How to Install Tensorflow 1.15 for Jetson™ Nano™?

How to Install Tensorflow 1.15 for Jetson™ Nano™?

Jetson Nano

29 July 2021
ENVIRONMENT

Hardware: DSBOX-N2 (with NVIDIA® Jetson™ Nano™ module)

OS: Ubuntu 18.04 LTS, JetPack 4.5

Installing Tensorflow 1.15 on Ubuntu 18.04 for NVIDIA® Jetson™ Nano™


In this blog post, we will install Tensorflow to NVIDIA® Jetson™ Nano™.

 First, we should update the system and install the necessary files;


sudo apt-get update 
sudo apt-get install libhdf5-serial-dev hdf5-tools libhdf5-dev zlib1g-dev zip libjpeg8-dev liblapack-dev libblas-dev gfortran
sudo apt-get install python3-pip
sudo pip3 install -U pip
sudo pip3 install -U pip testresources setuptools==49.6.0


The processes below may take a couple of minutes to finish;

sudo pip3 install -U pip testresources setuptools numpy==1.16.1 future==0.17.1 mock==3.0.5 h5py==2.9.0
keras_preprocessing==1.0.5 keras_applications==1.0.8 gast==0.2.2 futures protobuf pybind11
sudo pip3 install --pre --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v45 'tensorflow<2'

Now, the setup for Tensorflow is ready. When we import it in Python, we should enable the eagerly execution option, which is disabled by default. 


For starting up;

 

python3 
import tensorflow as tf
tf.enable_eager_execution()


Now let us test the Tensorflow with a basic operation;



We should see this result;



If you see the result above, Tensorflow is ready for use.


To check whether Tensorflow uses GPU or not, simply enter a command after starting Tensorflow and read the last lines;



Above command shows us that the Tensorflow device is working on physical GPU:0 device, which is what we are looking for.

Thank you for reading our blog post. 


Türkçe