Tensorflow GPU not working in WSL2

General Discussion
,
#1

Hi, so I installed TensorFlow in WSL2 following the methods mentioned in this article except for TensorRT. Well now TensorFlow is installed, and it detects my GPU, but when I run a TensorFlow model, it runs on my CPU. I am new to deep learning, I guess we don’t have to pass anything like "device=‘GPU’/‘cuda’ ", to get the model run on GPU, right?

When I run the model in VS Code in WSL, the CPU bar rises in the “Task Manager”, but the GPU bar is idle.

It will be help if you can make a video of this process, as going forward TensorFlow GPU support is only available in WSL.

Thanks in advance.

#3

Could you share the output of the snippet ?

import tensorflow as tf
tf.config.list_physical_devices('GPU')

If you would like a particular operation to run on a device of your choice please follow manual device placement. There you can use with tf.device to create a device context and all the operations within that context will run on the same designated device. Thank you.