SlideShare ist ein Scribd-Unternehmen logo
1 von 49
Top Azure Takeaways
from Microsoft Build
Azure webinar series
Welcome
How do I ask a question?
If you have a technical or content-related question, please use the Q&A window
We will address the questions as they come in
Can I view this presentation after the webinar?
Yes, this presentation is being recorded
A link to the recorded presentation will be sent to the email address you used to register
Tim Heuer @timheuer
Cloud Developer Advocate
David Smith @revodavid
Cloud Developer Advocate
Our
speakers
What did you miss at BUILD?
Project Brainwave
• FPGA: Field-programmable
gate arrays
• Hardware accelerated
inference
• Circuit board quality
detection
Model deployment
• Customvision.ai export
• ONNX
• Containers
• Azure AI Gallery
Azure ML packages
• Forecasting
• Computer Vision
• Text Analytics
• ML.net
1
What did you miss at BUILD?
AI Lab
• Drawing Bot
• JFK Files
• AirSim
• Drone Search and Rescue
AI for Accessibility
• Seeing AI
• Translator
• Custom Voice
• Custom Vision
• Vison AI Dev Kit
And more…
• Other announcements you
may have missed.
2
Top Azure Takeaways from Microsoft BUILD
Project Brainwave
1
Breakthroughs in deep learning
demand real-time AI
Deep neural networks have enabled major
advances in machine learning and AI
Computer vision
Language translation
Speech recognition
Question answering
And more…
Problem
DNNs are challenging to serve and deploy
in large-scale online services
Heavily constrained by latency, cost, and power
Size and complexity outpacing growth of commodity CPUs
Convolutional Neural Networks
ht-1 ht ht+1
xt-1 xt xt+1
ht-1 ht ht+1
yt-1 yt yt+1
Recurrent Neural Networks
Azure ML Hardware Accelerated Models, powered by Brainwave
• Create and deploy trained AI models as webservices
• Millisecond-scale round-trip inference with accelerated ResNet50
ResNet 50 on FPGA
(8 billion calculations per image)
Record speed:
<1.8 ms per image
Microsoft Build // Jabil Case Study:
Project BrainwaveMay 7-9, 2018 // Seattle, WA
FPGAs for ultra-fast inferencing
DATA BUILD TRAIN DEPLOY
Stored on
Azure Premium Storage
Azure Machine
Learning
Jabil Classification Model
Transfer learning
ResNet-50
Circuit board images
Jupyter Notebook
Azure Machine Learning
Ultra-fast Inferencing
using FPGAs
ResNet-50
Project Brainwave
Resources
WATCH: BRK3202 Hyperscale hardware: ML at scale on top of Azure + FPGA
https://mybuild.microsoft.com/sessions
Background on Field Gate Programmable Arrays (FPGA):
https://www.microsoft.com/en-us/research/project/project-catapult/
Documentation: deploying a model with FPGA:
https://cda.ms/tS
Github samples, and link to request quota:
http://aka.ms/aml-real-time-ai
Top Azure Takeaways from Microsoft BUILD
AI Model Deployment
2
ONNX
Open Neural Network Exchange Format
What is ONNX?
• Open standard to exchange trained deep neural network models
• Jointly developed by Microsoft and Facebook
• Natively supported as an export by many frameworks: Caffe2, Microsoft
Cognitive Toolkit, MXNet, and PyTorch
• Converters for CoreML and Tensorflow
• Supported by Windows ML: GPU-accelerated inference on Windows 10
• Supports a variety of DNN architectures (and other predictive models, too)
The flexibility to run your trained AI model anywhere:
on devices, on the cloud, or on the edge.
onnx.ai
Azure AI
Gallery:
models
gallery.azure.ai/models
Customvision.ai
“Not Hotdog”
customvision.ai
AI Model Deployment
Resources
WATCH: THR3106 Train your Computer Vision model in the cloud and export it to run
anywhere
https://mybuild.microsoft.com/sessions
Get Started with Custom Vision:
https://cda.ms/tW
Background on ONNX:
https://onnx.ai/
Overview of Windows ML:
https://cda.ms/tV
Top Azure Takeaways from Microsoft BUILD
Azure ML Packages
3
Forecasting
Text Analytics
Computer Vision
Azure ML Packages for Python
Python packages for Computer Vision, Forecasting and Text Analytics
Easily build and deploy models on Azure Machine Learning
Provides high level APIs for data preparation, augmentation, training,
evaluating and deployment.
Model experimentation and comparison, run history, model
management and deployment through Azure ML
Computer Vision Simplified with Azure ML Packages
After  Code example – Azure ML Package
for Computer Vision
Transfer learning example using the AML
Package for Computer Vision.
Before  Code example – Keras
Training a small CNN, screenshots taken from the file
