We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
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
Hi, thanks for releasing the code.
Why do you choose MPI adaptor instead of using default torch.distributed (just like in Mask2Former) ?
MPI adaptor
torch.distributed
The content you are editing has changed. Please copy your edits and refresh the page.
The text was updated successfully, but these errors were encountered:
No branches or pull requests
Hi, thanks for releasing the code.
Why do you choose
MPI adaptor
instead of using defaulttorch.distributed
(just like in Mask2Former) ?Tasks
The text was updated successfully, but these errors were encountered: