Indeed this is quite a new method and we have a policy to wait a bit to see if it's actually practically useful before including an implementation in the code base.
Michal, if you have replicated the results of the paper in Python it would be interesting to publish your code in a scikit-learn style class that implements the Transformer interface as SparseRandomProjection does. You can publish this code on your own github repo. Over time if people find it significantly more useful than SparseRandomProjection (assuming it address similar use cases, I have not read the paper yet) me might decide to include it directly into scikit-learn. -- Olivier ------------------------------------------------------------------------------ _______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
