[SciPy-user] Incorrect results in linalg.solve and friends

Pearu Peterson pearu at scipy.org
Mon Feb 23 17:04:36 CST 2004


Hi,

On Sun, 22 Feb 2004, Pearu Peterson wrote:

> This is heads-up for linalg crunchers. Incorrect results appear when
> the following condition holds:
> 
>   an input array is a *slice* of a complex array.

More accurate condition for the bug to show up is:

   an input array is an one-dimensional slice of a complex array

> This gives a workaround to avoid this problem until it gets resolved.
> Also, this is not due to some bug in Atlas but rather of copy_ND_array in
> fortranobject.c or of PyArray_CopyFromObject, but most probably of
> copy_ND_array.

This issue is now resolved. Upgrading f2py to the latest is highly
recommended, especially when computations use complex inputs to f2py
generated wrapper functions.

Regards,
Pearu



More information about the SciPy-user mailing list