qq_21686871的博客

AlexNet网络的基本流程为：class VGG16(nn.Module): def __init__(self, num_classes): super(VGG16, self).__init__() self.feature = nn.Sequential( nn.Conv2d(in_channels=3, out_channels=64, kernel_size=3, stride=1, padding=1),

import torchimport torchvisionimport torch.nn.functional as Fimport torchvision.transforms as transformsimport torch.nn as nnimport torch.optim as optimfrom torch.utils.data.sampler import SubsetRandomSamplerimport timeimport mathimport numpy as n

import torchfrom torchvision import transformsfrom torchvision import datasetsfrom torch.utils.data import DataLoaderimport torch.nn.functional as Fimport torch.optim as optimfrom matplotlib import pyplot as pltimport osimport sysnum_class=4inp

import torchimport torch.nn as nnimport torch.nn.functional as Fclass ResidualBlock(nn.Module): #残差块：将网络层进行组合，使得z=x+y（x与y同型） def __init__(self,channels):#输入通道数定为未知量，当实例化模型时可以调用 super(ResidualBlock,self).__init__() self.channels=chan

import torchimport torch.nn as nnfrom torchvision import transformsfrom torchvision import datasetsfrom torch.utils.data import DataLoaderimport torch.nn.functional as Fimport torch.optim as optim#1.构建数据集batch_size=64transform=transforms.Compose(

import torchimport torch.nn as nnimport torch.nn.functional as Fclass Inception(nn.Module): def __init__(self,in_channels):#输入通道数定为未知量，当实例化模型时可以调用 super(Inception,self).__init__() self.branch11=nn.Conv2d(in_channels,16,kernel_size=1

import torchimport torch.nn as nnfrom torchvision import transformsfrom torchvision import datasetsfrom torch.utils.data import DataLoaderimport torch.nn.functional as Fimport torch.optim as optim#1.构建数据集batch_size=64transform=transforms.Compose(

import torchinput=[1,2,3,4, 6,7,8,9, 1,2,3,4, 6,7,8,9,]input=torch.Tensor(input).view(1,1,4,4) #(b*c*h*w)maxpooling_layer=torch.nn.MaxPool2d(kernel_size=2, stride=2,) #默认步幅为2output=maxpooli

import torchinput=[1,2,3,4,5, 6,7,8,9,9, 1,2,3,4,5, 6,7,8,9,9, 1,2,3,4,5]#ToTensorinput=torch.Tensor(input).view(1,1,5,5) #(b*c*h*w)#定义卷积层conv_layer=torch.nn.Conv2d(in_channels=1, out_channels=1,

import torchin_channels,out_channels=5,10 #in 决定卷积核的channel；out 决定卷积核的个数width,hight=100,100kernel_size=3 #卷积核的边长 卷积核一般为奇数边长正方形batch_size=1input=torch.randn(batch_size, #random normal in_channels, hight,

import torchimport torch.nn as nnfrom torchvision import transformsfrom torchvision import datasetsfrom torch.utils.data import DataLoaderimport torch.nn.functional as Fimport torch.optim as optim#1.构建数据集batch_size=64transform=transforms.Compose(