LeNet是一种经典的卷积神经网络(CNN)结构,由Yann LeCun等人在1998年提出,主要用于手写数字识别(如MNIST数据集)。作为最早的实用化卷积神经网络,LeNet为现代深度学习模型奠定了基础,其设计思想至今仍被广泛采用。
LeNet由7层组成,包含卷积层、池化层和全连接层:
-
输入层
输入为32x32像素的灰度图像(如手写数字扫描图),经过归一化处理。
-
第一卷积层(C1)
- 使用6个5x5的卷积核,生成6个28x28的特征图。
- 通过局部感受野提取边缘、纹理等低级特征。
- 激活函数最初使用tanh,现代实现中常替换为ReLU。
-
第一池化层(S2)
- 采用平均池化(2x2窗口,步长2),将特征图下采样至14x14。
- 减少计算量并增强平移不变性。
-
第二卷积层(C3)
- 使用16个5x5的卷积核,生成16个10x10的特征图。
- 与前一层的连接并非全连接,而是通过特定组合降低参数量。
-
第二池化层(S4)
- 同样使用平均池化,输出5x5的特征图。
-
全连接层(C5、F6)
- C5层:120个神经元,将空间特征转换为向量。
- F6层:84个神经元,进一步提取高层特征。
- 通常加入Dropout防止过拟合(原版未使用)。
-
输出层
- 10个神经元(对应0-9的分类),使用Softmax激活函数输出概率分布。
python
net = torch.nn.Sequential(
nn.Conv2d(1, 6, kernel_size=5, padding=2), nn.ReLU(), # 第一卷积层
nn.AvgPool2d(kernel_size=2, stride=2), # 第一池化层
nn.Conv2d(6, 16, kernel_size=5), nn.ReLU(), # 第二卷积层
nn.AvgPool2d(kernel_size=2, stride=2), # 第二池化层
nn.Flatten(), # 展平
nn.LazyLinear(120), nn.ReLU(), # 全连接层
nn.Linear(120, 84), nn.ReLU(),
nn.Linear(84, 10) # 输出层
)
使用其进行基于MNIST的训练与识别代码如下:
python
import torch
import torch.nn as nn
import torchvision
import torchvision.transforms as transforms
import time
import matplotlib.pyplot as plt
class Accumulator:
"""在n个变量上累加"""
def __init__(self, n):
self.data = [0.0] * n
def add(self, *args):
self.data = [a + float(b) for a, b in zip(self.data, args)]
def __getitem__(self, idx):
return self.data[idx]
def reset(self):
self.data = [0.0] * len(self.data)
class Timer:
"""记录多次运行时间"""
def __init__(self):
self.times = []
self.start()
def start(self):
"""启动计时器"""
self.tik = time.time()
def stop(self):
"""停止计时器并将时间记录在列表中"""
self.times.append(time.time() - self.tik)
return self.times[-1]
def avg(self):
"""返回平均时间"""
return sum(self.times) / len(self.times)
def sum(self):
"""返回时间总和"""
return sum(self.times)
class Animator:
"""绘制训练数据折线图"""
def __init__(self, xlabel=None, ylabel=None, legend=None, xlim=None,
ylim=None, xscale='linear', yscale='linear',
fmts=('-', 'm--', 'g-.', 'r:'), nrows=1, ncols=1,
figsize=(3.5, 2.5)):
# 增量地绘制多条线
if legend is None:
legend = []
self.fig, self.axes = plt.subplots(nrows, ncols, figsize=figsize)
if nrows * ncols == 1:
self.axes = [self.axes, ]
# 使用lambda函数捕获参数
self.config_axes = lambda: self.set_axes(
self.axes[0], xlabel, ylabel, xlim, ylim, xscale, yscale, legend)
self.X, self.Y, self.fmts = None, None, fmts
def set_axes(self, axes, xlabel, ylabel, xlim, ylim, xscale, yscale, legend):
"""设置matplotlib的轴"""
axes.set_xlabel(xlabel)
axes.set_ylabel(ylabel)
axes.set_xscale(xscale)
axes.set_yscale(yscale)
axes.set_xlim(xlim)
axes.set_ylim(ylim)
if legend:
axes.legend(legend)
axes.grid()
def add(self, x, y):
"""向图表中添加多个数据点"""
if not hasattr(y, "__len__"):
y = [y]
n = len(y)
if not hasattr(x, "__len__"):
x = [x] * n
if not self.X:
self.X = [[] for _ in range(n)]
if not self.Y:
self.Y = [[] for _ in range(n)]
for i, (a, b) in enumerate(zip(x, y)):
if a is not None and b is not None:
self.X[i].append(a)
self.Y[i].append(b)
self.axes[0].cla()
for x, y, fmt in zip(self.X, self.Y, self.fmts):
self.axes[0].plot(x, y, fmt)
self.config_axes()
self.fig.show()
def load_data_fashion_mnist(batch_size, resize=None):
"""下载Fashion-MNIST数据集,然后将其加载到内存中"""
trans = [transforms.ToTensor()]
