book

AWS Certified Machine Learning - Specialty (MLS-C01) Certification Guide - Second Edition

by Somanath Nanda, Weslley Moura

February 2024

Intermediate to advanced

342 pages

8h 15m

English

Packt Publishing

Read now

Unlock full access

Second EditionContributorsAbout the AuthorsAbout the Reviewer
Who This Book Is forWhat This Book CoversHow to Use This BookOnline Practice ResourcesDownload the Color ImagesConventions UsedGet in TouchDownload a Free PDF Copy of This Book
Making The Most Out of This Book – Your Certification and BeyondComparing AI, ML, and DLExamining MLExamining DLClassifying supervised, unsupervised, and reinforcement learningIntroducing supervised learningThe CRISP-DM modeling life cycleData splittingOverfitting and underfittingApplying cross-validation and measuring overfittingBootstrapping methodsThe variance versus bias trade-offShuffling your training setModeling expectationsIntroducing ML frameworksML in the cloudSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing
Technical requirementsStoring Data on Amazon S3Creating buckets to hold dataDistinguishing between object tags and object metadataControlling access to buckets and objects on Amazon S3S3 bucket policyProtecting data on Amazon S3Applying bucket versioningApplying encryption to bucketsSecuring S3 objects at rest and in transitUsing other types of data storesRelational Database Service (RDS)Managing failover in Amazon RDSTaking automatic backups, RDS snapshots, and restore and read replicasWriting to Amazon Aurora with multi-master capabilitiesStoring columnar data on Amazon RedshiftAmazon DynamoDB for NoSQL Database-as-a-ServiceSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing
Technical requirementsCreating ETL jobs on AWS GlueFeatures of AWS GlueGetting hands-on with AWS Glue Data Catalog componentsGetting hands-on with AWS Glue ETL componentsQuerying S3 data using AthenaProcessing real-time data using Kinesis Data StreamsStoring and transforming real-time data using Kinesis Data FirehoseDifferent ways of ingesting data from on-premises into AWSAWS Storage GatewaySnowball, Snowball Edge, and SnowmobileAWS DataSyncAWS Database Migration ServiceProcessing stored data on AWSAWS EMRAWS BatchSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing
Identifying types of featuresDealing with categorical featuresTransforming nominal featuresApplying binary encodingTransforming ordinal featuresAvoiding confusion in our train and test datasetsDealing with numerical featuresData normalizationData standardizationApplying binning and discretizationApplying other types of numerical transformationsUnderstanding data distributionsHandling missing valuesDealing with outliersDealing with unbalanced datasetsDealing with text dataBag of wordsTF-IDFWord embeddingSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing
Visualizing relationships in your dataVisualizing comparisons in your dataVisualizing distributions in your dataVisualizing compositions in your dataBuilding key performance indicatorsIntroducing QuickSightSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing
Introducing this chapterStoring the training dataA word about ensemble modelsSupervised learningWorking with regression modelsIntroducing regression algorithmsLeast squares methodCreating a linear regression model from scratchInterpreting regression modelsChecking adjusted R squaredRegression modeling on AWSWorking with classification modelsForecasting modelsChecking the stationarity of time seriesExploring, exploring, and exploringUnderstanding DeepARObject2VecUnsupervised learningClusteringComputing K-Means step by stepDefining the number of clusters and measuring cluster qualityConclusionAnomaly detectionDimensionality reductionUsing AWS’s built-in algorithm for PCAIP InsightsTextual analysisBlazingText algorithmSequence-to-sequence algorithmNeural Topic Model algorithmImage processingImage classification algorithmSemantic segmentation algorithmObject detection algorithmSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing
Introducing model evaluationEvaluating classification modelsExtracting metrics from a confusion matrixSummarizing precision and recallEvaluating regression modelsExploring other regression metricsModel optimizationGrid searchSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing
Technical requirementsAnalyzing images and videos with Amazon RekognitionExploring the benefits of Amazon RekognitionGetting hands-on with Amazon RekognitionText to speech with Amazon PollyExploring the benefits of Amazon PollyGetting hands-on with Amazon PollySpeech to text with Amazon TranscribeExploring the benefits of Amazon TranscribeGetting hands-on with Amazon TranscribeImplementing natural language processing with Amazon ComprehendExploring the benefits of Amazon ComprehendGetting hands-on with Amazon ComprehendTranslating documents with Amazon TranslateExploring the benefits of Amazon TranslateGetting hands-on with Amazon TranslateExtracting text from documents with Amazon TextractExploring the benefits of Amazon TextractGetting hands-on with Amazon TextractCreating chatbots on Amazon LexExploring the benefits of Amazon LexGetting hands-on with Amazon LexAmazon ForecastExploring the benefits of Amazon ForecastSales Forecasting Model with Amazon ForecastSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing

