2 A complete example🔗ℹ

A full train-and-predict run is short. Build a DMatrix with labels, fit a booster, and predict:

(require xgboost)

(define dtrain

(make-dmatrix '((1.0 2.0 0.5)

(2.0 1.0 1.5)

(3.0 0.5 0.0)

(0.5 3.0 2.0))

#:labels '(3.5 3.5 6.5 2.0)))

(define booster

(train dtrain

#:objective "reg:squarederror"

#:max-depth 2

#:eta 0.2

#:verbosity 0

#:rounds 10))

(predict booster dtrain)

The train/predict pair is the single training interface; there is no separate scikit-learn-style estimator. The sections below break the workflow down step by step.