Improved Gauging Interface, TNO construction, Bug Fixes, Apply Function for a Vidal ITN

[JoeyT1994]

This PR adds a number of features in ITensorNetworks.

1. The gauging interface has been expanded significantly (via gauging.jl). Functions are now available for belief propagation gauging a tensornetwork, starting in either symmetric or Vidal form. Functions are available for switching between these two forms, and for measuring the degree of `canonicalness' (i.e how well the itn satisfied certain canonical conditions) of a given itensornetwork (either in Vidal or Symmetric form).
2. An additional apply(o::ITensor, psi::ITensornetwork, bond_tensors::Datagraph) function has been added to apply.jl which performs a simple update on a tensornetwork in the Vidal gauge. Ideally this will eventually share more code with apply(o::ITensor, psi::ITensornetwork) but for now I have left them more separate.
3. Functionality has now been added (see tensornetworkoperators.jl) for grouping up the ITensors in a vector{ITensor}object into groups of commuting ITensors (i.e. ones that act on different physical degrees of freedom). This is the workhorse of a get_tnos function which builds a vector of tensornetworkoperators from a vector of ITensors.
4. A few tweaks have been made in beliefpropagation.jl to improve efficiency and avoid excessive calls to forming the union of two itensornetworks (which can be quite expensive).
5. Extensive testing has been added for the above, including additions to tests of the apply function

[codecov-commenter]

Codecov Report

Merging #88 (94069c8) into main (472ceb6) will increase coverage by 1.20%.
The diff coverage is 96.38%.

❗ Your organization is not using the GitHub App Integration. As a result you may experience degraded service beginning May 15th. Please install the Github App Integration for your organization. Read more.

@@            Coverage Diff             @@
##             main      #88      +/-   ##
==========================================
+ Coverage   78.04%   79.25%   +1.20%     
==========================================
  Files          62       61       -1     
  Lines        3289     3442     +153     
==========================================
+ Hits         2567     2728     +161     
+ Misses        722      714       -8     

Impacted Files	Coverage Δ
src/ITensorNetworks.jl	77.77% <ø> (ø)
src/apply.jl	91.21% <91.07%> (-0.46%)	⬇️
src/tensornetworkoperators.jl	91.66% <91.66%> (ø)
src/beliefpropagation.jl	92.75% <96.15%> (+5.48%)	⬆️
src/ITensorsExt/itensorutils.jl	90.90% <100.00%> (+19.48%)	⬆️
src/abstractitensornetwork.jl	81.88% <100.00%> (+0.04%)	⬆️
src/gauging.jl	100.00% <100.00%> (ø)
src/itensornetwork.jl	87.20% <100.00%> (+0.53%)	⬆️

... and 13 files with indirect coverage changes

[JoeyT1994]

Okay I incorporated the changes you suggested above! Do you know what is going on with the checks? Suddenly, for some reason certain packages aren't compiling? Even though I don't think I have changed the manifest/ dependencies.
In fact, I notice that these also failed in Miles' latest commit #77

[mtfishman]

Looks like SciML/LinearSolve.jl#330.

[JoeyT1994]

I see! That makes sense.

[mtfishman]

We don't have any control over that issue, so as long as things pass in Julia 1.9 I think it's fine to merge this.

[mtfishman]

Looks good, thanks! This will be very useful to have, I'm sure we'll find many use cases for it.

[emstoudenmire]

Great to see this, thanks Joey and Matt!

On Fri, Jun 16, 2023 at 2:37 PM Matt Fishman ***@***.***> wrote: Merged #88 <#88> into main. — Reply to this email directly, view it on GitHub <#88 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAHJ3RLDVRKUFXDWVMUW5VTXLSRVNANCNFSM6AAAAAAYCM2D2M> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.*** com>

-- -=Miles Stoudenmire=- ***@***.*** ***@***.*** http://itensor.org/miles/