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

实现的疑问 #9

Open
JDanielWu opened this issue Jun 2, 2021 · 2 comments
Open

实现的疑问 #9

JDanielWu opened this issue Jun 2, 2021 · 2 comments

Comments

@JDanielWu
Copy link

你好 我想问下,按照原论文来说,应该有个FT(fine tune)过程,就是失真图片进行rank 孪生网络训练后,要用实际的有评分的图片进行fine tune,这样最后输出的是一个评分。你的实现我看到有3个任务,排序对比损失任务,分布的推土距离损失任务,回归损失任务。排序对比损失任务 我感觉比较接近第一步部分,FT 部分是回归损失任务? 我看样例里AVA好像也没用到回归损失。希望有空能解答下,谢谢。

@ayushrox
Copy link

ayushrox commented Jun 6, 2021

the same issue, want to implement original paper in Pytorch but having trouble in understanding in detail like does the siamese network work in the same way, how it is trained and where is the Ranking dataset. Would be grateful if anyone replies. Thank you!

@YilanWang
Copy link

对,ava就是rank的时候用用,rankiqa本质是给tid和live设计的,所以前边rank之后,再在tid上finetune regression任务

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