[
https://issues.apache.org/jira/browse/MATH-1137?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14061984#comment-14061984
]
Nigel Goodwin commented on MATH-1137:
-------------------------------------
How do I find out what tests have already been done? I'm a bit worried this bit
of code is never executed in my tests. Has a detailed comparison been done
between the Fortran and Java on a range of test problems? I know machine
precision is a problem, but has there been any attempt to compare the paths the
tests go through? I spent many hours comparing the results at all stages
between Fortran and java versions of L-BFGS-D, so I know how painful the
process is.
I am less concerned about the line we have been discussing here, than I am
about other bugs which we are not discussing and have not yet been found!
You can often have optimisation algorithms which have bugs but still find the
minimum, albeit inefficiently.
> BOBYQA incorrect indexing
> -------------------------
>
> Key: MATH-1137
> URL: https://issues.apache.org/jira/browse/MATH-1137
> Project: Commons Math
> Issue Type: Bug
> Affects Versions: 3.3
> Reporter: Nigel Goodwin
>
--
This message was sent by Atlassian JIRA
(v6.2#6252)
