[
https://issues.apache.org/jira/browse/MATH-423?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13012422#comment-13012422
]
Gilles commented on MATH-423:
-----------------------------
"@Retry" annotation committed in revision 1086539.
The failure in "CMAESoptimizer" still occurs now and then. To my knowledge,
there hasn't been any change to eliminate this behaviour.
This might be a case where the "retry" functionality is interesting; since the
randomization is an integral part of the algorithm, I'd think that, in the
test, it is better to let it behave as intended rather than use a fixed seed.
[Maybe we should find some way so that the Junit report indicates how many
retries were necessary to pass the test.]
Of course, as Phil had remarked, we should make sure that the intermittent
failure isn't caused by a bug.
> Upgrade all test code to Junit 4
> --------------------------------
>
> Key: MATH-423
> URL: https://issues.apache.org/jira/browse/MATH-423
> Project: Commons Math
> Issue Type: Task
> Reporter: Gilles
> Priority: Minor
> Fix For: 3.0
>
>
> Make all tests use the JUnit 4 annotation-based syntax.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
