End-to-end optimal quadcopter control through Reinforcement Learning - View it on GitHub