Distance image

Share:

Description

If you've done a model selection with selectmodel, this function can draw you a map, in which the distances of the original label vector and the estimated label vectors are shown. This is done by a filled.contour plot.

Usage

1
2
3
4
5
distimage(model,
          color.palette = terrain.colors,
	  log = TRUE,
	  plottitle = "Distance of Ys",
	  ...)

Arguments

model

list of model selection data as it has been returned by selectmodel. selectmodel must have been called with ydist parameter set to TRUE!

color.palette

color palette function to use, see rainbow

log

leave this TRUE, if the axis of the model parameter should be logarithmically scaled. Set this to FALSE if you want linear scaling.

plottitle

title of the plot

...

additional parameters for filled.contour

Author(s)

Jan Saputra Mueller

References

M. L. Braun, J. M. Buhmann, K. R. Mueller (2008) \_On Relevant Dimensions in Kernel Feature Spaces\_

See Also

selectmodel, modelimage, drawkpc, filled.contour, rainbow

Examples

1
2
3
4
5
6
## model selection with RBF-kernel and graphical illustration
## of the distances of the labels
d <- sincdata(100, 0.1) # generate sinc data
# do model selection
m <- selectmodel(d$X, d$y, ydist = TRUE, sigma = logspace(-3, 3, 100))
distimage(m) # distance image