# marcotcr/lime

Lime: Explaining the predictions of any machine learning classifier

Repository: https://github.com/marcotcr/lime
Canonical: https://ross.abutalabs.com/products/marcotcr-lime
Language: JavaScript
License: BSD-2-Clause
License Family: permissive
Last push: 2024-07-25T20:32:21+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3823, "days_push": 769, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12163, forks 1847 (observed 2026-08-28T04:10:51.888996+00:00)

## What it is
Lime (Local Interpretable Model-agnostic Explanations) is a Python library that explains the predictions of any machine learning classifier. It generates local explanations for individual predictions made by black-box models acting on text, tabular data, or images.

## Use cases
- explain predictions of a machine learning classifier
- interpret black box model predictions
- understand which features influenced a text classifier prediction
- visualize feature importance for tabular data models
- explain image classifier predictions
- generate local explanations for scikit-learn models
- debug machine learning model behavior on individual predictions

## When to choose
- you need model-agnostic explanations that work with any classifier exposing a predict-probability interface
- you want to explain individual predictions rather than global model behavior
- you work with text, tabular, or image classifiers and need interpretable output
- you use scikit-learn and want built-in support without extra wiring
- you want explanations embeddable in Jupyter notebooks

## When to avoid
- you need global model explanations or full surrogate models rather than per-prediction explanations
- you need explanations for deep learning models with tight GPU integration where newer tools like SHAP or integrated gradients fit better
- you require active development and frequent updates, since the project is in maintenance mode
- you need sub-second explanation latency for production inference pipelines, as LIME sampling adds overhead

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, data-visualization, nlp, image-processing
- domain: machine-learning, data-science, artificial-intelligence
- platform: python, cross-platform
- tags: explainable-ai, model-interpretability, local-explanations, black-box-models, feature-attribution, scikit-learn, jupyter-notebooks, xai, explainability, interpretability

## Member repositories
- marcotcr/lime (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:51.888996+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 2026-08-29T17:14:53.652355+00:00, confidence not recorded.
  - readme: https://github.com/marcotcr/lime (fetched 2026-08-28T04:10:51.888996+00:00, sha 5c951b2edfea)
- Data as of 2026-08-30T08:39:29.467469+00:00.