/ezCGP

A lightweight and low level framework to Cartesian Genetic Programming (CGP) in Python(3). Georgia Tech students working on using ezCGP to evolve neural network designs.

Primary LanguagePythonMIT LicenseMIT

ezCGP

A low level framework for Cartesian Genetic Programming in Python3.

Be sure to checkout the Wiki for more documentation and helpful guids.

Quickstart

Create environment and download dependencies

$ git clone https://github.com/ezCGP/ezCGP.git
$ cd ezcgp
$ conda env create -f python_envs/conda_environment.yml
$ conda activate ezcgp-py
$ conda deactivate

To run in a single process...

$ conda activate ezcgp-py
$ python main.py -p [problem_file].py

To run with MPI...

$ conda activate ezcgp-py
$ mpiexec -n [number_cores] python main.py -p [problem_file].py

Other arguments for main.py

  • -v or -d to set the logging level to 'verbose' or 'debug'. The latter will not print to stdout.
  • -s [int] to set the random seed
  • -t to change the name of the output folder for this run to lead with the word 'testing'
  • -n [str] to change the name of the output folder for this run to trail with a custom string tag