Skip to main content

Download Jetpack SDK

This article will guide you to install JetPack SDK and check the installation.

install JetPack SDK​

$ sudo apt update
$ sudo apt install nvidia-jetpack
$ apt depends nvidia-jetpack | awk '{print $2}' | xargs -I {} sudo apt install -y {}

( option ) update Jetpack to newer version​

$ sudo apt dist-upgrade
$ sudo apt install --fix-broken -o Dpkg::Options::="--force-overwrite"

install jetson-stats to manage package version​

$ sudo apt-get install python3-pip
$ sudo -H pip install -U jetson-stats

config cuDNN env​

$ echo ' ' >> ~/.bashrc
$ echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64' >> ~/.bashrc
$ echo 'export PATH=$PATH:/usr/local/cuda/bin' >> ~/.bashrc
$ echo 'export CUDA_HOME=$CUDA_HOME:/usr/local/cuda' >> ~/.bashrc

$ source ~/.bashrc

$ sudo apt-get install zlib1g libfreeimage3 libfreeimage-dev
restart docker service

We have installed the nvidia-container, need to restart the docker service to import the config

$ systemctl restart docker

test/ check the package​

check Jetpack version​

$ sudo apt show nvidia-jetpack

image(82).png

check jetson_release​

$ sudo jetson_release

image(83).png

test cuDNN​

$ nvcc -V

image(84).png

Building a cuDNN Dependent Program​

$ sudo cp -rv /usr/src/cudnn_samples_v8 ~/
$ cd ~/cudnn_samples_v8/mnistCUDNN/
$ sudo chmod 777 ~/cudnn_samples_v8/
$ sudo make clean && sudo make
$ ./mnistCUDNN

image(86).png

prepare jetson-containers​

$ git clone https://github.com/dusty-nv/jetson-containers
patch for jetson-containers

Lots of the AI demo need jetson-containers to prepare environment. But currently, the containers can't support the MIPI camera. Please download the patch, and run the below command.

$ cd jetson-containers
$ git apply 0001-add-v4l-subdev-vi-capture-channel-media-devices.patch
$ cd ../
$ bash jetson-containers/install.sh

Reference:​

How to Install JetPack
jetson-stats
cuDNN
FreeImage is not set up correctly. Please ensure FreeImae is set up correctly