Input data

[zhentingqi]

Hi! This is an amazing reproduction of DAG-GNN. Just a few questions: could you please help me make sense of the input data? i.e., what does each column/row in the CSV file mean? what does each number mean? thanks!

[ronikobrosly]

Hey thank you for the kind words @zhentingqi ! Hopefully this is helpful.
That's a good question. It's been a while since I looked at this code but I believe each column in the input data represents a unique variable. In other words, each column represents an eventual node in the graph. The rows are the observations for each variable. All variables are categorical, and so the values just correspond to those categorical values. In and of themselves, the values are arbitrary.