Update trainingyt.py: Migrate to transforms.v2, f-strings, and modern syntax

[Sai-prashanth-2006]

Fixes #3858

Description

This PR modernizes trainingyt.py by resolving deprecated API usage and aligning the codebase with current Python and PyTorch best practices:

• Migrated the torchvision.transforms pipeline to the v2 API, replacing the deprecated ToTensor transform
• Updated GarmentClassifier.__init__ to use the Python 3 super().__init__() syntax
• Replaced all .format() string literals with f-strings for improved readability and performance

No functional behavior is changed by this PR.

Checklist

• The issue being fixed is referenced in the description
• This PR addresses a single issue
• Labels from the original issue have been added to this PR
• No unrelated changes are included

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/tutorials/3871

• 📄 Preview Python docs built from this PR

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

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

[meta-cla]

Hi @Sai-prashanth-2006!

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 cla@meta.com. 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 Review this PR and compare it to the original issue #3858

[claude]

Claude finished @sekyondaMeta's task in 3m 39s —— View job

---

Review: PR #3871 vs Issue #3858

• Read the original issue Update Intro YouTube Training with PyTorch tutorial — v2 transforms, super(), formatting #3858
• Read the PR diff
• Read the full file for context
• Compare PR changes to issue requirements
• Provide review feedback

---

Summary

This PR aims to fix issue #3858 (deprecation of transforms.ToTensor()) by migrating to transforms v2, converting .format() to f-strings, and modernizing super() calls. The intent aligns with the issue, but there is a critical bug in the import that will break the tutorial at runtime.

---

Critical Issue: Incorrect Import

File: beginner_source/introyt/trainingyt.py, line 68

# What the PR does (WRONG):
import torchvision.transforms as v2

# What it should be:
from torchvision.transforms import v2

The PR simply renames the alias of the old torchvision.transforms module to v2. However, ToImage and ToDtype only exist in torchvision.transforms.v2, not in torchvision.transforms. This means v2.ToImage() and v2.ToDtype(...) on lines 76-77 will raise AttributeError at runtime.

The correct pattern (already used in this repo at beginner_source/serving_tutorial.py) is:

from torchvision.transforms import v2

Fix this →

---

Minor Issue: Pointless f-string

File: beginner_source/introyt/trainingyt.py, line 122

# PR changes this:
print('  '.join(classes[labels[j]] for j in range(4)))

# To this:
print('  '.join(f'{classes[labels[j]]}' for j in range(4)))

Wrapping classes[labels[j]] in an f-string (f'{classes[labels[j]]}') adds no value — it's equivalent to str(classes[labels[j]]), and since classes already contains strings, this is a no-op. The original code was correct. This change should be reverted.

---

What's Correct

• super().__init__() (line 139): Good modernization, this is standard Python 3 style.
• All other f-string conversions (lines 94-95, 181, 256, 281, 289, 310, 322): Correct and improve readability.
• v2.ToImage() + v2.ToDtype(torch.float32, scale=True) replacing ToTensor(): This is the correct v2 equivalent — once the import is fixed.

---

Comparison to Issue #3858

Based on the PR description, issue #3858 requests fixing the deprecated transforms.ToTensor() usage. The PR addresses this correctly in concept (replacing ToTensor with ToImage + ToDtype), but the broken import means it won't actually work. The additional f-string and super() changes are reasonable modernization but go beyond the strict scope of the deprecation fix.

---

Verdict

Needs changes before merge:

1. Must fix: Change import torchvision.transforms as v2 → from torchvision.transforms import v2
2. Should fix: Revert the pointless f-string on line 122 back to classes[labels[j]]

---

[Sai-prashanth-2006]

thanks for catching that, import fixed and f-string reverted