On 01/09/2012 12:19 PM, Mathieu Blondel wrote:
> On Mon, Jan 9, 2012 at 7:40 PM, Gael Varoquaux
> <[email protected]>  wrote:
>
>    
>> Can it be that if nu is None C is used, or do you thihnk that this is
>> confusing?
>>      
> That wouldn't work for NuSVR, which uses both C and nu (I know, it's 
> confusing).
>
> If we merge C-SVM and nu-SVM classes, users may think that they can
> obtain better results by switching from one formulation to the other.
> If this is not the case, this may be confusing.
>
>    
 From my understanding, the results should not really change.
I don't know if any one has any actual experience with that.

Alternative to "algorithm" (which I feel is not really appropriate
as both are solved either using SMO or CG), we could
use "parametrization" or "formulation".

I am undecided about using "nu" as the switching parameters.
Generally I think less parameters are always better.

Turning it around in my head, it might be a good idea (if
documented correctly - which should be pretty easy)
"""nu: If this parameter is set, use "nu-SVM" formulation instead of
standard "C-SVM" parametrization.
Trades of generalization and training error. See docs."""

Yeah, I could live with that.

------------------------------------------------------------------------------
Ridiculously easy VDI. With Citrix VDI-in-a-Box, you don't need a complex
infrastructure or vast IT resources to deliver seamless, secure access to
virtual desktops. With this all-in-one solution, easily deploy virtual 
desktops for less than the cost of PCs and save 60% on VDI infrastructure 
costs. Try it free! http://p.sf.net/sfu/Citrix-VDIinabox
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to