$P^2$GNN: Two Prototype Sets to boost GNN Performance

Benchmark Model Rank Results
node-classification-on-chameleon-60-20-20-random-splitsP2GNN–1:1 Accuracy: 69.87
node-classification-on-citeseer-60-20-20-random-splitsP2GNN–1:1 Accuracy: 82.37
node-classification-on-cora-60-20-20-random-splitsP2GNN–1:1 Accuracy: 89.89
node-classification-on-cornell-60-20-20-random-splitsP2GNN–1:1 Accuracy: 95.41
node-classification-on-film-60-20-20-random-splitsP2GNN–1:1 Accuracy: 43.05
node-classification-on-geniusP2GNN–Accuracy: 91.01
node-classification-on-non-homophilic-heterophilic-graphs-on-chameleon-60-20-20-random-splitsP2GNN–1:1 Accuracy: 69.87
node-classification-on-non-homophilic-heterophilic-graphs-on-cornell-60-20-20-random-splitsP2GNN–1:1 Accuracy: 95.41
node-classification-on-non-homophilic-heterophilic-graphs-on-geniusP2GNN–1:1 Accuracy: 91.01
node-classification-on-non-homophilic-heterophilic-graphs-on-penn94P2GNN–1:1 Accuracy: 86.40
node-classification-on-non-homophilic-heterophilic-graphs-on-texas-60-20-20-random-splitsP2GNN–1:1 Accuracy: 96.72
node-classification-on-non-homophilic-heterophilic-graphs-on-twitch-gamersP2GNN–1:1 Accuracy: 66.55
node-classification-on-non-homophilic-heterophilic-graphs-on-wisconsin-60-20-20-random-splitsP2GNN–1:1 Accuracy: 98.00
node-classification-on-penn94P2GNN–Accuracy: 86.40
node-classification-on-squirrel-60-20-20-random-splitsP2GNN–1:1 Accuracy: 56.47
node-classification-on-texas-60-20-20-random-splitsP2GNN–1:1 Accuracy: 96.72
node-classification-on-wisconsin-60-20-20-random-splitsP2GNN–1:1 Accuracy: 98.00