SlideShare a Scribd company logo
1 of 23
@elbruno
How a PoC at home can scale
to Enterprise Level using
Custom Vision APIs
Bruno Capuano
Innovation Lead @Avanade
@elbruno | http://elbruno.com
@elbruno
why should I care about AI and ML?
As a developer,
@elbruno
Some problems are difficult to solve using traditional algorithms and
procedural programming.
@elbruno
IBM slaps patent on coffee-delivering drones that can read
your MIND (link)
@elbruno
IBM slaps patent on coffee-delivering drones that can read
your MIND (link)
@elbruno
IBM coffee-delivering drones test footage
@elbruno
Microsoft AI platform
Azure AI Services
Azure Infrastructure
Tools
@elbruno
@elbruno
Play-offs de ascenso de LaLiga 1|2|3 2016-2017
Getafe 3 – 1 Tenerife
(24 de junio de 2017)
{
"tags":[
"person",
"grass",
"soccer",
"outdoor",
"field",
"playing",
"game",
"ball",
"player",
"sport",
"man",
"young",
"group",
"rugby",
"people",
"watching",
"match",
"crowd",
"girl",
"team",
"uniform",
"standing",
"holding",
"blue"
],
"captions":[
{
"text":"a group of
people watching a
football ball",
"confidence":0.9469714
}
]
}
[
{
"name":"person",
"confidence":0.9998956
},
{
"name":"grass",
"confidence":0.9998543
},
{
"name":"soccer",
"confidence":0.992252052
},
{
"name":"outdoor",
"confidence":0.982327938
},
{
"name":"field",
"confidence":0.958654761
},
{
"name":"athletic game",
"confidence":0.938705862
},
{
"name":"player",
"confidence":0.9206911
},
{
"name":"rugby",
"confidence":0.459357172
}
]
[
{
"age":37,
"gender":"Male",
"faceRectangle":{
"top":51,
"left":498,
"width":54,
"height":54
}
}
]
Computer Vision API
@elbruno
Custom models
Easier / Less Control Harder / Full Control
Pre-built models
TensorFlow
ML.NETVisionSpeech LanguageKnowledge Search
Artificial Intelligence Models
@elbruno
Prepare Your Data Build & Train Run
Build your own custom machine learning models
@elbruno
@elbruno
Hello Custom Vision!
MakeMagicHappen();
https://www.avanade.com/AI
@elbruno
Artificial Intelligence: Image Analysis
ONNX and WinML
@elbruno
Windows ML uses ONNX models
@elbruno
1. Developers can focus on their data and
their scenarios, using Windows ML for
model evaluation
2. Enables using ML models trained with a
diverse set of toolkits
3. Hardware acceleration gets fast evaluation
results across the diversity of the entire
Windows device ecosystem.
Windows ML solves three problems for you
Direct3D
GPU
CPU
DirectML
Model Inference Engine
WinML Win32 API
WinML UWP API
Win32 App
WinML Runtime
UWP App
@elbruno
Azure Machine Learning Services
gives you an end-to-end
solution to prepare data and
train your model in the Cloud.
WinMLTools converts existing
models from CoreML, scikit-
learn, LIBSVM, and XGBoost
Azure Custom Vision makes it
easy to create your own image
models - https://customvision.ai/
Azure AI Gallery curates models
for use with Windows ML -
https://gallery.azure.ai/models
How do I get ONNX models to use in my
application?
@elbruno
Custom Vision in Windows
10 using WinML and ONNX
MakeMagicHappen();
https://www.avanade.com/AI
@elbruno
Train Stations
Parking Spots
available
Waiting Lines
Objects in a counter
And more ….
Use Cases
#NoTrain #Train
#NoTrain #NoTrain
#NoTrain #NoTrain
#Train #NoTrain
#Train
#Train
@elbruno
Custom Vision Container in
Raspberry Pi 3
MakeMagicHappen();
https://www.avanade.com/AI
@elbruno
Don’t be scare of AI
Being a dev/programmer
give you 80% of your
required skills
Containers are hard, but
necessary
• You don’t need to be a
Data Scientist to create
amazing solutions
• The other 20% is fun (or
pain)
• Tears will come, that’s for
sure
Lessons learned
@elbruno
Bruno Capuano
Innovation Lead @Avanade
@elbruno | http://elbruno.com
Q&A
Thanks!

