villares / pinga

Interactive Genetic Algorithm in Processing (Python mode)

pinga

Interactive Genetic Algorithm in Processing (Python mode)

Contents

 handson: sketchbook for hands-on session (see below)
    bank: image bank (see below)
   faces: hands-on solution
sketches: another "species" using the same interactive GA engine

Hands-on sketchbook

The directory handson is incomplete. It already contains a visual frontend and some of the genetic-algorithm related routines.

The following routines will be implemented during the hands-on session:

• random_individual(), generates new individual from scratch
• draw_individual(), aka produce the "phenotype"
• mutate(), one of the genetic "operators"
• create_child(), another genetic operator

If there is time left, we will also implement the following routines (otherwise we paste them from sketchbook in the faces directory`):

• new_population_random(), generates new population of random individuals
• new_population_mutants(), generates new population of mutants from existing population
• new_population_children(), generates new population of children of selected individuals

Image bank

The directory bank contains images of eyes, noses and mouths. New images can be downloaded from the web and simply saved into this folder to be made available for the program. However, all images must start either with the word "eye", "nose", or "mouth" in order to be found by the program.