tidymodels2

ds1

tidymodels

prediction

yacsda

statlearning

error

string

Published

May 17, 2023

Aufgabe

Ein merkwürdiger Fehler bzw. eine merkwürdige Fehlermeldung in Tidymodels - das untersuchen wir hier genauer und versuchen das Phänomen zu erklären.

Aufgabe

Erläutern Sie die Ursachen des Fehlers! Schalten Sie den Fehler an und ab, um zu zeigen, dass Sie Ihn verstehen.

Startup

library(tidyverse)
library(tidymodels)

Data import

data("mtcars")

d_train <- mtcars %>% slice_head(n = 20)
d_test <- mtcars %>% slice(21:n())

Recipe

preds_chosen <- c("hp", "disp", "am")

rec1 <- 
  recipe( ~ ., data = d_train) %>% 
  update_role(all_predictors(), new_role = "id") %>% 
  update_role(all_of(preds_chosen), new_role = "predictor") %>% 
  update_role(mpg, new_role = "outcome")
rec1

── Recipe ──────────────────────────────────────────────────────────────────────

── Inputs

Number of variables by role

outcome:   1
predictor: 3
id:        7

d_train_baked <-
  rec1 %>% 
  prep() %>% 
  bake(new_data = NULL)

glimpse(d_train_baked)

Rows: 20
Columns: 11
$ mpg  <dbl> 21.0, 21.0, 22.8, 21.4, 18.7, 18.1, 14.3, 24.4, 22.8, 19.2, 17.8,…
$ cyl  <dbl> 6, 6, 4, 6, 8, 6, 8, 4, 4, 6, 6, 8, 8, 8, 8, 8, 8, 4, 4, 4
$ disp <dbl> 160.0, 160.0, 108.0, 258.0, 360.0, 225.0, 360.0, 146.7, 140.8, 16…
$ hp   <dbl> 110, 110, 93, 110, 175, 105, 245, 62, 95, 123, 123, 180, 180, 180…
$ drat <dbl> 3.90, 3.90, 3.85, 3.08, 3.15, 2.76, 3.21, 3.69, 3.92, 3.92, 3.92,…
$ wt   <dbl> 2.620, 2.875, 2.320, 3.215, 3.440, 3.460, 3.570, 3.190, 3.150, 3.…
$ qsec <dbl> 16.46, 17.02, 18.61, 19.44, 17.02, 20.22, 15.84, 20.00, 22.90, 18…
$ vs   <dbl> 0, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1
$ am   <dbl> 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1
$ gear <dbl> 4, 4, 4, 3, 3, 3, 3, 4, 4, 4, 4, 3, 3, 3, 3, 3, 3, 4, 4, 4
$ carb <dbl> 4, 4, 1, 1, 2, 1, 4, 2, 2, 4, 4, 3, 3, 3, 4, 4, 4, 1, 2, 1

Model 1

model_lm <- linear_reg()

Workflow 1

wf1 <-
  workflow() %>% 
  add_model(model_lm) %>% 
  add_recipe(rec1)

Fit

lm_fit1 <-
  wf1 %>% 
  fit(d_train)

preds <-
  lm_fit1 %>% 
  predict(d_test)

head(preds)

# A tibble: 6 × 1
  .pred
  <dbl>
1  22.6
2  17.2
3  17.4
4  12.1
5  14.9
6  28.2

Aus Gründen der Reproduzierbarkeit bietet es sich an, eine SessionInfo anzugeben:

sessionInfo()

R version 4.2.1 (2022-06-23)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Big Sur ... 10.16

Matrix products: default
BLAS:   /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] yardstick_1.2.0    workflowsets_1.0.1 workflows_1.1.3    tune_1.1.2        
 [5] rsample_1.2.0      recipes_1.0.8      parsnip_1.1.1      modeldata_1.2.0   
 [9] infer_1.0.5        dials_1.2.0        scales_1.2.1       broom_1.0.5       
[13] tidymodels_1.1.1   lubridate_1.9.3    forcats_1.0.0      stringr_1.5.0     
[17] dplyr_1.1.3        purrr_1.0.2        readr_2.1.4        tidyr_1.3.0       
[21] tibble_3.2.1       ggplot2_3.4.4      tidyverse_2.0.0    colorout_1.3-0    

loaded via a namespace (and not attached):
 [1] foreach_1.5.2       jsonlite_1.8.7      splines_4.2.1      
 [4] prodlim_2023.03.31  GPfit_1.0-8         yaml_2.3.7         
 [7] globals_0.16.2      ipred_0.9-14        pillar_1.9.0       
[10] backports_1.4.1     lattice_0.21-8      glue_1.6.2         
[13] digest_0.6.33       hardhat_1.3.0       colorspace_2.1-0   
[16] htmltools_0.5.6.1   Matrix_1.5-4.1      timeDate_4022.108  
[19] pkgconfig_2.0.3     lhs_1.1.6           DiceDesign_1.9     
[22] listenv_0.9.0       gower_1.0.1         lava_1.7.2.1       
[25] tzdb_0.4.0          timechange_0.2.0    generics_0.1.3     
[28] ellipsis_0.3.2      withr_2.5.2         furrr_0.3.1        
[31] nnet_7.3-19         cli_3.6.1           survival_3.5-5     
[34] magrittr_2.0.3      evaluate_0.21       fansi_1.0.5        
[37] future_1.33.0       parallelly_1.36.0   MASS_7.3-60        
[40] class_7.3-22        tools_4.2.1         data.table_1.14.8  
[43] hms_1.1.3           lifecycle_1.0.3     munsell_0.5.0      
[46] compiler_4.2.1      rlang_1.1.1         grid_4.2.1         
[49] iterators_1.0.14    rstudioapi_0.15.0   htmlwidgets_1.6.2  
[52] rmarkdown_2.25      gtable_0.3.4        codetools_0.2-19   
[55] R6_2.5.1            knitr_1.45          fastmap_1.1.1      
[58] future.apply_1.11.0 utf8_1.2.3          stringi_1.7.12     
[61] parallel_4.2.1      Rcpp_1.0.11         vctrs_0.6.4        
[64] rpart_4.1.19        tidyselect_1.2.0    xfun_0.40

Lösung

Definiert man das Rezept so:

rec2 <- recipe(mpg ~ hp + disp + am, data = d_train)

Dann läuft predict() brav durch.

Auch dieser Code funktioniert:

rec3 <- 
  recipe(mpg ~ ., data = d_train) %>% 
  update_role(all_predictors(), new_role = "id") %>% 
  update_role(all_of(preds_chosen), new_role = "predictor") %>% 
  update_role(mpg, new_role = "outcome")

Das Problem von rec1 scheint darin zu legen, dass die Rollen der Variablen nicht richtig gelöscht werden, was predict() verwirrt:

rec1 <- 
  recipe(mpg ~ ., data = d_train) %>% 
  update_role(all_predictors(), new_role = "id") %>% 
  update_role(all_of(preds_chosen), new_role = "predictor") %>% 
  update_role(mpg, new_role = "outcome")
rec1

── Recipe ──────────────────────────────────────────────────────────────────────

── Inputs

Number of variables by role

outcome:   1
predictor: 3
id:        7

Daher läuft das Rezept rec3 durch, wenn man zunächst alle Prädiktoren in ID-Variablen umwandelt: Damit sind alle Rollen wieder sauber.

Categories:

ds1
tidymodels
prediction
yacsda
statlearning
error
string