Hi,
The provided data is bit confusing few clarification will be helpful
1. Inside SleepSource filename is formatted like training_s0r1X.npy what is the significant os "S" and "r"? Does S denote the subject and "r" denote the recording number ?
2. What is the purpose of the data inside : LeaderboardSleep\sleep_target does it come from the subjects aged 65-79 and 80-95 ( the test set distribution) ? Can this be used for training ?
Thanks in advance.
Posted by: arijitx @ Aug. 8, 2021, 8:44 a.m.Thanks for your questions!
Your understanding is correct,'s' for the subject and 'r for recordings. For the 5 subjects with labels, they are the 'example subjects' of that age group to help your algorithm to release the transfer learning around age groups. yes, you can and we recommend using them. You can find the description in tutorial 1. In the data, ' First group contains 5 example subjects with labels, they are the example subjects from the testing group. The second group is the testing group, which contains the leaderboard subjects 6-17 without labels. You need to give predictions for those labels.'
hope that helps.
Best regards,
Xiaoxi Wei