Bayesian Time Series modeling in Python

The goal of this package is to be able to create Bayesian time series models at scale using Python and Jax. Eventually, this package should be usable for hierarchical regression modeling of time series with AR errors.

This is heavily based on the work in TimeSeers but with a jax + numpyro backend