Collection of PyTorch Lightning implementations of Generative Adversarial Network varieties presented in research papers.

$ pip install -r requirements.txt

The minimum code for training GAN is as follows:

from pytorch_lightning.trainer import Trainer
from models import GAN


model = GAN()
trainer = Trainer()
trainer.fit(model)

or you can run the following command:

$ python models/gan.py --gpus=2

• BEGAN: Boundary equilibrium generative adversarial networks (Berthelot et al.)
• DCGAN: Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks (Radford et al.)
• GAN: Generative Adversarial Networks (Goodfellow et al.)
• LSGAN: Least squares generative adversarial networks (Mao et al.)
• WGAN: Wasserstein GAN (Arjovsky et al.)
• WGAN-GP: Improved Training of Wasserstein GANs (Gulrajani et al.)

This repository is highly inspired by PyTorch-GAN repository.

• Goodfellow, Ian, et al. "Generative adversarial nets." Advances in neural information processing systems. 2014.
• Radford, Alec, Luke Metz, and Soumith Chintala. "Unsupervised representation learning with deep convolutional generative adversarial networks." arXiv preprint arXiv:1511.06434 (2015).
• Berthelot, David, Thomas Schumm, and Luke Metz. "Began: Boundary equilibrium generative adversarial networks." arXiv preprint arXiv:1703.10717 (2017).
• Mao, Xudong, et al. "Least squares generative adversarial networks." Proceedings of the IEEE international conference on computer vision. 2017.
• Arjovsky, Martin, Soumith Chintala, and Léon Bottou. "Wasserstein generative adversarial networks." Proceedings of the 34th International Conference on Machine Learning-Volume 70. 2017.
• Gulrajani, Ishaan, et al. "Improved training of wasserstein gans." Advances in neural information processing systems. 2017.