This is the github repo for the paper"Conservative semi-lagrangian finite difference scheme for transport simulations using graph neural networks"
We provide two simple numerical examples of this GNN-based method. You can run these .py files to train the model.
To test the model, you need install these packages. Make sure you have the corresponding cuda Pre-requisites:
Python 3.8.8 or higher
pip install torch==2.2.1+cu121 torchvision==0.13.1+cu121 torchaudio==2.2.1 -f https://download.pytorch.org/whl/torch_stable.html
pip install torch_geometric
@article{chen2024conservative,
title={Conservative semi-lagrangian finite difference scheme for transport simulations using graph neural networks},
author={Chen, Yongsheng and Guo, Wei and Zhong, Xinghui},
journal={arXiv preprint arXiv:2405.01938},
year={2024}
}If you have any questions, please contact 12335005@zju.edu.cn