ApplyCenteredRmsProp

public final class ApplyCenteredRmsProp

Update '*var' according to the centered RMSProp algorithm.

The centered RMSProp algorithm uses an estimate of the centered second moment (i.e., the variance) for normalization, as opposed to regular RMSProp, which uses the (uncentered) second moment. This often helps with training, but is slightly more expensive in terms of computation and memory.

Note that in dense implementation of this algorithm, mg, ms, and mom will update even if the grad is zero, but in this sparse implementation, mg, ms, and mom will not update in iterations during which the grad is zero.

mean_square = decay * mean_square + (1-decay) * gradient ** 2 mean_grad = decay * mean_grad + (1-decay) * gradient

Delta = learning_rate * gradient / sqrt(mean_square + epsilon - mean_grad ** 2)

mg <- rho * mg_{t-1} + (1-rho) * grad ms <- rho * ms_{t-1} + (1-rho) * grad * grad mom <- momentum * mom_{t-1} + lr * grad / sqrt(ms - mg * mg + epsilon) var <- var - mom

Nested Classes

class ApplyCenteredRmsProp.Options Optional attributes for ApplyCenteredRmsProp  

Constants

String OP_NAME The name of this op, as known by TensorFlow core engine

Public Methods

Output<T>
asOutput()
Returns the symbolic handle of the tensor.
static <T extends TType> ApplyCenteredRmsProp<T>
create(Scope scope, Operand<T> var, Operand<T> mg, Operand<T> ms, Operand<T> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, Options... options)
Factory method to create a class wrapping a new ApplyCenteredRmsProp operation.
Output<T>
out()
Same as "var".
static ApplyCenteredRmsProp.Options
useLocking(Boolean useLocking)

Inherited Methods

org.tensorflow.op.RawOp
final boolean
equals(Object obj)
final int
Operation
op()
Return this unit of computation as a single Operation.
final String
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()
org.tensorflow.op.Op
abstract ExecutionEnvironment
env()
Return the execution environment this op was created in.
abstract Operation
op()
Return this unit of computation as a single Operation.
org.tensorflow.Operand
abstract Output<T>
asOutput()
Returns the symbolic handle of the tensor.
abstract T
asTensor()
Returns the tensor at this operand.
abstract Shape
shape()
Returns the (possibly partially known) shape of the tensor referred to by the Output of this operand.
abstract Class<T>
type()
Returns the tensor type of this operand
org.tensorflow.ndarray.Shaped
abstract int
rank()
abstract Shape
shape()
abstract long
size()
Computes and returns the total size of this container, in number of values.

Constants

public static final String OP_NAME

The name of this op, as known by TensorFlow core engine

Constant Value: "ApplyCenteredRMSProp"

Public Methods

public Output<T> asOutput ()

Returns the symbolic handle of the tensor.

Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.

public static ApplyCenteredRmsProp<T> create (Scope scope, Operand<T> var, Operand<T> mg, Operand<T> ms, Operand<T> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, Options... options)

Factory method to create a class wrapping a new ApplyCenteredRmsProp operation.

Parameters
scope current scope
var Should be from a Variable().
mg Should be from a Variable().
ms Should be from a Variable().
mom Should be from a Variable().
lr Scaling factor. Must be a scalar.
rho Decay rate. Must be a scalar.
momentum Momentum Scale. Must be a scalar.
epsilon Ridge term. Must be a scalar.
grad The gradient.
options carries optional attributes values
Returns
  • a new instance of ApplyCenteredRmsProp

public Output<T> out ()

Same as "var".

public static ApplyCenteredRmsProp.Options useLocking (Boolean useLocking)

Parameters
useLocking If `True`, updating of the var, mg, ms, and mom tensors is protected by a lock; otherwise the behavior is undefined, but may exhibit less contention.