#' Boilerplate Workflow
#'
#' @family Boiler_Plate
#' @family cubist
#'
#' @author Steven P. Sanderson II, MPH
#'
#' @details
#' This uses the `parsnip::multinom_reg()` with the `engine` set to `cubist`
#'
#' @description This is a boilerplate function to create automatically the following:
#' - recipe
#' - model specification
#' - workflow
#' - tuned model (grid ect)
#'
#' @param .data The data being passed to the function. The time-series object.
#' @param .rec_obj This is the recipe object you want to use. You can use
#' `hai_cubist_data_prepper()` an automatic recipe_object.
#' @param .splits_obj NULL is the default, when NULL then one will be created.
#' @param .rsamp_obj NULL is the default, when NULL then one will be created. It
#' will default to creating an [rsample::mc_cv()] object.
#' @param .tune Default is TRUE, this will create a tuning grid and tuned workflow
#' @param .grid_size Default is 10
#' @param .num_cores Default is 1
#' @param .best_metric Default is "rmse". The only `.model_type` you can use with
#' `Cubist` is `regression` so use [healthyR.ai::hai_default_regression_metric_set()]
#' to get the available metrics. Because of this the `.model_type` parameter is
#' omitted from this function.
#'
#' @examples
#' \dontrun{
#' data <- mtcars
#'
#' rec_obj <- hai_cubist_data_prepper(data, mpg ~ .)
#'
#' auto_glm <- hai_auto_cubist(
#' .data = data,
#' .rec_obj = rec_obj,
#' .best_metric = "rmse"
#' )
#'
#' auto_glm$recipe_info
#' }
#'
#' @return
#' A list
#'
#' @export
#'
hai_auto_cubist <- function(.data, .rec_obj, .splits_obj = NULL, .rsamp_obj = NULL,
.tune = TRUE, .grid_size = 10, .num_cores = 1,
.best_metric = "rmse"){
# Tidyeval ----
grid_size <- as.numeric(.grid_size)
num_cores <- as.numeric(.num_cores)
best_metric <- as.character(.best_metric)
data_tbl <- dplyr::as_tibble(.data)
splits <- .splits_obj
rec_obj <- .rec_obj
rsamp_obj <- .rsamp_obj
model_type <- "regression"
# Checks ----
if (!inherits(x = splits, what = "rsplit") && !is.null(splits)){
rlang::abort(
message = "'.splits_obj' must have a class of 'rsplit', use the rsample package.",
use_cli_format = TRUE
)
}
if (!inherits(x = rec_obj, what = "recipe")){
rlang::abort(
message = "'.rec_obj' must have a class of 'recipe'."
)
}
if (!inherits(x = rsamp_obj, what = "rset") && !is.null(rsamp_obj)){
rlang::abort(
message = "The '.rsamp_obj' argument must either be NULL or an object of
calss 'rset'.",
use_cli_format = TRUE
)
}
if (!inherits(x = splits, what = "rsplit") && !is.null(splits)){
rlang::abort(
message = "The '.splits_obj' argument must either be NULL or an object of
class 'rsplit'",
use_cli_format = TRUE
)
}
# Set default metric set ----
ms <- healthyR.ai::hai_default_regression_metric_set()
# Get splits if not then create
if (is.null(splits)){
splits <- rsample::initial_split(data = data_tbl)
} else {
splits <- splits
}
# Tune/Spec ----
if (.tune){
# Model Specification
model_spec <- parsnip::cubist_rules(
committees = tune::tune(),
neighbors = tune::tune()
)
} else {
model_spec <- parsnip::cubist_rules()
}
# Model Specification ----
model_spec <- model_spec %>%
parsnip::set_mode(mode = model_type) %>%
parsnip::set_engine(engine = "Cubist")
# Workflow ----
wflw <- workflows::workflow() %>%
workflows::add_recipe(rec_obj) %>%
workflows::add_model(model_spec)
# Tuning Grid ---
if (.tune){
# Make tuning grid
tuning_grid_spec <- dials::grid_latin_hypercube(
hardhat::extract_parameter_set_dials(model_spec),
size = grid_size
)
# Cross validation object
if (is.null(rsamp_obj)){
cv_obj <- rsample::mc_cv(
data = rsample::training(splits)
)
} else {
cv_obj <- rsamp_obj
}
# Tune the workflow
# Start parallel backed
modeltime::parallel_start(num_cores)
tuned_results <- wflw %>%
tune::tune_grid(
resamples = cv_obj,
grid = tuning_grid_spec,
metrics = ms
)
modeltime::parallel_stop()
# Get the best result set by a specified metric
best_result_set <- tuned_results %>%
tune::show_best(metric = best_metric, n = 1)
# Plot results
tune_results_plt <- tuned_results %>%
tune::autoplot() +
ggplot2::theme_minimal() +
ggplot2::geom_smooth(se = FALSE) +
ggplot2::theme(legend.position = "bottom")
# Make final workflow
wflw_fit <- wflw %>%
tune::finalize_workflow(
tuned_results %>%
tune::show_best(metric = best_metric, n = 1)
) %>%
parsnip::fit(rsample::training(splits))
