run_training.py

# coding=utf-8
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Run masked LM pre-training for BERT."""

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import os
import modeling
import optimization
import tensorflow as tf

flags = tf.flags
FLAGS = flags.FLAGS

## Required parameters
flags.DEFINE_string("config_file", None,
                    "The config json file corresponding to the pre-trained BERT model. "
                    "This specifies the model architecture.")

flags.DEFINE_string("input_file", None,
                    "Input TF example files (can be a glob or comma separated).")

flags.DEFINE_string("eval_input_file", None,
                    "Input TF example file(s) for evaluation.")

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_integer("max_seq_length", 128,
                     "The maximum total input sequence length after WordPiece tokenization. "
                     "Sequences longer than this will be truncated, and sequences shorter "
                     "than this will be padded. Must match data generation.")


flags.DEFINE_bool("do_train", True, "Whether to run training.")

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

flags.DEFINE_integer("train_batch_size", 64, "Total batch size for training.")

flags.DEFINE_integer("eval_batch_size", 64, "Total batch size for eval.")

flags.DEFINE_float("learning_rate", 1e-4, "The initial learning rate for Adam.")

flags.DEFINE_integer("num_train_steps", 1000000, "Number of training steps.")

flags.DEFINE_integer("num_warmup_steps", 10000, "Number of warmup steps.")

flags.DEFINE_integer("save_checkpoints_steps", 10000,
                     "How often to save the model checkpoint.")

flags.DEFINE_integer("iterations_per_loop", 10000,
                     "How many steps to make in each estimator call.")

flags.DEFINE_integer("max_eval_steps", 100, "Maximum number of eval steps.")

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

tf.flags.DEFINE_string("tpu_name", None,
                       "The Cloud TPU to use for training. This should be either the name used"
                       " when creating the Cloud TPU, or a grpc://ip.address.of.tpu:8470 url.")

tf.flags.DEFINE_string("tpu_zone", None,
                       "[Optional] GCE zone where the Cloud TPU is located in. If not specified,"
                       " we will attempt to automatically detect the GCE project from metadata.")

tf.flags.DEFINE_string("gcp_project", None,
                       "[Optional] Project name for the Cloud TPU-enabled project. If not specified,"
                       " we will attempt to automatically detect the GCE project from metadata.")

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.")


def model_fn_builder(config, init_checkpoint, learning_rate,
                     num_train_steps, num_warmup_steps, use_tpu,
                     use_one_hot_embeddings):
  """Returns `model_fn` closure for TPUEstimator."""

  def model_fn(features, labels, mode, params):  # pylint: disable=unused-argument
    """The `model_fn` for TPUEstimator."""

    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"]
    target_ids = features["target_ids"]
    input_mask = features["input_mask"]

    is_training = (mode == tf.estimator.ModeKeys.TRAIN)

    model = modeling.BertModel(
        config=config,
        is_training=is_training,
        input_ids=input_ids,
        input_mask=input_mask,
        use_one_hot_embeddings=use_one_hot_embeddings)

    ###
    (lm_loss, lm_example_loss, lm_log_probs) = get_lm_output(
         config, model.get_sequence_output(), model.get_embedding_table(), target_ids, input_mask)
        
    total_loss = lm_loss

    tvars = tf.trainable_variables()

    initialized_variable_names = {}
    scaffold_fn = None
    if init_checkpoint:
      (assignment_map, initialized_variable_names
      ) = modeling.get_assignment_map_from_checkpoint(tvars, init_checkpoint)
      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, _lr = optimization.create_optimizer(
          total_loss, learning_rate, num_train_steps, num_warmup_steps, use_tpu)
      tf.summary.scalar("learning_rate", _lr)

      output_spec = tf.contrib.tpu.TPUEstimatorSpec(
          mode=mode,
          loss=total_loss,
          train_op=train_op,
          scaffold_fn=scaffold_fn)
        
    elif mode == tf.estimator.ModeKeys.EVAL:

      def metric_fn(lm_example_loss, lm_log_probs):
        """Computes the loss and accuracy of the model."""
        lm_log_probs = tf.reshape(lm_log_probs, [-1, lm_log_probs.shape[-1]])
        lm_predictions = tf.argmax(lm_log_probs, axis=-1, output_type=tf.int32)
        lm_example_loss = tf.reshape(lm_example_loss, [-1])
        lm_mean_loss = tf.metrics.mean(values=lm_example_loss)

