Crate neuralneat

neuralneat is an implementation of the NeuroEvolution of Augmenting Topologies (NEAT) described in the June 2002 paper titled “Evolving Neural Networks through Augmenting Topologies” by Kenneth O. Stanley and Risto Miikkulainen.

Much of this implementation was also guided by the NEAT 1.2.1 source code.

Modules

• activation
  The activation module contains basic activation functions that can be used with hidden and output layers of a Genome.
• defaults
  The defaults module contains the default values of all of the constants used by the Pool to create, mutate, and mate Genomes.

Structs

• Genome
  A Genome contains a network of Network of Neurons and Genes that when given the same number of inputs the containing Pool was given, will produce a Vec with the same number of outputs that the Pool was given.
• GenomeStats
  Basic statistics about a Genome’s fitness.
• Pool
  A “Gene” Pool that contains and manages a population of Genomes separated into one or more Species.
• PoolStats
  Basic statistics about a Pool and the Species and Genomes contained within it.
• Species
  A Species contains a number of Genomes that share common structure.
• SpeciesStats
  Basic statistics about a Species and the Genomes contained within it.