Gibbs-Helmholtz Graph Neural Network (GH-GNN)

This repository contains the code used in the paper Gibbs-Helmholtz Graph Neural Network: capturing the temperature dependency of activity coefficients at infinite dilution.

Do you want to use GH-GNN?

Check this notebook out to learn how to use GH-GNN for making predictions.

Do you want to run GH-GNN locally?

Install

Clone the repository and create a conda environment with all dependencies installed by running by running:

git clone https://github.com/edgarsmdn/GH-GNN.git
cd GH-GNN
conda env create -f enviornment.yml

If you use GH-GNN please cite the original publication as:

@article{sanchez2023ghgnn,
  title={Gibbs-Helmholtz Graph Neural Network: capturing the temperature dependency of activity coefficients at infinite dilution},
  author={Sanchez Medina, Edgar Ivan and Linke, Steffen and Stoll, Martin and Sundmacher, Kai},
  journal={Digital Discovery},
  DOI={10.1039/D2DD00142J},
  year={2023},
  volume={2},
  issue={3},
  pages={781-798},
  publisher={RSC},
}

If you use GH-GNN you could also give a star ⭐ to this repo

License

This material is licensed under the MIT license and is free and provided as-is. If you use the code provided in this repository please cite the original publication.