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This repository has been archived by the owner on Nov 17, 2023.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

[WIP] Onnx export API optional args and additional operator support, Fixes #12682 #12946

Closed
wants to merge 7 commits into from
Closed

[WIP] Onnx export API optional args and additional operator support, Fixes #12682 #12946

wants to merge 7 commits into from

Conversation

Roshrini
Copy link
Member

@Roshrini Roshrini commented Oct 23, 2018 •
edited
Loading

Description

ONNX export api takes optional arguments out_label and out_shapes
Adding operators - LogisticRegressionOutput, MakeLoss, BlockGrad, _rminus_scalar, r_div_scalar, identity

#12925

WIP - Need to add:

  • Test Cases for additional operators
  • Test case for multiple output model
  • Add comments

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

@Roshrini Roshrini requested a review from szha as a code owner October 23, 2018 23:28
@Roshrini
Copy link
Member Author

@mxnet-label-bot [ONNX, pr-work-in-progress]

@marcoabreu marcoabreu added ONNX pr-work-in-progress PR is still work in progress labels Oct 24, 2018
@vandanavk vandanavk mentioned this pull request Oct 29, 2018
Closed
@vandanavk
Copy link
Contributor

@Roshrini please add

Fixes #12682

in the PR description to automatically close the issue when PR is merged.

@Roshrini Roshrini changed the title [WIP] Onnx export API optional args and additional operator support [WIP] Onnx export API optional args and additional operator support, Fixes #12682 Oct 31, 2018
@kalyc
Copy link
Contributor

kalyc commented Nov 12, 2018

@Roshrini thanks for you contribution. There appears to be build failure could you take a look?

@stu1130
Copy link
Contributor

stu1130 commented Nov 21, 2018

@Roshrini ping again to address CI and description

@Roshrini Roshrini mentioned this pull request Nov 26, 2018
Open
vandanavk
vandanavk reviewed Nov 28, 2018
View reviewed changes
python/mxnet/contrib/onnx/mx2onnx/_op_translations.py
@@ -926,8 +966,14 @@ def convert_minus_scalar(node, **kwargs):
"""
return scalar_op_helper(node, 'Sub', **kwargs)

@mx_op.register("_rminus_scalar")
def convert_rminus_scalar(node, **kwargs):
"""Map MXNet's _rminus_scalar operator attributes to onnx's Minus operator.
Copy link
Contributor

Choose a reason for hiding this comment

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

nit: onnx's Sub operator

@vandanavk vandanavk mentioned this pull request Dec 3, 2018
Merged
@Roshrini
Copy link
Member Author

Will update and create new PR

@Roshrini Roshrini closed this Dec 10, 2018
@Roshrini Roshrini mentioned this pull request Dec 13, 2018
Closed
6 tasks
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Reviewers

@vandanavk vandanavk vandanavk left review comments

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

Assignees
No one assigned
Labels
ONNX pr-work-in-progress PR is still work in progress
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
5 participants
@Roshrini @vandanavk @kalyc @stu1130 @marcoabreu