Completed #41

R/create_report.r

@@ -1,57 +1,128 @@
 #' create_report Function
 #'
-#' This function generates the report of data profiling.
+#' This function creates a data profiling report.
 #' @param data input data
 #' @param output_file output file name. The default is "report.html".
 #' @param output_dir output directory for report. The default is user's current directory.
+#' @param y name of response variable if any. Response variables will be passed to appropriate plotting functions automatically.
+#' @param config report configuration with function arguments as \link{list}. See details.
 #' @param \dots other arguments to be passed to \link{render}.
 #' @keywords create_report
 #' @aliases GenerateReport
+#' @details \code{config} is a named list to be evaluated by \code{create_report}.
+#' Each name should exactly match a function name.
+#' By doing so, that function and corresponding content will be added to the report.
+#' If you do not want to include certain functions/content, do not add it to \code{config}.
+#' @details By default, there is a preset \code{config} object (refer to example).
+#' In case you would like to customize the report, copy and edit the code and pass it to \code{config} argument.
+#' @details All function arguments will be passed to \link{do.call} as a list.
+#' @note If there are multiple options for the same function, all of them will be plotted.
+#' For example, \code{create_report(..., y = "a", config = list("plot_bar" = list("with" = "b")))} will create 3 bar charts:
+#' \itemize{
+#' \item regular frequency bar chart
+#' \item bar chart aggregated by response variable "a"
+#' \item bar chart aggregated by `with` variable "b"`
+#' }
 #' @importFrom utils browseURL
 #' @importFrom rmarkdown render
 #' @export create_report GenerateReport
 #' @examples
 #' \dontrun{
-#' # load library
+#' #############################
+#' ## Default config file     ##
+#' ## Copy and edit if needed ##
+#' #############################
+#' config <- list(
+#'   "introduce" = list(),
+#'   "plot_str" = list(
+#'     "type" = "diagonal",
+#'     "fontSize" = 35,
+#'     "width" = 1000,
+#'     "margin" = list("left" = 350, "right" = 250)
+#'   ),
+#'   "plot_missing" = list(),
+#'   "plot_histogram" = list(),
+#'   "plot_bar" = list(),
+#'   "plot_correlation" = list("use" = "pairwise.complete.obs"),
+#'   "plot_prcomp" = list(),
+#'   "plot_boxplot" = list(),
+#'   "plot_scatterplot" = list()
+#' )
+#'
+#' # Create report
+#' create_report(iris)
+#' create_report(airquality, y = "Ozone")
+#'
+#' # Load library
 #' library(ggplot2)
 #' library(data.table)
-#'
-#' # load diamonds dataset from ggplot2
 #' data("diamonds", package = "ggplot2")
-#' diamonds2 <- data.table(diamonds)
 #'
-#' # manually set some missing values
+#' # Set some missing values
+#' diamonds2 <- data.table(diamonds)
 #' for (j in 5:ncol(diamonds2)) {
 #'   set(diamonds2,
 #'       i = sample.int(nrow(diamonds2), sample.int(nrow(diamonds2), 1)),
 #'       j,
-#'       value = NA_integer_)}
+#'       value = NA_integer_)
+#' }
 #'
-#' # generate report for diamonds2 dataset
-#' create_report(diamonds2,
-#'                output_file = "report.html",
-#'                output_dir = getwd(),
-#'                html_document(toc = TRUE, toc_depth = 6, theme = "flatly"))
+#' # Create customized report for diamonds2 dataset
+#' create_report(
+#'   data = diamonds2,
+#'   output_file = "report.html",
+#'   output_dir = getwd(),
+#'   y = "price",
+#'   config = list(
+#'     "introduce" = list(),
+#'     "plot_missing" = list(),
+#'     "plot_histogram" = list(),
+#'     "plot_density" = list(),
+#'     "plot_bar" = list("with" = "carat"),
+#'     "plot_correlation" = list("use" = "pairwise.complete.obs"),
+#'     "plot_prcomp" = list(),
+#'     "plot_boxplot" = list("by" = "carat"),
+#'     "plot_scatterplot" = list("by" = "carat")
+#'   ),
+#'   html_document(toc = TRUE, toc_depth = 6, theme = "flatly")
+#' )
 #' }
 
