Actually the other discussion was specifically about multiclass ROC AUC. I am not sure this could be generalized to a multilabel setting.
It is my understanding that multilabel problems are generally evaluated with information retrieval oriented metrics like precision, recall, f1 score which are already implemented in scikit-learn and support both the multiclass and the multilabel settings with micro and macro averaging. -- Olivier ------------------------------------------------------------------------------ Get your SQL database under version control now! Version control is standard for application code, but databases havent caught up. So what steps can you take to put your SQL databases under version control? Why should you start doing it? Read more to find out. http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk _______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
