[1/N] Refine beginner tutorial by accelerator api

[guangyey]

For #3113

Motivation

As titled, we aim to use accelerator API to refine the build model tutorial to make it device-agnostic.

Additional Context

This is my first PR to pytorch/tutorials

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 @subramen @albanD @jbschlosser

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/tutorials/3167

• 📄 Preview Python docs built from this PR

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

❌ 3 New Failures

As of commit e57d5f2 with merge base 540bd0c ():

NEW FAILURES - The following jobs have failed:

• Build tutorials / pytorch_tutorial_build_worker (13, 15, linux.4xlarge.nvidia.gpu) (gh)
  OSError: bert-base-uncased is not a local folder and is not a valid model identifier listed on 'https://huggingface.co/models'
• Build tutorials / pytorch_tutorial_build_worker (14, 15, linux.4xlarge.nvidia.gpu) (gh)
  AttributeError: module 'torch' has no attribute 'accelerator'
• Build tutorials / pytorch_tutorial_build_worker (2, 15, linux.g5.4xlarge.nvidia.gpu) (gh)
  RuntimeError: f2() Expected a value of type 'Tensor (inferred)' for argument 'y' but instead found type 'int'.

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

[guangyey]

@malfet @svekars, I was wondering if you could help review this PR.

[svekars]

Thank you for the fix. Can you check other tutorials to see where it needs to be updated? Also, we'll have to wait to merge this until 2.6 is available. We'll test against the 2.6-TEST branch after the first RC.

[svekars]

Suggested change

	# We want to be able to train our model on an `accelerator <https://pytorch.org/docs/stable/torch.html#accelerators>`
	# We want to be able to train our model on an `accelerator <https://pytorch.org/docs/stable/torch.html#accelerators>`__

[svekars]

Suggested change

	# like CUDA, MPS, MTIA, or XPU. If the current accelerator is available, we will use it, otherwise we use the CPU.
	# such as CUDA, MPS, MTIA, or XPU. If the current accelerator is available, we will use it. Otherwise, we use the CPU.

[guangyey]

Thanks, @svekars I will continue checking the other tutorials.

[albanD]

Sounds good!