Mateusz Baran

Nice to hear that you like `CovarianceEstimation.jl`. With #86 merged, we have vectorization, so `Manifolds.jl` is ready now 🙂 .

Great, thanks for staring this! > * maybe `mean` could be also but into base, so that `ManifoldML` can purely depend on base? Moving `mean` to base would mean adding...

Sure, there is no rush to make ManifoldsML.jl independent from Manifolds.jl.

"set precision of displayed value" is the thing I wanted to do, so to me this is the key part that needs documenting. If you thing this should be done...

I had a quantity that, when printed it to a string, it gave me way too many digits, more than the precision with which the quantity was determined. So it...

What's the current status of this? I've recently been reading about manifold-valued continuous normalizing flows and it appears that they work well without the need to have non-bijective bijectors, see...

> > What's the current status of this? > > I've started working on making the scope of Bijectors.jl a bit wider in #183 by: > > 1. Allowing non-bijective...

> > Not sure how that is going to influence the AD story which you ran into last time. > > jip, that is the biggest problem. The type cannot...

That's one challenge. There is also the problem of embedded manifolds where we have to differentiate between ambient space representation and embedded manifold representation (where we have `AnotherPlaneManifold` that violates...

Sort of yes, it just requires solving them one by one: writing a new method, and then making a test which is sometimes not straightforward.