README.md

TF-GANs

This repository supports following GAN implementations

• DCGAN: Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• WGAN: Wasserstein GAN
• WGAN-GP: Improved Training of Wasserstein GANs
• LSGAN: Least Squares Generative Adversarial Networks
• SNGAN: Spectral Normalization for Generative Adversarial Networks

Dependencies

• python libraries

$ pip install requirements.txt

• others
  • ffmpeg to generate animation

Examples

MNIST

$ cd tf_gans
$ python run_mnist.py

Visualization

This code automatically generates results directory and it contains a log for each experiment and includes TensorBoard log. So only you have to do is just call the results directory like following.

$ cd /path/to/results
$ tensorboard --logdir=results