From b7a5994c43ffa1f9ef69836c2f94c38cfd3d7226 Mon Sep 17 00:00:00 2001
From: Emil Hvitfeldt <emilhhvitfeldt@gmail.com>
Date: Tue, 17 Sep 2024 11:56:25 -0700
Subject: [PATCH 1/6] another early exit in prepare_data()

---
 R/predict.R | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/R/predict.R b/R/predict.R
index 4bc22bd42..16a8cb536 100644
--- a/R/predict.R
+++ b/R/predict.R
@@ -470,6 +470,10 @@ prepare_data <- function(object, new_data) {
   if (allow_sparse(object) && inherits(new_data, "dgCMatrix")) {
     return(new_data)
   }
+  if (allow_sparse(object) && is_sparse_tibble(new_data)) {
+    new_data <- sparsevctrs::coerce_to_sparse_matrix(new_data)
+    return(new_data)
+  }
 
   fit_interface <- object$spec$method$fit$interface
   switch(

From 9bf1d2e8a7c1b47829a35297c78996ef3a215a2e Mon Sep 17 00:00:00 2001
From: Emil Hvitfeldt <emilhhvitfeldt@gmail.com>
Date: Tue, 17 Sep 2024 11:56:44 -0700
Subject: [PATCH 2/6] error for formula interface with sparse data

---
 NEWS.md                              |  2 +-
 R/convert_data.R                     |  7 ++++
 tests/testthat/_snaps/sparsevctrs.md | 24 ++++++++++++++
 tests/testthat/test-sparsevctrs.R    | 49 +++++++++++++++++++++++-----
 4 files changed, 73 insertions(+), 9 deletions(-)

diff --git a/NEWS.md b/NEWS.md
index da2a95691..8c470547f 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -2,7 +2,7 @@
 
 * `fit_xy()` can now take dgCMatrix input for `x` argument (#1121).
 
-* `fit()` and `fit_xy()` can now take sparse tibbles as data values (#1165).
+* `fit_xy()` can now take sparse tibbles as data values (#1165).
 
 * `predict()` can now take dgCMatrix and sparse tibble input for `new_data` argument, and error informatively when model doesn't support it (#1167).
 
diff --git a/R/convert_data.R b/R/convert_data.R
index 2fcf04c9f..4f1fb8d06 100644
--- a/R/convert_data.R
+++ b/R/convert_data.R
@@ -48,6 +48,13 @@
     )
   }
 
+  if (is_sparse_tibble(data)) {
+    cli::cli_abort(
+      "Sparse data cannot be used with formula interface. Please use 
+      {.fn fit_xy} instead."
+    )
+  }
+
   if (remove_intercept) {
     data <- data[, colnames(data) != "(Intercept)", drop = FALSE]
   }
diff --git a/tests/testthat/_snaps/sparsevctrs.md b/tests/testthat/_snaps/sparsevctrs.md
index 797dd2285..ae7cbcb66 100644
--- a/tests/testthat/_snaps/sparsevctrs.md
+++ b/tests/testthat/_snaps/sparsevctrs.md
@@ -1,5 +1,13 @@
 # sparse tibble can be passed to `fit()
 
+    Code
+      lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
+    Condition
+      Error in `.convert_form_to_xy_fit()`:
+      ! Sparse data cannot be used with formula interface. Please use `fit_xy()` instead.
+
+---
+
     Code
       lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data[1:100, ])
     Condition
@@ -8,6 +16,14 @@
 
 # sparse matrix can be passed to `fit()
 
+    Code
+      lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
+    Condition
+      Error in `.convert_form_to_xy_fit()`:
+      ! Sparse data cannot be used with formula interface. Please use `fit_xy()` instead.
+
+---
+
     Code
       lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data[1:100, ])
     Condition
@@ -46,6 +62,14 @@
       Error in `predict()`:
       ! `x` is a sparse matrix, but `linear_reg()` with engine "lm" doesn't accept that.
 
