What it says in the title. A function to collect all the errors in a fast and easy way.
basic prototype
collect_notes <- function(x) {
x$.notes %>%
bind_rows()
}library(tidymodels)
set.seed(6735)
mtcars1 <- mtcars
mtcars1$disp <- "no value"
folds <- vfold_cv(mtcars1, v = 5)
spline_rec <- recipe(mpg ~ ., data = mtcars1) %>%
step_ns(disp) %>%
step_ns(wt)
lin_mod <- linear_reg() %>%
set_engine("lm")
control <- control_resamples(save_pred = TRUE)
spline_res <- fit_resamples(lin_mod, spline_rec, folds, control = control)
#> x Fold1: preprocessor 1/1: Error: All columns selected for the step should be num...
#> x Fold2: preprocessor 1/1: Error: All columns selected for the step should be num...
#> x Fold3: preprocessor 1/1: Error: All columns selected for the step should be num...
#> x Fold4: preprocessor 1/1: Error: All columns selected for the step should be num...
#> x Fold5: preprocessor 1/1: Error: All columns selected for the step should be num...
#> Warning: All models failed. See the `.notes` column.
collect_notes <- function(x) {
x$.notes %>%
bind_rows()
}
spline_res %>%
collect_notes()
#> # A tibble: 5 x 1
#> .notes
#> <chr>
#> 1 preprocessor 1/1: Error: All columns selected for the step should be numeric
#> 2 preprocessor 1/1: Error: All columns selected for the step should be numeric
#> 3 preprocessor 1/1: Error: All columns selected for the step should be numeric
#> 4 preprocessor 1/1: Error: All columns selected for the step should be numeric
#> 5 preprocessor 1/1: Error: All columns selected for the step should be numericCreated on 2021-03-28 by the reprex package (v0.3.0)
What it says in the title. A function to collect all the errors in a fast and easy way.
basic prototype
Created on 2021-03-28 by the reprex package (v0.3.0)