I believe random subspace ensembles are subsumed by the BaggingClassifier / BaggingRegressor estimators. See the class documentation. The proportion of features used is controlled by max_features.
M. On Mon, Aug 18, 2014 at 8:51 AM, Dayvid Victor <[email protected]> wrote: > Hello Everybody, > > I was looking for a Random Subspace implementation and found only a > Random Forest. Correct me if I'm wrong but the Random Subspace is a > generalization of the Random Forest (it works with classifiers not based > on decision tree). > > Is there any implementation of Random Subspace available? If not, is it > worth including into sklearn? And should this replace the random forest > (Random Forest would use inheritance)? > > PS: I just started working with Classifier Ensemble, so, I'm sorry if I > wrote something wrong here, just started reading the Kuncheva book (btw, > any other suggestion is welcome). > > Thanks, > -- > *Dayvid Victor R. de Oliveira* > PhD Candidate in Computer Science at Federal University of Pernambuco > (UFPE) > MSc in Computer Science at Federal University of Pernambuco (UFPE) > BSc in Computer Engineering - Federal University of Pernambuco (UFPE) > > > ------------------------------------------------------------------------------ > > _______________________________________________ > Scikit-learn-general mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general > >
------------------------------------------------------------------------------
_______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
