run_cut.py

#! usr/bin/env python3
# -*- coding:utf-8 -*-
"""
Copyright 2018 The Google AI Language Team Authors.
BASED ON Google_BERT.
@Author:jiangpinglei
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import collections
import os
import modeling
import optimization
import tokenization
import tensorflow as tf
import pickle
from evaluation import SegmenterEvaluation

flags = tf.flags

FLAGS = flags.FLAGS

flags.DEFINE_string(
    "data_dir", None,
    "The input datadir.",
)

flags.DEFINE_string(
    "bert_config_file", None,
    "The config json file corresponding to the pre-trained BERT model."
)

flags.DEFINE_string(
    "task_name", None, "The name of the task to train."
)

flags.DEFINE_string(
    "output_dir", None,
    "The output directory where the model checkpoints will be written."
)

## Other parameters
flags.DEFINE_string(
    "init_checkpoint", None,
    "Initial checkpoint (usually from a pre-trained BERT model)."
)

flags.DEFINE_bool(
    "do_lower_case", True,
    "Whether to lower case the input text."
)

flags.DEFINE_integer(
    "max_seq_length", 128,
    "The maximum total input sequence length after WordPiece tokenization."
)

flags.DEFINE_bool(
    "do_train", False,
    "Whether to run training."
)
flags.DEFINE_bool("use_tpu", False, "Whether to use TPU or GPU/CPU.")

flags.DEFINE_bool("do_eval", False, "Whether to run eval on the dev set.")

flags.DEFINE_bool("do_predict", False, "Whether to run the model in inference mode on the test set.")

flags.DEFINE_integer("train_batch_size", 32, "Total batch size for training.")

flags.DEFINE_integer("eval_batch_size", 8, "Total batch size for eval.")

flags.DEFINE_integer("predict_batch_size", 8, "Total batch size for predict.")

flags.DEFINE_float("learning_rate", 5e-5, "The initial learning rate for Adam.")

flags.DEFINE_float("num_train_epochs", 3.0, "Total number of training epochs to perform.")

flags.DEFINE_float(
    "warmup_proportion", 0.1,
    "Proportion of training to perform linear learning rate warmup for. "
    "E.g., 0.1 = 10% of training.")

flags.DEFINE_integer("save_checkpoints_steps", 1000,
                     "How often to save the model checkpoint.")

flags.DEFINE_integer("iterations_per_loop", 1000,
                     "How many steps to make in each estimator call.")

flags.DEFINE_string("vocab_file", None,
                    "The vocabulary file that the BERT model was trained on.")
tf.flags.DEFINE_string("master", None, "[Optional] TensorFlow master URL.")
flags.DEFINE_integer(
    "num_tpu_cores", 8,
    "Only used if `use_tpu` is True. Total number of TPU cores to use.")


class InputExample(object):
    """A single training/test example for simple sequence classification."""

    def __init__(self, guid, text, label=None):
        """Constructs a InputExample.

        Args:
          guid: Unique id for the example.
          text_a: string. The untokenized text of the first sequence. For single
            sequence tasks, only this sequence must be specified.
          label: (Optional) string. The label of the example. This should be
            specified for train and dev examples, but not for test examples.
        """
        self.guid = guid
        self.text = text
        self.label = label


class InputFeatures(object):
    """A single set of features of data."""

    def __init__(self, input_ids, input_mask, segment_ids, label_ids, ):
        self.input_ids = input_ids
        self.input_mask = input_mask
        self.segment_ids = segment_ids
        self.label_ids = label_ids
        # self.label_mask = label_mask


class DataProcessor(object):
    """Base class for data converters for sequence classification data sets."""

    def get_train_examples(self, data_dir):
        """Gets a collection of `InputExample`s for the train set."""
        raise NotImplementedError()

    def get_dev_examples(self, data_dir):
        """Gets a collection of `InputExample`s for the dev set."""
        raise NotImplementedError()

    def get_labels(self):
        """Gets the list of labels for this data set."""
        raise NotImplementedError()

    @classmethod
    def _read_data(cls, input_file):
        """Reads a BIO data."""

