GAN（Generative Adversarial Network）的复现

永不磨灭的fw

于 2023-04-10 23:14:47 发布

阅读量458

点赞数

文章标签：生成对抗网络深度学习 python

本文链接：https://blog.csdn.net/zjj666123/article/details/130072232

版权

GAN（Generative Adversarial Network）的复现

代码的复现是基于PyTorch-GAN/gan.py at master · eriklindernoren/PyTorch-GAN (github.com)，在一个新的数据集完成了复现

#gan.py 代码只要环境没问题是可以直接运行的
import argparse
import os
import numpy as np
import math

import torchvision.transforms as transforms
from torchvision.utils import save_image

from torch.utils.data import DataLoader
from torchvision import datasets
from torch.autograd import Variable

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

os.makedirs("images", exist_ok=True)

parser = argparse.ArgumentParser()
parser.add_argument("--n_epochs", type=int, default=200, help="number of epochs of training")
parser.add_argument("--batch_size", type=int, default=64, 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("--n_cpu", type=int, default=8, help="number of cpu threads to use during batch generation")
parser.add_argument("--latent_dim", type=int, default=100, help="dimensionality of the latent space")#论文中提到的Z，即生成满足高斯分布的随机噪声，是生成模型的输入。
parser.add_argument("--img_size", type=int, default=28, help="size of each image dimension")#图片的尺寸
parser.add_argument("--channels", type=int, default=1, help="number of image channels")#图片的通道数
parser.add_argument("--sample_interval", type=int, default=400, help="interval betwen image samples")#每400个iteration打印一组图片
opt = parser.parse_args()
print(opt)

img_shape = (opt.channels, opt.img_size, opt.img_size)

cuda = True if torch.cuda.is_available() else False


class Generator(nn.Module):#生成函数，从100维度的噪声中生成1*28*28的手写数字图片
    def __init__(self):
        super(Generator, self).__init__()

        def block(in_feat, out_feat, normalize=True):
            layers = [nn.Linear(in_feat, out_feat)]
            if normalize:
                layers.append(nn.BatchNorm1d(out_feat, 0.8))
            layers.append(nn.LeakyReLU(0.2, inplace=True))
            return layers

        self.model = nn.Sequential(
            *block(opt.latent_dim, 128, normalize=False),
            *block(128, 256),
            *block(256, 512),
            *block(512, 1024),
            nn.Linear(1024, int(np.prod(img_shape))),
            nn.Tanh()
        )

    def forward(self, z):
        img = self.model(z)
        img = img.view(img.size(0), *img_shape)
        return img


class Discriminator(nn.Module):#判别模型，目的是判别图片是生成的（0）还是真实的（1）
    def __init__(self):
        super(Discriminator, self).__init__()

        self.model = nn.Sequential(
            nn.Linear(int(np.prod(img_shape)), 512),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Linear(512, 256),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Linear(256, 1),
            nn.Sigmoid(),
        )

    def forward(self, img):
        img_flat = img.view(img.size(0), -1)
        validity = self.model(img_flat)

        return validity


# Loss function
adversarial_loss = torch.nn.BCELoss()#二分类交叉熵

# Initialize generator and discriminator
generator = Generator()
discriminator = Discriminator()

if cuda:
    generator.cuda()
    discriminator.cuda()
    adversarial_loss.cuda()

# Configure data loader
os.makedirs("../../data/mnist", exist_ok=True)
dataloader = torch.utils.data.DataLoader(#读入手写数字识别的数据
    datasets.MNIST(
        "../../data/mnist",
        train=True,
        download=True,
        transform=transforms.Compose(
            [transforms.Resize(opt.img_size), transforms.ToTensor(), transforms.Normalize([0.5], [0.5])]
        ),
    ),
    batch_size=opt.batch_size,
    shuffle=True,
)

# Optimizers
optimizer_G = torch.optim.Adam(generator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
optimizer_D = torch.optim.Adam(discriminator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))

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

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

for epoch in range(opt.n_epochs):
    for i, (imgs, _) in enumerate(dataloader):#_表示标签，GAN不需要使用手写数字识别的标签

