RPLIB

This repo contains the RPLib data, package, and web server.

Browse data (recommended)

Visit the RPLIB browser at https://birg.dev/rplib/.

View RPLIB Card (recommended)

You may view individual cards using the RPLIB browser or for a more customizable experience, use our sample Jupyter notebook that has been tested with Google's Colab: Click to launch on Colab.

Contribute a dataset (recommended)

We encourage the community to extend this repository with additional data. The recommended procedure for doing so begins with filling out this form: Click to launch add dataset form.

Generate artificial data (recommended)

We have also developed a sample notebook for those researchers interested in generating their own artificial data in a compatable format for RPLIB. Click to launch on Colab.

Installing pyrplib package

Prerequisites

The graphviz headers must be installed. To do this on Debian/Ubuntu:

apt-get install -y libgraphviz-dev

Recommended package installation

pip install pyrplib

Documentation

For documentation on pyrplib: https://birg.dev/pyrplib.

Post package installation: Gurobi License

The webserver does not require Gurobi to launch, but to use the full functionality of RPLIB, the Gurobi optimizer must be installed with a valid license. Gurobi provides free academic licenses and more information on obtaining and installing your license can be found here: https://support.gurobi.com/hc/en-us/articles/360040541251.

Verify installation

import pyrplib

Development notes

Launch a version of RPLIB browser locally

Production

cd RPLIB
USER=$(id -u) docker-compose up -d --build production

Visit http://localhost:7001/rplib to test your site.

Staging

cd RPLIB
git checkout staging
USER=$(id -u) docker-compose up -d --build staging

Visit http://localhost:7002/rplib_staging to test your site.

Development environments

cd RPLIB
git checkout <branch>
USER=$(id -u) docker-compose up --build dev

Visit http://localhost:7003/rplib_dev to test your site.

Authors

Paul Anderson, Ph.D.
Brandon Tat
Charlie Ward
Department of Computer Science and Software Engineering
California Polytechnic State University

Kathryn Pedings-Behling
Amy Langville, Ph.D.
Department of Mathematics
College of Charleston

Acknowledgements

We would like to thank the entire IGARDS team for their invaluable insight and encouragement.