Skip to content
/ gdfann Public

A Genetic Algorithm that tune the Artificial Neural Network Parameters (FANN library)

Notifications You must be signed in to change notification settings

jeiks/gdfann

Repository files navigation

Distributed Genetic Algorithm to Fast Artificial Neural Network

###About

This project is a Genetic Algorithm made in Python that works with library FANN (Fast Artificial Neural Network).

The Genetic Algorithm is able to tune the Artificial Neural Network (ANN) parameters.

The Genetic Algoritm can be used to another applications to.

###Configuration

First, you have to edit the file "config_cromossomo.py" and fill the matrix tipoGenes with the types of genes that can be used.

Next, in same file, you have to fill the function 'avaliacaoRNA'. This function will evaluate each chromossome. This result must be: "Resultado: FITNESS_NUMBER"

Next, you have to edit the file "config.py" to set the other Genetic configurations.

###IMPORTANT

1. This work is under development.

   It is working parallelized and distributed, but I will change the code to turn the process easy.

2. It is writed Portuguese.

3. We will translate it soon.

4. The first version was made in my master degree:

   <http://www.dpi.ufv.br/arquivos/ppgcc/dissertacoes/2011-ms-Jacson_Rodrigues_Correia_da_Silva.pdf>

About

A Genetic Algorithm that tune the Artificial Neural Network Parameters (FANN library)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published