Software implementing methods from: David Nahmias and Kimberly Kontson. Easy Perturbation EEG Algorithm for Spectral Importance (easyPEASI): A simple method to identifyimportant spectral features of EEG in deep learning models. In Review. (2020)
This work identifies frequency bands that are important in EEG-driven deep learning classifications.
To train and save deep learning models for all classifications
.\deepShell.shwith files in data subfolder.
Once models are generated,
python testModels.pyDeveloped and tested on Python 3.5, packages used found in software header. Deep learning implemented using PyTorch and Braindecode package. Make file coming soon.
- 0.0.1
- Initial commit of stable project
David Nahmias – Website – david.nahmias@fda.hhs.gov
Distributed under the public domain license. See LICENSE for more information.
Summited to conference, full citation forthcoming.
David Nahmias and Kimberly Kontson. Easy Perturbation EEG Algorithm for Spectral Importance (easyPEASI): A simple method to identifyimportant spectral features of EEG in deep learning models. In Review (2020)