def main(_):
tf.logging.set_verbosity(tf.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.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(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.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, FLAGS.max_seq_length,
FLAGS.max_predictions_per_seq, output_files)
def subtoken_split(task_id):
#robust_tokens = load_robust_token()
token_reader = get_token_reader()
doc_id_list = get_doc_task(task_id)
num_doc = len(doc_id_list)
vocab_file = os.path.join(cpath.data_path, "bert_voca.txt")
tokenizer = tokenization.FullTokenizer(vocab_file=vocab_file,
do_lower_case=True)
window_size = 256 - 3
skip = int(window_size / 2)
ticker = TimeEstimator(num_doc)
doc_seg_info = {}
for key in doc_id_list:
tokens = token_reader.retrieve(key)
fn = tokenizer.wordpiece_tokenizer.tokenize
sub_tokens = list([fn(t) for t in tokens])
def move(loc, loc_sub, skip):
loc_idx = loc
num_passed_sw = 0
loc_sub_idx = loc_sub
for i in range(skip):
num_passed_sw += 1
loc_sub_idx += 1
if num_passed_sw == len(sub_tokens[loc_idx]):
loc_idx += 1
num_passed_sw = 0
if loc_idx >= len(sub_tokens):
break
# only move in token level
if num_passed_sw > 0:
loc_sub_idx -= num_passed_sw
return loc_idx, loc_sub_idx
loc = 0
loc_sub = 0
interval_list = []
while loc < len(tokens):
loc_ed, loc_sub_ed = move(loc, loc_sub, skip)
e = (loc, loc_ed), (loc_sub, loc_sub_ed)
interval_list.append(e)
loc = loc_ed
loc_sub = loc_sub_ed
doc_seg_info[key] = interval_list
ticker.tick()
p = os.path.join(cpath.data_path, "adhoc",
"robust_seg_info_{}.pickle".format(task_id))
pickle.dump(doc_seg_info, open(p, "wb"))
def __init__(self):
self.stopword = load_stopwords()
self.stemmer = CacheStemmer()
vocab_file = os.path.join(cpath.data_path, "bert_voca.txt")
self.tokenizer = tokenization.FullTokenizer(vocab_file=vocab_file,
do_lower_case=True)
tprint("Loading inv_index for robust")
self.collection = RobustCollection()
tprint("Done")
self.num_candidate = 10
def gen_robust_token():
collection = load_robust_ingham()
vocab_file = os.path.join(cpath.data_path, "bert_voca.txt")
tokenizer = tokenization.FullTokenizer(
vocab_file=vocab_file, do_lower_case=True)
ticker = TimeEstimator(len(collection))
nc = {}
for doc_id, content in collection.items():
nc[doc_id] = tokenizer.basic_tokenizer.tokenize(content)
ticker.tick()
token_path = os.path.join(cpath.data_path, "adhoc", "robust_tokens.pickle")
pickle.dump(nc, open(token_path, "wb"))
def save_title_tokens():
meta_path = os.path.join(cpath.data_path, "adhoc", "robust_meta.pickle")
meta = pickle.load(open(meta_path, "rb"))
vocab_file = os.path.join(cpath.data_path, "bert_voca.txt")
tokenizer = tokenization.FullTokenizer(vocab_file=vocab_file,
do_lower_case=True)
head_tokens = {}
ticker = TimeEstimator(len(meta))
for doc_id in meta:
date, headline = meta[doc_id]
h_tokens = tokenizer.basic_tokenizer.tokenize(headline)
head_tokens[doc_id] = h_tokens
ticker.tick()
save_path = os.path.join(cpath.data_path, "adhoc",
"robust_title_tokens.pickle")
pickle.dump(head_tokens, open(save_path, "wb"))
def __init__(self):
tprint("Pipeline Init")
self.stemmer = CacheStemmer()
vocab_file = os.path.join(cpath.data_path, "bert_voca.txt")
self.tokenizer = tokenization.FullTokenizer(vocab_file=vocab_file,
do_lower_case=True)
self.iteration_dir = "/mnt/scratch/youngwookim/data/tlm_iter1"
if not os.path.exists("/mnt/scratch/youngwookim/"):
self.iteration_dir = "/mnt/nfs/work3/youngwookim/data/tlm_iter1"
self.seg_max_seq = 256
self.model_max_seq = 512
self.rng = random.Random(0)
self.masked_lm_prob = 0.15
self.short_seq_prob = 0.1
self.inst_per_job = 1000
self.stopword = load_stopwords()
self.pr = FeatureExtractor(self.seg_max_seq - 3)
self.tf_record_maker = None
self.code_tick = CodeTiming()
tprint("Pipeline Init Done")
def main():
num_inst = 1000 * 1000
max_seq = 256
vocab_file = os.path.join(cpath.data_path, "bert_voca.txt")
tokenizer = tokenization.FullTokenizer(
vocab_file=vocab_file, do_lower_case=True)
sp = StreamPickler("robust_segments_", 1000)
ticker = TimeEstimator(num_inst)
for i in range(num_inst):
r = get_random_sent()
s_id, doc_id, loc, g_id, sent= r
doc_rows = get_doc_sent(doc_id)
target_tokens, sent_list, prev_tokens, next_tokens = extend(doc_rows, sent, loc, tokenizer, max_seq)
inst = target_tokens, sent_list, prev_tokens, next_tokens, doc_id
sp.add(inst)
ticker.tick()
sp.flush()
def __init__(self):
self.d_num_sub_tokens = dict()
vocab_file = os.path.join(cpath.data_path, "bert_voca.txt")
self.tokenizer = tokenization.FullTokenizer(vocab_file=vocab_file,
do_lower_case=True)
def train_nli():
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)
processor = MnliProcessor()
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)
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:
# 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)
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)
# 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.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 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)
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