Model for Si FF energy¶
- Description: This is an AI benchmark to evaluate how accurately a machine learning force-field (MLFF) can predict the total energy of Si using the relaxation trajectories (energy and forces of intermediate steps) of the mlearn dataset, calculated with the PBE density functional. The dataset contains different types of chemical formula and atomic structures. Here we use mean absolute error (MAE) to compare MLFFs with respect to DFT (PBE) accuracy. External links: https://github.com/materialsvirtuallab/mlearn
Reference(s): https://github.com/materialsvirtuallab/m3gnet, https://github.com/materialsvirtuallab/maml, https://www.nature.com/articles/s41467-023-36329-y, https://pubs.rsc.org/en/content/articlehtml/2023/dd/d2dd00096b, https://doi.org/10.1021/acs.jpca.9b08723, https://github.com/CederGroupHub/chgnet, https://www.nature.com/articles/s43588-022-00349-3
Model benchmarks
| Model name | Dataset | Accuracy | Team name | Dataset size | Date submitted | Notes |
|---|---|---|---|---|---|---|
| snap_mlearn | mlearn_Si | 1.1141 | JARVIS | 239 | 01-14-2023 | CSV, JSON, run.sh, Info |
| matgl_pretrained | mlearn_Si | 5.8323 | Matgl | 239 | 01-14-2023 | CSV, JSON, run.sh, Info |
| nequip_mlearn | mlearn_Si | 83.841 | MIR | 239 | 07-22-2023 | CSV, JSON, run.sh, Info |
| chgnet_pretrained | mlearn_Si | 3.5833 | CHGNET | 239 | 08-07-2023 | CSV, JSON, run.sh, Info |
| allegro_mlearn | mlearn_Si | 37.8701 | MIR | 239 | 07-22-2023 | CSV, JSON, run.sh, Info |
| alignnff_mlearn_all_wt1 | mlearn_Si | 3.003 | JARVIS | 239 | 01-14-2023 | CSV, JSON, run.sh, Info |
| chgnet_mlearn | mlearn_Si | 1.7295 | CHGNET | 239 | 01-29-2024 | CSV, JSON, run.sh, Info |
| alignnff_mlearn_wt1 | mlearn_Si | 1.1766 | JARVIS | 239 | 01-14-2023 | CSV, JSON, run.sh, Info |
| matgl_mlearn | mlearn_Si | 0.8759 | Matgl | 239 | 01-14-2023 | CSV, JSON, run.sh, Info |
| m3gnet_pretrained | mlearn_Si | 6.9913 | M3GNET | 239 | 01-14-2023 | CSV, JSON, run.sh, Info |