colonyx 0.1.0

A Pythonic toolkit for Ant Colony, Particle Swarm, and Bee Colony optimization — written in Rust for high performance, designed for real-world use.

colonyx

colonyx is a Python library for solving optimization problems using swarm intelligence algorithms like Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), Grey Wolf Optimization (GWO), Firefly (FA), Simulated Annealing (SA), Cuckoo Search (CS), Bat Algorithm (BA), Glowworm Swarm Optimization (GSO), Bacterial Foraging (BFO), and Differential Evolution (DE).

While the interface is Pythonic and easy to use, the core is written in Rust to deliver better performance for larger or more complex problems.

Documentation: see docs/index.md for the library guide and API overview.

This is an early version — contributions, suggestions, and feedback are all welcome.

Features

• Ant Colony Optimization (ACO) — for discrete problems like TSP
• Particle Swarm Optimization (PSO) — for continuous function optimization
• Artificial Bee Colony (ABC) — inspired by bee foraging behavior
• Grey Wolf Optimization (GWO), Firefly (FA), Simulated Annealing (SA)
• Cuckoo Search (CS), Bat Algorithm (BA), Glowworm Swarm Optimization (GSO)
• Bacterial Foraging (BFO), Differential Evolution (DE)
• CMA-ES for covariance-adaptive continuous search
• Binary PSO, permutation GA, NSGA-II, and MOPSO for advanced search
• ACO variants including ACS, elitist, and MMAS behavior
• Simple, clean Python API
• Fast backend powered by Rust

Installation

pip install colonyx

(Install from source while release automation is being prepared.)

Example

All algorithms are used through the unified AutoColony interface, selected via the mode parameter.

Continuous optimization (PSO / ABC) — minimize an objective function over a box, given per-dimension bounds:

from colonyx import AutoColony

def sphere(x):
    return sum(xi * xi for xi in x)  # minimum 0 at the origin

opt = AutoColony(mode="pso", n_iterations=150, random_state=42)
opt.fit(sphere, bounds=[(-5, 5), (-5, 5), (-5, 5)])

opt.predict()  # best position, ~ [0, 0, 0]
opt.score()    # objective value at that position, ~ 0

# Artificial Bee Colony works the same way:
AutoColony(mode="abc", n_iterations=200).fit(sphere, bounds=[(-5, 5)] * 3)

Discrete optimization (ACO) — find a short tour through a square distance matrix (TSP):

import numpy as np
from colonyx import AutoColony

distance_matrix = np.array([
    [0, 1, 9, 9, 1],
    [1, 0, 1, 9, 9],
    [9, 1, 0, 1, 9],
    [9, 9, 1, 0, 1],
    [1, 9, 9, 1, 0],
], dtype=float)

opt = AutoColony(mode="aco", n_iterations=100, random_state=42)
opt.fit(distance_matrix)

opt.predict()  # best tour, e.g. [0, 1, 2, 3, 4]
opt.score()    # tour length (lower is better)

Use mode="auto" to let colonyx pick ACO for a square matrix or PSO for an objective function automatically.

For advanced optimizers, see docs/algorithms/advanced.md.

Documentation

• Library guide: docs/index.md
• AutoColony API: docs/autocolony-api.md
• CLI: docs/cli.md
• Getting started: docs/getting-started.md
• Algorithm overview: docs/algorithms.md
• Release and packaging notes: docs/release.md

License

MIT