Node Classification On Cora With Public Split
المقاييس
Accuracy
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
جدول المقارنة
| اسم النموذج | Accuracy |
|---|---|
| distilling-self-knowledge-from-contrastive | 81.19% |
| lanczosnet-multi-scale-deep-graph | 80.4 ± 1.1 |
| re-think-and-re-design-graph-neural-networks | 86.3% |
| convolutional-neural-networks-on-graphs-with | 78.0% |
| graph-random-neural-network | 85.4 ± 0.4 |
| gated-graph-sequence-neural-networks | 77.6% |
| graph-attention-networks | 83.0 ± 0.7% |
| from-cluster-assumption-to-graph-convolution | 86.9% |
| break-the-ceiling-stronger-multi-scale-deep | 83.26% |
| graph-entropy-minimization-for-semi | 83.05% |
| extract-the-knowledge-of-graph-neural | 85.3% |
| pre-train-and-learn-preserve-global | 84.31% |
| the-split-matters-flat-minima-methods-for | 83.26 ± 0.69% |
| distilling-self-knowledge-from-contrastive | 80.79% |
| diffusion-convolutional-neural-networks | 79.7% |
| graphair-graph-representation-learning-with | 84.7% |
| how-to-find-your-friendly-neighborhood-graph-1 | 84.3% |
| classic-gnns-are-strong-baselines-reassessing | 85.1 ± 0.7 |
| from-cluster-assumption-to-graph-convolution | 83.6% |
| distilling-self-knowledge-from-contrastive | 81.39% |
| convolutional-networks-on-graphs-for-learning | 74.6% |
| inductive-representation-learning-on-large | 74.5% |
| break-the-ceiling-stronger-multi-scale-deep | 83.19% |
| distilling-self-knowledge-from-contrastive | 81.05% |
| bridging-the-gap-between-spectral-and-spatial | 84.2% |
| semi-supervised-node-classification-via | 84.5% |
| learning-discrete-structures-for-graph-neural | 84.1% |
| break-the-ceiling-stronger-multi-scale-deep | 83.16% |
| simple-and-deep-graph-convolutional-networks-1 | 85.5% |
| a-flexible-generative-framework-for-graph | 82.9% |
| optimization-of-graph-neural-networks-with | 82.84 ± 0.87% |
| towards-deeper-graph-neural-networks | 84.4 ± 0.5 |
| graphmix-regularized-training-of-graph-neural | 83.94 ± 0.57 |
| lanczosnet-multi-scale-deep-graph | 79.5 ± 1.8 |
| graphmix-regularized-training-of-graph-neural | 83.94 ± 0.57 |
| data-augmentation-for-graph-neural-networks | 83.6 ± 0.5% |