Hastie et al. say that k-fold CV with a low k can overestimate the error if
there is still significant learning going on when the size of the training set
approaches n-n/k.
Leave-one-out CV, on the other hand, is free of this bias but has a high
variance, which is why Hastie et al. dislike it
Dear Alessandro,
it would be easier to give you a constructive answer if you provided at
least the code that you used.
The SAGA module that you mention does not (yet) have a front-end
function in the RSAGA package, so I assume that you are using the
low-level rsaga.geoprocessor function, whi
Hi all and thanks for help,
I have a shapefile "canopy_height.shp" with two column "a" and "b". I wish
to do a-b and to obtain a third column ("a-b"). I am using the R+SAGA
module: "Table calculator for shapes", and all run good, but when I open the
shape-file after (in ARCMAP) I don't see the
Hi Gerard,
following Dylan's hint, the following example should be easily reproducable:
library(maptools)
nc1 <- readShapePoly(system.file("shapes/sids.shp",
package="maptools")[1], proj4string=CRS("+proj=longlat +datum=NAD27"))
x100 = do.call(rbind, sapply(slot(nc1, "polygons"), spsample, n=1,
t
Hi All,
Do you know if R has a function to perform Iteratively Reweighted
Generalized Least Squares IRWGLS?
Thanks.
Faye
[[alternative HTML version deleted]]
___
R-sig-Geo mailing list
R-sig-Geo@stat.math.ethz.ch
https://stat.ethz.ch/mailman/lis
Thanks Edzer,
for some reason I had it in my head that n-fold was a variant of what
you describe; an independent randomly selected set to "check" the fit
if the model. I guess that's where I was heading with CV, sore form of
relative assessment of how "good" the fitted variogram model was/is
dde...@sciborg.uwaterloo.ca wrote:
> Hi list,
> A quick question regarding n-fold validation...
> I've seen several papers suggest the LOOCV is too optimistic. Is
> n-fold closer to a "true" validation?
I don't think "true" validation exists; could you explain what it is? If
you mean having a compl
Hi Gerald:
I have an example that may help: It demonstrates the sampling of a
polygon shape file with a regularly-spaced point grid,
and assignment of polygon values to the sampling points:
http://nceas.ucsb.edu/scicomp/GISSeminar/UseCases/SampleVectorPolygonsRasterGrid/SampleVectorPolysRastGr
On Monday 23 February 2009, Heuvelink, Gerard wrote:
> Dear list,
>
> I use readShapePoly to import a shape file. The file has multiple polygons
> (actually, a 'polygon' may consist of multiple polygons that have the same
> ID). I want to sample one location at random from each of the polygons. I
>
Dear list,
I use readShapePoly to import a shape file. The file has multiple polygons
(actually, a 'polygon' may consist of multiple polygons that have the same ID).
I want to sample one location at random from each of the polygons. I tried to
do this with spsample but did not manage (the opti
Hi list,
A quick question regarding n-fold validation...
I've seen several papers suggest the LOOCV is too optimistic. Is
n-fold closer to a "true" validation?
I am assuming that it uses the variogram that is constructed using ALL
data, so my assumption is that the variogram is not re-fit for
11 matches
Mail list logo
