Book Image

Hands-On Deep Learning with Apache Spark

By : Guglielmo Iozzia

Book Image

Hands-On Deep Learning with Apache Spark

By: Guglielmo Iozzia

Overview of this book

Deep learning is a subset of machine learning where datasets with several layers of complexity can be processed. Hands-On Deep Learning with Apache Spark addresses the sheer complexity of technical and analytical parts and the speed at which deep learning solutions can be implemented on Apache Spark. The book starts with the fundamentals of Apache Spark and deep learning. You will set up Spark for deep learning, learn principles of distributed modeling, and understand different types of neural nets. You will then implement deep learning models, such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long short-term memory (LSTM) on Spark. As you progress through the book, you will gain hands-on experience of what it takes to understand the complex datasets you are dealing with. During the course of this book, you will use popular deep learning frameworks, such as TensorFlow, Deeplearning4j, and Keras to train your distributed models. By the end of this book, you'll have gained experience with the implementation of your models on a variety of use cases.

Preface

Who this book is for

What this book covers

To get the most out of this book

Free Chapter

The Apache Spark Ecosystem

The Apache Spark Ecosystem

Apache Spark fundamentals

RDD programming

Spark SQL, Datasets, and DataFrames

Spark Streaming

Cluster mode using different managers

Deep Learning Basics

Deep Learning Basics

Practical applications of DL

Extract, Transform, Load

Extract, Transform, Load

Training data ingestion through Spark

Training data ingestion from a database with Spark

Data ingestion from S3

Raw data transformation with Spark

Streaming

Streaming data with Apache Spark

Streaming data with Kafka and Spark

Streaming data with DL4J and Spark

Convolutional Neural Networks

Convolutional Neural Networks

Convolutional layers

Fully connected layers

GoogleNet Inception V3 model

Hands-on CNN with Spark

Recurrent Neural Networks

Recurrent Neural Networks

Hands-on RNNs with Spark

Training Neural Networks with Spark

Training Neural Networks with Spark

Distributed network training with Spark and DeepLearning4j

Hyperparameter optimization

Monitoring and Debugging Neural Network Training

Monitoring and Debugging Neural Network Training

Monitoring and debugging neural networks during their training phases

Interpreting Neural Network Output

Interpreting Neural Network Output

Evaluation techniques with DL4J

Other types of evaluation

Deploying on a Distributed System

Deploying on a Distributed System

Setup of a distributed environment with DeepLearning4j

Spark distributed training architecture details

Importing Python models into the JVM with DL4J

Alternatives to DL4J for the Scala programming language

NLP Basics

Hands-on NLP with Spark

Textual Analysis and Deep Learning

Textual Analysis and Deep Learning

Hands-on NLP with DL4J

Hands-on NLP with TensorFlow

Hand-on NLP with Keras and a TensorFlow backend

Hands-on NLP with Keras model import into DL4J

Convolution

Object recognition strategies

Convolution applied to image recognition

Image Classification

Image Classification

Implementing an end-to-end image classification web application

What's Next for Deep Learning?

What's Next for Deep Learning?

What to expect next for deep learning and AI

Topics to watch for

Is Spark ready for RL?

DeepLearning4J future support for GANs

Other Books You May Enjoy

Other Books You May Enjoy

Leave a review - let other readers know what you think

Appendix A: Functional Programming in Scala

Appendix A: Functional Programming in Scala

Functional programming (FP)

Appendix B: Image Data Preparation for Spark

Appendix B: Image Data Preparation for Spark

Image preprocessing

Customer Reviews

5 star

0

4 star

0

3 star

0

2 star

0

1 star

0

Weights

CNNs share weights in convolutional layers. This means that the same filter is used for each receptive field in a layer and that these replicated units share the same parameterization (weight vector and bias) and form a feature map.

The following diagram shows three hidden units of a network belonging to the same feature map:

Figure 5.3: Hidden units

The weights in the darker gray color in the preceding diagram are shared and identical. This replication allows features detection regardless of the position they have in the visual field. Another outcome of this weight sharing is the following: the efficiency of the learning process increases by drastically reducing the number of free parameters to be learned.