Model for mepsy¶
- Description: This is a benchmark to evaluate how accurately an AI model can predict the static dielectric constant (y-direction) at the meta-GGA TBmBJ level of theory using the JARVIS-DFT (dft_3d) dataset. The dataset contains different types of chemical formula and atomic structures. Here we use mean absolute error (MAE) to compare models with respect to DFT (meta-GGA TBmBJ) accuracy.
Reference(s): https://www.nature.com/articles/s41524-020-00440-1, https://doi.org/10.1103/PhysRevMaterials.2.083801, https://www.nature.com/articles/s41524-023-01012-9;https://hackingmaterials.lbl.gov/matminer, https://doi.org/10.48550/arXiv.2305.11842, https://www.nature.com/articles/s41524-021-00650-1, https://hackingmaterials.lbl.gov/matminer/, https://github.com/aimat-lab/gcnn_keras
Model benchmarks
| Model name | Dataset | MAE | Team name | Dataset size | Date submitted | Notes |
|---|---|---|---|---|---|---|
| kgcnn_coNGN | dft_3d | 23.3299 | kgcnn | 16809 | 05-06-2023 | CSV, JSON, run.sh, Info |
| kgcnn_cgcnn | dft_3d | 26.5558 | kgcnn | 16809 | 09-26-2023 | CSV, JSON, run.sh, Info |
| matminer_rf | dft_3d | 25.5996 | UofT | 16809 | 05-22-2023 | CSV, JSON, run.sh, Info |
| kgcnn_dimenetPP | dft_3d | 31.0215 | kgcnn | 16809 | 05-06-2023 | CSV, JSON, run.sh, Info |
| cfid | dft_3d | 30.0578 | JARVIS | 16809 | 01-14-2023 | CSV, JSON, run.sh, Info |
| alignn_model | dft_3d | 23.6482 | ALIGNN | 16809 | 01-14-2023 | CSV, JSON, run.sh, Info |
| matminer_lgbm | dft_3d | 27.5274 | Matminer | 16809 | 01-14-2023 | CSV, JSON, run.sh, Info |
| kgcnn_megnet | dft_3d | 25.9523 | kgcnn | 16809 | 05-06-2023 | CSV, JSON, run.sh, Info |
| matminer_xgboost | dft_3d | 25.0706 | UofT | 16809 | 05-22-2023 | CSV, JSON, run.sh, Info |
| kgcnn_coGN | dft_3d | 24.1891 | kgcnn | 16809 | 05-06-2023 | CSV, JSON, run.sh, Info |
| cfid_chem | dft_3d | 31.0889 | JARVIS | 16809 | 01-14-2023 | CSV, JSON, run.sh, Info |
| kgcnn_schnet | dft_3d | 25.5455 | kgcnn | 16809 | 09-26-2023 | CSV, JSON, run.sh, Info |