xiaoqi-zhao-dlut/msnet-m2snet

(MICCAI 2021) Automatic Polyp Segmentation via Multi-scale Subtraction Network

Hi 👋, I'm ZXQ

• 📫 How to reach me [email protected]
• 👦 My personal homepage

msnet-m2snet's Issues

evaluate score

Hello author, I would like to ask that in the process of reproducing your code, after using your source code and data set, the same hyperparameters, and the training results for testing, it is estimated that the results in your paper are low, but using your There is nothing wrong with the weights provided. Could you please help me answer my doubts?

读取数据时出错:NoneType is not callable

作者你好，在复现代码时遇到点问题想请教，在从dataloader中读取数据时出错:NoneType is not callable，查看loader的长度是正确的，数据应该是读进去了，顺便问一下这份代码的环境配置是怎样的，谢谢。
Traceback (most recent call last):
File "D:/WH/MSNet-main/train.py", line 80, in
train(dataset_medical, MSNet, LossNet)
File "D:/WH/MSNet-main/train.py", line 60, in train
for step, (image, mask) in enumerate(loader):
File "C:\ProgramData\Anaconda3\envs\MSNet\lib\site-packages\torch\utils\data\dataloader.py", line 352, in iter
return self._get_iterator()
File "C:\ProgramData\Anaconda3\envs\MSNet\lib\site-packages\torch\utils\data\dataloader.py", line 294, in _get_iterator
return _MultiProcessingDataLoaderIter(self)
File "C:\ProgramData\Anaconda3\envs\MSNet\lib\site-packages\torch\utils\data\dataloader.py", line 801, in init
w.start()
File "C:\ProgramData\Anaconda3\envs\MSNet\lib\multiprocessing\process.py", line 105, in start
self._popen = self._Popen(self)
File "C:\ProgramData\Anaconda3\envs\MSNet\lib\multiprocessing\context.py", line 223, in _Popen
return _default_context.get_context().Process._Popen(process_obj)
File "C:\ProgramData\Anaconda3\envs\MSNet\lib\multiprocessing\context.py", line 322, in _Popen
return Popen(process_obj)
File "C:\ProgramData\Anaconda3\envs\MSNet\lib\multiprocessing\popen_spawn_win32.py", line 65, in init
reduction.dump(process_obj, to_child)
File "C:\ProgramData\Anaconda3\envs\MSNet\lib\multiprocessing\reduction.py", line 60, in dump
ForkingPickler(file, protocol).dump(obj)
TypeError: 'NoneType' object is not callable

M2SNet的源码在哪儿呢？

some question

Hello author, first of all thank you for providing such a good job. While reading this article, I noticed that you used a lot of visual features in the Fig 8 , which is a good highlight. I was wondering if you could share your code for feature visualization with me. I will quote your article in my article.

can you share SFA'code ?

Hi, I'm looking for SFAs' code now
but SFA's repo was deleted , can you share SFA's code ?

MSU单位

您好 我看您论文中说的MSU单位是由一个1×1卷积加上一个3×3卷积再加上一个5×5卷积的总和再卷积得到的，但我看您提供的代码里，self.conv_3 = CNN1(64, 3, 1)，self.conv_5 = CNN1(64, 5, 2)，是3×3卷积和5×5卷积 后续相加的时候也没有1×1卷积 是怎么回事啊 或者是不是我理解错误呀 希望您能回复 谢谢

test_score

Hello author, I have recently read MSNET and M2SNET written by you and your team, which benefited me a lot. But when I was looking at the code, I found that the trained model was not used in the test_score.py file. Very confused, can you tell me why? Thank you so much!

The dataset link is expired!

关于用于生成预测图部分的代码的一点问题

前辈您好！最近在研读您的论文（非常精彩的论文o(￣▽￣)ｄ）。然后尝试分析了一下您的代码，但是读到这里遇到了一点问题。

我理解您这里写的这个 crf_refine 方法是用来细化分割结果的对吗？
但是我回看您的论文发现并没有关于这个方法的任何描述，感觉很是奇怪。
请问这个是您用于调试时候留下的还是说您忘记在论文里对这一部分进行描述了？

pretrained models

Hi,
I couldn't find the pre-trained models, "res2net50_v1b_26w_4s-3cf99910.pth", used in this code.
Wish for your sincerely reply!

video polyp segment

I would like to ask the Settings of data sets for video polyp segmentation in the author's experiment. Should the two data sets be used separately for training or mixed together for training? Thank you for your answer