[SciPy-User] scipy 0.9 fails tests in windows XP

[k0ala]

Ralf Gommers <ralf.gommers <at> googlemail.com> writes:

> 
> This one can't be reproduced on most other systems, it may have something to
> do with your ATLAS or hardware. See
http://mail.scipy.org/pipermail/scipy-dev/2011-January/015868.html
> 
> 
> Ran 4728 tests in 69.475s
> FAILED (KNOWNFAIL=12, SKIP=42, failures=2)
> <nose.result.TextTestResult run=4728 errors=0 failures=2>
> Is this going to be a problem for using scipy? Do you have any recommendations
> for getting it to work, or for doing a cleaner install?
> 
> That one failure isn't likely to give you problems. If you want to investigate
> anyway, you can start with checking the questions by Pauli in the thread I
> linked to.Cheers,Ralf
> 


Hi Ralf, Thanks a lot for the quick reply! I will follow up on it, because if I
understand correctly, if this problem were to occur while I am using the
software, it would not raise any flags or warnings, just output incorrect
results. I'm not really comfortable using the library knowing that sometimes
eigenvector computations are way off...

If I find any solutions, I will follow up on this post.

Best,
k0ala