Gibbs–Helmholtz graph neural network: capturing the temperature dependency of activity coefficients at infinite dilution

Published:

A hybrid GNN model that predicts infinite dilution activity coefficients at varying temperatures. The code contains the training routines and the experiments presented in the paper Gibbs–Helmholtz graph neural network: capturing the temperature dependency of activity coefficients at infinite dilution.

To cite this work:

@Article{sanchez_medina_GHGNN_2023,
author ="Sanchez Medina, Edgar Ivan and Linke, Steffen and Stoll, Martin and Sundmacher, Kai",
title  ="Gibbs–Helmholtz graph neural network: capturing the temperature dependency of activity coefficients at infinite dilution",
journal  ="Digital Discovery",
year  ="2023",
publisher  ="RSC",
doi  ="10.1039/D2DD00142J",
url  ="https://doi.org/10.1039/D2DD00142J"}