Data Engineering Practice Exam 1

Question 1

Confusion matrix

Answer 1

A table that illustrates the number or percentage of correct and incorrect predictions for each class by comparing an observation’s predicted class and its true class

Question 2

Precision

Answer 2

The proportion of true positive predictions among all positive predictions

Question 3

Recall

Answer 3

The proportion of actual positives that are correctly identified

Question 4

Accuracy

Answer 4

The percentage of correct predictions out of all predictions made by the model

Question 5

Root mean squared error (RMSE)

Answer 5

A regression metric that measures the average magnitude of the errors between predicted and actual values

Question 6

AUC-ROC curve

Answer 6

A tool to evaluate a model’s ability to distinguish between classes across various thresholds; particularly useful in the presence of class imbalance

Question 7

Blue/green deployment

Answer 7

A strategy that deploys a new version of a model in parallel with the existing one; gradually shifting traffic to the new version while monitoring its performance

Question 8

Canary release

Answer 8

A deployment strategy where a small percentage of traffic is redirected to a new model version initially

Question 9

Amazon SageMaker Pipelines

Answer 9

A purpose-built workflow orchestration service to automate machine learning (ML) development

Question 10

Amazon SageMaker Data Wrangler

Answer 10

A service that reduces the time it takes to aggregate and prepare tabular and image data for ML from weeks to minutes

Question 11

Pipe input mode

Answer 11

A data streaming method where data is pre-fetched from Amazon S3 at high concurrency and throughput; and streamed into a named pipe

Question 12

File input mode

Answer 12

A data input method that downloads the entire dataset to the training instance before starting the training job

Question 13

FastFile mode

Answer 13

A data access method for scenarios where rapid access to data with low latency is needed; best suited for workloads with many small files

Question 14

Amazon SageMaker Serverless Inference

Answer 14

A deployment option that automatically scales compute resources based on incoming requests; cost-effective for workloads with idle periods between traffic spikes

Question 15

Amazon Bedrock

Answer 15

A fully managed service that offers a choice of high-performing foundation models (FMs) from leading AI companies through a single API

Question 16

SageMaker JumpStart

Answer 16

A machine learning (ML) hub that provides managed infrastructure and tools to accelerate scalable; reliable; and secure model building; training; and deployment of ML models

Question 17

XGBoost

Answer 17

An efficient implementation of gradient boosted trees algorithm that performs well in handling a variety of data types; relationships; and distributions

Question 18

Random Cut Forest (RCF)

Answer 18

An unsupervised algorithm for detecting anomalous data points within a data set

Question 19

Amazon SageMaker Feature Store

Answer 19

A fully managed; purpose-built repository to store; share; and manage features for machine learning (ML) models

Question 20

Script mode

Answer 20

A SageMaker feature that enables writing custom training and inference code while still utilizing common ML framework containers maintained by AWS

Question 21

Stacking

Answer 21

An ensemble learning technique where predictions from several base models are used as inputs to a meta-model to make the final prediction

Question 22

Area Under the (Receiver Operating Characteristic) Curve (AUC)

Answer 22

An industry-standard accuracy metric for binary classification models that measures the ability of the model to predict a higher score for positive examples compared to negative examples

Question 23

Amazon SageMaker Model Registry

Answer 23

A service for cataloging; managing versions; and tracking metadata of models; as well as managing approval status and deployment of models

Question 24

Bayesian Optimization

Answer 24

A technique based on Bayes’ theorem for hyperparameter optimization that builds a probabilistic model from a set of hyperparameters to optimize a specific metric

Question 25

Data drift

Answer 25

Changes in the distribution of the input data over time

Question 26

Model drift

Answer 26

Degradation in model performance because its assumptions or parameters no longer align with the real-world data

Question 27

Amazon SageMaker Model Monitor

Answer 27

A service for detecting data drift by tracking changes in data distribution

Question 28

Conditional Demographic Disparity (CDD)

Answer 28

A metric that measures the difference in positive prediction rates between demographic groups; while conditioning on relevant features

Question 29

Exploratory Data Analysis (EDA)

Answer 29

A process to understand data distribution; identify and address missing values; and assess the extent of class imbalance

Question 30

Amazon SageMaker Debugger

Answer 30

A tool that provides debugging capabilities for training jobs; addressing problems such as overfitting; saturated activation functions; and vanishing gradients

Question 31

AWS CloudFormation with nested stacks

Answer 31

A method to modularize infrastructure; making it easier to manage and reuse components while enabling communication between stacks