Skip to content
New issue

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

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

Already on GitHub? # to your account

關於ground truth的decode #32

Open
AlanChiuVP opened this issue Aug 12, 2022 · 1 comment
Open

關於ground truth的decode #32

AlanChiuVP opened this issue Aug 12, 2022 · 1 comment

Comments

@AlanChiuVP
Copy link

請問一下 關於下面這行ground truth

gt = movenetDecode(labels, kps_mask,mode='label')

請問為何需要經過decode??
我的理解是直接讀取jason file得到位置
但這邊跟predict一樣 似乎丟進model去decode
請問有特別的原因嗎? 這樣算loss時會不會有誤差?
謝謝

@fire717
Copy link
Owner

fire717 commented Aug 12, 2022

这里decode只是为了计算acc,用于计算loss的是decode之前的特征图,因此实质上对训练没有任何影响。

因为生成Dataloader的时候的labels转成了heatmap特征图的形式,用于计算loss,因此这里拿到的也是heatmap,计算acc自然就要转换回去。

而且pre和gt都经过decode可以保持流程一致,方便前期debug,前期编写decode流程的时候是先根据gt来测试验证一致(这也说明了不会存在误差。),再改到pre上,因为pre前期波动太大了不好测试。

当然你可以在dataloader多输出一个原始结果,用于计算acc,没有区别。

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

No branches or pull requests

2 participants