More Related Content

What's hot

What's hot (12)

How Daimler uses mobile mixed realities for training and sales - Unite Copenh...
How Daimler uses mobile mixed realities for training and sales - Unite Copenh...How Daimler uses mobile mixed realities for training and sales - Unite Copenh...
How Daimler uses mobile mixed realities for training and sales - Unite Copenh...
 
Virtual Reality in AEC
Virtual Reality in AECVirtual Reality in AEC
Virtual Reality in AEC
 
How Volvo embraced real-time 3D and shook up the auto industry- Unite Copenha...
How Volvo embraced real-time 3D and shook up the auto industry- Unite Copenha...How Volvo embraced real-time 3D and shook up the auto industry- Unite Copenha...
How Volvo embraced real-time 3D and shook up the auto industry- Unite Copenha...
 
Construction Industry Augmented Reality
Construction Industry Augmented RealityConstruction Industry Augmented Reality
Construction Industry Augmented Reality
 
Uninext solutions Portfolio
Uninext solutions PortfolioUninext solutions Portfolio
Uninext solutions Portfolio
 
Resume
ResumeResume
Resume
 
Designer classifications
Designer classificationsDesigner classifications
Designer classifications
 
Power BI Day : Monitor your IoT sensors using PowerBI
Power BI Day : Monitor your IoT sensors using PowerBIPower BI Day : Monitor your IoT sensors using PowerBI
Power BI Day : Monitor your IoT sensors using PowerBI
 
Turn Revit Models into real-time 3D experiences
Turn Revit Models into real-time 3D experiencesTurn Revit Models into real-time 3D experiences
Turn Revit Models into real-time 3D experiences
 
Ai on the edge... and containers
Ai on the edge... and containersAi on the edge... and containers
Ai on the edge... and containers
 
Advanced Analytics with Power BI 201808
Advanced Analytics with Power BI 201808Advanced Analytics with Power BI 201808
Advanced Analytics with Power BI 201808
 
Summarize IHMC Concept Mapping software
Summarize IHMC Concept Mapping softwareSummarize IHMC Concept Mapping software
Summarize IHMC Concept Mapping software
 

Similar to Make UofT - Using Azure Custom Vision from PoC to Enterprise

2018 .NET Conf - 利用Machine Learning .NET整合機器學習至應用程式
2018 .NET Conf - 利用Machine Learning .NET整合機器學習至應用程式2018 .NET Conf - 利用Machine Learning .NET整合機器學習至應用程式
2018 .NET Conf - 利用Machine Learning .NET整合機器學習至應用程式
Alan Tsai
 
Minko - Why we created our own Flash platform and why you should care
Minko - Why we created our own Flash platform and why you should careMinko - Why we created our own Flash platform and why you should care
Minko - Why we created our own Flash platform and why you should care
Minko3D
 

Similar to Make UofT - Using Azure Custom Vision from PoC to Enterprise (20)

Synthetic Data: From 3D model to AI on the Edge
Synthetic Data: From 3D model to AI on the EdgeSynthetic Data: From 3D model to AI on the Edge
Synthetic Data: From 3D model to AI on the Edge
 
Azure IoT Central
Azure IoT CentralAzure IoT Central
Azure IoT Central
 
2018 09 26 CTT .NET User Group - Introduction to Machine Learning.Net and Win...
2018 09 26 CTT .NET User Group - Introduction to Machine Learning.Net and Win...2018 09 26 CTT .NET User Group - Introduction to Machine Learning.Net and Win...
2018 09 26 CTT .NET User Group - Introduction to Machine Learning.Net and Win...
 
2018 .NET Conf - 利用Machine Learning .NET整合機器學習至應用程式
2018 .NET Conf - 利用Machine Learning .NET整合機器學習至應用程式2018 .NET Conf - 利用Machine Learning .NET整合機器學習至應用程式
2018 .NET Conf - 利用Machine Learning .NET整合機器學習至應用程式
 
Enhance and Accelerate Your AI and Machine Learning Solution | SIGGRAPH 2019 ...
Enhance and Accelerate Your AI and Machine Learning Solution | SIGGRAPH 2019 ...Enhance and Accelerate Your AI and Machine Learning Solution | SIGGRAPH 2019 ...
Enhance and Accelerate Your AI and Machine Learning Solution | SIGGRAPH 2019 ...
 
