Merge pull request #1364 from tidymodels/sparse-tibble-recipe

make sure `recipe()` accepts sparse tibbles

DESCRIPTION

@@ -34,6 +34,7 @@ Imports:
     Matrix,
     purrr (>= 1.0.0),
     rlang (>= 1.1.0),
+    sparsevctrs (>= 0.1.0.9000),
     stats,
     tibble,
     tidyr (>= 1.0.0),
@@ -62,6 +63,8 @@ Suggests:
     testthat (>= 3.0.0),
     workflows,
     xml2
+Remotes:  
+    r-lib/sparsevctrs
 VignetteBuilder: 
     knitr
 RdMacros: 

NEWS.md

@@ -2,6 +2,8 @@
 
 * Example for `step_novel()` now better illustrates how it works. (@Edgar-Zamora, #1248)
 
+* `recipe()` now works with sparse tibbles. (#1364)
+
 # recipes 1.1.0
 
 ## Improvements

R/sparsevctrs.R

@@ -0,0 +1,3 @@
+is_sparse_tibble <- function(x) {
+  any(vapply(x, sparsevctrs::is_sparse_vector, logical(1)))
+}

tests/testthat/helper-sparsevctrs.R

@@ -0,0 +1,26 @@
+# ------------------------------------------------------------------------------
+# For sparse tibble testing
+
+sparse_hotel_rates <- function() {
+  # 99.2 sparsity
+  hotel_rates <- modeldata::hotel_rates
+
+  prefix_colnames <- function(x, prefix) {
+    colnames(x) <- paste(colnames(x), prefix, sep = "_")
+    x
+  }
+
+  dummies_country <- hardhat::fct_encode_one_hot(hotel_rates$country)
+  dummies_company <- hardhat::fct_encode_one_hot(hotel_rates$company)
+  dummies_agent <- hardhat::fct_encode_one_hot(hotel_rates$agent)
+
+  res <- dplyr::bind_cols(
+    hotel_rates["avg_price_per_room"],
+    prefix_colnames(dummies_country, "country"),
+    prefix_colnames(dummies_company, "company"),
+    prefix_colnames(dummies_agent, "agent")
+  )
+
+  res <- as.matrix(res)
+  Matrix::Matrix(res, sparse = TRUE)
+}

tests/testthat/test-sparsevctrs.R

@@ -0,0 +1,31 @@
+test_that("recipe() accepts sparse tibbles", {
+  skip_if_not_installed("modeldata")
+
+  hotel_data <- sparse_hotel_rates()
+  hotel_data <- sparsevctrs::coerce_to_sparse_tibble(hotel_data)
+
+  expect_no_condition(
+    rec_spec <- recipe(avg_price_per_room ~ ., data = hotel_data)
+  )
+
+  expect_true(
+    is_sparse_tibble(rec_spec$template)
+  )
+
+  expect_no_condition(
+    rec_spec <- recipe(hotel_data)
+  )
+
+  expect_true(
+    is_sparse_tibble(rec_spec$template)
+  )
+
+  expect_no_condition(
+    rec_spec <- recipe(hotel_data, avg_price_per_room ~ .)
+  )
+
+  expect_true(
+    is_sparse_tibble(rec_spec$template)
+  )
+})
+