Compute SHAP values to explain model predictions with feature contributions.
npx skills add https://github.com/BKZhao/MIMIC-Papers-Repro-Agent --skill shap-bkzhao
Please help me install this Agent Skill. Skill: shap Source: https://github.com/BKZhao/MIMIC-Papers-Repro-Agent/tree/main/.codex/skills/shap Command: npx skills add https://github.com/BKZhao/MIMIC-Papers-Repro-Agent --skill shap-bkzhao
SHAP-based explanations provide faithful, local attributions for individual predictions by assigning each input feature a contribution to the model output, enabling debugging, trust-building, and fairness analyses across ML systems.
Train your model, create a suitable SHAP explainer, and compute SHAP values for a test set to begin interpreting predictions.
High-intent search queries and answers about installing and using this skill.
Explain machine learning model predictions by computing SHAP values to provide feature-level contributions, assigning each input feature an attribution for individual predictions and debugging model behavior.
Yes, SHAP explainers support tree-based models, neural networks, and linear models using TreeExplainer, DeepExplainer, LinearExplainer, and KernelExplainer to compute conditional expectations for explanations.
Visualize feature interactions and non-linear effects using SHAP dependence plots and heatmaps, which reveal feature relationships by plotting SHAP values against feature values across the dataset.
Generate local explanations for single predictions using SHAP waterfall and force plots, and global feature importance by aggregating local attributions across the dataset with beeswarm and bar summaries.
Yes, you need a representative background dataset to compute conditional expectations when applying SHAP explainers for model interpretation and feature attribution analysis.
Perform bias analysis and model debugging by comparing SHAP profiles across different cohorts, enabling trust-building and fairness analyses by inspecting feature contributions for various subgroups.
ManfronEnricoCompute SHAP values and visualizations for tree, deep, and linear models.
LeonChaoXCompute SHAP feature attributions and generate standardized plots for model predictions.
youyinnnExplain machine learning model predictions using SHAP values and plots.
tondevrelAnalyze scikit-learn model feature importance with SHAP, LIME, and partial dependence plots.
