Python nn.batch_normalization方法代码示例

# 需要导入模块: from tensorflow.python.ops import nn [as 别名]
# 或者: from tensorflow.python.ops.nn import batch_normalization [as 别名]
def batch_normalization(x, mean, var, beta, gamma, epsilon=1e-3):
  """Applies batch normalization on x given mean, var, beta and gamma.

  I.e. returns:
  `output = (x - mean) / (sqrt(var) + epsilon) * gamma + beta`

  Arguments:
      x: Input tensor or variable.
      mean: Mean of batch.
      var: Variance of batch.
      beta: Tensor with which to center the input.
      gamma: Tensor by which to scale the input.
      epsilon: Fuzz factor.

  Returns:
      A tensor.
  """
  return nn.batch_normalization(x, mean, var, beta, gamma, epsilon)


# SHAPE OPERATIONS 

# 需要导入模块: from tensorflow.python.ops import nn [as 别名]
# 或者: from tensorflow.python.ops.nn import batch_normalization [as 别名]
def normalize(self, inputs):
    """Apply normalization to input.

    The shape must match the declared shape in the constructor.
    [This is copied from tf.contrib.rnn.LayerNormBasicLSTMCell.]

    Args:
      inputs: Input tensor

    Returns:
      Normalized version of input tensor.

    Raises:
      ValueError: if inputs has undefined rank.
    """
    inputs_shape = inputs.get_shape()
    inputs_rank = inputs_shape.ndims
    if inputs_rank is None:
      raise ValueError('Inputs %s has undefined rank.' % inputs.name)
    axis = range(1, inputs_rank)

    beta = self._component.get_variable('beta_%s' % self._name)
    gamma = self._component.get_variable('gamma_%s' % self._name)

    with tf.variable_scope('layer_norm_%s' % self._name):
      # Calculate the moments on the last axis (layer activations).
      mean, variance = nn.moments(inputs, axis, keep_dims=True)

      # Compute layer normalization using the batch_normalization function.
      variance_epsilon = 1E-12
      outputs = nn.batch_normalization(
          inputs, mean, variance, beta, gamma, variance_epsilon)
      outputs.set_shape(inputs_shape)
      return outputs 

# 需要导入模块: from tensorflow.python.ops import nn [as 别名]
# 或者: from tensorflow.python.ops.nn import batch_normalization [as 别名]
def normalize_batch_in_training(x, gamma, beta, reduction_axes, epsilon=1e-3):
  """Computes mean and std for batch then apply batch_normalization on batch.

  Arguments:
      x: Input tensor or variable.
      gamma: Tensor by which to scale the input.
      beta: Tensor with which to center the input.
      reduction_axes: iterable of integers,
          axes over which to normalize.
      epsilon: Fuzz factor.

  Returns:
      A tuple length of 3, `(normalized_tensor, mean, variance)`.
  """
  mean, var = nn.moments(
      x, reduction_axes, shift=None, name=None, keep_dims=False)
  if sorted(reduction_axes) == list(range(ndim(x)))[:-1]:
    normed = nn.batch_normalization(x, mean, var, beta, gamma, epsilon)
  else:
    # need broadcasting
    target_shape = []
    for axis in range(ndim(x)):
      if axis in reduction_axes:
        target_shape.append(1)
      else:
        target_shape.append(array_ops.shape(x)[axis])
    target_shape = array_ops.stack(target_shape)

    broadcast_mean = array_ops.reshape(mean, target_shape)
    broadcast_var = array_ops.reshape(var, target_shape)
    if gamma is None:
      broadcast_gamma = None
    else:
      broadcast_gamma = array_ops.reshape(gamma, target_shape)
    if beta is None:
      broadcast_beta = None
    else:
      broadcast_beta = array_ops.reshape(beta, target_shape)
    normed = nn.batch_normalization(x, broadcast_mean, broadcast_var,
                                    broadcast_beta, broadcast_gamma, epsilon)
  return normed, mean, var 

# 需要导入模块: from tensorflow.python.ops import nn [as 别名]
# 或者: from tensorflow.python.ops.nn import batch_normalization [as 别名]
def normalize(self, inputs):
    """Apply normalization to input.

    The shape must match the declared shape in the constructor.
    [This is copied from tf.contrib.rnn.LayerNormBasicLSTMCell.]

    Args:
      inputs: Input tensor

    Returns:
      Normalized version of input tensor.

    Raises:
      ValueError: if inputs has undefined rank.
    """
    inputs_shape = inputs.get_shape()
    inputs_rank = inputs_shape.ndims
    if inputs_rank is None:
      raise ValueError('Inputs %s has undefined rank.' % inputs.name)
    axis = range(1, inputs_rank)

    beta = self._component.get_variable('beta_%s' % self._name)
    gamma = self._component.get_variable('gamma_%s' % self._name)

    with tf.variable_scope('layer_norm_%s' % self._name):
      # Calculate the moments on the last axis (layer activations).
      mean, variance = nn.moments(inputs, axis, keep_dims=True)

      # Compute layer normalization using the batch_normalization function.
      variance_epsilon = 1E-12
      outputs = nn.batch_normalization(inputs, mean, variance, beta, gamma,
                                       variance_epsilon)
      outputs.set_shape(inputs_shape)
      return outputs