Specifying outputs

The outputs section is where you explicitly list all the variables that will be used for the model training. If you have defined a group with some of the variables but you still want the original variables to be accounted for, you need to explicitly list them. The assignment section declares a list composed of the following: 

  • Groups
  • Assignments
  • Variables
  • Transformation (variable)

For instance, if you wanted the original body and subject of the e-mails as well as the bigrams you defined in assignments, you would need to declare the outputs as follows:

"outputs": [    "header",    "subject",    "body",    "bigrams"]

The following outputs declaration declares all the text variables and adds the bigrams assignment defined earlier on: ...

Get Effective Amazon Machine Learning now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.