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This repository has been archived by the owner on Nov 17, 2023.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

Export resize and support batch size #14014

Merged
merged 46 commits into from
Feb 1, 2019
Merged

Export resize and support batch size #14014

merged 46 commits into from
Feb 1, 2019

Conversation

stu1130
Copy link
Contributor

@stu1130 stu1130 commented Jan 29, 2019 •
edited
Loading

Description

  1. move the logic from python to C++ and change the Resize as HybridBlock
  2. support batch_size for resize
  3. add unit tests
  4. support resize GPU with Hang implementation

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

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

Comments

@sandeep-krishnamurthy
@zhanghang1989
@zhreshold

@stu1130 stu1130 requested a review from szha as a code owner January 29, 2019 00:12
sandeep-krishnamurthy
sandeep-krishnamurthy reviewed Jan 29, 2019
View reviewed changes
Copy link
Contributor

@sandeep-krishnamurthy sandeep-krishnamurthy left a comment

Choose a reason for hiding this comment

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

Thank you. Great work.

python/mxnet/gluon/data/vision/transforms.py Outdated Show resolved Hide resolved
src/operator/contrib/bilinear_resize.cu Show resolved Hide resolved
src/operator/image/image_utils.h Outdated Show resolved Hide resolved
src/operator/image/resize-inl.h Outdated Show resolved Hide resolved
@stu1130
add image resize operator and unit test
998c7ef
@stu1130
refactor the resize operator and address lint issues
4e63cad
@stu1130
address comment and add doc
feb661f
@stu1130
assert size is more than 2
f8ccd2f
@stu1130
add test case of 4D input
20b9c0d
@stu1130
use ndarray datatype
529121d
@stu1130
add inline to Shape
043584f
@stu1130
add 4D input example
ce0a447
@stu1130
refactor the duplicate code and separate the resize from image_random
26287b4
@stu1130
clean up the code
df8bb81
@stu1130
add resize implementation
d807326
@stu1130
delete the variable not used
20ed3a3
@stu1130
refactor the code with structure and enum to make code more understan…
1db74be 
…dable
@stu1130
fix the lint
20f288f
@stu1130
address comments
e365f9f
@stu1130
address comment 1. add description 2. refactor unit test and add dtype
546a81f
@stu1130
update data type check
c3ec485
@stu1130
lint
22583ae
@stu1130
move the common utitlity to image_utils
46b2e4d
@stu1130
add default value for keep_ratio
e5464f1
@stu1130
change the operator doc
d805b2e
@stu1130
update the image utility function
c2f8d61
@stu1130
fix lint
ecb3be6
@stu1130
use Hang implementation to achieve image resize operator GPU
1c3295d
@stu1130
update the check and doc
c9b2e80
@stu1130
refactor the caffe_gpu_interp2_kernel
d019090
@stu1130
update doc and fix the cpu compile error
ddf8d4e
@stu1130
update the comment
daa2b4b
@stu1130 stu1130 force-pushed the export_resize_and_support_batch_size branch from 4d189f2 to b31d19f Compare January 29, 2019 21:37
@stu1130 stu1130 changed the title [WIP] Export resize and support batch size Export resize and support batch size Jan 30, 2019
zhanghang1989
zhanghang1989 approved these changes Jan 31, 2019
View reviewed changes
Copy link
Contributor

@zhanghang1989 zhanghang1989 left a comment

Choose a reason for hiding this comment

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

LGTM

sandeep-krishnamurthy
sandeep-krishnamurthy approved these changes Feb 1, 2019
View reviewed changes
Copy link
Contributor

@sandeep-krishnamurthy sandeep-krishnamurthy left a comment

Choose a reason for hiding this comment

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

Thank you.

@sandeep-krishnamurthy sandeep-krishnamurthy merged commit 2a4634b into apache:master Feb 1, 2019
stu1130 added a commit to stu1130/incubator-mxnet that referenced this pull request Feb 1, 2019
@stu1130
Revert "Export resize and support batch size (apache#14014)"
69bf236 
This reverts commit 2a4634b.
@stu1130 stu1130 mentioned this pull request Feb 1, 2019
Closed
stu1130 added a commit to stu1130/incubator-mxnet that referenced this pull request Feb 1, 2019
@stu1130
Revert "Export resize and support batch size (apache#14014)"
3917cb2 
This reverts commit 2a4634b.
@stu1130 stu1130 mentioned this pull request Feb 1, 2019
Closed
5 tasks
stephenrawls pushed a commit to stephenrawls/incubator-mxnet that referenced this pull request Feb 16, 2019
@stu1130 @stephenrawls
Export resize and support batch size (apache#14014)
f77a2e8 
* add image resize operator and unit test

