Error When Running TensorFlow Recommenders Example in SageMaker: "ValueError: Cannot convert 'counter' to a shape" #4591
Unanswered
Its-suLav-D
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I've been following the TensorFlow Recommenders Basic Retrieval example and attempted to implement it in an AWS SageMaker environment. My setup includes TensorFlow version 2.13.0 with GPU support enabled.
However, I encountered a ValueError during the execution of the example code. The error occurs when attempting to initialize the Streaming layer within the factorized_top_k.py file. Here's the traceback for clarity:
I'm puzzled by this error because it suggests an issue with converting the string 'counter' into a shape, which seems to stem from the internals of TensorFlow/Keras rather than the example code directly. When I run the same code in Google Colab or my local computer I don't get that specific error.
Has anyone encountered a similar issue or have insights on how to resolve this error? Any guidance on troubleshooting this problem in a SageMaker environment specifically would be greatly appreciated.
Thank you for your help!
Beta Was this translation helpful? Give feedback.
All reactions