-
Notifications
You must be signed in to change notification settings - Fork 88
/
MUNIT.py
606 lines (458 loc) · 26.9 KB
/
MUNIT.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
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
from ops import *
from utils import *
from glob import glob
import time
from tensorflow.contrib.data import batch_and_drop_remainder
class MUNIT(object) :
def __init__(self, sess, args):
self.model_name = 'MUNIT'
self.sess = sess
self.checkpoint_dir = args.checkpoint_dir
self.result_dir = args.result_dir
self.log_dir = args.log_dir
self.sample_dir = args.sample_dir
self.dataset_name = args.dataset
self.augment_flag = args.augment_flag
self.epoch = args.epoch
self.iteration = args.iteration
self.gan_type = args.gan_type
self.batch_size = args.batch_size
self.print_freq = args.print_freq
self.save_freq = args.save_freq
self.num_style = args.num_style # for test
self.guide_img = args.guide_img
self.direction = args.direction
self.img_h = args.img_h
self.img_w = args.img_w
self.img_ch = args.img_ch
self.init_lr = args.lr
self.ch = args.ch
""" Weight """
self.gan_w = args.gan_w
self.recon_x_w = args.recon_x_w
self.recon_s_w = args.recon_s_w
self.recon_c_w = args.recon_c_w
self.recon_x_cyc_w = args.recon_x_cyc_w
""" Generator """
self.n_res = args.n_res
self.mlp_dim = pow(2, args.n_sample) * args.ch # default : 256
self.n_downsample = args.n_sample
self.n_upsample = args.n_sample
self.style_dim = args.style_dim
""" Discriminator """
self.n_dis = args.n_dis
self.n_scale = args.n_scale
self.sample_dir = os.path.join(args.sample_dir, self.model_dir)
check_folder(self.sample_dir)
self.trainA_dataset = glob('./dataset/{}/*.*'.format(self.dataset_name + '/trainA'))
self.trainB_dataset = glob('./dataset/{}/*.*'.format(self.dataset_name + '/trainB'))
self.dataset_num = max(len(self.trainA_dataset), len(self.trainB_dataset))
print("##### Information #####")
print("# gan type : ", self.gan_type)
print("# dataset : ", self.dataset_name)
print("# max dataset number : ", self.dataset_num)
print("# batch_size : ", self.batch_size)
print("# epoch : ", self.epoch)
print("# iteration per epoch : ", self.iteration)
print("# style in test phase : ", self.num_style)
print()
print("##### Generator #####")
print("# residual blocks : ", self.n_res)
print("# Style dimension : ", self.style_dim)
print("# MLP dimension : ", self.mlp_dim)
print("# Down sample : ", self.n_downsample)
print("# Up sample : ", self.n_upsample)
print()
print("##### Discriminator #####")
print("# Discriminator layer : ", self.n_dis)
print("# Multi-scale Dis : ", self.n_scale)
##################################################################################
# Encoder and Decoders
##################################################################################
def Style_Encoder(self, x, reuse=False, scope='style_encoder'):
# IN removes the original feature mean and variance that represent important style information
channel = self.ch
with tf.variable_scope(scope, reuse=reuse) :
x = conv(x, channel, kernel=7, stride=1, pad=3, pad_type='reflect', scope='conv_0')
x = relu(x)
for i in range(2) :
x = conv(x, channel*2, kernel=4, stride=2, pad=1, pad_type='reflect', scope='conv_'+str(i+1))
x = relu(x)
channel = channel * 2
for i in range(2) :
x = conv(x, channel, kernel=4, stride=2, pad=1, pad_type='reflect', scope='down_conv_'+str(i))
x = relu(x)
x = adaptive_avg_pooling(x) # global average pooling
x = conv(x, self.style_dim, kernel=1, stride=1, scope='SE_logit')
return x
def Content_Encoder(self, x, reuse=False, scope='content_encoder'):
channel = self.ch
with tf.variable_scope(scope, reuse=reuse) :
x = conv(x, channel, kernel=7, stride=1, pad=3, pad_type='reflect', scope='conv_0')
x = instance_norm(x, scope='ins_0')
x = relu(x)
for i in range(self.n_downsample) :
x = conv(x, channel*2, kernel=4, stride=2, pad=1, pad_type='reflect', scope='conv_'+str(i+1))
x = instance_norm(x, scope='ins_'+str(i+1))
x = relu(x)
channel = channel * 2
for i in range(self.n_res) :
x = resblock(x, channel, scope='resblock_'+str(i))
return x
def generator(self, contents, style, reuse=False, scope="decoder"):
channel = self.mlp_dim
with tf.variable_scope(scope, reuse=reuse) :
mu, var = self.MLP(style)
x = contents
for i in range(self.n_res) :
idx = 2 * i
x = adaptive_resblock(x, channel, mu[idx], var[idx], mu[idx + 1], var[idx + 1], scope='adaptive_resblock_'+str(i))
for i in range(self.n_upsample) :
# # IN removes the original feature mean and variance that represent important style information
x = up_sample(x, scale_factor=2)
x = conv(x, channel//2, kernel=5, stride=1, pad=2, pad_type='reflect', scope='conv_'+str(i))
x = layer_norm(x, scope='layer_norm_'+str(i))
x = relu(x)
channel = channel // 2
x = conv(x, channels=self.img_ch, kernel=7, stride=1, pad=3, pad_type='reflect', scope='G_logit')
x = tanh(x)
return x
def MLP(self, style, scope='MLP'):
channel = self.mlp_dim
with tf.variable_scope(scope) :
x = style
for i in range(2):
x = fully_connected(x, channel, scope='FC_' + str(i))
x = relu(x)
mu_list = []
var_list = []
for i in range(self.n_res * 2):
mu = fully_connected(x, channel, scope='FC_mu_' + str(i))
var = fully_connected(x, channel, scope='FC_var_' + str(i))
mu = tf.reshape(mu, shape=[-1, 1, 1, channel])
var = tf.reshape(var, shape=[-1, 1, 1, channel])
mu_list.append(mu)
var_list.append(var)
return mu_list, var_list
##################################################################################
# Discriminator
##################################################################################
def discriminator(self, x_init, reuse=False, scope="discriminator"):
D_logit = []
with tf.variable_scope(scope, reuse=reuse) :
for scale in range(self.n_scale) :
channel = self.ch
x = conv(x_init, channel, kernel=4, stride=2, pad=1, pad_type='reflect', scope='ms_' + str(scale) + 'conv_0')
x = lrelu(x, 0.2)
for i in range(1, self.n_dis):
x = conv(x, channel * 2, kernel=4, stride=2, pad=1, pad_type='reflect', scope='ms_' + str(scale) +'conv_' + str(i))
x = lrelu(x, 0.2)
channel = channel * 2
x = conv(x, channels=1, kernel=1, stride=1, scope='ms_' + str(scale) + 'D_logit')
D_logit.append(x)
x_init = down_sample(x_init)
return D_logit
##################################################################################
# Model
##################################################################################
def Encoder_A(self, x_A, reuse=False):
style_A = self.Style_Encoder(x_A, reuse=reuse, scope='style_encoder_A')
content_A = self.Content_Encoder(x_A, reuse=reuse, scope='content_encoder_A')
return content_A, style_A
def Encoder_B(self, x_B, reuse=False):
style_B = self.Style_Encoder(x_B, reuse=reuse, scope='style_encoder_B')
content_B = self.Content_Encoder(x_B, reuse=reuse, scope='content_encoder_B')
return content_B, style_B
def Decoder_A(self, content_B, style_A, reuse=False):
x_ba = self.generator(contents=content_B, style=style_A, reuse=reuse, scope='decoder_A')
return x_ba
def Decoder_B(self, content_A, style_B, reuse=False):
x_ab = self.generator(contents=content_A, style=style_B, reuse=reuse, scope='decoder_B')
return x_ab
def discriminate_real(self, x_A, x_B):
