Graph Regression On Kit

评估指标

R2
RMSE

评测结果

各个模型在此基准测试上的表现结果

Paper TitleRepository
PNA0.843±0.0000.430±0.843Principal Neighbourhood Aggregation for Graph Nets
ESA (Edge set attention, no positional encodings)0.841±0.0000.433±0.841An end-to-end attention-based approach for learning on graphs
GINDrop0.835±0.0000.441±0.835DropGNN: Random Dropouts Increase the Expressiveness of Graph Neural Networks
GAT0.833±0.0000.443±0.833Graph Attention Networks
GIN0.833±0.0000.444±0.833How Powerful are Graph Neural Networks?
GATv20.826±0.0000.453±0.826How Attentive are Graph Attention Networks?
GCN0.814±0.0000.469±0.814Semi-Supervised Classification with Graph Convolutional Networks
TokenGT0.800±0.0000.486±0.800Pure Transformers are Powerful Graph Learners
GraphormerOOMOOMDo Transformers Really Perform Bad for Graph Representation?
0 of 9 row(s) selected.
Graph Regression On Kit | SOTA | HyperAI超神经