-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmodels.py
142 lines (104 loc) · 5.16 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
from __future__ import print_function
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.autograd import Variable
class VggCutted(nn.Module):
def __init__(self, vgg_model, number_cutted_layer):
super(VggCutted, self).__init__()
self.features = nn.Sequential(
*list(vgg_model.features.children())[:number_cutted_layer+1]
)
def forward(self, x):
x = self.features(x)
return x
class ResidualBlock(nn.Module):
def __init__(self, in_channels=64, out_channels=64, kernel_size=3, stride=1):
super(ResidualBlock, self).__init__()
self.conv1 = nn.Conv2d(in_channels=in_channels, out_channels=out_channels, kernel_size=kernel_size, stride=stride, padding=1)
self.bn1 = nn.BatchNorm2d(num_features=out_channels)
self.prelu = nn.PReLU()
self.conv2 = nn.Conv2d(in_channels=in_channels, out_channels=out_channels, kernel_size=kernel_size, stride=stride, padding=1)
self.bn2 = nn.BatchNorm2d(num_features=out_channels)
def forward(self, x):
tmp = self.prelu(self.bn1(self.conv1(x)))
return x + self.bn2(self.conv2(tmp))
class SecondGeneratorBlock(nn.Module):
def __init__(self, in_channels=64, out_channels=256, kernel_size=3, stride=1):
super(SecondGeneratorBlock, self).__init__()
self.conv = nn.Conv2d(in_channels=in_channels, out_channels=out_channels, kernel_size=kernel_size, stride=stride, padding=1)
self.ps = nn.PixelShuffle(2)
self.prelu = nn.PReLU()
def forward(self, x):
return self.prelu(self.ps(self.conv(x)))
class SRGanGenerator(nn.Module):
def __init__(self, residual_blocks_number, second_blocks_number):
super(SRGanGenerator, self).__init__()
self.rbn = residual_blocks_number
self.sbn = second_blocks_number
self.conv1 = nn.Conv2d(in_channels=3, out_channels=64, kernel_size=9, stride=1, padding=4)
self.pr = nn.PReLU()
for i in np.arange(0, residual_blocks_number):
name = 'residual_blocks_number_%03d' % (i)
self.add_module(name, ResidualBlock())
self.conv2 = nn.Conv2d(in_channels=64, out_channels=64, kernel_size=3, stride=1, padding=1)
self.bn = nn.BatchNorm2d(num_features=64)
for i in np.arange(0, second_blocks_number):
name = 'second_blocks_number_%03d' % (i)
self.add_module(name, SecondGeneratorBlock())
self.conv3 = nn.Conv2d(in_channels=64, out_channels=3, kernel_size=9, stride=1, padding=4)
def forward(self, x):
x = self.pr(self.conv1(x))
tmp = x.clone()
for i in np.arange(0, self.rbn):
name = 'residual_blocks_number_%03d' % (i)
tmp = self.__getattr__(name)(tmp)
x = self.bn(self.conv2(tmp)) + x
for i in np.arange(0, self.sbn):
name = 'second_blocks_number_%03d' % (i)
x = self.__getattr__(name)(x)
return self.conv3(x)
class SRGanDiscriminator(nn.Module):
def __init__(self):
super(SRGanDiscriminator, self).__init__()
self.conv = nn.Conv2d(in_channels=3, out_channels=64, kernel_size=3, stride=1, padding=1)
self.lr = nn.LeakyReLU(0.2)
self.conv1 = nn.Conv2d(in_channels=64, out_channels=64, kernel_size=3, stride=2, padding=1)
self.bn1 = nn.BatchNorm2d(num_features=64)
self.lr1 = nn.LeakyReLU(0.2)
self.conv2 = nn.Conv2d(in_channels=64, out_channels=128, kernel_size=3, stride=1, padding=1)
self.bn2 = nn.BatchNorm2d(num_features=128)
self.lr2 = nn.LeakyReLU(0.2)
self.conv3 = nn.Conv2d(in_channels=128, out_channels=128, kernel_size=3, stride=2, padding=1)
self.bn3 = nn.BatchNorm2d(num_features=128)
self.lr3 = nn.LeakyReLU(0.2)
self.conv4 = nn.Conv2d(in_channels=128, out_channels=256, kernel_size=3, stride=1, padding=1)
self.bn4 = nn.BatchNorm2d(num_features=256)
self.lr4 = nn.LeakyReLU(0.2)
self.conv5 = nn.Conv2d(in_channels=256, out_channels=256, kernel_size=3, stride=2, padding=1)
self.bn5 = nn.BatchNorm2d(num_features=256)
self.lr5 = nn.LeakyReLU(0.2)
self.conv6 = nn.Conv2d(in_channels=256, out_channels=512, kernel_size=3, stride=1, padding=1)
self.bn6 = nn.BatchNorm2d(num_features=512)
self.lr6 = nn.LeakyReLU(0.2)
self.conv7 = nn.Conv2d(in_channels=512, out_channels=512, kernel_size=3, stride=2, padding=1)
self.bn7 = nn.BatchNorm2d(num_features=512)
self.lr7 = nn.LeakyReLU(0.2)
self.dense1 = nn.Linear(21*8*512, 1024)
self.lr8 = nn.LeakyReLU(0.2)
self.dense2 = nn.Linear(1024, 1)
self.sig = nn.Sigmoid()
def forward(self, x):
y = self.lr(self.conv(x))
y = self.lr1(self.conv1(y))
y = self.lr2(self.conv2(y))
y = self.lr3(self.conv3(y))
y = self.lr4(self.conv4(y))
y = self.lr5(self.conv5(y))
y = self.lr6(self.conv6(y))
y = self.lr7(self.conv7(y))
y = y.view(y.size(0), -1)
y = self.lr8(self.dense1(y))
output = self.sig(self.dense2(y))
return output