Follow up: Add check for threshold, warn if NA in predictors, and minor styling cleanup

R/pca-fit.R

@@ -26,7 +26,10 @@ apd_pca_impl <- function(predictors, threshold) {
     retx = TRUE
   )
 
-  # TODO: verify threshold \in (0, 1]
+  if (threshold <= 0 || threshold > 1) {
+    rlang::abort("threshold must be between 0 and 1: (0, 1]")
+  }
+
   eigs <- pcs$sdev^2
   cum_sum <- cumsum(eigs) / sum(eigs)
   num_comp <- sum(cum_sum <= threshold) + 1

R/pca-score.R

@@ -7,6 +7,21 @@ score_apd_pca_numeric <- function(model, predictors) {
     rlang::abort("The model must contain a pcs argument.")
   }
 
+  incomplete_rows <- which(!stats::complete.cases(predictors))
+
+  if (length(incomplete_rows) > 0) {
+    cols_with_na <- names(which(colSums(is.na(predictors)) > 0))
+    msg <- paste0(
+      "`new_data` contains missing predictor values; score columns may be `NA`. ",
+      "Rows with missing values: ",
+      paste(incomplete_rows, collapse = ", "),
+      ". Columns with missing values: ",
+      paste(cols_with_na, collapse = ", "),
+      "."
+    )
+    rlang::warn(msg)
+  }
+
   # Predict output and subset using `num_comp`
   predicted_output <- stats::predict(model$pcs, predictors)
   predicted_output <- predicted_output[, 1:model$num_comp, drop = FALSE]

tests/testthat/test-pca-fit.R

@@ -110,3 +110,11 @@ test_that("`apd_pca` is not defined for vectors", {
     expected_message
   )
 })
+
+test_that ("invalid threshold value throws error", {
+  error_message <- "threshold must be between 0 and 1: \\(0, 1\\]"
+  expect_error(apd_pca(mtcars, threshold = 1.1),
+               error_message)
+  expect_error(apd_pca(mtcars, threshold = 0),
+               error_message)
+})

tests/testthat/test-pca-score.R

@@ -78,3 +78,14 @@ test_that("`score_apd_pca_bridge` output is correct", {
     expected
   )
 })
+
+test_that("`score_apd_pca_numeric` warns for missing predictor values", {
+  model <- apd_pca(mtcars %>% dplyr::slice(1:15))
+  predictors <- as.matrix(mtcars %>% dplyr::slice(16:30))
+  predictors[1, 1] <- NA_real_
+
+  expect_warning(
+    score_apd_pca_numeric(model, predictors),
+    "`new_data` contains missing predictor values"
+  )
+})