Thanks to visit codestin.com
Credit goes to github.com

Skip to content
This repository was archived by the owner on Nov 17, 2023. It is now read-only.

[MKLDNN]Improve quantizeV2 and dequantize latency#14641

Merged
pengzhao-intel merged 9 commits into
apache:masterfrom
ZhennanQin:stateful_quantize
Apr 18, 2019
Merged

[MKLDNN]Improve quantizeV2 and dequantize latency#14641
pengzhao-intel merged 9 commits into
apache:masterfrom
ZhennanQin:stateful_quantize

Conversation

@ZhennanQin
Copy link
Copy Markdown
Contributor

@ZhennanQin ZhennanQin commented Apr 8, 2019 •
edited
Loading

Description

Change quantizeV2 and dequantize to stateful op to improve small batch size performance.

@pengzhao-intel @TaoLv @reminisce @zheng-da @anirudh2290

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

  • The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant JIRA issue created (except PRs with tiny changes)
  • Changes are complete (i.e. I finished coding on this PR)
  • All changes have test coverage:
  • Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
  • Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore)
  • Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL)
  • Code is well-documented:
  • For user-facing API changes, API doc string has been updated.
  • For new C++ functions in header files, their functionalities and arguments are documented.
  • For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
  • Check the API doc at http://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
  • To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change

Changes

  • Feature1, tests, (and when applicable, API doc)
  • Feature2, tests, (and when applicable, API doc)

Comments

  • If this change is a backward incompatible change, why must this change be made.
  • Interesting edge cases to note here

@pengzhao-intel pengzhao-intel mentioned this pull request Apr 8, 2019
Closed
@pengzhao-intel pengzhao-intel requested a review from TaoLv April 8, 2019 13:37
@piyushghai
Copy link
Copy Markdown
Contributor

piyushghai commented Apr 8, 2019

@ZhennanQin Thanks for your contributions. Can you look at the CI failures ?
@mxnet-label-bot Add [Quantization, pr-work-in-progress]

@marcoabreu marcoabreu added pr-work-in-progress PR is still work in progress Quantization Issues/Feature Requests related to Quantization labels Apr 8, 2019
@glingyan
Copy link
Copy Markdown
Contributor

glingyan commented Apr 15, 2019

@piyushghai ping for any further comments?

@glingyan glingyan mentioned this pull request Apr 15, 2019
Merged
7 tasks
@ZhennanQin ZhennanQin changed the title [WIP]stateful_quantize [MKLDNN]Improve quantizeV2 and dequantize latency Apr 16, 2019
anirudh2290
anirudh2290 reviewed Apr 16, 2019
View reviewed changes
Comment thread src/operator/quantization/quantize_v2.cc
Comment thread src/operator/quantization/mkldnn/mkldnn_dequantize-inl.h Outdated
Comment thread src/operator/quantization/mkldnn/mkldnn_quantize_v2-inl.h
Comment thread src/operator/quantization/mkldnn/mkldnn_quantize_v2-inl.h Outdated
Comment thread src/operator/quantization/mkldnn/mkldnn_dequantize-inl.h Outdated
pengzhao-intel
pengzhao-intel approved these changes Apr 18, 2019
View reviewed changes
Copy link
Copy Markdown
Contributor

@pengzhao-intel pengzhao-intel left a comment

Choose a reason for hiding this comment

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

LGTM

The change has been verified internally by our auto-ci with real models :)

@anirudh2290 Thanks for the review and really good suggestions.

@pengzhao-intel
Copy link
Copy Markdown
Contributor

pengzhao-intel commented Apr 18, 2019

Merging now.

@pengzhao-intel pengzhao-intel merged commit a083a61 into apache:master Apr 18, 2019
kedarbellare pushed a commit to kedarbellare/incubator-mxnet that referenced this pull request Apr 20, 2019
@ZhennanQin @kedarbellare
[MKLDNN]Improve quantizeV2 and dequantize latency (apache#14641)
61a8098 
* stateful_quantize

* fix lint

* Fix build

* fix gpu build

* Fix typo

* Move check to online calibration
@ZhennanQin ZhennanQin deleted the stateful_quantize branch May 31, 2019 02:07
haohuanw pushed a commit to haohuanw/incubator-mxnet that referenced this pull request Jun 23, 2019
@ZhennanQin
[MKLDNN]Improve quantizeV2 and dequantize latency (apache#14641)
209c7c2 
* stateful_quantize

* fix lint

* Fix build

* fix gpu build

* Fix typo

* Move check to online calibration
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Reviewers

@TaoLv TaoLv Awaiting requested review from TaoLv

2 more reviewers

@anirudh2290 anirudh2290 anirudh2290 approved these changes

@pengzhao-intel pengzhao-intel pengzhao-intel approved these changes

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

pr-work-in-progress PR is still work in progress Quantization Issues/Feature Requests related to Quantization

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

6 participants

@ZhennanQin @piyushghai @glingyan @pengzhao-intel @anirudh2290 @marcoabreu