Template of all NaNs

[lukebun]

Hello,

I've found a case where one of my templates is all NaNs, and yet 11 spikes were assigned to the cluster. I've tried reanalyzing my data multiple times but I consistently arrive at the same all NaN template. But only one template from this is all NaNs and I have only ever seen it in one file, despite using the same set-up for other files. I am recording with OpenEphys and NHP 1.0 probes. Do you have any ideas why this might be the case or things I can change to avoid this issue?

I realize that another user had run into a similar problem (#261), but this user appears to have run into a problem with catGT, something that is not in my pipeline.

Thank you!