Merge branch 'main' into refactor-eval-time-default

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: tune
 Title: Tidy Tuning Tools
-Version: 1.1.2.9000
+Version: 1.1.2.9001
 Authors@R: c(
     person("Max", "Kuhn", , "[email protected]", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0003-2402-136X")),
@@ -38,7 +38,7 @@ Imports:
     vctrs (>= 0.6.1),
     withr,
     workflows (>= 1.0.0),
-    yardstick (>= 1.2.0)
+    yardstick (>= 1.2.0.9001)
 Suggests: 
     C50,
     censored,
@@ -47,10 +47,13 @@ Suggests:
     kknn,
     knitr,
     modeldata,
+    scales,
     spelling,
     testthat (>= 3.0.0),
     xgboost,
     xml2
+Remotes:
+    tidymodels/yardstick
 Config/Needs/website: pkgdown, tidymodels, kknn, doParallel, doFuture,
     tidyverse/tidytemplate
 Config/testthat/edition: 3

NAMESPACE

@@ -138,6 +138,7 @@ export(.config_key_from_metrics)
 export(.estimate_metrics)
 export(.get_extra_col_names)
 export(.get_fingerprint)
+export(.get_tune_eval_times)
 export(.get_tune_metric_names)
 export(.get_tune_metrics)
 export(.get_tune_outcome_names)

NEWS.md

@@ -18,6 +18,8 @@
 
 * Improves documentation related to the hyperparameters associated with extracted objects that are generated from submodels. See the "Extracting with submodels" section of `?collect_extracts` to learn more.
 
+* An `eval_time` attribute was added to tune objects. There is also a `.get_tune_eval_times()` function.
+
 * `augment()` methods to `tune_results`, `resample_results`, and `last_fit` objects now always returns tibbles (#759).
 
 # tune 1.1.2

R/compat-vctrs-helpers.R

@@ -16,6 +16,7 @@ new_tune_results_from_template <- function(x, to) {
     x = x,
     parameters = attrs$parameters,
     metrics = attrs$metrics,
+    eval_time = attrs$eval_time,
     outcomes = attrs$outcomes,
     rset_info = attrs$rset_info
   )
@@ -105,6 +106,7 @@ new_resample_results_from_template <- function(x, to) {
     x = x,
     parameters = attrs$parameters,
     metrics = attrs$metrics,
+    eval_time = attrs$eval_time,
     outcomes = attrs$outcomes,
     rset_info = attrs$rset_info
   )
@@ -134,6 +136,7 @@ new_iteration_results_from_template <- function(x, to) {
     x = x,
     parameters = attrs$parameters,
     metrics = attrs$metrics,
+    eval_time = attrs$eval_time,
     outcomes = attrs$outcomes,
     rset_info = attrs$rset_info,
     workflow = attrs$workflow

R/iteration_results.R

@@ -23,18 +23,27 @@
 #' @rdname empty_ellipses
 #' @param parameters A `parameters` object.
 #' @param metrics A metric set.
+#' @param eval_time A numeric vector of time points where dynamic event time
+#' metrics should be computed (e.g. the time-dependent ROC curve, etc).
 #' @param outcomes A character vector of outcome names.
 #' @param rset_info Attributes from an `rset` object.
 #' @param workflow The workflow used to fit the iteration results.
-new_iteration_results <- function(x, parameters, metrics, outcomes = character(0),
-                                  rset_info, workflow) {
-  new_tune_results(
-    x = x,
-    parameters = parameters,
-    metrics = metrics,
-    outcomes = outcomes,
-    rset_info = rset_info,
-    workflow = workflow,
-    class = "iteration_results"
-  )
-}
+new_iteration_results <-
+  function(x,
+           parameters,
+           metrics,
+           eval_time,
+           outcomes = character(0),
+           rset_info,
+           workflow) {
+    new_tune_results(
+      x = x,
+      parameters = parameters,
+      metrics = metrics,
+      eval_time = eval_time,
+      outcomes = outcomes,
+      rset_info = rset_info,
+      workflow = workflow,
+      class = "iteration_results"
+    )
+  }

R/resample.R

@@ -160,6 +160,7 @@ resample_workflow <- function(workflow, resamples, metrics, control,
     x = out,
     parameters = attributes$parameters,
     metrics = attributes$metrics,
+    eval_time = attributes$eval_time,
     outcomes = attributes$outcomes,
     rset_info = attributes$rset_info,
     workflow = attributes$workflow

R/resample_results.R

@@ -14,14 +14,22 @@
 
 # ------------------------------------------------------------------------------
 
-new_resample_results <- function(x, parameters, metrics, outcomes = character(0), rset_info, workflow = NULL) {
-  new_tune_results(
-    x = x,
-    parameters = parameters,
-    metrics = metrics,
-    outcomes = outcomes,
-    rset_info = rset_info,
-    workflow = workflow,
-    class = "resample_results"
-  )
-}
+new_resample_results <-
+  function(x,
+           parameters,
+           metrics,
+           eval_time,
+           outcomes = character(0),
+           rset_info,
+           workflow = NULL) {
+    new_tune_results(
+      x = x,
+      parameters = parameters,
+      metrics = metrics,
+      eval_time = eval_time,
+      outcomes = outcomes,
+      rset_info = rset_info,
+      workflow = workflow,
+      class = "resample_results"
+    )
+  }

