Tensorflow Installation
System Requirements
Make sure your system is qualified for requirements (irrespective of operating system) on Requirements.
Select a image
While there are different releases (base tags) of Tensorflow, the instructions use the default latest version. In each base tag, there are multiple variants using two tags.
- tag-gpu - Specified tag release with GPU Support
- tag-jupyter - Specified tag release with Jupyter (includes Tensorflow tutorial notebooks)
You can mix both tags using tag-gpu-jupyter which includes Jupyter as well as GPU support. The following instructions uses this tag as default.
podman pull tensorflow/tensorflow:latest-gpu-jupyter
Info
If Podman prompts for selecting repository, select docker.io (Docker Hub) repository.
Check if podman image is working
Use the following command to check if image is working.
podman run --rm --security-opt=label=disable --hooks-dir=/usr/share/containers/oci/hooks.d/ docker.io/tensorflow/tensorflow:latest-gpu-jupyter python -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
In the following command, a container is created, SELinux is configured to ignore the container, NVIDIA hook is connected and runs the corresponding command:
python -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]
[], please check if you have missed any installation steps.
Created: July 7, 2023