Skip to main content

Python implementation of Genetic Algorithm

Project description

pyga

Python implementation of Genetic Algorithm

Installation

pip3 install pygenal

Usage

this test is able to find global maximum of 6th order polynom with two variables in 500ms with precision approximately 10-4 to 10-5 % (depending on hardware)

#!/usr/bin/env python3

# import required classes
from pygenal.ga import Population, Individual, Gene, geneTypes, Duration
import numpy as np
import time


# create population of your own, override fitness method to suit your needs.
# your system evaluation goes here, for sake of demonstration, use polynom
class Polynom(Population):
    def fitness(self, individual):
        x = individual.get("x").value
        y = individual.get("y").value
        color = individual.get("color").value

        # test against function with global maximum:
        # max{x + 3 x^2 - x^4 - y^6 + y/3 - y^3 4 - y - 5} ≈ 3.35864
        # at (x, y) ≈ (1.30084, -1.27413)
        # + favorize chocolate over shorter colors, just for demonstration
        return (x + 3*(x**2) - x**4 - y**6 + y/3.0 - (y**3)*4 - y - 5) + len(color)/10


# introduce non-number options such as color, good for selections
colors = ["brown", "green", "grey", "blue", "chocolate"]


if __name__ == '__main__':

    # spawn individual
    i1 = Individual()

    # construct "DNA"
    i1 += Gene("x", geneTypes.REAL, np.random.randint(-10, 10), min=-10, max=10)
    i1 += Gene("y", geneTypes.REAL, np.random.randint(-10, 10), min=-10, max=10)
    i1 += Gene("color", geneTypes.VALUE, np.random.choice(colors), availableOptions=colors, dominant=True)

    # create tribe with size of 100 individuals based on your first Individual
    population = Polynom(
            species=i1,
            size=100,
            chanceOfMutationStart=.5,
            chanceOfMutationStop=0.0001
        )

    tStart = time.time()

    # start evolution until:
    #   you didn't pass 1000 generations OR
    #   fittest didn't change for 200 generations OR
    #   2.5s didn't passes yet
    population.evolve(
            allowCrossover=True,
            generations=100,
            verbose=True,
            timeout=Duration(seconds=0, miliseconds=500),
        )
    print(f"Evolved in {time.time()-tStart}s, precision: {(4.258642115713602 - population.fittest.score)*100}%")
    print(f"Fittest: {population.fittest}, genes: {repr(population.fittest)}")

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pygenal-1.0.7.tar.gz (4.6 kB view details)

Uploaded Source

Built Distribution

pygenal-1.0.7-py3-none-any.whl (6.9 kB view details)

Uploaded Python 3

File details

Details for the file pygenal-1.0.7.tar.gz.

File metadata

  • Download URL: pygenal-1.0.7.tar.gz
  • Upload date:
  • Size: 4.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.21.0 setuptools/45.2.0 requests-toolbelt/0.9.1 tqdm/4.42.1 CPython/3.7.5

File hashes

Hashes for pygenal-1.0.7.tar.gz
Algorithm Hash digest
SHA256 114b3c28acff00b44d0923d5037c0bcff0213e59892173af119a4df55675adbc
MD5 aeccb6b6ea09fce6621d61981e867c4d
BLAKE2b-256 d629ffa70b55d65c9593a3f2fb03c634d1b41c4388dcaf29985a6af21d4680b6

See more details on using hashes here.

File details

Details for the file pygenal-1.0.7-py3-none-any.whl.

File metadata

  • Download URL: pygenal-1.0.7-py3-none-any.whl
  • Upload date:
  • Size: 6.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.21.0 setuptools/45.2.0 requests-toolbelt/0.9.1 tqdm/4.42.1 CPython/3.7.5

File hashes

Hashes for pygenal-1.0.7-py3-none-any.whl
Algorithm Hash digest
SHA256 3365a068dc512cd3fe43404ab4dbfb20d52f723e400225ebc8759709e11a1a17
MD5 cc4f7c711b81e8460e92e2f8921afd30
BLAKE2b-256 e82b7aa1a5ee6e013beec6219cd655a65d84cc5bfffa39ada8f6080186176d58

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page