-create_report <- function(data, output_file = "report.html", output_dir = getwd(), ...) {
-  ## Get argument list
-  args <- as.list(match.call())
+create_report <- function(data, output_file = "report.html", output_dir = getwd(), y = NULL, config = list(), ...) {
+  ## Check response variable
+  if (!is.null(y)) {
+    if (!(y %in% names(data))) stop("`", y, "` not found in data!")
+  }
   ## Get directory of report markdown template
   report_dir <- system.file("rmd_template/report.rmd", package = "DataExplorer")
+  ## Set report configuration if null
+  if (length(config) == 0) {
+    config <- list(
+      "introduce" = list(),
+      "plot_str" = list("type" = "diagonal", "fontSize" = 35, "width" = 1000, "margin" = list("left" = 350, "right" = 250)),
+      "plot_missing" = list(),
+      "plot_histogram" = list(),
+      "plot_bar" = list(),
+      "plot_correlation" = list("use" = "pairwise.complete.obs"),
+      "plot_prcomp" = list(),
+      "plot_boxplot" = list(),
+      "plot_scatterplot" = list()
+    )
+  }
   ## Render report into html
   suppressWarnings(render(
     input = report_dir,
     output_file = output_file,
     output_dir = output_dir,
     intermediates_dir = output_dir,
-    params = list(data = data, fun_options = list()),
+    params = list(data = data, report_config = config, response = y),
     ...
   ))
   ## Open report
   report_path <- file.path(output_dir, output_file)
   browseURL(report_path)
   ## Print report directory
+  args <- as.list(match.call())
   if (ifelse(is.null(args[["quiet"]]), TRUE, !args[["quiet"]])) message(paste0("\n\nReport is generated at \"", report_path, "\"."))
 }
 

R/plot_bar.r

@@ -75,7 +75,7 @@ plot_bar <- function(data, with = NULL, maxcat = 50, order_bar = TRUE, title = N
         } else {
           if (!is.numeric(data[[with]])) stop("`with` should be continuous!")
           x <- data.table(subset_data[, j, with = FALSE], "agg_by" = data[[with]])
-          agg_x <- x[, list(frequency = sum(agg_by)), by = eval(names(x)[1])]
+          agg_x <- x[, list(frequency = sum(agg_by, na.rm = TRUE)), by = eval(names(x)[1])]
         }
         if (order_bar) {
           base_plot <- ggplot(agg_x, aes(x = reorder(get(names(agg_x)[1]), frequency), y = frequency))

R/plot_prcomp.r

@@ -2,7 +2,7 @@
 #'
 #' Visualize output of \link{prcomp}.
 #' @param data input data
-#' @param variance_cap Maximum cumulative explained variance allowed for all principle components. Default is 80\%.
+#' @param variance_cap maximum cumulative explained variance allowed for all principle components. Default is 80\%.
 #' @param maxcat maximum categories allowed for each discrete feature. The default is 50.
 #' @param title plot title starting from page 2.
 #' @param ggtheme complete ggplot2 themes. The default is \link{theme_gray}.

R/plot_str.r

@@ -35,7 +35,7 @@
 #' plot_str(obj, type = "d", max_level = 2)
 #' }
 
-plot_str <- function(data, type = c("diagonal", "radial"), max_level, print_network = TRUE, ...) {
+plot_str <- function(data, type = c("diagonal", "radial"), max_level = NULL, print_network = TRUE, ...) {
   ## Declare variable first to pass R CMD check
   i <- idx <- parent <- NULL
   ## Capture str output
@@ -45,7 +45,7 @@ plot_str <- function(data, type = c("diagonal", "radial"), max_level, print_netw
   base_split <- tstrsplit(str_output[2:n], "\\$")
   nest_level <- (nchar(base_split[[1]]) - nchar(gsub("\ \\.{2}", "", base_split[[1]]))) / 3 + 1
   ## Handle max_level if exists
-  if (missing(max_level)) {
+  if (is.null(max_level)) {
     max_level <- max(nest_level)
   } else if (max_level <= 0 | max_level > max(nest_level)) {
     stop(paste0("max_level should be between 1 and ", max(nest_level)))