colab.pynb

import torch
from torchvision import datasets, transforms
from torch.utils.data import dataloader

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

# 定义残差块
class ResBlk(nn.Module):
    def __init__(self, ch_in, ch_out, stride):
        super(ResBlk, self).__init__()
        self.conv1 = nn.Conv2d(ch_in, ch_out, kernel_size=3, stride=stride, 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(
            nn.Conv2d(ch_in, ch_out, kernel_size=1, stride=stride),
            nn.BatchNorm2d(ch_out)
        )

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

# 定义ResNet18网络结构
class ResNet18(nn.Module):
    def __init__(self):
        super(ResNet18, self).__init__()
        self.conv1 = nn.Sequential(
            nn.Conv2d(3, 64, kernel_size=3, stride=3, padding=0),
            nn.BatchNorm2d(64)
        )
        self.blk1 = ResBlk(64, 64, stride=2)
        self.blk2 = ResBlk(64, 128, stride=2)
        self.blk3 = ResBlk(128, 256, stride=2)
        self.blk4 = ResBlk(256, 512, stride=2)

        self.outlayer = nn.Linear(512*1*1, 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)

        x = F.adaptive_avg_pool2d(x, [1, 1])
        x = x.view(x.size(0), -1)
        x = self.outlayer(x)

        return x

batch_sz = 128
learn_rate = 1e-3
train_num = 1000

# 导入训练集数据
train_data = dataloader.DataLoader(
    datasets.CIFAR10(root='data/', train=True, transform=transforms.Compose([
        transforms.Resize((32, 32)),      # 重新设置图片大小
        transforms.ToTensor(),      # 将图片转化为tensor
        transforms.Normalize(mean=[0.485, 0.456, 0.406],
                             std=[0.229, 0.224, 0.225])         # 进行归一化
    ]), download=True), shuffle=True, batch_size=batch_sz
)

# 导入测试集数据
train_test = dataloader.DataLoader(
    datasets.CIFAR10(root='data/', train=False, transform=transforms.Compose([
        transforms.Resize((32, 32)),
        transforms.ToTensor(),
        transforms.Normalize(mean=[0.485, 0.456, 0.406],
                             std=[0.229, 0.224, 0.225])
    ]), download=True), shuffle=True, batch_size=batch_sz
)

device = torch.device('cuda')           # 设置在GPU上进行训练

# 定义模型
model = ResNet18().to(device)

# 定义损失函数和优化方式
criteon = nn.CrossEntropyLoss().to(device)
optimizer = torch.optim.Adam(model.parameters(), lr=learn_rate)

# 训练模型
for epoch in range(10):
    model.train()
    for batch_idx, (x, label) in enumerate(train_data):
        x = x.to(device)
        label = label.to(device)

        logits = model(x)       # 经过模型得到的数据

        loss = criteon(logits, label)
        #print('logits:', logits[0])
        #print('label:', label[0])
        # 反向传播
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()
        if batch_idx == len(train_data) - 1:
            print(epoch, 'loss:', loss.item())

    # 进行测试
    model.eval()
    with torch.no_grad():
        total_correct = 0
        total_num = 0
        for x, label in train_test:
            x = x.to(device)
            label = label.to(device)

            logits = model(x)

            pred = logits.argmax(dim=1)

            correct = torch.eq(pred, label).float().sum().item()
            total_correct += correct
            total_num += x.size(0)

        acc = total_correct / total_num
        print('完成第{}轮!'.format(epoch+1))
        print(epoch, 'test acc测试准确率:', acc)