Python math_ops.reduce_sum方法代码示例

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _TileGrad(op, grad):
  """Sum reduces grad along the tiled dimensions."""
  assert isinstance(grad, ops.Tensor)
  input_shape = array_ops.shape(op.inputs[0])
  # We interleave multiples and input_shape to get split_shape,
  # reshape grad to split_shape, and reduce along all even
  # dimensions (the tiled dimensions) to get the result
  # with shape input_shape.  For example
  #   input_shape = [20, 30, 40]
  #   multiples = [2, 3, 4]
  #   split_shape = [2, 20, 3, 30, 4, 40]
  #   axes = [0, 2, 4]
  split_shape = array_ops.reshape(
      array_ops.transpose(array_ops.stack([op.inputs[1], input_shape])), [-1])
  axes = math_ops.range(0, array_ops.size(split_shape), 2)
  input_grad = math_ops.reduce_sum(array_ops.reshape(grad, split_shape), axes)
  # Fix shape inference
  input_grad.set_shape(op.inputs[0].get_shape())
  return [input_grad, None] 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.total_count, dtype=dtypes.int32)
    if self.total_count.get_shape().ndims is not None:
      if self.total_count.get_shape().ndims != 0:
        raise NotImplementedError(
            "Sample only supported for scalar number of draws.")
    elif self.validate_args:
      is_scalar = check_ops.assert_rank(
          n_draws, 0,
          message="Sample only supported for scalar number of draws.")
      n_draws = control_flow_ops.with_dependencies([is_scalar], n_draws)
    k = self.event_shape_tensor()[0]
    # Flatten batch dims so logits has shape [B, k],
    # where B = reduce_prod(self.batch_shape_tensor()).
    draws = random_ops.multinomial(
        logits=array_ops.reshape(self.logits, [-1, k]),
        num_samples=n * n_draws,
        seed=seed)
    draws = array_ops.reshape(draws, shape=[-1, n, n_draws])
    x = math_ops.reduce_sum(array_ops.one_hot(draws, depth=k),
                            axis=-2)  # shape: [B, n, k]
    x = array_ops.transpose(x, perm=[1, 0, 2])
    final_shape = array_ops.concat([[n], self.batch_shape_tensor(), [k]], 0)
    return array_ops.reshape(x, final_shape) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _sample_n(self, n, seed=None):
    n_draws = math_ops.cast(self.total_count, dtype=dtypes.int32)
    k = self.event_shape_tensor()[0]
    unnormalized_logits = array_ops.reshape(
        math_ops.log(random_ops.random_gamma(
            shape=[n],
            alpha=self.concentration,
            dtype=self.dtype,
            seed=seed)),
        shape=[-1, k])
    draws = random_ops.multinomial(
        logits=unnormalized_logits,
        num_samples=n_draws,
        seed=distribution_util.gen_new_seed(seed, salt="dirichlet_multinomial"))
    x = math_ops.reduce_sum(array_ops.one_hot(draws, depth=k), -2)
    final_shape = array_ops.concat([[n], self.batch_shape_tensor(), [k]], 0)
    return array_ops.reshape(x, final_shape) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _kl_categorical_categorical(a, b, name=None):
  """Calculate the batched KL divergence KL(a || b) with a and b Categorical.

  Args:
    a: instance of a Categorical distribution object.
    b: instance of a Categorical distribution object.
    name: (optional) Name to use for created operations.
      default is "kl_categorical_categorical".

  Returns:
    Batchwise KL(a || b)
  """
  with ops.name_scope(name, "kl_categorical_categorical",
                      values=[a.logits, b.logits]):
    # sum(probs log(probs / (1 - probs)))
    delta_log_probs1 = (nn_ops.log_softmax(a.logits) -
                        nn_ops.log_softmax(b.logits))
    return math_ops.reduce_sum(nn_ops.softmax(a.logits) * delta_log_probs1,
                               axis=-1) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _MinOrMaxGrad(op, grad):
  """Gradient for Min or Max. Amazingly it's precisely the same code."""
  input_shape = array_ops.shape(op.inputs[0])
  output_shape_kept_dims = math_ops.reduced_shape(input_shape, op.inputs[1])
  y = op.outputs[0]
  y = array_ops.reshape(y, output_shape_kept_dims)
  grad = array_ops.reshape(grad, output_shape_kept_dims)

  # Compute the number of selected (maximum or minimum) elements in each
  # reduction dimension. If there are multiple minimum or maximum elements
  # then the gradient will be divided between them.
  indicators = math_ops.cast(math_ops.equal(y, op.inputs[0]), grad.dtype)
  num_selected = array_ops.reshape(
      math_ops.reduce_sum(indicators, op.inputs[1]), output_shape_kept_dims)

  return [math_ops.div(indicators, num_selected) * grad, None] 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _BetaincGrad(op, grad):
  """Returns gradient of betainc(a, b, x) with respect to x."""
  # TODO(ebrevdo): Perhaps add the derivative w.r.t. a, b
  a, b, x = op.inputs

