Model for e_form¶
- Description: This is a benchmark to evaluate how accurately an AI model can predict the formation energy using the M3GNET dataset (PBE functional). 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 (PBE) accuracy. External links: https://github.com/materialsvirtuallab/m3gnet
Reference(s): https://doi.org/10.1063/1.4812323, https://www.nature.com/articles/s41524-021-00650-1, https://www.nature.com/articles/s41524-023-01012-9;https://hackingmaterials.lbl.gov/matminer
Model benchmarks
| Model name | Dataset | MAE | Team name | Dataset size | Date submitted | Notes |
|---|---|---|---|---|---|---|
| alignn_model | megnet | 0.0221 | ALIGNN | 69239 | 01-14-2023 | CSV, JSON, run.sh, Info |
| matminer_rf | megnet | 0.0658 | UofT | 69239 | 05-22-2023 | CSV, JSON, run.sh, Info |
| matminer_xgboost | megnet | 0.0508 | UofT | 69239 | 05-22-2023 | CSV, JSON, run.sh, Info |