training_data.tar.bz2

Contains a number of molecule structures with the following information calculated at the wB97xD/6-31G(d) level of theory using Gaussian 09: Coordinates [Å] Atomization Energy [kcal/mol] Total Energy [kcal/mol] Nuclear Forces [kcal/mol/Å] Dipole Moment [Debye] Number of atoms (Size) Number heavy (i.e. non-hydrogen) atoms (Heavy) Nuclear Charges Element types SMILES String Structures are created using Normal Mode Sampling from local mimima obtained from a conformational screening. Update 2026-02-28: Upload of the data in an alternative npz format with identical content