versions/1.6.0/api/r/docs/_sources/api/mx.nd.nag.mom.update.rst

``mx.nd.nag.mom.update``
================================================

Description
----------------------

Update function for Nesterov Accelerated Gradient( NAG) optimizer.
It updates the weights using the following formula,

.. math::

  v_t = \gamma v_{t-1} + \eta * \nabla J(W_{t-1} - \gamma v_{t-1})\\
  W_t = W_{t-1} - v_t

Where 
:math:`\eta` is the learning rate of the optimizer
:math:`\gamma` is the decay rate of the momentum estimate
:math:`\v_t` is the update vector at time step `t`
:math:`\W_t` is the weight vector at time step `t`





Arguments
------------------

+----------------------------------------+------------------------------------------------------------+
| Argument                               | Description                                                |
+========================================+============================================================+
| ``weight``                             | NDArray-or-Symbol.                                         |
|                                        |                                                            |
|                                        | Weight                                                     |
+----------------------------------------+------------------------------------------------------------+
| ``grad``                               | NDArray-or-Symbol.                                         |
|                                        |                                                            |
|                                        | Gradient                                                   |
+----------------------------------------+------------------------------------------------------------+
| ``mom``                                | NDArray-or-Symbol.                                         |
|                                        |                                                            |
|                                        | Momentum                                                   |
+----------------------------------------+------------------------------------------------------------+
| ``lr``                                 | float, required.                                           |
|                                        |                                                            |
|                                        | Learning rate                                              |
+----------------------------------------+------------------------------------------------------------+
| ``momentum``                           | float, optional, default=0.                                |
|                                        |                                                            |
|                                        | The decay rate of momentum estimates at each epoch.        |
+----------------------------------------+------------------------------------------------------------+
| ``wd``                                 | float, optional, default=0.                                |
|                                        |                                                            |
|                                        | Weight decay augments the objective function with a        |
|                                        | regularization term that penalizes large weights. The      |
|                                        | penalty scales with the square of the magnitude of each    |
|                                        | weight.                                                    |
+----------------------------------------+------------------------------------------------------------+
| ``rescale.grad``                       | float, optional, default=1.                                |
|                                        |                                                            |
|                                        | Rescale gradient to grad = rescale_grad*grad.              |
+----------------------------------------+------------------------------------------------------------+
| ``clip.gradient``                      | float, optional, default=-1.                               |
|                                        |                                                            |
|                                        | Clip gradient to the range of [-clip_gradient,             |
|                                        | clip_gradient] If clip_gradient <= 0, gradient clipping is |
|                                        | turned off. grad = max(min(grad, clip_gradient),           |
|                                        | -clip_gradient).                                           |
+----------------------------------------+------------------------------------------------------------+

Value
----------

``out`` The result mx.ndarray


Link to Source Code: http://github.com/apache/incubator-mxnet/blob/1.6.0/src/operator/optimizer_op.cc#L726