ok, i issued a pull request.
for CI reasons i also had to add the positivity option to the cross-validated estimators. furthermore i had to fix a couple of issues with doctest, also in doc. now all tests pass: https://github.com/scikit-learn/scikit-learn/pull/5131 cheers, michael On 18 Aug 2015, at 13:44, Joel Nothman <joel.noth...@gmail.com> wrote: > Please make a pull request. This looks like a small and useful change, > consistent with Lasso's support of non-negativity. > > On 18 August 2015 at 14:30, Michael Graber <michigra...@gmail.com> wrote: > > Dear all, > > I extended the lars_path, Lars and LarsLasso estimators in the scikit-learn > least_angle.py module with the possibility to restrict coefficients to be > 0 > using the method described in the original paper by Efron et al, 2004, > chapter 3.4. > > ( > https://github.com/scikit-learn/scikit-learn/compare/master...michigraber:nonnegative-lars > ) > > If you think this would be useful i could issue a pull request. > > I have not extended it to the cross-validated estimators yet but would be > willing to do so, if requested. > > Cheers, > Michael > ------------------------------------------------------------------------------ > _______________________________________________ > Scikit-learn-general mailing list > Scikit-learn-general@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general > > ------------------------------------------------------------------------------ > _______________________________________________ > Scikit-learn-general mailing list > Scikit-learn-general@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
------------------------------------------------------------------------------
_______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
