partition of data (train-test-split)
DataSplit(y, TrainRatio = 0.75, regression = TRUE, shuffle = FALSE, seed = 1)
y | is a numeric vector (response variable) |
---|---|
TrainRatio | is the percentage of train-data after the partition |
regression | is a boolean (TRUE, FALSE) indicating if it's a regression or classification task |
shuffle | is a boolean (TRUE, FALSE) indicating if the data should be shuffled or not (by default 5 times) |
seed | an integer specifying the random seed |
a list of indices (train-test)
if (FALSE) { data(iris) y = X[, 1] split = DataSplit(y, TrainRatio = 0.75, regression = FALSE, shuffle = FALSE) }