target_encode_apply: Applies a preparation map
In signaux-faibles/fte: Feature Target Encoding for classification

Description Usage Arguments

View source: R/fte.R

Applies a preparation map

1
2
3

target_encode_apply(data, group_variables, outcome_variable,
  preparation_map, holdout_type = "none", prior_sample_size = 30L,
  noise_level = 0, seed = 1793)

`data`	`data.frame()` Data to apply the target encode preparation map to.
`group_variables`	`character()` On which categorical variables to apply the target encoding?
`outcome_variable`	`character()` Name of the column representing the boolean outcome.
`preparation_map`	`data.frame()` Preparation map formatted as the expected output from target_encode_create.
`holdout_type`	`character(1)` The holdout type used. Must be one of: "LeaveOneOut", "None".
`prior_sample_size`	`integer(1)` If non-zero, then a bayesian blended average is computed. The prior_sample_size defines the weight of the population mean compared to the sample mean.
`noise_level`	`float(1)` The amount of random noise added to the target encoding. This helps prevent overfitting. Defaults to 0.01.
`seed`	`integer(1)` Seed for reproductible random values.