real_A_logit = self.discriminator(x_A, scope="discriminator_A")
real_B_logit = self.discriminator(x_B, scope="discriminator_B")
return real_A_logit, real_B_logit
def discriminate_fake(self, x_ba, x_ab):
fake_A_logit = self.discriminator(x_ba, reuse=True, scope="discriminator_A")
fake_B_logit = self.discriminator(x_ab, reuse=True, scope="discriminator_B")
return fake_A_logit, fake_B_logit
def build_model(self):
self.lr = tf.placeholder(tf.float32, name='learning_rate')
""" Input Image"""
Image_Data_Class = ImageData(self.img_h, self.img_w, self.img_ch, self.augment_flag)
trainA = tf.data.Dataset.from_tensor_slices(self.trainA_dataset)
trainB = tf.data.Dataset.from_tensor_slices(self.trainB_dataset)
trainA = trainA.prefetch(self.batch_size).shuffle(self.dataset_num).map(Image_Data_Class.image_processing, num_parallel_calls=8).apply(batch_and_drop_remainder(self.batch_size)).repeat()
trainB = trainB.prefetch(self.batch_size).shuffle(self.dataset_num).map(Image_Data_Class.image_processing, num_parallel_calls=8).apply(batch_and_drop_remainder(self.batch_size)).repeat()
trainA_iterator = trainA.make_one_shot_iterator()
trainB_iterator = trainB.make_one_shot_iterator()
self.domain_A = trainA_iterator.get_next()
self.domain_B = trainB_iterator.get_next()
""" Define Encoder, Generator, Discriminator """
self.style_a = tf.placeholder(tf.float32, shape=[self.batch_size, 1, 1, self.style_dim], name='style_a')
self.style_b = tf.placeholder(tf.float32, shape=[self.batch_size, 1, 1, self.style_dim], name='style_b')
# encode
content_a, style_a_prime = self.Encoder_A(self.domain_A)
content_b, style_b_prime = self.Encoder_B(self.domain_B)
# decode (within domain)
x_aa = self.Decoder_A(content_B=content_a, style_A=style_a_prime)
x_bb = self.Decoder_B(content_A=content_b, style_B=style_b_prime)
# decode (cross domain)
x_ba = self.Decoder_A(content_B=content_b, style_A=self.style_a, reuse=True)
x_ab = self.Decoder_B(content_A=content_a, style_B=self.style_b, reuse=True)
# encode again
content_b_, style_a_ = self.Encoder_A(x_ba, reuse=True)
content_a_, style_b_ = self.Encoder_B(x_ab, reuse=True)
# decode again (if needed)
if self.recon_x_cyc_w > 0 :
x_aba = self.Decoder_A(content_B=content_a_, style_A=style_a_prime, reuse=True)
x_bab = self.Decoder_B(content_A=content_b_, style_B=style_b_prime, reuse=True)
cyc_recon_A = L1_loss(x_aba, self.domain_A)
cyc_recon_B = L1_loss(x_bab, self.domain_B)
else :
cyc_recon_A = 0.0
cyc_recon_B = 0.0
real_A_logit, real_B_logit = self.discriminate_real(self.domain_A, self.domain_B)
fake_A_logit, fake_B_logit = self.discriminate_fake(x_ba, x_ab)
""" Define Loss """
G_ad_loss_a = generator_loss(self.gan_type, fake_A_logit)
G_ad_loss_b = generator_loss(self.gan_type, fake_B_logit)
D_ad_loss_a = discriminator_loss(self.gan_type, real_A_logit, fake_A_logit)
D_ad_loss_b = discriminator_loss(self.gan_type, real_B_logit, fake_B_logit)
recon_A = L1_loss(x_aa, self.domain_A) # reconstruction
recon_B = L1_loss(x_bb, self.domain_B) # reconstruction
# The style reconstruction loss encourages
# diverse outputs given different style codes
recon_style_A = L1_loss(style_a_, self.style_a)
recon_style_B = L1_loss(style_b_, self.style_b)
# The content reconstruction loss encourages
# the translated image to preserve semantic content of the input image
recon_content_A = L1_loss(content_a_, content_a)
recon_content_B = L1_loss(content_b_, content_b)
Generator_A_loss = self.gan_w * G_ad_loss_a + \
self.recon_x_w * recon_A + \
self.recon_s_w * recon_style_A + \
self.recon_c_w * recon_content_A + \
self.recon_x_cyc_w * cyc_recon_A
