*;;wC-+;zm0OKlq=6;R=cbjOg
zyxZs;)_*3Wz{8O0Z`m*07piU948HE)OG?f*ieZmdT+gPwjbiu^J0Dte+dp|E7zgv{
z1Vg|h1fxXc<#DuLpC4^%3^_+-dXlA%gd3A5*r}?DZp4x1o?(vqQ%mkq*U~OD#XXJ*
zG{}aIHEJBktHb)Z;Zq?~N9{!Hp@j=t40pj1%8@a$S8kU^!D_mSTRo6}m8TDlt|S%f
zc}L(~ud$K*H|T+jNw8`*S2-XoY=UaerdM+nl|rq9ssfh)S)`fldmqVSnNaJ>4xr(8
z%5A-AfPIIHYZzQx((O8(v>j)CYWu4ibVC&nH;Fx8(TAbVf*HL0%PKq4Kz0~d2*6T_
z1L<cM20n)_>xC+RI?aZ&%V|00FJIgjTUuJZ>hi;jIP~1s4@R=h9S+LSsCHF^m&hj}
z#^KH^C{Akh&ci6^DD~Df=ZHMGvmVCf<iF70&J|HM^(hY=JDy<>Lf#IuodkqnE6aO!
zPovN9_s3Z~<%S?KjEYV2f3B6zfOC89_!wZHEyLxRLg;wzsc}g`ejzqw#*9e+7Arc*
ztdUrKMj;qdB)%a*>jCcY%l+t=G>n)b<@C!GQJ*4$v>iyY8a5xHA|_p-KJvtd^Y>K#
zp)oP14GO_A2Cy|e?pb{7F}biFOcbG5X=L#T9@Chm#yJa|KJ7B=cW$$sX+<O}Y%WYu
zyW>AL;w03pMre#!{TU$>HPVCV)$Gz18p8i1jnF0!P*`bIMj51YySk)hNryO%8R+Ns
zgmiFUL^0J1hwNBIl`k<x#?aY0)%3Iw1W(Ffy7cg)FyMuje=Q>=aC(`Xv1#P;mPUq!
zEOXX#j%XG979HF0C#7=!F!S@3!{&lTn{t~T8H=-R-gngDn%C;1<3IJ_p}#YluCAkF
zVrLg?n&lqCYB_fXiY*fW3F6SKwY$vWxC0N=8{iT@Z26A_w;7@PNW%9}%Qp=6ivU3x
zkJk%(?IJZJ&FEqWPQ6Cyf4l&r%XYm8ZM!_Dxs8p<GU;q_!`qxT)xDueN>u3LWT<;s
zwtVjY1}Q-<%xbM86MS1I&LUA!(Ugpg*9Dp?Dgfc!c8ovWtc_pI>&i82`QUR6t}om{
zt@#uN{a~F?!4QE&7LS*}6{*`F{&2ZN-p=!BszLCN{Fl$<ZU^#V(mzJWqTCs!$41FG
z_;h#Ecwpu*)(7@kiYPTlRJA0Bd6l-Qi;7IYIC*7V1%9gd!#s+6eIiCNmzxUXG|I0s
z6Wujy#sUBs>AmRu?~?)HCMR#r@=*A%SiX7BwPUO`jx3Prq3~lF%nPk~W4Bs3{qIXC
z<dviN9TbXNu{*`Fv#F2Xe1Eh5KvkSc-!i-00|avVWG;3-v$POnbpY@Tm)IJt`{N@1
z$1=hfVzpQk+}GDfDBg81aOWdZEJo=49TaQasH~=8{R7>>XRNya8C*vA>;s50uw^OY
zm>xfuzFlg^^|?~rmf{+$XoBeYhjH7S4Tpi}CrsiLcXbs^vN!+@B($IP{vSF@&fy66
z!};MMAiiUXlB_yoJwBiII*tb~6(8n0Q>VgPE8BYF_LZ^ON8lVdQp$yRXp}ii*xvhp
z@jxH1Dq+!Sh9s4KhqmYW$sb;@+rogvX;Z{;@Yu!e3^$6AzFN^=3(@ywG=m2&6e*OR
zZyv!nD-%J+1D|AGyHRoPA=$?@15`MO)DYrN<{P2oVqI+r^RO{&JGnO-S9sAnaQ*b+
zc8CQP5uj281Ou#A+fu<L2s${wu0UzeH-qNYyYjVrI^Yy~6>{zZ`ECQpfXVX^1$E?&
zs`F<I{Lz`c6)LyCp7V|;OBc%HZp(H{(2Le0)Kc+iF{-Jpw$aj@A}<ZJ539UAsd>gG
zSwG{BK37ju?0*+1(CBuxHp#k5ajzS>XE%{Js{Z2rL(;pnuZ4!$YQdc3m@WL2w8z{i
ztIadA=y?x4q*Bi#eGn@G8Uo^NbH?vF#3af=V-=R+US=(Rl3-yu370iM6rZpZT=2^F
z?k-l2er9YT=v90ZCe*b*#iUXcHDzWvj~iKWETT$gt^Bx3rie|}#zyjCcw8>7)>><n
z%v@s5q;_OPWdQyv7O{)2(Oea$I4R_-I8cc-%Esop@scic5n$l*-=eep>xF3x#=MV|
zlSJC1TgXVKvUh12()*+L^i5*gQbAquOcOCmBT7SqT;GB=S)&*RMh5{QraLmKqIHHE
z!--C4%X6Q=G>)Tz$$=Sl0FWqNx-$71pVm8;@nb75z4mlm$Og*i_~z*TIrG^ru_;PQ
z7NCac+ufe&KmTKRiI4F<G6epM*QD6g6e{SzHg#|S3sivf6VMny+le5&mCnfC`}Lsh
zzMTlnCT|&ectFYiqX{6tKm!F3igE)(7wSE+Y(1Bjz{!6Ni$Mzb?Yb|6r2De_Zsr?6
zt$@!3yeU55LGqc+!MFsprvKv{1r9b)96=J-#UjIdnTs@O#RjTkfI;nu^-0Ksd|x@B
z`x*UR`xgLtv}~rV+%K<<)T!s~1N=&ivz3_XoJI}v*}Ls3$LLgdZl@ZftF;2Pah)sH
zru#P%-hSxUGNIvJ3W?Y9y+W*dP}fM?S{4rbkj*`OSHkpH34y{Sn(6b}PDq*Q{>SC$
z%A0^H0N1p|UN>jsD>QmW3LB_3g3)BM#dF}Rp|P%^6E#ea+Mm?$p=i1<V=}Y}?9(Gn
zFqaYfBgq-f8>)Hb@|Jh#sC4MuEGQlH{5ljpA{A^R3Y-vW`q;}p5679HTHMgbC*;Dg
zXpgt^|KbkrTTpE^08qwtTE#^oXFYJOEN|Ud5Epn2l4Yd)=cS6Ejg(k*(~-P-EVBCI
z({!^sVMPY{IIAlO^+DI<W3$%n)x0l$xG~82=%c9S+DVXUwkEl}XvoT>WN>{bnd!UX
zJ%`NT?5m!1HQYhFya~Ne`w?`w%kGQze8-2Bhl>`*=u=gR)L;X)R?X!1rLk6-$`wCo
zH6g<q*J~8>WO8DJOO_o6YH9tbm+xI}!whO7o09lQ@W%5EA9ChaFX6`yLTdMyeZFob
zIc?OcmTXr%Gqzj}^4o(IuSRbg-{&o;v8?v*+;@9<XP%vux0gk&ko{EGHOehc$gRv&
z#%iaR3BL|0;+^FTL=giAX3q^sx{w@`Un1y~6wtvqBEA0ZwKAx>x_T_Aa*A9)Ocr*U
z#eEGBxxi|$?eGzBO$nV0TFLy4@aTO3Z@fNLO;eN3n@E^AQDY@7-YdPs^Q!qR)aeS5
zw_K=aT)e=0Nb_+HA;0->?X+LhOuv-E`%sbUEB~Y<s@vORXKRye5PY>x`ptTBm-*N1
z!g>THqS4oj%eIN!-{P%Vibj`R=g@wnzvjI70;i=Y-B7JeP(}6g5c&X#&G6qL-?<ol
zR(puBy%DnykP?Ym)@&Mow(rIn6miRl1I(s0yG!az=-8_fBHHx*&(1V>Rf+@(=)X9)
zgpKgWS+Za2vEjDerO80!8^wso<(xe#P0t+C+iK-)Dq~GJ`I{7vC?9OW2TCgnl@gTH
z30QXo#Yuu_RH;d|TSs_1eT;QFndRF%f8w9KaL%QILyFbX{+SbMwm5r*qBVMkx6V#p
zHNlWc^JH<7L`%Hyg2~!@Ien1jX;_QtlJ?F+8F0DEq8NyUDPd9Qn;S`X=PjbkzwOn>
zRb2<x{Cird5M4J;#%~{EglDr_@Y4sZ-`MM%(9;7kujPmU!XY{GRKT3!zV82Fq25w;
zcNt_m*KS_3<0Z<af7d}66>KU4g^obI9j@U#|F+a%ou@_b=(EWkeF)MoOJOorjY}NS
zy`j99#Gn9jvCW+sgkX-XRCRR;(+W3Q=);5|J{k907gE7u(QEA;#UKX&=G|+}clJfz
z`*RTT(PDF$WyL}?B`_IJxhlL-VONNwstKk)pdFPvZ#R{~<6YXkb2PRp@5X1hF)UHo
z7-;5{O8mH5*Kd+*;9x`C%T7VD$ig$M?RuoIx68FyRM&jffHoqBw_DlqNag!-`z6cG
zC%$BnS~W8IT#5a7N3beTnIx<t6yF-4H#ADzJ~L5!z*RagYs8t&t*Br|ZJc~=&}b`a
zleP=dP%_M?az9HkGGo;1pg8C5yhwG1G$H0_k2Uekh2k*t%L`Ja+8>nNqMQh$jpE%k
z%g*wBb_X`RLh~O~5KE|P771!(5&tiFad*B&=-H;{l($$mc<BC4@0AJ#h(a38<v3tc
zK^~pW%FlX9&jR+OMT&{(A8b;+H7+pk=|38@BJl9&%NqLf!6Q!9!HqDKCgY|`qa8T2
zRo>i=)dmQK3V9zCrjw<!p@zFDu&dI$x_cSs`WaXl)#j#COB&By$rJwbcH1MqZAOCY
za9)Gkjrw(qeNX6v>2pqs$5x>?Z$pmWTUexTgJ}K2M$H`XaxZo^O!$N{%H>Y*c$eMF
z-tqQk1uI;C)8N;Vq%wvRew^&*rY{2LT4vl4k^1#G%TLarHnN**?NhHcpB<4h{Anj2
z_j3yYti$LIxxLjaNiCtS4ttz5w1BTKAK1Tu1&oHTyYYbMQ8s2&GITh`z`HEBlsZJ;
zsl&D3UxY;c#*<V`h{gQyffua!!sB*36M<xU6DCh|rm`c~GLmJ<?ARe-J*2PKwO*P{
zAiPB$_PxuN1tmJas-Nt$#nc59eh^tX%|zvNigT(@honinfQ5X&D*q?!{Kl==FU2Yc
zZjQz?En*jHfuS!}8OI;44GMZbnbYZ4@6HK$iRH0ifrTjbby5U8A-{AO7TXGv**j=r
zir`R#S>94-&GU1<m0jr`Imls=Ux6Imhl1-*Qb<E2k(Pc3QwlF$O0BNvCWE-ahOJqe
zv*UQH1ZO2}EPhqU$P&iFU6&C|MKRHENxFM@Ij`mM=vQZ!3!90_YieRCJ*(Y7_9q1f
zLxuhGJO5OMoB|n${(@Dz6b?<39O@Edgx@B~^6>9gps~d*2ExU3iDR5J$eh!+3j;AB
zjo{3`8^tI2jijLqqODs(iu&c6LuTb$V2{ww4LXHwP3E_M|5sB{O$5!^uDden6gPl;
zMd}>Wos3$pBdPr0(<GhZmg-pInf>Fl&;6x2FqH;k*(SeAtl-G<tEd2W0u(%gY$+oP
z$FJZ)I8bW=u5qxY5U9S2w3r~k3Jl~P9v)!F&2~>P96kz2FY04Mh3EsrJHQM)JfN8h
zI8?we599!F)mJq&iIc>=7R>^!_8;(%{5#=V9R>7fh<+6yN_XACHvpq+p*hfufY(iE
z9yAv}_@8mpV@6co6P7YW=agS=kTv9$LVSd*qvwQn)}}o3=AIV=rDy3CJ1e-q-42hw
z5b$vKUzlRu&M{-mAwjn3E1MSPpHJ9l(=_5}wL)cmN&8#!SE)(2DfDJ_^#`K^kHs@}
zh{MBey>n~xLGJ0lGqha}7nisPf~-xh52m+g_{{JXt!ezjAVy1%Hb8<x>!2e*<tI`>
zn=^er$Fq}XX?}C6aFch~lAMX#V68;mcvtJWqi*wn_r)=!p@=A`#o#Ce3&(#ek|0N1
zqL)j6AG^0n<I+ah_2&tQ<4vL52O%#MN^)ijf19t&d!;)5zptHsRP9jb!<^3Sq&y<w
zVj=54(5IfZY77n3S)qhf#Vv2<MM^n!h$i5I8A;ANUvA`626wrlYf4R-x!`O)Y0Ri|
z-I&=!@pOg-SAHn^8}_e1-q}G7-jJs&$j1*3C-tuDQFRndQ>95VedJ|jw3C1fhgZtN
zvT(UsLt<ZRGaz(17Njv+@AL_o;qpxl9$~xp^#0}A3D@2}AdI?(WQd1E0FU5$fW-VX
zW6_P(3H`FcrxjknGJ3`=KPxqqi@b3f26=f+vic9Rq1}^<PYGJr;o3^gbj~7ptU2Zh
zwzzctk0m(|De0MIClnU_Iuc^uB_^`rZqI<^INW^Va5??ZQ_8=V-*kxZ@KwNR1VV4G
z%UW($JsChHnYRF~sw>i%g>ani%IV<=;-QH#p?YHi9qMNyCq<{5L=jhG5m}gi``f>K
z&xZNmxgU;dxdBzgfIQgZo9u72VHF|d`=$TLkeogvMQEC3Y|A>Na#^N&Tx`}`@jXW0
z<>EPWd%ny5cdW8viNu4Z-t%5#dTwSJTi^JnQ(duARYUul!CGG>{d4h;thQB8$AL#p
z4KdA>NoHU1H4Ke4Fk&ua5&e<GYkEJeK3;qYBCP`J7k4%GB!ag*-kHP1o8ag&BSUDd
z2|%qk=638EQoOG~teMoiGedKo%Vjhm`9ATH1d^<lEgk6O@A%Fbo$%EaK8>=fFQt=Y
z*4ia-c95Q{ARwE4fg3-A5sru>hEuQ*<l*n;SBu}$OB?MF<jmY!ypYPr!nHb-Fx-*m
z-=yY6*e6v^tRsfTq1S|p*z&Y;k|I|Tk2{dt{DV3>m-0nZKbLmldytoJnrM5pfp~t!
zh6=ZEPuJKYOR^@U53bxjUON5fqn4j8F`E%rfzCHfB0rQaTMNzc_4Uc)0zcy<S1J=7
zVy=+*A18hropd)?Vq;&ndj~w@S}9Fk$<HTJk5dfKsKB5Kcx&Kb0O>VQfl^uu7VLoD
zou~UN@lS(BOrSFZwC1U(Sk!A!Dl*--111CX-Hc^?3Cx5Jbm194gf;_r7O;gCIRJdz
z=P$cYAh4}jEfqF4vLAhyTmU^+O4Ny<NDrSq)p`jS-a+{mkW)aK9++eAyt14)pIW{Q
zlA%&U7tdOPZf$so$1cjVcjz+2&}4#f@@UFKI7zcHsr%bq&QexuoMU0NN_c#lCSFk2
zw)M^O|J?rNk5|NTYR>f-EUupt@<dISUR?0;`DLz2mTUi(jdGZlk)KH(B-hH~-C=6o
z`uF0!|H+0^-r%8P1-sc=eCc2r5~MnhU0NM^A|aWGe~GB3cHO8TuS*bi9_vQRw~<AP
z%>Ow_M<k&HyQ~_lR(|b1Z^|)5=%N3FkMnU>-^T$~ng9K|<DICJccVt9Qhk4Mkz;uV
z*GV+EtNUpEuS_@3>T<fa`GbjxUBk)tx)C4;A=qq!kbJREI=Ik1z?w<v=)R8#GJ&ou
z2RT{UeuEML-K?~w-rT-_wvN<OGT;*+I$hYgAHS^!KFrm)KKZ}~a^kc8RD=JECxezt
zonmrpS&c19d;h#W7!VP0HYf+<a+-8`<$N4e(VOYoL&@8{)BlnNYbyj_(9;h(EhZ<g
zCVpKv^PxQcsgz{1x_uktSPtcTUv#GX3`a|=6&@{J)cd43YuBsIzI<j@G*=m#$u${I
zfy>_0;ho#zydmh8ad0)EeSPW!*I<M3IA<$yeY65c_`*>ug}Hxt9qV%~0J69}14Bkk
zPAPyrMUo@a1B+y$L0tB=ihQnIQx#*dgOG$$$Q5#7c=DL+8-;Hk%<{nJ>Kaq(KQ?PO
zccH$w9|*k81T@O$j7dZrg&_Dz%v2!Ku)K_B1<C}bn2X)%mK*pnAf~IZ39g!-ljkQj
zER@>VM#D@c@g9$`^VloNXDcuEbP0Et_s!>Q_w`T5e^SP-tGuJgQva>-HHA;zc(|TN
z>&3Q#@p=JO38Z9M<(08e%s47ycJDYk!fmKiT>MbQ=T64g>&UHV#i;2fFwnJ0wBGO2
zK+EO8276V8Z*rNXg_)#KeTgxm5HKl!Kh)=qfr=Hjk|6kY*~5>4?C^fNP6|;&qiDry
zx`O@uU6e&_7(Bd(@O$g#0L2UQ`LvWEMAUv~b84Ks$f#f=1zY)r454hR99*ITF;4|l
z?6-?;{w4erxv1aq|G8tN%AcjtSqb%H<U;SvX|rhOl+h=BLP_E;^S360_R0?4hh4lf
zXPMS>aPe<r+yh%N3UDDEuI%d_GDJychX#6Qe_sq0|1QQIVHq912PG$+(vh0ws#fH2
zxC1PsvR11!0+JH7i}0B`h20gItyGv103?7+EjYfwqYh}P*Y5i0zjo6y><X3h0E&QP
z?4(_!Q9fnH^4hcoG%!$23wo)=@}qxMfi79=rT-A7Ka{DRJz$-+#Uuy8n%G#y8k12R
z(51b9d}wTA6AemtaH2%T7}?<iE$aV-`Le(VzGhwu3k7pqkAG6jcTkAiX<A#%9vD|s
zDM|*o*rj$c6bP5kGSZ<5$;A(yZnap>*BU=zj}xSC?=#bh&KJi1=uYUjG&=Ts(yW<D
zg5&nI$m%EbjAcK{GBB`!*oV1y2ndatSEwU9^>Pdd&C1$c-eS_dn^PE*PHNgNe8bEq
z*&~}%3M~6@$dCdYE+pq{sDsKXxwAH;-bfJc)frduXejh6TF5*U49^U`iS)W$*Q$_p
zF!UM<7|JeKl?Ixp1VQpBf*nRkdqcQ{KG;4f;%onhjQGEQ`o<SiZ(ZD$&_<Y|V`6sI
z2|6xDphA9od3=f0Z6pYRB2MnLVv#_vFm+**@$K}><Pz$i^1kWrdLzIjvxIW==hMFX
zSy_(d0FS4d<j>vp3r^6kw?C~v^7o%Z${Yvl`<I(${G(Ct06E`VNbX0wrV%e*bT1YJ
z;&;+Y@uRS?!BVnJ#DY%@Oy)d~j9zC#0c(#4gHh}{%SgmMZawclthcU_CyWYu)4?MJ
z8oz~*fm?C9aFj23MDBNY!F=w2fp)E^WT<N|*TxOkvz*g95J*plo2_)>M4y8CkX!UY
zMU@JZ9G4xII?;|$g>^aD13wziAB|GaEowa>Te{bCRj6Y|aF{~RgUBhaU^@3KX}ni%
zVsxj{a@l}oBOAB|DH7C@t^(%UG9qX_v}eMmk4Zc~k<6!*mhQEz+J7-i69q{aCz_H{
z$=XFoorV90tFw-(vTM7&;zj}K1}S0FDc#cD-6h@Ksf2)ZNjE4+cO!_1bV-Bc2DX%R
zpT+Zj&v)K){6icA9mrtZ>%P{!=KM|3%~qD!b$6N|V|)Fu(7w<7QJLobq=;n2lHyp9
z)V^Wz`JeOIvKmm)*5U8vwaFt4S%D>Ay@Wl853-_DDqPp4!{=N*umkeqmKy>%e=<BT
zeZt;h!`;dEb&fU7M%z|_ltsNlmy;_qA1*!TTcPAj_b8(|_V@PrZr0TZ_U(%agY(Wn
z!I!*#pCtVj^R2#7w)~<Hvd5--oC6#C;yWa~IhNyP^ONSqdC35p{zuQZA%?_XKS){D
zA(x3@Vm?Zm^ccEg7w^zTB0(@L)IBCLwoR<yJt<LSS}M>ht7Q|CzHjAfKK4jP85-8q
zt>m{S`4Jo{o<yAy6I?;13#g7E3bA6nUln((%SP4=Qr4vhQPK45{l&2ulb#9|?N_UK
zVHsuCvsmMpU=)pCF)hv!iw{8H*ErTTcrWqXdwY%%%)hT0)P?!%A~?y!nfQ`b=w&8X
z)L49apZ_^6CWJhEnth#15zgz;ipbu)Sb^WiEfpvI#*0f@pHr)_xjj~9yy1)WAE^c0
zZ23y~-xw$VcLCRD0D1j_+x|Cz-2+cYg_gLSEO+Lf#6g8|R}OgI0Y42usDZNS`DgZP
ziyN$(>G@#o#SQHB=tI7~i~z9!tV;}kIpLp%Z65o8_#0?M1?E)XRONHu#`FTDSPa3#
zN6*kn@|GOeTODb8Sko11fgvBLDIiAkut~XU0K$2^8aprVkBJGje$rE4Ac26?uvYo$
zTl9);D@{#->tBvdSYl%`P;0AWRC}FXhXC5<`_(rjq*ce^F*Kd)uD(5e?3~LkOM3f$
z?87tU-eiU>M;o6(23YCM3+KrOAre8o!o?p)%S~uU=cDt20p{TmIi#h_CjjCp$lh1(
z9FXnH)y~bGIXuFvii;?z`HB<G_zL$#7NyVUXYAMU9ZEGL;ndc1i~_sQB9wf>Uh(*7
zMe!5}xEkTW#btKT_d@O4dB7VvCf9%SF73V$(Xq>g3vwJ_))zdReN#1e1+IGl7Tj&u
z*!iEcCFlg4E3TZyeK-D}MsMzDJLCA8j$v5G$81_iONNt^e-;Ct@^vMsd6KfHmC|m8
z{;Dp`5fgBEdn+dW^l^_(@0ADVUpt@uYM@T_ju+-r+1k)^8uBSwf5}K!CFGH|@wcG5
z-#ruhGp1w)tv?K_j7IXu*7|gNavD~UU=bGn*&bg?Do-PkM1aQo=7_EA@cWcHWTic>
z<`w`dwh+Cqb`E)mMk9JOlrg-?q9_aX*z_DOc4t6tj(kmr7TeT#wsHy5cCJD?r&#R>
z>GBFTOHjG=y|S4<Uyi#DXOtxkD#PjhbcH^rpmV^@i5nGyX=~VuK5gqg)cIEv03CkE
zIBC>G56HvuDcwsg*jGa?k=Hyt66vGF0DK9gr+hm`0D+s;&8;KR!XsLPF&nWw(o)r>
zvaAeHWU5Vjp8}L4U5XrtbRArODvJVQw&`P*1N&Acr6X#Z1!@(NM%9^&i}1mU!on$r
z$7R$5Xu;)8s+>5(ob)j~e1D5=Mo4wdj$|V1EhemaPd*Ueob-FCR<%IH`qIzReLg9A
zat?In?)F+=lU$jFeU{#$i#>n!TEDQk$DgBrukhXPmZ7Y&!wzNgK7lgTg)y^*JA{~S
zSkUdhY^?=po3iPo;=7^)1=AB`5}x1bPH8sep{H>7WZT5&7~Y?rxzBg7q;t`HuVARe
zBv7A#h+WOLO2~z<ekB=NV@*+=4qg^}+A-ffh8Ox$28!3{NZfr+t;Ig<+Sccew!!Wl
z_Q;J+9zSJRHoCxB72#Rce2*to3}t@$>3X!HWhQ#ZgxTk*qOg7w{(%DhH?8zCKIkw;
z#V_V>LnPkym=L}#FK37%$m695i>s0^A~vcY3hho}sOV#VamAt0PF&o=?tsY2!WU;W
z^0ndbWlD#c81Nc2E1>EtZs_gMMgLkN^&>#mV<q%c^CYnzpKV|&r}BM73fr`$tb(_K
z=>zytE3}@k9Pb>!OAS~+^~9zRS+N@OO9<Ga0L&N2w}2Dn<otZnwgJp7z)=cfv7jK;
zKvX#Q;QiZF3jqEgA}2qPP-6i&CqVdkfOr4}9IWLFrfu_)!fvn6-GFNh2pm90nVZuE
zYqqEYAgs!F7+Kxl-GZG=Ss4i1f}77I-^<AzPOy3bXGuZsYPn+D(!Tqr3S8E<uJW=d
zaQ}Km&hqR2QsvLW@WP6%>q>y~(M`+K8RLehReXKl%-_B6gfHH0iQd6ha#x2gB?Hfx
z3;j9Let&CWI2UgGb~lWF+W31}{4I|oNgk`yadtne^T`mZ@%`1Cc9>S=IEUQ)zUX`T
znz|g=YW;$B20AZcLz3X(28TcF=+$!llG#c_(<^<nHyl(hL+7_6yzpi(M7o}SbwLEE
zm<%Sx6JD1s&fLojm<~(F;aP@EQJaBtoBL2})%21}>CzYXUHYgVkcJLAkq(kLqy0mE
z`R~L+o+|bdHR~WY_P+m{6-ea*+L`gB2R;fS805eo{(}R#DU?*D$S&F0NtQF^G}WL!
zOEhO^6C&;O_95EZQW?n!%OU$4CiY2V8ouflf$`3vaL$uX#9{fv(ALL~Xr$)PFv8a~
zq4B^ndPKjAua0~1bA~I&-&3gTD5!FsxlRAmoR9ZDrd%M7<?E9(v7W<fg%k;~_<>-U
zT=*aC0II>P<+_kk@^cWZQf*OWe08|VT61*x)!7)IK>U@h%PZX4C`oHh6^UT5ztPsz
zF~VnJmUV3*?|xrwb0Kj4+t7n~I$&n9t~gkQeS?G1sL6{J)#>fpqZ8`inJb}}KFLxk
z_AGfi8awdklCNga8dwc6c-rPrPpb${N+h3+w|=6rWOrFyi;Ds;xcwFw*)Q|*_0501
z0F{P=fb;@_S%IRZ+UH*0Hl^J^Wxf&kp>OkY0z~9P&GbW;!^zI%%~jPsf2mS-D--*x
zjCLb>-!C7`{Ezlfs<9cON`9B~e<#jf<R_;O$WVGQ<Mv7cO;Y=nA9cC|x_q3nX&HA8
zoV|MZ+wQcg!@D9zn`uL(+%o6=8kApd)Rk9#9tNlRTt&Y=W?e@RuPYC;HR<%TKnZQn
zv1QPtQE14@)ddX1T@)r-NWypAcE=Hu*EM%1zS$6R5|H99^%qATlyi=@F};SdC9;^z
z=TYbq+ZFPdIA`{-Gv{1fLE(*i5;Gb_@f&WS8RDfz<brjcEpZ9`m<Jy26TwIv-jALL
zrE^j|T>cbPn&Ez&6Je0Zf)F;mMqI^=V>Cg@yak3R8l>b$^5ZT~;u2O7o`i9>w2RK!
ztWhh-ZSoI@WqeV8No38`W`Y!B(jo=_5WNsTl~x9wltoZxGj>?FMz@Qov!1YUb9%~^
z9Z}4Vwxfmt9aJP|EY4$06Nbw?x(M9yB#~;@nbA47qYP2doE|hIDV!bM@7Jb^V)#I!
z4ahlpw9NA?4{myl!g5uPw7~?qVqnvFfd4tnIoQuOhoq;ck7&~*nF7JovJUjqZMe25
z37UP@*!|*XX7*O#pdDxB90c`~<*4T?Jf%;@G|=i*zGZ+kkEJEp;E#J`18IcYVLr6P
z1~^a)$`}A61gvivQhGcLi;^Gg*&t#CE=U*qK+_Xx&?}y{4O6O{z-}=>A3ZjjL1*Z9
zvF0RZ_05Lki#<l|?dBYg{XJ#R?Sc@&^(3=ppE&LrxlihKZr0ZKY+?`X&4hK9gHGYE
zm5TJ}Ih8JZ(=_+j*UR58Hb~9`V0`Gg!lRv75fTi&7i!&^XZW+cA6MZRF-JG)J~a1N
zfq_@ybsy|!BT@(@=mui1E5MkA=%BWYyW+d8o5Rok92=zru}C9<z4=eZ3q$e3Bb4l`
z8tXV#&tzmw1Od|rlUP~GQ|Fy<h~bPtti&YSD^yY!Rqtb=xW;w&5rWZUXA+R$<j?i+
zf|%DkjJp`?>G}4#6#eT2xyJ?B>j#Wp!0^na+$+>|&HMWR_eq_HO%3=(5QPv45oS;m
zYeBu=HGU41tiNGq^5=emFt6XRAmJej`zG;e%Ru2Fnh)&g#i)BGn8GD+Qv<WQ$o8v8
zrK603e`N^q2qWDR-Z5eSpd^k_ak_3I$2&QjH?M<_)q6ni@t5y867L>I3UmzR22c;W
z^}A;%s^9`L+q=s1_<Da~=VKBWj*#GSQNI8Ujq1<MF~efIRj(Bc4q`vwJOin({l`5^
z>bSf<*mr$hxc)aY)X&h{#qybQOYd0kFzR{!l$gjP(92YwwYn2?WPgV=AAPT$*<0jG
z&K8h8X)&*35E5E7jT0m&A+j~<#2~+IGIi2@)*m8GhWtA2hf&%f&{E<6>;VXNppyP@
zF##r0yK88p`oRFV^ula#_l_^)#Q4&`=@Xt1KIAcIsWE3CXmnV3XtM__`-{$%^r@J@
zyTI%><@PUQ5|sm9QFP5o)Ir1&WM0amym)=6iYP-*XFp(i4*wB480q9qm(x44;0kHX
zn?^mf6E+f{NH`~)w`L2d{`n`8p|ELd_Hsm+Q{BR+QC9P*eMe_^s*nvsG)p5xdZ2BN
zR7YRQaKN8-YUM!1Xfy%!`elx9HwiRhrqRqiTPp02V={XF91(ZuW`FZdk(e)7ZDC?<
zGv~IoM#dQR`jjj;M5h197$NadSj`Um!f^+ko(KgZWqzSGwm65gMx^mMmRg?|bxC@U
zT_a|z1V8$d7St%JLqqc7y&7-RkT)>-KP&ZQ`DOK($*J%SJUn%@AH^cFx>JJ=mP(^_
zdu7@61f|tD^x<{*>?$gMIHPC>PIVyD+_Hlud!Smd$G>=Za&$Oep(n!~MfzAJAS}c9
zjcudCtF$+j`!DSro?iR0UKhfPX{NMakTc$Y)Z61qI4<`4B86U`0*^MrX{j-ak|{!Y
zpqD-!izGe`=%S*OxF%p`)e8b1b)%}|5DfoEgS8-E2jaOSxv#-O_Q^B!G0^i>p|#$5
zz5m9?CmnQ&z*~WFwJ13o>APOV_lFh&$Bt7iZpTF#fHLM({D*P&V#Wg`Zb0Cwa6C_k
zsen#Snzvko5&Sb?dwPJ8m=lW6T04W{12UftEXU4EOJHt=jbsBmypa7;Yf4TIncyf@
z<}e8H0ZbKDy1{8~;tXW;m_H$IoQuBom%DLS!Dv%k6-(48BoZm_qj7Ht<DfzFIYmD$
zQ~1^I8mc<`?Izyp!tT#*ns+CI%12IjO&<319ynf>SD%Ac$;e*6>TU3=!<|G%&p~ds
zb&Eak0-t+r?!0k=dw&e4_38b^uD!qTZR`h2*#<vl8uQOAK96MDoNQ=P4Jf$euJP#i
zERm;o?-MWkeQ9JRD1D6^=^p~Yq*H5|mGZ=*M?O4aHBzTcU?z@M?a#7F2Ob%BL&=2I
zAXJ8FhExkI$m1BP!bu+j(kAQMB?R+xo&R;Q>t>Tbn@O&OS4GPFPjs&i;bk`)W|<fz
z7JqGyANt*y)9Q!-tCm~E_tP0ruxYAfhG_ucb(^k_OZ?EeYRwfDR}Edbg3h~@$oy5@
zA>{O|9R}~^qoaMlh9VhSzIl_zdNwYN*ovN`y%FOST9NrAb<1ExDZ|^y+FJim<8Cd-
z2pn1nS5__7+HO62kv|7_Pn};Y;7CcSc=~FTvvJ^Pg{2Ixdfr{H;DfBN+_KLCBkxE$
zYC7Cg{PVXpPXB8YvPl3*gIg?XSA<8+{$-aMqseZRDed*Gk%7yVu7<xb=|^wg+QoDG
zb$|4_r&JivQWYP?v|rcQcv1GOu%Io-%0j~I{5*wJuAG_ns<M(&*{jW~6_KOg1pw_A
zx^5S4R>|*7`%kew4nQazuqhSFqz<BK>br86>r{9X>*dJ$ySV#Hng|FN9oFRH@`!$4
zqDaNPg(D6rdS6hF;pm<jbVYZ}Jm;L2J|JLHdM0=T!};Nod-@FJ!<|NmnNxJ%4t{A@
zhlFLF;(^q=)c{1x%1;xCDBCD;_dCbdjUz8sgk{$2BPyV5zV?6hYwO!lE}0pw?>%1c
zBSvc|i@0Wot+y$Vl~wy!hpT%KE$VOFC`5YLO(5x0b&dBB9;&*_dQ^tZSX1@&$vV=H
zCEDPso7$QtxH6hdZh3qh2=@Gq^$CH?3g<#N3Nm`s{oHSf*eH0xeP9>cRR1D2j#1<9
z4r=S?wWsJf&%v|CU_Yw6t^1;4cij&?kWu8^y*-jTQHWb?T365Z9RodTaY8{%Mm_Qt
z+*<eDkO)4Hitd+&?E#uub~TY#3VNqw(;IEzRXqNQ`xRH1*(=F;H5P&MAl#%~#(vfF
zi>S3VlKz(|a-d5@3olVy>Xj!>UzF4v!^wK3Z*QOvflXcQx;ao_q%aBTabpE>Rs|Y9
z06Tnu4*)_*0eD4U3xn+Le;8u&PBl$9ZUsBUKMoy70Enz^&U~rW1H|aQ_xHDKrs{(y
z9p3yK5Y2#I0%nlHanOVkuT?HttOf{(Npi*ip&u985I5+dNvi7T=ujjLsC?G2ZqPGu
z>BF;MXpDhZwpFh4vZKCJ`y5&chlq-N)+hkc7+3)u+@Ej9a@R4$Fd^Ba)K_0x4a=8i
z(zku_I{Mi@M0YrQh<x=>w<lg6Id#g~Y*S`-8RmYTg%Wgs$b0|yOwpd`*TotShTF;&
z(*gNn&y|DIs`$ZV^L36NpXZ3_alG_Adh@+=QUF{-*!9}C^RMC8-cgpL*;Qop-1PO5
za;xTjMV_%=>^dbOZGs~KD+1wVjP0AlnR5*=l%IRhMrPz2Vlv^0Hh0B+`~eoS#|ZTC
zu~G4+=H<8)MIw?r?ZUo6oSuck_3r%dN~T`v)yb5KwpODJ?iMcFPjt>6Z}BRXma|@w
zMJ9kMwZTo$4+8dOpEZ1$U*G+EQ~LK~E^6o_O-W7h8VhSWch7)*Wyp*%D-l^>!g+=M
zGxA$HU@FNEqtD>b<Aco$JtL<#77aGgF}v}n6O?qCu2-p3o0dmNuZ$=M7(@*Qe}{5B
zTiwbRon30k#&Dm*bY=Oc*U;zc!u{W&6)h;|feULJ%>%<ca09=0<|nA81agHQvQz5Q
zD$wYk9pMqH!%b}5_$87`ieZl2qtESr1#W6X4%+up2wlXo-dJ5kyjcJ|{rjM^WkH$P
z<LbXRo7uYO<6ecVS>Mp&A+qsn)G`HhQT*2W$xOPb(uuS^0s@V+f_4X$uxi2lS=B$@
zzcLPmJpYWyk&&8CCNo7wd{c));vdN_@(?d=x0Sc<e`!CDlfXxz5K*gz%GF&)f3kMh
z)Y{UXOE;$I)IJajF>L$&gxotRpnf4AY0i-fkp<XR(rQVFd&9#`W`-Z@H--jvF?yeB
zUQDF*#9UvQm_ul2dZd{WZ#Q=E(JS%PXNIRMYwVCQw7YnG<KV4=mm5ifCu8E)8Q*s%
zOMCZ=%R_6*SI#XF<}8}@F5<BcVR^niFJDvCINPTGl#RwG)+b}2q(xNn_lunfOxh&E
z2!QrE<PvU#>vLtq_KhD5O?ubF7$?p8tqDF6XiCyAPS8uT+)2l_x(h@;-xs`8edlp7
z^b#M!j`&WS07ghWMKl`^?V#zE(Wm92%Rj%6K}HVaEP%|TVSf%qm5tYboM@L1QN%>V
zUMG{4&ENDbLP2I=j91hAOZE6gBgQI=bb_<#=-8|Zf<lCCVOYiXmt;FsR2nP<a{6*g
zMp@#9r;T&pRj4I2rR^I(R&vL4A&O(_W>qEpqP{vNej<fEjcM}|>*8@hE9%oM{jyi7
z3U92z7Y1pma!f9?1v_@_T3+_9w;5w|1`4mn(uzE4P|yhKHh<G<hNdciKPMy|UbjQ9
zvxu{tG<T<v7S1lv`dnRB`b-ov9vh`{tq>{A>5eXuO|RXlLC?us(LiyE1JBTI93gNc
zbO&Fhb;(#&d%59kW(fy#=~+UiRAP}*azP~=NNT2p;9;Rj5>2%LWon%VUZFgyxS$CU
zFjA>xu~X#209Fy$RsPA<f6UDV{Yzn<ZHjes(A-f__atzA7(lK#*$hwxz{;jL2~H^B
zr&XoH#e_WWcOfJt-2}~vfWqtonwJ_KmckR*z@8Z#hM;ocp=bsK2iLBJTPuM+1TJ;i
zu}xqVnJ5DV{c50Oxz={NY($8(uhIQHxFcI4s7eMmW(}zgmh;Ju1w;~G4Ij75A4Mnv
zwmV;@<;*WM62VIZs}hpspnlsNL5<CtCtQyD+8c+h9__yF2@B{#VV0;Itv(p!qeb3-
zBM*+gh1h&QejJwk<Bo$tUNb^@iRVXSBx;;++faTI>|%c11A^e#>B7TTIJDw_O&6PY
z+zA+q@KjO-tw~H7q}u4B&%AqM10JVV(K$Gns*seG02mC!l!k_eDU7hc*~)vwKAX(G
z<aYI!gvrfteajoMyUtk~?>~PTjnZ{Da6Nk(8}XTocuFBmOPxX#q1<%Bnybd4luAr$
zCJ-R;K?VXe%k4VEEdT#CFPANwiO$sW`AA&6gtm0jaXq;=vLDO56?xL>dVo-W1#7D?
z8TVZBtTsUR{3FtOP3m7jVUpo|WJV`Ujh%mp!!_V-qyygd_rfOdUxTemP3!hso4O;(
z$BThgqL?DaB{3(t_sXt*=LLT*xxx>O`UCVenY1Z6{IK`jjKiK2h=nn`rc1x!8m`8z
zPz&+RPItgKYN*2FbA?U~7a+-q0&oJ)>xRIN0ar#*CuNb-S=48&XGiY>7QEMpQl30B
z9o*uX`}&+URBWcwFpus+rrFWke<h;9ZuJSqQv(ir_;cf6K;d!DD=e(GM6+LfA2i95
zacd`oY|1}%kJP?Lsq2M0lPO6g6@9n#x+<{4Z~St1CRm9_{p9Y4s2ku82P>TzH2Ll6
z^&?d3n0M_rLy&GBNv5f?vu7|+Afhx_B<yB=hzt%DYr|LJ+wb-dcJ7Sif*&bV43F)`
zAR%XOCUO1MeWLBA%e%bxN1poErr*a*y|16SYP6wm!$UOjbsE}4T>S8<tBJa-p8G0^
z$|5bER(-b_ZGDfNJJlRYSj6WswsBZUpxeMTTzP(;AhRvg+VlGtQ^x8#AHpS??8G|1
zeATQ*N!dqfQF1NSeffg($Uiq@gtau36qpd<SR2y~9MmeZ?y|-K2JC$~oQ3D{{*P5Z
zB~E$B6%&Ubg`?-d)qi|K8h{L@Flee4*BDp@1d$c9zKc&CSByBVjheILu$r4q;ctW|
z{E)loU=S7S=XTie&oRkYDq}GH#Z=U&_@-9&v<KVVcfiRY44YGzk8^NLg0)~*{WI^E
z`X^Z;-CX9W1gThpDI%#V9*;E=y3w|cq~b!$ejuMp5r&7@WIQsjn~M#`ZPC*JZ_;Tc
zWzDJFDbhm1zTt+2Jv%aSRa}M1KlTKXC<zz%eBb?74v_paxg84hD%M(A!1Y*ya-7ez
zeiZ0&KR2uU?SQ8xJCI#BRs6APB6vxgSXzdIMKj<Q#IDCH&<LErKez^6Z3>gX?gGdr
z;FKW!3i@M8SQH*`OQ2=ua@zP-(_qxX#s*nZ74*D<;NaEuRf0vjUI$3DK(UcQG~ger
zEyx2}<p!iuAc0P)ecm0)d=TdV8T=!`IDg^$MieU54<;v0NL9uA-hGZU>QEZbezT~8
zSr^3ApVU{i>-I#iyWfnC{LRtV$}ds#hqpw_1SXz$z4_>LZ;~qA<wrQ|on<m8ROHum
z6mp$E?f=0FBkN7C?YnF{t`U}9|I09Zp148(t$ePV^<~H?d3*O8)~v6Nm24BaFO!Cj
z2O_xHUv(x}a70(OeJnwQAdR&M7q}SPq2pnW6+`<+v)5Sf)>sA1tlX6fnwMmBo?uEm
z7R*g)aTC10)dq7e)gUvA!N;v%0t>i*iJ)H#8@wsAq7wsTb~qcP=)iZP$HKxAF66QK
z7lSDO#s7;#!NHjXuaKf(4HREr6Mw-et-F&^2w5PnNq;O)&^???JNixd<4gZf-Ag{S
zxX~YCdlY%XGP#JI;T3FxBvBSQ1pJMlg73+G>oGr`UZuQQ`PaNp69UdVEg5LAk*`PV
zA4sof4yDnF3^DpmbOAq2ucva;Kl(cv#3W<t=RRJh)X7O`$rr{$K_mo0(eLM9bF*PX
zuQH{Ll&T#)x_7iWNO%j~b~!b)^18?YWXtZ06`vAsCj~(CvR%1YBXyO=sR?|Yz{dUS
zeDC+<gM7gn-mIrJ;?1u1>dhtFYqw$I+4l6`sSCV+oWxk|z8nN3b?uB|Ikm|=bzj7C
zwqTt*UHelBV~8esqZ+!<5L#<p!;L*r5FlwL;dMSYjTNIk$!ua^cas(yMG>4BzD&x{
zQUfi5NO_(A*2B}npdJ59$#HGVJBcqhSR9{5S1~)(F}3mSJoC$3v6+!OaMPzWO0Srt
z>MdCwys0C(`VJMVkR5e+u|qmjv*f&t>GnA4lPVpZ>5b0dOAflpwvq0np4g@qb%&h$
zHRzTd+RssKB4U}kJ0boM>^yYn&Vk}@AK^;on)}S-sjeAlnpy%odJYZ8b(y%QL+ahD
z%S05`-`CE6B_bn*qi?Kp%;ox6cs`ACr`xn5rS^t@CFz{|)-9TfhalPewVzOivNREr
z5z48b@6}Gu`M4%tf>d$s%T#<OyI7gXmRJ~>EY(Vqm>dIV27d8pyr&U=y&a+ppou%I
z@djxU6Q)Pjk*U-*zY(+UmSulc{B*mz5N!H07fz?A(CLe>W+Qbq@t*lbAD)dKxA$W?
zxsoOBS1%}QPp|f?nrfRJGiVCv04GsgT>PP-fliKgf3(rzMBe!GdjsPK-`AcbR^$xM
z4}eATgw3A^bk*H&>&2O&J`Y0x5r2Sk22C<$Wt6hmAkq40<M?u<<0TS80i9aa3>aKU
zX*1xhAS?_JZz5p5Xkr5N971w(D`08?WmS>3S(%wsDRM6d*&k@rV7>vg=x47lMCFRr
z`YTeEN`VR2e~SY=l_O*JTmo+%Omd=a4q>^Iz>8C=1NKhV*4v47Wnf}51xJ-V_h34Q
z6_|SHqTW`*N&5|1(%|pIoSrS!L?xKu-+tJ_(oL<}7XPqSNw?3QaR3p`Tehy%ana!t
zjQTQiII%-a-s;8pEOyA{_XWb`9|q4a%15`ojQ+RdwBBQlM4hXiC_*oMX4*2Cw8}{M
zYh;|%HahjSTu#J7XpQ(rnKSW(PsjF9PYEu!yfBVqV5L5zd*zo67>-{R9L;K=!N1=*
z&lPVzTfL}?jr$<1eNI5s3R^8V{5ozGy)|=q&gREY@kmrm-WHbwxi+J(CavpT>W_(}
z1m3_6NRk`aMUt4>4@-qAfJ`KJCfK_kT!UEV*LTkZI{t6H%uI;}gA2em<jx=csxn@Y
z4SD|`=iMa{ERhW=D=OVjOo$X#>q<m+_o=(4@Nt!$?B4+10Dk{2rieHhl4##f$`Gb>
zRAy6g@oGm>1qIA)P`|0d^wYeDxA*=c2x9K$PNF1wNb@-uHd=RBQAlER{gTZ8z8J<%
zbOG-|?}xP>vffGRFq~~9tsX2X@kq7^F4HZj&h%RMTpw7i-@dt#!vpEVkD1RK7zf{C
z2qJ-AZtV(5!)6f`hU1L(=ozDTEpv~RvT%9bo-*JiXfaCH5z+7iiRszRbr1pBzK8#L
z+u_`3_?whX6-k4pSUuCMG$Zwvo$mnTXZJq*`SVp-|25=`HG*F3h8!}T4q|Rc=xj|=
zSdJozw-Ri>?Y>`6`Z^tRpc21h`4fD{m!)=prjj7vvw`Db5lHENk|Sf{%@OkD`lD)=
zAug@w=}{Pkh*IRqf!dJ}IVbWQKH+kPq;eUgL?mFKpnjIZ0eNGRICU?NN$XP<YoQ}~
zk&}#c*;9?Q-J*%HCb`YbAfNb_36io;W-#*!DwJv47y2ivki(<+&|t0ND5s{wWF{#Z
z#Dd^Wogot40-rCj#yqbQiO^ry{uomv!hV8XK=}J51u}L6Una*&hBQAt<%d@LdmZw>
zANyN6EJhA0>W<T%(Vtl(i9Mp!fGeUa&CISsaKk4IvMNHT;u$S#GMV2N#qKr8LUCkj
zgBiU)`J$}29?EfP42ljwd38AwFh62uUW7-f+fl+9>h0Mrg9Wp`%gKC-ecz!*3zvP*
z$n4*xNKwY5#pG)q*}hXKgWb^-$M(8Qp_Yt-n@Orx*L*;l&4FE))OS*$l5;!3U5Z`O
z_sZ}?;1lZHZh9V_kM#Mo!k0WPWR&NF(+Tf0N@~FEMTtiXz}eCj0LfJq7tjF^nFe>8
zYe6Utsm5ta0RQFSUp&c)J<k`axS&(ofMo_N6GZ!f0~cIV9Gh1Gy;u;0`2YkOxa$N+
zRT8`n%BTOKNh<PZZ^a7i=Yhb_;crpkY-t7(4siJwmzG96u;?qaKt4#S@GD>xfP9cB
z=4`76$fvvRFY*EMENI38!~$@>Nw>G|H7)>TK8OndGMjkKnEkr1={K_`Bw=>;9<U4t
zbC43~%*X^SG?WRTPD7<?X0F-gjkE?fBqWJeS&a@CRNi^E1q{#r%*9W*m?Vv(A6!jd
zT&nQqYdvL8Xk%&>eoN8qxG==ZTx06YRd_4yv*-TK*uSYKN1)N|Z0+1<wSjMcsUfPN
z9EzTu^%Wy$JwPn@perdSq(bmwA6xxc`EU<bN=QO?`23)aKsV$`KWrObaeNfF7&mQu
zoZ_^%4!SCCl)Lyx5Z&<JHbdS@{x;6he%F04R6MS;T5IiD(Q~}rvj2YUlRE15NdkSX
z7>glA!orYH+Xx3&Q#Z4|G+2$_?K+Y76KL#rRlLuK+4wMAqbrTuHqO?#y$}<QqkGTE
zDygBDI=!<Sy^jBH(6CYw;4oM27J{zTg)D1yHrznZgYxOe|Jhf)rc5(X&#+3AE8B3*
zzqIz|<zyco>P{%EwprIKh=qqI1eOt_$d~!*lBg#idTF9Ln5LuRqU6hPzj}Oodl3~?
z7A~aMN~YyBJQ_i1_$`xZuea6X!~HM*?^S`qyM#&5i}EfMuRG}=$jX)0M=IpBej*`G
zW)i!6OcbAmCMsfL;K<!ya=!8QW=T!WQwvisH5p7j6udC%c2$HOt_pXa+#X^UCyI2C
z(eUyseVUWRcx=SVFZIP3<G@#&@Uq#LYtftPYizbg7>W41$KvAJ4qdg_=v%LEX=mHc
zT}$<pmAxySnjdpJm2h=Nid!W?v8!z_n6nmgsg|`8z6k7(wavR-D9>7*k0Pn!3aDBv
zz4Bb@51<O!^X&)fV9X01*W+QHV*opF+ZsxrAr-(BQISw9L<wQV(GhN_!;c|~+>uMq
zVSZW&=VaNJ<2ctre2K*7GVlS2CeLKuh-O*x8@Y%(M`jywd+p5C=(DHhA9F7Xe;dJw
zeYvU0SwYIMvG+(H^GA`$w|)$$xuW=lq@cD4zWGxg8lt&;4!`cN_Bun^hWV=JV^@+$
zBK49Iv?6~YPiHK2jjUH^A6MWPX+~<v($EPs4-Y;+MMAi})s4;R3ZA2*^k#@k=wX9A
zrHz6ZJ#BJWPe;kwasBg0GW-}3C5%Z<t#=03U{W04{M%-*)-aT=;gAVK)*Z8!{Y)%<
zE2>-UvnWgHt7Q2^TnXY+{J&*LyZ?FtqAeoCs$wa`<v8NAAF&?(_(CKy|Iu_t0>SUI
zWc42vg^GeH+XO9%r`nsJZLv!V&3FU>U{?uc%GLuF>uso5#S+coiVk(00f%b4vHKo%
zk}Nno!ei9|F9ihbA1u0HRw129aaaKLlYpcKq!ejfNlry7aIP?<EFTI-#u6dFNl-~s
zpaC{ZaBTq@qK6DeRr_C*Cr_R{G*hp~@ccck3hJN|104k;o#3HnLj_tf$PWBj>B<G(
zJdhF-5)%4{ZD7E{At*S+Sk>MW&Dtpg^aOC({-?!m8_6F~SY9}_g98a%gi5r^L1sr5
ziVX-l9{Zr5Q^LuKb<b0vot71u1$6Dw;erFL&xHi=jE@DiD)azsUxyaK70A-SFb?>%
z;4bl_77(ma0p2>hvmP9oVC?Gt9`j=5@9o{X6s{z$q>mwl!hkQ%BbsO2-{(}2fIr{g
z_LO)p<nh&C*)#Ht%_)QEY@VfWx9al#w*yZ|1R8!wuS~vM7;)XcLAW~4StUS2{?<ll
z*p;a)>>P4`+m*HcV${eS-Y6;jch$>uKJK^zR!Vfh<9#l);C(cP$k0GT0JFOopvfpP
zox`t$8@+V*@T%~LtGkv*0^f6<!Ig1h0iB_l{+NrTRIYX+#2isWbL(vHX;TH3?8({j
z;^T-q^}*Sk(BQrUf6h^{y4;;~pEJ_*UM~{QKl6(5vTKOKd-5}JmGGHA75}TvAcKzb
zN7;18xtvZgg1P@8umnPIsofKYM^y-A#umy|D#D2sTS!nWa_{Q9dB{4#MGEQi;c_$L
z3G~?C3%n(JoBvQU;)_SmA3D=X(4(}HMVSc+<qMy7-8mryot5UR@AgMk_B!2r(ggy#
zY}k72_fnhHQJJCWHDqaNEUvJxqO+aZv3aF*hwsR{X#*vHW4wNbv0r(3+pl=HRrAuw
z2eHJGwUpWuXUDVPf(BuFcRD{0)|%}`Psbqs1JywO>f-Jkpf73N6D|XFn0N$9a^R83
zv-QY9$66y9;!MoX+mDxnn_gwXX?sRPlbF_q4fLHY>X-EAgngC0zUcBKY>(Z}zu0T?
zdm*Xn4Fa<l!z(w6zz7K-hpQN%)~gaz=8*(MLR4hTTu{2m_3PUbi7{tYLoBq&#o+H3
zIh@DW@?Vl6p)X_n0ynv(_J0<j%h~&^FgTl>>?7DJNh&t`|9SnVJB90S0LlPov*`2w
zs`WNq1_Fe6f=rpr30yZsEn-FN7S2lAHND_GRTB8-ua)@Ah|ydVpCF?14}*(+_$`qn
z7s4FU#Jtbh2aMtb?k{-<&!;|-I=HM|>UvL<G*VXYbv}NoO|;t5lH64GQ@G>S$5M`_
zZxx@)&NuH@sONKITuk%F%=uKs0pZi8pXQM=!qbvPkW^ww)G7vo|A;Q<IW7=-MAXxm
zE-=R`rhfW@1V;1tq{o59h>h=`{doJHsYJyCOZO~+#MuJNMy9vPKPyp^JuRQTE3~>;
zWTjbby*T=Ph`jQ55yyoKKE(BsK=hz<iYZ9AG+yF@?~fvg`?@c7fQ%Q6G&Gh1cx9Dn
zSI&3#(4Oc^OG^tl5{(|~bog=scqcH*;Yz;D9pzwS>juL|iH#9Y>dEf<b>~W?#fNWS
z6!^tO=y)H>bm*cUy7B-&_I<$*fcu>53JNOCrKjA*u49S@kwlPQd>~Q-`_qHn1{^S8
zx#c26(97n1`1P5<k|ki<{|AAsV`Q{-xY`SXiJ*-W^!RabOu7&it9>51{WxH5-LwEa
z#bD$Cs1YW7Fvz4C2c}P#YZsO=fT$7Z?*PDVu)+EQOvpg^aW{kk#TcM0fopijow(oS
zKbkY*U@hRDY4c_+qqiDHU?PK1WyU_?tfSXT!}mp(zShU7I~0+RqvfLwJ{lEyL4WTH
zdA(}TF*_(H@V=prF`l37ryBjr;Ww(v(r_``i$4+GHs`&-dN^tYy9l~Fg5IBYy&mnl
zu@8B+zcey6g&lMb2zT?(X6oD~AuGOH)JA-svHJUqp_?zJcDYFUJH!n3av0l|T$gf7
zGxtaFkwX#+hdh=l1KS=gcznzXjA0Gh!-QF-HV4RO@4=%^@Lb;~iqdGCPq$_-zD`~}
zf<ahDMqDoZW6ZY*?YNgAb0jGg+wK=0am@eb8n@;pjDKIt;zXPOXPkL$MJKn-?P{pj
zkMXM0k``N-!YdIejwEFCHorUfUUz1`oua$aaMe%0Vz5UCnkq?|Y#Q35$o)bX1rccR
zTw(UyQ=tkRi<|PSVMnnUPum>BSFS38b8>hh4gAXqrRa$F=5Y8c*wdX2!m*;P2h1;i
zz{FRbtdUexfU%DB&7~m7!MsW?RSIjc>bmwj?CT_{Bj8BRWUnQ$wJ#&i375ge=~Ll6
zzGbU)xroU!c#D}+SbncCb8X<YjC|ln-hQU%otL26KK=?S{>}iM+rFDfcX;S!jKO{d
za{Ne->RFd>M!FgfEJjdvjm=BYA7kCAed7)XW+ex`mc?ZDH|mgC=%DcRfQ{<Ul~N#q
z$yxYj72^R<SYiYA4IN6MIKhWRSmekH_4K2^bOMLJW<~Tgi^?ToTBL_KAiDBQt1zbz
zW5jR-riB!VznS}f{xY5C2ihfsQYNSSkS0vzks!v*u`?5au7Yajk<TWh+csC20_Es$
z$9HB7k&MVwKhaBbnJ0~^)2lA*yGnCoFwRU{(dFhvIs6X==h4!_j+wR4dk-xm8xGWD
z&IWaG=q*Q@td)Ge%$ru<b@+sSc=9$@<fD*4J-@V?7T;^*^}}>}#||Hyx0UL8AEjM9
zJ}4mcJFZjvf56?IWAe_UqjZ?*m7#Sernxi=4@+TiV~(g_e`%?qmneglppe_Mor+L~
zPl3WNx&$reg^+m{JZb$v4w=#ywowhQm1T2!smoo_f|n@w)I-K9e`0McI?0}7D2Y`>
zOU-A`4ADnC+Fq%*a)Fom`4V3GHhn>)lYhVG^P;yDKyZD9`tB4y11LacWd(F7iO>Y_
z0E4B*@NkXio*t;diiWp-G3p>5tOX@FU=IYQ1;AROi<-dt3i4oFaVtxo)8vZz`^j?&
z58i{~ujiATz<UQc*$;&X0pMN*?pd0eRn5wjv}3Iva24E#^-_WibQVAx!3Y7=prhjZ
zbNQP(<$I3shzR7jK)hi?eRp~3lbV|Lqlg3i<bkOe{4+3qfW>^~uvrET&~QMwz+}O5
zw#tM4O!c9aPzuKJ`tJ{_e_WfO{S&luJmjnouCtN)K~Nji0t1J!nXhjKM!>c>(%Xka
z60l(&cu#s2;SLLQQGXkw%oCycaNp8*Od0PgW1+Z^{K*$q3!c&nQ6$Yvl!+85&6N@B
zdQ)&k06&f{#+#UXwS2$tWx|IqJ<9pfuzHR&$HbkYx7;dqyz5fXi?yoS?%Z&(4JI6$
zWp_9g-7j0}E_@y7<qA7{d3Ux@^-8<3CBTn-bZ_((PZuQci7FF?Ji94veo)vGChWFW
z*e~mwnX4Z0la|80wskv!j)a_Ocq2XQNR5Vz^J1{L)!<@dK&bNlL`PP(k;yum-K_H<
zYmMwUCUuIHaWe1ghVIDY=jN`oM1)4snM!M|+!b1F>s$Y#7ydUYd5P7G%SZQTQI;R6
z*;F$r;fTp=5sL<ooA|eJ_E<69;pON%SXh9LDv$1zJpOI_l?#2MDn9kqwfK?EAU?49
z_Y`&8a{bvu;Pq87Ip2SD+kYyQo9X!W^)<)G9Q3f7{Ugt~g9$=vSxaX&+|EWvv^O}U
zw1Tu1TD&7#@}ih5cJFHa1NuLbvn<tlRs<ie>-`#n^?vFrrnYWC!j@$cQtqh?eJ2A|
z(a6?!3ZuYeA?P&3i5PGS3SqrDFVt9;bIz-cKE+r$zG@!KWq++rsax-y0dgc#+E@DC
z$|s#XaDKn?{y6+V5P4I@1-njw4+Gw=c2V+-xZl?ov9fjME?1u!<ObI_utILa(TW@~
z>1mSwT&CF}OC+2GBk4Uar&r^*OOP9Ygh!M<e}B>ZHjeem8{vIy#KF_4K{k)(;hRDH
z;wLogb~q5S_dhPwf`9vIlp2oQ8m6bLqgigHrUy7gV%#oEH7-p?WN65q(IDM$ANQwT
zEC*#41dLeTnvu5*GX(SsuQ4>kgLeAU8xkbQEv684`<_7&5XcP%d*dm}KYvX9v9_b@
z8-C36OQF6oV%eRp4PRHY#ECbih~c6nj1(dh8)tlcq=-pPUSp~HsJ<YilO|5bQ-}HD
z1l2-jgW7uGhuovY53JPAcIcg`X5+s%@^!OXzj^D?ypt^6cRm`z#u2}jDMS~;->qVz
z)D*<#v4~GZ@DwjHtZ6yiaN(|yTgSsqPL|j^k#tw2iC?+d-}ih5t~-E{2CDbW+2?j9
zGjenL$I$HiEhkG9byCAhY(T<zV*?shKr!gr?^9gH0PHQDox;E-w{l1vHYRGRiw_>s
z2m8vqXBc4p3sQ?kAe@+`3!1YAH^IW#vQD-_3y#D-epldvfq~)GIt9KOe**BN$2?aj
z-~~m2B{slO$1j%0H25H+C_r@q_jQsJDodF=>IR_Z;IDgc=SMu$kVasTOHFbjyrxKz
ztEsK+IXq+owI86L7+hw$?ytL)i`7IEp&(e@|7vDo0ptu|puZT5BY;NN0(6?@+p~oh
z?;|tX`5L3DNo!6}-$BLOVdx%h4k}Tx<hNTLrDdWF%Ew1#2-P(kI94t?<KTYVQ6UF>
z$y&rZjgs8O<Mu66!_<GjGb}g3l#CX=f=i7vN({<fK%uE`+I&`qjmr`FnX60w2y0gm
z!9j=JvnsNdAL=WwhvrvFnjBw9U<$=%XTeB$V%W!wH^%0C_A>-RegzR^ROZpSS}|!$
zGk88Xnt`W$*xD`^7%YHMiK>fG973@N1Tzq7KT%vzCZ@v91RB{~YKO}jo)ocnASHg^
z;*u=#eOpS;&ZoZGm!{-gwGqF)^dU*cRCgqfc^@ET{-0l4_h}Mv!MB)a7I?MhNW)r&
zFFNi5YDt8pq=PQR5uylf9TGy|SNeA!-$QC5Bb#ngx2f(FsANssDOcy@9*xcT=x$_e
z)z7F0X0QgG(%~uRd%>Cikie-;KN;|3tCCo5J9s919Zi>cn&#(!M{mHWbyUHU>eXch
zoR`4)*0uT2xCDwA1h2LWy>q-2zgA3hs!AfO;?Y&UlBsl@;9o2dT6aGDOg)i#C5!57
z{wsootb9i;!`^K>k^$Dm*ZZsO+3aP4u-a#c=x<7_-#+Le1i#xJi&3~-2xy|;H%V%t
zZO%Ebv$?94(EHT~WOm!EKLaJq-5d%T4Gc-AZ$NM`fyw&jZ&%A>YWUh9LILYbdg15A
zh*~s*OvB8RY|O$}5j=U8Po~~RkFzGlb2Pt|rjUXu1wV2(7EzP<I|r$=O0$_!PzykO
z8h%ALS{bhp4Q2PK4@Zn0{9~W}fzj387Rx++4<5lb!S0DU#xWM1=SK&s3|f5L#ou^)
zU>FsKdDTKfWTssY1rywj`i=vp6yBCts8{NppgdJupsbc`ZygrH!|g2$`a#(;OVOfY
zal9+)`ogz*L9<9QTA(oo5c?%oL{ov-mr0vntkWSJ;khLBjj^hVj$7q{PG#lmOT5t7
za`gSh0PGl;5KtaCYB4;^KL%O57JY&?^1#yp1PTzk5obs-1M#f~@$pkDl5VO1f+VzC
z0^uVd3C6~Ho&hB9TZ-y`y1qY_M8aD+iDQ&rg9R+8t^nMF0y=;^0m=h#-2>edNUxw7
z$8&#i0Ca4E7|$_mRQ<Dt_e|SL-^QgZh5PLEKP_5w_Nngt!}ASx4~j|w-2-tAaP-0c
zML}7)@KgRE<F2Z@CMX3AGbJQFVsqV-j1`>m0QZ{Lt(5?JFT<TFIm!9p(=Apr1q<We
z2R8~hX$*Omm818SBKG$yess)`F8DM&XT+uuFn;t1VMRO{+MU1DC{wI)b(P4;9MY=$
z-F^r6czrR*IMsfOk1W74>Q-7_(0PWpJi^xBYef)Hl|%(0!bhtD7gN1WM+C;pgW8>!
zC^&t@cKs7WESduw(e@+<n~^XWeVUbSn^KXUW`f1%b<gPUBr#mrBbDbCVf(^AwpX;J
z6@7CEqC#?_*XK}UR=S7vM_so6VBZ47ReCUOr`P*5<rzYuz{JrK@?ukd5Hx}vEfF7d
zyz1jwgqDtSU0yqpZxJk8nG{Kji_SG-wYDz*3x@r#kAX5HIa|q3ddY8|2;U!GoKlem
zefQRvtffGThwenUx`E#G3N63E>xjYVh8g{$4y}{5=_-%5@q&nTUVpg$%0uG};9WCP
zwzX1`JT`ddE-j_5>8qvx=gOX(psP^zegvVkGvKklbMw7dgbIhL``T1K1M7X-^Q_yz
zCk@E*Qbgwgk`M_AgmwxFOs+4~&#Ful_ajaPP}}SHCVVr@tA4r`%U<V2KzUD`lB==C
zCl6zl(T)x9u5x5v^75}0z0-9N?~%lpgs2vJc_a+`X&qgx+EgpxqZ6>bU{L!CEhfH5
zln53XLry{^*bP8n*kr$Tjm<KtbyLTnrO^hH(NMTYfX)CvSDhk7G$$PG=^jZS1vV5H
z`w}-5Ep=Kf-a>}3*#^0MuEfVOoUGTxX09j!{ji)H2cOeEDS$w`S}2ma0<{-McjFHF
zQz6D7fJiMPBr*BW0Y?&V02TV<gwPz9)a_ofz9g7E@tf1-g(lW$y=o$H&Wad^`<JN5
z(2(DQ-pmQz0mpV+*up@~ON4@!7{W8p78cg78+K3=mEGJ9y6(Zd@`0Ovb9rn&K?K#^
zJ@J(k(=Jxqk$B<R=33e5)q4G<z*~Bfb7ya-c+M77&;SuJyIB;B=}chELq;bTRse6%
z_ju!kPQYpo=6LxQ1DbRNurcKexZ(sg+!g>607xSafL12XpZ)z<5H!*_aQwj=gdLOC
zuwYl&FYLQVOICh*;n*Ak00Vz7{b6KohG`O%Ujbn;5eodVj~>hy>_njU7wl+tQ)P?Q
zz(c>z3w&hYL7uI(crM;6o=qq{<DqL{08JV&$=F&TYoY_o7xXe~!KH-Iw-oM+Z<dZN
z&JEttQBuC7q(nvsdoDU;T-^5V{!@RsV)o(GNCD1n=C*^co`Kz7=Q&$1lns9e1zLwr
zi^DG`-oSkRexA|M9P=J+bF?aEI82ewS--#ALL~Es5i0*FS1D(5y&eh7x%=u8n_E%6
z`5h0cD3NHPp&5`3hy_Klhp_8Fn41?)_PcG5=7|LSmO+w!(Nt=Cu)4y%Y>4a1)vFsS
z>tyOz8gjK>u>oEbP!`-V`Z|sYvEAJ|M6Tj(!G{_@b#luWBUzc7zw;*-+Zn^kN>uS)
z>^QXJ|JNh;uQ<5uGOw7IHY^Yr2x0erAz8cMcS9cjL)KFFm;Su)ZeOHCZPSDRKC^}~
zX*QiFA6hG7)V|s3{^+#sx2KrUj8qA`U!vNYDC;O^js)u>Z^P##lo_ovKH%~~97W_3
zuoSQgcV7}Q*lJNE$UhqIX?eeLlJ3(dcKR|!b1eEXeA!UKMN(lj|F2F=OP5>oMbWp5
zXTHt6+R|v|o<T)vITh52&m!{j&14ZO9mX3Caj{SpyMJsix}PKYb@^k<QATpvVDn+L
z;XALkOP5f@ld8yWKc+L%Xb-1&4yEz^M5ePkO_gN7*od#F#TM{v#>*Ol4P!~QQMYi1
zL#&jQ>U?@B61Qu()5Se4pm?EQKJXe8HB6cNw#b=8nrk4>oY0N3sL8y)2t!aChMi59
zJ2Pi$NJ+91HdjVF`4vgkF{P1~j52fNC?zTqF1EM4PyA(F#d(-&kSv@n5%;!3jf(B%
zp}+l%)W|2KoTxAJsCMkW%Gmvw35$09we(RFJ8Dr^`R2Nopfno1O-@cu@XXYNb4#af
zRm-QhcXmu&TxxJg=>{JjKY+vqIA~e$Spya?2)p47a1CWEl{pryP-#zq5Ify+lo+fz
z)dDm}VnISuQt&(vmSfg=tuuaM`4+;PL89dEAnWA9Li9Q>V2?_ung<5v0F}LMiN<Y!
zhqPr3Typ2YUhw+-09;A{f#m~T)I<Cunt=78I{WXGy7EyUYhVIM3xNq5l+MHX0aC$t
z95mcdPELY|Ef~!{0h02J|4QkSqZgz_#(C7dXTQ#XMLJ&vDceSUGW(PCV6Bd>u4pH8
zyFHsSy{rt19T^)uD^&UrTlaI9{_ZJoQ_$4_^N~pmrIURhY&Y|n%K>3rS`&dpS_BD#
z05GOoFaoTIt7b6&B0=J4b6*c@=Vrn<x~ii`ZzkA+Zl?(@uW5}pzmsM*-wW}$TwEZd
zXJ^1jixH3|*G?|k1++_lSa6|}@r_C^^T?_#yiKpJp5KejtXUwU8Ds|@5P&l(p9X<n
z!&m2CLEo(_V*{lq7cBCJikBQG>jizU8(puNaiEj@j9ahc6^tm-W#R`2ts|M<>2B|R
zj+ZM=NEQTC1CvhAaJeOOUz?DIVjFCs1?OydM(4k)!vA{wmJ#G?;)*ztUt0~+IAzM<
zGy8UGt}+zpFg>(s$+LExq&CLke{C}~dd;By+%KW+Nu{gD;Z-b$BtGA*Z8X6h+{Rql
z{JQNaVb{R#UvcPT*4Os!JaH8DE(pk@$5@&v6#0}^oWGW)`$vpPdsLw%Ms_bb1zn#Y
zqzmFo`{Qa#4cnUIB5I!7uxP6|V2HZJqX>&H{ME1`o<jd(GSXr$m=NRlU8kWVqGeJs
z>Ivd*iF_kVexz`Y3Dpk&y8fj9$;#ruPsHf#1Cl&LFQ3f>JoJEoD0Ab)<qDt1uGjrn
zh76dP5KPQG1>BE*4w0Vv8`}v)jZ3w$I4RVD&KH#Wzi?;kTM6p+-c){J<DGbx_+vbJ
z5Yi}OG<hVMK=xA9QsoiC6)K*pHB~uThY%l12T^@s_AxAtZrVW$PNr)pCP#b`qA-S{
z5+&RIW#n@)(XL{gW=6zBZttjVymrlkV!Nk^8rBhmA~7TPHf6dd(a<G7v8)QBf_<Wp
z&D0l#jy;#`pDnQ;u}-D<V&7TlKGZ&gP7kOoXe9z1%?DJkyL&y=uLnA+Z$~W52L!12
zv!ygOV@ddW0Guctn0Js6K<S!RdA<!X=s-*?mjZDrsH~XgbgT)khc>AEddQeRRIBdo
zSpbO#+(=#6mR$NtK~5a3@*j%T1YGjRZ_|}ZA7)5^Pz2Ax&Ip3FF}4JVX0f3@M7fv-
z*8n5~xLwA^R7$<d!G`~6O%l9==bD_zf$Dv;lY2-10J|6!JVph>Dp-4gTZu6E|IJ`^
z;DYj&Ulky3`Ur^OfaSVSOON{;2xtJl0fHc4i;oEbCjoeWT%0v^4U1z4g~D6G`%nq8
z3GSj%1!MNyRhZ2|^T1GKN~rz*O#<(f5Rd11tVe%~=WBBQQHTV6cg;dFqoP`PY{}<~
z)L>Xkb^uqV;ue0Y%G2xS<5<IFImI*5+|5gEtBaxjGqx_?LF0fuyw2;B=l(bA7;ZCr
z3%tIbu^s*JKh61R1Q^WK7K#{c+?JPH9E^fp2h@!RWYnfmYugyc#n$mzsO*|J`4y%p
zrZXsb0fZ1E*u>u54OSxbWt5rK-TqxI_#sYZc%<5a5(OXB{GWba)&@)%V=L<Br`}{=
z(=HATX&6gHgGv-X*AM9^XL8t752310%G<*$UNT(atvUZEJo?|i4`B?n3N#nOY|gse
zljEG`Z`{^tuG4p+evHqPr)^VajQsYPoq`%35lH%WU@cX7MH?+SKKmWDUB^Ql{Ya+Z
z+htlR@9hO=cqtpjb5utn%j0)GdX*7{vlQ=t)V|Na9ymj??<(xJe>s4Lff{dA^?$f}
z>!7&0rF)nV5-h<(kip&EEx234;0}YkI|L2x8r+>ga1A;T+%3VKpo8o0<azGB?^j>d
znSm<)VCMAMdw2I*y_VC~Ej!Ee8DDV9$eLZ&{nKySentNLeDa7Fp?Rc5KCCXFnxA@I
zlkz5VR9tBiis6}QaSQkTu=ew93dWp$qicH`3RmU8_s<p?w`KH%$0Wd}uqAsrj8Gj&
z&0o3lE3yvTOWFrlSq%{Hcdj3Av}BsK%(P||{PW$+d$q`!9K!=XFxSX#&uC99S(kb|
zQCV<!XW4#BtmX6KIX-9vo;ZUgkkN>bRih%20j*B6I__tYV>xrt<NMqrcHWGBtanjo
z`huz(BcIGmbJC^dtPAg51xY}%-RKfFY#&BSkPfJ|<;;g;Uf^nI7baqLGUr|Ge+m6^
z5)=~JXgZl<YerA3xVSSA{uW7*uwtfXy!?xhC03A7?ro{``^oA}!Glibd}caT93Y#+
ziR@R>HEz)bK;6HCk;RrsffJYmfJPnAIt3WIKyUY-2><{KDh4)FK%)VGmVmXAwI`}K
z|5@eRU+oBT315>0N;WPo$p&o*ut;o{>cfHF4#1-s6F4zFS?LgbyqE^iWWYM3*wQT@
zM23@oe}4~f`T)ltphg4|M~V4MrLDc8M4Fb(7y+Po6U+P2^8o*8`ctLW_fSBL=GPw(
zaA*@<1N;bOnv|1agC}<Ht3NOHe0Bh#-hxAZpjQkC>%D&c`d>Z&&ksef$2y;c0ntRK
zR=KMGasdDiAfS2=B(Z>(Ylm2U8_@LWwU~orUj!C=%VsA)yaCwDYL_J^q!DHjsPCBp
zH7<UbTvJfFPH9wd?V#k`;y5_I8m^8y5e@fjiQ$3@zO(6C@Ow17dRw$_S?^o1&&*kJ
zlb;k4WdX}NevyTy*g`ev+l0d8*W;bvNJ6L^XQZWY)4uYD{n4=CToSzfO-I0#-mbnI
zUZS688?FY(>t6!jq-ZlOirg-UbdAR!#v=f$vCOFX%0Xdnx<wobO8*_xQW`sT?P~)q
z(yd|_xkd5L@eh1|l|`9_aq~y|L`v@JY%7w+q7j~3v9K8e#as10hGahuTqFOXMqlOm
z;oLZ2Tog{7q!LmaWAZ;#>i_ycJJKJtpwdsXP_2i8t}WC%nwy!RHaw;F-$7o^D|5~L
z3e>%Kmxlu59gFbW4Xr&Z2Ui-f0_ISCrdB=73XdVSbx&fT-LFAJk!%0Me6&hyBt_A!
zLxLPR5WaP<V2C_rvp^1xJUsu$j)8vvUCLVm&~rEWcV)%|T;0NUMd}K&4KgGYV~jOK
z;TLFi^zlrEtU539*9#!7R*=ZT>e{mv@*RaB)O-2`uGF6)xY1_VLnwRW7Haxcu#dUl
z?vw^EG}9Pt(Vl+4shE4WcON1g)eg-Nl4RzO)C>%-zWGudNj?bJf90J;7HO0$oW?|h
zbhL`457y^;+?*agAL_;APiw*PrgJ}<d`3cL8<w_WbQ&k?6j7E#9T^F=;4zVwm~Eji
zC3`Iw)U2r^{Q{UOZ0L*GUUOYMy=fNyNFDOaWUY~jpu|dsDL(ht2ijuZGP*57Me05f
zFSFIk2Uee7KSZanbDJ>Av62`0Te``kb0gvub15ZlW+mfZ1+F6mGP^$I1EzpB3*Xiw
z(s*SP9GJr~Rkt<(Gtfp`0J{NrZ<f>4fama{LW&yy-{BsdzY4TKn*=Dy@@awQ4N%=x
zs8{tkf&k?Mgbx6U1F%SlCd&jem=C-nvQKc`Qd3gCF;+P1`5jdl^wu#6jsp%`sW#lp
zi`-ZRm5u#?N;Y8XWxP2Mohjm<4IFBDDti)yop%QnfF|-YAF$@u6LTIYTk`-{4WLc}
zW*zCXfQcUo)k44r3}7B50!Wi$qF09!_t$>bH<Yr*rlttJto-@E06h)>H5!Kjaxnn5
zY^a5M=|vO(SW|uj)C;mbcktnW%(sp)amX}|G92h3`>gK%YH<b})&Yqaf!fIG!_I2L
zLmI^&MYNK7o6mQfA_Mzwb;$7T-JokTnAQil7f1+S0zZ%jpXI$UE}b!eMUx^3=RDc6
z3AUe5ke;5_7-UTQ<iLGKe~xgl^mu__M7aJkFc^>T2Ye2x$7VIOqN|csP|#<w<6STf
zjZ?g<z)tl_&S9uyWO*?bf^hT{AMq{pJugRx0)G|+w!_C=)0ED_r^djAAZs|f_s03%
z=a6lbNO>wLeX!?l)mJ|D#Sb4pu6?Qf@c;KH#m0sy=RLq1JWa`~sf7(mKXW|asfI|9
zNq=k|yZWi*cO>E$K{W6ESkkvl&~$U!iw=vA_1QWi=uyr*mYmOC;XHX%$ha5o;MYy-
zY=75HwVCH&S;YH6Ooeg~4iDqUM3}VaZkV({<lbje46mkKZ(DvOF~@Mzv-3ogHN+Qz
z7OdyDRjE~s@JOQQ-S*M0uV)sMG3;DmZBop(h*8@MKG>%rx!+?y;9&G7`y)JwTpbH2
zRkw9sKObJ~3qZa$T4+Et5~y?%Y!_21bORmp(lDm&#ZYeJ1I3Ahe~~yO5*{mh#tp}A
zy+;y3dV!@7j{O72!~rT*oiMvxUiDK>$Gl{GfA<>tC(5lTnMnuo>Pgs>v4RgKwCiI|
zc`~-yKE8YZVo^ow>WPpvuK7yb{#CB)!G@Rn54Cgf@JmwCPzkeg=+aE_(okR`HK%JQ
z?~ANYh!!KRR?VDTu^$2#%3lRYL<A7~6n-ernyKlyMJ{IkHQZ28pOqw1c(8j_vhzor
zGzah}mHp-a{FsSa18JE=l}(^DKEpflUcDz+X6|}8$2SQiUE|2P9f_L%=Cu9uyRbVg
z&R^r2XeiLRw9hoTyKY^57+tj)@oZ#$1pEs#&peoCpB@KYTwSv~1TTO68n=s)X~^-J
zW^h037%kQ~Y^`2K3ie&4I2DTP?c02ffTcKu;FmiUFFUnrnV=Z9XD~7F&@bP7apU|3
zWJQLnIJJ_#vqX4YXkRq+*#ykl7Qi@uW^V!{m;b$j4L$<IF`u)@=T-bzjUtSrJaCE4
zH9R4U18u8xnO7nlo)fm>!NWkBqc@N67~Q=-Ez>@)|6TVo>qU!B;Yr#1*B&m9O67Ac
zz9jih8E73OG~ak_Z2NPHTGFUPZeYqjQ3}p?GcS{%07(;3Jo-0?evys7L?DlwFU&uY
zX2ep4`|-Y}5f0gsUZy-*NAaB!>1_Z=L60)D>`R4XgCG)V96>?+%kU4`1lvI@^Q+QI
zKK7;v-7~K|{mVBBaC>j3)N;986=X9OrJ0Z@1K0Ze8mwoYW<Zd*!llm6V9s~Mv#C0!
zOKjH9iG+!$h1$sxoR@wGpOzs?j7GeVIpU;>-gKXw-@tj&SC!8W&d{3`TArkHUJ?}E
z39!i0?2syt#^#>229;$U)PChmHhwLt;%ytilH-H6U-@*D-R`n$^O(*Y`pm6wi}v2D
z`ui)x%WE624{oijM`B~H+*qthEU=!ng_#2FR>IkL)yJ09L};Unzjl@`o*>f|0|)-q
z*H#rQLl!z{+1s03qnkn}&C%PD*Hcp|YYBGgLq6$6zh7htrD7?W;^7n1<SjnjwYvGN
z=5py9U;jWe9u*4SL{8FeIpdJ}zkni$df3Jg6a6(hn!=Cf+CumpZyjC{g1=1E)x!pZ
z$Q`r&`q6`~3ulUhvuF|J{CF?}K53bppZ-bGQS?^G>8WPZ_at7?X2}!5zMN|;;gxOs
zHIwzi!^8OPy#FKqt?^}Mmb_R#O?^aB9z36Y@Y^Dijn`Lno46r!=kX$Lxt=LO^);Ky
zPiAj3&pP&0wOIo1=UB?KO}ko)KV}Z}-(rJR65epsEatTLX)^6u3aS@Jr)ejgCdoh1
z{Om4;s&K;R?i)=+J%4o^Tf!?kYbIWyi%!~;;QRFlCW0B7g;zL&N1oL_d|@)PTKCC7
zZ8MgAO4dY4!ORRY7XCbCCa|(kj%}&kuvB>h(<%+)zZ6kkWgfA+^L9JGS{zV#_G9a)
zfkxt@pAfzDw6auyXID*rO5s(!+h5koX1=KS%48vgZO$*8hI0E-scKJvFIUi0|NE$?
z*GU6E7<?vJbEey2WplR*>DbNQ<kf5pf$&VW8Ir+tT<awemz(U&m*vBT3iI2)FPpA*
zi+0`@yr)Z&f4KXqQR`xao^#vd&g)=`ZLNNJp3>C*R!g<`Q4RNhD-rlbA9eM*mhmuZ
z;GjHoW1v>3sg~)-uBIzC#T;517o31C)6}aXTJ|pQ6K!Low}GE3D#wb<x&Le@<~%ar
zqs4MH$|m%k_#dpEJ?3I&W)Wg_3_uvcc4AJHA6+o7m+;)KM)8u>cH%F*;)168<g^pU
zj<@(S=R?TE>iAq*<E-W!Yc^Y^qrH4n1I-r?!RRi2Ec^z8@=F`%Ee>NcSauizyYuDj
zFLQpN-^}rJ4WY(96k=k{b^+skDS&w*(6R+|jac8{NLmKVUjCF=g%SGh9QdBvEZYUk
ziDG0<Ra8HLPs%h-iZn!O$}RE8UP`8P*MK|df6l-n0?P+K?Ov6pQ3Iu$RZ#>)TW$1!
z4xA{80q7Ro0ES-T?(Wx7a%}UisyIb#<NL6RYi~6i$N|?)b7z<qzVY%g6NmuaV>p?w
z@tA*P?Td8h)JSzlkU~uKck5M5#$lt~YVn(7<i<Zl1s@sf{CbIxn?3Qs2N}cun1Q57
zz|~X3yL#=YcRh=k{c2pG>;1vssI2R%QbzA<=n!@N{MHceXw5VIvd)EkBxl8&=srg2
zbdAMLl~(7bq{Z_=Olx6wq@m({_BAYt9{2eo8sJcXg!a=7Q(j?=$1qFEl#VhYRb}fh
z^>UrLpBrACm_;2;q+)AmkKHUNIY#ACh;RKwLnDYcMa}x1MafdvUoRm;jTGqcSvGTz
z;fq?un4EdW{^4Qo^`n$ZktG4ZS;c!DFB7anE5QqrO4*iDQ9m5t-sESl-UWy*DG;N-
zg%Nir2%3%=*Bx$E?FCnI$RCk!i1#{A<%@)%C?hTvR%EUWN{mTCDN3FS4jnCp=PSz9
zZkSJ{T)J3GpMz+=t_y9-PoS0?@Lcb6N(cD6C2Fa&sUIs5jV?Xsf_nDlt+-R%lSRd-
zb85bwsDvC1vC0cT&KsqQWs9h?M3GIs_)#>O3w07Wxg;&+OYo>klP6T_Su<+%)Mi8f
zd&_a>UB|uueos=Qt~YP^L@R3H!Ba0BUM3=Y2RK@zzu;gf$PXPuJkeC7C*HZLxVxD%
z(OKY0f}Bhu>z?SUt7fIkXzuRY`$qxxyQ}D>Y24L^n2~s;2Q6A06Rtv1-z7UQj?6u_
z-nWenu_;rNVQOekPz43@jDc&L0I%LjR()54;hMEqH7k;gO-G`gw<NKr8=rrb2|<&P
zIy6>BJ5GEBD#iru1#12c#*0@kZUSbI!i!VrD0z#V%87MD#K?f>ewq_l0jgGA9(wcr
zj+q1QP)9Sh*cRZN9>~Rwn4PVF79%=d#xY{XCXSg7nttqaAylj;&D$8nr(Ep)7VCU*
z>i0SpWMU%-;FI~rcg=2}w^JE+45?;${T}a!9=D1$Qf6~1!u+Nyh5;#<OFzivTBW+}
z&N?04(vtQBI>>m}G0WRdcjSCHgI3|q2LG$prGob7YT=(LqKwC%ch9qzb?_KmrrHeI
zv|(6<Cb!efCUk#qIdIiz0LAxT3!6ootH3PS0pNIOxQhim`oZ&J@h(f^72)Ma0m1JV
zqBM2z*_y`q{Wvvt<VAeKpU+n3&Z>3PPyfEAs>z;qo^oJ8mu=&Rx*Up_%m!1F9=g^1
zF78be88I&%Y&fCr#7iHz4a{bzm_I72!r67UKD#2CZ6;D>d)T3icsVhM_&Wv|FACnj
zcf!&`nJ%oaBlkj8lq0+!ef#ONmdD(gY(rM2z%LKo3#r;54gEXz>z?;FeY^9CFgQ-~
zeaf;@v~0zeGzyJM)|;M}K4#-^BzMEH->Z*6oj>MfgH3^N`d$sr?*r?sy-Lq31JQ&n
zDPDE-Y8O2k5Sj_z>ve}T724opt5WSF=_cD{fQzx~SrTk~S<bk6O=yJ2YU0^{Hx+$Z
z^zLmhDOSwH+>)3)y;$Gk(9>tt7;{)KPDsOirI{%&RJ?0pq`N?p+KDP+8|6#_w9JiM
ziK0=H5zBd50pQ7iP7@OR#u+bQ={>w{)=BtjUClSV-)md5<Id);E+OVq+jXHT=I$MW
zFy5IN`I2aPvBVhQ@==mZll{B5(_~q*(c*=>epgzfW~TkqZUKm2SXNql^X2u5So<ac
zT5J>`vtYen^iNy?gFW$dg-~>#wU^F9Z~ta`&9rz<@Zw5__mb2{K?|bcl^yPLwefKS
zG%Nnt0PzT)=;s{c;T8u;XV!44%W2cKNVH|ny_~IVLUUnLvj5Y<GQsjW1Xa2Q+@+|H
za%*64gZzA!_*RYSYokJ7oNC*#+ewy&!{JiVFQ3bVdBFN{aclhf>0Yqak4BuLMPqf{
zRfQ^x*#uu`fZo`NAV4yY!N4&`a)5xlRm4hAshkMU>Cc4#uK?i1cQ`b}ROvY$=JDLR
zT`!?Tz}^^iDRM)7EmE6)-&LDkA?!vJzZmP*{UuHT*r3Ib;ZscWeJ(<Qr(LV!OAk#U
zE$Ej+ns~;X=CR7WtBW`dR(RY{p{11^7IIkpyAZh2q+y`t3Yt$>_2udyFV}Zef37b7
zByvuL>}i$Wufi)ah=ds~VPR!Oo~)1{D;Gs_^J>-4NdOu%n^9fPUr3JDwV-c2{>0{V
zE4cj;U-HwL&o`z-of<o%dID)=D-v2Ya6hY6(Na}aRpGQHfbslJL(kMauO<F9yQ^7~
zRuL8&vq>uZkwWmL11Qz1NQ{0?OrCl`W@UG69kcR*tnp2hqYq=)VTdMMS5M<3#P4p#
ztz11M+U}@&rD13r$4OFm9@7``)qit$V5N>x0Ev1F^vn_OyIbc*op0(z{dxHHD(|7Q
zuO7be?dPw&teiuUZLxKS6AOG<B@#@9Iy1@XwZ79&n4OQCsd&W0&d=xuKkBVuImA}J
zR|X^ym-fL7dG9ZhW9?^itA6_>K9Anx*@?IF4OXb%9#A~(lbREZvR9O29r0E907W9C
z9Gox7kBndu<sX)I+$6t7iWx6D4>x@-#>TY5ESQG1uG@0RwV#l3ygh}!0tR-!K_#Z-
zl-R%{`2n6X!6>Nc=HX0QSc!T%>IAVgA~1gj9`FF?1oPoi#Ye=;QV(%Tvbmsx>B)z&
zN-@>A=8#eu8*oh((c?|E>@FC-c1BUW0HAx?Ld@9yFa4Xr(HLi%Q67w9GE8jWd?}tU
zd(H8)>dypoBsj(TzBss4{v6Z{@yQEj9F{Ov^;w0w+1%}vG+$*{r#bMmR>EVlL>>Lb
z<VqbX)GMN?aLhlU4;=^%0{=>_d#DaDACI2LL<uRTU&9f@3{vC&x#o~vTd5#cMb03c
zL5`_zYiyKNAbrgqKZ=Ed@9}wf;f8ls4td$_n+@W8*QvD@>c6}0Kkx0T3x4sc)&dy7
z`J?-y(cL2O`uXH}<4L*0d2SaXyHVqR!aV?c*ZFih{Lt$!<W6iRINfta%{aWS{JbMl
ztMsXRr<qv9!v|Fl_VJ;gWW{v=s^8`K+>&9|+cWV_p%qH22DMQ2)&#Yp=<&nW?6@h$
z^gqh&(mie-WY&^7$<fSLWeAZ9koA;v4Nr1|Of2v;J=CZ9t^X4mtA724465Zi$CB~_
zl`W|crvKy{9*j!t#N_m`AMwUG*x3UuwPflf*fEWap0_!vyi{;rO>>%G;8pOvx)wy|
ztu@VK9OkVK+H5wRQeE0Wi&w!1*G^p0yVdqZF1w#yL?ZHW>%!qbeCT%)%Et_W{KgWv
z@@}q^rk@MWrv=m*d`f;sYp_TtGf}C>pZXw=vJyN_Dog`}KQ=x-z?*=>tKdbnut{q*
z5B<bJx2Es@5U__{o+^XEE_yW~cG|%q9eOs)T-|$xvi9ExvgTQg_ZBEomrSRIa8em%
z&NdqH-c1+KVFFessL<qEM%3xbjaIbGV`jYn&)Py*uSB?M?oPj4{F$oI`8>k>bZ)JD
zD)O|<uO=xBQIK}djM?G!e+0A~{FN*C?{f_N6W+Fkd01XgM4j<eA3*e-14#oAiE}#8
zml!M~LibMOrg0`!w|RN0^*=1ud-~lQ4Nwm^%;U{?MIMRx*i|g+R}gQ0{P<H)dq)4y
z4^vMMqKLwdoz1A0Fmpk-myvoq*qVOG7DBFwCnK>{mi&gB?V-M7?=cgQ14&0dd$Cdu
zvzO-so_OmPNv7rDnu1crbfLOSqn{3&oYy8OO|T_7g`&#Kg6jjWL-HBB%420XC0Y#g
z$_dJ~e>5IR@$PsMw^kQezK&qTMi5T6dZh@Kv7!Ji2n<Rz2HWcJW-Irn;N#aMjDpVs
z;TD4|1OvC1`dky^D@V?i@WXi*6O=epda5&At_n4459kYLe73^LBg&04%qU1J)aav0
zYls~bH4NzMIxq6c<x4|DL&bl6QvsuDsn)jBMi{TX7EWMXM|?5Jo<UsH<c{x~74(N!
z*YLS?M32Yydud4)P4;3>qj;O<uZ{EWCg$CVG>PYYT-+-?(IK~5TR=qdRFI@EKl0xN
z)>ZFQ%O3FeBz%x7(?-jA9P?5bUO(}BR*XKBra>hnFxr%Wy9g0!`Rea$LGks97<#GX
zBh1UMy|$d=hklff=l}FsyR^~TA<F-5+<ei~p`ML73kh~>`8ZfC-%lZUb-UW{CWz}=
zTuT(&^}I9fbEPvL$B=~tm*eXxf-c;c-n80yP`MH`FE93*5=RyS*;1^z=e-|IRiU&U
zm`c-M%XDjzWSW+nS?*r>&9~=S1R)~2I>!G&CM&=5E>w^7pEa2F@-jdnff5HyMi4uh
zU`duPhixCh&cZq3FWeek!t*8}YY><Sl%Y!Gq)OOw7>psIQTwG}nA{y{LYQ^qhhtk_
zlU{uo+-vkkui}yo)dyHQ+FY_yrZGp01196dh8{LpP)DS|rMlR46(mZ0t?EnoX`FYn
zqu)!vM@u)oR*1sgQ%Y&}bVUtwdObAbblh1TDJfr_ZzwepZ+UL4#uVw(8DN<at-^=|
zA4vLPTB*ai=*e%eE|^ljR6jk=KkBoPbggZyf^$Y^r{s;4fK}-$sH>w8y5cl42$8kL
zr>wj2rLcFH9Ei^)DcmNWM8p?vy53Uu;9)lBcW+Hib4>WhD4)rE-Jix8U=Y-HMDgGL
z#?7``mHMt1N(mkyh)VK&M#8di3m{7FE1om}IjO^N7w2#C3M(__Yj3~i4U>jl)Mt6J
zRJpHu;S(apoIRGf$1|xr>c+l=)(PI8zgY2vp{QTG!+mS*3cBvJ+{P!-NXTt#C_B@f
z#}U5O;Ynlw-yXzrvujoQ64NCq4A97%`}}e~D(#5WH{lBPaU1PCR(hoKv=63%83=qQ
z!mT#;e88OZKXDeExRU<$y*%VoHN4lm*EtUaIsPthyzI<8?<$Kul>W(GjL>FMrIX?&
zT(gQLzL=H<>w<nTF5pDdx1lL*+_SeDVZ2gP>`xTzdA&sqz5V%8nS)(Dw~DIo$i6et
z^N8Wp2awZ3!Y4@3>1p_`HY^zkm+uw0<06pYnq(sS*<z3a3)w^mC4mu`WITLbSDug0
zUtXOjE38ptweB5z^dsR@b4>3-ur$>m%-UM~m(M0vy#-?LeNN#=F3~Auu1gHL`uu(T
zpZzEi`JBgBXTHx#*{0Jk=_OTJqcF-W!Y{{T;+XIeI{PLQMh)EpW{2rrt_D=aRX<3n
zQ#$qPEU<b6_cP%NRZ%aaYuPO5n;Er9xH-M~ZXARS5wxG!Z&Iy`h*DN!XX6XE%-E>@
zw;LI3VEm6agvi6`D<N}#|7ZN@uW4z#;~eR9Uq|cgSB3xC=a9NSUJ;F+_ly<P&{0N)
zay}i4+i!#pdwKY5z93%ymeX6{DSU@6QmoRb`CBBmf_Z<3v){62an<i;w7zKbp+A+R
zUr@2Dj(@dA?`7h)b638qOw43qeS(?*((swHlY$}&oAOD?#x|YTW=R!!g`(lR>!$%H
zZeLeKbiW;8Vype4uC@?U!H6A!B<F%RXvdIBB#W$Hmn&@^4abo8fq>qUD8^1)K5gFy
zfstpX-g4|WU&A!rGCBXSu^!LYi4Y>K<O!y%xV-oXYZo3w!KWz<Y|m5Q*pyBBrfNc(
z<S3(L>9Y#N+DKgKSR(V)ZVCk@+zGYl*;J;Tj6-|V$3&flx+XDyjijC8fIPPM(jOgc
zzNVz8i%BKPNQlqFyVwWX<o(cGDf-A4gZVbm!Ybd89+duOdTM&u1l@kke4Lm}fGs>v
zL$k3o=-u$0<&ky-&d}N|F(60ZYG;BxxcFBo-_oL0i591*sp1{vmt+bjJiN`g(eeso
zM0;e+u_WFZnyGnUs{0@F>pVe|QDF_n=x--!y=RppuETHbI<3eAu3{$V1%QQ5xM&lz
zzRp(H@;6bQcXia$P>Tl+s=u3Zf;OOeDM^B|-x_0tw|fZ!rf-vGUH<;fq0t6PVZiRF
zOQI;$z_Ykdng9Bv<&u4+#qQPfdBIyiD&H-8)iGG);U2NYOOj{x_QE@D)o-VJnL+hk
zdTht_Yo~sCgXf{)h2uXwrl}0L=pr32r`?(l>hyQU<xQ5*wj*5UoF|G{QKK0&@mk@*
z8IMcLhnVqSoB6NM>W(oA#$NcqbM9Z{SZmwe^X4RU4WvGA(!GYhUS#9;_=#6#`}X?b
z3eoePs-{DE>6TQC9oTI60DYM}nY1KTqs`&Ag*s=1X{S>m66o7zt8}1ws`v(Nc_Buc
zzSwCoAA`kSXyC_ZHZ?7ARO{ddgjOEerwNlkOC7OU-qN({L{d(`=ZoA$EUi=c^rpI{
zE^BfK1yR~ZI}v}0{gqc6_8*W-IwJvPA|1arhi=2KT)N6{-ZcZ``bZO(NTO?4$kMDm
zUQD`mOb#*MomY8vnodkSJn)vC7dy=}ZN2L%+pMAhO9f@hguXi4o^{p~2-$FnUv4}E
z9}J%4B<e-V>)Wlh0td^Z+)tWD22lu0(-dq)QWLrpL0xoHY9^5Br5C+T6T~LPrm;Dx
z@-Hrd382)Xy`tNy>nal`_DIab50RrZU$ZDvzvgO}U6MP4r2nbB|2>j{_xEKswLN3v
zCQ0SEWga(NCpC2JKx{wGF`w+5rJ}2_**=ZC&%sX6yGBZcJ%;(R4uK!2sf-re=~ZFM
zo_5n?6c6{;Ih*}Vh%K8Le#}R5upgnm?cX+8d1{U!p2{58&o0x4{5ACZjBgE;OFAzy
zUz<)B%_>b<RUONx8tk8l7F?&GpYibg8eWSbdIKV&&|u!o#Z+vZYIArue4kLl+{+vG
zPoswwn7B96cMg6yRp+)5Mu<@Q5I`wE7%||Wx|5Pjpcqb#qqVUFQH_ZvY0^#oFBf2I
zNQX>J8%Zmkt)8{4aG+QD$Pl^)i!x<MLZSK~mBBB~3+iDtt)pP?y1A9gpdcVq%u`9G
z%`j>;&05x|h<Xq5nKg1WkAa{j-Z+Ko-E*l?NO(&lI+4z&0B*6I4L{EtQ2@l&*n_WJ
z9U1-o_!Tge4=rCXrE#h^!z%Q)rv#xGsaJfzq6pHKF~fVEQI5s_rbh}v(RTZ^Gj*{Z
zLa3oPH8j<z-*}H%mq?8mt1vOm{^M+N6Q_c>dUw4|c=E=lT7N%|#g-z)s@I|=N<l7P
zw>p6VJZMH!rfv&b?y0raZBmE#Z{4Z+UqFFhdVbY{Vs(f5&h-7;qSOQx=!wm=lu<iL
z3@zxNp9XO^FoRY-4+)Idov|51T#~)*TcTW3S&Z-V*q+W9M6569yzPcb6^)|1o`aV6
zPwa~6a5ut~ZMIY1>HYdv_kJGM@>ECMdAAu2B4lCZROj9N+PX`aI%T38N}hGUH-rj8
zSy_m(^;_$>i_uP40nbmi9b_u+XJV731D^~8O#ev;O1Uqsdyj+q#mS8Ecw-_N<?+0N
zi#;NWOQn>57H$nj0n)MMXk@*l>9Uw}T-piLSSTWXcckEHsT4A#4=I#tlKrUgKC)(_
z(%4|0wa?I4=TQj0mp0C8s4aa;f3|3%xHvme6+KzUj{pLbwKF(zqFzR&++k**{vTv-
zF79tP3u}}J&aoXi>n8yf)ll>WdSd#l7Eq@1R?By)Q8t+e<_PCG97YSv)uD^Tso+4Q
zRJc3dYYEYEh>;ne@g^4#wmd_>F8W)1x%>)kK6y#s!4z4WMf+J}pyo{-EFemYD?V;d
z&XZS4U~&%6eKp~Ss>vIYVdFM5@b?JYWdOy8XaSQ|o4kK}TC*uZf3u_iU4E`17uuln
zdwad!{4q{(Y{@GS@n!_oZZ$_r`A-*4sBFJ@rS3$8z`NrrJ@#0jEi{o4!8PzSDKMY#
zg`iZ<cD87>^$;*xwxfG~_`Q4krzyMJZTycHSVAc~HK1}!@ms~+8|>pr{AI=GPaha~
z-_gf}nps%L9r2ZX?B?6#u#g~kHuyyS?;e_4P9I)OZJfu*OctfW^o`9WNJXZI3p>9_
zZ)fr7o&FM6XgM8aK3TOs)_NTkez{V{ag>}!`OkUctRFBcPOVoyHsk%MKBltd#F4;=
znWn3(KoGP}A;p&a4`ZEEp6Ez`&!m}?pmIN>vr99wEWF-$!6CsYAjKW5;9U8IQBu`4
z`iEn4`RaP@dxb0Fa`yOPBLO3RcHtxB^q?ZR=~~fHr~a+`*yD{g!}z8SNR-MBmXAUA
zwRsmzI)TzfPaF&nUlO2`W$NPsYO>l4?!jTo>a+{d<&u>N9opc%_4%;owHGS4n(r2o
z5*Qg(G1BYoTVBMN0$INoPO;r=Mr(v@X3az@FgDTtr!o6?jX9hFXWF;CypgHSO@*xp
z>cx#RUjderLqk`Q!cVG!aATL`;zr^WHkn8P{MuNVJ`pa)J9Qy`7~dePj@b*%&rvVg
z13GqK_12N$bY<Vu+4a|q5y3u!+{whoJw#%uZyEh}?%!j;p&g1m8+D#047A4jhoD!%
z#|RG62m?>DIlrszTaw2EX|00H+4}K}e^Z-)V#{E0qjnXR3Kl*&TGjsV3fddgFztNC
z0~I!+)ZL~wHG1mmlZ_T;Pl3MQ^LuZ$jVTN0%Mm+Z8@!Fm!Ww<Gwm##~YoVy-o`hSY
z>JG}$OAQxTP(-8;r8KwBnhK)?KYb!cL6S37%q~npX(t|n3<k^obuf(GES8bnXn#aW
ztP=m`*)=-&{gsOrAQp*%!OEs(%5LsF{F6E+S~MMQ9EKbsu36m6LFm3&aa!Hf|LvE$
z;J9r&`gAJa4Ch;LJe@frU;Qqg?K)-(^9XB(CDk-eUYk(VAM+Ns_~8AX>~5rDoe$IC
zk7XG93LG&cd1`bfT7W<ZH~j#mY>%U!?rN_isJ#q=tNec9-+Ts%qvfyBYUZJ3=0*zu
z3$BNebms0pT$G6TijLF#ji|jT>TXdp>_C`3oa2vlu_dWIh|TwDpFw21?Ct&HH}kzD
zqxCo?XXn7xwtE!e^TYYLF9f?DI~*M!U-Kj|#j~V8z^=1*pMBRB=Y4F3Szx;RFVv-0
zBqHY`FHp>#;jwwX5yrlRKR-soKF0h(afnt+TNnK9Y&cF+QX=MbqR)TCi&5`2Cy}rA
zU{Oa3qjOAeE&sn=gBP+1NO82UE$wvO37d3u7TRq^+QM}3IZV=^5U)hY24-~1&YKXi
z?3g(mrf{TKo2hi#MHvrTnaaKJgc)BqP#xP5LZ@CK-;0V=9uBSg1p)Y_7`C3);5@k~
zCESIMrk_mK+LlVCGI}sP7F^mb7~uW)=@WLXW|ULQ7vd>4F}jGD;m{6YXed+Q2P31e
z;@13qn%ssig3`0}u)YEK>y^g98WB#JvLxwwm!5Tu<c}f1oD?)dl4;~HdP5r*;d?uY
zXaJVlVyop#bm7`~2&&G_nt-|}?W^QX!OM*oSFz2yjjUrek4s<&Qyf`3jF^2!y|@9@
z-SuW>vxW$*D&s}&Y7Nx>gJk3CL*wNJboe^PNd3Qf0dUojCk{=lu8#Nh^}P}>H!uL`
z5xzI;VJv&U-~eLa-pNT<?!V<GPrgUrnqHIs&#d{(^%FgDs6jm(XR_LCu}QzNo5;-u
zUGLp{_tUwy>Wv!VTm7dy4)oJK`FoH1=+xXnbN8&$H1vtL$P^a`CQQpkDvb-?E`j$?
zIV5#QKai^L@9cM&y%YabKf$v&$dmznavbDv<ZC1>EHb`O?1evN)g^00C9;Z}ub3fS
z!J%UbdqtH|dT#?5hoW<tPRuOl+vq$J)!L`(IjkyMPF8Q`E@WN0vbt`Yn!1kXSvCHX
z?g)<ebJ|87--1@%Gf6s{6F;aT;bWj3)=PlZnIfIm(qc@YT9bqj!_uCymlRvhuQ+E(
z31Zo^=d~vCVe9Tp6e3#Nrxp<fQ2B?w{Q?|@7g;cz*I^Zvy1DV;Lm)mYY!BhDSE-(>
zU&TP6x%fvVDJkRN9!&B{-s)ZKa0yFWxv8u-3j)Th!fY|N%X=A8?8H08Tzc*G6@z0z
z;vhbD32)~pof?C_7;$12JLgh0TfuR15LzMw?qb_D+<pqP&AfO<N`h=c@X7r`V>HAo
zi5@aAKqFhi5e3$6zCQN8S!wF*`ScZ9rp7ej^v<r?_{{{XOLC*fm*bUJ2v$*6Wq}aD
z5i;_KxB1gB{m(kVf@@8oq#$jsbWJ<ee~wEdHvmkzKW?+HE;%dDpQJrM)D7KN09Sx@
z=Y20f|MRqDVgv2=r$4{vDRN{zsALgfOIFFISQ6x~t(i|u)&t!d?MM&|^>rdD<}YB-
z=Xt$fE{`bG&Qt?5C?Ho?8)#Oo;Kt$E_mdXNX{;)uk};lH#>tM{D=ml*r6DYZefcWZ
z;aQm{o?&2LZKFg@JzAi5;d2PRWm1Pm&E9ycxDD;|7du}*`CcB3go1XJ2WH_HqgFf+
z|Ac%VQU9%vb}}n5M>?SI=!t66^@n=jiNDJ#I=jQ#7BwGU>BzCHA?})ism*QTx{Z@o
zLYPKeHQ4{NJFH$?3)Kn|GVqKHjJ6GJGYPJW@o-M<<Ai&Jb|#4lqOj>v6ss)?VyG~u
zINC(Ywa5YUFB!54QH8vmYG%FF^g+^4HlkQ3@PMOksFc*udwNO=IxHh(#p$>JCvbim
z!(_4?)&)V5wzg{TYdg|Uj2g&->BD#VE+z0!hgvyuEef@2`E<D3Q|~HTt_EB$ScHC5
zKO*vRF7)>ttCQ<lx4=d%d&ttW?xA?Z<Db5vG{+2`O(I8ZBm1zhH@LBJ1k$fwBPZbv
zr(90AdB%x#Zw&3aPtRUaLZ->ABx=jS<aJSEpBt&eXCt|$sv^T_yy?;#_|z+?t9Lhe
z(PBPXzdLBPqrgjqv^YA7r)&?T{<AIoyEz0pI0DQ<O-*s(L~wv9K4&s1D8p5+*Fk3Y
zcbDVKL%ZgrWxLj>#zxBQUD($2vgbuO)32Lu*S)1hX{vPfa4USjr%$QI)!a<e4MXn$
z`6Jgk-FL4HPXdNeSyE~%kk!kYepqUU6E(#e-hw%Ya@Q0;I-!CZuS{;K8|}HOYZP~1
zp-KNCLbw4o1oTw1s%W%=Xw~H+MSBy+8!R<hio#IY(jg6By-<*`qX!=uUk}?MMuba-
zD^{%FV2<y`y^GJ*>djt;hVahbcY__4uJ_yGInR`b8#VkPFu8xqHs>GLzVNFwA9}dt
zx6UPh{eZ#*H>9HCssmYSy(P9a)u~T$w8`VF&MoZ5ScTCNOxF>y($f@lId0EBLC)Wy
zE;p8Da3@eIUT(9iLXz_wziw-oEpWya7<)xUN2@UNzy_tloVY%K^>8ohF;Va93n?VT
zU#Ay}F8*}dy1{F-HrXR+sysFwI_FYot>Qhy>qvFs=$ao$=86za8hc`Bp-ErQ5cghe
z_m11u7nWI=*gLQgoi)D6MjUxlbdaq7xo~>6EvZ4{&H3PXlk@3>Be9n;2Tv86<9&1_
z-gv{*?>`*sF&wInS{;<1Q=6l7hq}v^jCI_YNS%K#odxEq(sBD*N&@I_tmKD(=n4M|
z`Xo5rwF1|0&wQep9*f#SX})~0*wSA9obXyV?Bw2O?f2o{u~nX*=iv3rD-uvKRr&YI
zrtQj9ZG`4q=6A!vMM3N)y)R^l4IcNYiI+QLW1IqgOfj4ETGGtX?AOi;?k#6`Fu_=i
zZ&;|x`;_=r@=xpVV->r4$ak@>z;3K6i!mHkO$#piYYu};{FgF4Z$TfmIfU~1gSV0B
zs&2Wu8r6M?tL_AGgHth53Tix`(?Rt8>^Eoyd>{S^s=b0AM!@Ep9Q{x+9%hx8Ap)hb
z>Rp={P9^*JNoL{4MxbQU{b&NNk?U6~D5DSmGX68<bq5hL_sI~4DnWl`T3Sj72?A||
zb{nK0omPV$N6sizipz`t%@xs^<0xCXK4RMyF?pO?Qk+y?IR;t?QkbjStl7QaSI;3v
zFBQ`C2<Op=H$1xTFN;W{x%fk^+dLA2LBa7p$n#~SuQ2P2sp?s>p`%rLy0t#UMPius
zcm>;(!kZk&2W5zw0Oci}TeL1+z^*GiFCGMy8vY}cNtI}~>%Q`g3qQV5l*(NT(@c|V
z_yIIsZc5Yakf+(te9%|X3Dz}*Xe$tnL^0Mp^-aR16{r6>_5Qb(9pV8I@R%Khluc0;
z9ZVVUJC_FBj8`vstD)}xok6yVxc(PV`vM_87=g081_#O0Zyt(a_S(`(q(Ok$x%l1<
z7Q*rz?7mtbC7HGF%xzDBfSUT#h57WhqUI^{#NO|7xLk+{6dXE7vgP0^(&oGgVngS!
zA(D(&b4%jMm*XhNh?@JE(HGs9R6Nia2g)1XqpITmF6hO#{F9}&rtUE5n}V2Byy*wA
zWHmd*j^uAxROHg}h2PuIQa*CffD_)H4eoZ1e8NWK<&zQuR8D*Yl^@pLp&9Ss@u$2j
zirvG%uoqbU&oK{V-gUO{1<4iFdlWegYj@1@??YOA@_L(c8u>idkQ%o!l>1KquvRg6
z-voK@a&YpC(xy-*>3_!JUFdcO5xnK)77o{{i=#wdc8r4{%ckJrr{`+Ye>XyZSQp9?
z#B9`D`W+7OOsftv(b<a`GvtmX0k#+=N&+$@-5m(0@MX6=t&VDWp5)e}(nu|`&2o0F
zF8PeyIz(0~fl5$0PmV@irr$|kP2sdKKwFF(b_~)=iKj#!=B7)hKB_`(-W{DkZFu$x
zwVE}(UjZSa?WV8haq!MbaPx$dd(+k_47oViyq~VKmfV>N7dEM5<O!_Z7sU_Ar;Xl~
zPzE}6|0{fhBf}1FeGY4@9%kpW``oOBc51pR!wy-pVvFe)VxSy4=x2S!FPrs$aLw1O
z9SfL+>Xnzr+3+_USh_HduRix)^YL?ziTGc8Ce*T$v|MW&V?Jh%bXf8ETuKw(EQzzQ
zW!JzNE;5J3D4&n!Y?m2?I>8gtd}Dk3Ne8uvK&aH`9cq^e*p;tbbtY$wq-7(Qiy~|g
zXr`}?QS9F}`?Fl`K4E#0E7$O|H4Y?9Q)BJMKS@nw{!A*qsJ)DBZW`CE@msk7*67{b
zvMs1|^d|f-5yA)!sPK;#k{HmO);8~6q}LCab{Ff`;*Cq=XMYP1rYMaFI=KZvl@yDD
zvS;GVEndiQzl$#=U2ee%%F1*mnFyOwzY8m8NWwm#96lQ{zLOR727+1-NpNA&9Sl>y
z!c^wu@Opsr4poc+F~nst8tOm<=b@=oSg#om?2P)S-0BV*)QUJEZDLeB5w=<!Vs&@!
zS}E2-<HdLWp4E}q5)MUh-$+YOJ`u9W0%PFoyh>HlI%}U8aR6q<beqdt>l)rM?5>4w
zQ~mWJ^X$!H1|2-6=B-FcejPPkKB+bD=qHm#+f802iuwwFjoGCdf$<98(vN^IGoRyp
zqzm<-&cD!lyQmQ08}nzm=_D}A{%oWFm+QRV`#(Pa?B2RQp!v;xrXY^S$F=uuH{C35
zqx?@4{#Ps)*RB&?%`k8*5yFc`JyaL^lwRfR<=!^Cwj@nT=Q-HbW59`UL(ub6e~xPP
zSmh}S9Q4w_@mW6O>rEsf^Y+~>Q=u4lxu6o<uU2EM02|ruvm?4IH0d~>w|c5AWS;k0
z<M)@=>#eI#l=F|=yxyHdG*8zQl&fc|E1ePokIWHDbw#OT7YtD?0spR3M1YsTsfeYB
zOdWd!wCrs*Zi0gjUq_{q5maw1G1nEYc~ZyN*7~R-p6w{1p+~FIX*hyXN#SHs;<j<T
z_+`DF1M&m&LR6{HWl<b5*;5%se$;LvhWF>+3^6#?tN;9%rs>Xb(grnL?fm^FRw<ue
zxWxxqHt>}m%)<7hfBq|eMew%~z}FIG*69`0`OLRVwH91LD_1}dAbs$wP)9ulXrCKM
zcbsf(2K{B{3DA^mtv1Tw4jB@x9mB2GcE%UDCWesj*4aMOYA|2K5xtylJO#a<e>q(7
z%0}fr4jbV`_2FFkQt^({`NM%h0wq@<$G>ZHT@B!&KWx#6wB+t)rln#0?cxEEtC8i_
z=y4qTL!eA)^UZ2oEZf725N_N~pzrx^3EStktLvrS*AB!bk7|WZ8!lpK5C5nlCJtoh
zJY*R?>QuwP^G{O1pV`3#32(-}gDcL|h9ZmbOwBMA_=NbSh8oj8$T}0$)6PDyD$Lk)
z1MMZ0BF*kYBPjPL0k|egph1EvfGljxnM*5rZfd-|!nwXuxBon(MtH?h$&!MH=05=w
z9Pf1{GA0ZA-PU;aC$bGGq`;y)W|xmQ^3xzYj+`XQt>Dop@|=EWl2u$Gr(l#x8z7#5
zF*xW1xDo?K8Ts_I5mDqiG9Rn7e<tT6><cdy{BC}+C2vF-L7O2<S&3<F(EoEMq>|N{
zcTY`T%Vv6jQf?@Lb(k3}5D6|bPN!C%?RncSMLLVr7q}!?9QN^Y#jskV;^JJ8WEGv1
zx4~%`CJ2%iQw2qA!zV6c44EQ63Rw~hw-Zj&!8Xuxs&^2Adf831QhyK^*@Utwe#cx=
z6v?KP{l`F?V`ux`&t&DV$Xi4yHFhvhGoS+2k`3_7IMPQJS6?t7(Ii6B9c~^Bbm>xu
zcOOB4r&8p?n@I0b&u%4}(<repS72j=c!Hc$3+nI>76=u(-D$W<i`CG(s<4|Y3Y~>-
z-les^TK1^dzrBen?soFZP|Urw=&-ppXH&?HW~TDYV`Eh7^jfhog?*>xs?$=788AI(
zH!F`A9AwSYJqC9@l+1TE!@GQ)qQ1@absxZa@*vLiG<6Q5Fh}c$lXpDv4x;<O!`F@B
z*{zKT({_YjKTc+F2(0a}z}CmNZ`R(6jJei4IgK;?heSbos)7?<n6f$f8+g$1!LyEf
z=z30jhjzx#4k1Ib$vYkiO~2ctWMJe2b<}3FeXe>yQ+$-mH(Xpcp}4#VN9}S(z`!cm
z0PM4B&&X%7$KPu7A+8VR%g7Ui%?k`a$`z3pTmHk)ReYtVzgf{~YSfWY*=TA!vG)uX
zG57Vc%|xFXAvOfMO@WM1$5OQlQ)fOwk)A0@{LX6`{O@Ypw@fempO&jCSCgR)lprXy
zk+HIDo8<6Bwj-i)#d(DFjhDAKkk1eaxH$|nHN}nZ0?IkSfD|hm8`)g~au4AGFi7mu
zT|tSk?z4Y#vI|&Mem%5A>Iytaz5#%_^P2&l&TiPah{*gj2EyME(pKrW#34RCZgTLn
zGVA25D7XsB8)$@n%R8IiMZCFrPrS2MJA^OJgAOv`)Y-^t(v^FDFtgKcfYb8ZE>uW_
z*d17IJKMJgURfON6x7(RjC;%)tJ04xr?_+!&rjf#fyZ2nnE>cRaTQ5qu_^p<R?H+S
zI5<j7Gp3HxY2TXPFN=^D$e2E!qsXimcqH7G-}APRg$~dsWogkLjDWmz78LsJ^K9V%
zp-gyes+xw#Xome%jfqlI#X*KPA`#Nh>(zddm~Z$+@a9{gfCP0q^%rL2N>Zbg?+Y=%
zODD=Tlx9=WGM~US!|X3!AaJITCqeMF5+>jJN};orYP7TfQ3D3#8f|&t>%5S_BKE#o
zz5Cs1jKk4XZw}v?T|)3n{1yEHkuqWT@k`p|h+vA>wlJ@73u2R7_Iot-?mtCsk~Ee3
zNq_d#Fm||mn(w9}FZpB=)GExJ^?*(SAif}}3V25VxT_o6j4Ns&YsrDH+l(eY7d1Uz
zuz~)LLq#dx-^_*J>i9S6hf-gc7<2OW`0#O>9+r1tNh|9e`R8-i=Uoc-f8sHdO1wVd
zU^llOYrim_;@j)eNZEWmMQo~NENEaPsFD77TG#hbY(YW#@@B1Ea@X|WcQ{f`tKX~Q
zL$){}%!F46Z<eKNBe~hAL`x=N8L@h2a(~KfAcF*3o!h>YQ5s*8BLIqKq`}VpS17My
z$0G7R*k!-~<}$|jpMHU)s;rdH70Hm1Km(lYCttfZH%wo74R&p%4JnqQvI4RG5Ua%k
zHD*I_g-7;0Qh9NhsVhyG(q%;XzR|mx9typALbAyrGOZ5w;YM5F6a>iBPt&OknSKxZ
z9i0az#p9NDotJzHB%gGgYOfe(u6<E%;jI`)qBX8zz0ysX(fGXD=&8(+hRec@Gr^l_
znkkE=#uXaPfRYCilu|ZdjUM2|1FU&KJr2;U6vZ&qqM1v?MOcq!a(#qtkSCnV9}Uj0
za?F4Fd&8N<nC$(0s!K@QKmWbkA*?k*{m|zvVy~U1>{-Y25EMkTH`B9A!JA%!7k`KW
z=)}L@cfc6}T#ZL0)wW1Kav%-)F`8*#y_avYz>(utF*THS7KOe1aIh$5F{dx$J1-u7
zl<e&+X#5U;-z<^LOUD(&IJ)9rHFi+}V!y8KsE;Sa4Y~RAvqEcfNkUEJw;x~j-Nvqm
zq4TSHBqk<07w4qBN-23|ln(ZxJ<V4%Q$VR~t$DrErsaqEtBv_zScP<$J*FhVZYnL6
zZ6XQ7dVZv&X!9eAIzkD*C@UBrqJ<00IUm#qR1#avir0DBf4j>$G*a^&0YczgSzcaR
z?j;A&@U`so;#FT&ubIm5tKN6!Z6U8D-cm%LE<Q1EcY<UTTLOn!o{AVaQTO|fT<fw<
zxKsgexcsDUvQ_7oU3M-Jk~p*Az{@xVRU9=i)z{H{OYH)=v1?!BTD-=`C4w@P|CHAM
zob~Txt>+lmj|Jv1B184B*`C&6B3tWF9>?qLj$e&tY$ta!O$&AIH&NHuj~aHHXGiY3
z%KrS5NVQ{4)`>h*#p9dJu1P}4dVlC`O_V(u?it#U=rE6Ex)PJ-9wMM=E(0Fa*Bsi$
zSnd}!TUY(CI4a~@m$wTsLj4rW^b#q*hI5|xm-i#Zk88Vj&^z-SPYnDr55-jY>atIs
zhj|wN)K4vQK@S;~F#o|9U#X#on`#x?R9LJNufQ64MGaZ)Mop&B)^X|(I+Z_El~=5w
zR}#l%9XC%0ecvBu%LWi$D4?BEULM_PJRQp=WcZahO|gbl;FfC^+B<@wFwJ?~d2a{T
zX!qMo?zzS*fV+de8Z)T{ZCehXYDHRS)!0R^#DZ?2F=!h>)9l@JJrX?fY7J9OX}Oh2
znIsr&J_4;{tZqM-`&g!Cau2AT;SmxFP&ETY+yAS4nDuKdzQq1^N6&fq?%ug`$S;a<
z+Bdy?vp~~15Th4XQk~O{t!yi4d}Z}bgF+r_-o5E`wyCm+gIIXG1r-8*kXwpI>v0N}
zjHjyKJ8JG6Zsw5d9#wVyDVXs{ygBxVp{#yCOLy9Z;|0*QGWNo@t(|xtU&(?>gIDiK
z{70Rr6Nk99sLAl*jgTXakY(Vp;&U;QbJs%Ee_304&jPPY^^W<a@<-WwTHBh&<W@t5
z(q=gN(o&2NT1vS<j9>&Z4C*-Fx)I8Bj^UG7Suw1-L^xZKacB_Y2;VB~F*`kYWUCz*
z`aHViD{U-T5Hy>@HBVSzkIv77oHI0RClu2OM?;>8f;Zn+Xx?Tdul`OMyk(|Nj8i}&
z8w)Z^jXXFYu<{X(ITV|)a=EET(>*589UZ-CFX?3jgO`g*5%?9r3(0eEKabp+LL`5K
za~pDqYn@#+?^#FY_p>CHU;dX15F~oP|E`Uoi|fKFYpOOc5AZM!>Rtb_zSq1GD^*}w
zkpR2(vF%RDr`@~sIygKiWmhv{#hb8dXx3$R;U_9D-*9SiXwkJjENT#LnzL$|6FD01
zYS6RRDVdgac81}Pt0IvFV~G9prt49!{nkVp>I8e6dOm$$Lo^XRZm{bMRervgMCoMO
z|Mq(q#x}tqA=MFM4Sh=^JYzrx-nS|4-tH>Co*lAU;dOXp{Uw}`CdSZ1hE$cBLiDvz
z*7(XUtcL1!Zr;6hz;C`KkF+wf_t+MvInt%VQPu_aee0IFQ{b54UGkN*AJF8l=@^+a
z>kHyB@Jd{%bsalG`EMQZF8R&g#+u40c$~cCDa%Fe0j)4!P0`Wnze`26U;=SaC3?0b
zS6;8sJP*7peA5o+pduqf9gMN7VtqMRl&T*A-5fZC<<IVkZuTD9mc2eygUoiyRr-o_
zSo-T~kZJ68E?qAYus7sn)OK7QM!0g>-5bgkOge1zA;b0b_O@(9Fd(6#S_=sC^Jnlm
z?@Y|h%&cc>zigh%t+;f&e_B>1Tbw<dx#feOnVnSyfSoygRh+5mX*Pm!m_XfsZY&a6
zUsub5(K_Cn-%wossubDx(OV=e55c6<KdS#9S>GHSclW*>8*Oacw$nHb8ry1Y+g4*X
zY;4<&Z5taq-Qe9mU;Sp@&rJ4@&CbrAbMABdy0629CEn4swvU*Aa~s?TZF8+c9(6#E
z=<83#yA_X<3D);pS5h$H9Nt9!(irm}yb)ksapo=Id)lzn_;^+hFZQ=6sxA}tlAnw2
z8OPwcXj*?AYwjxtxspPiDJWXu6d_X4u(V#>aL#gPJNgt@iLLj<b7Pv^<{Z3On?G3D
zP}0B+sYG<a;GAxovq6IotyF;Doi;D=Aj&$^N{y44ndY?<v!NxylMeElv*5IX24K8b
z(Wk8`-r!(kj|e_(IO`k4FP##G{%@ge+^$X#=p!S}5`Qozj6^q^RWJynt7UPod4%%f
z1z)yrI)v>A9Edl|Y`?X+-!TnKg+bIFo4U5x2D_5J^0uUR9deq+HKCay<j<{6cuP>P
zXxNp`ts;q}EAQ^gCE*Z#C*$HtSRG(IY1-Tt5FW=HYW=Mw#0(69Q6zNkuMsJobzSeK
z9>I?n^LP6F3{}}~<0P~qKin#hCLgdEm15)M9NlK;BDTH|g;|XGE8q-wrVN=5el+CM
zp1?^b%1RGA{ICC92RPFAHz*v8xnudgdb#9Q3Ri(q6-jgqjDYIsLqQl7H8sH$HP<&G
zyS?XKLZY;k^|ER!OYifW7amQmk-jm$BqqGRxsDn#vR9Ww_1Lb!o|?BGVc$~@W==y+
z#9m&Xxw-}FuyfRdM*st5?ffyy>qWN!Lf{ObF$Rt^k(Do%c6L`+kSoria5u&|8_mT0
zrgfK~!Zkr7FIv6OO!9;>-DMLtUa!n~xxUm7J556J;f+KN4|ksXTK7_$sI4aenClzS
zm~NV6vea<%mQramM&TQT+r*JKd4(CAB;Q}_m$Q_R@K;^^Z@nR(#HqIMhF(2YaF^_C
zD@atpmi5@9Z}uyXZze!2Y9H?Y;UgbUX-^*3;*Yo&RFts?yLp>~YImjRTeKAO9Q5a2
zUD;)%^Kf&}Do1=0`kNZ}jcD41>yUP+(3njastah?XwnK+nA=$#Nee49Yc&yQ#!AnA
zw>S_ke7R#=Bg?KG@bls9oCkY;g3E%SiI0{7y8K2SSF!8qG9B*Fm|v&se$v2+gXPN^
z&Q_g*<}?=U6gFY_EUc;bwV&{P`zq+-9{11P|Cc)ZXJrwjDq3nIrCQ!Elj!aVey0==
zkjS9P|3SgbF`RqkWI?RZf2N1_Z;WePXI54=etun}#CG)%AQQ^G*$=xL*5QA?{4E0}
zCI->w;)WKDJvoRoQ#VIHW4W^zxnG8WOfC&Toi@*6E|jb)k*&Ipzqp$_u<hRmXeuv_
z<#Q@sT;Dv|$KB4l7bBr^-b{UUhPP^oTd&;_@$<OR%(mm6Huaaw>1f`r37_9L6jP;<
z`-7PTPAbFjGL$z8GCGa&MP`}Z+1syIJX;?W;RQ~xA=)z8iG2mUrKJV}c@z}r|7W2f
z9g|>X<K+1Ig>szunSEJTFnF9PXYjUzed@^y+yOA@?71_toWw(g5(|Sz@b&z)Ab?y-
z@xep)^_qt0rKWh}qfnyrH9mee-$?rwXU_a91q;hg%p*8B%r})`=UX!JQWP%mPeq7(
z?de(9e>%ZbTgOJr-!XKYM{!4_Xg*IM=P9OW_g$_r^BCLj{OSr^=Q2;mvfsT8!7(3{
zgm{mt54WJcIj)I<xzA*X7&{nB?lCdYVq`6Z1p0Q;_HW!c8uQ7a&#x0zU=CprmysAX
z9(KZ-{nb4G5l^JgHOq1)Ci^c!$e{Q3f_5CZZvL$EYxR(3<q6)RA+j@>Pu`&#E8U;C
zh=ayc(|75I&svJ8ZF&gkW9j9L4b2Z1EdDuWL_$Q|JHlrMs2G5QBcukf5NSJ_)Yx+3
zflQ|5+H^b|4IX5p4W3=h@zM%Et!?v-%wqV;CZ~rlQ5f`~<9O}*?DxiJpM4qRtWCF9
zi<G?Eaq>1Ft#6a;8WiZ*#)CyCa5Gl$h%EWD9yhz<;v<gU1(rx2okAP>_IdF?O+g@{
z-!OaccbMu$_Zy9`(Tf7_3ZKUFe1F_0bfdCj{NEGBF-=T-=WpA*tS*&91D*4Gj%Sw;
z9=@Ip(1!TweWYQd)s!nuFY`+)Z!2RZ9s*u$*vnnX?eX?Uf_uxs^=zr<>$@U;t!96^
z_vryCp^mFJk6UGoO6m82%aV+R{6$L~&=Q#ds9-cMKGEX<v5MJdL0H1}W^zf`xS?O1
z-wcqKd$hk4txh2>Pmr*Zg>;SA9OcFW>-eqVG|0lv+aBVh8n(doQ=IroNHhMY{d-b>
z|CYAvxM+6fJEBa}by67lE2C3?^&j`v>gbZ~%ZAf_@kB0fpjO#FE+?D$?_neqE+^pA
z5v%VSjwCfw;dO27h~JAl(eq7{ydu#+>lZ*m48YO5td{o|t^Hu-y)R?4qbR!fM&2i$
z%#hw$u!Z&uAC<fjSc#OCeZ9f(%-VDarJOh6?bd+L1Z7qywYK(J$TH=yD(hJIIvQ0}
z6|&Ha3T;xw=9}aK^=094S?4RV7x{7&ClH!R3Sw;km9_{`EL>dLpDCcF%}Q}T!o1)1
zhrb?%qPE$;h`qPoZjlly76!_oOV!?v^7DF@XDZfr8@;XP-Mz&5Zg`IPm`~vdpO|oQ
zx0gLv!#AJ_o%psO%x7hw3&nGH|A|;e@tJHi+VydfHd9vp(0IoqVu?E64}XdO;YiTH
z|J+=kx@drXlid)RXj_E@6^g6WLqDuu1(BbMsj^{xg|PIp%sM@HHIL`=fA7a%b+;FE
zeH3<1iaNR$O^DTZ(Ne_jeFWupjCDJ4wYBH(=ZH985{J9ssE}K3G@rUDZ)D!H1dAVm
zYs9{)JpY^C8`Z&g;t6gc!%O%VI-yrLEvWVPh`{DSp}WE^e`7Xe9bmmYCblLRpX7%6
zD}3qpcd`zvK~E6nzBk;V3(Q0+RORP4)XT7-5L0ncHRUdm05N+5^Cm6msvtJiZB%_s
zwF~i%p8~M|-Q&N%xBwMR@=IQt<46piBuEn(7YsOXwmEbNo;<W~p`JsgogFRHFDJSe
zB+S4e7DB*s2LKr4?T_lO$9d0QQ3{tr6y`sUSy?R+9$b-an0Tf}kG_E!LmJPzqtnx-
zI#NtGnq*|k&z&^zy*bN@)ss3n0k1sIF36!uUL&Y4b0n8>tHu&r^7b;Ss3S95f(*9U
zelCmHl4P&?n+;U}y=^UIJp)pn6!4Oq|10AEsG~o7saZ$aEp^CIxt?iX`z{{y=TY>V
z(xw6Xy)7eU4k^x+Hzx_<reHt+D1mi56OpaWd~j|-xKf#i>W#JhdCRsD>a)4CF%Qv`
zPZ0VAfF%2->E=JSntkg@O1k<fBF7&#`Yv~`YQwcNl`r>7k(ZrHMXtHl4k#0$aQp}e
zqauM+LNl+wKKs0R?&>5!9sQ<;khmlb|NB3x@>S)#ot<RGV(!&EkGPQ$shXPFv5kNh
z<X4c95N?Rv55JpgH%Q>2!U`{pR1^k=iA!)nKtewE=^?_g+<zdr?e2a<!OFGt7O;4f
z2+0ce765&hA<N0?^H$my$&8HT)P*ZYendE>mFwX`hVIj4pN9)OT_NnW5IxyiAyes6
z{nz-<!lb2c*yV~6v-cklci^1gG0Ti-C!*!0YGw<ITc&mC9+Ql1K(o(dyf@jBSp>nI
zd#1lT$VhWxG}bY+`jUF*Zhe^o*t0sH6Uu+jOZHT}y=jl9z4?;p2Z)u2!)aGh-c!*c
z*na44-{`po;IVw|OeEIB<#(a}uW%s+f;3TidAUQ!=CR8}_cLfDs^`O^M)<Q02)$BG
z#m;`09Y+h=u<WV?kUjRfQhOai#X`fzN9m26qpR+D_yBi!Yf5vNRHKa{FgjK7w*xud
zCWkifC-m?4X-~ZB-)5Wby>2bH^geJ{Q&~7Na|g-_hs+MH=*{;+JjnWkr~iayf4-Tu
zS$&4`_dXTZ{{tLR{XYCj{<qe{Dn7Hme7gwV#zP4G=QqBkaMTX%$-mCeY<R*ao~Xz^
zbT5HnPebx)p`O6kD4|O5;-cEUVo-216_W}{MY^gKsq>pq<7&6ey}{Gggs!La!KF#_
zu(_w<FhX`AdtFPQ&b31J@I8LDo-^E+GKHfYAuw?u-N6wix+YgrwKPG4anfiUxIRkH
zCt((hP$E*WN1fb#A~`qmX!N;Th4^l-UEr)=<z-gz^oyPc_r`xsu;5?TNE!x)qW1Pb
z*|Dq336-F!=YrH~c6Kem>~0j0q8~I=!Vq-d;)@^gcSHmSe~;cVJ9OJxPg-9(%5m>G
zD1Vu6J&t?vJ-eH)w14h{oL2U_op>7R?2^*-0?Ouo%*SW&bZ4C<VWC){D+=AgXji3P
zZKnisR{Q6H^UZ(f)Z8a~IW!q>-EHSU@!Q$)NKHFSGQSR7KOR1`9WDZx^9_LWobdFv
z#6d_x#{r(|P~w9NW5dCJxi9ZubKLs`u&1Tv=_z&-vSieuG#GN>wcFP;xHYIQwo!tv
zT~yZ}1*}Wou&@037Aj|j8$9ZP<qYq2FL2*Y@wPks#MZe*4=wx;GnSv`aRejpE3iM|
zpA_BB*wu4=q`ZaOAN<T?AZdPJclE@Yet+ad@iNZfvH(AwrCP|mTInLsT7GxPUi<J$
zA0qdr$*~dY*O|88%f}lJ0$v(wpD6YGHSzyk-@%>en3&F;Qk(_A0~8d}E(Y5(WWBT2
zS8C9Iky62eb8Bl!I2Zx`iwhjrfE2(B0FfA_bvo~Tub2sv4J;3}dyIdgVcLGTZTq8E
z@vX}9LB&p!4Ol5&|42nm$-)bdkYxQl@Zr-xs^W&%rJ%cIUJ_n{RtfNw{k`SkLxjSI
z@LY5zUJ_dLi6h`Wvs0nlYGwYhRko^Nc7EjzJk;}GwJ*CqoBC35<Q8P|l0m9o27fwe
z9lQ28gVCqc4RQX%n~2(Ad$6yBX0F?R|CAO7Q15wh@wOmQ+_1NM#ro3~nC(IpE<D7p
z!+o<<ogu)oo@btN{(CQ9#i+kZ_)yAO2ff;5PO`tyf0ERCofK<Juq=yn`0O`u#NfGA
z)u{ig@I9>ie>Ur1JI1@TW4X7<$jgJDH4AW6RbQ%ed$X`uJpM_h6A#p5;?`+7Y`8_m
z*Gy;4DT*Zb%Jy<xWju|M#c@+VUdSn;?{6#M3oE!1=zJU|I;SK8i4HFy!A4P!iT^^Y
zwF|N*l1i7jk7j?RIscsh_%jniHym*{(|W~&o9aZ-)W<RWZxN7XDIH`!n2PZgXm~Ms
ze=qf2AaX<F1Ohi!>=N0Iw+K);2Tphq6<|%@3vqLDhffr@_GXrW4pRyPP=njoa=AU~
z5{s^{3hc4C18}w|u=Mm+Axz@+$?FCOS8c8#AY*kko#y&q_5P2fWD5fk;_7O4G+Tho
z3Pb~g@Pdl5D+W^0;~w7+y&z&?sABNr^tT`SH8m!(#`IeB0!KN%sO{plnw^?0fJEo8
z+Md^IP9|OGwX(gDGmoU^onBWTTN_5`utB}hq|(YqN7~$dz065%_rbHnrUnZ7=Ltqo
z3?$KQXF`I_Z`wfvUuwi|&oIVtITC@`t8I?50o|-po0MNPzfypsX9P*!^rGNRuN0-7
zEbR~8+NIw-N#eyRsmp8}3cLc2RDO7Y*F;rBvn$7U4n^A&e<5rRAl?!}&M;HXW-m{U
z`iAcoju>M;F;J-qW7NM|z@4{u*hMu*tSKsqq?9w{+_NqF5@IIcs57^UO7f2k{Og6s
z9tlRUj2%sefc+Vo7{!9U=WPHcY$#o;-jXV6FJ4Y{ak6BRxAQ(;@nvhS?TbVx@}HHQ
zi*^5)+c&!-DSe3uA=BPB9s!icg~M-vYG{$dStQGH2{2^5JxD1wZXY3oCBL|FfU&nL
z7IH2OgXWZgp%|S^2>}ACnuswrkXZ>WDF;zj;O;2KXuLgma4K`8NMSdiVHW<^51&W8
z8rNm1WOR#h;l(5d{nv4w1GO@m-8!UbBy#E5!V%Gyp=4Lj*SFAkT4hP+eN=nX{uzaE
zCMdA)R!sO?D<bNCtZq$jM9_CDs%+H;!%3!h0YG<EqT**wp2w0UaKe>1Oq1xw!Z}yY
z;(xg+#)hhD7hSWSN}8G;Q^Qd)F%)c0*GuG|L09L?GE!7-mVN7w82Y3PR$qq1&yc8b
zUf;aRGqLqw(pTB34P!wQEw@JBgmu17X-E(-mx;^*W*%8E_YIZtyJ8mQ$yEK_0ypZ0
z$$x;bUVuBW1ftx&9|@;=&4Q=(bv=H}1F<*qNSPaE_!1_DPQN+u1dtCF50#Fq2j;(0
z;+9}scC)9982F|>kZaVs-1gQ0bLBugM=sbk^9M}OknEvb&BtHXUNGliUqhd~hAL{I
zL&IUD^4Gkmh&WfpJK&1o+$W+XsWR<dK|zgX{_>3#5i~T6Amjwq;vFydbzVIx3xff>
zxWV)fPi3O$jB1>%0g6`JYNbJjR9{v6%}fCleB;05r8S^!({AH3u*Qm2_5CK^A2jf>
z8WgzH_&w0WkAXfCL23@C)cU>8#jbmIj+7VxvRXKvv#4P{*||Sb77ONG1P@JM@rbo^
zU^w)J7=Q^2336~t;rYwY=n-^HNnY-RV1a8QObEG7l08Iuu*5<NoGDKEq<N6YFE&*^
z@jNC1Wkyzmc$GBTka(}V`O5gxjifMzObL?l$##p_i3UbS+CHRN{&yr=H~Z2t+~`Fv
z&W9RlBEy@w6!*JUsFd)dOkj?6%DtcI#sT>)GM}tb>ZV~RiW*vdMPMAxcv37vTIr;=
z-stUp0Tgaq2$#}{GFU^rN_8n}({{z0O$oOH;jRAFP5V!08}e9$s}toGOW&@C>6WP9
z<IMIcss5E+*b`&7+u*@s(v@~~ec;9aOWo^hr=zM06`)=rQ9<~eiSN7n;C18Gr3wO9
z{7t-yA=Tl?t?<JCo@#0Lq8^D$Yj}PCFmb#`7HoF)%rep>Y&aU(T`cQ-J^!p<d+6wB
zNv9%K^119wQ`4gLJwsG?Aeik4?V*4WsyCpgU`;P{`FJNY9h&eFoA-`q)+MAh01Bj`
zZP#-#1Tc~Dglb+If#{9cwJ%k%D*}wDatOZ}Pj}&IY2`ckgF}e&9%hmg&LKW!a^%de
zkrS*oB`2pbQ#8LF|0yTX_Xd&}HKf%#eCr4ppSQ|HyppNkn*52@>+IEVm$mG1Q;3Gs
zm43C&5-6$(BO9Ur3&UnjIuN}Q@k*GT#M%fiXVlkS{rXmveP5oCTB`Bb-uw+#>%d<l
zPFxMHK8B2zF}OJTTkSo&9KGffk2R`h+vTyGvXEcK13zNzsv1T^N|9We)B3}xCf`O?
zzVUijU@Y<RH>C|WzmSx{=wrvtJr}e!4LAPi(Y4mV3-74fRl<BkG6AZ<qVZ~z)7Qbc
zk8SPsBP?az3$Q|5>phM+Y-UPw<#!8&zivbXqK3Ko(Th;1J~Rvril_9_%1XrY`sBUX
zU&D!`$~t>&SKdvcM-p}9qPjOERagqY$jza4j~qu86`$hR*c!7@gP`>`7Y_0W#&k;~
zI5b%-U)6NIVu`9G<&s-IW2PYZ!pk%~=mH1ic)TjJ1JW8j$^{RWE(<o)hMQ)&KUIJE
zWL9gqkO{XGx;L<_{`u&!&Cvso=YS`msx3rnL`bA{?#4CuGgr#MRH2V%cdVh_C((8#
z`F0X8V~9N#`dt_Pmm=Os%9hPg$b9BSqqUPeeF+$}Bt4vhc`X6%EVXs+n{#KnX!a6x
zVwfgM;BPN?L{Cm-hIHZ+dwiiV_<<eiQ8)c{z97ovex>nllPBsdUem7@Yom+@YUk8f
zD;(xwsn<vJtEyg@#slA1U(8$Ra|5}etd^M!gLEsfEyh)(ABp%+(6nk<>PD<ECL6Ad
zhtA{H>6)t!Zqn`?2a6IMeW}z7gKr^MZG4?EJz=l*BE#8gANRvcwe{q0SgF?y{G#@5
zdFGgU;yBJR{OVFRkm+gNgvPEe<-hYPrtE!Hygk5|_>|Y*31F{-X`>NNpInMpPtLR7
z|7pNTJ__YwWAo}C-|%kvDKf3ZSuY4zqDke|6-&p!D=Fx9#N*FW>hTqW;RQPVsxID_
zy$z&P%FI4}`OD^uN>8@<In18jYSWkLzaHbiol2<Q>38RwfH?1~;0^tC8a%;5`2kQO
z>jy_J#iBr{hl$O)-{SF+sxx#rUfj3q3&-!@@!{a%31(gzk8eD9U-fA5x7UZL*Y94u
zrj5?d6cr89Re+S3)lL(C9PaT5sq{WYssHSJ!6>0Y=HBAUh-Su(E4cD-TnLXx(7*)(
zp4dovB0uTTglEI9>UAVTCn73yz<_}|ArSB$jop{Q9P)ogG+zHe@x^UK7>KRYCF0C7
zH-MHR42NF~MX&xccVabo;VkaXswGWkr9qYZ0C1_6k#-YRN_QAn^dvDQwD+wK#rE$h
zwvDH?!;8M^NC_>}DGc+#+Hxfy$=e=$Z~2ISvxi($k~Hw6f7)h=9Ig!}@cJOJ7kraL
z*{X*tMHe|V)yJat=7YYlEU5mGAdeI@W*h|wOs5!raL<wP#$xWtj28p}5d2-!gkvCD
z1%ekMq&yDZnkqLl#>Ab0j?|F~;PJ6OI#h}RZ82X8{hg1390i_c=;~W+O@Vt9Y!f06
zZcE)NA?#GaQhBY>MXAjbIa1#T_XX;W*sX<Sk52)W4rE#9g>*-4Pd6)FGNl8(5_dRH
zBPlkFjDtrq6r!+0_{+6zUr0}<s6Fmw<<`s0CS*(xnR875^!U<wo##QV&U<;OQW9#S
ztdIKNHLr*FD>t#f4nsaA5;>zxRS`C=g;n*(K|+d>VpTdGHR#7HZZ?FCY%@^oW{zkI
ztwXJ>ymlP_R#(w_pFw-+{$DI2oD<)2LCL}r0DV_dFq@5HMptom?bUhWqD@9Vt*d=<
zy4|&C(j(d~f(B8=5Z49~D>+bhBr6FujJ8@`Fb24rJ+U?ZVo1Q>;7rR{WjyF0huZZR
zTiy;L^6^uDf4BV(Ah0-h_aeaH@Cg?6Zcm!X*MN)WxbBd$I~?;=f#<+$KVRQ<yjI(^
z6T2sshm+mA%e?r(l_yd&%gVebdyf~Y?T)<7i@JNw0e0>AU<XQFQY_(Y{;W~*g&X}T
z^Ghg&M32?=QWAylmUqh|2qcR_J{E7<y3!OB{8IAH8rQEIx!NIYw0BbSW2$BJHAT(1
z5Ca7f2!eOFWo%C1r~)|(dM~u!vmoK4<H#Cazt2vDYjJKzs#y0If!$T(1zC`PK|MQv
zPttC{brxox;(t0)nt_U3w#Y^K4W=ZmnV_%-YfZb76Yp&2)l-V{wNs#wkp!~$hnQD=
z#<^`M2En({^fJTkWVt{D_%ucy!i1){yj;n>jv<W0Llwsy%xI}0=+<(22e8s@NyZ}K
zXPr46Vy~#*TeiUoE^N$K_vUzsl9FTEx5DG3z!f$$$)`J+NG|9af!`{5Jkp-mnCkcS
zA)(s#rJWE2xw-4eMEnxORTgjIs~A6AGwGI1DR~8xi<e82S$eZD7XCOz#bXvObbz6I
zWUtr`VQ^DB9E~vkH<=QFxP6u;yjLL@&a3GG;oC7*_Vwy&c-NAOjV#?j=#OB@S6*My
zs-5+H0!E=l+dTG3l?*Jt#FGHH0_67x2@}XEq@G4RWRRT1XhXU_D5$zRFp<6ZC|y~e
zeM*RpiQg;tRGDqc;_{mAE<89Q+)c_bX*HE3Th{etEt?;{U-QrDNd4YaKF}N#W*@;d
z=%_qHF$U^ixL#<vsBl`mo+@M1B@~RGElISvFSW#=f}w5>F$4c41p7ebeWj?{q;kD#
zUhjsfq-k7OK=kd>KM=D4vsb2%n%~F^M&@&uyDO8lv|FDtTx{Io1X5>(+l^ieBFpHS
z$=<7#?=)VocX-uGSQNyer>QN#MI)pI0z*l0t=5yI&Hy|c*ZMWX-eY!vA%M_3l(5#2
z-wqOkVW4o@QZ6Lr1dN96F5VtDLt%L3qH@w;a@s#v{A|Ls<f{ysRVAhh(Ajy7rW%Hf
z0K-2uv}=G3p6UDn!2^_!&SuemKh8xQeR@yCzC>S4!uU@wfS^=x8CP?)R}|-q8w+JZ
zJxfYaqIy3v(|6hE-_u(P@2QIzK6=03J{q!rARY_^{#E~9EI`S>E^a)E#Kgp78-~9p
zs+-=aPwcr(u&b09>lSLt!IOFI<qNvfokGO^Zv*jl!xDQATz5O-%u0IBM~nme!IF+C
zW+>_=NRqZ-+Fx5LzJn_r+wBBLO^?>_M-0sKtSD9mtC+cQ*-RboVE5RsAKBRa6%C4&
z2wIu|!{nQoeZRZXuPX|RF6&oGC-^VzOyQeN2DJ_5_-#$A9G+R)C2|5)tv66QV=c~a
zwH1-<Zi0tN!VLK1mh*61%pXu)CE<t-kGe~*M_SFPjXlbpdZbD!;5r5d;+Om{4F0b_
zorT6#H2&63keC{v*_D*_WJAOK)RK8K2?AtisjX?wsuO2tKhAJKPKKz3ox1i|vfAR|
zN5?8O)aQl6AHE2plKESK2>A_3FW2Z_?6s$X{I{cP35rWFy{4pAkFE?I#l>vDMgtM^
z$<iK}UAd|2g9diNB3nL&eqmJVjAr`qm(<B=-WbTqAXxPX$HrKvB>-<#1UeX_j}367
z_QX%KJIt*9sgQeVsHqKoykC9f&a+xfW)5AnZ2KvUV1SNkFmz=9=G*p+tMuWG$BQ9=
zNK*ekm<T8h*XU$BAs`8(4dgg#LEctY@=qsRG=O9g^h8QbB<&(tMD31K!Q|@L_@S|Q
zGo%56iQChG{S>V9H?MKMuGsDw&;i}k3!w&C(iS&tD%(%!!PHG^v7DaxR<~|%e1EQI
zPprONVC1@%%%8BFLTKT6Bbyg@pb_(HRpeP%!)sDmB=HKTP}FO+eFvUW$&%q}P$}1v
z=4pH+em$F>p8{ek|1F&d)7xO|ljZ+v59mZs@Gvj7<5xI$gKk(QmW6H#Q9niP>b*S>
z6`+o<0E}^^2@{Cl#8=S3^oEwZ&0=AuA>afnYy%A&lbr4lzS+T8$bfqSZ%0?_ayAsz
z?>r)dONplk%V}`HhMD%`d&UmVEts(^D&8_ptsWmLb(2x&ua0K3rgSwI+PsmcZ5ReS
z4Wa3^OU6~@oL8@>fCogi%9l#YVg<M<*;7dXcK*~z4I8t!W&{>W=XFIsSW`w9^KX~B
zQ{P<aAHPI@02?_q-LR6I93j`8qiL$$0%l`49q~@1(~=TBx`l=QI#YYw5<mpDD)n{o
z@jg)8LrZt|n9BQ)%Pj<|oc^%l;YJ!aG&R(%xeydW4UIF_XYh=ss;RIA_S0K@vL}4T
z6pSB<E^n_(QCIfq%-()85>LBOe{*F|o6L-$tp#w7Xm>rl+v)3e<P1w{qye)!zxt@(
z(RKj3i_|X$79vE7wZ-EEiK9t_L~=eO3u>&nRxm_Y0khLBwnorVOJ3}un8$d1s}Myj
zs7^~TUu*A$wVJ521T65Zj$hc)LQKUX`ikW!Y)eumuhxHiQD?<qT5YNi^SJj+koz<B
zdfgQYw9iDDrDM|cs+)(4R+ZMN=>HTm9zm#!`p%zk7<nC!NoWBcMZ;VbaGJCAdNy9>
z06|_@=b}t#5v54RC=}TknqgYylduS^A7vaJ#IqQ}gMd!q8WVK!tmJ9Q{#)90Hg*qW
zbHX?}i#7Fc&IzK7q;P%4s!MLL<(Hr?&`MfCOCVhRovc_>8UZy|tt)HZ2reaqqP4zs
ztLjJk^!6RvicvH}kUN|VbHbTzz1dYphT?7{2ZGN6+&WKl^k|L*?by7QVHWAp*r~<V
zVlvtDNA)wZJEfe`-^et=S*m6k9cSfDZ!);JySw{I@*fAe3ncm(pSj8u!867yzRftK
zYbe<CUf_31oOkIH8=U`1?E)!FYSg!?k><7+sg-9@rWXCYB@_PeLfr6%wh^wcTU42y
zUwwrX<EO>#Y=xpEM<B+{piW`jCS@3fq9#qux70QbP#@V7IeZDmG%#1_0*Im+!nS6%
zVW>i$Taq0nzs*e;PWxu40dDKuZ?=rLTXHSlQ)BfkNImO&HdeE<<qIH&`0Hz3g-ZCc
zIQ?}8j(mMPAPdM-f;M@1*6nXsey?kQw?hp#g;=B>SW%qh^H2k0Zs^^Pzny>s4ooWp
zE*cgqeVjN~EwbZq<y1fA|0&g4h$Z(d$31^B^*uaaP%pNecNS=2!q-V#v^M^Yn=F%M
z7k1`TkLvb^Y{mI?U9y;+nJ7P~%9=~5h^qD=k-GJ&1ZnQLB-wLaT;L|}RI}V!3urD>
z)rCQPSL2xGG0W4Q+1o(<AE&jM+GB5v(sbf0waJa1z@KNQq*zQFUo*&|;HL~wN$THW
z%ASF22-@(Egd5-Q-sNpUlZRRsP^O=ENj0mKX=Fbxy`ElrFCJ?j<^=`rPvl<DdkoGW
za7r68o?0$X>%K--MP+wh{c&aTJ$&%wUDOrr>M+{LX;q!cV0|)jlO4(jV_K+2+?%2;
z`)ek;6EIS1G^5_!0kj_el&D8kiCpv<P;#HbsQ8Cq>sQ-G!sab0_ebb|$Jp~SQ3V&L
zS=8MYg4E5S@oB#*+QbgS)bX1;SAx#x$)wP67zZ))ptzdj3=P%P5^CdNg}!j>6<vsu
zjOg@bZ4RuozPYw_tu|Xo|L5rb`p{9$Sh{B4EG?iU`1db(a@bU0p9o=^#kD$LkydJC
zOtKea_Ucp|S)w2KFK7nlWnv+fr1k4ViZ(b+KQ3f2fZvYvO#6z|l^RVTYy4E9Xh#_U
z_EBUokz%*msuOv(l>YAEX9ow(BIDisX<}-mr}Z2k92sV-ro?K>CFb>(E+fq8R!%Th
zj#@Sw$>M?38Kybt-dzY#pY1M>v6of8Jlf%h6eW9lmPQRnaIk$_O8U@3Ssh7DWbq16
z)cHRr^`l^xo!tk)yx874^1_~`K(dP81h_z3n4ZE%R9JUR`-z{PSGi9I-|cpfp4j=<
z^_RRjNAJ!bsf+h3B|9_lTL4OF{=6Tb*rQ4l3MmyL0Bv}a>bg-;xOavMR9qr|tJ+~9
zNv$DLc{0WTe!$CfG(<fYX5+dQmE+?#EWJD#SzREQC@30|@}#*pTe8{u?d;9fe8D<Y
zT6=%&-az39W|hPg(a4|H2dvO~LG+d@jt?^X!*=;0D%&wc`juk)4ToO7!t)`ysq)p>
zcKik*bu^>0)f<<V@v2Jpc2l5nqc^jqx#4lO(#<5JRza^n$~!e_#w%Gy9tKs=&hSDG
zu=+kGyQWQ#g?7wr@lw^5LK02NDHPk7US+cd+Sg3^dI%-a$7(TE6ErJ!oo>Dn+ZBQQ
zUqj~`C383XiI5#A4p=4?AqP0n#l`CDi`!AmrLlY8N%}=^EVNem7koNum+5!vOz0i^
zEi;_{*PvM-*rh)GoUh!~T;a6(((WeQ@%?e`T(F_YYODxqI40`i>Dl|XXP>*rHyO|S
zqr4sjud~?XU+q$8^XS7%9^^5>YP0W_eSbcraCh#T*J{j76K)orKuhw{E;t}a1hs;(
zx=cixZrst}KK<gU7%wZ+=(D8a8<T%h8#7O;*6FjaQK?HyF|iNIueD^*Q{AWdc!pkj
zYuyY>m(P8dGGE>dqi(pb6L!w&3cA(C4vVm4Ak~Ai<3prv5t1F7R5&-|ToHtsz(O=P
zISxZu;&s0{$?j0P!2Geyr_?0+1+~HOtU!lSD5rx+@tDa1x)}ks+?a}QI7ViPhX2DQ
zvS_lt-cv=07n*K#6ye8Vg0tojhY3*`W8bf}1F+sUdv@zb*X6XR0q}f7E?96BI=XBT
z;r>dhaR2hDDBY!=AM0Pqp~DAed;I>_toV+siy@FnBkmDzd$x%pRP!Eno++pYV|sT?
zdDk0X(`Slsvb-V@F8A`5$92z70{QD#xye)RKrmiXEOL}ocq)$xht;l%BT&FcDRJ`s
z&wz)IevnJxL}dCf6VR_D{43d<16G-uG@V?{hIC&EgZg`C#DNwq1#YyXhbkS49rd_f
z>9dj${}aakeY$iT`@TzJpC<*Voo0_fiA6$_^#Nc}8N@rxcY>)NwHnmKoO33hzAx{K
zx4m6j{1gU8g}9XHYePSk5wdDU(b=;ld}!>%_PAjaB|tTIMA0&&6jf(-(JrBQH=={?
zb>X~VNk7&M5qCK`;OVJ7?Wo>GSXxHOp@!j~vbJLOdZUr?+dcZf38OFD8I6H7nYo4W
z=wC~Jmv6<691EHlsG8C&Gssi=d3@B2ad5TqERHuuZaQ{!VGs!Py$4guXMWlSf_F)y
z5Kik2)JT-h7A#uMTE9jvOx9_9+tQsJJ(8`C%!tOqvBn#UsV%eQT+8LZNEGXjjd${&
z2bF(|(qC=4Zgn0p{OiLOZ6_Hh{OzHirrTJKmXiMIQ;bj~w>|J$x}5x(KPq-pZ5;8N
z#Mh~#zSl^-z_sm1+Nwj+j5Tr^EoDxT2IBAO?FpdNH!Z&QtTHjdAse3~M=p*@lcAbR
zGTy-`PACLA+aKP&PT(8L<o`{R-PC~cmqXr%>m5kHU9=e+Ec6C_g|!#1{6KENBIg4K
zcg35KdNhoWlA4O5=V1n*5M0;wMb+=^`s?d?TFqnG^lQ8L^R@AeTV+@@N65{%qcB9i
zcGV#^><4O&q+D@D)aov+h3YPDljMXf9A7Mvo;_*il>AtZVsax%Ue4+Z24RDdd#i&H
z$#0v!%=9m3=NxFvF+ND3Cd~+cUb~j8lB1(}e4iXV%XWVoHkduP4ptl_4*?<&6oP~V
z4Ad4rvfl@I0KI+V^=3~36AWIZ9t{EoBEe+GDtvh0iLC!=32Ua^^$D1U7FztcN`3kn
z5C8_6F%GzcK(M^ecMU9kGgin^-1r#8=n5JEfW9kG{U$yd!H)PjfA)g-_n}Sx&!(}q
zLyI<rLh-`c9XIdS&6amj$nI3^srB0yNk~5AlZT!8LRaoW6_;N`UH*B~r~X2Cc|AEb
zhXlTOYx)>@Ff%`ATUl}CpM6Sucyd#Z&+cAk0Q~QkOO_nZKHbe+UR^}7C3gVQskqPu
z-pERS%Dh+R0$Kr>RQh@8?|Ifk#YqHBiIc$k6)jNe1DF*lz}4Jcn>fdc%aD_I@)3tB
zV{*!wMeEU!9_#OudXl%T=;c$IUpXqdbXTG2|7D1cJHwGGz>z|cv*Dq^r@q@MiKD@n
zp!8d#6{im2C_@`9G5{>~1YPVE(YY^>`zHd9BV|ks-!?VHzeOZpG*}CsN*BIlc3-S8
zaIU2M&WTCL=U|!ez+-=u=_xz}s($a?8g`+jp)O;2$v>>B@%XqzRJ@2Z#xMt7$C<Z$
zd9rAxPPOAZZJpZ9=rJ%|fR#wj^dw-aQ0<pc>vW2d$=888yu#<X?vD93u~sz6bYm*n
zNU|lI*fZy+wLDs;jF_yXQJmkAc_ttmdBc3vHA#H^`$J&Yf9K*yLQ|NA7fCN{IqHNq
z9GK7bWdmUJ#TXkz6?g!OGtHkaH(x>IpOg#?F(8)jBy0np#DQ>B6&00W^@>9fqdGaL
zzugtEbkmUc7K3ls4j#V@`SbY$36^Rk=*`vt<ghCA4|o0}BhO(lnn-9kv0r+czE@~e
z6hcHN21J^8f8NfH#tuYw0KWgTDHz7<@#UlA0OxXY4|Dug_Tnd3{umCinY(LcM%6$j
z5;&8HG~ey@ewFIf<J^4k91z=l>E{EK-Mh+ZaAqi665T*0LDi1@c3fS3e?xQCoq|v4
z?t&ru66YQ9&Z~m#9{c0O0i$4X6-J7fx|PAZLcgFh0}6XtdVht6fziz7%RAh?#0wQS
zHev+#fr;qr6M?QyGPPn-Vxp*}B@Kx0JXUfZ>x<o1XGR9QMN)Qli~{wcgvB@7cN=7D
z6t^ZtQoAKr)0nI7_i;97qWz^R^JAT>{lT#$-bh0H6SUu*l(FX)4O+kn=U?0w2{xQR
z)iIOr`n0)@_@!eE=6?1w?5<&M7<4CD2#=dRn4s3XdeN1kqp{%Y@z_Wl_#}!eH%0M!
z;?iH*<g6)^dXAzcIV~r=wNNXc%VoGHLMLmB&c-2k{!#v1M9~0Z6e4|h{m~o8ID-_^
z3Nt+PaMLRLd~53f=BUvxq3p2Z%CiUl?Uht(Rei<1^`o??q5pj7JswW5Jal9E&EA_l
zwYN3>J)ZJk;`_X8&Klw<w{bWDhd&)zf^d(=Sd(J6FEYTI+4WY_jz^QWJR~}zyhXph
zWt$~u{?&A<I2mTaU)-0Rx$D-k1&A4&n;2N92{EB-oZCOR1GOdaK)by}Z07mu>Z6?^
zm$=U}E!>myd5NamA0U8He6<XQtPQFyAbpm_0ZSGiF<y9l)yWYpWT7=>ActUNh|LMF
zaqg3detVEwF=IBAn#6uB1sXAYd3xz&`rFue!(6hYKhkH9xNd9pGFmb~97BuuvHg=V
zP{l-UQ8hVi?T_|SrPjdNV#UiJKuu(HbYW*_c6~jYzULt(h+!~f)n@p@O+`y<v|O!1
zv(tkWB^<I!ef(!(xWl{?bs%;G)LB-qGK4-kls&k9c=>?RQLt=2%=dEZ=cz}+;t@qL
zS58f^5w{f$1iJT%Fupaq-kBZL$wmjdu9!lpIH%fo9@TT$7VNX`ATd;q#zy>5VN1kW
zu{|0ZF41{ihFukSTL1Xq&~OEf1l*?gF_sqW%uPl$;{bicYC}seOq3p(eFpM<fE<q|
zH)>o*uPYRl6)Dd15@f8<2QO@MCzcx{$UNl_InV|C72V`yaT^<kr~($76&45xh;#u@
zw$RYf!J(ngxVRvU+|RbQ3=j~J&3F)wjBLvIX2Ks5^F?QMZL)jNC_^yJG+jj|>y-`K
zJDbJpvlV%VCEJ})Ztuj|a!b9N8ke<BGyM_N+@)^criL~@QGNrcE?nx#=Lree9w+mE
zQeSH6ARWO-`!Xyjr>K$F6g9f@a|F54*Gh2RKubm);5H<-_j<_C!)ZIS3_uslmQvpN
z9cDkala4aHx!y8RE)+y~1-8p#jGG^vs1Yy=$@;z>`D+u>yZz&>M(59I!dhBFB??Lx
zYBAAF@xSWFpg4nPo<@bC><At#2JSwbUKr04S!c|d=e$u_$w1)Oek_-3SMKsurhri8
zGr=~lihYjT6ComuKs74g1i_-dya?3j&;4{lA<T-K`};3)CdVom+Cj<|B<y^#NyBxW
z`t2#qPKa7Qi$jCMA?~XN7EzMZMLVpf!xKiWbyq#<(15QSIQ)5qQ)nEPCw8~xv_DPK
z(fGnPA0g9S1Yb)|7^r|<lcv}hCKayvk3RzGMj1ZdoRX%U83RE1mDi3L>(>S%7;s4v
zbx5qdk2)Pl<a{qxxAT=*uERXZ!rf48{hjOGffXCBzcqDfX*joz9F<Z9WhJHFoBa{b
z>ps}G-B^78Lk$2@u$ss$D^4>+CS+d%vRSOsesJ<LeU-L_D$)RP`3qfd7fjrr*WGkj
z|Mb%3K&}<xs-GQZ?UJ!DXnSuTOjcmi+lE0fDBdLUkgVF*LjeWhE9gqqp`ME3>H_B;
z;0u~FR(<)w`KiR=3X8&K1=Gi<!w~oH-`QeOyDP#~!gr>iq~B0kE`uVWcV-4K7&)ws
zI<Qr%dZMN;Lw}_%8)r`u_aehSDpHY<1b_g7XgD}<ppF1gt>{y5U#|qV-WNSRLSmn5
zaP^9m+FF*ev9Z?Yz3A!mDCK9$%{z;w&H0vQ+NYlWc3{Vv3EA6mO~TgrO=`xthszOL
z_9EB!Gc^_obX(S0U!$}j*Q0Oh6S@dr-RS7fc{&k9KhJ84dGXNw|B*u3JS^<|Mz|(-
z%O;E-*|>U@i5By;IGiImd<gB`PS*0zGiM$**cgs`rx@67EBK!NqrMZvj~*4jxjL<h
z#af!fiGlHEkxM^AqzMCv;i;Y<Ay1?v{{~q^fFsO7*VA}m0Riz&1$9AuEe#{r<uzo;
z<|SbYC$&|p0Aq`XWuI>klnRO}HNLHzL~eeL0k(M;hhalj^0_|()V<y0@+hE&>u9`C
z0`;KeBSlNaK>=O3XzBIoF>tj3#OPAcKXdAp$EG}cCc`qs%8>7Wcj@h`Pp8MpBQPb@
zl!uimoxim5prFW#pXZnTAoWL+@D~q{yKBfMeq%+|Ro4^I;^LxZlZ9X3T20RW-AlW!
z+{p1=)5RTi>O0rPhlxu{fAjXzpQ|O>F>T@&;s#d(#SwR*<HwO)#GL7}H)n7jzgHd^
z85vV^^Md(P8d_Q-5V>SXSeP-_hxY=O%R680PglK4qQ<Tgp^qmm0}vHz9%{?_<c-Nl
z2ERL#^ZfCn*ThTdjiRC@m^>cKK5C&PI2Q63um=&`d}t7M!@+@Ne<a>wG>H;41<|l&
zofR)OBqZc|lsfMd7S`~=XfkWo<gapBklZ{xJiJHI+Kv-*zb61%QIYubMG?7(Ia=f;
zR$@K?3xCK5W^Px&Jp`4)g-PbEdObc{&iApy_!tCUarPqa!bD_3bY2QCm56b@m$OF0
zP?ZD&KmX!^2Nw>{WLp91qzj@yb|Io9_tZ0ufjKkGRQ78$DP4t`$A3f_1;F%UeB8$=
zV3uT(Qer_+^$n#c5kvgc_fJsY4G!U1@uxrefVrew0o=iwVeZa&ke_gi*<$wf^>J`=
z?%sIRYAb3GTXEolke8r(%S1YB@MIP@yW6=Y2OjdaACS?}5yV9yhs3E8f~n$<2#hgj
zxG`twYO&!?fCl{bBP0!ZrpDkV-rEYClU->+l+<5x_I@IwVSq+rIUKtbUI_pCupK^^
z!MaVHbV^P_D-OZ>ujJX_49jUu4A+ob8?zF&$Tw*SjW1u1H|v=lfC75WfcPwvn&Esw
zUzPfiW6F_t<UI9@f~|{_hvyb<)8sVw^y&Ji<en!uqo1NNg!obkWl!6g>OIe*Ex>y`
zy)<wwi^Ql5kQIlei@Tj`znIA24V+2nZGJ06H^$*)%Pbrj<0n`?cdaXZBq4_|$-ogP
zUv|{?y!k0U0B5pYqR3&1*|@1ECH!M(iJVSW9$|SiNz}=Q!`ws>5Q+oW4yOJCw=uy6
zWXRLyOPH2NPpzM4+I-!MBQKs@vD;mVdVB+xRPywi2rYTIX7<|=M%Tv2#tOQ-iQ-3s
z^=nGY%HVSpM*2e$aaoNA1BMA0KoW_Z*&I;LY>@YkJahYS&bV+UE)Bz2;|dtxOq;(K
zRdGhQ?|I{ztUeMVZ~&j#LmdtBc@JRFg*0ho!SmNFxp|?VXw07nNXyBQ^YEn8s+52b
zpM!yru<5KOF#K-ky*6Cz-xp{=2kh0AyE_nW!j-k!V1YV2%P1`U(5smb_n{i;w!<=W
z655-36>IcNzmlo5c5+h#Tf!RHf~$)+0~Js++cW8knqCcMg3HtJhG_{yA`%LMHfN2D
zxGp`tUo}h$;mH(ECYjl$?ro-`4{^DPWeburl2ZN<>xa%<7o%Y;9!K?RQ)(5tx)MYD
zkqUvWoeDK4!i6UN`&%+kWX_7?-dDGC>80-YX02(z4<7cyp(G7Y8K*5z-lVBVf#Zw^
z0T4{ZCq<oliRD-yF!`<zR@qAMsE4q3MKD`67Uu5{J@(l94g6kQi6AufTkAd>4a<%k
zbN?0(iKvLvS;E&}EcrpputNyvxIyKxJN0HW7l9CjHt2g)NYxNBFua#BFRb`<U#^Qp
z%R^!yw7H`Sl=Std=xaK~WMtrl@~(JTh4Es)si=taf$$fuC`1C%&dv?=KiXPbTRSn!
z!Z-mx?Wr3DAxG(R;>TVr{b8u&6PEj<*3QL{wTWFWCCWO%zbi1N>}Mx#AOcYWHlP?8
z69mA#bHn;RlRzC>8@)XoFh{|(^u6n`2*EpjR<#W~ioMgVwVY`o)Za!$H<{|<?Ixo!
zK&G#D=vJQ?ZG20TNsu!fO{q2JodM_92b0b5o9y=C{FDE{djbgf0&(!^OIH4AW|@Z^
z_J5x=S|}*l9}uZ9M|tB2(n5Aj>mnw#Q9z*oNZmp&ZZkB|>;T!7e4SJS$Ck8fPLc@B
z^>rctCyCEr=UKl-dkBP@X@fO3JvlFLeoRdJzqVjthu(OY>sP(*NZnzAgU@p*T2*h5
zkA)nP)RWvi`(IGz%$i4;a+)our4dmlWi(4@m<LR`Y+FT|?zGN8%&p2vc-cfyfHQsc
z<~drAM-J9SM?<5JIw`P>JLR(6wKXXmYnIk;+Kk|^6rxf)#Z9_ToQ~vP?C~uMk;1{1
zdT!0fv%W#I>xeYk;wxj^RST{+*Ev4cRV7CU9tXXytf$UPS(i^2kT6;Ln+tE<+3pLV
zP!l6&Ro1#kp{Cj9tli!98?hIoDX4&9_LapqX@*JSpV`8}&sd38L0KjD_rv8q4^m50
zCF6wkfe^#NMmQWC94u^XV!pnD_}q4;mX>fJ<G_MrYHS>2UDM?z+j_GX;dnI|sdGM9
zHs7>@YE1rxiewI1Gfu?@N9y3u?&eH0+9i<xnOWFF7!JKBBvPaI90q`V{fe;{FW&2>
zz-%je#Y!{s`OwMEin&%cK%9-jO=R>VXv;_wf}qZ5A<PcGx)pvw<LAqO(g$m;cfdC6
zBv=guCBo~Cx8XFnhq<sk4og9QFWe6S@RnZ+E~bI-^`~$_N9Zh%JPZSo*1G2WQ(XQ8
zMa-f-;D=KaIUr^nF;XI*Ytgas@v`!AGQ3zcBpCSc0n5zclM@<bWMrBsF)c0JY2gon
z@l?jFyD+G~f|SP1ou-Sr?k?(t+W06s?z5a|Qfps&XZ`jtpKXlSxP^E@WuxuCD^WiI
zzVeQKy#@7Pa?Rvj{P3S%fN}eQ5U!?fU;OiRMB;b+lUFH^JzS@hui!klNR1$Z?CZyj
z5l+6HI%e}6mmQyYFqWSWjvhDQt*3G`9Q<acNV)F<=e%sT!sw5J5Z-p-09m6E#ej!e
z^Hk|2cVYoA`{oa#>py++y@|H9`$C!@*bWy*4Y-V;-`&C%W@LynJjyMA=yXTI^v{Eh
zE;>4*k(SeWQm=Sl%crYo-I?4ikF?15_DLxIWVZ$&ESva^#pI=-u*1V_Z-<sxO3k9#
zRtrQddYiR<dGxdmqy$Ul_RqF%t~R?U=i_SRsCyx#hlBAcnqSimhuS%eR38%<di5KU
zF%}O1$TCTR5fN@oO%wS9DuSB4g20k!$mGcC1)e(d&2-WqLxBqUP{B0c68ihO<3(T)
zX_6*C6`hx{iIbI@&+x9+?RqGiHa~+VmXuf(0QgF*jy9R<8S)l(xt8mkgBDKt#?F0I
z4`^+3+@%hmK4*Wd^2C+-1QZ95cYj5HH<wanYUFV%a8BQvb756pc4ZhJi}tLiAdfO`
z_HQ#ma3(tb@ySV8VUJ{Def3D+{@B~;mgr!^dY-r<8sF%wG=hB`9Sp(t*AGJpII*Uk
z*na-pP$O^*NSzQDzF3+lo0ZzY+*~5GNT~h&eF7n$bVC+U-z{eng;6FIEDAw>Lj$^2
zs+En6OQ!tJAQKh@XRkT<`KQKhl()2ijIhkB-oXTNB4`I)BQtQ17rxdK^v8FIAT=c?
zasp4>B*lPEWVD*#pQ1Dnew%I3odaP>`Oa~43SCW@0n=fZ_Tt;zB65M|;yG~m0qXcv
zw$g%#glOk}pN6AJXA$Zso&}~_Ihlg8dXUNtdK>%^lVLD293H-TOy>nYpjFE8J4P$6
zT^B%0f_qT^lm%TU=z$q?vCj&yZ)6O}U{ls0)=D}$I?bl<nC0c)^e-`%&QDKwL5ZjR
zoGTydrlzKsU(+HaEEbhj(V%BMi`X8qk|tcwccvX#`-ql_rw+M&PK1}kV4^?fK-VVS
z_WEAm@(;eph~#zV%nONt4>LtxaOszmiva7M3|E&}QNYtqr0vWLB99Xy$Dv&|htaYy
zA4G;b=7tOR0yocvJVC(4cE}Q!$1A_m!%K!;=aRhzbUj+@UUF?Wbf?vc4_fX!yujj4
zgiAvWw@G|2vmZn`?2->cbt@G>H7j5VxKk!&IK51w37&;F_42D7WV|9HBnx|AFl)JN
zRvfn0oE-l?T9*Az0ceULqORwS1=tdz6a^uPA)!5GuUQ{|;I@HJ)Y8QzAg;)LTB7$w
zl)rzkLa4>XpZy-9r}M7ax9MqIihxIY!dWFdXi+H;Lcl}dv+&0){Ze)LA@;|U3PV-J
zPEJ;8!(B?D{P4VmtKANqwIjKN2p!MMFty3*AxF5uk)bs+hkA|=FgoO~Q8@TQr1R-k
zAFFTpvy=jv)1>!==h9(F#)h!oGHANjFAxM^v{RosPz0=IISc!{Lh)qVQl6|SIS+4c
zDP7@;>4+qW^=n;S10=QSdSgBMUXb8Cz6;ru<nOxCw)tMPBTGhlj7Z)-%gkWK`KJHJ
z(^rQ@)qc^^-QC?FEiK*M-Hn2zbazREfHaaKEiK)RG)Sj(cih+Scki9&IsT_Jo|&`X
zwbxpEucteMhA@UmhUnn=Ly+n$Nz5YFIi3@n{_dRLhgtFU`lyh3?;}-JQ}!7Fexs{B
zPo1)KH7R}uIzw7sCDzRgydqB^Yo}3e_;2{^>rO2lBPnj8qK=MS!IP#CqgMCx<_W*J
z%1W}|cBuykOyEO<T;cl0M%TYZUw$tx4t1_Wc;UM7R+AC*VRl}<QB{SECYOXAwwerd
zYiQ-HzI~h|iKu#pgEJ2IAcIkga3?(yo_tecoxTr0fuDE9fa5#w91nXMXjLVqa-)on
z=s6pF#kYWj0#N{w!KAy4VjlwSyugb{OAAzdwiRRr`_gNPMA)j5Ed1iiE=yjL-xT@X
zpHiKAikuc5&eoCHt@%Il8HEoaA^brKa+FYnJJz7^gPb#{ri3}C;+uh1k4{X)lvL|I
zy#Bqmrb3rYnJAkeU&Md_%gD@Jn3oqQ6@!Nv#YYyD6*7Z6WC&iX1)ip3KhF+SOwa80
zTS=L02H-L$AL^-hB^$-Zd~1U7F{O!NUf~*)Tg_^BpR28jj}F4A*}CT(dom4w98l@H
z;*)qo3Sdtz!zq2QJrr1%<)!F)XwA3FWWd#f<z0o%HgWmm>0nV!w^w`^{;oCadCnJ$
z$6=HeU2M$Z%ozP?e_3N-*o21GbYLP|z3AGb>oc#CU=2dWorWn4nGc%r3v9(}tijq?
zU!PTUaWUP?waM$XxJOUlzt@gZ&zKeQf_huDWBqGVVcDw2s$*hf3cBX1c*dSv%s_6I
zraWzGX2+4TK;hKvyU~`<xk4Mxb24e?i3e+fJMLfZn42n`_&dnFb2fCdbZ#0hlTCyM
z240sxF+_9Q7_oP3Q`5jJH=_oIW^7L7?=^jYNSj`AREB!vyRA~WH?sChx27sQN~%-!
zZN}p!;@x}}cH)u!c4%{Nq{h$zH%f>mQ&@R<WC8W)LW4z^V#zc@Fj)NRC9`<6HKv7q
zJc^-AWx1#!F0L(IZ-bzw;UhQ#%OaZfv+t^%H6_Ge|I5YkvTIfps_f$B==EH`G_RYR
z5(ohjBD3kI+dIOB>B3}`n&Lm<-pfjOwcCsR6jyB#66h(d-P`tfTHzO~VP28&>E_qp
z>N<<{p5|<CCsdM;!NUm}^_JV#2*OZlL3ZiA<{T^|n}DciiNR;v1t}4Q-@nJ~jIia~
z?ToQ7$$_1KJv}{OH8yBSRkdabKjiMu5O4+d@Nq=yV_vmQM;XNGV|YUc_XI*%^F*`5
z{vix0U;~*?40eGlhzLsJ{V-kVzu5_(LdDn1!V28a5LHon3o#NY@+HWfU}Qp>FU9s%
zx|Yy}LNH!k7wX{XS}3X~%4W5xdXm(eHeN%{U~EM~6Xo^^0yP=ie0hcHiBWnmmr$_L
z+EJ7`!eSxVi1X&e9!~R|VAO8|Rf2B38LBC5J86@sb7*<He+EEG@qi5Flar4~$jEk%
zjwOKF?2Ez2Cf@Ji#A0|mqE}-M+v;(;3n<@#!%xM341-4h{i1#-UFnxTz`CIcsD-A!
z;hHCoFg1aX%X9Au+utBmg<!T8T&0QBhFqbj3dWtSo5L5YA0&*}?ycWJJzFD#qyCId
z{SMiSF-10E#?(B$B6+zig6A9LPIcSP-80pE&+wz!#QNyY0jrFCA;O#q*X`f8D==O7
zBxaRd@5ZioH~;_3WeI0c>>wnygeEB}c+lL0Uk9&IRUkr?$Bu?zs()OHSA|e7p^3XT
z%PN0qdeuW=N+5iO!<b-s#8PMsnXO~of!9P?WsRZS-8aQm^#U_ZS##y#54P>n==BGa
zEs>LZq}CHP1RS{YQQqvbiY=A)7y63UPAw}uLz<8>elN^8jhI~*i!mLG_cs^!p+B->
z6z4xd1wr_Ckc;6|E7JROBoF1;5!JGgH?k$f+i@##vF#3mh=N_6xJAs^OUmt6-p8ks
zxkbyLd07Hhf(3DJ5V&p{i+>2tiv{p_S?Kx)C14+491CVQlw8x_5=16NuI)|4B&Ip!
zmosH3_7j?@360zA&D_;ivy^yJ21ha=MPhBkM<WxvHYF4+-_w2<7+&<Y>=V@@a;M3o
zfcfxN_l0^NM+}!kDH)6?8Q|dHR0q8JH#jUyv)~oh*QfmaFi-UegQD?`9jklq()nnC
z10;Rw)>=kU19cN)VpXpORTp{2+sqKI4k;8<w*%c<k)V;ZM6UxsRJz(;772qXOh0#P
z{eUbC5tj+Z7UDsX0xo}|OOh?tpc+ZF9r>9k7Q%vW^9w@hj_MV<YHsNE*yIZ05!A+~
z{hE9-RY`>Rd9Q*nxVk0~MSW*&P^&x({2Qa`_m&9`d_0iUjI#dZOz*pFg4GZ<SgYI7
z%4i}-C_G_eUs3ybt`HJdA+dc(YHGE?Cs#8v`fQi!yG>1=UsV;~l1ZCV6d{*d+1rp0
z6bo>YO)V|SySj2gWETG#%KH2tpe~W(ulN`=zPEl&Z}wqC(MEmV(X4w^nrqzT@u5Jx
zGI*aPj@Lw65lMSIR%(=T=O%;NN^1P%0}TcL?l%r%x^(REXMy#nY>NVB;-aQr%&E8@
zXa#LjwxqtLEN&YwM9DqOj%@*kFM01ELslkdxNqJ6hvWbE-u>a-J6BfH3*;j1-N(_X
z2)tj6hsw|Xm~>k#5v*hcxlZ*F$_5V}oT)1=TL!c_^tZH#;+}qQON%5CINp*@!O;0)
z*pF=C1A5&d&&-#64m48G!t%GjEJpGucL<C|YK!Tjt>aF(UPUf;^_V<E-8Sx34!fyu
zZGGEaHkNaFnBBw63Bw<9xuHt8!yf*TBSZ340iv#??{4RV<Jd;l@azHDN(xfe2eyXh
zrMVQLM`|xV)v}C;<Mch%fQ{^0S|G(U+>8PNiYz`tlmgW5OM#!`!(M62Zw0lXrb8PS
zUDu!x8WNtpyvMS%zZ{hj8KtrB9AK5aLu-am5_1}LnpS~b_R0cJPA%2mfMjn}_%t_a
z4|_tbc#-Lj@|yEwnH}!iALbc&1xH<bmm8>&vcEMYcgrG^S*J5@pfv5zkT7n<#}F#e
zf=MEB*#EqGR6n7cF(}ccZgjzugO7(J)pdG{jTCmFR{Z_F6`b@TBFuD33WGgRByDp2
z(%$*pBq$eACg!}KhSP^-AcD#XzuU<}>p8<D_?bse0P}5j-O=3|;vG|kZAg4trv~*I
z8{>cj9-8QuO};H}+nvF_(Eey3B99@`RjgR+&Wt!T{YN!ebwu4qmAthLgfO31HcJ))
zZc==GS*I5=g`SbBZcC07%|M*6uZptEXHNJtp(ONre*%q3aiD|!oP6HMQr($AZ)xq#
zSj?*aMofd1Vw6Q#=c1TJbB4T$PGAg&+xsPvW+ktnK)8Rr(Ru2xpmj+XZCTP(#4<87
zKYw{tHvv{;lEuR-P3zX%PI>u~{o{I?vK%ujBL<P#Ymm|6Iv<jbq7fHvu_(G|dB@(c
z7^QeB`%Y0#xUTS=3R_g4`Y+r<U7?QGjj>R&BoDq|Ws0D2sCz%vj+S0dFCec8!tSl2
z%zypYIQ;h-S<W(GZpfPQSL*6}Qrnqb@SAX=SmoH`fpv*Q*L)%S`GOp9yZ(r0G;rVq
zjmKla*+U}dcA;pGA5BsB)|-*??-dosM7YrwJbxBnxXKk3G8TE|%QK+Sf4vCi*w5#?
zk)A&Gc+^z3#I7T5;f3HBo8l8@QPUHSW+%`ibZ7M)x|~!jZuhb&O<yGXHAk7tC0R$(
zgN#HuEN~a}k}fx~gRxbt38al-Mxx%NStZIB0ea+~$Qq)&+*}W@%hXBU>FZp{w8&+~
zUZ#k$c2BzVs9medn~B*m0bVGz@gZwOp5~81$4!4&_BbX#y%YIZOSPpYgiw(FHg!?A
zVXJ>+*6*u-PQkt;-L9lOBu7AInnSYuHv+;~E}oQbPJ4ir!-^maaQ%J7M3NX(qFj*i
zV!Q;llCcrjSe#?(a(I&<gqJfgk@gdGQGZb~sGHG1&WQJ-|Jov8gkEm=M{y9KV8!&M
zXJtty(cGM_sYpmbfBp#<jw_Xv1h}2a)z!Gl!@DQp;*Zzk4L=|sS#)1%LUW+yA}oYu
zMg<@|VM>&VGBt8XIVpWnLM@an3oQ8u<l)RFt}!CJw>9zzHL7sQCu1}oM~53Sh{Kyu
ze*~bLJahW$ggoP=2XApBN3A1@O2caeSK=&~HQl&)){{a>jKA3nvAa7b@9H=X*I6=E
z4GV-q9PxrzvK-*u49b7cK<sYlBuwm$8i~06GkaVP%a&61CI0C%L@<S^EzwrhI~Chw
zJOKfL$xozJ^T0I&0|hAzCW5!Fy+Kzb^mNt(9!`F8yMuox{l%Pw@o6vm$$sFRL-mWh
zjn{>gZ7xK)+1LZ}LKpH7g3t}`<PyAO-V0r6c}O04g6N#-yhqlfS#h#!s@x_bYvAsk
z?HGJIj@UCT{i*TQl_T1x|9ZuLe|RaVSKw3&*IvcmR-NH#954!fydPGjX6{C1L7r1P
zd6Qu=aXC?|`o_wb{-max*h;T7?akDuNQ4yc=!DQZwp?{XoXxbuxvk3_PvvS8(tHu(
zE3K2yE}Xx_F)h>Xn8sTIApYYD^@Y#A{Am%>OR&g6AJ35GA)kh=`Aao$wZlK9FwIRz
zPcK)b4qz{h++c0&+9^<xcDArxoulq8qBOcfp`eB=lg`hb?U%=6?4t`z_3(x)IZp#G
zJcS|)BCYr_gUZWyE$~ZP)2$JVvVQg0u_m(PTiY)4+}9fJ{gzAf2;ZUMtad<tQ<x-Q
zazRx<fe>f9>)OkrhEcmL5XZj4u9}j;1~F0_rP7b-^nO4tI+kk4(=tPm4$yUcex!C+
z^`Vi9^6av71DKXyIjiueEP~!?eLWJFBtsg|+hesZ{v3~jvmPlf)`^qv#HRUKta#En
zb@mDSi2)AY-Z$W6i!1jYWy&_iCdVt>VYz`40g9HE_S>{wgX7wG+8cw(AkDU~ZY~6R
zY0Y5WAp;xp1W&FGcXlOBpct{bU5N;Fw8qE?1UE6%57U!Ry={=|9(Zlxmyb?2LY&VD
z_bjD|dqUih-Fi~<5amH1yX;2bG>MNO?zs8hcX@nTh}k0fIvST~DoB+@6Y^19Hr(Je
z`~kh^_G5U63@PuD6VW%~d}qR=C#r9Xdn8<Vy@6EpR5*uQs>Uc8Y{7P7KT*;kee8AL
z1n;+{mB|Lb3U)exTt*fIMG~*iyUW6o-%1Ilxihk|zO}Y~1r23`{o-cZoMfUC-)QUF
z^U0R6#eWC9BL|JH>zkto$g6j#A&vab{y$U%$$fU_#dnskjHAYL0TCkKbp2tH`!UlK
zMZd%K$)CS5v&_Z<e~-`I8Vx_4m$-(3Eo|*^^NE!k5Q50@<>h;As6Bk`Hv>a=e^mm8
z_xJbt_s9PHaULL*ES$QW%lThb_T`4b8)J0IW{PH*#%d15??=ni)5z@#sNVOMPZ95h
z#mK>n&b@ib?5OO()^k$i6C09I703rmg%fEL3uF^oXp0P2+fqnd<CA2qcUM-Knf_=!
zleKL|?O{GAeNmG7fcGt4o;{KZ^vwRQV@5gjiK97cd+(LiY?s!G?ULk+A_ul?GsD5p
z`PAdUaEHVUU<5gJ<C))|vWJG`e1*xS!lV|0xq^hbg?iil9uuqxd)UwJeF1$@zyPwq
z%FA(ZaW`%sLCCFK(nVryhN89f%+^)=IDQA8>-o94@(W8z6zh#I>?mG(8${ao^S2g7
z?~ZkdmurW<soN6bg1>a0%+JBHH>@FqWNMKK*OEyG%5iIx0Qq-YXzCgQ$NFDldS3i^
z6h1gGozz2?SVYjj5K(61sSqL;7G!Phf)Y1yq-ZHE^~5nM7&lyRE@jeR4*7`&^lI`Q
zktDMOERmG!^TIT4!NlOp(fdhlaR%62K^Py{NC^gl?LzWr22i-8sU##M+&nxqR8_&l
zP4dn=1*o{-JP0X-z5Dj@r0r4}kl5MC(z)wRJt(j=#G0s_orjC<I1qjorm2A*3(i4h
zh}D95lA-Rz#$z{nJsibyJc_-P&BRF1>$jq6{ZR=dqaV4VG$vSd_9w6c#S<Q~dC<0+
z#(be*Al}he7v;A@=FFNVqDSL*O>s-263m(Ee(FE^Ym;Z5h@HQW+gWXPT%?RzkCW#e
zM9I*Cj`XG!f@MX|5kgKhA!~oI6f&a_1s#2QWmd`o0xpz7$$!B8%?KxWmNz%e!BX>>
ziM_9w*u-cNMv4D*g$|<I`noRU^&S$3m<+dqBIi8^AK2VZSuyC`NS*wDw3(p=q2$e%
zE%JOdWtDd(pbKm+j*nQ1Gtt)FF=Qu%#^B1opo)Y7WjRC6!h*U+_b3*of~)8+P7M0V
zLQP1vphwQu>gYCa$2}?tjo<TfnEsF6uwr#rY8%(+sdDoBYh_d45!jg-4EIud^2@j5
za6b00Glb%@kW~;1#eLMEnRf<TiJ#fNGU~<BHfAO=s~*NYf#`3==~S}yH34#nta>4-
z#Ir}C3KI<(yLwaQ#8EcyvWqY0`Q-++%C*?k6iMt{B$4Dr{U(X{595(mU+vI3M!V*W
zZ7T#Wx3X;T6lq3eBDypgzwtNlFaCQ7hAd@iuyoU;#Ka);Rc*rynepS(T@%X6<(X2=
zot!X?=FhqL<EnL8`W4uIg;pMN^YCB+ocgJsTgk%LH_M3+90n7DZ(Tsu$#A4^6vC(F
z(^<RRGu4Zu-i0wAQ*<8Am=MWrX$;$$A*~Y|EnZZTLJ}y*t}Nfrj9c65*}4|kV&k*0
zVCMO&NHooZ;ea_+KlwYk_tC`0#XddH1vYFaHGKgEEq-chQQFy#W9kA6M@m{$l=oB8
zISlm*;wT?Oa3qZYG}QuFh0yY`;E;$DiIb|}4HT4h<?EsYq-0zb@m^A&g&<}KT^#az
z#h?dBU2c@6`SnAnnP;}U1BwEMi+*Cq?cBq!JPzFr4xAY%J3BidN7BZognDVPPm+m=
zk%2*U8pP6ZYeIg}?&_tdc!zSl7DmlDz)=k8oOKJYLNqKHsT=31jWEMpO-}Fb?7cDs
zl{XpQsXw_2KNW&;M@Yjt3F}Ei04)z0zaJT71hrr%GZ_u_E|7^YSPz#ylN4Pd5cf3X
zw@FtM<Z=L-Qhb+a)n!3F%K+`oU6rXt&~$DoIwzgs9Fz;Qn;u1q_|Qa`O&1krKP^FR
z{&FDJGUl)lkEc{Awpp3Do*bgJ40ARjx=b92+0Jn4Kwrw-{Cwe|fPaT8@uT52rqTbH
zLY%Hl)X5i=`VI(Yag6ECR$Ogkk|-*C80tsTF8mCi=fpKW^SPr;%PIW<*kX6lQv%5t
z>*IwqTuK{EJ!DKobHB+sxK!7%82a@=vdGG?eW7PSZu5Km@!HG&<9j)__v06rv)xdw
z@3^Deh*2W|_ig;LCt-d*K}@#IyHJvT%}rVKWCWgZ9~4-ngktV1q*fV5#KDS?x#-Vh
z(9kL_3X57SH`#`hjJIi%7^S;T1*nH40u-7Cd#%~%XS}E$E^^ho6>+Hwo<8rUhWz|g
zk~p&`Zl!j@>_0kluCMuR6$xD*3Hv|99MfoMeDsD48i~<9Ke_mhApfHAUIk0C+}G+T
z+cz730Ij;l2{tFWX=0bQIxhZc$LH?Lsc*6)1Nm;;cOCVX>hKoV?(P^%E*Q?ols~-u
zHrkbNddd_rl2pK(>}$lXhJ%SQi6gUK%YrP1p{kWCOC#>=%mG+u5<zzqH5@`xQd%Hv
z>*&aciFKhQDrjlZ(a~Le8i<QWs2{ZCyl=456aMJpJwEgvu;4Q9#)T*Nm5uqx$!ZPZ
zcOO=2WR<EKt!xD{Vx!kid-l#L<qHbOku{Xxs9_A1G6mb29SS5HMX351_ry!T9&$#F
z5^alOf+bc;N~mE{VS%6DARzEC1jD-_zvV(pgO?k>?)sZ$Z?MoJ-7iZ`gcZCkg#jxW
z^mI2a{Fd@?*kJP97Ywd}$Qb#LZ5smgs}sz8P@0|Yf0SZiF?{n35)eb4SrH!Bu^sa|
z01$FNFeMf8Ovub6QZK`4T=ixy(~KS;R|XBAUY)g)@wYVS!_tcKqRNKo?gH_TO_y+^
zv!B`TW%6z$bVS<NPnv3~OGvze2SZONYq?QlVBddzzu0m@+KcyAoLAYNYSH1BY06+E
z?O0^l2&E??j&iptR@y>^ego8EIPI`HMrEiX%6A_Tcermv*hny*&@w{ETanhuqD0z1
z^rJeQ8Rr~MDgEpb;_G3`>hR_DO}|502t-1qc;^ctoa!NF!a66{a%S(G?-*h>hnQNC
z;Jt}ZU;KOdZHI6D{Tqfv5@$|E0kqb@aP^a}<Vdyp@-Qf3Jgs?(=uMOwvcZnq?bQft
znRX!WfM4~`vJOJ3j*tc9>J{j4m=Bc!#4;~(7-bL-rcUUkykjIW?T@Ag#FEb#3KA~*
zz6EV~MA3PlP~2CbQFGm}vLFdU9M-}m6Eb)YAh<;@^*z1OkM7C1eKDCz=1IA-JqHGI
zoZk;UpNj;3PX<p+0PBsX7?9(4sK#;krseiFU8G~XDj(xLd=;=Lf~p9pW>tTZP&tMW
z<wpIgel|`pF#W}x)Z7-v;Sl3@&XX4LUYY2Tmt1(NCt4Mi^e^NC7Y7>slGi)u*jt^|
z(WNs^ahLVbN46U+C%Njl8#naww~xrL-51Z}5qAT)xdPuUXvVl5hRAXVR=Y}t5&81n
zAUnQX)Ruk!{CfX)IT8ovZmB~!+KNAUK}sI-3(N;)J&F;`1Yb>W1^&z<vzn@u7SWxI
zv6}ey&gTvx+26_EwsNYf5BHFQWGTDu>lKSK;-6m0tyf@`kN5^oyPETbRyooh;m1R(
ztn}^$EE^m&G&LnDFhrBS`BoG-@c$g%Jogm7P{NS{C`}=arEBXPBT5pW%+3Y}fY_z^
zNZjc%Y57mnYaIkT94Igbk=?|#YdDpX8ya{(Hbe~Go6HTKOHZMD*FmYg9l9z9o4zaT
z?8J`rd3cv4Gxx6Z?(zElo@QP?pHO=pSHD3y%ZAUz<pB7t9Zsms?HvXoni9}OMRD{$
zJR(M76&K}SlNQOU)>tK5wnfb7B3en|;z%oCNV*;n1Sfo&Sd^{B*TBbsL4+Y5X~3-A
zd#_70G)mL6r5Y%;fKRc0y`|PByQuljAd84Gk#NP(^XPsh1fO^|2H$x^X};R5XD1~!
zHJ@03!gLXDAnxb%?5q;Eo}xsLy0pw~&GY>pCTvvFVPyfa|Go47^8(aa1d|FHizmTI
zOG_h!WjPjuD4692Miv&(pSQHM*xK3UxBD|fb0mWMF;{6c^15(E*y7XDeBN*F?>$k4
zsA)LT`xaX=CPt1f8ALVBAJP(_a0+oEHN-EqsPm+L+-ANF$gt#$a#)6Wf&`fo!;b!!
z#6ZlQjF75q8!omzM~iEBPT8>{zOO2L4P%h4uV_w*D9_ovCY3S072D^^YlnIa!A2ON
zRi(X?hN?Uh`vJ!sl4p!svv;oScA1~^h5yclExAxFjZNWfRpbr<cqSATl`k$7*!bVV
zEXkA|m07yE;pk~IAVf?B++e-?^u>8!Bv*{-?Z)q)33seGe-Kki&+51tpq^6E$<)nq
zcbayXUgG-UThgCea#vp<g~-s;tfrQiN0L>9ho%Z;fH?zTAmGUW_ep7*fRuCx+2jr?
zQMUL0YHM_ZBW6m__99Tn5I?KEk6}+{V0V)~m<<@q3d7%~1E)|z;A-}yFqDoj;Wl|^
z1N5Xj`4Lf2`IxF6%-0u$rFQt5fpz$ZJTv}5yL~@IEY#|2IM2=Yt!2Ykx%~$e5`Ku=
z6;!3K`E59{QTDB*qL_x)C6%K8lMy)CuqPh~P;%`3m!MO1I(IxoHFmuIW&lN;U-Bon
z|I#B)qn2UUZh@AKTWrFS%?CG9hcaRCaCc^m_}i}AVM#}ONE@6*U9QPDakizH({`&o
zlI1h61W5bw<;L123-Pxl-)K7eXRyfqdGR5(w%0Kx+F;IomS{g;S=;{5pmjfNi=dh7
z{uPtI7q#fr^tM;bU8POAMqM=D1<r$oCW3##J7M(?DrT|rd$;r%-(_Zt{gSoe;JP!a
z#UmCmdzP<+X~Ta%{cw48`B<w+7gE3g&d#z(iO@FDH2D&cb4{N-gn<^3XvI%1Wvz#o
zM5Csr7Q_$@6AURSCd67W%B|#PG{7F)76Q_zXw<}~-!3gJEp!BkfF%~nU(0aH%Yn7#
z(!9$}W+0ZK|L-3E2R@ffJyxp8?)%54(!MM_YCexFJ)1k#(ND5_9>gs4MCxzfaIbXM
zAN!i17}0aCczqE>gtSMFRgAu?ajz&m9+w-;LvUt^|ICOeTPww|4~C(r5O^Ba`CeSq
zvYLtPhvZ0=V5M*_rK#;9XVw2shZE?iqgdGoV+%;>hfB4y;|8(&_l1N-WxQk7OTwVx
z{~$FMcMdgZ<%Ow}^>?*-0TyYbR+bLil9SNw-7SyBRwC}f`-9R-msJK@xq|Y7;4xyJ
zvxDzdoIFH@#dN)w6Q@PxOry0{BLH4>cZ=T~FGmb)nc3T;EH`)!2(F5GdGQZfX5G;!
z#(f8Fs>JztKyr(8+>-Ga`SSx1NNI>mg@FM0w{Z}BF+4bkU}0fl`##gOI(;cqUs^_n
zaG#taLlGc)Sz3*wKh4Rlr|n;V`P>%RLG^`6J>3n60CL*fr-k2b9|KV5adBwUC#06u
zRruP`?AARRoFt@ybQDZtHK?D-N{x^N_19A&d?|YOM=Y+wy!Vg4QX~`ZHrvRQ8ftcd
zV6afT%-GmiK8_?&6Gw7CKR-j~K3wq7)z>Fh=8%Emtvs>7E(_C#P6qgM+NR&Zy!Lm&
zXd!qkm>6x4OcoPUL~HRBul(iZoc#Pzm4~tM@pX=C!hp{Zx3gm`U~oBG7bEjrhg6*2
zzFBsZQNuy$;oI}b|94DhIo=#CO6cg|kByDZ%w!pH_Zc7|&i#Q9jN=>nmC7r6zuZQr
zeFpYm66CbA9aj5Lw)*n_xeFjsQ}YvPdwIJrxXm}h^TtE6cE5GwnR3Bj#O<fp2jW2~
z)d1{24`S<GUPM1a4PFu5p9R~VsOp6gel&j~V^~Ad@kIauSVCXwI{nDq0FSp!A@&?$
zI#$nbNDZA~w!e(*;FEt!=y@lldx`I~HHSQ}1(=h%0l4rP=NsjQ>>Am$!e&!O_{{-=
zsjR`s*diD9bGSEQ-&GirAT{sg7c`sVDBnt3XMSZNxH^Y#bTAv))rwSpLqX<aPmeBg
zv8i``tyoh{_*<G#)q*RH4cJ0D*1X1?mv3+4cc(YT@zgLFuW>Hz?a`kP5X6R5YScSF
z7ZD@KmZaOLJLBd)ydQBszrDo%c0|?GZ#3D|*NmKG74lRYEWxVIkt(04NOw72^GCpF
zo{^1B2B)trVfXUB`bU|7S#&|Q<xrAusQLIxksVQ#Wbnp@S@#Be5)I{u%j)N}o?4dh
zh=|L_-7L$N{OPHwE@Za0&WoLqvZ^YXG%KLk%wIoQHe2a=nW-evJo$?V=d!Ui?HqWM
z-ad^A<gswjlaC+!4qw}d)sz^s;!%a?W2$^dDbb><N6B;uV!&#Ul(%F^p+)JEu62d0
zu1KDknUsdZY!!$qW%Nx7{SjT76zt%mYL!}#Dy~m$pUzMln;;UN5D=GMKGf?(GJ+=t
z)0O133daqn4t;gErHhM+MF!a=yCG+bLdeu$rMio5YD-sMZn22>A{v+=5;$MT=V*<*
z-G&Xb(ThGK6<v8)s>f>@mu0|(f0>Bw?#dBez+nC1i7-^04j>MCvUnizEVOymQ5zPK
zjJ*yfQiF4SXC)2b7s{Lqpu^^?M&dGSpN*KjB<+lRxm@~C1$0zgTwKpz#^A0c(nJAP
zO1nb0YpRg+Xn9^qNeL3r3x7PfJ`26vZ4K5c1_X!zKjGGQlV_%okgb_H?X6qp-YsVu
zxy=yrCyH8@LOUXs#Eo=>(~BN@N%o{Tj2n@Ge8r$v91NWv6GuqyJ5m1qFI_SM1a~(S
z0}tXA@6)$G^I|<McRUk=sFsHZ@AB{8!53}=e3_yYq8=W+KYsi$^jJYnNn!0zFbToK
zq~cL@7r#5r=SC|)8IrfH&aXfv`<bU(pvg&~NhTHR$*zR0G!2vuz^(dU6_SH=oOnRK
zZJn!v9?8mz_W1Z%Q(GH2@T%G_kig^u2+(O3Uxe*#cw@iZJS95pFTH>uTM$@zct4+G
z2h`@2fM@=nyENmHHjOOUYHbMZZ?mXcDnn3{Sf~&-XYLiS((I^4IK%U8M81{?U;b(F
zaWsJuTQ4ig*#wt4NN}w%>B!mLwFa{&tZwtHd|d|X(~Vi1{gj@kg(VGk&n8`KFSCRm
zAEMtkS>*5(o<^v21iTo16^0IY%<$5dcj`l5kLK5&Z6klV<4NBS{|2|J9;y6&de<RH
z&>!u9Mi!Pr#pjWw18&s^H7}?<$YW^?h(Ap>uO#zsw38Q&lLc1GN+CGG-STJNyCO39
zUo`sul?pE(08Bu$I2(4-4@=#Gl)0PEZ1U>Q-eUI1Yc>zLaH|qX{_&6JYU}>yMZ@~f
zd1H%}mQQ<aq{LruG(|c{SCtetXmOY~xCF)l&}SBZ+s5aP(Zh2&AdH(>YC^i<R7<*&
zlX0)FqtgX1?1ZZ0X!IpFUbnGfBRI+uY8~?(R{`U9JlxCUDu05iy}RblS6@r`;$|m-
zU=}_*-zsI2uKuVNZu{ihHv)Zvcq!uL<;eIIB6~5gU#&jnNHzzI8jK8qeAy_Nb59a3
zLA7EAqymQB!_YFqfz$#!q9huAD#Ex(t4un;umgFY?^AazW>k6kZ%_(41F=o;5MVAg
z)5^=ey1>B?fO>Xzc7<+@dB0>`v7MW{JLshcz<8e;4lOyfT_#;}a`M&H)pjAilv;0t
z14t$_X2-5G|6@qB7<-<16T@!N<-447|IJK!50brR^3&I;rs2dP=N9g-p{bt*a8RLI
zqziTOk~nFm`|%_P=a}nYh?AV_8_<HwP@<{$Kk^a_i9?K~5M;2cMI%LdQt935&cWIW
zvWh_3V0l$LhAvDJP>EGrNm_EN(Y`5;iqPzhW_MqRr{()}!YNvnMq{e~t^YSoI9D&d
z76uDa43#Epw7NcC|K67bSVBPq8was;))Xb5Xp;(r;7*W?Ar2TgFsv}m&7p*?hQonE
zeTp&KO!N5gP+C^Dk*Z@nB=Z#3167l*WR^b)#tabQLV)WBlvZ#f|DD{xBbvr)bkJ0e
z<Nzm)gGMSKDMa?qDgy>{3Dj`(etEww(?sf6n>agTH?I0JJS-m_9ew60VCN_@Tx+Fs
zaX%42CZ)FU^8<ciA#hsMpY>v^<Q)naG}!m=Ax|zZ1CrbZ@O_0c@c=9^Ffj!ecZ6OA
zM5ugQH~4P4zmlNXIY%Cx0;LyPiRr0=ZvO{);A4;tH-hoRpeMSH0}3{hKw<*0t5RS^
z@%_9)LjDIRfS>jL!SorLTBx<~%TH#yC_^KJ$GRKUT*$Il2&GVe=rmw7)e+l9)3sE~
z7i$^rPOuWnP)U-dv2}E$&sHvEKq|IVH!)EG_YL@2q|~Y%z5-v_`wt%oiHJ5mIvz<u
zUH#{$2o4TLe%3T#ziGeY)~=7*RY)`3ugt>oX%JCj-`=Y+c~0wid1RYjoDE|<AR&#Z
zX2L2dZwLZ0#wHz(=;S`PR-R8}aI+8+M(|}UtQKP}vn4CL<7+PoUBS#=6T4P>>#=xv
z%e8mczt*}mex_Z?iX8Gd<X`H-7h*fVKRC=FAl_c`Fdj;de@#;&o}1#bX6JW8QE`UV
zXX@KU{bIbY;nWPFh%e*g?Y$4|uR0wMxSv0N=H}(C^MCg4-r#CYaPf-S43h0Z+(e~4
z!y7aF7T4*?Hi(LWO3L;TwVq#uI+eajRzR<nmDOhO>xSQCZN<~)LVWBR-B>_Ja|#G}
z8!s`lvhEt4de3{=jqLmwnee~ivR}4-ez^Wu{(yXmyiybc*G$y#0(~b8zkJ)>e##fD
z|4dS5B28W|4DUypAR9;$Y>Ulp1ieFlSFOFgzV13Xk4lK`yHjmRU7;)T>JhlW%2$gF
z2j9g`%IkfPWZTkxZ_Gh&XB*HTWv@(=<yX$rojF#*j)7?VN?PA2ZjOZ_f}<_*oT_T#
z^RzqkuKE<i^Q-Bz_r5R6%_(dlR%Po4{Wse9U+k`~ug{-P19CqT^VT|VNoQtej_Vmn
z91v6~wvE@SUK7Sx3J*yb<pmpo01K+cs7Z$qXTPMtZ5vJELi%@+{2xTbFsy0TAvIY9
zE^jSoe|tLHlb<r^yGlPIBS>yjgvqB<I-mr;9r@zaSG=_s!o8CB!y@4?Q>gW!B?~TJ
z9fxuT;qT&15%d1vp5B(ch}3<@F%4Kr(D;utcZ4<7+AzyGJ1rTExI68dCF+bcasX*>
zXB!tVY~DVqX{EZ+WydS0&@h_3GAGKKX3~N2To`CEm?=zCdwWnGL1hEa{ctKXBNNlG
zOfvY+Ie2)|JVvf=Z!MPHjYS9hM0QEoK$im((na`b+Qp-b9GJgB@9yhB=X8=j@q8&8
zSDI4Q&+{hp+$~pZS|M}srUKR`FzmCLL2S`m{Q-Ku@^TssOew|>^r=(0QDoH0zuBWA
z<&C@Q>cD^UK>GCWK+L+No;fsCG)3~9n1;OG@A|}x7H`R1bTDyJD_ew3gC)l@f{6?N
z5YF)y5uH-fptoBxS*u>0c*-Pj8($R1S}0c$`60=U?h*SN9Eu_-21|JO+uwHBRVLEs
zFbOooYJ9A{7ZvFzE=>z%rKO^2R^1y;FZ`$10%^L_+?UFz)7(HM2KWRH4(@I$-oJN1
z_=cmh{o#H8z<^#u1?KXN83B`jNABvXF>rZB_L+n3ft{a!=y#C<Aqj~jjuiKFPfeV#
zo6!XFIsi!^mmR%hO$Ep@zkdA!?G^a3*xK8F`|;zm*7vjmJ7G{NMee%L0r3hfD6U(d
zc`qhjpC){tw#fJPL|#>}5Ybapjluo>HyIA-2BF8JI#35iO-TRXh0PJTRfLw5L!t`T
z6QTWriEv;z3OsD-Ov_UUgCuorRBFnLjmG*42Cjje!obY@PZ`l=k=D?_2AW;`fCO+U
zmzPIpS6%(~x?=2J)#6@r<L8!5fD5-jJ9HX^7=h045qoA4Pn0~LrON2x!KM}M4wcLT
zdOEifH37;62r5aIrD0%T80_${%j^dgyp{#Xb)sfwW-sTz>s<{juY8kbixJWceOVbL
zG{i6_xxu}~pb1lDAFTcKg+qX+%SYG=9-?T<Jba6g^*bu3Fn1h+tvKXqhdZizUV8K}
zmGlx^^ZP9;p~p?(@$C=le>b%}h48<Knr6VYz@~e@)Gd&Mu=MKp)1UTRqSPDHf7@b)
ziIGdxaEK?0%Oht+Ztr~8rt+k&Fg)0*-lS%s`z!w)|K^(^D?y!ziR)T!9ND{JK(aKy
zzC7x+y1xN^7^rWpAN9XX%Y+=~hQA9@^%wL#tIWU`AId&EfXqT*xc&I)3vy@QlQ6+&
z=_pAmMMXtZJ!CZ$Q?vWI#&#(dJi+^LWLXgX#h)$f=5<INO$DGMzy}wT66iJdi?wa0
zY0Qil4gJtxPr|8l-6YH2EoXaf|GfhFc&o_RnRUPB(XFB4>_bRzyB<0MKkQ)?E_J}i
z3qu`L++WYhn9gfAT~A|V`%{m>*uO?=VB<oFc+#q3J8jD?yjUz;={1?VuPktQK5$i+
z^xSKp1z8F_9xWmns9;QI{eeY)roQ^T@#hX3iVO94qATZ-MCbVjdGOy2n<!{N%hHGC
zZ6XMrVbfO{e02FJ<RQiR<(`0?d@qVVXB}L5;D>@r%mYY2xbp8R9hI|D)zyDBEQC1k
zpc>h{>0{cariH@;J3rBWIX}$*BAaXZ@kr#6+~&7zt;Yq1!cz=_oEX;bd1}qgsYriZ
zb_D1)aC7*x^h;uNuCi)G(<BC8;Ll_l9d-p@9}8tS&s|qYje3O6O?+x97@v1!3zFYq
zM>IFvkLX>S36vL|AMJmrFM7B$o&8jSFgxCHekYdmbS&2Yj!gNjlLR7;y7tQ8utGV4
zVcy>^*iT72A9$3od5UAC!p#ztOE9GpP1z&o565IWcK$JsG>J~N%0O5K{u%-}YHDgX
zH#b#agPhot7~U<`)ivdZQU>FIsi~>=-oncDsau&KLEHuq9UW`&W(1%E0v41@<!VVP
zKn{iWY_!#(4#vFUaW~EbYD}nj{y(1|4OzV4&tOGZIgE?M0OgjvcNQ$gpEI$$_!I&&
zFn^{u1}$>IHcodn)cIAjqb6HKImaCksz>7^LL!!sWjG)%M8hj#%UNksQ98aj|73~D
z&oNH=5y8mAxsbHKuhS3+1eJ!c9W~urUs$V8O0vd;7#Nt<4wzlx8(oxi{n(KOfkm`u
z_lmyaP#Nsl<)2w&EF}}aLBLTZp;DRO+)<bp8t`nLqZI5&soTSJ_18OuhY#p+Ym1uo
z#=mj4ws^~|?R>+riULI#xOPg96$Lyr;D`X#kt$Jk6?K=duX~|Ei3Tgbv@{ZkvAbhm
z0cH6wVAR0i-=y5*jZv>T_RMS)i&0C8Df?i~J0DOmzo|32pROuxy;_`(Z5_6ZxUQW#
zD*5Daqo=(YNjuifp^C3bPKcUR4Bds(&l6S1KMqpQ`?Zs@yzH|7TpzJ|b5jqyq=1}4
z_wAb<1tV&THI?!vUxUSfhB@ff^=d7Z417L(i2ooIQ1R;vjm$;X(A4Gq!jezb-^TFg
z%Aad{emRZ5va#uv&o<EM?F<@F7f)<J&y^+Qm83`~;o~Fte?`h|7yg#k*8A1l_JIMB
z7iS_OqSDH^q!IonFd^^`>_g=O_W=|Dr`PA(%YlHW0o}B)a^cuwsUVPc?=o-Xaeg(e
z0wm7E`KpK(9n(`96n&>{7?vfN`CEx6EzjQ2SsoJT_kpl@6~07Gm==5ruTg$&E_YUn
z)885z($vdB%?SR<b(Aom;Y2}2t#R3rk8{IctQYQV|MI6j(h@pBXT{o&kT{dJNgi4c
z(iurSw@`kbL#s=2Jp38zBdgfsEk~?L?|=Tyg?c+|;CrU#z?Gu>I(Bi^gYF6vrjjT2
zfM+bdc<3M5T=g_u)%FvA(W<}UwtdaVK`(9ou`)Os)*<W)R_OZU&aa8RLR-QV66~Lr
zAs7qgd7(<3PJewVVH#BWyTqnQNk*i#t=_+XZ-SpoI<ULE_H6U9%^y`pnHDk8=D65k
ze>EW1kX42Pg8~D=X?qY8BD#-CZsa-j`nLj`*C8R{frVu$dSDN>gysM=SFwu2gfrqA
z|75%OcMo51>h9HVXMdb$4$aeK0=AYFpE4Gw=oX(tXP5>HpUW`<<UD3HVZERvDI=q7
z50$JqeB%Wv&bM-F`!usZF)bl)S6_^CJU0=k5)nY}*TYW0H;XQsuS)S{41%492Z^0J
zKNa2=tBx!e2@j-QeYy@4k)7B&j_0cCSv+Y5y}^2C5VX))Cv*<#;=wF_<!IJGD-MzU
z6RziRmD%0edajR|9(}0g;4io;VR&pj%{4fKD(~;`g7F^ItK-hqEiIWQ&$A|eJ&jzX
z_!6RanLLMFBRi5sB@aTo&SiX;g8o<gYX1(Rj;E0hC-ANO!`A?995mXxIbzp$AAC2x
zgSa{l`t6|cP4!}JW9=j)3GPEX#QX{hC2D%04tXg33+c@>Y$(xsCauPI5`PS`XUufO
zg*&_bKS#pIG`C$`CK1UZ5i;0kQEM#pif<Nf-+rOc&|eRq=*s+g!tF|6{K6x2%X0Lj
z5$Q)dH>2WkINQODZK>omaw=xCXw2yWk#i^T^z?JlV_66a{gn~raMF#p?Y7WLE9>Ou
zz0anB(}mbwynd4hk#ni-Nt3)qwGaHv3fu>QF7m+U(2tkt0~xntK|zrJh?R3-?ujf?
z_~MOJH&W!TM6Bmt5?{;jk9)1ZSsO`WKhR%aqI+6kL}jt-`O%lDaZ-oRPwbFoo2{Mi
z*PO?>b_g0;%P~_Ii#?7W@DA#1+<vbcH(@wfn7)vRPFTb+8TzS(@0i~#)E4=3YDbY^
zaXvA+B%1nZq!Zh>{$BgUeU~9CQ>Y(1{@rkAbC`csh<<Av`tyyWtBlVXm9)QI)6<TP
zRZVg_ZGg^S4m`Aftt=Vg4i+UVqGnA3poqR_In}eofXxtGe0;J#K98}zwtea~;(@a>
zGflmzc3%G90Izs+<L>S4J^5+nkP_~rd95mg)4>cQz<4#fqdxnoBhMzGPRDRodd2zq
zL8k0{*AOS`di+ZbMs$UtBZ*qx&c9wTnh%Tqw1S;YZWwXt9EH`{3YhUqA+7lZA2|$J
zR8=);Xd2hVy-?J>+NMkqw3-?s<4|j981NKGM1D(gzm+ST(7Khgx){Mj3wIiPfZ}ds
zm7e1E7*+1W)~4D=pZ+r$tHl>&QR4greK$kdof)+&$zDff@s~+jnKyAR*cKIve{<N#
zK~&ll88IITN3Sa}R6$(;4<X2I3Fo54XUs5kbKKd`r+;|OgOlmk0I;8cBXhrHytfwy
zH0U5%B~iIVukj;ku^lG|M^LeyqOx+W(}p;WTtQ6@<}^2Gfz3C%1;WFny`IjG?U(BD
zawlYfDK}9zGz=LZ)L%v%BtXIeKSC!fW-N;-pbY_&!Y&{n&zWHf&?_JsA08h5L194r
z{o6MrUo^PmV(4@HGcXr04c<OJ{GMkN(FI^6O}1qE-}p8kL!x|kaZyoNSC<$dZVehg
z(q=37i1_x5*+&U@+T%Z+s4eIR7xHYA>eBbviF!~_uFo*>ecK9nWq0`^1tz}-;>k;@
zssMzKr;!77m_tY?w(<~UYbPiY)^kfWoxhuRzFTM2kz<0-n+q0(*Cgx@TXw6~hmxS5
zawwSNr;&i-VLaIz@KeSA2EKE99t8%R0syAKs1P_D03uPO<99zI`0JJlK42(DXpb4#
zz#}Fj1D`*cY5<+4T;jDJ1V8-l`zYZ%W@dE9f{GItl7W~$2ZgjV!*dwuMKygF|CxNK
zFz?T59YewXjUAhVNf1dU)}EYFF=XP2mXh5Q^C!<whfqH;uyS9ns*>x><XVrh<i$64
zHq1_a04qLI^#i<hRp*wuQ~MA&czV(O&!FR<$`{IV<Y=@9=YTyEq!S{*g+wOgPHiJ7
zD|`G1RT>e@u}S*q{VTULUsYE5A|l`}B8TXcz+7c}bWRXQrSW&O<+SBOrL?~%C)Ts&
zTHyG1>vsVS(3gr}A-{(!AvyQ3P4T46a5Io(#Oi*$1STf0%PUI)==V6{7cp9K{q@4=
zPeYP@O!soi?Bt=e{c%1%t&f%;d5vz3buYg|Ysc2qn2s@8V;IG@ylgV$Gr&9<Ve1$L
z7chQ-15C&0`lMprRu_&0q7FR?k9$ZI&;6xSEBL1$lD}C?Hwzqd05(C$Ku*}oD_nMf
zYI~zf=Qw!~W9#J~z8}AopLy;AVb@-WXviaUL|{H(nVv8+4|40gnd5&m%5+_QeoMIw
z@8fMY5zv{bl1h)(-Hs1QIOl5}O#zMYBTTRZ2Hh8u&U$e#<hIKRQk=Jk&EGl`N-?B(
zc^z87ckeWB58mm*TGuUHv*!;y_HCz3!0~6<hw&}>9>ZHDN|=?E72tck{Kqf6{V?Yl
zsz+Pm&*&*|&#4oJkEz9)jUq=#7`OLWdaGq2fVZ6N<LPS;A5B$feQnM%Qv7&F(QJ6$
zHh;gMV%%0*qgkfbUVTT6|C1%q?CTMMNN0_&h=q4%+Afj4!FwC6M98N-jYm0^ZHO-v
ztpW|fx3l)N<%UFW{XL$2R6~0M;*L6<tc4%1@MSzLZ3T1RUe*~*>Qrq(lZRWtlkM$A
z{8pWnABfgIC>Jkn5c9r*F{&c5n2DUO`9yr|%<f|6ZQ_%m^Kg6FY2qJqmm84i;HaqZ
zko*6<04@e~BV=sMjb0pMc2kEDS|>4@fB({sT+9er>E#uaN>X70kdUDm4(4YH7{CMP
z*{fgV{XL0<gn(cj@Q!AE(YUsmyI3~G8+tSkzQPBGhrkN{ufhKM6>J(?eE^KxFB&s;
zoNu};pcaV<;oaicp~cBC(O@CYSJewCEp!vdI4ln~eAcW%%4&GEA)~j^=c=85N8u$D
z&ww2_F17epJLNrYqx1?p<06{4b^`OVlcBzB63J78+y`#y$r80e+!}d0VWl1(QatWA
zH07|>a1WT)kONTA5m;y?(R%V52qRQ#G?Y#x%fP3T5E_!8X+Fg_RV~H1tVA&rQT%N(
z^fMLBdk`XR&iq^kheIz)VM-A}f-=^f1k;UWiflj9U+ZQ`<7BB*79K(q@@*k7Xv?;#
zr4Cup5^Xxg!u4%RtFIr%8oOpB9)y${IC7}<0g#Y4FdznM>Obz)k`ve@#DE~3E38ze
zSx{UIH_iR`@UT#o0RyS4OrvMpWx^s8Ry-eL-p~~_N)k#h?H=&0<6sv=xEaRe@2i`k
ze**<91`V;t%Z)y3brKP1|8PBk$iP<xs;@dzDgg<JCD^?L6CzCJf5)-3Yye!FpozX*
z4?%}=+MlE>FP}3ffCF8y(AA^_m{n*$3u9~Q?ael0KMkE`B6%hBeFyuNXi#)8{Bx&@
zgY8w<Eb@9Ox$%3bU%BG>oiERipq;R+(>pjgXxa!T21Us2)2ABOJ!N|+7JH)}pN603
zOCMyWK4p6fq19l;W}Iaqe)r2OSKSYtrliuEprA($zge855-WjD+8O71QUWIcNKT{0
zidw(C*KPOZ2S(G89nXWZ67Vnkp2wJc15?{eOZs5GBEc#HoHQP><>%&k<xJuT7xzM7
z+Ecx38?yuVTDOm>{x`2WPzIO#dyDCCB;K^}v?SI`lx~}Dts{udO))CsMLh&(rA5z3
zl`{~ziqH(zDr&;rihDoinHyD<GxeE+CUhUuYF&6+J6au-R}68A5zewi8hGsF3gek>
zPnX!H6Peg;XO`ukjaKqQ6{z@(QT!(=R+5`q6`v=Jlz6@@9m{vG{_=VEVRx8o;OaIN
zq`UJh`?~)4ZjS0#d2D!ar9fz7<sF7_4A@(s++Cn7A1tQ9xB0cH^2JHeC2wTd<ovtx
zD=Xu!59g;B7v(gj^`1WLo>xj=e?C*-6}Q7kD%=sK(NbO6)TbZY>UKI??;>0~tn16I
z>uYXa?a_=)|FC55T*-NCxYcC1C6xQ@$+c<f=lAQ2YHoZ8z<tZaHVTS5o6<G9+y8dS
zfeWw*!-pyhOG^)QaaTX2N*gMchMUrYAd22O3Bl6jUG6S^F>ogib#L-(QN}aXgOGn*
z-wtj4GpuJ%;AgP@D;Jsz=GtmXdJSz*FRvo(ZQU%T?MchcRm;1P@Bq{UE1fqy1+B~a
z?!~Com9U$&R_wmpv4|E<751-f7nO@|uq=2bmi;E*!VScA7U<1&JXCrG+mr1@H2a_C
z-ut|)cZHAu^C0Nt0xSXU_&4OH^p$5UeV}BJTPm23?;DbU)Fst7$-SsjEK5ZXyWQ9f
ztJ7X)?)yW6X$>L4_K2h(b^b<zuXpGj3T2jOjccBv$IIxP!oqSjo+q0t(D2@@wnD;k
z)HILHCFjz<=I*oe_nmGmal3}Cl7Xq8K6B5~8|Ikd9WAW%SvEw9R@1e_qVueZD>e)L
z5D>7?arUhJn)B3oSo|yPu*8<Qj*Y<HN^c*NjIX#Bg#Pwdt;`G`mQHRm_`F;Pod58n
z{&N-pD(+7RFK_aD*Y*oXpSz*ULi^Ubo7$s+`_0;ZMfl=-=9X8jKTymfFib-&f6-ym
ztjzfFk~2^i?HjUuJ$|55CnD6cY~4OG;>F~=HQ!Ol8nXuFgZk@O0YfzK(BJ2))kvn$
zAOdIvn!|<$Jf-P2f7|kMpr^G|YA|dQs+15C5ahMD=aiQ-sWkKO@brVVIlFJF%9(0u
zG?-EW<ta&(m_ZZ6)mkvb7+{=X1(&>)7^6Y+b)eKpv3n}Ro00^+T9qV$G{}&j4Qjsj
zH^*6Ov}P2$Y#_~~B`zVMR-&D~VIYoDLIcZ1iO=~x?zU54a9|W7{96*qq!rRq7!9>%
zC{T7qFL5#YreS6gX;`*hF8KGECoxnKGf8MOZn0f8C8x~)G%%!8sdrE(#L&_VLCEzx
zce=xfW5I|M1tp-=8VNU`b@TI<Tnh;H4WH*ODmxxH3GCJUZX`&e$+siwGED#l!_A03
zL$TIk0Ifmfn&x#r;E_gRcJ}J##(U@rDAGb?^aLp0evfeQ@KFO>I7qOS9qq~Vxqp;i
zW;nEsTM3n>Z9jaV&QJuRFi@_6V6LEm+$=6)MDM1r&jQLbAmITn+}4mGWJu)$3e}=L
zo?aRtI-+!feown5NSZ_5w<+}QTzdVp7y^0Cz3a!3=YtyPIzXNNw<;HO+2^+bueTqi
zj8vg~D7m>iXMe0e;XV~2Vj`B$iQT=mg>lzQNLz=63|es-?Brgt>{H@h2`#n*SiJF6
zye&?`2#uJx>!qhK7WJ4qBlh?7bPss1(yA)>cj(YO--7(1iMrPfRk-UTt6f98v?N3^
zTi%o=<!J;Ohu%HU%=rWPu*PAz?!zoMDsI!qS*Ey|8C4d%^T%hf2ZLG;3ruNAQaMW{
zqvQTYHk{Sm$x)M|(W-O=hZ^@-%tSN`HWT0WQ}EU>4Ko$wsNoRp6s&A!eCTeWDbD!_
z5^_1wxdHphHmhA5nq&KeW)h7`mmv^jfvYG;v{Ru{OVcuS3)ad<eworzHZ(-cyZ=)x
z5qY91!E2P9<BB;E$1dMEt(HaZaBTJeA?homs_MG7rKG#%&?z2Lq+2?rLAs^8yQEXP
zLs}Y<Zajo^OSgcu2uQt)dwkFMe(=ZPAokg7t~sxoGhL^6fBziCgO>NW>+#{6H;IFM
zdI`8h&sLQKC=e_Ju{CXTq6aQ&eph}e3I;M`egOfUE`K5L0i{_o?m9{K%_CkwdW%hM
zZDS}OPQx88*;jF&s^UPQoiG;Y@V<xPO<3X`HQWm<{n;OoNIA-3)A4Vz+IRIHyU)S|
z%%kAtjM~!|Na^z$7jkoRbLdv__o4NI$OYEpRP23ND}lNXRpdf3uG&JAzv*3&3c;8s
z^qlR!H+ipg70k8xcx$jZrC1jm%Mj%YW8nMk_#3&{p_e%zjnNIY8p|NV@9Z~FV_dRK
zxP2Dvd@Wng@ImQE-ETowGK*+jmw~v8NJ7hKa?9^!BRV8re|}A(ubu$5t{z;t;5E$R
zaj*b|@yq+KQRcI_59dvnzfwo%4F!i+gLtY9f04WCUVER=uHJ6l_)`eq8-9+(z%MyG
z2Y3djW{m+xciQ86U;E>j&BBE$@!G#|*jPF9?JsEunomRZd&R8(Zl%ZhatSviZKFJD
zIp&;X!;UgPPvKf!$kb1oeN3!FO>LYTmraJ#`=s5+He=smi%2S3RP)j2h5as@&v9EV
z(uKe^Xcr>aE-iikmh3L>oz2&u;TzJtx%PV4zZGcy7rlgn_xW-RyY8r5#2i+~iecU2
z*?QLNv?#-QC1t>%XY?^@*lAhLUnf;4MH5PtmDrw@{loD~Tcs_bc>xpJnP;#Od*GIW
z#v<_bxNP>L0v@M*RgRpT9DG!`XkqQ2jdt=%tOlcE>=wacC6EM?Xywb|3hSG^U~^JH
z5do^L9?HSXJJ{VFIA6&`fL7;yX|1kN{EjJM$c$a_lP_z0$kt1$k~kPdjv+`oH_{4b
z&HsAq%p`Cdt)7(`^RT*TDjp`5h(v8n`-7(GOKKI}${BMp&4wx*V_M%ka|pU`)JG@*
z9U?sGt7lS*!F7>v2M2gKsD(_y43%iQ=ccCAs9GBu)c+tG>qfMwp8`d~^)e-hh~j96
zxhi7lY%6;4zi6|If45~wj@XnTDjI;N&A@0iQ`IG>RVV8vRvDsGlv1)}+EEC8ANVhm
zWYQ5uQWH1H(eBmwR-i_Nu0?iui$!My{wa}Y5*=vVljyvdhaMjMp(7?2O*V-L#xKN$
zxfF}OeM<|c@QyHi*fD&X3xS5W`VayLMDRaB%IqNE&~#c|UxKu)W#x-(UZ>#6NfkiD
z0nE7l@&07bBjCyd$c;eji5Jb(UqP%oqo1M-)J2DLRS0=`4+dI<NcQ7_CUp_UB)?jZ
zne|memc0*!eD?$%MdZ_1&F8E2LGi`I<ng*k@7rJp!0H$Mwp0)<-qyq!Uy^_Aar*8J
z*Eex#+PkaiQ(MesZvt`u1&bamgw}Y`$-oVRchv<=?r@8?so)o=cRSPsqQ04OwZTp#
z?HdHH#t!Vq#IC&w>a3MEkJo~NKS0>y{;ywD@f+`%J_b>calRCflrB5Rp7L?VL+kdy
zKNYDzZA2|WI3!O9*t~oVAteQG3{Z_r6^J%<QGP9t)aLuHj#3~;Qw%aw4y)UaS=xJ7
zwE@^JoF}cJ;VYb1Dp#Hmk+K;+e3>w^^G<_MQYBqR+3}|q9RUjGA`!i>E2M$CU&RYP
zl18aqMz=k#^9Kpiba$;n6GN(V?#8Cu^~vMeyr9pWxx4>H9@~i+FS6}YN8XLEaMIA{
zG{?#Wv(@|3Y{$xOTt^KK@xOmaB0UKsjUtkBzSqbZE$qMtWrSg+An&1vfx+^(K>t+h
zt1LPjd;8szEc^3+9-4SL(`(9)woUVv4>>lg=o$XI?e);{mfv6xT2o!_O)L1LYW$9D
zK>fb}@knD?{;Uu_=THaVEiz))J98b2zK@ts5o0N@v<!Y*Wy1hiiVA2`v!y1xhVO_*
zW7nz~+`6`KyD=^)R+iC{#1{{?hXV83dr>gUV^|A2B{C3fcdtDht--&JtavT^*@ZrC
zJlvOGm(z=)diHjGooZa@Dw`OAh_#(FuPu6uby0%gZdw?D9lQJ6s>iY0HiHF-P_fpx
z3`1HxuP<n5G*#_gPt(7J>0zO~t=sGmNVU7MJbU@dyc?dW+mtd8HpbEa`}$Bs&E(j#
z+U)K$T*S=L(NXCfg;n3qeu=wS6*W-g@j@ieE!RW9xdoyua?kbj1uE)pCOyk?n>DUi
z-)4BtpSrXbS<a=p4&i;`kl93T0%}u!dy0Nn1#k2kEkO;(*{CqRuO`$ytIcs)_Ilyf
z<LY8n+I2XR8UHr*0#>Zw6VG1sQ3{RcJuD(9C=pH25z5LCRN=t~DXWU1fOs#v$gf{w
zS>LuI2LD*Wjt=ULxO&*`Ch2;dMJ}s}r4FSc!aP5uYzeaAsd?6YqLvuCUj6T|T-EZ8
zNIP9)IP2!>)FoO?@<FrP5d<+KoL8&HZpR1vWH)7#oBJQ!*SK9~c4t24?yUW0PaWC%
zC~JlFzva#=@cQ-r96e38;I9V22B5grw6wy_b)JwkPEH9fA~1_R!FyEeY&=K=C~#3<
zz_bN4A^;v!0~#8-7h*)5SXdj53x|e9%m*<Qf{+Q8Z^Sy11x6$vi6I(_VVt;(bSU8=
z*Y{%QH#DZE=436Rff6(u8v9Sw41m@FeP9J#1qB5?@;{Xr4N8y|Puwy{8=MjnM0p3B
z<VhBHPLf!%qbO;cwpnS1dmAq>`{J(hUSTvqkTD~TSmRS(7D{ZEMTx7hZJ4>tyyUdg
zzzsG2^n!vmKMYk09!1uO))69_bPDT}&?Qr$fishoGQxH0mKxy32;Fq4L8#ft3~_`)
zjSG_ci1>S*Zs)a|LTGGIWW^KRHhY}J>m~EKqo4#8=#nN_+rMdyyC?@rX~eM}J_I^e
z^%{+%oCQjl`bE@=1k(7LBN^^wX5W55rEV8CJ1)pnjm#;>;B%lkvJ0%M<Dl^VMGXj!
zCOe>|eDmgw?MiD3&?h{vr~k3%uS5g+JhLzeIywm`(M9r0zyuvh@9}OsR}b&@N8$r$
z<#ZK$xwo3~kr?@(;gd)J^y^QQL=wIHHF%X33=N;WEE|J~#C-qW)6S|~wYjhPU7f9t
zYHdxcZ=Uf2*c;Tm>8YvUiVEJizF;JD?GFAR=)Ch`)i)I;GZ2onw`iF+0L>|^XpQR$
zvd8Gl2k|n1;)0HVdfZHl(q_H>NEX&g_xYJdI6dcwKnK+Rij7RJJ6_~T;cI%zMu$Q-
znWBb<Pg;h4DXQfXyby-GahLN(Y#o~cs4KX9K^%0m^a3F4o@^zQ@hu%=L;@3sv|Nj#
zTuW~qwBT&J_e4A49&WyE1l(hjkc8(gCa}NsThz$c{3o}QB>TlAaWELEnzO=}6)g-N
z#9xeXf@dgr|7V)yV^E=9#x5$MTTS(0ZiNC$(D>fPOh4$$eznH^7rp@GHF+b(^js3M
zrAEoT<}tPHtZ}?=dV2XOYoZA3E)};i$Tc)4>^|j}BqqsK7at)B!w6#Vu1;=HKMg>Y
z-H%Mnn|CnzC@jg}76gLtot)8RXR4EGDPWtq8$*|a6YOi9RMwTb5uwESTFk&LVg#+k
zKs@%j3_V?i6!gbb@_qL}*^mK+(D#N8eP7!?@-(}EwK(kt3rw52FJ6`5XHDYt+!})|
z7Xx8>mLq9wS9s!?+|pLZg1a2?JV;vlZnJGo%40j<0iyb_%_!of)}+=08vEVbgicHj
ze*WR>YpBl6NMOt0`RmAEulIx~Y3`l7x)Oo|ThQVFYCEv7upmiZ2`qLXO0fIq*M{rg
z`o<)}#}68r1+8}<_T6BzqD$v4HASt<nXv5Cc#*ZznJD~f?<S-Y<yYe(E9Umql9sp2
zx=JA>U-uZ=dNCVnu7`gRDTtvs^`MoZ>v-%K(mOyyPovVtH61;v2R;(>-|KN4=!GS^
zD2jNU)A?K-OGeCJF^drYTv}Reb$j&$?*nir+Y{+x=R&uMF(4BrWw5Q4>CiHvh8So#
z*IeLDt{iui<C4226!KYB2QRYxK@YIifaJy8)9a#>ZEWH`CClpsLHANO-+3N$;@_+(
z+0-U7!rbwb?`@;p@?|(V@tocSLs>i6Ds~nyQc4CT^?4TW&z(;E^c9m0o1K{dK8}?J
z1nIMR7&>Fuw>Yx9*nYp240;cR?W#N3!_P47uz!K<$H47P1n)R|v2&?PE=R$?QuKzO
zv54ilO?xnp&SKqQ7aC`B47E99gru=gW0J$1$|@`S_AU~~_c$;@#0vrw%(O^};@&Z*
zv_CUiYPywV(S0vTlSq2s&uq9UO4SY67;xn%&L)Iz<I<(tc^x2V<nZ$u%m$io@N%b`
zn`H6}=SAVHV}AgI4LCkBbb3|J^nB#;FmO~YKYRd8{%&lkGnDzT2FGIh)$J?97<Hu|
zXk-><R+e>l_+yO$#Gdwg=c-0<i>i@kt)V&0`IZ<ENuvyIfBjM~F@#tR4g$#mXo{YO
zW=67$&p#RSfoug($_J0v=@19uAw6c?v6~mPbj7|9L2|NQr`q2pi+@#|^-1WF>5zj$
zw|P;9dJ`d#5%D<!c;VOz&*Z=t{eM#;rBKbf-b$!xf1y9sL7Vh^)@A5cTxvNeUXm9+
z%$LxwUc;wV%p-r2({haIaN+(AOj8&b7@lj?k}89H7f{2_?CNS(px7eOBIM>l|45z%
zY$Jfwe-ge|S}@}e7OO_8GlMspL>vWNRuB1kyKb)*9)D5w_0fXLqA*QIfR;FFK}t$m
zP*;}-pjGxwQyOB^KaKLXi*flb8py2)z*u|f-~ZIMeEj$vl%j<;54Mri%Q{ypMny#i
zP&A%Q$alZbh|8*UB23v#mYQOpaBa;P5F>x5#f>UL7HZguR-cvMfc#*^h_Q*Mt;~wR
zhoVZ}2tiN_m%+s7PanimVNTI*ai$mWIR0E*jKs1mHt}MB%Q*%k`}^KS9iZBja`+{M
z^DLd6lb$RN<Q&)Zq%-V61>Bj+kitXbz9SgpGqqspTcc~Q>vK(Zv+L5w;4TiKS!Llo
z*pEGq*QDa3qXs|ZOQ#8<C7%*=0GbRUPl2Qt=v)8}gb43)E!W<+!R)0Utq+o>=!R&0
zf_ta{Gz#c_H@B~rTKl|M4@S=<=SB1C6K7o{)e)5>Ox0|W!asFfo2R6Dk`IZe)_&no
zAI$c7*RT$&5gd`Aj<zDbTl}=Pb4z(WY%Yw~VCt=*ONHZjk~b|QLEW$3BXg6hzQ3WP
zBkaN-xidc~i=`q@OR07`kN8sK)(JvhNiDwKj~G@L8C{*=bMVLiZ>e&A*7MAB3$*P$
zr{cou>M?ng`T6;-*}e?WM}opf?qCj4VEA<Rr&PpG?OO|ApdPlBAGQ@&b_K^`fef`_
z^(5s&#T@=MUV|lV+NvF|(`G+XqM4nYodE_)*e>ToTQB=_X{<K3tohc|kIy>Km5t3`
zE?>@;kB@^(x|VRbDSEF#!Ap0bSI2@`0oP-7fq7Bi@7Qj%Ny$*sy`*PO<YUu~cdd@o
zv?5QP*Tt?*Rm`{)e-1Xaun@)l=-M5ozP`TizPw?p3E|SX;6=>5%`I5vrLlu+T9r0I
zS!YHjDExOy@{<`}un$?y-pyGb!O_&03)?mS1jE~H!_!~1h8@40q-ew{DvO>0U^4#A
zB=%Wi$rapYeS%3^)W4r36*LzjZC!noUmw4l>z>fQRc`RuMEO+mDmMYgR;lqKWA18$
z_xPrAy0C$@a&8Vc?W*>+4%Xr}K;Ae;U#fVrakhsnYDc4zYx(B}w!nermSwqb`uF)(
zeUe#?DfaMZEG*ugiod{@v02AkPvbN6ZEjApeo^-P2AZktq5*d!nDK+IE?t*<gBR<+
zVB;Q99G#@Z&~4e*b|WSKajoOPdo!tS_k966QuvMuz{b)L(c(=VBLVIfz1+kN4r^;W
zjFsMcJB&U$iaeS@W(`Du!yO!0vr+SPIKN1r_0GrheH9u@*V2z8zOIf!f5MxKo6Q;w
z$+C7}sI8q){Dfa5`_T?LKrkpIXfv?zB`;jg7x<)#%_}5!@<`RXQ0l|7s3^>J6zB+D
z+IWl5cjoQ1XF;Ln@BifflDv?P3^k?VGndQ|pcvtNnaWv~zC(_OA}zjN0mp(|j}%>)
zBQ__M02K+tOLirb|HQ#EO}_w1AjFi#yMFs3^JUvo((iflvi;Q#c8vlnW3SL+9Qe--
zWw6zt3AY^3$^rgr5f}-<ObaSLctlTe+$=0CjjYYN#1C6}FQu$+`Wc1ASkwC0F@RML
z9LkxwIq&<wPggL|SAmx`RM$9(lD-O7R^|5TPm3PbQj?XnHBE}#Spl|R1So=7l!`83
z{sHP2DQDZ5(^C%T=sCRjw%|n&zXQmh$qxjwKw9S5;q%P_sPBh>m>mYnqQ;x5&#Hc3
z^cH0SI}Of);PraQ_J!A#7`-El?Hxz^9Nf(drs_aLhd>7*1m#S}O+M`o(l46Yb;!B(
z>t<-j!s3@3#mmkkEP%oR{-KY?b=Hd-GqZ@K6btB$F6-TJKul0#HTCo7@RQ)yOlMc(
z3*|WrB1Knj=0By@;79DcUBusw&*1#YzDJ$bJfekZ)LOiPV)gLo<8fw<swxyGgi+1R
zgwn}^rp+qz2~+^8lcy@&=_l|+aDleBA*MrB#Y;3tX-r-C6#*K>&1>)oxL)L_X7SZB
zQ5wx8y)(~9FDNv_P-S@$;vFTK8ejbI!aI8FGdiCF%%nM?fX|(1Ev6w~VDY45Rm3sb
z)fjIF(_-QktZ1sJN*h{em8n~|(|q3t0{L32+Yb8m#$M4R(xD1{4nBRWSz0a4s80O$
z8_RcV4FFURnM5WOa_S6n&aJ;oEBo$ww7k015Jz|=T&8<Lw<tJ<-LT?9C~3U)J=S))
zIdNxXyhy>50<bEAt393qupYCiDx%y6fq$_Q10BW`4kBD!KNxf~U<2#EGewaSy|mwn
zpxUPlF$YIsxhH)?@Hior_SydgSAR3S9;KFfKjE5NU+dX7<BU%fYN1ojg<*jrZY=sy
z8({){Kgc=0PG>HQ%)N?8R`g5uYpfQ126ky`0az2mbHm4sDGUzMBz`(rzFlX)!55HJ
zWME)Lk@oRo)X%_W^uNIa&`EhNtAWfeClz78h2IQ2x*x9^Hkn>X*WBY98vMfzKEIh-
zQ3IhDI!Bm(!YQKN;UzC!zv#L49Oj^tzK*xg3f<<Z{-BS`S=GbPdQm!Ix3!F(%wccF
zPL40B_tPaawlBK;ZF{<~#Lx5t!!^9Jjc>XRZ&L_-liD!7s&5wx$(o%ew*>zjAo%V)
zE{~S5XicxG%hP;@<Ns%t8Put@&}Adi@d;<yn}yt8M&!QK_hzXkL2IdH>*JjK&-myb
zvVWnvEec1RD#kiQ;7KYj46*8jB9xeK@^at!V!_-K2s2E|Weg2Pz=h7m!zPN-GckeF
zq8A4&>kx>-QvbwC==rY&CZHS>)zARcURhPuK<SNWb7ZS4|LFd7H4PmJ##lmTs~l2z
zq%%A}2}Th!Gw3~%%z<P=+Fd90O%aXUG?((|JK4FPO-C=c<%*N$AHSRziBEH>&wdU4
zqP^^u8o<q)Vq@ECxH&*)IKXu)ii|PMkn@sC9!*?kkcl@gCM-DittQU|i`cusFN1(T
zWl+Tjt=`5JvJ*7CF40)wl{}L=!$oL_`BHi>Rm#Yz^ZYAqYVC$X{4NyL#C|?3GX``z
zX~yzsIhkH@y!Cf~lls5At1C$N5JL;Bi>QWp#RnSSKo~cV!V~Ds%2hY0Y5AOW2ABjz
z<d*7Q?$cweh_W4+J5MXp$VbI|9vX}ar70g27boJx(~?5Us;B_^PhfrB-f)rmpBCUr
ziVhGV;FEc;s|#oWfD<&>Epfl|TXq`n_;Yy$b_BIEH-W5O&}+0-H~8K-*DpeWnon^W
zWGmzYbq`~}IsQ}|w&NY|wQu!Nj#VaLE>fX7;($p}7Cb}zw)}@8N)umE5inEJ+BsCD
z(JD})r}(k*@<;$ZDoppYV}Mk(xbZ#Or<ETKcrcFB+xy>ZU1{QzyPh7DMMW?bf-nF^
zc{_zY$j(0+t<zs8!^tBX&vQ2WAzE`S2p~>jO&ENfF3qP}pM`J4aw${>QAAHebgSE;
zcGsDv!Qso(ha~|ktCp-9-<?<0oUA0<bGi=2U%#dRZ>8g}Q(U>57mbIE<tn+K103o9
zeC?fubzf0Tq>7zNzvDqsA$M<w4{*a#$Mk>f08w|jL+j}b{aXqa4HN|@f333r;L!k~
z(%zGQ5a=TAR!{CJde*b_mCtsKy`9A9xiw`P<S1qJ=gZ5>#pd0>!|O4^>0=+veo?$a
zpxqQnw@9KPEVWHDpoc1}Ml#G08D5yLi^!GY>Yo(p>;7+!Z%tdkAp6vSEOjU??|~@b
zEO)K79oh6nYg_z9Lpa4Ye%>uP_@;4N>lZE9i9nkLb_8tk0(pjx%0a6VP*DNON;Zik
zhLqPS;{)r*pJ^rRoMP6p2@2(b<=Qe)QNc#d$)<cxa*QyG{(jXul%yDC0iJ#Hv=b|@
z4w(gy9fO*SX5aK&ruQIJnT*@k2vkKaSCwu-R%~bF^>w&>QtnmXoh9_qt>FOyYAm^=
zZJ~hn#B|1?-81{{MQ7}%?HvG61kA;OB4lz2Jp{WqK=`=Rz7=_nuB2pOE;aJA7_UDh
zGj+5+80vZya`5$?zWp@5u7-iVXWY*B^-1L(S#<7)F2^lq?M9+x;sRHae`pij^7ePv
z!fB!xRiPg_Ni|mvn3v^3*b{>PWxgho6B|#|`M!b{jmIiBcR^#@ZW*_A0j(v-+?8$g
z(gQj43v7m+%WARu5y2WfEKPYN3>v&w_rA4U-+8<#H%!r2<%>?HY=NBFWU(G;+3D8e
zxMlCXZ>F5j?anK9l6l63#Gm6iZu9z0Zn>y=pPhA&G!ajy6N$%uN`2sua*q_+*J@#K
zvO1!&g<0hD3WiW2@aoH*!D@tLq(6tTM4AtZZ}Kmh%vFlp?;_;gVoE(6P<oa;k_h;F
z$B?_&vNiv~>b#DOMTDEB>_6pk=_~cm5?BphaM#IeMxt8u<daI<%CJsHv44}DQ1a!7
zd3$Q`7ZWXQ-;DY^6{snh*Z0D%9ovfzVdB&P&~B~%q`=UzgV{%wH41Kbxq^29{lRv8
zh0d($5+>e5Tgq(Bs*{>XPYQFdg@Vvgo%ZHfN4ikcw^T+2R$lRg71~-iTx$xi2CWpy
zSVnre6u~6Y!TccC4Mt%wbnrs{%5ZkmPCT2>oa)s`ldaY%IO3I=Mya{@h|s*!p8$dB
z$J4!okm}bvIVRLP)+Vx$D|}M$Cy_;?&VtuS1Vl3EJ<U;d7bNg-ZKuW9%qL^v<5c@q
z&k6bwNgsJ5i}}9CzKw~2eL(qEScb+OJk+SDtQB`Bk$jMjtuY+J;uxyl`N0-~%YauS
ziOyGMO+k(5uhW}=KoEu<giV(~M@mh%Y+wohd-!rE&~)oraiNkwXSfDCE(3~mZ{3Pi
z96i0Xu)4%GJzggw2Yt*YliK_dm(6z%8s?75>W;tfUyzaA<gDG0Lq}*R;y=k1H}9nC
zY6FS{2%}nC`44PtWa6?*tX90~WKBjh-^9c;E7DBwb3FqjMQSP$NSXt3!_AT<d)h38
zsn^_MbW_{qz66pRfA*(Aamxb}qWa?I>8tVetMJ5Z2(2CYuf9^5;`(L3s}%s`sVNH$
z6iP-9S~P&^d0DL?HU<K#?}h3$2e!6)w<-_RIoCJ?bt_<dfMmYM2?8iFb_W0F)NrPJ
z*Jnx}&U&tV^A{(eMnQpQh{6*m?<uUiKcl_n28vO^t7VtQuV1N0&^k;H7aH9~5D*Z6
zQ7Rr}JgDmHj{@V9*Kx;{LFdh=gEP?A<ICwOy6Q&F?ZNirWN;pszaU$ymnDfQh)RU#
z!=sX6ARMFhy}8WyOZDZH<xcCH<oe7RsiLCL@n=dN5DX1~Xak<zap7BNsZ!2nD<D(y
zDMU`_1W(udmfdcB;N~f6RRh_Dnlw$0L7%(fNCI<e?*h9&DzG=klN}l4dE*pcmKZH@
z7raY&*qG+}xQRi;2_>CIf%$zlLkd_#-kJ<QW_&l>e{0`z)BatcV9%mmB=;cCbZi7c
zUd8JKS>*T174ua?lnFF%96x%XJ=ZLyFxgl5WTnOxCdFZU*T)BiK;ba?QX-g(TzAB#
zRD9{Py`%k|)rm0@N=!C9PpA!e6Sp3lk3PD@UH$T8Lm~MQqe9%m`*l@tRUb#(5k~G&
z3v7r$@V0ia(M~&Y`RtEkwom7N+0W-WTH3cszu`QO;ofnJ?+MCs$9u5vZmkRMo*nmQ
zMD7DEc0I84%+(kMND@ot67thwclQK@r(HGotNoU0J`Az^@eQ%s0A2Yw#BVK_-Y~3W
z|Hu>Sd3my4S1P(6YWSFkgV2mNr-vjWf^pl4Ins%CIW1F@OW7?;{vl|`9G=CUwowR0
zzwflStEV@9+;?uU1I5Px(d<{V-S9sZ9dvoTQ)@JX2RK(>`%IS4NcwAGxr^5bn3`)?
z$Uyyx-P5V`%QlBk{+z>e`-F-(4BMA1wS%K7V)u--C*x)u5b*INy3LLol2vMW!RBvN
zxS?gnT8~VrLLTC6W@L~|!5)#zN8&sKPk}e@f4eCey2O(cUEBN`&Ve`5xrKk&&^0t0
z*&SJ8<4KIb!9pIiKo{h1%!xb2Q<p6m(h{glI#S(8dy^Xeei68qjGX?oM{q7V@v<)t
z6pGR+K8!p<%9j{m7k!X%?2E09v}euIkF<C*qn)Dn-2PkR)w^?k;|;g7rWhC{vc%3L
zWud)Jo0rl1+HHx7hnYgUl3y?8ud^KbBfBKOb&os<oX5DodktQ)7QHdr2-_to9zFVQ
z&{!J12Du`}B>nhKL|<8@63xvPkA#35y%rh`M7vm&!U~{s0=i;g>$9_C0XhV5Y_G3*
z!O$nE0wiqzc`lw*S89nG1ZLA9kjE$jxSvX_c(HinV%JVGP0*-;#=<5>PlX>q@#E)c
zdcnmeZiIl(_N$`v#_UU<<f==$&N)8BG_RB>y?cYBmKsKk4p-Jooj;T?Vv0_hfE$OY
z3a3NfqbW_oI;6lqD=|v+qDUU)gV@Q&Xd>O_BvaGT)=W2D$J-UAS<>zFAlkzS?ZnpR
zP<FH!B}o5J$>}R&1^L3;^w9~cj05dgsctWlA>z&np`v(3e8Ogi7%T*Ax-T`=^|1z@
zrz2q$7@=cQw;+;=uP<<G%EFQ{fLc-$QJdSlDKzKX1Qc~qvs-0^xHJrhz1or~0zL^R
zH2GP@)=w@2g_E3bWPqyx#+)3>0BulYo_P3%GO(eqj~%q=<~_kk+DAtc=cb56EFh-^
zcv+Z;qMvqCJShd2@7qD62Aq#_#drYgTJk?5Cfz^UQ74KT5DXFxyy|+q2Re0t$2MP#
z2@LIBG_q7|3_9i5@VHI1SfQ`9<ykhH0b@9j<A4tetVyDLyoLcH8&J<_;~}Bdu;z{D
zSG#|g<)5p*6zBcTD!|YGUUN*nVaKnsh50?^3jE(pZFiL8)Pyva5JUxn;QqK<QL=Cb
zxJsmJGkzsDYF}-zhQ+ZDlV?H%ui$Ho3{!2Auke$l{wV{`W5taa2rz+Xxz6?GIpk|j
zpi7c@q$I|*y`^Lb;#ahOyY%Gs=a}cDAAZ8micSRElNAC9Cl!*C5y1g!8jG&E&fkcb
z$VgN$P<Dgc00j{q><a_b&^v;#cTC#@QLivDFTcvvqm@XOtM{_&5gHfg7idFOH8dnS
zGw48G_urO!s87_2=;ZF~DV<gdd^&x&k*c)pAQW?ED?a?!3qD!($pjW;8$oG~68sA1
zLxY1%B9-(KGY?3XR$PaaMBOiCCA@r@rKp#>LAgT(q+66j#a=94NjtM?%EKb$O$H`=
z_##_<d3OYWCTw$PHg#*aF>S3_WXAxiECMv@P$S7HWeaH7(SkvpjFm0kqH{Dl1lC_~
z3&mF$wQx3$SSWAjdA_(uo29H6J2b_&@Mf@XK5iZRUr$}jp&#qehqT*sKacsMyCWe(
z@3(P7g4Ix{PJJl7d}{k_pXXk8a1f*M8kAeevafLWRXh-{304#Qfb{^S`bs!}q?F(M
zTv)i)#NHbXL1KGEZLb)+`;rVd&D=F_G%4h`2ebUf{QC~_=N`_L{3Egm!bp>tA4@tu
z`zg(zKk%r{8$2V3I;2_w&%GaTlqdp`g%L)$#T8tJTB=#8o{@<lD5hYsqBfs(^_&`Z
zuOpo1uxc;v$60Nnw6`KS6V0q4%)2Qqy(6i<!GSzB$y@u?5T{llNrnSE{_E%S%g=ot
zR&rL%q9iIGsq{QZ{5!F8@8BVASVXL@X^IpdwXT9*_6@IPGY;-L?U!vEezRNpK3Am+
z>>VrgHEJpJ{$E{Q@_c>ceBnW9S?}t;B#Io~*~>2Q9oxn7{teTnZ*(7$$BbRUVSnGV
zFS>1^NTZg++-&4Fn&0YNW$|yZA;{~7&jUUxPc?JbrLsiMruq56_e@Q|A6qpiN`K7V
zP<Og+PZdb#laj$Yt=kn~y-jCDb559x{WjHlHg}P=M!8qTrPA@6PKVx2cY_DgT@MAk
zBG7t+g?*qDo_Q{-1;WkD%_!(tm@~XHBb3w}-u~pNFU}6SP>@m#ykhh5#f5`ti14{c
zg~lWjMJn+L!+RIk_x=QuXGv3pBiz86(oHEUL+F-H#>6ZX%xtD*jsZe6fvgX{TJDR9
z5Ma!~20au+3#QQ$e7}?}xF59ymE9|Gj<^1(`-RIyuNeD#dg|;TBXrwpINlV}#8fuv
zO1Sib3RK>BI^-7O&gr$i$<UDZEu=beflKx1Ft{bSAhak2DLT{}PDCyw+4B$-6Zz-j
zb%V-(U7THU3eauIRJYKbw8aWY7_uTaYh(*#B<Wy?WVX1X*LWxc;Rz*gZc@70T%m2E
zbAOG|F_9C(y6d1(pODb{`2GY^Wz|jizKEhLAi?sNmXv%>UF$TASS>z*h0_qS@DZy%
z)>L4je(U849d*aCZ~G^pM+US6kZgkvw7jaS7?=_fkg;klMqd<8G6A>)xH<QyO98Oh
z8wa&8dSo>l9gh#VGd=FSwFk8mXghr$F7jX}SEf2dly~dMi|t;oD=M6U_b=Dy!AsNC
zz20@!8W`?k3YmaG+I#KO@MR6zgb&)CznJurIDd5;DK#QU5;CByXY<>?jC5Uc=k@bW
z+TCTc`6jAn5}caKRE{d9!RLIu>KCP$MYOJ9vy+zjEDt*vjvbR%2#w`RN!RIS1`WA+
z#7`*qqgFh6wZ4dCvOidDa`qP?|D>@mSo3<;+o(2+SUIlVEnF;QW&L|1ZZt^UX+5&v
zPEZiV#u<SvTwdsd-_Chu?$GOCQu>45!VkEP@Dm^?x|?&jCj#XhJQCpjmj$bU^X|48
zTTVJ}c>yme9Z#{-v}E7s1nIm#a+qFQ8wUYtky4fp2{%fTZ}n5O$nj&PfYpg&(sYR%
zc%OhP2wV#H_xFHk0!{P;mt6&#OQGmdFiCN`D8<d%?+%Y_h=<?fV|M3dlRn$j{!BEz
zPLu7(%J=VqcD4vTN+<)Nf4h@Kui<2=f^a(F;6vFyi)Ek2_MkHKmyp!P`XrizWybRS
zDW8-)T37!zos9iXUZZRf@L<fzWHAl@eQ}m>?Cw5nX|ZQvAyPAbc*%_$P4sG0MJh*7
zMgKb+lf+wG&NB?m(bz_Fu22(FM45h5jF^~l>%{J$<38JFN0!Klm~E0kQLqjSFFuK9
zZ`XP$ZWKoOPk`e3-l2mAX1P;8V7pZJ!(~=Ok~Q6lcD#B+lYEfFr?n{U?1l?1toH}k
z^!f7-ym2As>5at%aWm$!(KyZ)Qq9AK5xgTqyu!I=o~{^@i7NDD1ykI8?CnlU(xuV{
ziCZc~Qn92oT0JGPrC0Axve@Fvq+(yZv7RYA)7=B2SGPl(N45clw%&wrL}wH{@`)xa
zO#@`~ET#Hj&E}h|+M_k2LyKRX#I2Pl<rA)idVj08Y`lk}H>v{A$NeeYZ9|UpM%#(?
z5VGvUF=3zeq^g5OV!BN|dV1_me>Y`ZHDWDXyEk87He{O+;yDDZA^2a=k}M%C{Z&uZ
z!O-%<ox|$pezuivTG0EQO?+GW^mk4EBlWL5w`>|4qrNsHQ@eGw*oHtVl&5^+>wC*@
zW17n~cvqFayXS8N9UUV|Z(+*(0b|V0>>`7!wv=|A-ka9sn=kqMkc<ktBE7%H$<n$I
z>58Pq#d))K`Lg)kn|@a1iz8T0jGErT5SBGn^gJU)@v#@VcjX&N&&_`ASJ0IaxSi>y
zu41geqUwE)5JNUJU7>8jM|A5fy^y#H``K!4Ia9*C?;K}#1ErL&6t%X^j<n4cJ+%JW
zYw2-m8Lb6P{QPrK$4wb%ja~IEt*xU#@D2@1ux?RVaTtgO1KJ)f+|SGJnE?>AVBbiw
z&R5<2)!)E*h#ur~d(&=LnrLZ2E2UI2N+Qj5ww+t#LYH{KAqROD&2NG*LKGN6=wyMl
zT$w6_^vy~=6d+0fAc*b+R=%fc6`213I@rOj^({??PLNn)_C`fN&#vXoRK-`<ikuaV
z4J8~z$hZmv)xy{h^9_0wo;b{bBqjblSe=4+a7g0VJEomFKiL;}YS;ICW<wgJP-=OD
zUyLJoG)jd@O*>%IL6a#h6`rhw6*k?32DvOww^J<=DjfP`6Z`P<;9@Ft_$1rS3(LVY
zMwuPOkBq^C1^mInNQ*)B#;Wyx5=Cw(NvgPCpTP+C5PQV%qX{r@b>E3icGs~x(|`0)
zBVeSiB8^aeF}sI@R>375jCk)aA=Y29aH@y?C1SHoA{%GN7q!4v4bN>mFy^xyZ*asB
zN<<3Q)?=i&&evo8udD{lDS*k0dvb&Y1Zaet(IlQ)6DuYw{A75}=e&&q%$Fdu5I8BH
zR@gqpC4r8wtYR1hGCk=j0c4hcGKiar0B7HJB<22__s7AkK4Qpl_7(#Qd?s2!{;Ni`
zrlTAfY|I0iL}x^Dyf)senGPXNV33$wLDUVvI8tP=vpfF!_L=e>zb^r6S^p_kyqNq}
z_DnM=^K=a7A4r%>>gXEQbm)Vzj093)4`&RP_?9@HSGWTITV8_Q0b)Z}^mKll^5|Ni
zkq3g#CkH2>ncng&-2ZqjaY)vDqSa+FU7`r=qUT$bk0byqBq1>}G75as2m&|UG+1qY
zF)jiun4lw195N#zB{eoP+xFmeu6s8~D~ncQ<oOk*FfEa~GT5)cuqFA+I&)X@1Pc?>
zOY#djvWcJMGekxD22t}~xds`V0^v=mhaEYu8Hx0eBI$kCgM>pmEj<n2{-W1?ryz)}
zif3)TGyP3j19;vqh2*_*k&_z@M86Gxm@2n%&v|*{%U07hOg^a*yPJK){AKmZqAf*n
z`VS4MhQ`1bJ$ytlQ5>W{{3)VCJu{f+==X?8L`topY+DJZFA-fyRkCkdxw$gv@p5Ie
z?d`306<J`!4rv4#zAX_YH(qzA?yl3KpJumqD^-fVFH$d{+9<QVZ2DhKd$AEvEINmJ
z@-U#IXDDQQ$7Rb!CG_luMus|~DcC)!hSC{X?=P{V&RCUl30eM|w>0jX=D}p^A2E%T
z`evB41LvMas-<hbUxPpCv?}p?55dzobD;M<mw1j&nA$ONpGFzz`ZI^fHf#`e*)(!#
z{nNLccA+=ckdYF_EQ86*fa~6!<>ta12ZvQym%WJVl$XzWNX&t({fD3Z#dR+42GmA~
z{AIVYPfaMYyn`(BmSBmUDnFOT?U#*gY=pq|xTM|sIacI$xhrwBHBtayDvEdtSsU_!
z5t?TpWu*MTq)g|Lz`B>gsp_(7bMeMedT(h@p-s@AHKdrM)kS)_KUa<2?(@z5!WM70
zSK{mF9EFqyLEe@OzZP1{#^p#RY=^DR$JTh5$Dc%I1SZ?N!5rUB6QuO@f`EV2<%-`f
z{$AFU1lWGm5^P$lhk`jSZVSZwg_l^-eUb%lBuM`QiEhCkxZX$dx-#c<Rr^#%(a9|o
z#?m);9)jImC01n>2W~`K0%#*UoIB2>k{SDD<7S*77Rl!_#Z~_*3hwIf-zkvX3!;;(
zK72?g^I}4Y2o{JEPLZnvfuL*^_GV^?OHKO+2fztEc^Cnl<gWaRb=q_gEXsLd$1t{z
z&j2Q^vm=}4I=k@&3UV&RtZBGo6!=0*TfPP+Q66FWe7z>wy=b#2{`$?Ls^Y{CgX8_%
zC@-nsvL^slIO29xl(A71sq+p86l~BfQZGcQtRhYzBppb48C@t)MWf0aXt$`{8X4S{
zO`oal{G26QdiKSc^tdDno8*MTux3!twsHL;=U$`rV8bTA{$5%jmCRsan30)m>9BSb
zcEg8HFJ@mVSTpqxrM>Bv8vMCkqG<M(agLg52p*mWzIuI!YSXAF#qVihio$a~I+_`>
zvQaT#G0G&S&3A5M9d!EV8j=K(prnI-2P6{V#Y(lT2cv@_5`d5pnDJzxB+xdOtDjp3
zrp$u%2ld9YXeYkpl$0Qz@*?-DJ$&4E_x1194l>2*>jSMWG|@$1WjxwhU+tjF%p!d;
zJTXi7HZ`8@ZgKh9(p9$?`T=f<G_||$-ap^Q#!x;_$0+j=JP5ySgF9x7Fj8@n600R{
zbPEs7h<i>rJ%V_Oj~e$$?k7ec(I(r47VJXfh%wVK=Bx>UXCuTL%wzbp6ZI&8K#QIm
z@ZbfaOGbgj5iDE*waDvhqrp|{ywgS)G1z7a7Ls$MkK)-`o7|l5Kn>c{bsaiV^jKkF
zY67|;;KT%$93WHxK1x7<gQC2CSrZ^6@^~{2aySfP%h%puovidtrPy=?A!&eWMj#33
zuBRQ=&%rb!;P6H<i!perheI*d`bcJuOHzT3s;t87k6=AjfwV^JYuPb$7O7`wTYevg
z_gC;<@7%RdM_VK=9p6K)?pL)8{{`{F6%|RkA0O?;&g*UYJ%Fh^gctmP|3d1R2$qqX
zDsFXf5W&d^Gsj)f5fwEvNtBx!tec{QQ805@ZR;FItIDKx*ISXeJ3{=%^eySwa51jO
z%2v!Pe-sSCY4Xf(7Fo}Dj&fiuQ6Rj=B3ngykA?N{Gxh~DVUHHJBH6~WywY8mm<WbX
zUg!oKlg_$LFcNVKhXmt$3+pTaG#}d^vFb^t6OCx8`p+ZrC7%-(T;cMlzD#pTvuZ<`
zGjM)!n-Q`KEBHp{Ov-+ba}-!TQc!tuy?7z8KH!(n5`syV|LDKaf&N3Y`LEv(!|c}o
zRn3C}akH~!?SW|;*Cfk$d@5M}KAp+Wb{U*~U(};YoZpU*vt?%@i3!XNP!-r4`S;gn
zxeuSD*of>x3XM<m`r->C(QJC7BL*1*{-q)&8V&fl{rN=#W&o|`!|HLz#p1o}H(MbU
z!;N1Vn(zM}$|zahERORum`Y<enHDXX+gkjqEg8{C`W}fc?O_>T%QAcP*N=R9Ttr9d
zBBzt`B6YX&L*)#i&4kJFlrO!scle-g@r@=QG8uAT&2-MM-Ljf^q1(86?f2bwEc0G3
zV7|@AO&H<l2n5Qe0;Pq%>#(~WtpGVg)C2-}l`wJZIZMNZkujE~g_f51-n=`^6P?F{
z`AtC|8l4>%w+ar6AI*b+Nx(tLf6Bjnnu9?Y7{K0v*?L@TCp804u;+)nEt{qs4pNP2
z?F-5`0>QYa{ECE<PWr}##M&D9@^pCrxm(V%4FNY^S%>(8w=-SE=D(0R;OGL`NToop
zWA);dR`ur=Fxa|dn|@UUZdU-EfmaP|a#4~lCKIxx8o~ovW?&mVc+2PAATVmY$R9Ie
zmC4N6<g^|olu=u}$rw4o2EBNq69jje!E6LnB$ux>6=<x$irXVXDFPKy_90zUf*dYp
z%yk+2KeRM5RH&TwzxoPP_eO|tIh+&*l}UM#5^(B};QA6(`iy_RC&Kw+BYI&qm=w#z
z%livaCpeiDK}FmV1F@$E7b1bE_bD@wPKGJ?nPjwci3EdHV&3_N)`3bg6+D$u2(3vL
zJHKXOC#&>Zng}X6BZ1~LN;2JND#%vwGq_~ZtO;=>ANUX~RkB5MBB+M#S$HAY)2yU7
zKlAliK|mT19~x}GFIP7Kp-~zdI?ayMC05=6S%8RMURG$)ld^j+nsdh0BO18ZWWRH0
z_{g7XewH!r`_YVZ=qY#Ti9_zg7Wq-|HhiJMGQKTV;AA@vER?2~-^9SO(5_tvY1p!%
z=K2ZH`A;$(o<C_a*4Xf|K<f5D+QWhDs?(wP-;=y;t*!%V67}o5y&9v@7Wtb$^Xqw?
zC>fA1U;e<9i;Yxnhvr3i(c}eI{*ucxBP_y0tA!JkFvML;n2N{uw){?f@`%6v1U(p_
zM?kn~HdrDAVzN#*quEU2V+&^HUvu!u&at<n%mGa|Qxv!;TmC)i?(OXvG8Di@MS!`I
z1M$CstAW1$%8A5LyJ^dj&<O%n2Gr{;NaGvHx@3ss;PU^p0Bl$o7_}cJX@GMGV6p&}
zcGXAL<4IAW>?WS6y)7kO@d<q=V;-<YaO;Cr>$#7odc%cO-2d^fk~<L=r-7@DZH~{-
zS4(pd?~*uGF1Ne$I%6`n_%L*>iJwi$&c(atmk<E#0@D^)W(b(AXR94G2fUT5^9+|<
zmr=pyDJp0*L9JBbM>L~H#y_7o5DYYW1mv7ReC~IMvo&>DV3`-S^Ldc+NV7NA?L--Y
z<Q{KZ35lvh$qhq~O|7%t9$&Ps!GH5Ry+$cmJ?Ulhf8Of$gHt5LPOE1$FSP%+7_*h<
z!N}?F>R}s^4|u=$c&~8;FHdUsl{e~5i$Y2^HpYA7gs+Mr@I%gNi%x`kZVE9|w&zRh
zoetfnTvS0s@RcYX%Dr)vJ-tioyB?z}-<Byn5A0D^pN%4Kbe%ZUCTyF(D`ZHTEqj`b
z|C@xJ7-R~&y{b^h>>@AB*2)iU2w8Fh4j%b(9xOU*f3GmcV!946Xpz<0g%@nR1S~Tu
zYpWxDg}1w!KbPXh_`{zouahGOqq8Xw6Oz|M0!^lB^_ZR09@x^+MT|Nlk#*!*`WYdR
zG3@%6O{vkQdsJxf9@s)MJyRVw(A)mFdX15#n>`2LK$m&e-V?Q!?by7n$fKp9EXmcb
zbr!~j>X40OK2;KvGS_qMAJe@DTb%(nFa36&&#ie2oHV-&*}e6+IP3EMV3FIM;=8T(
z_U>{%VCb!gSB%ZPq4Ltm+WE-l7l9#*tlSFLmy#-ib!E&1FQa@4HKluB)dy3dt%H@S
zzgiC)<>>Hal0u<F5y50vrHA|g!~=p>HfOncH{kLAb#!D7tZ3k0pik^FFsW5ljaKTv
z)|EX5+LD%*Mf_#k5h=VC_srD&S=v`Aa`pmdq#)7J(h{h86>}G!zb^V4(nE!eO#ya4
zB!el~M?9ZScA;q$9F``#>4}MT$*7bOa{xx?jnm2|y&@QM)@b=UF!uWsj$`~+E1d<b
zi26|4s6oP(5w?#qOjQIqsJ50b#2oF^u<%g07teWzrceru;AIn*|I%)vFqrq1=slCh
z6|JkYs~zFoR|%O3l=T}#=1h-^-IkyD>5Y~6Pcy+>>$=y?4V9_?1%gbZ1)U8>^h+F9
z99tZ)IfqGc+PM~rq=-f(;E&PMqe45Fv}VVX>S2!xPv9DqbBk5qSef`;0>6T;Z8Em-
z?GQIp%;#EZv``GfQs_Hwj6YLc!3gK|-pL7pbb}FCD6pSH23!9^D=GkV1(xn)mP}4e
zh_a@cf=2=J;6S?KQ{>O(!7R|M7?XenRyH=T22>-vZhAU=sZuu15Tu{>WP|6p3>1m0
zYl=W_3`|iBpN1TFQnc_QC3qWscrvM|qTs8lj6oeU8cOPSimPv>g7BlM428Qs=k5>{
z-57OvP^R!V3x-kp;|Biy%RW6ViA9kuf!<%6{V%HQpxT`uq4U!(0xWjkP5!~ZltJk8
zV}1dyQ?dHm3up^XuM#9*h~_v=soM}SNA&`jjO4|O!f)R|cysghX7v7Q8?m2BaBCBH
z0v!-DC@R`Fi7P8(V+;Mknfx$=e<@##r!0I!2T~=>ot>u$XE70UQbBwuh#mr2gCLkf
zbbXf6{7eowNI@tE06IJa=4X9|K>wJYo67(g{ZWhmRs?THMFc-{iY+|K*Q^QR3)u<A
zZYV<S!~IYe9b|v8IZ0u=L__mjfA1G*qXSH0JaWy|=yp(;^~m#Pz}}K}1g-wd`a+XO
z4ahwK+N!Dj#=XWY@)LnJiwq#bDdV+>Tg5_F<kp6APnjy?Od$<799Y6PRF|?*Z(gO(
z$CLm3o~5b(&ly_`RE#n1*^^AZg*(i_nh@Qt->X+YxFGH56wQ?1kHg95drHPKBw_xD
zA|1p(NNjyYY?~)oj6}X7P&{J&5qT~W+tVhZeN%7T6Ias=pRo+5<xenI?Xvb7wjL(1
zQO>xr$o6(ypw?XzGwC?s3yN8l<7nQG8PuVZ;5A=l3IX=V<BZ#NM#GOB2=5(&ZPwrv
zak~RIq}O@Y!X7Al*j}vLIb`c7X|A~Mf2zGC{oqUcW#yT$NcM)&#eaqF|5dzaIdwPz
z?mK*cR?|$`g#!4;>;4c*-+Vd_*pwe}C_P4ut8dKM)7l#Ms*zOr+`{)aDs|k-KCaGZ
zzRPiTHjgHcWkQRjKwy4kdPx6Q(^>s|UIW^lg)@3r#<KIW&(<f>hpko1=9}0({8P%c
zeVx3rtxfb)N*+Itrf_vW`QaZXi`>D_aQR{wm<4lva}8sJZt9ngCLb}zW5kxSM0!dc
z4g@!BgIrd7j^^!m*eI8WR%b37JBG`nEm<0D#>)4aANk_8Vl8UW6ru`&6%LS#z>E=T
zR>;&kMlN>R{~08)xBw6xqt9`GdE}4vG=$eN@X4YLW=t@}KH(z1zL31}a4HmVAc3M5
zB)tMhGh-H=tc%qAE3hc&X-BFt$@4Azt?hF@Al(=+nRHnKw=5V<Qq2L`2?AVPBzVMd
z5+t-_(4L1dMfE=ILI9^z8W?T+`ue~OmW52=8Ca$J&HK!Fc0s0V?WAQ9qp!Qs4g#_5
zDR{r3$cy0gz1;tK${JFr&MgZK8hlb2&fI$xS-Xg9z47ENDx)ow1m@3+0Mgr6M_b}_
zv|S<;VF~wDV&t<<dO<iuPZ(8>xSxn*CPjpsG$x7rf`bLW&sAoohALA#u#739@bJc2
z#0Fo5J&_(x<<j((B0&`j(%b%eB#?h(yOiuSNPuRBu7g+*p}}fI%3S}s6+UCDbrixw
ziz`il!<A8Pl6c%B_HpVZr&NaeWWIxKww;e*{*@8iucpIyi0W!mN~i=_N(IV0OiSip
zbley5`Cb`BC(F@;(KV_N=fB)n02#RImuf@L?=HsGphS@aCJKs*Kv7s{mjz}!T#V;W
z3t6v6=ZGB{!nHxNtd^b=Qi)Z)^(=GezuB7eZUk({{o=GK&`{l9t&JbAv{~u5lhCL1
z*4dfqfHf6}At4vU-g*=8;Yp}GtvJ`m9^V>l=Ff_t(2)(`X+<h2K?5Nrzb&qgo$lZ;
zVNTNF5TQ(|c^2V|4C(71rN=vYB8F5i+_i4Q^FFc&6x`6=thWdvBD|Y2zK_b-k3(1q
z&u3M~9ekEVi(?0GXHPv1ve)W?rV`NO)6>&HIccV2@%ZpWMhxTQ%ecAm7RoTEY_{y>
zUeIXx2~z^{2PiR<hUh4-yOEoZTMvPmec9)v<0+|4NN5yL9iRXvkC@v7R$KUQU3@`B
z*V1K<?NUi+=c-F0T<)i_>IeytFoc3AXguf!iT;4VH<+ByDbwu^*>1%N36q`jevTX7
zUQC;bRfQz7Q55#tJ*p&jS$6Gl`O~BLknat)6k~6_-jWpm(feyc$upX$O6*`5n!&Rw
zS|S>F+(qS?q&lC{U+KwpCC;$@*#@Mhv=P5H%C`YX@zS8e88A$<&tW*+Ze`r&LV`3-
zLNqP|;iRvAWQt&pucDVARtlqNb&OTTQr{Nr<2B#(5%&D1{^dT}nl8YM#X0L1ZKuui
zT2<LTIQEe;duhMgXX*EQ`o3CIbgG`8*yAW>&wAV(`$~H}f_uu?R*_64Q4CPO`8Lxe
zY%~>I#yVacQRz<H=1qu<(+@Ah?H{f;iG2K;)>j`3Yug93#sBZxy!BQW`x)7EB((+n
z!nJ<>xX2nhD%FTbr^1==9AQZ-$hP2%O_O{3HWWMHdh~NI=~GV@uV63m3Vv}!OwD|*
zX8(@?HiohaQ!7?uy{XgHM9p>EYTM2GlgBLwpOeis<ba;|kCdAab){q00>edHpFf!1
zd*zg7e`~134|8Hjviyq57TMtc-Se_NJ7cZV>7z!$_Z~R(8h1I32ysWkwR#=U+dtu5
z+uL-RCAjf04#iJEBoGI2LLa36Bn$MteS}(iVuBQwU<-u74|V>Wr|bwI{hwT3_SM78
zTMU-g_*D;$mZtCmF}dCUE)So~@-JdHd9lE*D^OErT`Tzec4uB>y*cmf=*R(X5)cB_
zj-@XN7HSBBoI&$N!UYL9n0tHrkhKu)Gcv6}S8~PDD^kzNxCoS{9ahUFDwK%AShYzh
zNjiu?5Cd!BZv|r%N!xB*sI7ca_8*RRw!U}X1csJ4UCINWe_a0|>HG9H!dWN`m0cmx
zv5Z`?f)Q68mmMdCVBn7W+&lGy33XD%Muhs`?{LzHHE}uvJyb|c$%nkI>(6lWWT7;s
z$*;{^tfGqXO3bNo-!`Thk#fwxp+O-mP&noHLYkCHHlF>dAwlD>E;%6DUB}}3mPQ{M
z$YLyS-@!R~nNAI_^!&>PyvbV}S@o6i?{z}BR{`)*g`UB#K*8K(_gYvuu8;{xIDi8U
z7(PI_!EB$;zx`5D3X0wO5mr-xo_(l$P86o$Zts^0a4L0;9Z;G;PQcSz7EGRHQ5Tm6
zMnS`X`%A8BU1@*c>~}7&>!DLXUgXIB*5`Iv+^_<EviZkW*J@u{XG`>GR(!pyRq*X`
zTWlCQtxQq`l_+=RMa|<s_^vy1>{k1vaWaBvJZ<j82q}i5*BmPA-^tR>)<eR9F@<9L
zaIH1$$Y;Y6yD!ggy&uM87+NQ*T|ZI9GpCL_jYdzB6F7iwG0Pzz7bt=_e5P+2jzzwy
zmiLc<XT|l?-T@GGDliX>G3JT_hV+1c`t4f)vARr35<7vfIuN=*Cws0}0+t4}@%mn~
zXO?(f5@=Pumjsz)PqDqgy9lgzK+ZOPeSQ5%((`{*y=7FDU)b$UDqT|2-CY9G4blzL
zCEeYf0@4lAut8c{I;BNQy1S*M-o^iU&N%1Y!{H}~BJ92Hb<Jze-@Gm+BigitM03S$
zg4T#SCg8pU<2cCfmRPNDtDPcH5@kp+rtbcx=X_PV-$$P0Sd%QEC|_g{i$OeYYBT%x
z*$(gLKiSoZefW->9zWl$ho-i<^7=?J1eeeXh$ruTb{R?U!Zf@n)S%P_{qLJpkwiB6
zH~9#0bQJU)spgy7oP)>3NM@31dDFT3CUK9ZeiFCKxE?mYJ-_33Aa}UWmG368M#THM
z=SBn|rmnInef~;HfCrVNoOx9&xeEPgMfR#<ryQ|FA8yVcdz6ezqvM&x@6`qF(4o<>
z?|ucOt;){Neypx`F<;hs*<>@k?VI`F0NaKX8Oz64ruT7PrdZzB&G(H519y~Il+__s
zZEI^(e~&`KE-Rn22?LPfvu*zGk#lPp$b?jB7xxPXxz@(@UqdQ8am?{P2}?|Mv5H{J
z3?0#h%i|RuR(!;Sl#5cwH^)GaAD6lVw5m&PbK*s+rMq$O39+O3hxeD96zjB^`aV71
zTAW^m#2=Zt3cxq3#Iag@QcbWO+!5(GO|kmAjkcjVZvSVSaa_kGv77zzbG@E%m{<^4
z-3;t%P@q;2d(Pmq7utf-FhI!6mII)<=ABHN<^xCt5RAdL95{<I<h`9jftLJ%r6m*|
z85bjdP;gFOfCM-WIQjVx+{Co#(_Yf$AkYoM70xY3k}81rF}1KT44`Qlxw+Rwc0d3<
zv~zF(_aU#G2#wbnOvCniR8>i|7zi)&1BhW+0pJ9XASw-hWP&>o2B-)sR?Q_8Nzn7i
zft1nJ1>5QHHUif~itfit;<Y!FauktOo>VeqBQR*VadxfjdWnb4HBOF1$fcE7*t*K2
zyai=zywGC(?y1pu&Z74?aK{*Ges9CC<Jnz}(rY_pCnbFgq?YF7OpUmW%VG2I#iq>6
z#^R5;@(*)ltQn(sJem#zjourzdDXi2ZyPV#Hx$rCh6cxKTSVe{9Pf4oJ${ZuPhd53
zWW&HA{zx#0C6$2Fi(^8ZPtl>-uk+eAa)9fAkPtDv*HRRfdruk}zTCmO$Yue_b+|;L
z1&f(JgpLxCBf9YGt}9WmgezRNV(#%gXHG}8Q>`sn#C^*1OKm%8jbZ-S=-Kl!CAFtd
zgO3^z{;-i=NY|G$)&41CS?|C%|K|&L1I%u>kMaJ|mi~!@qygDBRW0GHzU#;@KtX}K
zt)v&&$NQ7!5b)PwpuDoOvn#gE#HOMB;SLwVFjW>0*2=8fUNyq~cq;+#;)T3Jjg-w-
zJ7mY-pSHDcGv7EKetXu4O1yMB{o$}X*LkuI&L0)JbUjbe_XTt{zFsgdUDW1{c7#23
zwk4%8e^Gm`w7SKgk%Vx<71>&S8z{b9k!G7scyuC)Fd)NH+M{wRk!2CV1>gr?rp<%R
zX>nm;2zZx)L*>pl7r>x;Qu{A1Cup%_0R2_IR@)534_<)gpu+}CaU`q|ZrqrunVIdG
z56M(f;HI1gNcrHn0TKLlP?<*qkk>DaaBv{!;do5g;7oHR6m4FXRA2@JA&{jGlCJq}
z-%{7S?`Ok77sAcyZ1<~!l94`1)MbvhqY4~T45q`M{0WJe<g=Y~+pYzMs=af&`NyUd
zlfr&mjcG0K-i4~FVAnE@w0VrheZYo;_OZ(TcBLZ&hY6u5pj&qheXF}6t;c2h!<B2l
zd43?RH{kf+*oeW)O-L~isk?Rh)}w%x9*fI;{5iGS<ZG_os~mn5ezGQFH5Xe=pVS|i
zn)+x_azg9sdH0<2UK^PPQT;fjv-t8j%U1h^Dc(l3e{d(c3>VS6%hnssS8kxiCVxX~
za~@}jGdA=q*4X}!2=zbrn(_4OR;c5IuD7Ukxm>COY_%!uk9|FvtJO}I%xPzHhOIfN
zskaN32lP?>V(o`Ha~BEmOMr#8(Nw1p=qK3waFCEQsDK9#oj39LD|sYsZ=X0{`{B54
zF?m2Lc$U|vO=-qM)9AE{<fjQGYG+RTQBgZb2qmKhx#J-veRL1dB7id!DBac5-~Px7
zgT@YEklc3807mQ?X<!30kS?`b)K1-=RbY70u?M*nfa5WcX=JGGs2?%9Kn7f8Ky|gt
z2FA#C9qni6TuS*=^N9s;sPuyeY!^J99_XnD4EW7v7FL}^Fo_#Hn10{U(%^C;im>Gw
z0sj_Iyx{;gXWU>ARoK?fjvB~8BNs&Z3K;NAgZx$3>1ub+@@{YQFY?Id%c4IUstuP;
z^IE(WO~#}~BtjlhG+CgZe1uC5`&|0ZNe-#@f{!cX!*V;3sI&w$Ar%~!vmlN6FK_56
zo&+--Q7F*{fwn9<q|gvVI0cDpx6FmcSgVioE4dm#C6aQMs}uR)f-)6NAxoeAlO2-?
zv0D>U^7Y4bG38(L7Cp;TkR+5PU5hwTeX7q&DNd+Hn#D+2v6P5?gqaCxZ*Qe+hY-jj
zc~d3^_45LmqqQkyOx)58Hh<1Dfx}k#VI3WCSizD2JcX5*Z~<&|4f$5v<Bos)m_N-^
zU>vY_{_8v0b_VHu@Y!$KO8l^QaIkUubaIMN^=HU%wKe&PLNDMki?@`$$HV`@<;6n<
zd{Qq2r_~NWfq#32?Z<Wwl(n{Rhs7fzS#@AU=!#TXJ_6!RS+$3arQM1^R~&I+vYeMh
zOKy2NMKlE#n$6KXd(WhDs`B1$WI>nk)8^H~<7#J=N4qJr%^c79alm{m1?wQkMPvbC
zZHuuuIPKZw27`H{B6|Y_$sFgDL=)HAXU+nI*j;RGU;m5v`eWhovHbKA|7wWyFRC9_
z-m4M3G5#;*PG|7?cngRx{0$It0A_}b%RAtNqcY1l6#fehqZpMZbcuqe?}`bkMgWBq
zFs-evkyNk+wX9~=X=SnAyB~ecmdA?ef58E!$N?p7E$aFuz6Qb`pg3x|y082)^Q-rd
zsukL7Qwdg43{VJL48+Q;DhNX&)0=l@7b!ii^~CBL`9rp7P9h`5gfxxjoaDQfvD|qo
z3LLK^9*6z0@36&9h;EMxpGE4DcV=g*rIQqeo&!idKaWgw-X<?rxAp<l8&D1MCf*IY
zl|mSRSPF=_KpJI3QP=2iw4-nj#8E)z<>wn06A!CIn>VA%@}?gB`sGv5(Mo;z2HIFJ
z%=L$|51Msx0SNcAZ-loFcCH#C2*xe-fiH`f`&&~t>B=>SCfQh|acRgdLrmalk&9*x
znbup@Hv+?Hq<nBfh_1*NUSx2^)~&~~joVqDh+=Y~k-s9`tfJiH6-2doZ3u2fF6REf
znol+{e}QO+<CQndsqN08(y2LxA^$c9{D<8zSM(8>W3Dn4B_kKd+xUEGsCQ3iu@23F
zadE2#c_9;}b@q06#K)+5Uj6)a88A>8@<9`i{m+*>A80tYluU({(xQe@+ODtFevLHn
z3^s2CZxaz`Zj`*Uqf*8k4wo)n_+A?CyS^nWQ;nccs6!eMIrw~d+S0cErroqx08OMq
zWTN^x_tZ`?pQ$s$#c<o(>@1olvN<Ok2nkUVbZH|;%WY}k`2xCQwQG!^LC>NiFEI2|
z1C}H}?g57t2(5un66^S6Y?QMkP;Nme5k#)*7Ak4&p-%)FL$<YL3M5QT3wPN9Ga@Br
z6WdoS3J99r3?RhMYpvrD%JT~1S|qUP13a7US&}3irbVD|1GmD#Yd2R{vZzGbK*Z*R
zXy|}IfN-f6j5{`p;VIbUhmtE+-QH^vOHwOer~YHVwLxD_6k;7;O%_=Ok!Qj#&|`A`
zdc#J6_MsK`4FWvf9|A9NicOj1UqacnVP#&zwU)0TOr~<?j#Luec7bHkk&WS24A>}+
zY)){mWtrN<QYh88SP`+`B4dugEZ{oA)eyga4PE<dDrOUG4}xJE?Ti;*p-s!E)^0pH
z#wStpA|MO}hmvKGDH{D5lQEfh^Wj5lq*Ug)#Qq7}P(Uuj4wuiM`Uj!lW8ql!ki;Ut
zGJOIot&{6J$QWjqIulGS{}*iAcVn2c7~@sj7dR!r)B`s1ce{B?EqWH9F%s~Ez=#E@
z{DtB06=Rfh`7o{vAxsSHhd}%nSp7fRw}i3!pT__k2KXu6&UX6dRj{@to3%zR%>*gc
zWTdBm1z;k^s^(yjUGo0kOr;?SP%X7~p+8qOn07vS@{as<P0${lc-Pqy=5u;1<#=Ef
z%a2k5CqG?M!T2+k89<SdDzx4?U(Mu{LHc}FeOREfc-+pGPwi<}eveL^v;+(PE6n5M
zNi{Vh>n#;>V?TXfqq)BBdb4Qch<&|;e!-se(ew?t3#!R*@^kfP2|XK^lnBd9*E2%A
z<l}?rMEADDhw^}J;iUq@zIC5wKuQ{sCTZznD*nsLGmNmOJKmjGcUuP#urSdg!1M(Q
z41mG++%p?Y<x2h+Z#HW2Gbp8HGHGlG2tgwPnBcv{BJ37{AeK11b=r^wGEHi9*zLCq
zrtq99*e|RtJ(fZULuLsO8B?v;E0fFK)3{@Z#KN&(@`bK5F$N=19{cle(arNbWp5&_
zmre#jZax=Sv3bz-4&Uv}f{uvC`19?N)rmhP#WFo3oteObtXiH->#r4`1VqhCFBoX8
zFSu~Vagmxt*#pr8mDlf*=;Jiqb;vZahi6hO5nkmRbP<hAa>4M#(f-UqOEssE<#@09
z<Fey<Jns?5WBHbk{~*)gnjMHO0~HZ`hq$h-1gg{m(A#FnqyE(#omM&CA1@VdQq}U^
zmdpFAlrJ9`t0b9c%TzXPjTd=JtU3x7%V=i)cW|o4>9~`BSFD46MVvr3$6~)Oj>klK
zqUH5(kN3G_^NfBjiRqJ<<%QS548n(CyOqBdW5i(|ciT2zn}j?hTK4-*Kq1+Q9+~s5
zW**t;LDqt*dG3M=xGE;CGnJl*!VQpp_&24m=eeS~^0X6UX%$V5P}5wG*zybWU(D;=
zPnWZ>)ru{+=G+d?A+bjfrFIVh;I>mcTg()LE2CrnkK~c$xrR^Da6&Y^&iu^H`I1zx
z-^Q5tO6UG$SN#cb1`p>ZiLO~phYtHCTnHKrZEgF&H00I=XxHG}&xP8#xfgt`si9X#
zOsIk(7*wGfQKCCH;zShfJ>?QXUq@)1E>E+3TNP~txa~wiC}(}V`g7ot1>)O2=;<X^
z6ww<?o|3XJ+Z5SEfH1l=p-?Nw1|7>FK?3dpKPli9ru8tDpr$$4f4&mRdymVLBIjk_
z3PEO08^Mg}*DIYnn!2xF@}iY=jL(=`Ka3no{Jf)v81BntFe#?SB-M2)oh$)vd&311
z?#|fcuuhDTmh=b(q4e|hvl`#1cbbnORfZTC=7=g^YHX1g8yNC3eJYXeVERp<$Fz2j
z!?ktt4&j{JHyL@+!c4+#{b4>%E?c>qmxBLo)d2AU8MV`v@lP=rXyG~<Oq73YDs1&w
z<$tMaE#U&lLg^xN#g!b1%z7o>h{%=6xCd@%6AZFpVk6d4NB$;~m@&idle7CV&3jT!
z4y5(G$D>c6d=CJn*}+ka2x#lU0}_-O<tSdvN?^<&{Wlo_Y8C;e4jOb}c6C7kcv1VV
z;HmLKojGtBf#IS*WcAqt0EAnP|1i$K&ovpj&N-T|4FM*BKi_SD-g@gR^iYdW`QGDK
zt8vI&$?PnfsVN^)7{~L?v%%oy1bV{A+xa_s<v*B}R`xCC)c4m4=hJyBKT8yjFI%<H
zh~+LCg<U1<(f#%Wuq6w=bIa2e)$L*|Jz(BmqWAc9I@!d0pE&rNnf*#3BoMvz`AX5>
zao7H7Z{@v}L&`3aSvlet9E7|-hi$I=s`l;YQUn`L?t)j?-r}$(*2%}c(?rDhMj)5|
zzor2Ihz2e9tE)<&{2CDJ4*7u>;_=~fC=eUXWsGq4%VOvM@dB)fl&fz%-Jk9TBeRgD
z$bs?eXog|h?a@P*>WyTAGMz-Zdgdlol@vmB0bqWDoBFaal`IT&fhvYH{bsVl>4TG)
zko#a=BqCyeIJ`CS!Ugv0Z&;%GW-0>#x;%BeBoCO3EUj_1QBE*#a1$+!7VY&asnYaC
z8*Pn0@<i(mV8YSul?erjIVZ+n`M(_w`eziH>KH;O%!m;Boy=d-^i&zWee?Hm+3YuL
z$pnBM1=SO;w7{krU<A@eKvj_wgHvN}Qk8bm8s|_7^Tya!5t3Iri&4A<!u<wO;=q2g
z*dXt@MCEk*tEa#D(k<t1TB-z@78$A$m2~N;pQ(Oor$&<@_+F298aq|y)DVcIm{C|y
z+BI<{73=>90B(6InX;5_H1CDAZw`Kk``jg9<zJ1j%gtJ=9i=ajW6q%XUG<!b6KC17
zf8lauCQ@><Z!yy$M|Z}(q^CasTK{`uB)F4#rm!nq;@}PculzhmU>IP{TK-G^&uB@U
z)GU{j71G^ZuS?r)h+E9<W{g(d`st~#Ra4XU{NJqzd?p2Nij2)Jwc%wzDajUot`{c5
z<`41wtc4+q1%li-fy+K|Y=Sw0hl}|5*;^q27pz?GMh!`5l9-h}aSCuNBv_x|43kfG
ze1ttCfAk6-5xJKtxeAMOhq6yPAvU$2x>ucg+hMc5_kpFV7^nVptpqLV_+Id30{e3J
zXdlZk%|)Lg2Vjy;pld)^R~MM_s%*O4+v6r6y(5^!tT60ZW8TEi8VL|}2Ctl;8ShV^
zgM>+^J#=ShhmT_6XphD5J;~*rz<=eK><ZH^Eb3{n?QzH8hES#7h7i#U(jQnTuYf&I
z55P@cM2n6NAt)tJ49Ntb5Af&EU?YJa^RPud=$!zup+6siqZDFguV1?v-JUbByB0{v
zix9_lY0L}Lcua;!VaGfc`XD4RtPv|AO;b5=rleqUoGF#Vfg#4;P==U@k|HBJ=)}OG
zvt4Ruoc^kjh8&@QHjzILH8}9_2iB1JFBDWXj3}1ZRKPm!WSbpb!y1XD9#?zJi(NO3
zUyL1WM845FfdMB0{f=V_+7?q5H}YL%(WQ==WQmtCWqui{iaQ<7q0BWp2Rom<z*L|T
zG6f=5s;M$hbeZ{U%(vuZUz=q1+=Q5VMJ{dN2q?-_NE4YRoZ}I5m9HUmXP;!a2)04>
z_tiLG*BkpZCY|6-UU?<Oq%tKVY1<bM<azkJ_HLIP;PeL*%<aItj_wzTlUt4kp5bsF
z1BE7_@ch{K&TYN1za~8?Y-`H`z!XrhwoqdX3v#QG(PEYLNvaHCto)m4sBqXF7b&GL
zw$1Ktl-=CL9@iU;0v#%BG_W4Xde@mo{XvmQQSkAvkC$<t4}t9xFQ;>n@;=%9op3On
z*lKSTW+h}F*NraqYau+{Hlo1UsowK(sm>Dp1dgc7v{4p@;16~b@OXlLZ1XUMdsy<G
z;`gTc#r3_>>G=Cxo#(m93nq?`un+`m%{cb;mB94Fu28j76I9CMU?Hu9zU%HfIX5@>
z(tZr&o<Y!h2i}DMvvQjG8_aBo$DJ?!A}8Qr@UH&!J_FJT_O4c$D>fr_0A$>t)VP`b
zUP~mIJH=|DYXE1j|K}BifotHnB}`GNuO-b;X&Bq{=G*C-_^H>b?Cy7FWYNSS3l1;_
zR3Bfe{1}v{PRd2TH9HRnX)5s5f_4ocf~qrY_rHR5dF7^_QNC-DpzbD?+S8yzhJ`?B
z?m7YZGr@cN?OHKYPDERdP?7>aq{G}|@W6Jt^WUzIqF0>eO&i0l#-e|4hErEX6HnW$
zE>La!?<KO9Suu9#ikRPJ+-v`3XA8kVfyR=^3=)z-)v*3NR=@I)PWW90h2mtMAvk5R
zmJhn+jV+^~?*ms0Dg~DOTQ+V){co(s(k~kGCR2^=0tZX$Ucu@C4{Q5b&+5VT|DHFd
zQwlM8N5`~=JnY9A9luNe4K|<U1@v!trmwy)P5R34M$<gk^Vw~wGvJGzxSH+T^snlT
zac{<mh)#%ZsVC%5*G+Sp$YMB3OxgJ}w@%tf*Gt|Rj{f-`GoG0>+Gyw8g4L>$%c&0)
zEGq2(TrR!A*XTlHEKwQ@o2(EMkvp#-bb3JkB=@^Og@=BBJE?ljd@W5KTU5wgza<=b
zwL32_09pL?^{?r7eV(&WD%AKkU+wm-HCaDj{P{AdXOMu6sYRNFnFpcJ0%X;L&rt<?
ze9a&)2i^_9{W~9{)L@IOg2k9ig*Q?MOlK%)XfJae0BrJn(g&*xMb<@k%JuMwh=ryA
z(tJk*Fmi7yS4gO7yf*|UbKsf;tN)fINa2|=O+l?T(@S_xtD(a6H++aF*o6Tl+>$FA
z0(m)=%FF4I!h(09faY!Rsl;3d+y+>YJ|t3c<%RED=9kIM*lRfLXdtKL$Ziv9k(E6)
zFi&!k3U%t~G(i;*y5N5aX|O?jg?`*GtkU&4J$(wZp`Um@dTRlWxU1F<kgqs$%y_w@
ze$C5^AP19C7`34_%c+f&h_<AcI82EC*+iNu2}>`~j1p;JWKS-*ZzD4~pq5*p3N~-X
z;}Y#FM^eYR$*_CB7anRSN~cFdCo4hFBS9Dv868N*OQ$O+Bu(``r$~F$DD?wEaOfuf
z$-R(p2}&Ye!7#LMFzFPuY^w635VnS!zP;ELmr(uYlXW3oG%W;tAe3%%($mRsV*p85
zj3?{G#I^K%c%R`)v(RAn#zwRHe7ry2@5uC}ybpXHUm6rU07d+A%QU{OKuCfkY1_`h
z`%RuZT)3+?fcWP<KkWX9d<PXEFuv6t;v!L(<<JrF?PEmq#ZT}f`=G2<dOjuG8LHl#
zG!RFv=aZ2qvGKc5=np|Ro5&lKbfVBy7NLK0HJ;D{KZs|)1;%$GS3P6+9x546d*kN;
zvx~?C>_p<ZoAZ8gehaCM#ZR1}4dzn6N(~jWqczcM6ts(=rq#Yi%HH%~dBDp_qyn|e
zJ5gJJB4y<5oe89J;47jpQmr&>h?KOM`;{iAa(y?+6+18>g+{~+f_M4gj;$~a!r5TC
z1uUCDca~By_|TM49|M6@>a|=@GcW$}_GK;G)PKLb$@Q8PO80Br!xq^`Z?`Jl&fB8P
zryExPv65IqPtOTVEdQX{#<1+BX1f)y8zFDJiYnjNAN`*MOK7L%6>#ouSA7pd4^z!E
zW-7Gu3<~4(y_uHIbZ=UDb~7w;Kjdtt`U)_oGJOm(7&>2L1rz|VZ<Y<K%*D$i*GHRj
zailFU4Yqea^bzJd9X#KN`}VoSP3qM@3@QxUERT*u!P2=hAnjzkQP?oR{Kh8|Lisil
zmtl<M|4(SI$!K$K4Z>aHNIdjWWEECAW|Dj2_~{HyOJrJ8)MeX=V+xVrD|+u_17U_h
zgw@{iJ0;5Z#>S&KL*7kY15Klyti4JwAB=JrU$k%QY#h*b{SxIgJhu2h!yND1WizUb
z{yJiRH^)c^p=E_yfdK5Yb+goJfm%W+4J@zApJ+k>+qx&smhi$$3nLC_v6F%cCITM5
z%+-D^{@1&4&Sb(e!|eF@>1z%h7kMb?N*7_BzVDcsOBKWUaT<>&mzO61G~?Om{yM}m
zGq|LMBAE{9q5ITu?0)s8M6GoWTT+U@5qDH)Yvn%A)alb3H)G+>=h+X?o)#niCjpN4
zIy#*+46g75yYfEzk9?ICheZacdEW4<fsGo3Rv*8gS2paXY*ogXfdF2lSii<r+!GJ;
z$eW~g0@Jk_CV*#`n&#R#Z^kQ5Q7i@g0g!9i<OR4qV|+0FPtDCqgBwkXT>c?{kO&m;
z5Eann)&LN4KWNebG(t*jBw%q7Y)WMs;e^zMXerq?npQH{YUrz{nyb)Z<CBokfT`_2
zBY*^jC`iQmKaY}vUN^8z>5b`v@ROL-rsGlT;9RzEk|RVWM$*Y56dbyyA3Il&OsK`>
z=52dAm(7~Ec&MjRQK`7*7@<W5Mh1ez0*1V<E}Y^py#?#;gybla^uRQ-fH7INF0`z9
zDxr~@OqZg-lAO1(jYdUGtOv8Jj``;uxydyfxedPG>YL<|XJH=)5?U>d-N()eeVfw*
z$jYz%6W41$YGm0+3@T$c(`gk<C-vV`(%634(!|m#yQSj}1IE5mIdA*n;a~N=HCN$*
z4ib?;Sl~B1E}E7-{opod;C3oFLE6;X`W~4Ii)_!Y)8Oi$sx8;&;QHH^@A&Seek4|2
z^x0qycfZ}L3_DpHLO%#$OJc@l?Uj{LuV#vDNZISoW9Lw)?vvBn6=}x4wHM=7S6{cM
zEe^+w!<g$TZL5+POf8F!+vl8egVj(`4&AtJ*_f3P|NOg*5i!~H-@manGD7mCVKz`k
zSU{()sR^9?k;#bw2m}WAFfnL)>NfY|4Ud~=(1q$7lpjx3_N{^%A4Qc8`?d$KLj;Un
zK!BFd;0W8z^@(fhJd}%$5+dn1`~;%y7iYvxPE@9s`xCyYdMpvfpYud9BL`*O+%O?7
zoan4=wgdp(gW9y(5u0Xku-*N7H9`kJSPYF8x3KX3)@EM`i0kv;@UF+QmdyQ0d9mhi
z2^*g_(eIkP&}+A;W5tP$ER_vm=X?p=kHYh*h)x{g>k8oY{V=kdBG(G`&jN%INuSW#
zZ|Pi#Locg>GfUJ?cEuPZ*%OQw`thb~d<ZjH0}Bb@e1*V`%!!C21LbmO%ygqK3Ore{
zkizzOy|L?+zn13*l}N_yDS=i@;FK5#n-@@54`@RP|C_JU`Qgo-#bajzo)_D@MT6z!
zKb>l;b7iO$?bh}f?bC>lHZ9plKUygCG$PHr?-VWau;`6UcE`HfsGTVwa#)Sq;Gwhq
z!ZR%~BC79>y5Y-N;q>+Y_p3wesp=O@-vdD_0hJaVMM!>9ZfSXKLSx7BXtoeM5#iu<
zL9~Ftin2v-@WR=x;`{|=`Z;$Q@#UWNk2imiL?}m@5G`?RjDPCaG!gPF!77mAKj(e<
z>r%7NClc0@*4l#~*oaZ0*;6Izd9Ij#Uq0nFUvtA_DX0=134O^Lax5It*k(L}rTt65
zKk4ayy|WX2AEEtHnvL@BLE8YZ?fuVdlQrd7KjdNAA(v9CVZpQLH2H>c9<B4w!a(i{
z`P2TdVqgC;J)uqumgk4REGNq|10|G?gU2q1`G)a6)pSVyc~K=f8RvC&i)d%h&(ngy
z^`!;{IyBO`ZFWpE`Bla4r-XEM|NQ(us5tQXEh-n6dTqcnV?6aEJ(19#iS@!qJ@+XS
z3gfx<FT=PpxgvT3`eK>!$eWk9hb2w9jz1U_m*)B@vK{RMs?t(pC+e$dQ4^v@_3N_t
zXFkOvu8qc+nKTILSjo1~{ER%~a;LJDNJk12GY|S%#{{%lP(=qQet-c{SBEvt1;iG5
zwGw5}A@JX`F%586*zhGVk>XrmlBK)jIWOg)KG(J=y>!xC*EcVOL2x+-OBn#W|L^7{
zMTXJUcwDz>6PnO{0!X4@w^eU3)&n5iV;1Zvg-ovW(ZR`By=<U4Zm-oKE*@?!mS@A4
zu;uy}Yn2V+I)fIqicDN>VQv-obnHap$*v04eczBMwo2ox<Slu?wobyz4`d+b=<P)z
z5{fJ^!vymWB-+yr#c)2J4ckQh*B3h@vINS6y4p$@EL3tF+mDG#S0-p35;?O9=43IM
z_Pu|yx}u2533H+%-x>Y$<&}(C-@+gl$3Wau58o_G5@n)AFC!yrN{I`Oja84?TZa*o
zH3mC=l=aARgagdIDs~Cj{5{J!MY^@#NiGp5cu`3=Vp}mXjO}}sWT6Ol;keGGk=gd?
zdnN-1sf!cXASh6JA~kpAtY4R)MTczy9tbaKJaNvw`RU;=@?^zK@5g4V^gjSCJzn^U
z7Ye{JiE5cf9pR6Cbkpl>QPUD$zUk`f>XShC0aoZ1;vmm#iH{%oF3)zdw4iI8%_-g7
z$mP}GIbVHSRNlBUb1poOq<pv#y6w(!yr~@^K&<qoO?h^HyV&tn<xK>jfZaK)Pkk~T
zG_Zt3+~);;m}|!T%t{=bp=NZS%YZ9+c{b{xm%x+~nD;EGca}SSuzS=6<rfB%<6~&D
zn>q$}NZZD}Am>yEK(<WUHGu%l39jW};4f3X*a$&)UYXJ2^g1{3j*&cRJMEa#ji!EQ
z4X&6Ixq={7J(|H;UR5=~r2q`?;Ak;d8QZ*)<a`*-U_Zuby6-|IWqRb{udui*8Dr9K
z{M&5D2+8~ki9-KzVZ4fsAh}&FeEYM1#$wdX23IkMy}Dg~=k@SUDCpH&gi%WcL5H^a
zdM{7U?9XwAJ;JMQ){BsoDl)Ep4MhvR2Sm$9tOmOfLw^_xrVu39e@)#hKLSZp{a_BH
zcP7+({=|#l2)7<To*Sgs_r)p^c_X?D!dekpSBG+c7|5oBW2_E?6?T^VmPZo&7Fm>0
z<ObLBI3y*lr)0G`>*MEW%kOhtnK@^-xSoG$=9^w+W-j^Nd-PN>8<n%|N-2?Dp8cPf
z=l3n<EL!wjqgM~~b%pP{oARvJ%Nt3mK5~WG>?oO-t8p88S|YQ$qkU;oJd#tx4$I7>
zL&j4iClKU(u2C|-NcVp;{kuFe-1GT<?)>DYvY%_Ywz;YbIy$W2edHa-K=fNq0?Twy
zYij)`Y~PacKirqMvs$YGf9|a_gY#_=$cJ{2c*k!K)AwSWw&C7!v^KGg7XDZq{AOtE
z7rZtPznpcuZP&Tx^r-uE<x+w8Sx3V*ApO+;zSNc&zy2()O~?K5d504(m)C3mYG-ti
z`R7!|u6eC|pqjH;Z1WEZoi7-`L=T$1%JY=tA?4ti_-kfH5*!Pu=Io{eC@?{RAFD^$
z<i2Y#f#(nC?FC`h)UXDyEdeP{00{*&&tzf)%D5W8`)hFe`Iji>$dO@yEIX9`+z2eN
z#wJRbe2GFuO%EqrO$U-YFd&`YymXzAVTd4z3gwY%-J2KTjz*>fl{DNq`z%v(_=>9l
zpM>xe_rDtrU19sf>3adG0JnHD>4PyStk++PRAo8tV1#C<cIRXcf|JQ>kwZcYjeAf*
z0VXDwz9b!1kr>-nVL?wpeLy;FM0{dX%$`t=D^D1AO$5Crqeeaoq-LrpR(FqbAtS%-
zTb8danj81DGIN%^_ZMY-cu4u1zZzH<N+y=tqL8RWPWi-%kJ=RRn04&(udy+-Wfr7s
zJ+ZBnU^Pg8u+XdEz>80_p}dhSE1b%q4BonBA-S|rM@cf}SC(Nvd0dD##@}u=0n23o
zA}uSUTipbtP%yy&&in*G1dM`}0JxB_aChte(v2RFjjc57%#Dsl0+4=f5EA(xdJF*h
zo&I~kTeul5{3{zmub|NPelg=J2p#NG<7_^Vk$l=4!ekhT1Z5XWW_K6t=pq3KMZFEO
zNIqyiqr-(_3x<ZMm!G=+u-E>}L>lJ0-Q-WLnyw>#x~4o^VKiD<T%VrnPBrX(7l0tt
z>sDy)%;v+eg*+*Mc`>#->HqvFKjTM7e80YI$h)214`72O)$w~pvZJ{ii2vd!Gz~?-
ztry?}6{dm7a5Psi3;5DN(9&gRG{`K`lUIJaW%pvE(QY*NeHwVvyreGy^=03cV9*KZ
zmSAV%@^rTg?Bp5zZVceCwODEQDuc+?YX3c8&naH=9%06}&Me`p-xqYnudjP+UX!D+
zfT&k%K|cH&V|_9z2J%!~b6Kq4(0DWDZ^T-^J7GE;++w@%J#A?c0|%4T-Q>$9%MYRn
z#{G3C0&G2}J@uO4>h`hnuL@xmPnfxnh_(xGZT_6CEWGHND^MSHVpmH|nA<yHNS^tQ
zmf<Gs3G_el^wilH&-`Ooti?KKTqjsdX#46zYx_s%&nJe;M*=3x2n9!9Vf>mS8)YXv
zfqSwBK@>a6L-lt&&cm;b7J9T+NlE_it2%cO&GlkFNHJG%X+(0GhOL-mRH4G6zQQAt
z6aWUM);E*9!AGZS*S;gXslyFX3<31%;k{L~m~@K;w~aJn&C<^Wjw_jAqv;Dn4|hrX
zNBBky7a8V3C||3pu=llR;SE;LHynosANZ1*{~hl-pNGYyvtRwC!L5Mwd_UFh@5f{5
z;o^F_KJ46@Twifoblr{cv@m*?;vG2__4tq<u-klP*yQ{>u3gr4W3}T+Kld$^HM-Mp
zgR#E}Q5o0dPxtqouJCn^oH08}5};bad`^Yuam%o|?VPmP{U4zN9PcJYMi)+BsN<nN
zxu}MEk?YTK#$vNULZ=TP!VwQ58_p-uF6EJ$WW{FS0t>KHz);WmX}pL-H_HBiPK<t#
zG_yp<{p;2Bb*fDgAgR(QWO6Bzy%7VpdC<G~=@gv&S?DD3gY>$R&UZxO;)o#Z(V`cw
z&LFFy!L9lNK30`WWi?c6;E1M3uH$4$8v#Qan?gNP91apZ7!&+EJBf}x0eVCQ7gO3l
zm485=vCmQvpHHftk6JEKcUW|1R-)dCs;noLFImQsOpDI3rYYW*N?w0j;;1G{Sri>f
z{fEqQ-z$^v?5@3ZGcqp46dDO|Fv<>6GEw%q?=L8N4Er$OTK7ImTv~BSntXt-pc=T7
z%#Gw0VaTCL<c=(3(C`<QQ_1HwN@^r3!Vy=)MoJZ3IVrV7Oe}}9kW39MFWg6Jgb{<t
zysq>`Pfd}S5f2oRi88^P3JoqOv#;EcA#ODFk&R>hN+)---E1!QS>J0mjW$VDRYgUV
zmq-dh6vVV%z$eOdY*xRh02~uj4B!<(=M4@pVga$%|FUA$ajdO+J$~_N=!&#Gw>wA_
zblQX&<NpRm(O<uM_;$0!S@LLkd=N)`(e$m;`4hz7xti2&at6dIM|RI@ihciHHXw_n
z8_v+}M5m49rtyU_u#-y+_ixB?SG3LguwipS+UjdSBly<T%Z=A7{K(uu@=708!6`t|
zD)jkM<@V=|9-fYiY|>&aUg~&=-*3m&7(O8$Hj%U78xgaHp(xYE!tYmc`RT)7PEKfe
z><D;yeDeH1`@IF5!WS<5e6gjDh6c#jZLIj57T0#<Qoo_W$mEIr-31k0@RFZ<DI^8g
z7on<-f7alwFdl*~1h!<he;T4XIy&kouK*#58Vr%fJ#e5r`-NTA{BSWcDB14aelA1b
za*_0obEzXudD85wJwv$CC(o~`Z{JvO)-uSDbE5>TQ|L^G^80pRaAqAk6-nU0h-;7P
zw`0)95b?R*+6MG_du9`t)m+ooT+a}hk}Tb>jW}@ZGP3Pb9X3d5U*c#dIUgo%BOA?%
zwqI-wBC`ZBrUr5&Jn9?4T5$Kpf8%yXxzg&nIt|Qsm-(bm>Fv7}U{=3gf2dSf2+b+9
z*756@jcbbqc~-n%mL_9h6*f;$(4{3JVo;}N9<lO&4;cqgi!?cmO#CPzpEHwWeLMS)
zx-v&^WaY`Y<8lxe4TWlC@%kE#oh!Mdygc0Dsh9ZSB5rVK8%bRvc<Is5P{=PqzIiqp
z%oXoeZg~=ikA{h@qs5da$X?%&ny<851dFcb`A?O0WoMxJ9x|Txa=Fp`*5oi(MHO1T
zJ)bp?DXlvRzI!;TqE+(FjyDLv%)8HOn&z!VlEm~qoPCBKh%t*Y$!vSi^OcdISM%C@
z(QU80&C3D`hLd%7AYT7H!x!j&JM=S&kC-MEccbr~!p&7YyiSu=*H54Q7|~as1+;uE
z+Aoh{*SMWPq#cnGRv%LEwVVLNtbv|0u?x^+M1%PtHvuTXfGAd0RSs5FD&Uh*t6c<v
zL<8mqXb>c{Df=w76^o!6ZVc*0!XS|UjPbxB`Ck?+#U?rE@F*H&n{ov*fYtZEmTDs%
z+<p1z+!LZy4Jv%3gkcNM^M1Uy1agUrrP?6b4jz8`gc>|<3bf|MFaCT$kOMp?a7~Vq
zDA2pNE&65(UsrpaW(uEvp2K^R!>b=_O9Z1aJ#400#HmA(A9$FUNXE?jXGMHMId|yy
zzL{?df_V2A`=bI%<l7k}^}kbcJYwA`E4EbpLXy-5Zz+TY0*lShD$r^jG4qbe?4rIj
zF1o%}U&ml4f)OIc6cmxs5GliEmTmpg%VKtJqas#ootVMkeZM$d>LT-TkgaV>9a&Pw
zw*M791~0bDBg&PBDj=CoqE`B1e*4n<IdoDTl^pxVUqGseNb6IjEeYZ3u!GcKoL$?s
zDfx>j?*(iJQc|E*96ZJK=N9Tep@RZn`79pP#YW(j0mg9xg(=VL^-oqgzJb)!<K!gq
zF1tDZh4ny6YA}IP`XxSfyD;HD0MD3O=jzS*8b>&6@FXH;MUjg@uuAYdGA`Ti6`^qj
z9yATmp%#AE=VAIV$c%B;{nvs`F2GuiU~Ryo4XqB7s^*H?@;wr1$15J&ARBw&WA26Y
zpGd?__!ItmTgHGoi%MnmOX9}C>Y&xl_YCx%Yifm*LDSbF!`H#W3Y?wCyX@a|NEhWB
z>v7hHgv{TyGk$^oOI++V)_>&lSH;`M=cU9AG-Lz+O_y8hxRvEv4`cOK`v;cwpx2}l
zdw^L0JPM$K%IM}~8N^gx*0F$-qF+_a%gZYRs!E)QUXoxhBpfti{!qI`+tm)@kqc{%
z96JPZ%g^xSjrf7Gw4aqecUIZ)6GlmxN)&9uUtMZ4#7)NiA1}aXUCBf1LxWQqFden)
zieU(kslWM#Wo;-W|DK#dovSyGe}-&+uR}mU;OPO>Qh?Y<(pbj>)HtCi2re4l{nRp;
zb1`sVFxT_H+pa$369j)yKwz9JWB~^JA|B?s{)=|mkGymF6sG36I4YI;xE54b0cbVn
z-#^;#=5_k8ii=y->3Ha!MDA^tOgsg?KDpm<8uUNd&)KT;N@dDW6cD0$`{DCD9XJ>v
zn4lzNLt^5`hvn7f0l&^=5e<Q!n2-Lv<4*1++MVvudAF0)*=@G3Y7>oiX=aE!?c6lr
z1D-pp&Bg4fpP%Zb@#Tv~4qf#(A&i=53-N}cYNp&UyubKFD7@Om(T#@q&o9o2%nut2
z*FN^$3qN+$A?&CYnNP=11@5?=5;ax*d$_56y7}1BX3ppnQ=#+sRo3u$dy;&ht~1-P
z>UBW@-G81c?8PPrS|T(xa4L9synCsc0mCdnn-=MHewcFuod^Klm`QV|N5W{CUt#YM
z8%fw`x5<n4U!lxMPdWqw;@tVlT~66fCuj=O;8HTU>*C-jsUqjrFBZwq=0~H+9&f%-
z>?uu}Ad?PiuvyX=V+`wD{o;<3DM{Pf+Eg@f0AiC0i++yVN5^SL;%6G~({Ucd3}}IL
z4U_9-lcjpKcYKoHScOIoPb}R!f44ybqe<OugrZI?oj@lh>PCh8<%90LWR!ifE+w=i
zS#h|dZBf$TINDl$op!W%7p6pa%1luo{f|(LvYu{i9#L_VGMI8pIx!;|k@NgJ$dHXB
zl7CF)i4g-IQn%<Ao!@I1aq>Z~Uw(<oAP2=sVL0QqJds3ahEJSfqJ^<3On(#AJa|5T
z6#CUP7?bYcj?znpjr0;I0dx*|Aa^W?-u+fn0zq%!Z%Qu#-67C`cmkxP)zxPo2(3Ej
zp-ue|>P~reH7!Bx%PJ3KZvmwpbV^*r_?&dN98utA*MD9;{xh*LBTj3Vq)+F#97NQl
zc!5cH`Mc#pO}U}xVcJ}&-h0?<G#e!W!}fO8ZZvzHvi9!}d9Xdo_XrmyURF_AtF8|2
z5N$T%lP?Q*c&2?A!CQW;=bK+f%6sbCYe{^YuI@})5iaWc&lRfDH}~0l_x-+=)%pPQ
zN4gznl%rLp<_;1S2x&So5mlZ%02#pyhBxZ7S)sil4gp^&NUTbLK^R@=mJBR$VIaw)
zRaG?jG3R>&v0P(09{~Q5AhynanIBXUi2xRz%kFqFc<G4)7e?Zk#S8CgzSi{h{mtnv
zASMff*%_oK;)wWqz(F`Znd5xTCNLRmDiImJLUh8rKfi^HE2g31IO!Im#}hk2$|S5L
zMe~U+T8#fAnGplRw$k&@<HP-}#?|j<oz<Ot^7;cda<!H!bhzhP1PC~Tjda+|T+1T-
zRd0yshs`+QU__~m@Q`J&)fFV~IK63nJpX&OHv+W%gPY)LgVl6881L!r-Y1&CCfg)+
zrt-#pEL{FiHr=%CDlHH7<M-%1g<%?8kHh<Tw}AZG(V!%5t`^%aB1aIh<Mu3uSGCSx
zSJ30fF7r~O7pd@qGZavjz($3zDK9y3JzphZu+FZsV9bs5mM8wv82QdOfzO5Jy!&wO
zGM|v8o6S}I@o#eGr`(0^UUb2UH6u{p#wFbOh$(cjKB9PtPb-d~&0C!@5pZ$bei)5|
z|F1lj9uZmrRl|4lLqI?GskST!H-D7Ow0b51q`aCir`D2y;GZnk_W@$O?#y`<T&c`k
zx4E4uoPIR4>tG=pzLEGV#rX)!g%oHvikMWQk3O~()1}huLn?7-T?M);%HX`PWx_Hm
zmr8>`z>NSn&j7C!fjm;GP(G6@3KX?~#CM~e$47<!NmL+>rkdB9-lL^fYb%1QH?yB*
zqa7GzWlN`lFNT4E0cfPaH_Fb%6}oyVJ?D0D=gUW+qQ-68Us6y|&@QhrbMduAd5Ncq
z#Fm-}X&US@to{u|*$<o5wFdz8wz@N3h6*T?U2G^#uQIHR(rF+N3T!9&h~&OZ=VPXU
zlx~6u-r>{Ir8}}P2_&tuX;)H+MWS4>3O20%Oq7Pkw02VAd};nyP~d|V1VL2Spde1E
zDinmXhU50;ZY^1-E5?@@gkhBB|4y19(AAQXqTNW9H@YzbS|bR9)T*FM=&Eewzcc$P
zB8|}pTD02pBeudqeh)3chkrBnZeT&ndOys8QXtYtTbAtq;Etxiz>B}<j*96*Q#&Lf
z&lw7_PA>_vO8yt{*{1s8e!)=lNCTi}<{PZu#0y*?yqv)#@o$jAURKaCc&ru|Tf@L6
z=Xkc(ec?xh6c#B#0T&|r5~X%On(rp{f8_1{N9IB^zOL2z59e^MnifQlKxI~vl8`L$
z=S+<)@_9Tl;Z_O6oAAIcP~G+pe&6rUq(54$i8zh;biSQAwQZx-)f51SN;-Ahsl}@t
zZSuo>^>sNP$LO2Ykk#KyT^sTG$Tq9UBJ4Q>D|grqAf8e8;V3s^DKFudyt}?fA;Ej^
zrJJyg;rkDCNLM4NwFj4Zw8|R&lB@r4g>~w~hw%)C7N0hS9hs14*z~|Oeu}^~_DYf(
z91Lc|SS3n%ihuwDu8Fh>MyI)g4aL(;<z*t%T;Oua>v=}j=5fj*&A9?f!kBa$B0)He
zjf<-f2(xS`u!ByX2j!*dY{oEn2A<>~DnZw#iW~-mjK{(h^AN!Du!Fa{dx7qE9&0~9
zf4M|TQqp+^|2K1AVD%S>w4Sm1UvG&D{hiG?A{<gHwP-%PGpI`6$B62m?^s8)w#c6N
z^0(GxfYB^ECv-fOR(3b8`rbp6g8F&2^Eudj%;{A$K}Fg)rKR=KW7c5vHKgf$Dg2Y?
z;&$4_;roCgqm`}b7*;XO-!&eWuVC_d<Me^_RP69q<A2T=&vx0D#zug|@-56dufq>t
zSH$v0tz|l_be#rw8BfoL?e%GO3ZWeRJRe#d(+wz_UjOh5Ns{HcAVal%#chKfx<Vkh
z3i^rbxs(!;--dWpuGM9rF8!#FbzSeX`j8+K=#tu+q-(plE-n0zxXJg?-P=c+_ReA1
zJ20#9DKJau5*D;+*t9<qS3LjzFwjW+XVJ&dOI&15oq+*{x6zC@qq{CX{&Oym7tHhR
zcl4U;_LczWEZlJ$|9(5jCg!X@TV9`XW}joY$@%RQcGuAv+zRKpUdb!#$xtyF#>cjj
z^b+k*qso7`Lwy}Tj)<XprZaqk>#U|cv1K?1j;_Pc|3sT9SCOWUYy-FiP&#Ooa^>0f
zG||vsRwp3W%p12Et+A~wED99%!Es{jZxjFpS_pw4^|I{&pjc^X#Q)SAlvtT8x-Xo=
z7iMfxj&o7bL0)JUPDES^;$Sa9RainlpE@=<$@ZPw4bZ%~6j&GV-~pl;NG+gKB-R-J
zKWf9vhYjV`hlDRkVwiN0XlxU%4u{WTRiCxr^`~=a4MrLbwv8z&Q72IjdSE-+Ubx4p
zv&izl>kIwoE%t<AiIkEV@0%m$V2#kW*UUW{DzQdJghLe3ZOb}wD$pmeqpGD7vVKXP
zfF155OYKgP)2qIh_(grPNbyf?jqR6T-SqmM1q@};(VKx?-L=yQDvlGyWr2Mva6Py~
z+2TI=u|IT?>Fd#9sv<M$;WH|~avYYbA{+1FsNa54(Ps~lDMFgw6<5K5GszaAaTJcI
zg;~(vosS~2aIko(>I9Mx*idb(N03^0dZq(D{UoprgGRvl&H&-)f|oHL^wO!y<@#R8
zx}Pi!gZGNci+2RT(qGz+&)QylN_fE;jt|DZTNwf&S+9#>S}%a9c+p+~-bT_cQ(-Pg
zc)*|?4(rF7hfQzn_DYe_ry|?{hiKNh&lVhh2ak@uDHv?y>kETE>9s|1p4j#ejA5sb
z6hEu&!*-rJjYG8io?FIoW`r6WbC?aEnYbBQzCwQC#B@B=p8I_*KWlr2%CphTHy38g
z_1kz-b<6qQw4~~9_&z|0bfr0AZ1q`af5iZ(E^rXY5E>fAEB><?C5E;?FUf{nziS<E
zPJ}lw{gn@|Sh*y5x)Od)a^=_eA8%-E1UMcEh7<}QfoW+;^JGb?VB<3D_Dw4ZD@aRA
zx44^_no_3eIU|A#2`B>tr5c~@7D3F^!59d0d!>B!+qS!{N`6=0r>Z`16HheITJ3*i
zp=$c8$0h!${|-00#N`-bL51XEEiN-S@p@HPmff(gT4&VYl7;KN^eC@^h>*rkZUA%r
zs`D$>4f8L7UyL?qQQq5yu9cu0gluSkZ9cIcX>56_UI~L(={a`aK~=VG{6NqU$NYN0
za30RbH5bn7xb^!<Wq<wu_i}R2X9)HfwcGgVr2rS_Bs^ACzE*LFTiYx&FnWMw@8^v-
zL3F@(?Be_Ij?mC43whjdXs#11ghFH+Z5IZs{jUa{U=|2v)x>e~9)^n3!v{DgPK_=1
zZnb+ER~v&b$8)Y)8-P3JM_kH?#l_|GKgxK4+T-RcF$9{-6nWI<@rV!BxENCdMWQ)<
zOnD27J>-Oc>0L5>DeC{|s4HH7jxwW!%{DDOv|R*c-D$5Qfbk!yzrf}Y*w!3*_1xX~
z;zA1O{?oUqQse;A_XV~D_$dG@<9vH&46YOal&Mjn34US`RfpUB@<AeBm*PP6vE3qF
zv?{I`$f7wB)h*cGEm5-4>$6XM%9JQf*=tpavm$!i?oI@ZTTQ<JUr#{+b;>gNrShbu
zHb_jxQEvQOVgMo>Zye(+(MNCjj-Fd!KYt(glvc>Xtzz&hl`ioottGK_v1xOxI{$Fx
z5X{5gZ4_Ck#I!zMPC@zSA#XGucy6U%Ltl_(>x5!<uN}Hc#tU&pn9a&!chdQX$?!E3
zqGKuG#bo3=!wJ|NFN_eEbj`MX#6=P4bakQ?q1EWB#OOrn`F_`FhLe0ts$`F(iKeR}
zMa@UL5TneARmIC33yprYF{K;ajbAbmp^S~4pWXbln2vdCT8Q(UU4RSS+S&@rUqG?T
z|2W<NQtAxEuaic$(>Xt(fzLk$Rulj+0$7U=;8V7-p@e|n<3Ir`E{mRwg9B^Dg#K7u
zD)DHuEkBq|fM*>ecsPj`%z!{N+N>~qWY}famk+TQb%P~ZMt@TpZ!=A6;<)OVe0pOK
zH8AlL0UpZL-G8@@O8D8MGQcLX>04NgrWh~6YNreFY8$bTGqXS&4#YmYJ0i~~PP*|n
zed$PwH8--=`sNc0dQ!aG-wqTpl}Pz)nX!jEb;S%Zf6GX)m-VRDkNtKq`nVRMbKd!c
zw>CaYL_}nE+Gr<VRaNpXt=E|dQc*E7Q3S??7y2py=Dh4kKt{>!a2BlN`CRuf!2g^<
z*c32&Kw=wI>YmRwb)-a)ct?W$5m+NcV+0>ocjn%oc6Rz62^041)w6@bl+zzKskm(c
zozVrKV_p$^S;WhSVu-Ps|HKBE*SiZdWP0jo$VNqU;>m$ae09rtYnW}vLk<R7#NBTY
z#QLe=_i#Q4{LP@_gMm?>(G<wdKz*ZK{&d6@kz{ACg9~nRLi?ZzTNE06Sxu|{K3;j4
zH~If%1BmfRE362>IDK>f79XE1$Gvn{23pq+U+pr1@6X}*CS6OmXmEah1WIKP)4Nq)
z&TwO*)v~0s%e?38pFib*p!c)3S5*#N$_>v~j_o`Ggr;WPm-wU14t`s4<y3a$NBsX@
zgM`5l1dBQK8h2}AQb7+I7n}I86_*TsweF+LJE-4UaJ;9hBm$;(+8TT*-g!PNJvF^{
z!d{W$5v|QPb4t?-3jqV;oMK!A1r!7&6*d?e8fE2g1IEO~_cu+*pefd&DpDu5vcuGF
z_pT1dJ=+xV@+Q`*tN7!GlR$+IbM9h$nw~#b>IXS|CX2msbTu22Bxf@z#Dgy+d(^O|
zOJ<VSCrckN(}g~roU?F}P31uqB)6+O(0!7pdP7MOx`9N%3o=~b#$02UrA!9~2v+np
zoEXWXWFRwvJQf(df#8{(2tE)(c>u|zs-`B+do<or4Jv(r;o!v_3*HXwE<5tz;kRJh
zDDVZsz|bH3{!4)E&Q0VRb#W1h?}6{Bt3E4w89&<J8$c!=0Q1VWs5UuVYe}IM!G}~P
zV?aX!{h29}F#6M9v0?dUzbm`b4P%l^CBxC_oNz+(Qf?;Fim3E6?lj2JkC<eFLf<Y?
zBSy(Y&BWC8W7~bN)D9EcxX|la&G2Iv9U{Y^{m|j}R!E8j3UIZ>A&y*3{U($6CUof~
zdrcTZ#QXTX{=|H2^5$PfE~#=X!-8V_M4cr0;W3cn3Mf^Si$a)|iB4`SVuQtKvcz^&
z%lu#Cz)Lp6e)TF9i9a%u*{8oMj8-FPzzN)hHPlG|vqxoJL4W`kjNcjsvD_}}Nbot5
z*qG1DvN@6f#!xH`4bZm<$P9*GOG?6b^8#|g%UR(EI5O>CuwcMw3@SDOm?!m_2E3_8
z*gx-n<Bz9LD#0j2uQ~1Z12){3^@7DjE_jx{uGRkRcTD^jrL>%WYpJ>4gyioOZ-ST5
z{5$cl3I=kJwk%~kddurG7RyZ7u84{Ek?F!vKxEY-gKge1(dK=l$s9i6k5Q!Fe6x@6
zF5vY8jqMYTdNK+>x*%$RCtAm2%hVI^5@HhJ((18T(5y6x@j*Xcp`|HFKp4}0bw>oP
z;ZIesmHwlryl3)-8Oi%_)FL16bhd99I&*sB92#A<Qg(ylu(0yq!1#!EIe&fpU$EH(
z*q&@x+A{zIOI1fFYLgcz*5J5FP<39kYjV2VPD?9fN9{ENe+rn%z|k|sg{q+e00Gi+
z!+^;OQf;t$5l4BKeoVzwc@yy;Ef#seJpwqh3+TY``|^a!h1X&O%`)3hm||pZVZ<5Y
z@)Lx&HU&fbPH2ryYAFq0QuX6%1X69SBLg~*-FL=pi`poo%yGVSkR7rL#6Jc32jduL
zhMY)5VFJRNfksZ08qQ&H@^%!BDSS8Rp|8~jwkK>!dgMQpetdPdeO9Q6Xq|GrUxP49
zCi5hq)|rOPpf5lL%>MLGL{jn`#TS--HlNGSDrtIZ0Ig&~&Iv%uDfe%UM$9~u;6VR>
zU-9*bc_q$wmH!`CZy8kO+P{Ab(%phI(k<PMbmx+mMT2yAD%~B6kXke-9V*=*DkUP4
zf^_%)#QVOVncuTsY-jJ;;|#3pI?rz$$LBz~+bfPliklVd+!M-q?D})DK1{d>{gK#0
zL#*2J(r1-Qb&)rdol+=(o5DxJRWMMn^ycsq>wAr;&5x_yMI$@$xyziq>ksCaDSrZX
z_Lm1Y@Aftw=iZ|Zvc&&+HV@m|xa>pVtb)8{9f%9AB($r2V{_I$6g!M*C{6%^@(Ko9
ztfic+4~q%W%Fbw;_tZm#Gou@3?-op>XMZJlB4ht)Rp%UZ{TEE7Pc^1X6j@Y6yRf)u
z2VC?7(P<_e!1MFq<B667IbR%uJQ%bUae6%9L;&+Su%NgIC^`N7Pfr-rsHowL3y2P&
zeNdfYpFJkSSJl;fh-d`|1d#Pl^VJytYF@QPb>AH45M6wMVni%2O@|&S1D=?hpdPlQ
zKO!>Bg9?$u$pv^Nm;-0Y6g^ABLg|OY4X<01HNf(cgW`&}tMhA5y6Gl!bA|%*({OdB
zNpEp^TAg?50?Au{rCtil+>s_Y@#LkQutK0j2S#X1(n2_e#aF}u>5N(dNH?Gq`SdBd
z)=E_|dRGe0d`PQFtQoQ0zV)J*k2q47cGBD)k3*a0Q3bPMf_$)mbh&$4wQw9A1$}LY
zYGGj^4S#NNs{9t>3|5f?N=$LPdfM~DnAei6-K~~8DYtEeIBy`ph7Jm^ha;1l`=fL4
zwGkNKj07`*M;h#m9#nQzRMtS-2l6<mXZRjq#z1K*gZ9aw`-m9y@DEwrHNm!4hH9GB
z&V?U75QhMH2LMi!#A=BL);c`4%Xusc3%hh6_4mFe`p7qojY9yh3_TNfSw(EEddzB%
zM}33xm5<U}{M++${;)Kcon4n*hJ$2>7duG^j)CZV{sFr5Dpo(X`tiNJf%f7aV4yB8
zCi43Ri58n3hc))>ih}PEZZw=|7ndhTa-GZk2XVHoC#8m&<_~%oFTfkW6(pG$eCQ0h
z%g%jHfd17mPEO*%e;L#RLnmb9wj5C98#f1<(g(T=2uh!jybl2TioVrE$qCS1Jd9{A
zJ9lisPOKV?I#fllwHs~DUw2>hbAfRasLE6sW$BWC`7d8n0e@#zZTq(B5~mHrAjSe5
zPLMb_dR~hc2F}w)L*1Y=O16k6j@FEinGy2C(@8wrDEP@J2fcP)6{E2)9|k$A2^zNh
z;Iu(g(7#ErFVbYEGPfr>N+_t9F&$sd@D`bIB&??8&;Idfr{FaPYdw-dT|ExWSI+`@
zes{{E1x$0p%cjMr;w^IU2*KcU;mr&mBXYB4;<f4F|1@0x{Zoi6hwd#(oS+;HW+mej
zgGno8^Y=J6$cGTV3qPO6;7JxrKr*DZTq(9w&C&HA*{rEe(AL?-kdg^m7L<rAuXBF8
zQnZTs+R7ang|JdbVsa<47>#5SHr8{~iyd$_D!KkAzGGBoJ}G_O_BEb@@>f?``ETrI
z68RnbbA1}GyU16UswxdceKrzc^(^RDd6<TQ#^Q~13e;B&iQXaIP<;2vFgTCRgp$5r
zJkk8qN_p!XM$iaQWFqfLEdgOK7$rXdpPUx>z<a96>;gImu)^<$r&ocw*=r8CF%B(K
z*(5z!ed_3}e!mB_@c6gTN0$NJVj*TGCgQJ?G!iOc3BdF0da|WOG5856v6zXr_V!*d
zhxv#}sQYN^5zFh1B-A8lU%@wGc-Jwx+KM75uxAP4-U__xOw#`{{KB-H56hR897AwI
z2P3+kGU_Ps<cX*>Td{EM$j%mq>=|C}0*v-&Pr9{P()Ls8C;_SIFG^TtwxdK9#~QtS
zSV*617#QhuSf_3=a^OX_Bsc;x6=+0AzC?eL*rmKj%0>t-R6|*bD2U$CX;Flh*xDwO
zLSPdi{Hi#_^2~D?vY(YL{*dHs&`{LNSH?0hBRV2ee0V#B98+A(syr+$P&Po?NH39)
zXmX?Spcm&`v;NB(Bg2~h%EblGWELGY{NmX6WkFzun*W*017r<ccgGXn-Kh^;CU6>b
z2ME8xrDi0FZE%?71aH6qm<tSB^BT9A3!`4+a3?9d?bNh1zUs$#7*GQ4EcxnpG0b+{
zPkDM%%jk!zRXqB9{LiNrl&gK81{{0iv8>Y#Y!{Ye+cQefN|SDRB)Vh8AZpxDH1CJ9
zN#OL{<7WNDTE0f+QxECr5p8$6r%@%d6qyc?2xuUU)Lx@#<GYz{(X6|Eis^iT7ku$0
zbLgz2NhLwzUhISWrpvFN9fnuy&h@R$WaE6UXWdfc;mfpLK^*E%9F66ugRWT;fg-yK
zZ=%0f(`54sefdCcGONUxItG)3fd;54@Y<^-Ar}rp94IL%`CiZ&H@~L*ke)7f^BFU*
z#TNgK_wFk&l=}MhtIIG$W<79mv5)?qw}JIH?@IQx_FIx}qF8+uXNUl9r0MDDP`jM<
zml@AqSK=!i4k4kkFzDC5<wM5YDbxOFQTSa3dOTDeS%HK9Rw2><{dO7b_`9*s=VrT{
zs8f`)`?-B_mmUoT<3*s2(`A*nvSPRlxNB`y*3-jpoo+{c1!;tk!)i;=Lg99xJgb&j
zX&rj`zdPF{|9TjD1SX)a51uE-9~UDZlqTp~;wtU_{%3S>IKaau7OI!7bI4L+AInHY
zL2q)oq9Zc!_*_z<!Q6c?-)w+S!|Vh2OMV<zNxB`|)S=MiPc-;zh~7RPykD%;tXC{O
z9^{ZD*VK!YKhAACj5_VC-=+DZYKJIr^v+K9+mkNXsYYNps@E;<^6m9KnMskL`=$S^
zwz$_S6chKyna-X6suqtu|ITvrO@vllonT$|dr^H&9yLNeE$I+CB8oqnkU#C8S=rbs
zVTOsdZJzxPl_ikL0nN^ZDok|FZp3YE<?U8kLLa$uG`wOxM*>_;GDW{$aErL<cCn!w
z0*o9enZRugfRw6?z2J}luoW~UP|{bxUrGeEZKw^SufLrpyNS^>*0P6UHPGX=+h?(V
zt0tHH$SmBAGe0UaOd<U8W{(ert<*)+G1StC#)*&b-S1Xd=+`7xA&FScue2LF+;Mam
z0w^iyIv#ni7@E!GnGxUl*APTYNyBrlKO<Ol%(7BwIy&0%Kyy#7a<u|v<zmDu)Nv}6
zk#Gc_3NyT=%9m3!A=Ks{dU7=PMf%k4qSxYcKL%g_N7wZxZPgMaiy}p2okL4o5gAX1
zy!g>p_CWpM30IFgpqM5<|LM>5uv%+wCNY_9`%{(+s4x#)H3~7Gq(H7NxmH7{cAz)_
ztb&M@t>Rd)vH|u8w{ia2da#<qp%Qz$Dg>(b2hrs|3gKAJf~r-JWGXpRy~~QsUh66j
zpj`jQ9fHgXw<yH?Tg)bNIe<F49s0<R`|UGnNMizSAbM>uYF>{RVt$P#Z}hHYXZXNf
zdF4&!^{7_apf*`(Vh_3f@4f2@)3;F;ro%M)V$$&3!`0lj6P;Nd-j+4Q5u2iQGB;US
zEJ}$uZ8*SBM0RK7tJ)<@&+@8P)F#M+>-LuVUPx~oyE@wl*Sa7l$1mWH3(nQ`KP>>M
z_wDv`c1Umae3K&XYm1{k-_Fw;U&o($viM=g1~tE<nF<O!y7D)v1JjgQo?l;AHnU%S
zv7=gSj-QW2><|%cSXd-uCIQkZ89hBvRJVelXyEaFP)5a*3wFbY+66H!@1uz2Kk=yn
z?ZYS`J&<?+NWsiZxLpp@58RIiW}gaPh0X1iI1W)+tgp>yr80gI%=)-ZQI<tqE}*o+
zZs9oIki1|1szFhRf^+tA49a6^?=*7crLFum36}68Qk5~<-u9h4S&WQjXfWMnpvj!}
zFU4}>(c|=jQshxO;i*_Fh~i*Tt;H9IMTGwfn(9Qs#Dk32dTRN8L5?{8tqNDyNm|QN
ztLy5{f<Oqs&+u5P@<jgY{0GfMa^QGh&CFDiq_`z{k|$N$-ikE6`q6&v-+WwGlbC<h
z;MRtCzSI)x`DAH~8g@0S^K(BH=JzBQ#D_%Lc=7NJ@Mnw?+dAx6`&{@iL1@egu(D20
zu;YzhlRljuf@l_EPVt`#vHb2Xcw&d6>k1kFe8?6s<GY;Mqk59i;rnV)+RqJzUx%b&
zZ)xA=HqCNB+Vo{J1w#o+*7e@xL3g%IXnfqPCVjY;V0JO{o~joTbD<*gzaU-)5Kgiq
z65}#!ps$}H&-F;{+mr|PA4U2`W7%pkMRpd2ZNM_Kf~e+;q?>aq@?^^io*Z2^&9eg`
zKbQz;0U<TtQ+W>Tn8f#j`RZT^xs)RngOMXDCCU?%GYlQsF_9DgzCU+mFlfWgvyDP-
z(D8RIkXi1}TweR4s-nS!Oqdx8H=0%wa>Q>|pGFTkJwhZ|l$<2Ae&UZVjyFN!g_HEr
zzc~qOAu7r(RA~%)7#U%^%pv+Q-tSphMQ$6z)Xj*a%N}K@7GqHti@1`Sxjq_i9)9^0
zT|Deh6&)$g%5knDBve=T-J?a;DM8)Hg41N3V>AU8&z#4K9*Z=^dcS88nPq5H5VM&B
zgPw5?B;)wIV+X^?z(YFM<u3>hH{hCo$c5LhH3mXYVsH*v0j};12mvW;hGX;;n|pgE
z<3vt*<bxgvq&ZMMH0KYmI^c(ye6NPcf@wc++dt&5gDvCKj~^e2oh5Ngj%oC1C--FE
z03Gt{$2YVt?B|v$Co(1>mis@=#ndNjO}=s%encKEEP4%<C!)y3b-Loi{Jp=nOn`oK
zc+nboO?c2f`N`l8<>&d?v>$Z2(=I@=)8mn~24zom>-N)v&Z}G1_5(T@wpnR;;<MSk
z&2#Fm<lUYNWH>;?$iH(UB=X$`8MRM1jl(@C&Th_H(Bv&HnIwll3JOdY!26u(v!{N!
ztQfU#?b_M6xZuj5Exdf$`uFCz(zrfERbbQ-SG-Rn=t^hX-k||IJZ@5fm4KEEBYARj
zzSZ>^kbp8{(v;bO0z5jic4_mt!D&$dc&@W00{z-1X3Pk{@6<<39jpgPv3R(-rNEw<
zZ`}2Dx$Ey7ysbm-CLAK>JRLR57WvO<S(LJKUD0k!!-CIb(CCuUK#>m~K_F}C*nUx~
zGJEQ()ujFL4vy&c$#sW-gd?m7jW`fzxL0BHly~sOv&66BqWH~L(+jR?%aWbhIakc^
z5C<kI>g?_R-pv~z58(KZM(`E5z}WzG!b~rIPKP#k{7zog@IU4YArlg{x6gx5t9kuY
z)tK|1<WA%rM5I<5WK-sRjIc!F38@UVr^|tb{ywDpHtdobr@rqmK~-nI-sV2nz56Ip
zqB&A^*f)KgWUkfq9Q}O$@{0v$*IDi$2TZ&q6n49~Tn}AIo3ZcBrc0l2UEUXL^4&q$
zyV}Qo_odBs$Fry?mN1tLO=I(jyr*xy&Lic(Sl`T*qiPZ|xy@n|{$_0b1L3f8(;RO$
zKr0YnA*pCZl<@xI7-bc0{Bgw%q_=^$fk1cEvXKeTJV!XU-u?537B<<T9+rfYa`dav
zf)tP=is7u;j!_<FIRK*$<jTOn2T~&4oX8*bESr(6fmC$m2WVrNUHk8R-4PiT&P>3O
z0Cn7ww7mf(tBEtq#v~bkQ2J;>JlZQuo~{8L;qE+X)sYlA{fVZ4(je-1COKx^3)V{P
z@2Qi}`ZTPBdRG8w)#GA8=troep@^nK{^H6^f+2*cE8EIhn<77fFCdQ|eTP_CEL*6I
zIxZu^8m^qA50THTBT>Ol@_-B3>P|W2*6$;KP6$7czo!OH)6MUC%eOuI@G6s>z~!#G
zc?u?+r18)s0z@lJ3$`R}<(JG>{>nw{9Pg1aZmvqK*zgK*vEQpb%1siACH&pjKsea3
z8u<Hg%;yE~GrRXH=Rc2cTp8jARQ?JH;JGh%Mv?vD6xrO`9ubBeUj|KuIc>doM6Rie
zo3kgaPlSY6^=<~*tYzkCRju1swx*aQ^MZs``l;RF*{_+;%qbYJUc9TO9F$xmumGgF
za#&FN6$Bj#aqi(b1;7CPnwR}xyanKmX<}47QK3REkDLR=K{N*_7C;CQ@NRtLc}-ki
zHn?ix=opv#x;?r*1L1D>&e?o_UUvE8tIsPrT=c<xEZJMQy;~Nx`GM}4{McrJ;*o#s
ze!8mIS_47W@b_E#j~l@^EIpSpIphHGvAyw%QC4J!NdnuXhFc7!#%l8Qu8}FTjODE=
z2?TldOV#CZZjJ?1wEk&GQM^XuPPt@h;JR|->Z%7I>e~lIKqst!I=p5ij6Iy%ypN7D
z`0;r<A2l1p?8W!Rk((>K7l{KtUYn+!)=^)E(3XHBP0R<cy}iPyK7>WBBu$aK+ljp7
zP^fWou!^%ATudE_Ak-M3ZGn1FRaKQ+Na!uS5|9UV{-BNqkiG)n4Kh$YfuKO3a&kK!
z7o2^25=^Sw$N?UB(>g$SfPbDEPNGLwN>=jnnLhG;cgH+Gth8N*^g=-8(=!9-a;|=V
z`SoiMAfs0oY0PObK3RQIsh!b8mxC^ah8Zf?Y$%0{_3e#r-De+!tX+z#G<&i1TvaVS
z5%QeY)qFvnXt#8Fbd|YxDs9IzzEA$oCliyZu9#3)+!%k?(sT9mlYoH4mV(Gn`c%0h
zE-$Vfa+JBZEQi0`&FmY?GOw+=!|n;Gf?qu$ht2hGjN4>{dhrx2NsggPoKfFg7zHZ@
zPiEVcgg$dwKkDDMtSeeQ_DwST*f}fiAb8k~8yqWhxF)mWiH+F?-|H9oLk>9|Ffv^e
zo)F)3D$~ozM?VEcH%Pt#P!dh<h+@Do8V3L7f4)~*S2xKrSOJFIN@7%-L&?v85CG_m
zL#50!6!rgUg#co&tgH<Deu4c8%#8kJYXjj2Bm;OkL7Y8s2LZhtm`qJCE~)@8r&gMM
zvj)GVsusAkz-oVBU;uFYe8i|C@PN1<*n~2ygmL)vIAVFN;L6{}pv4h7F%3DQUaamA
zwae<&WO&SMww7x>i;OuN4{b>Sa?$!6a%%DyccL}Klvl+I1j+*%0yRmv4$SR{GBRn7
zM`bZXRm{!t#b^<)ENE#X;*wO;Qeu)T1}9jF9LZH+bSy}y;n;EhInswE$`~?Q!#>;D
z2@?!XiB0c?<%|pji_9t8xd5%%gq_yEv!d~t?g(5pye?k2pFuI2E_Fkw*@;#K1EI74
z#}c9DCxTczwNef~x0&C}k4;=!q*xM;hh}pfP#tR9`8h`EV3Po`S7lJQ5sFGBm0`tB
z%N*HD)CBnrTqV;f3_dqOou&^hf)(H!KCJ(M2>hX55XlFw2nS0x75~hJ$xSQQ%~y1T
zXEF}FKAq~aIqepI>ve*j^CN$1hwNf~Xd#pgAl0aNL9atJWc4=6qxNz!5jNGbZ{|t-
z#XtB(Ew_{cCPi%i+<gH4&x6c^cnuiT#e=I7_x5e^R0S%I#KEJIT)~npk-Cm!_2#=@
zsDH1&&-$$PCZ<kq%-=~m`25-vr)b-p4LFIuUk^<T%F=Imf+DEBy=%3<()**yi({se
z2wpn_rjcFP*g!k>aA{NdZP9|gy&U*4W<Y2(BH$~7u#-FhECiH$X#i6N^mmY{2|6Ta
zpI-6@ffazTc5ds{{q9)4<pjB9Lyf9n);9|<_$uFTJ(|&I0psTx9-405iZTww>iY2d
zv5#ZqJLVPnBBDttc9jNpdRiqWG?|1U_p5J%zu$QuEmIL^&Rtvxfm##H_8;V(IbsdU
z5Od~Ch%Qlt-L%oc^DT}Q15YFZmZ~8i32E=Iu}oTWXx9;9tr&{}lF<jVO#v%Xr@}px
z{xa9>7Vh-?UJ`FU_Wz&HI^xDidr(HCNH)2T#Val3y5i%EGh!S16znm(%@mal*Wpt4
zw$;aV{-5efcb*2H*$DFT3y=TolC_yJ{+J!^YdvVueYWp#F&peRNk-lB`c}YekLvd?
zcV74B;wJH9<hC}Z!EElR3lm#nr751up;%I>L|=7;hfROy`6rJ-;0g3fTkY!Hr1VsY
zykcTwZ}|YF8fe@>#Y6^(NV=>D3D*oj#tsDDfS&<m?;)4Rn}n<y#f!~q0=cDtt_h$L
z?4uw}>*2igYxm#!yVbXU{ETg)udsFpyz^j&3~aVw?9!8CVhPdzW@1hH!V@`YJqw`h
zhK3aRK_F>pHM88#_5taOcnr309-T+Cs>zbkguFv}?UBUiGsmN%>e;=kW;=XoX1Xb#
z%8~z|PsleA$qf6I9lI*)Q(06F`Ep5ML6;J#jtc3LbJUab)Qec9|8^GIT5I-FIqD%m
zzoy8^E-2)p;Z6LRQ*iHdCs|CVZySADzxK$qBX5aOn1RA-Nkzd%@|ffVKLJUTo=iz`
zPQXyVp)8>EnCe1Yx;^Z8`j9)Bs>E=pgjIxAL!2X2eLDb~xjjxOi1Qj(Fv{^^`MVKL
zeM7T+ch+KOZ!a^$S5{s=yzT}-Gzg&V9{U`LnVpx{$EOBPDggU4zN__~ZVM8B$dW47
zEe8B?@Xi8)aw1+B$XeN$giV6U+hspjkThrh>gjAxL3LWuR1;Rp^UbTO{YQ`}+t}kg
zC(*se)e#sNac3K11zXPJY+!$`1<9)Ew%vprEo^|cUMp$8?%DFWjw*B-y%l=s_+;J#
zLwTO!{&b<G-bWr)FmRQemz*KGJorrIXqet0c!T$5`*z>v(CFu9>igMXtywTt7nM*P
z-8NDJVGPHg?)ybLvoDM6ID?k?BeoxRuKqNexEg!C&=|?F*s^-+2M%sfX5@gE5@-`2
zvQEK^19mH525xI>TL^Sr4;y+Qxlx?~(P>1m$-z$FW3WQ-KNSawZQx7?Wrwl^tWKod
z&(AL_CkJpO&+|;UdbqGCg_Yf7-2;^e(G5YU!#{4R@q{aLJ|~+>7Rf>~eH49>uUitA
zQ0>pgDtOAgp06a#Z(FJObR6p>KgiKrL)^Q0Nz8$#O)RKR=gWI-#@5QpXrM6z+EqP0
z0KfJ8%Np3^)Y0Bk%se#s<s!_~!^DzRRE;rHR!6TfC;Igw?|Ps^LY-#$=De3<K!&0h
zYtZ2j!XFb(n>O6W|M$bm>;dU7-(tsTp5Ao$1#cG~I~44yaQ|*6QTz6+$@2<LcyyXZ
zj4IwmNOP&YQus5acV~h@j~|1Y+KPRD+mjKapKctexCsZBleY)&-Sx|#WpeRQsT_SQ
z5uNike|Ht&;GQ<S^uG8?s9>yqqr3d_9Q1l$bz}aVsOrI_KGF~zH9;+&WSV@PJYo>9
z?u2r-*Aig)Xr=M8X>lMRd@CGW{1~ISjOkX^l_qpfmyMLNmS&Tphh8^p#L<y83<U*v
zR$27&z760|0u?aG#xrE68|48VBp`T#<QFg;9^l6M4=0^M+%FAqsK8eRX73?7-kpZ*
z1Q<~Vmw~qJ`r~DAG&7)%047(1Mer8@A1eddD4AvY4<0?es%v75NVt*o3?E<bYFfxw
zK250HdNCf;6-GqXE?cU-pzh!wqVJ8XZ{DZWoXSX=$;hM=%e;8|b4cvziM3lK62lB5
ztf>Jj`}+oCKoE!=jp$I6jczZ)YA_5tY(N)lRG!;ZIPx1N!ce7F!iz_JYi_QHky1lY
zlsO=kY{3vm&q9Nlgq-pSU0PAkP^c$UacUC{N-#vgoDx~j0Mc-&5<@Nbk~E3alLTWF
z&ZZnsG}+fQp~Q)My$2UlV(OgvadAm<ZE@wibHQ)Ppo2OF27}w*s)W7%l2RqU0c&(P
z9ORq=u|GhL00kNdhpK|Ld~lcr8>68Ox4$BOX`g)ZzLq3@nVJsuJnX+&W0G8Pa`q<&
z?BV-M*uBN|*+Y;!|EzIu=_r>~FCn>5lBA;hB<tqW<y$0=ahPi&vAYZO$q9S*H~nB8
z!qn@t6$jfad>DP@4FcSki~35#mi$*S2vGJ8pm_Pz!sEwfKG$c@eS*$i>uAAlf=1tq
z(|nZ0y#UI}l@&;x&4wEvyWz)=Ri^z03ky@7rnMJ4iAl6$B|A)8&Tr&r*+Y$^ej0H_
zW(a=lS`PiN9M()|U;<1*zzPqP7{-2oO2K9j2D)M(|6c@bPjHX{?a_mD`R4Mt?P!F}
z7tH2Jsi{He20#S?R4kvMAYi+Li~43uLGl&QS%Bj2LCgraK|q)Rc5r}#4F()wA$O8z
z%G-mBwn%U0;<)lb_+@&h5-k!~<@n{J(OO4KHzh&&dT?^WvbnTI(nls{a%M#*Sk}+4
z0HIl^MlpkRojEIVb7?am>G7FFQ}*RCiC&XLpoH0}f0ZV49ye8bHzY&vnHUC-SAlMK
zNG%+QO010Jthku%peFL1Iaz08{+hHuzAM<<9>gt>v#G*H_c~ht&&T*9J9@Lb`zX&<
zVElWZ&zo{_c3;D9d?(p1R;J^tV|P)01@3obgL15*T|<b#eC6}&;^|+pejJdD*3<6}
zmm`_`bg9C=Vnf;H5CL4t=GaBa6Sd|&3aF0<yq)7I`azYa5j=5b9x=Dzr)1UGD^DX4
z4UzoKts(lJAQ!ek@$oTg+mbex&;2U?zPnXiCEz!-B$+%`zyE6peY0{lytB)a*!_0=
z<kB?H_VRj6m~fbt0l6x5F`+ShV|tqJg9yYKLNf12ve*o=3%1&E#-M5P|D1q8CI-Gw
zKqXHdE*!!DSYS}WbCY@~(<>*;@dw`p*bv~MLsIb3BIx04zn&#4dJ;5fztoy}5BD+1
z4ygfIz>Vw%DjbM3;8q`s`JGk+_e{`ZJasjOdhv&=>Y@%tKdZtXXLS})QmWeag_UEi
zvMi&nT+F>Yx_CyfD^I(Hy)t_{=Y?4srX&DU;PtqX#+7NtS4`*_w_H8Wy+G8$BU!0e
zZ~gfOZVz=z?NykO!@w|=b?@V*4>0A*hi5t}(d{+h4^?D@u|P;0<!tDss-Vv-g<qpU
zP}=;WHF-E37iCh?+=7vFSc_OqvFZ|>kVixvm}7KaHN1RLEtd7c@>C|WaXTVFcoEAq
zM_8U13MtcMt_LtVP=bP8Q3d>%i~?}@by*((TJG+=;P<5f=FVX84MxX@eqF|%-}Kmx
zn&ZIPDDw0vn(fnl{~e>@UumMdfByoD)x*MZXJ_?KIjl^VdD`*>&nBiq7d5svVl5(7
zg#`X4fCN!%NHx62*mRm_rP7ShH~>4l;W;7)$HcKaSLns_iYoN=u~OO{)%M_nDCxVM
zO0bW%h}K5$YuLQ#SR1_PG1~ual_~Xcvom;f`LvdBfDxwJc(`L$ZugDt&)sqGt6%L7
zSz|tjj+f5*C+7sYLkQz8%fTOisvR6s^OEJqo@XSwHZl(b>0NE>Zw!FN0D`q&&5N%s
z0h|8Y;n?OeaDf3H42Z^f$jJf43NZhKSU~`9iB^dKgx&$}BO@atSV912Sp<Ns3*-V4
zl^fVIKExO@GBJT6yvTm&S`SFMn(s@W?Ig%1V}bzJM`>wfL&Ng`iDmY~>a%n9Pgp~<
z{<Z~@9}`w?J*>pm+!@hA*_Gx#OW1wSm;^T8-S@jUCh0FJejFmbi6Ugr>S?{YvuT2;
z&Qu2l>+6ex8)n<vxHQeRv`zrA&DLU@K-?OQH{%69N?SdRD4uzB7LE476pwwuK^Kvv
z-y^-#yylv3>}vn%vNJKGMr3N2NixlXgh5Kx!j|{{{(A_D_{Z%5WMe%+7j<mB82Xvl
za=y;9dOv=2Vuc9HQdS$ck_J1QfA2Pe34dI^zfB~i*1)U5bUR;yadF8>9M6)I+S-gS
za2@rE?r&Z%_ncQRut)rAD<Yl5&d+*IsPPVB^eiPmW<FGNLKv4yV&GHn;K3GaPv<^M
zf#}oK)nH%DqC#9q25BP0{LR=W$fi_^?P)|?Z4hlvfJFwR+GQdiN<bhmA*D~DnNn;{
ziXnprejl|SbwbUogc*;(&;pY2AmIS3BAhqLR%A*kS%S_~%}xf9oJuK45fK)ope1*-
z5G}>WH_Xs;N}b)2N7Lev*UP4;5i7UjB`ir-GSJhrnP<x%dR|%PQOsvVR4rlDaqn76
z((f9qGG_dNDGka_Z5D(c%G~p{QZifbQ;qh3RtP=16!NzhT2QF2ZX&A6hXWrX&A*Km
zjG_>}Mx@5D+5J-I*Gx&W=?t4f5R1b+(<2u0%6DNEX)7t*V|3klazD5bcrxD6SWH)s
zcvKL~F*YNiEl%S^A`;AzWa(F09~q^UvZB+DGRxG8j6H%<Xt^~%TT7EYLLCNxmiab!
zrojZt)d6znce<<=IgH>u6T7=ueRxShNdq*_GN6BC)2{}L+y?)arsCgVQUUHa(CdKh
z52#(+MMSlD$-XanQ{TCef{GO|ZqeWLg8##<jY5B#P9}}?`BPgd3_+<@gcrDBa9I%q
z!e`u@uQYXqIw>XmF|X=W84ChQ%w!)o!bPqZzW-HUI+2J7DH`<V*{rI9Wd4k@;5$rQ
z4kg!rUw<}fI>9ga*;1o7tebGKud&cDtor-sFV5o9${!cIZ=NJ*p09DOptc;3{b*^t
z9Tx9AE_ze%+}7*Lx**VV)xk9dOeY3r!M}ZtjEulbTakefuvPR1EJ7dpX|M|PFi}7Y
zUEVM6>4BMacr}3aAMoOZIW2<4KoTfUpgd!_f{DWJ8+ah3Wj(_PUTT2=6sPm;o@~DY
zCfH&zIq6Xp2HH}x<sXA1Pm1PagXJI`7MwU34z@EI8bI1wb*+!YLyJTQ)r_kU<Lyk0
zQY0;Nzu@{KhsBV^r%d`xnBIM(74AdNmc=lbVamGZYrDi{cZ9hY>_*U$ChKg!npoGB
zRase?`u^@_X5s7T(A*j17J0Kn8BNPmn&}~fbFi`6p2TAJJSwS>#kp{uzIs?FhJJPT
zZM|8pRH|Ml{7nK2GU0JVW2QokR`JTr{f;J0)(}(u=%Bha!Pm>rn?7>?t=5B#{sJlo
zSpMDVn;j1`!93)7{=MUwn0YT?Z))Gd{=}&bu)e|!b)$;IjQ2}=(e;-|)Vk?dHn;U=
zbN1DuK!+Fh(N`V+%TKs_maBvvqh$L8DkS&lKxVJ3Z}Ne(ZiQALxloq}$l`w9Y&fW2
zzO2&cDEm-K1zR>>y43*2=y{v8e2x~g=ZSl^A$8*Iv4(!lRW+J?(Q2y4b<1s;)HhJ^
z(I@8zgTWrc<S>bkQEHR==SEMrf6I2&(N{cu0@Qb#+VoYfFVJbFGAKUvmu*j0R@>LD
z<)P8SWzy)CfLX@czG9s~vyoj?=4CuRh*X0QV$jx)r2dJPLwt~FK_QI!>b%0j;BxOX
zm=Itk!uV#%KtGnE1U$a`=dfR@6*P}%RTL2&eVYl4^2Mt4>{wAawOHDig-6tP(H-8@
zyeb#z_nDAbgW50rbw{TK2S}0w#S?7?hH-pWC)y;fq=+>%=#U;ho|Hb6Pe~#9){uc5
z(%68km=Z}f#r=%+)#qVySvN<w&`vq?g+9V0ZcpS`f~f2gNO*{*77@BpN<GJ-E01D=
ze~l(9&AXT9QL8c@FGu5&e2$7LB_y)VFJ&qM?;~%&=r4g*TiEY}8JNT#sPiC{sdfjP
zWK_Ou_~8Bm4qO{BDY7Z>jUspMOcwV)JNOljd$cgyo$+vQA08gsvwVr=>dsopAW6~D
z5j)y3oB0MZt?_865@&$bu1qZ#<n#O^tjZK8hzyXGM>B4=JM6v*K=G2?d#gQdZMzp$
z8+p(?8G`rT)C0dV8DEByQk^M=DWgfH>{klz#QH%?@Tb6CPamvSAN;>_I~G?9xGp=C
z^*fKSB!t{m1D-~`_+%qxLW+f3T~VWZF{H8b1{TI8`8lOb0wraJ{;fNgLtA<a?J!U9
z$++{YG<W>Ohk2^l=~o?a7lXnZBqgm6O5QyOx<|u|elTeuBO{|tSH?#};#Nxq8EwbF
zFaW~*W*>$)V6O9Ajc~N_z()*dBqs<i38RsKtq{NlV5NraJOTm(A7)KORpdJG44XaB
zbKvIbPqZIuV+d)K7V#iRj0NGJAj&4)y6^O*O+F(ON=rCAdQmW%41m~QTstk&H9I5h
zLQP1rSz|dK4HPWC*>HE<h$(Bh_^n4X{G)paMX5};@vVRWXq!+8Q8oU4^c^54I60lH
zl&BRdWxEZ@K7xHR#{-9oorp!%2TTcigaJ7!2oXKu+5fZvibL%$iXkiYBuu|gOLj~N
z{s@&BgxhWM8nUIx2Rv=`K2v53F%k8~JMew37JJur{AkD!+0jCWJN5?l;eMofCGZ3#
z=dv$-zpvcHaJDw~;WJ8&k$x{BxeVpWKE8X(`JYdZ|J8Q*&2me3-cC=T<TZ~aw1zII
zzmm#ZVpwh6uzO29tGQoSOwEz*xkHB&=Gm_3NO$1<xM0tv`wdn8t0K7~C|xB+n{lXM
z_n|$&yv}?*x=C@brdK+_eSer<*9nV`rVl<Xa(0;%Pt;<~-u$5P;r*MyfSUz|&ePu#
z-M4+F-FLZ5YsyV*f7Xu~jxGsyzB}72UF|aE$)*w)=it^AXYR6dEx{dZKDSNMhc{|I
zPc33d8hI`|K^Rk<G`c?Jxy0hcEKHal^Ui!UA-6IGIqKtZK0%mFkrX>g@8&VsoF~F7
zK)^=)_{j<1L+Vqi5)`rl9xS<?V&F2-9aYyG$zh3EeZ|Ak^ut)ex{@QBISb#opyjPu
zJh#pt%XKjsR#>@XD_mTbL9ix8O~9cvSl+feZuyXEI2Gqol`TK6Hb?bo+s(o^3O<`K
zq15peg8*jbHcCVa`AI}F$3^&Ao!cEE2@7xHvi5Um*VSDPT#C0zj+a1Rt6Y<38WV!(
zB&}PwhZdjmEI@cR)Lmk<Q48DZJ1ND!sSvTAg1o#^N4;c~O`Khh<9uV}@URjfB7yZ4
z7+{pyf&MWKY<Zc9KJT11+DuY`*)3?SR{_sHpF#8SSiUIO-vDWASxrs+oz*Ddt&gE2
z*<v6)SZK1cvMORo9${dN^QF>Azs2L>Aq^t+$Jx84I^jZH!mRU^r1*)@9RH4au1x8*
zJ%%&i%MzZ+&dOxGzpVzFn3C;rb9=G@o3bUHdslkz1$&{0I256yB)y@VANL~-Dd?mP
zMe4j<V(QoN4X-e=C?ZXILR)SrPdMd~!|omlyYbCvi_<UoZ^yyGoE7v6VA%l%F<_Ig
zL)^+r1_k~`euwYi>jg7aAO|x&g&P1!@52!XZt)C55HJ7Ga6eds0C*0>5&|k602i3M
zxFih^55uM0Xb+Dvdk{l>#BjJrEsR|e4BSq^I~^MG?7I+Kz}j|e?jMRzs#HjZYnn-R
z_55z0VUAzJnIKX;MOl-Cwhrd))8Tk7{rq*c&GH$<3mPfX=N*0f&ShQSE=_fbkU<uV
zb<?*ac)Ia8P%Ru?-mv4>q*z25^y_z#HmkB^Gt88wo@DV|e2!Doupq)9ZAXRZG^o=_
zaY7om-Z>fFUy#siBwe9MnsfzzyFRzI4@<1Re0sh=A5!Z#=lS;$VVj-AapU>aT4KS8
z@&8`Z9^JdHmEZsQxYf2+H-#$vyl0hYdd+4=+MYWrU@TMjQ?*TkQu<rCfp80PIuUel
z%4{GwQ9_>a_9oW7T~<QEf2O~B+hL<9@L7QUIElnfXr24Df6o|0=EQ3LegH#Nt#Myi
z<dW?|+*c5+H9oG@eIK6KbW#;~Ging@Laacf{8orEA72vIbNEWa_m=1NT_Wlne@3D0
zOC;=g)oJ<yj#suA*g?1qD%|PgI^6k}?~0uW*vh7?(k&yAOkRj1e3QzYOw2p!*yub_
zhTGDl8xttGjp?T#HAYl3&oz&K{YuoxzIuF>L`;S>_ZgV$c!96pf)v;s!S?*0Q)aHi
zn=8tK8}#O#QJE<ax_otD4+HXeA`IUn<iQt)PvU<aj;%F@@U|JcHmx7~0_|{Sl^cb#
zS7}#>qd^2$q4r+=7lln0`=Id?vz+1P9IHFOP?S>pXlYt&fD*$CV-3^Mvu&g-a<wL}
zfXgr%(dUy~JPsm)$Y)SUPm&N^zDo9C51quj2oNDKN6-GMoALO%pd|lA@f}_Yot7?g
zTcJ|LCmN!(26m7FwMB*1FO@=ULs9BhDTfYK`rIfj^Qpw&`gTyiMvCz-;kEHNPL<H9
zki+Bo>e)0QUe~>I%DUN6P;7!L>t;i+rwAB?w?`n&j%488Qb?f(-DWE&S0C);4|HI2
z^TMlnU|}h2Xn45m@jwDFlm`^@`gNw`KuZ7&0Bb2q?+;yNf+YK2xn>w4(<mk(p>J2X
z7x*l_F~9UYFffoZAveCYek;D2rq51&y~I}g>x>E3SK;hF!q<#FCSs%VxbxQfz}8{7
zj{3LzdEF}?tO9W)+-3ASAJLnu=bCyK<JK3M8We&;+4MKZi@)lEAMLv5dl}y+%&)%5
ztl}|9Qm4!vRuV-9UUn#`SHOF`*LkcCs7YX?)(MF9Ov#D|qh@z0pnC@B=Fv<p8{qE+
z5hWmCTZ<#sfF++aaj+_(5>&5^7QHCIYzFqEK#C4J?aWbNk29M?J|uY8oM-Ozy2J_s
z02<&FY<2=;Zhld1y>F1gxn`4Y+?PbFEO9As!tl(|*Ir1b24M$s%YW52J-D?La(`vA
zbsRsq+PiuPG&R8S2E@S+3Aw;f1qA3K0_kIu3;y=&*XD37#%N_rRT_~n)swBQRC%{~
z#8Y`(>3&(p)IlmGgC2zmxumY!d}q#mHh`{|dyf6_!*Xjrls2~BhG(QXZg=&K=>Gb%
z%%0N#tki$^@c;LZQ57^uN^5IZ!M)EaD~?9#yg1)Dp@PE2c+qoJOwMSJ2JR{4mb#!a
z=x^b|L4!$#mDgVHqF~-VMLtoJ%hy=33&l4(<aO^_B>K3#B?bklmU_Q)T+g+w;@v*p
z1p&t8WTdtVdNz}Zeqv8y7kx%#l=*9i*d252ryqrPdb59P5qAggjLy0E#2JcAGUQ|b
zrBXTg*p2VEgV<@U^&~Z0Y<a#l(H|xrV}*%hJ#cwcWQIUM*iZDhoAI^d%YpUepf%n+
z#nA*utC>`zd2eW*A-pedHYi1q!?n2|2mAN_-luba?cu{8VDv6mq|8lbm0Za<s-i^%
zq?kH7I&ldJ1mL&T8I_u^%cErT1q&CK@SmU+e3g~DGFn9cv4nz)TmohTHooBI1IWyg
z+ULYG>|n^MRI0PC?VezSUD|o2mI6*)@PGvMP|Is2F$@oXf@l4*UkGqRg^V8~{#}hz
zZKAL%|6)?G=?C}BALf5LKd<dCfsQVpe7Gf2>4-AJcX)+Ggj}66=7AHXz$Dx$_`=`l
z?H(02jFY>Zqs$9hKS;#LoN~-4-8{TAA?ImSk8Rc$8z)z3o}#L1xqwJ3O!$+4ppoub
zgid#(VWZUalxO*g{%&8Z&SuKf^10)aU9qq5Ko(#*99EDx!UsG3xkf8YFdaTQ6_z#u
zXPSh>KO;sX6g@xxA!>#Ph&28&MIMGz0?t3jy3T*9xVyXa&kQwFB6I-j4)DQ0)c3!4
z_Xiaj|7>r|2*2hrSyOrEy{pg;<+S)bl!F82Xd27dJ(t-!UWb;FQ63a@YCb#LhCth1
z5rq(iFrQjid$Evcv2GRo=?BThmB0DC*p}TJ{vC_O$rV8jVa5+^8NEfcS?Z4oDljVC
zC<9g=9rbi#ifpX-FK@m@YrXFcmZwesy|EwkBxJDXgyA<T?om7S?N;1?7;s|&F+-82
zIoM-6EjEL1E4u#`;C+Aw2EbFmk^TipPMsFPWTLMRsUJ)vWu}^U;cP<>S)8Ta-37X=
z{oo*c1QssAE^Iyy{ai{ZaP>9iTB}m5I^8CXU9diU)gIE?NZ8<uL?GNQCLNPDF_JE5
za<PP-{twaeXK%0e4toMINqL!VR-xa*fGqs#d2H965XTHc!uQO#8`v+JW=_=nwJd#;
zFp=1+wyA18kHx9m2eVFnG;V~G%*Dhb6?oO>?mx-q+MPuCZfXnK^!=ZGc0cmdTa@^q
zaN>OIxbow^@2ZE6?8CTGR7{0FK2|eywQzE&bPB!FVZs2d;`t)SIBKe;A)l4YC6oD-
z;aW@dff?VRwpFzKm%?v+$1TL*pZBJb!gW~As3|s<yIpbFIL)MjgF9wm1Kf`v?|wTD
zxZMf<lrPzuc;0=kVzac-B%hK;5a;|-LWje_%KO>A>2sU!K^(NyK_=^pbh2~?=4Osq
zRTL_G5Y>`V*2i}nGl?C<mc8daBj?mlq1GsBz1~jfQy&-lm(1=^1wM2|CkC&PY~%@f
zo)~TZ5FKXCHvYvd%FoGKX&kDhPxOxn1-?TvDl#Cn5Xm1{Jp{9pd?j!BB2DHbMS37@
z0hMNE_fH9=gc2H<H-#n0S%73zmHx(%w+VChfTooVFg|^nBbCcSuE$IQ3eQJbEF1<k
zI1n|8su%I&BeYds?#<lK@>MabhV9%%CgO~gaya(#zJKx{L7X4%D;aTMi;0=t=_LwN
z6;Tt&FM|@3=CZr^C|Wy;Dflh)CTH~@$mRH%;V&Cf?I=d0#m_0OxiMxq>FBZ_5GOUV
z`i0{{@aOas0=mJJ_R~&fFE=0SIy5pQKr4xD!KV{J*VC)5KbtcaG6{2xGp&{q*vb3C
z5^LZ64xHv77Yo>uANick+kpLno|ZMB>pyTb=jWp>NP(b$E?G*8NOfin&?(CP?GR;U
zWx*WkA#mxT$z1YY9{=4R3m8LLWy4{Sn4oY3ck~&^z8zoTF7n#M{PnBdQ|xTk`(V!o
z7eCX1!YXqa>4^!dWjOcyCn%g@8a+`xLC4;$yOuENnjfU7-clxPUxi=Sc4AXYpzJz_
z1R&KqhO0?Nas^-pojGjU@F2iiDZT1K&#VtJCao5^(<5xUF9g#XA+L=CIewfb##u}B
zPsp;KVF#;NyCf+H$jTDw2OiB-*wm-201V#Wg%6l&Gr%ZhJzXphe#{50=?8}=kASWY
z$Wg#D3{)P^C~$y#7$~0@k`;jg0etxrT~(SDF%$rxRyV5xK#X;DiT$xpczJjp09c&1
z7IS=R00Nt=`0mQMCKoHufg%r!Q*DM@jS!NY{r7nVdCT+irY6ctt3yRci7;j*+sE<H
z1ohtzv4sK(W}0~xWoeSP>YHT$(1T#=V`;wVrFW2O)t1u(xo%-LWLF|LjHEM>m+WWB
zy|W;0OOjPgIhIa_GG%J(u#m)T_M3=i<o{l7xuY;jxkzQGlek6rZ&CffKZ3dWkg+80
zM`DHevNuxCoh**0kZ}9X>Yc8bm3_n7#?g)s?+o(wE+viTZZ;26DcI;IFS5Xmxw#h@
z<z15}+qK-wWY(5f5{=b6jI|u($hCaG_Y#PdpD!$uBiO!9Xf(GFboQJ1X6!MYH<NdJ
z9v-{61BA`(X$`swm5!nLg9LvQ87?OV0w4WVq+FjrL5uf1F4$ye@VdR&zZ{T#{`SOW
zex1z_`ggSewhy0|FYn@&@fy^)yx)J>m$@vAx_jfb`#z61x!Uojl5X6uJJ0ENgop6w
zH*f+pB!Hy=ZW}r<qhp;0lr!*j@+*wJR|TM4u*0;?iCW=@*OGuS#N=jG`v@y%oG~Xj
zmm{88fbW`W5}19#2(Zj91#HZTFu;;qG}|^k{U{~1F7nZEt*)}z>kVBrCIponaaU%0
zuPP2?!(qfK;(qufrDW8H)Z}%9^h&C`;>x$`#C&>8TbWw)k&d8zGkSHP7!w{YBJ(AQ
zDBeTAe!nn&aO^pgK_RweBfD@IOj8SrPKNAGH+f=?Vdjojw4lnw{sleS@kOD9&N5Aj
zpK20jnzAOl(pD68-HP{7mU_nQfFpkOOPM2zSV;O&R#ToMmj$bD5ZLBFAn^V7yD~v#
zH}&O9m=)x~76sx&5JRQ-NB0{BsY+^Uu)zbrwI!ve2X@(Dvkh=?;AaLq-ri=X2W&1J
z{%~C_EvfwjtxN;uR?g&L_Xh{zW>2<$-576Rx#JU8;wD%pYc@Bp4~Yy3*$*H(6T=a_
zXNaTRfdADBSQXydUYTol;-mf%^YyC^@i32oSavw0<d2z+w@(92ew}~%-IgG9W?<&g
z*Wlghb(VDl<@n*r6|gz^x|f|OiSs2&)v|i-q>2%$RNy6Qqc15erNKcWp&9$bWe!2Y
zquH-tUlO16u5bg$6eG}jKY#uZP*-){0N`T(lKk}g2S)-`GORwYaLb^9;kaJDzutNt
zSor9i@`Ill%nPbAY&-FQs6~@mH_&(oD}4SOsctsbE(Z+6oX|hZBeSKuo;@LG#vA%F
z=AHUVnK>$oImfVucFfXFRmopka89-|*?oGvsq8zY0Sk-WqSv?Q`q!;=1$q$%N-220
zGs71Y8np2mqHP5-H0k?QrF%4oH1yg^8&)@IjHay)oF-#oo4)>`9FE;h>r??T6XZ{f
z{eQ~(n*3kIiK&Q(!eft!+kmsE^?=_78cl(G>R-n6TByh3n%XV!P322W&u1+1B(85<
zyuzq)C6xx)4|LrQkX!yJijCJS2WQoF-nlJ&mhjsNKlzGkhk5CH9fL5IGb{(m5b{`K
zcDU<-9l{zvk2-f_@TYG?UG7q3Wo=CLU5+wcUrszT{yzVa|8b4&NcLolVv3wipN35z
zs6rf2US^Mw7|t*CNdmV}*Q!!V>5pR=eM6`}RZNlRa^G^$w!p?k$nE{`0zW@p_VYKn
z5@$-vuv_w=tBc#eemfKy_X&E9?%VW)niD6inI`O#Ea^J$=oL)d3Nq;d3ugZuoV8Sm
zsJ+~_uU`-Boa)PlM&fPdsEg#6>9Xn__p>fK^oFMn(^q<Y%Kmuf2Ua!!Jp;;RH@6rl
zPc-1=a$^BOziC}d&L^bx#;A4c#X<Dv-|o_mN8sP+OIQ0o#q3nxNCHhJij9}ZI8BCk
ziv8fBJe|FhYr5A64=ZaB3^&obup$VG!JHDNsYYb+EFvmGM$0e@{*iNjC}tfxs)r&o
zG{+FxH^dj}lT?h?rw|KG<!L6S5kRbz!zn5|L~^?b2)eoBmN0vyHlrbO8OTceHz3CF
zNthH0fbIegApn{Jr4s-T4DzV~S;~V=@dN+Lv}cTU^5bySiw*Cd;n2sZZv|dJT-PyP
ziW)@^t{S^tGJrWyJs5a_S=cD2=?LM@td1s>k??t^bhwAac(FXDaDWfs{G)L^rO%By
zo}2BCt7ku%42o$VBE_s=Qb@p3*T;{fpa}fV(v>y&(MWRGi0Ioe)8ZMocU=&6OZB4E
z&X-vDE!<$`=W8E2%yuM0M$^QR7D>UGa|2@(wq?W?V^pf8ON7>4{s~<6{CS9Y+Z1Sq
zU^B^CQE#X@w3iLv06qC&LVg^js2pfH9Ro;d9XY$8=wNnFgNFvX^#@`kz)x8L;CaAF
z33ZLNy&lgytU8#7hz2te?bE}D(;6(e`1jRjJ%KvZLmIl7AERBV>nR1vh_r%Kz+>!k
zzNLf_(`unf7iqe<4c9kdki0MxFMUjy=2v1HuJi6?7t)JkG+F@rtrArJnS_QGxn`R;
z<#C6svR76=p|U)QL@Re%e$+}Af$$BPUgj;t5;E_+v4F3*zR9+%qc26fb(r#A8+>~M
z9I-R^gvWBt>Xc#n`*!DS5E3)rBc_|lZiR~h>|nKU!GKRc-u(YxSBe;RS5Gs{eIwyS
z+#g-8-vxB$0xo|ADyH<Ykv}>5S(}-E79fLGphQKhC2~y~f4?N~mR;#)(21*eiip0j
zj#kmEH<|(Mm?8F0Kr*;mhvO&l@WbkE4%dI2p3vUi%^FR%W0zYl4iTO2Q?7iOK-mtM
z?1;Hc12MH_ihQVDotMR?DVKEAcb8=|)IPRP-l$0EciyQbNO`fQdDRA88VC7pa)%F^
z`))$`kMmq7<aqwHO0rI{Dy0RFELxwX!mAe9)(e3aBBr<yZvV{$ipkWW0Q}}mOwn?5
z0Ch$$Y6BPg1RhwhKb>oq2MGyL7QYUCfT_y@^XL*0j4A*cR)~TT4{gjV`T1f4Ltk_t
zeBEdBhf?Ky<#t&rYjY#vd3wcL2c{z#y5R|r6j#QM7xEd%QFbK`PaC0@&z{j{4)zV`
zBgxB_Wi?JaYO*R<7`r-FthxLCZa~M#Q0J^HHe`)?7gCy}Hjid9Kz>GuH4D@#2c%Sk
z=>yR?wafPNy-%{w&#pnDCP<RS$A|>!4WPAtAWzmMW_%~Xa6`}?B5thqUkl<X#iG2W
z6uZbpy_7n8>@oT0mHaWr!@%i$_PjB8IS7cA{-iJjTqH<Z9na*Vu`+1CDR@C6loFJ%
z(X7r_Czp))*ts)XvTd6uWl~(BNPWXJNdC!E^)I)tBOFYIsLMYP^wA|6o@R7X5D`O>
zNsK?`_MEE~t7H~TwJ;K!k|TPFVeIThcn9OtsY8ZS(8*gq4cv!$2NRgyy7`<_!+Km3
zBFw*)q9mDvdzs?LCCE8_jD^jFhEyEUEA3&zL=X)Yy9a(<;3(mB{wW4HU^0wsq#9pK
zDnyjqkT_<97mRjn_y5vlQa$8>kFFp1Si4;VfoihiT9{|5-rjjSPZ%+N107LVD#GTe
zUv@zO<3HL|@}p;#$<N-fDv>_3Q-vay_->lo*uHy;eS;i2SZApYMT4V!hag3i#q@Nj
z@?@C=HU7q{qIpn<=(m7qR4d5000jj+LPJyReuB3tdJ?9E#37$pa`>49zl;%0jttz$
z;4~mp#NfK`a7oC<)+KmzbwrQ8eDZ4+Yu04rzq8)|{BeYuP&V{FsxtyRAKESCu&B%O
z_~SwYC6aFVEFV{IU7la}Sh>$g(~?c16=nei#pV<=CXyq0b7^*p2y0x$a!Q3^mc{vH
zMBK4xxy^i7@*j4#p6;lsYom*ELE~sTSYQvf-bz_yzL=jakn;d^G%4F|Di_V9HBVX^
z<HDkbGj?o5plw|0<q}hFo-9*#7Ji?(*QIpjRd6~%6n{e15$joup>?*t9!)Y7eVDVO
z6PgW%H7|{fre$7^Lt*Ye|1D#lJ&pf$jtU0-0FFpEN~6RlAX4||_!10oeZ&f~2~3&r
zmEdJ9jZ+Nv1(K<g+D>*-KmY;x*5iJSbJUiqQI7#eozZup2t<0W%97t5MRnBm1?m?H
zQs3)jSm7b6WmZ)2YT_{A*$QcA#V08%QjvE__kUL=D#O@Hu^FIc8lICWeMgj)7xyfh
zgvdx%54n-vk~Xxu<SAZ^4_ZVRDb-_ntz@=SdYi-p-%S~o8Xq3;)Gv1j0_(+teWO&n
z`sIfhz>9|}F;*Z+f4tC*yKo6hXvbkntu8D~=c9Ug%>Bf~10U!c8)Pk2anyhu)Sm-4
zK9teadk1#ZwI7)1)Bzq@M6qUXLBY;WoZVjt0oLdtkERx8FpK;$O1ai;ZXztA+GOuZ
z;qOANKIOy01OrIB+duhsu4-+WA{F^<zYZEyy0%8wq|d8vlr>sW*$)YBeUc>PhxPLc
z-)BwX;It<+>Ha41176r;RGfONQIc}@ze9LV92{#$0&T__wFM3bZx)0F1fq*S985*A
zbl<1iXs$(4kUc-FR6s`w{N~sSqVXV#Iz!5FiHShx3^aydf<uuw*gG|a=YRHh1?ZzX
z%;k9sU$WL~iyT}mgTmrv0Z+CnaP=kAld&1dDH+iy88W2We|d*@ffXO%faK1n<i{R+
z;J~*nk`iFeI<KSQXukJFr5+#6QJd@4x9QkYxBV}k`L7xb$E~V_2tpw;^Kre9&$CN%
z3!E-eO21}2+TSr#ynQUtY?aJ4uPFbgdM~*xfU3?S?^{kdR6_p$QS}y3S!VC|Hz3{J
zC0){;(%m2;-6_&79RdOZ(&+=zl0$cwba$6@NyB@X`TyRvzR#L9!^}9L-1j+G?ETrC
z=^Hp0&qBqeV<fiQy-$h1f1N<)38dpJutQPF;@WjHmj8EZv;A}Mp_g6finr2zbUeX9
zHBF%PaVgi85UZ?M$m>a`gZK8&x?A^XhwsRmR+{RAFISuA>L0F!)o~n4(-c%?aUJu>
zq({i2Gm~HA6+2Woo#vKuy8q|p?T{8LvE(@;iY*F<<`g5~Oi(0wPCih04#!lcKP({f
zs{mo!0kykgRxC+5Ar#L<T9<;W8I#UYD#!1<K2%Y$4`aU_l=$#b?(5be&5MpWPN9Ik
zS!P-UYQ#L-Mo@qWbZZQeL8H5pad<2y+KL1FjfSk-yM4P?0#tgsq=4t4JUQk@raW1^
zp)38Z<hx8MUOd2v&ZGA^S-=+1(cf;-KZkX6g{yjN$S2j`?#!MSsrnNeQL2$Q&Rbw-
zCv(|Y---zMolj!(Rw0ue`TeCB;f!OTLaJd2LV{ogkC3Ao<;q(crI|uPn6%7c@$rji
z^IHw5UiYd0hR4;FiTineMOi`n4G+%XI>$z^3PL%cvaisPpTG5~Byy?ry7BzLJfYLH
z7w$&St*!-8OWUF&2<|0kO+Nl5UZJ>{6X-MeN#TucZ6hD=uU`xr0Od4S_vt0GPNS$B
zD4c*x!|dtd798;}Obfs@I=FfnumINz-;(1iumuD<chK%+F{|09BRpPM5yumTkG7JF
zUZd;`DH0KAmZmo**gfAOc)n{J9iU$1rl_={PB6^~3!laEUrgrD;H<y-yQsQ1D~Gfq
z1k?5heqpzrfC8%e?RM8VxWI0>S<*2+3}=3%LeJ&c?W<brO%@tmglWBAWgkP+Dd%sh
zBqt-Y>RS<xcz~*18L2c6+&K=#&G%Tl*0o=G;E#@0zCnZDV2>59%F?jqN;QpJiAkS_
z1oUmbCPW1XMT33}I52%h`(h8T??Hk4!tGkE&0qOHEr746DhqzPw(W^TH`wz!Zx7;#
zMkS6GfPXwyY2(ZjH2J=wMs9~dHJlI|jY)`R5~i8J9m11WkzAw^<{199&&Bfj?VPt`
zyxawU<_8h~h0oM&eh;sC9ef6L)m~HTIVQEpX{)jed+f1`d~T^3cKplO)DXh$(w(H7
ze*3rWVbNyshYhigp%<-2gSit^qJ;LeEu*PC2=K1Hd3-9ie&_T(5m{O(TOo)ld7f56
zRBT=%Yfl)C*{HO!k$KbL$BfgXuW-E+SGj&o{PZsZ&Dn(zi*$x6jgPWNpX8nKt(0gt
zrlcV%`x~}4rdAPhrmpeIeRSVw3ov?o<5Zg;W4LVp|CPyqUkoW`$6H*Sc9<35_S+Ku
zR;gl`Q_&Kt99ez(O(;J7Hqk<Xhm}D)9;uidmi^;S<*4z<FAn31`YU@FCQ^@iw2MXE
zmYmEQq41GEq-@`tYRLiWsju9?dFBW{Wfs@cP?In_ANxFGi824`axB+=t&-=U!DeNe
zJInT?w=J_r`65O~$-a%i0bM57h*yJj(@tedjka+zJsj;i15Zo<RbT}O0RYG5v)M_@
zCQZ{Hu9FjG?K<aUtxQXAK_%<quXM}-G&DcrLRenEMj9hNC9WW#7pVB#%4Qd=i*U%p
zrb3t$N{eXL0z;z*^BRekdoVXOj?P_mm$bO4(YfB_x$yYUaq97aeAnZj$+TjU(FSb;
zw|n7TFF#O>e*O9tjOD;10ixT0MG?ZFp{*^=ln5y)K><zu(7wbZ&!M}JLtQ$EIi~&s
z6VH+e$C;yF2e0tzYdlRLrSWeh&uWP;)+`Q{t;Szx^y@cJ!K6-dmWYhZ!qL&uCQb=6
zB`d9y9c@R0C|XxB>%Cw<Qmh#LPDVepv1t$oGjgO0%)0$ZVId}x;QKbZ{>5;al&ldd
zs*USEr5};i^m3`v>72ql&5gdAHbV(u&Hvc6`)yim_GJ&(nGtn3I`CudFpOIFI(H=_
zRXimXGKvue{-Y3ES(X@n1GA-S^AeBVKMw{u%D}pHcX#(Pmc6|!5K{*Bf7awEV7am9
zB>`&{pwmlGDF71<Ahf>&n2gd=z_0^d=72dTkbk_;K!D=U3q+Oxa@vcTP{-&q$ZY5X
zgeyZpBp@Us>sfdFP7OVjm*24u9Zr&tY5WzJC|qt?+fMEnvGVio5MsGVhm!j@isjk<
z&l?DH)t!o?yR9{HG$g43sWe(Ml2Somp;fS2a}hl0F^pi8axb6aq+)!c{SKl|HwzTE
zM`)004WYAd8A^+TM}48W$yTc2R_e(7F1j$&+!P8pzbR4peEg9!jJ6{=in*mBqTkRP
z&LEhQ%2?s>t2^OUE=?2_E5hYZx|`cke--d^KRP-B5tOHMX0axO@L?uEHvyKxUm*Qy
zJa(R75)Fjdp!iA>I{~$Fs#4N*Lxwc)r@Yt}My*#tuwRDb;|`agZG_SN>8~&Gzx^Ml
zPBN>k^A7GHDg8L=&G&(t3o7ckydM`_6&4Mx%jU-b-1_0@#5XWfZZBkU$63JMU1+kB
z5mOpdN&<Q{elaafh0X}JMJt&*L1n|0Ec@!C$^U$uyf`xYPrE$sR2mI_E_eO7#0*uB
z#+bG5di@69X~qjudLh;D>3X)v<L-2j_C8*=c5)+<UHkWG*Lq#*xJt(3uA2GN&CZqY
z&cd|SQh_!WTbrLYGlJ~KWp})lZ-%BcpVE_?D&ON7_iFPqahik`U;epKnS-P=o7r=u
z{_t?R|8KFDziv;L7i3HBL(`?SXx?#3&KVJ{s4d|hXH%+Ko9!VImu2bmj_2mR7b600
z@ksP!b$ALqbvvY1JNjhoAiEvB-^oML!sJ%o-i^Y!2HM(Sy2hK12L2i_Y&tu~KPicU
zLtQzYcm2#0EcC(Ss?^oZuF$DhM6MenMHL8P*e1Gm$zAHzUx4Uh9}e;iSke}3UZy1=
z^3!7%9u^PQeq1an`PIc8-4lFU4N+fAo9;gkO15xknaL=5<(`CCeKm$Juw(hm7Qv|!
zK3I$g!y5a$TMBuy9|ws+D~!x@i4|5{%B{)qUHt$}g1H7`<N~JjBBsh)ZVOnSw2U(9
zJx+MRFtxYuQ0j$|dDFGrg(?c+c4+C}CK?TS=nhwbWLN-;d`ZV?4}4`mSMh_C?GC6!
z!kNqJgTVT-Y7Q_Z)ZEvAFb}|Y0a_3!sm8#{5BN~g!-E0Nm}Tfie*980gG_}NXyK<%
z^njPVxeWE!_ac36$sy+HkBkFj{$2$cw6O?D#i|Q|0Vxz^+m(!O`lodqAOEmsQI)=0
zBHrigRl})lv1cpKGc?b~Pb)tb^*f$PDK=J{e{nNl&MLAEf5zyny*dr`4-#^KS`EUO
zJ$b^zZ9^3ECT@KKs<Zp`=UdKh95sfZ;!d^|mPgmi&hk~wJKy4>5c=mww2d=i{|EY6
zC`x1t4{oST1UiPT{JrCt@s$-75FP`}wcD#E>bL-+1DKjGoDtCH1K@pLV<WLRCg`WP
zy`!pFxHvdoqN%u*ftnhmCxHw^Z1iwIAh4)g-6qEKW~5f&oYd?hC7PaZ-RYi|GWu*m
zfX1BW<>ycTx<E<$h^eg5=mWR+ebGt9ac`{i+D{5jE`gpc^ROoW3ye`&f{_=DpFLAY
z|9Eml$S1M&d)K^OvM~CnymQUMps;TWPDka!6<c*nS#|CSlQ~T#B;qjg><!PY7};$x
z@BjCsk3DA*(HlTgr%b0kmbBFSn(?sIdNXN|SJY!qzV=S1YNVR~pj=F-$8K_2YtF#e
zefkRa<Il>mINx}YGjW*<d2W~6n=T9o>#fuG8LIZ5)+cr9UMpJZ$UcfFYwJ&Hd5KF^
z39OK3Z5))7u2;|2;W*E$wssM%@BT6J`L+_lcF{e)>$OLb%~qaz^OxeR38nU{{Vi_W
zbw}^TRLtb&Qs0{q5m6N_%pDfR-$+#3QZheiR{E>H_HA}J)PcTzu=-TH2Dhz{0olEY
ztY8w%H>p7a%Yb>s;t}{!KS+Q<9;i#e@-cl_Ml{)qlQ*L5z_VpVmw?C^nO6Nbtmb=#
z9ITxLs!-aHs(IGo6Tb4SH#+ys0Gc{#bpOY9YL(NWs*0V9d2yKRRtxQ2!Pb`#mYGxo
zp3QW;BvV@b#%Xg>NKFRPm5c#_rkd(xtmkkP^ZIG2siYrFVOy+|^c|hi#Rqb!gN>>B
zrS+kq>qY~I9oTy8$rf>3cufzqqFOTp%nVG2{_4HIaU|MZZc<}Nk2*mKl>iGvpl$=f
z^?=jR=ym}0GV67ydlBV=h<f0nedzOlKma>9yYiMKM}8&1PKy=2V@S&%2PAM|@&aH+
z`;X8FM2*{KEUtDil!#v<JELPg;y#a#)?W=`XymlpHA`gJG<)9iu1bt_%6@+isoLj-
zIyMHj_NGEH{|EV-=|yDz);swpuLBV+UT*5ynyLx-L=MkWwzV|npV>zT>;7Uetn|Kj
ztrJg?8h2i6=$_9t6<6dpkE-pPqaBlM$x9~LjTjTUH$$u8*4qyaiwhFw)x<-ePe3js
zSpR_xCLr(wclL{j3E;b>s5-*XNM3FyC=gJ+Jz5SHj6k9X>U{w39^Eywv<#z`jlo0?
zEi5iJ1qUi{91d+u;7WsUw}9cPvwOV3AU~-gRcnHk&Ov-I(XxuqB(i;cOjql}f%ar$
z;Gq-&(PZq}^|=FE!?vTBqfTwMRzKP-H&!7-yxcT|1pdolRrbTdObr(|ue{<UayyGM
zXJy6zeCiL2&iQXX@%@ua0*}$QEL-c@tTwX|=_u)mSA$NWT!#5v@+{q2_4i475!Jll
zW_*vFyl}Z6*$R^()x7sye!?3oyVO~B`U8|<+8+~2l7fcLp?jTKswfzW#hFf}+IAl0
z#J<dDCC7OS2R*$$yA~W*mEZc`u%!lk<AQUYOj94ca0j49KZxGlwpkpwsGi=GwYZ{x
z#TmtgR(rrA)+w{ttf_OlsAcFwFf$|_NtJUiJs8BrXUWi`8<Bo9(!?R<VwjUsqDUu?
z4c?YXZY*gmb=;xj+=i>4cV;SD+Q9clsra)D82(>on&4X?E*Rb=&VUQhVN&WWO+`{J
ztq~~cOew~{(IQb$?p*X^cV|W|sWpo6Xug=vXSyAV#xrJWIWp>)V<|ZBcCfQsZZamj
zBqa9Ka2c!o!#w>93rmS;{<TS(yOesCvJ$Q(H47~)k`xk+bhm>GM2`KZ#rVV~VNjsy
z#p`;okHm%0tUDIp1Tz`7q3EjbglZ%app3$s{H=$v@>90N;AX2OLh%htt`}W*Z*pqT
z2QlXrfZYh57@%|mw-7h6g|TsfuW0P#J}{ppjP8O`4|p&EPC~EN{s#mSbmISf4>FG-
z2ZBI)WWU^)2#n}5va-OY1p$^UF!v}fe&KAf;17MYe(Y}PXF#18sk@=e-o}?9goj;O
z|9DuKl2UrtEt(Z`y!+^X?Lm|BS}G;Q3{`Rx(`xzs6_46UN|@2J-9ou18<ancZv}?)
zBjsgcNF`oPMO*F`?xJ_k#BA?L;HM`z{WdgFH@2lyC&yyU^soZMFt+EoFq@>2COf|7
z>lWrX)l){B>cl^~WB~FFjAt*5ED&`87Irwm$Ed9hzE@CSC<OqvbHKA=^n5ql@OQy(
zsD_}$knPNqR2>(n4FI_pjIM(y!9*IoU3nTs8V30iKMt#E>MIJ-`IN`EwOJ-vrWYE$
z5A^l(k{eC}Z!U2~{nzS8^Bg7jhrP8ehHkCTLuE0oKg`!3{;?N?ix0AQfy^lh`w0a>
zrMjls`=<P}PeXdjw$PU^_<r2c^g1i_zfaH4ulY7JBF+E)Sq~9Jjw*;!ZS0D6mQU~%
zcY{QKP&d30+Ng4?+&ebbNc}0L;8k5Ov_+VUO%xcI=n=~4Vvq5&G8DeEhGY7hqwBKm
zR%RV4#I+h`t0Qu~FEo*d`g(mh2=M#Sw2XUwtMqa~Uo<tK67t^hAGmlQA}<nCP2C$<
zeJ)L)y0vLN%N6$AlpO`x9MHSM4wYNKVVsKY#c<xfWmoQQfQa_7(?`B3#c{KLB#@6E
z>?HagQhB39YA_1`i)z-QGYUx-`m6-S#UrwBMru6X9bpb@EJ66&OarMMYRMEMH7^v_
zd_7=8oLA9)OD|?NQp>lqkjX-A;$~>0^w5d(|D~Z{^mkHb?fAZv$pu6Cz3jWdJt2<N
z{99Z&bNO!EU`hjVUOY*5G+ZA{bq$6`G)k%gP&Gs@J1`^ZRJ>(K|8y~NoRV;z+q@by
z)YP1CsD;bw@{W+3pq@ha8?u*8iFqXCCERjy-yGP8fwW4P90PPfj{G2>*@uvjuzlMZ
zM79AdT*2f(Z!`%A@V4d#nag=Uf5N2?hb=qM6QW`xhXNlwkO>-qoe6|NOGjt(;ugf+
z$!aU={v#;pKdOVcCi~+uO>97wo!!QiR(8KLI*L%UTXa#O<dL}Fig6*z^Gv<Lfowkl
zNqBntBJ+<Jtk^yL??`hfouja}dnPTfwTm`5wj<UE^Q-C8^~KjlP)$cEcGStRoD9G2
z2WE|}Q4JRnJeG`4r!gK`J)RzoI8*<gKdE-cK0%8*K1P&ZXVkMH!T=sjKJap7KR@gu
z1_4!+7{CRBtYNe2c`$zP-6UBxHa8DZW9SCgNnv3jaI=6QoXW)`i&=ZtWJ%nf^N;{7
zrqFzwFOYiv`00)K^|R+I9b7Wej1FSXePKM6IOF=rg1=YlcpukhKB9S2qY*Tv5+u&9
zdMCvEpov)b3(__@Zv0F?+R{qUkdrG%G^G@2chZb8pnzNR-*r*{HKpW@?`vQ{BiItg
z|M#;SZ?8cm@WVI_v-l0=;%*yXRHJ-j&JU8l*Z~=P2i_Y9gQ+}yeTy?sglWtunb&vb
z&p$Z`$%20z;!+eD#L5~T=2S&-x;vfmFUqk7Dq|P@6&y|9RDjawP4&MSb0>}L=QBKR
z=eM+f2%fgPnZ85ARnF{_K|ecsYAN`fb=8Vo8)_?ud(kY&^+mPj2NO1e%;rP0ONzbv
z=B8eHk*m;%`sfa$3L2IUL@I_c5dTE<t?-;g?lU%n4DXPDYkcB4PLF`cSP-(2Mfgl<
zKYLyC788yF&!DA&*iL28FwB|Md$l2Zhvkd}fK5n(1J{b>^BZ8!uxUk~X*RBUo5LZH
z|8x7FuKB);vu`YOkf|g%)8E(|^7Gb~thoI<2bph__mLx97RhmL`FHOG@m<Q|7c$yi
zuw%q}4|xc7eH!I3IiV*rCI>9xxN)SDoCG8biwRK4)Z;PXki;>q4BY4tO$S1GJw_UW
zTwP{O-ESkzwp3huQt&D)SJ^rTVr@n~SBxMTp{x>1>_6Hh5u_H(^+v^(9URVABgn^t
zXJl<V*@%u5ee=u{kaD_?++Eh%p&Of;Z0zk>W;>biqpZ{3;b(sW=wBd{aj0t|E#T+j
z0YVV)nsluLp$KYM4#=7Xp$)wEtK}DrF%umq%>e;#`3wpR?bj6YQ%vLKsY2vG4VnsH
zaNYD~=RWwWLIcKAO5X^ouD`*)74-9s>KJC&I?~Nd?Aw}PnnY3)qdV$7@{mL~7Lr87
zwTt$2gtY>;&H^zOh+!R(@_9uW+|y=>8diu1f+E&ecfEaCgVW_b=R`|GusLxK`N<Fo
z5dJ{)4X`zBtE!+pR4Kqwo&=i~03870z3kD+)zw&#N0mt%>_im@ETF)`ESimxJqjXd
z_%l_O96Y?@1A2aj+7OpE@0-!w360Ytg~#WlC~~lKci&y<y~n$HEjM$xUQ%Z?yaR!^
zw8;T)UZss2PU1SEF(<F61CsfF&pm!J8dz|@=Z?6R7@KfTrNRBT{dmy$(dc>!X+On6
z;Mzo1PS^Z0;q>bK3rc(!Y|hNOSwdTF^9tRdZ?hlD^%P4?YCJ4f6Ahc5v(;-IPdmcn
z{?oIQ*7j8qk}X9K2aCSQ$m-)w(l8s1DKzj^<?KBjK9V^MB{Ga;18h8{+6n5p(L4K1
zpI4vEOv&l<--Ob9FJ)Q`nij6N<yzfe9G$Er=k7A5)7h$yZR^mK4P#fnIZih?eC#sP
z{iiJ;fF^4lDq6Io?9=DePc7`-y{osp+4EJ77ICj)KS%g`5A`f8igg8F|7xM3A{b4T
zqZ{JnO+U&Du0sLGMq^3L=U`Ucp+U%E=>~Us_4g1IX?6VJL2N0Ljvt}YL5mM((Qs&j
z<0uZSm~}gT(4m{|sdi?%nCUo3u`WYzklk6XtW@ScO!azs$(Ww;`~r!ZpkrR90o@;S
zc{6tTtk3T&G4ZyY-|C$lC^0?yjcwbl@|4Is_ImbF8G73#lq_>rPdVF9toJ@2$)CkN
zt3;80qa-WJzR3#eD)Tk`(Y)MgHjhgJST!`*5!;K-SfG#q8f*aUcp-!1N^5CvgZ<8C
zxY)yM0LpzSh5+(oJXh9<pA@W(0RTQm=y(lyNTrf2K<xosMXXJEu;MQgUx3gDOrOzW
z?UKQ9e+ektI;prgr}fS#0;~vRruv{OB?5zx@2Hui9Z|8oGh>Pv?pnORniDsy{LRkp
zA8<`MJw8*Qdvp!?td)obkkodpbg}b)SZHZ&+`=v?7d>-&dJ;N4EIHk@THB;}fc2<(
zD;|9HkzE-57qh|%>~g>vp^{SaU%jSc8f%f~hYO2wcI*gtPEH9hm>3nlKpij#zjQ;O
ztpYe<(1Czr*K0GHul`SQi6^iSf!XH2>No0yfeMh~3zov5&jME^h|U7n44Auvq@{S8
zvknZuio>PqdTvbExEf|i1=R<`+A%D&jOoMm_xAB>48HEs&9L9_e&RMY1>EgD@RXES
zo8rA;`S-)h#xg$G`SRJ=LZqq>CVnFnuXMRUbnp&KO3faA;A@zXtNOudbthI%2ayyu
ztUaXL(w|J*eblMXWytCn%14Oft4Y$5-GDFF5ou#bwRH5ZknC-8GWuEfQy^_!k%ovp
z#*CByKbeKOxNwY7R1_tv{l@NLR`iYIDr)7+w2aMs8mgIHT|35ZjPEA4saXVW>y}$C
zl?(U<;}~KGVri-$Mt3n-9D@5gYQqo>wSV;b2y(s)d$syR^?Z2<%ixC_*}maKsY4TK
zbhzDaX0%Z*;@`3RDVp_jAxb~sZLnS*78+TC;YjtKm$Wn-s8c7UZP88@Ep0(Z3odG-
zLra#Z*#lZ!<d>Xep#29c(iiTa-<=H@no-=eBM=f1y|`Ui;<dH3zK!l;!9r_m13v~B
z-hv0MV%{E5w@A?ufDOj#sIg|xJu~Q6z?-~p%v+2qcJ<f}-G%+R4_`_e;1SSaV6Ahb
z|B~~ySr4ky^Q?Iu`fu8{VViB=rzm((dmf_7x)M7k@T^tR*pKd<u}+G01t2_QJG8>*
zZ=Cj&cP84ctz@1)(yvLLAgE`!`7wTz3-q|<)X=XkqWMQ#dTmiG?AOuhfKSIZ!<z##
z7=u@_D8<YQJQGljcP4at!@a*e;^ceOmfe;-Rx@2aJkG`>F@78f0$GT2L;tq(<>lqK
z0rQt`78vy{uHTDh%W7+bBE3ipSvLtxi(Wb!K;N*nwFPN{z<7@f#Ptf92ESSqzx7ev
zl7C$MWT}1Bke=uw%D7dvXDQF0WI+TN2^3^>nUmJ`g+FuB=V(9_7&Ql%B#@nCDXo7K
zcO&=TbZ2Tlzl&OGy$If|r{6`I%*!nyam|z)3tv9!+OObZ=U1Y{q#@7F=lvk*?@tl;
zjRH+5^pYDg97wZx+Q=3vAoc%F_W%1s!JwyLpCZCttX`QM)hwU(x|)hX*ca{{Xm;0>
zJ@q)N=B1<gUS+$e8(aiGX%Sl4xzuxs(t~C)A3@Evq0!^r`Iz<1nR)3{X5U183d0Tr
zyP1t|JLRb?UmhGL-O*|32}cVAQb-G|_Y`-^yl_c8Wc3<&g-<G)=(J>ib#<9_$=;5X
zAjq>6Iwk$azfIm1w7&R#AjF#9sfU2Ab)!}L3mwySvA*gr?qzmSvgc(e()X}*uhTZa
z474XRKDD{icB9iolC0e!qX$7B3jd&2e4it02n4*wne5F>oKiV}4+A(749k=6?!0Rq
zmIp6J#Z-atGgT>mbQkD;KrXsi4&alk<AQZ07^y=T6a-i+YHMo`-=T(gU8-K<M8b1$
z<IH|VX>#cMLWJlQsLvTriToS0AdJ}ers!vk(J4$gf45#q84Z0eG}7!ERH*_L6}HM3
z>*?wGRA^(3a8>j!3L4iZ`{{|tNi@$5X=|I1Yx}TB$APn}>s7ePfNO`$)mkdp%+G6T
zidX$(JaVG}9#4{w9kE?UvmJLr>-(R?IeCmmm2%8cIhDsQHHNvG$Y!|3HAP58xRuR*
z^|!X<enp+1UtBk*GgPhjenrAZ{xV$wFe9L(1K{7*R?*(RzJ{aPMKIk2Pa%ywu-PRW
zQD|$=H+ylB@Hyl`ATY^8JpkKQR9Fb2YjLF;wy4Ea3F;<8sf+{#y6LI26{`<^CKE;N
zIxVb8QZ%O1`&TB1nCu!(kMLy){{G}=Oj~}*e6#7pI0L)CPu|n!EK(bgxl?hpMK2I`
zrB^0`AtE3~7&ap0vq~v2p*pVJ^QLY|Np+XKNhoZ7sx>JqOU1Ut_wa<&)WAC^W}IdH
zkD!L82n{E3-O5c?)G{2(|9qk**D~X5j&F|v-rs`(+uvo;@>s78rfu`GS5gXzNoQLv
z2y?Yyts|DIxoMfWm}(sN?9{&R?BHD^%){-UYxQ5OVjbQ!^qmxS`a?QChP%^6;^$%R
z3cK1;ah4w6?hoGOro)FPaQq!s`Zj_5CFpInP`~V-*$vIZQrBBYg`ZB1ar%)bPY+O`
zYH6QV{D%irr>5m))lIrGDGVei^~SQ&4(hM(=;!WFT3ll~7otd6%p*W8l}rTa&ESmu
zb8xV~zyCsm1ZvKI{Xl+qliBe~bfBLDu2*o&1wk0V1czH0RIwlyV0?-yQdVZtO-ejR
z=9f5`3y?)I%pqGt!fev!9^JK)$+~hkyG9d3oYbovJ_kL2jSQ~mk47UA4?0ctXW`-I
zK+1xWj)X;oqJ&l_B3EtCu5zW!r|Ar)2v)XRn^6sB5f?<^g*$FTb9irjjuWC`Q1w(`
zWV+q2YW#S0dYZgep2hl!TJfh`PnMNJO&@$;j$O_cja^}?SPWHHFVD9D@e^a4+new0
zb9;#47$?Bf6+%+x6B%jzEtwAfea)Q=WS#}|umL9Dn7_EXwFa9IHsAdzv0to31M}P;
zYH1B1P7zZ&E-j7tCFc{k1jPZ<1V|;6=mdZ%5>Np^5UBHV{ZQzELyO(`MZ^0}Gcp&H
zO_W$X<lUq<Uwz;FN?G)1z*U7l<IeDEEHm?{Z#YZRivUcIbD-PHEa3ya5^~kh5XDN{
zD|JjyRm~VJ+gFCkzZd?vr8F?w8)7Nz-N9R`d^e12bj*A3H2K^;>4ZnFV|I=!uuSY3
z#Nf1RK=I+9(tJ*d#f2-Ve|{CqBCPl?^Fl#}>0U|Z{_|<Ew|$K8u962S)@EpXC-0yC
zoi^$Z;JLNlT+|KVyZN0g@X)WC6q(8CX33NNC^sq$93S~VEx>MBntSol(6(x(kBlX!
z?{>qhDbGx9>5->q&3$3jLy8o(KiopyXX6XZbHs%rKi=~8*{X(&mM-<Wt>X#v_<h$Y
zfX7F>poxSd|0C93Hb>dE{14J>P{e)>I!4=!CmdIg@I6k#L){~rF^tcK8LCBsC!P?R
z3fqA%)^SX&#;?YPTyH#o*Gr<O=g%!m(}uCWK~4};R<s+W8&y%F6^heWn|Utu%a_e&
z|2@cHJt86!%`yb`u3+2-LZv_r251%IZeS$&LahPH4d4(v1Ntb8g2jksi@v&rC=ge9
zJVI%yV5&rw3CM*~1z(0OCF5yun}YDi#;T-cFk(64vT{P-ZNX76Goi~6%1Gn~!Kr)a
zLrZ2>cq7=*sWDE^7OSs0a+Q)gI|53hOS_x?vlN!-Mcl+@mem?XE5c|CI)*<Hjm->4
z?Wwn%n)Zpw>6h>Y=1)QS(U=qn{yh*g8D_;;kmxLjPX#?!L$aW!s2=5F^g#uW*lKz3
zQ!(d#(}Vd1FREsa8iR4&qRj^^8jz5(ygWSq!~tGf#<XdM70hDEoZJgq-zUt(#Kf=B
z(ctNIY+44WaG++91>i^kpr8f`KHAzn(>g}OFE<TqGIjQ-k-JJ&Xx^mF3l<g%v}U>w
z@iY-&Ccf+dVgMlr-nuMP|AsRf`rTT9u=3rD#0^R5xTg=$y73Hz#H{l%U?jwp+5LK;
zo>XNCGDxv>-p-hQNdX`sq$&t{yr0BzrLnoV2*WzIx}|J_*Ef&Oqli8@+f=^aO0!cD
zrKV3FI_sy7H710IFeJg1hMpiza&w@*0z!URXoFcWc_lT4vF9YxEMTHm@86Acp=;RJ
zl6QQiAFXYE*GGU9gCTlD$-7uI`q%q%T$Teh7!X)G*KY-F!72@4CoHZ>_D&cd7wddW
z#c8mA{Q7nDebMy4-SF9vq-eI-(SQg+5D=Pd2W6H2{Mo}4{_o#(8Zd^*9bkzi+q1IL
z`#$sK<eyk9#|6hbZa-onDFwFER6L!{congqyIOL#UUVoIW(1Y4S+c9gn;zvxJ}OdG
zOy$9C^={0nC*%das8ti%oam(&?>0oD1#N#FvxDDs?(FLCYy1tga#?Z-yxbueHX1oA
z>)X^<Q9T1+mgT#T%@c**b{`uTne+xIOJ|Xxx(y&ZtE~S#<~V+@7w#Lh_@?KiuLn8|
zeq+oVhM7p(yD2OH@d@XndtikWCrZ>1s(d`yQ2`b1OHi+sl~qHG_RDMv_;L^p3=H!A
z!A0_csd@KPP%E0whES4DVIr(FamlobMH5W;hPF+CExPeeFh3Q(BagrIbw7;Tk=qpG
zRTKc@(;$)ZJFe+x;E`qrBn_ve?Cte!BZ^i2f>-VHL)5Rd7i&9{D?*<SrU<{b4>TvH
z>iAjbR;H^uf}f)Xo6d&1`DmPGegPxqOKeLOMP~ggnw=e4h_~|+mGp?Yhnf7IuWt5d
zC6i!eoL$b(kFPc-#SMM;u?`jF3tw&}09!aaa{}g6rO$~nM$ecn@_`X4@{R8$M;wg@
z<(p&pgVeAW1O){NNlCX7bqv>c(m&DRB4b6%fTY2u$KNQitt6oP0!KS<dJp^8s=|^3
z!63>U8(M@-{X=uQgbvGUAZ7@n{GLq^Db4<xCZxYfis>V{U>`gqOb{a_RopBpm0zDm
zG-V=N4g;@uQw)_naNth9Hv*Fm(tZpV9}5dFOqp4?_<=3mE1wZ{Po~*Ec2v|Zh7+#c
zGHlpDu;B*Xg&HoL^5me~B$)L4D3zL_kDuZER+~VU*_*#md7shULu!NSD;=yw7p?qG
z?a|>pv~wb91P27sda-(adp2~z1FP@tZ7>4}GNAvj?uarXm-vBM`QBc)?&(%=x9D_`
zx%?EIZYRQZmK=2}-pqB&?u&3D1^>;*2RvOEdI|iWc>TRM;opU-mK!!n#?||GAshZ7
z+db<-*jMSRh``25PamyWaUEH>ISS<S@+wZIE;^fBxyJCjB=+~}!?&kb`=;~?J3P~#
z`JZu(Cgk~V6;W37qlYmVo8QczESK;1<$7+Ka@mq+_wxsg|8{S(>xA2?)|2OaaJxVl
z*7NdQgYv&Q`E+_e`^zkqNnorwt|n`D{wlZpqI+<4qc%}-Dsz_0RrG?V^-6=(_LHUL
z){QbNcaoeXsF>9a42DNCg>L6Z{eb*#=eX&_a<<ICeyKjr`s4TPr-y=&&n9Ip`^4Vp
zf5)l|2;P)C<y|9M4(7F%ksHj;S6g9#a{)Lafp;4Wq^!CJqR7}Pc&lPclU@n+s}SXj
zW8=$0<;(vtqohpAG=;L3mQxYWiDxvGvLv(A6!etpPt8%F+JuW%NTSI0-xD*|^BLHm
zUeR|>XnBT}$uN%27Gb+~>Zw7Kixb>i=(0;7xfy*{mWnrC%^h3fZ~JV3cPQ{)&~FUG
z2P0YEQru{u68;;Au<6_6=DE|IJbZnaZa!-l43;Tn^DAFw4}dBNWVnFfWgx8rgOM|S
z&KnVdKTIy3hhE-yno1V$at?@FQ)D(32tdfy{=N;ECV+YdWM%?&z~)&7d(w&XTTEE}
zviK-R71HW?doX8oR~sJ2O`<^p4`CT6P>=M!=B()q6;7ZPjx+lT6Qq|jLqJoq%}8a6
zm_zkVp+I3W4ZnGuP6pw2J-rLph4|#7)3;#)hMftm*(vtSG?Vxn>3SyGoscPk*$oL$
z(S!e2pbT_hAPNJ%UEoFdR#v5ntF$ERoQG@^c?%On6KEsp-td0^l$|(s-IX7BkmVU;
z`%a)PZdN4>*jn&|7y?ax3yz}N|D8CwlXBQ`#U}fIQw^4p6a;e5fS4FsWS<Ori63J&
zLN(^Pq9?3;UF^U&3-D$a-sJzj5>Z^Qb*66XK8wJT0-eF8Hx=vEme8A=-UD;Rk14xN
zGG2E1-zNZE-mj;lfVeKZW^O7qZR={^@9!|G84CxbvPn!voX^upaTv1Q02v|3Uh3jA
z+BfFR+}&JEw?S>RLG@(FgNvNM<w$(eIkcEfLO~ko_$6*c7egZl1|zv&qWUap#dCVs
zMGsv1gXC)dP^Hdzxj4pawbnoM{=q+f9E1Ps>%xZd*w!`>Ye#HTRMK9~WQ;4{*SCLk
zWKM-7hvo3pKP)xc7-}`TEiz$ia;7N9g1;FjOWyIH1Qo#bf{KfMkL--PC{8!8yi$*|
zcM>UD6sY3bEvq>|iC%U9<UfYKXVPU&$AX{;0>QLE%S$5<rnBrhJ(CKybkcc@o!aT{
zr^EiY_`@Y>1k+d(W9*g&rY4<HB>Xcd<2&b?PJc<(H-jCcNLu+`8Nj{gVmi;HntFp{
za(|=K!6QV}VHuoCL)d%fP|m5sX7>t#h`gkr>_b3Ujs5}IHirMNFbWYgvX&@m&H%I$
z#UVYN_TK4lgWt3Grju=RTCQgFUMO225X08a&Z%ix&YM1T3bx5Z^(@Pd7bFh{)5w<^
zwPgX30no&N_f$R}IP+hynEU(LYtG7Mgmhu7-<0x^5G18?c@TrzK_;~d3w~~XKD=&{
z>ig>b<&n+>)3kh$d;+9Uz;rsfPm?|j^F3255(0jgZ(~XfPbk<U5f?BTHI7yA6OvY{
zY~t%3^p-LxbufM+atjViTi^)|zo={*M=1Bphon?ja(P@VDXb<Iv+6|#OI1fM$E?&S
zbyt1`^m?!Y9<~Hp9Iy`qV1YyKbds;BCF|S$gnI+9(EeXkydvd(OS%h&Qpw|nlAgtJ
zPl`q>ywa+}`Dh;5{B3;>#<E90(jN@wv8p&`>dMWYJE*P8*b_|%C|WMuO*J)}ImQDU
z+W&m|e|k2jRuIl`I9Iu8%C{Nv1>(7a^O{?^DaS$ZzbAjPhkLyQNMWv?<bza%KH?cZ
z?fiWsSMc8F;kwMA%S~B(>0)`8P<6Y1U-~aySPGMQ;k(#a38e!#KSB2t0<-09i?*5#
z+vdo@A6tZC<nu;`el$ECu1>euSpGMgvrv0a4=vjkP51P-f9*04|BzFsvYkKjHF|EI
zazED};fLfr^2Z(eI(iHf8)lPB1xQy$pv&*5e(-Sg_~$AkLl$3JOC4gIWV=Gawd#u%
zM?$f6@i+YDu1!>S*OA^QTqt%?rf8|%fZ`t4qn2%*y9x0<RcdIDMHoZm5-`;Qi@?Qc
z7ltOt79QH}JaPv{#iNf$psob;IdPC#4Y)KwVVb-wRK-&5abgJOgIxf?1!0IEwoFY+
zqasD06q!@r#Eah25Qrh)gVg*?1mPi1!Ag>5k&HBXn{b>4uL=<ICS+ryZy(m~KcCJf
zI5J>J<4mkDrJ>_{RqC4D1KoO=S<>|ETS2{Iidxdz9rwI58ci<4AW@-s>vJ(tf#GLQ
z_P2B>BEe&ra87?U4Ka<iMJ`d>BLy9WOx!f(8C(P?jToet1W03?j|zZRZ+v=M8UUKB
zs^UR$08Y=~;9$VwIDf(DZ6&C_jDY$!fes9~KOE|CBBVf`3a&HG*Vv)UOV<l<%>&u(
zpp$oTv0;1-{s^9nTZPi36fV#Rs?IC(f*uI$j*>vd4i-q~x4sRAMBiHrd!zydm&4SQ
za+>T(X?lrwk*RvvE6uUnhQ<5KAW<+XZV~m;u!mJpRZ$h*L3~rv;PwClC)Q*Xu)`g;
z1kJ*$@6VYyV>=L{U8yRiA8##L0-Hu8`QcgXNWwQL9VW+UU)4}<kSHYZDWgjEwM)Ha
zI1vknG&Wq%*rm$nM_9QzwEV<Zo*lC^dgb4)>M$26{g0;ygx+8ae`f6e+S7W}{^$?a
zd|>{UjNTM(3U#SeP;W9#sdqr7Rj+!?;=+Ip9ggPYo+0t}Egl{$bbb^O=iNh|C05|L
zZ#)~>XNPeP!^~r$T;6-U#E>Mb?QBusCBZuTT$w916OS`ClH+z0hCKSwJ?;YK;K@;p
z+NB}$*Ig^Aap&j5fwl|6KhA3e?=NmV{il)ruut-QID3ePjc@nGkjL%BMs~{^lYdYn
z>>ZMs3L8M#*TY~zm}jfpe`Z_wViCY8(o|Q}{{|-&n7p2Xh!GH!ELSiI5`lpHIf&8`
zB%=cLNKH)*_>Ka&dp#I6%mctwV_C8RL*Uy05ZeGIfjm4sFFQ2UpaU4CD^+!@Vq<_<
zjEF!=OHZ$yR|wL{;K!yiOk%L~+H!ku!1Tn^h59EbRx=hB;hYksq8HtO7a`2#)H9w|
zy2{K(58B|7_IkEz?r@4<c5l=Z=`O$#mkz$2tbIS=&eqe;f1l7LA}k(uoN3O3aHcLX
zsRm?Jh)cVoTt(|tG}aJ<>VP+Wo6m2P&vPf$3JgNV$JN2svZ$;~CdmSjae!)!bv9xV
z#Hp!8;%4}K20-VMk|-eW1%ZjFsi__NuE1#kLL;MP*zW74T{@<c1<ve7M6_A($GM|{
z0ET5~`=&zxD{dayLni-=O`04!iMEPZWf74ps=hm(w;Ac*9qXb(N#W%nz4r0cb(g8h
z$v$?4tK*&SZcpKuu62e!wZSSS&Qz8@?GZqrHQ3?4ihjMq`n6hAzvFZd(-vOe0YOGs
z+|-3y;dNBfZ;i#;$d2Gl+{hm3@;=fUwCD`snuX4d<-V!1Bl^|;e-|cTWn$3Vop91L
zr~EQ3;g88pKq9}HH9Fltdh;+cZTIC)MHK30^YJr{)#ek}n=t7R>VESyJ;ZF<4BkG%
zRfjCJcF%1QjUA$&1s1;}rx1~_vW=`+L5v=G>-9NzQZ<3OUqJ^d+_m|V>s#jfThomD
zx>6}FCBa_HIX*(O=z<{lg6E8XDuqOPV#6@1Xfd}lQ#t%N1m3QU-6a08JTA2nWsN*j
zn!_%Xk;>at%TiJi3zXxH+M;Q+L>cGN8@XHBO&{{fo9^65{`Q`w7lo}kptu;+0^k7z
zD%lre<=@3mAj=OF1oo?~nPB$%($oX$fHY9c%@1qo>dJ#60?3($Enj-3)e&P;OqYc`
zaIz}V1q1U3DZ1th-LSYj!mKYQ<css6Z@?w$R863ovv^NxeqitUpH~8fIynVUk8pz^
zyBz1#3Xk^eP>Ml;5nY|{?S~O-+JSFh)U|aKWn<fm;W1xnvwOnYz5~l)owDGLI1Mt+
zYyFH|jYxe`^6<2IB+n{Tl3BsjhDV2`70lEne#uM(EZD%8APn%X0P%(O$B)}tKF2=*
z-@g3wk=+Y60Vo&&)(t=x=Z@Tgm?l&L8W2Lj2@4d9QdHvL=k~8=7T~}D9c6BA&S@lT
zC|ZDie#tC_U<53RK#aRN0B}d?TZL8fFMM(!iws~Gz*WjT?tm_fsKwH>AVzVJiRTEJ
zR4U`<AY#s_-`l@!=Z->@VEiz_jtcNmFUi&_1=TE=`5KH5e*W_x38Q*W?{`1dexU!@
z{Ohang%F|0LO3xRHJ;|22<wJ;=;_<E?<SDC5JLVXKvzWR?wib2W62?Uw*={F3jd{7
zXLv~W{(S})m_K?|rW?N$0ZtJwskw&!k5^I4s3<50Pu-@v6L8Vb7nTH)T)M}rWBq6)
z#ZF97v{CC)VSN~A8vE~vly}%I`@Wz~=5dJ(c)+_~$TVk{x!RBWRXLv1&}qjPRW!_z
zm*!5ix;~bHHA(5U_V^aEGY!u`(%bL69(m1%xFP2a+><#%2E`bHNve|xnWh&X#1yn6
z_voVa)pt{8!*3P|R=2{8ar9u+%aCU_lRheD-D~L{6|E&ir>C-73Vb6(Si361`rF&L
zKT2~GoK*ds&iCh3Ew_;|jgIvU_p`Bg+=hH{62qUK>+ib%UHZe8Ad@Z~1jerb1ku@H
z_2`ECi(n$JpdcJ{I$>d9o?CIUz;yz?az4y1JJ|MY2R|5Orz3_+qQ;<^>KxTiI>gg}
zn))S72h6j;nCkj?`5%Xm7Ca^Lae|_jSb!ODMa%py3$#8Pr4H_0Pr~+YZzlq*s*qz+
z39p{g5|GW+CZ=ZUB%awz`Qv(in!KqUFeJ9q!BmwPD|qXMLzCZ5A&#ps9=-f(O>GyB
zZ`v@@nUpZF@y|pRU}Ul`D+;@(v|>dl33mOE6a)WpK4@!u`>he76LWyI5&k=81_)%-
zh!PDmdSEw)+vBWU1WwqOgVhdbmWE#9xp>p1lPr`b^ML##H5E+Vn@X3>`vjABAOy;j
zprqB%(b2*0*?}<ONv9I@T3!PJ?J!;ZfW?`Y@KUooisWQMUNcq4G0XHMjEb!7E#@{Y
zq3BN7frQhgH9ffzY<6F?O&XAL0e!urprV3<Xf}wG-~g2a5WY@wgIQ5peqwnPr97f2
zLXl~fZMu?^G0boYwsY~99#YNR1g+9S=?oRh5_DK0nuGz`Omq5OrN5OG^`0;FX?(er
z*~`<)s3ttm&?gO&cs7GFnZY;)Ih1womW5>m{(!s>9JkBuT6CjrM+YX)d))pFnp?d)
zw%_NW8{*89#QZHO#8y^gM@@1INw#VxQq;z$ytqW$)6W8FV@opg-Ze5H+MmhZIv(MH
z;o6#}3hAw<a$%sI1RDL)4oS<WD!byTtW?vO_Z-TbVQ(IGrhg$)&3&Yrt$la)R8uoY
zZ?84K%&?nN8Yx>_SzVRNw;{QNV=&}`BZ6vf#Xu<dTyR3>9s<$Fi_E(60u>k?Vzq$E
z5T;mr?(-e{=+Th{l^18IJqm{3g?Md7%lJ`gy|1k98x}u#5J!{;hO~t6Ku2+PFbm35
zFmgi-3RpbK2kAK4CGXB=^gaQ)6^JwdjXszI0f-z}Z~+-zVObenqo9J75q-_m2Pt)+
zoeKrpejq@0<R=9%Lwzg)^a1gt!wH!AEzFSOAyqeSE&SP_KMd_9)mI#ghmM;4@~u`B
z{?)&BzTf>?F!87NRDCC*@U9(rn|=~oLJ<98QNz$92t5`3PkCoL_oU<*P<}U7i?EJe
zPMjCZ?MHQNp1o*X(J4f~f>gqpI-D~7;J#Gpc$&Y5^IZVfJlFbJsJ!h-5DyRUh4M|}
zP#*~BHJqU19JXZcgO7=h2Gg0yPg8DW;a%$ZJ>aooT{K+^#~{luYVaCxbGcfHM1ef-
zz*_^*2>aLbb2B(HT>_5rl*qn_(ce%oz{3|LDU~u}@`V?6ZS3hY13F(xTAIq73QkJD
zQnt3yXh^XFiELBjS&f0Z`{+Q`SDM6QDsE-!e=Ujl2IbC8$~eDP`rtj_9yfDg{}}L_
zhjw|RDyvZ0VP_3e3JO9dWf%?%p4Zedcgpu$jf9o`JBATfJc|9z^(0dN8;@;=<l7`~
zwe0DDOhpu?cxc~s6r+j|&i|S#m)}3mi2TO6dbOjmeB+~=^Dk<2Vox7MM9d?JE0vxy
zvQ(vvLz&xSjL9IB?7IEmW>AE=Pk9lrH~;ZrZlG^;kNYyccB;vZDswKcq_h!Aih*%d
zz=uP>SM^!|t@UaDBUJM|Uq^EG^5|lhCa&wNbe7l1FkvRxio4eFK{xh)h!SCRk+n7E
z$P3Ma%hQBPbS~x}#Wt<I0Z-S<$vNOn;2#&N{!Gl4<nZS^Md1Yx8<o-z=>OHEK+>?}
zGAi=YTP&aWC4(3e&@BT*ug8j4!C^m0ssu{40@6xPz_Q+S<Oh>uK=~FH#W<EBC>p@M
z>hM;Y1RCd;oC#pj0b*ZpxmJC^&+e0zQDQR)M}5l{1c<F^G@%JzJc@klEh=-%usUvC
zpJ3PBF)lIEr-YIO*Ur9BZJn=O1lV*U;cVav8Efd!jtRkv@*#FGrR*VeVL(C{zC-Zt
z!^CG_3zP7Wo_4-R-H3?yWCy07*5k&kpPVTs)7pFx5~lerb@I=b>R`-!v$QMPr>$eq
zgW-4j3fSv_02`>%!1WI-F`&1qH0?nKB>*U>!M+m2_;nZqa$G0q2!TH5%-28FfCg7v
z9G9W*{>@hneZpJ%NK?kANe81Xx>st#tiMyu=iPx6^n<W8(^gLBHUf^mO5#c-ZkkOF
z%N$?UZ%T8XcbV_0U$wWt<wKKVU9#igRi`10NGb;n+iUXJ+ylXvDj*UisC{pb0SJ7+
zkYHIf8<VU*kZEx4JJjZ1n0YO%W$q#;ggAyO)l5qJ{w?#kr4EPRW`5ze#gAXJWCUt6
zo{k$|SN*rMd!Gi><K&)e(0~lUIwz2xoGM&{Tls!q-~V+y{>~r1K2wgD@ASVlYae0$
zfLN<9B2I!>lMQt|Llx(qxL5@Bq5U^lEOku6{*lwru2C0$0dW~NNGjMR;|Pm-n~(o(
z(^!^KU_R|MXG^m&HjSU=!PQ;sd)}w3EiY_8V34MFC`79)@%ajIsBl#AC@bWO8##Me
zb7-AxOD3UWKJ9mZD!fgzSS(6~Hxj*bzP$yd16haQ;l{&b?As!hbmi}-gYFsCSL3de
zYvb!G#o7cEcGhADI!CaipR)`;OMh=Xxalrm?U_l$${?7hb!Ymo_3ypAIw&eB+3X=O
z%4=vyB@=S5@VRza@jiG{*7oELu6`gW22jUgzE!2|XN!}O%`K+>hgE-Ip(|GTWnj>9
zqSsjM`Zuo{lTw@#(L`3fC;3$jC)c07(OZGv`h9UGQ|iqlrzFWlS210l<qTAj3^@T+
z2y!s;^uVjzQa(McZE){idoTMlDg+i=q7!@c{t8*Xc;p?u_VVy;)~;7^F~4i!DVz}|
z?BLLl)BQ3xLRw|dM?`XHo*dBnRKEWT&VSIigDOS4M5ofnkx!*50l5J?0>^hceNQYu
zQc1GwZyKA3k4wcEv^(@u3p}e}e71bm*P;W&iK5Ruc(8>m!NIR8PNO@~ji6-EB{n5&
znN2>iv7};rpJ_M7_~#89>w*?aHOC%F*k5YAX>AL3B8Zy6I~^eO!Fz%D;!3|n-z^@|
z;g42b(elxnd1>(scVUPrfp{GdImY@f61*NJ)$`zue*u-jq2RAM@)wqtngcV;#KeSY
zO%^5M_oO6K!d?M^gu^pgiuF4ZvB*Umv3!e-aq(2}FHT_8+Oyy!0tj0`;}geZXCGhi
zmCB`SeKO~u!MlJ*dnAH<beD?FtQ>eP80xH1S!5PJ)qD5B8{=4{%IK8=9~Lc=@-P8D
zChVc}dU#ax4=zq2c|EVLMR&hJrUQ`FjiJ504X9vX7g<z<q^%81!64Thx44J<*PQ$K
z?gg0(>vAMQSAaFEWbas|vSp$BP&fAEm$aC@>B{hp?{BOrZCe-KRS`;fGa*Xfsoth2
zm&E@D><5-4mtLPA2I*bv{0DsBEjUv*E!RJ=;79eC3b=3<kJM$&%SO-kTq&k`T2j8Z
zFZay1$?p*95^AvW+&vJ@9y9!9SRDZ)_H<uH?Z4<aaWoYF<p@?Rz0{h)?0(h2<+s_a
z@3YfN-BFv|Hj2wo3>~6zGR=p<X5Xtn)2=f~=X@B={m<Ta>Y|IaS$f1elv!&Z)~}O|
zhPBz275TI2;zuh)I0yQO2@+;_&Sgv}su(tE=VMDOxIlHsjTNGqe#9%AnMD^F#WTA)
znE%(qZ)UEjN`f<Lo79XDo{NhM<PTJNo|%Gv#=^p)2e}GN#RwKB@rtuuU!DdNLLjcx
zcEMvX4ndQWm#RPlPBw-}AjSv!b_Pbq4qPP(8N}>KSsHwqF<3u9iYoCz8T~&k0Fak~
zOc&$bkxZ?d&kW=8oZHG^y$;q6-fyXMlE)_|025h)4^?>AAPAp`8au*~l6xq?Sc58P
zb<<?4sGuV8J39}LGzi5^RCpap1!f`ODgo>3dDbax4S$J4<CyTQ3l5z4LLJKvc5!ZT
ztafNmXrc;nFKbxUUHCsWGQISshbuPpYNn9g>Ja}6!p(h&s4j3cDkv%vks_b}^CmzU
zk%I&Y_zK=v+?eXCL}+m;0ReLQFeeaH|2qYqSYTMcDDB+-X!_KQZEu<ciWq2MLjo_v
z3vB@W>H$w5NbtZ#j#DW>f)DNsu>gL!FAyLIpiWfyad0QU0(T;^YgCUE*O}QcZ(sYK
z%<WPdobN`>l5^Y6aYNgY)qSwA*te)?ZdZA|_v2ONVS2WPxp{1Xwtrmj6y;<k31!wv
z9`%_!<Tw>0>eOmeB4uyup6e;B41_4!>O<wIl&MlJ&Wg1OPsSSf&b>%+MeEMvD$45W
z2&6$mFEU=nx??^6Nm*$rs<t-BeVaQ96-w#751-$T!D3+kF?0E?q|Ssux;gBNZGyF;
zMdAB*J!O^9gBj?T=EMe`H8zmKpN>t)i8k|ELGOhlmbZJxsvNggiy_6t%VXop!N!(J
zaG&l|tKR?nGJqEayc6e!*KHIl9Y<|Gp*<G|Hl=UO%r`PqvBU}(*VaE@sD7mRJ#MS0
zF9!)=Xgxza!Fb%ZtSTOfHRSJp_Nodm@`RrZ`Op?coE=5-j{9Oh3P0g@(P4VV&#CQp
zhlDv|auYKzu8+s<PK*CUVV{&(uP5W&b0{U`l~wTwk6a^oRElsSwO7R}JlRd(GsfX5
zQO1w5r;lR){yB%A+RnzT@3}>HhCMId&~?>2{Q8Jf&fy{OTIDVLJ2}hB3>UB%Wovuj
zjHdzL7k1QK{z+uyA1{UuAW7~S#F184(@u;Rq*#4C_OLE<c+E`)?0%pgZ!Dt*n_V#K
zpC98!^H0bQn}83Kl*|hzV%7CwT)iFfNXg2woXl4S>-EZtPQ~l@fK`fqcX!9{d&B;Q
zIlZ!gw8uS|jJfGX%I%H6uLZ)ZSFgaPL#9e_HfvUz60<#Pxaz1K8u+bcz@wQx3Vfz7
z8M}%}EFihDH4}4A><Z755&iQpJC`P6{}c^BTp@JETNHL?7~Tv>cTbFA(q);IxjAf;
z71BpEm{+gD<<QBu29n(|M{Hk$hJc8b-2cIem{@PXnnx<%FfTtp%>Opo+wU%lfRMs3
zt|;O)m^3@okz|j`0<t-Abi5o519M4Ul{*r;70*?o6Swkq7sV6%3pg%OfkRMm5R@>H
z5cYrDLeu?;pQz~Q8oIgzAVwO<7AGb&z$*2{D#n_8km-|<2^aqND8D$LoZ+7`B^qpo
zNM|KlTs73sJ3A)7McAO8zRCCF(=P6HPbT*rSG@ktKQQf06cc7MwM5egK~zJHjyj+Z
z{vxX|^$|UE)c=mUTSGw2{(FN0V+nT!p53N0`smlzUy0_Blnz(iPIiM?P`XZmijP@1
zQ-vN>TsLU|IFLU4|55ec@m%lk|F}&EmA$jGlD$<3*&*4Hi0mypp^&}z$V_DKy*E)-
z_TDQi>vz4*`MiJM_xm5`oSSjHyq=H8bzS%CKIj7JoNAMSRIQ~b^x7%B!;B(2THh)E
z(S?8Vl|`WUl&>-<W&_#gq>t9d>68+Z#Ca7cn--_l#`W}(I-WdnJy>6T!Byn5(?Et6
z8oGgvPg7Ty0F-luJI4w%CX4S21za=`rQ+Yyf%pfr2`N)xYlo(FqmS3$N=!2H#AdFU
zltR7?+Bpql%8JL%Uw40(`6yEHF>7tQR9fp}c*~2s#<EMtoG$sJ7g!c9#nxiL;R`J^
z%c*FSu9mr*vo8AmZJ5`^cJGHsA-jZu-eFtu+SOZD){Wj-j2clrMFZc$r9WMZS)(@M
z9EXn{Y$YWx9PX1vXg5^#wMQ#1-SkEM#O0gEqO62}_{{iyQu-au!>)`bFZ^xfw6A6U
z@?R6ooT62njf1=2c7LPd4Vn&C&Pu<mx8MGO-3_%dn3j;Wab>a)Qa^~;lz|Ej@O3w1
ztPSMc{IiP~85(Lbes?>OrL?9rC_FrTp#rAWf`Wn<+h{}IbA4o4T4deP9G_>xB_$Vu
zV7>!+-^9r&hK;kvd)4!V;^_q+z~bZ-6o??-vGSKp3XmfL_bhH`=*Y}?6WgGe@5uVn
z#o=H;kFd${E<~e>`ZvS-406PE*{OCII)D87)dAT;FLHU!&+Z1B(6V~tx!8SA!ytb)
zm#Ok7hU8@*M*7n(-N!Q=DiQGr3>{p4N{##!c3<A{z{Hz#5H|aY&Ba5AXmnIqq}Yk+
zy-sPPHOzR6snICZ)b4!fsN|-Ie4ubcl@QbjRnt30E~;_KS+{X$nzrLtRPu=fzU}qP
zsuB|4*Y7)9O&x^c9WfjO$O=y#QzRT@h#FQFqy*>cG!|^UfJ_(a0mhr(aPIB9cCVJS
zpJW-5j$81<x@OD~HrWqKY5FqBg%iEQ)QUeTIW#{cW;9ad%y>u4$o5Qw`Xz&~<k!K^
z<nqKZ68VH;D$}z1_oG7D0&%2~cW*q|`V{fOtmUQ4g++4Uoga!K6Y7L4>OnQSg$u5*
zSjBp;;%?A~z8IK6*xHt#&)QL@CEWJRfxi3YE>iSQ<fb%rby7u{7jN#%*iCkKEC}L#
zxVL{!MorV^v_zdwQ8@lFpJc%Gt&E5x$2SHwvS0dX-e$yKufU-CCGnG2&6P)#TK`oi
zjD(ya(vl^KcJ5?v`Pz{KU$n$`)%nlu&pERP6+{7j8(ADtnAifS)|)$~+%Kwj7QV~r
zR@TK!|Mbjv_Zu?PctvJ)o*4S<LS`j0nYC8Y*ilXBYhq8R8)r=S;pvqvQ$}aQRk1?S
z)QMOJbE4+qr)-T*(bXtpPL{U|&duU>{z?Oa6_P)gbz=5uE^DE(n3_6lRC%o2j9k|>
z^a@otI_y%H=)Kf%ST0TCLisx0Cv)E|p_cmZjcIwSSJl^w^%u^;L=jr-t8vv&Zv!9$
zUN)SuUrCet{ZOEff-WgPpY~zbEjodS)|g6LP;fEGWu&M3z;Xt<-{KdzX)2GNuP_sb
z7^+$1^04L^WAha)*j0lSN+#eq;NzAs%_PzLDawre{A3sWQX{ewV@q<~@1D$_;J_8i
zpZNJeT7Kf}01pxT`y>;gu-p!Mb8-eMl0Q4+%{|d<)k_ibpqv3$R>1u_;py+{>%)c6
zv5pSIXP9yyxNmGYtmS)$hBP6)#H2$(+a|acAZ<j}b0atVvxyba2CoN{$OSK|k$+k=
z$;C~+ep?;9I=uP~0W%>yutY@UT}&v!Wy1bJX{jpIh(JZ@z55+G_;XIRfjqM-D|CJR
zKO-!K<1!x#wKikt=f7QFTWi`pf}`j^ySq9p15pTs|LUWG#|=^P%|k;&Q{1NW$t6zx
z=19EdTwdF2TE_N|5Q4GcH-wF|pq*wPpl2)7K@*FY9;m(XkrP)ZW?ta~I^Sp68mVL1
zw@)U%hJ`ZCb-lAUr$|s<7o?vML@HJ|2yVEFRNW$%7hXk8%l8}9;ZGFH{$a&QSvg?=
zZ_~6DWcHN3cpX!=9?T>eb*4LRau6r^NUG?))kYaD_jagSEVu2c#i^)%OEZB_Nr@3z
zJ?-Ro8rSnr5y5H^wyZsy)Q3eU;{AG>o;yiN>86NlzBe_)=5*WA-r-Nb>c3b274K)o
zfrB-*+Oxm29%08Ztm22E*EG}T*i;ruv9wqgYv!vIUS9M*&+w5_&kvMcc*`hSxz5`B
z&n~vJda<gga)^SH&1wuIx>m^I<o0^N*QVFl@$+hH>xEXUo$>y7;-<vUnwWz0ef2kc
zah=vyTgIJRgc4*EodpSE%ZuwF7;>AFpHkJ;s%os3dOoGvoXmQJFRy+}FKrIh@vJ`%
zigwT+8W2g&sz1E$Ar4v&zAb$#X9N*Rx9(s>ijkjBhXt0Q#Er9q^7rdOj4Nw(WB49{
z2!SLQj>ao!4)n%gXKpm#;s^V6sQ;krtsPVP6_Yi(%1m+p10H|kw-?2BP!8ZK;Np?P
zYtum7cfgK@4GOz%*~$1PnL992*q`nh>RJgDzlnOm+XH$*9Afq~Sj9XT9w;3Y?@gBs
z_|6=8nvht%o}CCf6}rf7$j>x`M003aAldog&GCt!H~WIWIuUR7+@z;0e~SY_2YT-C
z939mkn${hyINk2B44ZJxl~*+9q`*7^h2(4ztvoWr1jDU2bAd$33fX*C^+(T-ccy#c
zh6b6-c(DG#3Ljd}MsY8BM8DGROP;_8G~A3~mI#LVqj&f2-|w`l1eGSJ4Z+cAM4c+r
zx~bcuC8+b&0w|Tger+`=owut-a8Y!tBOr$sd~(~sD;P*7s~2fDj|kx0g1pgK3ksH$
zDno29D53!?1sbYRljEP=Y74vwp&w`L`U>V!0_zt3FNFQ3HrpBRRc&#v(|;pY?qmyZ
zMN-J2GhIPXr3=l$3(etTp8;95896{!jg5P+KA6Gg3xQ?DD`b-@XlQCiKFxv_^=`P4
zqa=kdJ$+Q!&s&>RGC?_A0?{u+atOpwe2s!#&4tvu37}Oorn8+~HD{$}`etG~mCS4~
z{!atbTl-v)as^Ux4yI}bLqhPxPpSv&znz!;+xhQAYVx&&Gf-Ey91Jk=7wR?nG~6f=
zIx4Gl{mIC78ryh{gtJ;N{Ax>btl*W~3DU^CU?BUj%7*xzfa@3qW^Pe0!LFBr1Fd58
zZKbRMlKW$N;&F;4eB3?R2P?0ri!|Nq8cUL?!fhk|e*H%5EbUuINbRMDJRzqj)_;Ar
zdQLWm{CLa7;v<!HS>5mU)xmR7y<M9Tb~c967XuIM?Yma;<t{uWU}l4|Bc2LQQ^3ef
z_;Qy8#YFvKjPb!)3rTnN29;LK1o--uu-nhLuI@iQVH2EI;!ek<4!{)Ch*MkHcL8u6
zHu4U7Iidu#bw|>8B5xrBOhHi*5$$R6>@G$HK6ezA5m*620GjnatJCFi>_>_*RaI5Q
z&<Ho`pKNuDcwA60#?5nS&N#sG@ebhnK)V7;8Y1IB*Yj*yT|?v1INockH~J{yp-nBZ
zX$!YbJ%s*Pd?+U0yTsn`Py2h16#vo0;DfJdNrn`IPR@u}K_2M6;f#zIi}=KEN^0;F
zkgvUxxn}u>BV2v9k~s0*5LFp+9ZE|}*;963tq$N(+7U`7XN^*2ml>fW8~8)&FQIgL
zaq%sL9G$&4DmogGIyA}~$PxVViJ;&MN%{IG-0>~EiZPseJ%A3&XlZHf<4wXe4U)y6
zZOp;fff;j$XMu);zwc{kgfoEGw{71!5z?cF&$muB;)kRNM|kfg-kQi!F8yA2<Brb*
zzUHkig?%wA(hjrK-gJ6YrCBQ(&DYUUQH0m*|LA`eM*=NXSM@W&G6VV|k<#G0vQ$@j
z60YC%G;$U5fWijiIDaFN6NpR3;|ZsEOn_!r{aeSgvAt9d@s>lUu`IT;hFwxE)M*--
zZD#HEGvS{mJjPj~{qA9<mm~U^L2&|(_35OU<zmJm@xf)$HJJtxW2dO+&n&j8>v>Dj
zivTOn^0yzY^(&7RC=blmlSB5uF(2=>JPfm8m3jX2qp31AHirE@YMkTZvM+&6s-@ka
zY?)Sjk~b~KeP}LQAt7$Rg0G*KX8)R`94Y4quKb2e=hv>Lf12iyvCY@--jS;kf(@Jk
zliEe976?6Qv5cc|@hGuWk*A05-Me>}l)VrCha=S^0@(GKfPo0mAS7+)tp7v?fg?O#
zSjtm+Y@=eo#otK<h$dH6r`hA!4h}q!R-qvPmO4>SLhqidyLzPt0QD?vY?6^uQrIMO
z_iMMf!HEv>JeX*R^@@;`*4W=F4!`A(N5hn$06q`E$1sF074a;>uCC>x?9<m~FPZ{<
zjAQw<N|chRmsO}$6m9is$1T3}1l>N)_$~?`3nd+hC+T-4N&~>`0J?7?M&WRmI|)u7
zFNY}|twg}~NN)^y+36J(F%U}!5X^+TDo^ckFb{?OkI4-rReI}Rs#@{P1X$nAf3iZp
z1E{^AS%PP>yqsu+7fg?{A5gcg_7>+bcMsD1_G|nCApr6qNr1#)_y))UPRK5H-kmXW
z{8_D_N?0<{qQ~xUNQ;Hi6JFtI-yCC>&{>({Fm)qTksjIK%!uKTz2_bl509*C@{^H=
zKR+7<byWVSu{&28G%GrP4rez%zx`FJ8wY6CV|yL@UfCO_nI-rQ?su`1Gtr;p8Di0{
zVen9+=A<t;*UAQD<R-l9Yv5IM8(HmcTIz}tOxd38R%GK8QM*6;@4YDkMPl9|<ucLW
z_NO+j*?@l^#S{PNpv2Q=y-MP1eLdC3j?y<Iu0xd%hX*}rG)87$ZojhBZS+2GILTX}
zEXj1$#rv&<i6j}0wtBs3eKyfKds3kpW#->ZgRi(Zd!KTlf2O1$5kpQa)am`Ee+li~
z2C79aR(_k_Re_|-%Ot~pK#o_iAKXW&W{_)_5J^^2*F&fI$U~QHWrX#_He~qsRzY>O
zCMZ5t35&b=-#vNv28%Z>W_g$%W5+w-O_|C!HGTa7sPFG-_wpJqa8gda8VA(|8y~{W
z1~xW^fb&5sXPo+Zlk<$rdD@W#ViF-8&4_KETU876VWt)qAVX~aRyh-BT>2p6^po<<
zyO=0gzP_hDy+PkjIBVAEz<SlmsX~SC)fIY~7@Fd(u3hw1mMBw0J*<@VfCNEY<t9=B
z?w9N^lRb%4)YMa+P0<7T)OScbC+5_Zr3-)QJcvb~=?G6h>q=#5u9V5YE@@FwQA0B`
zAyeBO+fE}h+ATx5vg}NAVUbn8g)>H#<TFFF<YO0X1i>SNQrx|2sD(4WZS@ebQ8zQY
zmz9+TGIXOQJhcg1z6i;jva+aw0VTKv1ACV<;6I?S132R}pc&EpmB^^*ru#~tnCB?=
zI1wh<4?mZcu97GEGNZVo^u2*H<=XDs{IkI}0FrDG-pu-$Q+}oM;DkIRWopGnFcA49
zDRXdm_)%;jQ4EiQ??G^I#*H_x$-C&{m@;Tln6OaF6&*>f54KGb>~E2|?=)X0vJoeT
zj_nUr-uwPf{rx|P;b4Z|=At~2qHyi8ZI|wKw{>P?bN>?9+-U499yd<5aL|aR-cH^8
za?YdZ@>;M>;=%;;+I^Wd--DCzZ0iJ-S0Qp053_o+c-mT+HZk%iUxaoFcEIL`)ot~t
zNh){C?swGWG%pS6xU3kynMoa(e&?WXN<TlF^k5zH+&2+6ur{8UYWwGk3j@+sg<jNa
zGF8GL%kkhc=zWAZ0%!?W!sH@8umqC&BY2o!zI-9o?#|-=ayMT_E<19G+mF}+sv~@o
zg^+bXG>FQ@4>M&JywqO75Cjw#=tM6sFL~D$gVjR`wYg(r?dOK>77Fb?<~&~@0j3hL
z^%3$r&@6pskR$!+3Eo~{MN|If*0=?`K|_)Se-MbZ2+&|ht%il1fcNkfX&cESYUKJ`
zXe0%_zUORv4m4X-KKH2bkjSt#_YC|^dz}cjB&1f)bYLxtc-EY@wT1aU8`DV%6B(Kg
zieq46RGnTNAqdECczfU_R&AB^BBCG}85z~lqz;6r9+@u$LOkKbLAhXbivT_J)$p?G
z>yr`5V?pg<uj4KFd+&Rl$rH1`cqmG}`u=cicqok^$TY9W8st-8&}bu^%&uP-XTdMe
zMd~h3q_XQ|AN|UHMcd0f0ce(z8lfEbVV3c8)yeQKq;Wh6YxQ3$yUW10<Q?1hXX}As
z;_W?HKcCe`@nwF;AK^V4KAzoEj-7UTcmEHEi>0ivCXs+<@(T?OM_ooTZx%|C)|*YE
zju#w6!ONz@DtWozK<90ASR;pFy2k(h_W`jL?^9zn_pKSPrQSsAOkJ0^k0w$d3tId?
zcnb>e5*qSz4oj2Ed{>Lj9T|<#A{KjArx^C!)l!<-zJMaFlkAGrv70;7o^jifcN-ke
zunP@lf_pukGNZa*fZf#b>nEm>IrX2nOJ7~A&N*G>zrLx+DG@>Y66Ky)sCiP|#=yOa
zD*Juyicx-hM?D;*8nc71#h&7go7E%Yg^RhWe`{*j#n^dLya$@%r~a&e|7Y95L-Mh|
z-zEX4WrZ_gYTGGEiP_(fP-kqt2CfB8e;`)U2hu&wLt#3t_XKze%ra}ILkvm#1_q!W
zg2D_Ep&XJwO`wrjHw1%`D-$ygPa=%3Fh?XP+&ptnIXJL`__!Mu?<6=F5svK8X)BJr
zC^sj9&nYmqJ`@cwwM)wDmV<i=q#Du$XfQ^2cp&m7U}=fNQ=2r4wu>8wg3J4O@3yQv
zX&spj&4=~abOB6y4C8xI)i2*;gv)*~v*wtwA&}8iv&DD}&JRS(s0#Uryxo0^l(`|p
zq;>XJ^DJPt3NF-<kcq^>(Y|a~ynL|y!Bhi}T4{Jjymx0uA1WmP6tMe-2_Kqa2xI{T
zIr(RxpCZ5qAQuBAlTgQCAaSe%A^Lm!mTh@?8J0kRb7|^1{xs8+sMlPp@$CEk9l$GK
z97Ql(Ts_WNsI~|D0PJRRlqbA;r}w&fhar*qK>&TUVFrVAqRN}GHwjEwbc<Q!j2U+n
zBe(=z6PLtlYqe@Dgc1sy!v}wO)$hr3#*ei}Y`8A*9VteA!=ZG&L%h2o#55-|F)1TH
zVs9zHm}es}0zfjvAi0R@Cn*zuegdo*m~p;%zmMytrAynycrSyI*p>D`e&8<cFQM+6
zD|;2ALq&sg2eEfeMv^wR<x%NhmGS?271t4NSMpeHq7;yy(_{&m;)&`sL$bH#B&r*K
z<|8Ows5ss61n{~;BqMI2-;yH?RWG$&+9JFXiN;>!a1(IIulS$))+sJt@!%pvvd(J+
z1e{;l5BI-$T@ajzm3<b&d@@n*^+9o!WA4-HvF&2?TNZqm>ivAazPI_pSss?!jbV_}
z;$y2G$9qe0zsPG`l09;Dan*4yYEUW3D30m+Y|9f*p}%3-QjCbat!c@6HCefQp_xnZ
znECOC!}!R;sU42=?rpIZ@>ry(DI)VED<+=2NE&bI*R!oBcGn}dqzQ8N?=KECo|0HC
zFC~gj6@>K}33UR!DBFxB@L64$6)6TNjY}uMS1SihtY9&PJIJdjIpzpKOc7sNc1u3I
zxpS{Ln(M&>)ugvEgR4PaDPPsq2+{Mrn9n;TgM&j5y%!QpFQB|sP*4D8U!YlL6%+)6
zvBsiy#rt0?Ow?7^tH14fIS>?5e2+Cr7;cIo87IovKRCbGq@SEuaS>cw*r`n+7pOmh
z4@x(9Owe-S*%5Oq$$mr2LdE#ivT(T<KrI*-5QR+cedG2^k|uOupyv92l2j}Ob|YlU
z>Lu+jIS2=7yal(h%>44`pGU~SJtl<>f6YCQOH*NI4FL&=oi)H0gnDCSq?GfWCnw=S
z;^<9~(++BN0Nj9&2}U1=4i1C?tv<86klOJbemelBe1P?af(-mNe2EJ#z_<~i))3P5
zPZZ>pWjy-65_lJa{fz>f6nrOaxo>|0Upw!XbN8L!n>C<!Z8KQ7kI@Lyg25G=(dhZF
z_6z(G?MHiEYgybJpO?qC@IF+}cw0YIR7%fhHK^W;Je8NnnL_y(PZ@HQY;#n7oR+gN
zFGFoIQSzEuPl@2ef>0O(EVscP2bS$6#@!JMt7;JpK&Srl1x+pj*aMHmWXyV$)#P#R
zWG*WF42oi=Tq8p*S7_cmrY&14{`zt-hK`(x!_o2MLY|)&CT74J%K}V|)d_r$*s+vf
zcam2SQ-%X!%BX^Z1F!ySPL4}TyZlP>kN>{m+l^dzDah~->uIjTf6D8GBr-C|(dqSS
zuWSG6@oBwN_ov6Ou*QM0@$WRARqa`ZKz)(2y9>P+Tqt7MtQ^r}<2=;!12RrVm7my)
z?q`0@Ti|Fc%#{8i<XLt^VBl479&Dk^=Xk|6J?QF#p*@XE{El@zHIB*jO;K%#TFLOn
zz4?O#6Ybv(uL@^BzHd7CfnAtuw%~<<pVnxYLlf;GaYk<7W^utFb8j*KjY<EJK*`Dy
z71yBf|3Lv9-Ov<(an^V~C+yRda=>H}2#gGJtXBy*sQ}Z$5M0|E5@Ac*CIwI{knJt_
zbu3&l38!WlBjrRh9=*J>`;d$<>+w8orLmj$BY`*JHnr!eA-~1=LrF>0#<PBFQW_eW
zyaGeRU20Dn@@&Wq1-8Uq{GdCAv}#~_z|^u-2?^(MDA*fy8v}n6k@E(f@zxf5sKk{Z
z_+hG}ntMA?|1oVIMvh0_8**|nqxi#k7mw5%MIB28ot+TT5^U)P3)hjPR@RLVs6;Tp
z!~~p*&`wE^If5rLE+ucZrzA$zzXRq!1o%0D#<QD&8AHH&PMscY4&9|<79I&=34|a;
z#3@FMi@?DHaUA5V6&xT8)j`+`q!}YWFuhqcl{hB@`*wt7GW<CT@I*DMbvdo7=KV*s
zPwEZ_MCKR$&8Y1$6>Mk7UeYD;(xSd7pa0oPQTiV)z$Ix-PefL4ZUKu|>+6#(_xOR=
zY_bfIeysAX?$eHT@q~Ce6eBAct3$MST3<5D*BvXrv0&4HqlvO-6SGaV87&rFk<$i0
z_Rm->ESa8mg>;oECZssO(7t;n#&y^4kC2ioP57vSV(RA26qy(Mveo~LvL_T~=D7+J
zbGS|~2LuF|nHMlq2SqAo?P2yPpH=<$Dx$c~`Q~T;@p9`p`TH*}YI@m|G<#cIVP+=^
z>p$z`h_xB>61DriIIkX4BwiDQ{?n8Ce6v^C?VS1G^Q`k>>o?57HGX7I(I21HN{YvI
z+Ii;F1#~M~=9lhI3QxRHm1**HH`C0UZdUxfJudyp;PX-Am4AlL*-_*o!Ha><G>zek
z+tIm$)BUTSoc2@B5`*N9Ni$TThjW=8>iYH+l~x6dRMavUFYd;-+YN-jjt>mP2AK^E
z`CuMw<Orl(NZVrXU9zh-Di6sf+F^i3HTZL=YGrQ7+7&4Rd$n9SIS0NQ;7ss+=_<LT
zD@lz7zG0k1m5E9@`)BTbeSLdA`0hw@5rPhD1mUuX$(kje;ZEoe0UkxLlm^fUFfVGi
ztp`An0~*Gt053fV+5Isf*f*oT9;^~<g_FSr3Gw(iCOZh<L~Sa#!AW(z<5S}GUQ56s
zVE!fXV^`Tkzvu4a$iirEDjy|BLJtM))6YZAaMul0{>w81@L}kO92ZceLNXPoK@<BC
zv=m5r8YfXIG2i6`i1hZSq0!L~FbaT%x4qhm#K8T*c0c!}5i?Qr5oE)FzcsjQfQ#x6
zh(84e1_Zx+e&PyZloyp&RPb{xD!8ypL7gx&qmMApX=*|O&g~|s9+U982mHq>9uexR
zaeL;s%+q?5^0O-2wW@RwrCGlWHvOBOdvpAarCH8_W6bpR_{sY3iyK?5p*%wT?Q>E|
zAw}*dij+U|?u3+axtCWcif*7$PLJ>X44CQrBcjS)=CPF;XE=j`x_&c%?e;q5eT!-T
zyX52O$f{xSl`r>ieEoGWZ0kfrnNdub;`LX$BqC8>V(&o=Uv%B4%zt34|E!J=^c`GW
z5@fEc65+;W;c$I>IH0_O80e07kW=2xH`xw;mR2)X|2+N5ka!FuY+z`*Hk5{vIb36~
zt0IIyDFkIGHNm6R)i_ALr>d;R=R{LrzNFq$mv>gD$8<7wA<Cl4z`EMU12>;rkLC<_
ztdKqGV^*lK6{OxD@RwecOs(nu!J4f*#M@cf{C(BHqZ98e{BJt}IkVjHUtkoLWq8aP
z%T|FK#qbHN$$+JfV50+o1fpfF8i#uc+3D}pZ>xPWD(~0r?Esk3kQCPO2oG2(I(Rmd
zCP?XdO4|vnUcN*c;dNclihWUS$A~ZXhX5Kr;3>ef2VszsmI;GPrwC!x)^u$wi21^=
z&r+|Oq0zOPu6Zod(tS^xvo6ZTT>#W^ur+@A^eJQwC@3lY7#?m0UqvP8XHTYGY1;Z9
zM~{EN*dgm0duWbzo5>jau=Bg>oAR~TM}>Z)(pA$!+S=OK(Y6^~P_qUzq`|tB5_)DJ
zkot{nE;pR-ET#ugx6IpcaSaW2?ZgUs^_GLpFBj>T+FFXl{tj5RLA$=1>Xm}Pu;k|A
zSLt~Lp5OW;6_=D`YyroULLh?$8`gk8CSr=5hj?BS*e$QDtQh^A)jfIvK4lQJ4e_99
zg1$Wm$N9MhISWTW3Ns7zhMqp-ISV@0WNQhrqH!%F8oRPCbNIlh=@Bzll=;rOdLbfm
zwpc?$tGQgWp5~0NzOJ8M>v(wUO1N-FSCPs3yL*2Jk8AG#@DXN~4_rUbUKJ<ya%%US
z3#3ecW%f)B8uVJYWXA1I8zSFdlyHqP758GK?u^g#`Qfm*gj0`hO&Esk?$|9HbJ{yD
zM6s|e8%D9Qs=6sE9vp?{b9-VSXDU`@M{u8olrQ>mG;7{hk^F!R2H*6Ie8R$M6uVYp
ztKWYFoi8?S9b4A<4LF}Z%OQ%WpUT^?ZQVB3;?!O9!dG#=!gs#Nh+X;G&}O+>k3$}O
zbt5jZlkk%iyc58>f_^O9?Ai0@{tR+(^r3iTybL1dK$t}p#I-BL^=*{4gDdnl5&9z*
zjMp&1fF*HBJT8%T5{Z8{Sm;X8frVOy`<Y#e$FY`)scC$2va~88XZ6zU2v?%?itutp
zK0g&D4Iv=u7@W3{K<38s-h3-k7YZe5>X-VA_p%J{r?wGZ0-!Smi#{_;OGe@#q7@!R
z)BvPr_CEgGH*I9Y_#Orrc?q?m$$8hIg*b7Ex4@jA02l^9O?&zG-WhmS3(lGqMhtM_
zzH@Vf>}80s1-+_YW<v_LDnjN7n0$)M!VOTVfHMo^W&AVyU_x+<pZru_zK&wkeXs_7
z1o%6+(ZLJC<+}8gfMmbftK&-OvVmQ%2*(j48RR37&*Tf_o39aDdge<_O6tQ(Pws>>
z#Cyv(sXgmxatRx@36!rD4kOyN-JH<%ah`iKOQy}Fbp3)<mCHtisE3tRHoNr8VyRT5
z5U*HggkRA23H?Jwr7#=e8T#4N|EZ)&s=oAht!gi+ug-P62Bay78w^hpD)ZlJ;W$k^
z(HHadk2PztVC}SFkdt|RgI`s2$7J>-VO`hXG^g`yrCE>iqfFIilTr<e#zklT=g}lu
zcKeS%Hl0mMdec_rObt9_pBo!zNZn~tjcreuN(`sj%(6xirWjaadv{n6wxws8bl&)<
z^5Wf}Q&VG2bVo}nc7xjZeu}C6M)@MD{7)bR3JAO73JClRp1&bMXlL<+1<a3tg#$b-
zeNXDh-oAZuaxzZBOT6*#&m!#f&yq4~8Vq_yU~z@eE{KbVumpCM=qA}d|97YUkru1K
zV7z1A2BoNSk7!_Ve+ZZ+VCJ~CzCNFwXtjVTaf%GCA)CUm#m_N6BR4<71+I}~03Pi6
zz*;DP`rSw1SpYr#4FpicE0j=>%zq0&1YE-%0mcl5qy;aEQJ^z~cOFnLKLU$FR#&UI
zh?m6FBJX0p$fai#^dWT;R$aq92c)qyf<BI8Y|+y>UrJ$zkI-=f4NH-oSW-g+Uet?H
zlbeud3Ph4W4>|ZYtg^SAfKmf__Xt^0FE?y)LA9H02HqIA0H~i3cycmh20VG6$I<%0
z+f9*x*@hOYUFADma9T09q!CqAlsc3stX>C5+?P8Z1u2@{by@O)-?zz@BA#y*KQhVZ
zG#)9@{*m33`p&mWZ(3~2WnXAAmYrFM@99)RZ@kzA<#~j-@H8ulRdt{-YtWdz(-iZ6
z57tJm)?}%HdzA*Yq1Ll!s+=!Gt_!w17oVR(N495uJ-DlUdSS!RO*DHcc0C4`;5M2R
zj7sNPA~BV%mLe&lJH=Gtv<7Q4J&o*o%8}=%6b%NUx;iphWxjg%w--CklB<xaGj2;v
zZ%xrGPilMFYVk#5Jo_@>p3^PR4-W{=7wi}V&s$V7g=8zG_cGKZ{~TE?kBRBHE)C!O
z{a2u}IPZ9xYh~a-NsZ8NWyJ*i5XnCy1_7Y<{WG7b5@wAT?hd6(0(M8*jCYS-QSCL)
z_1~iFz}ERDC>%mU!2VB%stK?myuENmx$pVXFvJFr(3)_Oiy$m)V2g+t@gOM=)*vt<
zcu!e3Qt6GU5~9~mE4AH83m1)GK(J`JNR5GM*Z$lO2}TD<QDP>{^n6Xi!NE~hSqZU1
za!iqc%R@b@0w)Eq39R}RbF1gW-SpN1<m^Yfx5G4N0!N7aOcnM%(oQs~r_Z1@KT<37
z`>c5_GJDz~0E-J)m^uo}_8Elss|wCQ;|6O~a36)48L%X!oJe>97iWjn#;wB7)^Sz!
zXn`d7j}w~f;R=W~;Va?F1DgiPPnvUoH#t|q#Z|Jjz4~C$3tyjz?yl@g(_L@P_jPy?
z-9L^t?@HeN9XYoeDuLVR7T0(_YOv^}m3~+oNPEz_>rr|-cB1D#A>zWdtFN~yFjVf@
zT8uw%%_SqCT86^?-yFLcoi_S!r??8YHzwr8jwwl(vz0wUKWR32{5qajH@2!X=V-ia
z1?kdy%8~Y~VM2+CnSd$7SFhY|9%Wgzjnox;nCW>-Hj{STI6hQ*$HY63qwkvFW#D@E
zP&CJ7%DEMTfAl)xoaUU#U~RT9=laS-m_m5{k9f;SV^!k0yw&T&_fJRjR;#Zo&sEmH
zw0Ir#r&vChJx)>SK42?+efV~k%lso%<0+?viwm2?1&NK<(TGDm&Ww7D^Do%0NL;Qf
zga247*J=BG&t`b{!J()Pu{{$0$CV6P?Kl1yS!oy=0|xKJA;kwe3`nqnc`XY7fXXR=
zy%BON#E$LhQ<wGOfH!&`)Y~&!&~3x<|Lt)yEtTn)9%j#hW?xNt=1z?i+v``G_xYI^
zD6HH0{SD4~G!Zxm1ph*YF>NpRw{PDlFBW&PrU*+~FCnv_)N+d7#1!Xn5JU`TqdDU3
z(Cgs${rN5-ipqCBmq*Eh!A^8{4CPzvUL0No*6XJ4s;3@)E$GE;@kQ<t5F-a8NCbmn
zMS00~ctGy@nlZ5LCMUJQ{Snqj;479!umF(>s^IF8QFUY2HQwWfFT%(L#F|zOXC`Qm
zfC9_I&D{c%0x$rf!2{m<JI}y#Pq#Sh>$%j2{Iuk#0^i&4EXSG)XIm9;dwl8%C%;a5
zGrVjunEkto*n$1;6{DGJTXXv7LfY6dITatDD@7@}+++Emv+|KBBQyC2PY5X(no#Yu
z6fkI$_<huLV~P`{PEfEvWgp8~zdW!}(CaSe`MNZRKv8^WQB45XA7@QjjPC-(D<bu^
zXD87lDGTI|7H7SQ6V7AxTNEiw9H<9&6}rC$ex;u5Z`L+;6@_X~mfNAeok3TVh5A*m
zP?)w!Z_r}x9xg&X-6Mu?VRXhU>{_0}{&becv)|Y2x3o1^U?7E;hsayb`jAb|H{yk7
zhGplscCM#K98Q!YpS}J~#Nl8y-uGG4N!3<8e6k^Rrn&xnvuX!NfKI{||3soObU!z6
zt0?pEg5qrLs5j3=@mxlPb#ZEEdB(sp%bC?L|I|#sA)Z14`m!-JxAYx@!@1oGY9YL^
z`sgwTtA)wLilMVfe8r5=POp7h<=w5<qUm&Qn`<vB<0*<}rAjx~EX7-V8>f02hRrgx
zY~0xLbiVudk6EN|QhsgfT|I0!XM#mAaOL4jf_F1n=~aSBdWnKsvF22_f=QSqcL_T{
zGWXcnARByRV*}zLATbV>cLVRyabUQ#OHHilv4hMBpi5+dgA0O|tn)^Y4GSgPETVTA
zZr)Zdh;RAA@XFuRrj`=WL}RTds)T5d(g1^j=M1;?<#dfBKpw**Bdu!Z=YgG6pQHe*
zhw7GX_UGsb2s^ayRKZz>3lw2D9;45dG%JbVy1lf;I9@euym9v>ZkYX$P%Gr#fItjX
z?r@bGEJzNQLh%Iv2JjKMuOZM*BWXW7pB1Ldi3u&h`AZs!UH7EnI07js6Ar232zYbA
zM2%Qf`H|><gw?2>-E9r3%x~0OqJE2(C9TriVR|^IkH(*2H$2^2q7_1Re#?c^^)N-#
zv_R|1SgMu&w3mE#z4d?genvt{kyEXf+c?7tcPbTER$DJAmD&cMI;smNu7*qOnDn;y
z&ioGaSH<d9GE9l@HCvtReSXl=>mho2uBq=_t*-w{G~vll<!htqh)E=lW&L-)#UnEv
z6B&t2V#QpgXOvk^o%g=c3s}z6G?XXp?6CC9aMq6L9Xh+qH3X!JRs=Pukn0jT6i&7D
z(@eD0rmu8D5j`}XEUJ!KR`v|tp;9j09Te8$M3tJ@Pj$9Au~%h{(QR{#<zIZB`KRHJ
zVCcmv)mGx)+s>Zf>q&F{QZAUBe~&2U=LhwVS*I(>J*f9$3%y>ja-yp96S1CEm=2$l
z|BU;9H)$o$t)XMcfrDbknTq1$kMowrEtkDS+^lOK>&uB@L#i+DcIFdNRfYMj=ASe^
z&0W1m?X{!dAXI3#nmqQy-FRx)a&|Ff*NRc2At|V^Cd%Z#6sFVODbLA4`3%AJUb)U%
z5>Dgo>5IKy=R=1!<yG{=>f%Wyhc_b1Vt*xf;3`OV{w-UIyG{i^#oC&quA-M;b^f_r
z=t;>@^?7w|ZR}Kor%2)q`lq9KUXo{d`GNf=uYisZ=JvE$(2J>Sa71CBKTLjhOEJ7o
z8*6xFk~=^DDGaG_yVXGg9@s9cXlo;|itsg-Lw?;B)Y(&vF<twc{~8u9T(E1!z&P`-
zt<d6rSiebkCFkSmO5g4{DRF~0)o&`NpvP@TDu6}<G~oygXQz6VdcjI!W{Dkmdhpzj
zZD!TgB|W5Vk|xoUQdY)a!h76-E=ROQr;aL{c7FHwjg)QLkXFxnuE;*a$DzpL(LF@b
zAjE;mo)njxORz2CuXcowIXvB<kANK|NP1KWf!+a(z!s}Y9wDJ_o<UKtgJ{lPelhlZ
z@JlvGoB>eUnr%!48yoQF3kLIT@ci%PhKb!JpS4*GyT4y0#pfXh+6b32sy$WRl3(Dx
zDlRmuo9_ujE`RqpAfL`x?xbyBY*x8UA2|P2QIg{o^zSto@VR<$P|T&bS+lmDDpfWW
zZ?I^{iX6@Paf7>hh<5qsPX$U*?;_i~SI_Prn5oa6?7WNkR(g1LfjO3XM&jU4R@0H}
z?CB?RwT|~Jo>IRua`t-POq={|Jei1so(xF|`66?pds1NFQ`QUaSOZ(e#)fNZe*I%)
z%D~`3%W~baf69v84)_wc?~wm9U?&AxWPI#>{*hdKQaJRp2H%B=MO;$hNaDx+(a|=t
zDNXUB{ilONiTK@fKhw4dS$ni8*sA9PIV+s;xy1S0$YtJ{lcQ3Lm*zzdUS0T=R)kg>
zu_rI-U^*XY9>%Eb9wbdZEfkJP)pY)YT4u0swo|uIK)Ik};%B^yV$Ix}n0CHLP=5!w
zR--%T)vMJ6SLLZI1A~?<OeKk&>lr&=4VGpCuDy;2L<b5>8m@Osr#(`uohuYf<V<dx
zU}-TUx7;Qkd93oD?wI*`GwZJ`)~m2zPs_E-CDtF8Zxkt74r?~EN`HDv_~<U-d#x8G
ziZSu*rT0$;>`t~V8YOo6`aduj)8Acko@sE27g`k>blr&*IY|*sT|GY&Z5bJJ_v6!e
z=uVF>QIhPkY~FDClZ<Zk=&Kg>U-_$pZW-4<?aRNtw$<-Pp!}ts!W`R+E}E7L%5-q)
z7tuc3xx_DCpEweAetWKZ941!F{rjwP_NC(XsE@=8>1p4SVyp8=3m3<msaq=Eq=_k$
zPm`YPoL*cOBwM*%I+ls-EcQ0O3%ObJ`_cK&EM%wa3q$@8RtNVBw%=|?fBJuY7qy`=
zxcZ4f?P17gP&d>6cF>wp3CH<zxxRoi$nfopk$o0}b3+m_%u<3Hmx9e~wl1(tB0vL&
zV=i3iu;A3fUI0uPx_Y!qBDMN-^YRtO9AP$~#4ClR7lWJ;gu#H@18jJvr>9}^S}eOq
zNh&)$!-^2eAWpd~>~KI+$pD_-y~_ZMfnF6tgrTH#DUm#3dDyeb2#OVD4_okfDBKi1
z`E3s+4s6;J1+1A!ai6gh-xI{`Grm2-TYYiRi?D6y<12pIi~^kV%2`hl3D^BdE>El|
zUMwdr$FMD0eaydlVV49fncm3AcOE~!xB9j!YCc9O2Q2O(l%T=mJWj)d>|o6h&Fdy`
zd0^j5OYm^a5h@z6NuWc6n*+GFuv3eI_%r#z($|~MQ{HVrrUxSOrag!RgmqwP1i-~g
ze<B(5lxHQZxwwywDC-=D(Em_yv7uw)+PFS6|G><+hK}+u{m*T|Gym91qCypp8e5Zw
zexCShA+wdz5VQA4acZ3(BBRDO6Z|1NE<%b)^C7{*+tOPV5s%HVxGB8O?%0*2&RpGy
z(VOwn?u%eE=6WkB^;`Jtg5?~;;){QW76DCKi4z_szS9p9D^s(@Jj@iO(zX1l_~>8C
z=QXdm?kRSR(&Py3So9@Tjht?RZ%kGYC)fhsl0qK3B`wRuWFwg5BFY(@H1ONsye^tA
zH(su3sbXq(ZvSY|{*p%DtwiTPNrrVChz)3YdG!AN{-M(#BEGcJP3xb}Qii7WD-D^L
zhKvP4w17gORB_kWJFl!K#c(qT8R?Vc@Zt)6-OWq7813J_$^9ytLr2y$9b38ZoI(8W
zTF@$AVjoj9t0t&G1cDb6G+#S@?qAIOtvlsn7ta2(d!%ur&Etrwpul4>yLfeMf3E_o
za=JE#c>VKK%3JpEtqj_STxs4ZcT><*)-NufFFcFm+tqb-=+02@bR4TZ9KYJOm?nPJ
z&tX_1Qt=}`AjY{!h%O8{ZRj*aqwM0ZDp+DMyKU!ZvXko+8t4&sA=x;7H&^vm9CRA4
zHh6s$NY8k`C*!BC^XUC2-qD)nhkM7C4m%s~L!4;)67FAz@wpP290xsqB~On%_3^Wm
z&y%6ri5d=!nP^>S*SXvZj!O4HHy!S(4NLDpb0p(W?dn+lq-s;i?#qceqTSXM;M$Nn
z2!ZdQ|ACMjmO!xM2_#}02gi3UG5NPTIC#z7d#7Ab%^nfJo)?66;j#=|ils?9$F;MU
zA@ZEnAB^rQJ7$D>Hh1q#*J@Q*kQLP*lY^`e;AdD_!}q*a9oRjIwQEp_+^-9)FI5l!
zH9M;ep!N%Pqsl3G95y!0VYv|(7e@<091w#5E9d?qXXm?Z+t9}0QHz2f6#|Zy4kCWQ
z&ky>B0C(&Uwpin7Yaghoww`qla1Egw;Tzw4b0^K5HBkXm`{`j8OV){W@r$-?Ctr)!
zWe4u~L4f|`5ZTolpr8VfsJZzO9;ILt0<&hg-OLqihpyXZX-dm4CmOl@ce5x51n!Um
zfcO%ag}@B?gTe?Vj%o7_L9QF^icgDS?kCAt1p4tT6OThyo)Q@g^nO?I*3-V1xka~t
zN^|w%Y!FT~2+pAz*Ay;7Xq_M>v?H(>O~?k*l^&;<n!-C+Xm3(@0<I_j;^r$q7I0*k
z2%)>d@I$V{o@_<9wTe(=<t6s|5A)*_sx<x=azRD@wN@^JstFTsteI<1EKS=Lz53E)
zJp;3e1s%A0+$64Cp+&|E4c)Zz@T#uQUN7StLrr^|Wuu)a)^{M(ltlm2))J@1Tcv_A
zx%<!V<frdKqNBoKV7)grdsV^i9hA|vw8Q{mCzEfB5LX&28yoE1Ay=OFGx*@Za^Tgg
zS0BX%*278Ps~L@|O6Bi0zeot1{%rnUH0cSh+pQq!W1$jV;@oBBF;Aq*%kj@>e!2;>
zb!7p*m~OcRcyxh#S9}5yy(BK3ij^5%DDTtKHf>t@G&9Y3MC+&?S-Gwz9E8>sqB{MW
zsi_%tPPh5(8pe>v{{24(Uv~`=aSR*Ym2cGk|JsA-yw~CUR>^_&;nd|V1<mw@gthkU
z?CC0V^$ESe{7JG4$MVf#-mou`&!aQ~P<aJDxVo;+FPcQ2QSz_&dyFra>J->d7+dL$
zcZC{=xAB}~4CWW9Egxq;xq7`!1YZ8UsjHt-^!a68)p2q6v;{koA^Tt_`jU6vBaCmW
zbzSMcIm(}^%RtZ9bxb8z7WyqKYokEFN9)x$UOmEG!W?7$#GABFl?{loZqHVRe-@Le
z<m^x<eEHpr0E=gw=VKts@Ywii!S-6m6I?=-08Fyc`1B2o+^_gFJQ1>vq#y5uw%<dS
zRHRCEw7^JsMs=6=kvn@x4=#xi{nOPLrFZ@dZ9J^`g)MLJSt+}XdhmYmAoGq=MaMoT
z*T$jZ;UDB^v}IJ@n(XqocK0u2WYcQMO+H57=Jn@%JSz+jYfT|UB+ADGsu1@8#1uZs
zg{~7Wc7Ino#u$`?85u^bZ1PzkaR$pdAdLVB50nfL2n~YiOx53#GlkO)(wX{zm+23I
z9!!cG)1K-etXE-|L@WfA$;b#Le}jeLb`>Qa=za6l^Vtz>#0(aw#em}j2!HW+tT7wl
z^0!6R3B`KUDEj!o@dB52kPJXl6)exc7Byh|-%z@d!S<*Q`wJ>pHS%rcN49LawjbkZ
zT^{0j3nJt2<&uQICX-erKtseN0G$l*_qK)1r6gpST;urO691Ec1qu_$CFbJif5H?A
z;Ng$)aamDnNWLq{q%~50qV;{v?zfcy+yv9L&ZY3c1J4U&UvCSm{VkMy9Ozvrnvy>F
z7Cn4`d%stO^1ToBkOK>;3Uf;dKNiaFI{Nv9rk-BcKZ_FCv&+nU#`*5W{V1Me=oRww
zpXuo6<O{MB%8p}?OEGKkl`pB8y~Ff>Lw(z<;HH{j(o(rBMmIAZ&Bw(*f?)?Jd3(ju
z&(bHWb#=xb+vNtuELS2i9cJs|c^yQiayjqlJ}=bOn)#;ALI?~pk*{OA?8I7Cd!rU8
z*<bHx<xPt>za(ho3VhV3@E<M!!pyMO5`GhaD*?%LbmVh#yNAg&qxo)#HGwWX$~%cO
zqCa=d=(ee1*n30TEW0jw3Nx4TSYr#qN21M{?{ihwONTu#Z{td$wt`y!0HmM5Z%6#p
z(8=MSBO_{83c`0X;^lOa@q1U;1sBEtO*;DjKMqK6YPDFGhPy;=1vs|O{gOCi=D#{2
zQOfh+*PK(nUchv}Y?#U|5<2>wyzwzLrh|3M>8y$_zVogcjYFoasrC7i^W<sYQEW-E
z+i3y!zzmiuL9S!<=~PHoU1lM%t@$+5g3Vs>*<sb!*PI4Z4UsbVlni<Kq}iWX9)2E?
z&YW_7uV0G0vr89)+JSNJS)lS{VO{g2+ufdDB=5Yj;&Gb7CvSIiVrSrn<kI@x2td+f
zjKuu&h+nKEM3K4bX}770kAgUAuEX^)rbFr)aR_dFyo9_}2nkuH;8VK9eCo>`r)tYi
zv#pMsZGl-0s1LWY97RtrKjW?tp_Ds*K((^qM`z5V_OKL}y!Fu5wq%KXEW+)y2laEo
z^#o7d=>Z1ybP2Kw7VmwdqgsikNyg4V4{K8pDJ|PYA2wxw?E-pO@x9Yclt0^BY^yeA
z2~U6^5P(PZc+3&VGGGt_j_=-ZAqTe8*{}57)3D82tjag#@w##cqo}^uv3ebAYisI6
zh}jpuUL}9%nFbLsEzJ^^lBv;JU&7u4P#W>mW_)g5-Zl_{K{x@p5yOZGaW<R60C%<q
z;E(Ja(pQ#6Zp<hGDY2%I$L-hx3Cc70l7wMO-T2V^@CD95f0G)mr4OondfMY5)K`^{
z|K85^AjEz!gpSvVe<!FNmcmf{p^0sUrkkU;kYV<S)mKA9T_d`J2R1?oYt;4wN&)aY
zNnnEiX-W$97AMJPdCna1rOcO=vFL4LZpX0J0lHc1>S1wh9BSTeCMO?ONjw&*c@-=R
zE0vNt-bciSN=V24w28>}w{+(P2WBtY9~Hk~pu~$sFIhh^qYH<uo)~pj5cxkOly<0(
zx?ZuXmSu`WV^Xka6=LfAyT{~^;1FMI9+{tp`pr7&*ia;c-B;46ks~zXX)F7k4`kFI
zyj!*Ys^1t)@aj8-R9sCSMVyB`=wEr39STzmzeKzyYvt<ENFDk43X~9Xx8oHs0PAIt
zTTXh+GI4h@2qRKM$-VoBY582)Jg#7OW`up<*7j6qT>{qwK99$Sy;ha>uRDi@>_9(e
ztRcjwJaSx=tySX47tG_;#e>Gh$^BLJUsL)2`xtah12p1Znbm>SU10c({H6TVp`4d&
zm>}9P|2oCP?0(}kZ)mM`BQ}!S*e693E_5GRzmwns@PgH%c7Mkl`%^=OMzq%5QI#Oh
zfM_lTz0}oYmv;UAyMrJlR5^k-I2flJ5$(D<Y97D&o#Ym*V>q*i7gt2wj(;Q$K$Ioh
z5@}yno--BmV<{Tr#YWLC|6%$7BMGmMn98qJ(uuY2duO@IGt`ojPIesMO)H81ML>J6
zipsm>qw>|}Wr1k-yBfEr<JykbC)K9I6qOvAB=Sy<MW*iryzR<ZvZ<~0Mp<H5yDa!E
zknSj#?wnJQIj}Dj-k<H=@Fz1rtk^wB4>0NS2z*xaq(26|LV%9Tn$H<VNay8O4!av)
z>^qrK+zcr~#YQFncwKXxu-CY7cO$`t4o3uR)L<9X4MQG+)37_KES-y@I08#<K}tNp
z%7G({a9d%ErUOwSsC(J9N<M)1h+m?0L;P}Y%rH&Rvj_Z705Fo2l!Q?b67LZCUA@f^
za&JibTBmv4{DBA9r*pvpgpTRclT1ip1#Kx{q_DR6ktLs3RD_ThLxFR16M@4v&b%<n
zuGa$rO3-aXc|rsPe238$?4*2txIHNJz2plkM9sS+DkOVjZW`y-t%KWf+bS<`t7jjU
z;NvCqxm}Aw)q-^=q#3{s4=jkaHijD`pPc>4w`>F<`ZW!x@4)7OjpV-`v0xEkQX2j=
zD^5it-+beZxw(;%(R(l-T|7A_qokCIS&oqYGKhX=MZlZ8@j;TQFFcJNeL|l0=kpSS
zzSVT28!V_ps>qFvxP3hAj9wJ5umBrRgrQsuV7{<Jm!k8@U;zOKu()rvKVy8!8QZ#e
z6p<T-73%9I7{t%G^}(L{ZyDN=A--M+Z@(&PqEU__K5Y#nQ)kBm#d-Au?OsuyJ#0<e
zpF#;fk<cuIJqN7mV-L}Cwn<)VJ<<4iun(-+iaUN04?&{g|Au>>oofTc)p$}`3pR#P
z!EuZWxVP?p6iVkd?@fMA{jYP}KmFmqP2%RTg71QsMUoh`(yDDBcZhHPhQ+k)$Ng?z
z<twYsTj70NWR&Eq<sO$G*xYq%jl{)w!}$ahN64t;e~9pg<Zp~ya`!}i%P6E+%9bi1
z!4KI>)v2lWI(lrpzeTJ{&Q|Y!sYk^U(WRD*N}4T~ooiR|;B_F&GsDWmln#u-k4Z10
zSvQ39%}VmK{b?L%ke-G_p`f^4x)>qRmfJuw?BmS_&Kf7ng=!8lm8r+mwtvd#$Hr{}
zbY8d~TC&$5nIpq!X{nr~DB&7Gt(R{5Gyk4Q8<yiObPOZh5R{wNza^_`@YalNcn7+d
zv*}Mhcc{M+s<GhazlBNIlAd|~DO4)ntPv|Mq%&@`nF@OmWpqcW=76AKKO_^K`=!mB
zu*@=hYbwM01QZO}kEi=qCLRgbE{vNeyvci)e7cM^@}oE~CPbs4`mgHF`6LFv*lhQ7
zkZ{$DJ`y77yAZMnrwZ0-AgP7N_&a!fN9;4SyNAO>{{2-a9u%MVv;;xzpx*qK+zxw6
ztx^-jWP$c*4!9od+I4Yo$%FX=<OF7CGl6XppfDiVg{K-GF-S6nO_ENFqGmjqpD1|5
zf)O+Dpux@&TpC~>2edk%hKx*4gR%C@fh3{ROHcoRfd5N^f}xQ?u4UN{C>oz<cXPnG
zw3Eh*QUL<i@;T-%D8FQe-M(wfnA_j<oMj93#JTB>&jMF)R{8`(>L8dlL6L=719E-9
z&<B2>Ck}fG8tbqM`+E*%4Adr56_#LP526N~SMZuZs|3sA{CsT5=TP&&9E-gd0Xv<X
zgE4b91dsPm*QT8;|1g*gdEEV!*W%&32jf;~&e#XiRbMEg<Z}C(e5eK4ePethf2lNx
z0WJg!7vW?eJfv-A(f%@{P?n3pLPG(572sD|g&B5&lrEH<2e(iw@t$O;56fnM40|wa
zB7<edzi^D=q}uU|s#A5oUFcp?1;%K!sKm@rCuniDoZPf%eV;k%z#mj|=2ON;u)Ekk
zwyTH*Qe(mesfZdFDk1TH*^;8C%bZ}y=3EHh8h?%69rkJYHrljd@ZyfmzqWV(@1vmO
zG2?Af3E_+Eqp=_A?W^K{19C=xaVgyH-g@BlDTC#El?Q)*_Tf^gNcVEk=$N`q`;Y1+
z9M(p+;z*irvweq&C$<xeN4g`!Pt`mak&^X%-FL%Rdps^Lc|^{B-^2-PFSRGX?d_`J
zx#QLOxX}8Ww16VD*AdsC<IVbcN2pDKvZQi?zuwCbrgd+=e#Fw1p@KDa<99FyH`%X~
z<Kf7sJYn5K(XuEi+cx{%G3aG3$^{fv<~nP+mg$+&ug`mZM;mt8eoxtM6rF$TxZOQ7
zVC*)2+PZtNSHb@Dl6>dNw<hln_O|?_YK5igNQaWuxyrdJipufN&eJhVo1c>+Yq{^6
zGNj8cawAhn__<vcwKq(zQWPu}Fs~;RPydz|XmK5qJj$}~eO{tjC>p_+jejs%NGi)w
z)}SOBWIZco+K}wj8NOLt;T$2I?x;VqsXxB?RBkcRWh<T<De&*8P+oMqB`1HwXm|Fu
z7O^Act=5J6mk?E%8QEnyJGcAG_{FvL=DK;EubZ`{kU4c!KG&_eQEdBJ@!WiEeOgUI
zKgcJ6t_#l0YqmBRZUAS1eb~nggw14iH5rIKc}(%ZMP6})g%BGd9dJLNn$3o?fj}BP
z&W|66iuOU}I_vbhkcnZ}LMUMZGr+EqOED6tqtHzu`Zz}LRb;^i@eNq<fS5r;18lm&
zWmQ-R5#*lx4X1Rl3<soW-bN5ycE6;jBUer-YHE^z-V^=>vY#p|Pd;KXBjKTLxf0;{
zkP!I3^d$Tt-E?!^gG(3**W2tpRv+iYIjDzbcRS~8UM4VERVG43{atMj%nb;!0!G(F
z@heOm5>8{FalpB2ou4~tycUDyHiT^f{RiA%U~><^F|Q)rMKI>U#iRlby#riRzpL+X
zyT&0s@xjV!Qt`rK;`AAwZKgvNCT~(`y~%oEM(lJ!92}&>pr%0tOkMF*%%^li4js?I
zNk*0Mak|Qc{?NKZMi;k_BpEkn+8a8$mwzoBFb{eCOoxfKGdXpAMdn(_?nl_l65i2D
z%CMV^e&%-P()tQV;wJneka@w@mAx&F>Vs5M2!BCX$0{(v3TZjN#FkGw&B6=oaos~n
z7KPlFP8(sLeqO9eR9W=T4tE3C6F$E1`%loaH9m;3g&I58wPivfhyfD2k}`)WQxg$J
zvA=#DzEl3!BQ;XhtkPc6k>idZ%H_l#t)peT$<rBu9iyMiogK!Mp*j3JOD9FtbEP!;
z{Tu?iPJL&~IElGmu?Z4o9|)4K7P^h8syjOE(xhcQXqlb4rd}D$GqjTYkk(j+art%C
ziJCDY`J<PcElna`v^!Q&f?B`K^8t~p2loq$UfOz~BWdz$?ibV4cl}CPyD`;RTkIq8
zwh?XJ<*Ojd2MZ@w%Huow5=s60-v;V03UUIvTf#}Rh^@WvS8y!$^dzW}3zUDY1_`FB
zlGyR~`zAQ0cx%2E=M$N2+2~A?jN*22g<{if2Xp2x`a8wFm^pZ!-qF*XZnhkBI2gBX
z==J?URZw5z=N<9-&6vIDe(C2wRhqWbG|GI1+Wk~x)=SN^q8+~YMAeHc%e)x0Q>Dqw
zUdL$vA6M@kk7XbBkBdYSDoGJ3D}>C;E<3V9$jZvzdsjsE%3dLRQ})cBSF*Er_R74B
z?{RhC_w#$6Uw_oAA)M!D9LM{3k1^eALz8YxtCwyzOdOC-ouqW$&3%PSl)0NtS(HMl
z!Dum4Wz(nnYygk1O6}l>-uU)LB!`Q#ZlPdf6S9S>CXQ1P6D+u7b4!AHmni3oqxjU+
zt`>{LI~^OUD->voO|WoXpV+-Kk*2f4fMqxCLiVUourtH+ly<`p&L)+Qj+VNoNIr$y
z0999!2Nk*%vzlg}CsY*O<Bf6YK^MtC%kCHL4X-W_VA*L*dh{uUHHOus8k31A5O)PE
zg58!|0YH}EP|ODRGVp~UYYzyu5YCjHlOv2K3O6s<Y!^7z2EdI4ggDsxCx>l9i&&J0
z7-&%s7<mJq2uLu}N3TDfz2klS_?-?Z7|#Om2Xl5HU_muqsZqf6BQ~Q0+D6!BXsQeU
zZZ|uc3kP)MVpV-m&;tPlZTfx53qBNXbl{5{G+vCqY%5kIJsm!C9jmo;+Vf&fF4fWf
zz%2QMV~PCNQ7_MKq<4M+y7E0-Advz%3*nc*;M0No7mPoiKSxvkKt&YG=V<Y8N=BhG
zDii=C;glyJJqE$tP0-^x)oKAr6Y6Q`EMXD@5j9P$QnPZ`5Y5+Q5LkB0L~Gt6+P((2
zjY6In;^1$vI7gLX6w2o$RDX;uyX+lq1Yk50Ja=!Cl0u>>nCf=j4+SrrC9CYC`!B>L
z(v)JN^KJW?l!~~2$u->k?C3`P9IIOQfl9>E&nz5vPqW`E9|9p^)2yrnk+I4PWNM)9
zK`#JbeR^d~Csy}@4HPF(BzS~8b;%*bXyW)ipgW-NeuzgD9J`j4*uc@4(QbEV4q<<7
z7oWfKe*%_nfw$wd3h*S=43d+WGZ}>8K!=DD`(O!MjuiR!>mC2ZOC#P?3TUV+H(C1P
zEF+WDD48>ZXrj4x22qneoQ#W!FMLMy?>BAzruN*+R(Bc>3fOWDp{!I?x%(x9p<%q0
zVI@cE!+pq8SDs1Zos}SNwo#Ip`hn0sK_%DDxO+~FBBPVsKGgYfM2v=aAr%u(PuvU%
z2`9_mOXtW9_FN?y_+W(VJSk6YdgWrw=0UPeGJA*7oz?yIr8XSGu)bg|x+`Oe8kDj2
zD~ZpTzkg$v+$*Xb?n!lKPgvM_TqcC#P$z7m`q4hSs?x9Es*~5Let&fAJtDDw;>l9r
z1AoRRUz}-73pILHOx1B~>ZD$n=&x%FzTIi^9o_2YtvkzIp2M5LHmx`iakQcKkxn@k
z(Vq91YC3t{<Wh%PIkw?0I40Pr*t8g`D`3O#=;<V!xGYyN*T~YGIdpXMyY$AqBIP>0
zO7zXapW9WA9#N|<L)(Sg>T)lWH(Y5B+*W!Y3>u^ij_=J_oGtsttDgVRE}>``YtK2D
zLQbeO+WMC7dabV>?VKi#A024x$ei9eJGrv8<;maJFXZ+yGqY{H#@|)Yg_}_XL+7RE
zhDXWyHdfNf9xsKPO)I(2E7iTQ>Ppv^mPiS%M}abi`XKiK{N7JAm5Cnq?f&{$EL@~8
z)TAW|fbI>bIar{|-*bTqlooJOAnbq^`Zw6=r>1`Oq5@Y3=iN7G*B`;;CD3Ls_>ftE
z8B|nmrcI@W^J9-+m<2G<gk22o6=)AexEujPveR6uTuOrM*isPq14(z;F0Rsc30M(@
zKc=QapsofLH!yGX)!sk{t6MBtta|Yx?v>9xI$uf1X=6@*oD)Akxm0q;Nb%f=>*2RH
z?*w#aN{en7uoMgp=~OhVv0<mW2_nO3q&v&PT=&;>HpVL@!$pf#MTstgn&gQyKitdk
zRRR1pw+-$l(6It%0TM7iWEg|!hTi`*mR*X{&j$*}9I~-n-|uNGc`?3vN+<#&t8|@t
zjKDZnXOAhlC9-~wuKh$GozJT-m3KOqcEMQ%l<FX7DY?gQUimwRAjRN{{@g>+`XKM9
z#s`w5RkF+xsTpj~?e0r8EToXK+%ut_^s!nNuzt7t<VG<wtqs>ot_hHq05>aGg^8FB
zDU5kRJP3C;_E%%FPU9{`h43GJ))nJf;eLurxHoqy#7huO%r?SNM<N7x;ZL<iMVL$w
zYqA$7ijqkFg+-ZF(HTf_oTig2IyF=!OwruLNsRbS2$n&+f5Np5-(!5mdCy;Ahg}~M
zwDV(ff_iZFSz&yKp<qoM@}$FrL1a?#dONF({)Nd?9O*24m(X;*d|bo(c$gH?iDKrZ
zawBi<CoOuqA8RkK3H~}`RUA;=t<BuxA)c(w^213UZA_X$5ImbbXVU&&?zo;NlsI!#
zpmSy`%{^7IH&Ba9*nIs2`Hs6Rhd<sRA60HKx7}2m)i!EHEgu)%|8(M<R2x;jXE!|6
za!^-OMZEFlpy-TDd$Aa+e-JZ%<Z$if2KZqQan<h$$9`M0f2!Dack1MtW5V)60m+ct
z30dBeWYmb{CwvRc8P<1mJ~P-x_3K(^8(fixMxUH0#?NqP8pvv%c?DF=z1y@_=NIQt
zXeRL|+NYQhw3@qh)EA^Eq=CB0J9&%6Ai47Xz=x2VnJWdS)SiwD6Y1Mh7glmG=@l`j
zS5WnB_M=QAnhSabQJii&mcv%a0n#mrEZ5VR+V#BC?y2IJJ=X)oI<oNmxg|A!&t7lK
zfZI{GBb)|3U4%>Q?@00mm<a(Tn<a*ioB;41ppIqsqwymQv`Moey21n&8}Lv-Q&VO(
z`u-j+7$1-Wi5%`)V-u6vQNv`Khz=M#K#!OKw(g)s!Ti)x4F5Z93NZip>^8N(zi;F7
z3VsXsBQ(z-8GJ>O0yk|cJ=pg{h51B*7$SyXr2p`;EN$pb;1T$3Hk{!*30^f3(%*Zb
z7IyLZlC*55<{j{KdwUNTkSwt6MJsx;^79w`84I2ex0JL4R0EE<Oeb(*!vqs-*rD5g
zO)9PyI=~SGmGO6Y5l`O2;1#ZEq`(1rD*bi54^mTX*I7?dsu({%X%Oof6ZyVYqbP7D
z=ep}Ha$~DRp(MPf0Xvahr-{Y0V=(t6aEb&31OOF-Fa0kwBW&kfB258?oN$srPBRQo
z(R$5$KaArW9zFApz^r{5^lGr@OQ>$PhQt6^a`nvER$X>|57oOq{TCnIn1;y-J}Z<6
z$G~v1U0zPX<Agwk)sTc;=h)T-OWoaI^8uB?Pm{Z!JbaavX=0zl=olh4z~&kstDggA
zm72p{g9+$hXWkLT2nnK#i>?gW6s5MTY|mcAH7L+1CYH_x8Y{JH<tUS#ws7>Tr<Vzw
zDiX|NofJAwSWRk`|BD*6w7jMFUofn&Z8^&O@~?~g4LAQFniMIvI3F9lsQ`=UyV3ov
zbb?M;+Q)0Z`m19L-d)y;zFWB3Gn)0&-szLH#mTHS5}adN3&KP90utR9=ECn4Rg_>J
z4Cp#!WNnwO3*;FETJ5Iq(JOw})Eex|93mIHU>5F89W|u)@^mgDkQUh^9%ICej51my
zeYaNf1e-!_hE65v)rR`*T3@;|bAbty7%}PiBAL6v^@pT#<njtRycOo_Hkdcmopn#S
zk`6m;^NzoN3f(8s_0vBW`dooK%za(ZS^5c;yt3mf%jYyi85>G0Ru{8O5zpqhNegFr
z>x%59^yD?XP37z<ChO2O9VCo{*C61REGiMuSr+id_U?4*v%z2i1ge!gkZneAm$w;m
zxY6nf7{D2XnKr)12ABpkg`LVeaT$G32a$wA-S{`L>}7Aj^@}M9v7esPLS^*V!+8$a
z`}(D$HXnvyfkJU<M#y72!~{zjJeyRRAhjIKZ85=w4mkCnF9^JO7~M_#eiWCUC=bUT
z_tjwdy>E#xz773Zs#Ci@4fPhVx1n33z79<q#Jq;|FW7*I9C{C_l*p(*uG{}*3L6%9
z&~OZcq6t7$P<#MXqgOuGr3;P?G{9HDegDL+|7EC5yW%l5vwY{X8}BBrn@Kh_rLjtT
z58M4_zbE~Sx*B(##GqQ3S-ZJF5@4cFw%f3i;xd4V4+C=ek6};(p@3*?YG6RAq>C8)
zji2Cr%{!y6{^^|u-ZlEJuD=~KwqGiRehi*1uUsa(oyxOGd*_<*RdK0mU1FRnkDU)6
z-WRYX$+5=W5h}bt`g+rYRl3?dBVS%ZTpXB7Oe__!UsCA}s8pg1L={BCzr6Tx8*}>=
zoyp^X*KvOs=GwKcY2He~byjlG;0-EuBdroomcUUIk9lDbzx1tSDUzE-fB4^ct^a<R
zb>isFnNL1DmOT-?rNV5=V~l!1{h{$p<iY%|7^{p1!3SrV!6y%U=gfXm?_fvPA1!Go
zIODMY4yHZa-Bb6=u&HsizLARISv#Zb(kA@@TLz*I^%wY=^B-Zm_FP5X=JAts$Y5Fb
z6qi7J;^lg<(Vu@sP1AnXzv>3t*GTg;dB^_RaZl;FfkuS7_HWV1?|Enaa>Wiwm`rs1
z&Xn4o9(-e^xoUeP!zHRvAAXwcO&nq7<1!+8GEq-yFq!=UBk9?cVUs$~M(HwY-5bJ(
zR>HwTGx`yDi5$K0*o~k1w0?H+={=GF>rsQ~#~=*RA%#u@*w2txfL3P%{0^KebTa`I
zNpE2)2#<t!X^`?LO!CM0cp!2KN+cLwD=6OtmrM|J4-dom9Kx+3#0sqa0R#kk7WiBO
z(;i@c_!BtQLDB(uKL)79Lcq1XJC@zoj<=`&993Tm*eHx0ph|+t6r`p3i9fjHL@3hK
z)`0(x4==d>t)m+mukIEj_m`Upj>{1)m2pgAlAn$5Xca<g9xi$00@v;mjOftGUu|>n
z=>aV@9WDTZ=Qe8UoI<7_Mi}iS&n+!kfd5u8VF8~CW*|?a3;;x59y9&rrItxO?);X>
zN*VjNt&SUe4VSWyMvO_c5Q}BnE_934sue~LeW*U=lNhMcMsz^Ix$VTikXK@bcn-l`
z0CQ>%OGhgD^nM9#(x#+V(_tf~)HiS;x!@aLH-1s@p{Rn&)9WAS^@+<Pisx}l??1(0
z@bWmuars>Q&Wu-v>)HePvG&GOSDlrox$8f%Ni*JAd^+Px9$9lRvoRj-`%1rv7GcyT
ztaH2bCnKZ9#=%aAsSDo7r_b>*h+ck`n@;3wA^Z0Ot_zmq_+#TYi8knur-U)jKg_s-
zFx%gWh)RjQSd+996W&2=QY^`<E&5XbV#8)7g3<23w(8$%Y?7Qd;DgQJz;=vKxsT`B
z3Z?C7XPM#(3s=%?q^9p}a=+b3-=x=8Up^_bWm0{Q7CHq9<or%c59|~ExyO_nA=}+o
zTze0Xhx_l!Y-OJB-RJYAEqZqU{J?YnGg&%M?A;w!?pcYXSM{|AKGM$H)9sg0aZ3k%
zvj>TN2yxpsZ6miYhC(XJL$hIW{dm^Y;!1>86RE8a+Da15HeaUL@GvgP9%TOb^}=^Y
zBhU!n2ocd&?+l9{Tj~!MWrOwJ;Gy7tv0m@h`nkKrj_aw5%b6k9KH#aBvD%E8BVNi>
z^B(-i1<3on;zt);$+i7Mgu4Df;N)P!vDLk;u#Uv6(e~#-CHaBhaoq@7xP-R-4Ca0L
z>k-wKpPXB&SqoR9I!A4hMI0^E=%|fqLJTnluu=dxjaW=)d$tAY;(NFTUE+xFcNz3}
zW961@96dJaU0?G7UN%x}4Cz0JH@>oDg*e>u71G#nteU+lHOT7!;uvEch6V@hAfp*~
zoLZ6|#``2Vm{BelbHAOE!OzI^-aD%IDYDACu?r^OmWxf|0P~k{_Q*JA$4_JW^?Xbq
zF2fH#(rvc~rq%_kHK>VAW|d7)NCV?D_hmi<vhhzsS4p>k<LP71le#g(z{m$Afn3-8
zZH)7LoVy)}>NmUJBUtobYS{|-Go{h90k#1*4Va1$;eO7cNdcn=>Sy<%zxwgx2@?s7
zp26<3QsbT4Gn_(Nl##JTLwd!J<C2(bd=&j1s^>)E^bw5ueo|^b!@VWG_LGIYd(9Ts
zJC|@Q&O~wr-=WrrJWqdTx{K6DO3pJOu1YON@qFxOb$~(X7;@}ne&>=1R_zlU;Nr+;
z!i@+ZhMi{eugA)r^bzR6GPn#tWd$uEhL_ge{LjVQ?9V@tf{~`^Q+j28Kg{=PE{E4K
z#E6aX2H6pU?>P3xbDXdccO_1~E8chzxFD)EHYj88Wq#d#;kz_)Ea5-;<iEexoH}{f
z#RO&~ZJnG_YBzn0MhY#QX^juHWY+&l#t&)@;q@ru?!-AiI8e~{WARua5k$>6PTNy*
z`~H43=Zd6ctl5+Ka+GQ4SghknnR+j@yWZMiYzoFTLEpX^+MTx%9whYH(+P9>-+M^U
z{*}{y%AraLx7#vPA+B#E14B%fJ#Hi6wV(bi`gb@K_gqlxxNjYO=sAuj4|HLL?VcU5
z!=EV$oH?lxcS6AYpZ&3Hx8GDY^{BhWgzg`=2t8Hn&<py@`D%TZMtShI5UP}YL;YmE
ze_Wk3-Jo@IGJX)fw6~AzNWm#Hv){ROn2%hdj){7?oo?7pY;g6Q|3>}!xIZ_#qbrRC
z=w|M#p!=J;$s60p!-A)~tA&XqSaKX(zD>bR+n18sm$-&e@AWDyS#0Ct<mbajZs1=6
zg(kRuK)z6<R3sA?_66WdrO2ZR2O7*0KsSn%4YRuQb?+C^++%{FoWnPRVKSLuaP?K|
z%d7p_t4ge6G(Y3XBsiY3kUW;558S)Qm|oxYBn$IAXN>h~;sUoj#@jCJnU9~Fj~r|A
zOh2+ZW~JI$qGO-_VQHeVCMN;K5l)#1G}gqx0MPueEb(NdsKZFi0-jBg>JxUEpDpCq
zq6tTdNIPk_I%KaOaRdiG)td_Q9ZD9e#g_Y0aAIrk&K?=poBrtP!V$oMOSZ56Jsf)<
z_~?XoCC2jymZm_4=p*&AgmTmeVJfJcY$rDiAxI9Q<cdQFIfoARcB*F@_1KEu>~tAw
zH5sc^%q23){#r_@qZe{J5%UqB10r1n(o6IDD0>)vxdLaR42$Eg#HNPFq*%#8ukn|@
zRl)UDf}nH+M-;Dd^~e6kZ-aw|Y#bN``31V3{7*e{tQu)Llz23{ZJdda4aqO8v8`9f
z{A9D<VrJRZ{dz`nYdzpsY-LXT?U1zDR`=un-3&eSvMM{;%|2e1yX}s2eH;FFVcj1d
zda<2v8962sAS})$jb2iRzMc)Ibq6yZQz+8vrKA#ijz=8I0%wSwwNb+f7~s0f3++%L
zaNLu-tdVPDkdF6#rlc!s&YrK&K0hIZ4j0ou6)tAM<ido;^c;r2dS|d{0;~@RX&Z^o
zN)2DTF4hv1I3!B;_3AthXqvg_+o)=XAt!_F@)9bF>n~4maR<0G-`}HDpv8XrBsBQc
zpS-&ngA~~u!sE#NEYOhjS=2?wehvlYZwc%%J#UJ~)*&*U#4s~g(A2~PShaA`e--{h
z#Sno%Hp=(do5$epLw`&K&Znx|Y@Q=NznN|92g!9$u|~)k9%nES2|6vg#w`1=YL0T>
z_%EKS;xH>K>mCs<NVU5x2ya~v;orf9PHt&039W7|D@z)7ef9UqsF~o2k=mI{!opuF
zu^id(%hZ}#Nc0_bwTj^!X3-xtkfAdpyDMF6YAJ-_Iq<7}!o0_cICjtX(qTw1R<ljI
zz5gRVf_^?kZ-P(zZ9kzIJj&?a+_}oBO}@qPg)pfoiSKTfNRO|>0*smz+0<(Y?(Z5O
zCVH^~NbV<!jr^*OF56u_ds-Fo4C7Vp1wlc_3(?1nRv+nDXc0cG;&<*cjCmfBj+v_5
zpm2_B)B5O@>y=B=9oi0@E{lx|D4RP?hWsboSAm@JRsp2?E15zD^8%Yx#TIvaa|?K-
z@OLKNxiOl`$1y)(-kdt)+Bm2!bgQ$O;BDQg^lc0#Ao4Exp}%K-*?@%P)}2R+0r=qm
ztUwIGYY+kmE_Y!04l`RAU5}oDoyc|kOHd|)74Q7+@25Co`fMJT)i|HU-QH=zj7}FI
z&!G>0MAE<^|C8Z@FQMjO=z2QKiNDq0!yv`v67}ad*_O<hgMIsHitJ`+7LqfuMM24T
z5ffUf?uXc>ujtwFy&xnOfI7IfI<sVTzu#NzN(<m6pa#%;)GbN>aR?z<T<u4WFr5+l
zpQ|iCqL@C}9$y6K`cU-=j>C;BLT^qtf;g&jV-_ykiYVf6Ci!z4>i)Oqd0>$5cDkx9
zR4A+@*0L<bh#(K%vF)l9Rvy$_oBFqWl9iQGJ`(d}DU=ON4Cj+O|8a53`S3Oa7DW`Z
z5r6vT;oaiPIhyMnD1q*6*CVSja;@Ot-N8rIBSNv%+VUaIi3zkRZ{sOX`vnH_4#yua
zEgrkO<>g0)X#5mrieB=eOBN@-XpyAZwj=ovXWegbczep%_2_5WdZy>@Fx3@I$8Tj@
zbi1#N{v@ci40CwI2idD=cjO~ag7p;)v`xbMgrmsF_Y8*$9cfGp+~2R)k2ZLD*6FN#
z7M}Yx_B$rHSk8WDq`a={7>nWEs!E?nT(RL2Caj7E^r|?{6rwp;D?JoE+t%xq3lM(a
z^}*1G6w+6!?5T(IoA!J?RY~IV>W0JA>d#{40I{<afSbE(0f<Df4I#^R*)Gbq9Ik5+
z`im(56#zH_pkAEYW7}ZACh8~I*Z5h$m&N$AH$VAH4hHuNfqs~`ll{h)g}FHb1I6j-
z5{Gufa-vDZcqIsuN_h|zjZf}2Vk|u8SBvVp0wH+-zkWn6F&ak5C<3>%pu4N0VZeOA
zi0e~crznOGZCi|9?w9WidRe3r`0*Sj4%8y^;T2^0p+XUc5*TIs8?(}Kr`ly#*P2*4
zw65*nr`oBmQ#m~G=QGYV<45@w|1;S~SSI*c*$akIXV<9x)_9y4Y5Mo1*Rgit=VNOx
z-(MG;aen$a!Tj-~k21~dZl`?p&`@tR9%$Hiw@wHvq?MY!?GU|m;V}2+6{CGu3@b)9
zi}X(|XNFgnwoiMp3^_d%h(0PEAI@eDZS8HEAy&p8O!yv(qWFJ#U0}HDAlN6EoTTK>
z!lczeo_y==tMY(VliX|vp4%NsmG0j#@{*L>GJi4alr8maKdj*(ofj&o^hcUcN_p}&
zMLU@5Sl{cOeVyIh#oE(P)J^<LXba<@nqtl`y87dd&wDF;6}wTb4`jX|1SYI8%^j^a
zQ6Xio-ExaWOUf^F);QDeE+%fBe5JR$?{+Y*w|B}rZaQ2y<9y{xK=~yFVz}yci`&7@
zF-0C%Oaag)F$JIz0zTcpEZWZjw=YcBL2Wyp8kDE-+R@{q^p{uorEl+be6-+eN${C4
zr^|{ZR&=H=PkhXdd~){3>_&O>zSq?%J@+M}OOr9iDXbJB?0|c~gzm2j9U4n$);wZ9
zQC|i`OU*)8B=jzh!;IqxqdPx8wo?SYP0qX8TJm_u<bv2I9u4!^&~=EXlws{fc&=j!
z9W(02J`<ghbgp}2WY_9)*j`xIm>D(M%ePd}Y+8@eo%#RogY4}3=l#wo1gqE6Z<NK}
z`m^o%l6mpl$i)lsK?h|V)1fqXC_UcyslI19Z$h%-M;sK^VpgwzObF|`HQGNaSK8FE
zRG|93C&<3-kLId_jpG>kihxBN4cV)LkMl~Qy;gcX$?jX?b?0-pB!LgI)3USeD7eSO
zGU0M#keN<<Q`K81uD!6j%^>C8`Dw6e<?)hSVqY!I)7LVmOlL=(UE2(nNnfciW9H`M
zI?qJKI~zIDO^ua?SZ=s~@Kkl<)XaWH@vGiTTf_a1bKZ*>vCUKJfJxW)y9F)dM}wBv
zjB&MzFmI!?3*ko#wl`B3BBLhmrr1+R_cc&Tk-rBC+DuUt!WoR2fLVgmLg$yK44DeC
z4k<Ft&p3IB@b8`*z3V7?tA`>@)f8tSR#?N!&cRe}jdj0EnA!UH@wMKH-j!}3q3g3f
z)6;t}x^|6l6ZUC_F+|jqv}YOL0TC&TJJ3#|aY1S+xMKZJV##!WNU~R##7#U9a1zBS
zU4J^Brq&EV&li_4YHS6bQr}k18@@5!nQ~5bS*C8N$mH#&#QMeIO_Z-~-6|9z|NF=6
zNm~3R9L3B2bnTosvELm%rOY)PE^=|`QCJeJwGymfr+jl?aXMvFNGt-`!x_yPjWm>)
zMeO63&J3+?jZHeaFSU*iZtdKA5W(Z}fZ4=Eq}x0&VShCn*ZC0XE7<9Is(VG~yyYdl
zQQ93baa*3GHM02!yG!@+zF%f6<S!Af9h31~{G-P&?7+7fCxEKJRP&@2;fTihJhYJ7
zy1<`bC-$@<m%he!-%N|#uf*6arf_4@pB|SukbBJJmzbv4>U>Y{gDA;2f9uIcsHL`@
z$di#c@@06$7cT%ig(Sd<Kgot9HYW#Ux}hC!M(&HM-Sho1z@IfemU-VfVJ)})$Gmw%
z*3Tc}Qs0AK{fH^Nw59C*1YzIQ;-`UW_SSfZh2=CbMEfYomg7cS8;8srWReV6suGKV
z>Y;=Q3~Ava1rs2^dwg_&VgiHXziQvL$)~i!#&<X=ZGz&vn=g9lrG1sWX=kceA$m<h
zyd|dnew9ASq8I7=_T{1Q*ci9<K?zx7sW#ioXVY!T3n9NI8-1D8ZM&Hn{||c?&Jq?L
z&7~PHO+&kN?GvH>4SvCb>bisTRi@Gq@t=VXqa}|z)YqK+<^*KPh2KYTT#S=$PyhNv
zdQ7|1{GE;T^6_EhS?8nl=qrCRegt8d-@Bah<n5YuLXY3Nlb?#37Pa&cTlkwVqlUc+
z3kPk*t?>y6zkKRHyWM&eVG>h3ORnN2kC3i{$;iQg!@Uo-2>ztSB>}9h@oSy6v7Jk%
z&*Ozu%ckwIO4vUThFHDcWvx!VkW*-+Gt}+!lEsvcXsd3qi&VBlv`<M%bHT^{N-<OX
ze1V6cZHt?%zRj8}_0jZNY@zGf@!JqW$KJ(FZ2_x3*CeUNuk@<4IIVPusNSu$E!Jwe
zo{-ZX=iS4@8d)vU;L@_<cdrT$QbVeXGU9?IZ&u&9?D2TuN<-?(Z?Z{k9*?){uEDIK
z?Fvb}l=j*AdFNZ=Z$O*$=8mK!tvAu^UGUuj4DauGY(<{=(?mgm7z&}_IWWro%R2lH
zK!Z$APq$*)$a24XV$f}!KMX?lQNfN9<gQOYQ3!~jKH{_$#o+Q8Wpf}p#RzG-VzC=e
z<58KYq}csXFu8~HxwwqS92}xB#DH3#9uKBB7-MC{<`BpNFx3m?MFgG-{;@o-Fn-Hi
zcMo<EE#9AAW<|EH0W}J<iO*ktaU7XSzMw5syTwU<@LY($?1M(-M<n#DdpFXVfWiH-
zEyPl9E^p!FfYBnAOMcHMnT*PtNEjFWQ%$Ln|Mxuk82xO!cQ*+hrgh5~JhA<~wnC?=
