nn.Conv2d的使用(卷积层)(附代码)

nn.Conv2d的使用

此处我们仍然使用官网自带的数据集进行训练,最后将其可视化
加载数据集和可视化部分在此处不在介绍,若需要了解:
加载数据集:torch.utils.data中的DataLoader数据加载器(附代码)_硕大的蛋的博客-CSDN博客
tensorboard可视化工具:Tensorboard 可视化工具的使用-史上最简单(附代码)_硕大的蛋的博客-CSDN博客

1. 导入相应的包和模块

   import torch
   import torchvision
   import torch.nn as nn
   from torch.utils.data import DataLoader
   from tensorboardX import SummaryWriter
   

2. 获取数据

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

3. 创建神经网络

   class Gsw(nn.Module):
       def __init__(self):
           super(Gsw, self).__init__()
           self.con1 = nn.Conv2d(in_channels=3, out_channels=6, kernel_size=3, stride=1, padding=0)
   
       def forward(self, input):
           output = self.con1(input)
           return output
   
   

4. 训练并将其可视化

   gsw = Gsw()
   writer = SummaryWriter('LOGS/011log')
   
   for step, data in enumerate(dataloader):
       imgs, target = data
       output = gsw(imgs)
   
       writer.add_images('input', imgs, step)
       output = torch.reshape(output, (-1, 3, 30, 30))
       writer.add_images('output', output, step)
   
   

完整代码

# 开发时间: 2021/11/21 17:51

import torch
import torchvision
import torch.nn as nn
from torch.utils.data import DataLoader
from tensorboardX import SummaryWriter

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


class Gsw(nn.Module):
    def __init__(self):
        super(Gsw, self).__init__()
        self.con1 = nn.Conv2d(in_channels=3, out_channels=6, kernel_size=3, stride=1, padding=0)

    def forward(self, input):
        output = self.con1(input)
        return output


gsw = Gsw()
writer = SummaryWriter('LOGS/011log')

for step, data in enumerate(dataloader):
    imgs, target = data
    output = gsw(imgs)

    writer.add_images('input', imgs, step)
    output = torch.reshape(output, (-1, 3, 30, 30))
    writer.add_images('output', output, step)