danijar/granular

Fast dataset format and loader

48
/ 100
Emerging

Granular is a tool for developers who work with large, complex datasets. It helps you store and load diverse data types like images, text, and numerical arrays efficiently. You put in raw data in various formats and get out a structured, performant dataset that can be easily accessed and processed, especially for machine learning workflows. It's designed for data engineers and machine learning engineers managing big data pipelines.

Available on PyPI.

Use this if you need a flexible and high-performance way to store and load custom datasets with diverse data types, especially when random access and resumable processing are important.

Not ideal if you're looking for a simple CSV or JSON file loader, or if your dataset is small and easily fits into memory.

data-engineering machine-learning-ops large-scale-data data-loading dataset-management
Maintenance 10 / 25
Adoption 6 / 25
Maturity 25 / 25
Community 7 / 25

How are scores calculated?

Stars

24

Forks

2

Language

Python

License

MIT

Last pushed

Mar 06, 2026

Commits (30d)

0

Dependencies

3

Get this data via API

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

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