MachineShop (version 2.9.0)

resample: Resample Estimation of Model Performance

Description

Estimation of the predictive performance of a model estimated and evaluated on training and test samples generated from an observed data set.

Usage

resample(x, ...)
# S3 method for formula
resample(x, data, model, control = MachineShop::settings("control"), ...)
# S3 method for matrix
resample(x, y, model, control = MachineShop::settings("control"), ...)
# S3 method for ModelFrame
resample(x, model, control = MachineShop::settings("control"), ...)
# S3 method for recipe
resample(x, model, control = MachineShop::settings("control"), ...)
# S3 method for MLModel
resample(x, ...)
# S3 method for MLModelFunction
resample(x, ...)

Arguments

x

input specifying a relationship between model predictor and response variables. Alternatively, a model function or call may be given first followed by the input specification and control value.

...

arguments passed to other methods.

data

data frame containing observed predictors and outcomes.

model

model function, function name, or call; ignored and can be omitted when resampling modeled inputs.

control

control function, function name, or call defining the resampling method to be employed.

y

response variable.

Value

Resamples class object.

Details

Stratified resampling is performed automatically for the formula and matrix methods according to the type of response variable. In general, strata are constructed from numeric proportions for BinomialVariate; original values for character, factor, logical, and ordered; first columns of values for matrix; original values for numeric; and numeric times within event statuses for Surv. Numeric values are stratified into quantile bins and categorical values into factor levels defined by MLControl.

Resampling stratification variables may be specified manually for ModelFrames upon creation with the strata argument in their constructor. Resampling of this class is unstratified by default.

Stratification variables may be designated in recipe specifications with the role_case function. Resampling will be unstratified otherwise.

See Also

c, metrics, performance, plot, summary

Examples

# NOT RUN {
## Requires prior installation of suggested package gbm to run

## Factor response example

fo <- Species ~ .
control <- CVControl()

gbm_res1 <- resample(fo, iris, GBMModel(n.trees = 25), control)
gbm_res2 <- resample(fo, iris, GBMModel(n.trees = 50), control)
gbm_res3 <- resample(fo, iris, GBMModel(n.trees = 100), control)

summary(gbm_res1)
plot(gbm_res1)

res <- c(GBM1 = gbm_res1, GBM2 = gbm_res2, GBM3 = gbm_res3)
summary(res)
plot(res)
# }
# NOT RUN {
# }