TimeFunction

All Superinterfaces:

java.io.Serializable
```
public interface TimeFunction
extends java.io.Serializable
```
A time-dependent function. When training a neural network model, it is often recommended to lower the learning rate as the training progresses. Besides the learning rate schedule, it may also be used for 1-dimensional neighborhood function, etc.

Method Summary

All Methods Static Methods Instance Methods Abstract Methods Default Methods
Modifier and Type	Method and Description
`double`	`apply(int t)` Returns the function value at time step t.
`static TimeFunction`	`constant(double alpha)` Returns the constant learning rate.
`static TimeFunction`	`exp(double initLearningRate, double decaySteps)` Returns the exponential decay function `initLearningRate * exp(-t / decaySteps)`.
`static TimeFunction`	`exp(double initLearningRate, double decaySteps, double endLearningRate)` Returns the exponential decay function `initLearningRate * pow(endLearningRate / initLearningRate, min(t, decaySteps) / decaySteps)`.
`static TimeFunction`	`exp(double initLearningRate, double decaySteps, double decayRate, boolean staircase)` Returns the exponential decay function `initLearningRate * pow(decayRate, t / decaySteps)`.
`static TimeFunction`	`inverse(double initLearningRate, double decaySteps)` Returns the inverse decay function `initLearningRate * decaySteps / (t + decaySteps)`.
`static TimeFunction`	`inverse(double initLearningRate, double decaySteps, double decayRate)` Returns the inverse decay function `initLearningRate / (1 + decayRate * t / decaySteps)`.
`static TimeFunction`	`inverse(double initLearningRate, double decaySteps, double decayRate, boolean staircase)` Returns the inverse decay function `initLearningRate / (1 + decayRate * t / decaySteps)`.
`static TimeFunction`	`linear(double initLearningRate, double decaySteps)` Returns the linear learning rate decay function that ends at 0.0001.
`static TimeFunction`	`linear(double initLearningRate, double decaySteps, double endLearningRate)` Returns the linear learning rate decay function that starts with an initial learning rate and reach an end learning rate in the given decay steps..
`static TimeFunction`	`piecewise(int[] boundaries, double[] values)` Returns the piecewise constant learning rate.
`static TimeFunction`	`polynomial(double initLearningRate, double decaySteps, double endLearningRate, boolean cycle, double power)` Returns the polynomial learning rate decay function that starts with an initial learning rate and reach an end learning rate in the given decay steps.

- Method Detail
  - apply
```
double apply(int t)
```
    Returns the function value at time step t.
    
    Parameters:
    
    t - the discrete time step.
  - constant
```
static TimeFunction constant(double alpha)
```
    Returns the constant learning rate.
    
    Parameters:
    
    alpha - the learning rate.
  - piecewise
```
static TimeFunction piecewise(int[] boundaries,
                              double[] values)
```
    Returns the piecewise constant learning rate. This can be useful for changing the learning rate value across different invocations of optimizer functions.
    
    Parameters:
    
    boundaries - A list of integers with strictly increasing entries.
    
    values - The values for the intervals defined by boundaries. It should have one more element than boundaries.
  - linear
```
static TimeFunction linear(double initLearningRate,
                           double decaySteps)
```
    Returns the linear learning rate decay function that ends at 0.0001.
    
    Parameters:
    
    initLearningRate - the initial learning rate.
    
    decaySteps - the decay steps.
  - linear
```
static TimeFunction linear(double initLearningRate,
                           double decaySteps,
                           double endLearningRate)
```
    Returns the linear learning rate decay function that starts with an initial learning rate and reach an end learning rate in the given decay steps..
    
    Parameters:
    
    initLearningRate - the initial learning rate.
    
    decaySteps - the decay steps.
    
    endLearningRate - the end learning rate.
  - polynomial
```
static TimeFunction polynomial(double initLearningRate,
                               double decaySteps,
                               double endLearningRate,
                               boolean cycle,
                               double power)
```
    Returns the polynomial learning rate decay function that starts with an initial learning rate and reach an end learning rate in the given decay steps. It is commonly observed that a monotonically decreasing learning rate, whose degree of change is carefully chosen, results in a better performing model.
    
    Parameters:
    
    initLearningRate - the initial learning rate.
    
    decaySteps - the decay steps.
    
    endLearningRate - the end learning rate.
    
    cycle - the flag whether or not it should cycle beyond decaySteps.
    
    power - the power of the polynomial.
  - inverse
```
static TimeFunction inverse(double initLearningRate,
                            double decaySteps)
```
    Returns the inverse decay function initLearningRate * decaySteps / (t + decaySteps).
    
    Parameters:
    
    initLearningRate - the initial learning rate.
    
    decaySteps - the decay steps that should be a small percentage of the number of iterations.
  - inverse
```
static TimeFunction inverse(double initLearningRate,
                            double decaySteps,
                            double decayRate)
```
    Returns the inverse decay function initLearningRate / (1 + decayRate * t / decaySteps).
    
    Parameters:
    
    initLearningRate - the initial learning rate.
    
    decaySteps - how often to apply decay.
    
    decayRate - the decay rate.
  - inverse
```
static TimeFunction inverse(double initLearningRate,
                            double decaySteps,
                            double decayRate,
                            boolean staircase)
```
    Returns the inverse decay function initLearningRate / (1 + decayRate * t / decaySteps).
    
    Parameters:
    
    initLearningRate - the initial learning rate.
    
    decaySteps - how often to apply decay.
    
    decayRate - the decay rate.
    
    staircase - the flag whether to apply decay in a discrete staircase, as opposed to continuous, fashion.
  - exp
```
static TimeFunction exp(double initLearningRate,
                        double decaySteps)
```
    Returns the exponential decay function initLearningRate * exp(-t / decaySteps).
    
    Parameters:
    
    initLearningRate - the initial learning rate.
    
    decaySteps - the decay steps that should be a small percentage of the number of iterations.
  - exp
```
static TimeFunction exp(double initLearningRate,
                        double decaySteps,
                        double endLearningRate)
```
    Returns the exponential decay function initLearningRate * pow(endLearningRate / initLearningRate, min(t, decaySteps) / decaySteps).
    
    Parameters:
    
    initLearningRate - the initial learning rate.
    
    decaySteps - the maximum decay steps.
    
    endLearningRate - the end learning rate.
  - exp
```
static TimeFunction exp(double initLearningRate,
                        double decaySteps,
                        double decayRate,
                        boolean staircase)
```
    Returns the exponential decay function initLearningRate * pow(decayRate, t / decaySteps).
    
    Parameters:
    
    initLearningRate - the initial learning rate.
    
    decaySteps - how often to apply decay.
    
    decayRate - the decay rate.
    
    staircase - the flag whether to apply decay in a discrete staircase, as opposed to continuous, fashion.

Interface TimeFunction

Method Summary

Method Detail

apply

constant

piecewise

linear

linear

polynomial

inverse

inverse

inverse

exp

exp

exp