z9HV?w_oZ^y3xmZkvh^Ej-^uXWG2U18wi)(Jw~P8%dU!I{vg>~Ra+UWDXOSLtM~4|j
za)!++Pd?dtBeB$o?~8Q_^Q_eLVxOyBe8z(!8p^lkylj6lubc?V=oz>N()7LBr)``1
zF-dn<+VD_Ko3f4VwnH!tf<KIRb-hPubJZ|Ubv~2sxI(X&=}dq5I{&SsjQQu4&wDwh
z{`|g5-;Zja+IAE|G<e%z>k?p$J(vmV_UK=#=Afp&A&k2n{8;Vbg20?6bA)XLsD%JB
zMTe}DkueVHgN%h7XY+MZ$HRiSVXQQ4yvIT<&##TK`WdzwgLzyK^=ElL4ch6!Ub2(p
zS36M`DulkS|K@8*K3gMmJ39`kU0Cgn4zM(;)gcv4+u(C%>H^bIpnt9Iy=2&=&D{he
zlboDX9R5I+XnOtY90Y2r!F+C<Ur-NKX(Y*1GyEb;Z(oVcqKIT)T-5sdNW^t&`hi9p
z-6aH<M>r+L<oBOd1TV{&^d0QkudKYe7FKES&uhY^YOW5J8$6uu=BbH$&Z_)0Y`x)i
zK|{*+fHT$T!hqJ@#Sj&yRDHDp@{fA^>3$iu^fXwS%B-jz!+b}ZOa2mA$-%#fQ%--$
zZj9L7OfkoLt#!gkN1<%su88GX?|`j6Y5TSZOXSvRm2{QHV!WoNlXB#PWscF3XQP7k
z=g1Aw_gHgtpUmPt`P{sbOs~A#P`{{Rv4|f9Q11QN*=ffuo1+-aG&xxb3=XIh3`bf`
zVd2CS@egQM`Y)8OT)tGJQ4}R(p1xpSa3AY(2YS_c0h9)+uJ|6^xHsvMpZ^2Ojf{*M
zN|*p^e)krIrF4(zb?b3D{hag8wX1DKm+_bAMZXij=M;V9HJamZE^J{Sp#1PgsVXgT
zO$<dl*Hrb=>iFJRa_Xc?5K=oRC30&;%j2Xog1`2hW>I3yh(OFZh=b@yt*Dz0W))%K
zVq>QB59O=!4DS+bR_dR5*6-n*EbZiKJJR{565)LSHf&6W+T|wN+*)F>M!~@)bp>M3
z!YtUtwzi4~yV@{y-?DxoHgG$Wq1nvWp)mQyw_ue=uT{qAwIbuhuXdHN2C?P1SU;Tg
z5_0{dUNH6K8=YwTYTMklt{RDd>g)(DJwvd0TF#v^y%b{#n^F9;x0GYW3JTWbI51oC
zwE5ULK)PUaxxUTUN2-}@HrLVpEOr0>e|z?0d2&I{_)p#lfjPYsL+#B^rY`+z6svb<
z;}$Q~{Q489^Mm+9>vT2&f*_Qc?MjeZ9R>j_f}?ja!nz=Eu;@!ctLB?gc_uvrX>v-b
zB6Ft%G8rL3WJ@`_(p!cQU*$dflfIu~fj0xzWCFF7<X@IB!M0TGmq!ZaY0Ic1ngwvG
zD$Z3cX$CNh$bH{0L|GejE|dy#-yE}cGB69QAFEET?V#3c=~wntQIa_NrnbuhA(;-|
z8*E=E&s5x&n@XMa>^&?g|5*omfISWsXreMGs*U<mrEB7!G1wNq3IT7%!TR%2HD!xJ
zgOb!$-4p4y@qkxLdoGnTK4i=$>y=R@e`EwUbbBlUXLg-OPewPd=MzA7Vb~$-;2<Y{
zvqUzYZ=Zmc$pb}|0Gb<V-ronC=S1j@639vQ#Z9{wDlY!d`KKX_P6S-cSiF4;zXQK(
z5#{TuN}jvOO;RSedX@p%c)sBe&t7*Cq<XL@)JOA#G1ToPDJwY(xY@S}*12Z+-n6$j
z)2K|~j=1>H|NDv$FYkTjH96VWwX&i183y*>%;HAh=g<!vu~-bWm>%UXX-!8KUljT9
z-TB=w0ToRN)%wFLOt);G2n0mcx?-}0hCYWw*+~;7E%yb&KJz8btsE6j7#_EU2sHWR
zMsG<Wadm{6b4*p8avb+7XMK~>JkDIo1RR7$pTB$FJ6|q%@S)54a`Z%waxOi|75<Yy
zU2Z2TQ>Zx8^7UU`I|kBD8!hHW6lC=*oxYPodwNv_YuH`$^G??$Hd9Jdq=ec2dn&V<
z!WSlmY)P_98qa^#`(AEBSoSz|+*7FeF<w&M`}J-aLC{WLn|8tu|M1rVTV$(pp8J;d
zPKKX$|6%h9L7mecTTNW+)W#^1(rohb6GhgBuJWn(+ODSVn1k%RSTbUY#4iV0Vy+R`
zSn;>uyg|;(MQN)Y)vP08_LXFq<WQ%KBroRpP(70b@~kpUZ7Fl_d|5Mbe_NQC%uyCp
zMg1VYpw>~wta`NQ$4T_lWkz=~tkz3Xk}HV3t+9kjNm;D6UB~&B?eppP^(m8*G)?LT
zV_aacK<GH-&RumjRQHYf>*jI|1(i1<A|v<k&+z#L`a3&T_t%N`b1rh=oN}Jku#nOo
z&4?%sa?f3jzWiy@Ku9bBF#287&H6L@;abPJOV7I&3-~NV6tqyA%{Oz0v5LA&<W7Ox
ziSzePMgO!(Ta{YEFw}5!@LLVX9U&Hdg`3#x*XtHhoszS8+LX_~9Cr44-NK(;!SwKO
zsj0utI~v?1nbKY$dOaj(r~1m$wV!*^Lj*q^$d7HETcy)GZBrQWZ0{?J@>{SDk`Qbk
z83d@RDvHR_inxBs**SDdX{c*PsQdUJbI(jJUoyt4ntS$2#4zM`EO=6~-9B62x)z+?
zABo3>UC#MN_29=m;m&I5t?A)jPxq<D8P{6PaZQGwh`|p%n#kiS!;!i<jCzNbR7V-k
zSa1FQ-)N#2C@Ij!>kzO~w@}n(O|IoMgAFuew1rKSLD+LzV+)HnR#u^qBL!>>`Pdmn
zu1@+KVe0FZqAGTEiF}Sb3oM?oEN(Mbj4at?%FW}^mWkjz2RyMEs~F&LKn<j+sfkua
z?e9B4*$lq{*E@ir!Cg-^O3d9|pbXAS2AG#Z(6!cVF0t^53NhZWjX}k}{jg^F%$;%i
z%mSOKzw)akA&32+2h$@YWUolHJM-(z;RN5k^K`bOk?PKhE~0Gx3?pO2h=OHQBY48q
zv9HnyyKkj*jQQBjvey3XzjYBV)udK{iHVxV4Fgx*KQ(w+&Ya2`*YYd%m#Q{W=%#)g
zt0_~(iM;DX*<2l4ln`(~$;u>rTU{7~OpuH*jlZL&s(u=Epi*XDS9NeN9><6A;6p9N
znMDuYWdBdQ$zygbwzoJTiPMaQ8*~Prkm{&2HTI+abZK1I-ry{$sFn~{xvhGwi8j?@
z)k^IPuP~{G202T=WCW|n8TS1qWQ!EdZV%Q^2pL-(xV0GDmFRhB8tAr{e8hJyadc?T
zXq<Q~ze#^RTqdna(|5GYoeUzC?3UG&uPQP`f(hLhpx^`6UgVB)036A5C=Y^5VSX!g
z)FCGsNOPDkcSgeJxB<>3J9M5qbkWl$8D-$$4Bj8U=3|z;)bV*}R~-KR5j_}ggAV#|
zo=)f)b2Qrhf4<xkRXz%#DPUvrrp5oJouebW=cF5j)7Axx`cnt^B;Z1akB|SOua9cb
zQ21ll#x|vgucYRcD7v?53<`G#{JbBK-|mL$DNlpLcV*;A=r{d}IwR7P+j*nd7qZ~l
z5K+(~^eN2EICauSckDKXoH_1qj5f7iU7gx6$NQi2>nb7^xn0|Y)2I9)KF*MpPJ?ry
z%Jfbg=_3*+vn^`uRl6T0C68M<AMHh(32nyqH7`aq71nQkawVj|`fMgMQMVP3Lx^SN
zE`7tXJGMnLv2?rUgRFI2s*Q;B4Rv1$M7Y#$?t1C(K&3vCBo{3dS+TPfZeh-~T~u#X
zz0oUvb`HCjPcChx#M-LeFrANz8ZG4WTw6YTru8iF^{-n-6)|pMs53pB8<T9cLZ?C}
zeR2X9(>T4E|NM+L<Vil<X;a(hs^_4k8s*qIat9yJu$7haBrYr%+yQ6!HW?WttZyv~
z9deA7S?U3$3|O3D4<rtOWR3nchR2o`hk$&V(BqttxucQ=0+6gv4!6q(Ec&d;!E6^g
zQP`ue)YXm6%;pcb=QLc`zo9*xY$~6Fe!KKd&wX-&v{l|KB+S$MW0r<6fQ7&ZFfjzV
z#TiWvFO_#poZc9#DUFwe!_SeQ&61lstzM5imsL70@~KwW+?}9wsxCh&L$zC;ZK&1v
zriW{2-pxN@Eph57S)Tgz|Lv{tB(u2s(Q+~wWkk5F^;n>qM*j~xtt&0@aq7lfs8NBW
z&)o5qW~ig{A?AgEF=bkOuXkoiXV1?#hch#?+T%Ley_)bBuee#sj;Z5y*eTDbyy^M;
zYiWU!&~AIV<%%kCzF76e2eeg+#=+@?DG82DzS;L1_UwDyti<_NR*7S4Sb3Z4?+UdW
zWn0aNRVomW^EW&(h?9(Uv-xyI5S3)jyb)VpLronoUOwSK)|%bu?W3cV6QA(VXD-n=
zr&2G1(h}@d#T1In0fPnad@_0En=<`C#{utX2+;y+Hs5B=jW%oW++0dj<l&^Z+Q~4!
z6PE!Z<P77NL&>NLD}=^qKq@_&+YSKV93oO6v9dn9V10b)sq$c_JrFrEcifKL1cA<-
zo(?G)Xp*k&QZM%Fy_La8B$OKPFBnlpdE7#C5(>m1cb4LyYBqhL6nT*U>S8WQrK5&t
z|5$a3qo2I}pFHhh<@!Cx=a>K69=a|dJ+mGzoWOn}e^4|`*^pPzQfOS0nZ1zrt&1sQ
zDnZD1<B#Y<U92({*RVw8{Z~B+lB6>3riCrcw&ZnXIuCpd*k92XTbd<cbr56kVwz)4
zy?w+vpY|UYAap}VLY*<d#I!g5XivQM*6*6f(t7uvOmnuXBs^5%GS0DFRJp~RY9^<7
zv@cFAJXmwbT;JM0&?My6+q6pJQZMD-oy+%raPv0g^zs(inC0YrSES4{evo|hQ>x;3
zXA#lW&9|4N;dpxh!;n%#!os>^h8EpxutsBI8p_N8WdQ)SpQD&5;6=75D=9S*F65rP
zn?oomR~qC*830?WkK-x;tZ2J6xojm^@clsjZw(?Fg+TVJSFeJyHXedOATj~r@`oN6
zXOTs+{TxtSz*uzK2^s?!Se3Z!nE(wM30%z4DtokOmpGuM^NgwMdzty{w$s8fvDnjL
z`;F>NLEvdwk^4f%CPpwh&-)DHPhAF}TqrWHz8(tiw{wAw-LMzE5kGY;NaeP6yftUr
zVvLj;DU$&$u#8O5_}M$c>xmOm4zDjFvjua(=u>&HD6jPVN54Jz^&1C#>z`!3mHBF^
zMJ1_$gC#))r0(0}wn)t(9*h4vM|9h;ZNUl3;WP03vmSDfOa1(rh&OXi_2!mul=I7m
zw9lUzru#Ygb1*GIH;L8Fw^uxe$SXC+?`ri()Qt>C&t(u)7W2F^s2@R6)3~SAHpARC
z8!Sx}DzlRXf0L^e@u+xxmtlcx6v7B=Cp{X6I|Vqu_SV>yE@T8i0nuH;1b_+1Eh6aT
zN+6<Qu{*XYEksbNQX@l&L4hHX;b{y=1<Pb;BU0q)<-zM(Sq(JQXpu%6RiZ3BA9C3w
z{%6IidEO@FlP-*~ctHrf^HZAT8s+U2rxgT@!Bgo!;bVnlp!rgN)nk<!w8VHo43zcs
z^rWG-CHIB%t&-oqVqko{8)$^URidH@QF&hsLNh>AfsO^yF!)WMopNwYSV{lpa}$9K
z0xT;5<N-fC;IYD+5=;((;$)-#++(pfv8!`oGyN0U($M+#l76z#$5~~>r2D66?a5XN
zkAq$Qv*f834m=oKn3$RxTUsu`8*UcfLtsY+ZB~0F3vJ&!65g*dRV^!uP<td(^K)>w
zD$?WsyfE#^g4KZm4L1Z^qPsqEgXhYA*OR#&A<w#p8z=)80wfLG52Z{2pud(%AA}Of
z7p(b8m|)R@C<by5L5`}yoVIFVwKqfP83W{aIBD9zSfz+5AT!w`4lIY5x^yRNoYCjM
z*+f+&5Om?J0U9xgXZyf3iTXN_{a+{c@mjOpi$59p`iQ<dQk5bE-Jaef0*c!;+=~Ey
zjg<9(hh~Pcmes}xL{m$GhXl|Y;myI9H3IpoQ3lHUe^mu&$|P(>ijaQLz<>oP_=!0Q
zXu+PcCWjJmC42t(cV8O0fD1a4ueI%N-AT%UqEk9H%}8t5k1K1Fwed$+7E^v*s^4Cv
zG;lJ4Y*R7jXua*!7N`*w>4}}T8aKC^)~A}RGi1lPHLHVrV;{jYwY^<;<-eZ!Z(;sX
z8`IL^S0CT!!TZoJtdqCtK$Kxu352{XWz!`K_j5))Qb(s+C!Z}RqxFSAFCH3txlyw#
ztL3sxgMPLbF`@i?`Ris9zjHY3>EOG3t?&SVng`inAU_yiDSF`te2s!t7|%<kU(d;T
znS&&6d)jYbLqX_~ocv<6q2xKNqz4Zk0Mtx`e+e>cQo*DKG=tL!0(58$Sa4Lm{xFE9
zyMlHf?OP9v6CMolSQ)TYHLD#!atJ#KzAO+n0FDhv1HKoCDV1{;;C2Uine-A^nnr7T
z#DXG|ou4272|MlC4W#Xq(H*ixJwE}o(8;|bUId3w0jPcEV=!C>t}1v9z%^8`N^jNp
z7JoYF+zD4%`LaX~OI!K7|GBgO9wW?LFzY;#IQeoeBRF0fj`Tp4=DEc;v0d)Sh&50H
z<5l?bAoAek<d%?OU#i$i@8$^sd;{p#Vd^kw>T`X4GJ5%en+~9Ipa%n&z32V&^BoZh
z2<!KJ8x>xhrb9m_CVC!(@C!3U>e|_T0?9UT)R#9m4I#;NU})&)*jRvpnuha2BwV(@
zKZKd3q&+8h-9arx211JZ2&gq7<9e(Nf|cMUppAM!h6)5p5SD<=NLvVHrV>M<qkHh&
zYc&#&ddjX6dF5{<%tW~0AjT?pRdQ{8y=8SzcZ@5_04*Zl-<$#hWOg?T7UVsnHwxOt
zSe{-~LIMY#^ZWEIuY!vL>1i=#jq)}apV=%QMRYtr;{53LTjcOPmA@~6*4FB$j)ho(
zaxXFILbomfs}KM8f%ZOteGExln!}(p2R0`Rae(U%c{xI#3krhan}-<d0OZnx=&tQ)
zFT&Iu1~?9z|3YNxFZ{rpkNxpuNL(j^WpketAHv_px3c8n5`;Mu=*RydmaPuu&!KBM
zFj#_c7YMm*t*NwiK1WT3z*2{(V(N%zXa+s(U3f~Msrjp#jRxas@CS!j<3i^k(AYu`
z29^s;yOlRh<G@}XsKIjx$eiPf2eM|)i#-R({04sVg3WDZ<%c8z@ZdouzYMYU5c#{x
z3)(qv5uM;AVhS?)8+e4aJ@#ZVBmCki@(4?zoE)7oi=Is6s^PmC17v&qkv7>)-7}&i
zn^?Ee*GE;H<5pcLhJ1?u9@|Im&a(9@y?^SIXr}@h&};Is`>Uf!p(Ebm`ejO6;ei^g
zSIkn9Ts_e9!RO9l0HX)DX)H>RI-KokpGTrwP`JN$CN@2y?N^6p4p5MAeCSpu#U~6E
z+$AqqVLR}kZ5AR`*uWAATp`gRYT)LC37)JV11LmFE9%{VHU<PsaL9r86WknNYzHSB
zWPSoW5Y)Cg4CwvMjkdl35%CPH$bT5&&>us#XRM*a#@<8!_r`HR@P*e04-&G%rMkr6
zh=c(E<mKA&24xsSY6#kYr?st(v2+9;Eo}PjZR??YbqP;TA>huKjh1|X_pj2ct)e2K
zDI!1-l6gHm+yDJK;_KIMyDHk}RN99V+EkC(n%^HuS337`lDix)ks7X(acL~g-PkMN
zim25*`D#ijbhhDc=DKp*!@Yk!`V{rw9n{Pct;%$%;zE?SD*BHO4T$ry=?)z%XCH-!
zn#F<tZ7O}Fe-6-f&CMVAr&|5_t-#bZTeVs+w-|2;FWqXsJxuE65CpEo@HsLg1yM|J
z#6TV$I?C-$Ph3l1ngqBWU_?6WE!d1El4U%}R?-E}ASe)^dV;BUB&!)LS8^eb1bwy+
zxUd&&DzAHwmi0h?KWa8~wT2iwO*Jn;j{(dA#T1+%yDR3g9{S(|k|p>*9~^)dos^8M
zsmvU17r<=cfP;ex{vIkq2tIH2yMf-ys!R}t{f`#|aBaG57@LT({acs_Pf&zV99Ehq
z4&dbe#vQVr+Q@u(6K~CB*%+^q2?K{<om!CX30JhBDKZ#qLet>P1{pCgm6>LVZTyY?
z-}`1Z2zq3<KWUVpACVp78O~E<H($E`&y&0dJ3c<o!-gCk0aA(-41n|H>od2rqa^@Q
z7D1~39Wtm9AReI28UjlUK+Fax>C60AEpUI2K(&cONTCmX04$%~{^T=a8Cz3g*wh>{
z_-W5+BXr=ofhZQR8K_)EbM8~OMew75aN_PEB@P0Ir^QNN^htq60J(P>%-}+iVLUti
zT73h*co?q(a=<Qu*g_PL`DEmcAaTH6HXf7v!pW(wPR3IE=Q`9x0Ne{Ew?UvOEbjI7
zg<S1Z2AHToT7eE}<mem2q=Tx}ms=s{e=_sb&iVAO-(Z^B^7!6y?T?-R>&TGi-$X1E
zs6w~AQ}MdW%{DNAv^tspgrcA*c{gl9^YFjRS|d=EsrA=4Q|RgcxwnV!HAQ8_npa<l
zC|Kp-;sOvIj(v1!5TtpO=rmmdmA><zJ_*Lsdbn(d?ZH8#a<dKvd<ocv=s<O_i{yb?
zK1KecJ{x>f+1`?6=5q*wGJ~+LrRCLCJvb|&E{Dy9E|Gq(twGokAm;7AN9bwNyyxB;
z#C?(}eKE1HdjySM*H?xMqe|-}7XYOKoC6FE?pk>@g0u>1TrzG8kORWS0u4c)R&AV3
zWj725gMxy<5g!=FM`s81AqA`Ode~y0r{}Dj5l968*)A0N)w}6&O7`1>l7+ezQM_QR
z#jr7>B3OJKBRL~t_yGq}!z(Ap?N33-4|#1*=emaf_UJ#4%Sk0vs>GOU^pZ8(Yli49
zfC)g@*aF&ViD2@S($X9CQ`)!JR^gj9S5#L1kdp(uV|TF7BNFZwP$Ph`4{EbM8IFAk
zi8i&6c>M0b7k*5jk<7}@23&|~Oc&Ug<or&dX3VQT+GSW^Y?GrTaUb$b-oqfIs=9g<
z$~?FwM#|O&&y>2FYzr-018>8X0?}MjaNmMEC2TVgE1;95&rXie_q-(a_ZsJ2hTV2_
z20e?%_LVb;eu@RBqp8h8Z@ncP{U5)xp5q5vBegxfB;BzVgeew^QEJ%(yPgO5v&i_b
zL(1%T>^^3qa4(Ii{Ve_{=Qc09zdqI0{+oWfOvx^$ib3}8s`>Y)=GBTP!^PReHe7Q2
z`8xsH^yESW%U@Ae^_1UrAQFhYpn6CX)t<KXt?(g&U|{I`(m5`iQtR%I4n&ioo3on7
z4Gb1$z#e6q-<4+)&c5h#A%ev?m%-@Hq1(nIQoBM<A_Qvg2}X^UV^C^E+if!2WXdM1
z5Jg2S)8X$K2Rw(T(KEBMN#0Y(j}AxLrXNtox&#@&FQ|?`q4|3d;Jd+C3F>5qdYBxd
zU5N$<K|h)P;xZH_#2-LbtENIv3_P!KtEQ{tF01+c$G_9ad2GA~b5+ophZGd8FnN5>
zsjSSX6Prp8S28`ZF6g76b+MXlzKm|Q;P`CXM$mIrwP7`Y{9LWpg|o~Yb~lgn4uE`@
z(4Der=d_8<WSsd}WT<pM2dp3EQ4vTZ0SGUWBm$~n(ff`4azZ^#ghGKjq#B;5dv;$P
zt<JFyA|$C{M~sy*&ZMT&e>V9{{|OIU93_jv>fJb;_*AM)M3Y*SE%RsP>)S!SW;?fq
z_^J+eiqDsBY<x@F5j5v4L}&(%!a)6hzb<C7E?r#}!om$By)iDzw&2Q{7ZNyb%hCjt
z$CidH^$Tw*78V`l-+37*Ff-s2nAq7jPTva*5uwGZOdt!t8DNpcsiv1oKfPTsGpv7)
z1h4Xf8mrUs+>?gWnyF16pM?@Ke`h=Sn^td6Q7QG5L-`9oef5v{Pv+^0OmEdJ32a~|
z+*w!B8QYqVfakq|eR@4Kbj=*H&Hj%683t9M>iTPr{vC=7<2&F>0wyS+MvgMLM&Jw6
zyT8I1^cn{OB}fxh!J`&VDkTQx%W1TRvA2ERL5=MXus9e=NA$eec6Nu8{Y`JeGZ4MV
z)1xa;nH!qJ(87Yi2g<Osgne&-y;l`IQRw7ou-k%CB%%W<DmZexN??oXfI~AB)6e_S
z$uSVNeUE6`bqYlRje(fLmP;*DjV@K4VtpSxosQ4l*5tZ(A1MtEhLVAra`X}&V!9!1
zEEsc|?6=Nb;RyrbRriP+58{zymJzSyI>u2^^#%W(G?Ns!Tpld(J-r_mQ)UZ6cE)_s
zXi(uz49Oa~#loP7pt$(BNp`FF;jA`~+ZLAP!G>0HL`OKK11V)8O{9%iv2f$nmgoJo
zqx;JYY*tt-V-D<@9h~Xm@j}@-uJIaraT`^t-B)p@P?M#lSxFS-RZQ0_n{U0OZy2t-
zq<tJ7zap?3s9sX$aXR{RbbHRCa;kD>v^oZ$h3EYoe~}i@0zz*FYB<}Ln=0~}nmqu_
z^l=2ieFyD5dawYG4+OG1;9vrI4`82Gv-SN7YzALAi{OOjb6Csjrj9o&$F*n~oaecz
zEY{BfZ5nLHX5C_`w}SvA0Imnh@|4vr9k@rqj0SRMKZ>UT5CI`M3@J$&#!j-z35lzb
zDq1O!XVK4*ou3a$WY7ZwH1!tn11RJwK^q9^uU7zl(18d%kR(c_Pk;h{wAPioL%-zu
zh)h{<!pV8x4Uy)yR_)!Rm+}>gr^I6(`^0?fJ)i061@C4$9rGN^9vOnm%ir1;5WQJ*
zut9_}G}6<uE&>dx&URSNZC)XryB?KWt>%}@uJS<X{E_=RQ&X?V)2ga*ygbQ^8>eMW
z88S;u<dkhkLfRtBYj3MJY*0jX6aHc)?w9uo?X3RL2L@K;6Cv?Cpf>D{Fof*2lK96m
zW%3F59Zais@vV%vnv^`#k+E}=%&{He-x{AD?w3ylWM^Bmn4uh0zpdmw=^m=3w_Qrj
zx*qPw4YeKs2Iy4EV%5OEjsS*ptwGSRL3@TSaiE$(UzZXT&fYK-0=5k)*bgMif^QAn
zozNpfYg=Xxu^e8=B|b(;o8H5TtSp9{9H<qIjE&)vi-2Yqupbav-v?+SpaoD4$Txx)
z1rHuA^M^ZEnuaa5TdghGg!VSXQWhwc-u6*2W?bHO0+1E}3owy|R{{a9LFg|6EK2$y
zXfR2@6|g&fHZ@~T9yJ~=XA7BrIPRKQd#hTr5eT>96YFtZC2Vtt2SW+GPIXs07uC;P
z{u@F#q0s6f(X>8eK5dRW1^~6(`c?*^TNrx-^=&NN{TvYIR9G($N1szQb;KXDyM_lF
zvX?G>PPtHb_U$1iTf@!=dhE1|^*c8!&OIuQ-WXXX^YX;eeh^K{m$>6}s8jwxhjxxc
zGd3fhohKM&DrW6fXsa!loRezC#%i0;|8y)>tZ3P$EQTp^I>q;M4qg<CFVcg=VbftL
z%$a}VH0+~#uR|}A^;C+C>Z-2)ca}C$bQ7NlP5hrf=)`20%6CCQ<h<30<9xE+2Krn&
ziVzT_JdBirHXp1H&=)_<iRRJyXwc#?X;c(l*OKuOi+FTLfPxU@3T)L^#-B(l6AL%q
z14saE(@9SZ7&%n8=nV?a-&sJE$;`xotE*9oHlnf2JWuhaRz$L*^4@!@(fK9VI~m4o
zdad{XyfC;1h~CfBs0>GEF&ZX0`)q7j0DE=8*BLIKlo5U0j8&Hfw@4K7HvVt5Z+8qQ
zF}qQlw&xrd_>)zAehLI{r}C9{Q%I*ob&X_P)c^Nb+qU4goU92T$e(1cDo_!>tihaN
ztlL>{3mc}UX)j2{8mXtC<=k@Bf8(g0pnRNd@73U@f!R7Lc6&S`tJ?AVVvog#{`zyi
zai_V3skPe^zc7B-S7{|pPx|sY9UC7GZ1T9R=~->o;%T4zka;X0FE^N$EX?W$wZ+y8
zmVTSnf0r;<QQBR1Qcj?vef8V7=;bOivJ$n!vHA~_THLr~x2@$0RxKD?tVZsJzEN9>
zZ29&T5g3^Be0%4ldT7CuF#^#wPne&pQkVgr8vr`ZAz<ESyEXk3atpx61GdC&;-;G~
zn3-=k5_*I_x%YN>6`&1BRec798FW-HVi$xs(>J@Vx|eJXgJpW@`t1{8`<acQVcEaJ
zPaR0|hI$+QAz)bm*H|z#w-kl9H2dKb0Coy(L02?O{dDyntvdg!D6#utuVG?@2=+1`
zLqh;QK2U;#tN^wksFq$sG89;KrR12<oB-kh8OPA>!cc>ToB(U%d3p9?%5uC#qidU|
z2+xCweaqVmoqk8p7k|lSyme=rA&HU!{9Ue#@8rKG0bMlA*8A(WPNU9WPW#c_ob)7f
zTjC>tk<Cos%g$i)uR-<xzXrgc%)%%_vSGrp$@2uKc`n+rG~+V&`1B=7521m<nUmi_
zm75C$N$w;T^>+n#Z<nvn+^YW_8fVr1<t~L2hTw@#R-t1PY^^tfrQv2aZ(=hlKd7$`
zHz~^Tj@#W3O^bf`_70_Q{SldJ8UCe9<NhUeg53A|ISzOVa%?P9LaP_s!%7Jrm!YWL
z+a67(q`jUZ&Zdrk10hGqCFfZ`I#Aa@i?Vx!^i0Z3azU5M*-Bs?4_*myy8Rt~qno$x
zn8#l3t_Ogm!eIpkq5=~MoT!oVCMC60@|TftQw7ffZVsIc7_Y^`G#zj-C?278hQ<e=
zv%;@R1Ij^QfP%(1Ap{ygpMQEx=S|lsX$1KjDA^UY(g0=(lPt2B;MH>57Nc<aO$Gfy
znYmsZEey1PnZC7UMI&Eo9#1131r-Qv{X7jQ66b^YHeN%I;I_c1oubMF!cN&&B@Prb
zIT$Rbd*`cfv<<(E*>;vzGE^ni{vzp&y6}X4dO^VSm-Cx3>k{X*k}@umg{kx13ukMD
zLdaY`NA20dn*R-voeY0salOsy2E`e)m50aq7w72kID!c(%+KTEsgfQ0(<ikK|76av
zoM8&?Tq`%6OV6OUKsI$?WUG3!oL*J&m^R@ui|L^g+>lqaJ!`V4>eFdaX65(-=8?;+
zW+g6b<MBCFV(MfC@8mpxQIS-856d%Ev9}adbzZICU-!yPdQX>Z=T$E}vqs%JXYExi
z*_=NTAg;4-#=XtgRBS^gP+O$CM>D=U_FTBc_>CF~*mgj4vCYY$wctS|5^g<)NQJ*n
zQ4Bgiun749=$P&0+njgbbfxU+=|Q77(1^jzD#}2Co)};U7*yON%DVU*3Qe2JwSA#s
zjzF7O@GXW*5k1=l)4iW0#c<<+fel=q@BvW4$Z$_IgLTz1PoFe&T2OF;gWbwtE@VZ2
zd6E6Fu#XiV&{j7$@-}NIS^jaDL_3H8<c{tpplV8kJ&8~+(Z8HwZ1exvdds*h*Ck$9
zQBfL^l9Wceq(e{?kPxK15$R4rltw`5Mmi*<yW^1(DQOT85b17s=W(5NetYlpep!BC
zk@t09GxM+c*?g6yC%@cf3V8gd;S?QqWDpfSR9>dZ8MA}h<%*85%T-H|=9WiudAZxt
zvhIj*QA6Q+qt4S#dC@6|O>vhhQEa+cB=Or;{OmdYYlH_oB3Ac58~lrB!B6JAFV8#r
z$7R@?uxaYz?#lJ25u^UHw_c1iuO(3QK0{dlbXYC4J|?VAQ&5Q}J*p=Np3hcG``}Sv
zczf};qUeuPHP5xwOB`i2y{)to;<Juap|gH5rLm^j^$C7~ikGDkYf~RA50ZMm>)JWu
zU!3*B_p?#|(p=amWZP_&M~qq9pOc?L@oDL>E)iN5=EH;pThqQyH2?lrPhI=m1BQL=
zt%*8a9za1H|5aZI+z4!4bUm62fUwO@4on#QC-P2|0P`?UD6go1S_p66-_7dC7Gg<R
zR^;v*xo<eT$DYwl%sQjcZNgG-nV-H3@Y(>wo<67C+6d4xlFf&0Du)}>@C0Dkfl~qw
zFMzax{L-$l6acgWMh!hZN`Ph4($aQ|`t6dsy^wb)BR-5DutWPXd=N?Y=jEip{_bR;
zZtS*xwsc=0m*K)R3E<e@?8V&tZs>7=I@B38H#et>msN!a)>eFgf*aPWueg45eERaW
zOeJYVy!>77ccHxb`gb>sPEb2*F76z@%~w7^w~n4I*`VRCd^f+FI%=cxpAWNradf80
z_DubKW95FYChroeQ4JZfKWS8S&sO^#>x$v3#>Gku%z(KVttr)@&#KAZ-}@4$3E%R~
z+>(<c60KQyNWVZawCdC!-w#;b)>_#YgKNTPU+R`>SATXh_q!j}yIEgVRf(=!?D}9i
zX#811(hi66mbU*vFoQyEB24d7%bD8#mAr8*p7-(ZB|D3$s_^lLJekgKSTfBm&cV+^
zd^vmkO<>O-3tM-Sz3U+taQe<5fq)e_T=Bm;No_JAk1j2Z3h-6<^MoT_pw&UE8JrII
z2QX9+3u3<z$~QPN0{aO70#?oc;-~}mbD5Wfn5g5`b~IX5)=tco_`uIIhr?5bSq{z#
zykY?G05Zafc@1F)hF7S{Wgwv8K>}{v(9Tra%mOVb1~5OwpS<k0P{!}TTC{yv5BKU9
zi%~XE$uz;rNMP5HxV1iDCZ4NWv-7qtd|?Y9OjZA)f0b|6lN-oLBya;~^GC5>c(yqV
z{8!*k3+r}p$nLyrvBg&FV|iJ{ba;bQ+0@dW*j<LoX<qqvmfJ1I-p3XNOD75AWJ1oq
z?`|G)TZmuV->0(P`mv>;!pun5`E%@`XG`baf7MwD9xOpYg6=2RDr@j7s)!<_`J9;W
zr!u=Lr?;3WTqdvosU<ah@Ia0{;@I<xNg3An<bSmQ7*h`6q__kjTW7o9C)^Ke6}$$q
zhUh6@Fx6Oullg?d?iEp>U!d0ptOUpVcgkf{{;jkAegV&=GW4#@#1c8xk8-HSW~xy-
z*od}gx!$wh8L#?xH!65|YYo(y7PIXaxnyOlN0;_%uWD!|3cW1Mx<P&WOcmSv=nU%#
zbu+ud6Ohq34)KI4<+Vdrp+=*p01P1B$IISnSru7>_ICF2qrdtpt)PNJ3ZDKyxiUcc
zD5O1yh(4GJYKv0$`6nxZBLM1zYn(+lJODZ*O)X7uPXX&T0PHIVaY3&HhMxok1uTyJ
z!ngr6zXa@mAPgySUe{#lN!MmWu8e|;3?R<3PHH+MFb8no1D6W&1t6;c%Ln*D4BB5H
zU?D;n8RUT+S1-@H!Ua5YL*NO+k${*R*fuupGTXPk0INaJ0j~TzTYdzjfP+|ga|(ei
z+FtSwC)r8QVA$00b9hrzhM&!ua0rRjoW22eW%NBm)4b35Ex{EItHqb}ha!hfU&9Gs
zYJK<J9$dh8&<MQF*3b9$KMQXlhLVaDEdffJ=_?_@DZ;n$?>(5Kr|?))``7JU^G|{%
zG{+hytw@hj4y#8@a%!u{t+b#1beRZrZg~7Ndf8x8xB9}JkE7J8Dv24__8OgS8%NVD
z)>nGwF{S%1Mik+vHIWQ++^Ifdr`&w{N-D}7eM#wxF7fF}<o+5l9^7(=rW&?+Wu*rK
z*v4j$h8^6Se)<P@Wu7L*XA6BUD7jtKNJ7Z&e61~k<j`$>!+Da&*|~M|@`7$=PAC8=
zIj|#BRxts>1%^#1lP{C!=tm&`AO4n>lk|Uga=|15?<?HFZUDSX)8eg7RD^(<qzu|0
zU{;X=Ho)EgCg1>~3OTd@>rV0}<SW7G63RCdShNbaYF8wKfZhCKI>+7Pb3l0EawxG1
z_Co=}6S84pgU}PU)QrmP82A~0D1O2eovzL@Z|n6^n=SUA#hFl*k9QV57f5FxDo!2v
zLkmC}5+R=hKTYpTb2?7J&K~xrUZ6buJbnByh?Mm7;9UNMH9EY0N<m5`ff{VUeIu8=
z!X06j>>_XVMiPsPVuhkjbU!B;Z%cowB7(vBz;^P1C~9Z=n)llMZq=aMlbK5|!@kuz
zD`61q+)qrq^XM%6(W9}*o1sa&Y$*RKg#HdlL+M#JlJp<Ecd2D13h5XUk6)<XZ87cl
zKX88e($<44KcDo?`t+)*d!fd)kGIy>(Gf}UNoy~Sw_BNSS<TL?284#!jhM$Q(rX6K
zrywe)If|6+-<2C~?{RRZ_G1m7dHvoDu<^Phx__UKv8^u?#YaZ;W|DQ@ON|Z4B{}5U
zabwewiIHmFUf|%GNA(RMtno;eU6sXnk5{0L_E~l4&_+SYrOy+ERP7N@@EvT}Oel&a
zC&yvrPnx|xv+i%s${PDJW0MSILflA-7b)TJI%=Z?<1_>~0d`hY1i3S~x30s7K&%xB
z?X_L*DlWhc$ajyqxE`Fa>Uq`Z*HM7*V8lmhUf5QCn+vh05i(iNm7||C`sUZt96+%E
zZp$(Y&uZwDH_c;DX0^jrSiXzVfEcy5=mm7?i}nEbfyOAd2YPl+IfrhGUehu-RVErB
z+k>K9kwcHegiIJwR+AiSd@Tf>Hni`AY}#J|Yn4TYjf@rmzzaJAU>%tWwCTtIIY_H9
zPl)cWb<p8smVH_n&Jdm-#lp9wB=o9FPvzNs4HaRm+{IClwXPH9!I+8Y=|_v2EGzoi
z;{jFPsh#`Qf0}}U_2V&_6x|&A7vTGQ*=Nr_q?cG&b90p)N={zfuPsCwGFPu1#`2t)
zm_pBe*o;-O<5sn^wirA8s#wcfys+48Lp#(lBetZ3d20r@ncK+!^jNtYs~fXT?n|i#
zpL*%~(A9W@YP8twnmKiq=t&O_JfAeG>O<Y5hS_)7*)myU@dU=Y#fhWS$5a8EUJ7Za
z9GEzDH9zRg>*D0w-r>})6U4H0Xp|Fk|BxEkd0we3`z)=13E%4LvHll#x=SS|22$!T
zqE`&C>Ir}6m}a?UU>#*}Z(|>%TY006Z2CR_`r6B;Zr#yoqAm$}B|zcs7&HE@tfJDP
z<1#JD6g@xVu_>vdf@Id_J3`<_r+FH#Kj}>OoS85LbP--BL#bc@06~j-PWGinjGuWz
zlnfo<eo|(G@Ki-pbBc?@p?L@L3Eo&B$brN3G=vSDu*DswUl6xcT3I4_x$d0&Dtfaw
zcm2i3?Y#pOM%<N$c)<J4BiDF&zBx2CCCe56cywQ8vK<LM!CImO<Z_@s1)C+;emg*R
zK?Kp@aX1KgRL~s;7^qpaD&H3s6?Hc@V@)PGm_|%DiTS7rk$<=vclzYH_c~gP<R-mj
z(5&;u;v1_gTN!2tcKRDvjI7Ru594Q-i;G{*-w%IMN=p_0pO=08o#9S=%Q_07a<Z(9
zYd?x4HEIU2N`DjwsFTZc{PgEdz(!|`S@hSvbI1D4K6G}>ozDZm1{1tj>}#ANi_s>3
zQoJrW-CgqB_F4U)IiKy26W@CGvO+&tqxkN9{_L=v+F`gcq5Z9pvF;C(y;YUnPUaI)
z@7ka1+^Nphi-gsJr<hfRlYK@PsQ$C3ie4rw(t9>yD3pyuSSK^@=L}r8C95LVCoDJ(
z`!aJ)sf!}}YPEV6r&u^XzM0W=U7;JRz&O~11z^)f!7%M(FYWq_eqPx3AJBM{(~yiS
z@V3LQ^@2+-IM`jfsT=7$L26Z^VfdI;0R2Kq$!%!6A$@!5)fZ%R;Uq7T>4cyK7?dEG
zv<>cHh~~<GXmJ&lnuRc-?>vR}z3Y=zYTvjadvSpuWGHZyKqGGgF0;Sfz&s?IX4cjM
zYl1|d{|4_c8w?w7Rl-_%<WQ)>G(a|Kz{1(NsrL!T9L8^`Q^nrrtMHY06HTBZz~KYg
zAP9TzFvx?T2joQ%#RY5+bITOSH(uJ<n4O>O1A|R~u=|#;38)Rofwof4*R^=8f|mrY
z#$I!-ZRg`U)E6{UOHC#wS#JN%?|qtg%e$^HwO=HaTBUO4j7aUj+IYIfLvf~cs_4*<
zUG0GkOIyjy#fJRYsVO*pg1%N&C(>dIbVhR53R|c+c4eq%P+s-h@!O#|Z)U$<c%n<9
zxWdXiC@^cMbthZNnmDGX=Bl<+g-26<ZM5-9xt0pyt?EK@vwmkYS2=&<u!Tx%&oeth
zEYH|Jy%XVs?gwu?0{?tj{C03MhZBD{WP5T{7h_dxE;7|o>#pcN?OES!bG@6*8X6Pf
zbmd28Rej#dh+bD>@)yLEF5IoMozw12H;jxp_-{Q8udOm1U={Y*mOQ*zGChoF(5(_O
zs=>^Q`nJ6w=zqI?`7;BV|LKp~Rj<1s{^|qq29PbVzZZ0KjCY}WBLRKHg0{=F6DT?$
zO%Mifh$h4hdJHQF2lH)9(|andPEJk$09H{sz8v@1J<9(hbb0AkSvNbCn2d{JfEFUo
zi#9R;mNflk$a<hN`FeRJxlq%OV`t0Ny~EwP_c{gg$?`5vov)@JGV`Z9-3+OI^ANWU
z?;hGSiG#(Z4YSjW?XCuC^?dcL=-K6CS2C*Q6c5p7MUVR<M^D%Ju>mm8C@6pboUvW4
zafjdv_*{m{J<%Qkj0Tz^_#h3mH4yR6g3(i|Uh8<jL>-iEwoi?P^+4fww&}1O&d}8v
zM&6@k)6fCS2(jNF77Rcbj!#4cL=<F+=#1WHVoHbHKfU0A<y}QsR-E=I02Y@*)-_xG
zmc(Ci8N|2P>BQRd_YNoI;rpRi|H=P6Gt|f&{C1W9c(CY(SF&+ui?eocXWJn2&))k^
z^)*!~@8kQLijCiY2moyoHSBZzkADP?`LXv7TE8U&^=3JBYxM`(@dAIk@0>4M`k%B>
zX6<P6C2T)S)ki}TzHMcU>_W-*6^X&Dti7qKt!X|SXx?Z(OjvYW%J!U*cVkS>&n5IF
zGv=pS-bsY9s^2qKl@9t~UV~~JbK}~z-{j<vQ#>#?&hr?Ds`Y&s{fN<9ZV+L{2As`a
z`muZ6W@5JM?(VW452`38@4s8N71dCs9G`XilEi^6k2q7%7qdsY#OG>>!ufgYtdmoa
zvk5u5p+Fc-_S7`%L1CTNEv`1)Qw<LimNS0OKQEpRzf%rVG1|xo@;vd5i08E;c@ZFM
zE6;mBID$wIye!4b-)f4OX}<@a))e&saI+^ae=gzx%be{G#?8VbED;cd;#s=kTD;uN
ztm`-+dHfoLB1XA0E6@-K=Zk4ruhVSiF|z%f=g&YW;-2FYNX}tAT5c4cZ>ZUR|01(8
z|9}6?A3c8scKqJNDzk`9V8@A*x>Zx=QY;nj<btuBY2IqLzaPY*E`BvCTH*(Ov0DTB
z$G1wI;X$^lKCKt3Us&D+ml~q9aO9i3H9~7M)sV2Ii`?1{Z!0(7+Ufb?DbO2t^&{g^
zeYGxS<x-0h7K>Vxneq#?_hsM2!}rrMJ0FL(ky_(&6@GdcSYb;%x=NFAL+T{>H`$&;
z^@|M4_nq$}4relDhwdfXY1FBT548_3*im)bx3Uo4@fe(B&eRZQJLqzfOstsldPO3e
zaNUa%Kt{GQQ*r&C&&}hMGlz%8@1u+v;=#}WZ}X1HEgw$-NrWBnp2xpp(;3xcK<xmR
zI7om*Q`x=0Ays|Ox=$c9*K@NJ^qMz_w71fVpaF1R8?zfep1nAzg0fvYZrM0x#1<OO
zoHqtqRS<Q$p2X+eXT9HdDAt3Pmy}e>5&{CPqSfEEVXL+9<)JBS^F>>5TPs0Gn+>XG
zOHF{T8|psngCYiCjfqH4IKJ|SH3{jRH^1{@>Qb#H;WHu7iz0mahPM27xi}6VsQ0K%
zvzKWbS^v+&lM1^RNi|Kr{JjOYug~`9n-3x}bk`Mx*)+rQ@;^vCvAOx@=apsJ)T}N4
zenpYuz}i})>#5r0*kt-^kKd^sD5H@0`FS}`hhO6-8LN)N&J^bO?YI8%m<QiGrmWQ{
zZr<-h+RqVjyL-{Z@-I!wejdUb-yP(u<9UOACZ9!jq;)~@xg&wrfRqNsiJ@S?LC7I0
zitP$s?Ay18Y+oN1SS`s)<>;suDF6P0hf<L9lop|MpEF0>rj*mj_j#|0_L=BWuh}Q%
z)$GC1o`@R_UtSSRiQP+gJ3VL_SBj7py$9>r6!GDz1t_Gt=oxP!9Wkb10)2iy9*B!k
z-@W<GY6=gGEid0a3V#<Ks>kxiCO>TdrGVXQLn9;mMR{zvhB8AuZxltE!=7$T(A4~j
zcy@HY>D>>rJ`h$&l~OGm2quwxL0}OwK%()mv@tL9_Y1Hv?O4F~0KEhcNnff$mJ*QX
zg2|?^T7c82^Na1=^<bHq=FM{+K=c3<{|=(l1%B=Apl9(s<BJEAo2KWVh4M)5d}`34
zgTtC6MZ0zE32pu8DiFtDs*0C&XPjHIoiNYY&Y(OGRCxX81^mMQdHEr)!mcjbWC)hh
zEQ=g#P+Ry;_0d-*NAd&>@<UXfkZo3ORlJ+cIQwoYQ=wa5chHh!jhbb-_u6HXFbjvz
z@}ZvneY$0D*s?IU6=#lK5+%imiTTh&h=;dzq5`$AkD$I*jXxuSMA3?qjBwL!m36FC
z@=TP>Umy3a_>HF7NB4eGk@>eu+@>#M8MxNF>UX@Wvy@6@tipeGQDt%A`JE?rS?^6w
ziW0x=!}q)U#Wjx3EtjHsTjTw$M2$a{#>#K#dfsNMzT(v4*;%CLv)qOF)TmpZm-BR3
zo8`Wb+phlptLi-^&bb$u_q*h|1XiM-g`japf}jX9{AMaLJr~58(AhwdnzY-=D{vG{
z-ZX=FM@Wyh<V_?p&o*Dmiie>{Ma2$oYS`4<?YYuw1QI?l3%-4ef_n2N{4q0>UNN)+
zeU4;(VI_=;iX^OR00m)ZrPVk71x#4y`&Hy{>K%xMJ-0hRwq0sIME_UH35-455y0%9
z6@#Fy{-Ser?_h7I`5Mkv!{+O7M|wGpAUQp~n;?z4{~{Ygvt{h;I12-?(^osr#4a6K
zC;zx`E|d?ADK(+UVSJG7gbf&R+3}kVONXb@Ci9;AC%eZBl%s~G7{?&9dXRrJ>HmGH
z<*|$A(muPQ+7#}7hYT&%?P@8>nj)#^HAVcL_@+nx=cJ}|7rqhexDSY3Ht@GgOR1!n
z_`i6IO}WxoPWU>Ej94ZfyO|D;l~>+(bj-uVMxTpeRu~&cV`WVznrL}PomSye%=m0<
zlG!~iIyPisDAl>f5l#E}{*5-=FzL~Uid_oT_1dlseHBM1YAj3NoNwKjo>NzAB~SUh
zaLXb_DiAMp)2GHkGQZbJ$Lq|<uT=sFXuHpsho@|kl7<#9t^Jq1uPIF3FVIm4{}P%&
zzox7Bk;A>C!*EEg0N0^@ryJ*F@=WciOljo%{QP6CU<d#Q$+V*4?KvvRf>rm%@^rZW
z@PcfCs~lW$6gB?=(plubpOLfunqTp&wHbJ3AQ{FdB)rfNk4aR6KR-ZdrkgX!5c3Z*
zS+-=$zLMi0;zU|+;D3~j<3R4Sdhz0g=mS1}{;#0@;B<WVod2;fh1s9sytT2CcYlYn
zvC7cV)xMAd=<+#M_cW&5mXnVwT(Lsi0*2xaM1dNIwRQmeK}^K``}YlI8?QM!k!L!0
zT2g3iGnKdot_-!PDmz;fN!Mt*NrJrQrI)UkIXUI&k@?wP*a&UpHGF*h)ZbO=JEELJ
zOLclYzH{6!)$K5Ftk{jV#ipCncW~?sDG4is%u)P(?AD%VH}i0Eekk1jpIhP5n>y0Y
z%_BCTP)jnt<HGCXR(ICdhk;x=PT!T@9Z9eS=Lbhtva)?sk)5no%uMq-;8q;{QWSRa
zEQWixo_e@Zx4a<L#|Id9iLm`c44dH@?83ryO9T0DSxV^z>7znxl}t?fmYz5RVNb9{
z`OugUK8P!l9s90<hw*1?qZB-HM@1pom3YA}pBg3})z)FK-hEV3;K?IzW%Fpd;e}Kl
zg5u6T?u+}mpYIT|jJ{bXCPtmw3V%*aF!M}Be0?L+xNj&omer>((O{A{Hy;o6OO@J`
z)~Fxx=UiOFl4|w$3y1Fz*W~1EXI}%3o#dQQy?y@;)bP8bPN2(lbad1=m8<org|!Os
z$Fx1#n)75YN>8T-L=HAE^&a+c$=w-s8gNRMrGI2+hvuIKWk3?vJ;7U#9HM0D-_33!
zWeVX4b1SRYqi&%ryXwly?Q?Abz=gSqJ5#5xK@#ZCg0gSvFuTKQZUA{MF5uRQh`@v1
z{;M%JKW6|ORXA}FQwNZ*_GsTB7jT|CJN7`R04hi?I&21!E!Y&nWiF|$!zAn_>O$|{
zO&gaI&ON`KoE(65bd6EwO{k4kN2QfS+C38!6=Uve!?G)BZcQQa1ot_6O1EyFG0G6E
z{;)PYa}AKQfd!4yT{qif^mg`z_&#_l5Ixn{YtSpv|C0KRO=@iOBCJW0WaE-v%(a9v
z9iwLZfHYS~HELp}?GfK}>rjfxl-;R4(`^4kXR*QnJp8cFv|G!@|9j!=RJ~mtoE&Q-
zEsov(CWKyCNIf0h7p#nriN-~4s@!?hmnnp{IZoXknjM6pr8PIACOK4U<N4@ubE`p|
z@ZH0=7~6db{%JMsI_&H3Ktkku_GjgGdO@Am*OCouszkD<Qk+Bh4l5gk>AlN$Hl^Q(
z;Bm@DjeIt+#r1G!-SB<2w@Zk}#B+4fge_*o*SD?y!}Ui)`cKudv6v8(8pb!H3rd92
z++9}G6Ql2UI@r<UD~xUmpK)Mfw+<<^4d)@I8*PajPe$gp2U{ct1MWQDY*1T!XAtQA
zvE^R<ZrBH7KR*r+UfCo@S`8j1-*g1cL0Q?0sC>_*yGPTT_wY|_8xI-oB8w0hKK>c)
zZ``;6kZy2rFw|_>B>s;e)dM^pR)_$yeg`{1U}`{Km;%;b!UNCB%0h~iZr&URYz~z9
z`W#>1-{l5sX7jVKe$iP5xU8zH<3I=@m<)2ICJXd+DIj-+d9A0Uq@~dy)CWoT!0v|8
zRTFUP3LpQ9gt)%d+P?rGnW!gH(FdyGzoAgDu?(p$iZ)Aq9WyvM=$VfFDU<f0TEH`J
zJ<1mSxSBG=Adyh>#x&t_3g1WLORAZfw55)WB2s_>Ld(A8)8UROtfpNl%3i2t0~{GJ
zE}HO*l!>9%mkA*z$z!#2-dBwXZI7d~jGrdSZ0I6(g-(J}8sj8ipwQ@|aXDf39`E#5
zANDYKq*J1tlU8re`3-wHVf(l=owSdaBv4L`_m<XKo7fEmaK8DUa<<cAD1^T15<4`;
z_)A&wBd*f<IW=uI<8zU#--I!WE*x(?yS#q#Lgc%%KEJ`pAd22;LbbYzUtjPOgF)Jf
zY`r&p>cOM_ti@*LB)D%2K3k|q>>mW{-9eiRv`6EeBz!%9it0}{o+5%98pG*vf)|{_
z-YlOTl!GN>s`7~@n<GM7pz~R5b-N+c<c#Xq%&U^Jvj-ky2H&XeCix*!vpqH8OsyXs
zSRxKP1)~Be3HbysJwLKAi>6a+T3-!DIXztPIJNQ|r%)8MJFqNCqX<|3y0=1Ic+jN~
z9HnlM_Q6OrH7+y#7I&Ofpv^-)${eL}X5?YY&yVOc)&RmB?ha^3;QS<HLxwZRGLex@
z;CAbQwK-qNZvw=`L(;*42|Xd$aFE-t?sA*jF2$RH!3Y_G3QwEYsPlbG<x_;kli)I%
z7Y2U=pgKA_WRnG>ks?yq&p}KKrST3^&F@+^P#64Tl+4WyluP6VLorZ(CLmvjK?F#O
zk5a!pa;;|S<GK5V0oy=;fDM{lfF^+`7CRXT2S^DHtBs9~S3z5V9QA-iZQa84ckhTr
z6KAsJ>EE+6L~o?Co6hh1?i%)$vYedg%$SdL3oLuRefW~qni>-4d|M4tJa?oM60z;8
z`<i)J<?Bfkf4%0e3^P$12^AbUF&63H{GH=C!xf{4cg`%g_UOsyOlKKNtl}9&clc~q
zMUMBDn6_pj$%>17GNtB!fC11CN4%hv^g$b8tD5GW@+c_ANjc8gdXRv+%#LS}KW`XX
z^qM<pwErIdl<=@)$XusM0)1v!Kf^Bt@vC$%LbjuV?R~O_f^)qUuV=Y?bPUY0hpEeE
zewB5oE>B;$Wlk85)58}26qB3dCj(hgyWP)v2OZB%Dod5;SC*4r?+yK)KCExu@c3b5
zQ1W_0m?G8vhRMK6Z>{6)U5QZFKYXn;;eo?yP8D&VvV(R4U%#dn&tWv2uEma|Zy76j
zt{ZlTm3rmmuI$Ph^T~CaBE9Cj%Srv2@mEpAZ-lFfj|Rv*VER;z9c<)d<zt@qOq$-U
z2}ZyL^cCX-dKi7c^@U7wmVt`-*S~k;<E^VrsZjO0JHr<DNA(z>xm8j^-XY+qyJ9(6
zi3NUeJ{yWf0wa#HbeHju3rp^4Ed%MYDk`sDy{g`*U4_}A{^Dd6S@0NxTLGHv1G{ZE
zm~qBHX$nl@#6-6*+MO}5JHu&Pb@xgJ$p~{%^;@KE5Go;Dd~mFx<iVFh($OFQZ)s^c
z6Vx8fsulbyN%Z=yNa(8V-H`OPg((NNj<m>kH*3wual#^-opj!|-19f!FmtA(qmvZg
zA7-wueCbRa^w{kr{=wRG*-eC<H0H>rhggtsSJ#5NNLO#AT!H(%8}GV#mJP?%Mn;p7
z3YL0DfoBSoR1*=b1zHWk6<P=1CsMDlN*TC4s94tR+6j5J6IA4X(`GJB$Asl#nN*#J
z$#0~4*X2CIY2}8t%MjPA9TUvc@!k1Xstx7Qonc<GZyzH@zWkhLuGvZYV?^75(p$4+
zSiRsP?0qRiICCKI%5&4|{Cr=bY$tp*cXm{|mFHHBe5Ui9VT_*c)4Z!=9_r_jr|NaZ
zH!R<cRL%AZW+Y_IHm3OH<T^zC7oPPoi6oaP$!Y1g!vZ1%cesSHU@t?S@lNHdUk=kr
zv_noiKNylbpON^pk@8`aX}M<kT($dkU4xpcRW-(`oL0#GmfW5DWvvoN0qiySgpK_j
zP5Xo(F&om<jQcTzy~49ng~6}I{%H4QIh%UtHH^k@=F9BHXY<4#XXxKw$d^0fIg)?=
zMEv=HWzF_FLS;aNiDE?Jfig?qQx@~@)1L<?P_eMSm}1q>Y~5YDAky{{kWIm)qf0l=
z=x3<psoF@3G4G|sILRA`kcGBA)I<YHB)Hj~bl~QKfEOuR2o7ntBZF!+@tl??mWz~_
zSby}JPLkV?_})ama4_U5DByt1^9s^EKR<6+TU*EBs2Ia3+@DSHuNDBHR)fJzY2z{t
zSgZ>T+0(2xFd@U`HRZ9{02VnwV0X;R@2osU1#OG(YAtd*j6R1H^+yQMEHmjLR95L&
z-d&_ivzi1=Aw*aJqIey}$8)nn?0iK*+rEoVc)R@$Ai3J3<ltifkf36nI$oCB)bh%j
z^qSQKxPS;N3)5GT)gE@TT!HWCeyYtg5>WKyyYN+I$+QL|K1wbAt2&hWPah#p|FqEy
z9Mjfmz8p`@6g#|@XF={V?asHaT((u8rLAqMe^8-#IiX2$B7Lop?;>z@`9`y~$$q$v
zfr-P1gfb#K6DRz(wH%XO!IaC?$(O&+M8(c5u~6yHby&;S_@v{WVp=CUlD|x&(~zU~
zE=`qdYV@o7arnah<XGFy?tyq&YA0XH@`~Q9#G}*0n*Dr|uoEG|fe!7{Jglj+pSx3s
z!xGjF`6s_E!al7orrP$r-ly$eRX@wiQa-_R+!$-#4WI7#p)+c~Di@MBoUWXr`pm=(
zL#OLr^w6t3G*zXg<DLVZ|AAwVwc0qYs9k@|zf)sf<QAVDG+J9XSF`L(jsZ^Zx%|)C
zCnp5hl!W$x&J{M<XAu%eD!(F=7G^8z5h>-xWMmJZi>+4>H=Q(WHcKU-@2>2SE7R~C
zCni|BH%8GtQyU=A_u1_rbk(gn{*hdT^l7`~QP0~?xjZ%jW1m`|>-x_d@}E0WHEOH%
z)-;d<o3lP?uq<B!9-|W0oMcoLmHF3c9~ke)Gjj^?GaAnWFXPh?E|~L`OH}X|o-RmI
z!$;qxGiDC0cm{Q74}ythFU?Ck#srAiQO_lk?w%f)(viml#24T{l!WdKSQYTNete7(
z?$u5KPHaAFGSJjO#3~%*qE!ky9~e4pn$CIQh6a~;<hRV(SVp*CsYybW)YK3lUw*QG
z*FlF!MP>ICGCjdt2zrp)n5eK~0S4C{%FAs^kVJFLdRzVywnr;&6pi&?*GcmDhFIBH
zH?}dl6JL;8u<|KSg6mJMJisB~-fR1`|2(4d2k+Y!+zvVl-uhafi;HuX21klKW#?f*
z<A}8k$;<4DVH}d-*9)gpTibKJKL+V4re2>i)gJH|ef!+OU+#m|M?bJ@{qt0>h@4O<
zHno2^_1t%PlEl2iBKJp_K(8I=s+#@m30u0^OmpF^w$z5%QhLar-1>D>3eDYn{iz<z
z&w8FVV7%#mR<ztbYHhlhQ0I11Kv`Pwr=Qi4{7^()!Mn@wVt=+Mi??bRgRtz=Z&PdC
zU0j@o+Bb>cn6ow_d%8B<QcgCM{=~eNKcEF|;(wO-q=f3CKmm=IK;gx&lJ)lHT#~n)
zEhc%#%B3G?y^&Edk)dzyc|8=^fr*DU_gVH1y}U2?Qp@>axR>e$j-IOP8Yfk<w%XTB
zSI=s?2|IHuf&la28GUZkz}<dP2#~FbGvm1RF|e9G(NES?{lKk(k%?ucIyEHX@Dqh?
z32FDwxqxE?Wf+ap6_*)KVYMOO%L__Ba-qi+U9Q@DdP0u35)wNEb=1z>f4(86Cww=k
z^pSqoy#s$@vg5eBgtDD)&E_Vu^Q*K9!H-|~x$ipgKv9$fR@WS!LQrS;)30U9OgA1v
zA%`_O`Ao1C2TlPO#4Uq~8U_c*3V0}mR7}(2-F=Ylook$7p5P{KGU?oCNp%ZNMOoR`
zY7!}{m&y9VLV9?;hxM(LrIVv0i0BkR^94HsN1b3M1}+ACouC|gE%m^xOat^uFf{-x
zxfr7b*RwH*wLyero&f&=d?m8$*Ns$@aYv`a(u<Lb>k>@<X2LPa)c8DqmBH~uA7J?c
zSdfm}$)4jMtWP6A8+@huG{q@1fUC6T^+)N_aKfeUPb-=}<2)hc#T>Oex1u>o9;es1
zH#sSZLVSU7FN@`5iJZmVJ7Z|;<jF~eQjBiN2Sx9^$De<lVZ73)sY?k>tW|e+TN$t!
zWzsmw%Cf9>dhq#T+5AsqQ&jiD8Hb$pq|_slSE3b;>L0G^KYzVV{p#Wf7t7{Hq_E8Z
zXSa~de;pnv+8oBNU%w{;gSl#_42yi?vc=Vi&Q6;`0;BqGrs4MVAYRya*>6~=Hwy_V
zpPWi|AyO}f7kjzGDK;7}?F2T3{ht1P+!$XPJRa9{E({J`JDkN~Z=jV`F`&|RP5YvD
z6Z@fD=m*b|I7Te<1Vfz~{zL@LFAa^mGU8V|zbBNxMNw~v@F<SIv791=UwG%YU{cK$
zUK{`EP->~s5xOv_=Br6YWMueUy_J5lJ<-~Gr;@D&pD~J><~h6+1Z-aoYehw@(IQ=^
zR=~J{R}5jdBT@^RA&_?*o^HE0KpzOQz}lka8wsNebkSWuIUNV{cwpZHCMtY%>=7OQ
z(->Na`T=n-xSfHkbZwaN47SL`R6;5cp@#%bH&{>6KMXw%PX86DQgcYAKn?3Ofd433
z1*;68TMq$%eH7(WZj}fWM*lJ-+rk+F?^@8~YFNDkD%NXXi>IA*c;tGXHBME7ghf|1
ztjNN6=!%BnJ5Kc08^<N~r*NeL3JXN{J={;AHs4<zUG9;me{rC=N__t|o%d-}ikqlI
zN=@Pk0pD)lI)*xj4Yq@PYR_<f^Y0zMJA@^!-I^5nO^G(%9yA!=lG>&;Du(6rXl>uF
zlo$w_3w*eazA{@8=ERzz*XJBh{BX5J^(S!zCI`Jv*FKTWe2kguc3lO=(P&86YaNSc
z%=D;^)+W<EPlx3lYn!j^+A<gN(zs{+FVddF5k_-;_~*T2!FNy5xr)`sQl8)H?tXmU
zvSUvyM99SSoTejb^>^=OLR-V-0OGPyv$x-C3GDwR0ekLq%zDlm2hvAFsh{$srLrO-
zT;f+Vr42<J8rt?TtVpxv?v(bHwp-k|(W<-nNAy(!vHqHR9EUh&)aQWQRf$}3LE6lT
zz<Z(DEgEJu#Hhog%shM{TAd3YWUyjG9LctSq-{vXWmn6Ob_|OiY7i9@NrV^qXr$h8
z5V`RIj!R^uLD6b@9!VvAgbBR&la+7_97n0hJ7*1<gNOLywRLX?$74+<hJ?S-Lx$P#
ztAkHs8BqZKfxRJx90*NQQ3=<0@*QG4H)rI;#jiqk!SP_A0KOz3u+aSLrjz5ql*$y{
zjlnd;vpc%1Wbf4^Hv6s*j^(!N(SEUaAB?g<1S3rdpnU{uLZ#L6(90SD2mt`sTZ1mt
z$%O2W@;L*_R*W|oWR&ysWsppY)a+C0+b}(NgN&xx>8_ee;M?w&cOq^Eyh^Atq7qST
z^VwP%1zu<7KddL3hlF^yc=Yi{zHF8Dmz|Z8eB2s@J0n@^zK1=X#gEZ4YZZ}I)vG>A
zuf)9QQtM1_k1(7p*I7kKU2|q8rKnHDLe%t1F>uWBiy$yxU2JF)+MHbN7jvduU7BuB
z|2m*XH;C43+#k|p)W5y7@a+BXXxJhDU$;wAYaIRA9M0D8i?&uY-)RA*4+QwPYLs=b
z%1l3|@CIO4y?;lJx*p0DQ~0Gouw%#@-Ql2y!}}tYQd7mxtnN>~OMJ${Lo<f8X^a=y
zf$i#otV>1a)J2QCLC2<Ct;!V@LC5rw5?9<CINZ2;lbK&VBqRtbeNWN9r5O^^A;|iS
zb_fqGPT&r{-}VBP_LY`5QsDrC%sF)YA~REeUsE(z`4ZnW8GH65=Gpe7h2rL^&30$M
z?ryBt=!Gfrlo%zL>i)?MPGEnOi$%aR;(zjCanJ;=cljK@D}&{pOfpfV_=dfeR%D2-
zSS4Yl4P>AWytnXajoMW2V;+VrKq8;Gnc1@9ea^qWQ6ww|SQn5hkfMTEA789xgM<XR
zp&e>=zJ5@+fpiX|Ajz}taK>Z2$y<!uwD%{(F(LJ7XJ^Y|IiRR+YW(QXeJ)BEvgnu6
zzl?CURDO(x0KipQQj%})3~5kt-RWYmBMayVvC!~Td|+cGy)|aE3f&Wz$bn*xBX&=t
zXn@Hm!^*^D^N*8IOZBFYPB=52DJ`43ZJ#ZCKkcX}x$K7y>tJ=g!?K*#?eE;T7~h+p
zp0ej?!LiYhSxhDWN_L)Qio18Ulg#>PU5;h1q5t^wmBz(k$e8Q0pz4L!9gN1?B7Y)2
zDP$rDrybcx``uC({`*is$w3LDIq^6@sG^pPh8I>-S;fFzn1bpP%GHlwNQ4MqS*#T6
zd|j}Bxmqi&9G{TYK3-$hezl6*b5)Nz$@}8eQ{Hm2Egk!zid~|X?Zkw1lpO6{7Rdp8
z%n|WXZEbu@wT^NXGPN6sPDV~NIXQ#jZ;z9Cv6ju6-n2>#vwKm{%TdG__+QOO2@XzY
zYD+JyBDz?geVkgh*fUo*o0VzEQgu1j;RJp1(flkkE#5qD&C~fB^ysckTPMDm-}ZAi
zJ~5#F{C<_QXd$pcMDm~e<WbS8p(|FYDIx3_%+Kc!d&HYaPKV4m=D>tkJ0bga=#$-`
zQn=oQDV~B7L1I;1x&$=F5XNI`%i-<KFL-~EEax%~WXuv&5d9*RBN>F4IyM>rh7?*F
z=^yZKem<y5j5Jg)zVn0vI7R^*dqMLKp-BHYIei}LE1Q2c3i=G+1l$e~);8)SO*{V>
zgL_Q%YQjc`UDBC4Fh<DFbo=$uQ=wdnM@l+K#yl@H|JBn_hx&bKQIwZgWpYMJTs6r<
z*tMG3n;eKpf7pE5-h7oSl78LbQ+|w}3g8Hq_gB4k<Thp+edKP&DrwxPG4wY*2s{-!
z-@`#U_J~&0Fl|m_(-eAxw(0%H$1QcamV32kwePfgbAj)<d%Nc|gfa?Nx%1w&r_Xr)
zuXtsr%Fjdtaeldu6O40u@wdA@lbf%FvuisfM@>QyjuWd%reVrCs?4m0QBq6daB$ah
z>c!UwqYoMRPsVq{y`0|f$Zrz#ynSHt$&1vQ|2-*sWKc{ApC%$P;9J^q-&}Bb+J{@g
zdHHc()KtHs%=3TOiu03w%FK+L=QiE#7Sj$G#UM)IW+_!BaCE93NkGtGKXU5YuofrP
z^5b>m5<a?Dzl{nj6^J_HI61t8%Phz%MW$|Xn_`ldk_ERKg)>7m6S}FYB~3olf%OlT
zk3WXwlSSE;B9e`P8YKyhkfq(VXHM(`iWG!SL-=SQb^;$m@o3^EBAQtsPL>{GNhT`?
zoJ91xIS$=nH0M%w7<STuPsY9VUPdSRR!yS;h*$w5CUQm8DOj^!fuSlm7#-LZc)5;s
z4ZzF-5~ddlB0!K#UfHJH2O}8-8o{2yaEOt8y00dLhp_wZ_z|ycjmL-?zo_;V=i$_@
zaQ0nDvRqt-p9|Tk0C;fR;4xQ&$~DeEUhZS1m7!$JC3m2hkt&BXi`Z?b&_3PoP!2oe
z78H&d<oNXmr?SzW52rA|Z?-zcFN5qw)L_Ited?q*@q1*~>mq+R-^uq6p*?8YX3~m$
z$nVnm$yt}RGuxfFz~X0jQ!%-U5@(GGUDNb+Gj=U`o#VPfx`)$5G2CW|X-ikz2Ng^-
z#$=C|d()QYP&I!PXY8k&WQ<PK`OKbs;bal97_F>i+(>YHd?fdO=j|ji`}3~~sFISB
z3?^iMPu4|#HQO3=qxDysj{ID=qNOm!uJr6vlg3kHpr(DnWn!|;me1*9GFy`Md%pE9
zIl-vLF)UHR;>dFX73F$5g`iNH7pu0d==);4h92jQE2(<dzd7oBEt%0(RN?%H`)tcS
zVA88lkNwxgYVjkL=v$5i_<=NJgimQvva+6QUymPJ`cy4Qb@}~-lyq)BL^2N{;LzS&
z=h8;hctxu+&@nDOP(<j`vnW9{ImDVfP!jQu0M~>%Bt#Ouztq2XJg+6MEF%0C6o@x(
zrLKbb9qiP=p-N{dMa9M<7s;8H9g7;(l)@$U$y`%6f00(X=#VHdDxqlAZ2+|`3$H=f
zrg8R1(5+9};|PE~g(><Osa%A8oM4M;iD3siXnT-?wlw{~q7Akbj}3ur`Q(IH4+zON
zY4gX*=oac+uyfv+>O$((5DWZa4pWcSM)==#gfMLF4cUB<Quo}ftJ)oYarv!00gu;8
z`sZ#?g3HLr$k|?w7<q+S(JB}2vmJyPiDGodD0L5;*KnXxe#O9%rid74B#T&6Ot;~+
z2pt}4^|sxcX?y8{*{>d6*K{dfXq-&D4EB>Pd~Y~*?zpar=vF)B>|`};QMVfn${EFV
zb-P2VHYFEY8J9P}qEq*y<j*4Q9fU;N&n-+8Gxp_fA`<oEDfZ)_$|MuYVyaiP=97&v
zDhdS|bk%{oO<NSPFUkW~izb>6FLHcP-{jvej%GJKzOf1mmzj^dbTnf0D&PEJq}BSP
zvD?nX>l9tGR@#rHV~LANs$cbGZX8QFg@RUHB!J}jvhUvsyer@vLiB($I3z@2!W9Sx
z1}EWv&K}d=VcYp(+eAex@8Y0Q826O6MJ8=ev<7B?<h#OPbiuO`vW<QFC;K-q&-h|^
zkH+t)SdLTEj`m;nWLb-{)I8=F#4Y?fXg-%mp_BH-y#m*WoLnNRFAiIc)n!1r<UN8y
zEwJNL9chY}rfWW@(We4gvaGE6++5rqRyGuG^_&`4dX=8HNA@0r!`DH-@FU$MD3Oon
z)}xzk4U-@;k&$6c5>0OPJ|^PGxuUO-Bq4+Ik=H3$CHX}bb|R8FM%Y0NE3_uBzrwG!
zOLs*FC|m3vwDi9!dC+M?9UI;Ee9OYnZqEfxZCHd1Ohf$G@Vn~;Z$It<encE#9Y$8x
zzS!5oGqZ1d`n*9j5ET^#cY64?AG@ayz(hzzJ|rOD74s4X5_qX%RJSKKtarNqZ_-+x
zG822g{V9ji4#avbw6?tCENb$)8!miu-C|T!7K*52;Z0KZ$tst*UDV_L%QFQKJv3&T
zSf9fVkv@LLINQgRy!XF~x+`B!%Xz*#b=~rdKC>SRB=XD})^(;C)f0)@NN^zM)7hMJ
z2`^qu(<>ltl*f23;BlAF%UodO(Mmyc{OI;bNbS-31lHNcH>@kfMnpR`-nA6{ZL^ta
zqhuTM*(zmgOPfYh-joVWwtQl)^e!`>_V%)-jt*0e<bO64=j@$X)Y!N6374OKy%i_E
zK6}Z`M=*uhOwsLQuPXS~gqNFdbX-+zq|s*^R(LSqlQ&)-A<};ykt$4NGH~ccM;HGu
zDQF)NC+PG?&jbCR(zw1*n_jH`^k#wj)%|^Hf+ha8tSdS$)xL%{#YciJ@+tvM{&Tv<
z0^%{BEr^32ieL3J62_0NCGOnL+4Dm4`w6p`bXLo%Pj`35BQ?=f{;`I{)uQCiJ#*WB
z5p<s`Q*Jpu^==#nXIuMf+H*Vm#mw!fb3d=%Mi+9$4G#82iVhaq7oIEYD@I$l#wk7;
zd7QjHnZvPP&m6dSh?BE!skK&Uj-9hPOhvQ29oRt1?Y~zO(iJ10-MmaAYEDQ#O+uaw
zafY_zCi8<Yt|HN&zv*re$-;;ZMNT@B-oRBjGeQ~&VyH-96afL!U-O`?4qvJ>$^8k;
zHqeu3KP1PKdU`N|9?%BL;=cob%6Cr!$+RzDK3wj4Um#&~5FYYcP3)OPs<PK_p+JoU
zUjZOTu;ID7s3+?F3-u~%@&Fu4Qy>wKDPGkmYx5@d&2q=`_N)(;bl<D2Odu1E%W|9j
z%-DS$o8k{WUI`k3U=ptpQ!T|$vCR{a81AX(GNTn~N<=iiV}TV*oy7MSXuZjq#Eu_P
zT>RSBD?aT!7d{KU%}}oNr>HTg_OslHkl>0!A~7&Bbc*j4>FlwX`fFmmMV2*tFW3Zc
z;>ErCP|BHgJ?x}~IBVmXSms88muB5bi`o+nw})RVH${5WY1nawzwE7w8Lz}fI##$u
z+><IVkILQreLU*5q9%NnC02G-lkS;KO==dxlT&6rn$I2)T~i^!q4dOj?|152XQs*2
z$!f>mq1pE141{{dzvt`!6ww_u!#)z%MS|Orf7S_>d92hw2!u(AEPCOjucfX^Fwr&2
zXZe#lEw;xaXs&C}Qd4UW<Q_Hj2Pl;NY^R9Cc#=J$&tJleq63f6zY%8Kxy9HH4@aHM
zaqH@<OWyz${K=h4WxHOHblFF$QT@HcN%DimZ#%b??S;8*M8}IY5s%0B)@Cns=8|0=
zyLEn(FL4UG?%O6f;v^Zx_h2rOC#c4OfMhl$%<PZlS!n2dtFaP~c+WZ!<};8AOW1Ye
zYknr;N*fDbc<52}tw;xCo&1FCF%hymqFr?H_IV|(trEb6j9n}x(%wNs2C@NMty|2;
zUIC&7KymIR%P()j5rp(gT)8(7blR5p$w!z$s|Cazcz*l8=NvqKp_FZ2>b7eNVGcDl
z0(z)B>W7N>eh=?`P)IaSh|z1??HB9cS?tlMctrrLpMad4T>bG<Vv^Gj4Jg`RQYHb{
z+>RkjV&9Ua=0;LFBV;<P)=~#8za3SR<%-Vy^5u;(-T`8$-Eb1^h&7+#I^2w0tGed*
z5}o)Dr^eSNt73KPYIiNZ7Sb@Zhn=bHD7M|x)Ro?wW^WISQM$|e`VixqA`?S>tk9A*
z*{AaSghSiqrVoeJsx%MF_pTdS{#&&EPv6}Y@MdT_BqA_lEQUjPe_z5>F5&u%oZzTP
zbM&CdK6LRI=A2t?r7Y2JTg0hg-+@6O*WK-kBAml}%xwM<c#7d)zC7nSaBu2=z1c+B
zxSMPc&*wmIIptc;$|YA)6srpsfBc>M#P}bJ2bFGLJ^kkLsphV(>(%|;RD{~eaw;u$
z-WAh6JVdDpTP%xWZZQ$;#b}w+H8nAz)8;R_splbd633auJNagVQv(0wT*7zd<Tr)&
zy~puIA^Z6F9B%T&n0~#KdfBXwR33-u;#J<4FV&6|k@7g%qNnPAwUE!BKYz1Y`<Ui-
zIx^|TNQ3b7a5E<>%MUUetn1c!Wn^T)^#R*I;R*kRZ3mom_HLG0%EDfg?e|Q6pbUUT
zsvu4Ahxdei;Yc?TwXW!yU^22$!6t2xpg@|cptV>jSq<JtP`V%}y_1keBcvGN^89cY
zL@Un~6~!UW15D=-e#7Z?=Bn$qNB|dGiIot0i@kk6j~rf5d#;^>bg?m0Q1+_=tx*ZZ
zv{oyzBdW{&)T&a0b2j&z5yYZWIjq6K)V+x=sI?SEhZ`x1Zy};-dpRdNqDwPF=HdTh
zx_Cjq*q66b_#K_+gAV7gUJ(=N_>5l5kG?(69E~TP<ESZBs_pi$hvecxBEgfF`VViB
zBW~+4L`tL|drE{Y{=toj#ZXjLruCirsL1+8U3PD^JCx6v<vl6iNrL|K%mG7AI~^_3
zvt6+^7{IIwKiI}&7U<!C!z2I)V}*xnfDpaKq8RI0*{EJXqV?NiPZR!VRV9>221<`>
z<WKkT;YEI+a6JCyD#hhHkN`9n=@?>IlCC{tunTj0Ba!uJs-MG`7UO#R=2QAjea8sd
z(GC(EO8<1~0~Nim8(6I?<id>dsWRc;vS@|6i2Xkj>A-7-OniW_K$&OHkimoC-25jk
zxf=4Q+)mf4Ks<u5<G`dz&Fs#~v(LKwH`ozC|1Rx}2C9fhCS*gv1UsJi{!%DHsI$!x
zfR0(3YhT?v?w*VcQ?LXq6JOoua@#S6_`U8#n|f)7XWT&FqurtBczxmsCJv}5;79*!
z`FNm*r=+3+dnJz$5b2A2+$nBLcL7ywj7`jEYDX2yO?m!sxA7`^t0Z(ob_-P(etk9%
zB4jfI36Jl;SLJdkij5eQ0E^VuBAtd~WzC|gjJK^FvBOHbs^7T1ryf4_ID5Ts^FJIA
z41rxGcgysJtpX6corJO3^l`aNvWb6`@G5St+!a<uq2s+rVwt5R@gx-AIEL;U9iGH5
zd_wg5F*0c8s1hHL?|t^I+bY^Dv?}~8MIxmV6KRhiH(;LMQEKmr&^iz35{v7KaAiCW
zgYE45JL>A@s+#<sQusV$yKRJx+58A6=%t$cha2w<*WRMo%~W<&l*E1TWUs#$Ci~=8
zbHA5r)GJ{^_C`jfmtZxLO$N;bhfPY)l)(1^w5r)6+>n;$H{xe^?$js?Jv7Lp#<Mgr
zY4iBBYDg}i_F;7)W>q?-t9jA}>=8VC7Uku)7~+`MkIX_wuqvO=5*mLizr9lJNVLx`
z*58@!6{yO6__kv>^nCI+1`XY~G`(6d2ZWu&qXR)aXf$TVTP~IeLBtHI8;Ibs-yc!q
zh6s0L>-g)_>}G$gGNAafydgjgxi^4}ojy{4X)zi+l@QC+oEs4e7(;2h`U*(VQMrlQ
zgt3Qm6p)^DmrW#I3p4rb*%IHe8_)yL+DfhW#zP09JNd;YKc5M1i`M4mhq~I58H#W^
zey^_50F0NV%66`Tt9Tr+%0?4E(Kh$vnrv=LH)YDvwL+^xMn+I2nV1m7>KxpC+fuax
z37W|VY5C#@o10XTx6>6yA9VL*#F-2Q6aA|N$PSW7GH#8qhn2Sf1G<Uo^I+w7iu0t~
zp0ErUS?pitzrVYXRy)x)7JP4!PSRHfFE>8w!-LA#dmsC-W3`hw#HBtN<DHz&)$jg%
z&+U6J<cXv^GlM@1#>AvwJ$k^^z*WU5b2D5PCh_S;EE1B(3xx#C?Zo2+#CO_HRaD4o
zYGC}y2*+mUVo0jj<+Q$cZ|S+Zl=BYeTKD}U5BE>Cn)N@1XntwxrE)y~tk(9U(Dryy
zV)!m<k`v+C5hDGU(bE8_VytL+mN+gy@9KTk+yY}uzaGkz=VMKj8+Ce;mpj)*b;y_B
zsC2a3Ej)?U?iIP#zpuLM>iZo7wZoH{>7hyxBJJ}tvW5o39<XTjPBYi6hmEY|)yC*e
zm#UWsm1wD-w`r-@@GTSYe%V+IWx2dK)ZkqyX7A2~TLpAP2E;%dfnXI<<y!s*SHFJ!
z8Zth?E*KAaF_6DGrjMCAdh-K&*~>{@Xk9Hx-0!_XB7c#ab6MmAEZ-}wpvCY_XWZ1J
z-&O~$7z-gyWEWt^;gjJkh?aLZA%xtvac_a#W$L3s%HmQL$mlgECm7k8GfGP%02*W=
z%&4zVf{h8#omBr$4{=>ic5#I?R%CcN2>G9}z`Z1BJ~*0ceD2TNDVIApzMNumd2xQ_
z*rK-9L*sPtwJ(idR8;c{cZl0c)}v}{*;HL|bMp^=7&-bAB#AR)ZQY0&`uN^3&4r)?
z+y9^YWE(A@gR=;oiAh`6RdKG5&)~!{I-Xmkx8MO|AmPsea^=z6s;0jjCZ<L!LoFu1
zINsMX^TclJ<8oNJzIANpM6Nbsi!}8;xAP60Xe9qp;KSYTi^2Yw_13mTKN&j4t9b7b
zOO!~yK>U%5a|tOdOuETIbE*z}6Mnx3_PRB@wW3Mg-9e?#Z^_6!r&rCumy|>)Dpk7m
z`K59%GYLWB;3$I(6E6?utiy%pOG|t^HekIu8BQSR>qk|y)3IlH+~Vc-Wl2R@g!c)9
z-%YP#4(^Hma&Cz)d@OiQX&>U$*lvT^tg+>jR2>-H1a*FAG>TGvb$9or4MBv{bgdDp
zPM0N!nBI5z7AIBoLhJG~vqr^eJ7e)ew+Tkzfwl3-AGrd}k{HQZQap2SN#vm!t4wBp
z%FoX)1y4G;FFlXxT?z4uG6@mYTM+yB=m)+OzY6iVoWkmr9nH@zhgWy!Ad@*u3G_1a
z+x)dfM&yrt&*{!9fZKuG#AY5j2-d5Uq4_u_^8&Qia7@JMBd<Q5W!D)6PO?C+DFp=8
zU~-+Rwu5Lnun@)zyK&_tJCi{4R~F0!0hC^GGcz-aCUg(vj9PusG6&Jw8^0vCM%W3T
zPC2YilyI{T*eez7f4`!WT({*eavg=DyIXT?Xv`;_q&G<*Ov8bJ>vLt2$K|Qq!2_pA
zxt!gg2SBZMHH4$5$0mfV3jRbve{@Mq=!KB!T0D@wZPRR`j@KBgtg~@`eE117l=V9g
z>+G_ec$ar)GL3h2!%cgmP(i-=6oDs252~(pf1j}59OPNg6?LIObe;d<|CZ~q?84K8
z$XJf4JCGU>JgXH{-Wl!d7jY?@TJF1h;^Q&3+4g-*X5%6_Lcwd<bcOmrO={t`Sn<>P
z<*yGyC#upsT5a7%ro8SyeeY7=|AWz=ey)~AG{Lr4?8VM(*+(MmLM>0{b~EV_U(YI5
zk72WicN@>X&V`ehwKq{eGGh7}+#hV;F7l*t$SFXKh(-(@_(3N0{Om5}|91X$rHYU0
zFUiIU41tDlE=UDkmmMFUF)Hfi(fLH+vO?>##fgYT%}B8SIUC}Ls*Ahuc|Ks(bSEH0
zzZw1c-tWUlw8ppzg5(d9{rP6z*v38N?`2>oMlfr&7FeaU3xovUA|gz)9A_~eoNH^t
ziIIIlhHy7cdS*<1zw6A4Dz&?3fLa`rft5DO9J`5k=j(Xx#l@Vig~ceg*x1go|7nI~
zhT;1Z3R|*0SF&>=F@uQ_J`Qiy2mOV+_J@?8kE(_i@N@IWqUIrL^x8GUJ1Q!-{y(zb
zIx5Pj?b}vCrKLey9J;%^ySuwvx}>FL=x&f05D*ZhyE~=3L^>tD?ftCxUGH;$*I%=i
ztdYgcwXeO;^LHLcqqX{l6CGWMHTIkfS$2NiR9>|c_lc%j;<Z(w&0>EE+Lz`FNyi_}
zK6~augmYZRw#D?1zqqtHN1ov-|GIVvIFS{&9O$7(@>?UVz!Lp8ngTKsL};KN4{WQ8
zH8S0=RdD`EeBj}}R{<jTJ8Y!<H~zoC4UzLD0#L=P4*w@J4q6(uwX6WIUt%-4yzJjO
zxxVadkP0dRAnF}+-}M_X*MNO5&bOt+#_#$kaDXV#4-F0lWkH57dBGe3c)YS3SnceW
zCdxys0qy&D-53}Y{D=JkZwa&oL$x65(ZvjCvtS?sH|6VTtio^m`Gi#V_b<ct7I#c_
zI1@xs72Ig(QP3Gr#-}Ia9B;l~6wvJ-`zPe&KyvSXD($W=&dy3|S<SuvkxKecVo+#8
z1-6~f#I_-*@^sNAducjY8Y;@+s{T>r@a0v}9}27NsHx3I=xgT5KutrKjI1aVIB03h
z3)Z-=#*Lu(n0y)Ib_3>7mFYcZUZ;Qk%xiV*7cB>h<*aRgNx7+E#cdYs^JbJJ$`^kf
zp^tiT^;wuO);H7-y=5=Fc;@S)sz$Of^QQu8FO+N$XTE<VD=#<3=S`Q#HJFly0Bp2E
z5iqLDrJM}2jRBzUfI8C9H)a&#moEd!K3eF2rx9-=B{m9bG>t@&s9M?Z=kFTI%BOdJ
zVxqq!R5aeHpo{8X;7kM6k1{p=Bk@=J7mv^OKfHj$V-JZ;qyk7@@6F3O#8+bGb#Fm<
z@C2VaN|raVcN|r906YI9Uf(xohu4cIYu_0&Xs^KA*)Mhu>fXQ^s!ROGBiK<$2y%v}
zT-J|FzQ@GGbN;I~LX9I!`@{~<&>_z;A<te^nkIY^bsuP;_gttYTKq}ax8M&9+PuC{
zuAYH~I>N=nWp0N$nz+RIIEo&P6pF#}B<d^0&t#D+EFo{@s~ejG^@Bgt<1Hju@U@Qd
zTsvDj?>+{!@!^HI{plLJrW;86L=Ba$QI}*xQJ)EO#OKzanvo^e8Q<SOlF-Cf%;A3i
zkxVyI!=M6`0+MY}QMwLjyz1%8sXyLEDQP4@*bSuS*vn^VgUQH-kSHlHZBeo0{vbg=
z9ysR_zAEoBAr`}CL8*9C*tMRgO;0X^nR&-m1FS|Qe`n3e(!imF`6Z-77=F+lRVQUU
zwqjQjeS)2MVp=!}gXJ@^B1%{HIO+#6DNUkTclXP`ifjtZsPby<<Chg>`a!ZB5kvc;
zD(1HCs50oEaUw-)y2~3lUjZM*wrkr?c|*;AUET~oJvyokD5c_%%vn=V^c=J6>H81d
z>Bco^br@QpG8L>FWophlF&f0DbVfl8dU-9xohCz6P5QHK9eMI<mC1XWq*W!B-62dm
znDn@ty~KRd)rZ^D)=8sRht?Pp@LgXg%`Wev()Io%Ht4&$q;G>1hD7?54dNGN1h*00
zlqAVuQ#2ZqMspx8pJCF+H4kbZ-0YFC!;+wDEw-7hF<`W73lC2Yu&e$SR3h`?DfRYn
z)&H{<$#2Mt{RR`0wwt|eZTDiW0evd$TB1<%wr{2a#u`0A&5<=-OF;U_ktjui;oMYt
zvP}ArA366I?vK|9mCT-qBgvk&H;lc1pU(Z$A7iCmR~v~icLSHCT_^<FUY<L0dErw-
zv5*-vwGR&Y`iF<b)JTjoVF8!D72qQ?tS&TcSukWw1;bFr%%7+M0F(qE0a7dgvr$e?
zZg_b3S;|MiOa(1hBDB}wIqlW8!yq$R-)d<*kl?V`@X+qx2#$!Q0KWr9462WQ`o(CK
zG!rRO85m^${?!x>)L5Vz01gjb%+B0I{0|pX;6DI#vvKyL415fU1bs#+2qqA!2lezw
z@bdCj#n$WFo+cv_MNADCczSHQ%3z-noHj5=53nal(^-uvB{0zd(4`wVE!rIeV1+gN
zU3hjESBPzW?{(`n%IbaZoXgK;Fc(9Ou=gXpYhO6U0zmu$=L!IpRV!vbCV#_rR_GiZ
z8Cn0pyT33W5;>5(PREV?I*sy&$+KVzLUFka#NbOm6O-quLjV%oqFLZ}*uX&Kb;a|F
z$<W-=BSp42oJ`tli8f3>!=6!zGV^E3njS6;3Yl1%?Cieb22JSF19Ovrpp_Lvb^|j%
z`+MD7f%rK&Gcz{F#%SI8&fmXZ06w1xb3#=~tho+2w3y=aI-KHO{b@-U^SfjFR3QB>
zS4tE6AC-Y!pYhF)D$vXX&kEyl=%itFgt~1)reka52imTsCHhOB*pd=U)iQQ~oO`C>
zPA)?pMV_e0lmPr9C~)A<?aTk^Be`6tnS%KgVphRN8b0sE#Z^TsEvnqR2O%<Sqyk#j
z+5$5DqnTajvb@aUk&!9vXjQtMGCN9%uI_bpWpj*(jrPMgwco!A7wej>!noW{Z6|BB
zhs_9hqv)sR=lh3$^OGr8&!9unO`F0G)XQcMZp5CaKxJN#ye9n{-xj%ew5{~GyUQ$(
z)%CU2?W~CiEdmam8v6{m2v2Pn=4qo`FO_19DTwT52+D^-=60=U5(_>vt$Fy;q#|J>
zzdXoz^+Fmd&5qs4j@s<_<_Ds?f*6n|7h^svHpPGsihtN~!nCtqI}bc-bC!Mq73*P_
zCqjEGQlY|z0xwhR^Z9@PEkd+L88SOBEY(L)BE_h}*<R-3+@w!WPoMWTJlU^>*4*w#
zP`=%#f>0BLX^vRi`fxI9F9X^rh$>9oWf0Tdm?m-i&(A+aZVx{C<%b%P0)+o{Ws;7>
z9Z^Q0Aww|fM5ttH{nYDi5o_&dk@QA}wKXuLB<WYp@Ke|mA&0vOpD6m49EaZD`k}|l
zc5j;S>Nbv_<e(*$so;}@m*!Fu35=k&JrWY}V|LW|5Y|j;<f}94>DAln@T&M736P3-
zISlk{Lv)y>TWb>i@{%iQ#t(+p5g=sQwf07WP7Vu|$yd6J(O;-%JLz!bpfphw;{*hi
zuCb`7vyVfR<Lh60t7q00tQM|?6Dj7=a%$t6JYeqnsY3)@i8b=3fkGFTB636Zn>-Y7
zr8>gbFK_mw6z;fmj#g4qZy%;(38aw*>3+q`L@2s7p>_Nr;Wb>K7|eghaZsPzgL9E+
zB3H|m4tD3%*gk5#ynZ()nXIGNZhgd_&Y%+2R^=b4PE-5O-5ROnw;|PfEoXE$d6+Xd
zsJyE6{tcL;yonc@AqoU^(GXi0Kv+#5y4xnRD$a<&Gz$$Y(Bs4x>h185k&)rWM3@;|
z`1twd?d&QvNYLasLBkdt9!RPqZg0IgKX^_N#MgaG>L5i>E>oXyk1Z~y07(?b#>}Bq
zFt%BXjl?6b_F5d&z<{dc{PvS%Cq4aeomrnp7TW6!O|iKHgkWhO8aIbFb=t70eMsgI
zEA!X%bviBJYZ@LMf$GXnOigtK;k_B3ni2!nrs(Ksc(HflDg~MqwcoyV&pPmZ(X5b$
zwhv#h>oqz^U`qh<&O|+LzVS13YTP04bw*JG6Qyd5sR<V8;MYCSQgaLld?u6fMRSnb
z976|tzSf&AF%gTMgA1yoYYHN`MQm?rw~<oA!}~2+w_M*g8<N&;bRV96*4O_gNC;De
z;CE9X(K8GiBdj_99Kbo#Jdg+Kr)Kjy9{Fm#MrswhWMu<J-q}jZx+Ig`Kg;TaIQr$l
zip58xMT>pr4UuUdXV1{p<MF?T7t>;N8EW<a1Pz^97e|LLDJivav?*ySOXwC2PRM`v
zYAm-e$)bFQSGQi=fXl^O-};g^c<p;eE$hS~Z>?A_JNv;ULA`D=Na27JygqD_$jO2x
zJ<r#Gs)TxCrXDwSNP;rc4EP0N1zN~HsSjBIfyqC2#QronC8c04=`u4g?Buf$Yy$W<
zMAZh&Pn7GFske<qP$S*&Gumn;x};nrj-$kiQihQY-rPTru|=H<E!UL55W$nDo#8Dp
zL3gmvw^9>YP!G=ZJStHyQz=v5(`}0rCzk=N{gQkYM!mVIGAt@*Ziu}2hY{NN{p;7`
zhZ+YzB1;aXOFBALadYRQsec6;I@tg3D}W%hnElcxH2>WYh{(=TBBYekV$(O)eMdMy
z*F>k!ZPbUFXk#)Yd?hKJN-5R#&}2;A{+N+rTwq{%dGJctOoV*I&gIfk6rWh~{;;iJ
zY;jQ30(m_)^Rp_A-N)lpi|UMTCkL+4Xnlhv)Oplj3?r@Ojm_1Kiw;UTN;VMT(cQFs
zb8wJmgX?tY7V3#Y*c9yUbw^}#>%VDpywTHphg(;a-e<<}K?l13!@aUJ*V3l%h}&Z&
zigpIQo<Ot-pa;9Ij!u@&$7gH=&>|QzztitvB6^)Z>xd=19eV?-9b|c-{{ucdySl9D
z%Xcxaiz&#YyYTY^lpLa>(9E1562tXT>JrxJ)3wuk6sDrNNPoOzg{@SXhI>10Mb{RU
zzPq{m``2d0d4@EDk1#(@&3FiwOaco#ap{zc0vWeBqRKzp2uPZXc6`>@m;xURZ4D?~
z+13+8b#*NLrU!=<KxHVW1u@b}gP-BjkU?Of0`9VIy-k!H(jw5a0JHW}WvqOB3QhXr
za~!h8<h5JE*f7l40{YFX?>0^j#T+X!CHnxShB5`gMwy&n>O=<3C@`v+mYoeI81wdd
zC!|Ar!(V|qC^(u;u_H=#erv4%Ck_OTPM*L=esBx3b9Mt1_9v*ckIlc`3=kVvP6BpG
z0SihO1pi{wwG5%)<J7*mdpQ~Y95e<r#p3%USapOoG$1lXx@CFNtjV%(OVx_i>DN4d
zMmF}Q=Vlicwx9;KkFAA8f0Sv}?mjx*K+o4^%<o5<kWLb<`5$yT86ss*llNx`n9x8g
z4X2j-*se+$O+NGqD$khNwW&FfiUjfoc6x+Ao_iiI_HD)s#TD=(hKr^tC;-CmM3W<G
zx;{y~&_}TpZSAzureYk46tJi$L;j)U@@o*Q$NDbjsnK*GS~Q!kKb48l%$JRQd@!pI
z#1ky&aCViy*zrRB>KDo?*Dmj#1<ArBxrr!oqbFy908-cDc%<=bk|I_t$7gwliIvtU
zFS#t+mfis@Dj=L->rloIWCkl73cwqJ7Bu+;La>SJdgsfXUA*ySML$8IaKvyM<^I$Y
zZ~lI?moiZFml!J0vO`|9(kOa5^)ns#yaK&qOyY7Z8SDh{$R8#a7j!eU3P;4<i&_ao
z+x@#p(zU=DWd-0VbX{6sY;1mM*1H`H(;#iLqD3Hg!84E3S%?Q!6cbU%i{xo?IhQ*E
za{78|X>LCFeg5BTgb69KGI2K-pxhX^rWb59UL<WNvqsCwr9k8;8B7|o$l->LlVmmD
z!1PCvj2W~fk|1=-GxG>)YRm$t6`DNOPVtk!h)UM)G`OvL5fl|_ljK-bkMk=^@!los
z$oMi0)gB+qSzcP}>`RrD&r%L4m#s;UY@S?AEt;LpM7%5^%=%CrKQT_!Ni$t3gNKv?
z8M1^)^lqjw>oF$GPrl&u5cX`O=nobdy64HBGheE>E43kbe1$i~Mj}5Vo{&$OzJ-(K
zY`<eC=VVs*uC|W`Lm6HliJLf&7j-ECUzOQ9+gzb;_U6r#@ulnErCr0tw93K|wEEHg
zK9cnhLr2{^xf=<1Y-=l{m;Tkt=65J#q$IC*(RfXTLXh{Ux>L+Kow$=_LDWBRW$Fgw
zp0L%00RdoOcXoFXef+tzFry$k>pT+V*|KiGZ(JtJu)z$cE=;z$KKryr#nQ?Oh#S7^
zGlCZYQ2;z4w8{TGlKcBKS&GT7#eh*+pj$S+G(B8nCIwK&dp`z1v2F%CR=7%8oQpj;
zh)maDcXE43UOpzlNY?FcSm(uap)z#>6)1Z3qlk{SHt0zM80n`TQ@|l+7T|D)1sVXI
zt3Km-y;R$)7xIdVXQ0%fYw>jE4JJefdL#yuS)mXXPTn9gS^V(0s<)w?0hp05MLV8#
z7=PqD{t<mKNsiWoSR%rVEebUUT~Lri2BTXQ7X?$1nzZX@p)Bx0J}sjiViC+AvekJd
z*fF>77wU$GM-xW$woxYBJ3E-0^ixcm#9N@fqzIts2%I`V{WY{*Q0C6@L$gr3{PPe_
zC1%<SuZQPRJukL(lbs<kHq!~Vlyz)w?Cjm>kg$a&oj6jV(TfZ0a<%Or*?j4lxx49B
zIVJh|VfLIus<b1IeI#bQ38E<$V7Iafy!6#ZePxMM;|K0hP4v`AEV-+^TH((q;U7$D
z<q)GBsp@r!>y*VW|9;u71vi=49E&{IHc94dL$J}@_Qv1oZo<@~Ut$hGUK?_Q)h1t+
zacXu&u}l@n>eTRmjIQ0*fi6&fCgr|*)C65BhIoQszVRg)HU&UeE0&6KCiZM!DZ@0G
zB`_j}#3)l+bb~b{xlq^@N)2b)={Ma5q|vCX*H|mkEVD%=q@WsEjCh-c087qeOK@><
zdA@wiFCv%=+B~d+ECWDjTDy^_N<%J(6wbKj+LQo!slp{aN>L{JLZte&NX2_@9NOS$
zW9nc>QOZ;?Hk6;!1^@3&3|@~Mu9q*+w)llt^Vfw}3vC0W<jO6YbJO~UM8$@&pi@)f
zHbZKi;hg&rvdrY*I{V0(Tx@LK3~X)(&)6El(h8QR&T8X;jQrx<0TTv${_EE(1eoC%
z?ORGHZS;y@L8ctS{x^%A+J$6Gn}YAAyO4ojDKFE9EGX)-6z!k@k2`HE-MG}L;JfVi
zKuE4wDoj}c1SsVP9fUFY;dcG?n5OkFk}-eJhfOqw-D#0X1k5cayt+fb7gW}OmuT1_
zm(h|FcGZBIjL;>D_^y-TRs1}?p@@>OgLQR*G8H+M(jXD{D?F*Z(mq=$OATM_Qy+_^
zKkm~6bL;e(-LE1NaPZD%VbwV1&sqhR+%~zmTb^75&y7gh?g~Wv{GF_R{ymtwCyV!c
zWV+dL&R%2Snka0(Vs_$efcRpS%^l+Z2<qDR>PnOw^8dD0U-m`oYUfS|V6lEwoiB_9
zq^#a%uZlp6NS{jPy!CI^&jYuA1x^RFA$=+kuPlR$XNsYl?#yr}v_U!WBBC8ArXa}3
zDAS6%xO6Det|W^jmutezwws&8k5#3{nLtep>*mkzrTpQ`g>mK!p0g-MP2F-#q?Uc|
za~LOoKZT0Ymz<;go(~a^-4YGXSwZFh`DF8o6SldZs^t7|$)(#Vjk#y-)O+lKj(B_f
z=W%!8r`boT)(7}qMnfNs)1lii$TO{~nC639R-n%`?8bg3&v|v>e7)=>qGDcBTB6gI
zpV}1m^6oXK%SFG6Y#6y#mHy5q=@Tkuq_<{$<GU!#{SQdbgY+DKX1iA`vY?~maEal)
zl3^&GAD@cuJKEVwck3=B;XGdMm7f5|#ZN`t{oB#zY=F_~%N$F<)ONx8=p0JyTh(`b
zozIIS*+<pn%+pc+DlurKK1L~~hFNH=%HU@tJ3A{E*O1!k1LL9Z-~Hpm?ePvgoyB^z
zwy~~}#@}M-Ym?(IL7sJfd?t*BCMGtHRZy^Z<?x!>YyZV!qeH2``|FV1G3T`&Jg?n{
zYc)o^e?PcLN|v2v<zq{d+9&4Z^d<z4sm6a%akkRY#MVBvLBj!p3;h5DJASr^8tw&i
zveHml?yc?Pwi_J5`=D#)uVtzB#iQAt%c#i|@&g0Oa;X!L*Lb5jCNl?$;wnW}W?Wc>
zo9vAZA!L)|=s=#tov4OL%Mp^s)edW1iF4dDpN!^zyx9cnn6-^<o;ocHN27FcT5+2v
z+=6_HmmD0U(KE`O`2)N(c{`$vG*chT@JEGA(6lX5t@!G_$T^zJ(|38jBHdcRBH$X7
zqK!X-jqrt4m$Bzp(9xIHXGTx60pdL&bJggGQrv7%U|jm@VStzH8mUpiN1beoZTkwN
z@l(mZ*ZuRebc(8<j(l3ZmQ5DNB>c$|Ss99g<MJaI#+I&GBQ1p5`tp-Ylc5m92P3|=
z&Q%%C|GI>54f=?hg&FP6m{J?h+S;~D@-bR36%JdEr?Kf7k%ou0O)C6Wpk`sgm(9BM
zp=o5=f`gkue=eQR87Pl^s6f4fb$WZ~<mtZcg_1x$=8~@C%Z6{}#bda5`LGcQJG%bb
zfz$H<%|4Hb7EA2jJ{WPz5CLP|68FgZ*R8ZUI}YJ<9~dG{RQ=fFgY$3vQ{Urlvp0@g
zc1pMSq>b=7)Ve>pICv1?iMWYof4NfN(_c9&HzyM(<3CU!@N`fFiG78{n^$?UY9m6E
z6_~s|KV@esDCUOLuvF-7tmR#5PgQOc3D{;wfNCM9)lkb4(o4Fp9(_m3wC<*oqMx%K
zI-n?FIdAS6+qMHjorv1x!^sXEy||9s8qF;5Xu58aQ95Fecd*_x|A5~dL)L?~$f(oe
z{Ih=ncCGhbe#A+0P!FfV)K=4O4j|Y$FFG`WdN=@5OGCw545>iWK<Gc4?v<}Bf%t(p
z5rO@H8PetSbmi3cu8$2Rr0pgeAlC^UfwGwtSR|6ccNFx9fioZm#9<*arXC*uPKW?>
zkB#&oJIE@N^r`v)_zLdOS08(?L9yufVjXcj%7QEE#QM9n>-WfHv}m<*HoxX>zSzAI
zYzz_GkiGpsSpe1I-~*)EBqrC>4T+Vfm1{D@S7ENnJ69U?MXb$74o-!tjO&*^=~*%x
zs1apbsn+(k<a)#B!g6#-?pLJTab&XkBqSaC>ifrkOwGTF)7A^uXB9e~Q$B7vwMmy5
zKAa-|714Ic-}`e!(OMqj$|yyi-o35abh<;xb22I7*m6!Xn#C;%)6B20$5E(Y;oy)k
zs2<0Rf8WpQZji@F4UqnW#VA%|M#erKIq&=8Y8?J@lH8q0Ne_>r+c^}lVV7vaRA%yq
zHaFL+144RJm?af<uAlba-2CypRx6v)PD?kIi;kxn%Yw{L%GGB`y}nuwpjvlEsAlS4
zaNB*Gn<0X=HtPFl{bB%Xcz7&Pv3P!Zv8&~Jsbl@URJxJ2v883kb=zO!7^Sg|^J`Ul
z%TwR%?}H+^xV`felk#S2SYVkWKb-x>!rDMEuue%}*S@c8Z8$Rmu*IMHw$077ULSwo
zMGNl>x7jyRdi&bIPOA`sHA+oXBfGm(1ec8xrjiiUdikxs5hxQ`h$-PRxe_yoK}qh{
z5+#p9ez7|a*l=tuv7@4@{P4Z&w{qjoW(Dn}BqSa*V$uyEAh?1+5rdzTLaG1l^afk5
zm*dN&vv|B?dV62ViwKVg-~aOcQ>xCW#;JhT>X)>*8ps8p#+#1OZZvx0&OTmaaBqMW
z#juUEbZ+b?iQng~dm@pDs1@}0OS3(uoJy-dC~FI?5PgB(oanU50YNhIt6piK85LVJ
z1yKT{?F{YujDk3!UXzyvl8E~h$tAN;?%LxufBV#`FI1LNgt1ZF%#%%bq$4HU7e70V
zeC@oZm7{;@{>8-C+{?q+jSt)_{YFmzT^tNdOmh#36iw?>?72N`!;yxpUk5mC?>abW
zhILToeC-}{-R@8wK(d}WOu=^Lg&l-|QHk7um7=EGdNY{b>!))!O48de?FsLhd2H(M
zXj8S~^yqc_P;Ou~93xwkJXFh$jTSX2QzlWhlbQ0^S{XUFeww+=bWjTc&!(zm14wf4
znBOatbIyyh<!1X8VM;1hM!ksD+8)^tp!(_h)9-DLucGM(_TC|_wUSKw)46$T*xY7K
zu$|&=6m`W+Wq^HCmiVgVo2qH{jE1Ov8+_vWL3!NhS?>x>n0oe<jZB@z?fg7~Bc4}F
zO67;2`Q)J(EsjOUOdL%y*k)@6q*iMNLR(!PPCE~Q7v2P#Jpu3F-+sI(`v<WUI+nSo
z1ykeC=YbpuG$RU{Yf{_6`5)}gKe)Stv*35XTUMa;0>|C=GME6(2h7@BABK<n*cE{P
zu*}XKaBg$#C@vm3fQKHu4~#_nl-tbAjQ@Jc4Qxh7a|EIxGW_=^UBE_PVgrg0irKsp
zfq_reMUaYFT6y`nLL1(x#MJp(3oN*Y7Wm+Iq7%&d%J_{#pdoPINu(9d=z2EoqUl*>
z^{%Nu9XZwOv3QCbJ62omy#tJ)zd$%&y`&75t;&d;MvX?bEt~CA+^uR$f_3s;KpHCF
zq;5%*-MfA2;$r>Dw!Ra17CacU_`2{O*Q4d-6<t?bJY8J8vQiBNWnr2>XW#vyS`dEr
z93XNWEq7eM$P!PD^LYNDZ(RC-EQQt*Q_xR8JT?L7bl2gr-<N62K4eHeZ3B04qx0^v
z=h<&Jru-iQ{~VNln|AT46?5YsP~j4F>w7PTQP#DS(ie1c9>5{*;hlJ1F^i}{we!yK
z+Xnl#INRdUE~5t|;L=|2J}oT5#gb>-;r9GPy)aD##;8OqG`%hoA~1rS1n(?xEql|m
zGp$_xf^VDNIxcw9Eg$|w_`9s%>KrG}DF=!auo2XyuWo#iTjb9JKhLZ`V^AR#&}(32
z<sAsJ$}MRsPCsn8bh{WAu}npRFJuv)f~4!4J9u&>D1V*><P0yi=rW0ef{<oQf5(0G
z4Cl%qb;=0ZxAYc+D2n^jNOkZnjQobA9okq>we7(Zgn&T&5#|J9M%xnMn$M-6Mw!i5
zEyix*V+&nq@_wNhd7=8sh`O`<N)x*@x~B(TixIWP(SZuol8sg=gP#i(P=8h|_F7v$
zZWlK9v&B?XrB1T*;&;sPyLuME7(4!}CS^-C1(2;{rGYn^!|qdP7ysEFmvvkt8((Tk
zwH5Kce$dL}h<r@UnsdLp!b_O0xx%zq2aZ@d9(pXVZi}Lt8{D>Ra=0`?eVVB)W8~N|
zIS-qJnAv;B`1qbMo&p6I&<e%F8GU28arN_<GI?O#)%(kgySmZmiF~=OUCaJd+REc2
zki2}94ZvHtKL8wV9_O`Z03Zsg=a0>$;2ZvbuX8n*t^R*=(Z1JzC~*J2w9dG2%F>tF
zXnWB4C`E!poxC=c%L%NDTE~e42r)73OG~opJ*F-a(8z-Aq6WvXv4CFlnQ~G=L&yu?
ztM<o-7w6}6GVER2G>dy5bdTB!R1jX~pN|Xa!h!mQirz}O#$ha41VtA)2AvO1w2l*6
zgqC^ws?!4@k_dfxnn6Jq(R(!yRiwuCQ-tPg0m|)KVMeRS%KUfSo};o;$wRaJhwgUs
zvKA7!DYA6&$S<|Ezsky%E6Nd2s$!Nhala)BjYO1d1=Y?P_;uUjUah50sD<5KBP#t@
zmqpn%D?ia9b*+=zusK|%4F9OMn{%orDD1s=A^T0hXE>FU*8<Hf9D6JHzaa~m;=nfy
zOb0+Gc)tE7@CH>VfH75UiXAEJ#j{jK19FmhHhz8}9|rN+ZB)|p*EQe2BkT}rojHR7
zHOQa=QwDJHf{Y`WHv>*&pm>=a9~U+<B7NWWp1qMBD^}cvH`}RyzAeVuJ$9f*#kGf5
zprwUf!XNvbXV?c(=CUA@JlG_!pM0PPB_KXit+a$X8M*ml<GJcSc0o}=?;I-|M~IYr
z0@IeuZD$sbN9f0NiyPOVdV@m*De_?oD14_1W+e8?j3IwHT+O?|?V;4q#7GPFIjtjl
z&b-8X;iD(R<TJL(qeEIZKLa0MC1-@+R)5ztO+X+4?GoM6{4!93k#v)zE9;S{T5ZZ!
zt_=F)p6%r?s9FYj0lfB~eE&4z8L%<e^NQh(o^^YMPoQF@%2^DhjAZe;BgtU@TX{Nf
z-OQ(G7!3#t%-pHe*rtk<s7}6{7oak5XXD+T2|WrB`f{#+Xae;>Iw`ca;dQyGZ}rvW
z=m@TT+1Kbch9l#*SoV8{V-uTH-w&&P+8|OB@$+?NU$=%t?<yK+yqd5^sUB0JO)si%
z!U*J+ojx=ctPwU-dmn-)`TgWfg!6b$Oh><<PGEE|`Q78TQR`@5;KQmKj={GE&J)kT
zhql|3*QDk%865Z}iupwt(VrlAn8Qr;%&eRuYI4ZF9)C76<NYK2$>kMC>Z)3TQZ$H@
zzIhMc61c0iPos)s)i%9IWqLEw;wTXLfuvEbb02klq6#{KgAi6&gjM(5BBV82k2*`B
z`$9C_wZ6A}#;$*tHA&V&cf+A*N(LfRpgUMtj@DqmC@E(alfG}`ar{fB>9d)O3&z0y
z&irX2_L)S(mA$c}X_jm8@CxBID<@0XhBIhK?e7zk)qM+{J1WqnnOa<wV#%~zY=1Rg
zQ3Rh*zF}TOE@ly_Ee4;TQLN7Kwv{!e@6fDFIbVN#T!Jc~G}_t>S+0qHz&rz%jf#)0
zY>L6U&088zv^7wgW9IKkWv5FZgH7B2=ULp@3lXJ{s`(!yZ#S7t|9dxL<g&zgbNw{B
z1$W#yDSJ5E;z6oWud141iCCaJg2|s@DZ`-CaIfac^{wIYhJ}xBz2eFKZ0+?+M*Xiv
z2{yW!)L=9buj&uz?nNmey@L6K7Ej8K^!Y4jd<4rtWpPiGNth@d1)j!?Yz9LrP4f?Q
zqx^O;H>N@+0Wifv7niv@s4u#qx7q-|GM$usR<2&zq(M7-zKyQOXob3Ytry9fI+W2?
z-g(rR{_Wex$aedcz<T>_SyC>Ax)M|T*;{(-%FmhfdQj$6xaHe}C@;sxfu{2p3=Gv(
zoQNL{P47E_|C)H~wE83FVP2`zM`Aq>=C3s&d3RU_xK0G1B$D_U68TSwg|C%)W205w
zPCD&$m@LrH+O|4--PU^F3M6QxQ|t*4(6{gg{%EOURTJ=d1d-S#{m1m8w}v{Ls@M{O
zw*yN5#7f1gLNli+Q&vN}01qAy@<{EpI!z!F{cOpR3oRl9;6CRnD<Vjs16tQLcQzpA
z;LLp=UbL}-A}&+J(i+t>gw-Df%r%q;KjOqxSWv!f)qP_RO*0%Wc#B5xKmx)IA$^DM
zi6Y5|SbeWnq$%q|h0u$&L(21&8B%%9CXttFi=%BZ*r)Qe`!VCfPs0nWb&0E2`|h{k
z9bhiKRFvs@X;xj~j<sbH;vME@o>Ysk@3@`!xQ!m~u#(E%+wVwUm)NvE9-iCAQ>7jN
z@us8I*dM~14ZJ6`u~}3UL+-J?J?`t(Kp~)Q%y&J-1SV8M#fT8qcM%XoKpPDOIty|+
z)^NwK=RdObn_MH`_?>j|KOQ-;a`J|QYD&i&|Ij{?znr^cIVGwzY85luZ6<JCU0p<8
zk8hy^FRxbo@L<FO=7VUh7fMFrn8vF+hJOx@Esy@_X%FLm;^3TI`qlBFb})y~sfP-=
zGB;BU_vI*wf86LOW3wZk!iYkaP+$sJp=%Cc*W-c9jL`GCe~sMld7Ia_I?FXt959xP
zAd+(FZyY@=c(^up$HpfcYC!+t4?4|WAJujl>--Nne{G^dCg0-5zA905?_B3)I7M4R
z3mZ5^ygCld=;=Y}Q_Qxq_DZ^L*<|202tX0bqd1__F(_zq6Tq&+j(+KvVA+{ZCb#gy
zYBVeV9j=AJ9IyIJ<RVO!7T5XbGPVX+m?(z8l4#(rB{Yn-UiW4Elt{g9&(xB5jdD?O
zeRf3Ky%N~VEqeYgXmnzWFcj?X-$N*;NB<Dr>~{vKw~dWhd3xsKifgum1b&hA^@z5?
zORgnh#>D)lGQ`|P5mOowz!n4Xk36-tW6}qLIRj7P4rdevt}pvIB);}So-4PwF@718
z^-=i9(BbGCF%cB3QH5b;IvnMexx#)-#?lb$aRnO>_z89y1QC#7<7;$SKYjSSuiF^c
z6b9)rP2U$!*_ThgnYA3GAm$UY{J)!y|9Nq~se-+C^nk&qUcrs>Wlp1f21lmEM)OP#
zoxW!CRJ#K<=SHW6o!>@kC}=$?M8?cwl1!CkVRwE@?;_dasNlRH-lR|0Tcui)k*-lO
zJN9eQ<l!zE;a6>wss^3Sj&tR-7<Q;#Y}Z!Yi`ptR>!V&9g70!R=1oEJuUOv>^|h#4
z<3y(nW5U}ZHDzu8Nr}MfNnXe4;gt#trsz-Ua-J#Y;IF)3qD!u<XHs&vMPiiqvB~kp
zI1OJWi~9CWyJfg)r8SU)q48X=jc|8*LF5$^3M)77=)P4S&%<vn5f`lEm0{thXK-+t
z`!+65p;`lDXrL$(i4fd=f6ov80O}PQpy#pe`nGf`60`=<;iG!MBN=#4K+SdQ+#57F
z3#ftItLjLvOdTY5Ko@gpTNE8$Tm^!SR09U$K$V}6%TYXK+wMgtbBx{D{_ByW2*f7R
zaOx;?)^sgIhzfesen}<(arwob+_`{W3z*^iGGqHc38sL(*1(h3f$gpSf5F-5xW2y4
zHCyP@T&2rx^q$oC($X*xya9uKwH>OEleh|KKdP3`F0g(@4m1nhnnqZ0hM!_i2w^qD
zO1z(0^-HKJd+YeagzI|w2@NDv!5!UTC$KJ^G_j6hD0qX}Bp>D9xZDdm7js8V=}&h}
zN^ZxC$E7vz!VjEHOD2=tY&frBW<EYqR}GqA!Yh)Ci-l8A#BPyRk%Uh+|MPB7Y8-2-
zbirHR1yPK{17?<hsIB~|{F4>WwrD=bmSowc_Tgdn+=r#z@2vd6CHq+do+-q0M{E5P
zW<`^7^o%8YzYn%LZ$-fD%5YQN?oE5h8**7prq&DoWZPIk>Q-jhj+A_%PolH7v{1Ko
zu7fSH4!Ha1T_2t4I}79ea1`de_t!*fE%3kpVKk4xQmg)tzB*HkmqbGH_-N-$K3tl5
zct6o9P3fl%8qCbnF|^8k89o@}KsPq$(f|c5=RJN@t$!cm!3?XK8XFP!2{~iLuyCLd
zXuufErDbHbQ^=(iMACfnJLgjZ8D0f;f}OkNpCF9N@4E(X$F(ZCezCcAZb5FC+_c0V
zQi|qs<s2^Xv)arq@Xdra9F#khjAuTeOV8yCufeiEJ;e^s!xHj8d*Fwc1hKq02JP#v
zYSg$cO>iLZcG?(tuJyY$fGIW>RXzp{!^r`_T!Dh_B&=YM65Y{~KY@I9T-dBzjy&CT
zEN6LGLE$6yX9L}h@%66m0t_%^Ve_dQ0bUG#3G;b;^u#G*>qGa{4V<9ImPcN>A*(I^
zz`luxM~*(HRnL!?&*Saa+&}(SPmoA#+Gkn0Y!_3nLp*C@U+9tWIBH3kXN}YxP*xaO
zkLAVBG}>A+O<s9UIwP90=(yB1-?XV7?JeJkSFhPqCf)3O1Hg4vmQ_#_noJAhyxgO7
zq6*$$JTAYv#{{qGeLa4gxUb@-&|2W)o;%$$egE&~Pc@ljzO?0l0;5$jk-lOAL#ibF
zkO0F_b}oD>a=@n9Nnh?xMiSIQ`{yfOUiqO{1FV&X4gyu`WF(CZ;je}`QBg&VNPjQe
z9pRzr?X#rXS*X_9E4X?mTA)jSgeg3`PS3PBw|_?S_bBtpGpDfNdc&@-v$ptj{tMzR
zJyRB}vtK8^Btw1s9w_t2XCyRftu<b!$A~jp4Sdxd;SKysK9Hg#Sn9}?#rx90X?e7K
zKqmCc16_ibPr>~5_0m}KQunOW?$zaD6LDUSiKX@Dsa<*-kK0ZW5+TpeeQofvmC-?9
z4f+#th5j@kSj$BO))+9`C&ErptXif_n+&KasYAd!18T7lnIDwiV37f-YSl7vYJ3nH
z1V4Z-fgzy+B2%tSG4w;+*^2U0&6M*%grcDVVw9lF3RshFcNaDwxC_K{2*Dq#4(Gb{
zz(FQn=<!AgG6<3Jfh~Q%o8JFW>%ZtRTWJ7H<O0CEo(B{sIw{pK0elqbh$RL*5F$hr
zP=k>Tu(JUN4?w?%`EA9U0$>KnaRS{b(8gPW85|G~1t750p?3%`6iX8y9|Qi40s%~G
zjc#{MapqkQi!GZFzx&nDv|K;@L555f{78ilk6Qq=23g8`2sX!-IILXt7UVE-ARf!j
z@k>~%C63nocTDlFi?Bvk7C}ihH{=kjv9Dj5hxAqhNbz33=JEK={dB(_2m-8iZhaoS
zzc#uv3E%!;k6GO<RHkPY$`ze^S=&T!aQKtQ=M2R#KC@5?&E<G;C3225zhFO$-}kw~
z6gqqY1DwBpn=hWxO<(nAI&4F|t}Cg^siU*J9|9<-@mMH#=@~gDULT06WG72o8*_!8
z?sLcI=K)MnZ~&R9xjpc%`$^#&EHgsz%*dJV>>E25Il>Fa_=v{lZJ6Kjz_BVlu5&qU
z0d-AUniWlUfj-S?nbHH=GNBKMI&m?KjC?6ah8bG;TfO#Nne1CDLn2NH>_g>RYA+kI
z;S{H<R6}Kb&gEc0n5LIzF{4XL=sOENp=0k3j!_z69$y{1=%&u(awt-sP0LoQu^8e+
zc|JJ98-Ye_dN%Ch;SMv^J#_&cvh~S5{Hd}RRB@zhJ9>_myE)5#*Qb3lrMOL(4L{Qq
z4!la<)c7iVt_v}%JhUx-w`Fm0rqgh|)b%HOF<npF_z0ApTAqtbY|lqd^>5G~%wX<u
zfcBHwyX@jZGY&?lXAxQjMd|P%&lwF#){EiBHeG$c2OKkG@df65)gsRw0n|x^8#r?}
zoh3)Sy7OrMz_ph?K7Z4cEj*5Xjf{^|yy$rLH`gy}@wgQ)b*QUE!T*Mz7CF*p))Tfn
z_VoLF$$v}$!uIkkjC{_h?e!EkK2i2s$)TsKor1;9q&H<>AHJQVx!IUx@JI!IW=;=7
zW)ht{=016jWu&Qu8BMCya)#wjFO&0fgCV9HGQMf3&5WEO=FL2x)7I~>cg?4`xRY;7
zpl^fYs5_kDBNk(AtSk4AN{gaVD)oj=wjf6gvFv}AhKI%i;++H@;j!x6#G~cxSL>7a
zq2YA+e_ha4K*;ZZT{{sb8P2EV!y_unX8IX>@6x|*jk~_7v5=uepnDF>{YI7s=@5I9
z-uyus=+WsSFV{kx4+Rrr37<_px0^GaO{MF4{v${SX8Pwx(65xkGE%{y^D4d1T0I#0
zw83KDGokP*?bzbN?ZU!R;#?5h3}IM<>K*f+k<W2en<tTEW@efP75WYEvvg`kLljgL
zh~+pHN_u3**@U@oc;_-gyXbanSh+iYo8t{^ML{a&U>_F+u7!=5{5CT+U5D9WNxxhJ
zBsB8N1_lH&A)S%%qH0*Nu_r%dscA$u{Cz)NAFY#ol!cxmhhzz!5h)S3cGg&h=syJs
zbO(;24%*9HQ5M<|Kq2pjSV6@Kq$)Nxz*4oPVZ&5C2Pj}xj*e&|=m=7sV66ZPM@9*V
zV1v9D*iJnIEQ@t#vLQNB8)|G*7ng*8Vihn}#+qJKSQrA_1|V4uteU5%r*`{w$9}tf
zPV>ALI+j4G^ufb@66CpZANO<rS$PB>Pvig7+y-c;`1JI*85tQ~2W6!|(F-DGDF8%Q
zW7hY5dq;0?WqgE6=yrgVFU2O~Uz{A2Mdhc8H0v-0z`VTG`Oe6{=kUM);L~RUrCsUN
zfSOgjnFo6ze9Ls})l$4d(gQ2mv+^mC`9naw!`gDASh)>4OixR<IvFIg<KyG@zC(Hc
z{=FC6b04mwf@)8=$Zn!~+ptSQPF6wA&MO$)lGl`!Ofjwvq7OG8cXx|ZOT3v>yxI`u
z$tMlaT`p5E?>1F1KbIhn9T^@~1o@5O;t3wkqbB5@?cL$X_e8umL~mM;12#rje*DtI
zC&=t(e@Eh!Z+1!RXZWT`6vNQ!iBbj~@#Y~0C7w(c354Htl5X04S36Qk9tn(HH*Q=4
z4jTiIRaCupf38Sb&GA0=Sv~SK{{ET0c>F!{K`9^t9N!4u@Sg|@UP=WH4@_CjyDTe0
zWK4#30tC<U-%Qrk5Nc@jZ_`~|Id0VSwGK>QbTw}?T%SKZJO)}kVa4YYa_sDu$dq9A
zGbWdb6LihAd%f%OQP~<S7BP4Fn6qLCLby{}jI11~ManTyc?K3XHi_n%1UfaFlu1Nz
zDtzh&eXB12>*c<MbFmBi6%jj+U^JXr{l?h$5on2Jv!_3nsmlqwn0Z`J$heMv#f8e)
zsU_uAD_U3v?`QRfE9iZyZ`hoU_PghLgj4$Wylp*)FVK|yz<t{ecWhy9pVi{qz%0Pi
z)qjEyHdI~$hWtV3a_5QolG#1xP&7D#%nf_(sZk-nO%=z#YhwCD{Lx+0IzJs=_Y@Nq
zR4UV^ewjHS@>+l94iCVF5s(UQNKmnTP<q*0KSQB)=cJB0tA**Gkw#CkPs0;pOkkUq
zZj@GPra%>V?rD^#SzTv0#wgh<>3ttbb=xl1)|(v3<HPBgR|ktMk^a0>B3VAe8$Gr8
zy;;T5(hAW_pwh4|YC|<Rv2<*5aZU`V%03LB7;r1?9@Gzy#j9~>WSGLT1idlggXq}m
z_H>gyn;#g4Eo{}@i37=`X<5d8)gLxFdk5#ATWQm7$#8H?CJ;~s-a_7Ev3>Z~>jFGT
z^pVYn*-m1b&D3X1s>rJUzp@dzEl0m)*%uK;%{kZg+To3jDE7(AWN2Wpv4fC9UzYH*
z$D6jyt#i%xis~^;W#h~9@<Bsd_b=+O(;=K7eKcI$we}8<EWrS}#&Dch<>=ADi)NP>
zx{e~0GWy5}))^E%vJ`aBQ^xehT1ZYDk(}?L?KNg8Uwtf&ziJJ(ZwoKBKjXg;Nfi3)
z7SthfSnqyVIakfX)6?Vnnrakesp>GTX3E_wuRLzLspPMIhhMk6arN%ETzk4F4YJA!
zJ9V+Mm%upT)|=mzMn}=4ce}z;9vgGMwau^xo}be&O!sHorGJ}OtpBZEpw(2bi&{D*
zldmj{4qq_wMdQyX^SFDqE)0zA0UkVHo7S`IKrr)lVXD;F;2*#tq!xI-0br+C6%p*3
ziu<2|_LTQuU&rmq6CN;}0rqELkgrUw6rY4dQbPkIQNbn-Of8Y>x}N6T2rMmqjh8BI
z9ZcvcthUR(^vM?bdoHvIn&n_@1#I|0=G|{!&d3z_V43{Zk-@qNMBkuaE`ko;uB5Rs
z6%?Lan~35CuJ8+}ALm_Te^jxtv5Dm?8+&=-f85Ot%&DniG1u8CPBkRY9AV|;1V_1V
z$=@?&@{(xLFUTd}_6CNPZe$)!$s@^^4HVQbP|a(-K{+6JFVEDA+uic>alr23lE+^I
zBCmaRf4w^%_d7_$8xpcTxAU6S`!;H>-cNmc+G)APFj&Zx+9tLfN2iKV+ZE7*e7Jeb
zDyJjc$q)DNpkBcNv^#;8hgXO5weEyng;?}BFKfNO*pQP{AKRL$me3Ha#}Rt6vI@x6
zSDoKrDV=Ry=j3ZP`Xsg(f=y@JRsNgP{=#_IsaP(IXvr_wb=AC~CSZsyS(vQ<W}dl2
z`Sa@Emep;&)qB0_#nk_k1>h`=BR0Jl+_p1|f_7$HX{x6Vc6USYA{fu5avzPh*kqa$
zIA3(UihF$e-Qt}@WaVG|q<;LDb6MbT>zH)mM5|t{`9SaA7Y#M2-6rQ0&tJ3&zi4{?
z0<j4(DD)oT9fZ=do~0LgzW*n^PObgXq_t#*|Ef>PbxTG0;iC%40{#FpVz~YRPw7~~
zk5!-5M%iVU+rj?WweOnCu;qI&BhJn~>r!=$vgrAz*S79-4er`(ng;a1Nqh48`2#`m
z?O}t4<e^NSJbC+=)9HPttn*FSL(!%RW6RmAtpZU!y|+Cd;N!b`KR_bw>%MMII)Gra
zG26lS&E6sJi?Rm@TQo{9>(W<QDOt=38ccT$3;sH;j{ZG(!D_7Nk)8$uu7}dAcN`zg
zD)dITT6fb0zvqTGSa=a>7ECbX6g&Q7ez^IP6xB9ofFAq$HFqRqsB|%wy+K=WPy*r(
zUKe~6=AjmBtw~JkbiNW^pHnrGTj*w@<@fhLx&#Nd&U_Bo#pKa!Q)#O2SR0<zREkq3
z7XHYOY^D}d5vLx6NC+Um$4Y{SyOHJJ>gK0HBg${hPY{R~3LmiUG#;*fx_!|1?qT$B
zSo}yoB>r9n35ow^PmO(g7L)qa7?g}s-1giMmvG-()*)7mDDvBw2AC$<8odgs`b3Vi
zzxVrNK@9a!Xg7VuoI}ZVKeE>dZf!wOOcIrGV;-BXjy6gn9>o<2Z?FCAz(JLyk5aa=
z$2-y83|LZ0i3+PpGX+kTfJ;^)Tj7N~!=$PaNbm`0y~^en;OogLHOh0-p&FXAzd3&R
zXlMPuulrVG=FgRSqx%@IBNP!(-c$^f7h_LX$T2X=uu9xs)}l)m{*wJIN{!Tw)p;9W
zIPv9Ig}p1LkI&lpyCCKM{@>&%Xrg+ANh#$5okVe3zY^XW;fmfi)AotJOUu#x8`@*)
z{MP{qTA9Md@2QC6Ece1e2KqMEHo33bupuIY)GvK)Q18o+qj)d|yZ+}G;eL6wtc=xr
z-{WozlAo6F&_{G}Rqj@zdO*_l;AH!&a{B_#d8*}Gb8{MbROM5D>D${?u9G%Fa_6%m
z+HHACIFg8Xh7J6kxLz&AJyA&M+uw)V;?=ahy_SY4rapv0cetF@M@4moiK7j?{B3Y;
zEOU^B#Tdm>F%bPfn5NOJ)CQ<g%4FGhQ6MylB?BG3exGRZRs_9&+YNi0+^2;kaUgWS
z94Iw_pQ>|y9v|!$XDYOS<~~q{mZo3-a-ZJHi2m%_Q{uTaWg;Ze38poAo)1@fUA^me
zl)D)Y$EB*wxI$+iSsY*rnx6tsOJJiZ6-OkZf?ZNt8feo@h5-1Ywsv++8&P!JUMFB%
zY=(~#K6j+>ikX?&`EhLB8UJIdp%3E6LaBJe3-0A<{_qrCl;_7!Y*3VGXeRa~eh>aO
z>ZSZ>FXD*8$y$6Hm`$Th98p%=-)$0j^MW;Y1O_k1Pai6ZlEo(#ITy>M+<P;R0B4qX
zarI{0hS_y%g%l$rbG<e@bwcCV%-)`;Zb6(k+kCb4pLdXzm0$jAw;XSh7>32H<9zpk
znUo9!3bA<qI_^>x-2xli{AE$GX=8Aw_fY*s9Rl~N725QeBAAa(wy?mC#yoO>X%ACA
z-V&}FJ6nG;t?>$IR-_uA{nZ&1!t~FB$k?{wN<er3My9y($NATt;j8B&J_aX6evV@@
zo%H@vNo`?mj#O9|Om0cAn8L<V+i&sHV{&py_-k{N7yW4MsE20tV*C3y0{tZ*$@k}g
z{B5)`r=Iu4L<F~|<1-bgjlDI5MCq&>t8Hz+G{@4>F|JHG?M=W5Bd%5N7bq@ExWu5}
zj29t+PK_%L2ZzS6;s`SrZ)D$*32k<V)uoBM*Qaji2S&Z1i7%za^~nsWzq{!_6%|JJ
zTT+SYFhps$n>L=u(Rr8%Z!C>NZiC|r`$oUoT{FIV9iiP;i(EGq@uiWOg-yvq=W|oK
z_itRn;0ejF)eu9PaV~BYpZ6^-aO#@V>j>2Lt1dJ!2V(JEs`SiHQ#18YdC}fHJp9~@
zqS~4;d&-anYfDGV7;cF{I@%hZql;XD(O)wapL*5cW$CP)4!dpZY6G9#fHWv0Ki`DJ
z_ksmzW+H#udo7W~pph2Bh5W-_=dDJM*gh`|OK-xz`sFHqflXt|MtGX3hA;cOsLSn>
zPdZX=gfe-I$f()%OB!jP63UX3l+*<JJsoZAbRN$*QH_@gS+S}{Z*l51Au^UO#!+<w
zEg#f*Yi2amESXp<u*Nk!oQtfY^4II8wNhP+0<&O*bL(e!-aGcB=ibCvO@c2hlF9Bf
zgp5Fc`hQ(F0;Oa~S@^E5H_R889m6rSsVPM&H8-+Qdqf@0L_;T?B$-LN%a<Z8T(_<$
z@+l_o;HTKii86DwwrA(@FRm+mp%$07Y*y0`8X6*GCsEXTwhCa^mYJo#>pVZhLh98f
zDi?isKaVdjFAOOdt)Gq`UjKNPLH=~Fh0*ZPW%Ta5Wd6u2-Y-?*V}Xh$$A&?XE@4~h
z(6f)R>wY+I_^USeSlmao<(H+H5h}Do1QW^HK6mTtmj9}ghBt9U>of2izl>CJJ`VIt
z@q7wOuM2sFMglKUllK#r4AU2<mq0sX3cMCWQ|b-Si88pRrMcpRlK66DKzuV3Ap6be
zd~Q1^Ln#b}%GAc}1QYuo(nc)7KI&D-^K3;aev;Dv%%?{c7&GI4)usojAP@>5CMK3w
zP+;NYloS(#%cHD#W-hb7_~iRp=<l^<gMeXAt5k_{l#@X*RX2Ao-z85K>_^J6<2_nn
z;1wwkhQO6w_EV2f4#0InWI%o1bxh!znU62JfZC}{>X)8G7OI(bRTugR)art(VRSKe
zC66jQX48EE<AV1>67EvN5I+CY)Sqtiaks8Zt^6rGZnrz^;%$p>4Goxa5U=$fyhy4<
zza1vQv5h7{(+Pw~aqQGZ5eNZA<`5eX+tA&JqH8)Npe=Va&O^NHQ<r6aQ+=%bhd5$A
zLLiZ@<NrHNQkzZ>)m{TxsZDbuAqfc)5uC#6k@&YLuk1CE2h<5f(7$~7VhqmZ#oFe8
z2x-6c4KbTub;I4Q=g$y+W*X3YCM^wFNIi##mzTtv!71)H4N70&ffpGA@w`4!>xAW>
z9ut$lqR0@i%@LzsV#Aw8vZJM@f?EQLdf?9(YD%IPf?BtrXrRHNqhcz`6+N^~miVc*
zp;xM%6cQ_J!`y!z;LIe@@_?O8E*TOuFn|cAO1QyC7`-kLTrAQn74G~C1yb<2_tTW6
z$(lWB@p3>^4T4S};;DL6N4`px*Bol784W$90F}}kR#whFed=%_bn^htZmp-W)SD}=
z$4C_xsg=!lSflUo&cN=OUyzA2BDrdNC_N+FY%%UADK+IYCmzO2VvoA8^_sV~hX<c5
zr_AeY^cnGZbk+x`atmr&!^3}AcCNd~Qv~d|y)384{gp}359xvrevW&=%bLz^Sr=5f
zAx>X}AXPwrtlzu0hrlp`dgDjYy@^<xjtvIwojwM+wRkN#=J<P%&aHY8>v1}D*s!_W
zeZbCK9t;fFb*yUL;mycI>2G4M`0+C>J(CJEqes<H(vL-)635b9GQYk8xrU5oJ}@pS
z!`5}3J@SV9ttYA)qChgq>tYIxBJrT6FlM!LnZbKU6jRJL7NwfXm@w<`o-miCIJD8L
zSY}?9u;ulFQ3ug}+yP`xN-C<aTrgu(MoAn0*R^CANx0T!u%d6}anM2A$ID%%I+t@w
z$JRUBW)i9&MqHcK`TzSm^&Ra-PHgt7jB6FtI0+Rk-_ntV8dTnB^FB~gEZlsz{r+ol
z{r@BDEu*shqjo`+4iV`tX&$;;knS$&29fSg>FySg?o?VDq)Qs4J0Cjd_dhdd*1U80
zBIsJe2kw3EePz7lzb^)=eQ7II1f^f90vM`)7}6O%nsssTqo}G@3*;-iFn>p`41rS@
z4bN|Rm-|ka$%QXdp0OP-y37n~?SUlZ+SR3*%G&xL^UKff5mbSnit^cl*J=I>s4utg
z*m6w9CoX5ld-h6nG;8ZF9%}<%t=((>7tQ8(FB8E0GKBTq4`0<fHnO@4o3DJ8(f;xd
z`}sC@)(7V;(~t7%0xe82w_EYq`-S7O__k9a>l}}7UT!PW<GVt=shyCCX&4acXR};l
z_$%&6hM9#6^HT%t-m)vUrk(@cN?~V^#u2Aw!qHbyTrjb=rbvcPMi9-$!t&9}i{A+f
z-E-lKlK22}^1Ov&A;Y_>MsMB(k{LE?ox2NwjRfSSdA8>O)w6_z1n9wl4H9u|UcS51
zcX$5gclYCW@rmv4oxkUiUj2T@?ZRA7c-z;G&vnpx@3Xzou!zZ*iz|1|KI3@Mi6BRf
z$&`xHx)P&M+-iwG_;TS{i#5o0@4wsWbFxMa_A7($SmI0<|C|zUi70T&ll5Aww+3Yq
z78;C*U5g_bTy<sC!y|K_A00duNGu+z7_6+V`VOx!B7<xAS1NRy3A7Vf@#J*fR^$b)
zp`mFR8I<W!%J0uu`2|OlHT9*}n#n=_pL#O(r9#Xm^DJ1~kVZr2YpgKywa5q}mEy$E
zojY}Om6a1sGv_Mv3L3*-9=o2}HWv@~j*hlyZl2?|hxxRA`ogf+I|)4XlE@7IY2LhE
zqQx<jfe{%gf{lz1<HJVQC{Grnm*-0??C8j;I#GYT!tD$bVg5ed8RX0rsgg8-<f*bi
zK~;!7GQ!Bhi9tc`buI<YWvFSQ5j_I~3qiDdyDwY5C@qKL7gb{Dhal(lX@~@kw5<5@
zHL|WPu6cd9F4I89EJJ!Gin5;GaFc8k0NJGVd_Fwn`Xn?fVP#Ga+wjB&hqRo$-HP-e
za%Gk%h{uSHWNdY#ka<&@g<m|-zc+!}?(@XKsiCT@C#jzl7L1FJN7QRrJLOfC3yg9&
zACzrd&f)8t1?uvPdpD^*#{oi@EYrcLQ42|YbAFVMe)thKs1PzX9`Kz;IW9dTjgm<J
z2koxI67ek&Xnf)mpfcO4r$fHu2=NYhqEQ5mP79zP>LJ%FBftq}{wr5i%f4QgfJ-3A
zOT+5wotc3J>z+_`=x8RlGCGP@g)0#n;jBwz5`xd=m7MZMnQ33fmD<m0BhQ*p>=gzF
ztxI=7mtIV|GV6nIaZx$N@d+ipKl58enO_cvES&b>h^sEFFR2Ny>#PODX(MosJduif
zB<d`wXm)C}D&2>z(=&;BO)zS3p<0j0a$+h+D-RzM7B~VS=a<v5syZ4n0}m%Z?5qP@
ziX5@BbEG&Vl>ayxo)&0D31#WWV}IdN(btZl)GD>p4H!1ct7^)aGCcD}Rsn+F|M9U|
z$*NKD0VgDnoU^czaxjHRBk#M=&B3d*zBfNt^9>t2#(LAEbZE{$kSZ5WND7xRv9Z0`
z-nI?`!AfaaqMok8DJ!Xyt0%@U&zdrfOvzFs=IZA^Xsz)Wvo$acWqo>(LsGrc%z)xw
z)r3Lgcc}0W-3A>a`RM!kGe_K5u)ZkM93O|+`C@zv%_~LvTNM%!VZhhau7_>!<~u$<
zmwN#nGP4$~EJukUQhxZs$z8tram8Zs=)+_~vSY#fZ2fAqx`mlYAr3M5(_9sktj)3|
zm^P&=Ci>P#Ts!HV3X|gqK4c%0OCg7M?TX7mj1zLQk|^24E6z-Ic_(fxGDXsLKhq0?
zwS;nM@K7<3LwFpw5P_<ZCU-n>czfdTYTiI&aKT1Hz6}dGjnNzx9tO;W`h}=IW04dy
zK;{EB6*U%h6ME;%%UJZ=>D^fEeI3U09_ay+Vno~J$OuC<%D+o3=>aFKH8L}52DX>K
zFJBWvD3|D)E(UMM4cs29Kl}uE6(~cW`TI8~cauAWNM0%HT|t}3c>Rks@F(+Nh>R4c
zXTMGSvk(sjF}FM9w(WP+Or16i5<&6t@vKL#4B0DR(xbQ2ax=GkWkW6PstTH_ND?~+
zXL`}|Y3q%z2u3tehCeUeP2QZCU$waWk|~$oVc(15Gp^;)$Rvghlk85?`0G$7+<lhR
zEziByJ+cMMw=$(UD*h}8_Zq84d*t#R+H`Y85NEk2I~QB`fbg|Lg*lAL<|dJ#ciM}~
zT&TyEg_l+o<?WCnDR39X<5DRFrw((8QY3Fo8VH2<92b_gWT?_8Vnl~%|1{@R_?a1_
zp2@q!?*Dx-Dhf??ZC<XlGuYWNFaiV9T!ealUWADwaP8~%(FC5jIC(Z>O-p;sc`bR-
z4vziYO#B?b72_2fAJ?Bn^>W%RWP!)Qp38TrN`u>X`%gD@R-5oy3N*38wV~mY%0wp^
zqV<?*5#VPc%pMwXHz4h6k$_!6ZB&2lIetCaVFM>pjn9i|WWI-ogQ-<P?Hnj)AxTIQ
znq2L(Y0zvlw~cqnXI);-nl=hD!#Bb?1{8CQjU6+t7-!Wq)!0chJ~Y&P^f4ifw~sQc
z35FRLlm`mrpR3ol9e0&4-)r+?ZaQ{dr@|363ef7iq#9R)2aNLl4C5>*dLE6_0CSRS
z_<KD2v^{9{kQ+UQx}+2`aC#NTaX`z!yep^3EEEe<ydr1G`lrKZd>Vw_%=A*~F=*I7
zw5|;qo?6KaGsIyvZXL`t!>+6l4w`TF8hc0-@9?_bqR;7i5sUcw=XZTP&}II5JRJ8!
zU~-S5F29Dw7f16erfYGw;qTBmT0?dMhXc5sI&EY}U_l*bCh_pkT=gMDsaM6KL`Q~e
zIXJ#d(PR}6=xcSYGA&^ql;e?8*Auk$!$){`WHI^&4Xxcdq4M+3n*93gaHXi<Wutg~
zv<meB4IBSrmh6mu%W?-ue1gkCNiAp+`MDePe{47}c>HekGqa`P#g-r<#}mb2SHZGG
zQ&rCZ0MnBN)t3}bt+QKy#TV)ka(oF$eAFtxSk`yF^D7~!h(WWo^fwzeoTaB8l@~)s
z%ZM-^VNR8iD^^EJmMYo!QhM~EjiPVL%4M7{`#y`d?47jAQ{sYmJQ}f&Pg&cr`D~Ee
zk<uZT&^_(4&rh9U2XhG@+aCB|*E%~8m!x*{jB8g!Sa4DOg1(zkeYhRK%5fL{uH;|&
z+0Vn9?3Ghp<YWTp46|&WDwazkTBzq)_nXH};Wst=MJ9&cG)bpQ18q2p+}tl#z@vzd
zBjLF(I6WLb-G13|hDUQF2=8Cnke)K8&;sJxbyr0Tp+A{NIi#<-OgIR@^0Md52jZYc
zM@O-c;V7j=tLIF(GHVxXUPVPmD=RBIbYQI~hz*;v?G7`^B#6~&Q6|ewE=(f%FI)`U
zO^7P&oC?qM^t>DxyIv;OIj<Vxr>-~(ezwE9>!-;z{MS2^$paJOC3z|y`szLs1=6ms
z=;Fc&st&avu0ro`TG$zo^Jy_6`)YJ5fr??lrbvmlt;&Y|z=a1woxi^<KA>waC{6ve
ztPCp>szB40NBNlRuYf*ri0Xc%uzY21o>*|S0;N@d{f;RWGjb)M$LK?vqLZs3*3aS5
zZ+~@+w{3kd2URKcB;I_{fX=K29WB&{X2h(RCc+?Mxp%vUbv`e3iVyYIBC`elO!)D}
z_0#*zK2e_dq6&eqv~j2Lr-vadHrOCyQx)k;K4R*gz7AVh=Qp%i9@cA6$mgag;g@N}
zvwxr7IqZ*R<GtlWrN#vE05O5F>G9(Mo1YF<ow=s;@;n4SykXalr0gGo!t`l>96g3c
zsu08brS(Q><NW+t`0>>yCZ-hJnY%1F9D!ec!8yX4y6>9{L#-Ag=pSI^WgBf~PrN-w
zBC<sQmN+T;?k*ES8||g&pS`iGM`V2L;tu2yv)jhvpNM71KY66*dY$_D2XwvXsxS}-
zS8t9Q&fTOWkbeqAW+`sE=pg?3FZZ?nX$ofp8+*hIl14|)h&->WAJGhJ9+_)#4chFR
z0~)_*IBxVJaoMbl;l6B)YoU&^cK#at5GxQkd`yiUtE__2d+5C3=xOg6)MmG9OYxx?
zh*En#>&H$S=ACXSH`!|Soz3%n(R3q6^rS8}M}mI@u+Uhn@0m{vpHA~5vF62cy<)wm
z8LmY8?+f+vyQ$R9?t>}?Ai#Zz1?_S0lKnMla=Ly`oIfyFHzv{VY+_6)0%l7J;>Jx_
zbhQ4i#w;E88M`T~>P6Dx@c2IR9CP}1heHx5Rn&C?ohqZ#)0K5Hbd<5)2W3(HYlGEh
za@^WujWffMkie6P(7NQ~TU?wxe9JM*oj1%(v2Dmou0n9OQJ*P=^V<rS1&iupfn%x3
zb|V1c20eCHdsRi@;ew#|mWX@RnW`rd9|!aO^Z3gPBMEr%xE~BzX#MQSlx?!5CeZ$2
zb;l<mAz9O!RZ(5KduNgJop$>dXx)^~66&oOis(6%td>n2jnrutfBQc_q75)blwed1
zZo5Ma44N@g7**#Rv<+P~jvv5l4x8yV(@Z?zvt0;Ht{M=|C97Y0cAaWsnqFN23-K=p
zLZ5V<4J#Oi7OW#%srF@DNTM>87FRiPZE7n?wt;BEOkHGpR7)^f$$)eM{U15%D;Fkb
zLcZj{+$U6hPic7vb&aygdZ%4AhT^&M(Oz)jEq?J}YYdOZr8T2PJeWwI6iDxZLAS|o
zTkiI`p~s3m;Pm12Y;RGUS{JEl6yd70UA0!Po=l)c3HgU0>>o<{&FeHouXSuK_6hrv
z>&@twh1Ir{WA0bRQJ`qZb3ALy)Q-7*Txev-f)svJN9a6Tu|~sh-D;Hy45ZJ$ZMeYj
z0lK7Yh0s3ZE<={@bBA;c3<br-p`v;IOUM1zjS0isEN#p6aw%r8A|DfeKnEaHO%)o4
zcfRJ;;we~TB;mO)t*eIr#yhX%Wvu03%ipmG<9kd$Xez}iPdJ^sRgdo$)Ym79QFNW1
zv8>zQzBR)w$%o{QgKt|}QUaJcaJ7lYV*6LGbcAHi&7QB?oo<%`2*{4SU1J32c9XxV
z|D@*p%8P_Lm7(oiq;vRxNrH-VeIrPHPKAMk!aLjl*xH&t5N{M`V+eK9GAhGSn8YB9
z#`hdg^Juke<_xis@k?67Hpunu?6G)$RCRW)`nNuShOUTK54~WpUyvi>bC0|KL&o^5
z;;a%Jeus2H%}(<3?Wfw`6BFos?ddyXoJ4->7T&vSoqd0aQ^|+L|NN2hGi&7&`CY=T
zq^~?SGJ>^jX2$V%cHNn0q1gms1-!sdecnyfwh@N*^Z~80R34&jYGl6F_qK|t=DRxs
z<PxMewBjK&IrBXu%Wt?>oY|w_j_zTF7<ym_4k-wwnt5K*VTpfl%g?#nS4OUjd9I~1
zhC(5B)f?#px!s0!8+4WV`DGbemhk0<hF4|k^Ofc~r})K1V*OmcSWdg<#=Pl0f}eQW
zUu&e!PPDtCxOJTR5D7UFWjCXik9Ok<-J~Ynulz`4wV@{6gc@b=YrIDu#ZVZ`BE;{L
zc|)~rW@Gavh=P@mPl1wSplqA?arefz_feFf9rm>SoXt2tB=_Y}8OZHf2`*t701Nrx
z2|so;aAMC^A7|L`Lg{zebu($i!7kkTkutror3K>>zKnrBEjPPwv!T|(hj6G5Ujlxt
z&H?OjAZ&frjg%cqu5hL3a0sl>Nmt(MP=k%%9W<aAI=CkiF(O!N5UhLD-i=q@S6>Wv
z?Sx+N?*~4?dJ6#!Wn?YtLyliS6*j}*)|42oMQP>N-}}yI-IJRYT9c!<w}=V(=%`7o
z{0tLwUIbedw$(wp)R9c-=APbZF-<sJE3OFKYtQ<tF2l#SGH>?c_i|%us@lI<Lc%N4
zC{wY*=$X3LY4_ix$-QA%TBNJ~`+i)+Lmm6P&uS{2W2iw+PY)Z%>03H}`otw<&PKau
zvWrDgz$xoEQ)9x*HLO`HK+ZX%@ylFI&99GZ&HCOXxVvLnMcu~KTH5mJdc-oyzb$J&
z$V@HWuFw4s#Cc-k_9?#z9v%S;OT3-NlFQ)tQdDfufH0b(VBumVmB`L{8^s{V#hRs8
zVc>G|tj*9B8NWV8<|8JHKpcYHwPnglU>k;XGtku)ex8iTmtj{{^9na(vnG>Mp6R5)
zC_^Z(3cNa;c{Z&!Z?b00`TX*#TmeBcwk+Sq<|w(yLW2P|2&+|d($1i#pHm`h%%Ghb
zEp##SsH#=&aUf6OdB>XNP@k^WopnTmm|PO3%M1=STcJHoHeqC2g%k?4-XF-c)4193
zNc<WVMRL%>t*WgmZT6Mer5MfeY!I)O+wIbhNIs!`bnogO>NtFN(W;d|Nh#fZ?3yT3
zXwR1oEUF+2QHnmf+mx;J>La!|W2+?>X0!xwk|auxdNL(D%p6`}A=^Pb{~TId>37ED
z<y5n96OebL-Ps8$Dq{`pLe(@hFc4w)j*rb89TRk!ZoZYWfxkCpo3G#k^$wuAgfdb3
z;_eY{0EI6EBQxjJ*O0;Q`Zueuz2U1n#vk2}`ZQ**nP<0O!%G&gO?f<5DMN60m*$2Q
z{q9wD7E@W-=3|#M16K$sF=Nr~(`z(f)3b8=jedd`Jn_s7XPwi48PNg0NcHr(3myyB
zh8PB`>*=0Ye2Phf&TY5BBOLdtCyG}Num9+NVFkxWd3CsR_rO5ZE`QRW#ld;!_(@yR
zHvIws+Jup)PtJ$1adJijCQum*>$eS?vbwT}hgF=g_Y3Jo`uHPn?)?_<k}U^|qNu5>
zgA@TeD6dj4D)Pj{{NULc?YO`d57f!2cf;C1=uTaD2_5S-?5~%BoBG2DhO{)I?TcZz
z`rTRxyXvNx30<#fjHT&&k@(v5i1{r_1v~%1`ruPv3)1mppRD3sj|78Scj*7t0?6^n
ze(xQZJoOZIX;jC9;B3SrN_0gIYsQqYD{o%JjrUIMA2~A_3~fAOIYCiw9}6qiY67~}
zS(%!bgJ+S|w+o}>75>?Kq=j2ZoAl*KI~p5CWc2fK+0qu)mnKXXD>x3c+EL-;d>Sd5
z@33NAI3x0V%yGL|9?v!2?AJhr7Zh*r&UWia%@w?0Z?at)+uH&u0H|ctzYo~J?+K5a
zdDP8wmmca>HWm4zc2=&Liv=eDa-5w5F#NEGYyyUw(bs=SiLk~+Se$w4fdbQ*HZ4O}
zzTh-zmSlPG;Q9}hYU?r~XzIDV?Fo9~P)lYbjQIC1rG7u*m`KPo=|06Q%rXAF)B`ps
zK##v^ik(I7v~*h6XBPX{G5mi#;@01#wa``5qT1T{t}Xzdm7&9Os(;?OF>>;_9eR3a
zKa}$J=$A9ikbCfYEja2CQf+i3&BWpIe;Be7;$?*W<1(eAg$JC#st>|^gVhI%SRzmy
z3>{n*@{WN-n=xT55Y>&zo2dIgi$#g%eGDIvkiBBNA4B_hy<aQ){%?iN?-LY5C~Qen
zM8#rz{CPN%4LrkF#>)8-_zDG@qN!JK(x%3a5mQr+2j%qRe!~BM;@$Gm+<oQQ1PVPX
zWbS%?Dl64tvqB5apkQDOY2n*Dpjc>3SDhr_<A-Zo)jVMw4$SFmNR|?lkKcAYP)oRs
zAp4xJr6#VO_dSLkZ-#Zh;UI0qyrjC*UbyZ8IUs(4bLK0N1TI>6BxKea3s(KaqaD$u
zk)y)ixWk_e#>G+=W!X>I-~Z0|sYgGk>y@ezLw;*^*-ju|{A`&%L#(8ad%N|gCaJ9r
zx;jmArl`rKxpJYgw=7LHTVzr0)~ySx?IA5|juEK;EjC8dmW+i(03$Z9$8Xmk@9l*J
z)q)L;jq9$htOU^@vXm@>!uBU2kWx}70;bzkTf&pK)UbX5gzRZm{^g!UsAwsKqslC-
z>;5y~D*RY<!-QC4WyU|ebZ~b=D(o=nFl3ql^g<d{y3(L;2x!9bq9p(u0IY1Cl$`p2
zvI2bACH&J}$JaaW=RhkqxZACpcgE}o&1PTUu6d73$gF5|bQFiSIiGXtkN&=eBo_+N
z2X(xhCqPmG8^swud{|h0ii(0#zidA}s-~tvj#oJt0m5uQG4DzB7wkEU+i<a~PP3!l
zT(R(R1rC_4|5-`>Amwn7nJHy~0w$otsLPzn!OgVNT5s<R2e#SkqTNq>ZzU?w-i50X
zm6jO(%yd|Ej(VGIVg?G<*>9H>U_}(e#}(PK#gt7~JvIU9BGeuEK88dvV%D4{b68dH
zSk!{W@p_r}Z-w?JpHDCz(&ny0BR;~u{s$R+8r3?{oxQ_O4>^(vR21<n`LyrP5I!nP
z6=7(7#aYzR!6PItXyX{%PBrM-ngLLnyjTzU;%C4yhVSoNMes2+iv<QL3J(}t#;7JD
z44PgvG&I$F?H1ZBG^B@U5sBFC*sNJ8<iwYLH5iB-_$J*d=j`@30Wmb3X$(E4x1^#X
ze^>5Z?){9#g4-S<R9WymDrz|WW3QZqecXr`dIWOhr(Lb~dkpFV-XlDKo&jCO`O~Ri
z>@lGR1>PlUYz)~hl;fw{RD%(mP->7C{U=ouOq6&Yepj2~{IVDW;2xwMh6Ll_<MBvy
zB^J~+Ca1=b6jSV(?JeVpR44aICCQ0UV#G|WmzxR^bz3?p$6d|LYOrzejj=!In=VnM
z#>I>nisd(zMfHj0#81BGvRe%svZPEL`{&%ezfan4)8wx(5P{9W^UIEz$pAfcHh96B
zB3aIu5n^9ejnFzB-eqa(>G|!Q-`=OFaR_=iL`7L&2HcsU`aHjvA1Ovh8Rv=7Ag(mn
zjiA+vxFC_8V@92on_p7Xf?t73+!@rqE`<oA|2DmTubq~kHx)T+-QVoPfYur$%k`Bk
zbsQk*9{acW-BjDZh|KT5eVd>I*{IxgX_EBj(gPc)<3+TzXmS!X9zj8J6oF(#XZJFD
z{L?`dv}%7jRP<G4jg4YB%jZ~x*@ynw1e+1dw_Q%X2Wk56QLwd`7l+jT2LKJe*yR2O
z>0uxL9ROopQLd$SNUT$74XraoFDuWhj24|EkSL^ITmr*oPEK`(mGCbnT6x`<(vf^f
z=Tz*`EN#U7V*-m)^S_JX^>2sC26QgBI)S6M$fe0zV_Dytc^mOF$x0G4<o-d)?`2qZ
z*5)C1Pj>4%c}1%rVPlW?Yf)L7uKWrk)k%W%+1-qEkCA%PK64F%L%=F7n&U+q7Up=T
zsk6pZ^S(-prt|d5iuATP<(eJLf7=Rpx+^po02c&qT=DnXDd0H=!8?<dy{2s0r?S-2
zQ+JL*`_2TQyqPuvxJLkd9Fec2uRr$h9~mfMjhq#whW3}%#OJVkKVS2QF-2h@6|T;#
z%1gkI{&L`6nsYlFB^CDVVvWep!gXJ_Ph+<j1BaM4XcyGe(-S!R4FiJ1<D@9{HFvh5
z9-wt>@8qQ4y#9PEQ4zRG#$~%SRpTPPw{mvVV=8x4Tlw!Y2{nwl@o%Sk2I9B4PeQl^
zfBJt%R7>sbvx`|1haD`M4#pJ)qzBWA1HNr>kT;tIee#dej&D-#@e`uK1HQGPH!BW=
z2^HTFCD2Hg2hPsOy8L6a`CTy}41LD637BTRqA9XW3xR3F4z~7jpwSb8XZ6);^XNCr
zW}>PZtiTpBhEQ(qw?%898gf4`>`A(_*|TlMR$3#aMQ&8*Y`clMv9H@jt<{PU6&f#G
zY6{9Hlck6tLq|<v<Kqh$wKjLOL2vE(^iap-`Eb9l`!`_ibTi{iRZ&Y*N@=(QiQfj2
z&gTqmfJzAe+uV&6`Zm1%JrFrUJiY9IAPNToJ~7153g5OBKS)1zdSA3*vG%pu>vwL;
zI<b_MJM-a~AgdsGwtUX<<Op9tNcc`IDW-@3XA}RTRJ^RrJ8gx8EinH3nj|E-S5rk)
zImPm8RWWV&9z_2_oUl-c!L=AAR@U734E62RES8Iec`sJMhaRH+^dE(#K@qdFd$59B
z<@clsKL*3_8EX4@Wk!@V)FmfP3pz$bB`_R3WO%rF6yoVp7(b5%?B3Laq@1aA$WyzZ
zuHTh<T6^+f2;R!qZ8i0jytzXj=UcpDg&ay|y7{jf66Uyg1o(w{d0F4SNyHEd56RMu
z*R_#<H2KO)U7s$>G)DEAIj`5~$LX>*sszl+N;!|m24%oUq|MDuAVR&w;g44Xv&X^P
zS2$W;K8qx`$GA~hQ&X~*Rv82Ii_e4qJFlrF&|5^bnmb;OdcNZc1LK^w4)Q*7L{)XY
zxV9+6UO<5&yUUu@VZ~)G4%bEsdGC_1><<GuZk{DO&{>ULX6H9o60_!t7h_KBM^UmP
zL2!s|b4f+U47m>{_DAQB%pvf^`<952uU3v5A!3O<+31S>zmLeZw0q`lG71nBF(ItA
zHrlxRVp&By)TXRKYr8vlFbKaqcdb>(nG-QFIexZ_#OVE>^W+m75-4Z3ZlaWK{AgaU
zBQ$ZU38!sA^QXlH!b3>wc1TBeH*1v?0>|dSJXvzyM(O2J+BdhlfdlJOYOMTL-{+_R
zSyTV6d>flNcqspq=6ml#()7Ok6N40Wx^36Qv)Og5k+c`Ds;TOu_Hx;P5@LFKwVu?W
z5*G?py2jkk9Ay!8DAIM!LCa9m8+OKmYpO2?=G%XgovZhX>brX2-Z*z+{zq{4gE^rH
zwj^G_0o6f1L{LZ+ypQYy?SKEFzZ?YMIssTPm>LT<*KJy$OpJ#J4P=oND`e+aRLIF&
z(CIexrixgI(U)Dj`Z^}u4GnR(p8}jC_n{z@l|c70y2)X0uT$5#;>>ZTPKg${gm}EJ
zA#ST4H2T%4U0os<cLE8N*^0e-h|&oxnWI3w0Yn)R1?P`#xD+&hyiWpp0PFZT&G_Z+
zNPYM{`t5_HV%(ojUxV-a&4_g>rkj6Ueb1L!yK8$he^GR1?zwncTbGrTRhb_zaxPzv
zWg?A_8-47y^q1lYpp^-$lLMOWT{-5J@;P(}gJg=BP9~vtqbo5^d9*Gc@WhERVX;uC
zCYgCty=E&`Q2neL(q&0O6a$1Zs@f!!x#P;TgWuW7zT1Nh2D%YnFcS3^Q_(ygJH5{y
zG4uNXmh9O(gD3iPEsUVaL9->UjVljnN)>^Z$JdVeBX!#hv{;l9-*(;4)<aqGqIA7_
zTJ8CZJ$zrFSdMhAdan2kt*r~~PW+$_`@hD&vki*EZS13Y7i{e0Sq_c`mOWUUon5-}
zk{@JrrimUYZViSLnRHU(#_{46uKa{twqn)XMjYvuwqN_cPm`N(YB{sXV+-x<UL^gh
zOiR)>bxwk}F{OTx&L$K$SokwPgIY+1f9Xh4-vd@WEORm?<30=S_t7CNROZ@M!CJZ;
z0rkNR`8_)<)2k{WUEir?iP88OQkZEHq~gOQYcrjb;GVJsmSR^h$+Ab+S<>$gg#98J
zmzd#RCxo*Q&Q)*UPWH$}#m1##C6r<%s*>O59+C@qWi<UYf7I0L8ysXAKV6Gmu>OD!
zDWv$f*_SA&HaEl0O5AV7?%1>oYffw`^A%5?dc>T4+f1m4=9`pJLCGFn7zT5@b6Bp>
zB|2&IVGpf%jW(m2re+>D-r2*G+UyJlS_o}rNzvB@$GUt`4vx^d!!9FMOxki|dwaZ}
z(;<_Frs|qB#!S&m`>@(#DU?}E%y<=QYQI1xdGEF>y1XpW;i1J{#ucF4mPwc+G3#ei
zJi^Q=GcrGW>`l^?(n0HtHmT+KLiR)h$F9+21UxC;hwrrLOk@rgny7Jel1IWx>{`%P
zHYp8KB2lsTZqk8Y#KG-utq_2~Q6;%Eis9_ULAA>m7~;g<hxSNNkZhm4Jd#7%*>!#I
zu)iNYfAypKpGsp(%g5(D+Ox#mlBdBG6;p|V3<{{HI@0WFOq~ZxP`_($TR6@Q+=(h~
z<Q6tN%6sDQV1g<gnFK~gs5PvJizE}St|y)Mh9VXL;aV0M3@3!z(`RBHu~@PEA$Gt(
zTd8`$z!r9xFqQ9-*~smC+K9DqledE4Ih8g}Y<Vn-G@Gpd$=A91FLyE;Om@_U52Low
zPLvfpFPKid!8*BGEntTSk1kNaoSy@wJA1S4i8m5>1OEq)SaKM15U}E*0!<$H7odw_
z!2|FKgo#N-MWq`kcmaJ4Fx?f~&VP%fy3o0_%bmGod0bI+SfQt`O!PhptZf|Ud8J@(
z1xv={L|rYde<!V{JE6jl<mUD1CnvTC2M0{_(4<ta>SI9R<m8Z;YJ-g<8#fBq;9bYh
z&fHG#Ggw~boqA2T^Y&JzLh{@Dh0*v@5B)v--#>7FU1faTbM9PcPG2K~F9<Nyrpj!x
zwSZEkIy~MwATLnDu_`KHA*E<2t6gfzdv>El{(x!&Z`j>_?5P>_Z{@Q0;W6x!ud~ei
z&=qkOJTay}Sv(Gb7@ef`qY_>Nj&B0k&`5Ybvf%Y@*O+i2;<+H7i$%%3KFWF3g+Z!$
zve;+w^2iQ+{;(g5=Qg?bw6~q10XHw?uEo(!F=z_8`vryVi9G{~Y#f}5@+O3p*(<*U
zY}<rLQ*G9yRb?3xHw$!~M%Bsug|#GjFpA^Gq4ko!3TEoWz~BZ$B<L92)KBlt#E+AR
zmZ<zuDZ+i(RUwP#&-^?s6}Ea*HvUJP2stK7uqCq+KoEZ$TapwEMTr#mY_b^5GHex6
zgnCJ8MfIj<W{~eSsFF4$Mlw$wac(wf(MeJ0ec>WAW8YFiV$lN05pU~7@(ReM#7jwo
z#>C9cvvj#Waby-~snH^7D63CP36QL>bzHgV3wS4GEL!Kdn=>~DO6mE$y`#94ADL{c
zh}_$F<H;&wrHx-}8Jv9l7IsBLqk-Fyx<^vz!su_Mfuf!ZoM%CN*`kbY${W)Ae+Z?e
zm1fq~1jRjl8MEbD^{&L6oGyduh<}Z7`nEQf7kn?$DSxhbqNJX7B#G)9oBM)XF-d`V
z)2AZ&5k?8EPIFEba!LB^?V?OOBXe7{)(-zqgOiI>-L;<(tZc1+Ti5?PU&JRRHL`K{
zfAG8YDYDp+GQTjruo}za>rNRxoIG?17KPr*GJLcnV-t*F!0kI>%B%JZ?#O#DLDPSp
zlBD$6Ad)iTR!<`eUhpn(@H~+w4%y>~nuR<OS6rh}ecj$G9*a0ID`<}bEi^lPU1YE?
zEX0Bpk?qboP|E#v;&gO%3yHX0eqm$y^}Qok#S!R~ftMbsKi#KE!{e8HX?CjE(;rP?
z%yd7?Luh`;Aw9b-h^;p}<RbAsl19rtFf(?&b{YO-{eR4kZ%SB=bXxR<*Vk`2#djP<
zJa{>{3o6!r<JTrnN@L<cLmq<Z&M7fBkqb6GF_2tjq&*9lH@-|B&LW3={d$&MP214a
z;UUu0tJV!SI;x~zm=eN4Fbj?Up?RP5)B>6H_deb+FnPf6z+`veuh)Kua?ggbs#BLN
zAI6wX%b?l4=z=R$``g_1<@e~3el;>1Yx8fdHE9JB-tPDhH2cI^)F}^FKVLFMWx)wU
z8xh!L_#2FkpG%+s5ZKK@at2>ET4Z0Kezl}B7FbMxD(m9H5o{s<%XI46n(Zy*r_W}_
zhQGMD2#N+l5?}-_?Kf+7zUG!~1+F>fGq>@Vq@480pAEa~L*`8T9U*@0r_%y1dy=&N
z&u^M;8{gPUU$=uay<aP~Ku1Q$sP6S{-<c061$?(9>hfu&5_-G4-%MIwwxKs@LBPQC
z-H$wevn8HhE$Ew3+lK|LKGapmRU36<pDpia-XZnX;|dgMK`u6Pk4?UMeUMl8ZDT@M
z*Se55$X%lXeK_oi1<s8ebv*phN?0YOv@}@qV5gYyJ3-~jOnxABk6Qw2$p0>qqgu%!
zQ>c58hQvT|4T0&-C#bv6>j*Z$kaO8A_D>A5BoGZ)H`?FsIb>uU+zt5M^Fg7iGjp?N
zldVE>`hf{`G7)z?3RJOUJLJi9aZxmY9d~xEqD&2UdYPxqJM~R<>p0Ytx)b=e&8X!p
zUZH^-{lzxl?v~QW;)EI<_3Z4^yNi>hQ$uslgioPpQ%(o42G5zD8w)-N?q^-OZ0v%O
zYX;rj-4r;P(QdkTXhP7C&cfy<P?*gf^kEZd@$nT;7bJx;`;FceNMXd+x9fUN^W*b+
zA(LwYAdxrRGASlSLf?BIo$>7~5SaY4f6%VCMPVl9mC#dv6+7MCGu{@quwU4aAIR*t
zI{>;^l$GZ`dWd6Z&bdOb?0j?g7tr~cJF})~6xf^=6z<1z3Y+rqtKsOI?6%JlNXS2<
z*DerUUR^n~>~b{Q#OSrUg3lmcH-d_#0}5Z|Ij^o}lXyn5@bN6h>k-^{)l(ghZmmG-
zv0wMPsvzV`iA3&jyqtxlrS-ow$n?VB7zm!$oVFs9kffOz?~qyoAWRaNrH(o_h`l!r
zI{IjY&shn|>1$d9mw#PlC#9*VB|n_s2@E33jf^_o+;j{Wyj-m*0nMSRqVRD1htMHv
z3h*Qk-H>_)_Q$xSIXcNPnrlZ%>`i_Cb-(Dw!pA0K_k#GD8|YRc31i{jk0zwMzq??7
z;7jPHulj72-!?Qq3IHU>j7IkVh@K+|#!zESnAsCFbsJnE1!2L;c9h9D3ojdV%lf2~
zL&_iqvyZ>Z9brkc;jEq}x9{L0j;Z?sd*ugC5n>Woyo?#kp!6SN>Wlg(=cWon)9<+%
z7mn7M{CM2RTTpz{>>%LMlC|+nlXW{0gCM)xfUJkyq7*ZL-n|%HpYZcl1u49|JdBNG
z3uWZvU$GtZr9hh9_`_u*0w{mr4q_N;*R7fLsNkeDJsN%zbwxs|A#8$8L06G-@+2y5
z)@&}(X^;LRm9c7Z^7O<SKW&2{D&^l^?X=TdTeY*wACcs=GYP$m3zNKi?tJtEZ&<8)
zs!(;=3_7_`o}W@ZsicqCrD_@Cm63>&n?I7{<nokgu|PM$|NX~C(A?zf(03MuO6hV<
z8)hv2!*JNx6H}r!Hst5h#F^Z~S+%L}V=)QRicH#4YFQEC>ly3R4z83V?I;l!1Q?j#
z*|sV7d+(48^Ou!ig+;Zns;irw_G}J)^C?xk_VrzhlS`?OZg3`ee&}3Z^{Owx%5|sk
zs(tlxPFmwg1=0F=xgB)L^Ou!3{Rt}?9u=OGI+37WJGRTP>w7Al7|$1&VZRGj->2VD
zBBC>BQ@J{!1M^Bh_MEO<<l&9C!?RRU+!0KXO6i0pNAUnv*{n7A02?a5poUWUY|iR2
zXJ@gE?_<7BB{dc@IMdQHGx=8rBL+5$mo)fRmS_@iK7X;y^iDSm`Nn>7<;hS#8K^<S
zwwnp_Z_MEN!#jaHOo7$BwTB#bP}};^Ukf4uR+U|-vaW8_j$7|u9Rx64KnC1x-Dq5^
zMa#q_QK-VoaXB`3wzvNJz9`81bW?U=SxZVU;HiQf|07#^k(+5?#)aqwH2YVCEdCMc
zZTsm}4>jT&U<Rd}R?<}#;e&gK&G}yO0~WwfoCAYV<C1=U%vEjPM=`nzSC{p+D+DLE
zld)301!|01e=@{SWrKcW4z#l9KoRfSf=xskbWo<**#{j7#)Cs~bNnzgW8R<k!d7$r
zoSK~MJK%KMDq&9}c3o7Ao@lUtJtlx_>qaMucDyTV@GHDHXHi5TksLGDXu|5%+exvE
zNT$UI%eeFd=!SiBNmUj3-c8<Y36I|)ocEoq0Tou7QU9UYP=1KBTNY@wg#Lzk_urLQ
zYt$nVAuE&S?i@G{h%dN4sx(FowBxm5nyQv4zsGy-U!WqHewj+E!Agyf5!dE{M`cQq
z2a7C4%0Ax$J(`w{Fic5Jd3<JzGRP;JV-A~*kG1EYZqmUwiH_$+J2EpB)aaO3ESRVt
zWbylDkO5t1=IEL5;QJ(dadB~buIENBjghKnYvyOcZ$k2pGg_XA$2l$3?}26N%i1Y}
zW>tQ-UVucw#{Czj0g5uMDP2YC4{@g>+o8Fw>HRw)u$yrOFrnIFs-ML>*i(6{aU;v)
zs#HHT{DRivrZ_(C`Px2g%-t<50ZQu<e4}Z7W{^AxTE2aqo8B~eA4!-O571TU{3%xy
zmg8XQDH{D8ugc1&;9T!=7;?1I8f(uZXXf%2NTa<kFmp{ktpg3#UJ&^WJdtJ!b)9XI
zxXMwIWegYDW26hnsR+=;-X81ufNoz5ZH&Fe`oy8s+|9p!LZUJST{&MQKNL#!i@F!J