        # Adversarial ground truths
        valid = Variable(Tensor(imgs.size(0), 1).fill_(1.0), requires_grad=False)#真实图片的标签，全为1
        fake = Variable(Tensor(imgs.size(0), 1).fill_(0.0), requires_grad=False)#生成图片的标签，全为0

        # Configure input
        real_imgs = Variable(imgs.type(Tensor))

        # -----------------
        #  Train Generator
        # -----------------

        optimizer_G.zero_grad()

        # Sample noise as generator input
        z = Variable(Tensor(np.random.normal(0, 1, (imgs.shape[0], opt.latent_dim))))

        # Generate a batch of images
        gen_imgs = generator(z)

        # Loss measures generator's ability to fool the discriminator
        g_loss = adversarial_loss(discriminator(gen_imgs), valid)

        g_loss.backward()
        optimizer_G.step()

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

        optimizer_D.zero_grad()

        # Measure discriminator's ability to classify real from generated samples
        real_loss = adversarial_loss(discriminator(real_imgs), valid)
        fake_loss = adversarial_loss(discriminator(gen_imgs.detach()), fake)
        d_loss = (real_loss + fake_loss) / 2

        d_loss.backward()
        optimizer_D.step()

        print(
            "[Epoch %d/%d] [Batch %d/%d] [D loss: %f] [G loss: %f]"
            % (epoch, opt.n_epochs, i, len(dataloader), d_loss.item(), g_loss.item())
        )

        batches_done = epoch * len(dataloader) + i
        if batches_done % opt.sample_interval == 0:
            save_image(gen_imgs.data[:25], "images/%d.png" % batches_done, nrow=5, normalize=True)

使用了一个新的数据集anime，数据集的链接在下方，数据集一共有5w多张动漫头像，都是 $3 * 96 * 96$ 的。

链接：https://pan.baidu.com/s/1rHwdw22bl-l5ID3OJMsldw
提取码：hb3j

首先为数据集建造dataset类，代码如下

#dataset.py
import torch
import math
import numpy as np
import random
import os
from PIL import Image,ImageFile
from torchvision import transforms
import torch.utils.data as Data

Transformes = transforms.Compose([
    transforms.Resize(96), transforms.ToTensor(), transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])
])#将图像进行归一化操作时，如果使用ImageNet的([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])，最终效果会变差，可能的原因是，这样会跟D生成模型带来误导，无法输出正确的颜色

class anime_dataset(torch.utils.data.Dataset):
    def __init__(self,path,shuffle=True):
        self.path = path#数据集的路径
        self.list = os.listdir(self.path)
        self.lenth = len(self.list)
    def __getitem__(self,index):
        image_name=os.path.join(self.path,self.list[index])
        image=Image.open(image_name).convert('RGB')
        image=Transformes(image)
        return image

    def __len__(self):
        return self.lenth
if __name__=='__main__':
    path = 'anime/'
    dataset = anime_dataset(path)
    print(dataset[0])

训练的代码如下，详细修改见代码注释

#gan_amine.py
import argparse
import os
import numpy as np
import math

import torchvision.transforms as transforms
from torchvision.utils import save_image

from torch.utils.data import DataLoader
from torchvision import datasets
from torch.autograd import Variable
import dataset#引用数据集处理的包（上一个代码块）

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

os.makedirs("image_anime", exist_ok=True)#输出图片路径

parser = argparse.ArgumentParser()
parser.add_argument("--n_epochs", type=int, default=200, help="number of epochs of training")
parser.add_argument("--batch_size", type=int, default=64, 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("--n_cpu", type=int, default=8, help="number of cpu threads to use during batch generation")
parser.add_argument("--latent_dim", type=int, default=100, help="dimensionality of the latent space")
parser.add_argument("--img_size", type=int, default=96, help="size of each image dimension")#修改图像大小
parser.add_argument("--channels", type=int, default=3, help="number of image channels")#修改通道数
parser.add_argument("--sample_interval", type=int, default=400, help="interval betwen image samples")
opt = parser.parse_args()
path = 'anime/'#数据集路径
print(opt)

img_shape = (opt.channels, opt.img_size, opt.img_size)

cuda = True if torch.cuda.is_available() else False


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

        def block(in_feat, out_feat, normalize=True):
            layers = [nn.Linear(in_feat, out_feat)]
            if normalize:
                layers.append(nn.BatchNorm1d(out_feat, 0.8))
            layers.append(nn.LeakyReLU(0.2, inplace=True))
            return layers

        self.model = nn.Sequential(
            *block(opt.latent_dim, 128, normalize=False),
            *block(128, 256),
            *block(256, 512),
            *block(512, 1024),
            nn.Linear(1024, int(np.prod(img_shape))),
            nn.Tanh()
        )

    def forward(self, z):
        img = self.model(z)
        img = img.view(img.size(0), *img_shape)
        return img


