gdalcubes (version 0.2.5)

chunk_apply: Apply an R function on chunks of a data cube

Description

Apply an R function on chunks of a data cube

Usage

chunk_apply(cube, f)

Arguments

cube

source data cube

f

R function to apply over all chunks

Value

a proxy data cube object

Details

This function internally creates a gdalcubes stream data cube, which streams data of a chunk to a new R process. For reading data, the function typically calls x <- read_chunk_as_array() which then results in a 4 dimensional (band, time, y, x) array. Similarly write_chunk_from_array(x) will write a result array as a chunk in the resulting data cube. The chunk size of the input cube is important to control how the function will be exposed to the data cube. For example, if you want to apply an R function over complete pixel time series, you must define the chunk size argument in raster_cube to make sure that chunk contain the correct parts of the data.

Examples

# NOT RUN {
# create image collection from example Landsat data only 
# if not already done in other examples
if (!file.exists(file.path(tempdir(), "L8.db"))) {
  L8_files <- list.files(system.file("L8NY18", package = "gdalcubes"),
                         ".TIF", recursive = TRUE, full.names = TRUE)
  create_image_collection(L8_files, "L8_L1TP", file.path(tempdir(), "L8.db")) 
}

L8.col = image_collection(file.path(tempdir(), "L8.db"))
v = cube_view(extent=list(left=388941.2, right=766552.4,
                          bottom=4345299, top=4744931, t0="2018-01", t1="2018-12"),
                          srs="EPSG:32618", nx = 497, ny=526, dt="P1M")
L8.cube = raster_cube(L8.col, v)
L8.cube = select_bands(L8.cube, c("B04", "B05"))
f <- function() {
  x <- read_chunk_as_array()
  out <- reduce_time(x, function(x) {
    cor(x[1,], x[2,], use="na.or.complete", method = "kendall")
  }) 
  write_chunk_from_array(out)
}
L8.cor = chunk_apply(L8.cube, f)
# }