You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Issue: The find_run_neighbours() method uses ICRS coordinate system when selecting the run neighbors, which may lead to runs with very different zenith distances to be stacked when evaluating the background.
Suggestion: use Alt/Az instead of ICRS in find_run_neighbours():
introduce az_bins and alt_bins keywords to the configuration file with the following behavior: if integers, they will define the Az/sin(Alt) grid to be used when searching for neighbors; if lists, they will define the Alt and Az grid bin edges;
I would suggest to follow the LST behaviour here, which is to match in sin(delta) and cos(zenith) where delta is the angle to the magnetic field. See for example cta-observatory/lst-sim-config#2
The only problem with that is that it then becomes nontrivial to define a distance threshold. We usually only select the n nearest nodes, but with no notion of an absolute distance...
One could of course create a grid and group runs that end up in the same cell
Issue: The
find_run_neighbours()
method uses ICRS coordinate system when selecting the run neighbors, which may lead to runs with very different zenith distances to be stacked when evaluating the background.Suggestion: use Alt/Az instead of ICRS in
find_run_neighbours()
:az_bins
andalt_bins
keywords to the configuration file with the following behavior: if integers, they will define the Az/sin(Alt) grid to be used when searching for neighbors; if lists, they will define the Alt and Az grid bin edges;BkgMakerBase.from_config_file()
;pointing_delta
argument offind_run_neighbours()
toneighboring _grid
or similar and provide the grid from step 2 here.The text was updated successfully, but these errors were encountered: