Remove BERT pretraining files for now
This commit is contained in:
VictorSanh
@@ -1,429 +0,0 @@
|
|||||||
# coding=utf-8
|
|
||||||
# Copyright 2018 The Google AI Language Team Authors.
|
|
||||||
#
|
|
||||||
# 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.
|
|
||||||
"""Create masked LM/next sentence masked_lm TF examples for BERT."""
|
|
||||||
|
|
||||||
from __future__ import absolute_import
|
|
||||||
from __future__ import division
|
|
||||||
from __future__ import print_function
|
|
||||||
|
|
||||||
import collections
|
|
||||||
import random
|
|
||||||
|
|
||||||
import tokenization
|
|
||||||
import tensorflow as tf
|
|
||||||
|
|
||||||
import argparse
|
|
||||||
|
|
||||||
parser = argparse.ArgumentParser()
|
|
||||||
|
|
||||||
## Required parameters
|
|
||||||
parser.add_argument("--input_file", default=None, type=str, required=True,
|
|
||||||
help="Input raw text file (or comma-separated list of files).")
|
|
||||||
parser.add_argument("--output_file", default=None, type=str, required=True,
|
|
||||||
help="Output TF example file (or comma-separated list of files).")
|
|
||||||
parser.add_argument("--vocab_file", default=None, type=str, required=True,
|
|
||||||
help="The vocabulary file that the BERT model was trained on.")
|
|
||||||
|
|
||||||
## Other parameters
|
|
||||||
parser.add_argument("--do_lower_case", default=True, action='store_true',
|
|
||||||
help="Whether to lower case the input text. Should be True for uncased "
|
|
||||||
"models and False for cased models.")
|
|
||||||
parser.add_argument("--max_seq_length", default=128, type=int, help="Maximum sequence length.")
|
|
||||||
parser.add_argument("--max_predictions_per_seq", default=20, type=int,
|
|
||||||
help="Maximum number of masked LM predictions per sequence.")
|
|
||||||
parser.add_argument("--random_seed", default=12345, type=int, help="Random seed for data generation.")
|
|
||||||
parser.add_argument("--dupe_factor", default=10, type=int,
|
|
||||||
help="Number of times to duplicate the input data (with different masks).")
|
|
||||||
parser.add_argument("--masked_lm_prob", default=0.15, type=float, help="Masked LM probability.")
|
|
||||||
parser.add_argument("--short_seq_prob", default=0.1, type=float,
|
|
||||||
help="Probability of creating sequences which are shorter than the maximum length.")
|
|
||||||
|
|
||||||
args = parser.parse_args()
|
|
||||||
|
|
||||||
|
|
||||||
class TrainingInstance(object):
|
|
||||||
"""A single training instance (sentence pair)."""
|
|
||||||
|
|
||||||
def __init__(self, tokens, segment_ids, masked_lm_positions, masked_lm_labels,
|
|
||||||
is_random_next):
|
|
||||||
self.tokens = tokens
|
|
||||||
self.segment_ids = segment_ids
|
|
||||||
self.is_random_next = is_random_next
|
|
||||||
self.masked_lm_positions = masked_lm_positions
|
|
||||||
self.masked_lm_labels = masked_lm_labels
|
|
||||||
|
|
||||||
def __str__(self):
|
|
||||||
s = ""
|
|
||||||
s += "tokens: %s\n" % (" ".join(
|
|
||||||
[tokenization.printable_text(x) for x in self.tokens]))
|
|
||||||
s += "segment_ids: %s\n" % (" ".join([str(x) for x in self.segment_ids]))
|
|
||||||
s += "is_random_next: %s\n" % self.is_random_next
|
|
||||||
s += "masked_lm_positions: %s\n" % (" ".join(
|
|
||||||
[str(x) for x in self.masked_lm_positions]))
|
|
||||||
s += "masked_lm_labels: %s\n" % (" ".join(
|
|
||||||
[tokenization.printable_text(x) for x in self.masked_lm_labels]))
|
|
||||||
s += "\n"
|
|
||||||
return s
|
|
||||||
|
|
||||||
def __repr__(self):
|
|
||||||
return self.__str__()
|
|
||||||
|
|
||||||
|
|
||||||
def write_instance_to_example_files(instances, tokenizer, max_seq_length,
|
|
||||||
max_predictions_per_seq, output_files):
|
|
||||||
"""Create TF example files from `TrainingInstance`s."""
