A DEVELOPER’S INTRODUCTION TO AZURE MACHINE LEARNING
JOHN GARLANDPRINCIPAL CONSULTANT & INSTRUCTORWINTELLECT@DOTNETGATOR
@ATLCODECAMPHTTPS://ATLANTACODECAMP.COM/2016
Platinum Sponsors
Gold Sponsors
ABOUT ME•Principal Consultant & Instructor at
ConsultingCustom software application development & architectureInstructor Led TrainingMicrosoft’s #1 training vendor in DevDivOn-Demand TrainingWorld class, subscription-based online training
• Author•Windows Store Apps Succinctly• Programming the Windows Runtime by Example
•Microsoft Azure MVP & Advisor
WHAT IS MACHINE LEARNING?• Using computer power to gain insight into
data that might otherwise be elusive• Credit-card fraud detection• Online shopping recommendations• Self-driving cars and more
• “Turning data into software”• Machine Learning Strategies• Supervised learning• Unsupervised learning
TRADITIONAL PROGRAMMING VS MACHINE LEARNING
Program Data Output Data Output Progra
m
Traditional Programming
Machine Learning
MACHINE LEARNING PROCESS
PrepareData
• Import data• Pre-process data
TrainModel
• Select Learning Algorithm & Build Model• Experiment/Iterate/Evaluate
MakeOperationa
l
• Prepare model for deployment• Deploy & call from applications
AZURE MACHINE LEARNING
• Fully managed cloud service for building and operationalizing ML models
Fully ManagedNo software to install, no hardware to manage, and one portal to view and update.
Integrated
Simple drag, drop, and connect interface. No need for programming for common tasks.
Best in Class Algorithms + RBuilt-in collection of best of breed algorithms. Support for R and popular CRAN packages.
Deploy in MinutesOperationalize models with a single click. Monetize in Machine Learning Marketplace.
AZURE MACHINE LEARNING HISTORY
AZURE MACHINE LEARNING STUDIO•https://studio.azureml.net•Online IDE to build, test, and deploy machine learning models•Drag and drop “modules”
Demo
PRICINGFree Tier Standard Tier
Azure Subscription
Not Required Required
Max Modules 100 Per Experiment UnlimitedMax Duration 1 Hour Per Experiment 7 Days Per Experiment, 24 Hours Per
ModuleMax Storage 10GB BYO (Azure Blob)Performance Single node Multiple nodesSLA No Yes
$9.99/Seat/Month$1/Studio Experiment Hour$2/Production API Compute Hour$.50/1,000 Production API Transactions
OTHER RESOURCES
Free e-book
http://bit.ly/a4r-mlebook
Not free book
http://bit.ly/pa-mlbook
http://bit.ly/Build2016Videos
Surveys and Prizes Please complete the session and event surveys!1 ticket per session survey1 ticket for the event survey1 ticket for completing the booth game
Drawing for prizes begins at 5pm in Q202
THANK YOU!
John GarlandPrincipal Consultant & InstructorWintellect@dotnetgator
A link to these slides will be posted on Twitter with a reference to @ATLCodeCamp