caret (version 2.27)

resampleSummary: Summary of resampled performance estimates

Description

This function uses the out-of-bag predictions to calculate overall performance metrics and returns the observed and predicted data.

Usage

resampleSummary(obs, resampled, index = NULL, keepData = TRUE)

Arguments

obs

A vector (numeric or factor) of the outcome data

resampled

For bootstrapping, this is either a matrix (for numeric outcomes) or a data frame (for factors). For cross-validation, a vector is produced.

index

The list to index of samples in each cross--validation fold (only used for cross-validation).

keepData

A logical for returning the observed and predicted data.

Value

• A list with:
• metrics
  A vector of values describing the bootstrap distribution.
• data
  A data frame or NULL. Columns include obs, pred and group (for tracking cross-validation folds or bootstrap samples)

Details

The mean and standard deviation of the values produced by postResample are calculated.

See Also

postResample

Examples

resampleSummary(rnorm(10), matrix(rnorm(50), ncol = 5))