Recommendation

• Our GAN based work for facial attribute editing - AttGAN.

News

• 8 April 2019: We re-implement these GANs by Tensorflow 2! The old version is here: v1 or in the "v1" directory.
• PyTorch Version

Tensorflow 2 implementations of DCGAN, LSGAN, WGAN-GP and DRAGAN.

DCGAN	LSGAN	WGAN-GP	DRAGAN

DCGAN	LSGAN
WGAN-GP	DRAGAN

WGAN-GP	DRAGAN

• Environment

  • Python 3.6

  • TensorFlow 2.2, TensorFlow Addons 0.10.0

  • OpenCV, scikit-image, tqdm, oyaml

  • we recommend Anaconda or Miniconda, then you can create the TensorFlow 2.2 environment with commands below

    conda create -n tensorflow-2.2 python=3.6
    
    source activate tensorflow-2.2
    
    conda install scikit-image tqdm tensorflow-gpu=2.2
    
    conda install -c conda-forge oyaml
    
    pip install tensorflow-addons==0.10.0

  • NOTICE: if you create a new conda environment, remember to activate it before any other command

    source activate tensorflow-2.2

• Datasets

  • Fashion-MNIST will be automatically downloaded
  • CelebA should be prepared by yourself in ./data/img_align_celeba/*.jpg
    • dataset link (find "img_align_celeba.zip"):
      • Baidu Netdisk (password rp0s) or
      • Google Drive
  • the Anime dataset should be prepared by yourself in ./data/faces/*.jpg
    • dataset link: https://www.kaggle.com/splcher/animefacedataset

• Examples of training

  • Fashion-MNIST DCGAN

    CUDA_VISIBLE_DEVICES=0 python train.py --dataset=fashion_mnist --epoch=25 --adversarial_loss_mode=gan

  • CelebA DRAGAN

    CUDA_VISIBLE_DEVICES=0 python train.py --dataset=celeba --epoch=25 --adversarial_loss_mode=gan --gradient_penalty_mode=dragan

  • Anime WGAN-GP

    CUDA_VISIBLE_DEVICES=0 python train.py --dataset=anime --epoch=200 --adversarial_loss_mode=wgan --gradient_penalty_mode=wgan-gp --n_d=5

  • see more training exampls in commands.sh

  • tensorboard for loss visualization

    tensorboard --logdir ./output/fashion_mnist_gan/summaries --port 6006