Stacked Autoencoder Neuroevolution

Paper

saene.pdf is a shortened version of the master thesis as a research paper.

Stacked Autoencoder Neuroevolution (SAENE)

This project implements a neuroevolution strategy for autoencoders. It implements a stacked autoencoder model using tensorflow.

Requirements

To run the code python3.6 is required. All package requirements are listed in the requirements.txt

Getting Started

I recommend to run this code in a virtual python environment. To do this, install virtualenv using pip with pip3 install virtualenv on Linux.

To create the virtual environment use virtualenv on Linux or venv on MacOs:

virtualenv -p python3 machine-learning
python3 -m venv machine-learning

Activate the virtual environment by running source machine-learning/bin/activate

Next install the requirements from the requirements.txt pip install -r requirements.txt

If the requirements change the requirements.txt can be updated using pip freeze > requirements.txt

docs

Sphinx project that contains the documentation for the code.

Update the documentation by running make html from the docs/ folder. The virtual environment has to be activated for this to work.

To regenerate the package documenation (after major changes) delete the contents of the source/ directory and run sphinx-apidoc -o source/ ../saene/ from the docs/ directory.