} else {
wflw_fit <- wflw %>%
parsnip::fit(rsample::training(splits))
}
# Return ----
output <- list(
recipe_info = rec_obj,
model_info = list(
model_spec = model_spec,
wflw = wflw,
fitted_wflw = wflw_fit,
was_tuned = ifelse(.tune, "tuned", "not_tuned")
)
)
if (.tune){
output$tuned_info = list(
tuning_grid = tuning_grid_spec,
cv_obj = cv_obj,
tuned_results = tuned_results,
grid_size = grid_size,
best_metric = best_metric,
best_result_set = best_result_set,
tuning_grid_plot = tune_results_plt,
plotly_grid_plot = plotly::ggplotly(tune_results_plt)
)
}
attr(output, "function_type") <- "boilerplate"
attr(output, ".grid_size") <- .grid_size
attr(output, ".tune") <- .tune
attr(output, ".best_metric") <- .best_metric
attr(output, ".model_type") <- model_type
attr(output, ".engine") <- "Cubist"
return(invisible(output))
}
Example:
> output
$recipe_info
Recipe
Inputs:
role #variables
outcome 1
predictor 10
Operations:
Zero variance filter on recipes::all_predictors()
$model_info
$model_info$model_spec
Cubist Model Specification (regression)
Main Arguments:
committees = tune::tune()
neighbors = tune::tune()
Computational engine: Cubist
$model_info$wflw
== Workflow ===============================================================================
Preprocessor: Recipe
Model: cubist_rules()
-- Preprocessor ---------------------------------------------------------------------------
1 Recipe Step
* step_zv()
-- Model ----------------------------------------------------------------------------------
Cubist Model Specification (regression)
Main Arguments:
committees = tune::tune()
neighbors = tune::tune()
Computational engine: Cubist
$model_info$fitted_wflw
== Workflow [trained] =====================================================================
Preprocessor: Recipe
Model: cubist_rules()
-- Preprocessor ---------------------------------------------------------------------------
1 Recipe Step
* step_zv()
-- Model ----------------------------------------------------------------------------------
Call:
cubist.default(x = x, y = y, committees = 99L)
Number of samples: 24
Number of predictors: 10
Number of committees: 99
Number of rules per committee: 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 ...
$model_info$was_tuned
[1] "tuned"
$tuned_info
$tuned_info$tuning_grid
# A tibble: 10 x 2
committees neighbors
<int> <int>
1 50 6
2 51 0
3 81 1
4 3 2
5 31 9
6 29 7
7 70 3
8 74 4
9 14 7
10 99 5
$tuned_info$cv_obj
# Monte Carlo cross-validation (0.75/0.25) with 25 resamples
# A tibble: 25 x 2
splits id
<list> <chr>
1 <split [18/6]> Resample01
2 <split [18/6]> Resample02
3 <split [18/6]> Resample03
4 <split [18/6]> Resample04
5 <split [18/6]> Resample05
6 <split [18/6]> Resample06
7 <split [18/6]> Resample07
8 <split [18/6]> Resample08
9 <split [18/6]> Resample09
10 <split [18/6]> Resample10
# ... with 15 more rows
$tuned_info$tuned_results
# Tuning results
# Monte Carlo cross-validation (0.75/0.25) with 25 resamples
# A tibble: 25 x 4
splits id .metrics .notes
<list> <chr> <list> <list>
1 <split [18/6]> Resample01 <tibble [60 x 6]> <tibble [0 x 3]>
2 <split [18/6]> Resample02 <tibble [60 x 6]> <tibble [0 x 3]>
3 <split [18/6]> Resample03 <tibble [60 x 6]> <tibble [0 x 3]>
4 <split [18/6]> Resample04 <tibble [60 x 6]> <tibble [0 x 3]>
5 <split [18/6]> Resample05 <tibble [60 x 6]> <tibble [0 x 3]>
6 <split [18/6]> Resample06 <tibble [60 x 6]> <tibble [0 x 3]>
7 <split [18/6]> Resample07 <tibble [60 x 6]> <tibble [0 x 3]>
8 <split [18/6]> Resample08 <tibble [60 x 6]> <tibble [0 x 3]>
9 <split [18/6]> Resample09 <tibble [60 x 6]> <tibble [0 x 3]>
10 <split [18/6]> Resample10 <tibble [60 x 6]> <tibble [0 x 3]>
# ... with 15 more rows
$tuned_info$grid_size
[1] 10
$tuned_info$best_metric
[1] "rmse"
$tuned_info$best_result_set
# A tibble: 1 x 8
committees neighbors .metric .estimator mean n std_err .config
<int> <int> <chr> <chr> <dbl> <int> <dbl> <chr>
1 99 5 rmse standard 2.64 25 0.149 Preprocessor1_Model10
$tuned_info$tuning_grid_plot
`geom_smooth()` using method = 'loess' and formula 'y ~ x'
$tuned_info$plotly_grid_plot
attr(,"function_type")
[1] "boilerplate"
attr(,".grid_size")
[1] 10
attr(,".tune")
[1] TRUE
attr(,".best_metric")
[1] "rmse"
attr(,".model_type")
[1] "regression"
attr(,".engine")
[1] "Cubist"
Function:
Example: