Skip to main content

API Reference

This page provides a detailed reference for the main classes and functions in RT-Sort.

ModelSpikeSorter​

ModelSpikeSorter.load(detection_model_path)​

Loads a pre-trained detection model.

  • Parameters:
    • detection_model_path: Path to a folder containing a ModelSpikeSorter object's files.
  • Returns: ModelSpikeSorter object

Sequence Detection​

detect_sequences(recording, inter_path, detection_model, **kwargs)​

Detects sequences in a recording.

  • Parameters:
    • recording: Recording loaded with SpikeInterface or path to a saved recording.
    • inter_path: Path to a folder for intermediate cached data.
    • detection_model: ModelSpikeSorter object or path to its folder.
    • **kwargs: Additional optional parameters.
      • recording_window_ms: Tuple (start_ms, end_ms) for section of recording to process. Default is None (entire duration).
      • delete_inter: Whether to delete the intermediate directory. Default is False.
      • verbose: Whether to print progress. Default is True.
      • num_processes: Number of CPU processes to use. Default is None (uses all available).
  • Returns: RTSort object or NumPy array of spike trains.

RTSort​

reset()​

Resets the RTSort object for online use.

  • Parameters: None
  • Returns: None

running_sort(obs)​

Sorts a stream of data in real-time.

  • Parameters:
    • obs: NumPy array with shape (num_frames, num_electrodes).
  • Returns: List of tuples containing sorted spike data.

Training​

train_detection_model(recordings, **kwargs)​

Trains a new detection model.

  • Parameters:
    • recordings: List of recordings for training.
    • **kwargs: Additional optional parameters.
      • kilosort_path: Folder where Kilosort2 is installed. Default is None.
      • input_scale: Multiplier for recording traces. Default is 0.01.
      • learning_rate: Learning rate for training. Default is 7.76e-4.
      • validation_recording: Recording to use as validation. Default is None.
  • Returns: ModelSpikeSorter object

Kilosort2 Integration​

run_kilosort2(recording_files, **kwargs)​

Runs Kilosort2 on a set of recordings.

  • Parameters:
    • recording_files: List of recordings to spike sort.
    • **kwargs: Additional optional parameters.
      • kilosort_path: Folder where Kilosort2 is installed.
      • results_folders: List of folders to store sorted results. Default is None.
      • compile_to_npz: Whether to save results as "sorted.npz". Default is True.
      • compile_to_mat: Whether to save results as "sorted.mat". Default is False.
      • intermediate_folders: List of folders for intermediate results. Default is None.
      • delete_inter: Whether to delete intermediate folders after sorting. Default is True.
  • Returns: None (results are saved to files)

Usage Examples​

Offline Sequence Detection​

rt_sort = detect_sequences(recording, inter_path, detection_model, recording_window_ms=(0, 5*60*1000))
sequence_spike_trains = rt_sort.sort_offline(recording, inter_path, recording_window_ms=(5*60*1000, 10*60*1000), verbose=True)

Online Sorting​

rt_sort.reset()
while not done:
    obs, done = env.step()
    sequence_detections = rt_sort.running_sort(obs)
    # Process sequence_detections

For more detailed information on each function and its parameters, please refer to the respective sections in the user manual.