Compute SHAP values and visualizations for tree, deep, and linear models.
npx skills add https://github.com/ManfronEnrico/thesis-manifold --skill shap-manfronenrico
Please help me install this Agent Skill. Skill: shap Source: https://github.com/ManfronEnrico/thesis-manifold/tree/main/.claude/skills/shap Command: npx skills add https://github.com/ManfronEnrico/thesis-manifold --skill shap-manfronenrico
SHAP provides principled, local- and global-level explanations for any ML model, helping you understand why predictions occur and which features drive outcomes.
Explain a trained model’s predictions by selecting the right SHAP explainer and generating feature attributions and visuals.
High-intent search queries and answers about installing and using this skill.
To explain predictions using SHAP values, apply the appropriate explainer to a trained model with a background dataset to generate local and global feature attributions. The Skill supports TreeExplainer, DeepExplainer, LinearExplainer, and KernelExplainer.
Feature attribution analysis visualizations include waterfall, beeswarm, bar, scatter, and heatmap plots. These visuals help communicate feature importance and interaction effects to stakeholders for debugging and fairness analysis.
SHAP explainers work with tree-based, deep learning, and linear models by applying TreeExplainer, DeepExplainer, LinearExplainer, or KernelExplainer respectively. Each requires a trained model and a background dataset to compute feature attributions.
You need SHAP for model interpretability when you require principled, local- and global-level explanations for any machine learning model. It provides unified feature attribution to understand why predictions occur and which features drive outcomes.
To compute SHAP explanations, you need a trained model, a background dataset, and an appropriate explainer such as TreeExplainer or KernelExplainer. These inputs produce SHAP values and visualizations for both global and local interpretations.
schneidermuCompute SHAP values and generate global and local model explanations.
youyinnnExplain machine learning model predictions using SHAP values and plots.
tomlupoExplain model predictions by attributing outputs to input features with SHAP values.
tondevrelAnalyze scikit-learn model feature importance with SHAP, LIME, and partial dependence plots.
