Repository files navigation

PyMC-BART extendsPyMC probabilistic programming framework to be able to define and solve models including a BART random variable. PyMC-BART also includes a few helpers function to aid with the interpretation of those models and perform variable selection.

• Installation
• Usage
• Contributions
• Code of Conduct
• Citation
• License
• Donations
• Sponsors

PyMC-BART is available on Conda-Forge. If you magange your Python dependencies and environments with Conda, this is your best option. You may also perfer to install this way if you want an easy-to-use, isolated setup in a seperate environment. This helps avoid interfering with other projects or system-wide Python installations. To set up a suitable Conda environment, run:

conda create --name=pymc-bart --channel=conda-forge pymc-bartconda activate pymc-bart

Alternatively, you can use pip installation. This installation is generally perfered by users who use pip, Python's package installer. This is the best choice for users who are not using Conda or for those who want to install PyMC-BART into a virtual environment managed by venv or virtualenv. In this case, run:

pip install pymc-bart

In case you want to upgrade to the bleeding edge version of the package you can install from GitHub:

pip install git+https://github.com/pymc-devs/pymc-bart.git

Get started by using PyMC-BART to set up a BART model:

importpymcaspmimportpymc_bartaspmbX,y= ...# Your data replaces "..."withpm.Model()asmodel:bart=pmb.BART('bart',X,y)    ...idata=pm.sample()

PyMC-BART is a community project and welcomes contributions.Additional information can be found in theContributing Readme

PyMC-BART wishes to maintain a positive community. Additional detailscan be found in theCode of Conduct

If you use PyMC-BART and want to cite it please use

Here is the citation in BibTeX format

@misc{quiroga2023bayesian,title={Bayesian additive regression trees for probabilistic programming},author={Quiroga, Miriana and Garay, Pablo G and Alonso, Juan M. and Loyola, Juan Martin and Martin, Osvaldo A},year={2023},doi={10.48550/ARXIV.2206.03619},archivePrefix={arXiv},primaryClass={stat.CO}}

Apache License, Version 2.0

PyMC-BART , as other pymc-devs projects, is a non-profit project under the NumFOCUS umbrella. If you want to support PyMC-BART financially, you can donatehere.