• 🍨 本文为🔗365天深度学习训练营 中的学习记录博客
• 🍖 原作者：K同学啊

1. CycleGAN 简介

CycleGAN (Cycle-Consistent Adversarial Networks) 是一种用于无配对图像到图像转换 (Unpaired Image-to-Image Translation) 的深度学习模型。

• 核心突破 ：传统的图像转换（如 Pix2Pix）需要成对的训练数据（例如同一位置的白天和黑夜照片）。CycleGAN 不需要一一对应的配对数据，只需要两个不同域的数据集（例如一堆莫奈的画和一堆风景照片），就能学习它们之间的风格转换。
• 应用场景 ：风格迁移（照片转油画）、季节转换（夏转冬）、物体转换（马转斑马）等。

2. 核心原理

CycleGAN 通过同时训练两个生成器和两个判别器，并引入 循环一致性损失 (Cycle Consistency Loss) 来保证转换质量。

2.1 模型架构

• 两个生成器 (Generators) ：
  G A B ​ : 将域 A（如莫奈画）转换为域 B（如照片）。
  G B A ​ : 将域 B（如照片）转换为域 A（如莫奈画）。
• 两个判别器 (Discriminators) ：
  D A ​ : 判断图像是真实的 A 还是由 [ o bj ec tO bj ec t ] G B A ​ 生成的伪造 A。
  D B ​ : 判断图像是真实的 B 还是由 [ o bj ec tO bj ec t ] G A B ​ 生成的伪造 B。

2.2 关键损失函数

1. 对抗损失 (Adversarial Loss) ：
   让生成器生成的图像尽可能骗过判别器，使生成的图像看起来逼真。
2. 循环一致性损失 (Cycle Consistency Loss) ：
   • 核心思想 ：如果把一张图片转换过去再转换回来，应该变回原样。
   • 公式 ： [ o bj ec tO bj ec t ] A → G A B ​ ( A ) → G B A ​ ( G A B ​ ( A )) ≈ A
   • 作用 ：防止模型随意改变图像的几何结构和内容（例如，把照片变成油画时，树的位置和形状不能变）。
3. 身份损失 (Identity Loss) (可选)：
   • 如果把域 A 的图片输入给 [ o bj ec tO bj ec t ] G B A ​ （本该把 B 转 A 的生成器），输出应该保持不变。这有助于保持色彩构成的稳定性。

3. 本项目代码结构分析

项目位于 cyclegan/ 目录下，基于 PyTorch 实现。

📂 cyclegan.py (主训练脚本)

项目的入口文件，负责训练流程的控制。

• 参数配置 ：
  • –n_epochs : 总训练轮数（默认 200）。
  • –batch_size : 批次大小（CycleGAN 通常设为 1）。
  • –lambda_cyc : 循环一致性损失权重（默认 10.0，权重很大，说明结构保持很重要）。
• 优化器 ：使用 Adam 优化器。
• 学习率调度 ：前 100 epoch 保持不变，后 100 epoch 线性衰减至 0。

📂 models.py (网络模型)

• 生成器 (GeneratorResNet) ：
  • 基于 ResNet (残差网络) 架构。
  • 包含下采样层、9 个残差块 (Residual Blocks) 和上采样层。
  • 相比 UNet，ResNet 在这种非配对转换任务中能更好地保留深层特征。
• 判别器 (Discriminator) ：
  • 使用 PatchGAN 架构。
  • 输出不是单一的 True/False，而是一个 [ o bj ec tO bj ec t ] N × N 的矩阵，对图像的每个 [ o bj ec tO bj ec t ] 70 × 70 区域进行真假判别，能生成更清晰的纹理细节。

📂 datasets.py (数据处理)

• ImageDataset 类 ：
  • 非对齐加载 ：设置 unaligned=True ，训练时随机从 A 文件夹和 B 文件夹取图，不需要配对。
  • 数据增强 ：包含 Resize (放大)、RandomCrop (随机裁剪) 和 RandomHorizontalFlip (水平翻转)，增加模型鲁棒性。

📂 utils.py (工具辅助)

• ReplayBuffer (经验回放) ：
  • 维护一个缓冲区，存储过去生成的 50 张假图。
  • 训练判别器时，不仅使用当前生成的假图，还随机从缓冲区采样历史假图。这能防止生成器和判别器之间的震荡，使训练更稳定。

4. 运行方法

在终端中进入 cyclegan 目录并运行脚本：

cd 
d:\my_project\python\365\Advanced_Camp\G6_CycleGAN_practi
ce\cyclegan
python cyclegan.py

