def linop_scale(w, op):
# We assume w > 0. (This assumption only relates to the is_* attributes.)
with ops.name_scope("linop_scale", values=[w]):
# TODO(b/35301104): LinearOperatorComposition doesn't combine operators, so
# special case combinations here. Once it does, this function can be
# replaced by:
# return linop_composition_lib.LinearOperatorComposition([
# scaled_identity(w), op])
def scaled_identity(w):
return linop_identity_lib.LinearOperatorScaledIdentity(
num_rows=op.range_dimension_tensor(),
multiplier=w,
is_non_singular=op.is_non_singular,
is_self_adjoint=op.is_self_adjoint,
is_positive_definite=op.is_positive_definite)
if isinstance(op, linop_identity_lib.LinearOperatorIdentity):
return scaled_identity(w)
if isinstance(op, linop_identity_lib.LinearOperatorScaledIdentity):
return scaled_identity(w * op.multiplier)
if isinstance(op, linop_diag_lib.LinearOperatorDiag):
return linop_diag_lib.LinearOperatorDiag(
diag=w[..., array_ops.newaxis] * op.diag_part(),
is_non_singular=op.is_non_singular,
is_self_adjoint=op.is_self_adjoint,
is_positive_definite=op.is_positive_definite)
if isinstance(op, linop_tril_lib.LinearOperatorLowerTriangular):
return linop_tril_lib.LinearOperatorLowerTriangular(
tril=w[..., array_ops.newaxis, array_ops.newaxis] *
op.to_dense(),
is_non_singular=op.is_non_singular,
is_self_adjoint=op.is_self_adjoint,
is_positive_definite=op.is_positive_definite)
raise NotImplementedError("Unsupported Linop type ({})".format(
type(op).__name__))
def _solve_linear_operator_diag_tril(linop_diag, linop_triangular):
return linear_operator_lower_triangular.LinearOperatorLowerTriangular(
tril=linop_triangular.to_dense() / linop_diag.diag[..., None],
is_non_singular=registrations_util.combined_non_singular_hint(
linop_diag, linop_triangular),
# This is safe to do since the Triangular matrix is only self-adjoint
# when it is a diagonal matrix, and hence commutes.
is_self_adjoint=registrations_util.
combined_commuting_self_adjoint_hint(linop_diag, linop_triangular),
is_positive_definite=None,
is_square=True)
def _matmul_linear_operator_tril_diag(linop_triangular, linop_diag):
return linear_operator_lower_triangular.LinearOperatorLowerTriangular(
tril=linop_triangular.to_dense() * linop_diag.diag,
is_non_singular=_combined_non_singular_hint(linop_diag,
linop_triangular),
# This is safe to do since the Triangular matrix is only self-adjoint
# when it is a diagonal matrix, and hence commutes.
is_self_adjoint=_combined_self_adjoint_hint(linop_diag,
linop_triangular),
is_positive_definite=None,
is_square=True)
def _add(self, op1, op2, operator_name, hints):
if _type(op1) in _EFFICIENT_ADD_TO_TENSOR:
op_add_to_tensor, op_other = op1, op2
else:
op_add_to_tensor, op_other = op2, op1
return linear_operator_lower_triangular.LinearOperatorLowerTriangular(
tril=op_add_to_tensor.add_to_tensor(op_other.to_dense()),
is_non_singular=hints.is_non_singular,
is_self_adjoint=hints.is_self_adjoint,
is_positive_definite=hints.is_positive_definite,
name=operator_name)
def _cholesky_linear_operator(linop):
return linear_operator_lower_triangular.LinearOperatorLowerTriangular(
linalg_ops.cholesky(linop.to_dense()),
is_non_singular=True,
is_self_adjoint=False,
is_square=True)