  # two cases: x is a scalar and a/b are same-shaped tensors, or vice
  # versa; so its sufficient to check against shape(a).
  sa = array_ops.shape(a)
  sx = array_ops.shape(x)
  # pylint: disable=protected-access
  _, rx = gen_array_ops._broadcast_gradient_args(sa, sx)
  # pylint: enable=protected-access

  # Perform operations in log space before summing, because terms
  # can grow large.
  log_beta = (gen_math_ops.lgamma(a) + gen_math_ops.lgamma(b)
              - gen_math_ops.lgamma(a + b))
  partial_x = math_ops.exp(
      (b - 1) * math_ops.log(1 - x) + (a - 1) * math_ops.log(x) - log_beta)

  # TODO(b/36815900): Mark None return values as NotImplemented
  return (None,  # da
          None,  # db
          array_ops.reshape(math_ops.reduce_sum(partial_x * grad, rx), sx)) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _ZetaGrad(op, grad):
  """Returns gradient of zeta(x, q) with respect to x and q."""
  # TODO(tillahoffmann): Add derivative with respect to x
  x = op.inputs[0]
  q = op.inputs[1]
  # Broadcast gradients
  sx = array_ops.shape(x)
  sq = array_ops.shape(q)
  unused_rx, rq = gen_array_ops._broadcast_gradient_args(sx, sq)
  # Evaluate gradient
  with ops.control_dependencies([grad.op]):
    x = math_ops.conj(x)
    q = math_ops.conj(q)
    partial_q = -x * math_ops.zeta(x + 1, q)
    # TODO(b/36815900): Mark None return values as NotImplemented
    return (None,
            array_ops.reshape(math_ops.reduce_sum(partial_q * grad, rq), sq)) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _RealDivGrad(op, grad):
  """RealDiv op gradient."""
  x = op.inputs[0]
  y = op.inputs[1]
  sx = array_ops.shape(x)
  sy = array_ops.shape(y)
  # pylint: disable=protected-access
  rx, ry = gen_array_ops._broadcast_gradient_args(sx, sy)
  # pylint: enable=protected-access
  x = math_ops.conj(x)
  y = math_ops.conj(y)
  return (array_ops.reshape(
      math_ops.reduce_sum(math_ops.realdiv(grad, y), rx),
      sx), array_ops.reshape(
          math_ops.reduce_sum(grad * math_ops.realdiv(math_ops.realdiv(-x, y), y),
                              ry), sy)) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _MaximumMinimumGrad(op, grad, selector_op):
  """Factor out the code for the gradient of Maximum or Minimum."""
  x = op.inputs[0]
  y = op.inputs[1]
  gdtype = grad.dtype
  sx = array_ops.shape(x)
  sy = array_ops.shape(y)
  gradshape = array_ops.shape(grad)
  zeros = array_ops.zeros(gradshape, gdtype)
  xmask = selector_op(x, y)
  rx, ry = gen_array_ops._broadcast_gradient_args(sx, sy)
  xgrad = array_ops.where(xmask, grad, zeros)
  ygrad = array_ops.where(math_ops.logical_not(xmask), grad, zeros)
  gx = array_ops.reshape(math_ops.reduce_sum(xgrad, rx), sx)
  gy = array_ops.reshape(math_ops.reduce_sum(ygrad, ry), sy)
  return (gx, gy) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _SquaredDifferenceGrad(op, grad):
  """Returns the gradient for (x-y)^2."""
  x = op.inputs[0]
  y = op.inputs[1]
  sx = array_ops.shape(x)
  sy = array_ops.shape(y)
  # pylint: disable=protected-access
  rx, ry = gen_array_ops._broadcast_gradient_args(sx, sy)
  # pylint: enable=protected-access
  # .op works with Tensors or IndexedSlices
  with ops.control_dependencies([grad.op]):
    # The parens ensure that if grad is IndexedSlices, it'll get multiplied by
    # Tensor (not a number like 2.0) which causes it to convert to Tensor.
    x_grad = math_ops.scalar_mul(2.0, grad) * (x - y)
  return (array_ops.reshape(math_ops.reduce_sum(x_grad, rx), sx),
          -array_ops.reshape(math_ops.reduce_sum(x_grad, ry), sy))


