⬢github Python · 4.2K ★ +23 since we first saw it · pushed 2 d ago · Apache-2.0
astroautomata/PySR
Symbolic Learning in Python and Julia
PySR is an open-source symbolic regression library for Python and Julia that searches for interpretable mathematical expressions fitting data. Built on the high-performance SymbolicRegression.jl engine, it can also 'distill' neural networks into analytic equations, making deep models easier to interpret. It installs via pip or conda and includes distributed search, scikit-learn integration, and an AI-agent skill.
Why now: It was featured on Hacker News, and the README now advertises a new skill for using PySR inside AI agent workflows.
Who it is for: Data scientists and researchers who want interpretable equations from data or neural networks instead of black-box models.
algorithmautomldata-sciencedistributed-systemsequation-discoveryevolutionary-algorithmsexplainable-aigenetic-algorithminterpretable-mljulia
Stars over our 35 snapshots: 4.1K to 4.2K, since 8 h ago.
Where people talked about it
API: https://socialmediatrends-api.osmike.com/v1/repos/astroautomata/PySR