TensorFlow implemetation of Transforming Auto-encoders

Implemenation of Transforming Auto-encoders (by Hinton et al.) using TensorFlow.

Paper: Hinton, Geoffrey E., Alex Krizhevsky, and Sida D. Wang. "Transforming auto-encoders." International Conference on Artificial Neural Networks. Springer, Berlin, Heidelberg, 2011.

Getting Started

These instructions describe how I setup my environment for this project. There are different ways to get it running using conda or traditional pip.

Prerequisites

• python3
• pipenv
• jupyter
• cuda 9.0 (If you don't have a GPU, you can replace tensorflow-gpu by tensorflow)

Installing

Clone the repository and install the python dependencies in a virtual environment using pipenv:

$ git clone git@github.com:HedgehogCode/transforming-autoencoders-tf.git
$ cd transforming-autoencoders-tf
$ pipenv install -d
$ pipenv shell
$ python -m ipykernel install --user --name transforming-autoencoders --display-name "Python (transforming-autoencoders)"

Running

Start jupyter notebook and open a notebook.

Similar Projects

• yingzha/Transforming-Autoencoders: Theano impemetation
• nikhil-dce/Transforming-Autoencoder-TF
• ndrplz/transforming-autoencoders
• ethanscho/TransformingAutoencoders
• ... (just search for Transfroming Autoencoders on GitHub)

TODOs

• Add other datasets
  • Fashion-MNIST should be easy to use

License

This project is licensed unter the 2-Clause BSD License - see the LICENSE.txt file for details.