zZ|&fPi)}Pi35K?N=^KsPy+qGuHl1IL71RA0-z^&VBcU^*$Wc&MQ4T7>R{jFJvr{y0
zf8r4p9UC7d$y?{d@6%&qQ({NGb(1Jv9+u&_f3V-LJ0)c#&iTAxWt?X<HH4Qgh7rWn
zTZo$@>!yg*kp^4$(aw#uTmygpd9O!JGdVsC_U<t{_j~aT7EI5kjaDmBV^y5N+nBn$
z{7NF?mQV)fg?{x^y@=71TO~h||IgF`!{;&AwAO#}_TENJc_c;~#3XyX&&%k$RM?(a
zc24m}m3k0GRaUT|kyGsSnAz}_y6&%vkVz&^O-qDv{mjMT#cYaC-Gf7ac!2(A*vJxb
zXQz&_Zb3UO&A-3j%|LnB^j`sK$m;6R+d(!tg>DOBl%LCQhA6}Bk)zpj_n&E7VGemZ
zP}){)NQyNq4E`)H+tiL)0^`BI2!^_w2aRVK;q61=g&<)VGGpiYn#_hcX$+*h^U!=$
zVz7L8Dx5Fa`GrK^ogTY*3e%A<AV893vR8Y6x){ihwIDd)i~~oTbOOk?1O=yHAH#`+
z1Fi&(ax)OX3QjVxLcJG1zpS-eKI9~=YfpwqyZ69QPCGo-vj6-sL(f?lvJJ7d+=kA1
z=UlI%*>-HB3=W$9xb#9EBqMYOTO4SKK6lSjq6I-h<mqxb4n*l>h(H6VQ2ZfT2C&Bt
z&AM3F*x(b8B68}j@UM3eHTA_5q1jvzvt#g7KtI9E_ijEC@_AAW8WD{bjRSiw9=Pfz
zG?$y9#4|D(!>Xgkl)_l`k;W7cAD?18Cuqs6kDI-x=_{^N!!aU!BHZ{T>dQH_wY0Q)
z{m(Fygk7D90vwA@C2Y)<apSgOf=Sj9shEi)8{dqip5BK>7Z(VGrM~u0SH~n`NOa_2
z?LtkF`H)lc29m_k5G!a2j$>#0qFF^bXDvRVZKczKaheqB_q=+e-Ra8+VtkC<Z2~pj
zPPg+TnZB2`C23K}OH04!#bF;<jB!}N)~vhv_nOERp$7&g2^&{AC0DVvDD(*djfInS
zsO#wg2{usoa#D7vMC@L&XrhE150&LGw)T5zdQOf$4x_b^yBIa5V%=e1acLQ?_{@F{
zuVec8fW4*$>#nT*(RM1`XuNXrU)Ie?48(7fgwaF-&cO*6Y+PCO>7;AE4A$77y==9s
zCswd~D2Bv**L#o+&xP8y^=j=Gw2I$~-!mzcq>VJ0%>L1ortS>=jWuv|)TWI(yEd_N
z@c8Rc)%-GhzpoYd<mKj6Rvnws64LhT^c7ceq&;odi_ov^3jq{xaTn+XdEZZSZ!rD&
zREfiqA!a%x!I7T-{d-tfQq(wqNb6t-#4>Ea1bFP~P50S~e3TuJc{$YP_Xsn3cLt3-
zWCcCtzo`jCYlJyI_Od?TJNe1?1X34KQ`^rw4obOm((wOpEdblPR>VgR1rICy$!0Fo
zW`bl0IU)g069n9nfm7bEgNgzc??bX|_@%70;l!Ev>5H*5m3u6C<n3Nw1tv)pW79EI
ze{ZeMUCn_BesU^Nqwjd0?Ht&Yt@Dcohx3RXsZR`c<!sggq3P8Hj@*A<AY>7^51)d9
zsov(S3gNUkbEqd94hDz)_z>nv%toJ5wH9I55ZTZruJ8L^qVgjpT*wJ^j?cA~r&sIz
zuX7>=7w1XWwn&g6*zqf)sR=0pS|QNsL_;`N^-s->hA$B&nhBw`75TF?r?GPs<8F4)
z!bvY_bVAS+rRw9F3-Dsww(J+ZdjMGN|K5W{&;2I7FOn53K9Lu&wuV-ehN~T_6a~M@
zYPKM$Mf8Vk=E~St=s9^Z&U=T;CakVfVg}EBVk>6<tmRb=`}!h6d`IPKi%A!8Wram3
z=j|_ZbHh0%tvX7te7A>Sb7-bpm>JvIveU%cI~<gZL%WaaD#4uaj0EYEwMWh}(r=Pq
z<>SD^!(N|`fAi-f)^!}h7p+r6h<&nt=bId9rs<b*$i?Koq+gfx6Resq8D0hg{q3EL
zJr~TC>pl8<MZu<|HI^JQJVgD*wSW!>#Z>k@VEd&4!(p{KS*8$OJRZpQK-|Ve31`{}
zHJCq>){T~pI#c_X`~F~ff!EoFc1L}+^|MzLeYlio!oL04S5jtX`fb%zpNzjPIEyVY
z7sCU-L(iBG@w`3V-8YucI*CJDncdw#RY0qtGau-{ZES4ZyYd9&8t9!85O}N-t@8+1
zgCFOIHGLH*BVOA2ZKJNQWQM@oNxuzKb&tbhHS@Z7#FQg%PFf6xB*%M;<uh0Az`sE}
z{8B_U0-~xn2ZYnh4~GV<hp`T`G(rP8sZG;Lb72xDz2+1DBFk%*B5Uy)iIy7^xspdB
ztB-c-xqV*0cHh~jc`uJcP?o~NwIpI-!5`c19XFcEHT7?Tg@reUw-Z^Uk{mAvBU(H?
z=iKpjPno{#kamz+lmVGig)T7Q6?J=zB<Whrs$u@m^2iM~ZY$~hVkl)aO2!Z~Ed>cO
z%hXzEVROhq8GQvWuQOV7ycD&S-6>h(kdmh6@Bs5O?Mm_e-Cynw@lVqf5fPE4m1U;R
zHo<>WG+J6)-*0g+eMUoM+n$ZybGCMLMh_2m(2<iHf!!xfWc}g1h2-1v?O~EP0-u0D
zv|v_sZekE*I=_uqN^@sZpeNa^9r(G)<l=N#`7`d%kBKk}>(}@92wHw))NK2xK0d{k
z6-<SGmPE-V1ubRYz-HBxk=T<h+>t$+`Lo9Q%6g*~&u#gNT{Mp@_hBnJN>Ltq@cbdF
zcf<SUORj~dIdA|xy#DklXbTr>dU{%xf~eP&h2`)arZmd!lhLI$XQ?}pNCMC5YP6o6
zh#ePtta#cb$p2Onha8C8jjcz-GHnjHJx3M&!SHsfM@KW}KB&dPI?sGKDQxJCu2fSr
zc%jQj3QyP&{ZDcbGdgw|jIf2eFP3DgM)p4PeYC`RRIITZ4m#?H5&CSesF0^bahT3Z
z@PZ^YapB>-kw%4nP;i7LkCeLfYd8gdUS6&l0!GXaFk;3?RjntUZS1~BaI=@rC+0P+
zi6~=hQ8E@}AG&N{FX06du8E9mPO*pyZns=C^VP$Duwj^&!)2FHzt8+6)t`o{U|gR*
zzcLvsoJX;-7As!8zWnwJJ|RAUM-vm|@G3n83VINYg9Q=L;uT~^M(-!BuZXPpBV@3&
zQ)E~LSw#kg2iCmTWsmr4a>EzUJqoSHl#bMCpYxk_nE)xHUwzZlf%X56m%Q5`a1H{<
zM)+v>c9@ktQ9gqs!)AKe(A~)%>YK1_Q<5ZS4iXqXntYgAyh6n%IG5yreBpU0wm>7=
z(K6+wSv0by=RamiA2GMJAZSFaKK;PJJbpsyB^fT(l+R@_eetl)-5H?hN06*zDxm6b
zvfvo}0aR$ww4yVTjOXmxT>WG}_pGm{SmV=UN)8NoO-nCYIqm<d|KT|@gYon2hFN!}
ztxAfKvzj4AMh`Pd{p9NN4$g|zrvb}Mn44sS^&OPRad+PQFRuxOpWe#FcWz2b-b~==
z^<=!`rBW5GdpGYb3jqO);5O<q<RQi>oVL33eA+jmk`^tPR8mt5)~W2@g@PEwQniE2
zU;Xx!q5|<NuEo>)U`Z+t;_|Q4;g0;vL}4)%u8bTv_+x0$?uD5reGFZ(@Yi+u9A-@L
z7=dBCfv@JrqDjmD;Bnv~u=9mx3S7^dfQDv=i!Et*c=#}hg44rBOBWq3gJ02)kH&H8
zmp5(~2V=N#ROuTV?0)Por3nEaSq7`J$Lo#F(yFQy?h)YAvd`-rXRRCZ;WHP+yex2&
zlhqrT^djx7(=Z3nFfT?8S@&z0Z2PS16QU%wYQ6D?Q>%X;W>XBTZ_Jc3EuxW9;zt^(
z7Sq-p@^Zl{7bIpgKVxX-WN*Nk&0FJ43-gBd*-)cj0-`WRmxi>{olp{ubYn12dB&R4
zTBL7l3(K=oN_LoHd3&435h`V78ra2Kw(=H{{;g+wyfbyrV)Sl}+Zs7CG&{<J8th|6
zBkWy%FXMt+Y0*;qSVEn-S6?(8F5C&J%*<|mge1Np&^l=T%gpYxt!LeZu1}OrhEwDd
z70kznDt{mgN1_<wan$D&FGGKC7bYYkrh+6;`^4(`fB+=@-Vgk1#>W|)N1iUYrKWF0
z3OCN;VIY<^hRR<Uo(+TVTde7w8n-OrAlJw#OrLwA=48dCS?eg3l=OSdQVqmkCdjh&
z^mN^Nq8l0@=GNvo$w{VD{E;69<XBRQia1gukhz=h`UIdO>|0u2_bxsEtRzFUTLRnK
zydfHdU%w8NR8;lZG)#FM;gM?Vj%)=7IP(pGcfo3u*l0IA0l^_yr-)9oNXuuxby;qS
zEh#tYxeGZt`Kl4TrBP^7)%+|^fm2w#i}j)B>6b1VK?V-%#GE1p=uSd31dg2DosAoW
zT0A1WQF8@X*ZjHtcuIWb84Og|QhNJ;#QP#9W5pd!@sWviz<_M2T^9vpFKz7Xa*9!^
zx^a>L!_q7mk||W^uqs28gqjtm-R$3jBf_|2&t@p|Cnc)N5vAelm3t<e60vZ%P1p!a
zN*&p!ydriER49<}<vB!jBQRs((<703;o^<PQE3!iYtA2I9gZeQzzc$9&t>n5?><z}
zl1)`EDrVg^VrzTx*IAp3z{RcWiQgp>Q|ploq)Nq+n9Dex<>C6o>@bC~t;UJwDjFq>
z>WA?CxN^vbBB+Jg7!ZCw5A^}T4gU|UBbs-sI<<gg@IXK$m{Cd2$=P)^*Hf7K>@b$9
z&?!W!Tpn^@&2uUMPb|B)kE>mO?bc=i60%~O!hM)c*(q~RuW+=f6MY6RmNHFs7?YBI
z<J$R#-URjc*LiidwAO#Ex^TQ*-ZivGC-K1&AdxHAQ2llErk~oH?yJqniDo8g^Q}U8
z5(d(*7vH_ZVg0SKY!*Jx6t&2s=;OuYw-_#bzb^jR)ZoPCR*fOm$)(GPFC<VwY;?pg
z{tmNd68LkEC&_sXoR>#xPhTjUmJ(WNH@!Nq2|btoenE`(^5MT}lSYOJmP}A}335HV
z*X<`j5sZ(|)4q8cWEgaBCJa#$M1Qm7@N7{xXG>Q&<(S@g22rhSrI-7xk%44wR?GP^
z{)8P@NHK!*F;;fp4B3494>QuP?T0)4PTzHO-W~ce!;1ViWdklQk*{AFJv=-vudnNT
zpFB_6pb7x?4(m6zAey|@2S39f1_te!79z>2NG;0lS`&o5{iBa5&jM@9IIsct_Yx_L
zdNNkh<S;NWM@vw8zHE}r(Xk4*h=p`~jY|Db0`VWKaqXKO5fOqKas|2;Im?x;-ljQL
zz*-ao0BbKACs;0OvCL85-RvEpaYn{QScHV6%#>e$*i?1CdRucBh=QM)w%&?>cHUS0
z*8~uUJdj&&C-8hF;O`xs4Cv{Pb$n8+d6@vQJDJ+cEn!ju;xd2N?Qx8ow$M@GS@xor
zp^*kPW>5{VZzb$=cqs6fng(hKj77O+EHv}Fe-+q`Fq74)_ok!gg;B{kwEY#3yXM24
zn3yQ`!HB->;LM#GZ||@xUDIC}pXx(bj76@wkq<EpOpva3myT!viz11@@0oeT`e_n~
zD}%;oriAbLB{R`0>s!UbcXUIhfojyFa7#=@6%|-+9mz2<n7}QDQA_*b6O84Q$IPtk
zk)UguI|b0Z84grEm8cV-#-olu&-v9^)Wo{8iHM6+y?I0Ux4N*RB+6-QMUHx>ps6Va
z2H_p65bJ3Ej`*_InqAGpk*?P$r<}Y8Qn08yb#aMvs*<*<xRVL-Dn98KjRqw%g*@%{
zKp4rG{MiF-*Hzx$p4VtOdmqcW&ySq=e_2jTEH_JZCwwGJ*s@=8Xv+CXQZJagZx|Q^
z${REafU-Safl67LP8W@^RtvRVp?CMDU@A<i6gFStdsT<O9GeK`d&vaTewt$RIHW=8
z2;vaSKy;J`?N-;fddE1+h2rkT#mZJ3!0-}fjv}-M3GDyKk(1xy!TPMt>?4@Hwc89T
zoAVZri)vVM!)qN(>BL>aF<&f6F)=ss<Z-fG8Ir}}+^HWX;EK|SGWr&lgJ}yVnq~I)
z*2eP-IipbD<|#-5k?V@g%Pp*UNQmz9xolW5TT;ugc6~l}yI-IZM}XI;MzC7&-@Z!F
zQpzsGD$S8%2yG^mrrgZxbU~T4d=srfiHE49p%ZZ*d$pwW{~0d;L-tef?yw<dh#!P!
zfq^-@7+x^lUv{ehT6Qz-eZ`Rcr>S9#7eR4__lUP*x8=)tM<}WxMe=Zcb;<nC(<{1A
zXzSoAUR?V*<Jp~ZWOMUE!ZQAlV-xRn5b1B98?$QG>@khm#l^Y-OO6G924T`&IP@ug
zDelI6IzPB?q4idJw9ePRQT`RC{=@AYgg1d!d(dB9-)EAUl2#wn%7$QuRkhh@aq(w9
zwP`^f#_w)64ZPoj4v!+vNPo{&f`libOi3|i6iVqQBpZf1K~Il}n;?J^gLdrceUO2J
z`u1%tkaR^vMgom7DD>mYmMEX4h@b#kVbF3*8&M2Xx4f)-0j(oE<{)?*NCq^1tm$%g
zC?(%Kzy12Zs5<MgsKTz@s~{mrH%JcM-Q6iIB{?7h(w)-X-O{C`v~<S+0!nvDOE;Xy
z^PTsc^X~r?uaRqJp1s$)*SdfA1!rT;(J|iC@7aNI1S21fse^3G=@;4~h@6h9B~yT_
z04vw37#ta)%2xRoOiSu_?*MKoe}Pz|AE{LEbD|cCygw#PCR>@PIy(?kl3UnVhe&@W
z=sS2Ho@ky95`{r^iHJ@LVV2kYJ?(f=pSv1q^vD79O&S8!pUcZ2(swm{fOLDpKafek
zCc=r=j+!v7+omQVUdr1aL*<=_?-LW}Mj%%!*H^2`p54*D<Eu5gQb_p-1p~^#;)#$z
zf&aH1zogsW)SzB&uJnEZ+%UoqI9<jVx~dvR?_#?$Mm~t#d4TyFNVJ4L`AhPhzWQnP
zOUhKO!EPmn&MLfb>c{;li;S>FuNkJmJQ_v$m;4_|2?0Wr;RIdAf6t$iX=rFMr7#13
z_HLdt!TCHWjl*Mpw-CbCAaEht+q0T;KeC{q5&E9}2Yxje-)OD#r`b7r?BBh{V@jvH
zHmTOjQ{K#dde@A!v^3+bTI#I5BMaLxML#3n@s8@iR3ndgORgwL5#jDW&A4q3r*I%_
z?Mr2G#AE^oQEKHzir`~*S-Nii312w$RRwb_^N!J=9FyE#FTuO_K5y!|U9yB60&<R2
zK%hYyswYdcCCW%DQ<r6J!aLROF?^%#>u}ex6;F|^qCOD#5mr&v9-$$;1TQHbu*RpD
z@^$-9jjPQ~@OKy2Gb>s2wgPMe`^T4A9UrmR<VcL@{7X62bqo|-T$mH!HJro+yh6)L
z^~D@?;f5@ODnuVKA@%XUFW%V$n23h1u0Ll-m=9vCSz{UnX0e9}mGB$d*^~3cuz;vJ
zDAe9Q>^sDm4WPbi24-9DaGmwupy<8B82zK8qY*S$DqAvDb90Y^0Z&*60qN@9-4cK%
zt@kY^J|2@sh}8G4^WogtB?0v=E&XzGj&qc;^)ip_SQ<5I!=6ud2+QYzO6cMsl0?WJ
zBUf~Au$PM_F^7t_B^{+N99f3ufkGl2lI)`V<91y|h!ow&%rM6F5+hn>*8PZPojzSh
z=1Z*!X36cH>Rc9q!Oh~`$kW=&(KX=3p?oZ@G9N%0cQRgKA^q_rV$n7w1%=zrB00@M
zJ9@lXTJ$<*4U<{%n6CK0CCtA$RiBZN?ih`He%kAR>Ai#eD&To~x~;Nl-Fv`7`=GB!
z%!6U4Cu1`uo6fgG<XQ5~`+joeP26)-v~u@7`n0T|5mpfYDONSIEp=9aH$IJ-?zdcb
zDQn7pz0JFp{BHS?UEWjpBeryy6b6>V3*U5OHR03z6dQNk*bl5s$pZTqUOq&H+bIWQ
zwKwr8Xa2M4EP55e-4fo>)^Y3M2Rms>bF1nVK4)@Dx?>1lqE8-|zRMD$u%4v!;>HEP
z-}@s9KNp-LaCi)wkL~{;iXjh)81K&P#qBybJZOQP;_GWZaE_%?%$)_^5Je+o!a|0i
z0uRSvmcL{@^2x%&0<7HP?;aglI649Vt1|F6B-C!SEfWR2S#<c^p?Cf6&l^17>(e*)
z=gGOXm-*Zu?%SUYvRltp|46fL)H<DbvwzcA#nJ~ATEOe!#Wu9Q_<?`sHTc5+8Gx+<
z6n#iY$Y0O4LiIA6DV5tE*)&2W8Nkuh)5EKhU`BGzvf0=TC0(mZ^@$y~yCx<2mFv)@
zo`MOFBoz?l(NB04Wjamlu^arM`Q~Y<w+H<JuT6Yl=5aUE!1Kf$lwc5gn6ls9lclAB
z90yoqrN!8ygDFR!JLixOe90^EOFLe6LBB(5)Yx?`(Do-UDy?FYlq>mX`TfJ^nO2hH
zp0tzek@HWMK#X<ydp?5baoLP7SCn74f%zkmD$Pd4aIrq}(O;u*5*a|~<~saDwUU#Q
zQt_%i&OAoPZvK#w=KUxS`Aw}j5%D5mY`A@+XVGt^s#@UerzJO&^(P6OUzSp@LEOlf
zva<BzZ41hHK5K1e8i1H8#l7xcS&D+0?5wSb$O++X$A|ijFjgH==A|^TH3liNG^gEr
z<I~a#mj=`kM3lQ*x*=LO7y!-;7hY!IWG%<RrGZ6BNn8SZWy`JS@W|4EyQ!dJCQ-cR
zYOStLn<IDqg^Tyvwx98pqF}g33E}b$j$ni6U@+MY!Okav*qBO{5E>eOZjRAFbkSaq
zOkV8BzUxD{D~BJ47*$6r18yB|D9gH{12H6PU5I3m^su^53p_0Io-8Z3lRLU|lY?XU
zIa@jl6asHa2iB_3Iy$c3YO{XY?!!0i6yzLf8Xjw}6mdOX>M8T50ElK;uDzc3mQWKH
z{!{~Jk$H_k5|3MR?J9lP!6qL|s9aBgOutZKn$s)7qhqMMbJ*lMr3#5)f5+0LPmbTZ
zTEwUo#}I)_eZBV+Vk0&ZJQ93QQfl(Vyyjc~p6-TknU*#>auAZ$n#0({{n{x(wtzkx
z2|P{IbM$;QWIhttQxZHY>++w(u+d=CUTigOJA;w_`Vrk`SMC9UmR&*uNo~R*xfc>b
z5|vZp<{vRjrvdZi?;%*>*t&7euu}M_*Yr6iEM^j=azIz-^Bd7bGdQ7}X(yZ%oDCM%
zqK95~!@1T}9VQ8y?CrrkNx9D1K{jS(7(tYZ#Wu~{$)4@y(cHig_NPUM{lhFi-vmpp
z2t6(iZcbUc)q?#$wgQGe36Y<k^RRv_{P7*9ey{C;N%OzPSJ#o4;hjT?bAJ^uYAMoX
z5+9s|&>C0JR<8T++}h(Y6Gmo!Hf|&MHeAjKRsU^0r&av;5oRHJ+-n@C+n}_~XcT|5
z_YT^A$(=r2Zo|XXk$UhAo+9%TXqxQ~qojcQ%fGw(Aa~^!o<ZvpY#(HrgZV~^M0ZN;
z4-+^how(Oiw|8|EtGe^hz1P<A+$_T(R-O<0d7J5@h^aK#9{diPJ|(qXX+P5*7H{GX
zqK~goBpNu}ic!DR2ER<fc@ujdh8XL?Z<drA&T-y7elvu|n$%d^>5wwB4|o0{3*r7o
zz2aAiTO!bKfSlXc*XQi!w&5)BHWmtEF@PNk<mYUh01$l`%z}!QH5QPTK*Qzh=em+8
zQ$UfeLM1;*C|Q2e?%;cy-0J>k46*rSf!DDQ_O51Uw+-5;(`NfHY5I7X!SR(}pxXqu
z!`Cq^`)(jyYCK1!gKXRFo3fG;M(hCChz!dQo3jyl!$gxPU$~-0?UOoLhO?#<hWhcm
z7=G<94lx63Z+<;Hex{f}&sc*fegOIE6|BZ-041x@<aO)i0o9Cc&yl_T_3HCIB4Rlj
zv-`u9`wQ#VT`1s$`S|z%2>dDoU2#iof@C&!($4P3N+&0<DTG(gVlk6vR?3Yfz~9{v
zDH70Uy8dSku8DEcT_{mgQxo%*Z1uw7wKwPe_Og))535wdOR4(skPy!nc-+Rvk*k`W
zw3``yyFOmQ9=CQkM$k_Myug=TQ2B!EqVrArl!PZr54YpymoP9a=GY4beK1Z52g1xx
zIz2VTn7$xAW@C;t2Z1c_lgOvegd#-;u&jDS2z9q4sB#P0E~bfiHFGxC(H<T+;3Wq=
z|EgQ}%15sc^L@nPks*SNu#VzLvuMt0?I(RC#1Bn7e#d66wi(Ah$GJdZU9+EUKCK;b
z=iA)TSzy<EvAOqp@u?t}G70~rsLC66yeU-t7=B;@P2-(b2!>u~zcT(w@7Rak#V1Pr
zcdTpf`ou^t!l7C5?WDFhV~zhf17lG#-=wp(PhfVwf%4;A@tru*!7N8jhbHa!7k=B=
zxlJi0IRfHll5ml!OxuI;C{25N)_|Fkb#Soe)wm+B>%q@sjRuyaOBxzN+|~0Br38m8
z<B}sYPXHTrVkL0O#AK|{ys!K<WT*IM#YnUh_npxYP>v&Y$M^bNN!y-iz4x$Z3Z{0|
zkg>E|6Y<eVsK(zT?!hajRBcW?3gh76);+CK06dF%Q7{xrm^c-Vqe5;DNvSeWCZ{o}
z5hTog&CV^N;E@+82Z2S6$I_J(l7f+e5t;SxY{{6S6lbRXpYL2bS2H*EtShORiT}L$
zv$c4peX`ni<tVPpSybnwnUt)yG~IhylDl(SQB{06sCHXoX#5mTF=-`;*EX2MI(mDt
zZyy8C_uY*btra)%mw0+B!yO7jg49QOOs5<Zv;)uhv)>{kDZfVM8XElb9(^B=j-PT$
zNPWXPr%{y^wT%z<P1H1~UXB@@a*BVCCv-fQi5cN(F!X#`A00hw(|~>_8}Q%7QYdZq
z7=pmv)rTGq@xc%lBISk*Fw*)!bhL<Ubp676i&6Dn<?id4L<$9l7z6!ArvvAcQ*7r(
z>`t{e|2;G_ihE~CSlQb8l6~9?R=KAD`O<*be)cInoAs5=3L5G7e#vhulO?^IuZdKg
z0U?j;(V9)@l&-GrH-Tr$tm1?1H2z|)n-s>5mg!&hLN_wVJ!5I_t`ZYJeEZ68U`1CQ
zxRmqem>}AM)v(CZ$A%L6^#0peh>48ZB~A9&!H>bLESt~#qK+rgxs392Ra@y}Rd#C;
zBlr`SqHly;7x3MOw~8C4gJGr;N$OJfQ^2>g{`$EoJN|no%;S@ahjw+p9l@2Gu}~k(
zr&J&b$IC%J9}J2{%Sb}7D2SpHGBXeH6cPY(n}3N3CA0?>6%{1b$m!}5A_V$_-3(yh
zD47CaX0X)b`OV@ZPv-NcjTpgYZ%0B(f40_!<Ms3>pY2aGG)VR%qD(vXcT6ic@`6`F
z^NJrjV%^tA1I4u%a93`BH!kB1KOB4z{g+j{TotWQ1Z*K1>goYgyS-m{SDpFO)6?@n
z;Csi4pC{Q|e>8E~*RPv7UPm`=<5cDl&(;>XUO0p6n)=yKt_<os<mDIl4Tmb*l$<58
zu%V%~;!Ioca!<=UH*9jXqeGWyF^z7jak+2bD^qjU`DW<r)$o&`US6;ZafCs}pF#pd
zK!1|q@BjR$^(vL?w|2GFjcG^;HtJicpB30p2;n!C@w`Apo$pZUbZ4SO=2U0O5S4G;
zWJv#Lt*hwpO?%~x1Vkk3icj{O{GRELiGpbtH_azgx%p?C?ERs)r7VnV8o4bY&Gr#<
z&ALo@*oyd#>-L-ZzX-(@%<A+A-n=RL(Gv1`{H?jm8-5GUoJST025X%bXF_z*Tgaz@
zSnNodnt76>ly|DSy5hRQnICd|Lu5&PaXxD@$SW#h$J+Tt=}2^QVw;`h3y`?JQA}m+
z!jHlZ@2lC;X}Yu%k-X}Ke-!(?W3;He@|SHFb@5{AX?4}#XBH+iW~3*p`;oqaCu`)T
zJ}>+3)lY7osJVomKdPefak<0g3OOq$Bp2Mjt|>8dydQOoX=s~oyQ7;T&-mu&-l`|4
zB3qqX`3pq*hlxW?%&>QCbztk}%!^llo&J1RNtsF0ZHLm|k_hg)qh>CR`BM$hF)(*=
zazrgzm|f+xG1(Y=F8#&MF)gd%^D>fXZxFY8(dk&y;I8Z~#qdWZ&o|2ra5#JL&eDBh
z>8LkaxTCj{md>6T_}nHFk(P031aHL5>{YXl#WaVc5(&8k-jFr_p<>c$!pPKp8#5$k
z^u*@c`r1vJzsP(`vt=nLz*iw+<j<Z!-|~_dm`%Lo@gmKUN6#;n9xkLCbW$_cW*6Wd
z;T-P}GuL|UT9K^8AT5&sx<0HJ8M^XP2vIl($L!noo*<mpc?_Y7s{h#qm?RAA$J(6D
zo;}c3ctzrF<df0e65NaB;5h(C@SYl{<<*eeyk#XlguWt<6U*b}4M7o~bkP<L9?nDy
zlyJyB3YIx$?yyf80i<oumbBB*6ZaB6ZXs!oWK(0yH1m;nSP3uH)48IfQPR>$cJ9x3
z;&Moee}A-))}olwL&=Z|7}}K|UevOzZ@F5m)cWtTt8V+MC)PJ<NgOX6s;_<)*|yBr
z%0>i35DgBlc0chFOofQsvF5}Ddy%o9&YI0-?W{+xt;c4!DP5=pP+V`ZzNDd9?Lstf
zunBP*A4=&?VG6FdUbW2p%+!<et@_K!=y-uTvgI4R!o)+%SkghB{p|r&E3P@2nS05U
zIx4z~itwGyvxyBo6)6V7dREiLHoU8geGEsY54KK2j;>^LgGlg!bND9hFL<BkB`<{#
z;BauePPgWu_7*g)qPCCMH^DY%igTLfHl-;FU9Wkef%T1ST9x^8M<pF@Vf?vd$3dY#
z*C;|{WKNXEF_i<-I-*8wZv`uLu35@`$etC+Up~*Y!y6qNOOS*7TX}Sg0WPem4Xj^Q
z+`tuPo1~?wNsa{euyNFI99pvE2*aTq^5Djn02#G7+r^OAjQq+1n4Gp0Z*#q)pfooz
zGc+TAw-(UYL~tJq+7UhgfeRp^0fhdSmGn354cAuSkj)(jfM~>8<uXy{#q}R9JIvzK
zVPPjEj?T?WAqa$7yKCPLTSRHooij8A<QXtGir%?7oeZ#;*dF1msS;!^iJRmk+A}H0
z=V1HMiP4b&=zEqd*(39f2MfdE7E~RJM>%e;R7{VJjT@hmbv*3XQU3lWyJo=WtoXrt
z0{NlOGLkM@uQfCh7kwALfsV}$sLvMJArajY5?ype^%0=98ic5sPm(OxhjLU94y@y4
zWa44G3XL}pH)fNQmq(U-fKwX3L6R-7lwvmDG6#Bu(Q0<iX3KsR6)t+sAr*eLy~!K&
z#Vc!rX0+mx;;@iF=kJY)VZk+eOyr+SfBwC%8^=UN*xmahWA}Z%g09D;5xl`$)WObI
zug|(;vYqR5du`n&{H~E^-ajFOVdBkr?g}h?V?bBcXrP4}MsjpO4j~?7jEq<5)opZz
zL-;mGTXj#Mp)qxi7Hz)*hi=9}bkkRLWY!h%vgx3ClVDWnWYUCTSn*?N`4_ioNse?0
zReuFDvD668i5XaLFHRe+fX<wOqPPiZPB_JPG+No1n28^j6dM|x-`$)YzaR23h+7;E
z;~G@Ea5@SoFdghHa*u#TEi)U-Fp_Ac>{h~D(cbWe9;UBVYf!ds|Lj&Zvc2v?{Pc_q
z3Udep<5E-%s)ni#QtfEdfMoL>B+nQlyGzPklJjVoL4`yl<f}Sp<iu<%QL?prOEoxW
z-Cjj<r&=;aEE{*>$J9E{aY)oa!oMqd-8sk%gv?c+2o?Szu_9;V?q2*JyJ&zvF8cHW
zsE;qDU-nWt0B|=#AX*(gVzdk-C5)5n#~c21F^TBkmc61oB|jj!QsprOR$#u`e_$^r
zB&L`rX5SmA5=J3!8}SN7zBmCjS|u(#B#uV5I-C;E1tI#G`AeSufqueF1yfe*)!PgF
z&vDzG%RYamDW|5eL0Wwe)X=&CI6%np!p~THsBoGbhkhZA2H&QKAC~;64lBR))V5oU
z7`iukl=}boqro5z<0IFqqq!{nSnNZlU3}TED%RU!C^^oHDqiAAoo_!jPOBReAosx+
z2i_{<M@w@h!y|$Dls5-d7%RcuFJ>?HGFMo6(v6o|4&VU)Iq~N79XaOHo41e=orF4!
zOjtwR_R5CuMyYiD?Zxkuam0-c9b)2}`<2pgnG5fS+}~%TihrkY3$naE;%JQ#O3;W8
z@^nQ+)PAe8=Ob3_->^gf#YI%DL%_zphG8>zkpE@%LX|0b*wC6TN!a(1Ncv>-`I~2y
zM^2(69ba1m@_=tD+T%o)vJiuP?$rKWL-m*b{#N~?^DfnYL$#UxzKs*(#szv-R$81_
zCR{}FB~t{b2<l}xwM@CTEXe;jgZ*pvP-4q~&)OB#$w~=IG-_p<1V;PzR$LiV%-_AO
zMcjP1Wg(pOU(G2C9@W-A5T$t{Y-t@I87{Znp#n%4FiZoNx2N0NQzA7Sz)A!iRa|cF
zTbjf*&fFW+?Cfm!HpB0F%pf1lxo7Oq)vR7`uE}BJ*d@mUp&$oktmBzDHO8Oxd}nc%
za6<I`q6<5aQ0tt3Bk3FA+X*lJrL;P1zp5?8lIc}x)Aqb|+KFe)ybaUG9})iXCimEI
zK~!qRFXX*NM1TF_mXZ)39%hL9MFf<H@?j@Ag}Aa(^OeeM9Q>9)M}NL7zVF)X<$JBn
zI9vBb4(4{c!_|&`j+y~Vva*Q#GoQ%g!xE!tOYxk^2c2{G`bxFpjZ~wS;{us+X_!if
zSXn7j6cWN{)3Y)<-y6yZr5EP4=ZyOt)fwEJC=UKo%<Ax<LKi(Hc<IqhJ^}{K>_>!%
zx|Ddx8;-Mlvwjgpim8}r!9P>?!u#^4dAZ_LiU`QAsvUnRa1e&e@kM=(D#da?_xIxW
z&X4T-#=N<;RpYQBbD&Ku6G-}@GEnMVk6|Q&-$gxg?u*Ghz9|D=pvtSWA?qqrM}_?6
zkIz{T=cAV_1}qvi40M8Iz7?025+o}~%j*(SOG}do`bJqYDC=p1;6&{h5`HD0_riM7
zavzx@=4`r6>*B3BI6TV(e~=J*nebPxTLt9VN(|8x;qy3b7%hG7GtF!C`}{`GRdw4{
zWA(enn(Nsnb}T@rTd3}zoTJrP+?Ym5B)E*|BZn5Cs(9!&LHXO-d^Wdn<BSZ3r-MS{
zyPgx_F^N_D;^5iNYc^+RDT$V%?AuIARbmL%#vvpmHc?YV!7U1&g;~JUM|1P0K0ZF@
z62R6Dh0|i_&v^)S*qjgVqYNXm{8aWq+>99!oh+}g^qJc4$ra&{x-W4>%r^&fZJ`;|
zin}7eB;`#o^|E9B$)K|RmX@Uh#$24-+~R7w{CJVUZLL?;7x!o3vIUR?LY2PyG!hT4
zMix_XALgT4^_T(ZX~c-??s=FecEQ7YolmX^LV=n|so2r|appSJ_UwAuxNhl*w+PsY
z_036|+yLV--unlMCz5dd9H)-6j>*cfcj7Ob$O;KwV*2ksW+^!Fgr9WGjQAVrf64di
z`LaDq;JE+a93R|_BAya8%V%V7!cARiOa!uhMiC_g>^(Ip8<$@Rx^}*IGk3FZ(+`JF
zZ`eSU+Mn$(8XuJ7@#4(%8;GgCnnSnq3We)7X}Wj(bKu=3jbMj1xJ_(Mq|8(5`B%L3
zYfuJJl#D;I2z{MTYv)^1R`Z|I93Q+94Kd<SA~E#zYQ^6PJ$kBMrfh6ynk2s#sIKU=
z7zwfddpiSeGQOV@h7M*}dGUuwR``2yJ~iq8UbXeWXI&0G7~ffX>cHUU=ANFO&a#!#
z{igkS>ag^tLv+f$a)wy*#ot@@vw@BikeMJoCCU{)3b|?C-a!%*^4YElKc*GOXac<5
z6^#W(uchu?ub-A&0KX2@kkt0Wo9(L>;j__&0h#s%r?P1VvF(I!&lsQ{)JEI=ytqR_
zGZI6h_~}XrZ7KsGsin-$Myx;`7`?+}sKWB`+O$We&$(6}FB=K3RLJ+DW^J@V>=a8$
zVt^^{ax(^vBwCG5z#2Pxtk4ZkTS>8+;`2T0Vcj2v$%z9{u&@K*>&?x&N;I6!Xo$(U
zww^7mA9cTnfk7Y|nFwlt7(;w<fft6C?Mv&ikNr8;)Og`v@HE5b59~-mM)t(&vo%#D
z-4vvkZ~$Lsqs>Sq`H*FEygpn}WojiroIzQ{U;TksztPtG_(=Gxeq#*a*}pnH0~DiP
zQ}$OVk)@U8G$d%?I|f)%*m?G8m!LwG##EaQwER+98fRVA7NS+5ZRYiDC`}FmAo@)2
zClzG9z#9=FMlROe8o(Tx@;C6<Csn>GR+VYUJk>WTMuqG_GI4rlrq-lK!1Z`p5mZYz
zH#a20o;aaX@+NBRoQAA2b-~h+c<fwUL5!@i-!gfpmY1Y~V|lIGjGhJJ$QI62M$bZ5
zn4b^tmLBO$DK3zm*W8R7*#|s}vgtWlP-pKfXNIA2?tMP@)n|8CZ)D7vzj&H3b9T|a
z6{*RNNbtTq5bcc;?F_ltXD0+l-ND9{5!;OF3l?2PDLD!1#;9_2bq(f+oXC$=sX$F1
zrGYDxN<bv@{G8wY0*}=;zpxNTizcw&lDwldNK@v8I84<|u$Y9A0r&bx;=%G3$g7Oz
zb8<>2I_!C$5UNF6<9y>7rHyDNd>9uoF+B~UmP}2(Nz^K+@ggzN5Oo`D=z%I&sNq^(
zVXaedq2+2|riO_SNE$g5HkZ-^C|*sERkoj%h}5QQ7gG=!o+s-;@S;y9Ifc(JuIe27
z)I($k*oj>rKTX|chsfjQ;ZIOq&i2|ROQY^|J8Ze-K9-j=mQ|Lm{axz;E;UfYAYVUl
zadCs@SDuex6GYKj%qcL69CLx)88b^RtLfwVfvW#1L{3sAaO(Uc4Re<ql9enZzud^o
z5tIEJC2?_{jg1SkVtlS-c~Pl{)@EE(K&9V3X<5tIkToW{{H@JegiPqx()Aw7bb0Gr
zbz^@J{XAVI)yJ}aE>`q`%X?2FF0m)zrTuP^{rRBoy95dp28-yTK^IpWJVaBl1th$)
zz5GM}AlhqU6F3X`NL2MQmK1rO*3)5gG`f6QpRSp8o+&HOS;xqR7eB_~%k)~_{1r4;
zkh^?-w}$RF_}*<Pm@Wd0nWhSGUjpzbgo_Ba`!nG97dW5Sk5?Vzr&}W{I0CQ^lguIL
za%xR*06Ck0z->dvO079cTU(o7C-#{jE$MLWD2#Vyw8drr`WkFI<*4M<F8nm|q_4R<
zl8liy65L(b+Sxff9brvOJY(Sn({iY8F~F6|lKBKFR{qGz8FQ6ExV6-2ij(6~7$k$<
zO!i47sOlMZt@lNG9NoMjA1vi2(j?+8bI#}<R*3ojh6)1}14YW=U6onSHub^0FYodY
zGEFD~JsWZSq&!|@gX20>5cIxLQL!~98~jq;CO}4T5XdGGbVben*?AgGw$|zIbKreR
zk1F%#*Aa(`v-8i1i7;}aKYKO<$vXkgQQQ4Qs^Q)bWE~NZo=<KMyl{bm_u0S1_t&*=
zz3%8n@1EuakFN*EZVg3Clyc)OroB?N^*&V@Hojl5uF_YkQ<h-o3xX_SAl(!?_}i~Z
z3@giZt_H;g*~1}3)!>^g+OJ%A&MaiQ{?e{6lp#_+-pRUpbl^((&2u@AV|L_m|5G7j
zYWXaOP+nx*+pmM8{qJta>OibWI0dSVtZaBl52V{<QH3^>*KT_vPZDs;5@6=LM|$!i
zRhX}Wb($TiX=BL%a-UO3=tV;Bj>waT5b1cg(_8YmIEZ~IgWqk9==SfG=D?@NZNCX#
zw$43RYJk<o99(GgUdF!V!SZN_-~Pfu(=AcUjGc+6DM`;~Z=R7g$xuR4!s6oMknL@n
zfG$FgIn)r7Y6}<<RF@`sWMXDU7SvuKq%3zzMOGi>@cn!Ez^N<|w%{xzOyW1xMl%8B
zbXkXj14pu?MHL0n^%d_NTXs-$pH}>Oy!rZ5MjEE()JIV6E-|yx*WQZU+alcE(z_US
z#gh7^ND0Y|?`qJM2RfH0B{ROObvrhcsH13U(f*uLRM<pdf`s~fXhP_t9XsbRPP)=7
zznWNpt4AXhQea6f`T9LbogYNB5GY9SBvR+3>NK_l>aX0OalMZlE#!Bp4P9y5@IlQM
zeQV}Ie|5h)VBm7d!eF)c7w;uf@b+x08*2p9)N*q`z4g1PnVEr3W6=ujs_pBV4b2%%
zr~5zbV^>!uZhWQX4hgWG>z{$5G;dxztk21w2xN~26*NUe3@N=fbsyd(PA^<4P@B<}
z)UZZ*5Ekb0`lPB26GqQ9AK7ATe!9D||I-)v)sCU0VNsSz;4Nh~WugK&WnI(G7_Cd<
zF6R8^{?Fii;0!v>h;F^c!6QLt>@_M@lyrO?!5=c?5$e>oW>E|P+xxF;St*FENJ;#j
zkHmtbERKc|GW@Eo_S8dT+xuaEV&WMLR^=1rb2m8aoi&zXe;w9)-9(skdh?|Wna9>W
zdH8dOTz|h3VZBu{eQU~D>=V+;O$_~zV{>Y%9upEr>(T-C_C*Hm>Q!bez8=$D=kDh*
z7N?(#@#8&aZ|O#1xKqDc;>TXA5FsKML!tD|@AzET_9g7TC#VNag(tzpX+9N?|Na)p
zZ4KleSiAW<1C!r1-@DpdMY@+lebE#x7Dj2?Z;*L9hK@NKA#+t26z~A84!l|5+Xn(k
zHYXKXrM0v`<I`;LZ*2n1;(aSO0By(lJ#D;Pwd<kYSdOqF$YBO|n3NPe5N^7=zb~n!
z1!n62NMrYhuS(Mv69h=bSaJ>T{i?wCD)YH&T`2<p8iXhR{-!^WBkc9ezS#i&#dE+K
zd|ql}XR;J6IClo~k{d{@YV!LGgd36E4*+{Q!qOTwHm1^k@y%}S>aMJ+YUlc7BXa-o
zG1M<Nd<YpV4uM~{u(dVwBt>TZoq_W9iR=)m&=qIwS>%ELWrdiQ7V!M*)LU^E1hx%I
zlsNrj_+bdV`w8VOR+H%ve@b30T=S+Zw95IKz`C{%YB=fs%bbI?y&WPC5WmxKQFEw!
z*kP)jiK7LAmjf+l+ow(L?0RC}{67Nb=37#x>@DxAOr32|-_O)690Gq7hP1Y-Rx<EM
zkVL>$!BpOvp%DN6y?%knx8Rl1zpah)OGca5(yFvz5>9f8(9_e?-YYUajbLX?P&5oy
ztg5lsyT!EK2Y8t+pHRx&1ByfR&`6I>G6zP=a;9u%t4UjzWc0m=!=VYd@3Ton9Y}({
zOp~qNw0R-RQn?%0F0$auVeS8}q)3D@dFqD3##%1+0q<3ji%1eTfD>+~>9N?YI^T)9
z-Bk4A^*;zPAf5oabEfG=JlWECNHp=Fxg!$A&*{m2(4->2ETLyc#-Kxv8-mg?Fyyxt
z1cGRswE|hZss%~(2l)3hySuw<E&G*b7Ay+7ow7_sTfKA7V@Ub?BTlg@Tb;2HX(Zr3
ze85N;ZCiO@+34x9Z02%W<J;$zzn?Sm3E5yhDj3||Tx#q{6%O_C*erX1PSLY0)rM!<
zh{TWzs_1*Q>MrBTS=8AyAE?`uM}j^jE1`D{%<&ITu#^iO7z4nkK*>)6c%=Y$G|f4`
z5;#$3DM6X3qP~_oDCk4Ci<1FAb38wE(<SbChk(&JJq_Rq(UJq>6U)E5;Y2+G+&TLE
zAMPwa9g?f+!lp-2t!JvJNMkn~TLyP(ovRFjO_0F-Eogja<2oE0ishhi?;A3XKGkNy
z-2K_w^=LifQ)c|SxP*rw^S*XM*npgQ?SZH>PNQ<M*P&-X1a{AMk;w&Kaj}g1T71bz
zJ1d#Dy<q>Rq<4Rlb;XO*=;_2zPk+9&;#<+qAIu22=dK37B!mQF86=&Yu*;YFa@y_&
z_ld|U$`~%{>q!KvbLabMZ0z!I%7g#=>SM*BSgjctJx?1k=Q-7589coyA~gu+(#YeT
zTkH8dH{M+$@{&>e$LV89vGF+RX$8*&w%z(1E#&r2dBB3eB6(mE2wD%5>1m%?eUO2~
zz+=mGSN#B6N!z*FO#FL)2CB*P+W=~DZMk@6d}v=^ahR1Q+qhpIw}y$H(QDn16g=^F
zw0zk->N33D`<zsiAvH-vjMetc@K_&*l$W<9QpM|#9bLtTS<-$soJjR7?Al(`7XS7i
zpw*xa4_0_DI`Cu8SNJGfe*GKt+44FP#r5b1ksgrz`Bvxz9wc$nGcz*-SO-Mj6x<2C
zMS!~0>Q}k3=s3c>=rZ#jWU-?Fqd548gpWPYjczAXXaA>S`WNaxKd;w*nq0+pMRcMi
z-{Aj@g5`U6M1I`(j*&Kz<F9fpB^8y^npmHIyY9eFntQ)VJJ;?b2smZykw*8FPp7O;
z5+6U(1EM+h)sh<-NPkECa#*X_bc=qxIU7s5KJg%WlWukkCS4<A*@FK9MVoH6GadLy
zz<Pl8H$7>V)PmB|Ffe-pFa9ra2OWKCZVqVCz?<f{vL-qEbYo3G8;f-}(qVABc8&#>
z3Hd?TygrS*1wntnG&YjSGmZr4uo5S6y=n&b^KP8a+r%$Nt59-bxyd66ey`Y2{qvDf
zYgF;DF(h6Li6v|XAHvr%8L7U|M0zekE-xT4D3=g`fsnw&{HTgA!JBJV9$^y}eDXp<
zLT&Vc03~*q=}EW!e36l4fu!Pj?e`{4@=hJTar3i_+R5Tz_0EK#I31FX^g-yHd+JzS
zbvzq<Kv8X82*%(x7aabpnJwh;)EIFj|G3ec6o@LsDTf;sDlZd1>Ol|B;$l)s%Y6R<
z?FFg{IrAuy1kXUE4oTqGK@0U!HoTxlovgQJ4py)Yxfelfarpfok^JeD!q^(|5sZ<X
z1f>nrz7{cxt+n88zt>HplIN)^s`*RuUR$Q2b7W*3FDFxR#QD7}uQRkSgo#iNi{Q4K
zJ6;P4he`;MB!`%p2o4GHzB+sM0(p+DV7C_9ZZ-NmqR(Nd7W$~%HR;rIPe>mZ7hi~y
zi&CX2G6T3I|Nphl8<A>>5D%jOo1Unso!5LyZ!-}XNkVi+lT3yRb$!`VIn7_E<(GaP
zRf76#bJ`D4<CD^6hdkVwjFU!LSM7LjX_I>DSR5Pjd0@%N!t~psqwVhw|DMF6lr#7>
zmsAd5g$?A2`CN-@{8d)TaNx;=B^gGTrZ6fMjXJ%YvY4~Mrx{RCQ0mgwfcx}mVkx>k
zAEmhgK|X$HT;@6voi=W7?|R3t1!({R4<Gm!(a(5Hm(OOb4w8<?{;%!1o(P?=^`NCn
z<e|hotA$`$Y>nhm)G05n^%U2&?$Xq)ufK7-kAUB-dd4V!$yjJ#aJ~xw9hL3=%gcs2
zevhDbMW3_5Yh3u0;^ifLrwI+R{jtu`!}cs0n^wi~@<2BuBZJRn4~Hc+a;epA<H8dd
ztE=dT6!vZNe%xqVAa-;u_HyELS1Qr;EV-?~z<~g|DGok9CkKZF=&D=YjzRuz==Qcn
z`&sreo3L<OQ+CeKXAJidMOIHV0G$$da9}SgDgu-MeE9^x=b?+Jorq)rz$~b*FR6S&
zSfA<UEn+odiRy3L&fm*2?TyFzu~)yvBvN3hjmGkL>FRJg7}&+7?5V@20iE&6^xYE8
z;&EaeZ3Eo}pXuk7o*KRzisvx<@q#c;xmr}Z?G$zKm~iuOLvxdHQ{5?9Gn4aR_Vj`G
z^KH!}n|&KhsP=g)i<2k87+;uJ4EeT$9Thl~rx=I7zIYj~<NlRg?^&UFjj?<}+^f6Q
z?g69Qi_`IuE9Q}rSUA_sU_vaEfS*~!7ey~D*7T(l@OO$=JD|G~{`B&bri9uWDH$WF
zv|0E?dmOqDR$pcc$WW3^f*2is&PD<}lR+$sb=mObq^H~$x#9KeR3o?^Huo~imQ}vb
zkde|_o0!+;dA&?1De5oG*YI;FpDT8&CQw5mP+i^c?xy0k6hkQMYUDxR|7PgMMV(I8
zvu4dtra=rHwf?JS1%G(#pmY`qnQy}DD9=EqQP$rR)Rs%wIxFVV^IaKW8y%p+<V^qd
z(`1<HndFx%v@bj49t)v_Lz)h&^y%2NqgJ6m|ERwlkl<7F*fO>3Q00h7ovX~&<coB;
zd$8Zb86C^WSDH#uU~1>8(v*wmC$X5UEFDaO!eC*)tkcStI{y~tYEc)<?q^?~xxf8v
z-Nt_Y6+81m%haF+jrGT~+$SMa^o}2+W@}X0AtkvqjYsOYW52K?)6qC_IF>Aoa(DT;
zVdtL^(b9a+gpX?=>3zgs?5Es#GJ;^4X`Y)M6Ln>|*16Qj?E(4U-lIOv`H`Q_zZ^k8
z9Ix{j@zc;eN8K6oi3kr5M@ccV(H+qF>p?X0NSG~nw6zLXF<2B>)vPR#O_Qy12C&M0
zCp9-W9?x!Xl_-9#tUT2os{cgFNGbCDrErGUaxyPIP*cO+-`}sQg{wSe*zlvV@d~$b
zDwag1`Q{Vdr^62R1f?mie_Wl_)zy3?2l-bKlzDPkx{@Wc`!rSiQtG-G0iVYsU-WY5
zax{~mg<_XXg)=$hz-JW3DHQ?R#$as}1w2~oYb~CzfiooOZj-vaE$(w~qQ(VDhHnM)
zxCgFvE6box`*$vZ9|bUcVHN}X7c7@sQ!YSC@*mI$f)IWVD&9JA|6$oYD)B;zB6miv
zIDs!4O$n0@Y_lPS_RvM>F0*IC+8x73gRgZ>(nq#TcgC{+Sqk|MTtQS^MkUn<he&R$
z6lKk9$FY_!X`0@(Avq}90doud7_c|At_tX0g_wNV7dSjTOiD>{xI7>P*|N2i8;;Gu
zj2O8=ALgPwg&_fW<alIf?n016`N`ey^j%QG#wlsabS|HGjM)QMO|8BO7w4tU{WAao
z{`U;!Buol63@68o9#Dk_YgN3<V_<lAfRpo#`C(!FETeMcvT10`5wN(x+Ff(7TQ6Ni
z)WV_oj0+BVCM;p<^?@rfT4*5fDU9z><-$w|1U!@__2pdV4_$>wS<5uBq}2wu9A|BK
z9Git~8&R~h<Y};+U0u|SjK-IRy}-w5($0aZ^q*Y-rgMW)>sj`RAaXk-8#LP5d~#I(
z!ofc+=|IF*(?ouaOVP>Z43*?%Eh{G<8SZX*A+5&FreU4#{A!79l$YwqIsY?2ovibH
zxSe_EM@0GWD{=+?#i4l?AM(C{^QKfcj!}!PLUjO02!%ODy;Rf2kZ-Xisw&Cl>3ABH
zCC3mW|9PwfD|~hK!5lP>;hY?%A=7xL_>s6I)OnIz2wS2B{xhsdv4}EIvaa8XmRQJ6
z&vwSr10!UVD<C8IDJe`zdd_^KT<PN<eusXo2t3m2)&D)gOVqiAkZMi8Xk6D7?qo>$
zL8fee{(2bhRp{DU=$b!b3UujI!Rn=WF<s=9(t9(X*)6NFY>jXk*rnCRMzOaN4QwQ3
zC{2q6oEfh$`#;NGd}+Jehh($jhl&I7MBCT#$fsIMHnvY&hJm?=fZm1-#m20y!`3uB
zGqlLLMf2o&2)RmjKjJ!Ke5ZaBuAlZOif&M?gcQAY_9S)}QH2>UB58RP&g&dyU=dP)
ze*EQA&(7@mB~_NkUSl}?RofbS2T@(mM&V8mj+VB@%v{%XdSvRmD#)N=^g&KrliO0z
z{PKxs=#8}^Pv#bo&Dd9NF5N~G{6CYu>`?5hoDWmPN~RW3*UkQ^N<b*Cp-f$l!S_&9
z_&V@xN0#o?v-HI4FAgrJ2r)&xLcBs@HU9y*f5%|xKpq>Je7+(N1YRXHQHoDazVj^;
zbG6_h-y5uu$}?7@r+2=m1xrFV=T3tW5^8FgAnl&H3hu?wL|Rs)s+xMh#N;M^f`uy}
zm$S}|dRxZCpAhOT%=5eKLBKlktY&alG2osLDA^;xH!a$3TLiLtv6O-ra=^^Hxe4Y7
z!AefXwOMG09tUU)uqg#XT=bTsi0>YJu}=6rl;J4;?Cs^3HHU))3Qi7z0Dz|(1cz*L
z7d*ccBO+eO$PVCP>j9qA?X4$hzDK`XP{62J72TfK=BsLGND3sIy4uEEwLSURELQus
z9zWvO&Y_lD^{}DR(9%r2e$xsI8Q3WL{xzw~UmQm|@9S68*nyt&b1sppR2@ojnnVOt
zo==NX(z^VjKI_Pi9%tS^L`2a-UxzApVPH`g74@z=L_~T6O&{xopop59FG~es{I#08
z_OPtzt(r}F^wae%uj(0ctp}W>ntbsT9S?Ct{yHOrskdpC4io5Tt;@ca8e7oFaS`?9
z%@gDj$ky=FGn}qpG^J&x=hfzCO?U@^Ujw{VNNL$9Ym5xVV7n5^AkC0YN%UGnb92DN
zja+`!L2#&0<Hr-QS{Zr68MW<K7+e0y2%#J)?e2^pi5s<YXB?hu^59(4$x5VES#uWn
zd+pte?J#-5XIl~tHS89Xm_Zfd88lCDR~)hD%NkWu-Z=3Lmo;k=lAa|&qoI+Nhphkc
z6i;AgJf?>iteh9OwyvNa`p;u{IUiUt!N4IJBLLld70&L`{(pY5>`3?O>t4ui_$vm1
z7S-2gWwhGjNn4J^9`Yrrn?FyJ5@8noso*)l^#vZSq<3fvYh%M=*2ZdBNtimZ*p2ar
zn@DcxjYs%RTJ)(=*G;7XvYDCf=^<lHbX3}D6?q>QOlVxm;$vX<4S#kqISTxjS}Z;N
zbi3uwyhnBIuNYk1=@lqT<eWUQOsSmw%ubtMGcyJ7w$1sQNkEo`(DfJkos=sdeP0``
z_qT)4xUsCfyA(z~XC-PJjva0SE)DZw{cj;AV`qy7;tACJe$Hff>pZSM>=5gRgC$Ko
z$*#ONOcEAE&E+{FQ_11a5XnmR=C9N|<^<C`FAalEhJ|YkBM+p)hD0A|qE{Z_UC7#k
zPg==Zj6K+{pRv4UTv|eR*dAR-&-6KnemytH;l4R_J=~(>@~iFO$wV;DOY+W>8&!E@
z`fbkUBzjvc<I(k-_N$@sHdo$m-Y_0#v`KoK1(v|pc2v;<n=Sol10GqLY{Ng|FDc|D
zk`D(%q?Xsde1Ky4eZkSH-*71}JR{=y)xTHCVkqPsRz9py*>d`7>?)`D(=UPfmr7A@
zC+6q7Pc~=)*BpuGmyCcpO?YvMY)u7&dC}G;YIySvQlGRT#~jX(&a0BC7zna~3Zh8W
z%t30AbzuQT(7M<RH%V;}_;v?<sA1&r_!SZ)PLo|+UK}IE0818Zn3{45xxGcGoNY+9
z<QfH8csmFos!&|;&?>Ckbn>WeK6vSrFIW6^>Xb;O1f-FP3nU{vACgK)!;@CiTD0Z;
zX_=%nk252Y)>i(z@_i)CqM)K8QaO|7WIdAAvF9}}J2$tqy!_|7598I=Gb}N01Z8Lv
zDKcT|0HW^U!7m$2TGUjU7K5hIzwGz)P`Wo6Ybc*EXu$!V+#Km#%?4;C<q2u%@l!8d
zO9Gpl`PIvkbib-j$@gOeE-`qL`3A(qpy@5>o!mYAyF+P|s5nd&gN62}DIi1vaU&2e
z&}OM6SWKQUq*7yLVo>9)@X#;822MfVSW(q(5Uj?i{rzl05<Px=CtRw#lt(c9pU8IT
z&^koO&ryb24olKfk*3auQ7%x^^ba71_7H@0albQyt#&@k<h7NO<IuoPG^qVVT(F_R
z#q&EJR!Db-7#1QKwBZ2h3o-fgz2Q5U@vz?_-pcozS?DQ_Y_PJ3t7(D=)-tIm8_Lju
z{l1#Gh>;z~&02ZroQz2B08n++;&8<U68n$r*@V4vlowbiNTVkfCnbdNsD;RiLG%${
z#+TB;*d<$9oGcSZ$E>lVWXJ}xwp*1c6dE+ItdAsXN<aWmW+5MPy<#R7Nrs@OFW4D_
zK&`@TX!M_YAhMAZz>x<?y`p~q{>|$}hxflPJ6#BJ2o1?5gP!HVL_838zU$<DHKj4j
zNR_Cn4Cr)d9!*L%0d~=@PmL7y-zTQwJFY0Yu-__p?+6z6$`>ioYBdQ{1>HP2t6G$^
z_D9V5*#C+#^7QH4vr@HKRMj#k`Vmmi2-#LIqbAgmi^s)H--0kQj2c3fmBVTs^`M-Z
z+G3p0F<FJd9hlSd!n-c+Hd%+^hiaGo$%x>VL}g_QGCmzE4sEd%ShIaA%tKo%Ej{mc
z;Hm%c_?GJNz6xi12etvNQ?#vH<;D&-nrFV_4K@;&R*8HuVwbYl%r*6P8P1jV*GwCk
zyhn4@k6HZ+Pll1{r+XfLY1xZr!v=4gO=~3;pXf2%r<kc<=e1j>(#yTL5*xE`SNhI7
z{WRvTV<JhG_Dilh#m{Y0CR^r<reUnlsYqzt#{H;@G)5L6ZAJ7+SU4y~iW0_1;=eJP
z4p$RiqNU=;mLQe%h(qGLRY?X_$V58GQWwC{bNzeJ?qk!or#H;N7!OUaOlFk-FzH;S
z7e2MyUtB7eAYqs&CTF4Buaw|S6L?^csaPB?wJl#<M^{cvfwN7;ewq=&BtMITd@h7W
z6)Bl4Z!KUL8q{<2!tGgRPhcS*5Ya4Zg9?zX%B_)LN^1J=LccPgqMmDL=qu}nqwA(r
z&nXkW`@?NM7U<t8#)%5-=q;6rhBTd!=83n8-A3vXS}w5s&eJC0>EGOGW$g0nKV`|x
zy-fqa-20?be?@v_^%?R14nmn+ULFGv<Tr2LI6FJrA6CAH*6I=w5r4G(l-FDo49;hR
z?NWpZ0l?@^?ZS^P`-3Rwe=~NUC7bd9lTG9?7zaPUJaGJKLlzD3%dHBg_w!r62T^w9
z9v!K)w6y5dn7*VcpfYIo5hSg`Z|(7?Wc;kegjoV!8<grbzCE50`47@NueJ+Jh%$+h
zli*9gmT#WViN`@f!aA7TpPV2&zyIKCyR4K}l(6YU6fqJ=gPKsr1&XBDk;s9ao;7<)
zij6BoI^-m*@Lqy9#3k){IoriD+Y%H!pO!5z9};k5a{0P$zv-8KEiY~}1H578*#orS
zl}2V`;v$ihToK+ZP-$^(sAD=4k33~g81Rq{TGWF0k~4ly`RC8em*knUWgA;DA{2lg
zm!iZbN>*Iq1d@*qA*N~(-hBqXTWF9>xW<v-axqI^G&=`ObKEU1Ztls+-(wSh_yBjG
zp5>&@^F1$k-hvW}rQ?8O(U!J6pFGIlN`oC0&MdDn{Xa5opMGlhQ)Ig-z1#mB^vK9=
zi@+mW0u%u`t;Ws>70ZwQ{W7KBxDz$}Khs;ky7K?TQ|D{qP;+Bamu_Lv^?9h4@Q3HX
z!pj*$>z15=PWt<KyzWH^K@jH}QMdBG;qUT4FH)}LIG5eyB((5f=u=Y{(^ZtQi@wuL
z+gW#JDQdB4D8h<NPTp>*uLr|d46MV$?Z=c@wW(3;(XG`93U`s37)_0t;K0A0dF`Y`
zM58__sevnYr21+p!%UuS+kA?O%4&#l3P0C#clwFpcfS+<<%}vaHZn0xMO^pAu1<h<
zNEqG@)aNL(U-bHI&@o8=QDmK*#KMPlkD$rEG|D(?rt?cY$|PSBJ&C9tV4Fm}eBw8w
zGNQ_Sqwj5yK&`h|-kgc>DEkLha%k2%H!zRg{aoX7Q2xE+(F^Nzl33qeYd&F8)Q4#5
z1D2F&bU(S<yingtGT#nYrdFmG6T#%P&vd&ET`BmTyMKC2zr=L1{gRq7PXk<fDlX>K
zZ-q|?DrUa!h<4m4|E@F)?>Smk;<KXZ<chYTk@?seS&5uQN><Lm&dnnsl%7|ahbdi;
zsYsoMK28PCN1{?^mooLTHX14rI#5OqCDw(3h#_)D6${BJP~)9Lv!oaRl$>5AG}W6X
z5;AnK{6GaBw*x;P#<4T3Bpmqo{Cv~wg@UKH{0$N`ImB7)+V`c9YE+ls>3QoSQH(yS
zqs$(<zMy^{z$u=EMmsF=M34N}T?qaj{KeEi*2dYq^}or`^e*Wb(2ggkJKvJ|MFGd=
z>EK&Z0BHKhVC~t0TG0R=9id8UqOjLxwZ#Py7+v!_|G_9P&)>gY=Yz;o=8m%(vQxmv
zI}^2zNS3V_yNW1*6lH%g!Vos3F{Skc__gMbBDX)Ey7Iv6nUK)j)iuQ_FmTh+qjJIJ
zjiDW83z}1&M}wgwvhWAbiPzaWTndya<T(mc)3bwM$;F{%5s(5X<XzQ=m}+f}6QM*4
z`a>mCav?D@yO31*4F`TUNB%4{#0V3N@DO$=((E(gSXia1uGVm#Br698N5IU!IS5N$
zQ^NsG&H?)P#sy+pW;VE~1r3n;qJ}OE0X|Uk4;erp0f<;*OFm6PWqfkt(ugQ!9=Vc)
zLMM<QG#vk(D}pI;d}qj_mXCxX+&LvHPgQ#pk{v-cmpyy<%8aUDlT+SPC7V2R=*tHU
zcv@P1Z(~vBPj67svU?^Rr@6Tt|E!9$%ryi{BBFD!^Ghm!V!pc+eygE@eRg(MIQ20o
z&*|@KWMKk*#Dnj2Qnr{peUIeM*YmXhbHw}Xo&|U8GRa$*{JT!hK;My9Jhfx+)yA@?
zWY8jDc$Z65F(JJ=VgTm#{U{isEq<DmlqRvWDQ{`5pY$;Sp6o`lE1eT7F;Ve({fn3K
zm)*X95gD2(1>ZgpU5y~}iJ(25zqL43OflI}X?nP%{&*#TX^XCIUPRc%09nUdW_l~o
zhSPKEsZ9TS(lp2V6LV53)xTyh_30rVJp_5&{U;Z$GUo_|$VpOEA&!&sR_#Tbq2(uT
z>9i$VQm-45{T6x(M#faAxo$fBumMXURcQ0W(u{@{bC{Izv0*sbz46B>iHrl@-M0|X
zy$>P1r^3$By<c8ay*@Ab&2}A)H0N=6@iNQM#ax_6@W*vF)5!gW=o4mzk9m^!5i(_!
z;!{L$Y$_gMTU^VnUPKn@SZpkFu^USYv?CNtBqrC7y1VP$`*$yv4VQbBc>30^nVM&A
zUVT!tYN~Oiy&5~BNz3$lzwp?i7AKaj)USl3fHjD9Vg6iiA)~C+j*^d9;vzW_1{LGr
zbGc`GXR8}LU|x(t^_;BLP4sgyu`}1q2E7(=0u;L7+SO!c&m>b3B7_`t9{pwGWtY2)
zU|G-m2$f~OcwK+LdUq8Ev%o|Okq!OH6?ogvLKBFbe<rT&3gFkp4t&|<qY9}vHVKZg
zfyR`QOo=<2;fYaEmmC6IgWL6t@mZwnL-6~0*78;!IIZZyuk{;j)<)+HBSsXFXw?fj
z2|)%0Au(}5W##kSaX@M~4(`@UP?Dz}6|qgX36Ww57Nf}?BA3SshhlMK(PYb?jn<8y
z9ypCt1Q}0lE0%uyCb_8|g_^rFu|^fNvjbBP`<!12lI2vj-ox6imPa->;Uy(^9A*!t
z@-Z<0fJ})Y9fZ$NeFi=J!>lGTa5F`vOHIP>^Vn{WrcXh+;UGG7#er*QJzw=*aBCr9
z1E!;6Dr$zCGK$jtqO)%WGQoc&S;+bE{t=6%_5M}y2b!Ob{0xi?Kb<^OvW0--Vr_ok
z4w4C5cf7OrTI3@^LnE1uwbHN`c(duGD<B}C>PY{C2RU~}njzDyXD@j)wA;9Djh!RR
zASXYBXVPg(Dj#`M?0*q;j?r~?+uv@}urV5Q$F|kjwr$&JlE$`eHfW57J9Zjd4H}#8
z{XFNK|GPe9jC+iH$sYS&S!>SUysj_8XndD9jm1g8pA8v`(!|N>vb?XsLKIO*g@l{6
zf*RT9g3KS%)frNg#bNB>B}tY@)o0jWYL_i2C~9NJEsQPN0FBg=7BOPAh6H2w#mscq
z2x#R2a}Pt!^3u|k>FL>|Tcai_rq;N}QsDmhYzb$c%bzWY;S$sApJw4gK%{s<rjFTx
zW1+`aN#T^hXS~}>Bsf5>)2yhc_Dv5uL<pE0S!4sU0@?xq#a_foj5A`1>&+J$keeQj
zSnWhm^`XEQs%y@U|F?^@Lod1Qz_rVHgqZT*vpd>jwxOwN;nI0gNyVp*RXd6*purPW
zN@c!TuRDohVzVgh(~;qP<tb}qhCRJ%!||>iy6M26?=PbX%ZiT~@fE0b2V`bRmj{&K
z{dvDNxELm7Vl$PlVYS#w)sfY@hvwqckkwK1ml97&$*`(TM7!pLcSFF#HGK6}c~Mi#
zIDv$ayu`k>{%oncax{%{`G$RH@jGcGT}8=q|M!B|HaA=tI&M}xh8cCm!F`W!>|!0~
zVQXPuHgTPY(%P5Dt%I!b*RFehhz`9SolTb4+}f<Sb51y!#MexxYq=Pdg;o&y7JNmw
z!HN^Ec4gb+ZvsBY+rtbIlIr)J=ctOx@zG`^>na<sw=MtKI+1}2an#E4LfI8D(av@<
z^gz8t&0Xq(@w2`9M<?gH-2hV#%(}zo$W`u~BXaB&pPAKH9*`88mmX=REQ7gW4g*#d
zLtC;;ZTCkH*L1a()4g7-shvwqW_Iq8iw7=|T{Yi_SPzvw+MVC48@=H7!$Z&QEhRQR
zlS4J$cr`q9sR-jd7)oyZGm!Z~KXzuQpBeq+mA9{Uav4(CQiC0NhsS$-Mw{04N;!w~
zBb`WpS9GqR2p#VB?^TD=+RDs}7rurGAu%%ozb*?)M++Pr9x%Kgdgq}?54h!h_XGhk
zn;hF=n}D~@;&w%?^$Qn2MsaN|ZU`|@GMoVpwp_uaq^2(1m1G%!k&z-#B5DwBkU>`J
zb3RoTY1SAkABQ|g+U`?~PV4s2=1%PWCK4c<7LAV4=j!>Ddw8#`j<!Wg<WF=cE)F*l
z#`tint))MiL*y$+DSv-)VG?JA4LxIOecAeDKgIWDN^lUlI{;ZC4!^LZ<h#zsoBZh5
zC|s!dhz;&rU>t#l9cA7AV`I+)8DUFQI}5Tl19;$+uf)tu)L1G~1_FUla3poo`tAl{
zuGyUSlZ6lL6co4X3mH9uUC55p3J=evx2~+LJQEAElq17HSf&mz$<FW6I0x{7{QL+Q
z34Id|Qq4E+d3z>$-VZsC2(XLmLBHfQ?7&#CidG)Z3+?XSp%zk7^3iRm4cfW4SsYF`
z8|&Ew{6fpIKXv_YfBvI0&wIN(ck=40KGqjAQNiP_?fBTCSJ2cHU2xscCOgmUc*RBZ
zxYc20YfYX#3p?b}`PzJsl8KxYE>Tf+adGiUFn`#{<m3x4byf8a?fZkruu+D`nj^iQ
zUXJA>KBuk+q8sh+q&V3vP|UHbNn-~#0e=2`!S+=eGdoskb47|NNM0b1GDG%q{u>F;
zN~dR(boc;i_7NU&j`Srs;04Tu;mCteHpgw>?oXN22+S!#gcw_adYPN;^{1A%y(p&k
zD*!Kq@=LY;>IizAfV7_<V&&7Z1KE%Nt%ut|nik9d92KI9N}xu^pyJ+DL0N7y%Bz86
zClXeG<AyeY!}gb@&Np`e&>O>B&^VWDi4<n_>&mNf>i0uyhX0i(=Di-Q@0Uc=@Aotv
zx<9<o2X>gaVD<FsUU>apj(N*{4sfsu*f|*(1Ny01J~;TI(O@bYfB7PF3x^K3)S2O{
z!aFp85v${#fxdfBx<-glhSz@(sEb6NrG1=~B3pQfC_j3Fjjd^IW8T6jH|5;N3?+qx
zVyP@Ev^Z=0yK}8|CfT@5p?`ui%K!$#d=`LxLlsmqr0DB@P24(vQ&&_5UH<yycvqc5
z*7(5fS;O|Na@4k`J^Qk3F1l3ZhO8f}>)hCYxkdL?htI=uN~~scT3Ln+yM*^-Z;Dc(
zvvxomT<e_4nR30i+0{OPhnG{#qA28N_61L)uH=60gUj_bmRs}>&jeiVa1Glb@j$*h
z0j#*tAx|GI^q(AztC6G2Scrvsd$hN<g~;CwV|>0$bq<nik;qd?9S91;HZ0QtDH3qz
zZJOel`cYgJ6%tW@e?L`ajw^)>p9zYrwW(T!A<y8zJpQt6Xc1VU!A1p(;FjI@bwi@{
z?Z6{{<#hgOwBm;d4W5uoJF!48L;r-v{Pxvn>rok;y!2oAq1>gQx;VVLuFn!2(PIZ6
zI>K~!@HdPF?|jp@<{{e#U@NDlB9GBHQb+joRR`j2!4_CeK5U7-)D43rOL`xT>$y0x
zq-ip?3c}ohag6Uq>EPoN{-_V3lX*Jn$X3fQl2I<E0`~qu18y?=uhyxW`c_Guj@}~R
zn18vDh+`#+HM~RB(as$^%!n&_{Coe(aPNDrpE$BN+Z{MM`kBLad8pn*;?H0Hev9|T
zL(hk82_RR$x7OwY^zMM5*T@08z3*zrg}3g~2r81;G(yg~-+P#D%Lz|=hencQ8TT@0
z#8@hMqAbY>4f}1(ylP4?$OQ;ivNgLp0zxbS-$Md?Ugz`_f|~J(@2$_ZjF%+=4?e~&
ztZXf{hJ%Qlj#K=b;t5|sGuVsz9Z>+AcRJt+lHcTdo>#b9Yk4H;`A%`Y6D{o+v|ZP4
z)q)yDY1!ruQ+8$5{j|WCG*MD<&HFf<b54?-;JIjbe$cV`D)8=oueauP2;<1D3M$~{
zFmQ%SXd_<zU3W0j#kf;xRWiH_BFM8~;F1$@fp?FK3S0y?!o)HGL&3ktYPXQzF7CNq
zjxpA0JASO%w(=8N|AY(j;}M990t~2G^s&`u398?J)Pmrl!m7gO_`BvG{z@Vt3#(gN
z(QJ#pqj2i0grR2c9TRA%D5*;`<dCru8S`*S=$Wv>AQ2Wf6o$+Uv%xLLZE6hZvWXo=
zVphHgj8@X5=SJ<wi~#jOq`1|H{>w2tXq$icgMV%hWBvbscNp9bjCJq5i4Gy-redHG
zQI{F<fpQXY%zu%V_h#QvH!G_b?fskJ$O9$c@vGo?i8hIe*Jda^Rf<K9Rq!+a2o3Fa
zlmUjSMsG=4COH-iw5-*gv}#Qa+xKj;jKtyHZ+S&DiYY8%@sY!=GQ<01SD*JVgv7t4
zqwwjq2hXHW@>b!iVve=yvP42?i-QP0iiC}hDoexj;^UjnF86Nup^=i7IwYB~-u$%B
zIoeL$WyHkh4vyZVq>mh-iBEuZl5%X_QJ(ZhUsKePc8iDx({4LTFR6a(N0Mic%Z~1`
zZQvt(*fW5$@A@KeLlQwMv2k(pqHQeIU_erL3;tP)w|2YV2t;j$&2FPGH4@1#E8=`J
z@7mP~SDzom?B-mm^Wruxq0EQC5tdkappH{y-nJ|F1-^%luDZ6`+7o<kHwd{o%ZduH
zf$l((&wYGbIA5tyD%aF5)_I}2A71}Ti@w`yYi+$CrfFGEDuc%dxnJ-z8Bv-2-J*g{
z)V+6!e|_CvHkk%=TCzh?AP~NVEQiTxd(sN=VD_frmvmPI|5)sgY~}z60XX=$J6%Mr
zNq4F0(l**ZqWJVXM&%a&<_F~GT~*?WN_{Gu)#tG7!Y>;OUR6OTV5T|r$$$e(snz`|
zDRF$1`N7MNFW{b-v4sz1WPuKmMK{xB#7dYJb6KR2ke)YGQgZ1XsYnLr3`gj@_3V4G
zBE!v@Rai>*mGGfvbyG%51U^gMZRZLK8~=2B5bpblefPIn5bxzMKG4*nEoPyXmKlYc
zgn~!^7zm;TU@4iFbbZs0Mj<5hp9jRWnibK`%M|z`Q>Th5IZiq7j%-6Up4=>90ZDCH
zRby~cnF1-9l@&k9@ZmMSfOCi<StL{RAWJk<Xz1kSe!vT;Od*dfNB(jvxWmhSiV)~M
zR;VV@>ofzs{)}9l{m_Xt)LfkY*e%}8E@Y|t#`9kv+uJo7tQV|DAplmxSj2%@cC9h+
z(&nyT<8z*NM%vVFj2uj)DW5VH_1r^BqG=fTQ@5;p(tG&r>HzXc1s4$lq>j<8HX*?8
zr(fISPTBSP+WF&MOyHrH*UHL{6g78$TY{)xOLwSV1vl?Ktj_s<v0K`#9^jIO<o=F-
z709j(R;3tANz)UiKr<bS)nl7aMIWUy{mF;NX%JMmBw;4U4|D+vzruwwmD|~Z5;Q6j
zS)$pZVcGwe1&Bs$yleO?J0-ir$Y5@3n*O`ly|7PSPKDL)^%K626VU|we)aJ&F?M$E
z5&h8qtx>r=8^UZGMtg$DZvsHB0n}h;+=4xMf=jKill+>J-KJU7J~+T>)Ee~mCs183
zTV-~l;k4vb7@~AQdST!mS4x3}G15{|;3j4jl|(J=k>Id$O_Qs8lCPW@XPU&%hxu~n
zajp`ytT}7NQ#T$uV*l4+q}$M~c8>bbHKa%L)LpX?8tJW!+AByX!c*7SdqtL#|0;a+
zf$K)T@9yvSn+zQHt7p69R30k(%Oa{-sQXh8YU8P)fJ0<!3hj3D>aRogG%T#Ff{MKF
zl?@nC4XY#b%{%ca8l8*}nW9CC`a;UB8M(^gqT{2H*VpFA^(tjsmteJngZzNc3_WYr
z?bxvCMzteS*CwkRt=3^UF<EtI2W^zLH#fi5vZ*2Yb4ya~vB*XcK0QtgMJGC&uiHh(
z$pA4T@Y&HfFws3T$HDFv_UV$*q?1;LRt%7L%e2neIr`WrzW=)hKFMUqweMK%g6N&W
z^ZdCu4@~;DVgzg>q#=R5y-PNAt$}o7Yd>Da?K{qL5b$w!%8&(n*Iov1=RZ$~!%m1Z
zJZZAqyy)I8{%w`cxe4e<YD7NYHnxAa_KHEb@OCWcu%@kCYPb3XmIn-U;XMb<jXS6>
zorrP%)M%UH`S0UyP)zZ$314tmBWaG@VW^btPUD~rR3HvSb9?8B@z(H;?pz&dE>|Ef
z7QbXc-c|<TMeH?zy9AKQ_sv3-z1I9Qd}+krmw_-1SyhLI^V1&n?>e2}#r;8Q_~`Ef
z((Sd|mlz3CHr40HqL(T8IQir9@l)0a#pEjo65)5b{Vi7fhr%u|uX+X+08#&Pi3eiY
zyr{MriJvc0M6H5csF4hIWh}gsFl2qOnYLy3cx4p2E^cN<NuMu|I5pf63?m3&Ux=%v
z|Cn*_WiaOg*2-4@MF#z~cZlHNPs!@)r11H~BOl-*7S`8oH(CLvW&mVa{{-UV=J+SU
z_VXaVADRyyuBMPuL7R^`n<d?Cu>l>=RlpY@jhHN0lxJ|x=bjkT9|xAi($?10-aS>7
zX3!)~AbP@_GDkjZ$9dTyMwMkGSkT+h2?5C1<}DM+X`>zeCCF)wlJ<Oi<Uief#xFQx
zw)b0g*mi5Tbg(n(4|NMMFT=@$1IB$R8Zy{~8_;7#D=KS9*f_(0?9%<&b46u^*={Jk
z?BwKbGwpn8T7%gfbWlBp!NY0~Ft_07zJ<SARP<6e<sZ3a8-E#L1&kHE%Xn$w*l^nJ
z*}ikvt~s%{bxuCcI0-yIGG0vTI8Z8Tj{pI74*9HJAN!7!ry&aIPagp+Dz~F^AWdFy
zW}-BOF>0>Y6E$wuv7%|p^LbC+bw4S#c5do{_pBbkN2uQNJ%^2{vEzIDvy0E~p8R*U
zWzOS{&ZSq2VYMM)!lr{78AC<yn6%Cc+Zzzv!$tNbZ}*-vg3`tS>Z}kO_TH#m6%IbK
zvbL5S7j98u6J$|oZsRo%l3HKQv1q>$!+KNDAK7qSU{a5U-7RKZJ_bEXOlpW-oh^kK
zJrlf3KeXFa-##hDf^WR3kY@%4{}wS$ecDkr=d<GZZ4YMt9o$%d#7X0B|KBsn<vX5w
zo7lh2iYNmlk4}Rh7Igzl>010wmc!K_xOi<o=XVP<`u#qon_tWt`ElheNk(REjWH?>
zb=-hAmuCBYy<@_3!Ql1F+M|%s=LW;+XgxL6^w7fvs+VpJwWely{82p1f`St!fmn6z
zK~+T~(*eFPz+03XdVM(IZcz}gReLdxbIm70QeMS6)mV+7tplm12H%07wRLKU6NbDB
z+-zt$a6_rV);31#D`o*@ri~Zltgs!yl6Ht9A_r@+F&SV1zZHB528+r1N`?t1vF~0l
zy7GlJ?NA<jRI^+8T6b6E)moq4?{y&JEHcQ)5H98bAmC-%=U_`@BZ4cye3W}m81R@)
zEqB^_%VNE`7fYT>=95%mZJFQQm?I+SckJeHa_hP5HL`HS&-2UMkKi5W?VzgG;IUqT
z`Ff$pYaL32Tx*BGK)-*@i)b=Yhq^X0qi!gwdZN+_H%va~tAUxKf!3fI9w#dkf1lxI
zo5nf=dFk+2qhu%+o`7>_^E}=}G3+6@ZPq|4X(CL}@bGQ=!sd=@mB2e5@eEN{&3i#d
zJ8=a1!R$*|bvzpz*@$d1lnAQ#kVzBAv)0M=befdX>XPi#h>2=S+U%1KXM!}D6wqKB
z!GMX)?mMA3=EyeW<C4%3L9B5OKWftg-F@C`@Gd0~d$<a2T28{x&NlX@`gK$QCq5Mi
z`aHox&J-_{$b7{Ox^84kc}$nCEWc9@QrUgL<AsBiUh=wS=kg`@-OuY0K8dt<U-x7+
z>HqYw=6=&T`mlsOCGZU4;s1b)EfA3V3ZuW(#y^!PZHCfo<;*BWwL;(?8R^T{a>xzv
zp_w30S9@ScqSI0oG1AX5+&b3RE}sJi)_hUIcJU-($8`y#8f1h)DU<P&rcT_tKza{t
zge%~+?R^3mf_DJ2^K{F)v9|U?lifFVU*OG{KAKAPG37TDCbmt}g%$)eT|9OsA40u;
z5{Zsca00Np4^axf_*#ihDalHw*M;RQ&jx(q>-M`_y}`o`z{HY`=i71_xAz__5V5$7
zfQ$44Xxi_{3CW-#?=U7;#6hhA7yy+1&{F4@VFHfbx%C`2z3z0}A_n*kr-F{@_4Q6b
zqQAVds-&_V_tQ5L1h#%dPuFZ$4h37=5LSyx;v<|D?M$k=>Y|4F;L1v#yEC8Al}eE5
za{m21HO&dGm8GSruXUy)+Yn%aY;`>`-|#!Bxj{tOvD6$fdB5Wy#nrwLY)k9j)_?gZ
z=+Lom=Q^KO-&U2gEE>7|NJW?zmoXwGPEm%9Ox|k0h^U(OIIT=wX>%M-t}H7zHLXvV
zfFN(IK^`eXlMywZ<n~Zr@g8v6<>86H-_{c_W$Q3&y^y6wC)2F9j>s{zWT^Xx{3FD-
z<WJlrP_W#TUGc*5uyACK5Stj*#E_ppMpMVu_jc(d(vKbMY1cHZY<L{pgFO@M5SHM=
zLS7EKYuo6~C&ziOtxwjT_!Mw*6F&OtbE?EZ$bCet_r(9~EG7Qw#-pQxj@JDAoB4|~
zg;UqJ(>`Ny9$wUHttJjmB_2EIM!{R0Sy`STVRLA4Jvs)qS?)9AB|Oto=si{%)2hP%
zuB*BbZF?oLf7U*NBt;5U8ZlKu`F#WqTu?*%99&md<S!?lFJMxst$SKy2^0+Q7X%t9
zW-av<Gc#U_)rt>vx>QM7w1&+6DH_9j2Zc*{?+?gJ-30zmDy`e0Be^=1q+^jeNoL{T
zdbF)*6I1FAELfJ*?fe!EZq70lR231CQSO*nnndZu<Pa@~I%mnC4;*;ox#TKp+{>TM
z6{*ah{l&_GU$`cQjl8C%3XF_2*##SqwrSIOj#;=gEz0U-v9$R4C4m!<$uVK6z@(9*
z_|LDScB6bcRCz0b^i0txuvB-KWad)`1Kg1a4HmCkK2r+sg!)H<gMZt-RP}#`_pxT5
zA>*kf`h{dyQfm2&K^?_H&}it&tl}x;&Aj>g6G`OSlniP2$94EiHy(ERL|IjHoaVJg
zOi5}JdL6uPyCUY@aW#-#58VsvUi`(wS@TbK6ZGGT?x6w_Yj<n}rma5$m(|$WlBe=p
zs?t39E;n{TsPV}0y)@Or>`q3!N?H_mA4Q&%oHp)afArOoOW5q8x9xW_t%xu^ZkYUa
zAtKhqFtl1h%bn(*5Pi8(aetS#q{oz=G!UPB(!h4G(Ri|fi=5|PnUjz=5|E35nI6S8
z4oY=N)03R*^^<HE=9ZZP+s`=fU63FF?SdJ%_g<a4F43O3iTOm((2G8V6YHFFB0fI$
zVD`b+G`;Z?J-LB0b`*I=vK}u+a&dmWdh8LCAUGjne5C6?=l(HbfFfJcL|P4TCfl0O
z`wulP@|uX79HL;yJ&C`|`_%Ho!O0SV`;hqapVLzc+yu|dIveU-4y@($2^)u(7^{7=
z)8)3Xp&^l_S=%-5leD(yGa{##Di=0fv$4Cs!wrQrrdsBvd#^<Im!c3YKIt`C8w2~e
zVr6r*<WKS>T0e0S^B(3AbzMhN*8CP*G6cFpdkYd+z0p(5GAh2Py<92;Y<K(k_rXsI
z#1O;cTU+n^L=4!y<=n{c=txo|>r*)M3w(bZ^h-yR;`@lDT{DcXz%SzC0|N!+wRYQ2
z#NshY*L=Pt?x8Tn`;6@QN!RId!IZlpV~#1fZv9Q|$(=)4No%O^J3JL_Mjw`aTLwFp
z!6@J=6z0O%+E4op^q2d3ODX)XPYhHHHh=X&S-2GcgnHDJ8u8@ot&CzY(>6l2;7mqC
z2=j9OYjeu$h*H#?qG=eev7(dC`ffYR+UV)@n%DLgN*OPvi3;mDhnP2%kP`!5?th8Y
z<(rg$d2^C3l{E^`<KE}cY_}9x=HMsE`Nr9DCZ6nT>uCe?_f@{zZ57Lk&olh$Xu(n}
z0U-(XFeELuNY1plOhqGoQdJI8M+(yh8NW22thX~1hmD<JfRjg<B_m&zbSFw|I(;r<
zhiZ^_MS^cnq+t(;x)G|=>4<{E!}?*e!Vh?<=*LO1vC4VwozX7Ita_c?PnX2pa%1{L
z0WHLVs)mbZUGKq<BHjfU#xMU}DREHNZDRjEq_7&!Q2=faD%xNN|0It>#j>x7$ydP)
zX(0PGd$Gyl4QCT(G2%AeNMP9CZ03YSP(d8msgOE%EkF#4?<qXG*HKpsfQgrc<}ZBq
z*hHM-hjBTeQew^eJMbyd%<P9LnlOpylT``yPBf09_EYa6VZ^VYm|<vibhAdw6O5yy
zZ@ofdwk4v=mxt72qh3$1(uw2t8vb`kFcEv!OlDLgEzDS_D^PZJqtxVaU$kYU*uU0C
zg&`$0EnVzQZAr$ZkV+(DEdL5*xqD~@;i1Esy@(3qOzkVr+ERmYq6l<A>ZL2Q!`S#j
z)l@3#Jmi=(hYV&`k1z^z)HBJ6uO9YKMuqyl&#>Gnnxo^R#vm#sCXam_8J0i`pS=B%
zy`&Jdf_uFwUG7hhe~c!xex@&SbbERyp0=7x`@v{GTw`@dbvQgkCs|S#7mFHfeyF!Z
z$Xb?WdF(ad;Ik!`(-Gql5Xku$gPPFuB<RT-0O~zK|D+qQB-uO}E?4DaAcdI-YJ!es
z>E|Qpn2e=-yi7JFW;Iz+Rc9~|2V8swj+$*+(zB}zy=GXo#9TaqUHS2vl6~l~2^kqN
zt0&~S;f*%;0)(%-M8Hf%u?;940axGqjyifkv)dFE$O@$MquPpJ#3G|0QI1q=hf{7@
zJSBrV#khTY2}(=$VHNoAVEUNJ*U2N}RWv0fB#;MFO0rc{hDuZ2)%ASgsu*=c>16PE
z=0hf|Nf}!LmmAGh40IIO(0XkfoJaQ4=4N(>&T%Yy`@VN}pVir}c4vsP2(94(01hLz
zOjrpJ<#;IBSm4X#?vimmT8t<YHez}-w6X_wjQEwavy(!!)9XSSNJgrGrSWm&KH%3s
z;YlFCxp9ldR7jk^a100C5Bx(0Dn)6KuY_!cp%q`|(zTU(Ll+m5;>ki(GFfr7LZcYE
zAPtygsW3{s4-m{b=T$WLV>Z{`;n}nau&TR*Ch+tNn!Ayt5~ojYY;9}-W${~>QCZaS
zaR%TxzX<;)XY!8*1|YM9i;+I~3LPKgO&_y0{cOIvnW5m};e8yOdJE%!zQ245<s6<E
zDX6&o5b*Nae{+A29xLAVaz)#E+A!97Jy|!mx~ZUHL8-T{_os<x(x5pTpa54^R{ltr
z9oQiO&5%9o@GN)xQ08@ZhmKxVn8GcyHM?KW<-LS=zW5$Y)|U0O1fQ<y8W*4l6=5D0
zpnB`LyNoTQp&&v_QIPq+6cJsAg`Jfh8KU30!WR+~Ticp4(t$0TcRA%mS{}Vbi*haL
z%3q+Na75<AgtEwRf1@Z!rn59Wt{QR@2BZgNMAesaV-w(-8)g2InSdFgt1?TQljxoh
zR+GZ53G)Ejh)JR{Aly<{Yn*k09Mi+SAI3)z#il_KGpJ|{u@Q#6{gX=2?K}^A>>AUX
z8}f@^8NctFMc;5Umza?b6<Q7i6zz=D{CB+zipV8al>f7`KN2DX9W6A8$gY5sEX!z*
zC!5FD$TVrFHd8HBj_n_y(Hrl<Da)UREWM7kU%jL_?so4eVHFJmW{J7Ee^{cPv+?<y
z(C<GW5&C+tNGIkOiEzUz;)boe<Iw?kx%HCyufBpjyd<1h4)WwRpR8P6shmfTRLSNa
zT4yWiuvJCpD<<bo)G!13AVNo`2;JO(^T-!nlX%&$8fm}Q1ywqiF2RiZd#s7Wg*E-Z
zW;n>XQxjcxcSrfk&}{Rg2_j@C>#1n&2aQJo|1~Q1=v0ZhILCg_gMeMyDVwezE8@AJ
z$oo}!;%QzC+4ddujJMu-o3y(=t`@fyogRBIIo85OFrz^KjCsl7oT%$gQ|QbdX8fOH
z(Oi~bM>gkUH#+Yo6#~rz%%r=#Ye`pc?A8P74~4xbGh5G5t#f7r+@r8XM4h0ml(^~C
zm(QFQ!^zgB+VWNCnZm>whB+NXjpe;*;XUW=iFB9qfAY8kj(ssV_$HC<Ir4?ZS~T+R
z9N*C(gGk;w@3{4~-#^c2A_$>;#HpGX8r`Q|a(n-@cZ*bC)f}0jp({_7xpyh}29)my
zJ6CYB`F$|8N_8H#75s(gwnZ$PaCufbTF)2nix%(zv@DMCnux70zK=U$sU_NBI#w4Z
zG|Uq<19k70Q%c4<6dP?^lC-PV-+-Szk2S8u1&akE{71*S*!dPL03+i%EfFn3b)|rx
ztE$QnaDlF&m!QhijtJ<J@p7UrlBfI|)~u|n9xGpD(%J*hO-=F{mrQHt?Kwb-?@}L4
z^}|P7On6!b&5_@B4y+z`wse^Ah2_4FdY->Q-OuYp7gYuu{bMKk!YGh@0z}AyH#+U@
z9r8G+af7>^7i)9X0gq6+Me@KHb;7s8LJ0fkItSK(w`Ke#kHDFk-jR?pI?`|uMG}CS
z45xzSd)+g!UAw@Q(AFMyHU&T>;H^K9n-6z-uD7=?tv_=J931j1E{Wj}3B@g#08<Vl
zngUrxUClP{<~s?pz{7{l*NuBA3UYvl|F~0QuotsvV6xgwFrLL3H^$f#WUzM4s=xW;
z+mOfva0{!`Z2E%__^#V-cGv;UZ(^=Jn19=BczL8<L*M9H$AOiLYnZ-et?v`EK%mXM
zl!(Zu``2R)zME+WtBuAK{H)N8H-X5qDWNo_py*Ty|972k_d`zXxbKgzdASybjv?J|
zLJVDZF2mZSVR!Z<Tmr)<CwImkTn}qywWbdDumN6YW|1<Zt$&+WEzgfjE+u&f2gPjH
zS`%I;CVXFGc>w_1o&<E;{!Zgj92{hy6C3y~(Tt6Ti9t(pj0N?x>{q5p!^r{A5H=1$
zae(R&{+~YW1fhdu>8nxN5gBjlv?1<>x~%LQ1ik|{HfU0_!NJqD)#2nmUr=J`#-#q0
z3qD3lG85k!sa_d5TC=m^e4GP7mMk=86X%Dc7H3Hnofe-tMj*PI)A{!z^v_+0R5Zo(
zKYNV@04-dqp7Y#n(lYAx&FB{RrYY;B0)*^JD~eU9in%zu^eeV3S*Wk9s;MH^mp`#u
zbRS}$neuc;UQ^fDtI;X()t=h!4vko9kE-X$9)D=>*j~`s4#f*UwH|mF%M&&u*Y&Lj
zt;OPh>*U+}Zn2%w72VH|xw4va6)Il_`bbGZ0j0&pL;XpJ`%qLE)vO@ixo4TikgL`&
zQyo4EOX01jUg^N22k?RAm%;eP@GGj17JNjvZvm1pEdDh72bWZ31TneZu_`5)KUbub
zFh5goC{%q3A7o2T9+O@q*HYiHa_2QQ?Nn$#-3_A|5fGMFvsbcVF&ddPEVde>Zpm=D
zQ|*3vO_5Zxv^w4$G3Xim8uuIx=3wlOm!*yXpBGj2oao(WNfU4pc4Le93Cwn*LY%~+
z`TyLF20#87nTFLhN-CclLy!gqz)NLI@prV6ZSzlDVQ)l=?5(Z$al;#j<1Tfods6m2
zITbosrq^!xqllJJW*~esjGvf#`(7oC_UZa0o!f!|Y8gi~3Aa>zfoa|=U3MgR2CumG
zm$051)Re<g5OJV*QAxW$5~dGF?Tk@~X?94lSmaQ>O;UU*DTV^@ecyGI!Fz1ue&DmG
zQ>X1q2aT{n$VS94V}~1Ylr{!z629GxYvUfDaG&?Fr_;IjiO3fwLPzxupMJNPS!mDd
zp9rI-rWLVaav4ck+`W&PnVabwc))IE!@KkhQlt{)Q@eiPpSEJt^&T0Nt;&udCUeq6
z<dUWMnn5arVr6Y#+Ew&7$8P76D@+^d8^Rn7?a}cu$ZPGBaH<)_+L_7gX!4s%5OLvx
zCG3N5XP4C{{k-P^v~rg87;#JnO#rL!K=ZnjEe~uhAH*f}T8AiyN5?5JViV0<cC)Za
zIcbxsYFm?^j-}oIF3aegpw_k@)3}P0#tB?W{4?z7^$5Vm&_2T)Sqe&@%&0>Y5Kq1A
zTR}rVFE@HHe)#q^9!uXhVcqBMa-q%_u5nP9E_bhvvCc%jAOFew8Ha$ttZ&!sHj|xg
zJoo#EDdb04ar6Siz|tY{$j{DjVZw9<?IDx79Kd$SB1N&ra8np4&=^=)*;t+vz%g0y
z5tIOHbdK%DN;{(86sYBb8P97*Q+q}V01z?hb-%i&21k$6g@RD&j*e7i^G5F0eSnCh
zL*b7ffnBRDEH2haMMWaX3*&|(ojkU&2u<GncYQ2RSx80;@Ou~-Xt0o%c#TY}JF@i>
z;h~*>uA?MP%?#%}u?bE|IB`J*3DbLDoE?`Gyt*Yte>j&<Nyz>-<gkPr=wE8Qj^CsD
zW*E*^^&RdWlAGLYT<;H<UqC)HJ(Qm`OOh;zbV8I4pNzpuva-YrppE{!1e(heW&hnF
z_7SEnDe!blXmq+@f<9ay_Ge{b;3QqCQ7hB;88e${P%D{p^j|Q!#${4xXMVHl`-F3+
z61zQ&kq?5Q!!(mB0fSJ*(#g>%ajWLP!38-JYSR~&(MGT-7Fq*zMxT3{il!ilboMAy
zeG-%3%CWBLmJS-TIu&S%`sb#~3~1cU^l2cq9VWM)?N4t9D~5{r3tRat9?$${bkieb
zXwu$v3-@&F{=`gPqhNzWdGcjtlQ7}Y?w{KGTnt=9*4Cz#RbPZ6$fytB`rxHs#?QmT
z2$c`+Hb4AR1Hpga_LjcH9U4L+9%Biu=6fOSDq9;eZ@E|_^0*vbd6kq>9PQ6}3%ePQ
zwl!ZIH*vf)o^}7LGCDW7nNx{2;f>&gE%2_?7nYsmfv@3wSu{k`my;1+mvBiTMwlC)
zA?g1BW771Qu1k69@J!m+Mf9?b{ZzMXw`DVYNAh_(>{(C@V-Vfc_Lw<MlI#y5QNQ!9
zMR~S*=#ZtT2wK+i_#>8{-0m5lc(!~40W_O3gdAU}Z#BN5zda$Af&UkdD!3w{xJ|4<
zb9T}}bl1CP6(v@zWmX}U++wDl55dclTxcY|Yj}lZ;Y~8oox=!jElL?C=0<v7u(EqX
zqUrDy1nxm{Fn3C-%HsXikGieS>|Op?KM}7{e|n!H!QjTEdh!i_^9$p0Ek>*w<n!7&
zCmjY}2CO=*eyFk3AUY+^SLdAssy<7i!Ve8(^)`#{NCeiZ#f!Yp%9(+wsVOetfPk>P
zja$7ZZy(AYI$HE-I@nR4KlRDG)6*z%uWuxs`iu#ZkSMXYp`U8h!{|m@!9%Dh@v$jE
z^OjA4vY-S=A;Wp@An=Kpcz$<t=vZAV1SDe7uKi8~Bvy^Sl({$$Yn26BdU`%!G-O_+
z^7HF5wJ>WmUy!b(DVtecRPgX%4Ddd-LIXd|kp#<Id_{rO*3tfRG|wuqZ8mTOfZbJ;
zlqO^Hx(U;4vU$&6cz8YjdY5_R69@6HPmX7>dhI9#7!R;d#+^Np)Jw)6rKnYN1wgGf
z+aBx-@=p$OZuUaisBhm#qai$$X9xs~vJL-f4a=X36RI-WJ*fi?_GkB)l0=y=7N(i7
z-6MejFS51mfOWm&ZVYFG#l3!fqOQ{j&F?y&42!q+SavK(DTPL(iSM3#LJ|^K<bzP6
z{X%PROy)Y!tL|LXwDSGLVCvByOiWA=^;CYO_6^tYk_QS##Q47w6T~70WyH-nf;YIl
zYeGA1VVu~COG<V#FVJJaRFaD`JY1qmN+f>4{mB~1bMtFL6>=aIbP>$k5QvYWHK{Xi
zgfx|vZv}^qk^fzzg2+m$0-II;d-vJ6C`J+)IA@03$B7)}$T+do>q0-lB+Id9j*R(G
ze@l|)Il7fb8{1WaRFw$(YGm|ru9~Gi$s@bch-!?B=bmPURZAT~vGTLps478Ev1CMp
zY7E%q^&m$454D!?P_p(-CRyU-SJ5V>)7fbsO(_RNWZL~NDo6ESza}VbUlbSfyq#C5
zd~5d9mGTw?5r+h!Bqz77Q=^FOYbyJ#Q&1#_<mi`?ft6bb7>AAP(N#>r3To2grN5<Q
zR3Mjc7xP-3`4WpYXe3;g{}#*L7;|^x1($ONQ;qd-RFXYY8a2ulb)T)e{(KENYN;)>
z*fo%E+c>zDd!OclE=rivb0Q6RUCqMKK1Gn_;!e5!+#EgaH?liCD|P>x)Qy~s2Pt5N
zPBlH;1QvVSOFD`Rcz~f*Ut{@S79dG0fBT>VDpi+|68HDTY(EbNANkxewo>sPST?al
zF_JF#;Uo(0Mv~P+G2IS~Cuclw6F!KJZ?__$=0;H3Co3+VnE=xFhE8$LJvy?|^&MPb
zoBzGz!JCliG&@Vz3Z`VrlKVq2h`_C93Hh-~i2^s)c*Y6`+jZyGdlSb%+>{K&!owq^
zq|Sr5-?sPJ$rdvM{Qkt(!Qn87j=;ikHe}h5&$$8zErM{<l-S&p5qY$*w|{tbNcr|v
z!vZETdjJcTDlu}m;d_H`5EFVJpvXHI7tAu0&Ce%}3$(zO5c;CjMWChFj0_C{loTK@
zi=A`9^1t@+mCh7mKAJAg0aS^kDndM-+;E^CivlJ}Up5uS5M$773q4$X`yM<odaWH#
zLP{EZ?tA~-M8H2vMD)o2!QF>v7!5dOe2RgoUz)W_?+WY(F4yBDyQ5=_`}JVL!^|p1
zi}!9n0O-}|an69$f?wL+o~g+u?xchY(2<5t5HD7qZU;)p&<PoX@&O%)fQ;L%Yy&GD
zfH&X8P@q06%~E%Hf4yfe>rNev!n*vkt8g{#@SfKF_7ONKK?f(q-nnzwE)8#|_5&z8
z_tOoqNB6zoi+~_jZF{(t+2ARw=2Dj(FTn7|BMTa?|In55*R*(b`VR}&$_hjfbQK24
zp1U%y521ZrT<>Jj9(VOXi7TD>r$-L{j%NGP+FIzSi6FXT&zqy4)jqEdz*~NNd}3*D
zXR>@F%~lvLh%FS1*TeWLx>CB+pD~SoyriNCS;CwL2#+i^ME>ALz@~g59t^iUb#0?8
zZyc=u18H3FLA@)YHMy8LfF?%*DS-rh(b1)e2=Vp-vQwZCiPydA+fm7*?V4x$(eEP2
z7<FR`Jfj*+8+S}@b$aa_OnKjs%75z$zz<(p#iakc1?VG-lL~wD7~9=i+0804UdOzL
zh7n>ewW*-;P-i63z+lHIP~^P$i$MVItFpE9Wqq<r<c!i%rxtYuc9dMImMnq6s46+y
z=!bJPVLCTEW;YaY&d!^9m~)AQ;sB6Q1r6!edcaqz2A!&sldBYyX;BrQ+Bj+!ulx=x
z@PG0l)UF&gV&hrPsxFs&Ts#lBU#gfBScWM@^QjQSNO{%uSnm*ghV>>GDx}Mn=aNRs
zxVFitVZAt4c2r6+eJ6`;dSlGca-p7j#Eel`lOx*`p0`feXsnwh3!=*Bv`H{VyIgu+
z{DhBtb{O#f(&M{ZS~s`0-aD41-)G5$#w}*hkOY-M44Ym9jQ~xLzS|pjI@ZgQHB3o|
z=55+6Ax4-LAO1>&&~=JSVDOj{IEight^%GCpq+Cf{`NA(5=?}qE^7Wq_C``lcm(pQ
ze)0AS$PFK0<j)LUaDyXo5%%`Xjef&L$04E7PEgkPQ(s!f>))TFLeYzn&rwq`fZ5l=
zMhuB5TLNdc{hghi!$tcocbr6Pay^wn3=!c2k`@qon-Y8+%R<T+M*42llL!Pgdi-qn
zu~!4_a;nK{sA+E59k$#snVCoDCq|3S6Kn)rfy>#!%oE=lEwJGI1a>00!ZNkfwdMkS
zjH`^(QRHWl`|7X4Sdb(r#!?vy8mOW|(O~>?v3tV@F49GH&ixtm{RkUAkdu?IU43V4
zFgwBgT^I2Ej}tiN!|SWx+VgeZ&Ml3wY^oW}sPyAZ+}RF%%SPlBk*a!1Qg~rta^{7}
zJJ5}_Z`)UQzPj?<nGp<hEbP>6vnHoy&=qmI#L)M#<Nq)@!#r;O>D7<FspTiL&zly$
zHIJm58cr>oV9W$PU5_DooZu}&oZuyf1aCWVpP=9!>z&}#i`zUg8LQQ!<FSAPVj;Rd
zw_>okpN?YbHdfk$71ka3k@+7!l&C~zFq=@YK-jn7rHu{tc{NU3i2(wUr8XaAu|bsw
zfWRv}IIn!wfdBC7-)y@+I*!9?WoKiudPiVtX1;^}=OB}P%#*hzJ|nLUDp)|Mc0I(0
z>DSLiy{Y)Q2z6RO2^t@4AL%leAy|cJqRt`syc|b8+;m+s<(rx43d6bSPgx`)!ZuzB
zHfkZ2LbKD;<9i#|=F>;MKDG%TNX+hV#W7Dp?b82e*FT#ffvkhMAD{xUe9z!!`w7YI
znIHV<Vv!UcicJJ>)UsN?I>)s(3sLydKy<8pOwK!0(nbT2f?>Z_Ye&LgFZnHrJto^0
z3Q@xOA<a$LH-LE^mm`nPPR(yewW1MS6HOFIQDT(XU3hveA%u#^so~0l{&U6$iU7M(
z(uq>maCZOcUmy=zhD0L_eAAC!<>-r^3fQ7pUUuC6KD`9#b4S%7IZ{0D?91I}sEDd)
z-*R;gmsa`e?|Gxd<*C@lYUT%!H|$FiO{uR_+YL}Qa>zdIH%p2#%Y&(g!%D?7EEYdw
zU{Eb!7b%f%AKcPJO_kZAA<@c6$Cq@jRm|$vr#DOt!=Ccn7B-mUsy*NJW^<qCvKd{W
zpu<Vxy99c9De6PV#3Z@6cM|h@o$jtSlpNqvR4M01t5?A`^$DBt3NeAK-~4L5DiTjg
z_a0b06ERFqeov{Wx(c){{$@Jg?l|&#G(&pZqYTljRk3#%<4DT2DQp#p>dU}ktNmN@
z`s)HFw!`fnvwQoJsmF+9^WlZl@~=RjN*&+C+W~)>?xyJsYfW?f<Du#g!y{E=<_zAt
zfMiD|@6lU9*(Qi;rf=cD;7KkfsSZ1%**H3#J{;{OuL=f!@I8acpBv%3mZ6_|<@v@R
zz;lb0JkxrqwikDaRS-`%lYx+w{1uZTNhvt^6#)Se9sW_1Dd1yH*E#|suNG6!x1MUh
z@AVt%>NT1}z{L?Z<OL%wh2%4wEwp8MhP;ghk=|>6GREGny8;R<WV*V3=p1{jIw{jJ
zJ~%#iNChu*`C?RNIPoR6Pg>Lj-`9t-P%O%@Nu}`tf`L3ri?wW)dv7Q;V`8-T<4PPG
z;020fsxcR^&~v1Y+qfpUlBq1Owl@UpIG`znSXnE3+l(YSDyN8j4DJnAui9QKIdldS
zc>FLKTiw$X?IU4yv9ZR%MlPfuD{>(lnVaev%4yLZOuSBY*&a+wNXeVJv(s^E7gtE1
z5(NoR4@-7=NdKb>4iqehWt7^<JgF^ftH>Y?L+&D__y~Akp`f7B>D9$0a*`angYt`C
z0k01vH1rcTp*ouHh-mby&t2n9_F!@f#`P=FdW=usL|o51KcVLqXOo-j)h|H3r(S0<
znoQk!vlAuR5Yk&-<d9E<hUU<Dhl~8)v&qcE(R)Q6Wtq+&1u!+5|1R|d=QxqVf)7<1
zwT6FJe*VLvm`y}0Okje^($Q@4WNtj16e^lw-EnO=+sA4<YgaIC$>8#s5pHlfA52K;
zZ|S);(4@yaYdaKp^5KD%AYHo|QA6P2{}cbml7AqQ9R4!I%sEoJ^?akFZChp0<R|b^
z=#uRS5x(#C0|3`=`!bj=t$(SsGH~W6vSSNMBN;btsDC?n1AWAE2{Q5wQMy<v%Rk3Q
zWyQ`E&^f8uFT#*QV_zMRahYT3*CXXgjip+yhe~T;3*dlJ8+5qGVYj6VnBDMyYlU;A
zsjM~Ay|skM>!ePzP^90@+_Kr{CJh$5q(%CDUzBkji11CetBrs6T!9)b7SIyNqj^wk
zga%l7WR0Yt(9!KnP09W^tb8xcM+hU$lVpn>iT=NP0~qD!TH(J>qd0~v)a)3%yxTqR
zfP6lTGEWsZq6<0bLyNzX`K@*90q%AVp&1i(nsNPza|qD0C>#Qt_H`94nVWtSR|T1|
zvyb-3mo+GJRf|?a8qH!55amRI6>PT?X?~`+(_^b9TBW2Y>cIb+Eny1i9$(dgY_?U3
zrN_ujHUXz;%93{}%b)F$X-Uh9fg&yEEHS_uiuvRgQY1KfdW!n~eZD2i?iGL0!FL4<
zeG3|6Y(!{|Z9vgcda@)niTkZvb@(jU-yky}?P19YXhe|i?IABFVieRl5cu79_WfFQ
zsNZWr-dVfC)SK#Z?DGzyQtT=Y4^>Deh6M>}gybkRvo7)YYwt=oh|bN3XU1||y_`&4
z8st4w<n<i8tZn?3y(t#eTJKc$e^^y8l$%3`Blo&45#hW0RT?K6|B=k};itv`)#sZb
zGQ!6z8slxQEKYXxnC_0WmCY!GHIFLHu+Hr1@EFTIuhfkju|3ahk*rPPOXgHlG?au8
zlaHyY6`bDy#jg2exnZXn9(@N#f{vO3i#-`Gl9Rf~mqW&M`ObGLQsWOwMTe{{{$WLO
zK2CcWjX%8*q^4Dr_6#je(-o;??L$*y?~1S|)cF!1p^mh@y>&eI>S5r%2T_KDgh&;&
zZNcdX%=V*u31LKrLS}^Y6Z(7!WN`1;UTxn^>3PyJ8V^|ixbEH#J&yksq@d-Ed3C5O
ztGR`R1>Equs45V{lCYMNU(&`WN6B)-m9>?Jr^~ab=PK*5Z!PEMmh0;xa28ml9w21L
z7MQwBk-U>;EhhdtJ2{@~L-cM%BvA+J*O8(sfLyA}upn}}{9bMjd${hh)k7Aup`w_}
zxGlNY#xIXZT2?0l{9chDg@_5`x~|P>BSFI~Hj4sd2NPN0;vkFzti+rYIW;jH;c*2J
zX1{g4A=@9Ia%nU+*(K}AM=Y~;eVymBYG(jso4aMl5;&lre3pLNz<gK#W?Pj)tDXaP
zAbf;3kBR{XN{?-<0abR7n1uo(AP{E^xQzPKq&;ci&XCu0q@pTYWxUx-fkXg6wFTeF
zH=oZu9uETig6sTmIrIZw)lwM+(;8{ZwpQ*b2)alP7`o!wbZ}Q&>`eYPP7zzW-nh0X
zW#>te;TJUfGnC%YwTBI9J)g3g9$;SVxJ)%(|D8}(li#;#@lCZdQ(10#7nsy>C#hm=
zOWvVwE7pmwZoT0d3iG*qf*7|c-?i~r8=0(eP~rE(wV252uX5;?n?67M-qSTw-FB_n
z==}wg0JzJHprA$LuYeV`1?nwB+LxoSZS98mnZq~O2;!!q*wK2c*b{T?_Z<(JN*=!>
zG<&!d`vo06GHhU@57Sl-*s~|ro*Q|~LF>n5!83MeYA(q*31JfU5K$)<n|q`B+@F_G
zaR%q23Of8<o2Zoe94|t6gUc>e`oDaZlNfzV`M(d&(AWslY~$A7>zS9j@P5~Rmtj-`
zX`BZyeCA#@Z2Z<=kW3IgNFjr2{WGbME`<UQ9}Qc|G=ZGf&VnrG+n?%2=F(ME?YLiC
z`Np|T+T_fsmF%9K#8HDA^drhgtU~G4x9O3g8M4?h(mQsyG&ns5TYp==2&K~0r%T6+
zfkqQ?Dac2ZRZ>z*5da=<NCC;dK@Jk#0iSvv-me77&N(c^z+GtyX*B7M{9S)2gI2MF
zUmv$aeyvrY*)n;pUhdO}g(-BM_3#2k292+YI2-lBhnTs3qsmD8h+gA>N~=r3ZO+9o
z*^=w>Y{RH5VwH8s<Lnt)y=q@C2w|&Ausx*4q3P<_0BLsNfDAX*uh`&oGza{3_l;Jz
z-8$D(tRXX=rNEj3-CeI4)l~8sKlhsJL_a=rw_JnvK;qffFB`^qu9(EM_t{QT@Bq4f
zCia=d=`_de9@7%Oq!E#%6eM@Vj4=g91Zjjh8ZIfWl0<+7yucK;c+nR<vcB#i@jF}F
z-Rrz}giFhizdjwqP2#Ai;pC9I?XJeZg-g{{XV-kULPTb*-As8;*$DrH1|b|wRFbkV
z2Tdl3Crm2PvBYwwhho5?%oQueXMsc0&Lm7m4>K3wN7zHc#rh?c(PGLa<fSctdXsSr
zu}~)|qt0lCyky(DpYcfEu3gXAApUv%_|*R72;*0bjA+E1B{h(Uk*o%R`&;1IVq>i-
zVY|SnZ#yX$JxJExpPJ#2X~vS6y$v>~yweCaX`e9Qs>=a24vZ2@J`hWpD@*)=qA?@;
z)D}>=10&pG7-9ga){#ddow+f!<Lrsq#heo*LIMXjC%-BnU6W7V#Dx_gLjte0v~;&5
z6E>(PKY4symPZGV1ht^qoWjaV-P`*KcJ=S6jkVd#j!T9*cT87@d!o4lCAAa)z@nio
zYdom5dAv@FOF;0#`G-3n5Yl`HU?qvF)N6pw>$%0maSEL-SKdagyEHsJ_JtU%xqKOm
zp0VkVG_yl`Bk-Nw%W>Mh<H%qYm%|7e5(h_v(VTfGm|xfHXrb6*ztMIOH@5kw^J@&J
zQoIReZanuAe`uRWxPrbPJ9F<qc3aePvll<E;nR!D$mkfD*81W(2ncqmjpt|yi8(P8
zIEE<b`jNM_L5<gB#g3Fq)GZpZ>EL)Mpddu&$54jLK$NJIZoxnO_<~`SEPCbAFm0J{
zvguFv_&xdHkcsP*j5MRz*qYkMzz(dfKit<2x+AhQU9u~#=E~C~8$BI`(D!qJ*N(wF
zFGuuqjP~t_>DDAvBZrE<!!PORg305Q^w<Aro&y{j@5q4q>%Z%fixf8K|6-Z0(<i<5
z8nS7tH8(IFe>psN;2ze`^P6t`IHxAQ=kS{9d&AkkCjOe9Hf^USA62$u$y1<~TiV<y
zV=?%Zo+Zvk@uP<|I5OMXxlk!YQ?{PJkV{LQS#6#A=bYt@s7eZ&DbKoTBWoIb_p=_?
zA%5$t`=C7m?>b&>8V;_q=<-9Wvo83aSf*r4S#b5Q<L!bXzskWuR?3B5B!Ez!ddFC5
zCb;_x)T@6bCTl9Ke{1un>8syO$dC;o$fJ=_Qit|<S3QWWgS)w-DMN*oo!qbRE-Wdf
zCjBJvto69bLVYLn&zVA;G*UB0`~yS+I^!@U=sZ>Z1OJ4)cUBJBPTBjm?!2WGV;u>7
zW8hmCp1dbCS)vo4{uFgf$8P)PZWe6LMMws_4W@>H-xsjKpIcA9HJ|gYGC#huvE)n|
z>q&zk3^~W`$M1pTjS6<pf~am^9xY7f4!4ZLSIlhEwUQOqEBDuayx@NTUI)*YOR(UL
z(>wR@o|&0qYSNJ7bhzcH7^9?Uq7x9jI!dO#i3U`MisR|7A(CXnd2bgs2U9swO6dJY
z5zI4llGEp;r&+{*?u{F?QEIF#UEy@m^=HZGRX)1_jr)?eqhE0Pz`U;_7lFeDJ^~M^
zpPvLa9`3>E;C?#D#8MVh94tQKUW|?)@gtqycjpQ3@iESSa;figCeIU7W_3uBK(T0|
zhdTp&IMo!qL}@U&Osex_kd`KE6sjzAv+agl4uvciPQLvk)WEJ1omB=Tc>UDV=2Oa|
zUW7$o{|}eFC4oV7bGfgEYwV(QnA@s_A{e5gN@N+bTbEuasPdB2a%V82LR(j3b*5))
zZ7(;Jf;;GWFJyoLz`^f23t&0o-yt|VUniKthV}8i-Q*ovIvfIAy^6XD^K~0M&x>A|
z|3}n0hS%9`U%ye4##Uq7X49CBZQEvJ+qTm<jh*b+wrw_P^xn^T{^z{wL#}Io+sL}t
zTyu;$erc)-Vz#;jI5-+DAk+?^D_B4^JGT@|1(X9He8Ue*Izt`7W=2iLiSbGU8IcS-
zb~T-0z53sfMF6tVa+3>O0GWcC)|m8}uE7`t5?ZWyI=>ai%F0S30e^&CsVFj31YAA!
z<dm4%ROzoNNwJJC(PakC&hQN*!wacZ`&G|(_QnS+esU0|k?C9fw~%7O6l%H-%s8LZ
z;SMK6FkRR#HYB~KV8~fdnWTtmN|Os34wwZ52DZ090ZPaA=gQ>WS3F23%&u+hu0_=t
z0v;gw`j|a$4)>b<O5f5JF@QhTo~-XT{#vX1pko_SEYT`U9Hc+IIHFN@bWjrqY-A)h
z4ncw6vZ!(b_2@VeI;;_M=ClUuHJ-&mA9s3EstHia4Zfbjy57Hpi<69}XH2ZMuYEbY
zO-nJOm?#+3UqqH88y+3$WoHVY$H_0@N|k2h3ccAuG(2R>yA*c6{9jx0wE<aF;eYlg
zM9p6b*bOV_f#|T*%`e`y7HHaT*{CjmKTYyTqp(}$Vwm&uoHVdy{73s&JT04>#52B?
z4^q<U9A6H@;HJdtWv@u1{`EK@ikt+OTrOr-?in_uS(rCRt-#!%u~OcfC$h1+wxiY^
zz;2aSEypdMw3m@4uc)v2YgTMrKW^(w^$2QFJp&Zfb&s8#&*s6w^>sgsU^n#QFEqdi
zPLyr0+&_o`&e2|<<DGS$Kqis%9gEHQ@opOm`WnVHjWsouBg11~0j%baNV)5)I9?$N
zCSyR`D$*-P_Rmruq>D?|wL_6}87(G+7s2{SBrrgv9H6VHp?n>&I(CX>rFemn6OjN3
ztar;)mX-CSBoCz$Y)Kh=haC=eNiUy~?Y_K7QA89B|9z;*vCpI%8|PNJ`<B<%hI<D+
zpXtMRszz7J#Ghy`ggIBq0t0Edf0UY-t>oF~3EZ&Nf2jhc@gKKnAkR5h*B3fHorHq~
z-Guj;Z+5);D3@mkqoAwnJzMjh&<qmik-d4EWuKy^HqD$*=-8YRYzi{Q36oUhbpV2(
zz}EloDC#}9ND*Xu4E*|}P}L(mL)*UMfN_vHiR5>yLN3Z)5h8y&OX^9NpJc?tUr6e%
z+r1U}2nrvj!RTTP+uC@R8?3O#RI5$B`=qtmb|1vb%I)-*xkGfLQ1Hx&DS3f=x`p*>
zS0YGQ92t@xoFoAq0t_`g&C&LoP)cC%QI@pJhV{6GD?ATEeh=4N>5&mIhQw@sa&cAA
zRuEB`9cV6#sq1$@Zu_)<FtphiWOSe`_X<u2kuQ&&tvcp$^sC35rQYL?;?-A>qR{_N
zt*WS~D5_iFgDx;oc$m_rhWoL2)v^h|3x|ms;qlqW?NTdo)qZ>e80qeOF?7OOUbKJS
zcWoKEj!9T9IDXr%MD!XPyXw$4v6|(vr;UtF4Cqha%#NL<X{yRGZJX@8dH_=!Hvp&t
z2Ie0PcCgrlPKOPtn^8hSakKH^GaA(MbEBIpb7ws&b-Mn$hM7XfS>C}g%H}!{Jf^s7
zFAZmNaq;9W@cUw2l8=Ctb4*BD94RJ?s_+dzYUIl!Ay1!{*6%`;?NwR^l5~md^|P<#
zzpV5h*BemYpSTJGFUWiXWhNvGr{h8Sh1qmAX(0a+VRJTI0-t+H22CHZ$Vsj6P2=sq
zXEoyUX(uO_<g~01GiLz0z3$`}Ck_sQ!CcD4ylS$dO5Bj8Qk?4_syY^Y3|7msUzMc|
zHM!@ziXa~p$VPbh#{`2W;xkr&{hU_iDphLAWXxv3q!~_{keMxH<OJ8c3K<d>8aMQF
zga{S(@5o%x^zy`0pb+9cq<V`x4uKz`se-g?HA?qd^yYa_E;e~VL!yRi4CSFP<uJT^
zAK3zA`hVriAzUG3<^Q`-IPx4jOZ^gQ*FllapDuE@w1VjVXe85c<#@e@wm-mNsZ@oF
zipo;QXC@zc^(FT?a23|AuKzXhnL`JUhpbjT$vjD}G%8b5k-1#t`-%n^pN6ACp~7O_
zO46M@S5oeTsR@26T#lDqxD_`Z5e!%M^j1wv&z1LA+ismhb%DbuYWm<0ALQzG)x!r)
z0Ko!@$xvIsF5`<CK15l{k&MXe?{y`5*9SL$zJ)}q)>rB`QGj6z|2h4|?^lumU)T8V
z4gE4hyNq0G_Z->ejtkQDc8*`3Xo4+4^HgWH$H)GC#y<uoKqY@JPIbJ@V3(+mDnp4Q
zd)`00{a`AfD&RDKy8Qmtu^!xG5N$N{rMp37?}ZrVste{7=<dC3=#jDU&<9YFEagOt
zk&;NV&`y4gwlB_EHUYo>rsvtOZY#V`*&rAgy3{L?Y4f<^yST3VXS78WNU*)Fm(fSY
zAow5<Kb}T+@a~!&`3WxN_T64A*>H0an!`h)*^JQ)Y*bL-PYF{~o+Vv3dY(?OFT>-t
zM(iipU-5hO<@#6FWHn8}M%05k)~)$$kuJlLf!&p*Z75f`puq79GjI^JI8lTA#X~Zv
zf7u!nPs3wBV>~K`OH#Ul6_bEiDYXaL6%E1D!|RM=w=QiK?L50LV)Q*Z*ZBQz)~Oa<
zhSXvfB^GifOe~55Vy?d71eng??fiM813fafI7=uI7te|1)g~OT+2;_uXLN-aK%~!t
z8%~J|WL(b9&OXnkc10~66}PtHbZ`BTlOxc3KILT6m;m-@bbkI08o?im#B@8KD+Ufb
z?`+zdmW0C+iuc>2QkW3o<qI5V*GAGYWPubSmsahL`&qqoMdh9qI4tBCku+96FrG^M
zZ1!lAu@scZ&*p-jNeh*#%mU24F*a*YH8p<G`c&UlT@kBi6l|Ykr*3ZBQcq4tOX^OT
zy37;)wrQs1x-qC`VgMF{!RMp+_#OSD93Q@q@x=T4dt&J?kU|u=(Rni`X)MKQjD9E5
zuYj&mR9c$K+#EZ4!akph8JBBOTv`eTNFW+_NBXrU)9c=8HoSindCgE(JfOUftn`Xc
zf>X_HEc@6>5hZHhi#cn>(2gfgs{Br%$Rxiv8XJu<^j_fV+fVm*b@B=UZRU<f0vi(`
z!nUhR04S{pVx97Ch%`<t&gmWXYkAe#UBEm$b1ahQJe=k@4C|wrRKd{-$7#p%+y;c@
zGJ?yhipwb@tcpU(BQ0;DkP_2(h;Sp7bdn^Q^uNgs*zaRZyEX>rfdUp1Lbi^Eu)-Ad
zl;n=$RoOzwG!|vFCk-~et>B!%Q5~X#>*6qEWyG+;NvxvOZKOP;|NlG&<NxL{<_64z
zUcISk0w(2(3DbJ>hKtnUb$xNFJ8rWbg9$}+@H^f!0D2<a`6_=p#oP%+zn+SUJj+2#
zG}#wVFASq(SbJ>v(J9Qx1Q^!zB=j^Qs()s~;Ha8Qi>XT(1ofUd`E^!wD^(r|VS9nF
zUG}AKF6{1H9V!3y0+1_hhkV^G`Dzjo@zf5razm$EMK){ODhi5sFw_85(6pRNJ>}*S
zCK^+^Ay`@f)G3S+H5~*@xPVW|$#Ka*xMyV*W^^Cx)Heg-b-(WMs?P_`Q}FoQ2McVk
z5Hav|=eao~5J0i8Z{DcMwi;`?ihzs4Mw0g;0g}GSmmu#fh*e_ry$U{gQVZ`!qo)>q
zztXZ-`5cfCwRTBFbcXT8@se9)S2BpqJ%PmHQKsrQ+aunpj7&=12uodFD77gqgB^@g
zFPnB@p|P>?u4}Y@{ZSkw%j|>*DatqnT8?=ft`bkX|DIl<i3W&uhn$TiYC;wC`fR48
zr!*#i$06o|30af-r8$)PGi$#sA;b<DGk0H4ZKOF}$|+>Yld;hE(gh=?v~UsdFkJfY
zlWZlE8Zv%ftca1xed(+?w!9Lv_&r-QF&w96HyUs!T+<T0+*ZiZB$|a_01M{F4@wRW
zSeBx$_f_i??Z@!1=BwSA5MFMm(PHL2Erg{`kf~!|zH#9h7`uOpmKz%gdI^Fopc@;m
zmoZi4z^YasX+6pj2Lkv{IMC#dU_!MtXU<&Cmz!R)UA9&b(~HRR7f#L1(_}OF;g-@e
z1FG28FUg#a#))Zb;_SI($nZc1k8$bAqGYJ~cKSc{1fR4VS^jYZc-;MJUatG|3F_Iy
z`)!gpji)n*9cx6UGA)8CV*ki^c41D|O^va^Y@)QY5MMIlA5L~nB{^<5YV4p!gC{!>
zy<O7X4Sn>DG1C=sW`R@2M+f34r#m9&-9Rjxv_-qn`gQFygLYdQJME-!!~l$y^Q*~*
zm~F=)N!4dsLp5FmH65{~Z6;3NzhY#a$%iHV^HWqN#$}n=?Y~JZ8}Y2y#i{TpJZ5tT
zfUJG+d?tuVg;QM9yVlwAAYo!%KO+by(<vjpHEdUL+__PD_kOwC2&u5A&NNwy^S?~(
z(1GI!A<Q@v*v=&Fk8BR_ubw2DC~hKs8ylj&2FwE8xdwV^c;Y+(%+c?U1SLO44F5J7
zqJnWM%}O0#UhYVYp<pubAT?9=4FttncV2(g{n>O#n0I;qXsBrC2~EEQvhygd(<VF&
zR!l||zA3|F!vs)ydQzTd@0@uE@?9nI${dHsqsf0m7y718I`)}kG9*1aJIm<Dk0a%i
zKjIVC`r!m=Wk-7ovFR7p*X0?TCpn(Jx>uYL6W@nTdTrm<e&lDQ{kuA{naAk=ebFI@
zhgen<s#2NPO(-S?58?E<BClkWmgD@_MVf^vX!;qeXNMG2^4zggfj)>VLhwiW^6Ob!
zT>()Rx*_XLgY{yZfp0f+NV>B+j|~jJEbH&`GNuH><YZ*{oFwA}XU8{pyvK9@TStML
z@bBCAi>7v;<tUX^vz}WfTv>=pD?m+L{&cz4-&9pW^1$d!mG63widygAhmc_iqhv;(
z8T&9570NkdyChJh;MCDaqwGfo6M1_0$l52cxS5rW)lpFm9F2xJh29fKRu+MX4L!)V
z?_2MrY?qe(<YUVsV9hBcXiK$Rd}MnH&LTj9R9gj67-}%0f0iwIs*;2;D;xX7k^LGo
zUVy}Hk|2eN3W%6`o{fRx*rmZ!62fo29?oyXS_}kwx(NIj)7km})3LFcpC2cor1k#L
z`Up2?mJ-d;ZITl~hWwt7e8$<4X-zU2LR-E2=LcWEvU9@b<es(*)V^|ZM1<v{whx9X
zd*GVw!SrtkI#idebfI-joYIY6x9%5+%fE}+BBdqA=O^OOcnp+Xec;L}LdZa>ZiB@b
zM@T5NWby2A+vL*|6=CPeo4RPYwLoNuYhU7*fNvC;hf0L){;L_9x2^DKXht1Z{8wc1
z*m-{XVO?EdZC-#{h$DmyNhaBM;b_>PNli;Htbp$O)K6%L0yVi+8}spc+6}-y`+$hL
zFj|K8v|{O$oA*Cmp_DnE8S!;rG-I3NCl8)heEzWd*+NKT(Gw+)BF7E^xtOnGsqciG
zTaL72Sd2wLj6U!V{(k=av`=YkcJ}+|gb;VPB-5@Szsq0m<F62jk`mwTa3K)hJ3Bos
zp`k&ovytt6D804))y*e#s_VswD&o7OC<25IJzWN+)0@lHF47M><AaFgMnt*#*N#Xk
z2KpU1H*Z7?%CFU(32(o$qhvv?JBM156mwPTO&O)R-XXrB<Gq1i3g}R5+1dgh6*U$b
z3)r~01?`>DPz^QPyJL)odx=wuDq5i`Ve$&fpl#!yH|`N_^(=jR{cK4q4%u3h>##_j
zJ7jM@oos-A1B(t)S}KepW)B@BKmg0AUl=_&0-#z%JwDFJqxW2<q(=mWM?DsN?qD0Y
zD1xPI!!vL#7xNG%O+-T*o&f^G7xs;z*0y|+7Y6=a`!<p3WUS2pEDxV!IV>q7GyrGg
z{jB|4*4lW2a;mq^5G-IWMmDbQ;hqFCuqG#r3raFTOQ7OzUI|=0CM)Utb4Oj4Dx2(F
z6iC0`TB0_+F|7nAcI0+MGkSz9DFt_ix_zT;B~9+DL*+sxf~2aVAq)%94F1WHNxWmc
z5@#$dJiQpwVxM0(nH!&85nm5KTJAKK>+#w35;79%mjOqy*^<Ex?&IVYX9^A-rn&WN
z1LBKJjNtf^AHCBAR*l(vF6W-p=MvyEczn%+16|#`=M%HZ8hLwXg=nY*QG(KW?F0(n
zHiEPfW11TaqM{4(;#HJE=$!t4lht<<!&cyJp`(5j)k+Y+`f!ZCGI;h;pS$wDgA=;F
z_Fu5Y-WPNk*B8U!t^Xl0nzG)hA7eB-s>TAakcGc02}37(M<0%EUhjC*&TRm^3OYF&
zDazMkD}5kxMV2CyM*&p<-T4)25FXyy;XW?4M3nYvWAAJ0b%HbwZf-$URRmCWUccsa
z>N_wAV47MwMghR*O|Lz1Zck&uXx(V5l=*zp>)p$208)ysqVz*q03kN^X6|Y$IBP19
zNWYo4<sGE8(smN-?ChFng&RiFnkAzvZ_ktc^y*8|Z-6e!tpKmM;N0fv`E;w+6=mF1
zuwUQ8Z%fgy1Q1*d>(U(L3W8~CIzp!2?^45JkO19XNLRsL3RN|7pf@n|d9t&6dofd>
zFQA{E?|#^b?{R=oK{<JH(zd>p4`51wahYA<F6K6*<i6e-fJhZC-*O3W2N`|BPa{}H
zA#@`Gq2OcT=Qx+AwzruU*c&9s=gvJG7@OC<zmA9`c{t&war&VQt?^uAVY2WDfF3W_
z93)iFRl_#}p#l3wcz-CO;_94~WPWO|c?xZ{Cnf$yAe}~6z&EA=fbRPHw~_b$VW!7Z
zc<ISnXR5S-JH#1-`jk>Ee@{FqJx~K5W;rd^Tdo-9cwK5Z3vqQi9n%uQfZwPm<>&jN
ztAK3l6t<Rs=KN4n+%B=Yx?_ts{&s8hWz}9&WR@ESGO)3kP~84+S#@>svX$A#P9aP+
z(b{9EMMb01c4kpWeSkWjXjyHVzCm1Xm)vN?YY{mu)B^(9=XUDfhZd<S;8F?SI3m0*
z4OfT$Qaecn@vfF<G=#C@)c^>y|E!TM_<#Og|J`!NCU9>Cg#7#hu7uUQr8jV`kH~cL
z2;54FRm^s(Ve%ZY*F5AnK!EkSL+GShH_h~h&2U}%yYa~WJw2HYCvMWkhF85_+MGoi
zk&a-ee`-Mhi7*3FIa!rPprT-Amyf}zT|11=BAs&$;l!m#Gw*a22U}!N2yvzeTU9>Y
zqCQGreE;;U;Xq?V<W>Q#7X_A1(On_omqr3F$po3DXaWGk-b!9M@>Buyx+3=QIlj=D
zjdO03nmOs`v)_(?M#A{OzD)oT3;+vp7>g#{o%^`gI1R$ZL!_O#MlLiW9p>Oo*@3~V
zGH;71Ia7C35(#Y|E+;t#Z>s<j$ApltS9P54Wq*ne9XQ5I4YzazxA;Jj8KTG~i3112
zZijUB{`QVhQ3*($t0tr$U$}?`Cj@$xnx=+WT~)eRZ*Mriu<<vZfg#K1HaP=@uygs-
z0S|SN@Qje)fybyi-vd<0Oc4zea(@1|^VPcSkX78S2faUWqbA&D4)FAu``_G<xK#{3
z`n&VXXoAL$dJXG{Dm^Yc|AbOi^p(>H&#|$b+<#$CoiKfR<h&w74JSHgGP~49*T^p|
z{VIgq3QT{s9xQ4-KiM?TEG)SE4PJ6_{-%oV>wXoBQc@F16-Jfexofs(zYG&fiA{*#
zYm!>4`76uovb9R5Awv}I@TTt(A3ipw&S1kcBvkAM>+{O*3y;_LPqe0-aS{XnM8dl?
znBK&=X%B8+!_d=<1AsXoLUQxNA8FdVB^WpZfvb`u&=X75@n8?!!JcV^KYl%&*_vKj
zi&jDR_*-tK(dbOA_dJ?<a(P-((FR+y)L$>pY;prxxR*H3F)k_*gOZL1O;?r?F?V5V
zZy(k{LRtoUU~&Ne*tl1}c6^Cn@Ml<e^c53xNeuwa#BQ7z8d|4q3C>KV*sPW4ak9Q0
z=@LF$hxqWi$*XDv>!ZcFz=$W@qhsyXWIzP`6s}L4pj3DO7a0kbR(4Q#*3^tKZ2nmd
zDu-{AFtdS#g~P@rDC*fWqaihNYb(!c$a+lSZ+Q(4O9(V4>nox_({4;s|6{O@N&E8H
zv~B-_eR8t@x{6vAVF=5D_U}sUHqfnxZvMYLl}?sue~ircC#wz~sXq`Y$62+wn4~gA
zv$gGp3!@jWO8utI5}oJ#+W*hjm-8#mqBy#*78TqL(YKN2xIz75KA;Lq<r+dL+6;yQ
z%efdCCQOx2qMvYvPmv#b#kQc%oqGhI@|3}$zah@wk057M;J(C@*|(mfdv2gJxRBuU
zLdNnLYX21JlgQriJR8X|M;y8*+TxRv%45=xva+O#PQgT;(O=_`pinrPr0h8Mu8^{+
z>1=2QXO|iyB-q~h1$}~08o!{Xu~dipM^6ZYLMb6Vts{EcUxq?hryoxlkE#ubNysNs
z|8C%)D>8@26Vpfn!4MkQwrW|$z?iXL&|MRHT7j#G7I+VjIM?n3UHtHX9<*M@4Tot5
zg{AAk@7f+~l=;yUN6~*z2Uj^Lr6T#s#FadQAz?6jY;SLnlHlk%A0o?X6N)a26Ke$1
za4O%Zv9U3J!W0=2EOstr<M7+u{7H@uNk4>9OO~7z9<FOm^6DweUuf}bOw5}nrLPut
zHXyxYql?qz^9`@G?3X{cH&cEHiVi`_Ny8E;{NlXnF9ax}Gw#1yVk;_uT?;1Ev-fzN
zT;m-LM2wjUzToV8=iE*4=B{So!otDIg$d!~H;)A_?U`6xqkoc}`++Au_oGYLykZn?
zEw{9$JJo55J2LnV3k4q1fhSvJj~PIviN_EOs$uJ&?21-tHbqpNY6L}mcYeITu3NJ;
zx&uWwHshD(^#I4wrwyI#vWfsvboJ@RpXz+QJB9QeX1Zvt9iuZTMu4^IAUQg#eTOgO
z&xN6-Wkd;J#b`7q_IdMJ#z)lDiiwjbQg3yH0?ILM=QE(_USj2A7IHZ8R$QTaM-^;i
zxXu*!llBmW0;Xuqugv{L&`FfWwKziG6Qpn%ys<DA7y=gQFz#@N!(H3N&|U!;abe*b
zSGmk*ou#em2^g?KH4%lL&<&b6DU)x_JC$WbC)a5AZQ-(rhRt4p-m~t`+qMxpb<#^?
zuB@8GN`Scc__SQh5=0acCAtR_kZf&?9@-oxLMAy_Whv!@SSutmgCk|`-Z8i=_VVR3
z?2HR%{3MfHT~>W7t%Vvp>|nRO>iYjxf&=>>K`V*l|9Q3l5HBaJ^%ai~1`SPK2VqIu
zh>JuLd<1r<Y#=*!e6Pf9NMm|Eq=Cd7J9nwqOrUUXf&97;E(8x%w5)2rPW`7kYG)Zt
zaamT?F<B<^$RT__N5gp06bnwjnP_3iFt0OAj?W@?cL6s2f|PXcg@8smb1r8WbiW&S
zw!FG|aY0qoDpW8!da41x>-@3-55m@4wF7eVSAkap;myu)F1+gl#*@b1upPBiLR&E7
zG;+8;+iVDRWFjL0)0v**OR+LzZ+b^(Q}CT2Pt<C8`pW9g7?o<lu^wHFfA<ygCsBo!
zx&sY{_IsA^z289;eNTbpL8W3dl~T(l3JT5=rnmP&DivqKQ89?MMlY^nXQ43MLeiz>
zwq*uv$;Gpm6pRsMP8P4ayjWR&0@-UhV$=I*eG(~m=WFe;U6-~XkhprAH3oYVhW5jx
zl}4Q-dqoE6anPne8GknZz9SBsjyO7GgfKZgJpA^fd+RR#(ria}y0a>YJZofFQy8ip
zOyxqv1O;<@UAzd*4+{Lf0NO1mxvx%YzbK6AkkhfrkDO;)^1|~VrE%4anlxMQXQfvf
zUElz?=e`RnWQ4SIog@dKCr`6sAf&121P<->pWGsc%lIQgK?M&-j1qo4!Ru^nfErPS
zdPu^&uM2aD&|+_Ud;~@3n$*q%%1>}1AOOefeVz9RiL<k9+X)3rE1%%|O{Z~<pP-tq
z3&tH5-R^cT1SG7tdJ|ppfr+)x4itg+H5iBBAG{lNcm-qtrq|fiL<|GI6V3$STnwDO
z@j~ZBDY7YQjyqc12(8p=#3Z5}FIU<5{)AF6bZmdGprHvHDpjxZ<_72rUYE}T1%py9
zF60Q&PF&3`33W3lx2-~Ck>3=0evl*%k#IZk;?wEP56nLyuW{HKpFh)U+q&G`*A=7d
zy&Eu;@kLHz`AJ&@Cu`0`U_@W}0e-YuPk7N3A>_>`YaO9)DHv6nDKY-cG;lWNju|Ya
za%3D5wy;?NF%u_-G?j)AYd5B)*yniQeqYFY!GUryM8?n^@`nRd7;5#d=IqQDmWc1V
zs>c?*-(4NywSBV=9+c<VwlP%r%E3N=Mor?Ws&a<yh@&Rv#vKd|PL9yX82ayB%MPbu
zz2M&(@C;z(<`7`xvDrpX<RX1Mvg4<o?O)VZGqh}f=hboxep;S@^1kjM1{a<Pi^S|O
zUA3I~uHLa+R>q~|=wM3X+2)mmvSocQdBaA`amSt-XuXJSL}JiL^!01t=0B{Ep!5pd
zE!qg;Cq4qB7v)i8D^@~a_>*){RsItO28oLc1MXQW(^);#(>-?I0GEs$Zo>EuQ`A~!
z6j9IL2mAfn$@&A5nHI$S6JLvrknKMPPH(}DKb)asZ3qH-P^;TdcsSTFN$t8be+ZH}
zoc;NTSGh&L0J3D<ez?ecKm1yEw}tXUP#Tw&G&Q{C`2Go9#lYJ->3eQu`7pX~P3ZAH
zdI7`i((I%<i9ZbtP#hY%<VCoSR-P@_nKAmg-+t+^rHqbtYPnps3LB!pk)~?NX5p|(
zuy&5q@yfuFH=Z6uvs108sJGi9$XRQNm*MCKz&20g7tPyF?sZ8Rdyk&TkjYqj>%vAh
z++K*HD3FD8JNd#B`;HorfcP5XPABH>9ij2y6kdF%>=DqSZ3`X@+*YEMG!zDO$zvC2
zcI$b~3Gwnud>Ja2bi~G3|IEzMBGWRM^6>`@TJ8uFz7jzrv7^Nfp)h?n3HB!<?AS4q
z;P|@Lezql9k^s}wCkmIImbP#8@s}L+269C^q_W0?QuZ!2PHK>gJ;t<76U8Ok$4fVh
zfzW(l!I_JAR_;~1q+8G>#Zpg?5J<7eoB-ARc2%v`9(?P5C5J{lKPp`6&OFsHIk&Mb
zE-Op3$>eM>=>H5UwC%1%8;QmE9i47GXHMdx$}&V8G<;wJU^?k-Y^u7qrBr|ML_#@P
zo?Lt-%Zy-cuZxlWnhSsfiN`!9HJWoa+FfVX)~05QrS02wl2nxIE#@%Ugp5Zbn#d?a
z)6{gvm+fhB=&A&*$aQ05C)V7qhwTUo6z7(Au50gZD5)=z?)=XL$LleG<h|f%>&>pr
z{2dfTeC~+9A#+BQcb>;Y8jx*bj>g&;kh2EbsKyN}@87^Md%u5%3wICDmxLn4|CMA+
zDG>n-ZzH8E!>S1}-^x=uDH%DFC*UoPnP+GP={N1VaM@#`$EV@-3ReD%AE7d4iGqih
zZzu}y(3KFDM5ri0#{b%9CONa=o3$6ub>(>k^+e@f_#q463;$j3Qpk|KGC(!;&+0(j
zbr9wuN0nEHIh$lQP8f7_%m#@aU(JP!?Oe8#=WuCD%ZftfW9`l_QE|ap!w@v1@Oym;
z__Xy>J$WPut=1(>lpb`#Kye;EVbX{n5fQalUO_x0hhdx^#r)fRzHFR}#}cM<WF*n2
zQ;}LY#~%fJ?1(r!K<2TC3}Vkg)IS1tFYDz6>hzSXq9THdnz|2b)QB=tb{1vBPF6+=
zIf^CasEU0uUU{_E4-q;YB&vRq(<~=YJO}EOZGIX5_RS5BzkXm8$veR+1?MOyhoO{8
z5|zP@UYemb`C0bI?L>`Fa{Ke=T)_{<WKZ63>Jb^4O|kK!2G^^Vp-CTc^DO>U&@Jlc
zKLjNl@=>#!KpO!pWEmF+2Xw(moGBS6RdoFXSG$4-9~HFH(HuP)!*RIoNt%&BqO2{W
ziorFc2I}=`2TUeh%x8+oB96~^Mtgfe%GufRKCj>__10^%U+_4Ja%}F-HTqGG8$lxf
zP*h3gT(?o;3E6SngvrY7TByfRH0Mnp+)FJSOJyp@jQ10ATSjSpMMAwkalxRRiw{Jg
z%}X8UJZS_E+FQ2FJ>I@>WMnvcO2Rc};yE@LSPUM&lsn1z{WlaS|A2t+n2dVT0+~Yf
zD?}#44fwpSabri46p5dJgpoSKK`ANevFS#l0QcK$y?7IXr{_+<4d^XEn5yRrrK6*B
zzS91Jfo3unsBh~-pGnV?-=<mZx8*hC5?xY4v2Kg>mDBD}E%rU9G<PFRP<gOa;8_w8
ze|@S#rE-QX`%mZ9Mq`s(;>1DgYyFtV<%%qwUTXs3$G)y%JeAdyK<2f&;Ey*4xY#$2
z(OYuaD!qn?Sl?@&M<8r|?T)eF7#!gN0nf^UTpfG61L0`L!3*dJk}-}&{SYg&GB?~y
zp*KI9%&%}Dt}B-W+>>tSE8V#{@5EA4QW_2K;HFdQ6PJV^NYJkQLmHKsXjD}DE3#T;
z<<$m9G1+AqkDxiDIThOAlNSt*c7%3_g0{q3&7M~&#wTycPTLsQhk{5|#dLTyr7w-k
zT>OtjU9+ft@sbg=H1ZfI>%mbWL-L)7#)D#4(@YJpP%ITv0_g=7H?TaKNI;<>B!@=B
z@F^6obV-&idFbFVFKHfS7HWPL{I3XD<sGB)%gX<1N!gwG5o=Swe=lo^YGTOny=-0_
z>+Xe<^n_4rwCkAER6lEcCR)l9N7Oye4Hy&4@v3ElOzuFntMu~P@B&*nYbnXS;&vh4
zx~2KO?$-UEvuBvOyw!;;$87{=q|fJ3BX3%WM<^%_yw~leRxH3MB0{vsOcAz>>w084
zPselp+K3>`uYT5ei*EOT842#xI$R11i`iGTEQ>Lj7ZyQ+7ff2Y>AuW8mrk9w==7Ax
zigilYV4d}OgxMZtI{ma>4Eto6;h;eF<@4uZ3)J5TW~36OXXM<;-+!YkDL6#nHA>`-
zi2bz_0}8BOm#;YomEArGDNc6!UBTT{|LI1y=bypHMqRHvO|iIbl4xqJmDRxCo_%y=
zE&|v#wdfPHaP=r)@+iJS<E}P>#A+$0|2CO69AxDEQvCrwBFa)yXEuab-TK%mQmB;l
zPk_uOXh`6J`tq%p|Nh}2$Jbo0WQC`_{rQkzus7$&^r8!6ymL;0d|1vU1DlXISCi!O
zeRF#({=u|bSI|$8C{~WNtO_+pbF4ib5UNH>fS)&fW_<CCrKG1RqT%sHhaaNIC^4(k
z6roegH?T~c8a63Ta(|@Ai7dfl^Wz=qdJNiEk|7o7ics{US-<ZY!6V$qOJILth4deh
zfD}S5K{EpfXsCyWZ^vyQ#;X@7r9=Z{k*FBsIGMp2muvudc^Ce}CuHJluF2HqM$NPM
z<ISb+VwKL=>>v=Rj~^8+OSD23(ECa7X;SC)Z~WGErMPdT?CRKnF{GUV%0{OT9$#&2
zUjecYe`i_6c!_N0(>nggFn;a_I1x!D?!iIvvM7%1b-nYO=Bc#xlaH+!1!Y(<`uj}`
z<Fw;FQzX_DC1zxtRH?>qZ*GE`(9sM}iaprM)Yg;Bi^0vgAC(;_SM8g!6uG5d`ay-i
zhDJ;=E?*8vsW5p(2b0k`8Hz8ii6`5S!4WT3M5&ZTWL3DQtY8bMNPZG4>)m9?2v%Di
z2gLjgZb+Fj+Nda&P(^89=;KXUxQ?VBs{PETUW`0b&U9jpjS#P28|GLVd3{Xv05@qe
z75PFnUr0s*LDnh9am6!^fmLNRf7xd<v-3n0fL&BlmD9EJ<#17rb#*Rg?toZbTgtmA
zwmuG;pEKChfw+wv{E<%77pYu3r(+3poD^1(QS9-*V=pEBdA!;J7jsPS=`#A?l2ebF
zHl@G0tDMyela0Oe{US@w{X?7#^-H+l%SQ*^d5q(8fzT+PMU`tiia0M+C5FI5IjhoU
zuq}>4BQu*Z8HjBHhIiyPJptfXs{_Yd!34TDs%Sq%tM+yJk27u0ll#filI^vvkS+ma
zQA`HC80Vpvr}+=sqK3Ke3~u#PfM17n6#nE;^2at6t@nguf*;LTiec7%;TwqPJ26=*
zq#bAoQ$<v8GiF`z`{s!_CjNjEQxV;vydN=Vb;Gv1h4uu6e+FeX|8v}aTwEBc?5EAd
z*d>ld68<^4A*S&fPc+brUr&-T1%(}{sT4Liz%mZLykyvPSP71TN0?n-8@#c`6Jk2#
zRJTEmZlLAL`2Jrn08P3c{<r!SdB(Ol!i_s0YYs{m_GE2`UdJw%CAZeyb**7J1cSOZ
zpozD08aZ{?I>CGG^;W}$s1pjX2}&lWa<H+!(Gjq}Po&oyo35W&XH<;R95M(*?(y#(
zXUHk}kcW>LtlMHG)(@nZ<(SsQ?EwLyM?LWz46Wymj~AOll5eO%n}smtOm~M1JFe|d
z*Rq60TRGW$cjtsS*m!`5w_8Ug_;JG@(tr|#c6%7Kj0y<}dE3g@?RAf0UC@Jd`9vw;
z@NywDm`Ir&!$1^9l0Y`IxdLLmd^tIJ@qk(n2+kF}NA@ou@{LV3vbo^<=yQF%*ytCx
z>i`o~EzeXDcQa$O7@YLlbb<q;p7?E}b$oix<?!o~fJ?$|3%U0jbME`ll+)IfAE;y^
zlgn&EWBn<AEUb*c)y^lP{F=*Zu}TchdH_wq^>TIaPuCxvz8y*5zma;=o>j@m%0cFQ
zgIzl9D5(qsS4fU0Sl#23^sx$tQ;87=t%n(Ivut)?v9E{3z8~#d@(Ma*2?u7RN9ed%
zbU<-|ly9;p0r01CJP>C{9A0diS-mdtINoolIqWuM_|gN=f*DI0IeFwl1MB5cmUMim
zPpmN&#uPMEK^ZF2@hZ{3cc@$jIiw=i?_CH>^V^b)`9wuMh)0&uJB0(sD`<|6;|n3B
zMyJUTHuU<gjgwqbwPjg&BGlyac*fM5Y7sS3aI(_|uI{EcHe{LPB^W3RON?OJ#Kcgr
z5ke)0=FZnT6HVZbSh_6)0w(;o>58gm0}Co@{yK5|r?n6Kc|JQl7%j;BSKH1CH}|Kl
zewu4`Yd7d`<*^xhdp%0@U2F_)J=|5`V;2P_)`V1}UBu3;;_mi<3HwSW%&3hUGWW@c
z2N93#I)n|v=O#_ylE6hh0Kzz;$x~>6p2JNYm-b(ay^)Kxo&3<e&J8AuAA%FBpy;f#
zXVh+7t&!<C1z!g-5IHJHP3Y4{A9w}im{uyq*wN1W%g&J?r`&P!u9qJ;nvaDi$clS<
zNkqIk*-aPD0;*&;ucNQSzPuyAa7-8R(srkf`_2JOu`CTd!@?rkQTBP{H<=?1LXpIX
z-D=#2vCMyNd7`T5o}O*bmq@-rKf_8ON?%7OkDS;Dtq*(RF(bxk%gSx^^J+pn3Ye`_
zA-X;8C9`Zf%d6y6CF4iM99Dke2`op+Lgr3|ilQS2r}};6QbF&Wxh1btk17<8sW<QC
zqiY-eriaugPdQ<ubFT&7h|m3V4t!`;6Ndo*;OX(``F5%MXVw%-1N6$xJZ8cqa;ly)
zM`9uxAnl^0XYg<%8febKfG;et15d~}$x`l(jW5b7Nsf;h3mtbEm1%&3Ke(4jDI7BI
z2W3$d!f%_6f%i8rz)-3#a-)K#1Rl4D-uCkM3&!W0XCJZzX!oH5HAO9r=(!!klllX;
z@$@nNkGoI#qX$uc+#J1&vVjPvLCfNzmbCJ6YDbUs$7%x4zkh+b_A3Kx=cv_F$C3gR
zp7yq5f}Gu8tluc^Q|;y!SNV2%vCbqyTzqnReuNp-H#K^1e(0%5_807G3VK?i&dxtq
zZ%IN(El5uA@O9qq{~#!8aE3ldrc5{5<ER394(>tl#MJc6ivPq8=)yI>$<Ayd%X2oF
zJt-xL1w-2D4S^FSGb@kgxc8v;W}o$|q?zJzTZn8uF)eICp3_6{8JFLMC9bJTe4>fB
zuAtx3o)}^?L8<Mex3Nyo&Mh-ajQsZRxeeBDZ?EJsQSgZqMe=OfUGZPCIB0j^;l=LI
zJp+wPoGG9`m)8Z?Npp~_z|SaW=^KDWe7+IM7%@)KnP9B-Tv4RcBmgMhJUIOaG(^fA
zsUy|45jwV2nmPITObmV~4VIBN9-A3$a}@twegq;jPg<aVp6dTBJNg2cR6Sd|Ry^-r
zG~!D?p5g4~`#QrF?D>Vx-QTMv6V_H1yJ|>6KVsua?w+GWGN;TWD%H=6@T5x-@{^}>
z>59#p0=AHR#bc7QyT!b%SzmX}u8#|A{kmrctV5hLwv{O(!2@4Q{K!X&h)0Z9uwTYW
zVB~de3p)<!=~`Pn*`uPcnK@)xQ`FBNY+klL5ZpGu9A!iQSvL~dmLpa_J2=}ybH5fk
zK3j0$lG}PUv3pkiWnyz+#P>SR*;;r8P)d%@()HMLpEoI2{tj>Oop;Aa9$}KkcJ=iC
zXe46ImG2!pKbngA=jQf~$X#5#WK|74H&GTm*F58bjfZDs>>kCfA{yoLu-&+F?RoBw
zhe`^;Oe0n4VL2x!A_UDXt~*JrPL@zk*Z&6b#Giw+<h(kpOEa*k?&AlhqE;u3T0z0G
zyd-KS&Z|^NdjIDERc;>{J$-cbGf>o_Lr12K7Q5fO{QyAT5{Kv%lzZ^-c;v}XY%fS0
zdvHne%E~e<$r%D}d!x(Nt^jPqZ%174iNL3`6X4{4C_y784adhV+6+~yw7jkzKgTXi
zh32%i8K8$&SM66XBvijt4hEEz52wS+N%mPZWVax)4lxI1{G$tsph(u==2&eF9zS8@
z>*&lcE>17Z$g%N^^ymu$W@;HR!z%UGkeEW=VObt7@8eI2@sdD#8&KwN_Cx~Nwk~HJ
zU2hj3VoEB1|GpqY#V6<!f6etkEFJ4LcVq1D1UM)i!ChTKJgaMALnJUTPo27KvgE0g
z^4oVt%`1T3?b5Um7203afJ^@?n~jhsVC`aO)@n1hgof>EZ35~@?<b;podHQkV&}`F
zBC7q|6$9A`f`fHSX4|TvZO0D9Z#&<Vj8T>t0=}PX7x@;G+JJKd2Ke|S03Vi$MO2qj
z2Vdvh&X$<*h?<&)gc+x#$3nbbliF4K-Z-J`1Uq3aHm~yD)m1-LWW5XAnjv_m?#m%V
zRw#LvND4jVn=Rg7>;%!FJs1=M915;V@QuA5isLfO@WtP9tL-l%`?q}U7PJIHl>FTO
z{b#*2D`7<#B%xZzmKLRX6?iHvL3g7Il^Tt==bYDV|NmUF+M9Qb{C7tITZ*+A8Xxg#
zWy)Du1%Loqao-vTz%50|>&6W#YFHK(%TWiuc07D1SO(8W=CkjX)5&(wyQ36DS*~A@
zhz;%!5znBm<JwC!J?xc0O-ky)svs=l!eouo(8@Geh1@aTmqCs)I^%Iee-FwO+!|U+
z%-sp6j56B40~Jt#JY;13-%Wj@u<?Vj^n!YYj*owq=CD|2(S4;XumBOcUIrn2%{#@g
z#4ec=`+L6ijsOlek5P%DbIE#19!cW;Vw-KZ^(OXNqZgxD=f{|zz&GhFE{iX5DfWZp
z(dX`EfkT(?B(d4`zpYj@RiQSA-?wo8b5^;5D&-urJZ_5>PciUR_|gqt7B7O%EjmOG
z=I51JtyKBbWinQxU7VYKGp%n#&`7b(FXW%l@VZ_eCX6xk@41rDdi)c-O3@G{l;zQL
z0=aaSo@>#;XfdFO<9f9j+z&f35vtMc%Wt7#>>*1(F#+wsts^l=&c=}=N{4BpZw&6I
zfI}WE)ImUL&eB%xPqbstHQjh3VBQu&B@@Y~b9B7ZFH%bKPfr2Kds~?A@364-O8pm=
z1+L8*Sv5B#&P{uWc!{7)yEgxv99=-vc1x5k)@ZRO2GRkeVq=xm|B1PVzH!g`mKQe|
znQDK%Ed<0gQ!yX{HHm+B-p_AD-}0x%Mw7?FMM_#;c(;rOBM$Z(G-BKu=Qid-=T5d+
zKb)ZpsBQ_jp4>)-E6p<mV{fjxz_%a%wj88rg$?ckJ1Sii4(oO|rxWm+D60VS0uxG3
zk71Mj;^{)stOg=J*Ea=)Pc`jXh4l2=_V3RvIt+zKnYQ=(*!W*XQJtq<*dh~~&N$^!
zoM$+hz5$wM@4JBMf^z<-$E}3BwyvTNM3qLnf7hgIN=^IEl>;G6q@Z{gFf}1!w7@B)
zQ@3g0dyo|6zB`%{tF8g<+0SguiLxXYb(P?@4^71{8YwTZ%WEr2vSdT*g}JK=q2~=;
z6*+Oq)V4%P&`Ft?uy^(sy6XfNDQb|!Gv%>jO6DESCShRZQ`0n=5NE{VhZYdY<-)~^
zlA`_~ZPg*$WeyP+8!l4f+F%e14UGc_15p-6IRicoSZ1?cZ$%Z8{RR(he?!j@=<=Ex
zGm;FV+^OyuwZ?yi>G1FoJzCRN(f?@HiQRZ9C#W%ng}U!YFRqIWQ*~`0T>iyPF#d@*
zg$<Ll<<(#{L&zrj>w+CmhS`V0;6t(fPc89OV@LUQGxmDovA83wqRfko^Q7+gMF!0s
zx_bha999x#>0ek`C&mD1SYue7v_PL<)%^4R1t6037j|Pzvl(I3U?qevwWz&#!i44*
zL-{@lV-s)y=0Lf{#POr(ng~Pe-XTIx>uvhxo_dqg)JmDs9j~oGJMVs`K&M0G%y!(^
zjfqT^GKGrP0LG1`pu<C|raAsGn(B9Oy%gWcw<)|LBNBva<hv#9-Uq_v9zCJB@qZ3%
zKeb~X-`H3!+qQHlI4taoMmZG4w(16_B?bb|ryaH7ft-L(t=dsof2@mZfu6bA{H}AG
zMw>PD1T+#V4TC{mGgjza_<|{=yt0Uj#uw<vbJLpyC_MgEz;6;274wM)GctILu`>vc
zu>+9H{wB<4$dR!OW^X<5RQ%Pv!UNB!V8Qrucg{M*<u&p>Njs&<YKa9STtOSJAg}gQ
z>R9l$0}2i85FtDZAD{n*X=znelD<B}OgS~A!glYA@4#z`hxe-y;oyNIxLExdNf=c{
zglxi7|6VMM*(2WKFTq@Giiz97yZ8gXkGs9x_({`~<I}_8nZohxjj;Y5veYpM@TR7w
z@pRr$o3-BxDmsJn^XTa^Aq54<(y0u^H8rsykQgSt?!fghjBX89<${!)Do$<`@MiiB
zQuRRRD^AFe>g~4p8Hlc#105w4-U5O{G5DmJ6U-Rl+jpL=e>2=rBFSgi@hepZOcQI>
zfBl@y0*PrOLPw_37q>SAzP}Sk;K_M<dWMPZ(P?(1Ns!5~VHR5WqDiOHnOJ*gRaX<h
zzJlSoeuW}{3!p;1PtkYdQJ423-$UD5)**o1h<<v5@_uRGDy*~kf?4_P;~El+fj`xl
zPtNL2m*C}ON~6gT>TL5t&a;ij;|c?di5W+H+9icvXLR@&Ci^1W9S+VRx_!iSu8Qx)
z`twUDRe5!x!6hFJ%_ufLK9tUots8PUrNt#)-tX#|+f$yBT?eTEFz^KhYfRo5cRZfe
zxOujW3MK)zpMso&Np#!PH2EDR1fnB}%^s5Fs*Y`9D*3VIYyo+r^Ye@Hx~iZYWjsBi
zZxvNpk2eld+H$ZF-?`3RS;EPrm#WAfy+i6FV6_o>+e|lQ{&zLg0?{>_8~(FyDGMsZ
zb}bCSFFTjq&kr*V4y5{tI_0c6%Bn-|7=GWh51r#xsP`48U^^Xw>{I@N^{XvR@Hnk~
z>#-M)58H{cwA^0Qa-QU!vKMc6>nDxvN&ErJb<b951D&2IXb3}9IwxU49pS(*mw56c
z(KtAGT<7vL-27r+(^bGxbN_PM&I&k~1lb2t0$pw>y_Wu+De1zYGwunKENhN*X*Akg
zV`4aq1_v2%t+U_$RHbPGZ)%D>LdN&eK|EPenctE1&GOSe+cJNR0D(pbWiH$K?vNm&
zsAx?`b!7m#nOfH0PtmDclS>RAXf=ac|5lbU(=N4As-{!6#t)||VkD>J<-cZT;!gx`
zbze9eD@My;>G)sx?oZ$yfw2Fz-!?`Mr|d;{G`(@T<-=T}V1kwm{cTOa9IbkJRaM_k
z?cD0BrSTA?|LCaHVwJ8>2pI%m?GY3CChE1oi6r!?nKG<ojT+%eqoFqCaqq&}#cghO
zHDc+AJatS_MP+1QnTa4Zv4{qZhPv;-)oHuOck!22Xv9w<^^;3WqnW<vET6}fWindC
z(CpJw4Zv)?ybPnHghEysd@X`4IthT8tQpkXvqV@1_kQq60+AnU&358TY#EtIq%hzo
zG8x}vk%W-lFISELt9uETa@5pOzhzSa6|8DyGBDW4k`52p0^RXqhum3dXy}IsD+;QK
z1BC14wY0{Tm$5TSc5d7Ookk2@^s@Uyh^E34n~-p?obuym0S%o_S3)P_<!(7;?*3l$
zh8RhX{+~!j!{)Vg6)kB%>5i8Ha@kbDgTdAX1O;j${d2gUoipTX!%+OdGAfw)_+;E2
zQ57IHMo6U)A33fHO6oHz5{Oho1FdPq9H_!X>l2gC0ZFpnY)!S==ZH5<BvD2-*xc@l
zh#SLnZ1ERYj7wad2iNAjoYfdg%Dmo4U5qP;f<gkc$M?QARJ)2#r*+s*2*ZMtZtP4a
zQ&yB#ZVa2c=3g+(WzA8)f~Q&u6+0Aoe2qf$ORRp48WFPRvQN)^h{n8ykD>mCdN$sG
zh3W}k)jHfE>(WuNUh1vNQgOy%)k}P2EErO_0sH)7;HZ`amZty_2w)}WVUAF#-@Em!
zmEZtQN)K@hoVQjw(PHxBjF`2Z@8|$gC1?qC=KtzV;9OWpJHY%`pnDfFWEPHaap~>(
zd86*c>izK&(&7=Q?M@7g<0Os={V2_Bft_4h7Q@FgaBs;{hZndP%I3ZNsnGy`^SRa(
zVLTTp_b)Vlek=YIYbLQ3yK&N(Nzt(BDP>Ls5w`5Ot|Nrg4mcHKXg=}YeQ62NL#Y&t
zs6Rfh{+X(90KJAGS-10sJ3}I~@$!BVbB^_*y7GcL*+`rRG*S-UPb%ES+BRHR*r>k=
zgM(r?Lz8lQ*fl}bUyBL+t|1D=*S3kgk9f0Y%FfaEhmjS^1LFSN4;60Ei$FA#)5pCB
z!QCzDOR=COc9T|N`ZpnBY56@+eY{pehj+>(0HfqU2R6UEy4vKkk+_@V%mP-!oKb0E
zWu|^vm^(>KNjWg>KHG}N>vq}05Wxj2?S33DuCR~SRt?MobBu^tNPfxWur<8O>12Gl
z%qwK|awTnhKixxx^#7(nAGxG0V-g-j5n67MHRLp9LxN4W>+S^&P7>~(piVO4ufyxW
zV;vP6;|K}o+MXQ!m9y}SB!Q~danN#6?+KA4oK7ppBa2jg?Wq^Z_d8$q?#tF2V+;pj
z&)LoF*3t;jYeXl{ClUDbBiGd49Wr)ktfSQlFnZ&L6)*uL)*;&lhmKBs6_qa-SaT~2
z!Gk1<=rkmmQ#WU!1VGlPvPynuVYW3x=z`_JC~b|gvwPad`;{HwkyLN72RmD;Ihy3X
z0*+3HMGZ5cq1n85)AIaw<@bXQbLg_xX>G`EugBk1ceyQiL?Q@D`@?r%`*Y=BAJ)gS
z-n0vXibA*u$@xYzQfB|8GG7Qogal@MawIPHg@qAW21)t^GtP<_N9N##>)wM$`Vw1g
z6<veV7BNq(8;X}FsZuywL!`F0+X(mi$2x;9-@2Ii4@M5rAt2)1mBzh8Kx)jgOtUG$
zn55V4?$3=fNniaNw7HGV;A!p6=Gmt7(Kci(|D%^kC5DO;4i5g{j=jE{7x(5&c-oYH
zQ`q?$24&`$K?65n09962iJv%>v{3zxXFg=LP-hCN!&2F;mkfL6dba*txe?5(EIUjb
z|2triJF_Ce5{N**hWviMwk2aBNuB)Q#O-O1g~gCIsN^M%tyVE|?^2*;quxN%uhU63
zbK~iwoSb~U8)Gw|WHAlZ#^W_yN%Qsw0X(VT$EY9b{=X*v$n?K48zD9X-#4qm$L;;1
zc-PtJMSM2S{+CVYDFG|8_{;XGTB(}Ebj66|<eCDulC6dXiKx&V`#Lk=KH#+l=X99W
z+O}jx7Y3GA6}VR)KRRk0Wg;dp!8Dh6ZZd1HTB!hr5+#fJrSNoTY_jQj>m#`RGJy9g
zD3XEa&O?PAG&8LqR)*71dL~vto6VtUuv2@}QGg9SW*Y4e2`;bkB{*2@R5l^XK|0~a
zE)~vOKp5um?mAKrd&}>D7Uaa}W^o}GRAI^G3J>t!o?<`TJtRxyYWI%Z?(S@a>E1$L
zT6_-5e~dQ1zGtpG^-R3d+8q&Y@#dxquI6G1aSC3fr@d|v(b8lT5SioFf5ODq?dR^K
z;K{S1aUBV$_;TFlOM+8kvFhx_T6mO36$Vpq^E7UUDLYrYc0m=v()s>j^Nk1F4yNEI
z%yHXKsxaX4Msz$kgn?2zL>Yk#lV<@AUQmY!O8cb>rlE>Lrz!&GEGw+SZq4yEqq?yC
z7Dr%;i|}hiK^OJWouY_>2)d}ThCrQN<X`~ZSXe<A5sdmTP^iDRKrGXSpe(YevMf6p
zK8UBSL{1L^MQ7<plo}9J-e6$n*&9Arns4?0==$ogD!1-i2|-#>>1NX%N`ruOmvnDR
zdehyYfW(HqX%Ho(Q@W9EknZm8zOU!}zI)I4J>PvE-H*W^d#!i9bB#I1m~-A(5w<6B
z^v5U1ni2Zcg{4x2Kiirrr1FqJ$}uFbi<-BS8sk>iKoI@)@pY`(`LD3o5giZ{!4C!g
z-+s^`Z})RN6@90n0Sp_d%L!x`GSHWL{@$6C`P(O>vaeQ*`8kx~c)yaaZn^91=#536
z4?huA9(jWhLH{kE=vzl10G>`pO+;hOKm|{=RU%Kpb%yCV-!M9@6?q*MoGYyeZC@Zi
zM^=}RsQe@i9njhw@cH~VAr5@rC0)&J3U>kh$3<>NqK`ZqOBw_I)~aS->+_6C;0Qj!
z5khWk%)LBKu5G22=~Pui%6Js?UZb{OAOJ>Ih}<E7J}=AJEgh3m;>x%my6}&WmvG#<
zlVoOh(9=vU*hTtTc3Yai<+FN5FTnxUP<hdxj<q`hI2KJm(SI$`U0=<pEOm_I_}o;E
zD||-@e3`SCCws_Y*sbk2?Z>O3k!l*N{AhuWk>Tjyx!3omLq8=q;eRU5cxAa=X}r?e
z83~#?9U{IhDhP5K3c2y5=Gbr`2(&Rzx+a{m>GU-t{H9_m67WRN35$)iv{)QW|HYz@
zyj0|fYpMdZu3jniR%W#FZqM-QrV}j&ed?&^&|$Y%?WHN_z@;C|D`S_2QI6a9Xz_Yx
z{q<coN8F!`N=oZJ!ie2dWKb}V#FAy*VCC!fZ@ZEoKi<?58gn$ZeP&}blT7ZnIrS5j
z5;^Y7SL%yVlanPHv@`GA{QTwl`(c;;_<R4aK^=A+sxE<<V&<SYdViyzsT>PQ9#6kn
z@;M6I!|4TF2HPT^d2Kf<r48&j@G9=ceci>S^W*0a9Cz*XRU$pLipJol2+WKf+%n^#
z?gVYe^Hf<Ghd;sI-JUcufBax&E{}*8F@B|;3fW6q)!L)0_S$}v6kEQ<b=|dX&F)jo
zZ}fs$t0-4~V8o(oQV}ln3i&;0i%BXYg*Nk+zy7PF*hajRffysGWkY;$QtXV<i%&Oa
zeexM_ueQ&tGH+);Kcd%t%}_b{{eQg%_DEcS0`V7?=T~|4{c`H15`oD!Y(c<fK87p6
zzpOzU%z5{?H}4UzexCODxY_tq)5S%TZJcSId!VU1$DiA-+vIm*4aodANcP)EqC(s?
z?aR2u_ZH<}M!+X&GMw6-zX%mhG#n!&Y0%%BNg+H%Cd$;V+x9t}DL6Qc=1KTCgQ9E3
zaeU9anVw%<=W@8YutLJyaC7f_y7k@Ra(>BGt;wFe+WVrS%bt?=og3{lm|8a%DM3=C
zo)Q(MsG(mo33*mvKXww;3$40Wb`3B5qlD(q+%}D+SnYU#fjZ8{a1H%0j^i4TlD&^V
z!Kw>q?rRJ2@!j7^A%wV1GRp8+LwY3;`!|k60etg8@BmNx*9*B#xfVQ3bEaY5J3g*k
zKd@B#Y^JWpYoLt+<t9v%?6p53$`!hIrDlfAb)*8PwcU?t2r1j?m7%WV7rp-d#hQTa
z%vOP##D|{4(=iOqleoi&@Bt;Efhvg%(-W(Mh98tks(w=aZtPc(sV?qFn<F#KNcz2m
za)j2ha{rgwwq_*Ql_<=cQ@3Q=)%Q*p4+Kbt<nCvyhP^5pigYu_QG8d8nZ9pP41Tjm
zd7bTEoK`;&@|!=mBoTSwh-wg7ieN84#0uj$A_*$-SQ^1%M=8GVE)FWT_|eC^<6^q>
zF>K-ThefAKU-@IC6?0C!sw~5KwURhzcD(AxL`?B~xOWqEz+kNH4<G$v8<*c+u$~qq
zobHASuxeXoqgUBPQ5S)^G+df!FMr6)dHR)XP8R#BBU&$12Cc`_i-y~&HmjHM71VaF
zhPrENLG_(fiBPzNUC@w-GyL}yejBDzhemvhakfK}l^!r>g#j}R&47)tIw5PB$}}m9
zB}l%LmlMLAHn`z_3_8pGI+)rc!YABOt;aqPD2U{ZjZuxEH;Abi7BXhM?4Xqx#>Hvn
z4>rzGQ^xJ0XX>)F@t-sZBIL@VSq17>1N>B<Es(Hj1c`qk^r(o^m%$%nYqt9Y%tzPo
zMbI4bOCC^~P@6=KV+uQelg3R1WdL?S_OA~t&Qj~3KYv~boo5~9u0Tk#lKsXt?6~3C
z$b^d1Mjb63V@42@Aa(m2gO7Llr9ANlWjH#e(2U7$)sp}f(;m`ACTo(ZD0bY~=cV7o
zW+dS8BraHLL;ekd3{wzZaH^KZ`tV~YJ;b&A1V*7?i@+?0t*dUSqVq8}0mFQb;x&gf
zEnhh)S_gKte8l^&^}6A8JG;gML=4FFmj>I&e6GyF?I@Yg*VcHSA!pGKti~d8)l&>^
