eggzec/pyswarm

Particle Swarm Optimization (PSO) for Python

What it solves

It provides a way to find the minimum of a function when you don't have gradient information (the slope of the function) or when it is too difficult to compute. This is useful for complex optimization problems where traditional calculus-based methods fail.

How it works

The library implements Particle Swarm Optimization (PSO), an evolutionary algorithm that mimics the social behavior of birds flocking or fish schooling. It searches for the optimal solution by moving a group of "particles" through the search space, where each particle adjusts its position based on its own best-known position and the rest of the swarm's overall best position.

Who it’s for

Developers and researchers who need a lightweight, gradient-free optimization tool for Python that supports constraints (lower and upper bounds).

Highlights

  • Lightweight and easy to use.
  • Built-in support for constraints.
  • Returns results in a format compatible with scipy.optimize.OptimizeResult.
  • Requires only Python 3.10+ and NumPy.

Related

  • Project
    facebookresearch/nevergrad

    A gradient-free optimization platform for minimizing functions, particularly useful for hyperparameter tuning in machine learning.

  • Project
    SimonBlanke/Gradient-Free-Optimizers

    A Python library for gradient-free optimization of black-box functions, providing a unified interface to 23 algorithms for hyperparameter tuning and simulation optimization.

  • Project
    CMA-ES/pycma

    A Python implementation of the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) for solving complex, derivative-free numerical optimization problems.

  • Project
    nnaisense/evotorch

    A scalable evolutionary computation library built on PyTorch that solves non-differentiable optimization problems using GPU acceleration and Ray distribution.

  • Project
    python-adaptive/adaptive

    A Python library for parallel active learning of mathematical functions that intelligently selects the best points to sample in a parameter space to reduce computational cost.