Lifelong Machine Learning Potentials for Chemical Reaction Network Explorations – 2

This repository contains the lifelong Machine Learning Potential (lMLP) software (version 2.0.0) presented in M. Eckhoff and M. Reiher, Lifelong Machine Learning Potentials for Chemical Reaction Network Explorations, arXiv:2504.11933 [physics.chem-ph] (2025) (10.1021/acs.jctc.5c01127, 10.48550/arXiv.2504.11933). Moreover, the chemical reaction networks, reference data sets, training and model settings, generated output of this work, and scripts for analysis and plotting are provided here.

Please see also the Zenodo repository M. Eckhoff and M. Reiher, Lifelong Machine Learning Potentials for Chemical Reaction Network Explorations – 1, Zenodo 10.5281/zenodo.14689929 (2025).