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

adding ROCm specifics for the tensorboard profiler tutorials #2684

Merged
merged 18 commits into from
Dec 21, 2023

Conversation

hongxiayang
Copy link
Contributor

@hongxiayang hongxiayang commented Nov 15, 2023

Fixes #2014

Description

The original tutorial is well written. But the issue #2014 described some problems with trace view. We verified the tutorial by running it on the ROCm platform.
The intention of this pull request is to give additional details of how the original tutorial works on the ROCm platform. It also addressed the original issues using an alternative approach.

Checklist

  • The issue that is being fixed is referred in the description (see above "Fixes #ISSUE_NUMBER")
  • Only one issue is addressed in this pull request
  • Labels from the issue that this PR is fixing are added to this pull request
  • No unnecessary issues are included into this pull request.

cc @aaronenyeshi @chaekit @jcarreiro @sekyondaMeta @svekars @carljparker @NicolasHug @kit1980 @subramen

Copy link

pytorch-bot bot commented Nov 15, 2023

No ciflow labels are configured for this repo.
For information on how to enable CIFlow bot see this wiki

Copy link

pytorch-bot bot commented Nov 15, 2023

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/tutorials/2684

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

✅ No Failures

As of commit 21c6911 with merge base 1285cab (image):
💚 Looks good so far! There are no failures yet. 💚

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

@hongxiayang hongxiayang changed the title adding ROCm specific for the tensorboard profiler tutorials adding ROCm specifics for the tensorboard profiler tutorials Nov 15, 2023
Copy link
Contributor

@svekars svekars left a comment

Choose a reason for hiding this comment

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

Thanks so much! A few editorial suggestions.

@hongxiayang
Copy link
Contributor Author

Thanks so much! A few editorial suggestions.

thanks for the review, i will take this into account when I update the diff.

@hongxiayang
Copy link
Contributor Author

HI, @svekars @malfet Can you help finish the review of this pull request? Thanks.

@hongxiayang
Copy link
Contributor Author

HI, @svekars @malfet : Thanks for helping review this pull request. The current checks failed on "Check spelling" step. However, I could not find any error message when clicking the "Details". I did run pyspelling in my local machine before my last commit. Any hints?

#
# .. code-block::
#
# I1115 15:23:49.014848 139961617598208 loader.py:57] started all processing
Copy link
Contributor

Choose a reason for hiding this comment

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

This needs to be indented because it appears incorrect in the html:
tensorboard_profiler_tutorial_issue

@svekars
Copy link
Contributor

svekars commented Dec 21, 2023

@hongxiayang Can you please update the indentation as suggested to fix the HTML? This might help to resolve the spellcheck issue as well.

@hongxiayang
Copy link
Contributor Author

hongxiayang commented Dec 21, 2023

@hongxiayang Can you please update the indentation as suggested to fix the HTML? This might help to resolve the spellcheck issue as well.
@svekars
I just removed the whole section related to "ValueError: Duplicate plugins for name projector" . That problem went away after I used the official pytorch wheel , and I think we can remove it.

@hongxiayang
Copy link
Contributor Author

@svekars I saw you reverted my changes to the en-wordlist.txt. However, my local run of pyspelling showed the ROCm is a misspelled word now. I will need to add ROCm back to the word dict. Any comments?
image

@svekars svekars merged commit b58b8a5 into pytorch:main Dec 21, 2023
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.

profiler ROCm tracing result was wrong
4 participants