[
https://issues.apache.org/jira/browse/MATH-1126?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gilles resolved MATH-1126.
--------------------------
Resolution: Cannot Reproduce
It seems that after several cycles of recompiling, I cannot reproduce different
outputs from the two implementations! Maybe I was using "stale" files
somewhere...
The results are almost exactly the same.
The performance degradation remains, although down to about 20% (vs 35% as
initially observed). I still get a different number of evaluations but it's
not caused by the core of the optimization algorithm...
Thus closing this report. Sorry for the noise.
> "LevenbergMarquardtOptimizer": Divergent behavior of new code
> -------------------------------------------------------------
>
> Key: MATH-1126
> URL: https://issues.apache.org/jira/browse/MATH-1126
> Project: Commons Math
> Issue Type: Bug
> Affects Versions: 3.3
> Reporter: Gilles
> Labels: regression
> Attachments: LM_cost_NEW, LM_cost_OLD
>
>
> The new implementation of "LevenbergMarquardtOptimizer" (package
> "o.a.c.m.fitting.leastsquares") behaves differently from the previous one
> (package "o.a.c.m.optim.nonlinear.vector.jacobian").
> This shows up not so much in the solutions respectively found by one and the
> other implementation; there are fairly similar, but in my use-case, the
> number of function evaluations is quite different. And this could explain an
> observed 35% performance degradation.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
