您现在的位置是：首页 >技术杂谈 >CGAN(条件GAN)网站首页技术杂谈

相比于GAN，CGAN给生成器和辨别器都添加了一个辅助信息，假设为y,y可以是标签类别或者其他模态的信息。
目标函数相比于GAN在输入端的x和z变为在y条件下生成的x和z。

模型框架可以表示为：

代码：

import argparse
import os
os.environ['CUDA_VISIBLE_DEVICES'] = '2'
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")
parser.add_argument("--n_classes", type=int, default=10, help="number of classes for dataset")
parser.add_argument("--img_size", type=int, default=32, help="size of each image dimension")
parser.add_argument("--channels", type=int, default=1, help="number of image channels")
parser.add_argument("--sample_interval", type=int, default=400, help="interval between image sampling")
opt = parser.parse_args()
print(opt)

img_shape = (opt.channels, opt.img_size, opt.img_size)#(1,32,32)

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


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

        self.label_emb = nn.Embedding(opt.n_classes, opt.n_classes)

        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 + opt.n_classes, 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, noise, labels):
        # Concatenate label embedding and image to produce input
        gen_input = torch.cat((self.label_emb(labels), noise), -1)#(64,110)
        img = self.model(gen_input)
        img = img.view(img.size(0), *img_shape)
        return img


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

        self.label_embedding = nn.Embedding(opt.n_classes, opt.n_classes)#10个查询向量，每个映射为10维

        self.model = nn.Sequential(
            nn.Linear(opt.n_classes + int(np.prod(img_shape)), 512),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Linear(512, 512),
            nn.Dropout(0.4),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Linear(512, 512),
            nn.Dropout(0.4),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Linear(512, 1),
        )

    def forward(self, img, labels):
        # Concatenate label embedding and image to produce input
        d_in = torch.cat((img.view(img.size(0), -1), self.label_embedding(labels)), -1)#(64,1034)
        validity = self.model(d_in)#(64,1)
        return validity


# Loss functions
adversarial_loss = torch.nn.MSELoss()

# 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(
    datasets.MNIST(
        "/home/Projects/ZQB/a/dataset",
        train=True,
        download=False,
        transform=transforms.Compose(
            [transforms.Resize(opt.img_size), transforms.ToTensor(), transforms.Normalize([0.5], [0.5])]
        ),#32×32
    ),
    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))

FloatTensor = torch.cuda.FloatTensor if cuda else torch.FloatTensor
LongTensor = torch.cuda.LongTensor if cuda else torch.LongTensor


def sample_image(n_row, epoch):
    """Saves a grid of generated digits ranging from 0 to n_classes"""
    # Sample noise
    z = Variable(FloatTensor(np.random.normal(0, 1, (n_row ** 2, opt.latent_dim))))#100,100
    # Get labels ranging from 0 to n_classes for n rows
    labels = np.array([num for _ in range(n_row) for num in range(n_row)])
    labels = Variable(LongTensor(labels))
    gen_imgs = generator(z, labels)
    save_image(gen_imgs.data, fp='/home/Projects/ZQB/a/PyTorch-GAN-master/implementations/cgan/result/result'+f"image_{epoch}.png", nrow=n_row, normalize=True)


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

for epoch in range(opt.n_epochs):
    for i, (imgs, labels) in enumerate(dataloader):#(64,1,32,32)/(64)

        batch_size = imgs.shape[0]#64

        # Adversarial ground truths
        valid = Variable(FloatTensor(batch_size, 1).fill_(1.0), requires_grad=False)#(64,1)
        fake = Variable(FloatTensor(batch_size, 1).fill_(0.0), requires_grad=False)#(64,1)

        # Configure input
        real_imgs = Variable(imgs.type(FloatTensor))
        labels = Variable(labels.type(LongTensor))

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

        optimizer_G.zero_grad()

        # Sample noise and labels as generator input
        z = Variable(FloatTensor(np.random.normal(0, 1, (batch_size, opt.latent_dim))))#(64,100)
        gen_labels = Variable(LongTensor(np.random.randint(0, opt.n_classes, batch_size)))#64

        # Generate a batch of images
        gen_imgs = generator(z, gen_labels)#(64,1,32,32)

        # Loss measures generator's ability to fool the discriminator
        validity = discriminator(gen_imgs, gen_labels)#(64,1)
        g_loss = adversarial_loss(validity, valid)#0.9160

        g_loss.backward()
        optimizer_G.step()

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

        optimizer_D.zero_grad()

        # Loss for real images
        validity_real = discriminator(real_imgs, labels)#(64,1)
        d_real_loss = adversarial_loss(validity_real, valid)#1.0134

        # Loss for fake images
        validity_fake = discriminator(gen_imgs.detach(), gen_labels)#(64,1)
        d_fake_loss = adversarial_loss(validity_fake, fake)

        # Total discriminator loss
        d_loss = (d_real_loss + d_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:
        sample_image(n_row=10, epoch=epoch)

1：首先从数据加载开始看：
指定数据集读取位置，如果没有提前下载，将download改为True。接着对数据集图片进行变换，首先resize到32×32大小，接着转换为tensor后归一化。最后通过Dataloader进行加载，batchsize为64.

2：定义两个优化器。两个配置一样。
3：用1填充validation，用0填充fake，判别器对于真实的image期望输出1，对于噪声生成的image期望输出0.
4：接着生成噪声和附加条件y=gen_labels。其中gen_labels为0到10的整数，batchsize为64，即生成64个。

5：将噪声和标签输入到生成器中，产生的图像再输入到判别器中。

查看生成器的构成：

将噪声和经过编码后的标签concat之后输入到生成器中，label=64.

第一个参数限制类别只能在0-9之间，第二个参数将每一个标签数字映射到10维空间，即经过编码之后标签的大小为（64,10）。噪声的大小为(64,100),则concat之后大小为（64,110）。最后经过5个线性层，最后一个输入为1024，输出为image_size的乘积。即将每一个batch的每一个像素输出一个值。最后经过resize到原图大小。


接着是生成器：

将生成的image和编码后的标签输入到模型中，输出一个概率值。
接着计算生成器损失：为了骗过辨别器，希望判别器对于噪声输出的图片判别为1。这里用的MSE，因此不需要对最后输出的概率值进行sigmoid。

接着训练判别器：对于真实的图片由其对应的label，我们希望输出为1，对于假的图片和随机生成的标签，我们希望输出为0，计算两个损失和。

与DCGAN相比，判别器的输入缺少了标签，也就是说生成器的输出是任意的，噪声只是拟合真实图片的分布，而不限制输出的数值等。

6：最后将每一个epoch的输出保存下来：
这里主要限制了label的数值，通过一个for循环，不再是随意的而是10行，每一行是0-9。

结果展示：