Merge pull request #113 from tidymodels/rc-0.1.0

Rc 0.1.0

Author: DavisVaughan

.Rbuildignore

@@ -8,5 +8,9 @@
 ^\.Rproj\.user$
 ^man-roxygen$
 ^graphics$
+^\.github$
 ^\.github/workflows/R-CMD-check\.yaml$
 ^\.github/workflows/pr-commands\.yaml$
+^LICENSE\.md$
+^cran-comments\.md$
+^CRAN-RELEASE$

CRAN-RELEASE

@@ -0,0 +1,2 @@
+This package was submitted to CRAN on 2019-12-06.
+Once it is accepted, delete this file and tag the release (commit 34ea0a04ea).

DESCRIPTION

@@ -1,18 +1,23 @@
 Package: hardhat
-Title: A Toolkit for the Construction of Modeling Packages
-Version: 0.0.0.9001
+Title: Construct Modeling Packages
+Version: 0.1.0
 Authors@R: c(
     person("Davis", "Vaughan", , "[email protected]", c("aut", "cre")),
     person("Max", "Kuhn", , "[email protected]", c("aut")),
     person("RStudio", role = "cph"))
 Maintainer: Davis Vaughan <[email protected]>
-Description: Provides infrastructure for building new modeling packages, 
-    including functionality around preprocessing, predicting, and validating 
-    input.
-License: GPL-2
+Description: Building modeling packages is hard. A large amount of effort
+    generally goes into providing an implementation for a new method that is
+    efficient, fast, and correct, but often less emphasis is put on the user
+    interface. A good interface requires specialized knowledge about S3 methods
+    and formulas, which the average package developer might not have.
+    The goal of 'hardhat' is to reduce the burden around building new modeling
+    packages by providing functionality for preprocessing, predicting, and
+    validating input.
+License: MIT + file LICENSE
 Encoding: UTF-8
 LazyData: true
-RoxygenNote: 6.1.1
+RoxygenNote: 7.0.1
 Roxygen: list(markdown = TRUE)
 Suggests: 
     recipes,
@@ -25,9 +30,8 @@ Suggests:
     devtools
 Imports: 
     glue,
-    rlang (>= 0.4.0),
+    rlang (>= 0.4.1),
     tibble,
-    zeallot,
     vctrs (>= 0.2.0)
 URL: https://github.com/tidymodels/hardhat
 BugReports: https://github.com/tidymodels/hardhat/issues

LICENSE

@@ -0,0 +1,2 @@
+YEAR: 2019
+COPYRIGHT HOLDER: Davis Vaughan

LICENSE.md

@@ -0,0 +1,21 @@
+# MIT License
+
+Copyright (c) 2019 Davis Vaughan
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

NAMESPACE

@@ -106,4 +106,3 @@ importFrom(vctrs,vec_cast.integer)
 importFrom(vctrs,vec_cast.list)
 importFrom(vctrs,vec_cast.logical)
 importFrom(vctrs,vec_cast.vctrs_list_of)
-importFrom(zeallot,"%<-%")

R/aaa.R

@@ -1,7 +1,6 @@
 #' @importFrom rlang abort is_missing arg_match := %||%
 #' @importFrom glue glue
 #' @importFrom tibble tibble
-#' @importFrom zeallot %<-%
 NULL
 
 # stats related imports
@@ -28,18 +27,3 @@ NULL
 #' @importFrom vctrs vec_cast.POSIXlt
 #' @importFrom vctrs vec_cast.vctrs_list_of
 NULL
-
-# for zeallot
-utils::globalVariables(
-  c(
-    ".predictors",
-    ".outcomes",
-    "predictors",
-    "outcomes",
-    "ptypes",
-    "ptype",
-    "extras",
-    "predictors_lst",
-    "outcomes_lst"
-  )
-)

