A Guide for Deploying AWS Panorama

This guide shows an end-to-end setup of an AWS Panorama deployment from the physical device setup to deploying a sample notebook. It includes steps to specify a deployment in a particular AWS region.

Deploy AWS Panorama

Prerequisites

AWS Account
IAM role in this account with the AWSPanoramaFullAccess permission

Create AWS Panorama Device

These instructions are also documented here: https://docs.aws.amazon.com/panorama/latest/dev/gettingstarted-setup.html

Login in to the AWS Management Console
In the top menu bar, region drop down, select your AWS region, e.g. Sydney (ap-southeast-2)
Search for and open AWS Panorama
Select Get Started
Select Add device
Select Begin setup

Specify Name

Enter a Name – should be descriptive and unique – e.g. project-name-panorama
Enter a Description
Add new tag – Key Name, Value = Panorama
Select Next

Power on

Plug in the cables, and power on the device
Select Next

Configure

Connect to a network via Ethernet in order for the device to access cameras on your local network and in order to manage the device from the AWS cloud.
Select Next

Download file

Download the configuration archive file. This is a zip file. Do not unzip it once it is downloaded.
The configuration file will expire in 5:00 mins. Proceed to the next step to complete this process.
Transfer configuration file to the USB storage device
Select Next

Done

Plug in the USB storage device on your device.
Once complete the device will appear in the list of devices in the AWS Management console

Setup the Development Environment

The setup below is based on the instructions in https://github.com/aws-samples/aws-panorama-samples/blob/main/docs/EnvironmentSetup.md

Create an Amazon S3 Bucket for the AWS Panorama Model Files

In the AWS management console, search for S3, select S3
Select Create bucket
Enter bucket name: project-name-panorama-bucket-aws-accountid
Select the AWS region: ap-southeast-2
Select Create bucket
Copy the bucket name in a notepad.

Subscribe to Ubuntu Bionic in the AWS Marketplace

In the AWS management console, search for Marketplace, select AWS Marketplace Subscriptions
Select Discover products
Search for Ubuntu 18.04 LTS – Bionic (Arm)
Select Ubuntu 18.05 LTS – Bionic (Arm)
Select Accept Terms
You should see a message: Thank you for subscribing to this product! We are processing your request.
Wait until message changes to: Thank you for subscribing to this product! You can now configure your software.

Create a PEM key

In the AWS management console, search for EC2, select EC2
In the left side menu, under Network and Security, select Key Pairs
Select Create key pair
Enter name: project-name-panorama-server-v1
Select Create key pair
A project-name-panorama-server-v1.pem file will download. Note its location.

Deploy the Amazon CloudFormation Stack

In the AWS management console, search for CloudFormation, select CloudFormation
Select Create stack, Select With new resources (standard)
Select Upload a template file
Select Choose file
In the window popup, find and select the ec2-instance-panorama-v2.yml file, select Open
Select Next
Enter a stack name: project-name-panorama-server-v1
Select KeyName: project-name-panorama-server-v1
Select a VpcId from the drop down list – a Virtual Private Cloud (VPC) is a logically isolated virtual network in your AWS account – your IT department may have a preference for where this should be deployed.
Select Next
In the Configure stack options screen, select Next
In the Review screen, at the bottom of the page check I acknowledge that AWS CloudFormation might create IAM resources.
Select Submit
The screen should change to an Events page with a CREATE_IN_PROGRESS message
Wait about 3 minutes until status is CREATE_COMPLETE
Select Outputs
Copy the publicName (the public DNS) value to a notepad. It will have a value in the format: ec2-12-34-567-89.ap-southeast-2.compute.amazonaws.com

Connect to the Amazon EC2 Instance

Locate your private key file. The key used to launch this instance is project-name-panorama-server-v1.pem
Run this command, if necessary, to ensure your key is not publicly viewable. chmod 400 ~/Downloads/computer-vision.pem
Use SSH to connect to your instance using its Public DNS ssh -i ~/Downloads/computer-vision.pem [email protected]
Check the bootup script is deploying tail -f /var/log/cloud-init-output.log
The installation process takes about 30 minutes.
When complete you should see the message echo 'INSTALLATION COMPLETE'

Launch Juypter Lab

Launch the Juypter Lab session sudo jupyter-lab --no-browser --allow-root
You should see a response similar to: http://ip-123-45-678-910:8888/lab?token=e#############################
Copy the port number (8888) and the token value ( e############################# ) to a notepad.
Open another SSH terminal on your computer. ssh -i ~/Downloads/computer-vision.pem -NL 8157:localhost:8888 [email protected]
Launch a web browser on your computer and open: http://localhost:8157/
Paste the token number
Select Log in
You should now be in the Jupyter Lab session

Creating Notebook in an AWs account separate from the AWS Panorama AWS account

This step is only required if the notebook is running in an account separate from the AWS Panorama AWS account.

Create an AWS IAM User

Create an IAM user with IAM permissions to deploy code to the AWS Panorama device.
Record the access key and secret access key for use in the Jupyter Lab session

Configure the AWS CLI

In a terminal session on this Jupyter notebook server, run aws configure. This allows this notebook server to access Panorama resources and deploy applications on your behalf.

Modify the AWS Panorama Sample - People Counter

Open the Juptyer notebook under: aws-panorama-samples/samples/people_counter/people_counter_tutorial.ipynb

Code changes

/aws-panorama-samples/samples/common/test_utility/panorama_test_utility.py

line 15 panorama_client = boto3.client('panorama', region_name='ap-southeast-2') /aws-panorama-samples/samples/people_counter/people_counter_tutorial.ipynb

Set up

s3_client = boto3.client('s3', region_name='ap-southeast-2') panorama_client = boto3.client('panorama', region_name='ap-southeast-2')

Notebook parameters

Device ID, should look like: device-oc66nax4cgzwhyuaeyifrqowue DEVICE_ID = 'device-1234567890' # input( 'DEVICE_ID (format: device-*)' ).strip()

Enter your S3 bucket info here S3_BUCKET = ‘project-name-panorama-bucket-aws-accountid' # input( 'S3_BUCKET' ).strip()

Enter your desired AWS region AWS_REGION = 'ap-southeast-2' # input( 'AWS_REGION (e.g. us-east-1)' ).strip() ML_MODEL_FNAME = 'ssd_512_resnet50_v1_voc'

Prepare model for testing with notebook server Compile the model to run with test-utility. This step takes 7 mins ~ 10 mins.

%run ../common/test_utility/panorama_test_utility_compile.py

--s3-model-location s3://{S3_BUCKET}/{app_name}/

--region 'ap-southeast-2'
--model-node-name model_node
--model-file-basename ./models/{ML_MODEL_FNAME}
--model-data-shape '{model_data_shape}'
--model-framework MXNET

Create New Camera

CAMERA_NAME = "test_rtsp_camera" CAMERA_CREDS = '{"Username":"root","Password":"password","StreamUrl": "rtsp://192.168.0.1/onvif-media/media.amp?profile=profile_1_h264&sessiontimeout=60&streamtype=unicast"}'

Match the camera login details
Test camera with VLC player first to check you can see the video stream

jacobcantwell / deploying-aws-panorama Goto Github PK

deploying-aws-panorama's Introduction