        def seg_long_sentence(words, labels):

            def tough_seg(words, labels):
                words_list = []
                labels_list = []
                while len(words) >  FLAGS.max_seq_length-2:
                    words_list.append(words[0:FLAGS.max_seq_length-2])
                    labels_list.append(labels[0:FLAGS.max_seq_length-2])
                    words = words[FLAGS.max_seq_length-2:]
                    labels = labels[FLAGS.max_seq_length-2:]
                if len(words) > 0:
                    words_list.append(words)
                    labels_list.append(labels)
                return words_list, labels_list

            words_result_list = []
            labels_result_list = []
            tmp = words.split("，")
            start = 0
            for each in tmp:
                if len(each) == 0:
                    start += 1
                elif len(each) <= FLAGS.max_seq_length - 2:
                    words_tmp_list, labels_tmp_list = tough_seg(each, labels[start: start + len(each)])
                    words_result_list.extend(words_tmp_list)
                    labels_result_list.extend(labels_tmp_list)
                    start = start + len(each) + 1
                else:
                    words_result_list.append(each)
                    labels_result_list.append(labels[start: start + len(each)])
                    start = start + len(each) + 1
            return words_result_list, labels_result_list
        lines = []
        with open(input_file, "r", encoding="utf-8") as f:
            for line in f.readlines():
                line = line.strip().split("\t")
                if len(line) != 2:
                    continue
                words = line[0]
                labels = line[1]
                if len(words) > FLAGS.max_seq_length-2:
                    words_list, labels_list = seg_long_sentence(words, labels)
                    for each_words, each_labels in zip(words_list, labels_list):
                        lines.append([each_words, each_labels])
                else:
                    lines.append([words, labels])
            return lines


class CutProcessor(DataProcessor):
    def get_train_examples(self, data_dir):
        return self._create_example(
            self._read_data(os.path.join(data_dir, "train.txt")), "train"
        )

    def get_dev_examples(self, data_dir):
        return self._create_example(
            self._read_data(os.path.join(data_dir, "dev.txt")), "dev"
        )

    def get_test_examples(self, data_dir):
        return self._create_example(
            self._read_data(os.path.join(data_dir, "test.txt")), "test")

    def get_labels(self):
        return ["s", "b", "m", "e", "X", "[CLS]", "[SEP]"]

    def _create_example(self, lines, set_type):
        examples = []
        for (i, line) in enumerate(lines):
            guid = "%s-%s" % (set_type, i)
            text = tokenization.convert_to_unicode(line[0])
            label = tokenization.convert_to_unicode(line[1])
            assert len(text) == len(label)
            examples.append(InputExample(guid=guid, text=text, label=label))
        return examples


def write_tokens(tokens, mode):
    if mode == "test":
        path = os.path.join(FLAGS.output_dir, "token_" + mode + ".txt")
        wf = open(path, 'a')
        lines = ""
        for i in range(1, len(tokens)):
            if tokens[i] == "[SEP]":
                break
            lines += tokens[i]
        wf.write(lines + '\n')
        wf.close()

def output_seg_result(output_dir, ori_labels, des_labels):
    token_file_path = os.sep.join([output_dir, "token_test.txt"])
    with open(token_file_path, "r", encoding="utf-8") as f:
        tokens = [each.strip() for each in f.readlines()]
    lenth = len(tokens)
    assert len(ori_labels) == lenth
    assert len(des_labels) == lenth
    seg_result_file = os.path.join(output_dir, "seg_result.txt")
    writer = open(seg_result_file, "w", encoding="utf-8")
    for i in range(lenth):
        token = tokens[i]
        ori_label = ori_labels[i]
        des_label = des_labels[i]
        ori_seg = " ".join([token[each[0]:each[1]] for each in ori_label])
        des_seg = " ".join([token[each[0]:each[1]] for each in des_label])
        writer.write(token + "\n")
        writer.write(ori_seg + "\n")
        writer.write(des_seg + "\n")
        writer.write("\n")
    writer.close()

