The first section of this chapter introduces the DeepLearning4j framework and then presents some use cases of training data ingestion from files using this framework along with Apache Spark.
Training data ingestion through Spark
The DeepLearning4j framework
Before jumping into the first example, let's quickly introduce the DeepLearning4j (https://deeplearning4j.org/) framework. It is an open source (released under the Apache license 2.0 (https://www.apache.org/licenses/LICENSE-2.0)), distributed deep learning framework written for the JVM. Being integrated since its earliest releases with Hadoop and Spark, it takes advantage of such distributed computing frameworks to speed up network training. It is written in Java...
