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

Where is the skeletonMix SkeletonMix strategy used in this paper by ta-cnn in the code #1

Open
wys2929 opened this issue Sep 15, 2022 · 4 comments

Comments

@wys2929
Copy link

wys2929 commented Sep 15, 2022

A skeletonMix strategy is used in the tacnn paper, but I didn't see this part of the code when I read the code of tacnn. I didn't find it. Could you please point me to which lines of code in which file?
thank you very much

@xiaoyong
Copy link
Collaborator

In this repo, we only port the Ta-CNN network architecture, which is the major contribution of the paper. Implementation of SkeletonMix is left as future work. And any contribution from the community is welcome.

@wys2929
Copy link
Author

wys2929 commented Sep 17, 2022 via email

@KailinXu
Copy link

Hello, the skeletonMix is implemented based on the 8th formula in the Ta-CNN's paper, and 1/16 of samples in a batch is generated with skeletonMix. The data is generated by randomly combining the upper body and the lower body of two different samples, and the label is the weighted sum of the one-hot encodings.
Both the code and the result are easy to reproduce. If you have any questions, please contact with me.

@shang766
Copy link

Hello, the skeletonMix is implemented based on the 8th formula in the Ta-CNN's paper, and 1/16 of samples in a batch is generated with skeletonMix. The data is generated by randomly combining the upper body and the lower body of two different samples, and the label is the weighted sum of the one-hot encodings. Both the code and the result are easy to reproduce. If you have any questions, please contact with me.

您好 非常感谢您提供这么优质的一篇论文 其中我对skeletonMix策略很感兴趣,看了您提供的公式,但是还是不知道如何实现 如果可以的话 您方便提供一下源码嘛?非常感谢!

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

4 participants