Suphabnk

IT-Dev blog

Setup Nvidia Jetson Orin Nano 8G - reComputer J3011 (Seeed Studio)

Posted

HOST OS

Ubuntu 22.04.5 LTS (Jammy Jellyfish)
Linux x86_64 GNU/Linux

Setup Nvidia SDK Manager

  1. https://developer.nvidia.com/sdk-manager#installation_get_started
sudo apt update
sudo apt install qemu-user-static sshpass abootimg nfs-kernel-server libxml2-utils binutils -y
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
sudo dpkg -i cuda-keyring_1.1-1_all.deb
sudo apt update
sudo apt list --upgradable
sudo apt upgrade
sudo apt -y install sdkmanager
sdkmanager

Upgrade JetPack 6.2

  1. https://wiki.seeedstudio.com/reComputer_J4012_Flash_Jetpack/
  2. https://collabnix.com/how-to-upgrade-jetpack-5-x-to-6-x-on-nvidia-jetson-orin-nano-super/
lsusb 
sudo vi /boot/extlinux/extlinux.conf
{
  ... apparmor=1 security=apparmor
}
sudo snap revert snapd
snap download snapd --revision=24724
sudo snap ack snapd_24724.assert
sudo snap install snapd_24724.snap
sudo snap refresh --hold snapd
sudo apt install firefox
sudo apt install chromium-browser

Setup Jetpack

  1. https://developer.nvidia.com/embedded/downloads
  2. https://github.com/azimjaan21/jetpack-6.1-pytorch-torchvision-
  3. https://pytorch.org/audio/2.1.0/build.jetson.html
  4. https://docs.nvidia.com/deeplearning/frameworks/install-pytorch-jetson-platform/index.html
  5. https://docs.nvidia.com/jetson/jetpack/install-setup/index.html
  6. https://pypi.jetson-ai-lab.io/jp6/cu128
cat /etc/nv_tegra_release
sudo apt-cache show nvidia-jetpack
sudo apt install nvidia-jetpack nvidia-jetpack-dev nvidia-jetpack-runtime
sudo apt install python3-pip 
sudo apt install wget curl openssl net-tools iputils-ping zip unzip
sudo apt install libopenblas-dev libopenmpi-dev libomp-dev libjpeg-dev libpng-dev libtiff-dev
sudo pip install jetson-stats
sudo su
jetson_clocks
jtop
free -h && sync && echo 3 > /proc/sys/vm/drop_caches && free -h
exit

Setup Pytorch – OpenCV (CUDA)

sudo apt-get -y update
sudo apt install -y python3-pip libopenblas-dev
sudo apt install ffmpeg libavformat-dev libavcodec-dev libavutil-dev libavdevice-dev libavfilter-dev libjpeg-dev libpng-dev
sudo pip install cmake ninja
wget https://developer.download.nvidia.com/compute/cusparselt/redist/libcusparse_lt/linux-sbsa/libcusparse_lt-linux-sbsa-0.5.2.1-archive.tar.xz
tar xf libcusparse_lt-linux-sbsa-0.5.2.1-archive.tar.xz
sudo cp -a libcusparse_lt-linux-sbsa-0.5.2.1-archive/include/* /usr/local/cuda/include/
sudo cp -a libcusparse_lt-linux-sbsa-0.5.2.1-archive/lib/* /usr/local/cuda/lib64/
python -m pip install torch torchvision torchao torchaudio --index-url=https://pypi.jetson-ai-lab.io/jp6/cu128
python
{
import torch
import torchvision
import torchaudio
print(torch.__version__)
print(torchvision.__version__)
print(torchaudio.__version__)
torch.cuda.is_available()
torchaudio.utils.ffmpeg_utils.get_build_config()
exit()
}

For OpenCV, follow this building steps

python
{
  import cv2
  cv2.cuda.getCudaEnabledDeviceCount()
}
sudo apt install libopencv
sudo apt install libopencv-dev
sudo apt install libopencv-python
pip install ultralytics
pip install opencv-contrib-python
pip install imutils schedule screeninfo pyserial pytz onnx
wget https://nvidia.box.com/shared/static/6l0u97rj80ifwkk8rqbzj1try89fk26z.whl -O onnxruntime_gpu-1.19.0-cp310-cp310-linux_aarch64.whl
pip install onnxruntime_gpu-1.19.0-cp310-cp310-linux_aarch64.whl
pip install onnxslim

Author