ingredients (version 0.3.1)

ceteris_paribus: Ceteris Paribus Profiles aka Individual Variable Profiles

Description

This explainer works for individual observations. For each observation it calculates Ceteris Paribus Profiles for selected variables. Such profiles can be used to hypothesize about model results if selected variable is changed. For this reason it is also called 'What-If Profiles'.

Usage

ceteris_paribus(x, ...)
# S3 method for explainer
ceteris_paribus(x, new_observation, y = NULL,
  variables = NULL, variable_splits = NULL, grid_points = 101, ...)
# S3 method for default
ceteris_paribus(x, data, predict_function = predict,
  new_observation, y = NULL, variables = NULL,
  variable_splits = NULL, grid_points = 101, label = class(x)[1],
  ...)

Arguments

x

a model to be explained, or an explainer created with the `DALEX::explain()` function.

...

other parameters

new_observation

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

y

true labels for `new_observation`. If specified then will be added to ceteris paribus plots.

variables

names of variables for which profiles shall be calculated. Will be passed to `calculate_variable_splits()`. If NULL then all variables from the validation data will be used.

variable_splits

named list of splits for variables, in most cases created with `calculate_variable_splits()`. If NULL then it will be calculated based on validation data avaliable in the `explainer`.

grid_points

number of points for profile. Will be passed to `calculate_variable_splits()`.

data

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

predict_function

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

label

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

Value

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

Details

Find more detailes in Ceteris Paribus Chapter.

References

Predictive Models: Visual Exploration, Explanation and Debugging https://pbiecek.github.io/PM_VEE

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(explain_titanic_glm, titanic[1,])
cp_rf
plot(cp_rf, variables = "age")

# }
# NOT RUN {
 library("randomForest")
 model_titanic_rf <- randomForest(survived ~ gender + age + class + embarked +
                                    fare + sibsp + parch,  data = titanic)
 model_titanic_rf

 explain_titanic_rf <- explain(model_titanic_rf,
                           data = titanic[,-9],
                           y = titanic$survived,
                           label = "Random Forest v7")

# select few passangers
selected_passangers <- select_sample(titanic, n = 20)
cp_rf <- ceteris_paribus(explain_titanic_rf, selected_passangers)
cp_rf

plot(cp_rf, variables = "age") +
  show_observations(cp_rf, variables = "age") +
  show_rugs(cp_rf, variables = "age", color = "red")

# }