R/exploratory.R

In snarles/introStatML: Introduction to statistics and machine learning

#' Samples rows from a table
#'
#' @param tab A data fram
#' @param k The number of rows to sample
#' @return Random subset of rows from the data frame
#' @export
sample_rows <- function(tab, k = 5, replace = FALSE) {
  tab[sample(nrow(tab), k, replace), ]
}

#' Get data from 2014 house of representatives and print metadata
#'
#' @param rolls Which rolls to obtain, 1-99
#' @param print Whether to print metadata
#' @return A dataframe
#' @export
roll_data <- function(rolls = 1:99, print = TRUE) {
  if (print) {
    print(roll2014meta[rolls, -5])
  }
  roll2014votes[, c(2:3, rolls + 3)]
}

#' Plot polynomial a + b * x + c * x^2 + d * x^3
#'
#' @param a intercept
#' @param b x term
#' @param c x^2 term
#' @param d x^3 term
#' @export
abpoly <- function(a = 0, b = 0, c = 0, d = 0, ...) {
  xl <- par("usr")[1:2]
  xs <- seq(xl[1], xl[2], length.out = 200)
  ys <- a + b * xs + c * xs^2 + d * xs^3
  lines(xs, ys, ...)
}

#' Split a data frame into training a test sets
#'
#' @param tab The data frame
#' @param frac Fraction of data for training set (will be rounded up)
#' @param seed Random seed
#' @return A list with train and test components
#' @export
data_split <- function(tab, frac = 0.5, seed = NULL) {
  n <- nrow(tab)
  if (!is.null(seed)) set.seed(seed)
  inds <- sort(sample(n, ceiling(frac * n), replace = FALSE))
  ans <- list()
  ans$train <- tab[inds, ]
  ans$test <- tab[-inds, ]
  ans
}