def convert_single_example(ex_index, example, label_list, max_seq_length, tokenizer,mode):
    label_map = {}
    for (i, label) in enumerate(label_list,1):
        label_map[label] = i
    label2idpath = './output/label2id.pkl'
    if not os.path.exists(label2idpath):
        with open(label2idpath,'wb') as w:
            pickle.dump(label_map, w)

    textlist = list(example.text)
    labellist = list(example.label)
    tokens = []
    labels = []
    unknow_index = [] #记录下标记为[UNK]在textlist当中的位置，用于还原
    for i, word in enumerate(textlist):
        token = tokenizer.tokenize(word)
        tokens.extend(token)
        label_1 = labellist[i]
        for m in range(len(token)):
            if m == 0:
                labels.append(label_1)
            else:
                labels.append("X")
            if token[m] == "[UNK]":
                unknow_index.append(i)
    assert len(tokens) == len(labels)
    if len(tokens) >= max_seq_length - 1:
        tokens = tokens[0:(max_seq_length - 2)]
        labels = labels[0:(max_seq_length - 2)]
    ntokens = []
    segment_ids = []
    label_ids = []
    ntokens.append("[CLS]")
    segment_ids.append(0)
    label_ids.append(label_map["[CLS]"])
    for i, token in enumerate(tokens):
        ntokens.append(token)
        segment_ids.append(0)
        label_ids.append(label_map[labels[i]])
    ntokens.append("[SEP]")
    segment_ids.append(0)
    label_ids.append(label_map["[SEP]"])
    input_ids = tokenizer.convert_tokens_to_ids(ntokens)
    input_mask = [1] * len(input_ids)

    while len(input_ids) < max_seq_length:
        input_ids.append(0)
        input_mask.append(0)
        segment_ids.append(0)
        label_ids.append(0)
        ntokens.append("**NULL**")

    assert len(input_ids) == max_seq_length
    assert len(input_mask) == max_seq_length
    assert len(segment_ids) == max_seq_length
    assert len(label_ids) == max_seq_length


    if ex_index < 5:
        tf.logging.info("*** Example ***")
        tf.logging.info("guid: %s" % (example.guid))
        tf.logging.info("tokens: %s" % " ".join(
            [tokenization.printable_text(x) for x in tokens]))
        tf.logging.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
        tf.logging.info("input_mask: %s" % " ".join([str(x) for x in input_mask]))
        tf.logging.info("segment_ids: %s" % " ".join([str(x) for x in segment_ids]))
        tf.logging.info("label_ids: %s" % " ".join([str(x) for x in label_ids]))

    feature = InputFeatures(
        input_ids=input_ids,
        input_mask=input_mask,
        segment_ids=segment_ids,
        label_ids=label_ids,
    )
    # 还原[UNK]的数据
    output_tokens = []
    for i, each in enumerate(ntokens):
        if each != "[UNK]":
            output_tokens.append(each)
        else:
            index = unknow_index[0]
            output_tokens.append(textlist[index])
            unknow_index = unknow_index[1:]
    write_tokens(output_tokens, mode)
    return feature


def filed_based_convert_examples_to_features(
        examples, label_list, max_seq_length, tokenizer, output_file, mode=None
):
    writer = tf.python_io.TFRecordWriter(output_file)
    for (ex_index, example) in enumerate(examples):
        if ex_index % 5000 == 0:
            tf.logging.info("Writing example %d of %d" % (ex_index, len(examples)))

        feature = convert_single_example(ex_index, example, label_list, max_seq_length, tokenizer, mode)

        def create_int_feature(values):
            f = tf.train.Feature(int64_list=tf.train.Int64List(value=list(values)))
            return f

        features = collections.OrderedDict()
        features["input_ids"] = create_int_feature(feature.input_ids)
        features["input_mask"] = create_int_feature(feature.input_mask)
        features["segment_ids"] = create_int_feature(feature.segment_ids)
        features["label_ids"] = create_int_feature(feature.label_ids)
        tf_example = tf.train.Example(features=tf.train.Features(feature=features))
        writer.write(tf_example.SerializeToString())


