Schema change to support dynamic shapes in ORT by ashbhandare · Pull Request #2955 · onnx/onnx

ashbhandare · 2020-08-12T00:52:37Z

This PR changes the schema of the following ops:
ReduceSum -> 'axes' attribute to input
Squeeze-> 'axes' attribute to input
Unsqueeze-> 'axes' attribute to input
Split -> 'split' attribute to input

It also modifies the function body of ops SoftmaxCrossEntropyLoss and NegativeLogLikelihoodLoss where they use the above ops with a modified schema.
The changes also include adding adapters for opset version conversion of these ops from 12->13 and 13->12.
This includes required changes to existing and new tests.

CLAassistant · 2020-08-12T00:52:44Z

All committers have signed the CLA.

gramalingam · 2020-08-27T21:08:23Z

It would be helpful if someone familiar with the optimizer and convertor could also review the PR, as it has updates to those too.

gramalingam · 2020-09-03T23:45:18Z

Can you fix the CI errors? Thanks!

askhade · 2020-09-04T16:37:55Z

Looks good to me. Please address the CI failures and make sure to regenerate the test data before checking in.

daquexian · 2020-09-19T01:54:47Z

Should we also update other ops (ReduceMax/Min/...) in the ReduceX family (they all have an 'axes' attribute) so that they can be consistent with ReduceSum?

* Changes to schema and python tests * Modify test * Remove attribute that is input also * Changes to optimizers, adapters and tests * Run flake8 * undo unrequired comit files, fix formatting, review changes * Fix ci test, cleanup * Fix narrowing conversion error * add missed test model Co-authored-by: G. Ramalingam <grama@microsoft.com> Signed-off-by: daquexian <daquexian566@gmail.com>

* Changes to schema and python tests * Modify test * Remove attribute that is input also * Changes to optimizers, adapters and tests * Run flake8 * undo unrequired comit files, fix formatting, review changes * Fix ci test, cleanup * Fix narrowing conversion error * add missed test model Co-authored-by: G. Ramalingam <grama@microsoft.com>

* Changes to schema and python tests * Modify test * Remove attribute that is input also * Changes to optimizers, adapters and tests * Run flake8 * undo unrequired comit files, fix formatting, review changes * Fix ci test, cleanup * Fix narrowing conversion error * add missed test model Co-authored-by: G. Ramalingam <grama@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com>

* remove wrong description for pow Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Add size check to make_tensor (#2987) Co-authored-by: Ke Zhang <linkerzhang@yeah.net> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Fix float16 data convert issue in numpy_helper.to_array (#3002) * handle f16 case for to_array * fix flake8 * nit: comment Co-authored-by: Ashwini Khade <askhade@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Deprecate Travis CI (#2773) * adding gtests for c++ api test * deprecate travis * adding the build badges for the new pipelines in Azure, deprecating travis build badge, renaming circleCI badge * updating badge label * removing - in badge names * c++ api changes for linux * update environment variables * update env variables, setup tools call * Update Linux-CI.yml for Azure Pipelines * revert changes to Linux and Mac CIs * delete last travis file Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Fix shape inference of scalar ConstantOfShape (#3005) When input shape is (0), we do not add any dim to inferred shape but we should initialize tensor_type.shape by calling mutable_shape(). Co-authored-by: Ashwini Khade <askhade@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * fix shape inference for loop (#3014) Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Fix ConvTranspose: enhance attribute check (#3000) * add check for using auto_pad and pads simultaneously * fix description for auto_pads == SAME_UPPER * update docs for operator * fix the old one as well * add a test * Revert "fix description for auto_pads == SAME_UPPER" This reverts commit e75e287. * Revert "update docs for operator" This reverts commit 70952c0. * Revert "fix the old one as well" This reverts commit 8a0482d. Co-authored-by: Ashwini Khade <askhade@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Schema change to support dynamic shapes in ORT (#2955) * Changes to schema and python tests * Modify test * Remove attribute that is input also * Changes to optimizers, adapters and tests * Run flake8 * undo unrequired comit files, fix formatting, review changes * Fix ci test, cleanup * Fix narrowing conversion error * add missed test model Co-authored-by: G. Ramalingam <grama@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * fix loop shape inference for ver 11 (#3023) Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Update softmax family ops behavior to align with other frameworks (fix #2289) (#2879) * Update softmax family ops behavior to align with other frameworks * Update logsoftmax, hardmax tests, regenerate docs and test data * fix wrong input name in function * regenerate test data * fix flake8 error * regenerate docs * regenerate docs * add missing type annotation for hardmax * add the math for softmax family operators * remove the 'description' field in docs as it is covered by the math * fix wrong format in axis attr * replace name with description * restore the name field for axis attr * regenerate docs * regenerate docs * add the missing name * regenerate docs * update reducesum to align with master * regenerate tests Co-authored-by: Wei-Sheng Chin <wschin@outlook.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Co-authored-by: Ke Zhang <linkerzhang@yeah.net> Co-authored-by: Ashwini Khade <askhade@microsoft.com> Co-authored-by: Vinitra Swamy <vinitras@gmail.com> Co-authored-by: Shinichiro Hamaji <shinichiro.hamaji@gmail.com> Co-authored-by: ashbhandare <ash.bhandare@gmail.com> Co-authored-by: daquexian <daquexian566@gmail.com> Co-authored-by: Wei-Sheng Chin <wschin@outlook.com>