Technical requirementsCreating notebooks in Amazon SageMakerWhat is Amazon SageMaker?Training Data Location and FormatsGetting hands-on with Amazon SageMaker notebook instancesGetting hands-on with Amazon SageMaker’s training and inference instancesModel tuningTracking your training jobs and selecting the best modelChoosing instance types in Amazon SageMakerChoosing the right instance type for a training jobChoosing the right instance type for an inference jobTaking care of Scalability ConfigurationsScaling Policy OverviewScale Based on a ScheduleMinimum and Maximum Scaling LimitsCooldown PeriodSecuring SageMaker notebooksSageMaker DebuggerSageMaker AutopilotSageMaker Model MonitorSageMaker Training CompilerSageMaker Data WranglerSageMaker Feature StoreSageMaker Edge ManagerSageMaker CanvasSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing
Factors influencing model deployment optionsSageMaker deployment optionsReal-time endpoint deploymentSolutionStepsExample code snippetBatch transform jobSolutionStepsExample code snippetMulti-model endpoint deploymentSolutionStepsExample code snippetEndpoint autoscalingSolutionStepsExample code snippetServerless APIs with AWS Lambda and SageMakerSolutionStepsExample code snippetCreating alternative pipelines with Lambda FunctionsCreating and configuring a Lambda FunctionCompleting your configurations and deploying a Lambda functionWorking with step functionsScaling applications with SageMaker deployment and AWS AutoscalingScenario 1 – Fluctuating inference workloadsAutoscaling solutionStepsExample code snippetScenario 2 – The batch processing of large datasetsAutoscaling solutionStepsExample code snippetScenario 3 – A multi-model endpoint with dynamic trafficAutoscaling solutionStepsExample code snippetScenario 4 – Continuous Model Monitoring with drift detectionAutoscaling solutionStepsSecuring SageMaker applicationsSummaryExam Readiness Drill – Chapter Review QuestionsExam Readiness DrillATTEMPT 1ATTEMPT 2ATTEMPT 3Working On Timing
How to Access These ResourcesPurchased from Packt Store (packtpub.com)Packt+ SubscriptionPurchased from Amazon and Other SourcesSTEP 1STEP 2STEP 3STEP 4STEP 5Troubleshooting TipsPractice Resources – A Quick TourA Clean, Simple Cert Practice ExperiencePractice QuestionsFlashcardsExam TipsChapter Review QuestionsShare FeedbackBack to the Book
Other Books You May EnjoyShare Your ThoughtsDownload a Free PDF Copy of This Book

Content preview from AWS Certified Machine Learning - Specialty (MLS-C01) Certification Guide - Second Edition

3 AWS Services for Data Migration and Processing

In the previous chapter, you learned about several ways of storing data in AWS. In this chapter, you will explore the techniques for using that data and gaining some insight from the data. There are use cases where you have to process your data or load the data to a hive data warehouse to query and analyze the data. If you are on AWS and your data is in S3, then you have to create a table in hive on AWS EMR to query the data in the hive table. To provide the same functionality as a managed service, AWS has a product called Athena, where you create a data catalog and query your data on S3. If you need to transform the data, then AWS Glue is the best option to transform and restore it to S3. Imagine ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Start your free trial

AWS Certified Machine Learning Specialty: MLS-C01 Certification Guide

Publisher Resources

ISBN: 9781835082201Supplemental Content

AWS Certified Machine Learning - Specialty (MLS-C01) Certification Guide - Second Edition

by Somanath Nanda, Weslley Moura

3

AWS Services for Data Migration and Processing

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

You might also like