def eig(tensor, name=None):
"""Computes the eigen decomposition of a batch of matrices.
The eigenvalues
and eigenvectors for a non-Hermitian matrix in general are complex. The
eigenvectors are not guaranteed to be linearly independent.
Computes the eigenvalues and right eigenvectors of the innermost
N-by-N matrices in `tensor` such that
`tensor[...,:,:] * v[..., :,i] = e[..., i] * v[...,:,i]`, for i=0...N-1.
Args:
tensor: `Tensor` of shape `[..., N, N]`. Only the lower triangular part of
each inner inner matrix is referenced.
name: string, optional name of the operation.
Returns:
e: Eigenvalues. Shape is `[..., N]`. Sorted in non-decreasing order.
v: Eigenvectors. Shape is `[..., N, N]`. The columns of the inner most
matrices contain eigenvectors of the corresponding matrices in `tensor`
"""
if tensor.dtype == dtypes.float32 or tensor.dtype == dtypes.complex64:
out_dtype = dtypes.complex64
elif tensor.dtype == dtypes.float64 or tensor.dtype == dtypes.complex128:
out_dtype = dtypes.complex128
e, v = gen_linalg_ops.eig(tensor,
Tout=out_dtype,
compute_v=True,
name=name)
return e, v
def eigvals(tensor, name=None):
"""Computes the eigenvalues of one or more matrices.
Note: If your program backpropagates through this function, you should replace
it with a call to tf.linalg.eig (possibly ignoring the second output) to
avoid computing the eigen decomposition twice. This is because the
eigenvectors are used to compute the gradient w.r.t. the eigenvalues. See
_SelfAdjointEigV2Grad in linalg_grad.py.
Args:
tensor: `Tensor` of shape `[..., N, N]`.
name: string, optional name of the operation.
Returns:
e: Eigenvalues. Shape is `[..., N]`. The vector `e[..., :]` contains the `N`
eigenvalues of `tensor[..., :, :]`.
"""
if tensor.dtype == dtypes.float32 or tensor.dtype == dtypes.complex64:
out_dtype = dtypes.complex64
elif tensor.dtype == dtypes.float64 or tensor.dtype == dtypes.complex128:
out_dtype = dtypes.complex128
e, _ = gen_linalg_ops.eig(tensor,
Tout=out_dtype,
compute_v=False,
name=name)
return e
def testMismatchedDtypes(self):
tensor = constant_op.constant([[0, 1], [2, 3]],
dtype=dtypes_lib.float32)
with self.assertRaisesRegex((ValueError, errors.InvalidArgumentError),
"Invalid output dtype"):
self.evaluate(
gen_linalg_ops.eig(
input=tensor,
Tout=dtypes_lib.complex128, # Expected dtype: complex64.
compute_v=True))
