Split a dataset into training and validation portions with optional
seed for reproducibility.
Usage
split_dataset(dataset, prop = 0.8, seed = NULL)
Arguments
- dataset
A data frame to split.
- prop
Proportion of data for training. Default is 0.8.
- seed
Random seed for reproducibility.
Value
A list with train and val data frames.
Examples
df <- tibble::tibble(x = 1:100)
split <- split_dataset(df, prop = 0.8, seed = 42)
nrow(split$train) # ~80
#> [1] 80
nrow(split$val) # ~20
#> [1] 20
