| 1 |
#' `teal` module: Distribution analysis |
|
| 2 |
#' |
|
| 3 |
#' Module is designed to explore the distribution of a single variable within a given dataset. |
|
| 4 |
#' It offers several tools, such as histograms, Q-Q plots, and various statistical tests to |
|
| 5 |
#' visually and statistically analyze the variable's distribution. |
|
| 6 |
#' |
|
| 7 |
#' @inheritParams teal::module |
|
| 8 |
#' @inheritParams teal.widgets::standard_layout |
|
| 9 |
#' @inheritParams shared_params |
|
| 10 |
#' |
|
| 11 |
#' @param dist_var (`data_extract_spec` or `list` of multiple `data_extract_spec`) |
|
| 12 |
#' Variable(s) for which the distribution will be analyzed. |
|
| 13 |
#' @param strata_var (`data_extract_spec` or `list` of multiple `data_extract_spec`) |
|
| 14 |
#' Categorical variable used to split the distribution analysis. |
|
| 15 |
#' @param group_var (`data_extract_spec` or `list` of multiple `data_extract_spec`) |
|
| 16 |
#' Variable used for faceting plot into multiple panels. |
|
| 17 |
#' @param freq (`logical`) optional, whether to display frequency (`TRUE`) or density (`FALSE`). |
|
| 18 |
#' Defaults to density (`FALSE`). |
|
| 19 |
#' @param bins (`integer(1)` or `integer(3)`) optional, specifies the number of bins for the histogram. |
|
| 20 |
#' - When the length of `bins` is one: The histogram bins will have a fixed size based on the `bins` provided. |
|
| 21 |
#' - When the length of `bins` is three: The histogram bins are dynamically adjusted based on vector of `value`, `min`, |
|
| 22 |
#' and `max`. |
|
| 23 |
#' Defaults to `c(30L, 1L, 100L)`. |
|
| 24 |
#' |
|
| 25 |
#' @templateVar ggnames "Histogram", "QQplot" |
|
| 26 |
#' @template ggplot2_args_multi |
|
| 27 |
#' |
|
| 28 |
#' @inherit shared_params return |
|
| 29 |
#' |
|
| 30 |
#' @examplesShinylive |
|
| 31 |
#' library(teal.modules.general) |
|
| 32 |
#' interactive <- function() TRUE |
|
| 33 |
#' {{ next_example }}
|
|
| 34 |
# nolint start: line_length_linter. |
|
| 35 |
#' @examplesIf require("ggpmisc", quietly = TRUE) && require("ggpp", quietly = TRUE) && require("goftest", quietly = TRUE) && require("MASS", quietly = TRUE) && require("broom", quietly = TRUE)
|
|
| 36 |
# nolint end: line_length_linter. |
|
| 37 |
#' # general data example |
|
| 38 |
#' data <- teal_data() |
|
| 39 |
#' data <- within(data, {
|
|
| 40 |
#' iris <- iris |
|
| 41 |
#' }) |
|
| 42 |
#' datanames(data) <- "iris" |
|
| 43 |
#' |
|
| 44 |
#' app <- init( |
|
| 45 |
#' data = data, |
|
| 46 |
#' modules = list( |
|
| 47 |
#' tm_g_distribution( |
|
| 48 |
#' dist_var = data_extract_spec( |
|
| 49 |
#' dataname = "iris", |
|
| 50 |
#' select = select_spec(variable_choices("iris"), "Petal.Length")
|
|
| 51 |
#' ) |
|
| 52 |
#' ) |
|
| 53 |
#' ) |
|
| 54 |
#' ) |
|
| 55 |
#' if (interactive()) {
|
|
| 56 |
#' shinyApp(app$ui, app$server) |
|
| 57 |
#' } |
|
| 58 |
#' |
|
| 59 |
#' @examplesShinylive |
|
| 60 |
#' library(teal.modules.general) |
|
| 61 |
#' interactive <- function() TRUE |
|
| 62 |
#' {{ next_example }}
|
|
| 63 |
# nolint start: line_length_linter. |
|
| 64 |
#' @examplesIf require("ggpmisc", quietly = TRUE) && require("ggpp", quietly = TRUE) && require("goftest", quietly = TRUE) && require("MASS", quietly = TRUE) && require("broom", quietly = TRUE)
|
|
| 65 |
# nolint end: line_length_linter. |
|
| 66 |
#' # CDISC data example |
|
| 67 |
#' data <- teal_data() |
|
| 68 |
#' data <- within(data, {
|
|
| 69 |
#' ADSL <- rADSL |
|
| 70 |
#' }) |
|
| 71 |
#' datanames(data) <- c("ADSL")
|
|
| 72 |
#' join_keys(data) <- default_cdisc_join_keys[datanames(data)] |
|
| 73 |
#' |
|
| 74 |
#' vars1 <- choices_selected( |
|
| 75 |
#' variable_choices(data[["ADSL"]], c("ARM", "COUNTRY", "SEX")),
|
|
| 76 |
#' selected = NULL |
|
| 77 |
#' ) |
|
| 78 |
#' |
|
| 79 |
#' app <- init( |
|
| 80 |
#' data = data, |
|
| 81 |
#' modules = modules( |
|
| 82 |
#' tm_g_distribution( |
|
| 83 |
#' dist_var = data_extract_spec( |
|
| 84 |
#' dataname = "ADSL", |
|
| 85 |
#' select = select_spec( |
|
| 86 |
#' choices = variable_choices(data[["ADSL"]], c("AGE", "BMRKR1")),
|
|
| 87 |
#' selected = "BMRKR1", |
|
| 88 |
#' multiple = FALSE, |
|
| 89 |
#' fixed = FALSE |
|
| 90 |
#' ) |
|
| 91 |
#' ), |
|
| 92 |
#' strata_var = data_extract_spec( |
|
| 93 |
#' dataname = "ADSL", |
|
| 94 |
#' filter = filter_spec( |
|
| 95 |
#' vars = vars1, |
|
| 96 |
#' multiple = TRUE |
|
| 97 |
#' ) |
|
| 98 |
#' ), |
|
| 99 |
#' group_var = data_extract_spec( |
|
| 100 |
#' dataname = "ADSL", |
|
| 101 |
#' filter = filter_spec( |
|
| 102 |
#' vars = vars1, |
|
| 103 |
#' multiple = TRUE |
|
| 104 |
#' ) |
|
| 105 |
#' ) |
|
| 106 |
#' ) |
|
| 107 |
#' ) |
|
| 108 |
#' ) |
|
| 109 |
#' if (interactive()) {
|
|
| 110 |
#' shinyApp(app$ui, app$server) |
|
| 111 |
#' } |
|
| 112 |
#' |
|
| 113 |
#' @export |
|
| 114 |
#' |
|
| 115 |
tm_g_distribution <- function(label = "Distribution Module", |
|
| 116 |
dist_var, |
|
| 117 |
strata_var = NULL, |
|
| 118 |
group_var = NULL, |
|
| 119 |
freq = FALSE, |
|
| 120 |
ggtheme = c("gray", "bw", "linedraw", "light", "dark", "minimal", "classic", "void"),
|
|
| 121 |
ggplot2_args = teal.widgets::ggplot2_args(), |
|
| 122 |
bins = c(30L, 1L, 100L), |
|
| 123 |
plot_height = c(600, 200, 2000), |
|
| 124 |
plot_width = NULL, |
|
| 125 |
pre_output = NULL, |
|
| 126 |
post_output = NULL) {
|
|
| 127 | ! |
message("Initializing tm_g_distribution")
|
| 128 | ||
| 129 |
# Requires Suggested packages |
|
| 130 | ! |
extra_packages <- c("ggpmisc", "ggpp", "goftest", "MASS", "broom")
|
| 131 | ! |
missing_packages <- Filter(function(x) !requireNamespace(x, quietly = TRUE), extra_packages) |
| 132 | ! |
if (length(missing_packages) > 0L) {
|
| 133 | ! |
stop(sprintf( |
| 134 | ! |
"Cannot load package(s): %s.\nInstall or restart your session.", |
| 135 | ! |
toString(missing_packages) |
| 136 |
)) |
|
| 137 |
} |
|
| 138 | ||
| 139 |
# Normalize the parameters |
|
| 140 | ! |
if (inherits(dist_var, "data_extract_spec")) dist_var <- list(dist_var) |
| 141 | ! |
if (inherits(strata_var, "data_extract_spec")) strata_var <- list(strata_var) |
| 142 | ! |
if (inherits(group_var, "data_extract_spec")) group_var <- list(group_var) |
| 143 | ! |
if (inherits(ggplot2_args, "ggplot2_args")) ggplot2_args <- list(default = ggplot2_args) |
| 144 | ||
| 145 |
# Start of assertions |
|
| 146 | ! |
checkmate::assert_string(label) |
| 147 | ||
| 148 | ! |
checkmate::assert_list(dist_var, "data_extract_spec") |
| 149 | ! |
checkmate::assert_false(dist_var[[1L]]$select$multiple) |
| 150 | ||
| 151 | ! |
checkmate::assert_list(strata_var, types = "data_extract_spec", null.ok = TRUE) |
| 152 | ! |
checkmate::assert_list(group_var, types = "data_extract_spec", null.ok = TRUE) |
| 153 | ! |
checkmate::assert_flag(freq) |
| 154 | ! |
ggtheme <- match.arg(ggtheme) |
| 155 | ||
| 156 | ! |
plot_choices <- c("Histogram", "QQplot")
|
| 157 | ! |
checkmate::assert_list(ggplot2_args, types = "ggplot2_args") |
| 158 | ! |
checkmate::assert_subset(names(ggplot2_args), c("default", plot_choices))
|
| 159 | ||
| 160 | ! |
if (length(bins) == 1) {
|
| 161 | ! |
checkmate::assert_numeric(bins, any.missing = FALSE, lower = 1) |
| 162 |
} else {
|
|
| 163 | ! |
checkmate::assert_numeric(bins, len = 3, any.missing = FALSE, lower = 1) |
| 164 | ! |
checkmate::assert_numeric(bins[1], lower = bins[2], upper = bins[3], .var.name = "bins") |
| 165 |
} |
|
| 166 | ||
| 167 | ! |
checkmate::assert_numeric(plot_height, len = 3, any.missing = FALSE, finite = TRUE) |
| 168 | ! |
checkmate::assert_numeric(plot_height[1], lower = plot_height[2], upper = plot_height[3], .var.name = "plot_height") |
| 169 | ! |
checkmate::assert_numeric(plot_width, len = 3, any.missing = FALSE, null.ok = TRUE, finite = TRUE) |
| 170 | ! |
checkmate::assert_numeric( |
| 171 | ! |
plot_width[1], |
| 172 | ! |
lower = plot_width[2], upper = plot_width[3], null.ok = TRUE, .var.name = "plot_width" |
| 173 |
) |
|
| 174 | ||
| 175 | ! |
checkmate::assert_multi_class(pre_output, c("shiny.tag", "shiny.tag.list", "html"), null.ok = TRUE)
|
| 176 | ! |
checkmate::assert_multi_class(post_output, c("shiny.tag", "shiny.tag.list", "html"), null.ok = TRUE)
|
| 177 |
# End of assertions |
|
| 178 | ||
| 179 |
# Make UI args |
|
| 180 | ! |
args <- as.list(environment()) |
| 181 | ||
| 182 | ! |
data_extract_list <- list( |
| 183 | ! |
dist_var = dist_var, |
| 184 | ! |
strata_var = strata_var, |
| 185 | ! |
group_var = group_var |
| 186 |
) |
|
| 187 | ||
| 188 | ! |
ans <- module( |
| 189 | ! |
label = label, |
| 190 | ! |
server = srv_distribution, |
| 191 | ! |
server_args = c( |
| 192 | ! |
data_extract_list, |
| 193 | ! |
list(plot_height = plot_height, plot_width = plot_width, ggplot2_args = ggplot2_args) |
| 194 |
), |
|
| 195 | ! |
ui = ui_distribution, |
| 196 | ! |
ui_args = args, |
| 197 | ! |
datanames = teal.transform::get_extract_datanames(data_extract_list) |
| 198 |
) |
|
| 199 | ! |
attr(ans, "teal_bookmarkable") <- TRUE |
| 200 | ! |
ans |
| 201 |
} |
|
| 202 | ||
| 203 |
# UI function for the distribution module |
|
| 204 |
ui_distribution <- function(id, ...) {
|
|
| 205 | ! |
args <- list(...) |
| 206 | ! |
ns <- NS(id) |
| 207 | ! |
is_single_dataset_value <- teal.transform::is_single_dataset(args$dist_var, args$strata_var, args$group_var) |
| 208 | ||
| 209 | ! |
teal.widgets::standard_layout( |
| 210 | ! |
output = teal.widgets::white_small_well( |
| 211 | ! |
tabsetPanel( |
| 212 | ! |
id = ns("tabs"),
|
| 213 | ! |
tabPanel("Histogram", teal.widgets::plot_with_settings_ui(id = ns("hist_plot"))),
|
| 214 | ! |
tabPanel("QQplot", teal.widgets::plot_with_settings_ui(id = ns("qq_plot")))
|
| 215 |
), |
|
| 216 | ! |
tags$h3("Statistics Table"),
|
| 217 | ! |
DT::dataTableOutput(ns("summary_table")),
|
| 218 | ! |
tags$h3("Tests"),
|
| 219 | ! |
DT::dataTableOutput(ns("t_stats"))
|
| 220 |
), |
|
| 221 | ! |
encoding = tags$div( |
| 222 |
### Reporter |
|
| 223 | ! |
teal.reporter::simple_reporter_ui(ns("simple_reporter")),
|
| 224 |
### |
|
| 225 | ! |
tags$label("Encodings", class = "text-primary"),
|
| 226 | ! |
teal.transform::datanames_input(args[c("dist_var", "strata_var")]),
|
| 227 | ! |
teal.transform::data_extract_ui( |
| 228 | ! |
id = ns("dist_i"),
|
| 229 | ! |
label = "Variable", |
| 230 | ! |
data_extract_spec = args$dist_var, |
| 231 | ! |
is_single_dataset = is_single_dataset_value |
| 232 |
), |
|
| 233 | ! |
if (!is.null(args$group_var)) {
|
| 234 | ! |
tagList( |
| 235 | ! |
teal.transform::data_extract_ui( |
| 236 | ! |
id = ns("group_i"),
|
| 237 | ! |
label = "Group by", |
| 238 | ! |
data_extract_spec = args$group_var, |
| 239 | ! |
is_single_dataset = is_single_dataset_value |
| 240 |
), |
|
| 241 | ! |
uiOutput(ns("scales_types_ui"))
|
| 242 |
) |
|
| 243 |
}, |
|
| 244 | ! |
if (!is.null(args$strata_var)) {
|
| 245 | ! |
teal.transform::data_extract_ui( |
| 246 | ! |
id = ns("strata_i"),
|
| 247 | ! |
label = "Stratify by", |
| 248 | ! |
data_extract_spec = args$strata_var, |
| 249 | ! |
is_single_dataset = is_single_dataset_value |
| 250 |
) |
|
| 251 |
}, |
|
| 252 | ! |
teal.widgets::panel_group( |
| 253 | ! |
conditionalPanel( |
| 254 | ! |
condition = paste0("input['", ns("tabs"), "'] == 'Histogram'"),
|
| 255 | ! |
teal.widgets::panel_item( |
| 256 | ! |
"Histogram", |
| 257 | ! |
teal.widgets::optionalSliderInputValMinMax(ns("bins"), "Bins", args$bins, ticks = FALSE, step = 1),
|
| 258 | ! |
shinyWidgets::prettyRadioButtons( |
| 259 | ! |
ns("main_type"),
|
| 260 | ! |
label = "Plot Type:", |
| 261 | ! |
choices = c("Density", "Frequency"),
|
| 262 | ! |
selected = if (!args$freq) "Density" else "Frequency", |
| 263 | ! |
bigger = FALSE, |
| 264 | ! |
inline = TRUE |
| 265 |
), |
|
| 266 | ! |
checkboxInput(ns("add_dens"), label = "Overlay Density", value = TRUE),
|
| 267 | ! |
collapsed = FALSE |
| 268 |
) |
|
| 269 |
), |
|
| 270 | ! |
conditionalPanel( |
| 271 | ! |
condition = paste0("input['", ns("tabs"), "'] == 'QQplot'"),
|
| 272 | ! |
teal.widgets::panel_item( |
| 273 | ! |
"QQ Plot", |
| 274 | ! |
checkboxInput(ns("qq_line"), label = "Add diagonal line(s)", TRUE),
|
| 275 | ! |
collapsed = FALSE |
| 276 |
) |
|
| 277 |
), |
|
| 278 | ! |
conditionalPanel( |
| 279 | ! |
condition = paste0("input['", ns("main_type"), "'] == 'Density'"),
|
| 280 | ! |
teal.widgets::panel_item( |
| 281 | ! |
"Theoretical Distribution", |
| 282 | ! |
teal.widgets::optionalSelectInput( |
| 283 | ! |
ns("t_dist"),
|
| 284 | ! |
tags$div( |
| 285 | ! |
class = "teal-tooltip", |
| 286 | ! |
tagList( |
| 287 | ! |
"Distribution:", |
| 288 | ! |
icon("circle-info"),
|
| 289 | ! |
tags$span( |
| 290 | ! |
class = "tooltiptext", |
| 291 | ! |
"Default parameters are optimized with MASS::fitdistr function." |
| 292 |
) |
|
| 293 |
) |
|
| 294 |
), |
|
| 295 | ! |
choices = c("normal", "lognormal", "gamma", "unif"),
|
| 296 | ! |
selected = NULL, |
| 297 | ! |
multiple = FALSE |
| 298 |
), |
|
| 299 | ! |
numericInput(ns("dist_param1"), label = "param1", value = NULL),
|
| 300 | ! |
numericInput(ns("dist_param2"), label = "param2", value = NULL),
|
| 301 | ! |
tags$span(actionButton(ns("params_reset"), "Default params")),
|
| 302 | ! |
collapsed = FALSE |
| 303 |
) |
|
| 304 |
) |
|
| 305 |
), |
|
| 306 | ! |
teal.widgets::panel_item( |
| 307 | ! |
"Tests", |
| 308 | ! |
teal.widgets::optionalSelectInput( |
| 309 | ! |
ns("dist_tests"),
|
| 310 | ! |
"Tests:", |
| 311 | ! |
choices = c( |
| 312 | ! |
"Shapiro-Wilk", |
| 313 | ! |
if (!is.null(args$strata_var)) "t-test (two-samples, not paired)", |
| 314 | ! |
if (!is.null(args$strata_var)) "one-way ANOVA", |
| 315 | ! |
if (!is.null(args$strata_var)) "Fligner-Killeen", |
| 316 | ! |
if (!is.null(args$strata_var)) "F-test", |
| 317 | ! |
"Kolmogorov-Smirnov (one-sample)", |
| 318 | ! |
"Anderson-Darling (one-sample)", |
| 319 | ! |
"Cramer-von Mises (one-sample)", |
| 320 | ! |
if (!is.null(args$strata_var)) "Kolmogorov-Smirnov (two-samples)" |
| 321 |
), |
|
| 322 | ! |
selected = NULL |
| 323 |
) |
|
| 324 |
), |
|
| 325 | ! |
teal.widgets::panel_item( |
| 326 | ! |
"Statistics Table", |
| 327 | ! |
sliderInput(ns("roundn"), "Round to n digits", min = 0, max = 10, value = 2)
|
| 328 |
), |
|
| 329 | ! |
teal.widgets::panel_item( |
| 330 | ! |
title = "Plot settings", |
| 331 | ! |
selectInput( |
| 332 | ! |
inputId = ns("ggtheme"),
|
| 333 | ! |
label = "Theme (by ggplot):", |
| 334 | ! |
choices = ggplot_themes, |
| 335 | ! |
selected = args$ggtheme, |
| 336 | ! |
multiple = FALSE |
| 337 |
) |
|
| 338 |
) |
|
| 339 |
), |
|
| 340 | ! |
forms = tagList( |
| 341 | ! |
teal.widgets::verbatim_popup_ui(ns("rcode"), "Show R code")
|
| 342 |
), |
|
| 343 | ! |
pre_output = args$pre_output, |
| 344 | ! |
post_output = args$post_output |
| 345 |
) |
|
| 346 |
} |
|
| 347 | ||
| 348 |
# Server function for the distribution module |
|
| 349 |
srv_distribution <- function(id, |
|
| 350 |
data, |
|
| 351 |
reporter, |
|
| 352 |
filter_panel_api, |
|
| 353 |
dist_var, |
|
| 354 |
strata_var, |
|
| 355 |
group_var, |
|
| 356 |
plot_height, |
|
| 357 |
plot_width, |
|
| 358 |
ggplot2_args) {
|
|
| 359 | ! |
with_reporter <- !missing(reporter) && inherits(reporter, "Reporter") |
| 360 | ! |
with_filter <- !missing(filter_panel_api) && inherits(filter_panel_api, "FilterPanelAPI") |
| 361 | ! |
checkmate::assert_class(data, "reactive") |
| 362 | ! |
checkmate::assert_class(isolate(data()), "teal_data") |
| 363 | ! |
moduleServer(id, function(input, output, session) {
|
| 364 | ! |
teal.logger::log_shiny_input_changes(input, namespace = "teal.modules.general") |
| 365 | ||
| 366 | ! |
setBookmarkExclude("params_reset")
|
| 367 | ||
| 368 | ! |
ns <- session$ns |
| 369 | ||
| 370 | ! |
rule_req <- function(value) {
|
| 371 | ! |
if (isTRUE(input$dist_tests %in% c( |
| 372 | ! |
"Fligner-Killeen", |
| 373 | ! |
"t-test (two-samples, not paired)", |
| 374 | ! |
"F-test", |
| 375 | ! |
"Kolmogorov-Smirnov (two-samples)", |
| 376 | ! |
"one-way ANOVA" |
| 377 |
))) {
|
|
| 378 | ! |
if (!shinyvalidate::input_provided(value)) {
|
| 379 | ! |
"Please select stratify variable." |
| 380 |
} |
|
| 381 |
} |
|
| 382 |
} |
|
| 383 | ! |
rule_dupl <- function(...) {
|
| 384 | ! |
if (identical(input$dist_tests, "Fligner-Killeen")) {
|
| 385 | ! |
strata <- selector_list()$strata_i()$select |
| 386 | ! |
group <- selector_list()$group_i()$select |
| 387 | ! |
if (isTRUE(strata == group)) {
|
| 388 | ! |
"Please select different variables for strata and group." |
| 389 |
} |
|
| 390 |
} |
|
| 391 |
} |
|
| 392 | ||
| 393 | ! |
selector_list <- teal.transform::data_extract_multiple_srv( |
| 394 | ! |
data_extract = list( |
| 395 | ! |
dist_i = dist_var, |
| 396 | ! |
strata_i = strata_var, |
| 397 | ! |
group_i = group_var |
| 398 |
), |
|
| 399 | ! |
data, |
| 400 | ! |
select_validation_rule = list( |
| 401 | ! |
dist_i = shinyvalidate::sv_required("Please select a variable")
|
| 402 |
), |
|
| 403 | ! |
filter_validation_rule = list( |
| 404 | ! |
