line::gen Namespace Reference

Classes
class	LayeredNetworkGenerator
	A random lqn::LqnBuilder<T> source, configured once and then drawn from. More...
class	MatlabRng
	One MATLAB global random stream, as left by rng(seed). More...
class	NetworkGenerator
	A random qn::Network<T> source, configured once and then drawn from. More...
struct	Range
	A closed interval the generator samples from. More...

Enumerations
enum class	TopologyKind { Rand , Cyclic }
	The two topology generators the reference ships, randGraph and cyclicGraph. More...

Functions
Matrix< double >	rand_spanning_tree (std::size_t num_vertices, MatlabRng &r)
	randSpanningTree(n): vertex i (i >= 1) is attached to a uniformly chosen earlier vertex, randi(i) in MATLAB's 1-based terms, and always a tree rooted at 0.
Matrix< double >	rand_graph (std::size_t num_vertices, MatlabRng &r)
	randGraph(n): a random strongly connected digraph on n vertices, as a random spanning tree closed up by strongConnect and then relabelled by a random permutation (without which vertex 0 would always be the DFS root).
Matrix< double >	cyclic_graph (std::size_t num_vertices)
	cyclicGraph(n): the single cycle 0 -> 1 -> ... -> n-1 -> 0.
std::vector< double >	randfixedsumone (std::size_t num_elems, MatlabRng &r)
	randfixedsumone(n): n probabilities summing to exactly 1.
std::vector< int >	randintfixedsum (int s, int n, MatlabRng &r)
	randintfixedsum(s, n): n STRICTLY POSITIVE integers summing to s.

Enumeration Type Documentation

TopologyKind

enum class line::gen::TopologyKind

strong

The two topology generators the reference ships, randGraph and cyclicGraph.

Enumerator
Rand
Cyclic

Definition at line 92 of file network_generator.h.

Function Documentation

cyclic_graph()

Matrix< double > line::gen::cyclic_graph ( std::size_t num_vertices )

inline

cyclicGraph(n): the single cycle 0 -> 1 -> ... -> n-1 -> 0.

Definition at line 187 of file network_generator.h.

References cyclic_graph(), and line::InputError::InputError().

Referenced by cyclic_graph(), and line::gen::NetworkGenerator< T >::set_topology().

rand_graph()

Matrix< double > line::gen::rand_graph ( std::size_t num_vertices, MatlabRng & r )

inline

randGraph(n): a random strongly connected digraph on n vertices, as a random spanning tree closed up by strongConnect and then relabelled by a random permutation (without which vertex 0 would always be the DFS root).

Definition at line 163 of file network_generator.h.

References line::InputError::InputError(), rand_graph(), rand_spanning_tree(), and line::gen::MatlabRng::randperm().

Referenced by rand_graph(), and line::gen::NetworkGenerator< T >::set_topology().

rand_spanning_tree()

Matrix< double > line::gen::rand_spanning_tree ( std::size_t num_vertices, MatlabRng & r )

inline

randSpanningTree(n): vertex i (i >= 1) is attached to a uniformly chosen earlier vertex, randi(i) in MATLAB's 1-based terms, and always a tree rooted at 0.

Definition at line 102 of file network_generator.h.

References rand_spanning_tree(), and line::gen::MatlabRng::randi().

Referenced by rand_graph(), and rand_spanning_tree().

randfixedsumone()

std::vector< double > line::gen::randfixedsumone ( std::size_t num_elems, MatlabRng & r )

inline

randfixedsumone(n): n probabilities summing to exactly 1.

The reference's own recipe, quirks kept: a uniform point of the simplex from Stafford's randfixedsum (which draws even at n = 1), each entry rounded UP to a multiple of 1e-3 (so the rounded vector oversums), and the FIRST LARGEST entry absorbing the whole residual. Rounding to three digits is what keeps the emitted model.json readable; the absorption keeps the row stochastic after it.

Definition at line 208 of file network_generator.h.

References line::gen::MatlabRng::randfixedsum(), and randfixedsumone().

Referenced by randfixedsumone().

randintfixedsum()

std::vector< int > line::gen::randintfixedsum ( int s, int n, MatlabRng & r )

inline

randintfixedsum(s, n): n STRICTLY POSITIVE integers summing to s.

The reference's recursion: draw the first part uniformly from 1..s-n (which leaves at least one unit for each remaining part) and recurse on the rest. There is no shuffle here: genCSMask permutes the result with randperm.

Definition at line 231 of file network_generator.h.

References line::InputError::InputError(), line::gen::MatlabRng::randi(), and randintfixedsum().

Referenced by randintfixedsum().