OGB/ogbg-molhiv
收藏资源简介:
`ogbg-molhiv`数据集是一个小型的分子属性预测数据集,由斯坦福团队从MoleculeNet改编而来,作为开放图基准(Open Graph Benchmark)的一部分。该数据集用于预测分子是否抑制HIV,是一个二分类任务,评估指标为ROC-AUC。数据集包含41,127个图,每个图包含节点特征、边索引、边属性和标签等信息。数据集遵循PyGeometric版本的数据分割,并提供了加载数据集的代码示例。
The `ogbg-molhiv` dataset is a small-scale molecular property prediction dataset adapted from MoleculeNet by a Stanford research team, and is included as part of the Open Graph Benchmark (OGB). This dataset targets a binary classification task aimed at predicting whether a molecule exhibits HIV inhibitory activity, with ROC-AUC serving as its evaluation metric. It consists of 41,127 molecular graphs, each containing node features, edge indices, edge attributes, labels and other relevant information. The dataset adheres to the data splitting strategy of the PyGeometric version, and provides code examples for loading the dataset.
数据集概述
数据集名称
ogbg-molhiv
数据集摘要
ogbg-molhiv 是一个小型分子属性预测数据集,由斯坦福团队从MoleculeNet改编,作为Open Graph Benchmark的一部分。
支持的任务和排行榜
- 任务类型:分子属性预测,旨在预测分子是否抑制HIV,属于二分类任务。
- 评估指标:ROC-AUC。
- 排行榜:
数据集结构
数据属性
- 规模:小
- 图数量:41,127
- 平均节点数:25.5
- 平均边数:27.5
- 平均节点度:2.2
- 平均聚类系数:0.002
- 最大强连通分量比率:0.993
- 图直径:12.0
数据字段
node_feat(列表: #nodes x #node-features)edge_index(列表: 2 x #edges)edge_attr(列表: #edges x #edge-features)y(列表: 1 x #labels)num_nodes(整数)
数据分割
数据遵循PyGeometric版本的数据分割,可通过以下代码获取分割信息: python from ogb.graphproppred import PygGraphPropPredDataset
dataset = PygGraphPropPredDataset(name = ogbg-molhiv)
split_idx = dataset.get_idx_split() train = dataset[split_idx[train]] # valid, test
附加信息
许可信息
数据集已根据MIT许可证发布。
引用信息
@inproceedings{hu-etal-2020-open, author = {Weihua Hu and Matthias Fey and Marinka Zitnik and Yuxiao Dong and Hongyu Ren and Bowen Liu and Michele Catasta and Jure Leskovec}, editor = {Hugo Larochelle and Marc Aurelio Ranzato and Raia Hadsell and Maria{-}Florina Balcan and Hsuan{-}Tien Lin}, title = {Open Graph Benchmark: Datasets for Machine Learning on Graphs}, booktitle = {Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual}, year = {2020}, url = {https://proceedings.neurips.cc/paper/2020/hash/fb60d411a5c5b72b2e7d3527cfc84fd0-Abstract.html}, }
贡献者
感谢 @clefourrier 添加此数据集。




