Evolutionary Keras: evolutionary strategies for Tensorflow-Keras

https://zenodo.org/badge/DOI/10.5281/zenodo.3630339.svg

This is the documentation for the latest release of the evolutionary_keras python module. This documentation includes a quick how-to use guide, provides several examples and collects changelogs of new releases.

What is evolutionary_keras

Tensorflow and its Keras API are some of the most widely used Machine Learning frameworks available in the market. It is a high-level API written in Python and that can run on multiple backends. The goal of Keras is to be able to build and test new TensorFlow models as fast as possible.

Keras models are trained through the usage of optimizers, all of which are Gradient Descent based. This module deals with that shortcoming of Keras implementing several Genetic Algorithms on top of Keras while keeping the main philosophy of the project: it must be easy to prototype.

Installing evolutionary_keras

evolutionary_keras is available in PyPI, conda-forge.

pip install evolutionary_keras

Furthermore, the code is available under GPL3.0 in github: N3PDF/evolutionary_keras.

How to cite evolutionary_keras?

When using evolutionary_keras in your research please cite the following zenodo publication.

https://zenodo.org/badge/DOI/10.5281/zenodo.3630339.svg
@software{evolkeras_package,
  author       = {Juan Cruz-Martinez and
                  Roy Stegeman and
                  Stefano Carrazza},
  title        = {evolutionary\_keras: a Genetic Algorithm library},
  month        = jan,
  year         = 2020,
  publisher    = {Zenodo},
  version      = {v0.9b2},
  doi          = {10.5281/zenodo.3630339},
  url          = {https://doi.org/10.5281/zenodo.3630339}
}