strata_i = shinyvalidate::compose_rules( |
| 405 | ! |
rule_req, |
| 406 | ! |
rule_dupl |
| 407 |
), |
|
| 408 | ! |
group_i = rule_dupl |
| 409 |
) |
|
| 410 |
) |
|
| 411 | ||
| 412 | ! |
iv_r <- reactive({
|
| 413 | ! |
iv <- shinyvalidate::InputValidator$new() |
| 414 | ! |
teal.transform::compose_and_enable_validators(iv, selector_list, validator_names = "dist_i") |
| 415 |
}) |
|
| 416 | ||
| 417 | ! |
iv_r_dist <- reactive({
|
| 418 | ! |
iv <- shinyvalidate::InputValidator$new() |
| 419 | ! |
teal.transform::compose_and_enable_validators( |
| 420 | ! |
iv, selector_list, |
| 421 | ! |
validator_names = c("strata_i", "group_i")
|
| 422 |
) |
|
| 423 |
}) |
|
| 424 | ! |
rule_dist_1 <- function(value) {
|
| 425 | ! |
if (!is.null(input$t_dist)) {
|
| 426 | ! |
switch(input$t_dist, |
| 427 | ! |
"normal" = if (!shinyvalidate::input_provided(value)) "mean is required", |
| 428 | ! |
"lognormal" = if (!shinyvalidate::input_provided(value)) "meanlog is required", |
| 429 | ! |
"gamma" = {
|
| 430 | ! |
if (!shinyvalidate::input_provided(value)) "shape is required" else if (value <= 0) "shape must be positive" |
| 431 |
}, |
|
| 432 | ! |
"unif" = NULL |
| 433 |
) |
|
| 434 |
} |
|
| 435 |
} |
|
| 436 | ! |
rule_dist_2 <- function(value) {
|
| 437 | ! |
if (!is.null(input$t_dist)) {
|
| 438 | ! |
switch(input$t_dist, |
| 439 | ! |
"normal" = {
|
| 440 | ! |
if (!shinyvalidate::input_provided(value)) {
|
| 441 | ! |
"sd is required" |
| 442 | ! |
} else if (value < 0) {
|
| 443 | ! |
"sd must be non-negative" |
| 444 |
} |
|
| 445 |
}, |
|
| 446 | ! |
"lognormal" = {
|
| 447 | ! |
if (!shinyvalidate::input_provided(value)) {
|
| 448 | ! |
"sdlog is required" |
| 449 | ! |
} else if (value < 0) {
|
| 450 | ! |
"sdlog must be non-negative" |
| 451 |
} |
|
| 452 |
}, |
|
| 453 | ! |
"gamma" = {
|
| 454 | ! |
if (!shinyvalidate::input_provided(value)) {
|
| 455 | ! |
"rate is required" |
| 456 | ! |
} else if (value <= 0) {
|
| 457 | ! |
"rate must be positive" |
| 458 |
} |
|
| 459 |
}, |
|
| 460 | ! |
"unif" = NULL |
| 461 |
) |
|
| 462 |
} |
|
| 463 |
} |
|
| 464 | ! |
rule_dist <- function(value) {
|
| 465 | ! |
if (isTRUE(input$tabs == "QQplot" || |
| 466 | ! |
input$dist_tests %in% c( |
| 467 | ! |
"Kolmogorov-Smirnov (one-sample)", |
| 468 | ! |
"Anderson-Darling (one-sample)", |
| 469 | ! |
"Cramer-von Mises (one-sample)" |
| 470 |
))) {
|
|
| 471 | ! |
if (!shinyvalidate::input_provided(value)) {
|
| 472 | ! |
"Please select the theoretical distribution." |
| 473 |
} |
|
| 474 |
} |
|
| 475 |
} |
|
| 476 | ! |
iv_dist <- shinyvalidate::InputValidator$new() |
| 477 | ! |
iv_dist$add_rule("t_dist", rule_dist)
|
| 478 | ! |
iv_dist$add_rule("dist_param1", rule_dist_1)
|
| 479 | ! |
iv_dist$add_rule("dist_param2", rule_dist_2)
|
| 480 | ! |
iv_dist$enable() |
| 481 | ||
| 482 | ! |
anl_merged_input <- teal.transform::merge_expression_srv( |
| 483 | ! |
selector_list = selector_list, |
| 484 | ! |
datasets = data |
| 485 |
) |
|
| 486 | ||
| 487 | ! |
anl_merged_q <- reactive({
|
| 488 | ! |
req(anl_merged_input()) |
| 489 | ! |
data() %>% |
| 490 | ! |
teal.code::eval_code(as.expression(anl_merged_input()$expr)) |
| 491 |
}) |
|
| 492 | ||
| 493 | ! |
merged <- list( |
| 494 | ! |
anl_input_r = anl_merged_input, |
| 495 | ! |
anl_q_r = anl_merged_q |
| 496 |
) |
|
| 497 | ||
| 498 | ! |
output$scales_types_ui <- renderUI({
|
| 499 | ! |
if ("group_i" %in% names(selector_list()) && length(selector_list()$group_i()$filters[[1]]$selected) > 0) {
|
| 500 | ! |
shinyWidgets::prettyRadioButtons( |
| 501 | ! |
ns("scales_type"),
|
| 502 | ! |
label = "Scales:", |
| 503 | ! |
choices = c("Fixed", "Free"),
|
| 504 | ! |
selected = "Fixed", |
| 505 | ! |
bigger = FALSE, |
| 506 | ! |
inline = TRUE |
| 507 |
) |
|
| 508 |
} |
|
| 509 |
}) |
|
| 510 | ||
| 511 | ! |
observeEvent( |
| 512 | ! |
eventExpr = list( |
| 513 | ! |
input$t_dist, |
| 514 | ! |
input$params_reset, |
| 515 | ! |
selector_list()$dist_i()$select |
| 516 |
), |
|
| 517 | ! |
handlerExpr = {
|
| 518 | ! |
params <- |
| 519 | ! |
if (length(input$t_dist) != 0) {
|
| 520 | ! |
get_dist_params <- function(x, dist) {
|
| 521 | ! |
if (dist == "unif") {
|
| 522 | ! |
return(stats::setNames(range(x, na.rm = TRUE), c("min", "max")))
|
| 523 |
} |
|
| 524 | ! |
tryCatch( |
| 525 | ! |
MASS::fitdistr(x, densfun = dist)$estimate, |
| 526 | ! |
error = function(e) c(param1 = NA_real_, param2 = NA_real_) |
| 527 |
) |
|
| 528 |
} |
|
| 529 | ||
| 530 | ! |
ANL <- merged$anl_q_r()[["ANL"]] |
| 531 | ! |
round(get_dist_params(as.numeric(stats::na.omit(ANL[[merge_vars()$dist_var]])), input$t_dist), 2) |
| 532 |
} else {
|
|
| 533 | ! |
c("param1" = NA_real_, "param2" = NA_real_)
|
| 534 |
} |
|
| 535 | ||
| 536 | ! |
params_vals <- unname(params) |
| 537 | ! |
params_names <- names(params) |
| 538 | ||
| 539 | ! |
updateNumericInput( |
| 540 | ! |
inputId = "dist_param1", |
| 541 | ! |
label = params_names[1], |
| 542 | ! |
value = restoreInput(ns("dist_param1"), params_vals[1])
|
| 543 |
) |
|
| 544 | ! |
updateNumericInput( |
| 545 | ! |
inputId = "dist_param2", |
| 546 | ! |
label = params_names[2], |
| 547 | ! |
value = restoreInput(ns("dist_param1"), params_vals[2])
|
| 548 |
) |
|
| 549 |
}, |
|
| 550 | ! |
ignoreInit = TRUE |
| 551 |
) |
|
| 552 | ||
| 553 | ! |
observeEvent(input$params_reset, {
|
| 554 | ! |
updateActionButton(inputId = "params_reset", label = "Reset params") |
| 555 |
}) |
|
| 556 | ||
| 557 | ! |
merge_vars <- reactive({
|
| 558 | ! |
teal::validate_inputs(iv_r()) |
| 559 | ||
| 560 | ! |
dist_var <- as.vector(merged$anl_input_r()$columns_source$dist_i) |
| 561 | ! |
s_var <- as.vector(merged$anl_input_r()$columns_source$strata_i) |
| 562 | ! |
g_var <- as.vector(merged$anl_input_r()$columns_source$group_i) |
| 563 | ||
| 564 | ! |
dist_var_name <- if (length(dist_var)) as.name(dist_var) else NULL |
| 565 | ! |
s_var_name <- if (length(s_var)) as.name(s_var) else NULL |
| 566 | ! |
g_var_name <- if (length(g_var)) as.name(g_var) else NULL |
| 567 | ||
| 568 | ! |
list( |
| 569 | ! |
dist_var = dist_var, |
| 570 | ! |
s_var = s_var, |
| 571 | ! |
g_var = g_var, |
| 572 | ! |
dist_var_name = dist_var_name, |
| 573 | ! |
s_var_name = s_var_name, |
| 574 | ! |
g_var_name = g_var_name |
| 575 |
) |
|
| 576 |
}) |
|
| 577 | ||
| 578 |
# common qenv |
|
| 579 | ! |
common_q <- reactive({
|
| 580 |
# Create a private stack for this function only. |
|
| 581 | ||
| 582 | ! |
ANL <- merged$anl_q_r()[["ANL"]] |
| 583 | ! |
dist_var <- merge_vars()$dist_var |
| 584 | ! |
s_var <- merge_vars()$s_var |
| 585 | ! |
g_var <- merge_vars()$g_var |
| 586 | ||
| 587 | ! |
dist_var_name <- merge_vars()$dist_var_name |
| 588 | ! |
s_var_name <- merge_vars()$s_var_name |
| 589 | ! |
g_var_name <- merge_vars()$g_var_name |
| 590 | ||
| 591 | ! |
roundn <- input$roundn |
| 592 | ! |
dist_param1 <- input$dist_param1 |
| 593 | ! |
dist_param2 <- input$dist_param2 |
| 594 |
# isolated as dist_param1/dist_param2 already triggered the reactivity |
|
| 595 | ! |
t_dist <- isolate(input$t_dist) |
| 596 | ||
| 597 | ! |
qenv <- merged$anl_q_r() |
| 598 | ||
| 599 | ! |
if (length(g_var) > 0) {
|
| 600 | ! |
validate( |
| 601 | ! |
need( |
| 602 | ! |
inherits(ANL[[g_var]], c("integer", "factor", "character")),
|
| 603 | ! |
"Group by variable must be `factor`, `character`, or `integer`" |
| 604 |
) |
|
| 605 |
) |
|
| 606 | ! |
qenv <- teal.code::eval_code( |
| 607 | ! |
qenv, |
| 608 | ! |
substitute( |
| 609 | ! |
expr = ANL[[g_var]] <- forcats::fct_na_value_to_level(as.factor(ANL[[g_var]]), "NA"), |
| 610 | ! |
env = list(g_var = g_var) |
| 611 |
) |
|
| 612 |
) |
|
| 613 |
} |
|
| 614 | ||
| 615 | ! |
if (length(s_var) > 0) {
|
| 616 | ! |
validate( |
| 617 | ! |
need( |
| 618 | ! |
inherits(ANL[[s_var]], c("integer", "factor", "character")),
|
| 619 | ! |
"Stratify by variable must be `factor`, `character`, or `integer`" |
| 620 |
) |
|
| 621 |
) |
|
| 622 | ! |
qenv <- teal.code::eval_code( |
| 623 | ! |
qenv, |
| 624 | ! |
substitute( |
| 625 | ! |
expr = ANL[[s_var]] <- forcats::fct_na_value_to_level(as.factor(ANL[[s_var]]), "NA"), |
| 626 | ! |
env = list(s_var = s_var) |
| 627 |
) |
|
| 628 |
) |
|
| 629 |
} |
|
| 630 | ||
| 631 | ! |
validate(need(is.numeric(ANL[[dist_var]]), "Please select a numeric variable.")) |
| 632 | ! |
teal::validate_has_data(ANL, 1, complete = TRUE) |
| 633 | ||
| 634 | ! |
if (length(t_dist) != 0) {
|
| 635 | ! |
map_distr_nams <- list( |
| 636 | ! |
normal = c("mean", "sd"),
|
| 637 | ! |
lognormal = c("meanlog", "sdlog"),
|
| 638 | ! |
gamma = c("shape", "rate"),
|
| 639 | ! |
unif = c("min", "max")
|
| 640 |
) |
|
| 641 | ! |
params_names_raw <- map_distr_nams[[t_dist]] |
| 642 | ||
| 643 | ! |
qenv <- teal.code::eval_code( |
| 644 | ! |
qenv, |
| 645 | ! |
substitute( |
| 646 | ! |
expr = {
|
| 647 | ! |
params <- as.list(c(dist_param1, dist_param2)) |
| 648 | ! |
names(params) <- params_names_raw |
| 649 |
}, |
|
| 650 | ! |
env = list( |
| 651 | ! |
dist_param1 = dist_param1, |
| 652 | ! |
dist_param2 = dist_param2, |
| 653 | ! |
params_names_raw = params_names_raw |
| 654 |
) |
|
| 655 |
) |
|
| 656 |
) |
|
| 657 |
} |
|
| 658 | ||
| 659 | ! |
if (length(s_var) == 0 && length(g_var) == 0) {
|
| 660 | ! |
qenv <- teal.code::eval_code( |
| 661 | ! |
qenv, |
| 662 | ! |
substitute( |
| 663 | ! |
expr = {
|
| 664 | ! |
summary_table <- ANL %>% |
| 665 | ! |
dplyr::summarise( |
| 666 | ! |
min = round(min(dist_var_name, na.rm = TRUE), roundn), |
| 667 | ! |
median = round(stats::median(dist_var_name, na.rm = TRUE), roundn), |
| 668 | ! |
mean = round(mean(dist_var_name, na.rm = TRUE), roundn), |
| 669 | ! |
max = round(max(dist_var_name, na.rm = TRUE), roundn), |
| 670 | ! |
sd = round(stats::sd(dist_var_name, na.rm = TRUE), roundn), |
| 671 | ! |
count = dplyr::n() |
| 672 |
) |
|
| 673 |
}, |
|
| 674 | ! |
env = list( |
| 675 | ! |
dist_var_name = as.name(dist_var), |
| 676 | ! |
roundn = roundn |
| 677 |
) |
|
| 678 |
) |
|
| 679 |
) |
|
| 680 |
} else {
|
|
| 681 | ! |
qenv <- teal.code::eval_code( |
| 682 | ! |
qenv, |
| 683 | ! |
substitute( |
| 684 | ! |
expr = {
|
| 685 | ! |
strata_vars <- strata_vars_raw |
| 686 | ! |
summary_table <- ANL %>% |
| 687 | ! |
dplyr::group_by_at(dplyr::vars(dplyr::any_of(strata_vars))) %>% |
| 688 | ! |
dplyr::summarise( |
| 689 | ! |
min = round(min(dist_var_name, na.rm = TRUE), roundn), |
| 690 | ! |
median = round(stats::median(dist_var_name, na.rm = TRUE), roundn), |
| 691 | ! |
mean = round(mean(dist_var_name, na.rm = TRUE), roundn), |
| 692 | ! |
max = round(max(dist_var_name, na.rm = TRUE), roundn), |
| 693 | ! |
sd = round(stats::sd(dist_var_name, na.rm = TRUE), roundn), |
| 694 | ! |
count = dplyr::n() |
| 695 |
) |
|
| 696 | ! |
summary_table # used to display table when running show-r-code code |
| 697 |
}, |
|
| 698 | ! |
env = list( |
| 699 | ! |
dist_var_name = dist_var_name, |
| 700 | ! |
strata_vars_raw = c(g_var, s_var), |
| 701 | ! |
roundn = roundn |
| 702 |
) |
|
| 703 |
) |
|
| 704 |
) |
|
| 705 |
} |
|
| 706 |
}) |
|
| 707 | ||
| 708 |
# distplot qenv ---- |
|
| 709 | ! |
dist_q <- eventReactive( |
| 710 | ! |
eventExpr = {
|
| 711 | ! |
common_q() |
| 712 | ! |
input$scales_type |
| 713 | ! |
input$main_type |
| 714 | ! |
input$bins |
| 715 | ! |
input$add_dens |
| 716 | ! |
is.null(input$ggtheme) |
| 717 |
}, |
|
| 718 | ! |
valueExpr = {
|
| 719 | ! |
dist_var <- merge_vars()$dist_var |
| 720 | ! |
s_var <- merge_vars()$s_var |
| 721 | ! |
g_var <- merge_vars()$g_var |
| 722 | ! |
dist_var_name <- merge_vars()$dist_var_name |
| 723 | ! |
s_var_name <- merge_vars()$s_var_name |
| 724 | ! |
g_var_name <- merge_vars()$g_var_name |
| 725 | ! |
t_dist <- input$t_dist |
| 726 | ! |
dist_param1 <- input$dist_param1 |
| 727 | ! |
dist_param2 <- input$dist_param2 |
| 728 | ||
| 729 | ! |
scales_type <- input$scales_type |
| 730 | ||
| 731 | ! |
ndensity <- 512 |
| 732 | ! |
main_type_var <- input$main_type |
| 733 | ! |
bins_var <- input$bins |
| 734 | ! |
add_dens_var <- input$add_dens |
| 735 | ! |
ggtheme <- input$ggtheme |
| 736 | ||
| 737 | ! |
teal::validate_inputs(iv_dist) |
| 738 | ||
| 739 | ! |
qenv <- common_q() |
| 740 | ||
| 741 | ! |
m_type <- if (main_type_var == "Density") "density" else "count" |
| 742 | ||
| 743 | ! |
plot_call <- if (length(s_var) == 0 && length(g_var) == 0) {
|
| 744 | ! |
substitute( |
| 745 | ! |
expr = ggplot(ANL, aes(dist_var_name)) + |
| 746 | ! |
geom_histogram( |
| 747 | ! |
position = "identity", aes(y = after_stat(m_type)), bins = bins_var, alpha = 0.3 |
| 748 |
), |
|
| 749 | ! |
env = list( |
| 750 | ! |
m_type = as.name(m_type), bins_var = bins_var, dist_var_name = as.name(dist_var) |
| 751 |
) |
|
| 752 |
) |
|
| 753 | ! |
} else if (length(s_var) != 0 && length(g_var) == 0) {
|
| 754 | ! |
substitute( |
| 755 | ! |
expr = ggplot(ANL, aes(dist_var_name, col = s_var_name)) + |
| 756 | ! |
geom_histogram( |
| 757 | ! |
position = "identity", aes(y = after_stat(m_type), fill = s_var), bins = bins_var, alpha = 0.3 |
| 758 |
), |
|
| 759 | ! |
env = list( |
| 760 | ! |
m_type = as.name(m_type), |
| 761 | ! |
bins_var = bins_var, |
| 762 | ! |
dist_var_name = dist_var_name, |
| 763 | ! |
s_var = as.name(s_var), |
| 764 | ! |
s_var_name = s_var_name |
| 765 |
) |
|
| 766 |
) |
|
| 767 | ! |
} else if (length(s_var) == 0 && length(g_var) != 0) {
|
| 768 | ! |
req(scales_type) |
| 769 | ! |
substitute( |
| 770 | ! |
expr = ggplot(ANL[ANL[[g_var]] != "NA", ], aes(dist_var_name)) + |
| 771 | ! |
geom_histogram( |
| 772 | ! |
position = "identity", aes(y = after_stat(m_type)), bins = bins_var, alpha = 0.3 |
| 773 |
) + |
|
| 774 | ! |
facet_wrap(~g_var_name, ncol = 1, scales = scales_raw), |
| 775 | ! |
env = list( |
| 776 | ! |
m_type = as.name(m_type), |
| 777 | ! |
bins_var = bins_var, |
| 778 | ! |
dist_var_name = dist_var_name, |
| 779 | ! |
g_var = g_var, |
| 780 | ! |
g_var_name = g_var_name, |
| 781 | ! |
scales_raw = tolower(scales_type) |
| 782 |
) |
|
| 783 |
) |
|
| 784 |
} else {
|
|
| 785 | ! |
req(scales_type) |
| 786 | ! |
substitute( |
| 787 | ! |
expr = ggplot(ANL[ANL[[g_var]] != "NA", ], aes(dist_var_name, col = s_var_name)) + |
| 788 | ! |
geom_histogram( |
| 789 | ! |
position = "identity", |
| 790 | ! |
aes(y = after_stat(m_type), fill = s_var), bins = bins_var, alpha = 0.3 |
| 791 |
) + |
|
| 792 | ! |
facet_wrap(~g_var_name, ncol = 1, scales = scales_raw), |
| 793 | ! |
env = list( |
| 794 | ! |
m_type = as.name(m_type), |
| 795 | ! |
bins_var = bins_var, |
| 796 | ! |
dist_var_name = dist_var_name, |
| 797 | ! |
g_var = g_var, |
| 798 | ! |
s_var = as.name(s_var), |
| 799 | ! |
g_var_name = g_var_name, |
| 800 | ! |
s_var_name = s_var_name, |
| 801 | ! |
scales_raw = tolower(scales_type) |
| 802 |
) |
|
| 803 |
) |
|
| 804 |
} |
|
| 805 | ||
| 806 | ! |
if (add_dens_var) {
|
| 807 | ! |
plot_call <- substitute( |
| 808 | ! |
expr = plot_call + |
| 809 | ! |
stat_density( |
| 810 | ! |
aes(y = after_stat(const * m_type2)), |
| 811 | ! |
geom = "line", |
| 812 | ! |
position = "identity", |
| 813 | ! |
alpha = 0.5, |
| 814 | ! |
size = 2, |
| 815 | ! |
n = ndensity |
| 816 |
), |
|
| 817 | ! |
env = list( |
| 818 | ! |
plot_call = plot_call, |
| 819 | ! |
const = if (main_type_var == "Density") {
|
| 820 | ! |
1 |
| 821 |
} else {
|
|
| 822 | ! |
diff(range(qenv[["ANL"]][[dist_var]], na.rm = TRUE)) / bins_var |
| 823 |
}, |
|
| 824 | ! |
m_type2 = if (main_type_var == "Density") as.name("density") else as.name("count"),
|
| 825 | ! |
ndensity = ndensity |
| 826 |
) |
|
| 827 |
) |
|
| 828 |
} |
|
| 829 | ||
| 830 | ! |
if (length(t_dist) != 0 && main_type_var == "Density" && length(g_var) == 0 && length(s_var) == 0) {
|
| 831 | ! |
qenv <- teal.code::eval_code( |
| 832 | ! |
qenv, |
| 833 | ! |
substitute( |
| 834 | ! |
df_params <- as.data.frame(append(params, list(name = t_dist))), |
| 835 | ! |
env = list(t_dist = t_dist) |
| 836 |
) |
|
| 837 |
) |
|
| 838 | ! |
datas <- quote(data.frame(x = 0.7, y = 1, tb = I(list(df_params = df_params)))) |
| 839 | ! |
label <- quote(tb) |
| 840 | ||
| 841 | ! |
plot_call <- substitute( |
| 842 | ! |
expr = plot_call + ggpp::geom_table_npc( |
| 843 | ! |
data = data, |
| 844 | ! |
aes(npcx = x, npcy = y, label = label), |
| 845 | ! |
hjust = 0, vjust = 1, size = 4 |
| 846 |
), |
|
| 847 | ! |
env = list(plot_call = plot_call, data = datas, label = label) |
| 848 |
) |
|
| 849 |
} |
|
| 850 | ||
| 851 | ! |
if ( |
| 852 | ! |
length(s_var) == 0 && |
| 853 | ! |
length(g_var) == 0 && |
| 854 | ! |
main_type_var == "Density" && |
| 855 | ! |
length(t_dist) != 0 && |
| 856 | ! |
main_type_var == "Density" |
| 857 |
) {
|
|
| 858 | ! |
map_dist <- stats::setNames( |
| 859 | ! |
c("dnorm", "dlnorm", "dgamma", "dunif"),
|
| 860 | ! |
c("normal", "lognormal", "gamma", "unif")
|
| 861 |
) |
|
| 862 | ! |
plot_call <- substitute( |
| 863 | ! |
expr = plot_call + stat_function( |
| 864 | ! |
data = data.frame(x = range(ANL[[dist_var]]), color = mapped_dist), |
| 865 | ! |
aes(x, color = color), |
