Optimizer

public abstract class Optimizer

Known Direct Subclasses AdaDelta, AdaGrad, AdaGradDA, Adam, Adamax, Ftrl, GradientDescent, Momentum, Nadam, RMSProp AdaDelta Optimizer that implements the Adadelta algorithm.  AdaGrad Optimizer that implements the Adagrad algorithm.  AdaGradDA Optimizer that implements the Adagrad Dual-Averaging algorithm.  Adam Optimizer that implements the Adam algorithm.  Adamax Optimizer that implements the Adamax algorithm.  Ftrl Optimizer that implements the FTRL algorithm.  GradientDescent Basic Stochastic gradient descent optimizer.  Momentum Stochastic gradient descent plus momentum, either nesterov or traditional.  Nadam Nadam Optimizer that implements the NAdam algorithm.  RMSProp Optimizer that implements the RMSProp algorithm.

Base class for gradient optimizers.

Nested Classes

class	Optimizer.GradAndVar<T extends TType>		A class that holds a paired gradient and variable.
class	Optimizer.Options		Optional attributes for Optimizer

Constants

String

VARIABLE_V2

Public Methods

Op	applyGradients(List<GradAndVar<? extends TType>> gradsAndVars, String name) Applies gradients to variables
<T extends TType> List<GradAndVar<?>>	computeGradients(Operand<?> loss) Computes the gradients based on a loss operand.
static String	createName(Output<? extends TType> variable, String slotName) Creates a name by combining a variable name and a slot name
abstract String	getOptimizerName() Get the Name of the optimizer.
<T extends TType> Optional<Variable<T>>	getSlot(Output<T> var, String slotName) Gets the slot associated with the specified variable and slot name.
final Ops	getTF() Gets the Optimizer's Ops instance
Op	minimize(Operand<?> loss) Minimizes the loss by updating the variables
Op	minimize(Operand<?> loss, String name) Minimizes the loss by updating the variables

Inherited Methods

From class java.lang.Object

boolean	equals(Object arg0)
final Class<?>	getClass()
int	hashCode()
final void	notify()
final void	notifyAll()
String	toString()
final void	wait(long arg0, int arg1)
final void	wait(long arg0)
final void	wait()

Constants

public static final String VARIABLE_V2

Constant Value: "VariableV2"

Public Methods

public Op applyGradients (List<GradAndVar<? extends TType>> gradsAndVars, String name)

Applies gradients to variables

Parameters

gradsAndVars	the list of (gradient, variable) pairs.
name	the name of the apply gradients operation

Returns

• an Op that applies the gradients to the variables.

public List<GradAndVar<?>> computeGradients (Operand<?> loss)

Computes the gradients based on a loss operand.

Parameters

loss	the loss operation

Returns

• the computed gradients

public static String createName (Output<? extends TType> variable, String slotName)

Creates a name by combining a variable name and a slot name

Parameters

variable	the variable
slotName	the name of the slot

Returns

• the combined name

public abstract String getOptimizerName ()

Get the Name of the optimizer.

Returns

• The optimizer name.

public Optional<Variable<T>> getSlot (Output<T> var, String slotName)

Gets the slot associated with the specified variable and slot name.

Parameters

var	The variable to lookup.
slotName	The slot name.

Returns

• The slot or empty().

public final Ops getTF ()

Gets the Optimizer's Ops instance

Returns

• the Optimizer's Ops instance

public Op minimize (Operand<?> loss)

Minimizes the loss by updating the variables

Parameters

loss	the loss operation that returns the value to minimize

Returns

• returns op that minimizes the loss by updating the listed variables

public Op minimize (Operand<?> loss, String name)

Minimizes the loss by updating the variables

Parameters

loss	the loss operation that returns the value to minimize
name	the name for the minimize operation

Returns

• op that minimizes the loss by updating the listed variables