Dear list members
I'm using the 'raster' package to load a rather large (~2 GB) file of
bathymetry data. To extract elevation/depth data for various positions,
I use the 'xyValues' function.
However, it turns that this operations is extremely slow for extracting
a even moderately large number of values. Here is in essence my code:
r = raster("file.ers")
x = seq(0, 20, length = 100)
y = seq(55, 70, length = 100)
xy = expand.grid(x, y)
z = xyValues(r, xy)
Am I doing something wrong here? Using 'readBin' or 'readBinFragments'
on the same file, to extract the same values, takes almost no time. But
then I have to calculate the indices, positions &c. myself, which is
tedious and easy get wrong (having to take into account half-pixel
shifts and other details).
Using Rprof indicates that a lot of 'self.time' is taken up by 'gc' (and
'=='), which does sound a bit strange.
--
Karl Ove Hufthammer
_______________________________________________
R-sig-Geo mailing list
[email protected]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
