QM9-OR: DFT Optimized Geometries and Optical Rotations for Selected QM9 Molecules

The QM9-OR data set consists of 121,416 molecules extracted from the original QM9 data set. For each molecule, the coordinate geometry is optimized using density functional theory, and the calculated specific rotations at three wavelengths using CAM-B3LYP/6-31G** are provided. The number of chiral centers and the absolute configurations are also provided for each molecule. Data set is a binary file in NumPy (npy) format. Each entry contains the following: index number, InChI identifier, Cartesian coordiantes, one-hot encoded atom type (H, C, N, O, F), chiral centers, and calculated optical rotations at 355 nm, 589.3 nm, and 633 nm.