Pytorch – GoogLeNet の仕組みと実装について解説

import torch
import torch.nn as nn
import torch.nn.functional as F


class BasicConv2d(nn.Module):
    def __init__(self, in_channels, out_channels, **kwargs):
        super().__init__()
        self.conv = nn.Conv2d(in_channels, out_channels, bias=False, **kwargs)
        self.bn = nn.BatchNorm2d(out_channels, eps=0.001)
        self.relu = nn.ReLU(True)

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

        return x

class InceptionAux(nn.Module):
    def __init__(
        self,
        in_channels,
        num_classes,
        dropout,
    ):
        super().__init__()
        self.pool = nn.AdaptiveAvgPool2d(x, (4, 4))
        self.conv = BasicConv2d(in_channels, 128, kernel_size=1)
        self.fc1 = nn.Linear(2048, 1024)
        self.fc2 = nn.Linear(1024, num_classes)
        self.dropout = nn.Dropout(p=dropout)
        self.relu = nn.ReLU(True)

    def forward(self, x):
        x = F.adaptive_avg_pool2d(x, (4, 4))  # (N, 512 or 528, 4, 4)
        x = self.conv(x)  # (N, 128, 4, 4)
        x = torch.flatten(x, 1)  # (N, 128 * 4 * 4)
        x = self.fc1(x)  # (N, 1024)
        x = self.relu(x)
        x = self.dropout(x)
        x = self.fc2(x)  # (N, num_classes)

        return x

class Inception(nn.Module):
    def __init__(
        self,
        in_channels,
        ch1x1,
        ch3x3red,
        ch3x3,
        ch5x5red,
        ch5x5,
        pool_proj,
    ):
        super().__init__()
        self.branch1 = BasicConv2d(in_channels, ch1x1, kernel_size=1)
        self.branch2 = nn.Sequential(
            BasicConv2d(in_channels, ch3x3red, kernel_size=1),
            BasicConv2d(ch3x3red, ch3x3, kernel_size=3, padding=1),
        )
        self.branch3 = nn.Sequential(
            BasicConv2d(in_channels, ch5x5red, kernel_size=1),
            BasicConv2d(ch5x5red, ch5x5, kernel_size=5, padding=2),
        )
        self.branch4 = nn.Sequential(
            nn.MaxPool2d(kernel_size=3, stride=1, padding=1, ceil_mode=True),
            BasicConv2d(in_channels, pool_proj, kernel_size=1),
        )

    def forward(self, x):
        branch1 = self.branch1(x)
        branch2 = self.branch2(x)
        branch3 = self.branch3(x)
        branch4 = self.branch4(x)

        out = torch.cat([branch1, branch2, branch3, branch4], 1)

        return out

class GoogLeNet(nn.Module):
    def __init__(
        self,
        num_classes=1000,
        aux_logits=True,
        dropout=0.4,
        dropout_aux=0.7,
    ):
        super().__init__()
        self.aux_logits = aux_logits

        self.conv1 = BasicConv2d(3, 64, kernel_size=7, stride=2, padding=3)
        self.maxpool1 = nn.MaxPool2d(3, stride=2, ceil_mode=True)
        self.conv2 = BasicConv2d(64, 64, kernel_size=1)
        self.conv3 = BasicConv2d(64, 192, kernel_size=3, padding=1)
        self.maxpool2 = nn.MaxPool2d(3, stride=2, ceil_mode=True)

        self.inception3a = Inception(192, 64, 96, 128, 16, 32, 32)
        self.inception3b = Inception(256, 128, 128, 192, 32, 96, 64)
        self.maxpool3 = nn.MaxPool2d(3, stride=2, ceil_mode=True)

        self.inception4a = Inception(480, 192, 96, 208, 16, 48, 64)
        self.inception4b = Inception(512, 160, 112, 224, 24, 64, 64)
        self.inception4c = Inception(512, 128, 128, 256, 24, 64, 64)
        self.inception4d = Inception(512, 112, 144, 288, 32, 64, 64)
        self.inception4e = Inception(528, 256, 160, 320, 32, 128, 128)
        self.maxpool4 = nn.MaxPool2d(3, stride=2, ceil_mode=True)

        self.inception5a = Inception(832, 256, 160, 320, 32, 128, 128)
        self.inception5b = Inception(832, 384, 192, 384, 48, 128, 128)

        if aux_logits:
            self.aux1 = InceptionAux(512, num_classes, dropout=dropout_aux)
            self.aux2 = InceptionAux(528, num_classes, dropout=dropout_aux)
        else:
            self.aux1 = None
            self.aux2 = None

        self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
        self.dropout = nn.Dropout(p=dropout)
        self.fc = nn.Linear(1024, num_classes)
        self._initialize_weights()

    def _initialize_weights(self):
        for m in self.modules():
            if isinstance(m, nn.Linear):
                nn.init.kaiming_uniform_(m.weight, nonlinearity="linear")
                if m.out_features == 1000:
                    nn.init.zeros_(m.bias)  # 出力層は0で初期化する
                else:
                    nn.init.constant_(m.bias, 0.2)
            if isinstance(m, nn.Conv2d) or isinstance(m, nn.Linear):
                nn.init.kaiming_uniform_(m.weight, nonlinearity="conv2d")

    def forward(self, x):
        x = self.conv1(x)  # (N, 64, 112, 112)
        x = self.maxpool1(x)  # (N, 64, 56, 56)
        x = self.conv2(x)  # (N, 64, 56, 56)
        x = self.conv3(x)  # (N, 192, 56, 56)
        x = self.maxpool2(x)  # (N, 192, 28, 28)
        x = self.inception3a(x)  # (N, 256, 28, 28)
        x = self.inception3b(x)  # (N, 480, 28, 28)
        x = self.maxpool3(x)  # (N, 480, 14, 14)
        x = self.inception4a(x)  # (N, 512, 14, 14)

        aux1 = self.aux1(x) if self.aux_logits and self.training else None

        x = self.inception4b(x)  # (N, 512, 14, 14)
        x = self.inception4c(x)  # (N, 512, 14, 14)
        x = self.inception4d(x)  # (N, 528, 14, 14)

        aux2 = self.aux2(x) if self.aux_logits and self.training else None

        x = self.inception4e(x)  # (N, 832, 14, 14)
        x = self.maxpool4(x)  # (N, 832, 7, 7)

        x = self.inception5a(x)  # (N, 832, 7, 7)
        x = self.inception5b(x)  # (N, 1024, 7, 7)

        x = self.avgpool(x)  # (N, 1024, 1, 1)
        x = torch.flatten(x, 1)  # (N, 1024)
        x = self.dropout(x)
        x = self.fc(x)  # (N, 1000)

        if self.aux_logits and self.training:
            return x, aux2, aux1
        else:
            return x


def googlenet():
    return GoogLeNet()