read_plexon() ValueError: When changing to a larger dtype, its size must be a divisor of the total size in bytes of the last axis of the array

[shunnnli]

Hi,

We're loading a .plx file using read_plexon() method. We know the code runs because a sample short (~1min) .plx file (acquired directly in .plx) runs in our code successfully.

However, when we try to load the actually file (~40min recording in .pl2 format, converted to .plx), spikeinterface returns following error (see below).

Thanks in advance!

My code to read plexon file:
raw_rec = si.read_plexon(recording_path,stream_name='Signals 0')

Spikeinterface error:
ValueError Traceback (most recent call last)
Cell In[11], line 2
1 # we do not load the sync channel, so the probe is automatically loaded
----> 2 raw_rec = si.read_plexon(recording_path,stream_name='Signals 0')
3 raw_rec

File ~.conda\envs\spikeinterface\lib\site-packages\spikeinterface\core\core_tools.py:26, in define_function_from_class..reader_func(*args, **kwargs)
24 @copy_signature(source_class)
25 def reader_func(*args, **kwargs):
---> 26 return source_class(*args, **kwargs)

File ~.conda\envs\spikeinterface\lib\site-packages\spikeinterface\extractors\neoextractors\plexon.py:29, in PlexonRecordingExtractor.init(self, file_path, stream_id, stream_name, all_annotations)
27 def init(self, file_path, stream_id=None, stream_name=None, all_annotations=False):
28 neo_kwargs = self.map_to_neo_kwargs(file_path)
---> 29 NeoBaseRecordingExtractor.init(self, stream_id=stream_id,
30 stream_name=stream_name,
31 all_annotations=all_annotations,
32 **neo_kwargs)
33 self._kwargs.update({'file_path': str(file_path)})

File ~.conda\envs\spikeinterface\lib\site-packages\spikeinterface\extractors\neoextractors\neobaseextractor.py:42, in NeoBaseRecordingExtractor.init(self, stream_id, stream_name, block_index, all_annotations, **neo_kwargs)
39 def init(self, stream_id=None, stream_name=None,
40 block_index=None, all_annotations=False, **neo_kwargs):
---> 42 _NeoBaseExtractor.init(self, block_index, **neo_kwargs)
44 kwargs = dict(all_annotations=all_annotations)
45 if block_index is not None:

File ~.conda\envs\spikeinterface\lib\site-packages\spikeinterface\extractors\neoextractors\neobaseextractor.py:24, in _NeoBaseExtractor.init(self, block_index, **neo_kwargs)
23 def init(self, block_index, **neo_kwargs):
---> 24 self.neo_reader = get_reader(self.NeoRawIOClass, **neo_kwargs)
26 if self.neo_reader.block_count() > 1 and block_index is None:
27 raise Exception("This dataset is multi-block. Spikeinterface can load one block at a time. "
28 "Use 'block_index' to select the block to be loaded.")

File ~.conda\envs\spikeinterface\lib\site-packages\spikeinterface\extractors\neoextractors\neobaseextractor.py:14, in get_reader(raw_class, **neo_kwargs)
12 neoIOclass = eval('neo.rawio.' + raw_class)
13 neo_reader = neoIOclass(**neo_kwargs)
---> 14 neo_reader.parse_header()
16 return neo_reader

File ~.conda\envs\spikeinterface\lib\site-packages\neo\rawio\baserawio.py:179, in BaseRawIO.parse_header(self)
166 def parse_header(self):
167 """
168 This must parse the file header to get all stuff for fast use later on.
169
(...)
177
178 """
--> 179 self._parse_header()
180 self._check_stream_signal_channel_characteristics()

File ~.conda\envs\spikeinterface\lib\site-packages\neo\rawio\plexonrawio.py:90, in PlexonRawIO._parse_header(self)
88 pos = offset4
89 while pos < data.size:
---> 90 bl_header = data[pos:pos + 16].view(DataBlockHeader)[0]
91 length = bl_header['NumberOfWaveforms'] * bl_header['NumberOfWordsInWaveform'] * 2 + 16
92 bl_type = int(bl_header['Type'])

ValueError: When changing to a larger dtype, its size must be a divisor of the total size in bytes of the last axis of the array.

[alejoe91]

Thanks for reporting this @shunnnli

Note that we'll have a plexon2 reader very soon which might fix the issue: #1736

If you have time, you can try to checkout that PR and beta test it :)

[alejoe91]

@shunnnli were you able tot try the plexon2 reader?

[zm711]

Just want to bump this one more time. @shunnnli, did the plexon2 reader work for you ?

[zm711]

I'll close this for inactivity. We have worked on Plexon1 and Plexon2 over on the Neo side. A Plexon engineer is helping update some functions at Neo. Feel free to open an issue over there if you run into more issues with plexon files so we can track there.

[ChauhanT]

Hi @zm711 , I am reading .plx files and am unable to read events from it. Using neo I can read stream_ids like EVT01 for events (digital in) and 0, 1, etc for analogue data. spikeinterface seems to only read analogue data. Am I missing something here ?

Thank you!