Positivity or box constraints on fitted coefficients?

pbreheny / biglasso

biglasso: Extending Lasso Model Fitting to Big Data in R

http://pbreheny.github.io/biglasso/

113 stars 29 forks source link

Positivity or box constraints on fitted coefficients? #8

Open tomwenseleers opened 7 years ago

tomwenseleers commented 7 years ago

In glmnet it is possible to impose positivity constraints on the fitted coefficients using the lower.limits argument and box constraints using the upper.limits argument. Would it be possible to incorporate this feature too in biglasso? Maybe just by clipping the coefficients to the specified bounds after each iteration or something?