Table of Contents
TensorFlow is a open-source machine learning framework mainly developed by Google. It can be used for verious machine learning tasks, e.g. deep learning. TensorFlow provides a high level API in python and other languages and is can run on CPUs as well as GPUs.
It is recommended to use Anaconda to create a virtual python environment and install the desired version of tensorflow within that environment.
module load anaconda3 source $ANACONDA3_ROOT/etc/profile.d/conda.sh conda create -n myenv python=3.8.8 conda activate myenv conda install tensorflow-gpu==2.2.0
If you do not want to use GPUs simply replace the last line with
conda install tensorflow==2.2.0
Testing the installation
To run TensorFlow on GPUs, load the correct modules and submit a job to the gpu partition.
#!/bin/bash #SBATCH -p gpu #SBATCH -t 1 #SBATCH --gpus-per-node 1 module load anaconda3 source $ANACONDA3_ROOT/etc/profile.d/conda.sh conda activate myenv python tftest.py
import tensorflow as tf tf.compat.v1.disable_eager_execution() hello = tf.constant('Hello, TensorFlow!') sess = tf.compat.v1.Session() print(sess.run(hello))
And then submit the job using Slurm:
The output file should contain
The output (if any) follows: b'Hello, TensorFlow!'
and also information about the GPUs selected.
Testing CPU only installation
If you want to test a CPU only installation, you can just run the tftest.py on a login node.
You can now use TensorFlow in your python scripts. Please read gpu_selection for more information about GPU usage.