Closed weiglszonja closed 4 months ago
No, you're right. Sorry it was not intentional.
ah no worries really, just wasn't sure. #16 should fix this.
I ran into this as well, thanks for the fix @weiglszonja!
Looks like CommandedVoltageSeries has a similar issue tho:
FiberPhotometryTable/commanded_voltage_series (acquisition/fiber_photometry_table/commanded_voltage_series): incorrect shape - expected '[None, None]', got '(4,)'
I just run
nwbinspector
on the converted sessions, and noticed that I got an incorrect shape error forFiberPhotometryResponseSeries
:In the spec file we have restricted that the data shape should always be 2D: https://github.com/catalystneuro/ndx-fiber-photometry/blob/2e80a7ac040ddca2873cce03aceaabd7c816a5f6/src/spec/create_extension_spec.py#L364
Shouldn't we allow it to be 1D as well? What do you think @pauladkisson @luiztauffer @alessandratrapani ? If this was intentional then we at least have to modify the description, because currently it says 1D is allowed.