| ## Instructions for Contributing to TSlib | |
| Sincerely thanks to all the researchers who want to use or contribute to TSlib. | |
| Since our team may not have enough time to fix all the bugs and catch up with the latest model, your contribution is essential to this project. | |
| ### (1) Fix Bug | |
| You can directly propose a pull request and add detailed descriptions to the comment, such as [this pull request](https://github.com/thuml/Time-Series-Library/pull/498). | |
| ### (2) Add a new time series model | |
| Thanks to creative researchers, extensive great TS models are presented, which advance this community significantly. If you want to add your model to TSlib, here are some instructions: | |
| - Propose an issue to describe your model and give a link to your paper and official code. We will discuss whether your model is suitable for this library, such as [this issue](https://github.com/thuml/Time-Series-Library/issues/346). | |
| - Propose a pull request in a similar style as TSlib, which means adding an additional file to ./models and providing corresponding scripts for reproduction, such as [this pull request](https://github.com/thuml/Time-Series-Library/pull/446). | |
| Note: Given that there are a lot of TS models that have been proposed, we may not have enough time to judge which model can be a remarkable supplement to the current library. Thus, we decide ONLY to add the officially published paper to our library. Peer review can be a reliable criterion. | |
| Thanks again for your valuable contributions. | |