| 866 | ! |
fun = mapped_dist_name, |
| 867 | ! |
n = ndensity, |
| 868 | ! |
size = 2, |
| 869 | ! |
args = params |
| 870 |
) + |
|
| 871 | ! |
scale_color_manual(values = stats::setNames("blue", mapped_dist), aesthetics = "color"),
|
| 872 | ! |
env = list( |
| 873 | ! |
plot_call = plot_call, |
| 874 | ! |
dist_var = dist_var, |
| 875 | ! |
ndensity = ndensity, |
| 876 | ! |
mapped_dist = unname(map_dist[t_dist]), |
| 877 | ! |
mapped_dist_name = as.name(unname(map_dist[t_dist])) |
| 878 |
) |
|
| 879 |
) |
|
| 880 |
} |
|
| 881 | ||
| 882 | ! |
all_ggplot2_args <- teal.widgets::resolve_ggplot2_args( |
| 883 | ! |
user_plot = ggplot2_args[["Histogram"]], |
| 884 | ! |
user_default = ggplot2_args$default |
| 885 |
) |
|
| 886 | ||
| 887 | ! |
parsed_ggplot2_args <- teal.widgets::parse_ggplot2_args( |
| 888 | ! |
all_ggplot2_args, |
| 889 | ! |
ggtheme = ggtheme |
| 890 |
) |
|
| 891 | ||
| 892 | ! |
teal.code::eval_code( |
| 893 | ! |
qenv, |
| 894 | ! |
substitute( |
| 895 | ! |
expr = {
|
| 896 | ! |
g <- plot_call |
| 897 | ! |
print(g) |
| 898 |
}, |
|
| 899 | ! |
env = list(plot_call = Reduce(function(x, y) call("+", x, y), c(plot_call, parsed_ggplot2_args)))
|
| 900 |
) |
|
| 901 |
) |
|
| 902 |
} |
|
| 903 |
) |
|
| 904 | ||
| 905 |
# qqplot qenv ---- |
|
| 906 | ! |
qq_q <- eventReactive( |
| 907 | ! |
eventExpr = {
|
| 908 | ! |
common_q() |
| 909 | ! |
input$scales_type |
| 910 | ! |
input$qq_line |
| 911 | ! |
is.null(input$ggtheme) |
| 912 |
}, |
|
| 913 | ! |
valueExpr = {
|
| 914 | ! |
dist_var <- merge_vars()$dist_var |
| 915 | ! |
s_var <- merge_vars()$s_var |
| 916 | ! |
g_var <- merge_vars()$g_var |
| 917 | ! |
dist_var_name <- merge_vars()$dist_var_name |
| 918 | ! |
s_var_name <- merge_vars()$s_var_name |
| 919 | ! |
g_var_name <- merge_vars()$g_var_name |
| 920 | ! |
t_dist <- input$t_dist |
| 921 | ! |
dist_param1 <- input$dist_param1 |
| 922 | ! |
dist_param2 <- input$dist_param2 |
| 923 | ||
| 924 | ! |
scales_type <- input$scales_type |
| 925 | ! |
ggtheme <- input$ggtheme |
| 926 | ||
| 927 | ! |
teal::validate_inputs(iv_r_dist(), iv_dist) |
| 928 | ||
| 929 | ! |
qenv <- common_q() |
| 930 | ||
| 931 | ! |
plot_call <- if (length(s_var) == 0 && length(g_var) == 0) {
|
| 932 | ! |
substitute( |
| 933 | ! |
expr = ggplot(ANL, aes_string(sample = dist_var)), |
| 934 | ! |
env = list(dist_var = dist_var) |
| 935 |
) |
|
| 936 | ! |
} else if (length(s_var) != 0 && length(g_var) == 0) {
|
| 937 | ! |
substitute( |
| 938 | ! |
expr = ggplot(ANL, aes_string(sample = dist_var, color = s_var)), |
| 939 | ! |
env = list(dist_var = dist_var, s_var = s_var) |
| 940 |
) |
|
| 941 | ! |
} else if (length(s_var) == 0 && length(g_var) != 0) {
|
| 942 | ! |
substitute( |
| 943 | ! |
expr = ggplot(ANL[ANL[[g_var]] != "NA", ], aes_string(sample = dist_var)) + |
| 944 | ! |
facet_wrap(~g_var_name, ncol = 1, scales = scales_raw), |
| 945 | ! |
env = list( |
| 946 | ! |
dist_var = dist_var, |
| 947 | ! |
g_var = g_var, |
| 948 | ! |
g_var_name = g_var_name, |
| 949 | ! |
scales_raw = tolower(scales_type) |
| 950 |
) |
|
| 951 |
) |
|
| 952 |
} else {
|
|
| 953 | ! |
substitute( |
| 954 | ! |
expr = ggplot(ANL[ANL[[g_var]] != "NA", ], aes_string(sample = dist_var, color = s_var)) + |
| 955 | ! |
facet_wrap(~g_var_name, ncol = 1, scales = scales_raw), |
| 956 | ! |
env = list( |
| 957 | ! |
dist_var = dist_var, |
| 958 | ! |
g_var = g_var, |
| 959 | ! |
s_var = s_var, |
| 960 | ! |
g_var_name = g_var_name, |
| 961 | ! |
scales_raw = tolower(scales_type) |
| 962 |
) |
|
| 963 |
) |
|
| 964 |
} |
|
| 965 | ||
| 966 | ! |
map_dist <- stats::setNames( |
| 967 | ! |
c("qnorm", "qlnorm", "qgamma", "qunif"),
|
| 968 | ! |
c("normal", "lognormal", "gamma", "unif")
|
| 969 |
) |
|
| 970 | ||
| 971 | ! |
plot_call <- substitute( |
| 972 | ! |
expr = plot_call + |
| 973 | ! |
stat_qq(distribution = mapped_dist, dparams = params), |
| 974 | ! |
env = list(plot_call = plot_call, mapped_dist = as.name(unname(map_dist[t_dist]))) |
| 975 |
) |
|
| 976 | ||
| 977 | ! |
if (length(t_dist) != 0 && length(g_var) == 0 && length(s_var) == 0) {
|
| 978 | ! |
qenv <- teal.code::eval_code( |
| 979 | ! |
qenv, |
| 980 | ! |
substitute( |
| 981 | ! |
df_params <- as.data.frame(append(params, list(name = t_dist))), |
| 982 | ! |
env = list(t_dist = t_dist) |
| 983 |
) |
|
| 984 |
) |
|
| 985 | ! |
datas <- quote(data.frame(x = 0.7, y = 1, tb = I(list(df_params = df_params)))) |
| 986 | ! |
label <- quote(tb) |
| 987 | ||
| 988 | ! |
plot_call <- substitute( |
| 989 | ! |
expr = plot_call + |
| 990 | ! |
ggpp::geom_table_npc( |
| 991 | ! |
data = data, |
| 992 | ! |
aes(npcx = x, npcy = y, label = label), |
| 993 | ! |
hjust = 0, |
| 994 | ! |
vjust = 1, |
| 995 | ! |
size = 4 |
| 996 |
), |
|
| 997 | ! |
env = list( |
| 998 | ! |
plot_call = plot_call, |
| 999 | ! |
data = datas, |
| 1000 | ! |
label = label |
| 1001 |
) |
|
| 1002 |
) |
|
| 1003 |
} |
|
| 1004 | ||
| 1005 | ! |
if (isTRUE(input$qq_line)) {
|
| 1006 | ! |
plot_call <- substitute( |
| 1007 | ! |
expr = plot_call + |
| 1008 | ! |
stat_qq_line(distribution = mapped_dist, dparams = params), |
| 1009 | ! |
env = list(plot_call = plot_call, mapped_dist = as.name(unname(map_dist[t_dist]))) |
| 1010 |
) |
|
| 1011 |
} |
|
| 1012 | ||
| 1013 | ! |
all_ggplot2_args <- teal.widgets::resolve_ggplot2_args( |
| 1014 | ! |
user_plot = ggplot2_args[["QQplot"]], |
| 1015 | ! |
user_default = ggplot2_args$default, |
| 1016 | ! |
module_plot = teal.widgets::ggplot2_args(labs = list(x = "theoretical", y = "sample")) |
| 1017 |
) |
|
| 1018 | ||
| 1019 | ! |
parsed_ggplot2_args <- teal.widgets::parse_ggplot2_args( |
| 1020 | ! |
all_ggplot2_args, |
| 1021 | ! |
ggtheme = ggtheme |
| 1022 |
) |
|
| 1023 | ||
| 1024 | ! |
teal.code::eval_code( |
| 1025 | ! |
qenv, |
| 1026 | ! |
substitute( |
| 1027 | ! |
expr = {
|
| 1028 | ! |
g <- plot_call |
| 1029 | ! |
print(g) |
| 1030 |
}, |
|
| 1031 | ! |
env = list(plot_call = Reduce(function(x, y) call("+", x, y), c(plot_call, parsed_ggplot2_args)))
|
| 1032 |
) |
|
| 1033 |
) |
|
| 1034 |
} |
|
| 1035 |
) |
|
| 1036 | ||
| 1037 |
# test qenv ---- |
|
| 1038 | ! |
test_q <- eventReactive( |
| 1039 | ! |
ignoreNULL = FALSE, |
| 1040 | ! |
eventExpr = {
|
| 1041 | ! |
common_q() |
| 1042 | ! |
input$dist_param1 |
| 1043 | ! |
input$dist_param2 |
| 1044 | ! |
input$dist_tests |
| 1045 |
}, |
|
| 1046 | ! |
valueExpr = {
|
| 1047 |
# Create a private stack for this function only. |
|
| 1048 | ! |
ANL <- common_q()[["ANL"]] |
| 1049 | ||
| 1050 | ! |
dist_var <- merge_vars()$dist_var |
| 1051 | ! |
s_var <- merge_vars()$s_var |
| 1052 | ! |
g_var <- merge_vars()$g_var |
| 1053 | ||
| 1054 | ! |
dist_var_name <- merge_vars()$dist_var_name |
| 1055 | ! |
s_var_name <- merge_vars()$s_var_name |
| 1056 | ! |
g_var_name <- merge_vars()$g_var_name |
| 1057 | ||
| 1058 | ! |
dist_param1 <- input$dist_param1 |
| 1059 | ! |
dist_param2 <- input$dist_param2 |
| 1060 | ! |
dist_tests <- input$dist_tests |
| 1061 | ! |
t_dist <- input$t_dist |
| 1062 | ||
| 1063 | ! |
validate(need(dist_tests, "Please select a test")) |
| 1064 | ||
| 1065 | ! |
teal::validate_inputs(iv_dist) |
| 1066 | ||
| 1067 | ! |
if (length(s_var) > 0 || length(g_var) > 0) {
|
| 1068 | ! |
counts <- ANL %>% |
| 1069 | ! |
dplyr::group_by_at(dplyr::vars(dplyr::any_of(c(s_var, g_var)))) %>% |
| 1070 | ! |
dplyr::summarise(n = dplyr::n()) |
| 1071 | ||
| 1072 | ! |
validate(need(all(counts$n > 5), "Please select strata*group with at least 5 observation each.")) |
| 1073 |
} |
|
| 1074 | ||
| 1075 | ||
| 1076 | ! |
if (dist_tests %in% c( |
| 1077 | ! |
"t-test (two-samples, not paired)", |
| 1078 | ! |
"F-test", |
| 1079 | ! |
"Kolmogorov-Smirnov (two-samples)" |
| 1080 |
)) {
|
|
| 1081 | ! |
if (length(g_var) == 0 && length(s_var) > 0) {
|
| 1082 | ! |
validate(need( |
| 1083 | ! |
length(unique(ANL[[s_var]])) == 2, |
| 1084 | ! |
"Please select stratify variable with 2 levels." |
| 1085 |
)) |
|
| 1086 |
} |
|
| 1087 | ! |
if (length(g_var) > 0 && length(s_var) > 0) {
|
| 1088 | ! |
validate(need( |
| 1089 | ! |
all(stats::na.omit(as.vector( |
| 1090 | ! |
tapply(ANL[[s_var]], list(ANL[[g_var]]), function(x) length(unique(x))) == 2 |
| 1091 |
))), |
|
| 1092 | ! |
"Please select stratify variable with 2 levels, per each group." |
| 1093 |
)) |
|
| 1094 |
} |
|
| 1095 |
} |
|
| 1096 | ||
| 1097 | ! |
map_dist <- stats::setNames( |
| 1098 | ! |
c("pnorm", "plnorm", "pgamma", "punif"),
|
| 1099 | ! |
c("normal", "lognormal", "gamma", "unif")
|
| 1100 |
) |
|
| 1101 | ! |
sks_args <- list( |
| 1102 | ! |
test = quote(stats::ks.test), |
| 1103 | ! |
args = bquote(append(list(.[[.(dist_var)]], .(map_dist[t_dist])), params)), |
| 1104 | ! |
groups = c(g_var, s_var) |
| 1105 |
) |
|
| 1106 | ! |
ssw_args <- list( |
| 1107 | ! |
test = quote(stats::shapiro.test), |
| 1108 | ! |
args = bquote(list(.[[.(dist_var)]])), |
| 1109 | ! |
groups = c(g_var, s_var) |
| 1110 |
) |
|
| 1111 | ! |
mfil_args <- list( |
| 1112 | ! |
test = quote(stats::fligner.test), |
| 1113 | ! |
args = bquote(list(.[[.(dist_var)]], .[[.(s_var)]])), |
| 1114 | ! |
groups = c(g_var) |
| 1115 |
) |
|
| 1116 | ! |
sad_args <- list( |
| 1117 | ! |
test = quote(goftest::ad.test), |
| 1118 | ! |
args = bquote(append(list(.[[.(dist_var)]], .(map_dist[t_dist])), params)), |
| 1119 | ! |
groups = c(g_var, s_var) |
| 1120 |
) |
|
| 1121 | ! |
scvm_args <- list( |
| 1122 | ! |
test = quote(goftest::cvm.test), |
| 1123 | ! |
args = bquote(append(list(.[[.(dist_var)]], .(map_dist[t_dist])), params)), |
| 1124 | ! |
groups = c(g_var, s_var) |
| 1125 |
) |
|
| 1126 | ! |
manov_args <- list( |
| 1127 | ! |
test = quote(stats::aov), |
| 1128 | ! |
args = bquote(list(stats::formula(.(dist_var_name) ~ .(s_var_name)), .)), |
| 1129 | ! |
groups = c(g_var) |
| 1130 |
) |
|
| 1131 | ! |
mt_args <- list( |
| 1132 | ! |
test = quote(stats::t.test), |
| 1133 | ! |
args = bquote(unname(split(.[[.(dist_var)]], .[[.(s_var)]], drop = TRUE))), |
| 1134 | ! |
groups = c(g_var) |
| 1135 |
) |
|
| 1136 | ! |
mv_args <- list( |
| 1137 | ! |
test = quote(stats::var.test), |
| 1138 | ! |
args = bquote(unname(split(.[[.(dist_var)]], .[[.(s_var)]], drop = TRUE))), |
| 1139 | ! |
groups = c(g_var) |
| 1140 |
) |
|
| 1141 | ! |
mks_args <- list( |
| 1142 | ! |
test = quote(stats::ks.test), |
| 1143 | ! |
args = bquote(unname(split(.[[.(dist_var)]], .[[.(s_var)]], drop = TRUE))), |
| 1144 | ! |
groups = c(g_var) |
| 1145 |
) |
|
| 1146 | ||
| 1147 | ! |
tests_base <- switch(dist_tests, |
| 1148 | ! |
"Kolmogorov-Smirnov (one-sample)" = sks_args, |
| 1149 | ! |
"Shapiro-Wilk" = ssw_args, |
| 1150 | ! |
"Fligner-Killeen" = mfil_args, |
| 1151 | ! |
"one-way ANOVA" = manov_args, |
| 1152 | ! |
"t-test (two-samples, not paired)" = mt_args, |
| 1153 | ! |
"F-test" = mv_args, |
| 1154 | ! |
"Kolmogorov-Smirnov (two-samples)" = mks_args, |
| 1155 | ! |
"Anderson-Darling (one-sample)" = sad_args, |
| 1156 | ! |
"Cramer-von Mises (one-sample)" = scvm_args |
| 1157 |
) |
|
| 1158 | ||
| 1159 | ! |
env <- list( |
| 1160 | ! |
t_test = t_dist, |
| 1161 | ! |
dist_var = dist_var, |
| 1162 | ! |
g_var = g_var, |
| 1163 | ! |
s_var = s_var, |
| 1164 | ! |
args = tests_base$args, |
| 1165 | ! |
groups = tests_base$groups, |
| 1166 | ! |
test = tests_base$test, |
| 1167 | ! |
dist_var_name = dist_var_name, |
| 1168 | ! |
g_var_name = g_var_name, |
| 1169 | ! |
s_var_name = s_var_name |
| 1170 |
) |
|
| 1171 | ||
| 1172 | ! |
qenv <- common_q() |
| 1173 | ||
| 1174 | ! |
if (length(s_var) == 0 && length(g_var) == 0) {
|
| 1175 | ! |
qenv <- teal.code::eval_code( |
| 1176 | ! |
qenv, |
| 1177 | ! |
substitute( |
| 1178 | ! |
expr = {
|
| 1179 | ! |
test_stats <- ANL %>% |
| 1180 | ! |
dplyr::select(dist_var) %>% |
| 1181 | ! |
with(., broom::glance(do.call(test, args))) %>% |
| 1182 | ! |
dplyr::mutate_if(is.numeric, round, 3) |
| 1183 |
}, |
|
| 1184 | ! |
env = env |
| 1185 |
) |
|
| 1186 |
) |
|
| 1187 |
} else {
|
|
| 1188 | ! |
qenv <- teal.code::eval_code( |
| 1189 | ! |
qenv, |
| 1190 | ! |
substitute( |
| 1191 | ! |
expr = {
|
| 1192 | ! |
test_stats <- ANL %>% |
| 1193 | ! |
dplyr::select(dist_var, s_var, g_var) %>% |
| 1194 | ! |
dplyr::group_by_at(dplyr::vars(dplyr::any_of(groups))) %>% |
| 1195 | ! |
dplyr::do(tests = broom::glance(do.call(test, args))) %>% |
| 1196 | ! |
tidyr::unnest(tests) %>% |
| 1197 | ! |
dplyr::mutate_if(is.numeric, round, 3) |
| 1198 |
}, |
|
| 1199 | ! |
env = env |
| 1200 |
) |
|
| 1201 |
) |
|
| 1202 |
} |
|
| 1203 | ! |
qenv %>% |
| 1204 |
# used to display table when running show-r-code code |
|
| 1205 | ! |
teal.code::eval_code(quote(test_stats)) |
| 1206 |
} |
|
| 1207 |
) |
|
| 1208 | ||
| 1209 |
# outputs ---- |
|
| 1210 |
## building main qenv |
|
| 1211 | ! |
output_q <- reactive({
|
| 1212 | ! |
tab <- input$tabs |
| 1213 | ! |
req(tab) # tab is NULL upon app launch, hence will crash without this statement |
| 1214 | ||
| 1215 | ! |
qenv_final <- common_q() |
| 1216 |
# wrapped in if since could lead into validate error - we do want to continue |
|
| 1217 | ! |
test_r_qenv_out <- try(test_q(), silent = TRUE) |
| 1218 | ! |
if (!inherits(test_r_qenv_out, c("try-error", "error"))) {
|
| 1219 | ! |
qenv_final <- teal.code::join(qenv_final, test_q()) |
| 1220 |
} |
|
| 1221 | ||
| 1222 | ! |
qenv_final <- if (tab == "Histogram") {
|
| 1223 | ! |
req(dist_q()) |
| 1224 | ! |
teal.code::join(qenv_final, dist_q()) |
| 1225 | ! |
} else if (tab == "QQplot") {
|
| 1226 | ! |
req(qq_q()) |
| 1227 | ! |
teal.code::join(qenv_final, qq_q()) |
| 1228 |
} |
|
| 1229 | ! |
qenv_final |
| 1230 |
}) |
|
| 1231 | ||
| 1232 | ! |
dist_r <- reactive(dist_q()[["g"]]) |
| 1233 | ||
| 1234 | ! |
qq_r <- reactive(qq_q()[["g"]]) |
| 1235 | ||
| 1236 | ! |
output$summary_table <- DT::renderDataTable( |
| 1237 | ! |
expr = if (iv_r()$is_valid()) common_q()[["summary_table"]] else NULL, |
| 1238 | ! |
options = list( |
| 1239 | ! |
autoWidth = TRUE, |
| 1240 | ! |
columnDefs = list(list(width = "200px", targets = "_all")) |
| 1241 |
), |
|
| 1242 | ! |
rownames = FALSE |
| 1243 |
) |
|
| 1244 | ||
| 1245 | ! |
tests_r <- reactive({
|
| 1246 | ! |
req(iv_r()$is_valid()) |
| 1247 | ! |
teal::validate_inputs(iv_r_dist()) |
| 1248 | ! |
test_q()[["test_stats"]] |
| 1249 |
}) |
|
| 1250 | ||
| 1251 | ! |
pws1 <- teal.widgets::plot_with_settings_srv( |
| 1252 | ! |
id = "hist_plot", |
| 1253 | ! |
plot_r = dist_r, |
| 1254 | ! |
height = plot_height, |
| 1255 | ! |
width = plot_width, |
| 1256 | ! |
brushing = FALSE |
| 1257 |
) |
|
| 1258 | ||
| 1259 | ! |
pws2 <- teal.widgets::plot_with_settings_srv( |
| 1260 | ! |
id = "qq_plot", |
| 1261 | ! |
plot_r = qq_r, |
| 1262 | ! |
height = plot_height, |
| 1263 | ! |
width = plot_width, |
| 1264 | ! |
brushing = FALSE |
| 1265 |
) |
|
| 1266 | ||
| 1267 | ! |
output$t_stats <- DT::renderDataTable( |
| 1268 | ! |
expr = tests_r(), |
| 1269 | ! |
options = list(scrollX = TRUE), |
| 1270 | ! |
rownames = FALSE |
| 1271 |
) |
|
| 1272 | ||
| 1273 | ! |
teal.widgets::verbatim_popup_srv( |
| 1274 | ! |
id = "rcode", |
| 1275 | ! |
verbatim_content = reactive(teal.code::get_code(output_q())), |
| 1276 | ! |
title = "R Code for distribution" |
| 1277 |
) |
|
| 1278 | ||
| 1279 |
### REPORTER |
|
| 1280 | ! |
if (with_reporter) {
|
| 1281 | ! |
card_fun <- function(comment, label) {
|
| 1282 | ! |
card <- teal::report_card_template( |
| 1283 | ! |
title = "Distribution Plot", |
| 1284 | ! |
label = label, |
| 1285 | ! |
with_filter = with_filter, |
| 1286 | ! |
filter_panel_api = filter_panel_api |
| 1287 |
) |
|
| 1288 | ! |
card$append_text("Plot", "header3")
|
| 1289 | ! |
if (input$tabs == "Histogram") {
|
| 1290 | ! |
card$append_plot(dist_r(), dim = pws1$dim()) |
| 1291 | ! |
} else if (input$tabs == "QQplot") {
|
| 1292 | ! |
card$append_plot(qq_r(), dim = pws2$dim()) |
| 1293 |
} |
|
| 1294 | ! |
card$append_text("Statistics table", "header3")
|
| 1295 | ||
| 1296 | ! |
card$append_table(common_q()[["summary_table"]]) |
| 1297 | ! |
tests_error <- tryCatch(expr = tests_r(), error = function(e) "error") |
| 1298 | ! |
if (inherits(tests_error, "data.frame")) {
|
| 1299 | ! |
card$append_text("Tests table", "header3")
|
| 1300 | ! |
card$append_table(tests_r()) |
| 1301 |
} |
|
| 1302 | ||
| 1303 | ! |
if (!comment == "") {
|
| 1304 | ! |
card$append_text("Comment", "header3")
|
| 1305 | ! |
card$append_text(comment) |
| 1306 |
} |
|
| 1307 | ! |
card$append_src(teal.code::get_code(output_q())) |
| 1308 | ! |
card |
| 1309 |
} |
|
| 1310 | ! |
teal.reporter::simple_reporter_srv("simple_reporter", reporter = reporter, card_fun = card_fun)
|
| 1311 |
} |
|
| 1312 |
### |
|
| 1313 |
}) |
|
| 1314 |
} |
| 1 |
#' `teal` module: Variable browser |
|
| 2 |
#' |
|
| 3 |
#' Module provides provides a detailed summary and visualization of variable distributions |
|
| 4 |
#' for `data.frame` objects, with interactive features to customize analysis. |
|
| 5 |
#' |
|
| 6 |
#' Numeric columns with fewer than 30 distinct values can be treated as either discrete |
