[SciPy-Dev] ANN: SciPy 0.8.0 beta 1

Vincent Davis vincent@vincentdavis....
Sat Jun 12 14:28:29 CDT 2010


On Sat, Jun 12, 2010 at 1:22 PM,  <josef.pktd@gmail.com> wrote:
> On Sat, Jun 12, 2010 at 3:02 PM, Vincent Davis <vincent@vincentdavis.net> wrote:
>> On Fri, Jun 11, 2010 at 6:41 PM,  <josef.pktd@gmail.com> wrote:
>>> On Fri, Jun 11, 2010 at 7:54 PM, Derek Homeier
>>> <derek@astro.physik.uni-goettingen.de> wrote:
>>>> Hi Josef,
>>>>
>>>>>> FAIL: test_stats.test_kstest
>>>>>> ----------------------------------------------------------------------
>>>>>> Traceback (most recent call last):
>>>>>>  File "/sw/lib/python2.6/site-packages/nose/case.py", line 186, in runTest
>>>>>>    self.test(*self.arg)
>>>>>>  File "/sw/lib/python2.6/site-packages/scipy/stats/tests/test_stats.py", line 1078, in test_kstest
>>>>>>    np.array((0.0072115233216310994, 0.98531158590396228)), 14)
>>>>>>  File "/sw/lib/python2.6/site-packages/numpy/testing/utils.py", line 441, in assert_almost_equal
>>>>>>    return assert_array_almost_equal(actual, desired, decimal, err_msg)
>>>>>>  File "/sw/lib/python2.6/site-packages/numpy/testing/utils.py", line 765, in assert_array_almost_equal
>>>>>>    header='Arrays are not almost equal')
>>>>>>  File "/sw/lib/python2.6/site-packages/numpy/testing/utils.py", line 609, in assert_array_compare
>>>>>>    raise AssertionError(msg)
>>>>>> AssertionError:
>>>>>> Arrays are not almost equal
>>>>>>
>>>>>> (mismatch 100.0%)
>>>>>>  x: array([ 0.007,  0.985])
>>>>>>  y: array([ 0.007,  0.985])
>>>>>
>>>>> maybe the precision (decimal 14) is too high for this test across platforms
>>>>>
>>>>> Could you check how large the difference is ?
>>>>>
>>>>> np.random.seed(987654321)
>>>>> x = stats.norm.rvs(loc=0.2, size=100)
>>>>> np.array(stats.kstest(x,'norm', alternative = 'greater')) -
>>>>>                np.array((0.0072115233216310994, 0.98531158590396228))
>>>>>
>>>>> (my line numbers differ, but this should be the right test given your numbers)
>>>>
>>>>
>>>> yes, just a decimal or two too high, if I got the numbers right:
>>>> # OS X 10.5 i386 / 10.6 x86_64:
>>>> array([  8.67361738e-18,   1.66533454e-15])
>>>>
>>>> # OS X 10.5 ppc:
>>>> array([  2.05955045e-13,  -7.16759985e-13])
>>>
>>> interesting that there are differences in the calculations, but for
>>> the test we can just reduce the precision to decimal=12 to avoid the
>>> test failure.
>>
>> I must be doing something wrong here becuase I don't get anything
>> close that what you have above.

>> In [4]: np.random.seed(987654321)
>>
>> In [5]: x = stats.norm.rvs(loc=0.2, size=100)
>>
>> In [6]: r1 = np.array(stats.kstest(x,'norm', alternative = 'greater'))
>>
>> In [7]: r2 = np.array((0.0072115233216310994, 0.98531158590396228))
>>
>> In [8]: r1-r2
>> Out[8]: array([ 0.03704986, -0.32866092])
>
>>>> np.random.seed(987654321)
>>>> xrvs = stats.norm.rvs(loc=0.2, size=100)
>>>> r1 = np.array(stats.kstest(xrvs,'norm', alternative = 'greater'))
>>>> r2 = np.array((0.0072115233216310994, 0.98531158590396228))
>>>> r1-r2
> array([  8.67361738e-18,   1.66533454e-15])
>
> Can you check mean and var to see if you have the same random  numbers?
>
>>>> xrvs.mean()
> 0.20830662128271851
>>>> xrvs.var()
> 1.1210385272356511

In [11]: x.mean()
Out[11]: 0.054996065027031464

In [12]: x.var()
Out[12]: 0.92731406990162746

I am cheating and using the enthought distribution, I just click install.
How do I run all of the tests for scipy or numpy when they are already
installed?

Vincent

>
> otherwise I have no clue, (but I guess your scipy.stats tests pass)
>
> Josef
>
>
>>
>> Vincent
>>
>>
>>>
>>> Thanks,
>>> Josef
>>>
>>>>
>>>> Cheers,
>>>>                                                Derek
>>>>
>>>> _______________________________________________
>>>> SciPy-Dev mailing list
>>>> SciPy-Dev@scipy.org
>>>> http://mail.scipy.org/mailman/listinfo/scipy-dev
>>>>
>>> _______________________________________________
>>> SciPy-Dev mailing list
>>> SciPy-Dev@scipy.org
>>> http://mail.scipy.org/mailman/listinfo/scipy-dev
>>>
>> _______________________________________________
>> SciPy-Dev mailing list
>> SciPy-Dev@scipy.org
>> http://mail.scipy.org/mailman/listinfo/scipy-dev
>>
> _______________________________________________
> SciPy-Dev mailing list
> SciPy-Dev@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-dev
>


More information about the SciPy-Dev mailing list