Governance Studio dashboard

HUGIML Governance Studio is a Streamlit dashboard for model validation and audit workflows.

Install the dashboard optional dependencies, then launch the dashboard console command:

pip install "hugiml-core[dashboard]"
hugiml-dashboard

For local source-tree development, the app can also be launched directly with Streamlit or through the module entry point:

python -m streamlit run src/hugiml/dashboard/app.py
python -m hugiml.dashboard.app

Workflow sections

  • Overview and validation metrics.

  • Representation audit across original features, HUG patterns, and augmented pairs.

  • Pattern inventory and case-level prediction review.

  • Missingness, drift, and fairness-oriented checks.

  • Configuration comparison and representation pruning analysis.

  • Governance summaries for model review packets.

Data inputs

The dashboard supports demo data and uploaded tabular files including CSV, TSV, Excel, and Parquet. Users can select target, identifier, and excluded columns before fitting and reviewing a model.