Models and data supporting the paper "Predicting neutron experiments from first principles: A workflow powered by machine learning"

This record accompanies the publication "Predicting neutron experiments from first principles: A workflow powered by machine learning". It comprises the machine-learned interatomic potentials (MLIPs) constructed and employed in that work with their respective training data as well as the experimental inelastic neutron scattering data for crystalline benzene presented in the publication.

Hydrogenated Sc-doped BaTiO3

• nep-BaScTiOH.txt – MLIP based on the neuroevolution potential (NEP) form
• nep-BaScTiOH.zip – model ensemble with the underlying training and validation data
• BaScTiOH-R2SCAN.db – database with reference data, in sql-lite format, readable using the ase package

Benzene

• nep-benzene.txt – MLIP based on the neuroevolution potential (NEP) form
• nep-benzene.zip – model ensemble with the underlying training and validation data
• benzene-CX.db – database with reference data, in sql-lite format, readable using the ase package
• reduced-benzene-tosca.zip – experimental inelastic neutron scattering data