Exam AI-900: Microsoft Azure AI Fundamentals Crash Course
Topic: Data
Artificial Intelligence (AI) is the capability of a computer to imitate human behavior. Microsoft Azure includes a suite of tools that enable AI engineers to ingest data, build machine learning (ML) models, and deploy those models as REST API web services. Azure also includes a number of platform-as-a-service cognitive services that give Azure developers easy access to pre-trained ML models.
Because AI/ML factor into most customer-facing software nowadays, coupled with the Azure AI/ML tools' steep learning curve, this training presents an excellent opportunity for IT professionals to learn the technology in a clear, low-stress manner.
Specifically, this Crash Course covers the objectives for Microsoft Exam AI-900: Microsoft Azure AI Fundamentals. This exam fulfills the requirements of the Microsoft Certified AI Fundamentals certification.
Note that Microsoft considers their 900-series exams to be beginner-to-intermediate level skill sets. The certification is intended for both technical and non-technical audiences, and covers the Azure AI skills from a mostly theoretical viewpoint.
What you'll learn-and how you can apply it
At course conclusion, you will be able to:
- Perform additional review and then pass Exam AI-900: Microsoft Azure AI Fundamentals
- Describe artificial intelligence workloads and considerations
- Describe fundamental machine learning concepts on Azure
- Describe computer vision workloads on Azure
- Describe conversational workloads on Azure
This training course is for you because...
- Azure AI certification candidates
- Future, present, or future data scientists or AI/ML engineers
- Azure professionals who will support Azure AI/ML projects
- Non-technical professionals who need to understand a bit of how Azure AI works
Prerequisites
The AI-900 is supposed to be a beginner/fundamentals skill set. Thus, the core prerequisite is that the learner has intermediate knowledge of Azure from any job role: administrator, developer, architect, business analyst, etc.
Ideally the learner is already a data scientist with Python, Scala, or R programming knowledge, but I'll teach to any level.
Course Set-up
To follow along with the demonstrations and practice on his or her own, the learner should have the following environment available:
- Windows or macOS computer
- Web browser and Internet connection
- Microsoft account to create an Azure trial subscription (free)
- Microsoft Azure 30-day trial (free)
If the learner wants to dive in and actually practice the skills, he or she should also have:
- A paid Azure subscription if they already used their trial
- Visual Studio 2019 (the free Community edition is fine)
- Visual Studio Code (free)
- Azure Command-Line Interface (CLI) (free)
- Docker Desktop software (free)
The course files will be available at Tim's GitHub repository:
https://github.com/timothywarner/ai900
Recommended Preparation
- Book: Pragmatic AI: An Introduction to Cloud-Based Machine Learning, First Edition, by Noah Gift https://www.oreilly.com/library/view/pragmatic-ai-an/9780134863924/
-
Book: Machine Learning with Python for Everyone, by Mark Fenner https://www.oreilly.com/library/view/machine-learning-with/9780134845708/
-
Book: Machine Learning in Production: Developing and Optimizing Data Science Workflows and Applications, First Edition, by Andrew Kelleher and Adam Kelleher https://www.oreilly.com/library/view/machine-learning-in/9780134116556/
Recommended Follow-up
- Book: Exam Ref AZ-900: Microsoft Azure Fundamentals, First Edition, by Jim Cheshire https://www.oreilly.com/library/view/exam-ref-az-900/9780135732199/
- Book: Microservices with Docker on Microsoft Azure (includes Content Update Program), by Boris Scholl, Trent Swanson, and Daniel Fernandez https://www.oreilly.com/library/view/microservices-with-docker/9780134218229/
- Video: Linux on Azure and LFCS Certification, by Sander van Vugt https://www.oreilly.com/learning-paths/learning-path-linux/9780135155417/
About your instructor
-
Tim Warner is a Microsoft MVP and cloud solutions architect based in Nashville, Tennessee. Tim has over 20 years of experience as an IT generalist and technical trainer. Reach Tim at Twitter (@TechTrainerTim) or his website, TechTrainerTim.com.
Schedule
The timeframes are only estimates and may vary according to how the class is progressing
DAY 1
Introduction (5 minutes)
- Set expectations
- Perform high-level course overview
- Provide strategies to maximize benefit from this course
Module 1: Identify Features of Common AI Workloads (40 minutes)
- Identify prediction/forecasting workloads
- Identify computer vision workloads
- Identity conversational and natural language processing (NLP) workloads
Module 2: Identify Guiding Principles for Responsible AI (40 minutes)
- Describe considerations for fairness and reliability in an AI solution
- Describe considerations for privacy and security in an AI solution
- Describe considerations for inclusiveness and transparency in an AI solution
Module 3: Identify Common Machine Learning Types (40 minutes)
- Identify regression ML scenarios
- Identify classification ML scenarios
- Identify clustering ML scenarios
(Break: 5 minutes)
Module 4: Describe Core Machine Learning Concepts (40 minutes)
- Identify features and labels in a dataset for ML
- Describe how training and validations datasets work
- Select and interpret model evaluation metrics
Module 5: Identify Core Tasks in Creating a Machine Learning Solution (40 minutes)
- Describe data ingestion and preparation features
- Describe feature selection and engineering features
- Describe model deployment and management features
Module 6: Describe No-Code Machine Learning in Azure Machine Learning (40 minutes)
- Describe the Automated Machine Learning tool
- Describe the Azure Machine Learning designer
Review, wrap-up: 10 minutes
DAY 2
Introduction (5 minutes)
- Briefly review Day 1 material
- Perform Day 2 content overview
- Answer any outstanding questions from Day 1
Module 7: Identify Common Types of Computer Vision Solutions (40 minutes)
- Identify image classification solutions
- Identify object detection solutions
- Identify facial detection solutions
Module 8: Identify Azure Tools and Services for Computer Vision Tasks (40 minutes)
- Identify Computer Vision service capabilities
- Identity Face service capabilities
- Identify Form Recognizer service capabilities
Module 9: Describe Features of NLP Workloads in Azure (40 minutes)
- Identify common NLP workload scenarios
- Identify entity recognition features and uses
- Identify language modeling features and uses
(Break: 5 minutes)
Module 10: Identify Common Uses Cases for Conversational AI (40 minutes)
- Identify webchat bot features and uses
- Identify telephone voice menu features and uses
- Identify personal digital assistant features and uses
Module 11: Identify Azure Services for Conversational AI (40 minutes)
- Identify QnA Maker service capabilities
- Identify Bot Framework service capabilities
Module 12: Exam AI-900 Strategy (40 minutes)
- Microsoft Online testing process
- Exam item tips and tricks
Review, wrap-up: 10 minutes