yandex-research/rtdl-num-embeddings

(NeurIPS 2022) On Embeddings for Numerical Features in Tabular Deep Learning

53
/ 100
Established

This project helps data scientists improve the accuracy of deep learning models when working with tabular data that includes numerical measurements. By transforming raw numerical inputs into a richer vector format before feeding them into the model, it allows for more nuanced pattern recognition. The output is a more performant predictive model, useful for anyone building machine learning solutions on datasets like customer behavior, financial metrics, or scientific readings.

408 stars. No commits in the last 6 months. Available on PyPI.

Use this if you are a data scientist or machine learning engineer building deep learning models on tabular data and want to improve predictive performance, especially when dealing with complex or irregularly distributed numerical features.

Not ideal if your primary goal is extreme model simplicity or if your dataset is very small, as the overhead of embeddings might outweigh the benefits.

predictive-modeling data-analysis machine-learning deep-learning tabular-data
Stale 6m
Maintenance 2 / 25
Adoption 10 / 25
Maturity 25 / 25
Community 16 / 25

How are scores calculated?

Stars

408

Forks

41

Language

Python

License

MIT

Last pushed

Apr 16, 2025

Commits (30d)

0

Dependencies

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/yandex-research/rtdl-num-embeddings"

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