I would like to apply for write access to the GRASS-Addons-SVN repository.
I was looking for a faster v.rast.stats implementation to calculate
univariate statistics based on vector polygons, because I need to do this for
large raster maps and the current v.rast.stats is just too slow.
In the GRASS-Addons-SVN repository I found r.univar.zonal [1] from Markus
Metz, so I adapted the script v.rast.stats to use r.univar.zonal. The result
v.rast.stats2 is a lot faster. See test with spearfish below:
# current v.rast.stats
time v.rast.stats -c -e vector=fields raster=elevation.10m col=elev
real 2m6.139s
user 2m1.616s
sys 0m4.464s
# v.rast.stats2 based on r.univar.zonal
time v.rast.stats_neu -c -e vector=fields raster=elevation.10m col=elev
real 0m3.249s
user 0m2.784s
sys 0m0.444s
I would be happy, if the script v.rast.stats2 and r.univar.zonal from Markus
Metz would finally, after an incubation phase, make it into the sources for
GRASS future versions.
On Mon, Jan 11, 2010 at 10:10 AM, Otto Dassau <otto.dassau@gmx.de> wrote:
Dear GRASS developers,
I would like to apply for write access to the GRASS-Addons-SVN repository.
I was looking for a faster v.rast.stats implementation to calculate
univariate statistics based on vector polygons, because I need to do this for
large raster maps and the current v.rast.stats is just too slow.
In the GRASS-Addons-SVN repository I found r.univar.zonal [1] from Markus
Metz, so I adapted the script v.rast.stats to use r.univar.zonal. The result
v.rast.stats2 is a lot faster. See test with spearfish below:
# current v.rast.stats
time v.rast.stats -c -e vector=fields raster=elevation.10m col=elev
real 2m6.139s
user 2m1.616s
sys 0m4.464s
# v.rast.stats2 based on r.univar.zonal
time v.rast.stats_neu -c -e vector=fields raster=elevation.10m col=elev
real 0m3.249s
user 0m2.784s
sys 0m0.444s
I would be happy, if the script v.rast.stats2 and r.univar.zonal from Markus
Metz would finally, after an incubation phase, make it into the sources for
GRASS future versions.
Am Tue, 12 Jan 2010 00:26:55 +0100
schrieb Markus Neteler <neteler@osgeo.org>:
(cc grass-psc)
On Mon, Jan 11, 2010 at 10:10 AM, Otto Dassau <otto.dassau@gmx.de>
wrote:
> Dear GRASS developers,
>
> I would like to apply for write access to the GRASS-Addons-SVN
> repository.
>
> I was looking for a faster v.rast.stats implementation to calculate
> univariate statistics based on vector polygons, because I need to
> do this for large raster maps and the current v.rast.stats is just
> too slow.
>
> In the GRASS-Addons-SVN repository I found r.univar.zonal [1] from
> Markus Metz, so I adapted the script v.rast.stats to use
> r.univar.zonal. The result v.rast.stats2 is a lot faster. See test
> with spearfish below:
>
> # current v.rast.stats
> time v.rast.stats -c -e vector=fields raster=elevation.10m col=elev
> real 2m6.139s
> user 2m1.616s
> sys 0m4.464s
>
> # v.rast.stats2 based on r.univar.zonal
> time v.rast.stats_neu -c -e vector=fields raster=elevation.10m
> col=elev real 0m3.249s
> user 0m2.784s
> sys 0m0.444s
>
> [1] http://grass.osgeo.org/wiki/GRASS_AddOns#r.univar.zonal
>
> I would be happy, if the script v.rast.stats2 and r.univar.zonal
> from Markus Metz would finally, after an incubation phase, make it
> into the sources for GRASS future versions.
>
> thanks a lot
> Otto
On Tue, Jan 12, 2010 at 8:19 AM, Otto Dassau <otto.dassau@gmx.de> wrote:
Am Tue, 12 Jan 2010 00:26:55 +0100
schrieb Markus Neteler <neteler@osgeo.org>:
(cc grass-psc)
On Mon, Jan 11, 2010 at 10:10 AM, Otto Dassau <otto.dassau@gmx.de>
wrote:
> Dear GRASS developers,
>
> I would like to apply for write access to the GRASS-Addons-SVN
> repository.
...
> I would be happy, if the script v.rast.stats2 and r.univar.zonal
> from Markus Metz would finally, after an incubation phase, make it
> into the sources for GRASS future versions.
>
> thanks a lot
> Otto