book

Building Data-Driven Applications with LlamaIndex

Name: Building Data-Driven Applications with LlamaIndex
Author: Andrei Gheorghiu
ISBN: 9781835089507

by Andrei Gheorghiu

May 2024

Intermediate to advanced

368 pages

10h 7m

English

Packt Publishing

Read now

Unlock full access

Building Data-Driven Applications with LlamaIndex
ContributorsAbout the authorAbout the reviewers
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:Introduction to Generative AI and LlamaIndex
Chapter 1: Understanding Large Language Models
Introducing GenAI and LLMsWhat is GenAI?What is an LLM?Understanding the role of LLMs in modern technologyExploring challenges with LLMsAugmenting LLMs with RAGSummary
Chapter 2: LlamaIndex: The Hidden Jewel - An Introduction to the LlamaIndex Ecosystem
Technical requirementsOptimizing language models – the symbiosis of fine-tuning, RAG, and LlamaIndexIs RAG the only possible solution?What LlamaIndex doesDiscovering the advantages of progressively disclosing complexityAn important aspect to considerIntroducing PITS – our LlamaIndex hands-on projectHere’s how it will workPreparing our coding environmentInstalling PythonInstalling GitInstalling LlamaIndexSigning up for an OpenAI API keyDiscovering Streamlit – the perfect tool for rapid building and deployment!Installing StreamlitFinishing upOne final checkFamiliarizing ourselves with the structure of the LlamaIndex code repositorySummary
Part 2: Starting Your First LlamaIndex Project
Chapter 3: Kickstarting Your Journey with LlamaIndex
Technical requirementsUncovering the essential building blocks of LlamaIndex – documents, nodes, and indexesDocumentsNodesManually creating the Node objectsAutomatically extracting Nodes from Documents using splittersNodes don’t like to be alone – they crave relationshipsWhy are relationships important?IndexesAre we there yet?How does this actually work under the hood?A quick recap of the key conceptsBuilding our first interactive, augmented LLM applicationUsing the logging features of LlamaIndex to understand the logic and debug our applicationsCustomizing the LLM used by LlamaIndexEasy as 1-2-3The temperature parameterUnderstanding how Settings can be used for customizationStarting our PITS project – hands-on exerciseLet’s have a look at the source codeSummary
Chapter 4: Ingesting Data into Our RAG Workflow
Technical requirementsIngesting data via LlamaHubAn overview of LlamaHubUsing the LlamaHub data loaders to ingest contentIngesting data from a web pageIngesting data from a databaseBulk-ingesting data from sources with multiple file formatsParsing the documents into nodesUnderstanding the simple text splittersUsing more advanced node parsersUsing relational parsersConfused about node parsers and text splitters?Understanding chunk_size and chunk_overlapIncluding relationships with include_prev_next_relPractical ways of using these node creation modelsWorking with metadata to improve the contextSummaryExtractorQuestionsAnsweredExtractorTitleExtractorEntityExtractorKeywordExtractorPydanticProgramExtractorMarvinMetadataExtractorDefining your custom extractorIs having all that metadata always a good thing?Estimating the potential cost of using metadata extractorsFollow these simple best practices to minimize your costsEstimate your maximal costs before running the actual extractorsPreserving privacy with metadata extractors, and not onlyScrubbing personal data and other sensitive informationUsing the ingestion pipeline to increase efficiencyHandling documents that contain a mix of text and tabular dataHands-on – ingesting study materials into our PITSSummary
Chapter 5: Indexing with LlamaIndex
Technical requirementsIndexing data – a bird’s-eye viewCommon features of all Index typesUnderstanding the VectorStoreIndexA simple usage example for the VectorStoreIndexUnderstanding embeddingsUnderstanding similarity searchOK, but how does LlamaIndex generate these embeddings?How do I decide which embedding model I should use?Persisting and reusing IndexesUnderstanding the StorageContextThe difference between vector stores and vector databasesExploring other index types in LlamaIndexThe SummaryIndexThe DocumentSummaryIndexThe KeywordTableIndexThe TreeIndexThe KnowledgeGraphIndexBuilding Indexes on top of other Indexes with ComposableGraphHow to use the ComposableGraphA more detailed description of this conceptEstimating the potential cost of building and querying IndexesIndexing our PITS study materials – hands-onSummary
Part 3: Retrieving and Working with Indexed Data

