NicolasHug / Surprise

A Python scikit for building and analyzing recommender systems
http://surpriselib.com
BSD 3-Clause "New" or "Revised" License
6.28k stars 1k forks source link

No timestamp data in trainset #469

Closed tseesurenb closed 1 year ago

tseesurenb commented 1 year ago

Hi all,

I recently discovered that the trainset class design has omitted the "timestamp" feature, which I believe is a significant oversight. This limitation restricts the utilization of this excellent framework for those interested in implementing temporal effects. Do you have any suggestions on how to redesign the Training class to include the timestamp?

Best regards, Tse

Description

Steps/Code to Reproduce

Expected Results

Actual Results

Versions

NicolasHug commented 1 year ago

Try https://github.com/NicolasHug/Surprise/issues/52#issuecomment-308640663

I added the timestamps originally thinking I'd implement predictors that use them, but ended up not doing so. So it's more of an unsupported feature at this point (and a wonky design but :shrug: ).