Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

Azure > Azure AI Lists 01 > Flashcards

Azure AI Lists 01 Flashcards

1
Q

Machine learning model types

A
  • Anomaly Detection
  • Classification
  • Clustering
  • Regression
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Steps in the machine learning process

A
  • Prepare data
  • Train model
  • Evaluate performance
  • Deploy a predictive service
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Metrics for evaluating regression models

A
  • Mean absolute error (MAE)
  • Root mean squared error (RMSE)
  • Relative absolute error (RAE)
  • Relative squared error (RSE)
  • Mean zero one error (MZOE)
  • Coefficient of determination (R-squared or R2)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Metrics for evaluating classification models

A
  • Accuracy
  • Precision
  • Recall
  • F1 score
  • Area Under Curve (AUC)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

AutoML machine learning operations

A
  • Classification
  • Regression
  • Time series forecasting
  • Natural language processing (NLP)
  • Computer vision
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Responsible AI principles

A
  • Fairness
  • Reliability and Safety
  • Privacy and Security
  • Inclusiveness
  • Accountability
  • Transparency
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Responsible conversational AI factors

A
  • Bot logic
  • Speech capture
  • Speech synthesis
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Cognitive Service main components

A
  • Anomaly Detector
  • Cognitive Search
  • Cognitive Service for Language
  • Cognitive Service for Vision
  • OpenAI
  • Speech
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Cognitive Service for Language components

A
  • Conversational Language Understanding
  • Language Understanding (LUIS)
  • Text Analytics
  • Translator
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Cognitive Services for Vision

A
  • Computer Vision
  • Custom Vision
  • Face
  • Form Recognizer
  • Video Indexer
  • Azure Spacial Analysis
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Cognitive Search features

A
  • Data from any source
  • Full text search and analysis
  • AI powered search
  • Multi-lingual
  • Geo-enabled
  • Configurable user experience
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Cognitive Search built-in skills

A

Natural language processing skills:
* Key Phrase Extraction
* Text Translation Skill

Image processing skills:
* Image Analysis Skill
* Optical Character Recognition Skill

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Cognitive Search projection types

A
  • Table
  • Object (JSON docs)
  • File (JPG images)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Cognitiive Search objects created

A
  • Data Source
  • Index
  • Indexer
  • Skillset
  • Knowledge store
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Cognitive Search for Language scenarios

A
  • Extract information
  • Summarize text-based content
  • Classify Text
  • Answer questions
  • Understand conversations
  • Translate text
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Cognitive Search model components

A
  • Utterances
  • Entities
  • Intents
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Cognitive Search entity types

A
  • Machine-learned
  • List
  • RegEx
  • Pattern.any
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Default features of a Language Service resource

A
  • Sentiment analysis
  • Key phrase extraction
  • Pre-built question answering
  • Conversational language understanding
  • Named entity recognition
  • Text summarization
  • Text analytics for health
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Custom Language Service features

A
  • Custom question answering
  • Custom text classification
  • Custom named entity recognition
  • Custom summarization
  • Custom sentiment analysis
  • Custom text analytics for health
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Text Analytics scenarios

A
  • Identify and categorize important concepts
  • Identify the main points in unstructured text
  • Better understand customer perception
  • Process unstructured medical data
  • Create a conversational layer over data
  • Automate workflow
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Image classification evaluation metrics

A
  • Precision
  • Recall
  • Average precision (AP)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Cognitive services for face analysis

A
  • Computer Vision
  • Video Indexer
  • Face
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Attributes generated by the Face service

A
  • Bounding box
  • Age
  • Gender
  • Emotion
  • Glasses
  • Hair
  • Facial hair
  • Makeup
  • Smile
  • Occlusion
  • Blur
  • Exposure
  • Noise
  • Head pose
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Spacial Analysis scenarios

A
  • People counting
  • Entrance counting
  • Social distancing and face mask detection
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

Speech recognition model types

A
  • Acoustic model
  • Language model
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

