Bayesian MAP / Variance Estimation

bayes_mapvar is a Python package that provides tools for Maximum A Posterior (MAP) estimation and posterior variance estimation. TensorFlow Probability and SciPy are leveraged to allow fast and efficient estimation for Bayesian models without using resampling methods..

Installation

1. pip install bayes_mapvar

or

1. Clone the repository.
2. Navigate to the project root directory.
3. Create the virtual environment with:

> . bin/setup_venv.sh

Documentation

The documentation for the latest release is at

https://cdlindsey.github.io/bayes_mapvar/mapvar.html

Usage and Examples

We predict oscar winners in this google collab notebook.

Testing

To run unit tests, run this from the project root directory:

> . bin/run_unit_tests.sh