Generator_B_loss = self.gan_w * G_ad_loss_b + \
self.recon_x_w * recon_B + \
self.recon_s_w * recon_style_B + \
self.recon_c_w * recon_content_B + \
self.recon_x_cyc_w * cyc_recon_B
Discriminator_A_loss = self.gan_w * D_ad_loss_a
Discriminator_B_loss = self.gan_w * D_ad_loss_b
self.Generator_loss = Generator_A_loss + Generator_B_loss + regularization_loss('encoder') + regularization_loss('decoder')
self.Discriminator_loss = Discriminator_A_loss + Discriminator_B_loss + regularization_loss('discriminator')
""" Training """
t_vars = tf.trainable_variables()
G_vars = [var for var in t_vars if 'decoder' in var.name or 'encoder' in var.name]
D_vars = [var for var in t_vars if 'discriminator' in var.name]
self.G_optim = tf.train.AdamOptimizer(self.lr, beta1=0.5, beta2=0.999).minimize(self.Generator_loss, var_list=G_vars)
self.D_optim = tf.train.AdamOptimizer(self.lr, beta1=0.5, beta2=0.999).minimize(self.Discriminator_loss, var_list=D_vars)
"""" Summary """
self.all_G_loss = tf.summary.scalar("Generator_loss", self.Generator_loss)
self.all_D_loss = tf.summary.scalar("Discriminator_loss", self.Discriminator_loss)
self.G_A_loss = tf.summary.scalar("G_A_loss", Generator_A_loss)
self.G_B_loss = tf.summary.scalar("G_B_loss", Generator_B_loss)
self.D_A_loss = tf.summary.scalar("D_A_loss", Discriminator_A_loss)
self.D_B_loss = tf.summary.scalar("D_B_loss", Discriminator_B_loss)
self.G_loss = tf.summary.merge([self.G_A_loss, self.G_B_loss, self.all_G_loss])
self.D_loss = tf.summary.merge([self.D_A_loss, self.D_B_loss, self.all_D_loss])
""" Image """
self.fake_A = x_ba
self.fake_B = x_ab
self.real_A = self.domain_A
self.real_B = self.domain_B
""" Test """
self.test_image = tf.placeholder(tf.float32, [1, self.img_h, self.img_w, self.img_ch], name='test_image')
self.test_style = tf.placeholder(tf.float32, [1, 1, 1, self.style_dim], name='test_style')
test_content_a, _ = self.Encoder_A(self.test_image, reuse=True)
test_content_b, _ = self.Encoder_B(self.test_image, reuse=True)
self.test_fake_A = self.Decoder_A(content_B=test_content_b, style_A=self.test_style, reuse=True)
self.test_fake_B = self.Decoder_B(content_A=test_content_a, style_B=self.test_style, reuse=True)
""" Guided Image Translation """
self.content_image = tf.placeholder(tf.float32, [1, self.img_h, self.img_w, self.img_ch], name='content_image')
self.style_image = tf.placeholder(tf.float32, [1, self.img_h, self.img_w, self.img_ch], name='guide_style_image')
if self.direction == 'a2b' :
guide_content_A, guide_style_A = self.Encoder_A(self.content_image, reuse=True)
guide_content_B, guide_style_B = self.Encoder_B(self.style_image, reuse=True)
else :
guide_content_B, guide_style_B = self.Encoder_B(self.content_image, reuse=True)
guide_content_A, guide_style_A = self.Encoder_A(self.style_image, reuse=True)
self.guide_fake_A = self.Decoder_A(content_B=guide_content_B, style_A=guide_style_A, reuse=True)
self.guide_fake_B = self.Decoder_B(content_A=guide_content_A, style_B=guide_style_B, reuse=True)
def train(self):
# initialize all variables
tf.global_variables_initializer().run()
# saver to save model
self.saver = tf.train.Saver()
# summary writer
self.writer = tf.summary.FileWriter(self.log_dir + '/' + self.model_dir, self.sess.graph)
# restore check-point if it exits
could_load, checkpoint_counter = self.load(self.checkpoint_dir)
if could_load:
start_epoch = (int)(checkpoint_counter / self.iteration)
start_batch_id = checkpoint_counter - start_epoch * self.iteration
counter = checkpoint_counter
print(" [*] Load SUCCESS")
else:
start_epoch = 0
start_batch_id = 0
counter = 1
print(" [!] Load failed...")
