hello lovely people.
i’m trying to feed a dense matrix into collab because my data is actually dense. i can’t see how to do it.
CollabDataBunch.from_df
expects user/product/rating rows obviously.
is there something incredibly clever in the datablock api which lets me do this? or anywhere else?
i could flatten it out but it’s just over 4 million ratings so i’d rather not if i can avoid it. this seems like something i should be able to do.