Implement confidence scoring #17
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
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.pywith a function likescore_dataframe(df: pl.DataFrame, annotation: Annotation) -> pl.DataFramethat adds confidence columns. Createpcdigitizer/quality/sources.yamlwith source-tier mappings for known depositors. Wire the scoring intoGetAnnotationPageoutput. 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.