GANs in High Physics

Project describtion

Project was created during participation in the Krakow Applied Physicsand Computer Science Summer School ’20. The aim of the project was to test the latest GenerativeAdversarial Network (GAN) models for their application in simulating physical events;

Participants

• Michał Kacprzak;
• Paweł Kopciewicz as supervisor;

Technologies

• Python 3.7

Installation

$ git clone https://github.com/MichalKacprzak99/reconstruction_particle_mass_spectra
$ cd reconstruction_particle_mass_spectra
$ sudo pip3 install -r requirements.txt

Example how to use

1. Open files/main.py.
2. Create object of GAN.
3. Call class method "train".

from GAN.gan import GAN
if __name__ == '__main__':
    gan = GAN()
    gan.train(30000)

Implementations

BGAN

Implementation of Boundary-Seeking Generative Adversarial Networks.

Code

Paper: https://arxiv.org/abs/1702.08431

DualGAN

Implementation of DualGAN: Unsupervised Dual Learning for Image-to-Image Translation.

Code

Paper: https://arxiv.org/abs/1704.02510

GAN

Implementation of Generative Adversarial Network with a MLP generator and discriminator.

Code

Paper: https://arxiv.org/abs/1406.2661

WGAN

Implementation of Wasserstein GAN (with DCGAN generator and discriminator).

Code

Paper: https://arxiv.org/abs/1701.07875

License

MIT