MolPool_GNN_model

MolPool GNN Models - CN and YSI Prediction

Model Description

These are Graph Neural Network (GNN) models for predicting Cetane Number (CN) and Yield Sooting Index (YSI) of fuel molecules, originally developed as part of the MolPool software package.

Original Source

These model weights are taken from MolPool, an open-source software package for molecular property prediction.

• Original Repository: [https://gitlab.kuleuven.be/creas/vermeiregroup/molpool-for-fuels-properties]
• Paper/Reference: [https://doi.org/10.1016/j.fuel.2024.133218]
• Original Authors: [Roel J. Leenhouts,Tara Larsson,Sebastian Verhelst b, Florence H. Vermeire]

License

These weights are redistributed under the original Creative Commons Attribution 4.0 International (CC-BY 4.0) license, as permitted by the original authors.

Usage in This Project

These models are used as part of a mixture property prediction tool developed as a capstone project. They are loaded via the HuggingFace Hub into a Gradio Space for predicting fuel mixture properties.

Citation

If you use these models, please cite the original MolPool work: [Roel J. Leenhouts, Tara Larsson, Sebastian Verhelst, Florence H. Vermeire, Property prediction of fuel mixtures using pooled graph neural networks, Fuel, Volume 381, Part B, 2025, 133218, ISSN 0016-2361, ]