# TeamHG-Memex/eli5

A library for debugging/inspecting machine learning classifiers and explaining their predictions

Repository: https://github.com/TeamHG-Memex/eli5
Canonical: https://ross.abutalabs.com/products/eli5
Homepage: http://eli5.readthedocs.io
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: scikit-learn, machine-learning, xgboost, lightgbm, crfsuite, inspection, explanation, nlp, data-science, python
Last push: 2026-04-08T21:01:12+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 76, release rhythm 35, longevity 100
- inputs: {"age_days": 3640, "days_push": 147, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2798, forks 324 (observed 2026-08-28T04:07:22.481056+00:00)

## What it is
ELI5 is a Python library for debugging, inspecting, and explaining machine learning classifiers and regressors. It supports scikit-learn, XGBoost, LightGBM, CatBoost, Keras, and black-box models via LIME and permutation importance, with text, HTML, DataFrame, and JSON output formats.

## Use cases
- explain predictions of a scikit-learn classifier
- show feature importances of an xgboost model
- visualize which words influenced a text classifier's prediction
- explain image classifier predictions with grad-cam
- compute permutation importance for a black-box estimator
- print a decision tree as text or svg
- debug a scikit-learn pipeline with hashing vectorizer

## When to choose
- you need to interpret or debug predictions from scikit-learn, xgboost, lightgbm, or catboost models
- you want LIME-based explanations for arbitrary text classifiers
- you need explanation output as HTML for notebooks or JSON for custom rendering

## When to avoid
- you need explanations for modern deep learning models beyond Keras image classifiers
- you require actively developed tooling with support for the latest model types
- you need model-agnostic explainers like SHAP with theoretical guarantees

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, nlp, data-science
- domain: machine-learning, data-science
- platform: python
- tags: explainability, interpretability, lime, scikit-learn, xgboost, lightgbm, feature-importance, grad-cam, natural-language-processing

## Member repositories
- TeamHG-Memex/eli5 (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:22.481056+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T08:14:37.733018+00:00, confidence not recorded.
  - readme: https://github.com/TeamHG-Memex/eli5 (fetched 2026-08-28T04:07:22.481056+00:00, sha c9821cd7976f)
- Data as of 2026-08-30T08:39:29.467469+00:00.