View Code? Open in Web Editor
NEW
Code accompanying the paper "TiZero: Mastering Multi-Agent Football with Curriculum Learning and Self-Play" (AAMAS 2023) 足球游戏智能体
Home Page: https://tizero.readthedocs.io/
License: Apache License 2.0
Python 99.58%
Makefile 0.36%
Shell 0.06%
tizero's People
Contributors
Watchers
tizero's Issues
[new feature] tizero eval support output detail information: such as pass ball, shot, etc
some thing is wrong python 3.8, need to test with python 3.8
[new feature] tizero eval support swtich left right agents
support offline rl from replay files
I didn't see the section on how to train the model.
[new feature] support convert dump to expert data