[AI] Taulli Tom, Deshmukh Gaurav / Таулли Том, Дешмукх Гаурав - Building Generative AI Agents / Создание генеративных агентов искусственного интеллекта [2025, PDF, ENG]
Главная »
Литература
» Книги FB2 » Учебно-техническая литература
|
| Статистика раздачи | |
| Размер: 5.4 MB | Зарегистрирован: 6 месяца 4 дня | Скачано: 517 раза | |
| Работает мультитрекерная раздача | |
|
Полного источника не было: Никогда |
|
|
| Автор | Сообщение | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MAGNAT ®  ![[Профиль]](./styles/templates/zlight/images/lang/ru/icon_profile.gif "Посмотреть профиль"){kind=icon} ![[ЛС]](./styles/templates/zlight/images/lang/ru/icon_pm.gif "Отправить ЛС"){kind=icon}
|
Building Generative AI Agents: Using LangGraph, AutoGen, and CrewAI / Создание генеративных агентов искусственного интеллекта: Использование LangGraph, AutoGen и CrewAI
Год издания: 2025 Автор: Taulli Tom, Deshmukh Gaurav / Таулли Том, Дешмукх Гаурав Издательство: Apress Media LLC ISBN: 979-8-8688-1134-0 Язык: Английский Формат: PDF Качество: Издательский макет или текст (eBook) Интерактивное оглавление: Да Количество страниц: 278 Описание: The dawn of AI agents is upon us. Tech visionaries like Bill Gates, Andrew Ng, and Vinod Khosla have highlighted the monumental potential of this powerful technology. This book will provide the knowledge and tools necessary to build generative AI agents using the most popular frameworks, such as AutoGen, LangChain, LangGraph, CrewAI, and Haystack. Recent breakthroughs in large language models have opened up unprecedented possibilities. After years of gradual progress in machine learning and deep learning, we are now witnessing novel approaches capable of understanding, reasoning, and generating content in ways that promise to revolutionize nearly every industry. This platform shift is as significant as the advent of mainframes, PCs, cloud computing, mobile technology, and social media. It’s why the world’s largest technology companies – like Microsoft, Apple, Google, and Meta – are making enormous investments in this category. While chatbots like ChatGPT, Claude, and Gemini have demonstrated remarkable potential, the years ahead will see the rise of generative AI agents capable of executing complex tasks on behalf of users. These agents already exhibit capabilities such as running test suites, searching the web for documentation, writing software, answering questions based on vast organized information, and performing intricate web-based tasks across multiple domains. They can autonomously investigate cybersecurity incidents and address complex customer support needs. By integrating skills, knowledge bases, planning frameworks, memory, and feedback loops, these systems can handle many tasks and improve over time. Building Generative AI Agents serves as a high-quality guide for developers to understand when and where AI agents can be useful, their advantages and disadvantages, and practical advice on designing, building, deploying, and monitoring them. What You Will Learn The foundational concepts, capabilities, and potential of AI agents. Recent innovations in large language models that have enabled the development of AI agents. How to build AI agents for launching a product, creating a financial plan, handling customer service, and using Retrieval Augmented Generation (RAG). Essential frameworks for building generative AI agents, including AutoGen, LangChain, LangGraph, CrewAI, and Haystack. Step-by-step guidance on designing, building, and deploying AI agents. Insights into the future of AI agents and their potential impact on various industries. Who This Book Is For Experienced software developers Рассвет агентов с искусственным интеллектом уже близок. Такие технические провидцы, как Билл Гейтс, Эндрю Нг и Винод Хосла, подчеркнули огромный потенциал этой мощной технологии. Эта книга предоставит знания и инструменты, необходимые для создания генеративных агентов искусственного интеллекта с использованием самых популярных фреймворков, таких как AutoGen, LangChain, LangGraph, CrewAI и Haystack. Недавние прорывы в области больших языковых моделей открыли беспрецедентные возможности. После многих лет постепенного прогресса в области машинного обучения и глубокого обучения в настоящее время мы наблюдаем новые подходы, позволяющие понимать, логически обосновывать и генерировать контент способами, которые обещают революционизировать практически все отрасли. Этот сдвиг платформы так же важен, как и появление мэйнфреймов, персональных компьютеров, облачных вычислений, мобильных технологий и социальных сетей. Вот почему крупнейшие технологические компании мира, такие как Microsoft, Apple, Google и Meta, вкладывают огромные средства в эту категорию. В то время как чат-боты, такие как ChatGPT, Claude и Gemini, продемонстрировали значительный потенциал, в ближайшие годы ожидается появление агентов с генеративным ИИ, способных выполнять сложные задачи от имени пользователей. Эти агенты уже обладают такими возможностями, как запуск наборов тестов, поиск документации в Интернете, написание программного обеспечения, ответы на вопросы на основе обширной систематизированной информации и выполнение сложных веб-задач в нескольких доменах. Они могут автономно расследовать инциденты, связанные с кибербезопасностью, и решать сложные задачи поддержки клиентов. Благодаря интеграции навыков, баз знаний, систем планирования, памяти и контуров обратной связи эти системы могут справляться со многими задачами и со временем совершенствоваться. Создание генеративных агентов искусственного интеллекта служит высококачественным руководством для разработчиков, помогающим понять, когда и где агенты искусственного интеллекта могут быть полезны, их преимущества и недостатки, а также практические советы по их проектированию, созданию, развертыванию и мониторингу. Что Вы узнаете Основные концепции, возможности и потенциал ИИ-агентов. Последние инновации в больших языковых моделях, которые позволили разработать ИИ-агенты. Как создать ИИ-агентов для запуска продукта, составления финансового плана, обслуживания клиентов и использования расширенной генерации результатов поиска (RAG). Основные платформы для создания генерирующих ИИ-агентов, включая AutoGen, LangChain, LangGraph, CrewAI и Haystack. Пошаговое руководство по проектированию, созданию и внедрению ИИ-агентов. Информация о будущем ИИ-агентов и их потенциальном влиянии на различные отрасли промышленности. Для кого предназначена эта книга Опытные разработчики программного обеспечения ОглавлениеAbout the Authors ................................................................................................ixChapter 1: Introduction to AI Agents ......................................................................1 What Are AI Agents? ............................................................................................4 Reflection ...........................................................................................................5 Tools ..................................................................................................................6 Memory ..............................................................................................................7 Planning .............................................................................................................9 Multi-agent Collaboration ......................................................................................10 Autonomy ...........................................................................................................11 UI and UX ...........................................................................................................12 New Approaches to Development ...........................................................................14 Flavors of AI Agents .............................................................................................16 Brief History ........................................................................................................17 LLMs, Copilots, and RPA ........................................................................................19 Use Cases ............................................................................................................21 Sierra ..................................................................................................................22 Enso ....................................................................................................................23 Asana ..................................................................................................................24 Conclusion ..........................................................................................................25 Chapter 2: Generative AI Foundations .....................................................................27 Pretrained Models ...............................................................................................28 Transformer Models ............................................................................................29 Transfer Learning ................................................................................................31 Alignment in Language Models ..............................................................................32 Multimodal LLMs .................................................................................................33 Types of Models ..................................................................................................34 Proprietary LLMs .................................................................................................35 Open Source LLMs and SLMs ................................................................................38 Prompt Engineering ............................................................................................41 Be Clear .........................................................................................................42 Details ...........................................................................................................42 Persona .........................................................................................................42 Use Delimiters ...............................................................................................43 Steps for a Task .............................................................................................43 Time to Think ................................................................................................44 Length of Output ............................................................................................45 Going Beyond the Transformer ............................................................................45 Conclusion ........................................................................................................46 Chapter 3: Types of Agents ..................................................................................47 Simple Reflex Agents ..........................................................................................48 Model-Based Reflex Agents ..................................................................................49 Goal-Based Agents ..............................................................................................50 Utility-Based Agents ............................................................................................51 Learning Agents ..................................................................................................52 Hierarchical Agents .............................................................................................53 Conclusion ..........................................................................................................55 Chapter 4: OpenAI GPTs and the Assistants API ....................................................57 Registering for the OpenAI API Key .....................................................................57 GPTs ..............................................................................................................58 Pricing and Tokens ...........................................................................................61 OpenAI API .....................................................................................................64 Assistants API ................................................................................................66 Playground ....................................................................................................68 Assistants API ................................................................................................72 Recent Advancements .....................................................................................76 Conclusion .....................................................................................................79 Chapter 5: Developing Agents ..........................................................................81 Jupyter Notebook, VS Code, and Google Colab ...................................................82 Jupyter Notebook ..........................................................................................82 Visual Studio Code (VS Code) ..........................................................................82 Google Colab ..................................................................................................83 How to Use Jupyter Notebooks ...........................................................................83 Google Colab .......................................................................................................86 Streamlit, Gradio, and Jupyter Widgets ...................................................................89 Hugging Face ......................................................................................................90 Languages ..........................................................................................................92 Using LLMs (Large Language Models) ...................................................................93 Using an API from an LLM Provider ......................................................................93 Using a Service like Ollama ................................................................................94 Using a Cloud Service like Azure, Google Cloud, or AWS .........................................95 Setting Up and Using Ollama ..............................................................................96 Using Ollama with Google Colab ..........................................................................97 Customizing LLMs ...............................................................................................98 Fine-Tuning .......................................................................................................98 Retrieval-Augmented Generation (RAG) ................................................................100 Conclusion ........................................................................................................101 Chapter 6: CrewAI ..............................................................................................103 The Basics .........................................................................................................104 Agents ...............................................................................................................104 