A simple stochastic OpenAI environment for training RL agents - View it on GitHub