[NumPy-Tickets] [NumPy] #1482: erroneous output of misleading exceptions whenever in lapack interface working with non-native byte order

NumPy Trac numpy-tickets@scipy....
Mon May 10 20:39:17 CDT 2010


#1482: erroneous output of misleading exceptions whenever in lapack interface
working with non-native byte order
-------------------------------+--------------------------------------------
 Reporter:  yarikoptic         |       Owner:  somebody
     Type:  defect             |      Status:  new     
 Priority:  high               |   Milestone:  2.0.0   
Component:  Other              |     Version:          
 Keywords:  lapack endianness  |  
-------------------------------+--------------------------------------------
 Original report is at
 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=581043

 but now I've tested with current master: trunk@8404 and issue present as
 well, e.g:

 {{{
 $> python -c "import numpy as N; print N.__version__;print
 N.linalg.cholesky(N.array([[ 5.7998084,  -2.1825367 ], [-2.1825367,
 9.85910595]], dtype='>f8'))"
 2.0.0.dev
 Traceback (most recent call last):
   File "<string>", line 1, in <module>
   File
 "/home/yoh/proj/numpy/build/lib.linux-x86_64-2.5/numpy/linalg/linalg.py",
 line 517, in cholesky
     Cholesky decomposition cannot be computed'
 numpy.linalg.linalg.LinAlgError: Matrix is not positive definite -
 Cholesky decomposition cannot be computed

 $> python -c "import numpy as N; print N.__version__;print
 N.linalg.cholesky(N.array([[ 5.7998084,  -2.1825367 ], [-2.1825367,
 9.85910595]], dtype='<f8'))"
 2.0.0.dev
 [[ 2.40827914 -0.        ]
  [-0.90626401  3.00629199]]
 }}}

 myself I consider it of higher than normal priority since the issue is
 quite nastily veiled (some functions, e.g. svd would just produce bogus
 results) and I am surprised noone seems hit it by now.

-- 
Ticket URL: <http://projects.scipy.org/numpy/ticket/1482>
NumPy <http://projects.scipy.org/numpy>
My example project


More information about the NumPy-Tickets mailing list