“cifar10_cnn.py” in the Keras github repository.
Proven & Extensible
Open Source
Developer Focused
Join at github.com/dotnet/machinelearning
ML.NET
Cross-platform Open Source Machine learning framework for .NET
Extensively used across Microsoft: Windows, Bing, Azure
High productivity throughout the entire ML workflow
Extensible to other frameworks (TensorFlow, CNTK…)
ML.NET usage at Microsoft
+ more!
Windows 10
Power Point
Excel
Bing
Azure ML Developer Libraries
Resources
WATCH: BRK3226 What's new with Azure Machine Learning
WATCH: BRK3203 Introducing ML.NET
https://mybuild.microsoft.com/sessions
Azure ML packages for Python
https://cda.ms/tY
ML.NET
http://dot.net/ml
Top Azure Takeaways from Microsoft BUILD
Azure AI Lab
4
Azure AI Lab
Complete implementations of AI applications
http://ailab.microsoft.com
Inspired by real-life problems
Implemented using Microsoft Cognitive Services: Vision, Speech, Language and Search
Try out applications online
Source code provided in Github
Easy to deploy your own instance to Azure
Azure
AI
Lab
JFK Files
Search the JFK files using Azure Cognitive Search
Data extraction, natural language processing (NLP), and image processing.
https://github.com/Microsoft/AzureSearch_JFK_Files
Azure
AI
Lab
Drones: Search and Rescue training with AirSim
Goal: train AI-powered drones to recognize targets for search-and-rescue
Train recognizer using real objects in controlled environment with Custom Vision
Simulate training data for drone control with AirSim (simulator based on Unreal)
https://github.com/Microsoft/AirSim
Azure
AI
Lab
Drawing Bot
Generative Adversarial Network to generate an image based on your description of a bird
Implementation of AttGAN in PyTorch, running in AKS (Azure Container Service)
drawingbot.azurewebsites.net
github.com/taoxugit/AttnGAN
Azure AI Lab
Resources
WATCH: BRK3218 Microsoft AI overview for developers
https://mybuild.microsoft.com/sessions
Azure AI Lab Projects:
ailab.microsoft.com
Microsoft Cognitive Services
https://cda.ms/tX
Top Azure Takeaways from Microsoft BUILD
AI for Accessibility
5
//
A new $25 million, 5-year program
aimed at harnessing the power of AI
to amplify human capability for the
more than one billion people around
the world with disabilities. Together,
we can create a more inclusive world.
AI for Accessibility
Microsoft translator and custom speech
Custom Speech Model Microsoft Translator
Vision AI Dev Kit
http://www.visionaidevkit.com
AI for Accessibility
Resources
AI for Accessibility
https://cda.ms/v2
WATCH: BRK2601 You Can’t Start a Fire Without a Spark: AI for Accessibility
WATCH: BRK3222 Computer Vision Made Easy: Consume or Build Your Own State-of-the-Art
Models And Use Them Everywhere
https://mybuild.microsoft.com/sessions
Microsoft Translator
https://www.microsoft.com/translator
Getting Started with Speech Service API
https://cda.ms/tZ
Microsoft Custom Speech Service
https://cda.ms/v0
Vision AI Dev Kit early access preview
http://www.visionaidevkit.com
Top Azure Takeaways from Microsoft BUILD
And more…
And so much more…
Azure Conversational AI
Develop together with Visual Studio and VS Code live sharing
Notepad now supports Linux line endings
Custom Excel functions
Azure Kubernetes Service general availability
…
Resources
Explore more BUILD presentations:
https://developer.microsoft.com/en-us/events/build/content
Azure Cognitive Services: Overview and Tutorials
https://cda.ms/v1
Free Azure subscription, with $200 free credits:
https://cda.ms/tT
Q&A
Tim Heuer @timheuer
Cloud Developer Advocate
David Smith @revodavid
Cloud Developer Advocate
Our
speakers
Thank you for joining us.
The power of Deep Learning on FPGA
Performance Flexibility Scale
Rapidly adapt to evolving ML
Inference-optimized numerical precision
Exploit sparsity, deep compression
Excellent inference at low batch sizes
Ultra-low latency | 10x < CPU/GPU
World’s largest cloud investment in FPGAs
Multiple Exa-Ops of aggregate AI capacity
Runs on Microsoft’s scale infrastructure
Low cost
$0.21/million images on Azure FPGA
Azure Cognitive Services
Powerful prebuilt AI models exposed as API services
Simple REST APIs with .NET, Java, Python, Node SDKs
Easily customize for highest accuracy
Train in the cloud and deploy anywhere
Vision
Speech
Language
Conversation
Bing Search
Knowledge
Docker container Export
You own
Prediction Endpoint
(Docker container)
Local or on-prem
Azure App Service

