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.

References

1. John Platt. Probabilistic Outputs for Support Vector Machines and Comparisons to Regularized Likelihood Methods. Advances in large margin classifiers. 10 (3): 61–74.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
PlattScaling(double alpha, double beta) Constructor.

Method Summary

Modifier and Type	Method and Description
static <T> PlattScaling	fit(Classifier<T> model, T[] x, int[] y) Fits Platt Scaling to estimate posteriori probabilities.
static PlattScaling	fit(double[] scores, int[] y) Trains the Platt scaling.
static PlattScaling	fit(double[] scores, int[] y, int maxIters) Trains the Platt scaling.
double	scale(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 alpha,
                    double beta)

Constructor. P(y = 1 | x) = 1 / (1 + exp(alpha * f(x) + beta))

Parameters:

alpha - The scaling parameter.

beta - The scaling parameter.

Method Detail

scale

public double scale(double y)

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

Parameters:

y - the binary classifier output score.

Returns:

the estimated probability.

fit

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

Trains the Platt scaling.

Parameters:

scores - The predicted scores.

y - The training labels.

fit

public static PlattScaling fit(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.

fit

public static <T> PlattScaling fit(Classifier<T> model,
                                   T[] x,
                                   int[] y)

Fits Platt Scaling to estimate posteriori probabilities.

Parameters:

model - the binary-class model to fit Platt scaling.

x - training samples.

y - training labels.