`isi_violations_ratio` incorrect name

AllenNeuralDynamics / NWB_Packaging_Units

Code Ocean capsule to export ephys units data to NWB

MIT License

0 stars 0 forks source link

`isi_violations_ratio` incorrect name #3

Closed bjhardcastle closed 3 months ago

bjhardcastle commented 3 months ago

Since moving to the pipeline sorting with NWB packaging, there are some discrepancies between the values in NWB.units and postprocessed/quality_metrics.csv, and some errors. I'll likely submit individual issues for each as I find them.

quality_metrics.csv contains isi_violations_count and isi_violations_ratio
NWB.units only contains isi_violations_ratio, but the values look like isi_violations_count

alejoe91 commented 3 months ago

Thanks for the report Ben. Add you check individual values in the CSV and nwb file or just visualized the histogram?

bjhardcastle commented 3 months ago

It looks like ratio and count are similar to each other in the csv:

alejoe91 commented 3 months ago

They are super correlated. Can you plot a scatter plot? Ratios should have many more values between 0 and 1, but they can also be greater than 1 for high contamination

bjhardcastle commented 3 months ago

Right ok, and it looks like both are in the NWB after all, it's just the figurl summary that was not showing count.

bjhardcastle commented 3 months ago

So all is fine - thanks for your help!