Skip to content

fix(datasets): SUN397 mirror fallbacks and clearer download errors#9528

Open
Nueramarcos wants to merge 7 commits into
pytorch:mainfrom
Nueramarcos:fix/sun397-download-mirrors
Open

fix(datasets): SUN397 mirror fallbacks and clearer download errors#9528
Nueramarcos wants to merge 7 commits into
pytorch:mainfrom
Nueramarcos:fix/sun397-download-mirrors

Conversation

@Nueramarcos

Copy link
Copy Markdown

Summary

The Princeton SUN397 download URL (vision.princeton.edu/.../SUN397.tar.gz) now 404s after HTTP→HTTPS redirects to cs.princeton.edu. MIT CSAIL mirrors (groups.csail.mit.edu/vision/SUN1old/SUN397.tar) are also down.

This PR:

  • Tries HTTPS then HTTP Princeton mirrors (MNIST-style fallback loop)
  • Raises an actionable RuntimeError listing failed URLs and pointing users to manual download + Torchvision dataset mirrors #7637 mirroring discussion when all mirrors fail

Fixes #9348

Test plan

  • SUN397TestCase (fake data injection) unchanged — no network in unit tests
  • Manually verified redirect chain: both mirrors return 404 today; error message now surfaces all attempts

Note

A full fix likely needs an official PyTorch dataset mirror (see #7637). Happy to help host/upload if maintainers want a download.pytorch.org path.

Nueramarcos and others added 7 commits June 18, 2026 07:04
* docs: Add fork notice to README

Co-authored-by: aider (ollama/qwen2.5-coder:7b) <aider@aider.chat>

* fix: correct typo in README.md

Co-authored-by: aider (ollama/qwen2.5-coder:7b) <aider@aider.chat>

* chore: sanitize artifacts for Add fork notice to README

---------

Co-authored-by: aider (ollama/qwen2.5-coder:7b) <aider@aider.chat>
Co-authored-by: aider (ollama/qwen2.5-coder:7b) <aider@aider.chat>
Co-authored-by: aider (ollama/qwen2.5-coder:7b) <aider@aider.chat>
Co-authored-by: aider (ollama/qwen2.5-coder:7b) <aider@aider.chat>
Princeton's HTTP mirror now 404s after redirects. Try HTTPS then HTTP,
and raise a actionable RuntimeError pointing to pytorch#9348 and dataset-mirror
discussion when all mirrors fail.

Fixes pytorch#9348
@pytorch-bot

pytorch-bot Bot commented Jun 25, 2026

Copy link
Copy Markdown

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/vision/9528

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:

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

@meta-cla

meta-cla Bot commented Jun 25, 2026

Copy link
Copy Markdown

Hi @Nueramarcos!

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!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

The URL for the SUN397 dataset isn't working

1 participant