About: The comparison method of experimental CEGCN and MSSG-UNET in the paper
renqi1998 opened this issue · 1 comments
renqi1998 commented
Hello, I have a question? That is, the comparison method in your paper CEGCN method is lower than the method proposed in this paper, but the actual running code is higher than mSSG-UNET accuracy, why? Thank you very much!
qichaoliu commented
In fact, I have optimised and added some tricks to the released code of CEGCN. But in the two papers, including both the CEGCN and MSSGUNet, I used the basic versions. By the way, the accuracies acquired by the released CEGCN code are also higher than those reported in the original CEGCN paper. The currently released code of MSSGUNet is also simple. If I have free time, I will optimise its code too. Thanks.