def file_based_input_fn_builder(input_file, seq_length, is_training, drop_remainder):
    name_to_features = {
        "input_ids": tf.FixedLenFeature([seq_length], tf.int64),
        "input_mask": tf.FixedLenFeature([seq_length], tf.int64),
        "segment_ids": tf.FixedLenFeature([seq_length], tf.int64),
        "label_ids": tf.FixedLenFeature([seq_length], tf.int64),
    }

    def _decode_record(record, name_to_features):
        example = tf.parse_single_example(record, name_to_features)
        for name in list(example.keys()):
            t = example[name]
            if t.dtype == tf.int64:
                t = tf.to_int32(t)
            example[name] = t
        return example

    def input_fn(params):
        batch_size = params["batch_size"]
        d = tf.data.TFRecordDataset(input_file)
        if is_training:
            d = d.repeat()
            d = d.shuffle(buffer_size=100)
        d = d.apply(tf.contrib.data.map_and_batch(
            lambda record: _decode_record(record, name_to_features),
            batch_size=batch_size,
            drop_remainder=drop_remainder
        ))
        return d

    return input_fn


def create_model(bert_config, is_training, input_ids, input_mask,
                 segment_ids, labels, num_labels, use_one_hot_embeddings):
    model = modeling.BertModel(
        config=bert_config,
        is_training=is_training,
        input_ids=input_ids,
        input_mask=input_mask,
        token_type_ids=segment_ids,
        use_one_hot_embeddings=use_one_hot_embeddings
    )

    output_layer = model.get_sequence_output()

    hidden_size = output_layer.shape[-1].value

    output_weight = tf.get_variable(
        "output_weights", [num_labels, hidden_size],
        initializer=tf.truncated_normal_initializer(stddev=0.02)
    )
    output_bias = tf.get_variable(
        "output_bias", [num_labels], initializer=tf.zeros_initializer()
    )
    with tf.variable_scope("loss"):
        if is_training:
            output_layer = tf.nn.dropout(output_layer, keep_prob=0.9)
        output_layer = tf.reshape(output_layer, [-1, hidden_size])
        logits = tf.matmul(output_layer, output_weight, transpose_b=True)
        logits = tf.nn.bias_add(logits, output_bias)
        logits = tf.reshape(logits, [-1, FLAGS.max_seq_length, num_labels])
        log_probs = tf.nn.log_softmax(logits, axis=-1)
        one_hot_labels = tf.one_hot(labels, depth=num_labels, dtype=tf.float32)
        per_example_loss = -tf.reduce_sum(one_hot_labels * log_probs, axis=-1)
        loss = tf.reduce_sum(per_example_loss)
        probabilities = tf.nn.softmax(logits, axis=-1)
        predict = tf.argmax(probabilities, axis=-1)
        return (loss, per_example_loss, logits, predict)


def model_fn_builder(bert_config, num_labels, init_checkpoint, learning_rate,
                     num_train_steps, num_warmup_steps, use_tpu,
                     use_one_hot_embeddings):
    def model_fn(features, labels, mode, params):
        tf.logging.info("*** Features ***")
        for name in sorted(features.keys()):
            tf.logging.info("  name = %s, shape = %s" % (name, features[name].shape))
        input_ids = features["input_ids"]
        input_mask = features["input_mask"]
        segment_ids = features["segment_ids"]
        label_ids = features["label_ids"]
        is_training = (mode == tf.estimator.ModeKeys.TRAIN)

        (total_loss, per_example_loss, logits, predicts) = create_model(
            bert_config, is_training, input_ids, input_mask, segment_ids, label_ids,
            num_labels, use_one_hot_embeddings)
        tvars = tf.trainable_variables()
        scaffold_fn = None
        if init_checkpoint:
            (assignment_map, initialized_variable_names) = modeling.get_assignment_map_from_checkpoint(tvars,
                                                                                                       init_checkpoint)
            tf.train.init_from_checkpoint(init_checkpoint, assignment_map)
            if use_tpu:
                def tpu_scaffold():
                    tf.train.init_from_checkpoint(init_checkpoint, assignment_map)
                    return tf.train.Scaffold()

