torch_cg
torch_cg copied to clipboard
Adding Support for BICGSTAB
Hi, I needed an implementation of batched BICGSTAB, and found your code helpful as a starting point. I am opening this pull request in case you find this code useful too.
I implemented batched BICGSTAB, an algorithm that solves large, sparse, unsymmetric PD linear systems. Here is the original reference.
I kept the interface the same as the cg_batch
function, with the exception of the preconditioner. I did not implement the version of the method with preconditioners, but that seems like a relatively simple extension.
I wrote tests in pytest
.