Open chenslcool opened 3 years ago
This project is a great framework. But I have some question to ask.
- I notice that in preprocessing phase, you crop each image into a box (larger than patch size(112, 112, 80)) according to label mask and get a LA region. But later, in Dataset LAHeart, you add a random crop transform to crop image to (112, 112, 80) again. Why this? I think just croping to (112, 112, 80) and remove the random crop in transform is enough. Random crop may lose some LA information.
- In the testing phase, why do you make several inference on patches of one image? If croping to (112, 112, 80) in preprocessing , I think it will be easier to make test with no need for patching. Besides, I think this patching and average method used in testing may act like "ensemble learning", which will improve performance.
这个项目是一个很棒的框架。但我有一些问题要问。
- 我注意到在预处理阶段,您根据标签掩码将每个图像裁剪成一个框(大于补丁大小(112、112、80))并获得一个 LA 区域。但后来,在 Dataset LAHeart 中,您添加了一个随机裁剪变换以再次将图像裁剪为 (112, 112, 80)。为什么这个?我认为只需裁剪到 (112, 112, 80) 并删除变换中的随机裁剪就足够了。随机裁剪可能会丢失一些 LA 信息。
- 在测试阶段,为什么要对一张图像的补丁进行多次推断?如果在预处理中裁剪到 (112, 112, 80) ,我认为不需要打补丁就可以更容易地进行测试。此外,我认为在测试中使用的这种修补和平均方法可能会起到“集成学习”的作用,这将提高性能。
- 我认为:第一,这样可以增加训练数据的多样性,第二,可以增加模型的鲁棒性(因为即使我们把预处理切得很好,测试集的数据也可能在这个范围内不完全匹配(112 , 112, 80))。
- “集成学习”,我想你已经回答了你的问题。😄
你好!请问你用其它数据集跑过这个项目吗?
This project is a great framework. But I have some question to ask.