Sorter does not recognize a probe is added to the recording

[AbhiSwamiUConn]

I've recently started to use spike Interface, and so far it's been working well. However, I am currently running into an issue when sorting my recording. My recording did not come with data about what probe was used. I learned from my professor however that the setup was three shanks each with a single electrode. So I created one using the generate_multi_shank method and set the probe to the recording. When using the get_probe method on the recording, the probe is returned, however when I go to sort, I get an error stating there is no probe attached to the recording. This seems to drastically lengthen the time my code takes to execute. Secondly, when I try to use the detect_peaks method using the method "by_channel" and the gather_mode "npy", I run into a string concatenation error. Am I doing something wrong?

import spikeinterface.full as si
import probeinterface as pi
import numpy as np
import spikeinterface.extractors as se
import spikeinterface.preprocessing as st
import spikeinterface.sorters as ss
import spikeinterface.exporters as sx
import spikeinterface.postprocessing as sp
import spikeinterface.sortingcomponents.peak_detection as ssp
import spikeinterface.sortingcomponents.clustering as ssc
from probeinterface import generate_multi_shank
from probeinterface.plotting import plot_probe

plx_file = "file_path"
output_folder1 ="folder_path"
output_folder2 = "folder_path"
output_folder3 = "folder_path"
output_folder4 = "folder_path"

"""#stream id 0 is spike channels; 109,989,908 data points, 40000hz sampling rate
#stream id 1 is wideband channels; 109,989,948 data points, 40000hz sampling rate
#stream id 2 is field potential channels; 27,497,477 data points, 10000hz sampling rate"""

plx_extractor = se.read_plexon(plx_file, stream_id="1")
print(plx_extractor)
print(plx_extractor.channel_ids)

probe = generate_multi_shank(num_shank=3, num_columns=1, num_contact_per_column=1)
probe.set_shank_ids(np.arange(3))
probe.set_contact_ids(np.arange(3))
probe.set_device_channel_indices(np.arange(3))
pi.write_probeinterface('my_probe_setup.json', probe)
recording_with_probe = plx_extractor.set_probe(probe, group_mode="by_shank", in_place=True)

print(recording_with_probe.get_probe())
plot_probe(recording_with_probe.get_probe(), with_contact_id = True)

recording_f = st.bandpass_filter(recording_with_probe, freq_min=300, freq_max=6000)
recording_cmr = st.common_reference(recording_f, "global")

sorting = ss.run_sorter("spykingcircus2", recording_cmr, output_folder= output_folder1, remove_existing_folder= True)
Sanalyzer = si.waveform_extractor.extract_waveforms(recording=recording_cmr, sorting=sorting, folder = output_folder2, overwrite= True)

templates_amplitudes = si.get_template_amplitudes(Sanalyzer, "pos", "extremum")
spike_locations = sp.compute_spike_locations(Sanalyzer)
spike_amplitudes = sp.compute_spike_amplitudes(Sanalyzer)
traces = recording_cmr.get_traces()
peaks = ssp.detect_peaks(recording_cmr, method="by_channel", gather_mode="npy", folder=output_folder4, names=["by_channel"])
clusters = ssc.find_cluster_from_peaks(recording_cmr, Sanalyzer, "circus")

print(recording_cmr)
print(traces)
print(peaks)
print(clusters)
sx.export_to_phy(Sanalyzer, output_folder=output_folder3, remove_if_exists=True)

Picture: (https://github.com/SpikeInterface/spikeinterface/assets/159394395/eeebaba2-34b5-4f34-b080-8b9627ccf686)
Error 1: C:\Users\User\anaconda3\Lib\site-packages\spikeinterface\core\baserecordingsnippets.py:244: UserWarning: There is no Probe attached to this recording. Creating a dummy one with contact positions
warn("There is no Probe attached to this recording. Creating a dummy one with contact positions")
Error 2: File ~\anaconda3\Lib\site-packages\spikeinterface\core\node_pipeline.py:614, in GatherToNpy.init(self, folder, names, npy_header_size, exist_ok)
612 self.final_shapes = []
613 for name in names:
--> 614 filename = folder / (name + ".npy")
615 f = open(filename, "wb+")
616 f.seek(npy_header_size)

TypeError: unsupported operand type(s) for /: 'str' and 'str'

[h-mayorquin]

Hi, sorry, you probably opened this here and then on issue, sadly, we don't check here as much as we do the issue. But this was fixed satisfactorly:

#2982