I just realized that I was setting max_iter in StratifiedKFold, not in SVC(), 
so problem solved when I do SVC(max_iter=100). Sorry for the confusion.

Ev


On Sunday, July 6, 2014 12:44 PM, Eva Nuenno <[email protected]> wrote:



I have a precomputed 250x250 kernel matrix, and I am running grid search with:
C_range =10.**np.arange(-2,9)param_grid =dict(C=C_range)grid 
=GridSearchCV(SVC(kernel='precomputed'),param_grid=param_grid,cv=StratifiedKFold(y=data_label,n_folds=10,max_iter=100))grid.fit(kernel,data_label)
However, the code never finishes (I waited for 6 hours), even though I 
specified max_iter option (so I was expecting that the optimization will be 
finished after it reaches to max_iter).
Is there any suggestion on how to fix it? The problem is that: since the code 
just hangs there, grid search never finishes to find a good C value. So instead 
of waiting forever, I want it to stop after some #iterations, and proceed to 
the other C values. How can I do that?
Note: I normalize the kernel with K[i, j]/sqrt(K[i,i]*K[j,j])

I am using scikit-learn '0.14.1'.

Thanks,
Ev

------------------------------------------------------------------------------
Open source business process management suite built on Java and Eclipse
Turn processes into business applications with Bonita BPM Community Edition
Quickly connect people, data, and systems into organized workflows
Winner of BOSSIE, CODIE, OW2 and Gartner awards
http://p.sf.net/sfu/Bonitasoft
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to