search

ibayer / fastFM

fastFM: A Library for Factorization Machines
http://ibayer.github.io/fastFM
Other
1.07k stars 206 forks source link

verbose #95

Open weichen1984 opened 7 years ago

weichen1984 commented 7 years ago

Am I missing something? I cannot seem to find where I can set a verbose flag? I would like to see after each iteration, how do the metrics improve, such as the output from the original implementation.

brawner commented 7 years ago

It does not appear that there is any logging capability currently. In libFM it prints out the training and testing regression accuracy or classification accuracy after each iteration. With scipy minimization you can specify after how many iterations details should be printed out. It's very helpful to me to understand how quickly things are converging.

It would be great to see some logging/output capability added. Thanks for this excellent tool though!

ibayer commented 7 years ago

The model parameter in fastFM can be inspected after each iteration without much overhead. This allows you to create your own logging output in python. This approach is much more flexible but requires some boiler plate code (see. http://ibayer.github.io/fastFM/guide.html#learning-curves).

brawner commented 7 years ago

Thanks for the tip! That will work just fine

On Tue, Jun 20, 2017 at 1:29 AM, ibayer notifications@github.com wrote:

The model parameter in fastFM can be inspected after each iteration without much overhead. This allows you to create your own logging output in python. This approach is much more flexible but requires some boiler plate code (see. http://ibayer.github.io/fastFM/guide.html#learning-curves).

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/ibayer/fastFM/issues/95#issuecomment-309648806, or mute the thread https://github.com/notifications/unsubscribe-auth/AA_x-h5hei9CS-xWhvKAiBAD3zSj5-aWks5sF1i_gaJpZM4NKj2h .