Pytorch学习11_神经网络-卷积层

1.创建神经网络实例

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

dataset=torchvision.datasets.CIFAR10("../dataset_cov2d",train=False,transform=torchvision.transforms.ToTensor(),
                                     download=True)
dataloader=DataLoader(dataset,batch_size=64)
class Xuexu(nn.Module):
    def __init__(self):
        super(Xuexu, self).__init__()
        self.conv1=Conv2d(in_channels=3,out_channels=6,kernel_size=3,stride=1,padding=0)

    def forward(self,x):
        self.conv1(x)
        return x

xuexu=Xuexu()
print(xuexu)

2.观察输出

输出图像的形状信息，以便检查输入和输出的张量形状是否符合预期。在深度学习中，了解输入和输出的形状对于调试和确保网络结构正确连接非常重要

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

# 加载 CIFAR-10 数据集
from torch.utils.tensorboard import SummaryWriter

dataset=torchvision.datasets.CIFAR10("../dataset_cov2d",train=False,transform=torchvision.transforms.ToTensor(),
                                     download=True)
dataloader=DataLoader(dataset,batch_size=64)# 创建数据加载器

# 定义神经网络模型
class Xuexu(nn.Module):
    def __init__(self):
        super(Xuexu, self).__init__()
        # 定义卷积层
        self.conv1=Conv2d(in_channels=3,out_channels=6,kernel_size=3,stride=1,padding=0)

    def forward(self, x):
        x=self.conv1(x)
        return x

xuexu=Xuexu()
# writer=SummaryWriter("./logs")
step=0
for data in dataloader:
    imgs,target=data
    output=xuexu(imgs)
    print(f"imgs:{imgs.shape}")
    print(f"output:{output.shape}")
    # writer.add_images("input",imgs,step)
    # writer.add_images("output",output,step)

    # step+=1

3.修改输出

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

# 加载 CIFAR-10 数据集
from torch.utils.tensorboard import SummaryWriter

dataset=torchvision.datasets.CIFAR10("../dataset_cov2d",train=False,transform=torchvision.transforms.ToTensor(),
                                     download=True)
dataloader=DataLoader(dataset,batch_size=64)# 创建数据加载器

# 定义神经网络模型
class Xuexu(nn.Module):
    def __init__(self):
        super(Xuexu, self).__init__()
        # 定义卷积层
        self.conv1=Conv2d(in_channels=3,out_channels=6,kernel_size=3,stride=1,padding=0)

    def forward(self, x):
        x=self.conv1(x)# 调用卷积层，并将输出赋给 x
        return x

xuexu=Xuexu()
writer=SummaryWriter("./logs")
step=0
for data in dataloader:
    imgs,target=data
    output=xuexu(imgs)
    # print(f"imgs:{imgs.shape}")
    # print(f"output:{output.shape}")
    # imgs:torch.Size([64, 3, 32, 32])
    writer.add_images("input",imgs,step)

    # output:torch.Size([64, 6, 30, 30])->[xxx,3,30,30]
    output=torch.reshape(output,(-1,3,30,30))
    writer.add_images("output",output,step)
    step+=1

终端运行

tensorboard --logdir="logs"

点击蓝色链接进入tensorboard网页

可以看到图片经过卷积之后得到的输出

参考

【PyTorch深度学习快速入门教程（绝对通俗易懂！）【小土堆】】 https://www.bilibili.com/video/BV1hE411t7RN/?p=18&share_source=copy_web&vd_source=be33b1553b08cc7b94afdd6c8a50dc5a