Skip to content
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

[air/train/benchmark] Add TF GPU 4x4 benchmark #26776

Merged
merged 3 commits into from
Jul 20, 2022

Conversation

krfricke
Copy link
Contributor

Signed-off-by: Kai Fricke [email protected]

Why are these changes needed?

Replaces #26712 after rebase.

Related issue number

Checks

  • I've run scripts/format.sh to lint the changes in this PR.
  • I've included any doc changes needed for https://docs.ray.io/en/master/.
  • I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/
  • Testing Strategy
    • Unit tests
    • Release tests
    • This PR is not tested :(

Signed-off-by: Kai Fricke <[email protected]>
@krfricke
Copy link
Contributor Author

The error we're seeing in the vanilla tf test seems to be related to multi gpu / mirrored strategy training: tensorflow/tensorflow#55250

It's curious though that it doesn't seem to come up for Ray AIR.

Also it could maybe be related to too small batches, so I've increased the batch size and will try to repro.

In any case will mark this test as unstable for now.

Signed-off-by: Kai Fricke <[email protected]>
@krfricke krfricke marked this pull request as ready for review July 20, 2022 17:29
Signed-off-by: Kai Fricke <[email protected]>
@krfricke
Copy link
Contributor Author

I've implemented a retry mechanism for the failing vanilla runs. Here is a run: https://buildkite.com/ray-project/release-tests-pr/builds/10268

Let's see if this passes

@richardliaw richardliaw merged commit 2e35d47 into ray-project:master Jul 20, 2022
avnishn pushed a commit to smorad/ray that referenced this pull request Jul 20, 2022
@krfricke krfricke deleted the air/benchmark-tf-gpu-rebase branch July 20, 2022 22:29
shrekris-anyscale pushed a commit to shrekris-anyscale/ray that referenced this pull request Jul 22, 2022
yaxife pushed a commit to alipay/ant-ray that referenced this pull request Jul 26, 2022
klwuibm pushed a commit to yuanchi2807/ray that referenced this pull request Jul 27, 2022
Catch-Bull pushed a commit to alipay/ant-ray that referenced this pull request Jul 27, 2022
Rohan138 pushed a commit to Rohan138/ray that referenced this pull request Jul 28, 2022
franklsf95 pushed a commit to franklsf95/ray that referenced this pull request Aug 2, 2022
gramhagen pushed a commit to gramhagen/ray that referenced this pull request Aug 15, 2022
gramhagen pushed a commit to gramhagen/ray that referenced this pull request Aug 15, 2022
Stefan-1313 pushed a commit to Stefan-1313/ray_mod that referenced this pull request Aug 18, 2022
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.

2 participants