Font size
S
M
L
XL
WorksheetsAI900-06
Total questions: 40
Worksheet time: 20mins
Name
Class
Date
1.
Classification is a type of __ in which models learn using training data and apply those learning to new data
a)
Supervised Learning
b)
Unsupervised Learning
2.
WHen using AutoML, the goal of classification models is to predict which categories new data will fall into based on learnings from training data. The possible classifications are
a)
binary classification
b)
multiclass classification
c)
multi-label classification
3.
Using AutoML, you would like to do holiday detection in your data. What would you use
a)
Classification
b)
Regression
c)
Time Series Forecasting
4.
When using AutoML, you want to make ensure high quality input data is being used to train the data. What would you use?
a)
Data Guard Rails
b)
ML Explainability (MLX)
c)
Experiementations
d)
Model Validation
5.
You are using AutoML, you would like to normalize the data for the model. Which are valid normalization options?
a)
StandardScaleWrapper
b)
MinMaxScalar
c)
MaxAbsScaler
d)
RobustScalar
e)
Principal component analysis (PCA)
6.
T/F AutoML will make recommendatioms on which machine learning althorithm model you should use
a)
"TRue"
b)
"False"
7.
WIth AutoML, what can be used to help understand and interpret a model's behavior?
a)
ML Explainability (MLX)
b)
Data Guard Rails
c)
Experiementations
d)
Model Validation
8.
__ is when you compare the results of the training dataset against the test dataset
a)
Model Validation
b)
ML Explainability (MLX)
c)
Data Guard Rails
d)
Experiementations
9.
Model Validation happens BEFORE/AFTER you train the model
a)
After
b)
Before
10.
T/F Custom Vision is a which of the following
a)
fully-managed
b)
no-code
c)
object detection
d)
programmable
11.
Custom vision is which of the following
a)
Supervised Learning
b)
Unsupervised Learning
c)
uses labeled data
d)
doesn't use labeled data
12.
With Computer Vision, when you want to apply many tags to an image, you use which classification
a)
Multi-label classification
b)
multiclass classification
c)
binary classification
13.
With Computer Vision, you want to classify photographs of dishes to determine what food is being shown. Which classification domain would you use?
a)
Food
b)
General
c)
Retail
d)
Landmarks
14.
With Computer Vision, you want to detect logos within images. What object detction domain would you use
a)
Logo
b)
Products and Shelves
c)
General
d)
General (Compact)
15.
With Computer Vision's___, you label the entire picture
a)
Image Classification
b)
Object Detection
16.
With Computer Vision's __, you draw a box around something within a picture and give the box a lebel
a)
Object Detection
b)
Image Classification
17.
With Computer Vision's model Advanced Training, you can select how much time to train the model by __
a)
moving a slider bar
b)
making selections in a drop down
c)
entering the time into the field
d)
you can not customize the time
18.
You have trained your Computer Vision model and received a report on the model's classification performance. What will the report show
a)
Precision
b)
Recall
c)
Average Preceision (AP)
d)
Worf, Data, Crusher
19.
You have trained your Computer Vision model and received a report on the model's Object Detection performance. What will the report show
a)
Precision
b)
Recall
20.
In Azure Machine Learning Workspace, what used to be called Experiements is now called __
a)
Jobs
b)
Pipelines
c)
Data
d)
Compute
21.
In Azure Machine Learning Workspace, what used to be called Datasets is now called
a)
Jobs
b)
Pipelines
c)
Data
d)
Compute
22.
The Microsoft AI Key Elements are which
a)
Machine Learning
b)
Anomaly Detection
c)
Computer Vision
d)
Natural Language Processing
e)
Conversational AI
23.
Pipelines are part of which Microsoft AI elemeent
a)
Machine Learning
b)
Anomaly Detection
c)
Computer Vision
d)
Natural Language Processing
e)
Conversational AI
24.
Data and Compute Management is part of which Microsoft AI element
a)
Machine Learning
b)
Anomaly Detection
c)
Computer Vision
d)
Natural Language Processing
e)
Conversational AI
25.
Automated Machine Learning uses which
a)
Supervised Learning
b)
Unsupervised Learning
26.
T/F Automated Machine Learning will provide the best selected model for you
a)
"TRue"
b)
"False"
27.
Which of the following are examples of a Regression Model
a)
Mean Absolute Error (MAE)
b)
Root Mean Squared Error (RMSE)
c)
Releative Squared Error(RSE)
d)
Releative Absolute Error (RAE)
e)
Coefficient of Determination (R2)
28.
This compares the predicted value to the actual value
a)
Confusion Matrix
b)
Regression Model
c)
Anomaly Detection
d)
Guard Rails
29.
With a confusion matrix, if you predict a positive and it actual is true, your result is
a)
True Positive
b)
True Negative
c)
False Positive
d)
False Negative
30.
The accuracy of a model is determined by
a)
(True Positive+True Negative)/Total NUmber
b)
True Positive/Total number
c)
True Negative/Total Number
d)
True Positive/(True Positive+False Positive)
31.
The precision of a model is determined by
a)
(True Positive+True Negative)/Total NUmber
b)
True Positive/Total number
c)
True Negative/Total Number
d)
True Positive/(True Positive+False Positive)
32.
which of the following are clustering models
a)
avaergae distance to other center
b)
average distance to cluster center
c)
number of points
d)
max distance to cluster center
e)
combined evaluation
33.
Semantic Segmentation is part of which Azure AI service
a)
Computer Vision
b)
Machine Learning
c)
Anomaly Detection
d)
Natural Language Processing
e)
Conversational AI
34.
Which Vision service can detect people
a)
Computer Vision
b)
Face
c)
Custom Vision
d)
Form Recognizer
35.
gpt is a __
a)
Nueral network
b)
Machine Learning
c)
pre-trained
d)
Language structure AI
e)
digital currency
36.
WHat does GitHub Copilot do
a)
Automatically creates apps
b)
Makes code suggestions as a developer is coding
c)
Interprits language
d)
Debugs compiled code
37.
Which AI product can modify and create images
a)
DALL-E
b)
Copilot
c)
ChatGPS
38.
If you use characteristics of an object to predict something about that object, it is known as
a)
Regression
b)
Object Dectection
c)
Confusion Matrix
d)
Unsupervised Learning
39.
__ let you organize, manage, and reuse complex ML workflows across projects and users
a)
Pipeline
b)
Confusion Matrix
c)
Regression Model
d)
Nueral network
40.
Pipelines are part of which Microsoft AI Element
a)
Machine Learning
b)
Anomaly Detection
c)
Computer Vision
d)
Natural Language Processsing
e)
Conversational AI
Reset
