Project-MONAI/research-contributions

difference brats2021 train test

Eight3H opened this issue · 0 comments

Brats stated in the paper that the number of experimental training rounds is 800, but in the code it is 300 rounds. Is there any reason for this? Additionally, the reasoning for sliding window approach with an overlapping of 0.7 in test.py is 0.6 in the code. Is there any reason for this?