Lecture11---Advanced CNN【GoogLeNet && Deep Residual Learning】

提示：刘二大人系列视频学习笔记

提示：以下是本篇文章正文内容，下面案例可供参考

Chapter 1 GoogLeNet

1.1 Implementation of Inception Module

1. Concatenate:

...

outputs  = [branch1x1,branch5x5,branch3x3,branch_pool]
        return torch.cat(outputs,dim=1) #concatenate

2. the effect of 1*1 conv:

1.2 Code

import torch
import torch.nn as nn
from torchvision import transforms #for constructing DataLoader
from torchvision import datasets #for constructing DataLoader
from torch.utils.data import DataLoader #for constructing DataLoader
import torch.nn.functional as F #for using relu() and pooling
import torch.optim as optim #for constructing Optimizer

#prepare dataset
batch_size=64
transform= transforms.Compose ([transforms.ToTensor( ),transforms.Normalize((0.1307,),(0.3081,))])#convert the PIL Image to Tensor[0,1]
                                                                                                  # normalization according to  mean and std of mnist

train_dataset=datasets.MNIST(root='./dataset',train=True,download=False,transform=transform)
train_loader=DataLoader(train_dataset,shuffle=True,batch_size=batch_size)
test_dataset=datasets.MNIST(root='./dataset',train=False,download=False,transform=transform)
test_loader=DataLoader(test_dataset,shuffle=False,batch_size=batch_size)

#design model using Class
class InceptionA(nn.Module):
    def __init__(self,in_channels):
        super(InceptionA, self).__init__()
        self.branch1x1 = nn.Conv2d(in_channels,16,kernel_size=1)

        self.branch5x5_1 = nn.Conv2d(in_channels,16,kernel_size=1)
        self.branch5x5_2 = nn.Conv2d(16,24,kernel_size=5,padding=2)

        self.branch3x3_1 = nn.Conv2d(in_channels,16,kernel_size=1)
        self.branch3x3_2 = nn.Conv2d(16,24,kernel_size=3,padding=1)
        self.branch3x3_3 = nn.Conv2d(24,24,kernel_size=3,padding=1)

        self.branch_pool = nn.Conv2d(in_channels,24,kernel_size=1)

    def forward(self,x):
        branch1x1 = self.branch1x1(x)

        branch5x5 = self.branch5x5_1(x)
        branch5x5 = self.branch5x5_2(branch5x5)

        branch3x3 = self.branch3x3_1(x)
        branch3x3 = self.branch3x3_2(branch3x3)
        branch3x3 = self.branch3x3_3(branch3x3)

        branch_pool = F.avg_pool2d(x,kernel_size=3,stride=1,padding=1)
        branch_pool = self.branch_pool(branch_pool)

        outputs  = [branch1x1,branch5x5,branch3x3,branch_pool]
        return torch.cat(outputs,dim=1) #concatenate
class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1,10,kernel_size=5)
        self.conv2 = nn.Conv2d(88, 20, kernel_size=5)#88=24*3+16
        self.incep1 = InceptionA(in_channels=10)
        self.incep2 = InceptionA(in_channels=20)

        self.mp = nn.MaxPool2d(2)
        self.fc = nn.Linear(1408, 10)

    def forward(self, x):
        in_size = x.size(0)
        x = F.relu(self.mp(self.conv1(x)))
        x = self.incep1(x)
        x = F.relu(self.mp(self.conv2(x)))
        x = self.incep2(x)
        x = x.view(in_size, -1)#change into one dimension
        x = self.fc(x)

        return x


model = Net()

# construct loss and optimizer
criterion = torch.nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)


# training cycle forward, backward, update


def train(epoch):
    running_loss = 0.0
    for batch_idx, data in enumerate(train_loader, 0):
        inputs, target = data
        optimizer.zero_grad()

        outputs = model(inputs)
        loss = criterion(outputs, target)
        loss.backward()
        optimizer.step()

        running_loss += loss.item()
        if batch_idx % 300 == 299:
            print('[%d, %5d] loss: %.3f' % (epoch + 1, batch_idx + 1, running_loss / 300))
            running_loss = 0.0


def test():
    correct = 0
    total = 0
    with torch.no_grad():
        for data in test_loader:
            images, labels = data
            outputs = model(images)
            _, predicted = torch.max(outputs.data, dim=1) #similar with dict
            total += labels.size(0)
            correct += (predicted == labels).sum().item()
    print('accuracy on test set: %d %% ' % (100 * correct / total))


if __name__ == '__main__':
    for epoch in range(10):
        train(epoch)
        test()

1.3 Result

[1,   300] loss: 0.766
[1,   600] loss: 0.209
[1,   900] loss: 0.137
accuracy on test set: 96 % 
[2,   300] loss: 0.110
[2,   600] loss: 0.101
[2,   900] loss: 0.090
accuracy on test set: 97 % 
[3,   300] loss: 0.082
[3,   600] loss: 0.078
[3,   900] loss: 0.068
accuracy on test set: 97 % 
[4,   300] loss: 0.065
[4,   600] loss: 0.060
[4,   900] loss: 0.061
accuracy on test set: 97 % 
[5,   300] loss: 0.057
[5,   600] loss: 0.055
[5,   900] loss: 0.050
accuracy on test set: 98 % 
[6,   300] loss: 0.050
[6,   600] loss: 0.051
[6,   900] loss: 0.045
accuracy on test set: 98 % 
[7,   300] loss: 0.045
[7,   600] loss: 0.046
[7,   900] loss: 0.044
accuracy on test set: 98 % 
[8,   300] loss: 0.037
[8,   600] loss: 0.039
[8,   900] loss: 0.045
accuracy on test set: 98 % 
[9,   300] loss: 0.033
[9,   600] loss: 0.040
[9,   900] loss: 0.039
accuracy on test set: 98 % 
[10,   300] loss: 0.036
[10,   600] loss: 0.032
[10,   900] loss: 0.036
accuracy on test set: 98 % 