# Logical operations have no gradients. 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _attention(self, query, attn_states):
    conv2d = nn_ops.conv2d
    reduce_sum = math_ops.reduce_sum
    softmax = nn_ops.softmax
    tanh = math_ops.tanh

    with vs.variable_scope("attention"):
      k = vs.get_variable(
          "attn_w", [1, 1, self._attn_size, self._attn_vec_size])
      v = vs.get_variable("attn_v", [self._attn_vec_size])
      hidden = array_ops.reshape(attn_states,
                                 [-1, self._attn_length, 1, self._attn_size])
      hidden_features = conv2d(hidden, k, [1, 1, 1, 1], "SAME")
      y = _linear(query, self._attn_vec_size, True)
      y = array_ops.reshape(y, [-1, 1, 1, self._attn_vec_size])
      s = reduce_sum(v * tanh(hidden_features + y), [2, 3])
      a = softmax(s)
      d = reduce_sum(
          array_ops.reshape(a, [-1, self._attn_length, 1, 1]) * hidden, [1, 2])
      new_attns = array_ops.reshape(d, [-1, self._attn_size])
      new_attn_states = array_ops.slice(attn_states, [0, 1, 0], [-1, -1, -1])
      return new_attns, new_attn_states 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _compute_euclidean_distance(cls, inputs, clusters):
    """Computes Euclidean distance between each input and each cluster center.

    Args:
      inputs: list of input Tensors.
      clusters: cluster Tensor.

    Returns:
      list of Tensors, where each element corresponds to each element in inputs.
      The value is the distance of each row to all the cluster centers.
    """
    output = []
    for inp in inputs:
      with ops.colocate_with(inp):
        # Computes Euclidean distance. Note the first and third terms are
        # broadcast additions.
        squared_distance = (math_ops.reduce_sum(
            math_ops.square(inp), 1, keep_dims=True) - 2 * math_ops.matmul(
                inp, clusters, transpose_b=True) + array_ops.transpose(
                    math_ops.reduce_sum(
                        math_ops.square(clusters), 1, keep_dims=True)))
        output.append(squared_distance)

    return output 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _covariance(x, diag):
  """Defines the covariance operation of a matrix.

  Args:
    x: a matrix Tensor. Dimension 0 should contain the number of examples.
    diag: if True, it computes the diagonal covariance.

  Returns:
    A Tensor representing the covariance of x. In the case of
  diagonal matrix just the diagonal is returned.
  """
  num_points = math_ops.to_float(array_ops.shape(x)[0])
  x -= math_ops.reduce_mean(x, 0, keep_dims=True)
  if diag:
    cov = math_ops.reduce_sum(
        math_ops.square(x), 0, keep_dims=True) / (num_points - 1)
  else:
    cov = math_ops.matmul(x, x, transpose_a=True) / (num_points - 1)
  return cov 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _define_full_covariance_probs(self, shard_id, shard):
    """Defines the full covariance probabilties per example in a class.

    Updates a matrix with dimension num_examples X num_classes.

    Args:
      shard_id: id of the current shard.
      shard: current data shard, 1 X num_examples X dimensions.
    """
    diff = shard - self._means
    cholesky = linalg_ops.cholesky(self._covs + self._min_var)
    log_det_covs = 2.0 * math_ops.reduce_sum(
        math_ops.log(array_ops.matrix_diag_part(cholesky)), 1)
    x_mu_cov = math_ops.square(
        linalg_ops.matrix_triangular_solve(
            cholesky, array_ops.transpose(
                diff, perm=[0, 2, 1]), lower=True))
    diag_m = array_ops.transpose(math_ops.reduce_sum(x_mu_cov, 1))
    self._probs[shard_id] = -0.5 * (diag_m + math_ops.to_float(self._dimensions)
                                    * math_ops.log(2 * np.pi) + log_det_covs) 

# 需要导入模块: from tensorflow.python.ops import math_ops [as 别名]
# 或者: from tensorflow.python.ops.math_ops import reduce_sum [as 别名]
def _define_diag_covariance_probs(self, shard_id, shard):
    """Defines the diagonal covariance probabilities per example in a class.

    Args:
      shard_id: id of the current shard.
      shard: current data shard, 1 X num_examples X dimensions.

    Returns a matrix num_examples * num_classes.
    """
    # num_classes X 1
    # TODO(xavigonzalvo): look into alternatives to log for
    # reparametrization of variance parameters.
    det_expanded = math_ops.reduce_sum(
        math_ops.log(self._covs + 1e-3), 1, keep_dims=True)
    diff = shard - self._means
    x2 = math_ops.square(diff)
    cov_expanded = array_ops.expand_dims(1.0 / (self._covs + 1e-3), 2)
    # num_classes X num_examples
    x2_cov = math_ops.matmul(x2, cov_expanded)
    x2_cov = array_ops.transpose(array_ops.squeeze(x2_cov, [2]))
    self._probs[shard_id] = -0.5 * (
        math_ops.to_float(self._dimensions) * math_ops.log(2.0 * np.pi) +
        array_ops.transpose(det_expanded) + x2_cov)