uber/orbit
Uber's Bayesian time-series forecasting library with probabilistic programming support.
Not currently ranked — collecting fresh signals.
star history
Orbit is a Python package for Bayesian forecasting built on object-oriented design. It provides probabilistic models for time-series forecasting including Damped Trend, Exponential Smoothing, and Logistic Regression variants. The library supports Bayesian inference via Pyro and PyStan backends, enabling uncertainty quantification in forecasts.
Frequently asked
- What is uber/orbit?
- Uber's Bayesian time-series forecasting library with probabilistic programming support.
- Is orbit open source?
- Yes — uber/orbit is an open-source project tracked on heatdrop.
- What language is orbit written in?
- uber/orbit is primarily written in Python.
- How popular is orbit?
- uber/orbit has 2.1k stars on GitHub.
- Where can I find orbit?
- uber/orbit is on GitHub at https://github.com/uber/orbit.