https://github.com/kstaats/karoo_gp

A Genetic Programming platform for Python with TensorFlow for wicked-fast CPU and GPU support.
https://github.com/kstaats/karoo_gp

Last synced: about 2 months ago
JSON representation

A Genetic Programming platform for Python with TensorFlow for wicked-fast CPU and GPU support.

• Host: GitHub
• URL: https://github.com/kstaats/karoo_gp
• Owner: kstaats
• License: other
• Created: 2015-10-22T10:29:52.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2022-10-31T06:43:54.000Z (over 1 year ago)
• Last Synced: 2024-03-30T23:40:20.187Z (3 months ago)
• Language: Python
• Homepage:
• Size: 5.75 MB
• Stars: 155
• Watchers: 28
• Forks: 62
• Open Issues: 21
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Lists

• awesome-python-data-science - karoo_gp - A Genetic Programming platform for Python with GPU support. <img height="20" src="img/tf_big2.png" alt="sklearn"> (Genetic Programming / Others)
• awesome-python-data-science - karoo_gp - A Genetic Programming platform for Python with GPU support. <img height="20" src="img/tf_big2.png" alt="sklearn"> (Genetic Programming / NLP)

README

        
# Karoo GP

Karoo GP is an evolutionary algorithm, a genetic programming application suite

written in Python which supports both symbolic regression and classification

data analysis. It has been used in radio astronomy, gravitational wave detector

characterisation and synthetic supernovae detection, and a variety of other use

cases in a diversity of fields.

You need only prepare your dataset according to the User Guide. No programming

required. Karoo is multicore and GPU enabled by means of the powerful library

TensorFlow. Karoo has three text cases built-in: Iris dataset, Kepler's law of

planetary motion, and a maths problem you can modify to various degrees of

challenge.

Karoo is launched from the command line with an intuitive user interface or

with arguments for full automation from bash or another Python script. The

output of each run is automatically archived and includes the configuraiton, a

summary, and the full suite of GP trees saved as .csv files for your review and

edit such that you can hand-build the starting block for your next run.

Be certain to read the User Guide for a starter's guide to Genetic Programming

and examples of all you can do with this unique body of code.

For an interesting read on scalar vs vector and CPU vs GPU performance with

Karoo GP: https://arxiv.org/abs/1708.03157 or to learn how Karoo applied to

supernova detection at LIGO: https://arxiv.org/abs/2002.04591

Learn more at kstaats.github.io/karoo_gp/ ...