[Model] LSTM

README.md

@@ -264,6 +264,7 @@ agent group. Here is a table of the models implemented in BenchMARL
 |------------------------------------------|:-------------:|:-----------------------------:|:-----------------------------:|
 | [MLP](benchmarl/models/mlp.py)           |      Yes      |              Yes              |              Yes              |
 | [GRU](benchmarl/models/gru.py)           |      Yes      |              Yes              |              Yes              |
+| [LSTM](benchmarl/models/lstm.py)         |      Yes      |              Yes              |              Yes              |
 | [GNN](benchmarl/models/gnn.py)           |      Yes      |              Yes              |              No               |
 | [CNN](benchmarl/models/cnn.py)           |      Yes      |              Yes              |              Yes              |
 | [Deepsets](benchmarl/models/deepsets.py) |      Yes      |              Yes              |              Yes              |

benchmarl/conf/model/layers/lstm.yaml

@@ -0,0 +1,15 @@
+
+name: lstm
+
+hidden_size: 128
+n_layers: 1
+bias: True
+dropout: 0
+compile: False
+
+mlp_num_cells: [256, 256]
+mlp_layer_class: torch.nn.Linear
+mlp_activation_class: torch.nn.Tanh
+mlp_activation_kwargs: null
+mlp_norm_class: null
+mlp_norm_kwargs: null

benchmarl/models/__init__.py

@@ -9,6 +9,7 @@
 from .deepsets import Deepsets, DeepsetsConfig
 from .gnn import Gnn, GnnConfig
 from .gru import Gru, GruConfig
+from .lstm import Lstm, LstmConfig
 from .mlp import Mlp, MlpConfig
 
 classes = [
@@ -22,6 +23,8 @@
     "DeepsetsConfig",
     "Gru",
     "GruConfig",
+    "Lstm",
+    "LstmConfig",
 ]
 
 model_config_registry = {
@@ -30,4 +33,5 @@
     "cnn": CnnConfig,
     "deepsets": DeepsetsConfig,
     "gru": GruConfig,
+    "lstm": LstmConfig,
 }

benchmarl/models/gru.py

@@ -4,12 +4,6 @@
 #  LICENSE file in the root directory of this source tree.
 #
 
-#  Copyright (c) Meta Platforms, Inc. and affiliates.
-#
-#  This source code is licensed under the license found in the
-#  LICENSE file in the root directory of this source tree.
-#
-
 from __future__ import annotations
 
 from dataclasses import dataclass, MISSING
@@ -167,7 +161,7 @@ def forward(
         h_0=None,
     ):
         # Input and output always have the multiagent dimension
-        # Hidden state only has it when not centralised
+        # Hidden states always have it apart from when it is centralized and share params
         # is_init never has it
 
         assert is_init is not None, "We need to pass is_init"
@@ -202,7 +196,7 @@ def forward(
         is_init = is_init.unsqueeze(-2).expand(batch, seq, self.n_agents, 1)
 
         if h_0 is None:
-            if self.centralised:
+            if self.centralised and self.share_params:
                 shape = (
                     batch,
                     self.n_layers,
@@ -243,8 +237,8 @@ def run_net(self, input, is_init, h_0):
             if self.centralised:
                 output, h_n = self.vmap_func_module(
                     self._empty_gru,
-                    (0, None, None, None),
-                    (-2, -2),
+                    (0, None, None, -3),
+                    (-2, -3),
                 )(self.params, input, is_init, h_0)
             else:
                 output, h_n = self.vmap_func_module(
@@ -283,8 +277,8 @@ class Gru(Model):
 
     The BenchMARL GRU accepts multiple inputs of type array: Tensors of shape ``(*batch,F)``
 
-    Where `F` is the number of features.
-    The features `F` will be processed to features of `hidden_size` by the GRU.
+    Where `F` is the number of features. These arrays will be concatenated along the F dimensions,
+    which will be processed to features of `hidden_size` by the GRU.
 
     Args:
         hidden_size (int): The number of features in the hidden state.
@@ -516,10 +510,9 @@ def is_rnn(self) -> bool:
         return True
 
     def get_model_state_spec(self, model_index: int = 0) -> CompositeSpec:
-        name = f"_hidden_gru_{model_index}"
         spec = CompositeSpec(
             {
-                name: UnboundedContinuousTensorSpec(
+                f"_hidden_gru_{model_index}": UnboundedContinuousTensorSpec(
                     shape=(self.n_layers, self.hidden_size)
                 )
             }