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

Learning Curves #667

Closed Blo0dR0gue closed 1 year ago

Blo0dR0gue commented 1 year ago

Hi,

it would be nice if you could add a partial fit function to the MF-models to better calculate learning cuves. I think it should be possible to outsource the learning part of the fit function to achieve this. Until now, you would have to perform multiple fits for different numbers of iterations to get a corresponding result. If I am wrong and there is a way, please let me know.

Blo0dR0gue commented 1 year ago

Ah, perhaps through the callback of the fit function per iteration, in which I can then calculate the corresponding metrics? I'll have to try it out. Nevertheless, I think a partial_fit function would not be bad.

Blo0dR0gue commented 1 year ago

nvm. fit does not reset the model.