jsat.lossfunctions

Interface LossMC

All Superinterfaces:

LossC, LossFunc, java.io.Serializable

All Known Implementing Classes:

HingeLoss, SoftmaxLoss

public interface LossMC
extends LossC

Specifies a loss function for multi-class problems. A multi-class loss function must support a raw vector of scores for each class, where positive values indicate preference for the class associated with the same index.

Calling process(jsat.linear.Vec, jsat.linear.Vec) on the raw scores is a mandatory first step, and will transform the raw scores into a usable form for the loss function.

Method Summary

Modifier and Type	Method and Description
void	deriv(Vec processed, Vec derivs, int y) Computes the derivatives with respect to each output processed and derivs may be the same object, and will simply have all its values altered if so.
CategoricalResults	getClassification(Vec processed) Given the processed predictions, returns the classification results for said predictions.
double	getLoss(Vec processed, int y) Computes the scalar loss for on the given example
void	process(Vec pred, Vec processed) Given the vector of raw outputs for each class, transform it into a new vector.

Methods inherited from interface jsat.lossfunctions.LossC

clone, getClassification, getDeriv, getDeriv2, getLoss

Methods inherited from interface jsat.lossfunctions.LossFunc

getConjugate, getDeriv2Max, lipschitz

Method Detail

getLoss

double getLoss(Vec processed,
               int y)

Computes the scalar loss for on the given example

Parameters:

processed - the vector of raw predictions.

y - the true class label in [0, k-1] for k classes

Returns:

the loss in [0, Inf)

process

void process(Vec pred,
             Vec processed)

Given the vector of raw outputs for each class, transform it into a new vector.
processed and derivs may be the same object, and will simply have all its values altered if so.

Parameters:

pred - the vector of raw predictions

processed - the location to store the processed predictions.

deriv

void deriv(Vec processed,
           Vec derivs,
           int y)

Computes the derivatives with respect to each output
processed and derivs may be the same object, and will simply have all its values altered if so.

Parameters:

processed - the processed predictions

derivs - the vector to place the derivative of the loss to.

y - the true class label in [0, k-1] for k classes

getClassification

CategoricalResults getClassification(Vec processed)

Given the processed predictions, returns the classification results for said predictions.

Parameters:

processed - the processed score/prediction vector

Returns:

the classification results