+# sparse data work with xgboost engine
+
+    Code
+      tree_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
+    Condition
+      Error in `.convert_form_to_xy_fit()`:
+      ! Sparse data cannot be used with formula interface. Please use `fit_xy()` instead.
+
 # to_sparse_data_frame() is used correctly
 
     Code
diff --git a/tests/testthat/test-sparsevctrs.R b/tests/testthat/test-sparsevctrs.R
index c1b2dfa09..bf0f69da2 100644
--- a/tests/testthat/test-sparsevctrs.R
+++ b/tests/testthat/test-sparsevctrs.R
@@ -1,5 +1,6 @@
 test_that("sparse tibble can be passed to `fit()", {
   skip_if_not_installed("xgboost")
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   hotel_data <- sparse_hotel_rates()
   hotel_data <- sparsevctrs::coerce_to_sparse_tibble(hotel_data)
@@ -7,8 +8,9 @@ test_that("sparse tibble can be passed to `fit()", {
   spec <- boost_tree() %>%
     set_mode("regression") %>%
     set_engine("xgboost")
-
-  expect_no_error(
+  
+  expect_snapshot(
+    error = TRUE,
     lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
   )
 
@@ -16,6 +18,8 @@ test_that("sparse tibble can be passed to `fit()", {
     set_mode("regression") %>%
     set_engine("lm")
 
+  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
+
   expect_snapshot(
     lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data[1:100, ])
   )
@@ -23,17 +27,21 @@ test_that("sparse tibble can be passed to `fit()", {
 
 test_that("sparse matrix can be passed to `fit()", {
   skip_if_not_installed("xgboost")
-
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
+  
   hotel_data <- sparse_hotel_rates()
-
+  
   spec <- boost_tree() %>%
     set_mode("regression") %>%
     set_engine("xgboost")
 
-  expect_no_error(
+  expect_snapshot(
+    error = TRUE,
     lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
   )
 
+  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
+
   spec <- linear_reg() %>%
     set_mode("regression") %>%
     set_engine("lm")
@@ -45,9 +53,14 @@ test_that("sparse matrix can be passed to `fit()", {
 
 test_that("sparse tibble can be passed to `fit_xy()", {
   skip_if_not_installed("xgboost")
-
+  
   hotel_data <- sparse_hotel_rates()
   hotel_data <- sparsevctrs::coerce_to_sparse_tibble(hotel_data)
+  
+  # materialize outcome
+  hotel_data$avg_price_per_room <- hotel_data$avg_price_per_room[]
+  
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   spec <- boost_tree() %>%
     set_mode("regression") %>%
@@ -57,6 +70,8 @@ test_that("sparse tibble can be passed to `fit_xy()", {
     lm_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
   )
 
+  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
+
   spec <- linear_reg() %>%
     set_mode("regression") %>%
     set_engine("lm")
@@ -68,6 +83,7 @@ test_that("sparse tibble can be passed to `fit_xy()", {
 
 test_that("sparse matrices can be passed to `fit_xy()", {
   skip_if_not_installed("xgboost")
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   hotel_data <- sparse_hotel_rates()
 
@@ -94,6 +110,11 @@ test_that("sparse tibble can be passed to `predict()", {
 
   hotel_data <- sparse_hotel_rates()
   hotel_data <- sparsevctrs::coerce_to_sparse_tibble(hotel_data)
+    
+  # materialize outcome
+  hotel_data$avg_price_per_room <- hotel_data$avg_price_per_room[]
+  
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   spec <- rand_forest(trees = 10) %>%
     set_mode("regression") %>%
@@ -105,6 +126,8 @@ test_that("sparse tibble can be passed to `predict()", {
     predict(tree_fit, hotel_data)
   )
 
+  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
+
   spec <- linear_reg() %>%
     set_mode("regression") %>%
     set_engine("lm")
@@ -122,6 +145,7 @@ test_that("sparse tibble can be passed to `predict()", {
 
 test_that("sparse matrices can be passed to `predict()", {
   skip_if_not_installed("ranger")
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   hotel_data <- sparse_hotel_rates()
 
@@ -151,6 +175,7 @@ test_that("sparse matrices can be passed to `predict()", {
 
 test_that("sparse data work with xgboost engine", {
   skip_if_not_installed("xgboost")
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   spec <- boost_tree() %>%
     set_mode("regression") %>%
@@ -161,7 +186,7 @@ test_that("sparse data work with xgboost engine", {
   expect_no_error(
     tree_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
    )
- 
+   
    expect_no_error(
      predict(tree_fit, hotel_data)
    )
@@ -169,7 +194,8 @@ test_that("sparse data work with xgboost engine", {
   hotel_data <- sparsevctrs::coerce_to_sparse_tibble(hotel_data)
 
