torch_geometric.datasets.QM9

class QM9(root: str, transform: Optional[Callable] = None, pre_transform: Optional[Callable] = None, pre_filter: Optional[Callable] = None, force_reload: bool = False)[source]

Bases: InMemoryDataset

The QM9 dataset from the “MoleculeNet: A Benchmark for Molecular Machine Learning” paper, consisting of about 130,000 molecules with 19 regression targets. Each molecule includes complete spatial information for the single low energy conformation of the atoms in the molecule. In addition, we provide the atom features from the “Neural Message Passing for Quantum Chemistry” paper.

Target

Property

Description

Unit

0

\(\mu\)

Dipole moment

\(\textrm{D}\)

1

\(\alpha\)

Isotropic polarizability

\({a_0}^3\)

2

\(\epsilon_{\textrm{HOMO}}\)

Highest occupied molecular orbital energy

\(\textrm{eV}\)

3

\(\epsilon_{\textrm{LUMO}}\)

Lowest unoccupied molecular orbital energy

\(\textrm{eV}\)

4

\(\Delta \epsilon\)

Gap between \(\epsilon_{\textrm{HOMO}}\) and \(\epsilon_{\textrm{LUMO}}\)

\(\textrm{eV}\)

5

\(\langle R^2 \rangle\)

Electronic spatial extent

\({a_0}^2\)

6

\(\textrm{ZPVE}\)

Zero point vibrational energy

\(\textrm{eV}\)

7

\(U_0\)

Internal energy at 0K

\(\textrm{eV}\)

8

\(U\)

Internal energy at 298.15K

\(\textrm{eV}\)

9

\(H\)

Enthalpy at 298.15K

\(\textrm{eV}\)

10

\(G\)

Free energy at 298.15K

\(\textrm{eV}\)

11

\(c_{\textrm{v}}\)

Heat capavity at 298.15K

\(\frac{\textrm{cal}}{\textrm{mol K}}\)

12

\(U_0^{\textrm{ATOM}}\)

Atomization energy at 0K

\(\textrm{eV}\)

13

\(U^{\textrm{ATOM}}\)

Atomization energy at 298.15K

\(\textrm{eV}\)

14

\(H^{\textrm{ATOM}}\)

Atomization enthalpy at 298.15K

\(\textrm{eV}\)

15

\(G^{\textrm{ATOM}}\)

Atomization free energy at 298.15K

\(\textrm{eV}\)

16

\(A\)

Rotational constant

\(\textrm{GHz}\)

17

\(B\)

Rotational constant

\(\textrm{GHz}\)

18

\(C\)

Rotational constant

\(\textrm{GHz}\)

Note

We also provide a pre-processed version of the dataset in case rdkit is not installed. The pre-processed version matches with the manually processed version as outlined in process().

Parameters:

  • root (str) – Root directory where the dataset should be saved.

  • transform (callable, optional) – A function/transform that takes in an torch_geometric.data.Data object and returns a transformed version. The data object will be transformed before every access. (default: None)

  • pre_transform (callable, optional) – A function/transform that takes in an torch_geometric.data.Data object and returns a transformed version. The data object will be transformed before being saved to disk. (default: None)

  • pre_filter (callable, optional) – A function that takes in an torch_geometric.data.Data object and returns a boolean value, indicating whether the data object should be included in the final dataset. (default: None)

  • force_reload (bool, optional) – Whether to re-process the dataset. (default: False)

STATS:

#graphs

#nodes

#edges

#features

#tasks

130,831

~18.0

~37.3

11

19