This repository contains code to train and use the PERCIVAL neural networks described in "Learning Bayes' theorem with a neural network for gravitational-wave inference" by A. J. K. Chua and M. Vallisneri (arXiv:1909.05966).
See also "ROMAN: Reduced-Order Modeling with Artificial Neurons" by A. J. K. Chua, C. R. Galley, and M. Vallisneri (Phys. Rev. Lett. 122, 211101 (2019) or arXiv:1811.05491).
The code is demonstrated in the notebooks/Percival.ipynb
Jupyter notebook, which can be opened in Google's Colaboratory by clicking on the badge above.
Your feedback, comments, and questions are welcome.