pnpAt: pnpAt

In benmack/oneClass: One-class classification in the absence of test data

Description

Calculate the probability of negative prediction at the minimum/lower whisker of the positive hold-out predictions for one or more models (parameter settings) stored in the trainOcc object.

Usage

pnpAt(x, modRow = NULL, u = NULL)

Arguments

x	an object of class trainOcc
modRow	the row in the results table corresponding to the model for which to derive the ppp.
u	a data which can be predicted with the x. If given the predictions made on u are used for the calculation instead of the held out predictions stored in x

Value

the lower whisker threshold and the corresponding probability of positive prediction at the Lower whisker

benmack/oneClass documentation built on Dec. 15, 2020, 7:38 p.m.