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

Add nvtx annotatations to groupby methods #12941

Merged
merged 4 commits into from
Apr 26, 2023

Conversation

wence-
Copy link
Contributor

@wence- wence- commented Mar 14, 2023

Description

Not sure if not annotating these was an oversight during implementation...

Checklist

  • I am familiar with the Contributing Guidelines.
  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

@wence- wence- requested a review from a team as a code owner March 14, 2023 17:48
@wence- wence- requested review from bdice and mroeschke March 14, 2023 17:48
@github-actions github-actions bot added the Python Affects Python cuDF API. label Mar 14, 2023
@wence- wence- self-assigned this Mar 14, 2023
@wence- wence- added improvement Improvement / enhancement to an existing function non-breaking Non-breaking change labels Mar 14, 2023
Copy link
Contributor

@bdice bdice left a comment

Choose a reason for hiding this comment

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

@wence- This would be a good opportunity to add a section to our Developer Guide about what methods should be annotated. (Searching "nvtx" returns nothing right now.) Should (all/some) private methods be annotated, or only public methods? Let's look at past NVTX PRs for guidance here, like these:

@shwina
Copy link
Contributor

shwina commented Mar 14, 2023

Should (all/some) private methods be annotated, or only public methods?

We have been super inconsistent about this so far, but I think all (and only) public methods should be annotated.

@galipremsagar
Copy link
Contributor

galipremsagar commented Mar 14, 2023

Should (all/some) private methods be annotated, or only public methods?

I'd vote for all public methods(& properties) to be annotated and select private methods on a case-by-case basis to be annotated. Historically we've annotated private methods for the sake of easier nvtx profiling while working on performance issues.

@bdice bdice changed the base branch from branch-23.04 to branch-23.06 April 23, 2023 21:28
@bdice bdice requested a review from a team as a code owner April 23, 2023 21:28
Copy link
Contributor

@bdice bdice left a comment

Choose a reason for hiding this comment

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

@wence- I updated this PR for 23.06. I think this is in a good state. Comments above mentioned that only "select" private methods should be annotated. I think the private methods that are annotated here are good candidates for annotation, so I would recommend merging this in the current state.

If we want to add documentation on when to add annotations, I think the discussion above is a good starting point but I would write that in a separate PR so this can merge without further delay.

@ajschmidt8 ajschmidt8 removed the request for review from a team April 24, 2023 17:00
@wence-
Copy link
Contributor Author

wence- commented Apr 26, 2023

/merge

@rapids-bot rapids-bot bot merged commit 62c43a8 into rapidsai:branch-23.06 Apr 26, 2023
@wence- wence- deleted the wence/fea/groupby-nvtx branch April 26, 2023 15:42
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
improvement Improvement / enhancement to an existing function non-breaking Non-breaking change Python Affects Python cuDF API.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants