Metadata-Version: 2.4
Name: bunobee
Version: 0.0.4
Summary: The time-series engine for forecasting with JAX and NumPyro
Author-email: Edwin Ng <edwinnglabs@gmail.com>
License-Expression: MIT
Project-URL: Homepage, https://github.com/edwinnglabs/bunobee
Project-URL: Repository, https://github.com/edwinnglabs/bunobee
Project-URL: Issues, https://github.com/edwinnglabs/bunobee/issues
Project-URL: Documentation, https://github.com/edwinnglabs/bunobee#readme
Keywords: time-series,forecasting,bayesian,jax,numpyro,kalman-filter
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Science/Research
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Scientific/Engineering
Requires-Python: >=3.11
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpyro>=0.20.0
Requires-Dist: jax<0.10.0,>=0.8.0
Requires-Dist: pandas>=2.0
Requires-Dist: arviz
Requires-Dist: xarray
Provides-Extra: test
Requires-Dist: pytest>=8; extra == "test"
Requires-Dist: black==26.1.0; extra == "test"
Provides-Extra: notebook
Requires-Dist: ipywidgets; extra == "notebook"
Dynamic: license-file

<div align="center">
  <img src="./bunobee.png" alt="Bunobee Logo" width="320" />
</div>

# Bunobee

Bunobee is an experimental Bayesian time-series library built on [JAX](https://github.com/google/jax) and [NumPyro](https://github.com/pyro-ppl/numpyro). It currently supports the following model families:

- **Damped Local Trend (DLT)** — an exponential smoothing model with a damped slope component
- **State Space Models (SSP)** — structured state space models with Kalman filtering and smoothing

**Fun fact:** "bunobee" is a word invented by my kids during their childhood to describe the action of the role they were playing.

