[
https://issues.apache.org/jira/browse/MATH-366?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12859762#action_12859762
]
Gilles commented on MATH-366:
-----------------------------
> I fear adding Trivariate versions is a path to QuadriVariate and so on [...]
No; I promise I'll stop there ;-)
The rationale is the same as for the 2D case: the data must be specified on a
grid. The existing "MultivariateInterpolator" interface does not enforce it.
Apart from that, the "Uni/Bi/Trivariate..." versions are probably more
intuitive to use. The more so for people dealing with surfaces and volumes in
real space. It is very common to provide data structures and algorithms
specialized for each of the low dimensional cases (see, for example,
[Vector3D|http://commons.apache.org/math/apidocs/org/apache/commons/math/geometry/Vector3D.html]).
The last point is "political": there was a request to add this feature in the
_GAIA_ project's toolbox, and whenever possible, I want to emphasize the
responsiveness of the Commons-Math project.
> TricubicSplineInterpolator
> --------------------------
>
> Key: MATH-366
> URL: https://issues.apache.org/jira/browse/MATH-366
> Project: Commons Math
> Issue Type: New Feature
> Reporter: Gilles
> Priority: Minor
>
> Similarly to the "BicubicSplineInterpolator", I propose to add a
> "TricubicSplineInterpolator" class.
> This will also require the following:
> # Add methods in "BicubicSplineInterpolatingFunction" (to compute the partial
> derivatives)
> # Add a "TrivariateRealGridInterpolator" interface
> # Add a "TrivariateRealFunction" interface
> # Add a "TricubicSplineInterpolatingFunction" class (to compute the spline
> coefficients from the sampled function, and its partial derivatives)
> Is it OK to perform the changes?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
