mattmar / rasterdiv Goto Github PK

I have a RasterLayer that contains NDVI values and I have a sampling grid as a SpatialPolygonsDataFrame. Here are the details of those two objects:

NDVI_ext
class : RasterLayer
dimensions : 2716, 2880, 7822080 (nrow, ncol, ncell)
resolution : 3.125, 3.125 (x, y)
extent : 270084.4, 279084.4, 1764175, 1772663 (xmin, xmax, ymin, ymax)
crs : +proj=utm +zone=16 +datum=WGS84 +units=m +no_defs
source : memory
names : layer
values : 0.4966216, 0.9459583 (min, max)

NDVI_spdf
class : SpatialPolygonsDataFrame
features : 232
extent : 270084.4, 279084.4, 1764175, 1772663 (xmin, xmax, ymin, ymax)
crs : +proj=utm +zone=16 +datum=WGS84 +units=m +no_defs
variables : 1
names : ID,
min values : 1,
max values : 232,

I want to calculate Rao's Q inside each cell of the grid using the field "ID", analogous to the use of the "CONTINENT" field shown in the example vignette (https://mattmar.github.io/rasterdiv/articles/rasterdiv_area_based_Rao.html). But when I do this, I receive an error:

paRao( x=NDVI_ext, area=NDVI_spdf, field='ID', window=9 )
Error in combn(p, m = 2, FUN = prod, na.rm = TRUE) : n < m

Is there a clear solution to this problem? I suspect there is an alternative approach in uses raster::extract to apply the paRao function to each sample unit in the grid, but it's not obvious how to do this either because of the moving window.

Any advice about this issue would be helpful. I am also new to posting to Issues so if this is not the appropriate venue or format, do let me know.

I am able to run Shannon, Renyi, or Roa on an EVI2 raster without np but get the following if I run np = 6 (my workstation has 6 cores):

Object x check OK:
Shannon output matrix will be returned.

################ Starting parallel calculation ##################
Error in registerDoParallel(cls) : could not find function "registerDoParallel"

Thanks

I used the Rao function on the NDVI image with a moving window of 3. It has been more than 5 days since the function running. Will the function generate output or it will keep going on like this. The NDVI image is of Sentinel-2 and covering around 500 km sq area.