You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@commons.apache.org by "greg sterijevski (Created) (JIRA)" <ji...@apache.org> on 2011/10/01 04:00:45 UTC
[jira] [Created] (MATH-678) NonLinear Optimizers seem to have a
hard time hitting NIST standards
NonLinear Optimizers seem to have a hard time hitting NIST standards
--------------------------------------------------------------------
Key: MATH-678
URL: https://issues.apache.org/jira/browse/MATH-678
Project: Commons Math
Issue Type: Bug
Affects Versions: 3.0
Environment: Java
Reporter: greg sterijevski
Assignee: greg sterijevski
Priority: Minor
As per a discussion on the mailing list, I am opening this ticket. In applying the nonlinear optimizers in commons, I noticed what I believe to be instability in the techniques. Further investigation investigation (both of my tests) and the code in prod is warranted.
I will be pushing a first set of tests which should illustrate what I am seeing.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (MATH-678) NonLinear Optimizers seem to have a
hard time hitting NIST standards
Posted by "Gilles (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MATH-678?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Gilles updated MATH-678:
------------------------
Assignee: (was: greg sterijevski)
Greg Sterijevski does not seem to be around this forum anymore.
> NonLinear Optimizers seem to have a hard time hitting NIST standards
> --------------------------------------------------------------------
>
> Key: MATH-678
> URL: https://issues.apache.org/jira/browse/MATH-678
> Project: Commons Math
> Issue Type: Bug
> Affects Versions: 3.0
> Environment: Java
> Reporter: greg sterijevski
> Priority: Minor
> Labels: NIST, NonLinear, Optimization
>
> As per a discussion on the mailing list, I am opening this ticket. In applying the nonlinear optimizers in commons, I noticed what I believe to be instability in the techniques. Further investigation investigation (both of my tests) and the code in prod is warranted.
> I will be pushing a first set of tests which should illustrate what I am seeing.
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MATH-678) NonLinear Optimizers seem to have a
hard time hitting NIST standards
Posted by "Gilles (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MATH-678?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13416262#comment-13416262 ]
Gilles commented on MATH-678:
-----------------------------
IMHO, this issue is too vague to be useful.
The unit tests referred to have been removed from the CM test suite (cf. MATH-763) because they were all failing (although such big problems were not revealed by the other unit tests). Nobody seems eager to determine whether there could be a problem in the test methodology (e.g. too stringent tolerances) or in some way clean them up so that the potential problems can be sorted out.
I would thus suggest to resolve this issue as "Won't fix".
> NonLinear Optimizers seem to have a hard time hitting NIST standards
> --------------------------------------------------------------------
>
> Key: MATH-678
> URL: https://issues.apache.org/jira/browse/MATH-678
> Project: Commons Math
> Issue Type: Bug
> Affects Versions: 3.0
> Environment: Java
> Reporter: greg sterijevski
> Assignee: greg sterijevski
> Priority: Minor
> Labels: NIST, NonLinear, Optimization
>
> As per a discussion on the mailing list, I am opening this ticket. In applying the nonlinear optimizers in commons, I noticed what I believe to be instability in the techniques. Further investigation investigation (both of my tests) and the code in prod is warranted.
> I will be pushing a first set of tests which should illustrate what I am seeing.
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira