6.5 R code examples

Source

Load DALEX packages

suppressMessages(library(DALEX))

Load Titanic imputed dataset.

data("titanic_imputed")
head(titanic_imputed)

##   gender age class    embarked  fare sibsp parch survived
## 1   male  42   3rd Southampton  7.11     0     0        0
## 2   male  13   3rd Southampton 20.05     0     2        0
## 3   male  16   3rd Southampton 20.05     1     1        0
## 4 female  39   3rd Southampton 20.05     1     1        1
## 5 female  16   3rd Southampton  7.13     0     0        1
## 6   male  25   3rd Southampton  7.13     0     0        1

Build a random forest model.

suppressMessages(library(randomForest))
model_titanic_rf <- randomForest(survived ~ gender + age + class + embarked + 
                                   fare + sibsp + parch,  data = titanic_imputed, 
                                 classification = TRUE)

## Warning in randomForest.default(m, y, ...): The response has five or fewer
## unique values.  Are you sure you want to do regression?

model_titanic_rf

## 
## Call:
##  randomForest(formula = survived ~ gender + age + class + embarked +      fare + sibsp + parch, data = titanic_imputed, classification = TRUE) 
##                Type of random forest: regression
##                      Number of trees: 500
## No. of variables tried at each split: 2
## 
##           Mean of squared residuals: 0.1402275
##                     % Var explained: 35.78

Build an explainer object.

explain_titanic_rf <- explain(model_titanic_rf, 
                                  data = titanic_imputed,
                                  y = titanic_imputed$survived, 
                                  label = "Random Forest",
                                  colorize = FALSE)

## Preparation of a new explainer is initiated
##   -> model label       :  Random Forest 
##   -> data              :  2207  rows  8  cols 
##   -> target variable   :  2207  values 
##   -> predict function  :  yhat.randomForest  will be used (  default  )
##   -> predicted values  :  No value for predict function target column. (  default  )
##   -> model_info        :  package randomForest , ver. 4.7.1.1 , task regression (  default  ) 
##   -> predicted values  :  numerical, min =  0.007568155 , mean =  0.3223656 , max =  0.9933732  
##   -> residual function :  difference between y and yhat (  default  )
##   -> residuals         :  numerical, min =  -0.7815098 , mean =  -0.0002088249 , max =  0.8965246  
##   A new explainer has been created!

Create a dataframe for Johnny D passenger

new_passenger <- data.frame(
     class = factor("1st", levels = c("1st", "2nd", "3rd", "deck crew", "engineering crew", "restaurant staff", "victualling crew")),
     gender = factor("male", levels = c("female", "male")),
     age = 8,
     sibsp = 0,
     parch = 0,
     fare = 72,
     embarked = factor("Southampton", levels = c("Belfast", "Cherbourg", "Queenstown", "Southampton"))
)

Decompose model prediction with predict_parts()

sp_rf <- predict_parts(explain_titanic_rf, new_passenger)
sp_rf

##                                       contribution
## Random Forest: intercept                     0.322
## Random Forest: age = 8                       0.210
## Random Forest: class = 1st                   0.057
## Random Forest: gender = male                -0.058
## Random Forest: fare = 72                    -0.060
## Random Forest: sibsp = 0                     0.001
## Random Forest: embarked = Southampton       -0.005
## Random Forest: parch = 0                    -0.028
## Random Forest: prediction                    0.440

BD plot

plot(sp_rf)

order argument

bd_rf_order <- predict_parts(explainer = explain_titanic_rf,
                       new_observation = new_passenger, 
                                  type = "break_down",
                   order = c("class", "age", "gender", "fare", 
                             "parch", "sibsp", "embarked"))

plot(bd_rf_order, max_features = 3)

keep_distributions argument

bd_rf_distr <- predict_parts(explainer = explain_titanic_rf,
                       new_observation = new_passenger, 
                                  type = "break_down",
              order = c("age", "class", "fare", "gender", 
                        "embarked", "sibsp", "parch"),
                    keep_distributions = TRUE)

plot(bd_rf_distr, plot_distributions = TRUE)

## Warning: The `fun.y` argument of `stat_summary()` is deprecated as of ggplot2 3.3.0.
## ℹ Please use the `fun` argument instead.
## ℹ The deprecated feature was likely used in the iBreakDown package.
##   Please report the issue at
##   <https://github.com/ModelOriented/iBreakDown/issues>.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.