What this book covers
Chapter 1, Machine Learning – an Introduction, discusses the basic machine learning paradigms. It will explore various machine learning algorithms and introduce the first NN, implemented with PyTorch.
Chapter 2, Neural Networks, starts by introducing the mathematical branches related to NNs – linear algebra, probability, and differential calculus. It will focus on the building blocks and structure of NNs. It will also discuss how to train NNs with gradient descent and backpropagation.
Chapter 3, Deep Learning Fundamentals, introduces the basic paradigms of deep learning. It will make the transition from classic networks to deep NNs. It will outline the challenges of developing and using deep networks, and it will discuss how to solve them.
Chapter 4, Computer Vision with Convolutional Networks, introduces convolutional networks – the main network architecture for computer vision applications. It will discuss in detail their properties and building blocks. It will also introduce the most popular convolutional network models in use today.
Chapter 5, Advanced Computer Vision Applications, discusses applying convolutional networks for advanced computer vision tasks – object detection and image segmentation. It will also explore using NNs to generate new images.
Chapter 6, Natural Language Processing and Recurrent Neural Networks, introduces the main paradigms and data processing pipeline of NLP. It will also explore recurrent NNs and their two most popular variants – long short-term memory and gated recurrent units.
Chapter 7, The Attention Mechanism and Transformers, introduces one of the most significant recent deep learning advances – the attention mechanism and the transformer model based around it.
Chapter 8, Exploring Large Language Models in Depth, introduces transformer-based LLMs. It will discuss their properties and what makes them different than other NN models. It will also introduce the Hugging Face Transformers library.
Chapter 9, Advanced Applications of Large Language Models, discusses using LLMs for computer vision tasks. It will focus on classic tasks such as image classification and object detection, but it will also explore state-of-the-art applications such as text-to-image generation. It will introduce the LangChain framework for LLM-driven application development.
Chapter 10, Machine Learning Operations (MLOps), will introduce various libraries and techniques for easier development and production deployment of NN models.