minionpy 1.6.0

MinionPy is the Python implementation of the Minion (C++) ibrary.

MinionPy

MinionPy is the Python interface to the Minion C++ optimization library. It focuses on single-objective, derivative-free optimization. The package includes several population-based and local optimization methods, along with CEC benchmark suites that can be used for testing and comparison.

Features

• Optimization Algorithms

  • Differential Evolution-based algorithms:
    • Basic Differential Evolution (DE)
    • JADE
    • LSHADE
    • ARRDE
    • and other DE variants
  • Other population-based algorithms:
    • Artificial Bee Colony (ABC)
    • Grey Wolf DE Optimization
    • Canonical PSO, SPSO-2011, Dynamic Multi-Swarm PSO (DMS-PSO)
    • CMA-ES (Covariance Matrix Adaptation Evolution Strategy)
    • BIPOP-aCMAES
    • RCMAES
  • Classical optimization algorithms:
    • Nelder-Mead
    • Generalized Simulated Annealing (Dual Annealing)
    • L-BFGS-B (vectorized & noise-robust)
    • L-BFGS (vectorized & noise-robust)

• Benchmark Support
  The library includes benchmark functions from the CEC competitions (2011, 2014, 2017, 2019, 2020, 2022), providing a standardized environment for algorithm development, testing, and comparison.

• Performance
  Most implemented algorithms are population-based, making them suitable for parallelization. MinionPy is optimized for vectorized functions, enabling efficient use of multithreading and multiprocessing capabilities.

• Cross-Platform Compatibility
  MinionPy is implemented in C++ with a Python wrapper, supporting usage in both languages. It has been tested on the following platforms:

  • Windows 11
  • Linux Ubuntu 24.04
  • macOS Sequoia 15

Applications

MinionPy is applicable in scenarios where single objective, bound-constrained/unconstrauned optimization is required, including engineering, physics, and machine learning. Its standardized benchmarks and high-performance algorithms make it suitable for developing and evaluating new optimization techniques as well as solving real-world optimization problems.

📖 Documentation

For full usage instructions, API reference, and examples, visit the official documentation:

• Minion Documentation

Citing Minion

If you use MinionPy in your research or projects, we would be grateful if you could cite the following publication:

Muzakka, K. F., Möller, S., & Finsterbusch, M. (2025).
Minion: A high-performance derivative-free optimization library designed for solving complex optimization problems.
Zenodo. https://doi.org/10.5281/zenodo.14794239