RGNN

This repo illustrates the RGNN model implementation in the paper EEG-Based Emotion Recognition Using Regularized Graph Neural Networks. The model is based on torch geometric v1.2.1

The EmotionDL regularizer is easy to implement and thus not included in the repo. More details can be found in the paper.

If you find the paper or this repo useful, please cite

@article{zhong2020eeg,
  title={EEG-Based Emotion Recognition Using Regularized Graph Neural Networks},
  author={Zhong, Peixiang and Wang, Di and Miao, Chunyan},
  journal={IEEE Transactions on Affective Computing},
  year={2020},
  publisher={IEEE}
}