DeeperGCN: All You Need to Train Deeper GCNs

Benchmark Model Rank Results
graph-property-prediction-on-ogbg-molhivDeeperGCN#22Test ROC-AUC: 0.7858 ± 0.0117Ext. data: No
graph-property-prediction-on-ogbg-molpcbaDeeperGCN+virtual node#21Test AP: 0.2781 ± 0.0038Ext. data: NoValidation AP: 0.2920 ± 0.0025
graph-property-prediction-on-ogbg-ppaDeeperGCN#7Ext. data: NoTest Accuracy: 0.7712 ± 0.0071
link-property-prediction-on-ogbl-collabDeeperGCN#10Test Hits@50: 0.5273 ± 0.0047Ext. data: No
node-property-prediction-on-ogbn-arxivDeeperGCN#65Test Accuracy: 0.7192 ± 0.0016Ext. data: No
node-property-prediction-on-ogbn-productsDeeperGCN#36Test Accuracy: 0.8098 ± 0.0020Ext. data: No
node-property-prediction-on-ogbn-proteinsDeeperGCN#13Ext. data: NoTest ROC-AUC: 0.8580 ± 0.0017
node-property-prediction-on-ogbn-proteinsGEN + FLAG + node2vec#15Ext. data: NoTest ROC-AUC: 0.8251 ± 0.0043