 
-  expect_no_error(
+  expect_snapshot(
+    error = TRUE,
     tree_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
   )
 
@@ -177,6 +203,11 @@ test_that("sparse data work with xgboost engine", {
     predict(tree_fit, hotel_data)
   )
   
+  # materialize outcome
+  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
+  hotel_data$avg_price_per_room <- hotel_data$avg_price_per_room[]
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
+
   expect_no_error(
    tree_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
   )
@@ -188,6 +219,7 @@ test_that("sparse data work with xgboost engine", {
 
 test_that("to_sparse_data_frame() is used correctly", {
   skip_if_not_installed("xgboost")
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
   
   local_mocked_bindings(
     to_sparse_data_frame = function(x, object) {
@@ -228,6 +260,7 @@ test_that("to_sparse_data_frame() is used correctly", {
 
 test_that("maybe_sparse_matrix() is used correctly", {
   skip_if_not_installed("xgboost")
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
   
   local_mocked_bindings(
     maybe_sparse_matrix = function(x) {

From e332d2b5b4c6aebd29a16bac4722b103112bb9d1 Mon Sep 17 00:00:00 2001
From: Emil Hvitfeldt <emilhhvitfeldt@gmail.com>
Date: Wed, 18 Sep 2024 15:38:15 -0700
Subject: [PATCH 3/6] update sparse_hotel_rates with tibble argument

---
 tests/testthat/helper-objects.R   | 14 ++++++++++++--
 tests/testthat/test-sparsevctrs.R | 23 ++++-------------------
 2 files changed, 16 insertions(+), 21 deletions(-)

diff --git a/tests/testthat/helper-objects.R b/tests/testthat/helper-objects.R
index f2e24a353..b3cbd8276 100644
--- a/tests/testthat/helper-objects.R
+++ b/tests/testthat/helper-objects.R
@@ -28,7 +28,7 @@ is_tf_ok <- function() {
 # ------------------------------------------------------------------------------
 # For sparse tibble testing
 
-sparse_hotel_rates <- function() {
+sparse_hotel_rates <- function(tibble = FALSE) {
   # 99.2 sparsity
   hotel_rates <- modeldata::hotel_rates
 
@@ -49,5 +49,15 @@ sparse_hotel_rates <- function() {
   )
 
   res <- as.matrix(res)
-  Matrix::Matrix(res, sparse = TRUE)
+  res <- Matrix::Matrix(res, sparse = TRUE)
+
+  if (tibble) {
+    res <- sparsevctrs::coerce_to_sparse_tibble(res)
+
+    # materialize outcome
+    withr::local_options("sparsevctrs.verbose_materialize" = NULL)
+    res$avg_price_per_room <- res$avg_price_per_room[]
+  }
+
+  res
 }
diff --git a/tests/testthat/test-sparsevctrs.R b/tests/testthat/test-sparsevctrs.R
index bf0f69da2..ed9e803e7 100644
--- a/tests/testthat/test-sparsevctrs.R
+++ b/tests/testthat/test-sparsevctrs.R
@@ -2,8 +2,7 @@ test_that("sparse tibble can be passed to `fit()", {
   skip_if_not_installed("xgboost")
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
-  hotel_data <- sparse_hotel_rates()
-  hotel_data <- sparsevctrs::coerce_to_sparse_tibble(hotel_data)
+  hotel_data <- sparse_hotel_rates(tibble = TRUE)
 
   spec <- boost_tree() %>%
     set_mode("regression") %>%
@@ -54,11 +53,7 @@ test_that("sparse matrix can be passed to `fit()", {
 test_that("sparse tibble can be passed to `fit_xy()", {
   skip_if_not_installed("xgboost")
   
-  hotel_data <- sparse_hotel_rates()
-  hotel_data <- sparsevctrs::coerce_to_sparse_tibble(hotel_data)
-  
-  # materialize outcome
-  hotel_data$avg_price_per_room <- hotel_data$avg_price_per_room[]
+  hotel_data <- sparse_hotel_rates(tibble = TRUE)
   
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
@@ -108,11 +103,7 @@ test_that("sparse matrices can be passed to `fit_xy()", {
 test_that("sparse tibble can be passed to `predict()", {
   skip_if_not_installed("ranger")
 
-  hotel_data <- sparse_hotel_rates()
-  hotel_data <- sparsevctrs::coerce_to_sparse_tibble(hotel_data)
-    
-  # materialize outcome
-  hotel_data$avg_price_per_room <- hotel_data$avg_price_per_room[]
+  hotel_data <- sparse_hotel_rates(tibble = TRUE)
   
