Merge branch 'master' into s3params_compat

NEWS.md

@@ -1,6 +1,7 @@
 # mlr3tuning (development version)
 
 * Compatibility with new paradox version 1.0.0.
+* fix: `TunerIrace` failed with logical parameters and dependencies.
 
 # mlr3tuning 0.19.2
 

R/Tuner.R

@@ -127,6 +127,7 @@ Tuner = R6Class("Tuner",
     #' @return [data.table::data.table()]
     optimize = function(inst) {
       assert_multi_class(inst, c("TuningInstanceSingleCrit", "TuningInstanceMultiCrit"))
+      inst$archive$start_time = Sys.time()
       inst$.__enclos_env__$private$.context = ContextOptimization$new(instance = inst, optimizer = self)
       call_back("on_optimization_begin", inst$callbacks, get_private(inst)$.context)
 

R/TunerIrace.R

@@ -145,9 +145,11 @@ target_runner_tuning = function(experiment, exec.target.runner, scenario, target
     # add configuration and instance id to archive
     set(configuration, j = "configuration", value = e$id.configuration)
     set(configuration, j = "instance", value = e$id.instance)
-    # fix logicals
-    configuration[, map(.SD, function(x) ifelse(x %in% c("TRUE", "FALSE"), as.logical(x), x))]
+    configuration
   })
+  # fix logicals
+  lgl_params = as.data.table(tuning_instance$search_space)[class == "ParamLgl", id]
+  if (length(lgl_params)) xdt[, (lgl_params) := lapply(.SD, as.logical), .SDcols = lgl_params]
 
   # provide experiment instances to objective
   tuning_instance$objective$constants$values$resampling = map(experiment, function(e) e$instance)