dolfinx.cpp.graph

Graph module

Classes

AdjacencyList_int32([order, writable])	Adjacency List
AdjacencyList_int64([order, writable])	Adjacency List
AdjacencyList_int_sizet_int8__int32_int32	Adjacency List
GeometricPartitioner	Cell partitioner using cell positions rather than the mesh dual graph, e.g. create_geometric_cell_partitioner's result, or the partition_morton, partition_hilbert and partitioner_parmetis_geom built-ins.
GeometricReorderer	Mesh cell reordering function that uses cell centroids.
GraphPartitioner	Cell partitioner using the mesh dual graph, e.g. as returned by partitioner, partitioner_scotch, partitioner_parmetis or partitioner_kahip.
HybridPartitioner	Cell partitioner returned by create_hybrid_cell_partitioner.
SCOTCHStrategy(*values)

class dolfinx.cpp.graph.AdjacencyList_int32(self, adj: ndarray[dtype=int32, shape=(*), order='C', writable=False])

class dolfinx.cpp.graph.AdjacencyList_int32(self, adj: ndarray[dtype=int32, shape=(*, *), order='C', writable=False])

class dolfinx.cpp.graph.AdjacencyList_int32(self, data: ndarray[dtype=int32, shape=(*), order='C', writable=False], offsets: ndarray[dtype=int32, shape=(*), order='C', writable=False])

Bases: object

Adjacency List

property array

(self) -> numpy.ndarray[dtype=int32, writable=False]

links(self, i: int) → numpy.ndarray[dtype=int32, writable=False]

Links (edges) of a node

property num_nodes

(self) -> int

property offsets

(self) -> numpy.ndarray[dtype=int32, writable=False]

class dolfinx.cpp.graph.AdjacencyList_int64(self, adj: ndarray[dtype=int64, shape=(*), order='C', writable=False])

class dolfinx.cpp.graph.AdjacencyList_int64(self, adj: ndarray[dtype=int64, shape=(*, *), order='C', writable=False])

class dolfinx.cpp.graph.AdjacencyList_int64(self, data: ndarray[dtype=int64, shape=(*), order='C', writable=False], offsets: ndarray[dtype=int32, shape=(*), order='C', writable=False])

Bases: object

Adjacency List

property array

(self) -> numpy.ndarray[dtype=int64, writable=False]

links(self, i: int) → numpy.ndarray[dtype=int64, writable=False]

Links (edges) of a node

property num_nodes

(self) -> int

property offsets

(self) -> numpy.ndarray[dtype=int32, writable=False]

class dolfinx.cpp.graph.AdjacencyList_int_sizet_int8__int32_int32

Bases: object

Adjacency List

property num_nodes

(self) -> int

property offsets

(self) -> numpy.ndarray[dtype=int32, writable=False]

class dolfinx.cpp.graph.GeometricPartitioner

Bases: object

Cell partitioner using cell positions rather than the mesh dual graph, e.g. create_geometric_cell_partitioner’s result, or the partition_morton, partition_hilbert and partitioner_parmetis_geom built-ins. Pass it as create_mesh’s partitioner argument; it is also directly callable for low-level use, e.g. via compute_cell_centroids.

class dolfinx.cpp.graph.GeometricReorderer

Bases: object

Mesh cell reordering function that uses cell centroids. Pass it as create_mesh’s reorder_fn argument; it is also directly callable for low-level use.

class dolfinx.cpp.graph.GraphPartitioner

Bases: object

Cell partitioner using the mesh dual graph, e.g. as returned by partitioner, partitioner_scotch, partitioner_parmetis or partitioner_kahip. Pass it as create_mesh’s partitioner argument; it is also directly callable for low-level use.

class dolfinx.cpp.graph.HybridPartitioner

Bases: object

Cell partitioner returned by create_hybrid_cell_partitioner. Pass it as create_mesh’s partitioner argument; it is also directly callable for low-level use, e.g. via compute_cell_centroids.

class dolfinx.cpp.graph.SCOTCHStrategy(*values)

Bases: Enum

balance = 1

none = 0

quality = 2

safety = 3

scalability = 5

speed = 4