Download src/casmd/predict/model.py from zzhaobz/HsingMD: direct link, hf CLI and curl.
- Browser
- Download file 722 Bytes
-
https://huggingface.co/spaces/zzhaobz/HsingMD/resolve/main/src/casmd/predict/model.py
- Command line
-
hf download hf://spaces/zzhaobz/HsingMD/src/casmd/predict/model.py
-
curl -L -o model.py https://huggingface.co/spaces/zzhaobz/HsingMD/resolve/main/src/casmd/predict/model.py
722 Bytes
| """Unified data model for prediction outputs across backends.""" | |
| from __future__ import annotations | |
| from dataclasses import dataclass | |
| from pathlib import Path | |
| class PredictionModel: | |
| """One predicted structure with its confidence metrics. | |
| iptm, ptm, plddt_mean are `None` when the backend does not report them. | |
| """ | |
| pdb_path: Path | |
| backend: str # 'boltz-2' | 'protenix' | 'af3' | |
| model_id: int | |
| seed: int | None | |
| iptm: float | None | |
| ptm: float | None | |
| plddt_mean: float | None | |
| class PredictionBundle: | |
| """All models from one prediction submission.""" | |
| backend: str | |
| raw_dir: Path | |
| models: tuple[PredictionModel, ...] | |