-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathtrain.py
151 lines (121 loc) · 6.78 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
import os
import sys
import shutil
sys.path.append("model")
sys.path.append("utils")
import numpy as np
import tensorflow as tf
from vae import VAE
from config import FLAGS
from batchloader import BatchLoader
def log_and_print(log_file, logstr, br=True):
print(logstr)
if(br):
logstr = logstr + "\n"
with open(log_file, 'a') as f:
f.write(logstr)
def main():
os.mkdir(FLAGS.LOG_DIR)
os.mkdir(FLAGS.LOG_DIR + "/model")
log_file = FLAGS.LOG_DIR + "/log.txt"
shutil.copyfile("config.py", FLAGS.LOG_DIR + "/config.py")
# gpu memory
sess_conf = tf.ConfigProto(
gpu_options = tf.GPUOptions(
# allow_growth = True
)
)
with tf.Graph().as_default():
with tf.Session(config=sess_conf) as sess:
batchloader = BatchLoader(with_label=False)
with tf.variable_scope("VAE"):
vae = VAE(batchloader, is_training=True, ru=False)
with tf.variable_scope("VAE", reuse=True):
vae_test = VAE(batchloader, is_training=False, ru=True)
saver = tf.train.Saver()
summary_writer = tf.summary.FileWriter(FLAGS.LOG_DIR, sess.graph)
sess.run(tf.global_variables_initializer())
log_and_print(log_file, "start training")
loss_sum = []
rnn_loss_sum = []
aux_loss_sum = []
kld_sum = []
lr = FLAGS.LEARNING_RATE
step = 0
for epoch in range(FLAGS.EPOCH):
log_and_print(log_file, "epoch %d" % (epoch+1))
if epoch >= FLAGS.LR_DECAY_START:
lr *= 0.95
for batch in range(FLAGS.BATCHES_PER_EPOCH):
step += 1
encoder_input, decoder_input, target = \
batchloader.next_batch(FLAGS.BATCH_SIZE, "train")
feed_dict = {vae.encoder_input: encoder_input,
vae.decoder_input: decoder_input,
vae.target: target,
vae.step: step,
vae.lr: lr}
aux_logits, rnn_logits, loss, rnn_loss, aux_loss, kld, merged_summary, _ \
= sess.run([vae.aux_logits, vae.rnn_logits, vae.loss, vae.rnn_loss,
vae.aux_loss, vae.kld, vae.merged_summary, vae.train_op],
feed_dict = feed_dict)
rnn_loss_sum.append(rnn_loss)
aux_loss_sum.append(aux_loss)
kld_sum.append(kld)
loss_sum.append(loss)
summary_writer.add_summary(merged_summary, step)
if(batch % 50 == 49):
log_and_print(log_file, "epoch %d batch %d" % \
((epoch+1), (batch+1)), br=False)
ave_loss = np.average(loss_sum)
log_and_print(log_file, "\tloss: %f" % ave_loss, br=False)
ave_rnnloss = np.average(rnn_loss_sum)
log_and_print(log_file, "\trnn_loss: %f" % ave_rnnloss, br=False)
ave_auxloss = np.average(aux_loss_sum)
log_and_print(log_file, "\taux_loss: %f" % ave_auxloss, br=False)
ave_kld = np.average(kld_sum)
log_and_print(log_file, "\tkld %f" % ave_kld, br=False)
loss_sum = []
rnn_loss_sum = []
aux_loss_sum = []
kld_sum = []
# train input, output
# output input and logits
sample_train_input, sample_train_input_list \
= sess.run([vae.encoder_input, vae.encoder_input_list],
feed_dict = feed_dict)
encoder_input_texts = batchloader.logits2str(sample_train_input_list,
1,
onehot=False,
numpy=True)
log_and_print(log_file, "\ttrain input: %s" % encoder_input_texts[0])
sample_train_rnn_outputs = batchloader.logits2str(rnn_logits, 1)
sample_train_aux_outputs = batchloader.logits2str(aux_logits, 1)
log_and_print(log_file, "\ttrain rnn output: %s" % sample_train_rnn_outputs[0])
log_and_print(log_file, "\ttrain aux output: %s" % sample_train_aux_outputs[0])
# validation output
sample_input, _, sample_target = batchloader.next_batch(FLAGS.BATCH_SIZE, "test")
sample_input_list, sample_latent_variables = \
sess.run([vae_test.encoder_input_list, vae_test.encoder.latent_variables],
feed_dict = {vae_test.encoder_input: sample_input})
sample_aux_logits, sample_rnn_logits, valid_aux_loss, \
valid_rnn_loss, merged_summary = \
sess.run([vae_test.aux_logits, vae_test.rnn_logits,
vae_test.aux_loss, vae_test.rnn_loss, vae_test.merged_summary],
feed_dict = {vae_test.target: sample_target,
vae_test.latent_variables: sample_latent_variables})
log_and_print(log_file, "\tvalid rnn loss: %f" % valid_rnn_loss)
log_and_print(log_file, "\tvalid aux loss: %f" % valid_aux_loss)
sample_input_texts = batchloader.logits2str(sample_input_list,
1, onehot=False, numpy=True)
sample_rnn_samples = batchloader.logits2str(sample_rnn_logits, 1)
sample_aux_samples = batchloader.logits2str(sample_aux_logits, 1)
log_and_print(log_file, "\tsample input: %s" % sample_input_texts[0])
log_and_print(log_file, "\tsample rnn output: %s" % sample_rnn_samples[0])
log_and_print(log_file, "\tsample aux output: %s" % sample_aux_samples[0])
summary_writer.add_summary(merged_summary, step)
# save model
save_path = saver.save(sess, FLAGS.LOG_DIR + ("/model/model%d.ckpt" % (epoch+1)))
log_and_print(log_file, "Model saved in file %s" % save_path)
if __name__ == "__main__":
main()