Skip to main content

Loading Detection Models

To use RT-Sort, you first need to load a detection model. Here's how to do it:

from braindance.core.spikedetector.model2 import ModelSpikeSorter

detection_model = ModelSpikeSorter.load(detection_model_path)

Parameters​

  • detection_model_path: Path to a folder containing a ModelSpikeSorter object's init_dict.json and state_dict.pt.

Returns​

  • ModelSpikeSorter: Represents the loaded detection model.

Pre-trained Models​

Pre-trained ModelSpikeSorter objects for Maxwell MEAs and Neuropixels used in the RT-Sort manuscript can be found in the BrainDance GitHub repository:

BrainDance/braindance/core/spikedetector/detection_models

Next Steps​

Once you've loaded a detection model, you can proceed to sequence detection or real-time application.