skpr v1.2.1: Rename and export normalize_design()

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: skpr
 Title: Design of Experiments Suite: Generate and Evaluate Optimal Designs
 Date: 2023-01-20
-Version: 1.2.0
+Version: 1.2.1
 Authors@R: c(person("Tyler", "Morgan-Wall", email = "[email protected]", role = c("aut", "cre")),
              person("George", "Khoury", email = "[email protected]", role = c("aut")))
 Description: Generates and evaluates D, I, A, Alias, E, T, and G optimal designs. Supports generation and evaluation of blocked and split/split-split/.../N-split plot designs. Includes parametric and Monte Carlo power evaluation functions, and supports calculating power for censored responses. Provides a framework to evaluate power using functions provided in other packages or written by the user. Includes a Shiny graphical user interface that displays the underlying code used to create and evaluate the design to improve ease-of-use and make analyses more reproducible. For details, see Morgan-Wall et al. (2021) <doi:10.18637/jss.v099.i01>.

NAMESPACE

@@ -13,6 +13,7 @@ export(gen_design)
 export(get_attribute)
 export(get_optimality)
 export(get_power_curve_output)
+export(normalize_design)
 export(plot_correlations)
 export(plot_fds)
 export(skprGUI)

R/eval_design.R

@@ -286,7 +286,7 @@ eval_design = function(design, model = NULL, alpha = 0.05,
   }
 
   #------Normalize/Center numeric columns ------#
-  run_matrix_processed = normalize_numeric_runmatrix(run_matrix_processed)
+  run_matrix_processed = normalize_design(run_matrix_processed)
 
   #-Generate Model Matrix & Anticipated Coefficients-#
   #Variables used later: anticoef

R/eval_design_custom_mc.R

@@ -167,7 +167,7 @@ eval_design_custom_mc = function(design, model = NULL, alpha = 0.05,
   model = rearrange_formula_by_order(model, data = run_matrix_processed)
 
   #------Normalize/Center numeric columns ------#
-  run_matrix_processed = normalize_numeric_runmatrix(run_matrix_processed)
+  run_matrix_processed = normalize_design(run_matrix_processed)
 
   #Remove skpr-generated REML blocking indicators if present
   run_matrix_processed = remove_skpr_blockcols(run_matrix_processed)

R/eval_design_mc.R

@@ -392,7 +392,7 @@ eval_design_mc = function(design, model = NULL, alpha = 0.05,
   }
 
   #------Normalize/Center numeric columns ------#
-  run_matrix_processed = normalize_numeric_runmatrix(run_matrix_processed)
+  run_matrix_processed = normalize_design(run_matrix_processed)
 
   #---------- Generating model matrix ----------#
   #Remove columns from variables not used in the model

R/eval_design_survival_mc.R

@@ -213,7 +213,7 @@ eval_design_survival_mc = function(design, model = NULL, alpha = 0.05,
 
 
   #------Normalize/Center numeric columns ------#
-  run_matrix_processed = normalize_numeric_runmatrix(run_matrix_processed)
+  run_matrix_processed = normalize_design(run_matrix_processed)
 
   #---------- Generating model matrix ----------#
   #remove columns from variables not used in the model

R/gen_design.R

@@ -350,10 +350,10 @@ gen_design = function(candidateset, model, trials,
         }
         if (length(contrastslisttemp) == 0) {
           advancedoptions$g_efficiency_samples = model.matrix(model,
-                                        normalize_numeric_runmatrix(advancedoptions$g_efficiency_samples))
+                                        normalize_design(advancedoptions$g_efficiency_samples))
         } else {
           advancedoptions$g_efficiency_samples = suppressWarnings(model.matrix(model,
-                                                         normalize_numeric_runmatrix(advancedoptions$g_efficiency_samples),
+                                                         normalize_design(advancedoptions$g_efficiency_samples),
                                                          contrasts.arg = contrastslisttemp))
         }
       } else {
@@ -578,7 +578,7 @@ gen_design = function(candidateset, model, trials,
       }
     }
     #Normalize
-    augmentnormalized = normalize_numeric_runmatrix(augmentdesign, candidateset)
+    augmentnormalized = normalize_design(augmentdesign, candidateset)
     augmentedrows = nrow(augmentdesign)
   } else {
     augmentedrows = 0
@@ -587,11 +587,11 @@ gen_design = function(candidateset, model, trials,
   #------Normalize/Center numeric columns ------#
   candidatesetnormalized = candidateset
 
