Baseline models and source code

The description of the baseline models are provided in the white paper (

Since submissions to Track I include a source code that will be run by the organizers, such submissions should include a “main” wrapper file that follows the exact structure of one of the “main” files presented in the codes attached below. Please use any of the wrapper files supplied and only change the relevant code lines.

There is no need to follow the structure of the code given below for the baseline in Track II, since no source code should be submitted. The codes here are only given for your convenience.

To run the code, please download the zip file of your favorite programming language and unzip to a single folder. It is very much possible you will have to change the path name at the top of the main source code to match the folder in which you installed.

The python code was written in Python 3.6.3 using packages Numpy 0.20.3, Pandas 1.13.3, and SciPy and Scikit-learn 0.19.1.

The source code for one of the baselines in Track II (factorization machine) is currently only available in R due to technical difficulties.


Baseline Models Examples:            Python Version     Matlab version     R version       SAS version

PsychForest:    Python version    Matlab version       R version

Track II:            Python version   R version