Question about dataset split method for comparative experiments

[SlamCabbage]

Hi,

I'm currently working on a comparative study related to E2E3DR and your method has drawn my great attention. When reproducing and comparing with your results, I have two questions about the dataset split strategy:

1. The paper only reports the size of the training and evaluation sets, but lacks details about the specific split method. Could you please elaborate on how the dataset was partitioned? For example:
   · Is it a random split, stratified split, or split based on specific rules (e.g., cross-validation folds, time sequence)?
   · Was a fixed random seed used to ensure reproducibility?
2. To ensure the fairness of the comparative experiment, I would be very grateful if you could share the exact dataset split scheme (e.g., split script, index files of train/eval subsets, or detailed step-by-step instructions).

Thank you very much for your excellent work and open-source contribution. Looking forward to your reply!
Best regards
llz