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
@@ -191,8 +182,7 @@ test_that("sparse data work with xgboost engine", {
      predict(tree_fit, hotel_data)
    )
 
-  hotel_data <- sparsevctrs::coerce_to_sparse_tibble(hotel_data)
-
+  hotel_data <- sparse_hotel_rates(tibble = TRUE)
 
   expect_snapshot(
     error = TRUE,
@@ -203,11 +193,6 @@ test_that("sparse data work with xgboost engine", {
     predict(tree_fit, hotel_data)
   )
   
-  # materialize outcome
-  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
-  hotel_data$avg_price_per_room <- hotel_data$avg_price_per_room[]
-  withr::local_options("sparsevctrs.verbose_materialize" = 3)
-
   expect_no_error(
    tree_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
   )

From efeafa4823c82714c0902912e9621146f6918a57 Mon Sep 17 00:00:00 2001
From: Emil Hvitfeldt <emilhhvitfeldt@gmail.com>
Date: Wed, 18 Sep 2024 15:50:02 -0700
Subject: [PATCH 4/6] split up test_that() chunks

---
 tests/testthat/_snaps/sparsevctrs.md | 16 +++++-----
 tests/testthat/test-sparsevctrs.R    | 45 ++++++++++++++++++----------
 2 files changed, 38 insertions(+), 23 deletions(-)

diff --git a/tests/testthat/_snaps/sparsevctrs.md b/tests/testthat/_snaps/sparsevctrs.md
index ae7cbcb66..5d9632f47 100644
--- a/tests/testthat/_snaps/sparsevctrs.md
+++ b/tests/testthat/_snaps/sparsevctrs.md
@@ -1,4 +1,4 @@
-# sparse tibble can be passed to `fit()
+# sparse tibble can be passed to `fit() - supported
 
     Code
       lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
@@ -6,7 +6,7 @@
       Error in `.convert_form_to_xy_fit()`:
       ! Sparse data cannot be used with formula interface. Please use `fit_xy()` instead.
 
----
+# sparse tibble can be passed to `fit() - unsupported
 
     Code
       lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data[1:100, ])
@@ -14,7 +14,7 @@
       Warning:
       `data` is a sparse tibble, but `linear_reg()` with engine "lm" doesn't accept that. Converting to non-sparse.
 
-# sparse matrix can be passed to `fit()
+# sparse matrix can be passed to `fit() - supported
 
     Code
       lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
@@ -22,7 +22,7 @@
       Error in `.convert_form_to_xy_fit()`:
       ! Sparse data cannot be used with formula interface. Please use `fit_xy()` instead.
 
----
+# sparse matrix can be passed to `fit() - unsupported
 
     Code
       lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data[1:100, ])
@@ -30,7 +30,7 @@
       Warning:
       `data` is a sparse tibble, but `linear_reg()` with engine "lm" doesn't accept that. Converting to non-sparse.
 
-# sparse tibble can be passed to `fit_xy()
+# sparse tibble can be passed to `fit_xy() - unsupported
 
     Code
       lm_fit <- fit_xy(spec, x = hotel_data[1:100, -1], y = hotel_data[1:100, 1])
@@ -38,7 +38,7 @@
       Warning:
       `x` is a sparse tibble, but `linear_reg()` with engine "lm" doesn't accept that. Converting to non-sparse.
 
-# sparse matrices can be passed to `fit_xy()
+# sparse matrices can be passed to `fit_xy() - unsupported
 
     Code
       lm_fit <- fit_xy(spec, x = hotel_data[1:100, -1], y = hotel_data[1:100, 1])
@@ -46,7 +46,7 @@
       Error in `fit_xy()`:
       ! `x` is a sparse matrix, but `linear_reg()` with engine "lm" doesn't accept that.
 
-# sparse tibble can be passed to `predict()
+# sparse tibble can be passed to `predict() - unsupported
 
     Code
       preds <- predict(lm_fit, sparse_mtcars)
@@ -54,7 +54,7 @@
       Warning:
       `x` is a sparse tibble, but `linear_reg()` with engine "lm" doesn't accept that. Converting to non-sparse.
 
