gumpy-bci / gumpy-deeplearning Goto Github PK

Hi, can you please explain how did you use kapre library to evaluate spectrograms and 2D-normalization. When I'm trying to run the code it says cannot found kapre.utils and cannot import Spectrogram from kapre.time_frequency and when I look at kapre library it seems that there is no function for computing spectrogram and also normalization. Can you please help me?

After running the example notebook (CNN_STFT_model.ipynb) my accuracy does not increase even after 100 epoch, accuracy is stuck at 49% (subject 1). This is quite different from what is reported in sensors journal entitled : Validating deep neural networks for online decoding of motor imagery movements from eeg signals could you rerun the notebook and share your result?

Hello, I used pCNN on MI dataset II 2a which is a four-class dataset. I could only get about 48% accuracy on average. Could you please share this part of code? It would be very helpful. Thanks!