Hugging Face's logo

OpenRL
/
tizero

Reinforcement Learning
PyTorch
English
deep-reinforcement-learning
OpenRL
Google Research Football
Model card Files
xet
 Community
tizero
File size: 2,518 Bytes
111384b
787ea75
111384b
787ea75
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33b9e1c
111384b
787ea75
 
 
 
 
 
 
 
 
 
 
 
b6f77bf
 
 
 
8ade906
b6f77bf
 
 
 
 
 
 
 
 
 
 
787ea75
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33b9e1c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
---
language: en
license: apache-2.0
library_name: pytorch
tags:
- deep-reinforcement-learning
- reinforcement-learning
- OpenRL
- Google Research Football
benchmark_name: Google Research Football
task_name: Google Research Football 11vs11 full game
pipeline_tag: reinforcement-learning
model-index:
- name: tizero
  results:
  - task:
      type: reinforcement-learning
      name: reinforcement-learning
    dataset:
      name: Google Research Football 11vs11 full game
      type: Google Research Football 11vs11 full game
    metrics:
    - type: trueskill
      value: 45.2
      name: trueskill

---


### 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](https://arxiv.org/abs/2302.07515)]  [[videos](https://www.youtube.com/watch?v=U9REh0otmVU)]  [[code](https://github.com/OpenRL-Lab/TiZero)]


### Installation

- Follow the instructions in [gfootball](https://github.com/google-research/football#on-your-computer) to set up the environment.
- `pip install gfootball`
- `pip install tizero` (or clone [this repo](https://github.com/OpenRL-Lab/TiZero) and `pip install -e .`).
- test the installation by `python3 -m gfootball.play_game --action_set=full`.

### 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](http://jidiai.cn/daily_6484285/daily_6484285.dump). 
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](http://www.jidiai.cn/env_detail?envid=34).

You can submit this agent to JIDI directly.


### Cite

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

```
@article{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},
  journal={arXiv preprint arXiv:2302.07515},
  year={2023}
}
```