Skip to content

Conversation

gujinghui
Copy link
Collaborator

The input shape checkers in conv/int8_conv operator is aims to avoid the issue when running with mkldnn winograd, the weigths has to be reordered each time if input shape changed.
However, the checkers result to big performance regression due to frequent reorder.

Meanwhile, in mkldnn-bridge, such case has been already fixed by correcting the prop_kind.
Therefore, we have to remove the useless checker to fix the performance regression.

because already done in mkldnn-bridge

Signed-off-by: Gu, Jinghui <jinghui.gu@intel.com>
@gujinghui
Copy link
Collaborator Author

@yinghai

Could you help review this PR...

Thanks a lot,
Jinghui

@jerryzh168 jerryzh168 requested a review from yinghai April 23, 2019 20:24
Copy link
Contributor

@facebook-github-bot facebook-github-bot left a comment

Choose a reason for hiding this comment

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

@yinghai has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

@facebook-github-bot
Copy link
Contributor

@yinghai merged this pull request in b675f07.

zhangguanheng66 pushed a commit to zhangguanheng66/pytorch that referenced this pull request May 6, 2019
Summary:
The input shape checkers in conv/int8_conv operator is aims to avoid the issue when running with mkldnn winograd, the weigths has to be reordered each time if input shape changed.
However, the checkers result to big performance regression due to frequent reorder.

Meanwhile, in mkldnn-bridge, such case has been already fixed by correcting the prop_kind.
Therefore, we have to remove the useless checker to fix the performance regression.
Pull Request resolved: pytorch#19608

Differential Revision: D15061169

Pulled By: yinghai

fbshipit-source-id: 649a43ae6fce989e84939210f6dffb143ec3d350
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