This is the repository for the Kaggle competition LEAP - Atmospheric Physics using AI (ClimSim). The goal of the competition is to develop machine learning models that accurately emulate subgrid-scale atmospheric physics in an operational climate model—an important step in improving climate projections and reducing uncertainty surrounding future climate trends.
In this repository, our development is based on the deep learning framework PyTorch, leveraging PyTorch Lightning to streamline the training process, poetry for dependency management. Our linting, type checking, and formatting tools include black, pylint, isort, and mypy.
git clone https://github.com/wyhwong/Kaggle-ClimSim-2024.git
cd Kaggle-ClimSim-2024See README.md for more details.