2012/7/29 Olivier Grisel <[email protected]>:
> 2012/7/29 JAGANADH G <[email protected]>:
>> Is there any support for string kernal + svm in scikit-learn
>
> Nope but if you can store the complete pairwise kernel matrix of your
> dataset you can use the kernel='precomputed' option of SVC, NuSVC, SVR
> and NuSVR.
>
> In that case you can use any python function that computes the kernel
> value for any pair of samples and store the results in a (n_samples,
> n_samples)-shaped numpy array to feed to your model.

Except that the code, as it stands, does try to convert predict's
argument to a NumPy array of floats...

-- 
Lars Buitinck
Scientific programmer, ILPS
University of Amsterdam

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to