[Octave-bug-tracker] [bug #65178] [octave forge] (statistics) pdist2 BIST failures

[Dmitri A. Sergatskov]

Follow-up Comment #8, bug #65178 (group octave):

See https://octave.discourse.group/t/pkg-test-package-fails/5160/53

Per Markus Mützel <mmuetzel>
<<<
If I understand correctly, some tests in BurrDistribution use random input:

## 'paramci' method
%!shared x
%! x = burrrnd (1, 1, 1, [1, 100]);
So, randomly failing tests might be a side effect.

Maybe, it would be better to set a seed so the “random” input is the same
for everyone.
>>>

Indeed:

for jj=1:100, [nn(jj),nmax] = test("BurrDistribution"); endfor
plot(nmax-nn)

Results are attached (tb.png).

Dmitri.
-- 



(file #55770)

    _______________________________________________________

Additional Item Attachment:

File name: tb.png                         Size:23 KB
    <https://file.savannah.gnu.org/file/tb.png?file_id=55770>


    AGPL NOTICE

These attachments are served by Savane. You can download the corresponding
source code of Savane at
https://git.savannah.nongnu.org/cgit/administration/savane.git/snapshot/savane-04b18e1179b14c0ef710c8e9c8e59d1e4bcd7bd0.tar.gz


    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?65178>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/