book

Google Machine Learning and Generative AI for Solutions Architects

Name: Google Machine Learning and Generative AI for Solutions Architects
Author: Kieran Kavanagh
ISBN: 9781803245270

by Kieran Kavanagh

June 2024

Beginner to intermediate

552 pages

16h 11m

English

Packt Publishing

Read now

Unlock full access

Google Machine Learning and Generative AI for Solutions Architects
Foreword
Contributors
About the authorAbout the reviewer
Preface
Who this book is forWhat this book coversTo get the most out of this bookDownload the example code filesConventions usedGet in touchShare Your ThoughtsDownload a free PDF copy of this book
Part 1:The Basics
Chapter 1: AI/ML Concepts, Real-World Applications, and Challenges
Terminology – AI, ML, DL, and GenAIA brief history of AI/MLAI/ML and cloud computingML approaches and use casesSLULRLA brief discussion of ML basic conceptsLinear algebraCalculusStatistics and probabilityMetricsCommon challenges in developing ML applicationsGathering, processing, and labeling dataOrganizational challengesOperationalization and ongoing management of AI/ML modelsEdge casesSummary
Chapter 2: Understanding the ML Model Development Life Cycle
An overview of the ML model development life cycleSDLC – a quick recapTypical ML project stagesRoles and personas in AI/ML projectsCommon challenges encountered in the ML model development life cycleFinding and gathering relevant dataPicking an algorithm and model architectureData labelingTraining modelsConfiguring and tuning hyperparametersEvaluating modelsDeploying modelsMonitoring models after deploymentBest practices for overcoming common challengesFinding and gathering relevant dataData labelingPicking an algorithm and model architectureTraining modelsConfiguring and tuning hyperparametersDeploying modelsMonitoring models after deploymentSummary
Chapter 3: AI/ML Tooling and the Google Cloud AI/ML Landscape
Why Google Cloud?Prerequisites for using Google Cloud tools and servicesSecurity, privacy, and complianceGoogle Cloud services overviewGoogle Cloud computing servicesGoogle Cloud integration servicesNetworking and connectivityGoogle Cloud tools for data storage and processingData ingestionData storageData managementData processingGoogle Cloud AI tools and AutoMLNLPThe Natural Language APIComputer visionAutoMLGoogle Cloud Vertex AIStandard industry tools on Google CloudChoosing the right tool for the jobSummary
Part 2:Diving in and building AI/ML solutions
Chapter 4: Utilizing Google Cloud’s High-Level AI Services
Prerequisites for this chapterCloning this book’s GitHub repository to your local machineThe Google Cloud consoleGoogle Cloud projectGoogle Cloud BillingGoogle Cloud ShellAuthenticationEnabling the relevant Google Cloud APIsStoring authentication credentials in environment variablesCreating a directory and cloning our GitHub repositoryDetecting text in images with the Cloud Vision APIUsing Document AI to extract information from documentsDocument AI conceptsPerforming OCR with Document AIUsing the Google Cloud Natural Language API to get sentiment analysis insights from textual inputsSentiment analysis with the Natural Language APIClassifying content with the Natural Language APIUsing Vertex AI AutoMLUse case – forecastingSummary

