search

Limingxing00 / Retinal-Vessel-Segmentation-ISBI2022

Accepted by ISBI 2022
MIT License
40 stars 4 forks source link

RCE模块有一些问题 #1

Closed huTao1030 closed 2 years ago

huTao1030 commented 2 years ago

复现过程中，在RCE模块里面遇到topk的部分会报错 Traceback (most recent call last): File "D:/Users/huTao/Desktop/Retinal-Vessel-Segmentation-ISBI2022-main/train.py", line 259, in train(i) File "D:/Users/huTao/Desktop/Retinal-Vessel-Segmentation-ISBI2022-main/train.py", line 140, in train x1, x2 = model(data, ratio) File "D:\ProgramData\Anaconda3\envs\yolo5\lib\site-packages\torch\nn\modules\module.py", line 889, in _call_impl result = self.forward(*input, *kwargs) File "D:\Users\huTao\Desktop\Retinal-Vessel-Segmentation-ISBI2022-main\network\twonet.py", line 179, in forward x2 = self.transform1(x1, x, ratio) File "D:\Users\huTao\Desktop\Retinal-Vessel-Segmentation-ISBI2022-main\network\twonet.py", line 160, in transform1 img = self.resultsErasing(img, x, ratio) File "D:\Users\huTao\Desktop\Retinal-Vessel-Segmentation-ISBI2022-main\network\twonet.py", line 39, in call max_list = input[n, i_c, ...].flatten().topk(int(f_pself.top_n))[0] # top max value RuntimeError: invalid argument 5: k not in range for dimension at C:/w/b/windows/pytorch/aten/src\THC/generic/THCTensorTopK.cu:24

请问有没有什么解决方法

Limingxing00 commented 2 years ago

您好，因为最近在忙毕业。我需要等到三月三十一号才能重新验证代码。但是您可以尝试将 Retinal-Vessel-Segmentation-ISBI2022/lib/xxx.yaml里面的RCE_RATIO: a改为RCE_RATIO: a/100 进行尝试嘛？

Hello, I'm busy graduating recently. I need to wait until March 31st to re-verify the code. But can you try changing RCE_RATIO: a in Retinal-Vessel-Segmentation-ISBI2022/lib/xxx.yaml to RCE_RATIO: a/100?

Limingxing00 commented 2 years ago

The issue has been unresponsive for a long time, so it is closed.