* refactor the resize operator and address lint issues

* address comment and add doc

* assert size is more than 2

* add test case of 4D input

* use ndarray datatype

* add inline to Shape

* add 4D input example

* refactor the duplicate code and separate the resize from image_random

* clean up the code

* add resize implementation

* delete the variable not used

* refactor the code with structure and enum to make code more understandable

* fix the lint

* address comments

* address comment 1. add description 2. refactor unit test and add dtype

* update data type check

* lint

* move the common utitlity to image_utils

* add default value for keep_ratio

* change the operator doc

* update the image utility function

* fix lint

* use Hang implementation to achieve image resize operator GPU

* update the check and doc

* refactor the caffe_gpu_interp2_kernel

* update doc and fix the cpu compile error

* update the comment

* fix lint

* add unit test for gpu

* address comments

* remove the crop and centercop utility function to make the PR clear

* fix the syntax error

* delete the warning

* add unit test with 4D

* fix typo

* add more unit test

* fix unit test

* set atol = 1

* fix missing numpy import

* fix the unit test

* delete test case

* fix unit test missing dependency

* fix error data type

* unify the style and add invalid interp

* update the doc
vdantu pushed a commit to vdantu/incubator-mxnet that referenced this pull request Mar 31, 2019
@stu1130 @vdantu
Export resize and support batch size (apache#14014)
d3fc292 
* add image resize operator and unit test

* refactor the resize operator and address lint issues

* address comment and add doc

* assert size is more than 2

* add test case of 4D input

* use ndarray datatype

* add inline to Shape

* add 4D input example

* refactor the duplicate code and separate the resize from image_random

* clean up the code

* add resize implementation

* delete the variable not used

* refactor the code with structure and enum to make code more understandable

* fix the lint

* address comments

* address comment 1. add description 2. refactor unit test and add dtype

* update data type check

* lint

* move the common utitlity to image_utils

* add default value for keep_ratio

* change the operator doc

* update the image utility function

* fix lint

* use Hang implementation to achieve image resize operator GPU

* update the check and doc

* refactor the caffe_gpu_interp2_kernel

* update doc and fix the cpu compile error

* update the comment

* fix lint

* add unit test for gpu

* address comments

* remove the crop and centercop utility function to make the PR clear

* fix the syntax error

* delete the warning

* add unit test with 4D

* fix typo

* add more unit test

* fix unit test

* set atol = 1

* fix missing numpy import

* fix the unit test

* delete test case

* fix unit test missing dependency

* fix error data type

* unify the style and add invalid interp

* update the doc
haohuanw pushed a commit to haohuanw/incubator-mxnet that referenced this pull request Jun 23, 2019
@stu1130
Export resize and support batch size (apache#14014)
c256e9f 
* add image resize operator and unit test

* refactor the resize operator and address lint issues

* address comment and add doc

* assert size is more than 2

* add test case of 4D input

* use ndarray datatype

* add inline to Shape

* add 4D input example

* refactor the duplicate code and separate the resize from image_random

* clean up the code

* add resize implementation

* delete the variable not used

* refactor the code with structure and enum to make code more understandable

* fix the lint

* address comments

* address comment 1. add description 2. refactor unit test and add dtype

* update data type check

* lint

* move the common utitlity to image_utils

* add default value for keep_ratio

* change the operator doc

* update the image utility function

* fix lint

* use Hang implementation to achieve image resize operator GPU

* update the check and doc

* refactor the caffe_gpu_interp2_kernel

* update doc and fix the cpu compile error

* update the comment

* fix lint

* add unit test for gpu

* address comments

* remove the crop and centercop utility function to make the PR clear

* fix the syntax error

* delete the warning

* add unit test with 4D

* fix typo

* add more unit test

* fix unit test

* set atol = 1

* fix missing numpy import

* fix the unit test

* delete test case

* fix unit test missing dependency

* fix error data type

* unify the style and add invalid interp

* update the doc
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Reviewers

@sandeep-krishnamurthy sandeep-krishnamurthy sandeep-krishnamurthy approved these changes

@zhanghang1989 zhanghang1989 zhanghang1989 approved these changes

@szha szha Awaiting requested review from szha szha is a code owner

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

Successfully merging this pull request may close these issues.

None yet
3 participants
@stu1130 @sandeep-krishnamurthy @zhanghang1989