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 | ResourceApplyCenteredRmsProp.Options | Optional attributes for ResourceApplyCenteredRmsProp
|
Constants
String | OP_NAME | The name of this op, as known by TensorFlow core engine |
Public Methods
static <T extends TType> ResourceApplyCenteredRmsProp | |
static ResourceApplyCenteredRmsProp.Options |
useLocking(Boolean useLocking)
|
Inherited Methods
Constants
public static final String OP_NAME
The name of this op, as known by TensorFlow core engine
Public Methods
public static ResourceApplyCenteredRmsProp create (Scope scope, Operand<?> var, Operand<?> mg, Operand<?> ms, Operand<?> 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 ResourceApplyCenteredRmsProp 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 ResourceApplyCenteredRmsProp
public static ResourceApplyCenteredRmsProp.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. |
---|