search

yngvem / group-lasso

Group Lasso implementation following the scikit-learn API
MIT License
105 stars 32 forks source link

automatic Ridge regression #37

Open MPenaR opened 1 year ago

MPenaR commented 1 year ago

Hi, I've been testing your library, and using GroupLasso setting both l1_reg=0 and group_reg=0 I get a solution which I think is not the minimizer of L(beta,X,y) but the minimization of that times a high Ridge regression penalization.

Is this the expected behaviour?

I not interesting only in adding Lasso regularization but not including any kind of Ridge regression.

Thank you very much for your repo