Transformers for Natural Language Processing and Computer Vision: Explore Generative AI and Large Language Models with Hugging Face, ChatGPT, GPT-4V, and DALL-E 3

How does the book prepare readers for a career working with Generative AI and Large Language Models?

From page 1, I take the reader into pragmatic approaches to Generative AI and LLMs. The goal is to understand the architecture, potential, and limits of several platforms, such as Hugging Face, Google Vertex AI, and OpenAI, before making a decision. It’s also important to understand the risks of using LLMs. There is a likelihood of losing control of superhuman AI. This book addresses several ways of mitigating risks, such as RAG, embedding-search, knowledge bases, and moderation models.

Moreover, working in this fast-paced, ever-evolving arena means finding the best process, model, and platform for your scenario. The best way is to follow the correct order with a baseline set of tasks. First, try a standard model and go as far as possible with prompt design. Then, try prompt engineering by controlling the input with augmented (RAG) inputs (documents, web scraping, and knowledge base). The next step would be to fine-tune (or build) a model and explore it with prompt engineering and RAG.