search

NicolasHug / Surprise

A Python scikit for building and analyzing recommender systems
http://surpriselib.com
BSD 3-Clause "New" or "Revised" License
6.42k stars 1.02k forks source link

How does Surprise compare to LightFM? #481

Open hichemfantar opened 5 months ago

hichemfantar commented 5 months ago

A nice breakdown of the pros and cons of each library would be great.

NicolasHug commented 5 months ago

Take this with a grain of salt but the chat-GPT answer is surprisingly accurate https://chatgpt.com/share/d58097c0-b087-4650-8fdc-640ca64476a0 (I didn't fact-check everything!!!!)

hichemfantar commented 5 months ago

@NicolasHug I believe surprise only handles explicit rating while lightFM handles both implicit and explicit, is that correct?

NicolasHug commented 5 months ago

yup

hichemfantar commented 5 months ago

yup

Are there any plans to support implicit rating in the future?

NicolasHug commented 5 months ago

nope

hichemfantar commented 5 months ago

Is it because of a limitation with the current implementation, or do you lack the time to implement it or do you simply not want to include it with Surprise? edit: i just read https://github.com/NicolasHug/Surprise?tab=readme-ov-file#development-status but still asking to see if the circumstances have changed.

NicolasHug commented 5 months ago

circumstances haven't changed

hichemfantar commented 5 months ago

Thanks for the info and your hard work !