OGB: Open Graph Benchmark

OGB: Benchmark datasets, data loaders and evaluators for graph machine learning. 

Open Graph Benchmark | A collection of benchmark datasets, data-loaders and evaluators for graph machine learning in PyTorch. (stanford.edu)

OGB上有多个数据集：

For Node Property Prediction

Scale	Name	Package	#Nodes	#Edges*	#Tasks	Split Type	Task Type	Metric
Medium	ogbn-products	>=1.1.1	2,449,029	61,859,140	1	Sales rank	Multi-class classification	Accuracy
Medium	ogbn-proteins	>=1.1.1	132,534	39,561,252	112	Species	Binary classification	ROC-AUC
Small	ogbn-arxiv	>=1.1.1	169,343	1,166,243	1	Time	Multi-class classification	Accuracy
Large	ogbn-papers100M	>=1.2.0	111,059,956	1,615,685,872	1	Time	Multi-class classification	Accuracy
Medium	ogbn-mag	>=1.2.1	1,939,743	21,111,007	1	Time	Multi-class classification	Accuracy

For Link Property Prediction

Scale	Name	Package	#Nodes	#Edges*	Split Type	Task Type	Metric
Medium	ogbl-ppa	>=1.1.1	576,289	30,326,273	Throughput	Link prediction	Hits@100
Small	ogbl-collab	>=1.2.1	235,868	1,285,465	Time	Link prediction	Hits@50
Small	ogbl-ddi	>=1.2.1	4,267	1,334,889	Protein target	Link prediction	Hits@20
Medium	ogbl-citation2	>=1.2.4	2,927,963	30,561,187	Time	Link prediction	MRR
Medium	ogbl-wikikg2	>=1.2.4	2,500,604	17,137,181	Time	KG completion	MRR
Small	ogbl-biokg	>=1.2.0	93,773	5,088,434	Random	KG completion	MRR
Medium	ogbl-vessel*	>=1.3.4	3,538,495	5,345,897	Random	Link prediction	ROC-AUC

For Graph Property Prediction

Scale	Name	Package	#Graphs	#Nodes per graph	#Edges per graph*	#Tasks	Split Type	Task Type	Metric
Small	ogbg-molhiv	>=1.1.1	41,127	25.5	27.5	1	Scaffold	Binary classification	ROC-AUC
Medium	ogbg-molpcba	>=1.2.2	437,929	26.0	28.1	128	Scaffold	Binary classification	AP
Medium	ogbg-ppa	>=1.1.1	158,100	243.4	2,266.1	1	Species	Multi-class classification	Accuracy
Medium	ogbg-code2	>=1.2.5	452,741	125.2	124.2	1	Project	Sub-token prediction	F1 score

For Large-Scale Graph ML

Task category	Name	Package	#Graphs	#Total nodes	#Total edges	Task Type	Metric	Download size
Node-level	MAG240M	>=1.3.2	1	244,160,499	1,728,364,232	Multi-class classification	Accuracy	167GB
Link-level	WikiKG90Mv2	>=1.3.3	1	91,230,610	601,062,811	KG completion	MRR	89GB
Graph-level	PCQM4Mv2	>=1.3.2	3,746,619	52,970,652	54,546,813	Regression	MAE	59MB‡

OGB没有装成功过。