Visual-Attention-Network/SegNeXt

Random seeds variance

baize97 opened this issue · 1 comments

baize97 commented

When I try the SegNeXt training with different random seeds, I find that the mIoU will have a large variance (up to 1% to 2%).

Tiny
image

Base
image

How do you think such impact on the comparison with other models and the reproducibility of results?

MenghaoGuo commented

Yes, it depends on benchmark. In all dataset, ADE20K dataset has the largest variance. If you want to get stable result, trying to use cityscapes.