-
Notifications
You must be signed in to change notification settings - Fork 58
/
shuffle_seg_skipnet.py
375 lines (290 loc) · 11.7 KB
/
shuffle_seg_skipnet.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
'''
Shuffle_Seg_SkipNet
Author: Zhengwei Li
Data: July 30 2018
'''
import torch
import torch.nn as nn
import torch.nn.functional as F
from collections import OrderedDict
from torch.nn import init
from model.layers import *
def conv3x3(in_channels, out_channels, stride=1,
padding=1, bias=True, groups=1):
"""3x3 convolution with padding
"""
return nn.Conv2d(
in_channels,
out_channels,
kernel_size=3,
stride=stride,
padding=padding,
bias=bias,
groups=groups)
def conv1x1(in_channels, out_channels, groups=1, bias=True):
"""1x1 convolution with padding
- Normal pointwise convolution When groups == 1
- Grouped pointwise convolution when groups > 1
"""
return nn.Conv2d(
in_channels,
out_channels,
kernel_size=1,
groups=groups,
stride=1,
bias=bias)
def channel_shuffle(x, groups):
batchsize, num_channels, height, width = x.data.size()
channels_per_group = num_channels // groups
# reshape
x = x.view(batchsize, groups,
channels_per_group, height, width)
# transpose
# - contiguous() required if transpose() is used before view().
# See https://github.com/pytorch/pytorch/issues/764
x = torch.transpose(x, 1, 2).contiguous()
# flatten
x = x.view(batchsize, -1, height, width)
return x
class ShuffleUnit(nn.Module):
def __init__(self, in_channels, out_channels, groups=3,
grouped_conv=True, combine='add'):
super(ShuffleUnit, self).__init__()
self.in_channels = in_channels
self.out_channels = out_channels
self.grouped_conv = grouped_conv
self.combine = combine
self.groups = groups
self.bottleneck_channels = self.out_channels // 4
# define the type of ShuffleUnit
if self.combine == 'add':
# ShuffleUnit Figure 2b
self.depthwise_stride = 1
self._combine_func = self._add
elif self.combine == 'concat':
# ShuffleUnit Figure 2c
self.depthwise_stride = 2
self._combine_func = self._concat
# ensure output of concat has the same channels as
# original output channels.
self.out_channels -= self.in_channels
else:
raise ValueError("Cannot combine tensors with \"{}\"" \
"Only \"add\" and \"concat\" are" \
"supported".format(self.combine))
# Use a 1x1 grouped or non-grouped convolution to reduce input channels
# to bottleneck channels, as in a ResNet bottleneck module.
# NOTE: Do not use group convolution for the first conv1x1 in Stage 2.
self.first_1x1_groups = self.groups if grouped_conv else 1
self.g_conv_1x1_compress = self._make_grouped_conv1x1(
self.in_channels,
self.bottleneck_channels,
self.first_1x1_groups,
batch_norm=True,
relu=True
)
# 3x3 depthwise convolution followed by batch normalization
self.depthwise_conv3x3 = conv3x3(
self.bottleneck_channels, self.bottleneck_channels,
stride=self.depthwise_stride, groups=self.bottleneck_channels)
self.bn_after_depthwise = nn.BatchNorm2d(self.bottleneck_channels)
# Use 1x1 grouped convolution to expand from
# bottleneck_channels to out_channels
self.g_conv_1x1_expand = self._make_grouped_conv1x1(
self.bottleneck_channels,
self.out_channels,
self.groups,
batch_norm=True,
relu=False
)
@staticmethod
def _add(x, out):
# residual connection
return x + out
@staticmethod
def _concat(x, out):
# concatenate along channel axis
return torch.cat((x, out), 1)
def _make_grouped_conv1x1(self, in_channels, out_channels, groups,
batch_norm=True, relu=False):
modules = OrderedDict()
conv = conv1x1(in_channels, out_channels, groups=groups)
modules['conv1x1'] = conv
if batch_norm:
modules['batch_norm'] = nn.BatchNorm2d(out_channels)
if relu:
modules['relu'] = nn.ReLU()
if len(modules) > 1:
return nn.Sequential(modules)
else:
return conv
def forward(self, x):
# save for combining later with output
residual = x
if self.combine == 'concat':
residual = F.avg_pool2d(residual, kernel_size=3,
stride=2, padding=1)
out = self.g_conv_1x1_compress(x)
out = channel_shuffle(out, self.groups)
out = self.depthwise_conv3x3(out)
out = self.bn_after_depthwise(out)
out = self.g_conv_1x1_expand(out)
out = self._combine_func(residual, out)
return F.relu(out)
class ShuffleNet(nn.Module):
"""ShuffleNet implementation.
"""
def __init__(self, groups=3, in_channels=3):
"""ShuffleNet constructor.
Arguments:
groups (int, optional): number of groups to be used in grouped
1x1 convolutions in each ShuffleUnit. Default is 3 for best
performance according to original paper.
in_channels (int, optional): number of channels in the input tensor.
Default is 3 for RGB image inputs.
num_classes (int, optional): number of classes to predict. Default
is 1000 for ImageNet.
