wyfzidane / gan-jittor

Jittor-GAN

Code based on Pytorch-GAN

Our GAN model zoo supports 27 kinds of GAN. This table is the latest citations we found from Google Scholar. It can be seen that since GAN was proposed in 2014, a lot of excellent work based on GAN has appeared. These 27 GANs have a total of 60953 citations, with an average of 2176 citations per article.

We compared the performance of these GANs of Jittor and Pytorch. The PyTorch version code uses the commit a163b8 on August 24, 2019 of the master branch of github repository: https://github.com/eriklindernoren/PyTorch-GAN. The picture below is the speedup ratio of Jittor relative to Pytorch. It can be seen that the highest acceleration ratio of these GANs reaches 283%, and the average acceleration ratio is 185%.

In another form of presentation, assuming that Pytorch's training time is 100 hours, we calculated the time required for GAN training corresponding to Jittor. Of these GANs, our fastest accelerating GAN takes only 35 hours to run, with an average of 57 hours.

Table of Contents

• Installation
• models
  • Auxiliary Classifier GAN
  • Adversarial Autoencoder
  • BEGAN
  • BicycleGAN
  • Boundary-Seeking GAN
  • Cluster GAN
  • Conditional GAN
  • Context Encoder
  • Coupled GAN
  • CycleGAN
  • Deep Convolutional GAN
  • DRAGAN
  • Energy-Based GAN
  • Enhanced Super-Resolution GAN
  • GAN
  • InfoGAN
  • Least Squares GAN
  • Pix2Pix
  • PixelDA
  • Relativistic GAN
  • Semi-Supervised GAN
  • Softmax GAN
  • StarGAN
  • UNIT
  • Wasserstein GAN
  • Wasserstein GAN GP
  • Wasserstein GAN DIV

Installation

$ git clone https://github.com/Jittor/gan-jittor.git
$ cd gan-jittor/
$ sudo python3.7 -m pip install -r requirements.txt

models

Auxiliary Classifier GAN

Auxiliary Classifier Generative Adversarial Network

Authors

Augustus Odena, Christopher Olah, Jonathon Shlens

[Paper] [Code]

Run Example

$ cd models/acgan/
$ python3.7 acgan.py

Adversarial Autoencoder

Adversarial Autoencoder

Authors

Alireza Makhzani, Jonathon Shlens, Navdeep Jaitly, Ian Goodfellow, Brendan Frey

[Paper] [Code]

Run Example

$ cd models/aae/
$ python3.7 aae.py

BEGAN

BEGAN: Boundary Equilibrium Generative Adversarial Networks

Authors

David Berthelot, Thomas Schumm, Luke Metz

[Paper] [Code]

Run Example

$ cd models/began/
$ python3.7 began.py

BicycleGAN

Toward Multimodal Image-to-Image Translation

Authors

Jun-Yan Zhu, Richard Zhang, Deepak Pathak, Trevor Darrell, Alexei A. Efros, Oliver Wang, Eli Shechtman

[Paper] [Code]

Run Example

$ cd data/
$ bash download_pix2pix_dataset.sh edges2shoes
$ cd ../models/bicyclegan/
$ python3.7 bicyclegan.py

Various style translations by varying the latent code.

Boundary-Seeking GAN

Boundary-Seeking Generative Adversarial Networks

Authors

R Devon Hjelm, Athul Paul Jacob, Tong Che, Adam Trischler, Kyunghyun Cho, Yoshua Bengio

[Paper] [Code]

Run Example

$ cd models/bgan/
$ python3.7 bgan.py

Cluster GAN

ClusterGAN: Latent Space Clustering in Generative Adversarial Networks

Authors

Sudipto Mukherjee, Himanshu Asnani, Eugene Lin, Sreeram Kannan

[Paper] [Code]

Run Example

$ cd models/cluster_gan/
$ python3.7 clustergan.py

Conditional GAN

Conditional Generative Adversarial Nets

Authors

Mehdi Mirza, Simon Osindero

[Paper] [Code]

Run Example

$ cd models/cgan/
$ python3.7 cgan.py

Context Encoder

Context Encoders: Feature Learning by Inpainting

Authors

Deepak Pathak, Philipp Krahenbuhl, Jeff Donahue, Trevor Darrell, Alexei A. Efros

[Paper] [Code]

Run Example

$ cd models/context_encoder/
<follow steps at the top of context_encoder.py>
$ python3.7 context_encoder.py

Rows: Masked | Inpainted | Original | Masked | Inpainted | Original

Coupled GAN

Coupled Generative Adversarial Networks

Authors

Ming-Yu Liu, Oncel Tuzel

[Paper] [Code]

Run Example

$ download mnistm.pkl from https://cloud.tsinghua.edu.cn/f/d9a411da271745fcbe1f/?dl=1 and put it into data/mnistm/mnistm.pkl
$ cd models/cogan/
$ python3.7 cogan.py

Generated MNIST and MNIST-M images

CycleGAN

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

Authors

Jun-Yan Zhu, Taesung Park, Phillip Isola, Alexei A. Efros

[Paper] [Code]

Run Example

$ cd data/
$ bash download_cyclegan_dataset.sh monet2photo
$ cd ../models/cyclegan/
$ python3.7 cyclegan.py --dataset_name monet2photo

Monet to photo translations.

Deep Convolutional GAN

Deep Convolutional Generative Adversarial Network

Authors

Alec Radford, Luke Metz, Soumith Chintala

[Paper] [Code]

Run Example

$ cd models/dcgan/
$ python3.7 dcgan.py

DRAGAN

On Convergence and Stability of GANs

Authors

Naveen Kodali, Jacob Abernethy, James Hays, Zsolt Kira

[Paper] [Code]

Run Example

$ cd models/dragan/
$ python3.7 dragan.py

Energy-Based GAN

Energy-based Generative Adversarial Network

Authors

Junbo Zhao, Michael Mathieu, Yann LeCun

[Paper] [Code]

Run Example

$ cd models/ebgan/
$ python3.7 ebgan.py

Enhanced Super-Resolution GAN

ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks

Authors

Xintao Wang, Ke Yu, Shixiang Wu, Jinjin Gu, Yihao Liu, Chao Dong, Chen Change Loy, Yu Qiao, Xiaoou Tang

[Paper] [Code]

Run Example

$ cd models/esrgan/
<follow steps at the top of esrgan.py>
$ python3.7 esrgan.py

GAN

Generative Adversarial Network

Authors

Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, Yoshua Bengio

[Paper] [Code]

Run Example

$ cd models/gan/
$ python3.7 gan.py

InfoGAN

InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets

Authors

Xi Chen, Yan Duan, Rein Houthooft, John Schulman, Ilya Sutskever, Pieter Abbeel

[Paper] [Code]

Run Example

$ cd models/infogan/
$ python3.7 infogan.py

Result of varying continuous latent variable by row.

Least Squares GAN

Least Squares Generative Adversarial Networks

Authors

Xudong Mao, Qing Li, Haoran Xie, Raymond Y.K. Lau, Zhen Wang, Stephen Paul Smolley

[Paper] [Code]

Run Example

$ cd models/lsgan/
$ python3.7 lsgan.py

Pix2Pix

Unpaired Image-to-Image Translation with Conditional Adversarial Networks

Authors

Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, Alexei A. Efros

[Paper] [Code]

Run Example

$ cd data/
$ bash download_pix2pix_dataset.sh facades
$ cd ../models/pix2pix/
$ python3.7 pix2pix.py --dataset_name facades

Rows from top to bottom: (1) The condition for the generator (2) Generated image <br>
based of condition (3) The true corresponding image to the condition

PixelDA

Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks

Authors

Konstantinos Bousmalis, Nathan Silberman, David Dohan, Dumitru Erhan, Dilip Krishnan

[Paper] [Code]

MNIST to MNIST-M Classification

Trains a classifier on images that have been translated from the source domain (MNIST) to the target domain (MNIST-M) using the annotations of the source domain images. The classification network is trained jointly with the generator network to optimize the generator for both providing a proper domain translation and also for preserving the semantics of the source domain image. The classification network trained on translated images is compared to the naive solution of training a classifier on MNIST and evaluating it on MNIST-M. The naive model manages a 55% classification accuracy on MNIST-M while the one trained during domain adaptation achieves a 95% classification accuracy.

$ download mnistm.pkl from https://cloud.tsinghua.edu.cn/f/d9a411da271745fcbe1f/?dl=1 and put it into data/mnistm/mnistm.pkl
$ cd models/pixelda/
$ python3.7 pixelda.py

Rows from top to bottom: (1) Real images from MNIST (2) Translated images from
MNIST to MNIST-M (3) Examples of images from MNIST-M

Relativistic GAN

The relativistic discriminator: a key element missing from standard GAN

Authors

Alexia Jolicoeur-Martineau

[Paper] [Code]

Run Example

$ cd models/relativistic_gan/
$ python3.7 relativistic_gan.py                 # Relativistic Standard GAN
$ python3.7 relativistic_gan.py --rel_avg_gan   # Relativistic Average GAN

Semi-Supervised GAN

Semi-Supervised Generative Adversarial Network

Authors

Augustus Odena

[Paper] [Code]

Run Example

$ cd models/sgan/
$ python3.7 sgan.py

Softmax GAN

Softmax GAN

Authors

Min Lin

[Paper] [Code]

Run Example

$ cd models/softmax_gan/
$ python3.7 softmax_gan.py

StarGAN

StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation

Authors

Yunjey Choi, Minje Choi, Munyoung Kim, Jung-Woo Ha, Sunghun Kim, Jaegul Choo

[Paper] [Code]

Run Example

$ cd models/stargan/
<follow steps at the top of stargan.py>
$ python3.7 stargan.py

Original | Black Hair | Blonde Hair | Brown Hair | Gender Flip | Aged

UNIT

Unsupervised Image-to-Image Translation Networks

Authors

Ming-Yu Liu, Thomas Breuel, Jan Kautz

[Paper] [Code]

Run Example

$ cd data/
$ bash download_cyclegan_dataset.sh apple2orange
$ cd models/unit/
$ python3.7 unit.py --dataset_name apple2orange

Wasserstein GAN

Wasserstein GAN

Authors

Martin Arjovsky, Soumith Chintala, Léon Bottou

[Paper] [Code]

Run Example

$ cd models/wgan/
$ python3.7 wgan.py

Wasserstein GAN GP

Improved Training of Wasserstein GANs

Authors

Ishaan Gulrajani, Faruk Ahmed, Martin Arjovsky, Vincent Dumoulin, Aaron Courville

[Paper] [Code]

Run Example

$ cd models/wgan_gp/
$ python3.7 wgan_gp.py

Wasserstein GAN DIV

Wasserstein Divergence for GANs

Authors

Jiqing Wu, Zhiwu Huang, Janine Thoma, Dinesh Acharya, Luc Van Gool

[Paper] [Code]

Run Example

$ cd models/wgan_div/
$ python3.7 wgan_div.py