This step method for update() takes named arguments as ... who's values
will replace the elements of the same name in the actual step.
# S3 method for step
update(object, ...)| object | A recipe |
|---|---|
| ... | Key-value pairs where the keys match up with names of elements in the step, and the values are the new values to update the step with. |
For a step to be updated, it must not already have been trained. Otherwise,
conflicting information can arise between the data returned from
bake(object, new_data = NULL) and the information in the step.
library(modeldata)
data(biomass)
biomass_tr <- biomass[biomass$dataset == "Training",]
biomass_te <- biomass[biomass$dataset == "Testing",]
# Create a recipe using step_bs() with degree = 3
rec <- recipe(
HHV ~ carbon + hydrogen + oxygen + nitrogen + sulfur,
data = biomass_tr
) %>%
step_bs(carbon, hydrogen, degree = 3)
# Update the step to use degree = 4
rec2 <- rec
rec2$steps[[1]] <- update(rec2$steps[[1]], degree = 4)
# Prep both recipes
rec_prepped <- prep(rec, training = biomass_tr)
rec2_prepped <- prep(rec2, training = biomass_tr)
# Juice both to see what changed
bake(rec_prepped, new_data = NULL)
#> # A tibble: 456 × 10
#> oxygen nitrogen sulfur HHV carbon_bs_1 carbon_bs_2 carbon_bs_3
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 42.9 0.41 0 20.0 0.421 0.313 0.0775
#> 2 41.3 0.2 0 19.2 0.423 0.309 0.0754
#> 3 46.2 0.11 0.02 18.3 0.431 0.290 0.0651
#> 4 35.6 3.3 0.16 18.2 0.441 0.258 0.0504
#> 5 40.7 1 0.02 18.4 0.436 0.278 0.0590
#> 6 40.2 2.04 0.1 18.5 0.440 0.262 0.0519
#> 7 38.2 2.68 0.2 18.7 0.434 0.283 0.0613
#> 8 39.7 1.7 0.2 18.3 0.439 0.265 0.0534
#> 9 40.9 0.8 0 18.6 0.426 0.301 0.0710
#> 10 40 1.2 0.1 18.9 0.434 0.282 0.0609
#> # … with 446 more rows, and 3 more variables: hydrogen_bs_1 <dbl>,
#> # hydrogen_bs_2 <dbl>, hydrogen_bs_3 <dbl>
bake(rec2_prepped, new_data = NULL)
#> # A tibble: 456 × 12
#> oxygen nitrogen sulfur HHV carbon_bs_1 carbon_bs_2 carbon_bs_3 carbon_bs_4
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 42.9 0.41 0 20.0 0.322 0.359 0.178 0.0330
#> 2 41.3 0.2 0 19.2 0.325 0.357 0.174 0.0319
#> 3 46.2 0.11 0.02 18.3 0.344 0.347 0.156 0.0262
#> 4 35.6 3.3 0.16 18.2 0.371 0.325 0.127 0.0186
#> 5 40.7 1 0.02 18.4 0.355 0.339 0.144 0.0230
#> 6 40.2 2.04 0.1 18.5 0.368 0.328 0.130 0.0193
#> 7 38.2 2.68 0.2 18.7 0.350 0.342 0.149 0.0242
#> 8 39.7 1.7 0.2 18.3 0.365 0.331 0.133 0.0201
#> 9 40.9 0.8 0 18.6 0.333 0.353 0.166 0.0294
#> 10 40 1.2 0.1 18.9 0.351 0.342 0.148 0.0240
#> # … with 446 more rows, and 4 more variables: hydrogen_bs_1 <dbl>,
#> # hydrogen_bs_2 <dbl>, hydrogen_bs_3 <dbl>, hydrogen_bs_4 <dbl>
# Cannot update a recipe step that has been trained!
if (FALSE) {
update(rec_prepped$steps[[1]], degree = 4)
}