TensorFlow 2016

From HPC users
Jump to navigationJump to search


TensorFlow™ is an open source software library for high performance numerical computation. Its flexible architecture allows easy deployment of computation across a variety of platforms (CPUs, GPUs, TPUs), and from desktops to clusters of servers to mobile and edge devices.

Installed version

The currently installed versions are

on the environment hpc-env/8.3


on the environment hpc-env/6.4

TensorFlow/1.13.1-foss-2017b-Python-3.6.3   (Not built with GPU support!)

on the environment hpc-uniol-env


Using TensorFlow

If you want to find out more about Tensorflow on the HPC Cluster, you can use the command

module spider TensorFlow

This will show you basic informations e.g. a short description and the currently installed version.

To load the desired version (e.g. 1.8), you must load the module as well as the environment:

module load hpc-env/6.4
module load TensorFlow/1.8.0


For more information about how to use TensorFlow, you can take a look at the tutorial.

Further information are also available on the homepage