R/blueprint-formula-default.R

@@ -14,6 +14,10 @@
 #'
 #' @param ... Not used.
 #'
+#' @return
+#'
+#' For `default_formula_blueprint()`, a formula blueprint.
+#'
 #' @details
 #'
 #' While not different from base R, the behavior of expanding factors into
@@ -89,7 +93,7 @@
 #'    to [mold()], then they are extracted with [model_offset()].
 #'
 #'    - If `intercept = TRUE` in the original call to [mold()], then an
-#'    intecept column is added.
+#'    intercept column is added.
 #'
 #'    - It coerces the result of the above steps to a tibble.
 #'
@@ -198,10 +202,8 @@
 #'
 #' # An informative error is thrown when `indicators = FALSE` and
 #' # factors are present in interaction terms or in inline functions
-#' \dontrun{
-#' mold(Sepal.Width ~ Sepal.Length:Species, train, blueprint = blueprint_no_indicators)
-#' mold(Sepal.Width ~ paste0(Species), train, blueprint = blueprint_no_indicators)
-#' }
+#' try(mold(Sepal.Width ~ Sepal.Length:Species, train, blueprint = blueprint_no_indicators))
+#' try(mold(Sepal.Width ~ paste0(Species), train, blueprint = blueprint_no_indicators))
 #'
 #' # ---------------------------------------------------------------------------
 #' # Multivariate outcomes
@@ -343,17 +345,22 @@ mold_formula_default_clean <- function(blueprint, data) {
 
 # mold - formula - process
 mold_formula_default_process <- function(blueprint, data) {
-
-  c(blueprint, predictors_lst) %<-% mold_formula_default_process_predictors(
+  processed <- mold_formula_default_process_predictors(
     blueprint = blueprint,
     data = data
   )
 
-  c(blueprint, outcomes_lst) %<-% mold_formula_default_process_outcomes(
+  blueprint <- processed$blueprint
+  predictors_lst <- processed$terms_lst
+
+  processed <- mold_formula_default_process_outcomes(
     blueprint = blueprint,
     data = data
   )
 
+  blueprint <- processed$blueprint
+  outcomes_lst <- processed$terms_lst
+
   # nuke formula environment before returning
   formula_empty_env <- nuke_formula_environment(blueprint$formula)
   blueprint <- update_blueprint(blueprint, formula = formula_empty_env)
@@ -470,17 +477,22 @@ forge_formula_default_clean <- function(blueprint, new_data, outcomes) {
 }
 
 forge_formula_default_process <- function(blueprint, predictors, outcomes, extras) {
-
-  c(blueprint, predictors_lst) %<-% forge_formula_default_process_predictors(
+  processed <- forge_formula_default_process_predictors(
     blueprint = blueprint,
     predictors = predictors
   )
 
-  c(blueprint, outcomes_lst) %<-% forge_formula_default_process_outcomes(
+  blueprint <- processed$blueprint
+  predictors_lst <- processed$terms_lst
+
+  processed <- forge_formula_default_process_outcomes(
     blueprint = blueprint,
     outcomes = outcomes
   )
 
+  blueprint <- processed$blueprint
+  outcomes_lst <- processed$terms_lst
+
   extras <- c(
     extras,
     out$extras$final(predictors_lst$extras, outcomes_lst$extras)

R/blueprint-recipe-default.R

@@ -14,6 +14,10 @@
 #'
 #' @param ... Not used.
 #'
+#' @return
+#'
+#' For `default_recipe_blueprint()`, a recipe blueprint.
+#'
 #' @section Mold:
 #'
 #' When `mold()` is used with the default recipe blueprint:
@@ -182,17 +186,20 @@ mold_recipe_default_process <- function(blueprint, data) {
   recipe <- recipes::prep(blueprint$recipe, training = data, fresh = blueprint$fresh)
   blueprint <- update_blueprint(blueprint, recipe = recipe)
 
-  c(blueprint, predictors_lst) %<-% mold_recipe_default_process_predictors(
-    blueprint = blueprint,
-    data = data
-  )
+  processed <- mold_recipe_default_process_predictors(blueprint = blueprint, data = data)
 
-  c(blueprint, outcomes_lst) %<-% mold_recipe_default_process_outcomes(
-    blueprint = blueprint,
-    data = data
-  )
+  blueprint <- processed$blueprint
+  predictors_lst <- processed$terms_lst
 
-  c(blueprint, extras) %<-% mold_recipe_default_process_extras(blueprint, data)
+  processed <- mold_recipe_default_process_outcomes(blueprint = blueprint, data = data)
+
+  blueprint <- processed$blueprint
+  outcomes_lst <- processed$terms_lst
+
+  processed <- mold_recipe_default_process_extras(blueprint, data)
+
+  blueprint <- processed$blueprint
+  extras <- processed$extras
 
   extras <- c(
     extras,
@@ -344,16 +351,22 @@ forge_recipe_default_process <- function(blueprint, predictors, outcomes, extras
     outcomes <- baked_data[, processed_outcome_names, drop = FALSE]
   }
 
-  c(blueprint, predictors_lst) %<-% forge_recipe_default_process_predictors(
+  processed <- forge_recipe_default_process_predictors(
     blueprint = blueprint,
     predictors = predictors
   )
 
