I. Valera and Z. Ghahramani, 
 "Automatic Discovery of the Statistical Types of Variables in a Dataset", 
 34th International Conference on Machine Learning (ICML 2017). Sydney (Australia), 2017.

function simLik(datasetC,Nits,KK,it)
%% runs proposed Bayesian method to infer the datatypes in a dataset.
% Inputs:
%   datasetC: name of the dataset to be inferred
%   Nit: number of interations of the Gibbs sampler
%   KK: low rank representation complexity (i.e., number of features)
%   itt: number of simulation
% Outputs: returns void but saves a file with the restuls, i.e., the
% test log-likelihood adn a vector with the inferred weights for the
% different datatypes in each dimension.

- Matlab 2012b or higher
- GSL library
    In UBUNTU: sudo apt-get install libgsl0ldbl or sudo apt-get install libgsl0-dev
- GMP library
    In UBUNTU: sudo apt-get install libgmp3-dev

Isabel Valera: isabel.valera.martinez (at) gmail.com