Best way to get started with ml agents

I'm interested in using ml agents for a side project I have, but haven't ever used them.

I was really hoping for an online tutorial or step by step set-up guide, but it seems every is outdated.

Are there any up to date date tutorials for a complete beginner. I don't fancy learning an old version if possible.

Looking for getting to grips with reward based and imitation based learning.

 

https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Getting-Started.md

and

https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Learning-Environment-Create-New.md

are both up to date..

 

Perfect thanks

 

I just finished this https://www.immersivelimit.com/tutorials/reinforcement-learning-penguins-part-1-unity-ml-agents and found it very informative.

It is outdated, but easy to update and the update process helps you learn as well.

Make sure you download Branch 2 of the ML agents repo and use the latest ML Agents package in unity. Then you'll find a few commands in the tutorial scripts that need to be updated.

If you go through that, be sure to note the questions people post at the bottom of each page, and also use this migration resource:
https://github.com/Unity-Technologies/ml-agents/blob/master/docs/Migrating.md

 

Thanks, I ended up watching the penguin video (a bit older) but it got me the fundamentals which is what I was struggling with.