Home > Testing and Tuning Models > Testing Classification Models > Test Metrics for Classifica... > Classification Model Test a... > Tuning Classification Models > Cost > ROC
ROC is only supported for binary models. For more information, see Receiver Operating Characteristics.
The ROC Tuning tab adds a side panel to the standard ROC Test Viewer. The following information is displayed:
Performance Matrix(s) in the upper right-hand pane, allows you to display these matrices:
Overall Accuracy - cost matrix for the maximum overall accuracy point on the ROC chart
Average Accuracy - cost matrix for the maximum average accuracy point
Custom Accuracy - cost matrix for the custom operating point
You must specify a custom operating point for this option to be available. See Select Custom Operating Point for details.
Model Accuracy - the current performance matrix (approximately) of current model.
The following calculation is used to derive this from the ROC result provided:
If there is no embedded cost matrix, find the 50% threshold point or the closest one to it. If there is an embedded cost matrix, find the lowest cost point. (For a model to have an embedded cost matrix, it must have either been tuned or it has a cost matrix or cost benefit defined by the default settings of the build node.)
The Performance Matrix Grid shows the performance matrix for the option selected.
Clicking Tune selects the current performance option as the one to use to tune the model. Clicking Tune implies deriving a cost matrix from the ROC result at that probability threshold. Tune Settings, the lower part of this panel, is updated to display the new matrix.
Clicking Clear clears any tuning specification and sets tuning to Automatic. In other words, no tuning is performed.