DALEX (version 1.2.1)

model_parts: Dataset Level Variable Importance as Change in Loss Function after Variable Permutations

Description

From DALEX version 1.0 this function calls the feature_importance Find information how to use this function here: https://pbiecek.github.io/ema/featureImportance.html.

Usage

model_parts(
  explainer,
  loss_function = loss_sum_of_squares,
  ...,
  type = "variable_importance",
  n_sample = 1000
)

Arguments

explainer

a model to be explained, preprocessed by the explain function

loss_function

a function that will be used to assess variable importance

...

other parameters

type

character, type of transformation that should be applied for dropout loss. variable_importance and raw results raw drop lossess, ratio returns drop_loss/drop_loss_full_model while difference returns drop_loss - drop_loss_full_model

n_sample

number of observations that should be sampled for calculation of variable importance. If negative then variable importance will be calculated on whole dataset (no sampling).

Value

An object of the class feature_importance. It's a data frame with calculated average response.

References

Explanatory Model Analysis. Explore, Explain and Examine Predictive Models. https://pbiecek.github.io/ema/

Examples

# NOT RUN {
 
# }
# NOT RUN {
library("ranger")
titanic_ranger_model <- ranger(survived~., data = titanic_imputed, num.trees = 50,
                               probability = TRUE)
explainer_ranger  <- explain(titanic_ranger_model, data = titanic_imputed[,-8],
                             y = titanic_imputed$survived)
vi_ranger <- model_parts(explainer_ranger, type = "raw")
head(vi_ranger, 8)
plot(vi_ranger)

titanic_glm_model <- glm(survived~., data = titanic_imputed, family = "binomial")
explainer_glm <- explain(titanic_glm_model, data = titanic_imputed[,-8],
                         y = titanic_imputed$survived)
logit <- function(x) exp(x)/(1+exp(x))
vi_glm <- model_parts(explainer_glm, type = "raw",
                        loss_function = function(observed, predicted)
                                     sum((observed - logit(predicted))^2))
head(vi_glm, 8)
plot(vi_glm)
 
# }
# NOT RUN {
# }