Download src/jd/__init__.py from hembad/reranker: direct link, hf CLI and curl.
- Browser
- Download file 647 Bytes
-
https://huggingface.co/spaces/hembad/reranker/resolve/main/src/jd/__init__.py
- Command line
-
hf download hf://spaces/hembad/reranker/src/jd/__init__.py
-
curl -L -o __init__.py https://huggingface.co/spaces/hembad/reranker/resolve/main/src/jd/__init__.py
647 Bytes
| """JD-parse phase — load JD → objective (keywords) + subjective → ONE combined jd_extraction.json (live, Qwen3). | |
| ``run_jd_parse`` writes the single JSON each run (objective keyword fields + ``subjective.ideal_summary``); | |
| ``extract_spec`` / ``read_subjective`` consume it. Live-only — when the JSON is absent, ``extract_spec`` | |
| returns an empty default spec (no committed fallback). Each part is independently testable. | |
| """ | |
| from .pipeline import run_jd_parse | |
| from .spec import JDSpec, embed_query, extract_spec, read_subjective | |
| __all__ = [ | |
| "JDSpec", | |
| "embed_query", | |
| "extract_spec", | |
| "read_subjective", | |
| "run_jd_parse", | |
| ] | |