                scaffold_fn = tpu_scaffold
            else:
                tf.train.init_from_checkpoint(init_checkpoint, assignment_map)
        tf.logging.info("**** Trainable Variables ****")

        for var in tvars:
            init_string = ""
            if var.name in initialized_variable_names:
                init_string = ", *INIT_FROM_CKPT*"
            tf.logging.info("  name = %s, shape = %s%s", var.name, var.shape,
                            init_string)
        output_spec = None
        if mode == tf.estimator.ModeKeys.TRAIN:
            train_op = optimization.create_optimizer(
                total_loss, learning_rate, num_train_steps, num_warmup_steps, use_tpu)
            output_spec = tf.contrib.tpu.TPUEstimatorSpec(
                mode=mode,
                loss=total_loss,
                train_op=train_op,
                scaffold_fn=scaffold_fn)

        elif mode == tf.estimator.ModeKeys.PREDICT:
            predictions = {
                "label_ids": label_ids,
                "predicts": predicts
            }
            output_spec = tf.contrib.tpu.TPUEstimatorSpec(
                mode=mode, predictions=predictions, scaffold_fn=scaffold_fn)

        else:
            raise ValueError(
                "Only TRAIN and PREDICT modes are supported: %s" % (mode))
        return output_spec

    return model_fn


def main(_):
    tf.logging.set_verbosity(tf.logging.INFO)
    processors = {
        "people": CutProcessor
    }
    if not FLAGS.do_train and not FLAGS.do_predict:
        raise ValueError("At least one of `do_train` or `do_predict` must be True.")

    bert_config = modeling.BertConfig.from_json_file(FLAGS.bert_config_file)

    if FLAGS.max_seq_length > bert_config.max_position_embeddings:
        raise ValueError(
            "Cannot use sequence length %d because the BERT model "
            "was only trained up to sequence length %d" %
            (FLAGS.max_seq_length, bert_config.max_position_embeddings))

    tf.gfile.MakeDirs(FLAGS.output_dir)

    task_name = FLAGS.task_name.lower()
    if task_name not in processors:
        raise ValueError("Task not found: %s" % (task_name))
    processor = processors[task_name]()

    label_list = processor.get_labels()

    tokenizer = tokenization.FullTokenizer(
        vocab_file=FLAGS.vocab_file, do_lower_case=FLAGS.do_lower_case)
    tpu_cluster_resolver = None
    if FLAGS.use_tpu and FLAGS.tpu_name:
        tpu_cluster_resolver = tf.contrib.cluster_resolver.TPUClusterResolver(
            FLAGS.tpu_name, zone=FLAGS.tpu_zone, project=FLAGS.gcp_project)

    is_per_host = tf.contrib.tpu.InputPipelineConfig.PER_HOST_V2

    run_config = tf.contrib.tpu.RunConfig(
        cluster=tpu_cluster_resolver,
        master=FLAGS.master,
        model_dir=FLAGS.output_dir,
        save_checkpoints_steps=FLAGS.save_checkpoints_steps,
        tpu_config=tf.contrib.tpu.TPUConfig(
            iterations_per_loop=FLAGS.iterations_per_loop,
            num_shards=FLAGS.num_tpu_cores,
            per_host_input_for_training=is_per_host))

    train_examples = None
    num_train_steps = None
    num_warmup_steps = None

    if FLAGS.do_train:
        train_examples = processor.get_train_examples(FLAGS.data_dir)
        num_train_steps = int(
            len(train_examples) / FLAGS.train_batch_size * FLAGS.num_train_epochs)
        num_warmup_steps = int(num_train_steps * FLAGS.warmup_proportion)

    model_fn = model_fn_builder(
        bert_config=bert_config,
        num_labels=len(label_list) + 1,
        init_checkpoint=FLAGS.init_checkpoint,
        learning_rate=FLAGS.learning_rate,
        num_train_steps=num_train_steps,
        num_warmup_steps=num_warmup_steps,
        use_tpu=FLAGS.use_tpu,
        use_one_hot_embeddings=FLAGS.use_tpu)

