def test_siamese_bert_model(self):
bert_config = modeling.BertConfig.from_json_file(
"../bert_model/data/chinese_L-12_H-768_A-12/bert_config.json")
tokenizer = tokenization.FullTokenizer(
"../bert_model/data/chinese_L-12_H-768_A-12/vocab.txt",
do_lower_case=True)
max_seq_length = 128
batch_size = 8
bert_siamese_model = run_pretraining.SiameseBertModel(
config=bert_config,
is_training=True,
init_checkpoint=
"../bert_model/data/chinese_L-12_H-768_A-12/bert_model.ckpt")
bert_siamese_model.compile(optimizer=tf.keras.optimizers.Adam(
learning_rate=0.0001))
siamese_processor = SiameseProcessor(tokenizer=tokenizer,
max_seq_length=max_seq_length)
lines = [("我想买保险", "我还想买保险", "0"), ("我想买保险", "我不想买保险", "1"),
("我想去北京玩耍", "我打算去北京玩耍", "0"), ("我想去北京玩耍", "我不想去北京玩耍", "1"),
("能不能借钱给我呢", "可以借钱给我吗", "0"), ("我想借你钱", "我不想借你钱", "1")]
examples = siamese_processor._create_examples(lines, set_type="dev")
data = siamese_processor.convert_examples_to_features(examples)
siamese_processor.write_features_into_tfrecord(
filename="./siamese_data", features=data)
d = siamese_processor.read_features_from_tfrecord(
filename="./siamese_data", batch_size=6)
d = d.repeat()
bert_siamese_model.fit(d, epochs=5, steps_per_epoch=20)
bert_siamese_model.evaluate(d, steps=3)
print(bert_siamese_model.predict(d, steps=1))
def main(_):
import logging
tf.get_logger().setLevel(logging.INFO)
tokenizer = tokenization.FullTokenizer(vocab_file=FLAGS.vocab_file,
do_lower_case=FLAGS.do_lower_case)
input_files = []
for input_pattern in FLAGS.input_file.split(","):
input_files.extend(tf.io.gfile.glob(input_pattern))
tf.get_logger().info("*** Reading from input files ***")
for input_file in input_files:
tf.get_logger().info(" %s", input_file)
rng = random.Random(FLAGS.random_seed)
instances = create_training_instances(input_files, tokenizer,
FLAGS.max_seq_length,
FLAGS.dupe_factor,
FLAGS.short_seq_prob,
FLAGS.masked_lm_prob,
FLAGS.max_predictions_per_seq, rng)
output_files = FLAGS.output_file.split(",")
tf.get_logger().info("*** Writing to output files ***")
for output_file in output_files:
tf.get_logger().info(" %s", output_file)
write_instance_to_example_files(instances, tokenizer, FLAGS.max_seq_length,
FLAGS.max_predictions_per_seq,
output_files)
def __init__(self,
vocab_file,
masked_lm_prob=0.15,
max_predictions_per_seq=20,
do_lower_case=True,
max_seq_length=128,
dupe_factor=10,
short_seq_prob=0.1,
do_whole_word=False,
random_seed=100):
self.vocab_file = vocab_file
self.do_lower_case = do_lower_case
self.max_seq_length = max_seq_length
self.sample = Sample(do_whole_word=do_whole_word)
self.dupe_factor = dupe_factor
self.masked_lm_prob = masked_lm_prob
self.max_predictions_per_seq = max_predictions_per_seq
self.short_seq_prob = short_seq_prob
self.random_seed = random_seed
self.tokenizer = tokenization.FullTokenizer(
vocab_file=self.vocab_file, do_lower_case=self.do_lower_case)
def test_full_tokenizer(self):
vocab_tokens = [
"[UNK]", "[CLS]", "[SEP]", "want", "##want", "##ed", "wa", "un",
"runn", "##ing", ","
]
with tempfile.NamedTemporaryFile(delete=False) as vocab_writer:
if six.PY2:
vocab_writer.write("".join([x + "\n" for x in vocab_tokens]))
else:
vocab_writer.write("".join([x + "\n" for x in vocab_tokens
]).encode("utf-8"))
vocab_file = vocab_writer.name
tokenizer = tokenization.FullTokenizer(vocab_file)
os.unlink(vocab_file)
tokens = tokenizer.tokenize(u"UNwant\u00E9d,running")
self.assertAllEqual(tokens,
["un", "##want", "##ed", ",", "runn", "##ing"])
self.assertAllEqual(tokenizer.convert_tokens_to_ids(tokens),
[7, 4, 5, 10, 8, 9])
