Regression on MoleculeNet Clearance
42.841RMSESPMM
Evaluation Results
| Method | Links | |
|---|---|---|
| SPMMPre-training=true, Fine-tuned=true2022.11 | 42.841 | |
| MolFormerFine-tuned=true2022.11 | 43.175 | |
| ChemBERTa-2Pre-training strategy=MTR, Pre-training size=77M2026.05 | 48.515 | |
| ChemBERTa-2Pre-training strategy=MTR, Pre-training size=10M2026.05 | 48.934 | |
| Chem-GMNetPre-training strategy=scratch, k=82026.05 | 49.36 | |
| D-MPNNFine-tuned=true2022.11 | 49.754 | |
| D-MPNN (Chemprop)Pre-training strategy=None2026.05 | 49.754 | |
| ChemBERTa-2Pre-training strategy=MTR, Pre-training size=5M2026.05 | 50.154 | |
| Chem-GMNetPre-training strategy=MLM, Pre-training size=10M2026.05 | 51.11 | |
| GCNPre-training strategy=None2026.05 | 51.227 | |
| N-GramRFFine-tuned=true2022.11 | 52.077 | |
| Random ForestPre-training strategy=None2026.05 | 52.077 | |
| ChemBERTa-2Pre-training strategy=MLM, Pre-training size=77M2026.05 | 52.754 | |
| SPMM (w/o pre-train)Pre-training=false, Fine-tuned=true2022.11 | 53.544 | |
| ChemBERTa-2Pre-training strategy=MLM, Pre-training size=10M2026.05 | 53.859 | |
| ChemBERTa-2Pre-training strategy=MLM, Pre-training size=5M2026.05 | 54.601 |