If that case, routeSampler.py should find those short routes with and without use of --optimize (the GEH mismatch comes only from rounding of fractional route counts as explained above). Also, using GEH as indicator should be OK. If there are only a few candidate routes for every counting location then there is also little ambiguity to be resolved by the algorithm. However, if your network is very small then you should make sure that your routes start and end at the fringe of the network rather than within it (in small networks most traffic is through-traffic).
regards, Jakob > I minimized my network, so it is necessary that there are also routes that > do only cross one junction and then disappear. > Does this mean that I can't benchmark my simulation with the GEH indicator > or can I use GEH with other detectors than routeSampler uses? Or should I > use a bigger network to use GEH? > > > > > > > -- > Sent from: http://sumo-user-mailing-list.90755.n8.nabble.com/ > _______________________________________________ > sumo-user mailing list > [email protected] > To unsubscribe from this list, visit > https://www.eclipse.org/mailman/listinfo/sumo-user >
_______________________________________________ sumo-user mailing list [email protected] To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
