[SciPy-user] negative values in diagonal of covariance matrix

David Cournapeau david@ar.media.kyoto-u.ac...
Thu Dec 11 22:49:55 CST 2008


Pauli Virtanen wrote:
>
> Ideally, the algorithm should be robust enough to know that it didn't 
> converge in this case, and signal failure (success > 4). leastsq is a 
> wrapper to MINPACK's LMDER, so this is not completely straightforward to 
> debug.
>   

Do you mean the python wrapper miss some diagnostic information
available in fortran ? Otherwise, would it make sense to check the ier
value from fortran and at least generate a warning about failed
convergence ?

David



More information about the SciPy-user mailing list