Mateusz Baran

> Yeah probably not really useful. My point there was that we shouldn't have `pdf(Uniform(Sphere()), [1, 0, 0])` equal to anything finite really: it would break > . Only explicitly...

I may draft something in the near future, with implementations of a few distributions on the sphere. I'm actually only interested in spherical ones for now, but interface would be...

I'm somewhat afraid that this PR may again aim to achieve too much to be finished in a finite amount of time :wink: .

Yes, it tests takes much more time than we would like. I'm not sure if you include compilation time for numpy as well? Most of the time spent in our...

I've taken a look at ManifoldsBase.jl type piracy and it is a tough task. There are a few "most likely move to ManifoldsBase.jl" but majority of cases are either related...

The unbound parameter can be fixed, here is how the method should look like: ```julia function TuckerPoint( core::AbstractArray{T,D}, factors::Vararg{

By the way, maybe we should add Aqua tests in a separate PR to not make this PR too large?

OK, let's see how it works out.

What did you do to get that `GroupManifold` error?

That is because you can't put aliases on the `exclude` list. I guess Aqua.jl could be a bit more clear about it.