To get started, make sure you're using Python 3 and then create an environment using something like
virtualenv env
source env/bin/activate
pip install -r requirements.txt
this will install Tensorflow and Keras (amongst other things), so it may take slightly longer than usual.
Link to the image data in the dropbox:
ln -s ~/Dropbox\ \(Datascope\ Analytics\)/DeepProduct/Data/ Data
With the environment active, view the notebook by launching the jupyter server
jupyter notebook
The code will need to download several files of order ~100MB when being first run, but will cache them for future execution.
For those working on developing the tentatively named couch_tinder application, you should also create a symlink within the static directory. Please call it to_dropbox, otherwise your web app will be Brokeymon.
cd static
ln -s ~/Dropbox\ \(Datascope\ Analytics\)/DeepProduct/Data/ to_dropbox