AllenInstitute/GRNN: GRNN v1.0.0 — Differentiable biologically-derived single-neuron models

Le résumé fourni par la source

Code and data accompanying Yeung et al., "A dataset of differentiable biologically-derived single neuron models" (Cell Reports Methods, 2026). This release provides: A database of over 1,000 fitted single-neuron firing-rate (GFR) model parameter sets, with per-cell provenance metadata (cell ID, Cre line, species, cortical area), fit to patch-clamp recordings from mouse and human cortex (Allen Cell Types Database). PyTorch implementations and reproducible pipelines for fitting GFR models to electrophysiology data, evaluating fit quality (explained variance ratio), and integrating GFR units as differentiable neurons in gradient-trained network architectures.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.