# ChenglongChen/tensorflow-DeepFM

Tensorflow implementation of DeepFM for CTR prediction.

Repository: https://github.com/ChenglongChen/tensorflow-DeepFM
Canonical: https://ross.abutalabs.com/products/tensorflow-deepfm
Language: Python
License: MIT
License Family: permissive
Topics: factorization-machine, deepfm, ctr-prediction, deep-ctr, ctr, click-through-rate
Last push: 2018-06-10T11:10:10+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3217, "days_push": 3006, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2066, forks 802 (observed 2026-08-28T04:06:10.133649+00:00)

## What it is
A TensorFlow implementation of the DeepFM model combining factorization machines and deep neural networks for click-through-rate (CTR) prediction. It supports classification and regression tasks, configurable FM/DNN components, early stopping, and refitting.

## Use cases
- predict click-through rates for online ads
- train a DeepFM model in TensorFlow
- build a CTR prediction model with factorization machines
- run deep learning based CTR models on sparse categorical features
- use FM or DNN components separately for recommendation tasks
- prepare Kaggle competition features for DeepFM

## When to choose
- you need a proven DeepFM implementation for CTR or recommendation ranking tasks
- you work with sparse categorical feature fields and want FM plus deep learning in one model
- you want a simple, well-documented reference implementation for learning or benchmarking

## When to avoid
- you need a maintained library compatible with modern TensorFlow 2.x
- you want production-grade feature engineering pipelines or serving infrastructure
- you need models beyond DeepFM/FM/DNN variants

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning
- domain: machine-learning, deep-learning, data-science
- platform: python
- tags: deepfm, factorization-machine, ctr-prediction, recommendation, tensorflow, click-through-rate

## Member repositories
- ChenglongChen/tensorflow-DeepFM (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:10.133649+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 2026-08-30T02:56:53.060923+00:00, confidence not recorded.
  - readme: https://github.com/ChenglongChen/tensorflow-DeepFM (fetched 2026-08-28T04:06:10.133649+00:00, sha 7602bf12ae4e)
- Data as of 2026-08-30T08:39:29.467469+00:00.