haghish/shapley
Weighted Shapley Values and Weighted Confidence Intervals for Multiple Machine Learning Models and Stacked Ensembles
This tool helps researchers and analysts in machine learning assess and select the most important features in their models more reliably. It takes the results from multiple machine learning models (like those from a grid search or stacked ensemble) and calculates more stable, weighted average SHAP values along with confidence intervals. This allows users to understand which features truly drive predictions, even when dealing with complex scenarios like class imbalance.
Use this if you need to identify consistent and reliable feature importance across multiple machine learning models or ensemble setups, especially when a single 'best' model is hard to define or SHAP values appear unstable.
Not ideal if you only work with single, simple models and don't need to account for variability or ensemble effects in feature importance.
Stars
18
Forks
3
Language
R
License
—
Category
Last pushed
Mar 04, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/haghish/shapley"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
shap/shap
A game theoretic approach to explain the output of any machine learning model.
mmschlk/shapiq
Shapley Interactions and Shapley Values for Machine Learning
iancovert/sage
For calculating global feature importance using Shapley values.
predict-idlab/powershap
A power-full Shapley feature selection method.
aerdem4/lofo-importance
Leave One Feature Out Importance