extract_importance_SL {flevr}

Extract extrinsic importance from a Super Learner object

Description

Extract the individual-algorithm extrinsic importance from each fitted algorithm within the Super Learner; compute the average weighted rank of the importance scores, with weights specified by each algorithm's weight in the Super Learner.

Usage

extract_importance_SL(fit, feature_names, import_type = "all", ...)

Arguments

fit

the fitted Super Learner ensemble

feature_names

the names of the features

import_type

the level of granularity for importance: "all" is the importance based on the weighted average of ranks across algorithmrithms (weights are SL coefs); "best" is the importance based on the algorithmrithm with highest weight. Defaults to "all".

...

other arguments to pass to individual-algorithm extractors.

Value

a tibble, with columns feature (the feature) and rank (the weighted feature importance rank, with 1 indicating the most important feature).

Examples

data("biomarkers")
# subset to complete cases for illustration
cc <- complete.cases(biomarkers)
dat_cc <- biomarkers[cc, ]
# use only the mucinous outcome, not the high-malignancy outcome
y <- dat_cc$mucinous
x <- dat_cc[, !(names(dat_cc) %in% c("mucinous", "high_malignancy"))]
feature_nms <- names(x)
# get the fit (using a simple library and 2 folds for illustration only)
set.seed(20231129)
library("SuperLearner")
fit <- SuperLearner::SuperLearner(Y = y, X = x, SL.library = c("SL.glm", "SL.mean"), 
                                  cvControl = list(V = 2))
# extract importance using all learners
importance <- extract_importance_SL(fit = fit, feature_names = feature_nms)
importance
# extract importance of best learner
best_importance <- extract_importance_SL(fit = fit, feature_names = feature_nms, 
                                         import_type = "best")
best_importance