-
Notifications
You must be signed in to change notification settings - Fork 2
/
tf_util.py
575 lines (490 loc) · 18.7 KB
/
tf_util.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
""" Wrapper functions for TensorFlow layers.
Author: Charles R. Qi
Date: November 2016
"""
import numpy as np
import tensorflow as tf
def _variable_on_cpu(name, shape, initializer, use_fp16=False):
"""Helper to create a Variable stored on CPU memory.
Args:
name: name of the variable
shape: list of ints
initializer: initializer for Variable
Returns:
Variable Tensor
"""
with tf.device('/cpu:0'):
dtype = tf.float16 if use_fp16 else tf.float32
var = tf.get_variable(name, shape, initializer=initializer, dtype=dtype)
return var
def _variable_with_weight_decay(name, shape, stddev, wd, use_xavier=True):
"""Helper to create an initialized Variable with weight decay.
Note that the Variable is initialized with a truncated normal distribution.
A weight decay is added only if one is specified.
Args:
name: name of the variable
shape: list of ints
stddev: standard deviation of a truncated Gaussian
wd: add L2Loss weight decay multiplied by this float. If None, weight
decay is not added for this Variable.
use_xavier: bool, whether to use xavier initializer
Returns:
Variable Tensor
"""
if use_xavier:
initializer = tf.contrib.layers.xavier_initializer()
else:
initializer = tf.truncated_normal_initializer(stddev=stddev)
var = _variable_on_cpu(name, shape, initializer)
if wd is not None:
weight_decay = tf.multiply(tf.nn.l2_loss(var), wd, name='weight_loss')
tf.add_to_collection('losses', weight_decay)
return var
def conv1d(inputs,
num_output_channels,
kernel_size,
scope,
stride=1,
padding='SAME',
use_xavier=True,
stddev=1e-3,
weight_decay=0.0,
activation_fn=tf.nn.relu,
bn=False,
bn_decay=None,
is_training=None):
""" 1D convolution with non-linear operation.
Args:
inputs: 3-D tensor variable BxLxC
num_output_channels: int
kernel_size: int
scope: string
stride: int
padding: 'SAME' or 'VALID'
use_xavier: bool, use xavier_initializer if true
stddev: float, stddev for truncated_normal init
weight_decay: float
activation_fn: function
bn: bool, whether to use batch norm
bn_decay: float or float tensor variable in [0,1]
is_training: bool Tensor variable
Returns:
Variable tensor
"""
with tf.variable_scope(scope) as sc:
num_in_channels = inputs.get_shape()[-1].value
kernel_shape = [kernel_size,
num_in_channels, num_output_channels]
kernel = _variable_with_weight_decay('weights',
shape=kernel_shape,
use_xavier=use_xavier,
stddev=stddev,
wd=weight_decay)
outputs = tf.nn.conv1d(inputs, kernel,
stride=stride,
padding=padding)
biases = _variable_on_cpu('biases', [num_output_channels],
tf.constant_initializer(0.0))
outputs = tf.nn.bias_add(outputs, biases)
if bn:
outputs = batch_norm_for_conv1d(outputs, is_training,
bn_decay=bn_decay, scope='bn')
if activation_fn is not None:
outputs = activation_fn(outputs)
return outputs
def conv2d(inputs,
num_output_channels,
kernel_size,
scope,
stride=[1, 1],
padding='SAME',
use_xavier=True,
stddev=1e-3,
weight_decay=0.0,
activation_fn=tf.nn.relu,
bn=False,
bn_decay=None,
is_training=None):
""" 2D convolution with non-linear operation.