|
| 7 |
#' or continuous with a checkbox allowing users to switch how they are treated(if < 6 unique values |
|
| 8 |
#' then the default is discrete, otherwise it is continuous). |
|
| 9 |
#' |
|
| 10 |
#' @inheritParams teal::module |
|
| 11 |
#' @inheritParams shared_params |
|
| 12 |
#' @param parent_dataname (`character(1)`) string specifying a parent dataset. |
|
| 13 |
#' If it exists in `datasets_selected`then an extra checkbox will be shown to |
|
| 14 |
#' allow users to not show variables in other datasets which exist in this `dataname`. |
|
| 15 |
#' This is typically used to remove `ADSL` columns in `CDISC` data. |
|
| 16 |
#' In non `CDISC` data this can be ignored. Defaults to `"ADSL"`. |
|
| 17 |
#' @param datasets_selected (`character`) vector of datasets which should be |
|
| 18 |
#' shown, in order. Names must correspond with datasets names. |
|
| 19 |
#' If vector of length zero (default) then all datasets are shown. |
|
| 20 |
#' Note: Only `data.frame` objects are compatible; using other types will cause an error. |
|
| 21 |
#' |
|
| 22 |
#' @inherit shared_params return |
|
| 23 |
#' |
|
| 24 |
#' @examplesShinylive |
|
| 25 |
#' library(teal.modules.general) |
|
| 26 |
#' interactive <- function() TRUE |
|
| 27 |
#' {{ next_example }}
|
|
| 28 |
# nolint start: line_length_linter. |
|
| 29 |
#' @examplesIf require("sparkline", quietly = TRUE) && require("htmlwidgets", quietly = TRUE) && require("jsonlite", quietly = TRUE)
|
|
| 30 |
# nolint end: line_length_linter. |
|
| 31 |
#' # general data example |
|
| 32 |
#' data <- teal_data() |
|
| 33 |
#' data <- within(data, {
|
|
| 34 |
#' iris <- iris |
|
| 35 |
#' mtcars <- mtcars |
|
| 36 |
#' women <- women |
|
| 37 |
#' faithful <- faithful |
|
| 38 |
#' CO2 <- CO2 |
|
| 39 |
#' }) |
|
| 40 |
#' datanames(data) <- c("iris", "mtcars", "women", "faithful", "CO2")
|
|
| 41 |
#' |
|
| 42 |
#' app <- init( |
|
| 43 |
#' data = data, |
|
| 44 |
#' modules = modules( |
|
| 45 |
#' tm_variable_browser( |
|
| 46 |
#' label = "Variable browser" |
|
| 47 |
#' ) |
|
| 48 |
#' ) |
|
| 49 |
#' ) |
|
| 50 |
#' if (interactive()) {
|
|
| 51 |
#' shinyApp(app$ui, app$server) |
|
| 52 |
#' } |
|
| 53 |
#' |
|
| 54 |
#' @examplesShinylive |
|
| 55 |
#' library(teal.modules.general) |
|
| 56 |
#' interactive <- function() TRUE |
|
| 57 |
#' {{ next_example }}
|
|
| 58 |
# nolint start: line_length_linter. |
|
| 59 |
#' @examplesIf require("sparkline", quietly = TRUE) && require("htmlwidgets", quietly = TRUE) && require("jsonlite", quietly = TRUE)
|
|
| 60 |
# nolint end: line_length_linter. |
|
| 61 |
#' # CDISC example data |
|
| 62 |
#' library(sparkline) |
|
| 63 |
#' data <- teal_data() |
|
| 64 |
#' data <- within(data, {
|
|
| 65 |
#' ADSL <- rADSL |
|
| 66 |
#' ADTTE <- rADTTE |
|
| 67 |
#' }) |
|
| 68 |
#' datanames(data) <- c("ADSL", "ADTTE")
|
|
| 69 |
#' join_keys(data) <- default_cdisc_join_keys[datanames(data)] |
|
| 70 |
#' |
|
| 71 |
#' app <- init( |
|
| 72 |
#' data = data, |
|
| 73 |
#' modules = modules( |
|
| 74 |
#' tm_variable_browser( |
|
| 75 |
#' label = "Variable browser" |
|
| 76 |
#' ) |
|
| 77 |
#' ) |
|
| 78 |
#' ) |
|
| 79 |
#' if (interactive()) {
|
|
| 80 |
#' shinyApp(app$ui, app$server) |
|
| 81 |
#' } |
|
| 82 |
#' |
|
| 83 |
#' @export |
|
| 84 |
#' |
|
| 85 |
tm_variable_browser <- function(label = "Variable Browser", |
|
| 86 |
datasets_selected = character(0), |
|
| 87 |
parent_dataname = "ADSL", |
|
| 88 |
pre_output = NULL, |
|
| 89 |
post_output = NULL, |
|
| 90 |
ggplot2_args = teal.widgets::ggplot2_args()) {
|
|
| 91 | ! |
message("Initializing tm_variable_browser")
|
| 92 | ||
| 93 |
# Requires Suggested packages |
|
| 94 | ! |
if (!requireNamespace("sparkline", quietly = TRUE)) {
|
| 95 | ! |
stop("Cannot load sparkline - please install the package or restart your session.")
|
| 96 |
} |
|
| 97 | ! |
if (!requireNamespace("htmlwidgets", quietly = TRUE)) {
|
| 98 | ! |
stop("Cannot load htmlwidgets - please install the package or restart your session.")
|
| 99 |
} |
|
| 100 | ! |
if (!requireNamespace("jsonlite", quietly = TRUE)) {
|
| 101 | ! |
stop("Cannot load jsonlite - please install the package or restart your session.")
|
| 102 |
} |
|
| 103 | ||
| 104 |
# Start of assertions |
|
| 105 | ! |
checkmate::assert_string(label) |
| 106 | ! |
checkmate::assert_character(datasets_selected) |
| 107 | ! |
checkmate::assert_character(parent_dataname, min.len = 0, max.len = 1) |
| 108 | ! |
checkmate::assert_multi_class(pre_output, c("shiny.tag", "shiny.tag.list", "html"), null.ok = TRUE)
|
| 109 | ! |
checkmate::assert_multi_class(post_output, c("shiny.tag", "shiny.tag.list", "html"), null.ok = TRUE)
|
| 110 | ! |
checkmate::assert_class(ggplot2_args, "ggplot2_args") |
| 111 |
# End of assertions |
|
| 112 | ||
| 113 | ! |
datasets_selected <- unique(datasets_selected) |
| 114 | ||
| 115 | ! |
ans <- module( |
| 116 | ! |
label, |
| 117 | ! |
server = srv_variable_browser, |
| 118 | ! |
ui = ui_variable_browser, |
| 119 | ! |
datanames = "all", |
| 120 | ! |
server_args = list( |
| 121 | ! |
datasets_selected = datasets_selected, |
| 122 | ! |
parent_dataname = parent_dataname, |
| 123 | ! |
ggplot2_args = ggplot2_args |
| 124 |
), |
|
| 125 | ! |
ui_args = list( |
| 126 | ! |
pre_output = pre_output, |
| 127 | ! |
post_output = post_output |
| 128 |
) |
|
| 129 |
) |
|
| 130 |
# `shiny` inputs are stored properly but the majority of the module is state of `datatable` which is not stored. |
|
| 131 | ! |
attr(ans, "teal_bookmarkable") <- NULL |
| 132 | ! |
ans |
| 133 |
} |
|
| 134 | ||
| 135 |
# UI function for the variable browser module |
|
| 136 |
ui_variable_browser <- function(id, |
|
| 137 |
pre_output = NULL, |
|
| 138 |
post_output = NULL) {
|
|
| 139 | ! |
ns <- NS(id) |
| 140 | ||
| 141 | ! |
tagList( |
| 142 | ! |
include_css_files("custom"),
|
| 143 | ! |
shinyjs::useShinyjs(), |
| 144 | ! |
teal.widgets::standard_layout( |
| 145 | ! |
output = fluidRow( |
| 146 | ! |
htmlwidgets::getDependency("sparkline"), # needed for sparklines to work
|
| 147 | ! |
column( |
| 148 | ! |
6, |
| 149 |
# variable browser |
|
| 150 | ! |
teal.widgets::white_small_well( |
| 151 | ! |
uiOutput(ns("ui_variable_browser")),
|
| 152 | ! |
shinyjs::hidden({
|
| 153 | ! |
checkboxInput(ns("show_parent_vars"), "Show parent dataset variables", value = FALSE)
|
| 154 |
}) |
|
| 155 |
) |
|
| 156 |
), |
|
| 157 | ! |
column( |
| 158 | ! |
6, |
| 159 | ! |
teal.widgets::white_small_well( |
| 160 |
### Reporter |
|
| 161 | ! |
teal.reporter::simple_reporter_ui(ns("simple_reporter")),
|
| 162 |
### |
|
| 163 | ! |
tags$div( |
| 164 | ! |
class = "block", |
| 165 | ! |
uiOutput(ns("ui_histogram_display"))
|
| 166 |
), |
|
| 167 | ! |
tags$div( |
| 168 | ! |
class = "block", |
| 169 | ! |
uiOutput(ns("ui_numeric_display"))
|
| 170 |
), |
|
| 171 | ! |
teal.widgets::plot_with_settings_ui(ns("variable_plot")),
|
| 172 | ! |
tags$br(), |
| 173 |
# input user-defined text size |
|
| 174 | ! |
teal.widgets::panel_item( |
| 175 | ! |
title = "Plot settings", |
| 176 | ! |
collapsed = TRUE, |
| 177 | ! |
selectInput( |
| 178 | ! |
inputId = ns("ggplot_theme"), label = "ggplot2 theme",
|
| 179 | ! |
choices = ggplot_themes, |
| 180 | ! |
selected = "grey" |
| 181 |
), |
|
| 182 | ! |
fluidRow( |
| 183 | ! |
column(6, sliderInput( |
| 184 | ! |
inputId = ns("font_size"), label = "font size",
|
| 185 | ! |
min = 5L, max = 30L, value = 15L, step = 1L, ticks = FALSE |
| 186 |
)), |
|
| 187 | ! |
column(6, sliderInput( |
| 188 | ! |
inputId = ns("label_rotation"), label = "rotate x labels",
|
| 189 | ! |
min = 0L, max = 90L, value = 45L, step = 1, ticks = FALSE |
| 190 |
)) |
|
| 191 |
) |
|
| 192 |
), |
|
| 193 | ! |
tags$br(), |
| 194 | ! |
teal.widgets::get_dt_rows(ns("variable_summary_table"), ns("variable_summary_table_rows")),
|
| 195 | ! |
DT::dataTableOutput(ns("variable_summary_table"))
|
| 196 |
) |
|
| 197 |
) |
|
| 198 |
), |
|
| 199 | ! |
pre_output = pre_output, |
| 200 | ! |
post_output = post_output |
| 201 |
) |
|
| 202 |
) |
|
| 203 |
} |
|
| 204 | ||
| 205 |
# Server function for the variable browser module |
|
| 206 |
srv_variable_browser <- function(id, |
|
| 207 |
data, |
|
| 208 |
reporter, |
|
| 209 |
filter_panel_api, |
|
| 210 |
datasets_selected, parent_dataname, ggplot2_args) {
|
|
| 211 | ! |
with_reporter <- !missing(reporter) && inherits(reporter, "Reporter") |
| 212 | ! |
with_filter <- !missing(filter_panel_api) && inherits(filter_panel_api, "FilterPanelAPI") |
| 213 | ! |
checkmate::assert_class(data, "reactive") |
| 214 | ! |
checkmate::assert_class(isolate(data()), "teal_data") |
| 215 | ! |
moduleServer(id, function(input, output, session) {
|
| 216 | ! |
teal.logger::log_shiny_input_changes(input, namespace = "teal.modules.general") |
| 217 | ||
| 218 |
# if there are < this number of unique records then a numeric |
|
| 219 |
# variable can be treated as a factor and all factors with < this groups |
|
| 220 |
# have their values plotted |
|
| 221 | ! |
.unique_records_for_factor <- 30 |
| 222 |
# if there are < this number of unique records then a numeric |
|
| 223 |
# variable is by default treated as a factor |
|
| 224 | ! |
.unique_records_default_as_factor <- 6 # nolint: object_length. |
| 225 | ||
| 226 | ! |
varname_numeric_as_factor <- reactiveValues() |
| 227 | ||
| 228 | ! |
datanames <- isolate(teal.data::datanames(data())) |
| 229 | ! |
datanames <- Filter(function(name) {
|
| 230 | ! |
is.data.frame(isolate(data())[[name]]) |
| 231 | ! |
}, datanames) |
| 232 | ||
| 233 | ! |
checkmate::assert_character(datasets_selected) |
| 234 | ! |
checkmate::assert_subset(datasets_selected, datanames) |
| 235 | ! |
if (!identical(datasets_selected, character(0))) {
|
| 236 | ! |
checkmate::assert_subset(datasets_selected, datanames) |
| 237 | ! |
datanames <- datasets_selected |
| 238 |
} |
|
| 239 | ||
| 240 | ! |
output$ui_variable_browser <- renderUI({
|
| 241 | ! |
ns <- session$ns |
| 242 | ! |
do.call( |
| 243 | ! |
tabsetPanel, |
| 244 | ! |
c( |
| 245 | ! |
id = ns("tabset_panel"),
|
| 246 | ! |
do.call( |
| 247 | ! |
tagList, |
| 248 | ! |
lapply(datanames, function(dataname) {
|
| 249 | ! |
tabPanel( |
| 250 | ! |
dataname, |
| 251 | ! |
tags$div( |
| 252 | ! |
class = "mt-4", |
| 253 | ! |
textOutput(ns(paste0("dataset_summary_", dataname)))
|
| 254 |
), |
|
| 255 | ! |
tags$div( |
| 256 | ! |
class = "mt-4", |
| 257 | ! |
teal.widgets::get_dt_rows( |
| 258 | ! |
ns(paste0("variable_browser_", dataname)),
|
| 259 | ! |
ns(paste0("variable_browser_", dataname, "_rows"))
|
| 260 |
), |
|
| 261 | ! |
DT::dataTableOutput(ns(paste0("variable_browser_", dataname)), width = "100%")
|
| 262 |
) |
|
| 263 |
) |
|
| 264 |
}) |
|
| 265 |
) |
|
| 266 |
) |
|
| 267 |
) |
|
| 268 |
}) |
|
| 269 | ||
| 270 |
# conditionally display checkbox |
|
| 271 | ! |
shinyjs::toggle( |
| 272 | ! |
id = "show_parent_vars", |
| 273 | ! |
condition = length(parent_dataname) > 0 && parent_dataname %in% datanames |
| 274 |
) |
|
| 275 | ||
| 276 | ! |
columns_names <- new.env() |
| 277 | ||
| 278 |
# plot_var$data holds the name of the currently selected dataset |
|
| 279 |
# plot_var$variable[[<dataset_name>]] holds the name of the currently selected |
|
| 280 |
# variable for dataset <dataset_name> |
|
| 281 | ! |
plot_var <- reactiveValues(data = NULL, variable = list()) |
| 282 | ||
| 283 | ! |
establish_updating_selection(datanames, input, plot_var, columns_names) |
| 284 | ||
| 285 |
# validations |
|
| 286 | ! |
validation_checks <- validate_input(input, plot_var, data) |
| 287 | ||
| 288 |
# data_for_analysis is a list with two elements: a column from a dataset and the column label |
|
| 289 | ! |
plotted_data <- reactive({
|
| 290 | ! |
validation_checks() |
| 291 | ||
| 292 | ! |
get_plotted_data(input, plot_var, data) |
| 293 |
}) |
|
| 294 | ||
| 295 | ! |
treat_numeric_as_factor <- reactive({
|
| 296 | ! |
if (is_num_var_short(.unique_records_for_factor, input, plotted_data)) {
|
| 297 | ! |
input$numeric_as_factor |
| 298 |
} else {
|
|
| 299 | ! |
FALSE |
| 300 |
} |
|
| 301 |
}) |
|
| 302 | ||
| 303 | ! |
render_tabset_panel_content( |
| 304 | ! |
input = input, |
| 305 | ! |
output = output, |
| 306 | ! |
data = data, |
| 307 | ! |
datanames = datanames, |
| 308 | ! |
parent_dataname = parent_dataname, |
| 309 | ! |
columns_names = columns_names, |
| 310 | ! |
plot_var = plot_var |
| 311 |
) |
|
| 312 |
# add used-defined text size to ggplot arguments passed from caller frame |
|
| 313 | ! |
all_ggplot2_args <- reactive({
|
| 314 | ! |
user_text <- teal.widgets::ggplot2_args( |
| 315 | ! |
theme = list( |
| 316 | ! |
"text" = ggplot2::element_text(size = input[["font_size"]]), |
| 317 | ! |
"axis.text.x" = ggplot2::element_text(angle = input[["label_rotation"]], hjust = 1) |
| 318 |
) |
|
| 319 |
) |
|
| 320 | ! |
user_theme <- utils::getFromNamespace(sprintf("theme_%s", input[["ggplot_theme"]]), ns = "ggplot2")
|
| 321 | ! |
user_theme <- user_theme() |
| 322 |
# temporary fix to circumvent assertion issue with resolve_ggplot2_args |
|
| 323 |
# drop problematic elements |
|
| 324 | ! |
user_theme <- user_theme[grep("strip.text.y.left", names(user_theme), fixed = TRUE, invert = TRUE)]
|
| 325 | ||
| 326 | ! |
teal.widgets::resolve_ggplot2_args( |
| 327 | ! |
user_plot = user_text, |
| 328 | ! |
user_default = teal.widgets::ggplot2_args(theme = user_theme), |
| 329 | ! |
module_plot = ggplot2_args |
| 330 |
) |
|
| 331 |
}) |
|
| 332 | ||
| 333 | ! |
output$ui_numeric_display <- renderUI({
|
| 334 | ! |
validation_checks() |
| 335 | ! |
dataname <- input$tabset_panel |
| 336 | ! |
varname <- plot_var$variable[[dataname]] |
| 337 | ! |
df <- data()[[dataname]] |
| 338 | ||
| 339 | ! |
numeric_ui <- tagList( |
| 340 | ! |
fluidRow( |
| 341 | ! |
tags$div( |
| 342 | ! |
class = "col-md-4", |
| 343 | ! |
tags$br(), |
| 344 | ! |
shinyWidgets::switchInput( |
| 345 | ! |
inputId = session$ns("display_density"),
|
| 346 | ! |
label = "Show density", |
| 347 | ! |
value = `if`(is.null(isolate(input$display_density)), TRUE, isolate(input$display_density)), |
| 348 | ! |
width = "50%", |
| 349 | ! |
labelWidth = "100px", |
| 350 | ! |
handleWidth = "50px" |
| 351 |
) |
|
| 352 |
), |
|
| 353 | ! |
tags$div( |
| 354 | ! |
class = "col-md-4", |
| 355 | ! |
tags$br(), |
| 356 | ! |
shinyWidgets::switchInput( |
| 357 | ! |
inputId = session$ns("remove_outliers"),
|
| 358 | ! |
label = "Remove outliers", |
| 359 | ! |
value = `if`(is.null(isolate(input$remove_outliers)), FALSE, isolate(input$remove_outliers)), |
| 360 | ! |
width = "50%", |
| 361 | ! |
labelWidth = "100px", |
| 362 | ! |
handleWidth = "50px" |
| 363 |
) |
|
| 364 |
), |
|
| 365 | ! |
tags$div( |
| 366 | ! |
class = "col-md-4", |
| 367 | ! |
uiOutput(session$ns("outlier_definition_slider_ui"))
|
| 368 |
) |
|
| 369 |
), |
|
| 370 | ! |
tags$div( |
| 371 | ! |
class = "ml-4", |
| 372 | ! |
uiOutput(session$ns("ui_density_help")),
|
| 373 | ! |
uiOutput(session$ns("ui_outlier_help"))
|
| 374 |
) |
|
| 375 |
) |
|
| 376 | ||
| 377 | ! |
observeEvent(input$numeric_as_factor, ignoreInit = TRUE, {
|
| 378 | ! |
varname_numeric_as_factor[[plot_var$variable[[dataname]]]] <- input$numeric_as_factor |
| 379 |
}) |
|
| 380 | ||
| 381 | ! |
if (is.numeric(df[[varname]])) {
|
| 382 | ! |
unique_entries <- length(unique(df[[varname]])) |
| 383 | ! |
if (unique_entries < .unique_records_for_factor && unique_entries > 0) {
|
| 384 | ! |
list( |
| 385 | ! |
checkboxInput( |
| 386 | ! |
session$ns("numeric_as_factor"),
|
| 387 | ! |
"Treat variable as factor", |
| 388 | ! |
value = `if`( |
| 389 | ! |
is.null(varname_numeric_as_factor[[varname]]), |
| 390 | ! |
unique_entries < .unique_records_default_as_factor, |
| 391 | ! |
varname_numeric_as_factor[[varname]] |
| 392 |
) |
|
| 393 |
), |
|
| 394 | ! |
conditionalPanel("!input.numeric_as_factor", ns = session$ns, numeric_ui)
|
| 395 |
) |
|
| 396 | ! |
} else if (unique_entries > 0) {
|
| 397 | ! |
numeric_ui |
| 398 |
} |
|
| 399 |
} else {
|
|
| 400 | ! |
NULL |
| 401 |
} |
|
| 402 |
}) |
|
| 403 | ||
| 404 | ! |
output$ui_histogram_display <- renderUI({
|
| 405 | ! |
validation_checks() |
| 406 | ! |
dataname <- input$tabset_panel |
| 407 | ! |
varname <- plot_var$variable[[dataname]] |
| 408 | ! |
df <- data()[[dataname]] |
| 409 | ||
| 410 | ! |
numeric_ui <- tagList(fluidRow( |
| 411 | ! |
tags$div( |
| 412 | ! |
class = "col-md-4", |
| 413 | ! |
shinyWidgets::switchInput( |
| 414 | ! |
inputId = session$ns("remove_NA_hist"),
|
| 415 | ! |
label = "Remove NA values", |
| 416 | ! |
value = FALSE, |
| 417 | ! |
width = "50%", |
| 418 | ! |
labelWidth = "100px", |
| 419 | ! |
handleWidth = "50px" |
| 420 |
) |
|
| 421 |
) |
|
| 422 |
)) |
|
| 423 | ||
| 424 | ! |
var <- df[[varname]] |
| 425 | ! |
if (anyNA(var) && (is.factor(var) || is.character(var) || is.logical(var))) {
|
| 426 | ! |
groups <- unique(as.character(var)) |
| 427 | ! |
len_groups <- length(groups) |
| 428 | ! |
