I am trying to train a simple Reinforcement Learning(RL) agent to avoid obstacles. The training environment gives me access to an array with of n labels(8) of a labeled top-