>Does the test time matter for Debian or Linux packaging?
I don't remember what is the cut-off for extremely long build-times but
it is hours (if not days), so I don't think that time issue is
relevant for pystatsmodels at the moment ;)
>With models that require a large number of iterati
On Thu, Mar 29, 2012 at 12:47 PM, Yaroslav Halchenko wrote:
> ah -- and found that you adopted it already:
> https://github.com/statsmodels/statsmodels/issues/109
>
> https://github.com/statsmodels/statsmodels/commit/590bd6238d7d2538a557ce664ee90f4b9116449c
>
5 months ago
>
> cool -- thanks ;)
ah -- and found that you adopted it already:
https://github.com/statsmodels/statsmodels/issues/109
https://github.com/statsmodels/statsmodels/commit/590bd6238d7d2538a557ce664ee90f4b9116449c
cool -- thanks ;)
On Thu, 29 Mar 2012, Yaroslav Halchenko wrote:
> d'oh forgotten about it -- thanks Grego
d'oh forgotten about it -- thanks Gregor
pystatsmodels people -- would you like to adopt this patch:
+--- statsmodels-0.3.1.orig/scikits/statsmodels/discrete/tests/test_discrete.py
2011-11-15 09:57:09.346314408 -0500
statsmodels-0.3.1/scikits/statsmodels/discrete/tests/test_discrete.py
20
On Wed, 14 Mar 2012 16:57:37 +0100, Jakub Wilk wrote:
> Source: statsmodels
> Version: 0.3.1-3
> Severity: serious
> Justification: fails to build from source
Ubuntu has a patch at
http://patches.ubuntu.com/s/statsmodels/statsmodels_0.3.1-3ubuntu1.patch
not sure if this is a good and/or future-pr
Source: statsmodels
Version: 0.3.1-3
Severity: serious
Justification: fails to build from source
statsmodels FTBFS on i386:
| ==
| FAIL:
scikits.statsmodels.discrete.tests.test_discrete.TestProbitCG.test_conf_int
| -
6 matches
Mail list logo
