Big announcement - Revealing my new book

Key Takeaways

hey it's been a while since I've posted a video on YouTube that is because I have been very busy with my new project uh I am writing a second book called time series forecasting using Foundation models uh and yes it is my second book the first one being time series forecasting in Python if you haven't read it uh make sure to grab a copy I will leave a link in the description below so uh as you know traditionally right we train models for a specific data set or scenario all right so if we need to forecast another series now we have to train another model now of course that takes time and requires knowledge and expertise uh to build those forecasting models with Foundation models this changes completely because now we have access to huge models train on billions of data points and capable of zero shot forecasting meaning that all we have to do is load a model feed our series and we get predictions right away so you can already see how using them can really speed up the entire forecast ing process since now we don't need to train a unique model for each series or scenario uh also it lowers the barrier of Entry to forecasting because now all of the heavy lifting was already done for us and it is really useful when you're forecasting a series that doesn't have enough data um to to train a model from scratch right so think of a situation where uh maybe you're trying to predict the sale of a new product right but you haven't only been selling that product for a very few days or very few months so none enough needa to build a model from scratch then using your foundation model it can help make some more reasonable forecasting uh than you know training it from scratch so this is why I decided to write this book so after reading it you will have all of the knowledge uh required to use Foundation models today uh but also to understand and implement the future work that will be done in the field and the book starts with the basics okay so we explore the concept of foundation models and we discover their advantages and dis advantages uh to make sure that you make the right decision for your projects and then we build our own tiny Foundation model uh mainly to experiment with the fundamental concepts like pre-training transfer learning and fine-tuning uh it also makes us appreciate how hard it is to actually build an actual Foundation model and then we start using large time models so Foundation models specifically built for time series forecasting uh we study their architecture their capabilities and we apply them in a Hands-On projects using Python and after that we experiment with large language models um because there is work that has been done to adapt uh these llms to forecasting tasks and finally we conclude the book with a Capstone project uh so that we can apply our learnings and compare Foundation models to the more traditional data specific methods so the book is currently in Early Access manning.com so at the time of making this video there are three chapters uh that are released already and uh so you can buy the book now at a very nice discount and you will get access to all future chapters and of course if you have any questions feel free to uh connect with me on LinkedIn on medium of course you're watching this on YouTube so uh feel free to drop a comment if you have any questions or you can go on my own website data science withth mar.com enjoy the read