[R] akima interp results to zero with less than 10 values

[PIKAL Petr]

Hallo Duncan

Thanks, I was not aware of this package.  I will try.

Petr

> -----Original Message-----
> From: Duncan Murdoch <murdoch.duncan using gmail.com>
> Sent: Thursday, January 26, 2023 3:44 PM
> To: PIKAL Petr <petr.pikal using precheza.cz>; r-help using r-project.org
> Subject: Re: [R] akima interp results to zero with less than 10 values
>
> The akima package has a problematic license (it doesn't allow commercial 
> use),
> so it's been recommended that people use the interp package instead.  When I
> use interp::interp instead of akima::interp, I get reasonable output from 
> your
> example.
>
> So that's another reason to drop akima...
>
> Duncan Murdoch
>
> On 26/01/2023 9:35 a.m., PIKAL Petr wrote:
> > Dear all
> >
> > I have this table
> >> dput(mat)
> > mat <- structure(c(2, 16, 9, 2, 16, 1, 1, 4, 7, 7, 44.52, 42.8, 43.54,
> > 40.26, 40.09), dim = c(5L, 3L))
> >
> > And I want to calculate result for contour or image plots as I did few
> > years ago.
> >
> > However interp does not compute the z values and gives me zeros in z 
> > matrix.
> > library(akima)
> >
> >> interp(mat[,1], mat[,2], mat[, 3], nx=5, ny=5)
> > $x
> > [1]  2.0  5.5  9.0 12.5 16.0
> >
> > $y
> > [1] 1.0 2.5 4.0 5.5 7.0
> >
> > $z
> >       [,1] [,2] [,3] [,4] [,5]
> > [1,]    0    0    0    0    0
> > [2,]    0    0    0    0    0
> > [3,]    0    0    0    0    0
> > [4,]    0    0    0    0    0
> > [5,]    0    0    0    0    0
> >
> > With the example from help page if less than 10 values are used, the
> > result is also zero interp(akima$x[1:9], akima$y[1:9], akima$z[1:9],
> > nx=5, ny=5)
> >
> > but with 10 or more values the result is correctly calculated
> > interp(akima$x[1:10], akima$y[1:10], akima$z[1:10], nx=5, ny=5) $x [1]
> > 0.0000  6.1625 12.3250 18.4875 24.6500
> >
> > $y
> > [1]  1.24  5.93 10.62 15.31 20.00
> >
> > $z
> >       [,1]     [,2]     [,3]     [,4]     [,5]
> > [1,]   NA       NA       NA       NA 34.60000
> > [2,]   NA       NA 27.29139 27.11807 26.60971
> > [3,]   NA 19.81371 19.63614 19.12778 18.61943
> > [4,]   NA 14.01443 10.66531 11.13750 10.62914
> > [5,]   NA       NA       NA       NA       NA
> >
> > Help page says
> > x, y, and z must be the same length (execpt if x is a
> > SpatialPointsDataFrame) and may contain no fewer than ***four*** points.
> >
> > So my understanding was that 5 poins could be used but I am obviously
> wrong.
> > Is it a bug in interp or in the documentation or is it my poor
> > understanding of the whole matter.
> >
> > Best regards
> > Petr
> >
> >
> > ______________________________________________
> > R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide
> > http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.