Title: TensorFlow/Keras Integration Error: A KerasTensor cannot be used as input to a TensorFlow function

[ShoumitroC]

Issue type

Build/Install

Have you reproduced the bug with TensorFlow Nightly?

Yes

Source

binary

TensorFlow version

2.13.0

Custom code

Yes

OS platform and distribution

Mac OS x86-64

Mobile device

Macbook

Python version

3.9

Bazel version

No response

GCC/compiler version

No response

CUDA/cuDNN version

No response

GPU model and memory

No response

Current behavior?

I'm working on an image processing project using InsightFace and OpenNSFW2. I'm encountering an error with TensorFlow and Keras integration. When I try to use tf.pad on a tensor, I get the following error:

ValueError: A KerasTensor cannot be used as input to a TensorFlow function. A KerasTensor is a symbolic placeholder for a shape and dtype, used when constructing Keras Functional models or Keras Functions. You can only use it as input to a Keras layer or a Keras operation (from the namespaces keras.layers and keras.operations). You are likely doing something like:

Standalone code to reproduce the issue

ValueError: A KerasTensor cannot be used as input to a TensorFlow function. A KerasTensor is a symbolic placeholder for a shape and dtype, used when constructing Keras Functional models or Keras Functions. You can only use it as input to a Keras layer or a Keras operation (from the namespaces `keras.layers` and `keras.operations`). You are likely doing something like:

x = Input(...)
...
tf_fn(x)  # Invalid.


I've tried wrapping tf.pad in a custom layer as suggested, but I'm still getting the error. Here's the relevant code snippet:

class MyLayer(Layer):
    def call(self, x):
        return tf.pad(x, [[0, 0], [3, 3], [3, 3], [0, 0]], "CONSTANT")

# Usage
x = MyLayer()(x)


Could someone please help me understand what I'm doing wrong and how to fix it?

Thank you!

Relevant log output

No response

[sushreebarsa]

@ShoumitroC Could you try to use keras.layers.Lambda layer that allows you to define a custom function using a Python lambda expression that can operate on KerasTensors within the model building process?
Please upgrade to the latest TF version and if the issue still persists then kindly provide the standalone code to replicate this issue?
Thank you!

[github-actions]

This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.

[github-actions]

This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.

[google-ml-butler]

Are you satisfied with the resolution of your issue?
Yes
No