HowieHwong/DataGen

[ICLR'25] DataGen: Unified Synthetic Dataset Generation via Large Language Models

24
/ 100
Experimental

This tool helps researchers, data scientists, and AI developers create new, high-quality text datasets for training language models or evaluating their performance. You provide an existing dataset description and configuration, and it generates diverse, accurate, and customized synthetic text data. This is ideal for anyone needing to expand or create specialized text datasets without the expense and time of manual data collection.

No commits in the last 6 months.

Use this if you need to generate diverse, accurate, and highly controllable synthetic text datasets to train or evaluate large language models, especially when real-world data is scarce or costly to obtain.

Not ideal if you need to generate non-textual data like images or numerical tables, or if your primary goal is simple data augmentation without advanced control over diversity, accuracy, or specific constraints.

AI-model-training dataset-creation NLP-research data-synthesis language-model-evaluation
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 8 / 25

How are scores calculated?

Stars

66

Forks

4

Language

Python

License

Last pushed

Mar 08, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/generative-ai/HowieHwong/DataGen"

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