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

少批量数据有用,大批量数据无效 #39

Open
guanyonglai opened this issue Aug 3, 2020 · 1 comment
Open

少批量数据有用,大批量数据无效 #39

guanyonglai opened this issue Aug 3, 2020 · 1 comment

Comments

@guanyonglai
Copy link

我把non-local 用在reid上面的,在resnet50的前三个layers后面都接了non-local层。用1000个ID进行训练是发现能提高一个点,而且收敛速度比原来快,但是当我用5000个ID训练时发现结果比原来低了一个多点。是不是有哪些参数我需要再调一下?烦请解答哦
PS:我没有发现其他网友提出的W权重为0的情况,我这边W还是挺正常的

@AlexHex7
Copy link
Owner

Hi @guanyonglai . 我没有做过reid的实验,因此不太确定(我猜测可能是跟数据有关)。
我做过其他研究方向的实验,在有些数据集上加入non-local block确实性能会更差一些。

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