提供一段Inception-BN的pytorch代码

以下是使用PyTorch构建BN-Inception网络的代码示例：

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

class InceptionBN(nn.Module):
    def __init__(self, in_channels, out_1x1, reduce_3x3, out_3x3, reduce_5x5, out_5x5, pool_proj):
        super(InceptionBN, self).__init__()
        
        # 1x1 conv branch
        self.branch1 = nn.Sequential(
            nn.Conv2d(in_channels, out_1x1, kernel_size=1),
            nn.BatchNorm2d(out_1x1),
            nn.ReLU(inplace=True)
        )
        
        # 3x3 conv branch
        self.branch2 = nn.Sequential(
            nn.Conv2d(in_channels, reduce_3x3, kernel_size=1),
            nn.BatchNorm2d(reduce_3x3),
            nn.ReLU(inplace=True),
            nn.Conv2d(reduce_3x3, out_3x3, kernel_size=3, padding=1),
            nn.BatchNorm2d(out_3x3),
            nn.ReLU(inplace=True)
        )
        
        # 5x5 conv branch
        self.branch3 = nn.Sequential(
            nn.Conv2d(in_channels, reduce_5x5, kernel_size=1),
            nn.BatchNorm2d(reduce_5x5),
            nn.ReLU(inplace=True),
            nn.Conv2d(reduce_5x5, out_5x5, kernel_size=5, padding=2),
            nn.BatchNorm2d(out_5x5),
            nn.ReLU(inplace=True)
        )
        
        # pooling branch
        self.branch4 = nn.Sequential(
            nn.MaxPool2d(kernel_size=3, stride=1, padding=1),
            nn.Conv2d(in_channels, pool_proj, kernel_size=1),
            nn.BatchNorm2d(pool_proj),
            nn.ReLU(inplace=True)
        )
        
    def forward(self, x):
        out1 = self.branch1(x)
        out2 = self.branch2(x)
        out3 = self.branch3(x)
        out4 = self.branch4(x)
        output = torch.cat([out1, out2, out3, out4], dim=1)
        return output
        
class BNInception(nn.Module):
    def __init__(self, num_classes=1000):
        super(BNInception, self).__init__()
        
        self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3)
        self.bn1 = nn.BatchNorm2d(64)
        self.pool1 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        
        self.conv2_1 = nn.Conv2d(64, 64, kernel_size=1)
        self.bn2_1 = nn.BatchNorm2d(64)
        self.conv2_2 = nn.Conv2d(64, 192, kernel_size=3, padding=1)
        self.bn2_2 = nn.BatchNorm2d(192)
        self.pool2 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        
        self.inception3a = InceptionBN(192, 64, 96, 128, 16, 32, 32)
        self.inception3b = InceptionBN(256, 128, 128, 192, 32, 96, 64)
        self.pool3 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        
        self.inception4a = InceptionBN(480, 192, 96, 208, 16, 48, 64)
        self.inception4b = InceptionBN(512, 160, 112, 224, 24, 64, 64)
        self.inception4c = InceptionBN(512, 128, 128, 256, 24, 64, 64)
        self.inception4d = InceptionBN(512, 112, 144, 288, 32, 64, 64)
        self.inception4e = InceptionBN(528, 256, 160, 320, 32, 128, 128)
        self.pool4 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        
        self.inception5a = InceptionBN(832, 256, 160, 320, 32, 128, 128)
        self.inception5b = InceptionBN(832, 384, 192, 384, 48, 128, 128)
        self.pool5 = nn.AdaptiveAvgPool2d((1, 1))
        
        self.fc = nn.Linear(1024, num_classes)
        
    def forward(self, x):
        x = self.conv1(x)
        x = self.bn1(x)
        x = F.relu(x, inplace=True)
        x = self.pool1(x)
        
        x = self.conv2_1(x)
        x = self.bn2_1(x)
        x = F.relu(x, inplace=True)
        x = self.conv2_2(x)
        x = self.bn2_2(x)
        x = F.relu(x, inplace=True)
        x = self.pool2(x)
        
        x = self.inception3a(x)
        x = self.inception3b(x)
        x = self.pool3(x)
        
        x = self.inception4a(x)
        x = self.inception4b(x)
        x = self.inception4c(x)
        x = self.inception4d(x)
        x = self.inception4e(x)
        x = self.pool4(x)
        
        x = self.inception5a(x)
        x = self.inception5b(x)
        x = self.pool5(x)
        
        x = x.view(x.size(0), -1)
        x = self.fc(x)
        
        return x

这段代码实现了BN-Inception网络，包括Inception模块和BN层的定义，以及整个网络的组合和前向传播过程。