[pymvpa] searchlight gets stuck, issues error

[Sergey.V.Fogelson at Dartmouth.edu]

Hello,

I tried using removeInvariantFeatures(dataset) as was suggested and  
got the same error. It is as follows:

[SLC] DBG:  Doing 22526 spheres: 9463 (58 features) [42%]WARNING:  
Obtained degenerate data with zero norm for training of  
LinearCSVMC(kernel_type='linear', svm_impl='C_SVC').  Scaling of C  
cannot be done.
  * Please note: warnings are printed only once, but underlying  
problem might occur many times *

After issuing this error the searchlight has continued operating... I  
will see if it saves the map, however, as ive also been getting a  
Memory Error (which I posted separately to the mailing list) after the  
searchlight finishes running.

Quoting Yaroslav Halchenko <debian at onerussian.com>:

>
>
>> I had upgraded to 0.4.4 when I reported this bug and whereas before
>> it just got stuck at 68% (with 0.4.2), the new version actually
>> issues the error I reported before.
> very good ;) no stuck -- good sign ;)
>
>> Thus, the fix via upgrade doesnt
>> seem to do the trick. i will try to do the mask checking as
>> recommended.
> or just use removeInvariantFeatures function/method of dataset to get
> rid of those invariant features -- see how many would it prune (I expect
>  >= 1 ;))
>
>
>> thanks,
>> Sergey.
> --
>                                   .-.
> =------------------------------   /v\  ----------------------------=
> Keep in touch                    // \\     (yoh@|www.)onerussian.com
> Yaroslav Halchenko              /(   )\               ICQ#: 60653192
>                    Linux User    ^^-^^    [175555]
>
>
>
> _______________________________________________
> Pkg-ExpPsy-PyMVPA mailing list
> Pkg-ExpPsy-PyMVPA at lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/pkg-exppsy-pymvpa
>