README

cgmodsel - a Python3 package to estimate conditional Gaussian (CG) distributions from data

Version 1.0, 5.3.2020

Content

This package contains solvers (mostly ADMM-based) for estimating various types of CG models (sparse pairwise, sparse + low-rank pairwise, mean parameters). Please see the Wiki for additional information.

Setup cgmodsel

For normal usage:

1. Clone this repository into a folder of your choice.
2. Open a terminal and navigate to the chosen directory using cd.
3. Install the cgmodsel package locally by typing pip install .

You can also use the code without doing the third step from inside the folder. For a quick start, execute the file quick_cgmod.py from the main folder.

Usage

If you use this software, please consider citing:

Nussbaum, F., & Giesen, J. (2020). Pairwise sparse+ low-rank models for variables of mixed type. Journal of Multivariate Analysis, 104601.

Contact

If you have questions, comments, bug reports, feature requests, etc. please contact me: frank.nussbaum@uni-jena.de or come and see me in my office #3335 EAP2 Jena, Germany.

Copyright

© 2020 Frank Nussbaum (frank.nussbaum@uni-jena.de). All Rights Reserved.