Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

如何进行FcaNet-TS的实验 #42

Open
Soso-developer opened this issue Mar 16, 2023 · 3 comments
Open

如何进行FcaNet-TS的实验 #42

Soso-developer opened this issue Mar 16, 2023 · 3 comments

Comments

@Soso-developer
Copy link

您好,请问如何进行LF、TS、NAS几个实验呢?尤其是TS,没有太明白code里是怎么实现的

@cfzd
Copy link
Owner

cfzd commented Mar 17, 2023

@Soso-developer
如果要跑LF的实验,那就是只选低频信号就好了,就行这个地方:

FcaNet/model/layer.py

Lines 15 to 17 in aa5fb63

elif 'low' in method:
all_low_indices_x = [0,0,1,1,0,2,2,1,2,0,3,4,0,1,3,0,1,2,3,4,5,0,1,2,3,4,5,6,1,2,3,4]
all_low_indices_y = [0,1,0,1,2,0,1,2,2,3,0,0,4,3,1,5,4,3,2,1,0,6,5,4,3,2,1,0,6,5,4,3]

如果要跑TS的实验,那么就需要做很多实验了:

  • 获取每个频率的性能,以ImageNet一般7x7的大小为例,总共有49个频谱,那么就拿这49个频段挨个做attention的实验
  • 然后把这49个频段的结果按照性能排序,选取最高的k个,如果k=2,4,8,16的话继续做4次实验
  • 最终得到性能最高的k,这个就是TS的结果

@Soso-developer
Copy link
Author

非常感谢您的回答。也就是说选用top1,挨个选取49个频谱来对比即可?

@yangwanglyb
Copy link

请问,SE注意力中的全局池化对应的是低频信号中的(0,0)吗

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants