pybind: expose VMEC preconditioner operator + preconditioned JFNK#579
Open
krystophny wants to merge 9 commits into
Open
pybind: expose VMEC preconditioner operator + preconditioned JFNK#579krystophny wants to merge 9 commits into
krystophny wants to merge 9 commits into
Conversation
Add a precondition flag to VmecModel.evaluate (default true, unchanged behaviour). With precondition=false the forward model returns at the INVARIANT_RESIDUALS checkpoint, so get_forces() yields the raw, unpreconditioned force: the gradient of VMEC's augmented functional (MHD energy plus the spectral-condensation and lambda constraints) with respect to the decomposed internal-basis state. This is the consistent state/gradient pair an external optimizer needs to minimise in VMEC's own basis. The native solver's preconditioned search direction (precondition=true) is a different vector; the raw gradient is the equilibrium residual and vanishes at convergence. Tests: raw force is finite and differs in direction from the preconditioned force, and drops by >1e6 from the initial guess to the converged equilibrium.
Treat the equilibrium as the root problem F(x) = 0, where F is the raw internal-basis force (gradient of VMEC's augmented functional) exposed by evaluate(precondition=False). Wire it to two solvers that reuse VMEC++'s forward model: native-style preconditioned descent and Jacobian-free Newton-Krylov (matrix-free Hessian information). Both reach the native solver's equilibrium. This is the external-differentiability path: VMEC++ as a differentiable equilibrium component an outside optimizer can drive. Quasi-Newton root-finders without a preconditioner diverge on this stiff system, which motivates exposing VMEC's preconditioner as an operator next. Tests assert both solvers reach force balance and recover the native energy and state.
Add VmecModel.apply_preconditioner(v): applies VMEC's preconditioner M^-1 (m=1, radial, lambda steps) to a vector in the decomposed basis. M^-1 is VMEC's hand-built approximate inverse Hessian; this exposes it as a reusable linear operator for preconditioned Krylov / quasi-Newton and for the Hessian solve in adjoint sensitivities. It requires a prior evaluate(precondition=true), which assembles the radial preconditioner. Validated exactly: apply_preconditioner(raw force) equals the native preconditioned search direction; the operator is linear and, once assembled, state-invariant. Use it as the inner Krylov preconditioner in Newton-Krylov: on solovev (ns=11) this cuts force evaluations from 2242 to 505 (4.4x) versus unpreconditioned JFNK, converging to the same equilibrium.
This was referenced Jun 14, 2026
The 'Compare benchmark result' step uses github-action-benchmark with comment-on-alert and the GITHUB_TOKEN, which is read-only for pull requests from forks -> 'Resource not accessible by integration'. Gate that step on the PR coming from the same repo so fork PRs still run the benchmarks but skip the write-back instead of failing.
The pinned vmec-0.0.6 cp310 wheel was f90wrapped against numpy 1.x. Under the numpy 2.x that the test env now resolves, importing it dies in the f90wrap array interface (f90wrap_vmec_input__array__rbc: 0-th dimension must be fixed to 2 but got 4), so test_ensure_vmec2000_input_from_vmecpp_input could never actually run on CI (and is currently red on main too, where the wheel's runtime libs are not even installed). Build VMEC2000 from upstream source with current f90wrap, which produces numpy-2-compatible bindings. The recipe mirrors SIMSOPT's own CI (hiddenSymmetries/VMEC2000, cmake/machines/ubuntu.json). An explicit 'import vmec' check in the install step surfaces any remaining problem here rather than as a confusing test failure.
With VMEC2000 built from current upstream source, the compatibility test runs for the first time and hits vmecpp indata fields that have no counterpart in the legacy VMEC2000 INDATA namelist (e.g. free_boundary_method), which raised AttributeError. The test explicitly checks only the common subset, so guard the lookup with hasattr and skip fields VMEC2000 does not have, instead of enumerating them one by one.
…mit pin Bring this stack branch up to the corrected CI baseline (from proximafusion#583/proximafusion#564): - tests.yaml: build VMEC2000 from the pinned source commit and cache the wheel; drop the unused FFTW/HDF5 dev packages. - benchmarks.yaml: skip the result upload on fork PRs (read-only token). - test_simsopt_compat.py: skip vmecpp-only INDATA fields. - CMakeLists: pin abseil to the 20260107.1 commit hash for Clang >= 21.
This was referenced Jun 16, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Expose VMEC's hand-built preconditioner as a reusable linear operator and use it
to precondition the external Newton-Krylov solver.
VmecModel.apply_preconditioner(v): applies VMEC's block preconditionerM^-1(the m=1, radial, and lambda blocks) -- its approximate inverse Hessian.Exact application:
M^-1 . F_raw == F_preconditioned. It requires a priorevaluate(precondition=true)(which assembles the radial block); the operatoris then state-invariant and can be frozen across a Krylov solve.
solve_newton_krylov(..., preconditioned=True): usesM^-1as the innerKrylov preconditioner.
This preconditioner is the metric for the preconditioned Krylov / quasi-Newton
solvers and the inner preconditioner for the Hessian solve in the adjoint
sensitivities (#11), and it is why first-order JFNK is so eval-efficient.
Verification (force evals counted in VMEC++, ns=11)
Preconditioning cuts JFNK from 2243 to 507 force evaluations on solovev (4.4x)
and converges cth_like where unpreconditioned JFNK does not. This is the
best-of-breed solver in the stack; the exact autodiff HVP (#23) is the operator
that needs no force evaluation per matvec.
Stacked on #8 (external optimizers).
Tracking: #590