|
|
||||||
writers = []
|
|
||||||
for output_file in output_files:
|
|
||||||
writers.append(tf.python_io.TFRecordWriter(output_file))
|
|
||||||
|
|
||||||
writer_index = 0
|
|
||||||
|
|
||||||
total_written = 0
|
|
||||||
for (inst_index, instance) in enumerate(instances):
|
|
||||||
input_ids = tokenizer.convert_tokens_to_ids(instance.tokens)
|
|
||||||
input_mask = [1] * len(input_ids)
|
|
||||||
segment_ids = list(instance.segment_ids)
|
|
||||||
assert len(input_ids) <= max_seq_length
|
|
||||||
|
|
||||||
while len(input_ids) < max_seq_length:
|
|
||||||
input_ids.append(0)
|
|
||||||
input_mask.append(0)
|
|
||||||
segment_ids.append(0)
|
|
||||||
|
|
||||||
assert len(input_ids) == max_seq_length
|
|
||||||
assert len(input_mask) == max_seq_length
|
|
||||||
assert len(segment_ids) == max_seq_length
|
|
||||||
|
|
||||||
masked_lm_positions = list(instance.masked_lm_positions)
|
|
||||||
masked_lm_ids = tokenizer.convert_tokens_to_ids(instance.masked_lm_labels)
|
|
||||||
masked_lm_weights = [1.0] * len(masked_lm_ids)
|
|
||||||
|
|
||||||
while len(masked_lm_positions) < max_predictions_per_seq:
|
|
||||||
masked_lm_positions.append(0)
|
|
||||||
masked_lm_ids.append(0)
|
|
||||||
masked_lm_weights.append(0.0)
|
|
||||||
|
|
||||||
next_sentence_label = 1 if instance.is_random_next else 0
|
|
||||||
|
|
||||||
features = collections.OrderedDict()
|
|
||||||
features["input_ids"] = create_int_feature(input_ids)
|
|
||||||
features["input_mask"] = create_int_feature(input_mask)
|
|
||||||
features["segment_ids"] = create_int_feature(segment_ids)
|
|
||||||
features["masked_lm_positions"] = create_int_feature(masked_lm_positions)
|
|
||||||
features["masked_lm_ids"] = create_int_feature(masked_lm_ids)
|
|
||||||
features["masked_lm_weights"] = create_float_feature(masked_lm_weights)
|
|
||||||
features["next_sentence_labels"] = create_int_feature([next_sentence_label])
|
|
||||||
|
|
||||||
tf_example = tf.train.Example(features=tf.train.Features(feature=features))
|
|
||||||
|
|
||||||
writers[writer_index].write(tf_example.SerializeToString())
|
|
||||||
writer_index = (writer_index + 1) % len(writers)
|
|
||||||
|
|
||||||
total_written += 1
|
|
||||||
|
|
||||||
if inst_index < 20:
|
|
||||||
tf.logging.info("*** Example ***")
|
|
||||||
tf.logging.info("tokens: %s" % " ".join(
|
|
||||||
[tokenization.printable_text(x) for x in instance.tokens]))
|
|
||||||
|
|
||||||
for feature_name in features.keys():
|
|
||||||
feature = features[feature_name]
|
|
||||||
values = []
|
|
||||||
if feature.int64_list.value:
|
|
||||||
values = feature.int64_list.value
|
|
||||||
elif feature.float_list.value:
|
|
||||||
values = feature.float_list.value
|
|
||||||
tf.logging.info(
|
|
||||||
"%s: %s" % (feature_name, " ".join([str(x) for x in values])))
|
|
||||||
|
|
||||||
for writer in writers:
|
|
||||||
writer.close()
|
|
||||||
|
|
||||||
tf.logging.info("Wrote %d total instances", total_written)
|
|
||||||
|
|
||||||
|
|
||||||
def create_int_feature(values):
|
|
||||||
feature = tf.train.Feature(int64_list=tf.train.Int64List(value=list(values)))
|
|
||||||
return feature
|
|
||||||
|
|
||||||
|
|
||||||
def create_float_feature(values):
|
|
||||||
feature = tf.train.Feature(float_list=tf.train.FloatList(value=list(values)))
|
|
||||||
return feature
|
|
||||||
|
|
||||||
|
|
||||||
def create_training_instances(input_files, tokenizer, max_seq_length,
|
|
||||||
dupe_factor, short_seq_prob, masked_lm_prob,
|
|
||||||
max_predictions_per_seq, rng):
|
|
||||||
"""Create `TrainingInstance`s from raw text."""
|
|
||||||
all_documents = [[]]
