Contents

BNNS.ActivationFunction.clampedLeakyRectifiedLinear(alpha:beta:)

An activation function that returns its input clamped to beta when that is greater than or equal to zero, otherwise it returns its input multiplied by alpha clamped to beta.

Declaration

case clampedLeakyRectifiedLinear(alpha: Float, beta: Float)

Discussion

This constant defines an activation function that returns values using the following operation:

if x < 0
    min(alpha*x, beta)
else
    min(x, beta)

The following illustrates the output that the activation function generates from inputs in the range -10...10, an alpha of 1 and a beta of 5:

[Image]

See Also

Activation Functions