For the documentation of PyParadiseo see this.

This README only gives a short introduction.

Table of Contents

• Installation
• Usage
• Components

Installation

The easiest way to get pyparadiseo is to install it via pip. Currently the following Python versions are supported: 3.6, 3.7, 3.8, 3.9, 3.10

Install with pip

You can install pyParadiseo with pip

    pip install pyparadiseo

Build from source

To build pyParadiseo, you'll need to have a few prerequisites installed on your system and set the corresponding paths in setup.py and CMakeLists.txt

Prerequisites

To compile the binary extension you need: - cmake >= 3.14 - python3 >= 3.6 - boost-python - boost-numpy

Docker image

Here is a manylinux2014_x86_64 Docker image_ with installed prerequisites

.. _Docker image: https://hub.docker.com/repository/docker/jangmys/manylinux2014_boost180_cp36_310

Getting Started

The documentation of PyParadiseo is available here

Example of running EO's simple GA (SGA) for the One-Max test problem

from pyparadiseo import Pop
from pyparadiseo.evaluator import FitnessEval,PopLoopEval

from pyparadiseo import evaluator
from pyparadiseo import operator
from pyparadiseo import population
from pyparadiseo import initializer

from pyparadiseo.eo import algo,select_one,continuator

import numpy as np

if __name__ == "__main__":
    #set solution type globally
    config.set_solution_type('bin')

    #make pyparadiseo evaluator from python function
    eval = evaluator.fitness(lambda sol: np.count_nonzero(sol))

    #generate and evaluate population
    init = initializer.random(size=20)
    pop = population.from_init(25, init)
    pop_eval=evaluator.pop_eval_from_fitness(eval)
    pop_eval(pop,pop)

    #assemble simple GA
    sga = algo.simpleGA(
        select_one.det_tournament(4),
        operator.OnePtBitCrossover(),.1,
        operator.DetBitFlip(),.7,
        eval,
        continuator.max_generations(self.NGENS)
    )

    #run algo on pop and print best individual
    sga(pop)
    print(pop.best())

Components

• EO (Population-based single-objective)
• MO (Trajectory-based single-objective)
• MOEO (Multi-objective)
• Encodings : Binary, Integer, Real, Permutation, Custom
• Genetic Operators : ...