Dataset split

[tommasocarraro]

Dear researchers, thank you for your great job!

I would like to compare your model with a model that I'm trying to implement. Since this model is really different from yours, it doesn't take features as input. For this reason, I kindly ask you to provide me with the datasets in the standard structure. I mean that the interactions should be divided into train and test split (your exact split), but instead of the features, I would like to have the original contextual fields, like in the original csv files of the datasets. In other words, I need the same datasets you provided with decoded one-hot encoded contextual fields.

Is it possible for you to help me with that? If not, I won't be able to compare my method with yours. Thank you!