Full check for dev version 0.2.20240109

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: ale
 Title: Interpretable Machine Learning and Statistical Inference with Accumulated Local Effects (ALE)
-Version: 0.2.20240105
+Version: 0.2.20240109
 Authors@R: c(
     person("Chitu", "Okoli", , "[email protected]", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0001-5574-7572")),
@@ -30,12 +30,14 @@ Imports:
     ellipsis,
     glue,
     grDevices,
+    insight,
     labeling,
     methods,
     purrr,
     stats,
     stringr,
     tidyr,
+    univariateML,
     yaImpute
 Depends: 
     R (>= 3.5.0),

NAMESPACE

@@ -2,11 +2,13 @@
 
 export(ale)
 export(ale_ixn)
+export(create_p_funs)
 export(model_bootstrap)
 import(assertthat)
 import(dplyr)
 import(ggplot2)
 import(purrr)
+importFrom(glue,glue)
 importFrom(stats,median)
 importFrom(stats,quantile)
 importFrom(stats,sd)

R/ale_core.R

@@ -200,6 +200,9 @@
 #'
 #' }
 #'
+#'
+#' @importFrom glue glue
+#'
 ale <- function (
     data,
     model,

R/calc_ale.R

@@ -60,10 +60,11 @@ calc_ale <- function(
   # local variables.
   `:=` <- NULL
   ale_y <- NULL
-  Var1 <- NULL
+  estimate <- NULL
   Freq <- NULL
+  p.value <- NULL
   statistic <- NULL
-  estimate <- NULL
+  Var1 <- NULL
 
 
 

R/model_bootstrap.R

@@ -50,7 +50,8 @@
 #' @param model_call_string character string. If NULL, [model_bootstrap()] tries to
 #' automatically detect and construct the call for bootstrapped datasets. If it cannot, the
 #' function will fail early. In that case, a character string of the full call
-#' for the model must be provided that includes `boot_data` as the data argument for the call. See examples
+#' for the model must be provided that includes `boot_data` as the data argument for the call.
+#' See examples.
 # Future arguments:
 # * y_col: name of y column in data. This would allow SD and MAD to be calculated.
 # * pred_fun,pred_type: allows the prediction function to be called; this would
@@ -127,7 +128,7 @@
 #'   attitude,
 #'   model_call_string = 'mgcv::gam(
 #'     rating ~ complaints + privileges + s(learning) +
-#'       raises + s(critical) + advance),
+#'       raises + s(critical) + advance,
 #'     data = boot_data
 #'   )',
 #'   boot_it = 3
@@ -255,14 +256,15 @@ model_bootstrap <- function (
   ale_y <- NULL
   ale_y_mean <- NULL
   ale_y_median <- NULL
+  aled <- NULL
+  estimate <- NULL
   it <- NULL
+  naler_max <- NULL
   name <- NULL
-  value <- NULL
-  term <- NULL
-  estimate <- NULL
+  p.value <- NULL
   statistic <- NULL
-  aled <- NULL
-  naler_max <- NULL
+  term <- NULL
+  value <- NULL
 
 
   # Create bootstrap tbl

R/stats.R

@@ -164,6 +164,10 @@
 #' # Plot the ALE data. The horizontal bands in the plots use the p-values.
 #' gridExtra::grid.arrange(grobs = ale_gam_diamonds$plots, ncol = 2)
 #' }
+#'
+#'
+#' @importFrom glue glue
+#'
 create_p_funs <- function(
     training_data,
     test_data,
@@ -178,6 +182,7 @@ create_p_funs <- function(
     silent = FALSE,
     .testing_mode = FALSE
 ) {
+
   # Validate arguments
 
   assert_that(training_data |> inherits('data.frame'))
@@ -245,6 +250,17 @@ create_p_funs <- function(
   assert_that(is.flag(silent))
 
 
+  # Hack to prevent devtools::check from thinking that masked variables are global:
+  # Make them null local variables within the function with the issues. So,
+  # when masking applies, the masked variables will be prioritized over these null
+  # local variables.
+
+  # This super-assignment might be problematic. See this tip from ChatGPT to resolve it:
+  # https://chat.openai.com/c/08b68562-c339-4c37-baab-3c71d2e9fb73
+  .rand_model <<- NULL
+  .rand_test <<- NULL
+  .rand_train <<- NULL
+
 
 
   # Determine the closest distribution of the residuals
@@ -294,7 +310,7 @@ create_p_funs <- function(
       else {  # use the automatically detected model call
         # Update the model to call to add .random_variable and to train on .rand_train
         model_call$formula <- model_call$formula |>
-          update.formula(~ . + .random_variable)
+          stats::update.formula(~ . + .random_variable)
         model_call$data <- .rand_train
 
         .rand_model <<- eval(model_call)
@@ -314,11 +330,11 @@ create_p_funs <- function(
       .rand_ale
     })
 
-  ale_y_norm <- ale:::create_ale_y_norm_function(test_data[[y_col]])
+  ale_y_norm <- create_ale_y_norm_function(test_data[[y_col]])
 
   rand_stats <-
     map(rand_ales, \(.rand) {
-      ale:::ale_stats(
+      ale_stats(
         ale_y = .rand$data$.random_variable$ale_y,
         ale_n = .rand$data$.random_variable$ale_n,
         ale_y_norm_fun = ale_y_norm,
@@ -336,11 +352,11 @@ create_p_funs <- function(
 
         # For aler_min and naler_min, the p-value is the simple ECDF
         if (stringr::str_sub(.name_stat, -4, -1) == '_min') {
-          ecdf(.stat_vals)(x)
+          stats::ecdf(.stat_vals)(x)
         }
         # For other statistics, the p-value is 1 - ECDF.
         else {
-          1 - ecdf(.stat_vals)(x)
+          1 - stats::ecdf(.stat_vals)(x)
         }
       }
     }
@@ -361,13 +377,13 @@ create_p_funs <- function(
         if (stringr::str_sub(.name_stat, -4, -1) == '_min') {
           .stat_vals |>
             quantile(probs = p) |>
-            setNames(p)
+            stats::setNames(p)
         }
         # For other statistics, the value is the quantile of 1 - p
         else {
           .stat_vals |>
             quantile(probs = 1 - p) |>
-            setNames(p)
+            stats::setNames(p)
         }
       }
     }
@@ -383,7 +399,7 @@ create_p_funs <- function(
 
   # Set S3 class information for the p_funs object
   class(p_funs) <- c('p_funs', 'ale', 'list')
-  attr(p_funs, 'ale_version') <- packageVersion('ale')
+  attr(p_funs, 'ale_version') <- utils::packageVersion('ale')
 
   return(p_funs)
 

inst/WORDLIST

@@ -30,13 +30,15 @@ interpretable
 Interpretable
 Jingyu
 MMDD
+Mmmm
 mtcars
 NALED
 NALER
 Noam
 noncrucial
 overfit
 PDP
+pkgdown
 preprocesses
 pretrained
 programmatically
@@ -56,6 +58,7 @@ www
 x1
 x2
 x3
+yyyy
 YYYY
 Zhu