I'm looking to do regularized regression with a non-negativity constraint. Scikit-learn's Lasso method has a 'positive' option that applies this constraint, so it seems like a good tool for the job. At the same time, the automatic tuning of the regularization parameter that is offered by LassoCV would also be useful for my problem. Unfortunately, the 'positive' option is not available for LassoCV (nor for alternative methods like LassoLarsCV or ElasticNetCV).
Is there a reason why this option could not be included in LassoCV? Apparently someone else had the same question: http://stackoverflow.com/questions/14324976/what-non-negative-linear-models-are-supported-planned-in-scikit-learn I hope it can be included in a future release. I suppose in the meantime I could try to write my own wrapper for Lasso to do the tuning by cross-validation using the scikit-learn cross-validation iterators, basically trying to replicate LassoCV myself but with the 'positive' option. Does anyone know where I can find more details about how LassoCV uses cross-validation? The LassoCV page itself gives some information but not enough for me to feel confident implementing it myself. ------------------------------------------------------------------------------ Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET Get 100% visibility into your production application - at no cost. Code-level diagnostics for performance bottlenecks with <2% overhead Download for free and get started troubleshooting in minutes. http://p.sf.net/sfu/appdyn_d2d_ap1 _______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