"""
super(ShuffleNet, self).__init__()
self.groups = groups
self.stage_repeats = [3, 7, 3]
self.in_channels = in_channels
# index 0 is invalid and should never be called.
# only used for indexing convenience.
if groups == 1:
self.stage_out_channels = [-1, 24, 144, 288, 567]
elif groups == 2:
self.stage_out_channels = [-1, 24, 200, 400, 800]
elif groups == 3:
self.stage_out_channels = [-1, 24, 240, 480, 960]
elif groups == 4:
self.stage_out_channels = [-1, 24, 272, 544, 1088]
elif groups == 8:
self.stage_out_channels = [-1, 24, 384, 768, 1536]
else:
raise ValueError(
"""{} groups is not supported for
1x1 Grouped Convolutions""".format(num_groups))
# Stage 1 always has 24 output channels
self.conv1 = conv3x3(self.in_channels,
self.stage_out_channels[1], # stage 1
stride=2)
self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
# Stage 2
self.stage2 = self._make_stage(2)
# Stage 3
self.stage3 = self._make_stage(3)
# Stage 4
self.stage4 = self._make_stage(4)
def _make_stage(self, stage):
modules = OrderedDict()
stage_name = "ShuffleUnit_Stage{}".format(stage)
# First ShuffleUnit in the stage
# 1. non-grouped 1x1 convolution (i.e. pointwise convolution)
# is used in Stage 2. Group convolutions used everywhere else.
grouped_conv = stage > 2
# 2. concatenation unit is always used.
first_module = ShuffleUnit(
self.stage_out_channels[stage-1],
self.stage_out_channels[stage],
groups=self.groups,
grouped_conv=grouped_conv,
combine='concat'
)
modules[stage_name+"_0"] = first_module
# add more ShuffleUnits depending on pre-defined number of repeats
for i in range(self.stage_repeats[stage-2]):
name = stage_name + "_{}".format(i+1)
module = ShuffleUnit(
self.stage_out_channels[stage],
self.stage_out_channels[stage],
groups=self.groups,
grouped_conv=True,
combine='add'
)
modules[name] = module
return nn.Sequential(modules)
def forward(self, x):
s0 = self.conv1(x)
s1 = self.maxpool(s0)
s2 = self.stage2(s1)
s3 = self.stage3(s2)
s4 = self.stage4(s3)
return s0, s1, s2, s3, s4
class Shuffle_Seg_SkipNet(nn.Module):
def __init__(self, groups=3, in_channels=3, n_classes=1):
"""ShuffleNet constructor.
"""
super(Shuffle_Seg_SkipNet, self).__init__()
self.encoder = ShuffleNet(groups=groups, in_channels=in_channels)
self.scorelayer = conv1x1(960, 1, bias=True)
self.bn_ = nn.BatchNorm2d(1)
self.up1 = BilinearConvTranspose2d(1, stride=2, groups=1)
self.stage3_down = conv1x1(480, 1, groups=1, bias=False)
self.up2 = BilinearConvTranspose2d(1, stride=2, groups=1)
self.stage2_down = conv1x1(240, 1, groups=1, bias=False)
self.up3 = BilinearConvTranspose2d(1, stride=2, groups=1)
self.stage1_down = conv1x1(24, 1, groups=1, bias=False)
self.up4 = BilinearConvTranspose2d(1, stride=2, groups=1)
self.stage0_down = conv1x1(24, 1, groups=1, bias=False)
self.deconv = BilinearConvTranspose2d(1, stride=2, groups=1)
self.init_params()
def init_params(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
init.kaiming_normal_(m.weight, mode='fan_out')
if m.bias is not None:
init.constant_(m.bias, 0)
elif isinstance(m, nn.BatchNorm2d):
init.constant_(m.weight, 1)
init.constant_(m.bias, 0)
elif isinstance(m, nn.Linear):
init.normal_(m.weight, std=0.001)
if m.bias is not None:
init.constant_(m.bias, 0)
def _make_stage(self, stage):
modules = OrderedDict()
stage_name = "ShuffleUnit_Stage{}".format(stage)
# First ShuffleUnit in the stage
# 1. non-grouped 1x1 convolution (i.e. pointwise convolution)
# is used in Stage 2. Group convolutions used everywhere else.
grouped_conv = stage > 2
# 2. concatenation unit is always used.
first_module = ShuffleUnit(
self.stage_out_channels[stage-1],
self.stage_out_channels[stage],
groups=self.groups,
grouped_conv=grouped_conv,
combine='concat'
)
modules[stage_name+"_0"] = first_module
# add more ShuffleUnits depending on pre-defined number of repeats
for i in range(self.stage_repeats[stage-2]):
name = stage_name + "_{}".format(i+1)
module = ShuffleUnit(
self.stage_out_channels[stage],
self.stage_out_channels[stage],
groups=self.groups,
grouped_conv=True,
combine='add'
)
modules[name] = module
return nn.Sequential(modules)
def forward(self, x):
s0, s1, s2, s3, s4 = self.encoder(x)
# to n_class heat_map
heat_map = self.bn_(self.scorelayer(s4))
heat_map = self.bn_(self.up1(heat_map))
s3_heat_map = self.bn_(self.stage3_down(s3))
heat_map = heat_map + s3_heat_map
heat_map = self.bn_(self.up2(heat_map))
s2_heat_map = self.bn_(self.stage2_down(s2))
heat_map = heat_map + s2_heat_map
heat_map = self.bn_(self.up3(heat_map))
s1_heat_map = self.bn_(self.stage1_down(s1))
heat_map = heat_map + s1_heat_map
heat_map = self.bn_(self.up4(heat_map))
s0_heat_map = self.bn_(self.stage0_down(s0))
heat_map = heat_map + s0_heat_map
heat_map = self.deconv(heat_map)
return heat_map