Projet-App-Sup/layers.py at main · Etienne-bdt/Projet-App-Sup · GitHub

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
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F


class Convolutional_Block(nn.Module):
    def __init__(self, in_channels, out_channels):
        super(Convolutional_Block, self).__init__()

        #on ne fait pas de rétro action pour l' instant!

        self.conv1 = nn.Conv2d(in_channels, out_channels,
                               kernel_size = (3,3), padding = "same")
        self.batch_norm1 = nn.BatchNorm2d(out_channels)
        self.relu1 = nn.ReLU()

        self.conv2 = nn.Conv2d(out_channels, out_channels,
                               kernel_size = (3,3), padding = "same")
        self.batch_norm2 = nn.BatchNorm2d(out_channels)
        self.relu2 = nn.ReLU()

    def forward(self,x):

        conv_1 = self.conv1(x)
        batch_norm_1 = self.batch_norm1(conv_1)
        relu_1 = self.relu1(batch_norm_1)

        conv_2 = self.conv2(relu_1)
        batch_norm_2 = self.batch_norm2(conv_2)

        output = self.relu2(batch_norm_2)

        return output

class DeConvolutional_Block(nn.Module):
    def __init__(self, in_channels, out_channels):
        super(DeConvolutional_Block, self).__init__()


        self.conv1 = nn.Conv2d(in_channels, out_channels,
                               kernel_size = (3,3), padding = "same")
        self.batch_norm1 = nn.BatchNorm2d(out_channels)
        self.relu1 = nn.ReLU()

        self.conv2 = nn.Conv2d(out_channels, out_channels,
                               kernel_size = (3,3), padding = "same")
        self.batch_norm2 = nn.BatchNorm2d(out_channels)
        self.relu2 = nn.ReLU()

    def forward(self,x):

        conv_1 = self.conv1(x)
        batch_norm_1 = self.batch_norm1(conv_1)
        relu_1 = self.relu1(batch_norm_1)

        conv_2 = self.conv2(relu_1)
        batch_norm_2 = self.batch_norm2(conv_2)

        output = self.relu2(batch_norm_2)

        return output

class Conv(nn.Module):
    def __init__(self, in_cn, out_cn):
        super().__init__()
        self.conv = nn.Conv2d(in_cn, out_cn, kernel_size=3, padding=1)
        self.bn = nn.BatchNorm2d(out_cn)
        self.relu = nn.ReLU()
        self.dropout = nn.Dropout2d(p=0.2)

    def forward(self, x):
        x = self.conv(x)
        x = self.bn(x)
        x = self.relu(x)
        x = self.dropout(x)
        return x

class DeConv(nn.Module):
    def __init__(self, in_cn, out_cn):
        super().__init__()
        self.deconv = nn.ConvTranspose2d(in_cn, out_cn, kernel_size=2, stride=2)
        self.bn = nn.BatchNorm2d(out_cn)
        self.relu = nn.ReLU()
        self.dropout = nn.Dropout2d(p=0.2)

    def forward(self, x):
        x = self.deconv(x)
        x = self.bn(x)
        x = self.relu(x)
        x = self.dropout(x)
        return x