دانلود کتاب پیش بینی سری های زمانی با استفاده از هوش مصنوعی مولد – استفاده از هوش مصنوعی برای پیش بینی دقیق

“Guru Brahma, Guru Vishnu, Guru Devo Maheshwara, Guru Sakshat Parabrahma, Tasmai Shri Gurave Namah” – a disciple expressing gratitude and reverence toward their guru (teacher). Grateful to my gurus who guided and supported me in the form of teachers and friends. A couple of years back while working on a project related to time series, we wanted to explore newer techniques in forecasting to improve precision. The advent of GenAI provides us with an opportunity to explore LLM-based models for forecasting. However, there was not enough material to help the team come up to speed. The research papers were difficult to understand for the team who came from diverse levels of mathematical backgrounds, so we had to go through a steep learning curve. We were looking for a resource that would equip us with the theoretical understanding of the models and practical implementation with python sample code. We could not find any, so that gave birth to the idea of writing this book. We present this book that is catered to the needs of working professionals to come up to speed. Those who wish to dive deeper may want to read the reference papers after reading this book. This book is primarily targeted toward intermediate to advanced time series forecasting modelers. So if you are a beginner, we suggest you to pick up a beginner-friendly book like Hands-on Time Series Analysis with Python by Vishwas and Ashish before reading this book. Researchers are suggested to read the provided references after going through this book. The book starts with a motivation to learn time series forecasting. Chapter 1 introduces different time series techniques, generative AI, large language models, evolution, and milestones to date. Chapters 2 and 3 discuss neural networks and transformer theory and implementation. You can use these chapters to refresh your knowledge and learn to implement them by leveraging modern tools. Chapters 4– 8 cover topics related to foundation models for time series forecasting. Each chapter discusses a new foundation model. We begin by understanding the technical overview, relevant concepts, and implementation using Python code and libraries. Techniques that help to understand forecasting by repurposing and reusing foundation models meant for NLP are explained. All chapters (except Chapter 1) discuss how to implement the models with a dataset and full code with explanation. Where possible and applicable, we try to implement the models for both univariate and multivariate scenarios.