Splits data into a training set and a test set. Parameter ratio determines the ratio of observation going into the training set (default: 2/3).

## Format

R6::R6Class inheriting from Resampling.

## Construction

ResamplingHoldout$new() mlr_resamplings$get("holdout")
rsmp("holdout")


See Resampling.

See Resampling.

## Parameters

• ratio :: numeric(1)
Ratio of observations to put into the training set.

as.data.table(mlr_resamplings) for a complete table of all (also dynamically created) Resampling implementations.

## Examples

# Create a task with 10 observations
task$filter(1:10) # Instantiate Resampling rho = rsmp("holdout", ratio = 0.5) rho$instantiate(task)
rho$train_set(1)#> [1] 4 9 6 7 2rho$test_set(1)#> [1]  1  3  5  8 10intersect(rho$train_set(1), rho$test_set(1))#> integer(0)
rho$instance # simple list#>$train
#>