[
https://issues.apache.org/jira/browse/MATH-1656?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17741732#comment-17741732
]
Gilles Sadowski commented on MATH-1656:
---------------------------------------
bq. [...] it is really part of the tests.
I understand but my point is that the "core" of a unit test is to compare
values (computed vs reference). IIUC, you have devised your own framework to
compare text files, whereas we (ideally) want to rely on Junit 5 for doing that
(see the [CSV
files|https://github.com/apache/commons-math/tree/master/commons-math-legacy/src/test/resources/org/apache/commons/math4/legacy/optim/nonlinear/scalar/noderiv]
used for testing the
[SimplexOptimizer|https://github.com/apache/commons-math/blob/master/commons-math-legacy/src/test/java/org/apache/commons/math4/legacy/optim/nonlinear/scalar/noderiv/SimplexOptimizerTest.java]
family).
We should strive to consolidate the whole codebase (tests included);
diversifying the approaches to performing the same task (in this case, testing
an optimizer) is a maintenance nightmare. As suggested, we should discuss
separately the benefits of alternative approaches, possibly combine them, and
then apply the decision to all the concerned code.
bq. It is not really an example or a mini application, [...]
I was referring to the generation of files for visualizing with gnuplot.
Note that I also created "debugging"/"trajectory tracking" code in
["SimplexOptimizerTest.java"|https://github.com/apache/commons-math/blob/master/commons-math-legacy/src/test/java/org/apache/commons/math4/legacy/optim/nonlinear/scalar/noderiv/SimplexOptimizerTest.java],
together with a [Gnuplot
script|https://github.com/apache/commons-math/blob/master/commons-math-legacy/src/test/gnuplot/legacy/optim/nonlinear/scalar/noderiv/simplex_2D.gnuplot].
So I dropped the ball there even though I think that a dedicated
"mini-application" would be more useful than a Gnuplot script generated as part
of running the unit tests. But admittedly, it's not obvious to decide what
should be under "src/test" and what should be moved under "examples".
> 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)
