Class RMSPropOptimizer

tensorflow/python/training/rmsprop.py:54–254

Optimizer that implements the RMSProp algorithm. See the [paper](http://www.cs.toronto.edu/~tijmen/csc321/slides/lecture_slides_lec6.pdf).

@tf_export(v1=["train.RMSPropOptimizer"])
class RMSPropOptimizer(optimizer.Optimizer):
  """Optimizer that implements the RMSProp algorithm.

  See the
  [paper](http://www.cs.toronto.edu/~tijmen/csc321/slides/lecture_slides_lec6.pdf).
  """

  def __init__(self,
               learning_rate,
               decay=0.9,
               momentum=0.0,
               epsilon=1e-10,
               use_locking=False,
               centered=False,
               name="RMSProp"):
    """Construct a new RMSProp optimizer.

    Note that in the dense implementation of this algorithm, variables and their
    corresponding accumulators (momentum, gradient moving average, square
    gradient moving average) will be updated even if the gradient is zero
    (i.e. accumulators will decay, momentum will be applied). The sparse
    implementation (used when the gradient is an `IndexedSlices` object,
    typically because of `tf.gather` or an embedding lookup in the forward pass)
    will not update variable slices or their accumulators unless those slices
    were used in the forward pass (nor is there an "eventual" correction to
    account for these omitted updates). This leads to more efficient updates for
    large embedding lookup tables (where most of the slices are not accessed in
    a particular graph execution), but differs from the published algorithm.

    Args:
      learning_rate: A Tensor or a floating point value.  The learning rate.
      decay: Discounting factor for the history/coming gradient
      momentum: A scalar tensor.
      epsilon: Small value to avoid zero denominator.
      use_locking: If True use locks for update operation.
      centered: If True, gradients are normalized by the estimated variance of
        the gradient; if False, by the uncentered second moment. Setting this to
        True may help with training, but is slightly more expensive in terms of
        computation and memory. Defaults to False.
      name: Optional name prefix for the operations created when applying
        gradients. Defaults to "RMSProp".

    @compatibility(eager)
    When eager execution is enabled, `learning_rate`, `decay`, `momentum`, and
    `epsilon` can each be a callable that takes no arguments and returns the
    actual value to use. This can be useful for changing these values across
    different invocations of optimizer functions.
    @end_compatibility
    """
    super(RMSPropOptimizer, self).__init__(use_locking, name)
    self._learning_rate = learning_rate
    self._decay = decay
    self._momentum = momentum
    self._epsilon = epsilon
    self._centered = centered

    # Tensors for learning rate and momentum.  Created in _prepare.
    self._learning_rate_tensor = None