Question Regarding Data Handling in REFUGE2 Dataset and Preprocessing in Trainning
#8
by
kato0116
- opened
I have a question regarding the use of the REFUGE2 dataset in your work.
The REFUGE2 dataset provides a total of 2000 fundus images (1200 for training, 400 for validation, and 400 for testing). However, in your paper, it is mentioned that 1200 images were used. Could you clarify how the dataset was handled in this case? Specifically:
How does the Adapter process or split the REFUGE2 dataset?
Since the REFUGE2 dataset consists of full fundus images, did you apply any cropping as a preprocessing step before training?
Thank you very much for your time and clarification.
Best regards,