# loop for epoch
start_time = time.time()
for epoch in range(start_epoch, self.epoch):
lr = self.init_lr * pow(0.5, epoch)
for idx in range(start_batch_id, self.iteration):
style_a = np.random.normal(loc=0.0, scale=1.0, size=[self.batch_size, 1, 1, self.style_dim])
style_b = np.random.normal(loc=0.0, scale=1.0, size=[self.batch_size, 1, 1, self.style_dim])
train_feed_dict = {
self.style_a : style_a,
self.style_b : style_b,
self.lr : lr
}
# Update D
_, d_loss, summary_str = self.sess.run([self.D_optim, self.Discriminator_loss, self.D_loss], feed_dict = train_feed_dict)
self.writer.add_summary(summary_str, counter)
# Update G
batch_A_images, batch_B_images, fake_A, fake_B, _, g_loss, summary_str = self.sess.run([self.real_A, self.real_B, self.fake_A, self.fake_B, self.G_optim, self.Generator_loss, self.G_loss], feed_dict = train_feed_dict)
self.writer.add_summary(summary_str, counter)
# display training status
counter += 1
print("Epoch: [%2d] [%6d/%6d] time: %4.4f d_loss: %.8f, g_loss: %.8f" \
% (epoch, idx, self.iteration, time.time() - start_time, d_loss, g_loss))
if np.mod(idx+1, self.print_freq) == 0 :
save_images(batch_A_images, [self.batch_size, 1],
'./{}/real_A_{:02d}_{:06d}.jpg'.format(self.sample_dir, epoch, idx+1))
# save_images(batch_B_images, [self.batch_size, 1],
# './{}/real_B_{}_{:02d}_{:06d}.jpg'.format(self.sample_dir, gpu_id, epoch, idx+1))
# save_images(fake_A, [self.batch_size, 1],
# './{}/fake_A_{}_{:02d}_{:06d}.jpg'.format(self.sample_dir, gpu_id, epoch, idx+1))
save_images(fake_B, [self.batch_size, 1],
'./{}/fake_B_{:02d}_{:06d}.jpg'.format(self.sample_dir, epoch, idx+1))
if np.mod(idx+1, self.save_freq) == 0 :
self.save(self.checkpoint_dir, counter)
# After an epoch, start_batch_id is set to zero
# non-zero value is only for the first epoch after loading pre-trained model
start_batch_id = 0
# save model for final step
self.save(self.checkpoint_dir, counter)
@property
def model_dir(self):
return "{}_{}_{}".format(self.model_name, self.dataset_name, self.gan_type)
def save(self, checkpoint_dir, step):
checkpoint_dir = os.path.join(checkpoint_dir, self.model_dir)
if not os.path.exists(checkpoint_dir):
os.makedirs(checkpoint_dir)
self.saver.save(self.sess, os.path.join(checkpoint_dir, self.model_name + '.model'), global_step=step)
def load(self, checkpoint_dir):
import re
print(" [*] Reading checkpoints...")
checkpoint_dir = os.path.join(checkpoint_dir, self.model_dir)
ckpt = tf.train.get_checkpoint_state(checkpoint_dir)
if ckpt and ckpt.model_checkpoint_path:
ckpt_name = os.path.basename(ckpt.model_checkpoint_path)
self.saver.restore(self.sess, os.path.join(checkpoint_dir, ckpt_name))
counter = int(next(re.finditer("(\d+)(?!.*\d)", ckpt_name)).group(0))
print(" [*] Success to read {}".format(ckpt_name))
return True, counter
else:
print(" [*] Failed to find a checkpoint")
return False, 0
def test(self):
tf.global_variables_initializer().run()
test_A_files = glob('./dataset/{}/*.*'.format(self.dataset_name + '/testA'))
test_B_files = glob('./dataset/{}/*.*'.format(self.dataset_name + '/testB'))
self.saver = tf.train.Saver()
could_load, checkpoint_counter = self.load(self.checkpoint_dir)
self.result_dir = os.path.join(self.result_dir, self.model_dir)
check_folder(self.result_dir)
if could_load :
print(" [*] Load SUCCESS")
else :
print(" [!] Load failed...")
