Closed Jankin66 closed 4 months ago
The split was done randomly: 80% train, 10% val, 10% test. I have added the exact split used here: data/pancreas/. Regarding the left atrium dataset, sadly, it is an in-house dataset at the University of Utah that I cannot publicly share.
I got it! Thank you so much!!!
I follow the script download.py to download the Pancreas dataset, but I do not exactly know how you split the dataset to training/testing. Besides, would you mind further providing the dataset of Left Atrium which you experiment in your paper.