R/tune_bayes.R

@@ -304,6 +304,7 @@ tune_bayes_workflow <-
         x = unsummarized,
         parameters = param_info,
         metrics = metrics,
+        eval_time = eval_time,
         outcomes = outcomes,
         rset_info = rset_info,
         workflow = NULL
@@ -476,6 +477,7 @@ tune_bayes_workflow <-
         x = unsummarized,
         parameters = param_info,
         metrics = metrics,
+        eval_time = eval_time,
         outcomes = outcomes,
         rset_info = rset_info,
         workflow = workflow_output

R/tune_grid.R

@@ -379,6 +379,7 @@ tune_grid_workflow <- function(workflow,
     x = resamples,
     parameters = pset,
     metrics = metrics,
+    eval_time = eval_time,
     outcomes = outcomes,
     rset_info = rset_info,
     workflow = workflow

R/tune_results.R

@@ -97,17 +97,26 @@ summarize_notes <- function(x) {
 
 # ------------------------------------------------------------------------------
 
-new_tune_results <- function(x, parameters, metrics, outcomes = character(0), rset_info, ..., class = character()) {
-  new_bare_tibble(
-    x = x,
-    parameters = parameters,
-    metrics = metrics,
-    outcomes = outcomes,
-    rset_info = rset_info,
-    ...,
-    class = c(class, "tune_results")
-  )
-}
+new_tune_results <-
+  function(x,
+           parameters,
+           metrics,
+           eval_time,
+           outcomes = character(0),
+           rset_info,
+           ...,
+           class = character()) {
+    new_bare_tibble(
+      x = x,
+      parameters = parameters,
+      metrics = metrics,
+      eval_time = eval_time,
+      outcomes = outcomes,
+      rset_info = rset_info,
+      ...,
+      class = c(class, "tune_results")
+    )
+  }
 
 is_tune_results <- function(x) {
   inherits(x, "tune_results")

R/utils.R

@@ -163,6 +163,20 @@ new_bare_tibble <- function(x, ..., class = character()) {
   res
 }
 
+
+#' @export
+#' @rdname tune_accessor
+.get_tune_eval_times <- function(x) {
+  x <- attributes(x)
+  if (any(names(x) == "eval_time")) {
+    res <- x$eval_time
+  } else {
+    res <- NULL
+  }
+  res
+}
+
+
 #' @export
 #' @rdname tune_accessor
 .get_tune_outcome_names <- function(x) {

inst/test_objects.R

@@ -1,4 +1,5 @@
 library(tidymodels)
+library(scales)
 library(sessioninfo)
 library(testthat)
 

tests/testthat/test-autoplot.R

@@ -133,6 +133,7 @@ test_that("parameter plot for iterative search", {
 
 
 test_that("regular grid plot", {
+  skip_if_not_installed("scales", "1.3.0")
   rcv_results <- readRDS(test_path("data", "rcv_results.rds"))
   svm_reg_results <- readRDS(test_path("data", "svm_reg_results.rds"))
 
@@ -166,6 +167,7 @@ test_that("regular grid plot", {
   expect_equal(p$labels$y, "rmse")
   expect_equal(p$labels$x, "deg_free")
 
+
   p <- autoplot(svm_reg_results)
   expect_s3_class(p, "ggplot")
   expect_equal(
@@ -183,7 +185,7 @@ test_that("regular grid plot", {
   expect_equal(p$labels$x, "Cost")
   expect_equal(p$labels$group, "%^*#")
 
-  expect_equal(class(p$scales$scales[[1]]$trans), "trans")
+  expect_true(grepl("^trans", class(p$scales$scales[[1]]$trans)))
   expect_equal(p$scales$scales[[1]]$trans$name, "log-2")
   expect_equal(unique(p$data$name), "Cost")
 })

tests/testthat/test-bayes.R

@@ -53,6 +53,7 @@ test_that("tune recipe only", {
   expect_equal(res_est$n, rep(10, iterT * 2))
   expect_false(identical(num_comp, expr(tune())))
   expect_true(res_workflow$trained)
+  expect_null(.get_tune_eval_times(res))
 
   set.seed(1)
   expect_error(

tests/testthat/test-grid.R

@@ -27,6 +27,7 @@ test_that("tune recipe only", {
   expect_equal(res_est$n, rep(10, nrow(grid) * 2))
   expect_false(identical(num_comp, expr(tune())))
   expect_true(res_workflow$trained)
+  expect_null(.get_tune_eval_times(res))
 })
 
 # ------------------------------------------------------------------------------

tests/testthat/test-last-fit.R

@@ -25,7 +25,7 @@ test_that("formula method", {
     nrow(predict(res$.workflow[[1]], rsample::testing(split))),
     nrow(rsample::testing(split))
   )
-
+  expect_null(.get_tune_eval_times(res))
 
 })
 

tests/testthat/test-resample.R

@@ -13,6 +13,8 @@ test_that("`fit_resamples()` returns a `resample_result` object", {
   expect_s3_class(result, "resample_results")
 
   expect_equal(result, .Last.tune.result)
+
+  expect_null(.get_tune_eval_times(result))
 })
 
 test_that("can use `fit_resamples()` with a formula", {