regr-tree03

statlearning

trees

tidymodels

string

Published

May 17, 2023

library(tidymodels)

── Attaching packages ────────────────────────────────────── tidymodels 1.1.1 ──

✔ broom        1.0.5     ✔ recipes      1.0.8
✔ dials        1.2.0     ✔ rsample      1.2.0
✔ dplyr        1.1.3     ✔ tibble       3.2.1
✔ ggplot2      3.4.4     ✔ tidyr        1.3.0
✔ infer        1.0.5     ✔ tune         1.1.2
✔ modeldata    1.2.0     ✔ workflows    1.1.3
✔ parsnip      1.1.1     ✔ workflowsets 1.0.1
✔ purrr        1.0.2     ✔ yardstick    1.2.0

── Conflicts ───────────────────────────────────────── tidymodels_conflicts() ──
✖ purrr::discard() masks scales::discard()
✖ dplyr::filter()  masks stats::filter()
✖ dplyr::lag()     masks stats::lag()
✖ recipes::step()  masks stats::step()
• Search for functions across packages at https://www.tidymodels.org/find/

Aufgabe

Berechnen Sie einfaches Prognosemodell auf Basis eines Entscheidungsbaums!

Modellformel: am ~ . (Datensatz mtcars)

Berichten Sie die Modellgüte (ROC-AUC).

Hinweise:

Tunen Sie alle Parameter (die der Engine anbietet).
Erstellen Sie ein Tuning-Grid mit 5 Werten pro Tuningparameter.
Führen Sie eine \(v=2\)-fache Kreuzvalidierung durch (weil die Stichprobe so klein ist).
Beachten Sie die üblichen Hinweise.

Lösung

Setup

library(tidymodels)
data(mtcars)
library(tictoc)  # Zeitmessung

Für Klassifikation verlangt Tidymodels eine nominale AV, keine numerische:

mtcars <-
  mtcars %>% 
  mutate(am = factor(am))

Daten teilen

d_split <- initial_split(mtcars)
d_train <- training(d_split)
d_test <- testing(d_split)

Modell(e)

mod_tree <-
  decision_tree(mode = "classification",
                cost_complexity = tune(),
                tree_depth = tune(),
                min_n = tune())

Rezept(e)

rec1 <- 
  recipe(am ~ ., data = d_train)

Resampling

rsmpl <- vfold_cv(d_train, v = 2)

Workflow

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

Tuning/Fitting

Tuninggrid:

tune_grid <- grid_regular(extract_parameter_set_dials(mod_tree), levels = 5)
tune_grid

# A tibble: 125 × 3
   cost_complexity tree_depth min_n
             <dbl>      <int> <int>
 1    0.0000000001          1     2
 2    0.0000000178          1     2
 3    0.00000316            1     2
 4    0.000562              1     2
 5    0.1                   1     2
 6    0.0000000001          4     2
 7    0.0000000178          4     2
 8    0.00000316            4     2
 9    0.000562              4     2
10    0.1                   4     2
# ℹ 115 more rows

tic()
fit1 <-
  tune_grid(object = wf1,
            grid = tune_grid,
            metrics = metric_set(roc_auc),
            resamples = rsmpl)

→ A | warning: 21 samples were requested but there were 12 rows in the data. 12 will be used.

There were issues with some computations   A: x1

→ B | warning: 30 samples were requested but there were 12 rows in the data. 12 will be used.

There were issues with some computations   A: x1
There were issues with some computations   A: x25   B: x1
There were issues with some computations   A: x25   B: x23
→ C | warning: 40 samples were requested but there were 12 rows in the data. 12 will be used.
There were issues with some computations   A: x25   B: x23
There were issues with some computations   A: x25   B: x25   C: x18
There were issues with some computations   A: x26   B: x25   C: x25
There were issues with some computations   A: x27   B: x25   C: x25
There were issues with some computations   A: x50   B: x25   C: x25
There were issues with some computations   A: x50   B: x48   C: x25
There were issues with some computations   A: x50   B: x50   C: x45
There were issues with some computations   A: x50   B: x50   C: x50

toc()

29.277 sec elapsed

Bester Kandidat

autoplot(fit1)

show_best(fit1)

# A tibble: 5 × 9
  cost_complexity tree_depth min_n .metric .estimator  mean     n std_err
            <dbl>      <int> <int> <chr>   <chr>      <dbl> <int>   <dbl>
1    0.0000000001          1     2 roc_auc binary     0.825     2   0.075
2    0.0000000178          1     2 roc_auc binary     0.825     2   0.075
3    0.00000316            1     2 roc_auc binary     0.825     2   0.075
4    0.000562              1     2 roc_auc binary     0.825     2   0.075
5    0.1                   1     2 roc_auc binary     0.825     2   0.075
# ℹ 1 more variable: .config <chr>

Finalisieren

wf1_finalized <-
  wf1 %>% 
  finalize_workflow(select_best(fit1))

Last Fit

final_fit <- 
  last_fit(object = wf1_finalized, d_split)

collect_metrics(final_fit)

# A tibble: 2 × 4
  .metric  .estimator .estimate .config             
  <chr>    <chr>          <dbl> <chr>               
1 accuracy binary         0.75  Preprocessor1_Model1
2 roc_auc  binary         0.833 Preprocessor1_Model1

Categories:

statlearning
trees
tidymodels
string