Weitere ähnliche Inhalte

Mehr von Revolution Analytics

The Value of Open Source Communities
The Value of Open Source CommunitiesThe Value of Open Source Communities
The Value of Open Source CommunitiesRevolution Analytics
 
Building a scalable data science platform with R
Building a scalable data science platform with RBuilding a scalable data science platform with R
Building a scalable data science platform with RRevolution Analytics
 
The Business Economics and Opportunity of Open Source Data Science
The Business Economics and Opportunity of Open Source Data ScienceThe Business Economics and Opportunity of Open Source Data Science
The Business Economics and Opportunity of Open Source Data ScienceRevolution Analytics
 
Taking R Analytics to SQL and the Cloud
Taking R Analytics to SQL and the CloudTaking R Analytics to SQL and the Cloud
Taking R Analytics to SQL and the CloudRevolution Analytics
 
The Network structure of R packages on CRAN & BioConductor
The Network structure of R packages on CRAN & BioConductorThe Network structure of R packages on CRAN & BioConductor
The Network structure of R packages on CRAN & BioConductorRevolution Analytics
 
The network structure of cran 2015 07-02 final
The network structure of cran 2015 07-02 finalThe network structure of cran 2015 07-02 final
The network structure of cran 2015 07-02 finalRevolution Analytics
 
Simple Reproducibility with the checkpoint package
Simple Reproducibilitywith the checkpoint packageSimple Reproducibilitywith the checkpoint package
Simple Reproducibility with the checkpoint packageRevolution Analytics
 
Revolution R Enterprise 7.4 - Presentation by Bill Jacobs 11Jun15
Revolution R Enterprise 7.4 - Presentation by Bill Jacobs 11Jun15Revolution R Enterprise 7.4 - Presentation by Bill Jacobs 11Jun15
Revolution R Enterprise 7.4 - Presentation by Bill Jacobs 11Jun15Revolution Analytics
 
Warranty Predictive Analytics solution
Warranty Predictive Analytics solutionWarranty Predictive Analytics solution
Warranty Predictive Analytics solutionRevolution Analytics
 
Reproducibility with Checkpoint & RRO - NYC R Conference
Reproducibility with Checkpoint & RRO - NYC R ConferenceReproducibility with Checkpoint & RRO - NYC R Conference
Reproducibility with Checkpoint & RRO - NYC R ConferenceRevolution Analytics
 
Reproducibility with Revolution R Open and the Checkpoint Package
Reproducibility with Revolution R Open and the Checkpoint PackageReproducibility with Revolution R Open and the Checkpoint Package
Reproducibility with Revolution R Open and the Checkpoint PackageRevolution Analytics
 
Performance and Scale Options for R with Hadoop: A comparison of potential ar...
Performance and Scale Options for R with Hadoop: A comparison of potential ar...Performance and Scale Options for R with Hadoop: A comparison of potential ar...
Performance and Scale Options for R with Hadoop: A comparison of potential ar...Revolution Analytics
 
