Python tensorflow.python.ops.array_ops.matrix_transpose() Examples

def _CholeskyGrad(op, grad):
  """Gradient for Cholesky."""

  # Gradient is l^{-H} @ ((l^{H} @ grad) * (tril(ones)-1/2*eye)) @ l^{-1}
  l = op.outputs[0]
  num_rows = array_ops.shape(l)[-1]
  batch_shape = array_ops.shape(l)[:-2]
  l_inverse = linalg_ops.matrix_triangular_solve(l,
                                                 linalg_ops.eye(
                                                     num_rows,
                                                     batch_shape=batch_shape,
                                                     dtype=l.dtype))

  middle = math_ops.matmul(l, grad, adjoint_a=True)
  middle = array_ops.matrix_set_diag(middle,
                                     0.5 * array_ops.matrix_diag_part(middle))
  middle = array_ops.matrix_band_part(middle, -1, 0)

  grad_a = math_ops.matmul(
      math_ops.matmul(l_inverse, middle, adjoint_a=True), l_inverse)

  grad_a += math_ops.conj(array_ops.matrix_transpose(grad_a))
  return grad_a * 0.5 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    self._check_x(x)
    return x 

def _batch_sqrt_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return diag_mat * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return math_ops.square(diag_mat) * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return diag_mat * x 

def assert_symmetric(matrix):
  matrix_t = array_ops.matrix_transpose(matrix)
  return control_flow_ops.with_dependencies(
      [check_ops.assert_equal(matrix, matrix_t)], matrix) 

def _batch_sqrt_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    self._check_x(x)
    return math_ops.sqrt(self._scale) * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    self._check_x(x)
    return self._scale * x 

def _SelfAdjointEigV2Grad(op, grad_e, grad_v):
  """Gradient for SelfAdjointEigV2."""
  e = op.outputs[0]
  v = op.outputs[1]
  # a = op.inputs[0], which satisfies
  # a[...,:,:] * v[...,:,i] = e[...,i] * v[...,i]
  with ops.control_dependencies([grad_e.op, grad_v.op]):
    if grad_v is not None:
      # Construct the matrix f(i,j) = (i != j ? 1 / (e_i - e_j) : 0).
      # Notice that because of the term involving f, the gradient becomes
      # infinite (or NaN in practice) when eigenvalues are not unique.
      # Mathematically this should not be surprising, since for (k-fold)
      # degenerate eigenvalues, the corresponding eigenvectors are only defined
      # up to arbitrary rotation in a (k-dimensional) subspace.
      f = array_ops.matrix_set_diag(
          math_ops.reciprocal(
              array_ops.expand_dims(e, -2) - array_ops.expand_dims(e, -1)),
          array_ops.zeros_like(e))
      grad_a = math_ops.matmul(
          v,
          math_ops.matmul(
              array_ops.matrix_diag(grad_e) + f * math_ops.matmul(
                  v, grad_v, adjoint_a=True),
              v,
              adjoint_b=True))
    else:
      grad_a = math_ops.matmul(
          v, math_ops.matmul(
              array_ops.matrix_diag(grad_e), v, adjoint_b=True))
    # The forward op only depends on the lower triangular part of a, so here we
    # symmetrize and take the lower triangle
    grad_a = array_ops.matrix_band_part(
        grad_a + array_ops.matrix_transpose(grad_a), -1, 0)
    grad_a = array_ops.matrix_set_diag(grad_a,
                                       0.5 * array_ops.matrix_diag_part(grad_a))
    return grad_a 

