Graph Property Prediction On Ogbg Molpcba
Métriques
Ext. data
Number of params
Test AP
Validation AP
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Ext. data | Number of params | Test AP | Validation AP |
|---|---|---|---|---|
| graph-convolutions-that-can-finally-model | No | 6147029 | 0.2979 ± 0.0030 | 0.3126 ± 0.0023 |
| graph-learning-with-1d-convolutions-on-random | No | 6115728 | 0.2986 ± 0.0025 | 0.3075 ± 0.0020 |
| how-powerful-are-graph-neural-networks | No | 3374533 | 0.2703 ± 0.0023 | 0.2798 ± 0.0025 |
| unlocking-the-potential-of-classic-gnns-for | No | 6016860 | 0.2981 ± 0.0024 | 0.3011 ± 0.0037 |
| semi-supervised-classification-with-graph | No | 565928 | 0.2020 ± 0.0024 | 0.2059 ± 0.0033 |
| parameterized-hypercomplex-graph-neural | No | 1690328 | 0.2947 ± 0.0026 | 0.3068 ± 0.0025 |
| flag-adversarial-data-augmentation-for-graph-1 | No | 2017028 | 0.2483 ± 0.0037 | 0.2556 ± 0.0040 |
| Modèle 8 | No | 10887085 | 0.3204 ± 0.0001 | 0.3353 ± 0.0002 |
| from-stars-to-subgraphs-uplifting-any-gnn | No | 3081029 | 0.2930 ± 0.0044 | 0.3047 ± 0.0007 |
| triplet-interaction-improves-graph | Yes | 47000000 | 0.3167 ± 0.0031 | - |
| nested-graph-neural-networks | No | 44187480 | 0.3007 ± 0.0037 | 0.3059 ± 0.0056 |
| semi-supervised-classification-with-graph | No | 2017028 | 0.2424 ± 0.0034 | 0.2495 ± 0.0042 |
| nested-graph-neural-networks | - | - | 0.3007 ± 0.0037 | 0.3059 ± 0.0056 |
| do-transformers-really-perform-bad-for-graph | - | 119529664 | 0.3140 ± 0.0032 | 0.3227 ± 0.0024 |
| towards-better-graph-representation-learning | No | 3842048 | 0.3031 ± 0.0026 | 0.3115 ± 0.0020 |
| directional-graph-networks-1 | No | 6732696 | 0.2885 ± 0.0030 | 0.2970 ± 0.0021 |
| recipe-for-a-general-powerful-scalable-graph | No | 9744496 | 0.2907 | 0.3015 ± 0.0038 |
| do-transformers-really-perform-bad-for-graph | Yes | 119529664 | 0.3140 ± 0.0032 | 0.3227 ± 0.0024 |
| principal-neighbourhood-aggregation-for-graph | No | 6550839 | 0.2838 ± 0.0035 | 0.2926 ± 0.0026 |
| ran-gnns-breaking-the-capacity-limits-of | No | 5572026 | 0.2881 ± 0.0028 | 0.3035 ± 0.0047 |
| graph-convolutions-that-can-finally-model | No | 6147029 | 0.2917 ± 0.0015 | 0.3065 ± 0.0030 |
| nested-graph-neural-networks | - | - | 0.2832 ± 0.0041 | 0.2915 ± 0.0035 |
| Modèle 23 | No | 5560960 | 0.3012 ± 0.0039 | 0.3151 ± 0.0047 |
| flag-adversarial-data-augmentation-for-graph-1 | No | 1923433 | 0.2395 ± 0.0040 | 0.2451 ± 0.0042 |
| flag-adversarial-data-augmentation-for-graph-1 | No | 565928 | 0.2116 ± 0.0017 | 0.2150 ± 0.0022 |
| Modèle 26 | No | 29440000 | 0.2054 ± 0.0004 | 0.2226 ± 0.0002 |
| flag-adversarial-data-augmentation-for-graph-1 | No | 5550208 | 0.2842 ± 0.0043 | 0.2952 ± 0.0029 |
| edge-augmented-graph-transformers-global-self | - | - | 0.2961 ± 0.0024 | - |
| flag-adversarial-data-augmentation-for-graph-1 | No | 3374533 | 0.2834 ± 0.0038 | 0.2912 ± 0.0026 |
| deepergcn-all-you-need-to-train-deeper-gcns | No | 5550208 | 0.2781 ± 0.0038 | 0.2920 ± 0.0025 |
| Modèle 31 | No | 10887085 | 0.3204 ± 0.0001 | 0.3353 ± 0.0002 |
| Modèle 32 | Yes | 119529665 | 0.3167 ± 0.0034 | 0.3252 ± 0.0043 |
| convolutional-neural-networks-on-graphs-with | No | 1475003 | 0.2306 ± 0.0016 | 0.2372 ± 0.0018 |
| how-powerful-are-graph-neural-networks | No | 1923433 | 0.2266 ± 0.0028 | 0.2305 ± 0.0027 |
| next-level-message-passing-with-hierarchical | - | - | 0.3129±0.0020 | - |
| Modèle 36 | No | 5511680 | 0.2994 ± 0.0019 | 0.3094 ± 0.0023 |