This is an example project to show how to use machine learning in context of FFRs.
main.ipynb
contains models for linear regression, logistic regression, support vector machine, k-nearest neighbors, and k-means clusteringneural_network.ipynb
contains a neural network modelFFR_data.csv
is a practice dataset based on this paper: "Hart, B. N. & Jeng, F.-C. (2021) A demonstration of machine learning in detecting frequency following responses in American neonates. Percept Mot Skills, 128(1) 48–58. www.doi.org/10.1177/0031512520960390". This dataset contains six FFR features (FrequencyError, SlopeError, TrackingAccuracy, SpectralAmplitude, PitchStrength, and RMSAmplitude) and a target response label (FFR).
This package is developed by the Auditory Electrophysiology Laboratory at Ohio University. This package is distributed to facilitate learning, but without any warranty. The user is free to use, modify, and re-distribute this package under the terms of an MIT license.