This repository contains the code of PolyGym, as presented in

Brauckmann, Alexander, Andrés Goens, and Jeronimo Castrillon.
"PolyGym: Polyhedral Optimizations as an Environment for Reinforcement Learning."
2021 30th International Conference on Parallel Architectures and Compilation Techniques (PACT). IEEE, 2021.

1. Build the Polyite dependency. For this, refer to the steps in https://github.com/PolyGym/polyite. Note that the version of Polly that shall be used is in https://github.com/PolyGym/polly.
2. Change several hard-coded paths in the polygym project to point to your build of Polyite.
3. Run a sample exploration with random agents using the run_random.sh script (Note that this also requires changing paths to Polyite). This script assumes to run on a AMD 3960X CPU with 24 cores.