[
https://issues.apache.org/jira/browse/MATH-1137?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14058852#comment-14058852
]
Nigel Goodwin commented on MATH-1137:
-------------------------------------
I know, I was being provocative, my bible of testing is Myers Art of Software
Testing. The purpose of testing is to find errors, not to pass tests.
The problem is I don't know what the expected behaviour is. I also struggle to
see where this bit of code fits into the algorithm and where it is described in
Powell's paper.
If I knew this was to do with matrix diagonal elements, and if I knew that all
diagonal elements are expected to be positive, and if I knew how to get the
code to go through this path, then maybe there is a chance of a test.
Do you have any idea where this code is described in the BOBYQA paper? Or is it
part of NEWUOA, or is it some line search function?
I can verify I have run BOBYQA hundreds of times on lots of problems, but have
no idea whether this bit of code has ever been exercised.
> 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)
