B站小土堆Pytorch视频学习——卷积层的基本操作（2023.06.14）

一.卷积操作

此为functional里的conv2d

import torch
import torch.nn.functional as F
input=torch.tensor([[1,2,0,3,1],
                    [0,1,2,3,1],
                    [1,2,1,0,0],
                    [5,2,3,1,1],
                    [2,1,0,1,1]])
kernel=torch.tensor([[1,2,1],
                     [0,1,0],
                     [2,1,0]])
#pytorch提供的尺寸变换
#通道为1，batchsize为1
input=torch.reshape(input,(1,1,5,5))
kernel=torch.reshape(kernel,(1,1,3,3))
print(input.shape)
print(kernel.shape)
output=F.conv2d(input,kernel,stride=1)
print(output)
output2=F.conv2d(input,kernel,stride=2)
print(output2)
output3=F.conv2d(input,kernel,stride=1,padding=1)
print(output3)

此为nn里的Conv2d

二.卷积意义

将图片卷积

```python
import torch
import torchvision
import torch.nn as nn
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter

#数据处理
dataset=torchvision.datasets.CIFAR10("./data",train=False,transform=torchvision.transforms.ToTensor(),download=True)
data_loader=DataLoader(dataset=dataset,batch_size=64)
#构建神经网络，此神经网络名称叫Tudui，输入的通道数为3，滤波器个数为6个，滤波器大小3x3，步数为1，不扩展，forward中进行卷积
class Tudui(nn.Module):
    def __init__(self):
        super(Tudui,self).__init__()
        self.conv1=nn.Conv2d(in_channels=3,out_channels=6,kernel_size=3,stride=1,padding=0)
    def forward(self,x):
        x=self.conv1(x)
        return x
#构建神经网络对象
tudui=Tudui()
print(tudui)
writer=SummaryWriter("logs")
i=0
for data in data_loader:
    imgs,targets=data
    #调用forwar进行卷积
    output=tudui(imgs)
    print(imgs.shape)
    print(output.shape)
    #因为无法add6通道，只能平铺变成3通道
    output=torch.reshape(output, (-1, 3, 30, 30))
    writer.add_images("before",imgs,i)
    writer.add_images("after",output,i)
    i+=1