UnravelSports/unravelsports

The unravelsports package aims to aid researchers, analysts and enthusiasts by providing intermediary steps in the complex process of turning raw sports data into meaningful information and actionable insights.

67
/ 100
Established

Provides specialized modules for converting multi-source tracking data (Kloppy-compatible soccer formats, NFL BigDataBowl) into Polars DataFrames with computed features like velocity and acceleration. Enables graph neural network training via PyTorch Geometric or Spektral with built-in data splitting strategies (by match/possession/sequence) to prevent leakage, plus domain-specific analytics including pressing intensity metrics and elastic formation identification via research-backed algorithms.

233 stars and 111 monthly downloads. Available on PyPI.

Maintenance 10 / 25
Adoption 15 / 25
Maturity 25 / 25
Community 17 / 25

How are scores calculated?

Stars

233

Forks

32

Language

Python

License

MPL-2.0

Last pushed

Jan 16, 2026

Monthly downloads

111

Commits (30d)

0

Dependencies

3

Get this data via API

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

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