Class GradientBoostedRegressor

java.lang.Object

com.quantfinlib.ml.GradientBoostedRegressor

public final class GradientBoostedRegressor
extends Object

Gradient-boosted regression over decision stumps (XGBoost-style additive boosting with squared-error loss), implemented in pure Java with no dependencies. Suited to small/medium tabular problems such as risk forecasting features.

Constructor Summary

Constructors

Constructor	Description
GradientBoostedRegressor(int rounds, double learningRate)

Method Summary

Modifier and Type	Method	Description
GradientBoostedRegressor	fit(double[][] x, double[] y)	Fits the model on x[sample][feature] / y[sample].
double	predict(double[] x)
double[]	predictAll(double[][] x)
double	rmse(double[][] x, double[] y)	Root mean squared error on a labeled set.
static GradientBoostedRegressor	withDefaults()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

GradientBoostedRegressor

public GradientBoostedRegressor(int rounds,
 double learningRate)

Method Details

withDefaults

public static GradientBoostedRegressor withDefaults()

fit

public GradientBoostedRegressor fit(double[][] x,
 double[] y)

Fits the model on x[sample][feature] / y[sample].

predict

public double predict(double[] x)

predictAll

public double[] predictAll(double[][] x)

rmse

public double rmse(double[][] x,
 double[] y)

Root mean squared error on a labeled set.