Book Image

C++ Game Development Cookbook

By : Druhin Mukherjee
Book Image

C++ Game Development Cookbook

By: Druhin Mukherjee

Overview of this book

<p>C++ is one of the preferred languages for game development as it supports a variety of coding styles that provides low-level access to the system. C++ is still used as a preferred game programming language by many as it gives game programmers control of the entire architecture, including memory patterns and usage. However, there is little information available on how to harness the advanced features of C++ to build robust games.</p> <p>This book will teach you techniques to develop logic and game code using C++. The primary goal of this book is to teach you to create high-quality games using C++ game programming scripts and techniques, regardless of the library or game engine you use. It will show you how to make use of the object-oriented capabilities of C++ so you can write well-structured and powerful games of any genre. The book also explores important areas such as physics programming and audio programming, and gives you other useful tips and tricks to improve your code.</p> <p>By the end of this book, you will be competent in game programming using C++, and will be able to develop your own games in C++.</p>
Table of Contents (20 chapters)
C++ Game Development Cookbook
Credits
About the Author
About the Reviewer
www.PacktPub.com
Preface
Index

Using neural network


Artificial neural networks (ANNs) are an advanced form of AI used in some games. They may not be directly used in-game; however, they may be used during the production phase to train the AI agents. Neural nets are mostly used as predictive algorithms. Based on certain parameters, and historical data, they calculate the most likely decision or attribute that the AI agent will distribute. ANNs are not restricted to games; they are used across multiple diverse domains to predict possible outcomes.

Getting ready

To work through this recipe, you will need a machine running Windows and Visual Studio.

How to do it…

Take a look at the following code snippet:

class neuralNetworkTrainer
{
  
private:

  //network to be trained
  neuralNetwork* NN;

  //learning parameters
  double learningRate;          // adjusts the step size of the weight update  
  double momentum;            // improves performance of stochastic learning (don't use for batch)

  //epoch counter
  long epoch;...