-  c(blueprint, outcomes_lst) %<-% forge_recipe_default_process_outcomes(
+  blueprint <- processed$blueprint
+  predictors_lst <- processed$terms_lst
+
+  processed <- forge_recipe_default_process_outcomes(
     blueprint = blueprint,
     outcomes = outcomes
   )
 
+  blueprint <- processed$blueprint
+  outcomes_lst <- processed$terms_lst
+
   extras <- forge_recipe_default_process_extras(
     extras,
     rec,

R/blueprint-xy-default.R

@@ -15,6 +15,10 @@
 #'
 #' @param ... Not used.
 #'
+#' @return
+#'
+#' For `default_xy_blueprint()`, an XY blueprint.
+#'
 #' @details
 #'
 #' As documented in [standardize()], if `y` is a _vector_, then the returned
@@ -88,9 +92,7 @@
 #' processed <- mold(train_x, train_y)
 #'
 #' # Can't do this!
-#' \dontrun{
-#' forge(test_x, processed$blueprint, outcomes = TRUE)
-#' }
+#' try(forge(test_x, processed$blueprint, outcomes = TRUE))
 #'
 #' # Need to use the full test set, including `y`
 #' forge(test, processed$blueprint, outcomes = TRUE)
@@ -106,9 +108,7 @@
 #'
 #' # This throws an informative error that tell you
 #' # to include an `".outcome"` column in `new_data`.
-#' \dontrun{
-#' forge(iris, processed_vec$blueprint, outcomes = TRUE)
-#' }
+#' try(forge(iris, processed_vec$blueprint, outcomes = TRUE))
 #'
 #' test2 <- test
 #' test2$.outcome <- test2$Species
@@ -178,14 +178,20 @@ get_mold_xy_default_function_set <- function() {
 
 # mold - xy - clean
 mold_xy_default_clean <- function(blueprint, x, y) {
-  c(blueprint, x) %<-% mold_xy_default_clean_predictors(blueprint, x)
+  cleaned <- mold_xy_default_clean_predictors(blueprint, x)
+
+  blueprint <- cleaned$blueprint
+  x <- cleaned$x
 
   # Special case `y = NULL` as a 0 column variation on `x`
   if (is.null(y)) {
     y <- x[, 0L, drop = FALSE]
   }
 
-  c(blueprint, y) %<-% mold_xy_default_clean_outcomes(blueprint, y)
+  cleaned <- mold_xy_default_clean_outcomes(blueprint, y)
+
+  blueprint <- cleaned$blueprint
+  y <- cleaned$y
 
   out$mold$clean_xy(blueprint, x, y)
 }
@@ -202,9 +208,15 @@ mold_xy_default_clean_outcomes <- function(blueprint, y) {
 
 # mold - xy - process
 mold_xy_default_process <- function(blueprint, x, y) {
+  processed <- mold_xy_default_process_predictors(blueprint, x)
 
-  c(blueprint, predictors_lst) %<-% mold_xy_default_process_predictors(blueprint, x)
-  c(blueprint, outcomes_lst) %<-% mold_xy_default_process_outcomes(blueprint, y)
+  blueprint <- processed$blueprint
+  predictors_lst <- processed$terms_lst
+
+  processed <- mold_xy_default_process_outcomes(blueprint, y)
+
+  blueprint <- processed$blueprint
+  outcomes_lst <- processed$terms_lst
 
   ptypes <- out$ptypes$final(predictors_lst$ptype, outcomes_lst$ptype)
   extras <- out$extras$final(predictors_lst$extras, outcomes_lst$extras)
@@ -266,9 +278,15 @@ forge_xy_default_clean <- function(blueprint, new_data, outcomes) {
 }
 
 forge_xy_default_process <- function(blueprint, predictors, outcomes, extras) {
+  processed <- forge_xy_default_process_predictors(blueprint, predictors)
+
+  blueprint <- processed$blueprint
+  predictors_lst <- processed$terms_lst
+
+  processed <- forge_xy_default_process_outcomes(blueprint, outcomes)
 
-  c(blueprint, predictors_lst) %<-% forge_xy_default_process_predictors(blueprint, predictors)
-  c(blueprint, outcomes_lst) %<-% forge_xy_default_process_outcomes(blueprint, outcomes)
+  blueprint <- processed$blueprint
+  outcomes_lst <- processed$terms_lst
 
   extras <- c(
     extras,