|
|
||||||
|
|
||||||
# Input file format:
|
|
||||||
# (1) One sentence per line. These should ideally be actual sentences, not
|
|
||||||
# entire paragraphs or arbitrary spans of text. (Because we use the
|
|
||||||
# sentence boundaries for the "next sentence prediction" task).
|
|
||||||
# (2) Blank lines between documents. Document boundaries are needed so
|
|
||||||
# that the "next sentence prediction" task doesn't span between documents.
|
|
||||||
for input_file in input_files:
|
|
||||||
with tf.gfile.GFile(input_file, "r") as reader:
|
|
||||||
while True:
|
|
||||||
line = tokenization.convert_to_unicode(reader.readline())
|
|
||||||
if not line:
|
|
||||||
break
|
|
||||||
line = line.strip()
|
|
||||||
|
|
||||||
# Empty lines are used as document delimiters
|
|
||||||
if not line:
|
|
||||||
all_documents.append([])
|
|
||||||
tokens = tokenizer.tokenize(line)
|
|
||||||
if tokens:
|
|
||||||
all_documents[-1].append(tokens)
|
|
||||||
|
|
||||||
# Remove empty documents
|
|
||||||
all_documents = [x for x in all_documents if x]
|
|
||||||
rng.shuffle(all_documents)
|
|
||||||
|
|
||||||
vocab_words = list(tokenizer.vocab.keys())
|
|
||||||
instances = []
|
|
||||||
for _ in range(dupe_factor):
|
|
||||||
for document_index in range(len(all_documents)):
|
|
||||||
instances.extend(
|
|
||||||
create_instances_from_document(
|
|
||||||
all_documents, document_index, max_seq_length, short_seq_prob,
|
|
||||||
masked_lm_prob, max_predictions_per_seq, vocab_words, rng))
|
|
||||||
|
|
||||||
rng.shuffle(instances)
|
|
||||||
return instances
|
|
||||||
|
|
||||||
|
|
||||||
def create_instances_from_document(
|
|
||||||
all_documents, document_index, max_seq_length, short_seq_prob,
|
|
||||||
masked_lm_prob, max_predictions_per_seq, vocab_words, rng):
|
|
||||||
"""Creates `TrainingInstance`s for a single document."""
|
|
||||||
document = all_documents[document_index]
|
|
||||||
|
|
||||||
# Account for [CLS], [SEP], [SEP]
|
|
||||||
max_num_tokens = max_seq_length - 3
|
|
||||||
|
|
||||||
# We *usually* want to fill up the entire sequence since we are padding
|
|
||||||
# to `max_seq_length` anyways, so short sequences are generally wasted
|
|
||||||
# computation. However, we *sometimes*
|
|
||||||
# (i.e., short_seq_prob == 0.1 == 10% of the time) want to use shorter
|
|
||||||
# sequences to minimize the mismatch between pre-training and fine-tuning.
|
|
||||||
# The `target_seq_length` is just a rough target however, whereas
|
|
||||||
# `max_seq_length` is a hard limit.
|
|
||||||
target_seq_length = max_num_tokens
|
|
||||||
if rng.random() < short_seq_prob:
|
|
||||||
target_seq_length = rng.randint(2, max_num_tokens)
|
|
||||||
|
|
||||||
# We DON'T just concatenate all of the tokens from a document into a long
|
|
||||||
# sequence and choose an arbitrary split point because this would make the
|
|
||||||
# next sentence prediction task too easy. Instead, we split the input into
|
|
||||||
# segments "A" and "B" based on the actual "sentences" provided by the user
|
|
||||||
# input.
|
|
||||||
instances = []
|
|
||||||
current_chunk = []
|
|
||||||
current_length = 0
|
|
||||||
i = 0
|
|
||||||
while i < len(document):
|
|
||||||
segment = document[i]
|
|
||||||
current_chunk.append(segment)
|
|
||||||
current_length += len(segment)
|
|
||||||
if i == len(document) - 1 or current_length >= target_seq_length:
|
|
||||||
if current_chunk:
|
|
||||||
# `a_end` is how many segments from `current_chunk` go into the `A`
|
|
||||||
# (first) sentence.
|
|
||||||
a_end = 1
|
|
||||||
if len(current_chunk) >= 2:
|
|
||||||
a_end = rng.randint(1, len(current_chunk) - 1)
|
|
||||||
|
|
||||||
tokens_a = []
|
|
||||||
for j in range(a_end):
|
|
||||||
tokens_a.extend(current_chunk[j])
|
|
||||||
|
|
||||||
tokens_b = []
|
|
||||||
# Random next
|
|
||||||
is_random_next = False
|
|
||||||
if len(current_chunk) == 1 or rng.random() < 0.5:
|
|
||||||
is_random_next = True
|
|
||||||
target_b_length = target_seq_length - len(tokens_a)
|
|
||||||
|
|
||||||
# This should rarely go for more than one iteration for large
|
|
||||||
# corpora. However, just to be careful, we try to make sure that
|
|
||||||
# the random document is not the same as the document
|
|
||||||
# we're processing.
|
|
||||||
for _ in range(10):
|
|
||||||
random_document_index = rng.randint(0, len(all_documents) - 1)
|
|
||||||
if random_document_index != document_index:
|
|
||||||
break
|
|
||||||
|
|
||||||
random_document = all_documents[random_document_index]
|
|
||||||
random_start = rng.randint(0, len(random_document) - 1)
|
|
||||||
for j in range(random_start, len(random_document)):
|
|
||||||
tokens_b.extend(random_document[j])
|
|
||||||
if len(tokens_b) >= target_b_length:
