On 11/03/2011 06:19 PM, Peter Prettenhofer wrote: > Can you try if it segfaults on X[5000:6000] and if so, it would be > great if you could send me the ndarray - then I can reproduce the > error and see what's wrong. > As I said, the features consist of three distinct parts and if I use any of those, it works. Since 5000:6000 is a subset of the second split (each is 13824/3), I guess it will work. I can still try, though.
> 2011/11/3 Andreas Müller <[email protected]>: >> On 11/03/2011 06:10 PM, Peter Prettenhofer wrote: >>> Hi Andreas, >>> >>> can you run the following line and send me the results (assume your >>> data is stored in `X`):: >>> >>> print X.flags >>> >> train_data.flags >> C_CONTIGUOUS : True >> F_CONTIGUOUS : False >> OWNDATA : True >> WRITEABLE : True >> ALIGNED : True >> UPDATEIFCOPY : False >> >> Seems reasonable to me. >> Thanks, >> Andy >> >>> 2011/11/3 Andreas Müller <[email protected]>: >>>> Hi folks. >>>> Today I ran across a segfault doing sgd multi class classification. >>>> I tracked the error down to sgd_fast but don't know how to proceed. >>>> My data has shape (50000, 13824) and is dense, the parameters >>>> of the classifier are: >>>> SGDClassifier(loss="hinge", penalty="l2") >>>> >>>> I ran it through scaler and the features seem to be in a reasonable >>>> range and there are no infs or nans. >>>> If I train only on the first 5000 of the 13824 features, I get >>>> reasonable results, for 6000 it segfaults. The features split >>>> into three parts of equal lenght and training on any of >>>> these parts works fine. >>>> >>>> My RAM is big and empty during the crash. >>>> >>>> Does anyone have any idea how to find the problem? >>>> >>>> Cheers, >>>> Andy >>>> >>>> ------------------------------------------------------------------------------ >>>> RSA(R) Conference 2012 >>>> Save $700 by Nov 18 >>>> Register now >>>> http://p.sf.net/sfu/rsa-sfdev2dev1 >>>> _______________________________________________ >>>> Scikit-learn-general mailing list >>>> [email protected] >>>> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general >>>> >>> >> >> ------------------------------------------------------------------------------ >> RSA(R) Conference 2012 >> Save $700 by Nov 18 >> Register now >> http://p.sf.net/sfu/rsa-sfdev2dev1 >> _______________________________________________ >> Scikit-learn-general mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general >> > > ------------------------------------------------------------------------------ RSA(R) Conference 2012 Save $700 by Nov 18 Register now http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