Args:
inputs: 4-D tensor variable BxHxWxC
num_output_channels: int
kernel_size: a list of 2 ints
scope: string
stride: a list of 2 ints
padding: 'SAME' or 'VALID'
use_xavier: bool, use xavier_initializer if true
stddev: float, stddev for truncated_normal init
weight_decay: float
activation_fn: function
bn: bool, whether to use batch norm
bn_decay: float or float tensor variable in [0,1]
is_training: bool Tensor variable
Returns:
Variable tensor
"""
with tf.variable_scope(scope) as sc:
kernel_h, kernel_w = kernel_size
num_in_channels = inputs.get_shape()[-1].value
kernel_shape = [kernel_h, kernel_w,
num_in_channels, num_output_channels]
kernel = _variable_with_weight_decay('weights',
shape=kernel_shape,
use_xavier=use_xavier,
stddev=stddev,
wd=weight_decay)
stride_h, stride_w = stride
outputs = tf.nn.conv2d(inputs, kernel,
[1, stride_h, stride_w, 1],
padding=padding)
biases = _variable_on_cpu('biases', [num_output_channels],
tf.constant_initializer(0.0))
outputs = tf.nn.bias_add(outputs, biases)
if bn:
outputs = batch_norm_for_conv2d(outputs, is_training,
bn_decay=bn_decay, scope='bn')
if activation_fn is not None:
outputs = activation_fn(outputs)
return outputs
def conv2d_transpose(inputs,
num_output_channels,
kernel_size,
scope,
stride=[1, 1],
padding='SAME',
use_xavier=True,
stddev=1e-3,
weight_decay=0.0,
activation_fn=tf.nn.relu,
bn=False,
bn_decay=None,
is_training=None):
""" 2D convolution transpose with non-linear operation.
Args:
inputs: 4-D tensor variable BxHxWxC
num_output_channels: int
kernel_size: a list of 2 ints
scope: string
stride: a list of 2 ints
padding: 'SAME' or 'VALID'
use_xavier: bool, use xavier_initializer if true
stddev: float, stddev for truncated_normal init
weight_decay: float
activation_fn: function
bn: bool, whether to use batch norm
bn_decay: float or float tensor variable in [0,1]
is_training: bool Tensor variable
Returns:
Variable tensor
Note: conv2d(conv2d_transpose(a, num_out, ksize, stride), a.shape[-1], ksize, stride) == a
"""
with tf.variable_scope(scope) as sc:
kernel_h, kernel_w = kernel_size
num_in_channels = inputs.get_shape()[-1].value
kernel_shape = [kernel_h, kernel_w,
num_output_channels, num_in_channels] # reversed to conv2d
kernel = _variable_with_weight_decay('weights',
shape=kernel_shape,
use_xavier=use_xavier,
stddev=stddev,
wd=weight_decay)
stride_h, stride_w = stride
# from slim.convolution2d_transpose
def get_deconv_dim(dim_size, stride_size, kernel_size, padding):
dim_size *= stride_size
if padding == 'VALID' and dim_size is not None:
dim_size += max(kernel_size - stride_size, 0)
return dim_size
# caculate output shape
batch_size = inputs.get_shape()[0].value
height = inputs.get_shape()[1].value
width = inputs.get_shape()[2].value
out_height = get_deconv_dim(height, stride_h, kernel_h, padding)
out_width = get_deconv_dim(width, stride_w, kernel_w, padding)
output_shape = [batch_size, out_height, out_width, num_output_channels]
outputs = tf.nn.conv2d_transpose(inputs, kernel, output_shape,
[1, stride_h, stride_w, 1],
padding=padding)
biases = _variable_on_cpu('biases', [num_output_channels],
tf.constant_initializer(0.0))
outputs = tf.nn.bias_add(outputs, biases)
if bn:
outputs = batch_norm_for_conv2d(outputs, is_training,
bn_decay=bn_decay, scope='bn')
if activation_fn is not None:
outputs = activation_fn(outputs)
return outputs
def conv3d(inputs,
num_output_channels,
kernel_size,
scope,
stride=[1, 1, 1],
padding='SAME',
use_xavier=True,
stddev=1e-3,
weight_decay=0.0,
activation_fn=tf.nn.relu,
bn=False,
bn_decay=None,
is_training=None):
""" 3D convolution with non-linear operation.
