Hi guys,
I was testing (version 0.15.1) the AdditiveChi2Sampler kernel approx method
and it seems that the output has a different size wrt the size cited in the
documentation.
The documentation said that it should be (n_examples, (2n+1)*n_features)
and I'm getting (n_examples, (2n-1)*n_features). Is an error on the
documentation or I missed something?
I checked part of code and it seems that if I use n>4, I need to provide
interval_samples. In case of an error, Is the sample_interval correct for
n<4??
Thank you for your effort building/maintaining this project.
--
VĂctor A. Escorcia Castillo, Eng.
Visiting MS Student at KAUST <http://www.kaust.edu.sa/>
Cellphone: +966-569485932
WhatsApp: +57-3012447606
mail: [email protected]
------------------------------------------------------------------------------
Slashdot TV.
Video for Nerds. Stuff that matters.
http://tv.slashdot.org/
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general