AlexNet+pytorch

AlexNet:

AlexNet是2012年ImageNet竞赛冠军获得者Hinton和他的学生Alex Krizhevsky设计的。也是在那年之后，更多的更深的神经网络被提出，比如优秀的vgg,GoogLeNet。 这对于传统的机器学习分类算法而言，已经相当的出色。其结构如下

 AlexNet网络相对于Lenet层数更深，同时第一次引入了激活函数ReLU,在全连接层引入了Dropout层防止过拟合。Dropout就是随机使一些神经元失效不参与网络的计算。

 

pytorch 网络代码：

import torch
import torch.nn as nn


class _AlexNet(nn.Module):
    def __init__(self):
        super(_AlexNet,self).__init__()
        self.conv1=nn.Sequential(
                nn.Conv2d(in_chananels=3,out_channels=64,kernel_size=11,stride=4,padding=2),#in 3*223*223   out 64*55*55
                nn.ReLU(inplace=True),
                nn.MaxPool2d(kernel_size=3,stride=2),     
                )
        self.conv2=nn.Sequential(
                nn.Conv2d(in_channels=64,out_channels=192,kernel_size=5,padding=2),
                nn.ReLU(inplace=True),
                nn.MaxPool2d(kernel_size=3,stride=2), 
                )
        self.conv3=nn.Sequential(
                nn.Conv2d(in_channels=192,out_channels=384,kernel_size=3,padding=1),
                nn.ReLU(inplace=True),
                )
        self.conv4=nn.Sequential(
                nn.Conv2d(in_channels=384,out_channels=256,kernel_size=3,padding=1),
                nn.ReLU(inplace=True),
                )
        self.conv5=nn.Sequential(
                nn.Conv2d(in_channels=256,out_channels=256,kernel_size=3,padding=1),
                nn.ReLU(inplace=True),
                nn.MaxPool2d(kernel_size=3,stride=2)
                )
        self.classifier=nn.Sequential(
                nn.Dropout(),
                nn.Liner(256*6*6,4096),
                nn.ReLU(inplace=True),
                nn.Dropout(),
                nn.Liner(4096,4096),
                nn.ReLU(inplace=True),
                nn.Liner(4096,10)
                )
     
    def forward(self,x):
        x=self.conv1(x)
        x=self.conv2(x)
        x=self.conv3(x)
        x=self.conv4(x)
        x=self.conv5(x)
        #x=x.view(x.size(0),-1)
        x=x.view(x.size(0),256*6*6)
        output=self.classifier(x)
        return output


AlexNet=_AlexNet()
AlexNet.cuda()

optimizer=torch.optim.Adam(_AlexNet.parameters(),lr=LR)
loss_func=nn.CrossEntropyLoss()