PyMC Survival

Overview

PyMC Survival is a collection of Bayesian parametric survival models written in Python using the scikit-learn API. The library is based on PyMC.

Create Docker container

docker build --build-arg GIT_ACCESS_TOKEN=[GITHUB_TOKEN] --target pymc-survival-paper -t ipaoluccimda/pymc-survival:initial-paper .

Installation

PyMC Survival requires Python 3.8 or higher (lower versions might work but are not tested).

Installation via pip

pip install pmsurv

Installation from source

pip install https://github.com/ipa/pymc-survival.git

Dependencies

PyMC survival requires ArviZ, NumPy, pandas, PyMC, and scikit-learn. All dependencies are listed in requirements.txt and in pyproject.toml. They will be installed automatically.

Example

In the following two examples we assume the following basic setup

    # Work in progress

Documentation

An official documentation is work in progress. See example notebooks for reference.

Citation

If you use PyMC Survival please cite:

Paolucci, I., Lin, YM., Albuquerque Marques Silva, J. et al. Bayesian parametric models for survival prediction in medical applications. BMC Med Res Methodol 23, 250 (2023). https://doi.org/10.1186/s12874-023-02059-4

Contributions

PyMC Survival started out of a research project. Contributions are welcome.

License

MIT License