Edge Impulse is a robust platform for building and deploying machine learning models on edge devices, enabling powerful on-device AI capabilities. Although ALPON X4 isn’t natively supported, this guide provides a streamlined workaround using a containerized deployment on ALPON Cloud. By leveraging a custom Dockerfile and an entrypoint.sh script, you can run the Edge Impulse runner with an HTTP server on your ALPON X4, unlocking the power of edge AI.

For advanced configuration details, refer to the Edge Impulse Docker Documentation.

Prerequisites

Before diving in, ensure you have the following ready:

• ALPON X4 Device: Your device must be powered on, operational, and connected to ALPON Cloud.
• Edge Impulse API Key: Obtain this from your Edge Impulse project settings to authenticate the runner.
• ALPON Cloud Access: Log in to your account at ALPON Cloud to manage your device and deploy containers.

Step 1: Prepare the Dockerfile

Create a file named Dockerfile on your local machine to define the containerized environment for the Edge Impulse runner:

FROM public.ecr.aws/g7a8t7v6/inference-container:b059854aa82274b16d242ced0892ef9fea15b4df

COPY ./entrypoint.sh /entrypoint.sh

RUN chmod +x /entrypoint.sh

ENTRYPOINT ["/entrypoint.sh"]

Important: Always check the Edge Impulse documentation for the latest inference container image tag to ensure compatibility and access to the most recent features.

Step 2: Create the entrypoint.sh Script

The entrypoint.sh script launches the Edge Impulse runner in HTTP server mode, enabling remote interaction with your machine learning model. Create a file named entrypoint.sh with the following content:

#!/bin/bash
node /app/linux/node/build/cli/linux/runner.js --api-key $API_KEY --run-http-server $HTTP_PORT --force-variant $VARIANT

Note: You can customize this script based on your application’s needs, such as adding additional parameters or modifying the runner behavior. Ensure the file has executable permissions (use chmod +x entrypoint.sh before building).

Step 3: Build the Docker Image

On your local machine, navigate to the directory containing the Dockerfile and entrypoint.sh, then build the Docker image optimized for ALPON X4’s architecture:

docker build --platform=linux/arm64 -t edge-impulse-runner:latest .

This command creates a container image tagged edge-impulse-runner:latest, ready for deployment.

Step 4: Upload the Image to ALPON Cloud

• Log in to the Sixfab Connect platform, navigate to the Sixfab Registry page
• Click on + Add Container and follow the prompts to push container to Sixfab registry.
  
  

  📘

  Manage and Deploy Applications

  Visit the Manage & Deploy Applications page for all the necessary details on pushing your container image to the Sixfab Registry.

Step 5: Deployment Configuration

1. Go to the Application section of your asset on Sixfab Connect.

2. Click the + Deploy button to configure and deploy the container.

3. In the Deploy Container window, use the following settings:
   
   

   • Container Name: edge-impulse

   • Image: Select the Datadog image and tag pushed to the Sixfab Registry.

   • Ports: Click "+ Add More" in the Ports section and add the following ports:

     From	To
     31337	1337

     
     

   • Environment: Click "+ Add More" in the environment section and add the following values:

     Key	Value
     API_KEY	ei_1234... (your Edge Impulse API key)
     HTTP_PORT	1337
     VARIANT	int8

     Note: The VARIANT must be set to int8, float32, or akida, depending on your model’s quantization requirements. Check your Edge Impulse project settings to confirm the appropriate variant.

   • Click the "+ Deploy" button to deploy Grafana.

Final Step: Access Your Edge Impulse Model

Congratulations! Your Edge Impulse runner is now deployed on your ALPON X4. To interact with your machine learning model:

Retrieve the IP address of your ALPON X4 from the ALPON Cloud device interface.

In a web browser or API client, navigate to http://<DEVICE_IP_ADDRESS>:31337 to access the HTTP server and interact with your model.

Test your model by sending inference requests as outlined in the Edge Impulse documentation.

With your Edge Impulse model running on ALPON X4, you’re ready to harness the power of edge AI for real-time predictions. Happy deploying!