def main(_):
absl.logging.set_verbosity(absl.logging.INFO)
layer_indexes = [int(x) for x in FLAGS.layers.split(",")]
bert_config = modeling.BertConfig.from_json_file(FLAGS.bert_config_file)
tokenizer = tokenization.FullTokenizer(vocab_file=FLAGS.vocab_file,
do_lower_case=FLAGS.do_lower_case)
"""
is_per_host = tf.contrib.tpu.InputPipelineConfig.PER_HOST_V2
run_config = tf.contrib.tpu.RunConfig(
master=FLAGS.master,
tpu_config=tf.contrib.tpu.TPUConfig(
num_shards=FLAGS.num_tpu_cores,
per_host_input_for_training=is_per_host))
"""
run_config = tf.estimator.RunConfig()
examples = read_examples(FLAGS.input_file)
features = convert_examples_to_features(examples=examples,
seq_length=FLAGS.max_seq_length,
tokenizer=tokenizer)
unique_id_to_feature = {}
for feature in features:
unique_id_to_feature[feature.unique_id] = feature
model_fn = model_fn_builder(
bert_config=bert_config,
init_checkpoint=FLAGS.init_checkpoint,
layer_indexes=layer_indexes,
use_tpu=FLAGS.use_tpu,
use_one_hot_embeddings=FLAGS.use_one_hot_embeddings)
# If TPU is not available, this will fall back to normal Estimator on CPU
# or GPU.
# modify begin
"""
estimator = tf.contrib.tpu.TPUEstimator(
use_tpu=FLAGS.use_tpu,
model_fn=model_fn,
config=run_config,
predict_batch_size=FLAGS.batch_size)
"""
#modify end
estimator = tf.estimator.Estimator(model_fn=model_fn,
config=run_config,
params={
"batch_size": FLAGS.batch_size,
})
input_fn = input_fn_builder(features=features,
seq_length=FLAGS.max_seq_length)
with codecs.getwriter("utf-8")(tf.io.gfile.GFile(FLAGS.output_file,
"w")) as writer:
for result in estimator.predict(input_fn, yield_single_examples=True):
unique_id = int(result["unique_id"])
feature = unique_id_to_feature[unique_id]
output_json = collections.OrderedDict()
output_json["linex_index"] = unique_id
all_features = []
for (i, token) in enumerate(feature.tokens):
all_layers = []
for (j, layer_index) in enumerate(layer_indexes):
layer_output = result["layer_output_%d" % j]
layers = collections.OrderedDict()
layers["index"] = layer_index
layers["values"] = [
round(float(x), 6)
for x in layer_output[i:(i + 1)].flat
]
all_layers.append(layers)
features = collections.OrderedDict()
features["token"] = token
features["layers"] = all_layers
all_features.append(features)
output_json["features"] = all_features
writer.write(json.dumps(output_json) + "\n")
def main(_):
logging.getLogger().setLevel(logging.INFO)
processors = {
"cola": ColaProcessor,
"mnli": MnliProcessor,
"mrpc": MrpcProcessor,
"xnli": XnliProcessor,
}
tokenization.validate_case_matches_checkpoint(FLAGS.do_lower_case,
FLAGS.init_checkpoint)
if not FLAGS.do_train and not FLAGS.do_eval and not FLAGS.do_predict:
raise ValueError(
"At least one of `do_train`, `do_eval` 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.io.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),
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)
# 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,
predict_batch_size=FLAGS.predict_batch_size)
if FLAGS.do_train:
train_file = os.path.join(FLAGS.output_dir, "train.tf_record")
file_based_convert_examples_to_features(
train_examples, label_list, FLAGS.max_seq_length, tokenizer, train_file)
logging.info("***** Running training *****")
logging.info(" Num examples = %d", len(train_examples))
logging.info(" Batch size = %d", FLAGS.train_batch_size)
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_eval:
eval_examples = processor.get_dev_examples(FLAGS.data_dir)
num_actual_eval_examples = len(eval_examples)
if FLAGS.use_tpu:
