Glmnet in Matlab

Lasso and elastic-net regularized generalized linear models

This is a Matlab port for the extremely efficient procedures for fitting the entire lasso or elastic-net path for linear regression, logistic and multinomial regression, Poisson regression the Cox model.

 
  • Features including

    • high efficiency by using coordinate descent with warm starts and active set iterations;

    • methods for prediction, plotting and k-fold cross-validation;

    • extensive options such as sparse input-matrix formats and range constraints on coefficients.

Two recent additions are the multiresponse gaussian, and the grouped multinomial.

License: GPL-2

Published: 2014-04-28

Citation Info
Glmnet for Matlab (2013) Qian, J., Hastie, T., Friedman, J., Tibshirani, R. and Simon, N.
http://www.stanford.edu/~hastie/glmnet_matlab/

Glmnet in R is also provided for R users, and many bells and whistles in the package are illustrated in the Glmnet Vignette.