Python tensorflow.python.training.training_ops.apply_rms_prop() Examples

The following are 15 code examples of tensorflow.python.training.training_ops.apply_rms_prop(). You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may also want to check out all available functions/classes of the module tensorflow.python.training.training_ops , or try the search function .
Example #1
Source File: rmsprop.py    From lambda-packs with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    if self._centered:
      mg = self.get_slot(var, "mg")
      return training_ops.apply_centered_rms_prop(
          var,
          mg,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op
    else:
      return training_ops.apply_rms_prop(
          var,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op
Example #2
Source File: rmsprop.py    From auto-alt-text-lambda-api with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    if self._centered:
      mg = self.get_slot(var, "mg")
      return training_ops.apply_centered_rms_prop(
          var,
          mg,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op
    else:
      return training_ops.apply_rms_prop(
          var,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op
Example #3
Source File: rmsprop_applier.py    From icra2017-visual-navigation with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    return training_ops.apply_rms_prop(
      var, rms, mom,
      self._learning_rate_tensor,
      self._decay_tensor,
      self._momentum_tensor,
      self._epsilon_tensor,
      grad,
      use_locking=False).op

  # Apply accumulated gradients to var.
Example #4
Source File: rmsprop.py    From deep_image_model with Apache License 2.0 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    if self._centered:
      mg = self.get_slot(var, "mg")
      return training_ops.apply_centered_rms_prop(
          var,
          mg,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op
    else:
      return training_ops.apply_rms_prop(
          var,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op
Example #5
Source File: rmsprop_applier.py    From a3c-distributed_tensorflow with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    return training_ops.apply_rms_prop(
      var, rms, mom,
      self._learning_rate_tensor,
      self._decay_tensor,
      self._momentum_tensor,
      self._epsilon_tensor,
      grad,
      use_locking=False).op

  # Apply accumulated gradients to var.
Example #6
Source File: rmsprop_applier.py    From a3c-distributed_tensorflow with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    return training_ops.apply_rms_prop(
      var, rms, mom,
      self._learning_rate_tensor,
      self._decay_tensor,
      self._momentum_tensor,
      self._epsilon_tensor,
      grad,
      use_locking=False).op

  # Apply accumulated gradients to var.
Example #7
Source File: rmsprop_applier.py    From Deep-RL-agents with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
        rms = self.get_slot(var, "rms")
        mom = self.get_slot(var, "momentum")
        return training_ops.apply_rms_prop(
            var, rms, mom,
            self._learning_rate_tensor,
            self._decay_tensor,
            self._momentum_tensor,
            self._epsilon_tensor,
            grad,
            use_locking=False).op

    # Apply accumulated gradients to var.
Example #8
Source File: rmsprop_applier.py    From Deep-RL-agents with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
        rms = self.get_slot(var, "rms")
        mom = self.get_slot(var, "momentum")
        return training_ops.apply_rms_prop(
            var, rms, mom,
            self._learning_rate_tensor,
            self._decay_tensor,
            self._momentum_tensor,
            self._epsilon_tensor,
            grad,
            use_locking=False).op

    # Apply accumulated gradients to var.
Example #9
Source File: rmsprop_applier.py    From Deep-RL-agents with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
        rms = self.get_slot(var, "rms")
        mom = self.get_slot(var, "momentum")
        return training_ops.apply_rms_prop(
            var, rms, mom,
            self._learning_rate_tensor,
            self._decay_tensor,
            self._momentum_tensor,
            self._epsilon_tensor,
            grad,
            use_locking=False).op

    # Apply accumulated gradients to var.
Example #10
Source File: RMSPropApplier.py    From Deep-RL-agents with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
        rms = self.get_slot(var, "rms")
        mom = self.get_slot(var, "momentum")
        return training_ops.apply_rms_prop(
            var, rms, mom,
            self._learning_rate_tensor,
            self._decay_tensor,
            self._momentum_tensor,
            self._epsilon_tensor,
            grad,
            use_locking=False).op

    # Apply accumulated gradients to var.
Example #11
Source File: rmsprop_applier.py    From async_deep_reinforce with Apache License 2.0 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    return training_ops.apply_rms_prop(
      var, rms, mom,
      self._learning_rate_tensor,
      self._decay_tensor,
      self._momentum_tensor,
      self._epsilon_tensor,
      grad,
      use_locking=False).op

  # Apply accumulated gradients to var.
Example #12
Source File: rmsprop_applier.py    From pathnet with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    return training_ops.apply_rms_prop(
      var, rms, mom,
      self._learning_rate_tensor,
      self._decay_tensor,
      self._momentum_tensor,
      self._epsilon_tensor,
      grad,
      use_locking=False).op

  # Apply accumulated gradients to var.
Example #13
Source File: rmsprop_applier.py    From thor-iqa-cvpr-2018 with Apache License 2.0 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    return training_ops.apply_rms_prop(
      var, rms, mom,
      self._learning_rate_tensor,
      self._decay_tensor,
      self._momentum_tensor,
      self._epsilon_tensor,
      grad,
      use_locking=False).op

  # Apply accumulated gradients to var.
Example #14
Source File: rmsprop.py    From Serverless-Deep-Learning-with-TensorFlow-and-AWS-Lambda with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    if self._centered:
      mg = self.get_slot(var, "mg")
      return training_ops.apply_centered_rms_prop(
          var,
          mg,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op
    else:
      return training_ops.apply_rms_prop(
          var,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op
Example #15
Source File: rmsprop.py    From keras-lambda with MIT License 5 votes vote down vote up 
def _apply_dense(self, grad, var):
    rms = self.get_slot(var, "rms")
    mom = self.get_slot(var, "momentum")
    if self._centered:
      mg = self.get_slot(var, "mg")
      return training_ops.apply_centered_rms_prop(
          var,
          mg,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op
    else:
      return training_ops.apply_rms_prop(
          var,
          rms,
          mom,
          math_ops.cast(self._learning_rate_tensor, var.dtype.base_dtype),
          math_ops.cast(self._decay_tensor, var.dtype.base_dtype),
          math_ops.cast(self._momentum_tensor, var.dtype.base_dtype),
          math_ops.cast(self._epsilon_tensor, var.dtype.base_dtype),
          grad,
          use_locking=self._use_locking).op