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来自“MoleculeNet: 分子机器学习的基准”论文的QM9数据集，包含约130,000个分子，具有19个回归目标。 每个分子包括分子中原子的单一低能量构象的完整空间信息。 此外，我们提供了来自“量子化学的神经消息传递”论文的原子特征。
目标
属性
描述
单位
0
\(\mu\)
偶极矩
\(\textrm{D}\)
1
\(\alpha\)
各向同性极化率
\({a_0}^3\)
2
\(\epsilon_{\textrm{HOMO}}\)
最高占据分子轨道能量
\(\textrm{eV}\)
3
\(\epsilon_{\textrm{LUMO}}\)
最低未占据分子轨道能量
\(\textrm{eV}\)
4
\(\Delta \epsilon\)
\(\epsilon_{\textrm{HOMO}}\) 和 \(\epsilon_{\textrm{LUMO}}\) 之间的差距
\(\textrm{eV}\)
5
\(\langle R^2 \rangle\)
电子空间范围
\({a_0}^2\)
6
\(\textrm{ZPVE}\)
零点振动能
\(\textrm{eV}\)
7
\(U_0\)
0K时的内能
\(\textrm{eV}\)
8
\(U\)
298.15K时的内能
\(\textrm{eV}\)
9
\(H\)
298.15K时的焓
\(\textrm{eV}\)
10
\(G\)
298.15K时的自由能
\(\textrm{eV}\)
11
\(c_{\textrm{v}}\)
298.15K时的热容量
\(\frac{\textrm{cal}}{\textrm{mol K}}\)
12
\(U_0^{\textrm{ATOM}}\)
0K时的原子化能量
\(\textrm{eV}\)
13
\(U^{\textrm{ATOM}}\)
298.15K时的原子化能量
\(\textrm{eV}\)
14
\(H^{\textrm{ATOM}}\)
298.15K时的原子化焓
\(\textrm{eV}\)
15
\(G^{\textrm{ATOM}}\)
298.15K时的原子化自由能
\(\textrm{eV}\)
16
\(A\)
旋转常数
\(\textrm{GHz}\)
17
\(B\)
旋转常数
\(\textrm{GHz}\)
18
\(C\)
旋转常数
\(\textrm{GHz}\)
注意
我们还提供了数据集的预处理版本，以防未安装
rdkit。预处理版本与process()中概述的手动处理版本相匹配。- Parameters:
root (str) – Root directory where the dataset should be saved.
transform (callable, optional) – A function/transform that takes in an
torch_geometric.data.Dataobject 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.Dataobject 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.Dataobject 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)
统计:
#图表
#节点
#edges
#特性
#任务
130,831
~18.0
~37.3
11
19