better call routing for errors (#1214)

* pass call through user-facing predict methods

* pass call for internal code; probably will never be surfaced by user

* show error is from autoplot() instead of map_glmnet_coefs()

* un-used bartMachine code

* fix bug in condense_control and route user-facing call

* unit tests for one-hot encodings

* pass calls through data conversion functions

* redoc

* small formatting changes

* route some glmnet checking calls

* route some spec updating calls

* some predict call routing

* make dev function as internal

* redoc

* revert passing in predict

* Apply suggestions from code review

Co-authored-by: Simon P. Couch <[email protected]>

* update snapshots

* redoc

---------

Co-authored-by: ‘topepo’ <‘[email protected]’>
Co-authored-by: Simon P. Couch <[email protected]>

Author: 3 people

NAMESPACE

@@ -202,7 +202,6 @@ export(bag_mars)
 export(bag_mlp)
 export(bag_tree)
 export(bart)
-export(bartMachine_interval_calc)
 export(boost_tree)
 export(case_weights_allowed)
 export(cforest_train)

R/arguments.R

@@ -258,7 +258,7 @@ make_form_call <- function(object, env = NULL) {
 }
 
 # TODO we need something to indicate that case weights are being used.
-make_xy_call <- function(object, target, env) {
+make_xy_call <- function(object, target, env, call = rlang::caller_env()) {
   fit_args <- object$method$fit$args
   uses_weights <- has_weights(env)
 
@@ -283,7 +283,7 @@ make_xy_call <- function(object, target, env) {
       data.frame = rlang::expr(maybe_data_frame(x)),
       matrix = rlang::expr(maybe_matrix(x)),
       dgCMatrix = rlang::expr(maybe_sparse_matrix(x)),
-      cli::cli_abort("Invalid data type target: {target}.")
+      cli::cli_abort("Invalid data type target: {target}.", call = call)
     )
   if (uses_weights) {
     object$method$fit$args[[ unname(data_args["weights"]) ]] <- rlang::expr(weights)

R/autoplot.R

@@ -41,14 +41,15 @@ autoplot.glmnet <- function(object, ..., min_penalty = 0, best_penalty = NULL,
 }
 
 
-map_glmnet_coefs <- function(x) {
+map_glmnet_coefs <- function(x, call = rlang::caller_env()) {
   coefs <- coef(x)
   # If parsnip is used to fit the model, glmnet should be attached and this will
   # work. If an object is loaded from a new session, they will need to load the
   # package.
   if (is.null(coefs)) {
     cli::cli_abort(
-      "Please load the {.pkg glmnet} package before running {.fun autoplot}."
+      "Please load the {.pkg glmnet} package before running {.fun autoplot}.",
+      call = call
     )
   }
   p <- x$dim[1]
@@ -89,9 +90,10 @@ top_coefs <- function(x, top_n = 5) {
     dplyr::slice(seq_len(top_n))
 }
 
-autoplot_glmnet <- function(x, min_penalty = 0, best_penalty = NULL, top_n = 3L, ...) {
+autoplot_glmnet <- function(x, min_penalty = 0, best_penalty = NULL, top_n = 3L,
+                            call = rlang::caller_env(), ...) {
   tidy_coefs <-
-    map_glmnet_coefs(x) %>%
+    map_glmnet_coefs(x, call = call) %>%
     dplyr::filter(penalty >= min_penalty)
 
   actual_min_penalty <- min(tidy_coefs$penalty)

R/bart.R

@@ -130,61 +130,13 @@ update.bart <-
     )
   }
 
-
 #' Developer functions for predictions via BART models
-#' @export
-#' @keywords internal
 #' @name bart-internal
 #' @inheritParams predict.model_fit
 #' @param obj A parsnip object.
-#' @param ci Confidence (TRUE) or prediction interval (FALSE)
 #' @param level Confidence level.
 #' @param std_err Attach column for standard error of prediction or not.
-bartMachine_interval_calc <- function(new_data, obj, ci = TRUE, level = 0.95) {
-  if (obj$spec$mode == "classification") {
-    cli::cli_abort(
-      "Prediction intervals are not possible for classification"
-    )
-  }
-  get_std_err <- obj$spec$method$pred$pred_int$extras$std_error
-
-  if (ci) {
-    cl <-
-      rlang::call2(
-        "calc_credible_intervals",
-        .ns = "bartMachine",
-        bart_machine = rlang::expr(obj$fit),
-        new_data = rlang::expr(new_data),
-        ci_conf = level
-      )
-
-  } else {
-    cl <-
-      rlang::call2(
-        "calc_prediction_intervals",
-        .ns = "bartMachine",
-        bart_machine = rlang::expr(obj$fit),
-        new_data = rlang::expr(new_data),
-        pi_conf = level
-      )
-  }
-  res <- rlang::eval_tidy(cl)
-  if (!ci) {
-    if (get_std_err) {
-      .std_error <- apply(res$all_prediction_samples, 1, stats::sd, na.rm = TRUE)
-    }
-    res <- res$interval
-  }
-  res <- tibble::as_tibble(res)
-  names(res) <- c(".pred_lower", ".pred_upper")
-  if (!ci & get_std_err) {
-    res$.std_err <- .std_error
-  }
-  res
-}
-
 #' @export
-#' @rdname bart-internal
 #' @keywords internal
 dbart_predict_calc <- function(obj, new_data, type, level = 0.95, std_err = FALSE) {
   types <- c("numeric", "class", "prob", "conf_int", "pred_int")

