# recommenders-team/recommenders

Best Practices on Recommendation Systems

Repository: https://github.com/recommenders-team/recommenders
Canonical: https://ross.abutalabs.com/products/recommenders
Homepage: https://recommenders-team.github.io/recommenders/intro.html
Language: Python
License: MIT
License Family: permissive
Topics: machine-learning, recommender, ranking, deep-learning, python, jupyter-notebook, recommendation-algorithm, rating, operationalization, kubernetes, recommendation-system, recommendation-engine, recommendation, data-science, tutorial, artificial-intelligence, ai
Last push: 2026-08-25T20:44:59+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 8, longevity 100
- inputs: {"age_days": 2905, "days_push": 8, "days_rel": 617, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 21864, forks 3319 (observed 2026-08-28T04:11:32.167852+00:00)

## What it is
A Python library and collection of Jupyter notebooks with best practices for building, evaluating, and operationalizing recommendation systems, from classic algorithms like ALS to deep learning models like xDeepFM. It is a Linux Foundation AI & Data project providing utilities for data preparation, model selection, offline evaluation, and production deployment.

## Use cases
- build a recommendation system in python
- compare recommender algorithms like ALS and xDeepFM
- evaluate recommendation models with offline metrics
- tune hyperparameters for a recommendation model
- deploy a recommendation model to production
- learn best practices for recommender systems
- split and prepare data for recommendation algorithms

## When to choose
- you want ready-made implementations and notebooks for classic and deep learning recommenders
- you need utilities for dataset loading, train/test splitting, and offline evaluation of recommendations
- you want to prototype and operationalize recommendation models on Azure or Kubernetes

## When to avoid
- you need a turnkey production recommendation service with no ML coding
- your use case is general machine learning unrelated to recommendations
- you require a framework outside the Python/Spark ecosystem

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, data-science, benchmarking, etl
- domain: machine-learning, data-science, artificial-intelligence, tutorials
- platform: python, cloud
- tags: recommender-systems, jupyter-notebooks, collaborative-filtering, deep-learning-recommendation, model-evaluation, hyperparameter-tuning, gpu, docker, kubernetes

## Member repositories
- recommenders-team/recommenders (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:32.167852+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 2026-08-29T16:57:46.565103+00:00, confidence not recorded.
  - readme: https://github.com/recommenders-team/recommenders (fetched 2026-08-28T04:11:32.167852+00:00, sha 375567c3c282)
  - homepage: https://recommenders-team.github.io/recommenders/intro.html (fetched 2026-08-29T07:56:07.083846+00:00, sha c2d88978cd27)
  - registry_pypi: https://pypi.org/pypi/recommenders/json (fetched 2026-08-29T07:56:07.086792+00:00, sha 1e1fccd30589)
- Data as of 2026-08-30T08:39:29.467469+00:00.