(AI Studio Core)
Synopsis
This operator searches the threshold that would cause the recall to rise or fall to the given level.Description
The given example set must contain a label and a compatible predicted label attribute together with all confidence attributes. Since Recall supports only one or two classes, this operator will only work on data sets with one or two label values at most. The label value that should be treated positive, can be defined in the operator. This operator will sort the given example set according to the confidence attribute of the positive label and searches the confidence threhsold, that would cause the desired recall. Of course this will affect the precision.
Input
example set (Data Table)This is an example set input port
Output
- example set (Data Table)
This is an example set output port
- threshold (IOObject)
Parameters
- min recallThe minimal desired recall on the positive class.
- use example weightsIndicates if example weights should be used.
- positive labelIf set, this value of the label attribute is treated as positive.