sjtu-marl / malib

A parallel framework for population-based multi-agent reinforcement learning.
https://malib.io
MIT License
499 stars 60 forks source link

Mappo+gfootball #29

Closed ziyuwan closed 3 years ago

ziyuwan commented 3 years ago

Runnable version for MAPPO on Google Research Football Env. This version is a simple demo for train against _built_inbot on "5_vs_5" scenario.

To run it python examples/mappo_gfootball/run_mappo_gfootball.py --config examples/mappo_gfootball/mappo_5_vs_5.yaml

Performance checking is needed. A more general API/config setting may also be needed.

@KornbergFresnel should we merge it to "evaluation worker" first?