• 训练过程 ：程序会实时打印 D loss (判别器损失) 和 G loss (生成器损失)。
• 结果查看 ：
  • 生成的样本图片保存在项目根目录的 images/monet2photo/ 下。
  • 模型权重保存在 saved_models/monet2photo/ 下。
    我的文件结构如下：

我的文件目录是这样的。
cyclegan.py程序如下：

import argparse
import itertools
from torchvision.utils import save_image, make_grid
from torch.utils.data import DataLoader
from models import *
from datasets import *
from utils import *
import torch

parser = argparse.ArgumentParser()
parser.add_argument("--epoch", type=int, default=0, help="epoch to start training from")
parser.add_argument("--n_epochs", type=int, default=200, help="number of epochs of training")
parser.add_argument("--dataset_name", type=str, default="monet2photo", help="name of the dataset")
parser.add_argument("--batch_size", type=int, default=1, help="size of the batches")
parser.add_argument("--lr", type=float, default=0.0002, help="adam: learning rate")
parser.add_argument("--b1", type=float, default=0.5, help="adam: decay of first order momentum of gradient")
parser.add_argument("--b2", type=float, default=0.999, help="adam: decay of first order momentum of gradient")
parser.add_argument("--decay_epoch", type=int, default=100, help="epoch from which to start lr decay")
parser.add_argument("--n_cpu", type=int, default=8, help="number of cpu threads to use during batch generation")
parser.add_argument("--img_height", type=int, default=256, help="size of image height")
parser.add_argument("--img_width", type=int, default=256, help="size of image width")
parser.add_argument("--channels", type=int, default=3, help="number of image channels")
parser.add_argument("--sample_interval", type=int, default=100, help="interval between saving generator outputs")
parser.add_argument("--checkpoint_interval", type=int, default=1, help="interval between saving model checkpoints")
parser.add_argument("--n_residual_blocks", type=int, default=9, help="number of residual blocks in generator")
parser.add_argument("--lambda_cyc", type=float, default=10.0, help="cycle loss weight")
parser.add_argument("--lambda_id", type=float, default=5.0, help="identity loss weight")
opt = parser.parse_args()
print(opt)

# Get root path
root_path = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))

# Create sample and checkpoint directories
os.makedirs(os.path.join(root_path, "images/%s" % opt.dataset_name), exist_ok=True)
os.makedirs(os.path.join(root_path, "saved_models/%s" % opt.dataset_name), exist_ok=True)

# Losses
criterion_GAN = torch.nn.MSELoss()
criterion_cycle = torch.nn.L1Loss()
criterion_identity = torch.nn.L1Loss()

cuda = torch.cuda.is_available()

input_shape = (opt.channels, opt.img_height, opt.img_width)

# 初始化生成器鉴别器
G_AB = GeneratorResNet(input_shape, opt.n_residual_blocks)
G_BA = GeneratorResNet(input_shape, opt.n_residual_blocks)
D_A = Discriminator(input_shape)
D_B = Discriminator(input_shape)

if cuda:
    G_AB = G_AB.cuda()
    G_BA = G_BA.cuda()
    D_A = D_A.cuda()
    D_B = D_B.cuda()
    criterion_GAN.cuda()
    criterion_cycle.cuda()
    criterion_identity.cuda()

if opt.epoch != 0:
    # 加载预训练模型
    G_AB.load_state_dict(torch.load(os.path.join(root_path, "saved_models/%s/G_AB_%d.pth" % (opt.dataset_name, opt.epoch))))
    G_BA.load_state_dict(torch.load(os.path.join(root_path, "saved_models/%s/G_BA_%d.pth" % (opt.dataset_name, opt.epoch))))
    D_A.load_state_dict(torch.load(os.path.join(root_path, "saved_models/%s/D_A_%d.pth" % (opt.dataset_name, opt.epoch))))
    D_B.load_state_dict(torch.load(os.path.join(root_path, "saved_models/%s/D_B_%d.pth" % (opt.dataset_name, opt.epoch))))
else:
    # 初始化权重
    G_AB.apply(weights_init_normal)
    G_BA.apply(weights_init_normal)
    D_A.apply(weights_init_normal)
    D_B.apply(weights_init_normal)

# Optimizers
optimizer_G = torch.optim.Adam(
    itertools.chain(G_AB.parameters(), G_BA.parameters()), lr=opt.lr, betas=(opt.b1, opt.b2)
)
optimizer_D_A = torch.optim.Adam(D_A.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
optimizer_D_B = torch.optim.Adam(D_B.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))

