User Tools

Site Tools


wiki:ghc:ghc18:iw533

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

wiki:ghc:ghc18:iw533 [2018/10/05 20:51] (current)
tina_wu created
Line 1: Line 1:
 +====== Hands on with Deep Learning and IoT - Conference Notes ======
 +
 +
 +===== Presenters =====
 +Shruti Karulkar, Anoush Najarian, Louvere Walker-Hannon
 +
 +===== Abstract =====
 +Let’s get started with Deep Learning and the Internet of Things! We’ll do hands-on exercises: you’ll use a webcam and a neural network to recognize images , aggregate data, and run real-time IoT analytics. Our goal to get you excited about IoT and Deep Learning, and to set you up for success with maker projects in your community after the conference.
 +
 +====== Session Notes ======
 +===== Deep Learning =====
 +  * Intelligently uses the foundation of machine learning
 +  * Automates feature selection
 +  * Workflow: ​
 +       - Gather data 
 +       - Train from scratch OR use pretrained network
 +       - Train until acceptable level of accuracy
 +       - Test network (eg. AlexNet)
 +  * Objects are not correctly classified all the time due to:
 +       * Lighting
 +       * Position
 +       * Other objects
 +       * Whether the network was trained well
 +
 +===== Internet of Things (IoT) =====
 +  * All around us
 +       * Eg. fitness trackers, Alexa, etc.
 +  * Workflow
 +       * Collect data
 +       * Analyze (trends, plots, ML)
 +       * Act (send alerts, social media)
 +
 +===== ThingSpeak =====
 +  * Will be used to aggregate all our data and analyze it
 +  * Open source IoT platform
 +
 +
 +Hands On Exercises:
 +> Create a Mathworks account at https://​www.mathworks.com/​mwaccount/​
 +> Run the following in MATLAB online
 +    unzip('​https://​tinyurl.com/​ghc18deeplearning'​)
 +
  
wiki/ghc/ghc18/iw533.txt · Last modified: 2018/10/05 20:51 by tina_wu