Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

IU > Introduction to Cloud Computing - Unit 5 > Flashcards

Introduction to Cloud Computing - Unit 5 Flashcards

(59 cards)

Start Studying
1
Q

What is MLflow?

A

An open-source platform to streamline the process of developing and deploying ML models in a version-controlled manner.

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

What components compose MLflow, and what do they mean?

A
  • MLflow Tracking: ability to track chosen parameters and models during different runs
  • MLflow Projects: portable projects consisting of ML models
  • MLflow Models
  • MLflow Model Registry: centralized location for trained models
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

When packaging ML models into MLflow Projects, which technologies are used?

A

Conda and Docker

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

What is a virtual environment in Python?

A

An isolated context with defined dependencies.

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

What is the relationship between an experiment and a run in MLflow?

A

An experiment can hold any number of runs.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

How is automatic logging acheived with MLflow in Python?

A

Through the .autolog() method.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the MLflow UI?

A

A locally-hosted webpage that allows visualization and comparision of different models.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is one of the main offers of Databricks?

A

Fully managed Spark clusters.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the source of the word Lakehouses? What are they in the context of Databricks?

A

A combination of data lakes and data warehouse. They are a unified data store solution for structured and unstructured data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What does the acryonym ACID have to do with, and what does it stand for?

A

Database principles to follow:

  • atomicity
  • consistency
  • isolation
  • durability
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are Delta Lakes used for?

A

Big data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What different tables exist in Delta Live Tables?

A
  • bronze: unprocessed raw data
  • silver: partly pre-processed, enriched data
  • gold: ready-to-use data for business needs
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is Unity Catalog used for, and ontop of?

A

For data governance ontop of Delta Lake.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Does Databricks have only one environment in which it executes?

A

No, it depends on the context (SQL, streaming, ML, etc.)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What specialized service does Google Cloud offer for ML?

A

Specialized VMs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is the Google Cloud AI Platform?

A

A centralized collection of services developed for typical ML and DS workflows.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What does the Google Cloud offering AI Platform Pipelines offer?

A
  • regular re-training of data
  • CI/CD pipelines
  • labeling service
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is Google Vertex AI?

A

A development environment designed to cover typical data science workflows.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What technologies, in conjunction with Google Vertex AI, can be used to develop an ML model without coding at all?

A

AutoML and a GUI

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What languages does Google Vertex AI Workbench support?

A

Python, SQL, or R

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is transfer learning in ML?

A

Adapting a pre-trained ML model to particular use cases.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is the meaning of the following ML offerings on Google Cloud:

  • teachable machine
  • natural language AI
  • dialogflow
A
  • teachable machine: transfer learning without writing code
  • natural language AI: extract numeric information from unstructured text data
  • dialogflow: used to create chat- and voice-bots
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What is Amazon Elastic MapReduce used for, and with which technologies?

A

Big data processing using Apache Spark, Hive, Presto, or another big data processing framework.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What does AWS SageMaker offer?

Study These Flashcards
A

A broad range of services and tools covering the complete lifecycle of a typical data science project.

25
What is Amazon SageMaker Studio Lab?
A free tier of Amazon SageMaker
26
What is Amazon SageMaker Autopilot?
A fully automatic version of Amazon Sagemaker where training is performed automatically on the given data, with no code needed.
27
What is Amazon SageMaker Canvas?
An interactive UI for working with Amazon SageMaker without code.
28
What is Amazon SageMaker Studio?
The entry point for all activities and the hub for all sub-services in Amazon SageMaker.
29
What is Amazon SageMaker JumpStart?
A quickstart coding experience in Amazon SageMaker.
30
What is Amazon SageMaker Data Labeling?
A labeling service for data so that supervised machine learning tasks may be performed.
31
What is Amazon SageMaker Data Wrangler?
A service to pre-process data for data science projects, such as outlier detection, imputation, etc.
32
What steps does Amazon SageMaker Data Wrangler pre-process data in?
1. import 2. prepare 3. analyse 4. export
33
What is Amazon SageMaker Feature Store?
A managed repository to store and share machine learning features.
34
What is the centerpiece data structure in Amazon SageMaker?
Notebook instances.
35
What is Amazon SageMaker Clarify?
A tool to evaluate model fairness/prejudice, and automatically detect class, prediction, and accuracy imbalances.
36
What is Amazon SageMaker Pipelines?
A framework for CI/CD pipelines in Amazon SageMaker.
37
What is Comprehend on AWS?
A Natural Language Processing (NLP) service providing pre-trained ML models.
38
What is Rekognition on AWS?
A Computer Vision service providing pre-trained ML models that focuses on facial recognition, OCR, etc.
39
What is Lookout for Vision and Panorama on AWS?
A Computer Vision service providing pre-trained ML models that focuses on site imagery and IoT use cases.
40
What is Textract on AWS?
A Computer Vision service providing pre-trained ML models that focuses on text-related tasks.
41
What is Translate on AWS?
A translation service providing pre-trained ML models.
42
What is Transcribe on AWS?
A speech-to-text service providing pre-trained ML models.
43
What is Polly on AWS?
A text-to-speech service providing pre-trained ML models.
44
What is Lex on AWS?
A service providing a quick way to configure chatbots.
45
What is Personalize on AWS?
A recommender service providing pre-trained ML models that provide user-specific real-time suggestions.
46
What is Forecast on AWS?
A time series analysis service providing pre-trained ML models.
47
What is Fraud Detector on AWS?
A time series analysis service providing pre-trained ML models that focuses on fraud detection.
48
What is Lookout for Metrics on AWS?
An anomaly detection service for KPIs.
49
What is Kendra in AWS?
A search and indexing engine.
50
What is Augemented AI (A2I) on AWS?
A machine learning service that allows human interaction to a part of the training process.
51
When creating an Azure ML workspace, what is generated in the same resource group?
* storage account (default data storage) * container registry * key vault * applications insights
52
What are compute targets in Azure ML?
Computation services to train and deploy ML models.
53
What choices exist for the underlying compute technology of compute targets in Azure ML?
Compute Instances or Compute Clusters.
54
What does Azure ML use to abstract the underlying external sources of data?
Datasets
55
What is Azure ML Designer?
A visual, drag-drop ML model generator.
56
How is an ML model deployed in Azure?
* register model in Model Registry * containerize * register container in Container Registry * deploy
57
What is the name of the managed set of pre-trained ML models in Azure called?
Azure Cognitive Services.
58
What is QnA Maker in Azure?
A service to create chatbots in Azure.
59
What is Speech Service in Azure?
A text-to-speech and a speech-to-text service.

IU (11 decks)

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