XGBoost🔗ℹ

Racket bindings for XGBoost, the gradient-boosting library.

xgboost is the high-level API for ordinary Racket data: build a DMatrix, train a booster, predict, and persist the model. It accepts lists, vectors, and f32vectors for common training and prediction workflows while keeping native XGBoost handles behind opaque Racket values.

For lower-level access, use xgboost/foreign, a contracted wrapper that exposes additional DMatrix constructors, metadata and dataset operations, booster inspection, CPU in-place prediction, and custom-objective training. For direct C FFI bindings, use xgboost/foreign/raw.

This documentation has three parts: a task-oriented User Guide that mirrors the upstream XGBoost tutorials, a set of literate Examples whose prose and code are the same source the test suite runs, and a complete API Reference.