Jetson Nano Getting Started Notes

page last updated on 1/4/2021

Notes Based on Getting started page from NVIDIA

https://developer.nvidia.com/embedded/learn/get-started-jetson-nano-devkit#intro

These notes are ideal for somebody that is familiar with the Raspberry Pi and should be used in conjuntion with the NVIDIA documentation.

SD Card Creation

After downloading the image. 

Used the Raspberry Pi Imager to write the SD card image which can be found here:

https://www.raspberrypi.org/blog/raspberry-pi-imager-imaging-utility/

Select “Use custom” and select JETSON NANO SD CARD IMAGE file and the target SD card.

Configure the operating system

Power up board with the new SD card

Follow the configuration step prompts

Select your language...

Select your timezone...

Configure your user account...

The operating system is ready to go...

Install the "Hello World" example

(this setup is without Docker)

We installed and built the “Hello World” project per the instruction found here:

https://github.com/dusty-nv/jetson-inference/blob/master/docs/building-repo-2.md#downloading-models

Command summary:

Run from the command prompt -

1. sudo apt-get update
2. sudo apt-get install git cmake libpython3-dev python3-numpy
3. get clone –recursive https://github.com/dusty-nv/jetson-inference
4. cd jetson-inference
5. mkdir build
6. cd build
7. cmake ../
8. make -j$(nproc)
9. sudo make install
10. sudo ldconfig

Some step details...

during the "cmake ../" command the following prompts will allow some selections..

Select models to download, we left the defaults and selected "OK"

Select PyTorch for installation.  This is unselected by default...

Select PyTorch, then "OK"

This step takes a while...

Next execute the Make command

Next run the "sudo make install" command

Finally execute the "sudo ldconfig" command

Confirm PyTorch is up and running...

Start python3 (python (2) will not work)

import torch   (imports the PyTorch library)

x=torch.rand(5,4) (creates a tensor with random values)

print(x)  (prints the tensor)

The basic PyTorch XOR example should also run.  

You should have a basic Jetson Nano system up and running.