smile.classification

Class PlattScaling

java.lang.Object

smile.classification.PlattScaling

All Implemented Interfaces:

java.io.Serializable

public class PlattScaling
extends java.lang.Object
implements java.io.Serializable

Platt scaling or Platt calibration is a way of transforming the outputs of a classification model into a probability distribution over classes. The method was invented by John Platt in the context of support vector machines, but can be applied to other classification models. Platt scaling works by fitting a logistic regression model to a classifier's scores. Platt suggested using the Levenberg–Marquardt algorithm to optimize the parameters, but a Newton algorithm was later proposed that should be more numerically stable, which is implemented in this class.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
PlattScaling(double[] scores, int[] y) Trains the Platt scaling.
PlattScaling(double[] scores, int[] y, int maxIters) Trains the Platt scaling.

Method Summary

Modifier and Type	Method and Description
static void	multiclass(int k, double[][] r, double[] p) Estimates the multiclass probabilies.
double	predict(double y) Returns the posterior probability estimate P(y = 1 | x).

Methods inherited from class java.lang.Object

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

Constructor Detail

PlattScaling

public PlattScaling(double[] scores,
                    int[] y)

Trains the Platt scaling.

Parameters:

scores - The predicted scores.

y - The training labels.

PlattScaling

public PlattScaling(double[] scores,
                    int[] y,
                    int maxIters)

Trains the Platt scaling.

Parameters:

scores - The predicted scores.

y - The training labels.

maxIters - The maximal number of iterations.

Method Detail

predict

public double predict(double y)

Returns the posterior probability estimate P(y = 1 | x).

Parameters:

y - the binary classifier output score.

Returns:

the estimated probability.

multiclass

public static void multiclass(int k,
                              double[][] r,
                              double[] p)

Estimates the multiclass probabilies.