search

opendilab / DI-engine

OpenDILab Decision AI Engine. The Most Comprehensive Reinforcement Learning Framework B.P.
https://di-engine-docs.readthedocs.io
Apache License 2.0
2.97k stars 364 forks source link

Call for contributors! #621

Open PaParaZz1 opened 1 year ago

PaParaZz1 commented 1 year ago

Introduction

Dear community,

:rocket: DI-engine aims to be a generalized decision intelligence engine, including various algorithm toolkits and application demonstrations. Due to the complexity of abstraction and maintainment, DI-engine has become a pretty big beast :dragon_face:. :star2: Therefore, We are calling for contributors now. For each applied task, we will arrange a internal support to help and review related work. We are looking forward to more support and inspiration from open-source community~

:checkered_flag: Here we list different types of tasks below:

Algorithm Section

No. Task Name Difficulty Assigned to Supporter Status Other Info
1 nstep/td-lambda support for SAC :star::star: @GoingMyWay @PaParaZz1 Apply
2 optimal baseline for MAPPO :star::star:
3 DRL algo: MPO :star::star::star:
4 multi-agent RL algo: QPLEX :star::star::star:
5 other RL algo: JSRL :star::star::star::star:

Env and Application Section

No. Task Name Difficulty Assigned to Supporter Status Other Info
1 Integration for metaworld :star::star:
2 Update for competitive-rl :star::star:
3 Integration for GNN + RL :star::star:
4 Integration for RL4LM :star::star::star:
5 Integration for BabyAI :star::star::star:

Doc and Tutorial Section

No. Task Name Difficulty Assigned to Supporter Status Other Info
1 English translation for MDQN doc :star: @TuTuHuss @ruoyuGao Finish https://github.com/opendilab/DI-engine-docs/pull/232
2 English translation for bitfilp doc :star: @deepNet-Chirag @zjowowen Finish https://github.com/opendilab/DI-engine-docs/pull/235
3 English translation for Acrobot doc :star: @deepNet-Chirag @puyuan1996 Apply
4 Chinese translation for PPG doc :star: @PommesPeter @zjowowen Finish https://github.com/opendilab/DI-engine-docs/pull/260
5 Chinese translation for IMPALA doc :star:
6 Chinese translation for SAC doc :star:
7 Best practice doc for multiple reward :star::star:
8 Chinese README :star::star:
9 RLHF overview doc :star::star::star:
10 Best practice doc for customizing policy :star::star::star:

Efficiency Section

No. Task Name Difficulty Assigned to Supporter Status Other Info
1 Integration for torch.compile :star::star:
2 Integration for torch.amp :star::star:
3 Integration for JAX env lib gymnax :star::star:

Code Quality Section

No. Task Name Difficulty Assigned to Supporter Status Other Info
1 Add more common typing lint type :star:
2 Add deprecated function decorator :star: @ooooo-create @PaParaZz1 Finish reference link
3 Simplify SAC related policies :star::star::star:

Test and Deploy Section

No. Task Name Difficulty Assigned to Supporter Status Other Info
1 Deployment of drone env docker :star:

:sparkling_heart: If you are interested, you can claim a task by replying to this issue in the following format:

Metric No. : <The Task No. you claim>
Status :  Apply | Submit | Finish
Links : The repo links you fork (Apply status) / The PR links you open (Submit status)

:sparkles: Thank you in advance to anyone who helps out! For some outstanding contributors, we will send them some souvenirs, like badge and T-shirt powered by OpenDILab.

Notes

  1. The documation of DI-engine can be found at: https://di-engine-docs.readthedocs.io/en/latest/index.html
  2. The contributing guides can be found at: https://github.com/opendilab/DI-engine/blob/main/CONTRIBUTING.md
  3. If you have more detailed problems, you can join our slack channel. Slack has an IRC bridge as well as a range of apps which makes idling and answering questions super easy.
GoingMyWay commented 1 year ago

Hi, I would like to implement Algorithm Section 1 and 4. Here is my profile: https://qiuwei888.github.io/

Metric No. : Algorithm Section 1 and 4 Status : Apply Links : TBD

TuTuHuss commented 1 year ago

Metric No. : Doc and Tutorial Section No.1 Status : Apply Links : https://github.com/opendilab/DI-engine-docs/pull/232

PaParaZz1 commented 1 year ago

Hi, I would like to implement Algorithm Section 1 and 4. Here is my profile: https://qiuwei888.github.io/

Metric No. : Algorithm Section 1 and 4 Status : Apply Links : TBD

Thanks for your application! Which task do you want to start first? Algorithm Section 1 needs some understanding about the data flow of DI-engine while Algorithm Section 3 requires more necessary knowledge about multi-agent RL. You can choose one first and open a draft PR, we will provide corresponding support.

GoingMyWay commented 1 year ago

Hi, I would like to implement Algorithm Section 1 and 4. Here is my profile: https://qiuwei888.github.io/ Metric No. : Algorithm Section 1 and 4 Status : Apply Links : TBD

Thanks for your application! Which task do you want to start first? Algorithm Section 1 needs some understanding about the data flow of DI-engine while Algorithm Section 3 requires more necessary knowledge about multi-agent RL. You can choose one first and open a draft PR, we will provide corresponding support.

@PaParaZz1 Thanks. I would like select Algorithm Section 1 first. I have over 5 years of RL but I am a beginner of DI-Engine . I will look into the data flow of DI-engine and open the PR.

deepNet-Chirag commented 1 year ago

Hello @PaParaZz1! I would like to work with DI-Engine.. As I am new to it, I would like to start with Doc and Tutorial Translation.

Metric No. : Doc and Tutorial Section No. 2 Status : Apply Links : https://github.com/opendilab/DI-engine-docs/pull/235

deepNet-Chirag commented 1 year ago

Hey @PaParaZz1! Here is the pull request thread link for Doc and Tutorial Section No. 2 : https://github.com/opendilab/DI-engine-docs/pull/235

Let me know if there is any issue regarding the same. If everything is good then I would like to apply for the section no. 3.

Metric No. : Doc and Tutorial Section No. 3 Status : Apply Links : TBD

RajeshDM commented 1 year ago

Hi @PaParaZz1, I am interested in working on the GNN + RL from the "Env + Application" Section. Could you please provide some details on what the contribution would entail

I am a Ph.D. student working with graph neural networks and classical AI planning (and have experience with RL).

Regards, Rajesh Mangannavar

PommesPeter commented 11 months ago

I would like to do some translation work first because I am also new to the DI-engine. Meanwhile, I also try to learn something about RL or MARL for my research direction. I previously made some contributions to Paddle in terms of operator implementation.

Metric No. : Doc and Tutorial Section No. 4 Status : Finish Links : https://github.com/opendilab/DI-engine-docs/pull/260/files

ooooo-create commented 6 months ago

Metric No. : Code Quality Section No. 2 Status : Submit Links : DI-engine/pull/778