Graph Regression On Esr2
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
比較表
| モデル名 | R2 | RMSE |
|---|---|---|
| graph-attention-networks | 0.666±0.000 | 0.510±0.666 |
| dropgnn-random-dropouts-increase-the | 0.675±0.000 | 0.503±0.675 |
| masked-attention-is-all-you-need-for-graphs | 0.697±0.000 | 0.486±0.697 |
| semi-supervised-classification-with-graph | 0.642±0.000 | 0.528±0.642 |
| how-attentive-are-graph-attention-networks | 0.655±0.000 | 0.518±0.655 |
| how-powerful-are-graph-neural-networks | 0.668±0.000 | 0.509±0.668 |
| principal-neighbourhood-aggregation-for-graph | 0.696±0.000 | 0.486±0.696 |
| pure-transformers-are-powerful-graph-learners | 0.641±0.000 | 0.529±0.641 |
| do-transformers-really-perform-bad-for-graph | OOM | OOM |