-
Notifications
You must be signed in to change notification settings - Fork 0
/
models.py
332 lines (279 loc) · 12.5 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
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
import torch
from torch import nn
import torch.nn.parallel
import torch.nn.functional as func
class ImageClassificationBase(nn.Module):
def training_step(self, batch):
if ( batch.shape[0] == 3 ):
images, labels, clabels = batch
images, clabels = images.to(device), clabels.to(device)
out = self(images) # Generate predictions
loss = func.cross_entropy(out, clabels) # Calculate loss
return loss
else:
images, labels = batch
out = self(images) # Generate predictions
loss = func.cross_entropy(out, labels) # Calculate loss
return loss
def validation_step(self, batch):
images, labels = batch
out = self(images) # Generate predictions
loss = func.cross_entropy(out, labels) # Calculate loss
acc = accuracy(out, labels) # Calculate accuracy
return {'val_loss': loss.detach(), 'val_acc': acc}
def validation_epoch_end(self, outputs):
batch_losses = [x['val_loss'] for x in outputs]
epoch_loss = torch.stack(batch_losses).mean() # Combine losses
batch_accs = [x['val_acc'] for x in outputs]
epoch_acc = torch.stack(batch_accs).mean() # Combine accuracies
return {'val_loss': epoch_loss.item(), 'val_acc': epoch_acc.item()}
def epoch_end(self, epoch, result):
print("Epoch [{}], train_loss: {:.4f}, val_loss: {:.4f}, val_acc: {:.4f}".format(
epoch, result['train_loss'], result['val_loss'], result['val_acc']))
class ImageClassificationBaseWithActivations(nn.Module):
def training_step(self, batch):
if ( len(batch) == 3 ):
images, labels, clabels = batch
images, clabels = images.to(device), clabels.to(device)
out, *_ = self(images) # Generate predictions
loss = func.cross_entropy(out, clabels) # Calculate loss
return loss
else:
images, labels = batch
out, *_ = self(images) # Generate predictions
loss = func.cross_entropy(out, labels) # Calculate loss
return loss
def validation_step(self, batch):
images, labels = batch
out,*_ = self(images) # Generate predictions
loss = func.cross_entropy(out, labels) # Calculate loss
acc = accuracy(out, labels) # Calculate accuracy
return {'val_loss': loss.detach(), 'val_acc': acc}
def validation_epoch_end(self, outputs):
batch_losses = [x['val_loss'] for x in outputs]
epoch_loss = torch.stack(batch_losses).mean() # Combine losses
batch_accs = [x['val_acc'] for x in outputs]
epoch_acc = torch.stack(batch_accs).mean() # Combine accuracies
return {'val_loss': epoch_loss.item(), 'val_acc': epoch_acc.item()}
def epoch_end(self, epoch, result):
print("Epoch [{}], train_loss: {:.4f}, val_loss: {:.4f}, val_acc: {:.4f}".format(
epoch, result['train_loss'], result['val_loss'], result['val_acc']))
def accuracy(outputs, labels):
_, preds = torch.max(outputs, dim=1)
return torch.tensor(torch.sum(preds == labels).item() / len(preds))
class Identity(nn.Module):
def __init__(self):
super(Identity, self).__init__()
def forward(self, x):
return x
class Flatten(nn.Module):
def __init__(self):
super(Flatten, self).__init__()
def forward(self,x):
return x.view(x.size(0), -1)
class Conv(nn.Sequential):
def __init__(self, in_channels, out_channels, kernel_size=3, stride=1, padding=None, output_padding=0,
activation_fn=nn.ReLU, batch_norm=True, transpose=False):
if padding is None:
padding = (kernel_size - 1) // 2
model = []
if not transpose:
