Package de.jungblut.math.activation

Interface Summary

Interface	Description
ActivationFunction	Squashing function interface to provide multiple activation functions, e.G.

Class Summary

Class	Description
AbstractActivationFunction	Implements the boiler plate code for applying functions on container classes like vectors and matrices by applying the function on every element.
ElliotActivationFunction	Implementation of the elliot activation function.
LinearActivationFunction	Linear activation function.
LogActivationFunction	Log activation function, guarded against NaN and infinity edge cases.
ReluActivationFunction	Rectified linear units implementation.
SigmoidActivationFunction	Implementation of the sigmoid function.
SoftMaxActivationFunction	Softmax activation that only works on vectors, because it needs to sum and divide the probabilities.
SoftplusReluActivationFunction	Smoothed approximation to a ReluActivationFunction.
StepActivationFunction	Classic perceptron-like step function.
TanhActivationFunction	Implementation of the Tanh activation based on FastMath.

Enum Summary

Enum	Description
ActivationFunctionSelector	Singleton helper to get the activation functions as singleton.