Altaheri / EEG-ATCNet

Attention temporal convolutional network for EEG-based motor imagery classification
https://ieeexplore.ieee.org/document/9852687
Apache License 2.0
173 stars 22 forks source link

model #8

Closed chenzhenwoaini closed 1 year ago

chenzhenwoaini commented 1 year ago

Hello! Regarding the model, I would like to ask why a convolution accuracy similar to EEGNet can reach around 81% after ablation experiments?

Altaheri commented 1 year ago

Hello. The convolutional (CV) block differs from EEGNet by using 2D convolution instead of separable convolution, which showed better performance. CV block also uses different parameter values than those used in EEGNet.