Deep Learning Algorithms: In PyTorch (From GPTs to MLPs)

By Ricardo A Calix

This is the Second Edition of my book "Deep Learning Algorithms". Now in PyTorch. It is now 2023 and Deep Learning (DL) has taken off like a rocket. This year some people believed we had reached the singularity. It is definitely an exciting time. Deep Learning techniques are widely used by companies now and the algorithms are doing things that are truly amazing. Generative models are now very popular and whole ecosystems and industries are gravitating towards them. This book will still cover the basics and focuses on building things from scratch. From Scratch today means PyTorch and as such the code examples in this book are now written in PyTorch. The goal of this book is to help you to learn, and to better understand, how to write your own deep learning algorithms from scratch using PyTorch. This is a book for everyone from those starting in deep learning to those with more advanced knowledge. The book starts with basic linear regression and progresses, chapter by chapter, to the more advanced DL algorithms such as CNNs, RNNs, encoders, GANs, Q-Learning, and Transformer based GPTs (to name a few). Additionally, as is necessary with progress, deep learning algorithms are now more complicated, with deeper and more resource intensive networks. This is best exemplified by the Transformer. The Transformer is much more complicated than previous deep learning algorithms. However, I have found that a simple solution to this problem is to write everything using Object Oriented Programming (OOP). This way you are still defining the classes from scratch but you can also build far more complicated algorithms thanks to classical OOP techniques. As such, a GPT can be easily built with a reasonable amount of lines of code if leveraging object oriented techniques. I have taken this very approach through out the whole book. I hope you enjoy the book. I sure have enjoyed learning PyTorch and writing it.

• Format: Paperback
• Language: English
• ISBN: B0CQYLGVYD
• ISBN13: 9798856475783
• Release Date: December 2023
• Publisher: Independently published
• Length: 351 Pages
• Dimensions: 9.3" x 0.8" x 7.5"