if resize:
trans.insert(0, transforms.Resize(resize))
trans = transforms.Compose(trans)
mnist_train = torchvision.datasets.FashionMNIST(
root="../data", train=True, transform=trans, download=True)
mnist_test = torchvision.datasets.FashionMNIST(
root="../data", train=False, transform=trans, download=True)
train_iter = torch.utils.data.DataLoader(
mnist_train, batch_size, shuffle=True, num_workers=4)
test_iter = torch.utils.data.DataLoader(
mnist_test, batch_size, shuffle=False, num_workers=4)
return train_iter, test_iter
def accuracy(y_hat, y):
"""计算预测正确的数量"""
if len(y_hat.shape) > 1 and y_hat.shape[1] > 1:
y_hat = y_hat.argmax(axis=1)
cmp = y_hat.type(y.dtype) == y
return float(cmp.type(y.dtype).sum())
def evaluate_accuracy_gpu(net, data_iter, device=None):
if isinstance(net, nn.Module):
net.eval()
if not device:
device = next(iter(net.parameters())).device
# 正确预测的数量,总预测的数量
metric = Accumulator(2)
with torch.no_grad():
for X, y in data_iter:
if isinstance(X, list):
X = [x.to(device) for x in X]
else:
X = X.to(device)
y = y.to(device)
metric.add(accuracy(net(X), y), y.numel())
return metric[0] / metric[1]
def train(net, train_iter, test_iter, num_epochs, lr, device):
def init_weights(m):
if type(m) == nn.Linear or type(m) == nn.Conv2d:
nn.init.xavier_uniform_(m.weight)
net.apply(init_weights)
print('training on', device)
net.to(device)
optimizer = torch.optim.SGD(net.parameters(), lr=lr)
loss = nn.CrossEntropyLoss()
animator = Animator(xlabel='epoch', xlim=[1, num_epochs],
legend=['train loss', 'train acc', 'test acc'])
timer, num_batches = Timer(), len(train_iter)
for epoch in range(num_epochs):
# 训练损失之和,训练准确率之和,样本数
metric = Accumulator(3)
net.train()
for i, (X, y) in enumerate(train_iter):
timer.start()
optimizer.zero_grad()
X, y = X.to(device), y.to(device)
y_hat = net(X)
l = loss(y_hat, y)
l.backward()
optimizer.step()
with torch.no_grad():
metric.add(l * X.shape[0], accuracy(y_hat, y), X.shape[0])
timer.stop()
train_l = metric[0] / metric[2]
train_acc = metric[1] / metric[2]
if (i + 1) % (num_batches // 5) == 0 or i == num_batches - 1:
animator.add(epoch + (i + 1) / num_batches,
(train_l, train_acc, None))
test_acc = evaluate_accuracy_gpu(net, test_iter)
animator.add(epoch + 1, (None, None, test_acc))
print(f'loss {train_l:.3f}, train acc {train_acc:.3f}, '
f'test acc {test_acc:.3f}')
print(f'{metric[2] * num_epochs / timer.sum():.1f} examples/sec '
f'on {str(device)}')
class Reshape(torch.nn.Module):
def forward(self, x):
return x.view(-1, 1, 28, 28)
net = torch.nn.Sequential(
Reshape(),
nn.Conv2d(1, 6, kernel_size=5, padding=2), nn.Sigmoid(),
nn.AvgPool2d(kernel_size=2, stride=2),
nn.Conv2d(6, 16, kernel_size=5), nn.Sigmoid(),
nn.AvgPool2d(kernel_size=2, stride=2),
nn.Flatten(),
nn.LazyLinear(120), nn.Sigmoid(),
nn.Linear(120, 84), nn.Sigmoid(),
nn.Linear(84, 10)
) # LeNet基本架构,经过两组卷积-池化后展平并进行全连接
batch_size = 256
train_iter, test_iter = load_data_fashion_mnist(batch_size=batch_size)
lr, num_epochs = 0.9, 10
train(net, train_iter, test_iter, num_epochs, lr, 'cuda:0')
LeNet验证了CNN在图像任务中的有效性,启发了后续模型(如AlexNet、VGG)。尽管现代网络更复杂,但其"卷积-池化-全连接"的基础架构仍源于LeNet。它标志着神经网络从理论走向实际应用,是深度学习发展的重要里程碑。