Skip to content

Add Acelerator API to ddp_tutorial #3508

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
Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Add Acelerator API to ddp_tutorial #3508

wants to merge 1 commit into from
+15 −6

Conversation

jafraustro
Copy link
Contributor

@jafraustro jafraustro commented Aug 5, 2025 •
edited by pytorch-bot bot
Loading

Motivation

  • Refine the DDP tutorial to make it device-agnostic by using the Accelerator API.

Description

  • Integrate the Accelerator API into the DDP tutorial to support multiple accelerators and improve backend initialization.
  • Align the implementation with this example from the PyTorch examples repository.

Checklist

  • The issue that is being fixed is referred in the description (see above "Fixes #ISSUE_NUMBER")
  • Only one issue is addressed in this pull request
  • Labels from the issue that this PR is fixing are added to this pull request
  • No unnecessary issues are included into this pull request.

CC: @gujinghui

cc @wconstab @osalpekar @H-Huang @kwen2501

@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Aug 5, 2025 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/tutorials/3508

Note: Links to docs will display an error until the docs builds have been completed.

⏳ No Failures, 15 Pending

As of commit 5f4478e with merge base fc8981b (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@meta-cla meta-cla bot added the cla signed label Aug 5, 2025
@sekyondaMeta sekyondaMeta requested a review from fegin August 6, 2025 16:32
@sekyondaMeta
Copy link
Contributor

@fegin could you take a look at this when you get a chance

@fegin
Copy link

fegin commented Sep 10, 2025

Sorry for missing this earlier. The tutorial looks good to me. As long as the example doesn’t break any existing CUDA use cases, I think it’s fine. I just wanted to confirm.

@jafraustro
Copy link
Contributor Author

Sorry for missing this earlier. The tutorial looks good to me. As long as the example doesn’t break any existing CUDA use cases, I think it’s fine. I just wanted to confirm.

Hi @fegin, the changes do not break the CUDA case. The example https://github.com/pytorch/examples/blob/main/distributed/ddp/example.py is derived from the tutorial and incorporates the Accelerator API effectively.

@fegin
Copy link

fegin commented Sep 12, 2025

Can you rebase to the main?

@fegin
Copy link

fegin commented Sep 18, 2025

@jafraustro I could not merge it since it is out-of-date again. Sorry about this.

@jafraustro
Add Acelerator API to DDP tutorial to support multiple accelerators a…
5f4478e 
…nd improve backend initialization

Signed-off-by: jafraustro <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@fegin fegin fegin approved these changes

@svekars svekars svekars approved these changes

Assignees
No one assigned
Labels
cla signed distributed
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@jafraustro @sekyondaMeta @fegin @svekars