Graph Classification On Proteins

Accuracy

Results

Performance results of various models on this benchmark

Model Name	Accuracy	Paper Title	Repository
R-GCN + PANDA	76	PANDA: Expanded Width-Aware Message Passing Beyond Rewiring
GIN	75.536±1.851	How Powerful are Graph Neural Networks?
Multigraph ChebNet	76.5%	Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules
ApproxRepSet	70.74%	Rep the Set: Neural Networks for Learning Set Representations
EigenGCN-3	76.60%	Graph Convolutional Networks with EigenPooling
Shortest-Path Kernel	76.4%	Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks
WEGL	76.5%	Wasserstein Embedding for Graph Learning
GIC	77.65%	Gaussian-Induced Convolution for Graphs	-
2-WL-GNN	76.5	A Novel Higher-order Weisfeiler-Lehman Graph Convolution
1-NMFPool	72.1%	A Non-Negative Factorization approach to node pooling in Graph Convolutional Neural Networks	-
SPI-GCN	72.06%	SPI-GCN: A Simple Permutation-Invariant Graph Convolutional Network	-
GraphGPS	77.143±1.494	Recipe for a General, Powerful, Scalable Graph Transformer
FGW sp	74.55%	Optimal Transport for structured data with application on graphs
GIN-0	76,2%	How Powerful are Graph Neural Networks?
GCAPS-CNN	76.40%	Graph Capsule Convolutional Neural Networks
DiffWire	74.91%	DiffWire: Inductive Graph Rewiring via the Lovász Bound
GANet	77.92%	Graph Representation Learning via Hard and Channel-Wise Attention Networks
GDL-g (SP)	74.86	Online Graph Dictionary Learning
GNN (DiffPool)	76.25%	Hierarchical Graph Representation Learning with Differentiable Pooling
PPGN	77.20%	Provably Powerful Graph Networks

0 of 99 row(s) selected.