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This repository has been archived by the owner on Nov 17, 2023.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

[MKLDNN] support using any format in pooling backward #17900

Merged
merged 2 commits into from
Apr 10, 2020
Merged

[MKLDNN] support using any format in pooling backward #17900

merged 2 commits into from
Apr 10, 2020

Conversation

ElaineBao
Copy link
Contributor

@ElaineBao ElaineBao commented Mar 24, 2020 •
edited
Loading

Description

Support using any format in pooling backward. For performance reasons, backward computations of pooling requires consistent memory format with the corresponding forward computations. Hence, using any format means it will align with forward memory format automatically, this avoids unnecessary reorder.

@TaoLv @rongzha1 @PatricZhao please review.

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

  • The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant JIRA issue created (except PRs with tiny changes)
  • Changes are complete (i.e. I finished coding on this PR)
  • All changes have test coverage:
  • Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
  • Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore)
  • Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL)
  • Code is well-documented:
  • For user-facing API changes, API doc string has been updated.
  • For new C++ functions in header files, their functionalities and arguments are documented.
  • For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
  • Check the API doc at https://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
  • To the best of my 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

@mxnet-bot
Copy link

Hey @ElaineBao , Thanks for submitting the PR
Once your PR is ready for CI checks, invoke the following commands:

  • To trigger all jobs: @mxnet-bot run ci [all]
  • To trigger specific jobs: @mxnet-bot run ci [job1, job2]

CI supported jobs: [centos-gpu, unix-cpu, unix-gpu, miscellaneous, clang, windows-cpu, website, centos-cpu, windows-gpu, sanity, edge]

Note:
Only following 3 categories can trigger CI :PR Author, MXNet Committer, Jenkins Admin.
All CI tests must pass before the PR can be merged.

@ElaineBao
Copy link
Contributor Author

@mxnet-label-bot add [MKL]

@lanking520 lanking520 added the MKL label Mar 24, 2020
@ElaineBao
Copy link
Contributor Author

@mxnet-bot run ci [all]

TaoLv
TaoLv reviewed Mar 24, 2020
View reviewed changes
src/operator/nn/mkldnn/mkldnn_pooling.cc Outdated
@@ -30,6 +30,11 @@
namespace mxnet {
namespace op {

// This can be replaced by mkldnn::memory::desc::data_type() after 1.2 release
Copy link
Member

Choose a reason for hiding this comment

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

The master branch is using 1.2.2 release of DNNL now, so let's use data_type() instead.

@TaoLv TaoLv added MKLDNN and removed MKL labels Mar 24, 2020
@ElaineBao
Copy link
Contributor Author

@mxnet-bot run ci [centos-cpu, unix-gpu, windows-gpu]

@mxnet-bot
Copy link

Jenkins CI successfully triggered : [windows-gpu, unix-gpu, centos-cpu]

@ElaineBao
Copy link
Contributor Author

@mxnet-bot run ci [windows-gpu]

@mxnet-bot
Copy link

Jenkins CI successfully triggered : [windows-gpu]

@ElaineBao
Copy link
Contributor Author

@mxnet-bot run ci [windows-gpu]

@mxnet-bot
Copy link

Jenkins CI successfully triggered : [windows-gpu]

@TaoLv
Copy link
Member

TaoLv commented Apr 4, 2020

The CI issue should be already addressed. Please rebase your PR and resolve the comments. Thanks.

@ElaineBao
Copy link
Contributor Author

@mxnet-bot run ci [centos-gpu, unix-cpu, unix-gpu, miscellaneous, clang, centos-cpu, windows-gpu]

@mxnet-bot
Copy link

Jenkins CI successfully triggered : [windows-gpu, unix-gpu, clang, unix-cpu, centos-cpu, miscellaneous, centos-gpu]

@ElaineBao
Copy link
Contributor Author

@mxnet-bot run ci [centos-gpu, windows-gpu, unix-gpu]

@mxnet-bot
Copy link

Jenkins CI successfully triggered : [windows-gpu, centos-gpu, unix-gpu]

@pengzhao-intel
Copy link
Contributor

Please rebase the code and trigger CI again.

@ElaineBao
Copy link
Contributor Author

@mxnet-bot run ci [centos-gpu, windows-gpu, unix-gpu]

@mxnet-bot
Copy link

Jenkins CI successfully triggered : [unix-gpu, windows-gpu, centos-gpu]

pengzhao-intel
pengzhao-intel approved these changes Apr 10, 2020
View reviewed changes
Copy link
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 and merging now.

@pengzhao-intel pengzhao-intel merged commit 8f82cd8 into apache:master Apr 10, 2020
@ElaineBao ElaineBao deleted the mkldnn-pooling-backward branch April 14, 2020 00:42
ElaineBao added a commit to ElaineBao/incubator-mxnet that referenced this pull request Apr 15, 2020
@ElaineBao
[MKLDNN] support using any format in pooling backward (apache#17900)
31c090a 
* use any format in pooling backward

* use data_type()
@ElaineBao ElaineBao mentioned this pull request Apr 15, 2020
Merged
7 tasks
pengzhao-intel pushed a commit that referenced this pull request Apr 16, 2020
@ElaineBao
[v1.x] backport #17900 "[MKLDNN] support using any format in pooling …
b56571d 
…backward" (#18067)

* [MKLDNN] support using any format in pooling backward (#17900)

* use any format in pooling backward

* use data_type()

* fix backport
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Reviewers

@TaoLv TaoLv TaoLv left review comments

@rongzha1 rongzha1 rongzha1 approved these changes

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

Assignees
No one assigned
Labels
MKLDNN
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
6 participants
@ElaineBao @mxnet-bot @TaoLv @pengzhao-intel @rongzha1 @lanking520