Make function fast_classification_parsnip_spec_tbl()

[spsanderson]

Function:

#' Utility Classification call to `parsnip`
#'
#' @family Utility
#'
#' @author Steven P. Sanderson II, MPH
#'
#' @details Creates a tibble of parsnip classification model specifications. This will
#' create a tibble of 32 different classification model specifications which can be
#' filtered. The model specs are created first and then filtered out. This will
#' only create models for __classification__ problems. To find all of the supported
#' models in this package you can visit \url{https://www.tidymodels.org/find/parsnip/}
#'
#' @description Creates a tibble of parsnip classification model specifications.
#'
#' @param .parsnip_fns The default for this is set to `all`. This means that all
#' of the parsnip __classification__ functions will be used, for example `bag_mars()`,
#' or `bart()`. You can also choose to pass a c() vector like `c("barg_mars","bart")`
#' @param .parsnip_eng The default for this is set to `all`. This means that all
#' of the parsnip __classification engines__ will be used, for example `earth`, or
#' `dbarts`. You can also choose to pass a c() vector like `c('earth', 'dbarts')`
#'
#' @examples
#' fast_classification_parsnip_spec_tbl(.parsnip_fns = "logistic_reg")
#' fast_classification_parsnip_spec_tbl(.parsnip_eng = c("earth","dbarts"))
#'
#' @return
#' A tibble with an added class of 'fst_class_spec_tbl'
#'
#' @importFrom parsnip linear_reg cubist_rules poisson_reg survival_reg
#'
#' @name fast_classification_parsnip_spec_tbl
NULL

#' @export
#' @rdname fast_classification_parsnip_spec_tbl

fast_classification_parsnip_spec_tbl <- function(.parsnip_fns = "all",
                                             .parsnip_eng = "all") {
  
  # Thank you https://stackoverflow.com/questions/74691333/build-a-tibble-of-parsnip-model-calls-with-match-fun/74691529#74691529
  # Tidyeval ----
  call <- list(.parsnip_fns) %>%
    purrr::flatten_chr()
  engine <- list(.parsnip_eng) %>%
    purrr::flatten_chr()
  
  # Make tibble
  mod_tbl <- tibble::tribble(
    ~.parsnip_engine, ~.parsnip_mode, ~.parsnip_fns,
    "earth","classification","bag_mars",
    "earth","classification","discrim_flexible",
    "dbarts","classification","bart",
    "MASS","classification","discrim_linear",
    "mda","classification","discrim_linear",
    "sda","classification","discrim_linear",
    "sparsediscrim","classification","discrim_linear",
    "MASS","classification","discrim_quad",
    "sparsediscrim","classification","discrim_quad",
    "klaR","classification","discrim_regularized",
    "mgcv","classification","gen_additive_mod",
    "brulee","classification","logistic_reg",
    "gee","classification","logistic_reg",
    "glm","classification","logistic_reg",
    "glmer","classification","logistic_reg",
    "glmnet","classification","logistic_reg",
    "LiblineaR","classification","logistic_reg",
    "earth","classification","mars",
    "brulee","classification","mlp",
    "nnet","classification","mlp",
    "brulee","classification","multinom_reg",
    "glmnet","classification","multinom_reg",
    "nnet","classification","multinom_reg",
    "klaR","classification","naive_Bayes",
    "kknn","classification","nearest_neighbor",
    "mixOmics","classification","pls",
    "xrf","classification","rule_fit",
    "kernlab","classification","svm_linear",
    "LiblineaR","classification","svm_linear",
    "kernlab","classification","svm_poly",
    "kernlab","classification","svm_rbf",
    "liquidSVM","classification","svm_rbf"
  )
  
  # Filter ----
  if (!"all" %in% engine){
    mod_tbl <- mod_tbl %>%
      dplyr::filter(.parsnip_engine %in% engine)
  }
  
  if (!"all" %in% call){
    mod_tbl <- mod_tbl %>%
      dplyr::filter(.parsnip_fns %in% call)
  }
  
  mod_filtered_tbl <- mod_tbl
  
  mod_spec_tbl <- mod_filtered_tbl %>%
    dplyr::mutate(
      model_spec = purrr::pmap(
        dplyr::cur_data(),
        ~ match.fun(..3)(mode = ..2, engine = ..1)
        #~ get(..3)(mode = ..2, engine = ..1)
      )
    ) %>%
    # add .model_id column
    dplyr::mutate(.model_id = dplyr::row_number()) %>%
    dplyr::select(.model_id, dplyr::everything())
  
  # Return ----
  class(mod_spec_tbl) <- c("fst_class_spec_tbl", class(mod_spec_tbl))
  class(mod_spec_tbl) <- c("tidyaml_mod_spec_tbl", class(mod_spec_tbl))
  attr(mod_spec_tbl, ".parsnip_engines") <- .parsnip_eng
  attr(mod_spec_tbl, ".parsnip_functions") <- .parsnip_fns
  
  return(mod_spec_tbl)
  
}

Examples:

> fast_classification_parsnip_spec_tbl(.parsnip_fns = "logistic_reg")
# A tibble: 6 × 5
  .model_id .parsnip_engine .parsnip_mode  .parsnip_fns model_spec
      <int> <chr>           <chr>          <chr>        <list>    
1         1 brulee          classification logistic_reg <spec[+]> 
2         2 gee             classification logistic_reg <spec[+]> 
3         3 glm             classification logistic_reg <spec[+]> 
4         4 glmer           classification logistic_reg <spec[+]> 
5         5 glmnet          classification logistic_reg <spec[+]> 
6         6 LiblineaR       classification logistic_reg <spec[+]> 
> fast_classification_parsnip_spec_tbl(.parsnip_eng = c("earth","dbarts"))
# A tibble: 4 × 5
  .model_id .parsnip_engine .parsnip_mode  .parsnip_fns     model_spec
      <int> <chr>           <chr>          <chr>            <list>    
1         1 earth           classification bag_mars         <spec[+]> 
2         2 earth           classification discrim_flexible <spec[+]> 
3         3 dbarts          classification bart             <spec[+]> 
4         4 earth           classification mars             <spec[+]> 

[spsanderson]

Pick up here: https://www.tidymodels.org/find/parsnip/

Page 3