This archive includes: 

1) the matlab/octave file "StoianovZorzi2012_data.mat", which contains unlabeled binary images and 
   corrisponding labels (in a separate matrix). The latter could be used for further analysis. 

2) Matlab/Octave script "stoianovzorzi2012_converter.m" that converts the unlabeled training data of Stoianov & Zorzi (2012)
   into 3D unlabeled training data for deep belief networks, and shows a random image and its parameters. 

   Author of this MATLAB/Octave script: 
%  Ivilin P. Stoianov, Computational Cognitive Neuroscience Lab, 
%  Department of General Psychology, University of Padova, Italy
%  Contact: ivilin.stoianov@gmail.com
%  Websites: http://ccnl.psy.unipd.it   &  http://www.stoianov.it

  Granted permissions:
% (a) to use the image data contained in "StoianovZorzi2012_data.mat" and this script if citing the article 
%     Stoianov, I., and Zorzi, M. (2012). Emergence of a “visual number sense” in hierarchical generative models.
%     Nature neuroscience 15, 194–6.
% (b) to use the parallel learning algorithms provided in the site http://ccnl.psy.unipd.it/research/deeplearning 
%     if citing the corresponding article: Testolin, Stoianov, De Filippo De Grazia, and Zorzi (2013).
%     "Deep unsupervised learning on a desktop PC: A primer for cognitive scientists", Frontiers in Cognitive Science.

  Date: 12 April 2013, Padova, Italy

