Another remark is that you set C=1e3. Depending on the scaling of your
data, this can be quite large. This means that the SVM is very lightly
regularized (=> hard SVM) and therefore the problem is ill-conditioned.
Mathieu
On Thu, Apr 21, 2016 at 11:51 PM, Mathieu Blondel <math...@mblondel.org>
wrote:
> By default, SVC stops only when the desired tolerance is reached. If the
> problem is poorly scaled, this can indeed take ages. You can however set
> max_iter to prevent this.
>
> http://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html
>
> We might want to change the default from -1 to something large like 1000.
>
> Mathieu
>
> On Thu, Apr 21, 2016 at 11:39 PM, Olivier Grisel <olivier.gri...@ensta.org
> > wrote:
>
>> Indeed, that is not a bug then.
>>
>> --
>> Olivier
>>
>>
>> ------------------------------------------------------------------------------
>> Find and fix application performance issues faster with Applications
>> Manager
>> Applications Manager provides deep performance insights into multiple
>> tiers of
>> your business applications. It resolves application problems quickly and
>> reduces your MTTR. Get your free trial!
>> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
>> _______________________________________________
>> Scikit-learn-general mailing list
>> Scikit-learn-general@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>>
>
>
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
