PyTorch快速入门教程【小土堆】-神经网络-最大池化的使用

1.pooling layer

nn.MaxPool1d	Applies a 1D max pooling over an input signal composed of several input planes.
nn.MaxPool2d	Applies a 2D max pooling over an input signal composed of several input planes. 二维最大池化   下采样
nn.MaxPool3d	Applies a 3D max pooling over an input signal composed of several input planes.
nn.MaxUnpool1d	Computes a partial inverse of MaxPool1d.
nn.MaxUnpool2d	Computes a partial inverse of MaxPool2d. 上采样

（1）nn.MaxPool2d

作用：本质是基于滑动窗口的思想，可以去除特征图中的冗余信息，降低特征图的维度。

torch.nn.MaxPool2d(kernel_size, stride=None, padding=0, dilation=1, return_indices=False, ceil_mode=False)

kernel_size :池化核

stride：步长  默认值 是池化核大小

dilation：空洞池化，原来正常的是紧挨在一起的，现在不是，如下图：

ceil_mode：设置为true时，用cell模式，false时用floor模式，floor舍，ceil入

例：步长是池化核的大小，cell_mode举例

（2） 代码演示

 

import tensorboard
import torch
import torchvision.datasets
from torch import nn
from torch.nn import MaxPool2d
#下载数据集
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter

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

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]],dtype=torch.float32)#dtype 将数据转为浮点数

input=torch.reshape(input,(-1,1,5,5))#参数，batch_size=-1,channel=1,大小是5*5
print(input.shape)
class Tudui(nn.Module):

    def __init__(self) -> None:
        super().__init__()
        self.maxpool1=MaxPool2d(kernel_size=3,stride=3,ceil_mode=False)

    def forward(self,input):
        output=self.maxpool1(input)
        return output

tudui=Tudui()
output=tudui(input)
print(output)
tu=Tudui()
writer=SummaryWriter("logs_maxpool")
step=0
for data in dataloader:
    imgs,targets=data
    writer.add_images("input",imgs,step)
    output=tu(imgs)
    writer.add_images("output",output,step)
    step=step+1

writer.close()