Re: [Scikit-learn-general] RandomForestRegressor max_features default

Hi Sebastian, Yes. This is intentional. The motivation comes from http://link.springer.com/article/10.1007/s10994-006-6226-1#/page-1 where it is shown experimentally that is a good default value on average. Gilles On 13 November 2015 at 11:17, Sebastian Raschka wrote: >

Re: [Scikit-learn-general] RandomForestRegressor max_features default

Oh, nice, that’s good to know! Many thanks! > On Nov 13, 2015, at 5:58 AM, Gilles Louppe wrote: > > Hi Sebastian, > > Yes. This is intentional. The motivation comes from > http://link.springer.com/article/10.1007/s10994-006-6226-1#/page-1 > where it is shown experimentally

[Scikit-learn-general] RandomForestRegressor max_features default

Hi, it’s probably intended, but I just wanted to mention that I just saw that the RandomForestRegressor defaults are set to “regular” bagging for regression. • If “auto”, then max_features=n_features.

Re: [Scikit-learn-general] Scikit-learn-general Digest, Vol 70, Issue 11

On Fri, Nov 13, 2015 at 6:01 AM, < scikit-learn-general-requ...@lists.sourceforge.net> wrote: > It's not a good idea, but i'll post it: > A man after my own heart :-) > Seems that it's possible to easily create StackingClassifier even in > current master. > But a Pipeline can't be used as a