Description Usage Arguments Details Author(s) See Also Examples
Fits a spatial prediction model via the fit.gstatModel
function (in the case of numeric variable), or some classification algorithm (factor-type variable), and generates predictions. By defaults uses machine learning method (random forest) as implemented in the ranger package.
1 2 3 | ## S4 method for signature 'SpatialPointsDataFrame,SpatialPixelsDataFrame'
autopredict(target, covariates,
auto.plot=TRUE, spc=TRUE, buffer.dist=TRUE, ...)
|
target |
object of class |
covariates |
object of class |
auto.plot |
logical; specifies whether to immediately plot the data via the plotKML function |
spc |
logical; specifies whether to generate spatial predictive components |
buffer.dist |
logical; specifies whether to use buffer distances as covariates |
... |
other optional arguments that can be passed to |
For factor-type variables, classes with <5 observations are automatically removed from analysis. Generation of principal components via the spc
function and buffer distances can be time-consuming for large rasters.
Tomislav Hengl
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 | library(sp)
library(gstat)
library(ranger)
library(plotKML)
library(raster)
## Ebergotzen data set:
data(eberg)
## subset to 8%
eberg <- eberg[runif(nrow(eberg))<.08,]
coordinates(eberg) <- ~X+Y
proj4string(eberg) <- CRS("+init=epsg:31467")
data(eberg_grid)
gridded(eberg_grid) <- ~x+y
proj4string(eberg_grid) <- CRS("+init=epsg:31467")
## predict sand content:
SNDMHT_A <- autopredict(eberg["SNDMHT_A"], eberg_grid,
auto.plot=FALSE, rvgm=NULL)
plot(raster(SNDMHT_A$predicted["SNDMHT_A"]), col=SAGA_pal[[1]])
## predict soil types:
soiltype <- autopredict(eberg["soiltype"], eberg_grid,
auto.plot=FALSE)
## Not run:
spplot(soiltype$predicted, col.regions=R_pal[[2]])
## most probable class:
eberg_grid$soiltype <- as.factor(apply(soiltype$predicted@data, 1, which.max))
levels(eberg_grid$soiltype) = names(soiltype$predicted@data)
spplot(eberg_grid["soiltype"])
## Meuse data set:
demo(meuse, echo=FALSE)
zinc <- autopredict(meuse["zinc"], meuse.grid[c("dist","ffreq")],
auto.plot=FALSE, rvgm=NULL)
spplot(zinc$predicted["zinc"])
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.