Grid Search

Grid search to optimise hyperparameters for FUN

gridsearch(x, y, searchspace, FUN, nfolds = 5, nrepcv = 2, ...)

Arguments

x

matrix/data.frame, feature matrix, see ranger() for details.

y

numeric/factor, classification labels, see ranger() for details.

searchspace

data.frame, hyperparameters to tune. Column names have to match the argument names of FUN.

FUN

function function to optimize.

nfolds

integer(1) number of cross validation folds.

nrepcv

integer(1) number of repeats.

...

further arguments passed to FUN

Value

data.frame with tested hyperparameters and metric

Examples

iris <- subset(iris, Species != "setosa")
searchspace <- expand.grid(
   mtry = c(2, 3),
   num.trees = c(500, 1000)
)
## nfolds and nrepcv are too low for real world applications, and are just
## used for demonstration and to keep the run time of the examples low
gs_rusranger(
    iris[-5], as.numeric(iris$Species == "versicolor"),
    searchspace = searchspace, nfolds = 3, nrepcv = 1
)
#>   mtry num.trees        Min         Q1     Median         Q3        Max
#> 1    2       500 0.00781250 0.00781250 0.00781250 0.00781250 0.00781250
#> 2    3       500 0.00390625 0.00390625 0.00390625 0.00390625 0.00390625
#> 3    2      1000 0.01384083 0.01384083 0.01384083 0.01384083 0.01384083
#> 4    3      1000 0.02249135 0.02249135 0.02249135 0.02249135 0.02249135