Test/train split of datasets

[anastasijamalyk]

Hi @zzxslp,
thanks a lot for providing the code to your paper :) I was wondering how the train.txt and test.txt files were created (in both validation and test folders). Since "targets" is a variable, simple 80/20 sequence split might not be sufficient (e.g. if T>20% of user item sequence).
Thanks,
Anastasia