The Bayesian Analysis Toolkit (BAT) is an MCMC sampling suite, and its current version BAT.jl is implemented in the Julia programming language. To bring this package, featuring many different samplers and integration algorithms, to the python user community, we started developing the batty package. It provides a python API and allows to use any python callable that evaluates the log-likelihood of your model for different parameter values. The project is currently in early development.

Further Information

Collaborators: Rasmus Orsoe (TUM), Dr. Oliver Schulz (MPP)

GitHub project page: https://github.com/bat/batty

PyPI project page: https://pypi.org/project/batty/

The project is currently funded by the PUNCH4NFDI consortium: https://www.punch4nfdi.de/