Reproducibility with Revolution R Open
Reproducibility with Revolution R OpenReproducibility with Revolution R Open
Reproducibility with Revolution R OpenRevolution Analytics
 
In-Database Analytics Deep Dive with Teradata and Revolution
In-Database Analytics Deep Dive with Teradata and RevolutionIn-Database Analytics Deep Dive with Teradata and Revolution
In-Database Analytics Deep Dive with Teradata and RevolutionRevolution Analytics
 
Batter Up! Advanced Sports Analytics with R and Storm
Batter Up! Advanced Sports Analytics with R and StormBatter Up! Advanced Sports Analytics with R and Storm
Batter Up! Advanced Sports Analytics with R and StormRevolution Analytics
 

Mehr von Revolution Analytics (20)

Reproducible Data Science with R
Reproducible Data Science with RReproducible Data Science with R
Reproducible Data Science with R
 
The Value of Open Source Communities
The Value of Open Source CommunitiesThe Value of Open Source Communities
The Value of Open Source Communities
 
The R Ecosystem
The R EcosystemThe R Ecosystem
The R Ecosystem
 
R at Microsoft (useR! 2016)
R at Microsoft (useR! 2016)R at Microsoft (useR! 2016)
R at Microsoft (useR! 2016)
 
Building a scalable data science platform with R
Building a scalable data science platform with RBuilding a scalable data science platform with R
Building a scalable data science platform with R
 
R at Microsoft
R at MicrosoftR at Microsoft
R at Microsoft
 
The Business Economics and Opportunity of Open Source Data Science
The Business Economics and Opportunity of Open Source Data ScienceThe Business Economics and Opportunity of Open Source Data Science
The Business Economics and Opportunity of Open Source Data Science
 
Taking R Analytics to SQL and the Cloud
Taking R Analytics to SQL and the CloudTaking R Analytics to SQL and the Cloud
Taking R Analytics to SQL and the Cloud
 
The Network structure of R packages on CRAN & BioConductor
The Network structure of R packages on CRAN & BioConductorThe Network structure of R packages on CRAN & BioConductor
The Network structure of R packages on CRAN & BioConductor
 
The network structure of cran 2015 07-02 final
The network structure of cran 2015 07-02 finalThe network structure of cran 2015 07-02 final
The network structure of cran 2015 07-02 final
 
Simple Reproducibility with the checkpoint package
Simple Reproducibilitywith the checkpoint packageSimple Reproducibilitywith the checkpoint package
Simple Reproducibility with the checkpoint package
 
R at Microsoft
R at MicrosoftR at Microsoft
R at Microsoft
 
Revolution R Enterprise 7.4 - Presentation by Bill Jacobs 11Jun15
Revolution R Enterprise 7.4 - Presentation by Bill Jacobs 11Jun15Revolution R Enterprise 7.4 - Presentation by Bill Jacobs 11Jun15
Revolution R Enterprise 7.4 - Presentation by Bill Jacobs 11Jun15
 
Warranty Predictive Analytics solution
Warranty Predictive Analytics solutionWarranty Predictive Analytics solution
Warranty Predictive Analytics solution
 
Reproducibility with Checkpoint & RRO - NYC R Conference
Reproducibility with Checkpoint & RRO - NYC R ConferenceReproducibility with Checkpoint & RRO - NYC R Conference
Reproducibility with Checkpoint & RRO - NYC R Conference
 
Reproducibility with Revolution R Open and the Checkpoint Package
Reproducibility with Revolution R Open and the Checkpoint PackageReproducibility with Revolution R Open and the Checkpoint Package
Reproducibility with Revolution R Open and the Checkpoint Package
 
Performance and Scale Options for R with Hadoop: A comparison of potential ar...
Performance and Scale Options for R with Hadoop: A comparison of potential ar...Performance and Scale Options for R with Hadoop: A comparison of potential ar...
Performance and Scale Options for R with Hadoop: A comparison of potential ar...
 
