The code has two CNN models (model 1 and 2) SGD and adam based.
Code for Wigner-ville distribution also included
Feature: RAW EEG, Short Time Fourier transform, Wigner-ville distribution
Deep learning parameters adaption: Bayesian Optimisation
platform : Matlab, Python
This code is for replicating experiment by paper titled "Can a Single Model Deep Learning Approach Enhance Classification Accuracy of an EEG-based Brain-Computer Interface?"
Cite these papers if you are using part of code:
-
Roy, Sujit, et al. "Channel Selection Improves MEG-based Brain-Computer Interface." 2019 9th International IEEE/EMBS Conference on Neural Engineering (NER). IEEE, 2019.
-
Roy, S., McCreadie, K. and Prasad, G., 2019, October. Can a Single Model Deep Learning Approach Enhance Classification Accuracy of an EEG-based Brain-Computer Interface?. In 2019 IEEE International Conference on Systems, Man and Cybernetics (SMC) (pp. 1317-1321). IEEE.
-
Roy, S., Chowdhury, A., McCreadie, K. and Prasad, G., 2020. Deep learning based inter-subject continuous decoding of motor imagery for practical brain-computer interfaces. Frontiers in Neuroscience, 14.