BinModels: Bin models function

In citronmeliss/predictionpackr: Predictionpackr

Description

This function bins model predictions using SupaLarnas gridsearch.breaks function.

Usage

BinModels(predictions, outcome, n_cores, grid, return_cps = TRUE,
  gridsearch_parallel = FALSE, is_sample = TRUE, maximise = FALSE)

Arguments

predictions	A numeric vector of predicted probabilites. No default.
outcome	A numeric vector of 0 and 1. No default.
n_cores	Number of cores to be used in parallel gridsearch. Passed to SupaLarna::gridsearch.breaks. As integer. Defaults to 2 (in gridsearch.breaks)
grid	All values to be combnd in gridsearch.breaks, i.e. used to find optimal cutpoints. No default.
return_cps	Logical. Function returns model cut_points if TRUE. Defaults to TRUE.
gridsearch_parallel	Logical. If TRUE the gridsearch is performed in parallel. Defaults to FALSE.
is_sample	Logical. If TRUE, only a tenth of possible cut points is searched. Defaults to TRUE.
maximise	Logical. If TRUE, grid search maximises performance metric. Passed to SupaLarna::gridsearch.breaks. Defaults to TRUE.

citronmeliss/predictionpackr documentation built on Feb. 10, 2020, 12:19 a.m.