Classification of Motor Imagery EEG Signals in Patients with High Uncertainty using a Spectral Transformer
This repository contains the code and documentation for the data science project developed as part of the course at the Polytechnic University of Madrid. The main objective of the project is to perform the classification of motor imagery EEG signals in patients with high uncertainty using a Spectral Transformer.
project.ipynb
: Jupyter Notebook file that contains the complete project explained step by step.utils.py
: File with utility functions required for the notebook.models.py
: Implementation of ATCNet and Spectral Transformer.
To run the project, follow these steps:
- Install the necessary dependencies:
pip install -r requirements.txt
(make sure you have the required libraries). - Open and run the
notebook.ipynb
in your Jupyter environment.
Important
Make sure you have download the BCI Competition IV 2a Dataset and it is added to your folder.
- ATCNet Paper: Link to the ATCNet paper and their original code
- Spectral Transformer Paper: Link to the Spectral Transformer paper
This project uses the following datasets:
- PhysioNet EEGMMIDB: Link to the dataset
- BCI Competition IV 2a: Link to the dataset and to the documentation
For any inquiries or collaborations, feel free to reach out:
- Email: [email protected]
- Twitter: @mixnikon
Contributions are welcome! If you wish to improve this project, please open an issue or send a pull request.