* Changes to schema and python tests * Modify test * Remove attribute that is input also * Changes to optimizers, adapters and tests * Run flake8 * undo unrequired comit files, fix formatting, review changes * Fix ci test, cleanup * Fix narrowing conversion error * add missed test model Co-authored-by: G. Ramalingam <grama@microsoft.com>

…x#2999) * remove wrong description for pow Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Add size check to make_tensor (onnx#2987) Co-authored-by: Ke Zhang <linkerzhang@yeah.net> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Fix float16 data convert issue in numpy_helper.to_array (onnx#3002) * handle f16 case for to_array * fix flake8 * nit: comment Co-authored-by: Ashwini Khade <askhade@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Deprecate Travis CI (onnx#2773) * adding gtests for c++ api test * deprecate travis * adding the build badges for the new pipelines in Azure, deprecating travis build badge, renaming circleCI badge * updating badge label * removing - in badge names * c++ api changes for linux * update environment variables * update env variables, setup tools call * Update Linux-CI.yml for Azure Pipelines * revert changes to Linux and Mac CIs * delete last travis file Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Fix shape inference of scalar ConstantOfShape (onnx#3005) When input shape is (0), we do not add any dim to inferred shape but we should initialize tensor_type.shape by calling mutable_shape(). Co-authored-by: Ashwini Khade <askhade@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * fix shape inference for loop (onnx#3014) Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Fix ConvTranspose: enhance attribute check (onnx#3000) * add check for using auto_pad and pads simultaneously * fix description for auto_pads == SAME_UPPER * update docs for operator * fix the old one as well * add a test * Revert "fix description for auto_pads == SAME_UPPER" This reverts commit e75e287. * Revert "update docs for operator" This reverts commit 70952c0. * Revert "fix the old one as well" This reverts commit 8a0482d. Co-authored-by: Ashwini Khade <askhade@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Schema change to support dynamic shapes in ORT (onnx#2955) * Changes to schema and python tests * Modify test * Remove attribute that is input also * Changes to optimizers, adapters and tests * Run flake8 * undo unrequired comit files, fix formatting, review changes * Fix ci test, cleanup * Fix narrowing conversion error * add missed test model Co-authored-by: G. Ramalingam <grama@microsoft.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * fix loop shape inference for ver 11 (onnx#3023) Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> * Update softmax family ops behavior to align with other frameworks (fix onnx#2289) (onnx#2879) * Update softmax family ops behavior to align with other frameworks * Update logsoftmax, hardmax tests, regenerate docs and test data * fix wrong input name in function * regenerate test data * fix flake8 error * regenerate docs * regenerate docs * add missing type annotation for hardmax * add the math for softmax family operators * remove the 'description' field in docs as it is covered by the math * fix wrong format in axis attr * replace name with description * restore the name field for axis attr * regenerate docs * regenerate docs * add the missing name * regenerate docs * update reducesum to align with master * regenerate tests Co-authored-by: Wei-Sheng Chin <wschin@outlook.com> Signed-off-by: Chun-Wei Chen <jacky82226@gmail.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Co-authored-by: Ke Zhang <linkerzhang@yeah.net> Co-authored-by: Ashwini Khade <askhade@microsoft.com> Co-authored-by: Vinitra Swamy <vinitras@gmail.com> Co-authored-by: Shinichiro Hamaji <shinichiro.hamaji@gmail.com> Co-authored-by: ashbhandare <ash.bhandare@gmail.com> Co-authored-by: daquexian <daquexian566@gmail.com> Co-authored-by: Wei-Sheng Chin <wschin@outlook.com>