z19fv>+Lyyp1q)-dd2i~vKj%2ClQkY}Sh5C$AFo9Uu!E)Hk>^O1C*t;oM0mY82rOq^
z3$BxKihn)(zgM0ox+lx&m0t2^@llj`LFr^s@ha11vJ@gX4<T`GxKPr=1PxD5p!y#e
z2q^;<Bs=nq628Xzj~AdPMJwFTb)huDK3Vw#N@h$*Y<GzK7m((1X+Op8VOZ7Mq)V5B
z+7klwo(4d8Fd8KBZn6>Ha%lPSv1c^2+|G*?hvkixbpqzdZb+M0(DoSCN9+tD5P@7a
zh@hfgMDjZn)zw9&x>G#N=+I!QS}Dq9>v^F8)|W5JI*bW18Eg{9trW<Jqllxsu+sij
zgj_X#n)5T>srpG#DwW<yLOcr?N#j8vF1~rzGa|;Yy9d3@=Ll6J4nW}}d2W0p=-mG8
z6JxQi+5J}|z<i|toQ99jZL5C(Cj)EDx}-G#>11xXJJiVaBlqyq^L;H&LicMkt@;n6
znevi|TwYw5GC@`L377lLKvU1u61$dKcZ4h1D8DFCZv%`zV`%Y@UyRJHL@6!&POLo@
zgPVT#`hB?%*EV{l1rF|d1lzC7rHIeYMGeWXulbm@O^%o4TAf?s8M=grJWdq`Q*a#4
znmHcsB3q+KRb$Nwa3g7Op^wwA!T0y7q7w_Ol-hC;g3<5S2+9b!-}C^j|6bvlOSP=8
zlaj)(y>Syg1R+2(L^O19%0$lVqFvY6;0Iy87Ud%Du0bP+15E+h{cEL3?^rqOP=aoF
zl=9^r+x32*KD!CWSbO_}ry>zn&wt(zhHR7(*9hQ<{ALdk!%gK)E1#yER%XFRtpW>@
z^+A7cbl_))GvW2Y)w5M`le5Q8lJ(Eidb)iD+}f&aM{$qSENyAlv4=-9#4Qq{5A4t~
zKNJ_P;G+($f3a9@3wG9HW~ibsX8omB{d%J4yF)7?eC}vak!MOGg5~BG^I{K1Rr~O;
zw@w>((on)i40UntKGkCOiWsxX<BsU~TA?<>v9T`mMw=cK?cs6avcBH)>1Qg5hwsBR
z-goP$iWX9VtVT1%!QsUWH#|*0Z~l~n+5Vu#2f=8VCRcHJV$9^zB+$J+%kRm_kr@o$
zHs&T*j@O6#@U-Xu3K8F6;J-G2k7WA6NBpUoYoCOdmw?|rlg$G2y2?vhX-;s0iUie(
zWdg7XQmAi}A{(k9G<xpKvl&crv5Yl7x6;{IL~=vmE`**yz`%&F)2Jhs4|G6mu<LzA
zAT(ptpZMfEQ*O+xYFL>Cj=E-ZuC_Hcs-6q1L;umF)T08~c)OB$M7VH%x;$6PE@Tm>
z!lGSD81d=#v}ut(Dlt!6vHo-^GmmDKvUF7-BH92thM92E<mp=sW|n~u`=r!sm~0Jc
z^@2Dj1tNuCi_4-0zPQ*=cWH2@F%>_OBxG>B&z`pW3S3mcBL8~qzd5$#|J)Hfwuw1s
zd#CPE*uANWG9D<Q;3WB~6YJ760U$yV){IXku(*YH(Yy$MpCzAPuiN_r_^A@A7Yr5U
z+VBQHcY`TM+o^XsB^io-H1%Vz8_E`j9_~433o=DFe1Hd8yuV*+7P01%mR8=s1s1W0
zB#^Otq(s}BJGgXaV!<%*(TMV!{~3lPfut(qexig@@78x6Cxh12bR`iHf&tosU`u-y
z*NEATI}M47Q{SB=prB%RYp$Kjyu-up!;9##?pbrh`)L(77n9GpJ?`GmJ}l8u%j2NA
zKhC{3C#mm)#Bz_;qCg=cFEIe7oeXwp%&m7ev6BAlmi%iU-h8UkRR{JYv}ZZytfQFL
zsh*~i=p(B(KZ{j=3tr^>yu*+XZs*#Dm<OMHDt(ti_L`O}4KN)$ORcsRk`Ct7wR+25
zxVsmeagS->M+)>&WXCMDjQnN)fI5`v$*#buF@Bcn^MdIjq0BWY;vRKPdAoAW?qaUQ
zT@P+Gbj_*YQB;Dn;)v9T&QSzXJQ}7X;5!kTmmLd!ya@?urwq}pl1C}7AN<yA=v!s4
zp)xYug@$6LTmD|SehnqUodhTxTEc&P>UGg=ySE)XYrR;$vm=VSvmKCe_ue&;Q$nmy
zd24^xjQ>tuK3Sq8OZ)KRLn`OLkNrL#OjS0|m{S3*d+#eV5$wT-H6d8<La$t2()VCD
zZohP`xRIBw_7#a@H=49|+*m+tk96S6Xp4fLn;Q_lwS5AFzFL4dy>^~6tgudpj1-#z
zt^U^T)B959xS-_<2SZne<M^Jc(C0&UZ9`SI4%bq~y{|CZgHg%3Ep1_HXXiaoQP#{{
zDMUtr>NpN7lys-?q>Z`o`iByB`-KrR{X>7_?*58QJ4eT3c5EZ<a(2HeSf-VE^!4?S
z?wkXX_i_Y^k`RsQB2;GkBcJ1n|LRoGaFtz%cE#61%z}ngdG`0CXG^Uu(vqJ*`T(7$
zhkdQZiMy=z)XILn?NH?YjoH6;!z2tmW2XYVA)dq+@|?|vh{e*s6r(}R<pn)^&Ursy
zSJ?)IU{Df(^-WNyrDSu=quHHmTUN^)pq=t`4q3Pad21uK_yYW(=Y0b(V`E?vvgM0N
zVmdSR?aFvxQq~ksg}CzHv=5~=L`rsyxfSB@+n(K}jg*S!SEoI@%nxUDR~>$YdsCDi
zDU(cce*16oTFUDMXG}*)u;Z2j(ITB}e|aBEBnoN^yXT)@wDB}i(tzXW<g9-oV2U^8
zEH2PZ-tHQ<a}5!AKj2}MxV1!;G-ESqLWP~)ey+ubLOf(Zn7Cm~LySyug%3kRrxjSN
ztXDVoS98uwEvs{mUE<tXgk+P<cFl*b=6s6Xux9Ij?ajCUds1Xl5D{dIOgLm>RT-vJ
z6+Ps0&8)417Vi~DvrD`~lDK5uS7d_R1yk~_y<-wAVp4WN&mC4SZyoEINe(bsmC&Vp
ziM|Vde#0tb((66JCnD>s5hIf@v4rmFz|TyQ#~UTh*7-Ycbd)2uW&Mof4O%7xdw!=$
z=UJuzs(~w^o@-La#+w(N8{98G{d}3m8L>Fe1~X5Je5E}dVmm_r{<|Gse@46e<=c~q
zl7xw1wOm=RHnom*D1zCfA=~CdzeU#j!u*fj^Y(RUKrL6%<HUAJdVe~f>C_}#d~=Le
zLIMiZ6skVgQbDkL=}_y4^Q=~{i%8>b{1{2pG&e}?KjAZ;=>04UBM?H`*BqpI;c?V5
zvQ(5Q^W{Sid};8H<?ebnc)}07mDQ^P_3#k#ru1w%taOhLu2?9>zn&%aX=quZa--N)
zBSm)wSoLPs*Dak-MqK#~MSp7`DMoKe<ao|n{%l=)4G(;TPv`iuO9362c~o(62!`j9
z(GH=s`^?EP4MmSNlqIP=Itnx$J}_n6_frW!aA5QWN|6WArL$|5iLpwB&yw_~afJ}B
zTK|wf?7H76z=aQ_USBvhDE@ZiFb-57vt3|jn5wjLc)TA?ha^@_cwBXq2mzC~4FQ_h
z@WF)LS1`C8*F*;%p>0^3v50f0yuR-J_|T_xMU!R|C2-hqIylz&?C<*oxK#2k;}Q&g
zli+YXJZ7zTTj~!eHNbUqUQJ5`n>GactA-i7t|#hiz=h|+hPW$eOi{6o8p>l^xa5+a
z^LAfP#yf4lsb$u)8D2znaX|rShll%8H#QPp3es1^L`wra#;cORs?w{g<xo2xe7CmW
z{Q0%dv;C!ZOU`p!=JvCv*coacZ#+Fk_3!9*$GfFo*lxQ>6Z8kyr_1IC(ypC^BerCz
zuiZo17Hwg|pv*ZSF|MqZvz(kUAO5{eOGMwMiPv_hXiPMrt3#gt3jQ=LO?AK`!Kd+>
zR`Bfk@5%NO-=~YW>6p}~fBb@Bvvxru_h^FxN{B)fBFKMkYu30I*3?pMO+F|iKfDLV
zavFEbte=2%cc;fvK|CT3GZ?-9IzZs@d<||Mo(hL$vA>{L-s<PB><|+_zqikj&@4x{
zzcckQ?`{#Aw|*7O8LJ__cs^4bb;o7hO|I5mnWTuvhR7TwBJ@<iO%`u*nn6Ps6{okR
ztW$D(&T`Yy^;kzRozxJ(kgYQhNVNB8Jgd1L$G3}3<$cRB*r7tN<F>x=_jo)wUY%x`
z&BI_<_b6tsvD=J;22$4FR#vqg%$M3Xi>g)QNYRIh%{bz3W;7fcQ3zz6d77ow)}8?M
zBGf!3&Ey6iPsw}<-r7QcJ5I7DiC>IFi!p}G+CFD=P}ZW?HdlE8dRcm_0F>Y*3b=iW
z*=t$7%V2x(w}ol4SFc>j*?2zK9bA=6G8nj%ji<-C(uGR={Yw4)j$!>L;6V6kA{ppo
z{5y+ZIq{b7zoveXXIH-HR=>^_5~A7&U;!dCZHLV?b`#_%^1CNezQPIm=8n^5(_=-3
zIG1zQ!^0FX)6c;T>9R7j?c!ls6~<JQ9?JBfFk4%g4<2h}_kfQ@f|lM#-1-d@)ne9E
zKW}AoUkNIQS=;yp7yRn^owX`mv&CZ8t1~Q={USv&k6&v)CmZxUt*|DQ_^uywN-^UO
z0*N5zo(>zEa(=(c%WcZ1CKKkVUU)e3n~gAic<0;XEi$Q(DhF?wDpS^rTy|E~#UM>P
z8kK|AHhRfWHQvX_jE+7<RvPYTmit%41#Y9H%ZuiBz<&H*6bTdGh<%MG%fEgs?UFb|
zhY)Ed^KN32&dh){0i>Cojl1*;PnJn+Xc_uV<vVEE^Gv1Y4&veGUojXsqr<0e$DVQy
z=7e^+xXGbdSU6MU>ef6*KVivH@N!EaJjEa8yEToMP;>thW_0U%bzYJ!)|ZxMi^qwX
z&U`$~u}k&gXJmHv8m8;Pne_AVIp%vpQo7upuV$ha(LtfoG-lr+Xs;8D$>tJRT6YiN
zrTq(@o@-twp6iVdzSqSMn|@2JsLOLs#t+_CZTM0B+`D`I3!0Wj(@MS_y%!)*y`>D@
z_T4?{-0`ntJ1{=h7yTsyxfU-~yLN$Z|3s`5IaY`|GJmjJBTJaL&Q10huJii3jD)Ia
zfZ*jV&B)GP{3|9p)cky-J9m1v1fTtwe(s9Xey9wg<Xd_WmqN|86CQxmY!(oC=D{Qk
zk+QTjn~wEnr|w`vr%so0hfMsj#{;#bhL`}3vf%e0g%^ATvcpW(UJ70V*E+2}>yame
z`-2B-Umtbjm5GV_GBFEe|M8{-qNIp87QFLOtA21PTjM&_Pm#<t=hYH=2)Jl}A2;;;
ztQ+r{w6r;@nf@OHpBDPlnr13R03VMv3rKDB%>BZ;6)1)jP6j-Q_@PfgDZ6=ov#`1!
zTJBFz+v(QdHc<cTO#I`e09BB`M9Y&5m(13&T9f|qqdSy!ET=W_>4~SP=vxXWewuX-
zOc`T&wYC&@L0vOf;LJpGS+rnH>I$A$w!&FdRq!HYI#*A7OhJgK0!~{>$%lCr8OGAe
z6Ixm!)h4eF?!3|BSB`C@&69DV)03v!ybmYLKn~v4Qcw)_QKV}3FIG(Cnq7GpBxC4@
z7AN`iXfuSJ2)lQw^4&Yc`qWgL@r*|EnA^ogwmefQ)VqsnWeo{VzVuOOK*DqL1&gMR
zhM{+m&7isCd9&c+hP1w}`l!ZtVJXTr$>D*kdrs*8AC%sLz9FWloZnK22`+5c)j<iM
zd8tKS8`yu!W@O;~Lef~c2K(}TWe`XE(ve~F6ObD1?XJn%Ehf5bn6L}Q6T@2~`DC!;
zttJ-L3%)cVl!3vL22p&zTDlE<L)kug`Ha;qBL-M6ikD(;LqZA@%-NN|Y{*dHDIN<{
zH$_pm;R@Q*tLemNEsM2G#Fn#|k)|>+M!k!PLc{D-zg{m*OS4yrS`WmX>iN!GR;^zy
zB27;(pTiuAk9c#2wy6Yd`Qxk=d(6`RSL*pUvRan0!2HI6dVIAkxaZIWm6zqvcB^*2
z@!Ci-c{mD_nQOow6d{+D1qqyZ(r5xfajGmL$<wptg|nKn=UsDzJ}W(X-=_*7!a<p1
zWBi$8{Z~w`j^;i@I@5bO=rY5Fj$48c0D_}G=r;4GhVbTbH=+rL$KrAf20BVMq_9+(
zY@g!e5SjahcFmdU^_+}}Va^!ZX1eL|5j9GQ%_*n-#4|`ES!9D?ZL6y?JcBK076Jj$
z_kq{?<Y;NkxD~ecJAR(9DaWHu8a~3kl5D2>igWOEc8Ij})Eq~P_ur5GHx5RLk%liq
zsF#rl*0o@;{d;ow2~dV%Jt4huvMJKHyV@nEY#z-*%q!JT76*&tifj%u<=D{Dz3LiL
zSTlSOing6y6Fjc5DJm*8mq^7Zt~xq<)0$%6yV+oHqiIhobV-63C$)T3yZ9i+V;3w%
z4#Vi~iCrqF7Z8ve*zW)E1qK)6O~vX#L_q7*(Ou7pB;ji?IuL@?F3ngoyYBg?dCf(V
ztBFxO|Meom<UNj`N@F8ubiaX<qGFYO_AC~)BmmS-BpUBH@7hHg-Z(^ZD>vsxt#Co<
z?g8`FKiQQi(Z4UYt^`Fn#owb;d~)_BcXQKd0|TXgB6|xghuV0R)$JGixH%f?(xKpJ
z8sIQ_@M2m-9^_UYK97$SzUbj1ta<b$S-d4ncwBxg^E75TKZp2waqBRnl(oIc`gZT$
zbfe*ZB6H-v)pP_gS!_&@XU8vT|K(Wd#H>rp^&3cua?_z_#f9>CB+(Y)Ir5SV7-5;^
zs;sOeaP%T0Wyb!!>5sRU;u_-&XgxLnh$-f`v~gUy+*~=|Sm{$*WsqK?3LxB0$h6zv
zXLtOjw`%Xlv^*sx!YGoG<LxM}-dfu^1l)i~#0L6G_povGW)aD#MwWl>0m_K%UzD`_
zo^1KV2TSWX_QDFaasVq@1x$_8+y<=4?(akK0d6V1oib&X#bfl-i<pen9D1IR>iXNM
ziiWznr4W3e{h_I!P_>^2He2(WQM+UwEpYa#>yE@CT4pp#O+p$HGoqWzKh4csvv*5t
ztvk2XI4bM2pv@B7cZW1y5DA_}2J35Rl(zuO@9T?O`cWX~%gPpJH3rfu?_1kThZ>aV
zef1gJ*~v;hcy)XYs-7*EXd-X-pVLFVz7~cc3eO!-C|aOYfI*Tf$?oe<=jtCk?P|uA
z#(zIIz^VCyeBk`5Yl)H5e^{{TiVY#%_l(2$qC~`W;^}E32fsW$fzeh&zrxoRzsO5P
z(tqzw(qH7$5hi9&ate)qnRxkys|VR<uz0z%PI#=RCy0J`H?G*XEOAZY`kJ$y83t*c
zUQOIVd@NI3m|+x<qAE3cx~^&Pd3LpBJN{JS`-$sKNOK!cGc!Z_A`Ty4%|sKjwL$#J
z+iG1!L2Eqk`(0&29OAdqy_?4X8>uSu8e47#OaQ%5NJh3urG#k+xVAuPa4P|{!E2t-
zRjW5WupHw;F6b^R&iPH{X18vtGB0(1cHQe)7!E4*`ox~RkMna^nzLK`#(;17f_jVL
zVK7rkyc`Q#H9dAaZ(j?E_g!b#^+7fHH&zs7srCa6LEnb+8gs{6pK7k%L>lkh(9l2Y
z{~tOsXbSc}SvY?aAX1d(&`$FU&^)QAwp2%Cy1i87B)kS%_^%Z8zQFsOPT(VeUA}Kb
zb~|UBoQz5J!l7P4GP>-DoQX!j;a&dElM?`XWaa=(G-{F#AG;-}D2s@j&lMtu<*dPm
z(Wv1OG)Iig@uc3uXoIq{mdWKYCSeg#fsWimq;lCMX-4@k0SEuPn95nH&TT4NUTIz0
zJkYN$@zunF?OjDKNGfWxVL5xQnwUs?jI~lO4JJ@s!iS8JPp|tE*5IKQ+EjJuM-CWF
zYi{joj>OOw7xK<`voF^gJl-Ewc!y<#k;D6nNDyLutwO9Ue2;6q4Q*{X>xxR0$TB=-
zA?+Ndn2b!B=G--?VY*h*$*gz2_+1D3eEv*FgC(1w;r|*=nSt4ZoaP1DG{lh--&L1h
zN6WnfKf*quN^Vy=Q#2UarCKu@e7iNmVS;C0+ips2%Xv)KsI;ZZv4%Q~Jg2j>2~IEw
z{6>}6JyVV2dq!<OpgUlS@Up>5X3f{AWoH~xegYvJ)KH<!0Qils%yg)2#*kA?Os1gr
z<(BogTj#-LCm7veie!i9RtaAVnneJ&=cc0jjTIdhmLhADR``fyWXwUND@T(4^u!ds
zJ|36V6I3p9#pSz9yN`w$Szq8!7w>52T!{fM(cxX+^?{r5Jik_T)cYUFng+2Hr_&b9
zw2JYE2bp15zLx@Mohy#o7F4mM6>#b9B86Nt85WkK&q`ncDKZrgzQxCZ^+WG3xk>*z
zS(fz@z$yFa&M#oR`(gVn5O|go^*^P__MEncOeaMKvmW!FfdrisBB>i5357WqJG#c2
zv$(W+4h@LJ=fjb>=bc}C&yRVvnHdK9dF+{Ombr*y+9liE5{@q8Ktx<&r3j*enYi+r
zqk@-9v$R^j27`S?);3>aeOP@%y0+OzDZrnZH}0541NLhU_D@S2)qxxu)LsaAt<@QN
ztF%=(uCB$0gh0^iaf{1dNPf^})}zlQRn;fGJIjR>jr`Eto%yM%?Rwn7ZrXX=98>jY
z{LD3HU41s#ep^+{@ETpt&Vk*dymhLqykHWeVzB(v`|`yUtCoTk%yds#esS732tmDq
zIp}2K@YMc)Z4AvHX84Ua4b76HI42BC8H(oa0SjhEn#I<E@5RZ>E}h-%qO50Y{7ygR
z2v4ap+Vc*&lSkzloR@4T!09LvK~K@m+RH0a5GKGB(|`r*hf?+^7-qzaoS(})c6}kH
zY31K#PUKVTtw+b8ApuU$TJJ7Qm8}H-&KKQD2$Q7BYzW8-;@dCIq%VF?(3Tu7vPFLm
z)>Bi8uB1VpZm!X4t5SXzj&A&*-Kef~V#>eBIU*<4{Kk4o<N*oDKasTHB;NZv+g_H*
zxab>eB)5GIykHI&7u7a^nXx5KU|H=KI9Te}jVjhp;eA0HI$PdIp{r|!<9zJLI9B*6
zec)!}W0h^#{w&Urn;Z@oNKR_&qh03DYI478Fk@i*RsX-KbBezqc{|rRW+E@;?Pc~o
zGc1_=fmra!UzCb+8ctd~Wqar8309w)Ij8MWPt1#BHTvCOd*zvoX6F~g@!{oc`=+r1
zLl(UQ?=#>6dv@XuzDlq%mm7O0#x)i%TOLLhm|_qA$eF7qyy;BldjK*`%;G}Hs|+wp
ztl?s3Z~kZpw_(JKyDO?E=I1ZxngZ?&-SQG9vV?a7`|$N&CeTG1+>@CAa~$eU;+im7
zbKg5tttxr!WLYyWb!_?939KP3x=ZO^SyXn<91TB3VH9UtuBhuNiLhR-4)A2JuWbsW
z4hbT?c_ZaxzMK9;zsSzpop)?rZ25b0ro#Xctw!Wpj$Fx7I+|OVTZyQl;l-^09K4D`
z=5@vKp-D$c6St+KknJIK)>{7^Q=jkg6Md%yhx4=HG91W%2lZXf{uZCw=mmiWXwr_O
zFO|TX`qLgC7-?WSwzi|i!|Y$Hhk7!J=*Fq7>67$BA0yxPs!|$EC5qeGA#`-860D#0
znByx<J5}zP{0Kr_zTPa*s@h{NmTsN4kkim)r#FU;F>F=<(wYP;L04)(WOoNT%oH1A
zRQO%Is_N}^Z7!Xzgt<1<^!&2VLSD^4E_;nbguFBDFh<+J^-!!2inBMzu&l+OcE9w_
zmC%5bAws7fueaA|6vUuL41Bev>;X~6Hm)SDl9v|ths%fU)UC_EK0{8)+@A~z;P^AV
z^hX<>+}xbjuu$XLhIAGge5oix8%r}@L1yUIt^Xj_Y<ho6UwcZ~bZQ^1Mvg?ndPj->
zLZ<gUm^fw9&ggW<lgjWGfk*xlXES<ACi`FcFrY+i99nIK0<W%1cu2(3PPDGgoYC-X
z3fA|%wjY-2fjMxzo={bWys3}2A-%HXrcwznb7GbBw#`o8()m49sB^*KZ9i0E66m_Z
zJF<EU8ub~+s_etCh}U?d1*s_KwnlK{r_=EE136=yE-u^eFtS*_k$%<ZUHtZAcquk}
z+NdA9m!E&lsR#U)yB7??+}rKLtLa!Rj~=rsbMqng_+EcZUf4HC(g*&UDJpEFS(k*Q
zD%Bn{IcJAfE&uYMnY^GO5CD+@hbtISb(<1lhaP^-7_ORPX4!KTC7OsoFI;7g?V1R4
z6k#}|52_I62pnzrT#Y+PgF*kGt8mj-B>Jpk#f^L{2K^z^X5Z8saN>4`R+~#ij@rl(
zh2eFA*|g6c{!xdY%~JiP;6S<5hf>R%Y)pL|w;jv>A^|Bzo(=|;%$Yk*>Z8O;S0yti
z#gowD_Ixl1K(_Mn!PUtg!xp3tT9c%b7otv(Mt0qY`T*MS4zm}YmX`qFC2?xIIDlo0
zS)o+}5$WR7MI?KM691dJ`4a+cIhBd2GAYeOgkLYA#&3A!UT8{%6;@YkMu}Lw<<%b@
zU{8wwDM2-TRNPFy(ume3gu1<0pZddI;Ne0a(h5zN*-4O<8MU}{t;9D;F7JN#4v!pL
zTn268jB7iC_vPZrI^;yxJ|r-ZdKVk{i7GU?!vu=CvDqiAuMYvhEVvb^T3vSH=}C<p
zso>F*xS{$GV!I=UMIZY{sV9579r@-zuDbWV@4t_C-JLFptn$N|b%xy1XZx#)X=gET
z|KkOa%i|r;$A9x*C~&$#4NWnAv*T<wVrMst+2@@ms)BDy{d*4>Evp&&&u@}m+5aNN
zR29JgxG9S;5y94KT|!v-Q)XGsx*!-FhAs2XT+_W{RW)y7mMOZ`=<V9EyU~raR%wgL
zNL%LQfwNX58CG6=Ym*Ny6yQ>B@Yq5#9_l5y(Jvoy_7?4r*5Z`6tvB@!`6K0tO1906
z&neB}q~99o5dvN;5i<p-3$pT1;>873Rms*(EV9WM^Rpmmv(qnYX(!oaXD4P-NQ9uw
zJwhuL8<s=Ai*f4ctVsroLFAhaWe?HO*Yp}2AyM7S+V(y^)6hC+{#kskw9o@P+Ang)
zJ-c%aM}ZuG3B%@NW&}E3hW34xA6dS2zN3ngHR`{l*4v1`MMg!EYM?VVo_T+%bk0(e
z2(kC!Efkn*ST**Q6gtg9NHgoG=Iffc(nS_DqeIWqu^A+RVcZ2Yeor6-yE6SMX64)X
z!qZk%-OE-_&|)EMUFsvyp9VDC9&0OK^9~7tijekITCtm<-(C9Y-dzvjp75oW-T_lb
zO*6B!CcG}2MelUoPByf|x=CZDI<82p9JSPgO-J4PM=)f(kihJYL=2G!2R1l>x!a?E
zC_xzz11D7(_#|aMpCypIk?mbRIRX9l+#sx-IvtFo7{L47GEsyZdF_3Yn396kwx`;d
zPBvXvzrMkj2I2Vp-qj&>o(<PSfNqlj2DX~+TIc8RGgo~5@3iM{)4^x6WB|4Z9v%cn
zu!U}B42A0CFhpONhnbx?`CE+@e`~XdwIGkiR0tyha^<d19iXw}Tbikxakl6Ofjiij
z!-GR-F2b%e_nLJ)pd?SUtCaZ@9Xl=>NrC2S+8Y=e1{(!bQ?i0ho~M@NqvL1;d7WXx
zUe9*Ca&aAh#!kNM2~(ZJxwDaTm7L{=545cxEOIfQ6y!s}_9x&UKVo{>tEKE#CW^qS
zgcp8$<P7I)D?$y0@v};+Q%`G12+{jl1@#AT1J5rln7VFmY1ntSV=oaAwDlNiDD^oR
zM75bwb|-;4n5A^LH?^W{xB_GY8hY{+jbChMbu&+R^Qd1`+2&DOeR$5wJ7|~t>ECGe
zuOO8sxS3(k;q~@zi!A8SM=2a;gH#Z&%Fq*@oDAYg=H|FvXk`JqSa`pXgA?>8Vd1qb
z8|{Cl3uAQg#D~gUt+N?l*zn%kpuw@p;&FMx#1$`g3bF#PE928!$uwVlDY9Ge?f_rf
z@j5)|?Q5OWRLajp6Qi7q>%#4JDRpI$Pv0{#utiu76!eyW@k&avcPVzJaw6qRIc*)E
z$1gzmCJ5DfUcHVR?4h>h<&Z~m(|)C?{Z!JPEn7mu-K~jsduiEgYQntCqoU_`-gPQR
zXP5vRPP<)+CGbMwLS0;BUIS9+<g3rp$|xsJ{8)b17}OVlV^GN_(cg`6iTB1}yx+vL
zmGgt}U4|e*Q`)XM_J|Y*Y34!F<C<ncy~CI0RKncaajJiY^K-nvC0zbmZ{e%W#~~`7
z$)zL<qYiWio4~zPU?3wuv)weNy-ir3Nb>z%lXPN%h_S<Q7}$L3ozn*9X@cH26+)9L
zi|EwheYi3`UK|B*4}2^JjM{A8qN9MtqT48@r-w$xKtW1_rKB2ud`B(>1b+`TIgEZG
zV!X(y9v0IFv4sPXwiN57cJk1AKXIpW|8>8p<qLk^-)rm2aWBW5O$ZF+EBqthgx_m8
zU7K>uX+6pGm3HZa?0K`kf~b9f%31kxEwJn^jYusw-A2tdCvh2dI!9;}U}9HZ(=$L;
zdl*_`WlVmV5|q(Pa#OWt%-faKXwAIMQ9aS&WGEt)lg$dW9U{xG^2Wit!=B@lT9lgv
z7dN@$rTCQv<Q~?-Uay-H;)f5z6jht=H1BL&@XuZ9Ew>Z`tCRam(uR(A2LAVp`WJVz
z^ZTkdfrV8E^YKg}$RY#Q-?<Ab4$8QX?4U`I-r{!0o@np@RC}*0uey3-*8`B1=avg>
zIOAe?e3dSi^Vpnf&BKI17{Ajj&VgmCGgbx{>-DHwlYhD|D%OCa>;B2nhGDABJR{#|
zb+hgAxfm^rKJGv|d%&Sxh<^PV;>AJ8q``Oag^loA9~>`0$8WSJVPSzbcbGu#6%>F`
zd)(a8sGY~F*<Z5e!$<ew{U3il3u1WLCSRB=w(!$n+b}5swZa^s+}`D<#JpWlF)M0-
zKT<+pMAz$<)!rp6u)}LwhpZLeIF<e;GQR2V*z&}Q4*(OI$Su>xb^B&>ctrKbh3f%C
zu)E~Jjc}G|cWRElq}t9<Vt3i+4t83z(n+JIbNWDFkn=By21pivCku2b#x1`-19QFb
z`VxqLtwLJ?kv;-oNiL<L1dlfrczsBoDQ=#pkr~V*>sGbwP2=&*(?2!86tOJbRA}*U
z(0jHf?bcRTXM@pgT@vDVmn^|gEukfmQ7`NI40$O`B=~eFtn7^{t_6-Z5>cH~&0$jB
z-n40qylhvymV6z_#f6wxJy<30Rl{*;a4=bP;cPkBSC<|ny0XxSJ!36ZR$7HTG#1^~
z$JTeTtt4(txomPZrgq{HEp?FFXSQ;gf*iDTS?^d2ao`6!W;jF>MflC}?Gaw15na$~
zx$C15wtsH7T5`&D?C=C-e~-XYDhlM<;Ihto?DoMLs*SQwb9ss-P_3f5_)g9qQJyE@
zj{UVM9hV3A)1{Ok?Y~sf|B=;mCvt^pxEg$&qn)q&2E44#r#@E}EDYRV65C88CKw<_
zv;JhqMEm-r=~`*pFusgg>lPADN2Yi*WcZW!wBJU{EV;#Ls1>#dl`R9SZg{ciqjQ-{
z`@L_E!2$euQmE@5_ujiqr50?0N2qkVyA69YG?dzafZs)}9Y2Ql`Zc?gPO>~mvU7Xo
z8fLe%^R>x9ar!s%{=V;E+BpYf&S4knt`IceWA3P|fcfD_DrnLmt~eYUb!|=hT-P97
z&y*AGxJHzD<6CZ5M=#8GW@f1uC8+=)&2?2`Gn^t?tS%}pvr#?nE^%JuCk;N05bj@@
zDGS!~Z&w0LI)GQl<)o?Q<@<m6I4kGB{ek%(lW4v7t?_((P^>2dCY>NdJo!X+_YD&K
z039vu!0j%CG_dO(?oscHy|@$Pd?T$vfE`w8XJbEfV51FnAPkNFOcQsaBZuLAJ4>+m
zC>(0W(Yd@dNXTVbe{4QO(xLSlrq%F<u%D-ne)a|*Xx>qCEP3I(iaWzE-{cY2x-$-K
zba4??i%XcPHBHi)mMP7CN-eb!8q&`Gt!?G9ndj&_YnX)yHk6bT?4++>{=1m9Ha-eF
z5A1a$`lB$oO$@E4%{Y+@HFfaGciJyBztyEC)3F2}`heffujNI6B3}Ho_A|(%Zt?29
z^uLzh-}=b)NBPT(8{7;ce>ifuikBC&$N+KCiBdfIE{SE@CH;Q%`#?%IV?XP;ENIQb
z%ZWE@j|m2CH90>owa@6M-rXNCP!mV0(JL?MLh4IUtAx-<0D76oW5pS~zb~#bzZyT?
z6TQlYQolI^-e>oV?^rP3TE17R1*X|@v9B6{A0e|fK5HFg5HnQxg5!>qR9X4SJ0-SA
zV${n8(-+u%kPprA8zR03Mq}ndQP}j1(Dk8$5I}v^C;+&o+SEbCq!f4x)9t%Y+LMbN
zh0#2v9_xl$7n2Fo-d_U-hVG*VhJ>WuM+G+wYZ5Uw)L%j{T}&6r|LdzUohg14m(eoW
zM4N4p+t>N!Xv!3d3xmm60$<VORfLQd-skS7JoUHIUdvAj=_h`J7~<ypM%EFVMd|^x
zMiAZJcq!)cMn3V0^`z_#OuA)j1Z3BUWqR{;EqI<Fbj?~0uC=z&iiAUOcV5D&$pS;H
zo~1Z!)640^Hrm(>GIrRe0q)Ck1i~Ir(uCs00V@!b>6S}7&NJYJCG?my9Vf4+o<e4;
zT9bd4Lb6*`9tBS{+3%smiM~Bp&f`TU>nE^6t?Dlm`%E5|x3XlOUz4>rnaz*CdB|fo
zXXM19wU0I-{698WO1sXP0VonT>w(fR>r`$zvq}B&XZZ=BafmnYh;HgXCX7}U9oaw9
zB2_BJYxT2+K`VRSP&9n*GM?rHG!u`;uTKI+J=kE-VSq9z0`>rqz}Gba0VY+N#K%xx
zt#=XvS#w9Kp40O<y9R+Hy4J6^M!=m)AU;pQ&vWdpftlhtQP?#bIzj7w>u(1V5%Beq
z$(wt>fl_^~*<`+EGYnK8me!oSR2&r0gw!ohHIp++5{Q#RTJCMlW_MmQ>o!3Lyy~O_
z`xyjQNm#XNK3#J$xi@p0{{9DnORl9`ZMXiJF1@9k8K}aAn;C24+1lDVJ39k?X5K3*
z4#rM{FN6&rP7(+m`TYv?|LlMWpA1LP`5b5D&?1#By4wMXFVv|YNk>@|m1iC~<*kgy
zl*s+dIS2*;$_F{7&ihMoxr#mO<k1^;I;pk-uqtAqrFzwaZ+%y0F!(QFnrHyUEEK_@
zBpB@B@f_W(`p%p1>NU2wHsJ?tzGG#Ev?RS5uIs(2-OQf@;7&~5+db38O&U2&G9P(3
zFbnQ<qxI*cRzKTRAfLy#eraN$z5i8@66QhEY`Fj@cE<5UtyLE;Uun|ozJxN!xj7NG
zTblFhNVAJTrkaEB?cjywJSFab<Ob&f7vm+qu$-J6<3ka<Io<=nY7|#Z0g$&l_5(fP
zgdd6m0XVkd>9g|kXm+uwdUbymwR!=4NQf>|L*i+w7b`#^D{ITFHi1I&%mr<+aNRph
zVif!E0N-Nww~@W)rA@+UoqI6=1*F3TaE`>1UIs}+bOBnT0l_8`64MTaS;&%6AR#+M
zFI8&o`Vmpn@1h_oQBB9PHRs}0XC@o_FsH#cQ;oV8X@Z1!tiehKX+~$C2txI-+_X3u
z64<xampxD^r(bM8!vdOMjQ90vRn?_`vc{`kbQ7phJZ<1i`^(1gm*kO$Q#muE157j>
zSv@c%=#2eX&IgndhMWEQPjZ{>N@V5Muj<Z4+SFXV!uM8a2@82bUQ4@%VVfDKxoS}<
zanuNl3yr{9uch&c4fC=<da@&*$QsI6BNx(b%(S>YooRrT0Vvy+I8Bi+J?&kG2`)4#
zXYVDZtTZ0BxMLZR1q0TZB#u=}SIsipg181fUW@dOA1dw54Q*mqN7+Zai_!tQrqm4=
zL=A$YGV}}LBWK^o*b{cIGH20XQNyL{W*>?TI&L&~Kc{{h!w&mL3Ul+0SW)_I>5J&U
z{_Quz@GORIwRuqXjc4AK=PFjsE-O6~@k80p&XSyA;CdfTY3~5+LzssLu3d+TlxtC5
znO60L1VygD<97%H`uXafkRIX0WCB*R2}j<mzMj~(;lddtfT78sjR<zB#(*{i2+}=(
zyRoQJy(%1?;7#~jO+H-%Cr-a634?|DdNX`|@E<X|u~D_Z%AVZDV*ot5_TrV}P6|40
zN_E*zh#9k&2iLj2jK_#!>sbU|nfr)$PR0f!=Rc(ua)<Txb@^oO&-=60&HGJbISSvG
zNO2C7|9Er+HXx#AG?4_9jmWExR%TbXi*lB(n`x?PE`rCrcT1?iFy_(n5pEjizM+JH
zbey+eH;qJv`%@W=)to9KlDJL@RB&5J_hR$Q14MydENKb=Ug#>nrRrQA4djT8ae>FT
z{7_+sfZ`L#(RvDd8S5H5Z01+8zTM5Ef+@69hv@k9xzTNHfxZabJk3Et0HnFR$Vmab
zFjhrCfkr?W^}NJ<9T0k(0ySoJV=en7ewdTtGwb{XJrp&b&a(d_i^{`k@HpMTVgjWA
zW!DT^UjXQl?NDFVsno*j@UAV#U5~L6&>g7uW~_~Oew+F=HIqNwFPiP{!x-5FD46JD
zSlH~lzZTUH*wu}!Oj}^V*?1B;+B>>?5yd1B^lI-_De<D^bS1nLwV7!*nVI6=^&cku
zl%zVDDb!|nEkp))bSSHe)%^(0d7yA95$Mh`9?iOR?c3-NNU^U1PE8)|?smY$hpBpt
zMATB|+G;?`)a?bU7O&=tghc!7)sTN7nd<NBOE^v#PqEq(y8Fh{zczEtY*3k~O?`j3
zJ)?;CuTY~2^f&OioR{KH>nvxEq(m7NAi=i%`KLxCjmn1u|GdXhK7AyVgB{cd1{zY-
ztl<KZRvsk3Fjzcn%|k8%bFFTJh#^xp8}zKApmWck`4tqP;kg;`1aK1xL%|;t^h|(>
z$|%+o@4V^(ZN@-0T*%!C!l5P;=}g3FrLD;b!y5Reg4Q7!O?$niRs+{V<+2bnAp4;B
zoGC`UElpg!B5r>++kLlgmpS#2qwM$^cw2(yXI2a!#wmzkJLLt%MCJt@a6I>82>^B6
z&d)MkX8+VE9^v>QVp0gR-ThE{5&|n+oOh}Dh<OJaOY&Pb!OIXyJDPoUn`ZY9bI}gc
z(CYD@UJr6?jn4f(8~sVZtOUN0f<Mn*xCRoath1k;Yl`ckCIG5LK>L`S4Bnb3)aKmW
zO|}NHr*J;~Q0QM=yw~L0j0|?={h%CU`_obrx<8NcU8+5iL#n8CJrHA4R}0<>utw=L
zaVQ=htUw#k9EK8YY-uDbBq$1~bPJ7W?+0JP_|L_)HuN88?^=-TkT`vyBKy<SX&@0-
zn1of&=URJgRPP-%Rm>`n2(X*o<*YY-8vM4_{EK>>#CO~anz-^7@(1O>ZX7@NmM|js
zyZXnlusy~%nxD!lSdAXOb@UUHulW2{Ij;F9v<jpK>RmLzPb@NgE*K1Affb*(!c#D?
z^o_maa#j^V;GD!h8XC2tB2|QmNmP-?T4W&?muZ|?q&nbn0}cuX0T2d?+376}{cN-0
zMh4c1uVZTn-f9=XbQ15)(ve?!(opn)L2}_;*%c;kJe8FMe_Y=4gK1D9eRs;=zom0K
z-o0yM9t%Dl{P`oFzwEN;>l2^>3l&kwIWf4g3&3bhCEbCv^1L@~P?<dT+YUC6WBnQ~
zLj@A9i~~&Gd?B!dS=bM6ytSHd=r-mb5m>dQ(rS#Jl{F_jd!6O*X^`k3|8C(}=^x|r
zPEg-IihYY{L4EQl7Ym82^v1bt&G-aup1NTMNIfaCe;mSF&Vtu&ER!QEx9(w_nI*}`
z2f+m9i)k9N-43NPP>xV!J|_{R4~y;q+rnvTRRmdAyFnCj_Y3ag5P)$;AOBDi8ioC%
z7@i#u@BLJsocz*nkp!FJ84^Q`K#n|$3CE2qPZ^Iht3ifnN6iOs31iGKal8y!ne~No
zvr<RuyQ?nM+Rbi@ha!>n@Yv<%r2MkXe}pIom#Mj0Cy$lSbPFeSgH(T<*^1Fd&*3Ew
z9IonEra37!@@o0N1j*3a_6!0q<V&4(7I2V3m`qG@=w*Ojn*#`un5Wlkr#TG9D!xsX
zc$Fd|(iBfJZn8Q;zn}oF+IQvz<zC#361Hh+oG3+^qGYR%)4)X&Yvp>L)J{2WAiT)X
zloa%b7PQ@8y;Op~V7}lmrERFJ2%nBSN%95;xF|k}lc|*=B|W+xlAG;N{&Xb$xZUXg
z^4&iZoA!^^)?t`EmE(V9cb;o&EmEu%K>N8gf=x{>Umc?MVmzl>&~SO7O|v+loH#jU
zr{>q`n1A^~L?g1f88n1rr&+aW%*n8BEF>iUAvU04?^IMSqGO^6ML6RwwMY8d#5@b3
zs&#=4Kw!63fmMC&tmAgkLe{w8E*}rqjrN5U9}dao+*h%*Zazs+JlRY~nqBNR^J{&o
z-myXekl%*iRcd1%9>$uFUj8c~YiDF)N@6!g6mZ?+J<*YoL5rhR;Q#VhtoVJNQ8jfH
z9`_=Sv4?OhMWG1rS9Pp7qSbN|T73t6^{bt&gcdPL-Y8`!Drdx^s@OP*w33}I#{PW>
ziuNegBA@%#JKfJU65s45?p_T`Jnt`aKi~@*ROEAEhGX^A)ze6oN6fr~$4i4P;D8_i
zQ8;}y-(dO-zg5ZiZlc9{^nHYomp6A5Uv&5k>(&l5q-#~R?9%OD*)#&M>(S^vmEnKN
z0{QPD5Tgs`2=&p3H{Vpv&8NA#fU{Bb(K;?G{~O|<;L9APFhVX~qY0o&E1}p9DU3R$
z=qcjK+#wGdiqNa4r^g<;%Pj~$3I#X#S?iw72r$33<+QX<*!xkURQ2#+0|(>lO;_{*
zUc2UM&2)j)`{6ife)VP2&c_I!+ta*)hM2NdS8>zt%z{Kh^Kt^2|BOpSU#qI%w+Hoy
z-9!A*e|^^|INsIG<Z@xfQYW$5hbfsAhj-%NKfhr8B{CX8$Qml$T#1}2j3gOJQMK&S
zFsT>d(y7jAw*LOK*WJvDt=SK8hFZAIm{85!!a6estPUW3wsU0-l`meKyPVw2&v`d=
ziW^Zj#&ra*yn^;;Y0kpmuQ?jMCX|+jD*_cio?d3Y+HSujK97O7P@;1atlEBjv~+nT
zzUn?z#WE{(LcqfCi1#nwrvKt$F0h300!=kFHE|c0^KAp@pXmB<zEg0t5>aZCRLQ4T
zzLu5N%T56+x7>GFIbSp34+MM1u(RCsW0ph;XZ-vkQ(SzZ3g2w8O#<^P4DR_rf6uJX
zncO2L$c6vP&j-KP24K1I^|dMprnt#o4q*nSCK}&etrq9jypPh*q=T7!4n_v&sk@R9
zcAL^t9|B5jQISJwPmi=c*Fy5i*rL}lL12hovRsZT=gI3oYKw!P)I(FT<X`ao&&LjO
zsW@lRUm<28w2K;TjM%=#LIgwP3V~Ul(FY9!J?s50vSWQ_7Qw|{i2ED`8PC>1=!W@J
z-FR!p!`jL)rTcdk9jYioB)~E_<=|g1<_FrwagZB|o+@kqTB2kgbbcnS0Th*~d(hyb
z9H51$!#o|cRJ*W3<i#PPWBs8<+m(Xk%yB&^fDSoI2zeBZ+m4Gkrj4WfiFnv}bcN}8
zQt&K<>FCN^b?|>3bW!ucR2fmxfWHCwtRp`S!-+cxVWyEdM<LGG@bL)eDD6;sdID{9
zl8(CFgd++We11NPRr+1L?MI`R@x3Qnksd2Sk|w4GEY;Bn$-=ApXfl19n;Q%?LR62k
zsFroEl}Ch2Ozm_%BNbWW!bIzY%J5Z_La_=#L3>`uHNlRgq~(J*X^vtGX}q1|(m&ku
zf0>j&>4P4Y07W<--?sageP?mdK6$vUT-l$$xve7twi}4#f0&+^S-{CzSy>rt#QgIN
z?g!QpOEk$)QR%HG+c4NthsOZQV+k0yfQ40R5p6Rtfcs7xj%@Yi3*e~)Ms<aGf0ynT
zPT(q>tew4)8f09M5@gM=vW)TB2V!~ocOl!_U$;=GFR7do&S%SHcRj>v^&P-?=Gxbx
z&PaSj%KJq3Vr?{D1fk%k?G{ct<~BDloy?*2X{iIH_s!VR;v)+SSD4P#!@cSayys4T
zfEwu^kTN?|U02tuF5xS%jK)Vkl2O!=ZEtIT6Yb&Y>&FT=V8(_1X4+=+=A4(cC_)J7
zFGburiSVqqvQ!Kaa4PNcBKLjpN)4T&C+ET7itEZY-I%uLoZ=AI^!#V8ZN=22te0J`
zxP9Ndk7;$wFTT8@3D2qtLgv{MK>v~Nb;x(xl_!Qkk=FsY3GfgZQv0Q<v6=3bxv3&_
zUU2E7pEr^6>k=53BS*Cym@}f{mXhQ+9sTpbnicisOV-&WbfU!C+n^tB3Tj!s(=!Th
zh^EzJr?oHV0uZ(i-3-|boR`OY;)X@wR8^W$DmOPT8zAbjeHuC}tf}HvN1V)<J>e3Q
z3yzl04EPX-$n%W#c*N{36zUZ<E8W<<myn9v>^*1h?;(6wx6NTC-cQp#I@7ulot%u~
zHy@gCp=gxB&&8ho`Ov%nP^5GIeIXF|`|-u5NAdW#O?_dIfXC?h|DozDprVYrwM7I$
z%Af=!6zT3pK%~37yBnkg1!U+Nx}}*R2N=2=hVD+0?w0@6?|%3Cuf<|9i&^uYIcM)@
z$9eYN_&#woD0UG04J}?<x)4<i<YH%x&Q+tC&y0ns_TLzFT2HUwZ4Y6YLrYNwjN{`6
ztoQd!K&`sRq(x6;TMP#0x2=Bmy85ZSx`WR_$_e;koAb(>(F~D`%@!7V-#vZ{knla}
z3E2(R_uB;4op?TDVsF2|-Mq@BY=HBJW{d??M|mvmysRk5bT4}MMS12zzSmqY4`6!z
zXXSsF|L2#gXeD_a2aOCBx(4(;w7<D11k*=`ZFP=*cnDsGA;#wDB@Yn0`l%U#YWL+e
zF_U)_6N+)f8<)ao4C{z=!+CrSPELV6x;|7?T}QXIZy06Ig2n>7jPrC$DgTrZ(7iU|
zMbqJn!lY|q)`;CaW#QbUTfDhh+3X-Sp|1r78+7?RbaR*78xz)N#6Ro^x5d0WSeuym
z;KB|R*iY+8xL!7;HJwv?4YtHtygD?TQ*YSTmQgdp6y#9cvGBq>_MagBpR@R%Z`6rA
zK#BM-7l19#?LDOqU)<nnlFcv580c#Mxu(O&0?Ds-G^2{IaNZ~%Kc^4s2WW-yHbauU
zRrHqvr6n$m<ieM{Z2q^k0{6+w+sw>M1O5*M?!y@_v#JtL8xMWK9tQz)j!&?0+B5Hy
zy$`UsTLmvasNp@D*2e%^O7r?^zvxK-hb0cguFdSzW9lh#ac>7!wuxxKFFeI!=eZX@
ztj8(jhCBs&db%IF?=u(-O~lJw_9HE|e+|jD$@4s+BoRJ1_(MfOs;>u{o`Gr?+bSm9
zK^@Vunsf7d!Kif^VE>x}$FsuuGqovaB~iBjisM?bxz^Ev;!a8Zp6?W0d}Y}!a=UcG
z@>eH>jw2$_q+;VzCW;=A+V5sVm&|0buaVi=O>u-V6u8^$YT-S1<R@M;4T+%>d^b-8
zsHk*DR2k|WG3Z`LPDK|v>#Q&E0xUj)tBHxaFTI3GN#y561}g;ivwaE=YdzQN<Cm;2
z%smgi89Xk|a!5VVNWGpkUkYm0Vjncdh1Msi?sn!l=A{`Wgbjr|iOdTgI!~HkKlWM0
z7<~F!8s5*o?4gGhCso^J`chzVaEKUM0IMMooWOaCQ*U@6-ToY?vIe^REPR2nl&Ur+
z$jtTcq$?kgZ_-sp5Qo~&cdT}vbHJXsx-n|q_P&^@MFfF1AK6<yuOF0>r21b?ot!Xh
zuEl7Fj5+KDE+-0yLYT2C>8gYlOcqZEkaRYM6vMC~{XK3H_W=m$KceLDDr1{2PMJ3G
zCSF)PH=elhd?SvDq+t_G#z4vMNAgJ`x!?VyeMenB&2@HYZ!a&wbmb&_Xfe)J=nT(u
z=RF~jGm?_nR~uqY?3iaLXDSMc{#WFg&8uY#CcjnTtveI68bfZ|>EAqBQeA8g$zjC_
zw#H7?Y(fY2*8;B99hnb7d3mpE?A;1Tf{(_@0+{`^jy()qz4Zy>Q37s%c|8Ix*uEj2
zvMYdFyt1~AO(=8M>qnZIgV`U(#WC?RTxN&oNVjf!F~Leo0FSgC!aV{|eQ&@|-n>b?
z{-VOG8*Y;L1ni*k0=@6Q8}z+IJI{Imyd<b>EuyJT>XxmV$0n;L3dt)`UAojJah*lk
zv^Ws*o(*K>G%%wQNd~meN$tqD5bD1F^i`;|w3>W?GVQd&5qoC9`R?kr-$4Y>bFpA$
zf!DmODY*%`#M3V`zpU(KEMC~bxfDDJ(qx{3aDR9|&pe#%EN*G-oS<^uIB)Y@^@cPu
z0{)yPGbGU<*mf#i#tnvXF$MW5c(`!}TAYy$2BW%3u}-FYv=%3T4Q`k5{Lezb4Tns4
zy1uIzj(dbam#B^mL6B<yN$(|YQ`5mnkf>RBDvSxyzisrH2=yuF<fN5-)BE#Z)7_`K
z^PLCw1UCs^+42?;7$v<vEiYps+9*1;l{pP5b9~;{*q~(0Y8YHSaudnX#bwN1>3>mk
zpl?k9HIcrlRPVW(mW(O+GYUFQke5uh;rF*;$%jyrV`TBS*;4tOevcc!IcI*ZNx+F<
zel|ym^9n#;sJ+Oe^K&49@r&Rm+>r`Rw>NgE-m#ldvtsQW@>=}h;;7@O_uY(ML-LXr
z&aM39>BKk`!_dEFAn?SiwtBm5Yv|})m{<Hhn_AM{;mLB(4QX+-ht=F5rmSpTbYXG7
zpp*UZVxlxWtEhpPh<K3~R@0s@x8blT*350V8$94}nPRiBzj#=aXI&mE7P2yHv28%P
z=)-;&^6;qTI+aCW&zXAJBS0HhglYTPOmK*v;zc$N+~u>}<RlZ?;N}8i^Gm6~UFqQE
zqSw^4m<KmOJATTm&rVx-;xla&6SG2B7qxmvS~@!D^X*6H_gIPO|H_zd-o4^`l2nJp
zLcL<Vu3%l#{2r|JdvMGt>pL-&d@8Ii0Yf~Iv1rJ-rR}lRj;vzU<I0GXU)FM2GMvWv
z;T4t+B9Aq!yMERfy=VWsl~jE{Y@}vJSLaT#u;884aZW8dRo$@1y0zF1UIR8Q*hv||
zw}&1`8hV0VQNC$yrrb}lf6PAQ45SZdxP>IA6swv*bayIqw&iO_oH}|4UkAi<l^!yU
zfHPWCi99G~&#Gmn>+5wbEilt~kpA|zV4^H3adBxNA`aCk+ngnMxd*?8h14)T>tzz7
zy}EwrTDt`y{O}{+d<BSZQ2^uMbWx@t5xdp@)&ay5yc0?o{`im+nxf3qYi$q3G-K~Z
zF%1g5)rJBoJKe0Qx&C4-sa^(sy0ow)?*o?ETyH%aX@rZ2*7LeUNK^4v?7x8KhF_y;
zI=aAWLU))>WA-|g1@lb&><v1vA4|^V$H60U8d>e?P$kW3bsAiJF<a5nu|m@fa6-8=
z#C&n<i`zW%T)l6lmfT7J>$4JF#r%X+Ev~YKNTszu*2MUlUc=CCc1^h?j9?KXa<T+q
zcWD^I`~-ubu(QbNs!$E|RC9|(%+g5goX7;(DL0pT;i1vQ)t^e-z||VU?SYugLjis$
zG~IQ)?6**9j`e|)ogK{VCxt|mJ9ecr#^K7j(u1N(B1#B9qTvpMAE53DFGc<KAd_(O
z7(%G4G~?{xDk+}UKd$6nvk*4F?v+@n^^-X+6Sjt5%=P=)Otoq>uXSe&-b-d&HqvL+
zTtBN?D0K^6?fiz(SLzl3HURm7#0j2(4es8@F+H-i%NwZ&={AFcQ@UhOIcWnM>;s^D
z8dG3qwN<k!rXbmZ_mqEx=`Rv?&M%JPhZ#nc`~(>%WrPFzxpsu{q5xYQGc7bZm17XU
z?FKX_FDD0kn;uNANRx;5u492=P`2ajhuK~Z1-I=_e$$x;)81e^q}|$^(z%~128ZxV
z3nuoI!C{R!DaAkbh{3xkUZ#kMSkI>Yf)rK1vA5s<l60`2G^ZOs{|;6DGVkZF&*$Y<
zSsq}YUB)C;Je43y0{yWx-u1wWW+5K`=9kf3tP9>acjpFLzjg`ARP?YOb`cFG`^#lN
zcfabY#D07$MbfZWxG?8<SVXE%G}i@Pr=JB^a&Q5J_xJlEV>B;mV$1ONno>@(ss|*g
zRxbT6$J}ad0UI;R#s&;qQKKw2FuOl?0xT(t2_qx)22wM%SX`o~$;H1L$mS@bA@Q`j
zI!gs9^bAjWIk0u;?SQ#nlBT^HFXv|Qq<)F1{p*-$+5aMC9Z1tgNt*Gh88>?N#PZp|
z$cMqu4vO1Mnv?!hig)+;2#wOcW6j0=(Fusv&xwf%`Z22@vPaKyM(U$&+u2wR3HRC0
zGDG2>Tw6b7#D_^hKl!(+_Aj33)KXhopG&G);pR<+o$eD1EX#=hF-a8*;&g5C%k8Sg
z$;We8K2_EPw~sqo$`lk!IL&28g$pI0fZy=z`!bg00aU%w@JRSHlZpxp&(E!>^UDVO
zuvf3_8PqMP?|KYcILLLZchv{(FX^|o7HqP0hs|UNnCoZF6$D~d+Jh2EQUd4gRcsf~
zAP`VK)pSC6E^;w@nP2$8ul*pv<z8JSUhyhFpPTk0SrILtETNw?>B>p1;1a~z-?z_C
zIcx~CgxJu+efg_lFV0Bc9WjRW&-8}}%xT-Yu+v2@rd&7PrVOajO+p`QAd>&Pm(O#6
zO}svYzz9{I<>2b_5S%nUJrVeA7?XQzqV8}s(@ftM{02#I-ZzDqbMm#^%6f*0({7Qt
z!S~Yl+%~UYv!w}=y6lC%eR>%z;Nu^&-K?_##i5!#ZR6OjrIxgcs4(~AN=|WTsUshu
zKh%26{s#!Ok_S5aR6rW~N&>eG0HK5Q)hm{;-I|%1pP;m*hvsSh61e#BDvr+YpH-k^
zCvHAlOEkb@Oipk77_bsR1Zm9ICUCPpC41ljIurQ(KDDpW<if}ZIiz>ruW3AsHTER@
z^hwuzVtU&=Fe1F}h4uYCFw|w>i#pWx_4tA=?K+>waGtYu-)=lhCn?`IYb&SIA^c#8
z&2MEzeQ(HPJma7|Or1nz9ux~;#-y81;TNr%swCN>++ngq!^RHHc;kzJi~AF&5odB~
zOWwq^bI5r2mop<F5<9Qh@x!gfomzT0v?ZZPi%da`N=2Lv*>CENP;Oa<hV4Hwp2iCs
zf|~t`(*(1)SHZ6&!hE)6ys(7{lo(p|gp*CjBcGITU2)WY{!xPGXJzSw3#T<b9iUYV
z12yBSNr`)>=48uP6-!iIm}kCY{zkG)c5*Zg3Rh^gJFHo_8h)ZC6*QG_u4+|r@*4`~
z)&{`gyT8?m7mx28itOzLe2)sWa92+LC1cWY9qYo);$t(=Iay8P-t9qS@iPzyyQHML
z*x{?tCI$>y0Mrc|bRI5LvTL-xC9G7gHs<@6XJ^rpbs@TqKkbo`p>FFlN%YN?Q|UcN
z<vRRy5e)3zW(NC9g!+V{3W%JhZdO?VulVl6`x==8p$9Jlp3T!E;9jE*nDAln;CA&@
z*wcl{Gkc}I)hEMpz}?)sJ}Cnot)rho+h&n>LyQoc`y_g4sMNvBqf|35JDVw|f>XkA
z>jy?Zv;o_POh<31XxBfm0)6PZ2XA~FYO@Z)0~OQN9_%qJ_}$_)NHXy+6$sz&C_EH$
z{|ATvYz5iE+L!R_57x#DNBv<GQ45(C4{;QsKg<V)kZN0)yKeT8AnBx&lSB&eepZ`>
zRMAsAB0H~qtxb;_RW<x8kZZ!*`5#ug*+mW~l0P`b?!l`V!F@_%bh*7#HnFrp`bohF
z+DvqAeau0BfI~Jss^b*VA;S#E{t9eup;Q4yTmV@M;w0np%an$uq5G6CBQ;CcTfT&>
zrZ<r!YH0#HQ)wB{Z>F;h3r<fPckPeb1poy(BJf|$$drbc1wE#W8=33ZR2^t0031;!
zN@61CAXOC=CInrG%%+YP|E7++FTL9&E8S~`l%70a9N6dgL2=ts0(WwWM^;z{H!tsw
z$r_MxpFxE;^5vic`yhffH#9NNRg`c77MCIHn;iMb^b=l}?jXPW`-8ii8(m4bzwq6-
zaNE6sTBGZSQQ@ql0ebUOQm*8jdR{rl6B=t<%#@Uo&_@UjpZio$2w8AP$RbRkK(BXZ
z`S5o)zq_vB8-lXfabPq4a|BC%_`&L%Qz3~}XF{;*1Vjm0P*+=Q*W?%Yx<tiVCq0G2
zol8c~?`+AMd6S+dboUl|scCC68;VGBhCtSfSSHahB66lWgf_qMb~s^mV6e8_J{IYI
z39>a2XOytwN*=Y?GhUfu=?W*E4|eK_?fkpl@d-N2#WL7`9{=5p{7>l2_bU+BME47j
z)DS*?<#?N+;dL@wb!6}x!**sR<7`BX>D%Wckzc#38vb2532TQBZE8MU6$4524XYM@
z-7f`h-T>6FKXI$%`#41%3@ogJ0HCs-L|DH1x{1X9SY18?!NR^i$F=1Qoh!a;%i?kO
znmViA4KIXY3<@Cb)8;a=#N@KHQwY)nH4@EE^vK@2?LGQ^co#`8NK6!%oZP>LjC=$~
zz^B4?r~Um2>J`7J6eJCouISdstus^e`pz*5^6S74L{@%N0a8D|9{P$2jRQ53)GaDG
z%{9@+UFFR#_oB4kq2B~DPU{f2%Hn88FmMU&BLZM5oaJPL_VV)0YbOyc5`X`KBX>%Q
zKPK+);T+>@5xF;MU!8XcR-oXc%5Glkci)qUJ6`C<Nc<K;oGznFaeFl$F8`SIllgG^
zEUM6u^$UzieS8kZ{&ZJU#;bWYFn!p+zg_@}xj@Vu6e#YkXi!$S-uz^TL*N4TR>7Z|
za-Jt_J8!m`T`CK2s<^+79aaa){quNm6pL&2z+gB=WviLm<mAXCditBPla29$MGSD6
z|6(5B{XNumXW9>_ee|%h4m7yNlR32N0vtAd2hR-pe(lql2MqtKOg(AbVCX_o#S3j6
zJ;lk96~9D8)k$B=`u4d<v->Ueb!v;!;1GO}7DT7R+Rez$zZ<CpeM^roBW~r#=>e&i
z+&?yPT#tYnV-aLQG&RD0S-bR*l0xGfMcxO{Upn3iz0M^(;q%I_4JN>`luvWa9YcLe
z@9hqGvFNqoe89!KGdx6i>1&XQ@hBJV#7lfN<r5clS=?n>YhvViUk2$x6Tz=t#zj4*
zX<{PvG~q+0^RBQ2dlU`ie}b7>MYyOe?VkxD)Cw6Jy)o}htF)l-Eu#8{26Sn$H~pvZ
zrRQdCbftkOm~4uK7`g9)4}Z&Ld{y|QI%#Qt>j<*rf62i?S6b28eU7gj-y&M{WN5O0
z(UM=)iq{qHA+D>Iq{dW^;UhWerQ1RuiiOAi6-<zc-N$A9i<rZP`~6~ow74TQc8~6J
ztvDbQR|J$GT|##5?e^D>Mnh1&BdAAH_rvlu<SWp=TSMPzM@A*Cgz0v^CCLBk=}c2<
zH;#W2V%OmW9!#xgY=#*c!Z5j<9Mp&IF=9F<SeuK?$Da)uLNTA5eE#GokNNx{W~#L?
z;!R6)NQ>X+H`MYp`kvV@_l6M1bT9)<u|y!9tu=hnc6~~kd#H;LJp7q6ppYeeEj80X
z;CG#<#N2{uF<kdM*2N{klrfSXeX{cjKzu59hhQ8oWfh4AItZk1tcK``1JfI9oA9~p
zzZu-2$mpcb-#YvTuO0X3!4@Y}H?I$n<)FWq)7|mRHh;m7L+OCQ6$%=+xRcvYkN9+E
zf$277ijLXZOxyBCK8+JnD|8h9>WdR2Iz7c~>k7Go;5~=I!?r(g)f|N5`%{dAgG154
zz`xJLVEcvi!<_P#X$5L^rK~(*Bq_tm;55d!7~}cjFX@+*R`#PbC{Ct|d4_{bdKf{<
z_y`v@lg+1ExKuTHHQ}HjadYW_sb?k(@H7!5tcVo5+9$G6gw5BNtRWdumiuTsnt0K&
z_4pA};K+(<n;(BwDTv3snQegZGKD-La2Q)mRQ!Wu*LFsdz1jZ3ykc@Vy_2<#%g%7`
zcNtUpv@grIf^3|erqEsYr?j}bvvZQ!%Gy&iXo7-FUw-Ym;^|>wF;U9oEtr7BUlK4^
zx~!${-Q@|0)hCEC>j)hm8d1rMaqa5Ir3ptBAPelrO{Cxi;W%#htK6-fW*T3)28Q_}
zWnNud5jiwX!qn~tT3p4(`XQtkC>|nPTXcEW7&2qQ!3(4WgIwN%pCt4w`q(X2xcxfD
zCec&5#pH(8TDqc^Ewo_Igf7^RGKF&i(AD@s5S$<lKX#f1zgVd0?pYmtE$8d|hTj2V
zvY$iXWq-R_TwO$4^iS+{xeDBr?=3bY9{OE2aBy)|I<0@Iv0vmPi)~`f>q#s|l%im2
zny?ZLpMMfKR}5%wPSU%~oAEM;jbzF+UNQ{t8h^G=%T3qu#p%WT-A8r^uCxvaMFtoC
zh#^LoXK;z>C%kZkC_JbLy+hIeqY&Bn6sh9!-tq7QZClV%l7>FRNoOs_ULC!ZO$k*x
zwAxKI55$As|Hl|@zK_&HQ6W^P3l9(I%YF-I&o!N@QPcmxP+|Xd>t_u1ltbu9ayUk9
ze!eam5GCO=;-KnymLEIu_$}lKYj?n=>+!z}3ff2HQ`ySlrxpd9f+TN0zL;^NSQ>ot
zJM7ky@BGnWTE@nRL_~=t(5oSl@P7DEvF_s2UcH+oiuU!0BL*|3e?sDI>a{}D-fEwY
zwU>cH6n!G#QrxDc0y-rPXJJTmF-UMW(`m**2>z*x&AfWD;OG6m=<}PvM>P&#g>P*$
zE^LSO;)<=aUe<V#Ha^@RpaRZ{{q;x)-n(}R?KPrpdUK)Z9y=g$jRbWS<U?PaQoV*1
zPAkG3`OJ5qDHh%x(aN12?h}Ka`ON>jMd-wl(qwoUL@}k^&hi8Ji;*PwIGCXjscKW<
ztbWDDhiYpgJ33C`6Gzz@d~55=!8xgt>5;%lv#8bga0=is-aeyXuoJftuR!ZDpmRR^
z(Ax`Cq^a9B&}LbPeelhdV684xP8HAla1F|34%+iA?G5M=J7EV;>WI4EX|_TeVr^Ua
z+m6bORy+r*a5iU8GboFg$TSAb>F~_{m`Y9c{x%t0HM(%Xsh&z)yGnjC|Ep(I4okzw
z&n?2Zw|cx@9lnvDR+m69@%b3CfZ=5P`1di{;#;zNn*6*39vzAJ;%q|GBn)3rjD(Xk
zyWl!q1r@bO303iKL*X)~gc}{zYr<Hk$c>rj#g?Nj$G5s?4ZMPy;s#~eMHKlIiX66}
zK3m2kVQzxHA0~-(5G~}PAtrMAL`yR^dj)oOCYn><!ZHsvXCTU+UIki*s^RS|T-TSm
zcyRmN)V{qseRv8~=hoN;DERink{Tc9Ef|ekB?=N%@zALB2bpp&p&i~j4b<4l;Pd!r
z6#0N$wigOvq3Hjr6i${dtGa#JRX2Svm`1~8w*OXgTfURTo7UWP#(BUy?#%YAsL08s
z_n8;rLDcIv-}^Q(C$;2MQWeC)Bv$CFo%dfNGC<eEiJ&xo>pr;vdF}4cVH?YknR*$e
zxMMUE_GLOJ>oLcb<7k2wJ3^iVBM|5=NhCFCYbt4y*mueSolubY^oskL?F`XK-Q&R4
z;mqK_z?Vr=tV)|8Tn7|+)_g&bKrA-91vDaibyufnEFY^(N7`E!GCYBDNP83Jp_@dB
zzBmydIx97Als-<FU=qMPZSGcAMoJ>t2-YQ3w@1=qsL#AS48b5E&(!eL2f6Zz({0wX
z*3^p0om+>yB~MfP%_=YBs7<EGc5iS0jxfIeT(_RC=5zEFj7R^aL%jAH4DM&C+Ohii
z%Q@_0j16&eA3&rpw<9jqTy2M$j-_uCtI@PB->1uzxbHjfHU`gWYXqtLLDWcN@WlZd
zbLQQT&4UK~TDvEP&au5Wd;5ngy)U@g>|H0!{Qn6g*KU<Fxbx#vS{R!S&$Vt2zTM6e
z^0=^UEl7Qprmb0eS)Q}lp~9k3eMTC^duQ95-0{L?JlaaJbkTms$K}H%T@-myQZk6w
zon@o-UGOXgPI$;@aB@KO+|>7rb5c1hp;3lY7domQtO5fr%*My{&sXS&CW4}u2(Ut?
zB^tMLk!;G&c`ssn=n;6l%fwaIP1(((xeM?LXVTcDKNp$<;v~%WFtmg1RU=Qo76j+M
zmpei5v3LDqffxA5w|A;tmNkGAsk@$YU>9FZT*$!c1rOh^$8}BKtE(ZY$G$&@+Bm1{
zKt;z+iE^8#_V?tDT2loNI5Ki_?^kGBE(l434esOD*2!x3W|1>cpP|fsz8rPmc;rcn
z^UwpkatcuW6n8yllvuy$%U!AyB9$*=S7!3vST*|*zqwL_d({I@b5F1!#PpyZ&w97K
z2)829bSj^=m~yDh&56vj6~$!Q0H@_D?4UCxWd)hc&=_$mmA5O|)IZ*Hk#1~ji)B=Z
zpI!@te&0v@c95~#H|x%QBM90N?5OCT_uP6tJzLqboPKc`x8IchPl~lMXrZxOI_unA
zBUY=VeCkW(LEJpbK#+#$iAiuICu#bwV6io>5~j*4@Sz_kqWQ2B^OA1tY9o}#?MJWd
z@W@#VZc%m5S^Vl?-}ADcF(XLQ-ZYqStASQt!g|Y@nxR<wLgyF6LA>8BzT}8{kxJ{t
zpZ6kKS6Ic8YO6~pT^&w&uF5wTj-#{fgw7-UcI1kkdDO9c`=`bKre@amZLK8DAtlNw
z8Nn(AE2Cm^)n!suTs4UUy$ddmNSaIqx-9Chwaz-Ny2B7|wzc+@IO^3SBaO+8TgR0z
z^AaxXXBYM45KY%)T^IJd?lK+qvgW&rj_1%8>?L@QnjX^N(0eaeYl1F3?y4f+*vJW*
zw$OK!`Ba9Be6|_{=4LqKbTanpDh%~5OoxJrU&rrvuj5Af>waVU7=9eZ{2^9b06cGj
z_o|l6;E1nW(-qy|dR)x^e)><iq*w|Wdb3i$3vh$HO0PRqXU(AoDbZRwJHADn%nFx|
zTj6gLT|CUHpfc;*zdOf99%0GI+=T}OAngre7HAx-#z>5j@I3PMwYS$|Lw`7L(YCgR
zW=vcBqin(-%haVV?jaVe-H=zo!D08oRlfKTj?|hb6O&dx2CV!hgw2LLr-6Aq8*x|y
zxX+fY&7Ydcld)mqXj<DiG7{==P_TAMW|tG19lvCq<&&#fYPt+8YK$5mDJ|UnmbDa3
z9EKL2WBE?CKYIU6k(7hWTD|O|2!hVzJ?D6wX0#659C0+_sbOLGlUVMY*AhBxFAg$M
zk8s+@4hhlwGTGE_A%4#Rqu*uHBL~q$4zsewnz>ewQB^29E^j;A--n{tF0GKK2^$ny
zEBB*Vva{RQOp*N#cz<rasU~BI1+h!`f~hM$*L3;#>MX;4V|Iz^<vwv?8O*c_f8VN`
z8T?hS7IC6kS;0Cz<#ede<0<KM5<h=A1%bXY(IGC?!8mO8=z9tb#UlAV+2Tw+v49y7
z0nJ-0Ry(+@=bN6ZC<vM7o)mH1fzR}^oDb5^WE%yWmt1fCCg$<w8`>+~J-eC}uDsh&
z(BFD1D_=js3PNMXcJaG^BXA;{vYp`}vVBzEFa^{TGuG<npV+>R`@ekf6~(Q{*NQQ^
zFX@^0*|@kEES^HfhAiDOG<@n_KZ-eMF1tVBYHc}#>mCJ6X0;K02Z0@j>x~Xjtt(SF
ziG9`W8QDDfsDv#>kX+AN=WGOtbU3a>cUtpIENnP++;0bm#nV31cH(lJPX;IG;0lVE
zeQ^9Q7eGfa@KLOlxJ!vdzSOw?5}PLlRmOwrb-T@5MgkxM^nLTK4swIL_S6{-seikg
zcS6ZPeu180^@C$<`n*Mv99Dg`dBBY2ur3;pQ#YRS&~W_<ghU`gtg=tagZPb^kSu)*
zHXZ(=OQqI`$-9xIv-->Bkw;`VD+%<s-s)lq>w%v%0@t{>VMHbi8Rq5-zwy-@A|mUr
z^?<m@#t?$>%!QZW$Zq!W?wqsZ+T&7#?U#kr8DC6#bjjTp-3~<3(?`Zcb>}~%^*?_q
zWy==PuNrd#_~^<ff}m-AL;$km4aW@b-hvhU7*!Mg>|#&r<ch{Q&&y)?Ibx7iZE?70
z_>I7WEvV^s0gU=|5jJeP(tK+;-d$n?F8#-stK540-iCsq^+>CuQ&m<aOhT-o8%Otf
zzFU!~Z5|c3HLZN7A+A%q;<TmJf#QTI-qZ)R1mpXyFMVX(X#%!MgjJNvjzd^N(+jwY
z%PUF@bK+w2ucU@<c75}mS>xDQyLiR6`G&Sq1-R@l(H|cPipw=ojb{@QhrT9e`6e#&
zwup&6PIiv**&l1vHdG5FL^}3LH!mMP9836bwC6a_w~Z=skN+wvz_Y36Y+Ez>u^~3J
zjws-D2?m<aU*E0i@G{UpHLt2l=;t(9Faa{lljA{R>!s3Og>$9+!wH`Sz)z5>Yg;b7
zxVYMH{YaY689g_=p`cQd*n~?h?0zaLL+k#1h(P8Lp@)GG4t;Sp+&ef0{`Kbq$K6h*
zvHfD`_27e4ofGl9E+X?B%4;4Gp)zy}-6P9zjFK9M1|L|O^2ovyVfXFb+j+caj>n}v
zw&f8qu2?;h*4Sk3`*?^<52Wo4O3If<F;T75e%BFdJB&~N!K?vgIA{<irb?6ZSoQ)*
zTwD$Eb{w<6p7$i6%xZsP&!;8oJdDJj$<$s~X&!0U<MUn}dsfjX#Eg*xas<E3C0z&H
z0;{jvw8gs|q2EO5gE;Z?qz|2}(8OdOuD_oam~{7K6;tF}G$3kpUR6VmlGCd+$rFOQ
zVJz<0TfvOiypoiYvFt9lfBdt~6D2j2hxq~qvTNF_-G}U-hat><70Ox{LFJJFx`RD_
z){&7gwXh5wm3=yBhW`?Zg3(XW@}OFPpHY~A{I0OtUBCmo`C5*UY-hEJ8g+JrT}n&>
zkG`g+v1mJasGemn*<xWj$eYohEI5IV&1P{urx(Qn8QJldYH&}Dtw~QeEmuwWi-(hC
zC*vRBu#nhrfgN_%)!`z6T}T=|Dw6Foc52zG9D2!1HKFR-(-Rkv{|z+(oAxefW;oNs
z$N(395w|;~{cG3%$fS8u^8Z*M3^LF1q(x$#6Evv;;k1%Bf3BA6rE`&(%de2Gq`I&g
zUIIuyd>ENb#5n;`bS*pAf_L|v>T=|9e5}}9*oi*%WmU4Yvb!vVOkZV`a8>06GD8-r
zdvNyXkM_P&P+zodd`Gw@ta?t3>Ao3tjuIv-L=l?c^m?K%tE!0mEP>l^3nBAA0xq-T
zzmaUMSxf47TdP-hl8jJ2m?C<-49tpUxLNkeC-U1ef%28=`l^{H`k0&VAY^K@a6K~=
z*hWRz%#MA}^ILw@xl!vjOsWMXh?VC-s}U!pi$a1vEg2tkuoX+`gZwu+nQMcUnZiq<
zelEi@w3<ys_4TX6{CYIow@hDBlZ);c;{)gRQvDg1u_^yA1_v`vgB58nAuzD<(uz08
zRQVMphjzzRl2Mkhxozi{K+^={(;*=>HIS*aQo`sDWvE&17z?`@&optxA^(ee3oKO4
zQ8Su7rBeOFILKtub#BMt52NJ{@kl}Bbqhh<l=k@5JrS|n{-mJ>EVaBQO9iC_CC`i$
z2NBon*QPJy|9YYcQ&4dOi|t+SMrCM(A=Rn*!5YNtO*SsTJ(^FgUQ30IT>5%0hk)Os
zc)(!7Df@Tj|2S<7-b~QSDdTW)t0Nx`J^W9O8y`APqc(Znc{;~au;zX2KTE_$=)}_4
zgueU!ZrXhGH0!~3Fc$}Z<`yTKcv=%n#IAEM4d^m9VK-g){o2spyx-Z{pmRd<v6-L`
z<hya#;5Q@NUN%9yLhCWhJjFV0<9K&Fya+9xuY-BM_+M|7T8t=xi~+#%`24SGgBkB_
z^csKQYSR;VzLv3N=!yI!a6K7~i~}GXu|BKU0*hOXWF^rQR3TcH<}GqSK`(*HFe%9M
zTBKu7^>IYEHq9{2a0?;3qu$^z8z1Ei3EZA6a>)<*;<*SN5IZWdjp*B#lfl1G<)~D!
zj~Bol&r4}x`X@|+nf=toc`5(7jcs)M_nbL(`oX#%z_Rr1BZ@SHCsn~JJz_Vv3~*K)
zyT$oO>_(l}als}bZ~6F<V*}bd4Fw6H;@R&rYY9t^O1rgXOl4eXrB6F#cZiomy3L~&
z52bRPUu@@~*^sN)X?n-?)bAn}mvD3%8q1LAIJ1LkYAg}YDZ!{FFT?(lC%=<n|C2~H
z8wNjvR^zBeJU31ndcBp;=;8u4M(d^#$U~CBr3HD>Y%pd3m+c?3b$U;6IptEe+(1W#
z6d5QGMm5gh{7H~)Zuhuhgt&KBH89-HzK&6Z1AS=!#dRaa{13RfMQaKpQa}OiCsY%_
zvUPr;Xn(G%rA!;7EMrm;wExHQ18D+JoBt0apuX9#e!N(%qgb$QVmqawc|RLBDm|Y-
z@h&D8%Qvd-N7}^J*4f?>5^fPx3C?Q9MrZ+tqs>beyMnA;KAjuno=6<-IyDO!SrCc1
z5ztJa8hD{JFS>Suljh*gGBB0O2ME(5B1mVo<=&2bj42VS`rtRk_L(l2@&_@;PBm)Y
zmd>a8WYY_*VM-m%2y=oveJ3jniNAS&3X*^Ktsl=2PA*9+`2&}cUNGgCz#wKg4k6be
zI$`~|p<U<9_1jM(It5#E=|a1;SGpuLIjFcr@j2O}Pi+7QtGm~uP0<EnUh3REQoryl
zo!?VgUU{KIqur5BEgDG{PDomAT2<~=iztpjqUU--xGu_U(h8=V0A-t54@UDlSm_p6
z{W?N>9HgA*WwtWBDZL7YY33=`2!stSLLAZ4)G&$>(smT|yD8$wuVws<g>rO`9E#e9
zv<KdI*|TxW{kg|5p8dx&)CAK}L`)qtd=wpB+8Lc)b*h+l+%oN~X?P5sPtU5*V)qp4
z#=wPPfm9RL@~Gc3WrU-P$#gs$g7IVSFAAx>7Po1jN|t9yt+LE3gs7OZrI2SQWpI%2
zYp(P?HI^r<8I-5Ji#Pks@Rp9oer_F}H1MSY-v(DinU%}7&XL00{H^)8bmr8sGnyM*
zac^v;uUOe9e<DXh({?9hj4v;*R8LneA79AIZ}eo*di0Q?W~}EV*2WLs^7XF-Y1;p*
zToukWCIIIl2R@tI(39K3ytP69LgigBlr%FkfpmSHRpimDSKwUcPXael6ehmU@v+?O
zSlM5J7cc4Rr>`hqf@;(eYARaaO~NXkOMpk)lD1pafY}!*ePN_TQD|aG7)1~b+j?5^
zbj2^uj1(1?3aosVKlw@730~KG3(WDIrTTF8BPHvdH8?LOSG?%%B(y1?GGyH>`L0d`
zX}nGq)tE9h5>NhS{73YkYhe5<QdS}ZBD#|K<*EgoZ&C#UeL-;{CQNDHt!&RMEiVeC
zZH!HLtC<lh%V@k;b#HAJb&v^qjH#$Z=QM6PU|tjEoY#%Xi1cvS50#YUwAHoQq;-TQ
zC{73+(J>90u|Bzf@yY^=ot;%JB6RG4L7Lte<!b?^FLPHCo%nS97rcExt3&)e{RX@2
z`X7sHYZ;(T9$DHZ%lu>*MRxK#vf=0`7u*~YqcmzZ3jCvZiIw+KH?PM;z}1=WchnGN
zR*Ofq${)5ato8#J!B3%5et*fI6{Y{Gwn;x71Ab5!_#`I-Lb_XaUgz0)bayFAG%rwz
zMHFOaPL<!nZ2kkG40&8VMf0JHh4UplJCf3izz?h5L=Zenf><x<80Nr>HjH@w3?)r+
zHr5??8MlIbzFZM4SxjwuWhovPu8o<1x9sm&Yn$n_TwoIQ(us7d3{%C@Y72qVEM4tz
z<AV@s%B`1;&gAOQ6S|M=8z2)|yx?S}@M!wT7<zdY;-Tj$yy3LQrA{jS(g+Rd)tx8k
zWgyuc@SG{ex!OL*kA_~V6s0+XiDo5Qwb|{QXY5h~e=QF9|2xRxp`CU$-`AGAXIizK
zD?gi)a+$xCHcb`4WX2bM1t`;CWjxLMhGxRpUpv7qMpc6L@|g)^hZt)Ia_}9=kK(M{
zfp?O0DcS<{wN_L;CRE#)G3u1W##0NWJF^FI@>by$G#_kex*a!Wqnk}A{JJY|GoAnd
z>0^-axcrb=)p<3keI`jX?VSdRNj*=$sb6&8-ShFF7a;TCLR^p%c1L_jSuD+_Hk~%+
zRQJakN!B!ze4cj41I*+dKBBSoYSqh;G%YEr!k%ebX@z3lJFl0~f9(bX^}j|{yQs@>
zhwPP3<s_+KsxQ~VFQ*d(28kC`(z`alsOw$ARJEITH6-*k?<<iXigVFck?<{i@ohRe
zHmp}Bthbp=j+O?4orQT&C8yJjFA+M(+6_L?27W{H#@_3A;0DJAkAEq_bp7Jd@>gRt
zzpU?~1Fi5R<fw15;jh*0F6vfmC}HObUZzu(g&<p}l1^uCF&*AT<*u}f7e)d-wmf~F
zzu$gkFCfg=Hz{V3rAR?3kwm2}eyLR%`nmXyfwm>(L&34zO9L+`QDoB0H{3N|^RT!<
z+@7RCoceQT3&JGcwqG~9JEe=Vdd2DA61$1`h0BKa3#uLZdjE6USpTT)rY!td#Wfi$
zRc@-Od;QHN74My5?`n@&wVJEl(az7486V%)g*3CzT!;}`v3&W8W2!Fx<y7PIPeAS6
z=E^CB!NYoVwhKtKa73!Z)ZxW%PF;VCz*LU{OWO&|%N{}#tNoK5&DV;05Hzu{0uBH(
zUE7X&3BvZgnN081qdc0q0q{Mfo0W#VN*K#aHa+;(Z-;9%+FLQ!#1sUwlZ5g}FLxph
z%aa)Mcm(V6h_ovxrCR61@<W1EpHB8sahnV(#HmVEV=H>noUMHr&<>9XwUuPeFyl%w
zHNC(jD2QU`aS8sNtk1P|J3CSxH=Hs=lSfc%@I$v>;{EZffegKg-Kq(57bWkZ+AYt`
zbq2wUv7SHBw)XVh|6KuAq-mk_C$JAGbX9^~rK>8bR}L~|ju%@P9d9H%k^MP)e1+3<
zoPxH3)6s!sL!XgM@iT2cj&wnxB0BNqw(m_*LE&dM?XPjn@TjWKGO>B*iN{^^WIZ$j
zGtQCr^+@q0^EgFTE>><q<S4!c?=rj!dsUW)$44P`o5N3Z?B!Q(8G7Rue$#t>;C<t=
ztTD$**ZO=mI%eTDYQDmVuy0@m9D|TpDWNnE;*RB3#AO{jWE%Kc*$)dny>lE%8*V78
zhVPLGb(Jh($T&HYj2^L4_5;(kM9!HK%c(7%$&qZdZ_pFI%8u^=6?0zkwfRKoz=`aW
zH}PeV32gRro?EUSmf`j@V7Y*Y?Y#wU8ub_#4|RO8F{YcxoWS#(xPU2G8vfUW<tcBD
z1s>8dVu;X->oxT-9!j$b8G2o0zrAf%$*R$>Fi{AEa+31f%dm0VDE7lXO*D(zBoVTv
z@xD|^lF2$&lzyQlW`qy;mBynvc&ci&%2qg#HX;IdT9oeApyniieYO(R1M2K{9OqZj
z_1F1eQ=qT--aTjpMMI%Rada?-K9Y^k8+@uFLD-jW#Fa+Asx}&}C1p|{7hH^O{Zfvy
z2{S|b5s@HS4V-0QliPf>0-ppoNR&3}gV$9ku2l;-<>R{m`EmaGsJ7%d$~z7<Kd@~j
ziUP`MiSE#=zUE9r^BotkPMry)i|}$8G<6fw>~AvYvvQTo(+t@2W9EQ=(;{j`Z#$SW
z_EDIwv|)6Hm$!^Yww^!zajz~CPY30QFBTF_7hy;{bWuE&-?D+`FQo<LFOnXgqUhp>
zjk5O96+%aUjHif>z6?G{nnRB0QlK$5dHp)_dWjn(aBi3}i{2sUfx7VfH>C|1%9D(C
z!Z3AfzsR{e0cKovErRG->7TeM0_!f@yuxnZP2M3`63D1_3apCs#XS~b9Qhit=tO9)
zj~#E_@RZDjp0FR(CI)LJFh%8=%w2Qh5*bhz$<(MUPJCf~QP6$q<g;O=Zu(i$)2jF@
zYWb#biFktVKwJAtm$nQEpyeal1}_J7i-wlR=&FNOBe%T<xE7LAQ%%gxQEeDGIXQof
z?~rLXh}!`Exyyz#Z@e-<cnBo-)Jrh`>qFoFD?SkAp#&HZ+v_B+=fz_%PC+k~$Yfur
zw!Jf8I++p1=ed;b6ckQ1-k*9BA(vA8L21H1UrQ!&YtF+{_M9M@aH{iD%wT6}<EQba
zDWLC_QW6S6<%yJd^b-L>Vs;WQij5ryT+gkT=<ZJg{7Z>|CQDa*j!@t$(k;xVhfq`$
zS6@0yt39iVC$0fUS?s)-M-){wVak$8&;q+=K37F-*;h*SY9d!EYebVcQ4}>uIi>Y<
z+h$*Yp(2#^;{kX7eh>;H!9YNIgF!9>A#s566`!2OJ=cj-q=}fqD1S|qY)*izVwCbt
zY)hYWc{QW4P^N{eYy9Y_Qc6lnVM$3UPH=ye7mTx2O&ALkGkRbXE$|B-y?V5C-jRUA
zEKEnheUC6o=!T}tIHuvSIdv$N`_+>+9llqkx^Q&{d|uBZa)j`MK=Yx?w4TT4<fJ%T
za^uF~A9v>a;lFpUl`xLR@>3+OYYRn6mGa7-9&d$|4>lLV+CPApR^#E81jxiy7<qrV
z<jbr#ms5k~*hC=S9LmMW<MMj6tA)c{Z}vPTTD$j=V4Nv5!84j$C*W)mF@J~7)q9E_
z{Yb*Su$^DZdsh`t`27Zlm}IEktjrU&aX$?uhQrOj#}|~~t!s6gGmmgPeW{8INe;^|
zEV0ge*N_1oB(jd9m-KpFeNsL(M9)4o9HO7y$F5KKo=kq-D$`X_#%+X!EE4_;*3X%S
zfb|DN!&KcZS-JA5MnB5rhh)1lcT`!tMqG}^?ojMl;H&-^;%@O_e;9E{cBWbPBHc!0
zZjVZkZN^BD8tu=)7QD98bv+i{Osi&P+RF#jlEjrdkB(Rj?$!u)_V@E3kO&<AtBz7_
z7^t|om~2U^N$Te2=4uM4JR2OuB<8X;1bU-5c#UW@gN+ECe@)Pm`rkRd3<Vg*_IWT^
zf+^c|adzGx7k`QHAA?&GjA7Z&xSI7cNp0G4%5H*E1&Bh;moGsii)6>`x#CP|ky`Lk
zi%F^|FrE<14b+X7vU><jcQ^cQVrj2Ys!tp(N#8ADvSrSo5(jSkopefVGTeykbRheY
zP)EC7(0wUp<wg))-L6nIx-IY-8-B2qyGZ$VaZj*wQ(8LDhr=t3POf~~aXF7ge7-7T
zDKD7i<5EsSVk3e<!f|PpaP!v#1}l3TBIeI>uk-_NirwogbqcYGj#bxUM=W-r`1iws
z6mO4WC-Su|q56FlG24LG!A2}s7j4@{3ERZii!bZy?L}j5JbHY&>@U2yxcFn-YHn_>
z*SzX9szH~<BwU2HtF!aTQ*07oYysKE&(U1lrN-oWdmyD9896O*F$OE+DLY_OQ`LBO
z1Qv2%t-u#LknTxIOWT+zQjwLHXXEANUH(+r2;>0$Ag$J``uh7<o1E)Bz&!B4Dhouc
zELC!PdO8kAS@p%PV3~&Z@3EG)1n6q}a?@13gDEp`8l5<){~)wl9!Y$F-c0{U;1oCm
zh7<uIwW+-oI91FbAPe=}35~XBR?D)1Nq8hG%xh<YBT}|5zFd&XML#NCy_-#1s5Ue1
zW%*2Jd=#F2XxbT6#p=R4Q>Rz8e(;2QA#S67_m%0`=h4z?X&FgQGHNZeitgbMnHa3s
z{po8XD;;U60%)!Ic(oPvtEed@>g%g7x+y+&S?*^?DA<~mqnfp;kj4dx$VcpNuC!j#
zbn;eTSeE{3B@$D(k5GLomsia0miz$_S}nEEfp|t=8%)}=6ktf>pTXx=4!YZ3566eT
zHM0cHy{qz|SV`+#cdv*&N;f=yuIF{3Vb$nkSu{jhXweqFuY8|ZJ7l2{aiEXgy&{i!
z!R`KP)|r91@*?4-xN>|QPM;RffY!mo*JKxtA2dDgai5$IaXvWlxD5M(M{`d*-1Nc@
zl?pd$)Lp8BKY!IS&ruq>psBPY?s0O|OHAc<aQ=Pgg-ycKIx_OST*$`?Tm8<_?DOXo
zgZp0^;Smur1Dl1D)|V*!Ln5-LcWnMmTfCMy&#Y2v&mZjl%C0quYwa=;slK7rI0%9L
zeL7gae|$Sr69L`bojNO97qPK*A0|djE_+m!%b^7r5&o<vy|bOTdn53(J_V6|-Rz%s
zXt)lyZ!GwI?r>*+kGOXe{zXa;%Ax7cOnKIaE3WiEKb!2E`iybUIf_dU1fwqmq&RX|
z1&qgHCTDy*SG##bbO=cB3-qX{NSO3%j{PXUYZ1j1wsO2*F`9m|S)8RD!-JRjG3=Af
zr}d(3xq1)6JpoG=9@8-}Cz{UrFIEv>)Yn%y=VmG}@W;>RH*6G7#JxisuIY7?eA7kk
z0%S|0>>wdK@_Vfl?-;=ON%WB_ZqiBh;%sSojC^16#ED(n1ax=Gb(=eUj;ZhpzUNzc
zj%6$A8B_Uxjhu1M`7|IQP)=~kM=c%j@oiVrrP~mtaenz33ui%RO4C^Hdty8%(Q>V`
z3mQvoQOu5&lg)uNJ~t3`rcbP=0O=s!X|0`~9#dj6pk8M?<E0P?3X-%-VWpF>CYiO}
zhDIT$zp!87M&`D;jjejj)>EF;Ah{tM&Zp9TUpDIPkFLL7I0$urox1l`(;!A>f4CY#
zngL!FK8IX4ST=uJmhgQ_jDTPNcDS=O8L4G0oVO+Wv*p(+Dip~%G{7<Z{qf1?5dLQW
z#N7E_qCWbgNuZ@Hb?nH`ea;!!^yYzG$owm&QxAa?408I3l18Fac#e{Y_VkF@b9qZw
z%%5i*ewDs_`+9t#KZ1*vF?tUWVFyDBaUvPiy&Bu_^p@6{HU5o~FjHgSvjUIt)CDok
zs0zgtL*q@7+EWe3Mz&z{7)%)Z%;)IgtLLm$D9=5|a`$kxs!k<W-*UX6Z9BG(Pk$V#
zA@+83B=A?6cihNSoZr<Xj}q;BB2+Mb*BFBUB7aMr3w|DDJl=KAFkozM9vnu%%qAo>
z4A_whP0YHMK&f#KDk>@-|K9@Z>+2KNu&2;@T{BgPx|X)Kv6EAba?!VZtLoa?xIR-!
z%9n(}FZR|$=ub~ial^!2^$JQ#Fesi}`(y+Kq5Oz9WqYlyqch480f+MzPtwImeygeB
zQrFP<x;|SY$CsWD*ehA8uSSHa>b;y#MJKHj5)$yXwzg;zq#t#zJE5YY0uR^~?Tond
zCBVn;-9CL(T~jkQHTCIj3_0#IK^C-mLJoU@0v{p7-JMUlJzMgRrlxdLHYpwwzm@uo
zZKaY)O<>7>ei<+rC&NGv@K50Jm>q^@W<lDJKOaIG6*sV{si*gz1bcjGNuGxUyS;pC
zJ$}FppMZciaey+e&rtB85f2c7O`~j4w*S>W@oNT~&q7QebJ>cI10HEB^FgfA8O2b$
zDy8g4&CKLuGZR%oq??>L^HAX*wG0uCn$)q+JK09Yw8|)DC^=;1Y^psQxu0USbmhcH
zTm-d+ayV0lcIN(qvZlB6M&w;4F)n1$bGA=Sb(~RpJRgmVPscM4<F2bXBR$<LT3Hag
zv3nj^m`rI+`65YD;0a#}%X73JXG9cXk!IgUtGjQ9@H~e6xr4mr&J}vQS%aQiwtpDM
zq<pEsC&ztTb$nzn?KU?*4-_q$SX+ZUmb^lShvkhcX7WX1x4m@X&OxW!h&aTyRGu=R
zT#L2PmQ1_(K?K5OcZv~e&@v3ssqNi#{#ie~vO+#zV`p-Ev2VLrk2g>m`0Ksgc%AFc
z%Ub)zx46%auhyefXmAP1$hNy9h?DaBV%|yc0Bz8~W=>m!ufd6lR(5t~j*iiP@OFQW
zlZu+UNR`gj&8@=s*4=$?rVaS`K2tU{=alQyDX1>I{d{*C?sv}z*TtrIk}MniP9n10
zW}4}<w>Nr7$EDus+S=Q=xH#Pg_XuKc`||SgM;|;qxc{ur((*f_PEE9Y!C`|2&a=fw
zU@$ldz>b^q-Gw@*ud}mS-<|`WeU06GHz5DW$jH?&*y1$)s1pf3KK`G@18)RCA7ynK
zc*_jj1z05r5N6<0{=5(q6I-Zv3uCt!j+Uh4ao&(Tdc#?pRXS<C(Bz#MN3Eb-q&fq$
zAt+L{9v@Np1KqzTR3-EO7e{C^wQ^!4u1_mJ3f5Ws1jmbHnOO#^%2MNv$J|Gdf0S-C
zJp8o#vHFIA8A<dCkC9yPmrL`}d;=$nxhVOM<(i`<R+Pu{5NB@xWQ-~Y0C2^)niEEc
zw|l>N8>P&b{g(?+QH_>yMl(0SkZ@S8n=C>;>W19oLc@<@j<J6B-aQT-=Q)?Cr0d!$
zS8O)6xw?aLULY06%T=1$nMX?OK_jjL)1T1Z$XTA-hbphFauU=p+F;NUx%53gzg2EY
zoU6aw^%hJ)`XU@ghx^QFr4<P{*}hF@U^tHOW*>DXW$Mrpf$J3npINWI6Y%_{3dZ-A
zTQWmK(Ova=+!hVJY2VjFtLo}{OUj!2ar};uySuvqJqHZt^FT>TQQ!;_0;ibEw@a6V
zgoNZI*iZ3nTJHB9yiSL>LqFvX4h}j3PVAEJr5*q(I|m1!(AqBc=R>=UNre5<fQJC=
zbXLr-mrLGJfS94G8D=Yfx8)kh5cIw}@MsxW?Fg~5vO<sCL|i!QC8`w29}n{%9^LGg
z0pjmOLT%|Xv$#0K-F%God&}W>a9dlO8zZqGV!h?zw&i+9!M}grp=GGQU)m88Y&0@A
z*H6%Vgre`YiI?I3`+B9}=y2cM?%Dn2vfr)U@8ySE|Hk8P;x+)4P<Dk*2iY6%P6qBa
z4SsL>viAU6=Cu!-zPj{(@PO(!_5b=sS7VoJMKo&pC&&s84n_+S{iB$F=J-ZNVgvm*
zuSTQ%zCm{Z1sz>nj4;!B)aI+Br?=UGBfN3D{NQ(V4ScnStA~KU$fctA|JknKN%|}z
z_Q-FgPt76}ETK&3`qxOB_q9s`xdRUxP<lm-IVw<h^ZDfy(NYJ?p9LndpT8(poDSmm
z@VIQN^~+}*Ovrzl%Y6IhK=c)A#y0u4U<w9V!Eeyls_C3Zt*5%thSvDO_Ssf^%Ae7)
z7{_`X=i>n1pijShqN7^;=}(3IcgB=2&^Q)FUp=N>{Cq1u^r#3W!Q%gs^&aqCwr}`w
zPborHRx&~%E7@f4P4?boW@l4bAuB6nZ?g9ep^WUk$qpf8#CzWT{_p!f&-;EpPoHPi
z*ZsY(>pIWlIL_lduS3kIH@Lc5B2t{on$W8?#K&1RC9*aY_!1Z2_hTD~q~l~}X5yn?
z|CYkv17nMjK+x3G1m|{Pt_i}1ZoP9bL@jFng$^Qv`^gHhp@>mTFQ&qmfsMl4E}zav
zK6@C*dz)Dk=(_lGw(;Y~XNM1lSX05bX(N^Qx9xnB_&tws;NoYG>9&M~JC1dvw{PG6
zZ2TCDobSudYy<p$q~}453fHQN1k%*ro&&R9y?@L*KUD1frO-!1<Ksaw?;wxvF9uP-
z+??Hmc8Uv}>1y}YglDg--}xv(UoyA*{(Wz6@AH$?OCI17_Y&k-xwyJ??3y|PFo@ii
z!?>ifmEq39^K0o!Xp|_0g@wq5IsZ;8xY!-b@MEIPFH<sN2v&Q|9SJT77;S`&4M|sS
z`1?YuA0*`Y<5~6^&E6Hq58x7^fZ3M1&0Y!3%>oxkc9(I*#mraC&CT8Z9FD^%Q_K~|
zJMAoPN0ZMW40pKbb+53NG@sH7UY_*dp`dtndGUu-!0Y|Rh0m4Ft}Y(WBSz@@_*`vk
zTt*x#bDa_eRs%Az<IW6^zUMZFzQ;!PLc^Y4YlG{q<KB`ufCsl9WnJsLyVwPNs@D<V
zv|sx4-<$B<U@2ZQctL#G?`*n$)UWu~V8o0~Pv_9m#kerXSiK4k_w%rmAr!8-IWvcC
z&g6Ws|IJEDdfnwEl=>No_VDJ$V{PgtQxly8rrFQY%z?$Y9C*&OLDB&g@3qlPo0^P!
zrG}G<X^cA`PBW})zwwWyB-{BsAR;`VmdC7A9v?OIz&J-L;7M7f)Ii++#n$cGFm%r8
ze4|a=Hgn_XyXO}BqO9`6Qn|x6ljQ-9b-kck^yI7L5|ff*im)2wt1iK*eg@=CmA(Tx
z^8i>%85#24?~gOb_du@LTTQZKWn*h^_C9l4iPe~>v7^}D-sTQ(UtgyK1T=rX%5^^O
z`-cc#BJALIUrp_Mt*Veizu}zNMeg|))|r)$PmU@o4Jf%1omi2&=rFrXHXeW~WV>ZQ
z@(#{T@d(-3d7c&XW3nO$BATMl{_Q#6OK<2O=!Wg;7y0qTEzpRCl@(^kxAek)3GP~E
zWMX>E&HbeXoK!44JYRPq@!R<vB*8i^xs$W8u^}N7a_0cVJdKu9wG^LIOWr3>_V^-B
zPTZ-Y(7GZh*8H#q47bLLwqAcR_mNtyXRfHI0O6r!VJKS`w%-99g#Gh_{^&k)ToD#7
zE=e7oyMpJ7*gW2+?7?EM^%`Cb4h@MEsha?ox=pKK4}b!B-tp!m4lD`*If(UFYV77|
z66BN_2|?y?B_=1w3ZEYz?*b@5#H_n8Rb?$FFCPVfQT;>8n#0;RGDsFc#cbxAlbf5H
zBR-7(&8)sM^!HW%AEFe~U+Q^Kov0t`nflrWMLXl^dew#Sk)+@_+*3sxHH0n3`Ej|Q
zBKmM##LvqmB|jEpMq5CZF~&U2OpyPp-g~OoYrIm6{b&cJzv#&JwkPy&$*-%<;sp|B
za13=AZu#J)S9`>2EWfWef5*alU&PieWthA9hTA<9wVE%ECs@mM!;D3<ozcwsQmCQ{
zU6!GkIGlKHLm2cw&qHxU7TnZisvhX}(`0l7Cw%R^&WRcxK_hrSTG@D-6OaYCW+ULf
zKZO*&uX9bFJb;~YEHf-gs<gvkMdl!_z@G3t-dTae-*~c`l%$SXR8-^u=dW0=K|)p*
zqoi>k{ZaF8V;q^Al<Vv3pV;MXrdLxhPg0Td?0NLV2%y{6x5p%5`@L-aKl|)_qG?my
zjAtDn#3{>QwhIlj19m?_v?Hf}en~q!JDf2D^~=!!4_N9Fym)M4V$$2!cbz~@hy}xT
zzS+C+*N|-E@1@Yz3jk=Xt-?NM)AOwS{C%H`bQqbLqNMKUsTa2L)wuC}TNAu=zc?+q
zOw?oyEZ%CZs;;h9ro;aYr$R#^nYRPuQRAm?$xr-_-rV(jFkM~edCXh_H-#$8gp>1k
zBMF(8m-n^H^z`%{Qc_b#$9G7q`@rq+y5+2)#RM=VoY^Yt@mjY%svOI|=LN}}&b(wZ
z(?>@yAele}N2H$L%>g(N<6rE~^(C-9Bffig89H*DU%bGJ=mN?D9|EcmvLloFSlQVt
zi%(1I0<sVW$cOn~5a-Zm)Ze+VrhhWtRN8xqR+esKGaJe5o0${ak#l#PLGIii^JiAq
z?|oFc;uO^Mvcdzc`sH?iG17jQD<ux-hi;0|B>75DqQ<D5X{wj%Is9{sEC|CjOw6{K
z%}iqbN5j(t1}i-@U&gE_2H+o}vu#N8ZIm~1_KkK#&-5neY!*{S^~TZ&d92&sNDd}L
z`_<__a+hiB2@ATFLh#DI9=kkF#gJH-GY-~R$>|_=Pgaa7J%MsQzMr4pufs!!gH?G*
zWuHMFgp6vDD8U@R+!lb*csQnGGhH3lMdAAaAcyXFhr<W?@kZ}6PIz~~7bGMl%NQ~G
z4_>p#46|3;Ot!(E!x2F#BH*gt4DB#Rb^)*g4$T%GrTun+m$2lpg#r8>PCBIot+}t)
zE5HUt#U}s$dvE#sfjeOB*G_vi7(wkn=9-eCqHtd>cnkjA*xk)7D!MgYt);1%R#*`Y
zD{_9$$H2hw_U&7UZ6Xkrc0wt5%__8B?96-x_>O^r0Zav0$LnQw1MHE)DahkMVUbh@
zwB=o&x!+h^V$SvBwSja|1W*m<Dzd@;+}~#nUuJpySh!Bt9KstZuk#z=oijw2Pu4d#
zjosYd+i-@);ElTsx{$rRhKgkm{N-D!VE<ja##Z>+kQP2WkV0y$bzDK0m6avs_lQA$
zn$-O<J-@%(MXqwbCHVhH2MG30l{ap;E?cA)HWAXImOC)8n~~)EY4>n9fZ(cdpU$$v
zB5QQv+Fo8+woA;;Z>%<tCH=|7fq{VF`~Vhb3>=Gf|1j<Zd%*+?K@wiqeGMML18afW
zGudz7nmhYqUA3GvzKpLTLP9{@cwOaBogDT4ryrTptZ#&Cm{pa3@^W}ruJ<Vxi=3RG
z=CjBKXTT<&GVdWO&62))-pyMYV{{{Nn)*qU=2eksRk*B6VfH@<v9EBdGg&4|`1%SO
zoOElv>018q#5Ej3_?5#e&wxW>3;ccw87!Ra3BrjIa~A7&yshhsW%E^yJv>N;*@4Gp
zDd)Rv6aR54Wef#=4r?DE@(Rd)8gK6<0Cd162mw{oxWKHm=O$Hsn~-zIYn7tE{9~H9
zNxG7ukB<O1wr>;ULP<$f7{Q;UE}E_UFk_HTS*9Qs9f&@CEL?ZhUTq88nu&=iN&V@a
zKw<bH@LOSXL4F2`-?U5W1wXW-L!7jWGWzSl&=8(#BuB|3|9kK+ASq-k8-Wz!B>>k~
zBYUVw9e5?wfbMBp3*=mqLEuA8J!`4?A5C1ee-F^>5!$To_zI9)IYmVi;v16Tr0JkD
zpppJhF36()ms%%&vK+mzsQi)1!DIvfq?@m3khwbR7?(>2*COnSNCVOPh0(2Au|JEA
zLWb)!4;p%vPq4V0a<v%cm)tn0%-@w)gro=M^FCM&z0DXyGw3hOJ0xxA=wHMnX`da+
zKxXy)y5tlZJIYleY(<YS>gN`zbV5Wo9=ImNc9V8&qVsR)-51#=ZmV~Fl^XUDvzAEP
z#o5{Z$He0~m@tb7JpfD*I9X<!v^8Gh1iKq<I{+wW-V?{vdf`($d=5Msa!q~lZS;WG
z-C3&zkMe>Z2O8mE1v!ugX%`Wy%5d1-4wLX(C_MS#ow8<>NVDfgI?T>>>$^XWzX&B}
z`=@H#?v;ISYgiRNa$3CL8COO@C00!8yw0nytFmI==LDN+9S67x^uXn)R3la#V`F2G
zDzMM>Ar%4MeE49|o_LsjH()^C`gDx|9^&OtYLR-8x}OYhd+p_5ny&%;jB-Z$*zWqn
z;SF|#_IHJoHNuS6w-FI+aIrW|^x{%SD3tNaCm8{IqnUp_8Ffdby#Eg_rSrYgcQcCO
z#)uAcw?~Nx1kmiNhn<9qxD`Xr9YGs!ZBbEDKK5e_57=kR_vy9o6<8!HNGYb7aH=0^
z=6`db^$pt?WjK6!UHYs3qJj~o9vj`5x~6u6Xnyr;vGP*R*2E?`-Vt%u8R5@zcooXu
zyLrRc?uj+InD7pX_sV3ZbA8i7nWY0->|&dudxBN1lgZYcoTZi;K$DH~ko@H3M;_C?
z!gsLAA?=&3eM<&7<N#Xr&Rmmhw(=q<rMfLX$^S<g0qj<0>p5#vv3e*sd><cu(Ve{7
z7PVF5;Dyh?cW$gvD-JYAP7KXNhP-ocE`7fieJ8cJ%c;v5yI1Q$B6oL#ajxo(54w2I
z`t%*!m~_Oq#NklYrmQY=rUq@**B0NMNnH)82|Ks-$C;>ZdOjOw{UrQ$rSva9KKbD{
z9mXIE&X`<Yusi9w>fIan^39IH#KPL|c;pj>LBwniDF>mst$wHo5`lyQ(69`?lHuHk
zd3|L>CL!niH-^Js!8Y{X>O6IFxom9^ODj`tZXelvAoC(++csh_Y9gElL``wDTiEgM
zD%z6I3F5|b=%%mB&1TTnCE3_8+$<!!b*rn;xWSCHMB>X^s%#A9rCK#Y8I_S4)@qH8
z5vKciU4>=Z`i5of(jZ~;bAD1<eaxP&lE;*3eS;~AIgl-4m}KL6r3Z%=^^7Q*Re784
zBF)Jr--bn8`0snTY+5Be2iK}DSDsvSjm$JZBE@ooqm`4Ji;&byU|tfi`BwXqKmoh0
z$Ojj*Qouzb(?{N0Li__E9mWX70Qso2zFr2zSNf9kI|hk4AY6h3`tjpOP?dm&2rrC4
zb){mDO>Z1TNk&FSr6P6EC(!D3$Hmcv0Z{)>;XFM%`)XcsTr^BtYKwWv`Gk37EzQNn
zC4Gw*$Rxx;c6p4~48Bj~hc$I|O(ASM^U4F}hIomsE0`b$vTC4ZA{-I)fUAi)@Pzd&
z11xZ1<yt~Q0wrGX!1S~%Z}N?_nwoffZaIj_FJ8noc^=<z8he@M!5H6zz;W1eerKy&
zJ3FgoUF6iHSZth}odB!=doBQly&o+NsKM_=V&7sSdN(X`Oi<XrD;nY_*EA_e(Vq6d
zH_dGOxEcr1ue_F1oOei=lT)hA>YA2a<{O-kqx&;Vx~r#4yPEmE#tmxhMoI-mOEN?Z
zWYh(;Z%CxyrDHlU$WK0`%$EHu6&ZYlQ=ts9m-5Hrg~cW5U}nrn$b(mffgVOrh<<rf
zVP2AP(@!-oS*|iQlDsgrB6306e_$_5w%9AFZb_bR<8kDi>`=F{1tn*-+4)BctUz}8
z<)4e~NI?Q_zuD_CxOg`<rVil>K)`pnDJHF{iGRX%*@p4bm75eM0AdfbC#0mT0<1*V
z*@B%!HRL8`t;-f6@R?68dI*fIQ}5gICKW-`9e)ttvj_44UTDYq&JLEq!C&Sj@O$A}
zNCy@0C&0+^5oKBdQ+@)d5GFyVj8e$014I`8$YAv#^^$MUg584crv2HGEufOl<R`9x
zUNV3}YBEaXtJasuU~L`@YHMq+Z*7@?iVsweg^f*AM@L7gXoVXKpp$NsM{IC#uwjli
z2<hsDWwz6tkO+X|A<!By8ECD+VpPH}-3~X^<~(=#VWU9Z^0><JZHMu%<6|~n-fp;m
zgU4YBk^e!+Kl(A*4u#eFVpO9n^sc`P9{b+~r_R^NxtFG-XXs!XQr;w^Xob3A&lN`~
zE>rS20(0yK!46?D#MIajovSE#rL(NTN>3(#t5bQrem^NAX><4FJC&1lv{ZlggBfW9
z52B|G!OV;WFA~a|LU3?6qO+*=?o(?y8>L}{s@6tp(uHk}4)}*>ix2wG<zXH~F6ocT
zb}+G4IldBci87^=_4S}Iq7LqCB7FI}UqNC_SA13D65w&!vwm`ffYmvz!8;cg;sn^q
zNQgln#fc$yen-2#z5UlOC|z-MY?ZVqC@NzA`SV9u2x<zyfljptQTRx4CYeDh0#Z?2
zC}Jhhg1RUAp#J~_bYR<%A}2u`2o}5A<p4+Wd$Q}?BK*2}wJ|K#l!A6)(oO#gAPqp<
zM?14U*|PE6E}QrbFSZdxs!D%ds7&*FQO(>^Zvxx@w1r8V7kUjHVO^Y<ejo`9D+=~K
z*==lx1Bi%A+|KJtN_AEB_1z~2YX~FHRu1+)I~tsvgldVz|1do0qHZ9p!U~H0oIU-=
zNFa5sYFvLAs@TU>ZNDzzbLD8-EBXa`d@t5~9RVjmlpAzOww<m9VF6~7iy`$9cjCNA
zIV6Vv?=Jz9F-4Sy+{cud4)Z@)I44aGP^Yl<8=4(UV;jc<m1H6UhL{{zp4>z6-gs&(
zLGtQTyZCM1l*;@GclGhryt>iT4=l4HGVzW@LS@x*9S^CP;!C5_Ah73+w(0KOz>}tr
z$!04fQ^Cd^S|(0x)m<b+`yzIB$6``a$N35wh6eAwds!Wi`*7;bN1k!JtLgT)ic)0<
zr3aNZp);&lK4)wU%@5j;!5k*I+d05^jwysvL`boX^9ehHGmuSL2Fhc*b4`3_KPo=(
zd&bFRi#&Tqt0-`CKVKDHe)x4HCMUKSC}L;FyK5V{spb$7Q!mfg06iXjk5*)5VR>mg
z-T3F|?=J|yDIS{-Q`}eYK$vzQMIB}rR2N!3m?l|u<nj5HN#Ov;7Gbxr26Tj>UO!t=
z5%_|Tl(kgf6v*iV!^0Ij`h4r!oKtT4uq{BFL&RoaPHtO8HMI~25El)|256MEka<9^
zc?J&z$F_Md%x++OoTxJ8moKCMP?_{{OG+Z_ytj0~oUjjb?7^V$walbzqRa$U1PwG5
zx5H5lsIQtH@64Y6nZM)*V%Vx%J2$5fLf-?ZeEd~Z1^)~SK4xLTL9qme1s?>W4e&d+
z6e3msljF`yfe5lEorplbF`TOf(ljtG%v-kx;Y$<;<e-1~|3`zBrz@_f(~<tc5$jjz
zNJh2#N;(3W&O3HHEDv*nrN;Q-UPtL)Ttshj##3Tm-cVlZCXrovNUB9i@I3dyOhCF$
zYdtx6;sDy+$L|(7D;pI?ixxe|hn%D}nLZb(Q%N*72hrZNemq63BNZ?ujE>pY#!6)&
zlF1+-H)CZcYjsne?Tr=_pyc-%`gJ!(3P1UqR`GTvHoe}CmDaq|CV{8tf0qa&)LLam
zleaH1`Ps8ApHt<A$A|V!r(qsC(v%i6g%Mc96cT--!al@H?)ogt1?~-?fPj4b#!Txz
z-c3YfP*LQsD&Cit3AUXE0QF;S*d-ZwF$i>W^s!6vE$0squg~Uap?|v(Uq5wf|M27_
zvMj#~TTU4yBqX2+&`JWxcA7KL2*sMq0Ir$X2I~&8`w%({m9Nx`U%BV!Yl2-4r04-6
z_5uc%r{g@%>jU*BK*p-HvvuKi&3^NPFPz`*9PElyMZpcMDe1b*2b+Z@j{D!Gea@FD
zZes=EiYU>?0{=zS9Ip*o17q-dLFUqpxYYYy(dh-52skhdsP_t5?us~QnEw-=3Mw_|
zC|5v47C~!1UwL!@qu}8RXhjfG{0@fL!$gCy{(&vn1wuEdEG8hGicuj2z6ky!Vr3Z^
ze=>i4t@QDFIb-O>#YGygJ=AAs8(fE{r$a3hVMqCUT9N-xL{|Q{h|E_YHf&e7xU0Bu
zktX;2_aBFEEBe~Bch~M^0lhV&?RPXI9I&|QiaV-eu^zTO#?a?K@Fs!DnbtM3gt1IH
zxjl+Ar*vJ$CYP-&tdzVfn*81AQ5<3CP<!iSkHmGI&r-mCD+4|48%YvbSbixr85fmh
z(~_ynxYnuEDirgDEyqd+)vwFE9Uz~V8IdYeP56GaDZU!Y^_8Z2AZbS2I*Vv<#M~kJ
z?%v++q^Qtv<-M0spjz#}_|b1+XScT3ML`?=^<yE5Ozto#pIhXv_pT?D7gAu9OnrU*
zfD1pGFhm~^)wsj2efsoC2#4>)`W9g58K?fJucq99n?Y&-A%|U_I#vcr)lm0BDjPQg
zM7G-U`@MIZpnwklGx{qErCi0TdK8qtkW#?Xql?$z_IfNJpuim8?m!Ch3yKm*O%7HS
z#i{9>1JP1yTK)iIV})y`KC`@a>H6urJKI3i<&dXJkBNzi6D)RxMLrsG4CDe?YikB-
zYHARd%c`pbV_+U<>!(jwK-zvA9{x@t*YAuNJpfU_k-Z2^AKKa(d(g*hEiD(pM}<H_
z2??O?nVO7Oq584`Nb%FFYseq{YWmVcX~n<*K0XMGECY-Ju!$U&J8!A0tD^<ABia;v
zo&UpLb1d<I6v{NOdIDeu`T?8KA1=Xhi%dyLIlKCpAo3w_J!}SocgWxW_g6Qyl^$in
z#P8zE`ipdf75qvx6w}O2xpH^Xv0vPc7xf!FH8{uP^3sWO7M0r0jU)k9;a)B`RcfSy
zo;<E%N%<vo8l~2h!DvjgA^ij8p_Xn^&T26(=Q7mm-vu9;SVl=4###Ir&NGi@eY3`=
zh3&)cOlpO1%>z_jk4sv6+y57F;y`|#c#irvC#Mrp!k!u^N|@(3n1)3Xv=52OmFSz&
z_};>fNUx?{m&3$;tSoHkR8pXD@AIFh%b96);)Zs1kD#c)cRb?+yN0nx1npJvmT+Nn
zdO9VHbh*T(MEOgZ$Ez$c14Kx99Q^^SUp<WT3N-pyC{)Q93cn9g?I65pg0`l~2)Z(0
zIam*obS0eTleL>aVWT<xT7P=w-PiB+>OM(o?jQiMCLon2MBoC;1ytOtZHK3(WN4x(
zNy0L=PQA}|rfVYLle1zz8<)w~;ELFP|AGscKS~(_!MI)t&TFW|1CjW<Z<ymv%*=q=
zlaY}@e58t!vM;(ggzo~SfL=&~I3hTpD7gID`mwQbb2WP9z#S6LOOGpY3b_af?U?N(
zj0I_+ryQyT=5S0vHL#h+YOWAT{{$Lgoo+bnG>DFi_HKIO@7LU+OooKSWTbEfQqJ+k
za&@OFl7s{a{<AAGr3CWqZA71l4_Zh{4$GBGun_voim>kvYBtwZ&75-o&?g;n)7(*{
zwkDURdGO}+Ly56VaH7sV^Fy4PDWT!0a3U1_dq02Th|n6DO{BQLyNF{gerAGpO(i1z
z<~5x4#BdtttY1H~LbvK(5`9R8OLoZ%N8O(@xAf1(GJtUhm#tN(hrGIqPh#Rtg>s6S
z;~^=AHP}C>>T{kJ#k^;lyGJPV)Fo6`k2`pKbYjZmeR{8oSlF*7Po~5v+2_ik$8;0K
z`XHwm5erW@_nA9x)GP2N?}X4k`M<pYSQm2>68+fyZHqqvF=)66H|eEXCT1xgzwTO<
zDelw$C7$^-V{kpk+79#UdB5H`lm<aBhphp(UVeO%ib}=+RoGD-3W8t$?Tap>j(7W|
z5f>Y*;L;v91d^@rmc0x1`L6kzV#`F>POt=Es65sim+)ce`ubHGd^LU{Rghg&Ki#B+
zgKuJE^A<c9Kovw`8{q|i#9-=KVFhVp8Pp7bwdtZkP(}pFXu(|{hrU2P6i#n<yIg?j
z|8tN-_9H!&I^rngW_xaYufJpVPC|2(IE8>O?`SP<maFzh^Hka4+M-4T2YKb$?LL1C
z-jBc5Vn^ql_=C!mBxSH=v2j_cj?Hs8zR9Bx3=WDl?TfX49ziT@*w`d&pZL08karbk
z+$8IHlVMGx9AY^3zIcOCW3{%)BgGoSPNjfp*-h=O_M5~Si5VN}yVI}n--YOO?paCk
z;qYEf&PX<KT;YpmAc`C^xmOjx2#fU=g+yDaiRNa=%Ghj;<k!wS+7wl@eya`~B?oCr
z<|f(W=Dv0_t`#QFF()~78r*#$=#E-(haY%G9Hu$>3~(mtb(ydIO0%Bm?JV8#t$%*N
z!PTE!s7jA;w3<1OP&k-K^#m9oWSK4&OwfjUtV^0YOML%W{}0XrRYSD~#U%_8`!<Fi
zAS&SV+4>GBxVSF{5J44M7(<cZEfOjtuKwq6P<LWyjg3G0dhZju$YWof%SGOMdf2)|
z8}(BC;`QW1F0BWj085~~!XXhE_oHCHg?WwrR=Q=F>2pyV+Ff6|9qFI~6_Klf=0;Xy
zP3%Puk4dfuOLpjGF5&0#S`%x(DZN9R(KR@bSNmXp)O}T!E1z3jj$e*<M%vRRGLl>z
zt?2QzoKm?{Z~XJJ+7}JvR7=LwYJ_}2E8UnIQbf3H!ZTJzRjd(PJQf-Hhsp&-h7BBy
zOd`G>F-%OTmJ!c|+jWnx^Cec9Z1Ly4-aNLw44T+k{#fDAr&#aJO-e%;iW@98wtX3d
zWl!$h<ohvleYkilHM4jNHZiEYP<w=O94dvE5O1=?xC@|eMC}Jh`?D1Hz}!@y#FP|r
zP-Dhj5Ql=D@0ku_S^E8(8wjG;UZ?>k{m=#NS{{;nE~o!fJi<<LIZ-j$iHl6~*?DGs
zaVC6aL9p`pT&cI!iFBrlwOm_Rz3_`RL5cbDsFfSXUEK!Uca~_gLKTd^l+cW{&=~v)
z!_vg5EY#7Ah>zRYCM}lfHf>1)^0<UPUT!bCair<r%1IyJDjhRJfA_Ab+i?X8BU=aT
zg_<5Qa+MKt-?1`0-g%kDCH3Uq>1mQ3WisBtEgxk1@lAxYIyRELOOz@gYcuZQ^mWqy
z&H;H8l6!l6sMAGMI8f>QbzdTnhrn^Afm_vKhF&fAHKj;pM-9sd_mG|6XmNx^{YnQq
zjL#XjJ^gJl**Gz{IIB<3*ta-aj?ZAqJ%LK9sqJM|@fN@d5VP(O6MtgwON(8m<vG6m
zdicD$&re$fOm2c3WfabeT*cGpKKJn!5sgF1&dwb`0FV^A_@35YZr`y<WV=Tz3P4do
zc0gngYTY2mFV{1J9KQ&R1|e?{yB6H%>aEp`{)Zy}<K^Tt-Mp^lldVP_Ygs&GRmt_s
z^Wv+-`h66OFdm_l<puTb+LNn0o2=i4@V+fwJyOQ1Q4%U9NY@qiU`{a*UgaRGaj;0=
zI*eo_ZI_~b%zEeT5u=18k-7B^{wT3f|1ARR6b(DQH}l0$uK<kPtQhqF+-{BH6ZR83
zqCEPU%?AE$hBZxk#YHzx5i8RS@7kdphUTy*9fdQRKB>Bg%u<?#EBU<EpGAtqdI?1a
zQQC$-`Q;_h5Nv+JC#=@#h-Ri_5|OP;a?Xk_#g(M(Rhm}G%G8^AsA~jaM>hKjP#-8W
zIHeu9vko$8N|2U3O<#?Vl)k(`y+s?VxP*O&&ladZl+lo~4jn$aSmx8Ez2Dw!Ef+Oj
z9~QQTKlJBnX`sDYa|fToUtABKJ~j|Lpjg<`)1z9b_CGl(v!x{!+z0?2STy9(S~Z&Y
z2@v<d70?60MF=0t*MIiq#XBTq@3Z_D!khXxj#EATa*EH`^6sN$jD`{u4T)$gUstQ1
zR;!-&XRVKywG0g0*?i^7HyR!%Nxv)=Ssx^BV~GrWHS+6Jw^-t3K+GXJ`?SOqN7nMf
zs$*^4gM#JdH%i2r9Db9d0i(ly3EN_d_i=J2-}|X!w>zib9?L|#8LrcQWWy&mfHibv
z)k|RK=y^!x)Zt*GW!cN>q?P{-{B+5Y<VIDzie+)4QUm^hp`xQ&nW}oNFBpWAq!jD#
zf5dFc*Ngc<$`I?$2X;s8b@}ZK{`!E(6sWr3QU@1naTm*v7K8f88ada0G)7(7Rc&mw
z=)oXWMrjinCUb`%ieELcg&MF6?0Xj|(6*O<6qFV^D%~rYwW}izlYoL5lsJ|0Rl$!A
zYU|&ciJ{h5{hurK4Gp02LTyWa_#$tc@o%XqLi#`Hu=HIRkBL-P7P;X``<I<RZ49yY
zq1Qj?hEF9wsl{-rtuvePBCmPhRj7^|_$o!g#N|cC0~L6w9^9`><p$&jZe`_A{_1SE
zReZgCZ<|Cv`>hWB5&?Q`z*YUSmks5wYwAPT1Vj0~?^fhSC|zEWkVN(13%Bt%tT)Hw
zP+2D)`9M#Qd*yem`GQPPY&RpridOd#eITwCiV}0Rj!{u{$ku4&(6yT>YgAPpJn`u|
z{%`5(&DqUMUIpFH>iXJQ{W|hv%v$6SwR!f}&YGyosX>n9@@QI(x^^3%w<2c2wUZ*g
zJ#;!cSi=MqG+AA{0g<hfV_WlMYSVuPo4r<s_X`uCkXV0w@yHshIB{BlBv6<>mI`p<
zF2_7W|G^GMD*&H*7jFtz^JIhx@X7r*7$=-Mt$Iqo7u@afou#{R?V*D;M}bPTs@+@~
zr_b%PBm7$CHgrl6BYRtaXIyiXqN-cg<K!azO~iYqnYJdwA806X>%(IX3)o3f``Dur
zKRH>st)T0d`8U+RZs1nj{>FZkY}p;H67xv^exnmn)L@gp<cqWpV<~pbJ}Ppn_7M~C
z(Cv?kXF30ItUE2^!+w<4Qy^$@oE1dlY^?rHj83$^T4YA<=|^U2J?V+J@6exZ$k|X^
z<y^V-kc@qE9G^{_SdQYeLY(kxW)Wf!e9HUQRiqUikBgZ<n3l1NPrPB9IpU_aO{FH=
zY#V4hR7&?zA!#2aA#e)|;kZ*7s!C4SCmbmlYHh+@>CkY=!1;~Y`eOQ^(WRp2Jo4!<
zyR%-)Zbe_U8rxs-M)E&W?urq$DGt?1dso)S{?B-5?6-c5suk4OsnmG{Z}uhCo~Zdh
zH#|Htb~&cPoj9nN;EKbw7N;Y0kP)X3416BIlHrgQDb;UFdfUJFX_8jji({;#A7rN4
zZ|9~$Ry2aeGjOn|b(Fu%>5EJU*%K0$=H}+lk>v8eAb9p7aO#(BMg;$>(YbI-ajAd?
zi(u!9TF$S!8#L2wL660WDK{y#C5;&>I+9|x!Muno$|e<1?))HdV$9=Boc*I5v}F(G
zK_!ghXVI8dgR!LPv}H-9Q9?2`PvvPyC^J!0KEuenS&_7-(PpdYCUQl(3rF1#3KW8i
zKco2-fthia+PpdqBU8U>wf~-9PDt~r2P^JKG=D+aC%=N10ty^T2S4(I9*|CQXBM6$
z3Vb^+zQhpNB<Hu74m*||u2l3k5TySy!p`J&3HKuY1qUksu6xA<i8tjgHC@RW6jfp`
zNXVkyD^Rm?&B>o3&U|4_>gj`Jl%Uz<{eBYb<IeM6s)<vxPgJ|Q(jq0ft2_o&iVWvT
z+jn+4HrK^M#UfK~tIF`4@U6(M<%jFFCnX6_aZ*KnCesx{BP+}$G9oz)+R9N&wI-A`
ztTbntn$DkVe8vAncICY8l&_-ctn~Ekrq1K~rD<qU(4?1VOKy=<^rFS34q#@~ZsWJH
z<rf=~c~`;=;9)3Ap}b6Z%0#E^7hSMD>NO@*DY0_56xq*mk?5{C71}tzvA1{mb;k#`
zun%5^L_e!y?XuaRf#t!smVGbres1LXsj}z#TRU2#s`*09+dk1f66#}OGsnA`)+T6s
zmW8j#S6P;g1Zsjha_b~hkD3MOR8=T*ki?Rk@o#^=yOb2k$y*DL%t3z~NG5vOu&1nv
zz8NXXEfvKMyptB0t~NP5G<o4{ob}GFSl0H4d2SQ^W9;VE&42W7amBYY;Qu7O8-^w#
zGr*p;teljT?s&-hZGuCrOyEhco~F>4l?77~vF-MHV7PO#hf>bQ8w*S(l8!H?13KFI
zx^;3ZL*n$$`d{Wyf5cJueWVv`L?{z2SCq2WIn>~P$UT_{R*qopqF=d3LnIp+syg6}
z{rnO0abZ9tp*R)TwV9&y-z_pqLA}bk&tJk*f`n}Vtw2?`;CX^EQ`gXx?P>_udxp8T
zSgc^ByZrgdHqzUmfkLAQ*OVsbl=dm*`_zQiXVrDaPNNFo!GK~9xE?UuYfIXs#$6bD
zs7S)z$z(%0<9&|%ya)YvCGyXgmrx#FaZCWqG!&b_y)`#C=W_53k?>c1{r+9FSuvtK
zC-m8$lVW{1$bPFqDa&gEFCdz>l$MsIoXp>Kn~Jp_yMjOTjTla1UXXtHHT@>>w6!{m
zuz|J%J~6ttgf}x2Xxi*taxA6B(_gG%&7ET!7rikjM$?lVDJZRf>*PwJRd~+?ce52A
zt^XRIP+9^_dTY`8S9}Rp>BYnGybpqYEVtf<V@+lHVP+jyXeX?lo<C&d4`jylz-t)D
zx@TDxUCQcKi}o;5WW>(Tc}vYcEUkYe!=jvA$2}$J6Vw(&h%FMs2iSE>M9S*hY#iS1
z#X9w_c4-x^NC)cN>E`Q<i@WChIS9@2>3L+4z_C2aSIs|j?9JDnUY<3VmrLq8H9XN9
zN1BgXF1A{@ea^YRn&LftP`q_xx@*}1>Uk5@ws#R(bGpWkHd&!WN9DRHc1hSa$0dH(
z-en6<cstbPeqS87^n(`44UMt$C*NGxLE?DLe&{O=HItM1wEyhmFH#Th-Xhu}FdYpL
zm)ZRj-W}WHcA}PnihY}$r1QtNsAA1MOtQtg1<s>U%-(OzZ4HtS;~Cse2ym`4#Zooy
zUP*8EVeb*~AXm0#cf}{s=KCj|h5V|~wA6S<G`TkS+qt0-mOcqJwdF-umKa$oe(dy{
zZ&lG>?6f_^&(m!ZIh-Q+*epOLO6~lhdfCm)YL3M9U_5_9JdU9PlrB6j9hNE<5*jJ&
zhY$2KKPp(ruegY$-Fzh^dd)wwHbr!lp<G?Ztj;UGqw1lf_3+2IwMb>y@Wxr&H|z04
zO{xBTF2al>q|{Vn&uQY&i^kt#=z6_ppZeIx;vcAocLRNT_U!|R0yA7CGUG0C!=KsZ
zFO~%_{1EqRyuL>;m}N10%6F2C4jrB)aisRpmT6r%kUl=UzgFv@J|?-{FxglgRwB63
zocg)*j#IU*)6(ag;CP)cg?<CBmJ@mC@e|@BZ?U6!Uvlyu8@-t`$?^!iyZc)Pmx+j%
zdv@0+zC8#V9FKc^*tJ~4z16l*^Mqo0j7+x6s~@y=M`5-ig4>sw5mURF#lFiY^ay7r
z{10btH=Y&Mex;Fb*xz4v)QYNFZ17JPC11JUmz}4uT)*5%oylHll#oW_J=`*iog5M|
zU1RnyUM^;&_7}4IiZj(rmRVFZw=d^$xXJNQc)XOiq4W$jY6abqM+5uYyPLgC(`dGL
z#NN|Lxj$mupXRnANgMUM-_!P`5_hr*m8-^|pG)>+U%c0bcF=|a{HzoDVS6jX_*SyI
z2D}|IW`S~9@lg95ku$M0a_$pDyXCKL=S(LogN=>)P>-}fXHzYYS6rrwAf$@4V~Pr#
zV$jVOL%yeUw`lSCl<~F7t{WhSf*G5AaE7Y1u4TpVkAd6yc0Dsh5^zExH4hLK5WFg6
zIL!&~4a!80i*47zOLPUSSs+B*4(<5K_XthrVxqORbHK{P@Zc=`(xDCCHw^W<M5_7r
ze9h(MuV(&ra-`Ga^p_<R9C_f-gfH{iW6gFt&HV5pcTRx{wb$>(AjGl=caj*iS8u1W
z_Fh6^68zb0m#IxpZiX)l>|S2(nB_d9z&tW5#kqI$^}W{Xx?-#`jN+9{g<1DQY##=1
z#}xk1b^QAEIEy}h#wobaIWpRDKl+DM91~B5x^l*S8C@ae`$8}81(V%>V|JtBcweyB
z+3AXMXmSe^tA{@Qn*}+=<C=ypwc?9)_Y_o$tir-jIO}-9VggRLeCbkOeAqlV`?z&~
z@jZ^n%>2Km2kbmt+0Y0#ds^@AH*?}-s@NQXhlx8eBWq!g-F$ju;LtHo{bs!q1BX7=
z0(V~PJqb@*ZCj=noTwqM$<Z?dsp+QbQJceW6@;|p-)ykE8?w{>A>oF&xZU&4^~pDb
zlL1TQ!`XB%Uz<!$oREfFW}LE*7^3->llhL{=#4c)&y0U)JO4^wMU-(w%I;6<!nmle
zjzGJOM-0PBM4gdtHqC&Td|E<j$qi9#aKjXA7W$iF#n~G1y344iG}g)F#28oEhmN*U
z)hUN$#CGML2H)7s$F3SCos|>Ki!N=!Y9SO+i|O7^=GZFXyHV21S=hLusuAs7c5mA!
z*OH{501pj?+v_*W{GT1y$!6~c4O?iGc%PcO21RW|+lG=MB0@t+LYj~KSfu<g{gs)P
zK0#EDrM31(5jGWLkM{8M{$zud=AB{(@6p7DlFjbg&EwhS`J%8?Pu=P2rlXwEF3pyf
zP4uEoCnv#UO9CxdjOl`Yjp0g;=^}$?hdin+tHo!}$H!(n`PK7P+t}rIt2PB^o%DU1
z<IIY*;5Vongl;;~$I8UWJlH+0?Q#GeTvYU$@6j_~34v5v(2YVn1UT$Z^@?N9lJlE)
zc3$jw%pBMsJJp;m7hF8A?RrtTSP=2(72IYrlltFf?`Aw0V{#M~HJy7Bo*`s^{QULw
zji00A%bE)A5^)7DGmMi~B4r0-Wu`kEr&-0v_d0IlD5}0FdTF~np^%N%yZAy+)4f<T
z&6Vw<niBx`fa-_^J`=vk;nqaX$~F`M@#Wi@@gLhUy}R4+Hd!H4Jo7h~&AoxqfRdtH
z-X0-82JZCUJ(%cY`Yl3vRog!-_|HfXZ-Ne)8PaN6uA&R~*xIH_sBk#*zf|Oa3G?V!
zSkM5XeO>0oKRuidq}TQKa7U;=Ukd{*h~585K1Ma}A=kn<l|{i$hheyh#&xur=S7+)
z$BC#j)>8~V9OPal*!DB_?YES!zPbXUKlB}BZ1MK<ACuB5e1J>FFhB|dQ|6NQw_6l|
zQDSAF&)S|e2%5mSNjujVG!E2I1*>f<>gz*nr+<41fM^kC29+>yc4n*6yQgBKT(uti
z;UkdR?-?t*Xg`7BKD5Dp#iu1ziSGOL2hVGx`plb5Q2c~52J^;69KBzD_feebuD9Xb
z+I~V32D@;;W%fzSX4<z~@z5x$L<hqwz<~un2V`xZmU#@r0zvCUXid1GruG=~T00B$
zs6lTlYeH|L^<6ZS#w}wlLJ$LOIAxjR586_6nY*B!Lm(CGKYJ%$zyJ7Pwre_t@9;a6
zKp!~NI9|8j3p8>&?q`<=v{h{jhULcPHAfzB+d%KnFgpONc9x%Kf545+NJx#3j*Ese
zhN3<yoNqe~{;Z*mO4lMN+rcM{TX{Eh=rT*@tBR<p!5ayIC&$;+*P=}A&)ni&R052*
z=xy#jn&$Zf%w9Cn>2PNO8VdYZaAi8{8QjJyzWfMdHqQM1HC>?TvqLF7$}`1iiaQd6
zE+T%OXf|`oDaU3yba@lmc&qz(*&CK0ozXc+JI4p!3=Zj9`sU3Pv--&dB&b=&Yx>OR
z-xpuGW3pnxP*1_M!$2WBY#-yO{hcmt-HON5i4<S0{b3xPeuKo(IZj6z?iV5@5lWfi
z+jIpXO9z>>R!{9q<KxSmv}Qz@^$t}`*l!QCja-lZh`%aL@Vs2RR(!74koUS=ga7vX
zg%}Fn!o-lx@j*2~A0{!O2%P;<oI_f)6DzdeODNt?F`3>JlL)!QzVURAEokZS`(fo;
zSX!`p<7Q&c;LHqk*ZdS=>?v<l5>-<}yxU$UtVECWO~G0N#;65uEGVx1Iyx#h>4Hr^
zU2T4(#hJ8%G?&H7433R;Uc=`5gqUl<7KE5Bpxgv59ZuV+&tpY8h^HrA>5GmIIO&mw
z8K`i8<(V}B{^(;qzFueo3N$LUn-@Szz4+pk$qG(?PF<kMkND1=MQAqtSX`{EqazFT
zJ%Et8EwJ}43m7(;4Vob&KI4YYt>NL}_c_vxgqe_Fpq^F=j|ek98gaVJ5^12k^+C`#
zWpU96^k<R36_rx=19PyPAYS^0hUqgo7~TTD5a=<2njF+y!J3P7T!2ptI=J%+WWd7&
zO|U<vszO20bJlwXF9b25IP>;F2R@=`134hlcY75fc=6tlMGEKzIAoApOrW_llMeB}
zBQ`R`SPyovk&&31Z=j4<+s%IicR2Xb+8p3AVM%vzth`T1SUf*D0Ef%<VRrYu4k8}c
z9n!x=OHhLi`&0N4_Z0v&Fjjq=nIQ#Z6;h;j<J-BekYnk%(oKLU?eN&(K7m(;7bbzs
zKLKA0RE5E)wwq`A1dM3V5)5Y^X(xs!2hW`$e4Xs<2XLsND)nQgE*h~w4Gg@@vDA-D
z_-cwgQ)Yy~5<HijoE*deu)0IJ(DNAX3TA?$^6ms)_)W^se<c5P4K?}F_XGSK+K0O5
zC{HRvS{j0cE?@kb(uUYu#)OY5bKvy7OX4AxLEE>ir|m{EU$OAbh?)a$;I~lx*i_DU
zyqA+l60Ur6Fvf8`n_Ho-QpTa%>62;^-z7tGr(ABS>;u&n3Iq9DA@f3m?D+Csy4o2m
z4q<ISX}J$7=&HktHYO(3(8=_O$&%Lx=?V%KO)Dk>GHOV#Ur)7KV8d*?(6$n_{Vr?C
znoWRi9{4)8NcAm2Q>2^<%u7_nSQrP9$p0ldO<K^P+BLPoo`2Ga-V{Z&7U*pMaabjr
z?Vl;g+$0PfWm*G)(t)(ZJN531KJYe;>DW1d!xfR>@zJ5KO4_vu9(dhGx2UqR$I!|;
z+2SiWXo0V$_7EQ(=}<)W!|JLzI7!`in~vY7q(DCnH2c6mtb%NB1pV1ivxVJZVP$o-
z+h_>cX3!KV4o-itEC6wZwoUNBLWL2U)|-U3Vdw+tlNWE1o&%Be1vLzKB*bUh5lVzi
z5<;jKcwQmoKKkwfHYOfmXn+s(OP3B#-X|t5L7o9ic(g+9U$k*(o=5}c1*D*jd>;HT
z@Q25lrEk@9ZEb_8#%oJw9vqYe?JV$Bd%u0#7rZ<WgbjtX&ueNzTSCgpiWx-h97_-D
z<`>%}ih_UG!A}duU3-{H{^%DNpR$$c?7=UI1Z{<{rtkjt7cf40gisNrg-an-Ak21J
z{{hJ-XkS5^+#mw*L3=QmH^Al%&Q5bLFQkzfk{R$92o&&ufkuW}gjH2la6pWvIiVc}
zaf=}>u*PLyzy;D2*~=~u@x<+Hy9RmVnp#>$E-uiZ|MSEPSqm6B!wP+XNXf+0^9e+%
z!cPs*DFqFlzjnr<xi%p&kt(V0UH*?}rl$VTRa;SB-iDN6cXm1*NCCQQl<2SYam^!h
zaKMV}HtN7-H;34X5yk|QiuOiS6v0-2MM6T76#}z(cU1m0#qjuw_)~kEyKGe=J3sF9
z*4^>yV?uG=?i`=cRPbWFD?V5ndVPOc<4ZDcu*)HOh>1}FFF2?37SXKkza~dpm{8|2
zq|cRG;S8qZ&c+M3&0Oa#$a%HAVEE6AI&Tw7NlS;|HYlU7J}WzLue69+h!Z$xVzMaX
z4k<RuQF+g`=J0%0%!yP<&1U+uG?PmG<Di!8Y|^oOUWKop%RcUZc$BT&vij{(nf<Ev
z+ca`oT)8;OBBN!)`0?-II8LI)GG8?KA7he|c6GK2J3KcSo-QJs-F(bNszVX?)o=Uo
z;Qd(ttAgijwWXy(^s)PGlr>?7;Lu!-5L^u)(qH%-d>iiG`bq1%Ppi){_TvZRZu1#4
z#4c!{L%Jdr6(ckT*`XH+X3jwM+XP;{&^;jy4I-G?bOk|2Ui)8Zk+xp&)gpF5Xr^l{
zU4v|ctUEM`LjgCSc#EBb1F-_ZST2j_&pX-WA+h~6q9RS8(C-q#xBtWar2qf+0-(WX
z-@`$R>@!E=G2;BZLq?XJpC1ffk)eSBA;@^MuJbFvlyAX<A|1$J?}a`Os7^q1{0R;w
zHXa`7GEEq|F^i#u7!gla;@#mFj59(eJHW?4ScVD^g5n`!z|(+-jc@PsZT%;o;d_Af
zM^j&)h9t}pDtbtxH8{e`G~dFVymRhrWZ@Ko$V3vIY8y&GK%pHIjqYq6#<<{K*V8+s
zKLwLiBj+3V25C7tH~#0c2GbH#%f(^I!&n)_@nXul5%2-&S`FixGXxJCkg>RaK|&L_
z#k<uIr^^9?3L`T!OkV<977Zzui54M0KmW+s*b+AuKsjgvM@qQxh944!&K+BWh(=`K
z6{TU|9Sh?h5Nk6~TNpF|l`(<ypP~@II>Di<q!x>a6VNfIs@73$3vP!yckUR0-wlQf
z;9}m`*xXzKe-m6=4tKvHCWjSJ8dQ0=n@Z@=fgU64dwJe}X$nsZ_LZaEc|{eK+hC@K
zz>4(f!Lfpb1os9H3=EX!B<G;Otxpb@k^k>~6SYKi=n4XgmM`LFHlU_M_bM#u@w@E>
z%@rQgf*T>PlNC(l6;1m4)AP|o%ng$W6?o+>%}G2D2;Ypy6}H*jRTWaRmkT}r?jlpc
z#$89M%@;m68K!?*a?quez-Bs&g-v|t*Tax_=EW=(?l@`bW0OaZmVZpjmU3ZAWM{}d
z{ZyW=N<UukaDn)rN_!OXW20$>y6wyYRS|}60_)nnn7htAb1+y+QcRUyX!*san$NHU
zDZNT}!DefW%w6eS$0hENnn&VwI@~$;7AI_X{Jh1VLdcr?Kp$_W1tFCMz*C9OAp^X3
zNp_O#+}t1w_7F2ISi^mT%R>X?<-zQa{pi$lQ~3a;bftyMKP{JC$kvCl{in%_XVUla
z5Yj=14lyr@nlM=G_g;i=FKjx9rB|S6e9+BMg4mv*?cS~wrjP)h`(Aa=`I%$b+qVdc
zLX5OOp#nNQJ1c(==N>U8G!!Cz%1HlIeeuCE#pQdb{7-{1-+j~`OlwQf<;o)&LB8@#
z@azsGC0ID<itakwZBc}y1?lB)w?9xJ$SBaFldN!n^b>4sSb}LLF(<u%9>5eFoxLI1
zN5F;L&iA@b&RQ-#t*ot^PK=>p`+TeAf+1Hib@gN|^_ijJrv`HPQ(L^Dp`j>O|7kvM
z+TDY`Ua$=>SKXW`H-j?y&#So#$xCpivokaO0M+inyF%%5_zitI!qCwHF&uF+z}t$1
zH@nO|r^_;3UB(!ilS6ayOYt%WJR^HQ+p&k?K4i5ReqiV4n;I?1{2<qP?|8`YoOV-U
zI@S!(a$9KdCj^oaL^(zNLwpE3Xn#%aXgB=kdfC;~o~jY(jW?yKJ;etzA0mR<HSXPj
z-aKT+6Vk$)nGcuNv49O4QuV=)ivE3o9DDHhq4{}5J-uXC0pNGmNfQGDpN^Z4`NK(h
zL;x&Y0sPu{ak_PWAb834tp9`2`V0@b&v)ESE9)k!j{N+GNQ6NqjDYI5P~-_17!u+B
z^!!B}e7}+mWtg(|NBUp5PS)jY(aO>?YUw-5Q8!-m9MkW2+j)VnROpw&29&SJ|6{`<
zPnWQAueZE=QRP7+{q9uj)9%H@<Q|m|NtB-Zq8}41KY0D~fcUe%H17KB>y{z{I1o(5
zN{O`!mf!Ja>KCg_ocG;(^S9oUk!fxx=cTiFe7s>$e$A<!aqK%vwMnKw_3!B?Xg2G@
zhbIrs*d}s<jZu`mn)gpU%=BVqN@R>z>7G+}!n~=Y8!O`kkBn6csF*vZza?{3SHCy$
zCZ<`O@ObXo0m$lWG%Z9hb}lY34%<Juf0vBR%)$b#WxgNUuiO`2q2`s8(8l*Xe)6OT
za({kF$r|*xkaF9<oSM@15=eD@8$|B$2Ao1{j~}-}(1P{@C=Yn32^ADTiui=6s37yn
z1CY>u5P9o{A7Z)$xB??k+`*d(h-<$sVzbT?oBtOMGWTE-`a`3nK4U~fPbM)52~~X0
z;QV~QYs;CQhldB)I_Ih;06hXELj*lcY-||h0sd(KnK0f3!aD40WaMN?$uWK2=H{jw
z<U~j>KxRu~ze9d|)}LYq>k3BEJun|5_bHzH2o@9oH{Wyn075RCpkP1n03Zc(bq$NL
z%iItS_W$hqrXnLLv4YBM@9K>s><#+b5f%ke_xW67aQ|9naWU2~J2-3C>%;%@cQ~NX
z;rOW#ztT3zl=W*3T&@Br5F}0bK##R&T7=MJh@>3r@#4oogSxxXeOf3|i~TadIy>M(
zrl!`kJTf|3lrO~?D`UN*Q3WUxNqLYUkv<}@R5OF+Uf*ylFdcbuFo0A)^c(OR&@O5W
zrmNvW=8o-QkRx&6tAYaf_An<4oCFJi=pNRTd`QaVjJlWF4@!v)b3CXAPYn!^G*Ykc
z^OgiFsA<w3!d+!$rCY`C0tRNd5nzVmU#e6ydkqEFlj^X#QXvO(67l2ong@fMn|ULe
zYUGXTRwY;esXX-<Ilb|tZqhMUmF3OcgnvubEe_(^rd=6gvux$>l7SEPPf_9sTPsT)
zl)MBg-7uTVaGCHvcbh-9V>X;^d}SUdG~DA5W4rXdd?J7^Tn%e=#GW|Q>6${m7$s?I
z8B=AV_S;*tzoWM`Lqjzg7*-0REbuvS#Ojh=+$q%`Ox2tC5F0G)%8csN=+JkIZq5h6
zRHkZqi1yImi41fyWPuGcw*}c0EXmN_^7Yj<)ZeG$Eeb%WVdo`{!&euxwPk`13Yek*
zq@>A&E<}Vk`RH{j?DA-t?8a^?M1cv5iz6iW`YD{4l++aG7m|U4h5`t9^%W}lLac_q
z(evlw5K`>sb1VTafF&NL$CSEksT-HQC+e&n)rTDnz~B*a0}xfR@%C~?*oe?!1&`$3
zswppj3);>9*K+~AardI6a4>JcqW*Q?MiTiv3=yG^g?rrIdI=~(Z_c<2Fy?U-vFD$b
zj?ZAb0}p}3)3iwk-Levy&aSR(+}!Ih2M8Irp{J(@$praUb%r}G`Ez4UxjWvj*U06E
zkz@Y!W+(*Q#svGF{_o%KpLmh?l-D{e-<Y(iCY9k?kx@GK{H~5toNBvPeII44ap#5(
zaErO#IuHVYR!FiC>)ytf$b1T7!eqS*19UYZMt<;;yE75V52L^=wF*!*!1L2&C3vNs
z#lprW3w%Lv1fu#(VGP@T!RBAliUmL}#H1!`3iJR_qJUL{<F}CpKiVE43yS|D3kHeP
z+ZUugZH-kg-L3y=O2;)YxoyS$)+C`u2Zt9cWH~LMo%k-7$r{FY{Ob*>pT-CL%PcWk
znf_QT;wUBbXIpT``MGZq2FS)6y5o+P39C@1b^+#hKcIOb5TUPfty(38Y;}TLrY=8y
zd17HAuN4RCVOGCeIiIRtH;~29rh3(4eGN!PK>Kg2EhTDPncL%~aq&ujZfM(*$#8$a
zZmPQKS7(b`*v_=Ot@w17+d;l5VM(=(GlA`>1*910*02`T{Hta^gMJ1>mM`<?ee-|u
z!{p?K+O4AEV(7b(=1h8!+K;e7z|Uj;s394VUrmw06)+11_NQ`w43WNO96);j50OxZ
zI{Tx4Ch5uqlc4}(vdf#mJ_Mv{cln%dbYw&fbZNjg7?wcT+2xTxTY%7Vq%Fkt7B3<p
zL6D3yD}xij9G_WM76oGpVA4!4aE{C05tmXxH$e7+ZTRl%eP}Cr2G?_gd8?!?4>&9g
zYk@gKKt%X1&kwAj!xp{17@3{$`*$su219%gxF;L!^Wm1^TX#9#{)su`sZle#kJrkA
z$(rz%Yq_VWO~;pFrmC(hXgXaDL?Z~(FdhwQJc6AFiWa~xID1HEWSrSkkVA!fXH)Oi
zs%eDRsl6WUdf{7AOjE%a3hifLJrhWc00;PeJgwW`>&8L*35TJ6peZ9^-L<cYGcgCq
z5tv_(r_}4`)#y;oU@8hIsrjn=1=BFrbpoIe5OE;3PQc4y!U=$KNNGdEaABXXAPmJI
z4qU7ni9ta1e|IBZ?ml+7pB|%eu8Pu^WouhWjDV;31YPAyu>~xB4oYM{xoXmGT!o}v
z%5<9_uU$hUf-p4pStJkETo>ux2G!5qW>0MzhNeoD-B-AjUjKui8m>aWSeV~BpO`f<
z@|uG!ScNYA_caTqFcjBYOS?ZLIIeQ%Xk9cGWCV!h!T6x+u#Am70g4yQvJO`0P7XH>
ztTCE%Br28ath(}hX;hRNU-ss!rVFVlEye~G*>=5XSnaLLJe50{c_Z&da%|+wLX1qX
z;yYc>gpNfw0)a<Ozwlb1S0^Dg^_ac^QauqfR5(fqjA){bAay~5;X=T@=U>Vojp*r-
z9UdMM2RcY*M;Q^rmUwi<QE*HBNnOz~q`gU-u_vg<2f*AQ_V%wG6pYQ75QU0^QUJ^a
zxDv>VD`XA4>t9S+p|c4hq=o~6y+d1mV?eM#+C>p6j5M)A2qz8{zK4?rTNA*h`?wNP
z_00l#pRMQ$@`R-S=wDX_!kLp3Xvngj+;W$_!}u}a;qi|y-a;$-L5FHqqtohzSYN>D
z3_A>|5s&M8hQYKh(Al9DV8OoG7m);=Z8kPPins9ux8EDw7jx{NFK4t<Nc1e;ahMU3
z89#JPl1z<|qY+J4`Z*W0f6Rk}25Gyow93PFt0cb1cICa)eYeX$$B489ycjU*^6?oA
zix?RnA08ZRb!|Q%Lzo+D!Y`4YO&K1T%zrHKuVKoMqz|c08k9aU<15*qG7*aIoe4ed
zKKK+fjjczZcAZp}`ejM;KTL#OWn`+eL7T^~3S!tEpmsOiB2yv&R%M{Vkev<n+??}I
z-1s7j)tOwJ8K>Ud*~&a_TR-l<pO(z8CeKw83X}Nx;0fBVqs79KM<uP*b|nUCN~%qy
zUX~=`+e(RvS=<}}CP^rsr~}q`boKg{6`JmJC3LPgOsttBk9UY_U#A`>{-&C_Cpzx(
ze;S5iq=NoSBs4+zM23W&01ujklLRvA+_PCFk*D~aHWQzq{|Lm}p8~sqWY6EEZSnT?
z^+A5hR9juIGkSA34&WQRyv}W7tvGRkRMhTQ5_I_c0$Udw2Pb2yx9Yky)J8E<K$->p
zneS-AME|a!tEa2M-x?tdQH&>=a)Hr2SV16K8$$8{;RO1hn+Ssu)eQPz0ZYPYsL4vp
z;lu<rc-&=Zr9Pt1Du~G4;{7{K4?fn~++CRGx5G0NHB`x#lEXzq;9;N+ER1Rxm_fb;
z`dAJ;4^n=9QPBhP2$Mvbl9mfz5WMtmzx%&>yYg@<*S<?CMarp74obsGX3AJfgp(AR
zDVb8{*^$T?rDSN}Br4O1Y*S>ODh*1892v?+$Z(KdBs1^2)%6|k^?iST{ZX4s_OqYo
zzJJ47>$h&-6YK_?0<z2|rpFDL&5BRfmDUeUq*AFH|A<Y&2rWPRDw?b2$Xtc7N89w<
z3~x3>9WXtJq{NK^9z}XWs@PfH-m%Dtq{AdRe`OFbp>ysN%^{g^c^+|Gn*5U7?nVV!
z-j+*QAs}#xPcPc-(X;Baq02%C^$i5q3dHg?NzhEgdJ8TWtk&GY<-*`VQB^+nZv8-8
zXtc?T3XcsdR)}7IGSIG*rQYb_(5TWNJzHA~_V`7L!LZIR(!7<Ov<sUflWn+L>ps8e
zi87W#LZx=aa2i-e&wf#@xo`a7D|c9jLY~WWd1I|=#*cjT-B&vUd7`F#6zES6;KblI
zc;)wy>0sco4z!}{7<NQOUgl+i>JsQH0_{1Vl2hLcANw9+&#h=T^$6&y`-)*VJU7B;
zWhkNhGc*KL^g>EwaA?TZ!J%wu4^1==r5AVw@LuezNJUk#Fh@UcZ^U<__;Kmhdu^vA
zIIrfw_Y)J*L;7{1A|evNe*I^w{J#+B5ypG&lV2>jd3a#e^GrgEauFGB6o6zyM@I#}
z9M!atC(kDKWp{UXtev=LX{U6UYhmWS4F<KYya;5(V(ONdNfMG6nZ_fe23kX87Gyfu
z<DxrtYq8YJetlQtY+-1@_>h~~#FP)7F|M9?<^drBbd_y#Dj(Pu8i&J)S_#F)@8dWo
zjyE}a8q>Yu`w+{ID1^?Nn4F^JN0nYyRs-CKxboq5^kdW-s)(^COzSYnHd3^btP0<y
zt71L#{oTsxqmepJoCB>+f<`-ZbQnyS^u?LaM0184aYh_haTSW!rCkatFw{tS(5t1M
zp0+j5lKF(joyzdIkF8vyqFk>}7qcbl+d2-kKJYs*|C0aP^vZ-$c4K{42Og@h8HKsZ
zl}}1OVdkOaub~@SUfrAKl3u^{Ve!V~^V>tSyspQexSehAzJYJZ?P|TYNqf&q!2-wg
z?v=RZ@5|52l6J-U@th36<kQ%r1ncWKngMJ8cTv`f28f~B{;E+e8*k`~B-z3@`?FGP
z(cC%Fik^Z=L&f+TcX-y+EZVVOxw%hyctpQ^`8#&dLm7u+W25d%nsCp9RS=j546SX6
zR!3*!Uv;A0AcF9{_wfV~j^QkYDhMDvUPy5Lh3Wuv%g!k+#KZ61I|>-w_<F!cs*O^h
zDvrYlqvvI~=gGK}V*UASVJh3=c?x_^Y;iAap?$vqR_|2KK<$)UkP<wMdfiSIxdiH@
zJ;#K>6h?B><-f~fRlt^Zmxg<-NY>z>s7B*M5qph`rXKp8?N(2UE4w7XTA{KbeP^_=
zAqJVuHvGAah+sfpA+bGrbaQ&zJ3v}g<ne<+`eB5JAB+O2p_^|#HNS>32Q?XTFJPcu
zD7FxwfaWKrr1a@x?rQ7v@*_(D3wQZ7x=f6eWSUsbiQ|=mf0i`|hbEoM`b*;0pE(8w
zE|j-+>nH8j4|GWwr4;fV($>mrbHBseHY)V^onnBZ^~U)=171Zou3q&#_1f4Zf#})V
z#FK$%XDqTU+oGEz8A5M0M(gh5Xrfou^}ULWxHk1=`C69Wo%x<tWIibIJHH_2b<L+f
z&FNUG?T6UohFWWnqf<XO{ZDE!c#)s(f`n|tyr2~K88oj58V&Y)im;c#UINxs+_T&p
zFf{Qvz>Vjd8#shIg`MT13oo-t3w(x<#OFVE&W9^*c>-g_GNDA&y~KQS9+((U0G0)z
zhumG5Vh6_!L#*MUqO1E1E~q3-Aw<m%TOa5#?wsaCf!i_r5Uo8Kwg)IU_2%*H5kNy~
z`!@H(qy*-Ov#h9F(9eAVdWKzC_Wn31Pt4JyJAW$`uCfi|Zm44_zUOBXw+=EmZD(iq
zAT8~yRc^HU4nV$y%)A{F1B1`eAimiEd_3t#t<8nkPV&7f>NneDS{xk+Ji<n*8r3)e
zIWpS{v^V=jqVfLLckiC#yDeEJJ@A+WVH;;7r4GgxX@j88D4CO}RN1l2z1J|-X)630
z5Z2%5CTg#{FJw4b7yFgt;F5ilAA^Wp0fDy&cm^lZYm2tm00jtBCY)VXi4Tr|`&V-E
zv-D5YwSzSJ)6_JDiqd&}Od^o`0(%eFFCUdQ`+n2L<pa0FC>~G@(949c#SYrS{zM|I
z6D|-WLNP9lP}`UlwYAuoL^)MgbH&K^<rp*$<a-1hw6(omIdBFM<sqkYmw?a-U#8z%
zR$eIg?GV|m%BaUym0M;frP|22LRHW(N-%Mi-2<vgl3{*d(H`TbN?xJj<4tWNHIJ%a
zMu)VWRXFDNNAT><{$0);PaeBUzWh_cN27MAuGO(vU_{!fDaNraCZ#w<{q@{_<O=wA
z5U?K5`6NmfA|5lF`Ou=qe-oLPjHmu^dL>gr#76ScrIox@WEbMRlFP6R20Kt5%5UiL
z<^_fn{k~o`bAfyfu@37o-7uRRpn<4Eej#w1lZ{R}Mk3yiGMWu3#r;^h=YXzZI<*f^
zvY&Vc!$>D_5w%uFd;f!5gM~W0Ko*Y&Sgeh?5W&i(2h#y+6Pt4pP7tqF*4Cs*hVL|z
zSuRrIq9i3A?1~#65h3rPL;#%Y(E2bC!y+&oL0e8%%Fx~`D%}y)aMgm81(L;M<sm7y
zfDRxpX_P8&k=95qF1`(R9&I^^*Jx7l1%ejFI3az2JOHa?9Z(C5qI_wLpN~S`!lqKj
z8-PUzOr5PG5?AXz<l52(vIx2X4Gj%qj04pjTy-&pjj=0|!7;MZq9??8RaM*u&Vc*A
zJYgd4aUg~x3p@SdbkbB{LIkE43WMG^w~Af5mXS65?X5C7WWz;n_^xxB9%?(U&i?UE
zripr;v`;3+o~4ilZDBXJwY@?J832wl&K*Yqa{YN`7Z7&j65{&~VS?#pqy2Pra?j_K
zmqyCJiW+X#hnO0N7~D*df(4HYlbP;@(E0?4T}t0;1l0}2l&v@pX=|U_H5|R;+lReA
zD4k0999UDe67upc->qe2@lN7|(5k`1Wq35QJIY_^@!rtVKig<lk(y_MV+^Nyi#)8?
zJ`fTx?;oS2^1r$_74!G7$O#js#PXS4B_-Mi6W!u|Z_2cDb>WrGGxpc}T4m5~rKzu<
zegC9Kek9LHM$`KZP>#GpxR--d3!mTQR^WjEuHd-l1sY(N#!WI#PrOCQt%QeWk^lsf
z&{*Qp0fBYVh*MV8#zRLAba$5|R&HR~p~(Pw2a`9@+&&6?FaYgxhCdY^a^%FlBmroh
zV4^0m#yKK}r=m!7gRcS~vKJxJFz8Q=JU}2QT3T*L=)x}m9}Px>C^86k8vgt0<;&pR
zP6!8cu9bvzBisjm!c=XywDiRuIV`BraAzMf|BvMxn)DFh>^!J0UX4aMEbfJQ`hq4v
zWaK<(r@$K^=MfMZp4-yyMX)h|MvLX*TMysbx%9PX@HP9YbNbdz9LB(0hcR?6Y?wvp
zp@!r`Uc?!^s%PK)WDgouwD(@$8V;e#T%@dfZ<+PMF&8!4(Z<tg5aD-v0TvwCKacvj
zM!X5^@%^(U7<b}Q2HhbUjYWG24-Ix)8E`N?{T=JCutT&8UTO~IJ8^Xa;RiRENq9)V
zIU-)Jf@T>t9+Y5v??Nv&3c`eVL0^OM=&avDEIm-ZKC47MGq$vSdgw!A<FoCN`+)X!
z15+f12*iUIz$32C!XgkYD4t%(IzB6uQur_;Z*DB263R=$?Zh(Gr!BOwh7P59P;)I=
ziL&`0?H!WNeP(GMrks|v@#AZi`EJKzZxuS*Bs}2Z!M?`)%<fM+vu++Z5bR~SS+(&i
zNE@4>v*->4s23dN1)|MWTq(TwPrkn|5cN!0HZQ$~NB4>F4);55du}HS2q+wg$g&U<
z(<$0_uy0a6<{wH$gji+B=50mA8+ULC6x!-h#97>VokO}eYqeWtp^rzHLLkBtOv%cs
zYg_5WY!{o7<G<hscmPEjF<WK1_|TE(nFMhOgNqJKvm%KB@__@#FlnnzflC3%@pZ(m
z7W)}!p?Hm(axTy`_=!%X;v{U6$Fbpw|D%)!`jRRr28e<P1hm?3-78e_;NT^aOcerV
z2G&9TgIG!bhAS$>pt`llbvwGk4xd~^A>3-P$)vM{1!)_rIW2%FK%``3a)D|>rcqm2
zNu=KZn^BUXxg*ZY;1o7QR+ZZik4?e+e-&yA(lj9U=H}%E+s2{~BFD&}m}o_9Odx%&
zUso+B?jOV-jl$B<($e4lc%1P)^fMR(s0gA`9X{mQYP@|9H+dYF(3)qv0RoQANV`Y~
z-a`>pJzSTbLW@!63rZ5*IO-7;kc&pThxmM>0JdNWV}pYF&i`XM1U22{*r>6-TC%R*
z+!7~eI6&TrBm{OR`^;#?AOpiXr^6~rdY^lNlOrpF<+rR|Bhywuco4L55M{a$NAI0G
zA>#(LOtuJ+K@=#6?&mtS3!Az@pn^}=Xtc4h`O<e=-+v=~Ie$z)01i<yHjZoYb43pf
z!XwAMUsKGq>=kNeRNgR6#k_=#t!*M;VpwIxV2e24yGQhhkxTT5w^SI*CSSR}wX^%;
z;i%a8R@v2Uy{IV@b!AC5jWb7YS^i1KE;K&2-3bXc+-JoOxw>5#X5RUbKdUPvMO*4F
zt$H`LUUAp5$-yCQ-VKFvHQF?)gzT2|&}>?0&Uhw=2VX>zd9*qu+cZ&P-%*R<-DeXO
z6a3`^9otqr(JpgKRIe*5FIH=^79Cm9B$>_tY`_bBEf!cpVqzdlTxrSIRB4nFXggtM
zt&Wx>C#QjRo1XrCv>s@R5rJCrKCv!ehNVlGTbA^3`&`F1R$0)Nl~}{LtyogP4>m{Y
zBDu=(no7D0ajhW=9jpq7*dc$CW~5D9t_V^BM+U^4*aFt~=ZL&X7o%ooj#CGz%3O~-
zk6~4Uhu9>47MAx2{if`kEEjP&TXJHeg@zjQ$BCKp1REEZxAG}jLwJtp)khwX%L3Y4
zfdduJyVs2vWd1d@H%`W}8`}^&4<Bx1RU}lU5JdilgHoV~rLf>yeSOPi7iuKdK-5P}
ze8CaekC=c_2@elPP4%|5^>Wq_a#hChNRU&{EI`eHX)OPK3u(Q>0~|boAR%K#(cBz-
z+9}!)+%1+Y_9)Tz+Qyo;9hs0X@tcoLv4v-(_tr$78<`6gRj_Fuq5Y#jhI@T(0yl`1
zM{>MY=H>xVwkpI=WVs7aU?Ku##K(5w!u6UY=c_Kw`N>+FxCH~ZR5TkLES<PEGpWqd
zV_5w(sxvb7!Hd@!r#6oDG6++Y($mG6%@p_y>@NQOjndcq%c~S^2UZG)y+`!l!)xXv
z#C+Jo69fe6KI{+>FkRcX;!M-_YN>qMi3Y(!11VkMC@ZU1k>OggaVr|PJv<{ExJiCj
ziA+M3Wv{tot+K6b?1qUOgD%+kUiy2Rp4Lq(LQu8l8C>3*Sk;<CmIK!=DcMWl%;BN2
zgTaGTX_(OS-^50vNe3G}7BAGW)YZLNI8u5ys1&CXfi4BO)|nGtN*}Z|SlGNa1R$h+
z-GWoRiT_c7`m#`X1Qr%%zklh-F);FM&ZNZy8xuju{(I?s1(crfmq#NRF_zr{^$VO`
zo69QvGO^`5%8mU^wdS)Rh3W9vG~x=eGBgUF`y+T>BB8hEG!NQ;H0_7yVSU+wp_a~f
z?>^O1XY%NQo%D%ik4qglBEs3(mGsPfDxjCO<!>MomFpyLEgs|=Jb6QI#6g^)N=aMa
zM|$<PneUdOVp>J}j*fh*nk@X5EK~ULqj_gcm6DRP#wmw?cD{XorF+8CTC(-X5jo*E
z4@yD@$GTeAQdqCm_fW3(*4Wi5`?~}Qi&qI9vWoVnQ1r|9pBIU{FDI$t=MdHXf)xqM
zTX{oZd+Dmr9YNU@YnvphB{PId-K%|h-^^tG<z8rg<Mp#%@y2Q9xriR~wgSV~y#@mr
z>yn$DhYaXN^x`J-5v*d<$uq|}0}AQK;y8LYgUoa0BL<lTvfDHFYL&LdQia15KJ$(g
z@K11!DA3McX|f0_t)0fOM}0|^%YpuThsKKy{IjMNis_{@Z>2s<(&Lf*y-qewq{Nn*
zjC<1y`UV|ozC}%LLoH*+tI|xJ$LBKEmX8!@jN5yb8jOrn6~0gZ>f~8!I`mk)X(p{A
zeq?`^!o=X*VRL?9%X~|*<tsW<RXwc<_5=B=E1B0Na;=<tl=?b;Pk3|d<!pDR=_g`&
zYLv{1&E*Aar740f$^ZC8X>l1hJ{9SX3Y80Gxz6RSyxlt~<&OUQx(`CT#Q#dZ6QzG}
zmuia^1(SYHPM%DXl2o;I2&??$QS*+)Zi}J~ooTuEjD*Y|iK>>pinXpnnY_YXaT6aW
z&k7i9s=Fdy-%br{W;XR~tMk=p%#$z3T)TCIGf%(pP)5fso@zMBj85%)ajK#1kjMQG
z?FZurJJXeK9&2-&+uP25-#dJ$u)Hn*1Rt#^-h^*7W1T5&jOs^sYSJ9)oXMML>AYEb
z*I(K*?pQj`NUfySC+>Binl`ozJ*gOQrDl(}*9|phHBGaBotWC}(>(6x@7Yo~e}eY;
zyg1|kn#A}uIwT`D^l?5%jmZPQAletzj2fYcGfAy7CqyE+*Cbwg2B0i7E4ih)oBm$S
zyv0dfQWx?lbqSS(^V{d|g%5l^S9~yIFwkXBx{lL=+()Aw!w#JCHxtn-pu}?bi}<@X
zPHt+Sx`bkav+#U`W>_8DBsWjPwqJPlGvD3QzZ;U9a{CkC*S(>gE%ecZg~rN9;?th8
z`&<{k&CC%!1(-ghmJ<Hc?(L2WALx(f>pzT)egEp!)n$Fr-RxU=r?p9ZVf(bDe~0v>
z$w)8V{Z41EkADlpp#RMN`LdoPTd+(PJeM2{PP6A_7UilO7(RGa1@e0g0p)OR6sW81
zmXkRvb~QgYjPn?{RRM)fX39)YP0ni#4d%G-$+MwYx{J>@|E4anT%Ti8gkz9nkCDMt
zs!^rjryK)aqc?A23ONNaH=LgvtSgZ4fcu0c)9*l)vYcie59eudskR8F+GBW}!C0`+
zG0NrWlT>E)Cifx$gQS;%bs+eY&V09jBFkSMcxg@dSh!XH*>gt=@P&E$lK4?%DA3Cp
zCfED+FdNq-jusmnL`U_?q(p7Ld&TI+x|;_)*GjKfwYNLyZ1=R27Mz{6HaU~6Jna`f
z4yXtsRkigtiLka6xQOsnNhKOR7vz*^Opn2(UoCv-)bXSwKcKuU>~6iq%pZBvzt6Vw
ztp4Uu+4(*q8~cT8rM^;H#s*qQRdlnZdU5Tx(C4SSAL)mfU9xcb^6{xOJv)8}ZZwY5
zUD&x~OP;?R6qF=Xc<7>F0=Hk-cg=6H+{@x*wC~(Yl#&ec8e7@-_qOIrw$NL8lz6_T
zRRLX0j|00UwLHx3KY7F0Rw*MoeCn4}t$>4!sYcHWj`CKC9aCey*l^za?a*KYPaOH%
z&c5N4YhFNL(vU!?yL_38urP)B#aGRw(UO@9xWW;I?EClhMivi<Ubj@=v{@(XxAf`K
zwn_;u#a*)xa=5&Gw_D|wpPb9jyNo4p_xOv~94_iR?%cSdE0WYIC=yaV&nuqprd1tn
zrEtIwle}vZf)CnUx7o8MJxwV)O|V{p^`h47N?^p=hF)GW_LbtPI?@IF@2&+X`?$vI
zOY>V`-{E1hFHxI}a71?mH%S{Ys25kPQPtND+KS<{@`?)H%N#5*X&f2Rl(T!oqctsi
zcWb?hHWoS|(;Dq%AtVsHZVl58c16}pTF)%r*(?|3R=xUME1V&;GbG8j*F0&xW)xG{
z=-z79vUby7RKHr?DE`cG|I$&PUm|*Z5*dcCI_+G1Or=YNlgLUDkNOU2N4at7Z(1oA
zU7gKxKR1&qV9sUEHIT!lnI2@>CReZ`g6-^7mFpe#1l8glNl94a=<$0$vgwO1c+CVv
zcRy;=zN<4%chmmxS$dD=nTye%)0rx=D^&z1Dh^*BE1fAjdlh#enB8A(NeAbDy2-JZ
zx9VxG2whrLyTav6?Iucf*G3^527UKRYJn?sU8y=N`iu;kG-;oWHbiCaneU&SOIOKU
z9@aFWvW$=9)ywCX9+b=X*1Goi-%@NQE%a3{=Vcj>6ssmp6dp5vXZ^WL@_-Qe@LQiR
zaQxRbIIY6q9ju?wXZTsg@Q-MOCVd>6f?U^=O&%H{f;DR_t)`$r2XtAPD*~A}s!jV7
zVxi}6^~j+-1bqv2A%S||y$iPBYAq1`@!+5P9$fMZBA@4qh`gyRa6|@xXDNKk+gS~t
z&jId&r~&9O{8EYBQp?OtTq$+&{e4xI{P%S=Y?|;TL%|yx%Z*kN;FzA!_>OGNSq0j)
z8v4YehYtg6V?E2nl@=G>e_#B%2j0y_@D&7DLNGswud=1%Yqpmz0Ee^S0#ZRvS+d|V
z`+noct)i>Bep<M~`m=yx_h*mjImZ}7rD6aK^QfS0VC1Nl-{6b!V90!cT%d`AM_pi6
z=e+av-U=4-A89j7mi>GH)mje+ASP;UiK{&on-Y9_Vpo-`mOkbTP}JGP@=rjGhy&df
zmK^B*^W7(_uirksxQNIz;Zqy;za{jjYTh|LYuim0CGta<=F+>d`sa_6J4e2k%{%_a
zG~nk~i{DyYhYm~D!2kT@zrpM04?Z`H__rThyd}6c;UzOLKd<L&^#6Ii|JHs#-`}Tq
zNiO>N9u;m&?$6KnFx&J0etBOP@~xSdG2qXI*-YkQsbvA14zDC%GCYa+jcM_P`Jo$(
zi!UB-f4%rGL+ka^KVA&0?f&uN>~FvRcp-md&EkuHe~sod{r7?l$Jj^7_YbRTs$?BB
GzwjTZ0LBXd

