[
https://issues.apache.org/jira/browse/MATH-1137?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14058781#comment-14058781
]
[email protected] commented on MATH-1137:
----------------------------------------------------
I have never done a unit test in my life! I wouldn't know how to do it.
To be honest, the only way I would check (in addition to reading) would
be to run the original Fortran and the Java and compare the two results.
I read the code very carefully, and walk through it, and make some notes
on pencil and paper, and walk through it again.
I have been using the corrected version for some years without any
problems. I am only going back to this issue because I am doing an
overall upgrade to 3.3. I have been using a corrected pre release
version of BOBYQA previously.
I wrote a lot more on the original MATH-621 (or was it 612?).
It may interest you to know I did a conversion of L-BFGS-B from Fortran
to Java, but it is a mess (I think you will understand why it is a
mess....).
> 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)
