Lessons Learned Reproducing a Deep Reinforcement Learning Paper http://amid.fish/reproducing-deep-rl