PyTorch快速入门教程【小土堆】-神经网络-卷积层

最新推荐文章于 2024-11-11 17:15:27 发布

润叶~

最新推荐文章于 2024-11-11 17:15:27 发布

阅读量354

点赞数 1

CC 4.0 BY-SA版权

分类专栏： PyTorch快速入门教程【小土堆】学习笔记文章标签： p2p 网络协议网络

本文链接：https://blog.youkuaiyun.com/weixin_42182290/article/details/124718164

PyTorch快速入门教程【小土堆】学习笔记专栏收录该内容

18 篇文章

订阅专栏

1.nn.Conv2d

参数：

CLASS

torch.nn.Conv2d(in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, bias=True, padding_mode='zeros', device=None, dtype=None)[SOURCE]

in_channels (int) – Number of channels in the input image 输入图像通道数
out_channels (int) – Number of channels produced by the convolution 输出图像通道数
kernel_size (int or tuple) – Size of the convolving kernel 卷积核大小
stride (int or tuple, optional) – Stride of the convolution. Default: 1 步长
padding (int, tuple or str, optional) – Padding added to all four sides of the input. Default: 0
padding_mode (string, optional) – 'zeros', 'reflect', 'replicate' or 'circular'. Default: 'zeros' padding填充已什么样的方式填充
dilation (int or tuple, optional) – Spacing between kernel elements. Default: 1
groups (int, optional) – Number of blocked connections from input channels to output channels. Default: 1
bias (bool, optional) – If True, adds a learnable bias to the output. Default: True 偏置

公式：

N是batch_size,Cin输入通道数，Hin 输入的高，Win 输入的宽

2.使用

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

#下载数据集
dataset=torchvision.datasets.CIFAR10("dataset",train=False,transform=torchvision.transforms.ToTensor(),download=True)
dataloader=DataLoader(dataset,batch_size=64)

#搭建简单的神经网络
class Lh(nn.Module):
    def __init__(self) -> None:
        super().__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

lh=Lh() #实例化
writer=SummaryWriter("pp")
step=0
for data in dataloader:
    imgs,targets=data
    output=lh(imgs)
    print(imgs.shape)
    print(output.shape)
    #torch.Size([64, 3, 32, 32]) 输入大小 batch_size=64 in_channel=3,32*32
    writer.add_images("input",imgs,step)
    #torch.Size([64, 6, 30, 30]) 输出大小
    #将形状变为[xxx,3,30,30]
    output=torch.reshape(output,(-1,3,30,30))
    writer.add_images("output",output,step)
    step=step+1

writer.close()