Summary

With the ubiquitous status of IoT, the data being generated is growing at an exponential rate. This data, mostly unstructured and available in vast quantities, is often referred to as big data. A large number of frameworks and solutions have been proposed to deal with the large set of data. One of the promising solutions is DAI, distributing the model or data among the cluster of machines. We can use distributed TensorFlow, or TFoS frameworks to perform distributed model training. In recent years, some easy-to-use open source solutions have been proposed. Two of the most popular and successful solutions are Apache Spark's MLlib and H2O.ai's H2O. In this chapter, we showed how to train ML models for both regression and classification in MLlib and H2O. The Apache Spark MLlib supports SparkDL, which provides excellent support for image classification and detection...