def __init__(self, **kwargs):
self.tokenizer = tokenization.FullTokenizer(
vocab_file=kwargs['vocab_file'],
do_lower_case=True)
self.tag_to_id = {'O': 0, 'I-BRD': 1, 'I-PRO': 2, 'B-PRO': 3, 'I-KWD': 4, 'B-BRD': 5, 'I-POP': 6, 'B-KWD': 7,
'B-POP': 8, 'I-PRC': 9, 'I-FLR': 10, 'B-FLR': 11, 'B-PRC': 12, '[CLS]': 13, '[SEP]': 14}
self.id_to_tag = {0: 'O', 1: 'I-BRD', 2: 'I-PRO', 3: 'B-PRO', 4: 'I-KWD', 5: 'B-BRD', 6: 'I-POP', 7: 'B-KWD',
8: 'B-POP', 9: 'I-PRC', 10: 'I-FLR', 11: 'B-FLR', 12: 'B-PRC', 13: '[CLS]', 14: '[SEP]'}
def __init__(self, **kwargs):
self.tokenizer = tokenization.FullTokenizer(
vocab_file=kwargs['vocab_file'],
do_lower_case=True)
self.max_seq_len = 70
self.ckpt_path = kwargs['model_dir']
self.init_checkpoint = kwargs['init_checkpoint_file']
self.bert_config = kwargs['bert_config_dir']
self.graph = kwargs["graph"]
with self.graph.as_default():
self.model = Model(init_checkpoint_file=self.init_checkpoint, bert_config_dir=self.bert_config)
self.saver = tf.train.Saver()
config = tf.ConfigProto(log_device_placement=False)
self.session = tf.Session(graph=self.graph, config=config)
self.load()
def __init__(self):
self.lstm_dim = 128
self.batch_size = 1
self.max_seq_len = 70
self.clip = 5.0
self.dropout_keep = 0.5
self.optimizer = 'adam'
self.lr = 0.001
self.tag_schema = 'iob'
self.ckpt_path = '..\\models'
self.steps_check = 10
self.zeros = False
self.lower = True
self.max_epoch = 2
self.num_tags = len(convert_samples.tag_to_id)
self.model = Model(init_checkpoint_file='D:\models\\albert_base_zh\\albert_model.ckpt'
, bert_config_dir='D:\models\\albert_base_zh\\albert_config_base.json')
self.saver = tf.train.Saver()
self.tokenizer = tokenization.FullTokenizer(vocab_file='D:\models\\albert_base_zh\\vocab.txt',
do_lower_case=True)
def __init__(self, **kwargs):
self.tokenizer = tokenization.FullTokenizer(
vocab_file=kwargs['vocab_file'],
# vocab_file='D:\\mygit\\NER_MODEL\\albert_tiny_489k\\vocab.txt',
do_lower_case=True)
self.max_seq_len = 70
self.ckpt_path = kwargs['model_dir']
# self.ckpt_path = 'D:\\mygit\\NER_MODEL\\models'
self.init_checkpoint = kwargs['init_checkpoint_file']
# self.init_checkpoint = 'D:\\mygit\\NER_MODEL\\albert_tiny_489k\\albert_model.ckpt'
self.bert_config = kwargs['bert_config_dir']
# self.bert_config = 'D:\\mygit\\NER_MODEL\\albert_tiny_489k\\albert_config_tiny.json'
self.graph = kwargs["graph"]
with self.graph.as_default():
self.model = Model(init_checkpoint_file=self.init_checkpoint,
bert_config_dir=self.bert_config)
self.saver = tf.train.Saver()
config = tf.ConfigProto(log_device_placement=False,
gpu_options=gpu_options)
self.session = tf.Session(graph=self.graph, config=config)
self.load()
def main(_):
tf.logging.set_verbosity(tf.logging.INFO)
processors = {
"mayi": SelfProcessor_test
}
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.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
print("###tpu_cluster_resolver:", tpu_cluster_resolver)
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) # TODO
print("###length of total train_examples:", len(train_examples))
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)
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")
train_file_exists = os.path.exists(train_file)
print("###train_file_exists:", train_file_exists, " ;train_file:", train_file)
if not train_file_exists: # if tf_record file not exist, convert from raw text file. # TODO
file_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_eval:
eval_examples = processor.get_dev_examples(FLAGS.data_dir)
num_actual_eval_examples = len(eval_examples)
if FLAGS.use_tpu:
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)
tf.logging.info("***** Running evaluation *****")
tf.logging.info(" Num examples = %d (%d actual, %d padding)",
len(eval_examples), num_actual_eval_examples,
len(eval_examples) - num_actual_eval_examples)
tf.logging.info(" Batch size = %d", FLAGS.eval_batch_size)
eval_steps = None
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)
# evaluate all checkpoints; you can use the checkpoint with the best dev accuarcy
steps_and_files = []
filenames = tf.gfile.ListDirectory(FLAGS.output_dir)
for filename in filenames:
if filename.endswith(".index"):
ckpt_name = filename[:-6]
cur_filename = os.path.join(FLAGS.output_dir, ckpt_name)
global_step = int(cur_filename.split("-")[-1])
tf.logging.info("Add {} to eval list.".format(cur_filename))
steps_and_files.append([global_step, cur_filename])
steps_and_files = sorted(steps_and_files, key=lambda x: x[0])
output_eval_file = os.path.join(FLAGS.data_dir, "eval_results_albert_zh.txt")
print("output_eval_file:", output_eval_file)
tf.logging.info("output_eval_file:" + output_eval_file)
with tf.gfile.GFile(output_eval_file, "w") as writer:
for global_step, filename in sorted(steps_and_files, key=lambda x: x[0]):
result = estimator.evaluate(input_fn=eval_input_fn, steps=eval_steps, checkpoint_path=filename)
tf.logging.info("***** Eval results %s *****" % (filename))
writer.write("***** Eval results %s *****\n" % (filename))
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 FLAGS.do_predict:
predict_examples = processor.get_test_examples(FLAGS.data_dir)
num_actual_predict_examples = len(predict_examples)
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)
tf.logging.info("***** Running prediction*****")
tf.logging.info(" Num examples = %d (%d actual, %d padding)",
len(predict_examples), num_actual_predict_examples,
len(predict_examples) - num_actual_predict_examples)
tf.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.gfile.GFile(output_predict_file, "w") as writer:
num_written_lines = 0
tf.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