注：书中对代码的讲解并不详细，本文对很多细节做了详细注释。另外，书上的源代码是在Jupyter Notebook上运行的，较为分散，本文将代码集中起来，并加以完善，全部用vscode在python 3.9.18下测试通过，同时对于书上部分章节也做了整合。

Chapter6 Convolutional Neural Network(CNN)

6.3 LeNet

LeNet模型中每个卷积块中的基本单元是一个卷积层、一个sigmoid激活函数和平均汇聚层。虽然ReLU和最大汇聚层更有效，但它们在20世纪90年代还没有出现。每个卷积层使用$5\times 5$卷积核和一个sigmoid激活函数，这些层将输入映射到多个二维特征输出，通常同时增加通道的数量。第一卷积层有6个输出通道，而第二个卷积层有16个输出通道，每个$2\times 2$池操作（步幅2）通过空间下采样将维数减少4倍。为了将卷积块的输出传递给稠密块，我们必须在小批量中展平每个样本。换言之，我们将这个四维输入转换成全连接层所期望的二维输入（第一个维度索引小批量中的样本，第二个维度给出每个样本的平面向量表示）。LeNet的稠密块有三个全连接层，分别有120、84和10个输出。因为我们在执行分类任务，所以输出层的10维对应于最后输出结果的数量。

import torch
from torch import nn
from d2l import torch as d2l
import matplotlib.pyplot as pltnet = nn.Sequential(nn.Conv2d(1, 6, kernel_size=5, padding=2), nn.Sigmoid(),nn.AvgPool2d(kernel_size=2, stride=2),nn.Conv2d(6, 16, kernel_size=5), nn.Sigmoid(),nn.AvgPool2d(kernel_size=2, stride=2),nn.Flatten(),nn.Linear(16 * 5 * 5, 120), nn.Sigmoid(),nn.Linear(120, 84), nn.Sigmoid(),nn.Linear(84, 10))X = torch.rand(size=(1, 1, 28, 28), dtype=torch.float32)
for layer in net:X = layer(X)print(layer.__class__.__name__,'output shape: \t',X.shape)batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size=batch_size)def evaluate_accuracy_gpu(net, data_iter, device=None): #@save"""使用GPU计算模型在数据集上的精度"""if isinstance(net, nn.Module):net.eval()  # 设置为评估模式if not device:device = next(iter(net.parameters())).devicemetric = d2l.Accumulator(2)#metric[0]:正确预测的数量，metric[1]:总预测的数量。with torch.no_grad():for X, y in data_iter:if isinstance(X, list):X = [x.to(device) for x in X]# BERT微调所需，将输入转移到GPU上（之后将介绍）else:X = X.to(device)y = y.to(device)metric.add(d2l.accuracy(net(X), y), y.numel())return metric[0] / metric[1]def train_ch6(net, train_iter, test_iter, num_epochs, lr, device):#@save"""用GPU训练模型"""def init_weights(m):if type(m) == nn.Linear or type(m) == nn.Conv2d:nn.init.xavier_uniform_(m.weight)net.apply(init_weights)print('training on', device)net.to(device)optimizer = torch.optim.SGD(net.parameters(), lr=lr)loss = nn.CrossEntropyLoss()animator = d2l.Animator(xlabel='epoch', xlim=[1, num_epochs],legend=['train loss', 'train acc', 'test acc'])timer, num_batches = d2l.Timer(), len(train_iter)for epoch in range(num_epochs):metric = d2l.Accumulator(3)#metric[i]分别是训练损失之和，训练准确率之和，样本数net.train()for i, (X, y) in enumerate(train_iter):timer.start()optimizer.zero_grad()X, y = X.to(device), y.to(device)y_hat = net(X)l = loss(y_hat, y)l.backward()optimizer.step()with torch.no_grad():metric.add(l * X.shape[0], d2l.accuracy(y_hat, y), X.shape[0])timer.stop()train_l = metric[0] / metric[2]train_acc = metric[1] / metric[2]if (i + 1) % (num_batches // 5) == 0 or i == num_batches - 1:animator.add(epoch + (i + 1) / num_batches,(train_l, train_acc, None))test_acc = evaluate_accuracy_gpu(net, test_iter)animator.add(epoch + 1, (None, None, test_acc))print(f'loss {train_l:.3f}, train acc {train_acc:.3f}, test acc {test_acc:.3f}')print(f'{metric[2] * num_epochs / timer.sum():.1f} examples/sec on {str(device)}')lr, num_epochs = 0.9, 10
train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
plt.show()

训练结果：