Brain-computer interfaces (BCIs) are reshaping how humans interact with machines. From restoring motor function in paralysis patients to hands-free device control, the ability to decode neural intent in real time is foundational to next-generation assistive technology and neuroscience research.

This project demonstrates the core ML pipeline behind such systems: signal generation, preprocessing, classification, and evaluation — built entirely in Python with production-ready modularity.

• Synthetic EEG Generation — Produces biologically-grounded signals across three frequency bands (delta/theta, alpha/beta, gamma) to simulate distinct cognitive states
• Automated Preprocessing — Applies standard scaling and an 80/20 train-test split with reproducible random seeds
• Random Forest Classifier — 100-estimator ensemble trained on 64-channel feature vectors across 1,000 samples
• Signal Visualization — Per-class waveform plots that expose frequency-domain differences between brain states
• Model Persistence — Exports a serialized .pkl model for downstream inference or deployment
• Modular Architecture — Each pipeline stage is an isolated function, making substitution of real EEG datasets or deep learning models straightforward

Raw EEG Signal (64 channels, 1s window)
             |
             v
    +------------------+
    |  Signal Generator |   <-- synthetic or real dataset
    +------------------+
             |
             v
    +------------------+
    |   Preprocessor   |   StandardScaler + train/test split
    +------------------+
             |
             v
    +------------------+
    | Random Forest     |   n_estimators=100, random_state=42
    | Classifier        |
    +------------------+
             |
       +-----+-----+
       |           |
       v           v
  Evaluation    Saved Model
  (accuracy,    (.pkl export)
  F1, recall)

# Clone the repository
git clone https://github.com/LakshmiSravyaVedantham/EEG.git
cd EEG

# Create and activate a virtual environment
python -m venv venv
source venv/bin/activate        # Windows: venv\Scripts\activate

# Install dependencies
pip install numpy pandas scikit-learn matplotlib seaborn scipy joblib

python EEG/main.py

This single command will:

1. Generate 1,000 synthetic EEG samples across 64 time-domain features
2. Plot representative waveforms for each brain state
3. Train and evaluate a Random Forest classifier
4. Print accuracy and a full per-class classification report
5. Save the trained model to neural_signal_classifier.pkl

Run python EEG/main.py to populate this table with your results. Overall accuracy typically exceeds 90% on the synthetic dataset.

Contributions are welcome. To get started:

1. Fork the repository
2. Create a feature branch: git checkout -b feature/your-feature
3. Commit your changes: git commit -m "Add your feature"
4. Push to the branch: git push origin feature/your-feature
5. Open a pull request

Please follow PEP 8 style conventions and include tests for any new functionality.

This project is licensed under the MIT License. See LICENSE for details.