dilation=1，称为空洞卷积，在卷积核相邻像素之间插入一个空白像素。

默认池化核：kernel_size = 3

Ceil_model=True or False： 是否对非完整像素进行保留(默认为False)

import torch
import torchvision.datasets
from mmcv import DataLoader
from mmcv.cnn import MaxPool2d
from torch import nn
from torch.utils.tensorboard import SummaryWriter

dataset = torchvision.datasets.CIFAR10(r"C:\Users\123\Desktop\python4.7\test03\data", 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)    # 经常使用的是浮点数，而这里计算机以为是整数，所以需要变一下
# input = torch.reshape(input, (-1, 1, 5, 5))   # Input: (N, C, H_{in}, W_{in})   N 为batchsize
#
# print(input.shape)


class LR(nn.Module):
    def __init__(self):
        super(LR, self).__init__()
        self.maxpool = MaxPool2d(kernel_size=3, ceil_mode=True)

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

lrp = LR()
# output = lrp(input)
# print(output)

writer = SummaryWriter("logs_maxpool")
step = 0

for data in dataloader:
    imgs, labels = data
    writer.add_images("input", imgs, step)
    output = lrp(imgs)
    writer.add_images("output", output, step)
    step = step + 1

writer.close()