Chooch Edge AI Device Setup Guide

 

Welcome to the Chooch Edge AI Setup.

 

Within the dashboard, this guide will walk you through the steps to setup an edge device running Chooch AI models both for Intel PCs and NVIDIA Jetson devices. The guide will explain how to use the Chooch AI dashboard to create a device, then add camera streams and add AI models to the device. Finally, you will run a script that installs the AI models on the hardware and access the Chooch Edge Control to see the AI in action.

 

Please note you will need a free account on the AI Platform. If you don’t yet have one, please sign up here. For the moment, you will also need an Enterprise account. Please request one here. Or learn more about Edge AI.

 

Hardware Requirements

CPU

Compatible Processors

• 8th to 10th generation Intel® Core™ > i7 (suggested i9)

• 3rd generation Intel® Xeon® Scalable processors

Compatible Operating Systems

• Ubuntu* 18.04.3 LTS (64 bit)

Memory

• min 16GB ram (suggested 32GB)

Disk

• 256GB SSD harddisk

3rd Party Software Requirements

• Docker

GPU

Compatible Graphics Cards

• Tesla P100

• Tesla T4

• Tesla V100

• DGX-A100

• DGX-1

• GeForce RTX 2080 Ti

• GP100

• GTX 1080

• GTX 1070

• GTX 1060

• GTX 1050

• GTX 1030

• Quadro RTX 4000

• Quadro RTX 6000

• Quadro RTX 5000

• RTX 2080

• RTX 2070

• Tesla A100 (GA100)

• Tesla P4

• Tesla P40

• Tesla Xp

• XNOR Tensor Cores

Compatible Operating Systems

• Ubuntu* 18.04.3 LTS (64 bit)

Memory

• min 16GB ram (suggested 32GB)

Disk

• 256GB SSD harddisk

3rd Party Software Requirements

• Nvidia Drives

• Nvidia Docker

Jetson Devices

• Jetson AGX Xavier

• Jetson Xavier NX

3rd Party Software Requirements

• Nvidia drivers and Jetpack 4.4.1

Installation Requirements

PC CPU Setup Guide

Installation requirements for Ubuntu Server for CPU:

Please follow the docker installation instructions
https://docs.docker.com/engine/install/ubuntu/

PC GPU Setup Guide

Notes:

-An Internet connection is needed to perform many of the following steps

-Many of the following command lines will require your system’s root or super user password

 

Installation requirements for Ubuntu server – GPU:

If Ubuntu has not been installed, please use the following link. If Ubuntu has been previously installed, please skip to next step.

 

https://releases.ubuntu.com/18.04/

 

1. Install Nvidia Driver

 

After installing and setting up Ubuntu, enter the following script in a command prompt to ensure your Ubuntu is up to date.

sudo apt update

Enter the following script in a command prompt to check for Ubuntu upgrades.

sudo apt upgrade

Enter the following script in a command prompt to show all driver packages that apply to the current system.

sudo ubuntu-drivers devices

2. NVIDIA GPU:

 

Enter the following script to check your NVIDIA GPU’s status.

nvidia-smi

Enter the following script to update your NVIDIA GPU’s drivers.

sudo apt install nvidia-driver-xxx ( which is recommended )

Enter the following script to install curl (data transfer tool).

sudo apt install curl

Enter the following script to install Docker-CE (SW package installer) via  Docker’s convenience script.

curl https://get.docker.com | sh 
&& sudo systemctl --now enable docker

Enter the following script to set up a stable repository (binary packages) and GPG key (encryption).

distribution=$(. /etc/os-release;echo $ID$VERSION_ID) 
&& curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - 
&& curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list

3. Updates and Installers

 

Enter the following script to update Ubuntu

sudo apt-get update

Enter the following script to update NVIDIA dockers.

sudo apt-get install -y nvidia-docker2

Enter the following script to restart the Docker daemon (container) to complete the installation.

sudo systemctl restart docker

Enter the following script to test NVIDIA docker installation.

sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

Jetson Setup Guide

Installation requirements for Jetson Device:

https://developer.nvidia.com/embedded/learn/getting-started-jetson
Please follow the installation guide of your device model and please be sure your Nvidia drivers and jetpack are installed before starting installation of Chooch Edge Service.

Jetpack 4.4.1:
Download Jetpack 4.4.1

Jetson NX:
https://drive.google.com/file/d/1a5v-kVXzrUNnkglhWC2KpL8sh_dGydHK/view?usp=sharing

Jetson AGX:
https://drive.google.com/file/d/1Q2ZR8MEO4-7VJs0YMYwb9Z7pDGXuOQ48/view?usp=sharing

Please follow the installation guide for your device’s model and please be sure the Nvidia drivers and Jetpack are installed before installing the Chooch Edge Service.

Installation Chooch Edge Service:

1. Login to your Chooch AI Dashboard

 

2. Go to Devices

3. Create a device

4. Add a stream to your device

5. Add Models to your Stream

6. Copy the Device ID of the device you want to connect. The Device list can be found here.

 

7. Enter the following script to install git (file coordination program).

sudo apt install -y git

8. In /home directory of your device run the following code on terminal:

Run the installation script.

git clone https://github.com/choochtech/chooch_ai_on_prem_install.git && 
cd chooch_ai_on_prem_install && 
./install.sh

This installation time may take approximately 60 minutes or more, based on your network upload speed.

You will be prompted for your Device ID, enter the Device ID of the Device you created.

The installation will pull the latest docker and create a chooch_predict_on_prem.service as a systemctl service.

 

9. You can access your Chooch Edge Control Panel on the device after installation, from http://localhost:8000

 

Some Useful commands start, stop and restart the service on device.

 

Start

sudo systemctl start chooch_predict_on_prem.service

Status

sudo systemctl status chooch_predict_on_prem.service

Restart

sudo systemctl restart chooch_predict_on_prem.service

Stop

sudo systemctl stop chooch_predict_on_prem.service
×

Oops, something went wrong.
Please try back later.

Click here Tap here to upload Upload your image
and Chooch will recognize it.
Supported formats are .jpg .jpeg .png

×