Reinforcement learning DQN /Double DQN/Multistep agent code of Pac-man environment.

Hi. I'd like to solve pacman environment using DQN /Double DQN/Multistep agent. There are many layout but smallGrid is our goal. All the environment codes are ready, you only need to add some function of [login to view URL] code (a few code of step function) with pytorch. I have a test code so you can test with your code. If you are interested in this project, please message me for the details :) (Btw, I've posted this project before but a freelancer ran away so I'm posting this project again.) Thank you.