# Learning rate update schedulers
lr_scheduler_G = torch.optim.lr_scheduler.LambdaLR(
    optimizer_G, lr_lambda=LambdaLR(opt.n_epochs, opt.epoch, opt.decay_epoch).step
)
lr_scheduler_D_A = torch.optim.lr_scheduler.LambdaLR(
    optimizer_D_A, lr_lambda=LambdaLR(opt.n_epochs, opt.epoch, opt.decay_epoch).step
)
lr_scheduler_D_B = torch.optim.lr_scheduler.LambdaLR(
    optimizer_D_B, lr_lambda=LambdaLR(opt.n_epochs, opt.epoch, opt.decay_epoch).step
)

Tensor = torch.cuda.FloatTensor if cuda else torch.Tensor

# Buffers of previously generated samples
fake_A_buffer = ReplayBuffer()
fake_B_buffer = ReplayBuffer()

# Image transformations
transforms_ = [
    transforms.Resize(int(opt.img_height * 1.12), Image.BICUBIC),
    transforms.RandomCrop((opt.img_height, opt.img_width)),
    transforms.RandomHorizontalFlip(),
    transforms.ToTensor(),
    transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),
]

# Training data loader
dataloader = DataLoader(
    ImageDataset(os.path.join(root_path, "data/%s/" % opt.dataset_name), transforms_=transforms_, unaligned=True),
    batch_size=opt.batch_size,
    shuffle=True,
    num_workers=opt.n_cpu,
)
# Test data loader
val_dataloader = DataLoader(
    ImageDataset(os.path.join(root_path, "data/%s/" % opt.dataset_name), transforms_=transforms_, unaligned=True, mode="test"),
    batch_size=5,
    shuffle=True,
    num_workers=1,
)


def sample_images(batches_done):
    """Saves a generated sample from the test set"""
    imgs = next(iter(val_dataloader))
    G_AB.eval()
    G_BA.eval()
    real_A = Variable(imgs["A"].type(Tensor))
    fake_B = G_AB(real_A)
    real_B = Variable(imgs["B"].type(Tensor))
    fake_A = G_BA(real_B)
    # Arange images along x-axis
    real_A = make_grid(real_A, nrow=5, normalize=True)
    real_B = make_grid(real_B, nrow=5, normalize=True)
    fake_A = make_grid(fake_A, nrow=5, normalize=True)
    fake_B = make_grid(fake_B, nrow=5, normalize=True)
    # Arange images along y-axis
    image_grid = torch.cat((real_A, fake_B, real_B, fake_A), 1)
    save_image(image_grid, os.path.join(root_path, "images/%s/%s.png" % (opt.dataset_name, batches_done)), normalize=False)


# ----------
#  Training
# ----------


if __name__ == '__main__':

    prev_time = time.time()
    for epoch in range(opt.epoch, opt.n_epochs):
        for i, batch in enumerate(dataloader):

            # Set model input
            real_A = Variable(batch["A"].type(Tensor))
            real_B = Variable(batch["B"].type(Tensor))

            # Adversarial ground truths
            valid = Variable(Tensor(np.ones((real_A.size(0), *D_A.output_shape))), requires_grad=False)
            fake  = Variable(Tensor(np.zeros((real_A.size(0), *D_A.output_shape))), requires_grad=False)

            # ------------------
            #  Train Generators
            # ------------------

            G_AB.train()
            G_BA.train()

            optimizer_G.zero_grad()

            # Identity loss
            loss_id_A = criterion_identity(G_BA(real_A), real_A)
            loss_id_B = criterion_identity(G_AB(real_B), real_B)

            loss_identity = (loss_id_A + loss_id_B) / 2

            # GAN loss
            fake_B = G_AB(real_A)
            loss_GAN_AB = criterion_GAN(D_B(fake_B), valid)
            fake_A = G_BA(real_B)
            loss_GAN_BA = criterion_GAN(D_A(fake_A), valid)

            loss_GAN = (loss_GAN_AB + loss_GAN_BA) / 2

            # Cycle loss
            recov_A = G_BA(fake_B)
            loss_cycle_A = criterion_cycle(recov_A, real_A)
            recov_B = G_AB(fake_A)
            loss_cycle_B = criterion_cycle(recov_B, real_B)

            loss_cycle = (loss_cycle_A + loss_cycle_B) / 2

            # Total loss
            loss_G = loss_GAN + opt.lambda_cyc * loss_cycle + opt.lambda_id * loss_identity

            loss_G.backward()
            optimizer_G.step()

            # -----------------------
            #  Train Discriminator A
            # -----------------------

            optimizer_D_A.zero_grad()

