Node Classification on Cora (3%)

15개 결과 · ⬇ CSV · JSON

Accuracy

Rank	Model	Accuracy	Paper	Code	Year
1	CPF-tra-GCNII	84.18%	Extract the Knowledge of Graph Neural Networks and Go Beyond it: An Effective Knowledge Distillation Framework	BUPT-GAMMA/CPF	2021
2	VCHN	83.1%	View-Consistent Heterogeneous Network on Graphs With Few Labeled Nodes	kunzhan/VCHN	2022
3	Truncated Krylov	81.92%	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks	PwnerHarry/Stronger_GCN	2019
4	Snowball (linear)	80.96%	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks	PwnerHarry/Stronger_GCN	2019
5	Snowball (tanh)	80.72%	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks	PwnerHarry/Stronger_GCN	2019
6	Snowball (linear + tanh)	79.52%	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks	PwnerHarry/Stronger_GCN	2019
7	MT-GCN	78.5%	Mutual Teaching for Graph Convolutional Networks	kunzhan/MTGCN	2020
8	AdaLanczosNet	77.7 ± 2.4	LanczosNet: Multi-Scale Deep Graph Convolutional Networks	lrjconan/LanczosNetwork	2019
9	DCNN	76.7%	Diffusion-Convolutional Neural Networks	jcatw/dcnn · RicardoZiTseng/dcnn-tensorflow · MindSpore-paper-code-3/code6	2015
10	LanczosNet	76.3 ± 2.3	LanczosNet: Multi-Scale Deep Graph Convolutional Networks	lrjconan/LanczosNetwork	2019
11	GGNN	73.1%	Gated Graph Sequence Neural Networks	dmlc/dgl · Microsoft/gated-graph-neural-network-samples · chingyaoc/ggnn.pytorch · +10	2015
12	GCN-FP	71.7%	Convolutional Networks on Graphs for Learning Molecular Fingerprints	HIPS/neural-fingerprint · debbiemarkslab/neural-fingerprint-theano · nrel/m2p · +5	2015
13	GraphSAGE	64.2%	Inductive Representation Learning on Large Graphs	pyg-team/pytorch_geometric · dmlc/dgl · dmlc/dgl · +17	2017
14	ChebyNet	62.1%	Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering	mdeff/cnn_graph · hazdzz/ChebyNet · selmiss/gp-tlstgcn · +2	2016
15	GAT	56.8%	Graph Attention Networks	labmlai/annotated_deep_learning_paper_implementations · dmlc/dgl · dmlc/dgl · +90	2017