MilesCranmer/PySR
PySR is a high-performance symbolic regression library that uses evolutionary algorithms to discover interpretable mathematical expressions from data.

PySR combines Python and Julia implementations to search for symbolic expressions that optimize a given objective function. It employs genetic algorithms and distributed computing to explore the space of mathematical formulas, allowing users to discover interpretable equations from observational data. The library integrates with scikit-learn and is designed for applications in scientific discovery, equation finding, and explainable machine learning.