Pretrain-model get higher accuracy than backbone.

[byhwhite]

Dear author: Appreciate to your amazing work first. I run your code on pancreas dataset, pretrain model could get 76.62% accuracy on test set. While the TABLE.Ⅴ in paper shows that the V-net accuracy is 69.96%, and I'm confused about the difference. Hope you could help me! Looking forward to your reply!

[koncle]

The accuracy reported in TABLE. V is the performance of the original version of V-Net. We have optimized it for our method, which gives higher accuracy for the backbone. We also test the accuracy of UA-MT with the same setting and does not get higher accuracy.

[wtkszzz]

The accuracy reported in TABLE. V is the performance of the original version of V-Net. We have optimized it for our method, which gives higher accuracy for the backbone. We also test the accuracy of UA-MT with the same setting and does not get higher accuracy.

Dear author, can you give details of the optimisation on the pancreas dataset, I can only train out the performance of the original version of Vnet using the pancreas dataset, thank you very much!

[GuozhengSui]

I would like to consult you about the missing module in the Train UA mt.py file about Train_Vnet.py. How do you run it successfully？