1.0 - 8.00 - ROC Arguments - Teradata Vantage

Teradata® Vantage Machine Learning Engine Analytic Function Reference

Product
Teradata Vantage
Release Number
1.0
8.00
Release Date
May 2019
Content Type
Programming Reference
Publication ID
B700-4003-098K
Language
English (United States)
OutputTable
Specify the name for the output table that the function creates. The output_table must not already exist.
ModelIDColumn
[Optional] Specify the name of the input table column that contains the model or partition identifiers for the ROC curves.
ProbabilityColumn
Specify the name of the input table column that contains the predictions.
ObsColumn
Specify the name of the input table column that contains the actual classes.
PositiveClass
Specify the label of the positive class.
NumThreshold
[Optional] Specify the number of thresholds for the function to use. The num_threshold must be a NUMERIC value in the range [1, 10000].
Default: 50 (The function uniformly distributes the thresholds between 0 and 1.)
ROCValues
[Optional] Specify whether the function displays ROC values (thresholds, false positive rates, and true positive rates).
Default: 'true'. See the following note.
AUC
[Optional] Specify whether the function displays the AUC calculated from the ROC values.
Default: 'false'. See the following note.
Gini
[Optional] Specify whether the function displays the Gini coefficient calculated from the ROC values. The Gini coefficient is a measure of inequality among values of a frequency distribution. A Gini coefficient of 0 indicates that all values are the same. The closer the Gini coefficient is to 1, the more unequal are the values in the distribution.
Default: 'false'. See the following note.
The valid combinations of ROCValues, AUC, and Gini arguments are those that specify one of the following:
  • ROCValues only
  • AUC only
  • Gini only
  • AUC and Gini

When specifying AUC only, Gini only, or AUC and Gini only, you need not specify ROCValues ('false'), but you must not specify ROCValues ('true').

If you specify an invalid combination (such as ROCValues ('true') and AUC ('true'), or all three 'false'), the function issues an error message.