ChatGPT for Conversational AI and Chatbots Learn How to Automate Conversations with the Latest Large Language Model Technologies
Explore ChatGPT technologies to create state-of-the-art chatbots and voice assistants, and prepare to lead the AI revolution Key Features Learn how to leverage ChatGPT to create innovative conversational AI solutions for your organization Harness LangChain and delve into step-by-step LLM application...
| Otros Autores: | |
|---|---|
| Formato: | Libro electrónico |
| Idioma: | Inglés |
| Publicado: |
Birmingham, England :
Packt Publishing
[2024]
|
| Edición: | First edition |
| Materias: | |
| Ver en Biblioteca Universitat Ramon Llull: | https://discovery.url.edu/permalink/34CSUC_URL/1im36ta/alma991009841738206719 |
Tabla de Contenidos:
- Cover
- Title Page
- Copyright and credits
- Contributors
- Table of Contents
- Preface
- Part 1: Foundations of Conversational AI
- Chapter 1: An Introduction to Chatbots, Conversational AI, and ChatGPT
- What are chatbots and conversational AI?
- A brief history of conversational AI
- An overview of chatbots and automated assistants
- Evolution of chatbots and conversational AI
- Understanding conversational AI applications
- Customer service
- Language translation
- Education
- Healthcare
- Banking and insurance
- Retail
- Human resources
- Conversational AI as a training tool - the emergence of digital humans
- Conclusion
- What is OpenAI's ChatGPT?
- Understanding the large language technology behind ChatGPT
- The role of tokens in LMs
- Understanding OpenAI's LMs
- Conclusion
- Capabilities and applications of ChatGPT
- Capabilities of ChatGPT
- How smart is ChatGPT?
- Applications of ChatGPT
- Limitations of ChatGPT
- Limitations
- Risks and security
- Summary
- Further reading
- Chapter 2: Using ChatGPT with Conversation Design
- Technical requirements
- Understanding conversation design
- Exploring the role of conversation designers
- Working with practical applications of ChatGPT in conversation design
- Intent clustering with ChatGPT
- Understanding utterance and entity generation
- Using ChatGPT to help write your dialogue
- Persona creation with ChatGPT
- Using ChatGPT for user research
- Creating our user and chatbot personas
- Simulating conversations
- What is a sample dialogue?
- Creating sample dialogue
- Creating sample dialogue with personas
- Testing and iteration in conversation design with ChatGPT
- Testing with ChatGPT
- Summary
- Further reading
- Part 2: Using ChatGPT, Prompt Engineering, and Exploring LangChain.
- Chapter 3: ChatGPT Mastery - Unlocking Its Full Potential
- Technical requirements
- Mastering the ChatGPT interface
- The Free and Plus versions of ChatGPT
- ChatGPT interface
- GPTs
- Exploring OpenAI Playground
- Getting started
- UI features
- Pricing for API and Playground
- Learning to use the ChatGPT API
- Getting started
- Calling the API directly
- Setting up with the OpenAI Python library
- Setting up with the OpenAI Node.js library
- Other ChatGPT libraries
- Summary
- Further reading
- Chapter 4: Prompt Engineering with ChatGPT
- Technical requirements
- Going through the concepts of prompt engineering
- Understanding the core components of a successful prompt
- Instruction
- Context
- Scope
- Role
- Audience
- Input data
- Output data
- Conclusion
- Working with a prompt engineering strategy
- Define clear goals
- Employ iterative prompt development
- Start simple
- Use follow-up prompts to test against multiple examples
- Use temperature when you need to
- Handling memory limitations in ChatGPT
- Conclusion
- Knowing the prompt engineering techniques
- Few-shot learning for a customer support chatbot
- Prompting to summarize data for a conversational agent
- Prompting to create your own chatbot powered by ChatGPT
- Summary
- Further reading
- Chapter 5: Getting Started with LangChain
- Technical requirements
- Introduction to LangChain
- LangChain libraries
- Core components of LangChain
- Working with LLMs in LangChain
- Prompt templates
- Using output parsers
- Understanding LangChain Expression Language
- What is LCEL?
- Key components of LCEL
- Runnable protocol
- A simple example of LCEL
- Creating different LangChain chains
- Basic chain example
- Creating a sequential chain to investigate conversational data.
- Utilizing parallel chains in LangChain for efficient multi-source information gathering
- Routing chains to answer questions effectively
- Summary
- Further reading
- Chapter 6: Advanced Debugging, Monitoring, and Retrieval with LangChain
- Technical requirements
- Debugging and monitoring LangChain
- Understanding tracing techniques
- Introducing LangSmith
- Leveraging LangChain agents
- What is an agent?
- What are LangChain tools?
- An introduction to OpenAI tool calling
- Plug-and-play LangChain tools for immediate integration
- An out-of-the-box tool example
- Using our tool with an agent
- Creating a custom weather tool
- Exploring LangChain memory
- Exploring the different types of memory applications
- Understanding memory challenges
- Introducing memory usage techniques
- Understanding an example of using memory in LangChain
- Summary
- Further reading
- Part 3: Building and Enhancing ChatGPT-Powered Applications
- Chapter 7: Vector Stores as Knowledge Bases for Retrieval-augmented Generation
- Technical requirements
- Why do we need RAG?
- Understanding the steps needed to create a RAG system
- Defining your RAG data sources
- Preprocessing our content and generating embeddings
- Chunking for effective LLM interactions
- Creating embeddings with OpenAI models
- Storing and searching our embeddings with a vector store
- Deciding which vector database to use
- Working through a RAG example with LangChain
- Integrating data - choosing your document loader
- Creating manageable chunks with text splitting
- Creating and storing text embeddings
- Bringing everything together with LangChain
- Summary
- Further reading
- Chapter 8: Creating Your Own LangChain Chatbot Example
- Technical requirements
- Scoping our ChatGPT project
- A holiday assistant use case.
- A persona outline for Ellie the explorer
- Ellie's conversational scope
- Technical features
- Getting our data ready for the chatbot
- Selecting our data sources
- Preparing the hotel data
- Creating our agent for complex interactions
- Creating the agent tools
- Bringing it all together - building Your own LangChain Chatbot with Streamlit
- Creating secrets and config management
- Creating our agent service
- Building our Streamlit chat app
- Running and testing Ellie, our chatbot application
- Ways to improve Ellie
- Summary
- Further reading
- Chapter 9: The Future of Conversational AI with LLMs
- Technical requirements
- Going into production
- Understanding the dangers of going into production
- Challenges of RAG systems
- Evaluating production systems
- Components of an evaluation system
- Learning how to use LangSmith to evaluate your project
- Application monitoring in production with LangSmith
- Advanced monitoring features with LangSmith
- Tracing and data management
- Monitoring tools
- Advanced monitoring features
- Alternatives to ChatGPT and LangChain
- Some alternatives to ChatGPT and OpenAI LLMs
- Some alternatives to LangChain
- Looking at the growing LLM landscape
- The growth of the small language model (SLM)
- Are LLMs reaching their limit?
- Enter SLMs
- SLMs versus LLMs - key differences
- Advantages of SLMs
- Disadvantages of SLMs
- Some examples of SLMs
- The transformative potential of SLMs
- Where to go from here
- Summary
- Further reading
- Index
- Other Books You May Enjoy.
