chkothe
commented
4 years ago Hi Izabel, I think this was a copy-paste error in the example.
On Wed, Jan 8, 2020 at 7:55 AM Izabel Cvetkovic notifications@github.com wrote:
I am bit confused with ml_trainridge.m documentation.
According to the explanation: % Ridge regression [1] is a generalization of basic multivariate regression which in addition % includes a regularization parameter. This should be a linear regression algorithm. When I use it, automatically chosen evaluation metric is mse, which also supports this.
But, this example from the same file
% Examples: % as before, but using a non-default regularization parameter (here: 0.1) % model = ml_trainlogreg(trials,targets,0.1)
seems like it is a logistic regression classification algorithm with ridge regularization. So, is it a classifier or regression?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/sccn/BCILAB/issues/65?email_source=notifications&email_token=ABISL2A3LVYF64KJW76CAPTQ4XZPBA5CNFSM4KEKPYRKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IEZY6AA, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABISL2A2GZTWFMOTISOXMUTQ4XZPBANCNFSM4KEKPYRA .
Izabel94
I am bit confused with ml_trainridge.m documentation.
According to the explanation: % Ridge regression [1] is a generalization of basic multivariate regression which in addition % includes a regularization parameter. This should be a linear regression algorithm. When I use it, automatically chosen evaluation metric is mse, which also supports this.
But, this example from the same file
% Examples: % as before, but using a non-default regularization parameter (here: 0.1) % model = ml_trainlogreg(trials,targets,0.1)
seems like it is a logistic regression classification algorithm with ridge regularization. So, is it a classifier or regression?