Add an attribute to disable custom kernels in deformable detr in order to make the model ONNX exportable #22918
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
|
The documentation is not available anymore as the PR was closed or merged. |
sgugger
reviewed
Apr 21, 2023
amyeroberts
approved these changes
Apr 21, 2023
| @@ -520,6 +520,10 @@ def __init__(self, embed_dim: int, num_heads: int, n_levels: int, n_points: int) | |||
| self.value_proj = nn.Linear(embed_dim, embed_dim) | |||
| self.output_proj = nn.Linear(embed_dim, embed_dim) | |||
|
|
|||
| # This option is necessary for the ONNX export, as the try/catch in the forward | |||
| # is not supported by PyTorch ONNX export | |||
| self.disable_custom_kernels = False | |||
There was a problem hiding this comment.
I'd remove the # Copied from statement for this method, as the comment and additional parameter don't make sense here
|
Thank you will update! |
|
@amyeroberts Let me know if this is better! |
amyeroberts
approved these changes
Apr 21, 2023
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
sgugger
reviewed
Apr 21, 2023
Comment on lines
592
to
594
| def __init__( | ||
| self, embed_dim: int, num_heads: int, n_levels: int, n_points: int, disable_custom_kernels: bool = False | ||
| ): |
There was a problem hiding this comment.
Let's pass the config here (I think you can also remove n_points from the args, which seems to always be config.num_feature_levels) and the read the flag from the config.
There was a problem hiding this comment.
Hopefully addressed in 4f3f041!
…com/fxmarty/transformers into fix-deformable-detr-for-onnx-export
sgugger
approved these changes
Apr 24, 2023
gojiteji
pushed a commit
to gojiteji/transformers
that referenced
this pull request
Jun 5, 2023
…r to make the model ONNX exportable (huggingface#22918) * add disable kernel option * add comment * fix copies * add disable_custom_kernels to config * Update src/transformers/models/deta/modeling_deta.py Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com> * Update src/transformers/models/deta/modeling_deta.py Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com> * Update src/transformers/models/deta/modeling_deta.py Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com> * style * fix --------- Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
novice03
pushed a commit
to novice03/transformers
that referenced
this pull request
Jun 23, 2023
…r to make the model ONNX exportable (huggingface#22918) * add disable kernel option * add comment * fix copies * add disable_custom_kernels to config * Update src/transformers/models/deta/modeling_deta.py Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com> * Update src/transformers/models/deta/modeling_deta.py Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com> * Update src/transformers/models/deta/modeling_deta.py Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com> * style * fix --------- Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.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.
As per title and reported in #22330 and huggingface/optimum#931
This option will allow us to patch the model on the fly during the export to avoid going into the try/catch logic that is not supported by PyTorch ONNX export.