# model += [ConvStandard(in_channels, out_channels, kernel_size=kernel_size, stride=stride, padding=padding
# )]
model += [nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, stride=stride, padding=padding,
bias=not batch_norm)]
else:
model += [nn.ConvTranspose2d(in_channels, out_channels, kernel_size, stride=stride, padding=padding,
output_padding=output_padding, bias=not batch_norm)]
if batch_norm:
model += [nn.BatchNorm2d(out_channels, affine=True)]
model += [activation_fn()]
super(Conv, self).__init__(*model)
class AllCNN(ImageClassificationBaseWithActivations):
def __init__(self, filters_percentage=1., n_channels=3, num_classes=10, dropout=False, batch_norm=True):
super(AllCNN, self).__init__()
n_filter1 = int(96 * filters_percentage)
n_filter2 = int(192 * filters_percentage)
self.conv1 = Conv(n_channels, n_filter1, kernel_size=3, batch_norm=batch_norm)
self.conv2 = Conv(n_filter1, n_filter1, kernel_size=3, batch_norm=batch_norm)
self.conv3 = Conv(n_filter1, n_filter2, kernel_size=3, stride=2, padding=1, batch_norm=batch_norm)
self.dropout1 = self.features = nn.Sequential(nn.Dropout(inplace=True) if dropout else Identity())
self.conv4 = Conv(n_filter2, n_filter2, kernel_size=3, stride=1, batch_norm=batch_norm)
self.conv5 = Conv(n_filter2, n_filter2, kernel_size=3, stride=1, batch_norm=batch_norm)
self.conv6 = Conv(n_filter2, n_filter2, kernel_size=3, stride=2, padding=1, batch_norm=batch_norm)
self.dropout2 = self.features = nn.Sequential(nn.Dropout(inplace=True) if dropout else Identity())
self.conv7 = Conv(n_filter2, n_filter2, kernel_size=3, stride=1, batch_norm=batch_norm)
self.conv8 = Conv(n_filter2, n_filter2, kernel_size=1, stride=1, batch_norm=batch_norm)
if n_channels == 3:
self.pool = nn.AvgPool2d(8)
elif n_channels == 1:
self.pool = nn.AvgPool2d(7)
self.flatten = Flatten()
self.classifier = nn.Sequential(
nn.Linear(n_filter2, num_classes),
)
def forward(self, x):
out = self.conv1(x)
actv1 = out
out = self.conv2(out)
actv2 = out
out = self.conv3(out)
actv3 = out
out = self.dropout1(out)
out = self.conv4(out)
actv4 = out
out = self.conv5(out)
actv5 = out
out = self.conv6(out)
actv6 = out
out = self.dropout2(out)
out = self.conv7(out)
actv7 = out
out = self.conv8(out)
actv8 = out
out = self.pool(out)
out = self.flatten(out)
out = self.classifier(out)
return out, actv1, actv2, actv3, actv4, actv5, actv6, actv7, actv8
cfg = {
'VGG11': [64, 'M', 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
'VGG13': [64, 64, 'M', 128, 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
'VGG16': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'M', 512, 512, 512, 'M', 512, 512, 512, 'M'],
'VGG19': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 256, 'M', 512, 512, 512, 512, 'M', 512, 512, 512, 512, 'M'],
}
class VGG(ImageClassificationBaseWithActivations):
def __init__(self, vgg_name, num_classes, num_channels = 3, return_activations = False):
super(VGG, self).__init__()
self.features = self._make_layers(cfg[vgg_name], channels = num_channels)
self.classifier = nn.Linear(512, num_classes)
self.return_activations = return_activations
def forward(self, x):
if not self.return_activations:
out = self.features(x)
out = out.view(out.size(0), -1)
out = self.classifier(out)
return out
activation_list = []
for layer in self.features:
x = layer(x)
if isinstance(layer, nn.Conv2d) and (x.numel() > 0):
activation_list.append(x)
x = x.view(x.size(0), -1)
x = self.classifier(x)
activation_list.append(x)
