R/plot.R

In bfgray3/bigo: Analysis of Empirical Algorithmic Complexity

#################
### plot.bigo ###
#################

#' @export
plot.bigo <- function(x, y, ...) {

  df <- x[["runtimes"]]

  if (ncol(df) == 2) {

    var_name <- setdiff(names(df), "elapsed")
    p <- ggplot2::ggplot(df, ggplot2::aes_string(x = var_name, y = "elapsed")) +
           ggplot2::geom_line() +
           ggplot2::labs(title = paste("Complexity of", x[["function_name"]]),
                         x = var_name,
                         y = "Runtime (s)")

    return(p)

  } else if (ncol(df) == 3) {

    var_names <- setdiff(names(df), "elapsed")
    p <- ggplot2::ggplot(df, ggplot2::aes_string(x = var_names[[1]], y = var_names[[2]], z = "elapsed")) +
           ggplot2::geom_contour() +
           ggplot2::labs(title = paste("Complexity of", x[["function_name"]]),
                         x = var_names[[1]],
                         y = var_names[[2]],
                         z = "Runtime (s)")

    return(p)

  } else {

    stop("Cannot plot complexity when there are more than two variables.",
         call. = FALSE)

  }

}

###