Ryan Ly

What is the issue with linking individual spikes to snippets? I saw a bunch of issues/PRs on UnitSeries and it looked like a reasonable solution was agreed upon https://github.com/NeurodataWithoutBorders/nwb-schema/issues/248

In Python: `np.testing.assert_array_equal` correctly tests whether `nan == nan`. We would have to replace all `self.assertEqual` calls for numeric values in PyNWB with `np.testing.assert_array_equal`. Or we could subclass `unittest.TestCase`, change...

Sounds good. I would also vote for a less-specific name, like 'Spectrum', or 'PowerSpectrum'. We could come up with specific subtypes and their parameterization as requested/needed.

This is related to https://github.com/NeurodataWithoutBorders/pynwb/issues/1889 but specific to matnwb. I think the slowness has to do with the number of groups in the file. @lawrence-mbf do you have any insights?

Are all references resolved on initial read? Can they be deferred?

@lawrence-mbf could you look into this?

Do you know whether the file was written with PyNWB or MatNWB? **And which part of the file says that it is version 2.0.2?** The version in nwb.namespace.yaml, the version...

@oruebel That would help with debugging but won't stop users from creating production files from the dev branch. I don't know about the particular situation regarding the file that @bendichter...

Ah, right, the schema was not cached. The `/nwb_version` attribute comes from `nwb.file.yaml` (it is a fixed attribute in the definition of `NWBFile`). So, the file must have been generated...

@himahuja how did you generate this file?