### PyTorch CNN MNIST 手写数字识别实现 以下是基于PyTorch框架实现卷积神经网络（CNN）来完成MNIST手写数字识别的完整代码示例： #### 1. 导入库 导入必要的库，包括PyTorch及其子模块`nn`和`optim`，以及用于数据可视化的`matplotlib`。 ```python import torch import torch.nn as nn import torch.optim as optim import torchvision from torchvision import datasets, transforms import matplotlib.pyplot as plt ``` #### 2. 配置设备 设置运行环境，默认优先使用GPU加速计算；如果不可用，则回退到CPU[^1]。 ```python device = torch.device("cuda" if torch.cuda.is_available() else "cpu") print(f"Using device: {device}") ``` #### 3. 数据预处理与加载 定义数据变换操作，并通过`DataLoader`加载MNIST数据集。这里我们将图片转换为张量形式并进行标准化处理[^4]。 ```python batch_size = 64 # 定义数据变换 transform = transforms.Compose([ transforms.ToTensor(), # 将图像转为Tensor transforms.Normalize((0.5,), (0.5,)) # 归一化 ]) # 加载训练集和测试集 train_dataset = datasets.MNIST(root='./data', train=True, download=True, transform=transform) test_dataset = datasets.MNIST(root='./data', train=False, download=True, transform=transform) train_loader = torch.utils.data.DataLoader(dataset=train_dataset, batch_size=batch_size, shuffle=True) test_loader = torch.utils.data.DataLoader(dataset=test_dataset, batch_size=batch_size, shuffle=False) ``` #### 4. 构建CNN模型 设计一个简单的两层卷积神经网络结构，包含两个卷积层、池化层以及全连接层[^1]。 ```python class CNN(nn.Module): def __init__(self): super(CNN, self).__init__() self.conv1 = nn.Conv2d(1, 32, kernel_size=3, stride=1, padding=1) # 输入通道数为1，输出通道数为32 self.pool = nn.MaxPool2d(kernel_size=2, stride=2) # 最大池化层 self.conv2 = nn.Conv2d(32, 64, kernel_size=3, stride=1, padding=1) # 输出通道数增加至64 self.fc1 = nn.Linear(64 * 7 * 7, 128) # 展平后的输入大小为64*7*7 self.fc2 = nn.Linear(128, 10) # 输出类别数量为10 def forward(self, x): x = self.pool(torch.relu(self.conv1(x))) # 卷积 -> ReLU激活 -> 池化 x = self.pool(torch.relu(self.conv2(x))) x = x.view(-1, 64 * 7 * 7) # 展平 x = torch.relu(self.fc1(x)) x = self.fc2(x) return x model = CNN().to(device) ``` #### 5. 定义损失函数与优化器 采用交叉熵作为损失函数，并选用随机梯度下降法(SGD)作为优化算法。 ```python criterion = nn.CrossEntropyLoss() optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9) ``` #### 6. 训练过程 迭代多个epoch，在每次循环中更新权重参数以最小化误差[^1]。 ```python num_epochs = 10 for epoch in range(num_epochs): model.train() running_loss = 0.0 correct = 0 total = 0 for images, labels in train_loader: images, labels = images.to(device), labels.to(device) optimizer.zero_grad() outputs = model(images) loss = criterion(outputs, labels) loss.backward() optimizer.step() _, predicted = torch.max(outputs.data, 1) total += labels.size(0) correct += (predicted == labels).sum().item() running_loss += loss.item() print(f'Epoch [{epoch+1}/{num_epochs}], Loss: {running_loss/len(train_loader):.4f}, Accuracy: {100 * correct / total:.2f}%') ``` #### 7. 测试评估 利用测试集合验证模型性能，统计分类准确率[^1]。 ```python model.eval() correct = 0 total = 0 with torch.no_grad(): for images, labels in test_loader: images, labels = images.to(device), labels.to(device) outputs = model(images) _, predicted = torch.max(outputs.data, 1) total += labels.size(0) correct += (predicted == labels).sum().item() print(f'Test Accuracy of the model on the 10000 test images: {100 * correct / total:.2f} %') ``` --- ###