-  candidatesetnormalized = normalize_numeric_runmatrix(candidateset, augmentdesign)
-  fullcandidatesetnorm = normalize_numeric_runmatrix(fullcandidateset, augmentdesign)
+  candidatesetnormalized = normalize_design(candidateset, augmentdesign)
+  fullcandidatesetnorm = normalize_design(fullcandidateset, augmentdesign)
 
   if (!is.null(splitplotdesign)) {
-    spdnormalized = normalize_numeric_runmatrix(splitplotdesign)
+    spdnormalized = normalize_design(splitplotdesign)
   }
 
   splitplot = FALSE

R/normalize_design.R

@@ -0,0 +1,46 @@
+#'@title Normalize Design
+#'
+#'@description Normalizes the numeric columns in the design to -1 to 1. This is important to do if your model has interaction or polynomial terms,
+#'as these terms can introduce multi-collinearity and standardizing the numeric columns can reduce this problem.
+#'
+#'@param design The design matrix.
+#'@param augmented Default `NULL`. If augmenting an existing design, this should be the pre-existing design. The column types must match design
+#'
+#'@return Normalized design matrix
+#'@export
+#'@examples
+#'#Normalize a design
+#'\dontrun{
+#'cand_set = expand.grid(temp = c(100,300,500),
+#'                       altitude = c(10000,20000),
+#'                       offset = seq(-10,-5,by=1),
+#'                       type = c("A","B", "C"))
+#'design = gen_design(cand_set, ~., 24)
+#'#Un-normalized design
+#'design
+#'
+#'#Normalized design
+#'normalize_design(design)
+#'}
+normalize_design = function(design, augmented = NULL) {
+  if(!is.null(augmented)) {
+    all_equal_classes = all(identical(lapply(design,class), unlist(lapply(augmented,class))))
+    if(!all_equal_classes) {
+      stop("Design to be augmented and new design must have identical column classes")
+    }
+    for (column in 1:ncol(design)) {
+      if (is.numeric(design[, column])) {
+        midvalue = mean(c(max(c(design[, column],augmented[,column])), min(c(design[, column],augmented[,column]))))
+        design[, column] = (design[, column] - midvalue) / (max(c(design[, column],augmented[,column])) - midvalue)
+      }
+    }
+  } else {
+    for (column in 1:ncol(design)) {
+      if (is.numeric(design[, column])) {
+        midvalue = mean(c(max(design[, column]), min(design[, column])))
+        design[, column] = (design[, column] - midvalue) / (max(design[, column]) - midvalue)
+      }
+    }
+  }
+  return(design)
+}

R/normalize_numeric_runmatrix.R

@@ -1,24 +1,27 @@
-#'@title Normalize Run Matrix
+#'@title Normalize Design
 #'
-#'@description Normalizes the numeric columns in the runmatrix
-#'@param RunMatrix The run matrix
+#'@description Normalizes the numeric columns in the design to -1 to 1. This is important to do if your model has interaction or polynomial terms,
+#'as these terms can introduce multi-collinearity and standardizing the numeric columns can reduce this problem.
+#'
+#'@param design The design matrix.
+#'@param augmented Default `NULL`. If
 #'@return Normalized run matrix
 #'@keywords internal
-normalize_numeric_runmatrix = function(RunMatrix, augmented = NULL) {
+normalize_design = function(design, augmented = NULL) {
   if(!is.null(augmented)) {
-    for (column in 1:ncol(RunMatrix)) {
-      if (is.numeric(RunMatrix[, column])) {
-        midvalue = mean(c(max(c(RunMatrix[, column],augmented[,column])), min(c(RunMatrix[, column],augmented[,column]))))
-        RunMatrix[, column] = (RunMatrix[, column] - midvalue) / (max(c(RunMatrix[, column],augmented[,column])) - midvalue)
+    for (column in 1:ncol(design)) {
+      if (is.numeric(design[, column])) {
+        midvalue = mean(c(max(c(design[, column],augmented[,column])), min(c(design[, column],augmented[,column]))))
+        design[, column] = (design[, column] - midvalue) / (max(c(design[, column],augmented[,column])) - midvalue)
       }
     }
   } else {
-    for (column in 1:ncol(RunMatrix)) {
-      if (is.numeric(RunMatrix[, column])) {
-        midvalue = mean(c(max(RunMatrix[, column]), min(RunMatrix[, column])))
-        RunMatrix[, column] = (RunMatrix[, column] - midvalue) / (max(RunMatrix[, column]) - midvalue)
+    for (column in 1:ncol(design)) {
+      if (is.numeric(design[, column])) {
+        midvalue = mean(c(max(design[, column]), min(design[, column])))
+        design[, column] = (design[, column] - midvalue) / (max(design[, column]) - midvalue)
       }
     }
   }
-  return(RunMatrix)
+  return(design)
 }