Package jsat.lossfunctions

Interface Summary

Interface	Description
LossC	Specifies a loss function for binary classification problems.
LossFunc	Provides a generic interface for some loss function on some problem that can be described with a single real prediction value and a single real expected value.
LossMC	Specifies a loss function for multi-class problems.
LossR	Specifies a getLoss function for regression problems.

Class Summary

Class	Description
AbsoluteLoss	The AbsoluteLoss loss function for regression L(x, y) = |x-y|.
EpsilonInsensitiveLoss	The ε-insensitive loss for regression L(x, y) = max(0, |x-y|-ε) is the common loss function used for Support Vector Regression.
HingeLoss	The HingeLoss loss function for classification L(x, y) = max(0, 1-y*x) .
HuberLoss	The HuberLoss loss function for regression.
LogisticLoss	The LogisticLoss loss function for classification L(x, y) = log(1+exp(-y*x)).
SoftmaxLoss	The Softmax loss function is a multi-class generalization of the Logistic loss.
SquaredLoss	The SquaredLoss loss function for regression L(x, y) = (x-y)2.