Graph Classification On Enzymes
평가 지표
Accuracy
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | Accuracy |
|---|---|
| dagcn-dual-attention-graph-convolutional | 58.17% |
| transitivity-preserving-graph-representation | 67.22±3.92 |
| spectral-multigraph-networks-for-discovering | 61.7% |
| masked-attention-is-all-you-need-for-graphs | 79.423±1.658 |
| evolution-of-graph-classifiers | 55.67 |
| capsule-graph-neural-network | 54.67% |
| template-based-graph-neural-network-with | 75.1 |
| how-attentive-are-graph-attention-networks | 77.987±2.112 |
| semi-supervised-classification-with-graph | 73.466±4.372 |
| panda-expanded-width-aware-message-passing | 46.2 |
| wasserstein-embedding-for-graph-learning | 60.5 |
| demo-net-degree-specific-graph-neural | 27.2 |
| hierarchical-graph-representation-learning | 63.33% |
| dissecting-graph-neural-networks-on-graph | 69.50% |
| online-graph-dictionary-learning | 71.47 |
| fea2fea-exploring-structural-feature | 48.5 |
| dynamic-edge-conditioned-filters-in | 52.67% |
| deep-graph-kernels | 53.43% |
| a-fair-comparison-of-graph-neural-networks-1 | 58.2% |
| improving-spectral-graph-convolution-for | 73.33 |
| a-fair-comparison-of-graph-neural-networks-1 | 59.6% |
| optimal-transport-for-structured-data-with | 71.00% |
| graph-star-net-for-generalized-multi-task-1 | 67.1% |
| graph-trees-with-attention | 59.6 |
| a-simple-yet-effective-baseline-for-non | 35.3% |
| 190910086 | 67.30% |
| improving-attention-mechanism-in-graph-neural | 58.45 |
| graph-isomorphism-unet | 70% |
| hierarchical-representation-learning-in-graph | 43.9% |
| graph-convolutional-networks-with | 65.0% |
| wasserstein-weisfeiler-lehman-graph-kernels | 59.13% |
| how-powerful-are-graph-neural-networks | 68.303±4.170 |
| hierarchical-graph-pooling-with-structure | 68.79 |
| bridging-the-gap-between-spectral-and-spatial | 78.39 |
| a-simple-baseline-algorithm-for-graph | 43.7% |
| principal-neighbourhood-aggregation-for-graph | 73.021±2.512 |
| panda-expanded-width-aware-message-passing | 43.9 |
| graph-classification-with-recurrent | 48.4% |
| spi-gcn-a-simple-permutation-invariant-graph | 50.17% |
| panda-expanded-width-aware-message-passing | 53.1 |
| dissecting-graph-neural-networks-on-graph | 70.17% |
| capsule-neural-networks-for-graph | 27% |
| hierarchical-graph-representation-learning | 62.53% |
| panda-expanded-width-aware-message-passing | 31.55 |
| gaussian-induced-convolution-for-graphs | 62.50% |
| graph-attention-networks | 78.611±1.556 |
| a-non-negative-factorization-approach-to-node | 24.1% |
| towards-a-practical-k-dimensional-weisfeiler | 58.2% |
| fine-tuning-graph-neural-networks-by | - |
| bridging-the-gap-between-spectral-and-spatial | 65.13 |
| dropgnn-random-dropouts-increase-the | 65.128±4.117 |
| recipe-for-a-general-powerful-scalable-graph | 78.667±4.625 |
| when-work-matters-transforming-classical | 67.50% |