SeaPearl (homonym of CPuRL, standing for constraint programming using reinforcement learning) is a flexible, easy-to-use, research-oriented, and open-source constraint programming solver able to natively use deep reinforcement learning algorithms for learning value-selection heuristics. The philosophy behind this solver is to ease and speed-up the development process to any researcher desiring to design learning-based approaches to improve constraint programming solvers.
The project is open-source and available on Github.