-# sparse matrices can be passed to `predict()
+# sparse matrices can be passed to `predict() - unsupported
 
     Code
       predict(lm_fit, sparse_mtcars)
diff --git a/tests/testthat/test-sparsevctrs.R b/tests/testthat/test-sparsevctrs.R
index ed9e803e7..b611de4e3 100644
--- a/tests/testthat/test-sparsevctrs.R
+++ b/tests/testthat/test-sparsevctrs.R
@@ -1,4 +1,4 @@
-test_that("sparse tibble can be passed to `fit()", {
+test_that("sparse tibble can be passed to `fit() - supported", {
   skip_if_not_installed("xgboost")
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
@@ -12,19 +12,21 @@ test_that("sparse tibble can be passed to `fit()", {
     error = TRUE,
     lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
   )
+})
+
+test_that("sparse tibble can be passed to `fit() - unsupported", {
+  hotel_data <- sparse_hotel_rates(tibble = TRUE)
 
   spec <- linear_reg() %>%
     set_mode("regression") %>%
     set_engine("lm")
 
-  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
-
   expect_snapshot(
     lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data[1:100, ])
   )
 })
 
-test_that("sparse matrix can be passed to `fit()", {
+test_that("sparse matrix can be passed to `fit() - supported", {
   skip_if_not_installed("xgboost")
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
   
@@ -39,7 +41,10 @@ test_that("sparse matrix can be passed to `fit()", {
     lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
   )
 
-  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
+})
+
+test_that("sparse matrix can be passed to `fit() - unsupported", {
+  hotel_data <- sparse_hotel_rates()
 
   spec <- linear_reg() %>%
     set_mode("regression") %>%
@@ -50,12 +55,11 @@ test_that("sparse matrix can be passed to `fit()", {
   )
 })
 
-test_that("sparse tibble can be passed to `fit_xy()", {
+test_that("sparse tibble can be passed to `fit_xy() - supported", {
   skip_if_not_installed("xgboost")
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
   
   hotel_data <- sparse_hotel_rates(tibble = TRUE)
-  
-  withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   spec <- boost_tree() %>%
     set_mode("regression") %>%
@@ -64,8 +68,10 @@ test_that("sparse tibble can be passed to `fit_xy()", {
   expect_no_error(
     lm_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
   )
+})
 
-  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
+test_that("sparse tibble can be passed to `fit_xy() - unsupported", {
+  hotel_data <- sparse_hotel_rates(tibble = TRUE)
 
   spec <- linear_reg() %>%
     set_mode("regression") %>%
@@ -76,7 +82,7 @@ test_that("sparse tibble can be passed to `fit_xy()", {
   )
 })
 
-test_that("sparse matrices can be passed to `fit_xy()", {
+test_that("sparse matrices can be passed to `fit_xy() - supported", {
   skip_if_not_installed("xgboost")
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
@@ -89,6 +95,10 @@ test_that("sparse matrices can be passed to `fit_xy()", {
   expect_no_error(
     lm_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
   )
+})
+
+test_that("sparse matrices can be passed to `fit_xy() - unsupported", {
+  hotel_data <- sparse_hotel_rates()
 
   spec <- linear_reg() %>%
     set_mode("regression") %>%
@@ -100,12 +110,11 @@ test_that("sparse matrices can be passed to `fit_xy()", {
   )
 })
 
-test_that("sparse tibble can be passed to `predict()", {
+test_that("sparse tibble can be passed to `predict() - supported", {
   skip_if_not_installed("ranger")
+  withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   hotel_data <- sparse_hotel_rates(tibble = TRUE)
-  
-  withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   spec <- rand_forest(trees = 10) %>%
     set_mode("regression") %>%
@@ -116,8 +125,10 @@ test_that("sparse tibble can be passed to `predict()", {
   expect_no_error(
     predict(tree_fit, hotel_data)
   )
+})
 
-  withr::local_options("sparsevctrs.verbose_materialize" = NULL)
+test_that("sparse tibble can be passed to `predict() - unsupported", {
+  hotel_data <- sparse_hotel_rates(tibble = TRUE)
 
   spec <- linear_reg() %>%
     set_mode("regression") %>%
@@ -134,7 +145,7 @@ test_that("sparse tibble can be passed to `predict()", {
   )
 })
 