if (len_groups >= .unique_records_for_factor) {
|
| 429 | ! |
NULL |
| 430 |
} else {
|
|
| 431 | ! |
numeric_ui |
| 432 |
} |
|
| 433 |
} else {
|
|
| 434 | ! |
NULL |
| 435 |
} |
|
| 436 |
}) |
|
| 437 | ||
| 438 | ! |
output$outlier_definition_slider_ui <- renderUI({
|
| 439 | ! |
req(input$remove_outliers) |
| 440 | ! |
sliderInput( |
| 441 | ! |
inputId = session$ns("outlier_definition_slider"),
|
| 442 | ! |
tags$div( |
| 443 | ! |
class = "teal-tooltip", |
| 444 | ! |
tagList( |
| 445 | ! |
"Outlier definition:", |
| 446 | ! |
icon("circle-info"),
|
| 447 | ! |
tags$span( |
| 448 | ! |
class = "tooltiptext", |
| 449 | ! |
paste( |
| 450 | ! |
"Use the slider to choose the cut-off value to define outliers; the larger the value the", |
| 451 | ! |
"further below Q1/above Q3 points have to be in order to be classed as outliers" |
| 452 |
) |
|
| 453 |
) |
|
| 454 |
) |
|
| 455 |
), |
|
| 456 | ! |
min = 1, |
| 457 | ! |
max = 5, |
| 458 | ! |
value = 3, |
| 459 | ! |
step = 0.5 |
| 460 |
) |
|
| 461 |
}) |
|
| 462 | ||
| 463 | ! |
output$ui_density_help <- renderUI({
|
| 464 | ! |
req(is.logical(input$display_density)) |
| 465 | ! |
if (input$display_density) {
|
| 466 | ! |
tags$small(helpText(paste( |
| 467 | ! |
"Kernel density estimation with gaussian kernel", |
| 468 | ! |
"and bandwidth function bw.nrd0 (R default)" |
| 469 |
))) |
|
| 470 |
} else {
|
|
| 471 | ! |
NULL |
| 472 |
} |
|
| 473 |
}) |
|
| 474 | ||
| 475 | ! |
output$ui_outlier_help <- renderUI({
|
| 476 | ! |
req(is.logical(input$remove_outliers), input$outlier_definition_slider) |
| 477 | ! |
if (input$remove_outliers) {
|
| 478 | ! |
tags$small( |
| 479 | ! |
helpText( |
| 480 | ! |
withMathJax(paste0( |
| 481 | ! |
"Outlier data points (\\( X \\lt Q1 - ", input$outlier_definition_slider, "\\times IQR \\) or |
| 482 | ! |
\\(Q3 + ", input$outlier_definition_slider, "\\times IQR \\lt X\\)) |
| 483 | ! |
have not been displayed on the graph and will not be used for any kernel density estimations, ", |
| 484 | ! |
"although their values remain in the statisics table below." |
| 485 |
)) |
|
| 486 |
) |
|
| 487 |
) |
|
| 488 |
} else {
|
|
| 489 | ! |
NULL |
| 490 |
} |
|
| 491 |
}) |
|
| 492 | ||
| 493 | ||
| 494 | ! |
variable_plot_r <- reactive({
|
| 495 | ! |
display_density <- `if`(is.null(input$display_density), FALSE, input$display_density) |
| 496 | ! |
remove_outliers <- `if`(is.null(input$remove_outliers), FALSE, input$remove_outliers) |
| 497 | ||
| 498 | ! |
if (remove_outliers) {
|
| 499 | ! |
req(input$outlier_definition_slider) |
| 500 | ! |
outlier_definition <- as.numeric(input$outlier_definition_slider) |
| 501 |
} else {
|
|
| 502 | ! |
outlier_definition <- 0 |
| 503 |
} |
|
| 504 | ||
| 505 | ! |
plot_var_summary( |
| 506 | ! |
var = plotted_data()$data, |
| 507 | ! |
var_lab = plotted_data()$var_description, |
| 508 | ! |
wrap_character = 15, |
| 509 | ! |
numeric_as_factor = treat_numeric_as_factor(), |
| 510 | ! |
remove_NA_hist = input$remove_NA_hist, |
| 511 | ! |
display_density = display_density, |
| 512 | ! |
outlier_definition = outlier_definition, |
| 513 | ! |
records_for_factor = .unique_records_for_factor, |
| 514 | ! |
ggplot2_args = all_ggplot2_args() |
| 515 |
) |
|
| 516 |
}) |
|
| 517 | ||
| 518 | ! |
pws <- teal.widgets::plot_with_settings_srv( |
| 519 | ! |
id = "variable_plot", |
| 520 | ! |
plot_r = variable_plot_r, |
| 521 | ! |
height = c(500, 200, 2000) |
| 522 |
) |
|
| 523 | ||
| 524 | ! |
output$variable_summary_table <- DT::renderDataTable({
|
| 525 | ! |
var_summary_table( |
| 526 | ! |
plotted_data()$data, |
| 527 | ! |
treat_numeric_as_factor(), |
| 528 | ! |
input$variable_summary_table_rows, |
| 529 | ! |
if (!is.null(input$remove_outliers) && input$remove_outliers) {
|
| 530 | ! |
req(input$outlier_definition_slider) |
| 531 | ! |
as.numeric(input$outlier_definition_slider) |
| 532 |
} else {
|
|
| 533 | ! |
0 |
| 534 |
} |
|
| 535 |
) |
|
| 536 |
}) |
|
| 537 | ||
| 538 |
### REPORTER |
|
| 539 | ! |
if (with_reporter) {
|
| 540 | ! |
card_fun <- function(comment) {
|
| 541 | ! |
card <- teal::TealReportCard$new() |
| 542 | ! |
card$set_name("Variable Browser Plot")
|
| 543 | ! |
card$append_text("Variable Browser Plot", "header2")
|
| 544 | ! |
if (with_filter) card$append_fs(filter_panel_api$get_filter_state()) |
| 545 | ! |
card$append_text("Plot", "header3")
|
| 546 | ! |
card$append_plot(variable_plot_r(), dim = pws$dim()) |
| 547 | ! |
if (!comment == "") {
|
| 548 | ! |
card$append_text("Comment", "header3")
|
| 549 | ! |
card$append_text(comment) |
| 550 |
} |
|
| 551 | ! |
card |
| 552 |
} |
|
| 553 | ! |
teal.reporter::simple_reporter_srv("simple_reporter", reporter = reporter, card_fun = card_fun)
|
| 554 |
} |
|
| 555 |
### |
|
| 556 |
}) |
|
| 557 |
} |
|
| 558 | ||
| 559 |
#' Summarize NAs. |
|
| 560 |
#' |
|
| 561 |
#' Summarizes occurrence of missing values in vector. |
|
| 562 |
#' @param x vector of any type and length |
|
| 563 |
#' @return Character string describing `NA` occurrence. |
|
| 564 |
#' @keywords internal |
|
| 565 |
var_missings_info <- function(x) {
|
|
| 566 | ! |
sprintf("%s [%s%%]", sum(is.na(x)), round(mean(is.na(x) * 100), 2))
|
| 567 |
} |
|
| 568 | ||
| 569 |
#' Summarizes variable |
|
| 570 |
#' |
|
| 571 |
#' Creates html summary with statistics relevant to data type. For numeric values it returns central |
|
| 572 |
#' tendency measures, for factor returns level counts, for Date date range, for other just |
|
| 573 |
#' number of levels. |
|
| 574 |
#' |
|
| 575 |
#' @param x vector of any type |
|
| 576 |
#' @param numeric_as_factor `logical` should the numeric variable be treated as a factor |
|
| 577 |
#' @param dt_rows `numeric` current/latest `DT` page length |
|
| 578 |
#' @param outlier_definition If 0 no outliers are removed, otherwise |
|
| 579 |
#' outliers (those more than `outlier_definition*IQR below/above Q1/Q3` be removed) |
|
| 580 |
#' @return text with simple statistics. |
|
| 581 |
#' @keywords internal |
|
| 582 |
var_summary_table <- function(x, numeric_as_factor, dt_rows, outlier_definition) {
|
|
| 583 | ! |
if (is.null(dt_rows)) {
|
| 584 | ! |
dt_rows <- 10 |
| 585 |
} |
|
| 586 | ! |
if (is.numeric(x) && !numeric_as_factor) {
|
| 587 | ! |
req(!any(is.infinite(x))) |
| 588 | ||
| 589 | ! |
x <- remove_outliers_from(x, outlier_definition) |
| 590 | ||
| 591 | ! |
qvals <- round(stats::quantile(x, na.rm = TRUE, probs = c(0.25, 0.5, 0.75), type = 2), 2) |
| 592 |
# classical central tendency measures |
|
| 593 | ||
| 594 | ! |
summary <- |
| 595 | ! |
data.frame( |
| 596 | ! |
Statistic = c("min", "Q1", "median", "mean", "Q3", "max", "sd", "n"),
|
| 597 | ! |
Value = c( |
| 598 | ! |
round(min(x, na.rm = TRUE), 2), |
| 599 | ! |
qvals[1], |
| 600 | ! |
qvals[2], |
| 601 | ! |
round(mean(x, na.rm = TRUE), 2), |
| 602 | ! |
qvals[3], |
| 603 | ! |
round(max(x, na.rm = TRUE), 2), |
| 604 | ! |
round(stats::sd(x, na.rm = TRUE), 2), |
| 605 | ! |
length(x[!is.na(x)]) |
| 606 |
) |
|
| 607 |
) |
|
| 608 | ||
| 609 | ! |
DT::datatable(summary, rownames = FALSE, options = list(dom = "<t>", pageLength = dt_rows)) |
| 610 | ! |
} else if (is.factor(x) || is.character(x) || (is.numeric(x) && numeric_as_factor) || is.logical(x)) {
|
| 611 |
# make sure factor is ordered numeric |
|
| 612 | ! |
if (is.numeric(x)) {
|
| 613 | ! |
x <- factor(x, levels = sort(unique(x))) |
| 614 |
} |
|
| 615 | ||
| 616 | ! |
level_counts <- table(x) |
| 617 | ! |
max_levels_signif <- nchar(level_counts) |
| 618 | ||
| 619 | ! |
if (!all(is.na(x))) {
|
| 620 | ! |
levels <- names(level_counts) |
| 621 | ! |
counts <- sprintf( |
| 622 | ! |
"%s [%.2f%%]", |
| 623 | ! |
format(level_counts, width = max_levels_signif), prop.table(level_counts) * 100 |
| 624 |
) |
|
| 625 |
} else {
|
|
| 626 | ! |
levels <- character(0) |
| 627 | ! |
counts <- numeric(0) |
| 628 |
} |
|
| 629 | ||
| 630 | ! |
summary <- data.frame( |
| 631 | ! |
Level = levels, |
| 632 | ! |
Count = counts, |
| 633 | ! |
stringsAsFactors = FALSE |
| 634 |
) |
|
| 635 | ||
| 636 |
# sort the dataset in decreasing order of counts (needed as character variables default to alphabetical) |
|
| 637 | ! |
summary <- summary[order(summary$Count, decreasing = TRUE), ] |
| 638 | ||
| 639 | ! |
dom_opts <- if (nrow(summary) <= 10) {
|
| 640 | ! |
"<t>" |
| 641 |
} else {
|
|
| 642 | ! |
"<lf<t>ip>" |
| 643 |
} |
|
| 644 | ! |
DT::datatable(summary, rownames = FALSE, options = list(dom = dom_opts, pageLength = dt_rows)) |
| 645 | ! |
} else if (inherits(x, "Date") || inherits(x, "POSIXct") || inherits(x, "POSIXlt")) {
|
| 646 | ! |
summary <- |
| 647 | ! |
data.frame( |
| 648 | ! |
Statistic = c("min", "median", "max"),
|
| 649 | ! |
Value = c( |
| 650 | ! |
min(x, na.rm = TRUE), |
| 651 | ! |
stats::median(x, na.rm = TRUE), |
| 652 | ! |
max(x, na.rm = TRUE) |
| 653 |
) |
|
| 654 |
) |
|
| 655 | ! |
DT::datatable(summary, rownames = FALSE, options = list(dom = "<t>", pageLength = dt_rows)) |
| 656 |
} else {
|
|
| 657 | ! |
NULL |
| 658 |
} |
|
| 659 |
} |
|
| 660 | ||
| 661 |
#' Plot variable |
|
| 662 |
#' |
|
| 663 |
#' Creates summary plot with statistics relevant to data type. |
|
| 664 |
#' |
|
| 665 |
#' @inheritParams shared_params |
|
| 666 |
#' @param var vector of any type to be plotted. For numeric variables it produces histogram with |
|
| 667 |
#' density line, for factors it creates frequency plot |
|
| 668 |
#' @param var_lab text describing selected variable to be displayed on the plot |
|
| 669 |
#' @param wrap_character (`numeric`) number of characters at which to wrap text values of `var` |
|
| 670 |
#' @param numeric_as_factor (`logical`) should the numeric variable be treated as a factor |
|
| 671 |
#' @param display_density (`logical`) should density estimation be displayed for numeric values |
|
| 672 |
#' @param remove_NA_hist (`logical`) should `NA` values be removed for histogram of factor like variables |
|
| 673 |
#' @param outlier_definition if 0 no outliers are removed, otherwise |
|
| 674 |
#' outliers (those more than outlier_definition*IQR below/above Q1/Q3 be removed) |
|
| 675 |
#' @param records_for_factor (`numeric`) if the number of factor levels is >= than this value then |
|
| 676 |
#' a graph of the factors isn't shown, only a list of values |
|
| 677 |
#' |
|
| 678 |
#' @return plot |
|
| 679 |
#' @keywords internal |
|
| 680 |
plot_var_summary <- function(var, |
|
| 681 |
var_lab, |
|
| 682 |
wrap_character = NULL, |
|
| 683 |
numeric_as_factor, |
|
| 684 |
display_density = is.numeric(var), |
|
| 685 |
remove_NA_hist = FALSE, # nolint: object_name. |
|
| 686 |
outlier_definition, |
|
| 687 |
records_for_factor, |
|
| 688 |
ggplot2_args) {
|
|
| 689 | ! |
checkmate::assert_character(var_lab) |
| 690 | ! |
checkmate::assert_numeric(wrap_character, null.ok = TRUE) |
| 691 | ! |
checkmate::assert_flag(numeric_as_factor) |
| 692 | ! |
checkmate::assert_flag(display_density) |
| 693 | ! |
checkmate::assert_logical(remove_NA_hist, null.ok = TRUE) |
| 694 | ! |
checkmate::assert_number(outlier_definition, lower = 0, finite = TRUE) |
| 695 | ! |
checkmate::assert_integerish(records_for_factor, lower = 0, len = 1, any.missing = FALSE) |
| 696 | ! |
checkmate::assert_class(ggplot2_args, "ggplot2_args") |
| 697 | ||
| 698 | ! |
grid::grid.newpage() |
| 699 | ||
| 700 | ! |
plot_main <- if (is.factor(var) || is.character(var) || is.logical(var)) {
|
| 701 | ! |
groups <- unique(as.character(var)) |
| 702 | ! |
len_groups <- length(groups) |
| 703 | ! |
if (len_groups >= records_for_factor) {
|
| 704 | ! |
grid::textGrob( |
| 705 | ! |
sprintf( |
| 706 | ! |
"%s unique values\n%s:\n %s\n ...\n %s", |
| 707 | ! |
len_groups, |
| 708 | ! |
var_lab, |
| 709 | ! |
paste(utils::head(groups), collapse = ",\n "), |
| 710 | ! |
paste(utils::tail(groups), collapse = ",\n ") |
| 711 |
), |
|
| 712 | ! |
x = grid::unit(1, "line"), |
| 713 | ! |
y = grid::unit(1, "npc") - grid::unit(1, "line"), |
| 714 | ! |
just = c("left", "top")
|
| 715 |
) |
|
| 716 |
} else {
|
|
| 717 | ! |
if (!is.null(wrap_character)) {
|
| 718 | ! |
var <- stringr::str_wrap(var, width = wrap_character) |
| 719 |
} |
|
| 720 | ! |
var <- if (isTRUE(remove_NA_hist)) as.vector(stats::na.omit(var)) else var |
| 721 | ! |
ggplot(data.frame(var), aes(x = forcats::fct_infreq(as.factor(var)))) + |
| 722 | ! |
geom_bar( |
| 723 | ! |
stat = "count", aes(fill = ifelse(is.na(var), "withcolor", "")), show.legend = FALSE |
| 724 |
) + |
|
| 725 | ! |
scale_fill_manual(values = c("gray50", "tan"))
|
| 726 |
} |
|
| 727 | ! |
} else if (is.numeric(var)) {
|
| 728 | ! |
validate(need(any(!is.na(var)), "No data left to visualize.")) |
| 729 | ||
| 730 |
# Filter out NA |
|
| 731 | ! |
var <- var[which(!is.na(var))] |
| 732 | ||
| 733 | ! |
validate(need(!any(is.infinite(var)), "Cannot display graph when data includes infinite values")) |
| 734 | ||
| 735 | ! |
if (numeric_as_factor) {
|
| 736 | ! |
var <- factor(var) |
| 737 | ! |
ggplot(NULL, aes(x = var)) + |
| 738 | ! |
geom_histogram(stat = "count") |
| 739 |
} else {
|
|
| 740 |
# remove outliers |
|
| 741 | ! |
if (outlier_definition != 0) {
|
| 742 | ! |
number_records <- length(var) |
| 743 | ! |
var <- remove_outliers_from(var, outlier_definition) |
| 744 | ! |
number_outliers <- number_records - length(var) |
| 745 | ! |
outlier_text <- paste0( |
| 746 | ! |
number_outliers, " outliers (",
|
| 747 | ! |
round(number_outliers / number_records * 100, 2), |
| 748 | ! |
"% of non-missing records) not shown" |
| 749 |
) |
|
| 750 | ! |
validate(need( |
| 751 | ! |
length(var) > 1, |
| 752 | ! |
"At least two data points must remain after removing outliers for this graph to be displayed" |
| 753 |
)) |
|
| 754 |
} |
|
| 755 |
## histogram |
|
| 756 | ! |
binwidth <- get_bin_width(var) |
| 757 | ! |
p <- ggplot(data = data.frame(var = var), aes(x = var, y = after_stat(count))) + |
| 758 | ! |
geom_histogram(binwidth = binwidth) + |
| 759 | ! |
scale_y_continuous( |
| 760 | ! |
sec.axis = sec_axis( |
| 761 | ! |
trans = ~ . / nrow(data.frame(var = var)), |
| 762 | ! |
labels = scales::percent, |
| 763 | ! |
name = "proportion (in %)" |
| 764 |
) |
|
| 765 |
) |
|
| 766 | ||
| 767 | ! |
if (display_density) {
|
| 768 | ! |
p <- p + geom_density(aes(y = after_stat(count * binwidth))) |
| 769 |
} |
|
| 770 | ||
| 771 | ! |
if (outlier_definition != 0) {
|
| 772 | ! |
p <- p + annotate( |
| 773 | ! |
geom = "text", |
| 774 | ! |
label = outlier_text, |
| 775 | ! |
x = Inf, y = Inf, |
| 776 | ! |
hjust = 1.02, vjust = 1.2, |
| 777 | ! |
color = "black", |
| 778 |
# explicitly modify geom text size according |
|
| 779 | ! |
size = ggplot2_args[["theme"]][["text"]][["size"]] / 3.5 |
| 780 |
) |
|
| 781 |
} |
|
| 782 | ! |
p |
| 783 |
} |
|
| 784 | ! |
} else if (inherits(var, "Date") || inherits(var, "POSIXct") || inherits(var, "POSIXlt")) {
|
| 785 | ! |
var_num <- as.numeric(var) |
| 786 | ! |
binwidth <- get_bin_width(var_num, 1) |
| 787 | ! |
p <- ggplot(data = data.frame(var = var), aes(x = var, y = after_stat(count))) + |
| 788 | ! |
geom_histogram(binwidth = binwidth) |
| 789 |
} else {
|
|
| 790 | ! |
grid::textGrob( |
| 791 | ! |
paste(strwrap( |
| 792 | ! |
utils::capture.output(utils::str(var)), |
| 793 | ! |
width = .9 * grid::convertWidth(grid::unit(1, "npc"), "char", TRUE) |
| 794 | ! |
), collapse = "\n"), |
| 795 | ! |
x = grid::unit(1, "line"), y = grid::unit(1, "npc") - grid::unit(1, "line"), just = c("left", "top")
|
| 796 |
) |
|
| 797 |
} |
|
| 798 | ||
| 799 | ! |
dev_ggplot2_args <- teal.widgets::ggplot2_args( |
| 800 | ! |
labs = list(x = var_lab) |
| 801 |
) |
|
| 802 |
### |
|
| 803 | ! |
all_ggplot2_args <- teal.widgets::resolve_ggplot2_args( |
| 804 | ! |
ggplot2_args, |
| 805 | ! |
module_plot = dev_ggplot2_args |
| 806 |
) |
|
| 807 | ||
| 808 | ! |
if (is.ggplot(plot_main)) {
|
| 809 | ! |
if (is.numeric(var) && !numeric_as_factor) {
|
| 810 |
# numeric not as factor |
|
| 811 | ! |
plot_main <- plot_main + |
| 812 | ! |
theme_light() + |
| 813 | ! |
list( |
| 814 | ! |
labs = do.call("labs", all_ggplot2_args$labs),
|
| 815 | ! |
theme = do.call("theme", all_ggplot2_args$theme)
|
| 816 |
) |
|
| 817 |
} else {
|
|
| 818 |
# factor low number of levels OR numeric as factor OR Date |
|
| 819 | ! |
plot_main <- plot_main + |
| 820 | ! |
theme_light() + |
| 821 | ! |
list( |
| 822 | ! |
labs = do.call("labs", all_ggplot2_args$labs),
|
| 823 | ! |
theme = do.call("theme", all_ggplot2_args$theme)
|
| 824 |
) |
|
| 825 |
} |
|
| 826 | ! |
plot_main <- ggplotGrob(plot_main) |
| 827 |
} |
|
| 828 | ||
| 829 | ! |
grid::grid.draw(plot_main) |
| 830 | ! |
plot_main |
| 831 |
} |
|
| 832 | ||
| 833 |
is_num_var_short <- function(.unique_records_for_factor, input, data_for_analysis) {
|
|
| 834 | ! |
length(unique(data_for_analysis()$data)) < .unique_records_for_factor && !is.null(input$numeric_as_factor) |
| 835 |
} |
|
| 836 | ||
| 837 |
#' Validates the variable browser inputs |
|
| 838 |
#' |
|
| 839 |
#' @param input (`session$input`) the `shiny` session input |
