MachineShop (version 1.1.0)

StackedModel: Stacked Regression Model

Description

Fit a stacked regression model from multiple base learners.

Usage

StackedModel(..., control = CVControl, weights = NULL)

Arguments

...

MLModel objects to serve as base learners.

control

MLControl object, control function, or character string naming a control function defining the resampling method to be employed for the estimation of base learner weights.

weights

optional fixed base learner weights.

Value

StackedModel class object that inherits from MLModel.

Details

Response Types:

factor, numeric, ordered, Surv

References

Breiman, L. (1996) Stacked Regression. Machine Learning, 24, 49--64.

See Also

fit, resample, tune

Examples

# NOT RUN {
library(MASS)

model <- StackedModel(GBMModel, SVMRadialModel, GLMNetModel(lambda = 0.01))
modelfit <- fit(medv ~ ., data = Boston, model = model)
predict(modelfit, newdata = Boston)

# }