7th CCDC CSP blind test training data for AIMNet2 machine-learned neural network potentials

Training data used by Group 16 for both phases (structure generation and structure ranking) of the seventh CCDC CSP blind test. These target-specific datasets contain molecular clusters (n-mers) and their properties computed with B97M/def2-TZVPP (meta-GGA DFT) method. Each data file contains different number of n-mers, depending on the target. DFT calculation performed with ORCA 5.0.3 software. Properties include energy, forces, atomic charges, and molecular dipole and quadrupole moments.

The results and conclusions of the 7th CCDC CSP blind test have been published in Acta Crystallographica:

1) The Seventh Blind Test of Crystal Structure Prediction: Structure Generation Methods – Hunnisett et al., J. Acta. Cryst. B80, Dec 2024.

2)The Seventh Blind Test of Crystal Structure Prediction: Structure Ranking Methods – Hunnisett et al., J. Acta. Cryst., B80, Dec 2024.