Machine Learning at the Edge
Machine Learning at the EdgeMachine Learning at the Edge
Machine Learning at the Edge
 
2018 11 14 Artificial Intelligence and Machine Learning in Azure
2018 11 14 Artificial Intelligence and Machine Learning in Azure2018 11 14 Artificial Intelligence and Machine Learning in Azure
2018 11 14 Artificial Intelligence and Machine Learning in Azure
 
Leverage the power of machine learning on windows
Leverage the power of machine learning on windowsLeverage the power of machine learning on windows
Leverage the power of machine learning on windows
 
2018 10-17 J1 1D - Unleash Office 365 with the power of cognitive services an...
2018 10-17 J1 1D - Unleash Office 365 with the power of cognitive services an...2018 10-17 J1 1D - Unleash Office 365 with the power of cognitive services an...
2018 10-17 J1 1D - Unleash Office 365 with the power of cognitive services an...
 
Unleash office 365 with the power of cognitive services and microsoft graph api
Unleash office 365 with the power of cognitive services and microsoft graph apiUnleash office 365 with the power of cognitive services and microsoft graph api
Unleash office 365 with the power of cognitive services and microsoft graph api
 
BUILD with Microsoft - Radu Stefan
 BUILD with Microsoft - Radu Stefan BUILD with Microsoft - Radu Stefan
BUILD with Microsoft - Radu Stefan
 
Azure Meetup: NovitĂ  CosmosDB modalitĂ  Serverless e Cognitive Services
Azure Meetup: NovitĂ  CosmosDB modalitĂ  Serverless e Cognitive ServicesAzure Meetup: NovitĂ  CosmosDB modalitĂ  Serverless e Cognitive Services
Azure Meetup: NovitĂ  CosmosDB modalitĂ  Serverless e Cognitive Services
 
Ehab_Essamuddin
Ehab_EssamuddinEhab_Essamuddin
Ehab_Essamuddin
 
Artificial Intelligence and Cognitive Computing
Artificial Intelligence and Cognitive ComputingArtificial Intelligence and Cognitive Computing
Artificial Intelligence and Cognitive Computing
 
Build 2019 Recap
Build 2019 RecapBuild 2019 Recap
Build 2019 Recap
 
AI at Microsoft for HEC
AI at Microsoft for HECAI at Microsoft for HEC
AI at Microsoft for HEC
 
Introduction to ML.NET
Introduction to ML.NETIntroduction to ML.NET
Introduction to ML.NET
 
Minko - Why we created our own Flash platform and why you should care
Minko - Why we created our own Flash platform and why you should careMinko - Why we created our own Flash platform and why you should care
Minko - Why we created our own Flash platform and why you should care
 
Real time analytics in Azure IoT
Real time analytics in Azure IoT Real time analytics in Azure IoT
Real time analytics in Azure IoT
 
Abdelrahman Al-Ogail Resume
Abdelrahman Al-Ogail ResumeAbdelrahman Al-Ogail Resume
Abdelrahman Al-Ogail Resume
 

More from Bruno Capuano

More from Bruno Capuano (20)

Let's code a drone to follow faces using Python 🐍
Let's code a drone to follow faces using Python 🐍Let's code a drone to follow faces using Python 🐍
Let's code a drone to follow faces using Python 🐍
 
Using Azure IoT to feed my squirrels ️
Using Azure IoT to feed my squirrels ️Using Azure IoT to feed my squirrels ️
Using Azure IoT to feed my squirrels ️
 
2021 12 01 Global XR Conference - My experiences adapting a Digital Twin WebG...
2021 12 01 Global XR Conference - My experiences adapting a Digital Twin WebG...2021 12 01 Global XR Conference - My experiences adapting a Digital Twin WebG...
2021 12 01 Global XR Conference - My experiences adapting a Digital Twin WebG...
 
2021 06 19 ms student ambassadors nigeria ml net 01 slide-share
2021 06 19 ms student ambassadors nigeria ml net 01   slide-share2021 06 19 ms student ambassadors nigeria ml net 01   slide-share
2021 06 19 ms student ambassadors nigeria ml net 01 slide-share
 
Global AI on Virtual Tour Oslo - Anomaly Detection using ML.Net on a drone te...
Global AI on Virtual Tour Oslo - Anomaly Detection using ML.Net on a drone te...Global AI on Virtual Tour Oslo - Anomaly Detection using ML.Net on a drone te...
Global AI on Virtual Tour Oslo - Anomaly Detection using ML.Net on a drone te...
 
2021 04 21 Azure Sydney User Group - Scaling a POC to an Enterprise using Azu...
2021 04 21 Azure Sydney User Group - Scaling a POC to an Enterprise using Azu...2021 04 21 Azure Sydney User Group - Scaling a POC to an Enterprise using Azu...
2021 04 21 Azure Sydney User Group - Scaling a POC to an Enterprise using Azu...
 
