Hi, If you want to get `sample_weights` working with the current master, the easiest is to take PR 3524 and either pass it through `fit_params` or just undo the last commit in the branch.
I needed to change a couple of things to get 1574 up to date with the current master, but nothing else is changed. Best, Vlad On Sun, Aug 3, 2014 at 6:47 AM, Joel Nothman <[email protected]> wrote: > You could use the implementation of sample_weight support in > cross-validation from > https://github.com/scikit-learn/scikit-learn/pull/1574, which should work > but doesn't have much in the way of tests. It may be superseded by > https://github.com/scikit-learn/scikit-learn/pull/3524 > > > On 3 August 2014 14:05, Michael Anuzis <[email protected]> wrote: >> >> Greetings scikit-learn-general, >> >> Long time fan of scikit-learn, but my first post on this list. Appreciate >> any insight or guidance you can share. >> >> I'm trying to implement my own scoring function for cross_val_score using >> the make_scorer factory function. The evaluation I'm trying to build is for >> a normalized weighted gini, which requires a "weights" array passed in to >> calculate the relative weight of each row. >> >> The function works fine when called manually with (y_true, y_pred), but >> raises an AssertionError when called as a parameter on >> cross_val_score(scoring=gini_function)... I think the error might be due to >> the weights array being its original size, while the (y_true, y_pred) arrays >> might be reduced sizes for cross-validation folds. >> >> Here's my code and the stack trace: >> >> https://dpaste.de/Zk6n >> >> Appreciate any suggestions, >> -Michael >> >> >> ------------------------------------------------------------------------------ >> Want fast and easy access to all the code in your enterprise? Index and >> search up to 200,000 lines of code with a free copy of Black Duck >> Code Sight - the same software that powers the world's largest code >> search on Ohloh, the Black Duck Open Hub! Try it now. >> http://p.sf.net/sfu/bds >> _______________________________________________ >> Scikit-learn-general mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general >> > > > ------------------------------------------------------------------------------ > Want fast and easy access to all the code in your enterprise? Index and > search up to 200,000 lines of code with a free copy of Black Duck > Code Sight - the same software that powers the world's largest code > search on Ohloh, the Black Duck Open Hub! Try it now. > http://p.sf.net/sfu/bds > _______________________________________________ > Scikit-learn-general mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general > ------------------------------------------------------------------------------ Want fast and easy access to all the code in your enterprise? Index and search up to 200,000 lines of code with a free copy of Black Duck Code Sight - the same software that powers the world's largest code search on Ohloh, the Black Duck Open Hub! Try it now. http://p.sf.net/sfu/bds _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
