R Deep Learning Projects

5 real-world projects to help you master deep learning concepts
Key FeaturesMaster the different deep learning paradigms and build real-world projects related to text generation, sentiment analysis, fraud detection, and more Get to grips with R's impressive range of Deep Learning libraries and frameworks such as deepnet, MXNetR, Tensorflow, H2O, Keras, and text2vec Practical projects that show you how to implement different neural networks with helpful tips, tricks, and best practicesBook DescriptionR is a popular programming language used by statisticians and mathematicians for statistical analysis, and is popularly used for deep learning. Deep Learning, as we all know, is one of the trending topics today, and is finding practical applications in a lot of domains.
This book demonstrates end-to-end implementations of five real-world projects on popular topics in deep learning such as handwritten digit recognition, traffic light detection, fraud detection, text generation, and sentiment analysis. You'll learn how to train effective neural networks in R—including convolutional neural networks, recurrent neural networks, and LSTMs—and apply them in practical scenarios. The book also highlights how neural networks can be trained using GPU capabilities. You will use popular R libraries and packages—such as MXNetR, H2O, deepnet, and more—to implement the projects.
By the end of this book, you will have a better understanding of deep learning concepts and techniques and how to use them in a practical setting.
What you will learnInstrument Deep Learning models with packages such as deepnet, MXNetR, Tensorflow, H2O, Keras, and text2vec Apply neural networks to perform handwritten digit recognition using MXNetGet the knack of CNN models, Neural Network API, Keras, and TensorFlow for traffic sign classification -Implement credit card fraud detection with Autoencoders Master reconstructing images using variational autoencoders Wade through sentiment analysis from movie reviews Run from past to future and vice versa with bidirectional Long Short-Term Memory (LSTM) networks Understand the applications of Autoencoder Neural Networks in clustering and dimensionality reductionWho this book is forMachine learning professionals and data scientists looking to master deep learning by implementing practical projects in R will find this book a useful resource. A knowledge of R programming and the basic concepts of deep learning is required to get the best out of this book.
Yuxi (Hayden) Liu is currently an applied research scientist focused on developing machine learning models and systems for given learning tasks. He has worked for a few years as a data scientist, and applied his machine learning expertise in computational advertising. He earned his degree from the University of Toronto, and published five first-authored IEEE transaction and conference papers during his research. His first book, titled Python Machine Learning By Example, was ranked the #1 bestseller in Amazon India in 2017. He is also a machine learning education enthusiast. Pablo Maldonado is an applied mathematician and data scientist with a taste for software development since his days of programming BASIC on a Tandy 1000. As an academic and business consultant, he spends a great deal of his time building applied artificial intelligence solutions for text analytics, sensor and transactional data, and reinforcement learning. Pablo earned his PhD in applied mathematics (with focus on mathematical game theory) at the Universite Pierre et Marie Curie in Paris, France.