|
| 840 |
#' @param plot_var (`list`) list of a data frame and an array of variable names |
|
| 841 |
#' @param data (`teal_data`) the datasets passed to the module |
|
| 842 |
#' |
|
| 843 |
#' @returns `logical` TRUE if validations pass; a `shiny` validation error otherwise |
|
| 844 |
#' @keywords internal |
|
| 845 |
validate_input <- function(input, plot_var, data) {
|
|
| 846 | ! |
reactive({
|
| 847 | ! |
dataset_name <- req(input$tabset_panel) |
| 848 | ! |
varname <- plot_var$variable[[dataset_name]] |
| 849 | ||
| 850 | ! |
validate(need(dataset_name, "No data selected")) |
| 851 | ! |
validate(need(varname, "No variable selected")) |
| 852 | ! |
df <- data()[[dataset_name]] |
| 853 | ! |
teal::validate_has_data(df, 1) |
| 854 | ! |
teal::validate_has_variable(varname = varname, data = df, "Variable not available") |
| 855 | ||
| 856 | ! |
TRUE |
| 857 |
}) |
|
| 858 |
} |
|
| 859 | ||
| 860 |
get_plotted_data <- function(input, plot_var, data) {
|
|
| 861 | ! |
dataset_name <- input$tabset_panel |
| 862 | ! |
varname <- plot_var$variable[[dataset_name]] |
| 863 | ! |
df <- data()[[dataset_name]] |
| 864 | ||
| 865 | ! |
var_description <- teal.data::col_labels(df)[[varname]] |
| 866 | ! |
list(data = df[[varname]], var_description = var_description) |
| 867 |
} |
|
| 868 | ||
| 869 |
#' Renders the left-hand side `tabset` panel of the module |
|
| 870 |
#' |
|
| 871 |
#' @param datanames (`character`) the name of the dataset |
|
| 872 |
#' @param parent_dataname (`character`) the name of a parent `dataname` to filter out variables from |
|
| 873 |
#' @param data (`teal_data`) the object containing all datasets |
|
| 874 |
#' @param input (`session$input`) the `shiny` session input |
|
| 875 |
#' @param output (`session$output`) the `shiny` session output |
|
| 876 |
#' @param columns_names (`environment`) the environment containing bindings for each dataset |
|
| 877 |
#' @param plot_var (`list`) the list containing the currently selected dataset (tab) and its column names |
|
| 878 |
#' @keywords internal |
|
| 879 |
render_tabset_panel_content <- function(datanames, parent_dataname, output, data, input, columns_names, plot_var) {
|
|
| 880 | ! |
lapply(datanames, render_single_tab, |
| 881 | ! |
input = input, |
| 882 | ! |
output = output, |
| 883 | ! |
data = data, |
| 884 | ! |
parent_dataname = parent_dataname, |
| 885 | ! |
columns_names = columns_names, |
| 886 | ! |
plot_var = plot_var |
| 887 |
) |
|
| 888 |
} |
|
| 889 | ||
| 890 |
#' Renders a single tab in the left-hand side tabset panel |
|
| 891 |
#' |
|
| 892 |
#' Renders a single tab in the left-hand side tabset panel. The rendered tab contains |
|
| 893 |
#' information about one dataset out of many presented in the module. |
|
| 894 |
#' |
|
| 895 |
#' @param dataset_name (`character`) the name of the dataset contained in the rendered tab |
|
| 896 |
#' @param parent_dataname (`character`) the name of a parent `dataname` to filter out variables from |
|
| 897 |
#' @inheritParams render_tabset_panel_content |
|
| 898 |
#' @keywords internal |
|
| 899 |
render_single_tab <- function(dataset_name, parent_dataname, output, data, input, columns_names, plot_var) {
|
|
| 900 | ! |
render_tab_header(dataset_name, output, data) |
| 901 | ||
| 902 | ! |
render_tab_table( |
| 903 | ! |
dataset_name = dataset_name, |
| 904 | ! |
parent_dataname = parent_dataname, |
| 905 | ! |
output = output, |
| 906 | ! |
data = data, |
| 907 | ! |
input = input, |
| 908 | ! |
columns_names = columns_names, |
| 909 | ! |
plot_var = plot_var |
| 910 |
) |
|
| 911 |
} |
|
| 912 | ||
| 913 |
#' Renders the text headlining a single tab in the left-hand side tabset panel |
|
| 914 |
#' |
|
| 915 |
#' @param dataset_name (`character`) the name of the dataset of the tab |
|
| 916 |
#' @inheritParams render_tabset_panel_content |
|
| 917 |
#' @keywords internal |
|
| 918 |
render_tab_header <- function(dataset_name, output, data) {
|
|
| 919 | ! |
dataset_ui_id <- paste0("dataset_summary_", dataset_name)
|
| 920 | ! |
output[[dataset_ui_id]] <- renderText({
|
| 921 | ! |
df <- data()[[dataset_name]] |
| 922 | ! |
join_keys <- teal.data::join_keys(data()) |
| 923 | ! |
if (!is.null(join_keys)) {
|
| 924 | ! |
key <- teal.data::join_keys(data())[dataset_name, dataset_name] |
| 925 |
} else {
|
|
| 926 | ! |
key <- NULL |
| 927 |
} |
|
| 928 | ! |
sprintf( |
| 929 | ! |
"Dataset with %s unique key rows and %s variables", |
| 930 | ! |
nrow(unique(`if`(length(key) > 0, df[, key, drop = FALSE], df))), |
| 931 | ! |
ncol(df) |
| 932 |
) |
|
| 933 |
}) |
|
| 934 |
} |
|
| 935 | ||
| 936 |
#' Renders the table for a single dataset in the left-hand side tabset panel |
|
| 937 |
#' |
|
| 938 |
#' The table contains column names, column labels, |
|
| 939 |
#' small summary about NA values and `sparkline` (if appropriate). |
|
| 940 |
#' |
|
| 941 |
#' @param dataset_name (`character`) the name of the dataset |
|
| 942 |
#' @param parent_dataname (`character`) the name of a parent `dataname` to filter out variables from |
|
| 943 |
#' @inheritParams render_tabset_panel_content |
|
| 944 |
#' @keywords internal |
|
| 945 |
render_tab_table <- function(dataset_name, parent_dataname, output, data, input, columns_names, plot_var) {
|
|
| 946 | ! |
table_ui_id <- paste0("variable_browser_", dataset_name)
|
| 947 | ||
| 948 | ! |
output[[table_ui_id]] <- DT::renderDataTable({
|
| 949 | ! |
df <- data()[[dataset_name]] |
| 950 | ||
| 951 | ! |
get_vars_df <- function(input, dataset_name, parent_name, data) {
|
| 952 | ! |
data_cols <- colnames(df) |
| 953 | ! |
if (isTRUE(input$show_parent_vars)) {
|
| 954 | ! |
data_cols |
| 955 | ! |
} else if (dataset_name != parent_name && parent_name %in% names(data)) {
|
| 956 | ! |
setdiff(data_cols, colnames(data()[[parent_name]])) |
| 957 |
} else {
|
|
| 958 | ! |
data_cols |
| 959 |
} |
|
| 960 |
} |
|
| 961 | ||
| 962 | ! |
if (length(parent_dataname) > 0) {
|
| 963 | ! |
df_vars <- get_vars_df(input, dataset_name, parent_dataname, data) |
| 964 | ! |
df <- df[df_vars] |
| 965 |
} |
|
| 966 | ||
| 967 | ! |
if (is.null(df) || ncol(df) == 0) {
|
| 968 | ! |
columns_names[[dataset_name]] <- character(0) |
| 969 | ! |
df_output <- data.frame( |
| 970 | ! |
Type = character(0), |
| 971 | ! |
Variable = character(0), |
| 972 | ! |
Label = character(0), |
| 973 | ! |
Missings = character(0), |
| 974 | ! |
Sparklines = character(0), |
| 975 | ! |
stringsAsFactors = FALSE |
| 976 |
) |
|
| 977 |
} else {
|
|
| 978 |
# extract data variable labels |
|
| 979 | ! |
labels <- teal.data::col_labels(df) |
| 980 | ||
| 981 | ! |
columns_names[[dataset_name]] <- names(labels) |
| 982 | ||
| 983 |
# calculate number of missing values |
|
| 984 | ! |
missings <- vapply( |
| 985 | ! |
df, |
| 986 | ! |
var_missings_info, |
| 987 | ! |
FUN.VALUE = character(1), |
| 988 | ! |
USE.NAMES = FALSE |
| 989 |
) |
|
| 990 | ||
| 991 |
# get icons proper for the data types |
|
| 992 | ! |
icons <- vapply(df, function(x) class(x)[1L], character(1L)) |
| 993 | ||
| 994 | ! |
join_keys <- teal.data::join_keys(data()) |
| 995 | ! |
if (!is.null(join_keys)) {
|
| 996 | ! |
icons[intersect(join_keys[dataset_name, dataset_name], colnames(df))] <- "primary_key" |
| 997 |
} |
|
| 998 | ! |
icons <- variable_type_icons(icons) |
| 999 | ||
| 1000 |
# generate sparklines |
|
| 1001 | ! |
sparklines_html <- vapply( |
| 1002 | ! |
df, |
| 1003 | ! |
create_sparklines, |
| 1004 | ! |
FUN.VALUE = character(1), |
| 1005 | ! |
USE.NAMES = FALSE |
| 1006 |
) |
|
| 1007 | ||
| 1008 | ! |
df_output <- data.frame( |
| 1009 | ! |
Type = icons, |
| 1010 | ! |
Variable = names(labels), |
| 1011 | ! |
Label = labels, |
| 1012 | ! |
Missings = missings, |
| 1013 | ! |
Sparklines = sparklines_html, |
| 1014 | ! |
stringsAsFactors = FALSE |
| 1015 |
) |
|
| 1016 |
} |
|
| 1017 | ||
| 1018 |
# Select row 1 as default / fallback |
|
| 1019 | ! |
selected_ix <- 1 |
| 1020 |
# Define starting page index (base-0 index of the first item on page |
|
| 1021 |
# note: in many cases it's not the item itself |
|
| 1022 | ! |
selected_page_ix <- 0 |
| 1023 | ||
| 1024 |
# Retrieve current selected variable if any |
|
| 1025 | ! |
isolated_variable <- isolate(plot_var$variable[[dataset_name]]) |
| 1026 | ||
| 1027 | ! |
if (!is.null(isolated_variable)) {
|
| 1028 | ! |
index <- which(columns_names[[dataset_name]] == isolated_variable)[1] |
| 1029 | ! |
if (!is.null(index) && !is.na(index) && length(index) > 0) selected_ix <- index |
| 1030 |
} |
|
| 1031 | ||
| 1032 |
# Retrieve the index of the first item of the current page |
|
| 1033 |
# it works with varying number of entries on the page (10, 25, ...) |
|
| 1034 | ! |
table_id_sel <- paste0("variable_browser_", dataset_name, "_state")
|
| 1035 | ! |
dt_state <- isolate(input[[table_id_sel]]) |
| 1036 | ! |
if (selected_ix != 1 && !is.null(dt_state)) {
|
| 1037 | ! |
selected_page_ix <- floor(selected_ix / dt_state$length) * dt_state$length |
| 1038 |
} |
|
| 1039 | ||
| 1040 | ! |
DT::datatable( |
| 1041 | ! |
df_output, |
| 1042 | ! |
escape = FALSE, |
| 1043 | ! |
rownames = FALSE, |
| 1044 | ! |
selection = list(mode = "single", target = "row", selected = selected_ix), |
| 1045 | ! |
options = list( |
| 1046 | ! |
fnDrawCallback = htmlwidgets::JS("function() { HTMLWidgets.staticRender(); }"),
|
| 1047 | ! |
pageLength = input[[paste0(table_ui_id, "_rows")]], |
| 1048 | ! |
displayStart = selected_page_ix |
| 1049 |
) |
|
| 1050 |
) |
|
| 1051 |
}) |
|
| 1052 |
} |
|
| 1053 | ||
| 1054 |
#' Creates observers updating the currently selected column |
|
| 1055 |
#' |
|
| 1056 |
#' The created observers update the column currently selected in the left-hand side |
|
| 1057 |
#' tabset panel. |
|
| 1058 |
#' |
|
| 1059 |
#' @note |
|
| 1060 |
#' Creates an observer for each dataset (each tab in the tabset panel). |
|
| 1061 |
#' |
|
| 1062 |
#' @inheritParams render_tabset_panel_content |
|
| 1063 |
#' @keywords internal |
|
| 1064 |
establish_updating_selection <- function(datanames, input, plot_var, columns_names) {
|
|
| 1065 | ! |
lapply(datanames, function(dataset_name) {
|
| 1066 | ! |
table_ui_id <- paste0("variable_browser_", dataset_name)
|
| 1067 | ! |
table_id_sel <- paste0(table_ui_id, "_rows_selected") |
| 1068 | ! |
observeEvent(input[[table_id_sel]], {
|
| 1069 | ! |
plot_var$data <- dataset_name |
| 1070 | ! |
plot_var$variable[[dataset_name]] <- columns_names[[dataset_name]][input[[table_id_sel]]] |
| 1071 |
}) |
|
| 1072 |
}) |
|
| 1073 |
} |
|
| 1074 | ||
| 1075 |
get_bin_width <- function(x_vec, scaling_factor = 2) {
|
|
| 1076 | ! |
x_vec <- x_vec[!is.na(x_vec)] |
| 1077 | ! |
qntls <- stats::quantile(x_vec, probs = c(0.1, 0.25, 0.75, 0.9), type = 2) |
| 1078 | ! |
iqr <- qntls[3] - qntls[2] |
| 1079 | ! |
binwidth <- max(scaling_factor * iqr / length(x_vec) ^ (1 / 3), sqrt(qntls[4] - qntls[1])) # styler: off |
| 1080 | ! |
binwidth <- ifelse(binwidth == 0, 1, binwidth) |
| 1081 |
# to ensure at least two bins when variable span is very small |
|
| 1082 | ! |
x_span <- diff(range(x_vec)) |
| 1083 | ! |
if (isTRUE(x_span / binwidth >= 2)) binwidth else x_span / 2 |
| 1084 |
} |
|
| 1085 | ||
| 1086 |
#' Removes the outlier observation from an array |
|
| 1087 |
#' |
|
| 1088 |
#' @param var (`numeric`) a numeric vector |
|
| 1089 |
#' @param outlier_definition (`numeric`) if `0` then no outliers are removed, otherwise |
|
| 1090 |
#' outliers (those more than `outlier_definition*IQR below/above Q1/Q3`) are removed |
|
| 1091 |
#' @returns (`numeric`) vector without the outlier values |
|
| 1092 |
#' @keywords internal |
|
| 1093 |
remove_outliers_from <- function(var, outlier_definition) {
|
|
| 1094 | 3x |
if (outlier_definition == 0) {
|
| 1095 | 1x |
return(var) |
| 1096 |
} |
|
| 1097 | 2x |
q1_q3 <- stats::quantile(var, probs = c(0.25, 0.75), type = 2, na.rm = TRUE) |
| 1098 | 2x |
iqr <- q1_q3[2] - q1_q3[1] |
| 1099 | 2x |
var[var >= q1_q3[1] - outlier_definition * iqr & var <= q1_q3[2] + outlier_definition * iqr] |
| 1100 |
} |
|
| 1101 | ||
| 1102 | ||
| 1103 |
# sparklines ---- |
|
| 1104 | ||
| 1105 |
#' S3 generic for `sparkline` widget HTML |
|
| 1106 |
#' |
|
| 1107 |
#' Generates the `sparkline` HTML code corresponding to the input array. |
|
| 1108 |
#' For numeric variables creates a box plot, for character and factors - bar plot. |
|
| 1109 |
#' Produces an empty string for variables of other types. |
|
| 1110 |
#' |
|
| 1111 |
#' @param arr vector of any type and length |
|
| 1112 |
#' @param width `numeric` the width of the `sparkline` widget (pixels) |
|
| 1113 |
#' @param bar_spacing `numeric` the spacing between the bars (in pixels) |
|
| 1114 |
#' @param bar_width `numeric` the width of the bars (in pixels) |
|
| 1115 |
#' @param ... `list` additional options passed to bar plots of `jquery.sparkline`; |
|
| 1116 |
#' see [`jquery.sparkline docs`](https://omnipotent.net/jquery.sparkline/#common) |
|
| 1117 |
#' |
|
| 1118 |
#' @return Character string containing HTML code of the `sparkline` HTML widget. |
|
| 1119 |
#' @keywords internal |
|
| 1120 |
create_sparklines <- function(arr, width = 150, ...) {
|
|
| 1121 | ! |
if (all(is.null(arr))) {
|
| 1122 | ! |
return("")
|
| 1123 |
} |
|
| 1124 | ! |
UseMethod("create_sparklines")
|
| 1125 |
} |
|
| 1126 | ||
| 1127 |
#' @rdname create_sparklines |
|
| 1128 |
#' @keywords internal |
|
| 1129 |
#' @export |
|
| 1130 |
create_sparklines.logical <- function(arr, ...) {
|
|
| 1131 | ! |
create_sparklines(as.factor(arr)) |
| 1132 |
} |
|
| 1133 | ||
| 1134 |
#' @rdname create_sparklines |
|
| 1135 |
#' @keywords internal |
|
| 1136 |
#' @export |
|
| 1137 |
create_sparklines.numeric <- function(arr, width = 150, ...) {
|
|
| 1138 | ! |
if (any(is.infinite(arr))) {
|
| 1139 | ! |
return(as.character(tags$code("infinite values", class = "text-blue")))
|
| 1140 |
} |
|
| 1141 | ! |
if (length(arr) > 100000) {
|
| 1142 | ! |
return(as.character(tags$code("Too many rows (>100000)", class = "text-blue")))
|
| 1143 |
} |
|
| 1144 | ||
| 1145 | ! |
arr <- arr[!is.na(arr)] |
| 1146 | ! |
sparkline::spk_chr(unname(arr), type = "box", width = width, ...) |
| 1147 |
} |
|
| 1148 | ||
| 1149 |
#' @rdname create_sparklines |
|
| 1150 |
#' @keywords internal |
|
| 1151 |
#' @export |
|
| 1152 |
create_sparklines.character <- function(arr, ...) {
|
|
| 1153 | ! |
return(create_sparklines(as.factor(arr))) |
| 1154 |
} |
|
| 1155 | ||
| 1156 | ||
| 1157 |
#' @rdname create_sparklines |
|
| 1158 |
#' @keywords internal |
|
| 1159 |
#' @export |
|
| 1160 |
create_sparklines.factor <- function(arr, width = 150, bar_spacing = 5, bar_width = 20, ...) {
|
|
| 1161 | ! |
decreasing_order <- TRUE |
| 1162 | ||
| 1163 | ! |
counts <- table(arr) |
| 1164 | ! |
if (length(counts) >= 100) {
|
| 1165 | ! |
return(as.character(tags$code("> 99 levels", class = "text-blue")))
|
| 1166 | ! |
} else if (length(counts) == 0) {
|
| 1167 | ! |
return(as.character(tags$code("no levels", class = "text-blue")))
|
| 1168 | ! |
} else if (length(counts) == 1) {
|
| 1169 | ! |
return(as.character(tags$code("one level", class = "text-blue")))
|
| 1170 |
} |
|
| 1171 | ||
| 1172 |
# Summarize the occurences of different levels |
|
| 1173 |
# and get the maximum and minimum number of occurences |
|
| 1174 |
# This is needed for the sparkline to correctly display the bar plots |
|
| 1175 |
# Otherwise they are cropped |
|
| 1176 | ! |
counts <- sort(counts, decreasing = decreasing_order, method = "radix") |
| 1177 | ! |
max_value <- if (decreasing_order) counts[1] else counts[length[counts]] |
| 1178 | ! |
max_value <- unname(max_value) |
| 1179 | ||
| 1180 | ! |
sparkline::spk_chr( |
| 1181 | ! |
unname(counts), |
| 1182 | ! |
type = "bar", |
| 1183 | ! |
chartRangeMin = 0, |
| 1184 | ! |
chartRangeMax = max_value, |
| 1185 | ! |
width = width, |
| 1186 | ! |
barWidth = bar_width, |
| 1187 | ! |
barSpacing = bar_spacing, |
| 1188 | ! |
tooltipFormatter = custom_sparkline_formatter(names(counts), as.vector(counts)) |
| 1189 |
) |
|
| 1190 |
} |
|
| 1191 | ||
| 1192 |
#' @rdname create_sparklines |
|
| 1193 |
#' @keywords internal |
|
| 1194 |
#' @export |
|
| 1195 |
create_sparklines.Date <- function(arr, width = 150, bar_spacing = 5, bar_width = 20, ...) {
|
|
| 1196 | ! |
arr_num <- as.numeric(arr) |
| 1197 | ! |
arr_num <- sort(arr_num, decreasing = FALSE, method = "radix") |
| 1198 | ! |
binwidth <- get_bin_width(arr_num, 1) |
| 1199 | ! |
bins <- floor(diff(range(arr_num)) / binwidth) + 1 |
| 1200 | ! |
if (all(is.na(bins))) {
|
| 1201 | ! |
return(as.character(tags$code("only NA", class = "text-blue")))
|
| 1202 | ! |
} else if (bins == 1) {
|
| 1203 | ! |
return(as.character(tags$code("one date", class = "text-blue")))
|
| 1204 |
} |
|
| 1205 | ! |
counts <- as.vector(unname(base::table(cut(arr_num, breaks = bins)))) |
| 1206 | ! |
max_value <- max(counts) |
| 1207 | ||
| 1208 | ! |
start_bins <- as.integer(seq(1, length(arr_num), length.out = bins)) |
| 1209 | ! |
