-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathresnet.py
180 lines (144 loc) · 5.93 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
'''ResNet in PyTorch.
For Pre-activation ResNet, see 'preact_resnet.py'.
Reference:
[1] Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun
Deep Residual Learning for Image Recognition. arXiv:1512.03385
'''
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
class BasicBlock(nn.Module):
expansion = 1
def __init__(self, in_planes, planes, stride=1):
super(BasicBlock, self).__init__()
self.conv1 = nn.Conv2d(in_planes, planes, kernel_size=3, stride=stride, padding=1, bias=False)
self.bn1 = nn.BatchNorm2d(planes)
self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=1, padding=1, bias=False)
self.bn2 = nn.BatchNorm2d(planes)
self.shortcut = nn.Sequential()
if stride != 1 or in_planes != self.expansion * planes:
self.shortcut = nn.Sequential(
nn.Conv2d(in_planes, self.expansion * planes, kernel_size=1, stride=stride, bias=False),
nn.BatchNorm2d(self.expansion * planes)
)
def forward(self, x):
out = F.relu(self.bn1(self.conv1(x)))
out = self.bn2(self.conv2(out))
out += self.shortcut(x)
out = F.relu(out)
return out
class Bottleneck(nn.Module):
expansion = 4
def __init__(self, in_planes, planes, stride=1):
super(Bottleneck, self).__init__()
self.conv1 = nn.Conv2d(in_planes, 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, self.expansion * planes, kernel_size=1, bias=False)
self.bn3 = nn.BatchNorm2d(self.expansion * planes)
self.shortcut = nn.Sequential()
if stride != 1 or in_planes != self.expansion * planes:
self.shortcut = nn.Sequential(
nn.Conv2d(in_planes, self.expansion * planes, kernel_size=1, stride=stride, bias=False),
nn.BatchNorm2d(self.expansion * planes)
)
def forward(self, x):
out = F.relu(self.bn1(self.conv1(x)))
out = F.relu(self.bn2(self.conv2(out)))
out = self.bn3(self.conv3(out))
out += self.shortcut(x)
out = F.relu(out)
return out
class ResNet(nn.Module):
def __init__(self, block, num_blocks, num_classes=10):
super(ResNet, self).__init__()
self.in_planes = 64
self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1, bias=False)
self.bn1 = nn.BatchNorm2d(64)
self.layer1 = self._make_layer(block, 64, num_blocks[0], stride=1)
self.layer2 = self._make_layer(block, 128, num_blocks[1], stride=2)
self.layer3 = self._make_layer(block, 256, num_blocks[2], stride=2)
self.layer4 = self._make_layer(block, 512, num_blocks[3], stride=2)
self.linear = nn.Linear(512 * block.expansion, num_classes)
def _make_layer(self, block, planes, num_blocks, stride):
strides = [stride] + [1] * (num_blocks - 1)
layers = []
for stride in strides:
layers.append(block(self.in_planes, planes, stride))
self.in_planes = planes * block.expansion
return nn.Sequential(*layers)
def forward(self, x):
out = F.relu(self.bn1(self.conv1(x)))
out = self.layer1(out)
out = self.layer2(out)
out = self.layer3(out)
out = self.layer4(out)
out = F.avg_pool2d(out, 4)
out = out.view(out.size(0), -1)
out = self.linear(out)
return out
def reg_loss(self):
reg_loss = 0.0
for m in self.modules():
if isinstance(m, nn.Linear) or isinstance(m, nn.Conv2d):
reg_loss += torch.sum(torch.abs(m.weight))
return reg_loss
def l1reg_loss(self):
reg_loss = 0.0
for m in self.modules():
if isinstance(m, nn.Linear) or isinstance(m, nn.Conv2d):
reg_loss += torch.sum(torch.abs(m.weight))
return reg_loss
def l12reg_loss(self):
reg_loss = 0.0
for m in self.modules():
if isinstance(m, nn.Linear) or isinstance(m, nn.Conv2d):
reg_loss += torch.sum((torch.abs(m.weight) + 1e-6).sqrt())
return reg_loss
def l23reg_loss(self):
reg_loss = 0.0
for m in self.modules():
if isinstance(m, nn.Linear) or isinstance(m, nn.Conv2d):
reg_loss += torch.sum((torch.abs(m.weight) + 1e-6).pow(2 / 3))
return reg_loss
def exact_sparsity(self):
nnz = 0
total_param = 0.0
for m in self.modules():
if isinstance(m, nn.Linear) or isinstance(m, nn.Conv2d):
total_param += np.prod(m.weight.data.shape)
nnz += torch.sum(m.weight.data != 0).detach().item()
ratio = nnz / total_param
return ratio
def sparsity_level(self):
nnz_2 = 0.0
nnz_3 = 0.0
nnz_4 = 0.0
total_param = 0.0
for m in self.modules():
if isinstance(m, nn.Linear) or isinstance(m, nn.Conv2d):
total_param += np.prod(m.weight.data.shape)
nnz_2 += torch.sum(m.weight.data.abs() >= 0.01).detach().item()
nnz_3 += torch.sum(m.weight.data.abs() >= 0.001).detach().item()
nnz_4 += torch.sum(m.weight.data.abs() >= 0.0001).detach().item()
ratio_2 = nnz_2 / total_param
ratio_3 = nnz_3 / total_param
ratio_4 = nnz_4 / total_param
return ratio_2, ratio_3
def ResNet18():
return ResNet(BasicBlock, [2, 2, 2, 2])
def ResNet34():
return ResNet(BasicBlock, [3, 4, 6, 3])
def ResNet50():
return ResNet(Bottleneck, [3, 4, 6, 3])
def ResNet101():
return ResNet(Bottleneck, [3, 4, 23, 3])
def ResNet152():
return ResNet(Bottleneck, [3, 8, 36, 3])
def test():
net = ResNet18()
y = net(torch.randn(1, 3, 32, 32))
print(y.size())
# test()