ebm_classify {interpret}

Build an EBM classification model

Description

Builds a classification model

Usage

ebm_classify(
  X, 
  y, 
  max_bins = 255,
  outer_bags = 16, 
  inner_bags = 0,
  learning_rate = 0.01, 
  validation_size = 0.15, 
  early_stopping_rounds = 50, 
  early_stopping_tolerance = 1e-4,
  max_rounds = 5000, 
  min_samples_leaf = 2,
  max_leaves = 3,
  random_state = 42
)

Arguments

X

features

y

targets

max_bins

number of bins to create

outer_bags

number of outer bags

inner_bags

number of inner bags

learning_rate

learning rate

validation_size

amount of data to use for validation

early_stopping_rounds

how many rounds without improvement before we quit

early_stopping_tolerance

how much does the round need to improve by to be considered as an advancement

max_rounds

number of boosting rounds

min_samples_leaf

number of samples required for a split

max_leaves

how many leaves allowed

random_state

random seed

Value

Returns an EBM model

Examples

  data(mtcars)
  X <- subset(mtcars, select = -c(vs))
  y <- mtcars$vs

  set.seed(42)
  data_sample <- sample(length(y), length(y) * 0.8)

  X_train <- X[data_sample, ]
  y_train <- y[data_sample]
  X_test <- X[-data_sample, ]
  y_test <- y[-data_sample]

  ebm <- ebm_classify(X_train, y_train)