-
Dear jax team,
this is just a friendly bump on the implementation of eigendecomposition and batched SVD on GPU. Are you planning on implementing these?
Should I want to implement it myself, wou…
-
Hi!
Our team is currently developing sparse symmetric and non-symmetric solvers in JAX (implicitly restarted arnoldi and lanzcos). I have it from hearsay that there are efforts to support this nati…
-
A comprehensive listing of methods in the references.
Please remove any methods are impractical/obsolete and add methods worth implementing to the list.
Linear systems
- Stationary methods
- […
-
**Submitting author:** @sarah-ek (Sarah El Kazdadi)
**Repository:** https://github.com/sarah-ek/faer-rs
**Branch with paper.md** (empty if default branch):
**Version:** v0.15.0
**Editor:** @jedbrown
…
-
This issue is meant for a central place for our Fortran (almost exclusively F77) codebase to track per module and if possible remove depending on contributor affinity to the subject matter and fortran…
-
An eigenvalue problem is solved with Arpack and ArnoldiMethod. The eigenvalues agree, the eigenvectors do not.
```
##################################################### …
-
Naively, I'd expect that computed eigenvalues are exactly same whether eigenvectors are computed or not. However, for the symmetric `dsyevr` solver, this does not seem to be the case. The difference i…
ev-br updated
7 months ago
-
## 🐛 Bug
It seems that `torch.lobpcg` (https://pytorch.org/docs/stable/torch.html?highlight=lobpcg#torch.lobpcg) just always breaks when trying to take gradients via `backward`.
## To Reproduce
…
-
Hi there, when running `make_test`, I receive the following error
`KeyError: 'honour_petsc_dir'`
How do I solve this?
-
It is not really a failure; some ARPACK implementations (such as the one used in Debian Linux when compiling the igraph package from source) do not respect the `nev` setting; they will return _at leas…