Skip to content

Setting seeds for test_batched_nms_implementations #4766

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 11 commits into from
Oct 28, 2021

Conversation

jdsgomes
Copy link
Contributor

@jdsgomes jdsgomes commented Oct 27, 2021

Fixes #4506

Checked locally 10k seeds and observed 6 fails: 1759, 3698, 3813, 4417, 8600, 8969

I am fixing the seed to avoid failures in the future despite they are seemingly rare.

cc @pmeier

@facebook-github-bot
Copy link

facebook-github-bot commented Oct 27, 2021

💊 CI failures summary and remediations

As of commit 16ffae3 (more details on the Dr. CI page):


💚 💚 Looks good so far! There are no failures yet. 💚 💚


This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

@jdsgomes jdsgomes requested a review from NicolasHug October 27, 2021 14:33
Copy link
Member

@NicolasHug NicolasHug left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks a lot @jdsgomes , LGTM when green :)

Checked locally 10k seeds

wow!

@datumbox
Copy link
Contributor

datumbox commented Oct 27, 2021

Awesome! Now we know this is due to the unstable sort. We actually get duplicate scores:

torch.random.manual_seed(1759)
...
len(set(scores.tolist()))
999

Changing to stable sort might help but we had been facing issues while trying to do it: #4491

Might worth putting more time on it.

@jdsgomes jdsgomes merged commit c786d12 into pytorch:main Oct 28, 2021
facebook-github-bot pushed a commit that referenced this pull request Nov 1, 2021
Summary:
* adding multiweight support for deeplabv3 prototype models

* adding default values for optional params

* fixing bug

* addressing PR comment

* fixing seed in test_batched_nms_implementations

* change seeds in test_batched_nms_implementations

* change seeds in test_batched_nms_implementations

Reviewed By: datumbox

Differential Revision: D32064711

fbshipit-source-id: f88cac21cb8a906237a9ef26bb53acca047dcace
cyyever pushed a commit to cyyever/vision that referenced this pull request Nov 16, 2021
* adding multiweight support for deeplabv3 prototype models

* adding default values for optional params

* fixing bug

* addressing PR comment

* fixing seed in test_batched_nms_implementations

* change seeds in test_batched_nms_implementations

* change seeds in test_batched_nms_implementations
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Fix flaky tests that are recently popping up
4 participants