ingredients (version 0.3.1)

ceteris_paribus_2d: Ceteris Paribus 2D Plot

Description

This function calculates ceteris paribus profiles for grid of values spanned by two variables. It may be useful to identify or present interactions between two variables.

Usage

ceteris_paribus_2d(explainer, observation, grid_points = 101,
  variables = NULL)

Arguments

explainer

a model to be explained, preprocessed by the 'DALEX::explain' function

observation

a new observarvation for which predictions need to be explained

grid_points

number of points used for response path. Will be used for both variables

variables

if specified, then only these variables will be explained

Value

An object of the class 'ceteris_paribus_2d_explainer'. It's a data frame with calculated average responses.

Examples

# NOT RUN {
library("DALEX")
# Toy examples, because CRAN angels ask for them
titanic <- na.omit(titanic)
model_titanic_glm <- glm(survived == "yes" ~ gender + age + fare,
                       data = titanic, family = "binomial")

explain_titanic_glm <- explain(model_titanic_glm,
                           data = titanic[,-9],
                           y = titanic$survived == "yes")
cp_rf <- ceteris_paribus_2d(explain_titanic_glm, titanic[1,])
head(cp_rf)
plot(cp_rf)
 
# }
# NOT RUN {
library("randomForest")
set.seed(59)

apartments_rf_model <- randomForest(m2.price ~ construction.year + surface + floor +
      no.rooms + district, data = apartments)

explainer_rf <- explain(apartments_rf_model,
      data = apartmentsTest[,2:6], y = apartmentsTest$m2.price)

new_apartment <- apartmentsTest[1, ]
new_apartment

wi_rf_2d <- ceteris_paribus_2d(explainer_rf, observation = new_apartment,
         variables = c("surface", "floor", "no.rooms"))
head(wi_rf_2d)
plot(wi_rf_2d)
# }