Model for n-Seebeck¶
- Description: This is a benchmark to evaluate how accurately an AI model can predict the n-doped Seebeck Coefficient 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 accuracy.
Reference(s): https://www.nature.com/articles/s41524-023-01012-9;https://hackingmaterials.lbl.gov/matminer, https://www.nature.com/articles/s41524-020-00440-1, https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.120.145301, https://doi.org/10.48550/arXiv.2305.11842, https://github.com/aimat-lab/gcnn_keras, https://www.nature.com/articles/s41524-021-00650-1
Model benchmarks
| Model name | Dataset | MAE | Team name | Dataset size | Date submitted | Notes |
|---|---|---|---|---|---|---|
| kgcnn_coGN | dft_3d | 39.2692 | kgcnn | 23210 | 05-06-2023 | CSV, JSON, run.sh, Info |
| alignn_model | dft_3d | 40.9214 | ALIGNN | 23210 | 01-14-2023 | CSV, JSON, run.sh, Info |
| matminer_xgboost | dft_3d | 44.2229 | UofT | 23210 | 05-22-2023 | CSV, JSON, run.sh, Info |
| kgcnn_cgcnn | dft_3d | 45.6596 | kgcnn | 23210 | 09-26-2023 | CSV, JSON, run.sh, Info |
| cgcnn_model | dft_3d | 49.3172 | CGCNN | 23210 | 01-14-2023 | CSV, JSON, run.sh, Info |
| kgcnn_schnet | dft_3d | 47.244 | kgcnn | 23210 | 09-26-2023 | CSV, JSON, run.sh, Info |
| kgcnn_coNGN | dft_3d | 40.0977 | kgcnn | 23210 | 05-06-2023 | CSV, JSON, run.sh, Info |
| kgcnn_megnet | dft_3d | 47.2813 | kgcnn | 23210 | 05-06-2023 | CSV, JSON, run.sh, Info |
| matminer_rf | dft_3d | 43.9839 | UofT | 23210 | 05-22-2023 | CSV, JSON, run.sh, Info |
| kgcnn_dimenetPP | dft_3d | 54.2759 | kgcnn | 23210 | 05-06-2023 | CSV, JSON, run.sh, Info |