Args:
inputs: 5-D tensor variable BxDxHxWxC
num_output_channels: int
kernel_size: a list of 3 ints
scope: string
stride: a list of 3 ints
padding: 'SAME' or 'VALID'
use_xavier: bool, use xavier_initializer if true
stddev: float, stddev for truncated_normal init
weight_decay: float
activation_fn: function
bn: bool, whether to use batch norm
bn_decay: float or float tensor variable in [0,1]
is_training: bool Tensor variable
Returns:
Variable tensor
"""
with tf.variable_scope(scope) as sc:
kernel_d, kernel_h, kernel_w = kernel_size
num_in_channels = inputs.get_shape()[-1].value
kernel_shape = [kernel_d, kernel_h, kernel_w,
num_in_channels, num_output_channels]
kernel = _variable_with_weight_decay('weights',
shape=kernel_shape,
use_xavier=use_xavier,
stddev=stddev,
wd=weight_decay)
stride_d, stride_h, stride_w = stride
outputs = tf.nn.conv3d(inputs, kernel,
[1, stride_d, stride_h, stride_w, 1],
padding=padding)
biases = _variable_on_cpu('biases', [num_output_channels],
tf.constant_initializer(0.0))
outputs = tf.nn.bias_add(outputs, biases)
if bn:
outputs = batch_norm_for_conv3d(outputs, is_training,
bn_decay=bn_decay, scope='bn')
if activation_fn is not None:
outputs = activation_fn(outputs)
return outputs
def fully_connected(inputs,
num_outputs,
scope,
use_xavier=True,
stddev=1e-3,
weight_decay=0.0,
activation_fn=tf.nn.relu,
bn=False,
bn_decay=None,
is_training=None):
""" Fully connected layer with non-linear operation.
Args:
inputs: 2-D tensor BxN
num_outputs: int
Returns:
Variable tensor of size B x num_outputs.
"""
with tf.variable_scope(scope) as sc:
num_input_units = inputs.get_shape()[-1].value
weights = _variable_with_weight_decay('weights',
shape=[num_input_units, num_outputs],
use_xavier=use_xavier,
stddev=stddev,
wd=weight_decay)
outputs = tf.matmul(inputs, weights)
biases = _variable_on_cpu('biases', [num_outputs],
tf.constant_initializer(0.0))
outputs = tf.nn.bias_add(outputs, biases)
if bn:
outputs = batch_norm_for_fc(outputs, is_training, bn_decay, 'bn')
if activation_fn is not None:
outputs = activation_fn(outputs)
return outputs
def max_pool2d(inputs,
kernel_size,
scope,
stride=[2, 2],
padding='VALID'):
""" 2D max pooling.
Args:
inputs: 4-D tensor BxHxWxC
kernel_size: a list of 2 ints
stride: a list of 2 ints
Returns:
Variable tensor
"""
with tf.variable_scope(scope) as sc:
kernel_h, kernel_w = kernel_size
stride_h, stride_w = stride
outputs = tf.nn.max_pool(inputs,
ksize=[1, kernel_h, kernel_w, 1],
strides=[1, stride_h, stride_w, 1],
padding=padding,
name=sc.name)
return outputs
def avg_pool2d(inputs,
kernel_size,
scope,
stride=[2, 2],
padding='VALID'):
""" 2D avg pooling.
Args:
inputs: 4-D tensor BxHxWxC
kernel_size: a list of 2 ints
stride: a list of 2 ints
Returns:
Variable tensor
"""
with tf.variable_scope(scope) as sc:
kernel_h, kernel_w = kernel_size
stride_h, stride_w = stride
outputs = tf.nn.avg_pool(inputs,
ksize=[1, kernel_h, kernel_w, 1],
strides=[1, stride_h, stride_w, 1],
padding=padding,
name=sc.name)
return outputs
def max_pool3d(inputs,
kernel_size,
scope,
stride=[2, 2, 2],
padding='VALID'):
""" 3D max pooling.
Args:
inputs: 5-D tensor BxDxHxWxC
kernel_size: a list of 3 ints
stride: a list of 3 ints
Returns:
Variable tensor
"""
with tf.variable_scope(scope) as sc:
kernel_d, kernel_h, kernel_w = kernel_size
stride_d, stride_h, stride_w = stride
outputs = tf.nn.max_pool3d(inputs,
ksize=[1, kernel_d, kernel_h, kernel_w, 1],
strides=[1, stride_d, stride_h, stride_w, 1],
padding=padding,
name=sc.name)
return outputs
def avg_pool3d(inputs,
kernel_size,
scope,
stride=[2, 2, 2],
padding='VALID'):
""" 3D avg pooling.
Args:
inputs: 5-D tensor BxDxHxWxC
kernel_size: a list of 3 ints
stride: a list of 3 ints
Returns:
Variable tensor
"""
with tf.variable_scope(scope) as sc:
kernel_d, kernel_h, kernel_w = kernel_size
stride_d, stride_h, stride_w = stride
outputs = tf.nn.avg_pool3d(inputs,
ksize=[1, kernel_d, kernel_h, kernel_w, 1],
strides=[1, stride_d, stride_h, stride_w, 1],
padding=padding,
name=sc.name)
return outputs
def batch_norm_template(inputs, is_training, scope, moments_dims, bn_decay):
""" Batch normalization on convolutional maps and beyond...
