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

ComplexAbs and Resize antialias errors #2368

Open
ZachZhao96 opened this issue Nov 20, 2024 · 0 comments
Open

ComplexAbs and Resize antialias errors #2368

ZachZhao96 opened this issue Nov 20, 2024 · 0 comments
Labels
bug An unexpected problem or unintended behavior

Comments

@ZachZhao96
Copy link

ZachZhao96 commented Nov 20, 2024

Hi, I tried the latest tf2onnx and used tf2onnx.convert.from_keras(model,opset=19) to convert my keras h5 model. Here is my partial code for preprocessing:
image

however, I got two error messages shown below:
image

The first error says that ComplexAbs doesn't expect the same dimesion input as tf.abs used. This is not an issue for opset -13, so i hope this is just a bug.

then, the second one says antialias for resize is not supported. I wonder if the antialias supported in Resize Op is the same as the tf.image.resize(..., antialias = True).
image

Here is the resize bilinear with antialias example in ONNX documentation.
image

Thanks for your time.

@ZachZhao96 ZachZhao96 added the bug An unexpected problem or unintended behavior label Nov 20, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug An unexpected problem or unintended behavior
Projects
None yet
Development

No branches or pull requests

1 participant