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

关于内存 #9

Open
KevinLight831 opened this issue Mar 29, 2021 · 3 comments
Open

关于内存 #9

KevinLight831 opened this issue Mar 29, 2021 · 3 comments

Comments

@KevinLight831
Copy link

作者可以透漏一下这个实验训练测试中所需要的内存吗?还有来自VLP巨大的region feature和之前SCAN中用的pre-comp feature有什么不同,只是单纯的框数量增到100吗?我看网络流程图中检测regions是用fasterRCNN,我以为是和前人一样用的BUTD的pre-comp,实际上是用的VLP的100框对吗?这个对于效果影响大吗?可以换回之前的pre-comp吗?因为我实验环境内存有限。🤦‍♂️允悲

@KevinLight831
Copy link
Author

DSRAN/data.py

Line 151 in 630d9dc

vis_pe = torch.randn(100,1601 + 6) # no position information

为什么GRU版本的data.py没用position information,但是BERT版本的data_bert.py用了

@kywen1119
Copy link
Owner

因为发现这个对结果影响不大

@liuyyy111
Copy link

你用SCAN的feature train出来了吗

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