Take It to the Edge: Classification and Cloud Connectivity

Many times, we focus on performing machine learning locally on-device, with no need for internet connectivity. This reduces latency, simplifies projects, and can sometimes reduce costs. However, some projects do require connectivity to the cloud in order to set a status, update a dashboard, log an activity or event, or allow a user to interact with a device. This month we're highlighting two example projects that use Edge Impulse plus Blues Wireless to send data to the cloud.

Follow the steps to complete this activity and be entered into our giveaway to win December's prize bundle - a Blues Starter Kit and Edge Impulse Swag.

Join the Edge Impulse community at hackster.io/EdgeImpulse
Sign up for an Edge Impulse account at studio.edgeimpulse.com/signup
Review resources below.
- Resource #1: Appliances Identifier Smart Energy Meter Project
- Resource #2: Running Faucent Alert System Project
Complete the quiz below to show us what you learned!

Submissions close January 1st at 11:59PM Pacific.

Name*

First Name Last Name

Email*

example@example.com

Hackster Profile URL Link*

Please include the entire URL link to your Hackster Profile: e.g. https://hackster.io/user-name

Email Used to Sign Up to Edge Impulse*

Please include the email address you used to create an Edge Impulse account.

Country*

Street Address

Street Address Line 2

1. In machine learning projects, using Transfer learning allows you to:*

Take an existing model and fine-tune it with your own data to classify to the labels you care aboutCreate a net-new model that has no existing layers in order to maximize training timeRemove existing data from a trained model in order to run it fasterNone of the above

2. A Mel Filterbank Energy (MFE) block is great for processing what type of machine learning data?*

Voice audioNon-voice audioColor depthImage features

3. Once the Edge Impulse CLI is installed, this command will launch the service to login and connect a Raspberry Pi to the Studio:*

edge-impulseedge-impulse-linuxedge-impulse-rpiedge-impulse-camera

4. In the Edge Impulse Studio, the Data Acquisition screen is where you can:*

Capture and load up data for your training setProcess data into a modelTransfer data from one device to anotherCreate a neural network

5. This service from Blues Wireless can receive data from a Notecard and route data to cloud applications and services.*

AirnoteNote CarrierNotehubAirhub

Download and read Take It to the Edge terms & conditions and check the eligibility requirements before completing your entry.

By checking the box below I agree that I have read and accept the terms & conditions for participation in Take It to the Edge, and that my provided personal information may be shared with Edge Impulse and used to facilitate my participation in this program and for program-related emails.*

I agree

I would like to receive marketing news updates on new products, design tools and resources from Hackster.io and Edge Impulse. I understand that I can change my communication preferences or unsubscribe at any time.

Sign me up to receive marketing updates