PyTorch深度学习实践第十一讲(完结）---卷积神经网络（高级篇）

Vinsada

已于 2022-10-20 16:50:52 修改

阅读量361

点赞数

分类专栏： Pytorch框架实践文章标签：深度学习 pytorch cnn

于 2022-10-20 16:40:45 首次发布

本文链接：https://blog.csdn.net/weixin_43224291/article/details/127421970

版权

Pytorch框架实践专栏收录该内容

11 篇文章 0 订阅

订阅专栏

Demo 11 :卷积神经网络高级篇
来源：B站刘二大人

本讲简介：在基础篇的基础上，接触经典的神经网络，并动手实现。
本讲目的：掌握复现大部分论文中出现的，自己没有见过的神经网络的能力

GoogleNet
在这里插入图片描述

网络中的Inception Moudel 说明：

卷积核超参数选择困难，自动找到卷积的最佳组合。
1x1卷积核的作用：进行不同通道的信息融合。使用1x1卷积核虽然参数量增加了，但是能够显著的降低计算量(operations) 传送：1 x 1 卷积核作用
nception Moudel由4个分支组成。要分清哪些是在Init里定义，哪些是在forward里调用。4个分支在dim=1(channels)上进行concatenate,通道数为24+16+24+24 = 88
GoogleNet的Inception(Pytorch实现)

在这里插入图片描述

代码说明：

使用类对Inception Module进行封装
先是1个卷积层(conv,maxpooling,relu)，然后inceptionA模块(输出的channels是24+16+24+24=88)，接下来又是一个卷积层(conv,mp,relu),然后inceptionA模块，最后一个全连接层(fc)。
408这个数据可以通过x = x.view(in_size, -1)后调用x.shape得到。

# demo 11
# 实现 inception module

import torch
import torch.nn as nn
from torchvision import datasets
from torchvision import transforms
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))])

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


# design model

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)  # 池化分支的1x1卷积核，不是池化卷积核

    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, branch3x3, branch5x5, branch_pool]
        return torch.cat(outputs, dim=1)  # b, c, w, h.  dim = 1 对应的是通道数


class Net(torch.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)
        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)


# train cycle

def train(epoch):
    running_loss = 0.0
    for batch_idx, data in enumerate(train_loader, 0):
        inputs, targets = data
        outputs = model(inputs)
        loss = criterion(outputs, targets)
        optimizer.zero_grad()

        loss.backward()
        optimizer.step()

        running_loss += loss
        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:
            inputs, labels = data
            outputs = model(inputs)
            _, 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()

接下来，GO DEEPER
在这里插入图片描述

ResNet

为什么网络层数更深反而效果更差：
梯度消失：在反向传播时需要根据链式法则把一连串的梯度乘起来，若每个梯度都小于1，则乘起来的结果会接近于0，导致权重在更新时得不到什么更新，进而导致最开始的这些块（离输入近的块）没办法得到充分的训练。梯度消失和梯度爆炸
Residual block：偏导数大于1，不会出现梯度消失
Resnet中为什么有虚线的跳连接：因为输入到该块的张量大小和输出的张量大小不一致，需要做特殊处理

说明：

要解决的问题：梯度消失
跳连接：H(x) = F(x) + x,张量维度必须一样，加完后再激活。不要做pooling，张量的维度会发生变化。

代码说明：
网络先是1个卷积层(conv,maxpooling,relu)，然后ResidualBlock模块，接下来又是一个卷积层(conv,mp,relu),然后ResidualBlock模块模块，最后一个全连接层(fc)。

import torch
import torch.nn as nn
from torchvision import datasets
from torchvision import transforms
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,))])

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


# design model
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)  # y : 竖直的分支

        return x + y  # x :跳跃的分支


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 = nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)


# train cycle
def train(epoch):
    running_loss = 0.0
    for batch_idx, data in enumerate(train_loader, 0):
        inputs, labels = data
        outputs = model(inputs)
        optimizer.zero_grad()

        loss = criterion(outputs, labels)
        loss.backward()
        optimizer.step()

        running_loss += loss
        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
    for data in test_loader:
        inputs, labels = data
        outputs = model(inputs)
        _, predicted = torch.max(outputs.data, dim=1)

        with torch.no_grad():
            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()