Skip to content

Conversation

@jakpiase
Copy link
Contributor

@jakpiase jakpiase commented Jun 15, 2021

PR types

New features

PR changes

OPs

Describe

Added split op bf16/fp32 oneDNN kernel. This PR caused 20% speed up on DPN68 model, measured on Intel(R) Core(TM) i9-9940X CPU @ 3.30GHz with 1000 repeated iterations. For bigger speed up keys optimizations are needed. For now OneDNN kernel will be used only if AxisTensor and SectionsTensorList are not set.

@paddle-bot-old
Copy link

Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

@jakpiase jakpiase marked this pull request as draft June 15, 2021 17:55
@jakpiase jakpiase marked this pull request as ready for review June 17, 2021 13:49
@jakpiase jakpiase requested review from jczaja and lidanqing-vv June 17, 2021 15:04
jczaja
jczaja previously approved these changes Jun 18, 2021
Copy link
Contributor

@jczaja jczaja left a comment

Choose a reason for hiding this comment

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

LGTM

lidanqing-vv
lidanqing-vv previously approved these changes Jun 21, 2021
Copy link
Contributor

@lidanqing-vv lidanqing-vv left a comment

Choose a reason for hiding this comment

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

LGTM

Copy link
Contributor

@wozna wozna left a comment

Choose a reason for hiding this comment

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

It would be good if you adapted these changes to the functionality of inference BF16.
This involves:

@jakpiase jakpiase dismissed stale reviews from lidanqing-vv and jczaja via 2bfbf62 June 22, 2021 10:12
Copy link
Contributor

@wozna wozna left a comment

Choose a reason for hiding this comment

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

LGTM

Copy link
Contributor

@lidanqing-vv lidanqing-vv left a comment

Choose a reason for hiding this comment

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

LGTM.

Copy link

@ghost ghost left a comment

Choose a reason for hiding this comment

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

LGTM

Copy link
Contributor

@wanghuancoder wanghuancoder left a comment

Choose a reason for hiding this comment

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

LGTM

@jczaja jczaja merged commit 6810650 into PaddlePaddle:develop Jun 23, 2021
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.

5 participants