Description
Tensorflow 2.13 not detecting in L40 server with cuda 12.2 and cudnn 8.9.5 and 535 nvidia driver
Environment
GPU Type: NVIDIA L40
Nvidia Driver Version: 535
CUDA Version: 12.2.2
CUDNN Version: 8.9.5
Operating System + Version: Ubuntu 20.04.2 LTS
Python Version (if applicable): python 3.8
TensorFlow Version (if applicable): Tensorflow 2.13
Baremetal or Container (if container which image + tag): nvidia/cuda:12.2.2-cudnn8-devel-ubuntu20.04
Relevant Files
Attached the error screenshot which we received
Steps To Reproduce
Run the below command and it give the error attached in screenshot.
python -c “import tensorflow as tf; tf.config.list_physical_devices(‘GPU’)”
And I was able to detect GPU with Torch but issue with tensor flow, If I downgrade to cuda 11.2 and cudnn 8.11 it works.
Please let us know tensorflow 2.13 officially support cuda 12.2 and cudnn 8.9.5 to detect GPU if yes please share the document or steps to get it done.