When I train and test with the resized Glas dataset，I got the dice result 0.71. Howerver，in author's paper the real result should be 0.83. = = Could some one to tell me how to solve this trouble ?
Perhaps my train model wasn't the best，so can you give me a better one ？
When I train and test with the resized Glas dataset，I got the dice result 0.71. Howerver，in author's paper the real result should be 0.83. = = Could some one to tell me how to solve this trouble ?
Perhaps my train model wasn't the best，so can you give me a better one ？