ycjuan/libffm

Health v2 · maintenance only

Flags: archived

Adoption not part of the score

1599 stars · 454 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

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

library · maturity maintenance

machine-learning cli machine-learning data-science windows cpp cli factorization-machines ctr-prediction recommendation openmp kaggle linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ycjuan/libffm")