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 analysis configurations definition JSON file are as follows.

[
    {
        "key": "algorithmName",
        "value": "LOGISTIC_REGRESSION"
    },
    {
        "key": "algorithmType",
        "value": "Classification"
    },
    {
        "key": "responseVariable",
        "value": "Class"
    },
    {
        "key": "trainDataFraction",
        "value": "0.7"
    }
]

Parameter definitions

The definitions of the parameters in the above sample analysis configurations definition JSON file are as follows.

Parameter
Description
Required
Default value
Example
algorithmName

Name of the machine learning algorithm to be used. Define one of the following values.

  • LOGISTIC_REGRESSION 
  • DECISION_TREE 
  • SVM 
  • NAIVE_BAYES 
  • LINEAR_REGRESSION 
  • RIDGE_REGRESSION 
  • LASSO_REGRESSION 
  • K_MEANS
YesNone

LOGISTIC_REGRESSION

algorithmType

Type of the algorithm. Define one of the following values.

  • Classification 
  • Numerical_Prediction 
  • Clustering
Yes None

Classification

responseVariable
Header name of the response variableYesNone

Class

trainDataFraction
 Proportion of the train dataset from the full dataset (0-1 scale) Yes None

 0.7

  • No labels