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

How can we use only 1 GPU for training and testing? For the current code implementation, needed 2 GPUs. #1

Open
HongminWu opened this issue Oct 25, 2020 · 3 comments

Comments

@HongminWu
Copy link

No description provided.

@CZ-Wu
Copy link
Collaborator

CZ-Wu commented Oct 31, 2020

Sorry for my late reply. The current version only supports 2 GPUs for acceleration in grasp proposal phase. I think it's ok to run on one GPU by setting larger forward number at this point:
https://github.com/CZ-Wu/GPNet/blob/59419f4505ca0732316a7d993bfe065547892847/tools/proposal.py#L384
and setting smaller batch size at these two points:
https://github.com/CZ-Wu/GPNet/blob/59419f4505ca0732316a7d993bfe065547892847/lib/gpnet.py#L345
https://github.com/CZ-Wu/GPNet/blob/59419f4505ca0732316a7d993bfe065547892847/lib/gpnet.py#L404
eg. change 5000 to 2000 or smaller.

@HongminWu
Copy link
Author

HongminWu commented Nov 2, 2020 via email

@HongminWu
Copy link
Author

HongminWu commented Nov 4, 2020 via email

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

2 participants