search

mgerstgrasser / super

suPER is a collaborative multi-agent RL algorithm
MIT License
9 stars 1 forks source link

show the result #1

Open czf456789 opened 4 weeks ago

czf456789 commented 4 weeks ago

Hello! It's a great project! Recently I was trying to run it, and now I have a problem, how should I apply the trained model to this battle environment and show it, that is, render it

czf456789 commented 4 weeks ago

@mgerstgrasser

mgerstgrasser commented 3 weeks ago

Hi @czf456789 and thank you for your interest!

Do you mean you want to show or save a video recording of the trained agents playing in the environment? That's not something we implement on our side, but RLLib has some examples of how to do that which you could adapt: https://github.com/ray-project/ray/blob/master/rllib/examples/envs/env_rendering_and_recording.py