Given the time it takes to write NWB recording extractors, we would greatly benefit from a new, high efficiency recording extractor that could also hold the sorting data and still allow us to be fully compatible with spikeInterface, ideally without running
recording = st.CacheRecordingExtractor(recording) first.
For reference:
Elapsed time for writing filtered NWB recording extractor to /stelmo/nwb/spikesorting/despereaux20191125000003.nwb/despereaux20191125.nwb_02_r1_11_mountainsort4_franklab_tetrode_hippocampus_30KHz.nwb: 1325.946664571762 sec
Given the time it takes to write NWB recording extractors, we would greatly benefit from a new, high efficiency recording extractor that could also hold the sorting data and still allow us to be fully compatible with spikeInterface, ideally without running recording = st.CacheRecordingExtractor(recording) first.
For reference: Elapsed time for writing filtered NWB recording extractor to /stelmo/nwb/spikesorting/despereaux20191125000003.nwb/despereaux20191125.nwb_02_r1_11_mountainsort4_franklab_tetrode_hippocampus_30KHz.nwb: 1325.946664571762 sec