Require documents for trainer/miner

[seefun]

I am a researcher on multi-modal area for 10 years and interest in any-to-any task. And I am also a contributor to the most popular repository in the field of computer vision timm
I really want to participate in this research in deep, but I can't get good documentation from this repo. How to join this subnet, task training and evaluation instructions and train/val data example , how to obtain training data, how to submit model, baseline training code examples, etc.