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.

Sign up for GitHub

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

Jump to bottom

Out of memory #289

Open
HarryTanNguyen opened this issue Feb 9, 2023 · 1 comment
Open

Out of memory #289

HarryTanNguyen opened this issue Feb 9, 2023 · 1 comment

Comments

@HarryTanNguyen
Copy link

In your implementation with loss_desc using triplet loss, you compute dot_product, which has a shape of Bs, 30, 40, 30, 40, C. It is quite difficult to use a large batch size and a big C due to the out of memory issue. Do you have any plans to reimplement this to be more GPU friendly?
@rpautrat
Copy link
Owner

rpautrat commented Feb 9, 2023

Hi, no there is no active development for this repo unfortunately. But for me, it was possible to train it with a batch size of 2 or 3, and this was enough to get a reasonable training.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@rpautrat @HarryTanNguyen