基于ResNet18残差网络完成图像分类代码

好的，以下是基于PyTorch搭建ResNet18模型完成图像分类的代码示例：

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

# 加载ResNet18模型
resnet18 = models.resnet18(pretrained=False)

# 替换最后一层全连接层
num_features = resnet18.fc.in_features
resnet18.fc = nn.Linear(num_features, num_classes)

# 定义损失函数和优化器
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(resnet18.parameters(), lr=learning_rate, momentum=0.9)

# 训练模型
for epoch in range(num_epochs):
    for images, labels in train_loader:
        # 将数据加载到GPU上
        images, labels = images.to(device), labels.to(device)
        
        # 前向传播
        outputs = resnet18(images)
        loss = criterion(outputs, labels)
        
        # 反向传播和优化
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()
        
    # 每个epoch结束后计算模型在验证集上的准确率
    with torch.no_grad():
        correct = 0
        total = 0
        for images, labels in val_loader:
            images, labels = images.to(device), labels.to(device)
            outputs = resnet18(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))

在上述代码中，我们首先加载了ResNet18模型，并替换了最后一层全连接层以适应我们的分类任务。然后定义了损失函数和优化器，并在每个epoch结束后计算模型在验证集上的准确率。在训练过程中，我们对训练集进行迭代，每次迭代都进行前向传播、反向传播和优化操作，直至达到预设的epoch数。