Hardware
- NVIDIA Jetson device. While in this guide we used Jetson Nano, any other model should work pretty much the same way.
- 5V 3A power supply
- 128GB SD card
Software
- Synpse cloud account (https://cloud.synpse.net)
- Jetson SD card image from https://developer.nvidia.com/embedded/downloads
Step 1: Open the terminal
Open the terminal in NVIDIA Jetson by clicking on the LXTerminal. From the start, I would suggest assuming root permissions as we will be installing some system packages:
|
|
Step 2: Install curl
In the terminal, run:
apt-get install curl
Step 3: Install synpse
Login into your account on https://cloud.synpse.net, then go into the “Devices” page. Click on “provision”:
And copy the given command into the terminal:
In a few seconds, you should be able to see a new device registered in Synpse dashboard.
Next Steps
Remote SSH access to Jetson
Synpse provides a simple SSH method into your Jetson via the web UI or the CLI:
Once Synpse and Jetson are up and running, you can move your device between buildings, change internet connections, you will always be able to connect to it as long as there’s a working internet connection.
File copy to and from Jetson
Install the Synpse CLI (https://docs.synpse.net/cli/install-cli) on your local machine (if you don’t have one yet). Then, you can use it to download and upload files directly from/to your NVIDIA Jetson:
|
|
You can read more about this command using:
|
|
This can be used to:
- Upload your ML scripts
- Upload ML models, etc
