testing and snapshot updates for tidymodels/parsnip#1162

tests/testthat/_snaps/glmnet-linear.md

@@ -25,6 +25,6 @@
       linear_reg(penalty = 0.01) %>% set_engine("glmnet") %>% fit(mpg ~ ., data = mtcars) %>%
         multi_predict(mtcars, type = "class")
     Condition
-      Error in `check_pred_type()`:
+      Error in `multi_predict()`:
       ! For class predictions, the object should be a classification model.
 

tests/testthat/_snaps/glmnet-logistic.md

@@ -36,6 +36,6 @@
         funded_amnt) + int_rate + term, data = lending_club) %>% multi_predict(
         lending_club, type = "time")
     Condition
-      Error in `check_pred_type()`:
+      Error in `multi_predict()`:
       ! For event time predictions, the object should be a censored regression.
 

tests/testthat/_snaps/glmnet-multinom.md

@@ -33,6 +33,6 @@
       multinom_reg(penalty = 0.01) %>% set_engine("glmnet") %>% fit(class ~ ., data = hpc_data) %>%
         multi_predict(hpc_data, type = "numeric")
     Condition
-      Error in `check_pred_type()`:
+      Error in `multi_predict()`:
       ! For numeric predictions, the object should be a regression model.
 

tests/testthat/_snaps/parsnip-case-weights.md

@@ -0,0 +1,46 @@
+# boost_tree - xgboost case weights
+
+    Code
+      print(wt_fit$fit$call)
+    Output
+      xgboost::xgb.train(params = list(eta = 0.3, max_depth = 6, gamma = 0, 
+          colsample_bytree = 1, colsample_bynode = 1, min_child_weight = 1, 
+          subsample = 1), data = x$data, nrounds = 15, watchlist = x$watchlist, 
+          verbose = 0, nthread = 1, objective = "binary:logistic")
+
+# decision_tree - rpart case weights
+
+    Code
+      print(wt_fit$fit$call)
+    Output
+      rpart::rpart(formula = Class ~ ., data = data, weights = weights)
+
+# logistic_reg - stan case weights
+
+    Code
+      print(wt_fit$fit$call)
+    Output
+      rstanarm::stan_glm(formula = Class ~ ., family = stats::binomial, 
+          data = data, weights = weights, seed = ~1, refresh = 0)
+
+# mars - earth case weights
+
+    Code
+      print(wt_fit$fit$call)
+    Output
+      earth(formula = Class ~ ., data = data, weights = weights, keepxy = TRUE, 
+          glm = ~list(family = stats::binomial))
+
+# mlp - nnet case weights
+
+    Case weights are not enabled by the underlying model implementation.
+
+# rand_forest - ranger case weights
+
+    Code
+      print(wt_fit$fit$call)
+    Output
+      ranger::ranger(x = maybe_data_frame(x), y = y, num.threads = 1, 
+          verbose = FALSE, seed = sample.int(10^5, 1), probability = TRUE, 
+          case.weights = weights)
+

tests/testthat/_snaps/parsnip-survival-censoring-model.md

@@ -21,5 +21,5 @@
       predict(alt_obj, time = 100)
     Condition
       Error in `predict()`:
-      ! Don't know how to predict with a censoring model of type: reverse_km
+      ! Don't know how to predict with a censoring model of type reverse_km.
 

tests/testthat/_snaps/parsnip-survival-censoring-weights.md

@@ -4,15 +4,15 @@
       .censoring_weights_graf("nothing useful")
     Condition
       Error in `.censoring_weights_graf()`:
-      ! There is no `.censoring_weights_graf()` method for objects with class(es): 'character'
+      ! There is no `.censoring_weights_graf()` method for objects with class <character>.
 
 ---
 
     Code
       .censoring_weights_graf(cox_model, lung)
     Condition
       Error:
-      ! There should be a single column of class `Surv`
+      ! There should be a single column of class <Surv>.
 
 ---
 
@@ -22,23 +22,23 @@
       .censoring_weights_graf(cox_model, lung_left)
     Condition
       Error in `.censoring_weights_graf()`:
-      ! For this usage, the allowed censoring type is: 'right'
+      ! For this usage, the allowed censoring type is right.
 
 ---
 
     Code
       .censoring_weights_graf(cox_model, lung2)
     Condition
       Error:
-      ! The input should have a list column called `.pred`.
+      ! The input should have a list column called ".pred".
 
