DFT energy prediction on ANI-Al
0.0816MAE (Config, eV)SchNet
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SchNetLabel Augmentation=true, Multi-task Pretraining=true2022.10 | 0.0816 | 0.0006 | 53.27 | |
| SchNetLabel Augmentation=true, Multi-task Pretraining=false2022.10 | 0.0845 | 0.0007 | 51.24 | |
| SchNetLabel Augmentation=false, Multi-task Pretraining=true2022.10 | 0.1296 | 0.001 | 26 | |
| CGCNNLabel Augmentation=true, Multi-task Pretraining=true2022.10 | 0.1392 | 0.0011 | 41.65 | |
| SchNetLabel Augmentation=false, Multi-task Pretraining=false2022.10 | 0.1693 | 0.0014 | — | |
| SOAPNetLabel Augmentation=true, Multi-task Pretraining=true2022.10 | 0.1697 | 0.0013 | 22.13 | |
| SOAPNetLabel Augmentation=false, Multi-task Pretraining=true2022.10 | 0.1744 | 0.0014 | 19.12 | |
| SOAPNetLabel Augmentation=true, Multi-task Pretraining=false2022.10 | 0.1786 | 0.0014 | 18.14 | |
| CGCNNLabel Augmentation=true, Multi-task Pretraining=false2022.10 | 0.1786 | 0.0014 | 25.44 | |
| SOAPNetLabel Augmentation=false, Multi-task Pretraining=false2022.10 | 0.2153 | 0.0017 | — | |
| CGCNNLabel Augmentation=false, Multi-task Pretraining=true2022.10 | 0.2206 | 0.0017 | 7.8 | |
| CGCNNLabel Augmentation=false, Multi-task Pretraining=false2022.10 | 0.241 | 0.0019 | — | |
| Best EIPLabel Augmentation=false, Multi-task Pretraining=false2022.10 | 46.4869 | 0.3561 | — |