Overview of Machine Learning and Deep Learning. Brief introduction to different types of BI Reporting tools like Power BI, SSMS, Cortana, Azure ML, TenserFlow and other tools.
8. What is Machine Learning?
• Using Computer Power to gain insight into
data that might other be elusive
• Credit Card Fraud Detection
• Online Shopping recommendations
• Self driving Cars and Tesla Motors
• Tuning data into Solutions
• Machine Learning Strategies
• Supervised Learning
• Unsupervised Learning
11. Prepare
Data
• Import data
• Pre-process data
Train
Model
• Select Learning Algorithm and Build Model
• Experiment/Iterate/Evaluate
Make
Operational
• Prepare Model for Deployment
• Deploy and call fro Applications
14. Bing maps
ships with ML
Traffic
Prediction
Service
Microsoft Kinect
can watch users
gestures
Computers
work on
users behalf,
filtering junk
email
Successful real-time
speech-to-speech
translation
Microsoft Search
Engine build with
Machine Learning
Enables data
mining of
databases
MICROSOFT MACHINE LEARNING HISTORY
15. SQL Server
Data Mining
Spam filtration Gestures
understanding
in Microsoft
Kinect
Azure Machine
Learning
Using Data
Mining in
search engines
Bing Maps started
to use ML for
traffic estimate
Voice recognition
Microsoft & Machine Learning
1999 201220082004 201420102005
19. Microsoft Azure
Machine Learning
Reduced Complexity
Access Through Web Browser, no
need to install any thing
Collaborate work with anyone
Visual composition, easy to use,
No Coding
Good storage of Algorithm (Use
in Bing search, Xbox..)
Have good support for R studio,
Python and Jupyter notebook
21. Azure ML Services
pre-configured environment for deep learning using GPU instances
Deploy in minutes
Operationalize models
as web services with a
single click; monetize in
Azure Machine Learning
Marketplace
Flexible
Built-in collection of
best of breed
algorithms with no
coding required. Drop
in custom R or use
popular CRAN packages
Integrated
Drag, drop, and connect
interface. Data sources
with just a drop down
run across any data.
Fully managed
No software to install,
no hardware to manage;
all you need is an Azure
subscription.
22. Azure Machine Learning Solution
Azure ML Studio
Browser-based
Designed for people without deep data science
backgrounds
Supports deep science scenarios – R support,
multiple models
Azure Marketplace
Drag-and-deploy
Fast monetization of ML solutions and APIs
Quick source for free and third-party Azure ML
APIs
Azure cloud services
No software to install or infrastructure needed
Nearly unlimited file repositories via Azure Storage
Supports Azure data-related services – HDInsight,
SQL Database
Azure ML API
REST-based web service
Supports best-in-class algorithms
Reduces time from model experimentation to
production
23. Azure ML Studio
Browser-based environment supporting general users
and data scientists
Immutable library of models including search, discover,
and reuse
Wide range of features, machine learning algorithms,
and modeling strategies
Ability to quickly deploy models as Azure web services
to the ML API service
New experiment flow
Streamlined experiment page
New visualization for data tables
Azure ML Studio
24. Azure ML API
Web-service and REST-based for easy creation and fast
deployment
Allows general users and data scientists to run models
as web services in minutes
Build apps that are easily accessible as web services,
app plug-ins, or even mobile apps
Supports advanced data science, including R coding
and 350 R packages included
Custom data ingress and egress
Extends ML Studio with customization
Rich functionality – rules engine, R
support, optimizer, simulation
25. AZURE MACHINE LEARNING STUDIO
• https://studio.azureml.net
• Online IDE to build, test, and deploy machine learning models
• Drag and drop “modules”
34. SSAS Vs Azure ML
Features Usability Cost Support
• End to end Product
• Canned algorithm
• Not possible to change
algorithm
• DMX Code
•More Visual
• Excel Version
•It’s not easy to start
•All users can use
If you
purchase
SQL Sever:
Free
Few books and
small online
community
• Current and up to date
algorithm
• Integration with R and
Python
• Cloud base
• REST format
• Hard to interpret
• Drag and Drop UI
• Customize the
Algorithm
Free version,
limited
options
More online
Community
Back in the 90s when the post office was wrestling with this issue, we were also working on Machine Learning, starting in 1991 when Microsoft Research was formed.
As early as 1999 they were using it to help create email filters by predicting which emails were junk, and which were relevant.
And as John Platt mentions—it’s a key technology that Microsoft uses to develop its own software. In 2004. Machine learning was part of Microsoft’s search engine
It is also used in Bing Maps as part of the traffic prediction service.
And many people know about how it was a key technology to make Kinect a reality, letting computers track people’s gestures and sort through what’s relevant and what’s not. Like filtering out a dog in the background to see a player’s movements.
And today, this technology that has been developed over decades is becoming available commercially as part of Azure
It’s this depth of experience with machine learning, testing and refining over years, using it to develop pretty much all Microsoft products, that makes Microsoft’s solution so robust.
What gives the Azure ML solution so much flexibility is largely the Azure ML API. This API allows customers to build powerful ML solutions, customize Azure ML Studio to their particular needs, and integrate Azure ML into other data analysis solutions and software.
And just like Azure ML Studio, the Azure ML API is accessible by users who are not sophisticated when it comes to advanced data analytics, but it also supports the needs of those who are. By enabling the API as a REST-based web service, we have made using it to run and publish models very easy. But by including richer functionality, including a rules engine, R support with 350 included packages, an optimizer, and simulation tools, we have also given it depth enough to address even the most advanced scenarios.