Reusing timestamps for TimeSeries objects in MatNWB?

[jccava]

Hi. I am using MatNWB. I have several channels of analog data that are all acquired using the same time values. I can easily create multiple TimeSeries objects by setting 'data' to the values acquired and 'timestamps' to the times. However, I noticed that PyNWB has the ability to reuse timestamps among TimeSeries objects:

data = list(range(101, 201, 10))
reuse_ts = TimeSeries(
    name="reusing_timeseries",
    data=data,
    unit="SIunit",
    timestamps=time_series_with_timestamps,
)

This tactic does not seem to work with MatNWB for me, and the validate_timestamps method for the TimeSeries object only appears to check for float64. Am I missing something, or is this 'reusing' functionality not available in MatNWB?

[lawrence-mbf]

this 'reusing' functionality not available in MatNWB

Yes, more advanced features like reusing is not present in MatNWB. I'm not entirely sure what goes on in "reuse" but you will have to just do that manually. I'm assuming there's more to reuse than just reassignment @rly?

[lawrence-mbf]

Please post the specific error message.

[jccava]

Working from home today. I'll post the matlab complaint in the morning.

Sheesh. You people are fast at responding!

James

[jccava]

Here's what happens:

Processing aiChannels
Processing convertInfo
Processing ops
Data from ops.mat not required
Processing rez
Data from rez.mat not required
Processing rez2
Data from rez2.mat not required
Processing sampsToSecsMap
Processing strobedEvents
Error using hdf5lib2
The HDF5 library encountered an error and produced the following stack trace information:

    H5G__traverse_real    component not found

Error in H5O.open (line 27)
output = H5ML.hdf5lib2('H5Oopen',loc_id,relname,lapl_id);

Error in io.writeAttribute (line 8)
oid = H5O.open(fid, path, 'H5P_DEFAULT');

Error in types.core.TimeSeries/export (line 407)
            io.writeAttribute(fid, [fullpath '/timestamps/interval'], obj.timestamps_interval);

Error in types.untyped.Set/export (line 181)
                    refs = v.export(fid, propfp, refs);

Error in types.core.NWBFile/export (line 843)
        refs = obj.acquisition.export(fid, [fullpath '/acquisition'], refs);

Error in NwbFile/export (line 62)
                refs = export@types.core.NWBFile(obj, output_file_id, '/', {});

Error in nwbExport (line 36)
    nwb(i).export(filename);

Error in processKilosortForNWB (line 92)
nwbExport(nwbOutputFile, nwbName);
 

[lawrence-mbf]

Yeah, it's trying to write with the assumption that timeseries exists as a dataset. That is a bug.

[jccava]

Do I get a lollipop??? Seriously, the work-around isn't all that bad. Not work stoppage here. But thanks for your attention to this.