Matrix equation solution via GenericLinearAlgebra with BigFloat not working

Specific Domains Numerics
1

I am trying to solve a system of linear equations:
A * x = B
where A and B are real symmetric positive definite matrices with type BigFloat.

I am trying to do the following:

using GenericLinearAlgebra
using LinearSolve

n = 100

A = big.(randn(n, n));
A = A * A';

B = big.(randn(n, n));
B = B * B';

c = big.(randn(n));

prob = LinearProblem(A, B);
solve(prob)

which yields the error

MethodError: no method matching ldiv!(::Vector{BigFloat}, ::LinearAlgebra.LU{BigFloat, Matrix{BigFloat}, Vector{Int64}}, ::Matrix{BigFloat})

If I use a vector instead of a matrix, i.e.

prob = LinearProblem(A, c);
solve(prob)

then it produces a result.

What is the workaround here to make it work for matrices? I can see in the GenericLinearAlgebra.jl package there are functions for matrix equations, but I am not sure how to access gthem.

2

Coming back to answer my own question, it turns out that the linear solution is done by

\(A, B)

for a pair of matrices.

1 Like