-
Notifications
You must be signed in to change notification settings - Fork 5
/
layer_utils.py
408 lines (366 loc) · 16.1 KB
/
layer_utils.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
import tensorflow as tf
import keras.backend as K
from keras.models import Model
from keras.engine import InputSpec
from keras.engine.topology import Layer
from keras.layers import Input, Conv2D, Activation, BatchNormalization
from keras.layers.merge import Add, Concatenate
from keras.utils import conv_utils
from keras.layers.core import Dropout
# def dense_block(input, nb_layers, filters, kernel_size=(3, 3), strides=(1, 1), use_dropout=False):
# concat_axis = 1 if K.image_dim_ordering() == "th" else -1
# feature_list = [input]
# for i in range(nb_layers):
# x = ReflectionPadding2D((1, 1))(input)
# x = Conv2D(filters=filters, kernel_size=kernel_size, strides=strides,)(x)
# x = BatchNormalization()(x)
# x = Activation('relu')(x)
# feature_list.append(x)
# x = Concatenate(axis=concat_axis)(feature_list)
# return x
def res_block(input, filters, kernel_size=(3, 3), strides=(1, 1), use_dropout=False):
"""
Instanciate a Keras Resnet Block using sequential API.
:param input: Input tensor
:param filters: Number of filters to use
:param kernel_size: Shape of the kernel for the convolution
:param strides: Shape of the strides for the convolution
:param use_dropout: Boolean value to determine the use of dropout
:return: Keras Model
"""
x = ReflectionPadding2D((1, 1))(input)
x = Conv2D(filters=filters,
kernel_size=kernel_size,
strides=strides,)(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
if use_dropout:
x = Dropout(0.5)(x)
x = ReflectionPadding2D((1, 1))(x)
x = Conv2D(filters=filters,
kernel_size=kernel_size,
strides=strides,)(x)
x = BatchNormalization()(x)
merged = Add()([input, x])
return merged
def spatial_reflection_2d_padding(x, padding=((1, 1), (1, 1)), data_format=None):
"""
Pad the 2nd and 3rd dimensions of a 4D tensor.
:param x: Input tensor
:param padding: Shape of padding to use
:param data_format: Tensorflow vs Theano convention ('channels_last', 'channels_first')
:return: Tensorflow tensor
"""
assert len(padding) == 2
assert len(padding[0]) == 2
assert len(padding[1]) == 2
if data_format is None:
data_format = image_data_format()
if data_format not in {'channels_first', 'channels_last'}:
raise ValueError('Unknown data_format ' + str(data_format))
if data_format == 'channels_first':
pattern = [[0, 0],
[0, 0],
list(padding[0]),
list(padding[1])]
else:
pattern = [[0, 0],
list(padding[0]), list(padding[1]),
[0, 0]]
return tf.pad(x, pattern, "REFLECT")
# TODO: Credits
class ReflectionPadding2D(Layer):
"""Reflection-padding layer for 2D input (e.g. picture).
This layer can add rows and columns or zeros
at the top, bottom, left and right side of an image tensor.
# Arguments
padding: int, or tuple of 2 ints, or tuple of 2 tuples of 2 ints.
- If int: the same symmetric padding
is applied to width and height.
- If tuple of 2 ints:
interpreted as two different
symmetric padding values for height and width:
`(symmetric_height_pad, symmetric_width_pad)`.
- If tuple of 2 tuples of 2 ints:
interpreted as
`((top_pad, bottom_pad), (left_pad, right_pad))`
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be "channels_last".
# Input shape
4D tensor with shape:
- If `data_format` is `"channels_last"`:
`(batch, rows, cols, channels)`
- If `data_format` is `"channels_first"`:
`(batch, channels, rows, cols)`
# Output shape
4D tensor with shape:
- If `data_format` is `"channels_last"`:
`(batch, padded_rows, padded_cols, channels)`
- If `data_format` is `"channels_first"`:
`(batch, channels, padded_rows, padded_cols)`
"""
def __init__(self,
padding=(1, 1),
data_format=None,
**kwargs):
super(ReflectionPadding2D, self).__init__(**kwargs)
self.data_format = conv_utils.normalize_data_format(data_format)
if isinstance(padding, int):
self.padding = ((padding, padding), (padding, padding))
elif hasattr(padding, '__len__'):
if len(padding) != 2:
raise ValueError('`padding` should have two elements. '
'Found: ' + str(padding))
height_padding = conv_utils.normalize_tuple(padding[0], 2,
'1st entry of padding')
width_padding = conv_utils.normalize_tuple(padding[1], 2,
'2nd entry of padding')
self.padding = (height_padding, width_padding)
else:
raise ValueError('`padding` should be either an int, '
'a tuple of 2 ints '
'(symmetric_height_pad, symmetric_width_pad), '
'or a tuple of 2 tuples of 2 ints '
'((top_pad, bottom_pad), (left_pad, right_pad)). '
'Found: ' + str(padding))
self.input_spec = InputSpec(ndim=4)
def compute_output_shape(self, input_shape):
if self.data_format == 'channels_first':
if input_shape[2] is not None:
rows = input_shape[2] + self.padding[0][0] + self.padding[0][1]
else:
rows = None
if input_shape[3] is not None:
cols = input_shape[3] + self.padding[1][0] + self.padding[1][1]
else:
cols = None
return (input_shape[0],
input_shape[1],
rows,
cols)
elif self.data_format == 'channels_last':
if input_shape[1] is not None:
rows = input_shape[1] + self.padding[0][0] + self.padding[0][1]
else:
rows = None
if input_shape[2] is not None:
cols = input_shape[2] + self.padding[1][0] + self.padding[1][1]
else:
cols = None
return (input_shape[0],
rows,
cols,
input_shape[3])
def call(self, inputs):
return spatial_reflection_2d_padding(inputs,
padding=self.padding,
data_format=self.data_format)
def get_config(self):
config = {'padding': self.padding,
'data_format': self.data_format}
base_config = super(ReflectionPadding2D, self).get_config()
return dict(list(base_config.items()) + list(config.items()))
class _Merge(Layer):
"""Generic merge layer for elementwise merge functions.
Used to implement `Sum`, `Average`, etc.
# Arguments
**kwargs: standard layer keyword arguments.
"""
def __init__(self, **kwargs):
super(_Merge, self).__init__(**kwargs)
self.supports_masking = True
def _merge_function(self, inputs):
raise NotImplementedError
def _compute_elemwise_op_output_shape(self, shape1, shape2):
"""Computes the shape of the resultant of an elementwise operation.
# Arguments
shape1: tuple or None. Shape of the first tensor
shape2: tuple or None. Shape of the second tensor
# Returns
expected output shape when an element-wise operation is
carried out on 2 tensors with shapes shape1 and shape2.
tuple or None.
# Raises
ValueError: if shape1 and shape2 are not compatible for
element-wise operations.
"""
if None in [shape1, shape2]:
return None
elif len(shape1) < len(shape2):
return self._compute_elemwise_op_output_shape(shape2, shape1)
elif len(shape2) == 0:
return shape1
output_shape = list(shape1[:-len(shape2)])
for i, j in zip(shape1[-len(shape2):], shape2):
if i is None or j is None:
output_shape.append(None)
elif i == 1:
output_shape.append(j)
elif j == 1:
output_shape.append(i)
else:
if i != j:
raise ValueError('Operands could not be broadcast '
'together with shapes ' +
str(shape1) + ' ' + str(shape2))
output_shape.append(i)
return tuple(output_shape)
def build(self, input_shape):
# Used purely for shape validation.
if not isinstance(input_shape, list):
raise ValueError('A merge layer should be called '
'on a list of inputs.')
if len(input_shape) < 2:
raise ValueError('A merge layer should be called '
'on a list of at least 2 inputs. '
'Got ' + str(len(input_shape)) + ' inputs.')
batch_sizes = [s[0] for s in input_shape if s is not None]
batch_sizes = set(batch_sizes)
batch_sizes -= set([None])
if len(batch_sizes) > 1:
raise ValueError('Can not merge tensors with different '
'batch sizes. Got tensors with shapes : ' +
str(input_shape))
if input_shape[0] is None:
output_shape = None
else:
output_shape = input_shape[0][1:]
for i in range(1, len(input_shape)):
if input_shape[i] is None:
shape = None
else:
shape = input_shape[i][1:]
output_shape = self._compute_elemwise_op_output_shape(output_shape, shape)
# If the inputs have different ranks, we have to reshape them
# to make them broadcastable.
if None not in input_shape and len(set(map(len, input_shape))) == 1:
self._reshape_required = False
else:
self._reshape_required = True
def call(self, inputs):
if self._reshape_required:
reshaped_inputs = []
input_ndims = list(map(K.ndim, inputs))
if None not in input_ndims:
# If ranks of all inputs are available,
# we simply expand each of them at axis=1
# until all of them have the same rank.
max_ndim = max(input_ndims)
for x in inputs:
x_ndim = K.ndim(x)
for _ in range(max_ndim - x_ndim):
x = K.expand_dims(x, 1)
reshaped_inputs.append(x)
return self._merge_function(reshaped_inputs)
else:
