Pre-trained Models¶
ALIGNN ships a large catalog of pre-trained checkpoints for direct use.
- Models are hosted on Figshare.
- Two CLIs load them:
pretrained.py(single-property predictors) andrun_alignn_ff.py(universal force-field). - Model names and sources are listed in
alignn/ff/all_models_alignn.jsonandalignn/ff/all_models_alignn_atomwise.json.
| Page | What it covers |
|---|---|
| Property predictors | Per-property checkpoints (formation energy, bandgap, …) |
| ALIGNN-FF | Universal force-field: EV curve, optimization, phonons, MD |
| ASE calculator | Drive ALIGNN-FF from Python via ASE |
See Performance for per-dataset accuracy numbers.