literal 0
HcmV?d00001

diff --git a/docs/images/vista3d.png b/docs/images/vista3d.png
new file mode 100644
index 0000000000000000000000000000000000000000..c8a94fbecdb30b74e667938283348cfc58f4495d
GIT binary patch
literal 87074
zcmeEtXINCr(yk&3f&v2wsH7Qk79?kIfFTPi0z*=Qqyfn}DS1dES)wqAY;u;2h~yv{
zhNR>)<Q(t9^X+rb)_eZmUmt(qVfCu6u6paO>grx0kJS`z6Wk-Xa^=cxMHp1`%9ZO@
zSFYd;;NJlLW3>L*2l$2SAglNkANcXaHwnISh5m{nROYFx!CKub2e@K|QPQr#l$PQ%
zc3ay?O9$1~=9C0i-n0d}TB+ly_z+6ohm|K&g3k~gj&3<~b<jRC`lEfN2Tb@lY*Dz1
z39)xrQH>%NTC~1+t81kWMhr?NFlijpC|_Jg`l~m8|FiJ}DV6x(FhcLYj_}3hyK0(`
z_vZik`lV1mBKq_^tgm7b;eVUN7dNJy_S%1cD(dT7QgMy>3-|BZ{a>9Q54+#`?@yPm
z;B;V!5^!t&A1nM@hR!PQe}9^O4WHFY%!qXDzb_LoLhIkM{xsJA8sYyJ5~wKr{sqPK
z6kI_;!F~)zS1h-zK$7_ZKYz@P8#j&<d>_UC!%@c}0KC%7#*3AJI2e6}6O0)e8rq#}
zk1ngPukX#*@sc7Dwwr<8yL)#z=!2;IZ!yx&k57yH3%lPbfLE2W-YqOH-YG|bEt!5t
z^mkj_usm$#U6<O+C;NWVpmGNXhaBDfkKu>5PU-K5@xlrw@g^h;=SPoWdaC>G_Py$5
z*JA#mtN{>AUROiZe}TPG4WFN^9Qn3$d8#e5jPLQ7;xii#Kb0RdD`WgslH9j{Q`JkM
zu_(4a$sUgr2=sI>T=H_Pv4!gITGC%-u#&j(iwumO<Gv5U1O$JovxQ;{T&tu12=c}@
z-N=8j#wf3!Sw`F-UMPkp(-H5tnMtD>pWw(3yFd7s+7Xv(%MY&Of4Hyw2lu=ueLO7q
zZ6^AcUPX8OE-kvSpjob~dVg!W>4zqCP|}&VIN$fE_782+<#0aAn1f=H6fdPpE9K_q
zhUyaI{;7;4=<)D6nRae0VB7CjxCZ3qbDsn<Dn=H@wti=k#-G4gT3Qcy+~rsS-1kU~
z!bi!a_jk)Of4AKC+qhji<F`vq=$HU1p@#Uy#>VnXNnPkgU_3w?HKLD2nhnOv{?z&a
zpGs%k!(k=((&%O3z8Mt=)_m9>B(kXquvk3A{@Uew&J?9J#?yAzKV44spF028#80^K
zb8~dO1enM#f@$;3CmLO)gFscU@5N+ht-q6eI?LtcWCKg|zb20aCimjZSNvo0A~Im|
zl<)TN%gH3vez<8Q^}hj?OGwTVssBD$^fI8L&A!%O>Lm=t*~^Ht;k?E5{`d#U;@6MA
zE1Z4-aO;U}4^8o<!j$;@p%^zgEMb_Z?C)|Ji>850aiThUd?~|G_^KQtNIhyY`X2!T
z$fyTDm=I!cO8@1bO&q_4jC~DVnospb`Tq)=e~vF|0HSizuzr~PGAfL3vRaBcTdpPi
z?xtO~tEPxFwc4)$4vv!mGc_zu_x+Au=~n=Fofm3HIWEm~A}_5`V{4wT{pUhha2E0=
z$s}EVfl43uo*Z`K`^##n(^2a3NrCWvt>52pj{!`oz%u;e64~aAn;D<<@ioNmXH@im
z&T05v3t)Y(@c~N(U72M5g+zY8`T6;>qN1q3hAeH|ZD1N6-yMuReupo;Up(l+D{LqJ
zQT3%KxLMLaVYhFVo%}xTyAUAT>){a5QykJI1)NZfcd}FKC8hi$lqZ#>b;k1_zG}GC
zpw?gdp~gugbve(6h0fpQ(ffU6NY9fH-><k7Vl<6!&1CspPsMsV0e%UIe+o7##MLvh
zF2tQQ1WY6y1TrTtSAWS{90bzH{e&$FDG~cq`i)EGfn9311O~LFeXkM(`KM5694`ex
zTbC#6zD)Ff228-hhZ3Fiei!wf2Uz$S-$29@K$Ha_DmB#rkL3?hvW>~*UVME-m!tG?
zctSBpZ5k(x`PB*98*0@If1ov$(Z~d$u9o}u9RRF=M&}IYOhbBMwcpGMEH;wl@$h~d
zRP+`AQy^w6|Bjh|D6lICz|^+4FcX*u08PffLGzylp0ptuAZl+{%oPxIA3%JK-SArJ
zAGZE653HBx)f~m8Aw&ZkhchbPcWI3C4wxwX?iOjB8c2HH<Q+j5U=@t7H*qT{%m2c>
zO#{LInZACJOgw<)I`Mo}z=MtTB##Hz3AH#bnfRZsb|eQ@qM#a-f&w0GLaxeTkt&|@
zzXt~6cO;->op5dhz*$Roak}NOdb%|93i&z2?q(S@!M{s|(F1HFaqDs}N7k!Q?99x|
zzqE3zg46>S2EM+4BIGp2KRB3G{-z569uM$?+W3^o=f4!#50W;9Vy%NQ3Za-YKSciT
zNCApoFyPf775~XgyFbR~M}B%X=<fN?5V?HIJq<9%!Qv<gC_eO|>AwmH55&q_?)Ty~
z?EmoW<-4N{R|n|VK<PD?A#YB3^$A@W6214Q)aqzJs@P(;0WeNNKE3Lb7b!C2nZ8P1
z>jLKIpTjF4M)&;l+YT=?s^2xpcUDV9ug9H0L`kHx72W+4i3$J`otybbZvyZ!yb0uT
z;-E!HM#W1gqJfq?0{**?j^hArY_q5Q0BwR)r8PW89~%9^D)fFm0D`c2M@2#a*3$r3
z?6*V3RQ{N}kqG~D<0z3)RnDcd&LrmCmcY7btn(lJJ}u+-?#p^>VXSlmkT?Q*Jm@?b
zGxp+7iGG(Ep33t6WdNoCAu}C$rIqJT+o$?kkW?(r=3crDBWq-(0av__u=t|%Css(8
zoB_^pKDmYQ0W36onNqw%*V6yMVmia+w&LZzHGiPA$OY=LI?>Vy<v-Wq4dmYahvmOY
z&qUvh;cGH~jG^~S0i3e#A2)tk#hRlUkr~cl32N_mT6jRnLS%)0=bfuJcWHnCn^vv5
z1|*^VBxwf?vD{CSwwuF$E`7-!=d1pdPtE0%^vqxU49OV&gG4V4Q^8<lf^~NE{1x0p
z086=C=~Vqa<pUbJ`uUdZUuTpA10szUe<eT?5zAN6pmMQS%d~&SlQ$Cpj*&!{3BZ-(
z_<4DG=zxHLr}<hADZv<^=B(H%hy3AK>4r<j>{Jm0$kfOXpWbruvNAuB9Z~vIp)<M0
z>n+Dt^)&f_E97yyq1dzFPx?;0#e$j@?0-tWAbC(y;j@520oFz^rhh^(sN`NU-#;k~
zMi20;JN>W71p@>|<X2Vo#~5iGE&zWO{_WwHZ;LY&Ra9ujXFBsH9yezGiRnwo1Eb$&
zH~~_<l+*bh3n(C_d+M+Kb5C<As_Z{__D|ISA4D}40QhqW=uY`v@5_h(dgXtG{Es32
zgEjxxCjV=b|BtcL_}U}4RPYZDx=dD$$H(sM#M;)@*1KY)mpN<qiXZUPZ2bE51F6HR
zNjaTyn~dTATcRGvNSh)UxSkL1WLDRM47+AK-O@^4kCKiO<9O0ghh5wzzKz~OeL1CC
zxQWCS4ZfmT0W>FoZphP|r`x?SxG4fp^_@OYk2$Me5fKCWDf062{Yy*c=arNFS~@kW
z%XSL?p(Z}P*<##C|5??LqmMV2gzhKw{LtRhkVP(l!cXwiO%-GZ)gp+p$OgVHn|OzP
zTrU2?k1j*dnOk85uA<krx9z>@ErGE>jTA0jMgJ|}6R@4;v~F!}-D$vvmOLBY$7syD
z{+qcPuaatquwvk)xy##U^HJro=rjrGL%|^z+|Q@Q_dsj8%0s)tv?H`c=u!pt<-I4y
z%Uvqa!&^0siqkaYH2g-tN<nWbxp5gX<L5wS;l=Qn`*D7L{(Ns;adB}4KHL;rubO!0
z-+L{;DQ4eKXUkKX$LnfPjc>gaH(@`FZbrG!wamuuGB#st7Mc6Mkr@+=zfHk`k2CtB
zrTe)#IoqHBvr12ceyF|KJ{lo<v6?kfFXm5O^zW=NDt1+qPA|k9(=e^8Ik--fEdVnb
z!EL*J#@HX_>EOXO=~?!E_?GRofU?Vk#8sQ?^2559CfMb;+7IF6;+hwq^x*2i)R8&@
zmB;_E`1c^(b;%s`#s()7<b)Go{hwso+B(WoQ*gCNii2~etxW}E;_47-MIWP;w>Iwu
zPJp^|4bQLS^NfKMacubsuh=l#zpEkl1OP!2XRa=6T+hhMCB(al@;mWKUWnaaCMl5A
zR(!^j;W)PD<l<6wHu9~x+JgvN2jo*)9KBjy{qPz+ND$T_w8WWbcj8}l^e@%HenEso
z&BiohCIGTT1FxXgi5EC5oX3_h5s5L9OZaRgfYLxVdd%QkG-<_-Ynrd{AC0KS2>gk}
z`~*q|h2{xu96SXh5y%u>4~G?T#Ahy$oZ02(f!^=l)eOtna@yN}XEneycuGGsphMNc
zeSVNi-kW7AM&;G1shn8serMc$hetPYuV%FUJ*{-TuEVf9|KIim%uUVeHy<U#E$oxt
zhET!bBGX3lgy6%Ge|vFKv5_BtQp9#&_uqMx`rg${OXAygx0Qcn;lh}(-(uol{;QJ%
z;Gy-hYp^yr%KLAh0dxGvbg=X^{96YDMl`v9&p^5)k3&o=4nrc5%Y!*jEMC6MxF2Mc
z_N%T*SH-2Epy;1%jT{&pd@5=`2fIhje<N4bJpXd{B>f-7`~Lu>NWgD(q%fPO;gnDR
z^J{S3`SI%V+E}U5c)s0H=%x8mnIDLWB}rbK?ufT2$8%>PDS6lt9_}qP$$)chFE_=4
zvk`xL)USY{sHEhb_xaJH<KD{9?tr>fme924L*;RY#h$FKt*wIlku;Zm4|=~zGNah7
zAY!_Y9M22p@S3zNYCKZcFR_+Ba9J25<kuqqz2MO|(x{3`+v%SE{cxeFHpB0+74R!N
z7MEs{P`P5E{5~0C6I+q5z7J61jICw4Znqdp>7}(b=WXI#m~Qr{7o2ZV9LDxdkNDOt
z$7y<ZbSuyu48cn?e)Q;(n_6<@tLRl0jY#SxqRX1AQ3f;`%=Gd3Lj04^p^r&%!i#sQ
zs9}#7zHU)<NWq=eeaRArj$&~S$XSErg9h3T+txW8dsFEj^B{N(>+{MGe7UtQ!@0Q@
z`6cfRX{e}F-hR#8s^Z6(Sf#P!3JMNCadHFuxV;(szNCGR+Vbq-E3dlue{m>p2|@Ph
zs0pUlrs0Mnf8{D$cyZwp8$K}hEI00GaH>ZVMM`Iu&G!&CW33A7%-sFCKF+sCvxGFH
z&7Eh_uN%oL-Bh+E{gH8lvO3XYTf*n?XOM-vyRM8YY(O=Xht|r<%EI2B&+B+O`{+$b
zmVA0;Pjd3JUoabej^lY(+p{dHz<Yl{*#v8Jk4OWlO-)oR?BvO=n)+VR_=#~U3HEj1
zaIV!dW-<PWX%%Uc@dFyKc~{YcpSfar-?}|wefynP6ZHu{?9c5-S{!Xn*7iw55h}D&
zX%Y3u+zp62W4Z>-8sS}GiO;9;@2zQQ6%~4FQ6JSRnR!U+I2?%kGt(OW3hC+#;rCkD
z<98~llq{t+ykO?ekmxW@7ItQ)sX2T>F+g#XTF}qkLU|<_f;e(9GiwQG6J)_xJ6NAo
zHuO1jqHrOHScR;Mc^#K62N0b|Pi60e$_+iY<a7-TdUm{am-=>x;kH>Q)G!Zh{mIC2
z2?kzileJNQ!u^ar_Vj3XtCfz!YBd7a6k%A;MSa<-_0uN}jvmh%`{uZsyPE%uMAAgq
zA3L|T%<tvSjBJT6OjFAQBe?QG+Jyz#V$;(ES4)xu14Evwbzp|S?+s~%_s1>KYclRS
zxcd)%V_EJS;u%<(_=+heu}v)xK1>#M0^g2Q-5c8K+cuhg=<T$u_`dG^ptjQp8Xgg*
zUdd^NijO^NRX?ChO;BTqF7}CR#{`H;Wo5Ox*za!XI*D*!sV&Sy9ML;`eR~PMb|*D|
zE+5L63W4J`(SC)l-d?E`qJGFkgO8Mi`z~vWl+QW+$7PHAE*n#-goNc{_kwrD-L_3A
zTI1r$=yfT6_MC!J+-90X_(d_NME81PM<bdVV6syjaXrkhcNV%iN7)o2s1>u-QdZW8
zDsx;qFoJ$%_kJPBT$aAa=P!<-U>lF*h4rpL;i@zWx~}Yp8x8Nw)aWcq-l_wQOAj!h
zV`s)g7`)zfe<0r!Ev21Wlcq&!Sy^fJ5wurTEu6$O-Zdt4@W$==3AaMC1rzReON(yY
zx7mj*4!Sok>DBKH;u|b04Pi!AOw9C7`+LV%pe3cFDD0jeGMpPZ?XQhp3zucCj$(au
z?C`TGfO;z@4}sN}!4(0O&e57peUA0!(CM^mC@YI$k~q?+N=HP@3=7jOXc>C#J*y9H
zTD(h{@;RZ8xC#+l9^^JSR&hBcH_HQI@q`F{VN`^naH2xfx8?BiHWzIjWO7ih2!YGi
zPfz@Jy4#0YV_A-~NGniP`Fn$tXA`;}_8&WMj(Xj6WcCBJ9E$x`woAW$Wi%|Et@NqR
zdt@&f4qEnMgMb-S&aQNxPNvOf3`K6;)lr>o?fDpOd(IcmW<~t^hP4vTnR2=T#-PTT
zOL^QL?*=M)f^>=};50$5gU<}jI)gX%9AZOBN$J2n@Rq%_w6sdSO&5Z9pJ`iWFZ2Gm
zOl?!d(S15PpK2q$(&Ct!aP7fR!qL_Un$=w6hP%-=t`SFEC2?)b-fSHh??j8%OY$|M
z!@t#XlRIO7&&Pe$0=o*f8i^{;9V**qEHIc(>jae(&!W-{bI>PJlif&@y7OM!8tq}p
zF61+%bQ0G7{hk!D5Ewi|Y07h_J2j$=$(9|I2h$-ynxNDjz4G5tQZr<)tn^#YeV6o3
z$5S$Olf-x@m|}mlxt_vBzV8F+6v$Bk&28YIDuUtm^81g<?m^<rXYJtvl~2j^bs%uJ
z+aiupB^CM#W~kaT#%Y(P2)8LL*KVzgDE6!`LNb=0Bt@_yrOT?&CO7ZjeO#iDD@da)
z=@vEWM3_PkowC9#ui!(JjgOav@`Ms}7-F5>ORc&HVUq34LdUM8m-`70TczBin()ZR
zcsD^u_NVgWvRiM9#a}U0s(MCZ0jEfj!PIYeaKNXyN2@Z&>R+wK4wVUe2yx%yTDYyi
z{L=0#L-xSwWGCw83ogi68aj)l3XjXGa>L205R$=$zcj^_viD^ug6C_U*z<@y_)UH#
zxDA5>W_!uWLfM~|jfAyLtRENDEh;1w?~wKm;`bE))Q{ec@b9|-UK1-6ZjcwKYCRHD
zJNA`Pwe^x2iM_8W=4uqYnRMd8yIGU=M6k(e?-T4mLpKy@|Djw*7Z(D(OL&&WGiQ`V
zBG7?2FDrd^4sYf@DgJ&hBf>dqe^AF?Kx#m0CF%Y94ksEZgPQM(KAja|Z+C93n83A3
zuC{9%GGEQPw(vk2{e&3NR1a-{Xx~-O$%V>4G-!FuC=ljzy49M!qb%zG+|Wy4L;0su
z&s%e{54T6Qb*It`9=j-lWz$B3BvF*{c*g!4oh>8UJYHEBUH4BkDMhmGig__;CLnr}
zMWQ3WXu3w6m(q=YPismNg{55GiO;Hq;3OWsxsT8!c4)V|MYv*=nw0f!blFawT*@n7
zY8|fM6)s+0)>@A)VBH`-K0jy}*<Se?D7jNLgakH0$kLUB%UaWs<Z3f^H<4j^2=$Zt
zna9b7ko`s+!^-}bd5`7Yq*wiqIc10Qiarm|%m(ngxCrtT1uSLJztKTkvr}*<iHg+C
zpJ(;x^=5uroQnSRDyJZ|k6%(EPHuZJCowL=Glb}))Q$$_GR<&?I@*Vd_I|{f!j5)i
zvc$W<m$mui#{wP?)$A9IJ--BzRfc5RL0o~{nvxI^+&FB_(N85`T1l6cWmH2^^(Mrg
zmiGQlwmU3X1><M?0}7Ay1Pr(0Di+sULhN^g1hX0Bt{a16<-#wNV26G%4?Di@=AF28
z#_z0%qc@F8ZFP)@Tn<8upXWp3GH6jMF?)6jUw59V!bk?pg?DW)h5ZzUVR1Snxfotc
z7iXL_CZ=;pbc-l^E3;VCbuMA@ug7$tnzpBNo6Q=MBh1?loqfz2b!Gbfd{5&a5lkF|
z58m?eQ>VQu-n@G<7{6!9w9z2$!gE{_!|;}Eu9}9HMtv6i0o3;h9yVLmA-WkKzGpdi
zCqlfbl(#-u++dt~`Z1@qyuozw@~j0v?IRvV%3>`YHIALg8QCAvWfiqlX<MCc+&6*(
zXG>kh!+WTSwRl-HHahFIpT{pe+Mf<tS{u_-5A8ibX|p06x6RiHcdSs-g(RQiEAZve
zT$J0c^qB_x-phhSa_e9`wlEM#8~09A5k9P08x$Qr3!U*gnvd5>1(%bgjJIPp8ZIQk
zX+Og0p?R@-xw|=>-11`he>I$8ZPjgi6;c_f@0)VH+e@;oZ<j@FJtlq^{f6Wbrq5^l
zv-jo~Jf7W1M54pCpC(D}YiWOwDF8|+z9yY4i55^#*hs#Te$UH|Yy5NA8h%{=49#<5
z*VRXiHb7+zes#5o-->ru=KeWeX}jnR*>WY<QKosombdh9mR*)Y=I9%xp~)2WI81S&
zU`sfN&CUNsmnVG_Dmm<J^=MEv%74@RnRL_Lb(qTMc*)mm-+7>%II-sq7w0GHrU(%i
z22$F-&V<ex^D_MisfB<8?wwI#W=!D0d}sVa+lad>rh3%zauE(c-#CJj?+Y#j2H>?y
zB#YFCmi6F0j&9oeQVc!Q!~gNwHug?GjanDi8E3%k`cAKu&#{;jyH!%-0o%WJCej?Z
zmN@p16t9g~PG*pAaC4Tzj)-bg-*ban8aJl7`8v7{bP(8)@}=1JL_rX17QHMgf=q#V
zU~5GE<A807GFdBgQfdNyfP)05&Hig}eOKX}<E>_!@UVzEcanxy0|}5uz1`lK>yEp*
z3z$NoW(V87duW!-8Zem0=>T`)>PNhGWt~&hvy<Yz!?$DZK0k{*S~O!(9t_aSD)5kM
zeztp(G*hYBJsznBe$n!1xCgyMSO31XR$qo{y1`Ru^5k4i*Ht)Pi$buK3UMpEx!ndf
zR~*&f`etX@rT5hsW|Ze}>g1<$QFRhSW^b(rcjSf^^ilkIxLkuga|;o4*q*QLCLhaN
zc*>k@D|*CWw)623>2O5xkn(tmrYjE&4ta%g)(MlB84RZrBP-c6sio^(TvU7%Rhj5R
zl6HUb{fr5?q;=#x)pS)Fzkw|N+zzf&^ZG*9-BSW0F+<qiyro@aV$Fvf$x^5s$h0zS
zl}q{TOBzXf7fF&Fjs<-oZ#Id0=TzcsDyj9kehZ;j6v~<ogaokRI8pncj^EXn^VtgA
zyt##>8-xVR?qtFVEWiuvdTN9R1uciPLngZmhfzykd0oDeDU>v)iX525Wre4~!<?A*
z<ve*LKcnPMkJ;0FzOXO6q_5v>y7@(vHaSAyRjTvE4ApMkE8a7CZ1-f42Pz>T5<x{r
zo2JIl7Ze>Y8PdCfS@dQpU3fTqarUz^WkT+x0uQH+x%o+rQe9b{-E7Mfi_Yom5cZ_H
z1I*CA%<$VhSz+)e);1L=eaS7}0P2}?Ql0A9Jh>%EUT7ZI9zuX(0a?M;CX7qo6n>`6
zZ#@=2GFo3xepe^;kuNwe%4V;SSG|kLX*)fVNq6UY!|}5SdhP(M5}j)V_F^|%ipigT
zHkT*a=k-)0)hJ$HkmtuUY(<tWzQm{?MTB89`!)#?`EexwoQzLsm?_-*e4fWAF;fSX
z9d<izFMxMQN}*x3H>JaSP1!=wcJsK?@cbR+oRw!?m#W(Zm}prH`5vqSXA7UOnC<mB
zcZk5}c4)7%T^5=^rJADh`)}mAHm2b@JUQO!Q=l_dG&Ed#xaWDcH<GCDC2^|f)76;d
zEOj&&y~wRi?j2So`OYKPZa__B-J)Sjc*;^L^%bo}wjG@}Pb=a0-nNVXaFcgM=91OF
zIv9Qw`1Ds;)YY;wi|Ox{#FdnMewR9t=p{2udKSu9do;9HMvZ(ItC^t^f-_&d>&iQ4
zq8Abw-2Yt4n+u|l7U>FMKNyZ58Nk}w_$TDZ!G>JQNs;MW^zX?*$_?D!?raJ%{nN*?
zgUfqThCO`I6=_a??GLyM*c2{&Nw6Vft4NqSjIVUi@2;PVBAr?6_3pr_s(($|dg#_A
zCm}*Aj9%8duHms*leCOU^C9YcEH>6X9#{(gD{91zQ|hKx81LL(Zdxsc{o)bXD~dq)
z?RI64zMs^4r*p%zyNB{~qZ2B7($I%|EA#{?3eOj1YdpY?58lsEd$vC=@;-BDDd-IQ
zv&T{)m|L>;E~SR$#$DX;!S?5~vK=PDW2F|(he@Zp?G`n$c}%1fiEVF8f;AvQl^j+e
zKOLR9Gs;R*wI2Vc`TE)~r6BUkN`abq8x9cDv~${bcl-1@?4j=^Yqv)@A1d=Bu;uT=
z<S<%pMNztFPkMsg-SG=*O1BxW$?HuOq$S5&;kG43X56V1^`0$0t6=zZx8~1Ul(Y*~
z!si$T9()H#H0mdzOQE<iwbL!4@VQ$LR+GYhK>hi4?6$FnuZkxRtHgaa7C#Fg#&eFY
zF21QR_9<XkFiok7n0+I%HYd9~wAbr1-L79|O9M-f&*CzEWEM+~4{e5kLMEU@+YkAo
z4~Y2iUw8n!ocJlN`f&>rl+dwn-fVI<zV4i`OI-~4EvWr)wz0G>#1hmw?|{#%;1#T4
zLh<I-az{gOEuHrv_(Juf+tzzW*577XYWwyJnBR4$ejCulZe%^_P-xz~3-oTueZP*1
zHW3ryixMu)M2D&tnEJ8*EhhKa9TLgNu|VD!xrR?~dHD>P2ojtV7>@UU2GN<l(9VFq
zxSD%hhKK*uwL;hEWTfF-KtY01N6CX4kD7%X=ca24VwvH~_16m4>mMHzH@!a}Fl;ew
zfzc`PTZP<sCI0$l#E-~96E)YE>Zm$s@-U%EgjGZc#PVl%l>z9m>!H&otsjTiOPkIU
zV<F@5PDyB|J?J_&?95){?YUkk>!hYLUxUYT?~AqEJc(N?6~FpP>NmsG!aim$Mr4hH
zU$DWNIW<-ZGzudNBvgbNYAFm0X3Gy-B<%RNP3pYaO>D7o;{G#Vt%zk!{JX-=HoTS3
z6*Dwu{RR6$k70?gH%^4zLdpoU!u#5RKbADn>yM#L+ozeI@N|@E^&!eZ43-Cb?UfLM
z)iG?_(7a{!*0&)?p$wztho++a%*n12HsMnf5#+eU1yJ(Mx1R~9`{R;p@uv^Y3nL+a
zS(SMQ>pZwP%_b@DvtfIf9XUsS&3It8D!jFlRvta|b@#`j$C2QK+4T@5lkM_e22rd`
zJ~u{muVtBfQViIpJ`o`4?h;0I5kqj|i`C`czwK*ZJlQJ?7v_CQEoT;{@E#u$KNjS}
zRj1&~qwHxnqYN^XJ<K@l5!-M~Da(f*zX`w_*~SHxcOZ6+2kgX*I3Ea#Dz2|R0nggZ
zvyRNOpeeMwG>T511(!%rkj9Bu?|{k`aoH%v16v?_9qoiVkS8}B=FG5!Od1~|3tF({
zF5d3!DXp?iYH99V3c>QV=6?lpLb(g%{U6LYOPb6Zg&nLZTnJ<<n8QPO&d?OT&p4PC
zaU-ym9QcEPCIi+mEi?<=Hp3I>_xiX+WG{^g_c*;mE^oF&WqqU=?#CIBg)JoZ>UA}B
zn+vMKv&UqZV!FQ4!H4&ns4c>!gH+WtKwe1=Ck{}_)&TsK9Oe8*DmufH6%h^P!R@(_
zwq<cQ#*dsfUvG^8xz~MtB#y3Q{Px5M<99jSds$qhk*wo35x`cI8uV5!6p|CAA?FWi
zaHTJgGHKIormlK)kV$_96%inial0=x@8obcGHt55muqXUgrXR($^M-u!{=h&@SKV8
z$sErfc=PbA5-2SP;FDv-?Q%8U1SfqI<C_O*-kV}hb_UYBEI1WVBM6S-a$`mCml7^4
z$@S1pV?);x!?Wu@AWn&71UrohIfZB@=_LrcMZEcQ3>Vv#1hSUfJQx{%zE{Du2U8v!
z7K}W6JEkIft2z(VMkzFu8_u`rE_k!AOSfqXdmEjcOkYSMUmwertw^1Im8yLWD!RMa
zviPHCT4IiI(Lk;Q0v2I*-MFLCd1xLaGP6jkgLKRyF3MKNx8G)sCiRoOujl)TvR*GC
z=$Rtui{1n0e%{8U3T3DRXeiukx_A06!%Rxk5!;3w!HycE!%;+k0<DSEiQ{)Cpd0`G
zGxqZ8+LAF2tZ6H;?e-lJ4N$lhO4b*ST_ptN9C%khAuHd)(cEp4s;8|?ZBaDG9}`Y%
zKcN=!5!b&^p0f$vy%9Vd1Xj{%%Y%lI-b9LXWyF2!>G5&}r?{V3tEKtEnEu)~M@xOa
z2P%LTzw7^TJ}<W1b+VWiT~1s$Md(JVp_7zB;dY8CoF>rm*v<*gQb*rCJoix`$`3xA
zZ^q|6-drpd$-1IZkTr3<B0u#pW|mSNJ(5|U^F+aZO3!3tD=53fy#A<CcfH#z==S!W
zTJQ?Q64giP<HiPAgK$A%CN_Ke$U%Y!;Gt`M9c`PfVB4QgHc8}!UA^XLHQ_rV0b@R#
z3tZVxg55dmE?YWryZDn3goF1;)d?7CRq|Tcf2nzEN2KH3n<d6OI3*NE@|IT_=XRGz
z4iK`fk*VPaeId!Y756k!pp7=xdEY!Nm?b@?d??uN@wu6!owxiwYw*92(sW@jSf)=+
z+a{4Z&Gy(pe2o_8gCw6`!jE9zMOh`aG!F(EkpnI7jmpeC1vQ<1a&gMDYSnL^>-lo*
zZ?FV?CA;O!mQT!HpY8#t2g|{XolWlaNNwoBR}Mq;x2OpyHxIoogXq0>)ep4&x-o6W
z8EAEA=$m%VFIG+O-J=SKy-%Ouj|aJp#|q02a};Z<r{u}aKmwJaTq2<k`e0|!>Sn;s
z=wQg9%EkDAVCDVnLdA4S{6S?nlnC@5MoT+M(HFQTBO5#ZmokArpE}bWoF<~5&osen
zW}qCc1zI;9T(i2ixmTlt@WarfA6okmGSD;h4OUI0GfI0c^3$UpP(Y{msh7ejg2@V0
zcUq@#V>sJKT2qrnqX254IZBZj$e%kM=lruXzz||!L7AvDzGn=@jdQ#{7=A0O{uuTc
zlU_LZ$dPRi;+BIHL^hh$K-Pg?#QMT0+S`QXEQ{VfBrd?Xj#J@dzq}%4Q^jjD6}hDa
z0$tA6KOrzhk;gu8slBF@Cg)A*qG1PFc?K(F7c*Z4^(y|4ra<G5)I#D4G`p2$cn?iP
z9J)?<UAjQxM308~3dLEe7k3jSx2V0iW-UPu&uu>DVLL2J2QI6<>KN(M4@{LDBCkN<
zb7yddtSgmcrBPa^k5+IXGHh*Mp4ifh^)3dKD1qH~p-{{_dn2|A!_cQUIS=JxmbZe{
zCW9}`Y$Zacg&g?gYFMVllX`p{!Oha1TM-{wp~VmeN(L70Duly5-=oiag0ypkz<PMt
z!TRmjl<x*F_FR1ZMqMYWqU+l&JGdD_(m|P#n&y3g2;=D|_NS`bnap=w$=4n=dom31
z-#B}$T#>d)NE?4XpR(9bN&-T<%1^wj|M4?c_q;nh9C7YTUr+Zk3FrCtI-fim!Nxh|
z%$SjQR~_-9tnHxy-BAmq?n@Q%Oz0laS+5S!lwU|C{&Krii&E4ouXVa6x>2h?j>mt`
zP*>E#aVR*BY0=!XXwC+O7uVLN87N3O2jSsFm5bbw4Ia;VLOISXbl<J0@eVNo@{Qc1
zqgR?7^wB^~Hx}8OR7DcRgunrPQ7}AtKsZmTp>+EzZ{YI7!hp?F#>tcQ107l$=t>}q
zP($RW<3s!5sFtk3hFPCXDcBH1%cb^$P|SbVM5g$M+6%RhJ+-N3Jq>4$p5l?#rUt{1
z?J82jue{GYK{UwMO!?R8^q>=xs_G!U>___$&s8SQu8*1C88OcnjAcFu3x3L#EXRFA
zAre9I@%0Pr=(oszp!aR%5SMm4hq}t<r-4ogwI`?>zKv&Fzb18g;S{b2pPRD59g_Jf
z+YBxyDLOi%&Io12xaS0s%!ojgP|w<gDi-5*Cfgd$s2KWCVQ-JpNfMEB)>?<~Zpbry
z=Q)~NFlFf&tI~$q2jCG4<J2k0ey}|kkgdXtY$xKpUyC=Av@6L#APuX8koRrsP=*Oq
z_#RG)HN08if?^qN6Xa9SBzP*@%xPcN=E*2wUv`({_-oTk>rOp-ahD{qSUa+3UB%B0
z6Ig(VXB6(Ln1POWUH4izG%}2sChhB@E|GHh%G+!O^Eo!u1V3Pgvd&P6bXv)&R9-E7
zw)V!i#F_H@gdEWS1@5Cs)2mUXhBgR~%rORCv^{~5FWEzr?N=oO$ge3tbju9}<KjR$
z=!Xg7yeXz!7R2nubQ(EBt|V2u_MO0}zFcUKV29$0$QEXq1V0yi&9bkcs;+zR8cetU
z&SRbcK~VXlq-BZ~#RuFJ8Vp=t*RaBgL^s39^9%A!WmI(tj+8q5H6s*=ri}~hKR4cE
z(TZ*j{^9j-&WV$|?CubJpP_Zl9tw_cd(kgNU$iQR7>o1%^1~d;&^yhb`iM)4t4j$J
zd~CP5<A^pDzW5Rl{C%<k-h6+2_n?OXI08|2$E{D#%Xb>KlR<M<mP8lEHQ_u%6VbDm
zlV_+ZWJ7}BP*qZ{YWbM?XuKmWx6}>)CLA5;X<Tw8XP|)BalZpjiMP0dp_@?i&x0*D
zN=iiVn~<%-XU|g!X4I&jc+m~FQP%P&nTjweh$-MAKz)1TM1h~oPSNtLZ|;Ps-n*WW
z1)7XzF$fohNT{KG>H9we)~w#fF)rrnr&fgP=MQNZusGeYpj^qcM4jps2$5sNw_37A
z^l|!+mgz!UZlCGnNGSO_2-<`%I6FcRAW3{U`UQ(ZHMlmohf^cE4ce@gY4ixT&58<G
zd)v6mr~}Vn5NZBlKv#@$9R)8^5tqbO&Dx3*n^+mkytla{({8ngs+Y?U9R9M=)_n{(
zso_|%^fnu-<l&FEoOU8MquJ<B=|19y5E+r4P-!mvGMyR^uY$QTu00^8O<t8&qmnn@
za2Ss&RtFp(8zIRNoMCCR6~S&1`jAXiaW}z)uz|84ShIqfuIx*(Z(f#7M%PiIIMCQ0
zaaAW@R+jKUeeXVAau5S<p-<lBsk}U!j`}v!(YW(G_*=iIAGPXzP&PRu1oVLkbIT1@
z#@X-pCb;%&s9f8BL1y3vNzR-P7ZVQ^1*&>5jdC87_iE+vDO7IAX4X=x@9A7q-RANt
zSj6%SGUIIYiQ=k48Df3oV{J-AZ9cTtW-rUc1Y%=|JWids{gbpL&ilc~z8nlfw+&-I
zBWDQXeVhjQQ17#u7&3_eHS-Zq8)olQlKGF4LrT5_o4E<5{?9soh!E|CJ<DLb<MP-%
z)3CX#6|gHfo;QDJiJvna8zWMdu&x`<Ffa=z4g%rO$G!bsO&-^5)&seFEgHNRRqiSI
zn&Yg04cmh%2NMZx&zP4JCl@D@mBWu`kw&GmChX&L@I!_=972cU*JH4`CVe7M-7cEh
zE(zPYxc+)GA;f$zp8(Vd-6KuGXx_A_-xZ#dpYuUz!LU)8fkIMAArW4lXGw?Dg^K$v
zUQr_)sQAj5J=jH~9)pz*W7zIZktP8{T<XxO+W3J7PZ^X1749qC!(-$`7J+x!1lVeh
z{Ae2n=qfk@YAd_vwgyHR5Rbf5*xa<b2hPr9aL%ODVQ@xzlwUYUugkq_{#GFbS+=kT
z6kIb96EB$?klV^KN;_he%Pz7uDs%8N5fdG1OOC1J++o|V^%IyWm0_D_c)d^5Ge@T^
zPqc99?BgUG<6@Lwe>7ym<K#nC{g7cqWUfvAg{>3sJyJzA4PL9nwuB{5=BFdF(slzR
zW8dTj+}W<ydEo)EQ1*jY34d|WOP7xNT7v%?SAaMIG1Ur#qq$qfK=bhS6Ep0I-pNf?
ztMT5!d%#Bqw9G{mhWiUQC<fmCWU;V38^d`jAwE|@#MRjeBQBz4At2AtUH!&Vxw7NV
z7TjO*_<3nWkwl~>$vfzx)^i4)CzCfjVd(3lN=hXg-`gJJCoklFgZPHL<O+znf@T{n
z2&2oLM~0sgxs?*IkwXr+4c9r8dl#30>&gYp88_l`osPVBnygXJ58BNLLR!v;>~W>w
zh}0$BmN?BFJE8TBjWbnWvi*12QRdM}&y6)AhkAIQp`{Zd^$N41avj%)BP~R5ZM=RW
z*^&0KO|tsRk?YM|8@5hRe)cnNO_U@g{(iuY!?l-5!L-bl!X@7p;$bvJAXyarMCX&W
zRVJ0lYDCJI!EoReIRo07O>g5;RUZ~e0_#Y=30DUBm@Ywv<N5kLr*6_6?)BK<=LRhI
z-Y@CZH<Cz5@U^Afad9{%L~|P}^pkqQ$;u}(I;KMoy>>R#jKu_UPSzfJyN1Cg7tYvD
zAl#1S;f!Z>Y54TTZMN-*eRS*C3XU_&1GXyhP8_-2n=*3V{bpwcl+H{pT;oDWZ}Fuq
zfsa?r<3!Vo)%Rpk4SwNH{5I1O(@A#ug_H$QN8m86QfCyxg!DHvG~ot)5WmM{B6d8I
z7vbL5zS`YS$3Je{kovY&jZt8u{TOC+Y$!!s#F&4r^;Um;3>D1Y-zH>+3$#?*QcLer
zC4EtHG{JM@%%HI3@B8LuD(Tf4s>JM(1Klf=W#XTcTZ9a1<x1u|jvXIm7*NtSuS_RJ
z1bx!o?vH!(I8REES(kmW1Oed~3p#T!SvQk8e+dPBzHikFCGFi1Tx8cQl?;8snxSXt
z+|S1+PmUe1nJ>GQRrE>M>+Y0F<mN;)*IEB-QoN4{Z4jHdUVJO5j6UIQs~ldTl_xmi
z@8a6Rs0AF(UXZvs>(MdOn$Tw0nC|P?yh1N}a=si|tJA&kmY<pTTBMWjh&<ET@?fzp
zF`5XwVk$*J<1zN~g?gsA(EXPxG3B2p(*{YHLiYl^Hv<yBmFluP(V34U8rXO%Cm36_
zr0OxwV982t60XIm$*!Z$inri5ov%2KEMGBP$8~cDZFU_E6t~Djry-6(fvqK*^92cJ
zOr1ZvVbc@xYa?r)-MptM-1x1&FEW1rhP%SVH4`-c5d>vINV5=y!|ST%YYbYjP=8)Q
z8;K2Eo$-Y2bjfXdj{Uv9SkDT9;o>3hlRfb@l;7nAk$ObhbjaGApeICBq?djkrvx>7
zJ4bD5vZb~XyKb17j0k&a<Kn@STL@b%p;<I;dm>z}X*5c8w{wE*dT+7ka(`Ab8RZLQ
z8#CIlyLO#BXyF$I+OG?WQ?)QINp-+2@b#1r0>D)zFKHtuFM&yYuPB@6GsI)Z_C1BV
z+j;w6Hpc~<l_s1o4qHk;2~rRwg~eRW5?4hQ9W^^bGfRwyp>qX@tk>h`qr2K_cON%2
z_FycTGOC<e;sRxFjQH5GxK|qN^RGTP5YJx^cvl#g9jK5R*`bddk8yrR!sV3V?)Z{Z
z|4?qCKn2&9Frc;;E=N{bRFkwT><6)Emurh>QCk!HN%7rm6y=O0E^L*1UwxqIdBnga
z?I3aB$u$?u6d=n2Tq_V=m9ia%5bViS9eG}iJV(X!rru|K&+JW;2A;;Q!o3aph(ulx
zm-)x!N9xi<);1{^p8cqrXd2%ON-pN4Zhf^mQI|FV-1EtkQ)=CS?ncl_(HvFv%IhMH
z-FP3QGVmR3l^2%QvfO@T+<wM(L)c(W*nRBP>2P01c=<^UuKU|7{cGhT2Aw6J4Emjq
zmt1+Gajxnp|8$6ieTmOPNNa&DKq;pxF<=p;r1=M2m8+G=Z_a5__@9p(K)Z+<S^d57
zB3rNZG3@?i;Yd_L7oN&89jd{${>ZGytE=h;Ds69uJh5#fzYT*Ot+N&{gQxAMb9Ph=
zc?WNo1m7FqJU4gtv@XBsOwH!m)96cnYqti`NxIV#Va|mbur_S=bjydX^s7%@)3XNi
zrIM5D4N7Mi)_aQ3J)}NC6M^5<C+5w6o+jLA;VsjBo;xluJRRP^Vf8;>UcFAojT@Pb
z99pT1DnA(5mdEaQhx-Lz3BR}@RM9|lE-i-n84i9xC&xcEk)n<AD9woNCjR>_K-B&H
zrK(p*tJ`w!e#fvU8B#jmR}TXgpToFLcJS`stTD)u%Oz?)e`{yY$1h6V_q2B*(2>Bj
zxY#U#q{zX=%<9PA_-#&`GP149;@-=j1-i{P2z3fP28?q})0<Zl_3(}*#@xmc@7zen
zg~a*lZ!HzMrCa6HG#Zq#DsyX4PMGsbcWHQ%<XK8`?P~Gp-Rryt$bw0G2g9XDICD|6
zginNE7(R(RHd!oq2=1-=gyi;jCAEw1RT(O$c_hM~70(*lq!@bJ^DK(cL{Jy;!~`Eo
zed57;#DL=Q_%Sw#Gh|9pk2&87TpDx#fJCEJQmPZ=9|hW2<jp?J8H4EZgf)b#7ny5&
zC0zr36QylRIBn`yIuBDwwUH-`eiV=QCiN({t93eh)Qq*olaT^0W`CZaaKRvw&8hJ@
zD~dFuWlv706USr^RVNbOtehaUn%fI>ByKw9qPZOW9y7d*8n0pF{2V}K@}Ub$07`@L
znA~}ToCaxUs||6#o)=S@k@@Kag>3(2upfIIqGetCt-cuResw%StlS%JV&-)}%O_mp
zgJ~L{w|6h)l57}lMP8P=DT?Dbz;`bBExB^|sS<(z;(E1}Z|i3Qsf8CTIoDQlWnsgW
zy>?$~5llv?BcJ1ewC$s-*y!2)uEhPx4_}itS)2xLY?FTp7tfb6=%7-vya{>9wi|%t
z2qT;l@7-#ux2@fhci*Tx*miYO-!!7_$nfyFIGJ{^B;a<8IUsU^_`)UDL5(~~NcX8Q
z+g!JT>4<|T4dkYXhOV+%;B!4^IU_p++oWDy()x4Sk3->$4RiBw_D2|91GOtvicoAg
zPl05&qH+O-;Z@`UY*aH)-ncIzF^b6!@`Z9U_9(XO(ENq64m>0|zw*MHL84Z(=Im%R
zw{@V3_@&e4r=+$%&8X}_xs<B)z!KnSYVvS(i2ov;;=}8$0U-!GS3yxw9u1PZd{5Ki
zd@7pLduImDqZQjK<ievhIjQpX0X1;S6-UNFA)34Q!Ff&mvnP-K>9_mg`W+36(GQ^x
zZF4QM)69|IGAzLQ4w9I*0px4Rot~<XEOv9-bw2Y3)qCIe^zNM0Z?{`<Da;5~Y;4rY
zr5p5b2ZZq8r`%J7tEa*gDMtR{ly{G1eTxz_VyEa5JzC(YNJ$_1rt9uoH7;e^S9P!7
zcE0_L-*ZKFirDSB{kQF7d4s%A&pe(5J~`@l%6o}-5Zhm(R=Wfik9F|#5=p32gQ?5G
z50qb64~@oMA9#OfC5MY!i>$iM0HSVT&8kVgxrFAUB6TtmU;j9I8urSsI5t25-hnA6
zSt=E3$mhbRyp|~H#tX5VzRlcO?8=a9G6I}dh}*1s&Zi_BR&FVCj=xT}zVqQ05AJI=
zWTb7@?9ib{@pUm=9{~k%7gC)Qh~GY`bOZ?-a3z#(p-qf&-kba5;M9~<?(Hw3>Fksf
zQ7K{BI%p#=x?q@zhAKpQj8aFL-Tz$!8HHTbw!02}uH8>=%%Zc;{$#;px-QhSwbIx!
zcFk~n7)Fj2@-xal{JO^eWpg{n8t8Altt71pKiL@B*v&PV;bs0jL@0tWyHDXeTde<-
zi!XNP&YWmffJ;(zym!>gbeQ}?Oma2)s=3Jx56y=$Zu1*zvw|M;tqDJxj;J*1+f*QK
zBHmTAImO>??fdK}G=Jy_JBWh~+)~~YEuKjT>{(9|v`BlPKlrkh#GS21eruVfbyGom
zc}WcR@^%#~B#p-&o0~1<b9#>)y40TITyvt3sn?&Z+k9VAL21^<3#4QOi*Im>Asi;8
zwT=!N8iUqjT1XA29r$oNhj9;$m7cCnZrEmz8yc!3i5GTS@LLieD)f&X7_>{p%XavM
zWJc@6a7*cgdc{9#o+8Zg;Iy6Aj-%z78=*Vbi!Pr9<tXP&&Pb)eN+0X!-qxr|UO#hW
zQB6&V;5q5-=Pu$h{7k3Nyb(PfnJQN$A0PTDC(AM+<QsAU^4^L|i^~vp@KQW8E#eiM
zo*~snc!PfTn<`IU;eE@z1=_x<EEVd+Hd%KOTwc?C;N+B%kz?lgr~|Sh$hf24aK?=s
zZNU+OZUc8!lkpM;i8&waA096|p3WdqZLdESSI6h-7E;?TyuY6$QZ{{Yly)JK#%ez^
zx=!}v(fvW%d*|yl8~uhC$0PCbi)&>e5Q`ALnYjiyBjPSgh9lS}h<GP44f2;GbB3jY
zth>$gs6evTo^RJ9I*%UtKE($h_{yR!-v8ZdnU+E=t-Xtrogg>2L>fl{Y18Zd<LTMb
zd}okJjbM9<(%gkEG?bM0L=O|#O%t+4lO{sDefYwg1|{Nh62!lK%p#&&R=;`oE6z)f
zzQUJ8Eb8_4Kl)0>!>hIgTBPow3h~F4IM)LY^*+0BWph(W*;m``L~WKHIZ=>gJ}A^l
zjeZ-aSDIuhymFs568Zc#<?Jv6-UAT?CEdiQ*|S=eR}H9;<uX3X%DHzJuIn!h%ECnP
zRvEcf!I2wY7eNV9+(eH1(-(x4JZK+}jSKINHad6Ym3RIPH|O`yox=+Z`XMi(q9<7`
zc1{GKWPG>#lwMf3M$*}`EwUJ!t>KG92;Ine1B7QZ(oSOGtst#<ZvBo7LW{Qj;L`g-
zHgh|PSf#o=6QG@hli)q*Q^0%U+~DCx81a|AMX3vs&c)+^DOcJMq%~(rxHssF`{;EU
zyb;YPTWC&GVe-b8AK7isl>J$#E8mR_%IJ4pKcRQBovKY7U~4SfvTbrU4#M6C+39A)
zhWvH5-O00~M+sK+f(-!W&O**1TJ6x`d~m&qO8BE|;a6YdM9D8CIl2Vz$PB2h$XjrF
zn-X~k37?r<#A^QtpZy?dpD<fHSCB=!U+8loM0-%CSO-1PBnH*OY|sz9a(d$@!fu}o
zD<9MOobpzV3A=0tNGwV~hx-qJ?*wSw&*#mIEf*a+FwXsD+`!d-7JT{_X-^mX_hR7|
z&IrkTxv8&aOxd!HyL)!+e1Z@y%JrcPOw<Ht`m=CqfCq;-g+uuCJ<F&JebXi@_=@ii
zjj^f~2R%V~Um0^<<*<KTusC?r_{m6rTOkA^0JULLE=4Y{yHu7ep`$V>Ziec~nb1o3
zaH<9*dwE`4E>5LAbrcG3v4T!Pgz*?uakgXIv{rON&NYMI$WiUM7YMD0o-HIsr4y5a
zKChybeze4o$a1QnpV+1miab4eqO`kNmu92TQ1}foBv7Sp`R4Lc(REd}L64R2hs!O7
z-aRZdu;Go1k&$KW#jy<9|5J+V)@gLq=jT#2)DY2n6KlKVaFu4+s~f4xQfIAt1y(<Q
zdR9}MG^R_sU@e0&9#Zn&MO0RI2ioSexk8UDdPP@*NCl*0zP{iI?7L|a{Fix{=$LXt
zHE<ozIl|3<dgg1A0J$ARin<>m&bZdu@t9LwK+sEEk}SgHW(@@@hC`hez0&Wgmid6{
zWh{JES;yV+q2#rC#a06^9;rwZJ)){g*9&(&&Z^W*@wC_lAId<8Xxz8lVNEVlN<U#f
zH-2R2cEG#fspV4JW{)E?UBkS|^NYik5gpB*WG{~Xnn7+7FnH0%FhcT7LNeMiO@Cs+
zXE*jB&A{CK?%S_!Ui9HN4%(wBOXlPwxW#KxfrYAa{EpDc7SEVB^gg5+?rzQxM@*ms
z(WTD8Pw()K?Cq42g4#Nv@{z!St?HK!>h7sSLE34PyBBK?yGKW0UH6&8qvXm2FId&)
zk8G*d69j8UnV}E6HL7Z^Ev07f(ZEDy&{v05tydpk^?PjatMe4^8PxZ4;gW#oofn;V
zs}j+2^=nY)@=Q8x-Zj4fTuzO|(@6$d*W7W;>bTiNn=YCT2iYKmuX#?vkVj$~Y&dsU
zlUM`^eS{$eDJBK8A8(eB<CiJJgO&C(^BYd*c|`1^cj)GZfNK;<*=O>s{Y0{1h~<F|
z?}HjWJY!?-5NoJCg`!H@XKY<c<?@De0ZDkUYD?f#`A<IkW3Sv)<T`knN1MFGG?eKQ
z82lj_B((UJ{#=|1xrV1Rl)WL{*yC_2xfq6PsGnV{i)lWotLWY?5O$DIH6k`r>fF^3
zhtNna&9z~Gxi*+zUYl<^EOL<qy~)moV<b!NB;pfzeQ|(^38xjWm8zA$*P-iy*20_F
zk5HDKdmBlb!V4n%7P*^Xduj`~&%{I9y}kiBZ+88$?=8uJPf16owinn2Q#Mtw#5D)$
z7ZDemK~m_#hI7$M#fx+7^f@Azx%$}66!Ba=Ed}`Y7^7+=0dc04W5kxnj*FDU)t15U
zG@Gaed~Xf#EOznvoML7xRnr{+_ZXS2@HAjhy*g82B<l$Q8J^7FUv6H_PDB31Mb#RW
z4u1H`R)XMWDG4e1o$1QUHl2N%klh=xcc7b|oK$3NH)os(P?ZV>A{2a?%<@NPCWCJW
znMx|qD)cnhP&H_lA0ioE@u}7ethaPLSMnW13=wCPF~*c?=;*%baioyF63+Ip!O>YH
zSeoIO-JcE7e*;ud)*6C4zjjg>C@!+qB+LUnzS;sE1aj)a?C#{n!829TAjd#(Usc<4
z=#EyvL~ZtzXwtfV{rwp;!Nq<ra?Ql*kH)9QV3YPI$`T?h8)hERxm3^D-6wsu&(0*L
zJ#d9{ynw#ZPJD$;7&<=<#K@EpXRbhi%8mwGCM~3Ry<l%gcrZ4vCvsYW+;TzRs>xrH
z=pdzOl0*_2tg8a*t4m^1{SV=x`%xZk{pH%DR)|<aVV(sUuMbtOFCG66U2h#0Ww*8u
z1ELJ5z#ySCNQW3S64EKs-QCjCpdck((n!NF^b7(L(uj18Gk{3ffV6<XcaQPx{qE=Y
z{pWGutb461&g)!tE)KVwBs5en(~siYt}{4>GkRWG*qtzF`*htxMaF}tj5=rL$GvFX
z{O5E(zw2eRxiiOs@ASJ%2W#_>3ysVr0F%+#E31ojkUICWPOsC?Mel!jitrH1a)h%o
zNVqS#-M4uot^7z#U8ACVla$dV@-+*R?7`sUk2sA{f^Em>o$;jg9sU9rhyBQQqv%Ey
z$5qMnl5}^M<*R*ysR(k{pq=Kgz1^)Yivy-h)N$hGkOZo?Vq`+bM~5c(9*?k@Zi4OU
z0uIz-X}LP8GMSyyi9+eXp+)r7UW}Yzt-iw3<STt#8xkyFm53X%`L1)}U5jca^L<|p
zEh)i;29a|4=8fDmPoigvb{?}y)44UMv4bl1{J?M!|Avj2-K{U$Lo#6f&n|VSlibZM
z>~l#vp-a=x2W`?4kLDOD%-Z+mg)Kb;=DzhfKcuUcB4*@Fh<(h~_t~1Ak#du*t?ryB
z=wx7qe=Id<!0`Tbb&*RLckjolz(Yvdio{%bOPVMV=e_&%S&@0KuSnAQL{C^c3l+cJ
zVU+VB3fk*uJqa+H>-~A%CBL@qbOXY*3x*pivqcxG$(_4)?p^|SkVILV?(por4RrLL
z8F$I|`f@GMfadr^+g#A)IknK|_iIFPZZ`lDp2^I7ZOeyl&$KzMh}Y-N;?qDQJyd_4
z#aG~7$`$5kF(4#fN*{!>jEH<n>V1oJm$epZRmkDwRb=7PZmnefYQIv+_(>d3@m&TI
zaLgb<qi&pxCAW`ZRY0JSk`kj-V|T8thOwM!;;R%x=x&-Yjo<Q%rXK@JU=^z9l(6ST
z&n~|l5A{+UDmIa^-|jnaYI>hR=kC48ziO!Z`P+)*!=)n8<8NUH8{%E|l^Vleot5>&
zuD87>Wt9YrQYDg^HQ#2xoyw?nWwh?qC>~>!l2O*<tn>MlgVU<7H(bB_m*bxDgCG8Y
zUhWx7Ph5zuw#m%UYxIW?w3g1r*hY)x`0jg<j~9m7aydrp1}>U>RnWOuJfpT!uK&I@
z_})`^Lz3~4ob-<MI+Y`f1-LevVL&x}DJ;8497L)xpzy#XQLu5DsV#E=Y46W|{i8NP
zhr9iwRY%T6$~(FE*e_VbbOPT982W#~*Xu*?NW&TJN*IgU7drP<_ls5xi*{dBcKO+E
z9CW`3lKHT;*yp6{?GR`|#}b=!Q{lvS0t{wlAO=@HW;;0l;Xt>X8@VCsx=!bA<UT5}
zM4#}Ei;28ga1QSJgcWyL1N%;ZASJkMY28UVWdsra^astCn<&`r#ZJ!MtK6a7?h46!
z>i3!QQ)<KqA7zTK4wWnd2Wf*MGh(|f_rltyAG$Lpe|xgTnHAr7vg3Xs*+PpWGKTK*
zuXO^S#ZVLXvCjSI_298YD;DO>Kk{Kdh1K3aI~Z_IN0r&8HL@kP)!Yk8lMnz-$k)r$
zbSXE;jZL}B5i;x$tCQv*D{7iT_BHas2}OP4mLv<25qgER3DIpAhgI&Obdn9ycT-Qz
z6iK-{qQgp~{GE;((oV(B55~DT6KoDzBI+1tHqK(=?;^?76i96&Ju2d?3u$m^0wk=b
z<m6MEHqInP29f3lIl$JkMX8<{>U8=?qpJaep8k#i7}h7Xas&REOs!&RxMT<Pd<D!h
zF^-VKWo%4Yr{hAvrP%~xBb`27#GG?kgrM6lHrvFkqMsMK9D>KFZ%Mx;QM<l)s);@L
zmRC%NP6g?xe28+Vei&5$Bm;}-VXs9%6n0Oy&+L~vo7{U-QwY{VZ@uSld?U)kG%S_`
zP!CvQQKxG&W30vn^Z`QKRL3^X3g13nNVlgRn~DN=QK;XFv_j}D24?t%r~(#nu3F@)
z6#n$uiIbh1J@wmn`7|p`dB%;oCzXcFv6<pwbNHl5t1&Zm+TvSUT7-Uv-f+IxW@)dH
zS5c>TbP|j>81$(&Si6q!?w$P(eFocJ#<;D7_%gX+nz+)^$O!uwl;-B%1V1X*u-azf
zD6o+I#F@@$S?L8A+bhHj*Q?RE$h!%3>|OWNkK_x9OHDv;1|#wU4#%e#2TnC7ZZ%ZR
zzLBqwC?34-=A=l5%m9V>tjzA~nI_KV>0QQWrCppr`@(&gX76SQc0FJRcU*leN`Tmr
z2Q5BIp#5ekuRX@jo1;0#lVibfUw7`u!fHgdU$Uy^8pY+!*Md1ho?M}WWiur%ch%wP
z6-D<1<N05D+^A-py)xPY(x+`~5o`=QQ(n3dJ^Og#@e;jBZpw)osVr&$eGhxN_L|Y|
zq*nWq`JI8>z*!Ncxswnh&A_e*CHqxf)AGrOw@pagI6}+h!-%iHfrfvF$WCnPgT8!}
z7rD>x&LP*#sGiUv9Xv(B=pIQd>UNR+9<iq(Z80~H_&^TM-$?vYUes&1;DsIAMswEw
z{0iIPZKcH{3vig<2GpeY>`1ecDl#&DX}ox5_=(6O1CPy6>z%m1Bzb}m7Bu`*!L`KU
zXq((ChC}Qp%?41Lg6-@6{)bh&wutnwk=U#tPM$1&@3Y>lB>ob5P|tD|hgw|$!vw49
z2g)F+xya%1I=%d~4#QmEv*~s4zWdI`O1@LgbCu5@=fB%hW(%pIj%PB3%zgP{$<&`s
z^c?>T;}A>I``|c#{!zc{(cy9<{+zPv{cpMq0i8A1>3~agKOdCLBO;<d9DdNZTJj<h
z<icsWXLns^Zd{t+GVXF$d{5(N>v&tx)+-OjraBP&2Adt}4~o&4dMPFna0Hi?xUq!+
zF6B%7(NIJIJ)6?K(bL6XUt;*oiKHJhCAfc*ZO-<}iE?%t<3_b~N-a@<sULElIyjJ5
ziM6{CYme|LDg4XE&){jnG`56(EC0iD9`QQv(Jm{4Q%OztfOj;;B|*}D<adVl52$39
zL!L1;k+OW4)1^2B4)D-%#lHxn=@rRRe8PKie0lLTSN7eW{-TesnG)~A(eZ<r#=+I~
z-8^naF*6C14Q8)Lv@ydU!`*8cMemGj&kVHIhY2p<{`SMPH39aS@2n$1EX4w%7qBcE
zEWMl`AY}bbN%YMt;rZd)Ck=X{P}*Z=oU90E#q?=e+Q&qj0+OGnhiNRzc;x4UPD_G(
zPsJ?mu??Og5V8#yV#KewR{dv%ul>$aRb1^UopJPJ^X~gxZGQX4OZ~Fv=AzDWF_+|c
zCbU{e#emsb6~4Wip|3%Ss=65zty$BFs3bY^kSJoAIsP1CzOR>G>iR}fB0`zbSdFd<
zK@w3{yOt0hO9D5(36cm|%Po2Z;qGc*wNtR<2DeX0*<*FBtyzMW5#DSYX0EmGbU|BI
z0n)dtq-lg;k_X&!Cp{KSr^#m|=ajiQrA4=2D9m%%T_=WcG06rjNzmT-rnJOuL6uE;
z)2rs0%+dvOw)6!K@i}t_yykumumsr#FnBHLyH(u^XU~@UWZr5J?MqxLmPeclo3>DA
z=}1Ot^sp=z?!0jSGK27H4r+Vv%db*7y{GBwHnpWB+l<;eIE{GvBlmJYNNlj>xb}iv
zVSwTRBe&Nn^P$z@rFf9v+m+PJ)&0OPt!UqD)M8{YS8bAEam@ALmJ03-`V@&q7|nwL
z99`xe_PrE&#AKgUv@#Pi>6l?G{%4u=YHjHqlx9TYvZ#RmKugqpM=pP75*hu$^;?py
zZZOuIfrs~(%ooKDs<gceL!Vf`rk<Hs(wa~qoJpwBjSqWVnV`tvA#LH_d@AZ5bAa90
zMPanJOyYyoFH`U!m+$20Oz3WYU64CS@G?u2BxT-!VL=edLYcZ-GyELnyn<VXQjU5k
z&b4>#j7!xbaBrrom73`D&9f>kMgmr(#EX65brwp!@izSc@Dqw&X(zB?my-M&L6*L7
z5nfr7Vz*CksEerFd)*5(cmlt<@Xx*S-|pD0IWU>CyL>5tCv**k%)n}RvM&%)+NDh}
zM!+5}_~brtCiotY$f{kDGyEI*E_C?10kp`kVIUX1`{i@niHmfS?btcBQNY>HH!uA3
zx>f>@Y*;<hTJP$vyVrp}XHQITJ4Iua7@vL;vTL3bXkO<G$Wwj%X;*v2kM(fU-KC;r
za(9x>JKS|<$MQpX4L74Nj~yi->-fL{bEw_pMo1BC*~8+{tBGQrh=U5cOv9B;_cbDE
zCMKkw_lJMIR_~3<G+`2x(|rX6c{S1p;EKL4+NvxB1W!34Kr1G#mUojHl0KjqDypJ$
z_Z3{FsdwsC>qBg1W=OOKF7%4GKQ`Z&WpXRysq<Cf-S@H(U7H5f07+c5rnVajxkWc|
zfo0{Q^MGp!I!dC?{BoxmpB{ggoqgiHzpBzom3Q=o%w>&>$9vzW?v*u>vqr1`3X1fl
zYrW#c_o1F%8{_Yh!k=1qJ1fsj7Gso$Tl&WuQM#NC-lFG5H5QOsxV%2&{l=80*+h8z
zhPNJy3@;`d*qI<qZa21q+7T0aG&utIgx5%M<SW8pMY{4^*2d4?_`^iFORGiwDQ4{4
z&s#(jnxu!EZ;7G^=vy6$ZtPrK_Ya_Qf^tf4jM$#36Vu0d`x<C-{+uax$EUYzM8~7B
z8e3w%`khQwcfO6}BGNdUD-pRPb@2h;3{e**UoTGtraJ(Za45bbdrNxGy8RcMx3ZcQ
z=G!s99&F;`tv5+LYR9vFg*m-;8Y*!%WPghLwiD=dQ!ZBik%Fh#`3JYVgV4bezI-1q
zwS{ZRS=T;d6x8OcL2PzJ(I3!&k{HR_V;q^W#O+}J%%%5^HchXFsA@zpvSK5d7?DGv
zs#rxp1MnqQgPvIBP!T7CJ;gMm%{(nRXmO>_U#T2(bgmuy*8CI)qn%!I2w75tK4>Ah
z*~<Obr6G(YX#HPF(5<`JR%o#fm_4u{R|l$)DSw5R(z|Y7D5k4#GIa9zSRE1M!nalA
zc?{?>o+d~n)GyI7&4whx>EdvHNqy$YVkr!8>=uf=_gGgq&YqLkk0Zp~3!GI2bTrT1
z*FeDhgdC(PP`*C&>D)t<{FXug^+>c)-8M5vL)g9MNX)(Z65RM&|2sj)v)=AZ=q~9B
zMe!}WtZ<BCV@M#=Ge$e{AM;n3oy2LeV)X#a_~7I$AO9p3kpOy3;&!z1Dci&S(78Oc
z-pSyBkrV2fegZC2wP=X<oDXXd105Zm&X{xDCmoC+%E9B1@e<XNuRm7=>?|#r#C1uB
ze@oZ3NCBM{abK4QzQ}^967I-XektLRYshnU>2=@458{qcEgX{Ib_7B<Q1_40(2U-<
zSu$EYj7fbl0&Cu@Qwa)rJNR5s?xF$C>G6V%Bw!2SoD&d+v~U>wWqF_vfN6gJGvtB+
z5VFf{^eQA-qr%}jgFQ1+7Qbhr;_h(W?yLI%r7lXKH}%TD)iAeM0fz;S!JL7J0b<VH
z2I9V6QUtLx3t$gKok^PW(1eD-cx0BY0k`K}z_;&E_w}0)&fkJJv<p$<Q)7j^@GE}z
z@rrF93sbd%sjV%W7UV3&S@|=jQ%%uW(Gq(Ir7x&CKf5mh7?ahfx1aX-`c|K^1vI=^
zo|opf7}2ixVSzpc9N~epSn#SQPo?B|>_Q<=9REfbl!ofa!8mB>;@cZv508WAl<;4&
zA~f<?YR%Ih7B0wk+9P2~Kx_`-$0r>fZJZyR?+x4eSe%$L+OJzj=o_JjRtH1@Cv3Ak
zx3}iKd8^e1qXZNS<PBm%zV~o|@bDfsMf=ulqo?JsR-QM6)OxMX%m7j?YY%WHiDA<^
zF1_aF)9O|QZg~ZeNPSOqh$NuuBR@ZapCuv;C~^R(DF*hEz)=AyOD~=EufJod`jL!#
z6pobuBS4*-IY0rkd4nY;_R{;>{Mcouw!?DHfP@Tx8a#m)$Ei97O=$+g;tJ1uf`IN2
zi-spa)6em#GjA)b#`Q#&j>5u4!0>w^*KjQEzFK$uAYjvDn9=oW(G*`<MMcJ!S~ZF&
zwFwH`f5?_KwX_7TDOV-Kvia+d<bZ4jhkOkA=wZHOduqAS;$e4?!Yg#Ywcz}?sixBF
zx6Cad_2W+$v@>Iw!RYNIK-ey(l!u4^9KF=F^xXG+>8((v?!!la&jJ+rJr5KKkOxKG
zm5cT86+q`=vMJ!Mgfx0-X(^|`em$?&>uYg;6mSCV6=QI9Hg$$FMURO>?q~mavA6!L
z+0<g>bkFhj!3}2Qp|3LRQ5Zi&w14h;c6tCS<K2*yrQzlDD$2#D&c6MW?!&``;+`X~
zO)K$79$S8Q?~94$#L|k|hoa~1H#GpD6^2U;8y_E!ysm?0yF#UgR7=2!7GhwVs%D8|
z9h^FqHa<Ivw;s5x#w84|<(cKo&6_)f@w|Z6tKu6Pie|D*JE-lkKFzX4DX8Wy%uJi@
z-+AHZ#T)?KdgZRHY@T0e9fP~!^Rl)J_4Sv1JLvB&^RV}wC`GQk^*_SG*W|#cZ|7?3
z=nz$FW&Y8ny@rQ3-qj@?xW%XH1YBK);fbfg*#JQ&dhCfOo-G-)S~lb8PrwOBVY=b^
zT}rJ=j*AkgXuBS)%$OC(+sZq#2r<3N;wqPWMVmxA3+8%M;ug)<!&*|>LvG!U?(-U>
z{ooEKSKD0g3sDWZ>HH-4?fpXZr%#Hha6o>iU$BPT^aN0q8PUQeuJ7#Z6cTXKQ?8g5
zYEFR}dCDX7fH`FY@Rxmu==HOr+=Z`2v)bbxZTZZfQ3%pFRYlWYpbCh{PJu<}?dW*4
z$7AGDsGUUGKBQ_-lQz&i5c0~m0nuCZzP>h|ATQ;1fzW*{JNEg<b2n&PPWnO)bwnbZ
zwPf-M2CV80My9TKrjN9Wf54>=xu6fOVCI%=mf=3IPb6n}SBZ`*=vM>6e-L!zwT$os
zzQkD7*upsD>^CZ;ut7mjq`QN|<~ni`V)xvZt>p;W{d6<u{7YiHgxKc&L@y<Y){!T;
zM;f*sBlpq2<fGsieao9gWS9g{HuY>tq?O+OFNBm3087wLih<7!w_f`|Dyxbv&)Y6_
zhG1Unhto~(29DJJKqr9YydkgjI9f`;!h%2*paNt-t7PYptY_m|FMDs$qv+sMza?|e
zWKDq`U?br0jZvc;vp8_u4Ep<6`7aCyGMCMPD6I_&Xu07@<dLYs;yB_oJPVL+4xLvz
z?aWOd-|P+o<lHcjEx|)lstsBHdU^hJe?%8+&AaW)h=*|e0p>0f0G&W`Rbo_Y%hzk`
z0Az9oRXA&oozoOc2tMS^7Rw3bIX6G;^l&}!k3iKxAC0XPx;C$;pZzHYgyC3fF!A#P
zjV3R&{x%jT4A9pamYt>@@<5aWE&nbWF8~?s&5xKW#v5`Nb68&ue@iZqYCyZ%*Df#o
z@NqWFwcfF%5ouqG1@lvnm4btdC4==J-xDHWhXr+3Z=HeA2EIn4fF)#*Z?A+GS$Q6J
z`6F)W6PiZI$_pS9zsDfPkmtIV6zTPrWK9Q0dU6i>cxLo_LVfS+*vAdS@`W;jqJVA`
z@1e7A(7~XGzY|N0<Bio|r2Z%Uq^WIC?QaeL_IF^3o(td(-1_O^=J4UiHKbG~3Dqy1
zuU~yo1&l6w8!%W7b2XUK(<gv=aNE?Z=3_)gHf5IlxRi!U<M-39o)^LzLJ7tOLTRzu
zHe(Vmkl6e5yvxf;zdAbNgS+8A$$#W*WyO5H8gx-I3X3uQY4lHcP!j_+5<3JAC>GDp
z<5%Y)UTqZAr%o9-`)8FR6AJ8}&o3+3hTmoAkf*GanUYOT{1X%->$vc!a_)1MrlXBX
zh)~e~36^<Gu;esy1L=S#{VRAY#OBl-`j7=R?-5T4jJ@-&HZ~^-&;F5YucpI=16vtr
zZXTWz@t@y_B4z%_$o=*wo|1%~U-btkKpt0+9>;;oA>qRG9=A(H{mjBzyRESB<0M%;
zIfJ?0n|~B)XC>#Y)L>TBveRAzZ;+~gd4z8Qz&t2yCNXhXM#4_KEk_WVO-fEK^1#D`
zwfQ}uVT<`(o&&JedO3V~BF=r7wuThH*o~8&hD#6p5}N%T-e-W$8mf0J1MmwWLwkq0
z(&m6rEqwAJ>4^<%ILlt=AC2EN8Jw>{mjO+#i?T{mca2^)Iv7YU%l?iY%!i-=FgbPG
zSpf;dO5CRgOObo$Xkc>S;BU*D9TC?3Srmo;$d>MRr)F{W`?XLA>&QxG#_9jk5Pp6i
zXaukGi#E`gD@@f*@C;f_eG^?rdO(u$<F1FL?FyEkoyMPteRa1wsT4WJiMs=Hm>wGz
zzx01j17rSc9;T61psj#FEJ1(#14?XiazE;IB5eU}16t(Tyu2{KKU2(&8x0@gs4$!x
z!+^rto%8Jn5UHuDV>Y<~sc$OLsbAo^f5+z`0th8H4CkE#`&>D9B?v<OCh<h_ABx>i
zjcu+*#J+9(EmQA76apxJR4yCBVhT7$VLFwDcK-ce|2PuU))An2>_wNSUj!(m3*t`Q
zl|y$R%`*%PY#Xn1UXil%wSA~^BNq1G-I6F$SHQ@T98b)4cXuD0tmd+I{daVl1_9A|
zoK^+HK$(1A6D?r+a1nb~E+%wjg2g>9_+5MF-^%>6lEJ0OOaYhevSxWCp!S#9?4?RN
z3lQsH2mmlXS%3>uvNix^QaGz!R(i6sqN+H2l9l^7Ijoyyufu$tduQ?W@Bba*%uwCb
zlmYB<el3cLi4pboF*9S3n<go9_@%b^E8R&D;(%!NZwic&=0xcdshtUG3Fu#v2=NPM
zJiS4uB9%b!YX*d&1bjH?h;nXa!0@nw&VSzE7%2nXm@WM12++g^h?1f_@@#TSgPvGR
z1W-tdy-D{)P2|t;NY<_az;ywDf(-z4126iG&eGa`cCR6w1+lgY{-UW7kbNkahD|R(
ze<V#sp(Q7}nvP~pWmas`<3rM?SN<szz?&e*J6Z;ac$|+juokaELE68eB^(1}-nezw
z1&}`?EYD_jQM(=Hi$Ygls=-8FqX~Z*Xk-^xTU#4Rt7ULaWm`YrA#w)x2l6o)P>)f3
zKUanVWlc~5hl>mT0iH;U#y}zOTNXT4H_N{c209T@-r~8bZJ<F<Dl0ub_ulpHIZz^6
zUJw3bA>f4%%<D=KcGb5q^#{MY8gc%qvz~ku22CCPm=IAentBcIC&)1+9rdeZNq7^}
zmmeLsT?Xizn0##@93CDn!7TU7PH$70asEZ?yj*-+j#Hl_8sN}vu!OO$BieP~F8lRp
z8osNN1D|J}{Z5+j_J*&4na*EZTX@u9VUm%REpE$-$H6rve`88MPYHmRX3PAoVvK@+
z$YDfzHe^VSGIW{3WM(1#TgbQ2Qkv*B{Xe>yd1)Xn3~B-}W#<GvY*z6S3CwqB`>}An
zz`^_a$ob!r7~f6~goQ!yE+GDxw~wzmuHR(fQ(X`ndsCVql!1aA1?Cm_HIk4@`sc)I
zu&cPEu<0rjRRBY~-<h%gO~^GyAVBYBaA9QjAK%4b9~g*GMh63KshVoU#E9kNV*S}}
z^q;R9JpuAZ{Cr&%c&>+5H%Q5Y+sEyfIPIUmqk(Ci6OkMy0|?(@U5^H9HPd38kRTuM
z+$}V$``Z6$Yo5Nq{yMSy|8(dfrbC(cn>n%ZCqNaII%A!mhRDe?Z&Rzvg<B(|p4(d7
z->v==NrchuV`J*4TXn;POgc`Bw^vtI#?njz)8AA&F)adm_ZV&hB%-M|F7J)z#T8>9
zIv7ZRAd{Q?QeRtxbmNgVH>+7pfTXS=TT(+JKGRTF1o%I#h$~7zWDi3dziVpy)xcJU
zL}a-U5GtND&9kFx$k{}9@xQ6SG?E2GvJgcsFk2PIjCmHa?Pyjr`f=3bu{wQy!3`p!
zlF`voE=M^5*|D09Hyv;q(?2TxUdpDft`7nLiBP-@2)rgqdOGlMQYZ35;~(1fi&?dR
zZ_lB&(X*Mh!IP6ycbAc<CfFefHWRY1Po5EHY|H>xSJ&~|rOl3L8>(rn8BLYa0x?Jg
z?&2+P*NCEBbJCUd_A66T;`W&@trJr!ENjPdm~uY9AAEnzq8HP0>u|PB#c6{Wx7}Ty
z_%T)`l%6p-z^n1-&ZiGA7F!{iaN{o#Ej5NLHn%EE5v5A6Mj~1CL>@j|2p>DHc!#X4
z<~yeG<Zd>bR6f3&DO|(T6O4sRj}<a^_j!gIEJT~P>8K}$ss%5BObQE!`1jvP!SX2;
zT8BtL#+dY*8GDN&ecHg6A3xsTym^zs(eIGhpuzyC<S8pgW;MRiw0I`>tO>kW#Aj>#
zp0weP@LZ9ULeN?casFy-ga#+dI)y$a3Fn=0)fou!UL}{YG4R~CV*AuP;RCGZ*(m;>
zuVRDhnd3S1P+mnJ@I!EM%O}gBRaIx4#BlufLy75Z{r+W$QB2F&8!KfD%rDkdMUL-2
zH)~#JbcDFR&L;xKgbk9I@52oQjv@E7YD)@4R!s7=2A!e4_R!|W{vH|dKd^9mS>yHf
z3|tc}9x2U17U_29W1f|+I=$CEAWHzl&pj`q%8l4~^vaB5TON>+XL2j+&7IG#zph5}
zs`b3WnC@32f7`pkA2NiBCfuFrPu}t}Yn3W^@R=Tb8+sGvXf<{q&;f6I+NxfKZ24rq
zrNVBp8_&tbpP3mIpfVN<LKSwc5&fUGfW8q2OK{?prNE%%pErpG-vr0?O=TNB2t}J!
z`>q9$ksClN&lb#GGYO(qR(hNSr_(#kp(`>7IAiDL-wyq1<aalj7=kNQB%@nJrLfFU
zkL1Yw-YJGWZ>T43ZWHFX+L7L&obM2eykq94>1K%c|L6!X*IG7S1K!AhANps5P+~&g
zpc_x~JIwV4khabfJw}#J4U`>lgTi#Ye?0j2<0HAa@vw8B%B>CT5a?Dc_mU1VTr^N_
z!jK^JWG`PU1;uDR&`b()Ke;ntRr{c7O<5q+$wu%Y!T(xU!ju@F-T`lS0?%<6gqjoo
zuGr#u%(<&UZ^^Yt5PkEDa^RjOe_XXF*_JZFzeWHo#PuO_#3lKO!zDz_L>~Lsc9$$v
z#EC)iOREuXR)cVR5vT~?hXPyV`p?i?@NUGkN7LO7__*{PqJIzQ5;LHno~`fL*r84?
za|AHoSs*x6>wicVkH9+2(d4hIo#eug|0g;y_ON9Nu5e=N*l%SCu12>k&XzhO?X5(Q
zI<#~h83n8@D&Ol;Qg-m)ENQ3w_m5kiVCkW#R=YSv<nvOhq0OM7goFgbp{{4W%z31s
z<4@Nst9AK~p0!r%n)R@ww#msI8YDuUM1(*5|N8T`JkDS!x~J#0O-QxMFt=J|ic@PX
zsoxV0F8(YkID1+ZzY`kBDgZ6God&E${~uB1k$_bTkFHMVFv896@T%Z9pR}JOhOY}3
zbbpFUu`fbIe`+j6ewj%oGa5qj@`pN6s)q4B`3o)oIB84@OD{90&Xl=C0?W*gV8r67
z9%Op!=|YFOPa2My@XC#n@tbCT!ZYb)|Mtaoz!xpg6)Se7@#!5^Np}+POdypgUrzx(
znXD(t1zryYoC_KcR%v7rJwj@ye`E4Ld3!AmtJdrLP!^2<LvXbd&T_Jex~9|Mc_d0x
zcgxXi$_hHz)v}~{1|xw#B1?0b4zMvENbEe)B#P)~_)jB$oCS1W59ogS!)#o@2tsLS
zWGuUh(+tZKNCdL~=ZhI7_0g*IYVIk5*VQWByVW2J0R11MN=yS0-W;2SKdy{rmKi&q
z&cmbE2w=}`6J0KGrxWF!03|6zkC?Tv<p2B-3->ffVkbSORh)$X1rnFZadrP*d7U}G
zHdT#1%$|y>xluz-b}X0Zu>OR`JCKaLS*JDE_HBMrz5COtBYP>t(MErb^S`@e19V4Z
zAZhe8v>i?X9{`~=1c#S5%AGVD%!eyF%r#CFsD^HKx3a|0fZ=wJ$)<4sap~XiUAu>E
z>E!M6$bblp7)p((77g2P(r9uy#N7vuUGb0sF$$A4Yp$!a=oS`oL_Js({dXtC!4gwP
z<94-av9}O0)y&M2y&vJXpF^6Q22x<;JGB0KVck9M1^fwAB0sF-T3=S9M_*Zd`9Eg;
zRpSVWC8oTv7^Mybu+<qocKmlzDJs;x`(IkX4Otv)_Unq*kb(9fRG_fR5zfDc5uu8u
zN0mz^iubnNSrOL_k+a18tV{>pngmloJ}b7ZJ1JvIJ5c`G8H%>|>a<VSAKnTN`?rti
zp9^AJqQAXa!^T#nU#ixcUwz@!1=KbtN5>z>z@r3t9?CdMDP*WhbQdl2X1={mRw?&N
zR87@745$#sA`hmn|Eptyv_OlyKU@!k!Os*!!+tsw8iYEn61Evvp^Si_3AC&E(AZ(#
z{bh)ovFI*?qlWK&uIv9d<^bb7#{rUv6)T0|EVI3`dR8Tpm<&!5TToDN-g0saIV<Zq
zt_|6&k<&4&iJW;KV}<uhpo9Mi`b6&EzEZsf`08M{=Epc=UNR^K4A0EW3={&PmP*`)
zm&>73i>oGDW;O2j%C*aguN5&k;w=|`Vf|mAzwr<l6Me3jpE!AlH9MFX-k*?aL^6N!
zNj;wwB$dvPxjhbmEa;jUq;1m6%d0|{2cc{=_BlYnWUw*cVw>pZzXk{}9H#q%W_B$;
zVim&6xL*E*C@X{EZQ054wYQ;8@o;vNK#tUgXU4VJ`s_#gwW+X2VRoFB*Y>?tNMI-a
zb8^p#Zr)7Kq1g1v^z87e6<%5U?j(9NDE<F1z$egmjet*tq00y4*eRXJ{?q_P+)$^z
zIuth-e=b`&|BgCc7W)u$Zbe+BEsNvzvrx42k`tX%I`n^`01NlpO8}YZW7Ab~m_nK~
zVwyX=$jmZ{59~b$`R$w4`d#k<mMr^O7-Jp2=&`dYh7Lo&kFJtI939?lU-_GUzbG6K
zG<iHgCZFx;e$i~e-y$#|qme?enB~@2za@yu=FD-WBmuJzfly4|R001+wRKj2Ib7bU
z>7Zr!{tS^w!Z4J;u8KCdV)i1yUH>xrxmsnFW#NYS{Q$yS@CbBBBN>A2e?5AD4WIxW
zHKTz%5s-Xc;?X`jx)p|yu2%X*F+>pjzC)w(y(7L?V>Q{;+2ygfA|JB};GBj2CNBRN
zCM1XT{@djVzY7FHA_HGqgYm6<!Xiu!q78w=V9RNu!onFV%idN_FyF!L6mmmNC+;Kv
z6XiLeGRMpYjR?zzL62Gipn13UHm7Q7%0lh|5v9lxCZkNBf!4l8(Ov%Nt!Wqts_T-K
z`Hy2q9ss-{_MIg?=rPlXGD4e!3^g;uUCg&8{*y~b+X>RTykz#FIBsv4oW+UF;Engk
z;3G){;o8SKCrolYz+d|sIc*XoYVA_zm`e1m`}Kb}%>Xn_7-D%F#7h#fQXM0UV00Xm
zo76V)07{0QU+eSKxcy$NDMiGP7R&jhG@>V1o&wWofU>EZ+ws>|VsnfJ|KXsbD#0dc
z$C%Wszh#@!sm%_4S+8<NE8NmLxB-W1PLl*M1u=@--)gLR%2VXh>yX;XZ!(NBwE9$B
zZXv(ZD(b_3?SH|g<tacoHd*^lHB{*{NX$f=Gp{{i2Vc`d%vEveH|7*p>-G#S59{)G
zHb9sJ{M^kl-8;OH_y$q2%w`HE|Hg&}kg0f+Ooq?33<+LlNlgStlUF>z*AL6NlgW+r
zYJRX2N2R|UB(uJDdLgv2VsPC+1`$i=sayG<G5!PU^PwPm2UQSPr(|A&1SA>sipi0J
za{zAi)lPbRuhDHX$2A}Rmb-qdlM0Q`OJHy~G)-nEVl@sz9esa6@h@~3vBDv}z4`EY
zF_k0EDA1U$$5u{sxr(~3viHErLi)#6JK^fHo_q@w3Plu(uBQ^U7-)nMe(oR!2zmql
zggqZ;+Q7xT17Tm|s_EvI8!JGEue}Yvpt_qms4te=U|wSzCi(Gz@-Hlf6aXMen@yu5
zIWLb3ShG@x-wE230Ps-Fz>Xfb0=b3HGH3W}c&z7Ge8Now`|ql=_kj93^MH7b28X0p
zL`0<clIE_R-|CDmJML);@FA%--mzn*^M;<jKF}E4geWuM7sry&0j$*gGF^ebI#+@>
z9GxeP@!WeT3n!OX&GV`50Fvy{{>yXTTW}jgMO?m6Csu_=<hV~?2j|;p<*5@f+Nr*N
zPykf1e$X0)JVUw3*R%sArgOI!FJ8Q5Ls_z|-~1gg-+|&W10oh=#u=rCpWk!jJdEN_
zPJ?R4ndnx;vm+gi>BdVuGb9SC2OFv3%^mRVP_%DJiN#|8m85W)o+Ph6jzALL9Jr71
zuCT1cn0BpKK)~#ELzyp$<nJtA+yVV^*tJlst3B72^1fILV{*U3M&O(gGFg6UW*8S}
zM*=JI6&m##BD3L68)R;JJ+zCDYX&K$fEk&Iprc|{y68f40B22u<bBuf_Cz2n8EhDV
z`QdHc4xY5++Au?8FIIoiujslT4~*uD&z7UG3T{3-_*4zCAwIGLz=>InJ>wR|by9LA
z+$~Txvtw^h&(B^`fb7Zz_c0T%k%Yt(spEuVatKfcW5asVW;yni_iqv`03PqLkx>O@
z-3|$72WwY+4Mpo)b@JP1MiN91HXh)0HX4PzFdw;txb^IC<-CJ`Q=O66Yyj95;04F$
zai=I=6m<V;tVLSlYxBVXOTL5O`h+g$FPzQd2g;>gF4n%(Ja&jJA?k1|HHfcJdZgFb
z4>k)KYEJ&p&&Y9)bVyCNA{uYb5b-pD<XyB)yC$)%oWvzLy$BM#0I)84VD;UO+3`eE
zLSHq)^LPAE0!3r|1xJOQDgKsfBmn--08x1y)-t6CtL0A7At58D>L)~bnAJ|hk^7sX
zY>rwUZ$PLVab5lq3S1_KliRK))XbT|?Ghv-B{&$f9yn6|a%QYNK$NAQ8(s@ZVyxzL
z^mb%u+frvdfLCY#LY6IyW(vbfq~U896;{t67}aqehW3dVf4~cI2FEW+cz8tf?*mAf
z&5~!rzbjfXuK8z{cAAbQTbQew)^f-d(qwSr1z!q<x*?8=tD%z)UTyVj1kv3*uajK+
ziH6A7xx6w2oZZco6hp@!-&%PcEGDbmDz|+F@ShI;_j8DjYOLq4MUZwofPdzY`>cdX
z+`vKU0R^_W+Ni7orR~J3W>!-drbqRO+#>$nfC7min_~(IwwYhz=Q~vY0vURoK7F9|
z!YqJcp<9wc#~d77sa#h(x)&Ogb9?5wsv4`A!_o6I*juFk``f3luxgKZOCPb~l8_)&
zBC8##L=#EOP%;yhno}4b48|fX10OToY06~^dDW3h4QK7otoe5x2fmLDIsjqgqkPgx
z-cf|2&w<(Z#-y|*9ea-F9fwDemGAcZZ`s~+RET`Ox0H_kzumY7%=Ns8RkKBpxv-0*
z<(sPk{&6yv$jL=+GPT0(Vng6A>UZzF_4}(1Z6TnIk)73{i+}NuGa$WYZHC?u3zCNz
znPtu;gOrx7P=QBJ3jdx3C}RcSsZ!A4#PLGc@cj=<d}>m7p|qgh!W+W>El{t3FcOI<
z8Org?k;u?soZ;bNej4)p9kA75{GL7Yx<QdN{q%F5EINmOQHg_F^vT@kHeCb^Bgqo-
zk)~vIHOL4p2(ZC}_#Pl9E1O~LKE~rb0e6JKo%xPgoi)^IMt|DQY#)eN;XFau-vLTN
zDrBcv`hJ?r3!w(DPY?@VAaJJfCpJ9|^z~~~QpgNUOpfF9p(~e0z&jQVz#Bqe4h}p}
zP5^;0FfxJ-4yp_+yInLzMn=M?Z**iXyf`~@y0xgyoz#yQ8HtaHjrCpe1^oUGk@`E2
zqQVtcq<8A@x&VcB139gsT~XlO-A^<Hr1eQ87|uGJdxw4`4i|WR9J5S`QMN^%pB)=G
zKgb;P)jOMUAN~S7Y~&Os`Axzk5_$P^`0~J9{cBaFNt?6y<;gJ70aZN(@DM!O<?_hI
z!pR`f%ijJ|j_(G$-?uP|5A8+&HFft02WI{vS?O1CSY_$CJK*Y72y?qHMH0{7wLUOl
z(5QZN!p;h8Aw))U@y!y$%Y+EXaexPz^4NyEyJgO1cLG=M+EW=sY~2{miz3vzyu8@C
zgsY03bQuHB;fMsyosY$VMr1vDeecZ;3=A9rH&4D~i+FGGT<-A%gg%7U;dTklpqj!P
z<3tA1M*jp^iy}Z*AgsMj48#odX3g;?HRH;W)oa(IQ%JIbI-a#$+bBIfA|#5Nfsbk=
z00R?r25p=Wk6ZDDO`e;n6fi{yNk(YO7=`lFr@l{0dBsV>H}2{|HV=+l{EoN20++|O
zY{bvfMfdvduFp9V=#;Q6_e6YgYy8kF@O<jd`M&sN9R(bC$%K7I!R4O9x-jsNh>C%M
z!T#~tBt!?@&zie6I9P4}w;~8*I;)m-<66l0-D=OM4(%^HE7=v$2}lh#TgH*$w;O7(
zn#iKM0sz0L$`uNij0jQmjyi6eE9S4q8lP)l#Rg>wV=wDnrUjj(1>}3Zc!6-9d2a^n
zDE@4HT*olBuNrikgn2W`c)pyy0S$0lIUcYor^#8_tH&KUO!g+e>&$2Us)4mT_W05d
z&VluKf5oWmY#N@Pl#GQSU_)xF>SCixd3r2XRl=}a&nNxRrA*zI-XU%{1^ma{=fGkI
zim0TImi~LNgeD?ZHY%1~Q9&FlSJMFEi12lK1fXMj2QWQRbyi`8>GX)oONg7oN#Kyi
z*?e!jss%EDc*o<*Yn}tIUNitN)`{LY72AWDoPW9p$d6#fT+n2ur|;Sk-825MR;7hq
zL+>hnx^~xZIYo!iu=jdDNS<<NaL{K8xSL#yIZFKy-!MBZKQo2oN!9p&fzH!d06Lct
z4<F(_4QmfH6WJ<=BZa;5Qt$9;t(^yU+#-+s7=lt~93vx&n%y6n`L#ZF@EyVYo|oiV
zVR1fz6eoG~g#|Fxb2`uusPfsW>HXQ}5cH#y?x#W2#<bj_N#M!yU73K_AKs0UUZ)gT
zBED&Nhj4(Ea_pLPZ)EJd2u;ygkyx>ZL1Uv(N|VN5tuFK{BC+pT%uW>K<VTj!i!(_-
zEsY?k(v5?ri`drf+{d;}#k*jH^kuB^t@9?=*!H_`?hI*6M|eKkRygIa6xmWSH(10Y
z{Pwf+4jxO)8R%T<6MhR_<)p(qSW~O(<D9?C_9H<6&YpW0Z-H>8o+3EJXU)XfSf1#X
zr#Lm%X0ToDcxH_#-cKr&kjH7lI|XJ&V!~8`dFYU$I%xpL25?({EJU~5Idn;M8xlKR
zE2z<CsPS9=M0U2Hd$}cETvgo0{5)lqUt#>1efdU<&wlZ2YXFKu2m@^DWtJH>j*g%2
z8NQzv&j~!*(!12JV>u~tZ9U!KJUc?U1aM%>+;s3SvU!`4QQJ!wG!t#T{4FKIVM;P^
z4hY|w>$Wjvt`jB>qk_QZ*YVjjkEzJXj?bbm(-YY0NPRCcKKGok6jZ|@38AGQ*6h9G
zi_deX3MP8CRVP*ymZ7^QEVBySg@*1jg-8*4)C4X>vOadNafXY_{0Ib|dGT}q@V(=@
z7*c<=#1(i}H(<iE&#$7UOL??|Qq|_{{P^qHA*5=|wX_OhPdEH{|IKC7?i-K6vz@k`
zs%b&Yj^lYwy~!ZNGwjUT+`OaGsCmVzVMt_4>PAY5cGdxGMSHx_LA1$bg7>)5eRWGt
z{4^ox9~3vO2Ix56=7%q6Ebv+?hU%n;Iwg9!+HSWKJyEDEaq-2Mc1LW4Yf}QKA0msB
zsF|4YP9eDk@I}n5zF69i#2ZPV<nJ0uh7K%8`h*K}@$?!AwccL?rdxrh?^-I&s3T=W
z$tfsW;*&g*;~hW8P#x_9z<JU)-owIB5^kBh*M0#!E^r9$LVDNRiAI@E`o8SCzOyN+
zU@6*qwk^W57q}a3vW4H*BP6nGNPVF6euRmIS$ndcHu5RH)#3Q2qQ!0xm5Xi!;pG1J
zo#4t8(DYog`}^HA9|oOD%c8=fP$*;3E7O>5IY#YT4L5BSNj+s==fyUW4Tzug=Fi=@
zpsgLMe)Vi`-YBg_d1^>iS4-4Xq^l;v-buLg`*j?&+)1Km%y!5~{k2&<UwMwp6@j*0
zha1N!rw&;?#jdcYe1lXpbu~Q0f2qxy3owDc5Hs--n8$spkC@uvGvKPyZEQX3G@zc@
zmdXRRp*Amq_Dk~xEk;;da65>ONmzfe=s}RYw4(6qVvwQ?FN+;XaiWWyQ({<RKw@-a
ztbJmMel(-@VKi%h9q|c!4eKro_6blo(YK(9E`U1CTu$jE>>C)j#Ig{#1UI*^awY!8
z8v<*5>6ly6Q0cO%!ZuGRt$sLD4pSXL;Q3x!+swe7KuvOJ<+wk?m_`5-F_jF=c$yYQ
zi&ojx5Etz@eN~zQTq&LW_Labuz+C-}N6SU(i^*xsA)J8S_g*TWdBfjJuip54ki!_j
z*QRd0HaG47Jyopbz^m=5%uhFfeLdgyzPK1MJ}4`q_u_Mj?}>VI(Hrzr?hNe`)GVO=
zzPJ80eITElphE!kfq{%`gFaE!KPVxNK?(QNEnFNG{CX1jxA5p<if(^bXsxpW-|+)g
zfcS`LysdfXxF46$L-Q)|*9>W-8J84&JL7^ovEK2`L(%nF1tfdZg9v>EsUq@cyfzV~
z$-F6fTJ?}t!i2Pql8E6ECY(d?3qG_IUE{ReI`bXo?SocEd$M!($db}$qP()2eeC&<
z3h#{g)|1P;-AFTG2nbdeU(=6)?A8a$tjpe&5c&9YK|@_*jF%U!f>WnA?&}f5RDV=d
zkBe+$`fyY+LClb$Z&P`k+$vGVz$6TFY2kUASVWSJ%<Vb-NEBGj)6w7>*lz??8HFqq
z2_0sv`~-9*j`xnj*&Gi=-B;eII)P8^@5!j*sa0;0DsT(Yk!_4NdeViT$pZF1DqquG
z)*!Sf`wEEUIb8zRg-n8;awL{aGZkpHbAz0VaD0Fr;B(T9y7b#FGa~;9O;t>2KGyNN
z8WMo}vU*VSA-_I~Ti9zRZdzevuBwR?gijD{WtfU|znX@_Af?IOfLSvOdBRMb0y7`E
z=et{Y+wP{K3vtHM0`O?!3bV1TewsSB{FY*3`MR*^%Ze1u<YWyxzLtAtY~LU(>8$A-
z{p{a37$C(FT9GaE4t%jow(^hYBY#Y2sQRcuSX^0a%3Mjy?hu5(ercj$1zj089PN8L
zSTN0O+%5a^fxK<MRHPDqx3$WS!;a!D2K@4XaF#|cFx5q~lwbTz@(7`d#b#q=(%QrY
z(*x!SJ{yRSXG^8e=VhEgpdLFqJ3G55XO|t3<a->xmWI0ZZ&Z|&lpLQLH+l5uhKE8|
zCa3jZ6}mx8nAQ#f0TG01;9inwI@fN+bh5|^-NiS$pSt$?Qfnp02P4!Qbe2$KjX)y|
z2L7h|5>swO5~dO$dBnD*BENkn&6nM$P)kS(A7lzr(~Ft>NISpjFt3Rm1p_k{iAylZ
zTtj5DtK@O4r2N7&^M|Cf$;r`Ct@n$W%9t_=%a<}3C8Q1<)^3SC<vC$@<qY9)EvsQZ
zsF#a*H_c?nS4A7s#(VQB>>%e<;jR6f`>iYElm&@u?&U#a#SWxCl$Pf2js22&r)JCo
z^xJSDmi7=^)t0K#XOT$T@yCfdqaf=@9ei{T%QyeS4K4A8rS~goTvPjAwc26pyxtZ~
zxC(|wtluD{H|-N^qb6b=PHbivEXo(-u_W&4OL6K!U>qt^yj3oUq!l$>IHSiTZ*d{4
zOZ4x`5jKcg7MM8SW|#}d!p$JjWn<;x@@f^dY5aWgl0yY=RpuAF1a@g-_8@BdUKWdq
zC7(p82m->Auj(riH$ExfaKRUfRPG-oJ}>ut9iiT@i2o$gkStqW%uuFdqhMF@X5J>*
zVA;eqdtX|WkUUq?7|15?Yur`LMONQ3_Q=x1vbM^z!>S<kUD?T(pAS_@`0>BlK1Eum
zDVcLL={(d-XoY4lTEK>Siu;TP!Ef31^+pTyz3W)+a1kz)ZQ@)vqKlb|F@{_~_X@**
zcD`$4vWLh!qZxc%?9u3Lm24YGfUVdrL9O<%)Tb!xvlYI-Y?+>z8Q6ag1Qv)$Zaw1#
z)}HwQa`^$+y*jZ|r)Mf2yoy1l)9BzrU0afOF?Qgl8UqD>8CgXfjximFAHd<G0H%`G
zR3=T{!Z{am6)K|wbyoS?uj95ig#)%s;-1MXtEHamUI2$0DEtS<$mOXyz02JouNk#C
zJc`2KuKPmu*h!f9q%^;)`PkizSunIvpxzYiS5tVVpr9~Rz>B_{VgIB<$oFEfXQH)>
zJHk@IO;o=hX>AcYov+r9K3#9=aTnq;Z@kmIMjcg1K5+jjZ~o6}*mtqNG5$&gn8zYk
zhqoD?GvN%}JOAhfY<=G2i0CkvTNsr2HMU5Jf{G;il#xV46l%hRk<Z`K$W4v0Ivv-E
zzqm3jREq<P46FgJ?upQYvi8Bfz^SM50TH!7PdBSteoi*=hPO^p`^sNaM7}oHjM-CD
zd(R<uh~WltpOxI~Ri4-e?RHZXhf6~lowf3mz#Xo~Gks4OwKpUA(K?&z2W9H_Y*oS)
z`g>dphPzer`FLYyIZ`Co3f$kwRzy8Ha6}HK6(2KYNsZs(U%`7iuCI6f@5t5}fxUxi
zpK3}_VNNhXG`F^<ZUu=S6gV~T<#AZI`IUmZ7}yhIp$D_sQAih#;jv9a@&PHW{%4T`
z&#INC%WvyK;yMLlAt%5y5u$#V=f~&5OzcwjIOp7>EM}!uDV<t$W-hFb%ql?#1#~Bx
zSR{3c&<v@3b%#k<qG!~cR#&N8_G{7`qcDZ|i=V6F4Zy>xCu}Ic<88Fe$u(@Q{tD0N
z%%w_oCMNJ~hU&}X2Rg}0GF4A&X~?sY_S<WLygh2Yc6?(0R7|W8USKJ&_{2k(3Rhj;
zw64j@VKPSg_XcDRT~W0z!Eh2YW|q9hk1g|b<L-H4f;~z4Cv=(o==(7>FeKN1FLHRd
zyf`k4Q_hiE4|3v>&1lbT&wP&z4Pn#3S1#g<e@5tV0=zzq4!S%J3dne6&D|$0lUJUX
z!c@To#jT1mV23{Gvw_rOMJ-Iejp~c36zDU>UvQYNJPAKy--R4*Ft~E*L4=q+V0a4V
zMoB#^ETc0Y7NCa>4(1dpck)__JcH0j40?1K+eNm8PR*My{&T*8-k}i>@P;%lIp=_O
zd5VQe0TC>R^F0uXO{sr22jbt4_5L6u<6LJ8DU(<0*CtC9u28VAghIb`baWW&$b3{p
zye5~1^til<&<cet*%7}af$_Bd$dCK%#BvoS<EhBSq0axNh>z`)&P#P4<HPv8<hOyV
z+1?*YCw=R^jBZ&%Tc<K4ldAxk{uyZ0)TC4bEb6mX25zJUlyj-U3BKJa=(QLi^GuGp
z2qDL8y+~PKVjF9kZNXMCg*dLyrAy6V@C~wuLJ2b0TfVow8lSth|Cj8<6XGC}JhHAp
zf+bXm*`_(*hQ>-jG`<B^sLi4}%tc*0xJWPs#2X|pCfi-Hv<X!FS?wh|t3>@zuAGfj
zZtoN|ImjgX+&Ml{>{Te@Duv2+XB67-HfeNds>9S?*xw_gN_um%(UodlzWM?)scpMF
zYdek`7CYOL2-Lz?inNtMzvknR<j)w;8m+^ucx?CnXl^?b;W49PuA4gk{Al>c+H7?j
zaAfw%Ipy?=jA58{Q3S}b+sMy}%!qT21o>o5@jx+9AZO5SoBcC4Hl1kaT82j(yRUFW
z;B1nj>nK8U&iruv@|x{0qK2uT5&*%Q!2?F+h%-P&A*%V+OV44mrb-B4b^a{bbi;32
z@PD5rkdVVK?VEs9wKlPSV`52}njaMSda!<lX`e}-xkhs0ov{X?PNaU3%CjWgq=SHs
zvXY}I)`UCU=FEDWAE}l~#%QIZt=B#KVWrfQ(^1Cf^RI}2skRtWhnhngZ7<p1wbuKS
zOohFXipfYfxhIZbvB%sts_9Lo0y~jp(<`<g@t6D`SWc*nRO?+RTge^z^<@YkcIS#;
z^FZH9u_$;VRmp6}ziezxGJQ))WsFe)4*@UVPGau@hyV5m1;nn(+6sRKme|lJ+4tUu
z`P%}s0)s=3WBhDX{u&~S3ww+{i_1hdT;eNnsoo%PdP*S~hat^1tec>&xXteO{|k|j
z$ln0&0T|CHyik}?7*Tk`A=;6oDz?s?T?jel7*l7|l<8)INhWb=C0Dx5zb<C`Y+Lj2
z@bY^nQGf@e>s?p+7@0hiVj{b+R-?(Gr-vB6Or%tRhd45yQQJa+T&dpvJTV5ieLxpl
z!vtM<#$+v@MjV4Xg)lRUnwS)>H;yS_)5q5vWS5N+Caa?l9L^QrIZjw%5fi?v@FG%K
zhHNvdTbV>>lR`vhMiCJX(m<PlhX)ze$E!`UKT^4T_*PKm`4kyGqyW6>LwEUpC+&)W
z^9+J@5f(T*{mFpZO<;5G$9V3=k~?kq1kvaOtsUZ#15h`A5v@EC%);!q&s@a7Nhzwz
zYJ{xQc==OgkCR3tK2`rs4PCB#_fIjvcAFk2(Et!sYD&{h%wi~i_E41_jO2^F7pabq
zp!+$Gcb2cj?7)nD*veGZqZL|#Tk5uav7ytpm#9qCuhy^IuODfuU>v|})bVUjc}#h+
zRF(H3e}zf_^9Iuz=^^L4Qvl~r`V?cTSoA8`W{J5uPRbqfg_QZ?j`CZPZ7aY8cp{~5
zaz-E0Wg6d|_)!r`ET;H1cG|9$BDBe}c}TBA+vIXO=sb;LRG0^?x=Na#dDwJR%;+%M
zw>&;0w>Yy<<tN>}S}!-7B)n*Ak2|{<RjRf7p{MxRgKjMi2id$$v&=UoWm9e<@$=B+
z(sR-3F+jC`S;!O%Gku?r|H6nHt0ZAZME+?2uEE(&%dtobjVt#I`jd1a*0@ZKOpYl6
z+F*Dhgd}k^8FzG($;&e)pO1NuwC46+`|H3b9t!aaW3RJdiCJcbG7e(LGwxzVFPBQ0
zGA+|XOA9l5MyR9QlG`Vdq<6}v-g?b{O5rrFRBXL^0(o#4!5>p*Zk-sfh$2>*%nMW%
zKd(1|<oihF7|JYdfSnRENm`<m$zR!?h6d13<?vQE@A@=5WVTSWTvXRERWeU@OC*d7
zTj1-T&%PIUK%znBpH$)On6X)X(!n=)3U7k%9`>xpY(Bd%Rq%-n$PTjDxSm}#Rdv?9
zOFOCCWKyp>8-8jPG9*@bF%p2>O1l!{Ffa7C*q<u^D}uSJpDO7`)NqRQzi4whXqWA*
z#795+@!Shhyq82u9(^TR43lfX{^-);08iW2aseX`^S<Pez5;}PkcpZ3|D)?FpsHNf
zw&GR+Q5uw%kOr0RmM-Z=K)O>>LO?)j)7{Oc_ol(3rF&CKO9~QF@(;$j-h2GlVx7f0
z=P)zh%sX#B?=$qQAaO)v$rep`UnW+1acw*5Ty0mj2%kOuMfqZ(cqt_@%`|oQUi$mw
zY%PvRO7Z~Vv2~n~!Izr7w4d{9uH(FhHiTBiH4P(azOd(yl|3leZ(bXprL*LSIDQ#G
z$v|UGV}YMn^k8p*-@|+^^sQQ`FSRoKZnlP#g?z`+%jU6SmzR;r2W4=$WwPSg51+zi
zJ?9)}DEt-mrkHk^=_~e@K85JvPc2cZb~Nhnl_*P*am!Qn^%3vF8XO{_JsaA$F6XPI
zHH2wu+)m-;=7wq2$LzDL%8z)3(&~>33w6J>`5(lL={Cdq6372ZWRUp1I*Dxk7U$2)
zWj7qXN1huhr8L!PJJ~zmoPYKV#5LSReojkwQi>Fg!}i6DOUPdkNw(@V*v)Nt*-Zg#
z<q6S=M1}!rgGoX#OT)a8_!wg0z30U~-=d<>DvTV&e5?{%@?Pzw=1qSNuj!^3roD`}
z$+DPTY`u3K+(z!vm%6OCTG9BvOAME2yG1N?f8ZIB<ajU~4vh&8Pl`Wm)Ad5hv2Q^d
zYaiy*+2rvLOctN@&Ykw^7vD&Wq%4Ayr$sMX;Y6p$JMnb6&^0t@z1WfN>zeqbRP||u
zO>D~3N!Eje6Hrz4*f*hJ#cBZK5Z}LOgSapMq_Xby>Q9K`393~5T2^6Kj4d?ac<Jf1
zfgu@!Oqy2!DWOWn88nNDF~7G77c^pV7-!v8z^>~EQQd-nHE#Yn56Fl!`Hf|0@J_nl
zAq5T?Hj0&6g3tgyK?w~V6X)g-Ifhv<9n%<<*@uU}_x-o>18Pm`IzXuK_W7v^Q^Y}3
zHLZ)Ppmh}!NvE%t=$YcX*`6x@o&=N~fB%}BaH)#3+3a06h5Z@?Kn>wlwo4u~2ilLE
zs5q0z3o10Njhz3t^7JF7ZG?1c^`rs&Tti)<>Z%Oi|0dDKJ}~*Vl#!vmHa=d-{mQR7
z{L-m1N#3vYXB<krk99iamRR+${mOLv_kS(Jd0ps?&lMw*>4vImiLc|J45vACn^y0{
zlRL6>y5oPa@Gyt`_wD>lYyWwO#^6htdWjZi%BH4B!`^l({%?@lfDK4k*y5TAf(oJE
z=vo6K6Cw^E60ChLTqDO8Cfwh1euIRzW%u^>wzc|#lKL;*`0*tK^*J9UH%ox95N;ul
zM5C&nJTR|jS}8ddnnH5~8~=%5|Jgr3+<U&I0#9CJVEXA(SE$V(_Pc{XIdY<hyeE7z
zk&z{y=BlW*>EAa{`J$M59aQiy`7L!9+l)i;2`&;<q6m9KCjB3a@>{@GdO-D`EW*XV
zynh?mxB!Q~4IGq8v{=ifnYTv!jvH1%RbrfR8JYha7gmO^3G-V{K~@j2IiD#r{jI$)
zWyh0140Ew|DJ~!=4gJBRBAM$D^%5%px19lK(dFG?&RFK;VKfKUte;!{o|<AXx@XDg
zAfFf<fuA$>z#%L6%f1(*Kej-);i9<iEm(RBMUO_QWL#HuGntV|JygJsC8m1vG<J#Y
zXkvWeBBgN&GLIXz+PQK>Mx1e5;4TP3u<7v7^9^`^7rwAwO8P$MSa8TD1emecL1Be@
zCAQieB2Oy209voO;<hsUibG_DG|23qfU0~n>C6hKzxux3L6aKVJzM`qa_48M_gWaJ
z?^g7s%!p>vT+S1iU}i$hR4FbH%Xr%7-?zD%ZAm=%g#@fexC3={;T@FpXQ5;d{j&Eq
z5fRbuT%%OL0AoQr?~mI>aTDyd1^3mlOMZA*bfYo|b*);*W*kors>vrQ_R(;eK;a+N
zO}`kbOm<YzRE#<mBle*aj2;Y@wRS>4kHw1-|LLc1vW*sX3AP=XcvF;u%nzT|5|@y`
zNf|Doq50}(SVAjHAznDTzU73qrCjv>ybak^_G<8A85^`%f;aE;^2(?(b6x*=5k8`V
zg-QCh1Iz{&O1E<3W4-E42=D@v{N<NurTqr%xUQV|U4lFYt!d)QmdBo+o>SdRRil>h
zk+Ot9-~a_k3M<#N84UusCKf!v?R4YkZR8aPw(o8f#uY#Pdlg%up*D4rFtYPt9|O$~
zc_K=)+CL@HIav5*h1y`{<D$ht8@Tp9ro{Bn#V0za_(EW0$1T*#CBZHC6m4o1{f~Zr
zx&d%HPL`-AuR|iFgag~u8LSX$Qy~JPV`#|wKO*Q`IH=?g)X8+Dwe!RQ%oXdt7+7iu
zDnzSfTUrV!zPgI>Gnr<7QFg?;!76HKk7~^)-KeftE93vxD5oBqxIa=^)7st^*A;;0
zbEAB7VT&)x{sFN3T#OK^$%!%X1`4o8po%kUJBHBnvN?_HXDD{P$W=)6x%{IM{-Y&@
zGwH#NQDJIA^?O61<T_YW!cy#7pdQ-%DVvpv^U2W#^6$@E4l~#4`Hcnmag~3roI4d%
z^6Ve&`3leBXqMY<5D-*<3v2^LEKa#6l`YM*6SsdWx<A%XAPLfC9vew_Pzq!+)yqk<
zDm2GTb<rUb<3G0jA6@iw3NRb5tZyh@+ESM}*6JM;9Tvq+4{s`vY6_W73~!t_n#5>b
z`J5DBn)SyNzkl7sxco}Cp9KF#ps*QQsURIwyvfMq+gN}ziHF?z3##%*OeQ|oegQVz
z!}JSPvuQ?H@Fjt73aS<ZU}j)G&Zn<|%>3^O{#mSFEH*_I=hN$N9VrHGqD)OfZTOUR
z+@8-P?r3OQzek&Tbt3i$aNv(ET;&G`v6A$V67pC+a~ZO!?m8tiDYN0S%xpGXFU_c=
z$;8R_7i2k5Ub*1TNQVrPqX;NurqWMEgq6bTg&d~DeLg&*X96Xsvmf&D$C+RJ`Z^lO
zZsd#bFbYsI;H>6j(Xcr!)ZTQyC-hVdk6?X9%xvjj-v$6Gkjv<=_TN9KltOh566R5H
z$!nV?2S$xEnb;NNS(p;K+L~YdtVCP{xc@b$B<NBRKw)R`6E4S@=-bd_&MAfO+z69j
z$+za_17e~t;5_yVwD!lg_zL#}X!u%TYAu7!mZp27S8f2~D?HeJ-3mEQr`+|Ev+Qn#
zwr(93NlUnCy=Y%(jzrIL$Pj<HlSUbm8Wz<x<qR;UJp@qQNB;&rUf4%JDlj~+`O+rT
z`+U|7;q3GY6}5M{EUo%RKfjR#t&PQXG$wC8yIP@!r=ySC9u;(K%zsU=*?qhBCZfqP
z8-gdR)AK&EaMb9Y(*oyOwq-R-+3#%3pC18cvabTi6k!L~g~Lxu5PLI&CYu9`6{8G<
z3J19!&-P%W%sQ@={NfWbw!=e;6bq0|_Q1BG=Srnmf03~k&@N=-63Y^$RK=h$EP2e9
z^vrwVl_WW;bFy%A*-P!`Djl}Q^~ylvV*km#|8fTYb0j37w*rCIx>Jqo`l+3Yb6&8D
zyx^u{__tIzJ8EwHYiysS`wNbvz(W1SJpgC>BPNySCX@`eKJHGHPtUO2WnmS`JVd2N
zE!<CGGz|laWE!7fK}UQ$jgv9xU(;V8xlBA=K9x>cALeY5QhXAmG$<b{)t6QGDxUm6
zsxut$=B(f_vbEGyb_vc2$tMHxNeX&yuTm3$SNykR{SZ{Fm`hKb9*EF<MpRbVel?hJ
zy9)1)?={`n?TOU(p|dM-)3Dx|l9<w(s@X+&vL?W@La>Q5h-<4A^(%Dmj=LuBs(0zN
z8kW>1l-AT6$31SV3Tz5jh-iqtNyW*@^(2O{0JArPk-l*=)Q8#2b8c)Krri~;!#F<e
z@lf4fxh;p~?A?b^o)r7f6b~m41P!dTk9@k~XN)~<dI+slOm^#ln1Fr^A+|wR#c(@y
z1fy+}gpRA{wAR!C2HQM4r}ewCsHr`wY9F~JG}|#wf-#a4_vd=EJ577@YW>DLDJO{K
zgNEwL%|n;ZZIaC6Mz2g44)(0jzK)tQTLJYz+<j|*523~X7cGkg`k*_JxZDrW-tm#o
z)(o{r;0BsHMP^i=7|g<HXfbLvYSRWvHG3th*5uKUZ<XG}+c6^-gqpW<O`JsTO@8Z5
zF>vqFcFk5FUypWp$ZhaJr%wRfGFop|<WfTFOD4{QB%G$N8FQ67+jBK(qk0y)C3S_`
zXacok)+{S+yJxIh6Vo|nn?qkUQk4V~YgNz2x5|-+&%ckw+p+Mv>+YTmtrHm!E`6EN
z>%Yp#=Ri~d;}-Fpg30b`pwzF8X_wRs@G3y7-D?XPGo}&Mih30n7xSIMM^1tdP&&*F
zCo!HBfD4&Bc`!rN99yQ{7P1@GulIl$RqKR;iARL&g@cBwTw9xC(m}RhU((RO8{jVV
z6onwry<X~-p6ZMWsJtRtvn)qeg6Gf4r<Ug%VgmVYl7BAOX+zWKdN)K%L&(Ex6xrgs
zDa^huK&#PuoS|s*E_^T%M_9QkbO@wG3;EXh6jp7)7S`LgRnE6?SjoEy9?!OTo-j^1
zdLM+-B&{^*_rjWUgJa647!@%1-nfs#q^-gbF0_`do?m5klpvKc#e?||T*vk&4i=;|
z8RrptF>+@IuyU`(3Thkk&|;Ts6Rr8d2P+*e&kejc9f)0{y9Sph`L>J<=$~$sa0%{X
z_r|3K8;>zuiPd;x$KhAx(p&%<K%;fP8W7d+iOv4q09`bz*x>qvYR27uh2C;+=2UjF
z>13ep@zt80Cpgj2;&&{fE?Y<*;<ThM%7vj6dZ(_lD`BHvbFGJIdXCr`ZBouKN{i+~
z7wnvtdiQ&M3X?~~gf=FYAHsF2NBMeOLET@!RK{f1%~7KWqpa!p+iVrb`SBUY$yf+-
z6S$3fJ&iAC*~a@CJG6j;Pn%|%>V`dSJe(wD^@_6;$<|!o!O@x@;pnB3E>Xo4DP^F1
zcCe<E;usAmzA7qfwz8eJk;kXWY_B($3fmjM?i{8|z!!7N!U)n`uFm$=k6=*^=QzS`
zQ`K4}S)29V@1{lCO^)Q<VdPh~PU;i()Ng?=Qbpmv;di&~4$6Nt!wa&#bAILzm(Y+E
zvfi0*jgBzph;R&Zeu8@cNByj#UVbHq`zn;^Y-3JnujuM6tLn(~N?BH2o1_-uaap21
zLD%-OmfgC3#uO)a(>RuosVCg$wCB4mJ%<g9PH|sdoGjigg)?t0r8O@>$|4KZoMz>V
z(8|s-*7c2o9H^N)`s9!W=f^FPhRk+xY4*)F2?p-gb|h$nQ{*$dIO_7_G9&sYLW=Hp
zN>1x#c<#IVARp=4PO#2>v*t1ymuP44VY6dXNVg>8fSG+-56*Udk2U}nSNV0E^*FUw
zruV*zBadCbVm;k#xlNb!gsaWx#;dGq;3VZ2=;Geh^)RROtYfTkx<Bb&Do(3m{l0FT
z;i*2@S*uqZ65<r2;1dZa(a(16tJL5P9-MS&wUf5oOmK~Qt!n%Bk<+^^b=iCFm=W)P
z_89-mDUe1Zv89TMn7Dq1uiV(OKq+cmleP0CgEv!;&+W|rr1LwjzTh$4^mW+B-3EQH
z?^Uo_oyl+|7wp{7-2KNR^$)v-`7*p(6Uv0$b!aR*^j?^wCdR4`6D^bRtvfEH!#y)*
zcPKSKU+><+&UTh%VSfE~{Lr<-kuZ|`jeA?yg1N^PSa0MK^;Y{({*2e5mC)GoZV)R)
zzKb-!9kX2n-Hcl6jgU!tjlK2x>d3l!vSmbXgOr{`2g5F6V@QauxK?ZUn~X}uJ_XXv
z*>qZZBTRo!5U-epwH&v!TgdYu<g4dre7vMm0c~O}&-VT28y$*U1T<m;oP7Sx<Bsvw
zMW-)rK0PStH}}N0`yJ~4_YOI~0)BNJoiT($-)nidjOo=&<o(TWlU>cL8qJ96%I4ba
z-nN^J_2yFp%er{mw&ww1>dHVt>h476Scdl9jDEvFvD|^2=*=o=sgCTc=Ljb!>1@VX
z+-_nFUA~cAUpRd;p-*^0l@UK<n{zQUZX?YB2fj*aNu07+MO}p+ZNqysC1=T)=eMl%
zi52c3(cKrp5$+Y&lvc-W6R2pnO~CF%IYD^SX|Nhb`4nH7&!=75Q5?3yryi%&KcPN`
zU({1a=@y;O50?pu%Bz?VWZ?(F+rPch8Bm9`5%rkPW2pN`DS^M_u1V}le<ijYuBRfX
zDUr!(CB{O;f!oE%aF-}~x1Rwz$~7*DT?%VgR^9jzr<FipXjI#n!n$Hol)Q_S2|Zpc
zFQb26axiAe)U5AZrNQG$^B8_jg1AIDDcf7^0~5Q%)~!Z%RL`sy&+QGLtd|qnQ_si3
z>kV#hwUj$w8}v7lo$Ec3ntp5lbyXq62u5F%=0fW+)u}DLsD5ItxhBjHGbv^%dT|oN
zlPLM^BjXjn7xI5QJiig(SaR}&zSSIjzW5xF`kxOhQ?EIgWKx^53z(}W7QNOmf85-C
zbE>NT7PP{H9g!RZdzx_VF1tk1tw!hTo5}4V`5a2Br6#ak>rUp9^{*`D<wHC*b|;6o
zH+5VGD@Ub>(DFDmTk7i+4JA)xNf=#3%|nmX{5Z8SP;Tka>o}0+Sz#hzYY3ZBgiV#8
zg7c;bf78EvCxMr$0D?CJnvq0<Y3jhp!h4+<)Q<aR@FoLMvXz|EP{Et>Cq10e4TH_5
zy1BNhr!rA8$7{H|`yTH!>z9?^IZRkg7kZ6y$Yfx-m9NP%myYG;Jv|H!tT(nSvv4af
zub)_1>Q3&_S%0i-g6ugRW81y;7YP9Ih%(5yM9P%i7{?DfBv;&9IG4`Xh`%cFIw*g2
z1k^&SD=pOsrHQ58J}zxud63*A@M<;=a_h({CERJkiil0ydGj<})^(@8hXpRquT?nR
zbTt1Yx;Yq@9XY3gYJ-x{GH(K-w3gJ}LJD+Ebv&E+@8r|3TPKAJOklD?&J*VDeWGMa
z%G&sGoig>oqN$Q8#Fu@6Imx!RK0$|BS?xRvy1MDgEa~_0V`KymtvCxI8xgzHg68#J
zW3kWG#~)GImMqQaB6{0je%015A|_iL+drETRQ+gejc{)<+^nuza$FI7i8rPl5fen&
zN90+rNKNWkY7@KuRHvvEDQ-IwI=So~V0=@Y_HVB4CBOF|mI#S<*+wz=D5o498sQwd
zq$4XRWL3R~v2QsO-;!2aKKtzhtCB=NHGiBN*E;1B$@~s4GEG)Ky~>IUYyA0$QN9C;
zI-@;V)th%yDE=`{VU(s1-S%!z!@O~{zokx(4#LN-7oUr7;e?Y@Vx_oFi{mjjGg8t>
zeXz1cI7w&f_-#W`g*KyBBfQDN;B=f8A=tVWTz|aA!FRqV!3wqL1%W_`OTu~ritMOb
zN=d)w;91s1iRypCBCzw$k-~(~DpxzUGfou;ug-jy4;waC-aa!e?(04IQ>6`}dWVAb
zHVk!nl5GO4x3;@ECDL!n*O`y9;eOY~-_!o?)3xR8Q}9`nSRv8pe74pWLqq8U2ihna
ztv824Syb((u0g#wYqP_;j0&f=ydlVmn*1s9>6*q2uemDJ#P*oPo67G~!iZhBho@LV
zzkPg0WB_$P`(FHf?ogR*CihcL_>Il4E72a-m97n1E+{Qk7NZ^<>d$%qpKl2Q?2q!?
z@mxKK2dxBr;|04VGMui*3*Vqop+0)`&5IgZPSE|v+E#>r%KFB_iu^+jN?Q;lv>eT`
zruypboz+APCOu#^7h`N#Rg=h+ab}RwII?T?T(M#`Dq!i|keIP$DVypQ^0r}ASU!HK
zUmK)Mt71kj-q#%#As2~!Ur(=d3LJ$IJ^Sa3n{RW<^}c>Bwvex#Yqh8qYN}^Vdpqm@
zme96ovc$}KC&gXo)U{&`o}AEUble%HahUy#Ba(aC5l73og#N1k*nHTlL2aBbm2fo2
z@vTSWWfRUeZ`*ZwA%DF(7o~X$s}y^E1I_CG^SO@7rIJu1J?|w4Rzbd*KAtxo=V6VU
z=SzA*HI!Pd&K8%b&4OtqnRGglV?+6IoNfoZbu!EPK1Y?_>l#}~HkRPc`SGtH8h;}_
z*Z`Z+tHk`iqeq;6d#6fmSp&OjsZ%r5qdC*xAO`c0Q;^HntURq+jaD)4u?TieXu95a
z+c&#ZS6O|;jsrq_yw}4WQ#f5sr<bRn$uv@6mlS^7o;~Vb*84uQGl|CfbibkRshPZ}
zn_E-$IIBnWw1xDnX_3H@`phKql^DT8%lDJ-E!}4j=u;raku{)6Vz~EL8%7GU8#XA)
zgK5xg2OamuMJhSPY-v@&vmUeNFRAC2wa+vsm;6SJ&5UcMG@}xrxl}Aw`$ZW->yDj$
zTLux7CVI0-tLHm4-bZs86Hwfe2HOsNB1TxFs1fGll%PuIor&oe0>?SDz{=Nq7hS{a
z@>oW?XN(_=WwbuPrz)xM80*e|s_u6Ev2*@+yZPzdFbK;8vRSRenF}ULRUpS{MVQgN
z@t8(+bGflGx*Ga=-A%*$g;V=kQT^GvE_?psMjl}qGLD=d!Rk@xpqsvp^M!W)y8)DM
zBrd!<uRAn+r5z+`AToc~Ml2U8lum{Tm(<k{sWAzIO+^f=_=^SYHe86ZuohOHsfnO=
z$&IA*ZHdNp$NJf$kMXcsy#u`iY3+nthHLXRdhYvZgf?rBuW(;RPOwL!N6dWNzahxF
zp_gDQ9N5rge(PSpv_PIB_u-;M0_P`acq4!9<msErSpQGmjFCsH`b)KX8c0-FS9qR#
zP={pN{dADL(S4YwKB4Q3-MAI`-g>NgZbG$bBbt!h#odKgZ|#K+VsVJ_8JfO(;cYc>
z>ks&PeB15(C%Lj$KB1>^ESJnYZRKk*dv0IcUUYrFez88+4T{j!s;&+sHL<}l-wXVk
z6VpKv<>1IH{d(p>KJV!eK-$a7HXdnlRq41^*AYv5e2eQ?zY960dk&kY<cb`qQmBxr
zy5>;mu&4a}%g*yW8gqJU(zBuXFl%pnbQwv|%xtOEtyMy&zBDqUHFsd7`n+A29aa|O
z98s3Dq&}_7@KBvu!>h*1y?RlF&Kyll!I$9Cjd-`fRy>g@#~Gq&Jte2Xilxm^1I3q*
zSR!umeBxUyPv}$^x6Z3cxh1k-0n@4-GleZ5k9C)|QU>e2x0}T08Q$-HQUZ4$Qy`NJ
zeCC`gzLCP5nPD(|nf41D+8B9L4-qD2#+>o#koU~Xs)Fk;qA&zFP*NhCq!8$|GuaW%
zFeL$N6STR)GPj525488e>+3$o&Rfdu=Bc#FP!VV)t}$od1Eqf4ZtcpXZp8`QcaaUw
zsCY!gg`R=EfAOg786c<$eE1t3sWmb4k1cNFi0dG<>j{fUZ`1OAucUS^3B2`fz20L?
zQfd6*PIjjT<2pilre<)X;wf8iPS$l2`l;Vwf&a2Xq*T7%pKr@t^B3+$<4zLPW!0`E
zJ9lH3sm>BL=G*^v_3*KR-_qZdz26890Gl0{Yu<+NDjL=;@1l)|hFsGxE=9V!bgW3`
z|Lt!;8I%O{gLYLuH9^gd@ExLfHKK=I54gmZJSO8ytklcuv)%tYt^a$iez;U1efppV
z1H;c5=XTLe+K4i}Ql=R6yJ;O=#s7Axu~w--)<=+64uRsw4T-H*ueZW~N?cpinMgM3
z(e&2Dp(F2~+T(wj<kw;-J2p2*Kw`Az?)`bkaf0<g@e?h;vJ@g}9FjPA=Who5H+vD*
zDlQ5X{*J}XW(*YF;`OpBVQZX$?ei%AsdN50my{GmOOZ79{TWa-+r#P9Y|nRG;DnNf
zHg^_}T30q0<!ST$ck$s5*Wowi$K3%9Nd7J^(9O~{1#LqQi?DF8bJ+goVgC6Sp_AG2
z4XfH`q_sBVZq|8Cws9UDTQqxr&x^bSzzJepyqSy6y?jTO22Q?IbE_L_8s??yGR`;C
z{!*?6>%0YM83a<X1t<n;A)srVGo7?L^#r#Traop<jd%+85cvP61DA;S3XUGdKsn@L
z(kuDKX*gB=x4`x__1d=Xs<?GLqV@3;|DQeN|E-7Yd4X40@Ged3ZWIwp3=}?z-9xQ8
z=;hvqu;Npg`u}Pw5}@Wp4}EpTkBla&f6>c%Q;E~Z*``fmltn`woS;nPR`S!|D7`;}
zQ@=gXd7Hm2r-=<|P%z}>`ivePPMz$QC7M+!{`z-C>yHz=*c6bq7Jy*ci5X&{zFG{j
z;pZ|vk`p{|9WP1AmxSwwdEX@c`)+CjSSP~t9Us<hI+Wq7OJ$@AH7np%(!lor@374Q
zW_K$2=C-h5CaRoihs}pP(ar`?6LEC!MHDlzKPsxM`|u0a@TWwAHELtOw0o)@{u;1!
z2^(N_npSbDB-VVgesJJZtPZD1Xhb~D8PK#g|Fb6cpO0ileXpoAX>>^+q`QxUUz_xS
zX~QSfnSE+D&UUK+mj~ScSNX?p3#V{=#*CPZzk!df`a)GalcI>}(tP&%>aQCHYUF^3
zAd*}|EVCRH0}wR<b{Cl1I82qX139JO-2sY?EOVPtOAqTc#|M5gvqeZ|0{8%28}xZC
zlE=xk{bV%&EC?pmT8wz;>P!S@R0%!q4zj-m{i8XM7|DfpH7o#2uRfC2aU86o38S^T
zDjL{sdP9ZNzjSlz#6KJ??=wXvowhV`bJOO0VEl)MDv%a#r9)0m?nHVC+JO@B_q!JV
zj7!0D*>UWiSvx+~suhayq%v`T_`Oi59;|fA)TPJLCA;DHc-hbi1bsy~QH5%d&Ew++
zz_LSOskqOc2#VO|kA&xP_Rgd+z2$$e6pUzM<7ldT*{|hZW+Y60r9d~UGHc?*`3|m4
z%LaP*=uzbE!~jUdfn@GoKME1N=0_KQ8l{eM+%_~e7H?x~OKCmqAyxX5U*|E9kCe5A
zCk|NLq$ny!4haFS6S_q&CvIpKmrl6C7Ha^5%&)qq-!n+;mhkQsW#UM}W}p^_X|aaj
zQHm#f{r>nr@_>LEAJFQIxmRkVItL01#<&f_aRJ)keR{bn>r@pkVwCp?*I4W)mO2q4
zdbcuh=JQ2m{No53-COlmLrT)OMdVj+{E#9)GN9!*E3X*WW33YSC8f#+MoAOsdA3%W
zKoLLf!YT=<?Z)>!bsTZMMNb7zMvsz0xx|1ITW^ZyqudoRF5n=!;36aJ=)6ZU;EvL*
z#TQL1t7>kY(_pc_C6ZOymvNm6nAZSV;i~w~nl}Q^U2pjTB60)HyH=iOcv?a~l&5nH
z$O)v>B;7y}5Ell*fX=|KE?2X5vLwN%V$Ouk#(SX>%q4}Z)G;r@0RR&DIl#`M=1=XL
zN^rq@5N)Q9x_9Za#8`x>B-8r$_ka~jZK~J?4Dk!F$o;~1SCEqTmf!m+aN#V`R!<UI
zBL9F+P_+1kaFJc?yC7v$4^U4wP*G!s`v@id>;X;knd{czx^U~ENJR`U^>6%nl~?Zp
zf#I@XBZPQFy>_-8dQ{bg9?q<;UaAW2FD4n6>|ZKUAMrpALJ?a~MGUY+VAQ9=W9rcu
zO?E32O4w0*MmY%ebpAgT^*R%1s9VRRufN~L3V3a;AJ`sbwC6Rk<m3K^0gp)b{j0t|
z*Y-<cU|di`7-k4H1`9Kx7#}GU_e3N+PPE%t#oe4Lw0-=~9{oqZ{21i^1*ON8=Ql20
zLJ7tNR?}3?oJobd>1zNPM@dq)m74u8+{ORSXwU8UO_~mqJd6zNG&yc!w_=^BG1UcN
zft0yXRSUkU-PP{MQ{c(wEf);?;k>+dKm{%HQ?s*z&M#9>4ypNXZBA$10wi((bMdrV
zOLx=gA6@{UUgEoQ&*!S^_~8X!g4;=dnwwvKp&>q&lz?9o<9*JUjGFzmpv@84$}yed
z1e?mH4eD*BY&`VJYX!JJ#BcAWA85iq0nlA}Xr)>VbsQ5Z5m8lT<%5*e)IChJju=av
zPPgq@_{rh6XVFn7Rpi9<^mItyWJvSI#>T{TaxX);_K9eQayl+9E~kJ%+HtdsKc?|N
zY5PC>1Rfc-;}bP%=+~rKPQps}F-HcU&XqGE>*q;88Q-kS@xr&t4fKU<?(SAKA1$X;
z##3AH@$#;<ecoRm-yF)5r6st-AbkS*HQb++xMz4DXnAhJ$4^f!Yyms}0qChs$@X8U
ziH^o@e72J<0A9LtJ=%TBn*R&b{NuhBi20u5@R4qPK=(_hz7Hiv07Qe$7^3uX`FmtQ
z=&K8@pUA8^_D}v~9?Sd{$XaAW8;OaD4*-F)XV*yp`RVI2-TL_)TfP~*tOmfj6nwMj
z>K;<M8*qc-XW)o_y@}cQ+)eMj73~mS&3sX_MI=CJ$1yWAPcoJ_8AS>lB;tp)DJ+nD
z+W`~<u*Z{U+q4qF>(>8SXXXW)U^FV|qVq#QmeGe1s#+n;RvZ(}L;muOc<7TDRS0I?
z*`JWQ^9QGslUjUkThoF>-NcN_4Gj&JILXl3P_HupeqL=`##Q7WbnsA7Q4#dlpwxW3
z_crhbErTW858&}X*v>R8<rv{}oDRvdCQ%wh4x5^qdM#SlNE5}&(DbW7ImNlpKVd%q
zVr@5F=a5QYkMQ$N1TC_S>9t!spw(hBjI{;zKY25IUywq-gmVErLt-a(hBx{@p(?YV
zh!~yD;0FEHrYWN~DbD02xgfMR`VrdSuY*um;Y}XBBmZts4(XDW7?}kP)ZmRm6;}2D
zL)8|YklqEE-bZWDb{xylk8u^FPD>vIx)2o=ABS1YTay{ZjmXRtB*<Dp+wVzG6u+wY
zCLl-P?+1@$w|~_&FsQwta&X;W8^t9Y?wC|z)n^=&QO@$zJ@WfwtM35i#m3MC%tQ@!
zl+$2k^^t{+OVLeN%N_J^HJIhsB^qE-p^XR{t=XS;!xe%zS?q7Lu(0rbVF4NnErdqk
z7y_)Vq4g*=rI^b1Np5*3nSr%IJNdUt<ZT)+3MlRHck1ivRs&(Vb?>uh*G>4k?>IYt
zSBT6zd*7VEE5NnzLj?dEI?xLA&6y`+0|Ssr7$WhBi0S)viqu`7w$I}e=@c?e<{k48
zcP%Pd>tX!FF`X$2LzNwO1|`Y&tJLKO3uMBvY;Y{P&x!@1kuo`%=4yK7UZnut>zuRO
zkHE2&!Oe8w;^Oiq&y$F{f9otRUSz?FtodFM<K^O)YI+5&9Oc@HDpq!{`qT{2R<Hy}
za`VSvsPkA4N+0y43rr277;4uZ>@CJ9CO@Vql=fxh5dEfYH7q!);qDB-wyEb`BMH-B
z0!+rI99f_DUuRF=dD)zZ;0<{69W){dzAADWoV3z2%5fN>U7QN<d#*X>eZx(0#Ngz{
z0;8kOVMH!M)BM~+QPkjTNLC+6=~eiIWLDM?Y)Z--SdmDum0T?gFA^defA91=9KHtV
zU|jF_CFzM~=0JjUD=UhoUlcvl<T1^9kSDKMR*H-|@e2|)CKb2^{!f5<w1@1nz#^y(
z%%Q1Jvn=T-8R*5!zJj8vj`G<|o!VWGyEv9I0TBA@N<SBKVczOYq~LL9%lUUt){*oJ
zU9(lc0;mdJOkL=_!>yUG<wX-o?Qt(u8m>wARg@_)$w9TSOP~=V$HlGa`4-^~_wZC=
zn7r0uwFi*Q?9k_+r`%a}EL3ZuW0#{9Qj`2*5fGwn@|ZW{z{>$CfIdU$m&)&E>F%Z(
zrK5LVhV=<skS3@3_1#bP6Mt%~%oYyABQz?iZ`1#3CEy`{+9tOxZEnTC;|-emo3Svz
zu9GPlqy-)nx9}Ep=r-I;3rr8LMm>z$B40|3yFKRFdem!oN~Zc!enDC=EPShtq5^*s
zwzJ1yQ{f|kdFXEfOKbu??JYVVk9!QhJ8tu~{I-OK_P|hIKirt)fka3E_k=*#GN7(X
zJ*H@2BIzwgV0(l_df1OhLb!?%RemsP1HDQf`enH{<p4A=xOWj{tBSQom8jt;h$;+i
z1O0h8U-n>L%{UN6w?W^Zg;)RODTZ<(jyuk?<k|=0rR`8J)#A%A5l1%h4?(fFls}ao
zjVOK%&VP`7GV7Uh#Q<p2ak${7@+&xNW!#I{Q}vkCS6dpoZ|}UcYh;U33QreNo#5O}
z$UWO?+&-NM_j;%Gvb=>r2j40N&t_a>S6#U+JX`ZHc(F9wXp}vEnJ}1+F`QY@!<-yC
z-R;B_Qj%Zz+5Y)ALNazeLVd#V_M;r{qn|`gz$<zKN@t{{UrlSVR&V<~NO<nbzaqG%
zHCJireHaXXcg>*0cA_DyJ*}@XGO9u7S7zb0GOAR&^ll(QHVw_7YUcU{8wsJAB?Cr&
z<jOkL+OalN4#mdDXI7Hek_3Nb$P))av|h-J_jyS2Q-3PYLXO!ewY!)}kxBad3`!Xq
z9x>HC{e)E25mT%eca2o6RAPKFaK02;S-YksLi9ZhtRBnwwcq>9@CnESx#tu5H@6*8
zgZ;Gz*#`7SMp#q*#>&0dFr3wSub`=g!+b(t-JEZB@X33jG<-X>Vs!d$o^P%zRzWCA
zyi~lsnt{Y7gMfr)>$18*nk;wNyImYvC94_UE4{EC_F!t*K~cF&(00?;2q8}iYb)|d
z{@s`7G``DYXV~K9Wg6nO$;Zm}r8_n5zo?3C>w*-?8v#$?{~BXQ0eY-WM;8u6DNL(Z
za$KPS51r>Q0PPp{De8sqiv)=7%L<HfnO*62nN#VPSEcEc-FD_BJU{0pG*zewp>c(1
zqcwwey0@}?65N1!BtgIG^TLj#4UQ|$%VX{)GVPdT@%am~im&}aN7QySXVgYCbToD}
zyWbatzQ2<d@`gq3x9)Z`>SGjJ2!#)}O6pnFvE0WU9`Lue<wW!eV>+mjjYk!5<>sNT
zP+cy`=P1d@mUp>^EI0AwzT!ZL8=~V4+tE(*OMF@W>~j`R0Es#;dYWfcZL*Fx%|{mk
zXjWGF`x5Hr^mRKljpjLp5v$%cr-k;LLg{0LU&nYXZSyJzo~_XyJ-={Ip2RhXyxP})
z*)+;;v=ftYYJ%y2wcDHc^`F##07x6~X|D9zUQZM*E0e2~j3lrOvSko@k~8^aswmEg
zkVvPrwC+0}NX%GpQkU;vzI-NJOfI*7h3zH^U_|bG>x!~ex8US`jJoO>Oi3<=_f+%A
z%h=e(n5Z~B)f>?<I4~SDXatd1w%kL-D!F+kcIgKp%FX4SGTyoR#*8%nCI*bkw}WHL
zE?9KVsKiItc%=o=G&g0fMvg`MRA7=gJYE?34>`r-*Q~;ENg4aV^SV+5T)QKOmN03H
zE3qvi>&nCDv((W#_>DszZ7REJ(OKbF;nad+FDei2K7r}BhTgv^?-{qDHr%A0(kDEx
z8mVqZam&xHCi0sl$tK%|n7@vZ+UI-JHexCUFlBGC!3TOdM^oPAgU?u-zW8Sdy&Sb?
zrAMkCDst=Oy-|WxZ3%f}4vRI73N50HSq(Pzc-~mN@sG(!2PWgA-Q`OdD3?Q!A;`j;
z<3Xpc`|qSmB5&XO6b%|N?g=fa5XY`Hj+S9Iwrm{tYQ%3i+7>SCeygZ=KVL+9QOI@n
zF|1U;eKz2Jz~Uo5zlW$8*K8u8+E6zf%dF2QAtuR-$z-2S!nQCaG-WzJ3r17KV5tpz
zhOWK}Mcz;UesiQ_R}Di3D&J~2bO?)5JSaR&rDUN&&2hZ{f&Oyw2nvpeCA1f*!j(+z
z%uZhREu?ykqR+43fLT9O{MK%rgEvkL4%}#JqJ~_0Aj(X`xX?g#B*@D7no+iq;a!@*
zF&xLYoP1ZpX|#}ZPNN#w_4C9gFp~<$LC*BR<+zt)f}#x}SO5*5vFg-&0yq&iLy)VU
z&@yB&IyBHzExN69!d2Y;cO$HyI`BqdonLq_9z+CBfyGG^7jMg8N~}3Etjt*ZAh!T<
zuquq(Xor%1yHORD5Zw&+>{XMmw<EtdJe<OE|D8}aqC6tLS2(i-vUf%$HR+1W_)xh@
zO*sYX1C55lMSR!1C8iL^hYAlhl;cCkM&L4$neO|C$|^UdD1)+}gj<b!WLphNp3}%R
z4pLL)@K~hu(vwBj$6jVY<aaKp%b&f6YvDGwvqj+vZD3HVE~NS$2gYKBe_j7jgBTE1
zwK5hG7}pY~@`rD|>l3xsMzap5?}F`Jf0Ft|Ho)}je{L9vu0a$Zacf&7vja_PB%<*p
z@{w{$XN<ZKH_gm$MOajtBaeBl)s%XO7KrT;PfJEQQhNH7%$exF#2#Q~t7FR_i@=i|
zM&DJZsCPNjS~BHnYK~!}$*FggDy;}9N4zX&Wr@MFy0=6<v6r5k_96?S{ZfDAFx`Bw
z@Ozq-IyED@lp`xr5=D(iGLpe+>dm-EvP2l7F<)idA|o6-pGb#)KV4I{n1PplaZ;dr
zn)k}i4dz!fwoGYes`j#}WD<#1tPV$OOi-V<s}<jI_6G&;&WJQNe<k}7-(5sv=8t?c
zUL-4j;y~TI><1nnB%-wBqqhu{4s7qkGqrw`$N7@c%4H^Ps7W^KzI;}tVBL6f2r_Ou
znYsRY6$10tH$24s`Ns$luwV9)$&Ym&?wBbzX!S9a@KB^wb5!1az_7B{ECpp{Z+1&`
zj;jfKWX!;z@fI<c?AXJAQeS!X2zJgHW&MhgkGKP^pT;W346EI4)S&#Z=Z^R`)ut?o
zvm)6&_o>QH|8QPk11S;L`~7Kphdh44MVR*v@&s3F3UA)-a5CnMCx$=n^?o&Q7BYO%
z1fVrj6>@4DxXwL3K<f=2tozJ|s}2E47K-}3!b#fnXOuhYL`$1b=0~67P(ZoVoyavb
z@98@evM33T=X{ZCsCncw&qrrAP}gss9GDB98Tt_=f^?Z7?y9xLw>PF;TX&<-Q7~1=
zrA7$OFw;Vrc<Gf(<X0YQrS$Y`3dV%>W?Gi5Rrk2)L?*3p9=vwmAi6AWGp@vj4WYma
z@Lf!Fj1IczPJk{2Z_pi#?5-wF&Ze4LqOsxVlWv~1I%*YLblStn_DXShjC^6<eJM@E
zIM%Xyeds{DMk!@1f8>_<X#Qrb=C5}gGPywA*K;9NlLlgh{Vee~(=(fs$(mE>>AT5K
z;ymVr-C=@631glRA_*8}v9$Mq=SpYj-2p`iHF_M`UDSt5<$C)1wpw45lrwOwJqiO2
zY|{~?;pl=PAM|k$G8R|@SEMI$%_u(kAk79go)bd*vYu&HAGV%<Kk+upd+QR;?130O
z?G8VC_M3kYY!XPt92`}j2vTA_Cf`^t+*A;^!83TEW6xtws~c-`x{BQN=ptZb6dymB
z@TPOfG>0q6$z{9pC`KU3_#VaI@M{5md5j_i=rPjH7op$QT<e^3IU^(E!Dj8;mc-Z^
zF5U1jOF{$j3VVw9Q>iwJXcGBh`{eadw%Q5$LmTcZXwqqKPEWmtduSY8uHCItA1^YT
zw#X2vZu9x}DLCIVKZ$?5HROVvEd9gi5xt_l@eQ6>j-dAJ-x1d((54y%?=Ad71o3$F
z`W*>8Mz2eG3cWv-kzM@*wQQhwt@E7a{xf%5o!*}H87eGsn9qH+7ueYd&y6II2{u<0
zRt{s}y~s1nDbM9lqXn4U&{#j2Kq8a94<^m@@rpCb)bz)15VF){<lfkp(Y&TrWZrkd
z)|E6VhvWK$CnO(B678!bnY+n0#@0-u8M*h~pde3Cb7@%_vLM69lFM#(nlGTfzCIty
zHb|*ZSiKc@L3Gt~ZvFz6yUw@WWE>V2=4s{(_{CV$h*QK7$p@_1&sX}=#GH;fBi{TI
zD;;MDKP(wK=byL>@rbzi&H_fus|phB_Z$~M^KSjCnjOf7hs*F7!#bxP_ct;5e$q!7
zthbe?R<3_IpcuFg!6OB|)RCTO`GxWlu0Ch`<N4DyY-}vG4JU~$lxhj_@m?8NgY<8{
zt>#1?cOr*vwFAbA?$LkyezJYCsw(l6LE^PjzraeXg4OUH{(}kQ^m6h1zK@{G{-FeK
za5D$rl@Pf!5w@<Rpyo9uz`1sVJOqaF3}+1~`EuYE_5g1qp@G+qa29WfwO+x45og8k
zASW@=$m)VA2;SVYm|*y)d;4STMOyLbQX&~j$K5aYB{x5ZX>U?3d{1TmfP~80D23l)
zF6sQxnCG6tAa}3Z&6C7dHzoB_HYX2T83=e96u}FlgOwL<1>m>x)Sr+k1O3U_pS`_>
z!j}Exs2d#?f)biUpsvrCFwL4kj_ETtbu}f=e$++`>ID_RV`i&1rS*hor=aWq^v>$=
z?TqK&-Fw)g3cZ(;%$iE&-`!#7dmf@L?IQu0=wN+fC|t?;qE-oV<AT{o798p6vPQDH
zj7ZJ#*#?&$p;%6l!MaYQ>7x5D$N2s;AgKXv%NuoysJZeZ*Ggkp>^7xr-*=axyMfNT
zVS*AygdTJK`VcWao*DTWvLLBY&ymM6!O2R!OZCKMY6IlsS1J9Xa@160EJ<n2in0{}
z3x_l-dz_@I1tiv3C;FAg3m(wU)ilxUv0wcB1%Jt)jTQ3~o8yKgUn{THO2ymdJ05!D
zrR9%#q&UO;ax+bwB+m1ByGUv!+}D!ZR~$OfN{RXlCL3~$Lt(j9eWH$4qj#hP*D<KM
z%601<KVoEmAUfjM%8E8?{2uO2oWT_VTfr$p24|<ny}g@^F5)D;Eb|RKs5q=nhnIZw
zQ;xZT=O)K_*o(lxtLL%-xn@(&hWfwTNI{_n^n<N&PkzY6k}4_0KflOzCS)LPCd_7n
zTplm9YTd21%3Q;V(0IQ<<q(ww5AGs1;u<N$CITvAQqCY;5IULcKXs+)!+ZKH^`11H
zLq2~^XppLpQb=V~hPp%(wb_1)pzEAWSzu(o?7d<OJ!OX?m)5-@S^Zlj()<XVBxreP
zLu{;SOx-YV3LKCwB#jkHeW2JIQCzonk#{CRn``|R);f_O2A|}2&{wHh1$nqnH-d|#
z-?vA9UUS<-a;T6YP0c8h%Gj`i)y*<_Xv$pTE$4@1_^nFYTVQ=^7(i`3CEp=#w;3IG
zvzIq3fB7UWWU+OZY~9*f0ovD)FjAlwiatZk`@FY&oyHPBsbst@>=!)=5N2_T1s|-)
zbE{)q)BN*~gU>${*_hOBET8w!rU-ID(^k@14+1Z_MU{G&mzCvd#A(l1Dt3bR9QceS
z<zBqQ{XUN>`OR+E?HtYR9#eRE*@(ladI_OR$>Ca=H>-MdTN<R&cy`4%SkC`aG{q$p
zH2wkzM>MzlZt9#g^+O<5-tU}9XDL%?FK=V7KFt~%`hdr_t9OjZl}#zT+;QgdDYmZE
zh;{pJeUK|EzJo|Zj2wpOCIcxv__g-{mjfji^Uh<5_x_iLq{KhF>3<^=wKn0oeK~ba
z6zg42D4dg8RN%?Jq0YSaKI$`vTY%0`@`5O)_Ug>@(~jrtZ`ls+yt=*<QK|UEz~4|u
zQM|0fWFUT_z1>IjAam#AGihIl6!<X|LQ$1C$a&W<mDz!+ndRDapXrlhr|#{K?+GI!
z^h0+D2JNrqym`{KTwP{7o-cbxi>0(Xx4r4<=bI=BSp;{P=|&FPn<@yYg;9F3ed~he
zI`xm!tZgB)<ui6p!nCD_htd&0Cjt*mC;7G}@O89AUqai!`Vy)XMm8mGd>M_MBe0)i
zP!luhEM?8JRjrJ`k}!bLdE{V<+KC?w`#(PdlD<YZ{qksh;wYYI&!h|C&5Dp9>N_Mj
z@)=cb!Y=Wpcw^Lh&g8sV3T(R5lzvFjP2QLgcxPNm%q7q@`rkjoR48~>#MtvI)PVAR
z|AjX7(CU&T?-jvTxoaR|MC+I)2&bcC-DRCoH*3Cym8mp!xzHN=W7mFv(!~z?LUeHq
zKWHapYKB7^L|z*l$Y$r>2`Vuf5iN9y38^RccC~e4;-<tl5Jz9ie#GyR&|_a#K|$Ls
zNxgW(rSarStv5RS3fB_J{U3`HCPgVyl_}G2W&jUOP~(gFT)LNAawRP6*KzqmkTPHc
zbR4Q4;FyzNMukgb-hfkJ9|W85vs0$<^d^@2!&0D(OHc2%GO>M%O<5})zK7BnsEz%;
zP!GLSc5uaZ7CCa0Q|#Ba3WyjwJF5bQ)h1F|A%d`d@zx169E!_7e*C`EzP^uoCe%Ta
zbsbIh2Hf6JoDzE{G-oV?AAd**Ri2h8#Vos0Y}zhOl$o>fwyt`XYq;p<-M-6Q5<f;C
z(1Cn`;6k*yEk74;!s?&c{oE|6%sDir(#^v=%is7VUzgL@r$%U_!UxSqVnmrgcAP!9
zgLcnFbqkYgy{(2~^0w^C#nh~|fyl=zTz$;{dxAd>bc6DvzQA2>9Po-D?tM|42ME%f
zv2{^q-zONupU~vlm%rPqKyh%xUA*jYwG0droOzk^9!n$!Dm7Y@!0{|GFAH&r{I8S&
zq4Pwo(@;Dp?Y;_r-=waf8i|mqBY8+!kU$Do;H2`)6+ctn`7BiXw6cOS>G7D2%a|4P
z(hDT0A)&+N9gSVUQ;wJKajac)v5omJiD?i6$LCh(R6<=5CM<LVe%t&tVU()9+^Kw>
z)s3a54@-n~1<t;g0+F|bpT<xRh@fCuo%y6!xnMm=xpx_7mEtl<>$Bma0N+XWpW-ry
z{W0nJoIL&YMGh48!Zf;92WdHDJcLDG>RwQ<dU$keJF(PKv^fs$^OjB|f??ZVX$C+%
z!Y|>FerglF_-uV@ER1XJ@4CWpI|P5;5$EtrZOF2$l^|(hxgqBFg@H=M4A)N>_C)qi
zgU+z;V1~2DnPWV`if1Q5k@ZAXDg+M(B1BPw<u8dczrRIrAv@reL?t@$%CWzUc|8s6
z(tBT2c%qefNgBGmB^~6@!7AE(!+mQhbDP2Dj!+;HVQ4}{J%Eq$?8R-P>oah@H=y!x
z39Ux>5?AuuPxsr91h*a4ev0TDD&b1kFUzS^TbEIY>U&V(F;_koy38v;=AnDE$ujH+
zZQX=@(T3}$y^VSkj3X>vj)wA~El{`XZRs(uFJTLpO1>n0R(T;+z6q#6yan%k%tJFa
z_<7kp{J}n?Cth<kib7R=f&BIw4W4LBeLLlrJo{3a#0Mnd)dnlfW6~n#kusG+0Xn5J
zNl+TUq+!EEkYhMS^L?5LElT(+Bzv&%V;dx-w|ar-rz#7O_chuy>=%-J-U+#;RR}Na
z5KEToC3KK)VS4JGWDx6B|Ivd~y%ejEdBj$!aOsvP3K@-WM@Vo`5JZ9F$M&v1MZvOI
z*x}ZLH}gwx5dRc}ew<9rscMyX5&OiK(cy7-ZXO3f(Zf07%$W=rZoi3d6Yx70wHfrL
zDwRf|9#DZ`-MU<9b$3D(@&sclJ(<ik2qGnn_A&o4W%T_HdXjR@UzgPvl1%%0<ZUk-
zSp5L|WyX<w_D41qBzU^^?;Lz%CPlp1F+$qYmzHx?T?yInw~&POSCi24F!<un(3A>i
zGBAZ`F0<lR_%pxI=C*dKPhI@#PLGHFn4z98-^2ZMgAeS`Hg<(H*4;iz1ZRX|z%!pI
zPquetz(RmCc@}r6O}o60<Ye;>3ZLj^B0BssXc|TFR0`CV(`CYnt~7i+Q`N)pE}poL
zW-V9NLW{n%%iO%~$(cx*aF%P7S%xt=H42s}Md<4`b(VGI#|BIn2Qa{l!rj{D5W+Ru
z$Z=(w<>%%JVS~+6tnRAZBk>DWB92jgQmd3AckepV;UjK_+;x?zC<6zYF(ad}wYL)4
z4?^MlZcga%{mC!57f($)m+!BCaMb;Iy1&j?#`l5Mb(dwHE$~Pcwq=JKkL7i}LioqK
z4ee(S#=l;2+k}QT0=2Qv8=x93x=GQ?*&*{U9R%Q5tEHRcNlSGNGMD-W#jzu_!KC-z
zZh!4@ex}R$&|-WGDOykWHQ9Vad{L^|^tvNYAkr_@^;YtQ#!0^CTh|U~+s~Ihng<7!
zk1igS9pBfA{zDsOx&Zs9W7bI^Wk?8)R-s6V*#3QCUXJT_1QcDqb77-&u{*Gtd`$Ck
zJoJM_cBSzr3%`jTO3|!%&&xmejSWITK%lE(#$+;HsQJgx$Va}Gl2b}mcX6=xht)~o
z!+4f@o41}k*SsgX5R{VQzsED_s7`?1V_SUaO^Ocp7HT7*_%Y&-ga4)FU;Bbn;WNbT
z`{Ww&Izow0MA8N~(M*&+;jvAga?BgbUW?zn0Om}GRUyoKW@_m0GoEr1{C^K+hzo|A
z_}V{M8{&&?Df=LaMYwQu#7Hm$T>2-g0qwrHZ|X3v<AyWZjtXa2?#~C65)PG(f36M1
zvplhR#6$MmDE>5!KmJ*Wm8sby;0MvBu3#QncDn%wkfiy;uZ&};tl#8`-C&JMsOFd<
z?aR9XpUd);rpzOOpYbnuhWeu(iYl}TaM`<KVpWK4%Gs7H-Ed<L<3qvCdVqE!y2W4x
zgMQbcW0G!JWO<OZ9^~0X8Hgmypm2@+;~;;#Bihu$l?Vx6U2l6dps)&s3-{NBh>tH_
z$W8Aip@@O7>Dw52SS%}{YK5M)B`K>$zS8>Ar19K8&c#2sUg|aJ<xuz*cO?PZ2IZwJ
z*Fs1Zr$iP){-nJo6yEzn6l_xcw0~^R66b!VTS}JFRHM_o^q<%EzihNHA29YdwDJfJ
zEC|;0CDBc6s*&QR!cVu}eySKyo<i<Po+;nC5ng8t32En)xueBgTCd*`hHrHP4tOT<
zo4H92lK%3YQikL}au5C9MM`BR3riG+cMURRxtf@~$iggsLi+Kna==-|3ZG+eB1}+$
z9idUE7rz9o&SRBS&A*QJnixVi=3$&Z(L;9kzG^Wl{J_DR+me0+VdwbB{~gX@hc4Gk
zabwvwwJyBsvhf|6HsDnf1!|YeQJer}`mZy+P9a=*aA}~f+Ym=e6%8IOCH_cK@^hih
z!q&&}D4gDVL;;PiE&Vv*0D#uHyqq5^gMWRqhvLE)nW`(Pr2mVktxAK>^>oypDAK38
zly#tKYsq&x9^cXvmT@+akz_A}a@=p@r+M<QmQIR$4D+g}4R2b$(iCP)*Mt8$jGvcF
zI0k$zC)%sg>yg5uF(D#_;MHRrvqQg3RyyD{tcS<KP4iRZ@=l@aE*@|oZ2D<;4#NI&
z2)BZPd3^Wj1MFGs>nLnXH>kdHp+IT~Kas8A;DoNvjy>;q7EGyd6HAj*_Ak3cMk!o5
zCKl6HM#ci%*be#;GL}6Uj0VL)`dUm`tsb<hG>+3GC(*T8fmJP6X-c;GP0Qat<`5M)
zlVqN@^}%QqkC+3ITM7e%9~D;7xCH3UH<0zG{wc)*YG8EA+;=E3A^Rhz6>R@LF>3on
zuqR~qY;1dQA=VEcOlbuPuJ+Z)2T9Uw+-RKT#tCOOpGJDoZ+oQKYD_ba#6MLIMCQFt
z#`8XlW`a$5)qDTt(7q6X8>FA|;Gi}KLNP-2^kHShL)WRgxs`;^UtUbD<?FZTMg{tF
z;e>DETWY8l>J=zWjl8OU_%Gdc32l!MjC974LDG8}?VEFRs8R~fX}GB5=H({{3yzGw
z_j1o3pU?G4Ed*hsKM|~@&U=W0zGq5rHD`$i?|D?nd|+BI$@rHAQuvA81(K?C--)VO
z?vl~RC<hR-z)%D5EIp0ev%eR<3Vx*LmR}muK9nc_<6^GdXA~`zFBo1{W4OyqBuZY2
z`T~g(h<uldZu>U@@qhvWNlztkU8}h8pXk8<i88zBdS6J=|Nq#!%CM-`wM{$p&>cfc
zgLH^AC@mpfLw86>Njo%xgwmjd3=KnfNeTi35>gUUf^>h2z4v#{{&B9$%O4_Zz3;Q0
zyq}U}YR;U-eV%mDT`-i_rss%TUk1F@>BB6||2XFsCvbh!zOV7VDJT)03LKouTP1rN
zLP86tDi1jX%<C*48|NDUuAQY`%_sPdw}ff6QAuR6j{pb=ND}5WdhD+L#}{9Q03!8R
z$Z*&3>i%#UBR|>t5|<xHGeD;hgl@*has}1zzgX+IkPA)B1_oD`dIK{Y)3jO?@)0_X
z-~aX+{%xxOZ{-8vt!Qd8NWKQ&dn*w*^_<hL*!9fJ%nuwAc3x(nY;ciLr-$$@`>U!5
zAu`O#uNyLLDz%J>aq1Vdh+H%OhlKfO0tEb#!*E(Lim{D-Z(o$^Y<{VuL0*3WAzX2Z
zeQkvBx}OJ6?{T_RzA!E^g>Ho-Qf$Nj$8E=-0Q`OW)x?tpA4HjtpiKF7?Wzn&3)Zs#
zG8E{2E|wo<r5<)Fs7~)=O?4rBGEMncDD$*<@;PTJ#$DPyPguOLqWG_U_<xh5l2?2m
z&i(pjxjCn3VO+warjpl6av37$(gN~D(D~(%gl4dyW6#iV0uIDUce69;WW|hwvf@t3
z;_U1!>l9$r3?vU~@c-9$cLF}qmA}8Z*nRBHaD_@7Plv6Zaq-gl&j)*8MbuA(|M1DS
z_`RZzOT+&k%Ar#RFwv6WNvX#0h^lmjt?)3ml7d3yq$JvgC$|u%rTOfM7!D_rmt3Fq
z#ZZ&kW*%9&ma6ieEVk$4|8X2ffD0EMT&9I!N%G6>y(4l&XA3o8Ats8FKG=WvAV?Ny
zqSX|NdTXm!V#>|!`*M65aJOGXYpW`|C|B2d{Er82NuqVN(-1vSKzjy+PRY@qK1a-1
zKXOnoCU+z5Tu_hdUw1f-Bo%f@&qVXxlM7ZXl1mBG!r9}kkC-Ei*Hq0%EPnh>|F<8;
zg9aoEl84tZI+WvMG9|`DfT3(7qocx($iRf9Edy@0=!mQHKNZxrtp7a+B|kvk2hQCO
zEE_fQMYh}l;Xk>W=R@M7O`1_L%Rzdt0n425WjL(GppKQydlFCr)7H?!{1pVjCtc2`
zY$<PFO1LCPcWDDBDTjTh{QD&6Xbi8fs{?__%xISN+79|~Q&ZE_*TC#zw<-9A6n)~b
zv0ED}us0#6>=Lf<FSDSPDf2a^KKMjQNp&BpgQTZVCtwOy5$8^c)CRnZ+*(3kZj%~Z
z>bX4l-NHK)@ygZxg1Cq&!$pO#nwvQvocczYBCQq%o?xZFo;l8e_M*}(6__rcLGxyh
z7wo%Ohx=l$w>QG1(LI0B)FR-&G6K{?`Rl{uQ<*$DeXW<-BG6qKP-zO&66Q_?_9D|5
z5eXP1fCpe*_;bIz&VYC5nTP44K6o$MMA{qKe1uJe^WTDyr;d@p@PyayYfh?L#|<*I
z1)-j~IE~lMkVjam`<Z04R{dXue~1>y7MNEQ;En~5u47YT8P~p`I=Tq!J>U9#nmt}M
z>WWZ7A)=JS!sf4E3MC+F48Cc`mLHF|GOd46Ve+egqv>DI1Ni9iI=pYt>>=s4<QLlU
z6%r6uI9nGcbc$a3s%Mr-4ifi3dA-_A|IY`R5p~(NzU~+`Cs1TwN4<P*v}Ow#jeAe7
z?IQtEl1_=g$wJ5MzB?-*?z?aDVxb<XCn+pEU42+>`L7$Wpp|ow10$UCeF6hpN6Tx(
z0q`Z51$_F(hDnI+tJhIQ)SGie=BGGriWfHDmfF|UWK&|u0^?Jf8@E}w{&5pKQL#Hs
zfyvvZhQ`K{7PzC8QJJeSJ=GU3N4t$!|Fwcs%m4~?^2J>oh!r(SKqW0AWLe|75&MMD
zSL#BDlIX!_>(4b4<FlFnL9J}D4Se@3rS(YTmEsH2D*$Xc_T<0*ZJsph8K?4%$T{}?
z8)zj^S*TSp){^m9Ox#*ZnD&eW%A*HGOe=e0{$Y@1GPE~TE#xc12MV;N^=98CXEpyr
zhDt!Y?4-W8dICKD6bvdU^Wl`Ls3~{*YIj2Z^O?Be2%nQygl8M4iN-&Cp2GqJev!Bo
z1gy3l-qS_TIW%jmivM-C7x$WU`n_cbpk}Fw0T;V^xk@2uCCVFE(AD&iD^hqc^-sl&
zrQ$y)IKT<*ai5tps0Ud61c&_eUqF)-fHIZa6~)l8q7s)VGXBIJ{MhWYNG|3mxE{w%
zroT|p?f?H5q(W0pq~B|75h0GwFsOey?Qh#)q5PjiNhtsidcziL3Os056=i=edN~6l
zC=>_^4U@UZbm69{1<lTe|CoimCD6-nrob2M*yEuk{QplVQ+Xd;I|+8*0u&cgL^*VX
zhQi_;#V{H^v9LY`Y)<`Z)V!X-|Hm^@^1xeSDfn=bL+V>P^#X3MUo|_<<EQQYH@fH~
zzgNX%z+0s>85vQZVzLV@W?ab|Ax0FSIk|oW|A-`Ss79mH%4a$T28#aA<+0fe|1}XZ
z!x#%(G(2uGf0+##I}jtO+#@lTLz$5itR#Eq`bOT&zPKIl|Ae%ya?4-dG6SVjm37mm
zfF=PxqP3^zH!jM5u><h|EpV)y%tt;M7%F+>p-F&0J#J8*xXnA<X*#R`^}aYj|NG0i
zk(~e1e9d|2*UA%~SBBuB{0{nmVgVGokKrIp`u8vz8fv4{Ko~ReX{3p-sQJ*aqow+R
z`Ckwa@Ka!3p`f5(zv+AeF~OT(pIsiPApV78;sbuM3kr+@fb{&rQHbLeukVq&=6!d=
zp=4f{p%@&*UueiZAd-l&TmcljJ381S^#b<h>eJ>{hW@J`Wzgd9L%0!`@IAfdP@?<U
z!lA?mm?mdpWfgOuRzJTRVx8HkV*2L>evv2(^}rS`$I=$}X|q;8Z8kPGBgZ#f|Aya%
z*nlevFeC4;2tw`nYL^_xioQ>$pvW=g?;$Vp%pox@d6DHGdENlk5x1mO%xxIdiu{*K
zb@wQE8Ag-zy;dNqitY!KcexCvdDgr$pUwl>G?Z2SZ5e^vM|B<_*O{~jUdY@;mXUk7
z{rdbEEM4(0+oKL(1M@<UD(-19i~&Zo6ia%S@npL>^TJbBKh_CQ^^yNLXP0j~ffD@8
z1k;^KcK!bT{-4{^WjC-0DAV6(1^x!g>4Cek7e{6T|7IU4Gs}<`%|;UMlJ-bV2#NEu
z;_pakd=elzFtPka<~FlQJ<Jyn8J*dG$EHR=aD-qDGy%fZJPK@#X;VpZ+P6u3Kbr#p
z-_gV%Rjz3ZzzF{#5h@>%eg8dI0t+ilT6{)T<*olgFz@BE^9XSL{DX1CQGnAqqQPFo
z@`DPg3A$W3o<2>9bNOHm{TCw?6&0BbRMb}(f&RlDe}%VifUL{jO7%t>@I->QGUF8Q
zJn$ozXI{Ii7QIXT)EzT~P!rlnquTigxO;=LPy?7j1PgI=kfb`W7v1k1^51Xw@?A&(
zSz+O##o9fWjhmsN=9(vmLa1<hO+-z`6`)Sr@KI91e=9m-RMCB-Mn|Hhv<f4>QSPh%
zU}fWc_kQiLnkOyrr~pvJ4Mbf?1Yvg(j4Ps7qPRP!^a;K{pT)=bDk$1x^;`TrjOq#d
zU*Bp=I#g-jC{S8Nq~D-^F_-i7KmM$R48ZPm<I$T{Pk|IE2H^hN^K%fro;`8*T5koi
zORUA%1CFV3rHnM|>V`P*!Qtf^=ks6P<$ck^_v8lBTq!8Mvx9?!&DBow`AHXeVLZ71
zy?avU3czgv=Oayk77vwY3Q>S+n=ACAGgmZ#2ArF-knW(J`5g<Y)$e1Q6zd*7`Z=j)
zU##2}k!TLyNb<TIx1Du=LAc642E~+l`Rr=Fyy3?jNlRL_V~Ysu!rxh>zd{pd;7!BX
zp#Te2Vlz(68o2A;;{=26JZnr$Qd^syiF$lQB-sfCHAi<G_9Nl%ke{}=np_f2{=Np_
z*n?!=segLzxjvL?dJ|SgKD_mLuk@Fd-9JAM!N7`>aqsLOUID;)JWA9KHfs6gYHR-d
zYWo)^@$G=G^gLK^h&qZqVJ_avsAfH@%-b;fA3<uB0hGMxQlJj7r4Ryit#ljIn`H=}
z{s%z>C_io}3920tNbU?F0YN4*nw>H^0Ec2^W5c?-N-ZBj<qrQ_mX@*6M10D<XRyL$
zn}B?MT!P41z^~}8nw0;lu>Zt*oY+a7T_B7pQ%PFd<i;l(&8WSleNX<QP+y<{6aRns
zt>IJA)2AICdNld$8Fdh(uKicMQNMyMC^5dOkx2q9=&psPB1C=&Sg>)}$J)#XE=9Ng
z<1gPr<NjY=NGaSK%xLUaK%odLEy1MTPJR>g#eZz;WjYYv-DVpn!~s$&%lJ^13L9_O
zYpIn~h|OglXSBgZelQB4b*m<$@x#GyO0r0i*yO?Vk6Np;H8XbgaMZaV5{gfG0?r+Q
zjX!o4ZSIaY+`s%zF)+xbj``K3F7<E)<u|3*z?M4Nr{_=qT(pwLDjr!u?sZf!V@T$r
zp9D7K<3ab84@FG<9^26AJ?AFKv}dJHcQ^Ovnp#?f=Va3PzyR)H0FlPnyf+<uKeKHs
zn``y(X@yoA(<9*;zYv+JV-McR&|*y~et$l1qV#Z7GACx)++Jo>D!uB{EqjCKgGq*C
za%gQbC`>LQnW-kc4Lnm;RDa3CF*+)&R%OZ&b6$moDPOCkynv@~E_hZ)3=25nea|lW
z({6oSr<ww}JtdCv-?f~y2Nk#U1}1;$BOb8IW{l@%UlC6ID?{!4v`c4P{w7XrA;yq-
z26s4xr_J{-sn*)ryY96N47kIkgD?g8P%NGUDDf6Is+hf~hzKHbZp<~l{N5@764|Nq
z^fVe(>ao^CQj9PcA!=bJS4o_{IUg-+8i0!)%cEHBI$Fx>KXa<7xKmFTVKp8SFf{l&
z0+s|aXh30^h;B*S;Y;}(?hTgWeItOKLVH-XbXR;VAM-UjBy*+x?0y*+wzr~z+OWZh
za%kn^G772(3;va20g|E126nkjolDgQaGtPx5-fR%NfT3|pt(<MJlo#lgx`l8$XNFY
zTOkn_OBAKA5J23L?q{DIca?vylX{*{{<;?MY=8x<?LiLS-+#P*pPCmX*C<9}rxLNR
z)+ggeW=VjYnNdGS|BKNnh0q!e>zkS!?o%?x@le%jpYc=1gmjF{IGIwDWE(OEyQ8BT
z!-aQo&GxF7iGTBQUM)rrx81q`Ks7nr9hPy7&^R2#xkf_2yr;t2Y)w+9CGi*QM?f-_
zM63bG{*HTLFwU+Tzzh&8?oJ&&d)P(eUa(mRKp0s_h@?KmT#Iq8Xj)x0-$%<fXALl8
zTWq1uvZjGv{+1PYHUN)q;iI|FR@lUF&N^DLH!lW2pJRQbaGx%kbcm9sOuE?M2ZU&w
zL<Y}T%5vQCAtjKIp*kV~oGU``L9}L6n7rpBPu=HT9(w|UkN}Ih*?W&UktD->dR&kb
z$DvW}{u9wNh5Ur7I<U|UpJ`RjrUtX4SfV97`kO>Af&AKAcf3k{fC==sJWPTAp2N8O
zBGgALBZ*{!*;D8pamTIXf?WNo`ztRefJGMB_!~Y<;(F0vH9sa?kj9>3N*YYJc4(`Q
z?KRr-q{0S*7_s`lgP2Z6;?7a7nEG(BBU(X#`{>d7EoorG)nDRhXI^OC(LD=&`$Arz
z>fTrFWF#{9&i(A|`}PM8L*n{`L|*`O_JrFsf&RmzufFwq+TBMeVt`pkC3XkP*k|bH
ziioiYhadeF0%x<Mbcs;IhKQElMMJMxu1spyO*TbG{>q>OiIp$pL)5WOomitQ$sx+1
zcS59SP^7<g&E>~TFxx-5KTZrVPN;xWlj!|XLc&p`Du`Yi6m_R~a{8;Sh8iee$iVGS
zWELg}YW@N@wopQeQ|Ab@?z7hPXpM>8mG%J!HDrH#yg)WQw1;4KHeSFp6CfA53d-2r
z;CN$J0<Gwwv!tFHDF&PH4yhU9e*wkliXoO@EXQVX17b>UG$4u5AI=<hC!M-kqh;XA
zg~IhMEhvz@vE1Sx1Ohnhzy9Do1blLIi9=cuY&;ym)pX0MM-~6<;Vl0s-8?Dz)=Bv=
zl`FbWh)G@Y1N2_jS+vqS3zZYC9UPvHWr;Hbkem0;;oojMP!Cbn*!b{k&f^rIE%bd(
zPMSqh`pM0c9a_M*0Y25Tc2JUSCX_W4m35h-g9Tth6FP+KYc1PauYy-2#zkER4|tfZ
z(SeunVvl_NI?SFM$OJ58NdjV>At0W3IM-}F;CO$8aW@PRQPI)gT|+}LS<*p9K$9gC
z(6>nHa^+HRpNR?$Wd*X{7R5Q}>_9gTils0)Y)E_QorH7vAE7G~&-Fl6uygAi(q<{G
zxhoYA6s!=LY6BsPb$6Y|uWwCUf3$K`w*J9UoJY+FAo$DTIueCg07<7{3ZU-PEvGfa
z9=F(q5nNJ_lm3rC7Xd(<iE5-R_o-tER1kvbwL(!(3R}Wn$}ZeFt>%I2tlRTgJvj(%
zBN1VR0p~lGw#Ih1g9oq2`SMi^<R<dY(MD{(4+*8K+N`%hK0P1Yb;&m{U>daNQuj0^
zB&jO6Z^s7;39XFlBIiZktNHs`kNURhj;Ojq6>ERY5Gt-6UGy^Gsq)iKL$y^7+9y9H
z4XZ$E*+Sjb!rnHzPPPnmKCjM;OUH-iF7d6{%U$QgoFe+<G(0oQnEYvv9)w+H@>_2d
zE<NLI<LnDu4`ll2ML*?nueHFPz9a=MY$*cyo^bmBU`5|r9i8<AxpSut6xu`EBbT3q
zhmZgMuEsADWy9OWMct}8qj$~YaOXEktSI*Kd8$V7sJDg~t;l9g*4H7E=kSva=}|Ml
zsL&>%@J{wn16TzB`4Tzj4*Pqo$&{C3Fb4OQ?_Jm5g<j4?r!vx3v>UuNW(aI}eFRp~
z{TOLalb~#Uyp<7`la-Wz*qaYyz49G1&Luy>Y`R&&CsV5#5@am(((&9b-fR;Fog(%y
zvI@$*=kE6^in%<Ht>a0LNjCH2^b&pI(5dfMdf_4UEt57{fa-$;t31g-y!&i)P17TG
zRl<(ck)hULbNNRe8|qKDM1q2UH!TeLuhcbNo0u(}hs%;_z4bZnwV@I}jp2HkD(=s8
z)bNC4VG;N9xTOE89#xL`rehGF-iMBpXzQ8umsvJcAEY2U=|4{DRDQFb?m9)v)Y}LA
z{_$8m;1mwTBkIpfr(|Lf7lCG-D?l@nXcACc8FzGkez>;aBBh2R`ndyGHfv~Lz~6n9
z>DJD{u8~FgC1{~%vT@N{iX-Z`L%vp$0u!X;dYzAz`V$9ch=f<s9AN<dz`dl*V1!ZQ
zSTcz@$}x3Xw&-eYYNlXf_hWrEG)|Kf;oISxF_{drZ~605+xE2|XW>Bq&vU0@77?YH
zx5gphFQQX}S*F7(GJSI^4s7Y3!>-_M**XBO6dr@X5KpcT!dox5%h)P-H2kmXWilRq
z4i|1gEW5(|(K%`**d53iYL(NicYl)INY^i|G&Kov(eN>p)sl;%2Gs~$D-<wuc=rZO
zoWSFzc^s)8Ol6yVacy36ada7-aI{HSj_A)ifew~xIf-?(iGCcg&uDwFrR4f#tn|ge
zPg0lk^z?P04BB&lp(&KqWo0THN?4G?(_}JMrZXJuMwF4AoyxA6lhM>S`A9QYIu!^u
z+NC#tZ4M4xaLF|-Z{W1&0<&&)0R&(@2cQyv6~;204x5D)e7KWcmN!B%MOOQ}=cYyb
zeMbFBZabYp#WXq}nQAgHU`8ZcFC7{6>f_yPX0i3EZVd0dqd}l1+5?tl;208Aq!T_A
zykeKa(I0uhJVT%4*2`~AIe*6IA>d;($<jGbf-i;f6Ty;<hjV4`wQ%Qoje?HN7wfZq
zrYoe{46%jtpb`98Dy1niJ^2)lDSLaLrfdQNg0ZT7H73Xp$9^v~F$1ej(r3&9MnT5m
zwDkFtS3PS3iI`7i4_l-pBpgk0U*d=qNsEfwS@Fx)^^9&n2NI^$KH`X$SJ)7E{C_RL
z;_K?%wTV~zU*j+Q+xioq+}g{T=P~vUm+BO^N`H*HDK?k~6LA4H;#d&1!)%-_99z7!
z+vOq6yUmOA=igqt;>a;>h2Oe<K6S~K5ql_DNuSD`?n}?XwGo^Am2u2cNuG^+_)iL}
zcipFW0<$mV*1rzJWIFqzr)Q(1+iiM6-^z%BbYl@;$P)&nSLzhu?ZG@SLP`JYWITqZ
zs>cecj`0s;2PS7W>A&=Mlf-8meOsGshD%*ch&c^>ZJj`7Dq#m<w8IPBZTCJ!zsQ_Y
zzFkq-vo?I?GaO;sDJvonq|_soWdT3XK(m`?0pgkyg?XTT))W9)9(sCP@UC1&h$Rd(
zC(?>r^?JnA$IHquD}BLf<f?XK`Z#%T@S}K8BS*?%wmUvLqh1A05#5;h=|>|wd?FTu
z5y`|WijzYQKq0ZUz0&uJUnUcq@tv#`I0ZUEh5eap$IwRu=dx1!dxGJ7i50`^gn`6@
z!Hj-J+C_SszR@D4Xq=kTqsfiKn+r+4pT8Os)j*is=I>3@ikGC4&$qxSHYbO-UHRUL
zfOd$*%?l=Xd3()Sl8?2_?M+UbXvM5U%<$y1Xu`3=-R;bV_K6%h@10$8mNeH&{#%5)
zuF6_Xf{YNd*JO(%*};2mJw^+&uQiH7#vWGA^1_7%OasJdTexUq<V+koBAH}jE^bTh
zE>>Ccb&(lgBBb{+ZDH{i__9Q;7X}TIh-FOI<pu`PMpvRjhLh?gxsvtUorLOiC(ETy
zT+FKhMp6l<2bO8#N`R@20vuVT8zOP)9P-}UpwH0*HS0HPMtZB$1c+J$xdEuLU52&G
zKr6dpB_F2n21-m5KfIXe`Yc#yjfgC!=KAH+oTWgv;$TmNzF>WNspL!qn)Bg%9h15`
zF0MesKD%S=>Q}}*TNKacFOmHo2nmsjC16*P%i%1gRSTquC)txx_56NV!lQ?00RjWe
zWNOUO9#q=tPCnj0Sl9%^%KKACUATX9A=8F0Z$y@uJ6e8!awUvz#NbOK$DS3F=b=BS
zr`VI&Xt<(WWMZt)KvO4ik3qnEgLjZTM?d?39dVWmainb@l;+-&&3TVFBE0q%TqWl{
z6VN+$!1gUGMUe+ToR;4DT=Ui!#@n6=Bi+;h*bS-#2Y^O?MhwRGs2j+80?p4BG)e4D
zo5{RL&7y+zeQJ#Slh`l81F<QbAOwkB@$tm6`m6Krrd}&qGn5D}&e@I1pF$TJdkRnb
zve-2J+qwA|MjVpM@i|Jyao@5^k*1y-m?$MAIDD(Y28K8CDb>|JYGH{bG<TGAYTnBt
zGFZV$aH^^8WvO|msiH9NqWkB4D?-rCER!pf^Zci}alXAuL3Bo3_WO5JzuwtOS3WM<
zK7BE5BjxL)kE}Nr-=lam!>+^p$kSSPK%^q^dCOI46jZ#eir33Gclc6jZU`shrxtes
zfo%|NfTKabu!cK}VL)6-hx=<Y@$!;`-AORn+9MDDXYz*dFWuyKGUJt&d<nn)EbjMh
z3mgXhCJM1{X>aGv%*?bB<X~16<NfwPu@1kV*niEwh6%Y%N+zrz2PuXa(SCv_%#LiU
z8ctSWQ<3?+1(%s{ZsV8aZC83kCupczZP(Fr^~ZT3nty)dsySWFiRt80TBeis*LKrY
z|4r;qlxDeiVM>yKn(x;SmJ}6%=i`~#H%7flHa4FMwi&t1_mgAl;UM{eI^STUC(rOP
zJ!fa7>f=fJs0R^i;jmfYUH*^9pO1M+>Cy(h+z`QU%KY=_{1TJMGKYbsMD%!Rh?KXO
zh4RD+iCfi|CB6NL_E{@)z>MDXKU8h9`C|^f569f#+v?rIJwugD{tq9q97aPQeXI^5
z`dlzES`BAYF`gFhyR{%5;Mo3hkRgF6u79FD8jnIw3=~Ch&2^VX(fI4v1p&JnZGimI
zCPt~opFX#>o;QT+>%-}`3&onbX@KI-RIb?bTsXiGyf%}Y>MuNlLN6t&*6C%o&|8<!
z<-57AHwu=uZ@xac3zWn|Ln;%GZ)4=y-`(C^Z4_}{hWmgs$$)TKNl7iYYSDjyi)FE!
zwt(?x_kFVvaUv5fdN-~`jsP3%4<aw|9B-%F1^<x#1YCr7qtI&V?CbB}ymA|Y;O47~
zKD3~klfCP5w6*Pz>oU|j>k5BRBz?|o#)EVD=VZp5hV3CG34PETT-xZPDq3%T^Tvy;
zXP@GjDH4a)yhyowq#(L-ZD?b&1Y_}9U@KICsPL|#CQ(YJrwn762cp~ZEQljdplwMI
zVT)_hHLJ<*Gs0o8rX<lHu?OLxkFY#uSo0!><LVnwTv1#mG^Yt~;2s8_NjCI|ITwhG
zJf7>k#UB$f;}@HPzPUM`C{(t6d?3ScwuVxOa!Pnia)sI-Y0wjovq5qNLe?HSRufp!
z+yT2tMa#k&<ZPE`r``G%$t*hJKdk))3<ZIJlxikUsY=mZP;zGKbhrQLoC+`E?0K85
z9<MIiik|`7*hgJ(mJY;vI*XJsZ7(LKI>oN&lz+~^B%R&-aZ+Reb($P>k$lBsaR|J9
zR{4v0H29T=7ZMgT@I#EXXR=UO(`v@NVsMS`_wctuOlG*m{P*#%Y4mH_>mi1BQzu^3
z-z<AUMfF2}CKX@hSsh&t>v1^dGA5;|8oL^p+D&&b#U`2$)mG9EHJ$NM2{f4*iyb!J
z9rC<7Tyy?pzAYl+@~a5a+}?WC4Jxw>Mv#Qp>gQH*=MHj|z4>!}Rh+C`g6&ckfmjAj
zk$g|JQ+?#CtVl~1nV9+0RVUyQlZg5-VzWL+)3iH-0XC3$7q|rK_#`(d_VF!cLSGBU
zXGLHI><!Wydm!4UEyeS9f{*Xlw}B%J3m&a<tz4yIPO8CPdWj@U!`r*rJZ}+ICpZ0P
z-`EYwG$&*IwaesweW_OU7hF${32Na(^Wwy*=<qA)-pRV)mA6-<#Dcr2^0t`}y@b?k
zTY=sTYX*<ooW)auJSc;D@OwvMmOza(M~T+k-vX2<dm{&<n!_PD=_gi=x^R<)f*U#4
z$>XIotv~}dU3htr!h8kt<jEtaSby|gr|%g7)~k5T0t=_rdYChHY%V>GQEZ9Bv!zb{
z+@G`ahzH7#zrnes&!0u^P}m0JYaY#qzM3+I`>UCl`*E)fh_ZMe(vdI~>Jy8NbQQ1_
z8o#<O92=g~H%`qsZHiaSPFJ>l{o&QgrtzLH9kZki+5;vF2EHh*mq#1wM*#=Gh$5=P
zPctSF52!7|QyEE_I4FDdNkRJHsH74SHO2(PLoz8X!V^fK+;~_c_8}e<O)u9}y@n#p
z3S1BW5;J=U{+wmSgABKTxyZuncA#{1v8Rw;v-ZI65rSbbV^c<ZNF63l@UXy}j+;HV
zQL*vYq;?DbpiD{z1)=MBm(@rw9cPNQ2u3VzQYOCMC4UWlSO$2y>i6scq5T8AaRLQd
zST5K_01H;3kJp#+8)!Vd{Pi<R+VU~%<HHORP`zJ%1=++QRqs=7<xm)`Nw%xmpna(c
zA43N;H~T=#y|8<Xt{+fhOO%HT`{?I*ft~YXxMTT@ouVl}@dC$Rj8M(VKY>`kKAupX
zpAXN7z7@m03_B9p6`XC@%@Q{ZXD4>gH&;&F{++pYjqZ~Fo06j6b(4N)UWDIn)~@<1
zL=9)7qZuDmIZJRE9$YXMY{j#IdlVfeUD!@jO}-?2jM=&y94B({xb{l&hx~4lfzv_F
zJ@z4U2c*eseWb*md(6g|os;RG;!#t4J3!VnGKVnuiy0~>CIY{YX|){HEK}Au7auA7
z>iFXE`<5FMD8K#!c1hv^j+Q86@~Yk!GBqiaI`sT{2}>%Ie8RhzFZdRwA)EFCr?UuV
z&em=Y+{5$T0E&3@=}O+`cCs{Q((UCHd}JTI=MferD%Kd5s0A__k<Z0^-L5xinvRFd
z9{D`a1`GEt1>AY-<OJQ#f+kvGwz0R$o=z2MtY$K7MIS5O%&L_fZN&yqqf@%<!9=#w
zH3xwC)+OxXeSxxRdP#~m*F|h?je&5-7L|V8?-g4J$eh(krt#p4*7uyweY0x8-_qyH
z=w1c!DNEZay<^VR4aR-ZrQTU}g~eiq2n}Ns{^ede_I3BUeP7GT1Q*}3_+*@rv5bmf
zG8FW#Aw|A!2p|kbi?<DN28}wL1GEe057SSs@|O;+AY8F$xA5p*&&!+kB3^%GG3=X)
z)Mv0!(+gfJI%Z-CoCtXs3^5j)nXB>?I!5kwp7VQ3cz}%7sg74pl>5nZkHc>7lqx)e
z<z%);Jpy*VYcPs-FIxB8wic`3t-~dhE$(MuIX#z_rq5!$3==Drob`xgUsaY`7=Lm7
zjB=BPSNg@sRDZg<_2UsN!fD=FH&1w^&hzOe!EVguP)We6+f%zoK^J_EF&(jl?GuQ0
z0RsDxj#|pM40|PF(pNXXM_-ri=$`FWEiJYAzG%N@V4yp-ZEBNeLRmp$GL*#aWoTuL
zk`0tM(SjksahDgqvpBM+qU2Cz)TQksdM?M+YmAfSt?(~V>joW{zSKLCifn=V&Lm-z
zxp&ugYZ5Pc-(*CKRP&l;r1N!Le}lz<HOeAor>f`s{l(sdjXTXppuF)Y$Sc>BzUYOy
zh+sWTgJ!xz<vMGba&Bn4=T3GmuL?J<$E^C~K^^uD%@&z{`EKX5YKLK4c!x6y)gNf)
zs^N#C2&XHz`P!XXU><SALzovEf8-G<lNC=W&7Pcs!s|{#@8wI32XPOM!hVN&nQ1+k
z_Z}P0y**~1#M<a{k<IkU?UDd%L8b~id;%!bz2>)uUzi8kDCmyF@?;pdj_25*hmv^3
z9!V_FR$;0!jzjmeuUL@-z6)a%(zRDn>#MYxUO#sIui_Uk7TsNaJG>9xFIMi+&R;IC
zF4#HlJ&B(??cRKe^tVi#RR$u53UKFIDRb9HBIlm(Nv&UiE@FZ{ocgwX{cW7aAdwjN
z2R)OLgMl%e1h%5SxeKzb3XNed`TCml$&)>v*K<#UNE=sk)cIV#)Y52LYctHhDQUX4
z)|RJY5ytiZCg!Kj_)Pk4<ri<hhy-o=uUJ9OD*F=L_~J~nALF=t>}(2X3ZwYmb<$U-
z)zS#5k1nhTWwh$!iF86A?0RriVjCcbW>0na$}OpLf^Sryhuej!oK@EP-#pI<pA0$O
z&1g>HH(XVlioMMJ<&!P9*LdonggJF?&>a!r=V7ORl{nzX)phJ-+2^C5TbW_)Qp6*6
zp$mR;D6vqR!imH>_E?Q;)tU&<f0g=<EUO>X!e3M%o>g;#vruRIWQIhHC9J14RA6J3
zABPuZx6t`bb(`!WmPVt;+fZC`7@`pzVhpOt3LY|xCkX)6&}^5~^;J?>gAW<~E6TjQ
z*T3(%<coZKI-r!$&V(JL`-1_8W?-jT!VrSB>K36a)BQ;7g?6RB3P!_~u9-Z$%$9sc
zlx2Jwu4n~k4ClDC&e?8wOlOG|s~kfd{p-HoUb-ho&3F!CSb8ke*>}|vH(UkkV&X-X
z&wkAM#9(y`w4r#-r`+sx^~GwTSgaSxaF4><78*Evy~N%$L>etj>9GIV8suUdLE%NK
zrBSW9K7a7t?_w6V@GFwtfA_<9l^|5&wd6om>-)m1WWyYtEY4?JZhoE{aJIkzy&nB#
z-vK*0rB3fyFm}=F?GP-x2t)qrs0N(=ra!t+xMalK>}=e7C+C)kOgicc)SE~7M4vG(
zf*u5^Fl^L_Gb5{#WNI<s_OSzJ5Th>Pb71t}Z~xtn?;>rwOEJsm_z~#f#CP%|VR=zY
z&l+VR^j+`$E{`47w!@7{3mAJ@S&kzzVwdQocnK$H3V4Anh}`m_=e7)l%BIff9c_&c
zI27D-VF#cIFcVL2El^4aqHlYXS(g)%(4JpuPZIX<lJEtWN#{A<04A(gf4LjMxDCNT
ziFn36Am-wS;!HuH97mLo=`8!0+ZU(aBFcvmTwqG`1C_uO6erdYt_9LBLEqd207171
z6Dee({}waS+m$dZ-)nw8+x;4qK7+3gDvxfFP!(@v#dG;cv{>^uhaIVVw~qi{W2j4{
z(FktuIxwOvGxK{%qQ~pQK8m8H<?vU%o70x`jwiLqM^zD8v1l{G3eW|XuB`mope~$x
zO0HAMW8AgyjQAVkBgaUC-SB1X^m1=KD2>>Re1JXrj#X~$jnX-0UFZ#g_i{x)Zji@X
zM7c0RKW)X9=oEc-`1N@Y1HuiI*D-6i=bLA)js>*Od=F>@%7JB;NJ0#RNnyj$CAVNl
zgNV&0@rkQ6f77DetumJvT|t^Pk7QU@!YrKVYu`wc^XhsBNZ?uxb@Mo7SAogCwg$*5
z%C%+m(*^a>(?!#0A6c^K5<%+ABd_>`$96I#zS^pJsdB>Pdh7RCo<Ch{k8~2DRFae?
zvtpD^_>z5hIRR^SAK-lq@Q(M|TJ+v8#d8!`7E(Jii(0^vpsHP+n|xB5wK)0DVr9E-
z=qKV0OH;aT(6qZrkB}Hi(&Gyw)2gOy4~ADKC0RGIs?@3i*0Zov=KKPCD^CO0yGk1G
zGlB?jzthDut+IN?cC+g~8!{_)UB8kzqMH~8lOH8fx{YTOJc3GMnG=;dGhdM|G;{LK
zQoI9tNjc;EWU_lS{UJ$2+&9W71w6c;m-{=hr4@{&o=2se`zK2KN1`c+_xhSPo_+dB
z<Zn@iQTX)6>?2D{yM-M1ovxUIY}`~e$&2HnF|x(Y=~ra~F;y`&Y#5<zT;cP?ykEm`
zK7w}vsMV$T4`?GaV@vVN_|K>BrS1?voI^=E{?Msv{5sQlBHxG(d#C8VWGhVW%94Lr
zFMP^-89x68cbkQW>9higU%CYf{Z<TN(Bh<cw5RPf@c@<&R$U}DG1>FD)h~x-)a+0*
z%u0m3Vbf6AZvC<RyeN2WdCsa*Ak6&7MK=g(PqNdiC#9~oCl*yh{Q*TFJYK8J<{V}w
zmK)S_P-Er)@B{zqRV)pIYm=MsMrBMFcsJW>?=#LBCcXn|(L<O{ToSe=S<v2f#;ciE
zeiZ^?zAe>dSc45L%i3tu1w8bYZ|rg3mgHz{P2}%yizJwWvG$mk^|!+LaRrFGX?ta0
ziQr?#I$69(##nC%SPmRHi}@BWEHfyJZh`SOUxu_t-2onsw)r`pIZ;A?x3Bqr2HuEB
zH4h9(JBYrnrF)h~?gfv%kfYUYw3*29-jNHemE-9T-;=YiTkCBGcT|lr+?^{=5`vPq
z%5?bqW+7Nc$~^rsfOma_1t(k~+gTQbb%@P`Sbzkmq^fD>yG0n2nIlje3(Ud9<p#fg
z>N~aiP$433NC=)g<9eaWPU8F=^RDJ8uBBuW!+2V(6?=JK4!jaBozLK7#z=X4InyCB
zFPrP(wdMx2ubu7uXwMC$a*dnKTyNZ{USaB_zKZppvYNl2TY$*!+@k1*Rfn_pqnv_G
zA+1o2aW~MEJ%HkIpNhirkpl52kdIJq143x6fJ+Mm$s`qVvv1%*8`L1G!aW}Rm|h#c
zc*fO3cwdu&k{+{te5m}>Pk|v`JAJP<tG{cKL5ht+a`p$R0C9!Bm^p+tP3RpeJ#K%W
zkOoUqos&yDu@73p)L`u-+xN7|CaekmBJ6%PVTbRRS!~A>S~$N_@q=WmI`0_W;h$w{
zPhfdbkc?)(DV7mWw%xEJQtbZca|ElsuG^Pj-<6M%{IUFKTO?LZOb}`ur1(XOSHI0R
zsl}nM5LLS1Z_{L&yWAK#<F(8eFLoxJ*bzGgH=o{6fA!ocHR`AlyYeFI%^DmPT96m3
zs0Q=V31d&>IKaY<J`vj?JlW~h*u4XR%4UqE@-Q%-J<afnfe2Jv^hlyxd~0?)dl}E+
zn{b@h_H~>zry0?p(@E&?Wo!tUOZg;uqG|W{5RywXa+j5Hp@NOH<ru@RThoyp%zB(U
zL-jeE<VY9nS72dCC#KNGZgUi6ND49L)3zu!&5T})U@}{mF3;X#x`}mrx4owR$UaQC
zv}@v7IjsJv_oPOSPQE@}=Iodcwtd_a?{7?4PT1Y~X<-tNiHy2_PqzO$qkRPTqPxb=
zM^^42I&|f@>$vLBHrUgV5h5SP@mVQVx79hx<dvoc5UjQq;=L0G<0U@~B+E6aM|M;d
z4av&A9QeCJ>zI@%YTwgMkRFv(hs2-nS;Y823w8V_wMd#BR{6zJRvLfMNysg>4AEk5
zjhv>gpHq<qrTCEs+66Q2cVpvk{$6dXq{f*n5&)1ZMgX}A2Bz8c3*#K_us^woCJNZj
zmiKedm^Y0oBkBlnH!`ZmV_0=-RD$@(VJK9iaYjgS`56k3=Y}WhysV|E9y8%;x6B0N
zLd=Yc&MSj3=&T?n_2=>Zr-$ZtOU^AmnXnfz;$fXYH#6L>Phv{31A-LKE-p!2-y%UO
zGeGv`>T3INUW*WLAValqswdvh;y@e1gtu2Cr9)^=%WRv(2Ml-I_(Q+MW{y#^yUNK^
zR`^5}+=+dymJ?inBg^q-;NJ5EJO>^UqI<2&f+M(A^92^Uovjj^9r_GTJm?P=XhRwk
zn%u{zE^#0<2?nIc>r<sV^9e@t^oUp*taCi?E|gWcU~g2d=?IsZfo*y?1IF{KMUfdB
zV^KE8la|S0ezxJs@4nZjZg;<5<`UH2Xk-o2+1@qr+}8kzk)nZ}nQyCTW7Smi;9Y+N
zy_De7_Hu7(ZrYb%w_5;`cRNypQ`Gc~R!K~eM5m{PcYz1@T8r64R@WTt1gT&OM6s%6
zN^`N*t1bNOMZx->S|ssMGSbLSvT%T%YsdZz%Db&ENjC<yDaCqQvP$PeCW&+mlZdFZ
z-a28fE&hXxFC!(~CHy!<)E-{A@56{nw3iV)Vyl}PWh7$BC3Fdlr1-;w4PPBi`Y4pH
zy~!h7LOPw-rm&b~siw#AH1d5~+E2NpNJU&tVfB=_Ncj%A>b7XkDl<<{h3DHP!UU#i
zq0Aw;1X+2;6XXe4VxO1Kx1w4KCNLNYeigCZ4ROCnemi^FI6huh?DCd;xaacx`seos
zmeMg*F{Yyr2g5O~ASW=xlOkSUWm=<weh4*zI&JFcC%a3!XFF$}>Yg7m;2&KzdpjSi
zgp!Duuwu;lgmH2jv9qT^LNlmz7)Zq;-#VAKl0hl<ZN&7PBHwi$N@r0DTe34EyG>)o
zs@;2ywCPr``!!5N66|s%3ZR98ExmfNBcR{gIZ=7>7s^kv2rNKdY{3WQn+n?>e;kCR
zj6*)U31dNgS4T3bRnIDjk{W(@pLAWjgIKiGJ7XllE-So|RTwYWBC>-y!5l!s>A4PG
z*uhq-2X^&WoPbgOkOOQ8QAhnl@GWDvI3~wDeJ^KV0%a3e0+MC5=m<d~m1)4Yj*AwN
zemL5*%?gnH=}UFspYR3DMkFtPz05Blbf1$jtp~|2lR9SKgpv@;@kYqh;kVZ#=J9&`
zy;u4b8&PQbQC7JWd^S&)_Z9EDNY;<joT!I8k+)+7<^DIltrjEt@cN@wHToBoiPOUZ
zXP|U#C*!nwvQuq}tCBYHWoKOH>LA!8*h#=GgBc!sgPPkFii=9I>DLG$BnM|h5XQ<v
zrwrWa4|cg8Q0bgdUbMdgI%YBf1pw5_VP(i@$wT7OQaBOo@oOS<B@CJ`Lj?9Dk3F>`
z<;Zlj-L>gKzZYbc)TRS%mEW1@<R1;(UBsI`{M}3ZQY2?-ghm?)z`6|%tg<s2tGp3t
zAa=)wP;Y3VK>U*c$E#J^eRD4V0(DR!qF49JqGqR1#nMFMIX$xgL7cI>tIbp@nE-o<
zAq%b-jOQttPeYvLSiat^{#&wqB1h~hcaUwvLVF3GD8C$=)Wx6MA3iU!Gi3Y_m<1x2
zdYW*>M-4;JFT$B?-KY=?MdJwN5FPXao&|{usa512S!HuHX4rgt;^NL3-)9dk9g5(B
zurgaweGlPE*ZF5wmCwGeq(vkZn<GESclnCt$?;l!l7sj8Hht{QA)+OGi0sBT-a(_I
z;}@`_<JH;0tN7l8g$txR9mLWfoH+fOWuhLlRD-r6IV}V~V^8;Mm{yz|=GeeIVh@PD
zp4@tTwKK3t{wSU`o~inE{rgOZAi;(aW`Sd{PD+pqnb@b;iv@%ME`%<@a0L6OGpxyp
znozXTN&?AMI!wLFOFY`8=ad}rzQ*`kWOXkMohCPm=+=s%_)+7_{=wGi$G2~Q>1Kdo
z1BNT<q`ar+)c`e|BtRb{Z=x1MK`&H6f&=K@A0$jmz_kRP==gH)$#5okd);6$CkR}R
z84es~X%38W|4;gH`csNmV8#_{r7kzAtYVlR&Rb#vjS)TsrU&_$nsL^e!cz7s4@TQf
z77|Pexy=^$>fdG87_e>=pnWNhVc+Wz35u@njE?h=`B-;VB8Y*OmT#V698p<?CV@yy
z(rR=tbzAb;&_`hynUsxCuQUJ3VZ`$en#mA+Zn?N;F*5taxhr7T4QxfUiUJ{p&y#x*
zSkkO;^D8f7+41(iymnD-J7jp7(3m3$aO<Ln)nDp&mM!m5j>2&2091J4$)J!d3U)PP
z)!b|)D!w2%X8)R$?C>HjDb`k-As;&{4DRKnbo3IGi|PGS?j3gt*a13S5d4rj*`nlD
zlU_q^8@mep8qzpN3~0vsULCuRID@j42hOrJyjOL+t0O>QGbq9>{F1p5dVjLtD<~rK
zpY`6Q9x2eEy);{_=Vn=s3TZQ_qVRmsB<5Y49gYm$5%`!m!Q6$c2>b2BZzUbB6)RTQ
zmDHa-Pb@RcxsV!@03QN73ICc6#(X-Mb*LW8IVE>4?>Fz0K7{!^9=#J)<ega{({N&N
zNef%TYM}u5|7!sZs}Bpk`;0)dl)fs^n`RC$G&}3_OsGB1;S@P&?-hP5ibuVx2kmd7
zqCDL>Lx!smtoHG7u`}U`HFfW0SL_<j)(<z{0gF5yeG_S30iVzkW+Vr$wJ;J{f4_!y
zkAQD0pQwHHK0HDvb|(&5BkaJ|n5bC7u2-!Ko1fsok<$vdH>%XV+b}xYl)XgHC*5I=
zf|Kz~P&oIT4!VB*u57o&h%_o@Pm{NQ8HH0q|3>_)etu&PrKxF0>e(zgZKcR9&n0#`
zhr%=15G9FpZG{Ca8T_FO^QtCAc-vyp{}Zxxsw`l9-NNN6Mr8bTV^p@BzxVuD@JfK=
zCJx#Yxj+jzu&_W_6mJ)IKI9AT3ORuT>=Nef1nt*%3;!*tnRlD&wAnEFS4Cim9d@VQ
z@HZ(`siBTbEA}qJWWC4&f`PJ}hZ36njP&Ox0@q5bC41P(7M66N^cIc0<KDeI9sTF*
z2Nlw#ohDzs?XE&8E5{yV6x8xev$vxv2Y$DWo*T!reZ0tsVb_4&8TMgolxN2Fs_zmA
z1KD@nr<n91^s#}#P`#S>Lu=t}{6?UX!Vh4)r(LJyfp;e*<rM7G1vKsRW0?Zy3sgQ9
zZpMyz)3fU>^wXl`mp@+o;@5ypddEhcf1i18;@c;|Aawa2-W(dCzr1nwz{d)6#^F0R
zLCID^Qx$9F3gL#^-CzXE!?L8EmjhO2H=IH=6wdtC3YNX%6?&Wl0Rw^E0*e;hHrM@C
zNx)E;kS#3Hz56Vo0)KIjdk8<RU&-`lm_2Arxnb4G+eD{<uFL1}-3T!LyoxTW>s5l2
z;)_PLYTErNBg|a!>@L?U)3S06?r=d;0^`h1{8c#qoyFSsIq{G(WK^7c!r<xfUTcW#
zV`Z;0DEx={!A%+2qkN<fqn7au?<MT0>3*!Cm_1~q`;~LwYpw;k<w(Z33uO>@#L8B^
zc!e-B*jT9=v1I%1?}QyaTaxOg^~K;=G()mIt8l_+9x$Oi!GqncY7g<FeMR++RCi%c
z(8FLd9ilSw3R7c7VhUmUiblbRy*YgqXxH5G={d@nUU8#N8PpJ;Z&~U>an~@QI<X3C
zEV(R3(YUP5Xn5pleqp)&BQ}cHuN0LM^~G_W%HhLR?Z%c38YCuCC_q+h&Oh;&HyviL
zL6Gj^d(sj0T*eW()>=%Bul`Iq<xtvee+U$pKGconp!t_-@eZDPmX`9EG+N6eg@u_O
z&JDjk5MPo?cGBR{jQT99Xy3@9{uZ{$7v9Iske#o;w-<DC{0@J$GDVwsZoiZGEZoE*
z_sm(`deb$SiEFzZn8LHYDoXw<dx>z;3|-*I%<|E;%faLO(J=~43w82r{=E$K1&}cg
zpGB6lZ4|=NCGcF<Dusd@%b%BArz;KTjH^`iZ|;D_3su?HcrKlKF@7~FRgnWf%dg1f
z*cD8r9+{iEo9&X~LiY3b%s94a<Q_A^Q6p5i#;xF>9qYyAk_QwO>#op?IgeIgKqy;a
z_>qJ~hd^_7$IV>HE&PLUyD!+72#Qqix*YQo9TCr}1y@35yR{{OBBq6Ym1Hd6a<VnZ
z&jMbP$(OUu`HMd!NS)?~Q_BCYyPSdm!yeX6N7W=cPGDu!Z^395Kffe2f~rNTQen8G
zj1`Vg*KE54jLVj-(Vh;*e2+mP0;Qfrd?x;evafIK=@9ST_?7t%<Xv@Oq4CHmm1J~+
zd5x1Pk+PT*L)%UzeD{Mbt*@Ak(}|OpVG4>1D@)HNgzQtfQEU?4*mJ&K`EsA^VfPv`
zpSs9B*>^t@Q=_q44Ly(-!fZVK>zHp#iEx8|1fz*@yvalx<LEV)j8mqVe-hKrBgg$x
zLi`XEDA00;$!?X;R8h(UhH;^pukCr)_K#0tM{DY@k|KAM>yH}E+I+aru7v3uqDrbh
z!Tf5jA~g6Z61Su3ExD?A6U-0lm`V*adgvJwcyPMS)XW+y1`Y+4IxAt+lKU^b7AB$g
z<zZD+8blwH4Bh;~EA%Bzl+bPz@^Vz>HK9@?)iR_naYDePkdo<VI8nhXQSSr~r)3pn
z>rjh7dJkJ^0D1YOUJ}8r#G~0aE>k2N)fKgsek1e{fo(Ea6Ijhf1r?Ow?zAtOpx^ge
zsvj68O4oy*pbbg{I~|(SkfV4rD$RMkh^;rE2!i_-QmbXJ{cw{ynH4fh5SQt2mX)#u
z!B0-f+|k`=B)U$QcpIt{s~6fkQJ&6rVp^R7tuN43N+{hfA0_m0HMz5KHkYJG!(L<g
z49unSwMI2!#twFK-_aT>%@}fCjeB-dQO?#;^Q4N1j9)o1zO=FNNCyK#GT7<-Fnq*F
zpJWxjQ#hKsTRr`-X2<Z=d9mcB%BjwgHW1QhUH%e}wl^`9+Hh0#=O{nK)e8x|D(%O-
z?C<^{b+et%R#*xewENM<D4+6_8CmdTOuAx>_ZM@We{#e}JK?#ohiwJ<^t;Te?u5!w
zjO4ohBn+w<Dc#R-Xx~yi85=6y+5G&*_D48ba+s`I1+TO>!2o+5&*SLuNs5A#p<0{w
zuTEwwQe0Zhw4ta^S`HQVU%wpYex?!srRmK4Em6A0!uH-#&}(8!(znPldir?6{;8%!
zzK6W-A~VNDqXl@S%(X~G&a8&-CyF;3`pFd-x*$BJqF>V3B@%7LyTs(`Xcco1DmB;}
z^6rYd0NT*1{*HJzI>LroyP5Rk#-hO`M_OaZhq0wwlOEay5!3sER<*1koq0<A&>(1f
zi1a%N4v+YbSS{>n%zdjA1$e?RrCg<#^vov1(Cn|76Q$K3h;bj@HY?)k_e&Jy#9M-B
z8gI!p+QTIzuJDH2dpzhnK9fX$CnJ)a`lY65j|Q^Q&>m|d5>vQ4dUEGwY<)f4ajm)*
z2ls}<XT-B~=6au(py%%c3w+oP^}VaRa|C(IYYC8j@g`o<WqaMFINF4p?e)QukkdtA
z25_s^xG#T*pCj&E&06R<sH-|P?1-wB-xl=FJt3?~=h!DorW2vH#4*zCaCx~?IYZ_I
zk-po0(eY?eVd?I&@@@{6v&U`WlG0#^oMqS^eX@ZTeaF~XAd}P%^pd|XzTXMMW=TEd
zRM}?RZEV@Tiwk|zL$lj){YRDgTel^??O2$<PYwiWm3i(~PIKkHy)T+R7Bq{q(;|X6
z>2EUMqo`~^Hk+9uUe^UyCdAwT!|CpOQnAs*1XO^m6#z<Wy9OU*6On2!=S8ac{oLLs
zp$1`3tYpZHD-1e_g<eR0wYfNy8$ER9tsL$j76wLN0?jhv+arBk;%|itTs~#PrLJzH
z;ZjS{=}t64?3IMVxD_!4?jx(1q0>>?ppuNNEc!(sxNC*;P)a4kHs<Az0lNw0ees>r
zjzOFCpM|LkcXSmka+(?EBV$at!|tB8r>x`1D4gk~n@d$dW`7qBi#3%-ye-Mv>W*wZ
zw=mmq9Yb12-d0GPF4$s9e!%|PDF;kuNR(b>S~I_0Hy>H^ZsafV{<i2=bwC+pZLeUS
z3ZHX#bsbqsKkhDkEFVeUAmXcuJk}22I#!HgA8j5r;s5Q>*oT>KzaB<4`cvfCL+|Aj
zI_5soJy4<Bco+9Qeh-kTxuV2^h&CjAv~*wIR+P%Y)AF71@rZga?7G>OL&_4K2~=6L
zK?3rkd0no5EE3FH2vr18e-GYfY^W}{>PZ<ADQhIxrG0(5e#bv%2CyhQn%J#5P;`BJ
zVt%K0d;9zDhxMB`_EV$K>=);&Z#}rqA^Ubb`8eEfXNPbQ;3Z(M%av*L$E>w_?CNoH
zvlEllD}nA;tY!$A>#)L_<~)MSqm?%du5VOj=r$LicOhS#7SoV#9r_IHGVKZF_JQ43
zK99<bc&)0cq!{PwKQt2DM7lc;zyyNM7JcV63}AhySP^@C8#cf~rK{@)E2vc(<pv0Y
z2i?WlB%p2$)b<ZMvP;mM<v=B8ds6pT2g_ZEEu%%=Q}N^ZQ{3Ee(f+7cwZl8C96hI6
zGy6S$<~MD={auPzjo-C+wyeZ6nBOV2%?^}6O&8B;te#xz(G^nUe$PZdZ_&3W;QrlE
zbM}AjU3oZ^-`i&l2F0MF>`TZlWesB~lC6|Id)BdsvXAU*B1?7&U%N159eYt|!DNUm
zWnZ!+>wC0(YkGgz`}h0DuWK&X(=+F}_xnEQIdjhE@aK(PdcOK&L&)tY0_MiP8IFS<
zP4H4JL&{%Z6DCpoejrNF<XY0~11|6P-JOjC@j-Punm6T&Vhfj!ItBD`_2oj@xw_{T
zo=Um0*o=m8UVE5TDKTVZqIcu<dK|x8QTrQ(;^7$;wN}HJVQTk9t|jgPV%@v^9+!#7
zlubsGCd-m6?^wO7wPItI+gYl3xxdmL1V^Cs7#X6MY4q}idg<A!%8#p>>7+L}k1+c#
zPNB!i<TqA_sNQ|s7-@fb5w+#B-c3fU>{(wRDD1l$HloWmm?7c+wMb~X+uvKUOXgM~
zYQD$Y&GmB|i)FJg85`4l1z*?0tHFe=5*M}>VyOR7j$Ll|eaUmrdQ9fpNAJ=%LdPYw
zNG%ozzIgh7^$ej>Qc62V>Gf@Qt?w0?lT1>N{I>w1wBxlrUE$?#pE>u7HKL6<C9^fQ
zX}yXId}76tK5vlpdd}rW&3kg*aH0H%!p-+%Md!smW8&8jyWs=>oMMn>ha>Ct1LvSO
zZe6X<5>I{it-ipQMRK(;=-egyvLqGPYl@<2BUV(|9<wdtUTp$}T?O4DN4ii8!&mhE
z*3)HewIoC2!;W^WhqMmTEAnnDSXSTK>+lVq9XSmrF}qUL(a_lNnC$s_<!FW7*9aV$
z`zJ?1nIQ|m4)yijP8PMdZOTH9HA%)kyh46kkM{>>mrbb1r<`Bz^&G5vlr2LR#t6yA
zuR88>A*yVxQOWm?tkl|)icl{X1wZm|EMvh1N`$SQ2(yE)EPZIShboRh-S<0NTVT{D
zn&%~YTgbw5^}yzt?#;VYi57ch&HKYu!I10|tv4&z0-q@v85oWmcZjjo8OYP(FcvGB
zJ~=Lpx@f<7w2m`*qh3%`Yjt?n^wn{Uk~cj^D<7f?9Ww~dJhH2of%!^&E%2b8l<6d?
zpq!YA=hjp+vV;|qHr|c$Lm;bN^wMM0+U-l0h&oYwENc+mbDg^#bSCtdda>1dhl_bd
zh0s|7>ov5*#b8D&lLyZC(Wj#Uya-Z!TIh^s2w@_xHM<GRfg@ZptSMrcwvJF5a+6k}
zRxOydv`F1%`LU(i$xV0H^4?nu8P%G8f=}?rTm?rv(F&J+@Jw^Kk?C#TR4x-X+v__r
zOVH&>vU945kMx;Kg}t!MZiu-9l_q8V_dREg8U&8L`D1c}UGpr<w+CEC4fiIGC<`dC
zq<J4a2q!Zv3Oe6_sf|dyG6xCPiWD=pgAv!7)6OLx?P}{BAQu_p{5(AT0UJ(sG%1U_
z%rN#F*R{5@PjYw|D^q*syxw>@&$2rTDrtn(!EIiLL<>d62XgTGb#{4jpJT{489%EV
z^W5U-!vJ@!gAXiDsbA%fjuLP$f11lX#^_*%%pa2!QsfoZ&D6@gGg(C)9!;c6@gUQ8
z=fKZwI55meFtLUvW$`$5+Dk3Zh4tJosxKz?BBVVn%x*UA`^zS=9gTyt#9U!Z8q%V%
z=MAwjuM&_aoJ?)HeY+;i$o_0y?MvhS@!qDjxGF)9e~MBaYr-NZPPgZFT$a*kFZ%kr
zlydXFIHk3n`F1fgpYhDAe7d@gr&2;Js$MQSOdGW>v}f6QnA?=qZWz*Nb8mF^>F@O}
zwcnJPDouUW(P?%4`L3Vz7Gc_BAtzrug<Mu8Ov=}Tf2==?A7grRpFjTwefq#;HhG0t
zl!3RO=4f>vlwImUU`zF*f&OD+%mSTJT4ai+K4Chs7iFRbdT%&d=z}F6avN5|_g$9Q
z*X1=j?!;f{4K9ClVLM`ff8;!lKWW42v6QQI^Nz_E1=@IG*8Pc2OTnAs2@2wy_vRcX
zq+eNBpMSW2cg&Ic_#)ejV@d`vuSLgZ|L5Y&WPT<kGhxH<K5*&)sT*VGvOtD8s(`c?
zKF+EkFA<y^GG%fnXf&b%KFt~^m7eYd&e}pjd66uEBJCnD8yi~lQzO~%PSzRuEVmF<
ztW#Bi|I&ygZRm+o{bOO_%{!yZQTIdp*aj!Adg1)n2`lW_zUa`E%a0Ya+KxWfBb^xc
zUT(Bz-gC-B1O@RiTDVHNyJ!$spc83NY={UYi>|Y*1u}5Jl}qC3A?KCU^NH5M#`!s{
z2;F4B6)YEF+qVm~2VMTVp|iB0dpXuCOa-qK(eTM!ZS`Kr94Bm{HKTSGCD>t4Hhs4D
z(pc2gq+>*aBGHT`JyVw>@@(s_!iA2eNb{kRCd%UTo;IWkW$Id?zF)v<<dA8(y|;2H
zdobb{>)JktRR7&cmg&h@Ik7#Gd;VLY{sjXK`AJucH`doH*1m85BZ966Wq+(LtGmph
z=+&C5FcV(WJ43v_uW=bxXFb+f5I7^Ce{lxVDBN3ws81-Rrj*y5xb``J`emOW%;;I~
z*~fvF`faB!?CxeDOA8cLaV@bvjvV&Jy@BmU`PI=JNwJHInVSc>oezU=Z?}`9-G}8<
zB`Pj#RIa&R>zY%$ZB-M>X*aarS2JC7u%B|0F5)sg&tID3c1v93KEK?}Yc6#(9B<!S
z?=q>kPfZ93_AGums{TNS8cCgbLP6ZPo?P%)-aDohx+Lg?SkE{$hWMnCfzg5_%Nd3X
zH1*0+TY3LP$_*$_$>C^_1NjF1sgm1;H)wc;H8Pq@9ru}(uatO&Rv_sW*21ly7RMOS
zW)OB+?!L7!Y2h4cjkEvyct7;DPz^3jlQI7CEh>~`WuQee7tLEvr&(mj$*LkInUJF_
zl0vkf{c^~7Ju{dQ>D2CxpRT_P?x#c@XZ&(1k3h)!q@)LR#T~WAiA1ovz6@@eZ|rN5
z;kzd7-fNe+!jWJ0fGg>3qGyfgM1<uHo8E$*n$=jdjl``CXx$Y82&EAg-kKKc?^&b!
z;e%AWio`x^hQ4n0@#ONCXI4IbrO$j1hix%T@EZ>h=e+Cl-*@9EWLYN>Y=7bT;jXgZ
z!Oj>%+)-ScKihqSxMR3r1j%Qic8U2n#Y<El6svX)L18lew*R?%jG5R$!3RxNy5I#L
z8*d0I(C5<-RFZbSXNrtB)OWP$k&3k6uxP#o=VRNI0`r?s;jHs6?dlZBoh}ytkxx@<
zy`M&yhqpR}{LHpLynx7mv6kCOagpIjES*LMyM9KO@WedxsQZU)OW|0@TLbv(ueh;(
zW{DQtD7UUT5sL$i>|~m!v<X@eYswO8$gp#*WQD|rLBw5ddxk?)*@j4HUZTc99?@0H
zmV=Yp5*n^jaN_>v`Pd<%dNixo;c$C8m`^OwRt~<zS)*Wc@6pBcgv}Yd9E5bt)v?hX
zx4tY?MP(+xQ`TWxCAKd$(hrM|TGfuHVTw+fG86JkzER=ssfK`MD4xQJxtm>zw__TO
z0fz|;2C?;P5x+S3YJYo_)_t%^*u^=W*jKG^XJaeX`rU@K@sI<#t}yiD>q*+PJFhq<
zHw6lK>g>I{Zz<et;Wm*gdEwNlmQ&?*T2h8z?pyO%yf@2O!ZX9ZSK*;&>c~6Aq(lYZ
z>c(_W%_jJ)jmeGeBV%WBHYOh{x_)bJ6zn^1ytq&kd&{@!0naN(h8!ogo9M~>Ypx6K
z(PQV)w4$E%F)>uDcRf}2nxuJGHDlq<5mf_}WU)ugJjFNYvpp3U>x(*tZV)bI5qG5p
z$;Pn6#M&W0r#>oPPGsiuFyU|OIC(OoA5N4+3^`*|Ge9hI-M2|qPEA25h#O(TL>xP9
zsLV4`hV^JPAMTgRJ1TQ&i>TV^_7lM-KDNcaxdo;db`PptDa9Xau|&<)m^jrxoHm^s
zF}@Ln-P0W`nfDOkHuUY%gLBl@neM(wkV-q3(9->u<<`b~henqhzDM5Xvz?1sSc#+H
zI!4G?IY4Yin^(5$G*`Z7fP1EYSpw~DbkViEAf!lTZu@e2{lI&}Q>nFIl*0OAzhQe|
z<}2Yf`=07C|HLlad7rLzt%=b5%$B!!l%>t7wdhpMruf|cz{~ga?g$IivO{TIJ=jfn
z^RzIGU7%@5Lxj@bVq}202PsornOTJWCfSQio0q~J9=mb3iZM1+ajKGC0D9%~D#^U!
zLJ)=`R3PZJ-Bp^>Yp@h>wrL)9pB-70MyO4rY_|W_ksLb5K?TOj93EN4did2?CTT0E
z&pOM}#bfKw617?8vzDDHEQ-VQDXjA<F}IFnQ!5A<SC15<_h;YcNB7V;TDIS;CL2z(
zXdSzgZMsu^N1`THn9%uH8<{;*(}n^ocB3C0;?nHIh;fdbh}*ZzrVbQqw+oBYkiR`N
zKHuQ-rE16au}c<n=NDc74P7oM(H-5&PDZohaXr4fjGe*LGvbc$>ycP=J(fyX7oL@V
zc|5JHB{^<}(u}=7MpTV%JHq$4rW4yS;wGX^D}ql$-<S}oa6b`0SVckP-L9QEuVmVV
zdS{}^OE4v$ce6vzwz4wM7Y?Fsr87l+KN`SPi)?(AhM~(^q?UKBWt98$z3wORlb7C<
z?S^!=sOvJ*Dw-GDoV8ORIj61q?LoMiVgJ_CTaB~~>Z}ZRElsXg8}Ec&jqHb9jMg-~
z{!yxU)swn4`@L-To<g2!TC)Uod-%KlT&^R-$}=TH-2CE{qw1E$W8E(cqgdaOc#>7@
zRQK=oX-wsvHZJ50Z6ot*&ImfMxLw&N8$Y(cS)<-j+(E{#@${oPDjuB}|0G3LVt#)+
z^`d^yg1V1q|E8CNjKuYissf(|rmynObOdk-T}Q#$sEYfC?R}S$_7f7s%{xxK>xj6E
zFdBBg&ZMQ`($r!~y-5%#H1axGG(M;#e4{8;$WHp3dX?_EE-P5utM&Z4#<n~DKK$%*
z&E?`fJ7&mK+{^_+iI*#>0bRkVXW(wEQJBjVA{`ONud1#V2?m|-Ca->euI5M}of6MH
zZ`G=0(Z!-q1|>Ifb#xym&VSC}mDTCuDrvOZdm<5A{i0<>JV<|p)p5vA$G<0i54wi3
zm*e~l5g#FNdU&Hq1|4&4B?RN7zswz?>qI;M{#)yqgzWeyg3iI0Y4XsT?DOSi=DuO;
z=~$<tuKu3omZ!(>LDK=0d;L!+_W(8mU3*JCTB@0%hZ9t6Z2iXKBl5%Rickk_!xkH@
z2d_7ct}Cj*s9J*amIM<~oB{C`Mk<J`j-m`cbJmZwjOUE%oSg<;o$HW+1C%P5GL0c;
zB({OlV0=89vy-xJJ7WXVQVu<L@)QNxoj3wN+G3T+u7Zz2$V*|X6UmLk)-VS#LA(5}
zfIUc{L1k(G-Zrr~XIku?4egc&%;?A!;$pUPTyiPtEJF-MEnENGD<4M9JEeE;pTznz
zMO5#seat@`#r*gHa2^SlP?a4pL8YC$(H>Rt9hIkF60*5!dSTP?;;=#NQ_oUT<PTWp
z@hpJm#GJr5^b^bPaQ;(}7SmCC+_vmyy_M4NoD%%z61%wOck<3*P~grLd2q^;=!;zV
zy#)3}HVuLwhfg>F2xjzA_<QJYFbGC2lwI5j$Ki^KO_@#KWQ&hz@GCy}7;qEBorty1
zl<^+qEKx9LiKr@ZyG8MRAa)e|fLbq~j{AkP1;inO2dG41O`ZFUpSgBO*q(XAOu(u8
zeeaYk8zK&^$@QQOOoOo!k{%+|G5z3h_!pe0hsEAgfe?cE!{Ihng*THn*zX{t^eTr~
zBkp@IQ$$_&w%sRlFaE@7VI^D!_)YVBl2fpM$!w<)JY6_Ec-j@5v=ve1D?7ya9p>yD
zd%y|_ouJWuuo(Zd>qI+QL{)Uas@vg=+zLIAggI5%Bk3pfl8WgH%n0DJt$fD<IyCX@
ztZhr=Pn;JyvaKCAz5jrq4m$N;pF5EN4#Mdbo%er2>-g#T{`^B=uGw~1z2CtAKsjYi
z@(rx&!7T?i%(9K4@KfqT84o88Nr0HYMh`z21%#3X5&31C&VMAc9NKvlVA7n%CO0tL
zlrhE3kQOTklM&-spad&#0T4qncFrZv+(F#>KJLdn<aeS}G%A;NA@X5BwXCnOAc*Wq
z^0l*NU#;@|y^kF^>vZ>^S5O(exAyY`{AoXbfKyXaGK!4V&-`U5;5GoB#ueH|lk3*E
zw)UKGof~EIocpojvA%6L4{x{YaVoju(w)ZN_e#^jpb$da4XzU%!-8sz;(y?R8*@2Z
za`KC~FTFOLpVqe=+leDs%I3_GPf-incDci-9ZM6ShbxRMRu1iS{7y<6I$jv8b)^D*
z23f53E0Ye<4tRl}`P`3bD6f?|JUl(bjq;$OkAO~X-O*!b^Ku=2@z7yWlR)sP&AB@N
z(|+GE(ufcQU2PsCn2EA2f(3rYUm$(Jl~t*jl<V(mzn&YS2XIUXDmcu(c$kBOO;Wa7
z`Hi=!6iF$CqM{-V`Rms*&a{1m0Mi0Vi=icf+qHdk?zw))YxJyw*uQCxfZS6iXzjP>
z4z+BBp>Ja;KQLpcztRwL9+y6QDlt;jg=}t&pVJu&WxaNUimGOJX4&j(D(?I58y;|s
z2wamMplZnT1nE!!yd5%E-=9GNyhyJFX*<QRw)mE8@>Arj#ZXD+BT60>%2X(&3c=UQ
z@a^$+C8bg(ma_l?2~h)My)^m!d6>&Xkr0O(D-`ZDCFD5=0RgSqWhMFRU<sNU1a3vW
z&3Yn|EJN64s3O)@fy(W~kL77UqfitqQY*`3N+cB${bIw$m-;McD)QEu#O|B}3y|bC
zqJ(k}k_x7|au1TCNTw3VtGl|`Z@)vQh^9)@6Y^mJJkblQN`RQ0Yb2OT_XFjMf-vl<
zSPDI3HO#-CQn_SWu>x(fc`WAM?;xvFB4DEy)aaTDBEQt9>-MxNBPkkr`Agy8moE)h
z;#*&Bo+hjvYCFXvb#kMpHiR`rH6&TBSjc)9cP}t3PK7h-Eq_waW!cOm^9q%gLEPK!
z55QMEgs`VNxfjzM8-}Oz+#z{-J)_aTIjUp{K+VmX(^8@y{gCoLIKr?zmPE1Y5PJ!j
z!dVj>Uol!~{s+kH4+H)AcjmCNFe&zg{0nIIZ=L#mfsp}?mh>yQewUAb>&U+sU}m;s
zI41j6+wWTOM`-=LIi3Nk&evkSKTuA8v-w`h!GEPLf&J-S{ZGw)N6?=l|F{O)nU{F$
z93F)72b%+bZ_lyR&WaAgJ13qC)J88>jU8~H!)b3?3^W*J(eV}1HDINCQqoEJ{Go{{
zZW;G1r}JMS0tg5pP#ByQ5r7aQG;<I$pr8fU|K9v{kyRg1j;?>mM|Rlq*UeuS_tg=4
zsnLV2Kj{4R?TjF6@uK6j>Q65C-&)eb2dy4g3;&V<@7p>Gn0>wBvj=!Lzl#d)yoUx=
zzqb)k`z6DlQlXY0t;0(?rJR4>{KewGYO*5!|9TT~6q1RqT+bY3q53;uFv`S1)hx%?
z{UFdn+GI5@Td8#BZDXY0WzYFO2yRZ6_!JkuzxP1=?1)#h6GAxXw$^90AN73>#dx68
z^Kctw5L{fJkvD$8n!*SVJUwBzyy2|Eyr{kO>YZ|rgF9Ycs<!ErsKxzN;k~$hN1exW
zx}x~c1z?0hyqTpfDg<D{ppWt-M{e+yGriWTR=^KV2n=RVbVQyE3QOQ-Ef;XI;{8kD
z*FeG7+_|z4Xe)WcajYMXDSTx?C8&H&oBmhy{>X(eF4p2`G3pICt5!k)23_ECCJ{eL
zf=<8(LzrVHU~nFJgsEX`!aEgwStfvOCi_%^p|E4DKo(OK)xVOZ1<2AuntK5P?WTrf
z6GizR(%@x2Dh%ixrhS!`5Yj=uK!kgbO#HO=Eju(nKR;pc%xI~<%<gDX%yDUhVElT5
zS^&>2w2OftyhMm{h}Iqw#n){sIiP&g!taB600Al<0>Q<fGsa^KS{s1e7vRZY2Q|wC
z;(%IA=b+x6sHViHL;o<K+~fM>yCZ}UYGNpwIzNXCKce882t5WTE?#zoX%OH;O?l%A
zzF)JfK|$=xO5{KUV9pSt7SEjQ!`Fon;!S8)(nkmwlN^I%DZ?E2@L6a9UC`yIGf4!f
zFT{S7$fTZIs;LP}gUw8*cEtALGE#4RQ=h|k3=CeYhtTsDNf8H?c6<^D7Rz9O!uM_z
zsJGJS@VQU|lsUAQ7;WO|i7g#5GknQ^Tos?%HXJ^<nQe0vh}EDy?qj;+ONT8Td-k-3
z`Y)S4A%odpNlo@5gVU!6U_5Sf48ZZtc_#_D<tFlj5^#$ju>lcoK+Z&J(|wK-=T}{F
z37@MuIn3Uo_u>sG>^e_56_PHpBYSPyHRTQ06K}BbEfXq$XGTN`hzJAjGA9UBfm|(L
z8|Co)1y2ON##40g!B667A|R6>4IFFSK5dR~Y2j0ZW{WnL{-K6IJ2_d)nckk;_$&MB
zfQI#w&khOSqkthvxog<)HLNWJ8W!$~aRYbXMQvg>*FLKIt2gJnSIRaA=>FC50mBDX
zcSNaXLt*cDSER~Eb9WwM(1U&Vn(!K$HUQmLdv&a#uqYLTsdSE+T`<08X~07n0juYL
zEIDLVz?a5L<-S;^@bYk-r(VB0yPt81q5w?kOrIqH5<@4pJWcE0BOhQkhqWu@@Ku>n
zM(E*UAKw7a3nv423fnQ@I~*ks()*q(E(%cEgQZa*57K0Fqd^*>(B$}}=(Rwsv7$==
zss4OSA8K)O(n1xt?#da6L2KID|3%ge2-qfpJJeAhI!uJawHZTj{8J}R<5%Y2oj1x1
zYD|j8t?$eopcavyoUKb;_wWq_4-*m&H_0JCwClJ4%EHv!%`F}jS)(Q@<bmGklNbN#
ziE)#aRqI>@5F@u~BuLPa9XY4V_qZ{ORre$RwrLTl%4MyN+Mw+eHI@{;Q*UoFX6D^o
z->hSrsytfwS5066H9;;$hYi#odemF6$sE~s<TBe|jV=Q72Tq&;BohRk8`=irU{5r!
z?cp`jiU*}3&vyPH0jdx>9II4|^PNqWMiy<e+Thpn2x|D?2~pl)P=3c<QIm17-txwg
zUTJ(aF#RCTvO5-)L3W~uH8Rb-KVRu0$p~KjtC=Lo5=o{q;5AcD$f7k?ipGp{a)1QS
z<xxt&;}7KR4*lRnRt0FyAHmg(H)!A|ISCDXkP_;^0F+dMOZajI3Nb|rvz*5pA0kE}
zm36ow89;xIR^o$M{j$$9qr63rR6mnZX4d^QncP?Z+k5oJnDpwlR9qb3uS*9gw=P}j
zLi?RJ@<ov2RV@@R@or>8cZK~U(DDouu;H_3m(|}gs%&~<_bX0dw|(%0Awbz5x;ooY
z6Nn6@mbe>If)RTjlpv3%*LhGT>>r_xLAdp^_j_=*CJlUUV?t{yvsZ=P{ktkruMk_@
zH7Wr`7nK8z@x(4g`uX^%*`~gcEgzNJe<YjhK6#}p;#7e-UQx(W0WlP3p?#S1+jMa3
zrjy-8AkrVgI){T&L?bW+d~P}gkt0R-OrLcSWAMbr_<R}$-O=w1K}i6|OH^u+0a^;M
zG%95Ca*UE*Xn{609zldD611yOzb*%E&wwETjwt8!pBzEW3kEGHT8x03Q3lXq2yQ1y
zh{-L~K_HU|yw3i?6XOgTek@)*><K2I`%Ks+$_n)^%;x;Mw}B*0c5Mo8@^AToSQDT(
znC!iD3vh<9^)qAlJqpxK8O)U;{v=9(iUZD3KC9vY%)pwDna7ANE|nk0emYkp259^i
z^&K~Baf#C>DKK=x)I*`@W%ISa`t^Q&kbNoLU}yseLz|70cx(tE-$04_hAaUu4ydc-
zFr>1ijyhfuLl6M(1hDZO_PI|EE3#*Axqw5|99{`SnF1S_5?wtUbEs0=97PUw2{G-&
z>AWgT(0O<H5C4333(Q_IO8(FxJlf+%raKPX95NKh9>(HAys#MO1vHDwafj7?l3L<8
zRs*fRy*y=mA$rbI7&Ys07Hcd9hCKh;7QqPw&9yB!EVjYqK+v!AeYOYxt9;IIjSoI_
z)F?fC6eSMbL@oNrYDRXjt8@7^HndFN`qQ{@))iHdd}S&Zz9+@<@#I%lLS~C=Yir&~
ztFjOwi2B?9Kh23a4|(PBwII(h8ZSbAiO5<IaFBoMws7~KV*mOwOAv^Eb8S!n&%HmB
z;IMm=Cw<J1FWe91=Rn~ZJnuC6pIE}LgE8HI8UEL#e@5wH#QonvGwxqKIW#xJr!htV
O{$0MPp;W418u&k<U(>4q

