# ycjuan/libffm

A Library for Field-aware Factorization Machines

Repository: https://github.com/ycjuan/libffm
Canonical: https://ross.abutalabs.com/products/libffm
Language: C++
License: BSD-3-Clause
License Family: permissive
Archived: true
Last push: 2024-08-16T04:32:57+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": 4105, "days_push": 747, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1599, forks 454 (observed 2026-08-28T04:05:09.470356+00:00)

## What it is
LIBFFM is a C++ library implementing field-aware factorization machines (FFM), a model especially effective for click-through-rate (CTR) prediction. It ships command-line tools for training and prediction, with OpenMP/SSE support and early-stopping via a validation set.

## Use cases
- train a field-aware factorization machine for CTR prediction
- predict ad click-through rates from categorical feature data
- build a click prediction model for display advertising
- compete in Kaggle CTR competitions like Criteo or Avazu
- train FFM models with early stopping on a validation set
- run fast multithreaded ML training with OpenMP and SSE

## When to choose
- you need a proven, competition-winning FFM implementation for sparse categorical data
- your task is CTR prediction or similar click/recommendation modeling
- you want a lightweight command-line trainer without heavy ML framework dependencies

## When to avoid
- you need deep learning or GPU-based models rather than factorization machines
- you want a Python-native API or integration with scikit-learn pipelines
- your features are dense continuous data rather than field-indexed sparse features

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, cli
- domain: machine-learning, data-science
- platform: windows, cpp, cli
- tags: factorization-machines, ctr-prediction, recommendation, openmp, kaggle, linux, macos

## Member repositories
- ycjuan/libffm (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:09.470356+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 2026-08-30T03:53:15.049262+00:00, confidence not recorded.
  - readme: https://github.com/ycjuan/libffm (fetched 2026-08-28T04:05:09.470356+00:00, sha 0ebcf3a9b547)
- Data as of 2026-08-30T08:39:29.467469+00:00.