Style package with new {mlr.styler} package

R/BenchmarkResult.R

@@ -270,7 +270,7 @@ BenchmarkResult = R6Class("BenchmarkResult",
       tab = insert_named(tab, scores)
 
       cns = c("uhash", "nr", "resample_result", "task_id", "learner_id", "resampling_id", "iters",
-          "warnings", "errors", "params", ids(measures))
+        "warnings", "errors", "params", ids(measures))
       cns = intersect(cns, names(tab))
       tab[, cns, with = FALSE]
     },
@@ -302,10 +302,11 @@ BenchmarkResult = R6Class("BenchmarkResult",
       learner_phashes = NULL
 
       filter_if_not_null = function(column, hashes) {
-        if (is.null(hashes))
+        if (is.null(hashes)) {
           fact
-        else
+        } else {
           fact[unique(hashes), on = column, nomatch = NULL]
+        }
       }
 
 

R/Learner.R

@@ -244,8 +244,9 @@ Learner = R6Class("Learner",
       }
 
       pdata = learner_predict(self, task, row_ids)
-      if (is.null(pdata))
+      if (is.null(pdata)) {
         return(NULL)
+      }
 
       as_prediction(check_prediction_data(pdata))
     },
@@ -269,8 +270,9 @@ Learner = R6Class("Learner",
       newdata = as.data.table(assert_data_frame(newdata, min.rows = 1L))
 
       if (is.null(task)) {
-        if (is.null(self$state$train_task))
+        if (is.null(self$state$train_task)) {
           stopf("No task stored, and no task provided")
+        }
         task = self$state$train_task$clone()
       } else {
         task = assert_task(as_task(task, clone = TRUE))
@@ -414,7 +416,10 @@ Learner = R6Class("Learner",
       switch(name,
         .param_set = value$clone(deep = TRUE),
         fallback = if (is.null(value)) NULL else value$clone(deep = TRUE),
-        state = { value$log = copy(value$log); value },
+        state = {
+          value$log = copy(value$log)
+          value
+        },
         value
       )
     }

R/LearnerClassifDebug.R

@@ -152,8 +152,13 @@ LearnerClassifDebug = R6Class("LearnerClassifDebug", inherit = LearnerClassif,
         if (!is.null(pv$predict_missing)) {
           ii = sample.int(n, n * pv$predict_missing)
           prob = switch(missing_type,
-            "na" = { prob[ii, ] = NA_real_; prob },
-            "omit" = { prob[ii,, drop = FALSE] }
+            "na" = {
+              prob[ii, ] = NA_real_
+              prob
+            },
+            "omit" = {
+              prob[ii, , drop = FALSE]
+            }
           )
         }
       }

R/MeasureDebug.R

@@ -51,8 +51,9 @@ MeasureDebug = R6Class("MeasureDebug",
   private = list(
     .score = function(prediction, ...) {
       na_ratio = self$param_set$get_values()$na_ratio
-      if (na_ratio > runif(1L))
+      if (na_ratio > runif(1L)) {
         return(NA_integer_)
+      }
       length(prediction$row_ids)
     }
   )

R/PredictionDataClassif.R

@@ -90,7 +90,7 @@ c.PredictionDataClassif = function(..., keep_duplicates = TRUE) {
   if (!keep_duplicates) {
     keep = !duplicated(tab, by = "row_ids", fromLast = TRUE)
     tab = tab[keep]
-    prob = prob[keep,, drop = FALSE]
+    prob = prob[keep, , drop = FALSE]
   }
 
   result = as.list(tab)

R/ResampleResult.R

@@ -138,7 +138,7 @@ ResampleResult = R6Class("ResampleResult",
         set(tab, j = "learner_id", value = ids(tab[["learner"]]))
         set(tab, j = "resampling_id", value = ids(tab[["resampling"]]))
         setcolorder(tab, c("task", "task_id", "learner", "learner_id", "resampling", "resampling_id",
-            "iteration", "prediction"))
+          "iteration", "prediction"))
       }
 
       if (conditions) {
@@ -213,8 +213,9 @@ ResampleResult = R6Class("ResampleResult",
     task = function(rhs) {
       assert_ro_binding(rhs)
       tab = private$.data$tasks(private$.view)
-      if (nrow(tab) == 0L)
+      if (nrow(tab) == 0L) {
         return(NULL)
+      }
       tab$task[[1L]]
     },
 
