model_name: SevenNet-MF-ompa model_key: sevennet-mf-ompa model_version: v0.11.0 # 2024-07-11 date_added: '2025-03-13' date_published: '2025-03-13' authors: - name: Jaesun Kim affiliation: Seoul National University orcid: https://orcid.org/0009-0000-6646-1318 - name: Jisu Kim affiliation: Seoul National University orcid: https://orcid.org/0009-0000-7380-6089 - name: Jaehoon Kim affiliation: Seoul National University orcid: https://orcid.org/0009-0006-3182-9411 - name: Jiho lee affiliation: Seoul National University orcid: https://orcid.org/0009-0008-7266-091X - name: Yutack Park affiliation: Seoul National University email: parkyutack@snu.ac.kr orcid: https://orcid.org/0009-0008-8690-935X - name: Youngho Kang affiliation: Incheon National University orcid: https://orcid.org/0000-0003-4532-0027 corresponding: true - name: Seungwu Han affiliation: Seoul National University, Korea Institute for Advanced Study email: hansw@snu.ac.kr orcid: https://orcid.org/0000-0003-3958-0922 corresponding: true trained_by: - name: Jaesun Kim affiliation: Seoul National University orcid: https://orcid.org/0009-0000-6646-1318 repo: https://github.com/MDIL-SNU/SevenNet url: https://figshare.com/files/52975859 doi: https://doi.org/10.1021/jacs.4c14455 paper: https://arxiv.org/abs/2409.07947 pypi: https://pypi.org/project/sevenn pr_url: https://github.com/janosh/matbench-discovery/pull/221 checkpoint_url: https://figshare.com/articles/software/28590722?file=53029859 license: code: GPL-3.0 code_url: https://github.com/MDIL-SNU/SevenNet/blob/8ce2c9d4/LICENSE checkpoint: GPL-3.0 checkpoint_url: https://figshare.com/articles/software/28590722?file=53029859 openness: OSOD trained_for_benchmark: false train_task: S2EFS test_task: IS2RE-SR targets: EFS_G model_type: UIP model_params: 25_734_966 n_estimators: 1 status: superseded superseded_by: sevennet-omni-i12 hyperparams: max_force: 0.02 max_steps: 800 ase_optimizer: FIRE cell_filter: FrechetCellFilter optimizer: Adamw loss: MAE/L2MAE/L2MAE loss_weights: energy: 1.0 force: 0.1 stress: 0.0001 batch_size: 512 initial_learning_rate: 0.0002 learning_rate_schedule: LinearLR - start_factor=1.0, total_iters=600, end_factor=0.0001 epochs: 2 n_layers: 5 n_features: - 128x0e - 128x0e+64x1o+32x2e+32x3o - 128x0e+64x1o+64x1e+32x2o+32x2e+32x3o+32x3e - 128x0o+128x0e+64x1o+64x1e+32x2o+32x2e+32x3o+32x3e - 128x0e+64x1o+32x2e+32x3o - 128x0e n_radial_bessel_basis: 8 graph_construction_radius: 6.0 # Å, from SevenNet-mf-ompa/hyperparams.yaml max_neighbors: .inf sph_harmonics_l_max: 3 requirements: torch: 2.2.1 torch-geometric: 2.5.2 ase: 3.22.1 pymatgen: 2024.6.10 numpy: 1.26.4 training_set: [OMat24, sAlex, MPtrj] training_cost: missing notes: Description: | SevenNet is a graph neural network interatomic potential package that supports parallel molecular dynamics simulations. The SevenNet-MF-ompa model used Multi-fidelity learning to simultaneously train on the OMat24, MPtrj, and sAlex datasets, achieving high accuracy despite differences in their DFT calculation settings. metrics: phonons: kappa_103: κ_SRME: 0.3171 pred_file: models/sevennet/sevennet-mf-ompa/2025-03-11-kappa-103-FIRE-dist=0.03-fmax=1e-4-symprec=1e-5.json.gz pred_file_url: https://figshare.com/files/53090603 κ_SRE: 0.21 geo_opt: pred_file: models/sevennet/sevennet-mf-ompa/2025-03-11-wbm-geo-opt-FIRE.jsonl.gz pred_file_url: https://figshare.com/files/52983491 struct_col: sevennet_structure symprec=1e-2: rmsd: 0.0639 # unitless n_sym_ops_mae: 1.7053 # unitless symmetry_decrease: 0.0467 # fraction symmetry_match: 0.8181 # fraction symmetry_increase: 0.128 # fraction n_structures: 256963 # count analysis_file: models/sevennet/sevennet-mf-ompa/2025-03-11-wbm-geo-opt-FIRE-symprec=1e-2-moyo=0.4.2.csv.gz analysis_file_url: https://figshare.com/files/53504708 symprec=1e-5: rmsd: 0.0639 # unitless n_sym_ops_mae: 2.0326 # unitless symmetry_decrease: 0.0439 # fraction symmetry_match: 0.7057 # fraction symmetry_increase: 0.2453 # fraction n_structures: 256963 # count analysis_file: models/sevennet/sevennet-mf-ompa/2025-03-11-wbm-geo-opt-FIRE-symprec=1e-5-moyo=0.4.2.csv.gz analysis_file_url: https://figshare.com/files/53504717 discovery: pred_file: models/sevennet/sevennet-mf-ompa/2025-03-11-wbm-IS2RE.csv.gz pred_file_url: https://figshare.com/files/52983488 pred_col: e_form_per_atom_sevennet unique_prototypes: F1: 0.901 # fraction DAF: 5.825 # dimensionless Precision: 0.89 # fraction Recall: 0.911 # fraction Accuracy: 0.969 # fraction TPR: 0.911 # fraction FPR: 0.021 # fraction TNR: 0.979 # fraction FNR: 0.089 # fraction TP: 30401.0 # count FP: 3739.0 # count TN: 178375.0 # count FN: 2973.0 # count MAE: 0.021 # eV/atom RMSE: 0.067 # eV/atom R2: 0.867 # dimensionless missing_preds: 0 # count full_test_set: F1: 0.884 # fraction DAF: 5.082 # dimensionless Precision: 0.872 # fraction Recall: 0.895 # fraction Accuracy: 0.96 # fraction TPR: 0.895 # fraction FPR: 0.027 # fraction TNR: 0.973 # fraction FNR: 0.105 # fraction TP: 39484.0 # count FP: 5799.0 # count TN: 207072.0 # count FN: 4608.0 # count MAE: 0.021 # eV/atom RMSE: 0.067 # eV/atom R2: 0.861 # dimensionless missing_preds: 2 # count most_stable_10k: F1: 0.985 # fraction DAF: 6.346 # dimensionless Precision: 0.97 # fraction Recall: 1.0 # fraction Accuracy: 0.97 # fraction TPR: 1.0 # fraction FPR: 1.0 # fraction TNR: 0.0 # fraction FNR: 0.0 # fraction TP: 9701.0 # count FP: 299.0 # count TN: 0.0 # count FN: 0.0 # count MAE: 0.019 # eV/atom RMSE: 0.071 # eV/atom R2: 0.888 # dimensionless missing_preds: 0 # count