Le 19 avril 2012 07:32, Paulo Henrique Junqueira Amorim
<[email protected]> a écrit :
> Hi,
>
> I'm trying to plot ROC curve from multiple multiclass. It seems to me that
> the library does not, how could get around it? the data I have are these:
>
>              precision    recall  f1-score   support
>          0       0.00      0.00      0.00         1
>          1       1.00      1.00      1.00         1
>          2       1.00      1.00      1.00         1
>          3       1.00      1.00      1.00         1
>          4       0.50      1.00      0.67         1
>          5       1.00      1.00      1.00         1
>              ...               ...
>        59       1.00      1.00      1.00         1
>        60       0.50      1.00      0.67         1
>
> avg / total   0.76      0.84      0.79        61

What mathematical definition of the multi-class ROC curve do you have
in mind? AFAIK it's only defined for the binary case.

-- 
Olivier
http://twitter.com/ogrisel - http://github.com/ogrisel
------------------------------------------------------------------------------
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to