jakubpeleska/redelex

ReDeLEx is a Python framework for developing and evaluating RDL models on relational databases via RelBench and CTU datasets.

48
/ 100
Emerging

This framework helps data scientists and machine learning researchers efficiently build and test deep learning models on complex relational databases. It takes a relational database (like PostgreSQL or MySQL) and transforms it into a graph representation, then applies various graph neural networks or other deep learning architectures. The output is a trained model capable of making predictions on tasks like classification or temporal events.

Use this if you need to develop, benchmark, and compare deep learning models on structured relational data for tasks like predicting outcomes or understanding relationships within your database.

Not ideal if your data is unstructured, purely tabular without meaningful relationships between tables, or if you only need classical machine learning models without exploring deep learning on graphs.

relational-database-modeling graph-neural-networks machine-learning-research predictive-analytics data-science-experimentation
No Package No Dependents
Maintenance 10 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

20

Forks

6

Language

Jupyter Notebook

License

MIT

Last pushed

Mar 05, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/jakubpeleska/redelex"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.