https://github.com/openrl-lab/tizero

Code accompanying the paper "TiZero: Mastering Multi-Agent Football with Curriculum Learning and Self-Play" (AAMAS 2023) 足球游戏智能体

https://github.com/openrl-lab/tizero

Science Score: 10.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
  • .zenodo.json file
  • DOI references
  • Academic publication links
    Links to: arxiv.org
  • Committers with academic emails
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (14.0%) to scientific vocabulary

Keywords

distributed-reinforcement-learning google-research-football multi-agent-reinforcement-learning pytorch reinforcement-learning self-playing
Last synced: 9 months ago · JSON representation

Repository

Code accompanying the paper "TiZero: Mastering Multi-Agent Football with Curriculum Learning and Self-Play" (AAMAS 2023) 足球游戏智能体

Basic Info
Statistics
  • Stars: 61
  • Watchers: 1
  • Forks: 7
  • Open Issues: 5
  • Releases: 0
Topics
distributed-reinforcement-learning google-research-football multi-agent-reinforcement-learning pytorch reinforcement-learning self-playing
Created about 3 years ago · Last pushed almost 3 years ago
Metadata Files
Readme License

README.md

License PyPI PyPI - Python Version Documentation Status

Introduction

Reinforcement learning agent for Google Research Football.

Code accompanying the paper "TiZero: Mastering Multi-Agent Football with Curriculum Learning and Self-Play" (AAMAS 2023). [paper] [videos].

Installation

  • Follow the instructions in gfootball to set up the environment.
  • pip install gfootball openrl "openrl[selfplay]"
  • pip install tizero (or clone this repo and pip install -e .).
  • test the installation by python3 -m gfootball.play_game --action_set=full.

Evaluate JiDi submissions locally

You can evaluate your agent locally using tizero:

bash tizero eval --left_agent submission_dir1 --right_agent submission_dir2 --total_game 10

For example, you can evaluate tizero with random agent as below:

bash tizero eval --left_agent submission/tizero --right_agent submission/random_agent --total_game 10

For evaluations for JiDi submissions on other games, please refer to the Arena of OpenRL and this example for the snake game.

Show a saved dump file

  • show detailed infomation of a match via: tizero show dump_file
  • show keypoints of a mactch via: tizero keypoint dump_file

You can download an example dump file from here.

Then execute: tizero show daily_6484285.dump or tizero keypoint daily_6484285.dump. Then you will see a GUI as below:

Convert dump file to video

After the installation, you can use tizero to convert a dump file to a video file. The usage is tizero dump2video <dump_file> <output_dir> --episode_length <the length> --render_type <2d/3d>.

You can download an example dump file from here. And then execute tizero dump2video daily_6484285.dump ./ in your terminal. By default, the episode length is 3000 and the render type is 2d. Wait a minute, you will get a video file named daily_6484285.avi in your current directory.

Submit TiZero to JIDI(及第评测平台)

JIDI is a public evaluation platform for RL agents. You can submit your agent of GRF at: http://www.jidiai.cn/env_detail?envid=34.

We provide several agents under ./submission/ directory, which can be submitted to JIDI directly:

  • ./submission/tizero: the final model of TiZero for JIDI submission, which ranked 1st on October 28th, 2022.
  • ./submission/random_agent: the random agent for JIDI submission.

Cite

Please cite our paper if you use our codes or our weights in your own work:

@inproceedings{lin2023tizero, title={TiZero: Mastering Multi-Agent Football with Curriculum Learning and Self-Play}, author={Lin, Fanqi and Huang, Shiyu and Pearce, Tim and Chen, Wenze and Tu, Wei-Wei}, booktitle={Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems}, pages={67--76}, year={2023} }

Owner

  • Name: OpenRL
  • Login: OpenRL-Lab
  • Kind: organization
  • Location: China

Open-sourcing advanced technology and exploring the forefront of AI.

GitHub Events

Total
  • Issues event: 1
  • Watch event: 12
  • Fork event: 2
Last Year
  • Issues event: 1
  • Watch event: 12
  • Fork event: 2

Committers

Last synced: over 2 years ago

All Time
  • Total Commits: 26
  • Total Committers: 3
  • Avg Commits per committer: 8.667
  • Development Distribution Score (DDS): 0.231
Past Year
  • Commits: 26
  • Committers: 3
  • Avg Commits per committer: 8.667
  • Development Distribution Score (DDS): 0.231
Top Committers
Name Email Commits
huangshiyu h****4@1****m 20
huangshiyu h****u@4****m 5
Shiyu Huang h****3@g****m 1
Committer Domains (Top 20 + Academic)

Issues and Pull Requests

Last synced: 10 months ago

All Time
  • Total issues: 11
  • Total pull requests: 15
  • Average time to close issues: 21 days
  • Average time to close pull requests: less than a minute
  • Total issue authors: 3
  • Total pull request authors: 1
  • Average comments per issue: 0.64
  • Average comments per pull request: 0.0
  • Merged pull requests: 15
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 1
  • Pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Issue authors: 1
  • Pull request authors: 0
  • Average comments per issue: 0.0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
Top Authors
Issue Authors
  • huangshiyu13 (8)
  • ardian-selmonaj (1)
  • 945716994 (1)
Pull Request Authors
  • huangshiyu13 (15)
Top Labels
Issue Labels
Pull Request Labels

Packages

  • Total packages: 1
  • Total downloads:
    • pypi 23 last-month
  • Total dependent packages: 0
  • Total dependent repositories: 0
  • Total versions: 3
  • Total maintainers: 1
pypi.org: tizero

Toolkit and agents for Google Research Football

  • Versions: 3
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Downloads: 23 Last month
Rankings
Dependent packages count: 7.3%
Average: 27.8%
Forks count: 30.4%
Stargazers count: 32.3%
Dependent repos count: 41.3%
Maintainers (1)
Last synced: 10 months ago