2021 02 23 MVP Fusion Getting Started with Machine Learning.Net and AutoML
2021 02 23 MVP Fusion Getting Started with Machine Learning.Net and AutoML2021 02 23 MVP Fusion Getting Started with Machine Learning.Net and AutoML
2021 02 23 MVP Fusion Getting Started with Machine Learning.Net and AutoML
 
2021 02 13 CodeGen Verona - Let’s code a drone to follow faces syncing everyt...
2021 02 13 CodeGen Verona - Let’s code a drone to follow faces syncing everyt...2021 02 13 CodeGen Verona - Let’s code a drone to follow faces syncing everyt...
2021 02 13 CodeGen Verona - Let’s code a drone to follow faces syncing everyt...
 
2020 11 19 MVP Days Israel 2020 - Introduction to Machine Learning.Net and Au...
2020 11 19 MVP Days Israel 2020 - Introduction to Machine Learning.Net and Au...2020 11 19 MVP Days Israel 2020 - Introduction to Machine Learning.Net and Au...
2020 11 19 MVP Days Israel 2020 - Introduction to Machine Learning.Net and Au...
 
2020 10 22 AI Fundamentals - Azure Machine Learning
2020 10 22 AI Fundamentals - Azure Machine Learning2020 10 22 AI Fundamentals - Azure Machine Learning
2020 10 22 AI Fundamentals - Azure Machine Learning
 
2020 09 24 - CONDG ML.Net
2020 09 24 - CONDG ML.Net2020 09 24 - CONDG ML.Net
2020 09 24 - CONDG ML.Net
 
2020 08 06 Global XR Talks - Lessons Learned creating a multiplatform AI proj...
2020 08 06 Global XR Talks - Lessons Learned creating a multiplatform AI proj...2020 08 06 Global XR Talks - Lessons Learned creating a multiplatform AI proj...
2020 08 06 Global XR Talks - Lessons Learned creating a multiplatform AI proj...
 
2020 06 27 Global AI On Tour Virtual GTA
2020 06 27 Global AI On Tour Virtual GTA2020 06 27 Global AI On Tour Virtual GTA
2020 06 27 Global AI On Tour Virtual GTA
 
2020 06 13 Best of Build 2020 - Canada Community Edition - Artificial Intelli...
2020 06 13 Best of Build 2020 - Canada Community Edition - Artificial Intelli...2020 06 13 Best of Build 2020 - Canada Community Edition - Artificial Intelli...
2020 06 13 Best of Build 2020 - Canada Community Edition - Artificial Intelli...
 
Global Azure AI Tour Buenos Aires Argentina, Drones and AI
Global Azure AI Tour Buenos Aires Argentina, Drones and AIGlobal Azure AI Tour Buenos Aires Argentina, Drones and AI
Global Azure AI Tour Buenos Aires Argentina, Drones and AI
 
2020 04 18 Global AI On Tour Monterrey - Program a Drone using AI
2020 04 18 Global AI On Tour Monterrey - Program a Drone using AI2020 04 18 Global AI On Tour Monterrey - Program a Drone using AI
2020 04 18 Global AI On Tour Monterrey - Program a Drone using AI
 
2020 04 10 Catch IT - Getting started with ML.Net
2020 04 10 Catch IT - Getting started with ML.Net2020 04 10 Catch IT - Getting started with ML.Net
2020 04 10 Catch IT - Getting started with ML.Net
 
2020 04 09 Global AI Community Virtual Tour - Drones and AI
2020 04 09 Global AI Community Virtual Tour - Drones and AI2020 04 09 Global AI Community Virtual Tour - Drones and AI
2020 04 09 Global AI Community Virtual Tour - Drones and AI
 
2020 04 04 NetCoreConf - Machine Learning.Net
2020 04 04 NetCoreConf - Machine Learning.Net2020 04 04 NetCoreConf - Machine Learning.Net
2020 04 04 NetCoreConf - Machine Learning.Net
 
2020 02 29 TechDay Conf - Getting started with Machine Learning.Net
2020 02 29 TechDay Conf - Getting started with Machine Learning.Net2020 02 29 TechDay Conf - Getting started with Machine Learning.Net
2020 02 29 TechDay Conf - Getting started with Machine Learning.Net
 

Recently uploaded

Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Victor Rentea
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
panagenda
 

Recently uploaded (20)

Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
 
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
 
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...
 
