Hi, does anyone have a best-practice code example on an ML pipeline from data ingestion from an external source to cleaning to feature generation to fitting and and subsequent use of the model using prefect? Wondering whether to split up these parts as tasks or individual flows and whether to store the interim results or not taking into account the possibility for (easier?) debugging in case they’re flows?

We wrote this post which may help you think through how to organize your ML stack. https://prefect-community.slack.com/archives/CL09KTZPX/p1660838471225999

Hi @Saad Zaheer ! Thanks for sharing that! I was wondering if you can share some high level code as well for a specific example to get a better feel.