06 神經網絡的基本骨架 —— 池化層

# 池化的作用就是，想保留特徵，並把數據量減小

 

1 池化層官方文檔

 

2 最大池化操作

就是取池化核覆蓋範圍內的最大值

# 池化的作用就是，想保留特徵，並把數據量減小

import torch
from torch import nn
from torch.nn import MaxPool2d


# 池化層操作，需要將數據類型轉爲dtype=torch.float32
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)

input = torch.reshape(input, (-1, 1, 5 , 5))   # -1可以自動調整前面的維度
print(input.shape)

class PoolingLayer_test(nn.Module):
    def __init__(self):
        super(PoolingLayer_test, self).__init__()
        self.maxpool_1 = MaxPool2d(kernel_size=3, ceil_mode=False)

    def forward(self, input):
        output = self.maxpool_1(input)
        return output

textPoolingLayer = PoolingLayer_test()
output = textPoolingLayer(input)
print(output)

 

# 池化的作用就是，想保留特徵，並把數據量減小

import torch
import torchvision
from torch import nn
from torch.nn import MaxPool2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter

dataset = torchvision.datasets.CIFAR10(root="D:/Learn/dataSet/Pytorch", train=False, transform=torchvision.transforms.ToTensor(), download=False)
dataloader = DataLoader(dataset, batch_size=64)
writer = SummaryWriter("logs_pooling")


class PoolingLayer_test(nn.Module):
    def __init__(self):
        super(PoolingLayer_test, self).__init__()
        self.maxpool_1 = MaxPool2d(kernel_size=3, ceil_mode=False)

    def forward(self, input):
        output = self.maxpool_1(input)
        return output

textPoolingLayer = PoolingLayer_test()

step = 0
for data in dataloader:
    imgs, targets = data
    print(imgs.shape)
    writer.add_images("input", imgs, step)

    output = textPoolingLayer(imgs)
    print(output.shape)
    writer.add_images("output", output, step)

    step += 1


writer.close()