# Issues with Pseudo-Inverse implementation using SVD

**URL:** <https://fortran-lang.discourse.group/t/issues-with-pseudo-inverse-implementation-using-svd/6004>\
**Category:** Uncategorized\
**Created:** [June 19, 2023, 2:53pm UTC](https://fortran-lang.discourse.group/t/issues-with-pseudo-inverse-implementation-using-svd/6004 "2023-06-19T14:53:34Z")\
**Posts on this page:** 1\
**Showing post:** 10

<div class="post-metadata">

**Author:** ![hkvzjal](https://yyz2.discourse-cdn.com/free1/user_avatar/fortran-lang.discourse.group/hkvzjal/32/3055_2.png) [@hkvzjal](https://fortran-lang.discourse.group/u/hkvzjal)\
**Post date:** [June 20, 2023, 7:16am UTC](https://fortran-lang.discourse.group/t/issues-with-pseudo-inverse-implementation-using-svd/6004/10 "2023-06-20T07:16:06Z")

</div>

> [@Ali](#):
>
> The matrices are sparse

If your matrices are sparse you can also consider ARPACK. You can try it from SciPy in python to see if it fits your needs [svds(solver=’arpack’) — SciPy v1.11.4 Manual](https://docs.scipy.org/doc/scipy/reference/sparse.linalg.svds-arpack.html). If it does, there are some discussions and sources for ARPACK here [Modernizing Arpack](https://fortran-lang.discourse.group/t/modernizing-arpack/3886)

---

_[View the full topic](https://fortran-lang.discourse.group/t/issues-with-pseudo-inverse-implementation-using-svd/6004)._