G&P example fails after turning collisions on (IndexError: index 199 is out of bounds for axis 0 with size 199) about pysdm HOT 6 CLOSED

Hello,

Yes, from what I understand the coalescence process results in removing super-droplets. It was not considered in this example because it concerned only condensation.

The code works when I change the _save function in simulation.py. Instead of

          for drop_id in range(self.particulator.n_sd): 
               attr[drop_id].append(attr_data[drop_id])

I wrote

            for drop_id in range(self.particulator.n_sd):
                if drop_id < len(attr_data):
                    attr[drop_id].append(attr_data[drop_id])
                else:
                    attr[drop_id].append(np.nan)

Then the length of attributes remains the same, but attributes of the "missing" super-droplets are np.nan. Maybe there is a more elegant way to do it. I'm not sure if it is possible to have a unique drop_id for each super-droplet, perhaps @slayoo can answer this.

from pysdm.

Thank you @AgnieszkaMakulska !
Upon coalescence or breakup, in a way the super-droplets inherently loose their "identity", and we do not have (yet) any drop-labeling attribute.

Sorting by one of the attributes before plotting could be a way to identify where, in a given attribute dimension, super-droplet are removed from the system?

from pysdm.

@huangynj, please confirm if that helps in your case, so we can add it to the main branch (and add a test with your sample code). Thanks

from pysdm.

@AgnieszkaMakulska @slayoo yes, filling in with np.nan can temporarily solve the issue. Thanks!

from pysdm.

Stale issue message

from pysdm.

Let me close this one as addressed. Thanks @huangynj for reporting, thanks @AgnieszkaMakulska for help!

from pysdm.