Tony Cao

Thank you for your reply. Well, that is a little weird then, I'm reproducing a keras code and the author used 'Renorm' in BatchNormalization. In fact, if I run the...

Thanks for your reply. The repository of the code I want to reproduce is https://github.com/jpviguerasguillen/corneal_endothelium_TVST. The part including `renorm` is: ```python def conv_block(x, growth_rate, name, data_format='channels_last'): """ A building block...

Still error there, maybe you can pack up a colab demo so that it can be reproduce more easily ``` Traceback (most recent call last): File "main.py", line 389, in...

> @caolonghao you can still use the default mmcv distributed in the code but with modifications from [open-mmlab/mmdetection#10720](https://github.com/open-mmlab/mmdetection/issues/10720) and [HarborYuan/mmcv_16@ad1a72f](https://github.com/HarborYuan/mmcv_16/commit/ad1a72fe0cbeead2716706ff618dfa0269d2cf4c). Then you should be good to go. Thanks, this solved...

> [@KumoLiu](https://github.com/KumoLiu) Opened a thread for this as discussed. Happy to help make the contribution if it aligns with the Auto3DSeg team's goals as I'm currently working on some Auto3DSeg...

Thank you for your reply, I'll try that.

By the way, is there any idea to do binary prediction with sccan? It is good to do regression, but my goal is classification. I don't know if it reasonable...

> Hmm, the online demo runs vllm vs sglang, but otherwise should be identical. There is some randomness is sampling, ex. have you tried running the PDF through several times...

2.0.0 preview 版本似乎解决了这个问题，等待进一步测试确认