# write html for visual comparison
index_path = os.path.join(self.result_dir, 'index.html')
index = open(index_path, 'w')
index.write("<html><body><table><tr>")
index.write("<th>name</th><th>input</th><th>output</th></tr>")
for sample_file in test_A_files : # A -> B
print('Processing A image: ' + sample_file)
sample_image = np.asarray(load_test_data(sample_file, size_h=self.img_h, size_w=self.img_w))
file_name = os.path.basename(sample_file).split(".")[0]
file_extension = os.path.basename(sample_file).split(".")[1]
for i in range(self.num_style) :
test_style = np.random.normal(loc=0.0, scale=1.0, size=[1, 1, 1, self.style_dim])
image_path = os.path.join(self.result_dir, '{}_style{}.{}'.format(file_name, i, file_extension))
fake_img = self.sess.run(self.test_fake_B, feed_dict = {self.test_image : sample_image, self.test_style : test_style})
save_images(fake_img, [1, 1], image_path)
index.write("<td>%s</td>" % os.path.basename(image_path))
index.write("<td><img src='%s' width='%d' height='%d'></td>" % (sample_file if os.path.isabs(sample_file) else (
'../..' + os.path.sep + sample_file), self.img_w, self.img_h))
index.write("<td><img src='%s' width='%d' height='%d'></td>" % (image_path if os.path.isabs(image_path) else (
'../..' + os.path.sep + image_path), self.img_w, self.img_h))
index.write("</tr>")
for sample_file in test_B_files : # B -> A
print('Processing B image: ' + sample_file)
sample_image = np.asarray(load_test_data(sample_file, size_h=self.img_h, size_w=self.img_w))
file_name = os.path.basename(sample_file).split(".")[0]
file_extension = os.path.basename(sample_file).split(".")[1]
for i in range(self.num_style):
test_style = np.random.normal(loc=0.0, scale=1.0, size=[1, 1, 1, self.style_dim])
image_path = os.path.join(self.result_dir, '{}_style{}.{}'.format(file_name, i, file_extension))
fake_img = self.sess.run(self.test_fake_A, feed_dict={self.test_image: sample_image, self.test_style: test_style})
save_images(fake_img, [1, 1], image_path)
index.write("<td>%s</td>" % os.path.basename(image_path))
index.write("<td><img src='%s' width='%d' height='%d'></td>" % (sample_file if os.path.isabs(sample_file) else (
'../..' + os.path.sep + sample_file), self.img_w, self.img_h))
index.write("<td><img src='%s' width='%d' height='%d'></td>" % (image_path if os.path.isabs(image_path) else (
'../..' + os.path.sep + image_path), self.img_w, self.img_h))
index.write("</tr>")
index.close()
def style_guide_test(self):
tf.global_variables_initializer().run()
test_A_files = glob('./dataset/{}/*.*'.format(self.dataset_name + '/testA'))
test_B_files = glob('./dataset/{}/*.*'.format(self.dataset_name + '/testB'))
style_file = np.asarray(load_test_data(self.guide_img, size_h=self.img_h, size_w=self.img_w))
self.saver = tf.train.Saver()
could_load, checkpoint_counter = self.load(self.checkpoint_dir)
self.result_dir = os.path.join(self.result_dir, self.model_dir, 'guide')
check_folder(self.result_dir)
if could_load:
print(" [*] Load SUCCESS")
else:
print(" [!] Load failed...")
# write html for visual comparison
index_path = os.path.join(self.result_dir, 'index.html')
index = open(index_path, 'w')
index.write("<html><body><table><tr>")
index.write("<th>name</th><th>input</th><th>output</th></tr>")
if self.direction == 'a2b' :
for sample_file in test_A_files: # A -> B
print('Processing A image: ' + sample_file)
sample_image = np.asarray(load_test_data(sample_file, size_h=self.img_h, size_w=self.img_w))
image_path = os.path.join(self.result_dir, '{}'.format(os.path.basename(sample_file)))
fake_img = self.sess.run(self.guide_fake_B, feed_dict={self.content_image: sample_image, self.style_image : style_file})
save_images(fake_img, [1, 1], image_path)
index.write("<td>%s</td>" % os.path.basename(image_path))
index.write("<td><img src='%s' width='%d' height='%d'></td>" % (sample_file if os.path.isabs(sample_file) else (
'../../..' + os.path.sep + sample_file), self.img_w, self.img_h))
index.write("<td><img src='%s' width='%d' height='%d'></td>" % (image_path if os.path.isabs(image_path) else (
'../../..' + os.path.sep + image_path), self.img_w, self.img_h))
index.write("</tr>")
else :
for sample_file in test_B_files: # B -> A
print('Processing B image: ' + sample_file)
sample_image = np.asarray(load_test_data(sample_file, size_h=self.img_h, size_w=self.img_w))
image_path = os.path.join(self.result_dir, '{}'.format(os.path.basename(sample_file)))
fake_img = self.sess.run(self.guide_fake_A, feed_dict={self.content_image: sample_image, self.style_image : style_file})
save_images(fake_img, [1, 1], image_path)
index.write("<td>%s</td>" % os.path.basename(image_path))
index.write("<td><img src='%s' width='%d' height='%d'></td>" % (sample_file if os.path.isabs(sample_file) else (
'../../..' + os.path.sep + sample_file), self.img_w, self.img_h))
index.write("<td><img src='%s' width='%d' height='%d'></td>" % (image_path if os.path.isabs(image_path) else (
'../../..' + os.path.sep + image_path), self.img_w, self.img_h))
index.write("</tr>")
index.close()