Stavros Theocharis

I think @NickNtamp can share some thoughts on that since he is going to check a bit Mlflow

Not exactly it was chosen to be used as it is. But i believe that the proposed way is better.

Do you believe that there is a time threshold (e.g. not take more than 20 seconds)? - Depends on when this pipeline runs. If it is in almost real-time, I...

I would keep this as an issue in the backlog, in order to further investigate it and implement an enhancement in the future

Have a quick analysis before the implementation.

Yes, it should be performed at the unprocessed dataset for sure. The only thing here is "where to do it" and then how to save them into the database. More...

 For now, let’s use it for processed training data. For inferences let’s keep it for a new feature.

For the name let's use the "descriptive statistics" where we use integrity, etc.: