Skip to content

Add int64, uint64 data type support for reduceL1, reduceProduct, reduceSum and reduceSumSquare - #750

Closed
shiyi9801 wants to merge 1 commit into
webmachinelearning:mainfrom
shiyi9801:reduce
Closed

Add int64, uint64 data type support for reduceL1, reduceProduct, reduceSum and reduceSumSquare#750
shiyi9801 wants to merge 1 commit into
webmachinelearning:mainfrom
shiyi9801:reduce

Conversation

@shiyi9801

@shiyi9801 shiyi9801 commented Aug 14, 2024

Copy link
Copy Markdown
Contributor

Fixes #694


Preview | Diff

@shiyi9801

Copy link
Copy Markdown
Contributor Author

Current Chromium implementation allows int64/uint64 for these reduce ops, so I think we should either change the spec or the implementation. PTAL, thanks :) @huningxin @fdwr

@fdwr fdwr left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

opSupportLimits for CoreML will be negative for these ops, but LGTM. Thanks for spotting the difference.

@huningxin

Copy link
Copy Markdown
Contributor

Thanks @shiyi9801 , is this PR duplicated by #695?

@fdwr

fdwr commented Aug 15, 2024

Copy link
Copy Markdown
Collaborator

Thanks @shiyi9801 , is this PR duplicated by #695?

Aah yes, that did look familiar.

@shiyi9801

Copy link
Copy Markdown
Contributor Author

Thanks @shiyi9801 , is this PR duplicated by #695?

Oh I didn't notice there is already a PR to fix this, yes it is duplicated, please continue yours :) I'll drop this PR.

@shiyi9801 shiyi9801 closed this Aug 15, 2024
@shiyi9801
shiyi9801 deleted the reduce branch August 15, 2024 06:44
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.

Consider adding int64/uint64 data type support for some reduce operators

3 participants