Publishing my first deep reinforcement learning model, lot to learn

initial commit