Tasks .................................................................................................................107 Tools ..................................................................................................................109 Crews ................................................................................................................111 Processes ..........................................................................................................112 Memory .............................................................................................................113 Financial Planning Agent ..................................................................................115 Product Launch Orchestrator ............................................................................123 Customer Call Center Processing ......................................................................130 Retrieval-Augmented Generation (RAG) ............................................................141 Connecting LLMs ..............................................................................................143 Conclusion ........................................................................................................145 Chapter 7: AutoGen ..........................................................................................147 ConversableAgent .............................................................................................148 Reflection Agent ................................................................................................150 Tool Use .............................................................................................................157 Group Chat ........................................................................................................162 Web Search Agent .............................................................................................165 Retrieval-Augmented Generation (RAG) ................................................................167 Using Ollama .....................................................................................................170 AutoGen Studio .................................................................................................171 Conclusion ........................................................................................................177 Chapter 8: LangChain .........................................................................................179 Background .......................................................................................................180 The Components ...............................................................................................181 Models ..............................................................................................................182 Prompt Templates .............................................................................................184 Output Parsers ..................................................................................................185 Document Loaders ............................................................................................188 Text Splitters .....................................................................................................190 Memory .............................................................................................................193 Key Concepts of LangChain Agents .......................................................................198 Types of Agents .................................................................................................199 Tool Calling Agent ........................................................................................199 XML Agent ...................................................................................................200 JSON Chat Agent .........................................................................................200 Structured Chat Agent .................................................................................200 Self-Ask with Search Agent ...........................................................................201 ReAct Agent ......................................................................................................201 Agent Program ..................................................................................................202 Conclusion ........................................................................................................208 Chapter 9: Introduction to LangGraph ...............................................................209 Benefits of Combining LangChain with LangGraph ...............................................210 Pros and Cons of LangGraph .............................................................................212 Graphs .........................................................................................................214 State ............................................................................................................215 Nodes ..........................................................................................................216 Edges ...........................................................................................................218 Reflection Agent ..........................................................................................219 Persistence ..................................................................................................225 LangSmith ...................................................................................................230 Assistant-UI .................................................................................................232 LangGraph Studio ........................................................................................234 Conclusion ...................................................................................................235 Chapter 10: Haystack ....................................................................................237 Haystack Program .........................................................................................238 Haystack Agent with Function Calling ................................................................242 Conclusion .....................................................................................................249 Chapter 11: Takeaways ....................................................................................251 Rethinking Software .........................................................................................253 The Challenges .................................................................................................255 AI Agent Frameworks ........................................................................................256 Conclusion ........................................................................................................260 Glossary ...........................................................................................................261 Index ...............................................................................................................267
|
|||||||||||||||||||
![[Код]](./styles/templates/zlight/images/lang/ru/icon_code.gif "Код"){kind=icon}
|
Главная »
Литература
» Книги FB2 » Учебно-техническая литература
|
Текущее время: 05-Дек 14:48
Часовой пояс: UTC + 5
Вы не можете начинать темы
Вы не можете отвечать на сообщения Вы не можете редактировать свои сообщения Вы не можете удалять свои сообщения Вы не можете голосовать в опросах Вы не можете прикреплять файлы к сообщениям Вы можете скачивать файлы |
