Skip to content

Profiler: "Compare two datasets" mode to detect representation drift #60

Description

@yakew7

Summary

Add a mode that profiles two CSVs (e.g. training vs. production) and reports how each demographic dimensions representation shifted.\n\n## Why\nDirectly operationalises the Distribution Shift explainer: a dataset that passed an audit can drift out of representativeness over time.\n\n## Tasks\n- [ ] CLI: faircode compare a.csv b.csv\n- [ ] Web: second dropzone + side-by-side score/dimension diff\n- [ ] Define the diff metrics in SPEC.md

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or requesthelp wantedExtra attention is needed

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions