-
Notifications
You must be signed in to change notification settings - Fork 66
/
Copy pathz_encoding_izif.py
368 lines (291 loc) · 15 KB
/
z_encoding_izif.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
"""
Encoder (izi_f) training
Copyright (c) 2018 Thomas Schlegl
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
"""
import os, sys
sys.path.append(os.getcwd())
import pdb
import re
import time
from tqdm import tqdm
from wgangp_64x64 import GoodGenerator, GoodDiscriminator, ResidualBlock
import numpy as np
import tensorflow as tf
import tflib as lib
import tflib.ops.linear
import tflib.ops.conv2d
import tflib.ops.batchnorm
import tflib.ops.deconv2d
import tflib.save_images
import tflib.img_loader
import tflib.ops.layernorm
import tflib.plot
class bcolors:
PINK = '\033[95m'
BLUE = '\033[94m'
GREEN = '\033[92m'
YELLOW = '\033[93m'
RED = '\033[91m'
ENDC = '\033[0m'
MODE = 'wgan-gp'
ZDIM = 128
DIM = 64 # Model dimensionality
BATCH_SIZE = 64
OUTPUT_DIM = 64*64*1 # Number of pixels in each image
ENCODING_SAVE_ITERS = [50, 100, 200, 500, 1000, 2000, 5000, 10000, 20000, 30000, 40000, 50000]
ITERS_ENC = 50000 #100000
N_GPUS = 1
Z_REG_TYPE = 'tanh_fc' # DEFAULT (paper version)
#Z_REG_TYPE = None # unconstrained ('linear') encoder training (see Appendix A.2.)
#
##Z_REG_TYPE = 'stoch_clip'
##Z_REG_TYPE = '3s_tanh_fc'
##Z_REG_TYPE = '05s_tanh_fc'
##Z_REG_TYPE = 'hard_clip'
##Z_REG_TYPE = '3s_hard_clip'
##Z_REG_TYPE = '05s_hard_clip'
print bcolors.GREEN + "\n=== ENCODER TRAINING PARAMETERS ===" + bcolors.ENDC
lib.print_model_settings(locals().copy())
DEVICES = ['/gpu:{}'.format(i) for i in xrange(N_GPUS)]
timestamp = time.strftime("%Y-%m-%d-%H%M")
z_reg_type_txt = ''
if Z_REG_TYPE is not None:
z_reg_type_txt = '_%s' %Z_REG_TYPE
## -- my loss functions
def l2_norm(x, y, axis=None):
if axis is None:
return tf.reduce_sum(tf.pow(x-y, 2))
else:
return tf.reduce_sum(tf.pow(x-y, 2), axis=axis)
def MSE(x, y, axis=None):
if axis is None:
return tf.reduce_mean(tf.pow(x-y, 2))
else:
return tf.reduce_mean(tf.pow(x-y, 2), axis=axis)
## --
def save(session, saver, checkpoint_dir, step):
print(" [*] Saving checkpoint (step %d) ..." %step)
if not os.path.exists(checkpoint_dir):
os.makedirs(checkpoint_dir)
saver.save( session,
os.path.join(checkpoint_dir, "model"),
global_step=step)
def load(session, saver, checkpoint_dir, checkpoint_iter=None):
print(" [*] Reading checkpoints...")
ckpt = tf.train.get_checkpoint_state(checkpoint_dir)
if ckpt and ckpt.model_checkpoint_path:
if checkpoint_iter is not None:
last_ckpt_iter = re.match(r'.*.model-(\d+)', ckpt.model_checkpoint_path).group(1)
target_ckpt_path = re.sub( 'model-%s'%last_ckpt_iter, 'model-%d'%checkpoint_iter, ckpt.model_checkpoint_path)
saver.restore(session, target_ckpt_path)
idxx = target_ckpt_path.rfind('/')
ckpt_name = target_ckpt_path[idxx+1:]
else:
saver.restore(session, ckpt.model_checkpoint_path)
idxx = ckpt.model_checkpoint_path.rfind('/')
ckpt_name = ckpt.model_checkpoint_path[idxx+1:]
return True, ckpt_name
else:
return False, ''
def get_ckpt_name(checkpoint_dir, checkpoint_iter=None):
print(" [*] Reading checkpoints...")
