displacement_energies.tar.bz2 (11.31 MB)

displacement_energies.tar.bz2

Name: displacement_energies.tar.bz2
Published: 2018-08-23T05:43:45+00:00
License: https://creativecommons.org/publicdomain/zero/1.0/
Keywords: computational chemistry, machine learning, quantum chemistry, Computational Chemistry

dataset

posted on 2018-08-23, 05:43 authored by Anders S. ChristensenAnders S. Christensen

Contains:

mc_dsgdb9nsd_XXXXXX_corrected_YYY_Z_TYPE.txt
...

Naming convention:

dsgdb9nsd: The initial structure is from the QM9/GDB9 set.
XXXXXX: The QM9 index. 
corrected: The structure was optimized at the wB97xD/6-31G(d) level of theory using Gaussian09.
YYY: The number (YYY) of samples of each of the molecules fragments was used to train the machine.
Z: The max number of heavy atoms in any fragments used to train the machine.
TYPE: "disp": file contains RMSD displacements along each normal mode.
      "true": file contains the true QM wB97xD/6-31G(d) energy in kcal/mol.
      "pred": file contains the predicted energy in kcal/mol.