Jupyter notebook of my autoencoder presentation
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
data initial commit Oct 30, 2016
imgs initial commit Oct 30, 2016
Autoencoders.ipynb explain that code snippet is not runnable Dec 4, 2016
README.md Update README.md Nov 2, 2016
deep_autoencoder_example.py.ipynb adding deep autoencoder example Jan 30, 2017



Slides for my autoencoder reading group seminar. Code based heavily on the keras library examples and keras blog (blog.keras.io).

Notebook can be viewed in github by clicking on Autoencoders.ipynb

or if you just want to look at the slide version: http://www.birving.com/presentations/autoencoders/index.html

How to run these slides and examples yourself

Setup python environment

  • Jupyter notebook
  • Requirements: numpy, keras, theano, jupyter
  • Python 3
  • Probably Python 2 as well but I haven't bothered to test -- everyone should be using python 3 ;)
  • Install RISE for an interactive presentation viewer


# Clone the repository
git clone https://github.com/benjaminirving/mlseminars-autoencoders
cd mlseminars-autoencoders


jupyter notebook

View in presentation mode using RISE
After running the notebook, click on the RISE button


jupyter nbconvert --to slides Autoencoders.ipynb --post serve

(for a non interactive version)