            # Real loss
            loss_real = criterion_GAN(D_A(real_A), valid)
            # Fake loss (on batch of previously generated samples)
            fake_A_ = fake_A_buffer.push_and_pop(fake_A)
            loss_fake = criterion_GAN(D_A(fake_A_.detach()), fake)
            # Total loss
            loss_D_A = (loss_real + loss_fake) / 2

            loss_D_A.backward()
            optimizer_D_A.step()

            # -----------------------
            #  Train Discriminator B
            # -----------------------

            optimizer_D_B.zero_grad()

            # Real loss
            loss_real = criterion_GAN(D_B(real_B), valid)
            # Fake loss (on batch of previously generated samples)
            fake_B_ = fake_B_buffer.push_and_pop(fake_B)
            loss_fake = criterion_GAN(D_B(fake_B_.detach()), fake)
            # Total loss
            loss_D_B = (loss_real + loss_fake) / 2

            loss_D_B.backward()
            optimizer_D_B.step()

            loss_D = (loss_D_A + loss_D_B) / 2

            # --------------
            #  Log Progress
            # --------------

            # Determine approximate time left
            batches_done = epoch * len(dataloader) + i
            batches_left = opt.n_epochs * len(dataloader) - batches_done
            time_left = datetime.timedelta(seconds=batches_left * (time.time() - prev_time))
            prev_time = time.time()

            # Print log
            sys.stdout.write(
                "\r[Epoch %d/%d] [Batch %d/%d] [D loss: %f] [G loss: %f, adv: %f, cycle: %f, identity: %f] ETA: %s"
                % (
                    epoch,
                    opt.n_epochs,
                    i,
                    len(dataloader),
                    loss_D.item(),
                    loss_G.item(),
                    loss_GAN.item(),
                    loss_cycle.item(),
                    loss_identity.item(),
                    time_left,
                )
            )

            # If at sample interval save image
            if batches_done % opt.sample_interval == 0:
                sample_images(batches_done)

        # Update learning rates
        lr_scheduler_G.step()
        lr_scheduler_D_A.step()
        lr_scheduler_D_B.step()

        if opt.checkpoint_interval != -1 and epoch % opt.checkpoint_interval == 0:
            # Save model checkpoints
            torch.save(G_AB.state_dict(), os.path.join(root_path, "saved_models/%s/G_AB_%d.pth" % (opt.dataset_name, epoch)))
            torch.save(G_BA.state_dict(), os.path.join(root_path, "saved_models/%s/G_BA_%d.pth" % (opt.dataset_name, epoch)))
            torch.save(D_A.state_dict(), os.path.join(root_path, "saved_models/%s/D_A_%d.pth" % (opt.dataset_name, epoch)))
            torch.save(D_B.state_dict(), os.path.join(root_path, "saved_models/%s/D_B_%d.pth" % (opt.dataset_name, epoch)))

dataset.py程序如下：

import glob
import random
import os

from torch.utils.data import Dataset
from PIL import Image
import torchvision.transforms as transforms


def to_rgb(image):
    rgb_image = Image.new("RGB", image.size)
    rgb_image.paste(image)
    return rgb_image


class ImageDataset(Dataset):
    def __init__(self, root, transforms_=None, unaligned=False, mode="train"):
        self.transform = transforms.Compose(transforms_)
        self.unaligned = unaligned

        self.files_A = sorted(glob.glob(os.path.join(root, "%sA" % mode) + "/*.*"))
        self.files_B = sorted(glob.glob(os.path.join(root, "%sB" % mode) + "/*.*"))

    def __getitem__(self, index):
        image_A = Image.open(self.files_A[index % len(self.files_A)])

        if self.unaligned:
            image_B = Image.open(self.files_B[random.randint(0, len(self.files_B) - 1)])
        else:
            image_B = Image.open(self.files_B[index % len(self.files_B)])

        # Convert grayscale images to rgb
        if image_A.mode != "RGB":
            image_A = to_rgb(image_A)
        if image_B.mode != "RGB":
            image_B = to_rgb(image_B)

        item_A = self.transform(image_A)
        item_B = self.transform(image_B)
        return {"A": item_A, "B": item_B}

    def __len__(self):
        return max(len(self.files_A), len(self.files_B))

model.py程序如下：

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


def weights_init_normal(m):
    classname = m.__class__.__name__
    if classname.find("Conv") != -1:
        torch.nn.init.normal_(m.weight.data, 0.0, 0.02)
        if hasattr(m, "bias") and m.bias is not None:
            torch.nn.init.constant_(m.bias.data, 0.0)
    elif classname.find("BatchNorm2d") != -1:
        torch.nn.init.normal_(m.weight.data, 1.0, 0.02)
        torch.nn.init.constant_(m.bias.data, 0.0)


