Random sample
Take a random sample from the cell values of a Raster* object (without replacement).
## S4 method for signature 'Raster' sampleRandom(x, size, na.rm=TRUE, ext=NULL, cells=FALSE, rowcol=FALSE, xy=FALSE, sp=FALSE, asRaster=FALSE, ...)
x
: Raster* objectsize
: positive integer giving the number of items to choosena.rm
: logical. If TRUE
(the default), NA
values are removed from random sampleext
: Extent object. To limit regular sampling to the area within the extentcells
: logical. If TRUE
, sampled cell numbers are also returnedrowcol
: logical. If TRUE
, sampled row and column numbers are also returnedxy
: logical. If TRUE
, coordinates of sampled cells are also returnedsp
: logical. If TRUE
, a SpatialPointsDataFrame is returnedasRaster
: logical. If TRUE
, a Raster* object is returned with random cells with values, all other cells with NA
...
: Additional arguments as in writeRaster
. Only relevant when asRaster=TRUE
With argument na.rm=TRUE
, the returned sample may be smaller than requested
A vector, matrix (if cells=TRUE
or x
is a multi-layered object), or a SpatialPointsDataFrame (if sp=TRUE
)
sampleRegular, sampleStratified
r <- raster(system.file("external/test.grd", package="raster")) sampleRandom(r, size=10) s <- stack(r, r) sampleRandom(s, size=5, cells=TRUE, sp=TRUE)