Clarify ImageFolder sample ordering and indexing

[yakshithkd23]

This PR clarifies the ImageFolder documentation by explaining:

• That samples are sorted alphabetically by class and file name.
• That samples from all classes are concatenated into a single list and indexed globally.

Fixes #400

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/vision/9353

• 📄 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 a956b23 with merge base 14e8cb6 ():
💚 Looks good so far! There are no failures yet. 💚

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

[zy1git]

@yakshithkd23 Thanks for the PR! You mentioned that your PR fixes #400, but I don't see the relationship between your PR and #400

Could you please clarify how they are related?

Thank you!

[yakshithkd23]

@zy1git Thanks for the question!

In this PR, I updated the ImageFolder docstring to clarify the behavior discussed in issue #400. Specifically, I added a Notes section explaining that:

Samples are sorted alphabetically by class name and file name.

Samples from all classes are concatenated into a single list, and dataset indexing (e.g., dataset[i]) refers to this global list rather than a per-class subset.

These points were unclear in the previous documentation and were the source of confusion described in #400. No functional changes were made—this PR only improves documentation clarity.

Please let me know if you’d like the notes moved or reworded. Thank you!

[zy1git]

@yakshithkd23 Thanks for your reply. May I confirm that the issue you linked is titled "Padding with reflection [feature request]"? That one is issue #400, which has been closed for a long time.