fix: removed setorder apllied to local optim df which caused reproducibility-issues

Author: kapsner

R/getNextParameters.R

@@ -1,5 +1,5 @@
 #' @importFrom dbscan dbscan
-#' @importFrom data.table fintersect uniqueN setorder
+#' @importFrom data.table fintersect uniqueN .SD
 getNextParameters <- function(
     LocalOptims
   , boundsDT
@@ -15,23 +15,21 @@ getNextParameters <- function(
 ) {
 
   LocalOptims <- LocalOptims[get("relUtility") >= acqThresh,]
-  LocalOptims <- LocalOptims[,c(boundsDT$N,"gpUtility"),with=FALSE]
-  
-  # using only one criteria to sort leads to not reproducible order
-  setorder(
-    LocalOptims,
-    cols = c("gpUtility", "Score", "inBounds"),
-    order = c(-1, -1, -1)
-  )
+  LocalOptims <- LocalOptims[
+    ,
+    .SD,
+    .SDcols = c(boundsDT$N,"gpUtility")
+  ]
+
   LocalOptims$acqOptimum <- TRUE
 
   # Mark clusters as duplicates if they have already been attempted. Note that
   # parameters must match exactly. Whether or not we should eliminate 'close'
   # parameters is experimental, and could cause problems as the parameter space
   # becomes more fully explored.
   LocalOptims$Duplicate <- checkDup(
-      LocalOptims[,boundsDT$N,with=FALSE]
-    , scoreSummary[,boundsDT$N,with=FALSE]
+      LocalOptims[, boundsDT$N, with = FALSE]
+    , scoreSummary[, boundsDT$N, with = FALSE]
   )
 
   # If we already have runNew non-duplicate local optims, use the best of those.