Update device selection to use torch.accelerator for device-agnostic code

[scinqyz]

Fixes #3876

Description

Updates suboptimal torch.cuda.is_available(), that checks only for CUDA while ignoring other accelerators, with device-agnostic torch.accelerator.current_accelerator().type.

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.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/tutorials/3883

• 📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• Long GPU queue (g5, g6) on LF fleet

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

[meta-cla]

Hi @scinqyz!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at [email protected]. Thanks!

[meta-cla]

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks!

[sekyondaMeta]

@claude

[claude]

Claude finished @sekyondaMeta's task in 1m 15s —— View job

---

PR Review: Update device selection to use torch.accelerator

• Gather context and review the PR diff
• Review code changes for correctness and style
• Post review feedback

---

Summary

The intent of this PR is good — updating from torch.cuda.is_available() to the device-agnostic torch.accelerator API aligns with the broader modernization effort across the tutorials repo. However, the change as implemented is incomplete and introduces a regression.

Issues

1. Missing tensor move and print statement (Bug)

The original code:

if torch.cuda.is_available():
  tensor = tensor.to('cuda')
  print(f"Device tensor is stored on: {tensor.device}")

The new code:

device = torch.accelerator.current_accelerator().type if torch.accelerator.is_available() else 'cpu'

The PR sets a device variable but never uses it. The tensor is no longer moved to the accelerator, and the print statement showing the device is removed. This means the tutorial no longer demonstrates how to move a tensor to a device — which is the entire purpose of this section (the comment above says "We move our tensor to the GPU if available").

2. Suggested fix

The replacement should be:

device = torch.accelerator.current_accelerator().type if torch.accelerator.is_available() else "cpu"
tensor = tensor.to(device)
print(f"Device tensor is stored on: {tensor.device}")

This keeps the tutorial's teaching purpose intact (demonstrating device transfer) while adopting the device-agnostic API.

Fix this →

---

[sekyondaMeta]

@scinqyz See review for bug to fix