R/condense_control.R

@@ -10,6 +10,10 @@
 #'
 #' @return A control object with the same elements and classes of `ref`, with
 #'   values of `x`.
+#' @param call The execution environment of a currently running function, e.g.
+#'   `caller_env()`. The function will be mentioned in error messages as the
+#'   source of the error. See the call argument of [rlang::abort()] for more
+#'   information.
 #' @keywords internal
 #' @export
 #'
@@ -20,16 +24,17 @@
 #'
 #' ctrl <- condense_control(ctrl, control_parsnip())
 #' str(ctrl)
-condense_control <- function(x, ref) {
+condense_control <- function(x, ref, ..., call = rlang::caller_env()) {
+  check_dots_empty()
   mismatch <- setdiff(names(ref), names(x))
   if (length(mismatch)) {
     cli::cli_abort(
       c(
-        "Object of class {.cls class(x)[1]} cannot be coerced to
-         object of class {.cls class(ref)[1]}.",
+        "{.obj_type_friendly {x}} cannot be coerced to {.obj_type_friendly {ref}}.",
         "i" = "{cli::qty(mismatch)} The argument{?s} {.arg {mismatch}}
                {?is/are} missing."
-      )
+      ),
+      call = call
     )
   }
   res <- x[names(ref)]

R/contr_one_hot.R

@@ -3,7 +3,8 @@
 #' This contrast function produces a model matrix with indicator columns for
 #' each level of each factor.
 #'
-#' @param n A vector of character factor levels or the number of unique levels.
+#' @param n A vector of character factor levels (of length >=1) or the number
+#' of unique levels (>= 1).
 #' @param contrasts This argument is for backwards compatibility and only the
 #'   default of `TRUE` is supported.
 #' @param sparse This argument is for backwards compatibility and only the
@@ -24,9 +25,13 @@ contr_one_hot <- function(n, contrasts = TRUE, sparse = FALSE) {
   }
 
   if (is.character(n)) {
+    if (length(n) < 1) {
+      cli::cli_abort("{.arg n} cannot be empty.")
+    }
     names <- n
     n <- length(names)
   } else if (is.numeric(n)) {
+    check_number_whole(n, min = 1)
     n <- as.integer(n)
 
     if (length(n) != 1L) {
@@ -35,7 +40,7 @@ contr_one_hot <- function(n, contrasts = TRUE, sparse = FALSE) {
 
     names <- as.character(seq_len(n))
   } else {
-    cli::cli_abort("{.arg n} must be a character vector or an integer of size 1.")
+    check_number_whole(n, min = 1)
   }
 
   out <- diag(n)

R/convert_data.R

@@ -40,18 +40,21 @@
                                     na.action = na.omit,
                                     indicators = "traditional",
                                     composition = "data.frame",
-                                    remove_intercept = TRUE) {
+                                    remove_intercept = TRUE,
+                                    call = rlang::caller_env()) {
   if (!(composition %in% c("data.frame", "matrix", "dgCMatrix"))) {
     cli::cli_abort(
       "{.arg composition} should be either {.val data.frame}, {.val matrix}, or
-      {.val dgCMatrix}."
+      {.val dgCMatrix}.",
+      call = call
     )
   }
 
   if (sparsevctrs::has_sparse_elements(data)) {
     cli::cli_abort(
-      "Sparse data cannot be used with formula interface. Please use 
-      {.fn fit_xy} instead."
+      "Sparse data cannot be used with formula interface. Please use
+      {.fn fit_xy} instead.",
+      call = call
     )
   }
 
@@ -84,7 +87,7 @@
 
   w <- as.vector(model.weights(mod_frame))
   if (!is.null(w) && !is.numeric(w)) {
-    cli::cli_abort("{.arg weights} must be a numeric vector.")
+    cli::cli_abort("{.arg weights} must be a numeric vector.", call = call)
   }
 
   # TODO: Do we actually use the offset when fitting?
@@ -175,10 +178,12 @@
 .convert_form_to_xy_new <- function(object,
                                     new_data,
                                     na.action = na.pass,
-                                    composition = "data.frame") {
+                                    composition = "data.frame",
+                                    call = rlang::caller_env()) {
   if (!(composition %in% c("data.frame", "matrix"))) {
     cli::cli_abort(
-      "{.arg composition} should be either {.val data.frame} or {.val matrix}."
+      "{.arg composition} should be either {.val data.frame} or {.val matrix}.",
+      call = call
     )
   }
 
@@ -244,9 +249,10 @@
                                     y,
                                     weights = NULL,
                                     y_name = "..y",
-                                    remove_intercept = TRUE) {
+                                    remove_intercept = TRUE,
+                                    call = rlang::caller_env()) {
   if (is.vector(x)) {
-    cli::cli_abort("{.arg x} cannot be a vector.")
+    cli::cli_abort("{.arg x} cannot be a vector.", call = call)
   }
 
   if (remove_intercept) {
@@ -279,10 +285,10 @@
 
   if (!is.null(weights)) {
     if (!is.numeric(weights)) {
-      cli::cli_abort("{.arg weights} must be a numeric vector.")
+      cli::cli_abort("{.arg weights} must be a numeric vector.", call = call)
     }
     if (length(weights) != nrow(x)) {
-      cli::cli_abort("{.arg weights} should have {nrow(x)} elements.")
+      cli::cli_abort("{.arg weights} should have {nrow(x)} elements.", call = call)
     }
 
     form <- patch_formula_environment_with_case_weights(

R/descriptors.R

@@ -103,22 +103,23 @@ NULL
 
 # Descriptor retrievers --------------------------------------------------------
 
-get_descr_form <- function(formula, data) {
+get_descr_form <- function(formula, data, call = rlang::caller_env()) {
   if (inherits(data, "tbl_spark")) {
     res <- get_descr_spark(formula, data)
   } else {
-    res <- get_descr_df(formula, data)
+    res <- get_descr_df(formula, data, call = call)
   }
   res
 }
 
-get_descr_df <- function(formula, data) {
+get_descr_df <- function(formula, data, call = rlang::caller_env()) {
 
   tmp_dat <-
     .convert_form_to_xy_fit(formula,
                             data,
                             indicators = "none",
-                            remove_intercept = TRUE)
+                            remove_intercept = TRUE,
+                            call = call)
 
   if(is.factor(tmp_dat$y)) {
     .lvls <- function() {
@@ -136,7 +137,8 @@ get_descr_df <- function(formula, data) {
         formula,
         data,
         indicators = "traditional",
-        remove_intercept = TRUE
+        remove_intercept = TRUE,
+        call = call
       )$x
     )
   }
@@ -263,7 +265,7 @@ get_descr_spark <- function(formula, data) {
   )
 }
 
-get_descr_xy <- function(x, y) {
+get_descr_xy <- function(x, y, call = rlang::caller_env()) {
 
   .lvls <- if (is.factor(y)) {
     function() table(y, dnn = NULL)
@@ -291,7 +293,7 @@ get_descr_xy <- function(x, y) {
   }
 
   .dat <- function() {
-    .convert_xy_to_form_fit(x, y, remove_intercept = TRUE)$data
+    .convert_xy_to_form_fit(x, y, remove_intercept = TRUE, call = call)$data
   }
 
   .x <- function() {

R/fit.R

@@ -157,7 +157,7 @@ fit.model_spec <-
     }
 
     if (all(c("x", "y") %in% names(dots))) {
-      cli::cli_abort("`fit.model_spec()` is for the formula methods. Use `fit_xy()` instead.")
+      cli::cli_abort("{.fn fit.model_spec} is for the formula methods. Use {.fn fit_xy} instead.")
     }
     cl <- match.call(expand.dots = TRUE)
     # Create an environment with the evaluated argument objects. This will be
@@ -307,7 +307,8 @@ fit_xy.model_spec <-
 
     if (object$engine == "spark") {
       cli::cli_abort(
-        "spark objects can only be used with the formula interface to {.fn fit} with a spark data object."
+        "spark objects can only be used with the formula interface to {.fn fit}
+        with a spark data object."
       )
     }
 

R/fit_helpers.R

@@ -27,7 +27,7 @@ form_form <-
 
     # if descriptors are needed, update descr_env with the calculated values
     if (requires_descrs(object)) {
-      data_stats <- get_descr_form(env$formula, env$data)
+      data_stats <- get_descr_form(env$formula, env$data, call = call)
       scoped_descrs(data_stats)
     }
 
@@ -86,7 +86,7 @@ xy_xy <- function(object,
 
   # if descriptors are needed, update descr_env with the calculated values
   if (requires_descrs(object)) {
-    data_stats <- get_descr_xy(env$x, env$y)
+    data_stats <- get_descr_xy(env$x, env$y, call = call)
     scoped_descrs(data_stats)
   }
 
@@ -96,7 +96,7 @@ xy_xy <- function(object,
   # sub in arguments to actual syntax for corresponding engine
   object <- translate(object, engine = object$engine)
 
-  fit_call <- make_xy_call(object, target, env)
+  fit_call <- make_xy_call(object, target, env, call)
 
   res <- list(lvl = levels(env$y), spec = object)
 
@@ -141,7 +141,8 @@ form_xy <- function(object, control, env,
     ...,
     composition = target,
     indicators = indicators,
-    remove_intercept = remove_intercept
+    remove_intercept = remove_intercept,
+    call = call
   )
   env$x <- data_obj$x
   env$y <- data_obj$y