# TPU requires a fixed batch size for all batches, therefore the number
# of examples must be a multiple of the batch size, or else examples
# will get dropped. So we pad with fake examples which are ignored
# later on. These do NOT count towards the metric (all tf.metrics
# support a per-instance weight, and these get a weight of 0.0).
while len(eval_examples) % FLAGS.eval_batch_size != 0:
eval_examples.append(PaddingInputExample())
eval_file = os.path.join(FLAGS.output_dir, "eval.tf_record")
file_based_convert_examples_to_features(
eval_examples, label_list, FLAGS.max_seq_length, tokenizer, eval_file)
logging.info("***** Running evaluation *****")
logging.info(" Num examples = %d (%d actual, %d padding)",
len(eval_examples), num_actual_eval_examples,
len(eval_examples) - num_actual_eval_examples)
logging.info(" Batch size = %d", FLAGS.eval_batch_size)
# This tells the estimator to run through the entire set.
eval_steps = None
# However, if running eval on the TPU, you will need to specify the
# number of steps.
if FLAGS.use_tpu:
assert len(eval_examples) % FLAGS.eval_batch_size == 0
eval_steps = int(len(eval_examples) // FLAGS.eval_batch_size)
eval_drop_remainder = True if FLAGS.use_tpu else False
eval_input_fn = file_based_input_fn_builder(
input_file=eval_file,
seq_length=FLAGS.max_seq_length,
is_training=False,
drop_remainder=eval_drop_remainder)
result = estimator.evaluate(input_fn=eval_input_fn, steps=eval_steps)
output_eval_file = os.path.join(FLAGS.output_dir, "eval_results.txt")
with tf.io.gfile.GFile(output_eval_file, "w") as writer:
logging.info("***** Eval results *****")
for key in sorted(result.keys()):
logging.info(" %s = %s", key, str(result[key]))
writer.write("%s = %s\n" % (key, str(result[key])))
if FLAGS.do_predict:
predict_examples = processor.get_test_examples(FLAGS.data_dir)
num_actual_predict_examples = len(predict_examples)
if FLAGS.use_tpu:
# TPU requires a fixed batch size for all batches, therefore the number
# of examples must be a multiple of the batch size, or else examples
# will get dropped. So we pad with fake examples which are ignored
# later on.
while len(predict_examples) % FLAGS.predict_batch_size != 0:
predict_examples.append(PaddingInputExample())
predict_file = os.path.join(FLAGS.output_dir, "predict.tf_record")
file_based_convert_examples_to_features(predict_examples, label_list,
FLAGS.max_seq_length, tokenizer,
predict_file)
logging.info("***** Running prediction*****")
logging.info(" Num examples = %d (%d actual, %d padding)",
len(predict_examples), num_actual_predict_examples,
len(predict_examples) - num_actual_predict_examples)
logging.info(" Batch size = %d", FLAGS.predict_batch_size)
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)
output_predict_file = os.path.join(FLAGS.output_dir, "test_results.tsv")
with tf.io.gfile.GFile(output_predict_file, "w") as writer:
num_written_lines = 0
logging.info("***** Predict results *****")
for (i, prediction) in enumerate(result):
probabilities = prediction["probabilities"]
if i >= num_actual_predict_examples:
break
output_line = "\t".join(
str(class_probability)
for class_probability in probabilities) + "\n"
writer.write(output_line)
num_written_lines += 1
assert num_written_lines == num_actual_predict_examples
else:
d = tf.data.TFRecordDataset(input_files)
d = d.map(lambda record: about_tfrecord.parse_example(
record, name_to_features))
d = d.map(lambda x: add_attention_mask(x, is_training))
d = d.batch(batch_size=batch_size, drop_remainder=True)
return d
if __name__ == "__main__":
input_file = "../bert/sample_text.txt"
output_file = "test_data/pretraining_data"
vocab_file = "./test_data/vocab.txt"
tokenizer = tokenization.FullTokenizer(vocab_file=vocab_file,
do_lower_case=True)
ptdm = PreTrainingDataMan(vocab_file=vocab_file,
max_seq_length=128,
max_predictions_per_seq=15)
if not os.path.exists(output_file):
ptdm.create_pretraining_data(input_file, output_file)
dataset = ptdm.read_data_from_tfrecord(output_file,
is_training=True,
batch_size=1)
i = 0
for data in dataset:
print(data)
ids = data["input_ids"].numpy()
print(tokenizer.convert_ids_to_tokens(ids[0]))
i += 1
print(i)