jfkirk

TensorRec: Build Recommendation Systems with TensorFlow

TensorRec is a Python recommendation framework built on TensorFlow. It lets developers customize recommendation and ranking algorithms using user features, item features, and interaction data.

View on GitHub Compare
Stars
1.3k
Forks
219
Stars, last 30 days
-1
Last commit
3.4 years ago
Contributors
6

Our take

Inactiveno commits in over a year

Health score

21/100

How it is scored
Activity
19
Community
25
Issues
21
Pull requests
20

-1 stars in the last 30 days

TensorRec provides a focused TensorFlow framework for customizable recommendation models, but its inactive development and highly concentrated contributions make it a maintenance risk for new production adoption.

Good fit if

  • You need a Python framework for experimenting with recommendation algorithms built on TensorFlow.
  • You can maintain or adapt the project yourself if compatibility issues arise.
  • You want to explore recommendation modeling with user, item, and interaction features.

Look elsewhere if

  • You need ongoing upstream development, releases, or responsive issue and pull request handling.
  • Your project depends on a recommendation tool with a broader distribution of maintainers.
  • You need a different recommendation system approach; the README points to TensorFlow Ranking, Spotlight, and LightFM as alternatives.
All health signals
Last commit2023-05-22 (3.4 years ago)
Commits, last 90 days0
Releases, last 12 months0
Contributors6 (top contributor: 92% of commits)
Issues closed, last 90 days0
Pull requests merged, last 90 days0
Project age9 years

Checked on 2026-10-16 with the GitHub API.

Project history

Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.

Selected snapshot

Stars

1,298

Forks

218

Change since Nov 17, 2026: 0 stars·0 forks

Overview

TensorRec is a Python recommendation system framework built on TensorFlow. It helps developers build and customize recommendation algorithms without having to implement the surrounding data handling, scoring, and ranking themselves.

A model learns from user features, item features, and interaction data. Developers can customize its representation and loss functions, then use it to score and rank recommendations. See the project wiki for usage details.

Key Features

  • Train recommendation models from user features, item features, and interaction data.
  • Customize representation or embedding functions with TensorFlow.
  • Customize loss functions to shape how the model learns.
  • Generate recommendation scores and ranked item lists.
  • Evaluate rankings with metrics such as recall at k.
  • Generate dummy data for experimentation.

Use Cases

  • A Python developer prototyping a recommender can start with generated data and the default model.
  • A machine-learning practitioner can experiment with different representations and loss functions for recommendation tasks.
  • A team with user, item, and interaction features can train a model to score and rank candidate items.
  • A developer teaching or exploring recommender systems can use the framework to test ranking and recall evaluation workflows.

What you need

Detected in the repository

  • Python (from setup.py)
  • A test suite

License in plain words

Apache-2.0permissive

  • Commercial use: yes
  • Modify and redistribute: yes
  • You must keep: the license, the NOTICE file and a note of your changes
  • Share your changes: no
  • Includes an explicit patent grant from the contributors:

A summary, not legal advice: the LICENSE file is what applies.

Getting Started

Install with pip:

pip install tensorrec

See the README for a basic training and prediction example.

Alternatives

  • spotlight: Spotlight offers matrix-factorization and sequence-based recommendation models, while TensorRec focuses on customizable models built around user and item features.
  • fastFM: fastFM provides factorization-machine solvers for ranking and other tasks, rather than TensorRec’s broader recommendation-focused framework.
ProjectLanguageLicenseStarsStatus
TensorRecPythonApache-2.01.3kInactive
spotlightPythonMIT3kInactive
fastFMPython–1.1kInactive

Considerations

The repository is marked inactive: it has had no commits in over a year and no recent releases. Although several contributors have participated, most of the code comes from one person, and there has been no recent issue or pull request activity. The repository lists Python as its runtime, includes tests, and has no CI configured in the supplied project facts. The README does not state a GPU or database requirement.

Found this useful?

Share it with someone who would like TensorRec.

Comparisons