ckpt = tf.train.get_checkpoint_state(checkpoint_dir)
if ckpt and ckpt.model_checkpoint_path:
if checkpoint_iter is not None:
last_ckpt_iter = re.match(r'.*.model-(\d+)', ckpt.model_checkpoint_path).group(1)
target_ckpt_path = re.sub( last_ckpt_iter, str(checkpoint_iter), ckpt.model_checkpoint_path)
idxx = target_ckpt_path.rfind('/')
ckpt_name = target_ckpt_path[idxx+1:]
else:
idxx = ckpt.model_checkpoint_path.rfind('/')
ckpt_name = ckpt.model_checkpoint_path[idxx+1:]
return ckpt_name
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
def my_Normalize(name, inputs, is_training):
if ('Discriminator' in name) and (MODE == 'wgan-gp'):
return lib.ops.layernorm.Layernorm(name,[1,2,3],inputs)
else:
#return lib.ops.batchnorm.Batchnorm(name,axes,inputs,fused=True)
return tf.layers.batch_normalization(inputs, axis=1, training=is_training, name=name)
def Encoder(inputs, is_training, dim=DIM, z_dim=ZDIM, rand_sampling='normal', reuse=None, z_reg_type=None, denoise=None):
with tf.variable_scope('Encoder', reuse=reuse):
if denoise is not None:
inputs = tf.nn.dropout(inputs, keep_prob=denoise)
output = tf.reshape(inputs, [-1, 1, 64, 64])
output = lib.ops.conv2d.Conv2D('Encoder.Input', 1, dim, 3, output, he_init=False)
output = ResidualBlock('Encoder.Res1', dim, 2*dim, 3, output, is_training=is_training, resample='down')
output = ResidualBlock('Encoder.Res2', 2*dim, 4*dim, 3, output, is_training=is_training, resample='down')
output = ResidualBlock('Encoder.Res3', 4*dim, 8*dim, 3, output, is_training=is_training, resample='down')
output = ResidualBlock('Encoder.Res4', 8*dim, 8*dim, 3, output, is_training=is_training, resample='down')
output = tf.reshape(output, [-1, 4*4*8*dim])
output = lib.ops.linear.Linear('Encoder.Output', 4*4*8*dim, z_dim, output)
if z_reg_type is None:
return output
elif z_reg_type == 'tanh_fc':
return tf.nn.tanh( output )
elif z_reg_type == '3s_tanh_fc':
return tf.nn.tanh( output ) * 3
elif z_reg_type == '05s_tanh_fc':
return tf.nn.tanh( output ) * 0.5
elif z_reg_type == 'hard_clip':
return tf.clip_by_value( output, -1., 1. )
elif z_reg_type == '3s_hard_clip':
return tf.clip_by_value( output, -3., 3. )
elif z_reg_type == '05s_hard_clip':
return tf.clip_by_value( output, -0.5, 0.5 )
elif z_reg_type == 'stoch_clip': ## IMPLEMENTS STOCHASTIC CLIPPING -->> https://arxiv.org/pdf/1702.04782.pdf
if rand_sampling == 'unif':
condition = tf.greater(tf.abs(output), 1.)
true_case = tf.random_uniform(output.get_shape(), minval=-1., maxval=1.)
elif rand_sampling == 'normal':
condition = tf.greater(tf.abs(output), 3.)