Reproducibility with Revolution R Open
Reproducibility with Revolution R OpenReproducibility with Revolution R Open
Reproducibility with Revolution R Open
 
In-Database Analytics Deep Dive with Teradata and Revolution
In-Database Analytics Deep Dive with Teradata and RevolutionIn-Database Analytics Deep Dive with Teradata and Revolution
In-Database Analytics Deep Dive with Teradata and Revolution
 
Batter Up! Advanced Sports Analytics with R and Storm
Batter Up! Advanced Sports Analytics with R and StormBatter Up! Advanced Sports Analytics with R and Storm
Batter Up! Advanced Sports Analytics with R and Storm
 

Kürzlich hochgeladen

Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...DianaGray10
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodJuan lago vázquez
 
Top 10 Most Downloaded Games on Play Store in 2024
Top 10 Most Downloaded Games on Play Store in 2024Top 10 Most Downloaded Games on Play Store in 2024
Top 10 Most Downloaded Games on Play Store in 2024SynarionITSolutions
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAndrey Devyatkin
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Miguel Araújo
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Drew Madelung
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfsudhanshuwaghmare1
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProduct Anonymous
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUK Journal
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesBoston Institute of Analytics
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MIND CTI
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 

Kürzlich hochgeladen (20)

Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
Top 10 Most Downloaded Games on Play Store in 2024
Top 10 Most Downloaded Games on Play Store in 2024Top 10 Most Downloaded Games on Play Store in 2024
Top 10 Most Downloaded Games on Play Store in 2024
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation Strategies
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 

