经典的卷积神经网络及其Pytorch代码实现
Posted candyren
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了经典的卷积神经网络及其Pytorch代码实现相关的知识,希望对你有一定的参考价值。
1.LeNet
LeNet是指LeNet-5,它是第一个成功应用于数字识别的卷积神经网络。在MNIST数据集上,可以达到99.2%的准确率。LeNet-5模型总共有7层,包括两个卷积层,两个池化层,两个全连接层和一个输出层。
import torch import torch.nn as nn from torch.autograd import Variable #方形卷积核和等长的步长 m1=nn.Conv2d(16,33,3,stride=2) #非长方形卷积核,非等长的步长和边界填充 m2=nn.Conv2d(16,33,(3,5),stride=(2,1),padding=(4,2)) #非方形卷积核,非等长的步长,边界填充和空间间隔 m3=nn.Conv2d(16,33,(3,5),stride=(2,1),padding=(4,2),dilation=(3,1)) input=Variable(torch.randn(20,16,50,100)) output=m2(input)
####LeNet的PyTorch实现 class LeNet(nn.Module): def __init__(self): super(LeNet,self).__init__() self.conv1=nn.Conv2d(3,6,5) self.conv2=nn.Conv2d(6,16,5) self.fc1=nn.Linear(16*5*5,120) self.fc2=nn.Linear(120,84) self.fc3=nn.Linear(84,10) def forward(self,x): out=F.relu(self.conv1(x)) out=F.max_pool2d(out,2) out=F.relu(self.conv2(out)) out=F.max_pool2d(out,2) #这句话一般出现在model类的forward函数中,具体位置一般都是在调用分类器之前。 #分类器是一个简单的nn.Linear()结构,输入输出都是维度为一的值,x = x.view(x.size(0), -1) #这句话的出现就是为了将前面多维度的tensor展平成一维 #x = x.view(batchsize, -1)中batchsize指转换后有几行, #而-1指在不告诉函数有多少列的情况下,根据原tensor数据和batchsize自动分配列数。 out=out.view(out.size(0),-1) out=F.relu(self.fc1(out)) our=F.relu(self.fc2(out)) out=self.fc3(out) return out
2.AlexNet
AlexNet具有更深的网络结构,使用层叠的卷积层,同时增加了Dropout和数据增强,并使用ReLU代替了之前的sigmoid函数,采用多GPU训练。
AlexNet共8层,前5层为卷积层,后3层为全连接层。
#####AlexNet的PyTorch实现 class AlexNet(nn.Module): def __init__(self,num_classes): super(AlexNet,self).__init__() self.features=nn.Sequential( nn.Conv2d(3,96,kernel_size=11,stride=4,padding=2), nn.ReLU(inplace=True), nn.MaxPool2d(kernel_size=3,stride=2), nn.Conv2d(96,256,kernel_size=5,padding=2), nn.ReLU(inplace=True), nn.MaxPool2d(kernel_size=3,stride=2), nn.Conv2d(256,384,kernel_size=3,padding=1), nn.ReLU(inplace=True), nn.Conv2d(384,384,kernel_size=3,padding=1), nn.ReLU(inplace=True), nn.Conv2d(384,256,kernel_size=3,padding=1), nn.ReLU(inplace=True), nn.MaxPool2d(kernel_size=3,stride=2), ) self.classifier=nn.Sequential( nn.Dropout(), nn.Linear(256*6*6,4096), nn.ReLU(inplace=True), nn.Dropout(), nn.Linear(4096,4096), nn.ReLU(inplace=True), nn.Linear(4096,num_classes), ) def forward(self,x): x=self.features() x=x.view(x.size(0),256*6*6) x=self.classifier(x) return x
3.VGGNet
VGGNet采用了几个3*3的卷积核代替AlexNet中较大的卷积核,模型由若干卷积层和池化层堆叠而成。
####VGGNet的实现 cfg={ ‘VGG11‘:[64,‘M‘,128,‘M‘,256,256,‘M‘,512,512,‘M‘,512,512,‘M‘], ‘VGG13‘:[64,64,‘M‘,128,128,‘M‘,256,256,‘M‘,512,512,‘M‘,512,512,‘M‘], ‘VGG16‘:[64,64,‘M‘,128,128,‘M‘,256,256,256,‘M‘,512,512,512,‘M‘,512,512,512,‘M‘], ‘VGG19‘:[64,64,‘M‘,128,128,‘M‘,256,256,256,256,‘M‘,512,512,512,512,‘M‘,512,512,512,512,‘M‘], } class VGG(nn.Module): def __init__(self,vgg_name): super(VGG,self).__init__() self.features=self._make_layers(cfg[vgg_name]) self.classifier=nn.Linear(512,10) def forward(self,x): out=self.features(x) out=out.view(out.size(0),-1) out=self.classifier(out) return out def _make_layers(self,cfg): layers=[] in_channels=3 for x in cfg: if x ==‘M‘: layers+=[nn.MaxPool2d(kernel_size=2,stride=2)] else: layers+=[nn.Conv2d(in_channels,x,kernal_size=3,padding=1),nn.BatchNorm2d(x),nn.ReLU(inplace=True)] in_channels=x layers+=[nn.AvgPool2d(kernel_size=1,stride=1)] return nn.Sequential(*layers)
4.
以上是关于经典的卷积神经网络及其Pytorch代码实现的主要内容,如果未能解决你的问题,请参考以下文章
[Pytorch系列-50]:卷积神经网络 - FineTuning的统一处理流程与软件架构 - Pytorch代码实现