Skip to content

Latest commit

 

History

History
21 lines (11 loc) · 849 Bytes

README.md

File metadata and controls

21 lines (11 loc) · 849 Bytes

epann

Evolutionary Plastic Artificial Neural Networks

A Python package for implementing Compositional Pattern Producing Network (CPPN) variants for indirectly encoding artificial neural networks for neuroevolution and artificial life experiments.
  • Purpose:
  1. Greater understanding of the HyperNEAT algorithm and indirect encoding variants for evolving a populations of neural networks.
  2. Practice in package management.
  3. General purpose research pipeline for appyling EPANN neuroevolution techniques to a variety of environments.
  4. Dissertation code and documentations.

Tutorials

You can find a collection of tutorials meant to introduce the major algorithms and features of the package in the docs/ subdirectory:

Start with the Tutorial Introduction