bundles / scipy 2.0.0.dev0+git20260907.41eeb59 / scipy / linalg / _flapack / dtgsen
lapack_function
scipy.linalg._flapack:dtgsen
source: build-install/usr/lib/python3.14/site-packages/scipy/linalg/_flapack.cpython-314-x86_64-linux-gnu.so
Signature
def dtgsen(select, a, b, q, z, ijob = 4, wantq = 1, wantz = 1, lwork = Ellipsis, liwork = Ellipsis, overwrite_a = 0, overwrite_b = 0, overwrite_q = 0, overwrite_z = 0)Summary
Reorder a generalized Schur decomposition to gather selected eigenvalues into a leading cluster (LAPACK dtgsen).
Parameters
select: ndarrayBoolean mask of length
n: the eigenvalues it marks are gathered into the leading block. A bool array works; LAPACK reads it as a Fortran LOGICAL.a: ndarrayUpper (quasi-)triangular factor of the pencil, shape
(n, n).b: ndarrayUpper triangular factor of the pencil, shape
(n, n).q: ndarrayLeft Schur vectors; updated when wantq is nonzero.
z: ndarrayRight Schur vectors; updated when wantz is nonzero.
ijob: int, optionalWhat to compute besides the reordering, 0 through 5; 0 reorders only. Default is 4.
wantq: int, optionalIf nonzero, the left transformation is accumulated into q. Default is 1.
wantz: int, optionalIf nonzero, the right transformation is accumulated into z. Default is 1.
lwork: int, optionalWorkspace size; -1 requests the query. Default is
4 * n + 16. The default is a bare minimum for the cheapestjobonly -- what is actually needed depends on the cluster size, so query the matching_lworkroutine.liwork: int, optionalInteger workspace size; -1 requests the query. Default is
n + 6.overwrite_a: int, optionalIf nonzero, a may be overwritten in place. Default is 0.
overwrite_b: int, optionalIf nonzero,
bmay be overwritten in place. Default is 0.overwrite_q: int, optionalIf nonzero, q may be overwritten in place. Default is 0.
overwrite_z: int, optionalIf nonzero,
zmay be overwritten in place. Default is 0.
Returns
as: ndarrayThe reordered first factor.
bs: ndarrayThe reordered second factor.
alphar: ndarrayReal parts of the eigenvalue numerators, length
n.alphai: ndarrayImaginary parts of the eigenvalue numerators, length
n.beta: ndarrayDenominators of the reordered eigenvalues, length
n.qs: ndarrayThe updated left Schur vectors.
zs: ndarrayThe updated right Schur vectors.
m: intSize of the selected cluster.
pl: floatLower bound on the reciprocal condition number of the left cluster.
pr: floatLower bound on the reciprocal condition number of the right cluster.
dif: ndarrayTwo estimates of the Difu and Difl separations.
info: int0 on success; if negative, the
-info-th argument had an illegal value.
Aliases
scipy.linalg._flapack.dtgsen