aum (version 2023.6.14)

aum_diffs: aum diffs

Description

Create error differences data table which can be used as input to aum function. Typical users should not use this function directly, and instead use aum_diffs_binary for binary classification, and aum_diffs_penalty for error defined as a function of non-negative penalty.

Usage

aum_diffs(example, pred, 
    fp_diff, fn_diff, 
    pred.name.vec)

Value

data table of class "aum_diffs" in which each rows represents a breakpoint in an error function. Columns are interpreted as follows: there is a change of "fp_diff","fn_diff" at predicted value "pred" for example/observation "example". This can be used for computing Area Under Minimum via aum function, and plotted via plot.aum_diffs.

Arguments

example

Integer or character vector identifying different examples.

pred

Numeric vector of predicted values at which the error changes.

fp_diff

Numeric vector of difference in fp at pred.

fn_diff

Numeric vector of difference in fn at pred.

pred.name.vec

Character vector of example names for predictions.

Author

Toby Dylan Hocking <[e-mail]> [aut, cre], Jadon Fowler [aut] (Contributed exact line search C++ code)

Examples

aum::aum_diffs_binary(c(0,1))
aum::aum_diffs(c("positive", "negative"), 0, c(0,1), c(-1,1), c("negative", "positive"))
rbind(aum::aum_diffs(0L, 0, 1, 0), aum_diffs(1L, 0, 0, -1))