-
Notifications
You must be signed in to change notification settings - Fork 2
/
train.py
196 lines (163 loc) · 8.34 KB
/
train.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
# -*- coding: utf-8 -*-
import tensorflow as tf
import numpy as np
import os
import datetime
import time
from crnn import CRNN
from sklearn.metrics import f1_score
import data_helpers
import warnings
import sklearn.exceptions
warnings.filterwarnings("ignore", category=sklearn.exceptions.UndefinedMetricWarning)
# Parameters
# ==================================================
# Data loading params
tf.flags.DEFINE_string("train_dir", "SemEval2010_task8_all_data/SemEval2010_task8_training/TRAIN_FILE.TXT", "Path of train data")
tf.flags.DEFINE_float("dev_sample_percentage", .1, "Percentage of the training data to use for validation")
tf.flags.DEFINE_integer("max_sentence_length", 100, "Max sentence length in train(98)/test(70) data (Default: 100)")
# Model Hyperparameters
tf.flags.DEFINE_string("word2vec", "GoogleNews-vectors-negative300.bin", "Word2vec file with pre-trained embeddings") #
tf.flags.DEFINE_integer("text_embedding_dim", 300, "Dimensionality of character embedding (Default: 300)")
# tf.flags.DEFINE_integer("position_embedding_dim", 200, "Dimensionality of position embedding (Default: 100)")
# tf.flags.DEFINE_string("filter_sizes", "2,3,4,5", "Comma-separated filter sizes (Default: 2,3,4,5)")
# tf.flags.DEFINE_integer("num_filters", 128, "Number of filters per filter size (Default: 128)")
tf.flags.DEFINE_string("layers", "100", "Size of rnn output, no (Default: 100")
tf.flags.DEFINE_float("dropout_keep_prob", 0.5, "Dropout keep probability (Default: 0.5)")
tf.flags.DEFINE_string("pooling_type", "max", "pooling method, max or att (Default: max)")
tf.flags.DEFINE_float("l2_reg_lambda", 3.0, "L2 regularization lambda (Default: 3.0)")
tf.flags.DEFINE_integer("f1", 2, "f1 filter size (Default : 2)")
tf.flags.DEFINE_integer("f2", 5, "f2 filter size (Default : 5)")
tf.flags.DEFINE_integer("n_channels", 100, "the number of channels-output vector size, nc(Default : 100")
# Training parameters
tf.flags.DEFINE_integer("batch_size", 64, "Batch Size (Default: 64)")
tf.flags.DEFINE_integer("num_epochs", 30, "Number of training epochs (Default: 100)") # 100 epochs - 11290 steps
tf.flags.DEFINE_integer("display_every", 10, "Number of iterations to display training info.")
tf.flags.DEFINE_integer("evaluate_every", 100, "Evaluate model on dev set after this many steps")
tf.flags.DEFINE_integer("checkpoint_every", 100, "Save model after this many steps")
tf.flags.DEFINE_integer("num_checkpoints", 5, "Number of checkpoints to store")
tf.flags.DEFINE_float("learning_rate", 1e-3, "Which learning rate to start with. (Default: 1e-3)")
# Misc Parameters
tf.flags.DEFINE_boolean("allow_soft_placement", True, "Allow device soft device placement")
tf.flags.DEFINE_boolean("log_device_placement", False, "Log placement of ops on devices")
FLAGS = tf.flags.FLAGS
FLAGS._parse_flags()
print("\nParameters:")
for attr, value in sorted(FLAGS.__flags.items()):
print("{} = {}".format(attr.upper(), value))
print("")
def train():
with tf.device('/cpu:0'):
# x_text, pos1, pos2, y = data_helpers.load_data_and_labels(FLAGS.train_dir)
x_text, y = data_helpers.load_data_and_labels(FLAGS.train_dir)
