-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathSwitchableNormalization.py
234 lines (204 loc) · 10.5 KB
/
SwitchableNormalization.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
import os
os.environ["KERAS_BACKEND"] = "tensorflow"
from keras.engine import Layer, InputSpec
from keras import initializers
from keras import regularizers
from keras import constraints
from keras import backend as K
from keras.utils.generic_utils import get_custom_objects
class SwitchableNormalization(Layer):
"""Switchable normalization layer
# Arguments
axis: Integer, the axis that should be normalized
(typically the features axis).
For instance, after a `Conv2D` layer with
`data_format="channels_first"`,
set `axis=1` in `BatchNormalization`.
momentum: Momentum for the moving mean and the moving variance.
epsilon: Small float added to variance to avoid dividing by zero.
center: If True, add offset of `beta` to normalized tensor.
If False, `beta` is ignored.
scale: If True, multiply by `gamma`.
If False, `gamma` is not used.
When the next layer is linear (also e.g. `nn.relu`),
this can be disabled since the scaling
will be done by the next layer.
mean_weight_initializer: Initializer for mean weight
variance_weight_initializer: Initializer for variance weight
beta_initializer: Initializer for the beta weight.
gamma_initializer: Initializer for the gamma weight.
beta_regularizer: Optional regularizer for the beta weight.
gamma_regularizer: Optional regularizer for the gamma weight.
beta_constraint: Optional constraint for the beta weight.
gamma_constraint: Optional constraint for the gamma weight.
# Input shape
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
# Output shape
Same shape as input.
# References
- [Differentiable Learning-to-Normalize via Switchable Normalization](https://arxiv.org/abs/1806.10779)
- [SN by pytorch](https://github.com/switchablenorms/Switchable-Normalization/blob/master/models/switchable_norm.py)
"""
def __init__(self,
axis=-1,
momentum=0.99,
epsilon=1e-5,
center=True,
scale=True,
mean_weight_initializer = 'ones',
variance_weight_initializer = 'ones',
beta_initializer='zeros',
gamma_initializer='ones',
moving_mean_initializer='zeros',
moving_variance_initializer='ones',
beta_regularizer=None,
gamma_regularizer=None,
beta_constraint=None,
gamma_constraint=None,
**kwargs):
super(SwitchableNormalization, self).__init__(**kwargs)
self.supports_masking = True
self.axis = axis
self.momentum = momentum
self.epsilon = epsilon
self.center = center
self.scale = scale
self.mean_weight_initializer = initializers.get(mean_weight_initializer)
self.variance_weight_initializer = initializers.get(variance_weight_initializer)
self.beta_initializer = initializers.get(beta_initializer)
self.gamma_initializer = initializers.get(gamma_initializer)
self.moving_mean_initializer = initializers.get(moving_mean_initializer)
self.moving_variance_initializer = initializers.get(moving_variance_initializer)
self.beta_regularizer = regularizers.get(beta_regularizer)
self.gamma_regularizer = regularizers.get(gamma_regularizer)
self.beta_constraint = constraints.get(beta_constraint)
self.gamma_constraint = constraints.get(gamma_constraint)
def build(self, input_shape):
dim = input_shape[self.axis]
if dim is None:
raise ValueError('Axis ' + str(self.axis) + ' of '
'input tensor should have a defined dimension '
'but the layer received an input with shape ' +
str(input_shape) + '.')
self.input_spec = InputSpec(ndim=len(input_shape),
axes={self.axis: dim})
# add mean/variance weight
self.mean_weight = self.add_weight(shape=(3,),
name="mean_weight",
initializer=self.mean_weight_initializer)
self.variance_weight = self.add_weight(shape=(3,),
name="variance_weight",
initializer=self.variance_weight_initializer)
# add gamma/beta weight
shape = (dim,)
if self.scale:
self.gamma = self.add_weight(shape=shape,
name='gamma',
initializer=self.gamma_initializer,
regularizer=self.gamma_regularizer,
constraint=self.gamma_constraint)
else:
self.gamma = None
if self.center:
self.beta = self.add_weight(shape=shape,
name='beta',
initializer=self.beta_initializer,
regularizer=self.beta_regularizer,
constraint=self.beta_constraint)
else:
self.beta = None
self.moving_mean = self.add_weight(
shape=shape,
name="moving_mean",
initializer=self.moving_mean_initializer,
trainable=False)
self.moving_variance = self.add_weight(
shape=shape,
name="moving_variance",
initializer=self.moving_variance_initializer,
trainable=False)
self.built = True
def call(self, inputs, training=None):
input_shape = K.int_shape(inputs)