AI in Action: Real World Use Cases by Anitaraj
AI in Action: Real World Use Cases by AnitarajAI in Action: Real World Use Cases by Anitaraj
AI in Action: Real World Use Cases by Anitaraj
 
Less Is More: Utilizing Ballerina to Architect a Cloud Data Platform
Less Is More: Utilizing Ballerina to Architect a Cloud Data PlatformLess Is More: Utilizing Ballerina to Architect a Cloud Data Platform
Less Is More: Utilizing Ballerina to Architect a Cloud Data Platform
 
Choreo: Empowering the Future of Enterprise Software Engineering
Choreo: Empowering the Future of Enterprise Software EngineeringChoreo: Empowering the Future of Enterprise Software Engineering
Choreo: Empowering the Future of Enterprise Software Engineering
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
 
ChatGPT and Beyond - Elevating DevOps Productivity
ChatGPT and Beyond - Elevating DevOps ProductivityChatGPT and Beyond - Elevating DevOps Productivity
ChatGPT and Beyond - Elevating DevOps Productivity
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
 
AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)
AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)
AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)
 
JavaScript Usage Statistics 2024 - The Ultimate Guide
JavaScript Usage Statistics 2024 - The Ultimate GuideJavaScript Usage Statistics 2024 - The Ultimate Guide
JavaScript Usage Statistics 2024 - The Ultimate Guide
 
WSO2 Micro Integrator for Enterprise Integration in a Decentralized, Microser...
WSO2 Micro Integrator for Enterprise Integration in a Decentralized, Microser...WSO2 Micro Integrator for Enterprise Integration in a Decentralized, Microser...
WSO2 Micro Integrator for Enterprise Integration in a Decentralized, Microser...
 
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
 
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)
 
Vector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptxVector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptx
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
 
Modernizing Legacy Systems Using Ballerina
Modernizing Legacy Systems Using BallerinaModernizing Legacy Systems Using Ballerina
Modernizing Legacy Systems Using Ballerina
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
 
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
 

Make UofT - Using Azure Custom Vision from PoC to Enterprise

  • 1. @elbruno How a PoC at home can scale to Enterprise Level using Custom Vision APIs Bruno Capuano Innovation Lead @Avanade @elbruno | http://elbruno.com
  • 2. @elbruno why should I care about AI and ML? As a developer,
  • 3. @elbruno Some problems are difficult to solve using traditional algorithms and procedural programming.
  • 4. @elbruno IBM slaps patent on coffee-delivering drones that can read your MIND (link)
  • 5. @elbruno IBM slaps patent on coffee-delivering drones that can read your MIND (link)
  • 7. @elbruno Microsoft AI platform Azure AI Services Azure Infrastructure Tools
  • 9. @elbruno Play-offs de ascenso de LaLiga 1|2|3 2016-2017 Getafe 3 – 1 Tenerife (24 de junio de 2017) { "tags":[ "person", "grass", "soccer", "outdoor", "field", "playing", "game", "ball", "player", "sport", "man", "young", "group", "rugby", "people", "watching", "match", "crowd", "girl", "team", "uniform", "standing", "holding", "blue" ], "captions":[ { "text":"a group of people watching a football ball", "confidence":0.9469714 } ] } [ { "name":"person", "confidence":0.9998956 }, { "name":"grass", "confidence":0.9998543 }, { "name":"soccer", "confidence":0.992252052 }, { "name":"outdoor", "confidence":0.982327938 }, { "name":"field", "confidence":0.958654761 }, { "name":"athletic game", "confidence":0.938705862 }, { "name":"player", "confidence":0.9206911 }, { "name":"rugby", "confidence":0.459357172 } ] [ { "age":37, "gender":"Male", "faceRectangle":{ "top":51, "left":498, "width":54, "height":54 } } ] Computer Vision API
  • 10. @elbruno Custom models Easier / Less Control Harder / Full Control Pre-built models TensorFlow ML.NETVisionSpeech LanguageKnowledge Search Artificial Intelligence Models
  • 11. @elbruno Prepare Your Data Build & Train Run Build your own custom machine learning models
  • 17. @elbruno 1. Developers can focus on their data and their scenarios, using Windows ML for model evaluation 2. Enables using ML models trained with a diverse set of toolkits 3. Hardware acceleration gets fast evaluation results across the diversity of the entire Windows device ecosystem. Windows ML solves three problems for you Direct3D GPU CPU DirectML Model Inference Engine WinML Win32 API WinML UWP API Win32 App WinML Runtime UWP App
  • 18. @elbruno Azure Machine Learning Services gives you an end-to-end solution to prepare data and train your model in the Cloud. WinMLTools converts existing models from CoreML, scikit- learn, LIBSVM, and XGBoost Azure Custom Vision makes it easy to create your own image models - https://customvision.ai/ Azure AI Gallery curates models for use with Windows ML - https://gallery.azure.ai/models How do I get ONNX models to use in my application?
  • 19. @elbruno Custom Vision in Windows 10 using WinML and ONNX MakeMagicHappen(); https://www.avanade.com/AI
  • 20. @elbruno Train Stations Parking Spots available Waiting Lines Objects in a counter And more …. Use Cases #NoTrain #Train #NoTrain #NoTrain #NoTrain #NoTrain #Train #NoTrain #Train #Train
  • 21. @elbruno Custom Vision Container in Raspberry Pi 3 MakeMagicHappen(); https://www.avanade.com/AI
  • 22. @elbruno Don’t be scare of AI Being a dev/programmer give you 80% of your required skills Containers are hard, but necessary • You don’t need to be a Data Scientist to create amazing solutions • The other 20% is fun (or pain) • Tears will come, that’s for sure Lessons learned
  • 23. @elbruno Bruno Capuano Innovation Lead @Avanade @elbruno | http://elbruno.com Q&A Thanks!