return x, activation_list
def _make_layers(self, cfg, channels=3):
layers = []
in_channels = channels
for x in cfg:
if x == 'M':
layers += [nn.MaxPool2d(kernel_size=2, stride=2)]
else:
layers += [nn.Conv2d(in_channels, x, kernel_size=3, padding=1),
nn.BatchNorm2d(x),
nn.ReLU(inplace=True)]
in_channels = x
layers += [nn.AvgPool2d(kernel_size=1, stride=1)]
return nn.Sequential(*layers)
def VGG16(num_classes = 10,num_channels = 3, return_activations = False):
return VGG('VGG16',num_classes=num_classes, return_activations=return_activations, num_channels=num_channels)
class LeNet(ImageClassificationBaseWithActivations):
def __init__(self, num_channels = 3, num_classes = 10):
super(LeNet, self).__init__()
self.conv1 = nn.Conv2d(num_channels, 6, kernel_size=5)
self.conv2 = nn.Conv2d(6, 16, kernel_size=5)
self.fc1 = nn.Linear(16*5*5, 120)
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, num_classes)
def forward(self, x):
x = self.conv1(x)
activ1 = x
x = func.relu(x)
x = func.max_pool2d(x, 2)
x = self.conv2(x)
activ2 = x
x = func.relu(x)
x = func.max_pool2d(x, 2)
x = x.view(x.size(0), -1)
x = self.fc1(x)
activ3 = x
x = func.relu(x)
x = self.fc2(x)
activ4 = x
x = func.relu(x)
x = self.fc3(x)
return x, activ1, activ2, activ3, activ4
# number of channels
nc=3
#checking the availability of cuda devices
device = 'cuda' if torch.cuda.is_available() else 'cpu'
# number of gpu's available
ngpu = 1
# input noise dimension
nz = 100
# number of generator filters
ngf = 64
#number of discriminator filters
ndf = 64
class Generator(nn.Module):
def __init__(self, ngpu, nc=3):
super(Generator, self).__init__()
self.ngpu = ngpu
self.main = nn.Sequential(
# input is Z, going into a convolution
nn.ConvTranspose2d(nz, ngf * 8, 4, 1, 0, bias=False),
nn.BatchNorm2d(ngf * 8),
nn.ReLU(True),
# state size. (ngf*8) x 4 x 4
nn.ConvTranspose2d(ngf * 8, ngf * 4, 4, 2, 1, bias=False),
nn.BatchNorm2d(ngf * 4),
nn.ReLU(True),
# state size. (ngf*4) x 8 x 8
nn.ConvTranspose2d(ngf * 4, ngf * 2, 4, 2, 1, bias=False),
nn.BatchNorm2d(ngf * 2),
nn.ReLU(True),
# state size. (ngf*2) x 16 x 16
nn.ConvTranspose2d(ngf * 2, ngf, 4, 2, 1, bias=False),
nn.BatchNorm2d(ngf),
nn.ReLU(True),
# state size. (ngf) x 32 x 32
nn.ConvTranspose2d(ngf, nc, 4, 2, 1, bias=False),
nn.Tanh()
# state size. (nc) x 64 x 64
)
def forward(self, input):
if input.is_cuda and self.ngpu > 1:
output = nn.parallel.data_parallel(self.main, input, range(self.ngpu))
else:
output = self.main(input)
return output
class Discriminator(nn.Module):
def __init__(self, ngpu):
super(Discriminator, self).__init__()
self.ngpu = ngpu
self.main = nn.Sequential(
# input is (nc) x 64 x 64
nn.Conv2d(nc, ndf, 4, 2, 1, bias=False),
nn.LeakyReLU(0.2, inplace=True),
# state size. (ndf) x 32 x 32
nn.Conv2d(ndf, ndf * 2, 4, 2, 1, bias=False),
nn.BatchNorm2d(ndf * 2),
nn.LeakyReLU(0.2, inplace=True),
# state size. (ndf*2) x 16 x 16
nn.Conv2d(ndf * 2, ndf * 4, 4, 2, 1, bias=False),
nn.BatchNorm2d(ndf * 4),
nn.LeakyReLU(0.2, inplace=True),
# state size. (ndf*4) x 8 x 8
nn.Conv2d(ndf * 4, ndf * 8, 4, 2, 1, bias=False),
nn.BatchNorm2d(ndf * 8),
nn.LeakyReLU(0.2, inplace=True),
# state size. (ndf*8) x 4 x 4
nn.Conv2d(ndf * 8, 1, 4, 1, 0, bias=False),
nn.Sigmoid()
)
def forward(self, input):
if input.is_cuda and self.ngpu > 1:
output = nn.parallel.data_parallel(self.main, input, range(self.ngpu))
else:
output = self.main(input)
return output.view(-1, 1).squeeze(1)