-
Notifications
You must be signed in to change notification settings - Fork 14
/
resnet.py
256 lines (196 loc) · 7.57 KB
/
resnet.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
from __future__ import absolute_import
'''Resnet for cifar dataset.
Ported form
https://github.com/facebook/fb.resnet.torch
and
https://github.com/pytorch/vision/blob/master/torchvision/models/resnet.py
(c) YANG, Wei
'''
import torch.nn as nn
import torch.nn.functional as F
import math
__all__ = ['resnet']
def conv3x3(in_planes, out_planes, stride=1):
"""3x3 convolution with padding"""
return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride,
padding=1, bias=False)
class BasicBlock(nn.Module):
expansion = 1
def __init__(self, inplanes, planes, stride=1, downsample=None, is_last=False):
super(BasicBlock, self).__init__()
self.is_last = is_last
self.conv1 = conv3x3(inplanes, planes, stride)
self.bn1 = nn.BatchNorm2d(planes)
self.relu = nn.ReLU(inplace=True)
self.conv2 = conv3x3(planes, planes)
self.bn2 = nn.BatchNorm2d(planes)
self.downsample = downsample
self.stride = stride
def forward(self, x):
residual = x
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)
out = self.conv2(out)
out = self.bn2(out)
if self.downsample is not None:
residual = self.downsample(x)
out += residual
preact = out
out = F.relu(out)
if self.is_last:
return out, preact
else:
return out
class Bottleneck(nn.Module):
expansion = 4
def __init__(self, inplanes, planes, stride=1, downsample=None, is_last=False):
super(Bottleneck, self).__init__()
self.is_last = is_last
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
self.bn1 = nn.BatchNorm2d(planes)
self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride,
padding=1, bias=False)
self.bn2 = nn.BatchNorm2d(planes)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.bn3 = nn.BatchNorm2d(planes * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
def forward(self, x):
residual = x
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)
out = self.conv2(out)
out = self.bn2(out)
out = self.relu(out)
out = self.conv3(out)
out = self.bn3(out)
if self.downsample is not None:
residual = self.downsample(x)
out += residual
preact = out
out = F.relu(out)
if self.is_last:
return out, preact
else:
return out
class ResNet(nn.Module):
def __init__(self, depth, num_filters, block_name='BasicBlock', num_classes=10):
super(ResNet, self).__init__()
# Model type specifies number of layers for CIFAR-10 model
if block_name.lower() == 'basicblock':
assert (depth - 2) % 6 == 0, 'When use basicblock, depth should be 6n+2, e.g. 20, 32, 44, 56, 110, 1202'
n = (depth - 2) // 6
block = BasicBlock
elif block_name.lower() == 'bottleneck':
assert (depth - 2) % 9 == 0, 'When use bottleneck, depth should be 9n+2, e.g. 20, 29, 47, 56, 110, 1199'
n = (depth - 2) // 9
block = Bottleneck
else:
raise ValueError('block_name shoule be Basicblock or Bottleneck')
self.inplanes = num_filters[0]
self.conv1 = nn.Conv2d(3, num_filters[0], kernel_size=3, padding=1,
bias=False)
self.bn1 = nn.BatchNorm2d(num_filters[0])
self.relu = nn.ReLU(inplace=True)
self.layer1 = self._make_layer(block, num_filters[1], n)
self.layer2 = self._make_layer(block, num_filters[2], n, stride=2)
self.layer3 = self._make_layer(block, num_filters[3], n, stride=2)
self.avgpool = nn.AvgPool2d(8)
self.fc = nn.Linear(num_filters[3] * block.expansion, num_classes)
for m in self.modules():
if isinstance(m, nn.Conv2d):
nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
elif isinstance(m, (nn.BatchNorm2d, nn.GroupNorm)):
nn.init.constant_(m.weight, 1)
nn.init.constant_(m.bias, 0)
def _make_layer(self, block, planes, blocks, stride=1):
downsample = None
if stride != 1 or self.inplanes != planes * block.expansion:
downsample = nn.Sequential(
nn.Conv2d(self.inplanes, planes * block.expansion,
kernel_size=1, stride=stride, bias=False),
nn.BatchNorm2d(planes * block.expansion),
)
layers = list([])
layers.append(block(self.inplanes, planes, stride, downsample, is_last=(blocks == 1)))
self.inplanes = planes * block.expansion
for i in range(1, blocks):
layers.append(block(self.inplanes, planes, is_last=(i == blocks-1)))
return nn.Sequential(*layers)
def get_feat_modules(self):
feat_m = nn.ModuleList([])
feat_m.append(self.conv1)
feat_m.append(self.bn1)
feat_m.append(self.relu)
feat_m.append(self.layer1)
feat_m.append(self.layer2)
feat_m.append(self.layer3)
return feat_m
def get_bn_before_relu(self):
if isinstance(self.layer1[0], Bottleneck):
bn1 = self.layer1[-1].bn3
bn2 = self.layer2[-1].bn3
bn3 = self.layer3[-1].bn3
elif isinstance(self.layer1[0], BasicBlock):
bn1 = self.layer1[-1].bn2
bn2 = self.layer2[-1].bn2
bn3 = self.layer3[-1].bn2
else:
raise NotImplementedError('ResNet unknown block error !!!')
return [bn1, bn2, bn3]
def forward(self, x, is_feat=False, preact=False):
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x) # 32x32
f0 = x
x, f1_pre = self.layer1(x) # 32x32
f1 = x
x, f2_pre = self.layer2(x) # 16x16
f2 = x
x, f3_pre = self.layer3(x) # 8x8
f3 = x
x = self.avgpool(x)
x = x.view(x.size(0), -1)
f4 = x
x = self.fc(x)
if is_feat:
if preact:
return [f0, f1_pre, f2_pre, f3_pre, f4], x
else:
return [f0, f1, f2, f3, f4], x
else:
return x
def resnet8(**kwargs):
return ResNet(8, [16, 16, 32, 64], 'basicblock', **kwargs)
def resnet14(**kwargs):
return ResNet(14, [16, 16, 32, 64], 'basicblock', **kwargs)
def resnet20(**kwargs):
return ResNet(20, [16, 16, 32, 64], 'basicblock', **kwargs)
def resnet32(**kwargs):
return ResNet(32, [16, 16, 32, 64], 'basicblock', **kwargs)
def resnet44(**kwargs):
return ResNet(44, [16, 16, 32, 64], 'basicblock', **kwargs)
def resnet56(**kwargs):
return ResNet(56, [16, 16, 32, 64], 'basicblock', **kwargs)
def resnet110(**kwargs):
return ResNet(110, [16, 16, 32, 64], 'basicblock', **kwargs)
def resnet8x4(**kwargs):
return ResNet(8, [32, 64, 128, 256], 'basicblock', **kwargs)
def resnet32x4(**kwargs):
return ResNet(32, [32, 64, 128, 256], 'basicblock', **kwargs)
if __name__ == '__main__':
import torch
x = torch.randn(2, 3, 32, 32)
net = resnet8x4(num_classes=20)
feats, logit = net(x, is_feat=True, preact=True)
for f in feats:
print(f.shape, f.min().item())
print(logit.shape)
for m in net.get_bn_before_relu():
if isinstance(m, nn.BatchNorm2d):
print('pass')
else:
print('warning')