|
JMSLTM Numerical Library 4.0 | ||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
Interface implemented by perceptron activation functions.
Standard activation functions are defined as static members of this
interface. New activation functions can be defined by implementing a method,
g(double x)
, returning the value and a method,
derivative(double x, double y)
, returning the derivative of g
evaluated at x where y = g(x).
Perceptron
Field Summary | |
static Activation |
LINEAR
The identity activation function, g(x) = x. |
static Activation |
LOGISTIC
The logistic activation function, . |
static Activation |
LOGISTIC_TABLE
The logistic activation function computed using a table. |
static long |
serialVersionUID
|
static Activation |
SOFTMAX
The softmax activation function. |
static Activation |
SQUASH
The squash activation function, |
static Activation |
TANH
The hyperbolic tangent activation function, . |
Method Summary | |
double |
derivative(double x,
double y)
Returns the value of the derivative of the activation function. |
double |
g(double x)
Returns the value of the activation function. |
Field Detail |
public static final Activation LINEAR
public static final Activation LOGISTIC
public static final Activation LOGISTIC_TABLE
This version of the logistic function differs from the exact version by at most 4.0e-9.
Networks trained using this activation should not use
Activation.LOGISTIC
for forecasting. Forecasting should be done
using the specific function supplied during training.
public static final long serialVersionUID
public static final Activation SOFTMAX
public static final Activation SQUASH
public static final Activation TANH
Method Detail |
public double derivative(double x, double y)
x
- A double
which specifies the point at which the
activation function is to be evaluated.y
- A double
which specifies y = g(x), the
value of the activation function at x. This parameter
is not mathematically required, but can sometimes be used to
more quickly compute the derivative.
double
containing the value of the derivative of
the activation function at x
.public double g(double x)
x
- A double
is the point at which the activation
function is to be evaluated.
double
containing the value of the activation
function at x
.
|
JMSLTM Numerical Library 4.0 | ||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |