Topology-Informed Graph Transformer

Benchmark Model Rank Results
graph-classification-on-cifar10-100kTIGT#8Accuracy (%): 73.955
graph-classification-on-mnistTIGT#8Accuracy: 98.230±0.133
graph-classification-on-peptides-funcTIGT#25AP: 0.6679
graph-regression-on-pcqm4mv2-lscTIGT#7Validation MAE: 0.0826
graph-regression-on-peptides-structTIGT#18MAE: 0.2485
graph-regression-on-zincTIGT#3MAE: 0.057
graph-regression-on-zinc-fullTIGT#3Test MAE: 0.014
node-classification-on-clusterTIGT#8Accuracy: 78.033
node-classification-on-patternTIGT#9Accuracy: 86.680