search

Ritam-Guha / Py_FS

A Python Package for Feature Selection
MIT License
58 stars 38 forks source link

Apply Py_FS for regression problems #31

Open hanamthang opened 3 years ago

hanamthang commented 3 years ago

Hi there,

Many thanks to your great work :-) I am wondering about the potential use of the Py_FS for any regression problems?

We work in the main theme of ecology remote sensing and most of the contents are for regression problems. So, it would be great if we have a chance to use your codes with our data.

Thang

Ritam-Guha commented 3 years ago

Hi Thang, Thanks for reaching out.

Currently, Py_FS does not allow any default regression function because that's how our objective function is defined. But, applying it to regression problems won't be difficult. You need to pass your regression objective function as a parameter while calling any algorithm from our framework. And when it asks for classification weightage, you need to provide the regression weightage. If you want, I can help you design the obj_function.

We are also coming up with a lot of updates to make it user-friendly.

Ritam

hanamthang commented 3 years ago

Hi Ritam-Guha,

It would be great if you can spend your time on designing the obj_function for regression problems. We are happy to use the Py_FS with our dataset for the feature selection. In addition, would you consider to add the Harris Hawk Optimization (HHO) into the Py_FS in the near future?

Many thanks, Thang

Ritam-Guha commented 2 years ago

Hey Thang, Sorry for the delayed response. I was quite busy over the past few months and could not work on this package. But, we are planning to add HHO to Py_FS.