Meta-Inverse Reinforcement Learning with Probabilistic Context Variables - View it on GitHub