search

lacava / few

a feature engineering wrapper for sklearn
https://lacava.github.io/few
GNU General Public License v3.0
50 stars 22 forks source link

few.model() and few.print_model() #4

Closed TheodoreGalanos closed 7 years ago

TheodoreGalanos commented 7 years ago

Hello!

Thanks for sharing your work, this is really cool!

I was wondering if you could provide a bit of explanation as to the difference between these two outputs of the algorithm.

Also, is there any (outside) documentation on all this?

Thanks in advance!

Kind regards, Theodore.

lacava commented 7 years ago

thanks for your interest! currently, the difference is that print_model() will return a model complete with coefficients generated from the paired ML, e.g. -2.5 sin(x_1x2) + -4.56 log(x_1)

whereas model() will just return a list of engineered features. however, these should probably be merged in the future or clarified - maybe renaming model() to features() or just storing a list of features as strings at the end of fit().

lacava commented 7 years ago

regarding the documentation, there should be some very soon!

TheodoreGalanos commented 7 years ago

@lacava Thanks a lot for the clarification, it makes sense.

I'll keep testing and come back with hopefully useful comments, if any.

Kind regards, Theodore.

lacava commented 7 years ago

a7dfb92 adds some outside documentation.