【PyTorch】生成对抗网络

生成对抗网络是什么

概念

Generative Adversarial Nets，简称GAN
GAN：生成对抗网络 —— 一种可以生成特定分布数据的模型
《Generative Adversarial Nets》 Ian J Goodfellow-2014

GAN网络结构

Recent Progress on Generative Adversarial Networks (GANs): A Survey

How Generative Adversarial Networks and Their Variants Work: An Overview

Generative Adversarial Networks_ A Survey and Taxonomy

GAN的训练

训练目的

1. 对于D：对真样本输出高概率
2. 对于G：输出使D会给出高概率的数据

GAN 的训练和监督学习训练模式的差异

在监督学习的训练模式中，训练数经过模型得到输出值，然后使用损失函数计算输出值与标签之间的差异，根据差异值进行反向传播，更新模型的参数，如下图所示。

在 GAN 的训练模式中，Generator 接收随机数得到输出值，目标是让输出值的分布与训练数据的分布接近，但是这里不是使用人为定义的损失函数来计算输出值与训练数据分布之间的差异，而是使用 Discriminator 来计算这个差异。需要注意的是这个差异不是单个数字上的差异，而是分布上的差异。如下图所示。

具体训练过程

step1：训练D
输入：真实数据加G生成的假数据
输出：二分类概率

step2：训练G
输入：随机噪声z
输出：分类概率——D(G(z))

DCGAN

Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks

Discriminator：卷积结构的模型
Generator：卷积结构的模型

DCGAN 的定义如下：

from collections import OrderedDict
import torch
import torch.nn as nn


class Generator(nn.Module):
    def __init__(self, nz=100, ngf=128, nc=3):
        super(Generator, self).__init__()
        self.main = nn.Sequential(
            # input is Z, going into a convolution
            nn.ConvTranspose2d(nz, ngf * 8, 4, 1, 0, bias=False),
            nn.BatchNorm2d(ngf * 8),
            nn.ReLU(True),
            # state size. (ngf*8) x 4 x 4
            nn.ConvTranspose2d(ngf * 8, ngf * 4, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ngf * 4),
            nn.ReLU(True),
            # state size. (ngf*4) x 8 x 8
            nn.ConvTranspose2d(ngf * 4, ngf * 2, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ngf * 2),
            nn.ReLU(True),
            # state size. (ngf*2) x 16 x 16
            nn.ConvTranspose2d(ngf * 2, ngf, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ngf),
            nn.ReLU(True),
            # state size. (ngf) x 32 x 32
            nn.ConvTranspose2d(ngf, nc, 4, 2, 1, bias=False),
            nn.Tanh()
            # state size. (nc) x 64 x 64
        )

    def forward(self, input):
        return self.main(input)

    def initialize_weights(self, w_mean=0., w_std=0.02, b_mean=1, b_std=0.02):
        for m in self.modules():
            classname = m.__class__.__name__
            if classname.find('Conv') != -1:
                nn.init.normal_(m.weight.data, w_mean, w_std)
            elif classname.find('BatchNorm') != -1:
                nn.init.normal_(m.weight.data, b_mean, b_std)
                nn.init.constant_(m.bias.data, 0)


class Discriminator(nn.Module):
    def __init__(self, nc=3, ndf=128):
        super(Discriminator, self).__init__()
        self.main = nn.Sequential(
            # input is (nc) x 64 x 64
            nn.Conv2d(nc, ndf, 4, 2, 1, bias=False),
            nn.LeakyReLU(0.2, inplace=True),
            # state size. (ndf) x 32 x 32
            nn.Conv2d(ndf, ndf * 2, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ndf * 2),
            nn.LeakyReLU(0.2, inplace=True),
            # state size. (ndf*2) x 16 x 16
            nn.Conv2d(ndf * 2, ndf * 4, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ndf * 4),
            nn.LeakyReLU(0.2, inplace=True),
            # state size. (ndf*4) x 8 x 8
            nn.Conv2d(ndf * 4, ndf * 8, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ndf * 8),
            nn.LeakyReLU(0.2, inplace=True),
            # state size. (ndf*8) x 4 x 4
            nn.Conv2d(ndf * 8, 1, 4, 1, 0, bias=False),
            nn.Sigmoid()
        )

    def forward(self, input):
        return self.main(input)

    def initialize_weights(self, w_mean=0., w_std=0.02, b_mean=1, b_std=0.02):
        for m in self.modules():
            classname = m.__class__.__name__
            if classname.find('Conv') != -1:
                nn.init.normal_(m.weight.data, w_mean, w_std)
            elif classname.find('BatchNorm') != -1:
                nn.init.normal_(m.weight.data, b_mean, b_std)
                nn.init.constant_(m.bias.data, 0)