caretEnsemble (version 4.0.1)

caretEnsemble: Combine several predictive models via weights

Description

Find a greedy, positive only linear combination of several train objects

Functions for creating ensembles of caret models: caretList and caretStack

Usage

caretEnsemble(all.models, excluded_class_id = 0L, tuneLength = 1L, ...)

Value

a caretEnsemble object

Arguments

all.models

an object of class caretList

excluded_class_id

The integer level to exclude from binary classification or multiclass problems. By default no classes are excluded, as the greedy optimizer requires all classes because it cannot use negative coefficients.

tuneLength

The size of the grid to search for tuning the model. Defaults to 1, as the only parameter to optimize is the number of iterations, and the default of 100 works well.

...

additional arguments to pass caret::train

Author

Maintainer: Zachary A. Deane-Mayer [copyright holder]

Other contributors:

• Jared E. Knowles [contributor]

• Antón López [contributor]

Details

greedyMSE works well when you want an ensemble that will never be worse than any single model in the dataset. In the worst case scenario, it will select the single best model, if none of them can be ensembled to improve the overall score. It will also never assign any model a negative coefficient, which can help avoid unintuitive cases at prediction time (e.g. if the correlations between predictors breaks down on new data, negative coefficients can lead to bad results).

See Also

Useful links:

• http://zachmayer.github.io/caretEnsemble/

• https://github.com/zachmayer/caretEnsemble

• Report bugs at https://github.com/zachmayer/caretEnsemble/issues

Examples

set.seed(42)
models <- caretList(iris[1:50, 1:2], iris[1:50, 3], methodList = c("rpart", "rf"))
ens <- caretEnsemble(models)
summary(ens)