Merged
Conversation
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
8556db1 to
5275f87
Compare
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
gramalingam
reviewed
Nov 1, 2022
gramalingam
reviewed
Nov 1, 2022
gramalingam
reviewed
Nov 1, 2022
gramalingam
reviewed
Nov 1, 2022
gramalingam
reviewed
Nov 1, 2022
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
gramalingam
reviewed
Nov 2, 2022
gramalingam
reviewed
Nov 2, 2022
Contributor
|
LGTM. I think there may be a minor interaction between this PR and #4512 (which is rationalizing all Reduce* functions to take axes as an input instead of attribute) which will need to be resolved when both get merged. |
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
Contributor
Author
|
@gramalingam Thanks for the suggestions. I've made the updates. Please review again. Thanks! |
Contributor
|
LGTM, thanks! I was wondering whether you had a chance to test the correctness of the function-body definition (eg., by running onnxruntime, or any other backend, on the generated expanded test-cases)? You may need to manually patch the opset of the generated test-cases to be the previous opset, and then it should be a valid test-case for the previous opset, I think, runnable by onnxruntime etc. |
Signed-off-by: Yuan Yao <yuanyao@nvidia.com>
…to dev-groupnorm
Contributor
Author
Thanks for pointing this out. I just tested and actually found a small error in the function body. Now fixed. |
gramalingam
approved these changes
Nov 4, 2022
justinchuby
pushed a commit
to justinchuby/onnx
that referenced
this pull request
Jan 27, 2023
* add GroupNormalization-18 Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * clean up function body Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix pytest error in function body Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error 2 Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error 3 Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error 4 Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * update md files Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * review feedback Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * update python test script per review feedback Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix python script lint issue Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix more python lint issues Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix Reshape error Signed-off-by: Yuan Yao <yuanyao@nvidia.com> Signed-off-by: Yuan Yao <yuanyao@nvidia.com> Co-authored-by: G. Ramalingam <grama@microsoft.com>
broune
pushed a commit
to broune/onnx
that referenced
this pull request
May 6, 2023
* add GroupNormalization-18 Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * clean up function body Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix pytest error in function body Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error 2 Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error 3 Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error 4 Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * update md files Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * review feedback Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix lint error Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * update python test script per review feedback Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix python script lint issue Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix more python lint issues Signed-off-by: Yuan Yao <yuanyao@nvidia.com> * fix Reshape error Signed-off-by: Yuan Yao <yuanyao@nvidia.com> Signed-off-by: Yuan Yao <yuanyao@nvidia.com> Co-authored-by: G. Ramalingam <grama@microsoft.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Signed-off-by: Yuan Yao yuanyao@nvidia.com
Description
Add new function op
GroupNormalizationto opset 18.Motivation and Context
Group norm is a commonly used normalization op, similar to some of the existing ONNX ops such as
BatchNormalization,InstanceNormalization, andLayerNormalization.Original paper that introduced the op: https://arxiv.org/abs/1803.08494
Related issue: #2020
Implementations in other frameworks:
https://www.tensorflow.org/addons/api_docs/python/tfa/layers/GroupNormalization
https://pytorch.org/docs/stable/generated/torch.nn.GroupNorm.html