Graph Regression On F2
評価指標
R2
RMSE
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | R2 | RMSE | Paper Title | Repository |
|---|---|---|---|---|
| TokenGT | 0.872±0.000 | 0.363±0.872 | Pure Transformers are Powerful Graph Learners | |
| Graphormer | OOM | OOM | Do Transformers Really Perform Bad for Graph Representation? | |
| GCN | 0.878±0.000 | 0.355±0.878 | Semi-Supervised Classification with Graph Convolutional Networks | |
| GIN | 0.887±0.000 | 0.342±0.887 | How Powerful are Graph Neural Networks? | |
| ESA (Edge set attention, no positional encodings) | 0.891±0.000 | 0.335±0.891 | An end-to-end attention-based approach for learning on graphs | - |
| PNA | 0.891±0.000 | 0.336±0.891 | Principal Neighbourhood Aggregation for Graph Nets | |
| DropGIN | 0.886±0.000 | 0.343±0.886 | DropGNN: Random Dropouts Increase the Expressiveness of Graph Neural Networks | |
| GAT | 0.886±0.000 | 0.343±0.886 | Graph Attention Networks | |
| GATv2 | 0.885±0.000 | 0.344±0.885 | How Attentive are Graph Attention Networks? |
0 of 9 row(s) selected.