R/input_handling.R

In TImA97/randfindR: Analysis of Randomness in Human Generated Sequences

#' convert vector to matrix of first order dependencies
#'
#' @param x vector of random numbers
#' @param options number of available options in sequence
#' @param order indicates displacement between response pairs (default is 1)
#' @param circ indicate whether to include wrap around from end to the beginning
#' of the sequence when computing response pairs
#'
#' @details helper function to transform a given vector to a matrix of first
#' order dependencies, specifying the frequency of all possible 2-digit long
#' sequences
#'
#' @noRd
convert_to_matrix <- function(x, options, order = 1, circ = TRUE) {
  matr <- matrix(data = 0, nrow = options, ncol = options)

  ## check whether to include wrap around in the computation of response pairs
  indizes <- 1:(length(x) - order)
  if (circ) {
    indizes <- 1:length(x)
  }

  ## compute number of response pairs depending on the specified order
  for (i in indizes) {
    current_value <- x[i]
    next_index <- i + order
    ## if wrap around occurs, compute next index with modulo
    if (next_index > length(x)) {
      next_index <- next_index %% length(x)
    }
    next_value <- x[next_index]
    matr[current_value, next_value] <- matr[current_value, next_value] + 1
  }

  return(matr)
}


#' function transforms given vector to numeric
#'
#' @param x vector
#' @return x_transformed
#'
#' @noRd
to_numeric <- function(x) {
  as.numeric(factor(x))
}


#' function returns the number of unique response in a vector
#' @param x vector
#' @return number of unique responses in \code{x}
#'
#' @noRd
get_number_unique_responses <- function(x) {
  length(unique(x))
}