使用pytorch构建一维三层CNN网络

下面是使用PyTorch构建一维三层CNN网络的示例代码。

import torch.nn as nn
import torch.nn.functional as F

class MyCNN(nn.Module):
    def __init__(self):
        super(MyCNN, self).__init__()
        
        # three convolutional layers with filters of size 3, 5, and 7
        self.conv1 = nn.Conv1d(1, 16, kernel_size=3)
        self.conv2 = nn.Conv1d(16, 32, kernel_size=5)
        self.conv3 = nn.Conv1d(32, 64, kernel_size=7)
        
        # max pooling layer
        self.pool = nn.MaxPool1d(2)
        
        # fully connected layers
        self.fc1 = nn.Linear(64*10, 128)
        self.fc2 = nn.Linear(128, 10)
        
    def forward(self, x):
        # pass input through convolutional layers
        x = F.relu(self.conv1(x))
        x = self.pool(x)
        x = F.relu(self.conv2(x))
        x = self.pool(x)
        x = F.relu(self.conv3(x))
        x = self.pool(x)
        
        # flatten the output
        x = x.view(-1, 64*10)
        
        # pass through fully connected layers
        x = F.relu(self.fc1(x))
        x = self.fc2(x)
        
        return x

这是一个简单的3层CNN网络，输入数据是一维的。网络的第一层卷积层包含16个3个长度的滤波器，第二层卷积层包含32个5个长度的滤波器，第三层卷积层包含64个7个长度的滤波器。每个卷积层之后都有一个最大池化层，将特征图的大小减半。最后，两个全连接层将特征映射到输出类别。