Logistic Regression with independently weighted regularization terms

Asked Nov 15 '19 at 16:40

Active Nov 15 '19 at 16:59

Viewed 32 times

For a standard logistic regression problem I'm interested in examining the impact of individual weighting of the parameters and in general the impact of regularization for completely separable data.

Does anyone know of a package in python which allows individual regularization weights for each parameter?

Thanks

(logistic regression, not SSE).

edited Nov 15 '19 at 16:59

asked Nov 15 '19 at 16:40

FourierFlux

That is using linear regression. – FourierFlux Nov 15 '19 at 16:53
If you read the `glmnet` documentation, you'll find that you can specify common GLMs very easily, including logistic regression. – Sycorax Nov 15 '19 at 16:55
I see, is there a python or matlab version? I don't use R. – FourierFlux Nov 15 '19 at 17:00
Yes. This is the first result for a Google search of "glmnet python." https://web.stanford.edu/~hastie/glmnet_python/ The bottom of the page also has a link to a Matlab version. – Sycorax Nov 18 '19 at 13:45

Logistic Regression with independently weighted regularization terms

0 Answers0