@@ -224,8 +225,9 @@ ResampleResult = R6Class("ResampleResult",
     learner = function(rhs) {
       assert_ro_binding(rhs)
       tab = private$.data$learners(private$.view, states = FALSE)
-      if (nrow(tab) == 0L)
+      if (nrow(tab) == 0L) {
         return(NULL)
+      }
       tab$learner[[1L]]
     },
 
@@ -234,8 +236,9 @@ ResampleResult = R6Class("ResampleResult",
     resampling = function(rhs) {
       assert_ro_binding(rhs)
       tab = private$.data$resamplings(private$.view)
-      if (nrow(tab) == 0L)
+      if (nrow(tab) == 0L) {
         return(NULL)
+      }
       tab$resampling[[1L]]
     },
 

R/ResamplingHoldout.R

@@ -72,7 +72,8 @@ ResamplingHoldout = R6Class("ResamplingHoldout", inherit = Resampling,
 
     .combine = function(instances) {
       list(train = do.call(c, map(instances, "train")), test = do.call(c, map(instances, "test")))
-    })
+    }
+  )
 )
 
 #' @include mlr_resamplings.R

R/ResultData.R

@@ -295,18 +295,20 @@ ResultData = R6Class("ResultData",
     #'   Returns `NULL` if the [ResultData] is empty.
     task_type = function() {
       tab = self$data$tasks
-      if (nrow(tab))
+      if (nrow(tab)) {
         tab$task[[1L]]$task_type
-      else
+      } else {
         NULL
+      }
     }
   ),
 
   private = list(
     get_view_index = function(view) {
-      if (is.null(view))
+      if (is.null(view)) {
         return(TRUE)
-       self$data$fact[list(view), on = "uhash", nomatch = NULL, which = TRUE]
+      }
+      self$data$fact[list(view), on = "uhash", nomatch = NULL, which = TRUE]
     },
 
     deep_clone = function(name, value) {
@@ -382,8 +384,8 @@ reassemble_learners = function(learners, states = NULL, param_vals = NULL) {
   learners = lapply(learners, function(l) l$clone(deep = TRUE))
 
   if (!is.null(states)) {
-     Map(function(l, s) {
-       l$state = s
+    Map(function(l, s) {
+      l$state = s
     }, l = learners, s = states)
   }
 

R/Task.R

@@ -351,8 +351,8 @@ Task = R6Class("Task",
 
         ci = self$col_info[list(keep_cols), on = "id"]
         data = do.call(data.table, Map(auto_convert,
-            value = as.list(data)[ci$id],
-            id = ci$id, type = ci$type, levels = ci$levels))
+          value = as.list(data)[ci$id],
+          id = ci$id, type = ci$type, levels = ci$levels))
 
         data = as_data_backend(data, primary_key = pk)
       } else {

R/TaskClassif_penguins.R

@@ -35,7 +35,7 @@ load_task_penguins = function() {
   penguins = as.data.table(palmerpenguins::penguins)
   setnames(penguins,
     old = c("bill_length_mm", "bill_depth_mm", "flipper_length_mm", "body_mass_g"),
-    new = c("bill_length",    "bill_depth",    "flipper_length",    "body_mass")
+    new = c("bill_length", "bill_depth", "flipper_length", "body_mass")
   )
 
   b = as_data_backend(penguins)

R/TaskGeneratorMoons.R

@@ -58,7 +58,7 @@ TaskGeneratorMoons = R6Class("TaskGeneratorMoons",
 
       data.table(
         y = factor(rep(c("A", "B"), c(n1, n2)), levels = c("A", "B")),
-        x1 = 5  * cos(x) + rnorm(n, mean = mu, sd = sigma),
+        x1 = 5 * cos(x) + rnorm(n, mean = mu, sd = sigma),
         x2 = 10 * sin(x) + rnorm(n, mean = mu, sd = sigma)
       )
     },

