Solving a maze with reinforcement learning - View it on GitHub