EEGNet

EEGNet: A lightweight convolutional neural network for EEG signal classification using PyTorch. This repository includes the model architecture and a training pipeline for efficient EEG signal processing.

EEGNet: A Compact CNN for EEG Signal Processing

This repository contains an implementation of EEGNet, a lightweight convolutional neural network designed for EEG (electroencephalography) signal classification. The architecture efficiently extracts temporal and spatial features from EEG signals and is optimized for real-time processing.

Files

• EEGNet.py: Defines the EEGNet model architecture using PyTorch.
• trainEEGNet.py: Contains the training pipeline for EEGNet, including dataset loading, training, and evaluation.

Features

• Compact and Efficient: Uses depthwise and pointwise convolutions for efficient feature extraction.
• Flexible Architecture: Customizable number of filters, channels, and pooling parameters.
• Classification Ready: Outputs predictions for EEG signal classification tasks with multiple classes.

Installation

Ensure you have Python 3.8+ and install the required dependencies:

pip install torch torchvision numpy

Usage

Training the Model

Run the training script to train EEGNet on synthetic or real EEG data:

python trainEEGNet.py

Modifying Parameters

You can modify model parameters like the number of filters, channels, and pooling factors in eegnet.py and dataset configurations in trainEEGNet.py.

Dataset

The training script uses a synthetic EEG dataset. Replace the EEGDataset class in trainEEGNet.py with actual EEG data for real-world applications.

License

This project is licensed under the MIT License.

Author

Developed by Abhishek Mishra