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/evaluate_final_model.Rd

@@ -0,0 +1,27 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/functions.R
+\name{evaluate_final_model}
+\alias{evaluate_final_model}
+\title{Final Model Evaluation (Months 6 & 7)}
+\usage{
+evaluate_final_model(
+  params,
+  bucket_name = "baf-fraud",
+  inputs_prefix = "05_model_input"
+)
+}
+\arguments{
+\item{params}{A named list of LightGBM hyperparameters with elements:
+\code{trees}, \code{tree_depth}, \code{learn_rate}, \code{loss_reduction}, \code{min_n}.}
+
+\item{bucket_name}{Character. Bucket name. Default "baf-fraud".}
+
+\item{inputs_prefix}{Character. Model input prefix. Default "05_model_input".}
+}
+\value{
+A tibble with columns \code{truth}, \code{prob}, and \code{pred_class}.
+}
+\description{
+Trains the winning strategy on the full training set (Months 0-5)
+and evaluates it on the unseen test set (Months 6-7).
+}