# Build vocabulary
# Example: x_text[3] = "A misty <e1>ridge</e1> uprises from the <e2>surge</e2>."
# ['a misty ridge uprises from the surge <UNK> <UNK> ... <UNK>']
# =>
# [27 39 40 41 42 1 43 0 0 ... 0]
# dimension = FLAGS.max_sentence_length
text_vocab_processor = tf.contrib.learn.preprocessing.VocabularyProcessor(FLAGS.max_sentence_length)
text_vec = np.array(list(text_vocab_processor.fit_transform(x_text)))
print("Text Vocabulary Size: {:d}".format(len(text_vocab_processor.vocabulary_)))
# Example: pos1[3] = [-2 -1 0 1 2 3 4 999 999 999 ... 999]
# [95 96 97 98 99 100 101 999 999 999 ... 999]
# =>
# [11 12 13 14 15 16 21 17 17 17 ... 17]
# dimension = MAX_SENTENCE_LENGTH
# pos_vocab_processor = tf.contrib.learn.preprocessing.VocabularyProcessor(FLAGS.max_sentence_length)
# pos_vocab_processor.fit(pos1 + pos2)
# pos1_vec = np.array(list(pos_vocab_processor.transform(pos1)))
# pos2_vec = np.array(list(pos_vocab_processor.transform(pos2)))
# print("Position Vocabulary Size: {:d}".format(len(pos_vocab_processor.vocabulary_)))
# x = np.array([list(i) for i in zip(text_vec, pos1_vec, pos2_vec)])
x = np.array([list(i) for i in text_vec])
print("x = {0}".format(x.shape))
print("y = {0}".format(y.shape))
# Randomly shuffle data
np.random.seed(10)
shuffle_indices = np.random.permutation(np.arange(len(y)))
x_shuffled = x[shuffle_indices]
y_shuffled = y[shuffle_indices]
# Split train/test set
# TODO: This is very crude, should use cross-validation
dev_sample_index = -1 * int(FLAGS.dev_sample_percentage * float(len(y)))
x_train, x_dev = x_shuffled[:dev_sample_index], x_shuffled[dev_sample_index:]
# x_dev = np.array(x_dev).transpose((1, 0, 2))
y_train, y_dev = y_shuffled[:dev_sample_index], y_shuffled[dev_sample_index:]
print("Train/Dev split: {:d}/{:d}\n".format(len(y_train), len(y_dev)))
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.333)
with tf.Session(config=tf.ConfigProto(gpu_options=gpu_options)) as sess:
model = CRNN(layers=FLAGS.layers, max_length=FLAGS.max_sentence_length, n_classes=y.shape[1], pooling_type=FLAGS.pooling_type,
vocab_size=len(text_vocab_processor.vocabulary_), embedding_size=FLAGS.text_embedding_dim,
f1=FLAGS.f1, f2=FLAGS.f2, n_channels=FLAGS.n_channels)
# Output directory for models and summaries
timestamp = str(int(time.time()))
out_dir = os.path.abspath(os.path.join(os.path.curdir, "runs", timestamp))
print("Writing to {}\n".format(out_dir))
# Checkpoint directory. Tensorflow assumes this directory already exists so we need to create it
checkpoint_dir = os.path.abspath(os.path.join(out_dir, "checkpoints"))
checkpoint_prefix = os.path.join(checkpoint_dir, "model")
if not os.path.exists(checkpoint_dir):
os.makedirs(checkpoint_dir)
saver = tf.train.Saver(tf.global_variables(), max_to_keep=FLAGS.num_checkpoints)
# Write vocabulary
text_vocab_processor.save(os.path.join(out_dir, "text_vocab"))
# pos_vocab_processor.save(os.path.join(out_dir, "position_vocab"))
sess.run(tf.global_variables_initializer())
# Pre-trained word2vec
if FLAGS.word2vec:
# initial matrix with random uniform
initW = np.random.uniform(-0.25, 0.25, (len(text_vocab_processor.vocabulary_), FLAGS.text_embedding_dim))
# load any vectors from the word2vec
print("Load word2vec file {0}".format(FLAGS.word2vec))
with open(FLAGS.word2vec, "rb") as f:
header = f.readline()
vocab_size, layer1_size = map(int, header.split())
binary_len = np.dtype('float32').itemsize * layer1_size
for line in range(vocab_size):
word = []
while True:
ch = f.read(1).decode('latin-1')
if ch == ' ':
word = ''.join(word)
break
if ch != '\n':
word.append(ch)
idx = text_vocab_processor.vocabulary_.get(word)
if idx != 0:
initW[idx] = np.fromstring(f.read(binary_len), dtype='float32')
else:
f.read(binary_len)
sess.run(model.W_emb.assign(initW))
print("Success to load pre-trained word2vec model!\n")
batches = data_helpers.batch_iter(list(zip(x_train, y_train)), FLAGS.batch_size, FLAGS.num_epochs)
max_f1 = -1
for step, batch in enumerate(batches):
x_batch, y_batch = zip(*batch)
feed_dict = {model.input_text: x_batch, model.dropout_keep_prob: FLAGS.dropout_keep_prob, model.labels: y_batch}
# max_pooling, convs = sess.run([model.max_pooing, model.conv], feed_dict=feed_dict)
_, loss, accuracy = sess.run([model.train, model.cost, model.accuracy], feed_dict=feed_dict)
# Training log display
if step % FLAGS.display_every == 0:
print("step {}:, loss {}, acc {}".format(step, loss, accuracy))
# Evaluation
if step % FLAGS.evaluate_every == 0:
print("\nEvaluation:")
feed_dict = {
model.input_text: x_dev,
model.labels: y_dev,
model.dropout_keep_prob: 1.0
}
loss, accuracy, predictions = sess.run(
[model.cost, model.accuracy, model.predictions], feed_dict)
f1 = f1_score(np.argmax(y_dev, axis=1), predictions, average="macro")
print("step {}:, loss {}, acc {}, f1 {}\n".format(step, loss, accuracy, f1))
# Model checkpoint
if f1 > max_f1 * 0.99:
path = saver.save(sess, checkpoint_prefix, global_step=step)
print("Saved model checkpoint to {}\n".format(path))
max_f1 = f1
def main(_):
train()
if __name__ == "__main__":
tf.app.run()