Re: resampling irregular sets

Hi Ugo,

Its a tall order, but all anyone can do with this sort of
problem is to look into what is happening in Field.resample()
and some appropriate *Set*.valueToInterp() using jdb or print
statements. It may be simpler to use Data.NEAREST-NEIGHBOR
and investigate *Set*.valueToIndex(). If you find any evidence,
please let us help interpret it.

Good luck,
Bill

On Fri, 21 Jan 2005, Ugo Taddei wrote:

> Hello,
>
> this is strange. I've got a small example that runs ok. There were a
> handful of points. In my app, however, I have some 2000 points.
> Triangulation is never a problem, but the thing refuses to resample. It
> returns NaNs.
>
> I've made sure the sets overlap and that the math types are the same.
> I'm not ruling a mistake by myself, but... Any ideas about what might be
> going wrong?
>
> Thanks in advance,
>
> Ugo
>
> Tom Rink wrote:
>
> > Hi Ugo,
> >
> > This operation is implemented, and should work.  There can
> > be problems sometimes with precision and coincident data
> > points.  If your willing to post a small program or the data to
> > the list we can take a look.
> >
> > TomR
> >
> > Ugo Taddei wrote:
> >
> >> Hi,
> >>
> >> there I go again on the subject of irregular sets...
> >>
> >> Is it possible to do something like
> >>
> >> FlatField bigField = ...// large field with Irregular2DSet as domain set
> >>
> >> Set smallSet = ...small irregular 2D set
> >>
> >> FlatField smallField = (FlatField)bigField.resample( smallSet,
> >> Data.WEIGHTED_AVERAGE, Data.NO_ERRORS);
> >>
> >> Before I had a Linear2DSet as my smallSet, and it worked fine (but
> >> "slow"). It is now returning "missing". The fields do (partially)
> >> overlap, however. I expected at least some values to return NaNs.
> >>
> >> I'm not sure if the mistake is somewhere else or if this is not
> >> possible/implemented.
> >>
> >> Thanks for the help,
> >>
> >> Ugo
> >>
> >
> >
>
>
> --
> l a t / l o n  GmbH
> Aennchenstrasse 19           53177 Bonn, Germany
> phone ++49 +228 184960       fax ++49 +228 732153
> http://www.lat-lon.de        http://www.deegree.org
>
>


  • 2005 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the visad archives: