This is the Repository of TSGL-EEGNet. TSGL-EEGNet is a kind of Convolutional Network for pre-processed EEG signals to predict classes. TSGL-EEGNet is based on EEGNet which you can find it here.

• BCI Competition IV 2a

You need to do some pre-processing works before using it. Maybe you need MATLAB and EEGlab tools to read these datas and transform them to *.mat file format.

• Cross Validation Framework
• Grid Search Framework
• Cropped Training with CV and GS
• Visualization Framework

• EEGNet [1]
• TSGL-EEGNet
• DeepConvNet [2]
• ShallowConvNet [2]
• Multi-branch 3D CNN [3]
• FBCSP

• Python >= 3.6
• tensorflow-gpu == 2.1.0
• scikit-learn >= 0.21.3
• scipy >= 1.3.1
• numpy >= 1.17.3
• pydot >= 1.4.1
• hdf5 >= 1.10.4
• h5py >= 2.9.0
• matplotlib >= 3.1.1
• graphviz >= 2.38
• mne >= 0.20.7 (pip)

It is recommended to use conda environment.

import TSGLEEGNet as tsgleeg
# TODO: finish the doc

If you use the EEGNet model in your research and found it helpful, please cite the following paper:

@article{Lawhern2018,
    author={Vernon J Lawhern and Amelia J Solon and Nicholas R Waytowich and Stephen M Gordon and Chou P Hung and Brent J Lance},
    title={EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces},
    journal={Journal of Neural Engineering},
    volume={15},
    number={5},
    pages={056013},
    url={https://stacks.iop.org/1741-2552/15/i=5/a=056013},
    year={2018}
}

MIT License