This repo contains an example implementation for training a deep learning model to predict machine failures on a training set that wouldn't fit on a single machine. Distributed data wrangling and model training is done by using Spark and Spark's ML library. AWS EMR service is used to create a cluster and run the Spark job.
The article is here.
Tags: spark, pyspark, aws, emr, bigdata, deeplearning, digitaltransformation