Initial commit: BAF Lakehouse fraud detection pipeline

End-to-end LightGBM fraud detection pipeline built as an R package,
orchestrated by targets with data stored in MinIO via Apache Arrow.
Includes 6-layer Lakehouse architecture, class imbalance tournament,
formally tuned hyperparameters (PR-AUC 0.198), and Quarto RevealJS slides.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

man/engineer_features.Rd

@@ -0,0 +1,37 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{engineer_features}
+\alias{engineer_features}
+\title{Engineer features for the BAF dataset}
+\usage{
+engineer_features(
+  in_prefix = "03_primary/variant=Base",
+  out_prefix = "04_feature/variant=Base",
+  bucket_name = "baf-fraud",
+  partitioning = "month",
+  existing_data_behavior = "delete_matching",
+  verbose = TRUE
+)
+}
+\arguments{
+\item{in_prefix}{Character. Input dataset prefix (e.g., "03_primary/variant=Base").}
+
+\item{out_prefix}{Character. Output dataset prefix (e.g., "04_feature/variant=Base").}
+
+\item{bucket_name}{Character. The S3/MinIO bucket name. Default "baf-fraud".}
+
+\item{partitioning}{Character vector. Columns to partition by. Default "month".}
+
+\item{existing_data_behavior}{Character. Behavior when data exists. Default "delete_matching".}
+
+\item{verbose}{Logical. Whether to print progress messages. Default TRUE.}
+}
+\value{
+Character. The output prefix path for downstream targets.
+}
+\description{
+Reads the primary BAF dataset and engineers new features, such as
+\code{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.
+}