|
|
||||||
break
|
|
||||||
# We didn't actually use these segments so we "put them back" so
|
|
||||||
# they don't go to waste.
|
|
||||||
num_unused_segments = len(current_chunk) - a_end
|
|
||||||
i -= num_unused_segments
|
|
||||||
# Actual next
|
|
||||||
else:
|
|
||||||
is_random_next = False
|
|
||||||
for j in range(a_end, len(current_chunk)):
|
|
||||||
tokens_b.extend(current_chunk[j])
|
|
||||||
truncate_seq_pair(tokens_a, tokens_b, max_num_tokens, rng)
|
|
||||||
|
|
||||||
assert len(tokens_a) >= 1
|
|
||||||
assert len(tokens_b) >= 1
|
|
||||||
|
|
||||||
tokens = []
|
|
||||||
segment_ids = []
|
|
||||||
tokens.append("[CLS]")
|
|
||||||
segment_ids.append(0)
|
|
||||||
for token in tokens_a:
|
|
||||||
tokens.append(token)
|
|
||||||
segment_ids.append(0)
|
|
||||||
|
|
||||||
tokens.append("[SEP]")
|
|
||||||
segment_ids.append(0)
|
|
||||||
|
|
||||||
for token in tokens_b:
|
|
||||||
tokens.append(token)
|
|
||||||
segment_ids.append(1)
|
|
||||||
tokens.append("[SEP]")
|
|
||||||
segment_ids.append(1)
|
|
||||||
|
|
||||||
(tokens, masked_lm_positions,
|
|
||||||
masked_lm_labels) = create_masked_lm_predictions(
|
|
||||||
tokens, masked_lm_prob, max_predictions_per_seq, vocab_words, rng)
|
|
||||||
instance = TrainingInstance(
|
|
||||||
tokens=tokens,
|
|
||||||
segment_ids=segment_ids,
|
|
||||||
is_random_next=is_random_next,
|
|
||||||
masked_lm_positions=masked_lm_positions,
|
|
||||||
masked_lm_labels=masked_lm_labels)
|
|
||||||
instances.append(instance)
|
|
||||||
current_chunk = []
|
|
||||||
current_length = 0
|
|
||||||
i += 1
|
|
||||||
|
|
||||||
return instances
|
|
||||||
|
|
||||||
|
|
||||||
def create_masked_lm_predictions(tokens, masked_lm_prob,
|
|
||||||
max_predictions_per_seq, vocab_words, rng):
|
|
||||||
"""Creates the predictis for the masked LM objective."""
|
|
||||||
|
|
||||||
cand_indexes = []
|
|
||||||
for (i, token) in enumerate(tokens):
|
|
||||||
if token == "[CLS]" or token == "[SEP]":
|
|
||||||
continue
|
|
||||||
cand_indexes.append(i)
|
|
||||||
|
|
||||||
rng.shuffle(cand_indexes)
|
|
||||||
|
|
||||||
output_tokens = list(tokens)
|
|
||||||
|
|
||||||
masked_lm = collections.namedtuple("masked_lm", ["index", "label"]) # pylint: disable=invalid-name
|
|
||||||
|
|
||||||
num_to_predict = min(max_predictions_per_seq,
|
|
||||||
max(1, int(round(len(tokens) * masked_lm_prob))))
|
|
||||||
|
|
||||||
masked_lms = []
|
|
||||||
covered_indexes = set()
|
|
||||||
for index in cand_indexes:
|
|
||||||
if len(masked_lms) >= num_to_predict:
|
|
||||||
break
|
|
||||||
if index in covered_indexes:
|
|
||||||
continue
|
|
||||||
covered_indexes.add(index)
|
|
||||||
|
|
||||||
masked_token = None
|
|
||||||
# 80% of the time, replace with [MASK]
|
|
||||||
if rng.random() < 0.8:
|
|
||||||
masked_token = "[MASK]"
|
|
||||||
else:
|
|
||||||
# 10% of the time, keep original
|
|
||||||
if rng.random() < 0.5:
|
|
||||||
masked_token = tokens[index]
|
|
||||||
# 10% of the time, replace with random word
|
|
||||||
else:
|
|
||||||
masked_token = vocab_words[rng.randint(0, len(vocab_words) - 1)]
|
|
||||||
|
|
||||||
output_tokens[index] = masked_token
|
|
||||||
|
|
||||||
masked_lms.append(masked_lm(index=index, label=tokens[index]))
|
|
||||||
|
|
||||||
masked_lms = sorted(masked_lms, key=lambda x: x.index)
|
|
||||||
|
|
||||||
masked_lm_positions = []
|
|
||||||
masked_lm_labels = []
|
|
||||||
for p in masked_lms:
|
|
||||||
masked_lm_positions.append(p.index)
|
|
||||||
masked_lm_labels.append(p.label)
|
|
||||||
|
|
||||||
return (output_tokens, masked_lm_positions, masked_lm_labels)
|
|
||||||
|
|
||||||
|
|
||||||
def truncate_seq_pair(tokens_a, tokens_b, max_num_tokens, rng):
|
|
||||||
"""Truncates a pair of sequences to a maximum sequence length."""