##############################
#           RESNET
##############################


class ResidualBlock(nn.Module):
    def __init__(self, in_features):
        super(ResidualBlock, self).__init__()

        self.block = nn.Sequential(
            nn.ReflectionPad2d(1),
            nn.Conv2d(in_features, in_features, 3),
            nn.InstanceNorm2d(in_features),
            nn.ReLU(inplace=True),
            nn.ReflectionPad2d(1),
            nn.Conv2d(in_features, in_features, 3),
            nn.InstanceNorm2d(in_features),
        )

    def forward(self, x):
        return x + self.block(x)


class GeneratorResNet(nn.Module):
    def __init__(self, input_shape, num_residual_blocks):
        super(GeneratorResNet, self).__init__()

        channels = input_shape[0]

        # Initial convolution block
        out_features = 64
        model = [
            nn.ReflectionPad2d(channels),
            nn.Conv2d(channels, out_features, 7),
            nn.InstanceNorm2d(out_features),
            nn.ReLU(inplace=True),
        ]
        in_features = out_features

        # Downsampling
        for _ in range(2):
            out_features *= 2
            model += [
                nn.Conv2d(in_features, out_features, 3, stride=2, padding=1),
                nn.InstanceNorm2d(out_features),
                nn.ReLU(inplace=True),
            ]
            in_features = out_features

        # Residual blocks
        for _ in range(num_residual_blocks):
            model += [ResidualBlock(out_features)]

        # Upsampling
        for _ in range(2):
            out_features //= 2
            model += [
                nn.Upsample(scale_factor=2),
                nn.Conv2d(in_features, out_features, 3, stride=1, padding=1),
                nn.InstanceNorm2d(out_features),
                nn.ReLU(inplace=True),
            ]
            in_features = out_features

        # Output layer
        model += [nn.ReflectionPad2d(channels), nn.Conv2d(out_features, channels, 7), nn.Tanh()]

        self.model = nn.Sequential(*model)

    def forward(self, x):
        return self.model(x)


##############################
#        Discriminator
##############################


class Discriminator(nn.Module):
    def __init__(self, input_shape):
        super(Discriminator, self).__init__()

        channels, height, width = input_shape

        # Calculate output shape of image discriminator (PatchGAN)
        self.output_shape = (1, height // 2 ** 4, width // 2 ** 4)

        def discriminator_block(in_filters, out_filters, normalize=True):
            """Returns downsampling layers of each discriminator block"""
            layers = [nn.Conv2d(in_filters, out_filters, 4, stride=2, padding=1)]
            if normalize:
                layers.append(nn.InstanceNorm2d(out_filters))
            layers.append(nn.LeakyReLU(0.2, inplace=True))
            return layers

        self.model = nn.Sequential(
            *discriminator_block(channels, 64, normalize=False),
            *discriminator_block(64, 128),
            *discriminator_block(128, 256),
            *discriminator_block(256, 512),
            nn.ZeroPad2d((1, 0, 1, 0)),
            nn.Conv2d(512, 1, 4, padding=1)
        )

    def forward(self, img):
        return self.model(img)

utils程序如下：

import random
import time
import datetime
import sys

from torch.autograd import Variable
import torch
import numpy as np

from torchvision.utils import save_image


class ReplayBuffer:
    def __init__(self, max_size=50):
        assert max_size > 0, "Empty buffer or trying to create a black hole. Be careful."
        self.max_size = max_size
        self.data = []

    def push_and_pop(self, data):
        to_return = []
        for element in data.data:
            element = torch.unsqueeze(element, 0)
            if len(self.data) < self.max_size:
                self.data.append(element)
                to_return.append(element)
            else:
                if random.uniform(0, 1) > 0.5:
                    i = random.randint(0, self.max_size - 1)
                    to_return.append(self.data[i].clone())
                    self.data[i] = element
                else:
                    to_return.append(element)
        return Variable(torch.cat(to_return))


class LambdaLR:
    def __init__(self, n_epochs, offset, decay_start_epoch):
        assert (n_epochs - decay_start_epoch) > 0, "Decay must start before the training session ends!"
        self.n_epochs = n_epochs
        self.offset = offset
        self.decay_start_epoch = decay_start_epoch

    def step(self, epoch):
        return 1.0 - max(0, epoch + self.offset - self.decay_start_epoch) / (self.n_epochs - self.decay_start_epoch)