        return {
            "lm_loss": lm_mean_loss,
        }

      eval_metrics = (metric_fn, [lm_example_loss, lm_log_probs])
      output_spec = tf.contrib.tpu.TPUEstimatorSpec(
          mode=mode,
          loss=total_loss,
          eval_metrics=eval_metrics,
          scaffold_fn=scaffold_fn)
    else:
      raise ValueError("Only TRAIN and EVAL modes are supported: %s" % (mode))

    return output_spec

  return model_fn


def get_lm_output(config, input_tensor, output_weights, label_ids, label_mask):
  """Get loss and log probs for the LM."""
  input_shape = modeling.get_shape_list(input_tensor, expected_rank=3)
  input_tensor = tf.reshape(input_tensor, [input_shape[0]*input_shape[1], input_shape[2]])
  
  with tf.variable_scope("cls/predictions"):
    # We apply one more non-linear transformation before the output layer.
    # This matrix is not used after pre-training.
    with tf.variable_scope("transform"):
      input_tensor = tf.layers.dense(
          input_tensor,
          units=config.hidden_size,
          activation=modeling.get_activation(config.hidden_act),
          kernel_initializer=modeling.create_initializer(config.initializer_range))
      input_tensor = modeling.layer_norm(input_tensor)

    # The output weights are the same as the input embeddings, but there is
    # an output-only bias for each token.
    output_bias = tf.get_variable(
        "output_bias",
        shape=[config.vocab_size],
        initializer=tf.zeros_initializer())
    logits = tf.matmul(input_tensor, output_weights, transpose_b=True)
    logits = tf.nn.bias_add(logits, output_bias)
    log_probs = tf.nn.log_softmax(logits, axis=-1)

    label_ids = tf.reshape(label_ids, [-1])

    one_hot_labels = tf.one_hot(label_ids, depth=config.vocab_size, dtype=tf.float32)

    # The `positions` tensor might be zero-padded (if the sequence is too
    # short to have the maximum number of predictions). The `label_weights`
    # tensor has a value of 1.0 for every real prediction and 0.0 for the
    # padding predictions.
    per_example_loss = -tf.reduce_sum(log_probs * one_hot_labels, axis=[-1])
    
    label_mask = tf.reshape(label_mask, [input_shape[0]*input_shape[1]])
    loss_mask = tf.dtypes.cast(label_mask, tf.float32)
    per_example_loss = tf.math.multiply(per_example_loss, loss_mask)
    loss = tf.reduce_mean(per_example_loss)

  return (loss, per_example_loss, log_probs)


def input_fn_builder(input_files,
                     max_seq_length,
                     is_training,
                     num_cpu_threads=4):
  """Creates an `input_fn` closure to be passed to TPUEstimator."""

  def input_fn(params):
    """The actual input function."""
    batch_size = params["batch_size"]

    name_to_features = {
        "input_ids": tf.FixedLenFeature([max_seq_length], tf.int64),
        "target_ids": tf.FixedLenFeature([max_seq_length], tf.int64),
        "input_mask": tf.FixedLenFeature([max_seq_length], tf.int64),
    }

    # For training, we want a lot of parallel reading and shuffling.
    # For eval, we want no shuffling and parallel reading doesn't matter.
    if is_training:
      d = tf.data.Dataset.from_tensor_slices(tf.constant(input_files))
      d = d.repeat()
      d = d.shuffle(buffer_size=len(input_files))

      # `cycle_length` is the number of parallel files that get read.
      cycle_length = min(num_cpu_threads, len(input_files))

      # `sloppy` mode means that the interleaving is not exact. This adds
      # even more randomness to the training pipeline.
      d = d.apply(
          tf.contrib.data.parallel_interleave(
              tf.data.TFRecordDataset,
              sloppy=is_training,
              cycle_length=cycle_length))
      d = d.shuffle(buffer_size=100)
    else:
      d = tf.data.TFRecordDataset(input_files)
      # Since we evaluate for a fixed number of steps we don't want to encounter
      # out-of-range exceptions.
      d = d.repeat()

    # We must `drop_remainder` on training because the TPU requires fixed
    # size dimensions. For eval, we assume we are evaluating on the CPU or GPU
    # and we *don't* want to drop the remainder, otherwise we wont cover
    # every sample.
    d = d.apply(
        tf.contrib.data.map_and_batch(
            lambda record: _decode_record(record, name_to_features),
            batch_size=batch_size,
            num_parallel_batches=num_cpu_threads,
            drop_remainder=True))
    return d

  return input_fn


def _decode_record(record, name_to_features):
  """Decodes a record to a TensorFlow example."""
  example = tf.parse_single_example(record, name_to_features)

  # tf.Example only supports tf.int64, but the TPU only supports tf.int32.
  # So cast all int64 to int32.
  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 main(_):
    tf.logging.set_verbosity(tf.logging.INFO)

    if not FLAGS.do_train and not FLAGS.do_eval:
        raise ValueError("At least one of `do_train` or `do_eval` must be True.")

    config = modeling.BertConfig.from_json_file(FLAGS.config_file)

    tf.gfile.MakeDirs(FLAGS.output_dir)
    # copy config file
    src = FLAGS.config_file
    dst = os.path.join(FLAGS.output_dir, FLAGS.config_file.split("/")[-1])
    os.system(f"cp {src} {dst}")
    # copy vocab file
    src = os.path.join(FLAGS.config_file.split("/")[0], config.vocab_file)
    dst = os.path.join(FLAGS.output_dir, config.vocab_file)
    os.system(f"cp {src} {dst}")
    
    input_files = []
    for input_pattern in FLAGS.input_file.split(","):
        input_files.extend(tf.gfile.Glob(input_pattern))

    tf.logging.info("*** Input Files ***")
    for input_file in input_files:
        tf.logging.info("  %s" % input_file)

    eval_input_files = []
    for eval_input_pattern in FLAGS.eval_input_file.split(","):
        eval_input_files.extend(tf.gfile.Glob(eval_input_pattern))

    tf.logging.info("*** Eval Files ***")
    for eval_input_file in eval_input_files:
        tf.logging.info("  %s" % eval_input_file)

    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))

    model_fn = model_fn_builder(
        config=config,
        init_checkpoint=FLAGS.init_checkpoint,
        learning_rate=FLAGS.learning_rate,
        num_train_steps=FLAGS.num_train_steps,
        num_warmup_steps=FLAGS.num_warmup_steps,
        use_tpu=FLAGS.use_tpu,
        use_one_hot_embeddings=FLAGS.use_tpu)
    # If TPU is not available, this will fall back to normal Estimator on CPU
    # or GPU.
    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)

    if FLAGS.do_train:
        tf.logging.info("***** Running training *****")
        tf.logging.info("  Batch size = %d", FLAGS.train_batch_size)
        train_input_fn = input_fn_builder(
            input_files=input_files,
            max_seq_length=FLAGS.max_seq_length,
            is_training=True)
        if FLAGS.do_eval:
            train_spec = tf.estimator.TrainSpec(input_fn=train_input_fn, max_steps=FLAGS.num_train_steps)
            
            eval_input_fn = input_fn_builder(
                input_files=eval_input_files,
                max_seq_length=FLAGS.max_seq_length,
                is_training=False)
            eval_spec = tf.estimator.EvalSpec(input_fn=eval_input_fn, steps=FLAGS.max_eval_steps)
    
            tf.estimator.train_and_evaluate(estimator, train_spec, eval_spec)
        else:
            estimator.train(input_fn=train_input_fn, max_steps=FLAGS.num_train_steps)

    if FLAGS.do_eval:
        tf.logging.info("***** Running evaluation *****")
        tf.logging.info("  Batch size = %d", FLAGS.eval_batch_size)

        eval_input_fn = input_fn_builder(
            input_files=eval_input_files,
            max_seq_length=FLAGS.max_seq_length,
            is_training=False)

        result = estimator.evaluate(
            input_fn=eval_input_fn, steps=FLAGS.max_eval_steps)

        output_eval_input_file = os.path.join(FLAGS.output_dir, "eval_results.txt")
        with tf.gfile.GFile(output_eval_input_file, "w") as writer:
          tf.logging.info("***** Eval results *****")
          for key in sorted(result.keys()):
            tf.logging.info("  %s = %s", key, str(result[key]))
            writer.write("%s = %s\n" % (key, str(result[key])))
            
            
if __name__ == "__main__":
  flags.mark_flag_as_required("input_file")
  flags.mark_flag_as_required("eval_input_file")
  flags.mark_flag_as_required("config_file")
  flags.mark_flag_as_required("output_dir")
  tf.app.run()