|
|
||||||
while True:
|
|
||||||
total_length = len(tokens_a) + len(tokens_b)
|
|
||||||
if total_length <= max_num_tokens:
|
|
||||||
break
|
|
||||||
|
|
||||||
trunc_tokens = tokens_a if len(tokens_a) > len(tokens_b) else tokens_b
|
|
||||||
assert len(trunc_tokens) >= 1
|
|
||||||
|
|
||||||
# We want to sometimes truncate from the front and sometimes from the
|
|
||||||
# back to add more randomness and avoid biases.
|
|
||||||
if rng.random() < 0.5:
|
|
||||||
del trunc_tokens[0]
|
|
||||||
else:
|
|
||||||
trunc_tokens.pop()
|
|
||||||
|
|
||||||
|
|
||||||
def main(_):
|
|
||||||
tf.logging.set_verbosity(tf.logging.INFO)
|
|
||||||
|
|
||||||
tokenizer = tokenization.FullTokenizer(
|
|
||||||
vocab_file=args.vocab_file, do_lower_case=args.do_lower_case)
|
|
||||||
|
|
||||||
input_files = []
|
|
||||||
for input_pattern in args.input_file.split(","):
|
|
||||||
input_files.extend(tf.gfile.Glob(input_pattern))
|
|
||||||
|
|
||||||
tf.logging.info("*** Reading from input files ***")
|
|
||||||
for input_file in input_files:
|
|
||||||
tf.logging.info(" %s", input_file)
|
|
||||||
|
|
||||||
rng = random.Random(args.random_seed)
|
|
||||||
instances = create_training_instances(
|
|
||||||
input_files, tokenizer, args.max_seq_length, args.dupe_factor,
|
|
||||||
args.short_seq_prob, args.masked_lm_prob, args.max_predictions_per_seq,
|
|
||||||
rng)
|
|
||||||
|
|
||||||
output_files = args.output_file.split(",")
|
|
||||||
tf.logging.info("*** Writing to output files ***")
|
|
||||||
for output_file in output_files:
|
|
||||||
tf.logging.info(" %s", output_file)
|
|
||||||
|
|
||||||
write_instance_to_example_files(instances, tokenizer, args.max_seq_length,
|
|
||||||
args.max_predictions_per_seq, output_files)
|
|
||||||
|
|
||||||
|
|
||||||
if __name__ == "__main__":
|
|
||||||
tf.app.run()
|
|
||||||
@@ -1,460 +0,0 @@
|
|||||||
# coding=utf-8
|
|
||||||
# Copyright 2018 The Google AI Language Team Authors.
|
|
||||||
#
|
|
||||||
# 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/next sentence 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
|
|
||||||
import argparse
|
|
||||||
|
|
||||||
|
|
||||||
def model_fn_builder(bert_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"]
|
|
||||||
input_mask = features["input_mask"]
|
|
||||||
segment_ids = features["segment_ids"]
|
|
||||||
masked_lm_positions = features["masked_lm_positions"]
|
|
||||||
masked_lm_ids = features["masked_lm_ids"]
|
|
||||||
masked_lm_weights = features["masked_lm_weights"]
|
|
||||||
next_sentence_labels = features["next_sentence_labels"]
|
|
||||||
|
|
||||||
is_training = (mode == tf.estimator.ModeKeys.TRAIN)
|
|
||||||
|
|
||||||
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)
|
|
||||||
|
|
||||||
(masked_lm_loss,
|
|
||||||
masked_lm_example_loss, masked_lm_log_probs) = get_masked_lm_output(
|
|
||||||
bert_config, model.get_sequence_output(), model.get_embedding_table(),
|
|
||||||
masked_lm_positions, masked_lm_ids, masked_lm_weights)
|
|
||||||
|
|
||||||
(next_sentence_loss, next_sentence_example_loss,
|
|
||||||
next_sentence_log_probs) = get_next_sentence_output(
|
|
||||||
bert_config, model.get_pooled_output(), next_sentence_labels)
|
|
||||||
|
|
||||||
total_loss = masked_lm_loss + next_sentence_loss
|
|
||||||
|
|
||||||
tvars = tf.trainable_variables()
|
|
||||||
|
|
||||||
initialized_variable_names = {}
|
|
||||||
scaffold_fn = None
|
|
||||||
if init_checkpoint:
|
|
||||||
(assignment_map,
|
|
||||||
initialized_variable_names) = modeling.get_assigment_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 = 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.EVAL:
|
|
||||||
|
|
||||||
def metric_fn(masked_lm_example_loss, masked_lm_log_probs, masked_lm_ids,
|
|
||||||
masked_lm_weights, next_sentence_example_loss,
|
|
||||||
next_sentence_log_probs, next_sentence_labels):
|
|
||||||
"""Computes the loss and accuracy of the model."""
