torchrec and RecTools

PyTorch RecSys is a low-level tensor computation framework for building recommendation models at scale, while RecTools is a higher-level library that simplifies the end-to-end workflow of building and evaluating recommender systems, making them complementary tools that could be used together in a production pipeline.

torchrec
82
Verified
RecTools
54
Established
Maintenance 22/25
Adoption 10/25
Maturity 25/25
Community 25/25
Maintenance 10/25
Adoption 10/25
Maturity 16/25
Community 18/25
Stars: 2,488
Forks: 618
Downloads:
Commits (30d): 139
Language: Python
License: BSD-3-Clause
Stars: 430
Forks: 54
Downloads:
Commits (30d): 0
Language: Python
License: Apache-2.0
No risk flags
No Package No Dependents

About torchrec

meta-pytorch/torchrec

Pytorch domain library for recommendation systems

This helps recommendation system engineers and machine learning scientists build, train, and deploy large-scale personalization models more efficiently. You provide user interaction data (like clicks or purchases), and it helps generate predictions for what users might like next. This is for teams developing recommendation features for products with many users and items, like e-commerce platforms or social media feeds.

recommendation-systems personalization machine-learning-engineering large-scale-data data-science

About RecTools

MTSWebServices/RecTools

RecTools - library to build Recommendation Systems easier and faster than ever before

This helps data scientists and machine learning engineers quickly build and evaluate recommendation systems. You provide historical user interaction data (like purchases or views), and it generates a list of personalized item recommendations for each user. This is ideal for those responsible for improving user engagement and conversion rates in products and services that offer a large catalog of items.

e-commerce recommendations content personalization data science machine learning engineering user engagement

Scores updated daily from GitHub, PyPI, and npm data. How scores work