Rob Wiederstein
33d0fc31c7
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>
33 lines
461 B
Plaintext
33 lines
461 B
Plaintext
# --- RStudio & System Basics ---
|
|
.Rproj.user
|
|
.Rhistory
|
|
.RData
|
|
.Ruserdata
|
|
.DS_Store
|
|
|
|
# --- SECURITY: Never commit these ---
|
|
.Renviron
|
|
.Rprofile
|
|
*.secret
|
|
credentials/
|
|
|
|
# --- DATA: The "Lake" Rule ---
|
|
# Since data lives on Smaug/MinIO, we ignore any local data attempts
|
|
data/*
|
|
!data/README.md
|
|
*.csv
|
|
*.parquet
|
|
*.rds
|
|
|
|
# --- WORKFLOW: Targets (for when you switch later) ---
|
|
_targets/
|
|
_targets.user/data/
|
|
.Renviron
|
|
|
|
/.quarto/
|
|
**/*.quarto_ipynb
|
|
docs
|
|
|
|
/dev/
|
|
/scratch/
|