Chapter 6: Querying Our Data, Part 1 – Context Retrieval
Technical requirementsLearning about query mechanics – an overviewUnderstanding the basic retrieversThe VectorStoreIndex retrieversThe DocumentSummaryIndex retrieversThe TreeIndex retrieversThe KnowledgeGraphIndex retrieversCommon characteristics shared by all retrieversEfficient use of retrieval mechanisms – asynchronous operationBuilding more advanced retrieval mechanismsThe naive retrieval methodImplementing metadata filtersUsing selectors for more advanced decision logicUnderstanding toolsTransforming and rewriting queriesCreating more specific sub-queriesUnderstanding the concepts of dense and sparse retrievalDense retrievalSparse retrievalImplementing sparse retrieval in LlamaIndexDiscovering other advanced retrieval methodsSummary
Chapter 7: Querying Our Data, Part 2 – Postprocessing and Response Synthesis
Technical requirementsRe-ranking, transforming, and filtering nodes using postprocessorsExploring how postprocessors filter, transform, and re-rank nodesSimilarityPostprocessorKeywordNodePostprocessorPrevNextNodePostprocessorLongContextReorderPIINodePostprocessor and NERPIINodePostprocessorMetadataReplacementPostprocessorSentenceEmbeddingOptimizerTime-based postprocessorsRe-ranking postprocessorsFinal thoughts about node postprocessorsUnderstanding response synthesizersImplementing output parsing techniquesExtracting structured outputs using output parsersExtracting structured outputs using Pydantic programsBuilding and using query enginesExploring different methods of building query enginesAdvanced uses of the QueryEngine interfaceHands-on – building quizzes in PITSSummary
Chapter 8: Building Chatbots and Agents with LlamaIndex
Technical requirementsUnderstanding chatbots and agentsDiscovering ChatEngineUnderstanding the different chat modesImplementing agentic strategies in our appsBuilding tools and ToolSpec classes for our agentsUnderstanding reasoning loopsOpenAIAgentReActAgentHow do we interact with agents?Enhancing our agents with the help of utility toolsUsing the LLMCompiler agent for more advanced scenariosUsing the low-level Agent Protocol APIHands-on – implementing conversation tracking for PITSSummary
Part 4: Customization, Prompt Engineering, and Final Words
Chapter 9: Customizing and Deploying Our LlamaIndex Project
Technical requirementsCustomizing our RAG componentsHow LLaMA and LLaMA 2 changed the open source landscapeRunning a local LLM using LM StudioRouting between LLMs using services such as Neutrino or OpenRouterWhat about customizing embedding models?Leveraging the Plug and Play convenience of using Llama PacksUsing the Llama CLIUsing advanced tracing and evaluation techniquesTracing our RAG workflows using PhoenixEvaluating our RAG systemIntroduction to deployment with StreamlitHANDS-ON – a step-by-step deployment guideDeploying our PITS project on Streamlit Community CloudSummary
Chapter 10: Prompt Engineering Guidelines and Best Practices
Technical requirementsWhy prompts are your secret weaponUnderstanding how LlamaIndex uses promptsCustomizing default promptsUsing advanced prompting techniques in LlamaIndexThe golden rules of prompt engineeringAccuracy and clarity in expressionDirectivenessContext qualityContext quantityRequired output formatInference costOverall system latencyChoosing the right LLM for the taskCommon methods used for creating effective promptsSummary
Chapter 11: Conclusion and Additional Resources
Other projects and further learningThe LlamaIndex examples collectionMoving forward – Replit bountiesThe power of many – the LlamaIndex communityKey takeaways, final words, and encouragementOn the future of RAG in the larger context of generative AIA small philosophical nugget for you to considerSummary
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 Building Data-Driven Applications with LlamaIndex

9 Customizing and Deploying Our LlamaIndex Project

Customizing Retrieval-Augmented Generation (RAG) components and optimizing performance is critical to building robust, production-ready applications with LlamaIndex. This chapter explores methods for leveraging open source models, intelligent routing across large language models (LLMs), and using community-built modules to increase flexibility and cost-effectiveness. Advanced tracing, evaluation methods, and deployment options are explored to gain deep insight, ensure reliable operation, and streamline the development life cycle.

Throughout this chapter, we’re going to cover the following main topics:

Customizing our RAG components
Using advanced tracing and evaluation techniques
Introduction ...

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

Building AI Applications with ChatGPT APIs

Publisher Resources

ISBN: 9781835089507

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Building Data-Driven Applications with LlamaIndex

by Andrei Gheorghiu

9

Customizing and Deploying Our LlamaIndex Project

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.