literal 0
HcmV?d00001

diff --git a/docs/source/whatsnew.rst b/docs/source/whatsnew.rst
index a12dbe6959..b1f6b2dac7 100644
--- a/docs/source/whatsnew.rst
+++ b/docs/source/whatsnew.rst
@@ -6,6 +6,7 @@ What's New
 .. toctree::
    :maxdepth: 1
 
+   whatsnew_1_4.md
    whatsnew_1_3.md
    whatsnew_1_2.md
    whatsnew_1_1.md
diff --git a/docs/source/whatsnew_1_3.md b/docs/source/whatsnew_1_3.md
index c4b14810b5..6480547eec 100644
--- a/docs/source/whatsnew_1_3.md
+++ b/docs/source/whatsnew_1_3.md
@@ -1,4 +1,4 @@
-# What's new in 1.3 🎉🎉
+# What's new in 1.3
 
 - Bundle usability enhancements
 - Integrating MONAI Generative into MONAI core
diff --git a/docs/source/whatsnew_1_4.md b/docs/source/whatsnew_1_4.md
new file mode 100644
index 0000000000..0fc82ff820
--- /dev/null
+++ b/docs/source/whatsnew_1_4.md
@@ -0,0 +1,68 @@
+# What's new in 1.4 🎉🎉
+
+- MAISI: state-of-the-art 3D Latent Diffusion Model
+- VISTA-3D: interactive foundation model for segmenting and anotating human anatomies
+- VISTA-2D: cell segmentation pipeline
+- Integrating MONAI Generative into MONAI core
+- Lazy TensorRT export via `trt_compile`
+- Geometric Data Support
+
+
+## MAISI: state-of-the-art 3D Latent Diffusion Model
+
+![maisi](../images/maisi_train.png)
+
+MAISI (Medical AI for Synthetic Imaging) is a state-of-the-art three-dimensional (3D) Latent Diffusion Model designed for generating high-quality synthetic CT images with or without anatomical annotations. This AI model excels in data augmentation and creating realistic medical imaging data to supplement limited datasets due to privacy concerns or rare conditions. It can also significantly enhance the performance of other medical imaging AI models by generating diverse and realistic training data.
+
+A tutorial for generating large CT images accompanied by corresponding segmentation masks using MAISI is provided within
+[`project-monai/tutorials`](https://github.com/Project-MONAI/tutorials/blob/main/generation/maisi).
+It contains the following features:
+- A foundation Variational Auto-Encoder (VAE) model for latent feature compression that works for both CT and MRI with flexible volume size and voxel size
+- A foundation Diffusion model that can generate large CT volumes up to 512 × 512 × 768 size, with flexible volume size and voxel size
+- A ControlNet to generate image/mask pairs that can improve downstream tasks, with controllable organ/tumor size
+
+## VISTA-3D: state-of-the-art 3D Latent Diffusion Model
+
+![vista-3d](../images/vista3d.png)
+
+VISTA-3D is a specialized interactive foundation model for 3D medical imaging. It excels in providing accurate and adaptable segmentation analysis across anatomies and modalities. Utilizing a multi-head architecture, VISTA-3D adapts to varying conditions and anatomical areas, helping guide users' annotation workflow.
+
+A tutorial showing how to finetune VISTA-3D on spleen dataset is provided within
+[`project-monai/tutorials`](https://github.com/Project-MONAI/tutorials/blob/main/vista_3d).
+It supports three core workflows:
+- Segment everything: Enables whole body exploration, crucial for understanding complex diseases affecting multiple organs and for holistic treatment planning.
+- Segment using class: Provides detailed sectional views based on specific classes, essential for targeted disease analysis or organ mapping, such as tumor identification in critical organs.
+- Segment point prompts: Enhances segmentation precision through user-directed, click-based selection. This interactive approach accelerates the creation of accurate ground-truth data, essential in medical imaging analysis.
+
+## VISTA-2D: cell segmentation pipeline
+
+![vista-2d](../images/vista2d.png)
+
+VISTA-2D is a comprehensive training and inference pipeline for cell segmentation in imaging applications. For more information, refer to this [Blog](https://developer.nvidia.com/blog/advancing-cell-segmentation-and-morphology-analysis-with-nvidia-ai-foundation-model-vista-2d/)
+
+Key features of the model include:
+- A robust deep learning algorithm utilizing transformers
+- Foundational model as compared to specialist models
+- Supports a wide variety of datasets and file formats
+- Capable of handling multiple imaging modalities
+- Multi-GPU and multinode training support
+
+A tutorial demonstrating how to train a cell segmentation model using the MONAI framework on the Cellpose dataset can be found in [`project-monai/tutorials`](https://github.com/Project-MONAI/tutorials/blob/main/vista_2d).
+
+## Integrating MONAI Generative into MONAI Core
+
+Key modules originally developed in the [MONAI GenerativeModels](https://github.com/Project-MONAI/GenerativeModels) repository have been integrated into the core MONAI codebase. This integration ensures consistent maintenance and streamlined release of essential components for generative AI. In this version, all utilities, networks, diffusion schedulers, inferers, and engines have been migrated into the core codebase. Special care has been taken to ensure saved weights from models trained using GenerativeModels can be loaded into those now integrated into core.
+
+Additionally, several tutorials have been ported and are available within [`project-monai/tutorials`](https://github.com/Project-MONAI/tutorials/blob/main/generation)
+
+## Lazy TensorRT export via `trt_compile`
+This release expands TensorRT optimization options for MONAI bundles with `trt_compile` API.
+The existing `trt_export` API requires the user to run a separate export script to prepare a TensorRT engine-based TorchScript model.
+`trt_compile` builds and saves a TensorRT engine the first time a bundle is run and provides limited dependency support.
+It also allows partial TensorRT export where only a certain submodule is being optimized, which improves usability.
+A few bundles in the MONAI model zoo, like the new [VISTA-3D](https://github.com/Project-MONAI/model-zoo/tree/dev/models/vista3d)
+and [VISTA-2D](https://github.com/Project-MONAI/model-zoo/tree/dev/models/vista2d) bundles, already come with `trt_inference.json` config files which use `trt_compile`.
+
+## Geometric Data Support
+
+MONAI introduces support for geometric data transformations as a key feature. As a starting point, ApplyTransformToPoints transform is added to facilitate matrix operations on points, enabling flexible and efficient handling of geometric transformations. Alongside this, the framework now supports conversions between boxes and points, providing seamless interoperability within detection pipelines. These updates have been integrated into existing pipelines, such as the [detection tutorial](https://github.com/Project-MONAI/tutorials/blob/main/detection) and the [3D registration workflow](https://github.com/Project-MONAI/tutorials/blob/main/3d_registration/learn2reg_nlst_paired_lung_ct.ipynb), leveraging the latest APIs for improved functionality.

From 5002fd9305c2440c1ec4c0f7a55bce34e1030d00 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 14 Oct 2024 18:36:50 +0800
Subject: [PATCH 165/183] Update citation (#8148)

Part of #8110

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 .github/workflows/weekly-preview.yml     | 2 +-
 CHANGELOG.md                             | 2 +-
 CITATION.cff                             | 4 ++--
 monai/transforms/regularization/array.py | 5 +++++
 4 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/weekly-preview.yml b/.github/workflows/weekly-preview.yml
index 8d8cccffad..f89e0a11c4 100644
--- a/.github/workflows/weekly-preview.yml
+++ b/.github/workflows/weekly-preview.yml
@@ -66,7 +66,7 @@ jobs:
         export YEAR_WEEK=$(date +'%y%U')
         echo "Year week for tag is ${YEAR_WEEK}"
         if ! [[ $YEAR_WEEK =~ ^[0-9]{4}$ ]] ; then echo "Wrong 'year week' format.  Should be 4 digits."; exit 1 ; fi
-        git tag "1.4.dev${YEAR_WEEK}"
+        git tag "1.5.dev${YEAR_WEEK}"
         git log -1
         git tag --list
         python setup.py sdist bdist_wheel
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 53f1049449..e073eac56a 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,7 +5,7 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ## [Unreleased]
 
-## [1.4.0] - 2024-10-15
+## [1.4.0] - 2024-10-14
 ## What's Changed
 ### Added
 * Implemented Conjugate Gradient Solver to generate confidence maps. (#7876)
diff --git a/CITATION.cff b/CITATION.cff
index b535a77a9f..3cd3d0e0b1 100644
--- a/CITATION.cff
+++ b/CITATION.cff
@@ -6,8 +6,8 @@ title: "MONAI: Medical Open Network for AI"
 abstract: "AI Toolkit for Healthcare Imaging"
 authors:
   - name: "MONAI Consortium"
-date-released: 2024-06-26
-version: "1.3.2"
+date-released: 2024-10-14
+version: "1.4.0"
 identifiers:
   - description: "This DOI represents all versions of MONAI, and will always resolve to the latest one."
     type: doi
diff --git a/monai/transforms/regularization/array.py b/monai/transforms/regularization/array.py
index 4bf6cff649..66a5116c1a 100644
--- a/monai/transforms/regularization/array.py
+++ b/monai/transforms/regularization/array.py
@@ -112,6 +112,11 @@ class CutMix(Mixer):
         the mixing weight but also the size of the random rectangles used during for mixing.
         Please refer to the paper for details.
 
+        Please note that there is a change in behavior starting from version 1.4.0. In the previous
+        implementation, the transform would generate a different label each time it was called.
+        To ensure determinism, the new implementation will now generate the same label for
+        the same input image when using the same operation.
+
         The most common use case is something close to:
 
     .. code-block:: python

From 4a4c25129f533e275764ee41a0303d5c5dec5b63 Mon Sep 17 00:00:00 2001
From: Boris Fomitchev <borisfom@users.noreply.github.com>
Date: Mon, 14 Oct 2024 22:05:32 -0700
Subject: [PATCH 166/183] Removed CPU randn() from schedulers (#8145)

Fixes performance issued due to extra CPU/GPU sync:
https://nvbugswb.nvidia.com/NvBugs5/SWBug.aspx?bugid=4904446&cmtNo=

---------

Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com>
Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/schedulers/ddim.py | 2 +-
 monai/networks/schedulers/ddpm.py | 8 ++++++--
 requirements-dev.txt              | 2 +-
 3 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/monai/networks/schedulers/ddim.py b/monai/networks/schedulers/ddim.py
index 2a0121d063..50a680336d 100644
--- a/monai/networks/schedulers/ddim.py
+++ b/monai/networks/schedulers/ddim.py
@@ -220,7 +220,7 @@ def step(
         if eta > 0:
             # randn_like does not support generator https://github.com/pytorch/pytorch/issues/27072
             device: torch.device = torch.device(model_output.device if torch.is_tensor(model_output) else "cpu")
-            noise = torch.randn(model_output.shape, dtype=model_output.dtype, generator=generator).to(device)
+            noise = torch.randn(model_output.shape, dtype=model_output.dtype, generator=generator, device=device)
             variance = self._get_variance(timestep, prev_timestep) ** 0.5 * eta * noise
 
             pred_prev_sample = pred_prev_sample + variance
diff --git a/monai/networks/schedulers/ddpm.py b/monai/networks/schedulers/ddpm.py
index 93ad833031..d64e11d379 100644
--- a/monai/networks/schedulers/ddpm.py
+++ b/monai/networks/schedulers/ddpm.py
@@ -241,8 +241,12 @@ def step(
         variance = 0
         if timestep > 0:
             noise = torch.randn(
-                model_output.size(), dtype=model_output.dtype, layout=model_output.layout, generator=generator
-            ).to(model_output.device)
+                model_output.size(),
+                dtype=model_output.dtype,
+                layout=model_output.layout,
+                generator=generator,
+                device=model_output.device,
+            )
             variance = (self._get_variance(timestep, predicted_variance=predicted_variance) ** 0.5) * noise
 
         pred_prev_sample = pred_prev_sample + variance
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 6d0ccd378a..72654d3534 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -22,7 +22,7 @@ isort>=5.1
 ruff
 pytype>=2020.6.1; platform_system != "Windows"
 types-setuptools
-mypy>=1.5.0
+mypy>=1.5.0, <1.12.0
 ninja
 torchvision
 psutil

From 46a5272196a6c2590ca2589029eed8e4d56ff008 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Thu, 17 Oct 2024 01:10:16 +0800
Subject: [PATCH 167/183] Update release data (#8155)

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 CHANGELOG.md | 2 +-
 CITATION.cff | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index e073eac56a..ffd773727f 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,7 +5,7 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ## [Unreleased]
 
-## [1.4.0] - 2024-10-14
+## [1.4.0] - 2024-10-17
 ## What's Changed
 ### Added
 * Implemented Conjugate Gradient Solver to generate confidence maps. (#7876)
diff --git a/CITATION.cff b/CITATION.cff
index 3cd3d0e0b1..86b147ce84 100644
--- a/CITATION.cff
+++ b/CITATION.cff
@@ -6,7 +6,7 @@ title: "MONAI: Medical Open Network for AI"
 abstract: "AI Toolkit for Healthcare Imaging"
 authors:
   - name: "MONAI Consortium"
-date-released: 2024-10-14
+date-released: 2024-10-17
 version: "1.4.0"
 identifiers:
   - description: "This DOI represents all versions of MONAI, and will always resolve to the latest one."

From a14a9d77165b11996c023a9bf8e649c3c5d6bbbc Mon Sep 17 00:00:00 2001
From: Smoothengineer <160827599+Smoothengineer@users.noreply.github.com>
Date: Fri, 18 Oct 2024 08:25:15 +0530
Subject: [PATCH 168/183] [DOC] Update README.md (#8157)

Fixes #8157 .

### Description

Fixed some typos in README.md

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: Smoothengineer <160827599+Smoothengineer@users.noreply.github.com>
---
 README.md | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/README.md b/README.md
index 498d3c6149..e5607ccb02 100644
--- a/README.md
+++ b/README.md
@@ -16,11 +16,11 @@
 [![codecov](https://codecov.io/gh/Project-MONAI/MONAI/branch/dev/graph/badge.svg?token=6FTC7U1JJ4)](https://codecov.io/gh/Project-MONAI/MONAI)
 [![monai Downloads Last Month](https://assets.piptrends.com/get-last-month-downloads-badge/monai.svg 'monai Downloads Last Month by pip Trends')](https://piptrends.com/package/monai)
 
-MONAI is a [PyTorch](https://pytorch.org/)-based, [open-source](https://github.com/Project-MONAI/MONAI/blob/dev/LICENSE) framework for deep learning in healthcare imaging, part of [PyTorch Ecosystem](https://pytorch.org/ecosystem/).
-Its ambitions are:
-- developing a community of academic, industrial and clinical researchers collaborating on a common foundation;
-- creating state-of-the-art, end-to-end training workflows for healthcare imaging;
-- providing researchers with the optimized and standardized way to create and evaluate deep learning models.
+MONAI is a [PyTorch](https://pytorch.org/)-based, [open-source](https://github.com/Project-MONAI/MONAI/blob/dev/LICENSE) framework for deep learning in healthcare imaging, part of the [PyTorch Ecosystem](https://pytorch.org/ecosystem/).
+Its ambitions are as follows:
+- Developing a community of academic, industrial and clinical researchers collaborating on a common foundation;
+- Creating state-of-the-art, end-to-end training workflows for healthcare imaging;
+- Providing researchers with the optimized and standardized way to create and evaluate deep learning models.
 
 
 ## Features

From 79f9ad9083808953cf7ffb01e592c4e77b3336ab Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 22 Oct 2024 11:56:23 +0800
Subject: [PATCH 169/183] Fix unittest (#8161)

Fixes #8160

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/utils.py       | 2 +-
 tests/test_bundle_download.py | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/monai/networks/utils.py b/monai/networks/utils.py
index d0150b4e5b..295a055390 100644
--- a/monai/networks/utils.py
+++ b/monai/networks/utils.py
@@ -697,7 +697,7 @@ def convert_to_onnx(
         torch.onnx.export(
             mode_to_export,
             onnx_inputs,
-            f=f,
+            f=f,  # type: ignore[arg-type]
             input_names=input_names,
             output_names=output_names,
             dynamic_axes=dynamic_axes,
diff --git a/tests/test_bundle_download.py b/tests/test_bundle_download.py
index 02a9f40846..399c61b117 100644
--- a/tests/test_bundle_download.py
+++ b/tests/test_bundle_download.py
@@ -250,7 +250,7 @@ def test_download_monaihosting(self, mock_get_versions):
                 mock_logger.warning.assert_called_once()
 
     @skip_if_quick
-    @patch("monai.bundle.scripts.get_versions", return_value={"version": "1.2"})
+    @patch("monai.bundle.scripts.get_versions", return_value={"version": "1.3"})
     def test_download_ngc(self, mock_get_versions):
         """Test checking MONAI version from a metadata file."""
         with patch("monai.bundle.scripts.logger") as mock_logger:

From 35b38946e5468acb3b2c2e8a10a06653d1d43c6a Mon Sep 17 00:00:00 2001
From: ANUSHKA KATHARE <anushkakathare07@gmail.com>
Date: Tue, 22 Oct 2024 11:13:20 +0530
Subject: [PATCH 170/183] added a top button because of file length (#8163)

Fixes # .

### Description

Added a top button because of file length as it improves readibility

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: ANUSHKA KATHARE <anushkakathare07@gmail.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 CONTRIBUTING.md | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 8db796637f..e780f26420 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -383,3 +383,8 @@ then make PRs to the `releasing/[version number]` to fix the bugs via the regula
 If any error occurs after the release process, first check out a new hotfix branch from the `main` branch,
 make a patch version release following the semantic versioning, for example, `releasing/0.1.1`.
 Make sure the `releasing/0.1.1` is merged back into both `dev` and `main` and all the test pipelines succeed.
+
+
+<p align="right">
+  <a href="#introduction">⬆️ Back to Top</a>
+</p>

From 052dbb4439165bfc1fc3132fadb0955587e4d30e Mon Sep 17 00:00:00 2001
From: Boris Fomitchev <borisfom@users.noreply.github.com>
Date: Tue, 22 Oct 2024 01:55:35 -0700
Subject: [PATCH 171/183] Streamlined Rearrange in SpatialAttentionBlock
 (#8130)

The Rearrange code failed dynamo export in 24.09 container:
https://github.com/pytorch/pytorch/issues/137629
While we can't still use dynamo export with TRT in 23.09, I also noticed
that my workaround improved runtime by about 1 second end-to-end for 100
seconds run.

### Description

Replaced einops Rearrange with reshape/transpose

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).

---------

Signed-off-by: Boris Fomitchev <bfomitchev@nvidia.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/networks/blocks/spatialattention.py | 23 +++--------------------
 1 file changed, 3 insertions(+), 20 deletions(-)

diff --git a/monai/networks/blocks/spatialattention.py b/monai/networks/blocks/spatialattention.py
index 665442b55e..60a89a7840 100644
--- a/monai/networks/blocks/spatialattention.py
+++ b/monai/networks/blocks/spatialattention.py
@@ -17,9 +17,6 @@
 import torch.nn as nn
 
 from monai.networks.blocks import SABlock
-from monai.utils import optional_import
-
-Rearrange, _ = optional_import("einops.layers.torch", name="Rearrange")
 
 
 class SpatialAttentionBlock(nn.Module):
@@ -74,24 +71,10 @@ def __init__(
 
     def forward(self, x: torch.Tensor):
         residual = x
-
-        if self.spatial_dims == 1:
-            h = x.shape[2]
-            rearrange_input = Rearrange("b c h -> b h c")
-            rearrange_output = Rearrange("b h c -> b c h", h=h)
-        if self.spatial_dims == 2:
-            h, w = x.shape[2], x.shape[3]
-            rearrange_input = Rearrange("b c h w -> b (h w) c")
-            rearrange_output = Rearrange("b (h w) c -> b c h w", h=h, w=w)
-        else:
-            h, w, d = x.shape[2], x.shape[3], x.shape[4]
-            rearrange_input = Rearrange("b c h w d -> b (h w d) c")
-            rearrange_output = Rearrange("b (h w d) c -> b c h w d", h=h, w=w, d=d)
-
+        shape = x.shape
         x = self.norm(x)
-        x = rearrange_input(x)  # B x (x_dim * y_dim [ * z_dim]) x C
-
+        x = x.reshape(*shape[:2], -1).transpose(1, 2)  # "b c h w d -> b (h w d) c"
         x = self.attn(x)
-        x = rearrange_output(x)  # B x  x C x x_dim * y_dim * [z_dim]
+        x = x.transpose(1, 2).reshape(shape)  # "b (h w d) c -> b c h w d"
         x = x + residual
         return x

From 684688a2877b07b0e1da98bb13f18a60ac73d426 Mon Sep 17 00:00:00 2001
From: binliunls <107988372+binliunls@users.noreply.github.com>
Date: Tue, 22 Oct 2024 20:18:02 +0800
Subject: [PATCH 172/183] Optimize VISTA3D (#8123)

Fixes #8122 .

### Description
As shown in [this
PR](https://github.com/Project-MONAI/model-zoo/pull/671), the memory
malloc and mask embedding for-loop are the bottlenecks that caused the
vista3d slow inference. Therefore, this PR fixed them by adding the
logic for malloc and replacing the for-loop with a tensor
multiplication.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: binliu <binliu@nvidia.com>
Co-authored-by: Yiheng Wang <68361391+yiheng-wang-nv@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/nets/segresnet_ds.py | 4 +++-
 monai/networks/nets/vista3d.py      | 8 +++-----
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/monai/networks/nets/segresnet_ds.py b/monai/networks/nets/segresnet_ds.py
index 1ac5a79ee3..098e490511 100644
--- a/monai/networks/nets/segresnet_ds.py
+++ b/monai/networks/nets/segresnet_ds.py
@@ -508,8 +508,10 @@ def forward(  # type: ignore
 
         outputs: list[torch.Tensor] = []
         outputs_auto: list[torch.Tensor] = []
-        x_ = x.clone()
+        x_ = x
         if with_point:
+            if with_label:
+                x_ = x.clone()
             i = 0
             for level in self.up_layers:
                 x = level["upsample"](x)
diff --git a/monai/networks/nets/vista3d.py b/monai/networks/nets/vista3d.py
index 4215a9a594..6313b7812d 100644
--- a/monai/networks/nets/vista3d.py
+++ b/monai/networks/nets/vista3d.py
@@ -639,12 +639,10 @@ def forward(self, src: torch.Tensor, class_vector: torch.Tensor):
         if self.use_mlp:
             class_embedding = self.mlp(class_embedding)
         # [b,1,feat] @ [1,feat,dim], batch dimension become class_embedding batch dimension.
-        masks = []
-        for i in range(b):
-            mask = class_embedding @ src[[i]].view(1, c, h * w * d)
-            masks.append(mask.view(-1, 1, h, w, d))
+        masks_embedding = class_embedding.squeeze() @ src.view(b, c, h * w * d)
+        masks_embedding = masks_embedding.view(b, -1, h, w, d).transpose(0, 1)
 
-        return torch.cat(masks, 1), class_embedding
+        return masks_embedding, class_embedding
 
 
 class TwoWayTransformer(nn.Module):

From a6f8f688146b8b7ed8e4bff0d1cd1ef2e6f3d7f4 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 23 Oct 2024 14:38:34 +0800
Subject: [PATCH 173/183] Skip torch trt convert test with torch newer than or
 equal to 2.5.0 (#8165)

Fixes # .

### Description
related bug
https://github.com/pytorch/pytorch/issues/138674

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 tests/test_trt_compile.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/test_trt_compile.py b/tests/test_trt_compile.py
index 2f9db8f0c2..5a56f0e4a2 100644
--- a/tests/test_trt_compile.py
+++ b/tests/test_trt_compile.py
@@ -21,7 +21,7 @@
 from monai.networks import trt_compile
 from monai.networks.nets import UNet, cell_sam_wrapper, vista3d132
 from monai.utils import min_version, optional_import
-from tests.utils import skip_if_no_cuda, skip_if_quick, skip_if_windows
+from tests.utils import SkipIfAtLeastPyTorchVersion, skip_if_no_cuda, skip_if_quick, skip_if_windows
 
 trt, trt_imported = optional_import("tensorrt", "10.1.0", min_version)
 polygraphy, polygraphy_imported = optional_import("polygraphy")
@@ -46,6 +46,7 @@ def tearDown(self):
         if current_device != self.gpu_device:
             torch.cuda.set_device(self.gpu_device)
 
+    @SkipIfAtLeastPyTorchVersion((2, 5, 0))
     def test_handler(self):
         from ignite.engine import Engine
 

From 90173c9b8676a5a414a5068e59935230f00a8abc Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 23 Oct 2024 23:47:01 +0800
Subject: [PATCH 174/183] Enable redirection of all loggers by configuring a
 FileHandler within the bundle (#8142)

Fix https://github.com/Project-MONAI/model-zoo/issues/658, part of #7513

### Description
Enable redirection of all loggers by configuring a FileHandler within
the bundle

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/bundle/scripts.py         | 3 +--
 monai/bundle/workflows.py       | 2 +-
 monai/handlers/stats_handler.py | 2 +-
 monai/networks/trt_compiler.py  | 6 +++---
 monai/utils/misc.py             | 2 +-
 5 files changed, 7 insertions(+), 8 deletions(-)

diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index 4251da0b6f..884723ed68 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -1945,7 +1945,6 @@ def create_workflow(
 
     """
     _args = update_kwargs(args=args_file, workflow_name=workflow_name, config_file=config_file, **kwargs)
-    _log_input_summary(tag="run", args=_args)
     (workflow_name, config_file) = _pop_args(
         _args, workflow_name=ConfigWorkflow, config_file=None
     )  # the default workflow name is "ConfigWorkflow"
@@ -1969,7 +1968,7 @@ def create_workflow(
         workflow_ = workflow_class(**_args)
 
     workflow_.initialize()
-
+    _log_input_summary(tag="run", args=_args)
     return workflow_
 
 
diff --git a/monai/bundle/workflows.py b/monai/bundle/workflows.py
index d728d7d930..ccdb08e208 100644
--- a/monai/bundle/workflows.py
+++ b/monai/bundle/workflows.py
@@ -316,8 +316,8 @@ def __init__(
                 else:
                     raise FileNotFoundError(f"Cannot find the logging config file: {logging_file}.")
             else:
-                logger.info(f"Setting logging properties based on config: {logging_file}.")
                 fileConfig(str(logging_file), disable_existing_loggers=False)
+                logger.info(f"Setting logging properties based on config: {logging_file}.")
 
         self.parser = ConfigParser()
         self.parser.read_config(f=config_file)
diff --git a/monai/handlers/stats_handler.py b/monai/handlers/stats_handler.py
index ab36d19bd1..c4971e9cac 100644
--- a/monai/handlers/stats_handler.py
+++ b/monai/handlers/stats_handler.py
@@ -74,7 +74,7 @@ def __init__(
         output_transform: Callable = lambda x: x[0],
         global_epoch_transform: Callable = lambda x: x,
         state_attributes: Sequence[str] | None = None,
-        name: str | None = "StatsHandler",
+        name: str | None = "monai.handlers.StatsHandler",
         tag_name: str = DEFAULT_TAG,
         key_var_format: str = DEFAULT_KEY_VAL_FORMAT,
     ) -> None:
diff --git a/monai/networks/trt_compiler.py b/monai/networks/trt_compiler.py
index 00d2eb61af..a360f63dbd 100644
--- a/monai/networks/trt_compiler.py
+++ b/monai/networks/trt_compiler.py
@@ -115,7 +115,7 @@ def __init__(self, plan_path, logger=None):
           logger: optional logger object
         """
         self.plan_path = plan_path
-        self.logger = logger or get_logger("trt_compile")
+        self.logger = logger or get_logger("monai.networks.trt_compiler")
         self.logger.info(f"Loading TensorRT engine: {self.plan_path}")
         self.engine = engine_from_bytes(bytes_from_path(self.plan_path))
         self.tensors = OrderedDict()
@@ -288,7 +288,7 @@ def __init__(
         self.fallback = fallback
         self.disabled = False
 
-        self.logger = logger or get_logger("trt_compile")
+        self.logger = logger or get_logger("monai.networks.trt_compiler")
 
         # Normally we read input_names from forward() but can be overridden
         if input_names is None:
@@ -563,7 +563,7 @@ def find_sub(parent, submodule):
         else:
             wrap(model, base_path)
     else:
-        logger = logger or get_logger("trt_compile")
+        logger = logger or get_logger("monai.networks.trt_compiler")
         logger.warning("TensorRT and/or polygraphy packages are not available! trt_compile() has no effect.")
 
     return model
diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index 6386aae713..ec9b1256a2 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -887,7 +887,7 @@ def run_cmd(cmd_list: list[str], **kwargs: Any) -> subprocess.CompletedProcess:
     if kwargs.pop("run_cmd_verbose", False):
         import monai
 
-        monai.apps.utils.get_logger("run_cmd").info(f"{cmd_list}")  # type: ignore[attr-defined]
+        monai.apps.utils.get_logger("monai.utils.run_cmd").info(f"{cmd_list}")  # type: ignore[attr-defined]
     try:
         return subprocess.run(cmd_list, **kwargs)
     except subprocess.CalledProcessError as e:

From 8dc3b4f14931233c573e46e6a2442cfc5439b5d7 Mon Sep 17 00:00:00 2001
From: James Butler <jamesobutler@users.noreply.github.com>
Date: Thu, 24 Oct 2024 11:26:16 -0400
Subject: [PATCH 175/183] Apply pyupgrade fixes for Python 3.9+ syntax (#8150)

### Description

Included is a commit here applying pyupgrade changes for Python 3.8+
syntax. This should have been included in
https://github.com/Project-MONAI/MONAI/commit/104a360f953b5d720b3eb8f5b1a0546540e5e391
when Python 3.7 support was dropped.

Also included is a commit here apply pyupgrade changes for Python 3.9+
syntax. This should have been included in
https://github.com/Project-MONAI/MONAI/commit/14b086b553693f5d344ff054f37d12ce6839da06
when Python 3.8 support was dropped.

I've also run the pre-commit autoupdate command to use the latest
versions of the various pre-commit hook repos. It appears that
pre-commit.ci bot has not been doing this on the quarterly schedule as
expected? It appears the last time it submitted the PR to update the
pre-commit hook repos was back in
https://github.com/Project-MONAI/MONAI/pull/6286 from April 2023.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).

---------

Signed-off-by: James Butler <james.butler@revvity.com>
---
 .pre-commit-config.yaml                       | 23 ++++++++-----------
 monai/__init__.py                             |  4 ++--
 .../detection/networks/retinanet_network.py   |  4 ++--
 monai/apps/detection/transforms/array.py      |  3 ++-
 .../maisi/networks/autoencoderkl_maisi.py     |  2 +-
 .../maisi/networks/controlnet_maisi.py        |  2 +-
 monai/apps/pathology/engines/utils.py         |  3 ++-
 monai/apps/pathology/inferers/inferer.py      |  3 ++-
 monai/apps/pathology/metrics/lesion_froc.py   |  3 ++-
 monai/apps/pathology/transforms/post/array.py |  3 ++-
 monai/apps/tcia/utils.py                      |  2 +-
 monai/apps/utils.py                           |  5 +---
 monai/apps/vista3d/transforms.py              |  2 +-
 monai/bundle/reference_resolver.py            |  4 ++--
 monai/bundle/workflows.py                     |  3 ++-
 monai/config/type_definitions.py              |  3 ++-
 monai/data/meta_obj.py                        |  3 ++-
 monai/data/meta_tensor.py                     |  3 ++-
 monai/engines/evaluator.py                    |  3 ++-
 monai/engines/trainer.py                      |  3 ++-
 monai/engines/utils.py                        |  4 ++--
 monai/handlers/clearml_handlers.py            |  3 ++-
 monai/inferers/utils.py                       |  4 ++--
 monai/losses/hausdorff_loss.py                |  2 +-
 monai/metrics/utils.py                        |  3 ++-
 monai/networks/nets/swin_unetr.py             |  2 +-
 monai/transforms/intensity/dictionary.py      |  3 ++-
 monai/transforms/lazy/functional.py           |  3 ++-
 monai/transforms/spatial/array.py             |  6 ++---
 monai/transforms/utility/dictionary.py        |  4 ++--
 monai/transforms/utils_morphological_ops.py   |  2 +-
 monai/utils/component_store.py                |  3 ++-
 monai/utils/decorators.py                     |  3 ++-
 monai/utils/dist.py                           |  7 +-----
 monai/utils/module.py                         |  4 ++--
 monai/utils/state_cacher.py                   |  3 ++-
 tests/test_dynunet.py                         |  3 ++-
 tests/test_network_consistency.py             |  2 +-
 38 files changed, 75 insertions(+), 67 deletions(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 3fff6ed631..2a57fbf31a 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -9,7 +9,7 @@ ci:
 
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.6.0
+    rev: v5.0.0
     hooks:
       - id: end-of-file-fixer
       - id: trailing-whitespace
@@ -26,33 +26,30 @@ repos:
         args: ['--autofix', '--no-sort-keys', '--indent=4']
       - id: end-of-file-fixer
       - id: mixed-line-ending
-  - repo: https://github.com/charliermarsh/ruff-pre-commit
-    rev: v0.3.5
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.7.0
     hooks:
     -   id: ruff
         args:
         - --fix
 
   - repo: https://github.com/asottile/pyupgrade
-    rev: v3.3.1
+    rev: v3.19.0
     hooks:
       - id: pyupgrade
-        args: [--py37-plus]
-        name: Upgrade code excluding monai networks
+        args: [--py39-plus, --keep-runtime-typing]
+        name: Upgrade code with exceptions
         exclude: |
           (?x)(
               ^versioneer.py|
               ^monai/_version.py|
-              ^monai/networks/|  # no PEP 604 for torchscript tensorrt
-              ^monai/losses/  # no PEP 604 for torchscript tensorrt
+              ^monai/networks/| # avoid typing rewrites
+              ^monai/apps/detection/utils/anchor_utils.py| # avoid typing rewrites
+              ^tests/test_compute_panoptic_quality.py # avoid typing rewrites
           )
-      - id: pyupgrade
-        args: [--py37-plus, --keep-runtime-typing]
-        name: Upgrade monai networks
-        files: (?x)(^monai/networks/)
 
   - repo: https://github.com/asottile/yesqa
-    rev: v1.4.0
+    rev: v1.5.0
     hooks:
       - id: yesqa
         name: Unused noqa
diff --git a/monai/__init__.py b/monai/__init__.py
index f6fc8b0646..d92557a8e1 100644
--- a/monai/__init__.py
+++ b/monai/__init__.py
@@ -11,12 +11,12 @@
 
 from __future__ import annotations
 
+import logging
 import os
 import sys
-import logging
 import warnings
-from ._version import get_versions
 
+from ._version import get_versions
 
 old_showwarning = warnings.showwarning
 
diff --git a/monai/apps/detection/networks/retinanet_network.py b/monai/apps/detection/networks/retinanet_network.py
index ec86c3b0e9..ca6a8f5c19 100644
--- a/monai/apps/detection/networks/retinanet_network.py
+++ b/monai/apps/detection/networks/retinanet_network.py
@@ -42,7 +42,7 @@
 import math
 import warnings
 from collections.abc import Callable, Sequence
-from typing import Any, Dict
+from typing import Any
 
 import torch
 from torch import Tensor, nn
@@ -330,7 +330,7 @@ def forward(self, images: Tensor) -> Any:
         features = self.feature_extractor(images)
         if isinstance(features, Tensor):
             feature_maps = [features]
-        elif torch.jit.isinstance(features, Dict[str, Tensor]):
+        elif torch.jit.isinstance(features, dict[str, Tensor]):
             feature_maps = list(features.values())
         else:
             feature_maps = list(features)
diff --git a/monai/apps/detection/transforms/array.py b/monai/apps/detection/transforms/array.py
index d8ffce4584..301a636b6c 100644
--- a/monai/apps/detection/transforms/array.py
+++ b/monai/apps/detection/transforms/array.py
@@ -15,7 +15,8 @@
 
 from __future__ import annotations
 
-from typing import Any, Sequence
+from collections.abc import Sequence
+from typing import Any
 
 import numpy as np
 import torch
diff --git a/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py b/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
index a52274b24a..6251ea8e83 100644
--- a/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
+++ b/monai/apps/generation/maisi/networks/autoencoderkl_maisi.py
@@ -13,7 +13,7 @@
 
 import gc
 import logging
-from typing import Sequence
+from collections.abc import Sequence
 
 import torch
 import torch.nn as nn
diff --git a/monai/apps/generation/maisi/networks/controlnet_maisi.py b/monai/apps/generation/maisi/networks/controlnet_maisi.py
index 269086d971..7c13fd7bc6 100644
--- a/monai/apps/generation/maisi/networks/controlnet_maisi.py
+++ b/monai/apps/generation/maisi/networks/controlnet_maisi.py
@@ -11,7 +11,7 @@
 
 from __future__ import annotations
 
-from typing import Sequence
+from collections.abc import Sequence
 
 import torch
 
diff --git a/monai/apps/pathology/engines/utils.py b/monai/apps/pathology/engines/utils.py
index 02249ed640..87ca0f8e76 100644
--- a/monai/apps/pathology/engines/utils.py
+++ b/monai/apps/pathology/engines/utils.py
@@ -11,7 +11,8 @@
 
 from __future__ import annotations
 
-from typing import Any, Sequence
+from collections.abc import Sequence
+from typing import Any
 
 import torch
 
diff --git a/monai/apps/pathology/inferers/inferer.py b/monai/apps/pathology/inferers/inferer.py
index 71259ca7df..392cba221f 100644
--- a/monai/apps/pathology/inferers/inferer.py
+++ b/monai/apps/pathology/inferers/inferer.py
@@ -11,7 +11,8 @@
 
 from __future__ import annotations
 
-from typing import Any, Callable, Sequence
+from collections.abc import Sequence
+from typing import Any, Callable
 
 import numpy as np
 import torch
diff --git a/monai/apps/pathology/metrics/lesion_froc.py b/monai/apps/pathology/metrics/lesion_froc.py
index f4bf51ab28..138488348a 100644
--- a/monai/apps/pathology/metrics/lesion_froc.py
+++ b/monai/apps/pathology/metrics/lesion_froc.py
@@ -11,7 +11,8 @@
 
 from __future__ import annotations
 
-from typing import TYPE_CHECKING, Any, Iterable
+from collections.abc import Iterable
+from typing import TYPE_CHECKING, Any
 
 import numpy as np
 
diff --git a/monai/apps/pathology/transforms/post/array.py b/monai/apps/pathology/transforms/post/array.py
index 035bce2c69..561ed3ae20 100644
--- a/monai/apps/pathology/transforms/post/array.py
+++ b/monai/apps/pathology/transforms/post/array.py
@@ -12,7 +12,8 @@
 from __future__ import annotations
 
 import warnings
-from typing import Callable, Sequence
+from collections.abc import Sequence
+from typing import Callable
 
 import numpy as np
 import torch
diff --git a/monai/apps/tcia/utils.py b/monai/apps/tcia/utils.py
index 5524b488e9..f023cdbc87 100644
--- a/monai/apps/tcia/utils.py
+++ b/monai/apps/tcia/utils.py
@@ -12,7 +12,7 @@
 from __future__ import annotations
 
 import os
-from typing import Iterable
+from collections.abc import Iterable
 
 import monai
 from monai.config.type_definitions import PathLike
diff --git a/monai/apps/utils.py b/monai/apps/utils.py
index 0c998146a3..c2e17d3247 100644
--- a/monai/apps/utils.py
+++ b/monai/apps/utils.py
@@ -136,10 +136,7 @@ def check_hash(filepath: PathLike, val: str | None = None, hash_type: str = "md5
         return True
     actual_hash_func = look_up_option(hash_type.lower(), SUPPORTED_HASH_TYPES)
 
-    if sys.version_info >= (3, 9):
-        actual_hash = actual_hash_func(usedforsecurity=False)  # allows checks on FIPS enabled machines
-    else:
-        actual_hash = actual_hash_func()
+    actual_hash = actual_hash_func(usedforsecurity=False)  # allows checks on FIPS enabled machines
 
     try:
         with open(filepath, "rb") as f:
diff --git a/monai/apps/vista3d/transforms.py b/monai/apps/vista3d/transforms.py
index 3e8145cd80..bd7fb19493 100644
--- a/monai/apps/vista3d/transforms.py
+++ b/monai/apps/vista3d/transforms.py
@@ -12,7 +12,7 @@
 from __future__ import annotations
 
 import warnings
-from typing import Sequence
+from collections.abc import Sequence
 
 import numpy as np
 import torch
diff --git a/monai/bundle/reference_resolver.py b/monai/bundle/reference_resolver.py
index 050cd75fa7..df69b021e1 100644
--- a/monai/bundle/reference_resolver.py
+++ b/monai/bundle/reference_resolver.py
@@ -13,8 +13,8 @@
 
 import re
 import warnings
-from collections.abc import Sequence
-from typing import Any, Iterator
+from collections.abc import Iterator, Sequence
+from typing import Any
 
 from monai.bundle.config_item import ConfigComponent, ConfigExpression, ConfigItem
 from monai.bundle.utils import DEPRECATED_ID_MAPPING, ID_REF_KEY, ID_SEP_KEY
diff --git a/monai/bundle/workflows.py b/monai/bundle/workflows.py
index ccdb08e208..3ecd5dfbc5 100644
--- a/monai/bundle/workflows.py
+++ b/monai/bundle/workflows.py
@@ -16,10 +16,11 @@
 import sys
 import time
 from abc import ABC, abstractmethod
+from collections.abc import Sequence
 from copy import copy
 from logging.config import fileConfig
 from pathlib import Path
-from typing import Any, Sequence
+from typing import Any
 
 from monai.apps.utils import get_logger
 from monai.bundle.config_parser import ConfigParser
diff --git a/monai/config/type_definitions.py b/monai/config/type_definitions.py
index 57454a94e1..48c0547e31 100644
--- a/monai/config/type_definitions.py
+++ b/monai/config/type_definitions.py
@@ -12,7 +12,8 @@
 from __future__ import annotations
 
 import os
-from typing import Collection, Hashable, Iterable, Sequence, TypeVar, Union
+from collections.abc import Collection, Hashable, Iterable, Sequence
+from typing import TypeVar, Union
 
 import numpy as np
 import torch
diff --git a/monai/data/meta_obj.py b/monai/data/meta_obj.py
index 0dccaa9e1c..15e6e8be15 100644
--- a/monai/data/meta_obj.py
+++ b/monai/data/meta_obj.py
@@ -13,8 +13,9 @@
 
 import itertools
 import pprint
+from collections.abc import Iterable
 from copy import deepcopy
-from typing import Any, Iterable
+from typing import Any
 
 import numpy as np
 import torch
diff --git a/monai/data/meta_tensor.py b/monai/data/meta_tensor.py
index 2df4da4a35..ac171e8508 100644
--- a/monai/data/meta_tensor.py
+++ b/monai/data/meta_tensor.py
@@ -13,8 +13,9 @@
 
 import functools
 import warnings
+from collections.abc import Sequence
 from copy import deepcopy
-from typing import Any, Sequence
+from typing import Any
 
 import numpy as np
 import torch
diff --git a/monai/engines/evaluator.py b/monai/engines/evaluator.py
index 523c3dcbf6..d70a39726b 100644
--- a/monai/engines/evaluator.py
+++ b/monai/engines/evaluator.py
@@ -12,7 +12,8 @@
 from __future__ import annotations
 
 import warnings
-from typing import TYPE_CHECKING, Any, Callable, Iterable, Sequence
+from collections.abc import Iterable, Sequence
+from typing import TYPE_CHECKING, Any, Callable
 
 import torch
 from torch.utils.data import DataLoader
diff --git a/monai/engines/trainer.py b/monai/engines/trainer.py
index bbcc9c880b..a0be86bae5 100644
--- a/monai/engines/trainer.py
+++ b/monai/engines/trainer.py
@@ -12,7 +12,8 @@
 from __future__ import annotations
 
 import warnings
-from typing import TYPE_CHECKING, Any, Callable, Iterable, Sequence
+from collections.abc import Iterable, Sequence
+from typing import TYPE_CHECKING, Any, Callable
 
 import torch
 from torch.optim.optimizer import Optimizer
diff --git a/monai/engines/utils.py b/monai/engines/utils.py
index 11a0000989..8e19a18601 100644
--- a/monai/engines/utils.py
+++ b/monai/engines/utils.py
@@ -12,8 +12,8 @@
 from __future__ import annotations
 
 from abc import ABC, abstractmethod
-from collections.abc import Callable, Sequence
-from typing import TYPE_CHECKING, Any, Mapping, cast
+from collections.abc import Callable, Mapping, Sequence
+from typing import TYPE_CHECKING, Any, cast
 
 import torch
 import torch.nn as nn
diff --git a/monai/handlers/clearml_handlers.py b/monai/handlers/clearml_handlers.py
index 1cfd6a33fb..0aa2a5cc08 100644
--- a/monai/handlers/clearml_handlers.py
+++ b/monai/handlers/clearml_handlers.py
@@ -11,7 +11,8 @@
 
 from __future__ import annotations
 
-from typing import TYPE_CHECKING, Any, Mapping, Sequence
+from collections.abc import Mapping, Sequence
+from typing import TYPE_CHECKING, Any
 
 from monai.utils import optional_import
 
diff --git a/monai/inferers/utils.py b/monai/inferers/utils.py
index bd99765348..edaf736091 100644
--- a/monai/inferers/utils.py
+++ b/monai/inferers/utils.py
@@ -12,8 +12,8 @@
 from __future__ import annotations
 
 import itertools
-from collections.abc import Callable, Mapping, Sequence
-from typing import Any, Iterable
+from collections.abc import Callable, Iterable, Mapping, Sequence
+from typing import Any
 
 import numpy as np
 import torch
diff --git a/monai/losses/hausdorff_loss.py b/monai/losses/hausdorff_loss.py
index 6117f27741..c58be2d253 100644
--- a/monai/losses/hausdorff_loss.py
+++ b/monai/losses/hausdorff_loss.py
@@ -79,7 +79,7 @@ def __init__(
                 Incompatible values.
 
         """
-        super(HausdorffDTLoss, self).__init__(reduction=LossReduction(reduction).value)
+        super().__init__(reduction=LossReduction(reduction).value)
         if other_act is not None and not callable(other_act):
             raise TypeError(f"other_act must be None or callable but is {type(other_act).__name__}.")
         if int(sigmoid) + int(softmax) > 1:
diff --git a/monai/metrics/utils.py b/monai/metrics/utils.py
index 340e54a1d7..96d60c9098 100644
--- a/monai/metrics/utils.py
+++ b/monai/metrics/utils.py
@@ -12,9 +12,10 @@
 from __future__ import annotations
 
 import warnings
+from collections.abc import Iterable, Sequence
 from functools import lru_cache, partial
 from types import ModuleType
-from typing import Any, Iterable, Sequence
+from typing import Any
 
 import numpy as np
 import torch
diff --git a/monai/networks/nets/swin_unetr.py b/monai/networks/nets/swin_unetr.py
index 714d986f4b..832135ad06 100644
--- a/monai/networks/nets/swin_unetr.py
+++ b/monai/networks/nets/swin_unetr.py
@@ -13,6 +13,7 @@
 
 import itertools
 from collections.abc import Sequence
+from typing import Final
 
 import numpy as np
 import torch
@@ -20,7 +21,6 @@
 import torch.nn.functional as F
 import torch.utils.checkpoint as checkpoint
 from torch.nn import LayerNorm
-from typing_extensions import Final
 
 from monai.networks.blocks import MLPBlock as Mlp
 from monai.networks.blocks import PatchEmbed, UnetOutBlock, UnetrBasicBlock, UnetrUpBlock
diff --git a/monai/transforms/intensity/dictionary.py b/monai/transforms/intensity/dictionary.py
index 5dbac485fe..f2b1a2fd40 100644
--- a/monai/transforms/intensity/dictionary.py
+++ b/monai/transforms/intensity/dictionary.py
@@ -17,7 +17,8 @@
 
 from __future__ import annotations
 
-from typing import Callable, Hashable, Mapping, Sequence
+from collections.abc import Hashable, Mapping, Sequence
+from typing import Callable
 
 import numpy as np
 
diff --git a/monai/transforms/lazy/functional.py b/monai/transforms/lazy/functional.py
index 6b95027832..a33d76807c 100644
--- a/monai/transforms/lazy/functional.py
+++ b/monai/transforms/lazy/functional.py
@@ -11,7 +11,8 @@
 
 from __future__ import annotations
 
-from typing import Any, Mapping, Sequence
+from collections.abc import Mapping, Sequence
+from typing import Any
 
 import torch
 
diff --git a/monai/transforms/spatial/array.py b/monai/transforms/spatial/array.py
index 6e39fb2e19..e4ed196eff 100644
--- a/monai/transforms/spatial/array.py
+++ b/monai/transforms/spatial/array.py
@@ -15,10 +15,10 @@
 from __future__ import annotations
 
 import warnings
-from collections.abc import Callable
+from collections.abc import Callable, Sequence
 from copy import deepcopy
 from itertools import zip_longest
-from typing import Any, Optional, Sequence, Tuple, Union, cast
+from typing import Any, Optional, Union, cast
 
 import numpy as np
 import torch
@@ -116,7 +116,7 @@
     "RandSimulateLowResolution",
 ]
 
-RandRange = Optional[Union[Sequence[Union[Tuple[float, float], float]], float]]
+RandRange = Optional[Union[Sequence[Union[tuple[float, float], float]], float]]
 
 
 class SpatialResample(InvertibleTransform, LazyTransform):
diff --git a/monai/transforms/utility/dictionary.py b/monai/transforms/utility/dictionary.py
index 79d0be522d..65c721e48e 100644
--- a/monai/transforms/utility/dictionary.py
+++ b/monai/transforms/utility/dictionary.py
@@ -18,9 +18,9 @@
 from __future__ import annotations
 
 import re
-from collections.abc import Callable, Hashable, Mapping
+from collections.abc import Callable, Hashable, Mapping, Sequence
 from copy import deepcopy
-from typing import Any, Sequence, cast
+from typing import Any, cast
 
 import numpy as np
 import torch
diff --git a/monai/transforms/utils_morphological_ops.py b/monai/transforms/utils_morphological_ops.py
index b3134c1865..61d3c5b858 100644
--- a/monai/transforms/utils_morphological_ops.py
+++ b/monai/transforms/utils_morphological_ops.py
@@ -11,7 +11,7 @@
 
 from __future__ import annotations
 
-from typing import Sequence
+from collections.abc import Sequence
 
 import torch
 import torch.nn.functional as F
diff --git a/monai/utils/component_store.py b/monai/utils/component_store.py
index d1e71eaebf..bf0d632ddd 100644
--- a/monai/utils/component_store.py
+++ b/monai/utils/component_store.py
@@ -12,9 +12,10 @@
 from __future__ import annotations
 
 from collections import namedtuple
+from collections.abc import Iterable
 from keyword import iskeyword
 from textwrap import dedent, indent
-from typing import Any, Callable, Iterable, TypeVar
+from typing import Any, Callable, TypeVar
 
 T = TypeVar("T")
 
diff --git a/monai/utils/decorators.py b/monai/utils/decorators.py
index 1c064468e8..a784510c64 100644
--- a/monai/utils/decorators.py
+++ b/monai/utils/decorators.py
@@ -15,7 +15,8 @@
 
 __all__ = ["RestartGenerator", "MethodReplacer"]
 
-from typing import Callable, Generator
+from collections.abc import Generator
+from typing import Callable
 
 
 class RestartGenerator:
diff --git a/monai/utils/dist.py b/monai/utils/dist.py
index c7ff988027..47da2bee6e 100644
--- a/monai/utils/dist.py
+++ b/monai/utils/dist.py
@@ -11,15 +11,10 @@
 
 from __future__ import annotations
 
-import sys
 import warnings
 from collections.abc import Callable
 from logging import Filter
-
-if sys.version_info >= (3, 8):
-    from typing import Literal
-
-from typing import overload
+from typing import Literal, overload
 
 import torch
 import torch.distributed as dist
diff --git a/monai/utils/module.py b/monai/utils/module.py
index df5fe873ae..1f7f8aecfc 100644
--- a/monai/utils/module.py
+++ b/monai/utils/module.py
@@ -18,14 +18,14 @@
 import re
 import sys
 import warnings
-from collections.abc import Callable, Collection, Hashable, Mapping
+from collections.abc import Callable, Collection, Hashable, Iterable, Mapping
 from functools import partial, wraps
 from importlib import import_module
 from pkgutil import walk_packages
 from pydoc import locate
 from re import match
 from types import FunctionType, ModuleType
-from typing import Any, Iterable, cast
+from typing import Any, cast
 
 import torch
 
diff --git a/monai/utils/state_cacher.py b/monai/utils/state_cacher.py
index d37e7abde4..60a074544b 100644
--- a/monai/utils/state_cacher.py
+++ b/monai/utils/state_cacher.py
@@ -15,8 +15,9 @@
 import os
 import pickle
 import tempfile
+from collections.abc import Hashable
 from types import ModuleType
-from typing import Any, Hashable
+from typing import Any
 
 import torch
 from torch.serialization import DEFAULT_PROTOCOL
diff --git a/tests/test_dynunet.py b/tests/test_dynunet.py
index f3c982056c..7c4882fcbb 100644
--- a/tests/test_dynunet.py
+++ b/tests/test_dynunet.py
@@ -13,7 +13,8 @@
 
 import platform
 import unittest
-from typing import Any, Sequence
+from collections.abc import Sequence
+from typing import Any
 
 import torch
 from parameterized import parameterized
diff --git a/tests/test_network_consistency.py b/tests/test_network_consistency.py
index 4182501808..bcfd448144 100644
--- a/tests/test_network_consistency.py
+++ b/tests/test_network_consistency.py
@@ -14,8 +14,8 @@
 import json
 import os
 import unittest
+from collections.abc import Sequence
 from glob import glob
-from typing import Sequence
 from unittest.case import skipIf
 
 import torch

From 5b2d78e8c6b97aa620fd7f84c261c48229605efe Mon Sep 17 00:00:00 2001
From: James Butler <jamesobutler@users.noreply.github.com>
Date: Thu, 24 Oct 2024 23:20:03 -0400
Subject: [PATCH 176/183] Fix Ruff Numpy2 deprecation rules (#8179)

### Description

This addresses expired numpy deprecations that happened with the numpy 2
release. This code is compatible with numpy 1 and 2 to support future
enablement of numpy2 for monai.

See https://docs.astral.sh/ruff/rules/numpy2-deprecation/

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).

---------

Signed-off-by: James Butler <james.butler@revvity.com>
---
 monai/metrics/fid.py                                |  2 +-
 monai/transforms/utils_pytorch_numpy_unification.py |  2 +-
 pyproject.toml                                      | 13 ++++++++++++-
 tests/test_compute_f_beta.py                        |  2 +-
 4 files changed, 15 insertions(+), 4 deletions(-)

diff --git a/monai/metrics/fid.py b/monai/metrics/fid.py
index d655ac1bee..596f9aef7c 100644
--- a/monai/metrics/fid.py
+++ b/monai/metrics/fid.py
@@ -82,7 +82,7 @@ def _cov(input_data: torch.Tensor, rowvar: bool = True) -> torch.Tensor:
 
 def _sqrtm(input_data: torch.Tensor) -> torch.Tensor:
     """Compute the square root of a matrix."""
-    scipy_res, _ = scipy.linalg.sqrtm(input_data.detach().cpu().numpy().astype(np.float_), disp=False)
+    scipy_res, _ = scipy.linalg.sqrtm(input_data.detach().cpu().numpy().astype(np.float64), disp=False)
     return torch.from_numpy(scipy_res)
 
 
diff --git a/monai/transforms/utils_pytorch_numpy_unification.py b/monai/transforms/utils_pytorch_numpy_unification.py
index 98b75cff76..365bd1eab5 100644
--- a/monai/transforms/utils_pytorch_numpy_unification.py
+++ b/monai/transforms/utils_pytorch_numpy_unification.py
@@ -88,7 +88,7 @@ def moveaxis(x: NdarrayOrTensor, src: int | Sequence[int], dst: int | Sequence[i
 def in1d(x, y):
     """`np.in1d` with equivalent implementation for torch."""
     if isinstance(x, np.ndarray):
-        return np.in1d(x, y)
+        return np.isin(x, y)
     return (x[..., None] == torch.tensor(y, device=x.device)).any(-1).view(-1)
 
 
diff --git a/pyproject.toml b/pyproject.toml
index c2ab92a43d..9dc9cf619b 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -39,7 +39,18 @@ exclude = "monai/bundle/__main__.py"
 
 [tool.ruff]
 line-length = 133
-lint.ignore = ["F401", "E741"]
+target-version = "py39"
+
+[tool.ruff.lint]
+select = [
+  "E", "F", "W", # flake8
+  "NPY",         # NumPy specific rules
+]
+extend-ignore = [
+  "E741", # ambiguous variable name
+  "F401", # unused import
+  "NPY002", # numpy-legacy-random
+]
 
 [tool.pytype]
 # Space-separated list of files or directories to exclude.
diff --git a/tests/test_compute_f_beta.py b/tests/test_compute_f_beta.py
index 43ebb6a6d5..be2a7fc176 100644
--- a/tests/test_compute_f_beta.py
+++ b/tests/test_compute_f_beta.py
@@ -59,7 +59,7 @@ def test_with_nan_values(self):
         metric = FBetaScore(get_not_nans=True)
         metric(
             y_pred=torch.Tensor([[1, 1, 1], [1, 1, 1], [1, 1, 1]]),
-            y=torch.Tensor([[1, 0, 1], [np.NaN, np.NaN, np.NaN], [1, 0, 1]]),
+            y=torch.Tensor([[1, 0, 1], [np.nan, np.nan, np.nan], [1, 0, 1]]),
         )
         assert_allclose(metric.aggregate()[0][0], torch.Tensor([0.727273]), atol=1e-6, rtol=1e-6)
 

From 82298adb01c27a6cbf0bf10a9916b09b41966bb0 Mon Sep 17 00:00:00 2001
From: czfy <545306097@qq.com>
Date: Fri, 25 Oct 2024 16:04:38 +0800
Subject: [PATCH 177/183] Fix torch.load() in PersistentDataset and GDSDataset
 (#8177)

### Description
Frequently getting warning massage in a newer Pytorch version (2.4.1 in
my case):
"_You are using `torch.load` with `weights_only=False` (the current
default value), which uses the default pickle module implicitly. It is
possible to construct malicious pickle data which will execute arbitrary
code during unpickling (See
https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models
for more details). In a future release, the default value for
`weights_only` will be flipped to `True`. This limits the functions that
could be executed during unpickling. Arbitrary objects will no longer be
allowed to be loaded via this mode unless they are explicitly
allowlisted by the user via `torch.serialization.add_safe_globals`. We
recommend you start setting `weights_only=True` for any use case where
you don't have full control of the loaded file. Please open an issue on
GitHub for any issues related to this experimental feature._"

This pull request fixes an issue with the `torch.load()` function in the
`PersistentDataset` and `GDSDataset` classes by adding
`weights_only=False`. The fix ensures that the `torch.load()` function
maintains load consistency in future versions of PyTorch.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: bnbqq8 <bnbqq8@i.smu.edu.cn>
---
 monai/data/dataset.py | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/monai/data/dataset.py b/monai/data/dataset.py
index 871b523289..8c53338d66 100644
--- a/monai/data/dataset.py
+++ b/monai/data/dataset.py
@@ -22,6 +22,7 @@
 import warnings
 from collections.abc import Callable, Sequence
 from copy import copy, deepcopy
+from inspect import signature
 from multiprocessing.managers import ListProxy
 from multiprocessing.pool import ThreadPool
 from pathlib import Path
@@ -371,7 +372,10 @@ def _cachecheck(self, item_transformed):
 
         if hashfile is not None and hashfile.is_file():  # cache hit
             try:
-                return torch.load(hashfile)
+                if "weights_only" in signature(torch.load).parameters:
+                    return torch.load(hashfile, weights_only=False)
+                else:
+                    return torch.load(hashfile)
             except PermissionError as e:
                 if sys.platform != "win32":
                     raise e
@@ -1670,4 +1674,7 @@ def _load_meta_cache(self, meta_hash_file_name):
         if meta_hash_file_name in self._meta_cache:
             return self._meta_cache[meta_hash_file_name]
         else:
-            return torch.load(self.cache_dir / meta_hash_file_name)
+            if "weights_only" in signature(torch.load).parameters:
+                return torch.load(self.cache_dir / meta_hash_file_name, weights_only=False)
+            else:
+                return torch.load(self.cache_dir / meta_hash_file_name)

From 3c252a819c8c785fb660c8208948b621b9aad0b9 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 29 Oct 2024 11:27:48 +0800
Subject: [PATCH 178/183] Fix the logging of a nested dictionary metric in
 MLflow (#8169)

Fix https://github.com/Project-MONAI/model-zoo/issues/697

### Description
Flatten the metric dict when the metric is a nested dictionary.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/handlers/mlflow_handler.py |  6 ++++--
 monai/handlers/stats_handler.py  |  5 ++---
 monai/networks/utils.py          | 13 ++++++-------
 monai/utils/__init__.py          |  1 +
 monai/utils/misc.py              | 13 +++++++++++++
 tests/test_handler_mlflow.py     |  9 ++++++++-
 6 files changed, 34 insertions(+), 13 deletions(-)

diff --git a/monai/handlers/mlflow_handler.py b/monai/handlers/mlflow_handler.py
index c7e293ea7d..3078d89f97 100644
--- a/monai/handlers/mlflow_handler.py
+++ b/monai/handlers/mlflow_handler.py
@@ -22,7 +22,7 @@
 from torch.utils.data import Dataset
 
 from monai.apps.utils import get_logger
-from monai.utils import CommonKeys, IgniteInfo, ensure_tuple, min_version, optional_import
+from monai.utils import CommonKeys, IgniteInfo, ensure_tuple, flatten_dict, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 mlflow, _ = optional_import("mlflow", descriptor="Please install mlflow before using MLFlowHandler.")
@@ -303,7 +303,9 @@ def _log_metrics(self, metrics: dict[str, Any], step: int | None = None) -> None
 
         run_id = self.cur_run.info.run_id
         timestamp = int(time.time() * 1000)
-        metrics_arr = [mlflow.entities.Metric(key, value, timestamp, step or 0) for key, value in metrics.items()]
+        metrics_arr = [
+            mlflow.entities.Metric(key, value, timestamp, step or 0) for key, value in flatten_dict(metrics).items()
+        ]
         self.client.log_batch(run_id=run_id, metrics=metrics_arr, params=[], tags=[])
 
     def _parse_artifacts(self):
diff --git a/monai/handlers/stats_handler.py b/monai/handlers/stats_handler.py
index c4971e9cac..214872fef4 100644
--- a/monai/handlers/stats_handler.py
+++ b/monai/handlers/stats_handler.py
@@ -19,7 +19,7 @@
 import torch
 
 from monai.apps import get_logger
-from monai.utils import IgniteInfo, is_scalar, min_version, optional_import
+from monai.utils import IgniteInfo, flatten_dict, is_scalar, min_version, optional_import
 
 Events, _ = optional_import("ignite.engine", IgniteInfo.OPT_IMPORT_VERSION, min_version, "Events")
 if TYPE_CHECKING:
@@ -211,8 +211,7 @@ def _default_epoch_print(self, engine: Engine) -> None:
 
         """
         current_epoch = self.global_epoch_transform(engine.state.epoch)
-
-        prints_dict = engine.state.metrics
+        prints_dict = flatten_dict(engine.state.metrics)
         if prints_dict is not None and len(prints_dict) > 0:
             out_str = f"Epoch[{current_epoch}] Metrics -- "
             for name in sorted(prints_dict):
diff --git a/monai/networks/utils.py b/monai/networks/utils.py
index 295a055390..cfad0364c3 100644
--- a/monai/networks/utils.py
+++ b/monai/networks/utils.py
@@ -16,6 +16,7 @@
 
 import io
 import re
+import tempfile
 import warnings
 from collections import OrderedDict
 from collections.abc import Callable, Mapping, Sequence
@@ -688,16 +689,17 @@ def convert_to_onnx(
             onnx_inputs = (inputs,)
         else:
             onnx_inputs = tuple(inputs)
-
+        temp_file = None
         if filename is None:
-            f = io.BytesIO()
+            temp_file = tempfile.NamedTemporaryFile()
+            f = temp_file.name
         else:
             f = filename
 
         torch.onnx.export(
             mode_to_export,
             onnx_inputs,
-            f=f,  # type: ignore[arg-type]
+            f=f,
             input_names=input_names,
             output_names=output_names,
             dynamic_axes=dynamic_axes,
@@ -705,10 +707,7 @@ def convert_to_onnx(
             do_constant_folding=do_constant_folding,
             **torch_versioned_kwargs,
         )
-        if filename is None:
-            onnx_model = onnx.load_model_from_string(f.getvalue())
-        else:
-            onnx_model = onnx.load(filename)
+        onnx_model = onnx.load(f)
 
     if do_constant_folding and polygraphy_imported:
         from polygraphy.backend.onnx.loader import fold_constants
diff --git a/monai/utils/__init__.py b/monai/utils/__init__.py
index 916c1a6c70..79dc1f2304 100644
--- a/monai/utils/__init__.py
+++ b/monai/utils/__init__.py
@@ -78,6 +78,7 @@
     ensure_tuple_size,
     fall_back_tuple,
     first,
+    flatten_dict,
     get_seed,
     has_option,
     is_immutable,
diff --git a/monai/utils/misc.py b/monai/utils/misc.py
index ec9b1256a2..b96a48ad7e 100644
--- a/monai/utils/misc.py
+++ b/monai/utils/misc.py
@@ -916,3 +916,16 @@ def unsqueeze_right(arr: NT, ndim: int) -> NT:
 def unsqueeze_left(arr: NT, ndim: int) -> NT:
     """Prepend 1-sized dimensions to `arr` to create a result with `ndim` dimensions."""
     return arr[(None,) * (ndim - arr.ndim)]
+
+
+def flatten_dict(metrics: dict[str, Any]) -> dict[str, Any]:
+    """
+    Flatten the nested dictionary to a flat dictionary.
+    """
+    result = {}
+    for key, value in metrics.items():
+        if isinstance(value, dict):
+            result.update(flatten_dict(value))
+        else:
+            result[key] = value
+    return result
diff --git a/tests/test_handler_mlflow.py b/tests/test_handler_mlflow.py
index 44adc49fc2..36d59ff1bf 100644
--- a/tests/test_handler_mlflow.py
+++ b/tests/test_handler_mlflow.py
@@ -122,6 +122,11 @@ def _train_func(engine, batch):
             def _update_metric(engine):
                 current_metric = engine.state.metrics.get("acc", 0.1)
                 engine.state.metrics["acc"] = current_metric + 0.1
+                # log nested metrics
+                engine.state.metrics["acc_per_label"] = {
+                    "label_0": current_metric + 0.1,
+                    "label_1": current_metric + 0.2,
+                }
                 engine.state.test = current_metric
 
             # set up testing handler
@@ -138,10 +143,12 @@ def _update_metric(engine):
                 state_attributes=["test"],
                 experiment_param=experiment_param,
                 artifacts=[artifact_path],
-                close_on_complete=True,
+                close_on_complete=False,
             )
             handler.attach(engine)
             engine.run(range(3), max_epochs=2)
+            cur_run = handler.client.get_run(handler.cur_run.info.run_id)
+            self.assertTrue("label_0" in cur_run.data.metrics.keys())
             handler.close()
             # check logging output
             self.assertTrue(len(glob.glob(test_path)) > 0)

From c1ceea3d4cbb0781eae4e209b80fe651a776fed2 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Tue, 29 Oct 2024 12:12:07 +0800
Subject: [PATCH 179/183] add wendell-hom to user list to trigger blossom-ci
 [skip ci] (#8184)

Fixes # .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/blossom-ci.yml | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/.github/workflows/blossom-ci.yml b/.github/workflows/blossom-ci.yml
index bf507bab3b..63ac5536d8 100644
--- a/.github/workflows/blossom-ci.yml
+++ b/.github/workflows/blossom-ci.yml
@@ -34,8 +34,7 @@ jobs:
       (
         github.actor == 'Nic-Ma' ||
         github.actor == 'wyli' ||
-        github.actor == 'pxLi' ||
-        github.actor == 'YanxuanLiu' ||
+        github.actor == 'wendell-hom' ||
         github.actor == 'KumoLiu'
       )
     steps:

From 530cc1f35e40e50d4b35de30f75b75ef6fb78d08 Mon Sep 17 00:00:00 2001
From: Eloi <eloi.navet@gmail.com>
Date: Thu, 7 Nov 2024 09:30:40 +0100
Subject: [PATCH 180/183] Fix ImageFilter to allow Gaussian filter without
 filter_size (#8189)

Fixes #8127

Update `ImageFilter` to handle Gaussian filter without requiring
`filter_size`.

* Modify `monai/transforms/utility/array.py` to allow Gaussian filter
without `filter_size`.
- Adjust `_check_filter_format` method to skip `filter_size` check for
Gaussian filter. Indeed Gauss filter is the only one in the list that
doesn't require a filter_size.
* Add unit test in `tests/test_image_filter.py` for Gaussian filter
without `filter_size`.
  - Verify output shape matches input shape.

Note that this method is compliant with the dictionnary version since
this one load the fixed version.

Signed-off-by: Eloi <eloi.navet@gmail.com>

---------

Signed-off-by: Eloi Navet <eloi.navet@labri.fr>
Signed-off-by: Eloi <eloi.navet@gmail.com>
Signed-off-by: Eloi eloi.navet@gmail.com
---
 monai/transforms/utility/array.py | 4 ++--
 tests/test_image_filter.py        | 6 ++++++
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/monai/transforms/utility/array.py b/monai/transforms/utility/array.py
index 72dd189009..1b3c59afdb 100644
--- a/monai/transforms/utility/array.py
+++ b/monai/transforms/utility/array.py
@@ -1609,9 +1609,9 @@ def _check_all_values_uneven(self, x: tuple) -> None:
 
     def _check_filter_format(self, filter: str | NdarrayOrTensor | nn.Module, filter_size: int | None = None) -> None:
         if isinstance(filter, str):
-            if not filter_size:
+            if filter != "gauss" and not filter_size:  # Gauss is the only filter that does not require `filter_size`
                 raise ValueError("`filter_size` must be specified when specifying filters by string.")
-            if filter_size % 2 == 0:
+            if filter_size and filter_size % 2 == 0:
                 raise ValueError("`filter_size` should be a single uneven integer.")
             if filter not in self.supported_filters:
                 raise NotImplementedError(f"{filter}. Supported filters are {self.supported_filters}.")
diff --git a/tests/test_image_filter.py b/tests/test_image_filter.py
index 76e38d94f4..fb08b2295d 100644
--- a/tests/test_image_filter.py
+++ b/tests/test_image_filter.py
@@ -134,6 +134,12 @@ def test_pass_empty_metadata_dict(self):
         out_tensor = filter(image)
         self.assertTrue(isinstance(out_tensor, MetaTensor))
 
+    def test_gaussian_filter_without_filter_size(self):
+        "Test Gaussian filter without specifying filter_size"
+        filter = ImageFilter("gauss", sigma=2)
+        out_tensor = filter(SAMPLE_IMAGE_2D)
+        self.assertEqual(out_tensor.shape[1:], SAMPLE_IMAGE_2D.shape[1:])
+
 
 class TestImageFilterDict(unittest.TestCase):
 

From 0bb20a88ec7869f6453aa58890df50ad6b2b6271 Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Mon, 11 Nov 2024 10:51:20 +0800
Subject: [PATCH 181/183] Update base image to 2410 (#8164)

Fixes # .

### Description

A few sentences describing the changes proposed in this pull request.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 .github/workflows/cron.yml | 16 ++++++++++------
 Dockerfile                 |  2 +-
 tests/test_trt_compile.py  |  2 +-
 3 files changed, 12 insertions(+), 8 deletions(-)

diff --git a/.github/workflows/cron.yml b/.github/workflows/cron.yml
index 6732ab7256..516e2d4743 100644
--- a/.github/workflows/cron.yml
+++ b/.github/workflows/cron.yml
@@ -16,7 +16,8 @@ jobs:
           - "PT110+CUDA113"
           - "PT113+CUDA118"
           - "PT210+CUDA121"
-          - "PTLATEST+CUDA124"
+          - "PT240+CUDA126"
+          - "PTLATEST+CUDA126"
         include:
           # https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes
           - environment: PT110+CUDA113
@@ -28,9 +29,12 @@ jobs:
           - environment: PT210+CUDA121
             pytorch: "pytorch==2.1.0 torchvision==0.16.0 --extra-index-url https://download.pytorch.org/whl/cu121"
             base: "nvcr.io/nvidia/pytorch:23.08-py3"  # CUDA 12.1
-          - environment: PTLATEST+CUDA124
+          - environment: PT240+CUDA126
+            pytorch: "pytorch==2.4.0 torchvision==0.19.0 --extra-index-url https://download.pytorch.org/whl/cu121"
+            base: "nvcr.io/nvidia/pytorch:24.08-py3"  # CUDA 12.6
+          - environment: PTLATEST+CUDA126
             pytorch: "-U torch torchvision --extra-index-url https://download.pytorch.org/whl/cu121"
-            base: "nvcr.io/nvidia/pytorch:24.08-py3"  # CUDA 12.4
+            base: "nvcr.io/nvidia/pytorch:24.10-py3"  # CUDA 12.6
     container:
       image: ${{ matrix.base }}
       options: "--gpus all"
@@ -80,7 +84,7 @@ jobs:
     if: github.repository == 'Project-MONAI/MONAI'
     strategy:
       matrix:
-        container: ["pytorch:23.08", "pytorch:24.08"]
+        container: ["pytorch:23.08", "pytorch:24.08", "pytorch:24.10"]
     container:
       image: nvcr.io/nvidia/${{ matrix.container }}-py3  # testing with the latest pytorch base image
       options: "--gpus all"
@@ -129,7 +133,7 @@ jobs:
     if: github.repository == 'Project-MONAI/MONAI'
     strategy:
       matrix:
-        container: ["pytorch:24.08"]
+        container: ["pytorch:24.10"]
     container:
       image: nvcr.io/nvidia/${{ matrix.container }}-py3  # testing with the latest pytorch base image
       options: "--gpus all"
@@ -233,7 +237,7 @@ jobs:
     if: github.repository == 'Project-MONAI/MONAI'
     needs: cron-gpu  # so that monai itself is verified first
     container:
-      image: nvcr.io/nvidia/pytorch:24.08-py3  # testing with the latest pytorch base image
+      image: nvcr.io/nvidia/pytorch:24.10-py3  # testing with the latest pytorch base image
       options: "--gpus all --ipc=host"
     runs-on: [self-hosted, linux, x64, integration]
     steps:
diff --git a/Dockerfile b/Dockerfile
index e45932c6bb..5fcfcf274d 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -11,7 +11,7 @@
 
 # To build with a different base image
 # please run `docker build` using the `--build-arg PYTORCH_IMAGE=...` flag.
-ARG PYTORCH_IMAGE=nvcr.io/nvidia/pytorch:24.08-py3
+ARG PYTORCH_IMAGE=nvcr.io/nvidia/pytorch:24.10-py3
 FROM ${PYTORCH_IMAGE}
 
 LABEL maintainer="monai.contact@gmail.com"
diff --git a/tests/test_trt_compile.py b/tests/test_trt_compile.py
index 5a56f0e4a2..6df5d520bd 100644
--- a/tests/test_trt_compile.py
+++ b/tests/test_trt_compile.py
@@ -46,7 +46,7 @@ def tearDown(self):
         if current_device != self.gpu_device:
             torch.cuda.set_device(self.gpu_device)
 
-    @SkipIfAtLeastPyTorchVersion((2, 5, 0))
+    @SkipIfAtLeastPyTorchVersion((2, 4, 1))
     def test_handler(self):
         from ignite.engine import Engine
 

From b6663b90f52b41eecd7e412ef5d8937ca640c6ad Mon Sep 17 00:00:00 2001
From: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Date: Wed, 13 Nov 2024 13:16:39 +0800
Subject: [PATCH 182/183] Add SM architecture version check (#8199)

Fixes #8198

NVIDIA Volta support (GPUs with compute capability 7.0) has been removed
starting with TensorRT 10.5. Review the [TensorRT Support
Matrix](https://docs.nvidia.com/deeplearning/tensorrt/support-matrix/index.html)
for which GPUs are supported by this release.
Add SM architecture version check to skip trt test before 7.0.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [ ] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [ ] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
---
 monai/bundle/scripts.py                       |  2 +
 monai/networks/trt_compiler.py                |  4 +-
 monai/utils/__init__.py                       |  1 +
 monai/utils/module.py                         | 41 +++++++++++++++++++
 tests/test_bundle_trt_export.py               |  9 +++-
 tests/test_convert_to_trt.py                  |  3 +-
 tests/test_trt_compile.py                     |  9 +++-
 ...version_after.py => test_version_after.py} | 22 ++++++++--
 tests/utils.py                                | 16 +++++++-
 9 files changed, 99 insertions(+), 8 deletions(-)
 rename tests/{test_pytorch_version_after.py => test_version_after.py} (72%)

diff --git a/monai/bundle/scripts.py b/monai/bundle/scripts.py
index 884723ed68..131c78008b 100644
--- a/monai/bundle/scripts.py
+++ b/monai/bundle/scripts.py
@@ -1589,6 +1589,8 @@ def trt_export(
     """
     Export the model checkpoint to the given filepath as a TensorRT engine-based TorchScript.
     Currently, this API only supports converting models whose inputs are all tensors.
+    Note: NVIDIA Volta support (GPUs with compute capability 7.0) has been removed starting with TensorRT 10.5.
+    Review the TensorRT Support Matrix for which GPUs are supported.
 
     There are two ways to export a model:
     1, Torch-TensorRT way: PyTorch module ---> TorchScript module ---> TensorRT engine-based TorchScript.
diff --git a/monai/networks/trt_compiler.py b/monai/networks/trt_compiler.py
index a360f63dbd..d2d05fae22 100644
--- a/monai/networks/trt_compiler.py
+++ b/monai/networks/trt_compiler.py
@@ -505,7 +505,9 @@ def trt_compile(
 ) -> torch.nn.Module:
     """
     Instruments model or submodule(s) with TrtCompiler and replaces its forward() with TRT hook.
-    Note: TRT 10.3 is recommended for best performance. Some nets may even fail to work with TRT 8.x
+    Note: TRT 10.3 is recommended for best performance. Some nets may even fail to work with TRT 8.x.
+        NVIDIA Volta support (GPUs with compute capability 7.0) has been removed starting with TensorRT 10.5.
+        Review the TensorRT Support Matrix for which GPUs are supported.
     Args:
       model: module to patch with TrtCompiler object.
       base_path: TRT plan(s) saved to f"{base_path}[.{submodule}].plan" path.
diff --git a/monai/utils/__init__.py b/monai/utils/__init__.py
index 79dc1f2304..8f2f400b5d 100644
--- a/monai/utils/__init__.py
+++ b/monai/utils/__init__.py
@@ -107,6 +107,7 @@
     InvalidPyTorchVersionError,
     OptionalImportError,
     allow_missing_reference,
+    compute_capabilities_after,
     damerau_levenshtein_distance,
     exact_version,
     get_full_type_name,
diff --git a/monai/utils/module.py b/monai/utils/module.py
index 1f7f8aecfc..d3f2ff09f2 100644
--- a/monai/utils/module.py
+++ b/monai/utils/module.py
@@ -634,3 +634,44 @@ def pytorch_after(major: int, minor: int, patch: int = 0, current_ver_string: st
     if is_prerelease:
         return False
     return True
+
+
+@functools.lru_cache(None)
+def compute_capabilities_after(major: int, minor: int = 0, current_ver_string: str | None = None) -> bool:
+    """
+    Compute whether the current system GPU CUDA compute capability is after or equal to the specified version.
+    The current system GPU CUDA compute capability is determined by the first GPU in the system.
+    The compared version is a string in the form of "major.minor".
+
+    Args:
+        major: major version number to be compared with.
+        minor: minor version number to be compared with. Defaults to 0.
+        current_ver_string: if None, the current system GPU CUDA compute capability will be used.
+
+    Returns:
+        True if the current system GPU CUDA compute capability is greater than or equal to the specified version.
+    """
+    if current_ver_string is None:
+        cuda_available = torch.cuda.is_available()
+        pynvml, has_pynvml = optional_import("pynvml")
+        if not has_pynvml:  # assuming that the user has Ampere and later GPU
+            return True
+        if not cuda_available:
+            return False
+        else:
+            pynvml.nvmlInit()
+            handle = pynvml.nvmlDeviceGetHandleByIndex(0)  # get the first GPU
+            major_c, minor_c = pynvml.nvmlDeviceGetCudaComputeCapability(handle)
+            current_ver_string = f"{major_c}.{minor_c}"
+            pynvml.nvmlShutdown()
+
+    ver, has_ver = optional_import("packaging.version", name="parse")
+    if has_ver:
+        return ver(".".join((f"{major}", f"{minor}"))) <= ver(f"{current_ver_string}")  # type: ignore
+    parts = f"{current_ver_string}".split("+", 1)[0].split(".", 2)
+    while len(parts) < 2:
+        parts += ["0"]
+    c_major, c_minor = parts[:2]
+    c_mn = int(c_major), int(c_minor)
+    mn = int(major), int(minor)
+    return c_mn > mn
diff --git a/tests/test_bundle_trt_export.py b/tests/test_bundle_trt_export.py
index 833a0ca1dc..835c8e5c1d 100644
--- a/tests/test_bundle_trt_export.py
+++ b/tests/test_bundle_trt_export.py
@@ -22,7 +22,13 @@
 from monai.data import load_net_with_metadata
 from monai.networks import save_state
 from monai.utils import optional_import
-from tests.utils import command_line_tests, skip_if_no_cuda, skip_if_quick, skip_if_windows
+from tests.utils import (
+    SkipIfBeforeComputeCapabilityVersion,
+    command_line_tests,
+    skip_if_no_cuda,
+    skip_if_quick,
+    skip_if_windows,
+)
 
 _, has_torchtrt = optional_import(
     "torch_tensorrt",
@@ -47,6 +53,7 @@
 @skip_if_windows
 @skip_if_no_cuda
 @skip_if_quick
+@SkipIfBeforeComputeCapabilityVersion((7, 0))
 class TestTRTExport(unittest.TestCase):
 
     def setUp(self):
diff --git a/tests/test_convert_to_trt.py b/tests/test_convert_to_trt.py
index 5579539764..712d887c3b 100644
--- a/tests/test_convert_to_trt.py
+++ b/tests/test_convert_to_trt.py
@@ -20,7 +20,7 @@
 from monai.networks import convert_to_trt
 from monai.networks.nets import UNet
 from monai.utils import optional_import
-from tests.utils import skip_if_no_cuda, skip_if_quick, skip_if_windows
+from tests.utils import SkipIfBeforeComputeCapabilityVersion, skip_if_no_cuda, skip_if_quick, skip_if_windows
 
 _, has_torchtrt = optional_import(
     "torch_tensorrt",
@@ -38,6 +38,7 @@
 @skip_if_windows
 @skip_if_no_cuda
 @skip_if_quick
+@SkipIfBeforeComputeCapabilityVersion((7, 0))
 class TestConvertToTRT(unittest.TestCase):
 
     def setUp(self):
diff --git a/tests/test_trt_compile.py b/tests/test_trt_compile.py
index 6df5d520bd..49404fdbbe 100644
--- a/tests/test_trt_compile.py
+++ b/tests/test_trt_compile.py
@@ -21,7 +21,13 @@
 from monai.networks import trt_compile
 from monai.networks.nets import UNet, cell_sam_wrapper, vista3d132
 from monai.utils import min_version, optional_import
-from tests.utils import SkipIfAtLeastPyTorchVersion, skip_if_no_cuda, skip_if_quick, skip_if_windows
+from tests.utils import (
+    SkipIfAtLeastPyTorchVersion,
+    SkipIfBeforeComputeCapabilityVersion,
+    skip_if_no_cuda,
+    skip_if_quick,
+    skip_if_windows,
+)
 
 trt, trt_imported = optional_import("tensorrt", "10.1.0", min_version)
 polygraphy, polygraphy_imported = optional_import("polygraphy")
@@ -36,6 +42,7 @@
 @skip_if_quick
 @unittest.skipUnless(trt_imported, "tensorrt is required")
 @unittest.skipUnless(polygraphy_imported, "polygraphy is required")
+@SkipIfBeforeComputeCapabilityVersion((7, 0))
 class TestTRTCompile(unittest.TestCase):
 
     def setUp(self):
diff --git a/tests/test_pytorch_version_after.py b/tests/test_version_after.py
similarity index 72%
rename from tests/test_pytorch_version_after.py
rename to tests/test_version_after.py
index 147707d2c0..b6cb741382 100644
--- a/tests/test_pytorch_version_after.py
+++ b/tests/test_version_after.py
@@ -15,9 +15,9 @@
 
 from parameterized import parameterized
 
-from monai.utils import pytorch_after
+from monai.utils import compute_capabilities_after, pytorch_after
 
-TEST_CASES = (
+TEST_CASES_PT = (
     (1, 5, 9, "1.6.0"),
     (1, 6, 0, "1.6.0"),
     (1, 6, 1, "1.6.0", False),
@@ -36,14 +36,30 @@
     (1, 6, 1, "1.6.0+cpu", False),
 )
 
+TEST_CASES_SM = [
+    # (major, minor, sm, expected)
+    (6, 1, "6.1", True),
+    (6, 1, "6.0", False),
+    (6, 0, "8.6", True),
+    (7, 0, "8", True),
+    (8, 6, "8", False),
+]
+
 
 class TestPytorchVersionCompare(unittest.TestCase):
 
-    @parameterized.expand(TEST_CASES)
+    @parameterized.expand(TEST_CASES_PT)
     def test_compare(self, a, b, p, current, expected=True):
         """Test pytorch_after with a and b"""
         self.assertEqual(pytorch_after(a, b, p, current), expected)
 
 
+class TestComputeCapabilitiesAfter(unittest.TestCase):
+
+    @parameterized.expand(TEST_CASES_SM)
+    def test_compute_capabilities_after(self, major, minor, sm, expected):
+        self.assertEqual(compute_capabilities_after(major, minor, sm), expected)
+
+
 if __name__ == "__main__":
     unittest.main()
diff --git a/tests/utils.py b/tests/utils.py
index 77b53cebb8..2a00af50e9 100644
--- a/tests/utils.py
+++ b/tests/utils.py
@@ -47,7 +47,7 @@
 from monai.networks import convert_to_onnx, convert_to_torchscript
 from monai.utils import optional_import
 from monai.utils.misc import MONAIEnvVars
-from monai.utils.module import pytorch_after
+from monai.utils.module import compute_capabilities_after, pytorch_after
 from monai.utils.tf32 import detect_default_tf32
 from monai.utils.type_conversion import convert_data_type
 
@@ -286,6 +286,20 @@ def __call__(self, obj):
         )(obj)
 
 
+class SkipIfBeforeComputeCapabilityVersion:
+    """Decorator to be used if test should be skipped
+    with Compute Capability older than that given."""
+
+    def __init__(self, compute_capability_tuple):
+        self.min_version = compute_capability_tuple
+        self.version_too_old = not compute_capabilities_after(*compute_capability_tuple)
+
+    def __call__(self, obj):
+        return unittest.skipIf(
+            self.version_too_old, f"Skipping tests that fail on Compute Capability versions before: {self.min_version}"
+        )(obj)
+
+
 def is_main_test_process():
     ps = torch.multiprocessing.current_process()
     if not ps or not hasattr(ps, "name"):

From 941e739c933691a2da2d111d3a693f47d6330939 Mon Sep 17 00:00:00 2001
From: Suraj Pai <bspai@bwh.harvard.edu>
Date: Wed, 13 Nov 2024 10:44:17 -0500
Subject: [PATCH 183/183] Add MedNext implementation (#8004)

Fixes #7786

### Description
Added MedNext architectures implementation for MONAI.

Since a lot of the code is heavily sourced from the original MedNext
repo, https://github.com/MIC-DKFZ/MedNeXt, I wanted to check if there is
an attribution policy with regarded to borrowed source code. I've added
a derivative notice bellow the monai copyright comment. Let me know if
this needs to be changed.

The blocks have been taken almost as is but the network implementation
has been changed largely to allow flexible blocks and follow MONAI
segresnet styling.

### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [ ] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Suraj Pai <bspai@bwh.harvard.edu>
Signed-off-by: Robin CREMESE <robin.cremese@pasteur.fr>
Co-authored-by: Robin CREMESE <robin.cremese@pasteur.fr>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: YunLiu <55491388+KumoLiu@users.noreply.github.com>
---
 monai/networks/blocks/__init__.py      |   1 +
 monai/networks/blocks/mednext_block.py | 309 +++++++++++++++++++++
 monai/networks/nets/__init__.py        |  19 ++
 monai/networks/nets/mednext.py         | 354 +++++++++++++++++++++++++
 tests/test_mednext.py                  | 122 +++++++++
 5 files changed, 805 insertions(+)
 create mode 100644 monai/networks/blocks/mednext_block.py
 create mode 100644 monai/networks/nets/mednext.py
 create mode 100644 tests/test_mednext.py

diff --git a/monai/networks/blocks/__init__.py b/monai/networks/blocks/__init__.py
index 47abc4a1c4..499caf2e0f 100644
--- a/monai/networks/blocks/__init__.py
+++ b/monai/networks/blocks/__init__.py
@@ -26,6 +26,7 @@
 from .fcn import FCN, GCN, MCFCN, Refine
 from .feature_pyramid_network import ExtraFPNBlock, FeaturePyramidNetwork, LastLevelMaxPool, LastLevelP6P7
 from .localnet_block import LocalNetDownSampleBlock, LocalNetFeatureExtractorBlock, LocalNetUpSampleBlock
+from .mednext_block import MedNeXtBlock, MedNeXtDownBlock, MedNeXtOutBlock, MedNeXtUpBlock
 from .mlp import MLPBlock
 from .patchembedding import PatchEmbed, PatchEmbeddingBlock
 from .regunet_block import RegistrationDownSampleBlock, RegistrationExtractionBlock, RegistrationResidualConvBlock
diff --git a/monai/networks/blocks/mednext_block.py b/monai/networks/blocks/mednext_block.py
new file mode 100644
index 0000000000..0aa2bb6b58
--- /dev/null
+++ b/monai/networks/blocks/mednext_block.py
@@ -0,0 +1,309 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Portions of this code are derived from the original repository at:
+# https://github.com/MIC-DKFZ/MedNeXt
+# and are used under the terms of the Apache License, Version 2.0.
+
+from __future__ import annotations
+
+import torch
+import torch.nn as nn
+
+all = ["MedNeXtBlock", "MedNeXtDownBlock", "MedNeXtUpBlock", "MedNeXtOutBlock"]
+
+
+def get_conv_layer(spatial_dim: int = 3, transpose: bool = False):
+    if spatial_dim == 2:
+        return nn.ConvTranspose2d if transpose else nn.Conv2d
+    else:  # spatial_dim == 3
+        return nn.ConvTranspose3d if transpose else nn.Conv3d
+
+
+class MedNeXtBlock(nn.Module):
+    """
+    MedNeXtBlock class for the MedNeXt model.
+
+    Args:
+        in_channels (int): Number of input channels.
+        out_channels (int): Number of output channels.
+        expansion_ratio (int): Expansion ratio for the block. Defaults to 4.
+        kernel_size (int): Kernel size for convolutions. Defaults to 7.
+        use_residual_connection (int): Whether to use residual connection. Defaults to True.
+        norm_type (str): Type of normalization to use. Defaults to "group".
+        dim (str): Dimension of the input. Can be "2d" or "3d". Defaults to "3d".
+        global_resp_norm (bool): Whether to use global response normalization. Defaults to False.
+    """
+
+    def __init__(
+        self,
+        in_channels: int,
+        out_channels: int,
+        expansion_ratio: int = 4,
+        kernel_size: int = 7,
+        use_residual_connection: int = True,
+        norm_type: str = "group",
+        dim="3d",
+        global_resp_norm=False,
+    ):
+
+        super().__init__()
+
+        self.do_res = use_residual_connection
+
+        self.dim = dim
+        conv = get_conv_layer(spatial_dim=2 if dim == "2d" else 3)
+        global_resp_norm_param_shape = (1,) * (2 if dim == "2d" else 3)
+        # First convolution layer with DepthWise Convolutions
+        self.conv1 = conv(
+            in_channels=in_channels,
+            out_channels=in_channels,
+            kernel_size=kernel_size,
+            stride=1,
+            padding=kernel_size // 2,
+            groups=in_channels,
+        )
+
+        # Normalization Layer. GroupNorm is used by default.
+        if norm_type == "group":
+            self.norm = nn.GroupNorm(num_groups=in_channels, num_channels=in_channels)  # type: ignore
+        elif norm_type == "layer":
+            self.norm = nn.LayerNorm(
+                normalized_shape=[in_channels] + [kernel_size] * (2 if dim == "2d" else 3)  # type: ignore
+            )
+        # Second convolution (Expansion) layer with Conv3D 1x1x1
+        self.conv2 = conv(
+            in_channels=in_channels, out_channels=expansion_ratio * in_channels, kernel_size=1, stride=1, padding=0
+        )
+
+        # GeLU activations
+        self.act = nn.GELU()
+
+        # Third convolution (Compression) layer with Conv3D 1x1x1
+        self.conv3 = conv(
+            in_channels=expansion_ratio * in_channels, out_channels=out_channels, kernel_size=1, stride=1, padding=0
+        )
+
+        self.global_resp_norm = global_resp_norm
+        if self.global_resp_norm:
+            global_resp_norm_param_shape = (1, expansion_ratio * in_channels) + global_resp_norm_param_shape
+            self.global_resp_beta = nn.Parameter(torch.zeros(global_resp_norm_param_shape), requires_grad=True)
+            self.global_resp_gamma = nn.Parameter(torch.zeros(global_resp_norm_param_shape), requires_grad=True)
+
+    def forward(self, x):
+        """
+        Forward pass of the MedNeXtBlock.
+
+        Args:
+            x (torch.Tensor): Input tensor.
+
+        Returns:
+            torch.Tensor: Output tensor.
+        """
+        x1 = x
+        x1 = self.conv1(x1)
+        x1 = self.act(self.conv2(self.norm(x1)))
+
+        if self.global_resp_norm:
+            # gamma, beta: learnable affine transform parameters
+            # X: input of shape (N,C,H,W,D)
+            if self.dim == "2d":
+                gx = torch.norm(x1, p=2, dim=(-2, -1), keepdim=True)
+            else:
+                gx = torch.norm(x1, p=2, dim=(-3, -2, -1), keepdim=True)
+            nx = gx / (gx.mean(dim=1, keepdim=True) + 1e-6)
+            x1 = self.global_resp_gamma * (x1 * nx) + self.global_resp_beta + x1
+        x1 = self.conv3(x1)
+        if self.do_res:
+            x1 = x + x1
+        return x1
+
+
+class MedNeXtDownBlock(MedNeXtBlock):
+    """
+    MedNeXtDownBlock class for downsampling in the MedNeXt model.
+
+    Args:
+        in_channels (int): Number of input channels.
+        out_channels (int): Number of output channels.
+        expansion_ratio (int): Expansion ratio for the block. Defaults to 4.
+        kernel_size (int): Kernel size for convolutions. Defaults to 7.
+        use_residual_connection (bool): Whether to use residual connection. Defaults to False.
+        norm_type (str): Type of normalization to use. Defaults to "group".
+        dim (str): Dimension of the input. Can be "2d" or "3d". Defaults to "3d".
+        global_resp_norm (bool): Whether to use global response normalization. Defaults to False.
+    """
+
+    def __init__(
+        self,
+        in_channels: int,
+        out_channels: int,
+        expansion_ratio: int = 4,
+        kernel_size: int = 7,
+        use_residual_connection: bool = False,
+        norm_type: str = "group",
+        dim: str = "3d",
+        global_resp_norm: bool = False,
+    ):
+
+        super().__init__(
+            in_channels,
+            out_channels,
+            expansion_ratio,
+            kernel_size,
+            use_residual_connection=False,
+            norm_type=norm_type,
+            dim=dim,
+            global_resp_norm=global_resp_norm,
+        )
+
+        conv = get_conv_layer(spatial_dim=2 if dim == "2d" else 3)
+        self.resample_do_res = use_residual_connection
+        if use_residual_connection:
+            self.res_conv = conv(in_channels=in_channels, out_channels=out_channels, kernel_size=1, stride=2)
+
+        self.conv1 = conv(
+            in_channels=in_channels,
+            out_channels=in_channels,
+            kernel_size=kernel_size,
+            stride=2,
+            padding=kernel_size // 2,
+            groups=in_channels,
+        )
+
+    def forward(self, x):
+        """
+        Forward pass of the MedNeXtDownBlock.
+
+        Args:
+            x (torch.Tensor): Input tensor.
+
+        Returns:
+            torch.Tensor: Output tensor.
+        """
+        x1 = super().forward(x)
+
+        if self.resample_do_res:
+            res = self.res_conv(x)
+            x1 = x1 + res
+
+        return x1
+
+
+class MedNeXtUpBlock(MedNeXtBlock):
+    """
+    MedNeXtUpBlock class for upsampling in the MedNeXt model.
+
+    Args:
+        in_channels (int): Number of input channels.
+        out_channels (int): Number of output channels.
+        expansion_ratio (int): Expansion ratio for the block. Defaults to 4.
+        kernel_size (int): Kernel size for convolutions. Defaults to 7.
+        use_residual_connection (bool): Whether to use residual connection. Defaults to False.
+        norm_type (str): Type of normalization to use. Defaults to "group".
+        dim (str): Dimension of the input. Can be "2d" or "3d". Defaults to "3d".
+        global_resp_norm (bool): Whether to use global response normalization. Defaults to False.
+    """
+
+    def __init__(
+        self,
+        in_channels: int,
+        out_channels: int,
+        expansion_ratio: int = 4,
+        kernel_size: int = 7,
+        use_residual_connection: bool = False,
+        norm_type: str = "group",
+        dim: str = "3d",
+        global_resp_norm: bool = False,
+    ):
+        super().__init__(
+            in_channels,
+            out_channels,
+            expansion_ratio,
+            kernel_size,
+            use_residual_connection=False,
+            norm_type=norm_type,
+            dim=dim,
+            global_resp_norm=global_resp_norm,
+        )
+
+        self.resample_do_res = use_residual_connection
+
+        self.dim = dim
+        conv = get_conv_layer(spatial_dim=2 if dim == "2d" else 3, transpose=True)
+        if use_residual_connection:
+            self.res_conv = conv(in_channels=in_channels, out_channels=out_channels, kernel_size=1, stride=2)
+
+        self.conv1 = conv(
+            in_channels=in_channels,
+            out_channels=in_channels,
+            kernel_size=kernel_size,
+            stride=2,
+            padding=kernel_size // 2,
+            groups=in_channels,
+        )
+
+    def forward(self, x):
+        """
+        Forward pass of the MedNeXtUpBlock.
+
+        Args:
+            x (torch.Tensor): Input tensor.
+
+        Returns:
+            torch.Tensor: Output tensor.
+        """
+        x1 = super().forward(x)
+        # Asymmetry but necessary to match shape
+
+        if self.dim == "2d":
+            x1 = torch.nn.functional.pad(x1, (1, 0, 1, 0))
+        else:
+            x1 = torch.nn.functional.pad(x1, (1, 0, 1, 0, 1, 0))
+
+        if self.resample_do_res:
+            res = self.res_conv(x)
+            if self.dim == "2d":
+                res = torch.nn.functional.pad(res, (1, 0, 1, 0))
+            else:
+                res = torch.nn.functional.pad(res, (1, 0, 1, 0, 1, 0))
+            x1 = x1 + res
+
+        return x1
+
+
+class MedNeXtOutBlock(nn.Module):
+    """
+    MedNeXtOutBlock class for the output block in the MedNeXt model.
+
+    Args:
+        in_channels (int): Number of input channels.
+        n_classes (int): Number of output classes.
+        dim (str): Dimension of the input. Can be "2d" or "3d".
+    """
+
+    def __init__(self, in_channels, n_classes, dim):
+        super().__init__()
+
+        conv = get_conv_layer(spatial_dim=2 if dim == "2d" else 3, transpose=True)
+        self.conv_out = conv(in_channels, n_classes, kernel_size=1)
+
+    def forward(self, x):
+        """
+        Forward pass of the MedNeXtOutBlock.
+
+        Args:
+            x (torch.Tensor): Input tensor.
+
+        Returns:
+            torch.Tensor: Output tensor.
+        """
+        return self.conv_out(x)
diff --git a/monai/networks/nets/__init__.py b/monai/networks/nets/__init__.py
index 0570c9fcc1..b876e6a3fc 100644
--- a/monai/networks/nets/__init__.py
+++ b/monai/networks/nets/__init__.py
@@ -53,6 +53,25 @@
 from .generator import Generator
 from .highresnet import HighResBlock, HighResNet
 from .hovernet import Hovernet, HoVernet, HoVerNet, HoverNet
+from .mednext import (
+    MedNeXt,
+    MedNext,
+    MedNextB,
+    MedNeXtB,
+    MedNextBase,
+    MedNextL,
+    MedNeXtL,
+    MedNeXtLarge,
+    MedNextLarge,
+    MedNextM,
+    MedNeXtM,
+    MedNeXtMedium,
+    MedNextMedium,
+    MedNextS,
+    MedNeXtS,
+    MedNeXtSmall,
+    MedNextSmall,
+)
 from .milmodel import MILModel
 from .netadapter import NetAdapter
 from .patchgan_discriminator import MultiScalePatchDiscriminator, PatchDiscriminator
diff --git a/monai/networks/nets/mednext.py b/monai/networks/nets/mednext.py
new file mode 100644
index 0000000000..427572ba60
--- /dev/null
+++ b/monai/networks/nets/mednext.py
@@ -0,0 +1,354 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Portions of this code are derived from the original repository at:
+# https://github.com/MIC-DKFZ/MedNeXt
+# and are used under the terms of the Apache License, Version 2.0.
+
+from __future__ import annotations
+
+from collections.abc import Sequence
+
+import torch
+import torch.nn as nn
+
+from monai.networks.blocks.mednext_block import MedNeXtBlock, MedNeXtDownBlock, MedNeXtOutBlock, MedNeXtUpBlock
+
+__all__ = [
+    "MedNeXt",
+    "MedNeXtSmall",
+    "MedNeXtBase",
+    "MedNeXtMedium",
+    "MedNeXtLarge",
+    "MedNext",
+    "MedNextS",
+    "MedNeXtS",
+    "MedNextSmall",
+    "MedNextB",
+    "MedNeXtB",
+    "MedNextBase",
+    "MedNextM",
+    "MedNeXtM",
+    "MedNextMedium",
+    "MedNextL",
+    "MedNeXtL",
+    "MedNextLarge",
+]
+
+
+class MedNeXt(nn.Module):
+    """
+    MedNeXt model class from paper: https://arxiv.org/pdf/2303.09975
+
+    Args:
+        spatial_dims: spatial dimension of the input data. Defaults to 3.
+        init_filters: number of output channels for initial convolution layer. Defaults to 32.
+        in_channels: number of input channels for the network. Defaults to 1.
+        out_channels: number of output channels for the network. Defaults to 2.
+        encoder_expansion_ratio: expansion ratio for encoder blocks. Defaults to 2.
+        decoder_expansion_ratio: expansion ratio for decoder blocks. Defaults to 2.
+        bottleneck_expansion_ratio: expansion ratio for bottleneck blocks. Defaults to 2.
+        kernel_size: kernel size for convolutions. Defaults to 7.
+        deep_supervision: whether to use deep supervision. Defaults to False.
+        use_residual_connection: whether to use residual connections in standard, down and up blocks. Defaults to False.
+        blocks_down: number of blocks in each encoder stage. Defaults to [2, 2, 2, 2].
+        blocks_bottleneck: number of blocks in bottleneck stage. Defaults to 2.
+        blocks_up: number of blocks in each decoder stage. Defaults to [2, 2, 2, 2].
+        norm_type: type of normalization layer. Defaults to 'group'.
+        global_resp_norm: whether to use Global Response Normalization. Defaults to False. Refer: https://arxiv.org/abs/2301.00808
+    """
+
+    def __init__(
+        self,
+        spatial_dims: int = 3,
+        init_filters: int = 32,
+        in_channels: int = 1,
+        out_channels: int = 2,
+        encoder_expansion_ratio: Sequence[int] | int = 2,
+        decoder_expansion_ratio: Sequence[int] | int = 2,
+        bottleneck_expansion_ratio: int = 2,
+        kernel_size: int = 7,
+        deep_supervision: bool = False,
+        use_residual_connection: bool = False,
+        blocks_down: Sequence[int] = (2, 2, 2, 2),
+        blocks_bottleneck: int = 2,
+        blocks_up: Sequence[int] = (2, 2, 2, 2),
+        norm_type: str = "group",
+        global_resp_norm: bool = False,
+    ):
+        """
+        Initialize the MedNeXt model.
+
+        This method sets up the architecture of the model, including:
+        - Stem convolution
+        - Encoder stages and downsampling blocks
+        - Bottleneck blocks
+        - Decoder stages and upsampling blocks
+        - Output blocks for deep supervision (if enabled)
+        """
+        super().__init__()
+
+        self.do_ds = deep_supervision
+        assert spatial_dims in [2, 3], "`spatial_dims` can only be 2 or 3."
+        spatial_dims_str = f"{spatial_dims}d"
+        enc_kernel_size = dec_kernel_size = kernel_size
+
+        if isinstance(encoder_expansion_ratio, int):
+            encoder_expansion_ratio = [encoder_expansion_ratio] * len(blocks_down)
+
+        if isinstance(decoder_expansion_ratio, int):
+            decoder_expansion_ratio = [decoder_expansion_ratio] * len(blocks_up)
+
+        conv = nn.Conv2d if spatial_dims_str == "2d" else nn.Conv3d
+
+        self.stem = conv(in_channels, init_filters, kernel_size=1)
+
+        enc_stages = []
+        down_blocks = []
+
+        for i, num_blocks in enumerate(blocks_down):
+            enc_stages.append(
+                nn.Sequential(
+                    *[
+                        MedNeXtBlock(
+                            in_channels=init_filters * (2**i),
+                            out_channels=init_filters * (2**i),
+                            expansion_ratio=encoder_expansion_ratio[i],
+                            kernel_size=enc_kernel_size,
+                            use_residual_connection=use_residual_connection,
+                            norm_type=norm_type,
+                            dim=spatial_dims_str,
+                            global_resp_norm=global_resp_norm,
+                        )
+                        for _ in range(num_blocks)
+                    ]
+                )
+            )
+
+            down_blocks.append(
+                MedNeXtDownBlock(
+                    in_channels=init_filters * (2**i),
+                    out_channels=init_filters * (2 ** (i + 1)),
+                    expansion_ratio=encoder_expansion_ratio[i],
+                    kernel_size=enc_kernel_size,
+                    use_residual_connection=use_residual_connection,
+                    norm_type=norm_type,
+                    dim=spatial_dims_str,
+                )
+            )
+
+        self.enc_stages = nn.ModuleList(enc_stages)
+        self.down_blocks = nn.ModuleList(down_blocks)
+
+        self.bottleneck = nn.Sequential(
+            *[
+                MedNeXtBlock(
+                    in_channels=init_filters * (2 ** len(blocks_down)),
+                    out_channels=init_filters * (2 ** len(blocks_down)),
+                    expansion_ratio=bottleneck_expansion_ratio,
+                    kernel_size=dec_kernel_size,
+                    use_residual_connection=use_residual_connection,
+                    norm_type=norm_type,
+                    dim=spatial_dims_str,
+                    global_resp_norm=global_resp_norm,
+                )
+                for _ in range(blocks_bottleneck)
+            ]
+        )
+
+        up_blocks = []
+        dec_stages = []
+        for i, num_blocks in enumerate(blocks_up):
+            up_blocks.append(
+                MedNeXtUpBlock(
+                    in_channels=init_filters * (2 ** (len(blocks_up) - i)),
+                    out_channels=init_filters * (2 ** (len(blocks_up) - i - 1)),
+                    expansion_ratio=decoder_expansion_ratio[i],
+                    kernel_size=dec_kernel_size,
+                    use_residual_connection=use_residual_connection,
+                    norm_type=norm_type,
+                    dim=spatial_dims_str,
+                    global_resp_norm=global_resp_norm,
+                )
+            )
+
+            dec_stages.append(
+                nn.Sequential(
+                    *[
+                        MedNeXtBlock(
+                            in_channels=init_filters * (2 ** (len(blocks_up) - i - 1)),
+                            out_channels=init_filters * (2 ** (len(blocks_up) - i - 1)),
+                            expansion_ratio=decoder_expansion_ratio[i],
+                            kernel_size=dec_kernel_size,
+                            use_residual_connection=use_residual_connection,
+                            norm_type=norm_type,
+                            dim=spatial_dims_str,
+                            global_resp_norm=global_resp_norm,
+                        )
+                        for _ in range(num_blocks)
+                    ]
+                )
+            )
+
+        self.up_blocks = nn.ModuleList(up_blocks)
+        self.dec_stages = nn.ModuleList(dec_stages)
+
+        self.out_0 = MedNeXtOutBlock(in_channels=init_filters, n_classes=out_channels, dim=spatial_dims_str)
+
+        if deep_supervision:
+            out_blocks = [
+                MedNeXtOutBlock(in_channels=init_filters * (2**i), n_classes=out_channels, dim=spatial_dims_str)
+                for i in range(1, len(blocks_up) + 1)
+            ]
+
+            out_blocks.reverse()
+            self.out_blocks = nn.ModuleList(out_blocks)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor | Sequence[torch.Tensor]:
+        """
+        Forward pass of the MedNeXt model.
+
+        This method performs the forward pass through the model, including:
+        - Stem convolution
+        - Encoder stages and downsampling
+        - Bottleneck blocks
+        - Decoder stages and upsampling with skip connections
+        - Output blocks for deep supervision (if enabled)
+
+        Args:
+            x (torch.Tensor): Input tensor.
+
+        Returns:
+            torch.Tensor or Sequence[torch.Tensor]: Output tensor(s).
+        """
+        # Apply stem convolution
+        x = self.stem(x)
+
+        # Encoder forward pass
+        enc_outputs = []
+        for enc_stage, down_block in zip(self.enc_stages, self.down_blocks):
+            x = enc_stage(x)
+            enc_outputs.append(x)
+            x = down_block(x)
+
+        # Bottleneck forward pass
+        x = self.bottleneck(x)
+
+        # Initialize deep supervision outputs if enabled
+        if self.do_ds:
+            ds_outputs = []
+
+        # Decoder forward pass with skip connections
+        for i, (up_block, dec_stage) in enumerate(zip(self.up_blocks, self.dec_stages)):
+            if self.do_ds and i < len(self.out_blocks):
+                ds_outputs.append(self.out_blocks[i](x))
+
+            x = up_block(x)
+            x = x + enc_outputs[-(i + 1)]
+            x = dec_stage(x)
+
+        # Final output block
+        x = self.out_0(x)
+
+        # Return output(s)
+        if self.do_ds and self.training:
+            return (x, *ds_outputs[::-1])
+        else:
+            return x
+
+
+# Define the MedNeXt variants as reported in 10.48550/arXiv.2303.09975
+def create_mednext(
+    variant: str,
+    spatial_dims: int = 3,
+    in_channels: int = 1,
+    out_channels: int = 2,
+    kernel_size: int = 3,
+    deep_supervision: bool = False,
+) -> MedNeXt:
+    """
+    Factory method to create MedNeXt variants.
+
+    Args:
+        variant (str): The MedNeXt variant to create ('S', 'B', 'M', or 'L').
+        spatial_dims (int): Number of spatial dimensions. Defaults to 3.
+        in_channels (int): Number of input channels. Defaults to 1.
+        out_channels (int): Number of output channels. Defaults to 2.
+        kernel_size (int): Kernel size for convolutions. Defaults to 3.
+        deep_supervision (bool): Whether to use deep supervision. Defaults to False.
+
+    Returns:
+        MedNeXt: The specified MedNeXt variant.
+
+    Raises:
+        ValueError: If an invalid variant is specified.
+    """
+    common_args = {
+        "spatial_dims": spatial_dims,
+        "in_channels": in_channels,
+        "out_channels": out_channels,
+        "kernel_size": kernel_size,
+        "deep_supervision": deep_supervision,
+        "use_residual_connection": True,
+        "norm_type": "group",
+        "global_resp_norm": False,
+        "init_filters": 32,
+    }
+
+    if variant.upper() == "S":
+        return MedNeXt(
+            encoder_expansion_ratio=2,
+            decoder_expansion_ratio=2,
+            bottleneck_expansion_ratio=2,
+            blocks_down=(2, 2, 2, 2),
+            blocks_bottleneck=2,
+            blocks_up=(2, 2, 2, 2),
+            **common_args,  # type: ignore
+        )
+    elif variant.upper() == "B":
+        return MedNeXt(
+            encoder_expansion_ratio=(2, 3, 4, 4),
+            decoder_expansion_ratio=(4, 4, 3, 2),
+            bottleneck_expansion_ratio=4,
+            blocks_down=(2, 2, 2, 2),
+            blocks_bottleneck=2,
+            blocks_up=(2, 2, 2, 2),
+            **common_args,  # type: ignore
+        )
+    elif variant.upper() == "M":
+        return MedNeXt(
+            encoder_expansion_ratio=(2, 3, 4, 4),
+            decoder_expansion_ratio=(4, 4, 3, 2),
+            bottleneck_expansion_ratio=4,
+            blocks_down=(3, 4, 4, 4),
+            blocks_bottleneck=4,
+            blocks_up=(4, 4, 4, 3),
+            **common_args,  # type: ignore
+        )
+    elif variant.upper() == "L":
+        return MedNeXt(
+            encoder_expansion_ratio=(3, 4, 8, 8),
+            decoder_expansion_ratio=(8, 8, 4, 3),
+            bottleneck_expansion_ratio=8,
+            blocks_down=(3, 4, 8, 8),
+            blocks_bottleneck=8,
+            blocks_up=(8, 8, 4, 3),
+            **common_args,  # type: ignore
+        )
+    else:
+        raise ValueError(f"Invalid MedNeXt variant: {variant}")
+
+
+MedNext = MedNeXt
+MedNextS = MedNeXtS = MedNextSmall = MedNeXtSmall = lambda **kwargs: create_mednext("S", **kwargs)
+MedNextB = MedNeXtB = MedNextBase = MedNeXtBase = lambda **kwargs: create_mednext("B", **kwargs)
+MedNextM = MedNeXtM = MedNextMedium = MedNeXtMedium = lambda **kwargs: create_mednext("M", **kwargs)
+MedNextL = MedNeXtL = MedNextLarge = MedNeXtLarge = lambda **kwargs: create_mednext("L", **kwargs)
diff --git a/tests/test_mednext.py b/tests/test_mednext.py
new file mode 100644
index 0000000000..b4ba4f9939
--- /dev/null
+++ b/tests/test_mednext.py
@@ -0,0 +1,122 @@
+# Copyright (c) MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import unittest
+
+import torch
+from parameterized import parameterized
+
+from monai.networks import eval_mode
+from monai.networks.nets import MedNeXt, MedNeXtL, MedNeXtM, MedNeXtS
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+TEST_CASE_MEDNEXT = []
+for spatial_dims in range(2, 4):
+    for init_filters in [8, 16]:
+        for deep_supervision in [False, True]:
+            for do_res in [False, True]:
+                test_case = [
+                    {
+                        "spatial_dims": spatial_dims,
+                        "init_filters": init_filters,
+                        "deep_supervision": deep_supervision,
+                        "use_residual_connection": do_res,
+                    },
+                    (2, 1, *([16] * spatial_dims)),
+                    (2, 2, *([16] * spatial_dims)),
+                ]
+                TEST_CASE_MEDNEXT.append(test_case)
+
+TEST_CASE_MEDNEXT_2 = []
+for spatial_dims in range(2, 4):
+    for out_channels in [1, 2]:
+        for deep_supervision in [False, True]:
+            test_case = [
+                {
+                    "spatial_dims": spatial_dims,
+                    "init_filters": 8,
+                    "out_channels": out_channels,
+                    "deep_supervision": deep_supervision,
+                },
+                (2, 1, *([16] * spatial_dims)),
+                (2, out_channels, *([16] * spatial_dims)),
+            ]
+            TEST_CASE_MEDNEXT_2.append(test_case)
+
+TEST_CASE_MEDNEXT_VARIANTS = []
+for model in [MedNeXtS, MedNeXtM, MedNeXtL]:
+    for spatial_dims in range(2, 4):
+        for out_channels in [1, 2]:
+            test_case = [
+                model,  # type: ignore
+                {"spatial_dims": spatial_dims, "in_channels": 1, "out_channels": out_channels},
+                (2, 1, *([16] * spatial_dims)),
+                (2, out_channels, *([16] * spatial_dims)),
+            ]
+            TEST_CASE_MEDNEXT_VARIANTS.append(test_case)
+
+
+class TestMedNeXt(unittest.TestCase):
+
+    @parameterized.expand(TEST_CASE_MEDNEXT)
+    def test_shape(self, input_param, input_shape, expected_shape):
+        net = MedNeXt(**input_param).to(device)
+        with eval_mode(net):
+            result = net(torch.randn(input_shape).to(device))
+            if input_param["deep_supervision"] and net.training:
+                assert isinstance(result, tuple)
+                self.assertEqual(result[0].shape, expected_shape, msg=str(input_param))
+            else:
+                self.assertEqual(result.shape, expected_shape, msg=str(input_param))
+
+    @parameterized.expand(TEST_CASE_MEDNEXT_2)
+    def test_shape2(self, input_param, input_shape, expected_shape):
+        net = MedNeXt(**input_param).to(device)
+
+        net.train()
+        result = net(torch.randn(input_shape).to(device))
+        if input_param["deep_supervision"]:
+            assert isinstance(result, tuple)
+            self.assertEqual(result[0].shape, expected_shape, msg=str(input_param))
+        else:
+            assert isinstance(result, torch.Tensor)
+            self.assertEqual(result.shape, expected_shape, msg=str(input_param))
+
+        net.eval()
+        result = net(torch.randn(input_shape).to(device))
+        assert isinstance(result, torch.Tensor)
+        self.assertEqual(result.shape, expected_shape, msg=str(input_param))
+
+    def test_ill_arg(self):
+        with self.assertRaises(AssertionError):
+            MedNeXt(spatial_dims=4)
+
+    @parameterized.expand(TEST_CASE_MEDNEXT_VARIANTS)
+    def test_mednext_variants(self, model, input_param, input_shape, expected_shape):
+        net = model(**input_param).to(device)
+
+        net.train()
+        result = net(torch.randn(input_shape).to(device))
+        assert isinstance(result, torch.Tensor)
+        self.assertEqual(result.shape, expected_shape, msg=str(input_param))
+
+        net.eval()
+        with torch.no_grad():
+            result = net(torch.randn(input_shape).to(device))
+        assert isinstance(result, torch.Tensor)
+        self.assertEqual(result.shape, expected_shape, msg=str(input_param))
+
+
+if __name__ == "__main__":
+    unittest.main()