Update to SI new API

examples/mountainsort4_example1.py

@@ -1,10 +1,10 @@
 #!/usr/bin/env python3
 
 import mountainsort4 as ms4
-import spikeextractors as se
+import spikeinterface.extractors as se
 
 def main():
-    recording, sorting_true = se.example_datasets.toy_example()
+    recording, sorting_true = se.toy_example(num_segments=1)
     sorting = ms4.mountainsort4(
         recording=recording,
         detect_sign=-1,

jinjaroot.yaml

@@ -1,5 +1,5 @@
 projectName: mountainsort4
-projectVersion: 1.0.0
+projectVersion: 1.1.0
 projectAuthor: Jeremy Magland
 projectAuthorEmail: [email protected]
 projectDescription: Spike sorting using MountainSort4 algorithm

mountainsort4/__init__.py

@@ -1 +1,2 @@
-from .mountainsort4 import mountainsort4
+from .mountainsort4 import mountainsort4
+from .version import __version__

mountainsort4/mountainsort4.py

@@ -6,11 +6,11 @@
 import numpy as np
 import math
 import multiprocessing
-import spikeextractors as se
+import spikeinterface as si
 
 
-def mountainsort4(*, recording: se.RecordingExtractor, detect_sign: int, clip_size: int=50, adjacency_radius: float=-1, detect_threshold: float=3, detect_interval: int=10,
-                  num_workers: Union[None, int]=None, verbose: bool=True) -> se.SortingExtractor:
+def mountainsort4(*, recording: si.BaseRecording, detect_sign: int, clip_size: int=50, adjacency_radius: float=-1, detect_threshold: float=3, detect_interval: int=10,
+                  num_workers: Union[None, int]=None, verbose: bool=True) -> si.BaseSorting:
     if num_workers is None:
         num_workers = math.floor((multiprocessing.cpu_count()+1)/2)
 
@@ -45,18 +45,14 @@ def mountainsort4(*, recording: se.RecordingExtractor, detect_sign: int, clip_si
         print('Cleaning tmpdir::::: '+tmpdir)
     shutil.rmtree(tmpdir)
     times, labels, channels = MS4.eventTimesLabelsChannels()
-    output = se.NumpySortingExtractor()
-    output.set_times_labels(times=times, labels=labels)
+    output = si.NumpySorting.from_times_labels(times_list=times, labels_list=labels,
+                                               sampling_frequency=recording.get_sampling_frequency())
     return output
 
 
-def _get_geom_from_recording(recording: se.RecordingExtractor):
-    channel_ids = cast(np.ndarray, recording.get_channel_ids())
-    M = len(channel_ids)
-    location0 = recording.get_channel_property(channel_ids[0], 'location')
-    nd = len(location0)
-    geom = np.zeros((M, nd))
-    for i in range(M):
-        location_i = recording.get_channel_property(channel_ids[i], 'location')
-        geom[i, :] = location_i
+def _get_geom_from_recording(recording: si.BaseRecording):
+    if 'location' in recording.get_property_keys():
+        geom = recording.get_channel_locations()
+    else:
+        raise AttributeError("mountainsort4 needs locations to be added to the recording object")
     return geom

mountainsort4/ms4alg.py

@@ -627,7 +627,7 @@ def numChannels(self):
         return len(self._recording.get_channel_ids())
 
     def numTimepoints(self):
-        return self._recording.get_num_frames()
+        return self._recording.get_num_samples()
 
     def getChunk(self, *, t1, t2, channels):
         channel_ids = self._recording.get_channel_ids()
@@ -642,14 +642,14 @@ def getChunk(self, *, t1, t2, channels):
                                                       t2=t2a, channels=channels)
             return ret
         else:
-            return self._recording.get_traces(start_frame=t1, end_frame=t2, channel_ids=channels2)
+            return self._recording.get_traces(start_frame=t1, end_frame=t2, channel_ids=channels2).T
 
 
 def prepare_timeseries_hdf5_from_recording(recording, timeseries_hdf5_fname, *, chunk_size, padding):
     chunk_size_with_padding = chunk_size+2*padding
     with h5py.File(timeseries_hdf5_fname, "w") as f:
         M = len(recording.get_channel_ids())  # Number of channels
-        N = recording.get_num_frames()  # Number of timepoints
+        N = recording.get_num_samples()  # Number of timepoints
         num_chunks = math.ceil(N/chunk_size)
         f.create_dataset('chunk_size', data=[chunk_size])
         f.create_dataset('num_chunks', data=[num_chunks])
@@ -672,7 +672,7 @@ def prepare_timeseries_hdf5_from_recording(recording, timeseries_hdf5_fname, *,
             aa = padding-(t1-s1)
             # Read the padded chunk
             padded_chunk[:, aa:aa+s2 -
-                         s1] = recording.get_traces(start_frame=s1, end_frame=s2)
+                         s1] = recording.get_traces(start_frame=s1, end_frame=s2).T
 
             for m in range(M):
                 f.create_dataset('part-{}-{}'.format(m, j),

setup.py

@@ -10,6 +10,6 @@
         'numpy',
         'h5py',
         'sklearn',
-        'spikeextractors>=0.9.5'
+        'spikeinterface>=0.90'
     ]
 )