sydney0zq / covid-19-detection Goto Github PK

请问hu_window的大小选择有没有参照相关的理论依据？

谢谢

您好！论文中写到groung-truth的masks是通过无监督学习方法得到的，但文章中并没有写到如何实现无监督方法获取masks。代码中也没有找到，请问这一步是在哪里实现的呢？谢谢！

245 行报错 model = import_module(f"model.{CLS_MODEL_UID}")
ModuleNotFoundError: No module named 'model' 。lesion_loc目录下缺少model文件夹

想问一下代码中
preprocess-obtain-lungmasks我用了自己的数据未能得到mask的结果，尝试了多个数据都是一样，想请问一下您我还有什么未注意到的地方吗

请问你们的3D图像的是什么形式的呢，有没有转化为nii形式呢？

Hello first, thank you for making your code available, can you help me?

I have a CT dataset and I don't know where I should put the images inside your project to run the code, can you explain what steps I should follow?

Thanks!

Where can I find the images? NpyData directories contain only empty files

when I run train.py, there is a error No such file or directory: 'NCOV-BF/size368x368-dlmask/ImageSets/ncov_train.txt'

Hi,
Thanks for sharing the code. But the train.py for deCoVnet is missing.

我使用cropresize文件将图片resize为224*336尺寸，但是在decovnet训练时发生错误：RuntimeError: Given input size: (64x1x56x84). Calculated output size: (64x0x56x84). Output size is too small，问题出现在baseline_i3d.py的248行，请问这可能是什么原因导致的

Hi, while train.py from deCoVnet I'm getting an error from dataset_ops.py at line 108
# Color Transforms def colorop(img, bright, contrast): _img = TF.adjust_brightness(img, bright) _img = TF.adjust_contrast(_img, contrast) return _img
the error is saying

ValueError: image has wrong mode

Image is a PIL image and bright is a float