Node Classification On Cora Full Supervised
Metriken
Accuracy
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
| Paper Title | ||
|---|---|---|
| GCNII | 88.49% | Simple and Deep Graph Convolutional Networks |
| IncepGCN+DropEdge | 88.2% | DropEdge: Towards Deep Graph Convolutional Networks on Node Classification |
| FDGATII | 87.7867% | FDGATII : Fast Dynamic Graph Attention with Initial Residual and Identity Mapping |
| ASGCN | 87.44±0.0034% | Adaptive Sampling Towards Fast Graph Representation Learning |
| Graph ESN | 86.0±1.0 | Beyond Homophily with Graph Echo State Networks |
| FastGCN | 85.00% | FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling |
| GraphSAGE | 82.2% | Inductive Representation Learning on Large Graphs |
| NCGCN | 73.42 ± 0.58% | Clarify Confused Nodes via Separated Learning |
| GraphMix (GCN) | 61.8% | GraphMix: Improved Training of GNNs for Semi-Supervised Learning |
0 of 9 row(s) selected.