spyking_circus error .

I’m not sure if this error is totally related to the sorter (spyking_circus), or spike_interface. In any case, I encountered this frequently. Sometimes this error would be resolved if I change (increase) the ‘detect_threshold’ from the sorter parameters.

But such tricks did not work in this case. Reason : unlike my previous successful trials which were on relatively short duration recordings (about 3 minutes), this file (which constitues my main data) consists of 30 minutes of recording. Hence I suspect this may be due to the high volume of input data, perhaps making some memory problem ?

ss.run_sorter( sorter_name='spykingcircus' , recording=rd_l_pps , detect_sign=0 , detect_threshold=8 , template_width_ms=2 , filter=False , auto_merge=0.6 , verbose=True , output_folder=r'/home/azare/groups/PrimNeu/Aryo/analysis/sort/srt_l_2')

I had activated the verbose=True while sorting : the following is the error from spyking circus itself :

Smart Search of good isolated spikes for the clustering (1/3)...


  0%|                                            |[00:00<?, ?it/s]

===================================================================================

=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES

=   PID 26456 RUNNING AT nib03.int.lin-magdeburg.de

=   EXIT CODE: 9

=   CLEANING UP REMAINING PROCESSES

=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES

===================================================================================

YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Killed (signal 9)

This typically refers to a problem with your application.

Please see the FAQ page for debugging suggestions


----------------------------  Error  -----------------------------

| Step "clustering" failed for reason Command '['mpiexec', '-np', '72', 'spyking-circus-subtask', 'clustering', '/home/azare/groups/PrimNeu/Aryo/analysis/sort/srt_l_2/recording.npy', '72', '0', 'False', 'False']' returned non-zero exit status 9.!

------------------------------------------------------------------

spykingcircus run time 3847.61s

Note: the initial steps of the sorting runs without any error. I did not copy all of the report (before the error) for the sake of conciseness.

And the subsequent error from python :

Traceback (most recent call last):

  File "/tmp/ipykernel_9621/1917434363.py", line 1, in <module>
    ss.run_sorter( sorter_name='spykingcircus' , recording=rd_l_pps , detect_sign=0 , detect_threshold=8 , template_width_ms=2 ,  filter=False , auto_merge=0.6 , verbose=True , output_folder=r'/home/azare/groups/PrimNeu/Aryo/analysis/sort/srt_l_2')

  File "/home/azare/anaconda3/envs/env_11/lib/python3.9/site-packages/spikeinterface/sorters/runsorter.py", line 142, in run_sorter
    return run_sorter_local(**common_kwargs)

  File "/home/azare/anaconda3/envs/env_11/lib/python3.9/site-packages/spikeinterface/sorters/runsorter.py", line 164, in run_sorter_local
    sorting = SorterClass.get_result_from_folder(output_folder)

  File "/home/azare/anaconda3/envs/env_11/lib/python3.9/site-packages/spikeinterface/sorters/basesorter.py", line 281, in get_result_from_folder
    sorting = cls._get_result_from_folder(output_folder)

  File "/home/azare/anaconda3/envs/env_11/lib/python3.9/site-packages/spikeinterface/sorters/spyking_circus/spyking_circus.py", line 169, in _get_result_from_folder
    sorting = SpykingCircusSortingExtractor(folder_path=Path(output_folder) / 'recording')

  File "/home/azare/anaconda3/envs/env_11/lib/python3.9/site-packages/spikeinterface/extractors/spykingcircusextractors.py", line 67, in __init__
    raise Exception(spykingcircus_folder, " is not a spyking circus folder")

Exception: (PosixPath('/home/azare/groups/PrimNeu/Aryo/analysis/sort/srt_l_2/recording'), ' is not a spyking circus folder')

[Aryo-Zare]

I just found a similar issue posted on the spyking-cirus page :

https://github.com/spyking-circus/spyking-circus/issues/377

The error reported by the sorter is similar. But seemingly the sorting process had already advanced further than in my case. Even though the user was implementing it via docker. The user was also using spike-interface.

Unfortunately that issue is still open.

[Aryo-Zare]

I found the solution to this : explained at #895.