# scikit-learn-contrib/lightning

Large-scale linear classification, regression and ranking in Python

Repository: https://github.com/scikit-learn-contrib/lightning
Canonical: https://ross.abutalabs.com/products/scikit-learn-contrib-lightning
Homepage: https://contrib.scikit-learn.org/lightning/
Language: Python
License Family: other
Topics: machine-learning
Archived: true
Last push: 2023-07-18T11:41:11+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 5348, "days_push": 1142, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1782, forks 217 (observed 2026-08-28T04:05:35.603350+00:00)

## What it is
A Python library for large-scale linear classification, regression, and ranking that follows scikit-learn API conventions. It supports dense and sparse data with computationally demanding parts implemented in Cython, offering solvers like CD, SDCA, SGD, AdaGrad, SAG, SAGA, SVRG, and FISTA.

## Use cases
- train linear classifiers on large sparse datasets
- fit linear regression models with L1 or group lasso penalties
- learn multiclass classifiers on text data like News20
- train online ranking models with PRank
- use SGD, SAGA, or SVRG solvers with a scikit-learn-compatible API
- perform feature selection via sparse linear models

## When to choose
- you need fast linear models on large-scale or high-dimensional sparse data
- you want scikit-learn API compatibility with more solver variety
- you need specialized penalties like l1/l2 group lasso or trace norm
- you want variance-reduced stochastic solvers like SAGA or SVRG

## When to avoid
- you need nonlinear models like gradient boosting or deep learning
- you need actively developed software with recent releases
- you need GPU acceleration
- you need a license file for compliance-sensitive use

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning
- domain: machine-learning, data-science
- platform: python, cross-platform
- tags: linear-classification, linear-regression, ranking, scikit-learn-api, cython, large-scale, sparse-data, optimization

## Member repositories
- scikit-learn-contrib/lightning (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:35.603350+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 2026-08-30T03:24:23.188508+00:00, confidence not recorded.
  - readme: https://github.com/scikit-learn-contrib/lightning (fetched 2026-08-28T04:05:35.603350+00:00, sha b6c4eb019667)
  - homepage: https://contrib.scikit-learn.org/lightning/ (fetched 2026-08-29T11:02:59.342066+00:00, sha 756808895c59)
- Data as of 2026-08-30T08:39:29.467469+00:00.