Open Emilia-Re opened 2 weeks ago
Thanks for checking out our work!
CUDA and version compatibility can indeed be tricky. This page specifies compatible versions across TensorFlow cuDNN CUDA etc.
I have personally been running the code in a container (as specified in the Dockerfile), which is probably the easiest way to avoid version conflicts.
Maybe you have installed a cpu-only build of TensorFlow? If you are installing with pip, try something like pip install tensorflow-gpu==2.14.0
or pip install tensorflow[and-cuda]==2.14.0
.
Let me know if this helps or if you run into any more issues!
Thank you for your work. I want to run your code on my mahcine but encountered many issues with tensorflow. Could you provice more information about your working environment ,like python version, cuda version, driver version,etc. The main problem is that tensorflow can't find gpus on my machine. my working environment is as follows
Python 3.11.4 Ubuntu 22.04.4 LTS
warnings when I run the code
It would be very appreciated if you coud provide some suggestions.