tensorflow::ops::ApplyPowerSign

Update '*var' according to the AddSign update.

Summary

m_t <- beta1 * m_{t-1} + (1 - beta1) * g update <- exp(logbase * sign_decay * sign(g) * sign(m_t)) * g variable <- variable - lr_t * update

Arguments:

• scope: A Scope object
• var: Should be from a Variable().
• m: Should be from a Variable().
• lr: Scaling factor. Must be a scalar.
• logbase: Must be a scalar.
• sign_decay: Must be a scalar.
• beta: Must be a scalar.
• grad: The gradient.

Optional attributes (see Attrs):

• use_locking: If True, updating of the var and m tensors is protected by a lock; otherwise the behavior is undefined, but may exhibit less contention.

Returns:

• Output: Same as "var".

Public attributes

Operation operation

::tensorflow::Output out

 ApplyPowerSign(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input var,
  ::tensorflow::Input m,
  ::tensorflow::Input lr,
  ::tensorflow::Input logbase,
  ::tensorflow::Input sign_decay,
  ::tensorflow::Input beta,
  ::tensorflow::Input grad
)

 ApplyPowerSign(
  const ::tensorflow::Scope & scope,
  ::tensorflow::Input var,
  ::tensorflow::Input m,
  ::tensorflow::Input lr,
  ::tensorflow::Input logbase,
  ::tensorflow::Input sign_decay,
  ::tensorflow::Input beta,
  ::tensorflow::Input grad,
  const ApplyPowerSign::Attrs & attrs
)

::tensorflow::Node * node() const 

 operator::tensorflow::Input() const 

 operator::tensorflow::Output() const 

Attrs UseLocking(
  bool x
)