1 core.parameters.MatrixParameter

core.parameters.MatrixParameter(name, values, symmetric=False)

A matrix of parameters for array-valued constants.

MatrixParameter stores a 2D array of constant values that can be updated between solves without rebuilding the problem structure. This is useful for: - Covariance matrices in portfolio optimization - Constraint coefficient matrices that change over time - Distance/cost matrices in routing problems

Unlike VectorParameter (which creates individual Parameter objects), MatrixParameter stores values directly for efficiency with large matrices.

1.1 Parameters

Name	Type	Description	Default
name	str	Name identifier for this parameter matrix.	required
values	ArrayLike	Initial 2D array of values.	required
symmetric	bool	If True, enforce and exploit symmetry. Only the upper triangle is stored, and updates must maintain symmetry.	False

1.2 Example

import numpy as np from optyx import VectorVariable, MatrixParameter

2 Portfolio with updatable covariance

cov = np.array([[0.04, 0.01], [0.01, 0.09]]) Sigma = MatrixParameter(“Sigma”, cov, symmetric=True) weights = VectorVariable(“w”, 2, lb=0, ub=1)

3 Access elements

Sigma[0, 1] # 0.01 Sigma.values # Full 2D array

4 Update for re-solve

new_cov = np.array([[0.05, 0.02], [0.02, 0.10]]) Sigma.set(new_cov)

4.1 Attributes

Name	Description
cols	Get the number of columns.
rows	Get the number of rows.
shape	Get the shape of the matrix.
symmetric	Check if this matrix is constrained to be symmetric.
values	Get the current matrix values.

4.2 Methods

Name	Description
col	Get a column as a 1D array.
row	Get a row as a 1D array.
set	Update the matrix values.
to_numpy	Get the matrix values as a numpy array.

4.2.1 col

core.parameters.MatrixParameter.col(j)

Get a column as a 1D array.

4.2.1.1 Parameters

Name	Type	Description	Default
j	int	Column index.	required

4.2.1.2 Returns

Name	Type	Description
	NDArray[np.floating]	1D array of column values.

4.2.2 row

core.parameters.MatrixParameter.row(i)

Get a row as a 1D array.

4.2.2.1 Parameters

Name	Type	Description	Default
i	int	Row index.	required

4.2.2.2 Returns

Name	Type	Description
	NDArray[np.floating]	1D array of row values.

4.2.3 set

core.parameters.MatrixParameter.set(values)

Update the matrix values.

4.2.3.1 Parameters

Name	Type	Description	Default
values	ArrayLike	New 2D array (must match original shape).	required

4.2.3.2 Raises

Name	Type	Description
	ValueError	If shape doesn’t match or symmetric constraint violated.

4.2.3.3 Example

Sigma.set(new_covariance_matrix)

4.2.4 to_numpy

core.parameters.MatrixParameter.to_numpy()

Get the matrix values as a numpy array.

4.2.4.1 Returns

Name	Type	Description
	NDArray[np.floating]	Copy of the internal 2D array.