forked from mankasto/iLRG
-
Notifications
You must be signed in to change notification settings - Fork 0
/
models.py
261 lines (231 loc) · 11 KB
/
models.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
import torch
import torchvision
import torch.nn as nn
from torchvision.models.resnet import BasicBlock, Bottleneck
# Models
# Only for MNIST
class DNN(nn.Module):
def __init__(self, n_hidden=1, dim=300):
super(DNN, self).__init__()
self.fin = nn.Sequential(nn.Linear(784, dim), nn.ReLU())
self.hiddens = []
for i in range(n_hidden):
self.hiddens.append(nn.Sequential(
nn.Linear(dim, dim),
nn.ReLU()
))
self.hiddens = nn.Sequential(*self.hiddens)
self.fc = nn.Linear(dim, 10)
def forward(self, x):
x = x.reshape(-1, 28 * 28)
x = self.fin(x)
x = self.hiddens(x)
y = self.fc(x)
return y, x
class LeNet5(nn.Module):
# MNIST:1,256,10; CIFAR-10:3,400,10; CIFAR-100:3,400,100;
def __init__(self, channel=1, hidden=256, num_classes=10, ns=0.01, bn=False, silu=False, leaky_relu=False):
super(LeNet5, self).__init__()
self.block1 = nn.Sequential(nn.Conv2d(channel, 6, kernel_size=5, stride=1, padding=0),
nn.BatchNorm2d(6) if bn else nn.Identity(),
nn.SiLU() if silu else nn.LeakyReLU(negative_slope=ns) if leaky_relu else nn.ReLU(),
nn.MaxPool2d(kernel_size=2, stride=2))
self.block2 = nn.Sequential(nn.Conv2d(6, 16, kernel_size=5, stride=1, padding=0),
nn.BatchNorm2d(16) if bn else nn.Identity(),
nn.SiLU() if silu else nn.LeakyReLU(negative_slope=ns) if leaky_relu else nn.ReLU(),
nn.MaxPool2d(kernel_size=2, stride=2))
self.nlt1 = nn.Sequential(nn.Linear(hidden, 120),
nn.SiLU() if silu else nn.LeakyReLU(negative_slope=ns) if leaky_relu else nn.ReLU())
self.nlt2 = nn.Sequential(nn.Linear(120, 84),
nn.SiLU() if silu else nn.LeakyReLU(negative_slope=ns) if leaky_relu else nn.ReLU())
self.fc = torch.nn.Linear(84, 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.init.constant_(m.weight, 1)
nn.init.constant_(m.bias, 0)
def forward(self, x):
out = self.block1(x)
out = self.block2(out)
emb = out.view(out.size(0), -1)
emb = self.nlt1(emb)
emb = self.nlt2(emb)
out = self.fc(emb)
return out, emb
class LeNetZhu(nn.Module):
# MNIST:1,588,10; CIFAR-10:3,768,10; CIFAR-100:3,768,100;
def __init__(self, channel=3, hidden=768, num_classes=10):
super(LeNetZhu, self).__init__()
self.body = nn.Sequential(
nn.Conv2d(channel, 12, kernel_size=5, padding=5 // 2, stride=2),
nn.ReLU(),
nn.Conv2d(12, 12, kernel_size=5, padding=5 // 2, stride=2),
nn.ReLU(),
nn.Conv2d(12, 12, kernel_size=5, padding=5 // 2, stride=1),
nn.ReLU(),
)
self.fc = nn.Sequential(
nn.Linear(hidden, num_classes)
)
for m in self.modules():
if hasattr(m, "weight"):
m.weight.data.uniform_(-0.5, 0.5)
if hasattr(m, "bias"):
m.bias.data.uniform_(-0.5, 0.5)
def forward(self, x):
out = self.body(x)
emb = out.view(out.size(0), -1)
out = self.fc(emb)
return out, emb
class VGG(nn.Module):
def __init__(self, block_nums, hidden, num_classes=10, dropout=True):
super(VGG, self).__init__()
self.block1 = self._make_layers(in_channels=3, out_channels=64, block_num=block_nums[0])
self.block2 = self._make_layers(in_channels=64, out_channels=128, block_num=block_nums[1])
self.block3 = self._make_layers(in_channels=128, out_channels=256, block_num=block_nums[2])
self.block4 = self._make_layers(in_channels=256, out_channels=512, block_num=block_nums[3])
self.block5 = self._make_layers(in_channels=512, out_channels=512, block_num=block_nums[4])
self.exactor = nn.Sequential(
nn.Linear(hidden, 4096),
nn.ReLU(inplace=True),
nn.Dropout(0.5) if dropout else nn.Identity(),
nn.Linear(4096, 4096),
nn.ReLU(inplace=True),
nn.Dropout(0.5) if dropout else nn.Identity()
)
self.fc = nn.Linear(4096, 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)
@staticmethod
def _make_layers(in_channels, out_channels, block_num):
blocks = [nn.Sequential(
nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=1, padding=1),
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True)
)]
for i in range(1, block_num):
blocks.append(nn.Sequential(
nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=1, padding=1),
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True)
))
blocks.append(nn.MaxPool2d(kernel_size=2, stride=2))
return nn.Sequential(*blocks)
def forward(self, x):
x = self.block1(x)
x = self.block2(x)
x = self.block3(x)
x = self.block4(x)
x = self.block5(x)
x = torch.flatten(x, start_dim=1)
emb = self.exactor(x)
out = self.fc(emb)
return out, emb
# ResNet implementation from https://github.com/JonasGeiping/invertinggradients
class ResNet(torchvision.models.ResNet):
"""ResNet generalization for CIFAR thingies."""
def __init__(self, block, layers, num_classes=10, zero_init_residual=False,
groups=1, channel=3, base_width=64, replace_stride_with_dilation=None,
norm_layer=None, strides=[1, 2, 2, 2], pool='avg'):
"""Initialize as usual. Layers and strides are scriptable."""
super(torchvision.models.ResNet, self).__init__() # nn.Module
if norm_layer is None:
norm_layer = nn.BatchNorm2d
self._norm_layer = norm_layer
self.dilation = 1
if replace_stride_with_dilation is None:
# each element in the tuple indicates if we should replace
# the 2x2 stride with a dilated convolution instead
replace_stride_with_dilation = [False, False, False, False]
if len(replace_stride_with_dilation) != 4:
raise ValueError("replace_stride_with_dilation should be None "
"or a 4-element tuple, got {}".format(replace_stride_with_dilation))
self.groups = groups
self.inplanes = base_width
self.base_width = 64 # Do this to circumvent BasicBlock errors. The value is not actually used.
self.conv1 = nn.Conv2d(channel, self.inplanes, kernel_size=3, stride=1, padding=1, bias=False)
self.bn1 = norm_layer(self.inplanes)
self.relu = nn.ReLU(inplace=True)
self.layers = torch.nn.ModuleList()
width = self.inplanes
for idx, layer in enumerate(layers):
self.layers.append(
self._make_layer(block, width, layer, stride=strides[idx], dilate=replace_stride_with_dilation[idx]))
width *= 2
self.pool = nn.AdaptiveAvgPool2d((1, 1)) if pool == 'avg' else nn.AdaptiveMaxPool2d((1, 1))
self.fc = nn.Linear(width // 2 * 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)
# Zero-initialize the last BN in each residual branch,
# so that the residual branch starts with zeros, and each residual block behaves like an identity.
# This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677
if zero_init_residual:
for m in self.modules():
if isinstance(m, Bottleneck):
nn.init.constant_(m.bn3.weight, 0)
elif isinstance(m, BasicBlock):
nn.init.constant_(m.bn2.weight, 0)
def _forward_impl(self, x):
# See note [TorchScript super()]
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
for layer in self.layers:
x = layer(x)
x = self.pool(x)
x = torch.flatten(x, 1)
y = self.fc(x)
return y, x
def get_model(model_name,
net_params,
device,
n_hidden=1,
n_dim=300,
batchnorm=False,
dropout=False,
silu=False,
leaky_relu=False):
num_classes, channel, hidden = net_params
if model_name == 'dnn':
model = DNN(n_hidden=n_hidden, dim=n_dim)
elif model_name == 'lenet5':
model = LeNet5(channel=channel, hidden=hidden, num_classes=num_classes, bn=batchnorm, silu=silu,
leaky_relu=leaky_relu)
elif model_name == 'lenetzhu':
model = LeNetZhu(channel=channel, hidden=hidden, num_classes=num_classes)
elif model_name == 'vgg11':
model = VGG([1, 1, 2, 2, 2], hidden=hidden, num_classes=num_classes, dropout=dropout)
elif model_name == 'vgg13':
model = VGG([2, 2, 2, 2, 2], hidden=hidden, num_classes=num_classes, dropout=dropout)
elif model_name == 'vgg16':
model = VGG([2, 2, 3, 3, 3], hidden=hidden, num_classes=num_classes, dropout=dropout)
elif model_name == 'vgg19':
model = VGG([2, 2, 4, 4, 4], hidden=hidden, num_classes=num_classes, dropout=dropout)
elif model_name == 'resnet18':
model = ResNet(torchvision.models.resnet.BasicBlock, [2, 2, 2, 2],
num_classes=num_classes, base_width=64, channel=channel)
elif model_name == 'resnet34':
model = ResNet(torchvision.models.resnet.BasicBlock, [3, 4, 6, 3],
num_classes=num_classes, base_width=64, channel=channel)
elif model_name == 'resnet50':
model = ResNet(torchvision.models.resnet.Bottleneck, [3, 4, 6, 3],
num_classes=num_classes, base_width=64, channel=channel)
elif model_name == 'resnet101':
model = ResNet(torchvision.models.resnet.Bottleneck, [3, 4, 23, 3],
num_classes=num_classes, base_width=64, channel=channel)
elif model_name == 'resnet152':
model = ResNet(torchvision.models.resnet.Bottleneck, [3, 8, 36, 3],
num_classes=num_classes, base_width=64, channel=channel)
else:
raise NotImplementedError('Model not implemented.')
model = model.to(device)
return model