Update introyt tutorial to torchvision v2 APIs

[saurabhkthakur]

Fixes #3858

Description

Updates the Training with PyTorch tutorial by:

• replacing deprecated torchvision transforms with v2 APIs
• replacing ToTensor() with ToImage() and ToDtype(...)
• updating Python 2-style super() usage
• converting .format() calls to f-strings

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

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/tutorials/3865

• 📄 Preview Python docs built from this PR

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

✅ No Failures

As of commit fe8a6a1 with merge base 5efb99a ():
💚 Looks good so far! There are no failures yet. 💚

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

[svekars]

@claude

[claude]

Claude finished @svekars's task in 1m 20s —— View job

---

Review of PR #3865

• Gather context and read the changed file
• Review the diff against origin/main
• Post review feedback

---

Summary

This PR correctly migrates from the deprecated torchvision.transforms to the torchvision.transforms.v2 API, modernizes super() calls, and converts .format() to f-strings. The changes are straightforward and the screenshot confirms successful execution.

Issues Found

1. Stale docstring reference (line 60)

The tutorial prose still says:

We use ``torchvision.transforms.Normalize()`` to
zero-center and normalize the distribution of the image tile content,

Since the code now uses v2.ToImage(), v2.ToDtype(), and v2.Normalize(), this description should be updated to reflect the new transform pipeline. A suggested replacement:

We use ``torchvision.transforms.v2`` to convert images to tensors,
normalize the distribution of the image tile content,

Fix this →

Looks Good

• v2.ToImage() + v2.ToDtype(torch.float32, scale=True) is the correct replacement for the deprecated ToTensor() per the torchvision migration guide.
• v2.Normalize((0.5,), (0.5,)) is unchanged in behavior.
• super().__init__() is the modern Python 3 style — correct.
• f-string conversions are all equivalent to the original .format() calls.

Verdict

Approve with one minor nit: update the tutorial prose at line 60 to not reference the old torchvision.transforms.Normalize() API. Otherwise LGTM.

---

[svekars]

@saurabhkthakur can you update the above comment?