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
Args
var
A Tensor of type resource. Should be from a Variable().
accum
A Tensor of type resource. Should be from a Variable().
lr
A Tensor. Must be one of the following types: float32, float64, int32, uint8, int16, int8, complex64, int64, qint8, quint8, qint32, bfloat16, qint16, quint16, uint16, complex128, half, uint32, uint64.
Learning rate. Must be a scalar.
grad
A Tensor. Must have the same type as lr. The gradient.
indices
A Tensor. Must be one of the following types: int32, int64.
A vector of indices into the first dimension of var and accum.
momentum
A Tensor. Must have the same type as lr.
Momentum. Must be a scalar.
use_locking
An optional bool. Defaults to False.
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.
use_nesterov
An optional bool. Defaults to False.
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.