Data & AI Essentials: Mastering Azure ML

https://learn.microsoft.com/training/challenges?id=1d37aaec-1ca7-450f-
b0b722ecd18bcec8&WT.mc_id=cloudskillschallenge_1d37aaec-1ca7-450f-b0b7-
22ecd18bcec8&wt.mc_id=studentamb_330281

Note: Answer – Answers are highlighted in Yellow color.

1. Introduction to machine learning

Q) What makes machine-learning algorithms different from traditional algorithms?

Machine-learning algorithms are always more complicated to build than traditional algorithms.
Machine-learning algorithms must be trained every time they're used.
Machine-learning algorithms are shaped by data directly as part of development. Traditional
algorithms are based almost entirely on theory or on opinions of the person writing the code.

Q) When do we want to perform training?

Whenever we want to use a model
Only when we want to improve the model
Every time we load a model from file

Q) What is the relationship between a model, an objective, and training data?

The training data is used to make changes to the model. These changes help the model get better at
achieving the objective.
The training data is used to make changes to the objective. These changes help the objective be more
like the model.
The model is used to make changes to the training data. These changes help the training data get
better at achieving the objective.

2. Fundamental AI Concepts

Q) You want to create a model to predict sales of ice cream based on historic data that includes daily ice
cream sales totals and weather measurements. Which Azure service should you use?

Azure Machine Learning

Azure AI Bot Service
Azure AI Language

Q) You work for a wildlife sanctuary and are considering using AI to identify bird species from images.
Which AI service should you use to prototype your idea?
Azure AI Vision
Azure AI Search
Azure OpenAI

Q) A predictive app provides audio output for visually impaired users. Which principle of Responsible
AI is reflected here?
Transparency
 Inclusiveness
Fairness

3. Introduction to Azure SQL

Unit 3 of 6
Q) Consider this scenario: you want to migrate to the cloud, but you use a third-party application that
requires access to the operating system (OS). Which Azure SQL deployment option will be easiest to use?

SQL Server in an Azure virtual machine

Azure SQL Managed Instance
Azure SQL Database, single database
Azure SQL Database, elastic database pool

Q) Consider this scenario: you want to migrate to the cloud and remove some of the management associated
with SQL Server, but your application uses CLR and Service Broker capabilities from SQL Server. Which
Azure SQL deployment option will be easiest to use?
SQL Server in an Azure virtual machine
Azure SQL Managed Instance
Azure SQL Database, single database
Azure SQL Database, elastic database pool

Unit 4 of 6
Q) You're moving an application and database to Azure, but your database is currently 62 TB and will
continue to grow. You don't currently use any instance-scoped features. Which Azure SQL deployment
option will be easiest to use?
SQL Server in an Azure virtual machine
Azure SQL Managed Instance
Azure SQL Database, single database
Azure SQL Database, elastic database pool

Q) Imagine you have Azure SQL Database with the serverless compute tier database deployed and an auto-
pause delay of two hours. After two hours of no activity, what happens to your database and incurred
charges?
Azure will pause your database to stop compute costs and charge you only for storage.
Database resources are reduced to a lower service tier to reduce costs.

4. Explore fundamental relational data concepts

Q) Which one of the following statements is a characteristic of a relational database?

All columns in a table must be of the same data type.
A row in a table represents a single instance of an entity.
Rows in the same table can contain different columns.
Q) Which SQL statement is used to query tables and return data?
QUERY
READ
SELECT

Q) What is an index?
A structure that enables queries to locate rows in a table quickly
A virtual table based on the results of a query
A predefined SQL statement that modifies data

5. Create and connect to an Azure Database for PostgreSQL

Q) Azure Database for PostgreSQL is a PaaS solution. If you ran PostgreSQL as IaaS, what added
responsibilities would you have?
Applying patches
Making backups
Managing virtual machines
All of the above

Q) What is the longest time the automatic backups can last for?

Seven days
3 days
35 days
20 days

Q) What is the cheapest pricing tier for the single server deployment option?
General
Memory optimized
Basic

6. Query relational data in Azure - No Test

7. Build a machine learning model

Q) Why did we choose a decision tree for our machine learning algorithm?
A decision tree is the most complex and accurate algorithm.
A decision tree is easy to visualize. It fits well because the model can make only two choices: yes or no.
A decision tree has lots of branches, and the model can make lots of choices.

Q) What is the purpose of splitting your dataset?

To make your model more accurate by getting rid of bad data.
To try out different algorithms with different data.
To have different data for training and testing your model.
8. Fundamentals of Generative AI

Q) What are Large Language Models?

Models that only work with one language.

Models that only work with small amounts of data.
Models that use deep learning to process and understand natural language on a massive scale.

Q) What is an example of a potential task a generative AI application can help solve?

Monitoring the temperature in a manufacturing facility.
Creating a draft for an email.
Collecting real time data and storing it in a database.

Q) What is the purpose of vector-based embeddings?

To represent semantic meaning of text tokens.
To create tokens that include multiple representations of a word in different languages.
To correct misspellings in the training data.

Q) What is the potential impact of copilots?

Copilots only impact applications used in professional settings.
Copilots can help with first drafts, information synthesis, strategic planning, and much more.
Copilots can only be used for certain natural language tasks like summarizing text.

9. Fundamentals of Azure OpenAI Service

Q) How are ChatGPT, OpenAI, and Azure OpenAI related?

Azure OpenAI is Microsoft's version of ChatGPT, a chatbot that uses generative AI models.
ChatGPT and OpenAI are chatbots that generate natural language, code, and images. Azure OpenAI
provides access to these two chatbots.
OpenAI is a research company that developed ChatGPT, a chatbot that uses generative AI models.
Azure OpenAI provides access to many of OpenAI's AI models.

Q) You would like to summarize a paragraph of text. Which generative AI model family would you use to
solve for this workload?
GPT.
Codex.
Dall-E.

Q) What is one action Microsoft takes to support ethical AI practices in Azure OpenAI?
Provides Transparency Notes that share how technology is built and asks users to consider its
implications.
Logs users out of Azure OpenAI Studio after a period of inactivity to ensure it's only used by one user.
Allows users to build any application, regardless of harmful effects, to ensure fairness.
10. Challenge project - Build a speech translator app

Q) What service provides speech-to-text functionality in this solution?

Language Understanding
Speech API
Canvas controls

Q) Where should API keys be stored in a Power Apps solution?

Directly in formula bars
As app settings
API connections

Q) How can the Speech API model be improved over time?

Increase API call volume
Train with more speech samples
Purchase more keys

11. Explore relational database services in Azure

Q) Which deployment option offers the best compatibility when migrating an existing SQL Server on-
premises solution?
Azure SQL Database (single database)
Azure SQL Database (elastic pool)
Azure SQL Managed Instance

Q) Which of the following statements is true about Azure SQL Database?

Most database maintenance tasks are automated
You must purchase a SQL Server license
It can only support one database

Q) Which database service is the simplest option for migrating a LAMP application to Azure?

Azure SQL Managed Instance

Azure Database for MySQL
Azure Database for PostgreSQL