AilurusUmbra/EEGNet-tutorial

EEGNet tutorial

EEGNet-tutorial

This is a naive Pytorch implementation of EEGNet We don't consider data preprocessing, visualization , ablation test, within/cross-subject experiments.

Setup environment

0. Clone this repo

git clone https://github.com/AilurusUmbra/EEGNet-tutorial.git
cd EEGNet-tutorial

1. From conda (recommended)

conda env create -f environment.yml

2. From pip

pip install -r requirements.txt

Run

1. Start to train EEGNet

python run.py

2. Distributed tuning hyperparameters

• Using ray tune

Noted that the path of dataset in dataloader.py should be modified to absolute path, since that ray tune generates several worker processes to run training function at different working paths.

python tuning.py

Notes