A Gentle Introduction To Offline Reinforcement Learning