A Hearthstone environment for reinforcement learning agents, subclassing OpenAI/Gym. Supports human input, and includes a default agent. - View it on GitHub