生成对抗网络（GAN）原理与实战

1. GAN 原理

复制代码

GAN 结构：
├── 生成器（Generator）：噪声 → 生成图像
├── 判别器（Discriminator）：判断图像真假
└── 训练过程：
    ├── 生成器：生成尽可能真实的图像
    └── 判别器：尽可能区分真假图像
    └── 两者博弈，最终达到纳什均衡

2. DCGAN 实现

python 复制代码

import torch
import torch.nn as nn

class Generator(nn.Module):
    def __init__(self, latent_dim=100):
        super().__init__()
        self.main = nn.Sequential(
            nn.ConvTranspose2d(latent_dim, 512, 4, 1, 0, bias=False),
            nn.BatchNorm2d(512),
            nn.ReLU(True),
            nn.ConvTranspose2d(512, 256, 4, 2, 1, bias=False),
            nn.BatchNorm2d(256),
            nn.ReLU(True),
            nn.ConvTranspose2d(256, 128, 4, 2, 1, bias=False),
            nn.BatchNorm2d(128),
            nn.ReLU(True),
            nn.ConvTranspose2d(128, 3, 4, 2, 1, bias=False),
            nn.Tanh()
        )
    
    def forward(self, x):
        return self.main(x)

class Discriminator(nn.Module):
    def __init__(self):
        super().__init__()
        self.main = nn.Sequential(
            nn.Conv2d(3, 64, 4, 2, 1, bias=False),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Conv2d(64, 128, 4, 2, 1, bias=False),
            nn.BatchNorm2d(128),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Conv2d(128, 256, 4, 2, 1, bias=False),
            nn.BatchNorm2d(256),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Conv2d(256, 1, 4, 1, 0, bias=False),
            nn.Sigmoid()
        )
    
    def forward(self, x):
        return self.main(x).view(-1)

3. 训练循环

python 复制代码

criterion = nn.BCELoss()
optimizer_G = torch.optim.Adam(generator.parameters(), lr=0.0002, betas=(0.5, 0.999))
optimizer_D = torch.optim.Adam(discriminator.parameters(), lr=0.0002, betas=(0.5, 0.999))

for epoch in range(num_epochs):
    for real_images, _ in dataloader:
        batch_size = real_images.size(0)
        real_labels = torch.ones(batch_size)
        fake_labels = torch.zeros(batch_size)
        
        # 训练判别器
        z = torch.randn(batch_size, 100, 1, 1)
        fake_images = generator(z)
        
        d_loss_real = criterion(discriminator(real_images), real_labels)
        d_loss_fake = criterion(discriminator(fake_images.detach()), fake_labels)
        d_loss = (d_loss_real + d_loss_fake) / 2
        
        optimizer_D.zero_grad()
        d_loss.backward()
        optimizer_D.step()
        
        # 训练生成器
        g_loss = criterion(discriminator(fake_images), real_labels)
        
        optimizer_G.zero_grad()
        g_loss.backward()
        optimizer_G.step()

总结

GAN 变体	特点	应用
DCGAN	卷积架构	图像生成
WGAN	Wasserstein 距离	训练稳定
CycleGAN	无配对图像翻译	风格迁移
StyleGAN	风格控制	高质量人脸