EA in Erlang
Erlang Shell Python
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.



Evolutionary Algorithm(s) in Erlang.

genetica is a finished project, but is neither idiomatic nor efficient: It was designed for learning both Erlang and subsymbolic AI. Considering its current state, you should not use this project for anything — It is just here so that I can be nostalgic in the future.

How to add in new problems to solve

Genetica has a framework which allows one to create new modules and let them be used in the main framework. This is done by following a protocol, which requires following functions to be available: parse_args, a function taking the arguments given to this module and converts it into "proper" arguments. random_genotype_fn takes parsed argument as input and returns a function generating random genotypes. genotype_to_phenotype_fn takes parsed arguments as input and returns a function which converts a genotype to a phenotype (A process). mutation_fn takes the input arguments as input and returns a mutated genotype. analyze_fn takes the input arguments as input, and returns a function which takes a list of phenotypes coupled with their fitness, and returns a list of numbers which may later be processed.

Phenotypes as processes

Phenotype ! {Pid, {gtype}}

Sends a message to Pid on the form {Phenotype, gtype, Genotype}, where Genotype is the genotype of the Phenotype.

Phenotype ! {Pid, {fitness, PopulationPids}}

Sends a message to Pid on the form {Phenotype, fitness, Fitness}, where Fitness is the fitness of the Phenotype.

Phenotype ! {Pid, {crossover, OtherPhenotype}}

Performs a crossover of Phenotype and OtherPhenotype, and sends the genotype of the crossover as a message on the form {Phenotype, crossover, Genotype} to Pid.

Phenotype ! {Pid, {shutdown}}

Sends the message {Phenotype, shutdown} to Pid and shutdowns.


Copyright © 2013 Jean Niklas L'orange

Distributed under the Eclipse Public License, the same as Clojure.