This repository contains code from our participation (Ketan Jaltare and Fabien Pavy) in the Harvard Medical School - Harmful Brain Activity Classification competition hosted on Kaggle.

The goal of the competition was to detect and classify seizures and other types of harmful brain activity from EEG recordings obtained from critically ill patients.

The repository includes Kaggle notebooks that:

• Use a combination of libraries such as NumPy, SciPy, PyTorch, Pandas, and Polars
• Perform feature extraction from EEG data
• Fit and evaluate various classification models, including:
  • Deep Neural Networks (DNNs)
  • Random Forests
  • Gradient boosted trees

• The code was adapted specifically for the competition setting on Kaggle.
• Some scripts may require adjustment if used outside of the Kaggle environment.