Graph Transformers without Positional Encodings

Benchmark Model Rank Results
graph-classification-on-cifar10-100kEIGENFORMERAccuracy (%): 70.194
graph-classification-on-mnistEIGENFORMERAccuracy: 98.362
graph-classification-on-peptides-funcEIGENFORMERAP: 0.6414
graph-regression-on-peptides-structEIGENFORMERMAE: 0.2599
graph-regression-on-zincEIGENFORMERMAE: 0.077
node-classification-on-clusterEIGENFORMERAccuracy: 77.456
node-classification-on-patternEIGENFORMERAccuracy: 86.738