Andreas Borgen Longva

Thanks for looking into this @CattleProdigy! Could you perhaps start a draft PR? It would be easier to discuss than a free-standing commit. I'm sorry but I don't have the...

Stats aren't really my forte, but looks good to me. However, I think if there's an improvement in precision, it would be good to actually try to verify this with...

> > it would be nice to have a few more tests that ensure that we're not introducing any new bugs here. > > Indeed, it would be nice to...

I will chime in here to say that I think the correct behavior here is for the SVD to succeed, not return `None`. Handling empty/zero-sized matrices is an important corner...

> Friendly ping? Would be nice to get this merged (and ideally, released) to reduce compile-times for downstream projects. I agree. However, if I run `cargo tree -p nalgebra --all-features`,...

Actually, I realized that I hadn't updated dependencies in my dev environment for a while. After running `cargo update`, I see that the only other dependency that has not been...

@sebcrozet: it's unclear to me how to handle this correctly with respect to release. I think we need to: - bump `nalgebra-macros` to `v0.2.2` (I believe patch version should be...

I think it would be good to have some very basic tests that verify that it indeed does work in some basic cases. I think just verifying that an arbitrary...

Thanks for putting in the effort for this, but as pointed out by @Ralith, `nalgebra` already has several variants of the LU decomposition. I'm therefore going to close this PR,...

While this is definitely a bug, I do not necessarily agree that the first result should give `-3`, but instead `NaN` to indicate that the result is not well-defined.