def _SelfAdjointEigV2Grad(op, grad_e, grad_v):
  """Gradient for SelfAdjointEigV2."""
  e = op.outputs[0]
  compute_v = op.get_attr("compute_v")
  # a = op.inputs[0], which satisfies
  # a[...,:,:] * v[...,:,i] = e[...,i] * v[...,i]
  with ops.control_dependencies([grad_e, grad_v]):
    if compute_v:
      v = op.outputs[1]
      # Construct the matrix f(i,j) = (i != j ? 1 / (e_i - e_j) : 0).
      # Notice that because of the term involving f, the gradient becomes
      # infinite (or NaN in practice) when eigenvalues are not unique.
      # Mathematically this should not be surprising, since for (k-fold)
      # degenerate eigenvalues, the corresponding eigenvectors are only defined
      # up to arbitrary rotation in a (k-dimensional) subspace.
      f = array_ops.matrix_set_diag(
          math_ops.reciprocal(
              array_ops.expand_dims(e, -2) - array_ops.expand_dims(e, -1)),
          array_ops.zeros_like(e))
      grad_a = math_ops.matmul(
          v,
          math_ops.matmul(
              array_ops.matrix_diag(grad_e) +
              f * math_ops.matmul(v, grad_v, adjoint_a=True),
              v,
              adjoint_b=True))
    else:
      _, v = linalg_ops.self_adjoint_eig(op.inputs[0])
      grad_a = math_ops.matmul(v,
                               math_ops.matmul(
                                   array_ops.matrix_diag(grad_e),
                                   v,
                                   adjoint_b=True))
    # The forward op only depends on the lower triangular part of a, so here we
    # symmetrize and take the lower triangle
    grad_a = array_ops.matrix_band_part(
        grad_a + math_ops.conj(array_ops.matrix_transpose(grad_a)), -1, 0)
    grad_a = array_ops.matrix_set_diag(grad_a,
                                       0.5 * array_ops.matrix_diag_part(grad_a))
    return grad_a 

def __call__(self, shape, dtype=None, partition_info=None):
    if dtype is None:
      dtype = self.dtype
    # Check the shape
    if len(shape) < 2:
      raise ValueError("The tensor to initialize must be "
                       "at least two-dimensional")
    # Flatten the input shape with the last dimension remaining
    # its original shape so it works for conv2d
    num_rows = 1
    for dim in shape[:-1]:
      num_rows *= dim
    num_cols = shape[-1]
    flat_shape = (num_cols, num_rows) if num_rows < num_cols else (num_rows,
                                                                   num_cols)

    # Generate a random matrix
    a = random_ops.random_normal(flat_shape, dtype=dtype, seed=self.seed)
    # Compute the qr factorization
    q, r = linalg_ops.qr(a, full_matrices=False)
    # Make Q uniform
    d = array_ops.diag_part(r)
    ph = d / math_ops.abs(d)
    q *= ph
    if num_rows < num_cols:
      q = array_ops.matrix_transpose(q)
    return self.gain * array_ops.reshape(q, shape) 

def assert_symmetric(matrix):
  matrix_t = array_ops.matrix_transpose(matrix)
  return control_flow_ops.with_dependencies(
      [check_ops.assert_equal(matrix, matrix_t)], matrix) 

def _batch_sqrt_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return diag_mat * x 

def _batch_sqrt_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return math_ops.sqrt(diag_mat) * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return diag_mat * x 

def assert_symmetric(matrix):
  matrix_t = array_ops.matrix_transpose(matrix)
  return control_flow_ops.with_dependencies(
      [check_ops.assert_equal(matrix, matrix_t)], matrix) 

def assert_symmetric(matrix):
  matrix_t = array_ops.matrix_transpose(matrix)
  return control_flow_ops.with_dependencies(
      [check_ops.assert_equal(matrix, matrix_t)], matrix) 

