Common video ids between split 0 and split 1 #2
Description
Good day,
I've been exploring this dataset and found collisions in video ids between split 0 and split 1. This may be inconsistent with what was stated in the paper regarding the posted official_splits: "To ensure a fair and robust evaluation, no video is shared between splits." https://www.nature.com/articles/s41597-021-00920-z
Could you shed some light on this? Thank you, and thanks for making your work available publicly.
Commin ids:['7ad22d50ebaf4596', '8885668afb844852', 'ad91cf7ca91440aa', '7a47e8eacea04e64', 'bca26705313a4644', '64440803f87b4843', '8ebf0e483cac48d6']