[
https://issues.apache.org/jira/browse/MATH-1656?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17741898#comment-17741898
]
François Laferrière commented on MATH-1656:
-------------------------------------------
{quote}I understand but my point is that the "core" of a unit test is to
compare values (computed vs reference).
{quote}
I totally agree with you. And that is exactly what DiffTest does, compare
computed test results vs reference. Except that you can compare result vs
reference with you favorite diff tool, so that you, then, have context (results
that where produces before the assert fails, result expected at failure point
and even result that could have been produced after first failure). With plain
junit, you just have a statement like "result is 5, should be 4" and to
understand what is going on, you have to run under debugger. I hated it when I
first use junit some ... 24 years ago. I started then to develop this little
framework, first in bash and makefile. This original "black box" version was
language independent and was use to test any binary (I used it for java, but
also C/C++, perl, bash etc.). Eventually "ported" it to junit/java and other
environments (nunit/c#).
I can provide a (bash) toolbox to compare and manage test references.
But I deeply feel that I won't ever convince you. *So I will rollback to a
plain junit test (without I/O or gnuplot stuff) focusing on results only.*
That may take some time.
> Classical multivariate optimizers (gradient descent, Raphson-Newton, BFGS)
> are missing
> --------------------------------------------------------------------------------------
>
> Key: MATH-1656
> URL: https://issues.apache.org/jira/browse/MATH-1656
> Project: Commons Math
> Issue Type: Wish
> Components: legacy
> Affects Versions: 4.0-beta1
> Reporter: François Laferrière
> Priority: Major
> Labels: features
> Attachments: MATH-1656-GradientDescent-Newton-BFGS-v2.0.zip,
> MATH-1658-GradientDescent-Newton-BFGS.patch, Screenshot from 2023-07-10
> 12-13-38.png
>
>
> Some classical multivariate such as
> * gradient descent,
> * Raphson-Newton,
> * BFGS
> are missing.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
