卷积层

介绍

卷积层对应前面的卷积操作，直接作为神经网络的一层。

import torch
import torchvision
from torch import nn
from torch.nn import Conv2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter

dataset = torchvision.datasets.CIFAR10("../data", train=False, transform=torchvision.transforms.ToTensor(),
                                       download=True)
dataloader = DataLoader(dataset, batch_size=64)


class Tudui(nn.Module):
    def __init__(self):
        super(Tudui, self).__init__()
        self.conv1 = Conv2d(in_channels=3, out_channels=6, kernel_size=3, stride=1, padding=0)  # 输入通道(这里RGB是3)、输出通道(卷积核数量)
    def forward(self, x):
        x = self.conv1(x)  # x 已经放到了卷积层 conv1当中了
        return x


tudui = Tudui()  # 初始化网络
print(tudui)

# 下面把每一张图像都进行卷积

writer = SummaryWriter("logs")

step = 0
for data in dataloader:
    imgs, targets = data  
    output = tudui(imgs)
    print("imgs.shape:", imgs.shape)  
    print("output.shape:", output.shape)  
    # torch.Size([64, 3, 32, 32])
    writer.add_images("input", imgs, step)
    # torch.Size([64, 6, 30, 30])  由于6个channel的图像，是无法显示的
    # torch.Size([xxx, 3, 30, 30])
    output = torch.reshape(output, (-1, 3, 30, 30))  # 强行转，无依据，正常可以汇聚层、奇异值分解等。
    writer.add_images("output", output, step)
    step += 1

writer.close()

参数说明:

​ in_channels 输入通道数

​ out_channels 输出通道数，对应卷积核数量

​ kernel_size 卷积核大小，这里卷积核大小是n*n的

​ stride 步长

​ padding 填充

后面没啥可介绍的，注意下细节就行。