Skip to content

[Inductor UT] update rtol for convoluton on XPU.#129782

Closed
etaf wants to merge 4 commits intogh/etaf/26/basefrom
gh/etaf/26/head
Closed

[Inductor UT] update rtol for convoluton on XPU.#129782
etaf wants to merge 4 commits intogh/etaf/26/basefrom
gh/etaf/26/head

Conversation

@etaf
Copy link
Copy Markdown
Collaborator

@etaf etaf commented Jun 28, 2024

[ghstack-poisoned]
@pytorch-bot
Copy link
Copy Markdown

pytorch-bot bot commented Jun 28, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/129782

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

❌ 1 New Failure, 5 Unrelated Failures

As of commit 533a665 with merge base 1ad6830 (image):

NEW FAILURE - The following job has failed:

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

UNSTABLE - The following jobs failed but were likely due to flakiness present on trunk and has been marked as unstable:

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

@etaf etaf marked this pull request as draft June 28, 2024 19:11
@etaf etaf requested a review from EikanWang June 28, 2024 19:11
@etaf etaf added the ciflow/xpu Run XPU CI tasks label Jun 28, 2024
@etaf
Copy link
Copy Markdown
Collaborator Author

etaf commented Jun 29, 2024

@pytorchbot label ciflow/trunk

@pytorch-bot pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label Jun 29, 2024
[ghstack-poisoned]
@etaf
Copy link
Copy Markdown
Collaborator Author

etaf commented Jul 2, 2024

@jansel @desertfire @gujinghui Hi, the xpu/ciflow has been blocked a few days by this newly added test case, can you please help have a review?

[ghstack-poisoned]
@etaf etaf changed the title [Inductor UT] Skip check fp16 accuray for convolution with OC=1 on XPU. [Inductor UT] Skip check fp16 accuray for convolution with OC< 3 on XPU. Jul 2, 2024

self.common(fn, (x, w))
check_lowp = True
# TODO: fix this. OneDNN convolution with output channel == 1
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Can you create an issue and reference this issue here ? To have some more context why this is needed ?

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

Hi, @atalman , actually it's it's a matter of precision. I've updated the code and also created a issue #129974 for it.

[ghstack-poisoned]
@etaf etaf changed the title [Inductor UT] Skip check fp16 accuray for convolution with OC< 3 on XPU. [Inductor UT] update rtol for convoluton on XPU. Jul 2, 2024
@etaf
Copy link
Copy Markdown
Collaborator Author

etaf commented Jul 3, 2024

unrelated fail job: pull / linux-focal-py3.12-clang10 / test (default, 4, 4, linux.2xlarge) (gh)
@pytorchbot merge -i

@github-actions github-actions bot deleted the gh/etaf/26/head branch August 2, 2024 01:58
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ciflow/trunk Trigger trunk jobs on your pull request ciflow/xpu Run XPU CI tasks Merged module: inductor open source topic: not user facing topic category

Projects

Archived in project

Development

Successfully merging this pull request may close these issues.

4 participants