-
Notifications
You must be signed in to change notification settings - Fork 1.1k
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
Optimize VISTA3D #8123
Optimize VISTA3D #8123
Conversation
Signed-off-by: binliu <[email protected]>
@binliunls : Why wasn't this PR integrated ? |
Need a reviewer to confirm these changes are reasonale. Will contact with them. |
Hi @heyufan1995 , this PR looks good to me, could you help to double confirm it? |
Hi @binliunls, do we want include this change in 1.4? |
I contacted with he vista3d author @heyufan1995 offline, and he confirmed that the changes look good |
/build |
Fixes #8122 .
Description
As shown in this PR, the memory malloc and mask embedding for-loop are the bottlenecks that caused the vista3d slow inference. Therefore, this PR fixed them by adding the logic for malloc and replacing the for-loop with a tensor multiplication.
Types of changes
./runtests.sh -f -u --net --coverage
../runtests.sh --quick --unittests --disttests
.make html
command in thedocs/
folder.