Create Model Explainer

Black-box models may have very different structures. This function creates a unified representation of a model, which can be further processed by various explainers.

explain.default(model, data = NULL, y = NULL,
  predict_function = yhat, link = I, ..., label = tail(class(model),
  1))

explain(model, data = NULL, y = NULL, predict_function = yhat,
  link = I, ..., label = tail(class(model), 1))

Arguments

model	object - a model to be explained
data	data.frame or matrix - data that was used for fitting. If not provided then will be extracted from the model
y	numeric vector with outputs / scores. Currently used only by variable_dropout() explainer.
predict_function	function that takes two arguments: model and new data and returns numeric vector with predictions
link	function - a transformation/link function that shall be applied to raw model predictions
...	other parameters
label	character - the name of the model. By default it's extracted from the 'class' attribute of the model

Value

An object of the class 'explainer'.

It's a list with following fields:

• model the explained model

• data the dataset used for training

• predict_function function that may be used for model predictions, shall return a single numerical value for each observation.

• class class/classes of a model

• label label, by default it's the last value from the class vector, but may be set to any character.

Details

Please NOTE, that the model is actually the only required argument. But some explainers may require that others will be provided too.

Examples

apartments_lm <- lm(m2.price ~ ., data = apartments)
apartments_lm_ex <- explain(apartments_lm, data = apartments, label = "apartments_lm")
apartments_lm_ex
#> Model label:  apartments_lm 
#> Model class:  lm 
#> Data head  :
#>   m2.price construction.year surface floor no.rooms    district
#> 1     5897              1953      25     3        1 Srodmiescie
#> 2     1818              1992     143     9        5     Bielany

 
library("breakDown2")
wine_lm_model4 <- lm(quality ~ pH + residual.sugar + sulphates + alcohol, data = wine)
#> Error in is.data.frame(data): object 'wine' not found
wine_lm_explainer4 <- explain(wine_lm_model4, data = wine, label = "model_4v")
#> Error in explain(wine_lm_model4, data = wine, label = "model_4v"): object 'wine' not found
wine_lm_explainer4
#> Error in eval(expr, envir, enclos): object 'wine_lm_explainer4' not found

library("randomForest")
wine_rf_model4 <- randomForest(quality ~ pH + residual.sugar + sulphates + alcohol, data = wine)
#> Error in eval(m$data, parent.frame()): object 'wine' not found
wine_rf_explainer4 <- explain(wine_rf_model4, data = wine, label = "model_rf")
#> Error in explain(wine_rf_model4, data = wine, label = "model_rf"): object 'wine' not found
wine_rf_explainer4
#> Error in eval(expr, envir, enclos): object 'wine_rf_explainer4' not found