Own reward function and environment, how to use model?

[onkarrai06]

Hi, I am a student trying to implement a project and want to use your models for my MARL project.

However, I am finding it incredibly confusing as to how I would use your model to train with my own reward function and my own environment. I am working on highway-env highway-v0 and I see that you have a .json file that is a config file with the details, but what if I want to use my own environment that has my own reward functions and not just changes like controlled_vehicles.

Another question is I have is that I saw upon running your code that it launches a simulated environment which shows how the ego-vehicles learn with time and I am not sure how I would do that with my own environment. I would appreciate if you could take some time to guide me on this.

Thank you.