我的P66实训记录和读书报告

`import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import DataLoader
from torchvision import datasets, transforms
import matplotlib.pyplot as plt

transform_train = transforms.Compose([
transforms.RandomCrop(32, padding=4),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465),
(0.2023, 0.1994, 0.2010))
])

transform_test = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465),
(0.2023, 0.1994, 0.2010))
])

train_dataset = datasets.CIFAR10(
root='./data', train=True, download=True, transform=transform_train
)
test_dataset = datasets.CIFAR10(
root='./data', train=False, download=True, transform=transform_test
)

train_loader = DataLoader(
train_dataset, batch_size=128, shuffle=True, num_workers=2
)
test_loader = DataLoader(
test_dataset, batch_size=128, shuffle=False, num_workers=2
)

classes = ('plane', 'car', 'bird', 'cat', 'deer',
'dog', 'frog', 'horse', 'ship', 'truck')

class SimpleCNN(nn.Module):
def init(self):
super(SimpleCNN, self).init()# 卷积层：3输入通道（RGB），32输出通道，5x5卷积核
self.conv1 = nn.Conv2d(3, 32, 5, padding=2)# 池化层：2x2最大池化
self.pool = nn.MaxPool2d(2, 2) # 卷积层：32输入通道，64输出通道
self.conv2 = nn.Conv2d(32, 64, 5, padding=2) # 全连接层：展平后连接1024神经元
self.fc1 = nn.Linear(64 * 8 * 8, 1024) # 32/2/2=8（两次池化） # 全连接层：输出10类（CIFAR-10）
self.fc2 = nn.Linear(1024, 10) # ReLU激活函数
self.relu = nn.ReLU()

def forward(self, x):
    x = self.pool(self.relu(self.conv1(x)))  # 卷积+激活+池化
    x = self.pool(self.relu(self.conv2(x)))
    x = x.view(-1, 64 * 8 * 8)  # 展平特征图
    x = self.relu(self.fc1(x))
    x = self.fc2(x)
    return x

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model = SimpleCNN().to(device)

criterion = nn.CrossEntropyLoss()
optimizer = optim.Adam(model.parameters(), lr=0.001)

epochs = 20

train_losses = []
train_accs = []
test_accs = []

for epoch in range(epochs):
model.train() # 训练模式
running_loss = 0.0
correct = 0
total = 0

for i, data in enumerate(train_loader, 0):
    inputs, labels = data[0].to(device), data[1].to(device)
    optimizer.zero_grad()
    outputs = model(inputs)
    loss = criterion(outputs, labels)
    loss.backward()
    optimizer.step()
    running_loss += loss.item()
    _, predicted = torch.max(outputs.data, 1)
    total += labels.size(0)
    correct += (predicted == labels).sum().item()
    if i % 100 == 99:
        print(f'[{epoch + 1}, {i + 1}] loss: {running_loss / 100:.3f}')
        running_loss = 0.0

train_acc = 100 * correct / total
train_losses.append(running_loss / len(train_loader))
train_accs.append(train_acc)

model.eval()  # 评估模式
correct = 0
total = 0
with torch.no_grad():  # 关闭梯度计算
    for data in test_loader:
        images, labels = data[0].to(device), data[1].to(device)
        outputs = model(images)
        _, predicted = torch.max(outputs.data, 1)
        total += labels.size(0)
        correct += (predicted == labels).sum().item()

test_acc = 100 * correct / total
test_accs.append(test_acc)
print(f'Epoch {epoch+1} - 训练准确率: {train_acc:.2f}%  测试准确率: {test_acc:.2f}%')

print('训练完成')

model.eval()
correct = 0
total = 0
with torch.no_grad():
for data in test_loader:
images, labels = data[0].to(device), data[1].to(device)
outputs = model(images)
_, predicted = torch.max(outputs.data, 1)
total += labels.size(0)
correct += (predicted == labels).sum().item()

print(f'测试集整体准确率: {100 * correct / total:.2f}%')

class_correct = list(0. for i in range(10))
class_total = list(0. for i in range(10))
with torch.no_grad():
for data in test_loader:
images, labels = data[0].to(device), data[1].to(device)
outputs = model(images)
_, predicted = torch.max(outputs, 1)
c = (predicted == labels).squeeze()
for label, prediction in zip(labels, c):
class_correct[label] += prediction.item()
class_total[label] += 1

for i in range(10):
print(f'类别 {classes[i]} 的准确率: {100 * class_correct[i] / class_total[i]:.2f}%')

plt.figure(figsize=(12, 4))

损失曲线

plt.subplot(1, 2, 1)
plt.plot(train_losses, label='训练损失')
plt.title('训练损失曲线')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.legend()

准确率曲线

plt.subplot(1, 2, 2)
plt.plot(train_accs, label='训练准确率')
plt.plot(test_accs, label='测试准确率')
plt.title('准确率曲线')
plt.xlabel('Epoch')
plt.ylabel('准确率 (%)')
plt.legend()

plt.tight_layout()
plt.show()`

我认为这份教程最核心的价值，在于打破了我对 CNN 的 “陌生感”。它没有抽象的理论推导，而是用通俗表达拆解 CNN 核心层：比如让卷积层 “提取图像特征”、池化层 “简化数据” 的作用变得直观，连激活函数的 “非线性转换” 也借实例褪去晦涩，也许能让我快速抓住本质？。更关键的是 “理论 + 实操” 的设计，这让 “理解” 不再停留在概念层面 —— 搭模型、跑算法的过程，能让人真切感受到各层如何协同工作，从 “知道是什么” 到 “明白怎么用”。最终，它不仅传递了 CNN 知识，更帮我搭建了一些cnn的框架，让我知道了一些知识点，能主动思考任务与技术的匹配，真正实现了 “入门” 的意义。