search

benfred / implicit

Fast Python Collaborative Filtering for Implicit Feedback Datasets
https://benfred.github.io/implicit/
MIT License
3.57k stars 612 forks source link

The latest version installed without save and load #559

Closed yexiao1107 closed 2 years ago

yexiao1107 commented 2 years ago

I use pip install implicit, but the latest version installed without save and load

benfred commented 2 years ago

The save/load functions will be in the next version of implicit - this functionality was added after the last release

yexiao1107 commented 2 years ago

Thanks reply, and I have another issues, when I use "partial_fit_users" , I find the user vector is very small after training, and the recommad result is same, I do not know how to solve this. Is there a length limit of the click items?

benfred commented 2 years ago

Thanks reply, and I have another issues, when I use "partial_fit_users" , I find the user vector is very small after training, and the recommad result is same, I do not know how to solve this. Is there a length limit of the click items?

Can you create a new issue for this - and maybe a sample script to replicate? I find that after partial_fit_users the recommendations are the same as if called during 'fit'

benfred commented 2 years ago

v0.6.0 should have the save/load functionality -