Skip to content

Home

Jump to bottom
cidelab edited this page Jun 7, 2017 · 27 revisions

Laga name comes from “Librería de Algoritmos Genéticos para Arquitectura”. The concept evolved but the name remains the same. The framework is designed for easy to use. Nevertheless you will need some programming knowledge to used efficiently and overall understand the concept of genetic algorithms. I’m open to discuss how the structure work and how could be improved. The framework currently have 3 main topics. Genetic Algorithms, Graphics and Interoperability.

Genetic Algorithms This section provides all the necessary tools to design and create Genetic Algorithms. The structure is very simple to use. Call the reference using Laga.GeneticAlgorithm; The structure of a generic GA works like this:

step operation Laga
1 Creates a random population Use the class GenrChromosome.cs and or GenrPopulation.cs. But it will depend on your objective.
Content Cell Content Cell

Step 1 Creates a random population Use the class GenrChromosome.cs and or GenrPopulation.cs. But it will depend of your objective. Step 2 evaluate Is up to you, it will depends of your problem Step 3 select the individuals with the highest evaluation Use the class NaturalSelection.cs Step 4 crossover the selected individuals to produce inheritance Use the class Crossover.cs Step 5 mutate the inheritance Use the class Mutation.cs Step 6 replace the original population Use the class Replacement.cs or develop your own method. Step 2 evaluate again. Back to step 2 until stop.

  1. Laga framework
  2. Getting started
  3. Genetic Algorithms
  4. Graphics
  5. IO

There are many ways to contribute: Use the framework, Review the code, Contribute with examples and documentation, Help improve this wiki.

Clone this wiki locally