true_case = tf.random_normal(output.get_shape())
print bcolors.YELLOW + "\nImplementing STOCH-CLIP with NORMAL z-mapping!\n" + bcolors.ENDC
return tf.where(condition, true_case, output)
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
def train_enc_izi(checkpoint_dir, checkpoint_iter, nr_valid_plt=9, nr_valid_btchs=10, lr=5e-5, loss_type='MSE', kappa=1.0, denoise=None):
np.random.seed(1234)
if '_unif' in checkpoint_dir:
rand_sampling = 'unif'
elif '_norm' in checkpoint_dir:
rand_sampling = 'normal'
loss_type_txt = loss_type
if loss_type == 'MSE':
loss_type_txt = '%s-k%.2f' %(loss_type, kappa)
if lr!=5e-5:
loss_type_txt = '%s-lr%.0E' %(loss_type_txt, lr)
if denoise is not None:
dn_txt = "_dn%.1f"%denoise
else:
dn_txt = ""
scriptname = os.path.basename(__file__).replace('.pyc', '').replace('.py', '')
checkpoint_dir_map = os.path.join('z_encoding_d/izi%s%s'%(z_reg_type_txt,dn_txt), checkpoint_dir.replace('wganTrain','wganTrain-iter%d'%checkpoint_iter).replace('/checkpoints', '-enc_iter%d_%s'%(ITERS_ENC,timestamp)), "%s-%s%s"%(scriptname,loss_type_txt,z_reg_type_txt), 'checkpoints')
if not os.path.isdir( checkpoint_dir_map ):
os.makedirs( checkpoint_dir_map )
samples_dir_map = checkpoint_dir_map.replace('checkpoints','samples')
if not os.path.isdir( samples_dir_map ):
os.makedirs( samples_dir_map )
log_dir = checkpoint_dir_map.replace('checkpoints','log_dir')
if not os.path.isdir( log_dir ):
os.makedirs( log_dir )
# Dataset iterator
train_gen,valid_gen = lib.img_loader.load(BATCH_SIZE, 'encoder_train')
def inf_train_gen():
while True:
for (images,) in train_gen():
yield images
(fixed_valid_img,) = valid_gen().next()
Generator = GoodGenerator
Discriminator = GoodDiscriminator
with tf.Session(config=tf.ConfigProto(device_count={'GPU':len(DEVICES)}, allow_soft_placement=True)) as session:
all_real_data_conv = tf.placeholder(tf.int32, shape=[BATCH_SIZE, 1, 64, 64])
if tf.__version__.startswith('1.'):
split_real_data_conv = tf.split(all_real_data_conv, len(DEVICES))
print "\n\nDEVICES: %s\n\n" %DEVICES
else:
split_real_data_conv = tf.split(0, len(DEVICES), all_real_data_conv)
mapping_losses = []
mapping_losses_img = []
mapping_losses_fts = []
for device_index, (device, real_data_conv) in enumerate(zip(DEVICES, split_real_data_conv)):
with tf.device(device):
real_data = tf.reshape(2*((tf.cast(real_data_conv, tf.float32)/255.)-.5), [BATCH_SIZE/len(DEVICES), OUTPUT_DIM])
z = Encoder(real_data, is_training=True, z_reg_type=Z_REG_TYPE, rand_sampling=rand_sampling, denoise=denoise)
recon_img = Generator(BATCH_SIZE/len(DEVICES), noise=z, rand_sampling='normal', is_training=False)
_,recon_features = Discriminator(recon_img, is_training=False)
_,image_features = Discriminator(real_data, is_training=False, reuse=True)
if loss_type=='l2Mean':
loss = tf.reduce_mean(l2_norm( real_data, recon_img, axis=1 ))
elif loss_type=='l2Sum':
raise
loss = l2_norm( real_data, recon_img )
elif loss_type =='MSE':
loss_img = MSE( real_data, recon_img )
loss_fts = MSE( recon_features, image_features )
loss = loss_img + kappa*loss_fts
mapping_losses.append(loss)
mapping_losses_img.append(loss_img)
mapping_losses_fts.append(loss_fts)
mapping_loss = tf.add_n(mapping_losses) / len(DEVICES)
mapping_loss_img = tf.add_n(mapping_losses_img) / len(DEVICES)
mapping_loss_fts = tf.add_n(mapping_losses_fts) / len(DEVICES)
