miniML models#
Currently, miniML includes the following models:
See also
Examples of using these models can be found in the paper.
miniML base model (GC mEPSC model)#
The base model trained from scratch using the GC mEPSC dataset. This model is used as a starting point for transfer learning. The labeled training originates from spontaneous mEPSC recordings of cerebellar granule cells in Delvendahl et al..
Transfer learning models#
1. miniML GC mEPSP model#
Transfer learning model for mEPSP detection.
2. zebrafish sEPSC model#
Transfer learning model for sEPSC detection. Training data was generated using data from Rupprecht et al..
3. Drosophila NMJ wt mEPSC model#
Transfer learning model for mEPSC detection in Drosophila NMJ two-electrode voltage-clamp recordings. Training data was generated using data from Baccino-Calace et al..
4. Drosophila NMJ glurIIa mEPSC model#
Transfer learning model for mEPSC detection in Drosophila NMJ two-electrode voltage-clamp recordings of GluRIIa mutants. Training data was generated using data from Baccino-Calace et al..
5. iGluSnFR optical minis model#
Transfer learning model for mEPSC detection in optical mini recordings using iGluSnFR. Training data was generated using data from Agarwal et al..
See also
Associated training datasets can be found on Zenodo.