local_interactions: Model Agnostic Sequential Variable Attributions with...

View source: R/local_interactions.R

local_interactionsR Documentation

Model Agnostic Sequential Variable Attributions with Interactions

Description

This function implements decomposition of model predictions with identification of interactions. The complexity of this function is O(2*p) for additive models and O(2*p^2) for interactions. This function works in a similar way to step-up and step-down greedy approximations in function break_down(). The main difference is that in the first step the order of variables and interactions is determined. And in the second step the impact is calculated.

Usage

local_interactions(x, ...)

## S3 method for class 'explainer'
local_interactions(x, new_observation, keep_distributions = FALSE, ...)

## Default S3 method:
local_interactions(
  x,
  data,
  predict_function = predict,
  new_observation,
  label = class(x)[1],
  keep_distributions = FALSE,
  order = NULL,
  interaction_preference = 1,
  ...
)

Arguments

x

an explainer created with function explain or a model.

...

other parameters.

new_observation

a new observation with columns that correspond to variables used in the model.

keep_distributions

if TRUE, then the distribution of partial predictions is stored in addition to the average.

data

validation dataset, will be extracted from x if it's an explainer.

predict_function

predict function, will be extracted from x if it's an explainer.

label

character - the name of the model. By default it's extracted from the 'class' attribute of the model.

order

if not NULL, then it will be a fixed order of variables. It can be a numeric vector or vector with names of variables/interactions.

interaction_preference

an integer specifying which interactions will be present in an explanation. The larger the integer, the more frequently interactions will be presented.

Value

an object of the break_down class.

References

Explanatory Model Analysis. Explore, Explain and Examine Predictive Models. https://ema.drwhy.ai

See Also

break_down, local_attributions

Examples

library("DALEX")
library("iBreakDown")
set.seed(1313)
model_titanic_glm <- glm(survived ~ gender + age + fare,
                       data = titanic_imputed, family = "binomial")
explain_titanic_glm <- explain(model_titanic_glm,
                           data = titanic_imputed,
                           y = titanic_imputed$survived,
                           label = "glm")

bd_glm <- local_interactions(explain_titanic_glm, titanic_imputed[1, ],
       interaction_preference = 500)
bd_glm
plot(bd_glm, max_features = 2)

## Not run: 
library("randomForest")
# example with interaction
# classification for HR data
model <- randomForest(status ~ . , data = HR)
new_observation <- HR_test[1,]

explainer_rf <- explain(model,
                 data = HR[1:1000,1:5])

bd_rf <- local_interactions(explainer_rf,
                 new_observation)

bd_rf
plot(bd_rf)

# example for regression - apartment prices
# here we do not have intreactions
model <- randomForest(m2.price ~ . , data = apartments)
explainer_rf <- explain(model,
         data = apartments_test[1:1000,2:6],
         y = apartments_test$m2.price[1:1000])

new_observation <- apartments_test[1,]

bd_rf <- local_interactions(explainer_rf,
                 new_observation,
                 keep_distributions = TRUE)

bd_rf
plot(bd_rf)
plot(bd_rf, plot_distributions = TRUE)

## End(Not run)

iBreakDown documentation built on May 29, 2024, 10:38 a.m.