rebuild docs, checks pass

Author: dajmcdon

DESCRIPTION

@@ -14,7 +14,6 @@ Depends:
     R (>= 3.5.0)
 Imports: 
     assertthat,
-    broom,
     cli,
     dplyr,
     epiprocess,

NAMESPACE

@@ -40,6 +40,9 @@ export(knn_iteraive_ar_args_list)
 export(knn_iteraive_ar_forecaster)
 export(knnarx_args_list)
 export(knnarx_forecaster)
+export(layer_naomit)
+export(layer_predict)
+export(slather)
 export(smooth_arx_args_list)
 export(smooth_arx_forecaster)
 export(step_epi_ahead)

R/epi_recipe.R

@@ -23,8 +23,7 @@ epi_recipe.default <- function(x, ...) {
 }
 
 #' @rdname epi_recipe
-#' @param vars A character string of column names corresponding to variables
-#'   that will be used in any context (see below)
+#' @inheritParams recipes::recipe
 #' @param roles A character string (the same length of `vars`) that
 #'   describes a single role that the variable will take. This value could be
 #'   anything but common roles are `"outcome"`, `"predictor"`,
@@ -38,19 +37,7 @@ epi_recipe.default <- function(x, ...) {
 #'  `cbind`; see Examples).
 #' @param x,data A data frame, tibble, or epi_df of the *template* data set
 #'   (see below). This is always coerced to the first row to avoid memory issues
-#' @return An object of class `recipe` with sub-objects:
-#'   \item{var_info}{A tibble containing information about the original data
-#'   set columns}
-#'   \item{term_info}{A tibble that contains the current set of terms in the
-#'   data set. This initially defaults to the same data contained in
-#'   `var_info`.}
-#'   \item{steps}{A list of `step`  or `check` objects that define the sequence of
-#'   preprocessing operations that will be applied to data. The default value is
-#'   `NULL`}
-#'   \item{template}{A tibble of the data. This is initialized to be the same
-#'   as the data given in the `data` argument but can be different after
-#'   the recipe is trained.}
-#'
+#' @inherit recipes::recipe return
 #'
 #' @export
 #' @examples

R/epi_workflow.R

@@ -9,6 +9,8 @@
 #' and numerous examples, see there.
 #'
 #' @inheritParams workflows::workflow
+#' @param postprocessor An optional postprocessor to add to the workflow.
+#'   Currently only `frosting` is allowed using, `add_frosting()`.
 #'
 #' @return A new `epi_workflow` object.
 #' @seealso workflows::workflow
@@ -81,7 +83,6 @@ is_epi_workflow <- function(x) {
 #'   `geo_value` columns as well as the prediction.
 #'
 #' @inheritParams parsnip::predict.model_fit
-#' @param forecast_date The date on which the forecast is (was) made.
 #'
 #' @param object An epi_workflow that has been fit by
 #'   [workflows::fit.workflow()]
@@ -113,7 +114,7 @@ is_epi_workflow <- function(x) {
 #'
 #' wf <- epi_workflow(r, linear_reg()) %>% fit(jhu)
 #'
-#' latest <- get_test_data(r, jhu)
+#' latest <- jhu %>% filter(time_value >= max(time_value) - 14)
 #'
 #' preds <- predict(wf, latest) %>%
 #'   filter(!is.na(.pred))

R/frosting.R

@@ -58,11 +58,22 @@ frosting <- function(layers = NULL, requirements = NULL) {
   out <- new_frosting()
 }
 
+#' Apply postprocessing to a fitted workflow
+#'
+#' This function is intended for internal use. It implements postprocessing
+#' inside of the `predict()` method for a fitted workflow.
+#'
+#' @param workflow An object of class workflow
+#' @param ... additional arguments passed on to methods
+#'
+#' @aliases apply_frosting.default apply_frosting.epi_recipe
 #' @export
 apply_frosting <- function(workflow, ...) {
   UseMethod("apply_frosting")
 }
 
