Enterprise AI is moving beyond text to systems that can understand images, documents, video, and language together. Vision-language models make this possible, but most real-world problems are not solved by models alone. They require well-designed pipelines, data flows, and integration with business systems. This book focuses on building those systems end-to-end. It covers the foundations of visual and language representations and moves into modern multimodal models. From there, it discusses visual reasoning, visual question answering, and multimodal retrieval. You will work with real code using frameworks such as PyTorch, Hugging Face, OpenCV, YOLO, and learn how to combine vision encoders and retrieval layers into a single workflow. As the book progresses, it focuses on full system design. You will build pipelines that handle data ingestion, pre-processing, model inference, and output generation. The book also covers practical aspects such as model optimization, quantization, and efficient deployment to reduce latency and cost. It explains how to integrate these components into enterprise use cases, evaluate performance, and make systems reliable and scalable. By the end of this book, you will be able to design and implement production-ready vision-language solutions that solve real business problems. WHAT YOU WILL LEARN ¿ Understand core concepts behind vision and language representations. ¿ Implement custom self-attention and cross-attention in PyTorch. ¿ Evaluate image captions and VQA benchmarks using CIDEr and CLIPScore. ¿ Develop visual question answering and multimodal retrieval systems. ¿ Apply model optimization techniques including quantization for efficiency. ¿ Design scalable and reliable enterprise-ready AI system architectures. WHO THIS BOOK IS FOR This book is intended for machine learning engineers, data scientists, AI engineers, and software developers who want to build multimodal AI systems using vision and language models. It is also relevant for architects and technical leaders designing enterprise AI solutions. Readers are expected to have a basic understanding of Python and machine learning concepts. Familiarity with deep learning frameworks will help in implementing the examples and understanding system design decisions across real-world applications.
Die Inhaltsangabe kann sich auf eine andere Ausgabe dieses Titels beziehen.
Shivashankar Chari is an AI and data science leader with over 18 years of experience in building and scaling enterprise AI systems across industries, including healthcare, insurance, manufacturing, people analytics, and consumer goods. He has led large cross-functional teams and delivered production-grade machine learning and generative AI solutions focused on real-world business outcomes. His work spans computer vision, natural language processing, and multimodal AI systems, with a strong emphasis on enterprise architecture, scalability, and deployment. He has designed AI platforms, reusable solution frameworks, and end-to-end pipelines that integrate vision and language for decision intelligence. His recent work also focuses on agent-based system design, including multi-agent orchestration and Model Context Protocol-driven architectures for building adaptive and modular AI systems. Shivashankar is a doctoral candidate in AI. He has also been invited to speak as an AI visionary leader at industry forums, sharing perspectives on enterprise AI adoption and system design.
„Über diesen Titel“ kann sich auf eine andere Ausgabe dieses Titels beziehen.
Anbieter: California Books, Miami, FL, USA
Zustand: New. Artikel-Nr. I-9788169619127
Anzahl: Mehr als 20 verfügbar
Anbieter: Home-set, Mannheim, Deutschland
Softcover/Taschenbuch. Zustand: Neu. Enterprise AI is moving beyond text to systems that can understand images, documents, video, and language together. Vision-language models make this possible, but most real-world problems are not solved by models alone. They require well-designed pipelines, data flows, and integration with business systems.This book focuses on building those systems end-to-end. It covers the foundations of visual and language representations and moves into modern multimodal models. From there, it discusses visual reasoning, visual question answering, and multimodal retrieval. You will work with real code using frameworks such as PyTorch, Hugging Face, OpenCV, YOLO, and learn how to combine vision encoders and retrieval layers into a single workflow. As the book progresses, it focuses on full system design. You will build pipelines that handle data ingestion, pre-processing, model inference, and output generation. The book also covers practical aspects such as model optimization, quantization, and efficient deployment to reduce latency and cost. It explains how to integrate these components into enterprise use cases, evaluate performance, and make systems reliable and scalable.By the end of this book, you will be able to design and implement production-ready vision-language solutions that solve real business problems.WHAT YOU WILL LEARN? Understand core concepts behind vision and language representations.? Implement custom self-attention and cross-attention in PyTorch. ? Evaluate image captions and VQA benchmarks using CIDEr and CLIPScore. ? Develop visual question answering and multimodal retrieval systems.? Apply model optimization techniques including quantization for efficiency.? Design scalable and reliable enterprise-ready AI system architectures.WHO THIS BOOK IS FORThis book is intended for machine learning engineers, data scientists, AI engineers, and software developers who want to build multimodal AI systems using vision and language models. It is also relevant for architects and technical leaders designing enterprise AI solutions. Readers are expected to have a basic understanding of Python and machine learning concepts. Familiarity with deep learning frameworks will help in implementing the examples and understanding system design decisions across real-world applications. Sprache: Englisch Gewicht in Gramm: 751. Artikel-Nr. IB-10824728
Anzahl: 2 verfügbar