[
https://issues.apache.org/jira/browse/MATH-513?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12995631#comment-12995631
]
Ole Ersoy commented on MATH-513:
--------------------------------
Gilles,
I guess I jumped the gun a little :). I already changed the tests as well and
updated the GaussianFitter per the notes. I'll attach the patches to 512.
Here are some test notes:
The biggest differences in the for the change to the GaussianFitter (Now 3
parameters) can be seen in fit04 and fit05 (Poor data fits).
The last test in the ParametricGaussianFunction test fails. It was expecting a
ZeroException, but now gets a NotStrictlyPositiveException...
> Moving the "ParametricRealFunction" interface
> ---------------------------------------------
>
> Key: MATH-513
> URL: https://issues.apache.org/jira/browse/MATH-513
> Project: Commons Math
> Issue Type: Improvement
> Reporter: Gilles
> Assignee: Gilles
> Priority: Minor
> Labels: api-change
> Fix For: 3.0
>
> Attachments: CurveFitter.patch, Gaussian.patch, HarmonicFitter.patch,
> ParametricGaussianFunction.patch, ParametricRealFunction.patch,
> PolynomialFitter.patch
>
>
> I'm suggesting that the "ParametricRealFunction" interface be moved from
> package "optimization.fitting" to the package "analysis" where other similar
> function interfaces (e.g. "UnivariateRealFunction" are defined). On the way
> it should also be renamed to "ParametricUnivariateRealFunction" (so as to
> allow the definition of the multivariate counterpart).
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
