Hi.
The CCA is not very well maintained unfortunately.
I am not very familiar with the method. It is part of the PLS module iirc,
which is in a non-optimal state.
I am not sure if any of the active developers know much about the CCA
implementation.
It would be great to bring this part of the code up to the standards for
the rest of the package.
If you have any insights or want to hack on the code, that would be very
welcome.
Sorry I can't be of more help atm.
Cheers,
Andy
On 02/09/2013 01:07 AM, ML Fan wrote:
Hi All,
This is relation to the earlier mail about CCA that I sent.
I have tried CCA on three different platforms -- R, MATLAB and
scikit-learn.
On R and MATLAB they almost match but the Python result is way way off.
The data I am using is from here:
http://www.ats.ucla.edu/stat/r/dae/canonical.htm
I have also shared my source codes in R, Python and MATLAB.
If anyone has any clue what's going on then I would be happy to
chat/discuss.
Thanks in advance !!
Thanks !!
------------------------------------------------------------------------------
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen firewall before the end March 2013
and get the hardware for free! Learn more.
http://p.sf.net/sfu/sophos-d2d-feb
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
------------------------------------------------------------------------------
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen firewall before the end March 2013
and get the hardware for free! Learn more.
http://p.sf.net/sfu/sophos-d2d-feb
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general