Package com.jml.optimizers

Class Optimizer

java.lang.Object

com.jml.optimizers.Optimizer

Direct Known Subclasses:

Adam, GradientDescent, Momentum

public abstract class Optimizer
extends Object

Interface for an optimizer which can be used to minimize a function by utilizing the gradients of that function.

Field Summary

Fields

Modifier and Type	Field	Description
String	name
Scheduler	schedule

Constructor Summary

Constructors

Constructor	Description
Optimizer()

Method Summary

Modifier and Type	Method	Description
abstract String	getDetails()	Gets the details of this optimizer.
double	getLearningRate()	Gets the learning rate for this optimizer.
void	setScheduler(Scheduler schedule)	Sets the learning rate scheduler for this optimizer.
abstract linalg.Matrix[]	step(boolean flag, linalg.Matrix... params)
abstract linalg.Matrix[]	step(linalg.Matrix... params)	Steps the optimizer a single iteration by applying the update rule of the optimizer to the matrix w.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

schedule

public Scheduler schedule

name

public String name

Constructor Details

Optimizer

public Optimizer()

Method Details

step

public abstract linalg.Matrix[] step(linalg.Matrix... params)

Steps the optimizer a single iteration by applying the update rule of the optimizer to the matrix w.

Parameters:

params - An array of Matrices strictly containing the relevant parameters for that optimizer.

Returns:

The result of applying the update rule of the optimizer to the matrix w.

step

public abstract linalg.Matrix[] step(boolean flag,
 linalg.Matrix... params)

getDetails

public abstract String getDetails()

Gets the details of this optimizer.

Returns:

Important details of this optimizer as a string.

setScheduler

public void setScheduler(Scheduler schedule)

Sets the learning rate scheduler for this optimizer.

Parameters:

schedule - Learning rate scheduler.

getLearningRate

public double getLearningRate()

Gets the learning rate for this optimizer.

Returns:

The learning rate for this optimizer.