labels_start <- as.character(as.Date(arr_num[start_bins], origin = as.Date("1970-01-01")))
|
| 1210 | ! |
labels <- paste("Start:", labels_start)
|
| 1211 | ||
| 1212 | ! |
sparkline::spk_chr( |
| 1213 | ! |
unname(counts), |
| 1214 | ! |
type = "bar", |
| 1215 | ! |
chartRangeMin = 0, |
| 1216 | ! |
chartRangeMax = max_value, |
| 1217 | ! |
width = width, |
| 1218 | ! |
barWidth = bar_width, |
| 1219 | ! |
barSpacing = bar_spacing, |
| 1220 | ! |
tooltipFormatter = custom_sparkline_formatter(labels, counts) |
| 1221 |
) |
|
| 1222 |
} |
|
| 1223 | ||
| 1224 |
#' @rdname create_sparklines |
|
| 1225 |
#' @keywords internal |
|
| 1226 |
#' @export |
|
| 1227 |
create_sparklines.POSIXct <- function(arr, width = 150, bar_spacing = 5, bar_width = 20, ...) {
|
|
| 1228 | ! |
arr_num <- as.numeric(arr) |
| 1229 | ! |
arr_num <- sort(arr_num, decreasing = FALSE, method = "radix") |
| 1230 | ! |
binwidth <- get_bin_width(arr_num, 1) |
| 1231 | ! |
bins <- floor(diff(range(arr_num)) / binwidth) + 1 |
| 1232 | ! |
if (all(is.na(bins))) {
|
| 1233 | ! |
return(as.character(tags$code("only NA", class = "text-blue")))
|
| 1234 | ! |
} else if (bins == 1) {
|
| 1235 | ! |
return(as.character(tags$code("one date-time", class = "text-blue")))
|
| 1236 |
} |
|
| 1237 | ! |
counts <- as.vector(unname(base::table(cut(arr_num, breaks = bins)))) |
| 1238 | ! |
max_value <- max(counts) |
| 1239 | ||
| 1240 | ! |
start_bins <- as.integer(seq(1, length(arr_num), length.out = bins)) |
| 1241 | ! |
labels_start <- as.character(format(as.POSIXct(arr_num[start_bins], origin = as.Date("1970-01-01")), "%Y-%m-%d"))
|
| 1242 | ! |
labels <- paste("Start:", labels_start)
|
| 1243 | ||
| 1244 | ! |
sparkline::spk_chr( |
| 1245 | ! |
unname(counts), |
| 1246 | ! |
type = "bar", |
| 1247 | ! |
chartRangeMin = 0, |
| 1248 | ! |
chartRangeMax = max_value, |
| 1249 | ! |
width = width, |
| 1250 | ! |
barWidth = bar_width, |
| 1251 | ! |
barSpacing = bar_spacing, |
| 1252 | ! |
tooltipFormatter = custom_sparkline_formatter(labels, counts) |
| 1253 |
) |
|
| 1254 |
} |
|
| 1255 | ||
| 1256 |
#' @rdname create_sparklines |
|
| 1257 |
#' @keywords internal |
|
| 1258 |
#' @export |
|
| 1259 |
create_sparklines.POSIXlt <- function(arr, width = 150, bar_spacing = 5, bar_width = 20, ...) {
|
|
| 1260 | ! |
arr_num <- as.numeric(arr) |
| 1261 | ! |
arr_num <- sort(arr_num, decreasing = FALSE, method = "radix") |
| 1262 | ! |
binwidth <- get_bin_width(arr_num, 1) |
| 1263 | ! |
bins <- floor(diff(range(arr_num)) / binwidth) + 1 |
| 1264 | ! |
if (all(is.na(bins))) {
|
| 1265 | ! |
return(as.character(tags$code("only NA", class = "text-blue")))
|
| 1266 | ! |
} else if (bins == 1) {
|
| 1267 | ! |
return(as.character(tags$code("one date-time", class = "text-blue")))
|
| 1268 |
} |
|
| 1269 | ! |
counts <- as.vector(unname(base::table(cut(arr_num, breaks = bins)))) |
| 1270 | ! |
max_value <- max(counts) |
| 1271 | ||
| 1272 | ! |
start_bins <- as.integer(seq(1, length(arr_num), length.out = bins)) |
| 1273 | ! |
labels_start <- as.character(format(as.POSIXct(arr_num[start_bins], origin = as.Date("1970-01-01")), "%Y-%m-%d"))
|
| 1274 | ! |
labels <- paste("Start:", labels_start)
|
| 1275 | ||
| 1276 | ! |
sparkline::spk_chr( |
| 1277 | ! |
unname(counts), |
| 1278 | ! |
type = "bar", |
| 1279 | ! |
chartRangeMin = 0, |
| 1280 | ! |
chartRangeMax = max_value, |
| 1281 | ! |
width = width, |
| 1282 | ! |
barWidth = bar_width, |
| 1283 | ! |
barSpacing = bar_spacing, |
| 1284 | ! |
tooltipFormatter = custom_sparkline_formatter(labels, counts) |
| 1285 |
) |
|
| 1286 |
} |
|
| 1287 | ||
| 1288 |
#' @rdname create_sparklines |
|
| 1289 |
#' @keywords internal |
|
| 1290 |
#' @export |
|
| 1291 |
create_sparklines.default <- function(arr, width = 150, ...) {
|
|
| 1292 | ! |
as.character(tags$code("unsupported variable type", class = "text-blue"))
|
| 1293 |
} |
|
| 1294 | ||
| 1295 | ||
| 1296 |
custom_sparkline_formatter <- function(labels, counts) {
|
|
| 1297 | ! |
htmlwidgets::JS( |
| 1298 | ! |
sprintf( |
| 1299 | ! |
"function(sparkline, options, field) {
|
| 1300 | ! |
return 'ID: ' + %s[field[0].offset] + '<br>' + 'Count: ' + %s[field[0].offset]; |
| 1301 |
}", |
|
| 1302 | ! |
jsonlite::toJSON(labels), |
| 1303 | ! |
jsonlite::toJSON(counts) |
| 1304 |
) |
|
| 1305 |
) |
|
| 1306 |
} |
| 1 |
#' `teal` module: Outliers analysis |
|
| 2 |
#' |
|
| 3 |
#' Module to analyze and identify outliers using different methods |
|
| 4 |
#' such as IQR, Z-score, and Percentiles, and offers visualizations including |
|
| 5 |
#' box plots, density plots, and cumulative distribution plots to help interpret the outliers. |
|
| 6 |
#' |
|
| 7 |
#' @inheritParams teal::module |
|
| 8 |
#' @inheritParams shared_params |
|
| 9 |
#' |
|
| 10 |
#' @param outlier_var (`data_extract_spec` or `list` of multiple `data_extract_spec`) |
|
| 11 |
#' Specifies variable(s) to be analyzed for outliers. |
|
| 12 |
#' @param categorical_var (`data_extract_spec` or `list` of multiple `data_extract_spec`) optional, |
|
| 13 |
#' specifies the categorical variable(s) to split the selected outlier variables on. |
|
| 14 |
#' |
|
| 15 |
#' @templateVar ggnames "Boxplot","Density Plot","Cumulative Distribution Plot" |
|
| 16 |
#' @template ggplot2_args_multi |
|
| 17 |
#' |
|
| 18 |
#' @inherit shared_params return |
|
| 19 |
#' |
|
| 20 |
#' @examplesShinylive |
|
| 21 |
#' library(teal.modules.general) |
|
| 22 |
#' interactive <- function() TRUE |
|
| 23 |
#' {{ next_example }}
|
|
| 24 |
#' @examples |
|
| 25 |
#' # general data example |
|
| 26 |
#' data <- teal_data() |
|
| 27 |
#' data <- within(data, {
|
|
| 28 |
#' CO2 <- CO2 |
|
| 29 |
#' CO2[["primary_key"]] <- seq_len(nrow(CO2)) |
|
| 30 |
#' }) |
|
| 31 |
#' datanames(data) <- "CO2" |
|
| 32 |
#' join_keys(data) <- join_keys(join_key("CO2", "CO2", "primary_key"))
|
|
| 33 |
#' |
|
| 34 |
#' vars <- choices_selected(variable_choices(data[["CO2"]], c("Plant", "Type", "Treatment")))
|
|
| 35 |
#' |
|
| 36 |
#' app <- init( |
|
| 37 |
#' data = data, |
|
| 38 |
#' modules = modules( |
|
| 39 |
#' tm_outliers( |
|
| 40 |
#' outlier_var = list( |
|
| 41 |
#' data_extract_spec( |
|
| 42 |
#' dataname = "CO2", |
|
| 43 |
#' select = select_spec( |
|
| 44 |
#' label = "Select variable:", |
|
| 45 |
#' choices = variable_choices(data[["CO2"]], c("conc", "uptake")),
|
|
| 46 |
#' selected = "uptake", |
|
| 47 |
#' multiple = FALSE, |
|
| 48 |
#' fixed = FALSE |
|
| 49 |
#' ) |
|
| 50 |
#' ) |
|
| 51 |
#' ), |
|
| 52 |
#' categorical_var = list( |
|
| 53 |
#' data_extract_spec( |
|
| 54 |
#' dataname = "CO2", |
|
| 55 |
#' filter = filter_spec( |
|
| 56 |
#' vars = vars, |
|
| 57 |
#' choices = value_choices(data[["CO2"]], vars$selected), |
|
| 58 |
#' selected = value_choices(data[["CO2"]], vars$selected), |
|
| 59 |
#' multiple = TRUE |
|
| 60 |
#' ) |
|
| 61 |
#' ) |
|
| 62 |
#' ) |
|
| 63 |
#' ) |
|
| 64 |
#' ) |
|
| 65 |
#' ) |
|
| 66 |
#' if (interactive()) {
|
|
| 67 |
#' shinyApp(app$ui, app$server) |
|
| 68 |
#' } |
|
| 69 |
#' |
|
| 70 |
#' @examplesShinylive |
|
| 71 |
#' library(teal.modules.general) |
|
| 72 |
#' interactive <- function() TRUE |
|
| 73 |
#' {{ next_example }}
|
|
| 74 |
#' @examples |
|
| 75 |
#' # CDISC data example |
|
| 76 |
#' data <- teal_data() |
|
| 77 |
#' data <- within(data, {
|
|
| 78 |
#' ADSL <- rADSL |
|
| 79 |
#' }) |
|
| 80 |
#' datanames(data) <- "ADSL" |
|
| 81 |
#' join_keys(data) <- default_cdisc_join_keys[datanames(data)] |
|
| 82 |
#' |
|
| 83 |
#' fact_vars_adsl <- names(Filter(isTRUE, sapply(data[["ADSL"]], is.factor))) |
|
| 84 |
#' vars <- choices_selected(variable_choices(data[["ADSL"]], fact_vars_adsl)) |
|
| 85 |
#' |
|
| 86 |
#' app <- init( |
|
| 87 |
#' data = data, |
|
| 88 |
#' modules = modules( |
|
| 89 |
#' tm_outliers( |
|
| 90 |
#' outlier_var = list( |
|
| 91 |
#' data_extract_spec( |
|
| 92 |
#' dataname = "ADSL", |
|
| 93 |
#' select = select_spec( |
|
| 94 |
#' label = "Select variable:", |
|
| 95 |
#' choices = variable_choices(data[["ADSL"]], c("AGE", "BMRKR1")),
|
|
| 96 |
#' selected = "AGE", |
|
| 97 |
#' multiple = FALSE, |
|
| 98 |
#' fixed = FALSE |
|
| 99 |
#' ) |
|
| 100 |
#' ) |
|
| 101 |
#' ), |
|
| 102 |
#' categorical_var = list( |
|
| 103 |
#' data_extract_spec( |
|
| 104 |
#' dataname = "ADSL", |
|
| 105 |
#' filter = filter_spec( |
|
| 106 |
#' vars = vars, |
|
| 107 |
#' choices = value_choices(data[["ADSL"]], vars$selected), |
|
| 108 |
#' selected = value_choices(data[["ADSL"]], vars$selected), |
|
| 109 |
#' multiple = TRUE |
|
| 110 |
#' ) |
|
| 111 |
#' ) |
|
| 112 |
#' ) |
|
| 113 |
#' ) |
|
| 114 |
#' ) |
|
| 115 |
#' ) |
|
| 116 |
#' if (interactive()) {
|
|
| 117 |
#' shinyApp(app$ui, app$server) |
|
| 118 |
#' } |
|
| 119 |
#' |
|
| 120 |
#' @export |
|
| 121 |
#' |
|
| 122 |
tm_outliers <- function(label = "Outliers Module", |
|
| 123 |
outlier_var, |
|
| 124 |
categorical_var = NULL, |
|
| 125 |
ggtheme = c("gray", "bw", "linedraw", "light", "dark", "minimal", "classic", "void"),
|
|
| 126 |
ggplot2_args = teal.widgets::ggplot2_args(), |
|
| 127 |
plot_height = c(600, 200, 2000), |
|
| 128 |
plot_width = NULL, |
|
| 129 |
pre_output = NULL, |
|
| 130 |
post_output = NULL) {
|
|
| 131 | ! |
message("Initializing tm_outliers")
|
| 132 | ||
| 133 |
# Normalize the parameters |
|
| 134 | ! |
if (inherits(outlier_var, "data_extract_spec")) outlier_var <- list(outlier_var) |
| 135 | ! |
if (inherits(categorical_var, "data_extract_spec")) categorical_var <- list(categorical_var) |
| 136 | ! |
if (inherits(ggplot2_args, "ggplot2_args")) ggplot2_args <- list(default = ggplot2_args) |
| 137 | ||
| 138 |
# Start of assertions |
|
| 139 | ! |
checkmate::assert_string(label) |
| 140 | ! |
checkmate::assert_list(outlier_var, types = "data_extract_spec") |
| 141 | ||
| 142 | ! |
checkmate::assert_list(categorical_var, types = "data_extract_spec", null.ok = TRUE) |
| 143 | ! |
if (is.list(categorical_var)) {
|
| 144 | ! |
lapply(categorical_var, function(x) {
|
| 145 | ! |
if (length(x$filter) > 1L) {
|
| 146 | ! |
stop("tm_outliers: categorical_var data_extract_specs may only specify one filter_spec", call. = FALSE)
|
| 147 |
} |
|
| 148 |
}) |
|
| 149 |
} |
|
| 150 | ||
| 151 | ! |
ggtheme <- match.arg(ggtheme) |
| 152 | ||
| 153 | ! |
plot_choices <- c("Boxplot", "Density Plot", "Cumulative Distribution Plot")
|
| 154 | ! |
checkmate::assert_list(ggplot2_args, types = "ggplot2_args") |
| 155 | ! |
checkmate::assert_subset(names(ggplot2_args), c("default", plot_choices))
|
| 156 | ||
| 157 | ! |
checkmate::assert_numeric(plot_height, len = 3, any.missing = FALSE, finite = TRUE) |
| 158 | ! |
checkmate::assert_numeric(plot_height[1], lower = plot_height[2], upper = plot_height[3], .var.name = "plot_height") |
| 159 | ! |
checkmate::assert_numeric(plot_width, len = 3, any.missing = FALSE, null.ok = TRUE, finite = TRUE) |
| 160 | ! |
checkmate::assert_numeric( |
| 161 | ! |
plot_width[1], |
| 162 | ! |
lower = plot_width[2], upper = plot_width[3], null.ok = TRUE, .var.name = "plot_width" |
| 163 |
) |
|
| 164 | ||
| 165 | ! |
checkmate::assert_multi_class(pre_output, c("shiny.tag", "shiny.tag.list", "html"), null.ok = TRUE)
|
| 166 | ! |
checkmate::assert_multi_class(post_output, c("shiny.tag", "shiny.tag.list", "html"), null.ok = TRUE)
|
| 167 |
# End of assertions |
|
| 168 | ||
| 169 |
# Make UI args |
|
| 170 | ! |
args <- as.list(environment()) |
| 171 | ||
| 172 | ! |
data_extract_list <- list( |
| 173 | ! |
outlier_var = outlier_var, |
| 174 | ! |
categorical_var = categorical_var |
| 175 |
) |
|
| 176 | ||
| 177 | ! |
ans <- module( |
| 178 | ! |
label = label, |
| 179 | ! |
server = srv_outliers, |
| 180 | ! |
server_args = c( |
| 181 | ! |
data_extract_list, |
| 182 | ! |
list(plot_height = plot_height, plot_width = plot_width, ggplot2_args = ggplot2_args) |
| 183 |
), |
|
| 184 | ! |
ui = ui_outliers, |
| 185 | ! |
ui_args = args, |
| 186 | ! |
datanames = teal.transform::get_extract_datanames(data_extract_list) |
| 187 |
) |
|
| 188 | ! |
attr(ans, "teal_bookmarkable") <- TRUE |
| 189 | ! |
ans |
| 190 |
} |
|
| 191 | ||
| 192 |
# UI function for the outliers module |
|
| 193 |
ui_outliers <- function(id, ...) {
|
|
| 194 | ! |
args <- list(...) |
| 195 | ! |
ns <- NS(id) |
| 196 | ! |
is_single_dataset_value <- teal.transform::is_single_dataset(args$outlier_var, args$categorical_var) |
| 197 | ||
| 198 | ! |
teal.widgets::standard_layout( |
| 199 | ! |
output = teal.widgets::white_small_well( |
| 200 | ! |
uiOutput(ns("total_outliers")),
|
| 201 | ! |
DT::dataTableOutput(ns("summary_table")),
|
| 202 | ! |
uiOutput(ns("total_missing")),
|
| 203 | ! |
tags$br(), tags$hr(), |
| 204 | ! |
tabsetPanel( |
| 205 | ! |
id = ns("tabs"),
|
| 206 | ! |
tabPanel( |
| 207 | ! |
"Boxplot", |
| 208 | ! |
teal.widgets::plot_with_settings_ui(id = ns("box_plot"))
|
| 209 |
), |
|
| 210 | ! |
tabPanel( |
| 211 | ! |
"Density Plot", |
| 212 | ! |
teal.widgets::plot_with_settings_ui(id = ns("density_plot"))
|
| 213 |
), |
|
| 214 | ! |
tabPanel( |
| 215 | ! |
"Cumulative Distribution Plot", |
| 216 | ! |
teal.widgets::plot_with_settings_ui(id = ns("cum_density_plot"))
|
| 217 |
) |
|
| 218 |
), |
|
| 219 | ! |
tags$br(), tags$hr(), |
| 220 | ! |
uiOutput(ns("table_ui_wrap")),
|
| 221 | ! |
DT::dataTableOutput(ns("table_ui"))
|
| 222 |
), |
|
| 223 | ! |
encoding = tags$div( |
| 224 |
### Reporter |
|
| 225 | ! |
teal.reporter::simple_reporter_ui(ns("simple_reporter")),
|
| 226 |
### |
|
| 227 | ! |
tags$label("Encodings", class = "text-primary"),
|
| 228 | ! |
teal.transform::datanames_input(args[c("outlier_var", "categorical_var")]),
|
| 229 | ! |
teal.transform::data_extract_ui( |
| 230 | ! |
id = ns("outlier_var"),
|
| 231 | ! |
label = "Variable", |
| 232 | ! |
data_extract_spec = args$outlier_var, |
| 233 | ! |
is_single_dataset = is_single_dataset_value |
| 234 |
), |
|
| 235 | ! |
if (!is.null(args$categorical_var)) {
|
| 236 | ! |
teal.transform::data_extract_ui( |
| 237 | ! |
id = ns("categorical_var"),
|
| 238 | ! |
label = "Categorical factor", |
| 239 | ! |
data_extract_spec = args$categorical_var, |
| 240 | ! |
is_single_dataset = is_single_dataset_value |
| 241 |
) |
|
| 242 |
}, |
|
| 243 | ! |
conditionalPanel( |
| 244 | ! |
condition = paste0("input['", ns("tabs"), "'] == 'Boxplot'"),
|
| 245 | ! |
teal.widgets::optionalSelectInput( |
| 246 | ! |
inputId = ns("boxplot_alts"),
|
| 247 | ! |
label = "Plot type", |
| 248 | ! |
choices = c("Box plot", "Violin plot"),
|
| 249 | ! |
selected = "Box plot", |
| 250 | ! |
multiple = FALSE |
| 251 |
) |
|
| 252 |
), |
|
| 253 | ! |
shinyjs::hidden(checkboxInput(ns("split_outliers"), "Define outliers based on group splitting", value = FALSE)),
|
| 254 | ! |
shinyjs::hidden(checkboxInput(ns("order_by_outlier"), "Re-order categories by outliers [by %]", value = FALSE)),
|
| 255 | ! |
teal.widgets::panel_group( |
| 256 | ! |
teal.widgets::panel_item( |
| 257 | ! |
title = "Method parameters", |
| 258 | ! |
collapsed = FALSE, |
| 259 | ! |
teal.widgets::optionalSelectInput( |
| 260 | ! |
inputId = ns("method"),
|
| 261 | ! |
label = "Method", |
| 262 | ! |
choices = c("IQR", "Z-score", "Percentile"),
|
| 263 | ! |
selected = "IQR", |
| 264 | ! |
multiple = FALSE |
| 265 |
), |
|
| 266 | ! |
conditionalPanel( |
| 267 | ! |
condition = |
| 268 | ! |
paste0("input['", ns("method"), "'] == 'IQR'"),
|
| 269 | ! |
sliderInput( |
| 270 | ! |
ns("iqr_slider"),
|
| 271 | ! |
"Outlier range:", |
| 272 | ! |
min = 1, |
| 273 | ! |
max = 5, |
| 274 | ! |
value = 3, |
| 275 | ! |
step = 0.5 |
| 276 |
) |
|
| 277 |
), |
|
| 278 | ! |
conditionalPanel( |
| 279 | ! |
condition = |
| 280 | ! |
paste0("input['", ns("method"), "'] == 'Z-score'"),
|
| 281 | ! |
sliderInput( |
| 282 | ! |
ns("zscore_slider"),
|
| 283 | ! |
"Outlier range:", |
| 284 | ! |
min = 1, |
| 285 | ! |
max = 5, |
| 286 | ! |
value = 3, |
| 287 | ! |
step = 0.5 |
| 288 |
) |
|
| 289 |
), |
|
| 290 | ! |
conditionalPanel( |
| 291 | ! |
condition = |
| 292 | ! |
paste0("input['", ns("method"), "'] == 'Percentile'"),
|
| 293 | ! |
sliderInput( |
| 294 | ! |
ns("percentile_slider"),
|
| 295 | ! |
"Outlier range:", |
| 296 | ! |
min = 0.001, |
| 297 | ! |
max = 0.5, |
| 298 | ! |
value = 0.01, |
| 299 | ! |
step = 0.001 |
| 300 |
) |
|
| 301 |
), |
|
| 302 | ! |
uiOutput(ns("ui_outlier_help"))
|
| 303 |
) |
|
| 304 |
), |
|
| 305 | ! |
teal.widgets::panel_item( |
| 306 | ! |
title = "Plot settings", |
| 307 | ! |
selectInput( |
| 308 | ! |
inputId = ns("ggtheme"),
|
| 309 | ! |
label = "Theme (by ggplot):", |
| 310 | ! |
choices = ggplot_themes, |
| 311 | ! |
selected = args$ggtheme, |
| 312 | ! |
multiple = FALSE |
| 313 |
) |
|
| 314 |
) |
|
| 315 |
), |
|
| 316 | ! |
forms = tagList( |
| 317 | ! |
teal.widgets::verbatim_popup_ui(ns("rcode"), "Show R code")
|
| 318 |
), |
|
| 319 | ! |
pre_output = args$pre_output, |
| 320 | ! |
post_output = args$post_output |
| 321 |
) |
|
| 322 |
} |
|
| 323 | ||
| 324 |
# Server function for the outliers module |
|
| 325 |
