File size: 5,371 Bytes
5adfc21
 
 
dc89944
5adfc21
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dc89944
5adfc21
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
"""
Implementation of YOLOv3 architecture
"""
from pytorch_lightning import LightningModule
import torch
import torch.nn as nn

""" 
Information about architecture config:
Tuple is structured by (filters, kernel_size, stride) 
Every conv is a same convolution. 
List is structured by "B" indicating a residual block followed by the number of repeats
"S" is for scale prediction block and computing the yolo loss
"U" is for upsampling the feature map and concatenating with a previous layer
"""
config = [
    (32, 3, 1),
    (64, 3, 2),
    ["B", 1],
    (128, 3, 2),
    ["B", 2],
    (256, 3, 2),
    ["B", 8],
    (512, 3, 2),
    ["B", 8],
    (1024, 3, 2),
    ["B", 4],  # To this point is Darknet-53
    (512, 1, 1),
    (1024, 3, 1),
    "S",
    (256, 1, 1),
    "U",
    (256, 1, 1),
    (512, 3, 1),
    "S",
    (128, 1, 1),
    "U",
    (128, 1, 1),
    (256, 3, 1),
    "S",
]


class CNNBlock(nn.Module):
    def __init__(self, in_channels, out_channels, bn_act=True, **kwargs):
        super().__init__()
        self.conv = nn.Conv2d(in_channels, out_channels, bias=not bn_act, **kwargs)
        self.bn = nn.BatchNorm2d(out_channels)
        self.leaky = nn.LeakyReLU(0.1)
        self.use_bn_act = bn_act

    def forward(self, x):
        if self.use_bn_act:
            return self.leaky(self.bn(self.conv(x)))
        else:
            return self.conv(x)


class ResidualBlock(nn.Module):
    def __init__(self, channels, use_residual=True, num_repeats=1):
        super().__init__()
        self.layers = nn.ModuleList()
        for repeat in range(num_repeats):
            self.layers += [
                nn.Sequential(
                    CNNBlock(channels, channels // 2, kernel_size=1),
                    CNNBlock(channels // 2, channels, kernel_size=3, padding=1),
                )
            ]

        self.use_residual = use_residual
        self.num_repeats = num_repeats

    def forward(self, x):
        for layer in self.layers:
            if self.use_residual:
                x = x + layer(x)
            else:
                x = layer(x)

        return x


class ScalePrediction(nn.Module):
    def __init__(self, in_channels, num_classes):
        super().__init__()
        self.pred = nn.Sequential(
            CNNBlock(in_channels, 2 * in_channels, kernel_size=3, padding=1),
            CNNBlock(
                2 * in_channels, (num_classes + 5) * 3, bn_act=False, kernel_size=1
            ),
        )
        self.num_classes = num_classes

    def forward(self, x):
        return (
            self.pred(x)
            .reshape(x.shape[0], 3, self.num_classes + 5, x.shape[2], x.shape[3])
            .permute(0, 1, 3, 4, 2)
        )


class YOLOv3(LightningModule):
    def __init__(self, in_channels=3, num_classes=80):
        super().__init__()
        self.num_classes = num_classes
        self.in_channels = in_channels
        self.layers = self._create_conv_layers()

    def forward(self, x):
        outputs = []  # for each scale
        route_connections = []
        for layer in self.layers:
            if isinstance(layer, ScalePrediction):
                outputs.append(layer(x))
                continue

            x = layer(x)

            if isinstance(layer, ResidualBlock) and layer.num_repeats == 8:
                route_connections.append(x)

            elif isinstance(layer, nn.Upsample):
                x = torch.cat([x, route_connections[-1]], dim=1)
                route_connections.pop()

        return outputs

    def _create_conv_layers(self):
        layers = nn.ModuleList()
        in_channels = self.in_channels

        for module in config:
            if isinstance(module, tuple):
                out_channels, kernel_size, stride = module
                layers.append(
                    CNNBlock(
                        in_channels,
                        out_channels,
                        kernel_size=kernel_size,
                        stride=stride,
                        padding=1 if kernel_size == 3 else 0,
                    )
                )
                in_channels = out_channels

            elif isinstance(module, list):
                num_repeats = module[1]
                layers.append(ResidualBlock(in_channels, num_repeats=num_repeats,))

            elif isinstance(module, str):
                if module == "S":
                    layers += [
                        ResidualBlock(in_channels, use_residual=False, num_repeats=1),
                        CNNBlock(in_channels, in_channels // 2, kernel_size=1),
                        ScalePrediction(in_channels // 2, num_classes=self.num_classes),
                    ]
                    in_channels = in_channels // 2

                elif module == "U":
                    layers.append(nn.Upsample(scale_factor=2),)
                    in_channels = in_channels * 3

        return layers


if __name__ == "__main__":
    num_classes = 20
    IMAGE_SIZE = 416
    model = YOLOv3(num_classes=num_classes)
    x = torch.randn((2, 3, IMAGE_SIZE, IMAGE_SIZE))
    out = model(x)
    assert model(x)[0].shape == (2, 3, IMAGE_SIZE//32, IMAGE_SIZE//32, num_classes + 5)
    assert model(x)[1].shape == (2, 3, IMAGE_SIZE//16, IMAGE_SIZE//16, num_classes + 5)
    assert model(x)[2].shape == (2, 3, IMAGE_SIZE//8, IMAGE_SIZE//8, num_classes + 5)
    print("Success!")