Metric for evaluating clustering models

A
  • Average distance to cluster center
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

Machine learning assets in a workspace

A
  • Compute
  • Datasets
  • Pipelines
  • Experiments
  • Models
  • Endpoints
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

Compute targets for machine learning

A
  • Compute instance
  • Compute cluster
  • Inference cluster
  • Attached compute
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

Languages used in Visual Studio Code for machine learning

A
  • .Net
  • R
  • Python (Jupyter notebooks)
30
Q

Machine Learning studio supported approaches

A
  • Automated machine learning
  • Python using Jupyter notebooks
  • Visual drag-and-drop designer
31
Q

Machine learning workspace data stores

A
  • Data used for training and evaluating models
  • Files such as logs and output files
32
Q

Machine Learning studio data sources

A
  • Files uploaded from a local computer
  • A datasource associated with the workspace
  • Files accessed via an HTTP URL
  • Open Datasets
33
Q

Machine Learning studio dataset types

A
  • Tabular
  • File
34
Q

Machine Learning studio tabular file types

A
  • CSV
  • TSV
  • Parquet
  • JSON
  • Output of a SQL query
35
Q

Categories of options for publishing a real-time inferencing model

A
  • Web service running in a Docker container
  • Open Neural Network Exchange (ONNX) platforms
36
Q

Options for publishing a real-time inferencing model as a web service running in a Docker container

A
  • Azure Kubernetes Service (AKS) cluster
  • Azure Container Instances (ACI)
  • Azure IoT Edge
37
Q

Steps for creating an AutoML run

A
  1. Select a dataset
  2. Name the experiment
  3. Select the compute for training
  4. Select the label
  5. Select the model type (classification, regression, or time-series forecasting)
  6. Select the primary metric
  7. Set the training time
38
Q

Machine Learning designer languages

39
Q

Steps for using a trained model in a real-time inference pipeline

A
  • Replace the dataset with a Web Service Input module.
  • Replace the evaluation steps with a Web Service Output module.
40
Q

Categories of Cognitive Services

A
  • Decision
  • Language
  • Speech
  • Vision
  • Web search
41
Q

Cognitive Services for Decision

A
  • Anomaly Detector
  • Metrics Advisor
  • Content Moderator
  • Azure Personalizer
42
Q

Cognitive Services for Language

A
  • Immersive Reader
  • Language Understanding (LUIS)
  • Conversational Language Understanding
  • Text Analytics
  • Translator
43
Q

Cognitive Services for Speech

A
  • Speech to Text
  • Text to Speech
  • Speech Translation
  • Speaker Recognition
44
Q

Cognitive Services for Web Search (unclear: services have changed)

A
  • Bing Web Search (maybe)
  • Cognitive Search (no longer part of Cognitive Service)
45
Q

Image classification capabilities

A
  • Describe an image
  • Categorize an image
  • Tag an image (object detection)
46
Q

Object detection capabilities

A
  • Detect common objects
  • Tag visual features
  • Detect faces (including age and sex)
  • Identify brands and products
  • Identify landmarks
47
Q

OCR capabilities

A
  • Extract printed text
  • Extract handwritten text
48
Q

Facial detection and recognition capabilities

A
  • Detect faces
  • Analyze facial features
  • Recognize faces
  • Identify famous people
  • Detect eyeglasses and goggles
  • Detect beards
  • Identify emotions
49
Q

Form Recognizer guidelines for best results

A
  • Images must be JPEG, PNG, BMP, PDF, or TIFF formats.
  • File size must be less than 50 MB.
  • Image size between 50 x 50 pixels and 10,000 x 10,000 pixels.
  • For PDF documents, no larger than 17 inches x 17 inches.
50
Q

Face service guidelines for best results

A
  • Image Format: JPEG, PNG, GIF, or BMP.
  • File Size: 6 MB or smaller.
  • Face Size Range: From 36 x 36 up to 4096 x 4096.
  • No extreme angles (frontal view)
  • No occlusion (objects blocking face)
51
Q

