P27 含并行连结的网络 GoogLeNet / Inception V3

1.名字来源于谷歌

2.最好的卷积层超参数是什么？GoogLeNet表示全部都要

3.Inception块是核心

和3x3，5x5的卷积相比，Inception块有更少的参数个数和计算复杂度

4.有五段，五个stage

5.后面还有其他变种，比如最著名的InceptionV3-ResNet，还有我用的Xception

6.金句：里面的数字不知道怎么来的（原话）

import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l


class Inception(nn.Module):
    def __init__(self,in_channels,c1,c2,c3,c4, **kwargs):
        #c1,c2,c3,c4是每一条path，要分别设置每一条线路的channel数
        super(Inception,self).__init__(**kwargs)
        #线路1
        self.p1_1 = nn.Conv2d(in_channels,c1,kernel_size=1)
        #线路2，有两个卷积层
        self.p2_1 = nn.Conv2d(in_channels,c2[0],kernel_size=1)
        self.p2_2 = nn.Conv2d(c2[0],c2[1],kernel_size=3,padding=1)
        #线路3
        self.p3_1 = nn.Conv2d(in_channels,c3[0],kernel_size=1)
        self.p3_2 = nn.Conv2d(c3[0],c3[1],kernel_size=5,padding=2)
        #线路4
        self.p4_1 = nn.MaxPool2d(kernel_size=3,stride=1,padding=1)
        self.p4_2 = nn.Conv2d(in_channels,c4,kernel_size=1)


    def forward(self,x):
        p1 = F.relu(self.p1_1(x)) #第一个path里加relu
        p2 = F.relu(self.p2_2(F.relu(self.p2_1(x)))) #放到第一层以后放第二层里，继续relu
        p3 = F.relu(self.p3_2(F.relu(self.p3_1(x))))
        p4 = F.relu(self.p4_2(self.p4_1(x)))
    #在通道维度上连结输出
        return torch.cat((p1,p2,p3,p4),dim=1) #在通道数是1（dim=1）的情况下contact起来

#实现每个模块

#第一个模块用64个通道，7*7的卷积层
b1 = nn.Sequential(nn.Conv2d(1, 64, kernel_size=7, stride=2, padding=3),
                   nn.ReLU(),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))

#第二个模块使用两个卷积层：1.64个通道，1*1卷积层 2.通道数量增加3倍=192，3*3卷积层
b2 = nn.Sequential(nn.Conv2d(64, 64, kernel_size=1),
                   nn.ReLU(),
                   nn.Conv2d(64, 192, kernel_size=3, padding=1),
                   nn.ReLU(),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))

#第三个模块，连接第一个和第二个模块，里面全部都是通道数
b3 = nn.Sequential(Inception(192, 64, (96, 128), (16, 32), 32),
                   Inception(256, 128, (128, 192), (32, 96), 64),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))

#第四个模块串联5个inception块，里面的数字不知道怎么来的（原话），都是通道数
b4 = nn.Sequential(Inception(480, 192, (96, 208), (16, 48), 64),
                   Inception(512, 160, (112, 224), (24, 64), 64),
                   Inception(512, 128, (128, 256), (24, 64), 64),
                   Inception(512, 112, (144, 288), (32, 64), 64),
                   Inception(528, 256, (160, 320), (32, 128), 128),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))

b5 = nn.Sequential(Inception(832, 256, (160, 320), (32, 128), 128),
                   Inception(832, 384, (192, 384), (48, 128), 128),
                   nn.AdaptiveAvgPool2d((1,1)),
                   nn.Flatten())


net = nn.Sequential(b1,b2,b3,b4,b5,nn.Linear(1024,10)) #输出类别是10，最后输出block是1024维度

#用Fashion-MNIST测试，把图片的高和宽变成了96

X = torch.rand(size=(1, 1, 96, 96))
for layer in net:
    X = layer(X)
    print(layer.__class__.__name__,'output shape:\t', X.shape)

Sequential output shape:     torch.Size([1, 64, 24, 24])
Sequential output shape:    torch.Size([1, 192, 12, 12])
Sequential output shape:    torch.Size([1, 480, 6, 6])
Sequential output shape:    torch.Size([1, 832, 3, 3])
Sequential output shape:    torch.Size([1, 1024])
Linear output shape:    torch.Size([1, 10])

#训练模型

lr,num_epochs, batch_size = 0.1,10,128
train_iter,test_iter = d2l.load_data_fashion_mnist(batch_size,resize=96)
d2l.train_ch6(net,train_iter,test_iter,num_epochs,lr,d2l.try_gpu())

#loss 0.236, train acc 0.909, test acc 0.898b

pycharm不能显示图片很遗憾