R/TaskSupervised.R

@@ -19,7 +19,7 @@
 #' @export
 #' @examples
 #' TaskSupervised$new("penguins", task_type = "classif", backend = palmerpenguins::penguins,
-#'     target = "species")
+#'   target = "species")
 TaskSupervised = R6Class("TaskSupervised", inherit = Task,
   public = list(
 

R/as_prediction_data.R

@@ -35,8 +35,9 @@ as_prediction_data.list = function(x, task, row_ids = task$row_ids, check = TRUE
   assert_names(names(x), subset.of = predict_types)
 
   x$row_ids = row_ids
-  if (inherits(task, "TaskSupervised"))
+  if (inherits(task, "TaskSupervised")) {
     x$truth = task$truth(row_ids)
+  }
 
   pdata = new_prediction_data(x, task_type = task$task_type)
   if (check) {

R/assertions.R

@@ -175,8 +175,9 @@ assert_measure = function(measure, task = NULL, learner = NULL, .var.name = vnam
 #' @rdname mlr_assertions
 assert_measures = function(measures, task = NULL, learner = NULL, .var.name = vname(measures)) {
   lapply(measures, assert_measure, task = task, learner = learner, .var.name = .var.name)
-  if (anyDuplicated(ids(measures)))
+  if (anyDuplicated(ids(measures))) {
     stopf("Measures need to have unique IDs")
+  }
   invisible(measures)
 }
 
@@ -281,7 +282,7 @@ assert_prediction_count = function(actual, expected, type) {
 
 assert_row_sums = function(prob) {
   for (i in seq_row(prob)) {
-    x = prob[i,, drop = TRUE]
+    x = prob[i, , drop = TRUE]
     n_missing = count_missing(x)
     if (n_missing > 0L && n_missing < length(x)) {
       stopf("Probabilities for observation %i are partly missing", i)

R/auto_convert.R

@@ -5,21 +5,35 @@ mlr_reflections$auto_converters = ee = new.env(parent = emptyenv())
 
 ## from: logical
 ee[["logical___integer"]] =
-  function(value, type, levels) { as.integer(value) }
+  function(value, type, levels) {
+    as.integer(value)
+  }
 ee[["logical___numeric"]] =
-  function(value, type, levels) { as.double(value) }
+  function(value, type, levels) {
+    as.double(value)
+  }
 ee[["logical___character"]] =
-  function(value, type, levels) { as.character(value) }
+  function(value, type, levels) {
+    as.character(value)
+  }
 ee[["logical___factor"]] =
-  function(value, type, levels) { factor(value, levels = union(levels, value), ordered = FALSE) }
+  function(value, type, levels) {
+    factor(value, levels = union(levels, value), ordered = FALSE)
+  }
 ee[["logical___ordered"]] =
-  function(value, type, levels) { if (all(value %in% c(NA_character_, levels))) factor(value, levels = levels, ordered = TRUE) else NULL }
+  function(value, type, levels) {
+    if (all(value %in% c(NA_character_, levels))) factor(value, levels = levels, ordered = TRUE) else NULL
+  }
 ee[["logical___POSIXct"]] =
-  function(value, type, levels) { if (allMissing(value)) .POSIXct(value, tz = "") else value }
+  function(value, type, levels) {
+    if (allMissing(value)) .POSIXct(value, tz = "") else value
+  }
 
 ## from: integer
 ee[["integer___logical"]] =
-  function(value, type, levels) { if (test_integerish(value, lower = 0L, upper = 1L)) as.logical(value) else value }
+  function(value, type, levels) {
+    if (test_integerish(value, lower = 0L, upper = 1L)) as.logical(value) else value
+  }
 ee[["integer___numeric"]] =
   ee[["logical___numeric"]]
 ee[["integer___character"]] =
@@ -35,7 +49,9 @@ ee[["integer___POSIXct"]] =
 ee[["numeric___logical"]] =
   ee[["integer___logical"]]
 ee[["numeric___integer"]] =
-  function(value, type, levels) { if (test_integerish(value)) as.integer(value) else value }
+  function(value, type, levels) {
+    if (test_integerish(value)) as.integer(value) else value
+  }
 ee[["numeric___character"]] =
   ee[["logical___character"]]
 ee[["numeric___factor"]] =
@@ -47,27 +63,40 @@ ee[["numeric___POSIXct"]] =
 
