Matt Fishman

Finally, in the signature of the function I don't think we should specialize `A::IsTreeOperator` and `psi0::IsTreeState`. Instead I think it should more generally be defined as: ```julia function ITensors.contract( ::ITensors.Algorithm"fit",...

@shinaoka I don't think using `init_guess=B` makes sense since it won't have the same external indices as `A * B`. I don't think there is any particular challenge to analyzing...

Also, we've decided that for now we won't explicitly support accepting `MPS`/`MPO` types in the functions in `ITensorNetworks.jl` (I'll change this throughout the package), so the signature should be: ```julia...

@shinaoka in #50 I changed this `contract` function to have the signature: ```julia function contract( ::Algorithm"fit", tn1::AbstractTTN, tn2::AbstractTTN; init, kwargs..., ) # [...] end ``` which helps us move towards...

@shinaoka that seems like a reasonable test to add. Does it seem like the current `contract(::TTN, ::TTN)` code addresses this original issue (i.e. can we close this issue)?

We could do it through a union type since we have a finite number of types in mind, but the more general solution is defining them as traits from [SimpleTraits.jl](https://github.com/mauro3/SimpleTraits.jl)...

An interesting perspective is that `ITensors.AbstractMPS` can be viewed as an implicitly defined graph, as in the graph layout is implicitly defined as a path graph by the type. The...

Also relevant is: https://github.com/cameton/LinearOrdering.jl with the associated reference https://arxiv.org/abs/2209.02895.

In recent versions of Julia it prints both: ```julia julia> Vector{Int} Vector{Int64} (alias for Array{Int64, 1}) julia> versioninfo() Julia Version 1.8.3 Commit 0434deb161e (2022-11-14 20:14 UTC) Platform Info: OS: macOS...

Note that `merge_vertices` is now implemented for `NamedGraph` in https://github.com/mtfishman/NamedGraphs.jl/pull/19, it still needs to be implemented for `DataGraph` for this to be implemented.