Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering

Benchmark Model Rank Results
graph-property-prediction-on-ogbg-molpcbaChebNet#26Test AP: 0.2306 ± 0.0016Ext. data: NoValidation AP: 0.2372 ± 0.0018
node-classification-on-citeseerChebNet#50Accuracy: 69.8%
node-classification-on-citeseer-05ChebyNet#10Accuracy: 45.3%
node-classification-on-citeseer-1ChebyNet#10Accuracy: 59.4%
node-classification-on-citeseer-with-publicChebyNet#32Accuracy: 70.1%
node-classification-on-coraChebNet#50Accuracy: 81.2%
node-classification-on-cora-05ChebyNet#15Accuracy: 33.9%
node-classification-on-cora-1ChebyNet#15Accuracy: 44.2%
node-classification-on-cora-3ChebyNet#14Accuracy: 62.1%
node-classification-on-cora-with-public-splitChebyNet#33Accuracy: 78.0%
node-classification-on-pubmedChebNet#52Accuracy: 74.4%
node-classification-on-pubmed-003ChebyNet#14Accuracy: 45.3%
node-classification-on-pubmed-005ChebyNet#14Accuracy: 48.2%
node-classification-on-pubmed-01ChebyNet#14Accuracy: 55.2%
node-classification-on-pubmed-with-publicChebyNet#36Accuracy: 69.8%
skeleton-based-action-recognition-on-sbuChebyNet#3Accuracy: 96.00%