学习笔记：动手学深度学习 28 卷积神经网络LeNet

LeNet

import torch
from torch import nn
from d2l import torch as d2l
class Reshape(torch.nn.Module):
    def forward(self, x):
        return x.view(-1, 1, 28, 28)#批量数-1不变，通道数1，
net = torch.nn.Sequential(
    Reshape(),
    #放进第一个卷积层，输入是1，输出是6，为了保证非线性性，在卷积后面加入sigmoid激活函数"""
    nn.Conv2d(1, 6, kernel_size=5, padding=2), nn.Sigmoid(),
    nn.AvgPool2d(2, stride=2),#均值池化层
    nn.Conv2d(6, 16, kernel_size=5), nn.Sigmoid(),#卷积层输入6，输出16
    nn.AvgPool2d(kernel_size=2, stride=2),
    nn.Flatten(),#拉成一维向量
    nn.Linear(16 * 5 * 5, 120), nn.Sigmoid(),
    nn.Linear(120, 84), nn.Sigmoid(),
    nn.Linear(84, 10))
"""检查模型"""
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)
    
Reshape output shape: 	 torch.Size([1, 1, 28, 28])
Conv2d output shape: 	 torch.Size([1, 6, 28, 28])
Sigmoid output shape: 	 torch.Size([1, 6, 28, 28])
AvgPool2d output shape: 	 torch.Size([1, 6, 14, 14])
Conv2d output shape: 	 torch.Size([1, 16, 10, 10])
Sigmoid output shape: 	 torch.Size([1, 16, 10, 10])
AvgPool2d output shape: 	 torch.Size([1, 16, 5, 5])
Flatten output shape: 	 torch.Size([1, 400])
Linear output shape: 	 torch.Size([1, 120])
Sigmoid output shape: 	 torch.Size([1, 120])
Linear output shape: 	 torch.Size([1, 84])
Sigmoid output shape: 	 torch.Size([1, 84])
Linear output shape: 	 torch.Size([1, 10])

模型训练

在模型使用 GPU 计算数据集之前，我们需要将其复制到显存中

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, torch.nn.Module):
        net.eval()  # 设置为评估模式
        if not device:
            device = next(iter(net.parameters())).device
    # 正确预测的数量，总预测的数量
    metric = d2l.Accumulator(2)
    for X, y in data_iter:
        if isinstance(X, list):
            # BERT微调所需的（之后将介绍）
            X = [x.to(device) for x in X]
        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):
    """用GPU训练模型(在第六章定义)。"""
    def init_weights(m):#初始化w
        if type(m) == nn.Linear or type(m) == nn.Conv2d:
            nn.init.xavier_uniform_(m.weight) #全连接层或者卷积层用xavier初始化
    net.apply(init_weights)
    print('training on', device) #在那个device上训练
    net.to(device)#把参数挪到gpu内存上
    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'])
    #动画效果
    timer, num_batches = d2l.Timer(), len(train_iter)
    for epoch in range(num_epochs):#对每一次数据做迭代
        # 训练损失之和，训练准确率之和，范例数
        metric = d2l.Accumulator(3)
        net.train()
        for i, (X, y) in enumerate(train_iter):#在每一次迭代中哪一个batc出来
            timer.start()
            optimizer.zero_grad()#梯度设0
            X, y = X.to(device), y.to(device)#输入和输出挪到gpu上
            y_hat = net(X)#前向操作
            l = loss(y_hat, y)#计算损失
            l.backward()#计算梯度
            optimizer.step()#迭代
            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}, '
          f'test acc {test_acc:.3f}')
    print(f'{metric[2] * num_epochs / timer.sum():.1f} examples/sec '
          f'on {str(device)}')
    
lr, num_epochs = 0.9, 10
train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
training on cuda:0
<Figure size 350x250 with 1 Axes>
<Figure size 350x250 with 1 Axes>


loss 0.465, train acc 0.826, test acc 0.783
17721.1 examples/sec on cuda:0