shap - Explain machine learning models with SHAP
Compute SHAP values and create explanations, visualizations, debugging analyses, model comparisons, and fairness assessments for machine learning models.
Tags
Updated: 2026-10-04machine learningmodel interpretabilityexplainable AISHAPfeature importancefairness analysismodel debugging
Capabilities
Typical Inputs
What this skill does
- Compute SHAP values
- Select model explainers
- Generate SHAP plots
- Explain individual predictions
- Analyze feature importance
- Debug model behavior
- Compare model explanations
- Analyze bias and fairness
- Implement production explanations
Inputs
- Machine learning model
- Training data
- Test data
- Background data
- Feature names
- Target labels
- Protected attributes
Outputs
- SHAP values
- Feature importance plots
- Individual prediction explanations
- Model debugging findings
- Model comparison results
- Fairness analysis results
- Explanation service endpoints
Requirements
- Python environment
- SHAP library
- Compatible model framework
- Model access