#138138th of 214 in ML frameworks and librariesFor data and documents, data analysis
pysr
Symbolic regression library that finds equations from data
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.2k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
PySR searches for interpretable symbolic expressions that fit a dataset, using an evolutionary search implemented in the Julia library SymbolicRegression.jl. It exposes a scikit-learn-style PySRRegressor in Python, with custom operators and losses written in Julia syntax. Fitted equations can be exported to SymPy, JAX or PyTorch, and runs can be distributed over a cluster such as Slurm.
Who it is for: Researchers and data scientists who want closed-form equations from data
Strengths
- scikit-learn-style fit/predict API with a table of equations by complexity and loss
- Exports equations to SymPy, callable functions, and differentiable JAX and PyTorch
- Custom operators, losses, constraints and per-operator complexity are configurable
- Multithreaded Julia search backend, with cluster support such as Slurm
Weaknesses
- Julia dependencies install on first import, which slows initial setup
- Suited to low-dimensional data; high dimensions need symbolic distillation of a neural net
- Possible GLIBCXX import crash needs an LD_LIBRARY_PATH workaround
- Warm-start state breaks if operators or search settings change between fits
What it needs
- no GPU
- Docker
- Needs Julia, SymbolicRegression.jl, SymPy, pandas
- Models: symbolic regression, scikit-learn-style estimator, JAX export, PyTorch export
Also in ML frameworks and libraries
See all 214| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | ultralyticsPython library and CLI for training and running YOLO vision models | 94 out of 100 |
| 3 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 94 out of 100 |
| #4 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |