pypsoa

A simple, zero-dependency Particle Swarm Optimization (PSO) library for Python.

pypsoa is a lightweight implementation of the Particle Swarm Optimization algorithm, designed for educational purposes and simple optimization tasks. It provides a clean, object-oriented interface with no external dependencies beyond Python's standard library.

Features

• Zero Dependencies: Only requires Python standard library
• Simple API: Easy-to-use classes for Vector, Particle, and Swarm
• Educational: Clear implementation perfect for learning PSO concepts
• Flexible: Customizable parameters for different optimization scenarios
• 2D Optimization: Optimized for 2-dimensional search spaces

Installation

pip install pypsoa

Or install from source:

git clone https://github.com/atasoglu/pypsoa.git
cd pypsoa
pip install .

Quick Start

import math
from pypsoa import Vector, Swarm

# Define a fitness function (minimize distance to target)
target = Vector(50, 50)
def fitness_function(pos):
    return math.sqrt((pos.x - target.x)**2 + (pos.y - target.y)**2)

# Create a swarm with 50 particles
swarm = Swarm.from_bounds(
    num_particles=50,
    x_bounds=(0, 100),
    y_bounds=(0, 100),
    w=0.5,  # inertia weight
    c1=1.0, # cognitive coefficient
    c2=2.0  # social coefficient
)

# Run optimization for 20 iterations
for _ in range(20):
    swarm.step()      # Update particle positions
    swarm.eval(fitness_function)  # Evaluate fitness

# Get best solution
best_position, best_fitness = swarm.g_best
print(f"Best position: {best_position}")
print(f"Best fitness: {best_fitness}")

See API Reference for more details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT License - see LICENSE file for details.