Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

opset 12 support #897

Merged
merged 13 commits into from Apr 28, 2020
Merged

opset 12 support #897

merged 13 commits into from Apr 28, 2020

Conversation

RandySheriffH
Copy link
Collaborator

@RandySheriffH RandySheriffH commented Apr 24, 2020

Adding support for opset 12 ops.

tf2onnx/onnx_opset/math.py Show resolved Hide resolved
tf2onnx/onnx_opset/math.py Outdated Show resolved Hide resolved
tf2onnx/onnx_opset/math.py Outdated Show resolved Hide resolved
tf2onnx/onnx_opset/math.py Outdated Show resolved Hide resolved
Copy link
Collaborator

@guschmue guschmue left a comment

Choose a reason for hiding this comment

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

For lessorequal - I think it would be ok to leave it as is

tf2onnx/onnx_opset/math.py Outdated Show resolved Hide resolved
tf2onnx/custom_opsets/ms.py Outdated Show resolved Hide resolved
tf2onnx/onnx_opset/logical.py Outdated Show resolved Hide resolved
tf2onnx/onnx_opset/logical.py Outdated Show resolved Hide resolved
tf2onnx/onnx_opset/logical.py Outdated Show resolved Hide resolved
class Inverse:
@classmethod
def version_12(cls, ctx, node, **kwargs):
utils.make_sure(node.get_attr('adjoint').i == 0, "adjoint must be false")
Copy link
Contributor

Choose a reason for hiding this comment

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

Why "adjoint must be false"? Does this parameter simply mean we should compute the inverse of the adjoint of this matrix if set to true? It seems we should be able to handle "true" as well.

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

Tested with inv(A) = adj(A)/det(A) approach, witnessed big diff. Will address adjoint case later.

@RandySheriffH RandySheriffH merged commit 40dd461 into onnx:master Apr 28, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants