Custom shuffle layer leaks memory when run on Apple M1 GPU with tensorflow-metal
#60616
Labels
comp:gpu
GPU related issues
stat:awaiting response
Status - Awaiting response from author
subtype:macOS
macOS Build/Installation issues
TF 2.12
For issues related to Tensorflow 2.12
type:bug
Bug
Click to expand!
Issue Type
Bug
Have you reproduced the bug with TF nightly?
No
Source
binary
Tensorflow Version
2.12.0
Custom Code
Yes
OS Platform and Distribution
macOS 13.0 (22A380)
Mobile device
Apple M1
Python version
3.10.10
Bazel version
No response
GCC/Compiler version
No response
CUDA/cuDNN version
No response
GPU model and memory
No response
Current Behaviour?
The following layer leaks memory during training with keras when running on an Apple M1 with
tensorflow-macos
andtensorflow-metal
installed:When run on the CPU alone it does not leak memory
Standalone code to reproduce the issue
The text was updated successfully, but these errors were encountered: