Full_Name: Jonathan Lees
Version: 2.0.1
OS: linux-gnu
Submission from: (NULL) (152.2.75.65)
there is a problem with calculating the convex hull in 2-D interpolation using
the codes interp fromt eh akima package:
x =c(0.6505304, -1.1821562, -0.2600792, 0.7913716)
y = c(1.0424226, 0.1754048, -1.4523334, 0.2349112)
z = c(0.000, 3.042, 0.370, 0.122)
EX = seq(from=min(x), to=max(x), length=100)
WHY = seq(from=min(y), to=max(y), length=100)
ZZ = interp(x=x, y=y, z=z, xo=EX, yo=WHY, extrap=FALSE)
plot(x,y, type='n')
image(ZZ, add=TRUE)
points(x,y)
text(x,y,labels=z)
BUG:
Notice the convex hull is wrong.
It includes only 3 of the 4 points provided.
If you add another point near by it does some
really wierd stuff.
x = c(x, -0.3950865)
y = c(y, -1.449117)
z = c(z, .37)
I do not think it should behave in this manner.
Please do read the posting guide and FAQ, as we do ask you to do before sending a bug report. 1) Your version of R is not current. 2) akima is a contributed package and you are not the maintainer. We do ask you not to `behave in this manner'. On Mon, 3 Oct 2005 jonathan_lees at unc.edu wrote:> Full_Name: Jonathan Lees > Version: 2.0.1 > OS: linux-gnu > Submission from: (NULL) (152.2.75.65) > > > there is a problem with calculating the convex hull in 2-D interpolation using > the codes interp fromt eh akima package: > > x =c(0.6505304, -1.1821562, -0.2600792, 0.7913716) > y = c(1.0424226, 0.1754048, -1.4523334, 0.2349112) > z = c(0.000, 3.042, 0.370, 0.122) > > > EX = seq(from=min(x), to=max(x), length=100) > WHY = seq(from=min(y), to=max(y), length=100) > > ZZ = interp(x=x, y=y, z=z, xo=EX, yo=WHY, extrap=FALSE) > > plot(x,y, type='n') > image(ZZ, add=TRUE) > points(x,y) > text(x,y,labels=z) > > BUG: > Notice the convex hull is wrong. > It includes only 3 of the 4 points provided. > If you add another point near by it does some > really wierd stuff. > x = c(x, -0.3950865) > y = c(y, -1.449117) > z = c(z, .37) > > > > I do not think it should behave in this manner.-- Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595
This message should be sent to the package maintainer, not to this list. Moreover, it works fine with my installation (i.e., I don't get the bug you describe): R: platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status beta major 2 minor 2.0 year 2005 month 09 day 24 svn rev 35666 language R akima: Package: akima Version: 0.4-5 Date: 2005-07-21 Best, Renaud 2005/10/3, jonathan_lees at unc.edu <jonathan_lees at unc.edu>:> Full_Name: Jonathan Lees > Version: 2.0.1 > OS: linux-gnu > Submission from: (NULL) (152.2.75.65) > > > there is a problem with calculating the convex hull in 2-D interpolation using > the codes interp fromt eh akima package: > > x =c(0.6505304, -1.1821562, -0.2600792, 0.7913716) > y = c(1.0424226, 0.1754048, -1.4523334, 0.2349112) > z = c(0.000, 3.042, 0.370, 0.122) > > > EX = seq(from=min(x), to=max(x), length=100) > WHY = seq(from=min(y), to=max(y), length=100) > > ZZ = interp(x=x, y=y, z=z, xo=EX, yo=WHY, extrap=FALSE) > > plot(x,y, type='n') > image(ZZ, add=TRUE) > points(x,y) > text(x,y,labels=z) > > BUG: > Notice the convex hull is wrong. > It includes only 3 of the 4 points provided. > If you add another point near by it does some > really wierd stuff. > x = c(x, -0.3950865) > y = c(y, -1.449117) > z = c(z, .37) > > > > I do not think it should behave in this manner. > > ______________________________________________ > R-devel at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel >-- Renaud LANCELOT Directeur Adjoint charg? des Affaires Scientifiques Deputy Director for Scientific Affairs D?partement EMVT du CIRAD, TA 30/B Campus International de Baillarguet 34398 Montpellier Cedex 5 - France Tel. +33 (0)4 67 59 37 17 Secr. +33 (0)4 67 59 39 04 Fax +33 (0)4 67 59 37 95