{"references": ["Welborn, M.; Cheng, L. & Miller III, T. F. Transferability in machine learning for electronic structure via the molecular orbital basis. J. Chem. Theory Comput., 2018, 14, 4772-4779. https://pubs.acs.org/doi/abs/10.1021/acs.jctc.8b00636", "Cheng, L.; Welborn, M.; Christensen, A. S. & Miller III, T. F. A universal density matrix functional from molecular orbital-based machine learning: Transferability across organic molecules. J. Chem. Phys., 2019, 150, 131103. https://aip.scitation.org/doi/full/10.1063/1.5088393"]}

Data curated by the QCArchive team, originally sourced from data.caltech.edu. 1,000 small organic molecules with up to 13 heavy atoms sampled from GDB-13 and thermalized with 350K MD at the B3LYP/6-31g* level of theory. Energies are provided at the HF/cc-pVTZ and MP2/cc-pVTZ levels of theory. All molecules are neutral singlets. For more information, see http://qcarchive.molssi.org/apps/ml_datasets/.