Pytorch【60天修炼计划】之第二阶段——进阶：卷积神经网络之LeNet

Dear 丶 2023-05-31 10:14 25阅读 0赞

LeNet是来自论文《Gradient-Based Learning Applied to Document Recognition》中提出的网络，它是CNN的开山鼻祖，对于手写数字的识别有了当时最先进的结果。

这是LeNet的网络结构：  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2thcnJ5X3p6ag_size_16_color_FFFFFF_t_70]

*  input \[ 1, 32 × 32 \] ( channels, weight \* height)
 *  Conv2d \[ 1, 6, 5 × 5 \] (in\_channels, out\_channels, kernel\_size)
 *  Maxpool2d \[ 6, 2 × 2, 2 \] (channels, kernel\_size, stride)
 *  Conv2d \[ 6, 16, 5 × 5 \]
 *  Maxpool2d \[ 16, 2 × 2, 2 \]
 *  FC \[ 16 × 4 × 4, 120 \]
 *  FC \[ 120, 84 \]
 *  FC \[ 84, 10 \]

## 项目结构图 ##

![20191024161150239.png][]

## 模型 ##

按照上面对于网络结构的分解，在pytorch构建如下网络：

class LeNet(nn.Module):
        def __init__(self):
            super(LeNet, self).__init__()
            self.conv = nn.Sequential(
                nn.Conv2d(1, 6, 5),
                nn.Sigmoid(),
                nn.MaxPool2d(2, 2),
                nn.Conv2d(6, 16, 5),
                nn.Sigmoid(),
                nn.MaxPool2d(2, 2)
            )
            self.fc = nn.Sequential(
                nn.Linear(16*4*4, 120),
                nn.Sigmoid(),
                nn.Linear(120, 84),
                nn.Sigmoid(),
                nn.Linear(84, 10)
            )
    
        def forward(self ,X):
            feature = self.conv(X)
            output = self.fc(feature.view(X.shape[0], -1))
            return output

## 训练数据 ##

def train(net, train_iter, test_iter, batch_size, optimizer, device, num_epochs):
        net = net.to(device)
        print("training on ", device)
        loss = nn.CrossEntropyLoss()
        batch_count = 0
        for epoch in range(num_epochs):
            train_l_sum, train_acc_sum, n, start = 0.0, 0.0, 0, time.time()
            for X, y in train_iter:
                # 放入设备中
                X = X.to(device)
                y = y.to(device)
                # 由模型得到的值
                y_hat = net(X)
                # 求损失
                l = loss(y_hat, y)
                optimizer.zero_grad()
                l.backward()
                optimizer.step()
                train_l_sum += l.cpu().item()
                train_acc_sum += (y_hat.argmax(dim=1) == y).sum().cpu().item()
                n += y.shape[0]
                batch_count += 1
    
            test_acc = evaluate_accuracy(test_iter, net)
            print(train_l_sum, train_acc_sum)
            print('epoch %d, loss %.4f, train acc %.3f, test acc %.3f,time % .1f sec' % (epoch + 1, train_l_sum / batch_count, train_acc_sum / n, test_acc, time.time() - start))

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2thcnJ5X3p6ag_size_16_color_FFFFFF_t_70]: /images/20230531/e3b0cc8ea779471cb698436c6fbf5258.png
[20191024161150239.png]: /images/20230531/7af3bc5134044346a69c05d74d8ae133.png