GNN-TL descriptors for organic molecules in QM9NMR dataset

This dataset is a part of supplemental data in arXiv:2402.18433.<br>GNN-TL descriptors from M3GNet, MEGNet, MACE-MP0 and MACE-OFF23Important columns are noted below to use each descriptor.molecule_formuladsgdb_idx (The name of molecule)nmr_shiftatomic_numberatomic_symbolx,y,z (atoms positions)atom_feature_vector1~N (N depends on the type of descriptor. For example, N is 64 in M3GNet.)<br><br><br><br><br><br>Please unzip using tar command.```tar -zxvf ***.tar<br>```The way of reading dataset in python.```import pandas as pddf = pd.read_csv(f'path/to/csv')```<br><b>Citations</b>Original QM9 dataset (R. Ramakrishnan, P. O. Dral, M. Rupp and O. A. Von Lilienfeld, Scientific data, 2014, 1, 1–7)QM9NMR dataset (A. Gupta, S. Chakraborty and R. Ramakrishnan, Machine<br>Learning: Science and Technology, 2021, 2, 035010)T. Shiota, K. Ishihara, W. Mizukami, Universal neural network potentials as descriptors: Towards scalable chemical property prediction using quantum and classical computers,<br>arXiv:2402.18433 [quant-ph]<br>