AdrianBZG/TabMDA

[ICML 2024] TabMDA: Tabular Manifold Data Augmentation for Any Classifier using Transformers with In-context Subsetting

21
/ 100
Experimental

When you have a small amount of crucial tabular data, like customer demographics or medical records, machine learning models often struggle to find reliable patterns. This project helps by taking your limited tabular dataset and intelligently generating more synthetic, yet realistic, data points. The output is an expanded dataset that helps your existing classification models perform better, making it easier for data scientists to get accurate predictions from scarce information.

No commits in the last 6 months.

Use this if you need to improve the performance of a machine learning classifier on a tabular dataset where data is scarce, and you want a method that doesn't require extra training.

Not ideal if you already have very large tabular datasets, as the performance gains might be less significant.

data-scarcity tabular-data machine-learning-performance dataset-expansion classification-models
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

9

Forks

Language

Python

License

MIT

Last pushed

Jul 26, 2024

Commits (30d)

0

Get this data via API

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

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