Jutho
Jutho
I've only browsed through this quickly so far. The design is definitely different than what I had in mind. What I had in mind was that "all" TensorKit functions would...
Also, as response to the question, yes also `mul!` is supposed to become part of MatrixAlgebraKit and potentially support different backends/algorithms (I guess Octavian is not going to happen, but...
I am definitely in favor of something that can be controlled via scoped values.
Diagonal only exists if domain == codomain, so the problem is that only `flip(d, (1,2))` can preserve the `DiagonalTensorMap` structure.
`transpose(d)` could be used to flip the domain and codomain, but this might not have the same semantics in terms of preserving contractions.
> Oh, that's also true... I think we can work around this in this particular case by doing the `svd` in the "other direction", but this is definitely a bit...
It is true that the whole planar contractions are a bit less expressive than the corresponding general tensor contractions (even when the restrictions of being planar are fulfilled), in the...
Yes it comes from the AD applied to the SVD; it tells you that in the backward process, there is component entering the adjoint Jacobian of the SVD process that...
I was a bit confused; this is coming from the full `svd` in TensorKit, right? I had KrylovKit in mind. In TensorKit, the current cutoff for printing this warning if...
Ok, this is now changed on master, hopefully this fixes it. I will tag a new release this weekend, but there are some breaking changes.