On 07/29/2012 01:06 AM, eat wrote:
Hi,

On Sat, Jul 28, 2012 at 7:48 PM, Jeffrey <zfyuan@mail.ustc.edu.cn> wrote:
On 07/29/2012 12:23 AM, Jeffrey wrote:
> Dear all,
>
>     The sentences bellow will always raise an Error or Exception just
> as follows, which is a little anomaly. Is this a bug?
>
>     >>> u1=numpy.random.rand(100000)
>     >>> u2=numpy.random.rand(100000)
>     >>> scipy.stats.kendalltau(u1,u2)
> ---------------------------------------------------------------------------
>
> AttributeError                            Traceback (most recent call
> last)
> /home/zfyuan/phd/paper1/pyvine_lap/<ipython-input-28-98f367090ed1> in
> <module>()
> ----> 1 sp.stats.kendalltau(u1,u2)
>
> /usr/lib64/python2.7/site-packages/scipy/stats/stats.pyc in
> kendalltau(x, y, initial_lexsort)
>    2673
>    2674     tau = ((tot - (v + u - t)) - 2.0 * exchanges) / \
> -> 2675                     np.sqrt((tot - u) * (tot - v))
>    2676
>    2677     # what follows reproduces the ending of Gary Strangman's
> original
>
>
> AttributeError: sqrt
>

Sorry, I didn't describe this bug with details. What I mean is that when
the two array have larger length, for example with length 100000, then
it is more possible that the Error would occur.

My scipy version is 0.9.0 and numpy is 1.6.2.

Thanks a lot for your answering.
I can confirm this, like
In []: os.sys.version
Out[]: '2.7.2 (default, Jun 12 2011, 15:08:59) [MSC v.1500 32 bit (Intel)]'
In []: np.version.version
Out[]: '1.6.0'
In []: sp.version.version
Out[]: '0.9.0'

In []: stats.kendalltau(rand(77929), rand(77929))
Out[]: (0.0060807135427758865, 0.010891543687108114)
In []: stats.kendalltau(rand(77939), rand(77939))
------------------------------------------------------------
Traceback (most recent call last):
  File "<ipython console>", line 1, in <module>
  File "C:\Python27\lib\site-packages\scipy\stats\stats.py", line 2675, in kendalltau
    np.sqrt((tot - u) * (tot - v))
AttributeError: sqrt

There really seems to be odd problem above a certain length of arrays.


My 2 cents,
-eat 

--

Jeffrey


_______________________________________________
SciPy-User mailing list
SciPy-User@scipy.org
http://mail.scipy.org/mailman/listinfo/scipy-user



_______________________________________________
SciPy-User mailing list
SciPy-User@scipy.org
http://mail.scipy.org/mailman/listinfo/scipy-user

Thanks eat. I found the reason is that numpy.sqrt cannot deal with too large number. When calculating kendalltau, assume n=len(x),then the total pair number is 'tot' below:

    tot=(n-1)*n//2

when calculating tau, the de-numerator is as below:

    np.sqrt((tot-u)*(tot-v))

u and v stands for ties in x[] and y[perm[]], which is zero if the two array sample from continuous dist. Hence (tot-u)*(tot-v) may be out of range for the C written ufunc 'np.sqrt', and an Error is then raised.

What about using math.sqrt here, or multiply two np.sqrt in the de-numerator? Since big data sets are often seen these days.

Thanks a lot !
  

-- 
Jeffrey