O'Reilly logo

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Distributed Deep Learning with Apache Spark

Video Description

End to end pipeline with different deep learning frameworks to be embedded in Spark workflows

About This Video

  • Use Apache Spark to create fast and parallel deep learning programs that can be leveraged in multiple systems and business domains
  • Understand how to leverage deep learning with anomaly-detection problems
  • Learn about deep learning layers and how to compose them with well-known ML algorithms

In Detail

Deep learning is a subfield of Artificial Intelligence and Machine Learning where a huge amount of data is processed in complex layers of neural networks. It has solved tons of interesting real-world problems in recent years. Distributed deep learning (DL) involves training a deep neural network in parallel across multiple machines. In this course, you will get started with implementing Deep Learning solutions easily with the help of Apache Spark.

You will begin with a short introduction on Deep Learning and Apache Spark and the principles of distributed modeling. With the help of real-world examples, you will investigate different types of neural network and work with DL libraries such as BigDL, Deeplearning4j, and the Deep Learning pipelines library to implement DL models and distributed computing on Spark. You will see how you can easily use a large dataset to implement efficient DL solutions to simplify real-world examples. You will also learn how to distribute the computationally heavy parts of DL into processes with the help of Apache Spark.

By the end of this course, you'll have gained experience in implementing Distributed Deep Learning for your models at work. Our examples will be based on real-world problems from the banking industry.

The code bundle for this course is available at https://github.com/PacktPublishing/Distributed-Deep-Learning-with-Apache-Spark

Downloading the example code for this course: You can download the example code files for all Packt video courses you have purchased from your account at http://www.PacktPub.com. If you purchased this course elsewhere, you can visit http://www.PacktPub.com/support and register to have the files e-mailed directly to you.

Table of Contents

  1. Chapter 1 : Leveraging Apache Spark Distributed Nature in Integration with DL4J
    1. Course Overview 00:02:07
    2. Understanding Spark Distributed Architecture 00:05:58
    3. Learning Spark API that Is Used by DL4J 00:04:12
    4. Configuring DL4J with Spark 00:03:25
    5. Adding ND4J for Distributed Feature Vector Processing 00:06:24
  2. Chapter 2 : Deep Learning RNN for Text Extraction
    1. Understanding and Using Word2Vec 00:05:19
    2. Downloading IMDB Movies Comments and Examine Data 00:04:47
    3. Creating Multilayer Network 00:02:56
    4. Using Word2Vec Transforming Iterator from DL4J 00:05:13
    5. Start Training and Validating Results 00:05:23
  3. Chapter 3 : Anomaly Detection
    1. Anomaly Detection Problem, Explained 00:04:32
    2. Extracting Features from Input Data Using Multilayer Approach 00:04:28
    3. Adding Layer that Finds an Actual Anomaly 00:04:06
    4. Testing and Validating Results from our Deep Learning Model 00:04:20
  4. Chapter 4 : Leveraging ND4J with DL4J for the Classification Problem
    1. How to Extract Features from Unknown Data? 00:04:42
    2. Defining Deep Learning Layer for Feature Extraction 00:03:39
    3. Defining Number of Classes (Clusters) 00:03:34
    4. Building Multilayer that Performs Classification into Classes 00:04:07
  5. Chapter 5 : Plugging Regression into Deep Learning Network
    1. Explanation of Regression Algorithm 00:03:25
    2. Constructing Regression 00:03:41
    3. Using Regression with DL4J Layer 00:03:22
    4. Running Our Network 00:04:10
  6. Chapter 6 : Tips and Tricks
    1. How to Tune Your Layer for Extracting Features 00:05:11
    2. How the Number of Iterations Affect Precision 00:03:51
    3. How Many Features Should a Layer Produce 00:03:49
    4. Most Common Mistakes When Using Deep Learning Algorithms in Production 00:03:45