> Schema change to support dynamic shapes in ORT > Squeeze-> 'axes' attribute to input Ref: onnx/onnx#2955

ashbhandare changed the title ~~Ort schema change~~ Schema change to support dynamic shapes in ORT Aug 20, 2020

ashbhandare marked this pull request as ready for review August 20, 2020 22:50

ashbhandare requested review from a team as code owners August 20, 2020 22:50

ashbhandare force-pushed the ort_schema_change branch from 34303f1 to eec2d91 Compare August 20, 2020 22:56

ebarsoum assigned wschin, gramalingam and postrational Aug 20, 2020

ebarsoum added the topic: operator Issues related to ONNX operators label Aug 20, 2020

ebarsoum requested review from gramalingam, postrational and wschin August 21, 2020 03:43

gramalingam reviewed Aug 21, 2020

View reviewed changes

Comment thread onnx/defs/reduction/defs.cc

gramalingam reviewed Aug 21, 2020

View reviewed changes

Comment thread onnx/defs/reduction/defs.cc Outdated

gramalingam reviewed Aug 21, 2020

View reviewed changes

Comment thread onnx/defs/tensor/defs.cc Outdated

ebarsoum requested review from a team, ebarsoum and gramalingam August 24, 2020 03:57

ashbhandare force-pushed the ort_schema_change branch 2 times, most recently from fe5bb8e to 8b484f8 Compare August 27, 2020 19:25

gramalingam approved these changes Aug 27, 2020

View reviewed changes

linkerzhang reviewed Sep 7, 2020

View reviewed changes

Comment thread onnx/defs/tensor/defs.cc

ashbhandare force-pushed the ort_schema_change branch from 82be6db to d79da37 Compare September 14, 2020 16:59

ashbhandare added 2 commits September 14, 2020 20:12

Changes to schema and python tests

dfdbb69

Modify test

6fd70dd

ashbhandare added 7 commits September 14, 2020 20:12

Remove attribute that is input also

aac0466

Changes to optimizers, adapters and tests

37b631d

Run flake8

80f5164

undo unrequired comit files, fix formatting, review changes

e004afe

Fix ci test, cleanup

9135527

Fix narrowing conversion error

64f4b69

add missed test model

520ca87

ashbhandare force-pushed the ort_schema_change branch from bde3d80 to 520ca87 Compare September 14, 2020 20:32

Merge branch 'master' into ort_schema_change

202eec8

askhade merged commit 269a030 into onnx:master Sep 15, 2020

gramalingam mentioned this pull request Oct 15, 2020

Advance NLL loss to opset 13 #3059

Merged

yan12125 added a commit to EMCLab-Sinica/Stateful-CNN that referenced this pull request Nov 17, 2020

Adapt to ONNX 1.8 + refactoring

06d502e

> Schema change to support dynamic shapes in ORT > Squeeze-> 'axes' attribute to input Ref: onnx/onnx#2955

masahi mentioned this pull request Apr 24, 2021

[Operator] Why Cumsum axis is not an attribute? #3420

Closed

jcwchen mentioned this pull request Feb 8, 2022

Can ReduceSum's axes be put into the onnx model as an attribute instead of an input. #3996

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Schema change to support dynamic shapes in ORT#2955

Schema change to support dynamic shapes in ORT#2955
askhade merged 10 commits intoonnx:masterfrom
ashbhandare:ort_schema_change

ashbhandare commented Aug 12, 2020 •

edited

Loading

Uh oh!

CLAassistant commented Aug 12, 2020 •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

gramalingam commented Aug 27, 2020

Uh oh!

gramalingam commented Sep 3, 2020

Uh oh!

askhade commented Sep 4, 2020

Uh oh!

Uh oh!

daquexian commented Sep 19, 2020 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants

Conversation

ashbhandare commented Aug 12, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

CLAassistant commented Aug 12, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

gramalingam commented Aug 27, 2020

Uh oh!

gramalingam commented Sep 3, 2020

Uh oh!

askhade commented Sep 4, 2020

Uh oh!

Uh oh!

daquexian commented Sep 19, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants

ashbhandare commented Aug 12, 2020 •

edited

Loading

CLAassistant commented Aug 12, 2020 •

edited

Loading

daquexian commented Sep 19, 2020 •

edited

Loading