Top Azure Takeaways from Microsoft Build

  • 1. Top Azure Takeaways from Microsoft Build Azure webinar series
  • 2. Welcome How do I ask a question? If you have a technical or content-related question, please use the Q&A window We will address the questions as they come in Can I view this presentation after the webinar? Yes, this presentation is being recorded A link to the recorded presentation will be sent to the email address you used to register
  • 3. Tim Heuer @timheuer Cloud Developer Advocate David Smith @revodavid Cloud Developer Advocate Our speakers
  • 4. What did you miss at BUILD? Project Brainwave • FPGA: Field-programmable gate arrays • Hardware accelerated inference • Circuit board quality detection Model deployment • Customvision.ai export • ONNX • Containers • Azure AI Gallery Azure ML packages • Forecasting • Computer Vision • Text Analytics • ML.net 1
  • 5. What did you miss at BUILD? AI Lab • Drawing Bot • JFK Files • AirSim • Drone Search and Rescue AI for Accessibility • Seeing AI • Translator • Custom Voice • Custom Vision • Vison AI Dev Kit And more… • Other announcements you may have missed. 2
  • 6. Top Azure Takeaways from Microsoft BUILD Project Brainwave 1
  • 7. Breakthroughs in deep learning demand real-time AI Deep neural networks have enabled major advances in machine learning and AI Computer vision Language translation Speech recognition Question answering And more… Problem DNNs are challenging to serve and deploy in large-scale online services Heavily constrained by latency, cost, and power Size and complexity outpacing growth of commodity CPUs Convolutional Neural Networks ht-1 ht ht+1 xt-1 xt xt+1 ht-1 ht ht+1 yt-1 yt yt+1 Recurrent Neural Networks
  • 8. Azure ML Hardware Accelerated Models, powered by Brainwave • Create and deploy trained AI models as webservices • Millisecond-scale round-trip inference with accelerated ResNet50 ResNet 50 on FPGA (8 billion calculations per image) Record speed: <1.8 ms per image
  • 9. Microsoft Build // Jabil Case Study: Project BrainwaveMay 7-9, 2018 // Seattle, WA
  • 10.
  • 11. FPGAs for ultra-fast inferencing DATA BUILD TRAIN DEPLOY Stored on Azure Premium Storage Azure Machine Learning Jabil Classification Model Transfer learning ResNet-50 Circuit board images Jupyter Notebook Azure Machine Learning Ultra-fast Inferencing using FPGAs ResNet-50
  • 12.
  • 13. Project Brainwave Resources WATCH: BRK3202 Hyperscale hardware: ML at scale on top of Azure + FPGA https://mybuild.microsoft.com/sessions Background on Field Gate Programmable Arrays (FPGA): https://www.microsoft.com/en-us/research/project/project-catapult/ Documentation: deploying a model with FPGA: https://cda.ms/tS Github samples, and link to request quota: http://aka.ms/aml-real-time-ai
  • 14. Top Azure Takeaways from Microsoft BUILD AI Model Deployment 2
  • 15. ONNX Open Neural Network Exchange Format What is ONNX? • Open standard to exchange trained deep neural network models • Jointly developed by Microsoft and Facebook • Natively supported as an export by many frameworks: Caffe2, Microsoft Cognitive Toolkit, MXNet, and PyTorch • Converters for CoreML and Tensorflow • Supported by Windows ML: GPU-accelerated inference on Windows 10 • Supports a variety of DNN architectures (and other predictive models, too) The flexibility to run your trained AI model anywhere: on devices, on the cloud, or on the edge. onnx.ai
  • 19. AI Model Deployment Resources WATCH: THR3106 Train your Computer Vision model in the cloud and export it to run anywhere https://mybuild.microsoft.com/sessions Get Started with Custom Vision: https://cda.ms/tW Background on ONNX: https://onnx.ai/ Overview of Windows ML: https://cda.ms/tV
  • 20. Top Azure Takeaways from Microsoft BUILD Azure ML Packages 3
  • 21. Forecasting Text Analytics Computer Vision Azure ML Packages for Python Python packages for Computer Vision, Forecasting and Text Analytics Easily build and deploy models on Azure Machine Learning Provides high level APIs for data preparation, augmentation, training, evaluating and deployment. Model experimentation and comparison, run history, model management and deployment through Azure ML
  • 22. Computer Vision Simplified with Azure ML Packages After  Code example – Azure ML Package for Computer Vision Transfer learning example using the AML Package for Computer Vision. Before  Code example – Keras Training a small CNN, screenshots taken from the file “cifar10_cnn.py” in the Keras github repository.
  • 23. Proven & Extensible Open Source Developer Focused Join at github.com/dotnet/machinelearning ML.NET Cross-platform Open Source Machine learning framework for .NET Extensively used across Microsoft: Windows, Bing, Azure High productivity throughout the entire ML workflow Extensible to other frameworks (TensorFlow, CNTK…)
  • 24. ML.NET usage at Microsoft + more! Windows 10 Power Point Excel Bing
  • 25. Azure ML Developer Libraries Resources WATCH: BRK3226 What's new with Azure Machine Learning WATCH: BRK3203 Introducing ML.NET https://mybuild.microsoft.com/sessions Azure ML packages for Python https://cda.ms/tY ML.NET http://dot.net/ml
  • 26. Top Azure Takeaways from Microsoft BUILD Azure AI Lab 4
  • 27. Azure AI Lab Complete implementations of AI applications http://ailab.microsoft.com Inspired by real-life problems Implemented using Microsoft Cognitive Services: Vision, Speech, Language and Search Try out applications online Source code provided in Github Easy to deploy your own instance to Azure
  • 28. Azure AI Lab JFK Files Search the JFK files using Azure Cognitive Search Data extraction, natural language processing (NLP), and image processing. https://github.com/Microsoft/AzureSearch_JFK_Files
  • 29. Azure AI Lab Drones: Search and Rescue training with AirSim Goal: train AI-powered drones to recognize targets for search-and-rescue Train recognizer using real objects in controlled environment with Custom Vision Simulate training data for drone control with AirSim (simulator based on Unreal) https://github.com/Microsoft/AirSim
  • 30.
  • 31. Azure AI Lab Drawing Bot Generative Adversarial Network to generate an image based on your description of a bird Implementation of AttGAN in PyTorch, running in AKS (Azure Container Service) drawingbot.azurewebsites.net
  • 33. Azure AI Lab Resources WATCH: BRK3218 Microsoft AI overview for developers https://mybuild.microsoft.com/sessions Azure AI Lab Projects: ailab.microsoft.com Microsoft Cognitive Services https://cda.ms/tX
  • 34. Top Azure Takeaways from Microsoft BUILD AI for Accessibility 5
  • 35. // A new $25 million, 5-year program aimed at harnessing the power of AI to amplify human capability for the more than one billion people around the world with disabilities. Together, we can create a more inclusive world. AI for Accessibility
  • 36.
  • 37. Microsoft translator and custom speech Custom Speech Model Microsoft Translator
  • 38.
  • 39. Vision AI Dev Kit http://www.visionaidevkit.com
  • 40. AI for Accessibility Resources AI for Accessibility https://cda.ms/v2 WATCH: BRK2601 You Can’t Start a Fire Without a Spark: AI for Accessibility WATCH: BRK3222 Computer Vision Made Easy: Consume or Build Your Own State-of-the-Art Models And Use Them Everywhere https://mybuild.microsoft.com/sessions Microsoft Translator https://www.microsoft.com/translator Getting Started with Speech Service API https://cda.ms/tZ Microsoft Custom Speech Service https://cda.ms/v0 Vision AI Dev Kit early access preview http://www.visionaidevkit.com
  • 41. Top Azure Takeaways from Microsoft BUILD And more…
  • 42. And so much more… Azure Conversational AI Develop together with Visual Studio and VS Code live sharing Notepad now supports Linux line endings Custom Excel functions Azure Kubernetes Service general availability …
  • 43. Resources Explore more BUILD presentations: https://developer.microsoft.com/en-us/events/build/content Azure Cognitive Services: Overview and Tutorials https://cda.ms/v1 Free Azure subscription, with $200 free credits: https://cda.ms/tT
  • 44. Q&A
  • 45. Tim Heuer @timheuer Cloud Developer Advocate David Smith @revodavid Cloud Developer Advocate Our speakers
  • 46. Thank you for joining us.
  • 47. The power of Deep Learning on FPGA Performance Flexibility Scale Rapidly adapt to evolving ML Inference-optimized numerical precision Exploit sparsity, deep compression Excellent inference at low batch sizes Ultra-low latency | 10x < CPU/GPU World’s largest cloud investment in FPGAs Multiple Exa-Ops of aggregate AI capacity Runs on Microsoft’s scale infrastructure Low cost $0.21/million images on Azure FPGA
  • 48. Azure Cognitive Services Powerful prebuilt AI models exposed as API services Simple REST APIs with .NET, Java, Python, Node SDKs Easily customize for highest accuracy Train in the cloud and deploy anywhere Vision Speech Language Conversation Bing Search Knowledge
  • 49. Docker container Export You own Prediction Endpoint (Docker container) Local or on-prem Azure App Service

