File size: 3,844 Bytes
984cdba
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import torch
import torch.nn as nn
import torch.nn.functional as F
import torchvision.models as models
# -------------------------------------------------
# Basic blocks
# -------------------------------------------------

class DoubleConv(nn.Module):
    """(Conv => BN => ReLU) * 2"""
    def __init__(self, in_ch, out_ch):
        super().__init__()
        self.net = nn.Sequential(
            nn.Conv2d(in_ch, out_ch, 3, padding=1),
            nn.BatchNorm2d(out_ch),
            nn.ReLU(inplace=True),

            nn.Conv2d(out_ch, out_ch, 3, padding=1),
            nn.BatchNorm2d(out_ch),
            nn.ReLU(inplace=True),
        )

    def forward(self, x):
        return self.net(x)

class Down(nn.Module):
    """Downscaling with maxpool then double conv"""
    def __init__(self, in_ch, out_ch):
        super().__init__()
        self.net = nn.Sequential(
            nn.MaxPool2d(2),
            DoubleConv(in_ch, out_ch)
        )

    def forward(self, x):
        return self.net(x)

class Up(nn.Module):
    """Upscaling then double conv"""
    def __init__(self, in_ch, out_ch):
        super().__init__()

        # bilinear upsample + conv (no checkerboard artifacts)
        self.up = nn.Upsample(scale_factor=2, mode='bilinear', align_corners=False)
        self.conv = DoubleConv(in_ch, out_ch)

    def forward(self, x, skip):
        x = self.up(x)

        if skip is not None:

            # pad if needed (for odd input sizes)
            diffY = skip.size(2) - x.size(2)
            diffX = skip.size(3) - x.size(3)

            x = F.pad(x, [
                diffX // 2, diffX - diffX // 2,
                diffY // 2, diffY - diffY // 2
            ])

            x = torch.cat([skip, x], dim=1)
        return self.conv(x)

class OutConv(nn.Module):
    def __init__(self, in_ch, num_classes):
        super().__init__()
        self.conv = nn.Conv2d(in_ch, num_classes, 1)

    def forward(self, x):
        return self.conv(x)

class ResNetEncoder(nn.Module):
    def __init__(self, in_channels=3, pretrained=True):
        super().__init__()

        resnet = models.resnet34(weights="IMAGENET1K_V1" if pretrained else None)

        if in_channels != 3:
            resnet.conv1 = nn.Conv2d(
                in_channels, 64, kernel_size=7, stride=2, padding=3, bias=False
            )

        self.initial = nn.Sequential(
            resnet.conv1,
            resnet.bn1,
            resnet.relu
        )                   # 128x128
        self.maxpool = resnet.maxpool  # 64x64

        self.layer1 = resnet.layer1    # 64x64
        self.layer2 = resnet.layer2    # 32x32
        self.layer3 = resnet.layer3    # 16x16
        self.layer4 = resnet.layer4    # 8x8

    def forward(self, x):
        x1 = self.initial(x)
        x2 = self.layer1(self.maxpool(x1))
        x3 = self.layer2(x2)
        x4 = self.layer3(x3)
        x5 = self.layer4(x4)
        return x1, x2, x3, x4, x5


# -------------------------------------------------
# Standard U-Net
# -------------------------------------------------

class model(nn.Module):
    def __init__(self, in_channels=3, num_classes=1, freeze_encoder=False):
        super().__init__()

        self.encoder = ResNetEncoder(in_channels, pretrained=True)

        if freeze_encoder:
            for p in self.encoder.parameters():
                p.requires_grad = False

        self.up1 = Up(512 + 256, 256)
        self.up2 = Up(256 + 128, 128)
        self.up3 = Up(128 + 64,  64)
        self.up4 = Up(64  + 64,  64)
        self.up5 = Up(64, 64)
        self.outc = OutConv(64, num_classes)

    def forward(self, x):
        x1, x2, x3, x4, x5 = self.encoder(x)

        x = self.up1(x5, x4)
        x = self.up2(x,  x3)
        x = self.up3(x,  x2)
        x = self.up4(x,  x1)
        x = self.up5(x,  None) 
        return self.outc(x)