Editor's Notes

  1. Everyone in the room is a Machine Learning developer! ML is everywhere and chances are you are already using a service that is possible through ML. We are going to show you how AI and ML can help you solve real problems in your own apps and services and how easy it is to get started with ML today in your own apps
  2. Any technology user today has benefitted from machine learning. Facial recognition technology allows social media platforms to help users tag and share photos of friends. Recommendation engines, powered by machine learning, suggest what movies or television shows to watch next based on user preferences. Manufacturing processes contain sensors that collect vast amounts of data and are able to predict failures and react in real time. Self-driving cars that rely on machine learning to navigate are already being tested and may soon be available to consumers. In our projects, we are often asked to solve very hard problems. These three problems have something in common – they are relatively easy for a human to solve, but are difficult to program a computer to do. This is especially true, when the only tool available for us is procedural code and traditional algorithms. This is where AI can help!
  3. https://www.theregister.co.uk/2018/08/23/ibm_coffee_drone_patent/ https://dronelife.com/2018/08/24/ibm-patents-coffee-delivering-drone/
  4. https://www.theregister.co.uk/2018/08/23/ibm_coffee_drone_patent/ https://dronelife.com/2018/08/24/ibm-patents-coffee-delivering-drone/
  5. https://www.youtube.com/watch?v=suFH2VEQ9Wo
  6. 10
  7. 11
  8. https://www.pexels.com/photo/dog-chihuahua-sobel-50718/
  9. NOTE TO PRESENTERS This deck contains a library of slides you can use to build a customized introduction to Avanade for external audiences, including clients, partners and recruits. The deck is not meant to be used in its entirety, but rather to provide options for how you introduce Avanade, allowing you to match your content to your audience and their interests. The slides at the front of the deck provide an overview of how Avanade sees the marketplace changing and how we’re helping clients respond to those changes. Your introduction can be extended with slides from the appendices, depending on your purposes, to provide additional information of specific relevance to your presentation audience.
  10. Industry standard format for Machine Learning model interchange A community project created by Microsoft and Facebook Defines an extensible computation graph, built-in operators, and standard data types. Supports a wide range of machine learning models including Classical and Deep Learning algorithms Learn more at http://onnx.ai/
  11. Let me explain how Windows ML solved 3 problems for me when building an application like the one I just showed you Windows ML provides an evaluation runtime, so you don’t have to worry about finding one and deploying it with your application. Just focus on identifying your question, gathering your data – and let Windows ML provide the platform. Likewise, Windows ML provides me with APIs that I can use in either my Win32 application, or my UWP application Windows ML supports ONNX, the industry standard for Machine Learning model interop. No matter what training framework you used to make your model, as long as you can save or convert it into ONNX – you can evaluate it with Windows ML. On Windows, your evaluation code is HW accelerated using DX12. DirectX as you know, has a proven history of providing excellent performance across a wide range of hardware. This means that you don’t have to change your code when running on different devices. This is a huge advantage compared to other platforms that depend on 3rd party device SDKs for evaluation acceleration. We’ve also optimized the CPU paths, using the instruction optimizations up to AVX-512.