Many-body machine learning models for water, acetonitrile, and methanol

GDML, GAP, and SchNet models trained on 1-, 2-, and 3-body energies and forces of water, acetonitrile, and methanol. Size-transferable NequIPs are trained on trimer data. Energies and forces were computed at the MP2/def2-TZVP level of theory in ORCA v4.2.0. Data sets, training scripts, and analyses...

Full description

Bibliographic Details
Main Authors: Maldonado, Alex M., Poltavsky, Igor, Vassilev-Galindo, Valentin, Tkatchenko, Alexandre, Keith, John A.
Format: Dataset
Language:English
Published: 2022
Subjects:
Orca
Online Access:https://zenodo.org/record/7464581
https://doi.org/10.5281/zenodo.7464581
Description
Summary:GDML, GAP, and SchNet models trained on 1-, 2-, and 3-body energies and forces of water, acetonitrile, and methanol. Size-transferable NequIPs are trained on trimer data. Energies and forces were computed at the MP2/def2-TZVP level of theory in ORCA v4.2.0. Data sets, training scripts, and analyses of these potentials are available here. Applications of these models on molecular dynamics simulations are found here. Changelog The format is based on Keep a Changelog, and this project adheres to Semantic Versioning. [0.0.2] - 2022-12-20 Added NequIPs trained for all solvents using 1000 trimers. [0.0.1] - 2022-09-25 Initial release!

Similar Items