Visual features from the Analyze operation

A
  • Adult (pornographic, racy, or gory)
  • Brands
  • Categories (86 categories)
  • Color
  • Description
  • Faces (coordinates, sex, age)
  • ImageType (clipart, line drawing, etc.)
  • Objects
  • Tags (list of words)
  • Celebrities
  • Landmarks
52
Q

Computer Vision supported files

A
  • Less than 4MB
  • Greater than 50x50 pixels
  • JPEG, PNG, GIF, or BMP
53
Q

OCR models

A
  • Read (images and PDFs, async)
  • OCR (older, images only, synchronous)
54
Q

Computer vision operations/capabilities

A
  • Analyze image
  • Describe image
  • Detect objects
  • Generate content tags
  • Identify domain-specific content (celebrities and landmarks)
  • Generate thumbnails
  • OCR
  • Moderate content
55
Q

Image classification domains

A
  • General
  • General [A1]
  • General [A2]
  • Food
  • Landmarks
  • Retail
  • Compact domains
56
Q

Object detection domains

A
  • General
  • General [A1]
  • Logo
  • Products on shelves
  • Compact domains
57
Q

Facial recognition operations

A
  • Verify
  • Identify
  • Find Similar
  • Group
58
Q

Form Recognizer pre-trained models

A
  • Business cards
  • Invoices
  • Receipts
59
Q

Text Analytics technique categories

A
  • Analysis of text
  • Language modeling
  • Analysis of speech
  • Translation
60
Q

Text analytics techniques

A
  • Tokenization
  • Statistics
  • Frequency
  • Part of speech tagging (PosTag)
  • Sentiment analysis
  • Language detection
61
Q

Language modeling techniques

A
  • Semantic modeling
  • Named entity recognition (NER)
  • Topic detection
62
Q

Language features for extracting information

A
  • Extract key phrases
  • Find linked entities
  • Named Entity Recognition (NER)
  • Custom Named Entity Recognition (custom NER)
  • Personally Identifiable Information (PII) detection
  • Personal Health Information (PHI) detection
  • Text analytics for health
  • Custom text analytics for health
63
Q

Language features for summarizing text

A
  • Document summarization
  • Conversation summarization
64
Q

Language features for classifying text

A
  • Analyze sentiment and mine text for opinions
  • Detect language
  • Custom text classification
65
Q

Language features for understanding conversations

A
  • Conversational language understanding
  • Orchestration workflow
66
Q

Major NLP workloads

A
  • Language modeling
  • Key phrase extraction
  • Named entity recognition (NER)
  • Sentiment analysis
  • Speech recognition and synthesis
  • Translation
67
Q

Named entity recognition (NER) entity categories

A
  • Person (name)
  • PersonType (job)
  • DateTime
  • Quantity (dimensions, age, temperature, etc.)
  • Location (landmark, building, city)
  • Organization
  • Event
  • Product
  • Skill
  • Address
  • Phone number
  • Email
  • URL
  • IP address
68
Q

Language Understanding (LUIS) prebuilt domains

A
  • Calendar
  • Communication
  • Email
  • HomeAutomation
  • Notes
  • Places
  • RestaurantReservation
  • ToDo
  • Utilities
  • Weather
  • Web
69
Q

Language Understanding (LUIS) prebuilt entities

A
  • Age
  • Currency (money)
  • DatetimeV2
  • Dimension
  • Email
  • GeographyV2
  • KeyPhrase
  • Number
  • Ordinal
  • OrdinalV2
  • Percentage
  • PersonName
  • Phonenumber
  • Temperature
  • URL
70
Q

Bot tools

A
  • Bot Framework Composer (build bots)
  • Bot Framework Emulator (test bots locally)
71
Q

Bot lifecycle steps

A
  • Plan
  • Build
  • Test
  • Publish
  • Connect to channels
  • Evaluate

Azure (4 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions