小土堆pytorch教程学习笔记P19

P19.神经网络-最大池化的使用

Pytorch官网 -> Docs -> Pytorch -> torch.nn -> Pooling Layers -> MaxPool2d最大池化&＃xff08;下采样&＃xff09;

class torch.nn.MaxPool2d(kernel_size, stride&＃61;None, padding&＃61;0, dilation&＃61;1, return_indices&＃61;False, ceil_mode&＃61;False)

Parameters

• kernel_size – the size of the window to take a max over

• stride – the stride of the window. Default value is kernel_size

• padding – implicit zero padding to be added on both sides

• dilation – a parameter that controls the stride of elements in the window

• return_indices – if True, will return the max indices along with the outputs. Useful for torch.nn.MaxUnpool2d later

• ceil_mode – when True, will use ceil instead of floor to compute the output shape

import torch
input &＃61; 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]])input &＃61; torch.reshape(input, (-1, 1, 5, 5)) # -1 means compute batch_size automatically
print(input.shape)

 torch.Size([1, 1, 5, 5])

import torch
from torch import nn
from torch.nn import MaxPool2dinput &＃61; 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&＃61;torch.float32) # 1 -> 1.0, 2 -> 2.0input &＃61; torch.reshape(input, (-1, 1, 5, 5)) # -1 means compute batch_size automatically
print(input.shape)class Tudui(nn.Module):def __init__(self):super(Tudui, self).__init__()self.maxpool1 &＃61; MaxPool2d(kernel_size&＃61;3, ceil_mode&＃61;True)def forward(self, input):output &＃61; self.maxpool1(input)return outputtudui &＃61; Tudui()
output &＃61; tudui(input)
print(output)

torch.Size([1, 1, 5, 5])
tensor([[[[2., 3.],
          [5., 1.]]]]) 

 RuntimeError: "max_pool2d_with_indices_cpu" not implemented for &＃39;Long&＃39;

解决方法&＃xff1a;

input &＃61; 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&＃61;torch.float32)

import torch
from torch import nn
from torch.nn import MaxPool2dinput &＃61; 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&＃61;torch.float32) # 1 -> 1.0, 2 -> 2.0input &＃61; torch.reshape(input, (-1, 1, 5, 5)) # -1 means compute batch_size automatically
print(input.shape)class Tudui(nn.Module):def __init__(self):super(Tudui, self).__init__()self.maxpool1 &＃61; MaxPool2d(kernel_size&＃61;3, ceil_mode&＃61;False)def forward(self, input):output &＃61; self.maxpool1(input)return outputtudui &＃61; Tudui()
output &＃61; tudui(input)
print(output)

 torch.Size([1, 1, 5, 5])
tensor([[[[2.]]]])

import torch
import torchvision
from torch import nn
from torch.nn import MaxPool2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriterdataset &＃61; torchvision.datasets.CIFAR10("dataset", train&＃61;False, download&＃61;True,transform&＃61;torchvision.transforms.ToTensor())dataloader &＃61; DataLoader(dataset, batch_size&＃61;64)class Tudui(nn.Module):def __init__(self):super(Tudui, self).__init__()self.maxpool1 &＃61; MaxPool2d(kernel_size&＃61;3, ceil_mode&＃61;False)def forward(self, input):output &＃61; self.maxpool1(input)return outputtudui &＃61; Tudui()writer &＃61; SummaryWriter("P19")
step &＃61; 0
for data in dataloader:imgs, targets &＃61; datawriter.add_images("input", imgs, step)output &＃61; tudui(imgs)writer.add_images("output", output, step)step &＃61; step &＃43; 1writer.close()

tensorboard --logdir&＃61;P19