Ref.: http://stackoverflow.com/questions/33949786/how-could-i-use-batch-normalization-in-tensorflow
Args:
inputs: Tensor, k-D input ... x C could be BC or BHWC or BDHWC
is_training: boolean tf.Varialbe, true indicates training phase
scope: string, variable scope
moments_dims: a list of ints, indicating dimensions for moments calculation
bn_decay: float or float tensor variable, controling moving average weight
Return:
normed: batch-normalized maps
"""
with tf.variable_scope(scope) as sc:
num_channels = inputs.get_shape()[-1].value
beta = tf.Variable(tf.constant(0.0, shape=[num_channels]),
name='beta', trainable=True)
gamma = tf.Variable(tf.constant(1.0, shape=[num_channels]),
name='gamma', trainable=True)
batch_mean, batch_var = tf.nn.moments(inputs, moments_dims, name='moments')
decay = bn_decay if bn_decay is not None else 0.9
ema = tf.train.ExponentialMovingAverage(decay=decay)
# Operator that maintains moving averages of variables.
ema_apply_op = tf.cond(is_training,
lambda: ema.apply([batch_mean, batch_var]),
lambda: tf.no_op())
# Update moving average and return current batch's avg and var.
def mean_var_with_update():
with tf.control_dependencies([ema_apply_op]):
return tf.identity(batch_mean), tf.identity(batch_var)
# ema.average returns the Variable holding the average of var.
mean, var = tf.cond(is_training,
mean_var_with_update,
lambda: (ema.average(batch_mean), ema.average(batch_var)))
normed = tf.nn.batch_normalization(inputs, mean, var, beta, gamma, 1e-3)
return normed
def batch_norm_for_fc(inputs, is_training, bn_decay, scope):
""" Batch normalization on FC data.
Args:
inputs: Tensor, 2D BxC input
is_training: boolean tf.Varialbe, true indicates training phase
bn_decay: float or float tensor variable, controling moving average weight
scope: string, variable scope
Return:
normed: batch-normalized maps
"""
return batch_norm_template(inputs, is_training, scope, [0,], bn_decay)
def batch_norm_for_conv1d(inputs, is_training, bn_decay, scope):
""" Batch normalization on 1D convolutional maps.
Args:
inputs: Tensor, 3D BLC input maps
is_training: boolean tf.Varialbe, true indicates training phase
bn_decay: float or float tensor variable, controling moving average weight
scope: string, variable scope
Return:
normed: batch-normalized maps
"""
return batch_norm_template(inputs, is_training, scope, [0,1], bn_decay)
def batch_norm_for_conv2d(inputs, is_training, bn_decay, scope):
""" Batch normalization on 2D convolutional maps.
Args:
inputs: Tensor, 4D BHWC input maps
is_training: boolean tf.Varialbe, true indicates training phase
bn_decay: float or float tensor variable, controling moving average weight
scope: string, variable scope
Return:
normed: batch-normalized maps
"""
return batch_norm_template(inputs, is_training, scope, [0,1,2], bn_decay)
def batch_norm_for_conv3d(inputs, is_training, bn_decay, scope):
""" Batch normalization on 3D convolutional maps.
Args:
inputs: Tensor, 5D BDHWC input maps
is_training: boolean tf.Varialbe, true indicates training phase
bn_decay: float or float tensor variable, controling moving average weight
scope: string, variable scope
Return:
normed: batch-normalized maps
"""
return batch_norm_template(inputs, is_training, scope, [0,1,2,3], bn_decay)
def dropout(inputs,
is_training,
scope,
keep_prob=0.5,
noise_shape=None):
""" Dropout layer.
Args:
inputs: tensor
is_training: boolean tf.Variable
scope: string
keep_prob: float in [0,1]
noise_shape: list of ints
Returns:
tensor variable
"""
with tf.variable_scope(scope) as sc:
outputs = tf.cond(is_training,
lambda: tf.nn.dropout(inputs, keep_prob, noise_shape),
lambda: inputs)
return outputs