+#' @inheritParams slather
+#' @rdname apply_frosting
 #' @export
 apply_frosting.default <- function(workflow, components, ...) {
   if (has_postprocessor(workflow)) {
@@ -74,6 +85,7 @@ apply_frosting.default <- function(workflow, components, ...) {
 
 
 
+#' @rdname apply_frosting
 #' @importFrom rlang is_null
 #' @importFrom rlang abort
 #' @export
@@ -113,7 +125,32 @@ add_layer <- function(frosting, object) {
   frosting
 }
 
-slather <- function(x, ...) {
+
+#' Spread a layer of frosting on a fitted workflow
+#'
+#' Slathering frosting means to implement a postprocessing layer. When
+#' creating a new postprocessing layer, you must implement an S3 method
+#' for this function
+#'
+#' @param object a workflow with `frosting` postprocessing steps
+#' @param components a list of components containing model information. These
+#'   will be updated and returned by the layer. These should be
+#'   * `mold` - the output of calling `hardhat::mold()` on the workflow. This
+#'     contains information about the preprocessing, including the recipe.
+#'   * `forged` - the output of calling `hardhat::forge()` on the workflow.
+#'     This should have predictors and outcomes for the `new_data`. It will
+#'     have three components `predictors`, `outcomes` (if these were in the
+#'     `new_data`), and `extras` (usually has the rest of the data, including
+#'     `keys`).
+#'   * `keys` - we put the keys (`time_value`, `geo_value`, and any others)
+#'     here for ease.
+#' @param the_fit the fitted model object as returned by calling `parsnip::fit()`
+#'
+#' @param ... additional arguments used by methods. Currently unused.
+#'
+#' @return The `components` list. In the same format after applying any updates.
+#' @export
+slather <- function(object, components, the_fit, ...) {
   UseMethod("slather")
 }
 

R/layer_naomit.R

@@ -1,3 +1,15 @@
+#' Omit `NA`s from predictions or other columns
+#'
+#' @param frosting a `frosting` postprocessor
+#' @param ... <[`tidy-select`][dplyr::dplyr_tidy_select]> One or more unquoted
+#'   expressions separated by commas. Variable names can be used as if they
+#'   were positions in the data frame, so expressions like `x:y` can
+#'   be used to select a range of variables. Typical usage is `.pred` to remove
+#'   any rows with `NA` predictions.
+#' @param id a random id string
+#'
+#' @return an updated `frosting` postprocessor
+#' @export
 layer_naomit <- function(frosting, ..., id = rand_id("naomit")) {
   add_layer(
     frosting,
@@ -13,7 +25,7 @@ layer_naomit_new <- function(terms, id) {
 }
 
 #' @export
-slather.layer_naomit <- function(object, components, the_fit) {
+slather.layer_naomit <- function(object, components, the_fit,...) {
   exprs <- rlang::expr(c(!!!object$terms))
   pos <- tidyselect::eval_select(exprs, components$predictions)
   col_names <- names(pos)

R/layer_predict.R

@@ -1,3 +1,18 @@
+#' Prediction layer for postprocessing
+#'
+#' Implements prediction on a fitted `epi_workflow`. One may want different
+#' types of prediction, and to potentially apply this after some amount of
+#' postprocessing. This would typically be the first layer in a `frosting`
+#' postprocessor.
+#'
+#' @seealso `parsnip::predict.model_fit()`
+#'
+#' @inheritParams parsnip::predict.model_fit
+#' @param frosting a frosting object
+#' @param id a string identifying the layer
+#'
+#' @return An updated `frosting` object
+#' @export
 layer_predict <-
   function(frosting, type = NULL, opts = list(), ..., id = rand_id("predict_default")) {
     add_layer(
@@ -17,7 +32,7 @@ layer_predict_new <- function(type, opts, dots_list, id) {
 }
 
 #' @export
-slather.layer_predict <- function(object, components, the_fit) {
+slather.layer_predict <- function(object, components, the_fit, ...) {
 
   components$predictions <- predict(the_fit, components$forged$predictors,
                                     type = object$type, opts = object$opts)