Electrode-counting ignores electrodes that are not clustered; similarity index is based on this count.

[ThePerson2]

Hello,

I think there might be a problem with the way electrode channels are indexed in the latest version of KwikGUI (v1.0.9). My suspicion is based on things of this sort:

This is a nicely-clustered spike, plotted with the three most-similar clusters by similarity index. The first similar cluster is noise; the others are spikes which don't have much in common with
the original cluster. I suspect that the code currently indexes each electrode for each cluster independently, ignoring the missing electrodes for that cluster. Comparisons are then based on this index.

Could you please fix this? The similarity index is a convenient feature which I miss.

Thanks,

Jeffrey Boucher