Node Classification On Citeseer Full
评估指标
Accuracy
评测结果
各个模型在此基准测试上的表现结果
| Paper Title | Repository | ||
|---|---|---|---|
| IncepGCN+DropEdge | 80.50% | DropEdge: Towards Deep Graph Convolutional Networks on Node Classification | |
| ASGCN | 79.66% | Adaptive Sampling Towards Fast Graph Representation Learning | |
| FastGCN | 77.60% | FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling | |
| GCNII* | 77.13% | Simple and Deep Graph Convolutional Networks | |
| FDGATII | 75.6434% | FDGATII : Fast Dynamic Graph Attention with Initial Residual and Identity Mapping | |
| Graph ESN | 74.5±2.1 | Beyond Homophily with Graph Echo State Networks | - |
| GraphSAGE | 71.40% | Inductive Representation Learning on Large Graphs |
0 of 7 row(s) selected.