You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm using ml_ms4alg in the context of SpikeTemplate. I have a simple 3 channel dataset which I'm sorting units on. When I save firings.mda or curate the results in phy through spikesorters.run_mountainsort4(), all units are assigned channel 0. The units however are on distinct channels (see attached image).
My three electrodes are 2 mm apart, so they are effectively independent. This occurs with adjacency_radius set to 0 or -1.
I was debugging and found that channel information is preserved within the firings array in ms4alg.MountainSort4(), but mountainsort4.py converts the sorting output to a spikeextractors.NumpySortingExtractor() object, which doesn't save channel information. Is this expected behavior within the SpikeTemplate ecosystem, or is there a way to pass channel information along? It would be helpful to have this information for curation (e.g. combining multiple channel clusters to a single unit in phy).
Would it be possible to potentially save MS4 results as a MdaSortingExtractor object, to preserve channel information? Please let me know if I'm asking too much.
Thank you.
The text was updated successfully, but these errors were encountered:
Hello -
I'm using
ml_ms4alg
in the context of SpikeTemplate. I have a simple 3 channel dataset which I'm sorting units on. When I savefirings.mda
or curate the results inphy
throughspikesorters.run_mountainsort4()
, all units are assigned channel 0. The units however are on distinct channels (see attached image).My three electrodes are 2 mm apart, so they are effectively independent. This occurs with
adjacency_radius
set to0
or-1
.I was debugging and found that channel information is preserved within the
firings
array inms4alg.MountainSort4()
, butmountainsort4.py
converts the sorting output to aspikeextractors.NumpySortingExtractor()
object, which doesn't save channel information. Is this expected behavior within the SpikeTemplate ecosystem, or is there a way to pass channel information along? It would be helpful to have this information for curation (e.g. combining multiple channel clusters to a single unit inphy
).Would it be possible to potentially save MS4 results as a
MdaSortingExtractor
object, to preserve channel information? Please let me know if I'm asking too much.Thank you.
The text was updated successfully, but these errors were encountered: