A preliminary platform for up to 1 million reinforcement learning agents - View it on GitHub