    estimator = tf.contrib.tpu.TPUEstimator(
        use_tpu=FLAGS.use_tpu,
        model_fn=model_fn,
        config=run_config,
        train_batch_size=FLAGS.train_batch_size,
        eval_batch_size=FLAGS.eval_batch_size,
        predict_batch_size=FLAGS.predict_batch_size)

    if FLAGS.do_train:
        train_file = os.path.join(FLAGS.output_dir, "train.tf_record")
        filed_based_convert_examples_to_features(
            train_examples, label_list, FLAGS.max_seq_length, tokenizer, train_file)
        tf.logging.info("***** Running training *****")
        tf.logging.info("  Num examples = %d", len(train_examples))
        tf.logging.info("  Batch size = %d", FLAGS.train_batch_size)
        tf.logging.info("  Num steps = %d", num_train_steps)
        train_input_fn = file_based_input_fn_builder(
            input_file=train_file,
            seq_length=FLAGS.max_seq_length,
            is_training=True,
            drop_remainder=True)
        estimator.train(input_fn=train_input_fn, max_steps=num_train_steps)

    if FLAGS.do_predict:
        token_path = os.path.join(FLAGS.output_dir, "token_test.txt")
        with open('./output/label2id.pkl', 'rb') as rf:
            label2id = pickle.load(rf)
            id2label = {value: key for key, value in label2id.items()}
        if os.path.exists(token_path):
            os.remove(token_path)
        predict_examples = processor.get_test_examples(FLAGS.data_dir)

        predict_file = os.path.join(FLAGS.output_dir, "predict.tf_record")
        filed_based_convert_examples_to_features(predict_examples, label_list,
                                                 FLAGS.max_seq_length, tokenizer,
                                                 predict_file, mode="test")

        tf.logging.info("***** Running prediction*****")
        tf.logging.info("  Num examples = %d", len(predict_examples))
        tf.logging.info("  Batch size = %d", FLAGS.predict_batch_size)
        if FLAGS.use_tpu:
            # Warning: According to tpu_estimator.py Prediction on TPU is an
            # experimental feature and hence not supported here
            raise ValueError("Prediction in TPU not supported")
        predict_drop_remainder = True if FLAGS.use_tpu else False
        predict_input_fn = file_based_input_fn_builder(
            input_file=predict_file,
            seq_length=FLAGS.max_seq_length,
            is_training=False,
            drop_remainder=predict_drop_remainder)
        result = estimator.predict(input_fn=predict_input_fn)

        seg_eval = SegmenterEvaluation(id2label)
        count = 0
        precision_avg = 0
        recall_avg = 0
        f1_avg = 0
        error_avg = 0
        ori_labels, des_labels = [], []
        for each in result:
            if count % 1000 == 0:
                tf.logging.info("Processing example: %d" % (count))
            label_ids = each["label_ids"]
            predicts = each["predicts"]
            precision, recall, f1, error, right, predict = seg_eval.evaluate(label_ids, predicts)
            precision_avg = (precision_avg * count + precision) / (count + 1)
            recall_avg = (recall_avg * count + recall) / (count + 1)
            f1_avg = (f1_avg * count + f1) / (count + 1)
            error_avg = (error_avg * count + error) / (count + 1)
            ori_labels.append(right)
            des_labels.append(predict)
            count += 1

        tf.logging.info("***** Eval results *****")
        tf.logging.info("  count = %s", str(count))
        tf.logging.info("  precision_avg = %s", str(precision_avg))
        tf.logging.info("  recall_avg = %s", str(recall_avg))
        tf.logging.info("  f1_avg = %s", str(f1_avg))
        tf.logging.info("  error_avg = %s", str(error_avg))

        output_seg_result(FLAGS.output_dir, ori_labels, des_labels)


if __name__ == "__main__":
    flags.mark_flag_as_required("data_dir")
    flags.mark_flag_as_required("task_name")
    flags.mark_flag_as_required("vocab_file")
    flags.mark_flag_as_required("bert_config_file")
    flags.mark_flag_as_required("output_dir")
    tf.app.run()