Graph Classification On Collab

Metrics

Accuracy

Results

Performance results of various models on this benchmark

Model Name	Accuracy	Paper Title	Repository
GFN-light	81.34%	Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification	-
GMT	80.74%	Accurate Learning of Graph Representations with Graph Multiset Pooling	-
G_DenseNet	83.16%	When Work Matters: Transforming Classical Network Structures to Graph CNN	-
DGCNN	68.34%	DGCNN: Disordered Graph Convolutional Neural Network Based on the Gaussian Mixture Model	-
DGCNN	73.76%	An End-to-End Deep Learning Architecture for Graph Classification
sGIN	80.71%	Mutual Information Maximization in Graph Neural Networks	-
PPGN	81.38%	Provably Powerful Graph Networks	-
GCN	80.6%	Fast Graph Representation Learning with PyTorch Geometric	-
GraphSAGE	73.9%	A Fair Comparison of Graph Neural Networks for Graph Classification	-
R-GIN + PANDA	77.8%	PANDA: Expanded Width-Aware Message Passing Beyond Rewiring	-
R-GCN + PANDA	71.4%	PANDA: Expanded Width-Aware Message Passing Beyond Rewiring	-
hGANet	77.48%	Graph Representation Learning via Hard and Channel-Wise Attention Networks	-
GCN + PANDA	68.4%	PANDA: Expanded Width-Aware Message Passing Beyond Rewiring	-
Graph U-Nets	77.56%	Graph U-Nets	-
1-NMFPool	65.0%	A Non-Negative Factorization approach to node pooling in Graph Convolutional Neural Networks	-
CT-Layer	69.87%	DiffWire: Inductive Graph Rewiring via the Lovász Bound	-
U2GNN (Unsupervised)	95.62%	Universal Graph Transformer Self-Attention Networks	-
U2GNN	77.84%	Universal Graph Transformer Self-Attention Networks	-
FactorGCN	81.2%	Factorizable Graph Convolutional Networks	-
NDP	79.1%	Hierarchical Representation Learning in Graph Neural Networks with Node Decimation Pooling	-

0 of 38 row(s) selected.