Explain model predictions with SHAP values and feature contribution attributions.
npx skills add https://github.com/swaruplab/operon --skill shap-swaruplab
Please help me install this Agent Skill. Skill: shap Source: https://github.com/swaruplab/operon/tree/main/src-tauri/protocols/shap Command: npx skills add https://github.com/swaruplab/operon --skill shap-swaruplab
SHAP provides a principled way to attribute a model's prediction to its input features, enabling both global and local explanations with theoretical guarantees.
Explain a trained model by selecting an appropriate SHAP explainer, computing SHAP values for a test set, and generating a waterfall or beeswarm visualization.
High-intent search queries and answers about installing and using this skill.
You explain machine learning model predictions by computing SHAP values to attribute feature contributions. This Skill applies SHAP to regression, classification, tree ensembles, and deep neural networks to generate global and local explanations.
The best way to visualize feature importance for tree-based models is using the TreeExplainer. It computes SHAP values to generate global visualizations like beeswarm and bar plots, as well as local explanations like waterfall and force plots.
Yes, you can use SHAP explainers with deep neural networks and linear models. The Skill supports DeepExplainer for neural networks and LinearExplainer for linear models to attribute prediction contributions across development and production pipelines.
You generate local explanations for individual predictions by computing SHAP values and creating waterfall or force plots. These visualizations isolate and display how each feature contributes to a specific prediction's output.
Yes, SHAP supports production-ready workflows for model debugging, model comparison, fairness checks, and feature engineering. It uses Explanation objects, background data, and baselines to maintain consistent interpretation across pipelines.
You should use KernelExplainer instead of TreeExplainer when interpreting models outside of tree ensembles, as KernelExplainer provides model-agnostic feature attribution while TreeExplainer is specifically optimized for tree-based models.
rocklambrosGenerate SHAP attributions for trained machine learning models.
dralkhCompute SHAP values and generate feature attribution plots for machine learning models.
jimmc414Generates SHAP values to explain feature importance and model predictions.
tondevrelAnalyze scikit-learn model feature importance with SHAP, LIME, and partial dependence plots.
