【深度学习】卷积网络代码实战ResNet

ResNet (Residual Network) 是由微软研究院的何凯明等人在2015年提出的一种深度卷积神经网络结构。ResNet的设计目标是解决深层网络训练中的梯度消失和梯度爆炸问题，进一步提高网络的表现。下面是一个ResNet模型实现，使用PyTorch框架来展示如何实现基本的ResNet结构。这个例子包括了一个基本的残差块（Residual Block）以及ResNet-18的实现，代码结构分为model.py（模型文件）和train.py（训练文件）。

model.py

首先，我们导入所需要的包

python 复制代码

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

然后，定义Resnet Block（ResBlk）类。

python 复制代码

class ResBlk(nn.Module):
    def __init__(self):
        super(ResBlk, self).__init__()
        self.conv1 = nn.Conv2d(ch_in, ch_out, kernel_size=3, stride=1, padding=1)
        self.bn1 = nn.BatchNorm2d(ch_out)
        self.conv2 = nn.Conv2d(ch_out, ch_out, kernel_size=3, stride=1, padding=1)
        self.bn2 = nn.BatchNorm2d(ch_out)

        self.extra = nn.Sequential()
        if ch_out != ch_in
            self.extra = nn.Sequential(
                nn.Conv2d(ch_in, ch_out, kernel_size=3, stride=1)
                nn.BatchNorm2d(ch_out)
            )

    def forward(self, x):
        out = F.relu(self.bn1(self.conv1(x)))
        out = F.relu(self.bn2(self.conv2(x)))
        out = self.extra(x) + out
        return out

最后，根据ResNet18的结构对ResNet Block进行堆叠。

python 复制代码

class Resnet18(nn.Module):
    def __init__(self):
        super(Resnet18, self).__init__()
        self.conv1 = nn.Sequential(
            nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1)
            nn.BatchNorm2d(64)
        )
        self.blk1 = ResBlk(64, 128)
        self.blk2 = ResBlk(128, 256)
        self.blk3 = ResBlk(256, 512)
        self.blk4 = ResBlk(512, 1024)
        self.outlayer = nn.Linear(512, 10)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        x = self.blk1(x)
        x = self.blk2(x)
        x = self.blk3(x)
        x = self.blk4(x)
        
        # print('after conv1:', x.shape)
        x = F.adaptive_avg_pool2d(x, [1,1])
        x = x.view(x.size(0), -1)
        x = self.outlayer(x)
        return x

其中，在网络结构搭建过程中，需要用到中间阶段的图片参数，用下述测试过程求得。

python 复制代码

def main():
    tmp = torch.randn(2, 3, 32, 32)
    out = blk(tmp)
    print('block', out.shape)
    
    x = torch.randn(2, 3, 32, 32)
    model = ResNet18()
    out = model(x)
    print('resnet:', out.shape)

train.py

首先，导入所需要的包

python 复制代码

import torch
from torchvision import datasets
from torchvision import transforms
from torch import nn, optimizer

然后，定义main()函数

python 复制代码

def main():
    batchsz = 32
    cifar_train = datasets.CIFAR10('cifar', True, transform=transforms.Compose([
        transforms.Resize((32, 32)),
        transforms.ToTensor()
        ]), download=True)
    cifar_train = DataLoader(cifar_train, batch_size=batchsz, shuffle=True)
    cifar_test = datasets.CIFAR10('cifar', False, transform=transforms.Compose([
        transforms.Resize((32, 32)),
        transforms.ToTensor()
        ]), download=True)
    cifar_test = DataLoader(cifar_test, batch_size=batchsz, shuffle=True)
 
    x, label = iter(cifar_train).next()
    print('x:', x.shape, 'label:', label.shape)
    
    device = torch.device('cuda')
    model = ResNet18().to(device)
    criteon = nn.CrossEntropyLoss()
    optimizer = optim.Adam(model.parameters(), lr=1e-3)
    print(model)
 
    for epoch in range(100):
        for batchidx, (x, label) in enumerate(cifar_train):
            x, label = x.to(device), label.to(device)
            logits = model(x)
            loss = criteon(logitsm label)
    
            optimizer.zero_grad()
            loss.backward()
            optimizer.step()
    print(loss.item())
 
 
    with torch.no_grad():
        total_correct = 0
        total_num = 0
        for x, label in cifar_test:
            x, label = x.to(device), label.to(device)
            logits = model(x)
            pred = logits.argmax(dim=1)
            total_correct += torch.eq(pred, label).floot().sum().item()
            total_num += x.size(0)
 
        acc = total_correct / total_num
        print(epoch, acc)