Graph Classification On Proteins

Metriken

Accuracy

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

		Paper Title
HGP-SL	84.91	Hierarchical Graph Pooling with Structure Learning
rLap (unsupervised)	84.3	Randomized Schur Complement Views for Graph Contrastive Learning
TFGW ADJ (L=2)	82.9	Template based Graph Neural Network with Optimal Transport Distances
ESA (Edge set attention, no positional encodings)	82.679±0.799	An end-to-end attention-based approach for learning on graphs
DUGNN	81.70%	Learning Universal Graph Neural Network Embeddings With Aid Of Transfer Learning
MEWISPool	80.71%	Maximum Entropy Weighted Independent Set Pooling for Graph Neural Networks
CIN++	80.5	CIN++: Enhancing Topological Message Passing
SAEPool	80.36%	Relational Reasoning Over Spatial-Temporal Graphs for Video Summarization
UGT	80.12 ±0.32	Transitivity-Preserving Graph Representation Learning for Bridging Local Connectivity and Role-based Similarity
U2GNN (Unsupervised)	80.01%	Universal Graph Transformer Self-Attention Networks
sGIN	78.97%	Mutual Information Maximization in Graph Neural Networks
QS-CNNs (Quantum Walk)	78.80%	Quantum-based subgraph convolutional neural networks
WKPI-kmeans	78.8%	Learning metrics for persistence-based summaries and applications for graph classification
PIN	78.8%	Weisfeiler and Lehman Go Paths: Learning Topological Features via Path Complexes
hGANet	78.65%	Graph Representation Learning via Hard and Channel-Wise Attention Networks
U2GNN	78.53%	Universal Graph Transformer Self-Attention Networks
DS-CNNs (Random Walk)	78.35%	Quantum-based subgraph convolutional neural networks
cGANet	78.23%	Graph Representation Learning via Hard and Channel-Wise Attention Networks
CAN	78.2%	Cell Attention Networks
GANet	77.92%	Graph Representation Learning via Hard and Channel-Wise Attention Networks

0 of 99 row(s) selected.

Command Palette

Graph Classification On Proteins

Metriken

Ergebnisse