Closed loomlike closed 5 years ago
@miguelgfierro and I went back and forth on this - let's discuss tomorrow. Maybe it's better to do categories via something else (like content based vs collaborative), or drop categories entirely?
Related to #516
I like the Type column where you can enter multiple attributes describing the algorithm. We could also add details like news data for DKN.
Description
Current main README includes a algo table, but the categorization is a bit vague, e.g. xDeepFM and DKN are both deep learning recommenders too but categorized under MS's recommenders.
Expected behavior (i.e. solution)
Need discussion
Other Comments