 ## from: character
 ee[["character___logical"]] =
-  function(value, type, levels) { if (all(value %in% c(NA_character_, "TRUE", "FALSE", "true", "false", "T", "F"))) as.logical(value) else value }
+  function(value, type, levels) {
+    if (all(value %in% c(NA_character_, "TRUE", "FALSE", "true", "false", "T", "F"))) as.logical(value) else value
+  }
 ee[["character___integer"]] =
-  function(value, type, levels) { if (allMissing(value)) as.integer(value) else value }
+  function(value, type, levels) {
+    if (allMissing(value)) as.integer(value) else value
+  }
 ee[["character___numeric"]] =
-  function(value, type, levels) { if (allMissing(value)) as.double(value) else value }
+  function(value, type, levels) {
+    if (allMissing(value)) as.double(value) else value
+  }
 ee[["character___factor"]] =
   ee[["logical___factor"]]
 ee[["character___ordered"]] =
   ee[["logical___ordered"]]
 ee[["character___POSIXct"]] =
-  function(value, type, levels) { x = try(as.POSIXct(value, ""), silent = TRUE); if (inherits(x, "try-error")) value else x }
+  function(value, type, levels) {
+    x = try(as.POSIXct(value, ""), silent = TRUE)
+    if (inherits(x, "try-error")) value else x
+  }
 
 ## from: factor
 ee[["factor___logical"]] =
   ee[["character___logical"]]
 ee[["factor___character"]] =
   ee[["logical___character"]]
 ee[["factor___factor"]] =
-  function(value, type, levels) { factor(value, levels = union(levels, levels(value)), ordered = FALSE) }
+  function(value, type, levels) {
+    factor(value, levels = union(levels, levels(value)), ordered = FALSE)
+  }
 ee[["factor___ordered"]] =
-  function(value, type, levels) { if (all(levels(value) %in% levels)) factor(value, levels = levels, ordered = TRUE) else NULL }
+  function(value, type, levels) {
+    if (all(levels(value) %in% levels)) factor(value, levels = levels, ordered = TRUE) else NULL
+  }
 ee[["factor___POSIXct"]] =
   ee[["character___POSIXct"]]
 

R/benchmark_grid.R

@@ -32,9 +32,8 @@
 #'
 #' # manual instantiation (not suited for a fair comparison of learners!)
 #' Map(function(task, resampling) {
-#'    resampling$instantiate(task)
+#'   resampling$instantiate(task)
 #' }, task = grid$task, resampling = grid$resampling)
-#'
 #' \dontrun{
 #' benchmark(grid)
 #' }

R/bibentries.R

@@ -15,15 +15,15 @@ bibentries = c(
 
   bischl_2012 = bibentry("article",
     title = "Resampling Methods for Meta-Model Validation with Recommendations for Evolutionary Computation",
-    author       = "Bernd Bischl and Olaf Mersmann and Heike Trautmann and Claus Weihs",
-    year         = "2012",
-    month        = "jun",
-    journal      = "Evolutionary Computation",
-    publisher    = "MIT Press - Journals",
-    volume       = "20",
-    number       = "2",
-    pages        = "249--275",
-    doi          = "10.1162/evco_a_00069"
+    author = "Bernd Bischl and Olaf Mersmann and Heike Trautmann and Claus Weihs",
+    year = "2012",
+    month = "jun",
+    journal = "Evolutionary Computation",
+    publisher = "MIT Press - Journals",
+    volume = "20",
+    number = "2",
+    pages = "249--275",
+    doi = "10.1162/evco_a_00069"
   ),
 
   breiman_1984 = bibentry("book",

R/default_measures.R

@@ -17,8 +17,9 @@
 #' default_measures("classif")
 #' default_measures("regr")
 default_measures = function(task_type) {
-  if (is.null(task_type))
+  if (is.null(task_type)) {
     return(list())
+  }
   assert_choice(task_type, names(mlr_reflections$default_measures))
   keys = mlr_reflections$default_measures[[task_type]]
   mlr_measures$mget(keys)