Strategies for Pre-training Graph Neural Networks

Benchmark Model Rank Results
drug-discovery-on-baceContextPred#5AUC: 0.845
drug-discovery-on-bbbpContextPred#4AUC: 0.687
drug-discovery-on-clintoxContextPred#4AUC: 0.726
drug-discovery-on-hiv-datasetContextPred#4AUC: 0.799
drug-discovery-on-muvContextPred#4AUC: 0.813
drug-discovery-on-siderContextPred#4AUC: 0.627
drug-discovery-on-tox21ContextPred#7AUC: 0.781
drug-discovery-on-toxcastContextPred#5AUC: 0.657
molecular-property-prediction-onPretrainGNN#5RMSE: 0.739
molecular-property-prediction-on-bace-1PretrainGNN#4ROC-AUC: 84.5
molecular-property-prediction-on-bbbp-1PretrainGNN#22ROC-AUC: 68.7
molecular-property-prediction-on-clintox-1PretrainGNN#15ROC-AUC: 72.6
molecular-property-prediction-on-freesolvPretrainGNN#20RMSE: 2.764
molecular-property-prediction-on-qm7PretrainGNN#7MAE: 113.2
molecular-property-prediction-on-qm8PretrainGNN#3MAE: 0.0200
molecular-property-prediction-on-qm9PretrainGNN#3MAE: 0.00922
molecular-property-prediction-on-sider-1PretrainGNN#12ROC-AUC: 62.7
molecular-property-prediction-on-tox21-1PretrainGNN#6ROC-AUC: 78.1
molecular-property-prediction-on-toxcast-1PretrainGNN#3ROC-AUC: 65.7