Implement confidence scoring #17

Open
opened 2026-04-12 16:49:45 -04:00 by aalexmmaldonado · 0 comments
aalexmmaldonado commented 2026-04-12 16:49:45 -04:00 (Migrated from github.com)

Once the design from Issue #16 is agreed on, this issue builds it.

Scope (tentative — will be refined after Issue 14)

Create pcdigitizer/quality/confidence.py with a function like score_dataframe(df: pl.DataFrame, annotation: Annotation) -> pl.DataFrame that adds confidence columns. Create pcdigitizer/quality/sources.yaml with source-tier mappings for known depositors. Wire the scoring into GetAnnotationPage output. Test with mock data.

Definition of done

Confidence information present in the output DataFrame. Source tier mappings cover at least the top 10 depositors by volume. Tests verify the scoring logic.

Once the design from Issue #16 is agreed on, this issue builds it. ### Scope (tentative — will be refined after Issue 14) Create `pcdigitizer/quality/confidence.py` with a function like `score_dataframe(df: pl.DataFrame, annotation: Annotation) -> pl.DataFrame` that adds confidence columns. Create `pcdigitizer/quality/sources.yaml` with source-tier mappings for known depositors. Wire the scoring into `GetAnnotationPage` output. Test with mock data. ### Definition of done Confidence information present in the output DataFrame. Source tier mappings cover at least the top 10 depositors by volume. Tests verify the scoring logic.
Sign in to join this conversation.