Hybrid Data Science (HDS) Modeling Approaches for Industrial and Scientific Applications

This book details the basics of Deep Learning, Machine Learning and numerical grid based methods such as finite element, finite volume and finite difference. Hybrid models combining grid based methods and machine learning in particular neural networks in covered in detail. It will take you through a step by step approach making the user to understand both the physics domain and data science to hybridize both. The book describes the methods to create hybrid data science models for industry and scientific applications. The book covers some practical applications where the hybrid techniques can be used. It also provides sample python codes for the several chapters discussed in the book. This book represents our attempt to make hybrid physics and deep learning approachable, meaningful and present the basic concepts, context, and the code. We believe that this might be the first book published using such hybrid physics/machine learning technology.