Skip to content

Improves performance of search reductions for small numbers of elements - #1464

Merged
ndgrigorian merged 2 commits into
optimize-small-size-tree-reductionfrom
sequential-search-kernels
Nov 3, 2023
Merged

Improves performance of search reductions for small numbers of elements#1464
ndgrigorian merged 2 commits into
optimize-small-size-tree-reductionfrom
sequential-search-kernels

Conversation

@ndgrigorian

@ndgrigorianndgrigorian commented Nov 3, 2023•
edited by oleksandr-pavlyk
Loading

Copy link
Copy Markdown
Collaborator

This pull request adds SequentialSearchReduction functor to improve argmax and argmin for small arrays.

  • Have you provided a meaningful PR description?
  • Have you added a test, reproducer or referred to an issue with a reproducer?
  • Have you tested your changes locally for CPU and GPU devices?
  • Have you made sure that new changes do not introduce compiler warnings?
  • Have you checked performance impact of proposed changes?
  • If this PR is a work in progress, are you opening the PR as a draft?

@github-actions

github-actionsBot commented Nov 3, 2023

Copy link
Copy Markdown

View rendered docs @ https://intelpython.github.io/dpctl/pulls/1464/index.html

@github-actions

github-actionsBot commented Nov 3, 2023

Copy link
Copy Markdown

Array API standard conformance tests for dpctl=0.15.1dev0=py310ha25a700_79 ran successfully.
Passed: 935
Failed: 65
Skipped: 119

@ndgrigorian
Adds SequentialSearchReduction functor to search reductions
5709f99
@ndgrigorian
Search reductions use correct branch for float16
119d43d
constexpr branch logic accounted for floating point types but not sycl::half,
which meant NaNs were not propagating for float16 data
oleksandr-pavlyk
oleksandr-pavlyk reviewed Nov 3, 2023
View reviewed changes
Comment threaddpctl/tensor/libtensor/include/kernels/reductions.hpp Outdated
@github-actions

github-actionsBot commented Nov 3, 2023

Copy link
Copy Markdown

Array API standard conformance tests for dpctl=0.15.1dev0=py310ha25a700_81 ran successfully.
Passed: 935
Failed: 65
Skipped: 119

@ndgrigorian
ndgrigorianforce-pushed the sequential-search-kernels branch from acd93ae to 119d43dCompareNovember 3, 2023 20:14
oleksandr-pavlyk
oleksandr-pavlyk approved these changes Nov 3, 2023
View reviewed changes

@oleksandr-pavlykoleksandr-pavlyk left a comment

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.

Looks great @ndgrigorian , thank you!

@ndgrigorian
ndgrigorian merged commit af28d98 into optimize-small-size-tree-reductionNov 3, 2023
@github-actions

github-actionsBot commented Nov 3, 2023

Copy link
Copy Markdown

Deleted rendered PR docs from intelpython.github.com/dpctl, latest should be updated shortly. 🤞

@github-actions

github-actionsBot commented Nov 3, 2023

Copy link
Copy Markdown

Array API standard conformance tests for dpctl=0.15.1dev0=py310ha25a700_83 ran successfully.
Passed: 935
Failed: 65
Skipped: 119

@ndgrigorian
ndgrigorian deleted the sequential-search-kernels branch November 28, 2023 18:03
Sign up for freeto join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

1 more reviewer
@oleksandr-pavlykoleksandr-pavlykoleksandr-pavlyk approved these changes
Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@ndgrigorian@oleksandr-pavlyk