用resnet18对四种天气分类，自定义归一化方法，采用均值方差归一化方法

好的，以下是使用ResNet18对四种天气进行分类并自定义归一化方法的代码：

import torch
import torch.nn as nn
import torchvision.models as models

# 自定义均值方差归一化方法
class MyNormalize(nn.Module):
    def __init__(self, mean, std):
        super(MyNormalize, self).__init__()
        self.mean = torch.tensor(mean).reshape(1, 3, 1, 1)
        self.std = torch.tensor(std).reshape(1, 3, 1, 1)

    def forward(self, x):
        return (x - self.mean) / self.std


# 加载ResNet18模型
model = models.resnet18(pretrained=True)

# 冻结模型所有参数
for param in model.parameters():
    param.requires_grad = False

# 替换最后一层分类器
num_ftrs = model.fc.in_features
model.fc = nn.Linear(num_ftrs, 4)

# 自定义均值方差归一化方法
normalize = MyNormalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])

# 定义损失函数和优化器
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(model.fc.parameters(), lr=0.001, momentum=0.9)

# 训练模型
num_epochs = 10
for epoch in range(num_epochs):
    for images, labels in train_loader:
        images = normalize(images)  # 归一化
        outputs = model(images)
        loss = criterion(outputs, labels)
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

    # 计算在验证集上的准确率
    correct = 0
    total = 0
    with torch.no_grad():
        for images, labels in val_loader:
            images = normalize(images)  # 归一化
            outputs = model(images)
            _, predicted = torch.max(outputs.data, 1)
            total += labels.size(0)
            correct += (predicted == labels).sum().item()

    accuracy = 100 * correct / total
    print('Epoch [{}/{}], Accuracy: {:.2f}%'.format(epoch+1, num_epochs, accuracy))

其中，`MyNormalize`是自定义的归一化方法，其输入为一个张量，输出为将每个像素减去均值并除以方差后的结果。在模型训练和验证时，需要先将数据归一化后再输入模型中。