Merge branch 'main' into release

Author: be-marc

R/HotstartStack.R

@@ -138,7 +138,7 @@ HotstartStack = R6Class("HotstartStack",
       hotstart_id = learner$param_set$ids(tags = "hotstart")
 
       set(self$stack, j = "cost", value = NA_real_)
-      cost = self$stack[list(.task_hash, .learner_hash), "cost" := map_dbl(get("start_learner"), function(l) calculate_cost(l, learner, hotstart_id)) , on = c("task_hash", "learner_hash")
+      cost = self$stack[list(.task_hash, .learner_hash), "cost" := map_dbl(get("start_learner"), function(l) calculate_cost(l, learner, hotstart_id)), on = c("task_hash", "learner_hash")
         ][, get("cost")]
       self$stack[, "cost" := NULL]
       cost

R/Learner.R

@@ -275,7 +275,7 @@ Learner = R6Class("Learner",
 
       pred_typs = replace(self$predict_types, self$predict_types == self$predict_type, paste0("[", self$predict_type, "]"))
       encapsulation = self$encapsulation[[1L]]
-      fallback = if (encapsulation != 'none') class(self$fallback)[[1L]] else "-"
+      fallback = if (encapsulation != "none") class(self$fallback)[[1L]] else "-"
 
       cat_cli({
         cli_li("Predict Types: {pred_typs}")

R/LearnerClassifDebug.R

@@ -196,14 +196,11 @@ LearnerClassifDebug = R6Class("LearnerClassifDebug", inherit = LearnerClassif,
       }
 
       model = list(
-          response = as.character(sample(task$truth(), 1L, prob = private$.get_weights(task))),
-          pid = Sys.getpid(),
-          id = UUIDgenerate(),
-          random_number = sample(100000, 1),
-          iter = if (isTRUE(pv$early_stopping))
-            sample(pv$iter %??% 1L, 1L)
-          else
-            pv$iter %??% 1L
+        response = as.character(sample(task$truth(), 1L, prob = private$.get_weights(task))),
+        pid = Sys.getpid(),
+        id = UUIDgenerate(),
+        random_number = sample(100000, 1),
+        iter = if (isTRUE(pv$early_stopping)) sample(pv$iter %??% 1L, 1L) else pv$iter %??% 1L
       )
 
       if (!is.null(valid_truth)) {

R/LearnerRegrDebug.R

@@ -64,7 +64,7 @@ LearnerRegrDebug = R6Class("LearnerRegrDebug", inherit = LearnerRegr,
     #' @return Named `numeric()`.
     importance = function() {
       if (is.null(self$model)) {
-       error_input("No model stored")
+        error_input("No model stored")
       }
       fns = self$state$feature_names
       set_names(rep(0, length(fns)), fns)

R/MeasureRegrPinball.R

@@ -28,8 +28,8 @@
 MeasureRegrPinball = R6Class("MeasureRegrPinball",
   inherit = MeasureRegr,
   public = list(
-   #' @description
-   #' Creates a new instance of this [R6][R6::R6Class] class.
+    #' @description
+    #' Creates a new instance of this [R6][R6::R6Class] class.
     initialize = function(alpha = 0.5) {
       param_set = ps(alpha = p_dbl(lower = 0, upper = 1))
       param_set$set_values(alpha = alpha)

R/MeasureRegrRQR.R

@@ -78,7 +78,7 @@ MeasureRegrRQR = R6Class("MeasureRQR",
           truth = prediction$truth,
           response = prediction$data$quantiles[, which(probs == alpha)],
           alpha = alpha
-         )
+        )
       )
 
       denominator = sum(
@@ -89,8 +89,8 @@ MeasureRegrRQR = R6Class("MeasureRQR",
         )
       )
 
-       1 - (numerator / denominator)
-     }
+      1 - (numerator / denominator)
+    }
   )
 )
 

R/PredictionClassif.R

@@ -104,7 +104,7 @@ PredictionClassif = R6Class("PredictionClassif", inherit = Prediction,
       weights = NULL,
       check = TRUE,
       extra = NULL
-      ) {
+    ) {
 
       pdata = new_prediction_data(
         list(row_ids = row_ids, truth = truth, response = response, prob = prob, weights = weights, extra = extra),

R/PredictionDataRegr.R

@@ -110,11 +110,18 @@ c.PredictionDataRegr = function(..., keep_duplicates = TRUE) { # nolint
     error_input("Cannot rbind predictions: Some predictions have extra data, others do not")
   }
 
-  elems = c("row_ids", "truth", intersect(predict_types[[1L]], c("response", "se")), if ("weights" %chin% names(dots[[1L]])) "weights")
+  nn = names(dots[[1L]])
+  elems = c("row_ids", "truth", intersect(predict_types[[1L]], c("response", "se")), if ("weights" %chin% nn) "weights")
   tab = map_dtr(dots, function(x) x[elems], .fill = FALSE)
-  quantiles = do.call(rbind, map(dots, "quantiles"))
 
-  extra = if ("extra" %chin% names(dots[[1L]])) {
+  quantiles = if ("quantiles" %chin% nn) {
+    quantiles = map(dots, "quantiles")
+    attrs = attributes(quantiles[[1L]])
+    quantiles = do.call(rbind, quantiles)
+    setattr(quantiles, "probs", attrs$props)
+    setattr(quantiles, "response", attrs$response)
+  }
+  extra = if ("extra" %chin% nn) {
     rbindlist(map(dots, "extra"), fill = TRUE, use.names = TRUE)
   }
 

R/PredictionRegr.R

@@ -65,7 +65,7 @@ PredictionRegr = R6Class("PredictionRegr", inherit = Prediction,
       weights = NULL,
       check = TRUE,
       extra = NULL
-      ) {
+    ) {
       pdata = new_prediction_data(
         list(row_ids = row_ids, truth = truth, response = response, se = se, quantiles = quantiles, distr = distr, weights = weights, extra = extra),
         task_type = "regr"

R/as_result_data.R

@@ -50,7 +50,7 @@ as_result_data = function(
   learner_states = NULL,
   data_extra = NULL,
   store_backends = TRUE
-  ) {
+) {
   assert_task(task)
   assert_learners(learners, task = task)
   assert_resampling(resampling, instantiated = TRUE)