Replies: 8 comments
-
the learning rate seems not right, |
Beta Was this translation helpful? Give feedback.
-
Why does this happen? I did not modify the scheduler settings.
|
Beta Was this translation helpful? Give feedback.
-
I'm afraid this is not the reason. The learning rate becomes zero in the benchmark too. |
Beta Was this translation helpful? Give feedback.
-
Hi @fingertap, |
Beta Was this translation helpful? Give feedback.
-
I really appreciate your work on building a large-scale benchmark like this! I borrow many ideas from mmcv when implementing and arranging my codes. However, the error is quite large to me. 82.25 is way higher than 81.9 on Cityscapes, as the models are approaching the performance ceiling of this dataset. |
Beta Was this translation helpful? Give feedback.
-
We are re-running the model and will feedback to you asap. |
Beta Was this translation helpful? Give feedback.
-
Hi @fingertap, |
Beta Was this translation helpful? Give feedback.
-
Hi @xiexinch , what a huge gap between my runs and yours! Actually, without --deterministic flag, I got an even worse score. I will attach the log later. Any ideas on this? If the device difference has such huge impact, the comparison between methods may be unfair. Actually, an improved version of OCRNet with 40k iters can achieve 81.93 at my machine. Segformer should outperform OCRNet i.m.o. |
Beta Was this translation helpful? Give feedback.
-
I cannot reproduce the 82.25 mIoU of Segformer on Cityscapes at my machine. I use 8 3090, with --seed 0 and --deterministic.
The log is attached.
Any ideas? Thanks in advance.
Beta Was this translation helpful? Give feedback.
All reactions