Package de.labathome.trustconstr.records

Class Sparsity

java.lang.Object

de.labathome.trustconstr.records.Sparsity

public final class Sparsity
extends Object

Pairing of a sparsity-pattern matrix A with a precomputed Curtis-Powell-Reid column grouping. Used by NumDiff.approxDerivative(java.util.function.DoubleUnaryOperator, double, de.labathome.trustconstr.records.FiniteDifferenceOptions) to drive the sparse-Jacobian finite-difference path.

Constructor Summary

Constructors

Constructor	Description
Sparsity(Matrix A, int[] sparsityGroups)

Method Summary

Modifier and Type	Method	Description
Matrix	A()
static Sparsity	of(Matrix A)	Convenience factory: derive the column grouping from A via NumDiff.groupColumns(Matrix).
static Sparsity	of(Matrix A, int[] order)	Convenience factory with a user-supplied column ordering.
int[]	sparsityGroups()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

Sparsity

public Sparsity(Matrix A,
 int[] sparsityGroups)

Parameters:

A - m x n sparsity-pattern matrix (non-zero entries mark structural non-zeros of the Jacobian)

sparsityGroups - length-n column-group assignment from NumDiff.groupColumns(Matrix)

Method Details

A

public Matrix A()

Returns:

the sparsity-pattern matrix A

sparsityGroups

public int[] sparsityGroups()

Returns:

per-column group indices

of

public static Sparsity of(Matrix A)

Convenience factory: derive the column grouping from A via NumDiff.groupColumns(Matrix).

Parameters:

A - sparsity-pattern matrix

Returns:

a fresh Sparsity

of

public static Sparsity of(Matrix A,
 int[] order)

Convenience factory with a user-supplied column ordering.

Parameters:

A - sparsity-pattern matrix

order - permutation array passed through to NumDiff.groupColumns(Matrix, int[])

Returns:

a fresh Sparsity