ExoGAN Guide

The ExoGAN code is a Deep Convolutional Generative Adversarial Network (DCGAN) trained to perform atmospheric retrievals of transmission spectra of exoplanets. ExoGAN can easily be modified to train on a diverse range of problems.

These sets of pages will be populated with installation instructions and how-to tutorials. If you have a specific question, please email Tiziano Zingales (tiziano.zingales.15@ucl.ac.uk) or Ingo Waldmann (ingo@star.ucl.ac.uk).

Code repository

ExoGAN is available on GitHub:


the training data (10 million transmission spectra) can be found here:



Zingales T., Waldmann I.P., AJ accepted, arXiv:1806.02906

Active developers

  • Tiziano Zingales
  • Ingo Waldmann


This work is licensed under the Creative Commons Attribution 4.0 International License.

We would like to draw your attention to Section 3 of the license and to:

  • retain identification of the creators by including the above listed references in future work and publications.
  • indicate if You modified the Licensed Material and retain an indication of any previous modifications

To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.