P24 AlexNet深度卷积神经网络

论文参考沐深的读论文视频

import torch
from torch import nn
from matplotlib import pyplot as plt
from d2l import torch as d2l

#根据教科书上AlexNet的结构来写
net = nn.Sequential(

    nn.Conv2d(1,96,kernel_size=11,stride=4,padding=1),nn.ReLU(),
    #这里用的Fashion-MNIST所以输入通道是1，灰度图；如果用的是RGB彩色图，输入通道是3；96是输出，stride是4，填充=1
    nn.MaxPool2d(kernel_size=3,stride=2),
    nn.Conv2d(96,256,kernel_size=5,padding=2),nn.ReLU(),
    nn.MaxPool2d(kernel_size=3,stride=2),
    nn.Conv2d(256,384,kernel_size=3,padding=1),nn.ReLU(),
    nn.Conv2d(384,384,kernel_size=3,padding=1),nn.ReLU(),
    nn.Conv2d(384,256,kernel_size=3,padding=1),nn.ReLU(),
    nn.MaxPool2d(kernel_size=3,stride=2),nn.Flatten(),
    #到这里5个卷积层结束
    #开始全连接层，拉成2-d的形状
    nn.Linear(6400,4096),nn.ReLU(),nn.Dropout(p=0.5),
    #dropout的概率是0.5,50%的概率把输出变成0
    #4096 = 256*16
    #6400 = 256*5*5
    nn.Linear(4096,4096),nn.ReLU(),nn.Dropout(p=0.5),
    nn.Linear(4096,10))
    #这里用的Fashion-MNIST所以输出分类是10，如果用ImageNet，输出分类是1000，这里用演示方便选了前者

打印参数

#打印参数
X = torch.randn(1,1,224,224)
for layer in net:
        X = layer(X)
        #print(layer.__class__.__name__,'output shape:\t',X.shape)

Conv2d output shape:     torch.Size([1, 96, 54, 54])
经过第一卷积层以后，就把输入从224*224变成了54*54，因为stride=4，就约等于缩小了4倍
ReLU output shape:  torch.Size([1, 96, 54, 54])
MaxPool2d output shape:     torch.Size([1, 96, 26, 26])
在经过maxpooling后再次减半，变成了26，这里看代码，stride=2
Conv2d output shape:    torch.Size([1, 256, 26, 26])
ReLU output shape:  torch.Size([1, 256, 26, 26])
MaxPool2d output shape:     torch.Size([1, 256, 12, 12])
Conv2d output shape:    torch.Size([1, 384, 12, 12])
ReLU output shape:  torch.Size([1, 384, 12, 12])
Conv2d output shape:    torch.Size([1, 384, 12, 12])
ReLU output shape:  torch.Size([1, 384, 12, 12])
Conv2d output shape:    torch.Size([1, 256, 12, 12])
ReLU output shape:  torch.Size([1, 256, 12, 12])
MaxPool2d output shape:     torch.Size([1, 256, 5, 5])
在进入全连接层之前，图片的高宽都变成了5，但是通道变成了256，最后把他拉成一个向量
Flatten output shape:   torch.Size([1, 6400])
Linear output shape:    torch.Size([1, 4096])
ReLU output shape:  torch.Size([1, 4096])
Dropout output shape:   torch.Size([1, 4096])
Linear output shape:    torch.Size([1, 4096])
ReLU output shape:  torch.Size([1, 4096])
Dropout output shape:   torch.Size([1, 4096])
Linear output shape:    torch.Size([1, 10])
最后的输出是10个分类

Fashin-MNIST的图像分辨与低于ImageNet，所以要把他们增加到224*224

batch_size = 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=224)
print(train_iter,test_iter)

开始训练

lr, num_epochs = 0.01, 10
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())

打印结果：

loss 0.513, train acc 0.809, test acc 0.873
4.7 examples/sec on cpu
 

基本上花了2个小时，这个4.7/example/sec是真的吗

问题：这样在pycharm里面跑，还是现实training on cpu