## ****************************************************
## *** TENSORBOARD
## ****************************************************
mapping_loss_sum = tf.summary.scalar("mapping_loss", mapping_loss)
mapping_loss_img_sum = tf.summary.scalar("mapping_loss_img", mapping_loss_img)
mapping_loss_fts_sum = tf.summary.scalar("mapping_loss_fts", mapping_loss_fts)
merged = tf.summary.merge_all()
train_writer = tf.summary.FileWriter(log_dir + '/train', session.graph)
valid_writer = tf.summary.FileWriter(log_dir + '/valid')
## ****************************************************
t_vars = tf.trainable_variables()
encoder_vars = [var for var in t_vars if 'Encoder' in var.name]
update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)
with tf.control_dependencies(update_ops):
train_op = tf.train.RMSPropOptimizer(learning_rate=lr).minimize(mapping_loss,
var_list=encoder_vars, colocate_gradients_with_ops=True)
saver = tf.train.Saver(max_to_keep=15)
session.run(tf.global_variables_initializer())
ckpt = get_ckpt_name(checkpoint_dir, checkpoint_iter)
ckpt_pth = os.path.join(checkpoint_dir, ckpt)
varmap = {}
for item in tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES):
if ('Generator' in item.name) or ('Discriminator' in item.name):
varmap[item.name.replace(':0', '')] = item
saver_restore = tf.train.Saver(varmap)
saver_restore.restore(session, ckpt_pth)
gen = inf_train_gen()
for iteration in tqdm(xrange(ITERS_ENC)):
start_time = time.time()
_data = gen.next()
## -- TRAIN encoder --
_mapping_loss, _, _ml_sum = session.run([mapping_loss, train_op, merged ], feed_dict={all_real_data_conv: _data})
## -- train LOGGING **
lib.plot.plot('z-mapping loss train', _mapping_loss)
lib.plot.plot('time', time.time() - start_time)
if (iteration+1) % 50 == 0:
train_writer.add_summary(_ml_sum, iteration+1)
## -- VALIDATION --
if ((iteration+1) == 100) or ( (iteration+1) % 500 == 0) or ((iteration+1) in ENCODING_SAVE_ITERS):
valid_map_losses = []
for n,_valid_data in enumerate(valid_gen()):
(images,) = _valid_data
_mapping_loss_valid, _ml_valid_sum = session.run([mapping_loss, merged ], feed_dict={all_real_data_conv: images})
valid_map_losses.append(_mapping_loss_valid)
if (n+1)>=nr_valid_btchs:
break
## -- validation LOGGING **
lib.plot.plot('z-mapping loss valid', np.mean(valid_map_losses))
valid_writer.add_summary(_ml_valid_sum, iteration+1)
_real_img, _recon_img = session.run([ real_data, recon_img ], feed_dict={all_real_data_conv: fixed_valid_img})
_real_img = _real_img.reshape((-1, 1, 64, 64))
inputs_plt = ((_real_img[:nr_valid_plt]+1.)*(255.99/2)).astype('int32')
_recon_img = _recon_img.reshape((-1, 1, 64, 64))
recons_plt = ((_recon_img[:nr_valid_plt]+1.)*(255.99/2)).astype('int32')
img_pairs = np.concatenate( (inputs_plt,recons_plt), axis=2 )
lib.save_images.save_images_as_row(img_pairs, '{}/samples_{}.png'.format(samples_dir_map, iteration+1))
if (iteration < 3) or ( (iteration+1) % 500 == 0) or ((iteration+1) in ENCODING_SAVE_ITERS):
lib.plot.flush(log_dir)
lib.plot.tick()
if (iteration+1) in ENCODING_SAVE_ITERS:
save(session, saver, checkpoint_dir_map, iteration+1)
# SAVE FINAL MODEL
save(session, saver, checkpoint_dir_map, iteration+1)
if __name__ == '__main__':
train_enc_izi("path-to-'checkpoints'-folder-of-WGAN-training", <checkpoint-iter>, loss_type='MSE', kappa=1.0)