Infer rank where reshape shape is inferred #5327
Merged
Conversation
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
Signed-off-by: Aditya Goel <agoel4512@gmail.com>
gramalingam
reviewed
Jun 16, 2023
Signed-off-by: Aditya Goel <agoel4512@gmail.com>
ff478cd to
90b92bc
Compare
Signed-off-by: Aditya Goel <agoel4512@gmail.com>
Signed-off-by: Aditya Goel <agoel4512@gmail.com>
adityagoel4512
commented
Jun 18, 2023
Signed-off-by: Aditya Goel <agoel4512@gmail.com>
9c81ad5 to
4b6b910
Compare
Signed-off-by: Aditya Goel <agoel4512@gmail.com>
…x#5246) ### Description The function implementation of CastLike does not propagate the attribute *saturate*. This PR fixes it and make the changes in the reference implementation to check the function implementation is working. This PR fixes two issues: 1. function implementation of CastLike (19) 2. reference implementation working context dependent function By default, the reference implementation choose a kernel if there is one, even if the operator defines an implementation. The PR introduces a mechanism to force the use of the kernel and expand the operator by its function. It delays that when the function is context dependant. It is then done when the node is executed and its inputs known. --------- Signed-off-by: xadupre <xadupre@microsoft.com> Signed-off-by: Aditya Goel <agoel4512@gmail.com>
e1e2cc9 to
0eb59b0
Compare
gramalingam
approved these changes
Jun 20, 2023
|
Does this require another approval? |
No. I have enabled auto-merge, so it should merge once CI completes, unless some other PR merges first (forcing another round of merging). |
adityagoel4512
added a commit
to adityagoel4512/onnx
that referenced
this pull request
Jul 28, 2023
Description This PR should address onnx#5324. In situations where the shape in a dynamic reshape operation itself has an inferred shape, this can be used to determine the rank of the output shape. Motivation and Context Closes onnx#5324 --------- Signed-off-by: Aditya Goel <agoel4512@gmail.com> Signed-off-by: xadupre <xadupre@microsoft.com> Co-authored-by: Xavier Dupré <xadupre@users.noreply.github.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Signed-off-by: Aditya Goel <agoel4512@gmail.com>
adityagoel4512
added a commit
to adityagoel4512/onnx
that referenced
this pull request
Jul 28, 2023
Description This PR should address onnx#5324. In situations where the shape in a dynamic reshape operation itself has an inferred shape, this can be used to determine the rank of the output shape. Motivation and Context Closes onnx#5324 --------- Signed-off-by: Aditya Goel <agoel4512@gmail.com> Signed-off-by: xadupre <xadupre@microsoft.com> Co-authored-by: Xavier Dupré <xadupre@users.noreply.github.com> Co-authored-by: G. Ramalingam <grama@microsoft.com> Signed-off-by: Aditya Goel <agoel4512@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Description
This PR should address #5324. In situations where the shape in a dynamic reshape operation itself has an inferred shape, this can be used to determine the rank of the output shape.
Motivation and Context
Closes #5324