aerdem4/lofo-importance
Leave One Feature Out Importance
When you're building a predictive model, it's crucial to understand which factors, or 'features,' truly drive its predictions. This tool helps you identify the most impactful features by systematically removing each one and seeing how your model's performance changes. It takes your dataset and your chosen predictive model, then outputs a clear ranking of feature importance, ideal for data scientists, machine learning engineers, and researchers who build and interpret models.
863 stars. No commits in the last 6 months. Available on PyPI.
Use this if you need an unbiased, robust way to determine how much each input feature genuinely contributes to your model's accuracy, especially when dealing with complex or highly correlated data.
Not ideal if you need an extremely fast, approximate measure of feature importance for a very large dataset and a pre-trained model, as it involves retraining the model multiple times.
Stars
863
Forks
83
Language
Python
License
MIT
Category
Last pushed
Feb 14, 2025
Commits (30d)
0
Dependencies
7
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/aerdem4/lofo-importance"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
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.
ReX-XAI/ReX
Causal Responsibility EXplanations for Image Classifiers and Tabular Data