|
|
||||||
masked_lm_log_probs = tf.reshape(masked_lm_log_probs,
|
|
||||||
[-1, masked_lm_log_probs.shape[-1]])
|
|
||||||
masked_lm_predictions = tf.argmax(
|
|
||||||
masked_lm_log_probs, axis=-1, output_type=tf.int32)
|
|
||||||
masked_lm_example_loss = tf.reshape(masked_lm_example_loss, [-1])
|
|
||||||
masked_lm_ids = tf.reshape(masked_lm_ids, [-1])
|
|
||||||
masked_lm_weights = tf.reshape(masked_lm_weights, [-1])
|
|
||||||
masked_lm_accuracy = tf.metrics.accuracy(
|
|
||||||
labels=masked_lm_ids,
|
|
||||||
predictions=masked_lm_predictions,
|
|
||||||
weights=masked_lm_weights)
|
|
||||||
masked_lm_mean_loss = tf.metrics.mean(
|
|
||||||
values=masked_lm_example_loss, weights=masked_lm_weights)
|
|
||||||
|
|
||||||
next_sentence_log_probs = tf.reshape(
|
|
||||||
next_sentence_log_probs, [-1, next_sentence_log_probs.shape[-1]])
|
|
||||||
next_sentence_predictions = tf.argmax(
|
|
||||||
next_sentence_log_probs, axis=-1, output_type=tf.int32)
|
|
||||||
next_sentence_labels = tf.reshape(next_sentence_labels, [-1])
|
|
||||||
next_sentence_accuracy = tf.metrics.accuracy(
|
|
||||||
labels=next_sentence_labels, predictions=next_sentence_predictions)
|
|
||||||
next_sentence_mean_loss = tf.metrics.mean(
|
|
||||||
values=next_sentence_example_loss)
|
|
||||||
|
|
||||||
return {
|
|
||||||
"masked_lm_accuracy": masked_lm_accuracy,
|
|
||||||
"masked_lm_loss": masked_lm_mean_loss,
|
|
||||||
"next_sentence_accuracy": next_sentence_accuracy,
|
|
||||||
"next_sentence_loss": next_sentence_mean_loss,
|
|
||||||
}
|
|
||||||
|
|
||||||
eval_metrics = (metric_fn, [
|
|
||||||
masked_lm_example_loss, masked_lm_log_probs, masked_lm_ids,
|
|
||||||
masked_lm_weights, next_sentence_example_loss,
|
|
||||||
next_sentence_log_probs, next_sentence_labels
|
|
||||||
])
|
|
||||||
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_masked_lm_output(bert_config, input_tensor, output_weights, positions,
|
|
||||||
label_ids, label_weights):
|
|
||||||
"""Get loss and log probs for the masked LM."""
|
|
||||||
input_tensor = gather_indexes(input_tensor, positions)
|
|
||||||
|
|
||||||
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=bert_config.hidden_size,
|
|
||||||
activation=modeling.get_activation(bert_config.hidden_act),
|
|
||||||
kernel_initializer=modeling.create_initializer(
|
|
||||||
bert_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=[bert_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])
|
|
||||||
label_weights = tf.reshape(label_weights, [-1])
|
|
||||||
|
|
||||||
one_hot_labels = tf.one_hot(
|
|
||||||
label_ids, depth=bert_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])
|
|
||||||
numerator = tf.reduce_sum(label_weights * per_example_loss)
|
|
||||||
denominator = tf.reduce_sum(label_weights) + 1e-5
|
|
||||||
loss = numerator / denominator
|
|
||||||
|
|
||||||
return (loss, per_example_loss, log_probs)
|
|
||||||
|
|
||||||
|
|
||||||
def get_next_sentence_output(bert_config, input_tensor, labels):
|
|
||||||
"""Get loss and log probs for the next sentence prediction."""
|
|
||||||
|
|
||||||
# Simple binary classification. Note that 0 is "next sentence" and 1 is
|
|
||||||
# "random sentence". This weight matrix is not used after pre-training.
|
|
||||||
with tf.variable_scope("cls/seq_relationship"):
|
|
||||||
output_weights = tf.get_variable(
|
|
||||||
"output_weights",
|
|
||||||
shape=[2, bert_config.hidden_size],
|
|
||||||
initializer=modeling.create_initializer(bert_config.initializer_range))
|
|
||||||
output_bias = tf.get_variable(
|
|
||||||
"output_bias", shape=[2], 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)
|
|
||||||
labels = tf.reshape(labels, [-1])
|
|
||||||
one_hot_labels = tf.one_hot(labels, depth=2, dtype=tf.float32)
|
|
||||||
per_example_loss = -tf.reduce_sum(one_hot_labels * log_probs, axis=-1)
|
|
||||||
loss = tf.reduce_mean(per_example_loss)
|
|
||||||
return (loss, per_example_loss, log_probs)
|
|
||||||
|
|
||||||
|
|
||||||
def gather_indexes(sequence_tensor, positions):
|
|
||||||
"""Gathers the vectors at the specific positions over a minibatch."""
