shekkizh/WassersteinGAN.tensorflow
Tensorflow implementation of Wasserstein GAN, a generative adversarial network variant using earth mover distance for stable training.
Not currently ranked — collecting fresh signals.
star history
This repository provides a Tensorflow implementation of Wasserstein GAN (WGAN), an alternative to standard GANs that replaces Jensen-Shannon divergence with Wasserstein (earth mover) distance for measuring distribution similarity. The approach addresses training instability issues in traditional GANs by providing a continuous, almost everywhere differentiable metric even when distributions have non-overlapping support. The critic network outputs scores rather than probabilities, enabling reliable gradient signals for the generator.
Frequently asked
- What is shekkizh/WassersteinGAN.tensorflow?
- Tensorflow implementation of Wasserstein GAN, a generative adversarial network variant using earth mover distance for stable training.
- Is WassersteinGAN.tensorflow open source?
- Yes — shekkizh/WassersteinGAN.tensorflow is open source, released under the MIT license.
- What language is WassersteinGAN.tensorflow written in?
- shekkizh/WassersteinGAN.tensorflow is primarily written in Python.
- How popular is WassersteinGAN.tensorflow?
- shekkizh/WassersteinGAN.tensorflow has 413 stars on GitHub.
- Where can I find WassersteinGAN.tensorflow?
- shekkizh/WassersteinGAN.tensorflow is on GitHub at https://github.com/shekkizh/WassersteinGAN.tensorflow.