loupiotes: a Bayesian starspot modelling tool

What is loupiotes ?

Using modern sampling method enabling GPU scaling, loupiotes is mainly dedicated to perform Bayesian starspot modelling. Building upon the powerful framework provided by the PyMC framework, it implements starspots model exploration through Maximum a-posteriori (MAP) analysis and Hamiltonian Monte-Carlo (HMC) sampling.

Getting started

Prerequisites

loupiotes is written in Python3. The following Python package are necessary to use it :

• pymc
• arviz
• numpy
• scipy
• matplotlib
• numba
• tqdm

Installation

loupiotes does not have a PyPI or conda-forge packaged version yet. You will have to clone the online repository and run at the root of the downloaded directory:

pip install .

Documentation

An online documentation with tutorials and API description is available.

Author

• Sylvain N. Breton - Maintainer - (INAF-OACT, Catania, Italy)

Acknowledgements

If you use loupiotes in your work, please provide a link to the GitLab repository.

References

The models implemented by loupiotes are described in the following publications:

• Lanza et al. (2007), Astronomy & Astrophysics, 464, 741L.
• Lanza (2016), Cartography of the Sun and the Stars, Lecture Notes in Physics, Volume 914.