R Utility Functions for cleaning, randomizing, sorting and creating test / training sets

dataset_utilities.R

clean_data <- function(dataset) {
  na.omit(dataset)
}

randomize <- function(dataset) {
  dataset$randomized <- runif(length(dataset[,1]))
  dataset
}

randomize_ordered <- function(dataset) { 
  randomized_dataset <- randomize(dataset)
  randomized_dataset[order(randomized_dataset$randomized), ]
}

split_dataset <- function(dataset, first_half = 1) {
  half <- length(dataset[,1]) / 2
  if (first_half == 1) {
    dataset[seq(1,half),]
  } else {
    dataset[seq(half + 1, length(dataset[,1])),]
  }
}

# Example
example_dataset <- read.csv("example_data.csv", header = TRUE, na.strings="")
training <- split_dataset(randomize_ordered(clean_data(example_dataset)), first_half = 1)
test <- split_dataset(randomize_ordered(clean_data(example_dataset)), first_half = 0)