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

        self.model = nn.Sequential(
            nn.Linear(int(np.prod(img_shape)), 512),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Linear(512, 256),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Linear(256, 1),
            nn.Sigmoid(),
        )

    def forward(self, img):
        img_flat = img.view(img.size(0), -1)
        validity = self.model(img_flat)

        return validity


# Loss function
adversarial_loss = torch.nn.BCELoss()

# Initialize generator and discriminator
generator = Generator()
discriminator = Discriminator()

if cuda:
    generator.cuda()
    discriminator.cuda()
    adversarial_loss.cuda()

# Configure data loader
# os.makedirs("../../data/mnist", exist_ok=True)

dataset_a = dataset.anime_dataset(path)#读入数据


dataloader = torch.utils.data.DataLoader(
    dataset_a,
    batch_size=opt.batch_size,
    shuffle=True,
)

# Optimizers
optimizer_G = torch.optim.Adam(generator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
optimizer_D = torch.optim.Adam(discriminator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))

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

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

for epoch in range(opt.n_epochs):
    for i, (imgs) in enumerate(dataloader):
        

        # Adversarial ground truths
        valid = Variable(Tensor(imgs.size(0), 1).fill_(1.0), requires_grad=False)
        fake = Variable(Tensor(imgs.size(0), 1).fill_(0.0), requires_grad=False)
        # print(valid.shape)
        # print(fake.shape)
        # exit(0)
        # Configure input
        real_imgs = Variable(imgs.type(Tensor))

        # -----------------
        #  Train Generator
        # -----------------

        optimizer_G.zero_grad()

        # Sample noise as generator input
        z = Variable(Tensor(np.random.normal(0, 1, (imgs.shape[0], opt.latent_dim))))

        # Generate a batch of images
        gen_imgs = generator(z)
        # print(gen_imgs.shape)64*3*96*96
        # exit(0)

        # Loss measures generator's ability to fool the discriminator
        g_loss = adversarial_loss(discriminator(gen_imgs), valid)

        g_loss.backward()
        optimizer_G.step()

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

        optimizer_D.zero_grad()

        # Measure discriminator's ability to classify real from generated samples
        real_loss = adversarial_loss(discriminator(real_imgs), valid)
        fake_loss = adversarial_loss(discriminator(gen_imgs.detach()), fake)
        d_loss = (real_loss + fake_loss) / 2

        d_loss.backward()
        optimizer_D.step()
        
        if i % 100==0:
            print(
                "[Epoch %d/%d] [Batch %d/%d] [D loss: %f] [G loss: %f]"
                % (epoch, opt.n_epochs, i, len(dataloader), d_loss.item(), g_loss.item())
            )

        batches_done = epoch * len(dataloader) + i
        # if  gen_imgs.shape[1]==3:
        #      gen_imgs =  gen_imgs[:, [2, 1, 0]]
        if batches_done % opt.sample_interval == 0:
            save_image(gen_imgs.data[20:45], "image_anime/%06d.png" % (batches_done//100), nrow=5, normalize=True)#这里查看的中间25张图像，源码看的是前25张图像

永不磨灭的fw

关注

0
点赞
踩
8

收藏

觉得还不错? 一键收藏
4
评论
GAN（Generative Adversarial Network）的复现

链接：https://pan.baidu.com/s/1rHwdw22bl-l5ID3OJMsldw。使用了一个新的数据集anime，数据集的链接在下方，数据集一共有5w多张动漫头像，都是。首先为数据集建造dataset类，代码如下。训练的代码如下，详细修改见代码注释。，在一个新的数据集完成了复现。
复制链接

扫一扫