T
- data type for out()
output@Operator(group="train") public final class SparseApplyMomentum<T> extends PrimitiveOp implements Operand<T>
Set use_nesterov = True if you want to use Nesterov momentum.
That is for rows we have grad for, we update var and accum as follows:
$$accum = accum * momentum + grad$$ $$var -= lr * accum$$
Modifier and Type | Class and Description |
---|---|
static class |
SparseApplyMomentum.Options
Optional attributes for
SparseApplyMomentum |
operation
Modifier and Type | Method and Description |
---|---|
Output<T> |
asOutput()
Returns the symbolic handle of a tensor.
|
static <T,U extends Number> |
create(Scope scope,
Operand<T> var,
Operand<T> accum,
Operand<T> lr,
Operand<T> grad,
Operand<U> indices,
Operand<T> momentum,
SparseApplyMomentum.Options... options)
Factory method to create a class wrapping a new SparseApplyMomentum operation.
|
Output<T> |
out()
Same as "var".
|
static SparseApplyMomentum.Options |
useLocking(Boolean useLocking) |
static SparseApplyMomentum.Options |
useNesterov(Boolean useNesterov) |
equals, hashCode, op, toString
public static <T,U extends Number> SparseApplyMomentum<T> create(Scope scope, Operand<T> var, Operand<T> accum, Operand<T> lr, Operand<T> grad, Operand<U> indices, Operand<T> momentum, SparseApplyMomentum.Options... options)
scope
- current scopevar
- Should be from a Variable().accum
- Should be from a Variable().lr
- Learning rate. Must be a scalar.grad
- The gradient.indices
- A vector of indices into the first dimension of var and accum.momentum
- Momentum. Must be a scalar.options
- carries optional attributes valuespublic static SparseApplyMomentum.Options useLocking(Boolean useLocking)
useLocking
- If `True`, updating of the var and accum tensors will be protected
by a lock; otherwise the behavior is undefined, but may exhibit less
contention.public static SparseApplyMomentum.Options useNesterov(Boolean useNesterov)
useNesterov
- If `True`, the tensor passed to compute grad will be
var - lr * momentum * accum, so in the end, the var you get is actually
var - lr * momentum * accum.public Output<T> asOutput()
Operand
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.
asOutput
in interface Operand<T>
OperationBuilder.addInput(Output)
Copyright © 2022. All rights reserved.