Various metrics related to model sculpting

Usage

calc_dir_var_imp(object, newdata = NULL)

calc_cumul_R2(object, newdata = NULL)

Arguments

object: sculpture
newdata: (Optional) Data to calculate the importance from. If omitted, the data that were provided to build the sculpture are used.

Value

data.table with direct requested metrics.

Functions

calc_dir_var_imp(): Direct variable importance
calc_cumul_R2(): Calculate cumulative approximation of R^2

Examples

df <- mtcars
df$vs <- as.factor(df$vs)
model <- rpart::rpart(
  hp ~ mpg + carb + vs,
  data = df,
  control = rpart::rpart.control(minsplit = 10)
)
model_predict <- function(x) predict(model, newdata = x)
covariates <- c("mpg", "carb", "vs")
pm <- sample_marginals(df[covariates], n = 50, seed = 5)

rs <- sculpt_rough(
  dat = pm,
  model_predict_fun = model_predict,
  n_ice = 10,
  seed = 1,
  verbose = 0
)

# show direct variable importance
calc_dir_var_imp(rs)
#>    feature  variance variance_total     ratio
#>     <fctr>     <num>          <num>     <num>
#> 1:     mpg 1936.4199       2296.917 0.8430518
#> 2:    carb  346.5405       2296.917 0.1508720
#> 3:      vs    0.0000       2296.917 0.0000000

# show cumulative approximation R^2
calc_cumul_R2(rs)
#>    feature       R2
#>     <fctr>    <num>
#> 1:     mpg 0.849128
#> 2:    carb 1.000000
#> 3:      vs 1.000000