|
|
||||||
sequence_shape = modeling.get_shape_list(sequence_tensor, expected_rank=3)
|
|
||||||
batch_size = sequence_shape[0]
|
|
||||||
seq_length = sequence_shape[1]
|
|
||||||
width = sequence_shape[2]
|
|
||||||
|
|
||||||
flat_offsets = tf.reshape(
|
|
||||||
tf.range(0, batch_size, dtype=tf.int32) * seq_length, [-1, 1])
|
|
||||||
flat_positions = tf.reshape(positions + flat_offsets, [-1])
|
|
||||||
flat_sequence_tensor = tf.reshape(sequence_tensor,
|
|
||||||
[batch_size * seq_length, width])
|
|
||||||
output_tensor = tf.gather(flat_sequence_tensor, flat_positions)
|
|
||||||
return output_tensor
|
|
||||||
|
|
||||||
|
|
||||||
def input_fn_builder(input_files,
|
|
||||||
max_seq_length,
|
|
||||||
max_predictions_per_seq,
|
|
||||||
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),
|
|
||||||
"input_mask":
|
|
||||||
tf.FixedLenFeature([max_seq_length], tf.int64),
|
|
||||||
"segment_ids":
|
|
||||||
tf.FixedLenFeature([max_seq_length], tf.int64),
|
|
||||||
"masked_lm_positions":
|
|
||||||
tf.FixedLenFeature([max_predictions_per_seq], tf.int64),
|
|
||||||
"masked_lm_ids":
|
|
||||||
tf.FixedLenFeature([max_predictions_per_seq], tf.int64),
|
|
||||||
"masked_lm_weights":
|
|
||||||
tf.FixedLenFeature([max_predictions_per_seq], tf.float32),
|
|
||||||
"next_sentence_labels":
|
|
||||||
tf.FixedLenFeature([1], 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 evaling 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(_):
|
|
||||||
parser = argparse.ArgumentParser()
|
|
||||||
|
|
||||||
## Required parameters
|
|
||||||
parser.add_argument("--bert_config_file", default=None, type=str, required=True,
|
|
||||||
help="The config json file corresponding to the pre-trained BERT model. "
|
|
||||||
"This specifies the model architecture.")
|
|
||||||
parser.add_argument("--input_file", default=None, type=str, required=True,
|
|
||||||
help="Input TF example files (can be a glob or comma separated).")
|
|
||||||
parser.add_argument("--output_dir", default=None, type=str, required=True,
|
|
||||||
help="The output directory where the model checkpoints will be written.")
|
|
||||||
|
|
||||||
## Other parameters
|
|
||||||
parser.add_argument("--init_checkpoint", default=None, type=str,
|
|
||||||
help="Initial checkpoint (usually from a pre-trained BERT model).")
|
|
||||||
parser.add_argument("--max_seq_length", default=128, type=int,
|
|
||||||
help="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.")
|
|
||||||
parser.add_argument("--max_predictions_per_seq", default=20, type=int,
|
|
||||||
help="Maximum number of masked LM predictions per sequence. Must match data generation.")
|
|
||||||
parser.add_argument("--do_train", default=False, action='store_true', help="Whether to run training.")
|
|
||||||
parser.add_argument("--do_eval", default=False, action='store_true', help="Whether to run eval on the dev set.")
|
|
||||||
parser.add_argument("--train_batch_size", default=32, type=int, help="Total batch size for training.")
|
|
||||||
parser.add_argument("--eval_batch_size", default=8, type=int, help="Total batch size for eval.")
|
|
||||||
parser.add_argument("--learning_rate", default=5e-5, type=float, help="The initial learning rate for Adam.")
|
|
||||||
parser.add_argument("--num_train_steps", default=100000, type=int, help="Number of training steps.")
|
|
||||||
parser.add_argument("--num_warmup_steps", default=10000, type=int, help="Number of warmup steps.")
|
|
||||||
parser.add_argument("--save_checkpoints_steps", default=1000, type=int,
|
|
||||||
help="How often to save the model checkpoint.")
|
|
||||||
parser.add_argument("--iterations_per_loop", default=1000, type=int,
|
|
||||||
help="How many steps to make in each estimator call.")
|
|
||||||
parser.add_argument("--max_eval_steps", default=100, type=int, help="Maximum number of eval steps.")
|
|
||||||
### BEGIN - TO DELETE EVENTUALLY --> NO SENSE IN PYTORCH ###
|
|
||||||
parser.add_argument("--use_tpu", default=False, action='store_true', help="Whether to use TPU or GPU/CPU.")
|
|
||||||
parser.add_argument("--tpu_name", default=None, type=str,
|
|
||||||
help="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.")
|
|
||||||
parser.add_argument("--tpu_zone", default=None, type=str,
|
|
||||||
help="[Optional] GCE zone where the Cloud TPU is located in. If not specified, we will attempt "
|
|
||||||
"to automatically detect the GCE project from metadata.")
|
|
||||||
parser.add_argument("--gcp_project", default=None, type=str,
|
|
||||||
help="[Optional] Project name for the Cloud TPU-enabled project. If not specified, "
|
|
||||||
"we will attempt to automatically detect the GCE project from metadata.")
|
|
||||||
parser.add_argument("--master", default=None, type=str, help="[Optional] TensorFlow master URL.")
|
|
||||||
parser.add_argument("--num_tpu_cores", default=8, type=int,
|
|
||||||
help="Only used if `use_tpu` is True. Total number of TPU cores to use.")
|
|
||||||
### END - TO DELETE EVENTUALLY --> NO SENSE IN PYTORCH ###
|
|
||||||
|
|
||||||
args = parser.parse_args()
|
|
||||||
|
|
||||||
tf.logging.set_verbosity(tf.logging.INFO)
|
|
||||||
|
|
||||||
if not args.do_train and not args.do_eval:
|
|
||||||
raise ValueError("At least one of `do_train` or `do_eval` must be True.")
|
|
||||||
|
|
||||||
bert_config = modeling.BertConfig.from_json_file(args.bert_config_file)
|
|
||||||
|
|
||||||
tf.gfile.MakeDirs(args.output_dir)
|
|
||||||
|
|
||||||
input_files = []
|
|
||||||
for input_pattern in args.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)
|
|
||||||
|
|
||||||
tpu_cluster_resolver = None
|
|
||||||
if args.use_tpu and args.tpu_name:
|
|
||||||
tpu_cluster_resolver = tf.contrib.cluster_resolver.TPUClusterResolver(
|
|
||||||
args.tpu_name, zone=args.tpu_zone, project=args.gcp_project)
|
|
||||||
|
|
||||||
is_per_host = tf.contrib.tpu.InputPipelineConfig.PER_HOST_V2
|
|
||||||
run_config = tf.contrib.tpu.RunConfig(
|
|
||||||
cluster=tpu_cluster_resolver,
|
|
||||||
master=args.master,
|
|
||||||
model_dir=args.output_dir,
|
|
||||||
save_checkpoints_steps=args.save_checkpoints_steps,
|
|
||||||
tpu_config=tf.contrib.tpu.TPUConfig(
|
|
||||||
iterations_per_loop=args.iterations_per_loop,
|
|
||||||
num_shards=args.num_tpu_cores,
|
|
||||||
per_host_input_for_training=is_per_host))
|
|
||||||
|
|
||||||
model_fn = model_fn_builder(
|
|
||||||
bert_config=bert_config,
|
|
||||||
init_checkpoint=args.init_checkpoint,
|
|
||||||
learning_rate=args.learning_rate,
|
|
||||||
num_train_steps=args.num_train_steps,
|
|
||||||
num_warmup_steps=args.num_warmup_steps,
|
|
||||||
use_tpu=args.use_tpu,
|
|
||||||
use_one_hot_embeddings=args.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=args.use_tpu,
|
|
||||||
model_fn=model_fn,
|
|
||||||
config=run_config,
|
|
||||||
train_batch_size=args.train_batch_size,
|
|
||||||
eval_batch_size=args.eval_batch_size)
|
|
||||||
|
|
||||||
if args.do_train:
|
|
||||||
tf.logging.info("***** Running training *****")
|
|
||||||
tf.logging.info(" Batch size = %d", args.train_batch_size)
|
|
||||||
train_input_fn = input_fn_builder(
|
|
||||||
input_files=input_files,
|
|
||||||
max_seq_length=args.max_seq_length,
|
|
||||||
max_predictions_per_seq=args.max_predictions_per_seq,
|
|
||||||
is_training=True)
|
|
||||||
estimator.train(input_fn=train_input_fn, max_steps=args.num_train_steps)
|
|
||||||
|
|
||||||
if args.do_eval:
|
|
||||||
tf.logging.info("***** Running evaluation *****")
|
|
||||||
tf.logging.info(" Batch size = %d", args.eval_batch_size)
|
|
||||||
|
|
||||||
eval_input_fn = input_fn_builder(
|
|
||||||
input_files=input_files,
|
|
||||||
max_seq_length=args.max_seq_length,
|
|
||||||
max_predictions_per_seq=args.max_predictions_per_seq,
|
|
||||||
is_training=False)
|
|
||||||
|
|
||||||
result = estimator.evaluate(
|
|
||||||
input_fn=eval_input_fn, steps=args.max_eval_steps)
|
|
||||||
|
|
||||||
output_eval_file = os.path.join(args.output_dir, "eval_results.txt")
|
|
||||||
with tf.gfile.GFile(output_eval_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__":
|
|
||||||
tf.app.run()
|
|
||||||
Reference in New Issue
Block a user