tidymodels-使用step_ns

如何解决tidymodels-使用step_ns

我正在尝试编写一个适合使用step_ns（）的配方重新采样的函数。由于某种原因，我收到错误消息：

Fold01: recipe: Error: Not all variables in the recipe are present in the supplied training set

，以此类推。然后

警告信息： All models failed in [fit_resamples()]. See the .notes column.

这是我的代码：

compare_basis_exp_to_base_mod <- function (data,outcome,metric,...) {

  outcome <- rlang::enquo(outcome)
  metric <- rlang::enquo(metric)
  
  pred_list <- colnames(data)
  
  outcome_str <- substring(deparse(substitute(outcome)),2)
  outcome_str_id <- which(colnames(data) %in% outcome_str)
  
  predictor <- pred_list[-outcome_str_id]
  
  data <- data %>% 
    rename(prediction = !!outcome)

  res <- tibble(splits = list(),id = character(),.metrics = list(),.notes = list(),.predictions = list(),pred = character())
  
  rec_without_splines <- recipe(prediction ~ .,data = data) %>%
    prep()
  
  rec_with_splines <- recipe(prediction ~ .,data = data) %>%
    step_ns(all_predictors(),...) %>% 
    prep()
  
  folds_without_splines <- vfold_cv(juice(rec_without_splines),strata = prediction)
  
  folds_with_splines <- vfold_cv(juice(rec_with_splines),strata = prediction)
  

  mod <- linear_reg() %>% 
    set_engine("lm")

  mod_without_splines <- fit_resamples(mod,rec_without_splines,folds_without_splines,metrics = metric_set(!!metric),control = control_resamples(save_pred = TRUE)) %>%
    mutate(pred = "no_splines")
  
  mod_with_splines <- fit_resamples(mod,rec_with_splines,folds_with_splines,control = control_resamples(save_pred = TRUE)) %>%
    mutate(pred = "with_splines")

  res <- mod_without_splines %>%
    bind_rows(mod_with_splines)
  
  return (res)
}

基本上，参数data包含两列表，而outcome是结果列的名称。除了使用此功能（我刚刚接触tidymodels之外）时，我只想了解是什么导致了此错误以及如何修复该错误。计算mod_with_splines时出错。

here遇到了类似的问题。但是我不知道这是否与我的问题有关。在将食谱传递给fit_resamples之前，我无法准备食谱。 （或者我认为）

任何帮助将不胜感激。谢谢。

解决方法

您的问题来自试图在已经通过同一配方运行的数据集上应用配方。

如果我们假设预测变量为X1和X2，那么rec_with_splines应该是这些变量。但是，由于folds_with_splines包含rec_with_splines的榨汁结果，因此folds_with_splines实际上包含X1_ns_1，X1_ns_2，X2_ns_1和X2_ns_2。不是X1和X2。

我建议使用workflows来组合预处理和建模步骤。并将原始数据传递到vfold_cv()。

library(tidymodels)
compare_basis_exp_to_base_mod <- function (data,outcome,metric,...) {

  outcome <- rlang::enquo(outcome)
  metric <- rlang::enquo(metric)
  
  pred_list <- colnames(data)
  
  outcome_str <- substring(deparse(substitute(outcome)),2)
  outcome_str_id <- which(colnames(data) %in% outcome_str)
  
  predictor <- pred_list[-outcome_str_id]
  
  data <- data %>% 
    rename(prediction = !!outcome)
  
  rec_without_splines <- recipe(prediction ~ .,data = data) %>%
    prep()
  
  rec_with_splines <- recipe(prediction ~ .,data = data) %>%
    step_ns(all_predictors(),...)
  
  mod <- linear_reg() %>% 
    set_engine("lm")
 
  wf_without_splines <- workflow() %>%
    add_recipe(rec_without_splines) %>%
    add_model(mod)
  
  wf_with_splines <- workflow() %>%
    add_recipe(rec_with_splines) %>%
    add_model(mod)
  
  data_folds <- vfold_cv(data,strata = prediction)

  mod_without_splines <- fit_resamples(wf_without_splines,data_folds,metrics = metric_set(!!metric),control = control_resamples(save_pred = TRUE)) %>%
    mutate(pred = "no_splines")
  
  mod_with_splines <- fit_resamples(wf_with_splines,control = control_resamples(save_pred = TRUE)) %>%
    mutate(pred = "with_splines")

  res <- mod_without_splines %>%
    bind_rows(mod_with_splines)
  
  return (res)
}

res <- compare_basis_exp_to_base_mod(mtcars,mpg,rmse)