Package | Description |
---|---|
org.bytedeco.tensorflow |
Modifier and Type | Method and Description |
---|---|
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.getPointer(long i) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.getPointer(long i) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.position(long position) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.position(long position) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.use_locking_(boolean setter) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.use_locking_(boolean setter) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.use_nesterov_(boolean setter) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.use_nesterov_(boolean setter) |
static ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.UseLocking(boolean x) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.UseLocking(boolean x)
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. |
static ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.UseLocking(boolean x) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.UseLocking(boolean x)
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. |
static ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.UseNesterov(boolean x) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.UseNesterov(boolean x)
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. |
static ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.UseNesterov(boolean x) |
ResourceSparseApplyMomentum.Attrs |
ResourceSparseApplyMomentum.Attrs.UseNesterov(boolean x)
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. |
Constructor and Description |
---|
ResourceSparseApplyMomentum(Scope scope,
Input var,
Input accum,
Input lr,
Input grad,
Input indices,
Input momentum,
ResourceSparseApplyMomentum.Attrs attrs) |
ResourceSparseApplyMomentum(Scope scope,
Input var,
Input accum,
Input lr,
Input grad,
Input indices,
Input momentum,
ResourceSparseApplyMomentum.Attrs attrs) |
Copyright © 2022. All rights reserved.