# Fold-safe ML Pipelines

Reproducible tabular classification for Kaggle Playground Series S6E8.

## Touhidul Alam Seyam's contribution

Created nested target encoding, matched-fold screening, five-fold out-of-fold evaluation, schema checks, and percentile-rank blends across LightGBM, XGBoost, and CatBoost.

- Status: ML pipeline · 2026
- Stack: Python, LightGBM, XGBoost, CatBoost
- [Primary artifact](https://github.com/Seyamalam/playground-series-s6e8)
- [All projects](https://seyamalam.vercel.app/projects)
- [Contact](https://seyamalam.vercel.app/contact)