Volledige modelworkflow

Modelleren met tidymodels in R

David Svancer

Data Scientist

Data resamplen

Trainings- en testsets maken

initial_split()
- Maak een datasplit-object
training()
- Bouw de trainingsset
testing()
- Bouw de testset

leads_split <- initial_split(leads_df, 
                             strata = purchased)


leads_training <- leads_split %>% 
  training()


leads_test <- leads_split %>% 
  testing()

Model specificeren

Specificeer model met parsnip

logistic_reg()
- Algemene interface voor logistische regressie
set_engine()
- 'glm'-engine
set_mode()
- purchased is een nominale uitkomstvariabele
- Mode moet 'classification' zijn

logistic_model <- logistic_reg() %>%

  set_engine('glm') %>%

  set_mode('classification')

Logistic Regression Model 
Specification (classification)

Computational engine: glm

Feature engineering

Specificeer feature-engineeringstappen met recipes

recipe()
- Modelformule en trainingsdata
step_*()-functies
- Sequentiële preprocessingstappen

leads_recipe <- recipe(purchased ~ .,
                       data = leads_training) %>%

  step_corr(all_numeric(), threshold = 0.9) %>% 
  step_normalize(all_numeric()) %>% 
  step_dummy(all_nominal(), -all_outcomes())

leads_recipe

Data Recipe
Inputs:
      role #variables
   outcome          1
 predictor          6

Operations:
Correlation filter on all_numeric()
Centering and scaling for all_numeric()
Dummy variables from all_nominal(), -all_outcomes()

Recipe trainen

Train feature-engineeringstappen op de trainingsdata

prep()
- Geef het recipe-object door aan prep()
- Voeg leads_training toe als trainingsdata

leads_recipe_prep <- leads_recipe %>% 
  prep(training = leads_training)

leads_recipe_prep

Data Recipe
Inputs:
      role #variables
   outcome          1
 predictor          6
Training data contained 996 data points 
and no missing data.

Operations:
Correlation filter removed pages_per_visit [trained]
Centering and scaling for total_visits ... [trained]
Dummy variables from lead_source, us_location [trained]

Trainingsdata preprocessen

Pas de getrainde recipe toe op de trainingsdata en sla de resultaten op voor het fitten

leads_training_prep <- leads_recipe_prep %>% 
  bake(new_data = NULL)

leads_training_prep

# A tibble: 996 x 11
total_visits  total_time   ... lead_source_email  lead_source_organic_search ...  us_location_west
     <dbl>      <dbl>                <dbl>                      <dbl>                    <dbl>        
 1    0.611      0.958      ...       0                          0            ...         1
 2    0.103     -0.747      ...       1                          0            ...         0
 3    0.611     -0.278      ...       0                          1            ...         1
 4   -0.151     -0.842      ...       0                          0            ...         1
 5   -0.659      1.19       ...       1                          0            ...         0 
# ... with 991 more rows

Testdata preprocessen

Pas de getrainde recipe toe op de testdata en sla de resultaten op voor evaluatie

leads_test_prep <- leads_recipe_prep %>% 
  bake(new_data = leads_test)

leads_test_prep

# A tibble: 332 x 11
 total_visits  total_time  ...  lead_source_email  lead_source_organic_search ...  us_location_west
     <dbl>      <dbl>                <dbl>                      <dbl>                    <dbl>        
 1    0.864     -0.984     ...        0                          0            ...         1
 2   -0.151      1.33      ...        0                          0            ...         0
 3   -0.405     -0.843     ...        0                          1            ...         1
 4   -0.659     -1.14      ...        1                          0            ...         0
 5    1.12       0.725     ...        0                          0            ...         1   
# ... with 327 more rows

Model fitten en voorspellen

Train een logistisch regressiemodel met fit()

Gebruik de voorbewerkte trainingsset, leads_training_prep

Krijg voorspellingen met predict()

Voorspel klassen en kansen
Gebruik de voorbewerkte testset, leads_test_prep

logistic_fit <- logistic_model %>% 
  fit(purchased ~ .,
      data = leads_training_prep)

class_preds <- predict(logistic_fit, 
                       new_data = leads_test_prep,
                       type = 'class')

prob_preds <- predict(logistic_fit, 
                      new_data = leads_test_prep,
                      type = 'prob')

Voorspellingsresultaten combineren

Combineer voorspellingen in een resultaten-set voor yardstick-metrics

Selecteer de echte uitkomstvariabele purchased uit de testset
Voeg voorspellingen samen met bind_cols()

leads_results <- leads_test %>% 
  select(purchased) %>%

  bind_cols(class_preds, prob_preds)


leads_results

# A tibble: 332 x 4
   purchased .pred_class .pred_yes .pred_no
   <fct>     <fct>           <dbl>    <dbl>
 1 no        no             0.257     0.743
 2 yes       yes            0.896     0.104
 3 no        no             0.0852    0.915
 4 no        no             0.183     0.817
 5 yes       yes            0.776     0.224
# ... with 327 more rows

Model evalueren

Evalueer modelprestatie met yardstick

De resultaten-set werkt met alle yardstick-metrics voor evaluatie
Confusion matrix, sensitiviteit, specificiteit en andere metrics

leads_results %>% 
  conf_mat(truth = purchased, 
           estimate = .pred_class)

          Truth
Prediction yes  no
       yes  77  34
       no   43 178

Laten we oefenen!

Modelleren met tidymodels in R