2012/1/14 Gael Varoquaux <[email protected]>:
> On Wed, Dec 21, 2011 at 06:36:03PM -0000, [email protected]
> wrote:
>> first thing I would like to say that I'm not so experienced with python
>> therefore I might do something really stupid which I cannot see.
>> Nevertheless I don't manege to understand why the following script throw
>> an assertion error.
>> Indeed, training the one class svm with a dataset or with a shuffled one
>> seems to give two different results.
>
> As Alex answered, the problem is that the algorithm didn't converge to
> the optimum to a high tolerance. So what you are seeing are optimization
> errors that depend on the starting point.
Do you think that the default tolerance set in scikit-learn is inadequate?
>>> OneClassSVM()
OneClassSVM(cache_size=200, coef0=0.0, degree=3, gamma=0.0, kernel='rbf',
nu=0.5, shrinking=True, tol=0.001)
--
Olivier
http://twitter.com/ogrisel - http://github.com/ogrisel
------------------------------------------------------------------------------
RSA(R) Conference 2012
Mar 27 - Feb 2
Save $400 by Jan. 27
Register now!
http://p.sf.net/sfu/rsa-sfdev2dev2
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
