Engineer features for the BAF dataset — engineer

Reads the primary BAF dataset and engineers new features, such as n_missing, which counts the number of missing values across key tenure and financial columns. This calculation is performed out-of-memory using Arrow compute.

Usage

engineer_features(
  in_prefix = "baf-fraud/03_primary/variant=Base",
  out_prefix = "baf-fraud/04_feature/variant=Base",
  bucket_name = "lake",
  partitioning = "month",
  existing_data_behavior = "delete_matching",
  verbose = TRUE
)

Arguments

in_prefix: Character. Input dataset prefix (e.g., "03_primary/variant=Base").
out_prefix: Character. Output dataset prefix (e.g., "04_feature/variant=Base").
bucket_name: Character. The S3/MinIO bucket name. Default "lake".
partitioning: Character vector. Columns to partition by. Default "month".
existing_data_behavior: Character. Behavior when data exists. Default "delete_matching".
verbose: Logical. Whether to print progress messages. Default TRUE.

Value

Character. The output prefix path for downstream targets.