Ganguly Arindam / Гангули Ариндам - Scaling Enterprise Solutions with Large Language Models / Масштабирование корпоративных решений с помощью больших языковых моделей [2025, PDF/EPUB, ENG]
Главная »
Литература
» Книги FB2 » Учебно-техническая литература
|
| Статистика раздачи | |
| Размер: 15.29 MB | Зарегистрирован: 6 месяца 4 дня | Скачано: 88 раза | |
| Список скачавших: Нет | |
| 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}
|
Scaling Enterprise Solutions with Large Language Models: Comprehensive End-to-End Generative AI Solutions for Production-Grade Enterprise Solutions / Масштабирование корпоративных решений с помощью больших языковых моделей: Всесторонние комплексные решения на основе генеративного ИИ для корпоративных решений производственного уровня
Год издания: 2025 Автор: Ganguly Arindam / Гангули Ариндам Издательство: Apress Media LLC ISBN: 979-8-8688-1154-8 Язык: Английский Формат: PDF/EPUB Качество: Издательский макет или текст (eBook) Интерактивное оглавление: Да Количество страниц: 458 Описание: Artificial Intelligence (AI) is the bedrock of today’s applications, propelling the field towards Artificial General Intelligence (AGI). Despite this advancement, integrating such breakthroughs into large-scale production-grade enterprise applications presents significant challenges. This book addresses these hurdles in the domain of large language models within enterprise solutions. By leveraging Big Data engineering and popular data cataloguing tools, you’ll see how to transform challenges into opportunities, emphasizing data reuse for multiple AI models across diverse domains. You’ll gain insights into large language model behavior by using tools such as LangChain and LLamaIndex to segment vast datasets intelligently. Practical considerations take precedence, guiding you on effective AI Governance and data security, especially in data-sensitive industries like banking. This enterprise-focused book takes a pragmatic approach, ensuring large language models align with broader enterprise goals. From data gathering to deployment, it emphasizes the use of low code AI workflow tools for efficiency. Addressing the challenges of handling large volumes of data, the book provides insights into constructing robust Big Data pipelines tailored for Generative AI applications. Scaling Enterprise Solutions with Large Language Models will lead you through the Generative AI application lifecycle and provide the practical knowledge to deploy efficient Generative AI solutions for your business. What You Will Learn Examine the various phases of an AI Enterprise Applications implementation. Turn from AI engineer or Data Science to an Intelligent Enterprise Architect. Explore the seamless integration of AI in Big Data Pipelines. Manage pivotal elements surrounding model development, ensuring a comprehensive understanding of the complete application lifecycle. Plan and implement end-to-end large-scale enterprise AI applications with confidence. Who This Book Is For Enterprise Architects, Technical Architects, Project Managers and Senior Developers. Искусственный интеллект (ИИ) является основой современных приложений, продвигая отрасль к созданию искусственного интеллекта общего назначения (AGI). Несмотря на этот прогресс, интеграция таких достижений в крупномасштабные корпоративные приложения производственного уровня сопряжена со значительными трудностями. В этой книге рассматриваются эти препятствия в области больших языковых моделей в корпоративных решениях. Используя технологию обработки больших данных и популярные инструменты каталогизации данных, вы увидите, как превратить проблемы в возможности, уделяя особое внимание повторному использованию данных для нескольких моделей искусственного интеллекта в различных областях. Вы получите представление о поведении больших языковых моделей, используя такие инструменты, как LangChain и LLamaIndex, для разумного сегментирования огромных наборов данных. Практические соображения имеют первостепенное значение и помогут вам эффективно управлять ИИ и обеспечивать безопасность данных, особенно в таких отраслях, как банковское дело, где важна информация. В этой книге, ориентированной на предприятия, используется прагматичный подход, обеспечивающий соответствие больших языковых моделей более широким корпоративным целям. От сбора данных до развертывания, в ней делается упор на использование инструментов искусственного интеллекта с низким количеством кода для повышения эффективности рабочего процесса. В книге рассматриваются проблемы, связанные с обработкой больших объемов данных, и дается представление о построении надежных конвейеров обработки больших данных, адаптированных для приложений с генеративным ИИ. Масштабирование корпоративных решений с помощью больших языковых моделей поможет вам пройти жизненный цикл приложений с генерирующим ИИ и даст практические знания для внедрения эффективных решений с генерирующим ИИ для вашего бизнеса. Что вы узнаете Изучите различные этапы внедрения корпоративных приложений с ИИ. Превратитесь из инженера по ИИ или специалисту по обработке данных в интеллектуального архитектора предприятия. Изучите возможности бесшовной интеграции ИИ в конвейеры больших данных. Управляйте ключевыми элементами, связанными с разработкой моделей, обеспечивая всестороннее понимание всего жизненного цикла приложения. Уверенно планируйте и внедряйте комплексные крупномасштабные корпоративные приложения искусственного интеллекта. Для кого предназначена эта книга Архитекторы предприятий, технические архитекторы, руководители проектов и старшие разработчики. ОглавлениеAbout the Author ..........................................................................................xiiiAbout the Technical Reviewer ..........................................................................xv Acknowledgments ..........................................................................................xvii Introduction ..................................................................................................xix Chapter 1: Machine Learning Primer .................................................................1 The Origins of Machine Learning .......................................................................1 Linear Regression ...........................................................................................3 Decision Tree .................................................................................................6 Ensemble Methods ..........................................................................................9 The Case of the Late Night Burglar ...................................................................10 Voting Classifier .............................................................................................10 Bagging and Pasting ......................................................................................12 Random Forest ..............................................................................................15 Boosting ........................................................................................................16 Stacking .........................................................................................................18 Metrics ...........................................................................................................19 Accuracy ........................................................................................................20 Precision ........................................................................................................21 Recall .............................................................................................................22 Confusion Matrix ...........................................................................................22 ROC AUC ........................................................................................................25 Mean Squared Error ......................................................................................25 Deep Learning .................................................................................................27 Sigmoid Neuron .............................................................................................29 Problems with Sigmoid Neuron .....................................................................31 Tanh ...............................................................................................................31 Vanishing Gradient Problem ..........................................................................32 ReLU ..............................................................................................................33 Leaky ReLU ....................................................................................................34 TensorFlow and Keras .........................................................................................36 Optimizers .....................................................................................................38 Unsupervised Learning .......................................................................................43 K-Means Clustering Algorithm .......................................................................44 Associative Rule Mining ................................................................................46 Dimensionality Reduction ..............................................................................47 Summary .........................................................................................................47 Chapter 2: Natural Language Processing Primer .....................................................49 Steps for an NLP Task .........................................................................................50 Data Gathering ...............................................................................................51 NLTK and Spacy .............................................................................................55 Cleaning Data ................................................................................................59 Tokenization ..................................................................................................67 Vectorization and Embedding ........................................................................69 Model Selection, Training, and Evaluation ........................................................73 Deep Learning in Natural Language Processing .................................................74 Pretrained Embeddings .................................................................................79 Summary .......................................................................................................82 Chapter 3: RNN to Transformer and BERT ...........................................................83 Sequence Modeling .........................................................................................84 Recurrent Neural Networks ...........................................................................85 Problems with Vanilla RNN ............................................................................88 Attention ........................................................................................................95 Encoder-Decoder Models ..............................................................................99 Self-Attention ..............................................................................................100 Transformers ...............................................................................................102 BERT ............................................................................................................112 HuggingFace Transformers ..........................................................................114 Summary .....................................................................................................127 Chapter 4: Large Language Models ...................................................................129 Language Models (LLMs) ..................................................................................130 Masked Language Modeling .........................................................................130 Sequence-to-Sequence Models ....................................................................132 Autoregressive Models ................................................................................132 GPT ..............................................................................................................133 Reinforcement Learning ...................................................................................135 OpenAI Gym ......................................................................................................137 Reinforcement Learning Through Human Feedback ..................................................139 Instruct GPT ......................................................................................................140 OpenAI ...............................................................................................................142 Prompting ...........................................................................................................144 OpenAI API .........................................................................................................150 Create Your API Key .....................................................................................150 Setting Up Postman .....................................................................................151 Handling Rate Limits ...................................................................................155 LLM API Best Practices ................................................................................156 Common Issues ...........................................................................................156 The IT Assistant ...........................................................................................157 Preparing the Database ...............................................................................159 Preparing the Backend and Orchestration Layer ..............................................161 Creating a Python File .................................................................................162 Creating Microservices ................................................................................166 Mission Accomplished ......................................................................................178 Summary ........................................................................................................181 Chapter 5: Retrieval Augmented Generation ........................................................183 Prompt Engineering ..........................................................................................184 Chain of Thought Prompting ........................................................................185 Vector Databases .........................................................................................188 LangChain ....................................................................................................197 Building Your First RAG Application ..................................................................204 Summary .....................................................................................................213 Chapter 6: LLM Evaluation and Optimization .......................................................215 The Need for LLM Evaluation ............................................................................216 LangGraph .........................................................................................................217 Hallucinations ...................................................................................................218 LLM as a Judge .................................................................................................219 Corrective RAG .............................................................................................220 Benchmarking ...................................................................................................231 MLFlow .............................................................................................................232 MLFlow for Scikit-Learn Models ......................................................................233 The Complete Intelligent Application with MLFlow Tracker ..................................239 Dockerfiles ..................................................................................................249 Tracking LLM and Generative AI Applications ...................................................257 Preparing Custom Generative AI Evaluation Metrics Using MLFlow ......................265 Portkey ..............................................................................................................269 Creating an Account ....................................................................................269 Using Portkey in Your Code ..........................................................................273 Load Balancing ............................................................................................279 Caching ........................................................................................................280 vLLM .................................................................................................................281 Prerequisites ...............................................................................................282 Steps to Install .............................................................................................282 Summary ..........................................................................................................284 Chapter 7: AI Governance and Responsible AI ........................................................285 AI Fairness ........................................................................................................286 Explainable AI ..............................................................................................288 Drift ...................................................................................................................294 Model Drift ...................................................................................................294 Data Drift .....................................................................................................295 Drift Detection .............................................................................................296 AI Regulations ................................................................................................299 LLM and Prompt Governance ............................................................................302 Langfuse ......................................................................................................303 Prompt Governance Using Langfuse ..................................................................310 Summary ..........................................................................................................316 Chapter 8: Adding Intelligence to Large Enterprise Applications ................................317 A Typical Chatbot ...............................................................................................318 The Need for AI Architecture .............................................................................320 Experimentation Environment ...........................................................................322 The Intelligent IT Assistant ................................................................................323 The Enterprise CRM ..........................................................................................325 Setting Up HubSpot ..........................................................................................325 Setting Up HubSpot Private App for REST API integration .......................................327 Setting Up the Knowledge Repository ..................................................................333 Agents ............................................................................................................334 Building the Bot ................................................................................................339 Setting Up the Vector Database ...........................................................................340 Developing Agents in LangChain ..........................................................................341 Summary ..........................................................................................................354 Chapter 9: Data Pipelines in Generative AI ............................................................357 A Closer Look at Data ........................................................................................358 File Formats ......................................................................................................359 JSON ............................................................................................................359 CSV ..............................................................................................................359 XML .............................................................................................................359 Avro and Parquet ..........................................................................................360 Data Models and Data Storage .........................................................................361 Data Processing Systems .................................................................................361 The Data-Intensive AI Assistant ........................................................................362 Setting up MinIO ..........................................................................................365 Upload File Application ................................................................................367 RAG from an S3 Bucket ...............................................................................374 Apache Kafka for Streaming ........................................................................383 Using Data Pipelines in AI Assistant ............................................................388 Summary .................................................................................................401 Chapter 10: Putting It All Together ..............................................................403 Option 1: Minimizing Cost while Maximum Efficiency ......................................404 Determining Optimal Intelligence ................................................................404 Small Language Models ...............................................................................409 Phi 3.5 .......................................................................................................410 Option 2: Getting the Best Performance with the Same Cost ..............................413 Fine-Tuning Large Language Models ................................................................413 Parameter Efficient Fine Tuning (PEFT) ............................................................414 Low Rank Adaptation (LoRA) ..........................................................................415 Implementing PEFT LoRA in Python .................................................................416 Long Context LLM and RAG ............................................................................422 Self-Routing .................................................................................................423 Summary ....................................................................................................424 Index .........................................................................................................427
|
|||||||||||||||||||
![[Код]](./styles/templates/zlight/images/lang/ru/icon_code.gif "Код"){kind=icon}
|
Главная »
Литература
» Книги FB2 » Учебно-техническая литература
|
Текущее время: 05-Дек 16:21
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 |
