pred_nestcv_glmnet: Prediction wrappers to use shapr with nestedcv

In nestedcv: Nested Cross-Validation with 'glmnet' and 'caret'

Prediction wrappers to use shapr with nestedcv

Description

Prediction wrapper functions to enable the use of the shapr package for generating SHAP values from nestedcv trained models.

Usage

pred_nestcv_glmnet(x, newdata)

pred_nestcv_glmnet_class(cl)

pred_train(x, newdata)

pred_train_class(cl)

pred_SuperLearner(x, newdata)

Arguments

x	a nestcv.glmnet, nestcv.train or nestcv.SuperLearner object
newdata	a matrix or data frame of new data
cl	integer representing which class to predict

Details

These prediction wrapper functions are designed to be used with the shapr package via nestcv.explain(). The functions pred_nestcv_glmnet and pred_train work for nestcv.glmnet and nestcv.train models respectively for either binary classification or regression.

For multiclass classification use pred_nestcv_glmnet_class(1), pred_nestcv_glmnet_class(2) etc for each class. Similarly pred_train_class(1), pred_train_class(2) etc for nestcv.train objects.

Value

prediction wrapper function designed for use with nestcv.explain() or shapr::explain()

Examples

if (requireNamespace("shapr") & requireNamespace("mlbench")) {
  library(shapr)

  # Boston housing dataset
  library(mlbench)
  data(BostonHousing2)
  dat <- BostonHousing2
  y <- dat$cmedv
  x <- subset(dat, select = -c(cmedv, medv, town, chas))

  # Fit a glmnet model using nested CV
  # Only 3 outer CV folds and 1 alpha value for speed
  fit <- nestcv.glmnet(y, x, family = "gaussian", n_outer_folds = 3, alphaSet = 1)

  # Generate SHAP values using shapr
  sh <- nestcv.explain(fit, pred_nestcv_glmnet)
                       
  # Plot overall variable importance
  plot_shap_bar(sh, x)

  # Plot beeswarm plot
  plot_shap_beeswarm(sh, x, size = 1)
}

nestedcv documentation built on July 14, 2026, 9:07 a.m.