What to do if the dataset I want to read has more than three parameters

[GustaMatos0]

Hey! I've recently started working on research pertaining to finding new uses for matrix factorization techniques. The biggest challenge is that it has proven quite hard not to overflow while testing with different datasets.
Tried using surprise but whenever I try to pass anything bigger than 3 collumns on the Reader, I get an error (ValueError: line_format parameter is incorrect.)

If there's any way to bypasss that restriction, or any resource you could point me to, I'd be delighted!
Thanks!

[NicolasHug]

Hi,

The recommendation algorithms available in surprise only care about user_id, item_id, rating. This is why you can only pass those 3 as columns of your dataset.

If what you're trying to do is to feed the algorithm some additional data sources (e.g. content-based information), then surprise probably isn't the right tool for you. HTH.