Generative Adversarial Networks

This repository contains implementation of various architectures of Generative Models.

Implemented architectures

• GANs
• Wasserstein GANs
• WGAN with gradient penalty
• StyleGAN

Prerequisites

• Pytorch
• Numpy

Usage of GPU is highly recommended.

Datasets used

• LSUN
• MNIST
• Celeb A

References

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks by Soumith et al.
• Generative Adversarial Networks by Ian Goodfellow et al.
• Wasserstein GAN by Soumith et al.
• Improved training of Wasserstein GANs by Arjovsky et al.
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks by Zhu et al.