Efficient Likelihood Inference for Stellar Ages
A Python package for Bayesian inference of stellar cluster parameters from Gaia photometry.
ELISA provides two inference engines:
- MCMC isochrone fitting — infers cluster age, metallicity, distance modulus, and extinction by fitting PARSEC isochrones with
emcee. - Binary-star SBI — estimates primary mass and mass ratio for individual stars using Simulation-Based Inference (SNPE neural posterior estimation).
git clone https://github.com/stevenalfonso/elisa.git
cd elisa
pip install -e .For SBI support (requires PyTorch):
pip install torch sbi isochronesFull usage guides and API reference are at astroelisa.readthedocs.io:
Bug reports and feature requests are welcome via GitHub Issues.
To contribute code, fork the repository and open a pull request against main. Please:
- Include tests for new functionality (
pytest) - Follow the existing code style (
black,ruff) - Describe what the PR changes and why
- PARSEC isochrones: Bressan et al. (2012)
- emcee: Foreman-Mackey et al. (2013)
- Likelihood formulation: von Hippel et al. (2006)
MIT License — see LICENSE for details.