Class Momentum

java.lang.Object

org.tensorflow.framework.optimizers.Optimizer

org.tensorflow.framework.optimizers.Momentum

public class Momentum
extends Optimizer

Stochastic gradient descent plus momentum, either nesterov or traditional.

See the paper for details of nesterov momentum.

Nested Class Summary

Nested classes/interfaces inherited from class Optimizer

Optimizer.GradAndVar<T>, Optimizer.Options

Field Summary

Fields

Modifier and Type	Field	Description
static final float	LEARNING_RATE_DEFAULT
static final String	MOMENTUM
static final float	MOMENTUM_DEFAULT
static final boolean	NESTEROV_DEFAULT

Fields inherited from class Optimizer

globals, graph, tf, VARIABLE_V2

Constructor Summary

Constructors

Constructor	Description
Momentum(Graph graph)	Creates a Momentum Optimizer
Momentum(Graph graph, float learningRate)	Creates a Momentum Optimizer
Momentum(Graph graph, float learningRate, float momentum)	Creates a Momentum Optimizer
Momentum(Graph graph, float learningRate, float momentum, boolean useNesterov)	Creates a Momentum Optimizer
Momentum(Graph graph, String name, float learningRate, float momentum, boolean useNesterov)	Creates a Momentum Optimizer

Method Summary

Modifier and Type	Method	Description
protected <T extends TType> Op	applyDense(Ops deps, Output<T> gradient, Output<T> variable)	Generates the gradient update operations for the specific variable and gradient.
protected void	createSlots(List<Output<? extends TType>> variables)	Performs a No-op slot creation method.
String	getOptimizerName()	Get the Name of the optimizer.
String	toString()

Methods inherited from class Optimizer

applyGradients, computeGradients, createName, createSlot, finish, getSlot, getTF, minimize, minimize, prepare

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Details

LEARNING_RATE_DEFAULT

public static final float LEARNING_RATE_DEFAULT

See Also:

• Constant Field Values

MOMENTUM_DEFAULT

public static final float MOMENTUM_DEFAULT

See Also:

• Constant Field Values

NESTEROV_DEFAULT

public static final boolean NESTEROV_DEFAULT

See Also:

• Constant Field Values

MOMENTUM

public static final String MOMENTUM

See Also:

• Constant Field Values

Constructor Details

Momentum

public Momentum(Graph graph)

Creates a Momentum Optimizer

Parameters:

graph - the TensorFlow graph

Momentum

public Momentum(Graph graph,
 float learningRate)

Creates a Momentum Optimizer

Parameters:

graph - the TensorFlow graph

learningRate - the learning rate

Momentum

public Momentum(Graph graph,
 float learningRate,
 float momentum)

Creates a Momentum Optimizer

Parameters:

graph - the TensorFlow graph

learningRate - the learning rate

momentum - hyperparameter that accelerates gradient descent in the relevant direction and dampens oscillations, Must be greater than or equal to zero. Default is 0.

Momentum

public Momentum(Graph graph,
 float learningRate,
 float momentum,
 boolean useNesterov)

Creates a Momentum Optimizer

Parameters:

graph - the TensorFlow graph

learningRate - the learning rate

momentum - hyperparameter that accelerates gradient descent in the relevant direction and dampens oscillations, Must be greater than or equal to zero. Default is 0.

useNesterov - Whether to apply Nesterov momentum. Defaults to false.

Momentum

public Momentum(Graph graph,
 String name,
 float learningRate,
 float momentum,
 boolean useNesterov)

Creates a Momentum Optimizer

Parameters:

graph - the TensorFlow graph

name - the name for this Optimizer

learningRate - the learning rate

momentum - hyperparameter that accelerates gradient descent in the relevant direction and dampens oscillations, Must be greater than or equal to zero. Default is 0.

useNesterov - Whether to apply Nesterov momentum. Defaults to false.

Method Details

createSlots

protected void createSlots(List<Output<? extends TType>> variables)

Performs a No-op slot creation method.

Overrides:

createSlots in class Optimizer

Parameters:

variables - The variables to create slots for.

applyDense

protected <T extends TType> Op applyDense(Ops deps,
 Output<T> gradient,
 Output<T> variable)

Generates the gradient update operations for the specific variable and gradient.

Specified by:

applyDense in class Optimizer

Type Parameters:

T - The type of the variable.

Parameters:

gradient - The gradient to use.

variable - The variable to update.

Returns:

An operand which applies the desired optimizer update to the variable.

toString

public String toString()

Overrides:

toString in class Object

getOptimizerName

public String getOptimizerName()

Get the Name of the optimizer.

Specified by:

getOptimizerName in class Optimizer

Returns:

The optimizer name.