 ---
 
     Code
       .censoring_weights_graf(cox_model, preds, cens_predictors = "shouldn't be using this anyway!")
     Condition
       Warning:
-      The 'cens_predictors' argument to the survival weighting function is not currently used.
+      `cens_predictors` is not currently used.
     Output
       # A tibble: 3 x 2
         .pred              surv
@@ -60,6 +60,6 @@
     Code
       .censoring_weights_graf(wrong_model, mtcars)
     Condition
-      Error in `.check_censor_model()`:
+      Error in `.censoring_weights_graf()`:
       ! The model needs to be for mode 'censored regression', not for mode 'regression'.
 

tests/testthat/_snaps/parsnip-survival-standalone.md

@@ -4,21 +4,21 @@
       parsnip:::.is_surv(1)
     Condition
       Error:
-      ! The object does not have class `Surv`.
+      ! The object does not have class <Surv>.
 
 # .check_cens_type()
 
     Code
       parsnip:::.check_cens_type(left_c, type = "right", fail = TRUE)
     Condition
       Error:
-      ! For this usage, the allowed censoring type is: 'right'
+      ! For this usage, the allowed censoring type is right.
 
 ---
 
     Code
       parsnip:::.check_cens_type(left_c, type = c("right", "interval"), fail = TRUE)
     Condition
       Error:
-      ! For this usage, the allowed censoring types are: 'right' and 'interval'
+      ! For this usage, the allowed censoring types are right or interval.
 

tests/testthat/_snaps/recipes1.1.0/recipes-nnmf_sparse.md

@@ -0,0 +1,36 @@
+# Correct values
+
+    Code
+      print(rec)
+    Message
+      
+      -- Recipe ----------------------------------------------------------------------
+      
+      -- Inputs 
+      Number of variables by role
+      outcome:   1
+      predictor: 4
+      
+      -- Operations 
+      * Non-negative matrix factorization for: all_predictors()
+
+# No NNF
+
+    Code
+      print(rec)
+    Message
+      
+      -- Recipe ----------------------------------------------------------------------
+      
+      -- Inputs 
+      Number of variables by role
+      outcome:   1
+      predictor: 4
+      
+      -- Training information 
+      Training data contained 150 data points and no incomplete rows.
+      
+      -- Operations 
+      * No non-negative matrix factorization was extracted from: Sepal.Length,
+        Sepal.Width, Petal.Length, Petal.Width | Trained
+

tests/testthat/test-glmnet-linear.R

@@ -265,7 +265,7 @@ test_that('multi_predict() with default or single penalty value', {
 
 test_that('error traps', {
   skip_if_not_installed("glmnet")
-  skip_if_not_installed("parsnip", minimum_version = "1.2.1.9001")
+  skip_if_not_installed("parsnip", minimum_version = "1.2.1.9002")
 
   expect_snapshot(error = TRUE, {
     linear_reg(penalty = 0.01) %>%

tests/testthat/test-glmnet-logistic.R

@@ -377,7 +377,7 @@ test_that("class predictions are factors with all levels", {
 
 test_that('error traps', {
   skip_if_not_installed("glmnet")
-  skip_if_not_installed("parsnip", minimum_version = "1.2.1.9001")
+  skip_if_not_installed("parsnip", minimum_version = "1.2.1.9002")
 
   data("lending_club", package = "modeldata", envir = rlang::current_env())
 

tests/testthat/test-glmnet-multinom.R

@@ -399,7 +399,7 @@ test_that("class predictions are factors with all levels", {
 
 test_that('error traps', {
   skip_if_not_installed("glmnet")
-  skip_if_not_installed("parsnip", minimum_version = "1.2.1.9001")
+  skip_if_not_installed("parsnip", minimum_version = "1.2.1.9002")
 
   data("hpc_data", package = "modeldata", envir = rlang::current_env())
 

tests/testthat/test-parsnip-survival-censoring-weights.R

@@ -174,6 +174,8 @@ test_that('compute Graf weights', {
 })
 
 test_that("error messages in context of .censoring_weights_graf()", {
+  skip_if_not_installed("parsnip", minimum_version = "1.2.1.9002")
+
   lung2 <- lung %>%
     dplyr::mutate(surv = Surv(time, status), .keep = "unused")
 
@@ -228,7 +230,7 @@ test_that("error for .censoring_weights_graf.workflow()", {
 
 test_that("error for .censoring_weights_graf() from .check_censor_model()", {
   # temporarily its own test, see above
-  skip_if_not_installed("parsnip", minimum_version = "1.1.0.9003")
+  skip_if_not_installed("parsnip", minimum_version = "1.2.1.9002")
   wrong_model <- fit(linear_reg(), mpg ~ ., data = mtcars)
   expect_snapshot(error = TRUE, .censoring_weights_graf(wrong_model, mtcars))
 })