# Transpose all inputs so that batch size is the last dimension.
# (batch_size, dim1, dim2, ... ) -> (dim1, dim2, ... , batch_size)
transposed = False
for x in inputs:
x_ndim = K.ndim(x)
if x_ndim is None:
x_shape = K.shape(x)
batch_size = x_shape[0]
new_shape = K.concatenate([x_shape[1:], K.expand_dims(batch_size)])
x_transposed = K.reshape(x, K.stack([batch_size, K.prod(x_shape[1:])]))
x_transposed = K.permute_dimensions(x_transposed, (1, 0))
x_transposed = K.reshape(x_transposed, new_shape)
reshaped_inputs.append(x_transposed)
transposed = True
elif x_ndim > 1:
dims = list(range(1, x_ndim)) + [0]
reshaped_inputs.append(K.permute_dimensions(x, dims))
transposed = True
else:
# We don't transpose inputs if they are 1D vectors or scalars.
reshaped_inputs.append(x)
y = self._merge_function(reshaped_inputs)
y_ndim = K.ndim(y)
if transposed:
# If inputs have been transposed, we have to transpose the output too.
if y_ndim is None:
y_shape = K.shape(y)
y_ndim = K.shape(y_shape)[0]
batch_size = y_shape[y_ndim - 1]
new_shape = K.concatenate([K.expand_dims(batch_size), y_shape[:y_ndim - 1]])
y = K.reshape(y, (-1, batch_size))
y = K.permute_dimensions(y, (1, 0))
y = K.reshape(y, new_shape)
elif y_ndim > 1:
dims = [y_ndim - 1] + list(range(y_ndim - 1))
y = K.permute_dimensions(y, dims)
return y
else:
return self._merge_function(inputs)
def compute_output_shape(self, input_shape):
if input_shape[0] is None:
output_shape = None
else:
output_shape = input_shape[0][1:]
for i in range(1, len(input_shape)):
if input_shape[i] is None:
shape = None
else:
shape = input_shape[i][1:]
output_shape = self._compute_elemwise_op_output_shape(output_shape, shape)
batch_sizes = [s[0] for s in input_shape if s is not None]
batch_sizes = set(batch_sizes)
batch_sizes -= set([None])
if len(batch_sizes) == 1:
output_shape = (list(batch_sizes)[0],) + output_shape
else:
output_shape = (None,) + output_shape
return output_shape
def compute_mask(self, inputs, mask=None):
if mask is None:
return None
if not isinstance(mask, list):
raise ValueError('`mask` should be a list.')
if not isinstance(inputs, list):
raise ValueError('`inputs` should be a list.')
if len(mask) != len(inputs):
raise ValueError('The lists `inputs` and `mask` '
'should have the same length.')
if all([m is None for m in mask]):
return None
masks = [K.expand_dims(m, 0) for m in mask if m is not None]
return K.all(K.concatenate(masks, axis=0), axis=0, keepdims=False)
class Multiply(_Merge):
def _merge_function(self, inputs):
# batchsize=4
tmp1 = inputs[0][0][0]*inputs[1][0]
tmp1 = tf.expand_dims(tmp1, axis=0)
tmp2 = inputs[0][1][0]*inputs[1][1]
tmp2 = tf.expand_dims(tmp2, axis=0)
# tmp3 = inputs[0][2][0]*inputs[1][2]
# tmp3 = tf.expand_dims(tmp3, axis=0)
# tmp4 = inputs[0][3][0]*inputs[1][3]
# tmp4 = tf.expand_dims(tmp4, axis=0)
tmp = tf.concat([tmp1, tmp2], axis=0)
return tmp
class Divided(_Merge):
def _merge_function(self, inputs):
# batchsize=4
tmp1 = inputs[1][0]/(inputs[0][0][0]+1)
tmp1 = tf.expand_dims(tmp1, axis=0)
tmp2 = inputs[1][1]/(inputs[0][1][0]+1)
tmp2 = tf.expand_dims(tmp2, axis=0)
# tmp3 = inputs[0][2][0]*inputs[1][2]
# tmp3 = tf.expand_dims(tmp3, axis=0)
# tmp4 = inputs[0][3][0]*inputs[1][3]
# tmp4 = tf.expand_dims(tmp4, axis=0)
tmp = tf.concat([tmp1, tmp2], axis=0)
return tmp
if __name__ == "__main__":
input = Input(shape=(512, 512, 3))
x = ReflectionPadding2D(3)(input)
model = Model(input, x)
model.summary()