[Numpy-discussion] ANN: NumPy 1.5.0 beta 1
Pauli Virtanen
pav@iki...
Sun Aug 8 06:20:52 CDT 2010
Sun, 08 Aug 2010 15:09:00 +0800, Ralf Gommers wrote:
> On Mon, Aug 2, 2010 at 10:30 AM, Alan G Isaac <alan.isaac@gmail.com>
[clip]
>> ======================================================================
>> FAIL: test_umath_complex.TestCsqrt.test_special_values(<ufunc 'sqrt'>,
>> 1, inf, inf, inf)
>> test_umath_complex.TestCsqrt.test_special_values(<ufunc 'sqrt'>, 1,
>> inf, inf, inf)
>> ----------------------------------------------------------------------
>> Traceback (most recent call last):
>> File "C:\Python27\lib\site-packages\nose\case.py", line 183, in
>> runTest
>> self.test(*self.arg)
>> File
>> "C:\Python27\lib\site-packages\numpy\core\tests\test_umath_complex.py",
>> line 539, i
>> n check_complex_value
>> assert_equal(f(z1), z2)
>> File "C:\Python27\lib\site-packages\numpy\testing\utils.py", line
>> 256,
>> in assert_equal
>> return assert_array_equal(actual, desired, err_msg, verbose)
>> File "C:\Python27\lib\site-packages\numpy\testing\utils.py", line
>> 686,
>> in assert_array_e
>> qual
>> verbose=verbose, header='Arrays are not equal')
>> File "C:\Python27\lib\site-packages\numpy\testing\utils.py", line
>> 596,
>> in assert_array_c
>> ompare
>> raise AssertionError(msg)
>> AssertionError:
>> Arrays are not equal
>>
>> (x and y nan location mismatch [ True], False mismatch)
>> x: array([ inf nanj])
>> y: array((inf+infj))
>> >> raise AssertionError('\nArrays are not equal\n\n(x and y nan
>> >> location
>> mismatch [ True]
>> , False mismatch)\n x: array([ inf nanj])\n y: array((inf+infj))')
[clip]
> Pauli or Charles, can you please have a look at these? Looks like
> that's related to your recent work on nans/infs.
These errors probably come from the fact that the platform's C library
does not handle special nan/inf values in csqrt in the way C99 standard
requires. The recent changes in umath since 1.4.x are unrelated.
I think the support for platform-provided complex functions was enabled
in r7986.
To be sure, we'd need to see the build log.
--
Pauli Virtanen
More information about the NumPy-Discussion
mailing list