[SciPy-User] SciPy+NumPy on 4 major linux distributions

[Ivo Maljevic]

So, numpy works very well, but scipy.test() fails, and I think it is the
same failure across the distributions.
Anyone knows what does this mean:

======================================================================
ERROR: test_implicit (test_odr.TestODR)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/usr/lib64/python2.6/site-packages/scipy/odr/tests/test_odr.py",
line 88, in test_implicit
    out = implicit_odr.run()
  File "/usr/lib64/python2.6/site-packages/scipy/odr/odrpack.py", line 1055,
in run
    self.output = Output(apply(odr, args, kwds))
TypeError: y must be a sequence or integer (if model is implicit)

----------------------------------------------------------------------
Ran 3395 tests in 51.646s

FAILED (KNOWNFAIL=3, SKIP=28, errors=1)
<nose.result.TextTestResult run=3395 errors=1 failures=0>




2009/9/4 Ivo Maljevic <ivo.maljevic at gmail.com>

> For what is worth, I've updated liblapack and libblas to 3.2.1 under
> openSUSE and numpy runs properly now.
>
> It looks like Lev Givon is making similar progress with Mandriva, so the
> situation is not as bad as I thought
> initially.
>
> 2009/9/3 Lev Givon <lev at columbia.edu>
>
> Received from Chris Colbert on Tue, Sep 01, 2009 at 12:55:34PM EDT:
>> > What I would like to see is a distribution build numpy and scipy with
>> > threaded atlas support.
>> >
>> > As it stands currently, Ubuntu "has atlas support", but its not
>> > threaded, and the packages are broken...
>> >
>> > Until that happens, I'll be rolling my own numpy and scipy from source.
>>
>> Mandriva's atlas packages consist of library rpms prebuilt for several
>> stock architectures [1] that are installed along with an ld.so
>> override that causes programs that dynamically link to a blas/lapack
>> shared library to call its atlas equivalent. The source srpm can also
>> be rebuilt on one's system so as to obtain a properly tuned library.
>>
>> I'm not sure whether the current prebuilt libraries are built with
>> thread support. I'm also not sure whether the current (3.8.3) prebuilt
>> libraries consistently provide any improved performance compared to
>> the netlib blas/lapack.
>>
>>                                                   L.G.
>>
>> [1] This scheme is also used in Debian's atlas packages.
>>
>>
>> _______________________________________________
>> SciPy-User mailing list
>> SciPy-User at scipy.org
>> http://mail.scipy.org/mailman/listinfo/scipy-user
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20090904/ca549be1/attachment.html>