srv_outliers <- function(id, data, reporter, filter_panel_api, outlier_var, |
|
| 326 |
categorical_var, plot_height, plot_width, ggplot2_args) {
|
|
| 327 | ! |
with_reporter <- !missing(reporter) && inherits(reporter, "Reporter") |
| 328 | ! |
with_filter <- !missing(filter_panel_api) && inherits(filter_panel_api, "FilterPanelAPI") |
| 329 | ! |
checkmate::assert_class(data, "reactive") |
| 330 | ! |
checkmate::assert_class(isolate(data()), "teal_data") |
| 331 | ! |
moduleServer(id, function(input, output, session) {
|
| 332 | ! |
teal.logger::log_shiny_input_changes(input, namespace = "teal.modules.general") |
| 333 | ||
| 334 | ! |
ns <- session$ns |
| 335 | ||
| 336 | ! |
vars <- list(outlier_var = outlier_var, categorical_var = categorical_var) |
| 337 | ||
| 338 | ! |
rule_diff <- function(other) {
|
| 339 | ! |
function(value) {
|
| 340 | ! |
othervalue <- tryCatch(selector_list()[[other]]()[["select"]], error = function(e) NULL) |
| 341 | ! |
if (!is.null(othervalue) && identical(othervalue, value)) {
|
| 342 | ! |
"`Variable` and `Categorical factor` cannot be the same" |
| 343 |
} |
|
| 344 |
} |
|
| 345 |
} |
|
| 346 | ||
| 347 | ! |
selector_list <- teal.transform::data_extract_multiple_srv( |
| 348 | ! |
data_extract = vars, |
| 349 | ! |
datasets = data, |
| 350 | ! |
select_validation_rule = list( |
| 351 | ! |
outlier_var = shinyvalidate::compose_rules( |
| 352 | ! |
shinyvalidate::sv_required("Please select a variable"),
|
| 353 | ! |
rule_diff("categorical_var")
|
| 354 |
), |
|
| 355 | ! |
categorical_var = rule_diff("outlier_var")
|
| 356 |
) |
|
| 357 |
) |
|
| 358 | ||
| 359 | ! |
iv_r <- reactive({
|
| 360 | ! |
iv <- shinyvalidate::InputValidator$new() |
| 361 | ! |
iv$add_rule("method", shinyvalidate::sv_required("Please select a method"))
|
| 362 | ! |
iv$add_rule("boxplot_alts", shinyvalidate::sv_required("Please select Plot Type"))
|
| 363 | ! |
teal.transform::compose_and_enable_validators(iv, selector_list) |
| 364 |
}) |
|
| 365 | ||
| 366 | ! |
reactive_select_input <- reactive({
|
| 367 | ! |
if (is.null(selector_list()$categorical_var) || length(selector_list()$categorical_var()$select) == 0) {
|
| 368 | ! |
selector_list()[names(selector_list()) != "categorical_var"] |
| 369 |
} else {
|
|
| 370 | ! |
selector_list() |
| 371 |
} |
|
| 372 |
}) |
|
| 373 | ||
| 374 | ! |
anl_merged_input <- teal.transform::merge_expression_srv( |
| 375 | ! |
selector_list = reactive_select_input, |
| 376 | ! |
datasets = data, |
| 377 | ! |
merge_function = "dplyr::inner_join" |
| 378 |
) |
|
| 379 | ||
| 380 | ! |
anl_merged_q <- reactive({
|
| 381 | ! |
req(anl_merged_input()) |
| 382 | ! |
data() %>% |
| 383 | ! |
teal.code::eval_code(as.expression(anl_merged_input()$expr)) |
| 384 |
}) |
|
| 385 | ||
| 386 | ! |
merged <- list( |
| 387 | ! |
anl_input_r = anl_merged_input, |
| 388 | ! |
anl_q_r = anl_merged_q |
| 389 |
) |
|
| 390 | ||
| 391 | ! |
n_outlier_missing <- reactive({
|
| 392 | ! |
req(iv_r()$is_valid()) |
| 393 | ! |
outlier_var <- as.vector(merged$anl_input_r()$columns_source$outlier_var) |
| 394 | ! |
ANL <- merged$anl_q_r()[["ANL"]] |
| 395 | ! |
sum(is.na(ANL[[outlier_var]])) |
| 396 |
}) |
|
| 397 | ||
| 398 |
# Used to create outlier table and the dropdown with additional columns |
|
| 399 | ! |
dataname_first <- isolate(teal.data::datanames(data())[[1]]) |
| 400 | ||
| 401 | ! |
common_code_q <- reactive({
|
| 402 | ! |
req(iv_r()$is_valid()) |
| 403 | ||
| 404 | ! |
ANL <- merged$anl_q_r()[["ANL"]] |
| 405 | ! |
qenv <- merged$anl_q_r() |
| 406 | ||
| 407 | ! |
outlier_var <- as.vector(merged$anl_input_r()$columns_source$outlier_var) |
| 408 | ! |
categorical_var <- as.vector(merged$anl_input_r()$columns_source$categorical_var) |
| 409 | ! |
order_by_outlier <- input$order_by_outlier |
| 410 | ! |
method <- input$method |
| 411 | ! |
split_outliers <- input$split_outliers |
| 412 | ! |
teal::validate_has_data( |
| 413 |
# missing values in the categorical variable may be used to form a category of its own |
|
| 414 | ! |
`if`( |
| 415 | ! |
length(categorical_var) == 0, |
| 416 | ! |
ANL, |
| 417 | ! |
ANL[, names(ANL) != categorical_var, drop = FALSE] |
| 418 |
), |
|
| 419 | ! |
min_nrow = 10, |
| 420 | ! |
complete = TRUE, |
| 421 | ! |
allow_inf = FALSE |
| 422 |
) |
|
| 423 | ! |
validate(need(is.numeric(ANL[[outlier_var]]), "`Variable` is not numeric")) |
| 424 | ! |
validate(need(length(unique(ANL[[outlier_var]])) > 1, "Variable has no variation, i.e. only one unique value")) |
| 425 | ||
| 426 |
# show/hide split_outliers |
|
| 427 | ! |
if (length(categorical_var) == 0) {
|
| 428 | ! |
shinyjs::hide("split_outliers")
|
| 429 | ! |
if (n_outlier_missing() > 0) {
|
| 430 | ! |
qenv <- teal.code::eval_code( |
| 431 | ! |
qenv, |
| 432 | ! |
substitute( |
| 433 | ! |
expr = ANL <- ANL %>% dplyr::filter(!is.na(outlier_var_name)), |
| 434 | ! |
env = list(outlier_var_name = as.name(outlier_var)) |
| 435 |
) |
|
| 436 |
) |
|
| 437 |
} |
|
| 438 |
} else {
|
|
| 439 | ! |
validate(need( |
| 440 | ! |
is.factor(ANL[[categorical_var]]) || |
| 441 | ! |
is.character(ANL[[categorical_var]]) || |
| 442 | ! |
is.integer(ANL[[categorical_var]]), |
| 443 | ! |
"`Categorical factor` must be `factor`, `character`, or `integer`" |
| 444 |
)) |
|
| 445 | ||
| 446 | ! |
if (n_outlier_missing() > 0) {
|
| 447 | ! |
qenv <- teal.code::eval_code( |
| 448 | ! |
qenv, |
| 449 | ! |
substitute( |
| 450 | ! |
expr = ANL <- ANL %>% dplyr::filter(!is.na(outlier_var_name)), |
| 451 | ! |
env = list(outlier_var_name = as.name(outlier_var)) |
| 452 |
) |
|
| 453 |
) |
|
| 454 |
} |
|
| 455 | ! |
shinyjs::show("split_outliers")
|
| 456 |
} |
|
| 457 | ||
| 458 |
# slider |
|
| 459 | ! |
outlier_definition_param <- if (method == "IQR") {
|
| 460 | ! |
input$iqr_slider |
| 461 | ! |
} else if (method == "Z-score") {
|
| 462 | ! |
input$zscore_slider |
| 463 | ! |
} else if (method == "Percentile") {
|
| 464 | ! |
input$percentile_slider |
| 465 |
} |
|
| 466 | ||
| 467 |
# this is utils function that converts a %>% NULL %>% b into a %>% b |
|
| 468 | ! |
remove_pipe_null <- function(x) {
|
| 469 | ! |
if (length(x) == 1) {
|
| 470 | ! |
return(x) |
| 471 |
} |
|
| 472 | ! |
if (identical(x[[1]], as.name("%>%")) && is.null(x[[3]])) {
|
| 473 | ! |
return(remove_pipe_null(x[[2]])) |
| 474 |
} |
|
| 475 | ! |
return(as.call(c(x[[1]], lapply(x[-1], remove_pipe_null)))) |
| 476 |
} |
|
| 477 | ||
| 478 | ! |
qenv <- teal.code::eval_code( |
| 479 | ! |
qenv, |
| 480 | ! |
substitute( |
| 481 | ! |
expr = {
|
| 482 | ! |
ANL_OUTLIER <- ANL %>% |
| 483 | ! |
group_expr %>% # styler: off |
| 484 | ! |
dplyr::mutate(is_outlier = {
|
| 485 | ! |
q1_q3 <- stats::quantile(outlier_var_name, probs = c(0.25, 0.75)) |
| 486 | ! |
iqr <- q1_q3[2] - q1_q3[1] |
| 487 | ! |
!(outlier_var_name >= q1_q3[1] - 1.5 * iqr & outlier_var_name <= q1_q3[2] + 1.5 * iqr) |
| 488 |
}) %>% |
|
| 489 | ! |
calculate_outliers %>% # styler: off |
| 490 | ! |
ungroup_expr %>% # styler: off |
| 491 | ! |
dplyr::filter(is_outlier | is_outlier_selected) %>% |
| 492 | ! |
dplyr::select(-is_outlier) |
| 493 |
}, |
|
| 494 | ! |
env = list( |
| 495 | ! |
calculate_outliers = if (method == "IQR") {
|
| 496 | ! |
substitute( |
| 497 | ! |
expr = dplyr::mutate(is_outlier_selected = {
|
| 498 | ! |
q1_q3 <- stats::quantile(outlier_var_name, probs = c(0.25, 0.75)) |
| 499 | ! |
iqr <- q1_q3[2] - q1_q3[1] |
| 500 |
!( |
|
| 501 | ! |
outlier_var_name >= q1_q3[1] - outlier_definition_param * iqr & |
| 502 | ! |
outlier_var_name <= q1_q3[2] + outlier_definition_param * iqr |
| 503 |
) |
|
| 504 |
}), |
|
| 505 | ! |
env = list( |
| 506 | ! |
outlier_var_name = as.name(outlier_var), |
| 507 | ! |
outlier_definition_param = outlier_definition_param |
| 508 |
) |
|
| 509 |
) |
|
| 510 | ! |
} else if (method == "Z-score") {
|
| 511 | ! |
substitute( |
| 512 | ! |
expr = dplyr::mutate( |
| 513 | ! |
is_outlier_selected = abs(outlier_var_name - mean(outlier_var_name)) / |
| 514 | ! |
stats::sd(outlier_var_name) > outlier_definition_param |
| 515 |
), |
|
| 516 | ! |
env = list( |
| 517 | ! |
outlier_var_name = as.name(outlier_var), |
| 518 | ! |
outlier_definition_param = outlier_definition_param |
| 519 |
) |
|
| 520 |
) |
|
| 521 | ! |
} else if (method == "Percentile") {
|
| 522 | ! |
substitute( |
| 523 | ! |
expr = dplyr::mutate( |
| 524 | ! |
is_outlier_selected = outlier_var_name < stats::quantile(outlier_var_name, outlier_definition_param) | |
| 525 | ! |
outlier_var_name > stats::quantile(outlier_var_name, 1 - outlier_definition_param) |
| 526 |
), |
|
| 527 | ! |
env = list( |
| 528 | ! |
outlier_var_name = as.name(outlier_var), |
| 529 | ! |
outlier_definition_param = outlier_definition_param |
| 530 |
) |
|
| 531 |
) |
|
| 532 |
}, |
|
| 533 | ! |
outlier_var_name = as.name(outlier_var), |
| 534 | ! |
group_expr = if (isTRUE(split_outliers) && length(categorical_var) != 0) {
|
| 535 | ! |
substitute(dplyr::group_by(x), list(x = as.name(categorical_var))) |
| 536 |
}, |
|
| 537 | ! |
ungroup_expr = if (isTRUE(split_outliers) && length(categorical_var) != 0) {
|
| 538 | ! |
substitute(dplyr::ungroup()) |
| 539 |
} |
|
| 540 |
) |
|
| 541 |
) %>% |
|
| 542 | ! |
remove_pipe_null() |
| 543 |
) |
|
| 544 | ||
| 545 |
# ANL_OUTLIER_EXTENDED is the base table |
|
| 546 | ! |
qenv <- teal.code::eval_code( |
| 547 | ! |
qenv, |
| 548 | ! |
substitute( |
| 549 | ! |
expr = {
|
| 550 | ! |
ANL_OUTLIER_EXTENDED <- dplyr::left_join( |
| 551 | ! |
ANL_OUTLIER, |
| 552 | ! |
dplyr::select( |
| 553 | ! |
dataname, |
| 554 | ! |
dplyr::setdiff(names(dataname), dplyr::setdiff(names(ANL_OUTLIER), join_keys)) |
| 555 |
), |
|
| 556 | ! |
by = join_keys |
| 557 |
) |
|
| 558 |
}, |
|
| 559 | ! |
env = list( |
| 560 | ! |
dataname = as.name(dataname_first), |
| 561 | ! |
join_keys = as.character(teal.data::join_keys(data())[dataname_first, dataname_first]) |
| 562 |
) |
|
| 563 |
) |
|
| 564 |
) |
|
| 565 | ||
| 566 | ! |
if (length(categorical_var) > 0) {
|
| 567 | ! |
qenv <- teal.code::eval_code( |
| 568 | ! |
qenv, |
| 569 | ! |
substitute( |
| 570 | ! |
expr = summary_table_pre <- ANL_OUTLIER %>% |
| 571 | ! |
dplyr::filter(is_outlier_selected) %>% |
| 572 | ! |
dplyr::select(outlier_var_name, categorical_var_name) %>% |
| 573 | ! |
dplyr::group_by(categorical_var_name) %>% |
| 574 | ! |
dplyr::summarise(n_outliers = dplyr::n()) %>% |
| 575 | ! |
dplyr::right_join( |
| 576 | ! |
ANL %>% |
| 577 | ! |
dplyr::select(outlier_var_name, categorical_var_name) %>% |
| 578 | ! |
dplyr::group_by(categorical_var_name) %>% |
| 579 | ! |
dplyr::summarise( |
| 580 | ! |
total_in_cat = dplyr::n(), |
| 581 | ! |
n_na = sum(is.na(outlier_var_name) | is.na(categorical_var_name)) |
| 582 |
), |
|
| 583 | ! |
by = categorical_var |
| 584 |
) %>% |
|
| 585 |
# This is important as there may be categorical variables with natural orderings, e.g. AGE. |
|
| 586 |
# The plots should be displayed by default in increasing order in these situations. |
|
| 587 |
# dplyr::arrange will sort integer, factor, and character data types in the expected way. |
|
| 588 | ! |
dplyr::arrange(categorical_var_name) %>% |
| 589 | ! |
dplyr::mutate( |
| 590 | ! |
n_outliers = dplyr::if_else(is.na(n_outliers), 0, as.numeric(n_outliers)), |
| 591 | ! |
display_str = dplyr::if_else( |
| 592 | ! |
n_outliers > 0, |
| 593 | ! |
sprintf("%d [%.02f%%]", n_outliers, 100 * n_outliers / total_in_cat),
|
| 594 | ! |
"0" |
| 595 |
), |
|
| 596 | ! |
display_str_na = dplyr::if_else( |
| 597 | ! |
n_na > 0, |
| 598 | ! |
sprintf("%d [%.02f%%]", n_na, 100 * n_na / total_in_cat),
|
| 599 | ! |
"0" |
| 600 |
), |
|
| 601 | ! |
order = seq_along(n_outliers) |
| 602 |
), |
|
| 603 | ! |
env = list( |
| 604 | ! |
categorical_var = categorical_var, |
| 605 | ! |
categorical_var_name = as.name(categorical_var), |
| 606 | ! |
outlier_var_name = as.name(outlier_var) |
| 607 |
) |
|
| 608 |
) |
|
| 609 |
) |
|
| 610 |
# now to handle when user chooses to order based on amount of outliers |
|
| 611 | ! |
if (order_by_outlier) {
|
| 612 | ! |
qenv <- teal.code::eval_code( |
| 613 | ! |
qenv, |
| 614 | ! |
quote( |
| 615 | ! |
summary_table_pre <- summary_table_pre %>% |
| 616 | ! |
dplyr::arrange(desc(n_outliers / total_in_cat)) %>% |
| 617 | ! |
dplyr::mutate(order = seq_len(nrow(summary_table_pre))) |
| 618 |
) |
|
| 619 |
) |
|
| 620 |
} |
|
| 621 | ||
| 622 | ! |
qenv <- teal.code::eval_code( |
| 623 | ! |
qenv, |
| 624 | ! |
substitute( |
| 625 | ! |
expr = {
|
| 626 |
# In order for geom_rug to work properly when reordering takes place inside facet_grid, |
|
| 627 |
# all tables must have the column used for reording. |
|
| 628 |
# In this case, the column used for reordering is `order`. |
|
| 629 | ! |
ANL_OUTLIER <- dplyr::left_join( |
| 630 | ! |
ANL_OUTLIER, |
| 631 | ! |
summary_table_pre[, c("order", categorical_var)],
|
| 632 | ! |
by = categorical_var |
| 633 |
) |
|
| 634 |
# so that x axis of plot aligns with columns of summary table, from most outliers to least by percentage |
|
| 635 | ! |
ANL <- ANL %>% |
| 636 | ! |
dplyr::left_join( |
| 637 | ! |
dplyr::select(summary_table_pre, categorical_var_name, order), |
| 638 | ! |
by = categorical_var |
| 639 |
) %>% |
|
| 640 | ! |
dplyr::arrange(order) |
| 641 | ! |
summary_table <- summary_table_pre %>% |
| 642 | ! |
dplyr::select( |
| 643 | ! |
categorical_var_name, |
| 644 | ! |
Outliers = display_str, Missings = display_str_na, Total = total_in_cat |
| 645 |
) %>% |
|
| 646 | ! |
dplyr::mutate_all(as.character) %>% |
| 647 | ! |
tidyr::pivot_longer(-categorical_var_name) %>% |
| 648 | ! |
tidyr::pivot_wider(names_from = categorical_var, values_from = value) %>% |
| 649 | ! |
tibble::column_to_rownames("name")
|
| 650 | ! |
summary_table |
| 651 |
}, |
|
| 652 | ! |
env = list( |
| 653 | ! |
categorical_var = categorical_var, |
| 654 | ! |
categorical_var_name = as.name(categorical_var) |
| 655 |
) |
|
| 656 |
) |
|
| 657 |
) |
|
| 658 |
} |
|
| 659 | ||
| 660 | ! |
if (length(categorical_var) > 0 && nrow(qenv[["ANL_OUTLIER"]]) > 0) {
|
| 661 | ! |
shinyjs::show("order_by_outlier")
|
| 662 |
} else {
|
|
| 663 | ! |
shinyjs::hide("order_by_outlier")
|
| 664 |
} |
|
| 665 | ||
| 666 | ! |
qenv |
| 667 |
}) |
|
| 668 | ||
| 669 | ! |
output$summary_table <- DT::renderDataTable( |
| 670 | ! |
expr = {
|
| 671 | ! |
if (iv_r()$is_valid()) {
|
| 672 | ! |
categorical_var <- as.vector(merged$anl_input_r()$columns_source$categorical_var) |
| 673 | ! |
if (!is.null(categorical_var)) {
|
| 674 | ! |
DT::datatable( |
| 675 | ! |
common_code_q()[["summary_table"]], |
| 676 | ! |
options = list( |
| 677 | ! |
dom = "t", |
| 678 | ! |
autoWidth = TRUE, |
| 679 | ! |
columnDefs = list(list(width = "200px", targets = "_all")) |
| 680 |
) |
|
| 681 |
) |
|
| 682 |
} |
|
| 683 |
} |
|
| 684 |
} |
|
| 685 |
) |
|
| 686 | ||
| 687 |
# boxplot/violinplot # nolint commented_code |
|
| 688 | ! |
boxplot_q <- reactive({
|
| 689 | ! |
req(common_code_q()) |
| 690 | ! |
ANL <- common_code_q()[["ANL"]] |
| 691 | ! |
ANL_OUTLIER <- common_code_q()[["ANL_OUTLIER"]] |
| 692 | ||
| 693 | ! |
outlier_var <- as.vector(merged$anl_input_r()$columns_source$outlier_var) |
| 694 | ! |
categorical_var <- as.vector(merged$anl_input_r()$columns_source$categorical_var) |
| 695 | ||
| 696 |
# validation |
|
| 697 | ! |
teal::validate_has_data(ANL, 1) |
| 698 | ||
| 699 |
# boxplot |
|
| 700 | ! |
plot_call <- quote(ANL %>% ggplot()) |
| 701 | ||
| 702 | ! |
plot_call <- if (input$boxplot_alts == "Box plot") {
|
| 703 | ! |
substitute(expr = plot_call + geom_boxplot(outlier.shape = NA), env = list(plot_call = plot_call)) |
| 704 | ! |
} else if (input$boxplot_alts == "Violin plot") {
|
| 705 | ! |
substitute(expr = plot_call + geom_violin(), env = list(plot_call = plot_call)) |
| 706 |
} else {
|
|
| 707 | ! |
NULL |
| 708 |
} |
|
| 709 | ||
| 710 | ! |
plot_call <- if (identical(categorical_var, character(0)) || is.null(categorical_var)) {
|
| 711 | ! |
inner_call <- substitute( |
| 712 | ! |
expr = plot_call + |
| 713 | ! |
aes(x = "Entire dataset", y = outlier_var_name) + |
| 714 | ! |
scale_x_discrete(), |
| 715 | ! |
env = list(plot_call = plot_call, outlier_var_name = as.name(outlier_var)) |
| 716 |
) |
|
| 717 | ! |
if (nrow(ANL_OUTLIER) > 0) {
|
| 718 | ! |
substitute( |
| 719 | ! |
expr = inner_call + geom_point( |
| 720 | ! |
data = ANL_OUTLIER, |
| 721 | ! |
aes(x = "Entire dataset", y = outlier_var_name, color = is_outlier_selected) |
| 722 |
), |
|
| 723 | ! |
env = list(inner_call = inner_call, outlier_var_name = as.name(outlier_var)) |
| 724 |
) |
|
| 725 |
} else {
|
|
| 726 | ! |
inner_call |
| 727 |
} |
|
| 728 |
} else {
|
|
| 729 | ! |
substitute( |
| 730 | ! |
expr = plot_call + |
| 731 | ! |
aes(y = outlier_var_name, x = reorder(categorical_var_name, order)) + |
| 732 | ! |
xlab(categorical_var) + |
| 733 | ! |
scale_x_discrete() + |
| 734 | ! |
geom_point( |
| 735 | ! |
data = ANL_OUTLIER, |
| 736 | ! |
aes(x = as.factor(categorical_var_name), y = outlier_var_name, color = is_outlier_selected) |
| 737 |