def main(_):
tf.logging.set_verbosity(tf.logging.INFO)
news_config = GroverConfig.from_json_file(FLAGS.config_file)
tf.gfile.MakeDirs(FLAGS.output_dir)
input_files = []
print(FLAGS.input_file.split(","))
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)
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,
keep_checkpoint_max=None,
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(
news_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,
)
# # 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.train_batch_size,
params={'model_dir': FLAGS.output_dir})
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,
seq_length=FLAGS.max_seq_length,
is_training=True)
print("Start trainning.............................................")
estimator.train(input_fn=train_input_fn, max_steps=FLAGS.num_train_steps)
def main(_):
tf.logging.set_verbosity(tf.logging.INFO)
news_config = GroverConfig.from_json_file(FLAGS.config_file)
tf.gfile.MakeDirs(FLAGS.output_dir)
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)
my_per_process_gpu_memory_fraction = 1.0
gpu_options = tf.GPUOptions(
per_process_gpu_memory_fraction=my_per_process_gpu_memory_fraction)
sess_config = tf.ConfigProto(gpu_options=gpu_options)
run_config = tf.estimator.RunConfig(
model_dir=FLAGS.output_dir,
session_config=sess_config,
save_checkpoints_steps=FLAGS.save_checkpoints_steps,
keep_checkpoint_max=None)
model_fn = model_fn_builder(news_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)
# # If TPU is not available, this will fall back to normal Estimator on CPU
# # or GPU.
estimator = tf.estimator.Estimator(model_fn=model_fn, config=run_config)
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,
seq_length=FLAGS.max_seq_length,
is_training=True,
batch_size=FLAGS.train_batch_size)
print("Start trainning.............................................")
estimator.train(input_fn=train_input_fn, max_steps=FLAGS.num_train_steps)
def main(_):
tf.logging.set_verbosity(tf.logging.INFO)
news_config = GroverConfig.from_json_file(FLAGS.config_file)
tf.gfile.MakeDirs(FLAGS.output_dir)
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)
tpu_cluster_resolver = None
if FLAGS.use_tpu and FLAGS.tpu_name:
tpu_cluster_resolver = tf.distribute.cluster_resolver.TPUClusterResolver(
FLAGS.tpu_name, zone=FLAGS.tpu_zone, project=FLAGS.gcp_project)
is_per_host = tf.compat.v1.estimator.tpu.InputPipelineConfig.PER_HOST_V2
run_config = tf.compat.v1.estimator.tpu.RunConfig(
cluster=tpu_cluster_resolver,
master=FLAGS.master,
model_dir=FLAGS.output_dir,
save_checkpoints_steps=FLAGS.save_checkpoints_steps,
keep_checkpoint_max=None,
tpu_config=tf.compat.v1.estimator.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(news_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,
)
# If TPU is not available, this will fall back to normal Estimator on CPU
# or GPU.
estimator = tf.compat.v1.estimator.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.train_batch_size,
params={'model_dir': FLAGS.output_dir}
)
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,
seq_length=FLAGS.max_seq_length,
is_training=True)
try:
estimator.train(input_fn=train_input_fn, max_steps=FLAGS.num_train_steps)
except KeyboardInterrupt:
def serving_input_receiver_fn():
"""Serving input_fn that builds features from placeholders
Returns
-------
tf.estimator.export.ServingInputReceiver
"""
number = tf.placeholder(dtype=tf.int32, shape=[FLAGS.max_seq_length + 1], name='feature')
receiver_tensors = {'input_ids': number}
return tf.estimator.export.ServingInputReceiver(number, receiver_tensors)
export_path = estimator.export_saved_model("./model_save", serving_input_receiver_fn)