Support for negative indices in 'Gather', 'GatherElements', 'ScatterElements', 'OneHot' #2260
Conversation
This file contains 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
BowenBao
reviewed
Aug 26, 2019
neginraoof
changed the title
|
Should I make a similar change for Gather_Nd. Alternatively, could you please include Gather_Nd when the op gets in (hopefully tomorrow) in this PR? That way all the negative indices spec updates can go in together ? :) |
|
@hariharans29 I can add GatherND in this PR |
Thanks @neginraoof. I actually had to make some formatting changes in the doc, so I went ahead and made the Gather_ND spec support negative indices (based on your change here). As part of your change, I think you can refine the sentence in GathertND if you want. Thanks! |
…gativeIndices # Conflicts: # docs/Operators.md # onnx/defs/operator_sets.h
|
@hariharans29 Thanks, I just rebased. The docs are consistent now. |
wschin
reviewed
Aug 31, 2019
wschin
reviewed
Aug 31, 2019
wschin
reviewed
Aug 31, 2019
wschin
reviewed
Sep 4, 2019
wschin
reviewed
Sep 4, 2019
wschin
reviewed
Sep 4, 2019
wschin
reviewed
Sep 4, 2019
wschin
approved these changes
Sep 4, 2019
Co-Authored-By: Wei-Sheng Chin <wschin@outlook.com>
Co-Authored-By: Wei-Sheng Chin <wschin@outlook.com>
|
@wschin Sure. I updated the PR as requested. |
spandantiwari
approved these changes
Sep 6, 2019
onnx/defs/tensor/old.cc
Outdated
| fail_type_inference("OneHot node must have three inputs."); | ||
| } | ||
| // Input 'depth' must be a scalar or a single-element vector. | ||
| // TODO: Ideally to match spec for this input only allow Scalar should be allowed. |
There was a problem hiding this comment.
This sentence is not phrased well. We should correct it.
|
cc @gramalingam @ebarsoum please review |
ebarsoum
approved these changes
Sep 10, 2019
…gativeIndices # Conflicts: # docs/Changelog.md # docs/Operators.md # docs/TestCoverage.md # onnx/backend/test/case/node/onehot.py # onnx/defs/operator_sets.h # onnx/defs/tensor/defs.cc # onnx/defs/tensor/old.cc
jcwchen
pushed a commit
to jcwchen/onnx
that referenced
this pull request
Sep 23, 2020
…lements', 'OneHot' (onnx#2260) * modified gather docs to support negative indices * added support for negatice indices to gather_elements and scatter_elements * fixed documentation formatting as per comments * Added negatice indices to docs for OneHot op * GatheND spec for negative indices editted * fix for comments * fixed formatting for gather op * Update onnx/defs/tensor/defs.cc Co-Authored-By: Wei-Sheng Chin <wschin@outlook.com> * Update docs/Changelog.md Co-Authored-By: Wei-Sheng Chin <wschin@outlook.com> * added print examples to onehot and gather as per comments * adding modifies model tests for onehot * updated doc files * updating unsqueeze test * typo fix as per comments
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.
We would need to support negative indices for these operators, since PyTorch does support this case. Using Select/Gather ops with -1 index is quite common among PyTorch models.