Home
last modified time | relevance | path

Searched refs:adjoint (Results 1 – 14 of 14) sorted by relevance

/petsc/doc/manualpages/MANSECHeaders/
H A DSensitivity4 The `TS` library provides discrete adjoint models (`TSAdjoint`) and tangent linear models (`TSForwa…
6 The adjoint solvers support gradient calculation for multiple cost functions, and the tangent linea…
/petsc/src/ts/tutorials/output/
H A Dex23fwdadj_1.out13 adjoint sensitivity:
/petsc/src/tao/tutorials/
H A Dex3.c48 Vec adjoint; member
267 PetscCall(VecDuplicate(user->data, &user->adjoint)); in CreateCtx()
283 PetscCall(VecDestroy(&user->adjoint)); in DestroyCtx()
321 PetscCall(KSPSolve(user->ksp_laplace, user->tmp2, user->adjoint)); /* adjoint solve */ in ReducedFunctionGradient()
324 PetscCall(MatMult(user->mass, user->adjoint, user->tmp1)); in ReducedFunctionGradient()
/petsc/doc/changes/
H A D312.md195 - Added second-order adjoint solvers based on Runge-Kutta methods
197 - Improved the usage of first-order adjoint solvers in an
204 checkpoint data. Users can rerun adjoint codes without worrying
208 approximate the Jacobians needed by the adjoint solvers.
H A D36.md151 (forwardsolve = 1) and adjoint solve (forwardsolve = 0).
152 - Added TSAdjointSolve() for adjoint sensitivity analysis
H A D315.md227 - Add forward and adjoint sensitivity support for cases that involve
H A D319.md212 - Add support for first-order adjoint calculation for `TSARKIMEX`
/petsc/src/snes/tutorials/
H A Dex13.c17 PetscBool adjoint; /* Solve the adjoint problem */ member
101 options->adjoint = PETSC_FALSE; in PLEXFE_QFUNCTION()
108 …ool("-adjoint", "Solve the adjoint problem", "ex13.c", options->adjoint, &options->adjoint, NULL)); in PLEXFE_QFUNCTION()
359 if (!user->adjoint) PetscFunctionReturn(PETSC_SUCCESS); in ComputeAdjoint()
/petsc/doc/manual/
H A Dts.md1091 state $u$, a boolean flag indicating forward solve (1) or adjoint
1109 The `TS` library provides a framework based on discrete adjoint models
1117 ### Using the discrete adjoint methods
1143 To perform the discrete adjoint sensitivity analysis one first sets up
1234 without having to use the adjoint solvers.
1236 To provide a better understanding of the use of the adjoint solvers, we
1290 In the adjoint run, $\lambda$ and $\mu$ are initialized as
1311 - discrete adjoint sensitivity using explicit and implicit time stepping methods for an ODE problem
1313 - an optimization problem using the discrete adjoint models of the ERK (for nonstiff ODEs)
1318 - an ODE-constrained optimization using the discrete adjoint models of the
[all …]
H A Dtao.md2306 The reduced gradient is obtained from one linearized adjoint solve
/petsc/doc/
H A Dindex.md69 - [DAFoam](https://dafoam.github.io) Discrete adjoint solvers
193 If you utilize the `TS` adjoint solver please cite
H A Dpetsc.bib414 …title = {An approximate factorization procedure for solving self-adjoint elliptic differe…
1183 title = {Optimal checkpointing for adjoint multistage time-stepping schemes},
1193 title = {Revolve-based adjoint checkpointing for multistage time integration},
4591 implemented in matrix-free fashion with adjoint-based computation of reduced
4594 solutions of the forward and adjoint wave equations and the evaluation of
5585 title = {A methodology for the development of discrete adjoint solvers using automatic
11249 adjoint mode of computational differentiation},
17780 title = {Parallel scalable adjoint-based adaptive solution of variable-viscosity {Stokes}
22468 adjoint formulation on distributed memory parallel computers},
/petsc/src/binding/petsc4py/src/petsc4py/PETSc/
H A DTS.pyx2855 """Set the number of steps the adjoint solver should take backward in time.
2873 """Set up the internal data structures for the later use of an adjoint solver.
2885 """Solve the discrete adjoint problem for an ODE/DAE.
2897 """Step one time step backward in the adjoint run.
/petsc/share/petsc/datafiles/meshes/
H A Dtestcase3D.cas5530 (adjoint/adaption/solution/initialized? #f)
5531 (adjoint/feature/goal-based-mesh-adaption? #f)
5532 (adjoint/solution-exists? #f)
5533 (adjoint/custom-turb-model/on? #f)
5534 (adjoint/on? #f)