I am searching for a Julia package, that can emulate
pinv(A) from LinearAlgebra.jl but has an all Julia implementation to leverage the AD ecosystem to evaluate the gradients with respect to the parameters used in my computation (where pinv is a subroutine). Any leads on this matter is highly appreciated.
You don’t need to use
pinv to solve a least squares problem,
A\b will do it.
But if you’re looking for something that lets you change algorithms to say GMRES and sparse linear solvers, you can use LinearSolve.jl
LinearSolve.jl recently added support for Enzyme and more AD backends are coming rather soon.