This documentation is for Machine Learner 1.0.0. View documentation for the latest release.
||
Skip to end of metadata
Go to start of metadata

The configurations included in a sample customized features definition JSON file are as follows.

[
    {
        "type": "numerical",
        "name": "numberOfLinks",
        "include": true,
        "imputeOption": "DISCARD"
    },
    {
        "type": "numerical",
        "name": "class",
        "include": true,
        "imputeOption": "DISCARD"
    }
]

Parameter definitions

The definitions of the parameters in the above sample customized features definition JSON file are as follows.

Parameter
Description
Required
Default value
Example
type
Type of the feature (numerical/categorical)YesNone

numerical

name
Name of the feature as it appears in the datasetYesNone

numberOfLinks

include
Whether this feature should be used for the analysis (true/false)YesNone

true

imputeOption

The action to be done if a value is missing. Define one out of the following values.

  •  DISCARD
  • REPLACE_WTH_MEAN
  • REGRESSION_IMPUTATION
YesNone

DISCARD

  • No labels