-test_that("sparse matrices can be passed to `predict()", {
+test_that("sparse matrices can be passed to `predict() - supported", {
   skip_if_not_installed("ranger")
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
@@ -149,6 +160,10 @@ test_that("sparse matrices can be passed to `predict()", {
   expect_no_error(
     predict(tree_fit, hotel_data)
   )
+})
+
+test_that("sparse matrices can be passed to `predict() - unsupported", {
+  hotel_data <- sparse_hotel_rates()
 
   spec <- linear_reg() %>%
     set_mode("regression") %>%

From a7ccbf427f105327ded9fb16df34cd370f304ee3 Mon Sep 17 00:00:00 2001
From: Emil Hvitfeldt <emilhhvitfeldt@gmail.com>
Date: Wed, 18 Sep 2024 16:05:24 -0700
Subject: [PATCH 5/6] use better names for fitted models

---
 tests/testthat/_snaps/sparsevctrs.md |  6 +++---
 tests/testthat/test-sparsevctrs.R    | 20 ++++++++++----------
 2 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/tests/testthat/_snaps/sparsevctrs.md b/tests/testthat/_snaps/sparsevctrs.md
index 5d9632f47..3619bf8dd 100644
--- a/tests/testthat/_snaps/sparsevctrs.md
+++ b/tests/testthat/_snaps/sparsevctrs.md
@@ -1,7 +1,7 @@
 # sparse tibble can be passed to `fit() - supported
 
     Code
-      lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
+      xgb_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
     Condition
       Error in `.convert_form_to_xy_fit()`:
       ! Sparse data cannot be used with formula interface. Please use `fit_xy()` instead.
@@ -17,7 +17,7 @@
 # sparse matrix can be passed to `fit() - supported
 
     Code
-      lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
+      xgb_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
     Condition
       Error in `.convert_form_to_xy_fit()`:
       ! Sparse data cannot be used with formula interface. Please use `fit_xy()` instead.
@@ -65,7 +65,7 @@
 # sparse data work with xgboost engine
 
     Code
-      tree_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
+      xgb_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
     Condition
       Error in `.convert_form_to_xy_fit()`:
       ! Sparse data cannot be used with formula interface. Please use `fit_xy()` instead.
diff --git a/tests/testthat/test-sparsevctrs.R b/tests/testthat/test-sparsevctrs.R
index b611de4e3..745d1a85e 100644
--- a/tests/testthat/test-sparsevctrs.R
+++ b/tests/testthat/test-sparsevctrs.R
@@ -10,7 +10,7 @@ test_that("sparse tibble can be passed to `fit() - supported", {
   
   expect_snapshot(
     error = TRUE,
-    lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
+    xgb_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
   )
 })
 
@@ -38,7 +38,7 @@ test_that("sparse matrix can be passed to `fit() - supported", {
 
   expect_snapshot(
     error = TRUE,
-    lm_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
+    xgb_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
   )
 
 })
@@ -66,7 +66,7 @@ test_that("sparse tibble can be passed to `fit_xy() - supported", {
     set_engine("xgboost")
 
   expect_no_error(
-    lm_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
+    xgb_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
   )
 })
 
@@ -93,7 +93,7 @@ test_that("sparse matrices can be passed to `fit_xy() - supported", {
     set_engine("xgboost")
 
   expect_no_error(
-    lm_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
+    xgb_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
   )
 })
 
@@ -190,30 +190,30 @@ test_that("sparse data work with xgboost engine", {
   hotel_data <- sparse_hotel_rates()
 
   expect_no_error(
-    tree_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
+    xgb_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
    )
    
    expect_no_error(
-     predict(tree_fit, hotel_data)
+     predict(xgb_fit, hotel_data)
    )
 
   hotel_data <- sparse_hotel_rates(tibble = TRUE)
 
   expect_snapshot(
     error = TRUE,
-    tree_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
+    xgb_fit <- fit(spec, avg_price_per_room ~ ., data = hotel_data)
   )
 
   expect_no_error(
-    predict(tree_fit, hotel_data)
+    predict(xgb_fit, hotel_data)
   )
   
   expect_no_error(
-   tree_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
+    xgb_fit <- fit_xy(spec, x = hotel_data[, -1], y = hotel_data[, 1])
   )
 
   expect_no_error(
-    predict(tree_fit, hotel_data)
+    predict(xgb_fit, hotel_data)
   )
 })
 