# Prepare broadcasting shape.
ndim = len(input_shape)
broadcast_shape = [1] * len(input_shape)
broadcast_shape[self.axis] = input_shape[self.axis]
# mean/variance of instance_normalization
reduction_axes_in = list(range(len(input_shape)))
del reduction_axes_in[self.axis]
del reduction_axes_in[0]
mean_in = K.mean(inputs, axis=reduction_axes_in, keepdims=True)
variance_in = K.var(inputs, axis=reduction_axes_in, keepdims=True)
# mean/variance of layer_normalization
reduction_axes_ln = list(range(len(input_shape)))
del reduction_axes_ln[0]
mean_ln = K.mean(inputs, axis=reduction_axes_ln, keepdims=True)
variance_ln = K.var(inputs, axis=reduction_axes_ln, keepdims=True)
# mean/variance of batch_normalization
reduction_axes_bn = list(range(len(input_shape)))
del reduction_axes_bn[self.axis]
def normed_training():
mean_bn = K.mean(inputs, axis=reduction_axes_bn,keepdims=True)
variance_bn = K.var(inputs, axis=reduction_axes_bn,keepdims=True)
mean = [mean_in, mean_ln, mean_bn]
variance = [variance_in, variance_ln, variance_bn]
# If the learning is either dynamic, or set to training:
self.add_update([K.moving_average_update(self.moving_mean,
K.reshape(mean_bn,(input_shape[self.axis],)),
self.momentum),
K.moving_average_update(self.moving_variance,
K.reshape(variance_bn,(input_shape[self.axis],)),
self.momentum)],
inputs)
return norm(mean, variance)
def normalize_inference():
mean_bn = self.moving_mean
variance_bn = self.moving_variance
mean = [mean_in, mean_ln, mean_bn]
variance = [variance_in, variance_ln, variance_bn]
return norm(mean, variance)
def norm(mean,variance):
mean_weight = K.softmax(self.mean_weight)
variance_weight = K.softmax(self.variance_weight)
norm_mean = mean_weight[0]*mean[0] + mean_weight[1]*mean[1] + mean_weight[2]*mean[2]
norm_variance = variance_weight[0]*variance[0] + variance_weight[1]*variance[1] + variance_weight[2]*variance[2]
normd = (inputs - norm_mean) / (K.sqrt(norm_variance + self.epsilon))
if self.scale:
broadcast_gamma = K.reshape(self.gamma, broadcast_shape)
normd = normd * broadcast_gamma
if self.center:
broadcast_beta = K.reshape(self.beta, broadcast_shape)
normd = normd + broadcast_beta
return normd
if training in {0,False}:
return normalize_inference
# Pick the normalized form corresponding to the training phase.
return K.in_train_phase(normed_training,
normalize_inference,
training=training)
def get_config(self):
config = {
'axis': self.axis,
'momentum': self.momentum,
'epsilon': self.epsilon,
'center': self.center,
'scale': self.scale,
'mean_weight_initializer': self.mean_weight_initializer,
'variance_weight_initializer': self.variance_weight_initializer,
'beta_initializer': initializers.serialize(self.beta_initializer),
'gamma_initializer': initializers.serialize(self.gamma_initializer),
'moving_mean_initializer': initializers.serialize(self.moving_mean_initializer),
'moving_variance_initializer': initializers.serialize(self.moving_variance_initializer),
'beta_regularizer': regularizers.serialize(self.beta_regularizer),
'gamma_regularizer': regularizers.serialize(self.gamma_regularizer),
'beta_constraint': constraints.serialize(self.beta_constraint),
'gamma_constraint': constraints.serialize(self.gamma_constraint)
}
base_config = super(SwitchableNormalization, self).get_config()
return dict(list(base_config.items()) + list(config.items()))
def compute_output_shape(self, input_shape):
return input_shape
get_custom_objects().update({'SwitchableNormalization': SwitchableNormalization})