Chapter 5: Building Custom ML Models on Google Cloud
Background information – librariesscikit-learnMatplotlibpandasXGBoostPrerequisites for this chapterVertex AI WorkbenchCreating a Vertex AI Workbench instanceVertex AI Notebook JupyterLab integrationsCloning the GitHub repositoryUML with scikit-learn on Vertex AIK-meansImplementing a UML workload in Vertex AIImplementing a regression model with scikit-learn on Vertex AIImplementing a classification model with XGBoost on Vertex AIDecision treesSummary
Chapter 6: Diving Deeper – Preparing and Processing Data for AI/ML Workloads on Google Cloud
Prerequisites for this chapterEnabling APIsIAM permissionsCloud Storage bucket foldersUploading dataFundamental concepts in this chapterIngesting data into Google CloudETL and ELTBatch and streaming data processingData pipelinesExploring, visualizing, and preparing dataBatch data pipelinesBatch data pipeline concepts and toolsBuilding our batch data pipelineCloud ComposerGoogle Cloud Serverless SparkStreaming data pipelinesStreaming data pipeline concepts and toolsBuilding our streaming data pipelineCreating a BigQuery datasetCreating a BigQuery tableCreating a Dataflow job from a templateVerifying data in BigQueryCreating a Dataflow notebookVerifying data in BigQuerySummary
Chapter 7: Feature Engineering and Dimensionality Reduction
Fundamental concepts in this chapterDimensions and featuresOverfitting, underfitting, and regularizationFeature selection and feature engineeringThe curse of dimensionalityDimensionality reductiont-SNEUsing PCA and LDA for dimensionality reductionFeature engineeringVertex AI Feature StoreIntroduction to Vertex AI Feature StoreOnline versus offline feature servingBuilding our feature storeHow features are used during online inferenceSummary
Chapter 8: Hyperparameters and Optimization
PrerequisitesEnabling the Artifact Registry APICreating an AI/ML service accountConceptsModel evaluation metrics used in this chapterWhat are hyperparameters?Hyperparameter optimizationMethods for optimizing hyperparameter valuesHands-on: performing hyperparameter tuning in Vertex AIVertex AI VizierSummary
Chapter 9: Neural Networks and Deep Learning
NN and DL conceptsNeurons and the perceptronBackpropagationActivation functionsLibrariesTensorFlowKerasImplementing an MLP in TensorFlowNN architectures, challenges, and optimizationCommon NN architecturesCommon NN challengesSummary
Chapter 10: Deploying, Monitoring, and Scaling in Production
How do I make my models available to my applications?Fundamental concepts for serving modelsOnline and offline model servingVertex AI Model RegistryVertex AI prediction serviceA/B testingCommon challenges of serving models in productionDeployment infrastructureModel availability and scaling in productionData qualityModel/data/concept driftSecurity and privacyModel interpretabilityTracking ML model metadataIntegration with existing systemsMonitoringMonitoring models in productionObjective model performanceMonitoring specifically for data driftAnomalous model behaviorResource utilizationModel bias and fairnessAddressing model performance degradationOptimizing for AI/ML at the edgeModel optimizationOptimization beyond model techniquesSummary
Chapter 11: Machine Learning Engineering and MLOps with Google Cloud
An introduction to MLOpsWhy MLOps is needed for deploying large-scale ML workloadsModel management and versioningProductivity and automationReproducibilityContinuous Integration/Continuous Deployment (CI/CD)Model validation and testingMonitoring and maintenanceCollaboration and communicationRegulatory compliance and governanceMLOps toolsPipeline orchestration toolsExperiment and lineage tracking toolsModel deployment and monitoring toolsModel interpretability and explainability toolsImplementing MLOps on Google Cloud using Vertex AI PipelinesPrerequisite: IAM permissionsImplementationSummary
Chapter 12: Bias, Explainability, Fairness, and Lineage
An overview of bias, explainability, and fairness in AI/MLBiasFairnessExplainabilityHow to detect and mitigate bias in datasetsData exploration and visualizationSpecific tools for detecting dependencies between featuresMechanisms incorporating model prediction resultsUsing explainability to understand ML models and reduce biasExplainability techniques, methods, and toolsFeature importanceReducing bias and enhancing fairnessAdditional librariesThe importance of lineage tracking in ML model developmentML metadata service terminologyLineage tracking in Vertex AISummary
Chapter 13: ML Governance and the Google Cloud Architecture Framework
ML governanceData governanceML model governanceOperationalization of ML governanceML governance in different industries and locationsKeeping up with the evolving landscape of ML governanceAn overview of the Google Cloud Architecture FrameworkPillar 1 – System designPillar 2 – Operational excellencePillar 3 – Security, privacy, and compliancePillar 4 – ReliabilityPillar 5 – Cost optimizationPillar 6 – Performance optimizationArchitecture Framework concepts about AI/ML workloads on Google CloudData collection and preparationModel building and trainingModel evaluation and deploymentSummary
Chapter 14: Additional AI/ML Tools, Frameworks, and Considerations
Prerequisite topics and stepsStaging files for serverless Spark MLlib activitiesBQMLUsing BQMLWhen to use BQML versus other tools for AI/ML use casesBigQuery StudioHardware considerations for AI/ML workloadsCPUs, GPUs, and TPUsAdditional open source tools and frameworks –Spark MLlib, Ray, and PyTorch on Google CloudSpark MLlibRayPyTorchLarge-scale distributed model trainingData parallelism and model parallelismDistributed training update processFederated learningTransitioning to Generative AICNNs and computer visionRNNs, LSTMs, and transformersSummary
Part 3:Generative AI
Chapter 15: Introduction to Generative AI
Fundamentals of GenAIWhat is GenAI?What is non-GenAI?Diving deeper into GenAI versus non-GenAIGenAI techniques and evolutionEmbeddings and latent spaceLLMsEvolution of LLMsBuilding LLMsSummary
Chapter 16: Advanced Generative AI Concepts and Use Cases
Advanced tuning and optimization techniquesPrompt engineeringTLAligning with human values and expectationsEmbeddings and vector databasesEmbeddings and similarity of conceptsVector databasesHow vector databases workCreating embeddingsRAGHow RAG worksMultimodal modelsWhy multimodality mattersMultimodality challengesGenAI model evaluationHuman evaluationBLEUROUGEISFIDAuto-raters and side-by-side evaluationsLangChainSummary
Chapter 17: Generative AI on Google Cloud
Overview of generative AI in Google CloudGoogle’s generative AI modelsOpen source and third-party generative AI models on Google CloudVector databases in Google CloudA detailed exploration of Google Cloud generative AIGoogle Cloud Vertex AI StudioA detailed exploration of Google Cloud vector database optionsImplementing generative AI solutions on Google CloudBuilding a Vertex AI Search and Conversation applicationSummary
Chapter 18: Bringing It All Together: Building ML Solutions with Google Cloud and Vertex AI
PrerequisitesUsing the Google Cloud ShellBuilding a RAG implementation piece by pieceTokens and chunksExample RAG solution architectureBuilding the RAG implementation on Google CloudAn example business use case and reference architectureAdditional background on our use caseThe reference architectureBuilding and implementing the use case on Google CloudRecap and next stepsNext steps
Index
Why subscribe?
Other Books You May EnjoyPackt is searching for authors like youShare Your ThoughtsDownload a free PDF copy of this book

Content preview from Google Machine Learning and Generative AI for Solutions Architects

10 Deploying, Monitoring, and Scaling in Production

Some people may read this book from beginning to end to gain an overall understanding of as many concepts as possible in the realm of AI/ML on Google Cloud, while others may use it as a reference, whereby they pick it up and read certain chapters on specific topics whenever they need to work with those topics as part of a project or client engagement. If you’ve been reading this book from the beginning, then you have come a long way, and we have journeyed together through the majority of the ML model development life cycle (MDLC). While model training is what often gets the most attention in the press – and that is where a lot of the magic happens – you know by now that training is just one ...

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.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Generative AI on Google Cloud with LangChain

Publisher Resources

ISBN: 9781803245270

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Google Machine Learning and Generative AI for Solutions Architects

by Kieran Kavanagh

10

Deploying, Monitoring, and Scaling in Production

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.