get_sl_input: Create input list for 'get_sl'

In benkeser/cvma: Cross-validation-based maximal association

Description

Create input list for get_sl

Usage

get_sl_input(split_Y, Yname, valid_folds, V, all_fit_tasks, all_fits,
  learners)

Arguments

split_Y	The outcome matrix split by relevant validation folds.
Yname	The names of the outcomes. Used to search all_fits and all_sl.
valid_folds	If not NULL, identifies validation folds used in the inner most super learner. Set to NULL when getting weights for outer cross-validation folds (where all folds are used for training super learner).
V	Number of folds.
all_fit_tasks	List of all learner fit tasks (faster to search over than search over all_fits).
all_fits	List of all learner fits.
learners	Vector of super learner vectors.

Value

List with each entry a list with entries: valid_fold (the number of the corresponding fold), Y (univariate outcome in this validation fold), pred (matrix of outcome predictions for this validation fold from learners fit in training folds). The length of the list varies according to what is input into valid_folds so that this function may be used both in the outer and inner layers of cross-validation.

benkeser/cvma documentation built on May 5, 2019, 1:37 p.m.