Gym for multi-agent reinforcement learning - View it on GitHub