Compute SHAP values and generate interpretability visualizations for machine learning models.
npx skills add https://github.com/shushuzn/Rairos --skill shap-shushuzn
Please help me install this Agent Skill. Skill: shap Source: https://github.com/shushuzn/Rairos/tree/main/skills/shap Command: npx skills add https://github.com/shushuzn/Rairos --skill shap-shushuzn
SHAP helps you understand and validate why a machine learning model produced a specific prediction or ranking, turning opaque model outputs into feature-level explanations you can trust and act on.
Use the shap skill to explain model predictions for your trained dataset by asking: “Compute SHAP values for my model and generate a beeswarm plot plus a waterfall plot for the first test row.”
High-intent search queries and answers about installing and using this skill.
Compute SHAP values by selecting an appropriate explainer for your trained model and dataset, generating feature attributions to interpret predictions across tasks like feature importance, local breakdowns, and bias checks.
SHAP explainers support tree-based models with fast explainers, linear models, deep learning through neural explainers, and black-box models using fallback model-agnostic explainers for broad compatibility.
Generate standard interpretability visualizations including beeswarm, bar, waterfall, scatter (dependence), heatmap, and force plots to visualize global and local feature attributions with correct output-units handling.
Feature attribution helps ML debugging by diagnosing unexpected feature importance, detecting potential data leakage, exploring feature interactions, and supporting fairness and bias analysis across different cohorts.
Build production-ready explanation workflows by saving and reusing explainers, batching SHAP value computations, and integrating explanation outputs into reporting or APIs for deployment validation.
Generating SHAP values requires a trained machine learning model and a corresponding dataset as inputs to compute feature attributions and produce visualizations for prediction interpretation.
ManfronEnricoCompute SHAP values and visualizations for tree, deep, and linear models.
felixboehmExplain machine learning model predictions using SHAP values and plots.
schneidermuCompute SHAP values and generate global and local model explanations.
tondevrelAnalyze scikit-learn model feature importance with SHAP, LIME, and partial dependence plots.
