volumential.wranglers.sumpy_backend#

The sumpy-based fpnd expansion wrangler.

fpnd = far field by Particle approximation, Near field Directly from precomputed interaction tables. This module owns the sumpy flavour of that wrangler plus its tree-independent data, and the default aliases exported as FPNDExpansionWrangler / FPNDTreeIndependentDataForWrangler.

class volumential.wranglers.sumpy_backend.FPNDSumpyTreeIndependentDataForWrangler(cl_context, multipole_expansion_factory, local_expansion_factory, target_kernels, exclude_self=True, use_rscale=None, strength_usage=None, source_kernels=None)[source]#

Bases: TreeIndependentDataForWranglerInterface, SumpyTreeIndependentDataForWrangler

Objects of this type serve as a place to keep the code needed for ExpansionWrangler if it is using sumpy to perform multipole expansion and manipulations.

Since SumpyExpansionWrangler necessarily must have a pyopencl.CommandQueue, but this queue is allowed to be more ephemeral than the code, the code’s lifetime is decoupled by storing it in this object.

get_wrangler(queue, traversal, dtype, fmm_level_to_order, source_extra_kwargs=None, kernel_extra_kwargs=None, self_extra_kwargs=None, *args, **kwargs) → FPNDSumpyExpansionWrangler[source]#

Build a queue-bound wrangler for traversal.

Returns:: an FPNDSumpyExpansionWrangler.

p2m(tgt_order: int)[source]#: Sumpy’s point-to-multipole translation for order tgt_order.

p2l(tgt_order: int)[source]#: Sumpy’s point-to-local translation for order tgt_order.

class volumential.wranglers.sumpy_backend.FPNDSumpyExpansionWrangler(tree_indep, queue, traversal, dtype, fmm_level_to_order, near_field_table, quad_order, potential_kind=1, source_extra_kwargs=None, kernel_extra_kwargs=None, self_extra_kwargs=None, list1_extra_kwargs=None, helmholtz_split=None, helmholtz_split_order=1, helmholtz_split_smooth_quad_order=None, helmholtz_split_auto_config=None, helmholtz_split_term_tables=None, helmholtz_split_order1_legacy_subtraction=False, translation_classes_data=None, preprocessed_mpole_dtype=None)[source]#

Bases: ExpansionWranglerInterface, NearFieldPayloadCacheMixin, HelmholtzSplitCorrectionMixin, SumpyExpansionWrangler

This expansion wrangler uses “fpnd” strategy. That is, Far field is computed via Particle approximation and Near field is computed Directly. The FMM is performed using sumpy backend.

For Helmholtz split mode, neighbor-list (list1) interactions are evaluated with the decomposition

\[G_k(r) = G_0(r) + S_p(r) + R_p(r),\]

where G_0 is tabulated by Laplace near-field tables, S_p is the sum of pretabulated non-smooth split terms, and R_p is an analytic remainder evaluated online by P2P.

Split terms kept in S_p are:

2D: \(r^{2n}\log r\), \(n=1,\dots,p-1\).
3D: \(r^{2j-1}\), \(j=1,\dots,p-1\).

Smooth polynomial terms (2D: constant and \(r^{2n}\); 3D: even powers including constant) remain in R_p.

source_extra_kwargs#: Keyword arguments to be passed to interactions that involve the source field.

kernel_extra_kwargs#: Keyword arguments to be passed to interactions that involve expansions, but not the source field.

self_extra_kwargs#: Keyword arguments to be passed for handling self interactions (singular integrals)

multipole_expansion_zeros(actx=None) → Any[source]#: Zeroed multipole-expansion storage, on actx or the wrangler’s own.

local_expansion_zeros(actx=None) → Any[source]#: Zeroed local-expansion storage, on actx or the wrangler’s own.

output_zeros(actx=None) → Any[source]#: Zeroed potential storage, on actx or the wrangler’s own.

reorder_sources(source_array: Any) → Any[source]#: Return source_array, given in user order, in tree source order.

reorder_targets(target_array: Any) → Any[source]#: Return target_array permuted by the tree’s target ordering.

reorder_potentials(potentials: Any) → Any[source]#: Return potentials, given in tree order, in user target order.

finalize_potentials(potentials: Any) → Any[source]#: Apply sumpy’s global scaling to the reordered potentials.

form_multipoles(level_start_source_box_nrs: Any, source_boxes: Any, src_weights: Any) → Any[source]#: P2M: form multipole expansions in source_boxes.

coarsen_multipoles(level_start_source_parent_box_nrs: Any, source_parent_boxes: Any, mpoles: Any) → Any[source]#: M2M: accumulate child multipole expansions into their parents.

eval_direct_single_out_kernel(out_pot: Any, out_kernel, target_boxes: Any, neighbor_source_boxes_starts: Any, neighbor_source_boxes_lists: Any, mode_coefs: Any, near_field_tables=None, list1_extra_kwargs=None) → tuple[Any, object][source]#

Near-field (list 1) contribution to out_pot for one target kernel.

The interaction is read out of the precomputed tables instead of being evaluated point by point. Returns (out_pot, event).

eval_direct(target_boxes: Any, neighbor_source_boxes_starts: Any, neighbor_source_boxes_lists: Any, mode_coefs: Any) → Any[source]#: Near-field stage: one table-based pass per target kernel.

multipole_to_local(level_start_target_box_nrs: Any, target_boxes: Any, src_box_starts: Any, src_box_lists: Any, mpole_exps: Any) → Any[source]#: M2L: translate list-2 multipole expansions into local expansions.

eval_multipoles(target_boxes_by_source_level: Any, source_boxes_by_level: Any, mpole_exps: Any) → Any[source]#: M2P: evaluate list-3 multipole expansions at the targets.

form_locals(level_start_target_or_target_parent_box_nrs: Any, target_or_target_parent_boxes: Any, starts: Any, lists: Any, src_weights: Any) → Any[source]#: P2L: form local expansions from the list-4 sources.

refine_locals(level_start_target_or_target_parent_box_nrs: Any, target_or_target_parent_boxes: Any, local_exps: Any) → Any[source]#: L2L: push each parent’s local expansion down to its children.

eval_locals(level_start_target_box_nrs: Any, target_boxes: Any, local_exps: Any) → Any[source]#: L2P: evaluate local expansions at the targets.

eval_direct_p2p(target_boxes: Any, source_box_starts: Any, source_box_lists: Any, src_weights: Any) → Any[source]#: Near-field stage done as a plain P2P, bypassing the tables.

class volumential.wranglers.sumpy_backend.FPNDTreeIndependentDataForWrangler(cl_context, multipole_expansion_factory, local_expansion_factory, target_kernels, exclude_self=True, use_rscale=None, strength_usage=None, source_kernels=None)[source]#

Bases: FPNDSumpyTreeIndependentDataForWrangler

The default tree-independent-data class.

class volumential.wranglers.sumpy_backend.FPNDExpansionWrangler(tree_indep, queue, traversal, dtype, fmm_level_to_order, near_field_table, quad_order, potential_kind=1, source_extra_kwargs=None, kernel_extra_kwargs=None, self_extra_kwargs=None, list1_extra_kwargs=None, helmholtz_split=None, helmholtz_split_order=1, helmholtz_split_smooth_quad_order=None, helmholtz_split_auto_config=None, helmholtz_split_term_tables=None, helmholtz_split_order1_legacy_subtraction=False, translation_classes_data=None, preprocessed_mpole_dtype=None)[source]#

Bases: FPNDSumpyExpansionWrangler

The default wrangler class.