Hinweis der Redaktion

  1. Replace cover image: go to Format Background > Picture or texture file Adjust gradient: go to View > Slide Master > Adjust shape fill options
  2. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  3. Speaker image: Insert/Paste image > Picture Format > Crop > Aspect Ratio 1:1 > Crop > Crop to shape (circle) > Resize to template *** Please retrieve speaker images from: https://developer.microsoft.com/en-us/advocates/
  4. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  5. FGPA Image Source: https://www.microsoft.com/en-us/research/project/project-catapult/ Real-time AI at cloud scale with industry-leading performance and lowest cost Models are easy to create and deploy into Azure Write once, deploy anywhere – to intelligent cloud or edge Record-setting DNN performance with accelerated ResNet50 Record speed: Object classification on FPGA in <1.8 ms per image Lowest cost: Only 21 cents per million images during preview More accelerated models coming soon
  6. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  7. Direct questions to Wee Hyong Tok or Lucas Joppa
  8. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  9. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  10. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  11. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  12. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  13. 21
  14. 23
  15. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  16. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  17. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  18. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  19. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  20. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  21. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  22. A new, $25M five-year program to put AI tools in the hands of developers to build solutions for people with disabilities. Seed grants of technology to developers, universities, NGOs, and inventors. Microsoft Cognitive Services APIs provide pre-built vision, speech and text capabilities. Extend these capabilities with AI frameworks.
  23. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  24. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  25. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  26. Replace background image: go to Format Background > Picture or texture file Adjust overlay: go to View > Slide Master > Adjust shape fill options
  27. Speaker image: Insert/Paste image > Picture Format > Crop > Aspect Ratio 1:1 > Crop > Crop to shape (circle) > Resize to template *** Please retrieve speaker images from: https://developer.microsoft.com/en-us/advocates/