Implementation of some different variants of GANs by tensorflow, DCGAN, WGAN, WGAN-GP, LSGAN, SNGAN, RSGAN, RaSGAN, BEGAN
Switch branches/tags
Nothing to show
Clone or download
Latest commit adb7164 Dec 8, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
BEGAN Add files via upload Dec 8, 2018
Image Add files via upload Dec 8, 2018
TrainingSet Add files via upload Jul 3, 2018
para Add files via upload Jul 3, 2018
result Add files via upload Jul 3, 2018
GANs.py Update GANs.py Nov 21, 2018
LICENSE Create LICENSE Jul 19, 2018
README.md Update README.md Dec 8, 2018

README.md

DCGAN_LSGAN_WGAN_WGAN-GP_SNGAN_RSGAN_RaSGAN_BEGAN

Implementation of some different variants of GANs

Introduction


This code is mainly implement some basic GANs about 'DCGAN', 'WGAN', 'WGAN-GP', 'LSGAN', 'SNGAN', 'RSGAN'&'RaSGAN', 'BEGAN'.

More details of these GANs, please see follow papers:

  1. DCGAN: Unsupervised representation learning with deep convolutional generative adversarial networks

  2. WGAN: Wasserstein gan

  3. WGAN-GP: Improved training of wasserstein gans

  4. LSGAN: Least Squares Generative Adversarial Networks

  5. SNGAN: Spectral normalization for generative adversarial networks

  6. RSGAN&RaSGAN: The relativistic discriminator: a key element missing from standard GAN

  7. BEGAN:BEGAN: Boundary Equilibrium Generative Adversarial Networks

How to use


Firstly, you should download the data 'facedata.mat' from Baidu Drive or Google Drive, then put the file 'facedata.mat' into the folder 'TrainingSet'.

Requirements

  1. python3.5
  2. tensorflow1.4.0
  3. pillow
  4. scipy
  5. numpy

Results of this code

This result is using DCGAN trained about 8000 iterations.

Compare LSGAN, WGAN, WGAN-GP, SNGAN, RSGAN of different iteration

Convergence of BEGAN