From 4cf99f038ae6e3b24759487b20b71a6afb1010fd Mon Sep 17 00:00:00 2001
From: Emil Hvitfeldt <emilhhvitfeldt@gmail.com>
Date: Wed, 18 Sep 2024 16:14:17 -0700
Subject: [PATCH 6/6] add comment about what sparsevctrs.verbose_materialize
 does

---
 tests/testthat/test-sparsevctrs.R | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/tests/testthat/test-sparsevctrs.R b/tests/testthat/test-sparsevctrs.R
index 745d1a85e..7759006e8 100644
--- a/tests/testthat/test-sparsevctrs.R
+++ b/tests/testthat/test-sparsevctrs.R
@@ -1,5 +1,7 @@
 test_that("sparse tibble can be passed to `fit() - supported", {
   skip_if_not_installed("xgboost")
+  # Make materialization of sparse vectors throw an error
+  # https://r-lib.github.io/sparsevctrs/dev/reference/sparsevctrs_options.html
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   hotel_data <- sparse_hotel_rates(tibble = TRUE)
@@ -28,6 +30,8 @@ test_that("sparse tibble can be passed to `fit() - unsupported", {
 
 test_that("sparse matrix can be passed to `fit() - supported", {
   skip_if_not_installed("xgboost")
+  # Make materialization of sparse vectors throw an error
+  # https://r-lib.github.io/sparsevctrs/dev/reference/sparsevctrs_options.html
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
   
   hotel_data <- sparse_hotel_rates()
@@ -57,6 +61,8 @@ test_that("sparse matrix can be passed to `fit() - unsupported", {
 
 test_that("sparse tibble can be passed to `fit_xy() - supported", {
   skip_if_not_installed("xgboost")
+  # Make materialization of sparse vectors throw an error
+  # https://r-lib.github.io/sparsevctrs/dev/reference/sparsevctrs_options.html
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
   
   hotel_data <- sparse_hotel_rates(tibble = TRUE)
@@ -84,6 +90,8 @@ test_that("sparse tibble can be passed to `fit_xy() - unsupported", {
 
 test_that("sparse matrices can be passed to `fit_xy() - supported", {
   skip_if_not_installed("xgboost")
+  # Make materialization of sparse vectors throw an error
+  # https://r-lib.github.io/sparsevctrs/dev/reference/sparsevctrs_options.html
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   hotel_data <- sparse_hotel_rates()
@@ -112,6 +120,8 @@ test_that("sparse matrices can be passed to `fit_xy() - unsupported", {
 
 test_that("sparse tibble can be passed to `predict() - supported", {
   skip_if_not_installed("ranger")
+  # Make materialization of sparse vectors throw an error
+  # https://r-lib.github.io/sparsevctrs/dev/reference/sparsevctrs_options.html
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   hotel_data <- sparse_hotel_rates(tibble = TRUE)
@@ -147,6 +157,8 @@ test_that("sparse tibble can be passed to `predict() - unsupported", {
 
 test_that("sparse matrices can be passed to `predict() - supported", {
   skip_if_not_installed("ranger")
+  # Make materialization of sparse vectors throw an error
+  # https://r-lib.github.io/sparsevctrs/dev/reference/sparsevctrs_options.html
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   hotel_data <- sparse_hotel_rates()
@@ -181,6 +193,8 @@ test_that("sparse matrices can be passed to `predict() - unsupported", {
 
 test_that("sparse data work with xgboost engine", {
   skip_if_not_installed("xgboost")
+  # Make materialization of sparse vectors throw an error
+  # https://r-lib.github.io/sparsevctrs/dev/reference/sparsevctrs_options.html
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
 
   spec <- boost_tree() %>%
@@ -219,6 +233,8 @@ test_that("sparse data work with xgboost engine", {
 
 test_that("to_sparse_data_frame() is used correctly", {
   skip_if_not_installed("xgboost")
+  # Make materialization of sparse vectors throw an error
+  # https://r-lib.github.io/sparsevctrs/dev/reference/sparsevctrs_options.html
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
   
   local_mocked_bindings(
@@ -260,6 +276,8 @@ test_that("to_sparse_data_frame() is used correctly", {
 
 test_that("maybe_sparse_matrix() is used correctly", {
   skip_if_not_installed("xgboost")
+  # Make materialization of sparse vectors throw an error
+  # https://r-lib.github.io/sparsevctrs/dev/reference/sparsevctrs_options.html
   withr::local_options("sparsevctrs.verbose_materialize" = 3)
   
   local_mocked_bindings(