def _SelfAdjointEigV2Grad(op, grad_e, grad_v):
  """Gradient for SelfAdjointEigV2."""
  e = op.outputs[0]
  v = op.outputs[1]
  # a = op.inputs[0], which satisfies
  # a[...,:,:] * v[...,:,i] = e[...,i] * v[...,i]
  with ops.control_dependencies([grad_e.op, grad_v.op]):
    if grad_v is not None:
      # Construct the matrix f(i,j) = (i != j ? 1 / (e_i - e_j) : 0).
      # Notice that because of the term involving f, the gradient becomes
      # infinite (or NaN in practice) when eigenvalues are not unique.
      # Mathematically this should not be surprising, since for (k-fold)
      # degenerate eigenvalues, the corresponding eigenvectors are only defined
      # up to arbitrary rotation in a (k-dimensional) subspace.
      f = array_ops.matrix_set_diag(
          math_ops.inv(
              array_ops.expand_dims(e, -2) - array_ops.expand_dims(e, -1)),
          array_ops.zeros_like(e))
      grad_a = math_ops.batch_matmul(
          v,
          math_ops.batch_matmul(
              array_ops.matrix_diag(grad_e) + f * math_ops.batch_matmul(
                  v, grad_v, adj_x=True),
              v,
              adj_y=True))
    else:
      grad_a = math_ops.batch_matmul(
          v,
          math_ops.batch_matmul(
              array_ops.matrix_diag(grad_e), v, adj_y=True))
    # The forward op only depends on the lower triangular part of a, so here we
    # symmetrize and take the lower triangle
    grad_a = array_ops.matrix_band_part(
        grad_a + array_ops.matrix_transpose(grad_a), -1, 0)
    grad_a = array_ops.matrix_set_diag(grad_a,
                                       0.5 * array_ops.matrix_diag_part(grad_a))
    return grad_a 

def _batch_sqrt_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return diag_mat * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return math_ops.square(diag_mat) * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return diag_mat * x 

def assert_symmetric(matrix):
  matrix_t = array_ops.matrix_transpose(matrix)
  return control_flow_ops.with_dependencies(
      [check_ops.assert_equal(matrix, matrix_t)], matrix) 

def _batch_sqrt_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    self._check_x(x)
    return math_ops.sqrt(self._scale) * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    self._check_x(x)
    return self._scale * x 

def _SelfAdjointEigV2Grad(op, grad_e, grad_v):
  """Gradient for SelfAdjointEigV2."""
  e = op.outputs[0]
  v = op.outputs[1]
  # a = op.inputs[0], which satisfies
  # a[...,:,:] * v[...,:,i] = e[...,i] * v[...,i]
  with ops.control_dependencies([grad_e.op, grad_v.op]):
    if grad_v is not None:
      # Construct the matrix f(i,j) = (i != j ? 1 / (e_i - e_j) : 0).
      # Notice that because of the term involving f, the gradient becomes
      # infinite (or NaN in practice) when eigenvalues are not unique.
      # Mathematically this should not be surprising, since for (k-fold)
      # degenerate eigenvalues, the corresponding eigenvectors are only defined
      # up to arbitrary rotation in a (k-dimensional) subspace.
      f = array_ops.matrix_set_diag(
          math_ops.reciprocal(
              array_ops.expand_dims(e, -2) - array_ops.expand_dims(e, -1)),
          array_ops.zeros_like(e))
      grad_a = math_ops.matmul(
          v,
          math_ops.matmul(
              array_ops.matrix_diag(grad_e) + f * math_ops.matmul(
                  v, grad_v, adjoint_a=True),
              v,
              adjoint_b=True))
    else:
      grad_a = math_ops.matmul(
          v, math_ops.matmul(
              array_ops.matrix_diag(grad_e), v, adjoint_b=True))
    # The forward op only depends on the lower triangular part of a, so here we
    # symmetrize and take the lower triangle
    grad_a = array_ops.matrix_band_part(
        grad_a + array_ops.matrix_transpose(grad_a), -1, 0)
    grad_a = array_ops.matrix_set_diag(grad_a,
                                       0.5 * array_ops.matrix_diag_part(grad_a))
    return grad_a 

def _batch_sqrt_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return diag_mat * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return math_ops.square(diag_mat) * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    diag_mat = array_ops.expand_dims(self._diag, -1)
    return diag_mat * x 

def _batch_sqrt_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    self._check_x(x)
    return math_ops.sqrt(self._scale) * x 

def _batch_matmul(self, x, transpose_x=False):
    if transpose_x:
      x = array_ops.matrix_transpose(x)
    self._check_x(x)
    return self._scale * x