1.4 More details

1.average pool

branch_pool = F.avg_pool2d(x,kernel_size=3,stride=1,padding=1)

2.view()

x = x.view(in_size, -1)

3.enumerate

for batch_idx, data in enumerate(train_loader, 0)

4.torch.max

_, predicted = torch.max(outputs.data, dim=1)

Chapter 2 Deep Residual Learning

2.1 Implementation of Deep Residual Learning

H (x) = f (x) + X, the tensors’ dimensions must be the same.

2.2 Code

import torch
import torch.nn as nn
from torchvision import transforms
from torchvision import datasets
from torch.utils.data import DataLoader
import torch.nn.functional as F
import torch.optim as optim

# prepare dataset

batch_size = 64
transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))])  # 归一化,均值和方差

train_dataset = datasets.MNIST(root='./dataset', train=True, download=True, transform=transform)
train_loader = DataLoader(train_dataset, shuffle=True, batch_size=batch_size)
test_dataset = datasets.MNIST(root='./dataset', train=False, download=True, transform=transform)
test_loader = DataLoader(test_dataset, shuffle=False, batch_size=batch_size)


# design model using class
class ResidualBlock(nn.Module):
    def __init__(self, channels):
        super(ResidualBlock, self).__init__()
        self.channels = channels
        self.conv1 = nn.Conv2d(channels, channels, kernel_size=3, padding=1)
        self.conv2 = nn.Conv2d(channels, channels, kernel_size=3, padding=1)

    def forward(self, x):
        y = F.relu(self.conv1(x))
        y = self.conv2(y)
        return F.relu(x + y)


class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 16, kernel_size=5)
        self.conv2 = nn.Conv2d(16, 32, kernel_size=5)

        self.rblock1 = ResidualBlock(16)
        self.rblock2 = ResidualBlock(32)

        self.mp = nn.MaxPool2d(2)
        self.fc = nn.Linear(512, 10)

    def forward(self, x):
        in_size = x.size(0)

        x = self.mp(F.relu(self.conv1(x)))
        x = self.rblock1(x)
        x = self.mp(F.relu(self.conv2(x)))
        x = self.rblock2(x)

        x = x.view(in_size, -1)
        x = self.fc(x)
        return x


model = Net()

# construct loss and optimizer
criterion = torch.nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)


# training cycle forward, backward, update


def train(epoch):
    running_loss = 0.0
    for batch_idx, data in enumerate(train_loader, 0):
        inputs, target = data
        optimizer.zero_grad()

        outputs = model(inputs)
        loss = criterion(outputs, target)
        loss.backward()
        optimizer.step()

        running_loss += loss.item()
        if batch_idx % 300 == 299:
            print('[%d, %5d] loss: %.3f' % (epoch + 1, batch_idx + 1, running_loss / 300))
            running_loss = 0.0


def test():
    correct = 0
    total = 0
    with torch.no_grad():
        for data in test_loader:
            images, labels = data
            outputs = model(images)
            _, predicted = torch.max(outputs.data, dim=1)
            total += labels.size(0)
            correct += (predicted == labels).sum().item()
    print('accuracy on test set: %d %% ' % (100 * correct / total))


if __name__ == '__main__':
    for epoch in range(10):
        train(epoch)
        test()

2.3 Result

[1,   300] loss: 0.544
[1,   600] loss: 0.141
[1,   900] loss: 0.103
accuracy on test set: 96 % 
[2,   300] loss: 0.088
[2,   600] loss: 0.074
[2,   900] loss: 0.071
accuracy on test set: 97 % 
[3,   300] loss: 0.063
[3,   600] loss: 0.061
[3,   900] loss: 0.052
accuracy on test set: 98 % 
[4,   300] loss: 0.046
[4,   600] loss: 0.046
[4,   900] loss: 0.048
accuracy on test set: 98 % 
[5,   300] loss: 0.039
[5,   600] loss: 0.040
[5,   900] loss: 0.041
accuracy on test set: 98 % 
[6,   300] loss: 0.038
[6,   600] loss: 0.036
[6,   900] loss: 0.034
accuracy on test set: 98 % 
[7,   300] loss: 0.027
[7,   600] loss: 0.032
[7,   900] loss: 0.034
accuracy on test set: 98 % 
[8,   300] loss: 0.029
[8,   600] loss: 0.026
[8,   900] loss: 0.029
accuracy on test set: 99 % 
[9,   300] loss: 0.025
[9,   600] loss: 0.024
[9,   900] loss: 0.026
accuracy on test set: 99 % 
[10,   300] loss: 0.022
[10,   600] loss: 0.026
[10,   900] loss: 0.019
accuracy on test set: 99 %