Apache Spark Deep Learning Cookbook

How it works...

This section explains how we prepare the dataset for the deep learning pipeline:

1. For ease of use inside the deep learning pipeline, it is best to clean up the column names and the order of the columns before the pipeline receives the data. After renaming the column headers, we can view the updated columns, as seen in the following script:

2. A bit of manipulation is performed on the ratings column to round up values of 0.5 increments to the next-highest whole number. This will assist when we are doing our multi-class classification within Keras to group ratings into six categories, instead of 11 categories.

 

3. To consume the ...